Login Register

Access the GIFSQ Portal

Select your user type to log in or register a new account.

Student Portal

Access your food safety courses and exams.

Instructor Portal

Manage courses, view student submissions, and grade quizzes.

Company Portal

Manage corporate setup, view employee logs, and access QA services.

GIFSQ ID: FTN-0108

← All subjects

Tea Coffee and Spice Technology

About this subject

Tea, coffee and spices reach the cup only after long chains of growing, picking, processing, grading and blending. This subject follows those chains from the plant to the market, covering the chemistry of tea and coffee, manufacturing technology, quality evaluation, process control and standardisation. It also covers tea growing in the field, major spice crops, post-harvest technology, plantation crops, and the effects of tea and spices on the body.

What you will learn

Units in this subject

  1. Unit 1: Tea and Coffee
  2. Unit 2: Chemistry of Tea and Coffee
  3. Unit 3: Manufacturing Technology of Tea and Coffee
  4. Unit 4: Quality Evaluation of Tea and Coffee
  5. Unit 5: Process Control and Standardisation
  6. Unit 6: Spices: Production, Chemistry and Technology
  7. Unit 7: Tea Breeding and Genetics
  8. Unit 8: Tea Propagation: Nursery, Seed and Grafting
  9. Unit 9: Young Tea, Pruning and Plucking
  10. Unit 10: Feeding, Watering and Mechanising Tea
  11. Unit 11: Protecting Tea: Diseases, Pests and Weeds
  12. Unit 12: Pesticide Residues and Organic Tea
  13. Unit 13: Tea Physiology and Biotechnology
  14. Unit 14: Tea Chemistry in Depth: Polyphenols, Theanine and Aroma
  15. Unit 15: Special Teas: Pu-erh, White and GABA
  16. Unit 16: Tea in the Body: Catechins, Antioxidants and Protection
  17. Unit 17: Tea, Weight and Metabolism
  18. Unit 18: Tea, Inflammation and Disease Prevention
  19. Unit 19: Tea Products and Flavour: Diversified Teas and Tea Drinks
  20. Unit 20: Spice Science: What a Spice Is, How It Is Classified and Graded
  21. Unit 21: Cooking with Spices and the Patterning of Spice Use
  22. Unit 22: Spices and Health: Physiological Effects, Antimicrobials and Antioxidants
  23. Unit 23: Major Spice Crops I: Black Pepper, Cardamom and Chillies
  24. Unit 24: Major Spice Crops II: Ginger, Turmeric and Seed Spices
  25. Unit 25: More Spice Crops: Bulbs, Bark, Buds and Leaves
  26. Unit 26: Post-Harvest Technology of Spices
  27. Unit 27: Plantation Crops: Coconut, Arecanut and Cashewnut
  28. Unit 28: Aromatic and Medicinal Plants
  29. Unit 29: Green Coffee: Processing, Grading and Blending
  30. Unit 30: Roasting and Grinding Coffee
  31. Unit 31: Instant Coffee: Extraction, Drying, Packing and Waste
  32. Unit 32: Decaffeination and Brewing the Cup

References and further reading

Books cited in the references of the units above.

Last reviewed: 10 November 2026. Curriculum: B.Tech Food Technology syllabus 2081.

Unit 1: Tea and Coffee

Note

Unit 1: Tea and Coffee

Tea and coffee are the two most widely drunk beverages in the world after water, and neither of them grows where most of the people who drink them live. Both come from a narrow belt of warm, high country, and both reach the cup only after a long chain of growing, picking, processing, grading and blending. This unit follows that chain from the plant to the market.

1.1 Two plants, two stories

Tea comes from Camellia sinensis, a member of the family Theaceae. Two basic varieties are recognised: the Chinese variety, C. sinensis var. sinensis, a small-leaved bush, and the Assamese variety, C. sinensis var. assamica, a larger-leaved tree. The plant originated in the highlands of East Asia.

Note on classification: the Handbook used in this unit recognises two varieties of the tea plant, C. sinensis var. sinensis and C. sinensis var. assamica. Ahuja, used in Unit 7, uses a three-type classification: China type Camellia sinensis, Assam type Camellia assamica, and Cambod type Camellia assamica ssp. lasiocalyx. The Assam plant is the same plant, ranked as a variety in one book and as a species in the other, and the Cambod type is the intermediate form between the two.

Its use as a beverage dates back to 350 CE, and it reached Europe in the 16th century. From its homeland it spread to the tea-growing regions of Asia, Africa and South America, and it is now grown wherever the climate suits it.

Coffee comes from the genus Coffea, in the family Rubiaceae, which holds over 70 species. Only seven have economic importance, and two dominate the trade: arabica (Coffea arabica) and robusta (Coffea canephora). A third species, liberica, was devastated in the 1940s by an epidemic of trachymycosis, a wilt disease caused by a fungus, and its commercial growing has effectively ceased since then. A hybrid of arabica and robusta, arabusta, has better flavour than robusta and more disease resistance than arabica, but its growing is limited by poor yield and small beans.

The discovery of coffee is told as a legend, and it should be told as one. A goatherd tending his flock in the hills around a monastery noticed that his goats, after chewing berries from the bushes, started prancing excitedly. A monk took some berries, roasted them and brewed them, and the brew kept his people alert during long prayers at night. The word coffee comes from the Arabic quahweh, a poetic term for wine. The wild coffee plant is indigenous to the highlands of north-east Africa, from where it spread to Arabia and nearby lands. The first written mention of coffee is by Razes, a 10th-century physician, and cultivation may have begun as early as the 6th century. For a long time the raw beans were not allowed out of Arabia without steeping in boiling water or heating, to destroy their power to germinate, and strangers were not allowed to visit the plantations. The beans spread to the rest of the world through travellers who carried a few seeds capable of germination, and planting followed in the tropical highlands of Asia and the Americas.

🤔 Pause and think: both tea and coffee spread from one small homeland to the whole world. What do the two plants have in common that let them travel so far and grow in so many places?

Quick check

Name the two commercial coffee species and the two recognised varieties of the tea plant.

Check your answer

Coffee: arabica (Coffea arabica) and robusta (Coffea canephora). Tea: the Chinese variety, C. sinensis var. sinensis, and the Assamese variety, C. sinensis var. assamica.

1.2 Where they grow: climate and soil

Tea is cultivated in tropical and subtropical regions with a temperature range of 13 to 29 °C, at altitudes up to 2,460 m above sea level, in acidic soil rich in iron and manganese, with a pH range of 3.3 to 6.0, preferably 4.5 to 5.5. Tea is the most widely consumed of ancient beverages: world production in 1996 was 2,622,000 metric tons.

Coffee grows in the tropical and subtropical belt, in highland country with warm days, cool nights and a clear dry season for harvesting. It is one of the most important agricultural products traded worldwide, grown and exported by over 70 developing countries in the tropical belt, while the industrialised countries import and consume most of it. Of those 70, 51 countries are responsible for more than 99% of world output and are exporting members of the International Coffee Agreement.

Both crops are therefore highland crops: they need warmth without great heat, rain without waterlogging, and altitude to slow their growth and build flavour. That is why the best teas and coffees come from hills, not plains.

Quick check

Tea grows best in soil of pH 4.5 to 5.5. Is that soil acid or alkaline, and why does it matter?

Check your answer

Acid: a pH below 7 is acid. Tea needs acidic soil rich in iron and manganese, and it grows poorly in alkaline soil.

1.3 Cultivation and variety: tea

Tea is grown from seed or from cuttings of selected clones. The bushes are pruned to a low, flat table so the pluckers can reach the new shoots. Only the rapidly growing shoot tips are plucked, down to about the second or third unfolded leaf, and the plucking is done by hand at intervals of 7 to 14 days throughout the year.

Variety matters because the leaf decides the cup. Long plucking intervals, or clones that stray from the desirable leaf standard, lower the levels of theaflavin, caffeine and volatile flavour compounds and give poor sensory properties, with sweet, flowery and grassy aromas. The chemical composition of the tea flush, as a percentage of dry weight, is: flavanols 25%, caffeine 4%, amino acids 4%, organic acids 0.5%, monosaccharides 4%, polysaccharides 13%, protein 15%, cellulose 7%, lignin 6%, lipids 3%, chlorophyll and other pigments 0.5%, ash 5%, and volatiles 0.1%. The flavanols, at a quarter of the dry weight, are the raw material of the colour and strength of black tea, as Unit 2 shows.

Quick check

Why are only the shoot tips, down to the second or third leaf, plucked for tea?

Check your answer

The young shoot tips hold the highest levels of flavanols, caffeine and volatile flavour compounds. Older leaves lower those levels and give poor flavour.

1.4 Cultivation and variety: coffee

Coffee is grown from seed in nurseries, then planted out in the field, often under shade trees. Arabica and robusta differ in more than name. Robusta is noted for its resistance to disease and holds more caffeine than arabica, about 2.2% against about 1.2% in the green bean, which makes it more economical for instant coffee. The hybrid arabusta holds about 1.72% caffeine. Arabica gives the finer flavour and the higher price; robusta gives body, strength and a higher caffeine yield.

The fruit is a cherry that ripens from green through yellow to red. It is picked when ripe, either by hand, picking only the ripe cherries, or by stripping whole branches. Hand picking gives the best quality because unripe and overripe cherries are left on the tree; stripping is faster but mixes ripe and unripe fruit, and the unripe fruit spoils the cup.

Quick check

Why does robusta suit instant coffee manufacture better than arabica?

Check your answer

Robusta holds about 2.2% caffeine in the green bean against about 1.2% for arabica, so it gives more caffeine per kilo, and it is also more resistant to disease.

1.5 Production: from field to factory

Tea production runs all year where the climate allows it. The plucked leaf must reach the factory the same day: the leaf is alive and its chemistry changes from the moment it is picked, and the withering that starts the manufacture must be controlled, not left to chance. The factory steps are plucking, withering, leaf distortion, fermentation, firing, grading, packing and storage, covered in Unit 3.

Coffee production is seasonal. The cherries are picked, taken to a central processing unit, and processed within 36 hours of picking: from picking in the early morning to late afternoon, prompt transport, pulping and removal of mucilage overnight, and washing, before the beans go to drying. The moisture of the fresh cherry, 65 to 70%, falls to about 45% in that first handling. The two routes, wet and dry, are covered in Unit 3: the wet route gives parchment coffee, the dry route gives cherry coffee.

In both crops the factory is near the field, because the raw material spoils fast and the first processing must happen where the crop grows. Roasting (coffee) and retail blending happen near the buyer; grading happens at the factory (tea) or curing works (coffee), near the field.

Tea and coffee: from field to market

Field: highland, warm, acid soil

↓

TEA
Pluck shoot tips every 7 to 14 days, then to the factory: wither, roll, ferment, fire; grade, pack. Tea blending is a separate, later step.
COFFEE
Pick cherries, then within 36 hours to the factory: pulp or dry, wash, dry to 10 to 12% moisture; hull, grade, pack. Roasting and blending happen near the buyer.

Figure 1.1: Tea and coffee production, from field to market. Tea and coffee are two parallel paths, not one sequence: both are processed near the field, grading happens at the factory (tea) or curing works (coffee), and roasting and retail blending happen near the buyer. Drawn from the production accounts in Nagalakshmi, Tea, and Ramalakshmi and Raghavan, Coffee.

Quick check

Why must coffee cherries be processed within 36 hours of picking?

Check your answer

The cherry is wet, 65 to 70% moisture, and warm. Left longer, it ferments uncontrolled, heats, and spoils, and the cup quality is lost.

1.6 Marketing: grading, blending and the auction

Tea and coffee are not sold as a single product. They are graded, blended and sold through a chain of brokers, tasters and buyers.

Tea is graded by particle size after manufacture: the grades fall into four broad categories, leaf, brokens, fannings and dust, based on appearance, size and liquor qualities. The graded teas are packed at 3% moisture in tea chests, preferably with aluminium foil or pliofilm lining, in humidity-controlled rooms. Brokers know the quality and value of each grade, and buyers blend the grades in different proportions for the needs of different markets. Retail tea is invariably a blend of 20 to 40 different components in various proportions, chosen on quality, flavour, strength, body, size and style of leaf, and price. Cheap neutral teas, called fillers, balance the cost. Because no tea grade has consistent quality, varying with processing and season, regional blending is done only by expert tasters: there is no objective method that replaces them.

Coffee is graded by size and density after hulling, and lots are sold by grade and by origin. Roasting usually happens near the buyer, because roasted coffee stales fast. The trade is governed by the International Coffee Agreement, whose exporting members supply more than 99% of world output.

Quick check

Why is retail tea always a blend of 20 to 40 different teas?

Check your answer

No single grade has consistent quality, because quality varies with processing and season. Blending many grades lets the buyer hold a steady flavour, strength and price all year.

1.7 The industries: how tea and coffee are organised

Tea and coffee are plantation industries: large areas of one crop. Tea has an estate factory that processes the harvest on the estate, and coffee has a curing works that prepares the product for sale. The work is seasonal for coffee and year-round for tea, and the labour is skilled: plucking tea by hand and picking only ripe coffee cherries cannot be done well by machine.

The outline for this unit names the tea and coffee industries of one place. The notes keep the worldwide view required for this subject, so the industry is described here in general terms, as it runs wherever tea and coffee are grown: tea through its estate factory, coffee through its curing works, then grading, blending, auction and export. The structure is the same across the growing belt, and the differences are in scale, in the share of smallholders against estates, and in how much processing happens near the field against near the buyer.

Quick check

Why is the tea factory built on the estate, while roasting usually happens near the buyer?

Check your answer

Fresh tea leaf spoils within hours and must be withered the same day, so the factory must be where the leaf grows. Roasted coffee stales fast, so it is roasted near the buyer and sold fresh.

1.8 Why highland leaf tastes different: slow growth and the plucking standard

Section 1.2 described where tea and coffee grow, and section 1.3 described how they are cultivated. This section joins the two: the flavour in the cup is decided in the field, by how fast the plant grows and by which part of it is picked. A factory can spoil good leaf, but it cannot make fine leaf out of coarse leaf, so the field standard is where quality starts.

Slow growth makes strong leaf. Tea grows best in warm, wet highlands, and the same warmth that makes it grow also decides its flavour. In cool highland air the shoot grows slowly, and the leaf has time to build up the flavour compounds, the polyphenols and the aroma substances, before it is picked. In hot lowland air the shoot grows fast, the leaf is larger and softer, and the flavour is thinner, because the same compounds are spread through more leaf. That is why highland teas carry the stronger, more complex character and why the plucking round matters: a shoot picked every 7 to 14 days, as section 1.3 notes, is picked young, while a shoot left for a month is coarse and woody.

The plucking standard. The standard is usually two leaves and a bud (down to the second, sometimes the third, unfolded leaf): the soft tip of the shoot, with the unopened bud and the two youngest leaves below it. The bud is the richest part, holding the most flavour compounds, and the young leaves are next; the older leaves further down the shoot are tougher, hold less flavour, and bring a coarse note to the cup. A plucker who takes three or four leaves, or who snaps the shoot low, raises the weight of the basket and lowers the quality of the tea, because every coarse leaf dilutes the fine ones. The factory pays for the standard, and the taster in Unit 4 can taste when it has slipped: the liquor loses its brightness and briskness, and the infused leaf shows coarse, dark pieces among the fine ones.

Coffee cherries: ripeness is the standard. Coffee has its own version of the same rule. The cherry must be picked ripe and red, not green and not overripe and black. A green cherry holds little sugar and gives a harsh, thin cup; an overripe cherry has started to ferment on the tree and gives a sour, over-fermented note. Pickers go through the same trees several times, taking only the red cherries each time, which is slow work and the reason good coffee costs more. Cherries must reach the factory within 36 hours of picking, as section 1.5 notes, because the sweet pulp around the bean starts to ferment at once, and a cherry that waits in a sack loses the clean fruit character that ripe picking earned.

What the grower controls, and what the grower cannot. The grower controls the plucking standard, the picking round, the shade and the pruning, and these decide the quality of the leaf and the cherry. The grower cannot control the weather, and a season that pushes growth too fast, or a drought that stresses the plant, will show in the cup whatever the factory does. That is why tea and coffee are bought by tasting, not by specification alone: the leaf and the cherry carry the season in them, and the cup test in Unit 4 is the check that the field standard was kept.

🤔 Pause and think: two gardens sit side by side, one plucking two leaves and a bud every 10 days, the other plucking four leaves every 25 days. The second garden harvests more weight per hectare. Which garden's tea sells for more per kilo, and why?

Quick check

Why does a slow-growing highland shoot give a stronger tea than a fast-growing lowland shoot?

Check your answer

The slow shoot has time to build up flavour compounds before it is picked, while the fast shoot spreads the same compounds through more leaf, so the flavour is thinner.

Try it yourself: Brew two teas if you can, one whole-leaf and one made of broken, dusty leaf, at the same weight and time. Taste them side by side: the whole-leaf cup should taste brighter and cleaner, because the leaf was picked young and handled whole, while the broken leaf gives up its flavour fast and flat.

1.9 Check yourself: quick self-check

Think about it first: a buyer wants the same tea flavour every month of the year. The harvest changes with the season, the grade changes with the factory, and no two lots are alike. How does the trade give the buyer a steady cup?

  1. What are the two recognised varieties of the tea plant?
Check your answer

The Chinese variety, C. sinensis var. sinensis, and the Assamese variety, C. sinensis var. assamica.

  1. Why is robusta more economical than arabica for instant coffee?
Check your answer

Robusta holds about 2.2% caffeine in the green bean against about 1.2% for arabica, so it gives more caffeine per kilo, and it resists disease better.

  1. Tea grows best at pH 4.5 to 5.5 and 13 to 29 °C, up to 2,460 m. What do those three numbers describe?
Check your answer

The soil acidity, the temperature range, and the altitude range for tea cultivation.

  1. Why must coffee cherries be pulped and washed within 36 hours of picking?
Check your answer

The cherry is 65 to 70% water and warm. Left longer it ferments uncontrolled and spoils, and the cup quality is lost.

  1. A retail tea holds 20 to 40 different components. Why so many?
Check your answer

No grade has consistent quality across seasons and factories, so many grades are blended to hold a steady flavour, strength and price, with cheap neutral filler teas balancing the cost.

Try it yourself: Buy two teas of different grades, one leaf grade and one fannings or dust grade, and brew both with the same weight of leaf and the same time. Compare the colour, strength and flavour. The difference is what the grader and the blender are working with.

Key ideas to carry forward

  • Tea is Camellia sinensis, in two varieties, Chinese and Assamese; coffee is Coffea, with arabica and robusta as the commercial species.
  • Tea grows at 13 to 29 °C, up to 2,460 m, in acid soil of pH 4.5 to 5.5; coffee grows in the tropical highland belt.
  • Tea leaf is plucked by hand every 7 to 14 days, shoot tips down to the second or third leaf; the flush holds about 25% flavanols and 4% caffeine on a dry basis.
  • Robusta holds about 2.2% caffeine and arabica about 1.2%; robusta suits instant coffee and resists disease.
  • Coffee cherries must be processed within 36 hours; tea leaf must reach the factory the same day.
  • Tea is graded into leaf, brokens, fannings and dust, packed at 3% moisture, and sold as blends of 20 to 40 components, mixed only by expert tasters.
  • Flavour is made in the field: slow highland growth concentrates flavour compounds, and the plucking standard of two leaves and a bud keeps coarse leaf out of the cup.
  • Coffee quality rests on picking only ripe red cherries and getting them to the factory within 36 hours, before the pulp ferments.

Glossary

Camellia sinensis

The tea plant, in the family Theaceae, with Chinese and Assamese varieties. Food tech example: shoot tips of Camellia sinensis plucked for black tea.

Arabica

Coffea arabica, the coffee species with the finer flavour and about 1.2% caffeine in the green bean. Food tech example: arabica beans roasted light for a fine, acid cup.

Robusta

Coffea canephora, the coffee species with more caffeine, about 2.2%, and more disease resistance. Food tech example: robusta beans used for instant coffee.

Flush

The young shoot tips of the tea plant that are plucked for manufacture. Food tech example: a flush of a bud and two leaves, holding about 25% flavanols on a dry basis.

Plucking

Picking tea shoot tips by hand, every 7 to 14 days. Food tech example: plucking down to the second or third unfolded leaf.

Parchment coffee

Coffee made by the wet route: the cherry is pulped, the mucilage removed, and the bean dried in its parchment skin. Food tech example: washed arabica sold as parchment coffee before hulling.

Cherry coffee

Coffee made by the dry route: the whole cherry is dried in the sun and then hulled. Food tech example: sun-dried cherries hulled to green beans at a ratio of two parts dry cherry to one part green bean.

Blending

Mixing teas of different grades and origins to hold a steady flavour, strength and price. Food tech example: a retail tea blended from 20 to 40 different components.

Filler

A cheap, neutral tea used in a blend to balance the cost. Food tech example: a filler tea that adds body without changing the flavour of the blend.

International Coffee Agreement

The agreement whose exporting members supply more than 99% of world coffee output. Food tech example: the framework under which most of the world's coffee is traded.

Plucking standard

The rule for which part of the tea shoot is picked, usually two leaves and a bud. Food tech example: a plucker who takes four leaves raises the basket weight and lowers the tea quality.

Briskness

The lively, fresh character of a tea liquor, felt on the tongue, which comes from fine young leaf. Food tech example: a tea that has lost its briskness was often plucked coarse.

References

  • Nagalakshmi, Srikantayya. Tea: An Appraisal of Processing Methods and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ramalakshmi, Kulathooran, and Raghavan, Bashyam. Coffee: A Perspective on Processing and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.

Unit 2: Chemistry of Tea and Coffee

Note

Unit 2: Chemistry of Tea and Coffee

A green tea leaf and a black tea leaf come from the same bush, yet one brews pale green and grassy and the other brews dark red and malty. Nothing was added to the black leaf: its own chemistry was switched on and left to run. This unit is about the chemicals that change, and the ones that do not.

2.1 What the tea leaf holds

The fresh tea flush holds, as a percentage of dry weight: flavanols 25%, caffeine 4%, amino acids 4%, organic acids 0.5%, monosaccharides 4%, polysaccharides 13%, protein 15%, cellulose 7%, lignin 6%, lipids 3%, chlorophyll and other pigments 0.5%, ash 5%, and volatiles 0.1%. Three groups in that list do most of the work in the cup: the flavanols, which become the colour and strength of black tea; caffeine, which gives the lift and the tang; and the amino acids, which build aroma.

The flavanols of tea are catechins, and they make up 25 to 30% of the dry matter of the fresh leaf. The major catechin is epigallocatechin gallate, and the gallated flavanols fall as the leaf ages. Tea quality is mainly related to the flavanol concentration of the fresh leaf: a leaf poor in flavanols cannot make a strong black tea, whatever the factory does to it.

🤔 Pause and think: two teas are made the same way in the same factory, but one is strong and bright and the other is weak and dull. The difference is in the leaf. Which chemical in the leaf decides it?

Quick check

Which group of chemicals in the tea leaf, at 25 to 30% of dry matter, becomes the colour and strength of black tea?

Check your answer

The flavanols, which are catechins. The major one is epigallocatechin gallate, and tea quality follows the flavanol concentration of the fresh leaf.

2.2 What fermentation does to tea: theaflavins and thearubigins

Black tea is made by letting the leaf's own enzymes oxidise its catechins. The process is called fermentation in the tea trade, although no yeast is involved: it is an enzyme oxidation, driven by polyphenol oxidase. The steps run like this:

  1. Catechins are oxidised to quinones. Quinone formation is the primary driving force of fermentation.
  2. A quinone from a simple catechin joins a quinone from a gallacatechin to form a compound with a benzotropolone ring, called a theaflavin.
  3. Most of the catechins go further, into a complex, poorly defined group of substances called thearubigins, known by their red-brown colour.

The numbers from the books show how the catechins divide: about 15% of the tea catechins remain unchanged, about 10% are accounted for by theaflavin, theaflavic acid and bisflavanol formation, and about 75% are converted to thearubigins. The thearubigins are the largest group of compounds in black tea, up to 20% of dry weight, and they give the colour, strength and mouth feel of the beverage. The total theaflavin concentration in black tea varies from 0.3 to 2.0%, and only 10% of the original catechin content of the flush is accounted for as theaflavins. Prolonging fermentation decreases the theaflavin content. Tannin falls during fermentation, from 20% in green tea leaf to 10 to 12% in fermented tea.

The theaflavins are bright orange-red in solution and give the brightness of the liquor and the brisk sensation on the palate. The thearubigins give the bright, vivid red colour. Together, theaflavins and thearubigins are the index of liquor strength, and the books use their balance to judge whether fermentation was done well: a tea with a good balance has good colour, strength, briskness and flavour.

A note on the ratio. The handbook gives the theaflavin to thearubigin (TF:TR) ratio as about 1:10 under ideal fermentation conditions, and about 1:10 to 1:12 in general during fermentation. The two statements agree: 1:10 sits inside 1:10 to 1:12. The balance varies with the leaf, the fermentation time and temperature, and the method of measurement, and it is judged by the liquor it gives, not by a fixed figure alone.

What happens to tea catechins during fermentation

Catechins in fresh leaf: 25 to 30% of dry matter

↓

Oxidised by polyphenol oxidase: quinones form

↓

About 15%
Stay as catechins
About 10%
Become theaflavins, plus theaflavic acids and bisflavanols (0.3 to 2.0% of black tea)
About 75%
Become thearubigins, up to 20% of dry weight

What they give: briskness, brightness, colour, strength, mouth feel.

Figure 2.1: What happens to tea catechins during fermentation. The three fates are three parallel shares of the same catechins, not successive steps: most become thearubigins; a small share becomes theaflavins. Drawn from the pigment chemistry in Nagalakshmi, Tea.

Quick check

Why does prolonging fermentation lower the theaflavin content of black tea?

Check your answer

Theaflavins are an intermediate product. Given more time, they are converted further into thearubigins, so their share falls while the thearubigin share rises.

2.3 Tea aroma: more than 600 compounds

More than 638 aroma compounds in tea have been identified. The most important groups are terpenes, terpene alcohols, lactones, ketones, esters and spiro compounds. They form during fermentation: amino acids break down by Strecker's degradation to aldehydes, and carotenes and unsaturated lipids oxidise to ionones, terpene alcohols, terpene aldehydes and their oxidation products, such as theaspirone and dihydroactinidiol.

The aroma differs by type and by origin. Orthodox teas hold higher concentrations of aroma constituents and volatile oil than CTC teas. CTC teas give a strong brew and body, more cuppage, and a malty aroma, while orthodox teas give a light brew with fine, flowery and rich aromas, from linalool, linalool oxides, methyl salicylate, geraniol, phenyl ethanol, cis-jasmone and beta-ionone. Famous teas carry their own markers: Darjeeling holds linalool, linalool oxides and geraniol; Keemun holds 2-phenylethanol and geraniol; Uva holds methyl salicylate, linalool and linalool oxides.

Other polyphenols in the leaf include gallic acid, flavanols such as quercetin, kaempferol and myricetin and their glycosides, and depsides such as chlorogenic acid and p-coumarylquinic acid. Theogallin, 3-galloylquinic acid, is unique to tea.

Quick check

Why does an orthodox tea smell more flowery than a CTC tea from the same leaf?

Check your answer

Orthodox tea holds higher concentrations of aroma constituents and volatile oil. CTC teas give strength and body and a malty aroma, while orthodox teas keep the flowery compounds such as linalool, geraniol and methyl salicylate.

2.4 Caffeine and tannin in tea

Tea leaves hold 2.5 to 4.0% caffeine, with traces of theophylline and theobromine. Caffeine is the important component of the nonvolatile fraction: it gives a modified pleasant taste to the infusion in the presence of polyphenols, it decides the level of tangy astringency in black tea infusion, and the refreshing property of tea is due to it. The effect of caffeine is most commonly linked to stimulation of the central nervous system. It increases muscular work, relieves allergy, controls obesity, increases urine flow and treats the common cold, but excess intake can lead to caffeism, causing anxiety and neurosis. Doses of 100 mg per kg of body weight are a serious poisoning hazard in children, and the acute fatal dose in humans is 170 mg per kg of body weight. Increased intake can raise the heartbeat, promote stomach acid, step up urine production and cause skin disorders.

Tannin in tea is the oxidised polyphenol complex, mainly thearubigins, and it has no relationship to tannic acid, which does not occur in tea. Processed black tea holds about 10% tannins, against 18 to 20% in green tea leaves. Tannin is estimated by Lowenthal's method: the tannins are extracted into water, precipitated with gelatin, and titrated with standard permanganate solution before and after precipitation, giving the oxidisable tannins. Caffeine is estimated by the Baily-Andrew method: the leaves are digested with heavy magnesium oxide, the caffeine is extracted into chloroform, nitrogen is measured in the crude caffeine by the micro-Kjeldahl method, and the result is multiplied by a conversion factor.

Quick check

Black tea holds about 10% tannins and green tea leaves hold 18 to 20%. Why the difference?

Check your answer

During fermentation the tannins fall from 20% in green leaf to 10 to 12% in fermented tea, as the polyphenols are oxidised into theaflavins and thearubigins.

2.5 Green tea, yellow tea and Oolong: the unfermented and partly fermented types

Green tea is unfermented: the enzyme is inactivated at the start, by steaming in the Japanese method or by roasting the leaves in a pan in the Chinese method, so oxidation cannot develop. Green tea holds larger amounts of catechins and vitamins than black tea, and it has a pleasant taste, flowery aroma and light green colour with an olive shade. Around 21% of total tea production is consumed as green tea. There are 18 different types, classed as pan-fixed and steam-fixed.

Yellow tea is unfermented but sits between black and green tea, closer to green. Its infusion is brighter than green tea, its taste milder and its aroma stronger. It holds more catechins, vitamins and extractives than black tea. It is made from a bud and two or three young leaves, withered, roasted, rolled, sorted and fired.

Oolong is the partially fermented tea. The books name it as one of the three conventional types, alongside totally fermented black tea and unfermented green tea, but they give it no manufacturing or chemistry section of its own. This unit therefore describes it only as the books do: a partially fermented tea, between green and black, with its own flavour from a short fermentation. The detailed chemistry of Oolong is not in the sources on disk, and it is not invented here.

Red tea is a fermented tea made by a controlled combination of enzymic and thermochemical processes, with the enzymic rate slower than in black tea manufacture. It has a pleasant, mild, astringent taste and a strong, stable aroma. Dark tea is an unfermented type in which piling, holding the tea warm and damp so microbes grow, oxidises the polyphenols and mellows the flavour; it holds no chlorophyll and lower catechins and free amino acids than yellow tea. White tea is made from buds and one or two leaves withered for 1 to 3 days, during which amino acids rise and sugars and polyphenols fall.

Note on the book's classification: the Handbook classes yellow and red tea under partially fermented Oolong in section 1, but calls yellow tea unfermented and red tea fermented in sections 2.3 to 2.4, and calls dark tea unfermented while describing microbial oxidation in piling. This unit follows sections 2.3 to 2.4, and the tension is the book's own.

🤔 Pause and think: the same leaf can become green tea, Oolong or black tea. What single decision in the factory decides which one it becomes?

Quick check

What is the difference between green tea and black tea in one sentence?

Check your answer

Green tea is unfermented because its enzymes are inactivated by steaming or roasting at the start, while black tea is fully fermented by its own enzymes oxidising the catechins.

2.6 What the coffee bean holds

Green coffee holds, as percentages: holocellulose, or fibre, 18%; lignin 2%; oils 13%; protein 13%; ash 4%; chlorogenic acid 7%; other acids 1%; trigonelline 1%; and caffeine 1%. The fuller comparison in the books is:

ComponentArabica greenArabica roastedRobusta greenRobusta roastedInstant powder
Caffeine0.9-1.2%about 1.0%1.6-2.4%about 2.0%4.5-5.1%
Total chlorogenic acid5.5-8.0%1.2-2.3%7.0-10.0%3.9-4.6%5.2-7.4%
Trigonelline1.0-1.2%0.5-1.0%0.6-0.7%0.3-0.6%-
Lipids12.0-18.0%14.5-20.0%9.0-13.0%11.0-16.0%1.5-1.6%
Proteins11.0-13.0%13.0-15.0%11.0-13.0%13.0-15.0%16.0-21.0%

Green coffee holds about 10 to 13% water; roasted and soluble coffees hold about 2 to 5%.

Caffeine is an alkaloid with a substituted purine ring system. Its content varies by species: robusta holds about 2.2%, arabica about 1.2%, and the hybrid arabusta about 1.72%. Climate and farming appear to have minimal effect on it. On roasting, caffeine is unchanged, though some is lost by sublimation. The fatal dose of caffeine is 10 g. Caffeine is broken down slowly in the body: demethylation yields theobromine, theophylline and xanthine, which are metabolised to urea.

Chlorogenic acid is the second most important component of coffee. Its level depends on species and is little affected by farming practice: robusta holds more than arabica, and roasting cuts it sharply, from 5.5 to 8.0% to 1.2 to 2.3% in arabica. Trigonelline falls on roasting too, from about 1% to 0.5 to 1.0% in arabica, and it is one of the compounds that break down to give volatile flavour compounds.

Quick check

Roasting cuts chlorogenic acid in arabica from 5.5-8.0% to 1.2-2.3%, but leaves caffeine almost unchanged. Why the difference?

Check your answer

Chlorogenic acid breaks down in the heat of roasting, while caffeine is heat-stable and is unchanged except for a small loss by sublimation.

2.7 What roasting does to coffee chemistry

Green coffee has very little aroma. Roasting develops it, and the changes are both physical and chemical: the bean changes colour, size and shape, and reactions of oxidation, reduction, hydrolysis, polymerisation and decarboxylation take place. The sugars and amino acids react, the chlorogenic acids break down, trigonelline decomposes into volatile compounds, and the oils move to the surface of the bean. The degree of roast decides the flavour: a very light roast at 200 °C tastes acid with little aroma; a medium roast at 230 °C gives optimal quality; a dark roast at 240 °C is preferred in Europe; and an Italian roast at 250 °C gives a dark colour and good taste. The weight loss rises with the roast, from 11 to 12% at very light to 24 to 25% at Italian.

Quick check

Why does a dark roast taste less acid than a light roast from the same beans?

Check your answer

Roasting breaks down the acids, including chlorogenic acid, and builds bitter, roasted flavours. The longer and hotter the roast, the more acid is lost and the more roast character appears.

2.8 Practical note: estimating caffeine and tannin

Estimation of caffeine in tea and coffee. Caffeine is extracted into an organic solvent such as chloroform, purified, and measured. The modern method is high-performance liquid chromatography: an aqueous extract of the coffee is prepared, cleaned by precipitation or solid-phase extraction, and injected into the chromatograph, where caffeine is detected by ultraviolet absorption at 272 nm. For tea, the Baily-Andrew method is used: digest the leaves with heavy magnesium oxide, extract the caffeine into chloroform, and measure the nitrogen in the crude caffeine by the micro-Kjeldahl method, multiplied by the conversion factor. The HPLC method for tea brews 2 g of tea in 177 ml of water for 6 minutes, which estimates the caffeine that dissolves into the cup in that time.

Estimation of tannin in tea. Tannins are extracted into water and precipitated with gelatin, then titrated with standard permanganate solution before and after precipitation. The difference gives the oxidisable tannins present.

Proximate analysis of tea and coffee. The routine analysis covers moisture, total ash, water-soluble ash, acid-insoluble ash, water extract and crude fibre, with limits set by standards bodies: total ash 4.0 to 8.0%, water-soluble ash at least 40.0% of total ash, acid-insoluble ash at most 1.0%, alkalinity of water-soluble ash as K₂O 1.0 to 2.2%, water extract at least 32.0%, and crude fibre at most 16.5%.

Quick check

Why measure caffeine by HPLC on a 6-minute brew instead of on the dry leaf?

Check your answer

The brew measures the caffeine that actually dissolves into the cup in 6 minutes, which is the caffeine the drinker takes in. The dry leaf figure is higher, because not all of its caffeine dissolves.

2.9 Measuring caffeine and tannin: the Beer-Lambert derivation

Section 2.8 named the practical work: estimating caffeine and tannin in tea and coffee. This section gives the maths that makes the estimation possible. The instrument is the spectrophotometer, and the law it uses is the Beer-Lambert law, the same law that the Beer-Lambert derivation in the Introduction to Analytical Techniques subject works through in full. Here it is applied to tea and coffee.

Step 1: light is lost as it crosses the solution. Shine light of one colour through a coloured solution, such as a tea extract, and some of it is absorbed. The more coloured molecules the light meets, the more is lost. Two things decide how many it meets: how concentrated the solution is, and how far the light travels through it. Double the concentration and the light meets twice as many molecules; double the path length and it meets twice as many again.

Step 2: absorbance is the measure. The instrument reads transmittance, the share of the light that gets through, and turns it into absorbance, A, which is the logarithm of the loss: A = log(1 ÷ T), where T is the transmittance as a fraction. The logarithm matters, because each layer of solution absorbs the same share of the light that reaches it, not the same amount, so the loss multiplies layer by layer. The logarithm turns that multiplication into addition, and the result is a straight line.

Step 3: the law itself. The Beer-Lambert law states the straight line:

A = ε × c × l

Here c is the concentration, l is the path length, usually 1 cm, and ε is the molar absorptivity, a number that belongs to the molecule and the colour of light used: it says how strongly one mole of that molecule absorbs that light. Double the concentration and the absorbance doubles; halve the path and the absorbance halves.

Step 4: from absorbance to concentration. The law is used backwards. The analyst makes a series of standard solutions of known concentration, reads their absorbances, and draws the calibration line of absorbance against concentration. The slope of the line is ε × l. Then the tea or coffee extract is read in the same instrument, at the same wavelength, and its concentration is read off the line, or worked out as c = A ÷ (ε × l). Caffeine and tannin are read at different wavelengths, because each absorbs most strongly at its own colour of light, which is how the two are told apart in the same extract.

Worked example: reading off the line. A caffeine standard of 10 mg per litre reads an absorbance of 0.25. A tea extract, prepared the same way and diluted tenfold before reading, reads 0.50. The absorbance has doubled, so the diluted extract holds twice the standard: 20 mg per litre. Undoing the tenfold dilution gives 200 mg per litre in the original extract. The dilution step is there because the law only holds on the straight part of the line: a very dark extract absorbs so much light that the reading leaves the line, so the analyst dilutes it into the range where the law works and multiplies back.

What goes wrong in practice. Three faults cover most bad readings. A cloudy extract scatters light, which the instrument reads as absorbance, so the extract must be filtered clear. A coloured impurity that absorbs at the same wavelength adds its own absorbance, so the wavelength is chosen where the target absorbs and the impurity does not. And a reading taken outside the straight range, too dark or too pale, is not trusted: the sample is diluted or concentrated and read again. The law is simple, and the skill is in knowing when its conditions hold.

🤔 Pause and think: two extracts read the same absorbance, but one was diluted tenfold and the other was read neat. Which holds more caffeine, and by how much?

Quick check

A standard of 20 mg per litre reads an absorbance of 0.40. An unknown reads 0.60 at the same wavelength and path length. What is its concentration?

Check your answer

The absorbance is 1.5 times the standard, so the concentration is 1.5 × 20 = 30 mg per litre.

Try it yourself: Brew two cups of tea, one with one spoon of leaf and one with two, for the same time. Hold them against white paper: the double-leaf cup should look about twice as dark, which is the Beer-Lambert law in the cup, absorbance rising with concentration.

2.10 Check yourself: quick self-check

Think about it first: a factory ferments its tea for twice as long as usual, hoping for a stronger tea. The tea comes out dark, flat and low in brightness. Which chemicals explain the result?

  1. What are theaflavins and thearubigins, and which one gives brightness?
Check your answer

Theaflavins are bright orange-red compounds formed when a simple catechin quinone joins a gallacatechin quinone; they give brightness and briskness. Thearubigins are the red-brown group formed from about 75% of the catechins, giving colour, strength and mouth feel.

  1. What theaflavin to thearubigin ratio does the handbook give?
Check your answer

About 1:10 under ideal fermentation conditions, and about 1:10 to 1:12 in general. The ratio varies with leaf, fermentation time and temperature, and measurement, and it is judged by the liquor it gives, not by a fixed figure alone.

  1. Green tea holds 18 to 20% tannins and black tea about 10%. What happened to the tannins?
Check your answer

During fermentation the polyphenols were oxidised into theaflavins and thearubigins, so the tannin level fell from 20% in green leaf to 10 to 12% in fermented tea.

  1. Robusta green coffee holds 1.6 to 2.4% caffeine and arabica 0.9 to 1.2%. What does that mean for a cup of each?
Check your answer

The robusta cup carries about twice the caffeine of the arabica cup from the same weight of beans, and instant coffee powder, at 4.5 to 5.1%, carries more still.

  1. Name two aroma compounds found in orthodox tea that give its flowery character.
Check your answer

Any two of: linalool, linalool oxides, geraniol, methyl salicylate, phenyl ethanol, cis-jasmone, beta-ionone.

Try it yourself: Brew a green tea and a black tea from the same weight of leaf for the same time. Compare the colour, the brightness and the astringency. The green tea is sharper and greener because its catechins are unchanged; the black tea is redder and brisker because its catechins became theaflavins and thearubigins.

Key ideas to carry forward

  • The tea flush holds about 25% flavanols and 4% caffeine on a dry basis; the flavanols are catechins, and tea quality follows their concentration.
  • Fermentation oxidises catechins to quinones: about 15% stay unchanged, about 10% become theaflavins and related compounds, and about 75% become thearubigins, up to 20% of dry weight. Theaflavins give brightness and briskness; thearubigins give colour, strength and mouth feel.
  • The theaflavin to thearubigin ratio is about 1:10 under ideal fermentation conditions, and about 1:10 to 1:12 in general; it varies with leaf, fermentation and measurement.
  • Tea holds 2.5 to 4.0% caffeine; black tea holds about 10% tannins against 18 to 20% in green leaf. More than 638 aroma compounds have been identified in tea.
  • Green tea is unfermented, Oolong is partially fermented, black tea is fully fermented; the books give Oolong no section of its own.
  • Green coffee holds about 1% caffeine, 7% chlorogenic acid and 13% oils; robusta holds about 2.2% caffeine and arabica about 1.2%; roasting cuts chlorogenic acid sharply and leaves caffeine almost unchanged.
  • The Beer-Lambert law, A = ε × c × l, turns a light reading into a concentration, through a calibration line of standards read at the same wavelength.
  • The law only holds on the straight part of the line, so dark extracts are diluted into range and the dilution is multiplied back.

Glossary

Flavanol

A group of polyphenols in tea, also called catechins, that become the colour and strength of black tea on fermentation. Food tech example: a flush holding 25% flavanols making a strong black tea.

Catechin

The main flavanol of tea, including epigallocatechin gallate, the largest single catechin in the fresh leaf. Food tech example: catechins oxidising into theaflavins and thearubigins during fermentation.

Theaflavin

A bright orange-red compound formed during tea fermentation, giving brightness and briskness to the liquor. Food tech example: a black tea with a good theaflavin level showing a bright, brisk cup.

Thearubigin

A red-brown group of compounds formed from about 75% of the tea catechins, giving colour, strength and mouth feel. Food tech example: thearubigins making up to 20% of the dry weight of black tea.

Polyphenol oxidase

The enzyme in tea leaf that oxidises catechins to quinones and drives fermentation. Food tech example: withering bringing polyphenol oxidase to maximal activity before rolling.

Caffeine

An alkaloid that stimulates the central nervous system, giving tea and coffee their lift and tang. Food tech example: robusta coffee at about 2.2% caffeine against arabica at about 1.2%.

Chlorogenic acid

A compound in coffee, second in importance to caffeine, that breaks down sharply on roasting. Food tech example: arabica chlorogenic acid falling from 5.5 to 8.0% to 1.2 to 2.3% on roasting.

Trigonelline

A nitrogen compound in coffee that breaks down on roasting into volatile flavour compounds. Food tech example: trigonelline falling from about 1% in green arabica to 0.5 to 1.0% when roasted.

Tannin

In tea, the oxidised polyphenol complex, mainly thearubigins. Food tech example: black tea at about 10% tannins against green tea leaves at 18 to 20%.

Oolong

A partially fermented tea, between green tea and black tea. Food tech example: an Oolong with a short fermentation, giving a flavour between green and black tea.

Absorbance

The logarithm of the light lost crossing a solution, which rises in proportion to concentration. Food tech example: a tea extract with twice the caffeine reads twice the absorbance at caffeine's wavelength.

Molar absorptivity

The number that says how strongly one mole of a substance absorbs light of one colour, written ε. Food tech example: caffeine and tannin have different absorptivities at different wavelengths, which is how they are told apart.

Calibration line

The straight line of absorbance against concentration drawn from standard solutions of known strength, used to read unknown concentrations. Food tech example: a tea extract's caffeine read off the line drawn from caffeine standards.

References

  • Nagalakshmi, Srikantayya. Tea: An Appraisal of Processing Methods and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ramalakshmi, Kulathooran, and Raghavan, Bashyam. Coffee: A Perspective on Processing and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ranganna, S. Handbook of Analysis and Quality Control for Fruit and Vegetable Products. Tata McGraw-Hill, 2nd edition, 1986.

Unit 3: Manufacturing Technology of Tea and Coffee

Note

Unit 3: Manufacturing Technology of Tea and Coffee

The same tea leaf can become a light, flowery orthodox tea or a strong, malty CTC tea, and the same coffee cherry can become a clean washed parchment coffee or a heavy natural cherry coffee. The factory decides, step by step, and this unit walks through each of those steps.

3.1 Making black tea: the eight steps

Black tea is made from young leaves and unopened buds. The steps are plucking, withering, leaf distortion, fermentation, firing, grading, packing and storage.

Plucking. The fresh leaves are harvested by hand, every 7 to 14 days, taking only the rapidly growing shoot tips down to about the second or third unfolded leaf.

Withering. The leaves are spread in thin layers of 0.3 to 0.7 kg per square metre in trays, troughs or lofts, for 16 to 20 hours. During withering the leaf changes: amino acids, simple carbohydrates and caffeine rise, polyphenol oxidase reaches maximal activity, pectinase activity is lost, and chlorophyll breaks down. Hot air is blown from below the trays. Short withering of 12 hours at a low temperature of 10 to 15 °C gives good flavour; longer withering of 20 to 30 hours at 25 to 30 °C gives good colour but harms the chemical and flavour quality. Trough withering is the most popular method because it is cheap to build and maintain.

Leaf distortion, or rolling. The withered leaf is rolled to wring out the juice and twist the leaf. A roller is a circular table with a cylindrical box and a cap that applies pressure. The leaf is bruised and twisted, broken into small pieces by rising pressure, and sifted. Each roll lasts 15 to 60 minutes, and the number of rolls varies from two to five, depending on the degree of wither, the tea wanted, and the roller charge, speed and temperature. Leaf distortion immediately after plucking is not advised: the leaf must wither first.

Fermentation. The rolled leaf is spread in thin layers 5 to 8 cm deep on the factory floor or on racked trays in a fermentation room. The time varies between 45 minutes and 3 hours, and the temperature between 24 and 27 °C. Low temperature, 15 to 25 °C, improves flavour. At the end the leaf turns from green to coppery red and develops a pleasant aroma. Fermentation is stopped by firing.

Firing. The fermented leaf is fired in a special dryer at 90 to 95 °C for 20 minutes, which cuts the moisture by 3 to 4% and inactivates the enzymes. The hot air turns the coppery red leaf black. Firing loses 70 to 80% of the essential oil, which weakens the aroma, and that loss is one of the known weaknesses of the traditional method.

Grading and storage. The tea is winnowed to remove stalky material and sieved into grades by particle size. Chemical changes continue in storage: the tea loses its residual greenness and harshness within a few weeks. Tea stays sound and full of flavour for more than a year if kept cool and protected from moisture and oxygen.

🤔 Pause and think: withering is the slowest step in black tea making, taking 16 to 20 hours. Why not skip it and roll the fresh leaf straight away?

Quick check

Why does a short, cool withering give better flavour than a long, hot one?

Check your answer

Short withering of 12 hours at 10 to 15 °C keeps the flavour quality. Longer withering of 20 to 30 hours at 25 to 30 °C gives good colour but harms the chemical and flavour quality.

3.2 Orthodox tea against CTC tea

The difference between the two black teas is in the leaf distortion step.

Orthodox tea is rolled in the traditional way described in section 3.1. The leaf is twisted and broken gently. That uneven damage gives a non-uniform rate of oxidation and loses polyphenols. Traditional leaf distortion generally, rolling and CTC, leaves 20 to 25% of the leaf tissues undamaged. But orthodox tea keeps more of its volatile oil and aroma compounds, so it brews light, with fine, flowery and rich aromas.

CTC tea is made with machines that cut, tear and curl the leaf: the CTC machine has two engraved metal rollers running close together, one at 70 rpm and the other at 700 rpm, working like a mangle. The soft withered leaf is cut, torn or rolled in the small gap between the serrated surfaces. Other machines in the same family are the Legg-cut, the Rotorvane, the Triturator, the Tocklai continuous roller, the Barbora leaf conditioner and the continuous rolling expresser. The aim is intensive maceration, so fermentation is rapid and complete. CTC teas give a strong brew and body, more cuppage, and a malty aroma, but they hold fewer volatile aroma compounds than orthodox teas.

Two ways to distort the leaf

The same withered leaf

↓

ORTHODOX: roller, twist and bruise
2 to 5 rolls, 15 to 60 minutes each. The leaf is twisted and bruised gently.

Orthodox cup: light brew, flowery, rich aroma
More volatile oil survives, so the tea brews light with fine, flowery aromas.

CTC: two rollers, 70 rpm and 700 rpm
The leaf is cut, torn and curled in the gap between the serrated rollers. The maceration is intensive, so fermentation is rapid and complete.

CTC cup: strong brew, malty, more cuppage
More cups per kilo, a stronger brew and a malty aroma, but fewer volatile aroma compounds.

Traditional leaf distortion generally, rolling and CTC, leaves 20 to 25% of the tissues undamaged.

Figure 3.1: Orthodox against CTC manufacture. Orthodox and CTC are two parallel, alternative ways of distorting the same withered leaf, not successive steps. The same leaf gives a light, flowery tea by rolling and a strong, malty tea by cutting, tearing and curling. Drawn from the leaf distortion methods in Nagalakshmi, Tea.

Quick check

Why does CTC tea brew stronger than orthodox tea from the same leaf?

Check your answer

CTC machines macerate the leaf intensively, cutting, tearing and curling it, so fermentation is rapid and complete and the liquor extracts fast and strong. Traditional leaf distortion generally, rolling and CTC, leaves 20 to 25% of the tissues undamaged, so orthodox fermentation is uneven.

3.3 Green tea and the other types

Green tea is unfermented. The steps are plucking, steaming or roasting, primary heating and rolling, rolling, secondary rolling, drying, refining, firing, sorting and packing. In the Japanese method the enzyme is inactivated by steaming; in the Chinese method it is inactivated by roasting the leaves in a pan. Green tea is made from a bud and two to three leaf flushes, and there are 18 types, classed as pan-fixed and steam-fixed.

Yellow tea is made from a bud and two or three young leaves by withering, roasting, rolling, sorting and firing. Red tea is made by withering, rolling, roasting, firing, sorting and final firing. White tea is made from buds and one or two leaves withered for 1 to 3 days, then fired, sorted and packed. Dark tea is withered, rolled, piled, dried, steamed and compressed; piling holds the tea warm and damp so microbes grow and oxidise the polyphenols, mellowing the flavour. Green brick tea is made by pressing the processed leaf into bricks under steam and pressure, and it is eaten as a soup cooked with water or milk.

Note on the book's classification: the Handbook classes yellow and red tea under partially fermented Oolong in section 1, but calls yellow tea unfermented and red tea fermented in sections 2.3 to 2.4, and calls dark tea unfermented while describing microbial oxidation in piling. This unit follows sections 2.3 to 2.4, and the tension is the book's own.

Quick check

Why is green tea steamed or pan-roasted before rolling, while black tea is rolled before any heating?

Check your answer

Green tea must not ferment, so its enzymes are killed by heat at the start. Black tea must ferment, so its enzymes are kept alive through withering and rolling and only killed at the firing step.

3.4 Making coffee: cherry coffee and parchment coffee

Coffee is processed by two routes, and the books give both in full.

Cherry coffee, the dry route. The harvested cherries are spread evenly on concrete patios, on trays, or on paved ground, and dried in the sun for 3 to 5 weeks, turned regularly so the lower layers do not ferment, and covered at night. Drying on bare ground is not advised: the cherries pick up dust and earth and take an earthy smell and taste. Drying is complete when the moisture is 11 to 12%, the outer shell is dark brown and brittle, and the bean rattles inside the husk. The dried beans are bagged and stored for several weeks, during which the green beans inside lose a little more moisture and even out. The ratio of dry cherry to green beans is usually 2 to 1.

Parchment coffee, the wet route. The cherries are pulped to remove the skin, and the sticky mucilage around the bean is removed by fermentation, by alkali, by enzymes, or by attrition. Natural fermentation takes about 24 to 36 hours for arabica; robusta has a thicker and more sticky mucilage, and fermentation is often not complete even after 72 hours, so warm water, alkali or attrition is used. Prefermentation of pulped coffee for 12 hours in the case of arabica and 24 to 36 hours in the case of robusta is desirable before use of an aquawasher. Alkali treatment uses about 1 kg of sodium hydroxide dissolved in 10 litres of water for 1,000 to 1,200 litres of wet parchment, trampled for about half an hour until the parchment is no longer slimy and rattles. Pectinolytic enzymes shorten fermentation, and attrition machines rub the mucilage off by friction. The washed parchment may be soaked under water for 12 to 24 hours overnight, which improves the quality of substandard coffees. Cup tests show no difference in cup quality between coffees processed with and without prolonged natural fermentation, but coffee made without prolonged fermentation dries faster.

Drying. The wet parchment, at 50 to 55% water, must be dried to 10% moisture; cherry coffee, from 65% water in the fruit, is dried to 11 to 12%. The book gives 11% for safe storage and about 12% as the drying end-point. In the sun the layer is 2 to 4 inches (5 to 10 cm) thick and raked often; washed coffee dries in about 1 week and natural coffee in about 3 weeks. In machine drying, hot air passes through a bed of coffee, at 60 °C in the first stages and falling slowly. Under-dried parchment turns mouldy and bleached in storage. The beans, at 11 to 12% moisture, are packed in sacks and sent to the curing works.

Quick check

What is the difference between cherry coffee and parchment coffee?

Check your answer

Cherry coffee is made by the dry route: the whole cherry is sun dried for 3 to 5 weeks and then hulled. Parchment coffee is made by the wet route: the cherry is pulped, the mucilage is removed by fermentation or other means, and the bean is dried in its parchment skin.

3.5 Hulling, grading and storage of green coffee

Hulling removes the parchment hulls and silver skin from parchment coffee, and the husk from dried cherry coffee. Polishing takes off any silver skin still clinging to the bean after hulling.

Grading separates the beans by size and density. The beans are passed over screens and through air classifiers, and the lots are labelled by grade. A lot can be labelled as graded only when it meets the grade standard: size, density and freedom from defects are all checked.

Storage. The green beans are packed in sacks and stored in a dry, cool store. Green coffee holds about 10 to 13% water, and if it picks up moisture it goes mouldy. The curing works also bulks the lots, blending beans of the same grade from different estates into one even lot for sale.

Quick check

Why must green coffee be stored dry, and what happens if it picks up moisture?

Check your answer

Green coffee holds about 10 to 13% water. If it picks up more moisture it turns mouldy and bleached in storage, and its quality is lost.

3.6 Roasting and grinding

Roasting is the first step in making any consumable product from green coffee. Green coffee has very little aroma; the characteristic aroma develops only through roasting. The beans are heated to 200 to 250 °C, driving off free and bound moisture. The degree of roast is critical to flavour:

TemperatureRoastWeight lossCup quality
200 °CVery light11-12%Acid taste, less aroma
220 °CLight14-15%Better aroma, less astringent, acidic cup
230 °CMedium16-20%Optimal quality
240 °CDark21-23%Preferred in Europe
250 °CItalian24-25%Dark colour, good taste

Roasting equipment runs from hand-rotated cylinders to continuous roasters. The conventional roaster is a metal container in which the beans are heated while continuously rotated, by conduction from hot metal, convection from hot air, or both. Continuous roasters use a perforated drum or cylinder; fluidised-bed roasters heat and cool in the same vessel and give uniform roasting; pressure roasting is considered more efficient and is thought to increase the acidity of the brew. The beans must be heated quickly and evenly and stirred continuously.

Grinding. Roasted beans cannot be ground at once: they are too soft and would be crushed and flattened. Water quenching cools them first, so they become hard and brittle and can be ground. The grind is set by the brewing method: regular grind, drip grind and fine grind. Fine grind suits steeping methods and vacuum coffee makers. The extraction and keeping quality of the powder depend on the particle size.

Quick check

Why quench roasted coffee beans with water before grinding them?

Check your answer

Freshly roasted beans are too soft and would be crushed and flattened in the grinder. Quenching cools them so they become hard and brittle and grind cleanly.

3.7 Instant tea and instant coffee

Instant tea is made by dehydrating a tea infusion into a dry powder. The raw material can be green leaves after fermentation, black tea, low-grade teas, crude leaves from pruning, or tea waste. The leaf is withered, distorted to a size that suits extraction, and fermented to the right time for body. A rotorvane with a CTC machine is the best choice for distortion, and coarse cutting prevents clogging in the extractors. Extraction uses demineralised or distilled water, in co-current or counter-current systems, building the solids to 7 to 10%. Part of the first extract is kept back and added before spray drying to make up for flavour loss. The extract is concentrated to 30% solids in a vacuum or film evaporator at low temperature and short time, then dried, usually by spray drying in a tall tower. Decreaming removes the cream, the haze that forms when tea cools, which is a caffeine-polyphenol-protein complex. Aroma is stripped with steam or inert gas straight after extraction and added back at the end. The powder is highly hygroscopic, so it is packed at once in triple-laminated aluminium foil, and anti-caking agents are added. Hot-water-soluble instant tea holds tannins 40.50 to 49.00%, total catechins 55.20 to 89.40%, caffeine 3.50 to 5.10%, total amino acids 7.00 to 8.00% and total nitrogen 2.90 to 4.00%.

Instant coffee is made by extracting roasted, ground coffee with hot water, concentrating the extract, and drying it by spray drying or freeze drying. Robusta suits it because of its higher caffeine and lower cost. The powder holds 4.5 to 5.1% caffeine, against about 1.0% in roasted arabica and about 2.0% in roasted robusta, because the soluble solids are concentrated. Coffee-chicory mixtures are also made into instant coffee.

Quick check

Why is part of the first tea extract kept back and added just before spray drying?

Check your answer

Flavour is lost during concentration and drying. Adding back the first extract, which holds the most aroma, makes up for that loss.

3.8 Packaging and storage

Tea is packed at 3% moisture in tea chests with aluminium foil or pliofilm lining, in humidity-controlled rooms. During the 3 to 6 months before it reaches the consumer, it loses flavour and astringency, loses theaflavins, amino acids, sugar and pigments, and gains moisture. High moisture and high temperature speed all of it. Instant tea must be packed at once in triple-laminated foil, and its moisture kept well below 5% to stop caking and darkening.

Coffee is packed as green beans in sacks, as roasted beans or ground coffee in airtight packs, and as instant coffee in sealed jars. Roasted coffee stales fast because its oils oxidise in air, so the pack must keep air out, and grinding is best done near the time of brewing.

Quick check

Why is tea packed at 3% moisture, and what happens if it picks up moisture in storage?

Check your answer

At 3% moisture the tea keeps its flavour and stays free of mould. If it gains moisture, it loses flavour, astringency, theaflavins and pigments, and develops off-taints.

3.9 Practical note: making cherry and parchment coffee

Preparation of cherry and parchment coffee. Take one lot of ripe cherries and dry them whole in the sun, turning them daily and covering them at night, until the moisture is 11 to 12% and the bean rattles in the husk: that is cherry coffee. Take a second lot, pulp it, ferment it overnight to remove the mucilage, wash it clean, and dry it in a thin layer until the moisture is about 10%: that is parchment coffee. Hull both lots, grade them by size, and brew them side by side. The parchment coffee brews cleaner and brighter; the cherry coffee brews heavier and fuller. That difference is the wet against the dry route, tasted in the cup.

Quick check

Why cover sun-drying coffee cherries at night?

Check your answer

Night dampness and dew wet the cherries again, and the lower layers can ferment. Covering them keeps the drying even and stops fermentation.

3.10 The black-tea factory in order: a layout that follows the leaf

Section 3.1 gave the eight steps of black tea making, and section 3.2 compared orthodox and CTC tea. This section puts the steps into a factory: the layout follows the leaf, so each step sits next to the one that feeds it, and the leaf never travels backwards. A factory that is laid out in the order of the process is faster, cleaner, and easier to control than one where the leaf is carried across the building between steps.

Black tea: the factory in order

Withering troughs
Fresh leaf loses moisture in warm air until it is soft and floppy.

↓

Rolling or CTC
Rollers twist and bruise the leaf, or CTC rollers cut, tear and curl it.

↓

Fermenting room
The leaf lies thin in cool, humid air while enzymes build colour and flavour.

↓

Dryer, firing
Hot air stops the ferment and dries the leaf to about 3% moisture.

↓

Sorting and grading
Sieves and winnowers sort the tea into leaf, broken, fannings and dust.

↓

Packing and storage
Graded tea is packed at once in dry, airtight packs, away from damp and strong smells.

Figure 3.2: The black-tea factory in order. Each step feeds the next, and the leaf never travels backwards. Drawn from the account of black tea manufacture in Nagalakshmi, Tea: An Appraisal of Processing Methods and Products.

Why the order cannot be changed. Withering must come first, because a fresh, stiff leaf shatters in the roller instead of twisting. Rolling must come before fermenting, because it is the bruising that lets the enzymes meet the air and the leaf juices. Fermenting must come before firing, because firing kills the enzymes and fixes the colour: fire too early and the tea is green and raw, fire too late and the tea is over-fermented and dull. Sorting must come last, because the leaf changes size and weight at every step, and grading a wet leaf would sort nothing. The control points in Unit 5 sit on this line: withering time and moisture, rolling pressure, fermenting temperature and time, and the firing temperature are each checked where the step happens, not at the end, because a fault made at one step cannot be fixed at the next.

Roasting: what the bean loses and gains. Coffee roasting, in section 3.6, follows the same logic of order, with its own landmarks. The green bean at 10 to 12% moisture is heated, and the water inside it turns to steam: the bean swells and loses weight all the way, mostly water. The roast is judged by the book's temperature and weight-loss table in section 3.6: final bean temperature 200 to 250 °C, with weight loss from 11 to 12% for a very light roast to 24 to 25% for an Italian roast, and cup quality, from acid with little aroma at 200 °C to optimal quality at a medium roast of 230 °C. A roaster who weighs the batch before and after knows the roast by the loss as well as by the colour. The rule for the cup is the same as for tea: the factory can spoil good raw material, but it cannot make fine coffee out of poor cherries, so the roast is there to develop what the cherry already holds, not to hide what it lacks.

🤔 Pause and think: a tea factory fires its leaf before fermenting, by mistake, and the tea comes out green and raw. Which step was skipped in effect, and why can no later step fix it?

Quick check

Why must withering come before rolling in black tea making?

Check your answer

A fresh, stiff leaf shatters in the roller instead of twisting. Withering softens it until it is floppy, so the roller can twist and bruise it without breaking it.

Try it yourself: If you brew tea, watch the leaf after brewing: whole-leaf tea unfurls into recognisable leaves, while CTC tea stays as small grains. The shape tells you which rolling method made it, and the cup tells you the difference: the whole leaf gives a lighter, more fragrant cup, and the CTC gives a stronger, maltier one.

3.11 Check yourself: quick self-check

Think about it first: a tea factory withers its leaf for 30 hours at 28 °C and fires it hard. The tea is dark and strong but has lost its flowery aroma. Which two steps explain the loss, and what should change?

  1. Name the eight steps of black tea manufacture in order.
Check your answer

Plucking, withering, leaf distortion or rolling, fermentation, firing, grading, packing and storage.

  1. What is the difference between orthodox and CTC tea?
Check your answer

Orthodox tea is rolled gently and brews light and flowery; traditional leaf distortion generally, rolling and CTC, leaves 20 to 25% of the tissues undamaged. CTC tea is cut, torn and curled between rollers at 70 and 700 rpm, ferments fast and completely, and brews strong and malty.

  1. How long does natural fermentation take for arabica and for robusta parchment coffee?
Check your answer

About 24 to 36 hours for arabica. Robusta has a thicker and more sticky mucilage, and fermentation is often not complete even after 72 hours. Prefermentation of 12 hours for arabica and 24 to 36 hours for robusta is desirable before use of an aquawasher, and that is prefermentation, not natural fermentation.

  1. A medium roast at 230 °C loses 16 to 20% of its weight and gives optimal quality. What does a very light roast at 200 °C taste like?
Check your answer

Acid taste with less aroma, losing only 11 to 12% of its weight.

  1. Why is instant tea powder packed at once in triple-laminated foil?
Check your answer

The powder is highly hygroscopic: it picks up moisture from the air, cakes, darkens and loses flavour. Foil packing at once, with moisture well below 5%, stops that.

Try it yourself: Brew a cup of tea from loose leaf and a cup from a tea bag of the same type, using the same weight and time. Compare the strength and the clarity. The bag holds smaller leaf, fannings or dust, which extracts faster and stronger: that is grading, tasted in the cup.

Key ideas to carry forward

  • Black tea is made by plucking, withering 16 to 20 hours, rolling in 2 to 5 rolls of 15 to 60 minutes, fermenting 45 minutes to 3 hours at 24 to 27 °C, firing at 90 to 95 °C for 20 minutes, and grading, packing and storage.
  • Orthodox tea is rolled gently and brews light and flowery; CTC tea is cut, torn and curled at 70 and 700 rpm and brews strong and malty.
  • Green tea is unfermented because its enzymes are killed by steaming or pan roasting at the start.
  • Cherry coffee is sun dried whole for 3 to 5 weeks to 11 to 12% moisture, at a ratio of 2 parts dry cherry to 1 part green bean; parchment coffee is pulped, fermented naturally for about 24 to 36 hours for arabica (robusta often not complete even after 72 hours), washed and dried to 10% moisture.
  • Roasting runs from 200 °C, very light, to 250 °C, Italian, with weight loss rising from 11 to 12% to 24 to 25%; medium roast at 230 °C gives optimal quality.
  • Instant tea is extracted to 7 to 10% solids, concentrated to 30%, and spray dried; instant coffee is extracted, concentrated and dried from roasted ground coffee.
  • The tea factory is laid out in the order of the process, withering, rolling, fermenting, firing, sorting, packing, and each step's control point sits where the step happens.
  • Coffee roasting is judged by the book's temperature and weight-loss table: 200 to 250 °C, with weight loss from 11 to 12% for a very light roast to 24 to 25% for an Italian roast, and a medium roast at 230 °C giving optimal quality.

Glossary

Withering

Spreading plucked tea leaves in thin layers for 16 to 20 hours so they lose moisture and their chemistry changes to suit rolling and fermentation; the first step in black tea making, in which fresh leaf loses moisture in warm air until it is soft and floppy enough to roll. Food tech example: withering at 0.3 to 0.7 kg per square metre in troughs, with leaf withered until it bends without breaking.

Orthodox tea

Black tea made by rolling the withered leaf gently, giving a light, flowery brew. Food tech example: orthodox leaf tea with whole, twisted leaves.

CTC tea

Black tea made by crushing, tearing and curling the leaf between rollers at 70 and 700 rpm, giving a strong, malty brew. Food tech example: CTC granules that brew fast and strong.

Firing

Heating fermented tea leaf at 90 to 95 °C for 20 minutes to stop fermentation, cut moisture by 3 to 4% and turn the leaf black. Food tech example: firing the coppery red leaf into black tea.

Mucilage

The sticky layer around the coffee bean inside the cherry, removed by fermentation, alkali, enzymes or attrition. Food tech example: prefermenting pulped arabica for 12 hours before an aquawasher to remove its mucilage.

Hulling

Removing the parchment hulls and silver skin from parchment coffee, or the husk from dried cherry coffee. Food tech example: hulling dried cherry coffee to green beans.

Roasting

Heating green coffee beans to 200 to 250 °C to develop their aroma and flavour. Food tech example: a medium roast at 230 °C giving optimal cup quality.

Quenching

Cooling roasted coffee beans with water so they become hard and brittle enough to grind. Food tech example: quenching beans before grinding them for drip coffee.

Decreaming

Chilling tea extract to remove the cream, the haze of caffeine, polyphenol and protein that forms as tea cools. Food tech example: decreaming an extract before making cold-soluble instant tea.

Curing works

The works where dried coffee is hulled, graded, bulked, packed and stored before sale. Food tech example: parchment coffee sent to the curing works at 12% moisture.

References

  • Nagalakshmi, Srikantayya. Tea: An Appraisal of Processing Methods and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ramalakshmi, Kulathooran, and Raghavan, Bashyam. Coffee: A Perspective on Processing and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.

Unit 4: Quality Evaluation of Tea and Coffee

Note

Unit 4: Quality Evaluation of Tea and Coffee

A tea buyer can look at a handful of dry leaf and tell its grade, smell it and tell its freshness, and taste the brew and tell its strength, briskness and faults, all in five minutes. No instrument can do all of that. This unit is about the cup test, the laboratory tests that back it up, and the tricks used to cheat both.

4.1 What quality means for tea and coffee

Quality in tea and coffee has four parts, and all four are checked:

  • Physical quality. Grade, particle size, bulk density, colour and appearance of the dry leaf or bean, and freedom from stalks, fibre and foreign matter.
  • Chemical quality. Polyphenols and tannins, total solubles, caffeine, pigments, sugars, minerals and lipids, and the limits for ash, water solubles and fibre set by standards bodies.
  • Sensory quality. What the taster finds in the cup: colour, strength, briskness, flavour and faults.
  • Microbiological quality. Important in liquid tea products such as carbonated beverages and canned or frozen tea concentrates.

No single test covers all four. Tasting is more economical than chemical analysis, but the two are complementary: the analytical results are useful when tasting reports contradict each other, and they help to correct processing faults. The chemical composition of the liquor relates to the cup like this: colour follows theaflavins and thearubigins; strength follows total solids, theaflavins and thearubigins; briskness follows theaflavins; and quality follows caffeine, theaflavins and thearubigins.

🤔 Pause and think: a laboratory report says a tea is excellent, but the taster says it is dull and flat. Which one is right, and why would they disagree?

Quick check

Why are tasting and chemical analysis used together instead of one replacing the other?

Check your answer

Tasting is cheaper and catches aroma and appearance, which cannot be measured chemically. Analysis settles disputes between tasters and helps correct processing faults. The two are complementary.

4.2 The cup test: tea tasting step by step

The cup test, or tea tasting, is a fixed ritual, so that one tea can be compared fairly with another. The books give the method in full:

  1. Weigh the tea. Place 2.83 g of the graded or blended tea sample in a mug of 0.142 litre capacity, in earthenware, china or plastic, preferably earthenware.
  2. Pour boiling water into the mug.
  3. Wait 5 to 6 minutes. The brew reaches 2% solids in that time.
  4. Judge the liquor. The expert evaluates colour (depth and brightness), quality (good or coarse), strength (strong or weak), briskness (pungent or soft), flavour (fine or poor), manufacturing faults, taints, and cream (bright or dull).
  5. Judge the infused leaf. Its colour and appearance (bright copper, bright greenish, mixed, dull or dark) and the smell of the wet, hot leaf.
  6. Judge the dry leaf. Its grade (uniformity of size and form), colour (black or greyish), make and style (general appearance as desirable for the grade), aroma (smell, warmth and dampness), and feel (spongy or lacking density).

The taster's terms for dry leaf, infused leaf, liquor and defects are defined in a glossary used by producers, buyers, blenders and brokers. Aroma and appearance, which cannot be determined chemically, play a major role in tea quality and price. To meet changing market needs, and the variation in quality from processing, season and region, sensory assessment by expert tasters is essential for routine evaluation of commercial samples and estate varieties.

The cup test for tea

Weigh the tea
2.83 g tea in a 0.142 L mug

↓

Pour boiling water
Pour boiling water into the mug

↓

Wait 5 to 6 minutes
The brew reaches 2% solids in that time

↓

Judge the liquor
Colour, strength, briskness, flavour, taints, cream

↓

Judge the infused leaf
Colour, appearance, smell

↓

Judge the dry leaf
Grade, colour, make, aroma, feel

Figure 4.1: The cup test for tea. The fixed weight, volume and time make every tea comparable. Drawn from the tea tasting method in Nagalakshmi, Tea.

Quick check

Why must the cup test use exactly 2.83 g of tea and 5 to 6 minutes of brewing?

Check your answer

The fixed weight, volume and time make every tea comparable. Changing any of them changes the strength of the brew, so two teas could not be judged fairly against each other.

4.3 Cup testing coffee

Coffee is cupped in the same spirit as tea, though the books give no fixed protocol for it as they do for tea. The coffee chapter judges cup quality by the roast and by the taster: the roast table in Unit 3 links temperature and weight loss to cup quality, from a very light roast that tastes acid with little aroma to a medium roast of optimal quality and a dark roast preferred in Europe. The taster judges the brew for aroma, acidity, body, flavour and faults, and for the cleanness of the cup.

The practical lesson is the same as for tea: use the same weight of coffee, the same grind, the same water temperature and the same brewing time for every sample, or the comparison means nothing. A fine grind suits steeping methods and vacuum makers, and the grind must match the brewer, or the cup will be over-extracted and bitter, or under-extracted and thin.

Quick check

Why must the grind match the brewing method in a coffee cup test?

Check your answer

The grind sets how fast the water extracts the coffee. A fine grind in a slow brewer over-extracts and turns bitter; a coarse grind in a fast brewer under-extracts and tastes thin.

4.4 Chemical tests that back up the cup

Theaflavins and thearubigins. They are extracted into ethyl acetate from the water layer, and separated further by shaking with sodium bicarbonate solution. Their optical densities are measured at 380 nm for theaflavins and 460 nm for thearubigins. Their balance is used as an index of liquor strength and of efficient fermentation. As in Unit 2, the theaflavin to thearubigin (TF:TR) ratio is about 1:10 under ideal fermentation conditions, and about 1:10 to 1:12 in general, varying with leaf, fermentation time and temperature, and measurement, and it is judged by the liquor it gives, not by a fixed figure alone.

Tannins. They are estimated by Lowenthal's method: extract into water, precipitate with gelatin, and titrate with standard permanganate solution before and after precipitation. The difference gives the oxidisable tannins.

Caffeine. In tea it is estimated by the Baily-Andrew method: digest the leaves with heavy magnesium oxide, extract the caffeine into chloroform, and measure the nitrogen in the crude caffeine by the micro-Kjeldahl method, multiplied by a conversion factor. In coffee the preferred method is high-performance liquid chromatography, with detection by ultraviolet absorption at 272 nm. A brew test for tea uses 2 g of tea in 177 ml of water for 6 minutes, estimating the caffeine that dissolves into the cup.

Ash, extract and fibre. The standard limits for tea, as a percentage by mass, are: total ash 4.0 to 8.0%; water-soluble ash, at least 40.0% of total ash; acid-insoluble ash, at most 1.0%; alkalinity of water-soluble ash as K₂O, 1.0 to 2.2%; water extract, at least 32.0%; and crude fibre, at most 16.5%. High acid-insoluble ash points to sand and dirt; low water extract points to spent leaf.

Quick check

A tea tests low in water extract and high in acid-insoluble ash. What does that suggest?

Check your answer

Low water extract suggests spent leaf that has already been brewed, and high acid-insoluble ash suggests sand and dirt. Both point to adulteration or poor handling.

4.5 Adulteration of tea and how it is detected

Tea is adulterated in the ways the books list, and each has a test:

  • Spent tea. Tea that has already been brewed, dried and sold again, alone or mixed with fresh tea. It is detected by its low water extract and flat cup: it gives little colour and no strength.
  • Spent tea with colour. Spent tea mixed with coal-tar dyes, catechu, caramel or curcuma after drying, to restore its colour. The dyes are detected by chemical tests on the leaf and by the unnatural colour of the liquor.
  • Blackgram husk and sawdust. Mixed in after colouring, to add weight. They are detected by inspection, by the microscope, and by the ash and fibre figures.
  • Foreign leaves. Astringent leaves of other plants, dried and mixed in. They are detected by the taster, who finds the wrong flavour, and by leaf inspection.
  • Metal contamination. Copper, lead and iron from the equipment used in processing. They are detected by chemical analysis, and their limits are fixed in several countries.

The general defence is the combination of tests in this unit: the dry leaf is inspected, the liquor is tasted, and the ash, extract and fibre are measured. Any one test can be fooled; the three together are hard to fool.

Quick check

Why is spent tea the hardest adulterant to spot by looking at the dry leaf alone?

Check your answer

Dried spent tea looks like fresh tea, especially when it is recoloured with dyes. Only the cup, which is flat and weak, and the water extract, which is low, give it away.

4.6 Adulteration of coffee and how it is detected

Coffee is adulterated with roasted substitutes that look and grind like coffee: chicory, barley, malt and rye are the substitutes named in the books, and coffee-chicory mixtures are also sold openly as a product in their own right. The difference between a mixture and an adulteration is the label: a mixture declared on the pack is a product; the same mixture sold as pure coffee is a fraud.

Detection uses the same three lines of defence as tea. The dry powder is inspected and tested under the microscope: chicory and cereal starches look different from coffee under magnification, and starch can be tested with iodine. The brew is tasted: chicory adds a bitter, burnt note and a heavy body that pure coffee does not have. The chemistry is checked: caffeine and chlorogenic acid levels fall when the coffee is diluted with substitutes, and the ash and extract figures shift. Green coffee is also checked for moisture, which should be 10 to 13%, and for extraneous matter such as stones, nails and wood splinters, which are removed by screens, air blasting and magnetic separators before roasting.

Quick check

A coffee-chicory mixture is sold in one shop as a blend and in another as pure coffee. Which sale is adulteration?

Check your answer

The second one. A declared mixture is a product in its own right; the same mixture sold as pure coffee is fraud, because the buyer is paying coffee prices for a substitute.

4.7 Practical note: cup testing tea and coffee

Cup testing of tea and coffee. Carry out the tea cup test of section 4.2 with two teas of different grades, using 2.83 g in a 0.142 litre mug, boiling water, and 5 to 6 minutes. Record the liquor colour, strength, briskness, flavour, cream, the infused leaf and the dry leaf for each. Then cup two coffees of different roasts with the same weight, grind and brewing time, and record aroma, acidity, body and flavour. The point of the exercise is the discipline: the same weight, the same water, the same time, every time, so the differences you taste are in the tea and coffee, not in your method.

Quick check

Why record the infused leaf as well as the liquor in a tea cup test?

Check your answer

The infused leaf shows the quality of the leaf itself: its colour, whether it is bright copper or dull and dark, and its smell. A good liquor from a poor leaf is a warning, and a poor liquor from a good leaf points to a processing fault.

4.8 Making the cup test fair: replication, blinding and scoring

Sections 4.2 and 4.3 gave the steps of cup testing tea and coffee. This section answers the harder question: how does the taster know the result is true, and not a trick of the day, the order, or the taster's own expectations? A cup test is a measurement, and like any measurement it needs controls, or it measures the taster more than the tea.

Replication: taste it twice. A single cup can mislead: the water may have been a little cooler, the leaf a little short, or the taster's palate tired from the last sample. The guard is replication: the same tea is brewed in two or more cups, and the taster judges each on its own. If the two cups agree, the result is trusted; if they disagree, the test is repeated, because something in the brewing, not the tea, went wrong. The same rule applies across days: a tea that is praised on Monday and faulted on Friday has probably been tasted differently, not made differently.

Blinding: hide the name. A taster who knows the tea is the expensive lot, or the new supplier's sample, will taste it differently, however honest the intention. The guard is blinding: the cups are coded, so the taster judges the liquor, not the label. The code is broken only after the scores are written down. Blinding also stops the order effect: a strong tea tasted first dulls the palate for the mild tea that follows, so the order of the cups is changed between rounds, and the taster rinses and rests between samples.

Scoring: write numbers, not adjectives. A taster's notes say "bright" or "flat", but a buyer needs numbers that can be compared across lots and across weeks. The guard is a scoring sheet: each attribute is scored on its own scale, and the scores are added. A simple sheet for tea scores five attributes out of 10: colour of the liquor, strength, briskness, flavour, and the infused leaf. A tea that scores 8, 7, 8, 7 and 8 totals 38 out of 50. The number does not replace the taster's judgement: it records it in a form that can be filed, compared, and argued about at the buying table.

Worked example: reading a score sheet. Two teas are tasted blind, each in duplicate. Tea A scores 38 and 37 out of 50; Tea B scores 33 and 34. The duplicates agree within one point, so the test is trusted, and the gap between the teas, about 4 points, is real. If Tea A had scored 38 and 30, the test would not be trusted at all, and the tasting would be repeated: a gap of 8 points between two cups of the same tea says the brewing or the taster failed, not the tea. That is the whole discipline in one line: the duplicate tells you whether to believe the score, and the score tells you whether to buy the tea.

🤔 Pause and think: a buyer tastes a new supplier's tea knowing it is the new supplier's, scores it high, and buys it. The next week, tasted blind among the old lots, it scores lowest. What went wrong in the first tasting, and which guard would have caught it?

Quick check

Why are the cups coded in a cup test, and why is each tea brewed twice?

Check your answer

Coding blinds the taster to the name, so the liquor is judged, not the label. Brewing twice checks the result: if the two cups agree, the score is trusted, and if they disagree, the test is repeated.

Try it yourself: Brew the same tea twice, in two identical cups, and score each cup on colour, strength, briskness and flavour out of 10, without looking at the other cup's scores. If your two scores agree within a point or two, your tasting is steady; if they differ widely, the brewing varied, and you have found why replication matters.

4.9 Check yourself: quick self-check

Think about it first: a merchant offers a tea at half the usual price. The dry leaf looks black and even, the liquor is dark, but the cup is flat and the infused leaf is dull. Which tests would confirm your suspicion, and what would they show?

  1. In the tea cup test, how much tea, how much water, and how long?
Check your answer

2.83 g of tea in a mug of 0.142 litre, boiling water, 5 to 6 minutes, giving a brew of 2% solids.

  1. What three things does the taster judge in the cup test?
Check your answer

The liquor (colour, strength, briskness, flavour, taints, cream), the infused leaf (colour, appearance, smell), and the dry leaf (grade, colour, make, aroma, feel).

  1. What theaflavin to thearubigin ratio does the handbook give?
Check your answer

About 1:10 under ideal fermentation conditions, and about 1:10 to 1:12 in general. The ratio varies with leaf, fermentation time and temperature, and measurement, and it is judged by the liquor it gives, not by a fixed figure alone.

  1. Name three adulterants of tea and one test that catches each.
Check your answer

Spent tea, caught by low water extract and a flat cup; coloured spent tea, caught by chemical tests for dyes; blackgram husk or sawdust, caught by inspection, microscope, ash and fibre figures; foreign leaves, caught by the taster and leaf inspection.

  1. Why is a declared coffee-chicory mixture not adulteration, while the same mixture sold as pure coffee is?
Check your answer

A declared mixture is labelled honestly, so the buyer knows what is in the pack. Sold as pure coffee, the buyer is deceived and pays coffee prices for a substitute.

Try it yourself: Brew the same tea twice, once for 2 minutes and once for 10 minutes, and taste both. Then brew it once with half the leaf and once with double. The differences you taste are the reason the cup test fixes the weight and the time: without them, no two cups can be compared.

Key ideas to carry forward

  • Tea and coffee quality has four parts: physical, chemical, sensory and microbiological, and tasting and analysis are complementary.
  • The tea cup test uses 2.83 g of tea in a 0.142 litre mug, boiling water, 5 to 6 minutes, giving a brew of 2% solids, and judges liquor, infused leaf and dry leaf.
  • Coffee cup quality follows the roast and the taster, and the grind must match the brewing method.
  • Theaflavins are measured at 380 nm and thearubigins at 460 nm after ethyl acetate extraction; tannins by Lowenthal's method; caffeine by the Baily-Andrew method or by HPLC at 272 nm.
  • Tea limits: total ash 4.0 to 8.0%, water-soluble ash at least 40.0% of total ash, acid-insoluble ash at most 1.0%, water extract at least 32.0%, crude fibre at most 16.5%.
  • Tea is adulterated with spent tea, dyes, catechu, caramel, curcuma, blackgram husk, sawdust and foreign leaves; coffee with chicory, barley, malt and rye. The defence is inspection, tasting and chemistry together.
  • A cup test is a measurement: replication checks the result, blinding stops the label and the order from shaping it, and scoring records judgement as numbers that can be compared.
  • The duplicate tells you whether to believe the score: cups of the same tea that disagree say the test failed, not the tea.

Glossary

Cup test

The fixed tasting ritual for tea: 2.83 g in a 0.142 litre mug, boiling water, 5 to 6 minutes, judging liquor, infused leaf and dry leaf. Food tech example: a buyer cupping two grades side by side with the same weight and time.

Briskness

The pungent, lively quality of a tea liquor, linked to theaflavins. Food tech example: a brisk tea that feels sharp and clean on the palate.

Cream

The haze that forms as tea cools, from a caffeine, polyphenol and protein complex; judged bright or dull in the cup test. Food tech example: a bright cream showing a sound, strong liquor.

Taster

An expert who judges tea and coffee by the cup, using defined terms for leaf, liquor and defects. Food tech example: a taster blending 20 to 40 grades into a retail tea.

Spent tea

Tea that has already been brewed, dried and sold again, alone or mixed with fresh tea. Food tech example: spent tea detected by its low water extract and flat cup.

Lowenthal's method

The method for estimating tannins: extract, precipitate with gelatin, and titrate with permanganate before and after precipitation. Food tech example: tannins in black tea estimated at about 10%.

Baily-Andrew method

The method for estimating caffeine in tea: digest with magnesium oxide, extract into chloroform, and measure nitrogen by micro-Kjeldahl. Food tech example: caffeine in a tea sample reported as a percentage.

Water extract

The soluble matter extracted from tea by water, at least 32.0% by the standard. Food tech example: a low water extract pointing to spent leaf.

Acid-insoluble ash

The ash that does not dissolve in acid, mainly sand and dirt, at most 1.0% in tea. Food tech example: high acid-insoluble ash showing a tea contaminated with dirt.

Chicory

A roasted root used as a coffee substitute and extender. Food tech example: a declared coffee-chicory mixture, and the same mixture sold as pure coffee, which is adulteration.

Blinding

Hiding the identity of a sample from the taster, by coding the cups, so the liquor is judged and not the label. Food tech example: a new supplier's tea tasted blind among the old lots.

Replication

Repeating a test on the same sample, so agreement between repeats shows the result is steady. Food tech example: the same tea brewed in two cups and scored separately.

Scoring sheet

A sheet that scores each quality attribute of a tea or coffee on its own scale, so judgement is recorded as numbers that can be compared. Food tech example: colour, strength, briskness, flavour and infused leaf scored out of 10 each.

References

  • Nagalakshmi, Srikantayya. Tea: An Appraisal of Processing Methods and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ramalakshmi, Kulathooran, and Raghavan, Bashyam. Coffee: A Perspective on Processing and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ranganna, S. Handbook of Analysis and Quality Control for Fruit and Vegetable Products. Tata McGraw-Hill, 2nd edition, 1986.

Unit 5: Process Control and Standardisation

Note

Unit 5: Process Control and Standardisation

A tea factory makes the same tea every day for a year, and a coffee roaster sells the same blend every week. The leaf changes with the season, the cherry changes with the rain, and the machines wear. What keeps the product the same is not luck: it is control, measured at fixed points, written down, and acted on. This unit is about those points.

5.1 What process control is

Process control means holding a manufacturing process inside its set limits, by measuring it at fixed points and correcting it before the product goes wrong. Standardisation means making every batch meet the same standard, so the buyer gets the same product every time.

The outline for this unit gives no list of sub-topics, so this unit is built from the control points named across the books for tea, coffee and spices: the temperatures, times, moistures and tests that appear in Units 1 to 4 and in Unit 6. Each control point in this unit is one of those points, gathered here so they can be run as a system.

The system has four parts: set the standard for each step, measure at the control point, compare the result with the standard, and correct the process before the next batch. A measurement that is written down and never read is not control. A correction made after the batch is sold is not control either.

🤔 Pause and think: a factory measures the moisture of every batch of tea but files the readings without looking at them. Is that process control? What is missing?

Quick check

What are the four parts of process control?

Check your answer

Set the standard for each step, measure at the control point, compare the result with the standard, and correct the process before the next batch.

5.2 Control points in tea manufacture

Every step of black tea manufacture in Unit 3 has a control point, a measurement, and a limit:

StepWhat is controlledStandard from the books
PluckingShoot standardTips down to the second or third unfolded leaf, every 7 to 14 days
WitheringTime, temperature, layer16 to 20 hours; 0.3 to 0.7 kg per square metre; short, cool withering for flavour
RollingRolls, time2 to 5 rolls of 15 to 60 minutes each
FermentationTime, temperature, depth45 minutes to 3 hours; 24 to 27 °C; layers 5 to 8 cm deep
FiringTemperature, time90 to 95 °C for 20 minutes; moisture cut by 3 to 4%
GradingParticle sizeSieved into leaf, brokens, fannings and dust
PackingMoisture3% moisture in foil or pliofilm-lined chests
StorageTemperature, humidityCool, dry, protected from moisture and oxygen; keeps over 1 year

The most sensitive points are withering and fermentation. Withering decides the chemistry the factory has to work with: if the leaf is under-withered it is too wet to roll, and if it is over-withered it is too dry to ferment evenly. Fermentation decides the liquor: too short and the tea is green and harsh, too long and the theaflavins fall and the tea goes dark and flat. The factory controls both by clock, thermometer and the maker's eye on the colour of the leaf.

Quick check

Why is fermentation the most sensitive control point in tea manufacture?

Check your answer

It decides the liquor: too short and the tea stays green and harsh; too long and the theaflavins fall and the tea goes dark and flat. The window is 45 minutes to 3 hours at 24 to 27 °C, and it must be judged batch by batch.

5.3 Control points in coffee processing

Coffee has its own control points, from the cherry to the cup:

StepWhat is controlledStandard from the books
Picking to processingTimeWithin 36 hours of picking
Wet processFermentation timeNatural fermentation about 24 to 36 hours for arabica; robusta often not complete even after 72 hours. Prefermentation 12 hours for arabica and 24 to 36 hours for robusta before an aquawasher
DryingMoistureParchment to 10%; cherry to 11 to 12%; sun layer 2 to 4 inches
Machine dryingAir temperature60 °C in the first stages, falling slowly
Green coffeeMoisture10 to 13% water; measured by air oven or dielectric meter
RoastingTemperature, weight loss200 to 250 °C; 11 to 12% loss very light, 16 to 20% medium, 24 to 25% Italian
GrindingParticle size, quenchingQuench before grinding; regular, drip or fine grind to suit the brewer
PackingAir, moistureAirtight packs; roasted coffee stales fast in air

Moisture measurement deserves a closer look, because it is the control that fails most often. Green coffee is measured by the air oven method: drying for 6 hours, resting 15 hours in a desiccator, then drying again for 4 hours. Roasted and soluble coffees hold only about 2 to 5% water, so the air oven method is not accurate enough for them, and the Karl Fischer method is used instead, extracting the water with methanol first. Moisture meters that work on the dielectric constant are used in the field and in the curing works, but they must be calibrated against the air oven reference method, or their readings drift.

Quick check

Why must a dielectric moisture meter be calibrated against the air oven method?

Check your answer

The meter reads an electrical property, not water itself, so its readings drift unless they are checked against the reference method. Once calibrated, it is fast and handy in the field and the curing works.

5.4 Standards: the numbers a product must meet

A standard is a written set of limits that a product must meet to be sold under its name. The books name standards bodies and limits for tea, and the same idea runs through every product in this subject:

  • Tea. Total ash 4.0 to 8.0%; water-soluble ash at least 40.0% of total ash; acid-insoluble ash at most 1.0%; alkalinity of water-soluble ash as K₂O 1.0 to 2.2%; water extract at least 32.0%; crude fibre at most 16.5%. These limits are set to protect the consumer, and they catch spent leaf, sand and dirt.
  • Coffee. Green coffee holds 10 to 13% moisture, and roasted coffee holds 2 to 5% moisture. Roasting runs from 200 to 250 °C, with weight loss from 11 to 12% for a very light roast to 24 to 25% for an Italian roast.
  • Spices. Dried spices hold about 10% moisture on a wet basis, low enough to stop microbes growing. Irradiation decontaminates herbs, spices and food ingredients at high doses of 10 to 50 kGy, and irradiation of any food up to a total dose of 10 kGy is judged to present no toxicological danger.
  • Instant tea. Instant tea powder is highly hygroscopic, so it is packed at once in triple-laminated foil, and its moisture is kept well below 5% to stop caking and darkening.

Standardisation also covers the pack: tea at 3% moisture in lined chests, instant tea with moisture well below 5%, green coffee at 10 to 13% moisture in sacks, roasted coffee in airtight packs, and dried spices packed at once in closed packs away from air, light and damp. A product that meets the recipe but fails the pack standard will fail on the shelf.

🤔 Pause and think: a tea meets its ash and extract limits, but the chests are packed damp and left unsealed overnight. The tea meets the standard and still goes mouldy. Which standard failed?

Quick check

A tea meets its ash and extract limits but goes mouldy on the shelf. What else could have failed?

Check your answer

The pack standard: the tea must be packed at 3% moisture in lined chests, sealed at once, and stored cool and dry. A correct recipe in a bad pack still fails, because mould grows where air and moisture meet.

5.5 Sampling, testing and records

A standard is useless without sampling. The rules for sampling in the books are:

  • Sample from the lot, not from the top. Take from different parts of the batch, because a lot varies from one end to the other.
  • Sample every batch. A test on last week's batch says nothing about this week's.
  • Test what matters. For tea: moisture, grade, ash, extract, fibre, caffeine and tannin, and the cup test. For coffee: moisture, grade, roast and cup. For spices: moisture at about 10%, irradiation dose, volatile oil, and the test for added starch in spice powders.
  • Write it down. The record is the memory of the factory: it shows drift before the drift becomes a failure, and it proves what was done if a buyer complains.
  • Hold the batch until the test is passed. A batch that is shipped before its test comes back cannot be called back cheaply.

In tea and coffee, the hold is built into the routine: a batch is held until its moisture, grade and cup test pass, so a faulty batch shows itself before packing and sale.

Quick check

Why sample from different parts of a lot instead of taking one scoop from the top?

Check your answer

A lot varies from one end to the other: the top may be drier, cooler or cleaner than the bottom. One scoop from the top can miss the fault that sits in the rest of the lot.

5.6 Blending as standardisation

Blending is standardisation by mixing. Tea buyers blend 20 to 40 grades to hold a steady flavour, strength and price, and the blending is done only by expert tasters, because no grade has consistent quality across seasons and factories. Instant tea is used to fortify regular tea: adding instant tea raises the extractives, including polyphenols and caffeine, from 35 to 40% up to 60 to 70%, and adding 15% instant tea to a regular tea does not affect its taste or flavour.

The same idea runs through coffee: the curing works bulks beans of the same grade from different estates into one even lot for sale. In every case the standard is a number, and the blend is the way of hitting it.

Quick check

Why is blending done by expert tasters rather than by a formula?

Check your answer

No tea grade has consistent quality, because quality varies with processing and season. Only a taster can judge the flavour, strength and body of each grade and mix them to a steady result.

5.7 The factory visit: what to look for

A visit to a tea or coffee processing industry is a practical in this subject, and it is the best way to see process control working. Go with a checklist drawn from this unit:

  1. At the intake, is the raw material graded and tested, or taken on trust?
  2. At each step, is there a clock, a thermometer and a written standard?
  3. Is moisture measured, and is the meter calibrated?
  4. Is there a laboratory, and are its results written down and acted on?
  5. Are batches held until their tests pass?
  6. Is the pack checked for moisture, air and sealing?
  7. Is the store cool, dry and clean, and is stock used in order?

Write a report on what you saw, step by step, with the numbers you measured. The report should say what the factory controls well and where its control is weak, and it should give the evidence for each judgement.

Quick check

Why visit a factory with a checklist instead of just looking around?

Check your answer

A checklist makes sure every control point is checked and nothing is missed, and it turns looking into evidence: a written record of what was measured and what was found.

5.8 Control charts: deriving the limits that catch drift early

Section 5.5 covered sampling, testing and records. This section turns the records into a tool: the control chart. A factory that only checks each batch against a specification finds trouble after it has happened. A factory that charts its results finds trouble while it is still small, because the chart shows the drift before the limit is crossed.

Step 1: the centre line. Take a measurement that is made on every batch, such as the moisture of withered leaf, and collect the results over a period when the process is running well. The average of those results is the centre line of the chart: it is where the process sits when nothing is wrong. The centre line comes from the factory's own good running, not from a book, so it fits the factory's leaf, its weather and its machines.

Step 2: the spread. The results are never all equal: they scatter around the centre line. The scatter is measured by the standard deviation, which is roughly the average distance of the results from the centre line. A process with a small standard deviation is steady; a process with a large one is jumpy, even if its average is right.

Step 3: the control limits. The upper control limit is the centre line plus three standard deviations, and the lower control limit is the centre line minus three standard deviations. Three is chosen because, for results that scatter in the usual bell-shaped way, almost all good results, about 99.7 in 100, fall inside three standard deviations. A result outside the limits is therefore a signal, not a coincidence: something has changed, and the process is stopped and checked. A run of results creeping up toward the limit is a signal too: the process is drifting, and it is fixed before it crosses.

Worked example: withering moisture. Ten batches of withered leaf, in a period of good running, read 68, 70, 69, 71, 70, 68, 69, 70, 71 and 69 percent moisture. The average is 69.5 percent, and the standard deviation is about 1 percent. The centre line is 69.5, the upper control limit is 69.5 + 3 = 72.5, and the lower limit is 69.5 − 3 = 66.5. A batch that reads 74 percent is outside the limits: the withering was too short or the air too damp, and the batch is held. A batch that reads 72 percent is inside the limits but high, and three such batches in a row would say the withering is drifting long or damp, and the troughs are checked before a batch fails.

Why more samples help. Section 5.5 said samples must be taken properly. The chart shows why the number of samples matters: the average of several samples scatters less than single samples do, by the square root of the number of samples. Averaging four samples halves the scatter, so the control limits tighten and smaller drifts are caught. That is the reason a factory tests several samples per batch instead of one: one sample can mislead, and the average of four rarely does.

🤔 Pause and think: a factory's withering moisture reads 71, 71.5, 72 and 72.5 percent across four batches, all inside the control limits. The specification allows up to 75 percent. Should the factory act, or wait for a batch to fail?

Quick check

A process has a centre line of 50 and a standard deviation of 2. What are the control limits, and what does a result of 57 mean?

Check your answer

The upper limit is 50 + 6 = 56 and the lower limit is 50 − 6 = 44. A result of 57 is outside the limits, so something has changed and the process is stopped and checked.

Try it yourself: Measure the same thing ten times, such as the weight of ten spoonfuls of sugar from the same jar, and work out the average and the rough spread. Plot the ten results with the average as a centre line and the average plus and minus three times the spread as limits. The chart should hold all ten points, and it shows how a control chart is built from ordinary measurements.

5.9 Check yourself: quick self-check

Think about it first: a roaster buys green coffee at 14% moisture and stores it in a damp shed. The roast is uneven, the cup is musty, and the customers complain. Which control points failed, in order?

  1. Name the four parts of process control.
Check your answer

Set the standard, measure at the control point, compare with the standard, and correct the process before the next batch.

  1. A tea factory fires at 80 °C instead of 90 to 95 °C. What goes wrong?
Check your answer

The enzymes are not inactivated and the moisture is not cut by 3 to 4%, so fermentation goes on in the pack and the tea loses its quality in storage.

  1. Why is the Karl Fischer method used for roasted coffee moisture instead of the air oven method?
Check your answer

Roasted coffee holds only about 2 to 5% water, and the air oven method is not accurate enough at that level. Karl Fischer extracts the water with methanol first and measures it directly.

  1. A tea tests within its ash and extract limits. Is that enough to guarantee it will keep?
Check your answer

No. The composition standard must be met together with the pack standard: pack at 3% moisture in lined chests, seal at once, store cool and dry. A correct recipe in a bad pack still goes mouldy.

  1. Why hold a batch of tea or coffee after processing instead of packing and shipping it at once?
Check your answer

The hold lets a faulty batch show itself, through its moisture, grade or cup test, before it reaches the buyer. A fault caught in the hold costs a batch; a fault caught by the buyer costs the factory's name.

Try it yourself: Pick one food you make regularly at home and write down its control points: the temperature, the time, the weight and the test that tells you it is done. Cook it twice, once following your list and once by feel. Which batch was more consistent, and what did the list catch that feel missed?

Key ideas to carry forward

  • Process control is four parts: set the standard, measure at the control point, compare, and correct before the next batch.
  • Tea control points run from plucking standard to firing at 90 to 95 °C for 20 minutes and packing at 3% moisture; the sensitive points are withering and fermentation.
  • Coffee control points run from processing within 36 hours to drying to 10 to 12% moisture, roasting at 200 to 250 °C, and quenching before grinding; green coffee moisture is 10 to 13%.
  • Standards are written limits: tea ash, extract and fibre; coffee moisture 10 to 13% green and 2 to 5% roasted; spice moisture about 10%, with irradiation up to 10 kGy judged safe for any food; tea packing at 3% moisture and instant tea moisture well below 5%.
  • Sampling must cover the whole lot and every batch, and results must be written down and acted on; batches are held until their tests pass.
  • Blending is standardisation by mixing: 20 to 40 tea grades in a retail blend, and 15% instant tea added without changing taste.
  • A control chart centres on the average of good running, with limits at the average plus and minus three standard deviations; a result outside the limits is a signal that something has changed.
  • Averaging several samples tightens the chart, because the scatter of an average falls as the square root of the number of samples.

Glossary

Process control

Holding a manufacturing process inside its set limits by measuring at fixed points and correcting it before the product goes wrong. Food tech example: checking fermentation time and temperature in every batch of black tea.

Standardisation

Making every batch meet the same written standard, so the buyer gets the same product every time. Food tech example: blending 20 to 40 tea grades to hold a steady flavour.

Control point

A step in a process where a measurement is taken and compared with a standard. Food tech example: the firing temperature of 90 to 95 °C in black tea manufacture.

Karl Fischer method

A method for measuring small amounts of water, used for roasted coffee at 2 to 5% moisture. Food tech example: measuring moisture in roasted coffee where the air oven method is not accurate enough.

Dielectric moisture meter

A meter that measures water content from the dielectric constant of the material, used for green coffee in the field. Food tech example: a Kappa meter calibrated against the air oven method.

Batch

One lot of product made in one run, tested and recorded as a unit. Food tech example: holding a batch of tea until its tests pass before packing.

Water extract

The soluble matter extracted from tea by water, at least 32.0% by the standard. Food tech example: a low water extract pointing to spent leaf in a tea lot.

Blending

Mixing grades or lots to hold a steady standard. Food tech example: adding 15% instant tea to regular tea without changing its taste.

Control chart

A chart of a process measurement over time, with a centre line at the average of good running and control limits three standard deviations away. Food tech example: withering moisture charted batch by batch.

Standard deviation

The measure of how far results typically scatter from their average. Food tech example: a withering moisture with a standard deviation of 1 percent is steadier than one with a standard deviation of 3 percent.

Control limits

The lines at the centre line plus and minus three standard deviations, beyond which a result signals that the process has changed. Food tech example: withering moisture limits of 66.5 and 72.5 percent around a centre of 69.5.

References

  • Nagalakshmi, Srikantayya. Tea: An Appraisal of Processing Methods and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ramalakshmi, Kulathooran, and Raghavan, Bashyam. Coffee: A Perspective on Processing and Products. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Farkas, Istvan. Control Aspects of Postharvest Technologies. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.

Unit 6: Spices: Production, Chemistry and Technology

Note

Unit 6: Spices: Production, Chemistry and Technology

A pinch of black pepper can change a whole pot of food, and a pinch of dried ginger can change it differently. The power of a spice sits in a few drops of volatile oil, and those drops are easily lost by heat, by air and by time. This unit is about keeping them.

6.1 What a spice is

A spice is a plant part used in small amounts to flavour food: a berry, a bud, a bark, a rhizome, a seed or a pod. The main attribute of a spice is the improvement of food flavour and taste by its characteristic components, especially its volatiles, the aromatic compounds that evaporate easily and reach the nose. Those components are very heat-sensitive and must be preserved to the maximal level during all processing steps.

Spices differ from herbs mainly in the part used and the strength of flavour, but the technology is the same: harvest at the right stage, dry gently, keep the volatile oil, and pack so air and moisture cannot get in. The six spices in this unit are ginger, turmeric, chillies, cardamom, cinnamon and black pepper. The books cover them unevenly, and this unit says plainly where the coverage ends.

🤔 Pause and think: a spice is used by the pinch, not by the kilo. Why does that make its drying and packing more demanding, not less, than drying a vegetable?

Quick check

What gives a spice its power to flavour food, and why is it easily lost?

Check your answer

Its volatile aromatic components, which evaporate easily and reach the nose. They are very heat-sensitive, so hot drying, air and long storage drive them off or break them down.

6.2 Production: how spices are grown and harvested

Spices are grown in the tropical and subtropical belt, mostly as smallholder crops with hand harvesting. The part harvested decides the whole process that follows:

  • Berries. Black pepper is the berry of Piper nigrum, picked and dried whole.
  • Rhizomes. Ginger is the rhizome of Zingiber officinale, dug, peeled and dried.
  • Bark. Cinnamon is the bark of coppiced shoots of Cinnamomum zeylanicum or C. cassia, stripped, fermented and rolled into quills.
  • Pods. Chillies are the pods of Capsicum frutescens, picked and cured and dried whole or ground.
  • Flower buds. Cloves are the buds of Syzygium aromaticum, picked and dried whole.
  • Seeds and leaves. Coriander, mustard, dill and others are harvested as seed or leaf and dried.

The books give the drying characteristics of these spices in a table, and section 6.4 works through the six spices of this unit against that table. Turmeric and cardamom do not appear in that table, and the books give no drying schedule for them: that gap is recorded in section 6.6 rather than filled in.

Quick check

Why does the harvested part of the plant decide the drying method?

Check your answer

A berry, a bark, a rhizome and a pod hold their volatile oil in different tissues and dry at different rates. A method that suits a thin bark would scorch a thick rhizome, and a method that suits a berry would not dry a pod evenly.

6.3 Chemistry: flavour components in general terms

The flavour of a spice comes from its volatile oil and its nonvolatile pungent compounds. The books study the volatiles of many spices: cardamom, ginger, black pepper, cloves, coriander, dill, garlic, rosemary and others all have published work on their volatile compounds, and the references for ginger run to five separate studies of its volatile and pungent compounds. The pungent compounds of ginger and the volatile components of ginger oil have been analysed by gas chromatography, and cardamom has been studied for supercritical carbon dioxide extraction.

What the books do not give is a full chemistry for each spice in this unit: no composition tables for ginger, turmeric, chillies, cardamom, cinnamon or black pepper individually, no oleoresin extraction yields or procedure, and no grading standard for any one of them. Section 6.6 records that gap. The chemistry that can be stated from the books is general: the flavour sits in the volatile oil, the oil is heat-sensitive, and every step of processing must be chosen to keep it.

Quick check

What are the two kinds of flavour compounds in a spice?

Check your answer

The volatile oil, which gives aroma, and the nonvolatile pungent compounds, which give heat and bite. Both must survive drying and storage for the spice to keep its character.

6.4 Processing and drying: the six spices

Drying is the main processing step for spices, and the rule is low temperature, chosen individually for each spice. Sun and field drying is the oldest and still the most typical method. Direct sun can discolour the tissue, so shadow, open-air or cabinet drying is used where colour matters. Larger-scale drying uses tunnel and conveyor dryers, as for fruits and vegetables, with multistep hot air falling as drying advances, for example 110 °C at first, then 60 °C and 40 °C at the end.

The books give these drying characteristics for the spices of this unit:

  • Black pepper. Berries are fermented as a pretreatment, then dried in the sun on mats or concrete floors, in about 20 hours, to whole corns or powder. White pepper is made from the same berries by soaking them for 8 days and removing the skin, which is decortication.
  • Chillies. Pods are cured in shelters, then dried in the sun for 5 to 15 days, or artificially in tunnels or on steel belts in about 30 hours, to whole or ground spice. About 85% of the water must be removed.
  • Cinnamon and cassia. Coppiced shoots are fermented, then dried in shadow in the first stage and in the sun as a second stage, over 3 to 4 days, until the quills are smooth and pale brown, then sold as quills or powder. Artificial drying is not practical for them.
  • Ginger. Rhizomes are peeled and boiled, then dried in the sun for 8 days to whole rhizomes, called hands. Artificial drying is not practical for them by the book's table.
  • Turmeric. No drying entry appears for turmeric in the books on disk: no part used, no pretreatment, no drying time. It is named in the outline but not covered in the sources.
  • Cardamom. No drying entry appears for cardamom in the books on disk either. The books cite studies of cardamom volatiles and of supercritical carbon dioxide extraction of cardamom, but they give no production, drying or grading procedure for it.

The final moisture level should be low enough to stop microbes growing, but not so low that the tissue over-dries and loses colour and gains hardness. An average of 10% moisture content on a wet basis is desirable for most spices.

Drying spices

Fresh spice
Berry, bark, rhizome, pod, bud, seed, leaf

↓

Pretreat
Ferment, peel, boil, cure or soak, as the spice needs

↓

Dry gently: sun or shadow
Sun or shadow drying, with no temperature schedule
Dry gently: tunnel or conveyor
110 °C first, then 60 °C, then 40 °C

↓

Dry to about 10% moisture
Low enough to stop microbes growing, not so low that the tissue over-dries

↓

Pack at once
In closed packs, away from air, light and damp

Figure 6.1: Drying spices. Sun and shadow drying carry no temperature schedule; the 110 °C, then 60 °C, then 40 °C schedule belongs to tunnel and conveyor drying only. The heat falls as drying advances, so the volatile oil survives. Drawn from the spice drying methods in Grabowski et al., Drying of Fruits, Vegetables, and Spices.

Quick check

Why does tunnel drying start hot at 110 °C and finish cool at 40 °C?

Check your answer

Early in drying the spice is wet, and evaporation cools it, so hotter air is safe. Late in drying the spice is nearly dry and heats up fast, so cool air is needed to protect the volatile oil and the colour.

6.5 Storage, decontamination and quality

Measuring moisture in spices is harder than in other foods. The usual oven test measures weight lost on heating, but the volatile oils of spices are also lost on heating, and that loss is counted as moisture. The trade therefore uses a codistillation method for most spices: the spice is covered with toluene, the toluene is raised to its boiling temperature, the moisture codistills with it, and the water is separated from the condensed toluene and measured. Karl Fischer titration is used for some spices, and a standard oven test is sufficient for paprika and other capsicum products. Vacuum drying is also used to evaporate moisture at lower temperatures.

Decontamination. Spices carry a heavy load of microbes from the field and from sun drying on the ground. Irradiation is used to decontaminate herbs, spices and food ingredients at high doses of 10 to 50 kGy. The international bodies concluded that irradiation of any food up to a total dose of 10 kGy presents no toxicological danger and introduces no particular nutritional or microbiological problem. Low doses up to 1 kGy inhibit sprouting and disinfect from insects; medium doses of 1 to 10 kGy kill spoilage microbes and extend shelf life.

Quality and adulteration. The books give specifications for spices and condiments, and tests for them, including a test for added starch in spice powders: starch is cheap, it bulks the powder, and it is detected by chemical test. Other adulterations are spent spice, exhausted of its oil, and foreign matter such as husk, sand and dirt. The defence is the same as for tea and coffee: inspection, microscopy, chemical tests and the nose. A spice that smells of nothing has lost its volatile oil, whatever the label says.

🤔 Pause and think: an oven moisture test on a spice reads 14% moisture, but the codistillation test reads 10%. Which reading is right, and why do they differ?

Quick check

Why is the oven test misleading for spice moisture?

Check your answer

The oven drives off the volatile oils as well as the water, and the weight lost is counted as moisture. Codistillation with toluene separates the water from the oil and measures only the water.

6.6 What the sources do not cover

This unit is honest about its limits, because the outline asks for more than the books give:

  • Individual-spice chemistry. The books give no composition tables for ginger, turmeric, chillies, cardamom, cinnamon or black pepper individually. The flavour chemistry is covered only in general terms and in cited studies of volatiles. Section 6.8 now names the key flavour carrier of each of the six spices, but no quantities are stated, because the books give none.
  • Oleoresin extraction. Section 6.8 now describes steam distillation, solvent extraction and supercritical carbon dioxide extraction in principle, and how to choose between them. No yields and no procedure are given for any spice, because the books give none.
  • Grading. The books give no grading standard for any of the six spices: no size, colour or quality grades, and no sampling standard. Grading is not invented here.
  • Turmeric and cardamom drying. Neither appears in the handbook's spice drying table, so no drying schedule is given for them.
  • Production figures. The books give no production statistics for the six spices, and none are presented here.

What the sources do cover is the drying of black pepper, chillies, cinnamon and ginger, the general chemistry of spice volatiles, moisture measurement by codistillation, decontamination by irradiation, and the tests for adulteration. That is the content of this unit, and the gaps above are left open for a source that covers them.

Quick check

Why does this unit describe oleoresin extraction only in principle, with no yields or procedure?

Check your answer

Section 6.8 covers steam distillation, solvent extraction and supercritical carbon dioxide extraction in principle only. The books on disk give no yields or procedure for any spice, so giving them would mean inventing them, which breaks the rule that the notes follow the books.

6.7 Practical note: drying and testing a spice at home

Dry a small batch of chillies or black pepper by the book's method: spread them thinly in the sun, turn them daily, and bring them in at night. Weigh them each day until the weight stops falling: that is the point where they have reached about 10% moisture. Grind a little and smell it at once, then grind a little more and leave it open for a week. The fresh-ground spice smells strong and the week-old powder smells flat: that is the volatile oil leaving, and it is why spices are sold whole and ground near the time of use.

Test a bought ground spice for added starch: mix a little with water, heat it gently, and add a drop of iodine solution. A strong blue-black colour shows starch. A pure spice powder gives little or no such colour.

Quick check

Why are spices sold whole rather than ground, when ground spice is easier to use?

Check your answer

Grinding exposes a large surface and lets the volatile oil escape fast. Whole spice keeps its oil until it is ground, so it stays strong far longer.

6.8 Oleoresin and essential oil: getting the flavour out

Section 6.3 covered the flavour chemistry of spices in general terms, and section 6.4 covered drying. This section covers the step the outline asks for that those sections did not: extracting the flavour itself, as essential oil and as oleoresin. The methods below are covered in principle only: no yields and no procedure for any spice, because the books give none, and the flavour carriers are named without quantities. A spice sold whole keeps its flavour locked in its tissue. Extraction takes that flavour out in a concentrated form that a food factory can dose by the drop, and the method chosen decides what comes out and what is left behind.

Essential oil: steam distillation. The oldest method is steam distillation. Steam is passed through the ground spice, and the volatile oils, the light aromatic compounds that carry the smell, evaporate with the steam. The mixed vapour is cooled, and it condenses into two layers: water below and oil above, which are separated. The oil that comes out is the essential oil: the smell of the spice, concentrated, with no colour and no bite. The heat is the limit: steam is hot, and the most delicate aroma compounds can be changed or lost in it, so the distillation is run gently and the oil is kept cool and dark afterwards, because light and air spoil it as fast as heat does.

Oleoresin: solvent extraction. An oleoresin is the whole flavour, not just the smell: the volatile oil plus the non-volatile compounds that carry the bite, the heat and the colour, such as the pungency of pepper and chillies and the colour of turmeric. It is made by solvent extraction: the ground spice is washed with a food-grade solvent, which dissolves both the oil and the non-volatile flavour compounds, and the solvent is then evaporated off, leaving a thick, dark extract behind. The oleoresin is far stronger than the ground spice, so it is dosed in tiny amounts, and it keeps longer, because the flavour is sealed in the extract instead of spread over a large surface. The risk is the solvent: it must be fully removed, and the extraction must be run so that no residue is left in the product.

Supercritical carbon dioxide: the cold solvent. The third method uses carbon dioxide held above its critical point, where it behaves like both a gas and a liquid. In that state it flows through the spice like a gas and dissolves flavour like a liquid, and when the pressure is released it turns back into gas and leaves, carrying nothing with it. The handbook cites a study of supercritical carbon dioxide extraction of cardamom, and the reason the method is used is in its properties: it runs cool, so delicate aromas survive, and it leaves no solvent residue, because the solvent is a gas at ordinary pressure. The cost is the equipment: the pressures are high, and the plant is dearer than a still or a solvent extractor, so the method is used where the flavour is delicate and the price supports it.

Choosing between the methods. The choice follows the product, not the fashion. Steam distillation suits a spice whose value is its smell, such as cardamom or cinnamon, and where a colourless oil is wanted. Solvent extraction suits a spice whose value is its bite or colour as well as its smell, such as chillies, black pepper and turmeric, where the oleoresin carries the whole character in one extract. Supercritical carbon dioxide suits a delicate, high-value spice where heat would spoil the aroma and where no solvent residue can be accepted. A buyer who wants the smell buys the essential oil; a buyer who wants the whole spice character in a drop buys the oleoresin.

The six spices: what carries each flavour. The books give no composition tables for the six spices individually, and none are invented here. What standard food science names as the key flavour carrier of each is:

The six spices and their key flavour carriers

Ginger: gingerol
The pungent compound that gives fresh ginger its heat.
Turmeric: curcumin
The yellow compound that gives turmeric its colour and much of its character.
Chillies: capsaicin
The pungent compound that gives chillies their heat.
Cardamom: 1,8-cineole
The cool, camphor-like compound that leads cardamom's aroma.
Cinnamon: cinnamaldehyde
The warm, sweet compound that leads cinnamon's aroma.
Black pepper: piperine
The pungent compound that gives black pepper its bite.

Figure 6.2: The key flavour carrier of each of the six spices. Each compound is named, not measured: the books give no composition tables, so no quantities are stated.

🤔 Pause and think: a sauce maker wants the heat of chillies without the red colour and without any solid particles. Which extract should the maker buy, the essential oil or the oleoresin, and why?

Quick check

Why does an oleoresin carry more of a spice's character than its essential oil?

Check your answer

The essential oil holds only the volatile aroma compounds, while the oleoresin holds the volatile oil plus the non-volatile compounds that carry the bite, the heat and the colour.

Try it yourself: Put a few drops of a food-grade spice extract, such as vanilla or peppermint, in warm water and smell it, then compare it with the ground spice. The extract should smell stronger and cleaner, because the flavour is concentrated and the fibrous tissue is gone: that is what extraction does for a factory, at a larger scale.

6.9 Check yourself: quick self-check

Think about it first: a spice merchant dries chillies fast in very hot air, grinds them at once, and packs them in open sacks. The chillies lose their colour, the powder loses its smell in a month, and the sacks pick up damp. Which step failed at each point?

  1. Why must spice drying be low-temperature and chosen individually for each spice?
Check your answer

The flavour sits in heat-sensitive volatile oils. Hot drying evaporates or breaks them down, and each spice holds its oil in different tissue and dries at a different rate.

  1. Black pepper dries in about 20 hours and ginger in 8 days. Why the difference?
Check your answer

Pepper berries are small and dry fast on mats in the sun. Ginger rhizomes are thick and wet, peeled and boiled first, and take 8 days in the sun to reach a safe moisture.

  1. Why is codistillation with toluene used to measure spice moisture?
Check your answer

An oven test drives off volatile oils as well as water and counts the oil loss as moisture. Codistillation separates the water from the toluene and measures only the water.

  1. What dose of irradiation decontaminates spices, and what dose is judged safe for any food?
Check your answer

High doses of 10 to 50 kGy decontaminate herbs, spices and food ingredients. Irradiation up to 10 kGy is judged to present no toxicological danger for any food.

  1. Name two things the books do not cover for the six spices of this unit.
Check your answer

Any two of: individual-spice chemistry with quantities, oleoresin extraction yields or procedures, grading standards, drying schedules for turmeric and cardamom, or production figures. Extraction methods are covered in principle only in section 6.8, not with yields or procedures.

Try it yourself: Buy whole black pepper and ready-ground black pepper of the same brand. Grind the whole pepper fresh and compare the smell of the two. Leave both open for a week and compare again. The gap between them is the volatile oil, and it is the reason whole spice keeps and ground spice fades.

Key ideas to carry forward

  • A spice flavours food by its volatile oil and its pungent compounds, both of which are heat-sensitive and easily lost.
  • Drying is the main processing step: low temperature, chosen per spice, with hot air falling from 110 °C to 60 °C to 40 °C as drying advances.
  • Black pepper dries in about 20 hours, chillies in 5 to 15 days in the sun or 30 hours artificially, cinnamon in 3 to 4 days in shadow then sun, and ginger in 8 days in the sun. An average of 10% moisture is desirable for most spices.
  • Spice moisture is measured by codistillation with toluene, because oven drying also drives off volatile oils.
  • Irradiation at 10 to 50 kGy decontaminates spices; up to 10 kGy is judged safe for any food.
  • The books give no individual-spice chemistry with quantities, no oleoresin extraction yields or procedures (extraction is covered in principle only in section 6.8), no grading, and no drying schedules for turmeric and cardamom, and this unit does not invent them.
  • Essential oil is the volatile aroma taken out by steam distillation; oleoresin is the whole flavour, aroma plus bite and colour, taken out by solvent extraction.
  • Supercritical carbon dioxide extracts cool and leaves no residue, because it turns back into gas when the pressure is released, so it suits delicate, high-value spices.

Glossary

Volatile oil

The aromatic compounds of a spice that evaporate easily and give its smell and flavour. Food tech example: the volatile oil of black pepper that escapes when the spice is ground and left open.

Oleoresin

A concentrated extract of a spice holding its volatile oil and pungent compounds: the whole flavour, volatile oil plus non-volatile bite and colour, taken out by solvent extraction as a thick extract. Food tech example: oleoresin used to flavour foods where a standard strength is wanted, such as chilli oleoresin carrying both the heat and the red colour; no extraction yields or procedure for it are given in the books on disk, and section 6.8 covers extraction in principle only.

Codistillation

Measuring spice moisture by distilling it with toluene, so the water separates from the oil and is measured alone. Food tech example: a spice testing 10% moisture by codistillation against 14% by the oven test.

Quill

A rolled tube of cinnamon bark, dried until smooth and pale brown. Food tech example: cinnamon quills sold whole and ground near the time of use.

Curing

Holding a harvested spice under cover before drying, so its flavour develops. Food tech example: chilli pods cured in shelters before sun drying.

Decortication

Removing the outer skin or husk from a spice, as in making white pepper from black pepper berries by soaking. Food tech example: white pepper made by soaking berries for 8 days and removing the skin.

kGy

The kilogray, the unit of irradiation dose. Food tech example: spices decontaminated at 10 to 50 kGy.

Added starch

Starch mixed into ground spice to bulk it out, detected by chemical test. Food tech example: a spice powder that turns blue-black with iodine, showing added starch.

Essential oil

The volatile aroma compounds of a spice, taken out by steam distillation as a concentrated oil. Food tech example: cardamom oil carrying the smell of cardamom with no colour and no bite.

Supercritical carbon dioxide

Carbon dioxide held above its critical point, where it flows like a gas and dissolves like a liquid, and leaves no residue when the pressure is released. Food tech example: cardamom extracted by supercritical carbon dioxide, as cited in the handbook.

References

  • Grabowski, Stefan, Marcotte, Michele, and Ramaswamy, Hosahalli S. Drying of Fruits, Vegetables, and Spices. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Lacroix, Monique, Marcotte, Michele, and Ramaswamy, Hosahalli S. Irradiation of Fruits, Vegetables, Nuts, and Spices. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Alvo, Peter et al. Potential Applications of Volatile Monitoring in Storage. In Chakraverty, Amalendu et al. (eds.), Handbook of Postharvest Technology: Cereals, Fruits, Vegetables, Tea, and Spices. Marcel Dekker, 2003.
  • Ranganna, S. Handbook of Analysis and Quality Control for Fruit and Vegetable Products. Tata McGraw-Hill, 2nd edition, 1986.
  • Srivastava, R. P., and Kumar, Sanjeev. Fruit and Vegetable Preservation. International Book Distributing Co., 3rd edition, 2002.
  • Fellows, P. Food Processing Technology: Principles and Practice. Woodhead Publishing, 2nd edition, 2000.

Unit 7: Tea Breeding and Genetics

Note

Unit 7: Tea Breeding and Genetics

Every cup of tea starts in a decision about which plant to grow. A tea bush lives and yields for decades, so a good choice pays for years. This unit is about how breeders find, keep, cross and test better tea plants, and how DNA tools help them. Leaf chemistry and manufacture belong to Units 2 and 3.

7.1 The plant the breeder works with

Tea is Camellia sinensis, a woody evergreen in the family Theaceae, a family of about 30 genera and some 500 species. Three types of tea plant are recognised: the China type, Camellia sinensis; the Assam type, Camellia assamica; and the Cambod type, Camellia assamica ssp. lasiocalyx. All three cross with each other, so many garden plants sit somewhere between the types. In a garden it is kept as a low bush in continuous vegetative growth, and the part picked is the young shoot: the apical bud with two young leaves.

Note on classification: the Handbook in Unit 1 uses the two-variety classification, C. sinensis var. sinensis and C. sinensis var. assamica, while Ahuja here uses the three-type classification above. The Assam plant is the same plant, ranked as a variety in one book and as a species in the other, and the Cambod type is the intermediate form.

Two facts shape everything a breeder does. First, tea is strongly outbreeding and self-sterile: a plant crossed with its own pollen sets little or no seed. Second, tea is a perennial with a long generation time: five to ten years pass from one generation to the next.

The chromosome number is steady at 2n = 30 in the three types, with a base number of n = 15. The genome is large, about 4,000 Mb.

China type
Small leaf, hardy in cold
Assam type
Large leaf, warm country
Cambod type
Between the two, crossing freely with both

↓

Garden tea: mostly hybrids
2n = 30, genome about 4,000 Mb

Figure 7.1: The three types of tea plant. The three types are three parallel types, not a sequence: all three feed into garden tea, which is mostly hybrid between them. Drawn from Ahuja et al., Science of Tea Technology, Ch. 2.

Quick check

Name the three recognised types of tea plant and give the chromosome number of cultivated tea.

Check your answer

China type (Camellia sinensis), Assam type (Camellia assamica) and Cambod type (Camellia assamica ssp. lasiocalyx). Cultivated tea is diploid with 2n = 30.

7.2 Where tea came from, and why its gene pool is wide

Tea is native to East and South Asia, and probably arose around the meeting point of the lands of north-east India, northern Burma, south-west China and Tibet. The south-west Chinese province of Yunnan is regarded as the centre of origin of Camellia, where the greatest diversity and the most species are found. Tea drinking began in China, and tea spread widely only from about the middle of the seventeenth century. The China type spread to cold-climate regions, while the Assam type did well in tropical and equatorial regions.

The spread left traces that DNA studies can now read. Molecular marker studies support the picture built earlier from leaf shape and the terpene index of the aroma: Korean tea shows little genetic diversification since its introduction from China, Japanese tea arose from a limited set of introductions from China and then developed through local selection and crossing, and Kenyan clones trace back to collections made in the Assam region.

Diversity across tea populations runs in a clear order, from most to least diverse: wild Camellia, then Indian tea, Chinese tea, Kenyan tea, Sri Lankan tea, Vietnamese tea, Japanese tea and Taiwan tea. The widest choice of genes sits near the centre of origin, and it shrinks as you move away from it.

Quick check

Which tea population is the most diverse, and which holds the least diversity in the order reported by Wachira and co-workers?

Check your answer

Wild Camellia is the most diverse, and Taiwan tea the least in that order.

7.3 Germplasm: the living library, and why it is shrinking

Germplasm is the whole set of genetic material a breeder can draw on: cultivated varieties, wild species, weedy relatives, old seed-grown landraces, improved clones and seed stocks, and breeding lines and hybrids. Worldwide, over 600 cultivated tea varieties are available, and many carry single striking traits.

Table 7.1: Tea cultivars with special characters (examples from the book)

Special characterCultivar example
Drought resistanceUPASI-9
Frost toleranceBS-26
Blister blight toleranceTRI-2043, DT-1
High caffeine content (6.96%)Wild tea at Yunnan

That library is shrinking. Old seed-grown plantations are uprooted and replanted with a small number of recently released, high-yielding clones. The book is blunt: genetic diversity is being narrowed at an alarming rate, and the diverse material should be conserved before it is lost.

Conservation happens in gene banks, called ex situ repositories, where living collections are held outside their original habitat. National repositories in China held about 2,665 accessions by the end of 2003, and one major collection in north-east India holds more than 2,532 accessions. The numbers sound large, but the book warns that they are far from adequate for a self-incompatible, outcrossing species that is losing its diversity so fast.

Quick check

Give two reasons the book gives for the loss of tea genetic diversity.

Check your answer

Old seed-grown fields are replanted with a few improved clones, and diversity is lost through deforestation, shifting cultivation and industrialisation.

7.4 Why breeding tea is slow and hard

Tea breeding has not reached the stage of repeated cycles of recombination. The book lists the reasons below.

  1. Perennial nature: the plant lives for decades, so each trial runs for years.
  2. Long gestation periods: years pass before a seedling can be judged.
  3. High inbreeding depression: crossing related plants weakens the offspring.
  4. Self-incompatibility: a plant will not set seed with its own pollen.
  5. Low variability for biotic and abiotic stresses: little resistance to pests, diseases, drought, cold and poor soil sits in the available material.
  6. Lack of distinct selection criteria: for a long time breeders had no clear test for what makes a good bush.
  7. Low success rate of hand pollination.
  8. Seed maturation: capsules reach physiological maturity in 10 to 11 months, and complete seed development and maturation is generally quoted as 12 to 18 months (about a year).
  9. Clonal differences in flowering time and fruit bearing make crossing hard to arrange.

Add to that the risk of a narrow genetic base. Many commercial clones are related to each other, and when a whole region grows closely related vegetative clones, one new pest, disease or cold snap can hit them all at once. Breeding strategy now aims to widen the base by deliberately crossing parents that are genetically far apart.

Try it yourself

Sort the nine obstacles above into two piles on paper: biology of the plant, and how people grow and test tea. Which pile is bigger, and which obstacle would you tackle first? Write one sentence giving your reason.

7.5 What breeders are trying to achieve

The two great aims of tea breeding are higher yield and better cup character in the made tea. Beyond that, the aims change with the place, because the limiting problem changes with the place. Cold tolerance matters most where low temperature injures the plant. Drought tolerance matters most where tea grows as a rain-fed crop. Resistance to blister blight, stem canker, pests and nematodes matters wherever those attackers do serious damage. Other aims include shorter winter dormancy, hail resistance in hilly country, waterlogging tolerance in low-lying plains, suitability for a particular type of manufacture, low caffeine and high catechin content for particular product lines, varieties suited to organic growing, and long internodes, which suit mechanical harvesting.

Table 7.2: Region-specific breeding objectives in tea (shortened from the book)

ObjectiveWhy it mattersWhere it matters most
Improving qualityQuality is directly linked to profitabilityBlack tea producing countries
Increasing yieldExtending the planted area is limited worldwideWorldwide
Drought toleranceDrought cuts productivity in all tea regionsWhere tea grows as a rain-fed crop
Cold hardinessReduced productivity in winterCold-climate regions
Disease resistance (blister blight, stem canker)Blister blight attacks only young leavesWarm, wet tea regions

Quick check

Why is long internode length a breeding objective?

Check your answer

Long internodes suit mechanical harvesting and easier hand plucking, cutting picking labour, which can be up to 60% of field input costs.

7.6 How yield and quality traits are inherited

Work on inheritance in tea is thin, and the book says so plainly: study is complicated by the outbreeding habit, the long life cycle and the fact that most traits of interest are governed by many genes at once. What is known comes in single findings, and each one is worth holding on to.

Leaf size in China by Assam hybrids is quantitative with additive gene action: offspring sit between the parents. Long internode of the young shoot is completely dominant over short internode. Yield in biclonal crosses shows a positive link between the mean yield of the parents and that of their progeny when the parents are widely different, but the link breaks down when morphologically similar clones are crossed. Heterosis, the extra vigour of the hybrid over its parents, ranged from 21 to 85% over the mid-parent value for yield.

Fermentation ability, measured as polyphenol oxidase activity, is inherited with normal plants dominant over non-fermenter parents. The caffeine-less character is different: two plants with 0.2% caffeine or less were found in a population from a cross with a wild relative, and the trait appears to be inherited recessively from the wild parent.

Quick check

A breeder crosses two high-yielding clones that look very much alike. Should she expect the progeny yield to follow the parents closely?

Check your answer

No. The link held when the parents were widely different, but it broke down for morphologically similar clones.

7.7 Selection: reading the bush

Selection is the oldest breeding tool in tea, and it starts with looking. In commercial plantings, about 10% of the bushes produce only 2% of the total crop, while about 0.5% of bushes produce as much as, or more than, 300% of the yield of an average bush. That spread is the breeder's raw material: somewhere in an ordinary field, a few bushes are already far better than the rest.

Mother bushes are first picked by eye, using leaf size, shape, pose and colour. But a high-yielding mother bush is not always a genetically high yielder: one study found no link between the yield of a mother bush and that of its vegetative offspring. Eye selection is the start, not the verdict.

Two traits deserve a closer look. Density of plucking points, the number of shoots actually harvested from the plucking table, is the most important criterion for quick selection by eye.

Then comes the environment. Clones respond differently to different sites, a pattern called genotype by environment interaction, so a clone tested in one place can fail elsewhere. A Kenya study found 23 yield QTLs at two sites. No QTL was detected at both sites, showing strong genotype-by-site interaction.

Look: pick out the best bushes by eye

↓

Multiply: raise the candidates as clones

↓

Test: trial the clones at several sites over several years

↓

Release: the clones that hold up everywhere go to gardens

Figure 7.2: The clonal selection pipeline, from picking mother bushes by eye to releasing a tested clone. Drawn from Ahuja et al., Science of Tea Technology, Ch. 2.

Quick check

Why can a high-yielding mother bush still be a poor choice as a clone parent?

Check your answer

Its high yield may come from position and history rather than genes: one study found no link between a mother bush yield and its offspring yield.

7.8 Breeding for tough conditions: drought, cold, pests and disease

A tea bush in the field faces drought, cold, waterlogging, fungi, insects and nematodes, and breeding for tolerance of these is a large part of the work.

Drought tolerance can be picked up in the field with xylem water potential measurements. Drought-tolerant clones put a greater share of their dry matter into the root system and frame, so breeders look for a high root to shoot ratio, fast root growth and deep roots. One practical answer is grafting a high-yielding clone onto the rootstock of a drought-hardy clone, with seven such combinations released in one programme. Grafting is covered in full in Unit 8.

Cold tolerance differs sharply between types. In observations on 816 cultivars, frost tolerance was highest in China types, then Assam, then Cambod: plants from China were resistant to freezing, while those from Sri Lanka, Burma and other South-east Asian countries were less hardy. Tea stays dormant for 3 to 4 months in winter in cold regions, and early flushing is a useful selection criterion there.

The most serious leaf disease is blister blight, caused by the fungus Exobasidium vexans. Even below a 35% disease threshold, one third of the yield can be lost, along with quality. Some clones resist it through several small defences together: more epicuticular wax, a thickened epidermis and cuticle, stomatal frequency on both surfaces, compact palisade cells, higher chitinase, more (-)-epicatechin and less (-)-epigallocatechin. For insect pests, resistance to mulberry scale, which lives inside the tree crown where sprays cannot reach, is selected in Japan with the marker MSR-1, linked to the resistance gene.

Quick check

Blister blight attacks young leaves. Name two features of a resistant clone that help it, and state the yield loss the book reports even below the 35% disease threshold.

Check your answer

Any two of: more epicuticular wax, thickened epidermis and cuticle, stomatal frequency on both surfaces, compact palisade cells, higher chitinase, more (-)-epicatechin and less (-)-epigallocatechin. Even below a 35% threshold, one third of the yield can be lost.

7.9 Breeding for quality: what the cup rewards

Quality is an inherent character of a clone, and it is the second great breeding aim after yield. The China type is described as higher in quality relative to strength, while the Assam type is high in strength relative to quality. Flush with fine hairs, pubescence, on the underside of young leaves and the terminal bud generally gives better tea than glabrous, hairless, flush, and the hairy part of the leaf holds large amounts of polyphenols, amino acids, organic acids, caffeine and carbonyl compounds.

Chemical measures also help. The ratio between the enzyme polyphenol oxidase and its substrate, the catechins, is used to characterise the quality potential of clones: the catechins mix with the enzyme during oxidation to form theaflavins, thearubigins and high polymerised substances. The chemistry itself is covered in Unit 2. Caffeine content is itself a breeding target: the book reports cultivars ranging from 0.14% caffeine to 6.96%, and caffeine-less plants at 0.2% or less, found through a cross with a wild relative.

Quick check

Why does a breeder look at the fine hairs on a young tea shoot?

Check your answer

Hairy flush generally gives better quality tea than hairless flush, and the hairy part holds large amounts of polyphenols, amino acids, organic acids, caffeine and carbonyl compounds.

7.10 Crossing: hybridization and the seed bari

Tea is cross-pollinated and its progeny are highly heterogeneous. Selfing a clone generally sets no seed, or very little, so pure lines cannot be built as in self-pollinating annual crops. The parent plants are the most important choice in a cross, and their genetic make-up should be known.

Hand pollination follows a fixed routine. Flowering branches are enclosed in fine muslin cloth or polythene bags after removing opened flowers, immature fruits and very small buds, keeping only buds about to open in one or two days. Pollen from bagged flowers is applied to the receptive stigma, and the flower is bagged and tagged. The stigma becomes receptive 24 hours before the flower opens and stays receptive for 24 to 48 hours. Tea pollen is large and sticky and loses viability fast in heat, so it suits wind pollination poorly and hand pollination gives far better seed set.

From flower to seed takes about a year in almost all tea regions. The fruits (capsules) reach physiological maturity in 10 to 11 months, and complete seed development and maturation is generally quoted as 12 to 18 months (about a year). The fruits hold one to three seeds. Fruit drop is heavy: about 70-80% falls during flowering, 11 to 16% while the fruit develops, and 1-6% as nearly mature capsules.

Seed orchards, called seed baris, are where improved seed is made at scale. Selected parents are planted in alternate rows in isolation and allowed to grow as trees. Seed from a cross between two clones is biclonal seed, and that variation is the material from which new clones are selected. Seed from a group of selected clones planted together and left to cross naturally is polyclonal seed: its performance is less predictable and less stable than biclonal seed. The clone-seed cycle, selecting clones, crossing them for seed, selecting new clones from the seed progeny, is the engine of continued improvement.

Quick check

What is the difference between biclonal and polyclonal seed, and which is more predictable in performance?

Check your answer

Biclonal seed comes from a cross between two chosen clones, so parentage is known. Polyclonal seed comes from a group left to cross naturally, so parentage is mixed and performance less predictable.

7.11 New tools: mutation, polyploidy, tissue culture and transformation

When crossing is slow, breeders reach for tools that create new variation or move genes directly. Mutation breeding uses radiation: irradiating single nodal cuttings with gamma radiation at 2 to 6 kR gave varietal and clonal differences in response, with survival lower than untreated cuttings, and 2 krad appeared to be the upper limit for survival.

Polyploid breeding works with chromosome number. Polyploids were more vigorous with profuse rooting, but tetraploid and natural triploid quality was inferior to diploids. Two triploid clones have been released for cultivation, TV-29 and UPASI-03.

Tissue culture opens two more doors. Somaclonal variation, variation appearing in plants raised from somatic embryos in culture, gave five variants from one clone. Gametoclonal variation aims at haploid plants, which are valuable because doubling a haploid gives a fully homozygous plant in one step. The first successful plants from tea anthers came from the cultivar Fuyun-7, and of 12 plantlets examined, 9 were aneuploids with chromosome numbers from 2n = 18 to 22. The technique still needs standardising.

Genetic transformation in tea is difficult, partly because the high polyphenol content of tea explants acts against Agrobacterium tumefaciens. The most striking result in the book is a low-caffeine tea made with RNA interference: a construct targeting caffeine synthase, delivered into somatic embryos, cut caffeine by 44-61% and theobromine by 46-67% compared with controls.

Quick check

Why are haploid plants valuable to a tea breeder, even though a haploid plant itself is not a finished cultivar?

Check your answer

Doubling a haploid gives a 100% homozygous plant in one step, making trait genetics far easier to study.

7.12 Reading the DNA: fingerprints, markers, maps and QTLs

This is where Chapter 3 takes over from Chapter 2. DNA markers are free from environmental influence and show differences at an early stage of the breeding programme, even in a seedling.

The first molecular work on tea diversity used RAPD markers, and the picture that emerged has held up across many studies: outbreeding woody plants keep most of their variation within populations rather than between them. In one large study of 123 commercial accessions, 85% of the genetic variation sat within populations and only 15% between them. In a national survey of 1,644 accessions, the material clustered into six major groups, and a core collection of 105 accessions captured 98% of the diversity. That core set is the practical payoff: a breeder who cannot test 1,644 plants can test 105 and still cover almost all the diversity.

DNA fingerprints give each cultivar a unique identity. They can check parentage: the registered parents of two Japanese cultivars were found to be wrong, and the genuine parents were identified by RAPD analysis. They can tell the China, Assam and Cambod types apart: nuclear SSR alleles classify genotypes into their groups, and 5S rDNA sequencing can separate China tea from Assam and Cambod. They can also authenticate dried tea and check uniformity in tissue culture.

SSR markers, short repeated sequences in the DNA, are the workhorse here. They are frequent, co-dominant, multi-allelic, reproducible and easy to detect by PCR. Development grew from 15 SSR loci to 625 polymorphic markers from enriched genomic libraries.

Linkage maps turn markers into a tool for breeding. The first map was built with RAPD markers, with markers linked to theanine content, bud sprouting date, anthracnose resistance and cold tolerance. A yield study built a map of 30 linkage groups spanning 1,411.5 cM, with a mean interval of 14.1 cM between loci, and found 23 putative QTLs for yield. Note: the book gives 14.1 cM in its text and 14.7 cM in its Table 3.6 for this map; this unit keeps the text value, 14.1 cM, consistently.

Table 7.3: Genetic linkage maps in tea (from the book)

Markers usedMap lengthWhat it gave
RAPD and AFLP1,349.7 cM126 markers, average distance 11.7 cM between loci
RAPD, AFLP, SSR and ISSR1,411.5 cM30 linkage groups; 23 putative yield QTLs
Take a leaf sample from each candidate

↓

Read its DNA fingerprint with markers

↓

Check: right cultivar? Uniform? Carries the marker?

↓

Keep the plants with the right markers, and test only those

Figure 7.3: How marker-assisted selection works in tea, from leaf sample to field testing. Drawn from Ahuja et al., Science of Tea Technology, Ch. 3.

The book also names the risks of leaning too hard on clones. In Japan, more than 90% of tea gardens are under improved clonal cultivars, with one dominant clone, Yabukita, accounting for 85% of cultivation.

Quick check

A breeder has 1,644 accessions to choose parents from. How does the core collection of 105 help, and what share of the diversity does it capture?

Check your answer

The core collection of 105 captures 98% of the diversity, so the breeder screens a manageable number and still covers almost all the variation.

7.13 What comes next

The book ends its breeding chapters with a list of things still to do, and it reads like a work plan. Build a worldwide germplasm appraisal through a common consortium. Study functional markers linked to the traits breeders care about, including EST-SSR and SNP markers. Build a high-density linkage map so QTLs for important traits can be found and used in marker-assisted selection. Develop early selection markers that work at the nursery stage for stresses, and doubled haploid plants, which would be 100% homozygous. Widen the genetic base through distant hybridization. Redefine the ideal tea plant, the ideotype, with a higher harvest index, selecting for high shoot density, high photosynthetic efficiency and low photorespiration. And define chemical quality parameters for black, green and oolong teas against tasters' scores and prices.

Practical note: if you ever visit a tea garden, ask which clones are planted and how many there are. A garden with five or six clones and some seed-raised plants is following the book's advice. A garden with a single clone across the whole estate is efficient, and the most exposed to whatever comes next.

Key ideas to carry forward

  • Tea is an outbreeding, self-sterile perennial with 2n = 30 and a genome of about 4,000 Mb, and breeding takes 5 to 10 years per generation.
  • The widest genetic diversity sits near the centre of origin, and it is shrinking as old seed-grown fields are replanted with a few clones.
  • Breeding aims are practical and local: yield and cup quality first, then cold, drought, hail, waterlogging, disease, pests and suitability for machines.
  • Selection starts by eye but must be confirmed by testing clones at several sites over years, because genotype by environment interaction is strong.
  • Hybridization, mutation, polyploidy, tissue culture and transformation all create or move variation, and DNA markers let breeders check identity, uniformity and trait genes years before the field can.
  • A core collection of 105 accessions captures 98% of the diversity of 1,644.

Glossary

Germplasm

The genetic material available for breeding: varieties, wild species, landraces, clones and breeding lines. Food tech example: a gene bank holding thousands of tea accessions.

Clone

Plants raised vegetatively from one mother plant, all genetically identical. Food tech example: a named tea clone planted across a garden.

Accession

One distinct sample held in a gene bank, such as one cultivar or one wild collection. Food tech example: a gene bank listing 2,665 tea accessions.

Self-incompatibility

The inability of a plant to set seed with its own pollen. Food tech example: a tea flower that sets seed only after cross-pollination.

Heterosis

The extra vigour of a hybrid over the average of its parents, often seen as higher yield. Food tech example: a tea hybrid yielding up to 85% more than the mid-parent value.

Biclonal seed

Seed from a cross between two chosen clones, with known parentage on both sides. Food tech example: biclonal tea seed from two high-yielding parents.

Polyclonal seed

Seed from a group of selected clones left to cross naturally, with mixed parentage. Food tech example: polyclonal tea seed of less predictable performance.

Seed bari

A seed orchard where selected tea plants are grown as trees, in isolation, to produce improved seed. Food tech example: a seed bari with two clones in alternate rows.

Genotype by environment interaction

How the same clone performs differently at different sites. Food tech example: a clone that yields best at one site and poorly at another.

Pubescence

The fine hairs on the underside of young tea leaves and on the terminal bud. Food tech example: hairy flush giving better quality tea than hairless flush.

Blister blight

A fungal disease of tea, caused by Exobasidium vexans, that attacks young leaves and can cut yield by a third. Food tech example: a clone with thick wax resisting blister blight.

Marker-assisted selection (MAS)

Selecting plants using a DNA marker linked to a gene for a wanted trait. Food tech example: mulberry scale resistant tea selected with the MSR-1 marker.

DNA fingerprint

A pattern of DNA markers unique to one cultivar, used to identify it. Food tech example: a fingerprint showing a cultivar's registered parents were wrong.

SSR marker

A short repeated sequence in DNA, used as a genetic marker because it varies between plants and is easy to detect. Food tech example: SSR markers separating China type tea from Assam and Cambod types.

QTL

A quantitative trait locus, a region of the genome linked to a trait governed by many genes, such as yield. Food tech example: 23 yield QTLs found in a tea mapping study.

Linkage map

A map of marker order and distance along the chromosomes, used to locate genes. Distances are measured in centimorgans (cM). Food tech example: a tea map of 1,411.5 cM with 30 linkage groups.

Haploid

A plant with a single set of chromosomes, from which a fully homozygous doubled haploid can be made. Food tech example: a haploid tea plant from anther culture.

RNA interference (RNAi)

A method that switches off a chosen gene, used here to lower caffeine by silencing caffeine synthase. Food tech example: transgenic tea with 44-61% less caffeine.

Core collection

A small set of accessions representing almost all the diversity of a much larger collection. Food tech example: 105 accessions capturing 98% of the diversity of 1,644.

Ideotype

The description of an ideal plant type for a breeding programme. Food tech example: a tea ideotype with high shoot density and low photorespiration.

References

  • Ahuja, P.S., Gulati, A., et al., eds., Science of Tea Technology. Chapter 2: Singh, S., Kumar, A., Karthigeyan, S. and Ahuja, P.S., Genetic improvement of tea.
  • Ahuja, P.S., Gulati, A., et al., eds., Science of Tea Technology. Chapter 3: Sharma, R.K., Bhardwaj, P., Kumar, R., Sharma, H. and Ahuja, P.S., Tea applied genomics: utilizing genetic variability and current genomic tools to enhance productivity.

Unit 8: Tea Propagation: Nursery, Seed and Grafting

Note

Unit 8: Tea Propagation: Nursery, Seed and Grafting

A tea garden is only as good as its planting material. Every bush starts as a seed or a cutting in a nursery, and the nursery is the most vulnerable stage in the life of the plant. This unit follows the young plant from the mother bush and the seed orchard through the nursery bed, the sleeve and the grafting bench, and into the biology of the seed itself. Breeding, the choice of which plants to multiply, is Unit 7. Here the question is how to multiply them well.

8.1 Why propagation matters

The aim of propagation is simple to state and hard to achieve: produce healthy, well-grown plants in the shortest possible time with the fewest losses. Raising a nursery plant takes over a year, 10 to 18 months in many tea regions.

Tea is multiplied in two ways. Seed is the oldest way, and the single-leaf nodal cutting was first demonstrated in 1931. Seed-raised plants give a mixed gene pool, with long tap roots and good adaptation to hard conditions. Clonal plants give uniformity. The book's guideline for a garden is to hold both clones and seed-raised plants in equal proportion, keep no single clone on more than 10% of the area, and hold five or six clones in each garden.

🤔 Pause and think: if every bush in a field were genetically identical, what would happen when a new disease arrived?

Quick check

Why does the book recommend a garden hold both clones and seed-raised plants, with no single clone on more than 10% of the area?

Check your answer

For stable productivity: a single clone everywhere is vulnerable to whatever it cannot handle.

8.2 What a nursery needs: weather, soil and water

A nursery succeeds or fails on three things: weather, soil and water. Nursery tea grows best in a hot, humid climate. The optimum temperature range is 30-35°C and should never exceed 35°C. The optimum night temperature is 15°C and should not fall below 5°C. Light intensity should sit between 25,000 and 40,000 lux. Relative humidity above 90% gives the best rate of plant establishment.

Soil should be neither too clayey nor too sandy. Clayey soil causes over-callusing and poor rooting, while sandy soil needs frequent watering and lets roots grow deep, so they break when plants are moved. Sandy loam is generally preferred. For cuttings the book recommends two sleeve layers: subsoil and sand in equal proportion for rooting, and topsoil and sand at 3:1 for growing.

Soil pH should sit between 4.5 and 5.5: below 4.5, slaked lime is applied, and above 5.5, aluminium sulphate solution brings it down. Nitrogen, phosphorus, potash and organic matter should all be high.

Water is the third need. A regular irrigation supply is a key factor. Over-watering is a common cause of failure, because tea is highly sensitive to waterlogging.

Table 8.1: Nursery conditions for young tea plants (from the book)

FactorValue
Day temperature, optimum30-35°C, never above 35°C
Night temperature, optimum15°C, not below 5°C
Relative humidityAbove 90% for best establishment
Soil pH4.5-5.5
Nematodes in nursery soilNot more than 6 per 10 g soil

Quick check

A nursery soil tests at pH 6.0. Is that suitable for tea cuttings, and what does the book say to use to bring it down?

Check your answer

No, it is above the 4.5-5.5 range. Aluminium sulphate solution is used to lower the pH.

8.3 Nursery structure, beds and sleeves

A nursery structure exists to build a small climate: temperature, humidity and shade tuned for young plants. The nursery should sit in a warm, sheltered position, close to water, sunny and bright, on raised, well-drained land, and not prone to hail or high wind.

Low shade sits only 20 cm above the ground, with dense bamboo or straw mats, and suits humid areas. High shade is a polythene tunnel with one side 90-100 cm high and the other 125 cm, covered with bamboo or coir matting on a slope. Overhead shade stands about 1.8 m above the ground, with thatch, coir mat or plastic net letting through 50% of the light or less.

Beds are 1.5 m wide, with 0.6 m between beds. A drain 15 cm deep and 15 cm wide is dug on the upper side, the bed is raised about 15 cm with a slight camber in the centre to avoid waterlogging, and sand is added for a sandy loam texture. Rooting beds take cuttings under a polythene tent; where the air is already above 90% humidity, the tent is not needed.

Sleeves are open-ended polythene bags of 150 gauge, open at both ends so water cannot lodge. Cuttings go in sleeves 10 cm lay-flat by 25-30 cm long, and seedlings in sleeves 12-15 cm lay-flat by 20-25 cm long. The soil is sieved, mixed with superphosphate at 500 g per cubic metre, and packed evenly, so the filled sleeve neither bulges nor holds gaps. Filled sleeves are stacked on staking beds, watered, and mulched.

Quick check

Why are both ends of a nursery sleeve left open?

Check your answer

So water cannot lodge in the sleeve: tea does not tolerate very wet soil, and a closed bottom would waterlog the young plant.

8.4 Mother bushes and making cuttings

Mother bushes are the bushes from which cuttings are taken, and all the progeny of one mother bush form a clone. A mature mother bush gives 50-300 cuttings a year, and at least 250 bushes supply cuttings for 1 hectare.

Mother bushes are planted 1.0-1.2 m apart and fed with NPK in split doses. Primary shoots are collected at pencil thickness, cut at 40 cm, and the bush is pruned 3-5 cm above the last mark. A mother bush left unpruned moves into flowering, and cuttings from a bush that has already formed flower initials root poorly and grow weakly, so mother bushes are kept in vegetative growth by frequent pruning.

Only the central, semi-hard part of the shoot is used. Semi-hardness is tested by hand: the test part of the shoot is held between fore and middle fingers and pressed with the thumb, and if the shoot bends with some pressure, it is semi-hard. A single-node cutting is made with a short slant cut close to the axillary bud, parallel to the mother leaf, leaving about 3 cm of stem below the leaf. Fresh cuttings are kept moist and shaded, never in the sun.

Leaf area matters: cuttings with a whole single leaf did best, cuttings with three-quarters of a leaf came next, and cuttings with a quarter or a half leaf struck poorly.

Mother bush: 50-300 cuttings a year

↓

Primary shoot, pencil thick, cut at 40 cm

↓

Single-node cutting: one whole leaf, active bud
About 3 cm of stem, slant cut at the bottom

Figure 8.1: Making a single-node tea cutting, from mother bush to cutting. Drawn from Ahuja et al., Science of Tea Technology, Ch. 4.

Quick check

How do you test whether a tea shoot is semi-hard, and which part of the shoot is used for cuttings?

Check your answer

Hold the shoot between fore and middle fingers and press with the thumb: if it bends with some pressure, it is semi-hard. Only the central, semi-hard part is used; the soft tip and hard base are rejected.

8.5 Planting, rooting and hardening

Cuttings are planted first in rooting beds. The bed is compacted and watered thoroughly the day before, and cuttings are set at 5 cm by 5 cm, pushed into the soil in one motion by holding the node between thumb and index finger, until the base of the leaf sits just above the ground. The mother leaf must stand upright and not sag on the soil: big or drooping leaves are cut to half their length. No water is given immediately after planting. The bed is then covered with a polythene sheet, sealed airtight with soil.

After rooting, cuttings are moved into sleeves. Rooted cuttings in bags of 2, 5 or 8 kg of dry soil grew best in the largest bags, and established better in the field.

Aftercare is steady work: keep the beds weed-free, spray insecticide when the cuttings carry 2-3 leaves, repeat monthly, and add copper fungicide if blister blight threatens. One warning: sprouting does not always mean rooting. Some clones sprout well but survive poorly.

When the shoots finish their first flush, the polythene sheet is loosened on one side. After the second flush, the second side is loosened. After the third flush, both sides are folded up, and after a few days the sheet comes off altogether. The plants are moved to the open field, and after 2-3 months of hardening they are ready for transplanting.

Quick check

A cutting has sprouted well in the nursery. Does that prove it has rooted? What does the book say?

Check your answer

No. Sprouting is not always accompanied by rooting: some clones sprout well but survive poorly, and nursery leaders may not hold their lead in the field.

8.6 Feeding, fungi and hormones

Phosphate is the fertiliser the book returns to most often for nursery plants. In one trial, phosphate at 0.15 g per kg of substrate during rooting, then 0.2 g of phosphate and 0.1 g of potash per kg of soil with nitrogen at 0.15 g applied three times as top dressing after transplanting, gave the best growth. In other trials, triple superphosphate or rock phosphate raised root dry matter.

Mycorrhizal fungi are the quiet helpers of the nursery. Seedlings inoculated with the arbuscular mycorrhizal fungus Scutellospora calospora grew greater biomass and better seedling quality than other mycorrhizal seedlings. In another study, Glomus fasciculatum improved rooting, shortened the nursery period, and raised leaf GABA content.

Hormones are a different story: the results are inconsistent and clone-specific. In one study, a 20-hour soak in IBA at 50 to 150 ppm gave 100% rooting after 4 months, while IAA, NAA and mixtures did worse. In another study, IBA and NAA gave lower rooting than untreated cuttings. In a third study, IBA at 8,000 ppm gave the best rooting, but IBA with NAA worked against it. The lesson is plain: there is no hormone dose that works everywhere, and any dose should be tested on the clone in hand before it is used at scale.

Try it yourself

Draw one single-node cutting life-size on paper, with the leaf, the bud and the 3 cm stem, and label the slant cut at the bottom.

Quick check

Why does the book refuse to recommend one hormone dose for rooting tea cuttings?

Check your answer

Results are inconsistent and clone-specific: IBA gave 100% rooting in one study, worse than untreated cuttings in another, and different doses won in different studies.

8.7 Raising seedlings from seed

Seedlings start in the seed orchard, not the nursery. Tea is cross-pollinated, so seed from an open-pollinated bush is mixed in character. For uniformity, seed should come from an authentic biclonal seed orchard, a seed bari, where two chosen clones are the parents. The seed bari itself is covered in section 8.10 below.

Collected seed is stored cool and moist, then sieved. Only seed more than 12 mm in diameter is kept. The kept seed is soaked in water for 2-3 hours: sinkers are kept, floaters go back in moist soil for two days and are tested again. Floaters, diseased and damaged seed are discarded.

The hard seed coat needs cracking before sowing, a step called pre-germination. Healthy sinkers go into a pit 30 cm deep and 1.2 m wide, in a sunny, warm place near the nursery. The pit is filled with river sand to 20 cm, the seed spread in one layer, covered with 10 cm of sand and mulch, and watered often. Cracking starts within a week and goes on for up to a month. Pre-germinated seed is sown 15-20 cm apart and 1-2 cm deep in beds, with the micropyle pointing downwards, or sown directly in filled sleeves. No polythene cover is needed for seedlings.

Quick check

A batch of tea seed is soaked for 2-3 hours. Which seed is kept, which is tested again, and which is thrown away?

Check your answer

The sinkers are kept. The floaters are put back in moist soil for two days and tested again. Floaters that still float, and diseased or damaged seed, are discarded.

8.8 Centring: shaping the young bush

A single-stemmed nursery plant, once it reaches the field, is turned into a multi-branched, multi-stemmed bush by pruning, nipping and tipping. Doing that shaping while the plant is still in the nursery sleeve is called centring, and it aims to save time in the field by starting the bush shape early. The idea sounds attractive, but the book shows that practice disagrees about it.

Reports from Bangladesh favour centring. When seedlings were pruned, side branches, shoot length and weight, stem diameter and root weight all rose; pruning at 15 cm usually gave the best response, though weak seedlings were harmed.

In south India and Malawi, centring brought lateral branches but hurt the root and shoot system and cut field survival. After transplanting, the early advantage of more laterals in centred plants had disappeared within 6 months. Centring trades root against branch: a bushier top at the cost of a weaker root.

Quick check

Centring gives a nursery plant more side branches. Why might a grower still choose not to centre?

Check your answer

Centring can weaken the root and shoot system and cut field survival, and the early advantage in laterals had disappeared within 6 months after transplanting in one study.

8.9 Grafting: two plants in one

Grafting joins two plants into one composite plant: the shoot system of one clone on the root system of another. The usual aim is a high-yielding, high-quality shoot on a hardy root, most often a drought-tolerant rootstock. For nursery grafting, a single-node scion is cleft-grafted onto a single-node unrooted rootstock cutting. The union is wrapped tightly with polyethylene tape, and the composite cutting is planted for rooting in the usual way, under a polythene tent. After 4-5 months the tent is removed gradually over 4 weeks, then the tapes. Success rates of up to 90% can be achieved.

Other methods were compared. Splice grafting gave only 1.5% survival after 6 months in one trial, while chip budding gave 81% after 6 months and 50% after 10 months. Camellia japonica can also serve as a rootstock, with 80-84% success against 90% in the control.

ROOTSTOCK
Single-node cutting of a drought-hardy clone
SCION
Single-node cutting of a high-yielding quality clone

↓

Composite cutting under a polythene tent: up to 90% success

Figure 8.2: Making a grafted tea plant, from scion and rootstock to one composite plant. Rootstock and scion are two parallel inputs joined into one plant, not successive steps. Drawn from Ahuja et al., Science of Tea Technology, Ch. 4.

Grafted plants should not be transplanted until 18 months old, though some combinations are ready at 9 months. Their survival, growth, drought tolerance, recovery from pruning and leaf yield were all better than own-rooted plants, with no loss in quality. The quality of the made tea follows the scion, not the rootstock.

Quick check

What does the scion contribute to a grafted tea plant, and what does the rootstock contribute?

Check your answer

The scion gives the shoot, yield and quality of the made tea. The rootstock gives the roots, hardiness, drought tolerance and field survival.

8.10 The seed orchard: biclonal and polyclonal seed

Seed orchards, called seed baris, are the factories of improved tea seed. Tea plants are set aside and allowed to grow into seed-bearing trees without pruning or plucking. In a seed bari, the selected plants are spaced 3-4 m by 3-4 m for the China type, and 5-6 m by 5-6 m for the Assam type, following their normal spread, and are usually allowed to grow indefinitely. Only the seeds that drop onto cleared ground are collected, every day.

Two kinds of seed come out of a seed bari. Biclonal seed comes from a cross between two superior cultivars, and polyclonal seed comes from a series of crosses with a number of superior cultivars. Biclonal seed has known parents on both sides, while polyclonal seed has mixed parentage and less predictable performance.

Quick check

How far apart are seed bearers planted in a seed bari, for China type and for Assam type?

Check your answer

China type: 3-4 m by 3-4 m. Assam type: 5-6 m by 5-6 m.

8.11 Flowering, pollination and fruit set

Flower buds develop from the axils of bud-scales in the lower part of the bush, and the whole process from bud to open flower generally takes 120-150 days. Flowering runs through the year near the equator, and narrows as latitude rises. Tea is appreciably self-sterile: seed set is better only after cross-pollination. The flowers are visited by bees, wasps and small flies. Few insects visit, partly because peak flowering does not match peak insect activity, and partly because the pollen is heavy, sticky and clumped. There are also too few insects for full pollination, so fruit set is poor despite an abundance of flowers.

The first sign of fruit set is the withering of the corolla and anther lobes, 24 to 48 hours after pollination. Fruit set is very poor, and the capsules drop continuously from flowering to maturity. From flowering to complete seed development takes about 12-18 months (about a year); the capsules reach physiological maturity in 10 to 11 months. A mature fruit is about 26.5 mm across and about 6.69 g in fresh weight, and usually holds 2-3 healthy seeds.

Quick check

Why is fruit set poor in tea even though the bushes carry an abundance of flowers?

Check your answer

The pollen is heavy, sticky and clumped, few insects visit, peak flowering does not match peak insect activity, and there are too few insects for full pollination. Tea is also self-sterile, so cross-pollination is needed.

8.12 Inside the seed: development and shape

A tea seed holds an embryo, nutritive tissue and a protective coat. At first the embryo is tiny, white and less than 0.1 mm across, wrapped in a brittle white coat and bathed in transparent liquid endosperm that fills 98% of the whole volume of the seed. As the embryo grows, the endosperm turns jelly-like and viscous, and by the time the embryo matures it fills only 50% of the seed volume. In the last phase the embryo grows at the expense of the endosperm and fills the whole seed, leaving only a thin, fine, papery layer stuck to the inner wall of the seed coat. The mature seed is non-endospermic: a linear, thick embryo guarded by a thin, brittle coat.

Growth regulators steer this development. Gibberellins abound in the liquid endosperm, peaking at the globular stage, and abscisic acid peaks at stage 8 of the embryo.

The mature seed is almost the size of a playing marble, dark brown to black, spheroidal, slightly flattened on the side of the hilum. The seed coat is hard and brittle. The linear embryo holds two large cotyledons and an embryonic axis. Seeds generally range from 10 to 22 mm in diameter, and within that range they can be sorted into grades.

Young seed: embryo under 0.1 mm, endosperm 98%

↓

Growing seed: endosperm falls to 50%

↓

Mature seed: embryo fills the whole seed, 10-22 mm across

Figure 8.3: How a tea seed develops inside, as the endosperm shrinks from 98% to almost nothing. Drawn from Ahuja et al., Science of Tea Technology, Ch. 5.

Quick check

A mature tea seed is cut open. What will you find inside, and what has happened to the endosperm?

Check your answer

A thick linear embryo with two large cotyledons. The endosperm, which filled 98% of the young seed, has been consumed: the mature seed is non-endospermic.

8.13 Germination: how the seedling starts

Almost 100% of tea seeds are viable and can germinate in large numbers, provided they hold about 30.6% moisture on a fresh weight basis. But tea seeds are recalcitrant: they lose viability as fast as they lose moisture, which is the subject of section 8.14.

Germination in tea is hypogeal and takes about 2-3 weeks for the shoot to emerge. The optimum temperature for germination is 20-25°C. The seed coat acts as a mechanical barrier, so intact seed should be soaked in water for 24 hours before sowing in sand. One study recommends using only freshly harvested seed, soaking for 24 hours, separating floaters from sinkers, and sowing the sinkers after the coats crack in a day or two.

Here the book disagrees with itself, and the reader should know it. Section 8.7, from Chapter 4, says to sow with the micropyle pointing downwards, so the young root goes straight into the soil. Chapter 5 says shoot emergence is early and the seedling is straight when seed is planted with the micropyle pointing upwards, and a third account in the same chapter says to sow with the hilum, the eye of the seed, horizontal, buried under 1 inch of moist, coarse vermiculite, and left under 80% shade for 1-2 months. The book gives no ruling between these three positions. The position of the seed changes the time the shoot takes to emerge, but not whether the seed germinates, so the practical rule is to follow one method consistently and watch the emergence time in your own nursery.

Quick check

What is the optimum temperature for tea seed germination, and how long does shoot emergence take?

Check your answer

The optimum temperature is 20-25°C, and the shoot takes about 2-3 weeks to emerge.

8.14 Why tea seed will not wait: recalcitrant storage

Tea seed was first placed in an intermediate category, sharing features of both recalcitrant and orthodox seed types, but later work showed it is truly recalcitrant: highly sensitive to drying and to chilling. The critical moisture content is high, 28-30%, and seed dried below it loses viability.

Drying damages the seed in a measurable way. Desiccation treatment cuts germination because hydrogen peroxide, H₂O₂, builds up in the embryos, and the damage is partly eased by a rise in ascorbic acid and catalase.

Storage is therefore difficult, and the book records a series of attempts. Seed stored at 7°C in polythene bags kept for more than three months. Treating seed with Jalshakti, a starch polymer, at 0.625 and 1.25% by weight and storing it in high-density polythene lined hessian bags gave about 88% germination even after 6 months.

Table 8.2: Tea seed storage methods reported in the book

MethodResult reported
0°C and 100% relative humidityStored for 7 months
Jalshakti at 0.625 and 1.25%, in polythene lined hessian bagsAbout 88% germination after 6 months
Moist sand with PEG at -5 bar, plus 0.05% mercuric chlorideAbout 70% germination after 7 months

Freezing the seed right down is harder still. Seeds exposed to liquid nitrogen were badly damaged by ice crystal formation and membrane injury, and neither cryoprotectants nor rehydration after retrieval helped. Encapsulated embryonic axes, dried for 1 hour and then cryopreserved, revived better on Murashige and Skoog medium, while non-encapsulated axes did not survive.

Quick check

Tea seed is recalcitrant. What does that mean for a nurseryman who wants to store seed for next season?

Check your answer

The seed cannot be dried below its critical moisture of 28-30% without losing viability, so it must be kept cool and moist, not dried like grain seed.

8.15 What else the seed is good for

Tea seed has uses beyond raising plants. In China, seed collected from wild populations is cold-extracted to give an edible oil and a residual cake or powder. The residue left after oil extraction holds saponins, and the cake or powder is used in organic farming as an insecticide or molluscicide in rice fields, fish culture and horticulture. It does not hamper plant growth and is a source of nutrients. The seed cake is also used as a fish poison, especially in shrimp farming, to clear unwanted fishes and harmful insects from the ponds.

Practical note: nursery seed must stay moist and alive, while oil seed is pressed and its cake is spread on fields. Keep the two lots apart and label them.

Key ideas to carry forward

  • Propagation aims at healthy plants in the shortest time with the fewest losses, and the nursery is the most vulnerable stage.
  • A nursery needs hot, humid air, sandy loam at pH 4.5-5.5, and steady water without waterlogging, under shade that cuts about half the light.
  • A single-node cutting with one whole leaf, from the semi-hard middle of a pruned mother bush, is the standard clonal unit.
  • Sprouting does not prove rooting, and centring buys branches at the cost of roots.
  • Grafting puts a quality shoot on a hardy root, with up to 90% nursery success, and the quality of the tea follows the scion.
  • Tea seed is recalcitrant: it must stay moist, near 30% moisture, and cannot be dried and stored like grain.
  • From flower to ripe seed takes 12-18 months (about a year), with capsules reaching physiological maturity in 10 to 11 months, and the mature seed is non-endospermic, 10-22 mm across.

Glossary

Mother bush

A bush selected to supply cuttings for vegetative propagation. Food tech example: a mother bush giving 50-300 cuttings in a year.

Multiplication plot

A plot of mother bushes of one clone. Food tech example: a multiplication plot holding at least 250 bushes.

Single-node cutting

A cutting holding one node, one leaf and one bud, the standard unit of tea propagation. Food tech example: a single-node cutting with about 3 cm of stem.

Semi-hard wood

Wood that bends with some pressure when held between finger and thumb, neither soft nor hard. Food tech example: the central part of a tea shoot.

Callus

Healing tissue that forms over the cut end of a cutting before roots appear. Food tech example: cuttings held in a callusing bed for 3-4 weeks.

Sleeve

An open-ended polythene bag used as a pot for a nursery plant. Food tech example: a cutting sleeve 10 cm by 25-30 cm.

Hardening

Exposing nursery plants to open conditions before field transplanting. Food tech example: loosening the polythene sheet after each flush.

Centring

Shaping a single-stemmed nursery plant into a multi-branched bush while still in the sleeve. Food tech example: a nursery plant pruned at 15 cm.

Scion

The shoot part of a graft, which becomes the top of the composite plant. Food tech example: a single-node scion of a quality clone.

Rootstock

The root part of a graft, which supplies the root system of the composite plant. Food tech example: a drought-hardy clone used as rootstock.

Composite plant

One plant made by grafting a scion onto a rootstock. Food tech example: a composite plant with a quality shoot on hardy roots.

Seed bari

A seed orchard where selected tea plants grow as trees for improved seed. Food tech example: a seed bari with bearers spaced 3-4 m apart.

Biclonal seed

Seed from a cross between two chosen clones, with known parentage on both sides. Food tech example: biclonal tea seed from a seed bari.

Hypogeal germination

Germination in which the cotyledons stay below the soil while the shoot emerges. Food tech example: tea seed germinating with the shoot emerging in 2-3 weeks.

Recalcitrant seed

Seed that loses viability when dried below a critical moisture content. Food tech example: tea seed with a critical moisture of 28-30%.

Endosperm

The nutritive tissue inside a seed that feeds the growing embryo. Food tech example: tea endosperm filling 98% of the young seed.

Saponin

A bitter compound in tea seed cake with insecticidal, molluscicidal and antiseptic effects. Food tech example: tea seed cake spread in rice fields.

Mycorrhiza

A fungus that lives with plant roots and helps the plant take up nutrients. Food tech example: inoculated nursery seedlings growing better than uninoculated ones.

References

  • Ahuja, P.S., Gulati, A., et al., eds., Science of Tea Technology. Chapter 4: Sud, R.K., Plant propagation.
  • Ahuja, P.S., Gulati, A., et al., eds., Science of Tea Technology. Chapter 5: Bhattacharya, A., Ogra, R.K., Prakash, O. and Shekhawat, R.S., Tea seed and its biology.

Unit 9: Young Tea, Pruning and Plucking

Note

Unit 9: Young Tea, Pruning and Plucking

Picture a tea plant left alone. It grows into a tree 8 to 12 metres tall, flowers, sets seed, and gives you almost nothing to drink. Now picture the same plant cut back, shaped into a low flat table, and picked every few days. It stays young, keeps pushing out soft shoots, and fills basket after basket. The difference between those two plants is this unit: raising young tea, pruning, and plucking.

🤔 Pause and think: the leaf you drink is the plant's food factory. Every shoot you pick is a leaf the plant loses. So how can picking leaves every week make a bush produce more leaf, not less? The answer sits in the leaves left behind, in sections 9.6 to 9.7.

9.1 Raising young tea: shaping the bush, mulch, roots and water

A newly planted tea plant is a single stem growing straight up, and a single stem gives a single line of shoots. The grower's first job is to turn it into a wide, many-branched bush with a flat top, because a wide flat bush carries far more plucking points than a tall thin plant.

Centering. When the young stem is 1.5 to 2.5 cm thick at the collar, the point where stem meets soil, the plant is cut across at 45 to 50 cm from the ground. Then the main stem is cut again, lower, at 15 to 23 cm, just above two or three side branches. That second, lower cut is the centering cut: it forces branching low down, so the bush spreads wide from near the ground. Later practice moved the first cut earlier: plants are often pruned when the stem reaches about 1 cm thick at the collar, cut across at 35 to 45 cm and centered at once between 10 and 20 cm, preferably above three side branches. Before cutting, a few plants are tested for root starch with a rapid iodine test (section 9.4): if the roots are short of starch and the plant is flushing, pruning waits.

How high to cut depends on the plant. Pruning height follows the length of the shoot joints, the internodes. China type plants and China hybrids have short internodes, so they can be pruned and centered lower than Assam type plants without losing branches, and tipping and plucking can start lower on them too. Getting young bushes into bearing early shortens the payback period for planting new ground or replanting old fields.

Other shaping routines. In East Africa, sleeve-grown plants are pruned at 15 cm when they reach 30 cm, side branches are cut at 28 cm when 1 cm thick, the bush is pruned again at 41 cm, then tipped at about 61 cm. In Sri Lanka the first prune comes at 23 cm about 9 months after planting, then at 35 to 38 cm, then tipping at 45 cm, and a final frame-forming prune is done at 45 cm in the seventh year. One prune alone does not make a well-formed bush. Bending and pegging bends the main stem and branches toward the ground and pins them there with pegs. Pegging works best when the main stem is 40 to 60 cm tall and no thicker than 8 to 10 mm at the collar: a thick stem snaps when bent. Trials found pegged bushes out-yielding pruned ones in the early years, but pegging needs regular watching and correction.

Tipping, timing and growth regulators. Tipping is done at 35 to 50 cm from the ground, depending on plant type, land shape and climate, to push branches to grow evenly across the frame. Prune in the dormant period or in cold weather, provided the soil holds some moisture, and never in hot, dry weather: during active growth the roots hold little carbohydrate, which may not be enough to back new shoots after pruning. Several plant growth regulators push branching by holding back apical dominance, the pull of the growing tip: ethephon (Ethrel), CCC (Cycocel), SADH (Alar), Goal (oxyfluorfen), morphactin and paclobutrazol. Paclobutrazol at 500 ppm increased side branches low on young tea, and polyamines at 350 ppm improved water use and drought tolerance.

Mulch, intercropping, roots and water. Covering the soil with cut grass, straw, pruned material, factory waste, peat or black plastic holds moisture, steadies soil temperature and stops weeds. Black plastic also gave complete weed control and held the root zone at 22 to 25 °C better than peat or factory waste. In one trial on steep land, clean weeding lost 89 tonnes of soil per hectare in the first year and 51 in the second, while Guatemala grass mulch lost 7 and 1 tonnes and Mana grass mulch lost 5 and 2 tonnes. In another trial, Guatemala grass mulch raised tea yield to 1,939 kg of made tea per hectare against 1,497 kg with no mulch. Mulch has limits: grass mulches used for four years made roots grow shallow, which left the tea more open to drought, and the yield gain over no mulch had almost vanished by the fifth year in one long trial. In the first two years the space between young tea rows can carry short-term crops such as pulses and grains, adding income, feeding the soil and holding weeds down, but after about three years the effect fades as the bushes close over. In young tea the fine white roots do most of the feeding, and between 74 and 94% of all roots sat within 90 cm of the surface, with a thick mass of fibrous roots in the top 30 cm. Deep hoeing and fork hoeing once or twice a year gave the highest average annual yield in one trial of young clonal tea, but the land must be levelled after hoeing, or the hollows hold water and invite root-rot pathogens. In a pot trial under glass, young tea wilted when soil moisture fell below 20%, and the plants died at 15% soil moisture: young tea has no deep root reserve to fall back on, so moisture in the top soil is its lifeline.

Quick check

Why is a young tea plant cut twice, first high and then low, instead of once?

Check your answer

The first cut stops the upright growth. The second, lower cut, called centering, is made just above two or three side branches so branching starts low down. A bush that branches low spreads wide from near the ground and carries more plucking points.

9.2 Pruning: what it is and why a tea bush needs it

Left alone, a tea plant grows into a tree 8 to 12 m tall in the Assam type, or a shrub 3 to 5 m tall in the China type, and spends its energy on flowers and fruit. A free-growing plant like that has a bigger leaf area and stores more dry matter than a managed bush of the same age, because its leaves catch light better. But it gives no commercial crop. To keep the plant in its leafy, vegetative phase and producing, its upright growth is held down by cutting the main stems and branches at a set height at set intervals. That operation is pruning.

Pruning holds the bush at a proper plucking height and keeps it able to push out active shoots, and it is repeated when the bush starts turning from leaf growth to flowering. After a prune, shoot growth is very active: shoot density is low but shoot size is large. With time the growth rate falls, shoot density rises, shoot size shrinks, and many shoots go banjhi, dormant, and fail to grow. At that stage the bush flowers heavily. That point arrives in 4 to 6 years depending on the bush type, and China type bushes flower earlier.

There is a second reason. With continuous plucking, the branch ends form many-pronged knots called crows' feet. Knots block the easy movement of sap, carrying nutrients up from the roots and sugars down from the maintenance leaves to the growing shoots, so shoot growth is held back. Pruning removes the crows' feet and lets the sap flow freely again. And even after pruning, the plucking table keeps rising as plucking goes on. A bush that grows too tall gives fewer shoots, and pickers work slower on it, so output and productivity fall. Pruning brings the height back down.

Pruning costs a crop in the year it is done, so it is sometimes called a necessary evil. It is also the second most labour-hungry job in tea, after plucking. A sound pruning policy has four aims, as set out by the Tocklai research station: keep the bushes healthy and feeding well so the maximum number of healthy shoots appears from the pruning sticks and most reach tipping height on time; harvest as even a crop as possible at acceptable quality all season; cut down pests and diseases; and keep bush height within limits for efficient plucking. Pruning, tipping and plucking work as one sequence, and the severity of one is balanced by the leniency of another. The eight objectives of pruning, in short: push vigorous leafy growth and hold back flowering; set the plucking table and hold a convenient height for picking; renew the maintenance foliage and keep enough of it; put the wood-to-shoot ratio right and remove knots and crossed branches; remove dead, diseased and worn-out wood and renew the frame; spread the crop evenly and avoid a rush in peak periods; hold quality by pushing active shoot growth and cutting down banjhi shoots; and hold back pests and diseases.

9.3 Types of pruning and skiffing

Three pruning systems are used at different times in a bush's life: formative prune, maintenance prune and corrective prune.

Formative prune. The formative prune, also called the frame-forming prune, trains the free-growing young plant into a table-topped bush. It cuts the upright growth, breaks the dominance of the leading shoot, and turns a single-stemmed plant into a many-branched, many-stemmed bush. The sequence is: thumb or lung prune the leading stem at 15 to 20 cm; decentre the leading stem at 8 to 15 cm, cut the side branches across at 30 cm, or peg them instead; tip the new shoots at 55 cm, pluck after that, and skiff to level the table in later years; cut across at 35 cm and remove crossed branches; then tip at 55 cm and keep plucking. Trials comparing cut-across pruning with lung pruning, where a leafy branch is left on the bush, found slightly higher yield with lung pruning, likely because the lung shoots kept feeding sugars to the plant.

Maintenance prune. This keeps the bush vigorous by removing the knots or crows' feet that form on first-order branches from repeated plucking, and by bringing down a table that has risen through 3 to 5 years of plucking. By severity it is light, medium or deep. Light prune: a uniform cut across, normally 3 to 5 cm above the previous light prune mark, done when the top hamper gets congested and shoots come small with more banjhi shoots. After the final frame-forming prune at 35 to 45 cm, three to four light prunes can follow at 60 to 65 cm on cycles of 3 to 5 years. For China type bushes the light prune sits at 40 to 50 cm. A thick, strong frame is a precondition: a thin weak frame cannot carry active shoot growth. Medium prune: done to renew the frame and cut the bush height, which should not pass 80 cm. The bush is cut below the first light prune level but above the original frame-forming height, usually at 45 to 60 cm, after 3 to 5 light prune cycles, that is at intervals of 12 to 18 years. Deep prune: a hard cut at the level of the frame nearest the ground, at or below 35 cm for Assam type bushes, and at or below ground level for China type bushes. Its severe form is called rejuvenation pruning or collar pruning, used when the root system is healthy but the top is damaged, knotty, thin or diseased, after about 20 years of normal pruning cycles. It can kill up to 25% of Assam type bushes, so it is done only after checking plant health and root starch reserve, and it is only a stopgap, not a substitute for replanting: a garden with more than 25% vacancies should be uprooted and replanted. For Assam bushes some leafy branches must be left at heavy pruning to serve as breathers or lungs, feeding sugars to the roots until new shoots grow; at least two lungs are recommended on the south-west side of the bush, for shade against sun scorch and feeding sugars to the roots.

Corrective prune. This sets up a top hamper of clean straight wood. The bush is cut 5 to 10 cm below the knotty hamper, and any knot below the general pruning level is removed on its own merits, at 55 to 65 cm in old sections or 35 to 45 cm in newly established tea. Because it removes the knotted frame and replaces it with new wood, it is also called rejuvenation prune when done low. A healthy root system and adequate root reserves are essential. It causes severe crop loss in the year of the prune, then a steady rise in productivity over time.

Skiffing. Skiffing is a lighter cut across the top of the bush. It levels the table, removes an undue rise in the plucking table during the season, and takes off knotted stems and crows' feet. It is also used to skip a prune that is due that year, or to stretch the gap between two prunes so the garden crops more. It comes in four grades, and the cut gets deeper as time passes since the last prune: deep skiff, midway between the current table and the last pruning level, clearing congestion and making most primaries fork; medium skiff, 15 cm above the pruning mark or 5 cm below the tipping level, removing the crows' feet at the top of the table; light skiff, a shallow cut in the red stems, removing old plucking points and overgrowth; and level-off skiff, a cut in the green stem made only to level the table for the first plucking. Earlier and deeper autumn skiffing brings the first plucking of the next year earlier, and lighter skiffing gives a larger total crop.

The pruning cycle: about 20 years, starting with the frame

1. Frame-forming prune: 35 to 45 cm
The height the whole cycle builds from

↓

2. Light prunes: 60 to 65 cm
3 to 4 prunes on cycles of 3 to 5 years, 3 to 5 cm above the last light prune mark. Light prunes sit HIGHER than the frame-forming prune.

↓

3. Medium prune: 45 to 60 cm
Below the first light prune level, after 3 to 5 light cycles, that is at intervals of 12 to 18 years

↓

4. Deep prune: at or below 35 cm (Assam type), at or below ground level (China type)
After about 20 years of normal pruning cycles, then the cycle starts again

This is a cycle over about 20 years, not a single descent: light prunes sit higher than the frame-forming prune.

Figure 9.1: The pruning cycle in order, from the frame-forming prune through light and medium prunes to the deep prune. Light prunes sit higher than the frame-forming prune, so the cycle is not a steady descent, and the deeper the cut, the longer the bush needs to recover. Drawn from Ahuja et al., Science of Tea Technology, Ch. 7.

Quick check

A bush has been light pruned four times and now stands over 80 cm tall with a knotty frame. Which prune comes next, and at what height?

Check your answer

A medium prune, usually at 45 to 60 cm, below the first light prune level but above the original frame-forming height. It renews the frame and brings the height back down.

9.4 Level, time and the starch reserve

Level. How low to prune depends on the existing height, the last pruning height, pests and diseases, the crop and market expected, and the vacancies in the field. In one ten-year trial, the lower the pruning height below 60 cm, the lower the yield: crop lost in the first two years from pruning at 15 and 30 cm, and in the first three years from pruning at ground level, was never recovered by the end of ten years. Deep pruning below 30 cm harmed bush survival and recovery and cut production the next season, while pruning at 40 cm or above showed no crop loss. So for healthy Assam type bushes the pruning level should be 40 to 45 cm, with deeper cuts kept for stems badly damaged by sun scorch, canker, fire or other causes. The time to bud break tells the same story: bushes pruned at 60 cm broke bud in 21 days, against 60 days for severe prunes. Severe pruning also harms root growth, both fibrous and main roots.

Time. Where winter dormancy rules, in regions away from 16 degrees north or south latitude, winter is the best pruning time: the plant is not flushing, so the cut costs little growth. Note: the book gives this latitude threshold differently in different chapters: Unit 10 §10.8 puts it at 19 degrees. Treat both 16 degrees and 19 degrees as approximate chapter figures, not as a single rule. Where there is no winter dormancy, near the equator to 16 degrees north or south, pruning should come just before the rush period, when starch reserves are high. After the rush the starch reserve is low, so pruning then should be avoided. The deeper the prune, the earlier it should be finished within the season. Shade helps recovery: bushes cut across in November or December recovered better and faster under shade than unshaded bushes.

The starch reserve. Root starch is what carries a pruned bush until new shoots grow. Starch is the main reserve sugar: root wood holds the most starch on a dry weight basis, and the pruned sticks themselves hold large amounts. On average 37% of the total reserve starch is used within one and a half months of pruning to make new growth. After pruning, root reserves fall fast, and rebuilding starts only after about two months; full rebuilding takes about 10 months. The lowest workable threshold of carbohydrate reserve, expressed as dextrose, was 11% of total dry matter. On average the reserve rose 2% with each 300 m of elevation, reaching 24% at 1,830 m. Resting the bush before pruning builds the reserve: one to two months of rest raised root reserves, and a rest of 90 days raised leaf and root total non-structural carbohydrate (TNC) to 294 and 230 g per kg respectively. The best resting period for higher TNC, regrowth and yield is between 45 and 90 days. Pruning without an adequate carbohydrate reserve can be disastrous, so the reserve is tested before the cut. Adequate soil moisture matters too, because water carries the starch from the root store to the growing points, and waterlogging slows starch use and recovery.

The iodine test. For a quick field test of starch, an iodine solution is made by dissolving 1 gram each of iodine and potassium iodide in distilled water and making the volume up to 100 ml. Tincture of iodine diluted with hot water 1 to 8 can be used instead. The solution is put on the cut ends of pencil-thick roots, and it turns blue. The depth of blue shows how much starch is there.

Table 9.1: Reading the iodine test on pruned tea roots

Iodine test resultWhat it means
Dark blueAdequate starch: safe to prune
Light blueMedium starch content
YellowVery low or no starch: do not prune yet

9.5 After the cut: care, cycles and old gardens

Post-pruning care. Three jobs follow every prune. Knife clean out cuts out the mature banjhi shoots, thin shoots and other unproductive shoots that sit under the maintenance foliage and never add to the crop, and repairs uneven cuts. Bush sanitation protects the open cuts: copper fungicide paste is put on the cuts, the frame is lime washed, and where moss or lichens infest the stems, micronised shell lime at 1 to 10 is applied. Pruning litter should be buried or laid between the rows: 100 kg of pruning litter equals 200 kg of farmyard manure in nutrient value. Keeping the litter, especially where no fertilizer is used, raises yields, and in the first two months after prunings are dug in they break down fast, raising nitrogen and potassium availability and organic matter in the soil.

What pruning does inside the plant, and pests. Pruning changes the chemistry of the bush, not just its shape. After pruning, a smaller share of carbon went to spring and summer shoots than in unpruned plants, and none at all to autumn shoots, while a large share went to stems and to polyphenols. Polyphenol content fell almost in a straight line from 17.6% in the first year to 11.1% in the fifth year after pruning, and fatty acids in young shoots also fell with time from pruning, which helps explain why flavour quality improves as the next prune approaches. Hydrogen cyanamide sprayed on pruned frames at 0.5% in one trial and 1 to 2% in another triggered early bud break and tipping and cut crop loss in the pruned year. Fresh growth after pruning is tender and open to attack, so pruning time is also a disease decision. Blister blight, caused by the fungus Exobasidium vexans, infects recovering bushes easily because the new leaves are more susceptible; avoiding shade and avoiding June pruning reduced it in one study. Low pruning beside bark injuries encourages Hypoxylon wood rot, whose inoculum lives in the soil. The shot-hole borer, which tunnels in stems and farms a fungus inside them, is held back by changing the pruning cycle and burning or burying the pruned material, and rejuvenation pruning can fight canker. Mites sit thickest on the bottom leaves of the bush and are denser in the third and fourth years of a cycle than in the first half, so a shorter cycle with removal of bottom leaves helps control them.

Vacancy infilling and the pruning cycle. After pruning, gaps are marked and filled with fast-growing cultivars, at the rate of 2n + 1, where n is the number of vacancies: three empty bushes call for seven infills. In one trial, infilling after rejuvenation pruning at 40 to 45 cm raised yield in the seventh year by 9% with no infilling, 47% with one plant per vacancy, 56% with two and 71% with three plants in staggered hedge planting. The interval in years between two prunes is the pruning cycle: a prune followed by skiffs in the following years until the next prune. Its length depends on the growth pattern, the cultivar, the climate, the quality and quantity of crop wanted, labour for plucking and market conditions, and it usually runs 2 to 6 years. The prune year always loses crop in that section, so the garden balances pruned against unpruned sections to keep the total crop steady. One long comparison of 2, 3, 4 and 6 year cycles over 12 years found the 4-year cycle gave the highest total yields. Quality cycles are short, such as light prune, deep skiff, unpruned; quantity cycles are longer, with more level-off and light skiffs between prunes; combined cycles mix the two. Even inside a 4-year cycle, swapping a deep skiff with a light skiff changed the cycle's productivity, so the order of operations matters, not just their number. In one study, yield rose through the first three years of a cycle and fell in the fourth, as both shoot number and shoot weight fell. The length of a cycle can be estimated from elevation:

Length of pruning cycle in years = elevation in thousand feet ÷ 1000, plus or minus 1

Assam type: add 1. China type: subtract 1. China hybrid: add 0.

Example: tea at 5,000 feet gives 5000 ÷ 1000 ± 1, that is a 6-year or 4-year cycle.

And the area to prune each year follows from the cycle length:

Area to be pruned (hectares) = total area (hectares) ÷ length of pruning cycle

Bringing an old garden back. Semi-hard pruning at 35 cm with mulching by the prunings beat both leaving the plantation unpruned and unplucked for one to two years and hard pruning with deep soil cultivation, on yield, quality, shoot regrowth, bush size and economics. For restoring an aged plantation the book's advice is: remove as much unproductive frame as possible by pruning low; clean out all diseased and damaged wood; set up conditions for quick regrowth of shoots and roots; and hold plant density by infilling. The plantation should rest at least two months, and fertilizer should go on five to six months ahead of the prune. Rejuvenation of a 65-year-old seedling plantation by normal cut-across pruning to 66 cm with tipping 10 cm above the prune gave significantly greater yields, and increases of two to four times in productivity from rejuvenation pruning, infilling and consolidation are recorded. But the warning is plain: rejuvenation and interplanting old tea with clonal plants can hold back decline and pay back early, yet they are not a substitute for uprooting and replanting with high-yielding clones, because an old plantation with infilling in unrehabilitated soil is unlikely to stay productive for long.

9.6 Plucking, tipping and shoot growth

Plucking is the harvesting of the growing shoot tips above a set plucking level on the bush. It is the most important job in tea growing, for six reasons: it is the harvest itself; it removes apical dominance and pushes new shoots to grow; it checks the upright growth of the bush and keeps it leafy; it decides the quantity and quality of the crop; it affects the health of the bush; and it is the costliest job, taking 70 to 80% of the total cost of green leaf production and 63% of all the field energy used.

A cultivated tea bush has a permanent frame of branches called pruning sticks. When pruned, the cut ends of the sticks lie in one flat plane. New shoots grow from dormant buds on the sticks: these are the primary shoots, or primaries. When a primary reaches a set height it is decapitated, tipped, and after tipping new shoots grow from the side buds on it. That regrowth is called regeneration, and the regenerated side shoots are plucked as crop when they grow above the tipping level.

Why leaves must be left behind. The leaves are the plant's food factory, and in tea the leaves are also the crop. That sets up a balance: some leaves are picked, some must stay to feed the bush. The leaves below the plucking table are the maintenance foliage, and the shoots above the table are the crop. Young leaves photosynthesise weakly. The relative photosynthetic rate of the first leaf, second leaf, third leaf and fourth leaf (the mother leaf) is 21%, 28%, 47% and 76% of a maintenance leaf. So most of the sugars the bush and its growing shoots need are made by the maintenance foliage, not by the young shoots themselves. Tracer studies showed the same story from the other side: the growing shoots are the strongest sinks on a plucked bush, and the relative sink strength of the bud, first leaf, second leaf and third leaf is 100%, 70%, 40% and 30%. Cut the maintenance foliage down and the whole bush suffers, and the crop of shoots falls. The top 10 cm layer of maintenance foliage, two to three leaves deep, contributes about 75% of the total crop, while the bottom 10 cm gives 20 to 25% and helps thicken the stems. A leaf on a tipped primary or on a plucked stub lives at most 18 months after full expansion, and its peak photosynthetic efficiency lasts about six months after full expansion, so maintenance foliage must be added to and renewed on a regular cycle.

What a new shoot looks like, and tipping. After plucking, the side buds on the stub swell and lengthen, then break and unfold two small scale leaves called janams, then a small round non-serrated leaf called the fish leaf, then normal serrated leaves one after another. Tipping is the decapitation of the growing primaries at a set height to set the plucking table, to hold the depth of maintenance foliage, to break apical dominance and wake the dormant side buds into pluckable shoots, and to spread the bush well. On a long trial of annually pruned tea, the lowest tipping measure of 10 cm gave the highest first yield, but the longest tipping measure of 20 cm gave the highest sustained yield. Work at Tocklai concluded that the ideal tipping height is the natural banjhi horizon, the common height of dormant shoots on a pruned bush. In a normal Assam tea bush that is 20 cm, equal to the average height of five leaves on a primary shoot. For China hybrid tea the height of five leaves is lower. Tipping low in the prune year, below the five-leaf measure, makes forks and knots form at the level of the next prune, and knots cut the frame's ability to crop. The five-leaf rule goes with hard plucking systems such as janam plucking. With lenient systems such as mother leaf plucking, the first tipping is done over two leaves, the fish leaf and the first leaf, and after every second plucking one leaf is added to the table. Tipping height also falls as pruning height rises: in one recommendation, pruning at 45 to 55 cm calls for four tiers of leaves, 55 to 60 cm for three tiers, 60 to 75 cm for two tiers, and 75 cm and above for one tier, for Assam tea. Tipping should start when the primaries have reached green semi-hard wood in at least 30% of the bushes and the tipping-in material is bigger than three leaves and a bud. If tipping is late, the primary stems harden and cannot be broken with the fingers.

From a plucked stub to a pluckable shoot

Plucked stub: side bud wakes and swells

↓

Two scale leaves unfold: the janams

↓

Small round leaf: the fish leaf

↓

Normal serrated leaves, one after another: bud, 1st leaf, 2nd leaf, 3rd leaf

↓

Pluckable shoot at 2 leaves and a bud; the cycle starts again

Figure 9.2: Regeneration of a tea shoot after plucking. Each pluck wakes a side bud, and the new shoot repeats the same sequence of janams, fish leaf and serrated leaves. Drawn from Ahuja et al., Science of Tea Technology, Ch. 8.

Quick check

A picker strips every leaf off a bush, leaving no maintenance foliage at all. The next flush is poor. Why?

Check your answer

The young shoots photosynthesise weakly: the first leaf manages only 21% of a maintenance leaf's rate. Without maintenance foliage there is almost no sugar being made, and the growing shoots are the strongest sinks on the bush, so they starve. The top 10 cm of maintenance foliage alone supplies about 75% of the crop.

Shoot growth, flushes and the banjhi shoot. Tea shoots come in two kinds. Aperiodic shoots are the fast-growing primaries from the pruned frame, which after tipping become the maintenance foliage. Periodic shoots grow from freely growing side buds and pass through alternating phases of active growth and dormancy. A shoot is dormant, or banjhi, when its terminal bud is less than half the length of the leaf below it. The growth a shoot makes between two dormant states is called a flush. The ratio of banjhi shoots to active shoots is the dormancy index: during the growing season a bush always carries some of each, and regular swings in the crop are thought to follow regular swings in the dormancy index. Short nitrogen, short water or other nutrients, hormone imbalance and weather can all push buds into dormancy, and the wood just below a banjhi tip is poorly developed, with weak water-carrying tissue. Commercial growing aims to raise active shoots and cut the banjhi share, because dormant shoots hurt both yield and quality. The shoot regeneration rate is the days a side bud on a plucked shoot needs to grow into a pluckable shoot. Reported times range from 36 to 95 days depending on bush type, climate and altitude, with faster regrowth, 28 to 32 days, in the warm wet months and 40 days or more in other months in one set of observations. The gap between the unfolding of two leaves in a row on a growing shoot is the leaf period, or phyllochron: reported values run from 3 to 10 days in warm conditions and stretch to 11 to 19 days in a cool dry season, and the leaf period lengthens with altitude as mean air temperature falls. Shoot growth rate is nearly proportional to temperature between 12.5 and 30 °C, provided soil moisture stays above 50% of field capacity, day length is more than 12 hours 15 minutes, and the afternoon vapour pressure deficit stays below 23 mbar. The heat a shoot needs to reach pluckable size is reported at about 108 to 481 degree days, varying with place and type. In a pruned bush the shoots are fewer but bigger and heavier, easy to pick and better in quality; in an unpruned bush the shoots are dense, light and small, and many fail and go banjhi, partly because crows' feet block water and nutrients and partly because the growing tips lose vigour with age. In the pruned year a pluckable shoot is usually twice as heavy as the same standard of shoot in an unpruned year.

9.7 Systems, standards and the plucking interval

Systems: which leaf stays behind. The plucking system decides which leaf is picked as crop and which leaf stays on the bush as the base for the next crop. A system that leaves new leaves on the table is lenient; one that leaves no new leaf behind is hard. Hard plucking can raise yield but harms bush health over time: it pushes shoot number up and shoot weight down. A lenient system does the opposite, and many shoots go dormant below the plucking surface. The growing bud and one-leaf-and-a-bud shoots are the strongest sinks, so they should stay on the table to pull extra photosynthesis out of the maintenance foliage. The three common systems are janam plucking, fish leaf plucking and mother leaf plucking. Janam plucking: shoots are plucked at the table down to the janam, the scale leaf, so the table does not creep upward. It suits places where shoots grow fast. Standard plucking takes all shoots at the table except very fine ones such as small one-leaf-and-a-bud shoots and unopened buds, at 5 to 9 day intervals. Black plucking takes all shoots and leaves only unopened buds behind: it is harder, done at 8 to 11 day intervals, and used as a short-term measure when plucking runs out of control for lack of labour against fast growth, during severe pest attack or blister blight, or when dormancy rises toward the end of a flush in unpruned tea. Fish leaf plucking: all pluckable shoots above the fish leaf are taken except immature one-leaf-and-a-bud shoots and unopened buds, and the fish leaf stays on the stub. It suits places where shoots grow more slowly. Mother leaf plucking: the most lenient system. Shoots are plucked above the mother leaf, the first leaf, and the table is stepped up leaf by leaf after every second plucking. In one comparison, black plucking gave 14 to 32% more production than standard plucking, and fish leaf plucking gave 12 to 15% more, while mother leaf plucking gave the lowest production, especially in the first two years. But fineness was highest with standard plucking except in the pruned year, when mother leaf plucking was finest.

Standards: how big a shoot is picked. The standard of plucking is the size of shoot taken as crop, and it decides both quantity and quality.

Table 9.2: The three standards of plucking

StandardWhat is pluckedEffect
FineOne leaf and a bud, two leaves and a bud, single soft banjhi shootsCrop falls sharply; quality very high, fetches a very high price; short intervals
StandardTwo leaves and a bud, large one leaf and a bud, single banjhi shoots; larger shoots broken backGood quality tea; a weekly interval suits it
CoarseTwo leaves and a bud and larger shoots, double banjhi shoots, no breaking backReached when the interval is long; yield rises, quality falls

Coarse plucking raises the cellulose, hemicellulose and lipid content of the shoot and lowers the soluble solids, while fine plucking gives the highest soluble solids. With coarse plucking the sum of Group I volatile flavour compounds, which give an inferior aroma, and C6 aldehydes and alcohols rises, the flavour index falls, and total theaflavins fall, with the galloylated theaflavins, especially theaflavin digallate, falling most. Yield and picker output rise as the standard drops from one leaf and a bud to three leaves and a bud, while total organoleptic scores fall, though the quality loss from a coarse standard varies by clone.

The plucking interval. Tea is harvested many times at gaps of some days, and the gap between two pluckings is the plucking interval, or plucking round. For the maximum number of two-leaves-and-a-bud shoots, one early suggestion was a round equal to the leaf period, since one-leaf-and-a-bud shoots are usually left on the table. That proved impractical for labour, so a general round of 7 days was adopted, though the suitable interval in practice runs from 7 to 14 days. Shorter rounds give more plucking points per unit area. As the interval lengthens from 6 to 22 days, fresh shoot weight rises from 0.65 to 0.89 g per shoot. A three-leaves-and-a-bud shoot carries 55% more dry matter than a two-leaves-and-a-bud shoot, gained over 3 to 5 days. So a heavier shoot can be harvested without losing fineness by setting the interval from the leaf period:

Plucking interval (days) = (2 × leaf period in days) − 1

Under continuous plucking the plucking level rises steadily, a creep of the table. The creep should not pass 5.6 cm in a pruned year and 4.0 cm in an unpruned year, or crop is lost. Shorter intervals give less creep, and the share of active shoots falls as the round lengthens. Interval and yield: the evidence points both ways. A 4-day round gave 16% more crop than a 6-day round in one study, and shorter rounds won in Central Africa, Mauritius and western Kenya, because each pluck breaks apical dominance and wakes new shoots. But in north-east India a long trial from 1978 to 1985 found the 13-day round gave the highest yield and the 5-day round the lowest, in the order 13, 9, 11, 7, 5 days. In Malawi, raising the interval from 7 to 10 or 11 days raised yield, holding at 14 days, then falling sharply at 21 days. In one study, stretching the interval from 3 to 17 days cut the number of rounds to one-sixth but made the shoots about nine times heavier, and yield rose by 50% in the unpruned year and 80% in the pruned year. The best interval depends on clone, place and season, and must be standardised locally. Interval, quality and economics: longer rounds cut fineness, because coarse overgrown shoots bigger than two leaves enter the crop and dormant shoots rise. With coarse plucking the share of bud, first leaf and second leaf falls from 83% with fine plucking to 43%. Polyphenol content falls from the bud down through the leaves to the stem, and caffeine is highest in the bud and lowest in the stem, so tender shoots from short intervals carry more of both, and more theaflavins. In one study the fineness of the crop fell from 77% to 49% in the unpruned year and from 87% to 35% in the pruned year as the interval rose from 3 to 17 days. Plucking takes around 60 to 85% of labour cost and about 25% of the cost of cultivation. A 14-day round cost about 64% of a 7-day round, but gross revenue was higher with the shorter round. So the interval is a three-way bargain between yield, quality and cost, and no single number suits every garden.

Try it yourself

Take one shoot and mark its leaves: bud, 1st, 2nd, 3rd. Lay out three baskets labelled fine, standard and coarse, and sort ten imaginary shoots into them using Table 9.2. Then ask: which basket would a tea taster pay most for, and which would a garden manager fill fastest? That tension is the whole of this section.

9.8 Harvesting a quality crop

Quality in tea comes from the garden, not the factory. The quality of made tea depends on the size of the shoots plucked, the share of active shoots in the harvest, the chemistry of the shoots, and how the shoots are handled after picking. A quality harvest needs three things: a bush carrying actively growing shoots with the lowest dormancy index, careful picking for the maximum fine shoot share and minimum loose leaves, and safe storage and transport with minimum physical damage.

Flushes, in general terms. A tea season divides into flushes, waves of shoot growth. The first flush is the quality flush of the whole season: shoots grow fast, and after unfolding two to three leaves they turn banjhi, so the flush comes into full bloom within two to three weeks. Start standard plucking early, even when only a few bushes are flushing, and repeat at 3 to 4 day intervals, then weekly; black plucking at 9 to 10 days is the fallback when labour is short and banjhi shoots are many. The main flush, in the warm wet months, is the bulk of the crop: shoots grow fastest, especially in pruned and deep skiffed sections. Standard plucking at 6 to 8 days in unpruned sections and 5 to 6 days in pruned or deep skiffed sections suits it, and black plucking is best avoided except in the same emergency. The backend flush, at the end of the season, slows down, shoots harden fast, banjhi shoots rise quickly and the table grows uneven: longer intervals suit it, black plucking at 10 to 12 days in unpruned sections and standard plucking at 7 to 9 days in pruned sections, with care not to damage the top layer of maintenance foliage. A shorter pruning cycle gives a higher share of active shoots, heavier shoots, lower shoot density and fewer banjhi shoots. Total crop may suffer against a longer cycle, but shoot quality and fineness improve. A 4-year cycle of light prune, level-off skiff, deep skiff and level-off skiff, pruning 25%, deep skiffing 25% and leaving 50% level-off skiffed or unpruned each year, was found better for both quality and yield in one comparison.

Feeding, shade, pests, picking and handling. Active shoot growth needs balanced feeding with nitrogen, phosphorus and potash, and foliar feeding with urea, zinc and magnesium can help shoot size and yield, while potash with the same nutrients at weekly intervals after plucking helps the bush through drought. Shade trees help quality shoots: they improve succulence, leaf area up to a limit and shoot number; they cool the microclimate and hold humidity, which suits both shoot growth and pickers; they set a temperature near the optimum for photosynthesis, so more sugars go to the pluckable shoots; they save soil moisture by cutting evaporation; and they add organic matter and nutrients to the soil. Too much shade is bad, especially in the wet months, when it favours blister blight. Thrips, aphids and mealy bug suck the sap of shoots in flush periods and cut growth, yield and quality, and blister blight in the rains distorts young shoots and spoils leaf style, briskness, brightness, colour and aroma in proportion to the disease. Weeds grow fast in the wet months and cut garden efficiency and quality when plucked by mistake with the tea. Sprays should not be used without care: integrated pest management is the aim, so that residues do not reach the made tea. For quality shoots the standard should be fine, aiming at two leaves and a bud, and coarse plucking should be avoided, with a minimum of 60% fine crop as the target. The interval should sit between 5 and 8 days depending on shoot growth. Stripping shoots and long intervals should be avoided. When labour is short, black plucking can stretch the interval without losing quality. Shoots grown past the two-leaves-and-a-bud size should be broken back and the unwanted growth discarded: it improves the harvest and protects the bush against grey blight. The plucked leaf should not be hard pressed during picking. It should be stored in a cool shady place, spread on a clean floor in layers no more than 15 to 20 cm thick, with loose leaf, coarse leaf and hard banjhi shoots sorted out, and turned often so it does not heat. It should reach the factory as soon as possible, loosely packed and never pressed in transport. Leaves should not be compressed at any stage and should never be allowed to heat up to 40 °C. Acceptable quality follows if leaves reach the factory within 5 hours of plucking and their temperature stays below 35 °C at all times.

Practical note

If you walk a tea garden, you can read its pruning and plucking history in the bushes. A flat table at waist height with clean straight wood below it means the cycle is on track. A table that has crept high, knots like fists on the branches, and many small shoots with tiny dormant buds means the prune is overdue. And a basket of leaf tells its own story: two leaves and a bud, soft and even, is a fine pluck; long stems, hard banjhi shoots and loose leaves mixed in is a coarse one, and no factory step can put back the quality that was lost at the picking.

Key ideas to carry forward

  • A young tea plant must be shaped, not left to grow: centering at 10 to 20 cm above three side branches, or bending and pegging, builds a wide low bush with many plucking points.
  • Mulch holds moisture, steadies soil temperature and cuts soil erosion sharply, but grass mulch can make roots grow shallow, and its yield benefit fades by the fifth year.
  • Young tea wilts below 20% soil moisture and dies at 15%: the top 30 cm of soil is its lifeline, holding most of its fine feeding roots.
  • Pruning holds a tree 8 to 12 m tall down to a table-topped bush, clears crows' feet that block sap flow, and must return every 4 to 6 years as the bush turns to flowering.
  • Light prune sits at 60 to 65 cm, medium at 45 to 60 cm after 3 to 5 light cycles, deep at or below 35 cm, and deep pruning can kill up to 25% of Assam type bushes.
  • Skiffing is the lighter cut in four grades, deep, medium, light and level-off, used to level the table and stretch the time between prunes.
  • A pruned bush lives off its root starch: about 37% of the reserve is spent within six weeks, rebuilding starts after two months and finishes in about 10 months, and the iodine test, dark blue, light blue or yellow, tells the grower whether it is safe to cut.
  • The pruning cycle runs 2 to 6 years and can be estimated from elevation: cycle length in years equals elevation in thousand feet divided by 1000, plus or minus 1, with Assam type adding 1 and China type subtracting 1.
  • Plucking is the costliest job in tea, 70 to 80% of green leaf cost, and it works only because maintenance foliage is left behind: young leaves photosynthesise at 21 to 76% of a maintenance leaf, and the top 10 cm of maintenance foliage supplies about 75% of the crop.
  • Tipping sets the table, ideally at the five-leaf measure, about 20 cm in Assam tea, and tipping below that in the prune year breeds knots at the next pruning level.
  • Janam, fish leaf and mother leaf plucking differ in which leaf stays behind, and fine, standard and coarse standards differ in shoot size: fine gives the best quality, coarse gives the most weight.
  • The plucking interval is a three-way bargain: longer rounds give heavier shoots and lower labour cost, but fineness, polyphenols, caffeine and theaflavins all fall, and the best interval must be set locally.
  • Quality comes from the garden: active shoots, a fine standard with at least 60% fine crop, a 5 to 8 day interval, and leaf at the factory within 5 hours below 35 °C.

Glossary

Centering

Cutting the young tea stem a second time, low down at 10 to 20 cm, just above two or three side branches, so branching starts near the ground. Food tech example: a centered bush spreads wide and low, carrying many plucking points within a picker's reach.

Bending and pegging

Bending the young stem and branches toward the ground and pinning them there with pegs, so new growth springs from the base and the bush spreads wide. Food tech example: a pegged young bush that flushes faster and covers the ground sooner than a pruned one.

Tipping

Cutting off the tip of a growing primary shoot at a set height to set the plucking table and wake side buds into pluckable shoots. Food tech example: tipping primaries at 55 cm to fix a flat table for picking.

Maintenance foliage

The mature leaves left below the plucking table that photosynthesise and feed the growing shoots. Food tech example: the top 10 cm of maintenance foliage supplying about 75% of the crop.

Crows' feet

Many-pronged knots that form on branch ends from repeated plucking and block the flow of sap. Food tech example: knots on an old bush that must be cut away at pruning so nutrients reach the new shoots.

Lung (breather)

A leafy branch deliberately left on the bush at heavy pruning to feed sugars to the roots until new shoots grow. Food tech example: two lung branches kept on the south-west side of a heavily pruned bush to shade the frame and feed the roots.

Skiffing

A light cut across the top of the bush to level the table, remove knots or stretch the time between prunes. Food tech example: a level-off skiff that evens the table before the first plucking of the season.

Pruning cycle

The interval in years between two prunes, made of one prune followed by skiffs until the next prune. Food tech example: a 4-year cycle of light prune, level-off skiff, deep skiff and level-off skiff.

Banjhi shoot

A dormant shoot whose terminal bud is less than half the length of the leaf below it, so it has stopped growing. Food tech example: a banjhi shoot in a basket of plucked leaf that lowers the fineness of the crop.

Flush

The growth a shoot makes between two dormant states, and by extension a wave of shoot growth across the garden. Food tech example: the first flush of the season, when shoots grow fast and turn banjhi after two to three leaves.

Dormancy index

The ratio of banjhi shoots to active shoots on a bush, used to judge whether the crop is flushing or stalling. Food tech example: a rising dormancy index warning the grower that the flush is ending.

Janam

The small scale leaves, also called birth scales, that unfold first when a new shoot grows after plucking. Food tech example: janam plucking, which cuts the shoot down to the janam so the table does not creep up.

Fish leaf

The small round non-serrated leaf that follows the janams on a new shoot. Food tech example: fish leaf plucking, which leaves the fish leaf on the stub when the shoot is picked.

Mother leaf

The first normal serrated leaf on a shoot, also called the fourth leaf counting from the bud. Food tech example: mother leaf plucking, the most lenient system, which picks above the mother leaf.

Leaf period (phyllochron)

The time between the unfolding of two leaves in a row on a growing shoot. Food tech example: a leaf period of 3 to 7 days in warm weather, used to set the plucking interval as twice the leaf period minus one.

Apical dominance

The pull of the growing tip that holds side buds dormant until the tip is removed. Food tech example: plucking a shoot tip wakes the side buds below it into new shoots.

Table creep

The steady rise of the plucking level as plucking goes on through the season. Food tech example: a table that has crept more than 5.6 cm in a pruned year, signalling crop loss ahead.

References

  • Ahuja, P. S., Gulati, Arvind, et al. Science of Tea Technology. Chapter 6, Young Tea Management; Chapter 7, Tea Pruning; Chapter 8, Tea Plucking.
  • Barua, D. N. Science and Practice in Tea Culture. Tea Research Association, Calcutta, 1989. Cited in Ahuja et al. for young tea raising, plucking physiology and pruning practice.
  • Eden, T. Tea. 3rd edition, Longmans, London, 1976. Cited in Ahuja et al. for carbohydrate reserves and elevation.
  • Tocklai Experimental Station, Tea Research Association. Annual Scientific Reports and field management notes. Cited in Ahuja et al. for pruning policy, tipping measures and plucking trials.

Unit 10: Feeding, Watering and Mechanising Tea

Note

Unit 10: Feeding, Watering and Mechanising Tea

A tea bush is picked every week, pruned every few years, and fed and watered all year. Every one of those jobs costs labour, and labour is the hardest thing to find on a tea garden. This unit is about the three answers the book gives to that problem: machines that do the heavy work, fertiliser that feeds the bush, and water managed on purpose instead of left to the rain. The three fit together: a machine-plucked bush needs the right food to recover, and food only works when the soil holds the right amount of water.

🤔 Pause and think: a machine can pick 200 to 300 kg of leaf in a day where a person picks 15 to 20 kg. So why is most tea in the world still picked by hand? Keep that question in mind through section 10.3: the answer is in the basket, not the engine.

10.1 Why machines: the labour problem and the five jobs

Tea growing is hungry for labour. Plucking alone takes 70 to 80% of the cost of green leaf production (Unit 9), pruning is the second most labour-hungry job, and spraying, weeding and fertilising all need hands too. As production has risen year after year, recruiting and keeping workers has grown harder: wages rise elsewhere, schooling rises, and garden work is rated a low-standard job. That squeeze is the reason mechanisation matters, and it matters most for the routine field jobs that repeat all year.

Five field operations can be mechanised: fertiliser application, chemical spraying, mechanical weed control, pruning and skiffing, and plucking. The simple machines built for these jobs have limited capacity and are restricted to one job each. The power sprayer is designed as an all-purpose machine for plantation crops, while plucking, skiffing and pruning machines are designed for tea only. This section follows the machines in the order a garden meets them: spraying first, then harvesting, then pruning and uprooting, then turning the waste into compost.

Quick check

Why did mechanisation become urgent in tea even as production kept rising?

Check your answer

Production rose, but workers became harder to recruit and keep: wages rose elsewhere, schooling rose, and garden work is rated a low-standard job. The same hands are needed for plucking, pruning, spraying and weeding, so machines are needed to cover the work.

10.2 Spraying: engine sprayers and aircraft

Engine-operated sprayers. These are usually portable single-block engines mounted on the tank and carried by hand. The engine drives the sprayer pump and builds enough pressure to spray insecticide and pesticide. Some sprayer pumps can be attached to the drive shaft of a tractor, power tiller or any movable engine, which builds the pressure for spraying chemicals. These pumps are also used in tea plantations.

Aeroplane and helicopter sprayers. A centrifugal pump forces the spray liquid through the nozzles, working under a pressure of 3.0 to 8.5 kg per cm². The pump is driven by a wind-driven propeller with four to six blades. To cut dead weight, plastic tanks store the liquid, with capacities from 450 to 2,200 litres depending on the size of the carrier. The system was used for applying fertiliser and chemicals on unshaded tea. It is expensive, although a large area can be covered in a short time, but the application efficiency is low compared with a self-propelled ground machine. Aerial application of major nutrients has been done in Papua New Guinea and Malaysia, and zinc has been sprayed from the air in Malawi, but the results of aerial pesticide application were not encouraging. Spraying from the air is a tool for covering ground fast, not for placing chemicals exactly.

10.3 Harvesting machines: shears to self-propelled

Shear harvesting. The shear harvester is a hand tool for holding a proper bush table level. It has 25 to 30 cm shear blades and weighs 700 to 800 g. It can be used from the tipping level to the production stage, and only during high cropping months. The whole available crop can be harvested effectively during the rush period. Plucking average can rise by 50 to 100%, and labour requirement can fall by 30 to 40% compared with hand plucking. A proper plucking table can be held through the whole pruning cycle. Since 70% of the total cost of production is the cost of leaf plucking, and crop productivity rises while labour availability falls, motorised mechanical harvesting is the best option the book offers.

Motorised harvesting. Motorised harvesters fall into four kinds: hand-held one-man machines, hand-held two-men machines, trailed machines driven by one or two men, and self-propelled machines. In a one-man machine a small motor drives the cutting action, and a collection bag sits behind the shearing blade, where the cut leaves are carried by a blast of air. It lifts the output of hand plucking, particularly when the flush is very high and not enough workers are available.

Early trials gave mixed results. One trial with an electrically driven portable motor found no significant change in yield or loss in quality. Others found a 30% fall in yield with a decline in quality and harm to bush vigour, and lower yield and quality than hand-plucked bushes: the machine leaf was low quality, holding immature and damaged leaves, and the unopened buds were damaged, cutting the quantity and quality of leaf. In one comparison the average yield of tea plucked by machine, shears and hand was 7,858.3 kg, 7,435.6 kg and 6,318.5 kg per hectare respectively: the machine gave the most leaf, but the best quality tea was hand plucked. With careful pruning and ideal cultural practices, the quality of machine or shear harvested tea can also be improved. In another comparison, only 28% of mechanically plucked leaf was good quality leaf, and a roll-type plucking machine damaged 30 to 50% of leaves in the first crop of long shoots and 50 to 80% in the third crop of short shoots. The damaged leaves cut the quality of the plucked leaf, although the roll-type machine gave equal or better quality than a shearing tool or a commercial cutting machine, because it left out more hard woody stem parts. A battery-operated machine harvested three times more shoots than hand plucking. The pattern is steady across the trials: machines pick more leaf faster, and hand picking gives the finer leaf.

Japanese models. The common single-man motorised harvester was developed by Kawasaki in Japan. It is a set of reciprocating shears on a frame, with the engine mounted on the back and connected to the shear by a flexible cable, turning rotary motion into reciprocating motion. The blade is 30 to 60 cm long. The leaf cut by the shear is carried by an air blast into a collecting bag. The engine is 27 cc, giving 1.75 HP at 6,000 rpm, and runs on a fuel and oil mixture at 25 to 1. The whole machine weighs about 9 kg. The two-men machines, models H120 and H140 from Kawasaki, have straight or curved blades. Each worker walks on one side of the row. Their engines are 40.6 and 41.5 cc, running on a petrol and 2T oil mixture at 25 to 1, with fuel use of 800 ml and 1,000 ml per hour, cutting spans of 120 cm and 140 cm, weights of 17 kg and 29 kg, and green leaf harvesting of 800 kg and 1,000 kg per hour. Each machine can harvest about 0.5 hectare per day. With a half-width cutter bar the machine is carried along the same row twice in different directions: lighter, but twice the time. A curved-blade two-men machine harvested 113 kg of medium-sized leaves in 4 hours in one trial. In another trial, machine plucking was fastest, with little difference between hand and shear plucking, and tea quality was best with hand plucking. At 21, 42 or 63 day intervals, the highest yield, 5,927 kg per hectare, came from hand plucking, but it was not significantly different from mechanical harvesting at 63-day intervals with a 1 cm table rise at each harvest. At 42, 56 or 70 day intervals, mechanical plucking gave double the yield and at least 35 times faster working than hand plucking.

Other machines. The Picky power harvester is battery-operated, with a rechargeable battery of 24 watts [book's wording; watts is a power rating, not a battery rating] and a shear blade of 27.5 by 25 cm. One battery set runs for 2.3 to 3.00 hours. The shear has two semi-circular knives and a rotation guide that pushes the leaf into a bag at the rear. The harvester weighs about 9 kg including battery and machine, its capacity is 100 bushes per hour. The Tal-Power mobile harvester from India comes in a single-man model, about 10 kg with a 1.8 HP engine at 6,200 rpm and a 45 cm blade, and a two-men model, about 18 kg with a centrally mounted engine and a 100 cm reciprocating blade, covering about 0.4 hectare and harvesting 400 kg of green leaf. A two-men machine developed by the Institute of Himalayan Bioresource Technology with the Central Mechanical Engineering Research Institute has an electronic start, fuel-efficient air-cooled 2-stroke 40.6 cc petrol engine giving 2 HP at 7,200 rpm, running on a petrol and 2T oil mixture at 25 to 1 at 800 ml per hour, with a self-sharpening blade, low vibration and noise, a 1,200 mm cutter bar, and a green leaf capacity of 800 kg per hour. The Agromatica DK harvester from Italy is pneumatically operated: an engine coupled with a compressor and five harvesting heads, each with reciprocating blades 30 cm long and a leaf tray 38 by 38 by 18 cm. Each head weighs 2.8 kg and the tray holds up to 1.5 kg of green leaf. The engine runs on diesel, giving 10 HP from a single-cylinder 406 cc motor. The compressor gives a pressure of 13 kg per cm² at the blower and 9 to 10 kg per cm² at the nozzles. The machine needs five workers to run it and two more to collect leaf and run the engine, and its capacity is 500 to 600 bushes per hour.

Trailed and self-propelled machines. In a trailed machine the cutting and collecting system sits on a light four-wheeled frame. It is not self-propelled, and the light assembly is pushed along the row by two people. These machines were built in Brazil and are used in South America and South Africa. Self-propelled machines have been developed and tried in many countries, but their use has been discontinued in several places. Early Soviet machines worked on slopes of 8 and 25 degrees, and later comparisons found the machines caused considerable damage to the bushes and gave lower yields than hand plucking, although one improved model gave higher yields than an older one. A large self-propelled machine in one report was giving a profit. A multipurpose tracked machine reduced ground pressure and could be used for herbicide and pesticide application, fertiliser application and cultivation. A half-track machine was used in Mauritius, and British manufacturers built the Spencer and Cab-Craft machines, used in Uganda and Kenya. In Kenya a Bobard Jeune Bob M55 harvester cut both yield and quality in varying degrees compared with hand plucking. The self-propelled route shows the same trade as the hand-held machines, only bigger: more leaf per hour, less fineness in the basket.

Table 10.1: Machine plucking against hand plucking (from the book, Tables 9.1 and 9.2)

MeasureOne-man machineTwo-men machineHand shearHand plucking
Plucking cost (Rs. per kg)2.00-2.500.96-1.452.35-2.734-5.30
Green leaf plucked (kg per manday)80-100200-30030-3515-20
Efficiency by weight of leaf plucked (%)500-5331300-1500175-200Control
Efficiency by cost of plucking (%)200-212365-400170-194Control
Time per hectare, lean flush (hours)20-254-5100Control
Time per hectare, full flush (hours)30-356-8125-140Control

Table 10.1 note: The book gives no hours for hand plucking here; hand plucking is the baseline (Control) against which the machines are compared.

The leaf itself tells the cost of speed. In mechanical harvesting the average composition of the leaf was: bud 3 to 5%, 1 leaf and bud 8 to 10%, 2 leaves and bud 28 to 30%, 3 leaves and bud 20 to 25%, 4 leaves and bud 12 to 16%, 5 leaves and bud 5 to 9%, cut leaf 8 to 12%, and fine leaf 42 to 52%. Fine leaf (42-52%) is not an extra share on top of the rest; it is the subtotal of the fine shoots (bud plus one- and two-leaf shoots) already counted above. That 8 to 12% of cut leaf, and the damaged buds and small shoots, is why machine-plucked leaf suits teas where fineness matters less, and why the book's chapter on plucking (Unit 9) ends by calling for selective machines that can leave unopened buds, small one-leaf-and-a-bud shoots and mature leaves behind.

Harvesting machines: five alternatives, ranked by output

Alternative 1: Hand plucking
15 to 20 kg per manday, finest leaf, slowest
Alternative 2: Hand shear
30 to 35 kg per manday, 25 to 30 cm blades, 700 to 800 g, fine leaf
Alternative 3: One-man motorised harvester
80 to 100 kg per manday, 27 cc engine, about 9 kg, mixed leaf quality
Alternative 4: Two-men motorised harvester
200 to 300 kg per manday, 800 to 1,000 kg per hour, 0.5 ha per day, mixed leaf quality
Alternative 5: Self-propelled machine
Fastest of all, coarsest leaf, most damage to bushes and buds, use discontinued in several places

These are five alternative machines, not five steps: no leaf passes from one machine to the next.

Figure 10.1: Five alternative harvesting machines, ranked by output per worker. Output per worker rises across the row, and leaf fineness falls with it: no leaf passes from one machine to the next. Drawn from Ahuja et al., Science of Tea Technology, Ch. 9.

Quick check

A two-men machine plucks 200 to 300 kg per manday against 15 to 20 kg by hand, yet the book still calls hand plucking the best quality. What does the machine put in the basket that the hand leaves behind?

Check your answer

Cut leaf, 8 to 12% of the machine harvest, plus damaged unopened buds, small one-leaf-and-a-bud shoots and mature leaves. The hand takes only the shoots at the standard; the machine cuts everything at the table level.

Try it yourself

Draw five baskets in a row: hand, shear, one-man machine, two-men machine, self-propelled. Under each, write its kg per manday from Table 10.1. Then write one word under each for leaf quality: finest, fine, mixed, mixed, coarsest. That row of baskets is the whole argument of this section.

10.4 Machines for pruning, skiffing, uprooting and compost

Mechanised pruning. Pruning is done once in four to five years, and it is recommended at that interval. Because workers are short, pruning is often delayed, and delayed pruning gives poor recovery of the bushes. Mechanising the job helps finish pruning on time. The Kaaz Java cutter is a shoulder-mounted double-handled machine with a two-cycle TL 43 engine of 42.7 cc displacement. The one-man machine has a shaft 1.54 m long, a chip saw blade on a spinning disc 25 cm across, a total weight of about 7 kg, a maximum speed of 8,500 rpm, and runs on a gasoline and 2T oil mixture at 25 to 1. It uses 800 ml of fuel per hour at optimum speed, and one worker can prune 850 bushes with it. Two one-man pruners with two-stroke motors were found satisfactory and economic for skiffing at 85 cm and above, and another model was effective and economic for skiffing at 75 cm and above.

The gain over hand work is large. With the one-man pruning machine a worker prunes 2,000 to 2,200 bushes a day at light prune, 1,400 to 1,600 at medium prune and 1,200 to 1,300 at deep or rejuvenation prune, against 100 to 120, 80 to 90 and 40 to 50 bushes a day by the conventional method. Fuel use is 9 to 10 litres per hectare for light prune, 12 to 15 for medium prune and 15 to 18 for deep or rejuvenation prune. The benefit-cost ratios against the conventional method are 3.5 for light prune, 3.24 for medium prune and 4.25 for deep or rejuvenation prune. Skiffing shows the same pattern: with the one-man machine a worker skiffs 6,000 bushes a day at level-off skiff, 4,000 at light skiff and 2,900 at medium skiff, against 800, 400 and 200 by hand; with the two-men machine the figures are 15,000, 10,600, 7,100 and 5,300 bushes a day for level-off, light, medium and deep skiff, against 800, 400, 200 and 100 by hand. Fuel use for skiffing is 5, 8 and 11 litres per hectare with the one-man machine for level-off, light and medium skiff, and 4, 6, 9 and 12 litres with the two-men machine for level-off, light, medium and deep skiff. The benefit-cost ratios for skiffing run from 1.06 to 1.48 with the one-man machine and 1.16 to 1.67 with the two-men machine, rising with the severity of the skiff.

Mechanical uprooting. Replanting is an important job for the long-term development of a garden and for raising productivity. But of late it has become hard to do, because manual workers are not available in time. To overcome this, JCB and Proclaim machines have been used successfully. The JCB can uproot 150 bushes per hour and the Proclaim 250 bushes per hour. The operating cost of the machine per hour is on a par with the cost of manual uprooting. The advantages of mechanical uprooting over manual uprooting are: the heavy-duty machines save time, and all the work can be finished in dry weather; soil erosion measures can be taken before the rainy season; the shortage of manual labour can be overcome; the soil can be loosened down to the root zone; and the roots can be removed completely. It can be used effectively on land with mild irregular slopes. One precaution is absolute: this work should never be done during the monsoon, as it could cause massive soil erosion. It should be backed by adequate soil erosion measures.

Bio-composting. In the present economic climate, low-yielding fields of tea become an economic burden. Rejuvenation or total replanting can in some cases reverse the economics, so chemical-free tea may be produced from these gardens. A lot of biomass is produced as pruning litter. It can be chopped and powdered by machine, which will raise the nitrogen content and also the water-holding capacity of the soil, and that may lead to higher yield and profitability. The prunings that leave the bush as waste come back to the soil as food: Unit 9 gave the exchange rate, 100 kg of pruning litter equals 200 kg of farmyard manure, and section 10.7 gives the nutrients in a hectare of prunings.

10.5 Feeding the bush: nutrients and what shortage looks like

Until the end of the nineteenth century, manures were seldom used in tea plantations in south-east Asia. Inorganic fertilisers were not tried in tea until the First World War. The first inorganic fertiliser used in tea was ammonium sulphate. Gradually potassium, calcium, phosphorus and magnesium became the important plant nutrients for tea plantations. Fertilisers can be an expensive input in tea production, accounting for up to 15 to 25% of expenditure, and productivity is linked to fertiliser input. Fertiliser recommendations aim at holding back the reproductive phase and pushing vegetative growth: the bush should grow leaves, not flowers and seed.

How well fertiliser works depends on climate (rainfall, temperature, relative humidity, solar radiation and day length), soil (pH, organic matter, moisture, texture, parent material, clay mineral and slope), the plant itself (root growth, leaf orientation, root cation exchange capacity, branching, spread and growth habit) and cultural practices (competition from weeds and shade trees, pruning, shade management, plucking style and intervals, rate, method, time and number of splits of fertiliser application, use of proper nutrient ratios and bush health). In south India, tea production rose from 685 kg per hectare in 1946 to 2,451 kg per hectare in 1991, despite only small increases in tea acreage. It is estimated that 20% of the increase in crop yields was due to good fertiliser use and the other 80% was due to improved crop management practices.

Deficiency symptoms. Removing nutrients with the crop rises as tea productivity rises. A plant also takes up a nutrient in large quantities, far above what it needs for growth, when the supply is plentiful. A shortage of one nutrient can look like a shortage of another, and multi-nutrient deficiencies have become common in tea gardens. A nutrient deficiency can be put right by applying the deficient element. The deficiency symptoms of the major plant nutrients and important trace elements, as given by Barua, are in Table 10.2.

Table 10.2: Deficiency symptoms of nutrient elements in tea (Barua 1989, as cited in the book)

ElementWhat a short plant looks like
NitrogenStunted growth; leaves pale yellow; lower leaves fall first, then upper leaves
PhosphorusMature leaves with smooth surface, at times slightly purple; leaves yellow at the tip
PotassiumFoliage with distinct marginal scorch; internodes longer with thin stem
CalciumMature leaf curves downward, tip shows marginal scorching; leaves drop and plant is stunted; lower leaf surface shows translucent patches turning into necrotic spots
MagnesiumMature leaf shows distinct inter-veinal chlorosis; defoliation starts from the lower leaves
SulphurDistinct chlorosis; young leaf turns pale green to yellow, but veins stay green against a pale-yellow background
BoronShoot apex dies; axillary buds abort and clusters of small shoots appear in the axils; leaf thick and dark green; upper side of leaf stalk shows cork formation spreading to main veins and laterals
ManganeseInter-veinal chlorosis; irregular mottling, veins and surrounding areas staying green; necrotic spots within the chlorotic patches near the leaf margin; severe deformation of young leaves and buds
ZincLeaf smaller, darker and sickle-shaped from unequal growth of the blade on either side of the midrib; internodes shorter

The nutrients one by one. Nitrogen drives yield, and the response to nitrogen is similar in all parts of the world: the yield gain falls as the rate rises, because other nutrients such as phosphorus run short. The acidifying effect of nitrogen fertilisers makes this worse and can also lead to manganese building up in the leaves on manganese-rich soils. In a trial with NPK 20:10:10, leaf potassium fell as nitrogen rates rose, even though potassium application also rose: high nitrogen rates either block potassium uptake or raise potassium leaching from the soil. At Tocklai, raising nitrogen from 100 to 300 kg per hectare raised nitrogen in the topsoil and in the plucked shoots, gave higher yield than the control in all years, and 200 kg of nitrogen per hectare was the optimum rate. In mature tea, the optimum rate of nitrogen was reported at 225 to 235 kg per hectare. In the western Himalaya, total yield rose linearly with nitrogen, giving 31.0, 41.5 and 70.3% yield increases at 90, 180 and 270 kg of nitrogen per hectare over the control. In one long trial, nitrogen at 250 kg per hectare gave an average yield as high as 3,691 kg per hectare, and the highest average yields reached 5,259 kg per hectare with nitrogen at 375 kg per hectare, although tea quality and value tended to fall as the nitrogen level rose. The common forms of nitrogen used in tea gardens are urea, ammonium sulphate and calcium ammonium nitrate. Studies in south India found ammonium sulphate gave better yields than calcium ammonium nitrate or urea; calcium ammonium nitrate gave good results as winter applications, and urea was generally the least satisfactory nitrogen source. In sand culture, growth and leaf colour were better with ammonium nitrogen than with nitrate nitrogen.

Phosphorus is needed in much smaller amounts than nitrogen and potassium, and the response to phosphatic fertiliser is more erratic. Phosphate fertiliser dug into the soil at planting time raised the growth rate of the plants, showing that tea in acid soils has a demand for phosphate. Of several application methods, broadcasting between the rows followed by digging into the soil was the most effective, with the best time 40 to 45 days before plucking. Placing phosphatic fertiliser at 5 to 10 or 15 to 25 cm depth gave better results than broadcasting with or without dibbling in, and rock phosphate was as effective as other forms of phosphatic fertiliser. In two cultivars, 50 kg of P₂O₅ per hectare raised nitrogen uptake as well as tea yield.

Potassium is used by the tea plant in large quantities. Responses to potassium fertiliser that improved the return from nitrogen fertiliser occurred where the soil pH is 4.9 or lower. Potassium below 1.75% in the first leaf with bud and 1.57% in the third leaf is a clear sign that lack of potassium is limiting yield. The level of potassium application should follow the crop yield, and the full benefit of potassium comes only when other nutrients, particularly phosphorus, are not limiting. In Georgia, potassium fertilisers in chloride and sulphate forms gave 21 to 29% higher yield over untreated controls. The contents of polyphenols, theaflavins and thearubigins, important quality components of brewed black tea, were also raised by potassium application. In a 5-year study with potassium oxide at levels up to 120 kg per hectare and nitrogen at 80 kg per hectare on old and mature tea, the response was small in the first two years, but yields then rose with potassium oxide at 70 or 90 kg per hectare. Tea shoots hold high concentrations of potassium, 2.51 to 2.71%: the potassium taken up in producing 1,000 kg of commercial tea is 26 to 27 kg, and the potassium in a cup of tea liquor is 48 to 53 mg.

Calcium in excess cuts potassium uptake and brings on potassium deficiency, with a consequent fall in crop yield. The effect builds up and can follow repeat applications of small amounts of calcium, such as are present in superphosphate, and potash fertiliser counteracts it. Calcium builds up in older tissues while magnesium builds up in younger tissues, and the calcium to magnesium ratio rises fast as tissues age. The quality of made tea fell as the calcium to magnesium ratio in the leaves rose. Shortage of either element, or an imbalance in their ratio, cut growth, yield and tea quality.

Magnesium significantly affected yield, with the highest yield, a mean of 2,901 kg per hectare, at 15 kg of magnesium per hectare. Quality parameters (theaflavin content, thearubigin content, brightness, total colour and the theaflavin to thearubigin ratio) rose with the magnesium rate up to 15 kg per hectare. Magnesium with potassium gave a synergistic effect on tea yield. Foliar application of magnesium fertiliser raised tea yield by 20.8%, and raised amino acids, caffeine and water extracts by 5.8, 9.2 and 2.3% over the control. Magnesium deficiency can come from a lack of soil magnesium, or from an excess of potassium despite adequate magnesium in the soil.

Sulphur removal with the harvested crop is small, only 10 to 15 kg even with a harvest of 3,000 kg of made tea per hectare per year. Out of nine sources of sulphur at 20 kg per hectare, gypsum gave the highest yield, followed by ammonium sulphate. Sulphur application improved tea quality as measured by theaflavin and thearubigin contents, brightness and tasters' scores, and some sulphur treatments raised caffeine in the shoots. Sulphur also improved plant metabolism, raised photosynthetic rate, chlorophyll content and total nitrogen and phosphorus contents, but held back potassium uptake.

Micronutrients. Zinc deficiency was first found in Sri Lanka, and within two decades zinc became an indispensable fertiliser for tea throughout the world. Aerial application of zinc oxide at 1.35 kg per hectare in October and February raised yield by 12%. A single application of copper sulphate at 2.0 kg per hectare in October raised yield by 5%, and by 14% when applied with zinc oxide. In Central Africa, zinc oxide in four equal splits of 1.25 kg per hectare each, in August, November, February and March, raised average yield from 2,915 to 3,366 kg per year, significantly better than two splits a year. Foliar sprays of 0.2 to 0.5% zinc sulphate and soil application of 100 kg per hectare were recommended. Manganese-rich leaves made poor quality tea, with higher polyphenol oxidase activity than normal leaves, and the older leaves held more manganese than the younger ones. Foliar application of manganese was more effective than soil application, and 10 kg of manganese per hectare per year gave the highest tea yield. Foliar application of 0.1% copper sulphate gave maximum activity of polyphenol oxidase and nitrate reductase in isolated tea shoots. Boron uptake fell as calcium application rose, and rose with boric acid up to 6 kg per hectare, higher with foliar application, although the highest spray concentration was toxic to the plant. Foliar application of boric acid up to 1 kg of boron per hectare per year raised yield by 12%, with no effect on tea quality. The tea plant takes up abnormally large amounts of aluminium without any apparent harm to the plant. Selenium-enriched green tea kept its colour better in storage, with higher sweetness and aroma and lower bitterness than normal tea, so selenium application can extend the shelf life of green tea. A chelate mixture of iron, manganese, zinc and molybdenum with basal fertiliser raised leaf yield by 19% in young tea over the control, and leaf quality improved in the year of application and the two following years, but both yield and quality fell in the fourth year.

Quick check

A bush shows pale yellow leaves, stunted growth, and the lower leaves falling first. A second bush shows distinct yellowing between the veins of mature leaves, starting from the lower leaves. Which nutrient is short in each case?

Check your answer

First bush: nitrogen, pale yellow leaves, stunting, lower leaves falling first. Second bush: magnesium, inter-veinal chlorosis on mature leaves with defoliation starting from the lower leaves. The two look alike at a glance, which is why the book warns that one shortage can look like another.

10.6 How much, when and how: rates, timing, method and quality

How much. Rates of application differ widely between areas, and no single dose suits every garden. The evidence in the book points to an optimum nitrogen rate near 200 kg per hectare in one long trial and 225 to 235 kg per hectare in mature tea, with the yield gain falling as the rate rises. Phosphorus is needed in much smaller amounts than nitrogen and potassium, and its response is erratic: it matters most where soil phosphorus is low. Potassium should follow the crop yield, with 70 or 90 kg of potassium oxide per hectare raising yields in a 5-year study, and a leaf potassium below 1.75% in the first leaf with bud is the sign that potassium is limiting yield. Magnesium gave its highest yield at 15 kg per hectare. Zinc oxide at 1.35 kg per hectare by air, or in four splits of 1.25 kg per hectare, raised yields by 12 to 15%. Boron at 1 kg per hectare per year by foliar spray raised yield by 12%. Manganese at 10 kg per hectare per year gave the highest yield. Sulphur at 20 kg per hectare, best as gypsum, raised yield and quality. The age of the plantation matters too: in one long study, phosphorus had an adverse effect on yield during the first eight cropping years, then improved yield but only up to 90 kg per hectare, while potassium helped yields in all years, and the highest yields came with nitrogen, phosphorus and potassium at 135:90:180 kg per hectare. The peak of productivity came in the 10th to 12th year and then declined: the fall was 51% in the 18th to 20th year on plots with no phosphorus and potassium, and only 9 to 14% on plots receiving phosphorus and potassium. In another study, mean annual made tea yield was higher for clonal tea than for seedling tea, yield rose with age within genotypes, and tea yield stayed unaffected by age of plantation up to 80 years from planting: improved genotype and judicious nitrogen application cut the chance of production falling with age.

Fertiliser and the cup. Feeding changes the tea as well as the yield. Nitrogen up to 200 kg per hectare raised yield and leaf tannin, caffeine and extractives, but higher rates harmed leaf quality. Nitrogen at 150 kg per hectare had no adverse effect on tea chemical composition, but 300, 500 or 700 kg per hectare cut catechin content. High rates of nitrogen fertiliser raised caffeine content and the compounds giving both inferior and superior flavour, but the flavour index, the ratio of superior to inferior flavour compounds, fell as nitrogen rates rose. Theaflavins generally rose while thearubigins fell with nitrogen rates, in a curved relationship. Foliar urea at 1% raised chlorophyll and amino acid contents in new leaf, and that showed in better made tea quality. Potassium and magnesium together raised free amino acids, polyphenols, caffeine, theaflavin and thearubigin contents, and potassium and magnesium applications raised the content of some important aromatic compounds, including nerolidol. The quality of marketed oolong tea was closely tied to its magnesium content. But potassium also cut theanine, and raising potassium from 45 to 180 kg per hectare cut theanine, glutamine, valine and total amino acids, cut reducing and non-reducing sugars and the theaflavin and thearubigin contents of the made tea, and raised fibre and protein in the shoot. Sulphur showed a linear positive link with theaflavin content and brightness of made tea. Long-term mineral fertilisation over more than 10 years harmed leaf mineral content and ratios and impaired leaf quality. The lesson is plain: more fertiliser is not more quality, and the dose that gives the most leaf is not always the dose that gives the best cup.

When to apply. Fertilisers should not be applied when the plant is leafless. No fertiliser should go on pruned bushes until the bushes have grown leaves. Fertilisers should go on moist soil: on dry soil the nutrients stay out of the plant's reach and nitrogen is lost as ammonia gas. Fertilisers should not be applied when heavy rain is expected.

How to apply. Manual broadcasting is the normal method in mature tea, with care for even distribution over the whole area including the spaces between rows. In young tea the fertiliser is applied in a circular basin around the plant, a method also used for wide-spaced mature tea and gardens with moderate vacancies. In young tea, care must be taken to keep a gap between the plant and the fertiliser strip, or the soft stems of young plants are likely to be scorched by contact with soluble salts such as ammonium sulphate and potassium chloride. On steep slopes in high-rainfall areas, placement of fertilisers should be used, to stop soluble fertilisers being lost by leaching.

Single or split application. Loss of nitrogen by volatilisation and leaching of other soluble fertilisers, particularly in high-rainfall areas, rises if the amount applied at one time is large. The danger of damage to feeding roots from concentrated salt solution also rises at high levels of application. So in some areas, applying fertilisers in divided doses gives better results, although the response to split application differed in different areas. For better yield, 60% of nitrogen, from a total of 300 kg per hectare, should go on in March as dry sulphate of ammonia dug 5 to 7 cm deep into the soil, and the remaining 40% in two doses of 60 plus 60 kg per hectare in June and July, or three doses of 40 plus 40 plus 40 kg per hectare in June, July and August, as urea with irrigation. Another recommendation is nitrogen at 300 kg per hectare as a split application, 60% in March to April and 40% in July. Split applications of ammonium sulphate or urea tended to raise leaf yields more than single applications in mature seed-propagated tea. Split application of fertilisers from summer to autumn raised nitrogen in mature leaf, stem and root, and then plant growth. Potassium oxide at 120 kg per hectare in four splits gave 4,034 kg of tea per hectare, against only 3,664 kg per hectare from putting all the potassium on before the monsoon, and a second trial also found four splits significantly better than one application or two splits. In contrast, one trial on mature clonal tea under rainfed conditions found urea at 275 kg of nitrogen per hectare as single, double and triple doses gave no significant differences in made tea quality or leaf nitrogen content. Splitting helps where leaching is heavy and matters less where it is not.

Pruning and fertiliser. Inadequate nitrogen and phosphorus before pruning harmed recovery of the bush. Phosphorus pentoxide and potassium oxide at 70 and 140 kg per hectare, with 110 kg of nitrogen per hectare, was most profitable in the pruning year, when yields reached 1,000 kg of made tea per hectare. But adding 120 kg of nitrogen, 100 kg of phosphorus pentoxide and 200 kg of potassium oxide per hectare in the pruning year, and 260 kg of nitrogen, 100 kg of phosphorus pentoxide and 200 kg of potassium oxide in the following year, gave 3,100 kg of made tea in the latter year. Pruning litter also feeds the bush: incorporating prunings into the soil raised nitrogen and potassium availability and organic matter in the surface (0 to 7.5 cm) and sub-surface (7.5 to 15.0 cm) soil, cut leaching of added nitrogen and potassium fertilisers, and adding organic matter as pruning litter during light prune and deep skiff gave the highest nitrogen contents in topsoil and plucked shoots. Nitrogen fertiliser also changes the roots: raising nitrogen from 100 to 300 kg per hectare pushed horizontal root spread toward the spaces between rows, mainly of fine roots, giving better bush growth, while potassium oxide at 100 to 200 kg per hectare supported conducting and fibrous root growth, although 300 kg per hectare of potassium oxide harmed root growth.

10.7 Reading the plant and the soil

Leaf analysis. Analysis of a leaf or any other plant part is a useful tool for judging the nutrient status of a tea bush. Nutrient composition can show deficiencies and give a first guide to fertiliser use. Mineral composition of the tea plant varies with elevation, plant type, age, leaf position, soil fertility, time and method of sampling, method of analysis, and genetic, environmental and cultural factors. Nitrogen, phosphorus and potassium are normally higher in young leaves than old leaves, manganese and calcium are higher in old leaves, and magnesium is similar at both ages, while copper, zinc and sodium show no clear trend with age. The mineral contents of the leaf track closely with those in the soil, so the mineral status of the crop can be judged from the available nutrients in the surface soil or sub-soil. Shoots on the well-lit parts of the bushes suit nutritional diagnosis better than mature leaves, and the best results come from sampling between 8 and 11 a.m. during late April and early May, late June and early July, and late August and early September. For judging magnesium need, chemical analysis of the mature leaf is preferred over the third leaf or bud, because it is a more reliable index. With leaf maturity, nitrogen, potassium and phosphorus in the leaf fall while calcium, manganese, iron and aluminium rise, so the first mature leaf is not an entirely reliable guide for confirming visual observations, particularly when magnesium deficiency is widespread. Weather also moves nutrients: high temperature and high evaporative demand help calcium uptake, while high relative humidity and high rainfall cut it; low temperature and high humidity cut manganese uptake; and high weekly evaporation, humidity and rainfall depress uptake of zinc, copper and iron into green tea shoots.

Soil testing. In precision agriculture, soil testing is an important diagnostic technique, used on soils before planting and then at regular intervals. Reading soil analysis data with the crop need, the physiological stage of the crop and the expected yield in mind helps move toward the best practices of scientific agriculture. For tracking nutrient changes in tea, soil and leaf samples should be taken in May, July and September. A positive link was found between ammonium nitrogen and hydrolysable nitrogen in the soil and total nitrogen in the leaf. Only a very low soil phosphorus content showed in an appreciable fall in leaf phosphorus, and high available potassium in the soil showed in high leaf potassium. Water-soluble soil potassium tracks the potassium need of the plant better than either exchangeable or total potassium. Intensive cultivation systems such as tea plantations need information on nutrient dynamics, the annual nutrient supply, and the amount of nutrients needed to hold a set production level, because soil fertility falls as crop productivity rises. Targeted yield equations for nitrogen and potassium have been built from soil test crop response studies using correlation and regression analysis. Statistical analysis of cumulative made tea yield showed the optimum ratio of nitrogen to potassium manuring would be 1 to 1, particularly when the rate of nitrogen is between 300 and 450 kg per hectare. At the same fertility status, the calculated nitrogen dose for a given target yield varies for different cultivars: one cultivar may respond to nitrogen more than another.

What prunings return to the soil. In tea plantations, fallen leaves and pruning litter left in place return substantial quantities of nutrients to the soil. About 30 tonnes per hectare of organic material is returned to the soil annually as prunings, made of 33% easily decomposed foliage with a carbon to nitrogen ratio of 15 to 1, and the rest woody matter with a 55 to 1 ratio.

Table 10.3: Nutrients in one hectare of prunings returned to the soil each year

Nutrientkg per hectare in prunings
Nitrogen (N)518
Phosphorus pentoxide (P₂O₅)142
Potassium oxide (K₂O)340
Calcium oxide (CaO)223
Magnesium oxide (MgO)60
Manganese oxide (MnO)58

Most of those nutrients sit in the foliage, which breaks down fast, so the prunings are not waste: they are the garden feeding itself. When plants are pruned, large shares of the nutrients locked up in the plant return to the soil, and some, particularly from leaves, can be used quickly and become available within a short time after pruning. A large uptake of potassium was observed, suggesting that current potassium application rates are inadequate on potassium-deficient soils. Mulching with plant material also changes nutrient uptake: tea mulched with Eragrostis curvula or Napier grass showed higher leaf phosphorus than unmulched controls.

10.8 Water: how much tea drinks, and getting it on and off the field

How much tea drinks. Tea is a perennial crop needing a high amount of water through the year, but in most areas the crop depends on rain water. Rainfall, which often falls most evenly in the rainy season, also leaves deficits in dry periods. Irrigation is essential not only as insurance against drought but also for producing a higher crop and for cutting seasonal dormancy during deficit and cold dry periods. Both over-irrigation and under-irrigation affect soil moisture and the environment, and the expected benefit can turn into a loss, so the right amount of water must go on at the right time. More than half of the tea area suffers from either waterlogging or drought, or both.

Well-distributed rainfall is essential for even crop production. In most tea areas 150 mm of rainfall per month will ensure continuous crop production, that is an annual rainfall of 1,800 mm. Unless supplementary irrigation is available, tea should not be grown where annual rain is less than 1,150 mm. In a dry area with a single rainy season, irrigation during the six-month dry season doubled the annual yield of tea and evened its distribution over the year to some extent. In general, total yields fall as irrigation gets less frequent and less water is applied, although tea can withstand a soil moisture depletion of 25% of available moisture, with yield falling considerably. The most economical use of water came when the soil was not filled to field capacity during irrigation, although yields were slightly less. Seedling tea bushes were generally less stressed than the clones raised from them by vegetative propagation during dry weather, and clones differed in their response to dry conditions. Water in the uppermost layer of soil matters most for production. Yield can rise by 9% if rainfall deficiency is replenished by irrigation during the dry months, and irrigation raised crop yield by 12.8 to 29.4% in one set of trials.

Winter dormancy and water. Tea plants growing within 19 degrees of the equator and south of it have no pronounced dormant period, but further from the equator a lull of about two to three months is experienced. Note: the book gives this latitude threshold differently in different chapters: Unit 9 §9.4 and Unit 13 §13.4 put it at about 16 degrees. Treat both 16 degrees and 19 degrees as approximate chapter figures, not as a single rule. Winter dormancy is due to short day length and low night temperature, which disturb the hormonal balance between growth-promoting and growth-retarding substances. Bud break resumes only after the balance is restored, either by the plant itself or by growth promoters such as gibberellic acid. Water alone cannot produce enough leaves in the dormant period for large-scale manufacture of tea. In one trial, the fresh weight of shoots was 372.7 g with irrigation and illumination, 244.5 g with irrigation alone, and 218.1 g with neither. When plants were kept cold at 5 ± 2 °C in summer, no flush was produced, while the same plants kept hot at 27 ± 2 °C in winter produced two flushes. Foliar sprays of Alar, Cycocel, ethephon and growth promoters such as gibberellic acid also helped give an early flush, and tea plants showed early bud break when night temperature was held at 20 ± 2 °C. So water, controlled temperature at 27 ± 2 °C, artificial illumination and growth promoters such as gibberellic acid can help bring an early bud break during the dormant period, but water on its own is not enough.

Measuring crop water need. Crop evapotranspiration, ETcrop, is generally estimated from field water balance or from soil water depletion, and it can be predicted for a given climate from weather data and crop factors. Various types and sizes of lysimeters measure ETcrop precisely, and weighing lysimeters give the most accurate measurement, including over short periods, although they cost a lot to install. Simple low-cost mechanical weighing lysimeters are preferred over hydraulic weighing types, which need elaborate equipment to keep the measurement accurate and remove temperature effects. The effect of climate on crop water requirement is given by the reference crop evapotranspiration, and there are several climate methods for calculating it: Blaney-Criddle, Thornthwaite, Radiation, Penman and Modified Penman. The Blaney-Criddle method links the water used by the crop to mean monthly temperature and daylight hours, with a locally developed crop coefficient. The Thornthwaite formula assumes an exponential link between mean monthly temperature and consumptive use. The Penman method combines energy balance and wind, and the Modified Penman method of Doorenbos and Pruitt, based on climate data, gives the best results, with a possible error of plus or minus 10% in summer and up to 20% under low evaporative conditions. The pan method comes next with a possible error of 15%, depending on where the pan sits. The radiation method can err by up to 20% in summer. The Blaney-Criddle method should only be used for periods of one month or longer in humid, windy and mid-latitude winter conditions, and it can over- or under-predict by about 25%.

In one measurement, skiffed tea transpired on average 2.74 cm more water per month from a 90 cm soil profile than pruned tea during the dry months of December to April. Soil moisture depletion over the period was 35 cm under pruned tea and 47.1 cm under skiffed tea, while total reference evapotranspiration over the same period was 43 cm by Penman's method. So for skiffed tea a crop factor of 1.09 was estimated, by dividing total soil moisture depletion by reference evapotranspiration. The crop factor rises steadily from the end of December as the crop grows, but falls as moisture depletion rises, and from the end of March it holds steady because of considerable rainfall. In a drought year, the net irrigation requirement, the gap between crop water need and effective rainfall, was highest in the dry months and fell to nearly nothing once the rains came.

Getting water on: irrigation methods. In tea gardens where the slope is mild, border irrigation can be used, while sprinkler irrigation is used for water application on undulating and high-slope land. Sprinkler systems are of two types by the way they spray water: rotating head systems and perforated pipe systems. By portability there are five types: portable, where the whole unit moves from field to field or pumping site; semi-portable, where all units move except the water source and pumping unit; semi-permanent, with portable laterals, permanent main lines and sub-main lines, and a stationary water source and pumping unit, with mains buried and risers with nozzles at set distances; solid set, where the laterals are laid in the field early in the crop season and stay for the season; and permanent, with permanently laid mains, sub-mains and laterals and a stationary water system, used in perennial crops such as tea.

The design of a sprinkler system uses a few working formulas from the book. The discharge needed from one sprinkler is Q = S₁ × Sₘ × I ÷ 360, where Q is the required discharge in litres per second, S₁ is the spacing of sprinklers along the laterals in metres, Sₘ is the spacing of laterals along the main in metres, and I is the optimum application rate in cm per hour. The capacity of the system is Q = A × d ÷ (F × H × E), where Q is the discharge capacity of the pump in litres per second, A is the area to be irrigated in hectares, d is the net depth of water application in cm, F is the number of days allowed to finish one irrigation, H is the number of operating hours per day, and E is the water application efficiency in percent. The discharge of a nozzle is q = C × a × √(2gh), where q is the nozzle discharge in m³ per second, a is the cross-section area of the nozzle in m², h is the pressure head at the nozzle, g is acceleration due to gravity, and C is the coefficient of discharge. The radius of the wetted area of a rotating head sprinkler is R = 1.35 × √(dh), where R is the radius in metres, d is the diameter of the nozzle in metres, and h is the pressure head at the nozzle in metres. The average rate of application of a single nozzle is Rₐ = q ÷ 360 × A, where Rₐ is the water application rate in cm per hour, q is the discharge in litres per second, and A is the wetted area in m². The uniformity coefficient, which measures how evenly a sprinkler spreads water, is Christiansen's Cu = 100 × (1 − ΣX ÷ mn), where m is the average value of all observations of application rate in mm, n is the total number of observation points, and X is the numerical deviation from the mean in mm. The hydraulic design of a sprinkler aims at uniform coverage, the desired rate of application, break-up of drops to protect the soil surface, efficiency to cut energy use, and maximum area covered.

Getting water off: drainage. Tea is a water-loving plant, but it cannot stand excess water, known as free water or gravitational water. It must be drained to create favourable soil conditions for growth. Removing excess water from the land surface is surface drainage; removing water that has saturated the pore spaces of the soil is sub-surface drainage. The excess water drained is gravitational water, which is not available to plants. Drainage does not remove capillary water, and there is no danger of over-draining the soil by placing the drains too close. Excess water can come from heavy precipitation, over-irrigation, seepage, or surface and sub-surface flow. Waterlogging is mainly a problem where the land slope is marginal and rainfall intensity is very high.

Surface drainage uses the potential energy of land elevation to give a hydraulic gradient for water movement. The system has three parts: a collection system, a disposal system and an outlet. Water in individual drains is collected, moved through the disposal system and drained off at the outlet. There are four types of surface drainage system: random field drains, parallel open drains, parallel open ditches, and bedding. Random field drains suit water collected in low spots in the field and run-off from adjoining higher ground, removed by random field ditches or surface drains. The parallel field drain system is the most effective method of surface drainage and suits drainage of irrigation and precipitation: field drains discharge into a lateral bordering the field, and the lateral in turn discharges into the main. The parallel open drainage system suits soils needing both surface and sub-surface drainage. The bedding type of drainage is a land-forming process.

The rate at which open drains must remove water depends on rainfall, size of drainage area, characteristics of the drainage area and the nature of the crop. The drainage coefficient is the depth of water in cm to be removed in 24 hours from the whole drainage area, expressed in cm per day. The intensity of rainfall and its duration are inversely proportional to the time allowed for removing the water. Two formulas from the book estimate design discharge: the Cypress Creek formula, Q = C × A^p, where Q is design discharge, C is a coefficient depending on the degree of protection, p is a coefficient, and A is area; and the Boston Society formula, Q = C × A^½, where Q is peak discharge, C is a coefficient and A is area. The section of a channel is designed using Manning's formula, and the section may not be the most efficient hydraulic section. Drainage sections are rectangular, trapezoidal or parabolic. Sub-surface drainage removes excess water at the root zone through drainage pipes, and is designed by the Hooghoudt equation. The layouts of drainage systems are random, herringbone, gridiron and interceptor types. In any drainage project a combination of surface and sub-surface methods is used, and a combination of open drains and tile drains drains an area effectively.

Water in, water out: the tea garden water balance

Rain falls: 150 mm a month keeps the crop going; below 1,150 mm a year, irrigation is needed

↓

Crop drinks it: skiffed tea used 47.1 cm in the dry months, pruned tea 35 cm

↓

Top up the gap: border irrigation on mild slopes, sprinklers on steep and uneven land

↓

Drain the excess: surface drains for free water on top, sub-surface pipes for water at the roots

Figure 10.2: Water in and water out of a tea garden. Rain and irrigation supply the crop, and drainage removes the free water the roots cannot stand. Drawn from Ahuja et al., Science of Tea Technology, Ch. 11.

Note: the book's own figures here do not exactly agree: 2.74 cm more per month over the five dry months December to April gives 13.7 cm, while the book's totals give 47.1 minus 35, that is 12.1 cm. Both figures are reported as printed.

Practical note

If you stand in a tea garden after rain, you can read its water story in the ground. Puddles sitting between the rows and yellow, drooping bushes mean the drains are not working, and the roots are sitting in free water they cannot use. Dry, cracked soil with wilted shoot tips means the irrigation gap was never filled. And a basket of machine-plucked leaf next to a basket of hand-plucked leaf tells the other story of this unit: one is full and fast, the other is fine and slow, and the garden has to choose between them every single round.

Key ideas to carry forward

  • Five field jobs can be mechanised: fertiliser application, chemical spraying, mechanical weed control, pruning and skiffing, and plucking.
  • Aerial spraying covers ground fast at 3.0 to 8.5 kg per cm² through 450 to 2,200 litre tanks, but places chemicals less exactly than ground machines.
  • Shear harvesting lifts plucking output by 50 to 100% and cuts labour by 30 to 40%; a two-men motorised harvester plucks 200 to 300 kg per manday against 15 to 20 kg by hand.
  • Machines pick more leaf faster, but hand picking gives the finer leaf: machine harvests carry 8 to 12% cut leaf plus damaged buds and small shoots.
  • A one-man pruning machine prunes 2,000 to 2,200 bushes a day at light prune against 100 to 120 by hand; JCB and Proclaim machines uproot 150 and 250 bushes per hour.
  • Never uproot during the monsoon: it can cause massive soil erosion.
  • Fertilisers are up to 15 to 25% of production expenditure, and recommendations aim at leafy growth, not flowers and seed.
  • Nitrogen near 200 kg per hectare is the optimum in one long trial, but the yield gain falls as the rate rises, and high nitrogen cuts the flavour index even as it raises caffeine.
  • Potassium below 1.75% in the first leaf with bud is the sign that potassium is limiting yield; magnesium gave its best yield at 15 kg per hectare.
  • Zinc, boron and manganese each raised yields by about 12% at the rates in this unit, and sulphur as gypsum raised yield and quality.
  • Do not fertilise leafless pruned bushes, dry soil or before heavy rain; keep a gap between young plants and the fertiliser strip; split applications where leaching is heavy.
  • Leaf analysis and soil testing diagnose feeding: nitrogen, phosphorus and potassium run higher in young leaves, calcium and manganese in old leaves, and the best nitrogen to potassium ratio is 1 to 1 at 300 to 450 kg of nitrogen per hectare.
  • Prunings return about 30 tonnes of organic material per hectare per year, holding 518 kg of nitrogen, 142 kg of P₂O₅ and 340 kg of K₂O.
  • Tea needs about 150 mm of rain a month, 1,800 mm a year, and should not be grown below 1,150 mm a year without irrigation; it can stand 25% depletion of available soil moisture, with yield falling.
  • Skiffed tea used 47.1 cm of soil water in the dry months against 35 cm for pruned tea; the Modified Penman method estimates crop water need within about 10% in summer.
  • Drainage removes free gravitational water the roots cannot use, through surface drains and sub-surface pipes, and never removes capillary water.

Glossary

Mechanisation

Doing field work with machines instead of by hand. Food tech example: a two-men harvester plucking 200 to 300 kg of leaf per manday against 15 to 20 kg by hand.

Shear harvesting

Cutting tea shoots with hand shears of 25 to 30 cm blades to hold the bush table level and harvest the rush crop. Food tech example: shear harvesting during high cropping months, lifting plucking output by 50 to 100%.

Cut leaf

Leaf sliced by a machine cutter instead of plucked whole, which does not roll properly in manufacture. Food tech example: 8 to 12% cut leaf in a machine harvest, against none in a hand pluck.

Skiffing machine

A motorised cutter that levels the bush table or removes knots, doing in a day what hand skiffing takes many days to do. Food tech example: a two-men skiffing machine clearing 15,000 bushes a day at level-off skiff.

Bio-composting

Chopping and powdering pruning litter by machine so it breaks down into compost that raises soil nitrogen and water-holding capacity. Food tech example: prunings powdered and returned to the soil instead of being burnt or carted away.

Macronutrient

A nutrient the plant needs in large amounts: nitrogen, phosphorus, potassium, calcium, magnesium and sulphur. Food tech example: nitrogen at about 200 kg per hectare as the main fertiliser dose for tea.

Micronutrient

A nutrient the plant needs in small amounts: zinc, manganese, copper, boron, aluminium and selenium. Food tech example: zinc oxide at 1.35 kg per hectare raising tea yield by 12%.

Chlorosis

Yellowing of leaf tissue from a shortage of chlorophyll, often from a nutrient deficiency. Food tech example: inter-veinal chlorosis on mature leaves pointing to magnesium shortage.

Necrotic spot

A dead patch on a leaf where tissue has died, often at the margin in nutrient shortage. Food tech example: necrotic spots near the leaf margin in manganese deficiency.

Marginal scorch

Browning and drying of the leaf edge, a classic sign of potassium shortage. Food tech example: distinct marginal scorch on foliage when potassium falls below the level the bush needs.

Chelate

A form of a trace element held in an organic cage so the plant can take it up more easily. Food tech example: a chelate mixture of iron, manganese, zinc and molybdenum raising young tea yield by 19%.

Split application

Dividing a fertiliser dose into two or more smaller doses through the season instead of putting it all on at once. Food tech example: nitrogen at 300 kg per hectare split 60% in March and 40% in July.

Broadcasting

Scattering fertiliser evenly by hand over the whole area, including the spaces between rows. Food tech example: manual broadcasting as the normal method of fertiliser application in mature tea.

Leaf analysis

Testing the mineral content of leaves to judge the nutrient status of the bush and guide fertiliser use. Food tech example: sampling well-lit shoots between 8 and 11 a.m. to diagnose magnesium need from the mature leaf.

Evapotranspiration

The water lost from a crop by evaporation from the soil and transpiration from the leaves. Food tech example: skiffed tea transpiring 2.74 cm more water per month than pruned tea in the dry season.

Crop factor

The ratio of a crop's water use to reference evapotranspiration, used to work out irrigation need. Food tech example: a crop factor of 1.09 for skiffed tea in the dry months.

Field capacity

The amount of water soil holds after free drainage has stopped, the full point for irrigation. Food tech example: tea withstanding 25% depletion of available moisture below field capacity, with yield falling.

Drainage coefficient

The depth of water in cm to be removed in 24 hours from a whole drainage area, in cm per day. Food tech example: sizing open drains from the drainage coefficient and the rainfall intensity.

Free water (gravitational water)

Excess water that drains freely through the soil and that roots cannot use, which must be removed by drainage. Food tech example: waterlogging where free water sits in the root zone after heavy rain.

References

  • Ahuja, P. S., Gulati, Arvind, et al. Science of Tea Technology. Chapter 9, Mechanization in Tea; Chapter 10, Plant Nutrition; Chapter 11, Water Management in Tea.
  • Barua, D. N. Science and Practice in Tea Culture. Tea Research Association, Calcutta, 1989. Cited in Ahuja et al. for deficiency symptoms and nitrogen response.
  • Doorenbos, J., and Pruitt, W. O. Crop Water Requirements. FAO, Rome, 1984. Cited in Ahuja et al. for the Modified Penman method.
  • Carr, M. K. V. The Climatic Requirements of Tea. Experimental Agriculture, 1972; and irrigation studies 1974-1977. Cited in Ahuja et al. for rainfall needs and irrigation response.

Unit 11: Protecting Tea: Diseases, Pests and Weeds

Note

Unit 11: Protecting Tea: Diseases, Pests and Weeds

A tea bush is a standing invitation. It keeps soft new shoots coming all year, it grows in warm wet air, and it stands in the same ground for decades. That is exactly what fungi, insects, mites and weeds want too.

🤔 Pause and think: a tea bush is plucked every week or so, and the plucked shoot is the very part that is eaten and sold. How does that short gap between spraying and picking change the way a grower must think about disease and pest control, compared with a crop like wheat that is harvested once a year?

11.1 A bush under attack from three sides

Tea grows in warm, humid country, from about 12 °C to 40 °C, with annual rainfall from 938 to 6,000 mm and relative humidity from 30 to 90%. It covers more than 2.71 million hectares in over 34 countries, and it is a perennial monoculture: the same bushes stand for decades, putting out soft new growth again and again. The book counts about 1,031 arthropod species associated with tea (the book gives 1,031 in one chapter and 1,034 in another: see Unit 12 §12.1), and close to 250 insect species attack it.

The diseases fall into three groups by the part of the bush they hit: leaf, stem and root diseases. Most are caused by fungi, with a few caused by bacteria, viruses and one alga. Leaves fall, bushes weaken, and whole bushes can die. Weeds steal water, nutrients and light, slow the pluckers, and shelter pests. Weed control is the second most expensive input in tea, after plucking.

Quick check

Why does a perennial monoculture like tea face more pest pressure than a crop that is replanted every year?

Check your answer

The bushes stay in place for decades and keep producing soft new shoots, so pests always have food and shelter in the same spot. There is no fallow year or change of crop to break their life cycle.

11.2 Leaf diseases: blister blight, the big one

Blister blight is the most important tea disease in the major tea growing countries. The fungus is Exobasidium vexans, a biotrophic pathogen that feeds on living tissue, and it attacks the tender shoots harvested for making tea.

An infected leaf tells its story plainly. Blisters form with a shiny, hollowed upper surface and a white, powdery, bulging lower surface. When blisters join, the leaf twists and curls. If the tender stem is infected, it withers and snaps off. Bushes just pruned suffer worst, because every leaf is young and tender, and dense shade in the rainy season encourages the disease.

The losses are heavy: reports quoted in the book put crop loss in unprotected areas at about one fifth to one quarter in one case, about one third in another, and as much as half the annual crop in a third. Tea from infected shoots loses leaf style, briskness, brightness, colour and aroma, and the shoots hold less sugar, nitrogen, amino acids, protein, polyphenols and catechin.

The disease moves fast. The fungus completes its life cycle in 11 to 28 days, so several generations run through one blister season, which usually matches the monsoon. Only basidiospores spread the infection. Overcast sky, low sunshine, mist and moderate temperature favour it, high humidity is needed for spore production and infection, and long leaf wetness is the key condition.

Wet weather: overcast sky, mist, dew, long leaf wetness

↓

Basidiospores released, mostly at night

↓

Tender shoot infected: blister forms in 11 to 28 days

↓

New blisters release spores: the cycle repeats through the season

Figure 11.1: The blister blight cycle. What it shows: wet weather starts the cycle, spores spread at night, and each round takes 11 to 28 days. Drawn from Ahuja et al., Science of Tea Technology, Ch. 12.

Control starts with spraying, but spraying alone is not the whole answer. Copper fungicides based on copper oxychloride or cuprous oxide, with 50 per cent metallic copper, worked best, yet as protectants they fail when the disease is already severe. Nickel chloride was once recommended, but it can scorch young growth, and the book notes it has been banned by the US EPA. The EBI fungicides tridemorph, bitertanol, cyproconazole, fenpropimorph, hexaconazole and propiconazole stop spores forming and cut their viability.

Weather can guide spraying: one rule says stop when the average relative humidity over three days stays below 83 per cent, and another says stop below 80 per cent over five days and restart above 80 per cent over three days. These rules did not work everywhere, because in a strong monsoon the weather stays so favourable that spraying can never safely be paused. In the field, thin out heavy shade and pluck hard for two to three months in the peak blister season.

Quick check

Why does blister blight hurt both yield and quality at the same time?

Check your answer

It attacks the tender shoots that are plucked for tea, so infected shoots are lost outright, and the shoots that survive carry less sugar, amino acids, polyphenols and catechin, which weakens the briskness, brightness, colour and aroma of the made tea.

11.3 Leaf diseases: black rot, grey blight and the rest

Blister blight is not the only leaf disease, and a grower needs to tell them apart by what the leaf looks like.

Black rot (Corticium theae and Corticium invisum) attacks mature leaves and stems, worst in badly ventilated patches where dead snags have not been cleaned out. It makes slightly raised, uneven patches, brown to chocolate brown and grey on top, brown or grey underneath. In wet weather infected leaves turn black as they rot. Dead leaves stay hanging on the bush, held by fungal cords, and the fungus carries over as sclerotia in bark cracks. Control means copper sprays, cleaning out dead twigs, and thinning shade.

Grey blight (Pestalotiopsis theae) has grown in importance with mechanical harvesting, because the fungus enters through wounds. The book puts the reported crop loss at about 17 per cent. It can be managed with mancozeb at 0.3 per cent, or carbendazim or thiophanate-methyl at 0.05 per cent, with the bushes drenched well. Biocontrol agents such as Trichoderma, Gliocladium and Pseudomonas with a food base also gave satisfactory control. Resistance is governed by two genes, Pl(1) and Pl(2).

Anthracnose (Colletotrichum theae-sinensis) makes reddish brown lesions, and diseased leaves usually fall; benomyl and thiophanate-methyl sprays were effective. Net blister blight (Exobasidium reticulatum) makes white net-like lesions under the leaf along the veins; in severe outbreaks leaves fall and leaf stalks die back, cutting the first crop of the next year. Bird's eye spot (Cercospora theae) attacks mature foliage, especially mother leaves. Systemic fungicides beat biocontrol at disease control, but yield rose more with biocontrol.

Smaller leaf troubles include brown blight, leaf scab, sooty mould, red rust (an alga), Phyllosticta leaf spot, bacterial shoot blight, bacterial canker, crown gall and the virus phloem necrosis. The book gives no control schedule for them, and this unit does not invent one.

Leaf diseases described in detail in the book

DiseaseOrganismWhat you see
Blister blightExobasidium vexansShiny hollow blisters above, white powdery bulge below; leaf twists; shoots snap
Black rotCorticium theae, C. invisumRaised brown to grey patches; leaves rot black in wet weather; dead leaves hang by fungal cords
Grey blightPestalotiopsis theaeGrey blight on leaves, entering through wounds; worse with mechanical harvesting
AnthracnoseColletotrichum theae-sinensisReddish brown lesions; leaves fall
Net blister blightExobasidium reticulatumWhite net-like lesions under the leaf along the veins; leaf stalks die back
Bird's eye spotCercospora theaeSpots on mature foliage, mother leaves and maintenance leaves

11.4 Stem diseases: cankers, rots and dieback

Stem diseases move slowly, and that is what makes them dangerous. A leaf disease shows itself in days. A stem disease can work for years before the bush dies. The pattern is the same across all six: the fungus needs a wound to get in, so the pruning cut is both the grower's main tool and the disease's main doorway.

Wood rot (Hypoxylon species) enters branches and stems of mostly old bushes through sun-scald injury. The wood turns dry, brittle and dead, and the fungus forms whitish-grey to dark crusts on the wood. Control is hands-on: prune out dead branches, paint large cut ends with copper oxychloride paste in raw linseed oil, and use rejuvenation pruning.

Branch canker (Macrophomina theicola) shows as slightly sunken patches on branches, ringed by callus growth; dry weather and poor soil favour it. Do not prune in dry weather, cut out cankered branches, and dress the cut ends. Collar canker (Phomopsis theae) strikes young plantations, 2 to 8 years old, killing the branch or the whole bush as the bark dies. Stem canker (Poria hypobrunnea) enters through pruning cuts, spreads slowly to the root, and kills the bush in about 8 to 15 years. Thorny stem blight (Tunstallia aculeata) enters through wounds and forms black thorny projections on the bark. Dieback (Leptothyrium theae) kills back branches in bushes recovering from pruning; copper paste on the cut ends right after pruning protects them.

Quick check

Why do so many stem diseases enter through pruning cuts, and what single habit cuts the risk for all of them?

Check your answer

Pruning opens fresh wounds in the wood, and the fungi enter through those wounds, through sun-scald cracks and torn branches. Dressing the cut ends promptly, and avoiding pruning in dry weather, closes the doorway for all of them.

11.5 Root diseases: the hidden killers

Root diseases are the hardest problem in the book: hidden killers, very hard to treat and to wipe out. They are common wherever tea has been planted after clearing forest. The fungi spread to healthy bushes by direct root contact with diseased wood or stumps left from clearing or uprooting, and by airborne spores.

The cruel part is the delay. A bush can carry root disease for 2 to 3 years with no sign above ground, while the infection quietly spreads to its neighbours. Then the bush dies suddenly, with the withered dry leaves still hanging on it. Depending on spacing, one to nine bushes may die in a single patch.

The book draws a line that decides the whole response. Primary root diseases kill the bush outright. Secondary root diseases attack bushes already weakened by poor drainage, low root starch, untimely pruning, over-plucking, or heavy leaf disease, and they hasten the death. Telling the two apart matters, because uprooting healthy-looking neighbours is only justified for primary root disease.

Their signs are read on the roots and the wood: charcoal stump rot (Ustulina deusta): white fan-shaped mycelium under the bark, double black lines in the wood, charcoal-like fruiting bodies at the collar; black root rot (Rosellinia arcuata): black woolly strands on the root surface, white star-shaped patches, black lines in the wood; brown root rot (Phellinus noxius): tawny brown mycelium under the bark, brown net-like lines, roots soft and spongy in advanced disease; red root rot (Poria hypolateritia): red to black strands on the root surface, red lines in the wood, roots crusted with soil; Armillaria root rot (Armillaria mellea): lengthwise split at the collar, white mycelium in bark cracks, black or brown ribbon rhizomorphs; violet root rot (Sphaerostilbe repens), secondary, favoured by waterlogging: enlarged root pores, inky black root tips, violet rhizomorphs; diplodia root disease (Lasiodiplodia theobromae), secondary: bluish wood, coal-black hairy cushions on the root surface, more common in sandy loam soils.

Infection starts in the roots: no sign above ground for 2 to 3 years

↓

Fungus spreads underground to neighbours: 1 to 9 bushes die in one patch

↓

Bush dies suddenly, dry leaves still hanging on it

↓

Diagnosis from the roots: mycelium colour, lines in the wood, rhizomorphs

Figure 11.2: Root disease as a hidden killer. What it shows: years of silent spread underground, then sudden death, then diagnosis from root signs. Drawn from Ahuja et al., Science of Tea Technology, Ch. 12.

11.6 Managing root disease: uproot, isolate, rebuild

There is no easy cure here. Soil fumigation with Vapam, metham-sodium and methyl bromide worked more or less well against the main root rot fungi, but the treatments are uneconomic, kill without selection, and harm useful soil microflora. Soil drenching with mancozeb or carbendazim controlled black root rot effectively in the trials the book cites.

The field measures are physical and slow: uproot the dead bush with a ring of apparently healthy neighbours, the suspects, and remove all wood debris; rest the area for 18 to 24 months under non-host plants such as Guatemala grass or Mana grass before replanting; dig a trench 45 cm wide and 120 cm deep around a large diseased pocket at once;

Biological control is the hopeful thread. In laboratory tests, Trichoderma harzianum was best against Poria and Armillaria, Trichoderma viride best against Rosellinia, and Gliocladium virens showed antibiosis against all the pathogens. But laboratory success against Armillaria did not repeat in living plants in one study, and field-scale results remain to be seen.

Try it yourself

Sketch the trench from section 11.6 on paper: 45 cm wide, 120 cm deep, ringing a diseased patch. Why must it include a ring of healthy-looking bushes? Use the 2 to 3 year silent period from section 11.5 in your answer.

11.7 Weeds: the second most expensive input

Weeds are plants in the wrong place at the wrong time, and in tea they are expensive twice over: they cost money to control, and they cost yield when left alone. The book ranks weed management as the second most expensive input in tea, after plucking. Grasses dominate the weed flora, followed by broad-leaf weeds, while sedges are not serious. Common weeds include Imperata cylindrica, Paspalum conjugatum, Cynodon dactylon, Borreria hispida, Eupatorium odoratum and Mikania micrantha.

Why do weeds get in so easily? First, the shape of the bush: China hybrid bushes carry many stems from the collar, and the soil pockets between them shelter weeds, especially where fertiliser has been wrongly placed in the centre of the bush frame. Second, wide spacing leaves open ground. Third, the pruning cycle: the heaviest infestation appears in pruned or deeply skiffed sections, where the ground lies exposed.

Timing matters most in young tea. The book calls April to September the critical period of weed competition in young tea: delay in weeding harms branching, growth and yield. In one study, the critical period was 8 to 16 weeks after planting, with the threshold at about 12 weeks. Weeds were estimated to cause yield losses of 10 to 25 per cent. In young tea, competition caused nearly 50 per cent fewer primary branches and a second-year yield about three and a half times lower. In one trial, weeds built up 13,090 kg of dry matter per hectare in the first year and 8,400 kg in the second, taking away 252 kg and 151 kg of nitrogen per hectare.

Weeds hurt in other ways too. The heavy crop period matches the period of active weed growth, so labour is pulled away from plucking just when it is needed most. In the nursery, weeds carrying root-knot nematodes (Meloidogyne incognita and M. javanica) can infect seedlings. Thorny weeds such as Bidens pilosa and Rubus species slow the pluckers, and weed shoots get harvested by mistake with tea shoots, harming made tea quality.

11.8 Weed control: chemicals, mulch and muscle

Weed control uses four tools: chemicals, physical and mechanical methods, biological methods, and cultural practice. The common herbicides are paraquat, glyphosate, simazine, 2,4-D, diuron, dalapon and MSMA. Even approved herbicides can damage tea if the dose is too high, the spray misses its target, spray drifts, or the bush is young. Pre-emergence herbicides stop weeds before they appear: in the nursery, simazine, atrazine, fluchloralin, oxadiazon and methazole were recommended at 2 kg per hectare in April, three weeks before planting cuttings. In young tea, oxyfluorfen at 0.125 kg per hectare before emergence, followed by oxyfluorfen at 0.06 kg with paraquat at 0.24 kg or 2,4-D at 0.8 kg per hectare after emergence, controlled most problem weeds. Dalapon and diuron are not recommended for tea younger than 3 years.

Glyphosate was rated the most effective herbicide in tea, at 1.5 to 6 kg per hectare, normally 1.5 to 2.5 litres per hectare (kg of active ingredient in trials; litres of formulated product in normal use). One study found it safe on bushes at 1.68 kg per hectare, but another found it could damage seed-raised China hybrid tea at 1.03 kg per hectare. 2,4-D controlled paraquat-resistant broad-leaf weeds at 1.12 kg per hectare. Paraquat is a fast contact herbicide, rain-fast, and repeated paraquat at 0.3 kg per hectare, weekly or fortnightly, can kill even persistent weeds. Prolonged use of one herbicide lets resistant weed species grow, so the book recommends rotating herbicides. The rates above are the rates reported in the book's trials, not a prescription for any garden. Even the spray water matters: glyphosate worked best with pond and rain water, 2,4-D best with tap, stream or pond water and worst with rain water, and paraquat best with tap water.

Black plastic sheet mulch suppressed weeds totally in the inter-rows of young tea on slopes. Rubber mulch raised mature tea yield by about 39 per cent. But deep hoeing on steep, high-rainfall ground causes erosion and damages feeding roots, which sit in the top 10 to 15 cm of soil, and rotary cultivation deeper than 3 to 5 cm is not recommended. Biological weed control has had little work. Clean weeding lost 89 and 51 tonnes of soil per hectare in the first and second years, against 7 and 1 tonnes with Guatemala grass mulch and 5 and 2 tonnes with Mana grass mulch. A soft weed cover left by selective weeding cuts erosion in high-rainfall areas. The book cites a benefit to cost ratio of 10 or more for herbicide weed control.

11.9 Insect pests: sucking pests and caterpillars

Tea pests are grouped by how they feed: sucking pests, caterpillars, beetles, soil pests and mites. Tea mosquito bug (Helopeltis theivora, H. antonii, H. schoutedeni) sucks sap from tender stems, young leaves and buds, leaving round marks that turn dark brown within hours. Damaged leaves curl upward and dry out, new shoots fail to start, and the crop can be lost entirely. A female lays 4 to 10 eggs a day, 220 to 224 in her lifetime, inside tender stems. Eggs hatch in 4 to 20 days, and the life cycle runs 12 to 40 days, peaking from June to September. Control begins without chemicals: pluck every 10 to 15 days to carry eggs away, pull out damaged buds, and regulate shade. In severe attack the book records thiamethoxam at 1 g per litre or profenofos at 2 ml per litre.

Mealybug (Nipaecoccus viridis) is covered in white powdery wax. Nymphs and adults suck sap on shoots, stems and branches, drying them. In severe cases they secrete honeydew, black sooty mould grows on it, and the plant can die. Control includes destroying ant colonies, removing alternate hosts such as hibiscus, okra, custard apple and guava, a strong jet of water, sticky bands, and predatory ladybirds, including Cryptolaemus montrouzieri. Thrips (Scirtothrips dorsalis) lay eggs inside young tissue, hatching after 7 to 8 days. Attacked leaves turn silvery underneath with black spots, shoots are stunted, and leaves fall. Control uses sticky traps, 10 to 15 per hectare, and neem sprays. Tea aphid (Toxoptera aurantii) forms dark colonies on tender shoots; feeding wrinkles the leaf into a boat shape, and honeydew leads to sooty mould. Scale insects feed in one place with mouthparts inside the plant, and their scale shields them from contact chemicals; start with scale-free planting material. Leafhoppers and the painted bug are listed among the sucking pests, but the chapter gives no damage or control detail for them.

Caterpillars. The tea tortricids (Homona coffearia, H. magnanima, Adoxophyes honmai) tie leaves into webs and feed on growing shoots; Homona coffearia is reported to cause about 50 per cent yield loss in one tea growing area. The leaf roller (Gracilaria theivora) rolls leaves and feeds inside. The red borer (Zeuzera coffeae) is recognised by sawdust under the bush; the larva, about 3.5 cm long, bores into young stems and tunnels down to the root over a 4 to 5 month larval stage. Control is to collect and destroy egg masses and larvae, cut and destroy borer-attacked branches, and release the egg parasitoid Trichogramma chilonis six times at 20,000 parasitoids per release at fortnightly intervals. Neem, profenofos and deltamethrin are recorded for caterpillars, and newer molecules are spinosad, novaluron and flubendamide, and acetamiprid and thiacloprid for sucking pests.

11.10 Mites, beetles and soil pests

Mites are the most serious pests of tea in almost all tea producing countries. The red spider mite, Oligonychus coffeae, first reported on tea in 1868, prefers the middle zone of the bush, about 30 cm below the plucking surface. It tears open leaf cells, making tiny reddish brown marks on mature leaves, and the book puts the crop loss at 17 to 46 per cent. It grows best at 30 °C, stops below 10 °C, and needs 232.6 degree-days to complete its life cycle. The scarlet mite (Brevipalpus phoenicis) gathers under mature leaves; infested leaves show brown scurf, then the stalk splits and the leaf falls, with losses of about 8 to 17 per cent. The pink mite (Acaphylla theae) damages only the top 10 to 15 cm of tender leaves, which turn pale, curl upward, then leathery and brown. The purple mite (Calacarus carinatus) turns mature leaves coppery brown. The broad mite (Polyphagotarsonemus latus) attacks the top young leaves and the bud, making them rough, brittle and curled downward, like herbicide injury. The Kanzawa spider mite (Tetranychus kanzawai) is a sixth mite pest listed in the book: a spider mite like the red spider mite, important in Japan, China, Taiwan and the Philippines. Control starts with shade and mulching to raise humidity, water sprays, and protecting natural enemies. Sprays recorded include azadirachtin, fenpyroximate, dicofol, fenazaquin, propargite and ethion.

Beetles. The shot hole borer (Xyleborus fornicatus), most serious of eight to nine scolytid species, reaches infestation of 91 to 100 per cent in one area reported in the book. Females bore galleries in stems, grow the ambrosia fungus Monacrosporium ambrosium inside, and branches break. Control is to prune attacked twigs 2 to 3 inches beyond the shot hole and burn them, and spray the fungus Beauveria bassiana at 5 g per litre. White grubs (Holotrichia species) are white, C-shaped, live 5 to 20 cm deep in the soil, and late grubs eat roots and cause wilting. Control is night collection of beetles, light traps at 2 to 3 per hectare, neem, and neem cake at 100 kg per acre. Termites are social insects whose workers feed on roots and stems and cause wilting and drying. Control is good water management, destroying mounds, and drenching soil with chlorpyrifos 40 EC at 4 ml per litre. The leaf miner mines young leaves in the nursery and after pruning; control is to destroy attacked leaves and spray neem or deltamethrin.

Four pest groups, acting at the same time in different places

Group 1 of 4: Sucking pests
Mosquito bug, mealybug, thrips, aphid, scale: sap feeders on shoots and leaves. Response: regular plucking, shade regulation, spray only in severe attack.
Group 2 of 4: Caterpillars
Tortricids, leaf roller, red borer: web, roll and bore into leaves and stems. Response: collect egg masses and larvae, cut and destroy attacked branches, release parasitoids.
Group 3 of 4: Beetles and soil pests
Shot hole borer, white grubs, termites: stems, roots and soil. Response: prune out attacked twigs and burn them, light traps, soil drenching where the book records it.
Group 4 of 4: Mites
Red spider, scarlet, pink, purple, broad, Kanzawa: cell contents of young and mature leaves. Response: shade and mulching to raise humidity, water sprays, protect natural enemies.

The four groups are parallel, not steps in a sequence: no pest turns into the next group.

Figure 11.3: The four pest groups of tea. What it shows: each group feeds in a different place and in a different way, so each needs a different response. The four groups attack at the same time in different places: they are parallel groups, not steps in a sequence. Drawn from Ahuja et al., Science of Tea Technology, Ch. 14.

11.11 Putting it together: integrated protection

No single method carries the garden. Combine cultural, physical, mechanical, biological and chemical control, chosen by what is attacking the bush.

Practical note

Every dose in this unit is a dose reported in the book's trials, not a recommendation. Check the current rules where you grow before using any chemical named here.

Key ideas to carry forward

  • Blister blight attacks tender shoots, runs an 11 to 28 day cycle in wet weather, and cuts both yield and cup quality.
  • Stem diseases enter through wounds, above all pruning cuts, and can take 8 to 15 years to kill a bush.
  • Root diseases are hidden killers: 2 to 3 years with no sign above ground, then sudden death.
  • Weeds are the second most expensive input: the critical window in young tea runs April to September.
  • One herbicide used without a break shifts the weed flora towards resistant species.
  • Pests feed in different places, so the response must match the feeding habit; mites are the most serious pests.
  • Integrated protection combines cultural, physical, mechanical, biological and chemical methods.

Glossary

Biotrophic pathogen

A pathogen that feeds on living plant tissue. Food tech example: blister blight fungus feeds on living tender tea shoots.

Basidiospore

The only spore the blister blight fungus makes. Food tech example: basidiospores released at night in wet weather start new blisters on tender shoots.

Canker

A dead, often sunken patch of bark and wood. Food tech example: branch canker leaves sunken patches ringed with callus on tea branches.

Primary root disease

A root disease that kills the bush outright. Food tech example: charcoal stump rot can kill a whole tea bush suddenly.

Secondary root disease

A root disease that attacks a bush already weakened by another cause. Food tech example: violet root rot attacks tea bushes weakened by waterlogging.

Critical period of weed competition

The window when weeds do the most harm. Food tech example: in young tea the critical period runs April to September.

Pre-emergence herbicide

A herbicide applied before weeds appear. Food tech example: simazine applied in April, three weeks before planting tea cuttings.

Integrated pest management

Combining cultural, physical, mechanical, biological and chemical control. Food tech example: against tea mosquito bug, regular plucking, shade regulation and natural enemies are used together, with a spray only in severe attack.

Sclerotium (plural: sclerotia)

A hard resting body of a fungus. Food tech example: black rot survives the winter as sclerotia tucked into bark cracks.

Degree-day

A measure of heat over time used to predict pest development. Food tech example: red spider mite needs 232.6 degree-days above 10 °C to complete its life cycle.

References

  • Ahuja, P. S., Gulati, A., et al. (eds.), Science of Tea Technology. Chapter 12, Disease Management in Tea, by Arvind Gulati and R. K. Sud; Chapter 13, Managing Weeds in Tea, by R. D. Singh; Chapter 14, Insect and Mite Pests of Tea and their Management, by Gireesh Nadda, S. G. Eswara Reddy and Adarsh Shanker.
  • Gulati, A. and Sud, R. K., Disease Management in Tea, in Ahuja et al., Science of Tea Technology, Ch. 12.
  • Singh, R. D., Managing Weeds in Tea, in Ahuja et al., Science of Tea Technology, Ch. 13.
  • Nadda, G., Reddy, S. G. E. and Shanker, A., Insect and Mite Pests of Tea and their Management, in Ahuja et al., Science of Tea Technology, Ch. 14.

Unit 12: Pesticide Residues and Organic Tea

Note

Unit 12: Pesticide Residues and Organic Tea

More than one and a half billion cups of tea are drunk every day around the world, and tea is the second most consumed drink after water. Every one of those cups began as a leaf in a garden where pests and diseases were fought, often with chemicals. This unit follows the chemical from the sprayer to the cup, then turns to organic tea farming, which tries to grow tea without those chemicals at all.

🤔 Pause and think: a cup of tea is mostly hot water poured over a few grams of dry leaf. Does a chemical on the leaf follow the water into the cup, or stay in the leaf?

12.1 Why residues matter in a cup of tea

Tea is a hard crop to keep clean. The book counts about 1,034 arthropod species and 82 nematode species active in tea fields (the book gives 1,034 in this chapter and 1,031 in another: see Unit 11 §11.1), and pest and disease problems cause 10 to 20 per cent of annual yield losses. Chemical pesticides have been the main defence in almost all tea growing countries for nearly three decades, but their heavy, careless use contaminates soil, water and air and worries consumers.

Tea has three features that make residues a special problem. First, it is perennial and mostly grown as a monoculture. Second, the surface area per unit weight of tea is large. Third, the gap between spraying and harvest is short, because shoots are plucked every week or so. At the same application rate, the residue level on tea is usually higher than on other crops.

That is why authorities around the world set maximum residue limits, or MRLs: the highest residue level allowed in the traded product. Zero residue is not practically possible while pests and diseases take 10 to 20 per cent of the crop. The message is that residues must stay far below the limits, and that growers must understand what happens to a chemical between the sprayer and the cup.

Quick check

Why does tea carry higher residues than other crops sprayed at the same rate?

Check your answer

Tea leaves have a large surface area for their weight, and shoots are plucked only days after spraying.

12.2 What happens to a spray on the bush

Most of a sprayed pesticide stays on the leaf and the shoot at first: that is the initial deposit. What happens next depends on the kind of chemical. Fat-loving pesticides, such as chlorinated hydrocarbons, synthetic pyrethroids and some organophosphates and carbamates, sit on the waxy layer of the leaf. Many systemic pesticides pass into the leaf quickly and are carried to all parts of the plant. Much of what is sprayed is lost in the field, and the loss follows first order kinetics: the amount falls by a steady fraction over time.

Five forces drive the loss. Growth dilution: young shoots grow fast, and the book estimates that the weight of all shoots rises about 13 times in a week and about 80 times in two weeks. As the bud grows into a pluckable shoot, the chemical is spread through much more tissue, so its concentration falls. Studies cited in the book put growth dilution at about 45 per cent of the total residue decline, and at around 40 to 45 per cent for organochlorine and pyrethroid pesticides. Note: the 13x figure is the book's estimate for the weight gain of all shoots over a week, while the 45% figure is the book's estimate of growth dilution's share of the total observed decline: the rest is photodegradation, evaporation, wash-off and breakdown. Neither number is changed here. Photodegradation: all pesticides break down in light to some degree. In experiments, quinalphos broke down fastest and cypermethrin slowest. Evaporation: for pesticides of similar molecular weight, the rate of loss depends mainly on vapour pressure, and evaporation is the main route of loss for deltamethrin and fenitrothion. pH: some organophosphates, such as chlorpyrifos, are stable in acid but break down in alkaline conditions, and its breakdown in water rose steadily as pH rose from 7 to 12 at 25 °C. Tea shoots are acidic, and pesticides are generally more stable at low pH, though fenazaquin breaks down sharply at low pH. Rainfall: rain shortly after spraying washes pesticide off the leaf, and the mechanical washing of rain matters more than the dissolving action of the water. For highly water-soluble pesticides such as dichlorvos, malathion and dimethoate, the wash-off rate was over 85 per cent under rain. What matters most is how soon after spraying the rain falls. Residues also fall faster in the wet season.

Five forces shrink the residue, all at the same time

Spray lands on the leaf: the initial deposit

↓

Force 1 of 5: Growth dilution
Shoots grow 13 times heavier in a week, spreading the chemical thin
Force 2 of 5: Photodegradation
Sunlight breaks bonds
Force 3 of 5: Evaporation
Vapour pressure lifts chemicals away
Force 4 of 5: pH
Acid leaf chemistry breaks some chemicals down further
Force 5 of 5: Rainfall
Rain washes chemicals off the leaf

↓

Residue falls by plucking

The five forces act in parallel, at the same time, not one after another.

Figure 12.1: The fate of a pesticide on the tea bush. What it shows: five forces act in parallel to shrink the residue between spraying and plucking. Drawn from Ahuja et al., Science of Tea Technology, Ch. 15.

Quick check

A grower sprays on Monday and it rains on Tuesday. Another grower sprays on Monday and it rains the following Monday. Who loses more residue to the rain, and why?

Check your answer

The first grower. Rain shortly after spraying washes the chemical off, and the washing effect fades with time, so the amount of rain matters less than how soon after spraying it falls.

A shoot weighs 1 unit on the day of spraying and 13 units a week later, and no chemical is lost at all. What happens to the residue concentration in that week? Now add the book's figure that growth dilution accounts for about 45 per cent of the real decline, and explain in one sentence why the real decline is faster than dilution alone.

12.3 How long residues last: half-lives

The half-life of a pesticide is the time for half of it to disappear from the leaf. The book gives a table of average half-lives on tea, and the spread is wide: from a fifth of a day to eight days.

Average half-life of some pesticides on tea

PesticideHalf-life (days)
Dichlorvos0.20
Phoxim0.27
Malathion0.50
Phoxim under shade0.87
Dimethoate0.96
Quinalphos1.20
Dimethoate + omethoate1.20
Phosmet1.20
Fenitrothion1.30
Bifenthrin2.50
Cypermethrin2.50
Permethrin3.20
Fenvalerate3.20
Deltamethrin3.20
Ethion3.40
Chlordimeform3.60
Dicofol3.90
Demeton8.00

Three patterns stand out. First, the organophosphates dimethoate, fenitrothion and quinalphos all fall below two days. Second, ethion at 3.4 days and dicofol at 3.9 days last rather longer. Third, the synthetic pyrethroids last a middling time, 2.5 to 3.2 days, but the book notes they are marked by very low toxicity to mammals. The book's headline finding is that no pesticide on tea shoots has a half-life longer than eight days, with demeton the longest at 8.00 days.

Take dimethoate, with a half-life of about one day. After one day, about half the residue is gone. After two days, about a quarter is left. After four days, about one sixteenth is left. The same wait means very different residues for different chemicals.

Quick check

Two pesticides are sprayed at the same dose. One has a half-life of 1 day, the other 4 days. After 4 days, roughly how much of each is left?

Check your answer

About one sixteenth of the first, after four half-lives, and about half of the second, after one half-life. The same wait gives very different residues, so the chemical's half-life decides the risk, not the calendar alone.

12.4 What processing does to residues

Green shoots are not drunk as they are picked. They go through withering, rolling, fermentation and drying to become black tea, or withering, rolling and drying to become green tea, and each step can cut the residue further. Drying runs above 100 °C, where heat breaks some pesticides down.

Drying removes water, so the dry leaf weighs much less than the green leaf, and the book puts the concentration factor at 4: in theory, the residue in the dry made tea could be four times the residue in the green shoot at harvest. In practice, the residue in made tea does not match that value, because so much is lost during processing. The book reports that 30 to 40 per cent of what remains is lost during drying, because drying drives off water and carries pesticide with it, and heat breaks chemicals down. The biggest single-step fall in the table below is at withering, not at drying. The higher the vapour pressure of a pesticide, the more of it is lost during processing.

The table below, from a study of orthodox black tea manufacture, shows the residue left at each step, with the green leaf set at 100. Dimethoate falls to about half at withering and to about 30 per cent at drying, while deltamethrin, which is far less water-soluble, stays near 84 per cent through withering, rolling and fermentation and still holds about 68 per cent after drying. Green tea manufacture shows the same pattern, with even lower residues left at the end: dimethoate falls to 23.4 per cent after drying, quinalphos to 28.6 per cent, dicofol to 36.6 per cent and deltamethrin to 56.2 per cent, from the same starting point of 100 in the green leaf.

Residue left during orthodox black tea manufacture (green leaf = 100)

PesticideWitheringRollingFermentationDrying
Dimethoate52.048.048.030.5
Quinalphos42.541.642.836.0
Dicofol66.064.064.948.0
Deltamethrin84.084.283.268.0
Green leaf: residue 100

↓

Withering: hot air, large weight loss, residue falls

↓

Rolling and fermentation: little further change

↓

Drying above 100 °C: further loss, heat breaks chemicals down, water vapour carries them off

Figure 12.2: Residue through black tea manufacture. What it shows: the residue falls most at withering, holds near-steady through rolling and fermentation, and falls again at drying, where about 30-40% of what remains is lost. Drawn from Ahuja et al., Science of Tea Technology, Ch. 15.

Quick check

Drying should concentrate the residue four times, because the leaf loses so much water. Why does the made tea end up with less residue than the green leaf, not more?

Check your answer

Drying removes the pesticide as well as the water: heat breaks chemicals down, and water vapour carries them off.

12.5 What reaches the cup: the infusion

Made tea is not eaten. It is infused in boiling water, and only the part that moves into the brew is actually drunk. The book reports that generally 3 to 25 per cent of pesticides move into the brew, and that at most about 40 per cent moves across, in the case of water-soluble compounds.

What decides how much moves across is mainly water solubility, with the partition coefficient, Kow, playing a part too. A large Kow means the chemical binds into the leaf tissue and stays out of the water. The table below shows the pattern clearly: dimethoate, soluble at 25,000 mg per litre, moves 25.8 per cent into the infusion, while dicofol, soluble at only 0.8 mg per litre, moves 0.10 per cent, and deltamethrin, soluble at 0.002 mg per litre, moves almost nothing. Phosphamidon, completely miscible with water, moves 33.0 per cent, the highest in the table. In this table the book gives no transfer value for deltamethrin and permethrin (marked '-'); cypermethrin is given as 0.06%.

Transfer of pesticides into tea brew, and the properties behind it

PesticideWater solubility (mg/L)KowTransfer to infusion (%)
PhosphamidonCompletely miscibleNot given33.0
Dimethoate25,0000.625.8
Monocrotophos10,0001.619.78
Malathion145.0156.012.14
Methyl parathion55.60447.49.96
Quinalphos22.01,2088.04
Chlorpyrifos2.016,255.53.14
EndosulfanInsolubleNot given1.80
EthionInsolubleNot given0.8
Dicofol0.843,954.00.10
Cypermethrin1.034,506.40.06
Permethrin0.234,434.2-
Deltamethrin0.00234,506.0Not given in the book

Brewing time matters too. Longer brewing moves more pesticide across, though the rise is small for chemicals that barely dissolve. At 2 minutes, phosphamidon moves 35.5 per cent, dimethoate 26.2 per cent and malathion 10.2 per cent; at 10 minutes those figures rise to 40.2, 32.5 and 13.8 per cent. For the pyrethroids the numbers stay near zero: deltamethrin moves 0.0 per cent at 2 minutes and 0.28 per cent at 10 minutes, cypermethrin 0.1 and 0.32 per cent, and permethrin 0.0 and 0.48 per cent. Longer brewing and continuous heating help release pesticides from the tissues. The book also notes that organophosphates lose 30 to 70 per cent through their high vapour pressure during brewing, so the hot cup itself drives some chemicals off.

Dry leaf: residue sits in the tissue

↓

Boiling water: water-soluble chemicals move into the brew, fat-loving chemicals stay in the leaf

↓

The cup: generally 3 to 25 per cent of the leaf residue, at most about 40 per cent

Figure 12.3: What reaches the cup. What it shows: solubility decides how much residue leaves the leaf, and most of it stays behind. Drawn from Ahuja et al., Science of Tea Technology, Ch. 15.

Quick check

A tea carries equal residues of dimethoate and dicofol on the dry leaf. Which one reaches the drinker in larger amount, and why?

Check your answer

Dimethoate. It dissolves at 25,000 mg per litre and moves 25.8 per cent into the brew, while dicofol dissolves at 0.8 mg per litre and moves only 0.10 per cent.

Brew two imaginary cups from the same leaf, one for 2 minutes and one for 10 minutes, using the book's figures for phosphamidon and deltamethrin. How much does the extra 8 minutes add for each? What does that tell you about which chemicals care about brewing time?

12.6 How residues are measured

The sample is extracted, impurities are removed, and a broad range of pesticides is measured in one step. The usual tools are gas chromatography, GC, with electron capture detection, flame photometric detection, mass spectrometry, MS, or tandem MS. Polar and heat-sensitive compounds do not suit GC, so liquid chromatography with MS, LC-MS, is used for them. Newer extraction methods are liquid-liquid extraction, solid-phase extraction and solid-phase microextraction. But separations are not always complete, pesticides break down into products that are still unknown, and no one set of methods works in every situation. The book recommends testing for broad groups of compounds, such as total organic chlorine or total organic phosphorus, and using an established method for a known pesticide.

Quick check

Why can one laboratory method not cover every pesticide in tea?

Check your answer

Some pesticides evaporate and suit gas chromatography, while polar, heat-sensitive or non-volatile ones need liquid chromatography instead.

12.7 MRLs and world trade

A maximum residue limit, MRL, is the highest residue level allowed in the traded product. MRLs are fixed from field trial data on how well a pesticide works, to find the highest residue expected when it is used according to Good Agricultural Practice, or GAP. The book's table of MRLs covers eleven authorities, and the numbers differ widely between them, a serious problem for trade.

The Codex Alimentarius Commission, set up by the Food and Agriculture Organization and the World Health Organization, is the international body that sets MRLs, and its values give a fair picture of the scale involved. The table below lists the Codex values from the book for a set of pesticides. Where the book's table leaves a cell blank, no value is given here either.

Codex maximum residue limits for selected pesticides in tea (mg/kg)

PesticideCodex MRL (mg/kg)
Bromopropylate5.0
Cartap2.0
Cyhexatin2.0
Dicofol8.0
Endosulfan30.0
Ethion5.0
Fenitrothion0.5
Methidathion0.1
Parathion-methyl0.2

Two points in the table deserve attention. First, the spread is enormous: endosulfan at 30.0 mg per kg against methidathion at 0.1 mg per kg, a three-hundred-fold difference, because limits reflect each chemical's own behaviour and toxicity, not a single standard. Second, the book gives no Codex value for captan, cypermethrin, deltamethrin, malathion or fenazaquin. The book also notes that the same pesticide can carry very different limits under different authorities: dicofol, for example, is listed at 8.0 under Codex and at 45.0 under the US EPA, while ethion is listed at 5.0 under Codex and at 0.1 under one European authority.

Present MRLs are fixed for made tea, the dry processed leaf, but made tea is not consumed as it is: the drink is the infusion. It is the pesticide that moves into the brew that is actually drunk, so the book argues that realistic MRLs should be based on what is in the brew rather than on what is in the made tea. Its closing advice is consistent with that: keep residues far below the MRL, use persistent broad-spectrum pesticides only with the greatest care, bring in safer biopesticides and newer formulations at much lower doses, and watch residues continuously whenever a new compound or formulation enters the pest management programme.

Practical note

This unit reports limits and rates exactly as the book states them. It does not advise any grower or drinker on safety. MRLs differ between authorities and change over time, so a number here is a record of what the book published, not a current legal limit.

Quick check

Why does the book argue that MRLs should be set on the brew rather than on the made tea?

Check your answer

Only the part that moves into the hot water infusion is drunk, generally 3 to 25 per cent of the leaf residue.

12.8 Organic tea farming: the idea

Organic farming starts from a different question. It asks whether the garden can be run without those chemicals at all. The book defines organic farming, following Lampkin and Padel, as an approach to agriculture that creates integrated, humane, environmentally and economically sustainable production systems, relying as much as possible on farm-derived renewable resources and on managing ecological and biological processes.

The foundations are: keep soil fertility by building biomass in place, rely on genetic diversity, recycle nutrients, conserve the soil and the ecosystem, limit fossil energy and non-renewable inputs, and treat the farm as a near-closed system with few inputs from outside.

The need for it comes from the record of intensive farming. Intensive farming raised production but damaged the ecosystem, while agrochemicals stayed a necessary evil. In plantation crops the use of agrochemicals has been much higher than in field crops, so the threat from pesticide residues is high, especially in made tea. Tea is harvested about weekly, too short a time for many agrochemicals to break down, and the importing markets have fixed MRLs on made tea. Steady use of pesticides, above all persistent herbicides applied to the soil, pollutes groundwater and harms useful soil life such as vesicular-arbuscular mycorrhizal fungi and plant growth promoting rhizobacteria. The organic sector is still small: under 0.5 per cent of the agricultural industry in most countries.

The book lists what proper organic practice can give: better tasting tea, healthier plant growth, improved soil fertility and soil aeration, less soil erosion, greater water retention, rejuvenation of depleted soil, recycling of organic wastes, less environmental pollution, less pest and disease attack, and higher demand, price and saleability of the tea. The switch is not instant: the transitional period between stopping outside inorganic inputs and moving fully to organic farming is generally 2 to 3 years, and the change should be planned properly rather than rushed. Tea grown in neglected or abandoned gardens does not count as organic tea, because there is no deliberate organic system behind it. A 1991 international conference on organic tea and coffee proposed sustainable plantations kept up by biomass, nutrient recycling, and returning the nutrients taken away in the harvested leaf.

Quick check

Why does the book say the change to organic farming should take 2 to 3 years, not happen overnight?

Check your answer

The soil, nutrient supply and pest balance must be rebuilt after years of inorganic inputs, and the transition generally takes 2 to 3 years.

12.9 Organic practice: planting material, manuring, mulching

Planting material. The first choice in an organic garden is what to plant. The book observes that genetically uniform clonal tea, chosen because it responds well to high inorganic inputs, is not suited to organic farming. It assumes instead that seed stocks, or seed-borne tea plantations, can be brought under organic farming effectively. The reasoning follows the definition from the last section: an organic system relies on genetic diversity, and a single clone is the opposite of diversity.

Manuring. Manure use grew as soil fertility fell: green manure first, then cattle manure, then concentrated manures such as oil cakes. The modern picture in the book is that plant nutrient needs can be met by compost, bone meal, neem cake and bio-fertilizers, and by recycling farm residues and wastes. Enough organic matter must be kept up by periodic supply, or neither nutrient uptake nor microbial activity reaches its best. One pot trial with tea seedlings compared organomineral manures against NPK at 5:1:1 applied at 150 kg per hectare of nitrogen. The organomineral treatment gave a higher tea yield than the NPK treatment, and nutrient uptake, residual soil nutrients and organic matter build-up were all higher with the organic-based fertilizers.

Mulching. Mulching matters in organic farming because it improves nutrient and moisture status, holds soil temperature steady and controls weed growth. In one trial, leaf number three months after transplanting one-year-old tea plants was highest with tea waste mulch at 10 tonnes per hectare, and plant height was highest with Guatemala grass mulch. Fresh tea yield was 78.1 kg per plot with fresh weeding mulch, 90.8 with tea waste, 94.5 with cow dung, and 67.7 with no mulch. In a study of china hybrid tea, farmyard manure with humic acid gave leaves with clearly higher polyphenols and catechins, and made tea with higher theaflavins, thearubigins and caffeine, than the control, though the flavour profile did not change.

Six steps for building organic tea production are listed in the book, following Xu: soil and fertilizer management; cultivation techniques; processing and packaging; storage and marketing; testing organic tea; and changing normal plantations into organic ones.

Quick check

Why does the book say seed-raised tea suits organic farming better than a single high-yielding clone?

Check your answer

Because the clone was chosen for responding to high inorganic inputs, while organic farming relies on genetic diversity and on the farm's own resources. Seed stocks carry that diversity, so they fit the organic system better.

Use the mulch trial numbers above. How much more fresh tea did the cow dung mulch plot yield than the no-mulch plot, in kg per plot and as a percentage? Which of the four treatments would you test first on a new organic plot, and why?

Try it yourself

Use the mulch trial numbers in section 12.9. How much more fresh tea did the cow dung mulch plot yield than the no-mulch plot, in kg per plot and as a percentage?

12.10 Organic tea: how well it works, and where it struggles

Organic tea is a knowledge-heavy, labour-heavy system. The book calls organic farming highly knowledge intensive, labour oriented and more diverse and complex than conventional farming, needing motivation, commitment and inventive ability.

On the technical side, the results are encouraging. In a study of the technical efficiency of organic tea small holdings, the book reports great potential to raise production by 55 per cent through efficient use of the present technology. The study found a clear link between compost application and labour on organic tea production, and a clear positive link between technical efficiency and training of farmers.

The soil life itself tells a story. A study comparing ground arthropods in organic and chemically intensive tea plots found the arthropod community in the organic tea numerically and taxonomically richer. Beetles were the most diverse group, and the ground-level fauna suggested the soil of the organic plantation was healthier than that of the conventional plot.

The main hurdle, the book says, is marketing. The bottlenecks are the absence of testing and certifying agencies, the lack of an organised market, higher premium costs, low demand, and the higher perishability of organic farm produce. Organic farming carries long-term benefits, but building the system, from soil to market, is still unfinished.

Practical note

Organic tea is not simply tea grown without spraying. The book's definition requires a whole system: soil fertility built from biomass, nutrients recycled, diversity kept, and a planned transition of 2 to 3 years.

Key ideas to carry forward

  • Tea is sprayed often and plucked soon after, so residues on tea are usually higher than on other crops at the same application rate.
  • Five forces shrink residues on the bush: growth dilution, sunlight, evaporation, pH and rain, and no pesticide on tea shoots has a half-life longer than eight days.
  • Half-lives differ widely: dimethoate lasts about a day, dicofol about four days, and the same wait after spraying means very different residues for different chemicals.
  • Processing cuts residues further: withering gives the biggest single-step fall, and drying removes about 30 to 40% of what remains; green tea manufacture leaves less residue than black tea manufacture for the four pesticides the book tables.
  • Only part of the leaf residue reaches the cup: generally 3 to 25 per cent, at most about 40 per cent, and water solubility decides how much moves, with longer brewing moving more.
  • Residues are measured by gas or liquid chromatography with mass spectrometry, and no single method covers every pesticide.
  • MRLs differ widely between authorities, and the book argues they should be set on the brew that is drunk rather than on the made tea.
  • Organic tea farming replaces outside chemical inputs with biomass, recycling, diversity and soil life, needs a 2 to 3 year transition, and its main hurdle is marketing, not growing.

Glossary

Maximum residue limit (MRL)

The highest residue level allowed in a traded product, fixed from field trials under Good Agricultural Practice. Food tech example: the Codex MRL for dicofol in tea in the book's table is 8.0 mg per kg.

Half-life

The time for half of a pesticide to disappear from the leaf. Food tech example: dimethoate has a half-life of about one day on tea, so after two days about a quarter of it is left.

Growth dilution

The fall in residue concentration as the shoot grows heavier and spreads the chemical through more tissue. Food tech example: tea shoots grow about 13 times heavier in a week, so the same amount of pesticide is spread far thinner.

Photodegradation

Breakdown of a chemical by sunlight. Food tech example: quinalphos breaks down in sunlight faster than cypermethrin on tea shoots.

Codex Alimentarius

The international food standards body set up by the Food and Agriculture Organization and the World Health Organization, which sets MRLs. Food tech example: the Codex MRL for endosulfan in tea in the book's table is 30.0 mg per kg.

Organic farming

Farming that relies on farm-derived renewable resources and ecological processes, with few outside inputs and no synthetic chemicals. Food tech example: organic tea farming builds soil fertility with compost, mulch and green manure instead of inorganic fertilizer.

Infusion

The drink made by pouring hot water over tea leaves. Food tech example: the infusion, not the dry leaf, is what the drinker actually consumes, so the book argues MRLs should be set on it.

Transition period

The time between stopping inorganic inputs and running a full organic system. Food tech example: the transition period for organic tea farming is generally 2 to 3 years.

References

  • Ahuja, P. S., Gulati, A., et al. (eds.), Science of Tea Technology. Chapter 15, Pesticide Residues in Tea, by Yogesh B. Pakade and Adarsh Shanker; Chapter 16, Organic Tea Farming, by R. D. Singh.
  • Pakade, Y. B. and Shanker, A., Pesticide Residues in Tea, in Ahuja et al., Science of Tea Technology, Ch. 15.
  • Singh, R. D., Organic Tea Farming, in Ahuja et al., Science of Tea Technology, Ch. 16.

Unit 13: Tea Physiology and Biotechnology

Note

Unit 13: Tea Physiology and Biotechnology

A tea bush can sit still all winter, with no new shoot in sight, and then burst into growth in spring. The same bush can lose a large share of its yield in a dry spell, and yet a single leaf taken into a laboratory jar can grow into a whole new plant. This unit follows the living bush first, then the laboratory work that tries to copy, protect, and improve it.

Questions to think about as you read: Why does a tea bush stop growing in winter even when the days are still mild? Where does the food made by a mature leaf actually go? Why is it so hard to root a tea shoot in a jar?

13.1 Introduction: the bush as a factory and a storehouse

Tea, Camellia sinensis, grows in a wide range of climates, from warm humid lowlands to cooler highland sites. The book lists about 82 species in the genus Camellia, with three main cultivated types: the China type, the Assam type, and the Cambod type. Seed-raised plants vary a great deal because tea outcrosses freely, so breeders select chosen plants and multiply them by cuttings to keep them uniform.

A tea bush needs, on average, at least five hours of sunshine a day and air humidity between 70 and 90% while it is growing actively. When the air is too dry, shoots tend to form dormant buds, called banjhi buds. Yield in tea is measured as made tea per unit of land per year, from shoots of usually two or three leaves and a bud, after withering, rolling, oxidation, and drying. Yield depends on two plain things: how many plucked shoots grow per unit area, and how heavy each shoot is. Behind both sit four processes: how much carbon dioxide the leaves fix, how the food made is shared out to the harvestable shoots, how fast shoots start and extend, and how the plant manages water in its tissues when soil water changes.

Quick check

What two things decide tea yield in the field?

Check your answer

The number of plucked shoots per unit of land, and the mean weight of each shoot. Shoot number follows the rate of shoot initiation, and shoot weight follows the rate of shoot expansion.

13.2 Effect of environmental factors on tea

Tea leaves build their photosynthetic capacity slowly as young leaves expand, and reach their peak when shoots are ready for harvest. The book reports low rates of ribulose bisphosphate carboxylation, low amounts and activity of the enzyme Rubisco, and high mesophyll resistance to carbon dioxide diffusion in tea leaves. Net photosynthetic rate, written Pn, responds to light, temperature, carbon dioxide level, cultivar, leaf age, and management. One study reported maximum Pn of 13.95 µmol m⁻² s⁻¹ at a light level of 1,400 µmol m⁻² s⁻¹, falling to 6.92 µmol m⁻² s⁻¹ at 200 µmol m⁻² s⁻¹. Pn was highest between 25 and 30°C in several studies, and fell fast at 37°C. At 17°C, Pn was about 17% of the value at 28°C in one study. Heat-tolerant cultivars had a higher temperature optimum, higher stomatal conductance, and higher water use efficiency.

Too much strong light can injure photosynthesis, a problem called photoinhibition, seen most at midday. In one study Pn showed no late afternoon recovery, only overnight recovery. Giving nitrogen above 375 kg ha⁻¹ yr⁻¹ reduced photoinhibition but did not remove it. Xanthophyll cycle pigments (violaxanthin, antheraxanthin, and zeaxanthin) found in fresh tea leaves point to a photoprotection role. Nitrogen also raises chlorophyll in flush and mature leaves, and nitrogen is needed to build chlorophyll and Rubisco, so leaf nitrogen tracks Pn, growth, and yield, above all where nitrogen is short.

Tea is believed to have started as an understorey plant in tropical rainforest, and its photosynthetic machinery works well under shade. Shade trees and shrubs are a common practice in tea gardens to cut photoinhibition, save soil moisture, and keep stomata open so carbon dioxide can enter. Shade is less needed in cooler highland sites, and too much shade can cut yield. One study in a humid lowland zone found an optimum shading level between 30% and 40% for tea growing up to 600 m above sea level. The book gives a temperature range of about 13°C to 32°C as favourable for growth, with the exact range shifting by site and cultivar. Water stress cuts stem diameter, shoot extension, branching, and leaf expansion, and a dry spell can unbalance the crop: buds go dormant in drought, then a sudden flush follows rain. Day length below about 11 hours for about six weeks is linked with dormancy in tea, with the book's Ch. 21 passage putting the critical length at 11 hours 15 minutes with minimum temperature below 13 °C.

Studies with carbon-14 labelled photosynthate found about 11% going to pluckable parts such as two leaves and a bud, and 19% and 25% going to maintenance leaves and branches. The top 10 cm of canopy supports about 75% of the harvested shoots, and the next 10 cm supports the rest, about 25%. The top leaf is the most efficient in photosynthesis because of its age, its place in the light, and its nearness to growing shoots. Young leaves do not export food: what they make stays in them and is used to build flavanols such as catechins. Mature leaves, the maintenance foliage, supply sugars to young buds, developing shoots, and roots. Even the fifth leaf from the top sends food to the pluckable shoots above it. Because young shoots are picked before they can export food, a canopy of mature leaves, four to five layers deep, is left on plucked bushes. Short exposure to raised carbon dioxide raised Pn, and a free-air enrichment study reported Pn up by 17.9% at 550 µmol mol⁻¹ and by 25.8% at 750 µmol mol⁻¹, compared with ambient carbon dioxide. The book still warns that there is no clear link between maximum photosynthetic rate and yield potential across tea genotypes.

Nitrogen is the most important nutrient for yield. More nitrogen is usually needed in the second and third years of a pruning cycle. Zinc application in unpruned years is reported to raise yield by raising the number of actively growing shoots. Phosphorus matters for active shoot growth when nitrogen is adequate, but its amount needed is usually low. Potassium helps build and move starch and sugars, and it counters the adverse effect of high nitrogen levels. When nitrogen is short, the root to shoot ratio rises, because roots become stronger sinks for carbohydrate.

Sunlight, carbon dioxide, and water enter the leaf

↓

Photosynthesis in mature maintenance leaves makes sugars

↓

Sugars move to growing shoots, buds, and roots

↓

Only about 11% reaches the pluckable two leaves and a bud

Figure 13.1: From light to pluckable shoot. What it shows: the chain from photosynthesis to the harvested shoot, and the small share that reaches the pluckable part. Drawn from Ahuja et al., Science of Tea Technology, Ch. 17.

13.3 Dry matter production

Dry matter content, written DMC, is the share of dry matter in the shoot tissue. It depends on photosynthesis, respiration, and water content. Dry periods tend to give higher DMC, and wet periods give lower DMC, because wet-season shoots hold more water and have high turgor pressure. Low DMC means a low recovery of made tea from green leaf, and the extra water puts load on factory dryers, uses more energy, and needs more labour. A high green-leaf weight built mainly from water is therefore a poor bargain.

Plucking at a seven-day interval gave 38% less yield than plucking at 14 days in one study, linked to immature plucking and low DMC. Longer plucking rounds can harm quality by raising crude fibre. Plucking after 11 am gave a higher recovery percentage than morning plucking, because sinks are empty in the morning and food made early in the day is used up fast. Three leaves and a bud weigh more than two leaves and a bud, so picking it can raise productivity and allow a longer plucking interval, but it can dilute quality through extra fibre and loss of constituents. Total dry matter production falls as the bush ages. One study found 46% of respiratory loss was from permanent woody tissue and 28% from large branches. The book is plain that dry matter production and partitioning in tea are still not clearly understood. Harvest index is the dry matter in plucked shoots as a share of the total dry matter produced by the plant. The levers the book names for raising it are: more photosynthesis, less respiratory loss, more food moved to new shoots, and fewer non-productive plant parts.

Try it yourself: Take two shoots from the same bush on the same day, one picked in the morning and one after 11 am, and weigh each fresh, then dry them fully and weigh again. Work out dry matter as a share of fresh weight for each shoot. The book predicts the afternoon shoot will show the higher recovery: check whether your two shoots agree, and note the weather, because a wet morning changes the result.

13.4 Winter dormancy

Near the equator, tea yields almost the same amount each month. Further from the equator, the winter harvest falls, and beyond about 16° north or south latitude there is almost complete winter dormancy. Note: the book gives this latitude threshold differently in different chapters: Unit 10 §10.8 puts it at 19 degrees. Treat both 16 degrees and 19 degrees as approximate chapter figures, not as a single rule. The trigger is day length: bushes go dormant when day length stays below about 11 hours for about six weeks, with the book's Ch. 21 passage putting the critical length at 11 hours 15 minutes with minimum temperature below 13 °C. Low temperature alone is not the whole story: one site with winter maximum temperatures of 22 to 26°C and minimum temperatures of 10 to 16°C still had three to four months of dormancy, while another site with temperatures of 19 to 25°C by day and 8 to 9°C by night had no dormancy. The difference was day length. The first flush after winter dormancy usually fetches a higher price for its flavour, while leaf plucked in the rainy season is seen as less desirable in aroma and is often sold in bulk for blends.

Winter dormancy by latitude, generalised from the book table, with no single site named:

Latitude bandDormancy period reported
Near the equatorAlmost none, yield steady through the year
About 20° from the equatorAbout 2 to 3 months
About 26 to 30° from the equatorAbout 3 to 4 months
About 37 to 42° from the equatorAbout 5 to 6 months

Table 13.1: Winter dormancy length rises with distance from the equator. Values are the ranges given in the book, grouped by latitude.

Quick check

What day length threshold is linked with the start of winter dormancy?

Check your answer

Day length below about 11 hours for about six weeks, with the book's Ch. 21 passage putting the critical length at 11 hours 15 minutes with minimum temperature below 13 °C, at latitudes beyond about 16° north or south.

13.5 Dormancy and plant growth regulators

Dormancy is a balance between promoters and inhibitors inside the shoot. Abscisic acid, written ABA, is linked with starting and holding dormancy: high levels of free ABA were found in dormant shoots, and the levels fell as dormancy released. Gibberellic acid, written GA, was low during dormancy and rose sharply at bud break, and spraying GA was effective in bringing bud break earlier, though the response was not the same every year on the same clone. Cytokinins were low in winter shoots and rose slowly before bud break. Phenolic acids may also take part, because some phenolics slow cell division. Polyamines also shift: in two clones studied, putrescine rose and spermidine and spermine fell as dormancy set in, while high spermidine and spermine were linked with dormancy release. Brassinosteroids were reported from dormant shoots, and at a very low concentration, 5 mM in the study cited [the book prints 5 mM and calls it very low; the concentration should be checked against the cited source before reuse], they could move food out of maintenance leaves faster, raising dry matter recovery and harvest index.

Short days, below about 11 hours, for about six weeks

↓

ABA rises, GA falls, cytokinins fall, putrescine rises, spermidine and spermine fall

↓

Bud growth stops: winter dormancy

↓

Longer days in spring: ABA falls, GA and cytokinins rise, spermidine and spermine rise, bud break follows

Figure 13.2: The regulator balance behind winter dormancy. What it shows: how day length shifts ABA, GA, cytokinins, and polyamines between dormancy and bud break. Drawn from Ahuja et al., Science of Tea Technology, Ch. 17 and Ch. 21.

13.6 Plant growth regulators for crop improvement

Plant growth regulators, written PGRs, have been used to raise yield in horticultural and ornamental crops, and in tea the response is encouraging but often inconsistent, because weather shifts and tea plants are highly variable. Auxins, gibberellins, and cytokinins have found only limited use in tea, in specific cases. The regulators of commercial interest named in the book are triacontanol, written TRIA, a long-chain fatty acid alcohol, and products holding hydrolysed protein complexes. The book reports that these regulators help above all in the drought season by raising water use efficiency, and that brassinosteroids give similar responses at a very low concentration [the book prints 5 mM and calls it very low; the concentration should be checked against the cited source before reuse]. They move food out of maintenance leaves faster, raise dry matter recovery and harvest index, and allow faster growth and leaf expansion. About 10 to 15% of photosynthate is expected to go into dry matter accumulation. Growth retardants have a different job in young tea: they suppress apical dominance and push latent buds to grow. Paclobutrazol at 500 ppm and ethaphon at 1,000 ppm, sprayed on foliage one month after planting, were found effective in strengthening the root system and promoting lateral branching in young tea.

13.7 Tea glutamine synthetase: gene isolation, characterisation, and exploration

Glutamine synthetase, written GS, is a key enzyme of ammonium assimilation in the GS/GOGAT pathway in plants. Researchers cloned a cDNA for tea glutamine synthetase, called CsGS, from Camellia sinensis, and made the recombinant protein in Escherichia coli, purifying it by nickel affinity chromatography and confirming it by blotting with anti-His antibodies. The kinetic values reported are: with glutamate as substrate, Km 9 mM and Vmax 93 U/mg protein; with ATP, Km 6 mM and Vmax 70 U/mg protein. The purified enzyme showed a pH optimum of 8. Cations raised activity, with magnesium, Mg²⁺, giving the largest rise. Activity rose by 25% with DTT and fell by 18% with PMSF. The Km values for CsGS were higher than those reported for rice, Arabidopsis, maize, and human, which points to lower affinity for its substrates. Tests ruled out a ping-pong mechanism and favoured a random mechanism of catalysis, with ADP acting as a competitive inhibitor against ATP and a non-competitive inhibitor against glutamate and ammonium.

Expression of the CsGS gene in the tea bud was stimulated by ABA, salicylic acid, and hydrogen peroxide, H₂O₂, while gibberellic acid, GA₃, had no influence on it. Expression and activity rose with ammonium and nitrate, and with cadmium and salt stress, but fell with copper, aluminium, cold, and heat stress. Expression fell during dormancy compared with the active growth phase, and was highest in the apical bud and lower in the mature fourth leaf. In light, both nitrate and ammonium raised CsGS expression in the apical bud; in darkness, only ammonium raised it, while nitrate had no influence, and under darkness ammonium raised enzyme activity while nitrate lowered it.

Try it yourself: Draw a simple two-column card for CsGS: one column headed "raises CsGS", one headed "lowers CsGS". Fill it from this section only: ammonium, nitrate, cadmium, salt stress, ABA, salicylic acid, and hydrogen peroxide on one side; copper, aluminium, cold, heat, and dormancy on the other.

13.8 Molecular diagnostics using DNA markers

Tea is a popular and high-priced product, and adulteration has become a serious problem, so checking that a sample is really tea matters. DNA markers are more suitable than the older ash test, because ash content varies widely between tea types and their hybrids. The book describes a ribosomal RNA gene approach that amplifies the spacer regions to build species-specific PCR primers for common plant adulterants, naming babul, bamboo, and cashew. The method was used to detect adulterants in processed and made tea samples, with a PCR profile showing a band for each adulterant and for tea. The point for a food technologist is plain: a DNA test reads the plant itself, not a bulk property such as ash that shifts with variety.

Molecular markers also help in a wider job: judging genetic diversity, sorting gene bank material, finding parent groups, and checking the identity of important germplasm. Morphological traits shift with the environment and may not show true genetic diversity, while DNA markers read variation at the DNA level. Early work used RFLP with phenylalanine ammonia-lyase cDNA probes, then RAPD, ISSR, and AFLP markers. One RAPD study of 38 accessions reported an average of 62% variability, and one AFLP study reported 79% of variation within populations and 21% between populations. The book notes that molecular tools in tea were still limited when the chapter was written, and that the first markers were reported in 1994.

13.9 Physiological and molecular basis of winter dormancy

Winter dormancy shows as growth arrest of the apical bud and its two associated leaves. The length of the dormant period rises steadily with distance from the equator, reaching up to six months at the highest latitudes in the book table. Bushes enter dormancy when day length falls below about 11 hours for about six weeks, with the book's Ch. 21 passage putting the critical length at 11 hours 15 minutes with minimum temperature below 13 °C. Exposing bushes to a 13-hour photoperiod in winter hastened bud break and cut flowering sharply in one study.

Free ABA is high in dormant shoots and falls on release, while bound ABA rises before release, which points to conversion between free and bound forms. Free gibberellin activity is very low at the start of dormancy and stays low, while conjugated gibberellin activity stays high; later in dormancy, the conjugated forms fall and free gibberellin rises. Cytokinin glucosides are higher during dormancy, and free cytokinins rise at bud break. In direct tests, BAP was not strong at forcing bud break but stimulated shoot growth in spring; ABA and NAA slowed shoot growth, while GA₃ brought bud break earlier by about two weeks and also raised amino acids, protein, and caffeine content.

Cold brings a second problem: oxidative stress. Low temperature with light raises reactive oxygen species, written ROS, because chilling-induced photoinhibition lets molecular oxygen take electrons in place of NADP⁺. ROS can damage lipids, proteins, and DNA, so plants carry enzymes that clear them. In tea clones with different dormancy lengths, bud growth, net photosynthesis, and the ratio Fv/Fm all fell in winter, while superoxide anion rose in all clones. The clearing enzymes superoxide dismutase, written SOD, ascorbate peroxidase, and glutathione reductase rose at the start of dormancy and at peak dormancy. The clone with the largest rise in clearing enzyme activity held the least superoxide, and the clone with the least superoxide had the shortest dormancy. The book draws the line plainly: managing oxidative stress should be central to managing winter dormancy, and manganese SOD was closely linked with dormancy in tea.

Gene work adds names to the story. Differential display found a QM-like protein gene, CsQM, which codes for ribosomal protein L10. It was strongly expressed in apical buds during active growth, fell with leaf age, and was severely down-regulated in all leaves during dormancy. It was down-regulated by drought and ABA, and up-regulated by GA₃. Histone H3, CsH3, behaved in the same way. A wider transcriptome study used two subtractive libraries with 1,102 sequences in total. The forward library, genes switched on during dormancy, gave 161 unigenes from 602 sequences; the reverse library, genes switched on during active growth, gave 329 unigenes from 500 sequences. Stress-responsive genes were switched on during dormancy, while genes for the cell cycle and DNA processing were switched off. Low temperature at 4°C switched gene expression in actively growing shoots in the same way as dormancy did, and growth temperature at 25°C switched it back toward the active-growth pattern. One more gene, alpha-tubulin CsTUA, was switched on during dormancy and is a low-temperature inducible gene: it was switched on within 24 hours of low-temperature treatment, was not switched by ABA, GA₃, indole-3-butyric acid, or 6-benzylaminopurine, and was also switched on by sodium chloride, polyethylene glycol, and hydrogen peroxide. The book suggests it may help the evergreen tea plant keep its leaves through winter instead of shedding them like a deciduous tree.

13.10 Response of tea to drought

Drought is a yield-limiting factor in tea growing areas around the world. Growing areas face water deficit for 2 to 3 weeks to 5 months, and this cuts yield by up to 40% in mature plants. Water deficit at the seedling stage harms survival and establishment. Using differential display, a thaumatin-like protein gene, written TLP, was found as a prominent drought-responsive gene, and it was not switched by ABA. A wider study built two subtracted libraries: a drought-stress library of genes switched on during drought, and a library of genes switched off during drought. From 1,892 clones sequenced, 1,189 high-quality expressed sequence tags were found, giving 965 unigenes after removing repeats: 645 from the drought library and 320 from the other library. Cell rescue and defence genes and cellular transport genes were strongly switched on, more than twofold, during drought. Among 49 genes with more than three sequences in the drought library, cell rescue and defence held the most, 14 genes, or 29%, followed by cellular transport with 7, metabolism with 6, and energy with 5. Genes of the flavonoid pathway were switched off during drought, which fits the fall in catechin building reported in the next section.

Quick check

Which two gene groups were most strongly switched on during drought?

Check your answer

Cell rescue and defence genes, and cellular transport genes, both switched on more than twofold during drought stress.

13.11 Molecular basis of catechin biosynthesis in tea

Catechins give black tea its astringency and bitterness, and they make up 25 to 30% of the dry mass of young tea leaves. They are built through two linked pathways: the phenylpropanoid pathway, written PP, and the flavonoid pathway, written FL. The PP pathway starts from phenylalanine, which comes from the shikimate pathway. Phenylalanine ammonia-lyase, written PAL, turns phenylalanine into cinnamic acid. Cinnamate 4-hydroxylase, written C4H, turns that into 4-coumaric acid, and 4-coumarate:CoA ligase, written 4CL, turns it into 4-coumaroyl CoA, the starting material for the flavonoid pathway.

The FL pathway starts when chalcone synthase, written CHS, joins one molecule of 4-coumaroyl CoA with three molecules of malonyl CoA to make a C15 chalcone, naringenin chalcone. Chalcone isomerase, written CHI, cyclises it to naringenin, the first flavonoid. Hydroxylation steps then give dihydroflavonols, which dihydroflavonol 4-reductase, written DFR, reduces to leucoanthocyanidins. From there, leucoanthocyanidin reductase, written LAR, gives 2,3-trans-(+)-catechins, while anthocyanidin synthase, written ANS, gives anthocyanidins, which anthocyanidin reductase, written ANR, reduces to 2,3-cis-(-)-epicatechin. Flavonoid 3′-hydroxylase and flavonoid 3′,5′-hydroxylase, written F3′H and F3′5′H, set the hydroxylation pattern that decides which catechin forms. Catechins are made in the cytoplasm and stored in the vacuole, and moving them there needs glutathione S-transferase and a glutathione pump from the ATP-binding cassette family. Tea is unusual because it uses these pathways mainly to make catechins, and its leaves store as much as 25 to 30% catechins on a dry weight basis.

All the genes studied were switched on more in younger leaves, up to the third node below the apical bud, than in older leaves. Most of them were switched off by drought and by ABA, while CHS rose slightly on longer exposure. GA₃ lowered most of them, while F3′5′H rose slowly in expression and DFR was unchanged by GA₃. Note the contrast for F3′5′H: its expression rose under GA₃, but its correlation with catechin content under GA₃ is -0.990, meaning expression rose while catechin content fell. Wounding raised the expression of the genes tested. The book names CHI, F3H, DFR, ANS, and ANR as the critical regulatory genes for catechin content under the cues tested so far. The full book table is given below.

GeneDroughtABAGA₃WoundingLeaf position
PAL0.9960.7330.9190.0650.968
C4H0.9110.6190.8570.4300.985
4CL0.7050.7090.7380.3510.962
CHS0.91-0.2190.4930.4040.998
CHI0.980.4520.8820.9320.98
F3H0.7240.9430.7870.1370.609
F3′5′H0.8930.423-0.9900.2170.996
DFR0.8780.9070.3270.4440.800
ANS0.780.4690.5150.0830.906
ANR0.960.790.0960.690.958

Table 13.2: Pearson correlation between each gene's expression and the catechin content, under each cue. A high positive value means expression and catechin content rise and fall together; it does NOT mean the cue switches the gene on. Most genes are down-regulated by drought and ABA even though their expression still tracks catechin content.

Phenylalanine from the shikimate pathway

↓

PAL, then C4H, then 4CL: 4-coumaroyl CoA

↓

CHS joins it with three malonyl CoA: naringenin chalcone

↓

CHI, F3H, F3′H, F3′5′H, DFR: leucoanthocyanidins

↓

LAR gives (+)-catechins; ANS then ANR gives (-)-epicatechin

Figure 13.3: The catechin biosynthesis pathway in plain steps. What it shows: the phenylpropanoid pathway feeding the flavonoid pathway, with the enzyme at each step. Drawn from Ahuja et al., Science of Tea Technology, Ch. 21.

13.12 Tea micropropagation

Tea is traditionally grown from cuttings, but many old plantings came from seed, and seed plants vary widely. Vegetative cuttings give uniform plants, but they grow slowly, depend on the season, and root differently from clone to clone. Micropropagation, growing small shoots in sterile jars, gets around those limits and can supply improved clones faster. Many explants have been used: zygotic embryos, immature and mature cotyledons, stem segments with or without the outer layer, flower stalks, leaf petioles, shoot tips, nodal segments from young or adult plants, and leaf pieces for adventitious buds. Most reports used agar-gelled medium; only one report cited used liquid medium for shoot growth, and that route is attractive for commercial use because liquid medium with longer subculture periods saves agar, medium, and labour. The genetic fidelity of micropropagated plants was checked with RAPD and microsatellite markers.

13.13 Rooting, hardening, and field transfer of tissue culture raised plants

Rooting is the hard step in tea micropropagation. Reduced salts, light, and a low pH of 4.5 to 4.6, with IBA treatment at 100 mg/l followed by hormone-free medium, favoured rooting in the work cited. Ex vitro rooting, rooting shoots directly in soil mix after IBA treatment, was seen as better than in vitro rooting. Moving plantlets from jar to greenhouse is the most delicate phase. Researchers built special hardening chambers and tuned the conditions: shoots harvested in spring and early summer, shoot size, soil pH 5.4, IBA at 500 mg/l for 30 minutes, carbon dioxide enrichment, and light. The result was up to 71.60% root induction, 73% survival, and successful field transfer.

Single-node cuttings took 70 to 85 days to root and 510 days from rooting to field establishment, with average height 32.50 cm, 8.60 leaves, stem thickness at the collar 12.50 mm, and 91.2% survival in the field after one year. Microshoots took 60 days to root and the same 510 days to field establishment, with average height 20.00 cm, 16.00 leaves, stem thickness 4.5 mm, and 73.3% survival. The cuttings were taller and thicker; the microshoots carried more leaves but survived less well in the field. Four-node and eight-node cuttings rooted with auxins plus phenolics gave 100% rooting with NAA plus phenolics, and the cuttings were ready for the field in only ten months, against 15 to 16 months for single-node cuttings.

Micro-grafting was also reported for the first time in tea: in vitro microshoots grafted onto seedling rootstocks, for vigour and for hardening and field transfer in one step. Scions and rootstocks were pretreated with BAP at 5 mg/l and NAA at 5 mg/l for 10 minutes, and grafted plants were kept in carbon dioxide enriched air at high light in hardening chambers. Grafts on four-month-old rootstocks did best, at 88.33% establishment, against 71.00% on two-month-old rootstocks and 47.00% on twelve-month-old rootstocks for one stock type. Grafts made in spring established best, at 92.00%, against 74.50% in winter, 83.50% in early summer, and 7.50% in peak summer. Field survival of grafted plants was 88.33%, and success came within 10 months, against the usual 20 to 24 months for cuttings.

Try it yourself: Sort the field-transfer numbers above into a small table of your own: rooting days, field survival, and height for cuttings against microshoots. Then write one sentence saying which route you would pick for a new planting and why, using only those numbers.

13.14 Somatic embryogenesis

Somatic embryogenesis is a second route to new plants: ordinary body cells are pushed to form embryos without fertilisation. It suits mass propagation, and it is also a tool for improvement work. The first report was in 1981, and many workers followed, using cotyledons, immature embryos, and other explants, with stress treatments and changes in nutrients and growth regulators to trigger embryos. One laboratory protocol used half-strength MS medium with 2% sucrose, BAP at 2.0 mg/l, and IAA at 0.2 mg/l, and the embryos germinated when GA₃ at 0.2 mg/l was also present. A systematic study of salts found that potassium nitrate at 9.39 mmol and ammonium nitrate at 10.31 mmol, with only 1.5 mmol potassium sulphate, gave the most synchronous secondary embryos: 20 to 25 secondary embryos per primary embryo in 91.6% of responsive explants. The best mix of regulators was 8.88 µmol BAP, 0.98 µmol IBA, and 10 mmol L-glutamine. Germination, at 52%, came only when sulphate salts of potassium were removed from the medium. Poor and abnormal germination was a common fault, and the book traces it to a lack of reserve food built up during embryo maturation. Adding trans-cinnamic acid, a fatty acid precursor, and maltose, a readily used sugar, to the germination medium raised normal germination to 70%. Adding ABA to the maturation medium also helped normal germination by switching on the programme for storing reserves.

Cotyledon or embryo explant on medium

↓

Callus, then primary somatic embryo

↓

Secondary embryos: globular, then heart shaped, 20 to 25 per primary embryo

↓

Germination with reserves built up, then hardening and field transfer

Figure 13.4: Somatic embryogenesis in tea, from explant to plantlet. What it shows: the stages from explant through secondary embryos to a germinated plantlet. Drawn from Ahuja et al., Science of Tea Technology, Ch. 21.

13.15 Synthetic seeds

A synthetic seed is a somatic embryo, or a small rooted shoot, packed in a clear gel that holds nutrients, growth substances, and fungicides, so it can be stored and moved like a seed. Researchers were the first to make viable synthetic seeds of tea by enclosing somatic embryos in sodium alginate beads. Among four cultivars tested, one gave the best result, with 66.70% of explants responding and 120 days for development, though each genotype responded best at the mature cotyledon stage. Cold treatment cut germination, and the cut was larger in uncoated embryos than in coated ones, so the alginate coat protected the embryo in cold storage. Alginate beads dried for one hour before cryopreservation tolerated liquid nitrogen better, while uncoated embryonic axes that were pretreated with sucrose and dried did not survive.

Because tea embryos are recalcitrant and convert poorly, a second system used axillary buds as the propagule. A solution of 4% sodium alginate and 75 mM calcium chloride, written CaCl₂, gave perfectly rounded beads with nodal explants. An initial treatment of 5 µM TDZ and 10 µM NAA, followed by transfer to hormone-free basal MS medium, raised shoot growth. Storing these artificial seeds at 4°C cut the time to sprouting by 15 days in both coated and uncoated buds, and it also improved sprouting and growth.

Quick check

What does the alginate coat do for a synthetic seed?

Check your answer

It packs the embryo or bud with nutrients, growth substances, and fungicides in a small, firm bead that is easy to store and transport, and it protects the embryo against cold storage better than no coat at all.

13.16 Tea transgenics

Conventional breeding in tea is slow because the crop is highly heterozygous, long-lived, and cross-pollinated. Biotechnology offers other routes: transgenics, somaclonal variants, somatic hybrids, artificial seeds, cryopreservation of elite clones, and embryo rescue. Agrobacterium mediated transformation was the first route to transgenic tea plants carrying the gus and nptII genes. Two disarmed strains were tested, both carrying a binary plasmid with nptII and gus-intron genes. Pre-culture, wounding, and acetosyringone treatments were inhibitory, while bacterial growth phase at optical density OD₆₀₀ of 0.6, cell density of 10⁹ per ml, co-cultivation for five days, and co-cultivation medium pH 5.6 helped transformation. After co-cultivation, globular somatic embryos were placed on medium with 50 µg per ml kanamycin, then moved to maturation and germination medium at 75 µg per ml kanamycin. An average of 40% transient expression was found by the GUS test. Integration into the tea nuclear genome was confirmed by PCR with nptII and gus primers and by Southern hybridisation with an nptII probe.

Getting rid of leftover Agrobacterium after co-cultivation matters. Sporidex and carbenicillin at 400 mg/l, and cefotaxime at 200 mg/l, were needed to control bacterial growth without harming germination of somatic embryos, and the most germination came at 400 mg/l sporidex and carbenicillin. Most commercial cultivars stayed outside the host range of Agrobacterium, in part because of high levels of bactericidal polyphenols in the leaves, but adding autoclaved L-glutamine, or filter-sterilised L-glutamine with glutamic acid, overcame the block in the work cited. Agrobacterium rhizogenes was also used to make hairy roots from leaf explants, confirmed with rol C gene primers, with higher catechin levels found by HPLC in the hairy roots.

Biolistic transformation, shooting DNA into cells, was also tried. Osmotic pretreatment with 0.25 M sorbitol, the distances in the gun, burst pressure, and DNA amount all mattered. About 40% of the GUS-positive transformants that survived a lethal dose of kanamycin showed PCR bands for both nptII and gus, and Southern hybridisation confirmed stable transformation, but those plants could not express the transgene, probably because the transgene landed in the wrong place in the genome. A later biolistic study used osmotin and chitinase genes for stress tolerance. From 99 lines made, only 33% had intact osmotin inserts by Southern hybridisation, and only 15% of those gave stable expression by Northern hybridisation. All of those lines came from one setting: a target distance of 9 cm and a burst pressure of 1,100 psi. Those lines were healthier, multiplied faster, and tolerated osmotic stress from polyethylene glycol better than controls. Their desiccation tolerance was also tested: somatic embryos that normally behave like recalcitrant seed germinated normally even after drying stress, with 40% germination after desiccation stress against 57 to 60% without stress. Lines carrying the chitinase gene deterred insects, and larvae feeding on transformed leaves died at higher rates than larvae on untransformed leaves. Vegetative growth under a polyhouse stayed healthy, but reproductive ability was highly suppressed, which the book reports without explaining the cause.

13.17 Practical note: what a grower and a laboratory worker can take from this unit

For the grower, the unit points to four levers that are already in hand: keep a deep maintenance canopy, because it feeds the next flush; watch day length, because dormancy follows short days more than cold alone; pick in the afternoon where the book shows a higher recovery, and do not chase green-leaf weight that is mostly water; and treat PGR sprays as helpers that vary with season and clone, not as a fixed recipe. For the laboratory worker, the unit points to a chain that must all work: clean multiplication, reliable rooting, careful hardening, and field transfer. A protocol that multiplies well but hardens badly will fail in the field, as the survival numbers in section 13.13 show. DNA markers give a fast check on identity and adulteration, and the catechin pathway genes give targets for future breeding, but the book is clear that the critical promoters and transcription factors for winter dormancy are still not understood, so that work is open.

Key ideas to carry forward

  • Yield is shoot number times shoot weight, and both depend on photosynthesis in mature leaves, sharing of food to shoots, and water management.
  • Only about 11% of labelled photosynthate reached the pluckable shoot; the top 10 cm of canopy carries about 75% of the crop.
  • Winter dormancy follows day length below about 11 hours for about six weeks; ABA rises and GA falls in dormancy, and the pattern reverses at bud break.
  • Drought can cut yield by up to 40%, and drought switches on cell rescue and transport genes while switching off flavonoid pathway genes.
  • Catechins are built by the phenylpropanoid pathway feeding the flavonoid pathway, with PAL, C4H, and 4CL at the start and LAR, ANS, and ANR at the end.
  • Micropropagation, somatic embryogenesis, synthetic seeds, and transgenics all work in tea, but rooting, hardening, and field survival decide whether the laboratory plant becomes a field bush.

Glossary

Apical bud

The growing tip at the end of a shoot, with the youngest leaf. In tea it is the bud that stops growing in winter dormancy and breaks in spring. Food tech example: the plucked shoot is usually the apical bud plus two leaves.

Banjhi bud

A dormant bud on a tea shoot that is not growing, often formed when the air is too dry. Food tech example: a flush with many banjhi buds gives fewer pluckable shoots.

Catechin

A flavanol in tea leaves, such as EGCG, that gives astringency and bitterness and is oxidised during black tea making. Food tech example: young tea leaves hold 25 to 30% catechins by dry weight.

Clone

Plants grown vegetatively from one selected parent, so they are genetically the same. Food tech example: a tea estate plants one high-yielding clone across a whole block for a uniform crop.

Cytokinin

A plant growth regulator that promotes cell division and shoot growth. Food tech example: cytokinin levels in tea shoots are low in winter and rise before bud break.

Dormancy

A resting state in which visible growth stops, as in tea bushes in winter. Food tech example: winter dormancy stops plucking for months at higher latitudes.

Dry matter content (DMC)

The share of dry matter in fresh plant tissue, after water is removed. Food tech example: a dry-period shoot has higher DMC and gives a better recovery of made tea than a wet shoot of the same fresh weight.

Explant

A small piece of plant tissue taken into sterile culture to grow a new plant. Food tech example: a tea nodal segment is a common explant for micropropagation.

Gibberellic acid (GA)

A plant growth regulator that promotes shoot growth and can bring bud break earlier. Food tech example: spraying GA in late winter can advance bud break by about two weeks, but the result varies by year.

Glutamine synthetase (GS)

An enzyme that helps the plant take up ammonium and build glutamine. Food tech example: tea GS, CsGS, has a pH optimum of 8 and is raised by ammonium and nitrate.

Hardening

The step of slowly adapting tissue-culture plantlets to greenhouse and field conditions. Food tech example: hardening in special chambers gave up to 73% survival of tea microshoots.

Harvest index

The dry matter in plucked shoots as a share of the total dry matter made by the plant. Food tech example: raising harvest index means more of the plant food ends up in the plucked shoot.

Maintenance foliage

The mature leaves left on a plucked bush to make food for the next flush. Food tech example: four to five layers of maintenance leaves are left after plucking.

Micropropagation

Growing many uniform plants from small shoot pieces in sterile jars. Food tech example: micropropagation can supply a new tea clone faster than cuttings.

Photoinhibition

Damage to photosynthesis by too much strong light, seen most at midday. Food tech example: shade trees cut photoinhibition in tea grown in strong sun.

Reactive oxygen species (ROS)

Highly reactive forms of oxygen that can damage lipids, proteins, and DNA, above all in cold plus light. Food tech example: tea clones that clear ROS better with SOD show shorter winter dormancy.

Somatic embryogenesis

Forming embryos from ordinary body cells in culture, without fertilisation. Food tech example: one tea protocol gave 20 to 25 secondary embryos per primary embryo.

Synthetic seed

A somatic embryo or bud coated in alginate gel with nutrients and growth substances, for storage and transport. Food tech example: a 4% sodium alginate bead with 75 mM calcium chloride gave rounded synthetic seeds of tea.

Transgenic

A plant carrying a gene moved in from another source by genetic engineering. Food tech example: transgenic tea lines carrying osmotin tolerated osmotic stress better than controls.

References

  • Ahuja, P. S., Gulati, A., et al. Science of Tea Technology. Chapter 17, Plant physiological approaches for enhancing productivity in tea.
  • Ahuja, P. S., Gulati, A., et al. Science of Tea Technology. Chapter 21, Some recent innovations in tea biotechnology.

Unit 14: Tea Chemistry in Depth: Polyphenols, Theanine and Aroma

Note

Unit 14: Tea Chemistry in Depth: Polyphenols, Theanine and Aroma

Unit 2 gave you the map: the leaf holds catechins, fermentation turns them into theaflavins and thearubigins, aroma holds more than 600 compounds, and caffeine and tannin shape the cup. This unit goes under the map. You will follow one catechin molecule as it is oxidised, meet the theasinensins that Unit 2 did not name, watch theanine being built in the root, and learn how a laboratory actually measures each of these compounds, and where each method can mislead you.

Questions to think about as you read: Why does a mixture of two catechins oxidise faster than either one alone? Why is black tea chemistry so much harder to pin down than green tea chemistry? And why can two honest laboratories report different catechin numbers for the same tea?

14.1 Introduction: from the leaf to the cup, in more detail

Fresh tea shoots are rich in polyphenols, above all the flavanols called catechins, which may make up to 30% of dry leaf weight. Other polyphenols include flavonols and their glycosides, such as quercetin and myricetin, and depsides such as chlorogenic acid, coumarylquinic acid, and theogallin, which is 3-galloylquinic acid and is unique to tea. Caffeine sits at an average level of 3%, with very small amounts of theobromine and theophylline, and the amino acid theanine, 5-N-ethylglutamine, is also unique to tea.

Two tea types differ in chemistry. The broad-leaf Assam type holds more flavan-3-ols, with flavonols present mainly as diglycosides, while the small-leaf China type holds fewer flavan-3-ols, with flavonols present as triglycosides and more volatile flavour compounds. True China and Assam types are now rare, and most plantings are hybrids with mixed characters.

Black tea is made by crushing the leaf so its enzymes meet its polyphenols, in a step the trade calls fermentation, although no micro-organisms take part. Green tea is steamed at once after harvest so the enzymes are switched off at the start, and its polyphenol mix stays close to that of the fresh leaf. Black tea accounts for almost 80% of world tea production in the book, and its polyphenol mix is far more complex. Much of it is still not fully described, and that gap is the theme of this unit.

Quick check

Why does green tea chemistry stay simple while black tea chemistry turns complex?

Check your answer

Green tea leaves are steamed at once, so the enzymes are switched off before they can oxidise the catechins. Black tea leaves are crushed, so enzymes and catechins mix and a cascade of oxidation and coupling reactions runs on.

14.2 Polyphenols and catechins in depth

What the fresh shoot holds

The book table for fresh tea shoots gives the composition below. Use it as the reference for this unit: every later section is about what happens to the compounds in this table.

Compound% dry weightRole in the cup
Total polyphenols25 to 30Astringency
(-)Epigallocatechin gallate, EGCG8 to 12Main catechin
(-)Epicatechin gallate, ECG3 to 6Catechin
(-)Epigallocatechin, EGC3 to 6Catechin
(-)Epicatechin, EC1 to 3Catechin
(+)Catechin, C1 to 2Catechin
(+)Gallocatechin, GC3 to 4Catechin
Flavonols and flavonol glycosides3 to 4Polyphenol
Leucoanthocyanins2 to 3Polyphenol
Polyphenolic acids and depsides3 to 4Polyphenol
Caffeine3 to 4Briskness
Theobromine0.2Methylxanthine
Theophylline0.5Methylxanthine
Amino acids4 to 5Brothiness
Organic acids0.5 to 0.6Acidity
Monosaccharides4 to 5Sweetness
Polysaccharides14 to 22Body
Cellulose and hemicellulose4 to 7Fibre
Pectins5 to 6Body
Lignin5 to 6Fibre
Protein14 to 17Nutrition and flavour precursors

Table 14.1: Chemical composition of fresh tea shoots, exactly as given in the book. The catechin rows are the starting material for black tea chemistry.

Table 14.1 note: the two books disagree on the methylxanthines: Ahuja Table 18.1 gives theobromine 0.2% and theophylline 0.5%, while Ho Ch. 6 gives theobromine about 0.02% and theophylline 0.2-0.4%. Both figures are reported as printed; they have not been reconciled.

There are six major catechins in tea: (+)-catechin, (+)-gallocatechin, (-)-epicatechin, (-)-epicatechin gallate, (-)-epigallocatechin, and (-)-epigallocatechin gallate. In the shorthand used through this unit, EC is (-)-epicatechin, EGC is (-)-epigallocatechin, ECG is (-)-epicatechin gallate, and EGCG is (-)-epigallocatechin gallate. The gallate forms carry a galloyl group, written G, attached by an ester link. Catechins with a pyrogallol-type B-ring, EGC and EGCG, carry three hydroxyl groups on that ring; catechins with a catechol-type B-ring, EC and ECG, carry two. That one extra hydroxyl group changes how easily each catechin oxidises, and it decides which products form, as the next sections show.

Catechins are built by the shikimic acid pathway through chalcones, with phenylalanine and tyrosine as precursors. They are formed first and are later epimerised and galloylated during storage in the epidermal tissue. They show seasonal change: EGCG and ECG rose with the rainy season in one study cited, and in shoots grown in a warm climate EGCG was higher in the warm harvest season while EGC was higher in cool months.

The eight major catechins listed in the analytical chapter are: (+)-catechin (C), (-)-epicatechin (EC), (-)-gallocatechin (GC), (-)-epigallocatechin (EGC), (-)-catechin gallate (CG), (-)-gallocatechin gallate (GCG), (-)-epicatechin gallate (ECG), and (-)-epigallocatechin gallate (EGCG). The six named above plus (-)-catechin gallate (CG) and (-)-gallocatechin gallate (GCG) make the eight major catechins counted in analytical work. In green tea they make up to 30% of dry leaf weight. In black tea, catechin content is lower, at 10 to 12%, with theaflavins at 3 to 6% and thearubigins at 12 to 18%. Oolong tea, partly fermented, holds a mix of catechins, theaflavins, and thearubigins.

How catechins were first measured

Early methods for total polyphenols used Folin-Denis reagent, vanillin reagent, Price-Butler reagent, ceric sulphate, potassium permanganate, and diazotised amines. These reagents are non-specific: they cannot tell flavonoid classes apart and are best used for total soluble polyphenols. Flavanols were first separated by two-dimensional paper chromatography and seen under ultraviolet light. A highly specific diazotised sulphanilamide reagent was later made for tea catechins: it forms a yellow colour with catechins, and that colour is the basis of a spectrophotometric method for total and individual catechins.

Gas liquid chromatography needs the catechins to be turned into trimethylsilyl derivatives first, because they are poorly volatile. Today the method of choice for catechins is C18 reversed-phase HPLC. Early work used isocratic elution with acetic acid, methanol, dimethylformamide, and water, with UV detection at 254 nm. Later gradient HPLC with an acetonitrile, water, and acetic acid mobile phase on a 3 µm C18 column gave highly resolved flavanol peaks. In HPLC work, caffeine is measured at the same time as the catechins. Detection wavelengths tried include 275, 280, and 231 nm, and 210 nm to improve signal to noise. Electrochemical detection at +550 mV was reported to be 1,000 times more sensitive than UV detection at 280 nm. One gradient method cut the run time to 21 minutes, with the last peak at 17 minutes, against 60 minutes for an isocratic method in another study.

14.3 Theaflavin formation, step by step

Theaflavins are the most important reddish orange pigments of black tea. They hold a 1′,2′-dihydroxy-3,4-benzotropolone ring, and they form by oxidative coupling between a catechol-type catechin, EC or ECG, and a pyrogallol-type catechin, EGC or EGCG. The four major theaflavins are theaflavin, theaflavin-3-gallate, theaflavin-3′-gallate, and theaflavin-3,3′-digallate.

The steps, as worked out in model fermentation experiments, run like this:

  1. The enzyme oxidises the B-ring of EC to its quinone, called 1a. The enzyme prefers EC, so this step is fast.
  2. Quinone 1a then oxidises the B-ring of EGC to its quinone, called 2a. In doing so, 1a is reduced back to EC, so EC is regenerated and appears to be oxidised very slowly, while EGC falls fast. This is the coupled oxidation mechanism.
  3. The two quinones, 1a and 2a, join and rearrange to form the benzotropolone ring of theaflavin, called 5.

The proof for step 2 came from a neat test. When EC and EGC were treated separately with a banana fruit homogenate, EC fell faster than EGC. When the two were mixed, EGC fell faster than EC, which fits the coupled mechanism. In a second test with glutathione present, EC was fully turned into glutathione adducts of quinone 1a, while about 70% of EGC stayed unchanged. The mechanism is explained by higher enzyme specificity for EC and a lower oxidation-reduction potential for EGC.

Theaflavin itself can be oxidised further. One product is theanaphthoquinone, called 6, a yellow pigment whose structure was fixed by two-dimensional NMR and chemical derivatisation. In the model mixture, the colour first turned reddish orange as theaflavin rose, then changed sharply to greenish black when EC ran out and theaflavin began to be oxidised, probably through a quinhydrone-type complex between theaflavin and quinone 1a, and then faded to light brown as theaflavin was used up. Theaflavins carrying galloyl groups at the C-3 or C-3′ positions behaved differently: no greenish black colour appeared, and no galloyl esters of theanaphthoquinone were found, probably because the galloyl groups cover the benzotropolone ring and block the complex with quinone 1a.

Other theaflavin oxidation products found in model work are bistheaflavin A, called 7, dehydrotheaflavin, called 8, and bistheaflavin B, called 9, formed by auto-oxidation of theaflavin in phosphate buffer at pH 7.3. Compound 9 is thought to form by coupling of the reduced form of theanaphthoquinone with theaflavin quinone, which shows that theanaphthoquinone can also form without an enzyme.

Step 1: Enzyme oxidises EC, a catechol-type catechin, to quinone 1a

↓

Step 2: Quinone 1a oxidises EGC, a pyrogallol-type catechin, to quinone 2a, and EC is regenerated

↓

Step 3: Quinones 1a and 2a join and rearrange to the benzotropolone ring of theaflavin

↓

Step 4: Theaflavin can oxidise further to theanaphthoquinone, bistheaflavin A, dehydrotheaflavin, or bistheaflavin B

Figure 14.1: Theaflavin formation by coupled oxidation. What it shows: the four steps from EC and EGC to theaflavin and its oxidation products, in order. Drawn from Ho, Lin, and Shahidi, Tea and Tea Products, Ch. 5.

Quick check

Why does EGC fall faster than EC when the two are oxidised together, even though EC falls faster on its own?

Check your answer

Because of coupled oxidation: the enzyme oxidises EC to its quinone, and that quinone oxidises EGC, being reduced back to EC in the process, so EC is regenerated while EGC is used up.

How the model experiments were done

Because black tea polyphenols are so hard to separate directly, researchers built model fermentation experiments. A tea leaf homogenate from which polyphenols had first been removed by adsorption on polyvinylpolypyrrolidone, or a homogenate of pear or banana fruit, was used as the enzyme source. Early tests showed that pear or banana homogenates gave oxidation products much the same as tea leaf homogenate for the major products, and large-scale work was easier with fruit. A peroxidase and hydrogen peroxide system was also used. The enzymes mainly did the first step, oxidising o-diphenols to o-quinones, and the later complex reactions of the quinones ran without enzymes, so polyphenol oxidase and peroxidase from different plants gave similar products. The book is honest about the limits: some differences remain between model experiments and real tea leaf fermentation, above all for minor products and polymeric substances, and those differences are still to be clarified.

14.4 Theasinensins and the oxidation of pyrogallol-type catechins

Pyrogallol-type catechins, EGC and EGCG, make up more than 70% of total tea catechins, and their B-rings oxidise more easily than catechol-type rings because of their lower oxidation-reduction potential. When only EGC and EGCG are oxidised, two B-rings join by oxidative coupling.

The first products are quinone dimers, the most important being dehydrotheasinensins C, called 12, and A, called 13, which build up in crushed fresh tea leaves. When the leaves are heated and dried, these turn into theasinensins, called 14 to 17. Theasinensins are major constituents of black tea, and their total amount is comparable to that of theaflavins. The same change happens in water at neutral pH. The reaction looks like an oxidation-reduction dismutation, because theasinensins, the reduction products, appear together with oolongtheanins, called 18, and compound 19, the oxidation products.

Direct derivatisation of the dehydrotheasinensins in crushed leaves with o-phenylenediamine gave phenazine derivatives with only an R biphenyl bond, which shows that the quinone coupling is highly stereoselective: it picks one shape, not a random mix. A newer pigment, dehydrotheasinensin AQ, called 21, was found in commercial black tea and comes from isomerisation of 13, while isomerisation of 12 gives dehydrotheasinensin E, called 22, whose presence in black tea has not yet been confirmed. Compound 13 was also made by auto-oxidation of EGCG in phosphate buffer at pH 7.4, but it is unstable at pH 7 and was seen only at the start of the reaction.

Other quinone dimers of EGCG were also found. EGCG quinone dimer A, called 23, was unstable in hot water and gave EGCG and compound 24, which holds the A- and C-rings of EGCG and was isolated from black tea. Its formation is explained by a retro-Aldol-type reaction of 23 giving quinone 4a, followed by oxidation-reduction dismutation. EGCG quinone dimer B, called 25, has a caged structure with two carbonyl groups. Compound 26 was first isolated in a model fermentation with tea leaf homogenate and later found in commercial black tea; its formation needs a more complex mechanism including decarboxylation, not simple dimerisation of quinone 4a.

Two EGC or EGCG molecules oxidise to hydroxy o-quinones 2a and 4a

↓

Two quinones join B-ring to B-ring: dehydrotheasinensins 12 and 13 build up in crushed leaves

↓

Heating and drying, or neutral pH water: dismutation gives theasinensins 14 to 17 plus oolongtheanins 18 and compound 19

↓

Isomerisation gives dehydrotheasinensin AQ, 21, found in commercial black tea

Figure 14.2: From pyrogallol-type catechins to theasinensins. What it shows: the B-ring to B-ring coupling, the dismutation on heating, and the isomerisation product. Drawn from Ho, Lin, and Shahidi, Tea and Tea Products, Ch. 5.

Try it yourself: On one sheet, draw two columns. Left column: theaflavin route, one catechol-type catechin plus one pyrogallol-type catechin, joined through a benzotropolone ring. Right column: theasinensin route, two pyrogallol-type catechins, joined B-ring to B-ring, then changed by heating. Under each, write the one fact that sets it apart: theaflavin needs the mixed pair and the coupled oxidation trick, and theasinensin needs the heat step to turn the dehydro form into the final form.

14.5 Other routes: catechol oxidation, benzotropolone by a second path, and the theanine link

Oxidation of catechol-type catechins alone

When only catechol-type catechins, (+)-catechin or EC, are oxidised, the first B-ring o-quinone acts as an electrophile and reacts with the phloroglucinol A-ring or the catechol B-ring of another molecule. The dimeric products, typified by dehydrodicatechin, called 10, can react further to give oligomers and polymers, such as the trimer 11. These products were not isolated from fermented tea leaves, because catechol-type catechins are present at relatively low levels. A theaflavin-related benzotropolone pigment holding a procyanidin unit was also made in vitro: although proanthocyanidin dimers have two B-rings, the joining happened only at the extension, upper, unit, because of steric effect.

Another pathway for benzotropolone formation

Epitheaflagallin, called 29, and its 3-O-gallate carry a 1′,2′,3′-trihydroxy-3,4-benzotropolone ring, a different hydroxylation pattern from theaflavin. They were first made in vitro from epigallocatechin and pyrogallol or gallic acid, and it was assumed they formed the same way in crushed leaves. The book reports a different finding: compound 29 is made directly from epigallocatechin, through an unstable quinone dimer called proepitheaflagallin, called 30, which breaks down on heating to give 29 plus a new pigment, hydroxytheaflavin, called 31. The contribution of 30 to total catechin oxidation was relatively low, but this is the first reported oxidation route for benzotropolone pigments from epigallocatechin alone.

A polyphenol with a theanine Strecker aldehyde part

Theanine is a characteristic amino acid of tea, and in the heating step of tea making, aldehydes that add to tea aroma are thought to form by a Strecker reaction between amino acids and the carbonyl groups of sugars or catechin quinones. Researchers isolated a new polyphenol, called 28, in which an N-ethyl-2-pyrrolidinone part, equal to the theanine Strecker aldehyde, called 27a, was attached to the A-ring of theasinensin A, called 16. The structure was confirmed by partial synthesis, and the joining of the aldehyde with the A-ring was shown to be reversible. Related adducts of 27a with single catechins were not found in black tea, probably because they are less stable than 28, and the link may be partly stabilised by the second catechin unit in the structure of 28.

Why black tea polyphenols are so complex

The book sums up the complexity in one cascade. Simple catechin quinones form first, then cross-coupling of quinones gives many products, and some of those products are unstable and break down into still more products. The confirmed members in commercial black tea are theaflavins and their galloyl esters, theasinensins 14 to 17, oolongtheanins 18, and compounds 21, 24, 26, 28, and 29. The rest of black tea polyphenols show as a broad hump on the HPLC baseline, a complex mix of larger oxidation products that do not give clear spots on thin-layer chromatography and make up the majority of black tea polyphenols.

Two points limit how far the B-ring coupling story can go. First, there is no evidence so far that B-ring coupling alone builds trimers and larger molecules at the start of fermentation, and mass spectrometry of the total oxidation mix of EGC suggested a trimer and a tetramer, but their peaks were much smaller than the dimer peaks. Second, galloyl groups take part: oxidation of galloyl groups can join a theasinensin to another EGCG B-ring through a biphenyl bond, giving trimeric to oligomeric pigments such as compound 33, in which one of the two galloyl groups of theasinensin A is linked to another EGCG B-ring. Oxidation of pyrogallol-type B-rings runs much faster than oxidation of galloyl groups. A further route, electrophilic substitution at catechin A-rings giving dehydrodicatechin-type products, may also take part in polymer formation. A proanthocyanidin-type C-4 to C-8 link in the polymer fraction has been proposed, but the book states plainly that there is, to date, no evidence that such structures can be made from flavan-3-ols by chemical or enzymatic means. The book closes the chapter by saying that chemical evidence so far is still not enough to build a general structure for the polymeric oxidation products, and researchers are still a long way from a full chemical understanding of all black tea polyphenols.

Quick check

What two mechanisms extend black tea polyphenols from dimers toward larger molecules?

Check your answer

Oxidative coupling of galloyl groups, which links a theasinensin galloyl group to another catechin B-ring, and possibly electrophilic substitution at catechin A-rings. B-ring coupling alone has no evidence so far for building trimers at the start of fermentation.

14.6 L-theanine in depth

Identity and taste

L-theanine is γ-glutamylethylamide, also written 5-N-ethylglutamine, with the formula C₇H₁₄N₂O₃. It was first found in tea leaves in the early 1950s, and the only other natural source reported in the book is the edible mushroom Xerocomus badius. It is a non-protein amino acid, and it makes up more than 50% of the total free amino acids in tea. Its content in tea is between 1.5 and 3% of dry weight, and it gives tea a sweet, brothy, savoury taste called umami. Its taste threshold was about 0.06%, against about 0.15% for other amino acids, which may be why theanine is used as one of the most important quality indices in world tea markets. Theanine and glutamic acid are the most important umami substances in green tea, and theanine mellows and broadens a flavour that is otherwise dominated by the bitter taste of polyphenols and caffeine.

Pure theanine forms colourless needles, melts at 217 to 218°C (the natural product) with decomposition, and is very soluble in water: one gram dissolves in 2.6 ml water at 0°C and in 1.8 ml at 100°C, but it does not dissolve in ethyl alcohol or ether. It gives a strong ninhydrin reaction. Natural theanine in tea is mostly the L-form, with small amounts of the D-form also found. Hydrolysis of theanine gives L-glutamic acid and ethylamine.

Theanine is stable under acid conditions: it showed little change at pH 3 for one year, and little difference in hydrolysis between pH 3 and pH 7. It is far less stable under alkaline conditions: at pH 11 it hydrolysed much more to glutamic acid and ethylamine, and at pH 12 and 40°C it broke down to glutamic acid plus ethylamine. In water it slowly racemises, and it undergoes hydrolysis, above all at basic pH. Heating green tea infusion at 120°C did not change theanine and other amino acid levels in one study cited.

Theanine is about 2% of dry leaf weight in general, and higher-grade shaded green teas hold about 2% on a dry weight basis. In shaded leaves, theanine builds up in large amounts while catechin content falls, compared with unshaded leaves. Younger shoots, the bud and first leaf, hold more theanine than mature second, third, and fourth leaves. Theanine content falls as fermentation rises and as drying temperature rises. In the roots, theanine rose with rising ammonium fertiliser, and high levels were found in roots in winter months when bushes were dormant.

Biosynthesis in the plant

Theanine is made in the root of the tea plant by the enzyme theanine synthetase, also called L-glutamate:ethylamine ligase, EC 6.3.1.6, using ATP and Mg²⁺. The reaction is: ATP plus L-glutamate plus ethylamine gives ADP plus phosphate plus N5-ethyl-L-glutamine, which is theanine. The enzyme has a high affinity for ethylamine and little affinity for ammonia, D-glutamate, and L-aspartate, and theanine synthesis needs ATP but is independent of glutamine synthesis. The enzyme is quite labile, and the book reports no purification and property study of it.

The ethylamine part of theanine comes from L-alanine, through free ethylamine formed by alanine decarboxylase in the root. After it is made in the root, theanine moves to the tops of the plant, where it serves as the main source of soluble nitrogen for building other carbon skeletons, such as flavanols. The first leaf was the main site for building polyphenolic compounds from L-theanine. The N-ethyl carbon of theanine can be turned into the phloroglucinol nucleus of catechins, and that step is controlled by light. During germination, theanine level reached a peak after 45 days. Overall amino acid content ran in the order: roots, then stems of new shoots, then buds plus first leaves, then second leaves, then third leaves, with roots holding about two to nine times more amino acids than other organs.

In the root: L-alanine gives ethylamine, by alanine decarboxylase

↓

Theanine synthetase, EC 6.3.1.6, joins L-glutamate and ethylamine with ATP and Mg²⁺

↓

L-theanine moves from root to shoots and young leaves

↓

In the leaf it stores nitrogen, feeds flavanol building, and gives umami taste

Figure 14.3: How theanine is made and where it goes. What it shows: the root synthesis from glutamate and ethylamine, and the move to young leaves. Drawn from Ho, Lin, and Shahidi, Tea and Tea Products, Ch. 16, and Ahuja et al., Science of Tea Technology, Ch. 18.

Making theanine outside the plant

Theanine is hard to isolate from tea, and isolation is expensive, so several production routes have been tried. Biosynthesis with theanine synthetase is held back by the instability of the enzyme and the need for an ATP generating system. One route coupled baker's yeast preparations with bacterial glutamine synthetase to make theanine from glutamate and ethylamine, but the yield was not satisfactory because glutamine synthetase reacts poorly with ethylamine and pH control was hard. Growing tea cells in culture took four weeks. An enzymatic route with glutaminase from Pseudomonas nitroreducens was reported, and a bioreactor was built with immobilised cells of that bacterium, held in kappa-carrageenan and packed in columns, to make theanine continuously from glutamine and ethylamine. After immobilisation the enzymes became stable, and the reactor kept its activity for at least several weeks. A bacterial γ-glutamyltranspeptidase route gave a 60% conversion from glutamine.

Chemical synthesis was first done with a 9% yield from pyrrolidonecarboxylic acid treated with 33% ethylamine solution at 37°C for 20 days. Glutamic acid can be turned into that starting acid with a 70 to 90% yield. A large-scale synthesis from γ-benzyl glutamate gave white crystals, melting point 214 to 216°C (the large-scale synthesis product), with a 33.9% yield. A later industrial-scale route used L-glutamic acid, phthalic anhydride, and ethylamine in four steps, with a yield of over 50%. The book notes that most of these attempts stayed at the developing stage because of low yields, low purity, and complex processes, and that theanine products are now available in the market but are expensive.

14.7 What theanine does after drinking: what the book reports

Read this section as a report of the studies in the book, not as health advice. The studies are a mix of animal work, small human tests, and cell work, and the book itself flags the limits, such as a relaxation test done only in eight young women.

After drinking, L-theanine is absorbed through the brush-border membrane of the small intestine and spread to tissues. It crosses the blood-brain barrier through the leucine-preferring amino acid transporter system. It does not appear to build up: it is hydrolysed by enzymes to glutamic acid and ethylamine in blood, kidney, liver, and brain, most of the ethylamine is excreted in urine at once, and theanine is fully absent 24 hours after dosing. Gut absorption of D-theanine was far lower than that of L-theanine, and the D-form seems to be excreted with little metabolism, while the L-form is preferentially reabsorbed and metabolised by the kidney. No toxic effects have been reported for theanine, even in large amounts, in the book.

The book reports that theanine raises dopamine, lowers serotonin, and raises γ-aminobutyric acid, written GABA, in the brain, and that these shifts may lie behind its effects on mood, memory, learning, and blood pressure. In a human brain-wave test, 50 female university students aged 18 to 22 were screened, and eight women, four with high anxiety and four with low anxiety, took a theanine solution once a week at a fixed time. A dose of 200 mg of theanine in 100 ml water gave alpha brain waves, at 8 to 13 Hz, in the occipital and parietal regions about 40 minutes after drinking, while water alone gave only a small amount. The 200 mg dose gave more alpha waves than a 50 mg dose, above all in the high-anxiety group, and theta waves did not change, which the book reads as relaxation without drowsiness. The book states plainly that this effect has not yet been confirmed in other human populations.

In animal tests, theanine improved performance in active and passive avoidance tests of memory and learning. In one study, weanling male rats given 180 mg of theanine per day for four months did better in an operant test, where feed comes out when a lever is pushed and a light turns on, and the theanine group stayed longer in the light room in a passive avoidance test and reacted faster in an active avoidance test. Theanine also lowered blood pressure in spontaneously hypertensive rats at a high oral dose, and raised dopamine in the striatum, hypothalamus, and hippocampus.

For neuroprotection, theanine is similar in structure to glutamate, an excitatory neurotransmitter, and it can act as a competitive antagonist at glutamate receptors, guarding brain cells against glutamate toxicity. In gerbils, theanine given through the lateral ventricle 30 minutes before a three-minute blockage of the carotid arteries prevented ischaemic neuronal death in the hippocampus in a dose-dependent way, at doses of 50, 125, and 500 mM. These findings are from animal work, and the book presents them as a possible clinical use, not a proven one.

The book also reports work on theanine with the cancer drug doxorubicin. In tumour cells, theanine lowers glutamate uptake through glutamate transporters, which lowers glutathione synthesis and the formation of the glutathione-doxorubicin conjugate, and so lowers the pumping of the drug out of the tumour cell, raising drug levels in the tumour. In normal cells, the opposite happens: theanine is turned into glutamate, glutathione rises, and doxorubicin is pumped out more, which lowers toxicity. Theanine also reduced tumour growth in an animal liver cancer cell line, suppressed angiogenesis, and slowed invasion. These are animal and cell findings, and the book presents them as a hope for future cancer chemotherapy, not as an established treatment.

Two more reported effects are worth noting. Theanine can soften the stimulating effects of caffeine, by raising dopamine and GABA and shifting serotonin, with relaxation felt about 30 to 40 minutes after drinking. With green tea catechins and caffeine, theanine reduced body weight, fat build-up, triacylglycerols, and free fatty acids in animals, possibly by raising dopamine release and lowering serotonin, which may help control appetite. Finally, ethylamine, the breakdown product of theanine, primed gamma-delta T cells: in a study cited, 50% or more of tea drinkers showed strong gamma-delta T cell responses within two weeks of tea drinking, lasting through the fourth week, while coffee drinkers showed no notable response.

Quick check

Why does the book treat the human relaxation finding with caution?

Check your answer

Because the test was done in only eight women, all aged 18 to 22, and the book states plainly that the effect has not yet been confirmed in other human populations.

14.8 Caffeine in depth

Caffeine adds to astringency with catechins and their derivatives in green tea, and it gives briskness and creaming in black tea infusion. Cream is a thick turbid layer that forms on the surface when a good cup cools: a complex of theaflavins, thearubigins, and caffeine, held by weak hydrogen bonds, which changes the taste of both caffeine and theaflavins for the better.

Tender younger leaves hold more caffeine than older leaves. Caffeine falls from about 4.7% in the bud to about 2.9% in the third leaf, and leaf stalks hold low levels, so fine plucking gives high caffeine in the brewed tea. There is a positive link between caffeine in the green two leaves and a bud shoot and tasters' preferences for black teas. Caffeine shows seasonal change, and high relative humidity was the only climate factor significantly linked with caffeine in one study cited.

Extraction of caffeine into the cup follows first-order rate kinetics. The activation energies were 45.2 kJ mol⁻¹ for loose tea and 43.2 kJ mol⁻¹ for tea-bag tea. During black tea making, caffeine rises during withering in a time-dependent way, independent of the rate and degree of drying, falls during fermentation in line with time and temperature, and is lost in small amounts by sublimation during drying. Tea stored for six months held more caffeine than freshly made tea. The ability to make and store caffeine in green leaf differs by clone.

The caffeine biosynthesis route runs in four steps with three methylation steps, using S-adenosyl-L-methionine, written SAM, as the methyl donor. SAM is turned into S-adenosyl-L-homocysteine, written SAH, which is hydrolysed to L-homocysteine and adenosine. The route is: SAM to SAH to adenosine to adenine to AMP to IMP to XMP to xanthosine to 7-methylxanthosine to 7-methylxanthine to theobromine to caffeine. Caffeine synthase, EC 2.1.1.160, carries the final two steps, adding two methyl groups in turn to 7-methylxanthine to give theobromine and then caffeine. The gene for caffeine synthase from young tea leaves has been cloned. Theobromine is the direct precursor of caffeine, and its conversion to theophylline is the rate-limiting step in purine alkaloid breakdown, so caffeine builds up in the leaf.

Decaffeination has been tried by solvent extraction, column chromatography, and hot water treatment. Chloroform and methylene chloride remove caffeine but are not recommended because of their toxicity. Hot water treatment removes soluble solids and secondary metabolites with the caffeine and is expensive. Supercritical carbon dioxide, written CO₂, has been used to decaffeinate green and black tea, with lower energy costs than conventional solvent distillation, but most flavour volatiles fell during the process. In a separate laboratory route, an RNAi construct built from a partial fragment of caffeine synthase was put into somatic embryos, and the transformed embryos showed lower caffeine synthase transcript levels and lower caffeine content.

14.9 Lipids and aroma biogenesis during manufacture

Lipid content rises with shoot maturity, from 30 mg/g in the bud to almost 72 mg/g in the fourth leaf. The order of fatty acids in leaves and stalk is: linolenic acid, then linoleic acid, then palmitic acid, then stearic acid, then oleic acid, then palmitoleic acid, whatever the plucking standard or shoot part. The stem holds the least fatty acids. Linolenic acid is the most abundant, and it rises with leaf maturity and coarser plucking. Total fatty acids, total unsaturated fatty acids, and linolenic and linoleic acids are higher in coarse pluck than in standard pluck, which is why quality falls with the higher amount of Group I volatile compounds from coarse plucking.

Flavour is taste plus aroma. Non-volatile compounds give taste, and volatile compounds give aroma. Volatile compounds are present at only 0.01% to 0.03% of total dry weight, and more than 600 volatile compounds have been found in tea. The book table groups them by chemical class: hydrocarbons 72, alcohols 89, aldehydes 68, ketones 75, acids 69, esters 82, lactones 25, phenolic compounds 22, oxygenated compounds 36, nitrogenous compounds 86, and sulphur compounds 14. Aroma has a high impact on flavour because smell is more developed than taste, and the mix, amount, and odour threshold of the volatiles decide the final flavour and the value at auction.

Volatiles fall into two groups. Group I holds mainly non-terpenoids with a grassy aroma, such as hexenols, hexanals, hexenals, pentanol, and 1-penten-3-ol, with notes such as fresh green, resinous, fatty, biscuity, soapy, and balsamic. Group II holds mainly terpenoids with a sweet flowery aroma, such as linalool and geraniol, plus benzaldehyde, phenylacetaldehyde, methyl salicylate, benzyl alcohol, and β-ionone, with notes such as sweet, almond, honey, hawthorn, rosy, lemon, peppermint, and violet. Group II compounds are highly desirable. The ratio of Group II to Group I, called the flavour index, is used to grade black teas for flavour quality: quality rises as the flavour index rises.

How aroma is made during manufacture

Black tea making runs through plucking, withering, rolling, fermentation, and drying, and volatiles are freed from their precursors when the leaf structure is broken. Three kinds of mechanism take part.

First, oxidative degradation. Carotenoids are turned into β-ionone derivatives by oxidative degradation during fermentation and drying, and carotenoid content falls during the process. They are present in tiny amounts but matter because their perception threshold is very low. Methyl epijasmonate, a flowery compound, comes from carotenoid degradation: the (+) Z form is 400 times more flavoury than its (-) Z isomer, methyl jasmonate, which is reported to be odourless. The cis form is unstable and is easily changed to the trans form by heat.

Second, thermal generation. Aldehydes, furans, pyrazines, and pyrroles are made from sugars and amino acids during drying, the final step, and they give a slight roast aroma. Short-chain volatile acids such as propanoic, butanoic, hexanoic, and heptanoic acids are thermal breakdown products of fats or long-chain fatty acids. Lactones are made by dehydration of hydroxyacids during drying and add to oolong tea aroma. β-ionone and β-cyclocitral are heat-made products of β-carotene.

Third, enzyme-catalysed reactions. The green odour of green leaves comes from (E)-2-hexenal and (Z)-3-hexenal, made from unsaturated fatty acids such as α-linolenic and linoleic acids through their hydroperoxides. Four enzymes take part: lipolytic acyl hydrolase, lipoxygenase, hydroperoxide lyase, and alcohol dehydrogenase, with lipoxygenase and hydroperoxide lyase the most important. Lipoxygenase adds oxygen to the fatty acids to give 13-(S)-hydroperoxides, and hydroperoxide lyase cuts the bond between C-12 and C-13 to give C6 aldehydes, which give a grassy odour and are undesirable. Shoots with high fatty acid content give poorer teas.

Many flavour volatiles are stored in the leaf as non-volatile β-D-glycosides, with the sugar part as a monoglucoside or a diglycoside such as primveroside, vicianoside, or rutinoside. Terpenoid alcohols are freed by one-step enzymatic hydrolysis of their β-glucoside or β-primveroside precursors, by β-glucosidase and by the specific β-primverosidase. Rolling breaks the cells, bringing the precursors in the cytoplasm into contact with the glycosidase enzyme in the vacuole, so linalool and geraniol are freed by hydrolysis of their bound forms in the broken shoots. (Z)-3-hexenol, linalool oxide I, methyl salicylate, benzyl alcohol, and 2-phenylethanol are also freed from their glycosides in the same way.

The book adds a wider point about flavour and climate. Extra-chloroplastidic building of terpenoid compounds from leucine leads to flavoured tea, while intra-chloroplastidic fixation of CO₂ and building of terpenoids from acetate gives teas without flavour. The leucine route runs in climatic stress that limits CO₂ fixation, while the acetate route runs in vigorous growth.

Stored precursors: glycosides in the cytoplasm, enzymes in the vacuole

↓

Rolling breaks cells: enzymes meet precursors

↓

Three routes run: enzyme action frees linalool and geraniol, carotenoids oxidise to ionones, heat makes roast notes in drying

↓

More Group II than Group I gives a higher flavour index and better flavour quality

Figure 14.4: How tea aroma is made during manufacture. What it shows: the three routes from stored precursors to finished aroma, and the flavour index. Drawn from Ahuja et al., Science of Tea Technology, Ch. 18.

Try it yourself: Smell a green tea and a black tea side by side, then sort the words below into Group I or Group II using this section only: fresh green, grassy, rosy, lemon, honey, almond, biscuity, peppermint, violet. Check your sort against the table in section 14.9, and note which tea smells more flowery and which smells more grassy.

14.10 Polyphenol oxidase: the enzyme that starts fermentation

Polyphenol oxidase, written PPO, is a copper-holding enzyme of tea that catalyses the change of catechins to theaflavins. There are three isozymic forms, and the active purified enzyme is 72 kD. It stays active from pH 4.03 to 7.00, with an optimum temperature of 30°C. The purified enzyme could not oxidise monophenols or p-quinol, but it could oxidise catechol. Catechin was the best substrate, with a Km of 0.49 mM. The enzyme is inhibited by tropolone. It sits in the chloroplast and is solubilised with Triton-X-100.

The ratio of PPO to catechins helps decide the quality potential of tea. PPO activity is higher in tender leaves than in mature leaves, and it falls during withering and storage of fresh leaves. Very hard withering, or a heavy loss of moisture, causes a heavy loss of 50 to 55% of PPO activity. Enzyme activities differ by clone, season, and shoot maturity, and field practices such as plucking standard and pruning have a large effect on them. The activities link positively with tasters' scores. The loss of activity during withering could be restored by rehydration, and residual activity was found in made tea. Adding enzymes raised black tea quality markedly in cuppage and creaming properties.

The book also reports nitrogen assimilation enzymes in tea. With nitrogen at 100 to 300 kg N per ha, nitrate reductase and nitrite reductase activities fell with nitrogen above 200 kg N per ha, while glutamate dehydrogenase activity rose up to 300 kg N per ha. Potassium raised nitrate reductase activity. Phosphate up to 50 kg P₂O₅ with 150 to 200 kg N per ha raised the activities of nitrate reductase, nitrite reductase, glutamate dehydrogenase, glutamate synthase, and acid phosphatase in shoot and roots, while higher phosphate lowered them. High nitrogen at 200 to 300 kg N per ha raised amide, mainly theanine, in feeder roots and lowered starch, and with adequate potassium the amide build-up fell and nitrate reductase activity and starch rose. NPK at 330, 60, and 290 kg per ha harmed catechin building in shoots and so lowered black tea quality, because theaflavins, thearubigins, brightness, and total colour all depend on catechins.

Intact leaf: catechins and PPO are kept apart

↓

Rolling breaks cells: PPO meets catechins, with oxygen

↓

PPO oxidises catechins to quinones: best substrate catechin, Km 0.49 mM, optimum 30°C, active pH 4.03 to 7.00

↓

Quinones join to give theaflavins, theasinensins, and thearubigins: drying stops the enzyme

Figure 14.5: Polyphenol oxidase at work. What it shows: how rolling brings enzyme and catechin together, and the conditions the enzyme needs. Drawn from Ahuja et al., Science of Tea Technology, Ch. 18.

14.11 Analytical methods: getting the sample ready, and HPLC

Tea used to be graded by tasters, by flavour and colour, but chemical analysis gives more accurate and reliable measures of quality, linked to leaf age and degree of fermentation. Analysis is also needed for health-promoting food products, which must deliver the same bioactive compounds in a complete and consistent way, and for studying what happens to tea compounds in the body.

Sample preparation

Catechins are usually extracted with water or organic solvents, then filtered and used directly. Water is popular, but a water and organic solvent mix extracts better, and organic solvents extract catechins without unwanted epimerisation. Water alone can still work if time and temperature are set with care. The highest extraction in one study came from boiling aqueous acetone above 95% and 50% acetonitrile for two hours, but a similar yield of 97 to 98% came from water alone at 80°C for 20 minutes. Longer extraction gave lower yields because catechins broke down, and extraction above 80°C can promote epimerisation, such as forming GCG from EGCG, which changes the catechin mix of the sample. Lower pH helps avoid that problem, and alkaline conditions are not recommended because they greatly speed catechin breakdown.

After extraction, catechins are often cleaned up to remove caffeine, which can interfere with analysis. Liquid-liquid partition with chloroform takes caffeine out of a tea infusion, and catechins left in the water layer can be cleaned further with an ethyl acetate and water partition. Other routes are solid-phase extraction on C18 cartridges and precipitation of catechins with calcium carbonate.

Methylxanthines extract poorly in water or pure organic solvents such as acetone, methanol, and acetonitrile: in those extracts theophylline was not found. A water and organic solvent mix works better, with the best result from 70% methanol, and boiling water also extracts all three alkaloids. Polyphenols can be removed with chloroform and 2-propanol at 80:20, or by passing extracts through a basic aluminium oxide column or a heavy magnesium oxide column, or by adding polyvinylpolypyrrolidone, written PVPP, which adsorbs polyphenols well. A PVPP precolumn can do the clean-up on line, with better efficiency than batch treatment.

Theaflavins are extracted much like catechins, with boiling water or organic solvents. They are stable under slightly acid conditions at pH 6.5, break down slowly at pH 7.0 and 7.5, and fast at pH 9.0, although one study found breakdown in water extracts was small even after 815 minutes.

HPLC: the workhorse

HPLC is the method of choice for catechins, because it separates well and can be joined to many detectors. A reversed-phase C18 column is used in most cases, with aqueous methanol or acetonitrile as the mobile phase, and acid in the mobile phase is needed for full resolution and to stop peak tailing. The benchmark for full separation was complete resolution of the eight major catechins, first reported with a C18 column, a water, acetonitrile, and phosphoric acid gradient, and UV detection at 231 nm, with a limit of detection of about 0.2 ng for all catechins.

Later work separated twelve catechins in one run, including minor catechins and methylated derivatives, by HPLC-MS on a C18 column with a water and acetonitrile gradient holding 0.05% trifluoroacetic acid, for 35 minutes, with sensitivity down to femtomole to low picomole levels. Another study separated twelve catechins by HPLC with electrochemical detection, with limits of detection of 10 to 40 pmol per ml.

Catechins absorb UV light with maxima at 210 nm and 270 to 280 nm, so UV and diode array detectors are the most popular. But UV has two weak points: it cannot tell an analyte from a co-eluting interference, so it is less useful in complex matrices such as human plasma and saliva, and it needs baseline separation of all analytes, which gives long run times above 60 minutes and low throughput. That is why more sensitive and selective detectors were brought in.

Quick check

Why can hot, long extraction give a wrong catechin result?

Check your answer

Because catechins break down on long extraction, and heat above 80°C promotes epimerisation, such as forming GCG from EGCG, which changes the catechin mix away from the original sample.

14.12 Analytical methods: the detectors and the alternatives

HPLC-MS

Mass spectrometry gives molecular weight and characteristic fragments for identifying catechins. Electrospray ionisation, atmospheric pressure chemical ionisation, and thermospray ionisation are the common interfaces between LC and MS. In selected ion monitoring mode, only a few catechin ions are scanned, which raises signal to noise and sensitivity. One study found gallocatechin co-eluting with gallic acid, yet the selective mode saw only the gallocatechin ion, so the interference did not affect the result. Limits of detection of 5 to 15 ng per ml were reported for catechins in cosmetic and skin extracts in 15 minutes, and a solid-phase microextraction method gave limits below 0.5 ng per ml, a tenfold gain over direct injection.

Tandem MS, written MS/MS, with collision-induced dissociation lets researchers assign structures. For catechin, retro Diels-Alder fragmentation of the A-ring gives an ion at m/z 139, and a mass shift of 14 Da in that ion shows methylation of ring A. B-ring methylation shifts B-ring ions from m/z 123 and m/z 165. In negative ion mode, EGCG gave major ions at m/z 125 for an unmodified A-ring, m/z 169 for loss of an intact gallic acid anion, and m/z 305 for neutral loss of gallic acid, and a shift of m/z 169 to m/z 183 shows a methylated gallic acid part.

LC-MS is also used for catechins in plasma and urine. One capillary LC-MS method found EGCG in human plasma with a detection limit of about 20 ng. A tandem LC-LC-MS method found six catechins in human plasma with limits of 0.6 to 2 ng per ml. The first LC-MS method for all eight major catechins in biological fluids showed that all eight were absorbed into human plasma after drinking a commercial tea beverage. Catechins are turned in the body into glucuronide and sulphate derivatives, and monoglucuronides and monosulphates of EGC and EC were the main conjugates in urine, with ring-fission metabolites M4 and M6 also found.

HPLC with electrochemical, fluorescence, and chemiluminescence detection

Catechins are easily oxidised, so an electrochemical detector, written ECD, sees them with high sensitivity and more selectivity than UV, at lower cost than MS. One study reported a sensitivity of 40 pmol, 1,000 times higher than UV detection. Caffeine, which co-elutes with EGC and blocks its measurement by UV, is electrochemically inactive, so ECD removes that interference. ECD was reported to be about 45 to 320 times as sensitive as UV. In one study, eight major catechins were separated on a C18 column, with EGCG and catechin read at 400 mV and the rest at 900 mV, and limits of detection of 0.1 to 1.5 ng per ml. Maximum response comes at about 750 to 800 mV, but at that voltage many other compounds are also oxidised, so about 500 to 600 mV is recommended. A microbore C18 column method gave limits of 290 to 460 pg per ml for the eight catechins at 600 mV.

Fluorescence detection, at 280 nm excitation and 310 nm emission, suits catechins in biological fluids because of its selectivity. Limits of detection and quantification for (+)-catechin in plasma were 8 and 20 ng per ml in the first report. Fluorescence was much more sensitive than UV, with limits of about 2 to 3 ng for catechins and methylated derivatives, but only C, EC, and their methylated derivatives show much fluorescence, so the method is limited to those compounds.

Chemiluminescence detection works by two routes. In the first, polyphenols with vicinal hydroxyl groups give light when reacting with hydrogen peroxide and acetaldehyde, and EGCG was measured at picomole levels in plasma, with a limit of about 2 pmol, similar to ECD. In the second, aromatic compounds give light on UV irradiation when mixed with tris(2,2′-bipyridine)ruthenium(III), and EC and EGCG were measured with limits of 0.8 and 1.2 pmol. The first route does not work for methylated catechins, while the second may work for all major catechins because they all hold an aromatic ring.

Capillary electrophoresis, NIRS, NMR, HSCCC, planar chromatography, GC, and spectrophotometry

Capillary electrophoresis, written CE, gives short run times under 20 minutes, low running costs, and less solvent than HPLC. Three forms are used: capillary zone electrophoresis, written CZE, micellar electrokinetic chromatography, written MEKC, and microemulsion electrokinetic chromatography, written MEEKC. MEKC resolution was better than CZE for catechins. In CZE, borate forms complexes with vicinal hydroxyl groups on catechins, giving one negative charge per complex and longer elution times. One CZE method separated five catechins, caffeine, theanine, and ascorbic acid in under 11 minutes with 20 mM borax buffer at pH 8. One MEKC method separated seven catechins, three methylxanthines, and gallic acid within 4.5 minutes, with limits of detection of 0.0011 to 0.0051 µg per ml, though caffeine and (+)-catechin were not fully resolved. MEEKC separated six catechins in 10 minutes, with limits of 0.391 and 0.781 µg per ml for C and GC, and it uses acid conditions, while CZE and MEKC often use alkaline conditions that promote catechin breakdown. MEEKC had higher efficiency, at 246,000 to 1,302,000 plates per m, but limits about three times higher than MEKC.

Near-infrared reflectance spectroscopy, written NIRS, is fast and non-destructive and suits large numbers of samples. It predicted EC, ECG, EGC, and EGCG with R² above 0.85 against HPLC-UV in one study of 95 leaf samples, and analysis can take as little as one minute. A second study gave R² above 0.83 for EGCG but above 0.44 for ECG, which was seen as too low for accurate measurement, possibly because ECG was present at low levels.

NMR can identify catechins from their proton and carbon shift patterns. A high-resolution proton NMR method found five main catechins and five unknown catechins in a green tea extract, with catechin signals at H-3, 3.80 to 5.20 ppm, and H-4, 2.50 to 3.10 ppm, and it separated high-quality Longjing-type tea from most other teas in the study. In urine after tea drinking, catechin metabolites were seen at 6 to 7.3 ppm, but further work was limited by their low amounts.

High-speed countercurrent chromatography, written HSCCC, is a liquid-liquid method that keeps the stationary phase in a coil while mobile phase flows through it. A one-step method separated multimilligram amounts of EGCG, ECG, and epiafzelechin gallate in 300 minutes with a hexane, ethyl acetate, methanol, and water system at 1.5:5:1.5:5, at 2.8 ml per min. A fuller profiling study separated catechins, theaflavins, gallic acid, and purine alkaloids in 80 minutes, but only ECG and EGCG were baseline separated.

Planar chromatography, thin-layer chromatography, is cheap. Five catechins were separated on silica gel and read by UV at 254 nm. High-performance TLC on cellulose plates with a vanillin and phosphoric acid spray separated all eight major catechins in one solvent system, though only well enough for qualitative work, while a second solvent system gave the best resolution of the five major catechins in green tea extracts. Two-dimensional paper chromatography separated six catechins, and diazotised sulphanilamide spray gave a limit of detection below 1 µg of catechin on the paper.

Gas chromatography needs catechins to be turned into volatile trimethylsilyl derivatives first. GC-MS for C, EC, and four methylated derivatives gave limits of 2 to 3 pg, over a thousand times more sensitive than HPLC with fluorescence or electrochemical detection, but it cannot directly measure polar conjugated metabolites in biological fluids, and GC, CG, GCG, and other minor catechins have not been separated by GC so far.

Spectrophotometry gives total values, not individual compounds. Diethyl amino cinnamaldehyde, written DMACA, forms a green complex with catechins with peak absorbance at 637 nm, with a limit of detection of 0.05 µmol per litre, but it also reacts with theaflavins and thearubigins, so black tea results can be wrong. Diazotised sulphanilamide forms a coloured complex with a maximum at 425 nm by reacting with the A-ring of catechins, and it excludes flavonols and phenolic acids.

MethodWhat it is best forLimit reported in the book
HPLC-UVRoutine catechin separation, eight major catechinsAbout 0.2 ng; cannot tell co-eluting compounds apart; runs above 60 minutes
HPLC-MSTrace catechins, metabolites in plasma and urineBelow 0.5 ng per ml with microextraction; needs standards for firm assignment
HPLC-ECDSensitive catechin work, caffeine does not interfere0.1 to 1.5 ng per ml; high voltage loses selectivity
HPLC-fluorescenceC, EC, and methylated derivatives in fluidsAbout 2 to 3 ng; most catechins show little fluorescence
CE, MEKCFast routine catechin analysis4.5 minutes for 11 compounds; caffeine and (+)-catechin not fully resolved
NIRSFast, non-destructive screening of many samplesAbout 1 minute; ECG prediction weak at R² above 0.44 in one study
GC-MSVery sensitive C, EC, and methylated derivatives2 to 3 pg; needs derivatisation; cannot measure conjugated metabolites directly
SpectrophotometryTotal catechins or total theaflavins, cheap and fastCannot measure individual compounds; DMACA also reacts with theaflavins and thearubigins

Table 14.2: Choosing an analytical method for tea catechins. Values are the limits and run times reported in the book, not a ranking of quality.

14.13 Analytical methods: theaflavins, methylxanthines, and thearubigins

Theaflavins

HPLC is the main technique for theaflavins. Early methods used C14 or ODS packing with 29% aqueous acetone holding 1% acetic acid, but resolution was poor and all theaflavins overlapped, so only total theaflavins could be estimated. A later method on a Hypersil ODS column with citrate buffer separated the four major theaflavins plus four minor ones, and the book calls it the most comprehensive HPLC method for theaflavins to date. The four major theaflavins absorb at about 280, 373, and 450 nm, and measurement at 450 nm was the most specific with the best repeatability. At 205 nm, limits of detection were about 0.5 µg per ml.

LC-MS for theaflavins in biological fluids gave a limit of 6 ng per ml in selected ion monitoring mode, watching m/z 562.9, but background noise was high in biological samples. In multiple-reaction monitoring mode, watching m/z 565 to m/z 139, the limit fell to 0.05 ng per ml, and theaflavin was found in urine and plasma. MALDI-MS found adduct ions and characteristic losses: loss of a gallate ester at 169 Da, retro Diels-Alder loss of 138 Da, and cleavage of the ester in theadibenzotropolone A giving an ion at m/z 548.

HPLC with electrochemical detection separated the four major theaflavins in biological tissues and fluids, with limits of 5 to 10 ng per ml at 70 mV, and found theaflavins in saliva. CE for theaflavins was harder: MEKC gave no peaks, probably because theaflavins were adsorbed or broken down on the capillary wall. A CZE method separated the four major theaflavins within six minutes, but reproducibility was poor, with relative standard deviation below 32.2%. A non-aqueous CE method separated the four major theaflavins in 10 minutes with limits of 21 to 25 µg per ml and intra-day and inter-day variation below 1.6% and 3.5%.

The Flavognost method is a simple spectrophotometric test for total theaflavins: a tea infusion is mixed with an equal part of isobutylmethylketone, the theaflavins move into that layer, react with Flavognost reagent, and absorbance is read at 625 nm. Theaflavin levels by this method link strongly with liquor brightness of black tea extracts, but the method cannot measure individual theaflavins and has largely been replaced by HPLC and CE.

Methylxanthines

Caffeine at 1.5 to 4%, theophylline at 0.2 to 0.4%, and theobromine at about 0.02% are the three major methylxanthines in tea, and their amounts change little with fermentation. Note: the two books disagree here: Ahuja Table 18.1 gives theobromine 0.2% and theophylline 0.5%, while Ho Ch. 6 gives theobromine about 0.02% and theophylline 0.2-0.4%. Both figures are reported as printed; they have not been reconciled. HPLC with UV detection on a C18 column is the common route, with absorption maxima at 210 nm and about 274 nm, the second being more specific. One method separated five catechins and three methylxanthines in 18 minutes with a water, formic acid, and acetonitrile gradient, while an isocratic method for the same compounds took 60 minutes. With a PVPP precolumn for clean-up, caffeine and theobromine were measured in 10 minutes. Ion exchange methods also separated the three alkaloids in 10 minutes with no sample pretreatment, and limits of detection were in the lower µg per ml range.

LC-MS gave limits of 0.02 to 0.03 ng per injection for the three methylxanthines, watching m/z 195 for caffeine and m/z 181 for theophylline and theobromine in selected ion monitoring mode. A simplified LC-MS/MS method without sample clean-up found three methylxanthines and eight metabolites in urine, with limits of 0.05 to 1 µM. CE methods are faster: one MEKC method separated five methylxanthines with a limit below 1.5 µg per ml and no sample preparation beyond filtration, and a CZE method separated five methylxanthines within two minutes, though caffeine stays uncharged and moves with the electro-osmotic flow, so other neutral compounds can interfere.

Thearubigins

Thearubigins are polymeric catechins in fermented black tea and partly fermented oolong tea. They make up a major share of polyphenols in black tea, at 47% in the book, yet their structures stay poorly described. HPLC of thearubigins often shows an unresolved convex Gaussian hump, because the mix cannot be separated into single peaks.

One way to sort them is by solubility: an aqueous extract is taken in turn with chloroform, ethyl acetate, and n-butanol, and the ethyl acetate and butanol extracts are purified by precipitation to give five fractions, TB1 to TB5. Another way is by HPLC elution: group I pigments are excluded from the column, group II are resolved, and group III stay unresolved. Most work estimates total thearubigins by spectrophotometry at 380 nm, using the method of Roberts and Smith, but flavonol glycosides are also extracted and measured, so thearubigin content is overestimated. Clean-up with a C18 cartridge, a cellulose column, or solid-liquid extraction can cut the interference, but thearubigins stick strongly to solid supports, so a support-free method such as countercurrent chromatography has an advantage: after clean-up on an Amberlite XAD-7 column, group I thearubigins were separated from other compounds in two hours.

Thearubigins can also be estimated indirectly by the Folin-Ciocalteu assay, which measures total polyphenols, minus the monomeric polyphenols measured by HPLC, but the response factor of thearubigins to the Folin-Ciocalteu reagent is not known, so errors follow. MALDI-MS of a black tea extract found many peaks above m/z 1,000, including a peak at m/z 1,173 that fits a condensation of a gallate ester of a catechin dimer at 914 Da with the B-ring of EC or C at 290 Da, minus 32 Da, and a peak at m/z 1,431 that fits a tetramer made the same way. Given how hard thearubigins are to separate, the book concludes that direct MS analysis may be an important tool for their analysis and structural assignment.

Quick check

Why does HPLC show thearubigins as a broad hump instead of sharp peaks?

Check your answer

Because thearubigins are a complex mix of polymeric catechins that cannot be separated into single compounds, so they elute as an unresolved hump on the chromatogram.

14.14 Practical note: reading a tea analysis with care

When you read a catechin or theaflavin number, ask four questions from this unit. First, how was the sample extracted, and at what temperature and time, since heat above 80°C and long extraction change the catechin mix. Second, was caffeine removed, since it co-elutes with EGC in UV detection. Third, which detector was used, since UV cannot tell a co-eluting interference apart, while MS and ECD can, at a cost in price or in voltage settings. Fourth, what was actually measured: a single compound, a total by spectrophotometry, or an unresolved hump called thearubigins. A number without those four answers is a claim, not a measurement.

For the maker, the unit points to the same four levers: keep the leaf rich in catechins, because the factory cannot add what the leaf lacks; control rolling and fermentation so PPO meets catechins under the conditions it needs; do not overheat the drying step, since heat changes both aroma and theasinensin formation; and pick the analytical method that fits the question, not the method that is nearest at hand.

Key ideas to carry forward

  • Theaflavins form by coupled oxidation: EC quinone oxidises EGC, EC is regenerated, and the two quinones join to make a benzotropolone ring.
  • Theasinensins form by B-ring to B-ring coupling of two pyrogallol-type catechins, then dismutation on heating and drying; their total amount rivals theaflavins in black tea.
  • Black tea polyphenols stay complex because quinone coupling gives many products, some unstable, and galloyl coupling extends dimers toward oligomers, while the polymeric thearubigins are still not fully described.
  • Theanine is made in the root from glutamate and ethylamine by theanine synthetase with ATP, moves to young leaves, gives umami taste, and is stable in acid but not in alkali.
  • Aroma comes from three routes: enzyme release from glycosides, carotenoid oxidation, and heat in drying; Group II terpenoids over Group I non-terpenoids gives the flavour index.
  • PPO is a copper enzyme, optimum 30°C, active pH 4.03 to 7.00, best substrate catechin at Km 0.49 mM, and hard withering can cut its activity by 50 to 55%.
  • No single analytical method does everything: HPLC-UV is the workhorse, HPLC-MS and ECD add sensitivity and selectivity, CE adds speed, NIRS adds speed for screening, and spectrophotometry gives totals only.

Glossary

Astringency

Check your answer

A dry, puckering feeling in the mouth, given by polyphenols such as catechins and theaflavins. Food tech example: a black tea rich in thearubigins feels strong and astringent in the cup.

Benzotropolone

Check your answer

A ring structure formed when two catechin quinones join during fermentation, found in theaflavins. Food tech example: the benzotropolone ring gives black tea its reddish orange colour.

Catechin

Check your answer

A flavanol in tea leaves, such as EGCG, EGC, ECG, or EC, that gives astringency and is oxidised during black tea making. Food tech example: fresh shoots hold 25 to 30% catechins by dry weight.

Catechol-type catechin

Check your answer

A catechin whose B-ring carries two hydroxyl groups, such as EC and ECG. Food tech example: EC is oxidised first in the coupled oxidation that makes theaflavin.

Coupled oxidation

Check your answer

A reaction in which one quinone oxidises a second compound and is itself reduced back to its starting form. Food tech example: EC quinone oxidises EGC and is regenerated, so EGC falls faster than EC in a mixture.

Dismutation

Check your answer

A reaction in which one compound splits into a reduced product and an oxidised product. Food tech example: dehydrotheasinensins turn into theasinensins plus oolongtheanins by dismutation on heating.

EC, ECG, EGC, EGCG

Check your answer

Short names for the four main tea catechins: (-)-epicatechin, (-)-epicatechin gallate, (-)-epigallocatechin, and (-)-epigallocatechin gallate. Food tech example: EGCG is the main catechin in fresh shoots, at 8 to 12% of dry weight.

Epimerisation

Check your answer

A change in the shape of a molecule at one carbon atom, which can happen when catechins are heated. Food tech example: extraction above 80°C can turn EGCG into GCG, changing the measured catechin mix.

Flavour index

Check your answer

The ratio of Group II terpenoid volatiles to Group I non-terpenoid volatiles in tea. Food tech example: a higher flavour index means a more flowery, better-flavoured black tea.

Galloyl group

Check your answer

A gallic acid part attached by an ester link to a catechin, as in EGCG and ECG. Food tech example: galloyl groups can cover the benzotropolone ring and block the greenish black colour change in theaflavin oxidation.

Glycoside

Check your answer

A compound in which a flavour volatile is bound to a sugar, so it has no smell until the sugar is cut off by an enzyme. Food tech example: linalool is stored as a glycoside in tea leaves and freed when rolling breaks the cells.

HPLC

Check your answer

High-performance liquid chromatography, a method that separates compounds in a liquid by passing them through a column. Food tech example: C18 reversed-phase HPLC is the method of choice for tea catechins.

Mass spectrometry (MS)

Check your answer

A detector that measures the mass of ions, giving molecular weight and fragments that identify a compound. Food tech example: an MS fragment at m/z 139 comes from the A-ring of catechins.

Methylxanthine

Check your answer

A purine alkaloid such as caffeine, theobromine, or theophylline. Food tech example: caffeine is the main methylxanthine in tea, at 1.5 to 4% of dry weight.

Polyphenol oxidase (PPO)

Check your answer

A copper enzyme in tea that oxidises catechins to quinones and starts black tea fermentation. Food tech example: PPO works best at 30°C and is active from pH 4.03 to 7.00.

Pyrogallol-type catechin

Check your answer

A catechin whose B-ring carries three hydroxyl groups, such as EGC and EGCG. Food tech example: pyrogallol-type catechins make up more than 70% of total tea catechins and oxidise easily.

Quinone

Check your answer

An oxidised form of a catechin, made by an enzyme, that can join with other quinones to build larger pigments. Food tech example: EC quinone and EGC quinone join to form theaflavin.

Theaflavin

Check your answer

A reddish orange pigment of black tea, made by joining a catechol-type and a pyrogallol-type catechin, with a benzotropolone ring. Food tech example: the four major theaflavins are theaflavin, theaflavin-3-gallate, theaflavin-3′-gallate, and theaflavin-3,3′-digallate.

Theanine

Check your answer

A non-protein amino acid unique to tea, made in the root from glutamate and ethylamine, that gives umami taste. Food tech example: theanine makes up more than 50% of free amino acids in tea and about 1.5 to 3% of dry weight.

Thearubigin

Check your answer

A polymeric catechin oxidation product in black tea, poorly described, that gives red-brown colour and body. Food tech example: thearubigins make up 12 to 18% of black tea and show as a broad hump on HPLC.

Theasinensin

Check your answer

A dimer made by joining two pyrogallol-type catechins B-ring to B-ring, then changing by dismutation on heating. Food tech example: theasinensins are major black tea polyphenols, comparable in amount to theaflavins.

Umami

Check your answer

A brothy, savoury taste, given in tea mainly by theanine and glutamic acid. Food tech example: theanine mellows the bitter taste of polyphenols and caffeine in green tea.

Volatile flavour compound

Check your answer

A compound that evaporates easily and gives tea its smell, present at only 0.01 to 0.03% of dry weight. Food tech example: more than 600 volatile compounds have been found in tea, including linalool and geraniol.

References

  • Ahuja, P. S., Gulati, A., et al. Science of Tea Technology. Chapter 18, Tea biochemistry.
  • Ho, C.-T., Lin, J.-K., Shahidi, F., et al. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 5, Production of theaflavins, theasinensins, and related polyphenols during tea fermentation.
  • Ho, C.-T., Lin, J.-K., Shahidi, F., et al. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 6, Analytical methods for bioactive compounds in teas.
  • Ho, C.-T., Lin, J.-K., Shahidi, F., et al. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 16, Chemistry and biological properties of theanine.

Unit 15: Special Teas: Pu-erh, White and GABA

Note

Unit 15: Special Teas: Pu-erh, White and GABA

Three teas sit apart from the everyday cup. One is pressed into cakes and left to ripen for years. One is barely touched at all, withered in the sun and dried. One is made by starving fresh leaves of oxygen until a single compound, GABA, builds up inside them. This unit follows each one from the leaf to the cup, and then looks at what the book's studies actually found about them.

Questions to think about as you read:

  • Why would a tea maker press leaves into a cake instead of leaving them loose?
  • White tea gets almost no processing, yet its chemistry changes a lot during withering. What is doing the changing?
  • GABA tea is made by treating leaves with nitrogen gas. What is the leaf doing while it has no oxygen?

Units 2 and 3 already cover the basics: the main catechins, theaflavins and thearubigins, caffeine and tannin, and the standard steps of black, green and oolong making. This unit does not repeat that ground. It goes deep on three teas those units only touch in passing.

15.1 Pu-erh tea: a tea that ripens like a stored crop

Pu-erh is a traditional tea made from large leaves of Camellia sinensis var. assamica. The book places its origin in the districts of Xi-shuang-ban-na, Si-mao and the Lan-chuang-jiang valley, and notes that its pressed products are made mainly in Yunnan. Some pu-erh is also made from small-leaf tea plants. In the market it is sold as loose pu-erh tea, pressed pu-erh tea, and pu-erh tea bags. Every pressed form has to be broken into pieces before brewing with boiling water.

A common belief the book records is that the older the pu-erh, the better it is. The book adds a warning with it: the hardest job for a buyer is telling the exact age of a pu-erh product, and quality control differs a lot between makers.

Pu-erh is sorted in two ways. By shape there is loose tea, the dried leaf in its original curled strip shape, and pressed tea, the loose leaf pressed into a bowl shape, a brick shape or a ring cake shape. By degree of fermentation there are three kinds: green tea-like pu-erh, made without fermentation or storage under high humidity; fermented pu-erh, made by microbial fermentation of piled fresh loose tea or green pressed tea and then stored at room temperature for a long period; and postprocessed pu-erh. Two manufacturing routes are used: the classical pressing method, which dates back centuries to the Tang Dynasty, and the wet-piling method, developed recently, from 1938 to the present.

The classical pressing method: seven steps

1. Plucking. The best fresh leaves are plucked in early spring, in March. Leaves plucked in April and May are still good for pu-erh.

2. Blanching. The fresh leaves are baked on a hot pan with constant manual mixing, as soon as possible after plucking. As soon as the leaves turn very soft and give off a unique grass fragrance, blanching stops at once. At this stage the enzyme reactions in the leaves are thought to be stopped.

3. Rolling. Rolling is done by hand, to let the juice of the leaves out onto the leaf surface, so the flavour comes out easily when brewed with hot water. A skilled tea maker can make about 5 kg of this tea in a day by hand rolling.

4. Sun drying. The rolled leaves are dried under sunlight with constant mixing for two working days. Heat and infrared rays from sunlight pass into the leaves, remove moisture and drive chemical reactions in the leaf. The moisture of the finished product should reach 12 to 15%. The product at this point is called dried raw tea.

5. Storage in a dry place. The dried raw tea is packed in a large, clean gunny bag and stored in a stockroom where temperature and humidity are controlled.

6. Pressing into shapes. The dried raw tea is picked over and graded by quality and evenness, and stems and foreign matter are removed. A cake of pu-erh tea generally weighs 357 to 400 g. Then comes steaming, the step the book calls critical for good quality. The dried raw tea is packed in a cloth bag and steamed, often with a bronze cylinder, for 40 to 50 seconds. If steaming runs too long, the tea becomes too ripe and turns yellow. The water used for steaming matters a lot, and mountain-stream water is generally used. The steamed, soft tea in its cloth bag is pressed between two stone templates, with the tea maker standing on the top template and twisting to spread body weight over the bag.

7. Long-term incubation for ripening. This step is essential for good quality pu-erh. The general rule the book gives is that the longer the storage, the better the quality: in the same batch, a 50-year-old pu-erh should be better than a 5-year-old one for drinking and for health care function, and the prices differ accordingly. In ancient times, the book notes, this ripening happened on horseback during transport.

Classical pressing: seven steps from leaf to cake

1. Pluck in March
April and May leaves still good

↓

2. Blanch on a hot pan
Stop when leaves turn soft and smell of grass

↓

3. Roll by hand
About 5 kg a day per skilled maker

↓

4. Sun dry for two days
Moisture falls to 12 to 15%

↓

5. Store dry in gunny bags

↓

6. Steam 40 to 50 seconds, press
Cake of 357 to 400 g

↓

7. Ripen in storage
50 years beats 5 years in the same batch

Figure 15.1: Classical pressing method for pu-erh tea. Seven steps from plucking to long ripening. Drawn from Ho et al., Tea and Tea Products, Ch. 2.

The wet-piling method: fermentation in piles

The wet-piling method starts the same way as the classical method for its first four steps: plucking, blanching, rolling and sun drying. Then it takes a different road. Both enzymatic and nonenzymatic oxidation shape the tea, and the reactions run very slowly and keep going, helped by microbes from inside and outside the leaves and by oxygen. The book is plain that the sources of the microbes are poorly understood, and the origin and share of internal and external microbes in the fermentation still need investigation.

Step 5: Fermentation by wet piling. Sun-dried tea holds 9 to 12% water. Water is showered on until the moisture reaches 20% or higher, judged by the maker's experience, and the mass is mixed well during showering. The water must be clean and free from harmful microbes, and high mountain stream water is the usual choice. The wet mass is piled 1.0 to 1.5 m high, about 10 tons to a pile, and covered with wet gunnysack to hold moisture and temperature. The pile temperature may reach 50 to 60 °C and must not pass 65 °C. The piles sit in a ventilated clean room.

Step 6: Repiling and mixing. Temperature and moisture decide how far fermentation goes and set the quality, so both are strictly controlled. The piles are broken down, mixed and piled up again every 2 days, with more water showered on during repiling. It takes five to eight rounds to finish fermentation. The end point is judged by an expert from the colour of the fermented mass, which should be reddish brown, and the taste of the extract: it should be friendly to the mouth, free from astringent taste, deep red-brown in colour, and holding a persistent old-tea flavour.

Step 7: Air drying. The fermented tea is broken down into heaps for air drying, not sun drying. The heaps are broken down and spread again every 3 to 5 days, repeating until the moisture reaches 14%.

Step 8: Selection and grouping. The air-dried tea is strictly selected and grouped by colour, size, fragrance and appearance. Step 9: Packaging and pressing. The selected tea is packed in gunnysacks and stored in a dry stockroom, or pressed into shapes such as cubes, bricks, bowls and ring cakes, then stored.

Wet piling: fermentation in big warm piles

Steps 1-4 as in the classical method: pluck, blanch, roll, sun dry

↓

Sun-dried leaf, 9 to 12% water

↓

Shower with clean water to 20% moisture or more

↓

Pile 1.0 to 1.5 m high, about 10 tons, cover with wet sacking
50 to 60 °C, never over 65 °C

↓

Break down, mix and repile every 2 days
Five to eight rounds, until reddish brown and free from astringency

↓

Air dry to 14% moisture, select, pack or press

Figure 15.2: Wet-piling method for pu-erh tea. Water, warmth and microbes do the fermenting. Drawn from Ho et al., Tea and Tea Products, Ch. 2.

Quick check

Why does the wet-piling method need repiling every 2 days?

Check your answer

Temperature and moisture decide how far fermentation goes and set the quality. Breaking the piles down, mixing and piling them up again every 2 days, with more water showered on, keeps temperature and moisture even through the whole pile.

15.2 What the studies found about pu-erh

The book opens this part with a clear warning: studies on the health benefits of pu-erh tea are quite scarce, and the experimental and clinical evidence is incomplete. Folk medicine ascribes nine benefits to pu-erh: it promotes imagination, eye health, mental activity and alertness, digestion and nutritional conditions, an antiobesity effect, detoxifying of certain toxicants, an antidiarrheal action, bone formation, and longevity and well-living. The book states plainly that the evidence for these folk claims is limited and incomplete, and that proper experimental and clinical data are still needed. Do not read the folk list as proven fact: it is a list of traditional claims, not study results.

Antioxidant work in the laboratory. In tests, pu-erh water extract at 100 µg/ml protected plasmid DNA from strand breakage caused by the Fenton reaction, as well as the control. Another study found pu-erh extract with far fewer catechins (8.01 mg/ml) than green tea extract (79.1 mg/ml) could still chelate metal ions, scavenge DPPH radicals, and cut nitric oxide production in activated RAW 264.7 macrophages. Both are laboratory results, not human results.

Fatty acid synthase. Fatty acid synthase (FAS) is a key enzyme of lipogenesis, the making of fat in the body. In rats fed pu-erh tea leaves, FAS expression in the liver was significantly suppressed, and gains in body weight, TAG and total cholesterol were also suppressed. In human hepatoma HepG2 cells in culture, pu-erh extracts suppressed FAS at both the protein and mRNA levels. The book's reading is that this may work through downregulation of the PI3K/Akt and JNK signalling pathways.

Xanthine oxidase and uric acid. In laboratory tests, six tea polyphenols were tested for inhibition of xanthine oxidase, the enzyme that makes uric acid, and theaflavin-3,3'-digallate (TF3) was the most potent inhibitor of the six. The book links this to pu-erh and gout: by slowing uric acid formation, pu-erh polyphenols may help where uric acid runs high. This is laboratory work, not a human trial result.

Quick check

A friend says folk medicine proves pu-erh promotes longevity. What does the book actually say?

Check your answer

The book lists longevity among nine folk medicine claims for pu-erh, then states plainly that the evidence for these benefits is limited and incomplete, and that adequate experimental and clinical data are still needed.

Try it yourself: Pick up a pu-erh cake or a loose pu-erh in a shop and read the label. Note the shape (loose, cake, brick, bowl, ring), any age stated, and whether it says fermented or raw. Compare what the label claims with the two methods in section 15.1 and ask which method the tea most likely came from.

15.3 White tea: the tea that is barely made

White tea was first discovered during the Tang Dynasty (618-907 AD), and soon became the choice of royal courts. It changed a lot after 1885, when specific tea bush varieties were selected to make Silver Needle and other specialty white teas. Export of these fine white teas began in 1891.

White tea is named for its look: the buds are covered with fine white hair. Its special quality comes from a special process the book sums up as "no frying and no crushing". Annual production is relatively small, just 2,000 tons, about 0.1% of that of black tea. More than 90% of world white tea is made in Fuding, Zhenghe, Jianyang and Songxi. White tea is harvested when the buds are still covered with fine white hair and the leaves have not yet fully opened. After plucking, the fresh leaves are sunshine withered and dried, and they stay fresher than green or black tea leaves. The taste is lighter and sweeter, closer to fresh leaves than green tea is.

There are four grades, set by plucking: Silver Needle with White Hair, White Peony, Gongmei and Shoumei. Silver Needle is made of the pure bud. White Peony is made of the bud and one or two leaves of Dabaicha or Shuixian varieties. Gongmei is made from the Qingzhong variety, and Shoumei from fresh leaves without the bud. White Peony is the most dominant grade.

Silver Needle: sunshine and a slow bake

Silver Needle is named for its colour and shape: white like silver, long like small needles. The material is new bulky buds. Two methods dominate, from Fuding and Zhenghe. In Fuding, the more common method, the buds are spread on sieves or withering mats right after plucking and put directly under the sun. One whole day under strong spring sunshine nearly dries them, and then they are baked on a slow fire until fully dry: about 30 minutes for 250 g of tea buds at 40 to 50 °C in a baking basket. On hot and wet days, two days of sunshine are needed; on rainy or foggy days, direct baking on slow fire is necessary. In Zhenghe, the buds are aired under shade or mild sunshine before direct sunshine drying, and the whole process often takes 2 to 3 days.

White Peony is made from the bud and one or two leaves, and Gongmei and Shoumei are processed in a similar way. White Peony is processed in two simple steps, withering and basket drying, and withering is the key step that decides final quality. Three withering methods are used. Airing is the most used: in spring the room is held at 18 to 25 °C with relative humidity (RH) of 67 to 80%, and in summer and autumn 30 to 32 °C with RH of 60 to 75% is more acceptable, for 52 to 60 hours. Low-temperature withering is used on rainy days: the room is heated by pipeline, kept at 20 to 32 °C and RH of 65 to 70%, with 29 to 30 °C most suitable, for more than 36 hours. Sunshine withering is used on sunny spring days: fresh leaves are put under slight sunshine for 25 to 30 minutes in the morning or at dusk, at 25 °C and RH of 63%, then moved into a prewarmed withering room, for two to four rounds. After withering comes basket drying: 15 to 20 minutes at a fire temperature of 70 to 80 °C, or twice, first at 100 °C and second at 80 °C, if the leaves are only half dried.

Quick check

Why is withering, not drying, called the key step for White Peony?

Check your answer

Withering decides the final quality. It runs 52 to 60 hours by airing in spring, and during withering the water, enzymes, polyphenols and other constituents change, setting the colour, taste and aroma that drying then fixes.

15.4 White tea chemistry: big changes from gentle handling

White tea looks untouched, but its chemistry moves a lot during withering. Tea polyphenols, amino acids and water are the key components, and withering drives much of the change in all of them.

Water. Water content matters because it affects enzyme activity and chemical reactions. During withering, water evaporates at a rate set by humidity and temperature.

Table 15.1: Water in white tea during withering

Withering time (hours)010203040
Water content in tea leaves (%)6962423216
Loss of water (%)010.139.153.676.8

Half the water is gone after about 30 hours of withering, and only about 10% is lost after 10 hours.

Enzymes. Polyphenol oxidase (PPO) and peroxidase (POD) are the two important enzymes here. In green tea making they are unwanted, because they oxidise catechins and harm green tea quality. In black tea making they are wanted, because they turn catechins into theaflavins and thearubigins. White tea is withered and then dried, with no rolling or fermentation step; during its long withering, PPO and POD slowly oxidise the polyphenols. White tea gets no shaping or fermentation treatment. During withering, PPO activity rises faster than POD at first. After 4 hours, PPO is inhibited by quinones formed by catechin oxidation, while POD is not affected. After 8 hours PPO rises again and stops after 20 hours, while POD reaches its maximum at 12 hours. During drying, PPO and POD are denatured by heating at 80 °C or higher. The best condition for white tea is slight and slow oxidation of polyphenols below 30 °C.

Amino acids, carbohydrates and chlorophyll. Amino acids give freshness and briskness to the liquor. Their content rises gradually during withering, especially early on, and after 24 hours it rose by about 71% in Fuyunzhong No. 6 (2.41 to 4.13) and by about 33% in Fuding Dabaicha (3.12 to 4.14) compared with fresh leaves. Carbohydrates fall early in withering, rise later as polysaccharides are hydrolysed by enzymes, and fall again during heating at drying. Chlorophyll, the key factor for white tea colour, falls steadily during withering and drying. The white hair itself holds more amino acids than the tea: 3.3% against 2.7% in one sample, and 3.2% against 2.5% in a second, while polyphenols were lower in the hair (25.0% and 23.9%) than in the tea itself (32.1% and 29.6%).

🤔 Pause and think: white tea is made with almost no heat and no rolling, yet its catechins fall by more than half during withering. What is doing that work, if no machine is crushing the leaf?

15.5 What the studies found about white tea

Without high temperature and fierce kneading or rolling, white tea keeps more beneficial components than other teas, the book says, and new studies suggest it offers the benefits of green tea and may be the most potent among teas for human health promotion. That suggestion is the start of the story, not the end of it.

What is inside it. White and green teas are considered especially beneficial because they hold higher amounts of EGCG, a biologically active polyphenol. White tea also holds the highest amounts of flavones, 14.2 to 21.4 fold higher than other teas, and flavones have the strongest antioxidative activities. White tea also carries less caffeine: only 15 mg per serving, against 40 mg in black tea and 20 mg in green tea. Among processed teas tested by electron spin resonance, white tea held the lowest amount of stable free radicals, with green tea second.

In Apc(min) mice, after 12 weeks, mice given white tea, green tea or sulindac, a nonsteroidal anti-inflammatory drug known to be highly effective in these mice, had significantly fewer tumours than the controls. Protection from 1.5% green or white tea was comparable to that from 80 ppm sulindac, and mice given white tea plus sulindac had significantly fewer tumours than either treatment alone. In rats given PhIP, a heterocyclic amine, white tea cut aberrant crypt foci, early lesions in the colon, from 5.65 plus or minus 0.81 per colon to 1.31 plus or minus 0.27 per colon over 8 weeks. Then comes the sentence that must not be skipped: although many studies point to white tea as a potential anticarcinogen, the data are highly preliminary and cannot be extrapolated to human cancer prevention or treatment, and further studies in animal models and in human trials are needed.

Antioxidant and antimicrobial work. White tea extract could limit and prevent DNA damage in skin cells after exposure to sunlight, in the studies the book cites. In a model of oxidative damage of human red blood cells started by cumene hydroperoxide, white tea extract at 4 g per 150 ml water significantly decreased formation of malondialdehyde, an oxidation product. A study found white tea has better antimicrobial function than green tea: white tea extract may slow the growth of bacteria that cause Staphylococcus infections, Streptococcus infections, pneumonia and dental caries, was very effective in inactivating bacterial viruses, and totally inactivated Penicillium chrysogenum spores and Saccharomyces cerevisiae yeast cells. White tea extract has now been added to several toothpastes in China and the United States. All of this is laboratory and early-stage work, and the book does not present white tea extract as a medicine.

Why does the book warn against reading the white tea cancer studies as proof for people?

Check your answer

The studies were done in the Salmonella assay, in cell models, in Apc(min) mice and in rats given PhIP. The book states plainly that these data are highly preliminary and cannot be extrapolated to human cancer prevention or treatment.

Try it yourself: Brew a white tea and a green tea side by side from the same weight of leaf. Look at the dry leaf first: find the white hair on the buds. Then compare the liquor colour and taste. The white tea should taste lighter and sweeter. Ask yourself which processing step, or lack of one, explains the difference.

15.6 GABA tea: a tea built by taking oxygen away

GABA tea, also called Gabaron tea in Japan, is a new type of specialised tea with a high content of gamma-aminobutyric acid, GABA. The book's definition is a number: GABA tea contains more than 150 mg of GABA per 100 g of tea. Besides health benefits related to antioxidation, anticancer and blood lipid-lowering effects, the book says, GABA tea has special biological functions due to GABA itself.

The story starts in 1987. Tsushida and colleagues were looking for a way to extend the storage time of fresh tea leaves for making green tea, and they treated the fresh leaves anaerobically, in the same way as fresh vegetables are treated. In the 1987 storage trial they found GABA content up 8.9 fold and alanine up 5.2 fold, while 79% of glutamic acid and 91% of aspartic acid had disappeared. GABA was already known since the 1950s as an important chemical transmitter for the brain nerves, and researchers had reported that it could reduce blood pressure in experimental animals and in humans. Tsushida's group kept working on the making of GABA tea, and with Omori at Otsuma Women's University they tested it in animals. After confirming that GABA tea could reduce the blood pressure of the spontaneously hypertensive rat (SHR), GABA tea was mass produced and sold in Japan.

What GABA is, and what the book links it to

GABA is found in bacteria, insects, plants and animals, and is widely spread in living cells of the cerebral cortex and hypothalamus. In the body it is made from glutamate by the enzyme glutamate decarboxylase (GAD). It is an important inhibitory neurotransmitter in the brain and central nervous system: it lowers the excitation of nerves, opens more chloride channels in the nerve membrane, and increases chloride penetration. Low GABA in plasma is reasonably specific to Parkinson's disease, epilepsy, alcoholism, depression and menopause, the book reports.

The book reviews what GABA itself, not GABA tea, has been linked to in the literature: blood pressure, depression, sleep, alcoholism, growth hormone release, diabetes, Parkinson's disease and epilepsy. For blood pressure, GABA has been reported to reduce blood pressure in experimental animals and humans. For depression, low GABA concentrations in the brain are linked with depression, palpitation, insomnia, learning disorders and memory failure. For sleep, GABA is described as a therapy medicine for sleeplessness. For alcoholism, plasma GABA was lower in alcoholics than in nonalcoholic drinkers. For diabetes, a fall in GABA in pancreatic islet cells was associated with a fall in insulin synthesis and release. For Parkinson's disease, GABA and GAD concentrations were reduced in patients' brains and plasma. For epilepsy, low brain GABA levels were associated with recent seizures and poor seizure control. Read that list with care: it is a review of GABA research, mostly in animals and in patients with disease, and it is not a set of trials of GABA tea in healthy people.

The content of GABA in plant tissues is usually around 0.3 to 32.5 µmol/g. In tea, Tsushida and Murai showed with 15N-labelled glutamic acid that GABA is formed from glutamate. In the 15N-labelling test after 3 hours of anaerobic incubation of tea leaves, GABA rose about 17.7 fold and alanine about 6.8 fold. About 95% of the 15N in glutamate and 90% of the 15N in aspartate disappeared in the first 3 hours, while 15N in alanine and GABA rose markedly up to 6 hours, showing the 15N in the accumulated GABA came from glutamate.

The mechanism is a blocked exit, not a faster entry. GAD activity did not increase under anaerobic conditions. Instead, the activity of succinic semialdehyde dehydrogenase fell, so succinic semialdehyde could not be further metabolised to succinate, and GABA piled up. In plain terms, glutamate is turned into GABA, and the normal route that would clear GABA away is switched down, so GABA accumulates.

How GABA builds up in tea leaves without oxygen

Glutamate
The starting amino acid in the fresh leaf

↓

Glutamate decarboxylase (GAD)
Best at pH about 5.8. Turns glutamate into GABA, releasing CO₂. Note: GABA builds up most at 40 °C in the book's trials; the 40 °C enzyme optimum comes from radish-leaf work cited in the book.

↓

GABA
Builds up when oxygen is taken away

↓

Normal exit route: blocked
Without oxygen, succinic semialdehyde dehydrogenase activity falls, so GABA is not cleared to succinate. Withering lowers cell pH, which favours GAD.

Figure 15.3: Formation of GABA in tea leaves. Glutamate becomes GABA, and the anaerobic treatment blocks the route that would clear GABA away. Drawn from Ho et al., Tea and Tea Products, Ch. 4.

GABA also accumulates in plants under stress: lack of oxygen, loss of water, mechanical damage, lower pH and changes in calcium levels. During withering, the leaf loses water, hydrolysis rises, theanine and glutamine fall, glutamate rises greatly and GAD activity is induced, so a further anaerobic treatment then accumulates GABA. Withering also lowers the pH inside the cell, which favours GAD, whose best pH is around 5.8, and not GABA-transaminase, whose best pH is around 8.9.

Making GABA tea: the anaerobic step is the key

Anaerobic treatment is the key step, and it runs 5 to 10 hours. The basic procedures differ by tea type. For GABA green tea, fresh leaves go straight to anaerobic treatment, then panning or steaming, rolling and drying. For GABA oolong tea, the leaves get solar withering and indoor withering with shaking, then anaerobic treatment, then panning, rolling and drying. For GABA black tea, the leaves are withered, then given anaerobic treatment, then rolled, fermented and dried. For fermented types such as oolong and black teas, anaerobic fermentation must come after solar and indoor withering, or GABA is easily lost. To prevent loss of the GABA made by anaerobic fermentation, pan blanching, rolling and drying must follow right after it, to fix the GABA content in the finished tea.

The order matters, and the book gives numbers for it. In oolong-type GABA tea, leaves incubated in nitrogen after withering held 245.9 mg of GABA per 100 g, against 135.9 mg per 100 g when nitrogen came before withering. In black-type GABA tea the figures were 176.5 against 81.2 mg per 100 g.

Table 15.2: Amino acids in oolong and black GABA teas made with nitrogen before or after withering (mg/100 g)

GABA teaAspartic acidGlutamic acidAlanineTheanineGABA
Oolong 1, nitrogen before withering85.747.163.3491.1135.9
Oolong 2, nitrogen after withering39.146.258.5451.1245.9
Black 1, nitrogen before withering95.4176.131.4411.281.2
Black 2, nitrogen after withering29.522.488.7377.0176.5

Which gas? Carbon dioxide gave higher GABA than nitrogen: after 10 hours, 290.9 mg% with CO₂ against 233.9 mg% with N₂, while air gave only 28.5 mg% and oxygen 12.7 mg%. Even so, nitrogen is recommended, because tea treated with N₂ still holds a high amount of GABA, and N₂ is cheaper and safer than CO₂. Caffeine and tannin barely changed under any gas.

Which leaves? Glutamate is the precursor of GABA, so leaves high in glutamate give high GABA. Any variety can be used, but the highest GABA content found, 663 mg%, came from the Assam variety. Less mature leaves give higher GABA, and amino acids fall in the order stem, bud, second leaf, third leaf and old leaves, so GABA tea made with stem holds the most. Spring tea gives the highest GABA, followed by summer and fall. Leaves from above 1,000 m gave on average 100 mg% more GABA than leaves from 200 to 500 m. Shading tea bushes for 14 days raised GABA by 20 to 80 mg%. Four oolong varieties, TTES 12, TTES 13, CS Oolong and SJChue, all gave more than 150 mg% GABA with a repeated anaerobic-aerobic process, with TTES 12 highest.

New techniques for raising GABA. Cycling anaerobic and aerobic treatment raises GABA further: under aerobic treatment GABA falls slightly but glutamate rises significantly, so the next anaerobic round turns that glutamate into GABA. Three cycles of 3 hours anaerobic with 1 hour aerobic after each, then one more anaerobic round, raised GABA 1.5 fold in spring tea and 2.33 fold in summer and fall tea, compared with a single anaerobic treatment. But three cycles can turn the leaf and infusion red and create an unpleasant odour, so the book suggests limiting treatment to three anaerobic rounds with two aerobic rounds. In one trial, GABA finished at 154.99 mg/100 g after three anaerobic and two aerobic cycles, above the commercial standard of 150 mg/100 g. Microwave irradiation raised GABA to 167 to 202 mg/100 g, with the best power 0.3 kW and the best time 20 minutes. Infrared irradiation raised GABA in a first-picked crop from 1.73 to 2.70 mg/g, with the best time 30 to 50 minutes. Soaking leaves in sodium glutamate solution with infrared irradiation raised GABA to 2.89 mg/g for spring tea and 2.81 mg/g for summer tea.

Why is nitrogen, not carbon dioxide, recommended for GABA tea, even though carbon dioxide gave more GABA?

Check your answer

After 10 hours, CO₂ gave 290.9 mg% GABA against 233.9 mg% with N₂, but N₂ is recommended because tea treated with N₂ still holds a high amount of GABA and N₂ is cheaper and safer than CO₂.

15.7 What the studies found about GABA tea

The book says the most special characteristic of GABA tea is not the original functions of tea, but blood pressure moderation, helped by both GABA and catechins together. Catechins alone lower blood pressure by inhibiting angiotensin I converting enzyme (ACE), and GABA tea adds GABA to that. Most research on GABA tea has concentrated on blood pressure. The studies below are animal studies and small human studies, and they should be read as early evidence, not as proof that GABA tea treats high blood pressure.

Rats first. Omori and colleagues used 24 spontaneously hypertensive rats (SHR), split into three groups fed GABA tea, green tea or deionised water. After 2 weeks, blood pressure in the GABA group was significantly reduced and their urine held a lot of sodium, which led Omori to suggest drinking GABA tea helped the body excrete sodium and so reduced blood pressure. Blood pressure in the GABA-fed rats was 25 to 30 mmHg lower than in the other groups, with no significant difference in body weight. After 9 weeks, when GABA tea was replaced with deionised water, blood pressure rose again to a hypertensive level, while rats switched the other way, from water to GABA tea, had significantly reduced blood pressure. The authors' reading was that daily drinking must continue to keep blood pressure down, in the same way as taking hypertension medicine.

In Dahl salt-sensitive rats fed a 4% salt diet for 3 weeks, then given water, ordinary tea or GABA-rich tea for 4 weeks, old rats on GABA tea averaged 176 plus or minus 4 mmHg, against 207 plus or minus 9 for water and 193 plus or minus 5 for ordinary tea, with an average GABA intake of 4.0 mg per rat per day.

Small human studies. Omori's group gave 13 hypertensive patients diagnosed in hospital one 3-g bag of GABA tea per day for 3 months, after stopping hypertension medicine. Their blood pressures fell from 180 and 184 mmHg to 165 and 140 mmHg. Note: the book reports two readings, 180 and 184 mmHg falling to 165 and 140 mmHg, without labelling them; do not present them as a systolic/diastolic pair. A further test with 15 hypertension patients not diagnosed in hospital, also drinking GABA tea for 3 months, gave significant blood pressure reduction in 8 of the 15, and all felt relaxed and well. These are small studies with no control group described in the book, and stopping prescribed medicine to drink tea is not something this unit recommends: the book reports what the authors did, and a reader on blood pressure medicine should talk to their own doctor, not copy a study. A clinical study with 50 patients split into two groups, one given two 5-g bags of GABA tea a day and the other two 5-g bags of green tea a day, found after 20 days that the GABA tea group showed a significantly greater blood pressure reduction than the control group, with no side effects reported.

How it may work. In laboratory tests, GABA, alanine, theanine and gamma-hydroxybutyric acid all inhibited ACE activity, and GABA was the best of them.

Table 15.3: Inhibition of ACE activity by the main components of Gabaron tea in vitro, each at 40 mM

ComponentACE activity (%)Inhibition (%)
GABA10.5389.47
Alanine23.0876.92
Theanine34.0064.00
Gamma-hydroxybutyric acid42.5157.49

In rats, the rise of ACE activity caused by extra saline was significantly inhibited by 3% added GABA, which suggests ACE inhibition is one of the major mechanisms of Gabaron tea.

A long rat study. In 2005, Ou and colleagues compared Taiwan GABA tea and green tea in SHR over 12 weeks. Blood pressure in rats fed high-dose (300 mg/100 g) or low-dose (150 mg/100 g) GABA tea was reduced significantly, more than with pure GABA or green tea. Systolic pressure in the high-dose GABA tea group fell from 236.71 to 207.14 mmHg, while the water control group rose from 238.00 to 252.40 mmHg. Serum triacylglycerol, cholesterol, LDL cholesterol and HDL cholesterol fell in all groups except the water control. When feeding was extended to 82 weeks, only the rats fed GABA tea were all alive, at 100%, against 50% for the control group, 57.14% for green tea, and 66.7% for low-dose pure GABA solution. The book presents this as a strong survival difference in rats, and it is a rat result, not a human result.

Practical note. If you buy GABA tea, the number to look for is the one the book uses as its definition: more than 150 mg of GABA per 100 g of tea. The book's own evidence says the GABA content is set by the leaf, the season, the elevation, the gas, the temperature and the timing, so two GABA teas can differ widely. Nothing in this unit is medical advice, and GABA tea is not presented here as a treatment for high blood pressure or any other condition.

Quick check

In the Omori rat study, what happened when GABA tea was taken away after 9 weeks?

Check your answer

Blood pressure in the rats rose again to a hypertensive level once GABA tea was replaced with deionised water, which is why the authors said daily drinking would have to continue.

Self-check: three teas, three ideas

  1. Pu-erh is pressed into cakes of 357 to 400 g and steamed for 40 to 50 seconds. What goes wrong if steaming runs too long?
Check your answer

The tea becomes too ripe and yellow colouration develops, so steaming time is kept within 40 to 50 seconds.

  1. White tea loses more than half its catechins during withering, yet its amino acids rose by about 71% in Fuyunzhong No. 6 (2.41 to 4.13) and by about 33% in Fuding Dabaicha (3.12 to 4.14) in 24 hours. Which change helps freshness and briskness?
Check your answer

The amino acid rise helps freshness and briskness. Amino acids in Fuyunzhong No. 6 rose by about 71%, from 2.41 to 4.13 in 24 hours, and in Fuding Dabaicha by about 33%, from 3.12 to 4.14.

  1. GABA tea must hold more than 150 mg of GABA per 100 g. Name two things a maker can change to raise GABA content.
Check your answer

Any two of: use leaves high in glutamate such as stem or less mature leaves or the Assam variety; treat anaerobically after withering rather than before; use nitrogen or carbon dioxide instead of air; hold the anaerobic treatment near 40 °C for about 6 hours; use spring leaves or leaves from above 1,000 m; shade the bushes; cycle anaerobic and aerobic treatment.

  1. In the Dahl rat study, old rats on GABA tea averaged 176 mmHg against 207 mmHg on water. What kind of evidence is this, and what can it not prove?
Check your answer

It is an animal study in Dahl salt-sensitive rats on a high-salt diet. It cannot prove that GABA tea lowers blood pressure in humans.

  1. White tea holds 15 mg of caffeine per serving, black tea 40 mg and green tea 20 mg. A reader says this proves white tea is the healthiest tea. What is wrong with that claim?
Check your answer

Caffeine is one measurement among many, and the book's health findings for white tea come from laboratory and animal studies that the book itself calls highly preliminary and not transferable to human prevention or treatment.

Try it yourself: Make a small comparison card for the three teas. For each one, write the key making step in one line (pressing and ripening, sunshine withering, anaerobic treatment), one number from this unit, and one sentence saying what kind of study the health finding came from (folk claim, laboratory test, rat study, small human study). Keep the card where you brew tea, and use it the next time a label makes a big health claim.

Key ideas to carry forward

  • Pu-erh is made from large leaves of Camellia sinensis var. assamica, sold loose or pressed into bowls, bricks and ring cakes, and sorted by shape and by degree of fermentation.
  • The classical pressing method runs plucking, blanching, hand rolling, two days of sun drying to 12 to 15% moisture, dry storage, steaming for 40 to 50 seconds and pressing into 357 to 400 g cakes, then long ripening in storage.
  • The wet-piling method showers sun-dried leaf to 20% moisture or more, piles about 10 tons 1.0 to 1.5 m high at 50 to 60 °C, never over 65 °C, and repiles every 2 days for five to eight rounds until the mass is reddish brown and free from astringency, then air dries to 14%.
  • Pu-erh folk claims are listed in the book as claims with limited and incomplete evidence. Rat studies found lower plasma TAG, cholesterol and LDL cholesterol, suppression of liver fatty acid synthase, and laboratory inhibition of xanthine oxidase, with theaflavin-3,3'-digallate the most potent of six polyphenols tested.
  • White tea is made with no frying and no crushing: Silver Needle is pure bud, sun dried and slow baked, and White Peony is bud and one or two leaves, withered for 52 to 60 hours by airing and basket dried at 70 to 80 °C. Spring tea is premium, autumn next, summer lowest.
  • During white tea withering, water falls from 69% to 16% over 40 hours, catechins fall by more than half, amino acids rose by about 71% in Fuyunzhong No. 6 and by about 33% in Fuding Dabaicha in 24 hours, and PPO and POD drive a slight, slow oxidation below 30 °C.
  • White tea holds 15 mg of caffeine per serving against 40 mg in black tea and 20 mg in green tea, and its flavones run 14.2 to 21.4 fold higher than other teas. Its anticancer findings come from the Salmonella assay, cell models, Apc(min) mice and PhIP rats, and the book calls them highly preliminary.
  • GABA tea is defined by more than 150 mg of GABA per 100 g of tea. GABA forms from glutamate through glutamate decarboxylase, and anaerobic treatment blocks its clearance, so it accumulates.
  • GABA tea is made by anaerobic treatment for 5 to 10 hours, after withering for oolong and black types, then fixed at once by pan blanching, rolling and drying. Nitrogen is recommended over carbon dioxide for cost and safety, 40 °C gives the highest GABA, and 6 hours at 40 °C or 8 hours at room temperature is recommended.
  • In rats, GABA tea lowered blood pressure and serum lipids, and in a long feeding study to 82 weeks the GABA tea groups were all alive at 100%, against 50% for the water control. Small human studies reported blood pressure falls, and the book reports GABA inhibited ACE activity by 89.47% in laboratory tests at 40 mM.

Glossary

Pu-erh tea

A tea made from large leaves of Camellia sinensis var. assamica, sold loose or pressed into cakes, bricks, bowls and ring cakes, and ripened or fermented. Food tech example: a pressed cake of 357 to 400 g steamed for 40 to 50 seconds before pressing.

Wet piling

A pu-erh fermentation method in which wet tea is piled about 10 tons at a time at 50 to 60 °C and repiled every 2 days. Food tech example: a wet pile held under wet sacking until the mass turns reddish brown and loses its astringency.

White tea

A tea made with no frying and no crushing, from buds covered with white hair, withered and dried. Food tech example: Silver Needle made from pure buds sun dried for a day and baked at 40 to 50 °C.

Silver Needle

The top grade of white tea, made from pure buds that look white like silver and long like needles. Food tech example: Fuding Silver Needle baked for about 30 minutes for 250 g of buds at 40 to 50 °C.

White Peony

A white tea grade made from the bud and one or two leaves. Food tech example: White Peony aired for 52 to 60 hours in spring at 18 to 25 °C, then basket dried at 70 to 80 °C.

Withering

The slow loss of water from plucked leaves, during which enzymes and chemistry change the leaf. Food tech example: white tea withering for 52 to 60 hours while water falls from 69% toward 16%.

Polyphenol oxidase (PPO)

An enzyme that oxidises catechins and drives colour formation during withering. Food tech example: PPO rising during white tea withering and being denatured by heating at 80 °C or higher.

Peroxidase (POD)

An enzyme that works with PPO in oxidising tea polyphenols. Food tech example: POD reaching its maximum activity at 12 hours of white tea withering.

GABA

Gamma-aminobutyric acid, an inhibitory neurotransmitter in the brain and central nervous system that lowers nerve excitation. Food tech example: GABA tea defined by more than 150 mg of GABA per 100 g of tea.

Glutamate decarboxylase (GAD)

The enzyme that turns glutamate into GABA. Food tech example: GAD working best near pH 5.8 during anaerobic treatment of tea leaves; GABA builds up most at 40 °C in the book's trials.

Anaerobic treatment

Treating fresh tea leaves without oxygen, usually under nitrogen gas, so GABA accumulates. Food tech example: fresh leaves held under nitrogen for 5 to 10 hours to make GABA tea.

Angiotensin I converting enzyme (ACE)

An enzyme linked with blood pressure regulation, inhibited in laboratory tests by GABA and tea catechins. Food tech example: GABA inhibiting ACE activity by 89.47% in a laboratory test at 40 mM.

References

  • Ho, Chi-Tang, Lin, Jen-Kun, and Shahidi, Fereidoon (eds.). Tea and Tea Products: Chemistry and Health-Promoting Properties. CRC Press, 2008. Chapter 2: Chen, Huai-Yuan, Lin-Shiau, Shoei-Yn, and Lin, Jen-Kun. Pu-erh Tea: Health Benefits. Chapter 3: Jiang, He-Yuan. White Tea: Its Manufacture, Chemistry, and Health Effects. Chapter 4: Ou, Andi Shau-mei, Tsai, Yung-sheng, and Wang, Hsueh-fang. Biological Functions and Manufacturing of GABA Tea.

Unit 16: Tea in the Body: Catechins, Antioxidants and Protection

Note

Unit 16: Tea in the Body: Catechins, Antioxidants and Protection

You drink a cup of green tea. Four catechins go in, but almost none of them arrive in your blood unchanged. Your gut, your liver and the microbes in your colon rebuild them first, and what protects your cells is often the rebuilt form, not the original. This unit follows that journey, then looks at how tea polyphenols act as antioxidants, and at one more quiet job they do in a test tube: trapping methylglyoxal, a reactive compound linked with diabetes.

Questions to think about as you read:

  • If only a tiny share of tea catechins reaches the blood unchanged, why do researchers still study them?
  • A catechin can be changed in three ways after you drink it. What does each change do to it?
  • Tea polyphenols can act as antioxidants and, in some conditions, as prooxidants. What decides which way they act?

Unit 2 introduced the catechins, theaflavins and thearubigins. This unit picks up where it left off: what happens to those compounds inside the body, how they fight oxidation, and what the book's methylglyoxal experiments showed.

16.1 The four catechins

Green tea and black tea are made from the same plant, Camellia sinensis, by different processing. Green tea is made by drying the leaves at high temperature to prevent polyphenol oxidation, so its major components stay close to those in the fresh leaf. The four main catechins are (-)-epicatechin (EC), (-)-epigallocatechin (EGC), (-)-epicatechin gallate (ECG) and (-)-epigallocatechin gallate (EGCG). Together the four make up approximately 25% of green tea on a dry weight basis. EGCG is the most abundant and the best studied, accounting for 65% of the total catechin content in green tea. Black tea is made by withering, rolling and crushing the fresh leaves, which starts fermentation of polyphenols and turns some catechins into theaflavins and thearubigins. Total flavanol level falls from between 35 and 50% in green tea to 10% in black tea, while theaflavin content in black tea is 3 to 6% and thearubigin content is 12 to 18%.

The four catechins have relatively high molecular weights, 290 to 458, and five or more hydroxyl groups (EC has five; EGCG has eight). Because of their large apparent size, they have a relatively low bioavailability. That single fact shapes everything that follows in this unit: the body takes in far less catechin than the cup holds, and most of what it does take in is changed on the way.

The book also notes a gap worth remembering. Little research has been done on the bioavailability and metabolism of theaflavin, the major component of black tea, because it is highly polymerised and less well characterised. Most of this unit is therefore about catechins, and the theaflavin story is thinner, because the book's story is thinner.

Quick check

Why does black tea hold far fewer catechins than green tea from the same leaf?

Check your answer

Black tea making starts fermentation of the polyphenols, turning catechins into theaflavins and thearubigins. Total flavanol level falls from 35 to 50% in green tea to 10% in black tea.

16.2 Absorption: most catechins never arrive unchanged

Bioavailability means the concentration of a compound, or its metabolite, at the target organ. Because organ sites cannot be reached directly in living humans, exact values cannot be calculated. Absolute bioavailability is used instead: the exact amount of a compound that reaches the systemic circulation, calculated as the fraction of the area under the curve (AUC) after oral ingestion compared with the AUC after intravenous administration. Bioavailability covers five linked processes: liberation, absorption, distribution, metabolism and excretion, shortened to LADME. The book is plain that there are no data on the exact amount of catechins absorbed, but the plasma profile and the amount excreted in urine suggest only small proportions are absorbed. Urinary excretion does not give an accurate measure of absorption either, because metabolism and possible elimination in bile change the amounts excreted in urine.

The main absorption site is the small intestine. EC, ECG, EGC and EGCG are handled at the intestinal epithelium by similar efflux and metabolism mechanisms. In the human Caco-2 cell line, a laboratory model of the intestinal wall, the absorption transport of the four catechins at different doses was similar, but green tea polyphenol mixtures could affect the secretion transport of individual catechins. The book suggests transporter competition may be responsible for the reduced efflux of EC, and metabolic competition may reduce the formation of EGC sulfate and methylated EGC sulfate.

How small is small? In humans drinking black tea, approximately 1.68% of the ingested catechins were found in the plasma, urine and faeces, and the apparent bioavailability of the gallated catechins was lower than that of the nongallated forms. Plasma levels of intact tea catechins and other flavonoids did not pass 1 µM after consumption of catechins in amounts similar to those in the diet. The total peak plasma concentrations of EGCG, EGC and EC, free plus conjugated forms, were around 2 to 3 µM or lower. These concentrations are much lower than those used in many laboratory cell line studies, which is a warning to carry through the whole unit: a result seen in a dish at a high dose may not happen in a body at a tea-drinking dose.

The catechin journey: LADME

Cup of tea
Four catechins, about 25% of green tea dry weight. EGCG is 65% of them.

↓

Liberation and absorption: small intestine
Only small proportions are absorbed. Intact catechins in plasma stay below about 1 µM.

↓

Distribution and metabolism: liver and gut wall
Glucuronidation, sulfation and methylation rebuild most of what is absorbed.

↓

Colon microbes
What is not absorbed is broken down by gut microbes into valerolactones and phenolic acids.

↓

Excretion: urine and bile
Total urinary excretion of catechins in humans was 0 to 9.8%. Most leaves as metabolites.

Figure 16.1: The catechin journey through the body. Most catechins are changed or broken down before they can act. Drawn from Ho et al., Tea and Tea Products, Ch. 7.

The book compares the four catechins on an equal footing by converting the published pharmacokinetic data to a single 50 mg dose of each, assuming the values rise in a straight line with dose. Only free and conjugated intact catechins were used, and methylated metabolites were mostly not analysed, so the mean values are probably underestimated.

Table 16.1: Pharmacokinetic data for tea catechins in humans, converted to a single 50 mg dose (mean values from the book)

CatechinPeak plasma concentration, Cmax (µM)Time to peak, Tmax (hours)Area under the curve, AUC (µM·h)Half-life, T1/2 (hours)Absorption in plasma (%)Excretion in urine (%)
EC0.282.151.393.360.835.11
ECG0.086.740.886.90.050.01
EGC0.241.841.031.470.331.25
EGCG0.101.950.503.090.030.01

Table 16.1 note: These excretion figures come from different studies in the book and do not agree: Table 16.1 (50 mg dose, intact catechins) gives EC 5.11% and EGC 1.25% in urine, while §7.2.4 of the book gives EC 18.5% and EGC 11.1%, and the book's own total range is 0-9.8%. Do not quote one set without saying which study it is from.

Read the table as a comparison, not as exact values for any one person: large differences between people, and within the same person, were reported in humans. Two patterns stand out. First, the peak concentrations are tiny, a fraction of a micromole per litre. Second, the gallated catechins, ECG and EGCG, show far lower absorption in plasma and excretion in urine than the nongallated EC and EGC. The book also notes that higher doses may saturate the metabolic pathways: after oral intake of 200, 400, 600 and 800 mg of EGCG, the AUC for unconjugated EGCG rose disproportionately at 800 mg, probably because the conjugation pathways were saturated.

🤔 Pause and think: EGCG is the most abundant catechin in green tea and the most studied, yet its absorption in plasma in this table is the lowest of the four. What does that say about judging a compound by how much of it is in the leaf?

16.3 Three changes: glucuronidation, sulfation and methylation

After absorption, tea catechins are metabolised in the body mainly through glucuronidation and sulfation, two phase II metabolic reactions that change catechins into more water-soluble compounds. Methylation and microbial metabolism also occur. Three enzymes run these reactions: microsomal UDP-glucuronosyltransferases (UGT) catalyse glucuronidation with UDP-glucuronic acid; cytosolic sulfotransferases (SULT) catalyse sulfation with 3'-phosphoadenosine 5'-phosphosulfate (PAPS); and catechol-O-methyltransferases (COMT) catalyse methylation with the methyl group from S-adenosylmethionine (SAM). The three reactions can occur at different sites of the same catechin molecule, and a catechin can be methylated and then glucuronidated or sulfated at other sites.

In human plasma, more EGCG was in the free form than EGC and EC, and EGC and EC were mostly in the conjugated form. At 1 hour after ingestion, 77% of EGCG was in the free form, while 31% of EGC and 21% of EC were free. After decaffeinated green tea or pure catechins, more than 80% of green tea catechins in the plasma were in conjugate form, with two-thirds in sulfate form and one-third in glucuronide form. The two findings come from different measures: the 77% figure is for EGCG alone at 1 hour after ingestion, while the figure of more than 80% conjugated is for all green tea catechins taken together after decaffeinated green tea or pure catechins, where EGC and EC, which are mostly conjugated, outweigh EGCG. ECG was not detected in human plasma. More than 90% of EGC and EC in urine were in the conjugate form, mainly as sulfates.

Two methylated metabolites show why the changed forms matter. One major metabolite in human blood and urine after green tea was 4'-O-methyl-EGC (4'-O-MeEGC). Its plasma level peaked within the first 2 hours of tea ingestion, at four to six times higher than EGC itself. The half-lives of EGC and 4'-O-MeEGC in the blood were 1.02 plus or minus 0.07 and 4.39 plus or minus 1.14 hours, so the methylated form lasts much longer. The amount of 4'-O-MeEGC excreted in urine was about three times higher than EGC, and 88% of it was excreted within 8 hours. An EGCG metabolite, 4',4"-di-O-methyl-EGCG, reached peak plasma values of 20.5 plus or minus 7.7 nM, against 145.4 plus or minus 31.6 nM for EGCG, after 2 hours, with half-lives of 4.1 plus or minus 0.8 and 2.7 plus or minus 0.9 hours. Its concentration was 15% of EGCG in human plasma, and the amount excreted in urine over 24 hours was about five times higher than EGCG, though both together accounted for only about 0.1% of the ingested EGCG. The methylated EGCG had only about one-tenth of the activity of EGCG in scavenging DPPH radicals, but stronger activity than EGCG in inhibiting the spontaneous release of arachidonic acid in human colon cancer HT-29 cells. Methylation changes what the molecule does, not just how long it lasts.

Three changes rebuild a catechin

Catechin absorbed from the gut
EC, EGC, ECG or EGCG

↓

Parallel change 1 of 3: Glucuronidation (UGT)
Adds glucuronic acid. Makes the catechin more water soluble.
Parallel change 2 of 3: Sulfation (SULT)
Adds a sulfate group. Two-thirds of plasma conjugates were sulfates.
Parallel change 3 of 3: Methylation (COMT)
Adds a methyl group. 4'-O-methyl-EGC peaks 4 to 6 times higher than EGC and lasts about 4.4 hours.

The three changes are parallel, not steps in a sequence: they can occur at different sites of the same molecule.

↓

Conjugated catechin in blood and urine
Mostly conjugated: over 80% in plasma, over 90% of EGC and EC in urine.

Figure 16.2: The three phase II changes to tea catechins. The rebuilt forms, not the originals, dominate the blood. Drawn from Ho et al., Tea and Tea Products, Ch. 7.

The book also reports that EGC is methylated more than ECG and EGCG, that EGC is partly O-methylated and excreted in the urine at a large percentage (13.6%), and that 4'-O-methyl-EGC accounted for 30 to 40% of the total metabolites of EGC. ECG and EGC increased the antioxidant activity of human plasma, but EGCG did not increase it significantly, in the study the book cites. The biological activities of the glucuronide, sulfate and methylated catechins and their ring-fission metabolites still need to be determined, the book says, in order to understand the link between tea drinking and human cancer risk.

Quick check

Why does the book say the methylated form 4'-O-methyl-EGC matters more than its size suggests?

Check your answer

It peaked at four to six times the level of EGC, lasted longer with a half-life of about 4.4 hours against 1.0 hour for EGC, and about three times as much was excreted in urine. The changed form, not the original, dominates.

16.4 Gut microbes finish the job

Tea catechins that are not absorbed in the small intestine reach the colon. Animal and human colons hold around 10¹² microorganisms per cubic centimetre. Bacterial enzymes break tea polyphenols down into simple phenolic acids and their conjugates. The low absorption of green tea catechins in the intestine means most catechins are metabolised by colon microbes in the large intestine and then absorbed. Some microbial metabolites may have a physiological effect, and identifying them is an important field of research, with data in humans still limited.

After a single dose of green tea, two major tea catechin metabolites, M4 and M6, were identified in human urine and blood. M4 is (-)-5-(3',4',5'-trihydroxyphenyl)-gamma-valerolactone, and M6 is (-)-5-(3',4'-dihydroxyphenyl)-gamma-valerolactone, with a related form M6', (-)-5-(3',5'-dihydroxyphenyl)-gamma-valerolactone. The cumulative amounts of M4 and M6 excreted in urine were 8 to 25 times higher than those of EGC and EC in some subjects, and accounted for 6 to 39% of the amounts of ingested EGC and EC. During and after repeated consumption of green tea, the cumulative renal excretion of M4 and M6 was more than nine times higher than that of EGC or EC. Their peak levels in plasma were lower than those of EGC and EC. Renal excretion of M4 or M6 had a 3-hour lag time and peaked at 7.5 to 13.5 hours after a single dose of green tea, which shows they are formed in the colon, absorbed, and excreted in urine mostly in the conjugate form. M4, M6 and M6' together accounted for 1.5 to 16% of ingested catechins. M4, M6' and M6 were all seen after ingestion of pure EGCG or EGC by human subjects, but only M6 was detected after EC ingestion. The microbial metabolites seem to have a lag time and longer half-lives than catechins, and so increase the action time of catechins.

Hippuric acid was identified as a major metabolite associated with black tea consumption, and about 43% of the tea phenols were metabolised to hippuric acid in one black tea study. The authors calculated the amounts of catechins, theaflavins, gallic acid and flavonols their volunteers consumed and showed these simple polyphenols could not fully account for the increased urinary hippuric acid, so they suggested the complex thearubigins were also converted to hippuric acid, a suggestion later confirmed. Green tea and black tea consumption result in similar amounts of microbial degradation products absorbed by the body, and the authors speculated that these microbial metabolites, not the native tea flavonoids, may be responsible for at least some of the health effects attributed to tea drinking. But hippuric acid can also come from other sources besides catechins, such as quinic acid and aromatic amino acids, and wine, cider and coffee also raise urinary hippuric acid. The book is plain that hippuric acid is not a suitable biomarker of tea catechin intake, and that 3-hydroxyhippuric acid may be a more valid biomarker of polyphenol intake.

Try it yourself: Take the numbers in this section and sketch the timeline of a cup of green tea. Mark the catechin peak at about 1 to 2 hours, the methylated metabolite lasting about 4.4 hours, and the microbial metabolites peaking at 7.5 to 13.5 hours. The cup is gone in minutes, but its products are still moving through the body half a day later.

16.5 Where catechins go: tissues, urine, and the human-rat gap

Bioavailability in tissues may matter much more than bioavailability in plasma, but little published data are available on tissue distribution of catechins in humans after tea drinking, even in animals. Henning and colleagues studied tissue bioavailability in humans: twenty men consumed green tea, black tea or a caffeine-matched soda control daily for 5 days before radical prostatectomy. Tea catechins, free and conjugated EGC, EGCG and the total of all four catechins, in prostate samples of men drinking black tea and green tea were significantly higher than in men drinking the caffeine-matched soda. That is one of the few direct human tissue measurements the book reports, and it shows catechins do reach tissue, not just blood.

Metabolites of tea catechins are excreted in bile or urine. Small conjugates, such as monosulfates, tend to be excreted in urine, and extensively conjugated metabolites are more likely to be excreted in bile. The total amount of metabolites excreted in urine correlated roughly with maximum plasma concentrations. The exact half-lives of tea catechins in plasma were calculated to be in the order of 2 to 3 hours, except for EGCG, which is eliminated more slowly. Note: the book's text (§7.2.4) says 2-3 hours except EGCG slower, but its own Table 7.2 (reproduced as Table 16.1) gives ECG 6.9 h as the longest and EGC 1.47 h. The two do not agree. The urinary excretion rates of EC and EGC were 18.5 and 11.1%, while EGCG, which is mostly excreted in bile, had a urinary excretion rate of only 0 to 0.1%, so its absorption was underestimated by urine measurements. Note: these excretion figures come from different studies in the book and do not agree: Table 16.1 (50 mg dose, intact catechins) gives EC 5.11% and EGC 1.25% in urine, while §7.2.4 of the book gives EC 18.5% and EGC 11.1%, and the book's own total range is 0-9.8%. Do not quote one set without saying which study it is from. The low cumulative excretion of tea catechins in human urine, 0 to 9.8%, suggested they were extensively metabolised in the human body.

Humans and rats differ. The chemical structures of EC metabolites in human and rat urine are different. In humans, EC is conjugated with glucuronide mainly at the 3' position of the B ring, forming (-)-EC-3'-O-glucuronide, 4'-O-methyl-(-)-EC-3'-O-glucuronide, and 4'-O-methyl-(-)-EC-5 or 7-O-glucuronide. In rats, glucuronidation of EC seems to occur at the 7 position of the A ring, forming 3'-O-methyl-(-)-EC, (-)-EC-7-O-glucuronide, and 3'-O-methyl-(-)-EC-7-O-glucuronide. Hippuric acid was detected in urine of humans consuming green and black tea, but studies in rats indicated these animals do not seem to produce hippuric acid from pure catechin. Chow and colleagues found no significant differences in the pharmacokinetics of EGCG using pure EGCG or Polyphenon E in humans, but a significant difference was found in rats: in rats, EGCG in decaffeinated green tea showed a 3.6 times higher absorption rate constant than pure EGCG. The lesson is direct: a rat result about catechins is not a human result, and the book keeps the two apart.

Pure form, mixed form, and food. The form the catechins arrive in changes the numbers. For 1,050 mg EC in the pure form, peak plasma concentration was 1.53 µM, while for 75 mg EC in decaffeinated green tea form, peak concentration was 0.65 µM. For 400 mg pure EGCG, AUC was 1.71 µM·h on day 1, while for EGCG in Polyphenon E form (200 mg EGCG, 37 mg EGC and 31 mg EC), AUC was 2.59 µM·h. Henning and colleagues concluded that polyphenols given as a green tea extract supplement showed enhanced bioavailability compared with green tea or black tea: the supplement gave significantly higher peak concentrations and AUC, the time to peak was extended by more than 1 hour for all four flavanols, and more EGC and EC were excreted in urine. Food itself matters too, though it has not been studied in much detail. In one study, black tea without milk raised ferric reducing antioxidant power (FRAP) by 65% between 9:00 a.m. and 12:00 p.m., and the value at 3:00 p.m. was 76% higher than at 9:00 a.m., while tea with milk produced no significant change. The book reports this as one study, not as a rule about milk in tea.

Quick check

Why can urine measurements underestimate how much EGCG was absorbed?

Check your answer

EGCG is mostly excreted in bile, not urine, with a urinary excretion rate of only 0 to 0.1%. Urine alone misses most of it, so absorption is underestimated.

16.6 What tea polyphenols are made of

Young tea leaves hold a high concentration of polyphenols and oxidative enzymes, which is why young leaves are better for tea manufacture. Tea polyphenols, previously called tea tannins and also known as tea flavonoids, are directly or indirectly linked with the colour, taste and aroma of tea. Among the polyphenols in fresh tea leaves, catechins are the predominant form, accounting for 12 to 24% of the dry weight.

Table 16.2: Polyphenols in young tea shoots and their typical contents (dry weight basis)

PolyphenolsContent
Total polyphenols18 to 36%
Flavan-3-ols (catechins)12 to 24%
Flavonol and glycosides3 to 4%
Anthocyanins and leucoanthocyanidin2 to 3%
Phenolic acids and depsidesAbout 5%

Catechins. Catechins are a group of flavonoids belonging to flavan-3-ols. In fresh tea leaves the principal catechins are (-)-epicatechin (EC), (-)-epigallocatechin (EGC), (-)-epicatechin gallate (ECG), (-)-epigallocatechin gallate (EGCG), catechin (C) and gallocatechin (GC). EGCG is the most abundant, followed by EGC, ECG and EC. Catechins are responsible for the astringent taste and strength of green tea infusion. During green tea manufacture, most catechins are preserved because the endogenous enzymes are inactivated by dry heating or steaming at the first step. Green tea quality correlates positively with the concentration of polyphenols, but a high concentration of polyphenols or catechins, which makes the infusion strongly bitter and astringent, is not necessarily required for high-quality green tea. High-quality green tea is characterised by high contents of free amino acids with appropriate concentrations of catechins and caffeine.

Table 16.3: Catechins in young tea leaves (dry weight basis)

CatechinContent
(-)-Epicatechin (EC)1 to 3%
(-)-Epigallocatechin (EGC)3 to 6%
(-)-Epicatechin gallate (ECG)3 to 6%
(-)-Epigallocatechin gallate (EGCG)8 to 12%
(+)-Catechin (C)1 to 2%
(+)-Gallocatechin (GC)3 to 4%

Flavonols and glycosides. In fresh tea leaves, 20 flavonols and glycosides have been reported. The flavonols are kaempferol, quercetin and myricetin, at 0.14 to 0.32%, 0.27 to 0.48% and 0.07 to 0.2% of dry weight. Flavonol glycosides are flavonols bound to a monosaccharide or a disaccharide. The principal ones are rutin, quercetin glycoside and kaempferol glycoside, at 0.05 to 0.15%, 0.2 to 0.5% and 0.16 to 0.35% of dry weight. Flavonols are often bright yellow and are thought to contribute to the yellowish green colour of green tea liquor, and some flavon-3-ol glycosides give a silky, mouth-drying sensation and amplify the bitterness of caffeine.

From catechins to theaflavins and thearubigins. During tea fermentation, polyphenols inside the vacuoles of intact leaf cells are released and oxidised by polyphenol oxidases in the cytoplasm. Polyphenol oxidase can use any of the catechins as a substrate to form complex polyphenolic constituents. The catechins undergo enzymatic and chemical oxidation leading to oxidised, condensed and polymerised polyphenols known as theaflavins and thearubigins, which contribute to the colour and taste of black tea liquors. Theaflavins account for 1 to 3% of the dry weight of black tea. Note: Ho Ch. 7 gives total theaflavins in black tea as 3-6%, while the precursor-by-precursor table in Ch. 8 sums to 1-3% for the four major theaflavins; the two chapters also give green-tea flavanols as 35-50% (Ch. 7) against catechins 12-24% (Ch. 8). The figures are reported as printed in each chapter and have not been reconciled. Thearubigins are by far the major components of black tea extract, making up as much as 10 to 20% of the dry weight of black tea, and up to 75% of catechins in tea leaves may ultimately find their way into thearubigins during black tea processing. The chemistry of thearubigins is poorly understood, the book says, because they are hard to separate.

Catechins carry a di- or trihydroxyl group on the B-ring and a meta-5,7-dihydroxyl group on the A-ring. Both rings can be oxidised, with the B-ring more easily oxidised. One-electron oxidation generates a free radical that can join with other radicals to form C-C or C-O bonds, from which theasinensins and related compounds are probably derived. The major theaflavins in black tea are theaflavin (TF), theaflavin-3-gallate (TF-3-G), theaflavin-3'-gallate (TF-3'-G) and theaflavin-3,3'-digallate (TF-3,3'-DG). Their precursors and contents are: EC plus EGC give TF at 0.2 to 0.3%; EC plus EGCG give TF-3-G and ECG plus EGC give TF-3'-G, together at 1.0 to 1.5%; and ECG plus EGCG give TF-3,3'-DG at 0.6 to 1.2%. Compared with EGCG or other catechins, theaflavins have higher molecular weights, more phenol groups, and a benzotropolone moiety, and the book says it is reasonable that they have similar biological activities to their precursor catechins.

Quick check

Which polyphenol group dominates young tea shoots, and what share does it hold?

Check your answer

Catechins, the flavan-3-ols, dominate at 12 to 24% of dry weight, out of 18 to 36% total polyphenols. EGCG is the largest single catechin at 8 to 12%.

16.7 How tea polyphenols act as antioxidants

Free radicals matter in food and in the body. In food, lipid auto-oxidation and rancidity run by a free radical chain mechanism through initiation, propagation and termination, and lipid peroxidation causes off-flavours and undesirable chemical compounds. In the body, free radical damage to cell membrane lipids can lead to cell death, and is an accepted factor in cancer, cardiovascular disease and ageing. The potential health benefits of tea drinking have been partly attributed to the antioxidant properties of tea polyphenols.

What makes a catechin a good antioxidant. In Table 16.4 the lowest first reduction potential is EGC (0.09 V), and the gallate EGCG (0.14 V) is lower than the non-gallate EC (0.19 V); the book's prose generalisation does not hold for every pair in its own table. The 5'-OH group in the B-ring of gallocatechins and catechins without gallate ester is thought to give better antioxidant activity. The reduction potentials of theaflavins are similar to those of simple catechins and higher than those of gallocatechins. TF is not the lowest first reduction potential in the table, but the TF gallates have higher antioxidant activity than TF in the lipid phase, showing that redox potential is an indicator, but not a perfect one, of antioxidant activity.

Table 16.4: First redox potentials of catechins, gallocatechins and theaflavins, against the saturated calomel electrode (V)

ComponentFirst redox potential (V)
Epigallocatechin (EGC)0.09
Gallocatechin (GC)0.13
Epigallocatechin gallate (EGCG)0.14
Gallocatechin gallate (GCG)0.15
Theaflavin (TF)0.16
Epicatechin (EC)0.19
Theaflavin-3'-gallate (TF-3'-G)0.19
Theaflavin-3,3'-digallate (TF-3,3'-DG)0.19
Epicatechin gallate (ECG)0.20
Catechin (C)0.20
Theaflavin-3-gallate (TF-3-G)0.20

The structural features thought to be responsible for antioxidant activity are the o-dihydroxy catechol (3',4'-OH) arrangement on the B-ring, either di- or trihydroxy-substituted catechins or flavonols, and the C2-C3 double bond in the C-ring joined with a C4 carbonyl group in flavonols. The antioxidant activity of catechins is set by the B-ring catechol structure and is further lifted in gallocatechins by the 5'-hydroxyl group on the B-ring. Studies have concluded that a gallate group in the 3-position and a trihydroxy B-ring play the most important role in the free radical scavenging abilities of catechins. Antioxidant function is also thought to be promoted by C5 and C7 dihydroxylation on the A-ring. The A-ring is generally very insensitive to oxidation, but Zhu and colleagues reported that the A-ring of catechins may also serve as an antioxidant site. There is a synergistic effect on antioxidation among catechins, and between catechins and vitamin E or Trolox.

The three mechanisms. The book proposes three mechanisms by which tea polyphenols act as antioxidants.

Three ways tea polyphenols fight oxidation

1. Scavenge reactive oxygen species
Each EGCG traps six superoxide or hydroxyl radicals in laboratory tests. EC traps two.
2. Chelate metal ions
Catechins bind iron and copper, the catalysts of free radical reactions, and so prevent radicals from forming.
3. Regulate enzymes and genes
Tea polyphenols lift the body's own defences, such as superoxide dismutase, catalase and glutathione S-transferase, and damp down prooxidant enzymes such as xanthine oxidase.

Figure 16.3: The three antioxidant mechanisms of tea polyphenols. Scavenging, chelation and regulation work together. Drawn from Ho et al., Tea and Tea Products, Ch. 8.

1. Scavenging reactive oxygen species. Tea polyphenols scavenge superoxide radical, singlet oxygen, hydroxyl radical, nitric oxide, nitrogen dioxide and peroxynitrite. Catechins trap peroxyl radicals and so suppress radical chain reactions and stop lipid peroxidation. They also inhibit metmyoglobin-started peroxidation of low-density lipoproteins (LDLs) and the consumption of alpha-tocopherol. Among tea catechins, EGCG is most effective in reacting with most reactive oxygen species. Yang and colleagues reported that the centres of the scavenging reaction of EGCG are the B, D and A rings, and each EGCG traps six superoxide anions or hydroxyl radicals in laboratory tests as shown by electron spin resonance, whereas EC only traps two. Theaflavins also show strong antioxidant activity against lipid oxidation in rabbit red blood cell ghosts and rat liver homogenates, against LDL oxidation in mouse macrophage cells, in preventing DNA oxidative damage in cell-free systems, and through hydrogen peroxide scavenging.

In the DPPH test system, catechin and epicatechin are oxidised on the B-ring to a phenoxyl radical, which then joins with another catechin unit to form dimers: the B-ring and the A-ring are both principal sites of antioxidant activity. For epigallocatechin and EGCG, the trihydroxyphenyl B-ring, rather than the gallate moiety, is the active site. For theaflavin, the benzotropolone moiety is the active site, forming theanaphthoquinone. Against peroxyl radicals, the B-ring is again the first site of reaction. Against hydrogen peroxide, the A-ring is the site, which proves the A-ring may also serve as an antioxidant site. The main site of antioxidant action depends on the oxidant used, and different oxidants give distinctly different oxidation products.

2. Chelating metal ions. Catechins can chelate metal ions, mainly iron and copper, which catalyse free radical reactions, because of their neighbouring dihydroxy or trihydroxy structures, and so prevent the generation of free radicals. Green and black tea polyphenols reduced cell-mediated LDL oxidation induced by Cu2+ in laboratory tests. Guo and colleagues found that EGC, EGCG, ECG and EC could protect synaptosomes from lipid peroxidation damage in part through iron-chelating activity, with chelated ratios with iron (III) of 3:2, 2:1, 2:1 and 3:1. Sugihara and colleagues found each catechin varied markedly in antioxidative potency depending on the metal ion added, ferrous, copper or vanadium, while butylated hydroxytoluene, a typical lipid radical scavenger, showed similar potency with all metal ions. That difference suggests the metal-chelating property of catechins may play a major role in their antioxidative activity in cultured liver cells. Theaflavins can also chelate iron (III), forming a complex that is then oxidised to o-quinone and converted to dehydrotheaflavin, theanaphthoquinone and polymers.

3. Regulation of enzymes and genes. Tea polyphenols enhance the expression of the body's own antioxidants such as glutathione, glutathione reductase, glutathione peroxidase, glutathione-S-transferase, catalase and quinone reductase. Lin and colleagues found that long-term oral feeding of green tea leaves to Wistar rats lifted the activities of superoxide dismutase (SOD) in serum and of glutathione S-transferase (GST) and catalase in liver. An experiment on human leukaemia HL-60 cells proved that TF-3,3'-DG, TF-3-G, TF and EGCG inhibited the prooxidative enzyme xanthine oxidase, with TF-3,3'-DG most effective. Tea polyphenols also inhibited cyclooxygenase COX-2 and 5-, 12- and 15-lipoxygenase, enzymes involved in enzymatic lipid peroxidation. In RAW 264.7 mouse macrophages, theaflavins, in particular TF-3,3'-DG, inhibited activation of the transcription factor NFκB, preventing expression of the inducible nitric oxide synthase (iNOS) gene and so cutting nitric oxide generation.

🤔 Pause and think: the same EGCG molecule can scavenge six radicals in one test and generate hydrogen peroxide in cell culture medium. What does that say about calling any food compound simply "an antioxidant"?

16.8 Methylglyoxal: a reactive compound tea polyphenols can trap

Methylglyoxal (MG), also called pyruvaldehyde, is one of the reactive carbonyl species. In water it exists in three forms in rapid equilibrium: unhydrated (about 1%), monohydrate (71%) and dihydrate (28%). MG in food and beverages may come from sugars, from intermediates of Maillard reactions, and from lipids, and microbes growing during processing or storage might generate it as well. MG is formed from carbohydrates by auto-oxidation, or by fragmentation and retro-aldol reaction through Maillard reaction intermediates.

MG matters because of glycation. Nonenzymatic glycation is a complex series of reactions between reducing sugars and amino groups of amino acids, peptides and proteins, lipids and DNA. Free amino groups of proteins react with the carbonyl group of reducing sugars such as glucose to form fructosamines through a Schiff base by Amadori rearrangement. The Schiff base and the Amadori product then react further through dicarbonyl intermediates, glyoxal (GO), methylglyoxal (MG) and 3-deoxyglucosone, to form advanced glycation end products (AGEs). GO and MG are the two major alpha-dicarbonyl compounds found in the human body. They are extremely reactive and readily modify lysine, arginine and cysteine residues of proteins. Dicarbonyl compounds are more reactive than reducing sugars and more important for cross-linking proteins in glycation. Higher levels of GO and MG are present in the plasma of diabetic patients than in healthy people: in one recent report, MG in diabetic patients was 16 to 27 µg/dl, compared with 3.0 to 7.0 µg/dl in normal subjects. The rate of MG formation is approximately 120 µM per day, which is about 0.1% of the flux of glucose under normal conditions measured in laboratory red blood cells. Even such a small fraction matters, the book says, because of the high reactivity of MG.

Tea and diabetes: what the book reports. Tea is the most widely consumed beverage in the world and may be effective in the prevention and treatment of diabetes, the book says, with tea polyphenols considered the active components. In human studies, green tea promoted glucose metabolism in healthy humans in oral glucose tolerance tests, and plasma glucose and fructosamine levels of type 2 diabetes patients fell significantly after daily consumption of oolong tea. In animal studies, EGCG lowered serum glucose in Sprague-Dawley rats and in lean and obese Zucker rats, green tea lowered blood glucose in genetically diabetic db/db mice 2 to 6 hours after administration at 300 mg/kg body weight, and tea catechins slowed kidney changes in streptozotocin-induced diabetic rats. In the body, EGCG or green tea reduced carbohydrate absorption by suppressing intestinal alpha-amylase, sucrase or alpha-glucosidase, and green tea improved insulin sensitivity in rats. None of this is presented by the book as proof that tea treats diabetes in people: the human evidence is a small set of studies, and the detailed mechanisms still require further study.

Theaflavins, the main black tea components, were more reactive toward MG than the other polyphenols tested. TF1 (theaflavin) cut MG by 63.1%, TF2 (theaflavin-3 and -3'-gallate) by 60.1%, and TF3 (theaflavin-3,3'-digallate) by 66.7%. All tested theaflavins cut MG by about 66%, which implies that one theaflavin molecule can trap two MG molecules. The book concludes that theaflavins will be excellent candidates for MG scavenging in future studies inside the body.

Table 16.5: Fall in methylglyoxal after 1 hour at 37 °C with tea polyphenols (mole ratio MG to polyphenol 3 to 1)

CompoundFall in MG (%)
MG alone, control5.8
Gallic acid17.1
Pyrogallol27.8
Catechins (most)About 33
EGC, highest catechin45.7
Theaflavin (TF1)63.1
Theaflavin-3 and -3'-gallate (TF2)60.1
Theaflavin-3,3'-digallate (TF3)66.7

The major adduct between EGCG and MG has been identified. The reaction between EGCG and MG occurs dominantly at the C8 position in the A ring of EGCG. EGCG also showed a significant inhibitory effect of 69.1% on MG-mediated protein glycation. The book's final word on this work is careful: tea catechins are able to effectively trap MG, the most important reactive carbonyl species, in laboratory tests. Whether tea catechins can trap reactive carbonyl species inside the body and so reduce the formation of AGEs and prevent the development of diabetic complications needs to be further studied. Measuring MG is itself hard work: MG is very volatile, so it must be derivatised first, for example with O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride (PFBHA) to form oximes, and more than 90% of MG in biological samples is bound to proteins, so perchloric acid is needed to free it.

Trapping methylglyoxal: one catechin traps one, one theaflavin traps two

MG alone at 37 °C for 1 hour
Falls only 5.8%

↓

Add a catechin (EC, ECG, EGC or EGCG)
MG falls about 33%. One catechin traps one MG. EGC is best at 45.7%.

↓

Add a theaflavin (TF1, TF2 or TF3)
MG falls about 66%: 63.1%, 60.1% and 66.7%. One theaflavin traps two MG.

↓

EGCG meets MG at the C8 position of its A ring
EGCG also inhibited MG-mediated protein glycation by 69.1%. All of this is laboratory work: whether it happens inside the body still needs study.

Figure 16.4: Trapping of methylglyoxal by tea polyphenols at 37 °C for 1 hour. Catechins trap one MG each, theaflavins trap two. Drawn from Ho et al., Tea and Tea Products, Ch. 15.

Quick check

Why does a fall of about 33% in MG mean one catechin traps one MG molecule?

Check your answer

The starting mole ratio of MG to polyphenol was 3 to 1. A fall of one-third, about 33%, means one of the three MG molecules was trapped by each polyphenol molecule. A fall of about 66% for theaflavins means two of the three were trapped.

Self-check: what the body does with tea

  1. In humans drinking black tea, about 1.68% of ingested catechins were found in plasma, urine and faeces. What does that number say about catechin absorption?
Check your answer

Only small proportions of catechins are absorbed. Most are changed by metabolism or broken down by gut microbes, and plasma levels of intact catechins stay below about 1 µM.

  1. 4'-O-methyl-EGC peaked at four to six times the level of EGC and lasted about 4.4 hours against 1.0 hour for EGC. Why does this matter?
Check your answer

The methylated metabolite, not the original EGC, dominates the blood for longer. Judging tea by the catechins in the cup misses what the body actually carries.

  1. M4 and M6 excretion peaked at 7.5 to 13.5 hours after a single dose of green tea, with a 3-hour lag. Where are they made, and why the delay?
Check your answer

They are made by microbes in the colon from catechins that were not absorbed in the small intestine. The delay is the time for the catechins to reach the colon and be broken down.

  1. Each EGCG traps six superoxide or hydroxyl radicals in laboratory tests, while EC traps two. Yet EGCG is unstable in cell culture medium, with a half-life under 30 minutes. What follows from these two facts together?
Check your answer

EGCG is a strong radical scavenger in test systems, but it also auto-oxidises and can act as a prooxidant, generating hydrogen peroxide. Whether a polyphenol acts as antioxidant or prooxidant depends on the conditions.

  1. Theaflavins cut methylglyoxal by about 66% in the book's test, while most catechins cut it by about 33%. What does each number mean in molecules?
Check your answer

The starting ratio was 3 MG to 1 polyphenol. A 33% fall means one catechin traps one MG, and a 66% fall means one theaflavin traps two MG molecules.

Try it yourself: Brew a cup of green tea and a cup of black tea. Write down what is in each cup from this unit: green tea holds about 25% catechins by dry weight with EGCG at 65% of them, while black tea holds 3 to 6% theaflavins and 12 to 18% thearubigins. Then write down what the body does with each: low absorption, three phase II changes, microbial breakdown, and the theaflavins trapping two methylglyoxal molecules each in the test tube. Keep the note with the tea, and use it when a label claims a tea "floods your body with antioxidants".

Key ideas to carry forward

  • The four main catechins are EC, EGC, ECG and EGCG. Together they make up about 25% of green tea dry weight, with EGCG at 65% of them. In black tea, flavanols fall to 10%, with theaflavins at 3 to 6% and thearubigins at 12 to 18%.
  • Bioavailability covers liberation, absorption, distribution, metabolism and excretion. Only small proportions of catechins are absorbed: about 1.68% of ingested catechins were found in plasma, urine and faeces in humans drinking black tea, and intact catechins in plasma stay below about 1 µM.
  • After absorption, catechins are rebuilt by glucuronidation (UGT), sulfation (SULT) and methylation (COMT). Over 80% of catechins in plasma are conjugated, and the methylated form 4'-O-methyl-EGC peaks four to six times higher than EGC and lasts about 4.4 hours.
  • Gut microbes break unabsorbed catechins into valerolactones M4, M6 and M6' and phenolic acids. M4 and M6 in urine were 8 to 25 times higher than EGC and EC, peaking at 7.5 to 13.5 hours, and hippuric acid, at about 43% of tea phenols in one black tea study, is not a suitable biomarker of tea intake.
  • Catechins reach human tissue: in prostate samples after 5 days of tea drinking, catechin levels were significantly higher than with a caffeine-matched soda control. EGCG is mostly excreted in bile, so urine underestimates its absorption.
  • Humans and rats metabolise EC differently, at different ring positions, and rats do not seem to produce hippuric acid from pure catechin. Rat results about catechins are not human results.
  • Tea polyphenols in young shoots run 18 to 36% of dry weight: catechins 12 to 24%, flavonol and glycosides 3 to 4%, anthocyanins and leucoanthocyanidin 2 to 3%, phenolic acids and depsides about 5%. Up to 75% of catechins may end up as thearubigins in black tea.
  • Antioxidant activity rests on the B-ring catechol or pyrogallol group, the A-ring meta-5,7-dihydroxyl group, and the galloyl D-ring, plus the benzotropolone skeleton in theaflavins. Each EGCG traps six radicals in laboratory tests, against two for EC.
  • Tea polyphenols act by scavenging reactive oxygen species, chelating iron and copper, and regulating enzymes and genes, including SOD, catalase, GST and xanthine oxidase. They can also act as prooxidants, especially with copper (II) or ferric (III) ions.
  • Methylglyoxal is a reactive dicarbonyl that drives AGE formation. In diabetic patients it ran 16 to 27 µg/dl against 3.0 to 7.0 µg/dl in normal subjects. In the book's test at 37 °C, catechins trapped one MG each and theaflavins trapped two, and EGCG inhibited MG-mediated protein glycation by 69.1%. All of this is laboratory work, and whether it happens inside the body still needs study.

Glossary

Catechin

One of the four main flavanols of tea: EC, EGC, ECG and EGCG. Food tech example: the four catechins making up about 25% of green tea on a dry weight basis.

EGCG

(-)-Epigallocatechin gallate, the most abundant and best studied tea catechin, at 65% of total catechins in green tea. Food tech example: EGCG at 8 to 12% of the dry weight of young tea leaves.

Bioavailability

The concentration of a compound, or its metabolite, at the target organ, often measured as the share reaching the blood. Food tech example: only about 1.68% of ingested catechins found in plasma, urine and faeces in humans drinking black tea.

LADME

Liberation, absorption, distribution, metabolism and excretion: the five linked processes that decide bioavailability. Food tech example: a catechin moving through LADME from the cup to the urine.

Glucuronidation

A phase II change that adds glucuronic acid to a catechin, making it more water soluble, run by UGT enzymes. Food tech example: EGC glucuronide as the major glucuronide metabolite of EGC in human urine.

Sulfation

A phase II change that adds a sulfate group to a catechin, run by SULT enzymes. Food tech example: two-thirds of plasma catechin conjugates in sulfate form.

Methylation

A phase II change that adds a methyl group to a catechin, run by COMT enzymes. Food tech example: 4'-O-methyl-EGC peaking four to six times higher than EGC in human plasma.

Valerolactone (M4, M6, M6')

Ring-fission products made by gut microbes from catechins, found in human blood and urine after tea drinking. Food tech example: M4 and M6 in urine at 8 to 25 times the amounts of EGC and EC.

Hippuric acid

A major urinary metabolite linked with black tea drinking, formed by gut microbes, but not a suitable biomarker of tea intake. Food tech example: about 43% of tea phenols metabolised to hippuric acid in one black tea study.

Theaflavin

A bright orange-red polyphenol formed from catechins during black tea fermentation, carrying a benzotropolone ring. Food tech example: theaflavins at 3 to 6% of black tea, trapping two methylglyoxal molecules each in the book's test.

Thearubigin

A red-brown group of polyphenols formed from catechins during black tea fermentation, making up 10 to 20% of black tea dry weight. Food tech example: up to 75% of leaf catechins ending up as thearubigins.

Free radical

A reactive species with an unpaired electron that can damage lipids, proteins and DNA, and drive rancidity in food. Food tech example: peroxyl radicals starting the chain reaction of lipid peroxidation in stored food.

Antioxidant

A compound that slows oxidation, in tea mainly by scavenging radicals, chelating metals, or regulating enzymes and genes. Food tech example: EGCG trapping six superoxide or hydroxyl radicals in laboratory tests.

Chelation

Binding a metal ion, such as iron or copper, so it cannot catalyse free radical reactions. Food tech example: catechins chelating iron (III) at ratios of 3:2, 2:1, 2:1 and 3:1.

Methylglyoxal (MG)

A highly reactive dicarbonyl, also called pyruvaldehyde, that modifies proteins and drives formation of advanced glycation end products. Food tech example: MG at 16 to 27 µg/dl in diabetic patients against 3.0 to 7.0 µg/dl in normal subjects.

Advanced glycation end product (AGE)

A damaged protein product formed when sugars and dicarbonyls such as MG react with proteins. Food tech example: AGEs forming through glyoxal, methylglyoxal and 3-deoxyglucosone intermediates.

References

  • Ho, Chi-Tang, Lin, Jen-Kun, and Shahidi, Fereidoon (eds.). Tea and Tea Products: Chemistry and Health-Promoting Properties. CRC Press, 2008. Chapter 7: Sun, Ting, Ho, Chi-Tang, and Shahidi, Fereidoon. Bioavailability and Metabolism of Tea Catechins in Human Subjects. Chapter 8: Wan, Xiaochun, Li, Daxiang, and Zhang, Zhengzhu. Antioxidant Properties and Mechanisms of Tea Polyphenols. Chapter 15: Tan, Di, Lo, Chih-Yu, Shao, Xi, Wang, Yu, Sang, Shengmin, Shahidi, Fereidoon, and Ho, Chi-Tang. Trapping of Methylglyoxal by Tea Polyphenols.

Unit 17: Tea, Weight and Metabolism

Note

Unit 17: Tea, Weight and Metabolism

A cup of tea carries catechins and caffeine, and both of them touch the way the body spends energy. This unit follows what the book's studies actually measured: heat production in rat tissue, energy use in men and women in a calorimeter room, body weight in rats fed tea leaves for 30 weeks, and the enzymes that build fat. Every number here comes from those studies, and every study is named by its setting, so you can tell a rat result from a human result at a glance.

A question to start with: if two cups hold the same amount of caffeine, can one of them still raise energy use more than the other? Keep that question in mind. The studies in this unit answer it, and the answer is not what most people guess.

17.1 What is in the cup: catechins, caffeine, and the difference between teas

Tea comes from Camellia sinensis. The book gives world production as about 2.5 million metric tons of dried tea leaves a year. Of all the tea drunk in the world, 78% is black tea, 20% is green tea, and about 2% is oolong tea. Green tea is made by heating fresh leaves fast so fermentation does not start. Oolong tea is partly fermented. Black tea is made by crushing the leaves so fermentation runs before the final drying.

That difference in making changes what is in the cup. Green tea infusions hold about 30 to 40% catechins and other polyphenols and flavonols. Black tea infusions hold about 3 to 9%. During black tea manufacture, catechins fall by 85%, and only 10% can be accounted for as theaflavins and theaflavic acids. The four main catechins in fresh tea leaves are epigallocatechin gallate (EGCG), epigallocatechin (EGC), epicatechin gallate (ECG), and epicatechin (EC). A typical catechin mix, as the book gives it, is 10 to 15% EGCG, 6 to 10% EGC, 2 to 3% ECG, and 2% EC. EGCG is the most plentiful catechin and the one most studied. Caffeine accounts for 3 to 6% of the dry weight of brewed tea. The leaves also hold theobromine, theophylline, gallic acid, and flavonols such as quercetin, in smaller amounts.

Catechins are colourless, water soluble compounds that give tea infusions their bitterness and astringency. The book notes that almost all the character of tea, its taste, colour, and aroma, is tied to the catechins. Their mix shifts with where the plant grows, the variety, the season of harvest, and the process conditions.

One more fact shapes everything that follows. EGCG is not well absorbed: EGC and EC are absorbed faster than EGCG, and EGCG on its own is poorly bioavailable, though it seems better absorbed as a tea infusion than in pure form. After absorption, catechins change through condensation, methylation, glycosylation, and sulfation. Galloyl esters are more active than catechins without a galloyl group, because they have lower redox potentials. Keep that in mind when you read the animal doses: a dose injected into a rat is not a cup of tea drunk by a person.

Quick check

Green tea infusions hold about 30 to 40% catechins. Black tea infusions hold about 3 to 9%. Which tea holds more catechins, and why?

Check your answer

Green tea holds more. Black tea manufacture ferments the leaf, and catechins fall by 85% during that step, with only 10% accounted for as theaflavins and theaflavic acids.

17.2 Thermogenesis: how the body turns fuel into heat

Body weight rests on a balance. Energy taken in must match energy spent, and even a small lasting gap can shift weight over time. The book puts it plainly: there are only two basic ways to treat obesity, reduce energy intake and increase energy expenditure. Total daily energy use has three parts: the heat effect of physical activity, the resting metabolic rate, and the heat effect of eating, called diet-induced thermogenesis.

Thermogenesis is heat production, under the control of the sympathetic nervous system. Its messenger is norepinephrine, shortened to NE. When NE is released at a nerve ending, it acts on alpha and beta adrenergic receptors, raises the level of cyclic adenosine monophosphate (cAMP) inside the cell, and drives heat production, either through higher ATP use or through mitochondrial oxidation poorly coupled to ATP synthesis, so more of the fuel's energy leaves as heat.

Two enzymes set the pace of that signal. Catechol-O-methyl transferase, shortened to COMT, breaks down NE by methylation, and so ends its action. Phosphodiesterase breaks down cAMP inside the cell. This is where tea's two active parts meet the pathway, each at a different step:

Inputs from tea, acting at the top of the pathway

EGCG and caffeine
EGCG blocks COMT. Caffeine blocks phosphodiesterase.

↓

EGCG blocks COMT, so NE lasts longer →
Sympathetic nerve releases norepinephrine (NE)

↓

Caffeine blocks phosphodiesterase, so cAMP lasts longer →
NE acts on receptors: cAMP rises inside the cell

↓

Heat production rises: thermogenesis and fat oxidation

Figure 17.1: How EGCG and caffeine act on the norepinephrine pathway. EGCG slows the breakdown of NE, caffeine slows the breakdown of cAMP, and the two effects add up. Drawn from Ho et al., Tea and Tea Products, Ch. 13.

EGCG inhibits COMT, so NE stays active longer at the nerve ending. Caffeine inhibits phosphodiesterase, so cAMP stays higher inside the cell. The book's phrase for the pair is that they act together to prolong and strengthen the sympathetic signal. Neither one creates NE. They stretch out the signal that is already there.

Try it yourself

Draw the pathway above on paper from memory, then check it against Figure 17.1. Cover the two side boxes and say aloud what EGCG does and what caffeine does. If you mix the two up, you are not alone: the mix-up is the most common error in this unit, and the diagram is the fix.

17.3 What the rat tissue studies showed

The clearest tests of this pathway used brown adipose tissue from rats, the tissue that specialises in heat production. The measure was MO₂, the respiration rate of interscapular brown adipose tissue (IBAT). Dulloo and colleagues tested green tea extract, caffeine alone, and EGCG, at matched concentrations.

Caffeine alone, at 100 and 250 µM, did not raise basal IBAT respiration. Millimolar concentrations would be needed for caffeine alone to stimulate it. Green tea extract holding the same amount of caffeine did raise MO₂ above basal levels, in a dose-dependent way. When the animals were chemically sympathectomised, so their sympathetic nerves were depleted of NE, the green tea effect was greatly blunted. The blunting was not blamed on damage to the tissue after the nerve treatment. It was credited to the loss of NE stores. That result points the same way as the pathway diagram: without NE to prolong, there is little for EGCG to prolong.

EGCG alone at 200 µM gave a small but statistically significant rise in MO₂, about 40% above basal values. Caffeine at 100 µM on its own had no effect. Together, EGCG at 200 µM plus caffeine at 100 µM raised MO₂ 2.4-fold over basal values. The pair did more than either part alone.

The team then added ephedrine, which increases NE release, at low doses. Ephedrine at 0.1 µM alone had no effect on MO₂. Add caffeine, and MO₂ rose 84% over basal. Add caffeine plus EGCG at 200 µM, and the rise was 2.8-fold. Ephedrine at 0.25 µM alone raised MO₂ by 50 to 100%. Add caffeine, and the rise reached 4-fold. Add caffeine plus EGCG, and it reached 7.4-fold. EGCG at 100 µM, which had no effect on its own, raised MO₂ when ephedrine was present. The book's reading: EGCG does not increase the release of NE, it prolongs the action of NE once it is released, and the effect grows as more NE is present.

EGCG blocks COMT directly. Using EGC and 3,4-dihydroxy-L-phenylalanine as substrates, EGCG, 4''-O-methyl-EGCG, and 4',4''-di-O-methyl-EGCG were all potent inhibitors of COMT, and the methylated forms inhibited it noncompetitively, which suggests they bind at sites other than the catechin binding site. As EGCG levels rose, COMT activity fell.

One caution from the chapter belongs here. Catechins bind strongly to proteins rich in proline, such as caseins in milk, gelatin, and salivary proteins. Hydrophobic interactions and hydrogen bonding form protein-catechin complexes. Those complexes can reduce the absorption of tea catechins and the digestibility of dietary protein. A cup of tea drunk with milk is not the same test as a pure extract in a lab dish.

Quick check

In the rat tissue study, caffeine at 100 µM alone had no effect on MO₂. EGCG at 200 µM plus caffeine at 100 µM raised MO₂ 2.4-fold. What does that tell you about the pair?

Check your answer

The pair does more than either part alone. EGCG prolongs the NE signal and caffeine prolongs the cAMP signal, so together they strengthen thermogenesis in a way neither achieves on its own at those doses.

17.4 What the human studies measured: 24-hour energy use

Rat tissue is a clean test of a mechanism. It is not a person. The human evidence in the chapter comes from small, careful studies in calorimeter rooms, where every kilojoule in and out is measured. Read the numbers with the sample sizes beside them.

Dulloo and colleagues tested ten healthy men. Each man received, in turn, a placebo, caffeine alone at the level found in the tea extract, and a green tea extract rich in catechin polyphenols and caffeine. Diet and activity were kept the same across all treatments. The results are in Table 17.1.

Table 17.1: Energy expenditure in ten healthy men: placebo, caffeine, and green tea extract (kJ per period)

PeriodPlaceboCaffeineGreen tea extract
Diurnal energy expenditure6463 ± 3866547 ± 3836754 ± 352
Nocturnal energy expenditure3075 ± 1493053 ± 1453112 ± 140
Total 24-hour energy expenditure9538 ± 5219599 ± 5189867 ± 488

Values from Dulloo et al. 1999, as given in Ho et al., Tea and Tea Products, Ch. 13, Table 13.2.

Diurnal energy use with green tea extract was higher than with placebo by 4.5%, and higher than with caffeine by 3.2%. Total 24-hour energy use with green tea extract was higher than with placebo by 3.5%, and higher than with caffeine by 2.8%. Six of the ten men showed a rise of 266 to 836 kJ after green tea extract, against only two of the ten after caffeine. Caffeine at the matched level did not change energy use, respiratory quotient, or substrate oxidation. The book's reading: the effect cannot be explained by caffeine content alone, and the catechins act together with caffeine to promote catecholamine-induced thermogenesis.

How big is a 4% rise? The chapter works it out. Thermogenesis contributes up to 8 to 10% of daily energy use, about 760 to 950 kJ, in a typical sedentary man. A 4% rise in total 24-hour energy use, about 328 kJ, would mean a 35 to 43% rise in the thermogenesis part of daily energy use. The chapter then gives a long-term arithmetic example: a lower energy use of 75 kJ per day would, in theory, lead to nearly 8 lb of weight loss per year. That is arithmetic, not a measured outcome, and the chapter presents it as such. But the book's own conversion does not check out: 75 kJ a day is about 18 kcal a day, about 6,570 kcal a year, which is under 2 lb a year at 3,500 kcal per lb, not nearly 8 lb. The figure is quoted here as the book prints it, and the conversion is flagged as faulty. No study in the chapter followed people for a year to test it.

Worked example

Take the total 24-hour values from Table 17.1. Green tea extract minus placebo is 9867 minus 9538, which is 329 kJ, about 3.5%. Caffeine minus placebo is only 61 kJ, about 0.6%. The gap between the full extract and caffeine alone is the story of this unit.

Quick check

Why did the researchers give one group caffeine alone at the same level as in the green tea extract?

Check your answer

To separate the caffeine effect from the catechin effect. Caffeine alone did not raise 24-hour energy use, while the full extract did, so the extra effect is credited to the catechins acting with caffeine, not to caffeine on its own.

17.5 Oolong tea and energy use in men and women

Two more human studies in the next chapter tested oolong tea, the partly fermented tea. Both were small, and both measured energy use in a calorimeter.

In the first, twelve men took four treatments in turn, each for three days: water, full-strength oolong tea brewed from 15 g of tea a day, half-strength oolong tea brewed from 7.5 g of tea, and water holding 270 mg of caffeine, the same amount as in the full-strength oolong tea. They avoided caffeine and flavonoids for four days before the study, and ate a typical diet matched to each man's needs. On the third day, energy use was measured by indirect calorimetry in a room calorimeter.

Against water, energy use rose by 2.9% with full-strength oolong tea and by 3.4% with caffeinated water. That is an extra 281 kJ a day with oolong tea and 331 kJ a day with caffeinated water. Fat oxidation was 12% higher with full-strength oolong tea than with water. The authors' reading, as the chapter reports it, is that oolong tea stimulated energy use and fat oxidation in normal-weight men, and could help a person hold a lower body fat level. Note the honest limit in the numbers: caffeinated water raised energy use slightly more than oolong tea did in this study, while only oolong tea raised fat oxidation by that 12%.

In the second study, eleven healthy, normal-weight women drank water, oolong tea, and green tea in random order, and resting energy use was measured over two hours. Resting energy use was similar before the drinks, and stayed low after water and after green tea, but rose significantly after oolong tea. The cumulative rises over resting energy use across the two hours were 110.7 kJ (26.5 kcal) after oolong tea, 49.5 kJ (11.8 kcal) after green tea, and 11.2 kJ (2.7 kcal) after water. Because oolong tea held less caffeine and less EGCG than green tea, the authors credited the rise to the more polymerised polyphenols found in oolong tea. That is their interpretation, and the chapter reports it as such.

Quick check

In the women's study, oolong tea raised two-hour energy use by 110.7 kJ, green tea by 49.5 kJ, and water by 11.2 kJ. Oolong tea held less caffeine and less EGCG than green tea. What did the authors credit the oolong result to?

Check your answer

To the more polymerised polyphenols in oolong tea. That is the authors' interpretation as reported in the chapter, not a direct measurement of those polyphenols in the study.

17.6 Body weight over 12 weeks: the open-label study

The longest human study in the chapter followed moderately obese patients for 12 weeks on a green tea extract. Chantre and Lairon recorded body weight and waist circumference. Mean body weight fell by 4.60% and waist circumference by 4.48% over the 12 weeks. The numbers are in Table 17.2.

Table 17.2: Body weight and waist circumference over 12 weeks on green tea extract

MeasureBaselineWeek 4Week 8Week 12
Number of people70686366
Mean body weight (kg)75.99 ± 8.8574.64 ± 8.9373.97 ± 9.0272.49 ± 8.97
Mean waist circumference (cm)94.41 ± 11.7290.6 ± 10.8689.98 ± 10.9188.27 ± 10.8

Values from Chantre and Lairon 2002, as given in Ho et al., Tea and Tea Products, Ch. 13, Tables 13.3 and 13.4.

Read the cautions beside the numbers. The study used an open, uncontrolled design: no placebo group, and everyone knew what was being taken. Without a control group, the 4.6% fall cannot be credited to the extract alone, since diet, activity, and simply being in a study can all move weight. The chapter also records that in the Dulloo human study there were no side effects and no significant differences in heart rate across treatments. Green tea extract is distinct from sympathomimetic drugs, the chapter notes, because those drugs are limited by adverse cardiovascular effects. That is a comparison of drug classes, not a safety promise for any product.

The chapter's own closing is careful. It calls green tea extract a natural substance for the management of obesity in clinical studies, and at the same time says the bioavailability and biotransformation of EGCG are not well understood, and further studies are necessary, especially in patients with severe obesity. It also says further studies are required to find the best dosage, and pure components in capsules might not deliver the active catechins as well as drinking tea. Nothing in this unit is advice to drink tea to lose weight.

17.7 Fermented teas in rats: oolong, black, pu-erh, and green

The next chapter asks a different question. Green tea holds the most catechins, so a simple story would say green tea should work best on body weight. The rat feeding study it reports says otherwise.

Three-week-old male Sprague-Dawley rats were fed a basal diet, or the same diet with 1.5% or 4% green, oolong, black, or pu-erh tea leaves, for 30 weeks. Food intake differed little between the five groups. At 30 weeks, with 1.5% tea leaves, green tea gave no reduction in body weight, while oolong tea, black tea, and pu-erh tea all gave significant reductions. With 4% tea leaves, all four teas reduced body weight. The figures at 30 weeks were about 6% lower than the basal diet group for green tea, 11% lower for oolong tea, about 7% lower for black tea, and about 13% lower for pu-erh tea. The chapter's overall ranking for suppression of body weight is oolong tea first, then pu-erh tea, then black tea, then green tea. That ranking is the chapter's summary across the whole study, and it sits beside the single-dose figures above, where pu-erh tea showed the largest percentage fall at the 4% dose. Both are reported here as the chapter gives them.

The same rat work measured blood fats. All four teas suppressed plasma triacylglycerol, cholesterol, and low-density lipoprotein (LDL) cholesterol in the animals. Pu-erh tea and oolong tea lowered triacylglycerol more significantly than green tea and black tea. Pu-erh tea and green tea were more efficient than oolong tea and black tea in lowering total cholesterol. The fermented teas, oolong, black, and pu-erh, were more effective than unfermented green tea in suppressing body weight and lipogenesis in rats. That is the chapter's headline, and it is a rat result.

Purified EGCG was also tested on its own in rats and mice. Intraperitoneal injection of EGCG, but not EC, EGC, or ECG, caused acute body weight loss in Sprague-Dawley rats within 2 to 7 days. Female rats injected daily with 12.5 mg of EGCG, about 92 mg per kg of body weight, lost 10% of their body weight against their starting weight, and 29% against the control weight, after 7 days. In mice, EGCG in the diet reduced the body weight gain seen on a high-fat diet in a dose-dependent way: control mice gained twice as much body weight during four weeks as animals given 1% EGCG in the diet. Food intake did not change, the reduced gain came only from less body fat, and lean body mass was not affected. Injection into the abdomen is not drinking tea, and a 1% EGCG diet is not a cup of tea.

Quick check

At the 4% dose after 30 weeks, which tea gave the largest percentage fall in rat body weight, and which tea gave no fall at the 1.5% dose?

Check your answer

At 4%, pu-erh tea gave the largest fall, about 13% below the basal diet group. At 1.5%, green tea gave no reduction in body weight, while oolong, black, and pu-erh teas all gave significant reductions.

17.8 Lipogenesis and fatty acid synthase: how fat is built

Lipogenesis is the making of fat. Its key step is the synthesis of fatty acids, the complete synthesis of palmitate from acetyl-CoA in the cytosol. In the rat, the pathway is well represented in adipose tissue and liver. In most mammals, glucose is the main substrate for lipogenesis. The enzyme at the centre of it is fatty acid synthase, shortened to FAS. FAS carries out all the reactions in the conversion of acetyl-CoA and malonyl-CoA to palmitate, using seven active sites. Its level responds to nutrition and hormones in liver and adipose tissue, mainly through changes in FAS gene transcription.

Tea polyphenols act on FAS in two settings, and the chapter keeps them apart. In vitro, in a test tube, EGCG is an effective inhibitor of FAS from chicken liver. Its inhibition has a reversible fast-binding part, through which 52 µM of EGCG can inhibit 50% of the activity of FAS. The marked inhibition of ketoacyl reduction points to the beta-ketoacyl reductase part of FAS. Protection by NADPH, and competitive inhibition of NADPH for ketoacyl reduction, suggests EGCG may compete with NADPH for the same binding site.

In cells, the picture is about gene expression, not just enzyme blocking. FAS is overexpressed in the human breast carcinoma cell line MCF-7, and EGF raises it further. Green and black tea extracts at 120 µg per ml blocked that EGF-induced expression of FAS, while at 30 µg per ml they slightly stimulated it. EGCG suppressed FAS expression, but catechin, EC, and EGC did not. Theaflavins TF-1, TF-2a, TF-2b, and TF-3 suppressed FAS at both protein and mRNA levels. The route, as the chapter traces it: EGCG and TF-3 block EGF binding its receptor, block the PI3K and Akt pathway, reduce the DNA-binding of the transcription factor Sp-1, and so lower FAS gene expression. A PI3K inhibitor, Ly294002 at 5 µM, could block FAS expression completely. All of this was shown in cultured human cancer cells, not in people.

Input: EGCG and TF-3 ⊗ block EGF binding its receptor
Tea polyphenols act at the receptor step, at the top of the chain

↓ blocked at the receptor

EGF binds its receptor (EGFR)

↓

PI3K and Akt pathway switches on, but less strongly

↓

Sp-1 binds DNA in the nucleus, but less strongly

↓

FAS gene expression falls: fat synthesis drops

Figure 17.2: How tea polyphenols suppress fatty acid synthase (FAS) in the chapter's cell studies. EGCG and theaflavin-3,3'-digallate (TF-3) act on the EGF receptor, PI3K, Akt, and Sp-1 route. Drawn from Ho et al., Tea and Tea Products, Ch. 14.

Try it yourself

Take a food label that lists fat per serving, and work out how much of that fat the body would need to make from scratch if none came from the diet. That is the job FAS does. Now list the three places the chapter says tea polyphenols touch that job: the enzyme itself in a test tube, the gene in cultured cells, and blood fats in rats and hamsters. Keep the three settings apart in your notes.

17.9 Uncoupling proteins and the mitochondria

The second half of the chapter's mechanism sits in the mitochondria, where fuel is burned. Uncoupling proteins let fuel oxidation run without making ATP, so the energy leaves as heat. UCP-1 is the original uncoupling protein, and its job is thermogenesis. UCP-2 and UCP-3 appear to work in reactive oxygen species handling and in fatty acid metabolism. In brown adipose tissue, cold exposure increases substrate use so glucose and fatty acid oxidation detach from the usual coupling to ATP synthesis, and the result is thermogenesis.

17.10 A practical note: reading a weight claim about tea

Tea products are sold with weight claims. This unit gives you a short test for any such claim, built only from the studies above.

First, ask what was tested: a cup of tea, a tea extract, a pure compound, or an injection. The rat injection study used EGCG injected into the abdomen, and the mouse diet study used 1% EGCG in feed. Neither is a cup of tea. Second, ask who or what was tested: rat tissue, rats, mice, hamsters, cultured cells, or people. The strongest human numbers in this unit come from ten men and twelve men in calorimeter rooms, and eleven women over two hours. Third, ask whether there was a control group. The 12-week weight study had none, so its 4.6% fall cannot be credited to the extract alone. Fourth, ask whether the dose is stated. The chapter says the best dosage for tea consumption is not yet known, and further studies are required to find it. Fifth, check the size of the effect against the size of the study. A 3.5% rise in 24-hour energy use is real in the data, and it is small in absolute terms, about 329 kJ a day in that study.

A food technologist does not need to settle whether tea helps anyone lose weight. The job is to read the study, name its setting, quote its numbers, and refuse to stretch a rat result into a human promise.

Quick check

A label says a tea extract "burns fat, proven in studies". Name two questions you would ask before accepting that claim, using this unit.

Check your answer

Any two of these: was it tested in people or only in rats, mice, hamsters, or cells? Was there a control group? What was the dose, and was it a cup of tea or a pure extract or an injection? How big was the effect, and how many people were studied?

Key ideas to carry forward

  • Green tea holds far more catechins than black tea, about 30 to 40% of the infusion against 3 to 9%, because fermentation cuts catechins by 85%.
  • EGCG inhibits COMT, so norepinephrine acts longer. Caffeine inhibits phosphodiesterase, so cAMP stays higher. Together they strengthen thermogenesis more than either does alone at the same doses.
  • In rat brown adipose tissue, EGCG at 200 µM plus caffeine at 100 µM raised respiration 2.4-fold over basal, and with low-dose ephedrine the rise reached 7.4-fold.
  • In ten healthy men, green tea extract raised total 24-hour energy use by 3.5% over placebo and 2.8% over matched caffeine, about 329 kJ a day. Caffeine alone did not raise it.
  • In twelve men, full-strength oolong tea raised energy use by 2.9% and fat oxidation by 12% over water. In eleven women, oolong tea raised two-hour energy use by 110.7 kJ, against 49.5 kJ for green tea and 11.2 kJ for water.
  • In rats fed tea leaves for 30 weeks, fermented teas beat green tea on body weight: at 4%, pu-erh tea about 13% lower, oolong tea 11% lower, black tea about 7% lower, and green tea about 6% lower than the basal diet group.
  • Tea polyphenols suppress fatty acid synthase: EGCG at 52 µM inhibits 50% of FAS activity in a test tube, and EGCG and TF-3 suppress FAS gene expression in cultured MCF-7 cells through the EGFR, PI3K, Akt, and Sp-1 route.
  • The human evidence is small, short, and mixed in design. The chapter calls for further studies on dose, bioavailability, and severe obesity. No result in this unit is a recommendation to use tea for weight loss.

Glossary

Thermogenesis

Heat production by the body, often from burning fuel without making ATP. Food tech example: the heat a brown fat cell gives off when its mitochondria burn fatty acids is thermogenesis at work.

Catechin

A type of tea polyphenol, mainly EGCG, EGC, ECG, and EC. Food tech example: catechins give a green tea infusion its bitterness and astringency.

EGCG

Epigallocatechin gallate, the most plentiful catechin in green tea. Food tech example: EGCG is the catechin named on green tea extract labels, and the one tested at 200 µM in the rat tissue studies.

Theaflavin

A polyphenol formed when catechins change during fermentation of black tea. The main ones are TF-1, TF-2a, TF-2b, and TF-3. Food tech example: theaflavins give black tea its colour and part of its brisk taste, and TF-3 is the theaflavin that suppressed FAS in the cell studies.

Norepinephrine (NE)

A messenger of the sympathetic nervous system that raises heat production. Food tech example: NE is the signal that EGCG prolongs by blocking the enzyme that breaks NE down.

COMT

Catechol-O-methyl transferase, the enzyme that breaks down norepinephrine and other catechol compounds by methylation. Food tech example: a tea extract that inhibits COMT slows the clearing of NE, so the heat signal lasts longer in the tissue test.

cAMP

Cyclic adenosine monophosphate, a messenger inside the cell that carries the NE signal onward to heat production. Food tech example: caffeine raises cAMP by blocking phosphodiesterase, the enzyme that breaks cAMP down.

Phosphodiesterase

The enzyme that breaks down cAMP inside the cell. Food tech example: caffeine's thermogenic action in this unit works by inhibiting phosphodiesterase.

Brown adipose tissue (BAT)

A tissue that specialises in turning fuel into heat. Food tech example: the rat studies in this unit measured the respiration rate of interscapular brown adipose tissue, called IBAT, as a direct test of thermogenesis.

MO₂

Respiration rate, the rate at which a tissue uses oxygen. Food tech example: MO₂ of IBAT was the readout in the rat tissue studies, rising 2.4-fold with EGCG plus caffeine.

Energy expenditure (EE)

The energy the body spends, usually measured in kilojoules per day. Food tech example: 24-hour energy use in a calorimeter room is the gold-standard measure used in the ten-man study in this unit.

Lipogenesis

The making of fat in the body, mainly the synthesis of fatty acids from acetyl-CoA. Food tech example: lipogenesis in liver and adipose tissue is the process that fatty acid synthase drives, and that tea polyphenols suppress in the studies here.

Fatty acid synthase (FAS)

The enzyme that turns acetyl-CoA and malonyl-CoA into palmitate. Food tech example: EGCG at 52 µM inhibits 50% of FAS activity in a test tube.

Triacylglycerol

The main form of fat stored in the body and carried in the blood. Food tech example: plasma triacylglycerol fell in rats fed oolong, black, pu-erh, and green tea leaves, with pu-erh and oolong teas lowering it most.

LDL cholesterol

Low-density lipoprotein cholesterol, the blood fat fraction often called bad cholesterol in plain speech. Food tech example: LDL cholesterol fell in rats fed tea leaves in the 30-week study.

Uncoupling protein (UCP)

A mitochondrial protein that lets fuel oxidation run without making ATP, so energy leaves as heat. UCP-1 drives thermogenesis. Food tech example: UCP-1 in brown fat is the protein behind the heat production measured as MO₂ in the rat tissue studies.

Polymerised polyphenol

A larger polyphenol built by joining smaller catechin units together during fermentation. Food tech example: oolong tea holds more polymerised polyphenols than green tea, and the authors of the women's study credited them for oolong tea's larger rise in energy use.

Calorimeter room

A sealed room that measures all the heat and gases a person gives off, so total energy use can be counted exactly. Food tech example: the ten-man and twelve-man studies in this unit measured 24-hour energy use in a calorimeter room.

References

  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 13, Green Tea-Induced Thermogenesis Controlling Body Weight, by John Shi, Sophia Jun Xue, and Yukio Kakuda.
  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 14, Fermented Tea Is More Effective Than Unfermented Tea in Suppressing Lipogenesis and Obesity, by Jen-Kun Lin and Shoei-Yn Lin-Shiau.

Unit 18: Tea, Inflammation and Disease Prevention

Note

Unit 18: Tea, Inflammation and Disease Prevention

Inflammation is the body's alarm: redness, swelling, pain, and heat after injury. When the alarm never switches off, it damages tissue, and the book links that long-running damage to several diseases, including some cancers. This unit follows what tea polyphenols did in the book's studies: in mouse ears, in cultured cells, in rats, and in one human trial. Every finding is labelled with its setting, because a result in a dish is not a result in a person.

A question to start with: if a compound calms swelling in a mouse ear, does that mean a cup of tea calms swelling in a person? Hold that question through the whole unit. The chapter's own closing answers it, and the answer is careful.

18.1 What inflammation is, and when it turns harmful

Inflammation is the body's reaction to tissue injury or damage. The chapter defines it by four signs: redness, swelling, pain, and heat. Injury can come from heat, chemicals, radiation, microbial infection, trauma, or immune processes. The body answers in two layers: a primary defence that acts at once, which includes inflammation, and a secondary defence run by the immune system.

Acute inflammation follows a set order. Blood vessels narrow briefly, then widen, and capillary walls let more fluid through, which brings redness, swelling, and heat. White blood cells, leukocytes, travel to the injured site by chemotaxis to attack the intruder. During this process, mediators are released, along with reactive oxygen species (ROS) and reactive nitrogen species (RNS), to help control the process. If acute inflammation lasts too long because its events fall out of control, it becomes chronic inflammation: sustained tissue damage, with a build-up of inflammatory mediators, growth factors, ROS, and RNS. Those can damage DNA, protein, and cell membranes, and the chapter links that damage to arthritis, ulcerative colitis, inflammatory bowel disease, and cancer.

The link between inflammation and cancer was first proposed in 1863 by Rudolph Virchow, who noted that cancer arises from sites of chronic inflammation. The chapter reports that about 15% of cancers worldwide are caused by chronic inflammation, and gives examples: gastric cancer from Helicobacter pylori, liver cancer from hepatitis C, colorectal cancer from inflammatory bowel disease, lung cancer from prolonged cigarette smoking, oesophageal cancer from reflux oesophagitis, and mesothelioma from asbestos exposure. Those are associations reported in the literature the chapter cites, not findings from tea studies.

Three groups of mediators carry the story. Cytokines are short-acting peptide messengers that regulate growth, cell division, inflammation, and immunity. Tumour necrosis factor, IL-6, and IL-8 are named as cytokines that promote tumour growth. Arachidonic acid metabolites work through the cyclooxygenase (Cox) and lipoxygenase (Lox) pathways, linked to cell proliferation, blocking of apoptosis, and growth of new blood vessels. ROS and RNS, such as hydrogen peroxide, nitric oxide, and superoxide anion, damage proteins, DNA, RNA, and lipids, and can switch on pathways such as c-MYC, c-FOS, NF-κB, and MAPKs. The same free radicals can also push cells toward apoptosis by opening pores in mitochondria and releasing cytochrome c.

Quick check

Name the four classic signs of inflammation, and say in one line what chronic inflammation adds to the picture.

Check your answer

Redness, swelling, pain, and heat. Chronic inflammation adds sustained tissue damage, with a build-up of inflammatory mediators, growth factors, ROS, and RNS that can damage DNA, protein, and cell membranes.

18.2 The cast of tea polyphenols

The polyphenols in tea fall into families, and the chapter keeps them apart because they do not behave the same. Green tea's main polyphenols are catechins: catechin itself, gallocatechin, epicatechin (EC), epigallocatechin (EGC), epicatechin-3-gallate (ECG), and epigallocatechin-3-gallate (EGCG), with EGCG the most plentiful. Black tea's main polyphenols are theaflavins, formed as catechins change during fermentation: theaflavin (TF-1), theaflavin-3-gallate (TF-2a), theaflavin-3'-gallate (TF-2b), and theaflavin-3,3'-digallate (TF-3). Oolong tea holds an intermediate mix, including theasinesin A, a polyphenol isolated from oolong tea and confirmed by synthesis from free radical oxidation of EGCG. Pu-erh tea, highly fermented, holds polymerised catechins of high molecular weight that are poorly characterised because they dissolve poorly in most solvents. Its water extract is active in blocking lipopolysaccharide-induced inducible nitric oxide synthase activity.

Two extracts appear by name in the animal work. Black tea theaflavins, shortened to BTTs, hold mainly TF-1, TF-2, and TF-3. Green tea catechins, shortened to GTCs, hold mainly catechins such as EGCG. In a transgenic mouse model of mammary cancer, with inactivated tumour suppressor genes p53 and Rb, intervention with either extract increased survival: control mice lived 143.5 ± 8.25 days, BTT mice 154 ± 16.83 days, and GTC mice 151.2 ± 8.02 days, and mammary carcinogenesis fell. That is a mouse model result.

Four parallel products from one leaf

Fresh leaf: catechins, mainly EGCG

Each tea below is made from the same leaf, at a different level of fermentation. No tea on this row turns into the next one.

Green tea
Unfermented
Catechins, mainly EGCG
Oolong tea
Partly fermented
Catechins, theaflavins, theasinesin A
Black tea
Fermented
Theaflavins TF-1, TF-2a, TF-2b, TF-3
Pu-erh tea
Highly fermented
Large polymerised catechins

Figure 18.1: Tea polyphenol families by tea type. Green, oolong, black and pu-erh teas are four parallel products made from the same fresh leaf at different fermentation levels, not stages in one sequence. Drawn from Ho et al., Tea and Tea Products, Ch. 9.

Try it yourself

Take three tea labels from a shop shelf: one green, one oolong, one black. Write down which polyphenol family each tea is richest in, using Figure 18.1. If the label names EGCG, it is pointing at the green tea family. If it names theaflavins, it is pointing at the black tea family.

18.3 How tea polyphenols act: the antioxidant step

The first mechanism the chapter describes is the simplest to picture. Tea polyphenols are strong scavengers of superoxide, hydrogen peroxide, hydroxyl radicals, nitric oxide, and peroxynitrite. They also block the enzyme xanthine oxidase (XO), which makes uric acid and superoxide. TF-3 acts as a competitive inhibitor of XO and is the most potent inhibitor of the enzyme among the compounds examined, and it also blocked superoxide production in HL-60 cells. The chapter's reading: the antioxidative activity of tea polyphenols comes not only from scavenging superoxides, but also from blocking XO and the oxidative signals tied to it.

The one human trial in this part of the chapter tested that idea in people. It was a randomised, double-blinded, placebo-controlled phase IIa chemoprevention trial with green tea polyphenol capsules in 124 individuals. The participants were seropositive for both HBsAg and aflatoxin-albumin adducts, and took capsules daily at 500 mg or 1000 mg, or a placebo, for three months. The marker was 8-hydroxydeoxyguanosine (8-OHdG) in urine, a marker of oxidative DNA damage.

At the start, the three groups were alike. After one month, 8-OHdG levels did not differ between the groups: medians of 1.83, 2.08, and 1.86 ng per mg of creatinine for placebo, 500 mg, and 1000 mg groups. After three months, 8-OHdG fell significantly in both treated groups: medians of 2.02, 1.03, and 1.15 ng per mg of creatinine for placebo, 500 mg, and 1000 mg groups, with p equal to 0.007. Urinary EGC and EC levels rose significantly and in a dose-dependent way in both treated groups. The chapter's reading is that urinary EGC and EC can serve as a practical marker of green tea consumption in people, and that chemoprevention with green tea polyphenols is effective in reducing oxidative DNA damage in this trial. Note the limits: 124 people, three months, a specific high-risk group, and a urine marker, not a disease outcome.

Quick check

In the human trial, after three months the median urinary 8-OHdG was 2.02 ng per mg of creatinine with placebo, 1.03 with 500 mg of green tea polyphenols, and 1.15 with 1000 mg. Did the marker rise or fall with treatment?

Check your answer

It fell in both treated groups, roughly by half against placebo. The chapter reads that as a reduction in oxidative DNA damage in this trial, measured by a urine marker over three months.

18.4 Switching off growth signals: MAPK, PI3K, and the cell cycle

Cancer cells grow because growth signals stay switched on. The chapter describes several points where tea polyphenols switch those signals off, all shown in cell studies.

First, the growth factor receptors. In human epidermoid A431 cells, EGCG significantly blocked DNA synthesis, measured by tritiated thymidine uptake, and strongly blocked the protein tyrosine kinase activities of the EGF receptor, the PDGF receptor, and the FGF receptor in a test tube, with an IC50 of 0.5 to 1 µg per ml. EGCG scarcely blocked the protein kinase activities of pp60src, PKC, and PKA. The block was selective, not general.

Second, the MAPK pathway. Overactivity of the transcription factors AP-1 and NF-κB is a feature of some cancer routes, including UVB-induced skin tumour formation, and it can come from activation of MAPK pathways. EGCG and theaflavins blocked TPA- and EGF-induced transformation of JB6 mouse epidermal cells in a dose-dependent way, and the block tracked with lower JNK activation and less AP-1 binding to its recognition site. In Ras-transformed mouse epidermal cells, treatment with 20 µM of EGCG or TF-3 lowered the levels of phosphorylated Erk1/2 and MEK-1/2. EGCG blocked the link between Raf-1 and MEK-1, while TF-3, but not EGCG, increased the breakdown of Raf-1. Both compounds also blocked the kinase activity of Erk1/2 directly, by competing with its substrate ELK-1 for the enzyme's active site. In normal human epidermal keratinocytes exposed to UVB radiation, pretreatment with EGCG blocked UVB-induced hydrogen peroxide production and the phosphorylation of MAPK pathways that followed it.

Third, the PI3K pathway. In mouse epidermal JB6 cells, pretreatment with EGCG or TF-3 blocked UVB-induced PI3K activation, and the downstream effectors Akt and ribosomal p70S6 kinase were also dampened. The block on p70S6 kinase was direct, while the block on Akt was only weak, so the compounds do not switch off the whole pathway evenly.

Fourth, the cell cycle. EGCG blocked cell cycle progression at the G1 phase in asynchronous MCF-7 human breast cancer cells. Over 24 hours of exposure, the Rb protein changed from its hyperphosphorylated form to its hypophosphorylated form, and G1 arrest developed. Cyclins D1 and E fell slightly. EGCG blocked the activities of cyclin-dependent kinases 2 and 4 (Cdk-2 and Cdk-4) in a cell-free system in a dose-dependent way, and when cells were exposed to EGCG at 30 µM over 24 hours, both Cdk2 and Cdk4 activities fell gradually. EGCG also raised the Cdk inhibitor p21, in step with a rise in p53 levels, and raised p21 mRNA. Within six hours of EGCG treatment, the Cdk inhibitor p27 protein rose. The chapter's reading: EGCG slows growth either by acting on G1 control proteins such as Cdk2 and Cdk4, or by raising the inhibitors p21 and p27.

Worked example

IC50 is the concentration that gives 50% inhibition. EGCG blocked the EGF receptor kinase in A431 cells with an IC50 of 0.5 to 1 µg per ml. Compare that with the 30 µM used in the MCF-7 cell cycle study, and the 20 µM used in the Ras-transformed cell study. The numbers sit in the same broad range, low micromolar, which is why the chapter treats these cell studies as a coherent set. None of these numbers is a blood level in a tea drinker, and the chapter does not present them as one.

18.5 The inflammation switches: iNOS, NF-κB, and IKK

Inflammation and cancer prevention meet at a small set of switches, and tea polyphenols act on them in the chapter's cell and animal work.

Inducible nitric oxide synthase, shortened to iNOS, makes nitric oxide in activated immune cells. Gallic acid, EGC, EGCG, TF-1, TF-2, and TF-3 all blocked nitrite production, iNOS protein, and iNOS mRNA in macrophages activated by thioglycolate and lipopolysaccharide. In LPS-activated macrophages treated with EGCG or theaflavins, the 130 kDa iNOS protein and the 4.5 kb iNOS mRNA fell significantly. EGCG blocked the activation of NF-κB, the transcription factor needed to switch iNOS on, and the theaflavins blocked the disappearance of its inhibitor IκB from the cytosol. The chapter's reading: the compounds lower iNOS mRNA, possibly by stopping NF-κB from binding the iNOS promoter, and so blocking iNOS transcription.

NF-κB itself is controlled by IκB kinase, shortened to IKK. In LPS-activated mouse macrophages of the RAW 264.7 cell line, TF-3 blocked IKK activity more strongly than the other polyphenols tested. It strongly blocked both IKK1 and IKK2, and stopped the breakdown of IκBα and IκBβ in activated macrophage cells. TF-3 also blocked phosphorylation of IκB and held down both NF-κB activity and iNOS levels. The chapter suggests the block could act directly on IKKs, or on events upstream in the signal pathway.

Quick check

TF-3 blocked IKK activity more strongly than the other polyphenols tested. What two things did that block stop in the macrophage cells?

Check your answer

It stopped the breakdown of IκBα and IκBβ, and it held down NF-κB activity and iNOS levels in the activated macrophages.

18.6 The proteasome, the matrix, and apoptosis

Three more mechanisms complete the chapter's list, and each one is a different way to stop a cancer cell from surviving and spreading.

The proteasome is the cell's protein shredder, a large complex that destroys short-lived proteins. The 26S proteasome holds a 20S catalytic core and two 19S regulatory caps, with three activities: chymotrypsin-like, trypsin-like, and peptidyl glutamyl peptide hydrolase (PGPH). Its targets include the tumour suppressors p53 and p27Kip1, and destroying them can promote tumour growth. EGCG blocked the chymotrypsin-like activity of the 20S proteasome in leukemic, breast cancer, and prostate cancer cell lines, leading to a build-up of p27kip1 and IκB, followed by cell cycle arrest and blocking of NF-κB activity. EGCG also selectively blocked topoisomerase I, but not topoisomerase II, in human colon cancer cell lines, at 10 to 17 µM. Chapter 11, covered later in this unit, goes deeper into the proteasome work.

Matrix metalloproteinases, shortened to MMPs, are enzymes that cut through the basement membrane and the extracellular matrix, the first step in a cancer spreading. EGCG blocked the activity of secreted MMP-2 and MMP-9 at concentrations of 8 to 13 µM. At 1 µM, EGCG raised the expression of TIMP-1 and TIMP-2, proteins that block activated MMPs. Because such concentrations are physiologically significant, the chapter calls this an attractive mechanism for the anti-invasive and anti-angiogenic activities of EGCG seen in animals. TF-3 appeared to be the most active of the compounds examined for MMP blocking.

Apoptosis is programmed cell death, and some tea polyphenols push cancer cells toward it. Certain green tea catechins and black tea theaflavins are very potent blockers, with Ki in the nanomolar range, of the anti-apoptotic Bcl-2 family proteins Bcl-XL and Bcl-2. Theasinensin A, TF-1, and TF-2a and TF-2b showed strong growth-blocking effects against human histolytic lymphoma U937 cells, with an IC50 of 12 µM, but were less effective against Jurkat cells. Treatment with tea polyphenols switched on caspase-3, but not caspase-1, and increased cleavage of poly(ADP-ribose)-polymerase (PARP). Theasinensin A also caused loss of mitochondrial potential, release of cytochrome c into the cytosol, caspase-9 activity, and breakdown of DFF-45, so the DNA-cutting enzyme could enter the nucleus.

The chapter's Figure 9.3 ties these threads into one map: tea polyphenols can stimulate the inactivation of pro-carcinogens by conjugation enzymes, block their activation by cytochrome P450 enzymes, and so reduce DNA adducts and tumour initiation. Figure 9.4, described in the text, shows tea polyphenols suppressing MAPK and NIK pathways while switching on the caspase cascade, a combination that points toward apoptosis in the target cells. The chapter is plain that the molecular mechanisms of cancer chemoprevention by tea extract are not fully worked out.

Four separate bodies of work, side by side

Antioxidant work
Tea polyphenols scavenge ROS and block xanthine oxidase
Growth signal work
They block growth signals: EGF receptor, MAPK, PI3K, and the G1 cell cycle step
Inflammation switch work
They block inflammation switches: IKK, NF-κB, and iNOS
Proteasome, MMP and apoptosis work
They block the proteasome and MMPs, and switch on apoptosis through caspases

Figure 18.2: The mechanisms of cancer chemoprevention by tea polyphenols, as listed in Ho et al., Tea and Tea Products, Ch. 9. Each box is a separate body of cell and animal work, not a single proven chain in people. Drawn from Ho et al., Tea and Tea Products, Ch. 9.

18.7 Tea and inflammatory disease: skin, joints, lungs, gut, and heart

Chapter 10 turns from mechanisms to whole tissues and animals, and to a small number of human observations. Tea and its constituents have been reported as helpful in several inflammatory conditions, including osteomyelitis, psoriasis, arthritis, and inflammation of the respiratory system. Each example below carries its setting.

In bone, osteomyelitis is inflammation of the bone with bone loss, and infected osteoblasts release inflammatory mediators such as cytokines. Treatment with EGCG blocked the release of certain cytokines in osteoblasts infected with Staphylococcus aureus, in cell studies. In skin, EGCG prevented damage caused by UVA radiation and by TPA application, through blocking NF-κB DNA binding and the expression of MAP kinase and Cox-2. Green tea polyphenols eased symptoms of psoriasis in mice. Psoriasis is an inflammatory skin disease with overgrowth and faulty differentiation of epidermal keratinocytes.

In joints, arthritis affects nearly one in three adults in the United States, in forms such as osteoarthritis and rheumatoid arthritis. The usual anti-inflammatory drugs are slow acting and can cause long-term toxicity. In normal and arthritic cartilage tissue, catechins were effective in easing symptoms of arthritis such as collagen breakdown. EGCG was effective in osteoarthritis through blocking the production of pro-inflammatory mediators, nitric oxide and PGE2, in osteoarthritic tissue. Giving green tea polyphenols to mice lowered the incidence of arthritis, with a fall in inflammatory mediators such as Cox-2, interferon-γ, and tumour necrosis factor. An epidemiological study showed that older women drinking more than three cups of green tea a day had a significantly lower risk of rheumatoid arthritis than those who drank no tea. That is an observational link, not a trial, and the chapter reports it as one.

In the lungs, cigarette smoke given to guinea pigs led to lung damage, oxidative stress, and inflammation, and those were prevented by giving black tea beverage. Tea catechins eased airway inflammation by blocking the production of inflammatory mediators in nasal fibroblasts and bronchial epithelial cells, with EGCG and ECG the most effective of the catechins tested. EGCG was also helpful in a mouse asthma model, by blocking inflammatory cell migration and ROS production.

Try it yourself

Pick one condition from this section, for example arthritis or asthma. Write two columns: what was tested, and in what setting. If your notes say "green tea helps arthritis" without a setting, rewrite the line until the setting is in it. That rewrite is the whole skill of this unit.

18.8 The mouse ear test: theaflavins against TPA inflammation

The chapter's own experiment is the cleanest test in the unit, and it gives numbers you can check. The team tested black tea theaflavin derivatives on skin inflammation in mice, induced by TPA, 12-O-tetradecanoylphorbol-13-acetate. Both ears of the mice were treated with acetone or the test compound in acetone, 20 minutes before TPA in acetone was applied. The mice were sacrificed after five hours. Ear punches 6 mm across were weighed, and the tissue was tested for inflammatory mediators. Female CD-1 mice, 7 to 8 weeks old, six mice per group, were used.

All the compounds tested, theaflavin mixture, theaflavin monogallate (TF-2), and theaflavin-3,3'-digallate (TF-3), were effective in blocking ear swelling and the levels of inflammatory mediators, in a dose-dependent way. The theaflavin compounds blocked both the cytokines, the peptide mediators, and the eicosanoids, the lipid mediators. Table 18.1 gives the theaflavin mixture results.

Table 18.1: Theaflavin mixture against TPA-induced ear inflammation in female CD-1 mice

TreatmentEar punch weight (mg)IL-1β (pg per mg tissue)IL-6 (pg per mg tissue)PGE2 (pg per mg tissue)LTB4 (pg per mg tissue)
Acetone (no TPA)8.10 ± 0.300.95 ± 0.150.34 ± 0.211000 ± 751.9 ± 0.17
TPA (0.8 nmol)14.10 ± 0.4888.50 ± 4.003.95 ± 0.022400 ± 6169.8 ± 0.21
Theaflavins (0.25 µmol) plus TPA10.90 ± 0.6512.80 ± 1.052.40 ± 0.032100 ± 2602.53 ± 0.21
Theaflavins (0.5 µmol) plus TPA9.78 ± 0.507.10 ± 2.051.95 ± 0.04833 ± 1442.12 ± 0.18
Theaflavins (1 µmol) plus TPA8.31 ± 0.351.92 ± 0.150.40 ± 0.031277 ± 811.18 ± 0.04

Values from Ho et al., Tea and Tea Products, Ch. 10, Table 10.1. Asterisks in the book mark values significantly different from the TPA group at p less than 0.05.

Read the dose trend down the table. Ear punch weight falls from 14.10 mg with TPA alone to 8.31 mg with 1 µmol of theaflavins plus TPA, close to the acetone value of 8.10 mg. IL-1β falls from 88.50 to 1.92 pg per mg of tissue. IL-6 falls from 3.95 to 0.40. LTB4 falls from 9.8 to 1.18. PGE2 does not fall smoothly: it is 2400 with TPA, 2100 at the low dose, 833 at the middle dose, and 1277 at the high dose. That uneven PGE2 line is in the book's table, and it is reported here as it stands.

The chapter ends this section with a warning that belongs in bold in your notes. The effect seen in a dish and in certain animal experiments is not reproducible in the human situation. That gap could come from confounding factors in epidemiological studies, or from the poor bioavailability of catechins. More detailed mechanistic studies are warranted to understand the effects of tea catechins in animals so they can be made applicable to the human situation. A mouse ear is not a human joint, and the chapter says so.

Quick check

In Table 18.1, which mediator fell most sharply in absolute terms when theaflavins at 1 µmol were added to TPA: IL-1β, IL-6, or LTB4?

Check your answer

IL-1β. It fell from 88.50 to 1.92 pg per mg of tissue, a fall of about 86.6. IL-6 fell from 3.95 to 0.40, and LTB4 from 9.8 to 1.18.

18.9 Breast cancer cells and the proteasome

Chapter 11 goes deep into one mechanism: the proteasome. The proteasome is the cell's system for destroying selected proteins, and it matters for cancer because it destroys tumour suppressors such as p53 and p27Kip1. Blocking the proteasome lets those protectors build up, and the cell stops dividing or dies. The chapter tested theaflavins and EGCG against the proteasome in purified form and in extracts of human cancer cells.

The cell lines were human leukemia Jurkat T cells, human histiocytic leukemia U937 cells, promyelocytic leukemia HL60 cells, and human breast cancer MCF-7 cells. After 24 hours of treatment, cell viability was measured by ATP content, since ATP is present in all metabolically active cells and falls fast when cells die. The theaflavins blocked growth in all four cell lines in a dose-dependent way. The leukemic lines Jurkat T, U937, and HL60 were more sensitive than the breast cancer line MCF-7. MCF-7 cells were more resistant, with an estimated IC50 of 23 to 30 µM. TF-3 was the most effective, with an estimated IC50 of 12 µM in Jurkat T cells and 6 µM in U937 cells. TF-2a and TF-2b behaved alike in the four cell lines. Theaflavins showed stronger growth-blocking activity than EGCG in all cell lines.

Against purified proteasomes, theaflavins and EGCG blocked the chymotrypsin-like activity of the recombinant 20S proteasome, the eukaryotic 20S proteasome, and the eukaryotic 26S proteasome. The pattern was similar across the three sources, so the compounds block both the 20S and 26S forms in eukaryotic cells. In tumour cell extracts, the order of sensitivity was U937, then Jurkat T, then MCF-7, then HL60. TF-3 at 5 µM blocked about 80% of the proteasomal chymotrypsin-like activity in U937 cell extracts, while EGCG had only a slight blocking effect. The concentrations needed to block the 26S proteasome in cell extracts were higher than those needed for purified 20S proteasome, which suggests that enzymes present in cell extracts may break down theaflavins.

Table 18.2 gives the comparison across polyphenols at 5 µM against the purified recombinant 20S proteasome, as chymotrypsin-like activity left, in percent of control.

Table 18.2: Chymotrypsin-like activity left after polyphenols at 5 µM (percent of control)

FlavonoidActivity left (%)TanninActivity left (%)Tea polyphenolActivity left (%)
Rutin77.0 ± 5.5Tannic acid30.0 ± 11.1EGCG56.6 ± 1.5
Quercetin52.1 ± 3.1Geraniin42.0 ± 3.7TF-158.3 ± 1.6
Apigenin92.0 ± 2.2Corilagin57.0 ± 2.1TF-2a28.3 ± 9.8
Genistein94.2 ± 0.7Ellagic acid104.4 ± 1.9TF-2b27.1 ± 9.0
Genistin94.6 ± 6.0TF-316.2 ± 3.8

Values from Ho et al., Tea and Tea Products, Ch. 11, Table 11.3.

TF-3 left only 16.2% of activity, the strongest block in the table. The pattern the chapter draws from this table is about structure. Polyphenols carrying ester bonds blocked the chymotrypsin-like activity more strongly than polyphenols without ester bonds. TF-3 at 5 µM blocked about 85% of proteasome activity, while 0.3 µM of pentagalloylglucose (5GG) blocked 80%, with an estimated IC50 of 73.2 nM. But TF-1 and quercetin, which hold no ester bonds, also blocked the activity, so other structural features must contribute as well. Gallic acid and n-propyl gallate showed no significant block, though n-propyl gallate, which holds one gallate esterified to propanol, blocked slightly more than gallic acid. The chapter's reading is that the number of galloyl-ester groups may raise the blocking effect.

18.10 Prostate cells: aromatase, 5α-reductase, and the androgen receptor

Chapter 12 studies prostate cancer cells, and its logic turns on two enzymes. Testosterone is converted by 5α-reductase into dihydrotestosterone (DHT), the most potent androgen inside the prostate, which binds the androgen receptor (AR) and drives androgen action. Aromatase converts androgens, androstenedione and testosterone, into the oestrogens estrone (E1) and estradiol (E2). Blocking either enzyme lowers the hormone signal that drives the cancer cells. Tea polyphenols block both, which is unusual: the chapter notes that tea polyphenols differ from 5α-reductase blockers and aromatase blockers by holding both blocking activities.

Aromatase was tested in microsomes from rat ovary and human placenta. TF-1, TF-2, TF-3, and EGCG all blocked it in a concentration-dependent way. The IC50 values for rat ovarian microsomes were 5.72 µM for TF-1, 7.33 µM for TF-2, 20.37 µM for TF-3, and 18.27 µM for EGCG. For human placental aromatase, they were 4.17, 3.23, 3.45, and 13.79 µM in the same order.

The 5α-reductase results are in Table 18.3, measured in rat liver microsomes.

Table 18.3: Blocking of 5α-reductase by polyphenols in rat liver microsomes

CompoundIC50 (µM)
TF-34.9
5GG7.8
TF-2b7.9
TF-2a20.2
EGCG92.6
TF-1More than 100
Gallic acidMore than 100
n-Propyl gallateMore than 100

Values from Ho et al., Tea and Tea Products, Ch. 12, Table 12.1. IC50 is the concentration giving 50% inhibition.

TF-3 was the most effective blocker among the polyphenols tested, at 4.9 µM. EGCG, the green tea catechin, needed 92.6 µM, nearly twenty times as much. TF-1, gallic acid, and n-propyl gallate did not reach 50% blocking even at 100 µM.

The chapter then followed the signal past the enzyme. After testosterone becomes DHT, DHT binds the androgen receptor. Of six polyphenols tested, 5GG and TF-3 significantly reduced the amount of AR protein at a dose of 40 µM. TF-2a, TF-2b, TF-1, and EGCG had only minor effects at the same concentration. Prostate-specific antigen (PSA) is an androgen-responsive gene, and its promoter holds a functional androgen responsive element. PSA is the most commonly used serum marker for diagnosing prostate cancer, and serum PSA usually rises in patients with prostate carcinomas. Treatment of LNCaP human prostate cancer cells with 5GG or TF-3 significantly suppressed PSA levels in a dose-dependent way.

Finally, the chapter tested growth. LNCaP cells are androgen dependent. Testosterone alone raised LNCaP cell numbers by an average of 50% above the untreated control over four days. Treatment with 5GG, TF-3, TF-2b, or EGCG at 10 to 40 µM blocked that growth in a dose-dependent way, bringing cell numbers back toward the level of the no-androgen controls at the higher concentrations of 20 to 40 µM. The chapter's summary is that TF-3, EGCG, and 5GG block androgen action in prostate cancer cells by two routes: blocking 5α-reductase activity, and lowering AR protein levels. It presents them as compounds with potential as chemopreventive and chemotherapeutic agents for prostate tumours, and that word, potential, is doing the work: all of this is cell and microsome work, not a trial in men.

Testosterone

Two parallel routes from the same starting material. Neither enzyme box leads into the other.

Left branch: aromatase route
Aromatase makes estradiol (E2), which acts on the oestrogen receptor. Tea polyphenols block aromatase.
Right branch: 5α-reductase route
5α-reductase makes DHT, which acts on the androgen receptor. Tea polyphenols block 5α-reductase and lower AR protein.

Both routes converge here:

LNCaP prostate cancer cell growth falls, and PSA levels fall

Figure 18.3: The two hormone routes tea polyphenols block in the prostate cell studies. Aromatase and 5α-reductase are two parallel routes from testosterone, not steps in one chain, and both lead to lower LNCaP growth and lower PSA. One compound family acts on both routes. Drawn from Ho et al., Tea and Tea Products, Ch. 12.

Quick check

TF-3 blocked 5α-reductase with an IC50 of 4.9 µM. EGCG needed 92.6 µM. Which is the stronger blocker, and by roughly how much?

Check your answer

TF-3 is stronger. Its IC50 is about nineteen times lower than EGCG's, so it reaches 50% blocking at about one nineteenth of the concentration.

18.11 A practical note: reading a disease claim about tea

Health claims about tea are everywhere, and this unit gives you a short test for them, built only from the chapters above.

First, name the setting. Was the result in a dish, in a mouse, in a rat, in a guinea pig, or in people? The proteasome work was in purified enzymes and human cell lines. The ear swelling work was in mice. The arthritis link in older women was observational. Only the 8-OHdG trial, 124 people over three months, was a randomised human trial, and it measured a urine marker, not a disease.

Second, ask what dose was used, and in what form. Cell studies used 5 to 40 µM of pure compounds. The mouse ear test applied theaflavins to the skin. None of those is a cup of tea, and the chapters do not give a conversion from a dish dose to a drinking dose.

Third, ask whether the chapter itself claims a human effect. Chapter 9 says the molecular mechanisms are not fully worked out. Chapter 10 says the effect seen in a dish and in animals is not reproducible in the human situation, and names poor bioavailability of catechins as a possible reason. Chapter 12 calls the compounds potential agents. The honest summary of this unit is that tea polyphenols show many interesting actions in cells and animals, and the human evidence is thin.

Fourth, watch for the word prevention. Chemoprevention means stopping cancer before it starts, and the chapters use it for laboratory and animal work. It is not a promise that drinking tea prevents cancer in people. No result in this unit is medical advice, and none of it recommends tea as a treatment for any disease.

Quick check

A product page says "green tea prevents cancer, proven by science". Using this unit, give two reasons that sentence overstates the book.

Check your answer

Any two of these: the book's evidence is mostly in cells and animals, with only one small human trial measuring a urine marker; the book says the mechanisms are not fully worked out; the book says the dish and animal effects are not reproducible in people, possibly because catechins are poorly absorbed; the book calls the compounds potential agents, not proven treatments.

Key ideas to carry forward

  • Inflammation has four signs, redness, swelling, pain, and heat. When it persists, it damages DNA, protein, and membranes, and the chapter links chronic inflammation to about 15% of cancers worldwide.
  • Green tea is rich in catechins, led by EGCG. Black tea is rich in theaflavins TF-1, TF-2a, TF-2b, and TF-3. Oolong tea holds a mix, including theasinesin A.
  • In a human trial of 124 people over three months, green tea polyphenol capsules at 500 mg and 1000 mg lowered urinary 8-OHdG, a marker of oxidative DNA damage, to medians of 1.03 and 1.15 against 2.02 for placebo.
  • Tea polyphenols block growth signals in cell studies: the EGF receptor, the MAPK pathway, the PI3K pathway, and the G1 step of the cell cycle, through Cdk2, Cdk4, p21, and p27.
  • They block inflammation switches: IKK, NF-κB, and iNOS. TF-3 blocked IKK most strongly in mouse macrophages.
  • In the mouse ear test, theaflavin mixture at 1 µmol brought ear weight from 14.10 mg with TPA alone down to 8.31 mg, and IL-1β from 88.50 to 1.92 pg per mg of tissue.
  • Theaflavins block the proteasome's chymotrypsin-like activity. TF-3 left only 16.2% of activity at 5 µM, and blocked about 80% of the activity in U937 cell extracts. The galloyl-ester structure is linked to that blocking power.
  • Tea polyphenols block both aromatase and 5α-reductase in microsome tests. TF-3 blocked 5α-reductase with an IC50 of 4.9 µM, against 92.6 µM for EGCG. 5GG and TF-3 lowered androgen receptor protein and PSA in LNCaP prostate cancer cells.
  • The human evidence is thin, the mechanisms are not fully worked out, and the book says dish and animal effects are not reproducible in people, possibly because catechins are poorly absorbed. Nothing here is a recommendation to use tea to treat or prevent any disease.

Glossary

Inflammation

The body's reaction to tissue injury: redness, swelling, pain, and heat. Food tech example: the mouse ear test in this unit measures inflammation as ear swelling after TPA is applied to the skin.

Acute inflammation

Short-term inflammation that clears the injury and then stops. Food tech example: the widening of blood vessels and arrival of white blood cells after a cut is acute inflammation.

Chronic inflammation

Long-running inflammation that keeps damaging tissue. Food tech example: the chapter links chronic inflammation to arthritis, bowel disease, and about 15% of cancers worldwide.

Cytokine

A short-acting peptide messenger of inflammation and immunity. Food tech example: IL-1β and IL-6 are cytokines measured in the mouse ear test, falling sharply when theaflavins were applied.

Interleukin

A type of cytokine that carries signals between white blood cells. Food tech example: IL-1β rose to 88.50 pg per mg of tissue in the TPA mouse ear, and fell to 1.92 with theaflavin mixture.

Eicosanoid

A lipid mediator of inflammation made from arachidonic acid, including prostaglandins and leukotrienes. Food tech example: PGE2 and LTB4 are eicosanoids measured in the mouse ear test.

ROS

Reactive oxygen species, highly reactive molecules such as superoxide and hydrogen peroxide that can damage cells. Food tech example: tea polyphenols scavenge ROS, and the human trial measured 8-OHdG, a marker of oxidative DNA damage.

RNS

Reactive nitrogen species, reactive molecules containing nitrogen, such as nitric oxide and peroxynitrite. Food tech example: iNOS makes nitric oxide in activated immune cells, and tea polyphenols block iNOS.

NF-κB

A transcription factor that switches on genes for inflammation, including iNOS. Food tech example: TF-3 blocks NF-κB by blocking IKK, the enzyme that frees NF-κB from its inhibitor.

IKK

IκB kinase, the enzyme that breaks down IκB and so frees NF-κB to act. Food tech example: TF-3 blocked both IKK1 and IKK2 more strongly than the other polyphenols tested in mouse macrophages.

iNOS

Inducible nitric oxide synthase, the enzyme that makes nitric oxide in activated immune cells. Food tech example: EGCG and theaflavins blocked iNOS protein and mRNA in activated macrophages.

Chemoprevention

Stopping cancer before it starts, using food or drug compounds. Food tech example: the chapters use this word for laboratory and animal work with tea polyphenols, not as a promise about drinking tea.

Apoptosis

Programmed cell death, the cell's orderly self-destruction. Food tech example: theasinensin A pushed lymphoma cells toward apoptosis by switching on caspase-3 and caspase-9.

Proteasome

The cell's protein shredder, which destroys selected short-lived proteins, including tumour suppressors such as p53 and p27. Food tech example: TF-3 blocked the proteasome's chymotrypsin-like activity, leaving only 16.2% of it at 5 µM.

MMP

Matrix metalloproteinase, an enzyme that cuts through tissue barriers and helps a cancer spread. Food tech example: EGCG blocked MMP-2 and MMP-9 at 8 to 13 µM in the chapter's cell work.

Aromatase

The enzyme that converts androgens into oestrogens, estrone and estradiol. Food tech example: theaflavins blocked aromatase in rat ovary and human placenta microsomes, with IC50 values as low as 3.23 µM.

5α-reductase

The enzyme that converts testosterone into DHT, the most potent androgen in the prostate. Food tech example: TF-3 blocked 5α-reductase with an IC50 of 4.9 µM in rat liver microsomes.

Androgen receptor (AR)

The protein that DHT binds to inside cells, switching on androgen-driven growth. Food tech example: 5GG and TF-3 lowered AR protein in LNCaP prostate cancer cells at 40 µM.

PSA

Prostate-specific antigen, an androgen-responsive gene product and the most commonly used serum marker for prostate cancer. Food tech example: 5GG and TF-3 suppressed PSA levels in LNCaP cells in a dose-dependent way.

IC50

The concentration of a compound that gives 50% inhibition of a target. Food tech example: a lower IC50 means a stronger blocker, so TF-3 at 4.9 µM is stronger than EGCG at 92.6 µM against 5α-reductase.

References

  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 9, Mechanisms of Cancer Chemoprevention by Tea and Tea Polyphenols, by Jen-Kun Lin.
  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 10, Effect of Tea and Tea Constituents on Inflammation, by Divya Ramji, Mou-Tuan Huang, Fereidoon Shahidi, and Chi-Tang Ho.
  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 11, Inhibition of Breast Cancer Cell Proliferation by Theaflavins and Epigallocatechin 3-gallate through Suppressing Proteasomal Activities, by Jen-Kun Lin, Yu-Wen Chen, and Shoei-Yn Lin-Shiau.
  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 12, Suppression of Prostate Carcinogenesis by Tea Polyphenols through Inhibiting Microsomal Aromatase and 5α-Reductase, by Jen-Kun Lin.

Unit 19: Tea Products and Flavour: Diversified Teas and Tea Drinks

Note

Unit 19: Tea Products and Flavour: Diversified Teas and Tea Drinks

A tea factory throws away 30 to 40% of its leaf as sub-standard produce, and a bottled tea drink can taste different after sterilisation than it did before. This unit covers both problems: turning low-grade leaf into new products, and keeping a ready-to-drink tea tasting like tea after heat treatment. The first half is about new products built on tea. The second half is about flavour chemistry, and why a few parts per billion of one compound can change a whole drink.

A question to start with: a bottled green tea is heated twice, once to sterilise it and once to fill the bottle. Which step changes the flavour more, and which single compound would you measure first to find out? Keep that question in mind. The chapter answers it with numbers.

19.1 Why diversify: the sub-standard leaf problem

Tea is the most consumed beverage in the world after water, with per capita consumption of about 120 mL a day, as given in the chapter. Of all the tea produced and consumed in the world, 78% is black, 20% is green, and about 2% is oolong. Every tea processing unit faces the same problem: 30 to 40% of its produce is sub-standard. That leaf still holds all the essential characteristics of quality tea, but in smaller amounts. Throwing it away wastes it. Converting it into diversified products puts it to work.

Both black tea and green tea can be turned into value-added products with a ready world market. The chapter notes that consumer interest in the health-promoting and disease-preventive effects of foods has exploded, and those foods carry several names: functional foods, nutraceuticals, healthy foods, healthful foods, and wellness foods. The Institute of Medicine's Food and Nutrition Board defined functional foods as any food or food ingredient that may provide a health benefit beyond the traditional nutrients it contains. A shorter working definition in the chapter is that functional foods are foods or food components that provide a health benefit beyond basic nutrition. A still shorter one is a food that has medicinal effects. The idea is not new: the chapter quotes Hippocrates from about 2,500 years ago, "Let food be thy medicine and medicine be thy food."

Tea fits that idea well because it is a storehouse of nutraceutical chemicals. Research on tea and human health has been fuelled by the growing need for naturally healthy diets that include plant-derived polyphenols. The chapter says there is a need to work out how known functional components in foods could expand the role of diet in disease prevention and treatment. That is the research programme behind this whole unit.

The market has already moved that way. The chapter reports that the specialty tea sector rose in value by 50% and the herbal and fruit tea sector by 30% between 2002 and 2004, fuelled by consumers trading up to premium specialty teas and by healthier lifestyles. Ready-to-drink tea consumption is increasing, and the typical ready-to-drink buyer is younger and prefers sweeter, flavoured drinks, while higher quality flavoured and unflavoured teas are a growing trend of their own.

Quick check

A factory grades 1,000 kg of made tea a day. If 30 to 40% is sub-standard, how many kilograms a day is that, and what does the chapter say to do with it?

Check your answer

300 to 400 kg a day. The chapter says such teas hold all the essential characteristics of quality tea in smaller amounts, and can be converted into quality diversified products holding all the nutraceutical values.

19.2 Tea as a nutraceutical base

Tea polyphenols are the core of the nutraceutical case. The chapter cites growing evidence that tea polyphenols reduce the risk of heart disease and cancer in humans, and notes studies linking tea with antiallergic action and antimicrobial properties. It also reports that co-administration of drugs with catechins, EC and EGCG, inhibits glucuronidation and sulfonation of orally administered drugs, so raising the bioavailability of those drugs. Epidemiological studies have linked tea consumption with a lower risk of several types of cancer, including those of the stomach, oral cavity, oesophagus, and lungs. The chapter's reading is that tea appears to be an effective chemopreventive agent for toxic chemicals and carcinogens. Those are associations and study findings as the chapter reports them, not a recommendation to use tea as a medicine.

Bottled green tea is by far the most popular soft drink in its home market, ahead of carbonated soft drinks, and beverage companies are building value-added products on that popularity. Adding functionality is one way to do it. The chapter names one launch as the most successful product in functional green tea: a healthy green tea that sold $100 million a year in 2002. The figure is a sales figure for a single product in a single year, as the chapter gives it, and it shows the scale a functional tea can reach.

Tea companies in the main tea-growing regions have branched out into functional teas. The chapter gives examples of green tea sold with added nutraceuticals or herbal extracts: green tea with glucosamine, green tea with French pine bark extract, green tea with soy isoflavones, green tea with caffeine and citrus, and a green tea approved as a food for specified health uses for people with diabetes. Collagen, ceramide, CoQ10, and omega 3 are named as the possible next line of green tea fortification. Tea majors have launched ranges of tea and herb blends promising various physical and mental benefits. High doses of ECG and other catechins are reported in the chapter as particularly effective in preventing inflammation and skin cancer when used as a topical lotion: topical ECG reduced the release of inflammatory prostaglandins, which play a role in generating free radicals and promoting tumour growth. That is a topical use in research, not a drink.

Green tea weight loss supplements are described in the chapter as holding both caffeine and EGCG. When the two act together, the chapter says, metabolism rates increase along with the body's 24-hour energy expenditure, and EGCG triggers the release of the hormone noradrenaline, an appetite suppressant. That last claim is contested between the books used in this subject, and it is reported here as the chapter states it, not as settled fact. The book used in Unit 17, Ho et al., Ch. 13, says the opposite: EGCG does not increase the release of NE, it prolongs the action of NE once it is released, by blocking the enzyme COMT. Ahuja et al., Ch. 20, says EGCG triggers the release of noradrenaline. The two books disagree, so neither wording is presented here as the settled mechanism. The human studies behind that claim are covered in Unit 17, with their small samples and careful limits. This unit reports the product claim as the chapter states it, and Unit 17 is where the evidence is weighed.

Sub-standard leaf: 30 to 40% of factory produce

↓

Extract the value: polyphenols, theanine, theaflavins, concentrates

↓

New products: ready-to-drink teas, fortified green teas, tea wines, caffeine-free polyphenols

Figure 19.1: From sub-standard leaf to diversified products. The leaf that cannot be sold as whole tea still holds the compounds that new products are built on. Drawn from Ahuja et al., Science of Tea Technology, Ch. 20.

Try it yourself

Walk a tea shelf and sort what you see into three piles: plain tea, flavoured or blended tea, and tea with an added functional ingredient named on the pack. Count the piles. The chapter's market figures say the third pile is the fastest growing, and your shelf is a quick check of that claim in your own shop.

19.3 Theanine: tea's own amino acid

Tea holds an amino acid that is very specific to tea: theanine, or 5-N-ethyl glutamine. It is made in the tea roots and moved to different parts of the plant, mainly the growing shoots. The chapter lists the activities reported for theanine:

  • Promotes relaxation without causing drowsiness
  • Reduces stress and anxiety
  • Improves learning ability
  • Supports the immune system
  • Lowers cholesterol

Those are reported activities as the chapter lists them. The chapter does not give doses, trial sizes, or human outcomes for theanine, so no strength of effect can be stated here. What it does give is the research programme for making theanine. Work is underway on eco-friendly green technology for extraction of theanine from tea, and on systems for producing theanine in vitro. Three routes for biotransformation are listed: developing hairy root cultures and comparing theanine content in vitro and ex vitro in induced tea roots, callus, and cell suspension cultures; microbial biotransformation using suitable microbes with glutamine or glutamic acid as substrates; and enzymatic biotransformation using the enzyme theanine synthase with glutamine or glutamic acid as substrates.

Theanine is a good example of the diversification logic. A compound that exists in useful amounts only in tea becomes a product in its own right, extracted or made by biotechnology, and sold beyond the cup.

Quick check

Theanine is made in the tea roots and moved mainly to which part of the plant?

Check your answer

The growing shoots. That is where theanine is concentrated after being made in the roots.

19.4 Technologies developed for commercialisation

The chapter lists four technologies developed under a programme on product diversification from tea, for commercialisation:

  1. Ready-to-drink (RTD) tea concentrates, both black and green tea, and formulations for ready-to-drink teas
  2. Tea wines and tea-based herbal health wines made with Himalayan berries
  3. Tea polyphenols free from caffeine, amino acids, and soluble sugars
  4. High-grade bioprocessed theaflavins

Each one solves a different problem. Concentrates let a bottler make a steady drink from variable leaf. Tea wines turn tea into a fermented drink in its own right. Caffeine-free polyphenols let a maker sell the polyphenol fraction to people who want to avoid caffeine. Bioprocessed theaflavins give a high-grade black tea fraction for nutraceutical use.

The chapter also lists the forces driving more tea consumption: scientific validation of healthy attributes covered in major media; an endless diversity of flavours, scents, and accessories; better education by trade associations, tea houses, and cafes about proper preparation and serving; and a broader worldwide tea supply that keeps the cost per serving generally lower than coffee.

The future trends the chapter names are: a proliferation of flavoured and unflavoured green teas and their use in new consumer products; specialty ethnic teas; and nutrients and herbal concentrates added to tea to fortify human health. There is also a trend toward more exotic, fruity flavours, and toward caffeine-free hot beverages, which is boosting the popularity of fruity herbal infusions. Customers at all levels are demanding more accountability on safety, quality, and efficacy of new formulations, which makes the formulator's job more complicated as products hold more ingredients beyond vitamins and minerals. The chapter's summary of the market is that it is becoming increasingly diversified and segmented, as manufacturers try to revive it.

A factory makes 1,000 kg of tea a day and 350 kg is sub-standard. If a concentrate line takes that 350 kg and turns it into ready-to-drink tea, the factory sells 1,000 kg of value instead of 650 kg. That is the arithmetic of diversification: the waste stream becomes a product stream, and the nutraceutical value of the leaf is what makes the stream worth selling.

19.5 How a tea drink is made: cans and PET bottles

A bottled tea drink is not a cup of tea poured into a bottle. It goes through extraction, mixing, filling, and sterilisation, and the sterilisation step is where flavour is most at risk. The chapter gives the manufacturing processes for green, oolong, and black tea drinks, and they share the same skeleton.

First, extraction: tea leaves are extracted, then the extract is filtered, cooled, and centrifuged. Then mixing, filling, and sterilisation, in two main routes. For canned drinks, the extract is mixed with antioxidants and other ingredients, filtered, heated, filled into the can, sealed, and then put through a retorting process, which is heating under pressure, and cooled. For PET bottle drinks, the extract is mixed, filtered, heated, and put through UHT sterilisation, ultrahigh temperature treatment, then cooled, and the pasteurised PET bottle is filled by hot filling or aseptic filling, sealed, and capped.

Black tea drinks with milk add steps: milk and emulsifier are mixed, heated, and homogenised, then mixed with the extract, sugar, and flavour, filtered, heated, and homogenised again before filling and sterilisation. Black tea drinks without milk follow the same route as green tea drinks, with sugar and flavour added at mixing.

The heat numbers matter. Canned drinks are low-acidity drinks, with pH above 4.6, and the retorting process requires heating at 120 °C for more than 4 minutes. PET bottle drinks are sterilised by UHT at 135 to 140 °C for 30 to 60 seconds, followed by hot filling or aseptic filling of the pasteurised bottle. The retort heats longer at a lower temperature. UHT heats briefly at a higher temperature. That difference in heat history is why the two routes change flavour differently, as the next section shows.

Flavour stability has three stages, and the chapter lists them: flavour stability of tea leaves during storage, flavour stability during processing of the tea drink, and flavour stability during display of the tea drink in a hot vending machine or under light. Quality control is the most important criterion and the most difficult problem for all drink makers, and flavour stability is the most important factor in tea drink quality. PET bottles are growing against cans because they are light, easy to carry after opening, and give a sense of security by protecting the contents from the outside.

Quick check

Canned tea drinks are sterilised by retorting at 120 °C for more than 4 minutes. PET bottle drinks are sterilised by UHT at 135 to 140 °C for 30 to 60 seconds. Which route heats for longer?

Check your answer

Retorting. It heats for more than 4 minutes, while UHT heats for only 30 to 60 seconds, though at a higher temperature.

19.6 Green tea drinks: what heat does to flavour

The chapter's green tea work compares three drinks: an unheated control, a retorted drink heated at 121 °C for 10 minutes, and a UHT drink heated at 134 °C for 30 seconds. The control had an odour intensity of 3 and a pH of 6.2. After retorting, the pH fell to 5.6. After UHT, the pH was 6.1. The retorting process brought about a greater flavour change than UHT sterilisation, as expected from the difference in heat history. The green, potato-like, floral, sweet, heavy, and spicy notes all shifted, and the retorted drink's odour profile moved furthest from the control.

The tool used to find which compounds matter is aroma extract dilution analysis, shortened to AEDA. In AEDA, an extract is diluted step by step and smelled at each step, and the flavour dilution (FD) factor records how far a compound can be diluted and still be smelled. A high FD factor means a compound has a strong effect on flavour even in tiny amounts. The chapter's point is practical: components with a higher FD factor after heating than before heating are the ones to watch.

Seven compounds stand out in the heated green tea drink, with their smells and their odour thresholds in water:

Table 19.1: Key flavour compounds in heated green tea drinks, with odour thresholds in water

CompoundSmellOdour threshold in waterPrecursor
2,3-Butanedione (diacetyl)Butter-like10 ppbSugars, or citrate via pyruvate
MethionalPotato-like0.2 ppbL-methionine
PhenylacetaldehydeSweet, honey-like4 ppbL-phenylalanine
LinaloolFloral6 ppbLinalyl glycoside (primeveroside)
GeraniolSweet, honey-like7.5 ppbGeranyl glycosides
β-DamascenoneSweet, honey-like0.002 ppbNeoxanthin via allenic triol glucoside
2-Methoxy-4-vinylphenolDusty, spicy5 ppbFerulic acid

Values from Ho et al., Tea and Tea Products, Ch. 17, Figures 17.4 to 17.6. Thresholds are in water, as the chapter gives them.

Read the thresholds, not just the names. β-Damascenone can be smelled at 0.002 ppb in water. Methional can be smelled at 0.2 ppb. That means a trace of β-damascenone, one hundred times smaller than the trace of methional that can be smelled, still changes the drink. The chapter reports that β-damascenone and methional both rose after the heating process, and theorizes that they have a remarkable effect on tea drink quality. That is the chapter's word, theorized, and it is kept here.

The precursors explain where each compound comes from. 2,3-Butanedione is a characteristic component of fermented dairy products, made from citrate via pyruvate, with a threshold of 10 ppb in water. Under non-enzymatic conditions, alpha-dicarbonyl compounds such as 2,3-butanedione, glyoxal, or methylglyoxal form from monosaccharides such as D-glucose or D-fructose, or disaccharides such as maltose or maltulose, during heating. Adding L-glycine under those conditions raises the yield of the alpha-dicarbonyl compounds. Methional forms from L-methionine, and phenylacetaldehyde from L-phenylalanine, by Strecker degradation. In black tea, the oxidised tea flavanols, the ortho-quinones formed by catechol oxidase during fermentation, drive that degradation, and the yield of aldehydes depends on the catechins present. Green tea manufacture has no fermentation step making those ortho-quinones, so in green tea drinks the aldehydes from amino acids are supposed to form by another route during heating at raised temperature.

Linalool and geraniol come from glycosides, sugar-bound forms stored in the leaf. The linalool precursor, linalyl 6-O-β-D-xylopyranosyl-β-D-glucopyranoside, called β-primeveroside, was isolated from oolong tea leaves. The geraniol precursors, geranyl 6-O-β-D-xylopyranosyl-β-D-glucopyranoside and geranyl 6-O-α-L-arabinopyranosyl-β-D-glucopyranoside, called β-vicianoside, were isolated from oolong and green tea leaves. Glycosides are most plentiful in young leaves and fall as the leaves age. Under acid conditions, they break down into many terpene derivatives, so heating under acid conditions makes a range of off-flavour compounds from the same starting material.

β-Damascenone comes from neoxanthin, through 3-(2,4-dihydroxy-2,6,6-trimethylcyclohexylidene)-1-methylprop-2-enyl-β-D-glucopyranoside, the allenic triol glucoside. During heating under acid conditions, β-damascenone is assumed to form from neoxanthin by that route. 2-Methoxy-4-vinylphenol forms from ferulic acid by thermal fragmentation, a route also reported in cooked corn and cooked asparagus.

Heat during sterilisation
One shared driver, four separate precursor routes below it
Route 1: amino acids
L-methionine and L-phenylalanine → methional (potato-like, 0.2 ppb) and phenylacetaldehyde (sweet, honey-like, 4 ppb), by Strecker degradation
Route 2: sugars
Sugars → 2,3-butanedione (butter-like, 10 ppb)
Route 3: glycosides
Glycosides → linalool (floral, 6 ppb) and geraniol (sweet, honey-like, 7.5 ppb)
Route 4: neoxanthin
Neoxanthin → β-damascenone (sweet, honey-like, 0.002 ppb)

Figure 19.2: How heat turns flavour precursors into off-flavour compounds in green tea drinks. Four separate precursor routes, all driven by the same heat. Each precursor gives a compound with a very low odour threshold, so a trace changes the drink. Drawn from Ho et al., Tea and Tea Products, Ch. 17.

Worked example

Compare two thresholds from Table 19.1. β-Damascenone: 0.002 ppb. 2,3-Butanedione: 10 ppb. Divide 10 by 0.002 and you get 5,000. β-Damascenone can be smelled at a concentration 5,000 times lower than 2,3-butanedione. That is why the chapter watches β-damascenone so closely: a compound the drink holds in traces can still dominate the flavour.

19.7 The eleven potent compounds and the refining step

AEDA found eleven components that matter most in a green tea (sen-cha) drink, each with an FD factor greater than 500 in the unretorted drink: 4-methoxy-2-methyl-2-butanethiol (meaty), (Z)-1,5-octadien-3-one (metallic), 4-mercapto-4-methyl-2-pentanone (meaty), (E,E)-2,4-decadienal (fatty), β-damascone and β-damascenone (sweet, honey-like), (Z)-methyl jasmonate (floral), indole (animal-like), 1-octen-3-one (mushroom-like), methional (potato-like), and (E,Z)-2,6-nonadienal (cucumber-like). Three sulfur compounds in that list were identified from green tea for the first time. Among the eleven, β-damascenone and methional rose after heating, and the chapter theorizes that they have a remarkable effect on tea drink quality.

Refining is the heating step that turns crude green tea, ara-cha, into refined green tea, sen-cha, the green tea usually sold as green tea. The refining conditions in the chapter's comparison are 110 °C for 15 minutes. Refining matters because it builds the roasted and sweet flavour of good green tea, and it makes the flavour more stable under later heating. The compounds that carry that roasted character have very low thresholds: 2-acetyl-1-pyrroline, threshold 0.1 ppb in water, and 2-acetyl-2-thiazoline, threshold 1.3 ppb in water. Their analogues are far weaker: 2-acetylpyrrole has a threshold of 170 ppm in water, and 2-acetylthiazole has a threshold of 10 ppb. The gap between 0.1 ppb and 170 ppm is the gap between a compound that defines a drink and one that barely registers.

The precursors of 2-acetyl-1-pyrroline are 1-pyrroline and 2-oxopropanal, formed from L-proline and 1-deoxyglucosone during heating. For 2-acetyl-2-thiazoline, the precursors are 1-amino-2-mercaptoethane, also called cysteamine, and 2-oxopropanal, and it forms through 2-(1-hydroxyethyl)-4,5-dihydrothiazole as the key intermediate during heating. Other refining products are Maillard reaction products: 2-ethyl-3,5-dimethylpyrazine, threshold 0.04 ppb, 2,3-diethyl-5-methylpyrazine, threshold 0.09 ppb, and 4-hydroxy-2,5-dimethyl-3(2H)-furanone, called furaneol, threshold 60 ppb. L-rhamnose and L-proline are reported as important precursors for furaneol. 3,7-Dimethyl-1,5,7-octatrien-3-ol gives a roasted and sweet flavour in high-grade oolong tea and black tea, and also in green tea, and it forms from 2,6-dimethyl-3,7-octadiene-2,6-diol by dehydration during refining.

The sulfur compounds deserve a note of their own because their thresholds are the lowest in the chapter: 4-mercapto-4-methyl-2-pentanone at 0.0001 ppb in water, and 4-methoxy-2-methyl-2-butanethiol and 1-p-menthen-8-thiol at 0.00002 ppb. A slight change in these compounds can produce a huge change in flavour quality. 4-Mercapto-4-methyl-2-pentanone is reported to form from S-4-(4-methylpentan-2-one)-L-cysteine by cysteine conjugate β-lyase, which was reported to work at 75 °C.

Kamairi-cha is a green tea made differently: the raw leaves are rolled directly on a hot pan, without the steaming step used for sen-cha. The difference in heat history between sen-cha and kamairi-cha is postulated to cause the different FD factors for each flavour component. (Z)-1,5-Octadien-3-one is made from α-linolenic acid by auto-oxidation, as are 1-penten-3-ol and (E,E)-2,4-heptadienal. Eicosapentaenoic acid, found mainly in marine animals and plants, also serves as its precursor. 2-Isobutyl-3-methoxypyrazine, threshold 0.002 ppb, is made from 2-isobutyl-3-hydroxypyrazine by S-adenosyl-L-methionine-dependent O-methyltransferase in grapes, but whether a similar enzymatic O-methylation occurs in green tea drinks is not known, and the chapter says so plainly.

Quick check

2-Acetyl-1-pyrroline can be smelled at 0.1 ppb in water. Its analogue 2-acetylpyrrole needs 170 ppm. Which one matters more for the flavour of refined green tea, and why?

Check your answer

2-Acetyl-1-pyrroline. Its threshold is far lower, so a trace of it is smelled while the analogue needs a much larger amount to register at all.

19.8 Black tea drinks: a different flavour change

Black tea drinks change differently under retorting. In a Darjeeling black tea drink, the intensity of the off-flavour, described as sweet, clove-like, heavy, and putrid, rose after the retorting process at 121 °C for 10 minutes. The control pH was 5.4, and after retorting it was 5.1. The FD factors and the structures of the flavour components that rose by retorting are given in the chapter's figures, and they differ from the green tea pattern. No alcohol formation was noted in black tea drinks, while alcohols were noted in green tea drinks. That is a clear split between the two teas under the same heat treatment.

AEDA of black tea drinks from Darjeeling, Assam, and Uva found potent flavour components with FD factors greater than 210 in the unretorted drink. They include (Z)-4-heptenal (hay-like), (Z)-3-hexenol (green), 2-ethyl-3,5-dimethylpyrazine and 5-ethyl-2,3-dimethylpyrazine (nutty), linalool (floral), 2-acetylpyrazine and 2-acetyl-3-methylpyrazine (roasty), phenylacetaldehyde (sweet, honey-like), 3-methylbutanoic acid (buttery, rancid), methyl salicylate (minty), β-damascenone (sweet, honey-like), hexanoic acid (green, acid), geraniol (floral), guaiacol (smoky, burning, sweet), 2-methoxy-4-vinylphenol (dusty, spicy), and vanillin (vanilla-like). A second AEDA set, from two other black teas, found components with FD factors greater than 64, including 3-hydroxy-4,5-dimethyl-2(5H)-furanone and 4-hydroxy-2,5-dimethyl-3(2H)-furanone (caramel-like), (Z)-1,5-octadien-3-one (metallic), 3-methylnonane-2,4-dione (strawy, fruity), bis(2-methyl-3-furyl)disulfide (boiled, meat-like), and (Z)-4-heptenal (biscuit-like).

The chapter's closing summary ties the two teas together. Maintaining flavour stability is important for manufacturers. Commercially available tea drinks go through many manufacturing steps that differ from homemade tea drinks, and those steps cause flavour deterioration, especially during sterilisation, retorting and UHT. AEDA is an effective method to find the delicate change in flavour components before and after heating, and the components with a higher FD factor after heating have been the focus of study. The non-volatile flavour precursors, glycosides, carbohydrates, fatty acids, and amino acids, are considered to generate off-flavour compounds on heating. The amount of precursor left depends on how the tea leaves were made. In green tea leaves, enzymes are deactivated early in manufacture, so the precursor gives a variety of breakdown products under heating, compared with black tea leaves, where fermentation has already changed the leaf.

Try it yourself

Brew two cups from the same green tea. Heat one cup in a pan at a simmer for ten minutes, and leave the other as brewed. Smell both, and write down three words for each. Then check your words against Table 19.1: potato-like points to methional, sweet and honey-like points to phenylacetaldehyde or β-damascenone, butter-like points to 2,3-butanedione. You have just run a home version of the chapter's sensory test.

19.9 A practical note: keeping a tea drink tasting like tea

A drink maker reading this unit needs a short list of decisions, all drawn from the chapters above.

First, choose the leaf with the flavour in mind, not just the price. The precursor left in the leaf decides what heat will make: glycosides give linalool and geraniol, amino acids give methional and phenylacetaldehyde, sugars give 2,3-butanedione, carotenoids give β-damascenone and β-ionone, ferulic acid gives 2-methoxy-4-vinylphenol. Young leaves hold the most glycosides, and the amount falls as leaves age.

Second, choose the sterilisation route with the heat history in mind. Retorting at 120 °C for more than 4 minutes changes flavour more than UHT at 135 to 140 °C for 30 to 60 seconds, in the chapter's green tea comparison. The retorted drink's pH fell to 5.6 from a control of 6.2, while the UHT drink stayed at 6.1. A maker who must use retorting should expect a bigger flavour shift, and a bigger rise in methional and β-damascenone.

Third, respect thresholds. A compound at 0.002 ppb can matter more than a compound at 10 ppb. Measuring the big peaks on a chromatogram and ignoring the trace sulfur compounds misses the flavour. AEDA exists for that reason: it finds the compounds that smell, not just the compounds that are plentiful.

Fourth, remember that refining is a flavour step, not just a drying step. Refining crude green tea at 110 °C for 15 minutes builds the roasted and sweet notes, 2-acetyl-1-pyrroline and 2-acetyl-2-thiazoline, and gives a more stable drink. Skipping or shortening refining leaves a drink that changes more under sterilisation.

Fifth, keep the product claims honest. A fortified green tea is a food product with added ingredients, and the chapter lists what is added: glucosamine, pine bark extract, soy isoflavones, caffeine and citrus, collagen, ceramide, CoQ10, omega 3. The claim on the pack should name the ingredient and its amount, and should not borrow the language of the laboratory studies in Units 17 and 18, which tested pure compounds in cells and animals, not the bottled drink.

Quick check

A maker switches a green tea drink from retorting to UHT sterilisation. Using this unit, name two flavour changes you would expect.

Check your answer

Any two of these: a smaller overall flavour change, because UHT heats for a shorter time; a smaller rise in methional and β-damascenone; a pH closer to the control, 6.1 against 5.6 after retorting.

Key ideas to carry forward

  • Tea factories discard 30 to 40% of produce as sub-standard leaf, and that leaf still holds the compounds that diversified products are built on.
  • Functional foods are foods or ingredients that give a health benefit beyond basic nutrition. Tea is a strong base for them because it holds polyphenols and theanine.
  • The chapter's commercial technologies are ready-to-drink tea concentrates, tea wines and herbal health wines, caffeine-free tea polyphenols, and high-grade bioprocessed theaflavins.
  • Theanine, 5-N-ethyl glutamine, is made in tea roots and moved to the growing shoots. The chapter lists relaxation without drowsiness, less stress and anxiety, better learning, immune support, and lower cholesterol as its reported activities, without doses or trial outcomes.
  • Tea drinks are made by extraction, filtration, cooling, and centrifugation, then mixing, filling, and sterilisation. Cans are retorted at 120 °C for more than 4 minutes. PET bottles are UHT sterilised at 135 to 140 °C for 30 to 60 seconds.
  • Retorting changes green tea flavour more than UHT. In the chapter's test, the retorted drink's pH fell to 5.6 from a control of 6.2, while the UHT drink stayed at 6.1.
  • Off-flavour compounds come from precursors in the leaf: methional from L-methionine, phenylacetaldehyde from L-phenylalanine, 2,3-butanedione from sugars, linalool and geraniol from glycosides, β-damascenone from neoxanthin, and 2-methoxy-4-vinylphenol from ferulic acid.
  • Odour thresholds decide what matters: β-damascenone at 0.002 ppb and methional at 0.2 ppb can dominate a drink in traces, while 2,3-butanedione needs 10 ppb to be smelled.
  • Refining crude green tea at 110 °C for 15 minutes builds roasted notes, 2-acetyl-1-pyrroline at 0.1 ppb and 2-acetyl-2-thiazoline at 1.3 ppb, and gives a more stable drink.
  • Black tea drinks change differently under retorting: sweet, clove-like, heavy, and putrid notes rise, and no alcohol formation was noted, unlike green tea drinks.

Glossary

Diversified tea product

A new product made from tea or its compounds, beyond plain leaf tea. Food tech example: ready-to-drink tea concentrates and caffeine-free tea polyphenols are diversified products made from leaf that could not be sold as whole tea.

Functional food

A food or food ingredient that may give a health benefit beyond the traditional nutrients it contains. Food tech example: a green tea with added glucosamine is sold as a functional tea.

Nutraceutical

A food or food component with a health benefit beyond basic nutrition, often sold for that benefit. Food tech example: tea polyphenols sold free of caffeine are a nutraceutical ingredient.

Theanine

5-N-ethyl glutamine, an amino acid very specific to tea, made in the roots and moved to the growing shoots. Food tech example: theanine is extracted from tea for use in functional products, and its reported activities include relaxation without drowsiness.

Theaflavin

A polyphenol formed during fermentation of black tea, including TF-1, TF-2a, TF-2b, and TF-3. Food tech example: high-grade bioprocessed theaflavins are one of the chapter's commercial technologies.

Ready-to-drink (RTD) tea

Tea sold as a finished bottled or canned drink, rather than as leaf to brew. Food tech example: RTD tea concentrates let a bottler make a steady drink from variable leaf.

Retorting

Sterilising a sealed can by heating it under pressure. Food tech example: canned tea drinks are retorted at 120 °C for more than 4 minutes, and the long heat changes flavour more than UHT does.

UHT sterilisation

Ultrahigh temperature sterilisation, heating a liquid briefly at very high temperature. Food tech example: PET bottle tea drinks are UHT sterilised at 135 to 140 °C for 30 to 60 seconds.

Aseptic filling

Filling a sterilised container with a sterilised drink in a sterile environment, so the drink is not heated in the pack. Food tech example: PET bottle tea drinks are filled by hot filling or aseptic filling after UHT sterilisation.

Flavour stability

How well a drink keeps its intended flavour through storage, processing, and display. Food tech example: flavour stability is the most important factor in tea drink quality, and sterilisation is its biggest threat.

AEDA

Aroma extract dilution analysis, a method that dilutes a flavour extract step by step and smells each dilution to find which compounds matter. Food tech example: AEDA found eleven potent compounds in green tea drink, each with a flavour dilution factor above 500.

FD factor

Flavour dilution factor, how far an extract can be diluted before a compound can no longer be smelled. Food tech example: a compound with an FD factor above 500 still smells after the extract is diluted 500 times.

Odour threshold

The lowest concentration at which a compound can be smelled, usually given in water. Food tech example: β-damascenone has a threshold of 0.002 ppb in water, so a trace of it changes a tea drink.

Strecker degradation

A reaction in which amino acids break down under heat to give aldehydes. Food tech example: L-methionine gives methional and L-phenylalanine gives phenylacetaldehyde by Strecker degradation during heating.

Glycoside

A compound in which a flavour molecule is bound to a sugar, storing it in the leaf until heat or acid frees it. Food tech example: linalool and geraniol are stored as glycosides in tea leaves and freed during heating.

Maillard reaction

Browning reactions between sugars and amino acids under heat, which build roasted and caramel notes. Food tech example: pyrazines and furaneol in refined green tea are Maillard reaction products.

Sen-cha

Refined green tea, the green tea usually sold as green tea, made by heating crude green tea. Food tech example: refining ara-cha at 110 °C for 15 minutes gives sen-cha and its roasted, sweet flavour.

Ara-cha

Crude green tea, before the refining heating step. Food tech example: ara-cha is refined into sen-cha, and the refining step makes the flavour more stable.

Kamairi-cha

A green tea made by rolling raw leaves directly on a hot pan, without steaming. Food tech example: kamairi-cha has a different heat history from sen-cha, so its flavour compounds differ.

References

  • Ahuja, D. K., Gulati, A., et al. Science of Tea Technology. Chapter 20, Diversified Tea Products, by H. P. Singh, Brajinder Singh, and Mahesh Gupta.
  • Ho, C. T., Lin, J. K., and Shahidi, F., editors. Tea and Tea Products: Chemistry and Health-Promoting Properties. Chapter 17, Flavor Stability of Tea Drinks, by Hideki Masuda.

Unit 20: Spice Science: What a Spice Is, How It Is Classified and Graded

Note

Unit 20: Spice Science: What a Spice Is, How It Is Classified and Graded

Paprika and saffron both colour a dish yellow-red. Swap them and the dish still looks right, but the taste is wrong, the colour washes out of the wrong food, and one costs a hundred times more than the other. Why do two spices that look like twins behave like strangers? This unit starts with a harder question: what counts as a spice at all?

20.1 What a spice is

The definition of a spice differs according to the country or region where it is used. Some definitions sort spices by where they grow, some by whether they are dry or wet, and some by their history, meaning when people started using them as spices. None of those ways of sorting is always right when you look at what a spice does in food and what it is made of.

The working definition the book settles on is short: a spice is the dry part of a plant, such as a root, a leaf or a seed, which gives food a certain flavour and a pungent stimulus. A spice is a part of a plant, not the whole plant, and it works by flavour and by a stimulating bite, not by feeding you.

Unit 6 covers the basics of what a spice is, how the six main spices are produced and dried, and how their flavour is kept through processing. This unit goes deeper on the other side of the subject: the definition itself, the line between spice and herb, the botanical classification, and how spices are specified and graded.

20.2 Spice or herb: where the line sits, and where it does not

The word herb comes from the Latin herba, meaning a medicinal plant. In the narrow sense, an herb is a plant that does not last: it withers after blooming, and its stems never become woody. Perennial herbs are used for dyeing and gardening as well as for medicine and food, and in Europe herbs have long been used for medical purposes.

No spice definition draws a clean line between a spice and an herb. Some edible herbs belong in the spice category. Even some herbs that carry poisonous components can count as spices if the poisonous part can be neutralised by heating and other cooking steps. The difference is one of viewpoint, not of botany. An herb is classified by what kind of plant it is, a perennial plant. A spice is classified by how it is used in cooking. And a spice should be edible.

The book draws its own line for its own pages, and this unit follows it. All edible leaves that people usually call herbs go in the spice category, and the book calls them "spice leaves" or "edible herbs". Herbs used for medicine and not for cooking stay in the herb category. So basil on a plate is a spice leaf, and a medicinal herb that never enters a kitchen is not. The broad definition the book gives is worth learning by heart: a spice is a compound that has a pungent flavour or colouring activity, or one that increases appetite or enhances digestion. It comes from the seeds, berries, buds, leaves, bark and roots of plants growing mainly in the tropical, the subtropical and the temperate zones.

Quick check

What is the difference between the meaning of herb and the meaning of spice?

Check your answer

Herb is a botanical idea: a plant, often perennial, used for medicine, dyeing or gardening. Spice is a cooking idea: a part of a plant used in food for flavour, pungency or colour.

20.3 The plant parts that become spices

Only a part of the plant does the work. Leaves give us basil, sage and thyme. Bark gives us cinnamon. Buds give us clove. Fruits give us allspice, nutmeg, mace and Sansho, the Japanese pepper. Roots and rhizomes give us ginger and turmeric. Seeds give us mustard, cumin and coriander.

The book sums the whole idea of a spice up in three points: many plants used as spices grow in the tropical, the subtropical or the temperate zone; not the whole plant but a part of it is effective as a spice; and the effect of a spice is characterised more or less by its stimulating flavour.

Looked at from the botanical side, the book's Table 1.1 shows that many kinds of Labiatae, Umbelliferae and Zingiberaceae plants count as spices. The table itself, with its full list of species, is printed as an image in the book and its rows cannot be read from the copy on file, so only the three families named in the text are stated here, and no other values from that table are reproduced.

20.4 Spice seasonings and condiments

Spices are used to season foods either alone or as part of a complex seasoning that also holds sugar, salt or other ingredients. Two names cover those mixtures, and the difference between them is when they meet the food. A spice seasoning is a mixture holding one or more spices or spice extracts that improves the flavour of the original food, and it is added during processing by the food industry or during cooking at home. A condiment is the more common word, and it means a mixture added to food when it is eaten, not during cooking. It is made of one or more spices and spice extracts, and it improves the flavour of the food on the plate. Pickles, most often eaten as a side dish, usually sit in this category.

Quick check

What separates a spice seasoning from a condiment?

Check your answer

Timing. A spice seasoning is added during processing or cooking. A condiment is added when the food is eaten, at the table.

20.5 The four basic effects, and the effects behind them

Spices do four basic things in cooking: they give flavour, they give pungency, they give colour, and they deodorise or mask smells. The deodorising and masking job can overlap with the other three, but it is often wanted on its own. Each spice does at least one of these four jobs.

Behind those four sit other effects. Spices have antioxidant, antimicrobial, pharmaceutical and nutritional properties. And beyond those direct effects, there are complex or secondary effects that come from using spices in cooking: salt reduction, sugar reduction, and improved texture for certain foods.

Cinnamon in bread is the book's example of a secondary effect. Cinnamon has strong inhibitory activity against most microbes. In bread making, yeast provides the fermentation that makes bread rise, but too much fermentation cuts the yield of the finished product. Cinnamon lends its typical flavour to the bread, its direct effect, and it also suppresses yeast activity through its antimicrobial action, holding fermentation back and improving the texture and the yield. Flavour is the direct effect; texture is the secondary one.

Two spices that seem interchangeable show why the four effects must be understood one by one. Saffron, the most expensive spice, is essential in pilaf and bouillabaisse. Its yellowish-red colour comes from a component called crocin, and crocin does not dissolve in oil, so saffron cannot colour oil-based or fat-based foods. Paprika also gives a yellowish-red colour, but its colouring component is β-carotene, which dissolves in oil and not in water, so paprika is used for colouring oil-containing foods such as margarine, salad dressing and beef stew. Same colour on the plate, opposite solubility, and the two cannot be swapped.

20.6 Flavour: the volatile oil and the terpene family

A spice is a plant made of fibre, sugar, fat, protein, ash, gum, essential oils and other components. Among these, it is the volatile essential oil that gives a spice its particular flavour. An essential oil is not one chemical compound but a number of compounds, and spice aromas and flavours differ subtly with the amount of essential oil and the share of each compound in it. That is why two samples of the same spice can smell delicately different, and why two spices in the same botanical category can differ in flavour with where they grow and when they are harvested, even when they look alike.

The major chemical components of spice essential oils are terpene compounds, made of carbon, hydrogen and oxygen. Compounds with 10 carbons are called monoterpenes, with 15 carbons sesquiterpenes, and with 20 carbons diterpenes. A monoterpene generally has a strong aroma and is very volatile, and monoterpenes are used as ingredients in perfumes and essences. Green plants give off oxygen during carbon assimilation and also release terpene compounds, which are said to strengthen the anti-cold property of spices and to work as insect repellents.

Essential oils sit in special tissues and cells such as oil cells, oil glands and glandular hairs. A strong aroma results, and the oil is released, when those tissues and cells collapse. The book shows this with a photomicrograph of a peppermint leaf: an oily membrane, oval in shape, holds the essential oil, and the membrane breaks to release its aromatic compounds when the mint leaves are crushed or sliced.

How a spice keeps its flavour locked away, and how cooking lets it out

Oil cells, oil glands and glandular hairs
Essential oil sits inside these tiny structures in the leaf

↓

Crush, slice or tear the leaf
The oily membrane breaks open

↓

Essential oil escapes
Terpenes reach the nose as aroma

↓

Heat drives the oil off
Flavour is lost fast if heating goes on too long

Figure 20.1: How spice flavour is stored and released. Essential oil sits in oil cells, glands and glandular hairs; crushing breaks the membrane and frees the aroma, and heat drives it off. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 1.

Some spice compounds stand up to heat and others do not. Flavour compounds are somewhat volatile even at room temperature, and they are certainly volatilised swiftly on heating, so heat and heating time should be limited as far as possible to keep a spice's fresh flavour. The same rule works in reverse: more heat will reduce a spice's flavour, if that is what you want.

Fresh basil loses its aroma when cooked at high temperature, for example when fried in oil. Sesame and poppy seeds, which do not have a strong aroma, become much more flavourful when roasted. Sesame seeds hold little volatile essential oil but much nonvolatile oil, from 45 to 55%, made mainly of oleic acid, stearic acid and palmitic acid, and this oil has a slight almond-like aroma. Commercial sesame oil is made by pressing roasted sesame seeds, and the major flavour compounds in roasted seeds are aldehyde, acetyl pyrazine, hexanal and furfural. Red pepper holds almost no volatile essential oil but many nonvolatile compounds, so it is roasted to reach its typical flavour, and roasted red pepper is a major ingredient in Shichimi, the Japanese traditional blended spice mixture. Mustard seed is also roasted for use in Indian chutney or curry.

Fresh leaves often carry a grassy smell that many people dislike. Drying can fix that: the flavour and aroma of sage and thyme are actually intensified by drying, which removes the compounds causing the grassy smell and much of the moisture, leaving a stronger flavour through a concentration of low-volatility compounds.

Many spices have a bitter, astringent or harsh taste, especially fresh. Alkaloids, their glycosides, and organic or inorganic salts are known to have a bitter taste. An inorganic salt content above 1.5% in spice leaves will give a bitter taste, and the bitterness is stronger if potassium is a major part of that inorganic salt. Frying in oil is one cooking technique used to break down harshness compounds in spices, especially fresh leaves. Boiling sometimes makes spices tastier by removing harshness and grassy flavour as those compounds dissolve in the boiling water.

Try it yourself: take two mint leaves. Smell one whole, then crush the second between your fingers and smell it again. The crushed leaf smells far stronger: you have just broken the oil membranes shown in Figure 20.1.

Quick check

Why does drying make sage and thyme taste stronger instead of weaker?

Check your answer

Drying removes the compounds that cause the grassy smell and much of the moisture, so the low-volatility flavour compounds that stay behind are concentrated and taste stronger.

20.7 Pungency: hot, sharp, and the chemistry that makes them

Taste the essential oil of any spice and you feel a pungent sensation, stronger or weaker. That pungency is directly tied to the concentration of the essential oil: most pungent sensations come from the essential oil stimulating a taste bud on the tongue, and the more concentrated the oil, the stronger the sensation. Some spice compounds are felt as pungent even at very low concentrations.

The pungent compounds in spices fall into four groups: acid-amide compounds, carbonyl compounds, thioether compounds and isothiocyanate compounds. Pungent sensations divide into two kinds. The "hot" sensation spreads into the mouth, and it comes from spices holding acid-amide or carbonyl compounds. The "sharp" sensation stimulates the mucous membrane of the nose and the mouth, and it comes from spices holding thioether or isothiocyanate compounds. Most compounds giving the "sharp" sensation are volatile, while those giving the "hot" sensation are generally nonvolatile. That means most "sharp" pungent compounds are the same as the spice's flavour compounds, while its "hot" pungent compounds are usually different from its flavour compounds.

The four groups of pungent compounds and the two kinds of pungent sensation

Pungent compound groupSensationVolatile or nonvolatile
Acid-amide compoundsHot, spreads in the mouthGenerally nonvolatile
Carbonyl compoundsHot, spreads in the mouthGenerally nonvolatile
Thioether compoundsSharp, stimulates nose and mouthMostly volatile
Isothiocyanate compoundsSharp, stimulates nose and mouthMostly volatile

Table 20.1: The four groups of pungent compounds. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 1.

Some spices hold their flavour and pungency as a glycoside, and an enzyme held in the same tissue breaks the glycoside down to make the flavour and pungent compounds. The Allium plants, garlic, onion and leek, work this way. Their typical flavour comes from alkyl disulfide compounds, made from alkyl cysteine-sulfide by enzyme action. In garlic, the precursor glycoside is called alliin. Alliin itself has no smell or pungency, but it easily produces allicin. Cutting the vegetable collapses its cells, so an enzyme called alliinase breaks alliin down to allicin. Allicin and its analogues are not stable in air: they slowly break down into disulfide or trisulfide compounds, which are the major flavour compounds in Allium plants.

Each Allium plant can be told apart by its main sulfide compound. Three main R-groups matter: the methyl group (CH₃-), the propyl group (CH₃CH₂CH₂-) and the allyl group (CH₂=CH-CH₂-). Compounds with a propenyl group (CH₃CH=CH-) are also found in onion. Pickled scallion holds mainly methyl groups, while propyl groups predominate in leek and onion, and the chief flavour compounds of garlic hold allyl groups. Plants holding mainly methyl groups are generally thought to have a strong smell, while ones with propyl groups have a relatively moderate smell. Cutting an onion makes the eyes water: the volatile compound behind that is made by the breakdown of (+)-S-(1-propenyl)-L-cysteine-sulfoxide.

A cross section of an onion shows that its stalk holds twice as many flavour precursors as its leaves, and the inside of the leaves holds more than the outside. That fact matters in the kitchen: the outside, which is less pungent, can be used as a salad ingredient, while the inside is better for stewing. Chopping an onion fine brings flavour precursors into contact with the enzyme in the onion tissue, so fine chopping strengthens its typical pungency compared with simple slicing. The precursors themselves have no flavour but do have a sweet taste. Heating easily deactivates the enzyme in onion tissue, so heating stops the breakdown of the precursor and reduces the odours of garlic and onion. Onion pungent compounds joined with protein will neutralise meat smell, so it is most effective to cook onion together with meat rather than to add it later.

Mustard, horseradish and radish belong to the isothiocyanate group, and their pungent compounds sit in the plant tissue as glycosides. The glycoside itself has no pungency, but when an enzyme called myrosinase breaks it down it produces an isothiocyanate compound that shows pungency. It is best to grate these finely so the enzyme and the glycosides mix well. It generally takes some time to break down the glycosides of mustard or horseradish, so a while passes after grating before the pungent sensation develops. Acid, salt and heat will reduce the pungency of these spices because they hold back the breakdown of the glycoside. Rice vinegar is sometimes added to grated horseradish, mustard and radish in Japanese foods to make them less pungent. Onion, leek and radish gain a sweet taste when boiled, because pungent sulfide compounds are deoxidised to make sweet mercaptan compounds.

Ginger, red pepper and black and white pepper count among the "hot" pungent spices, and their pungent compounds are relatively heat stable compared with those in onion or mustard, so they can be used in cooking that needs heat. Ginger, whose pungency is due to volatile carbonyl compounds, is a little less stable than red and black or white pepper, whose pungency is due to nonvolatile acid amides. Red pepper, the most stable against heat, has little aroma because it holds little essential oil. Mustard and horseradish are not suited to cooking that needs heat, because their pungent compounds usually disappear and sometimes change into bitter compounds. Sansho, the Japanese pepper, has a "numbing" type of pungency and is popular in Japan and Korea, where it is often used for its aroma rather than its pungency, expecting its deodorising effect on fish.

Quick check

Why does a finely chopped onion taste more pungent than a thickly sliced one?

Check your answer

Fine chopping breaks more cells, so more flavour precursor meets the enzyme in the onion tissue, and more pungent compound is made.

20.8 Colour: the pigments and how to use them

Many colouring compounds in spices are used in food processing as natural colourants, and using them well means knowing each colouring component and its tint. The colouring components of spice include carotenoids, flavonoids and chlorophyll.

Chlorophyll gives plant leaves their green colour. A bright green tint can be had by using green leaves as they are, but cooking time must be kept short, because chlorophyll will turn brown on heating. Green herbs are often added at the end of cooking to keep the green colour, which also stops flavour loss from heating.

Carotenoids are oil-soluble and so suit use as a colouring agent in salad dressing, and they can be used in cooking because of their heat-resistant nature. Many types of carotenoids are found in paprika, for example β-carotene, capsanthin, capsorbin and lutein.

Flavonoids and anthocyanins are water-soluble. Anthocyanins are red, blue, purple and pink, while flavonoids are yellow and yellowish-white. Anthocyanin is also soluble in acids. When perilla or hibiscus holding anthocyanin is put in vinegar, the solution becomes red. Anthocyanin pigments become stable and give a bright colour when they bind with aluminium or iron, which is why a nail is often put in pickled vegetables to hold their bright colour.

The book sums up the use of spice colour in seven points: each spice holds a water-soluble or an oil-soluble colour component, so choose the right type to suit the cooking purpose; some spice colour components change their colour tone with the pH of the solution, so you may change the pH of the solution the spice is dissolved in; some spice colour components are stabilised when used with metals; most spice colour components change their colour tone on heating, so heating conditions must be carefully controlled; for spices whose colour component does not seep out during cooking, you can use the spice directly when cooking; you can enjoy the bright green colour of fresh leaves by straining them to add to sauces; and you can also enjoy the colour tone of dried spices such as parsley and green pepper when these are added to sauce or soup.

The easiest way to colour a food with a spice is without dissolving it: the spice can be used as it is, or after crushing into a paste. You can enjoy the green colour of parsley just by sprinkling it on a dish, but it can also be boiled and used to colour a soup or sauce, though then its colour becomes a dull green. The colour component in hibiscus belongs to the anthocyanin family: it appears red when protonated, meaning acidified, but it becomes colourless when the pH is raised. The yellow colour component in chestnut and sweet potato belongs to the flavonoid family, and it becomes bright when boiled with alum, because the aluminium in alum bonds to the component and stabilises the yellow colour of the flavonoid.

Quick check

Why is paprika used to colour margarine while saffron is not?

Check your answer

Paprika's colouring component, β-carotene, dissolves in oil, so it colours oil-based foods. Saffron's crocin does not dissolve in oil, so it cannot colour oil-based or fat-based foods.

20.9 Carotenoids: the colour family that feeds animals their colour

Carotene was first extracted from carrot by Wackenroder in Germany in 1831, and the compound was named carotene after the carrot. Carotenoids include hydrocarbon compounds called carotene and their oxygenated derivatives called xanthophylls. Some microbes and plants can make carotenoids, but animals cannot, so a carotenoid found in an animal can be assumed to come from carotenoids in the animal's feed. The red colour of a goldfish and of a varicoloured carp is due to the carotenoid group, and flamingo, shrimp and sea bream also take their colour from carotenoids they eat.

More than 1,000 kinds of carotenoids have been found. Among them are α-, β- and γ-carotene, lycopene, zeaxanthin, lutein, capsanthin, cantaxanthin and astaxanthin. Red pepper and paprika hold many carotenoid compounds. Their main ones are said to be capsanthin, lutein, zeaxanthin, α-carotene and cryptoxanthin, though some reports say 20 to 30 kinds of carotenoid compounds are present. The kind and amount of carotenoid varies with the kind of red pepper and where it is grown.

Carotene is found chiefly in carrot as provitamin A, which is converted to vitamin A in the human body. Too much vitamin A is thought to be harmful, but carotenoids have no harmful effect, even when they build up in the body. Paprika holds a large amount of provitamin A. β-carotene, the most common provitamin A, is widely used in medicines, cosmetics, food and feed, and in food it has been used as a colourant in cheese, butter, margarine, ice cream and some confectionery.

Fish and seafood can be given colour by feeding them carotenoid-holding spices. In many countries of Europe, North America and South America, astaxanthin, which can be made by the oxidation of carotene, is added to livestock feed for Pacific salmon, silver salmon, rainbow trout and Arctic char to improve their colour. The same technique is applied to "black tiger" prawns harvested in Southeast Asia. Black sea bream weighing 800 g to 1 kg become reddish in 2 to 4 months on feed holding astaxanthin, fish meal and flour. A wild sea bream becomes red when it eats shrimp and small fish holding red colourant, but hatchery fish fed on feed with almost no colourant slowly turn black, and they tend to get a kind of sunburn near the head and lose market value because they swim in shallow water and are exposed to the sun. Black hatchery fish can regain a bright red colour when fed a carotenoid such as astaxanthin, and the change reaches the inside meat as well as the surface. It is also reported that the hatching rate of a reddish sea bream is higher than that of a black sea bream, and carotene is confirmed to raise the breeding rate of the cow and is listed as an additive for livestock feed.

20.10 Natural colour extracts from spices

Recent technical advances have made it possible to extract colour components from spices and use them as natural colourants. The book lists eight of them, and each one behaves differently.

Spice extracts used as natural colourants, with their colouring components and behaviour

Spice extractColouring componentWhat the book says about it
Turmeric (Curcuma longa L.)Curcumin (C₂₁H₂₀O₆)Yellow at acid to neutral pH, reddish-brown in alkaline pH; melting point 183 °C; soluble in ethanol, propylene glycol and acetic acid, not in water or ether; stains protein well; fairly stable to heat, not to light; turns dark with metal, especially iron; curcumin is about 3% of the spice
Paprika and red pepper (Capsicum annuum var. cuneatum)CapsanthinRed to orange; oil-soluble; soluble in alcohol and oil, not in water; colour does not change with pH; fairly stable to heat; slightly unstable to light, improved by adding an antioxidant
Saffron (Crocus sativus L.) and gardeniaCrocin and crocetinWater-soluble, unusual for the natural world; very beautiful yellow; not stable against acid or light, so cannot be used in beverages; fades with iron and copper ions
Red perilla (Perilla frutescens)Sisonin, peraninRed at acid pH, purple at neutral pH, blue at alkaline pH; the purple and blue fade and turn brown quickly; resistant to heat and light at acid pH
Onion (Allium cepa L.)QuercetinBrown to reddish-brown; soluble in water and in alkaline solution, not in acid solution, not in oil; stable against heat and light; strong dyeing effect
Hibiscus (Hibiscus sabdariffa L.)AnthocyaninsBright red at acid pH, purple at neutral and alkaline pH; not stable against heat and light; used as a pink colour in soft drinks and alcoholic beverages
Annatto (Bixa orellana L.)Bixin, norbixinBixin: melting point 198 °C, soluble in oil and alkali solutions, not in water, slightly unstable to light. Norbixin: melting point over 300 °C, soluble in acetic acid and pyridine, not in oil, resistant to heat, slightly resistant to light
Others: Western madder, Japanese indigo plant, litmus moss, cacao, European grape, green plant leaves, red beet, safflowerAlizarin, purpurin, indigo, orchil, betanin, safflower yellow, carthaminAlizarin is an oil-soluble orange colourant from madder root; purpurin is reddish-purple and comes from stored root, not fresh root; indigo is a blue colourant used outside food processing; orchil is bluish-purple for beverages; betanin from beet root is a water-soluble red colourant; safflower yellow colours noodles and fruit juice, and its oxidised red form, "beni", was once used as a cloth dye

Table 20.2: Spice extracts used as natural colourants. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 1.

Spice colourants are not very stable compared with artificial colourants. Oxygen and ultraviolet light especially tend to fade their colour or oxidise them. Paprika or red pepper, in which carotenoid compounds give the colour, are said to suffer oxidation especially easily, so when these colourants are used in processed foods it is essential to pack the products in light-impermeable materials or in packages in which inactive nitrogen is put in place of air. Heat can also sometimes damage their colour. Three things matter in choosing a colourant for food: how attractive it will make the food, how natural the food colour will be, and how stable the food colour will be.

Antioxidants can sometimes be used to stabilise colour. Carotenoid compounds become more stable when vitamin C and enzyme-treated rutin are added, because they work as antioxidants for carotenoids. Lease and Lease tested other antioxidants, BHA, BHT and α-tocopherol, on keeping the red colour of ground paprika, and found that BHA was effective at a concentration of 0.1%, while BHT and α-tocopherol were not effective even at a concentration of 0.5%.

Quick check

A food maker wants to colour a clear drink yellow with a spice extract. Why is saffron a poor choice?

Check your answer

Crocin and crocetin, the colouring components of saffron, are not stable against acid or light, so they cannot be used in beverages.

20.11 Why spices need specifications

A spice can be defined as the dried aromatic parts of natural plants, and its colour and constitution may vary with the year of harvest, the place of harvest, and other factors. The quality of processed spices can also vary with differences in the separation and milling processes used. For those reasons it has been thought necessary to set quality standards or specifications for spices.

There are no unified standards or specifications worldwide. Nations that export spices often have their own quality standards to keep their own reputations, while nations that import and consume spices set specifications for consumer safety. The quality standards most used as international guidelines are those of the American Spice Trade Association (ASTA) and the United States Federal Specifications for ground and whole spices and spice blends. The International Organization for Standardization set its own standards in 1969 for the quality of various spices, and there are also specifications for spices imported into and consumed in the United Kingdom and Canada.

Why one spice lot is not like the next, and what a specification does about it

Harvest year and place change the spice
Colour and constitution vary with year and place of harvest

↓

Separation and milling change it again
Processed spice quality varies with the process used

↓

No single worldwide standard exists
Exporting nations guard their reputation; importing nations guard consumer safety

↓

Specifications set limits
Extraneous matter, moisture, ash, volatile oil and microbes are measured against a written limit

Figure 20.2: Why spice specifications exist. Quality varies with harvest and processing, and no single worldwide standard covers it, so exporting and importing nations each set their own limits. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 2.

20.12 Specifications in spice-consuming nations

The ASTA specifications cover unprocessed spices imported into the United States, including edible herbs. They were published first in 1969, with several revisions through 1975, and they are now used in many other nations, including Japan, as a kind of world standard. They deal specifically with the cleanliness of the spice product, setting limits on extraneous matter such as insects, insect excrement, stones, stems and sticks. Imported spices that do not meet these specifications should be reconditioned at the port of entry, while domestic spices can be reconditioned before they are processed into a consumable product.

The ASTA rules work in numbers. Amounts of rodent and other animal excrement are set by weight. Spices in which a certain number of insects are found alive must be reconditioned, for example by fumigation. Mites and psocids, whose numbers are confirmed by a flotation test that makes these insects float in a heated organic solvent, must not go above the standard. Spices are not acceptable if mould is present above the stated values, given as percent by weight of the total number of subsamples, or if the total sample quantity goes above the stated values for insect-bored or otherwise defiled seeds, leaves or roots. Light berry content of black pepper, though not counted as extraneous matter, should not go above 4% by weight. The standard for extraneous matter is set for 33 kinds of spice on the basis of past results.

Sampling is also set down. The number of samples drawn must equal the square root of the containers in the lot, and the sample size should be one pound for high-density items and a well-filled two-pound paper bag for low-density items. Testing methods such as sifting and hand-picking for extraneous matter and light berry determination are described for black and white pepper. For nutmeg there are detailed guidelines for inspecting moulds and insects, which may get in through broken surfaces and cause mould contamination.

The United States Federal Specifications go further than ASTA, which regulates the cleanliness of unprocessed spices: they set quality standards for both ground and whole spices. Total ash, acid-insoluble ash, volatile oil, moisture and other characteristics are standardised for major spices such as allspice, cardamom, coriander, cumin, cinnamon, black pepper, nutmeg, mustard and ginger. They also set a standard for a "defective lot" by the amount of stems or other contaminants found in certain spices. For example, bay leaves or basil leaves holding 3% or more stem material count as defective under these specifications.

The International Organization for Standardization set specifications in 1969 for 68 items, using extraneous matter, moisture content, total ash and other chemical characteristics. In the United Kingdom, the British Standard was set in the early 1970s on the basis of ISO specifications, setting the amount of extraneous matter and setting limits mainly for total ash, acid-insoluble ash, volatile oil and moisture for several major whole spices, and also setting a limit for crude fibre for some ground spices. Canada has a quality standard setting total ash and acid-insoluble ash for more than 20 kinds of spices, under the Canadian Specification for Spices, Dressings and Seasonings. The amounts of salt and anticaking agents that can be used for spice seasonings such as celery salt and garlic salt are also regulated in this specification: the amount of salt used in celery salt, onion salt and garlic salt is limited to less than 75%, and anticaking agents in onion salt are limited to 2%. Japan, one of the biggest spice-consuming nations, has no quality specifications for spices, but import of spices treated with ethylene oxide or contaminated with a certain amount of aflatoxin is prohibited under the Japanese Sanitation Law.

Quick check

A lot of black pepper holds 5% light berries by weight. Does it meet the ASTA limit?

Check your answer

No. Light berry content of black pepper should not exceed 4% by weight under the ASTA specifications.

20.13 Grading in spice-exporting nations

Most spice-exporting nations have their own exporting specifications, which also regulate the testing methods. In India, the Indian Standards Institution states quality standards for 36 kinds of both unprocessed and processed spices, from major exported items such as celery, coriander, cumin, fennel, fenugreek and turmeric to items particular to India such as ajowan seed and kokum. These specifications mainly regulate the maximum moisture content, and they include sampling and testing methods. The Government of India has also set standards for almost all exported spice items and graded each item using "Agmark" grades, under the Directorate of Marketing and Inspection, which administers quality control and preshipment inspection.

Grade specifications are set for age-old, familiar trade names. Alleppey Finger turmeric, Cochin ginger, Malabar pepper and Sannam chilies each have individual specifications that set them apart from other turmeric, ginger, pepper and chilies. Each specification states limits for moisture, volatile oil, total ash, acid-insoluble ash and starch, in addition to standards for extraneous matter, so spices are inspected for each chemical and physical quality before export. Black pepper, one of the most important import items, is classified into more than 10 grades, depending on the share of light berries, the harvest place (Malabar or others), the moisture content, and other factors. Tellicherry black pepper in particular is classified by size. Curry powder, a mixture of spices, is graded by the amount of spice or salt it holds: curry powder holding 85% or more spice and less than 5% salt is graded "standard", and one with 70% or more spice and 10% or less salt is graded "general".

In Malaysia, the Pepper Marketing Board set grade specifications for Sarawak pepper, which accounts for more than 90% of the total pepper production of Malaysia. The grade of black pepper is fixed by the amount of light berries present, extraneous matter, moisture and other characteristics. White pepper is graded by the amount of light berries, moisture, extraneous matter and black pepper present. In general, higher grade black and white pepper holds less moisture and fewer light berries, as well as less extraneous matter.

Sarawak pepper grades, with their colour labels

GradeColour labelRank
Standard Malaysian Black Pepper No. 1BrownHighest
Sarawak Special BlackYellowSecond
Sarawak FAQ BlackBlackThird
Sarawak Field BlackPurpleFourth
Sarawak Coarse FieldGreyLowest
Standard Malaysian White Pepper No. 1CreamHighest
Sarawak Special WhiteGreenSecond
Sarawak FAQ WhiteBlueThird
Sarawak Field WhiteOrangeFourth
Sarawak Coarse WhiteGreyLowest

Table 20.3: Sarawak pepper grades. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 2.

Nutmeg from the two major origins, Indonesia and Grenada, is graded rather than set to an export limit. Nutmeg can be classified largely into "sound nutmeg", which has taken no injuries, and "substandard nutmeg". Sound nutmeg is also graded as "80s" and "110s" by the number of nutmegs per pound: "80s" means there are 80 pieces in one pound. Substandard nutmeg, which is exported from Indonesia, can be shrivelled and "BWP", meaning broken, wormy and punky. In Grenada, broken or injured nutmeg is graded "defective". BWP and defective nutmeg are often used as raw material for ground nutmeg, but broken or injured nutmeg is thought to be more likely to suffer insect infestation, which sometimes causes microbial problems such as aflatoxin production.

Spain and Hungary are among the major nations exporting paprika since the early 20th century. In Spain, paprika is defined as the product made by dehydrating and then grinding clean, fully ripe berries of Capsicum annuum and Capsicum longum, and both the sale and the use of biologically altered paprika are prohibited. Spanish paprika is classified into three grades by moisture content, total ash, ether-soluble extract, acid-insoluble ash and total fibre: Extra grade is made only from the peel, with all seeds and placenta removed; Select grade allows 10% seed content; Ordinary grade allows a 30% seed content. In Hungary, grade and quality standards are set by the Hungarian Office of Standard. Paprika is classified by three qualities and eight grades, by appearance, pungency and other characteristics such as total ash and amount of ether extract. First-quality grades are Special Paprika, Table Quality Mild Paprika (nonpungent), Table Quality (mildly pungent) and "Hot" Table Paprika. Second-quality grades include Semi-sweet Paprika, and third-quality grades include Pink (rose) Paprika and Pungent Paprika.

Other exporting nations run their own systems too. Sri Lanka regulates the quality of cardamom, coriander and other spices and the testing methods used, under the Bureau of Sri Lanka Standards. The minimum volatile oil content is relatively high at 4%, and six grades are set by appearance and colour. Zanzibar and Madagascar grade cloves for export by appearance, moisture content, extraneous matter and other factors.

Try it yourself: take the Sarawak table above and cover the colour labels. From the grade names alone, put the five black pepper grades in order from highest to lowest, then check yourself against the labels. The names carry the order: Standard No. 1, Special, FAQ, Field, Coarse Field. Now do the same for white pepper.

Quick check

Why is broken nutmeg a bigger problem than sound nutmeg, beyond the loss of appearance?

Check your answer

Broken or injured nutmeg is more likely to suffer insect infestation, and that sometimes causes microbial problems such as aflatoxin production. Sound nutmeg has no broken surfaces for moulds and insects to enter through.

20.14 Insects, fumigation and other ways to kill them

Insects harmful to farm products, including spices, are usually controlled by agricultural chemicals during cultivation. But spices can also be damaged by insects, including mites, during storage, because spices, like flour and beans, are generally stored for relatively long periods. Such pests are called "stored grain insects". Of the many harmful insects, moths and beetles are most damaging to spices. How fast the insects develop and breed depends on the air temperature, the kind of spice, and the kind of insect. Red pepper and basil are among the spices that often suffer from harmful insects during storage; parsley, garlic and oregano do not.

The cigarette beetle (Lasioderma serricorne) is found in many areas from tropical to temperate zones. The Indian meal moth (Plodia interpunctella) is one of the most common insects attacking flour and rice. The coffee bean weevil is known to breed on nutmeg. Both the flour beetle (Tribolium confusum) and the red flour beetle (Tribolium castaneum) also breed on red pepper. So-called book lice (Liposcelis bostrychophila) breed on paper and can sometimes be moved to spices from corrugated cartons or Kraft bags during transport. Miyazima and colleagues tested the pull of Tyroglypid mites to several spices, putting red pepper, coriander, garlic, laurel and mandarin in trap vessels and counting the mites drawn to each vessel. Red pepper and laurel drew the mites the most after both 2 and 48 hours, while garlic drew them far less.

Insects found on spices breed and multiply very quickly, and the result is a big problem unless action is taken early. Indian meal moths lay about 200 eggs at one time, and the life cycle of this moth is only around 40 days, so the total number can be expected to rise explosively within a short period.

The most common means used to control insects in the warehouse is fumigation, which has the advantage that it can reach every part of the storage warehouse and act uniformly. The two chemicals most widely used on spices for insect disinfection are methyl bromide and phosphine.

The two fumigants most used on spices

FumigantHow it worksStrengthsWeaknesses
Methyl bromideGas; boiling point 3.6 °C, so it can be used even in winter; efficacy follows K = C × T, where K is fumigation efficacy, C is gas concentration and T is fumigation timeShort fumigation time, several hours to a couple of days; relatively harmless to humans; used in warehouses for almost 50 years, so its effects are relatively predictableNot always as effective as phosphine, especially for pupae and eggs of some insects, despite strong effect on adult insects; believed to react with ozone and deplete the ozone layer
Phosphine (from aluminium phosphate)Aluminium phosphate in tablet form breaks down on reacting with water in the air into phosphine: AlP + 3H₂O → Al(OH)₃ + PH₃Very strong effect on many sorts and forms of insects, including eggs and pupae; easy to handle because it is a solidGenerated very slowly, so fumigation takes relatively long; said to be more toxic to humans than methyl bromide

Table 20.4: The two fumigants most used on spices. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 2.

Methyl bromide was listed as an ozone-depleting substance in November 1992 at the Montreal Protocol held at Copenhagen. With that in mind, the United States Environmental Protection Agency decided that the use of methyl bromide would be banned in a few years, and other industrial countries planned to phase out its use afterwards. The use of aluminium phosphate is expected to increase as methyl bromide is restricted.

Other ways of killing insects exist. Most insects will die in air holding less than 2% oxygen. Controlled-atmosphere storage, in which carbon dioxide or nitrogen is put in place of oxygen in air, has been studied and is practised in some nations. Carbon dioxide has a particularly lethal effect against some insects and can be expected to be used more widely. Most insects from tropical regions cannot survive at 15 °C or lower, except for mites, which multiply even at 10 °C, so low-temperature storage is considered an effective method of killing them.

Quick check

Why is the fumigation equation K = C × T useful to a warehouse manager?

Check your answer

Because efficacy rises with a longer fumigation time or a higher gas concentration, so the manager can trade one against the other: a lower concentration needs more time, and a shorter time needs a higher concentration.

20.15 Microbes, moulds and aflatoxin

Most spices are grown and harvested in tropical and subtropical regions, and dehydration, separation, milling and storage are not always done hygienically. In most cases, harvested spice plants are spread on the ground and sun-dried for several days until the wanted moisture level is reached. A large number of microbes, mainly spore-forming bacteria such as Bacillus species, live in the soil at levels of around 10⁶ to 10⁸ cells per gram. These can be a direct cause of contamination during sun-drying, and they also explain the relatively high counts of microbes on ginger and turmeric, which are dried rhizomes of tropical origin.

The number of bacteria on spices is generally higher than the number of moulds. In general, spore-forming bacteria like Bacillus species predominate, and Staphylococcus and Streptococcus species are also found in large numbers in most kinds of spices. It is unusual for a spice to be infected by pathogenic bacteria such as Escherichia coli or Salmonella. Clove is a special case: it holds only small mould counts, and its total microbe counts are also relatively low. That low count is thought to come from eugenol in the essential oil of clove, which has strong antimicrobial activity and seeps out onto the surface of ground clove.

Moulds need more attention than bacteria, because some moulds make toxic substances. Typical examples are Aspergillus flavus and Aspergillus ochraceus, which make hazardous toxins called aflatoxin and ochratoxin. Aflatoxin is especially known as a cause of cancer and is classified into B1, B2, G1 and G2 types. Aflatoxin B1 is the most toxic of all aflatoxin types. Many European nations limit aflatoxin in spices to no more than 10 parts per billion, and the types of spices likely to go above this limit are limited to red pepper and nutmeg. In Japan, aflatoxin was detected in peanut butter for the first time in 1970, and the Ministry of Welfare the following year set an upper limit for aflatoxin B1 of 10 parts per billion and required importers to check aflatoxin levels in peanuts and peanut products. The Ministry made known its intention to strengthen supervision of aflatoxin contamination in nutmeg in 1985 and in red pepper somewhat later.

Aspergillus glaucus and Aspergillus niger are the main moulds found contaminating spices. Aspergillus flavus, which includes strains able to make aflatoxin, has also been confirmed on many kinds of spice, though in most cases A. flavus does not outnumber A. niger or A. glaucus. Large numbers of Penicillium species have been detected on fenugreek, ginger and anise seed. Moulds found on spices are generally called "storage fungi" because they usually stick to spices and increase in number while the spice is sun-dried or stored after harvesting. Some types of beetles carry moulds, including Aspergillus species, to some types of spices, such as red pepper. So it is important not only to control the relative humidity of the storage warehouse, but also to kill insects by fumigation, to stop mould growth and aflatoxin production during storage.

Seenappa and colleagues studied the effect of relative humidity on the growth of Aspergillus species on different parts of red pepper. At 28 °C and 85% relative humidity, A. niger was the main mould found on stalks during storage, but when the relative humidity was raised to 95%, A. flavus and A. ochraceus predominated on the stalks and the pods, and the chance of aflatoxin production was raised. Moulds grew inside mechanically damaged pods. As moulds grow inside the pods, they slowly become yellowish, and the colour of the seeds inside the pods changes from the normal ivory white to olive green in pods holding A. flavus and to black in pods holding A. niger. That means red peppers heavily contaminated by Aspergillus species can be easily recognised by their yellow discoloration, and removing all discoloured pods after storage would be a useful way of stopping aflatoxin-contaminated red pepper from reaching the buyer.

Quick check

Why is clove unusually clean compared with other ground spices?

Check your answer

Because eugenol in the essential oil of clove has strong antimicrobial activity, and the oil seeps out onto the surface of ground clove, holding microbe counts down.

20.16 Sterilising: ethylene oxide, irradiation and steam

Three methods of sterilisation are discussed in the book: the ethylene oxide gas method, irradiation, and steam sterilisation.

Ethylene oxide has been widely used as a cold sterilisation method, especially for spices, because it has a minimal effect on the very sensitive flavours and aromas in spices. It has a strong sterilising and fumigating effect because it reacts with active parts of protein, including carboxyl groups, sulfhydryl groups, amino groups and hydroxyl groups, changing the protein structures in living insects and microbes. Sterilising effects were confirmed for many of the spices tested, such as black pepper and paprika. Koizumi and colleagues studied the effect of temperature on sterilising black pepper with ethylene oxide and found that a temperature range of 25 to 30 °C is most effective for killing microbes on this spice, and sterilising efficiency falls drastically at temperatures below 10 °C.

Ethylene oxide has real drawbacks. It is known to be toxic to humans in the vapour concentration used when sterilising, and because of its toxic nature many nations restrict its use in foods. It does not stay in spices for long, because its boiling point is around 10 °C, and usually the ethylene oxide residue falls to less than one-tenth after a week of storage. It sometimes reacts with nutrients and changes them, because its reactivity is very high: ascorbic acid (vitamin C) and riboflavin (vitamin B2) were reported to break down in reaction with ethylene oxide. The book prints riboflavin as vitamin B1, which is a misprint in the book: riboflavin is vitamin B2, and thiamine is vitamin B1. Treatment was not always wanted for some spices, such as paprika, which may turn a dark, dull colour.

Irradiation uses gamma radiation made from Cobalt-60 and Cesium-137, and electron and X-radiation from an electron accelerator. In 1980, the Joint Expert Committee on the Wholesomeness of Irradiated Food concluded that irradiation of any food commodity up to an overall average dose of 10 kGy introduces no toxicological hazard, and in 1983 a Codex General Standard for Irradiated Foods and a Recommended International Code of Practice for Operation of Radiation Facilities Used for Treatment of Food were adopted. To date, 37 nations permit the use of food irradiation for 40 or more kinds of foods, with the foods allowed varying by country. For spices, 32 governments have regulations allowing irradiation, but in Japan spice irradiation is still banned.

Irradiation doses reported for spices

What was measuredDose found
Thermophilic bacteria on spices, population 10³ to 10⁵ cells per gramSterilised by 10 kGy
Aerobic spores on spices, population 10² to 10⁵ cells per gramSterilised by 4 kGy
Coliform bacteria and fungi on spices, standard plate counts brought below 3,000 cells per gramAverage doses of 6.5 to 10 kGy
Flavour change in coriander and cardamomChanged at 7.5 kGy; other spice flavours unchanged below 10 kGy
Volatile compounds in black pepperα-pinene and carene resist irradiation; α-terpinene, γ-terpinene and terpinolene tended to fall in quantity even at 7.5 kGy

Table 20.5: Irradiation doses reported for spices. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 2.

Peroxide values for extracted lipids from some irradiated spices, especially nutmeg, were found to rise when relatively high doses were applied, but the rises were slight when the dose was below 10 kGy. The book's finding is clear: negligible irradiation doses of between 7 and 10 kGy can cut the populations of microbes on spices without changing their essential qualities.

Steam sterilisation works differently. Heat sterilisation is the most popular sterilising method for liquid foods such as sauces, but for low-moisture products such as spices this method is inadequate, because the heat conductivity inside such dried products is very low. When Clostridium sporogenes, a thermophilic spore bacterium, is to be sterilised by heat, the D-value, meaning the time needed to cut the population of microbes by one digit, at 120 °C with wet heat is 0.14 to 1.4 minutes, while with dry heat it is 115 to 195 minutes. That is why steam sterilisation with saturated or supersaturated steam is considered most effective for cutting microbes on spices.

Kikkoman has made an "air stream sterilising system" in which superheated steam is used as the heating medium. Superheated steam is a kind of dry steam, which does not make the target dry ingredient wet or stick together during sterilisation. In this system, the dry ingredient is delivered to the air stream tube, where superheated steam sterilises it while it is being carried along. One advantage is that even thermophilic spore bacteria can be sterilised quickly. Although the essential oils of some spices fell in content, total microbe counts were found to fall. The sterilisation time is very short, but heat does not get into the target as well as with ethylene oxide or radiation, so the sterilising effect may vary widely within a lot, especially when the surface of the spice is not smooth, as with whole seed. Another drawback is that green-coloured spices such as basil and parsley tend to become brown at high temperature.

20.17 Chemical quality standards: pungency, flavour and colour

Beyond water content specifications for stopping mould growth, chemical standards for quality control are also needed. Official import and export specifications do not always set such standards, but most spice-processing companies set them themselves to give the consumer a high-quality product.

Pungency standards. The major pungent compound in black and white pepper is an alkaloid compound called piperine. Besides this, pepper holds small amounts of other alkaloid compounds such as piperittine and piperyline, but piperine accounts for around 98% of the total alkaloid compounds in these spices. Piperine content tends to be affected by many environmental factors, such as climate, temperature and especially where it was grown: Brazilian black pepper tends to hold a relatively low amount of piperine compared with black pepper of other origins. Ways to measure piperine include the ultraviolet spectrophotometric method used by ASTA, the gas-chromatographic method, infrared analysis, the Kjeldahl method and some colorimetric methods. The Kjeldahl method, which measures nitrogen levels in piperine, is simple, but its results tend to read high because it also counts chavicine, piperittine and some amino acids in black and white pepper. Colorimetric methods measure the formaldehyde made by reacting chromotropic acid with the CH₂O₂ group of piperine in concentrated sulfuric acid. The ultraviolet spectrophotometric method is used most often because, besides being simple and fast, it has high specificity for piperine and does not tend to pick up other compounds such as chavicine and piperine isomers, and it can also measure piperettine content at a different ultraviolet region. The gas-chromatographic method uses a reaction that releases formaldehydes, so it measures not only piperine but also other compounds such as dextrose. High-performance liquid chromatography has also been studied, and its results are said to be a little lower than with the ultraviolet method because its resolution is superior.

The pungent principles of red pepper are capsaicinoid compounds, which are vanillylamides of monocarboxylic acids. The chain length of each capsaicinoid has a strong effect on the degree of pungency. The major pungent compound is capsaicin, which accounts for over 50% of the total capsaicinoids; dihydrocapsaicin, norhydrocapsaicin, homodihydrocapsaicin and homocapsaicin are also present in relatively high amounts. The degree of pungency differs widely with species and harvest place, and also with growing and drying conditions and harvest year: an almost twofold difference in pungency levels was reported depending on geographical location. Maturity of the plant also affects pungency level strongly. For those reasons, red pepper pungency is usually checked quantitatively by spice-processing companies before they use it.

The Scoville heat value method measures pungency by finding a distinct pungent sensation, and it was developed early in the 20th century. It is still widely used and has been adopted as an official ASTA method because of its simplicity in apparatus and theory, but it has drawbacks, including the long time needed to train panels and the difficulty of keeping accuracy because of panel fatigue. More than 10 kinds of pungent compound have been found in red pepper, with a different level of pungency for each, so it is hard to measure the overall pungency of the pepper and to measure each pungent compound quantitatively. One colorimetric method measures the blue colour made by the reaction of vanadium oxytrichloride with the hydroxyl group of capsaicin or its isomers, based on the molybdenum blue reaction. Several ultraviolet methods have also been reported, and one ultraviolet method was compared with the Scoville method and a very close correlation was found between the two. Gas chromatography methods have also been developed, and one was found to give a lower value than the colorimetric method, possibly because the capsaicin analogue is also measured by the colorimetric method. High-performance liquid chromatography methods for capsaicin analysis have also been studied, and one has been adopted as an ASTA official method. As with black pepper, the HPLC method is considered more accurate because of its high resolution, especially for capsaicin analysis in curry powder or curry roux, which holds impurities. All these quantitative methods have the drawback of taking a long time, because they are all based on extraction, and simpler and easier direct analytical methods are expected to be developed.

Mustard's pungent principle is a range of isothiocyanate compounds, which sit in plant tissue as a glycoside. The major pungent compound of black mustard is allyl isothiocyanate, made by the reaction of the enzyme myrosinase with the glycoside sinigrin, while that of white mustard is p-hydroxybenzyl isothiocyanate, which is less pungent than allyl isothiocyanate and is made by myrosinase acting on the glycoside sinalbin. Because these pungent compounds sit in the tissue as glycosides, which have no pungency, pungency levels usually do not fall even during making processes such as drying or milling, so the pungency level of mustard does not vary widely. Both allyl isothiocyanate and p-hydroxybenzyl isothiocyanate can be measured quantitatively with gas chromatography, and the colorimetric method for allyl isothiocyanate is used by ASTA, though with that method p-hydroxybenzyl isothiocyanate and some other isothiocyanate compounds cannot be measured quantitatively. Because these pungent compounds are considered stable during storage, pungency levels and allyl isothiocyanate content are not always measured as a routine quality control activity at spice companies.

Other pungent spices are not routinely checked quantitatively. Sansho, or Japanese pepper, is characterised by its "numbing" pungency due to sanshool compounds. Since both compounds break down quickly after the pepper is ground, it is usually stored in its whole form. The major chemical compounds behind the typical pungency of ginger are gingerol, zingerone and shogaol. Gingerol, the most pungent of the three, is thought to change to shogaol by dehydration of the β-hydroxyl ketone during the drying process, and shogaol is broken down into nonpungent polymers under the influence of heat. The pungency of ginger varies with the drying conditions.

Flavour and aroma standards. Spice flavours are expressed by their volatile essential oils, and some official import and export specifications set lower limits for volatile oil content for some spices. Since flavour is the most important property of most spices, many spice companies set their own standards without official specification, or set higher standards than the official ones. Because volatile oils tend to fall in amount during processes like grinding or sterilisation, minimum amounts in ground spices are usually lower than in whole spices. The analytical procedure for essential oil content is called the Clevenger method, which extracts the oil by distillation. In some cases the volatile oils in spice leaves (herbs) are not listed in specifications, because their contents are generally much lower than those of seeds and tend to vary widely.

It is also possible to measure a specific compound in an essential oil to check the flavour character of a certain spice. In cinnamon, cinnamaldehyde, which accounts for 60 to 90% of the whole composition of its volatile oil and is considered to represent its flavour character, is sometimes listed in specifications in addition to volatile oil content. The amount of cinnamaldehyde is measured by indirect titration of its steam volatile oil. Even within one species of spice, qualitative and quantitative differences in individual volatile composition exist with product origin or spice grower: Russell and colleagues found volatile differences between the Lampong and Sarawak cultivars of Piper nigrum (black pepper). Though these differences should affect the character of spice flavours, few standards exist to control their delicate flavours.

Colour standards. The green colour of parsley is a very important property of this spice. This green colour, expressed by chlorophylls, is usually checked by an organoleptic method because quantitative measurement is difficult, and the colour tone of parsley can be judged by appearance, because it is not concentrated as the colour of paprika or turmeric is. Paprika is used mainly for its colouring properties, and its colour tone matters in fixing the value of this spice. Its typical yellowish-red colour is formed by about 10 different kinds of carotenoids, with capsanthin predominating at more than 30% of the total carotenoids. The other carotenoids, which account for 0.1 to 0.8% of the total weight of this spice, include zeaxanthin, capsorbin, β-carotene and cryptoxanthin. The colour value is generally found by measuring an acetone extract spectrophotometrically with an absorbance at 460 nm, a method described in ASTA. Paprika loses its colour during storage through oxidation, helped by light, of carotenoid compounds. Since oxidation speeds up with rising temperature, some manufacturing firms keep dehydrated paprika in a cold place until it is packaged. The antioxidant ethoxyquin is sometimes added to stop rancidity and slow down carotenoid breakdown, but ethoxyquin use is not allowed in most European nations or in Japan.

The colour of turmeric, like that of paprika, matters to its commercial value. The yellow pigment from the rhizomes of this spice is curcumin, one of the diketones. Curcumin, accounting for about 3% of this spice, is not water soluble but is soluble in alcohol. The curcumin content of turmeric differs with harvest region: Alleppey turmeric in India is characterised by its high amount of curcumin, and export specifications for curcumin content of turmeric for this region are high, at 6.5% minimum. Colour value is generally found spectrophotometrically with absorbance at 425 nm, which has been adopted as an official method of ASTA. The commercial value of saffron is fixed mainly by its typical flavour, but its yellowish-red colour is also an important property. The principle of this colour is a terpene glycoside called crocin, a water-soluble yellowish-red pigment. Crocin is hydrolysed to the sugar gentiobiose and crocetin, which belongs to the carotenoids. The amount of crocin in saffron is found by observing the colour of its water extract.

Other quality standards. In addition to the standards above, there are cases in which total acid, acid-insoluble ash or water-insoluble ash are measured, mainly to limit the amount of extraneous matter in spices. For example, Sansho, a ground pericarp fruit skin, will have a very high acid-insoluble ash value if too many seeds or other parts of the plant are mixed with the pericarp when processed.

Practical note: reading a spice label like a grader

Pick up a jar of ground spice and read what the label does not say. A good specification would state the volatile oil, the moisture, the ash and the extraneous matter, but most retail labels state none of them. Use the book's tests instead: smell the powder at once after opening, then leave a spoonful open for a week and smell it again. The flat smell of the week-old powder is volatile oil lost to the air. Look at the colour against a fresh jar: paprika that has faded toward brown has oxidised, and no label date will tell you that. Check the powder for light berries, stems and grit, the same extraneous matter the ASTA limits, by spreading a spoonful on white paper. And remember the grading lesson of this unit: a grade name such as Tellicherry or Standard No. 1 is a promise about size, light berries and moisture, and a buyer who cannot read that promise is buying blind.

Key ideas to carry forward

  • A spice is the dry part of a plant that gives food flavour and a pungent stimulus, and no definition draws a clean line between a spice and an herb: the book counts edible leaves as spice leaves and keeps medicinal herbs outside cooking in the herb category.
  • Spices do four basic things, flavouring, pungency, colouring and deodorising or masking, and behind those sit secondary effects such as salt reduction, sugar reduction and better texture.
  • Flavour sits in the volatile essential oil, held in oil cells, glands and glandular hairs and freed when those tissues break; heat drives it off, so heat and time must be limited.
  • Pungency divides into "hot" (acid-amide and carbonyl compounds, generally nonvolatile) and "sharp" (thioether and isothiocyanate compounds, mostly volatile), and enzyme-made pungency in onion, garlic, mustard and horseradish is broken by heat, acid and salt.
  • Colour components are oil-soluble (carotenoids) or water-soluble (anthocyanins, flavonoids), and their tone changes with pH, metals and heat; natural colourants fade with oxygen and light, and packing and antioxidants hold them.
  • There is no single worldwide spice standard: ASTA and the United States Federal Specifications serve as international guides, while exporting nations grade by their own systems, from Agmark grades to Sarawak pepper labels, nutmeg counts per pound, and Spanish and Hungarian paprika grades.
  • Insects, microbes and moulds are the three quality enemies in storage; fumigation, controlled atmosphere and cold each kill insects differently, and aflatoxin, limited to 10 parts per billion in many European nations, is the mould risk that matters most for red pepper and nutmeg.
  • Ethylene oxide, irradiation at 7 to 10 kGy, and superheated steam each sterilise spices with a different trade-off between microbe kill and flavour, colour and nutrient loss.
  • Chemical quality is measured by piperine in pepper, capsaicinoids in red pepper, volatile oil and cinnamaldehyde in cinnamon, and colour value at 460 nm for paprika and 425 nm for turmeric.

Glossary

Spice

The dry part of a plant, such as a root, leaf or seed, that gives food flavour and a pungent stimulus. Food tech example: cinnamon bark or a clove bud used by the pinch to flavour a dish.

Herb

In the narrow sense, a plant that withers after blooming without its stems becoming woody, used for medicine, dyeing or gardening. Food tech example: a medicinal herb that never enters cooking stays an herb, while an edible leaf used in cooking counts as a spice leaf.

Spice leaf (edible herb)

An edible leaf usually called an herb, counted in the spice category in this book. Food tech example: basil or sage leaves dried and used as a spice.

Spice seasoning

A mixture holding one or more spices or spice extracts, added during processing or cooking to improve flavour. Food tech example: a seasoning blend added to a soup during cooking.

Condiment

A mixture of one or more spices and spice extracts added to food when it is eaten, not during cooking. Food tech example: pickles served as a side dish, or a sauce put on the table.

Essential oil

The volatile aromatic compounds of a spice, held in oil cells, glands and glandular hairs. Food tech example: the oil freed when a mint leaf is crushed, carrying the mint aroma.

Terpene

A compound of carbon, hydrogen and oxygen in a spice essential oil; 10 carbons makes a monoterpene, 15 a sesquiterpene, 20 a diterpene. Food tech example: monoterpenes giving a spice its strong, very volatile aroma.

Pungency

The hot or sharp sensation a spice gives in the mouth and nose. Food tech example: the hot bite of black pepper against the sharp bite of mustard.

Acid-amide compound

One of the four groups of pungent compounds, giving a hot sensation and generally nonvolatile. Food tech example: piperine in black pepper.

Isothiocyanate compound

One of the four groups of pungent compounds, giving a sharp sensation and mostly volatile, made when myrosinase breaks down a glycoside. Food tech example: allyl isothiocyanate in black mustard.

Alliin

The precursor glycoside in garlic that has no smell or pungency until the enzyme alliinase breaks it down into allicin. Food tech example: a whole garlic clove that smells of little until it is cut.

Carotenoid

An oil-soluble colour compound in spices, including carotene and its oxygenated forms, the xanthophylls. Food tech example: capsanthin giving paprika its red colour, or β-carotene colouring margarine.

Anthocyanin

A water-soluble colour compound that is red, blue, purple or pink, and changes tone with pH. Food tech example: hibiscus turning red in vinegar and purple at neutral pH.

Curcumin

The yellow pigment of turmeric, soluble in alcohol and not in water, yellow at acid to neutral pH and reddish-brown in alkaline pH. Food tech example: turmeric colouring a curry yellow.

Crocin

The water-soluble yellowish-red pigment of saffron, a terpene glycoside. Food tech example: saffron colouring rice in pilaf, where it cannot colour an oil-based food.

Extraneous matter

Material that should not be in a spice, such as insects, insect excrement, stones, stems and sticks, limited by specification. Food tech example: stems in bay leaves making a lot defective at 3% or more.

Light berry

A light, unfilled black pepper berry, limited to 4% by weight under ASTA rules. Food tech example: a pepper lot rejected for holding 5% light berries.

Agmark grade

An Indian grade for exported spices, set by the Directorate of Marketing and Inspection. Food tech example: curry powder graded "standard" at 85% or more spice and less than 5% salt.

BWP nutmeg

Substandard nutmeg that is broken, wormy and punky. Food tech example: broken nutmeg sold as raw material for ground nutmeg, with a higher risk of insect and aflatoxin problems.

Fumigation

Killing insects in stored spice with a gas such as methyl bromide or phosphine. Food tech example: methyl bromide fumigation of a spice warehouse, with efficacy following K = C × T.

Aflatoxin

A toxic substance made by moulds such as Aspergillus flavus, known as a cause of cancer, limited to 10 parts per billion in spices in many European nations. Food tech example: aflatoxin risk in red pepper and nutmeg stored damp.

kGy

The kilogray, the unit of irradiation dose. Food tech example: spices sterilised at 7 to 10 kGy without changing their essential qualities.

D-value

The time needed to cut a microbe population by one digit at a given temperature. Food tech example: Clostridium sporogenes at 120 °C, 0.14 to 1.4 minutes with wet heat against 115 to 195 minutes with dry heat.

Piperine

The major pungent alkaloid of black and white pepper, about 98% of their total alkaloids. Food tech example: piperine content measured by the ultraviolet method to grade pepper pungency.

Capsaicinoid

The pungent compounds of red pepper, of which capsaicin is the major one at over 50% of the total. Food tech example: capsaicin measured by the Scoville heat value method.

Clevenger method

The analytical procedure for essential oil content, extracting the oil by distillation. Food tech example: measuring volatile oil to check whether a ground spice still meets its flavour standard.

References

  • Hirasa, Kenji, and Takemasa, Mitsuo. Spice Science and Technology. Marcel Dekker, 1998. Chapters 1 (Spices and Herbs: Basic Concepts) and 2 (Spice Qualities and Specifications).

Unit 21: Cooking with Spices and the Patterning of Spice Use

Note

Unit 21: Cooking with Spices and the Patterning of Spice Use

A cook adds garlic to a steak before it meets the heat, sprinkles pepper on it after it leaves the heat, and would never do it the other way round. Another cook puts the same garlic on raw fish and wonders why the fish still smells of fish. Both cooks used the right spice. One of them used it at the right time, in the right form, with the right partners. This unit is about those choices: when a spice goes in, what form it takes, what it is mixed with, and how the patterns of whole cuisines can be read, counted and used to invent new dishes.

21.1 The four effects meet the kitchen

Unit 20 set out the four major effects of spices used in cooking: flavour, pungency, colouring, and deodorising or masking. In general, each spice fulfils one major function, but some spices fulfil more than one. This unit follows those four effects into the kitchen, where heat, time, form, blending and company decide how much of each effect actually reaches the plate.

Several factors affect spice function, including flavour, when spices are used for cooking. The book lists eleven of them, and they are worth learning as a set, because nearly every cooking decision in this unit comes back to one of them.

Eleven factors that influence spice flavour in cooking

No.Factor that influences spice flavour
1Spices hold both volatile and nonvolatile oils. Most spice flavours come from components in the volatile oil, but volatile flavour compounds can be made even from nonvolatile oil when heated. Flavour components tend to weaken or disappear at high temperatures, and they become more volatile as temperature rises, so the flavour sensation of each spice can be changed by controlling cooking conditions, including temperature and time.
2Oily components, including volatile oils, are generally soluble in alcohol, so the flavour of a spice can be raised effectively by cooking with alcohol.
3The oily component of any spice is oil soluble, so flavour characteristics can be raised by adding the spice to the oil phase of a salad dressing or to the oil used in cooking.
4It is not always necessary to eat the spice itself to get its wanted effect: a deodorising or neutralising effect can be had without using the spice flavour as such.
5It is sometimes effective to put a spice directly on the fire when grilling, giving a smoking effect.
6Each spice holds different pungent components with different pungent characteristics, so there are differences in pungent sensation, in how long each can be kept, and in how each appears. It matters to put each pungent spice to its proper use. Some spices change in pungency with heating, and spices whose pungent components work through enzymes are not suited to cooking with heat.
7Colouring compounds in spices are oil-soluble or water-soluble, and each type must be put to its proper use to get the colouring effect wanted.
8The flavour of food can be changed by using the proper spice at the proper time. When the deodorising effect of a spice is wanted, it is more effective to use the spice at the preparation stage or the finishing stage rather than during cooking.
9In most cases it is best to use a blend of spices rather than only one kind. The total amount of spice used is then greater than if only one spice were used, because the typical flavour of each spice is generally weakened when blending.
10The flavour of a spice blend will be milder after a period of time than if it were used right after blending.
11There are synergistic effects and suppressive effects for some spices: flavour characteristics, including flavour strength, change when a spice is used with another spice or seasoning.

Table 21.1: Eleven factors that influence spice flavour in cooking. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 3.

🤔 Pause and think: a salad dressing has an oil phase and a water phase. Which phase would you put paprika in, and which phase would you put saffron in, and why?

Quick check

A cook fries basil in hot oil and the basil loses its aroma. Which factor from Table 21.1 explains it?

Check your answer

Factor 1: flavour components are volatile even at room temperature and are lost swiftly on heating, so high temperature and long heating drive the flavour off. Factor 6 is also at work: basil's flavour compounds do not stand up to heat.

21.2 Forms of spice: from fresh leaf to coated granule

Spices can be used either fresh or in a dried form. A spice is generally characterised by its original fresh flavour, but the grassy smell of some spices makes them unwanted in the fresh form. Drying or milling may also change the original flavour of a spice, or may set off an enzyme activity that gives an unwanted flavour. When a spice is used for cooking, its overall quality must be judged, including microbial status, dispersion, heat resistance and colour tone reached, as well as flavour.

Mashing a fresh spice gives a spice paste. A spice that is dried is called whole spice, and grinding it gives ground spice. These "half-processed" spices can be processed further to suit various food-processing uses. Ground spice is processed further to give a natural essential oil by steam distillation, or an extracted spice such as oleoresin using alcohol or acetone. Such processed spices have the advantage of being uniform in quality and can be processed further to make other types of processed spice.

The forms of spice, from fresh leaf to coated granule

Spice formHow it is madeWhat the book says about it
Fresh spiceUsed as harvestedCarries its original fresh flavour, but some spices have a grassy smell in this form
Spice pasteFresh spice mashedA paste of the fresh spice
Whole spiceSpice driedThe dried spice, before grinding
Ground spiceWhole spice groundGround for cooking; can be processed further
Essential oilGround spice steam distilledNatural essential oil, uniform in quality
OleoresinGround spice extracted with alcohol or acetoneExtracted spice, uniform in quality
Liquid spiceOleoresin or essential oil dissolved in vegetable oil or alcoholA liquid form of the spice
Emulsified liquid spiceEssential oil or oleoresin emulsified in water with emulsifier and stabiliserNot always stable: the oil phase can separate from the water phase, and spice components can recrystallise
Absorbent spiceOleoresin or essential oil absorbed onto glucose, dextrin or saltThe spice component is sometimes easily volatilised or spoiled, depending on the absorbent used
Coating spice, water-soluble single layerOleoresin or essential oil emulsified in a water solution of a high molecular substance such as dextrin, sodium caseinate or gum arabic, then spray-dried into powderA coated powder
Coating spice, water-insoluble single layerOleoresin or essential oil mixed in heat-melted solid fat, then sprayed and cooledA coated powder
Coating spice, double layerOleoresin or essential oil mixed in heat-melted solid fat, then emulsified in a previously emulsified solution holding essential oil or oleoresin, then spray-dried into powderNo microorganisms can survive in it, and it can be kept for long periods; covered with both minute solid fat particles and water-soluble high molecular substances, so it keeps for especially long periods

Table 21.2: The forms of spice, from fresh leaf to coated granule. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 3.

Grinding changes flavour and pungency, and the size of the grind matters. Mori studied this using cardamom and white pepper, milling both spices to powders of different particle sizes for use in sausage, then checking how spice flavour differed by particle size. Flavour was significantly stronger when the medium ground spice (28 mesh) was used than when the finer ground spice (80 mesh) was used. The coarse ground spice in the study (14 mesh) seemed too large to spread evenly, so no significant difference could be found, though there was a tendency for more flavour when coarse ground spice was used.

Mori also compared emulsified spices and coating spices for their deodorising effect on fish smell. He used granules coated with hardened oil as the coating spice and an emulsion made with gum arabic as the emulsified spice, each used in sausage to judge by taste how well the fish smell was neutralised. The emulsified spice was so well spread out that it held back the fish smell much more than the coating spice. But the flavour and pungency of the coating spice were rated better than those of the emulsified spice. Form is a trade-off: the form that spreads best is not always the form that tastes best.

Try it yourself: take whole black peppercorns and divide them into three piles. Leave one pile whole, crush the second coarsely, and grind the third very fine. Put each pile in a separate cup of hot water, cover the cups, wait one minute, and smell each in turn. The coarse crush usually smells strongest, the fine powder weakest, because the finest grind has already lost its volatile oil to the air: the same pattern Mori found in sausage.

Quick check

Why did the emulsified spice beat the coating spice at removing fish smell, but lose on flavour and pungency?

Check your answer

The emulsified spice was so well dispersed through the sausage that it met the fish smell everywhere and held it back more. The coating spice kept its flavour and pungency better, because its oil was protected inside the coating instead of being spread thin.

21.3 Supercritical carbon dioxide: extracting flavour without cooking it

Supercritical fluid extraction has recently been applied to flavour extraction from spice. Its major characteristic is the use of carbon dioxide gas as the extract solvent in place of the organic solvent used in solvent extraction. Steam distillation and organic solvent extraction have both been widely used to extract compounds from spices, but both tend to damage the original spice flavour and lose volatile compounds because of the high temperatures used. With supercritical fluid extraction, the extraction temperatures can be low, so a high-quality extract with good flavour can be had.

The characteristics of supercritical carbon dioxide used as an extract solvent are: its density is close to that of a liquid, but its viscosity is very near to that of a gas, and its diffusion ability is 100 times that of a gas; both the critical temperature and the critical pressure are relatively low; simple control of temperature and pressure lets the operator easily control the fluid density, making this extraction method superior in selectivity, such as separation; since this fluid is inactive, and extraction can be done at relatively low temperature, the extract is not greatly broken down or oxidised during extraction; its ignitability is low, and it is nonpoisonous, odourless and tasteless; this fluid does not stay behind in the extract; and carbon dioxide is inexpensive and easy to obtain.

The book's Figure 3.2 compares what each extraction method pulls out of a spice. Steam distillation can separate essential oils, but not higher terpenes, free fatty acids, fats, wax resins or colourants. Alcohol can extract parts of some higher terpene compounds, though some flavour top notes disappear during the evaporating process. Many compounds can be extracted with both methylene chloride and supercritical carbon dioxide. Methylene chloride extraction has the drawback of scattering some flavour compounds during the evaporating process, while supercritical carbon dioxide can extract essential oils at relatively low temperatures, and higher terpenes, free fatty acids, fats, waxes and colourants can be extracted as the extraction temperature is raised. When this technique is used for spice, the typical spice flavour, which is usually very sensitive to high temperature, can be extracted without much damage because the temperature used is so low. The method can also separate out specific components such as colour and specific flavours by controlling extraction temperature and pressure: when a colour component is extracted from a spice, solvent extraction may also pull out some flavour components besides the expected colour, while the extract from supercritical fluid extraction will hold only the target colour component. The method still has drawbacks, such as a high processing cost, but its use is expected to rise as more research is done into the specific components of each spice, including physiologically active compounds.

Quick check

Why does supercritical carbon dioxide give a better flavoured extract than steam distillation?

Check your answer

Because extraction temperatures can be low, so the heat-sensitive spice flavour is not damaged and volatile compounds are not lost, while steam distillation uses high temperatures that damage the original flavour. It also leaves no solvent behind in the extract.

21.4 When to add the spice: before heat, during heat, after heat

Most flavour compounds in spices sit in their volatile oils and are changed or damaged on heating, so the timing of adding a spice when cooking is one of the most important parts of making full use of spices. The cooking process can be divided into three parts: preparation before heating, cooking using heat, and final preparation after the food is taken off the heat.

Black pepper shows how timing changes flavour. It can be sprinkled over meat with salt before baking or barbecuing, it can be added to a sauce or stew during the cooking period, and it can be sprinkled over beef after it has been taken off the heat. Each timing gives a different flavour. A spice will lose less of its delicate aroma if it is used after the food is taken off the heat, because most components of essential oils in spices are sensitive to heat.

Timing matters even more when a spice is used to deodorise fish or meat. The deodorising effect of a spice against a fish or meat smell is due mainly to a chemical reaction, so the right time to add the spice must be found. In general, a relatively high temperature is needed for spice components to react with the smelled compounds, so spices used for deodorising should be added either during the heating process or even before heat is applied. That is the opposite of the rule for delicate flavour, and it is the central tension of this section: flavour wants late adding, deodorising wants early adding.

Garlic shows the tension clearly. Garlic is often used to deodorise the smell of meat. If grated garlic were sprinkled over meat already baked or barbecued, you would get a strong garlic smell. But if it is used either during heating or before the heating process, garlic works well to get rid of the meat smell. Heating cuts the typical strong garlic smell as well as deodorising the meat smell, so a relatively large amount of garlic can be used. Beef is sometimes baked with garlic slices in a steak restaurant because this process neutralises the meat smell and changes the garlic flavour to a milder, roasted one.

Three moments to add a spice: pick ONE moment for the job

Each box below is a separate choice for one job, not a step in a sequence. One addition does not pass through all three moments.

1. Before heating, at preparation
Black pepper and salt on meat before baking; garlic with meat before cooking: the spice reacts with the smell compounds and the heat tames its own sharpness
2. During heating, in cooking
Pepper in a sauce or stew: flavour develops in the dish, and deodorising reactions have the heat they need
3. After heating, at finishing
Pepper sprinkled on beef off the heat: the delicate aroma survives, because the volatile oil is not driven off

Figure 21.1: The three moments to add a spice. Early adding suits deodorising, late adding suits delicate flavour. Each box is a separate choice for one job, not a step in a sequence. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 3.

Try it yourself: cook two small pieces of chicken the same way. Rub one with grated garlic before cooking and leave the other plain, then sprinkle grated garlic on the plain piece after it comes off the heat. Taste both. The first has little garlic smell and no strong meat smell; the second smells strongly of raw garlic. That is timing, not amount.

Quick check

Why should a deodorising spice go in early while a flavouring spice goes in late?

Check your answer

Deodorising works through a chemical reaction between spice components and smell compounds, and that reaction needs relatively high temperature, so the spice must be in during or before heating. Flavour compounds are volatile and are driven off by heat, so a flavouring spice keeps more of its delicate aroma when added after the food is off the heat.

21.5 Blending: why one spice is rarely enough

Statistically, there are very few cases where only one kind of spice is used for cooking. In most cases, a spice is used in combination with one or more other kinds of spice. When bouillon or soup is made by extraction from beef, pork or chicken bones, many spices are used to deodorise their typical smells. In general, four or more spices such as bay leaves, thyme, celery and parsley are added together during the extraction for these meats. This "grouping" technique is also used in processed foods such as ketchup and Worcestershire sauce.

It is hard to fix the amount of a certain spice for a cooking, because of differences in personal preference. Even a pleasant flavoured spice can have a medicine-like smell or taste if used to excess. But such a medicine-like flavour or aroma of one spice can be cut down by blending with other spices. Each spice has a typical flavour, and the quality of that flavour cannot be changed when only that spice is used for cooking, though the strength of its flavour can be controlled by adjusting the amount added. If a spice is grouped with other spices, the total spice combination will have a more delicate flavour than when each is used alone. This is called the "spice-blending" effect.

When blending is used to cut down the specific flavour of one spice, it is most effective to use a spice having similar flavour characteristics. The addition of celery or celery salt to tomato juice can remove the typical "greenish" flavour of tomato, because celery has a similar flavour. Besides celery, carrot and parsley are often added to tomato juice or tomato-based vegetable juice to make it more drinkable. One advantage of this technique is that the amount of spice added can be raised to cut down the unwanted typical flavour. Most people who dislike spices, especially children, say they do not like strong spice flavours, and spice-blending techniques can solve this problem.

The blending effect can be easily confirmed by comparing alcoholic drinks holding different combinations of spices. The typical spice flavour of a drink holding only one kind of spice tends to be felt as stronger than that of a drink to which three or more spices have been added. Whole spices tend to show the blending effect more clearly than ground spices. If clove flavour in a brandy holding only clove is compared with that in another brandy holding not just clove but also cinnamon and allspice, the first would give a stronger clove sensation than the second, because the similar flavours of cinnamon and allspice mask the typical flavour of clove.

The many kinds of blended spices found worldwide include chili powder, used in Mexican cooking, quatre epices and bouquet garni, used in French cooking, and herb seasoning, used in Mediterranean cooking. Japanese shichimi, the seven spice mixture, and Chinese five spice powder are also blended spices. Blending spices from the same botanical family gives a well-rounded flavour: Perilla family herbs are often blended together to make Italian herb mix or pizza spice mix.

Quick check

Why does adding more spices sometimes make a dish taste milder, not stronger?

Check your answer

Because of the spice-blending effect: when spices with similar flavours are grouped, each one's typical flavour is weakened and masked by the others, so the total combination tastes more delicate than any one spice alone, even though the total amount of spice is greater.

21.6 Aging: the blend that mellows

The flavour of blended spices becomes milder, or "rubbed", after they are stored for a period of time. Right after blending, the flavour of the spice mixture is a kind of assembly of separate flavours. As time passes, the separate flavours join into a single flavour. This is called the aging effect.

Aging techniques are widely used to make whiskey, miso and other products, especially fermented foods, and a special aging method is devised for each product. This aging process matters very much in creating the special flavours of wine and whiskey, which are aged in wooden barrels under controlled temperature for years. A wine aged for a longer period tends to be of higher quality.

Curry powder and chili powder, the most well-known blended spices, gain their characteristic flavour qualities with the aging process. For most manufactured curry or chili powders, aging conditions such as temperature and aging period are not specifically fixed: the aging effect would be had simply by leaving the blended spice in its container for several months.

The aging effect is thought to come from slow chemical and physical changes of the essential oils of each spice. So it is better to age a blended spice in a cool, dark place than at warmer temperatures, because the volatile flavour components would change slowly under mild conditions. Ultraviolet light and high temperatures, which spoil the properties of spice essential oils, should be avoided. Curry powder, for example, is better roasted in a pan at low heat if its flavour is kind of "rough": this process actually promotes the aging effect, while roasting at too high a temperature will evaporate the volatile compounds making up the typical curry flavour and give a product with poor flavour. The roasting process is also used in most spice companies for making curry powder, not only to promote the aging effect but also to create a delicate flavour characteristic specific to each curry product.

Quick check

Why should a blended spice be aged in a cool, dark place rather than a warm, bright one?

Check your answer

Because aging works through slow chemical and physical changes in the essential oils, and ultraviolet light and high temperatures spoil those oils. A cool, dark place lets the separate flavours join slowly into one flavour without losing the volatile compounds.

21.7 Synergy and suppression: when spices change each other

The taste of a food can often change when it is combined with another food or drink. A synergistic taste effect is when the taste of one food component is raised by association with other food components. A suppressive effect is when a certain taste is lowered in strength by combining with other components, also known as the "offset effect".

It is well known that sodium glutamate shows a remarkable synergistic effect with inosinic acid, a kind of taste illusion. Similar things can be seen with some spices. Ginger in soft drinks is one example: ginger has a typical earthy smell, but it also has a fresh citrus flavour and a fresh stimulus, and these raise the "freshness" of some soft drinks when ginger is added. Cinnamon has the effect of raising the sweetness of sugar-holding foods and has been widely used in baking and confectionery. Cinnamon holds a sweet aroma but no substance that gives a sweet taste, yet when it is combined with a sweet food, the sweet sensation is raised because of the synergistic effect between the sweet taste of sugar and the sweet aroma of cinnamon. On the other hand, when the salt concentration in a soup goes above a certain level, usually around 1%, adding black or white pepper is thought to hold back the salty taste.

Synergy also happens between spices. Take the addition of vanilla to fruit cake. Vanilla essence is made by combining alcohol with the essential oil of the vanilla bean to make it soluble in water, but the flavour of vanilla, when used for baking, is weakened when heated to over 100 °C, causing its essential oil to evaporate. In this case, adding a small amount of clove or nutmeg will create the impression of a stronger vanilla flavour, because the faded flavours of these spices when heated are close to that of vanilla.

Ito and colleagues showed how synergy and suppression work in deodorising, by testing how spices hold back mutton odour. They added a phased concentration of diluted solution of a single spice to an emulsion of 1% mutton oil and found by taste the amount of diluted spice solution needed to deodorise the smell of mutton. They also used dilute mixed solutions of two or more kinds of spice. One spice with a strong deodorising effect is sage, followed by thyme, clove and caraway. When sage and thyme, the strongest and second strongest, were mixed, the actual amount needed was more than the calculated amount: the two spices held each other back, a suppressive effect. They sorted the effects of two or more spices into three kinds: synergistic, suppressive and arithmetic, meaning the effect is simply the sum of the parts.

Quick check

Cinnamon holds no sweet-tasting substance, yet it makes sweet food taste sweeter. How?

Check your answer

Through synergy: the sweet aroma of cinnamon joins with the sweet taste of sugar, and the sweet sensation is raised by the two working together, even though cinnamon itself adds no sweet taste.

21.8 Less salt, less sugar: what the tests show

Judicious use of spices often lets a cook cut the amount of sugar or salt used, and the more healthful foods made this way could be useful in settings such as hospitals. Taste is a general term for what is felt when a food is seen, taken into the mouth and then into the stomach. It has three sides, psychological, physical and chemical, and spices can influence all three.

Sweet, tart, salty and bitter tastes are made by substances that dissolve in water and stimulate certain nerves. The components of a spice are felt as a physical stimulus in the mouth. How good food tastes is influenced not only by its taste, smell and texture, but also by psychological factors felt through sight, like its colour and shape. Of the factors that fix how good food tastes, those that spices can influence directly include pungency, flavour and colour.

Taste sensitivity falls with age. Higher minimum amounts of sugar, salt and sulfate quinine were needed for detection by the 75 to 89 year age group than by the 15 to 29 year age group, which points to fewer taste buds: the number of taste buds in one vallate papilla is said to drop suddenly at about the age of 70. Using spices to raise the flavour of foods while cutting salt or sugar could be especially useful for diseases of adults, including diabetes and kidney disease.

Saltiness is made by a chemical reaction on the tongue, but the pungent sensation from some spices is a physical sense thought to be similar to pain, so it is generally considered hard to replace salt completely with the pungency of spice. But the amount of salt needed can be expected to fall if the pungency, flavour and colouring effects of spices are used well. Using a pungent spice or spice leaves in a salad can make it very tasty, and using coloured spices may raise the total deliciousness of a salad. Such dishes can be called low in salt because the total amount of salt used is cut by using spices.

Ohta tested the effect of white pepper on salty taste. He made five salt solutions with concentrations of 1.38, 1.20, 1.04, 0.91 and 0.80%, and also made two salt-pepper solutions by adding white pepper at 0.08 and 0.16% to the 1.04% salt solution, comparing the salty sensation of each salt solution with each salt-pepper solution by taste, using several panels. The book prints 0.08% for the fifth solution in its list, and 0.80% in its results; the list figure is evidently a misprint, and 0.80% is used here. Eleven panels were asked one by one which salt solution matched the salt-pepper solution holding 0.08% white pepper and 1.04% salt. In the first attempt, 6 of 11 panels said its salty sensation matched the 1.04% salt solution, one panel each said it matched the 1.20 and 0.91% salt solutions, and 3 said the 0.80% salt solution matched. The assumed salt concentration of the salt-pepper solution was calculated to be 0.97% at the first attempt, and the average over three attempts was 1.01%. But when white and black pepper were added to the 0.8% salt solution, the assumed salt concentration rose to 0.96% and 0.91% respectively. These results show that salty sensation is raised by adding pepper when the salt concentration is low, but there is almost no effect if the salt concentration goes above 1.0%.

The effect of pepper on salty taste in Ohta's panel test

Solution testedWhat the panels said it tasted like
1.04% salt with 0.08% white pepper, first attempt6 of 11 panels said it matched 1.04% salt; 1 said 1.20%; 1 said 0.91%; 3 said 0.80%. Assumed salt concentration 0.97%
1.04% salt with 0.08% white pepper, average of three attemptsAssumed salt concentration 1.01%
0.8% salt with white pepperAssumed salt concentration 0.96%
0.8% salt with black pepperAssumed salt concentration 0.91%

Table 21.3: The effect of pepper on salty taste in Ohta's panel test. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 3.

Goto and colleagues studied the effect of capsaicin on the preference for salty taste, first in rats and then by testing the chorda tympani, the nerve that carries taste from the tongue. Rats were split into two groups: one fed a capsaicin-holding diet, the other a diet without capsaicin. Salt water at 0.5, 0.9 and 1.4% concentrations was prepared. One-tenth of the 8-week-old rats fed the diet without capsaicin were seen to drink the 1.4% salt water, while the rats fed the capsaicin diet did not drink this salt water during the 8-week testing period. The average amount of salt taken by rats fed the diet without capsaicin for 8 weeks was 7 g per 100 g of body weight, while the amount taken by capsaicin-fed rats was 5 g per 100 g of body weight. To check whether the change came from a numbed tongue, Goto and colleagues fed capsaicin directly into the stomach, 3 mg per rat per day, for 10 days, and still found that rats taking capsaicin tended not to prefer high-concentration salt water. Testing the chorda tympani gave a similar result: the nerve of rats taking capsaicin reacted to 0.5% salt water, while rats on normal diets reacted to 1.0% salt water, meaning the nerve of capsaicin-fed rats had been stimulated twice as much. Other research studied SHR-SP rats with high blood pressure, which prefer twice as much salt as normal rats, and confirmed that the amount of salt taken fell when capsaicin was fed, with a smaller rise in blood pressure. These results suggest that capsaicin can cut the preference for salt and so the amount of salt used in cooking.

The same kind of thinking applies to sugar. The sweet aroma of some spices can raise the sweet sensation, letting a cook cut the amount of sugar in some foods. Sucrose and glucose, important energy sources, are the most popular sweet compounds. The sweet sensation is affected not only by the amount of sugar but also by the texture of the food: sweetness needs to be stronger for hard-textured foods such as caramel than for soft-textured foods such as soft drinks or ice cream. Sugars are classed roughly into mono-, di-, tri-, oligo- and polysaccharides. All monosaccharides have a sweet taste, most disaccharides such as sucrose have a sweet taste as well, though some have just a slight sweetness, and polysaccharides have almost no sweet taste. Sugar alcohols, amino acids and peptides also give a sweet sensation, but their characters differ: some have a slight acid taste, others are slightly bitter, and the time taken to feel sweetness differs, sometimes at once on tasting, sometimes more as an aftertaste. Temperature also matters in feeling sweetness: fructose is 1.4 times as sweet at 0 °C but 0.8 times as sweet at 60 °C as sucrose at the same temperatures, taking the sweetness of sucrose at each temperature as 100. It is hard to raise the sweet sensation of ice cream, because the sweetness of sucrose is weak at low temperatures and adding extra sucrose changes the balance of total taste and flavour. In that case it is more effective to add vanilla essence, which has no sweet taste but does have a sweet aroma, than to add extra sucrose, because the aroma of vanilla can raise the total sweet sensation spread into the mouth together with the taste of sugar compounds. Spices in general hold several sugar compounds, and some spices like onion and garlic actually have a sweet taste. Anetol, one of the compounds with a strong sweet aroma, is held in anise seed, fennel seed and star anise, all of which can be used for raising the sweet sensation of foods. Cinnamon and cardamom are also used in sweet foods.

🤔 Pause and think: a cook wants a low-salt soup to taste as salty as a normal one. The soup is at 0.8% salt. Would adding white pepper help, and would it still help if the soup were at 1.2% salt? Use the numbers in Table 21.3.

Quick check

Why did capsaicin-fed rats drink less salt water, and what did the nerve test rule out?

Check your answer

They took less salt, 5 g per 100 g of body weight against 7 g for rats on the diet without capsaicin, and they did not drink the 1.4% salt water at all. Feeding capsaicin straight into the stomach for 10 days gave the same result, so the change was not caused by a numbed tongue: the chorda tympani of capsaicin-fed rats reacted to 0.5% salt water against 1.0% for normal rats.

21.9 Deodorising and masking: three types, and a fourth

Spices are sometimes used to deodorise or mask the smell of raw materials, and in that case the flavour of the spice itself matters less than its ability to mask an unwanted odour. Whether masking is wanted often depends on the people in question: Japanese people tend to dislike the typical smell of soymilk, so soymilk products sold in Japan are usually deodorised by some means, while Chinese people do not like soymilk products from which the smell has been completely removed.

Deodorising and masking divide into three types: chemical, physical and sensation.

The types of deodorising and masking

TypeHow it worksExample from the book
Chemical deodorisingSmelled compounds are changed to nonvolatile compounds or to odourless substances through a chemical reaction such as neutralisation, oxidation or reductionFish smell deodorised by soaking the fish in lemon juice or vinegar: alkali trimethylamine, one of the odorous compounds, is neutralised by the acid to become a nonvolatile compound
Physical deodorisingUnpleasant smelling compounds are absorbed by porous active carbon or zeoliteActive carbon used as a refrigerator deodorant because of this absorbing effect
Sensation deodorising, masking in the narrow senseA strong spice flavour covers an unpleasant smell; the actual odour does not fall in strength or disappear, it is simply not felt as readilyA strong spice aroma covering a fish smell
Sensation deodorising, offset deodorisingTwo compounds with different odours become odourless when mixedTwo smells cancelling each other when mixed

Table 21.4: The types of deodorising and masking. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 3.

Some spices have a fourth deodorising activity: the mucous membrane inside the nose, which works as a sensor for smell, is actually numbed so that the odour is not felt. Spices with this kind of activity include those holding volatile pungent compounds, such as wasabi or mustard. Mustard is often used in natto, fermented soybeans, a very popular side dish in Japan, and wasabi for raw fish. Their deodorising effect comes through isothiocyanate compounds, the flavour and pungent components of wasabi and mustard, which stimulate the mucous membrane of the nose when eaten. The mucus given off numbs the sensor function, giving a short inability to feel any odours.

The same thinking applies to bad breath. Bacteria in the mouth break down saliva and food sediment into bad smelling compounds, including volatile sulfur compounds, volatile nitrogen compounds, low fatty acids, alcohol, aldehydes and acetone, and the actual amounts of volatile sulfur compounds, such as hydrogen sulfide, methyl mercaptan and dimethyl sulfide, are proportional to the strength of the bad breath. Recent research found that the concentration of methyl mercaptan tracks bad breath more closely than the other compounds, showing that the major cause of bad breath is methyl mercaptan. Three means have been considered for deodorising bad breath: killing bacteria in the mouth, seizing methyl mercaptan with a deodoriser by chemical or physical deodorising, and masking the bad smell with flavour compounds.

Sodium copper chlorophyllin, a form of chlorophyll, is known to have a chemical deodorising function and is often used in breath-freshening chewing gum and some kinds of oral-care products such as toothpaste. Tokita and colleagues studied the deodorising effects of various plants, including spices, against methyl mercaptan, looking for a plant with a stronger deodorising effect than sodium copper chlorophyllin. Only a few kinds of spice showed a deodorising effect greater than sodium copper chlorophyllin. Thyme has a deodorising effect almost as strong as sodium copper chlorophyllin, while the effect of rosemary was found to be six times stronger and that of sage three times stronger. The effective deodorising compounds of rosemary and sage are carnosol and rosmanol, which also have an antioxidant function, and these compounds when isolated were confirmed to be many times stronger in deodorising function than sodium copper chlorophyllin. The major deodorising compound was found in the decolourised and de-essential-oiled ethanol extract section: the nonvolatile section was also found to seize methyl mercaptan.

There is a limit to what chemical deodorising alone can do, and the Weber-Fechner law explains why. The strength of an odour felt by the sense of smell is proportional to the logarithm of the concentration of the smelled compounds: the felt strength is proportional to the logarithm of the actual strength of the stimulus. Even if 99% of the total smelled compounds is removed chemically, the felt strength is cut by only 66%. So it is more effective to use an aromatic spice to deodorise the remaining 1% through the masking function. Judging from these facts, the spices in the Labiatae family, including rosemary, are most effective at deodorising, because they have both a chemical and a sensational deodorising function and raise the deodorising and masking effect very effectively.

Other studies tested the deodorising effect of spices against bad smelling compounds. Trimethylamine oxide is reduced to trimethylamine by microbial action, and trimethylamine rises as freshness is lost, so it is used as a reverse measure of seafood freshness. Kikuchi and colleagues tested the masking effect of each spice by taste, adding a spice solution in phased concentration to a trimethylamine solution. Sensory testing found the essential oils of onion, bay leaf and sage to be the most effective in holding back the odour of trimethylamine, followed by those of thyme, caraway, ginger, clove and cassia. It was also confirmed that the amount of trimethylamine measured by gas chromatography fell when the essential oils of sage and other effective spices were mixed with trimethylamine, and that trimethylamine falls slowly with time when the essential oils of ginger and sage are added. Niwa and colleagues chose the essential oils of onion and bay leaf for further research, and found that the phenol portion of the essential oil of bay leaf was very effective in masking trimethylamine, and on refining this phenol section the effective compounds were found to be eugenol and p-allyl-phenol. Onion, though confirmed by taste to hold back trimethylamine, showed no change on gas chromatography, which means its effect is sensational rather than chemical. Shiraishi and colleagues studied the holding-back compounds in some fermented seasonings against fish smell, and took the view that this holding back was due to flavour compounds changing to nonvolatile ones through the forming of a π-complex as well as salt formation. They tested phenol compounds, including vanillin, and terpene compounds against trimethylamine, and found that some compounds such as vanillin show agreement between the sensory test and the fall in trimethylamine, while other compounds such as acetovanillone do not. Some other compounds such as anisole were found to be effective by taste despite no fall in the amount of trimethylamine measured, showing that a holding-back mechanism against bad smell exists besides salt formation. The deodorising and masking effect of vanillin is taken to come from its bonding with fish smell compounds, since a vanillin solution absorbs trimethylamine very well.

Kawamura and colleagues tested cooking aromas from heated chicken soup and reported that the aromas were held back by adding bay leaves or cloves. They also tested aromas from chicken soup holding thyme, thyme and rosemary, or thyme and clove, and the aromas from the soups holding two kinds of spice were found to be simply mixtures of the flavour compounds of each spice, proving that there were no chemical reactions between the different flavour compounds.

Research on masking body odours, including methyl mercaptan, has also drawn attention. Nutrients from food are broken down by digestive enzymes in the stomach and intestines, absorbed through the intestinal wall into the blood, and the residues are sent to the colon and changed to faeces. Nutrients taken in are broken down by intestinal bacteria or enzymes in the digestive organs, and some of these broken-down nutrients hold bad smelling components. Some mercaptan compounds, hydrogen sulfide and indole compounds, all made in the intestines, are absorbed into the blood through the intestinal wall and sent to the liver. If the working of the liver is weakened by a lesion or by aging, it cannot properly break down such substances, so they build up, giving bad breath and body odour. It is said that a high concentration of ammonia in the blood may be linked to certain diseases such as cancer, arteriosclerosis or even Alzheimer's disease. Abe and colleagues wrote about the deodorising effects of a plant extract, from mushroom, against both ammonia in blood and foul smelling compounds in faeces such as methyl mercaptan. They added this plant extract to livestock feed and fed it by mouth to rats for 30 days. Both the ammonia concentration in blood and the gas concentration in faeces were remarkably lower after the 30-day feeding period, and they suggested that the ammonia concentration in blood is linked to that in faeces. As mentioned, some plant extracts were found to hold back body odour, and it is expected that the use of spices or spice extracts to deodorise or mask bad breath will draw attention in the future.

🤔 Pause and think: a fish smells of trimethylamine. Onion holds the smell back by taste, but the gas chromatography reading does not fall. Bay leaf holds it back and the reading does fall. Which one is working chemically, and which one is working through sensation?

Quick check

Why is it not enough to remove 99% of a bad smell chemically?

Check your answer

Because of the Weber-Fechner law: the strength of an odour felt is proportional to the logarithm of its concentration, so removing 99% of the smelled compounds cuts the felt strength by only 66%. An aromatic spice is needed to mask the remaining 1%.

21.10 The patterning theory: five ways to find a new combination

The use of spices in cooking varies greatly by area of the world. Sometimes spices not used in a native cuisine are brought in and used in a new way. Ginger is used in Japan in raw fish meals, such as finely chopped horse mackerel or lightly roasted bonito, while in Europe it is often used with pork. The combination of ginger and pork also occurs in Japanese cooking, such as grilled pork with ginger. Sliced ginger may be added to stewed giblets to remove the unpleasant smell of giblets.

The relationship of spices to cooking ingredients and techniques can be thought of as a kind of pattern. The "patterning theory" of spice use considers the suitability of a spice and certain raw materials to be the result of a "synthesis" that happens in the mouth. A preference for a specific spice is fixed by individual judgement, and the suitability of any spice and any other ingredient rests on individual preference. Flavour preference is judged using the senses of taste and smell: if a spice and an ingredient tasted together are well received, it follows that these two can be combined in cooking. The theory gives five substitution moves for finding new combinations.

Move 1: synthesis in the mouth. The Japanese confection Daifuku, holding sweet bean paste, is considered compatible in taste with green tea, and the two are often taken together, so it is not surprising that products such as a sweet jellied bean paste holding both ingredients are sold. In the same way, if drinking coffee while eating a cake holding certain spices is pleasurable, it would make sense to mix coffee with those spices, and along the same way of thinking, new types of spiced teas or drinks could be made.

Move 2: substitute an analogous ingredient. Ground meat is a major ingredient in many foods, and nutmeg is often added to ground meat to deodorise or neutralise its meat smell. This use for nutmeg can be applied across the broad range of raw materials that have meat or meatlike flavours, such as milk and cream. So a spice used to lift meat flavour can be used in a stew that holds milk or a cake to which fresh cream is added, though the timing or amount of spice may need adjusting. For example, if pepper or ginger suits meat, it should be possible to make a kind of ice cream to which pepper or ginger is added.

Move 3: substitute an analogous spice. Spices with similar flavours tend to hold similar kinds of flavour compounds, so patterning can mean replacing one spice with others that have similar flavour characteristics.

Move 4: substitute a spice from the same botanical family. Spices in the same family tend to hold similar series of flavour compounds and similar flavour characteristics, so one can stand in for another for a specific cooking ingredient. Onion, garlic and leek all belong to the Liliaceae family, and their similar flavour characteristics let them stand in for one another. Garlic is usually used in Gyoza, a Chinese dish in which minced pork is wrapped in a small pancake and steamed or fried. Nira (Allium ampeloprasum L.), which belongs to the same botanical family as garlic, can stand in for garlic in this dish.

Move 5: substitute a spice with similar pungency. The pungent spices of the Cruciferae family all have similar pungent characteristics and can be used interchangeably. Horseradish is often used to season roast beef. Radish, which belongs to the same family, can be used instead if a Japanese flavour accent is wanted. Soy sauce is also used to cook Japanese-flavoured beef steak because its flavour is known to be compatible with that of radish. Horseradish powder is used as an imitation "wasabi" powder by adding a green colour, but it also tends to be acceptable to the Japanese palate if Japanese mustard, which also belongs to the Cruciferae family, is added.

A sixth move works on seasonings rather than spices. A common cooking technique uses vinegar, wine or spirits to season certain cooked foods. If a certain spice soaked in wine vinegar is found to be a good combination, Japanese rice vinegar could also be used. Japanese sake or shochu, Japanese spirits distilled from sweet potatoes, can be used in place of liqueur, another kind of alcoholic drink. There are many kinds of alcohol to which spices are added: gin is usually used as the base of a cocktail, and several kinds of spices are added during the brewing of gin. Gin or rum is often added to fruitcake, and spices can be added to make the flavour of fruitcake more characteristic. Red pepper alcohol, made by soaking red pepper in Japanese shochu, can be used as a new kind of seasoning.

Five alternative moves: how to invent a new spice combination

Each box below is one alternative move, not a step in a sequence. A cook uses one move, not all five in order.

Move 1: Synthesis in the mouth
Two things tasted together and liked: mix them. Daifuku with green tea; coffee with the spices in its cake
Move 2: Analogous ingredient
Nutmeg suits ground meat, so it suits milk and cream too; pepper and ginger for meat can go into ice cream
Move 3: Analogous spice
Spices with similar flavours hold similar flavour compounds: swap one for another
Move 4: Same botanical family
Onion, garlic and leek are all Liliaceae: nira can stand in for garlic in Gyoza
Move 5: Similar pungency
Cruciferae spices share pungency: radish can stand in for horseradish on roast beef

Figure 21.2: The five patterning moves for finding new spice combinations. Each move is an alternative way to invent a combination, not a step in a sequence. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 4.

Try it yourself: pick a dish you know well, for example a stew with milk or cream. Write down the spice it uses. Now use Move 2: what meat dish uses the same spice for the same reason? Then use Move 4: what spice from the same botanical family could stand in for it? You have just designed a new combination the way the patterning theory does, and the only test left is tasting it.

Quick check

Garlic is the usual spice in Gyoza. Which spice can stand in for it under the patterning theory, and why?

Check your answer

Nira, because it belongs to the same botanical family as garlic, the Liliaceae family, and spices in the same family tend to hold similar flavour compounds and similar flavour characteristics.

21.11 Frequency patterning analysis: counting 9,000 recipes

When thinking about what spices to use in cooking, the name of the dish or the cooking method is one important factor. But it is not necessarily true that a specific spice would suit only some specific meals or specific ingredients: any spice can be used in many cooking applications. Choosing a spice for a certain food or cooking technique is often influenced by the preference of the people of the nation where the combination will be used.

The book's authors tested the patterning theory with a method called "Frequency Patterning Analysis". The suitability of a spice for any cooking ingredient or technique is fixed in this method from the range of combinations of that spice with those ingredients or techniques in a range of recipes in which various kinds of spices are used. The name comes from the frequency of each of 40 spices being analysed so that a pattern of spice use for each nation, each cooking ingredient and each cooking technique can be seen.

The steps are:

  1. Choose 40 spices that appear in recipes worldwide, and analyse statistically how often they appear by nation, cooking ingredients and cooking technique; then judge the suitability of each spice by each of these factors.
  2. Divide cooking ingredients into vegetable and animal ingredients, then divide further into nine categories: meats, fish, milk, cereals, potatoes, fruits, beans, seeds and beverages. Count the number of times each spice appears under each ingredient. This number is the frequency of appearance of that specific spice.
  3. Divide cooking techniques after adding spices into heat and nonheat cooking, then divide further into eight categories: boiling, baking (including grilling), frying, deep-frying, steaming, pickling, dressing with sauce, and not cooking. Count the number of times each spice appears in each category.
  4. Divide types of cooking or meals into French, Italian, German, British, American (Western-style) and Southeast Asian, Chinese, Japanese, Indian and others (Eastern-style). Count the number of times each spice appears in each category.
  5. Calculate the amount of each spice, as a percent value, for cooking ingredient, cooking technique and nation, from the frequency of appearance found for each case.
  6. Calculate the suitable index of each spice for each of the categories, cooking ingredient, cooking technique and nation, from the percent value above.
  7. Calculate the mean value of the spice from the values found above. A spice with a higher value than the calculated mean value is considered "suitable". Any spice with a much higher value is considered "very suitable".

Points to keep in mind during analysis: remove and try not to count items that are hard to interpret; do not count any spice more than once in the same category; new data can always be added, and the reliability of the analysis rises with the amount of data added; when blended spices are used, count each spice separately. Frequency patterning analysis rests on 40 of 88 botanically classifiable spices commonly used worldwide. If a different number or selection of spices is used, a new set of data needs to be gathered and analysed.

A worked example makes the method plain. The book uses a recipe for white kidney bean soup. Onion, leek, pepper, parsley, celery and bay leaves are the spices to be analysed. Chicken stock, pork, butter, carrot, kidney beans and other ingredients are also used, and one point each is given to meat (chicken stock, pork), milk (butter), vegetable (carrot) and beans (kidney beans). Cooking technique is analysed after adding a spice: one point each for onion and leek is given to "frying" because both spices are added before the frying process, and another point each for onion, leek, parsley, celery and bay leaves is given to "boiling" because these spices are added as a bouquet garni before the boiling process.

The results of the analysis, shown in the book's Tables 4.3 to 4.5, come from 9,000 recipes from around the world. The tables themselves are printed as images in the book, and their cell values cannot be read from the copy on file, so this unit reports only the findings the book states in its text, not the table numbers.

The results can be used in three ways. First, a spice can be picked to fulfil a certain function, using its major function (A-value): spices can be used to flavour, deodorise (mask), colour or add pungency. Second, a spice may have subfunctions (B-value) as well as its major function: the major function of red pepper is to give pungency, but it also has the subfunction of acting as a colourant, while paprika has no flavouring, deodorising or other function other than its colouring function. Third, picking spices can be made easier by cross reference: to find a spice suited to boiling meat in an Italian-style meal, the cooking ingredient, cooking technique and national indexes can be consulted together.

Quick check

In the white kidney bean soup example, why does onion get a point for frying and a point for boiling?

Check your answer

Because onion is added before the frying process, so it counts for frying, and it is also added as part of the bouquet garni before the boiling process, so it counts for boiling too. Technique is counted after the spice is added, at each stage where it is present.

21.12 What the patterns show: nations

The book's national findings, arranged by the four functions, show clear East and West patterns. When the incidence of spice use in Eastern cooking is compared with that in Western cooking, sesame seed, star anise, fenugreek and cardamom are often used to flavour Eastern dishes, while tarragon, basil, allspice, marjoram and vanilla are used more often in Western foods. This makes sense when the origins of the spices are taken into account: star anise from China, cardamom from India, fenugreek and sesame seed from the Middle or Near East, and allspice from South and Central America, though there are exceptions.

Rosemary, sage, oregano, thyme and celery are commonly used in Western cooking for deodorising. Rosemary, sage and oregano, all belonging to the Labiatae family, are routinely used in Mediterranean countries such as Italy and Greece. There are no great differences between Western and Eastern cooking in the use of spices as colourants or for pungency. One exception is ginger, a spice of Eastern origin: it is often used in Japanese and Chinese cooking to mask the smell of seafood and to lend pungency, while in Western countries it is used most often in cakes to raise the sweet flavour.

The national profiles are:

  • American cooking: basil, tarragon, dill, anise, paprika, mint, parsley, allspice and pepper are suitable, while star anise and Japanese pepper, which are eaten often in China and Japan, are almost never used.
  • Indian cooking: fenugreek, cardamom, cumin, mace, turmeric, saffron, coriander, fennel, cinnamon, clove, red pepper and nutmeg were found suitable. These spices are grown and processed mainly in India. The spice mixture garam masala, widely used in Indian cooking, consists of a blend of five or more of these spices.
  • Japanese cooking: sesame, leek, garlic, onion, horseradish, Japanese pepper, mustard, red pepper and ginger were found very suitable. Sesame, Japanese pepper, red pepper and ginger are often used for making a Japanese-flavoured spice mixture.
  • Chinese cooking: star anise, ginger, sesame and coriander show very high suitability. The Chinese five-spice mixture includes star anise, Japanese pepper and others.
  • Southeast Asian cooking: red pepper and cumin.
  • British cooking: marjoram, sage, mint, allspice, rosemary, parsley, fennel and nutmeg.
  • German cooking: savory, caraway, nutmeg and paprika.
  • Italian cooking: vanilla, oregano, rosemary, celery and saffron.
  • French cooking: thyme, bay leaves and celery.

The book also gives a suitability profile for each of the 40 spices, one by one, summarised in the table below.

Suitability profile of each of the 40 spices, by nation

SpiceWhere it suits best
ParsleyMainly Western cooking; less often Eastern, except Chinese and Southeast Asian cuisine, where it tends to be used as an alternative to coriander leaves, also called Chinese parsley
CeleryFar more suited to Western cooking than Eastern, except Southeast Asia; used particularly well in Italian and French cooking
CinnamonSuits both Western and Eastern dishes, but used most in Indian and Southeast Asian cooking; suitability for Italian cooking is especially low
AllspiceMostly Western cooking, less suited to Eastern cooking; high values for British, American and German cooking
DillMainly Western cooking, such as American; almost no Eastern nations except India
MintMost suited to Western cooking, such as British and American; only India among Eastern cuisines shows suitability
TarragonBest suited to Western cooking, such as American and French; almost no suitability for Eastern cooking
CuminMost suited to Eastern cooking, such as Indian and Southeast Asian; no suitability for any Western cooking except American
MarjoramMost suited to Western cooking, such as British, Italian and German; no suitability for Eastern cooking
Star aniseMost suited to Eastern cooking, such as Chinese and Southeast Asian; no suitability for Western cooking
BasilMost suited to Western cooking, such as American and Italian; almost no suitability for Eastern cooking
AniseMost suited to Western cooking, such as American; almost no suitability for Eastern cooking except Indian
MaceMost suited to Indian cooking; almost no suitability for Western cooking except in British foods
FennelMuch more suited to Indian and British cooking than to any other nation's cooking
Sesame seedMost suited to Eastern cooking, such as Japanese and Chinese
VanillaWestern cooking, especially Italian; almost no suitability for Eastern cooking
FenugreekDoes not show great suitability for any cuisine except Indian
CardamomMost suited to Indian cooking, but does not show above-average suitability for any other national cuisine
GarlicAlmost equally suited to Western and Eastern cooking, with slightly higher suitability for Italian and Chinese cooking
OnionSuited to both Eastern and Western cooking, especially American and French
Bay leafMost suited to Western cooking, such as French and Italian; no great suitability for any Eastern cuisine except Southeast Asian
CloveOverwhelmingly suited to Indian cooking
NutmegFar more suited to Western cooking, such as German and British, than Eastern cooking except for Indian
LeekMore suited to Italian, Japanese and Southeast Asian cooking than to other national cuisines
ThymeFar more suited to Western cooking, especially French and Italian, than Eastern cooking
RosemaryFar more suited to Western cooking, especially French, British and Italian, than any Eastern cooking
CarawaySuited to Western cooking, especially German
SageSuited to Western cooking, especially British
OreganoSuited to Western cooking, especially Italian
SavoryMuch used for Western cooking, especially German
CorianderMost suited to Eastern cooking, especially Chinese and Indian
PepperAlmost no difference between Eastern and Western cooking, but very high suitability for American cooking
Red pepperSuited to Eastern cooking, especially Southeast Asian and Indian
MustardSuited to both Western and Eastern cooking, with slightly higher suitability for Japanese cooking
GingerMost suited to Eastern cooking, especially Chinese and Japanese
HorseradishSuited to Japanese cooking among Eastern nations and to British cooking among Western ones
Japanese pepperSuited to Eastern cooking, especially Japanese
PaprikaSuited to Western cooking, such as American and German, but no suitability for Eastern cooking except Southeast Asian
TurmericSuited to Eastern cooking, especially Indian
SaffronSuited to Indian and Italian cooking and no other nation's cuisine

Table 21.5: Suitability profile of each of the 40 spices, by nation. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 4.

A nation's typical cuisine tends to change slowly, being affected by regional cooking trends around the world. In Japan, a type of regional cooking called "ethnic meals" has become popular, but the flavours are usually altered to taste more "Japanese". In Western cooking, there are cases where flavours characteristic of French food are fitted to another nation's cuisine. Creole cuisine, found in and around New Orleans, is basically a kind of French cooking, but American flavour characteristics are added using chives, green pepper and poppy seed, all routinely used in American cooking. Chinoise cuisine is a kind of Chinese-French cooking in which typical Chinese spices like ginger, coriander, Japanese pepper and Chinese five-spice mixture are added to basic French food. TexMex cooking is another example of a flavour mixture of two nations, in which anise, coriander, cumin and chives, often found in Southwestern food, are used in addition to spices normally used in Mexico.

Quick check

Why does star anise show high suitability for Chinese cooking and almost none for Western cooking?

Check your answer

Because spice use follows origin: star anise comes from China, so it is built into Chinese cooking, including the five-spice mixture, while Western cuisines did not build their dishes around it. The patterning analysis counts how often each spice actually appears in each nation's recipes.

21.13 What the patterns show: ingredients

Spice trends by cooking ingredient are arranged by flavouring, deodorising, colouring and pungency functions. There is only a small difference between the suitability of spices for animal as opposed to plant materials. Exceptions include fenugreek, in many cases used only for animal materials for its flavouring function, and coriander, used specifically for animal materials for deodorising. For colouring and pungency, no difference is found.

Spice suitability by cooking ingredient

IngredientSpices with high suitability, and why
MeatGarlic, rosemary, sage and Japanese pepper. Garlic is widely used for many kinds of meat such as beef, pork, lamb and chicken, so it has high suitability for most kinds of meats. Most spices suited to meat also have a deodorising function. Japanese pepper also shows high suitability to meat as a flavouring spice and is blended with salt in many fried meat dishes in Eastern nations.
SeafoodFenugreek, Japanese pepper, horseradish, anise, fennel and saffron. Fenugreek is an essential ingredient of curry powder or garam masala in Indian cooking. Japanese pepper and horseradish are routinely used for Japanese seafood dishes. Fennel is often used in fish dishes in Western nations, and is also used in China to stop people becoming ill after eating fish. Saffron is often used to cook bouillabaisse, risotto, pilaf and paella, in which many kinds of seafood are used.
MilkCaraway and vanilla. Vanilla is essential for making ice cream holding fresh cream, and caraway seed is mixed with many kinds of cheese products.
EggsThyme and leeks. Both are known to have a deodorising function, and are used to mask the typical egg smell, for example when cooking omelettes.
GrainsMost spices were found suited to grain, with caraway showing especially high suitability. Caraway is widely used in Europe, especially in Germany, in breads, cookies and cakes.
VegetablesMost spices were found suited to vegetables, with coriander judged most suitable. Its frequency of use is higher in Eastern nations than elsewhere, and it is especially routinely used in the cooking of India and China, where its fresh leaves, called "shiang cai" in China, are often used in salads or other vegetable dishes.
FruitsA relatively large number of spices were found suited to fruits, but only a few show very high suitability. Vanilla is most suitable, because many fruits are used in desserts in which vanilla is often used.
Beans and seedsCardamom, savory, cumin, turmeric and ginger show high suitability. All of these spices except savory are ingredients mainly used in curry powder, which is often cooked with some kind of beans. These spices do not have a deodorising function, so another spice with a deodorising or masking function is needed when the curry dish holds meats or seafood. Savory is used in bean dishes in German and other Western cooking.
BeveragesBay leaves, celery, parsley, coriander and thyme show high suitability. Celery and parsley are common parts of vegetable juice, doing a deodorising job or giving a "hidden flavour" rather than a flavouring function. Bay leaves, coriander and thyme are expected to do a deodorising or masking job.

Table 21.6: Spice suitability by cooking ingredient. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 4.

Quick check

A curry of beans uses cardamom, cumin, turmeric and ginger. What is missing if meat is added to that curry?

Check your answer

A spice with a deodorising or masking function. The four named spices do not have a deodorising function, so another spice that masks meat smell is needed when the curry holds meat or seafood.

21.14 What the patterns show: cooking techniques

Spice trends by cooking technique are shown for flavouring, deodorising, pungency and colouring. Of spices with a flavouring function, fenugreek, sesame seed and oregano can be used for cooking with heat. Fenugreek and sesame seed hold nonvolatile oils that tend to change on heating so that their typical flavours are raised. Oregano, which differs from other Labiatae spices such as basil and marjoram in its flavour characteristics, holds a "dull" flavour that can be weakened by heating. Among spices with a deodorising or masking function, rosemary, thyme and star anise are suited to cooking with heat, because heating weakens their unwanted flavours. Most pungent spices such as mustard, horseradish and Japanese pepper are more suited to cooking without heat: both mustard and horseradish get their pungency by the action of an enzyme, which is easily switched off by heating, and the typical flavour compounds in Japanese pepper tend to evaporate on heating. Among spices with a colouring function, saffron shows great suitability for cooking with heat: since its colouring compounds dissolve in water, it is particularly effective for boiling dishes, such as those holding rice, like pilaf and paella.

Spice suitability by cooking technique

Cooking techniqueSpices with high suitability
BoiledSaffron, star anise, celery, marjoram, bay leaves, thyme, basil, fennel, onion, clove and rosemary. Most of these have either a deodorising or a flavouring function.
BakedSesame seed, fenugreek, vanilla, rosemary, fennel, garlic, bay leaves, sage, onion, nutmeg, leek, oregano and pepper. Sesame seed is well known to give an appetising flavour on baking.
FriedSesame seed, clove, thyme, garlic, onion, nutmeg, rosemary, mace, leek, oregano, ginger and turmeric. Most of these also show high suitability for baked foods.
SteamedCinnamon, allspice, turmeric, vanilla, thyme, dill, ginger, cardamom and sage. Most of these are used for making confectionery and are characterised by relatively mild flavours. Cinnamon, dill and cardamom are also suited to nonheat cooking.
Deep-friedCumin, sesame seed, oregano, sage, thyme, rosemary and leek. These spices, except for sesame, are characterised by a strong flavour, which may be made milder by interaction with heat.
Sauces, no heatMustard, horseradish, Japanese pepper, turmeric, red pepper, caraway, parsley and celery. Many of these are picked for their pungent characteristics.
PickledAnise, horseradish, tarragon, mustard, Japanese pepper, mace, celery, allspice, dill, nutmeg, coriander and red pepper. Most of these have either a pungent or a flavouring function. Coriander, whose major function is deodorising or masking, is used in pickled foods to raise their appearance.
Fresh foods, no cookingMint, horseradish, parsley, tarragon, basil, Japanese pepper, star anise, mustard, coriander, celery, fennel, vanilla and cardamom. The kinds picked are very similar to those used for pickled foods, except for vanilla, mint and parsley, which suit fresh foods but not pickled foods.

Table 21.7: Spice suitability by cooking technique. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 4.

Try it yourself: take the eight rows of Table 21.7 and sort them into two piles, cooking with heat and cooking without heat. Then find the spices that appear in both piles: mustard, horseradish, Japanese pepper, cinnamon, dill and cardamom all cross the line. Ask why each one can cross it: mustard and horseradish lose pungency to heat yet still appear in hot dishes, and cinnamon, dill and cardamom are mild enough to suit both.

Quick check

Why are mustard and horseradish better suited to cold sauces than to hot cooking?

Check your answer

Because they get their pungency from an enzyme breaking down a glycoside, and heat switches that enzyme off, so their pungency disappears in hot cooking and can even turn bitter.

21.15 Practical note: designing a dish with the patterns

Pick a dish you have never cooked and design it with the three indexes the book gives: ingredient, technique and nation. Say you want to boil meat in an Italian-style meal. Look at the ingredient index for meat, where garlic, rosemary, sage and Japanese pepper score high, then at the technique index for boiling, where rosemary, thyme, bay leaves, basil and sage score high, then at the national index for Italian cooking, where rosemary, oregano, celery, vanilla and saffron score high. The spice that appears in all three lists is rosemary. That is the cross-reference method the book teaches: the answer is the spice that survives all three filters.

Now test it the other way. You want a cold sauce with bite. The technique index for sauces without heat gives mustard, horseradish, Japanese pepper, turmeric, red pepper, caraway, parsley and celery, and the pungency rule from section 21.14 tells you why: enzyme-made pungency survives cold and dies in heat. Pick mustard, and you have a sauce that keeps its bite. Try the same spice in a long boil and the bite is gone: that is not a failure of the spice, it is a failure to read the pattern.

Key ideas to carry forward

  • Flavour compounds are volatile and are lost by heat, so timing is a cooking decision: deodorising spices go in early, because their reaction needs heat, while delicate flavouring spices go in late, after the food is off the heat.
  • Spice form changes what a spice does: medium grinding at 28 mesh gave stronger flavour than fine grinding at 80 mesh in Mori's sausage test, emulsified spice spreads and deodorises best, and coating spice keeps flavour and pungency best.
  • Supercritical carbon dioxide extracts spice flavour at low temperature, pulling out essential oils, higher terpenes, free fatty acids, fats, waxes and colourants as temperature rises, with no solvent left behind, though processing costs are high.
  • Blending weakens each spice's typical flavour and makes the total more delicate, and blending works best with spices of similar flavour or from the same botanical family; aging then joins the separate flavours into one, slowly, in a cool, dark place.
  • Synergy raises a taste and suppression lowers one: cinnamon raises sweet sensation without adding sweetness, pepper raises salty sensation only when salt is low, below 1.0%, and capsaicin-fed rats took less salt, 5 g per 100 g of body weight against 7 g.
  • Deodorising works three ways, chemical, physical and sensational, plus a fourth, numbing the nose's sensor, and the Weber-Fechner law means removing 99% of a smell chemically cuts the felt strength by only 66%, so masking must finish the job.
  • Rosemary is six times and sage three times stronger than sodium copper chlorophyllin at seizing methyl mercaptan, through carnosol and rosmanol, and the Labiatae family is most effective because it deodorises both chemically and sensationally.
  • The patterning theory finds new combinations by five moves: synthesis in the mouth, analogous ingredient, analogous spice, same botanical family, and similar pungency, plus swapping Western seasonings for Japanese ones.
  • Frequency patterning analysis counts how often 40 spices appear in 9,000 recipes by nation, ingredient and technique, and a spice above the mean is suitable, while one far above the mean is very suitable; the spice that survives the ingredient, technique and national indexes together is the right pick.

Glossary

Volatile oil

The aromatic compounds of a spice that evaporate easily and give its smell and flavour. Food tech example: the oil lost when basil is fried in hot oil.

Whole spice

A spice that has been dried but not yet ground. Food tech example: whole black peppercorns ground near the time of use.

Oleoresin

A concentrated extract of a spice holding its volatile oil and pungent compounds, made with alcohol or acetone. Food tech example: oleoresin used where a uniform strength is wanted.

Liquid spice

A spice oleoresin or essential oil dissolved in vegetable oil or alcohol. Food tech example: a liquid spice dosed into a dressing for even flavour.

Emulsified liquid spice

Essential oil or oleoresin emulsified in water with an emulsifier and stabiliser. Food tech example: the gum arabic emulsion that spread through sausage and held back fish smell better than a coating spice.

Coating spice

Essential oil or oleoresin coated in a water-soluble layer, a fat layer, or both, then dried to powder. Food tech example: double layer coating spice that holds no living microorganisms and keeps for especially long periods.

Supercritical carbon dioxide extraction

Extracting spice compounds with carbon dioxide held above its critical point, where it flows like a gas and dissolves like a liquid, at low temperature and with no residue. Food tech example: extracting a spice colour alone, without its flavour compounds, by controlling temperature and pressure.

Mesh

A measure of grind size: the number of openings per inch in the sieve the powder passes through, so a higher mesh number means a finer powder. Food tech example: 28 mesh medium ground spice giving stronger flavour than 80 mesh fine ground spice in sausage.

Spice-blending effect

The more delicate flavour that results when spices are grouped, because each spice's typical flavour is weakened and masked by the others. Food tech example: clove tasting weaker in a brandy that also holds cinnamon and allspice than in a brandy holding clove alone.

Aging effect

The mellowing of a blended spice in storage, as its separate flavours join into a single flavour through slow chemical and physical changes in the essential oils. Food tech example: curry powder left in its container for several months gaining its characteristic flavour.

Synergistic effect

A taste raised by association with another food component. Food tech example: cinnamon raising the sweet sensation of a sugared food, though it holds no sweet-tasting substance.

Suppressive effect (offset effect)

A taste lowered in strength by combining with another component. Food tech example: black or white pepper holding back salty taste in a soup above about 1% salt.

Chorda tympani

The nerve that carries taste from the tongue. Food tech example: the nerve of capsaicin-fed rats reacting to 0.5% salt water, against 1.0% for rats on normal diets.

Chemical deodorising

Changing smelled compounds to nonvolatile or odourless substances by a chemical reaction such as neutralisation, oxidation or reduction. Food tech example: lemon juice neutralising trimethylamine in fish.

Physical deodorising

Soaking up unpleasant smelling compounds with a porous material such as active carbon or zeolite. Food tech example: active carbon as a refrigerator deodorant.

Sensational deodorising (masking)

Covering an unpleasant smell with a strong spice flavour, or cancelling two odours by mixing them, so the smell is not felt though it is still there. Food tech example: a strong spice aroma covering a fish smell.

Weber-Fechner law

The rule that the strength of an odour felt is proportional to the logarithm of its concentration. Food tech example: removing 99% of smelled compounds chemically cuts the felt strength by only 66%.

Methyl mercaptan

A volatile sulfur compound, the major cause of bad breath. Food tech example: methyl mercaptan seized by rosemary and sage extracts, which beat sodium copper chlorophyllin.

Trimethylamine

The compound behind fish smell, made when trimethylamine oxide is reduced by microbial action as freshness is lost. Food tech example: trimethylamine held back best by the essential oils of onion, bay leaf and sage.

Patterning theory

The idea that the suitability of a spice and a raw material is a pattern, a synthesis in the mouth, and that new combinations can be found by substituting analogous ingredients, analogous spices, spices from the same botanical family, or spices with similar pungency. Food tech example: nira standing in for garlic in Gyoza.

Frequency patterning analysis

A method that judges how suited a spice is by counting how often it appears with each ingredient, technique and nation in recipes, here 9,000 recipes and 40 spices. Food tech example: white kidney bean soup scored one point per spice per ingredient and per technique.

Major function (A-value)

The main job a spice does: flavouring, deodorising or masking, colouring, or pungency. Food tech example: red pepper, whose major function is pungency.

Subfunction (B-value)

A second job a spice does besides its major function. Food tech example: red pepper acting as a colourant as well as giving pungency.

Bouquet garni

A bundle of herbs and spices added before boiling, taken out after cooking. Food tech example: onion, leek, parsley, celery and bay leaves added as a bouquet garni to white kidney bean soup.

References

  • Hirasa, Kenji, and Takemasa, Mitsuo. Spice Science and Technology. Marcel Dekker, 1998. Chapters 3 (Cooking with Spices) and 4 (The Patterning Theory of Spice Use).

Unit 22: Spices and Health: Physiological Effects, Antimicrobials and Antioxidants

Note

Unit 22: Spices and Health: Physiological Effects, Antimicrobials and Antioxidants

A spoon of chilli sauce can make your forehead sweat, a clove of garlic can slow the microbes in a sausage, and a pinch of rosemary can keep a tub of lard from going rancid for longer. Each effect comes from a named compound, acting through a named route, in a study you can check. This unit follows those studies, and it keeps the study setting in view at every step, so a result in rats stays a result in rats and a result in a petri dish stays a result in a petri dish.

Questions to think about as you read: Why does chilli make you feel hot and then, after you sweat, cooler? Why do some spices slow microbes in food while others do almost nothing? Why would a food maker choose rosemary extract over a synthetic antioxidant?

22.1 What the body does with a pungent spice

Kim and co-workers fed red pepper as part of food and found that only a small quantity of capsaicin and its related compounds was absorbed in the gut. Kawada and co-workers, working with male Wistar rats given 3 mg of capsaicin by stomach tube, found more than 80% of the dose absorbed within 3 hours, taken up in the stomach quickly and also in the small intestine, in the duodenum and the jejunum.

Once absorbed, capsaicin sets off a chain. It activates the primary afferent neurons, which release the transmitter substance P from the spinal cord. That step switches on sympathetic efferent nerves and raises the secretion of adrenal catecholamines, above all epinephrine. Epinephrine then acts on the liver and on fat tissue: liver glycogen breaks down to glucose, which rises rapidly in the blood, and fat breaks down to free fatty acids, which rise gradually, and both are carried to muscle and burned as body heat.

Capsaicin: from the gut to body heat

Eaten in food
Capsaicin, piperine, zingerone

↓

Absorbed in the gut
Rats: more than 80% of a 3 mg capsaicin dose within 3 hours

↓

Nerves switch on
Substance P released, epinephrine secreted

↓

Fuel is freed and burned
Glycogen to glucose, fat to free fatty acids, burned in muscle as heat

Figure 22.1: The capsaicin chain from gut to heat. Each step is a finding from the rat studies reported in the book. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 5.

Pungent spice also changes digestion itself, in three ways: chemical digestion is promoted by more digestive juice, physical digestion by more movement of the gut, and absorption by more blood flow in the gut. A person normally secretes 1 to 1.5 litres of saliva a day, mostly water with protein, including alpha-amylase and mucin, and its secretion is promoted by pungent spice, above all by the nonvolatile compounds.

Quick check

Why does the book report two different findings on how much capsaicin is absorbed?

Check your answer

Kim and co-workers, feeding red pepper as part of food, found only a small quantity absorbed in the gut, while Kawada and co-workers, giving rats 3 mg of capsaicin by stomach tube, found more than 80% absorbed within 3 hours. The compound, the dose, the food around it and the animal studied all change the result, so the book reports both.

22.2 Hot, then cool: skin temperature after pungent food

Eat a hot curry and your skin warms, you feel hot, you may start to perspire, and then your skin temperature falls again. If the pungent compound is taken in a large quantity, the skin temperature after sweating can drop lower than it was before the food was eaten. In general, the more pungent the food, the bigger the gap between the highest temperature recorded before sweating and the lowest temperature after sweating.

Skin temperature after pungent food

High, medium and low pungency are three separate test conditions, shown as three curves on one graph in the book, not steps in a sequence. In each case the skin warms, then falls after sweating, sometimes below the starting point.

High pungency (A)
Biggest rise, then the deepest fall after sweating
Medium pungency (B)
Smaller rise, smaller fall
Low pungency (C)
Smallest rise, smallest fall
Separate measurement, not a next step: core temperature
Measured rectally, it does not change greatly

Figure 22.2: Skin surface temperature after pungent food, for high, medium and low pungency. The three pungency levels are separate test conditions, and core temperature is a separate measurement, not the next step after low pungency. The fall after sweating can go below the starting temperature. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 5, Fig. 5.7.

🤔 Pause and think: a cook in a hot climate adds plenty of chilli to a meal, and the eaters sweat and then feel cooler. A cook in a cold climate adds the same amount. What goes wrong, on the book's account?

Quick check

After a very hot meal, why can your skin end up cooler than before you ate?

Check your answer

The skin warms first as blood flow to the surface rises, then sweating discharges heat from the skin. With a large quantity of pungent spice, the fall after sweating can go below the starting temperature. Core temperature, measured rectally, does not change greatly.

22.3 Weight, fat and endurance: what the studies found

Kawada and co-workers fed rats a high-lipid feed made mainly of lard, with 0.014% capsaicin added, which the book notes is the average daily intake in Southeast Asia, for 10 days. The weight of perirenal adipose tissue and the serum triglyceride concentration were significantly lower in the capsaicin-fed rats, and perirenal lipoprotein lipase activity was increased, although the rate of weight gain and the rate of lipid absorption were not affected.

Henry and Emery studied 12 panels of 20-year-olds who were accustomed to eating spiced food, measuring metabolic rate by oxygen consumption. Half the panels were given a breakfast with no spice and the rest the same breakfast with chilli sauce and mustard sauce added, on consecutive days, and resting metabolic rate was measured at intervals for 3 hours.

Table 22.1: The capsaicin weight studies reported in the book

StudySubjects and doseWhat the study found
Kawada and co-workers, 1986Rats, high-lipid feed with 0.014% capsaicin for 10 daysPerirenal fat and serum triglyceride significantly lower; lipoprotein lipase activity increased; weight gain and lipid absorption unchanged
Henry and Emery, 198612 panels of 20-year-olds used to spiced food; spiced breakfastResting metabolic rate rise over 3 hours: 158% with spice against 128% without; metabolic rate increased 25%
Yoshida and co-workers, 1988Rats, capsaicin and isothiocyanate injected into muscleBrown adipose tissue temperature and oxygen consumption rose, as with ephedrine; long-term confirmation still needed
Kim and co-workers, 1995Mice, capsaicin 6 mg per kilogram, swimming testMice swimming 3 hours after dosing swam significantly longer; almost no effect at 1 hour

Quick check

In the Kawada rat study, what changed and what did not change with 0.014% capsaicin in the feed?

Check your answer

Perirenal adipose tissue weight and serum triglyceride fell, and lipoprotein lipase activity rose. The rate of weight gain and the rate of lipid absorption did not change.

22.4 Garlic, onion and the blood

Garlic and onion belong to the lily family, and their sulphur compounds act on the blood in several ways, all of it in animal or laboratory studies. Allicin and diallyl sulfide are sulphur compounds formed from alliine in garlic tissue.

When a vessel is injured, platelets aggregate, blood coagulates around the clump, and the clot can block a narrowed vessel, slowing and then stopping flow, which can lead to myocardial or brain infarction. Ajoene, found in onion and garlic, hinders thrombus production by preventing platelet aggregation, and it acts together with methyl allyl trisulfide, known as MATS. MATS blocks the formation of thromboxane A2, which drives platelets to aggregate, and it makes up only 5% of the essential oil of garlic.

Scordinin was tested in rabbits and in rats. In rabbits it was confirmed to decrease cholesterol levels, and it lowered both glutamic oxaloacetic transaminase (GOT) and glutamic pyruvic transaminase (GPT) levels in blood, which the book reads as an ability to maintain normal liver function.

Quick check

How does MATS in garlic act on platelets, and how much of garlic essential oil is MATS?

Check your answer

MATS inhibits the formation of thromboxane A2, which is produced when arachidonic acid is metabolised and which drives platelets to aggregate. MATS makes up only 5% of the essential oil of garlic.

22.5 Cell culture, mice and cancer: reading the limits first

This section needs its warning before its content. Everything in it comes from cell culture or from mice, and the book says so at each step. None of it shows that eating a spice treats or prevents cancer in people, and this unit does not say that it does.

Yamazaki studied the Labiatae family, the mint family, against HIV in cell culture. Ground herbs or plants were extracted with acetone, ethanol and 70% ethanol in that order, and the residues were freeze-dried as water extract. Each extract was added at various concentrations to MT-4 cells partly infected by HIV (LAV-1) and cultivated for 5 days. The acetone and ethanol extracts, which contain the flavour compounds, showed no effect. The water extracts of 30 plants, and the 70% ethanol extracts of some plants, suppressed HIV in the cells, most of them below 31 micrograms per millilitre, and lemon balm was effective against both the LAV-1 virus and the KK-1 virus obtained directly from a patient with AIDS.

For cancer, the book reports two mouse findings. A phenol compound called MSP, obtained by extracting mace with acetone and fractionating with another solvent, inhibited the propagation of cancer cells in vitro in mice, and an anticancer effect was also observed. In a second study, mice injected with breast cancer-inducing cells and fed basil, sage, oregano, peppermint and perilla lived longer than mice without the spices, though the book gives no numbers, no dose and no mechanism.

Quick check

In the Yamazaki HIV study, which extracts worked and which did not?

Check your answer

The water extracts of 30 plants, and the 70% ethanol extracts of some plants, suppressed HIV in MT-4 cells, mostly below 31 micrograms per millilitre. The acetone and ethanol extracts, which contain the flavour compounds, showed no effect.

22.6 Antimicrobials: whole spices, oils and single compounds

Spices have been used to keep food for centuries: cinnamon, cumin and thyme were used in mummification in ancient Egypt, spices were used in ancient India and China to preserve foods and for medicinal purposes, coriander was used in ancient Greece and Rome to extend the preservation of meat, and mint was used to prevent milk from spoiling. Research on the antimicrobial properties of spices began in the 1880s, and mustard, clove and cinnamon were soon proven to have antimicrobial effects.

The early oil studies used a phenol coefficient, comparing each spice oil with phenol. Clove oil and cinnamon oil were effective in retarding microbial growth, and oregano oil and peppermint oil also had relatively strong effects.

Mace extract delayed slime formation on sausage, and the preventive effect was greater at the lower storage temperature. Garlic had a strong antimicrobial effect against all eight microorganisms tested by Dole and Knapp, including Salmonella typhi and Shigella dysenteriae, and onion and clove were effective against all but B. subtilis.

Thyme essential oil inhibited the growth of Aspergillus parasiticus and its aflatoxin production at 0.2 mg per millilitre. Bullerman found that adding 0.02% cinnamon powder suppressed aflatoxin production by 21 to 25%, and aflatoxin production was almost completely inhibited at 2.0% cinnamon. Cinnamon bread usually contains 0.5 to 1.0% cinnamon, which the book reads as enough to almost completely inhibit mould growth and aflatoxin production. The ethanol extract of cinnamon was stronger still: 0.02% of it inhibited the production of aflatoxins B1, B2, G1 and G2 by 74 to 89%, although mould growth was not inhibited at that concentration.

Three levels of antimicrobial action

Whole spice, essential oil and single compound are three levels of analysis, not a process sequence. No level turns into the next one.

Level 1: Whole spice
Garlic strong against all eight test microbes; cinnamon powder 0.02% cuts aflatoxin 21 to 25%
Level 2: Essential oil
Thyme oil 0.2 mg per millilitre stops Aspergillus parasiticus growth and aflatoxin
Level 3: Single compound
Eugenol 200 to 800 parts per million by microbe; cinnamaldehyde 0.66 millimolar against most fungi

Figure 22.3: Antimicrobial action at three levels, from whole spice to single compound. The numbers are the study concentrations reported in the book. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 6.

Mustard oil is formed when milled mustard seed is mixed with water, and its major antimicrobial principle in black mustard is allyl isothiocyanate, produced by the enzyme myrosinase acting on the glycoside sinigrin. Wasabi and horseradish also produce allyl isothiocyanate when their tissues are disrupted. These compounds are volatile and lose their effectiveness quickly, although their antimicrobial activities are in general stronger than those of ginger and red pepper.

Table 22.2: Antimicrobial concentrations reported in the book

Spice or compoundTestConcentration and result
Sage or rosemaryBacteria in the Shelef study0.3% bacteriostatic; 0.5% bactericidal; allspice MIC twice as high
Cinnamon powderAflatoxin production by Aspergillus parasiticus0.02% suppressed aflatoxin 21 to 25%; 2.0% almost completely inhibited it
Cinnamon ethanol extractAflatoxins B1, B2, G1, G20.02% inhibited production 74 to 89%, without inhibiting mould growth
Thyme essential oilAspergillus parasiticus growth and aflatoxin0.2 mg per millilitre inhibited both
EugenolAcinetobacter and yeast; Bacillus megaterium and Pseudomonas; Aspergillus flavus and A. versicolor200 parts per million; 800 parts per million; 250 parts per million complete inhibition
Cinnamaldehyde18 fungi0.66 millimolar inhibited most fungi tested
Allyl isothiocyanate vapourBacteria and yeasts31 to 470 micrograms per dish, equal to 16 to 110 nanograms per millilitre of vapour
Garlic juiceTyphoid bacillus; most microorganisms0.5% inactivated typhoid bacillus completely in 5 minutes; 3.0% inhibited almost all kinds tested

Try it yourself: Make two small pots of plain cooked rice. Stir a pinch of ground cinnamon into one and leave the other plain. Cover both loosely and leave them at room temperature for two days, then compare smell and visible mould. The cinnamon pot should stay cleaner for longer. That is the book's cinnamon finding at kitchen scale, and it is not a preservation method for food you plan to eat: throw both pots away after looking.

22.7 Antioxidants: keeping fat from going rancid

Fat goes rancid because it reacts with oxygen in the air to generate peroxides, which are further oxidised and break down into low molecular weight alcohol and aldehyde compounds. The free radicals formed in this process are also thought to work in the human body to damage DNA and promote cancer and aging.

Studies in the 1930s found that some spices retarded the generation of peroxide in peanut oil and inhibited rancidity in meats, and the first patent for spice antioxidants, approved in 1938, specified that spice oils such as clove oil could prevent the oxidation of cooking oils. Chipault and co-workers evaluated 72 kinds of spices and found 32 that retarded the oxidation of lard. Ground sage and rosemary were particularly strong, and oregano, thyme, turmeric and nutmeg possessed relatively strong antioxidant activities.

From rosemary, carnosol was isolated first, then rosmanol, epirosmanol and isorosmanol. Rosmanol, an odourless and colourless compound, showed antioxidant activity more than four times greater than BHA and BHT. From oregano came a phenolic glucoside, caffeic acid, protocatechuic acid, rosmarinic acid and 2-caffeoyloxy-3-phenylpropionic acid, more effective antioxidants than alpha-tocopherol, with the phenolic glucoside and 2-caffeoyloxy-3-phenylpropionic acid comparable to BHA. Mace is more effective than nutmeg in retarding the autoxidation of lard, although the chemical components in each are almost the same.

Rosemary extract has been commercialised as an antioxidant agent under the name HSE: SP-100. In the AOM test at 97.8 °C, the rosemary agent retarded the oxidation of lard to the same degree as BHA at a concentration of 200 parts per million. In soybean oil, BHA and tocopherol did not show antioxidant activity at 200 parts per million, while the rosemary extract did, and its effect rises in proportion to its concentration while tocopherol plateaus, with a synergistic effect when it is used with tocopherol.

Antioxidant strength, as the book ranks it

  1. Rank 1, strongest: Sage and rosemary: stronger than BHA; rosmanol more than four times BHA and BHT
  2. Rank 2, strong: Oregano and thyme; mace myristphenone two to four times BHA in lard; turmeric curcuminoids stronger than alpha-tocopherol
  3. Rank 3, useful in some foods: Eugenol and thymol best on dehydrated pork; black pepper strong in mayonnaise, weak in lard
  4. Rank 4, weak here: Linalool and cineol: no marked antioxidant properties on dehydrated pork

Figure 22.4: How the book ranks spice antioxidants, from sage and rosemary down. The ranking depends on the food and the test, so black pepper can be weak in lard and strong in mayonnaise. Drawn from Hirasa and Takemasa, Spice Science and Technology, Ch. 6.

Quick check

Why is rosemary extract attractive to a food maker compared with BHA and tocopherol, on the book's tests?

Check your answer

At 200 parts per million it matched BHA on lard and worked on soybean oil where BHA and tocopherol did not. Its effect rises in proportion to its concentration, while tocopherol plateaus, and it works synergistically with tocopherol, which BHA does not.

22.8 Aroma, the brain, stress and memory

The four types of brain waves, distinguished by frequency per second, are alpha waves at 8 to 13, indicating the subject is awake and resting; beta waves at 18 to 30, indicating intense nervous system activity; delta waves below 3.5, indicating deep sleep or brain disorder; and theta waves at 4 to 7, usually found in adults under stress. Aroma compounds travel through the nostrils, dissolve in the mucus secreted by the Bowman's glands in the olfactory epithelium, and are picked up by chemical receptors made of olfactory neurons and connecting olfactory cells, which send electrical signals to the olfactory bulb in the brain.

Torii measured the psychological effect of aroma with contingent negative variation, known as CNV. In the test, a subject hears a buzzer, then sees a light a few seconds later, and learns to anticipate the light and press a button when it appears. With no essential oil as the control set at 100, jasmine gave a higher CNV than the control, marking it as a stimulant, while lavender gave a lower one, marking it as a sedative.

A peppermint study tested whether aroma could keep a drowsy driver awake. Subjects on the verge of sleep were woken by a buzzer, and the degree of wakefulness fell as they became used to the buzzer sound. Then the aromas of peppermint, jasmine, lavender and lemon were blown at them. Peppermint had the strongest stimulating effect, and subjects stayed awake for more than 10 minutes after the buzzer and peppermint, against about a minute for either alone.

Aroma also touches memory and stress. Sakurai and co-workers showed 100 test subjects single, unrelated words while exposing them to one of several aromas, or to no aroma, and retested them the following day. Subjects exposed to rosemary or chocolate retained more than subjects exposed to no aroma. In mice, Saito and co-workers found that saffron improved memory scarring. In mice restrained for 15 minutes, lymph production fell, but immune function recovered when the animals were exposed to the aroma of alpha-pinene.

Quick check

In the CNV test, what did jasmine and lavender do, and what did peppermint do for a drowsy subject?

Check your answer

Jasmine raised CNV above the control of 100, marking it as a stimulant, and lavender lowered CNV, marking it as a sedative. Peppermint was the strongest stimulant for drowsy subjects, and buzzer plus peppermint kept them awake for more than 10 minutes, against about a minute for either alone.

22.9 Practical note: what a food maker can use, and what nobody should claim

Read the three uses apart, because they answer different questions. As a flavour, a spice is used by the pinch, and its effect is immediate and certain. As an antimicrobial or antioxidant in food, a spice works at a measured concentration, in a named food, against named microbes, and the book gives the numbers in Table 22.2 and section 22.7.

Try it yourself: Take two identical small jars of cooking oil. Add a sprig of fresh rosemary to one and leave the other plain. Keep both on a kitchen shelf for two weeks, then smell each. The rosemary jar should smell fresher for longer. That is the book's rosemary finding at kitchen scale, and it is a test of the oil, not a test of your health.

22.10 Check yourself: quick self-check

  1. Capsaicin is absorbed in rats, then epinephrine rises. What happens next in the liver and the fat tissue?
Check your answer

Epinephrine acts through beta-adrenergic receptors: liver glycogen breaks down to glucose, which rises rapidly in the blood, and fat breaks down to free fatty acids, which rise gradually. Both are carried to muscle and other tissues and burned, producing body heat.

  1. In the Kawada rat study, 0.014% capsaicin was added to a high-lipid feed for 10 days. What fell, and what stayed the same?
Check your answer

Perirenal adipose tissue weight and serum triglyceride concentration fell, and lipoprotein lipase activity rose. The rate of weight gain and the rate of lipid absorption did not change.

  1. Sage and rosemary were tested at 0.3% and 0.5%. What is the difference between those two results?
Check your answer

At 0.3% either spice was bacteriostatic, holding microbes still, and at 0.5% it had a bactericidal effect, killing them, in the conditions of that study.

  1. Why can garlic juice both kill E. coli and, at a much higher dilution, let it grow?
Check your answer

Allicin gives garlic its antimicrobial activity, and heating destroys the enzyme that makes allicin. At 750-fold dilution no E. coli grew, but at 15,000-fold to 150,000-fold dilution E. coli was observed, because scordinin in garlic promotes the propagation of E. coli.

  1. Name the strongest spice antioxidants in the book, and name one compound from rosemary that beats BHA.
Check your answer

Sage and rosemary are by far the most effective, stronger than BHA at the same concentration. Rosmanol from rosemary showed antioxidant activity more than four times greater than BHA and BHT.

Try it yourself: Pick three spices from your kitchen, one leafy herb such as oregano, one bark such as cinnamon, and one seed such as mustard. For each one, write down which section of this unit gives it an antimicrobial or antioxidant number, and what that number is. If a spice has no number in the unit, write that down too: a gap is a finding, not a failure to look.

Key ideas to carry forward

  • Capsaicin, piperine and zingerone are absorbed fast in rats, raise epinephrine through substance P and the sympathetic nerves, free glucose and fatty acids, and produce body heat.
  • Skin temperature rises after pungent food, then falls with sweating, sometimes below the starting point, while core temperature does not change greatly.
  • In rats, 0.014% capsaicin in a high-lipid feed for 10 days lowered perirenal fat and serum triglyceride without changing weight gain. In 12 young people, a spiced breakfast raised resting metabolic rate by 25% over 3 hours. Both are short studies, not weight-loss proof.
  • Antimicrobial strength runs from whole spice to oil to single compound, and gram-positive bacteria are generally more sensitive than gram-negative bacteria. Cinnamon at 0.02% cut aflatoxin production by 21 to 25%, and its ethanol extract at 0.02% cut aflatoxins B1, B2, G1 and G2 by 74 to 89%.
  • Sage and rosemary are the strongest spice antioxidants in the book, stronger than BHA at the same concentration. Rosmanol is more than four times as strong as BHA and BHT. Rosemary extract at 200 parts per million matched BHA on lard and worked on soybean oil where BHA and tocopherol did not.
  • Aroma changes brain waves measurably: jasmine stimulates and lavender sedates by CNV, peppermint plus a buzzer kept drowsy subjects awake for more than 10 minutes, and rosemary aroma helped memory retention in 100 subjects tested the next day.
  • No finding in this unit is medical advice, and no spice in it is recommended as a treatment for any disease.

Glossary

Capsaicin

The pungent compound of red pepper. Food tech example: the compound that raises epinephrine and body heat in the rat studies in section 22.1.

Piperine

The pungent compound of black pepper. Food tech example: 85% absorbed from rat jejunal loops in one hour, and a weaker trigger of epinephrine than capsaicin.

Zingerone

A pungent compound of ginger. Food tech example: 95% absorbed from rat jejunal loops in one hour.

Thermogenesis

Production of body heat, here driven by pungent compounds through the sympathetic nerves and the adrenal gland. Food tech example: the warmth felt after a hot meal, and the rise in metabolic rate after a spiced breakfast.

Substance P

A neurotransmitter released when capsaicin activates primary afferent neurons. Food tech example: the first step in the chain from eating red pepper to epinephrine secretion.

Epinephrine

Adrenaline, a hormone secreted by the adrenal medulla that frees glucose from the liver and fatty acids from fat. Food tech example: its secretion rose sharply in rats infused with capsaicin at 200 micrograms per kilogram.

Allicin

The sulphur compound of garlic that gives its antimicrobial activity. Food tech example: garlic juice at 0.5% inactivated typhoid bacillus completely in 5 minutes, and heating destroys the enzyme that makes allicin.

Scordinin

A compound found in garlic. Food tech example: it lowered cholesterol in rabbits, and it also promotes the growth of E. coli at high dilution.

Thrombus

A blood clot formed inside a vessel by platelet aggregation and coagulation. Food tech example: the clot that ajoene and MATS from garlic hinder by preventing platelet aggregation.

Bacteriostatic

Holding bacteria still, preventing growth without killing them. Food tech example: sage or rosemary at 0.3% was bacteriostatic in the Shelef study.

Bactericidal

Killing bacteria. Food tech example: sage or rosemary at 0.5% had a bactericidal effect in the Shelef study.

Aflatoxin

A toxin made by Aspergillus moulds, including Aspergillus parasiticus and Aspergillus flavus. Food tech example: cinnamon powder at 0.02% suppressed its production by 21 to 25%.

Antioxidant

A substance that slows the oxidation of fat and oil, delaying rancidity. Food tech example: rosemary extract at 200 parts per million retarding the oxidation of lard as well as BHA did.

Contingent negative variation (CNV)

A measure of anticipatory brain waves in the frontal lobe, used to test whether an aroma stimulates or calms. Food tech example: jasmine raised CNV above the control of 100, lavender lowered it.

References

  • Hirasa, Kenji, and Takemasa, Mitsuo. Spice Science and Technology. Marcel Dekker, 1998. Chapters 5, 6 and 7 are the sources for this unit: Chapter 5, Physiological Effects of Spice Components; Chapter 6, Antimicrobial and Antioxidant Properties of Spices; Chapter 7, The Physiological Effects of Flavor/Aroma.

Unit 23: Major Spice Crops I: Black Pepper, Cardamom and Chillies

Note

Unit 23: Major Spice Crops I: Black Pepper, Cardamom and Chillies

Three spices sit at the top of the spice trade: black pepper, cardamom and chillies. One climbs a tree, one hides its flowers at the base of the plant, and one is picked again and again from the same bush. Unit 6 gave the general picture of spice production and what happens after harvest; this unit gives the crop by crop detail.

23.1 The three crops at a glance

This unit covers the first three of the five major spices; ginger and turmeric follow in Unit 24 with the seed spices.

Table 23.1: The three crops at a glance.

CropFamilyPart usedPlant habitYield of dry spice
Black pepperPiperaceaeBerry (drupe)Evergreen perennial woody vine800–1,000 kg/ha from a 7–8 year old vine
Small cardamomZingiberaceaeCapsulePerennial herbaceous rhizome plant, 2–5 m tall50–70 kg/ha on average
Large cardamomZingiberaceaeCapsulePerennial herbaceous rhizome plant, to about 2.8 m tall3–10 q/ha at 6–7 years
ChilliesSolanaceaePod (fruit)Warm season annual500–1,000 kg/ha rainfed; 1,500–2,500 kg/ha irrigated

One quintal (q) is 100 kg, so 3–10 q/ha means 300–1,000 kg/ha.

Quick check

Which crop is a climbing vine, and which two belong to the ginger family?

Check your answer

Black pepper is the climbing vine. Small cardamom and large cardamom both belong to the ginger family, Zingiberaceae. Chillies belong to Solanaceae, the same family as tomato and potato.

23.2 Black pepper: the vine and its varieties

Black pepper is Piper nigrum, of the family Piperaceae, called the king of spices. The plant is an evergreen perennial woody vine that grows to about 4 m on a support tree, pole or trellis, and can reach 10 m or more where the support allows. It roots where trailing stems touch the ground, and it is not a parasite.

The leaves are oblong, 5–10 cm long and 3–6 cm across. The flowers are small and grow on pendulous spikes, 4–8 cm long at the leaf nodes, lengthening to 7–15 cm as the fruit matures. Most flowers carry both sexes. The fruit is a drupe, and when dried it is a peppercorn.

The vine has three kinds of runner. The main stem, or primary runner, forms the permanent stem. Secondary runners are round, long shoots with long internodes that climb high and later droop down. Tertiary runners are short, sturdier branches that spread sideways. The climbing shoots make roots at every node, and those shoots are the planting material.

Growers keep many local cultivars, among them Aimpiryan, a high yielder suited to higher elevations, Arakkulammunda, Balankotta, Karimunda and Neelamundi. The best known improved variety is Panniyur-1, the most popular hybrid black pepper, with 11.8% oleoresin, 5.3% piperine and 3.5% essential oil. Piperine gives pepper its bite, and the oleoresin holds the flavour.

23.3 Black pepper: climate, soil and planting

Black pepper needs about 250 rainy days and 2,000–3,000 mm of rain a year, growing up to 1,500 m above sea level and thriving best at 500 m or below. The crop tolerates 10–40°C, ideally 23–32°C, with humidity of 75–95%. One dry spell matters most: a dry spell of 30–45 days before flowering is needed for successful fruit setting. Waterlogging harms the shallow root system and favours foot rot.

The soil should be well drained and rich in organic matter. The vine grows on a wide range of soils, from heavy clay to light sandy clays, provided they are friable, rich in humus and hold water well without staying wet, thriving best on red laterite or alluvial soils rich in humus. A pH of 4.5–6.0 is ideal, and soil with a pH above 7.5 holds back growth.

Propagation is by rooted cuttings. Plants raised from seed grow slowly and do not always breed true. One or two leaf cuttings are taken from secondary runners in September, rooted in mist beds and transplanted after about 9 months.

Planting is done at the onset of the monsoon. Two or three rooted cuttings are planted together on the northern side of the standard, or support tree, in a pit 50 × 50 × 50 cm dug 30 cm away from the standard. The pit is filled with topsoil and 10 kg of farmyard manure, with 1 kg neem cake and 70 g rock phosphate mixed in 15 days before planting. The cutting is set semi-horizontally, at an angle of 30°–45° towards the standard, not upright. Recommended spacing is 2.7 × 2.7 m, giving about 1,100 vines per hectare in a monocrop and 540–560 vines per hectare in a mixed crop.

The standards are planted one or two years before the pepper, most often Erythrina indica, with silver oak, Gliricidia sepium and Ailanthus species also used. The vine is tied to the standard as it grows, vines are pruned to one main shoot and two lateral upright shoots, and standards are pruned each March–April and kept to about 6 m.

23.4 Black pepper: feeding, watering and field care

Pepper is a surface feeder, so fertiliser must be placed where its shallow roots can reach it.

Each vine gets 10 kg of compost or farmyard manure in May, with nitrogen 100 g, P₂O₅ 40 g and K₂O 140 g per year, half in April and half in August–September, plus lime at 500 g in April–May in alternate years where the soil is acidic.

One third of the dose is given in the first year, two thirds in the second, and the full dose from the third year onwards, spread 30 cm around the vine at 15 cm deep.

Watering follows the rain. Where vines are trailed on Erythrina indica, the books give drip irrigation at 7 litres of water per day per vine from October to March. Good drainage is part of watering: the root system is shallow and waterlogging favours foot rot. Shade is regulated by lopping the standards, shade in hot weather and sunlight in cool weather, and mulch of green leaves, sawdust or coir dust is given towards the end of the north-east monsoon.

23.5 Black pepper: pests and diseases

Five insect pests are named. The pollu beetle (Longitarsus nigripennis) is the most destructive pest of black pepper and is worse in the plains and below 300 m altitude. Adults feed on tender leaves and spikes; the females lay eggs on tender spikes and berries, and the grubs bore in and feed on the inner tissues, so infested spikes turn black and decay and infested berries turn black and crumble when pressed. The leaf gall thrips (Liothrips karnyi) makes tubular galls on the leaves. The scale insect (Lepidosaphes piperis), the top shoot borer (Cydia hemidoxa) and the mealy bug (Planococcus minor) complete the list.

The diseases named are foot rot, leaf rot and blight, and basal wilt. Foot rot attacks the collar and roots in wet soil, and it is why the books insist on drainage.

23.6 Black pepper: harvest and yield

The vine flowers in May–June and takes 6–8 months from flowering to ripening, with 30–45 days more at high elevations. Vines start yielding in the third or fourth year and reach full bearing in the seventh or eighth year. Harvest runs from November to February in the plains and January to March in the hills.

The picking rule is simple and visual: when one or two berries on a spike turn red, the whole spike is plucked. Picking before proper maturity cuts the yield by 15–20%. The stage picked decides the product. For black pepper the berries are gathered young, while still green. For white pepper the berries are harvested at a slightly more advanced stage, when they turn red or bright orange, and the outer skin is then removed. The berries are removed by rubbing, scrubbing or trampling and dried in the sun for 3–4 days, turned over as they dry, until the skin turns dark brown to black and shrivels.

A full bearing vine yields about 1 kg of dry pepper, and a seven or eight year old vine gives 800–1,000 kg/ha of black pepper. About 100 kg of fresh berries gives 26–39 kg of black pepper of commerce.

Black pepper: one year on the vine

May–June
Vine flowers on pendulous spikes

↓

6–8 months
Berries fill and ripen

↓

November–February
Pick the spike when 1–2 berries turn red

↓

Green berries: black pepper
Red berries: white pepper, skin removed

↓

Sun dry 3–4 days
100 kg fresh berries give 26–39 kg black pepper

Figure 23.1: Black pepper: one year on the vine. The picking signal is one or two red berries on a spike, and the stage picked decides black or white pepper. Drawn from Das, Spice Crops Production Technology, Black Pepper chapter, and Dashora et al., Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants, Black Pepper chapter.

23.7 Cardamom: two cardamoms, one name

The name cardamom covers two different crops. Small cardamom is Elettaria cardamomum, called the queen of spices. Large cardamom is Amomum subulatum. Both belong to the ginger family, Zingiberaceae, and both are grown for the capsule. They differ in size, climate and handling, and mixing them up is the most common mistake a reader makes with this crop.

Small cardamom is a perennial herbaceous plant, 2–5 m tall, with an underground rhizome as its true stem and an aerial pseudostem of encircling leaf sheaths. The leaves are long, alternate and lance-shaped. The flowers are bisexual, arranged in clusters called cincinni on long panicles that rise straight from the underground stem on long floral stalks. Flowering runs through the year, peaking in May–June, and the fruit takes 5–6 months from flowering to maturity. The plant starts bearing in 2–3 years and reaches full bearing at seven or eight years. Small cardamom is grown as three types: Malabar, early maturing; Mysore, standing sun better; and Vazhukka, with dark green capsules.

Large cardamom is a taller forest crop, with plants to about 2.5 m, and one variety, Bharlangey, to 2.8 m. The fruit is on average 70% seeds and 30% skin, and the seeds carry 2.8% volatile oil. Its varieties are Ramshai, tall with small fruits; Sawney, with bold brown fruits, favoured below 1,500 m because it suffers less from Furkee virus disease; Golsey, with bold roundish capsules of 40–50 seeds; and Bharlangey, with bold elliptic capsules of 40–60 seeds.

Released varieties range from IISR Coorg Suvasini at 409 kg/ha to Njallani Green Gold at 1,600 kg/ha, with PV-2 at 982 kg/ha and 10.45% essential oil. IISR Avinash is noted for rhizome rot infested areas, and a sister selection, IISR Vijetha, for mosaic infested areas.

Quick check

A trader offers you "cardamom" and the capsules are large, dark pink brown and smoky smelling. Which cardamom is it, and why does it smell smoky?

Check your answer

Large cardamom. Its dried capsules are dark pink brown with a smoky odour, because large cardamom is cured in a kiln over heat, while small cardamom is dried to keep its green colour.

23.8 Cardamom: climate, soil and planting

Small cardamom wants a cool, constantly moist, never waterlogged forest climate. The books give a temperature range of maximum 14–33°C and minimum 4–22°C, with annual rainfall of 150–400 cm well spread through the year, at an altitude of 600–1,675 m. That altitude and rainfall are the range across the books: Dashora et al. give 600–1,500 m and 150–400 cm, while Das and Major Spices of India give 765–1,675 m and 200–250 cm. The soil is acidic forest loam, pH 4.2–6.8, well drained and full of humus.

Large cardamom grows from 600 to 1,800 m on slopes under the canopy of tall evergreen forest trees. It needs high humidity and cool surroundings with moderate shade, an optimum temperature of 6–25°C and annual rainfall of 2,000–3,500 mm. It is highly sensitive to wind and drought. Its soil is well drained forest soil rich in humus, never waterlogged, pH 5.8–6.5.

Propagation of small cardamom is by sucker and by seed. For vegetative propagation a section of rhizome called a bulb, with not less than three shoots, is taken from healthy 1.5–2.0 year old, regularly high yielding mother clumps and planted at 0.6 m spacing in trenches in March–October. Plants raised this way come to bearing one year earlier than seedlings. For seed, ripe capsules are collected in September–October and sown at once, since fresh seed gives uniform and early germination; seedlings move to a second nursery for at least a year.

Large cardamom is propagated the same two ways, by rhizome slips and by seed. Vegetatively raised plants come to yield two years earlier than seed raised plants, and germination from seed is only 30–50%. Seedlings are raised in a primary nursery, then a secondary nursery, and are ready for the main field at 12 months.

Planting of small cardamom puts one or two seedlings in the main field in June–July or in September–October, 2–3 inches deep, the soil pressed well around them, and the seedling tied to a stout wooden stake against monsoon rain and wind. Pits measure 45 × 45 × 45 cm and are partly filled with rich surface soil and compost or farmyard manure. Spacing is 2 m × 2 m. Large cardamom is planted in June–July, in pits 30 × 30 × 30 cm at 150 × 150 cm for tall varieties and 90 × 90 cm for small varieties, with 100 g of rock phosphate per pit.

23.9 Cardamom: feeding, watering and field care

Small cardamom is a surface feeder: fertiliser is spread at a radius of 30 cm and covered with a thin layer of soil. Compost or farmyard manure is given at 10–15 t/ha. The fertiliser recommendation is 30 kg nitrogen, 30 kg P₂O₅ and 60 kg K₂O per hectare under rainfed conditions, and 75 kg nitrogen, 75 kg P₂O₅ and 150 kg K₂O per hectare under irrigated conditions. The dose is split between May and late September. Only half the dose is given in the first year and the full dose from the second year onwards.

On depleted soils large cardamom gets 20 kg nitrogen, 30 kg P₂O₅ and 40 kg K₂O with 30 tonnes of farmyard manure or compost per hectare each year; virgin forest land needs no fertiliser. Phosphorus goes on as a full dose in April, nitrogen and potassium in three equal splits in April, June and September. One third of the dose is given in the first year, two thirds in the second and the full dose from the third year onwards.

Small cardamom is irrigated every 10–15 days from the last week of January until the monsoon, since irrigation helps panicle initiation, flowering and fruit set, and sprinkler and drip irrigation are found beneficial. Large cardamom does not tolerate drought: a continuous dry spell of about two months can cause heavy mortality, so plantations are irrigated from December to April at 10 day intervals.

Weeding is done in May–June, August–September and December–January in the first years, then 1–3 times a year as needed, and the removed weeds serve as mulch. Mulching with fallen shade tree leaves is done in November–December to soften summer drought. Trashing, the removal of old and drying shoots, is done once a year with the onset of monsoon rains under rainfed conditions and 2–3 times under irrigation. From the first week of June the panicles are lifted above the mulch layer so bees can visit them for pollination.

23.10 Cardamom: pests and diseases

The thrips (Sciothrips cardamomi) is the most destructive and persistent insect pest of small cardamom. Adults and nymphs damage the panicles, flowers and capsules. The shoot and panicle borer (Conogethes punctiferalis), the capsule borers (Jamides species), the beetle borer, the root grub (Basilepta fulvicorne and Holotrichia serrata), hairy caterpillars (Eupterote cardamomi, E. canarica and E. fabia), the white fly (Dialeurodes cardamomi) and the cardamom aphid (Pentalonia nigronervosa) are also named. The diseases named for small cardamom are azhukal, or capsule rot, clump rot and damping off, rhizome rot, leaf blight and mosaic. For large cardamom the disease named is Furkee virus, worst at lower altitudes.

23.11 Cardamom: harvest and yield

Small cardamom plants start bearing in two to three years and reach full bearing at seven or eight years. The peak harvest is October–November, with picking every 15–30 days.

Each capsule is picked just before it is fully ripe, when the seed inside has turned brownish black and the skin is still green. Fully ripe fruits split on drying and lose their colour; immature fruits give a shrivelled, badly coloured sample. The capsules are picked with their stalks, and cutting off whole fruiting shoots and stripping the capsules is criticised in the books as giving a poor sample.

After harvest the fleshy capsules must be cured before marketing. They are spread thinly and stirred often for even drying, in a fuel kiln, an electric drier or the sun. In a drier the temperature is 45–50°C for 14–18 hours, and in a kiln overnight drying at 50–60°C is used. Curing cuts the moisture of green cardamom from 80% to 8–12% while keeping the green colour. The green colour can be protected by soaking freshly harvested capsules in a 2% washing soda solution, Na₂CO₃, for 10 minutes before drying.

Large cardamom is different. Seed raised plants fruit after five years and vegetatively raised plants after three years. Flowering runs from April–May to June–July, and the fruits are ready in August–September. Harvesting is done when the seeds of the topmost capsule turn brown, by cutting the mature panicles with a knife. The fruits are shelled and cured in a kiln in the plantation, on a bamboo mesh over low heat, taking about three days. Dried recovery is only 33% of the fresh weight, and the dried capsules are dark pink brown with a smoky odour.

Yield figures differ between the books. For small cardamom Das gives an average of 50–70 kg/ha of dry capsules, and Dashora et al., in the Small Cardamom chapter, give 200–350 kg/ha from a fully grown healthy plantation. For large cardamom the yield is less than 25 kg/ha from a three year old plantation, rising to a maximum of 3–10 q/ha in the sixth or seventh year. Plants can yield for 25–30 years, but the economic period is the fifth to tenth year, after which yield falls and replanting is done gradually after the ninth year.

Small cardamom and large cardamom: tell them apart

Two different crops side by side. No arrow joins them: neither one becomes the other.

Small cardamom
Elettaria cardamomum
Green capsule, picked just before ripe
Altitude 600–1,675 m, rain 150–400 cm (range across books: see section 23.8)
Cured at 45–50°C to keep the green colour
Large cardamom
Amomum subulatum
Dark pink brown capsule, smoky odour
Altitude 600–1,800 m, rain 2,000–3,500 mm
Cured in a kiln over low heat for about 3 days

Figure 23.2: Small cardamom and large cardamom: tell them apart. Two species, two climates and two curing methods, under one common name. Drawn from Das, Spice Crops Production Technology, Cardamom chapter, and Dashora et al., Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants, Large Cardamom and Small Cardamom chapters.

23.12 Chillies: the plant and its varieties

Chillies are Capsicum frutescens, of the family Solanaceae, and the books also name Capsicum annuum for the capsicum group. The fruit is a pod, and it is the part used. Bird chillies, or tabasco chillies, are Capsicum frutescens, and paprika types sit in the same group. The crop is grown for green pods eaten fresh, for red pods dried and ground, and for extraction of its pungent and colour compounds.

Improved varieties named in the books include NP 46, Pusa Jwala and G-3. The books also list varieties for different uses: some for eating fresh, some for pickles, some for export and some for extraction of capsaicin, the pungent compound. "Chilli" is not one crop with one harvest: the variety and the intended use decide when it is picked and how it is handled.

23.13 Chillies: climate, soil and planting

Chillies grow in tropical and subtropical regions in a warm and humid climate at 20–25°C. Low moisture in the soil during blossom development and fruit formation makes the buds wither and the fruit drop. Excessive rainfall is harmful because it brings defoliation and rotting of the plant. Sweet capsicum grows at the same 20–25°C with moderate rainfall of 60–120 cm a year.

Sandy loam, clay loam and loam suit chillies best. The soil must be well drained and well aerated, and acidic soils are not suitable. For capsicum, sandy and loam soils are best, with soil pH 5.5–6.8.

The land gets 2–3 ploughings, and compost or farmyard manure at 10–12 tonnes per hectare is mixed in at least 15–20 days before planting. Seed is sown almost all year round: summer crops in January–February, winter crops in July–August, hybrids in August–September and December–January. The seed rate is 1,000–1,250 g/ha for general varieties, 300–325 g/ha for hybrids and 650–750 g/ha for open pollinated varieties.

Seedlings are raised in well prepared raised nursery beds with decayed leaf mould or farmyard manure. Seed is sown 0.5 cm deep, covered with fine soil and then with straw or an old gunny bag, removed as soon as germination starts. Watering is by hand with a can, every day except on rainy days, and overwatering must be avoided, since it brings the fungal disease called damping off, which kills the seedlings.

Seedlings 6–8 weeks old and 15–20 cm high are ideal for transplanting. Topping the seedlings in the nursery bed on a sunny morning 8–10 days before transplanting makes thicker seedlings that root quickly, and it also keeps off the curl disease caused by chilli thrips on tender leaves. Spacing is 45–60 cm between rows and 45 cm between plants, varying with variety, season and soil fertility.

23.14 Chillies: feeding, watering and field care

The books give two feeding schedules for chillies: one for general varieties and one for hybrids. Both start with farmyard manure and Azophos at 15 kg/ha, plus zinc sulphate 25 kg/ha, borax 10 kg/ha and ammonium molybdate 0.5 kg/ha.

For general varieties, farmyard manure is 10 t/ha, with nitrogen 90 kg/ha, P₂O₅ 60 kg/ha and K₂O 60 kg/ha. All the manure, Azophos and micronutrients go on as a basal dressing with all the phosphorus and half the nitrogen and potassium. One fourth of the nitrogen and potassium is top dressed at 21 days after planting and the rest at 40 days after planting. For hybrids, farmyard manure is 25 t/ha, with nitrogen 135 kg/ha, P₂O₅ 50 kg/ha and K₂O 75 kg/ha. The basal dressing carries all the phosphorus with 45 kg of nitrogen and 35 kg of potassium, and the rest is top dressed in five equal splits of 18 kg nitrogen and 8 kg potassium at 45, 60, 75, 90 and 105 days after planting.

Watering is frequent and even. The first irrigation is given after transplanting, then at 5–7 day intervals in summer and the rainy season and every 10–15 days in winter. Even soil moisture is essential to prevent blossom and fruit drop, which rises with high temperature, low moisture and shading. Spraying with Planofix, that is NAA, at 10 ppm at flower initiation and again 15 days later at peak flowering reduces flower drop and improves fruit set.

Field care is 2–3 shallow hoeings in the early stages to kill weeds and give a soil mulch. Chillies are also grown with brinjal as a companion crop.

23.15 Chillies: pests and diseases

Four insect pests are named. Chilli thrips (Scirtothrips dorsalis) attacks the tender leaves and causes leaf curl. The pod borer (Spodoptera litura and Spodoptera exigua) bores into the pods. Aphids (Aphis gossypii) suck sap from tender parts, and mites (Hemitarsonemus latus) attack the foliage. The diseases named are damping off in the nursery, bacterial leaf spot or bacterial blight, anthracnose, leaf curl, dieback, chilli mosaic, powdery mildew and fruit rot.

Try it yourself: take three chillies: one green, one red and one dried. Write one line for each: what stage was it picked at, and what would you use it for? Check your lines against section 23.16.

23.16 Chillies: harvest and yield

Chillies are picked green but full grown for the vegetable, red for canning, and full ripe for drying. The crop is ready in about 60–70 days in the north and 90–105 days in the south, with picking continuing for about two months in about six pickings.

The yield of dry chillies is 500–1,000 kg/ha from a rainfed crop and 1,500–2,500 kg/ha from an irrigated crop. The proportion of dry to fresh ripe chillies varies from 25–40%.

Practical note: three picking tests you can do with your eyes

Three picking tests use sight alone. On a pepper vine, one or two red berries on a spike is the signal to pick the whole spike. On a cardamom clump, the seed should be brownish black and the skin still green. On a chilli plant, decide the use before you pick: green for the vegetable, red for canning, full ripe for drying.

What the books do not settle

Yields differ between books, as section 23.11 shows, and no area or production statistics are given here, since this unit uses the general growing science and leaves country figures out.

Key ideas to carry forward

  • Black pepper is a climbing vine that needs a support tree, a monsoon planting, a ring of fertiliser 30 cm from the stem and a picking signal of one or two red berries on a spike.
  • Small cardamom and large cardamom are two different species with different climates and different curing: green capsules dried at 45–50°C against dark pink brown capsules cured in a kiln.
  • Cardamom is a shade plant and a surface feeder: shade, mulch, trashing and shallow fertiliser placement are the core jobs.
  • Chillies are picked by purpose: green for the vegetable, red for canning, full ripe for drying, with dry weight 25–40% of fresh weight.

Glossary

Drupe

A fleshy fruit with a single seed inside. Food tech example: a black pepper berry is a drupe, and when dried whole it becomes a peppercorn.

Standard

A support tree or pole for a climbing crop. Food tech example: a pepper vine trained on an Erythrina standard, pruned to about 6 m.

Piperine

The compound that gives black pepper its bite. Food tech example: Panniyur-1 pepper carrying 5.3% piperine, the bite factor buyers test for.

Oleoresin

A concentrated extract of a spice holding its volatile oil and pungent compounds. Food tech example: pepper oleoresin at 11.8% in Panniyur-1, carrying the flavour of the berry.

Capsule

The dried fruit of cardamom, holding the seeds inside. Food tech example: small cardamom capsules picked just before ripeness, when the seed inside is brownish black and the skin is still green.

Trashing

Removing old and drying shoots from a cardamom clump. Food tech example: trashing in May–June to let new shoots grow and bees reach the panicles.

References

  • Das, P.C. Spice Crops Production Technology. Scientific Publishers, Jodhpur, 2014. Chapters: 1 Cardamom, 2 Black Pepper, 25 Capsicum, 26 Chilli.
  • Pruthi, J.S. Major Spices of India: Crop Management and Post-Harvest Technology. Indian Council of Agricultural Research, New Delhi, 1993. Chapters: 2 Black Pepper, 3 Cardamom (Small and Large), 4 Chillies or Capsicums.
  • Dashora, L.K., Dashora, Abhay, and Lakhawat, S.S. Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants. 2nd edition, Agrotech Publishing Academy, Udaipur, 2013. Chapters: 9 Black Pepper, 10 Large Cardamom, 11 Small Cardamom.

Unit 24: Major Spice Crops II: Ginger, Turmeric and Seed Spices

Note

Unit 24: Major Spice Crops II: Ginger, Turmeric and Seed Spices

Ginger and turmeric are rhizome spices: the spice is the underground stem, planted as pieces with a live bud. The seed spices are different again: the spice is the dry fruit of a small plant, usually sown as seed in the cool season and harvested when the seed head turns colour. This unit covers ginger, turmeric and nine seed spices: coriander, cumin, fennel, fenugreek, dill, anise, ajwain, celery and mustard. For the general picture of how spices are dried and processed, see Unit 6.

🤔 Pause and think: ginger is planted as a piece of rhizome, turmeric the same, and coriander as a seed. What does that mean for how much planting material a grower must keep back from each harvest?

24.1 Ginger and turmeric at a glance

Ginger and turmeric look alike in the field and in the books: both are rhizome crops of the family Zingiberaceae, both are planted in May–June, both are mulched and earthed up at 45 and 90 days, and both are harvested at 8–9 months when the leaves turn yellow. They differ in what the rhizome is for: ginger is valued for its pungent volatile oil and is sold fresh, dry or as oil, while turmeric is valued for its yellow colour, curcumin, and is sold as a boiled, dried, polished rhizome.

Table 24.1: Ginger and turmeric at a glance.

GingerTurmeric
Botanical nameZingiber officinaleCurcuma longa
Plant heightAbout 1 m60–90 cm
Planting materialRhizome bits 20–25 gMother and finger rhizomes 35–45 g
Seed rate1,500 kg/ha1,500–2,500 kg/ha
HarvestFresh from 5 months; dry at 8–9 months7–9 months by variety
Yield12–15 t/ha fresh; dry recovery 20–22%4–5 t/ha cured; fresh 17.4–42.0 t/ha by variety

24.2 Ginger: plant, climate, soil and planting

Ginger is Zingiber officinale, of the family Zingiberaceae. The rhizome is the spice: its inner core is pale yellow to bluish and the outer part light yellow. The pseudostems die every year and new shoots grow from the rhizome. The flowers are yellow with dark purplish spots on a spike that grows from the rhizome, and the fruit is rarely produced and does not set seed, so ginger is always grown from rhizome pieces. Cultivars named in the books are Assam, Nadia, Karakkal and Taffingiva.

Ginger needs a warm, humid climate and plenty of rain: 125–250 cm of rain, 28–35°C, and it can be grown up to 1,500 m. It needs deep, well drained, friable loam rich in humus, and light sandy, sandy loam and loam soils suit it best. Planting is in the first fortnight of May under rainfed conditions, when the pre-monsoon showers start, and in mid-February to early March under irrigation. Rhizome bits of 20–25 g are planted 20 × 20 cm to 25 × 25 cm apart, 4–5 cm deep, with the bud pointing up. Seed rhizomes are treated before planting with Carbendazim, Captan, Mancozeb or Trichoderma viride and kept in shade for at least 24 hours. The first mulching is done at planting, with green leaves 10–12 t/ha or dry leaves 5–6 t/ha.

24.3 Ginger: feeding, watering, pests and harvest

Feeding is farmyard manure 25–30 t/ha as a basal dressing, with nitrogen 100–120 kg/ha, P₂O₅ 75–80 kg/ha and K₂O 100–120 kg/ha. Half the nitrogen and potash with all the phosphorus goes on as a basal dose, and the rest of the nitrogen and potash is split at 45 and 90 days after planting. Sulphur is applied at land preparation to correct deficiency. Irrigation is critical at three stages: sprouting, rhizome initiation and rhizome development. Under irrigation there are 2–4 waterings at 7 day intervals in April–May, then water every 15 days from mid-October to the end of December.

Weeding is done twice, earthing up at 45 and 90 days, and mulching repeated at 7.5 t/ha at 45 and 90 days. The pests named are the shoot borer (Conogethes punctiferalis), the leaf roller (Udaspes folus), the rhizome scale (Aspidiella hartii) and thrips (Panchaetothrips indicus). The diseases named are soft rot or rhizome rot, leaf spot and bacterial wilt.

Harvest depends on the use. Fresh ginger can be taken from the fifth month, when it is tender with low pungency and fibre. For dry ginger and oil the crop is taken at full maturity, 8–9 months, when the leaves turn yellow; green ginger is taken within 195 days and dry ginger at 245–255 days, not beyond 260 days, since fibre rises and volatile oil falls after that.

Ginger: from planting to harvest

Plant in May
Rhizome bits 20–25 g, 4–5 cm deep, bud up; mulch at planting

↓

Feed, weed and earth up at 45 and 90 days
Split nitrogen and potash; mulch again 7.5 t/ha each time

↓

Fresh from 5 months, dry at 8–9 months
Leaves turn yellow; dry ginger at 245–255 days, not beyond 260

Figure 24.1: Ginger: from planting to harvest. The same 45 and 90 day rhythm that turmeric follows: feed, weed, earth up and mulch. Drawn from Das, Spice Crops Production Technology, Ginger chapter.

Quick check

Why must dry ginger be harvested by 260 days, and what goes wrong if it is left longer?

Check your answer

After 260 days the fibre rises and the volatile oil falls, so the dry ginger loses quality. The window for dry ginger is 245–255 days.

24.4 Turmeric: the plant and its varieties

Turmeric is Curcuma longa, of the family Zingiberaceae. The spice is the rhizome, boiled, dried, cleaned and polished. The plant is 60–90 cm tall, with 7–12 leaves whose blades are 30–40 cm long and 8–12 cm wide, and a central spike 10–15 cm long carrying about 30 flowers. The rhizome has a central mother rhizome and fingers that branch off it.

The books give a variety table that shows how much varieties differ. Three varieties in the books' table make the point: IISR Prathibha (39.1 t/ha, 188 days, 6.2% curcumin), Rajendra Sonia (42.0 t/ha, 225 days, 8.4% curcumin) and Roma (20.7 t/ha, 250 days, 9.3% curcumin). Roma has the highest dry recovery rate (31%) and the most curcumin (9.3%), while Rajendra Sonia gives the most cured spice per hectare (about 7.6 t/ha) because its fresh yield is twice Roma's.

Table 24.2: Three turmeric varieties, as printed in the books.

VarietyFresh yield (t/ha)DaysDry recovery (%)Curcumin (%)Oleoresin (%)Oil (%)
IISR Prathibha39.118818.56.216.26.2
Rajendra Sonia42.022518.08.415.05.0
Roma20.725031.09.313.24.2

24.5 Turmeric: climate, soil and planting

Turmeric needs 20–35°C and at least 1,500 mm of rain, from sea level to 1,200 m, and growth stops below 20°C. The soil should be sandy loam to clay loam or alluvial, rich in organic matter and of uniform texture; it cannot stand water stagnation or alkalinity, and well drained sandy loam suits it best. The soil pH range is 4.5–7.5. The land is deep ploughed and planked.

Planting material is whole or split mother and finger rhizomes of 35–45 g, healthy and disease free. Seed treatment before planting uses Carbendazim, Captan, Mancozeb or Trichoderma, with shade for at least 24 hours. Planting is in May–June or July–August, or April–May where rain comes early: short duration varieties in the second fortnight of May, medium in the first fortnight of June, long in the second fortnight of June to the second fortnight of July. The pit method uses pits 25 × 30 cm with cattle manure or compost. Spacing is 45–60 cm between rows and 25 cm between plants, or 45 × 22.5 cm on black heavy soils, 30 × 15 cm on red loamy soils, and 25–35 cm each way on beds. Emergence takes 2–4 weeks.

24.6 Turmeric: feeding, watering, pests and harvest

Feeding is farmyard manure 25–30 t/ha, with nitrogen 60 kg/ha, P₂O₅ 50 kg/ha, K₂O 120 kg/ha and zinc 5 kg/ha. All 50 kg of P₂O₅ goes on as a basal dose, and 30 kg nitrogen with 60 kg K₂O is top dressed at 45 days after planting and again at 90 days after planting, finishing within 120 days, with light hoeing, earthing up and irrigation after fertilising. Irrigation starts before or after planting and again 3–4 days after planting, then every 8–10 days: about 15–23 irrigations on clayey soils and 40 on sandy loam, avoiding water stagnation. Weeding is done three times, at 60, 90 and 120 days after planting. Mulching is green leaves 12–15 t/ha at planting, repeated at 7.5 t/ha at 45 and 90 days after planting, the same rhythm as ginger.

The pests named are the same four as ginger: shoot borer, leaf roller, rhizome scale and thrips. The diseases named are soft rot or rhizome rot, leaf spot and bacterial wilt.

Harvest is by variety: early varieties at 7–8 months, medium at 8–9 months, late at 9 months, in January–March. The land is ploughed, the clumps lifted with a spade, and the rhizomes handpicked. Yield is 4–5 tonnes of cured turmeric per hectare.

24.7 The seed spices: one habit, nine crops

Most seed spices belong to one family, Apiaceae, the old Umbelliferae: coriander, cumin, fennel, dill, anise, ajwain and celery. Their flowers grow in umbels, flat topped clusters like tiny umbrellas, and the spice is the dry fruit, usually called the seed. Two of the nine sit outside that family: fenugreek is a legume, Fabaceae, and mustard is a brassica, Brassicaceae. All nine are short season crops, most of them sown in the cool season and harvested in 110–180 days, and all of them are judged at harvest by the colour of the seed head.

Table 24.3: Seed spices: seed rate and yield as printed in the books.

CropSeed rate (kg/ha)Yield
Coriander15–20 for seed; 40 for green leaves500–800 kg/ha rainfed; 1,200–2,000 kg/ha irrigated
Cumin12–15600–700 kg/ha
Fennel2–2.5 kharif; 2.5–3.0 rabi1,800–2,000 kg/ha kharif; 1,500–1,800 kg/ha rabi
Fenugreek10–151,800–2,000 kg/ha
Dill3–46–9 q/ha
Anise5–10600–700 kg/ha
Ajwain4–512–20 q/ha irrigated; 4–6 q/ha unirrigated
Celery21,000–1,200 kg/ha
Mustard and rapeseed6.25–7.5 pure crop; 2.5–3.75 mixed crop1,250–2,000 kg/ha by variety

Notice how little seed these crops need against 1,500 kg/ha of rhizome for ginger: the seed is small, the plant is small, and the yield is counted in hundreds of kilos, not tonnes.

24.8 Coriander and fenugreek

Coriander is Coriandrum sativum, of the family Apiaceae: a thin stemmed bushy herb, 25–50 cm tall, with umbels of five smaller umbels. The seeds are pale white to light brown, globular, 3–4 mm across, and break into two locules when pressed. The green leaves are a herb and the dry fruits the spice. It is a tropical crop grown in the cool season, free from severe frost at flowering and seed formation: cool weather suits growth, warm dry weather suits maturity, and seed quality and oil are better in colder regions. Optimum temperature is 18°C. Sandy loam and clay suit it best; it does not tolerate alkaline and saline soils, and the suitable pH is 4.5–8.0, optimum 6.3.

Sowing is 15th October to 15th November. Seed is rubbed to split it in two and soaked overnight. Feeding differs by water: rainfed, farmyard manure 15–20 t/ha with N20 P₂O₅ 30 K₂O 20 kg/ha; irrigated, farmyard manure 20 t/ha with N60 P₂O₅ 30 K₂O 20 kg/ha. One third of the nitrogen with all the phosphorus and potash goes on basal, the rest top dressed in two splits. Weeding is at 25–30 days rainfed and 40–45 and 50–60 days irrigated, with thinning to 5–10 cm. Irrigation runs to 4–6 waterings, the first within 30–35 days, then at 50–60, 70–80, 90–100, 105–110 and 115–125 days.

Green leaves are cut before flowering, from about 20 cm. For seed, harvest is at full maturity, when 50% of the seeds turn yellow, at 110–140 days; the crop is dried in partial shade, the seeds beaten out, winnowed and graded. Varieties named include Gujarat Coriander-1 and CO-1. The pest named is the coriander aphid, the most destructive insect of coriander, and the diseases are wilt, powdery mildew, stem gall and bacterial leaf spot.

Fenugreek is Trigonella foenum-graecum, of the family Fabaceae: a cool season legume that needs cool weather during growth and warm dry weather at maturity, and tolerates frost. Clayey loam suits it best, optimum pH 6.0–7.0. Sowing is mid-October to the first fortnight of November, the seed soaked 6–8 hours and dried in shade; germination takes 6–8 days. Feeding is farmyard manure 10–15 t/ha with N40 P₂O₅ 40 kg/ha: all the phosphorus and half the nitrogen basal, the rest top dressed at 30 days. Weeding is at 25–30 and 40–50 days, thinning to 8–10 cm. Irrigation runs to 4–6 waterings, the first at thinning and then every 20–25 days, avoiding stress at pod and seed development. Green leaves are nipped 2 cm above the ground after 3–4 weeks, with 2–3 cuttings. For seed, the crop is ready in 120–150 days, when leaves and pods turn yellow; the plants are dried, threshed by beating or trampling, and winnowed. Varieties named include Kasuri and Rajendra Kranti (RM-16). The pest is the aphid, and the diseases are powdery mildew, downy mildew and bacterial blight.

24.9 Cumin, fennel and dill

Cumin is Cuminum cyminum, of the family Apiaceae, a tropical plant that also grows well in subtropical climates, in a dry cool climate free from frost. It grows in all soil types, but well drained sandy loam and medium soils suit it best. Sowing is in the first week of December. Seed is soaked for 8 hours before sowing, dried in shade, and sown by broadcasting or drilling in lines 30 cm apart; deep sowing harms germination. Feeding is farmyard manure 15–20 t/ha with nitrogen 30 kg/ha and P₂O₅ 15 kg/ha: all the phosphorus and half the nitrogen basal, the rest top dressed at 30 days after sowing. Weeding needs 2–3 hand weedings, the first with hoeing at 30–40 days. Irrigation runs to 4–6 waterings: the first light irrigation right after sowing, the second after 6–10 days, then after 30, 45, 65 and 80 days. Irrigation at flowering and fruit set is essential, and it is stopped at maturity. The crop matures in 110–115 days, when the plants turn yellowish brown. It is notorious for ripening unevenly, so the plants are harvested as soon as one seed head is dry enough to crack when pinched, early in the morning, by cutting or uprooting the whole plants. The seeds are threshed out with sticks, dried to 10% moisture, and winnowed. Varieties named include Gujarat Cumin-1 and GC-2. The pests named are the aphid and the leaf eating caterpillar, and the diseases are Fusarium wilt, Alternaria blight and powdery mildew.

Fennel is Foeniculum vulgare, of the family Apiaceae: a hardy perennial umbelliferous herb, erect and glaucous green, growing to 2.5 m with hollow stems. The flowers grow in terminal compound umbels 5–15 cm wide, each with 20–50 tiny yellow flowers, and the fruit is a dry seed 4–10 mm long. Fennel needs a cool dry climate: frost prone areas must be avoided, and rain at maturity spoils the colour and lowers the quality of the seed. It is grown as a kharif crop and as a rabi crop. Well drained loamy soil and black cotton soils suit it. Sowing for the kharif crop is in the nursery in the middle of June, and for the rabi crop in the last week of August to the first week of September. Seedlings are transplanted at about 15 cm height: kharif at 90 × 60 cm or 100 × 60 cm around 15th August, rabi at 45 × 10 cm. Feeding for kharif is farmyard manure 20 t/ha with nitrogen 100 kg/ha and P₂O₅ 60 kg/ha: all the phosphorus and 40 kg nitrogen basal, and the remaining 60 kg nitrogen in two splits of 30 kg at 30 and 60 days after transplanting. For rabi it is nitrogen 90 kg/ha and P₂O₅ 45 kg/ha, with the remaining 45 kg nitrogen in two splits of 22.5 kg at 30 and 60 days. Hoeing is done three times from 20–25 days after transplanting, and earthing up at 60–75 days checks lodging. Irrigation is 10–12 waterings for kharif and 8–10 for rabi, the first on the day of transplanting and then at 15–20 day intervals up to February, with irrigation at seed development critical. The umbels are cut as they reach physiological maturity, when the seeds are fully filled and still green, and dried in partial shade. The crop matures in 170–180 days, with umbels plucked in 3–4 rounds at 10–15 day intervals, since not all umbels mature at once. Varieties named are PF-35 and Gujarat F-1. The pest named is the aphid, and the diseases are damping off, collar rot and Ramularia blight.

Dill has two forms in the books: European dill (Anethum graveolens) and Indian dill (Anethum sowa), which is bolder and has stronger, more flavourful seeds. Dill is grown as a rabi or winter crop for seed. It needs at least 6–8 hours of direct sunlight a day and resists frost only to a limited extent in the vegetative stage. Dry and relatively high temperatures suit seed production. It grows best in well drained, slightly acidic soil rich in organic matter. Sowing is October–November, best mid-October to mid-November, by seed drill at 25–30 cm between rows and 2–3 cm deep, and thinning is done at 21 days after sowing to 8–10 cm between plants. The books give no manuring recommendation for dill, and none is given here. Weeding starts at 21–28 days after sowing. Irrigation is given as the plant needs it, keeping the soil moist but not wet, the first immediately after sowing for germination, then at flowering and seed development, in the cool hours of morning or evening. Flowering takes about 70–80 days. For seed, the flower stalks are cut just before the seeds ripen and turn tan, hung upside down in a warm well ventilated room, with a small paper bag around the flower heads to catch the seeds as they drop. The crop matures in 150–160 days, harvesting when the seeds on the tertiary branches have just turned yellowish brown, cutting the plants 40 cm above the ground and drying in shade. The pest named is the aphid.

Quick check

Fennel umbels do not all mature at once. How does the harvest handle that, and what happens if the grower waits for all of them?

Check your answer

The umbels are cut in 3–4 pluckings at 10–15 day intervals, each one as it reaches maturity. Waiting for all of them over-ripens the early ones and lowers market quality.

24.10 Anise, ajwain, celery and mustard

Anise is Pimpinella anisum, of the family Apiaceae: an herbaceous annual growing to 1 m or more, grown chiefly for its fruit, called aniseed. The flowers are yellowish white, about 3 mm across, in dense umbels, and the fruit is an oblong dry schizocarp, 3–6 mm long, greyish white. Anise is a tropical plant that also grows in subtropical climates, in a dry cool climate free from frost. It likes poor, light, well drained soil and full sun, and it should be planted where it is sheltered from wind. Sowing is October–November in subtropical areas and March–April in temperate areas, broadcast half an inch deep or drilled in rows 2.5–3 ft apart. Seed soaked overnight germinates better, and the books note that seed sown near coriander germinates faster and sets seed better. Feeding is farmyard manure 15–20 t/ha with nitrogen 30 kg/ha and P₂O₅ 15 kg/ha: all the phosphorus and half the nitrogen basal, the rest top dressed at 30 days after sowing. Weeding needs 2–3 weedings, the first with hoeing at 30–40 days after sowing. Irrigation is a pre-planting watering and then four irrigations from November to March. The crop is harvested when the plants turn yellowish brown, cut or uprooted early in the morning, dried in the threshing yard, threshed with sticks, dried to 10% moisture, winnowed, and the crisp dry seeds rubbed between the palms to separate seed from hull and sieved. The books state that aniseed has no serious pest problems and no serious disease problems, and that aniseed oil is said to repel insects.

Ajwain, or bishop's weed, is Trachyspermum ammi, of the family Apiaceae. Its seed carries 2–4% volatile oil, yellow brown in colour. It is a winter season crop that needs a dry and cool climate during growth and a warm climate at maturity. High atmospheric humidity and poor drainage cut the yield. It grows in all soils except sandy and heavy clay soils, and sandy loam with organic matter suits it best. Sowing is mid-September to the last week of October, usually by broadcasting, though line sowing is preferred: rows 45 cm apart, plants 15–20 cm apart, seed 1.5–2.0 cm deep. Feeding is farmyard manure 10–15 t/ha with nitrogen 40 kg/ha and phosphorus 20 kg/ha: all the phosphorus and half the nitrogen at field preparation, the rest of the nitrogen at 30 days after sowing. Weeding and hoeing are done two or three times, the first at 30 days after sowing, with thinning to 15–20 cm in the row. Irrigation runs to 4–5 waterings: a light watering after sowing if soil moisture is short, then at 15–25 day intervals. The crop matures in 160–180 days, when the flower heads turn brown, and is harvested by cutting the plants 40 cm above the ground, dried under shade, and the seeds separated by beating bundles with wooden sticks and winnowing. Varieties named include Lam Selection-1 (135–145 days, 8–9 q/ha) and AA-01-19 (150 days, about 13 q/ha, powdery mildew resistant). The pest named is the aphid, and the disease named is powdery mildew.

Celery is Apium graveolens, of the family Apiaceae: a herbaceous plant, usually 60–120 cm tall, annual in the plains and biennial in cold climates and on the hills. The flowers are white in compound umbels, and the seeds are very small, 1–2 mm long, oval and greenish brown. Celery is a rabi crop needing warm days and cool nights, low humidity and plenty of sunshine, with a long cool season. It needs cool weather during growth and warm dry weather at maturity, and a week long cold spell in mid season can induce bolting and lower the seed yield. It grows in almost all soils, but thrives on silt loam to loamy soil at pH 6–8.5 with moderate organic matter; clayey soils and waterlogging do not suit it. Celery seed is sown in a nursery in mid-September to mid-November, in beds 3 × 1 m, with about 200 sq m of nursery for one hectare. The nursery bed is given farmyard manure 500–600 kg with nitrogen 1.6 kg, P₂O₅ 0.8 kg and K₂O 0.4 kg per 200 sq m. Seedlings 12–15 cm tall and 60 days old, with 7–8 cm of root, are transplanted at 30 × 30 cm or 40 × 25 cm. Flowering comes in March and the seeds mature by the end of April, with the crop in the field for 120–140 days. Feeding in the field is farmyard manure 25–30 t/ha with nitrogen 80 kg/ha, P₂O₅ 40 kg/ha and K₂O 20 kg/ha: all the phosphorus and half the nitrogen basal, the rest of the nitrogen top dressed at 30 days after transplanting. Hoeing is done 3–4 times, the first 20–25 days after transplanting and then at 20–25 day intervals. Celery is shallow rooted and its moisture needs are high: the first irrigation is on the day of transplanting, then at fortnightly intervals, weekly near seed maturity. Harvesting is done when about 80% of the umbels turn light brown; the plants are dried on a clean threshing yard, threshed by beating with sticks, winnowed, and the seed dried in bright sunlight. The books state that no major insect pests and diseases attack celery.

Mustard and rapeseed are brassicas, of the family Brassicaceae, and the books treat them as one group with several types: Indian mustard, or rai, Brassica juncea, with medium round dark brown or black seeds; toria, Brassica rapa var. toria, with bold large dark brown seeds; brown sarson, Brassica rapa var. brown sarson, with bold light reddish seeds; yellow sarson, Brassica rapa var. yellow sarson, with smaller ovoid yellow seeds; and rapeseed, or gobhi sarson, Brassica napus, with large brownish black seeds; and taramira, Eruca sativa. Mustard is a crop of tropical and temperate zones that needs cool dry weather, clear dry weather with plenty of bright sunshine and adequate soil moisture for good oil yield. It is grown in the rabi season from September–October to February–March, needing a fair supply of moisture while growing and dry clear weather at maturity. It is not drought tolerant: it needs annual precipitation of 35–45 cm and moderate temperatures of 21–25.7°C. It grows on a wide range of soils from sandy loam to clay loam, thriving best on light loam soils, and the soil must drain well, since the crop cannot stand waterlogging. The soil pH of 6.0–7.5 suits it. Seed is usually broadcast, but line sowing is better: lines 30 cm apart, plants 8–10 cm apart, seed 4–5 cm deep. Feeding is farmyard manure 5 t/ha with Azophos 15 kg/ha, and the books stress soil testing first, since the crop takes heavy nutrients from the soil. Sulphur is given at 30 kg/ha, with borax 10 kg/ha and ammonium molybdate 0.5 kg/ha where deficiency shows. Under rainfed conditions all the nitrogen, phosphorus and potassium go on as a basal dressing; under irrigation all the phosphorus and half the nitrogen and potassium go on basal, and the rest is top dressed. Weeds cut the yield by about 20–30%, and one hoeing in the third week after sowing is enough for toria, with 1–2 hoeings for mustard and sarson. Thinning is done three weeks after sowing to 10–15 cm between plants. Flowering and pod filling are the most critical stages for water. Harvesting is done when the leaves turn yellow and the pods are just ripe, with a hand sickle or by uprooting; sarson can be left to a drier ripe stage since it shatters less. The plants are dried on the threshing floor for 4–5 days, threshed by beating or trampling, and winnowed in a natural air current. Maturity runs from 70–100 days for toria, 105–160 days for Indian mustard, 120–155 days for yellow sarson, 100–235 days for brown sarson and 140–150 days for taramira. Cleaned seed is sun dried for 2–3 days until the moisture falls to 8% and stored in gunny bags or bins in a dry store. The pests named are the mustard aphid, Bihar hairy caterpillar, mustard sawfly, painted bug, diamondback moth, leaf miner and cabbage butterfly. The diseases named are Alternaria blight, Sclerotinia stem rot, downy mildew, bacterial rot, phyllody and white rust.

Seed spices: one harvest rule for nine crops

Sow in the cool season
October to December for most seed spices; fennel also in June

↓

Thin, weed and water at the right stages
Flowering and seed filling are the stages that must not go dry

↓

Watch the colour turn
Seeds yellow, umbels brown, pods just ripe: that is the harvest signal

↓

Dry, thresh, winnow
Beat with sticks, winnow in the air, dry seed to 8–10% moisture for storage

Figure 24.2: Seed spices: one harvest rule for nine crops. Different sowing months and different yields, but the same sequence of thinning, watering at seed filling, watching the colour turn and drying the seed. Drawn from Das, Spice Crops Production Technology, Coriander, Cumin, Fennel, Fenugreek, Dill, Anise, Celery and Mustard chapters, and Dashora et al., Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants, Ajwain chapter.

Practical note: a seed spice bed you can run in one season

Ginger and turmeric need 8–9 months and a rhizome bed, so they are not a first trial. Start with coriander or fenugreek, which the books put at 110–150 days from sowing to seed. Sow coriander in rows 30 cm apart in mid-October to mid-November, rub the seed to split it, soak it overnight, and thin to 5–10 cm before the first irrigation. Water at 30–35 days, then at 50–60, 70–80 and 90–100 days. Cut green leaves from 20 cm onwards, and leave part of the bed uncut for seed: harvest that part when half the seeds turn yellow, dry it in partial shade, and beat the seed out lightly.

What the books do not settle

Four gaps are recorded here rather than filled in. First, the turmeric yield section in one book prints 20.0–25.0 q/ha, while the same book's own variety table gives fresh yields of 17.4–42.0 t/ha and a second book gives 4–5 tonnes of cured turmeric per hectare: the variety table and the cured yield are used here, and the printed figure is left as a conflict. Second, the sulphur dose for ginger is printed as a very wide range, 20.0–250 kg/ha, which reads as a misprint, so this unit gives no figure. Third, the anise harvest paragraph in one book repeats the cumin harvest text word for word, including the words "cumin crop", so the harvest description here follows the anise steps the book gives elsewhere. Fourth, dill has no manuring recommendation in the books used, and mustard fertiliser doses are given state by state, so this unit gives mustard farmyard manure, sulphur and micronutrients and says the NPK dose must follow a soil test. No yield or area statistics by country are given anywhere in this unit.

Key ideas to carry forward

  • Ginger and turmeric are rhizome crops: plant pieces with a live bud, mulch and earth up at 45 and 90 days, and harvest by the leaves, fresh ginger from 5 months and dry ginger and turmeric at 8–9 months.
  • Turmeric varieties trade fresh yield against dry recovery and curcumin: Roma has the highest dry recovery rate (31%) and the most curcumin (9.3%), while Rajendra Sonia gives the most cured spice per hectare (about 7.6 t/ha) because its fresh yield is twice Roma's.
  • Seed spices are mostly one family, Apiaceae, with fenugreek in Fabaceae and mustard in Brassicaceae, and they share one harvest rule: watch the seed head turn colour, then cut, dry, thresh and winnow.
  • Seed rates are tiny and yields are counted in hundreds of kilos: celery is sown at 2 kg/ha, fennel at 2–3 kg/ha, and the whole group is judged by seed colour and dryness at harvest.

Glossary

Rhizome

An underground stem that stores food and bears buds, from which the plant grows. Food tech example: a ginger rhizome bit of 20–25 g with one live bud, planted bud up at 4–5 cm deep.

Finger

A branch of the turmeric or ginger rhizome, also called a secondary rhizome. Food tech example: turmeric fingers harvested at 7–9 months and dried to a recovery of 18–31%.

Curcumin

The yellow colouring compound of turmeric. Food tech example: Roma turmeric carrying 9.3% curcumin, the highest in the books' variety table.

Dry recovery

The share of fresh weight that remains after drying, in percent. Food tech example: ginger with a dry recovery of 20–22%, so 100 kg of fresh rhizome gives 20–22 kg of dry ginger.

Earthing up

Drawing soil up around the base of the plant to cover the rhizomes and give them room to grow. Food tech example: earthing up ginger at 45 and 90 days after planting, right after weeding and fertiliser.

Umbel

A flat topped flower cluster in which the stalks spread from one point like the ribs of an umbrella. Food tech example: a coriander compound umbel of five smaller umbels, and a fennel umbel with 20–50 tiny yellow flowers.

Bolting

Running to flower and seed early, before the crop has grown properly. Food tech example: celery bolting after a week long cold spell in mid season, which lowers the seed yield.

Shattering

Seed falling out of the pod or head before or during harvest. Food tech example: mustard left too long after the pods ripen, losing seed to shattering on the threshing floor.

References

  • Das, P.C. Spice Crops Production Technology. Scientific Publishers, Jodhpur, 2014. Chapters: 3 Ginger, 4 Turmeric, 5 Coriander, 6 Fenugreek, 7 Cumin, 8 Anise, 9 Fennel, 10 Dill, 15 Celery, 20 Mustard and Rapeseed.
  • Pruthi, J.S. Major Spices of India: Crop Management and Post-Harvest Technology. Indian Council of Agricultural Research, New Delhi, 1993. Chapters: 5 Ginger, 6 Turmeric.
  • Dashora, L.K., Dashora, Abhay, and Lakhawat, S.S. Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants. 2nd edition, Agrotech Publishing Academy, Udaipur, 2013. Chapters: 12 Bishop's Weed (Ajwain), 13 Coriander, 14 Cumin, 15 Dill (Sowa), 16 Fennel.

Unit 25: More Spice Crops: Bulbs, Bark, Buds and Leaves

Note

Unit 25: More Spice Crops: Bulbs, Bark, Buds and Leaves

Open a spice box and look at what is inside. A garlic bulb, a curled stick of cinnamon bark, a dark clove bud, a green curry leaf, a few red threads of saffron.

Pause and think: a clove is a flower bud that was never allowed to open, and saffron is the stigma of a flower that opened for a day. What does that tell you about the harvest window for each one?

Quick check

Why does this unit group the crops by plant part rather than by flavour?

Check your answer

The plant part decides the whole production story: how the crop is planted, when it is picked, and how it is dried. A bulb is dug, a bark is peeled, a bud is picked before it opens, and a leaf is plucked fresh. Flavour alone would hide all of that.

25.1 The bulbs: garlic and onion

Garlic and onion are the two bulb crops of this unit, and they are close cousins. Both belong to the genus Allium and the family Liliaceae in the books.

Garlic: a bulb made of cloves

Garlic, Allium sativum, is a compound bulb made of small bulblets, the cloves. Das calls it one of the most important bulb crops, used as a spice or condiment. It is native to west Asia and the Mediterranean area.

Climate and soil. Garlic is a cool season crop. It does best in a mild season without extremes of heat or cold, at 13 to 24°C.

Planting. Garlic is propagated by cloves. Well grown, compact bulbs of uniform shape and size are selected, and the bigger cloves from the outer side of the bulb are used.

Nutrition and irrigation. The book gives a basal dressing of 10.0 to 15.0 tonnes of compost or farmyard manure per hectare, with 50 kg each of nitrogen, phosphorus as P₂O₅ and potassium as K₂O per hectare, plus 25 kg of nitrogen per hectare at 25 days after planting and again at 42 days. Zinc sulphate at 25.0 kg per hectare, borax at 10.0 kg per hectare and ammonium molybdate at 0.5 kg per hectare are given once in 3 to 4 crops, on the basis of a soil test. Garlic is shallow rooted and needs water about every 8 days while it grows and every 10 to 15 days while it matures, with the last watering 2 to 3 days before harvest.

Harvest and yield. Garlic takes four and a half to five months. The right time to lift is when the bulbs have reached full size but the cloves have not started to separate. The sign is plain: the main stem above the bulb softens, the top falls, and about 75 percent of the leaves turn yellow.

Pests and diseases. Thrips, the onion maggot and the black garlic fly attack garlic in the field and in store. The diseases the book lists are white bulb rot (Sclerotium cepivorus), brown rust (Puccinia allii), pink root (Pyrenochaeta terrestris), Fusarium bulb rot, Cercospora leaf spot, purple blotch (Alternaria porri), downy mildew (Peronospora destructor) and powdery mildew.

Cloves dibbled 5 to 7.5 cm deep, growing end up, 15 cm by 7.5 cm

↓

Cool moist growth at 13 to 24°C, water every 8 days, then every 10 to 15 days

↓

Top falls, 75% of leaves yellow, bulbs at full size, cloves not yet separating

↓

Lift, tie in bundles, cure in shade 2 to 3 days: 100 to 200 quintals per hectare

Figure 25.1: Garlic from planting to harvest. The fall of the top and the yellowing of the leaves mark the lifting day. Drawn from Das, Spice Crops Production Technology, Garlic chapter.

Onion: the bulb that starts in a nursery

Onion, Allium cepa, is the most widely cultivated spice of the genus Allium in the book, and one of the most important commercial vegetable crops. The bulb is the commercial part.

Climate and soil. Onion is a cool season crop, grown in winter and harvested before the real summer begins. It needs different temperatures at different stages: 15 to 21°C for vegetative growth, 20 to 25°C for bulb formation and for seedling growth.

Propagation and planting. Onion is propagated by seedlings and by bulbs. Seed is sown in a raised nursery bed about 3 m long, 0.6 m wide and 20 to 30 cm high, at 8.0 to 10.0 kg of seed per hectare. Seedlings are ready to transplant in about eight weeks, when they are about 15 cm tall, at 7.5 by 10 cm.

Nutrition. The book gives 10.0 tonnes of compost or farmyard manure per hectare, with 150 kg nitrogen, 100 kg P₂O₅ and 150 kg K₂O per hectare. All the phosphorus and half the nitrogen and potassium go in as a basal dressing, and the rest is top dressed in two parts, at 21 days and at 40 days after transplanting or planting. The book gives no irrigation schedule for onion.

Harvest and curing. The bulb crop is ready in 3 to 5 months, depending on the variety. Maturity shows when the tops fall over while the leaves are still green.

Pests and diseases. The insect pests are thrips, the onion maggot, the black onion fly and the onion leaf eating caterpillar (Spodoptera exigua). The diseases are damping off, onion smut (Urocystis cepulae), basal rot, bacterial brown rot (Pseudomonas aeruginosa), black mould (Aspergillus niger), bacterial soft rot and Fusarium rot, and aster yellows, carried by the aster leafhopper.

Quick check

Garlic and onion both need dry weather at maturity. Why would watering a mature garlic crop after a dry spell be a mistake?

Check your answer

Watering after a long dry spell splits the bulbs, and too much water at maturity makes them sprout. That is why the book stops watering down to a last light watering 2 to 3 days before harvest, and why onions are cured dry for weeks before storage.

25.2 Bark and buds: cinnamon and cloves

Cinnamon: the bark that becomes a quill

Cinnamon, Cinnamomum verum, in the family Lauraceae, is the dried inner bark of an evergreen tree. Das calls it "sweet wood" and the earliest known spice, native to Sri Lanka and the Malabar Coast, and seen wild in the forests of the Western Ghats. A full grown tree reaches 6 to 15 m.

Climate and soil. Cinnamon suits a wide range of climates and grows well at about 200 m above sea level, with 200 to 250 cm of annual rainfall and an average temperature of 27°C. It is grown as a rainfed crop, and long dry spells do not suit it.

Propagation and planting. Seed propagation and vegetative propagation through cuttings and air layers are both practised. Seeds are sown in sand beds or polythene bags in a mixture of sand, soil and well powdered dried cow dung in the ratio 3:3:1.

Nutrition and irrigation. Per plant, the book gives 20 g nitrogen, 18 g P₂O₅ and 25 g K₂O in the first year, rising by the third year after planting to 29.0 kg compost or farmyard manure, 4.0 kg neem cake, 150 g nitrogen, 75 g P₂O₅ and 150 g K₂O. Manures go on in May to June and fertilisers in two equal split doses, in May to June and September to October. Seedlings are watered until established, and after that as the plant needs it, since cinnamon is a rainfed crop.

Harvest and yield. Shoots are ready 2 to 3 years after planting, when the plant is 1.5 to 2 m tall with 3 to 4 shoots and the bark has turned brown. The best time to harvest shoots is September to November.

Pests and diseases. Gall forming mites make pinkish galls on the lower leaf surface that turn green with age. The cinnamon butterfly (Chilasa clytie) eats tender and slightly mature leaves until only midribs and veins are left. The shoot and leaf webber webs leaves and terminal shoots into clusters and scrapes the leaf surface.

Seedling cut back to 30 cm at 2 to 3 years, side shoots trained into a low bush

↓

Shoots 2 years old, 1.0 to 1.25 m long, bark brown: cut close to the ground, September to November

↓

Bark peeled, rolled into quills, dried: 60 to 120 kg per hectare at 3 to 4 years, 225 to 300 kg at 10 to 11 years

Figure 25.2: Cinnamon from bush to quill. Cutting the shoot back young is what makes the straight, peelable shoots. Drawn from Das, Spice Crops Production Technology, Cinnamon chapter.

Cloves: the bud picked before it opens

Cloves are the aromatic dried flower buds of Syzygium aromaticum, in the family Myrtaceae. The name comes from the Latin clavus, a nail, for the shape of the dried bud.

The tree and the soil. Clove is an evergreen tree 8 to 12 m tall, with large leaves and flowers in many flowered terminal clusters. The buds start pale, turn green, and then bright red when they are ready to pick. It is native to the Molucca islands, the "Spice Islands".

Propagation and planting. Clove is propagated by seed from ripened fruit, known as "mother of clove", taken from trees more than 15 years old that bear regularly. Seeds are placed flat about 2.5 cm deep at 12 to 15 cm spacing.

Nutrition and irrigation. Manures rise with the age of the tree: 15.0 kg of manure per plant at 1 year, 30.0 kg at 2 years and 50.0 kg at 15 years, with nitrogen, P₂O₅ and K₂O at 20, 20 and 60 g at 1 year, 40, 35 and 100 g at 2 years, and 300, 250 and 750 g at 15 years. Fifty grams each of Azospirillum and Phosphobacterium are added one month after manuring. Water is given in the early stages and afterwards as the crop needs it.

Harvest and yield. The tree starts to yield in the seventh year, and full bearing comes after 15 to 20 years. Flowering is September to October in the plains and December to February at high altitudes.

Pests and diseases. The stem borer (Sahyadrassus malabaricus) is the most important insect pest: its caterpillars bore into the main stem, the plant dries above the point of attack, and the plant dies in the end. Scales (Lecanium psidii) suck sap from leaves and tender shoots and can kill young seedlings, and mealy bugs suck sap from tender shoots until the parts dry up; ladybird beetles are the book's biological control for both.

Quick check

A cinnamon shoot and a clove bud are both picked by hand. What is the grower watching for in each case?

Check your answer

For cinnamon, the bark turning brown on a shoot about 2 years old, 1.0 to 1.25 m long: that is the peeling stage. For cloves, the base of the calyx turning from green to pink on a fully developed bud: that is the last moment before the bud opens, and after it opens the spice is lost.

25.3 Two spices from one fruit, and one spice that tastes like three: nutmeg and allspice

Nutmeg gives two spices from a single fruit, and allspice gives one berry that smells like three spices at once. Both are evergreen trees, both are sun dried, and both are picked green or ripe by hand.

Nutmeg: seed and mace

Nutmeg, Myristica fragrans, in the family Myristicaceae, is a large evergreen tree native to the Molucca Islands. It produces two spices: nutmeg, the seed kernel inside the fruit, and mace, the dried aril around it.

Climate and soil. Nutmeg needs a hot, humid climate without a marked dry season, with 150 cm or more of annual rainfall, and it prefers partial shade. It grows well from about 1,300 m above sea level.

Propagation and planting. Nutmeg is raised from seedlings. Fully ripened, tree burst fruits are selected, the fleshy rind and the mace are removed, and the seeds are sown at once in beds 100 to 120 cm wide and 15 cm high, in garden soil and sand at 3:1, dibbled about 12 cm apart.

Nutrition and irrigation. Organic manure starts at 10.0 kg per seedling in the first year and rises year by year until a well grown tree of 15 years and above gets 50 kg per year. Fertilisers start at 20 g nitrogen, 18 g P₂O₅ and 50 g K₂O per plant in the first year, rising to about 500 g nitrogen, 250 g P₂O₅ and 100 g K₂O per year at 15 years and above, in two equal split doses, in May to June with the organic manure and in September to October. Water is needed soon after planting and in the summer months, and otherwise as required.

Harvest and yield. Fruits are plucked from the tree or allowed to drop. The nuts are dried in the sun for six to eight days until they rattle in their shells, then stored in a warm, dry place before shelling. Dried nutmeg and mace are used direct as spice and for their derivatives.

Allspice: the berry that smells of three spices

Allspice, Pimenta dioica, in the family Myrtaceae, is the dried unripe fruit of a mid canopy tree. Its other names are Jamaica pepper, pimento, myrtle pepper and clove pepper. The English coined "allspice" in 1621, the book says, because they thought it combined the flavours of cinnamon, nutmeg and cloves. The leaves and fruits smell of cloves, black pepper, nutmeg and cinnamon together. It is native to the Greater Antilles, southern Mexico and Central America.

The tree and the soil. The tree is small and evergreen, 7 to 13 m tall, with leathery, smooth, grey bark and glossy, dark green, aromatic leaves up to 15 cm long. The white flowers are about 0.6 cm across, borne in many flowered pyramidal cymes from the leaf axils.

Propagation and planting. Allspice is propagated by seed. Seeds are sown in raised beds 15 to 20 cm high and 1 m wide, in loose soil and sand with sand in the top layer.

Harvest and yield. Plants start flowering at 7 to 10 years, but the peak harvest comes at 15 to 20 years. Berries are picked when fully grown but still green, then sun dried for about 3 to 12 days, turned often, until they are crisp and give a metallic sound when shaken. That sound is the curing test. The book gives no irrigation schedule for allspice.

Nutmeg fruit: fleshy rind outside, red mace netting the seed inside

↓

Rind and mace removed, seed dried in the sun 6 to 8 days until it rattles in its shell

↓

Two spices: nutmeg, the kernel, and mace, the dried aril: 800 kg of nutmeg per hectare

Figure 25.3: Nutmeg, one fruit and two spices. The mace is the lacy coat around the seed, not part of the seed itself. Drawn from Das, Spice Crops Production Technology, Nutmeg chapter.

25.4 Leaves and fruit rind: curry leaf and Garcinia

Curry leaf: the leaf that flavours the pot

Curry leaf, Murraya koenigii, in the family Rutaceae, is a small tree 4 to 6 m tall with a trunk up to 40 cm across. Its leaves are pinnate, with 11 to 21 leaflets, each 2 to 4 cm long and 1 to 2 cm broad, and highly aromatic.

Climate and soil. Curry leaf tolerates temperatures up to 37°C, but growth is affected below 16°C, and it needs about 20°C to germinate, taking a long time to do so. The best temperature range is 26 to 37°C.

Propagation and planting. Curry leaf is mainly propagated by seed. Well ripe fruits are collected from high yielding plants and sown in a nursery or in polythene bags in sand, soil and farmyard manure at 1:1:1. Seeds germinate in three weeks.

Nutrition, water and pruning. Per plant per year the book gives 10.0 kg of compost or farmyard manure, with 60 g nitrogen, 80 g P₂O₅ and 40 g K₂O, applied at the onset of the monsoon. The crop is watered at once after planting, again on the third day, and then once a week.

Harvest and yield. Leaves are first picked 15 months after planting, and commercial harvest starts with three year old plants. With good management a normal yield runs up to 20 to 25 years of age.

Garcinia: the sour rind

Garcinia, Garcinia cambogia, in the family Clusiaceae, is also called Malabar tamarind, brindall berry and vrikshamla. The commercial part is the fruit.

Climate and soil. The book says very little here: Garcinia can be grown in different agro-climatic conditions, and in a variety of soils, with sandy loam, loam and clay loam best suited. No temperature, rainfall or altitude figures are given, and this unit records that gap.

Propagation and planting. Garcinia is propagated by seed and by grafting, including soft wood grafting, and planted in July to August. Seedlings are raised from well ripe fruits of high yielding plants, sown in a nursery or polythene bags in sand, soil and farmyard manure at 1:1:1.

Care and yield. Weeding is done as needed, usually 15 to 20 days after planting, and fast growing branches are pruned from the third year onwards to keep the weight of the plant balanced. Water is given as the plant needs it, and standing water must be avoided.

25.5 The flower spice: saffron

Saffron is the most precious and most expensive spice in the world, the book says, and the only spice that comes from a flower. It is the stigma of Crocus sativus, in the family Iridaceae, a perennial, low growing herb with a globular corm 0.5 to 5 cm across.

Climate and soil. Saffron does well in sub temperate regions from 1,500 to 2,400 m, at a day temperature of 20 to 22°C, with an optimum of 12 hours of light a day for growth and flowering, and 30 to 40 cm of rainfall. Dry weather during flowering is essential for higher yields.

Propagation and planting. Saffron is propagated by planting corms. The mother corm reproduces every year and gives four to six daughter cormlets.

Nutrition. The book gives 20.0 tonnes of compost or farmyard manure per hectare, with 50.0 kg nitrogen, 40.0 kg P₂O₅ and 20.0 kg K₂O per hectare. The book gives no irrigation schedule for saffron.

Harvest. The flowering season lasts three weeks, from the middle of October to the first week of November. Flowers are picked daily in the morning, and the stigmas and styles are trimmed at once. About 160,000 flowers must be hand picked to produce 1 kg of good quality dried saffron.

Uses and disease. Saffron is used as a culinary seasoning and for colouring food, with a unique taste that adds a distinctive flavour. It is also an ingredient for cosmetics and perfumes, and a dye for textiles.

Corms planted June to July, 15 to 20 cm deep, 1,600 to 2,000 kg per hectare

↓

Purple flowers for three weeks, mid October to first week of November, picked daily in the morning

↓

Red stigmas trimmed at once and dried: about 160,000 flowers for 1 kg of dried saffron

Figure 25.4: Saffron from corm to thread. The harvest is three weeks of daily hand picking, and the yield is counted in flowers, not in tonnes. Drawn from Das, Spice Crops Production Technology, Saffron chapter.

Quick check

Why is saffron the most expensive spice in the world, in one sentence drawn from this section?

Check your answer

About 160,000 flowers must be hand picked, one by one, in the morning, over a three week flowering season, to make 1 kg of dried saffron, and only the red stigma threads are used.

25.6 The seed spice from Dashora: ajwain

Ajwain, bishop's weed, Trachyspermum ammi, in the family Apiaceae, is the tenth crop of this unit and the only one drawn from Dashora rather than Das. It is a minor seed spice, grown for its seeds, its herb and its volatile oil.

Climate and soil. Ajwain is a winter season crop. It needs a dry and cool climate while it grows, and a warm climate at maturity. High atmospheric humidity and poor drainage cut the yield. It grows in all types of soil except sandy and heavy clay soils, and sandy loam with organic matter suits it best.

Sowing. The field is brought to fine tilth with a deep turning plough followed by 2 to 3 light ploughings, with planking after each one to conserve moisture. Sowing is from mid September to the last week of October.

Nutrition, weeding and water. At field preparation, 10 to 15 tonnes of well decomposed farmyard manure or compost are applied, with 40 kg nitrogen and 20 kg phosphorus per hectare. All the phosphorus and half the nitrogen go in at field preparation, and the rest of the nitrogen is given 30 days after sowing. The crop needs 4 to 5 irrigations, the first a light one after sowing and later ones at 15 to 25 day intervals.

Harvest, yield and varieties. The crop matures in 160 to 180 days after sowing. The fruits are ready when the flower heads turn brown. Plants are cut 40 cm above ground level, dried under shade, and the seeds are separated by beating the bundles with wooden sticks, followed by winnowing.

Pests and diseases. Aphids suck sap from tender parts and flowers, the plant turns yellow, and the seeds shrivel and form poorly. Powdery mildew starts as a white powdery mass on the leaves and later covers the whole plant with whitish powder, and the infected plant gives seed of lower quality.

25.7 The ten crops side by side

The table below puts the harvest and the yield of all ten crops in one place, so the differences stand out. Read it with care: the yields are not comparable across crops, because a kilogram of saffron threads and a kilogram of onion bulbs are different kinds of harvest. The point of the table is the timing and the test for readiness, not a ranking.

Table 25.1: Harvest part, readiness test and yield of the ten crops, as given in the books

CropPart harvestedReadiness testYield as given in the book
GarlicBulbTop falls, 75% of leaves yellow, cloves not yet separating100 to 200 quintals per hectare
OnionBulbTops fall over while leaves are still greenNo yield given in the books
ClovesUnopened flower budBase of calyx turns green to pink2.0 kg per plant; 400.0 kg per hectare
CinnamonInner barkBark brown on a shoot about 2 years old60 to 120 kg quills per hectare at 3 to 4 years; 225 to 300 kg at 10 to 11 years
NutmegSeed kernel and maceFruit plucked or dropped; nut rattles in its shell after drying800 kg per hectare
AllspiceUnripe berryFully grown but still green; crisp with a metallic sound when shaken50 to 60 kg dry berries per plant
Curry leafLeafPicked from 15 months; commercial harvest from 3 years20.0 to 25.0 tonnes green leaves per hectare
SaffronStigmaFlowers picked daily in the morning, stigmas trimmed at onceAbout 160,000 flowers per 1 kg dried saffron
GarciniaFruit rindRipe fruit8.5 tonnes ripe fruit per hectare
AjwainSeedFlower heads turn brown, 160 to 180 days after sowing10 to 15 quintals per hectare irrigated; 4 to 6 unirrigated

Questions to think about

  • Garlic, onion and saffron all need dry weather at the end. What goes wrong if each one is watered or rained on at that stage?
  • Cloves and allspice are both picked green and unripe. What is lost if the picker waits?
  • Cinnamon, cloves, nutmeg and allspice are all trees that take 7 to 15 years to reach full bearing. What does that mean for a farmer who plants them today?

A visual sorting game: write the ten crop names on ten cards and the ten plant parts (bulb, bulb, bud, bark, seed and mace, berry, leaf, stigma, fruit rind, seed) on ten more cards. Shuffle both sets face down and turn them over two at a time, matching each crop to its part.

Practical note: buying, drying and keeping these ten

Buy whole, not ground, wherever you can: whole cloves, whole nutmeg, whole allspice berries, cinnamon quills, saffron threads and whole dried Garcinia rind keep their volatile oil far longer than powders, because the oil sits inside intact tissue. Dry every one of these crops before storage, and dry them in the sun or in the shade as the book directs for that crop: garlic and onion need shade curing and weeks of dry ventilation, cloves and allspice need sun drying until crisp, nutmeg needs sun drying until the seed rattles, and saffron threads need drying at once after trimming.

Key ideas to carry forward

  • The plant part decides the production story: bulbs are dug, bark is peeled, buds are picked unopened, leaves are plucked, stigmas are trimmed by hand.
  • Garlic and onion are cool season bulb crops that need dry weather at maturity, shade or windrow curing, and careful lifting without wounds.
  • Cinnamon is ready when the bark turns brown on a two year old shoot; cloves are ready when the calyx base turns pink; both are picked by hand and sun dried.
  • Nutmeg gives two spices from one fruit, seed and mace; allspice gives one berry that smells of cinnamon, nutmeg and cloves.
  • Curry leaf is pruned to a one metre bush and picked every two and a half to three months; Garcinia rind is dried for its sour acids.
  • Saffron is counted in flowers, about 160,000 per kilogram, and that hand work is what sets its price.
  • Ajwain is a winter seed spice with 2 to 4 percent volatile oil, sown in lines in autumn and harvested when the flower heads turn brown.
  • Where the books give no value, no variety and no pest list, this unit says so: onion bulb yield, allspice manuring, and the pest lists for curry leaf and Garcinia are gaps in the books, not in the reading.

Glossary

Bulb

An underground storage organ made of fleshy scales or cloves, as in garlic and onion. Food tech example: a garlic bulb pulled apart into cloves for planting, or an onion bulb cured for storage.

Clove (garlic)

One of the small bulblets that make up a garlic bulb, used as planting material. Food tech example: garlic cloves dibbled 5 to 7.5 cm deep with the growing end up.

Quill

A roll of dried cinnamon bark. Food tech example: cinnamon shoots peeled and rolled into quills, yielding 60 to 120 kg per hectare at 3 to 4 years.

Mace

The dried aril, the lacy coat around the nutmeg seed. Food tech example: mace removed from the nutmeg fruit before the seed is sun dried.

Curing

Drying and resting a harvested crop so it keeps, such as garlic in shade for 2 to 3 days or onions in windrows for 3 to 10 days and then in ventilated storage for 3 to 4 weeks. Food tech example: onions cured until the tops are dry before the tops are cut off.

Sunscald

Damage to bulbs or fruit from direct strong sun after harvest. Food tech example: onion bulbs left in windrows with their tops covering them to prevent sunscald.

Volatile oil

The aromatic oil in a spice that evaporates easily and carries its smell and flavour. Food tech example: cloves holding 13.2 percent volatile oil, and ajwain seed holding 2 to 4 percent.

Stigma

The pollen catching tip of a flower, the part harvested as saffron. Food tech example: red saffron stigmas trimmed from the flower at once after morning picking.

References

  • Das, P. C. Spice Crops Production Technology. Scientific Publishers (India), Jodhpur, 2014. ISBN 978-81-7233-873-2. Chapters 11 (Cloves), 12 (Cinnamon), 13 (Garlic), 14 (Onion), 16 (Nutmeg), 17 (Allspice), 18 (Curry Leaf), 19 (Garcinia) and 22 (Saffron).
  • Dashora, L. K., Dashora, Abhay, and Lakhawat, S. S. Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants. 2nd edition, Agrotech Publishing Academy, Udaipur, 2013. ISBN 978-81-8321-275-5. Chapter 12 (Bishop's Weed, Ajwain).

Unit 26: Post-Harvest Technology of Spices

Note

Unit 26: Post-Harvest Technology of Spices

A spice is made or ruined after it leaves the plant. The same pepper berries can come out black, glossy and sharp, or grey, mouldy and flat, and the difference is not the field: it is the day of picking, the minutes in boiling water, the turning on the mat, and the bag they are packed in. Unit 6 gave the basics of drying and storage for six spices. This unit goes deeper into the post-harvest chain itself: harvesting, drying and curing, cleaning, grading, packaging, storage and decontamination, with the methods and the numbers the books give.

Questions to think about as you read: Why does picking a cardamom capsule a week early change its colour, its size and its price? Why does a chilli drier run at 65 °C and not at 80 °C? Why is a warehouse, not a drier, the place where a spice merchant makes or loses money? Each section answers one of those questions.

26.1 The post-harvest chain, in order

Post-harvest technology covers everything that happens after harvesting: drying, cleaning or garbling, grading and quality evaluation, packaging, storage and distribution or marketing. That list comes from the black pepper chapter of the Major Spices book, and it fits every spice in this unit.

🤔 Pause and think: a farmer dries pepper perfectly, then packs it damp into a sealed bag. A second farmer dries it poorly, then cleans it thoroughly and packs it dry. Which lot keeps, and why does the order of the steps decide the answer?

Quick check

What are the steps of post-harvest technology, in the order the pepper chapter lists them?

Check your answer

Drying, cleaning or garbling, grading and quality evaluation, packaging, storage and distribution or marketing. A careless early step cannot be fixed by a later one.

26.2 Harvesting: the right stage, judged by eye and hand

Harvest timing is decided by the part that will be sold, and the books judge it by visible signs, not by the calendar alone. Black pepper is picked when the spikes start yellowing and one or two berries on a spike become red.

The book states that no amount of cleaning and re-drying at a central packing centre will restore the quality of a carelessly dried primary crop, so only well-matured berries must be harvested.

Small cardamom is picked fruit by fruit, and the stage matters more than for any other spice in this unit. The fruits ripen over an extended period and are usually gathered once a month, at an interval of 30 to 40 days, and each clump is visited periodically.

Chillies are judged by ripeness and by withering on the plant. Pods left to ripen and to partially wither on the plant are superior in pungency, initial colour and colour retention to pods picked when fully coloured but succulent.

Ginger is harvested in stages, according to what it will be sold as. For tender rhizome sold as green ginger, for preserve or for pickling, the crop is ready from the fifth month after sowing. Shrivelling, yellowing and withering of leaves, with drying and lodging of the aerial stems, indicate maturity, and the crop is harvested in about 7 months after planting.

The crop for green ginger should be harvested within 195 days of planting, because after that stage crude fibre increases and protein and fat decrease. Dry ginger is harvested at 245 to 255 days and never beyond 260 days, because after that the fibre keeps increasing and the percentage of volatile oil falls.

Harvest signs the books use: five separate tests, one per spice

Black pepper
Spikes yellowing, 1 or 2 berries red; 6 to 8 months from flowering
Small cardamom
Picked fruit by fruit at the paling stage; ripe splits, unripe shrivels
Chillies
Ripe and partly withered on the plant; heaped indoors 2 to 3 days at 22 to 25 °C to finish colour
Ginger
Green ginger within 195 days; dry ginger at 245 to 255 days, never beyond 260 days
Garlic
Top fallen, 75% of leaves yellow, cloves not yet separating; bundles cured in shade 2 to 3 days

Figure 26.1: Harvest signs for five spices. Each panel is a separate test for one spice, judged by eye and hand, and each one decides what drying can and cannot do next. Drawn from Pruthi, Major Spices of India, Ch. 2 to 5; Das, Spice Crops Production Technology, Ch. 1 and 13.

Quick check

Why must dry ginger be harvested by 260 days after planting, and not later?

Check your answer

After 260 days the fibre keeps increasing and the percentage of volatile oil falls. Green ginger should be harvested within 195 days, before fibre rises and protein and fat fall.

26.3 Drying and curing: sun, kiln, flue and machine

Drying is the main post-harvest step, and the books describe four families of methods: sun drying, kiln and flue curing, solar drying, and mechanical drying. The choice is set by the spice, the weather, and the scale, and the books give the numbers for each.

Black pepper is the sun-drying case. The berries are spread on bamboo mats or concrete platforms in the sun and turned over periodically, because without turning, mould contamination may result in a poorly dried product with a greyish, unattractive appearance. Uniform drying is more likely when round berries are separated from spikes before drying.

The cascade drier works in two passes with a 24 to 28 hour rest between them, because the surface dries fast in the first pass while moisture from the core needs time to diffuse out, and the second pass brings the lot to 10% moisture.

Small cardamom is the curing case, and colour is the judge. In large plantations, curing is done in a drying house in which the fruits are heated by pipes carrying hot air from a furnace. A hot-air chamber of 4.5 metres by 4.5 metres, with a ceiling about 2.5 metres high, holds enough for 1800 to 2000 kilograms of cardamom a year.

Three cardamom curing schedules, side by side: a batch uses one of the three

Schedule 1: Conventional kiln chamber
About 30 hours at about 82 °C; trays turned, top and bottom exchanged after 12 hours
Schedule 2: Three-stage flue curing
54 °C for 3 hours, then 43 °C for about 30 hours, then 54 °C for 3 hours: about 36 hours in all
Schedule 3: Das book drier and kiln
Drier at 45 to 50 °C for 14 to 18 hours; kiln overnight at 50 to 60 °C
Judge the result, whichever schedule was used
Hard capsules, good green colour, no bleaching by strong light, no splitting

Figure 26.2: Three reported curing schedules for small cardamom, plus the judging criteria. A batch follows one schedule only: they are alternatives, not steps. They disagree on temperature and time, and the books do not reconcile them. Drawn from Pruthi, Major Spices of India, Ch. 3; Das, Spice Crops Production Technology, Ch. 1.

Solar driers were built to fix the faults of the conventional chamber, above all its waste of firewood and its guesswork over temperature. In the conventional chamber the room is heated with hot smoke from burning firewood, there is no way of determining the temperature except by guesswork, and overheating discolours or chars the cardamom.

Large cardamom is cured by smoke, and the book is frank about the cost. Fresh fruits are dried in a kiln made by local growers in the plantation itself, using wet and fresh cut wood that does not burn well but gives a lot of smoke, so that the temperature does not rise high.

Chillies carry the fullest drying comparison in the books. Fresh chillies have a moisture content of 65 to 80%, depending on whether they were partly dried on the plant or harvested while still succulent, and this must be reduced to 10% to prepare dried spice.

Artificial drying of chillies is described from the United States, where processors wash and inspect the fruits, spread them on trays whole or sliced in 2.5-centimetre lengths, and dry them in tunnel driers or stainless steel continuous-belt driers in a forced current of air at 50 to 60 °C, reducing moisture to 7 to 8%. Some processors use a two-stage method, drying first to 12 to 20% moisture, storing at 0 °C, and finishing the drying to 7 to 8% when the pods are needed for grinding. Slicing or slitting pods reduced drying time by half, with no loss of initial colour on drying for 72 hours at 60 to 75 °C.

Lease and Lease found the optimum drying temperature for a good quality product to be 65 °C: single layers of succulent whole fruits dried to 8% moisture in 12 hours, and in 6 hours when sliced. One source in the book recommends 45 to 50 °C instead, and the book does not reconcile the two.

The CFTRI method dips fresh chillies in Dipsol, a water-based emulsion of potassium carbonate at 2.5%, refined groundnut oil at 1%, gum acacia at 0.1% and butylated hydroxyanisole at 0.001%, for about 5 minutes, then dries them on multitier trays at 5 to 10 kilograms per square metre. The method dries in about a week against 15 to 21 days traditionally, uses less space, retains colour and pungency better, gives a more hygienic product, and yields 2% more finished product.

Ginger curing is scraping, washing and sun. In scraped ginger, the epidermal layer of the fresh rhizome is scraped off with a sharpened bamboo splinter, the rhizomes are washed in water and dried in the sun for 7 to 10 days, turned uniformly during drying.

Turmeric curing is boiling, then sun. In the traditional method, the cleaned rhizomes are boiled in water, just enough to immerse them, for 45 to 60 minutes. Boiling is stopped when froth comes out and white fumes appear, giving out a typical odour.

Nutmeg and mace are the delicate case, and the Shanmugavelu book gives the numbers. Fresh nutmeg has a volatile oil content of 15% and fresh mace 20%, with nonvolatile ether extract of 20% and 30% respectively, and those values are high compared with the dried spices, so there is considerable loss of these ingredients during processing and storage, which needs to be minimised. The kernel is dried along with the shell, because at harvest it is firmly attached to the outer shell and cannot be separated without injuring it, and slicing the kernels for quicker drying cannot be recommended, because it loses volatile oil and any injury may lead to discolouration through oxidative degradation of polyphenols.

Table 26.1: Drying and curing numbers from the books

SpiceMethodTime and temperatureMoisture result
Black pepperSun on mats or concrete, turned often2 to 4 days after a 1-minute boiling water dipCascade drier: 25% to 10% in two stages, 24 to 28 hours rest between passes
Small cardamomKiln chamber; three-stage flue curingAbout 30 hours at about 82 °C; or 54 °C for 3 hours, 43 °C for about 30 hours, 54 °C for 3 hoursHard capsules, good green colour; total about 36 hours
Large cardamomSmoke kiln; improved flue-pipe systemAbout 3 days smoking; flue-pipe about 24 hoursAbout 33% of fresh weight recovered; flue-pipe gives deep maroon colour
ChilliesTraditional sun; CFTRI Dipsol method; Laul et al. tray trial5 to 15 days; CFTRI Dipsol method about a week; Laul et al. tray trial 14 daysFresh 65 to 80% down to 10%; Laul et al. trial 72 to 74% down to about 6%
ChilliesArtificial, optimum65 °C; whole fruits 12 hours, sliced 6 hours, to 8% moistureAt 80 °C pungency, initial colour and colour retention fall
GingerScraped, washed, sun dried7 to 10 days in the sunDry ginger recovery 12 to 23% of green weight by type (China 12-15%, Rio-de-Janeiro about 16%, Nadia/Narasapattam 20-22%, Karakkal 23%)
TurmericBoiled, then sun driedBoiled 45 to 60 minutes; sun dried 10 to 15 daysFingers dried in 5 to 7 centimetre layers on bamboo mats
NutmegMechanical at 50 °C; sun28 hours mechanical; 7 days sun with 18.5 hours direct sun36% down to 9% mechanical, 8.5% sun; finish at 6 to 7%
MaceMechanical at 50 °C; sun; shade5 to 6 hours mechanical; 15 hours sun; 24 to 28 hours shadeAll to 5 to 5.5% moisture

Quick check

Why is pepper dried in two passes with a rest of 24 to 28 hours between them, instead of in one long pass?

Check your answer

In the first drying the surface dries fast, but moisture from the core needs time to diffuse out to the surface. After one pass the pepper is stored for 24 to 28 hours, then dried again to 10% moisture.

26.4 Cleaning, garbling, polishing and bleaching

Before packing, dried pepper is cleaned to get rid of dirt, grit, stones, stalks and leaves, and graded by size of berry or density. Pneumatic separators with magnetic separators remove debris by relative specific gravity, and magnetic separators remove metallic contamination such as iron filings and stray nails.

Bleaching is a separate step, and the books are doubtful about its value. For cardamom, only capsules that lack a uniform green colour and fetch a low price are put through bleaching, to attain uniform colour and appearance. The agents used are bleaching powder, sulphur dioxide and hydrogen peroxide.

Bleaching is costly and time-consuming, and no material benefit is reported from it except an improvement in the outer appearance of the pericarp, which is often thrown away, and a slight improvement in keeping quality. It reduces the percentage of volatile oil in the seeds and imparts a sulphurous odour to the produce. Most traders and processors consider it an unnecessary luxury.

Quick check

Bleaching makes cardamom whiter. What does it cost the spice?

Check your answer

It reduces the percentage of volatile oil in the seeds and imparts a sulphurous odour. Most traders and processors consider it an unnecessary luxury: costly and time-consuming, improving only the outer appearance of the pericarp, which is often thrown away, with only a slight improvement in keeping quality.

26.5 Grading: how a lot is sorted and named

Grading sorts a dried lot by the qualities a buyer can see and measure: size, colour, weight, and the share of empty, broken, shrivelled or immature pieces. For black pepper, well-defined trade grades and quality characteristics were established by the Agmark authorities, the Directorate of Marketing and Inspection, under a pre-shipment inspection scheme and a compulsory quality control act, and the standards are implemented under a compulsory Agmark grading scheme, particularly for exports.

The whole pepper grades are Malabar Garbled grades 1 and 2, Malabar Ungarbled grades 1 and 2, Tellicherry Garbled Special Extra Bold, Tellicherry Garbled Extra Bold, Tellicherry Garbled, Garbled Light Black Pepper in Special and grades 1 and 2, Ungarbled Light Black Pepper in Special and grades 1 and 2, Pinheads in Special and grade 1, and Black Pepper Non-specified, grade X.

For small cardamom, the Agmark grade designations were laid down in the Cardamom Grading and Marking Rules of 1962, under the Agricultural Produce (Grading and Marking) Act of 1937. About 22 specifications were laid down for different qualities, such as Alleppey Green, Coorg Green, Bleached, Half Bleached, Bleached White and Mixed cardamoms, and cardamom seeds. The Alleppey Green grades are AGEB for Extra Bold, AGB for Bold, AGS for Superior, AGS 1 for Shipment Green 1, AGS 2 for Shipment Green 2, and AGL for Light.

Green cardamom's age is easy to judge, because fading of the green colour closely parallels deterioration in flavour strength as volatile oil evaporates, while bleaching makes colour uniform and hides age.

For ginger, elaborate grades were formulated under the Ginger Grading and Marketing Rules of 1964, with 12 specified and 4 non-specified grades, and the requirements are set out in the Indian Standards specification IS: 1908-1961, covering the size of the rhizome, the lime content as calcium oxide, and the permissible limit of extraneous matter for each grade. Export of ginger that does not come up to the specifications is prohibited. The named grades include Garbled Non-bleached Calicut, Ungarbled Non-bleached Calicut, Garbled Non-bleached Cochin, Ungarbled Non-bleached Cochin, Garbled Bleached Cochin and Ungarbled Bleached Cochin.

In each case the extraneous matter limit is 2.0%, garbled grades allow no very light pieces, ungarbled grades allow up to 10% very light pieces by count, and bleached grades allow up to 1.0% lime as calcium oxide by weight. In international markets, ginger is usually received in three grades: 3-fingered rhizomes, 2-fingered rhizomes, and pieces, and the European market prefers the first two grades, mixed in the ratio 20 to 80.

Quick check

A buyer is offered two lots of small cardamom, one green and one bleached, at the same price. Which lot tells its age more honestly, and why?

Check your answer

The green lot. Fading of the green colour closely parallels deterioration in flavour strength through volatile-oil evaporation, so the age of green cardamom is more readily seen than that of bleached cardamom, where the colour has been made uniform by bleaching.

26.6 Packaging: keeping water, air, light and smell out

Spices are hygroscopic: they absorb moisture from the air, notably during the rainy season with high humidity, and that can result in mould attack and insect infestation in whole spice, and in caking, hydrolytic rancidity, mould and insect infestation and loss of aroma in spice powder. Pepper needs protection against ingress of moisture and oxygen, loss of aroma or flavour, gain of foreign odour, and protection against light and heat or high storage temperature.

For cardamom, the factors that preserve quality in storage are moisture content, avoidance of physical damage and excessive heat, and, for green cardamom, exclusion of light. Green cardamoms should be packed in wooden boxes or tins lined with heavy-gauge black polyethylene, metal foil or waterproof paper, and new jute bags with a waterproof lining are also used where presentation is less significant, as in the packaging of lower grades. Ground pepper packed in laminated foil kept for about 6 months at 37.8 °C and 90% relative humidity, and about 1 year at 21.1 to 23.9 °C and 80 to 85% relative humidity.

🤔 Pause and think: a merchant packs ground pepper in a paper bag and whole pepper in a foil-lined pouch. Which one loses its smell first, and which enemy is the paper bag letting in?

Quick check

Why does ground pepper need foil, while whole pepper can be packed in a polyethylene pouch?

Check your answer

Ground pepper has far more surface exposed to air and moisture, so it cakes, goes rancid, grows mould and loses aroma faster. Foil blocks moisture, oxygen and light, and ground pepper in laminated foil kept for about 6 months even at 37.8 °C and 90% relative humidity.

26.7 Storage: the warehouse, the stack and the fumigant

Storage is one of the important functions in orderly marketing of any crop, and spice growers who own only small holdings are often forced to dispose of their produce immediately after harvest, sometimes even before the harvest, because they lack suitable storage facilities and need money at once. That immediate disposal causes a slump in prices from large arrivals in the market, and the grower generally receives a very low price.

A scientific warehouse is built to keep the enemies out. Godowns are generally of high plinth with cemented floor, well-plastered walls and leak-proof roof, and they are rat-proof, bird-proof and damp-proof, with controlled ventilation. Different crops, and even different spices, require different disinfestation measures, so they are stored in separate rooms.

Table 26.2: Optimum storage conditions for fresh ginger, from the CFTRI technique reported in the book

Storage temperatureRelative humidityStorage life
2 to 5 °C90%120 days
21 to 30 °C60 to 90%70 days
46 °C, accelerated storage temperature40%4 days

A scientific spice store: four aspects of one store

Building
High plinth, cemented floor, plastered walls, leak-proof roof, rat-proof, bird-proof, damp-proof, controlled ventilation
Inside
Separate rooms for different spices; pepper and ginger kept away from chillies and odour-producing goods
Stacking
Wooden dunnage under the bags, block stacking, 75-centimetre alleyways between stacks and walls
Care
Periodical inspection, prophylactic fumigation, dusting or spraying; warehouse receipt can be pledged for cash advances

Figure 26.3: A scientific spice store, as the pepper chapter describes it. The building, the separation, the stacking and the inspection each do one job. Drawn from Pruthi, Major Spices of India, Ch. 2.

Quick check

Why does a warehouse keep pepper and ginger in a different room from chillies?

Check your answer

Chillies and other odour-producing commodities pass their smell to neighbouring goods. Pepper and ginger are kept away from them to maintain their mild flavour.

26.8 Decontamination: fumigation, steam and irradiation

Spices carry a heavy load of microbes from the field and from sun drying on the ground, and the books give three lines of defence: fumigation in storage, heat and steam in processing, and irradiation, which Unit 6 covers in full. This section deepens the first two and points to Unit 6 for the third. To prevent infestation during storage, fumigation with methyl bromide or phosotoxin, or dusting with pyrethrum, is adopted.

To prevent the drug-store beetle from attacking the plant, pepper is fumigated with chlorosol. Turmeric and coriander are fumigated with an EDCT mixture, besides other fumigants.

Quick check

Name the fumigant the book gives for pepper against the drug-store beetle, and the fumigant mixture it gives for turmeric and coriander.

Check your answer

Pepper is fumigated with chlorosol against the drug-store beetle. Turmeric and coriander are fumigated with an EDCT mixture, besides other fumigants. The book gives no doses, so none are stated here.

26.9 Practical note: drying a small batch and judging it

Dry a small batch of chillies by the book's traditional method: spread them thinly in the sun on a clean mat or concrete, stir them during the day, heap them and cover them at night, and bring them in if rain threatens. Weigh them each day until the weight stops falling: that is the point where they have reached about 10% moisture, the target Unit 6 records.

Try it yourself: Buy a small bag of whole black pepper and a small bag of ground black pepper of the same brand. Pack half of each in a sealed foil pouch and half in an open paper bag. Leave all four on a shelf for a month, then grind the whole pepper fresh and compare the smell of the four lots. The foil-packed whole pepper should smell strongest, the paper-packed ground pepper weakest, and the gap between them is the packaging lesson of section 26.6 in your own kitchen.

26.10 Check yourself: quick self-check

  1. A pepper grower harvests unripe berries, dries them badly, then cleans and re-dries them carefully at a packing centre. Will the lot recover its quality?
Check your answer

No. The book states that no amount of cleaning and re-drying at a central packing centre will restore the quality of a carelessly dried primary crop, and only well-matured berries must be harvested.

  1. Small cardamom is cured at about 82 °C in one schedule and at 54 °C, then 43 °C, then 54 °C in another. What does each schedule protect, and why does the book report both?
Check your answer

Both protect the green colour and the volatile oil by controlling heat and moisture: the kiln schedule dries for about 30 hours at about 82 °C with trays turned, and the three-stage schedule fixes colour at 54 °C for 3 hours, dries at 43 °C for about 30 hours, and finishes at 54 °C for 3 hours. The books do not reconcile them, so this unit reports both as reported practices.

  1. The CFTRI chilli method dips chillies in Dipsol. What is in Dipsol, and what does the method gain over traditional sun drying?
Check your answer

Dipsol is a water-based emulsion with potassium carbonate at 2.5%, refined groundnut oil at 1%, gum acacia at 0.1% and butylated hydroxyanisole at 0.001%. The method dries in about a week against 15 to 21 days, uses less space, retains colour and pungency better, gives a more hygienic product, and yields 2% more finished product.

  1. Why does artificial drying of chillies stop at 65 °C, and what happens at 80 °C?
Check your answer

At 65 °C, whole fruits dry to 8% moisture in 12 hours and sliced fruits in 6 hours, with good colour. At 80 °C there is a lowering of pungency, of initial colour and of colour retention. One source in the book recommends 45 to 50 °C instead, and the book reports both.

  1. A warehouse keeper stacks pepper bags flat on a damp floor, against the wall, with no gaps. Name three things wrong with that stack, from section 26.7.
Check your answer

Any three of: no wooden dunnage under the bags, so they absorb moisture from the floor; no 75-centimetre alleyways between stacks and walls, so there is no ventilation or access; spices stacked in the same room as chillies or other odour-producing goods; or no periodical inspection and prophylactic measures.

  1. Bleached cardamom looks white and uniform. Why do most traders in the book still call bleaching an unnecessary luxury?
Check your answer

It is costly and time-consuming, it improves only the outer appearance of the pericarp, which is often thrown away, and a slight improvement in keeping quality, while it reduces the volatile oil in the seeds and imparts a sulphurous odour.

Try it yourself: Visit a spice shop and look at three lots: whole green cardamom, whole black pepper, and ground turmeric. For each lot, write down the colour, whether the pieces are whole or broken, and whether you can smell the spice through the pack. Then match each observation to a step in this unit: harvest stage, drying, grading, packaging or storage. The lot that fails on smell has usually failed on drying or packaging, not on grading.

Key ideas to carry forward

  • Post-harvest technology runs in order: drying, cleaning or garbling, grading, packaging, storage and marketing, and a careless early step cannot be fixed later.
  • Harvest stage decides quality: pepper at 1 or 2 red berries a spike, cardamom fruit by fruit at the paling stage, chillies ripe and partly withered on the plant, ginger for drying at 245 to 255 days and never beyond 260 days, garlic when the top has fallen and 75% of leaves are yellow.
  • Drying methods differ by spice: pepper sun drying with a 1-minute boiling water dip, cardamom kiln curing at about 82 °C for about 30 hours or three-stage curing at 54 °C, 43 °C and 54 °C, chillies at 65 °C for 12 hours whole or 6 hours sliced, turmeric boiled 45 to 60 minutes then sun dried 10 to 15 days, nutmeg at 50 °C for 28 hours to 9% moisture, mace at 50 °C for 5 to 6 hours to 5 to 5.5% moisture.
  • Grading is a named system for pepper, small cardamom and ginger, a trade practice of 3 or 4 grades for chillies, and absent for large cardamom, where sales are made on sample or consignment.
  • Packaging must block moisture, oxygen, light and foreign odour: foil-lined packs for ground pepper keep about 6 months at 37.8 °C and 90% relative humidity, and black polyethylene-lined packs keep green cardamom colour.

Glossary

Garbling

Cleaning dried spice to remove dirt, grit, stones, stalks, leaves and light or broken pieces. Food tech example: pepper garbled by winnowing and by sieve-cum-air classifier before grading.

Curing

Holding a harvested spice under controlled heat, and sometimes smoke or steam, before or during drying, so its colour and flavour develop and fix. Food tech example: cardamom cured in a kiln at about 82 °C for about 30 hours.

Blanching

A short dip in boiling water before drying. Food tech example: pepper berries dipped for about 1 minute, which turns them black uniformly, speeds drying and cuts microbes.

Case hardening

A hard dry skin forming on the outside of a piece during fast hot drying, which traps moisture inside and slows further drying. Food tech example: nutmeg dried at 65 °C, where rattling kernels still held 11% moisture.

Flue-pipe curing

Drying in a room heated by hot air passing through pipes from a fireplace, so smoke never touches the spice. Food tech example: large cardamom cured in about 24 hours to a deep maroon colour, against about 3 days in a smoke kiln.

Dipsol

A water-based emulsion used to dip chillies before sun drying in the CFTRI method. Food tech example: potassium carbonate 2.5%, refined groundnut oil 1%, gum acacia 0.1% and butylated hydroxyanisole 0.001%.

Agmark

The grading and marking system of the Directorate of Marketing and Inspection, with named grades and specifications for each spice. Food tech example: Tellicherry Garbled Special Extra Bold pepper, and Alleppey Green AGEB cardamom.

Dunnage

Wooden crates, mats or polyethylene sheets placed under stored bags to keep them off a damp floor. Food tech example: wooden crates with one layer of mat, or polyethylene sheets sandwiched between two layers of mats, under spice bags.

Fumigation

Treating stored spice with a gas or vapour to kill insects and other pests. Food tech example: pepper fumigated with chlorosol against the drug-store beetle, and turmeric and coriander with an EDCT mixture.

Hygroscopic

Absorbing moisture from the air. Food tech example: pepper absorbing moisture in the rainy season, leading to mould, caking and loss of aroma if the pack lets damp in.

References

  • Pruthi, J. S. Major Spices of India: Crop Management and Post-Harvest Technology. CAB International. The post-harvest sections of the crop chapters are the sources for this unit: black pepper, small cardamom, large cardamom, chillies, ginger and turmeric.
  • Das, P. C. Spice Crops Production Technology, 2014. Cardamom harvesting, curing and bleaching; turmeric curing, drying and polishing; garlic harvesting and curing.
  • Shanmugavelu, K. G., Kumar, N., and Peter, K. V. Production Technology of Spices and Plantation Crops. Cardamom harvesting and curing; nutmeg harvesting, curing, drying and packaging; vanilla curing.

Unit 27: Plantation Crops: Coconut, Arecanut and Cashewnut

Note

Unit 27: Plantation Crops: Coconut, Arecanut and Cashewnut

A coconut palm, an arecanut palm and a cashew tree can stand in the same district and look like neighbours, but they are three different kinds of farming. One is a tall palm whose fruit is picked every month of the year.

A plantation crop, as Dashora defines it, is a crop grown on an extensive scale in a large contiguous area, owned and managed by an individual or a company. The major plantation crops in the book are coconut, arecanut, cashew, oil palm, coffee, tea and rubber, with betelvine and cocoa as minor ones.

Pause and think: a coconut palm gives fruit all year, a cashew tree gives fruit for a few months, and an arecanut palm gives bunches that are picked for three different products. How would each one change the way a farming family plans its year?

Quick check

What makes a crop a plantation crop in the book's definition, and why does that matter for how long a grower must wait?

Check your answer

A plantation crop is grown on an extensive scale in a large contiguous area, owned and managed by an individual or a company. The crops in this unit are perennial trees and palms that take years to bear and then crop for decades, so the grower invests years before the first real income.

27.1 Coconut: the palm that gives all year

Coconut, Cocos nucifera, in the family Arecaceae, is called Kalpavriksha in the book, the tree that gives everything. Every part of it is useful: food, drink, shelter and raw material for industry. The products of commercial importance are copra oil, oil cake, desiccated coconut, fibre, shell and timber. The fruit is a fibrous drupe weighing 1.2 to 2.0 kg. A mature dehusked nut holds three raw materials: wet meat or kernel at about 50 percent, water at about 17 percent, and shell at about 33 percent. The kernel is the endosperm, and the part used is the endosperm as copra.

The nut and its uses. The kernel is an important food wherever coconut grows. The milk or cream got by squeezing grated kernel goes into foods of commercial importance. Tender coconut water, the liquid endosperm of a seven month old nut, is a popular refreshing drink, with a caloric value of 17.4 per 100 g of water.

Climate and soil. Coconut is a tropical crop that grows well in a warm, humid area from sea level up to 1,000 m. It does not tolerate extremes of temperature, and it fails where long dry spells and severe cold prevail.

Varieties. The book divides coconut varieties into tall, dwarf and hybrid. Tall palms come to bearing in 6 to 8 years and live and yield for 70 to 75 years. Dwarf palms bear in about 4 years. Hybrids bear in the 4th year and yield more than either parent type in many cases.

Table 27.1: Coconut varieties, as given in Dashora, with yield and copra figures

VarietyTypeBearing startsYield (nuts per palm per year)Copra per nut and oil
East Coast TallTall6 to 8 years75 to 100154 g; 62 to 65% oil
Laccadive Ordinary (Chandrakalpa)TallNot givenAbout 127155 g; 72% oil
Double Century (Philippine Ordinary)Tall6 to 8 years130200 g; 66% oil
West Coast TallTallNot given20 to 100, average 70180 g; 68% oil
GangabondamSemi dwarfAbout 4 yearsAbout 60Not given; sweet water, medium nuts
ChennangiDwarf4th year70 to 80Not given; round green nuts
Chowghat Dwarf OrangeDwarfNot given50, range 40 to 120125 g; 66% oil
Malayan Dwarf YellowDwarfNot givenAbout 70130 g; 66% oil
Malayan Dwarf OrangeDwarfNot given50150 g; 67% oil
Tall x Dwarf hybridsHybrid4th year80Not given
Dwarf x Tall hybridsHybrid4th year100 to 200Not given

Shanmugavelu gives a plainer summary: tall varieties give on average 60 to 80 nuts a year, Dwarf Green about 70, and Dwarf Orange about 90. That summary matches the table for tall varieties and Dwarf Green, but the books differ on Dwarf Orange: Dashora gives 50 nuts per palm per year for Chowghat Dwarf Orange and for Malayan Dwarf Orange, while Shanmugavelu gives about 90.

Propagation and planting. Coconut is propagated commercially by seed nut. Because it is cross pollinated, it does not breed true, so choosing the seed nuts and then choosing the seedlings in the nursery matters more than anything, and the result can only be judged years after planting. Seed nuts are taken from high yielding, middle aged, regular bearing mother palms giving not less than 100 nuts per year, from the October to March harvests, fully mature, medium sized, with little water. They are kept in shade in layers, covered with sand and sprinkled with water so they do not dry.

Nutrition and irrigation. Potash is the main nutrient need of the palm. Deficiency shows as yellowing or bronzed leaflets on short fronds and poor yielding bunches. Coconut husks are rich in potash and may be used as mulch around the palms, buried in trenches, or burnt and the ash applied. About 80 percent of the roots sit within 2 m of the trunk, so fertiliser is applied within that limit. Farmyard manure is given at 12.5 tonnes per hectare before planting and 50 kg per palm per year after planting, and the fertiliser dose per palm rises with age, reaching 170 g nitrogen, 120 g phosphorus and 400 g potassium before the monsoon and 330 g, 200 g and 800 g after it in the fourth year. Young palms need 45 litres of water once in four days in summer for the first three years, and with drip irrigation 30 to 35 litres go through 4 emitters at about 1 m from the palm.

Flowering and fruit set. The palm is monoecious, with separate male and female flowers on each spadix, borne in the axil of each leaf of a bearing palm. Flowering starts at 6 to 12 years of age. An inflorescence is about 1 to 2 m long, with a central axis and about 40 lateral branches called spikelets.

Harvest and yield. An inflorescence is produced about every month, and the fruit takes 12 months from setting to full maturity, so harvesting goes on all year. Harvest is normally every 30 days in summer and every 45 days in the rainy season, and in some places three times a year. Nuts are harvested by experienced climbers, who climb with a rope ring round the feet or ankles, examine the nuts for maturity and cut the bunches down. The maturity index for harvesting nuts is 18 percent total soluble solids.

Intercropping. Seedlings are planted 8 m apart, so the wide space between them invites other crops. Note that the two spacings in this unit come from different passages of the book: 9 by 9 m for tall palms and 7.5 by 7.5 m for dwarf and hybrid palms in the varieties section, and 8 m apart in the intercropping paragraph. In heavy rainfall areas the annual intercrops are potato, tapioca, elephant foot yam, chillies and other vegetables, and the perennial ones are banana and pineapple from the start, with cocoa, clove, nutmeg and cinnamon in plantations more than 12 to 15 years old.

Pests and diseases. The black headed caterpillar (Opisina arenosella) scrapes the green matter from leaves and lives in galleries of silk and excreta on the lower leaf surface; in severe attack the whole plantation looks burnt up from a distance. The red palm weevil (Rhynchophorus ferrugineus) has grubs that feed on soft tissues in the trunk of 2 to 10 year old palms, with brown fluid oozing from holes and chewed fibre sticking out. The rhinoceros beetle (Oryctes rhinoceros) bores into unopened tender fronds and spathes and throws out chewed fibrous mass, and the adults breed in manure heaps.

Seed nut from a mother palm giving 100 nuts a year, germinated 3 to 4 months, seedling chosen at 9 to 12 months

↓

Planted 9 by 9 m (tall) or 7.5 by 7.5 m (dwarf and hybrid), first fruit at 4 years (dwarf/hybrid) to 6-8 years (tall)

↓

A new inflorescence every month, fruit ripe in 12 months, harvest every 30 to 45 days all year

↓

Nut split into kernel (50%), water (17%) and shell (33%): copra dried to 5 to 6% moisture, 80 to 100 nuts per palm per year

Figure 27.1: Coconut from seed nut to copra. One palm flowers every month, so the harvest never has a season. Drawn from Dashora and others, Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants, Coconut chapter.

27.2 Arecanut: the slender palm with three products

Arecanut, Areca catechu, in the family Arecaceae, is also called betel nut or supari. It is a slender palm, and its fruit is a berry whose kernel is the part used. The book says chewing arecanut has a tonic effect on the body, and nerve stimulating and vermicidal properties, and the extracts got by boiling the nuts hold tannin, which is used for tanning leather.

Climate and soil. Arecanut is a humid tropical palm. It does well with rainfall of 200 to 375 cm and temperatures of 15 to 30°C, and it can be grown at elevations of about 80 m to 1,000 m from sea level. It can also be grown where rainfall is less, if the temperature is not high. Red loam, lateritic loam or alluvial loam with adequate drainage suits it. Heavy, alkaline or waterlogged soils are to be avoided.

Varieties. Indigenous tall types are commonly grown, and the book names six improved varieties with yields.

Table 27.2: Arecanut varieties and yields, as given in Dashora

VarietyRecommended forRipe nuts (quintals per hectare)Chali (quintals per hectare)
MangalaCoastal Kerala and Karnataka20028
SumangalaCoastal Kerala and Karnataka23645
Sree MangalaKarnataka and Kerala21442
MohitnagarNorthern West Bengal, coastal Karnataka and Kerala20750
CAL 7Andaman and Nicobar Islands25960
SAS 1Hilly areas of Karnataka239Not given

Mangala is semi tall with a partly drooping crown, early bearing and high nut set, and dark yellow to orange ripe nuts. Sumangala is tall with a partly drooping habit, flowers in 4 to 5 years, and gives deep yellow to orange, oblong to round, bold nuts.

Propagation and nursery. Arecanut is propagated only through seed. Seed nuts are collected from selected high yielding mother palms, 5 years after their first bearing. Besides high yield, early first bearing and a nut set above 50 percent matter in selection.

Planting. Land is ploughed, harrowed and levelled. Good perennial shade is provided on the western and southern borders with dense leafy shady trees such as jackfruit, neem, casuarina or pongamia. Pits 90 cm cube are dug at 2.7 to 4.0 m by 2.7 to 4.0 m, filled with topsoil mixed with compost, and seedlings are planted in the centre 7.5 to 15 cm below ground level to make watering easy.

Nutrition and irrigation. Manuring is done in basins 15 to 20 cm deep and 1 m wide around the palm. Farmyard manure or compost is given at 25 tonnes per hectare in nursery beds and 20 kg per plant per year after transplanting, and the book also recommends 25 kg of organic manure as green leaf, compost or farmyard manure in a single dose in May to June. Fertilisers per plant per year are 100 g nitrogen, 40 g P₂O₅ and 140 g K₂O, in two split doses, in May to June and September to October, broadcast around the base and forked in. Arecanut is generally grown as a rainfed crop, but watering raises the yield: once in 7 days in November to December, once in 6 days in January to February, and once in 3 to 5 days in March to May, with drip irrigation recommended.

Harvest, yield and processing. Bunches are harvested when fully ripe if the end use is chali or kotapak, the dried ripe nuts. Nuts about 6 to 7 months old, dark green and soft, are harvested to make kalipak. In a regularly spaced garden the climber climbs a palm at one end, harvests the bunch, then pulls the nearest palm toward it and swings across to it: one climber may harvest up to 100 palms by swinging from palm to palm at a stretch before coming down.

The three trade types are chali or kotapak, kalipak and scented supari. For chali, ripe nuts are sun dried for 35 to 40 days, then dehusked and sold as whole nuts. The grades of chali, in decreasing order of size, are moti, srivardhan, jamnagar and jini.

Multiple cropping. Cocoa is planted at 2.7 m between alternate rows of standing arecanut palms, and black pepper trained on arecanut is another popular system. Cinnamon, coffee, betel vine and cardamom are also grown with arecanut in some areas, and banana, pineapple, elephant foot yam, tapioca, dioscorea, sweet potato, ginger and turmeric are ideal intercrops, with cocoa the most popular.

Pests and diseases. Mites suck the lower leaf surfaces and turn the leaves yellow and bronzed. The spindle bug sucks sap from the tender spindle, so the palm loses vigour and dies.

One palm, three products: the age of the nut decides which one

The palm
Bunches picked at two different ages, for two different products
Branch 1: Ripe nut
Fully ripe nut: sun dry 35 to 40 days, dehusk: chali, the dried ripe nut
Branch 2: Green nut
Green nut, 6 to 7 months, dark green and soft: dehusk, cut, boil, coat with kali: kalipak
Both branches feed into scented supari
Chali or kalipak broken into bits, scented with spice oils, rose essence or menthol: scented supari

Figure 27.2: One palm, three products. The age of the nut at harvest decides which product it becomes: ripe nuts become chali, green nuts become kalipak, and either one can become scented supari. Drawn from Dashora and others, Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants, Arecanut chapter.

Quick check

Why does the age of the nut at harvest matter so much for arecanut?

Check your answer

Because the product is decided by age: fully ripe nuts are sun dried into chali, and 6 to 7 month old dark green, soft nuts are cut and boiled into kalipak. Pick at the wrong age and the nut cannot become the product the buyer wants.

27.3 Cashewnut: the nut below the apple

Cashewnut, Anacardium occidentale, in the family Anacardiaceae, is native to Brazil. Its English name comes from the Portuguese caju, from the Tupi word acaju. The part used is the nut kernel.

Climate and soil. Cashew is a hardy, drought resistant tropical crop. It does well in the hot, humid climate of the west coast, and can be grown up to 750 m above sea level with 500 to 3,750 mm of annual rainfall. It is very sensitive to low temperature.

Varieties. The book names varieties with tree yields and shelling percentages, the share of kernel in the nut.

Table 27.3: Cashew varieties, yields and shelling percentage, as given in Dashora

VarietyYield (kg nuts per tree)Shelling (%)Notes as given
Ullal-119.630.7Early bearing, high sex ratio; 144 nuts per kg; yellow apple
Ullal-217.930.0Early flowering and bearing; suits hilly areas; 164 nuts per kg
Ullal-314.6830.0Flowers November to January, 60 to 70 days; harvest January to March
BPP-542.126.7Selection from Bapatla
BPP-642.524.0Selection from Bapatla; highest yielder named
Selection S-2 and S-38 to 1029.0Released from Puttur for Karnataka

Other varieties named are Chintamani-1, Jhargram-1, Ana-1, Hybrid Dhara, NRCC-1, NRCC-2, V-1, V-2, VRI-1 and VRI-2. Ullal-1 flowers from the last week of November to the middle of March, and harvesting is finished by May, which lets it escape the monsoon showers that would otherwise lose the crop.

Propagation. Cashew is propagated by seed and vegetatively. Seedlings are raised in polythene bags 15 by 20 cm, sown with the stalk pointing up, and ready to plant in 2 to 3 months. The vegetative methods are air layering, stooling, epicotyl grafting and softwood grafting.

Planting and nutrition. Pits 60 cm cube are dug 8 m apart, filled with topsoil and 10 to 15 kg of organic manure per pit, and seedlings or air layers are planted in the centre during the monsoon. Fertilisers are applied in September to October, after the heavy rains are over. Before planting, farmyard manure or compost at 2.5 tonnes per hectare is mixed in, and a similar dose is applied every year. Fertilisers go into trenches 1.0 to 1.5 m from the base, rising with age to 250 g nitrogen, 60 g phosphorus and 60 g potassium at each of two dressings in the fourth year, and manganese deficiency is met with Mancozeb or manganese sulphate at 0.2 percent.

Harvest and yield. Economic yield starts in the third year, rises markedly up to the 10th year, and the plant bears continuously for a further 20 years. The shell of the nut reaches full size in 3 to 4 weeks after flowering, then the pedicel swells to make the cashew apple. The fruit matures in 2 to 3 months, and the fully ripened nuts are removed from the apples and sun dried for about 2 to 3 days before storage.

Processing. The share of whole kernels at the end runs between 55 and 95 percent, depending on method and operation, and 65 percent recovery is considered satisfactory. Manual shelling in the book's account gives a considerably higher yield of whole kernels, 90 to 95 percent, than mechanical processing. The steps are: passing the nuts over a sieve or through a perforated drum to remove stones, twigs and foreign matter; humidifying and roasting, after grading by width and thickness; conditioning the nuts to about 16 percent moisture before oil bath roasting, so the cells holding cashew nut shell liquid burst during roasting; roasting itself, in the old way in a perforated open pan over a wood fire, where the liquid drips through the holes and burns with irritant fumes, or in the newer hot oil bath method, where conditioned nuts in baskets are immersed in a tank of cashew nut shell liquid at about 170 to 200°C for about 90 seconds, giving more uniform roasting and a higher yield of white whole kernels, with 6 to 8 percent excess liquid recovered; centrifuging after roasting, or mixing with sawdust in a revolving drum, to remove the liquid still sticking to the shells; cooling in air; shelling, by hand on a flat stone with a wooden mallet, with gloves and wood ash to protect the hands from the liquid, at about 10 nuts a minute, or about 5 kg of kernels in an eight hour day, with experienced shellers producing twice as much and up to 90 percent whole kernels, or by semi mechanised knives; drying the shelled kernels from more than 6 percent moisture to about 3 percent, so the testa shrivels and peels easily; peeling by gentle rubbing with the fingers, or with bamboo knives where the testa still sticks, at 10 to 12 kg of kernels per person per day; and grading by size and colour of the whole kernels and pieces. Roasting is a delicate operation, Shanmugavelu adds: it adds flavour and taste, makes the shells brittle and extraction easier, and a slight under roasting or over roasting harms quality and recovery.

Pests and diseases. The stem borer (Plocaederus ferrugineus) is a serious pest: its grub bores in the bark and feeds on the tissues under the skin and the vascular tissues, making tunnels. The tea mosquito (Helopeltis antonii) is also serious: nymphs inject toxic saliva, the affected part turns brown, and young leaves and inflorescence dry up. The leaf miner (Acrocercops syngramma) mines tender leaves at the time of new flushes.

Flower, November to February: shell reaches full size in 3 to 4 weeks

↓

Pedicel swells into the cashew apple above the true nut: fruit ripe in 2 to 3 months

↓

Nut removed, sun dried 2 to 3 days, conditioned to 16% moisture, roasted at 170 to 200°C for 90 seconds

↓

Shelled, dried to 3% moisture, peeled, graded: 65% whole kernel recovery is satisfactory, hand shelling reaches 90 to 95%

Figure 27.3: Cashew from flower to graded kernel. The apple is the swollen stalk, the nut is the true fruit, and the shell liquid is why roasting comes before shelling. Drawn from Dashora and others, Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants, Cashewnut chapter.

Try it yourself: buy a few raw cashew nuts in the shell if you can find them, or look at a photo of one, and a cashew apple if a market near you sells them. Draw the fruit as it hangs: the kidney shaped nut at the bottom, the swollen apple above it. Label which part is the true fruit and which part is the swollen stalk, and mark where the shell liquid sits. Then write one sentence explaining why the nut is roasted before it is shelled.

27.4 Three crops side by side

Table 27.4: Coconut, arecanut and cashewnut compared, from the books

PointCoconutArecanutCashewnut
Botanical nameCocos nuciferaAreca catechuAnacardium occidentale
FamilyArecaceaeArecaceaeAnacardiaceae
Origin as givenMalaysia, Indonesia (South East Asia)Malayan ArchipelagoBrazil
Part usedEndosperm (copra)Kernel (fruit)Nut (kernel)
First bearing4 years (dwarf/hybrid) to 6-8 years (tall)Flowers at 4 to 5 years3rd year; full bearing by 10th year
Harvest rhythmAll year, every 30 to 45 daysBunches as they ripenSeasonal, peak March to April
Typical yield80 to 100 nuts per palm per year200 to 260 quintals ripe nuts per hectareAbout 3 kg nuts per tree per year; 17 kg from high yielding clones
Main productsCopra, oil, desiccated coconut, fibre, shell, tender waterChali, kalipak, scented supariKernel, cashew apple, shell liquid
Key pestRed palm weevil, rhinoceros beetleSpindle bug, mitesStem borer, tea mosquito
Key diseaseBud rot, basal stem rotKoleroga, bud rot, yellow leaf diseaseDieback (pink disease), damping off

Questions to think about

  • Coconut is harvested all year and cashew is harvested in a short season. Which crop needs more labour at one time, and which one needs labour all year?
  • Arecanut gives three products from one palm, and the product is decided by the age of the nut. What would happen if a grower picked every bunch at the same age?
  • Cashew shell liquid is both a hazard and an industrial raw material. How does the processing chain turn that hazard into a product?

A visual sorting game: make nine cards, one for each product: copra, tender coconut water, coir fibre, chali, kalipak, scented supari, cashew kernel, cashew apple, cashew shell liquid. Shuffle them, then sort them into three piles by crop, and then into two piles by use: food and drink, and non food or industrial use. The fibre, shell and shell liquid cards are the test: they belong to the non food pile even though they come from food crops.

Practical note: handling the harvest without losing it

Three rules run through all three crops. First, pick at the right stage and test it: a coconut is ready at 18 percent total soluble solids and rattles when shaken if it is fully mature; an arecanut bunch is picked ripe for chali and green at 6 to 7 months for kalipak; a cashew nut is picked when the apple is fully ripe and the nut is removed before drying. Second, dry promptly and to the book's moisture: coconut copra to 5 to 6 percent, cashew nuts in the sun for 2 to 3 days before storage and kernels to about 3 percent before peeling, arecanut chali in the sun for 35 to 40 days.

Key ideas to carry forward

  • A plantation crop is a perennial grown on a large contiguous area, and the three crops here take 3 to 7 years to bear and then crop for decades.
  • Coconut is a palm that flowers every month, so it is harvested all year, and its nut splits into kernel, water and shell, with copra dried to 5 to 6 percent moisture.
  • Tall coconut palms bear in 6 to 8 years and live 70 years or more, dwarfs bear in about 4 years, and hybrids bear in the 4th year and yield 80 to 200 nuts a year.
  • Arecanut is propagated only by seed, planted 2.7 to 4.0 m apart, and one palm gives three products: chali from ripe nuts, kalipak from green nuts, and scented supari from either.
  • Cashew is a tree whose true fruit is the nut hanging below a swollen stalk, the apple, and the kernel is won by conditioning, roasting at 170 to 200°C, shelling, drying to 3 percent and peeling.
  • Potash rules coconut nutrition, and the fertiliser dose rises with the age of the palm; cashew answers to manganese, and arecanut to drainage, lime in acid soils and stem protection from the sun.
  • Bud rot kills coconut and arecanut palms from the spindle downwards, and dieback takes cashew flowers before fruit can form: in all three, removing and destroying the affected parts is the first control.

Glossary

Plantation crop

A crop grown on an extensive scale in a large contiguous area, owned and managed by an individual or a company. Food tech example: coconut, arecanut and cashew, grown as perennial palms and trees that bear for decades.

Copra

The dried kernel of the coconut, from which oil is pressed. Food tech example: coconut cups sun dried until the endosperm reaches 5 to 6 percent moisture, then scooped out as copra.

Endosperm

The food store inside a seed or nut. Food tech example: the coconut kernel, the endosperm of the nut, eaten fresh or dried as copra.

Drupe

A fleshy fruit with a single seed inside a hard shell, such as a coconut. Food tech example: a coconut fruit weighing 1.2 to 2.0 kg, with fibrous husk outside and shell inside.

Button drop

The shedding of young coconut fruits before they mature. Food tech example: button drop checked by spraying 1 percent Bordeaux mixture in the rainy season.

Chali (kotapak)

Dried ripe arecanut, sun dried for 35 to 40 days and dehusked. Food tech example: chali graded by size as moti, srivardhan, jamnagar and jini.

Kalipak

Arecanut made from immature nuts, cut, boiled and coated with concentrated extract until glossy. Food tech example: dark green, soft nuts of 6 to 7 months boiled into kalipak.

Scented supari

Chali or kalipak broken into bits and scented with spice oils, rose essence or menthol. Food tech example: scented supari packed in butter paper or foil pouches.

Cashew apple

The fleshy, swollen stalk below the cashew nut, juicy and edible, not a true fruit. Food tech example: cashew apple juice used for syrup and juice, and for cashew fenny.

Cashew nut shell liquid

The caustic oil in the cashew shell, an industrial raw material. Food tech example: shell liquid that must be driven off by roasting at 170 to 200°C before the nut is shelled.

Shelling percentage

The share of kernel in the whole cashew nut, by weight. Food tech example: Ullal-1 with a shelling percentage of 30.7 percent.

References

  • Dashora, L. K., Dashora, Abhay, and Lakhawat, S. S. Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants. 2nd edition, Agrotech Publishing Academy, Udaipur, 2013. ISBN 978-81-8321-275-5. Part A, chapters 1 (Introduction), 2 (Coconut), 3 (Arecanut) and 4 (Cashewnut).
  • Shanmugavelu, K. G., Kumar, N., and Peter, K. V. Production Technology of Spices and Plantation Crops. ISBN 81-7754-161-7. Chapters on Coconut, Arecanut and Cashew, used only where the scanned text is legible.

Unit 28: Aromatic and Medicinal Plants

Note

Unit 28: Aromatic and Medicinal Plants

Smell a vanilla pod and you smell a flavour that was not in the pod when it was picked. Take a piece of cinchona bark and you hold a medicine that was not made in a factory, but grown in the bark of a tree.

Aromatic plants are grown for their scent and flavour, and medicinal plants are grown for compounds that act on the body. The two groups overlap: many aromatic plants are also used as medicine, and many medicinal plants are aromatic.

An honest note belongs at the start of this unit, not at the end. Dashora's book is divided into three sections, and its third section is meant to cover the aromatic and medicinal plants: opium poppy, safed musali, isabgol, ashwagandha, guggal, mehandi, cheti gulab and mint. In the copy of the book supplied for this subject, that third section is missing: the text ends with the fennel chapter, and the aromatic and medicinal chapters are not in it.

Pause and think: vanilla has no smell when it is picked, and quinine has no use until it is taken out of the bark. What does that tell you about where the value of an aromatic or medicinal plant actually sits: in the field, or in what happens after harvest?

Quick check

What is the difference between an aromatic plant and a medicinal plant, and why do the two groups overlap?

Check your answer

An aromatic plant is grown for scent and flavour, and a medicinal plant is grown for compounds that act on the body. They overlap because the same compound can do both: vanilla's vanillin flavours food, and cinchona's quinine treats fever, while many aromatic oils are also used as medicine.

28.1 How the flavour and the medicine are got out

Three ideas run through this unit, and each one is shown by a worked example in the sections that follow. First, the compound is often not present in its final form at harvest: vanilla pods hold no vanillin when they are picked, and the vanillin is made by enzyme action on glucosides in the bean during curing. Second, the compound is often locked into the tissue with other substances and must be freed before it can be used: cinchona alkaloids sit in the bark combined with organic acids, and the factory process frees them with lime and caustic soda, takes them into a solvent, and then moves them into acid water before they crystallise out.

Questions to think about

  • If a vanilla pod has no vanillin at harvest, what is the grower actually harvesting?
  • Why would a factory dry cinchona bark quickly and at low temperature, at 65°C, rather than drying it fast and hot?
  • Wattle bark is graded by tannin percentage and by thickness. Why would a thicker bark hold more tannin?

28.2 Vanilla: the aromatic orchid

Vanilla is an orchid grown for its pleasant flavour. It belongs to the tropical genus Vanilla in the family Orchidaceae. Over 50 species are described, the book says, but only three matter as sources of vanilla: Vanilla planifolia, Vanilla pompona and Vanilla tahitensis. Vanilla planifolia, formerly known as Vanilla fragrans, is the most preferred commercially and the most widely grown. It is native to the Atlantic coast from Mexico to Brazil. The word vanilla comes from the Spanish vaina, a sheath or pod, with the diminutive meaning little pod.

The plant. Vanilla is a climbing orchid with opposite, sessile leaves 10 to 22.5 cm long and oblong in shape. Aerial roots grow at the nodes.

Climate and soil. Vanilla does well in a humid tropical climate at 700 to 1,200 metres elevation, with 250 cm of annual rainfall. It needs warm, moist conditions at 21 to 25°C.

Propagation and planting. Vanilla is generally propagated by stem cuttings. Vines 60 to 120 cm long are selected as planting material. Trials at the Horticultural Research Station, Thadiankudisai, found that 1.0 m long cuttings rooted 100 percent, better than four, three, two and single noded cuttings.

Because vanilla is a climbing vine, it needs a support and partial shade of about 50 percent sunlight. Low branching trees with rough bark and small leaves are grown for this: jatropha, plumeria, casuarina, and species of Erythrina and Glyricidia. If the support is a legume, it also enriches the soil.

Feeding. The decomposed mulch is the main source of nutrient to vanilla. It also holds moisture and gives a loose structure for the roots to spread in, so easily decomposable matter is put around the vines 2 to 3 times a year. Animal manures are generally not favoured by orchids, the book warns. Nitrogen as leaf compost or farmyard manure at 112 g per vine may be applied in two doses, with 112 g of nitrogen as ammonium sulphate or calcium ammonium nitrate and 112 g of potash as muriate of potash per vine in two or three split doses.

Flowering and hand pollination. In Mexico, the original home of vanilla, natural pollination is done by bees of the genus Melipona and by some humming birds. Vanilla starts flowering from the fourth year onwards, during December to March and sometimes in October to November. Vines usually flower in about the third year, and if longer cuttings were used, some may flower in the second year. Note: the book gives both timings, flowering from the fourth year onwards and flowering in about the third year, and this unit reports both as the book gives them.

Each flower must be pollinated by hand, because the pollen is covered by a hood, the anther cap, and the stigma is protected by a lip called the rostellum, and the two together prevent natural pollination. The book advises limiting the number of pods per plant, since overproduction exhausts the vine and the pods are of poorer quality: 6 to 8 flowers in a cluster are hand pollinated. A skilled worker can hand pollinate 1,000 to 1,500 flowers a day, and hand pollination may account for 40 percent of labour cost.

Harvest and curing. When immature, the bean is dark green. When ripe, yellowing starts from the far end, and that is the best time to harvest. Left on the vine, the beans turn yellow all over and start to split.

Three curing methods are described. In the Peruvian process, the pods are dipped in boiling water, the ends tied and hung in the open, dried for twenty days, coated with castor oil and tied in bundles. In the Guiana process, the pods are dried in the sun until they shrivel, wiped and rubbed with olive oil, the ends tied to prevent splitting, and bundled. In the Mexican process, the pods are killed, warmed, sweated for 7 to 12 days, dried slowly, and conditioned for months. The sweating is repeated through those 7 to 12 days, and it is during this curing that the enzymes turn the glucosides into vanillin: the Mexican process gives 4.15 to 4.40% vanillin in the cured bean, the figure shown in Figure 28.1.

Yield and diseases. Yield starts in the third year and rises until the seventh or eighth year, then falls slowly until the vines are replanted after another 7 to 10 years. Under a reasonable level of management, a middle aged plantation gives 300 to 400 kg of cured beans per hectare. Diseases are more serious than pests.

Green pod picked as yellowing starts from the far end: no vanillin in it yet

↓

Curing: kill the pod, warm it, sweat it for 7 to 12 days, dry it slowly, condition it for months

↓

Enzymes turn glucosides into vanillin: dark brown, soft, flexible bean, 4.15 to 4.40% vanillin

Figure 28.1: Vanilla from green pod to cured bean. The flavour is not grown, it is made during curing. Drawn from Shanmugavelu and others, Production Technology of Spices and Plantation Crops, Vanilla chapter.

Quick check

Why must vanilla flowers be pollinated by hand outside Mexico, and why only 6 to 8 flowers per cluster?

Check your answer

The anther cap covers the pollen and the rostellum shields the stigma, so the flower cannot pollinate itself, and the bees that do it in Mexico are not present elsewhere. Only 6 to 8 flowers per cluster are pollinated because overproduction exhausts the vine and gives pods of poorer quality.

28.3 Cinchona: the fever bark tree

Cinchona is the source of quinine, and it has been one of the most important medicinal plants of the last two centuries, the book says. It is native to the highlands of South America, from Bolivia to Colombia and from Panama to Costa Rica, and it is found in Ecuador, Peru, Bolivia, Sri Lanka, Java and, in the book's account, in a few parts of West Bengal, Tamil Nadu and Karnataka.

The tree and its species. Cinchona is a beautiful evergreen tree, 10 to 12 m tall, with a stem 15 to 30 cm across, sparsely branched. The leaves are of various sizes and shapes, generally shining bright green with crimson veins. Quinine is taken from the stem bark or twig bark, and the root bark is richer than the stem bark. More than 25 alkaloids are found in the bark, of which quinine, quinidine, cinchonine and cinchonidine are the most important, and the alkaloids sit in the bark combined with quinic acid, C₇H₁₂O₆, at 5 to 8 percent, and cinchotannic acid at 3 to 4 percent.

Table 28.1: Cinchona species grown commercially, as given in Shanmugavelu

SpeciesBark of commerceQuinineOther marks as given
Cinchona ledgerianaYellow barkHigh; most productive in quinineShort lived, matures about 12th year, best at 915 to 1,370 m
Cinchona officinalisCrown barkNext to C. ledgerianaHardy, long lived, matures about 25th year, best at 1,830 to 2,440 m
Cinchona robustaNot namedGenerally lowFast growing, more bark, hardy, used as rootstock for budding
Cinchona succirubraRed barkLow; other alkaloids highHardiest, fastest growing, longest lived, matures about 40th year
Cinchona hybridaNot namedAverage, 2.0 to 2.5% alkaloidDry bark yield 6,000 to 6,250 kg per hectare after 16 years

The book gives dry bark yield after 16 years as 2,750 to 5,250 kg per hectare for C. succirubra, 2,365 to 4,150 kg for C. ledgeriana, and 3,090 kg for C. robusta. One outstanding clone, No. 701, holds 12 percent quinine sulphate in the bark and is propagated on a very extensive scale in the Nilgiris and Anamalais.

Climate and soil. Cinchona grows well where the average minimum temperature is 13.5°C and the average maximum is 21°C, with a mean relative humidity of 83 percent. It grows poorly below 7°C or above 26°C. Annual rainfall should be around 2,000 mm, spread evenly through the year.

Propagation and planting. Cinchona is propagated by seed and vegetatively, by cuttings, stooling, layering and patch budding. Seeds are sown in April and take 20 to 30 days to germinate. About 50 percent germinate, but only 10 percent of seedlings are fit to transplant after weak, lean and lanky seedlings are removed. Seedlings go into baskets or polythene bags at about four months old, and into the main field in the middle of May of the next year, with the outbreak of the rains.

Nutrition and aftercare. Fertiliser use on cinchona is a recent innovation, the book says. Liming with dolomite or limestone is needed where soil pH is 5 or lower. Triple superphosphate, muriate of potash and ammonium sulphate are applied at 115 kg nitrogen, 105 kg P₂O₅ and 115 kg K₂O per hectare.

Harvest. Trees are coppiced when 8 to 10 years old, depending on growth. Coppicing means pruning the tree at 5 cm from ground level. The stump throws up many shoots, but only two or three are kept and allowed to grow, and the rest are removed.

Extraction of the alkaloids. Cinchona bark is judged by manufacturers mainly on its quinine percentage, and also quinidine, the other alkaloids being regarded as of little or no value. The bark holds about 12 to 16 percent moisture when it reaches the factory. It is ground to a fine powder of 60 mesh and mixed with slaked lime holding over 60 percent calcium hydroxide, and water. The lime fixes the organic acids as insoluble calcium salts.

Uses. The demand for quinine products is rising, the book says, because of their wide use not only as an anti malarial but also in analgesic preparations for common colds, cough, influenza and other fevers. Quinidine, got during processing and also by changing levo quinine into dextro quinidine, is used widely as a remedy for cardiac ailments.

Pests and diseases. Crickets cut young seedlings and drag them into their haunts in the seed beds. The tea fly sucks juice from tender leaves in nurseries and in the field, makes them curl, and weakens the plant if the damage is heavy.

Tree coppiced at 8 to 10 years, cut at 5 cm: 2 to 3 shoots kept, bark beaten and peeled

↓

Bark dried quickly at 65°C to 10% moisture, then powdered: quinine 3 to 4% in seed raised bark, nearly double in clonal stock

↓

Lime frees the alkaloids from their acids, solvent oil takes them up, acid water takes them back, quinine sulphate crystallises at pH 5.5

Figure 28.2: Cinchona from bark to quinine sulphate. The bark is the storehouse, and the factory frees what the tree locked away. Drawn from Shanmugavelu and others, Production Technology of Spices and Plantation Crops, Cinchona chapter.

Try it yourself: take three glasses of water. Stir a spoon of sugar into the first, a spoon of salt into the second, and a spoon of baking soda into the third, and taste a drop of each on a clean spoon. The sugar dissolves unchanged, the salt dissolves unchanged, but the baking soda changes the water. That is the idea behind the cinchona factory: the alkaloids are not simply washed out of the bark, they are freed from their natural salts with lime before the solvent can take them.

28.4 Wattle: the tannin tree

Wattle belongs to the genus Acacia, in the family Fabaceae. Four species give wattle: black wattle (Acacia mearnsii, also called A. mollissima), green wattle (A. decurrens), silver wattle (A. dealbata) and golden wattle (A. pycnantha). They are natives of Australia.

The four species. Black wattle is the most widely grown for high yield and quality. Its foliage is dark green, its leaflets short, flat and closely crowded, its young shoots golden tinged, its flowers yellow at first and later pale cream, and its seed pods narrow.

Climate and soil. Black wattle does well at 1,950 to 2,500 m elevation, with well spread rainfall of 150 cm and 100 rainy days. In regions of high rainfall the bark may be infested with lichens, which lowers its quality. Winds and frosts in the early part of the rainy season, June to July, harm the growth of the trees.

Propagation and planting. The common method is by seed. Before sowing, seeds are given a hot water treatment, then washed with cold water four or five times to remove the gummy mucilage, dried in shade and stored in lead lined boxes, where they keep their viability for six months to one year. Seeds are sown in nursery beds in September, start to germinate in about eight days, and finish within a month. Seedlings of 9 to 12 months are used for planting, and about 450 g of seeds gives 3,500 to 4,000 seedlings, enough to plant 3 to 6 hectares.

Rotation, yield and stripping. A 12 year rotation is used on good sites in South Africa, and 8 to 10 years on poor sites; in the Nilgiris a 10 year rotation is used. Black wattle stands are felled at about ten years, sometimes after a pre felling thinning about 18 months before the end of the rotation, to remove diseased and non strippable trees and ease the main felling. Felling is done by axe or bowsaw at the time of year when the bark strips most easily, when the sap is rising, in the warm, wet summer months. At clear felling the trees are 15 to 20 cm across at breast height and up to 18 to 22 m tall. At the Wattle Research Institute, Pietermaritzburg, the highest yield of tannin per acre came from 11 year old trees at 600 per acre, and both phosphate and potash gave significant and profitable increases in yield.

Seed sown September, seedling planted June to July at 3 by 3 m, thinned in year 3

↓

Black wattle grown 10 years to 15 to 20 cm across and 18 to 22 m tall, bark thickest on the largest trees

↓

Bark stripped in the warm wet months when sap is rising, cut to 1.2 m, bundled at 50 kg, sent fresh to the factory: 30% tannin or more

Figure 28.3: Wattle from seed to tannin bark. The bigger the tree, the thicker the bark, and the thicker the bark, the more tannin it holds. Drawn from Shanmugavelu and others, Production Technology of Spices and Plantation Crops, Wattle chapter.

Quick check

Why must wattle bark reach the factory fresh, and why is a thick bark worth more than a thin one?

Check your answer

Delay causes oxidation, which gives an excessively red coloured extract, or fungal mould, both of which spoil the product. A thick bark is worth more because tannin percentage rises with bark thickness, and bark thickness rises with the diameter of the tree.

28.5 What the books do not cover

This section is part of the unit, not an afterthought, because a reader who trusts this unit should know exactly where its coverage ends. Dashora's third section names eight aromatic and medicinal plants, and none of them is covered in the supplied text, because those chapters are missing from it: opium poppy, safed musali, isabgol, ashwagandha, guggal, mehandi, cheti gulab and mint. This unit does not describe their production, their uses or their compounds, and it does not borrow descriptions from other books, because the task for this unit limits its sources to Dashora and Shanmugavelu.

Table 28.2: The three plants covered, and how their compounds are obtained

PlantCompoundWhere it sitsHow it is obtainedMeasure of quality
VanillaVanillinIn glucosides in the green podEnzyme action during curing, then maceration in alcohol for essenceVanillin 4.15 to 4.40% by the Mexican process
CinchonaQuinine, quinidine, cinchonine, cinchonidineIn bark, combined with quinic acidLime frees the acids, solvent oil extracts, acid water takes the alkaloids back, crystallisation at pH 5.5Quinine percentage of bark; 12% quinine sulphate in clone No. 701
WattleTannins, chiefly catechin and its derivativesIn the barkBark stripped fresh and taken to the tannin extraction factoryTannin 30% or more in dry black wattle bark; 35% in the book's account

A visual sorting game: make twelve cards: vanilla pod, vanillin, curing, quinine, cinchona bark, solvent oil, tannin, wattle bark, catechin, glucoside, lime, and pH 5.5. Shuffle them and sort them into three piles, one for each plant in this unit, matching each compound and each step to its plant.

Practical note: handling scent and medicine with care

Three working rules come out of these three plants. First, do not judge the harvest by the smell: a vanilla pod has no vanillin when it is picked, and a cinchona bark shows nothing of its quinine, so the harvest test is colour, age and stage, not scent. Second, dry and cure exactly as the book directs, because the compound is made or lost after harvest: vanilla needs its four curing stages, and cinchona bark must be dried quickly at about 65°C to 10 percent moisture, or the alkaloids are lost.

Key ideas to carry forward

  • Aromatic plants are grown for scent and flavour, medicinal plants for compounds that act on the body, and the two groups overlap.
  • Vanilla pods hold no vanillin at harvest: enzymes make it from glucosides during curing, and the Mexican process gives 4.15 to 4.40 percent vanillin.
  • Vanilla flowers last one day and must be hand pollinated, 6 to 8 per cluster, by a worker who can do 1,000 to 1,500 flowers a day.
  • Cinchona bark holds more than 25 alkaloids, with quinine the most valued, and the bark is judged on its quinine percentage.
  • Quinine is freed from the bark with lime, taken up in solvent oil at 97.5°C, moved into dilute sulphuric acid, and crystallised as quinine sulphate at pH 5.5.
  • Cinchona trees are coppiced at 8 to 10 years, cut at 5 cm, and finally uprooted in the 30th year, giving 4,000 kg of dry bark per hectare at the coppicings and 6,000 kg at uprooting.
  • Black wattle bark holds about 35 percent tannin, dry bark must hold at least 30 percent, and tannin rises with bark thickness and tree diameter.
  • The books name eight more aromatic and medicinal plants that this unit does not cover, because their chapters are missing from the supplied copies, and this unit says so rather than inventing their production.

Glossary

Aromatic plant

A plant grown for its scent and flavour compounds. Food tech example: vanilla, grown for the vanillin that develops in its cured pods.

Medicinal plant

A plant grown for compounds that act on the body. Food tech example: cinchona, grown for the quinine in its bark.

Alkaloid

A bitter, nitrogen holding compound in a plant, often with a strong effect on the body. Food tech example: quinine, quinidine, cinchonine and cinchonidine in cinchona bark.

Vanillin

The flavour compound of vanilla, made from glucosides in the pod during curing. Food tech example: cured vanilla beans holding 4.15 to 4.40 percent vanillin by the Mexican process.

Glucoside

A compound in which a substance is joined to a sugar, and freed when the sugar link is broken. Food tech example: the glucosides in a green vanilla pod that enzymes turn into vanillin during curing.

Curing

The controlled drying and resting that develops flavour or quality after harvest. Food tech example: vanilla pods killed, warmed, sweated for 7 to 12 days, dried slowly and conditioned for months.

Quinine

The main alkaloid of cinchona bark, used against malaria and in tonic drinks. Food tech example: quinine sulphate crystallised from cinchona bark extract at pH 5.5.

Coppicing

Cutting a tree down to a short stump so it throws up new shoots. Food tech example: cinchona cut at 5 cm from the ground at 8 to 10 years, with 2 to 3 shoots kept.

Tannin

A bitter plant compound that binds protein and is used to tan leather. Food tech example: black wattle bark holding about 35 percent tannin.

Catechin

The main tannin type in wattle bark, a polyphenol of the catechol group. Food tech example: catechin and its derivatives in black wattle bark.

References

  • Dashora, L. K., Dashora, Abhay, and Lakhawat, S. S. Production Technology of Plantation Crops, Spices, Aromatic and Medicinal Plants. 2nd edition, Agrotech Publishing Academy, Udaipur, 2013. ISBN 978-81-8321-275-5. Preface and About the Book, on the importance and scope of aromatic and medicinal plants. The book's third section, on opium poppy, safed musali, isabgol, ashwagandha, guggal, mehandi, cheti gulab and mint, is not present in the copy supplied for this subject.
  • Shanmugavelu, K. G., Kumar, N., and Peter, K. V. Production Technology of Spices and Plantation Crops. ISBN 81-7754-161-7. Chapters on Vanilla, Cinchona (chapter 29) and Wattle (chapter 30), used only where the scanned text is legible.

Unit 29: Green Coffee: Processing, Grading and Blending

Note

Unit 29: Green Coffee: Processing, Grading and Blending

A coffee cherry arrives as a soft red fruit and leaves the curing works as a hard green bean that can travel the world and wait months for the roaster. What happens in between decides whether the cup is clean or spoiled. This unit follows the cherry through drying or wet processing, curing, grading, storage and blending.

Unit 1 followed coffee from plant to market in general terms, and Unit 3 gave a short account of cherry coffee, parchment coffee, hulling, grading and storage. This unit goes deeper: the two processing routes, the machines that pulp, ferment, wash, dry, hull and sort, how a lot is graded and specified, the conditions that keep green coffee sound in store, and how lots are chosen and blended. Unit 5 looks at blending again as standardisation; here blending is about choosing the green coffees that go into the roaster.

29.1 Two routes from the same cherry

Two main methods turn harvested cherries into the intermediate products that curing then treats in the same way. Dry processing gives dried cherry coffee. Wet processing gives dry parchment coffee. Dry processing is simple and asks little of the harvest, because all the cherries are dried at once, straight after picking. Wet processing asks for ripe cherries only, and for clean water and equipment, because unripe cherries and cherries that have partly dried on the tree cannot be handled by pulping machines. Dry processing is generally used for robusta coffee, and also widely for arabica. Wet processing is used for arabica and gives the so-called mild coffee, when fermentation is part of the preparation. Only ripe fruit has developed the full sugars and aroma precursors that roasting will later turn into flavour, so the harvest itself is a quality step.

🤔 Pause and think: if a farmer picks every cherry on the branch in one pass, ripe, unripe and overripe together, what does the dry route gain, and what does the cup lose?

Quick check

Why does wet processing need ripe cherries only, while dry processing can take the whole harvest at once?

Check your answer

Pulping machines cannot handle green berries or berries that have partly dried on the tree, so wet processing needs ripe fruit only. Dry processing dries the whole cherry at once, so it can take ripe, unripe and overripe fruit together, though the cup suffers for it.

Two routes, one green bean

Harvested cherries
Ripe fruit holds the sugars and aroma precursors the roast will need.

↓

Dry route: dry the whole cherry
Sun or machine drying, then curing. Gives dried cherry coffee.
Wet route: pulp, ferment, wash, dry
Most outer coverings come off before drying. Gives dry parchment coffee.

↓

Curing, in both routes
Redry, clean, hull, size grade, density sort, colour sort.

↓

Green coffee
Flat beans and peaberries, graded, stored and blended for the roaster.

Figure 29.1: Dry and wet processing from the same harvest. Both routes end at curing and green coffee. Drawn from Vincent, Coffee, Volume 2: Technology, Ch. 1.

29.2 Dry processing: drying the whole cherry

Dry processing dries the whole fruit at once, straight after harvest. It suits places where labour is short, because there is no need to pick only ripe fruit. The book notes a trend to strip picking, taking ripe, unripe and overripe fruit in one pass. In some cases as much as 80% of the fruit picked this way can be unripe. Coffee picked this way cannot give a fine brew, and it is often described as hard.

Sun drying needs no equipment, but it needs a large area, because the berries dry slowly and their pulp holds sugars and pectins. Drying is done on clean dry ground, on trays, or on a solid concrete surface. The berries must lie in a thin layer only 30 to 40 mm thick, above all at the start, so that fermentation does not start. Any undue heating at this stage can cause discoloration, a brown silverskin, and a poor brew. Frequent raking is needed in the early wet stage to stop moulds growing, and drying areas should sit in well ventilated places. Moulds can grow in the field, for example Colletotrichum coffeanum and Fusarium, or on the drying surface, for example Aspergillus niger, Penicillium species and Rhizopus.

Drying time depends mainly on the weather and on the size, ripeness and moisture of the berries. Floating berries, overripe or dried, hold 20 to 50% moisture, while non-floating berries, green, yellow and red, hold 50 to 70%. Ripe berries are mostly large and dried berries are mostly small: 51% of ripe berries by weight are 12 mm across, while 56% of dried berries are 10 mm. Drying takes 3 to 4 weeks. For robusta berries, with bulk density from 645 down to 440 kg per cubic metre, a layer 30 mm thick carries 19.4 down to 13.2 kg per square metre. Drying brings the moisture to about 12% as is.

Large estates dry by machine. Temperature is the limiting factor in how fast drying can go. Very high temperatures at the start can form so-called stinker beans, and temperature needs even stricter control in drying cherry coffee than in drying parchment coffee. Several authors advise low temperatures, for example 55 °C, as advised by Roelofsen. Vincent's tests confirm that a poor product results, reddish beans with brown silverskin, giving unpleasant to stinking brews, if high temperatures are used, and the same holds even at 60 °C if there is too little airflow. Even at low moisture, above 50 °C, measured as the temperature of the ingoing air, taste falls off. Below 60 °C the drying period grows noticeably longer. Coffee temperature matters more than air temperature, but air temperature is far easier to measure. In the later stages, part of the exit hot air can be recirculated. Hot air can come from wood and dried husks, or from liquid fuel and steam, heated indirectly through a heat exchanger or directly, mixed with combustion gases or smoke.

A static drier is often rustic, with a tray of perforated sheet metal about a metre above an exchanger fired by solid fuel: cheap, but hard to control, with a fire risk, and needing constant watching and hand raking. A static drier with mechanical raking adds a rotary rake on a perforated circular platform with hot air blown up from below. A rotary drum drier holds the coffee in a rotating perforated drum at about 2 to 4 rpm. Types named are the Guardiola drier, with four compartments for small batches at different moisture contents; the Torres drier, much used for cherry coffee; and the Okrassa drier. Vertical driers let coffee fall inside vertical chambers in contact with hot air, including the Moreira drier and the American vertical grain drier, often run batchwise with recirculation of the coffee until dry.

Quick check

Why is a thin layer of 30 to 40 mm so important at the start of sun drying?

Check your answer

A thick layer heats in the middle and starts to ferment, which discolours the beans to a brown silverskin and spoils the brew. A thin layer, raked often, dries evenly and keeps moulds from growing.

29.3 Wet processing: pulp, ferment, wash, dry

Wet processing gives a much higher quality product, and it is always used for mild arabica coffee. It needs equipment for cleaning, classification, pulping, fermentation, washing and drying, a large supply of clean water, and ripe fruit only. Most of the outer coverings of the bean are removed before drying.

The wet processing line

Receiving tank
Floaters off, sand and stones out, leaves and twigs out.

↓

Pulping
Skin and most flesh torn off under running water. Pulp out.

↓

Separation and classification
Light beans floated off; heavy beans kept. Small unpulped berries to the repasser.

↓

Fermentation, then washing
Mucilage hydrolysed, then washed away so microbes cannot grow on it.

↓

Draining, pre-drying, drying
Moisture 60% down to 53%, then to 45% in the sun, then to 11% dry parchment coffee.

Figure 29.2: The wet processing line, from receiving tank to dry parchment coffee. Drawn from Vincent, Coffee, Volume 2: Technology, Ch. 1.

29.3.1 Receiving and pulping

The crop must reach the processing centre as fast as possible: heating of the mass in transit can cause damage that cannot be undone. The crop is unloaded into a receiving tank, 1.5 to 2 cubic metres for a ton of berries, fitted with an overflow weir to remove the floaters, and a syphon of more than 75 mm diameter, which carries the heavy berries to the pulper. The tank separates floating from non-floating berries, takes out sand and stones, and removes leaves and twigs. The floaters, berries dried on the tree and insect-attacked berries, are generally processed by the dry method.

Pulping removes the exocarp, the outer skin, and the major part of the mesocarp, the fleshy part, which is called the pulp. It is a delicate step: if the bean itself is damaged, it is open to microbial attack. Pulpers work by tearing off the skin and flesh under running water. Ordinary pulping does not remove all of the mesocarp: the mucilage stays, clinging to the parchment that surrounds the bean, with the testa. The yield, as dry green coffee, varies with the species: arabica 15 to 19%, robusta 18 to 23%, and arabusta 13 to 16%.

Disc pulpers move the berries on by discs with a bulbed surface, generally 18 in (46 cm) in diameter, on a shaft turning at 120 rpm. For arabica, a space of 2 mm between disc and plate and about 5 mm between disc and pulping bar is a starting position. A typical flow rate is about one metric ton of berries per hour per disc, with 1.0 hp for a single disc and 2.5 hp for four discs. Drum pulpers use a rotating drum of punched copper sheet against an adjustable breastplate. A typical drum is 15 in (38 cm) in diameter and 12 in (30 cm) wide, with a capacity of one metric ton per hour at 120 rpm, needing 1.0 hp, and doubling the width doubles the capacity. A pulper-repasser line holds a pulper, a sieve and a repasser: the pulper bars are set a little less tightly, small berries pass through, the sieve rejects them, and the repasser pulps them. The Raoeng pulper is built to pulp and also to remove the mucilage, with a horizontal perforated rotating drum inside a fixed cylinder and water under pressure fed by a hollow shaft. Only one size was available, the Aquapulpa Major, 3,000 kg of berries per hour at 400 rpm, needing 25 hp. Water use from receiving to pulping is about 40 cubic metres per metric ton of green coffee, or 8 cubic metres per metric ton of berries.

29.3.2 Separation, fermentation, washing and drying

The receiving tank does not sort the berries perfectly, so there is a second chance after pulping, again by flotation. Floating parchment coffee holds very few healthy beans. Draining uses small vertical boards as a spillway in running water: the lightest beans travel furthest. The Aagaard densimetric grader works on the jigging principle: beans sit on a sieve under water, the sieve moves to make rising and falling currents, and with each pulse the beans are lifted and settle again, the heaviest to the bottom and the lightest to the top. Placed between pulper and repasser, it splits the mass into small berries not yet pulped, heavy beans taken away by a scraper conveyor, and light beans.

Fermentation aims to hydrolyse the mucilage, so washing can remove it. If mucilage stays on during drying, there is a risk of further fermentation that harms quality. The pectins are hydrolysed by a pectinase already in the fruit, speeded by microbes such as Saccharomyces. The rate depends on temperature, so fermentation time must be fitted to the weather, and pH control matters, to avoid excess formation of acids such as propionic acid. Fermentation can be dry or under water; mixed fermentation means dry fermentation at first, to acidify the surroundings fast, then soaking in water. Wootton showed that the pulped bean loses about 1% of its dry matter during fermentation, and some water-soluble substances, polyphenols and diterpenes, pass out by exosmosis, cutting brew hardness and bitterness. Fermentation is most often done in concrete tanks, with grids and lift gates that let out water alone, or water and coffee together. In low altitude regions fermentation is very short, for example 16 hours, while at high altitudes it generally takes about 48 hours. Beans damaged during pulping can develop tainted to foul flavours, with a foxy appearance, if the water has been too alkaline or holds too much organic matter.

All mucilage must be washed away, so no microbes can grow on it later. Manual washing is often done in the fermentation vats, but if the mass is too deep, say 0.6 to 0.7 m, the work becomes hard. Wet parchment may also be washed in drains, say 60 m long, moved along by paddles. Washing and gravity grading need some 25 to 30 cubic metres of water per metric ton of green coffee. Mechanised washing uses a batch horizontal washer with a perforated drum at about 90 rpm, a vertical washer with blades at 18 rpm, at first with no water, and the Kivu pump, which moves coffee and water at a volume ratio of 40:60, at 20 to 70 cubic metres per hour, with 7.5 hp giving a lift of 9 m.

In draining and pre-drying, moisture falls from 60% to 53%, in a draining tower or on trays, or the mass can be left in the sun for two or three days, when the moisture can reach 45%. Drying the parchment coffee has a wet stage, from 60% to 30% moisture, and a hygroscopic stage, below 30%. Gibson distinguishes the white stage, when moisture falls to 30%, done slowly in the shade to stop the parchment cover cracking or breaking, and the soft black stage, from 30% to 23% as is, done in sunlight. Artificial drying can be done without danger at the start, and again at the end, when the final moisture of 11% is being reached. Sun drying is the most widely used method, usually on trays, with coverings for the hottest hours. With the coffee spread at 10 to 15 kg per square metre, drying takes 10 to 15 days. Drying on concrete is not advised, because cracks give microbes places to grow. In the hygroscopic stage, the temperature must not go above 55 °C.

🤔 Pause and think: parchment drying has a white stage in the shade and a soft black stage in the sun. Why would the same coffee need shade at 30% moisture and sun at 25%?

Quick check

Why must the white stage of parchment drying be done slowly in the shade?

Check your answer

At the white stage the moisture is falling to 30%, and fast drying then cracks or breaks the parchment cover of the beans. Slow drying in the shade protects the parchment, and the soft black stage that follows, from 30% to 23%, is the stage done in sunlight.

Try it yourself: Put two handfuls of dried beans, one of whole dried cherries and one of parchment coffee, side by side on a tray. Feel the outer covering of each, shake the tray, and listen for the bean rattling inside the dried cherry. The covering you feel is what the huller must take off in section 29.4.

29.4 Curing: from dried cherry and parchment to green coffee

The steps that follow dry or wet processing are together called curing, and they are most often done centrally, away from the growing and processing areas. The steps are redrying, cleaning, hulling, size grading, density sorting and colour sorting.

Curing: the green bean takes shape

Redry and clean
Dry to 11% moisture. Screens, magnet and air take out stones, metal and dust.

↓

Hull and polish
Parchment or husk off the bean. Silverskin off too, if polishing is wanted.

↓

Size grade, then density sort
Screens sort by size. Air and gravity tables sort by density.

↓

Colour sort
By hand, or by electronic eyes that eject the discoloured bean.

Figure 29.3: Curing, from dried cherry or parchment coffee to sorted green coffee. Drawn from Vincent, Coffee, Volume 2: Technology, Ch. 1.

Coffee arriving from farms is not always dry enough. Apart from giving storage stability, a moisture content of 11% lets the husk and parchment come off more easily. Dried cherry coffee may hold pebbles, pieces of metal and other foreign bodies, taken out by a first cleaning, with screens for large and medium impurities, a magnetic separator for metal, and a cleaner-separator that joins sifting with pneumatic dust removal. An air-float separator is an inclined flat-bed unit, with air blown up and across it, sorting particles of the same diameter but different densities.

In wet processed coffee, the hull, the dried parchment layer at once around the beans, must come off. In dry processed coffee, the whole of the dried outer coverings, the husks, must come off. Hulling may or may not take off a final layer closest to the beans, now called the silverskin, from the testa. If silverskin removal is wanted, it may need polishers after hulling. The Africa huller, based on the Engelberg patent, holds a steel screw whose pitch grows towards the outlet, a rotor inside a closed horizontal cylinder with a perforated grid, and a knife fixed to the outer frame; husks are sucked through the grid by a fan. The Smout huller-polisher uses a rotating screw inside a casing with spiral grooves, the spirals turning in opposite directions, with pressure adjustable by a floodgate. The Okrassa huller-polisher has two compartments, the first like the Smout huller and the second built for polishing, with models from 1,000 to 2,000 kg of parchment per hour. The Descascador huller, of Brazilian origin, has a cage of perforated sheet metal in which a blade with knives turns, giving excellent results, with few beans broken and no overheating; similar machines have flow rates from 150 to 9,000 kg per hour.

Typical capacities of Africa hullers for different models

Model No.Parchment (kg per hour)Husks (kg per hour)Power (hp)Rotor speed (rpm)
0100030 to 35
165044010 to 12400
1.54102808 to 10400
23202206 to 8450
51701104550

The blank cells are as the book gives them: no parchment flow rate and no rotor speed are stated for model 0. Smout hullers run from 2,000 kg per hour for the largest model, casing 54 in, 32 hp, 60 rpm, down to 140 kg per hour for model 3, casing 18 in, 3 hp, 120 to 140 rpm.

After the outer coverings are off, it pays to size-grade the green beans. A higher price can be got for the larger beans, peaberries, called caracoli, can be separated from normal flat beans, later density sorting is easier, and the roaster can roast more evenly. Two types of size grader work on the sieving principle: screens set in drums, a reel grader, and a vibrated flat bed. Screen apertures are either round, with diameters in steps of 1/64 in, or oblong for peaberries.

Pneumatic sorting can be done with a rising current of air, or with a gravimetric table, and it depends on density and diameter. Size separation first is wanted, because two spheres of different diameters can have the same settling velocity when their densities differ. The book works an example: a bean of 17/64 in and density 1.20, and a second of 16/64 in and density 1.27, would not separate, because the two settling velocities match. The Catador works with a rising current of air made by a radial blade fan turning at about 550 rpm inside a vertical chamber, and the Duplex model can split normal beans into heavy and medium heavy. Even with size grading first, separation is imperfect: beans are not perfect spheres, air velocity is lower near the walls, and the air flow changes with the resistance of the bean surface.

Discoloured beans can be sorted out by hand, but electronic sorters have made great progress since 1959. Black beans above all need to be taken out. Sorting by hand is very tedious, but about 80 beans can be taken out per minute using both hands, usually on belt conveyors with people on both sides. In electronic sorters, beans pass one by one past electronic eyes that judge colour, which control a mechanical ejector. The sorters use monochromatic or bichromatic light, and the flow rate is of the order of 100 to 150 kg per hour per trough. For monochromatic machines, the defects that can be taken out are black beans, grey or dull beans, and some insect-attacked beans. For bichromatic machines, the list is black, grey, brown, foxy, unripe green, white or marbled beans, some insect-attacked beans, and dried cherries. How far to sort is a matter of judgement: Wilbaux gives the bookkeeping, that for 100 kg of unsorted coffee, with x the percentage of waste taken out, c the cost of sorting per kilogram, P the ex-factory cost of unsorted coffee, and p the added value of the sorted coffee per kilogram, sorting pays when (100 minus x) times (P plus p) is at least 100 times (P plus c), adding the value of the waste itself if it has one.

Quick check

Why must size grading come before density sorting?

Check your answer

Density sorting by air depends on both density and diameter. A small dense bean and a large light bean can settle at the same speed, so they cannot be separated. Size grading first makes the beans in each lot the same size, so density alone decides where they fall.

29.5 Marketed grades: how a lot is described

Green coffee is sold by specification, covering species, processing, origin, crop year, liquoring and flavour, bean size and shape, defective beans, extraneous matter, colour, roasting characteristics, bulk density and overall grade. Not every item will appear in every contract, and some are covered by the words according to sample. At a minimum, country of origin, species, type of green processing and usually the bean size grade will be clearly set out, and that information is stencilled on the bags with the net weight. The systems in use go back to the early system set up by the New York Sugar and Coffee Exchange for green coffees from Brazil and Central and Latin America. Full details for different producing countries are available through the marketing channels, and the book points the reader to Jobin for very full details from all countries.

Proper sampling sits at the centre of specifications, at the port of entry and before export. An ISO standard suggests sampling from at least 10% of the bags, to give at least three samples of 300 g each, though in practice the level of sampling may well be higher. The third sample, fully sealed, can be used for arbitration in cases of dispute. Moisture content is not usually part of a specification, but in practice it will not go above about 13% by weight. The sample for moisture measurement must go into a moisture-proof package at once after sampling. Caffeine content is also not a matter for specification, except for decaffeinated green coffees.

Flavour is the main and most important test of coffee quality, that is, of the brew made after roasting, grinding and infusion with hot water, so-called liquoring or cupping. For the first assessment, the brewing method of steeping is almost universal: some 10 g of roast and ground coffee per 150 ml of boiling water, left for 5 minutes, generally with a medium roast coffee made in a sample roaster and with a fairly fine grind. Washed arabicas, or milds, are expected to have a fine acidity with aromatics, while dry processed robustas are expected to be neutral, though straw-like, with varying degrees of harshness. Brazilian coffees are officially classed by a flavour scale, strictly soft, soft, softish, hard, Rioysh, Rio, which marks the degree of absence of Rio or medicine-like, iodoform, flavour. Overall quality terms are often used, such as usual good quality, UGQ, and fair average quality, FAQ.

Green beans are graded for size by large-scale screening and sold in set size grades, which usually follow laboratory screen analyses, described as large, medium or small, or by letters, A, AA and so on, or by numbers, grades 1, 2, 3. Arabica beans, above all wet processed, tend to be flat, so-called flats, one surface flat with a central cleft, though they may hold peaberries, from a false embryony. Robusta beans tend to be more rounded. Very large arabica beans, called margaropipes or elephants, are also known. Round-hole screens are generally numbered from No. 21 down to No. 10, for hole diameters from 21/64 to 10/64 in, at 1/64 in intervals, with 1/128 in intervals for half sizes. Usually about four screens are used in a nest, shaken by hand. ISO 4150, published in 1980, sets out nominal aperture diameters, in English numbering but stated in millimetres, with allowed tolerances. The results of a screen analysis will generally fall as a good straight line, at least centrally, when aperture size is plotted on a linear probability scale against percentage cumulative amount by weight held at each size. The size at which there is a 50% cumulative amount shows the average size, and the slope shows the size distribution. A typical specification is Ivory Coast Grade 2: all pass through a No. 16 screen, ISO 6.30 mm, with an allowed tolerance of 20% held, all held by a No. 14 screen, ISO 5.60 mm, with 6% passing allowed, and all held by a No. 12 screen, ISO 4.75 mm, with only 1% passing allowed.

Defects is the trade term for defective beans and extraneous matter, taken together or separately, and it matters more for dry processed than for wet processed coffees. The main types are black bean, sour bean, stinker bean, insect-damaged bean, immature or quaker bean, broken bean, bean fragment or brisure, floater bean, shells, husk, twig or stick, and stone, with dried coffee cherry as well. In systems that count physically, types are judged by the number of defects in a sample, 300 g in metric countries, or 36 cubic inches or about 1 lb in the USA. Not all defects count the same. The New York Coffee and Sugar Exchange devised the black-bean-equivalent count: all other defects are judged by how many of them equal one black bean, from 1 to 15. As a guide, a black bean counts as 1, a sour bean as 1 or 1/2 depending on the system, an insect-damaged bean as 1/5 to 1/10 in some systems and 1/2 in others, an immature quaker bean as about 1/5, a broken bean as about 1/5, and a large twig or large stone as 2 to 3 or even 5. The book prints full equivalency tables for several national systems, and notes that the information is continuously updated, so primary sources should be consulted. The black bean mainly comes from a dead bean inside the cherry on the tree, and the sour bean, with the stinker bean, is linked with overfermentation in the wet process.

Type numbers are given by the maximum defects allowed, and the systems differ in detail even among robustas of different origins. As a guide, one system runs from 4 defects for its finest type to 79 for a good type, per sample, and another runs from 15 for Extra Prima to 120 for Limite. There is now a move towards weighing defects, where good balances are available, because it is simpler and faster. With wet processed coffees the number of defects is generally quite low: the New York system allows not more than 13 defects per 1 lb sample for Central and Latin American coffees under the name Extra.

Wet processed arabicas are generally blue-green to grey-green and even, with the blue-green colour prized above all, while dry processed robustas tend to be brownish and uneven. Storage changes colour: dry processed Brazilian coffees turn whitish on long storage. ISO recommends the terms bluish, greenish, whitish, yellowish, brownish. The look after sample roasting, generally to a light or medium colour, is also used: one system gives fine roast for 0 to 1 defects per 100 g of roasted beans, good to fine for 2 to 5, good roast for 8 to 10, and poor for more than 12, including so-called pale beans. Bulk density is easy to measure, by free-flow or vibrational methods, and it matters for the bulk density of roasted coffee and for volume in packs. Crop year is a further factor: differences are expected between new crop and old crop, for example faded beans with a thinner and often woody flavour.

Quick check

A sample holds 5 black beans and 150 insect-damaged beans. Which defect counts more in the black-bean-equivalent system, and why?

Check your answer

The 5 black beans count as 5 black-bean equivalents, while 150 insect-damaged beans at 1/10 each count as 15 equivalents in that example, so the insect damage counts more in total. The book gives the weight example as about 0.24% for the 5 blacks in 2,100 beans, against about 5% by weight for the 150 insect-damaged beans in 3,000 beans.

Try it yourself: Take a kitchen sieve and a colander with clearly different hole sizes, and a handful of dried beans of mixed sizes. Shake a small lot through the coarser holes first, then the finer, and weigh what each screen holds. The share held on each screen is a screen analysis in miniature, and the screen that holds half the weight marks the average size, just as in the book.

29.6 Storage of green coffee: moisture, mould and insects

The moisture content for satisfactory storage should not be over 11% by weight. At that level, mould growth and enzyme activity are minimal. When the relative humidity of the air is over 74%, matching an equilibrium moisture content of about 13% by weight, certain moulds develop: Aspergillus niger, Aspergillus ochraceous and Rhizopus species. A relative humidity of more than 85% is needed for yeasts and bacteria to multiply, matching about 18% by weight in the coffee. Mould growth is likely to start above 13% moisture by weight, matching about 67% equilibrium relative humidity at 25 °C in Stirling's data, which fits the general finding that a water activity of at least 0.65 is needed to start mould growth, and 0.85 for best growth. Green coffee is in a mild state of respiration during storage, giving off some heat. Note: the two chapters use different data for the same 13% moisture figure. Chapter 1 matches about 13% moisture to above 74% relative humidity, while Chapter 2, in Stirling's data, matches 13% moisture to about 67% equilibrium relative humidity at 25 °C. Both figures are reported here as the book gives them.

Among the insects that can spoil green coffee by boring holes are Araecerus fasciculatus, the most dangerous, Lasioderma serricornis, Tribolium castaneum and Carpophilus species. A traditional storage building suits green coffee in sacks, as long as it is heat insulated and perhaps also ventilated. Sacks are stored on pallets placed at least 0.5 m from walls, the building should be fully clean, with smooth walls, and all openings fitted with devices to keep insects out. Coffee can also be stored in bulk, as for buffer stocks in a roastery, and bins should be fitted with anti-breakage devices such as chutes. Vacuum storage in containers has been studied, with promising results, and disposable plastic containers of 25 to 1,000 litres are envisaged, which can be stored in the open air and need no building. Otherwise insects have to be destroyed, and the most widely used insecticide is methyl bromide, generally applied under tarpaulins outside the building, at 15 g per cubic metre.

Both Stirling and Multon found that wet processed arabica coffee should keep near its original quality for six months, provided the coffee had first been dried, in its parchment form, to not more than 11.0% by weight by Stirling, or 13% on a dry basis, equal to 11.5% by weight, by Multon, that the relative humidities are 50 to 70% by Stirling, or 50 to 63% by Multon, and that the air temperature is held at 20 °C or less by Stirling, or 26 °C or less by Multon. Stirling also found that controlled ventilation was better than sealed conditions, above all at higher temperatures, and that parchment coffee showed the greatest stability.

Green coffee is moved and held in bags, usually of jute, though other natural fibres have been used, and also polypropylene. Net contents are generally 60 kg, but 70 kg is common in some producing areas, and some areas bag coffee at 75, 80 and even 90 kg. For statistical purposes, exports and imports are recorded in 60 kg units. Pallets measuring 5 ft by 6 ft by 0.5 ft hold 20 bags in four layers piled in alternate directions, and stacks are usually up to four to five pallet loads high, that is, some 100 bags. Jute bags are preferred to polypropylene for pallet storage, because they do not slip. Silos of steel or galvanised iron are also used for short term storage inside a large roastery, often divided into compartments with a cone bottom. A figure of 40 tons of refrigeration is offered for a warehouse holding some 8,000 bags of green coffee.

Keeping green coffee sound in store: four conditions at once

Condition 1: Moisture at or below 11%
Mould growth and enzyme activity are minimal at this level.
Condition 2: Air 50 to 70% relative humidity, 20 °C or less
Stirling's six month guide for wet processed arabica. Multon allows 50 to 63% and 26 °C or less.
Condition 3: Watch the danger lines
Above 13% moisture, mould is likely. Above 74% relative humidity, moulds grow. Above 85%, yeasts and bacteria multiply.
Condition 4: Clean, ventilated store, sacks on pallets
Pallets at least 0.5 m from walls, smooth walls, openings screened against insects.

Figure 29.4: Storage conditions that keep green coffee sound. Drawn from Vincent, Coffee, Volume 2: Technology, Ch. 1, and Clarke, Coffee, Volume 2: Technology, Ch. 2.

Quick check

Why is parchment coffee more stable in storage than hulled green coffee?

Check your answer

The parchment, and the husk in dried cherry coffee, gives protection against insects and a barrier against moisture transfer. Parchment coffee therefore stands up better to unfavourable temperature and humidity, and hulled coffee with its silverskin kept is also more stable than fully cleaned coffee.

29.7 Pre-treatments and blending

Before blending and roasting, it is often needed to take out the small fractional percentage by weight of extraneous matter in the green coffee delivered in bags: stones, nails, coins, light contaminants, wood splinters, fibrous materials and general dust, some of which may have settled to the bottom of the bags. It is convenient to tip the bags, of the same kind of green coffee in one run, through a hooded grill onto conveyors leading to cleaning, usually by air currents to cyclones, and destoning equipment, including the Gump cleaner, though apparently no longer sold. It is convenient to weigh the total extraneous material collected as a percentage in a consignment, since that figure will show the amount more truly than bag sampling. Since roasted beans are a good deal less dense than green beans, and therefore than stones, destoning is often more conveniently done, or finished, after roasting.

A crop year is taken to start in July and to finish in July of the next year. Inside the crop year, harvesting of the main crop will generally take place at set periods lasting a few months, often followed by subsidiary crops. Green processing may take a few months, so with shipment time a harvested crop may take several months to reach its buyers. The book prints a table of average harvesting and exporting dates for a number of producing areas, and the reader who needs a date for one origin should consult that table in the book.

Choosing green coffee is first a matter of judging what each coffee can add to a blend for consumer taste needs at an acceptable cost. Some roasters also sell single coffees, named by country of origin, for example 100% Colombian coffee. Various combinations of milds may be used, but Brazilian coffees, soft qualities, are generally included because they blend well and cost less, for example in 50:50 blends. Smaller roasters may use margaropipes and other large beans, or peaberries only, and mocha coffees, which are in small supply. While makers try to hold marketed brands at a steady flavour quality, as with tea, blending gives room to change the actual recipe, because particular coffees change in availability through the year, change in flavour quality by crop or storage, and change in relative price. Recipes are generally secret, though arabica contents are often declared, and in some places they are legally required. Similar principles apply in making instant coffee, and robusta coffees are widely used in blends of instant coffee. No special arrangements are normally made for blending, since the roaster itself gives the needed blending action for the different green coffees, as long as they have been correctly weighed in. Sometimes the single green coffees are first roasted separately, and then separate blending equipment is needed before grinding and packing, or before extraction in instant coffee making.

🤔 Pause and think: a roaster loses its supply of one mild coffee in the middle of the crop year. Which two levers does blending give the roaster, and which one is closed if the brand must taste exactly the same?

Quick check

Why does the roaster itself do the blending in most plants?

Check your answer

If the different green coffees are weighed in correctly, in the wanted proportions, the turning of the roaster mixes them fully. Separate blending equipment is only needed when the coffees are roasted separately first.

In one market above all, the indigestibility of normal coffee brews has been a matter of continuing interest, shown in so-called commercially made health or Sanft, meaning smooth, coffees, in which components reported or said to be responsible for stomach upset in some buyers are taken out of the green coffee. Two main processes have been used: the dewaxing of green coffee, and the old Lendrich patented steaming process. Dewaxing, of the outer layers, will occur at the same time as solvent decaffeination, described in Chapter 3 of the book. Less severe washing by solvents is used to take out only the waxy substances, of which the carboxy-5-hydroxytryptamides, C-5-HT, are a component and are used as a measure of how well the unwelcome components have been removed. In untreated coffee, 22% of C-5-HT was broken down on roasting, 70% appeared in the spent grounds, and about 6% in the brew, though all would be held back by a paper filter. Dewaxing took out 64% of the starting C-5-HT, and roasting a further 12%; on brewing, 26% was present in the spent grounds, and only 1% in the brew. The Lendrich process is a low temperature steaming step on the green beans, giving 3-methoxy-4-hydroxystyrene in the steam condensate, formed by the breakdown of ferruloylquinic acid, and a general fall in the chlorogenic acid complex and its phenolic breakdown products. The book presents these findings as reported results, not as proof of a health claim, and this unit does the same.

29.8 Practical note: reading a lot the way a buyer reads it

Take a small lot of green coffee, if you can get one, and read it in the order a buyer would. First look at the bag stencil: species, processing, origin, size grade and net weight. Then take a sample, spread it on a white tray, and count what you see: whole flat beans, peaberries, broken beans, black beans, insect-damaged beans and stones. Then shake a handful through two sieves of different hole sizes and note what each holds. Then smell the dry beans, and note the colour: blue-green and even, or brownish and uneven.

Try it yourself: Make a defect tray. On a sheet of white paper, lay out ten beans in a row, and label each one from section 29.5: black, sour, insect-damaged, immature, broken, floater, shell, husk, twig and stone. Ask a friend to mix in ten sound beans and pick out the defects without looking at the labels. The beans you miss are the ones that would slip through in a real lot.

29.9 Check yourself: quick self-check

Think about it first: a curing works takes in dried cherry coffee at 14% moisture and parchment coffee at 13% moisture. Both lots look dry. What must happen before either lot is hulled or stored, and why?

  1. What are the two main processing methods for coffee cherries, and what intermediate product does each give?
Check your answer

Dry processing gives dried cherry coffee, and wet processing gives dry parchment coffee. Both then go through the same curing steps.

  1. Why must sun-dried cherries lie in a layer only 30 to 40 mm thick at the start?
Check your answer

A thicker layer heats and starts to ferment, giving discoloured beans with brown silverskin and a poor brew, and moulds grow in the wet stage without frequent raking.

  1. What does fermentation do in wet processing, and how long does it take?
Check your answer

Fermentation hydrolyses the mucilage, by pectinase in the fruit and by microbes such as Saccharomyces, so washing can remove it. It takes about 16 hours at low altitude and about 48 hours at high altitude, and the pulped bean loses about 1% of its dry matter.

  1. Name the curing steps in order, and state the moisture content wanted for storage.
Check your answer

Redrying, cleaning, hulling, size grading, density sorting and colour sorting. Moisture for satisfactory storage should not be over 11% by weight.

  1. What is a black-bean-equivalent count?
Check your answer

All other defects are judged by how many of them equal one black bean, from 1 to 15. A black bean counts as 1, an immature quaker bean as about 1/5, and a large twig or stone as 2 to 3 or more.

  1. At what moisture and humidity does mould growth become likely in stored green coffee?
Check your answer

Mould growth is likely to start above 13% moisture by weight, matching about 67% equilibrium relative humidity at 25 °C. Above 74% relative humidity moulds develop, and above 85% yeasts and bacteria multiply.

Key ideas to carry forward

  • Dry processing dries the whole cherry at once and gives dried cherry coffee; wet processing pulps, ferments, washes and dries, and gives dry parchment coffee. Both end in the same curing steps.
  • Sun drying needs a layer 30 to 40 mm thick, frequent raking and moving air, and takes 3 to 4 weeks to about 12% moisture. Artificial drying must stay low in temperature, about 55 °C as advised, because high temperatures make stinker beans.
  • A disc pulper runs at 120 rpm with discs 18 in across, and a drum pulper of 15 in by 12 in gives about one metric ton per hour. The Raoeng Aquapulpa Major gives 3,000 kg per hour and also takes off mucilage.
  • Fermentation hydrolyses mucilage, takes about 16 hours at low altitude and 48 hours at high altitude, and costs the bean about 1% of its dry matter. Washing must take all mucilage away.
  • Parchment drying runs from 60% moisture through a white stage to 30% in the shade, a soft black stage from 30% to 23% in the sun, and a final moisture of 11%, with the hygroscopic stage kept at or below 55 °C.
  • Curing is redrying to 11%, cleaning, hulling, size grading, density sorting and colour sorting. Electronic sorters pass 100 to 150 kg per hour per trough, and hand sorting takes out about 80 beans per minute with both hands.
  • A specification covers species, processing, origin, crop year, flavour, size, defects, colour, roasting character, bulk density and overall grade. Sampling takes at least 10% of bags, for at least three 300 g samples.
  • Defects are counted in black-bean equivalents. Screens run from No. 21 to No. 10, 21/64 to 10/64 in. Wet arabicas are blue-green and even; dry robustas are brownish and uneven.
  • Green coffee keeps for about six months near its original quality when dried to 11% or less, held at 50 to 70% relative humidity and 20 °C or less. Mould is likely above 13% moisture.
  • Blending holds a steady cup when crops, prices and availability change. The roaster itself mixes the blend, if the green coffees are weighed in right; separately roasted coffees need separate blending equipment.

Glossary

Green coffee

The dried, hulled coffee bean before roasting, ready for grading, storage and blending. Food tech example: green coffee at 11% moisture stored in sacks on pallets.

Dry processing

Drying the whole coffee cherry at once, in the sun or by machine, to give dried cherry coffee. Food tech example: robusta cherries sun dried in a layer 30 to 40 mm thick.

Wet processing

Pulping the cherry, fermenting and washing away the mucilage, then drying the bean in its parchment to give dry parchment coffee. Food tech example: mild arabica pulped, fermented for about 48 hours at high altitude, washed and dried.

Parchment coffee

Coffee dried in its parchment skin after wet processing, before hulling. Food tech example: parchment coffee dried to 11% moisture before curing.

Dried cherry coffee

Coffee dried as the whole cherry by dry processing, before hulling. Food tech example: dried cherry coffee at about 12% moisture after 3 to 4 weeks of sun drying.

Pulping

Tearing the skin and flesh off the coffee cherry under running water, leaving the bean in its parchment with mucilage clinging to it. Food tech example: a disc pulper at 120 rpm with discs 18 in across.

Mucilage

The sticky layer around the coffee bean inside the cherry, removed by fermentation and washing. Food tech example: mucilage hydrolysed in fermentation and washed away before drying.

Fermentation

Holding pulped coffee so pectinase and microbes hydrolyse the mucilage, making it washable. Food tech example: fermentation for about 16 hours at low altitude and about 48 hours at high altitude.

Curing

The central preparation of dried cherry or parchment coffee, redrying, cleaning, hulling, grading and sorting, to give green coffee for use and export. Food tech example: curing works hulling parchment coffee and size grading the green beans.

Hulling

Removing the parchment hull from wet processed coffee, or the husk from dry processed coffee. Food tech example: an Africa huller taking the husk off dried cherry coffee.

Silverskin

The thin final layer closest to the bean, from the testa, taken off by polishing if wanted. Food tech example: silverskin left after hulling, removed by a polisher.

Peaberry

A single round bean from a cherry that held one bean instead of two, from a false embryony. Food tech example: peaberries, called caracoli, separated from flat beans by size grading.

Flat bean

A normal coffee bean with one flat surface and a central cleft. Food tech example: wet processed arabica beans, mostly flat beans with some peaberries.

Black-bean equivalent

A count in which every defect is valued by how many of it equal one black bean. Food tech example: an immature quaker bean counting as about 1/5 of a black bean.

Liquoring

Tasting the brew, or cupping, to judge the flavour quality of a coffee after roasting and grinding. Food tech example: liquoring with 10 g of roast and ground coffee per 150 ml of boiling water for 5 minutes.

Equilibrium relative humidity

The relative humidity of air in balance with the moisture content of the coffee at a given temperature. Food tech example: 12% moisture in green coffee matching about 60 to 77% equilibrium relative humidity, depending on the coffee and the method used.

Water activity

A measure of free water available for microbes to grow, with mould growth starting at about 0.65. Food tech example: green coffee at 13% moisture, matching a water activity near the mould growth limit.

Destoning

Taking stones and other heavy foreign matter out of green or roasted coffee. Food tech example: destoning after roasting, when the beans are lighter than the stones.

Dewaxing

Washing the outer waxy layers off green coffee, used for health coffees. Food tech example: dewaxing taking out 64% of the C-5-HT in green coffee.

Blending

Mixing different green coffees in set proportions to hold a steady flavour and cost. Food tech example: a 50:50 blend of mild coffees with soft Brazilian coffees.

References

  • Vincent, J.-C. Green Coffee Processing. In Clarke, R. J., and Macrae, R. (eds.), Coffee, Volume 2: Technology. Elsevier Applied Science, 1987. Chapter 1.
  • Clarke, R. J. Grading, Storage, Pre-treatments and Blending. In Clarke, R. J., and Macrae, R. (eds.), Coffee, Volume 2: Technology. Elsevier Applied Science, 1987. Chapter 2.

Unit 30: Roasting and Grinding Coffee

Note

Unit 30: Roasting and Grinding Coffee

Green coffee has very little aroma. The characteristic flavour and headspace aroma of coffee develop only through roasting, and grinding is what lets hot water reach the soluble solids and flavour volatiles inside the bean. This unit follows the bean through the roaster and the grinder, with the heat, the mass loss, the gas, the swelling, the colour, the machines and the grind sizes the book gives.

Unit 3 gave a short account of roasting and grinding in section 3.6, and Unit 2 touched on roasting in section 2.7. This unit gives the full treatment from Chapter 4 of the book: the mechanisms and methods of roasting, the chemical and physical changes, the heat factors, how roast degree is measured, emission control, the roaster types, and the mechanism, size analysis, bulk density and equipment of grinding. Extraction, which follows grinding, is a later unit.

30.1 Roasting: a time and temperature process

Roasting is a time-temperature-dependent process, in which chemical changes are induced in the green coffee beans, and marked physical changes in the structure of the coffee are also evident. There is a loss of dry matter, primarily as gaseous carbon dioxide and water, over and above the moisture already present, and other volatile products of pyrolysis. Roasting is normally carried out under atmospheric conditions with hot combustion gases and excess air as the primary heating agent, though heat may also be provided by contact with hot metal surfaces, solely in more primitive methods, but more generally as a supplement to convection from the hot gases. Some heat may also be supplied by radiation, and in some designs even as the primary source by radiative heating of various kinds.

The roasted whole beans are characterised by the roasting process to which they have been subjected. Their so-called degree of roast is reflected in their external colour, the flavour developed, the dry mass loss that has occurred, and the chemical changes in selected components. In simple terms, roast coffee can be described as having a light, medium or dark roast. Coffees may also be described as fast roasted, that is, a time of roasting of a few minutes or even less, or conventionally roasted, a time in the order of 12 to 15 minutes, with many in an intermediate time of 5 to 8 minutes. Degree of roast plays a major part in determining the flavour characteristics of extracts subsequently brewed from roasted coffee, whatever the blend, and speed of roasting is associated with the production of so-called high-yield coffees.

The horizontal rotating drum has been the most usual method of providing contact of hot air in a tumbling coffee bed, but roasters have been designed and are available based on other mechanical principles. Horizontal rotating drums may provide either batch or continuous roasting, whereas the vertical drum and bowl are usually batch. Fluidised-bed roasters generally operate in the batch mode, but the latest designs are often continuous. All these roasters must provide a cooling facility to bring the roasted coffee at once after the desired level of roast to ambient temperature, usually by contact with cold air, a process often called quenching. Quenching is a term also applied when water is added by spraying within the roasting section for a similar purpose, but also to provide benefits in grinding, covered in section 30.5.

A further important feature is the use either of once-through flow of hot gases or of recirculated hot gases, since the latter can have considerable advantages in the rate of heat transfer and overall economy. Both methods are tied to the need to reduce the levels of organic volatile substances in the exit gases from the roaster to the stack and finally to the atmosphere, which may otherwise be regarded as pollutants subject to legal constraints, covered in section 30.3.3.

🤔 Pause and think: a roaster can reach the same roast colour in 4 minutes or in 14 minutes. The book says speed is tied to high-yield coffees. What else, besides colour, would you expect to differ between a fast roast and a slow roast?

Quick check

What does degree of roast show in, and what is the difference between fast and conventional roasting times?

Check your answer

Degree of roast shows in external colour, flavour developed, dry mass loss, and chemical changes in selected components. Fast roasting takes a few minutes or even less, conventional roasting takes 12 to 15 minutes, and many roasts take an intermediate 5 to 8 minutes.

Roasting: what goes in, what comes out

Green coffee blend, weighed in
Moisture about 12%, dry matter, sucrose and other precursors.

↓

Hot gases, excess air, tumbling bed
Convection first, conduction as a supplement, some radiation. Time and temperature set the roast.

↓

Roasted whole beans, cooled at once
Colour, flavour, dry mass loss, carbon dioxide held inside, chaff released.

↓

Exit gases to stack
Water, carbon dioxide, organic volatiles and chaff, cleaned before release.

Figure 30.1: Roasting, from green coffee to roasted beans and exit gases. Drawn from Clarke, Coffee, Volume 2: Technology, Ch. 4.

30.2 Chemical changes in roasting

The chemical changes for each individual compound in green coffee, giving the characteristic flavour and aroma of roasted coffee and its brews, are discussed in detail in Volume 1 of the book. In processing, two compositional factors matter above all: dry matter roast loss, and carbon dioxide evolution and retention.

30.2.1 Dry matter roast loss

The dry matter loss observed with different degrees of roast is given below. In practice, accurate measurement can be difficult, because it needs true measurements of the moisture content of both the green and the roasted coffee, though arbitrary values can be used. The roasted whole bean may also contain small but significant weights of entrapped carbon dioxide, which is subsequently released. The total, or as is, mass loss will include the moisture in the original green coffee and is unfortunately the figure very often unhelpfully given in many roasting studies. The moisture content is a known variable, generally around 12%, but it may well fall to about 8% in a dry, warm laboratory.

Approximate percentage dry mass loss for different degrees of roast, on a dry basis of green coffee weight

Degree of roastPercentage dry mass loss
Light1 to 5
Medium5 to 8
Dark8 to 12
Very darkMore than 12

The composition of the gases released from roast and ground coffee, and therefore presumably also from roast whole beans, has been reported as 87% carbon dioxide, 7.3% carbon monoxide and 5.3% nitrogen by volume. For a large continuous roaster, carbon dioxide makes up about 8% by volume of the actual gases leaving the roaster, with nitrogen from combusted and excess air, and oxygen from the excess air used in combustion, forming the remainder. It is not clear what share of any particular volatile compound generated is either lost to the roaster gases or retained within the roasted beans, though it is believed that a major share of the total volatile substances generated are retained. The roaster gases clearly contain a number of acidic substances in an amount and type not necessarily wanted in the roasted coffee.

30.2.2 Carbon dioxide: made, held, then released

Roasted whole beans contain entrapped carbon dioxide, retained from that generated, together with evolved water, in the roasting process by pyrolysis. The amount depends on the degree of roast and on the blend, and also on the method of determination. At once after roasting, whole beans contain a quantity of the order of 2 to 5 ml of carbon dioxide, measured at NTP, 20 °C and 760 mmHg pressure, and probably more, per gram of roast coffee. This carbon dioxide is subsequently slowly released, so that the total amount can be assessed by collection methods over substantial periods of time, for example some 2,500 hours, though more rapidly if ground, for example some 360 hours or less according to the degree of grinding. Barbara has described a more rapid method of collection and reported preliminary figures as high as 10 ml of carbon dioxide per gram of roast coffee. A quantity of 2 to 5 ml per gram corresponds to only some 0.4 to 1% of roasted coffee weight, and slightly less of green, which, though significant in many respects, appears to be a relatively small share of the dry mass lost on roasting. This carbon dioxide must be held under considerable pressure within a roasted bean, which can be calculated to be several atmospheres, for example 6.4 atm in the book's worked example, from 5 ml of carbon dioxide in a bean volume of 0.780 ml per gram.

About half of the total carbon dioxide generated is retained in the roast whole bean. For a large-scale continuous roaster, taking 3,000 lb (1,361 kg) per hour of green coffee and producing 2,450 lb (1,111 kg) roasted coffee at once after roasting, the 18.4% total mass loss, on as is green coffee weight, is made up as follows: water, 16.7%, made up of say 12% initial moisture, with say 4.7% generated moisture from dry matter; and carbon dioxide and other gases, 1.7% loss during roasting. The total dry matter loss on original green coffee weight is therefore 4.7 plus 1.7, or 6.4%, matching 7.2% on a dry green coffee weight basis. A further 1.7% of carbon dioxide is assumed retained in the roasted coffee, so that the total roasting loss, when finally assessed on 2,400 lb per hour of roasted and ground coffee, with virtually total release of carbon dioxide, will be about 20% on an as is green coffee weight. The total possible dry-matter loss on a dry green coffee basis will therefore be as much as 9.1%. The amount of at once entrapped carbon dioxide on dry roasted coffee weight is 1.7 divided by 0.816, or 2.1% by weight, which matches 10 ml of carbon dioxide at NTP per gram of coffee, consistent with the data of Barbara.

A further small contribution to mass loss is the chaff released from either the centre fold of the bean or its surface. Dry processed green robusta beans often contain tenaciously held silverskin, which becomes so-called chaff after roasting. Wet processed arabicas carry much less silverskin and may indeed be polished. Few reliable data are available for actual quantities.

The origin of the evolved carbon dioxide and water is not clear. Exothermic heat is generated during roasting, covered in section 30.3, but the reaction involved is not understood, though it most certainly involves anhydro formation and other degradative changes in the saccharides present, above all the sucrose. Arabica coffees contain a higher percentage of sucrose than robusta, and this is largely converted on roasting into caramelised substances. Although the exothermic heat is thought to be lower for robusta, the carbon dioxide content of roasted robusta is in fact believed to be higher than for arabica, at similar roast degree.

30.2.3 Soluble solids, and what too much heat does

The percentage soluble solids contents of roasted coffees are an often quoted parameter, though a distinction should be drawn between the value arrived at by exhaustive Soxhlet extraction with water at 100 °C and that obtained under conditions of some type of household brewing or by an open-pot method, in which, say, 1 litre of boiling water is poured over 53 g ground coffee and stirred for 5 minutes. Results will differ also according to the degree of grind. Reliable data are scarce, though Kroplien provides considerable data using the same extracting method, 2 g ground coffee simmered with 200 ml water at 100 °C for 5 minutes, for some 36 arabica coffees, calculated average 31.2%, and seven robusta coffees, calculated average 30.3%, though of somewhat differing roast colours. In general, robusta coffees provide some 3 to 4% more soluble solids than arabica for the same roast colour, and the percentage is also greater for darker roasts of the same coffee, though not necessarily when expressed on an original green basis. Fast roasted coffees are believed to provide higher percentages of soluble solids than those conventionally roasted, though not necessarily in exhaustive extraction, reflecting a greater availability to extraction of these solids by opening up the bean structure. Maier has provided data for both a robusta and an arabica coffee, showing an increasing yield of water-soluble solids determined by an open-pot method with decreasing roasting time, though there are also differences in percentage roast loss and reflectance colour in the coffees.

Excessive temperature in roasting, above all where the heat transfer is from very hot metal surfaces, is known to cause unwelcome chemical changes, above all at the bean surfaces, where the composition is different from that of the body of the bean. Charring and carbonisation affect the bean surfaces, though they may also extend to the whole of the bean. Decreasing air temperatures and a decreased share of conductive heat is a feature of recirculatory-type roasters. Sivetz and Desrosier discuss the formation of tars depositing on beans, contributing a harsh taste, and on the roaster cylinder and other surfaces. Such depositions will occur in once-through roasters, though fluidised-bed roasters, once-through, may well provide cleaner-surfaced roasted beans.

Quick check

A dark roast loses 8 to 12% of its dry mass. Why is the as is weight loss at the roaster higher than that?

Check your answer

The as is loss also includes the moisture in the original green coffee, generally around 12%, plus the carbon dioxide retained in the beans and released later. In the book's large roaster example, the total loss at the roaster is 18.4% as is, made of 16.7% water and 1.7% carbon dioxide and other gases.

30.3 Heat factors, physical changes and roast degree

After the removal of the original moisture present in the green coffee beans, roasting proper starts at a green bean temperature of about 200 °C, after which, through exothermic reactions, escalation of effect readily occurs, requiring considerable control of the process for a given degree of roast. Reaction in green arabica coffee appears to start at a temperature as low as 160 °C, but not until 210 °C is reached do these reactions peak, and they fall off at 250 °C, in the thermal analysis work of Raemy and Lambelet.

The specific heat of green coffee varies with its moisture content and with temperature. A dry green arabica coffee from Mexico has a specific heat of 0.395 and a robusta coffee from Togo a lower figure of 0.32, both at 30 °C, assuming a specific heat for water of 1.0. At normal moisture contents of around 12%, these figures will therefore be higher, for example 0.47 for arabica. Between 20 °C and 95 °C, the specific heat value will increase by about 30%.

Heat requirements have been calculated from experimental data, taking into account the exothermic heat generated. Raemy and Lambelet found a measured enthalpy or heat content for a Mexican arabica coffee, 7.5% moisture on green, of between 60 and 100 kcal per kilogram of roasted coffee at roasting. Vincent and colleagues found an energy of roasting of 87 to 98 kcal per kilogram of green coffee, in a very small-scale fluidised roaster with once-through flow of hot gases, at a constant inlet temperature of 260 °C, after deducting a thermal loss of 119 to 126 kcal per kilogram of green coffee. Sivetz and Desrosier calculated 532 Btu per pound of green coffee, about 12% moisture, or 295 kcal per kilogram, for the net heat input requirement in a large continuous drum roaster, 3,000 lb per hour, to a total 18.3% weight loss at the roaster, with a roasting bean temperature of 196 °C, initial hot air at 249 °C, and a datum temperature of 27 °C. In that calculation the exothermic heat generated provides some 11% of the total gross heat requirement, or 35.5 kcal per kilogram of green coffee, equal to 40.3 on dry green and 43 kcal per kilogram on roasted coffee. Surface heat losses are about 19% of the net heat requirement, giving some 81% heat utilisation for the roasting process itself, and the heat loss from the exit gases is some 25% of the net heat input requirement. In practice, net heat requirements are best assessed from fuel-oil consumption data from manufacturers, for example 320 kcal per kilogram of green coffee quoted for a Gothot Rapido-Nova roaster, with or without recirculation. Heat is also needed to burn or remove organic pollutants in the exhaust gases, in the same order as the amount needed for roasting itself: Gothot quote 370 kcal per kilogram of green coffee for combustion in an external afterburner at 450 °C, and higher for higher combustion temperatures. Fans and mechanical arrangements also need electrical energy, for example typically 0.0217 kVA per kilogram of green coffee, equal to 18.7 kcal per kilogram, assuming a power factor of 1.

The fluidised roaster data of Vincent and colleagues indicate gas to coffee ratios of 7.3 to 12.8 to give roasting times of 125 to 192 seconds for a light roast, with an inlet temperature of 250 °C. To cause fluidisation, certain minimum air velocities must be reached: an actual air flow of 45 to 44 cubic metres per hour measured at 250 °C for green coffee, or 39 to 37 for roasted coffee, over a surface area with a diameter of 100 mm, giving a fluidising velocity of 159 to 131 cm per second for a small bed. Sivetz and Desrosier calculate a value of 1,500 cm per second at 204 °C, a true velocity of gas between beans, equal to a superficial velocity of about 750 cm per second. In once-through drum roasters, air to coffee weight ratios are of the order of 3 to 1 to give roasting times of 10 to 12 minutes with relatively high inlet temperatures. The ratio of conductive to convective heat applied has been estimated to be about 1 to 3 to 2 to 3 in a once-through system, compared with 1 to 4 to 3 to 4 in a recirculating system.

Heat transfer rates decide the time taken to achieve a given degree of roast: Q divided by t equals U times A times the mean temperature difference, where Q is the heat required in time t, excluding exothermic heat, U is the overall heat transfer coefficient, A is the surface area of the beans, and the mean temperature difference is between the heating air and the roasting coffee. An increase in the hot gas to coffee weight ratio generally produces a marked decrease in roasting time, or enables a lowering of inlet gas temperatures, as in a change from once-through flow to recirculation. The primary hot air to coffee ratio is only some 0.5 to 1, while the recirculating gas to coffee ratio is of the order of 11 to give a roasting time of some 5 to 6 minutes for a gas temperature of about 250 °C but an actual roasting temperature of 196 °C. Sivetz and Desrosier suggest U is about 0.5 Btu per hour per square foot of bean surface per degree Fahrenheit, or 1.24 cal per hour per square centimetre per °C. Fluidised roaster data indicate gas to coffee ratios of 7.3 to 12.8 to give roasting times of 125 to 192 seconds for a light roast, with an inlet temperature of 250 °C, and a fluidising velocity of 159 to 131 cm per second for a small bed. In once-through drum roasters, air to coffee weight ratios are of the order of 3 to 1 to give roasting times of 10 to 12 minutes.

Swelling of beans progressively occurs, including a popping phase, leading to considerably decreased density, as a function of the degree of roast, but also of the speed of roasting. Oil exudes to the surface with increasing severity of roast. Uniformity of roast colour at any given degree of roast is generally wanted, though this may be frustrated by inherent differences in individual beans, above all of the dry processed type, for example immature beans giving rise to pales in the roast. Charred or carbonised beans at local hot spots are generally to be avoided. Green coffee bean densities are of the order of 1.25 g per ml, or 0.80 ml per g, though high-grown arabica are generally higher, 1.30, or 0.77 ml per g. For medium roast arabicas, bean densities are given as 1.428 ml per g, or 0.70 g per ml, with a dry substance density of 0.755 ml per g, or 1.32 g per ml, matching a porosity, volume of void space over volume of bean, of 0.47. The degree of swelling on roasting is of the order of 40 to 60%, that is, percentage of original density, though robustas swell less than arabicas for the same degree of roast, and are thus said to roast more slowly. A light roast coffee may have a free-flow bulk density of 23 lb per cubic foot, 369 kg per cubic metre, but only 18 lb per cubic foot, 289 kg per cubic metre, when darkly roasted. Fluidised-bed roasted coffees had a bulk density of 320 to 350 kg per cubic metre for a Costa Rica arabica, and 270 to 330 for a Cameroon robusta, while in the green form both had a value of 700, though the method of measurement is not given. The cell walls are believed to remain intact, but there is a softening of the structure, reflected in the variable compressibility of grounds after water extraction.

Degree of roast is qualitatively assessed from external colour. In large-scale practice, a final quantitative assessment is made by colour reflectance readings by a suitable meter, after grinding finely to a specified size fraction and compressing the sample at about 1,000 psig, or 70.3 kg per square centimetre, to give a consistent surface. The meters available, notably the Photovolt, Gardner, Agtron, EEL and Hunterlab meters, rely on photoelectric cells. The Gardner meter measures reflected light in three parameters, percentage reflectance and deviation from grey towards redness or blueness, by tristimulus filters, green, amber and blue, onto three photocells, standardised by an opaque white tile. The Hunterlab meter also gives three parameters, L, a and b, while the Photovolt meter uses one colour to a single photocell, giving a single reading referenced to coloured tile surfaces. Other methods are accurate measurement of dry matter mass loss, and laboratory determination of trigonelline and chlorogenic acids, which are destroyed according to the degree of roast: total chlorogenic acids fall to about 40% of their original value for a medium roast on an equivalent dry green basis.

Emission control matters because exhaust gases from large roaster installations are regarded as a public nuisance and in some countries as pollutants subject to legal restraint. The afterburner recombusts these gases at high temperature in a separate furnace in the exhaust stack. During a conventional 6 minute batch roast in a Gothot Rapido roaster, without cleaning except chaff removal, total organic carbon rose to a maximum of some 1,200 mg of carbon per cubic metre of gas at NTP, with an average of 158. The German Federal Republic law of 1968 stipulated a maximum of 300 mg per cubic metre. With gas recirculation, at a furnace temperature of 600 °C the average was down to 45, maximum 220, and at 700 °C the average was 15, maximum 60. With a conventional external afterburner at 450 °C, the average was 54, and 350 °C was unsatisfactory. Catalytic converters can reduce carbon levels at about 350 °C. German Federal law also stipulates a limit of 150 mg per cubic metre for dust, specifically chaff, solved by efficient cyclones before afterburners or recirculation furnaces.

Quick check

Why does a reflectance meter need a ground, compressed sample, rather than whole beans straight from the roaster?

Check your answer

Whole beans vary in colour from bean to bean, and the surface is curved. Grinding finely to a specified size fraction and compressing the sample at about 1,000 psig gives a flat, even surface, so the reflectance reading is consistent.

30.4 Roasting equipment

Roasting plant is supplied by various companies specialising in its construction. Horizontal drum roasters are perhaps the oldest and most well established type, above all batch operated, made by Probat-Werke, Jabez Burns, and G. W. Barth. There has been a gradual evolution from 19th century designs, with a decreasing share of conductive heat from hot metallic surfaces, improved furnaces and their siting, and lower hot gas temperatures, through to recirculatory systems.

Batch drum roasters. The Probat, Burns Thermalo and Barth Tornado roasters are characteristic of this type. The Probat has a double-walled solid drum, so hot gases from the furnace enter through the back end, through the tumbling beans and away to the stack. The four-bag roaster, 240 kg batch, is typical of the Probat GO-series, with a roasting time of some 10 to 12 minutes and a turn-round time of 15 to 18 minutes. Heating is programmed in two stages, from a controller monitoring the bean temperatures: a high rate of heating at first, followed by a lower rate until exothermic reaction occurs, and at a set roasting temperature the furnace is cut off and the beans discharged into a tray with upflow of cold air. Inlet gas temperatures will be of the order of 450 °C or less. The GO-series has been replaced by the R and RR series, which recirculate the exit roaster gases via a cyclone, for chaff removal, back to the furnace, with excess returned to the stack after cleaning through a catalytic converter or afterburner. The Burns Thermalo differs in that the drum is perforated, letting hot gases pass through the tumbling beans and through the perforations back to the furnace after chaff removal.

Continuous drum roasters. Burns was the first company to offer a successful continuous roaster, in 1940, again using a perforated drum for both the roasting and cooling sections. The cylinder of a 3,000 lb per hour roaster rotates at speeds from 2 to 7 rpm, according to the residence time wanted, which at the highest speed is some 7 minutes, of which some 5 minutes will be in the roasting section. The beans are compartmented within the cylinder by a spiral, and a central disc separates the heating part from the cooling part. This roaster has had a built-in hot gas recirculation system from early on, and with the high gas to coffee ratios the gas temperature need only be of the order of 250 °C. Probat introduced a continuous roaster, type RC, in about 1980, again with recirculation and chaff removal, with hot gases distributed over the tumbling beans by a nozzle distributor running along the horizontal axis of the drum, and the cooling section sited underneath the roasting section. Roasting times are reported to be between 3 and 6 minutes, again as a consequence of high velocities of the hot gases.

Other roaster types. The Gothot Rapido-Nova series uses a vertical fixed drum with paddles: a set of paddles to increase the rate of heat transfer between beans and between bean and gas, rotating on a horizontal axis, with exhaust gases taken through a fan at the top of the vertical cylindrical vessel to cleaning devices, and discharge at the end of the roast at once into an integral vessel for cooling. With a relatively high gas to coffee ratio, the roasting time is of the order of 6 minutes. The series also has roasters with in-built recirculation, with the furnace operated at 600 to 700 °C, so that the exhaust gases can be almost completely cleaned to a low residual organic level, and a variant with a catalytic converter. The Probat RZ, Radial-Turbo, series uses a rotating bowl: beans fed into the centre of a rotating horizontal bowl with a vertical shaft are carried to the periphery through centrifugal force assisted by high-temperature air, and on reaching a fixed multi-plate ring, they fall back to the centre in spiral-shaped circuits surrounded by hot air. Short roasting times, about 4 to 6 minutes, are employed. The first commercial fluidised-bed roaster of reasonable size is generally considered to be the Lurgi Aerotherm roaster, introduced in 1957, with a largest batch size of 110 lb, 50 kg, and a roasting time from 2 to 4 minutes from once-through hot air indirectly heated, with an air to coffee weight ratio of about 10. The Wolverine Jet Zone roaster is a further fluidised-bed type, continuous in operation, with jets of hot air directed into the beans fed onto an oscillating conveyor. Pressure roasting was developed from the Alcalor batch pressure roasters to the Smitherm roaster, continuously operated using nitrogen gas pressure, in a chamber designed for 150 psig pressure and 440 °F, 227 °C, with a recirculating gas system. The roasting time is given as 5.33 minutes, and the ratio of hot gases to coffee is reported to be 3,500 cubic feet per minute, at 80 psig, to 6 lb coffee per minute, suggesting a weight ratio of 70 to 1. Compared with conventional atmospheric roasting, for the same reflectance roast colour, pressure roasting gave roasted coffee which on brewing had a higher acidity, somewhat higher yield of soluble solids and somewhat higher bulk density.

A complete roaster installation is more than the roasters themselves. For batch roasters, batching must be carried out with green coffees drawn from the silos, with either weighing or volumetric input from a control panel. The cooled roasted coffee has to be conveyed to storage vessels or silos awaiting grinding, usually via a destoner, and it is useful at this stage also to be able to weigh the roasted coffee. The need for afterburners, if not in-built, and chaff removal devices has already been mentioned.

Roaster types at a glance: five alternative machines, a roastery uses one of them

Type 1: Horizontal drum, batch or continuous
Probat, Burns Thermalo, Barth. Tumbling bed, hot gases through the beans. Batch 10 to 12 min, continuous 3 to 7 min.
Type 2: Vertical fixed drum with paddles
Gothot Rapido-Nova. Paddles lift heat transfer. About 6 min, furnace 600 to 700 °C in the recirculating form.
Type 3: Rotating bowl
Probat RZ. Centrifugal force carries beans to the rim and back in spirals. About 4 to 6 min.
Type 4: Fluidised bed
Lurgi Aerotherm 2 to 4 min, 50 kg batches. Wolverine Jet Zone, continuous.
Type 5: Pressure roaster
Smitherm pressure roaster, nitrogen, 5.33 min, 150 psig, 440 °F (227 °C).

Figure 30.2: Roaster types, five alternative machines: the horizontal drum, the vertical drum with paddles, the rotating bowl, the fluidised bed and the pressure roaster. Drawn from Clarke, Coffee, Volume 2: Technology, Ch. 4.

Quick check

Why does a continuous drum roaster need a gas temperature of only about 250 °C, while a batch Probat uses inlet gas of about 450 °C?

Check your answer

The continuous roaster uses a high gas to coffee ratio, with built-in recirculation, so a large flow of moderately hot gas carries the heat. The batch roaster uses a lower gas flow and a higher inlet temperature to drive the heat in over 10 to 12 minutes.

30.5 Grinding: cutting, not crushing

Roasted whole beans require a cutting rather than a tearing or crushing action to provide a ground coffee with particles of suitable size and shape for subsequent brewing, and this is best produced by means of cutting rolls of various designs. Various impact-type grinders have been used for producing finer grind coffee, and recently so-called flaking rolls have been used to produce particles of a flaked shape, which, it is claimed, gives benefits on brewing.

The finer the average particle size, the more rapid will be the rate of extraction by water of soluble solids and volatile components. In practice, the method of brewing and equipment used will dictate the degree of grinding wanted, as in large-scale extraction for instant coffee manufacture. There can be a problem in home percolators with fines. This is negligible with filters, but the coffee bed should not consist of such fine particles as to impede percolation.

Coffee grinds are qualitatively defined as being coarse, medium or fine, or, as in US usage, according to the brewing device for which they are intended, for example percolator, drip-pot, filter, vacuum makers or espresso. There is no consensus of agreement, whether national or international, as to the average particle size and range that makes up the required sizes in this nomenclature, but various recommendations are available in terms of screen analyses. Coarse or medium grinds are used for household percolators, and generally somewhat coarser for large-scale percolators, whereas fine grinds are used in filter devices, and often somewhat finer in espresso machines. US grinds tend to be coarser for each named grind than in Europe.

Average particle size in micrometres in coffee grinds

Brewing deviceEuropeUSA
Percolator8501130
Filter430800

Achieving as uniform a grind as possible is important. Uniformity is generally best obtained by multi-stage grinding devices, that is, with a series of grinding rolls, progressively reducing particle size. Uniformity is also dependent on the brittleness of the roasted coffee, its moisture content and degree of roast. Grindability is improved by the prior addition of water at an optimum quantity of around 7%, as especially practised in ground coffee for subsequent large-scale percolation. That is one of the benefits of water quenching in section 30.1.

Really fine grinding, say to an average particle size of about 50 micrometres or less, is never really required for home brewing or large-scale percolation, though the requirements of Turkish-type coffee, and of specific processes and products in instant coffee manufacture, need to be met. Such grinding is best carried out under cryogenic conditions, that is, at sub-zero temperatures by use of solid carbon dioxide or liquid nitrogen, above all using impact-type mills. Cryogenic grinding has also been recommended for maximising the retention of flavour volatiles within the roast and ground coffee, which are otherwise released in conventional grinding, above all when accompanied by the generation of heat. So-called grinder gas may be collected by means of condensing devices attached at a suitable position in the exit passages of the grinder. Such condensates will contain useful quantities of aroma and flavour volatiles, which may be incorporated later in manufacture and packing. The major part of these gases is, though, the residual carbon dioxide in the roast whole beans and water vapour. At once on grinding, depending on the degree of grind, about half of this residual carbon dioxide is released, causing some potential hazard if not properly vented. The remainder is subsequently slowly released.

Grinding will also release a further quantity of chaff, either that which is not released from the surface on roasting or that remaining in the centre fold of the whole beans. This material can be disposed of, but it is often incorporated with the ground coffee in a so-called normaliser, by mixing under the action of mixing blades. Above all with dark roast coffees, this process is advantageous in absorbing any exuding oil on the surface and in keeping the ground coffee reasonably free-flowing in subsequent handling. Normalising has a further function in that it is a means of controlling the bulk density, covered in section 30.6, though only within certain limits, by variation of blade speed and disposition.

30.5.1 Size analysis: sieves, plots and uniformity

Screen mesh sizes have been rationalised through the International Standards Organization. For woven wire mesh screens, the relevant standard is ISO 565, 1983, based on preferred numbers: the R40 series, successive apertures in the ratio 1.059, the R20 series, ratio 1.122, and the R10 series, ratio 1.258. The selection for ISO 565 is the R20/3 series, approximate ratio 1.42, as principal sizes. National standards, such as Tyler and ASTM in the USA, once designated screens by numbers, and now, in the UK, France, Germany, the former USSR, Italy and Japan, use aperture size designations in millimetres or microns, though the selections differ. Round sieves are generally standardised at 8 in diameter, or 200 mm.

It is not possible to grind coffee to a single uniform particle size, and a distribution around an average is to be expected. The results of a size analysis are most conveniently plotted graphically with the screen mesh size, that is, aperture in microns, on a linear scale on the y-axis and the weight cumulative oversize on the x-axis. The average particle size may be approximately assessed as that of the aperture at which there is a 50% cumulative oversize. Such plots are generally found to be quite linear, except at the extremes. The slope of the line indicates the degree of uniformity of particle size, quantitatively assessable as a coefficient of variation, CV. The CV value is obtained by determining the difference between the apertures for oversize amounts of 16% and 50% and dividing by the aperture for 50% retention, and expressing this ratio as a percentage figure. Single-stage grinding, once through a single grinding device, will show a relatively high CV figure, but multi-stage grinding will be found to give lower CV values, typically 30% for roasted coffee grinds. The use of the R20/3 series of screens gives a satisfactory separation between aperture dimensions, sequential ratio of about 1.41, for a reasonable spread of retained weights on the screens. BS 410 shows an allowable tolerance on the average size of apertures of about plus or minus 3.2% on each screen.

Grind standards are often given in terms of the screen analyses themselves. The table below shows recommended grinds in the USA, by the US Department of Commerce, established as long ago as 1948, but still generally acceptable. The screen analyses are to be conducted with a Rotap machine, which imparts a circular and tapping motion to the set of sieves, with a timer to cut off the motion, at 5 minutes for these analyses, and a 100 g sample of the roast and ground coffee.

Screen analyses for recommended coffee grinds in the USA

Grind descriptionCombined amount retained on No. 10 plus 14 Tyler screens (%)Combined amount retained on No. 20 plus 28 Tyler screens (%)Amount passing through No. 28 Tyler screen (%)Approximate average particle size (micrometres)
Regular335512 plus or minus 31130
Drip77320 plus or minus 4800
Fine07030 plus 10 minus 5680

30.5.2 Bulk density of ground coffee

The bulk density of ground coffee can be assessed in one of two ways, free fall or packed measurement. In the former, the weight of coffee required to fill a container in free fall, under specified conditions and after levelling the surface, is determined. In the latter, the weight to just fill a similar container after applying vibration or jogging in a clearly defined manner is determined. Different values will be found according to blend, degree of roast, decreasing with increasing severity, grinding degree, increasing with increasing fineness, moisture content and treatment in a normaliser. Values will of course increase from free fall to packed measurements, about 15 to 20% increase. A fine grind coffee may have a bulk density of 0.39 to 0.47 g per ml in free fall. These values will be related to the originating whole roast beans, and to the weight to volume ratios experienced in actual packing, which need to be assessed directly in practice.

From whole bean to grind size

Roasted whole beans, cooled
Cutting rolls, not crushing. Water quenching at about 7% helps grindability.

↓

Multi-stage grinding
Series of rolls, progressively smaller gaps. Uniform grind, CV typically 30%.

↓

Grind set by the brewer
Percolator coarse, about 850 micrometres in Europe, 1,130 in the USA. Filter fine, about 430 in Europe, 800 in the USA.

↓

Normaliser and packing
Chaff mixed back in, oil absorbed, bulk density controlled. Fine grind 0.39 to 0.47 g per ml, free fall.

Figure 30.3: From roasted whole beans to a grind set by the brewer. Drawn from Clarke, Coffee, Volume 2: Technology, Ch. 4.

Quick check

Why does multi-stage grinding give a more uniform grind than single-stage grinding?

Check your answer

Each pair of rolls reduces the particle size a little further, so the spread of sizes narrows. Single-stage grinding, once through one device, gives a relatively high coefficient of variation, while multi-stage grinding typically gives a CV of 30% for roasted coffee grinds.

30.6 Grinding equipment

For small-scale grinding, variants of the disc mill are commonly used. The discs, one of which rotates, carry serrated surfaces and are separated by an adjustable gap to provide the degree of grind required. Feeding is at the centre, and discharge is from the periphery of the discs. For large-scale grinding, of the order of 1,000 kg roast and ground coffee per hour, the multi-roll mill is almost entirely used. The type of serration on the rolls is important. The original patented invention of Lepage for roast coffee provides for slanting U-shaped corrugations running lengthwise in the fast roll and a straight U-shaped corrugated ring around the slow roll. A pair of rolls run at differential speeds with adjustable gaps between them. In the Gump grinder using these rolls in a number of models at different capacities, the housing carries three pairs of rolls, enabling reduction down to the drip grind of the table in section 30.5.1, and an additional pair of rolls is required for fine grinding. The Probat-Werke company also manufactures a multi-roll grinder with a number of special features. Normalisers can be fitted to the multi-roll grinders so that the chaff released can be incorporated with the roast and ground coffee by means of a trough with mixing blades. The combination of grinding and normalising is commonly known to take place in a Granulizer.

For intermediate capacity grinding for the finer grinds, single-passage grinding is used, though classification and recycle are possible. The Fitzmill mill is available, which has vertically rotating blades of various designs, cutting and hammer, within a housing with a perforated screen at the base, of different available hole sizes, through which the ground coffee passes to a collecting container. The Entoleter of Henry Simon is another type of mill that is used, also for killing insects in wheat, and is based on a high-velocity impact principle.

30.7 Practical note: judging a roast and a grind by hand and eye

Take three small lots of roasted coffee, if you can get them: a light roast, a medium roast and a dark roast. Lay them side by side on white paper. Note the colour, the size, and the oil on the surface of the darkest lot, and weigh equal spoonfuls: the darker lot will weigh less for the same volume, because its bulk density has fallen, as in section 30.3. Then grind a small lot coarse and a small lot fine, and shake each through a kitchen sieve. The coarse lot stays on the sieve, the fine lot passes through, and the share that passes is the grind, judged by hand. The book gives the numbers, the reflectance meters and the screen analyses, but the judgement starts with the eye, the hand and the scales.

Try it yourself: Grind the same roasted beans at two settings, coarse and fine, and brew both with the same weight of coffee and the same water, in a filter. Taste the two cups side by side. The fine grind extracts faster and stronger, because the water reaches more surface, and the coarse grind extracts slower and lighter. That is the link between particle size and extraction, tasted in the cup.

30.8 Check yourself: quick self-check

Think about it first: a roaster takes a batch from 12% moisture green coffee to a dark roast, and the as is weight loss at the roaster is 18%. How much of that loss is water, and how much is dry matter and gas, in the book's large roaster example?

  1. What are the dry mass losses for light, medium, dark and very dark roasts?
Check your answer

Light 1 to 5%, medium 5 to 8%, dark 8 to 12%, very dark more than 12%, on a dry basis of green coffee weight.

  1. How much carbon dioxide is held in roasted whole beans at once after roasting, and how is it released?
Check your answer

Of the order of 2 to 5 ml per gram at NTP, up to 10 ml per gram by a more rapid method. It is slowly released over some 2,500 hours whole, or some 360 hours or less when ground, with about half released at once on grinding.

  1. At what bean temperatures do roasting reactions start, peak and fall off for green arabica?
Check your answer

Reactions start at as low as 160 °C, peak at 210 °C, and fall off at 250 °C, with roasting proper starting at about 200 °C after the original moisture is removed.

  1. Name four roaster types and one roasting time for each.
Check your answer

Horizontal drum batch, 10 to 12 minutes; horizontal drum continuous, 3 to 7 minutes; Gothot vertical drum with paddles, about 6 minutes; rotating bowl, about 4 to 6 minutes; fluidised bed, 2 to 4 minutes. Any four, with the times given in the unit.

  1. What is the average particle size of a percolator grind and a filter grind, in Europe and in the USA?
Check your answer

Percolator: 850 micrometres in Europe, 1,130 in the USA. Filter: 430 in Europe, 800 in the USA.

  1. How is grind uniformity measured, and what CV is typical for multi-stage grinding?
Check your answer

By screen analysis, plotted as aperture against cumulative oversize, with the coefficient of variation from the apertures at 16% and 50% oversize. Multi-stage grinding typically gives a CV of 30% for roasted coffee grinds.

Key ideas to carry forward

  • Roasting is a time-temperature process: fast roasting takes a few minutes or less, conventional roasting 12 to 15 minutes, and many roasts 5 to 8 minutes. Degree of roast shows in colour, flavour, dry mass loss and chemical change.
  • Dry mass loss is light 1 to 5%, medium 5 to 8%, dark 8 to 12%, very dark more than 12%, on a dry green basis. The gases released are about 87% carbon dioxide, 7.3% carbon monoxide and 5.3% nitrogen by volume.
  • Roasted beans hold 2 to 5 ml of carbon dioxide per gram at once after roasting, up to 10 ml per gram, under several atmospheres of pressure. About half the carbon dioxide generated is retained.
  • Roasting proper starts at about 200 °C, reactions start at 160 °C, peak at 210 °C and fall off at 250 °C. Net heat is about 295 kcal per kilogram of green coffee in the book's large roaster calculation, with exothermic heat providing about 11% of the gross requirement.
  • Beans swell 40 to 60% on roasting, and bulk density falls from 369 kg per cubic metre for a light roast to 289 for a dark roast. Roast degree is measured by colour reflectance, after grinding and compressing the sample.
  • Emission control uses afterburners and recirculation: a 6 minute batch roast peaks at about 1,200 mg of carbon per cubic metre, average 158, against a 1968 limit of 300, and recirculation at 700 °C cuts the average to 15.
  • Roaster types are horizontal drum, vertical drum with paddles, rotating bowl, fluidised bed and pressure roasters, with times from 2 to 12 minutes.
  • Grinding needs cutting, not crushing. Percolator grind is 850 micrometres in Europe and 1,130 in the USA; filter grind is 430 in Europe and 800 in the USA. Multi-stage grinding gives a CV of typically 30%.
  • Fine grind bulk density is 0.39 to 0.47 g per ml, free fall. Cryogenic grinding, with solid carbon dioxide or liquid nitrogen, keeps flavour volatiles in very fine grinding.

Glossary

Degree of roast

How far roasting has gone, shown by external colour, flavour, dry mass loss and chemical change. Food tech example: a medium roast losing 5 to 8% of its dry mass.

Dry mass loss

The loss of dry matter on roasting, not counting the moisture in the green coffee. Food tech example: a dark roast losing 8 to 12% of its dry mass.

Quenching

Cooling roasted coffee at once to ambient temperature, usually with cold air, or by spraying water in the roasting section. Food tech example: quenching beans into a cooling tray with upflow of cold air.

Exothermic heat

Heat given out by the roasting reactions themselves, once roasting proper starts. Food tech example: exothermic heat providing about 11% of the gross heat requirement in the book's large roaster calculation.

Chaff

The thin silverskin and other light material released from the bean on roasting and grinding. Food tech example: chaff from dry processed robusta, with tenaciously held silverskin, removed by cyclones.

Reflectance

The light reflected from a ground, compressed sample of roasted coffee, used to measure roast degree. Food tech example: a Photovolt meter reading from a sample compressed at about 1,000 psig.

Afterburner

A furnace in the exhaust stack that recombusts roaster gases at high temperature to clean them. Food tech example: an external afterburner at 450 °C cutting the average organic carbon in exhaust gases to 54 mg per cubic metre.

Fluidised bed

A roaster in which hot air lifts and suspends the beans, giving fast, even heating. Food tech example: the Lurgi Aerotherm roaster, 50 kg batches in 2 to 4 minutes.

Cutting rolls

Pairs of rolls with corrugations that cut roasted beans, rather than tearing or crushing them. Food tech example: Lepage rolls with slanting U-shaped corrugations on the fast roll.

Normaliser

A trough with mixing blades that mixes chaff back into ground coffee, absorbing surface oil and controlling bulk density. Food tech example: normalising dark roast coffee to keep it free-flowing.

Coefficient of variation

A measure of grind uniformity, from the apertures at 16% and 50% cumulative oversize in a screen analysis. Food tech example: multi-stage grinding typically giving a CV of 30%.

Cryogenic grinding

Grinding at sub-zero temperatures, with solid carbon dioxide or liquid nitrogen, to keep flavour volatiles in very fine grinding. Food tech example: cryogenic grinding for Turkish-type coffee at about 50 micrometres or less.

Grinder gas

The gas released on grinding, mainly residual carbon dioxide and water vapour, with aroma volatiles that can be condensed and kept. Food tech example: about half the residual carbon dioxide released at once on grinding.

Free fall bulk density

The bulk density measured by filling a container in free fall, without vibration. Food tech example: a fine grind coffee at 0.39 to 0.47 g per ml, free fall.

Rotap machine

A sieve shaker that gives a circular and tapping motion to a nest of sieves, used for grind screen analyses. Food tech example: a Rotap test of 100 g of ground coffee for 5 minutes.

References

  • Clarke, R. J. Roasting and Grinding. In Clarke, R. J., and Macrae, R. (eds.), Coffee, Volume 2: Technology. Elsevier Applied Science, 1987. Chapter 4.

Unit 31: Instant Coffee: Extraction, Drying, Packing and Waste

Note

Unit 31: Instant Coffee: Extraction, Drying, Packing and Waste

A spoonful of instant coffee looks simple. Behind it sits a factory that pulls flavour out of roast coffee with hot water under pressure, catches the smell before it escapes, dries the liquid without cooking it, packs roast coffee that keeps breathing out gas, and finds a use for tonnes of wet grounds. Unit 3 gave you the short version: instant coffee is extracted, concentrated and dried. This unit goes inside each step, with the numbers the book gives.

Questions to think about

  • Why does a factory need 15 to 25% solids in its extract when your cup at home holds only 1 to 3%?
  • If hot water pulls out more coffee, why not simply use the hottest water and the longest time?
  • Why does roast coffee puff up a sealed pack, and why does instant coffee go hard and cakey instead?
  • What is left in the grounds after extraction, and why is it worth burning or feeding to animals?

31.1 From roast coffee to extract: why this is a factory job

A home brew is a single stage: hot water passes through ground coffee once and you drink a dilute liquid of about 1% soluble solids. A factory cannot dry that. It would have to boil off almost all the water, at huge cost, to get a spoonful of powder. So factory extraction is arranged in many stages, with water flowing against the coffee, so the extract that comes out is strong: 15 to 25% soluble solids by weight. Liquid water is the only solvent used in practice for extracting coffee solids.

Three main methods exist. Percolation batteries are the most commonly used: roast and ground coffee sits as a static bed in five to eight vertical columns, hot water enters at the most exhausted column and flows countercurrently, and strong extract is drawn off in batches from the freshest column. An exhausted column is isolated, emptied and refilled. Continuous countercurrent screw extractors carry the coffee through one inclined or horizontal vessel on a rotating screw against flowing hot water, so production is continuous. Stage slurry systems use stirred tanks with centrifuges between the stages; at least three stages are needed, and a plant of the same capacity costs more to build and to run.

Hot water in at the most exhausted column

↓

Middle columns, water getting stronger

↓

Fresh coffee column: strong extract drawn off here

↓

Spent column isolated, grounds discharged, refilled with fresh coffee

Figure 31.1: A percolation battery in five to eight columns. Water enters at the exhausted end and extract leaves at the fresh end. Drawn from Clarke, Coffee, Volume 2: Technology, Ch. 5.

Feed water can enter at 100 °C, but in industrial practice temperatures up to 180 °C are used. A temperature profile forms across the battery: the liquor touching the freshest coffee is around 100 °C, while the feed end is much hotter. Higher feed temperatures give higher yields and stronger extracts. The price is pressure: water at 175 °C needs 120 psig, or 828 kPa, of overpressure to stay liquid. Two working numbers run the battery: cycle time, from putting a fresh column on stream to finishing the draw-off from it, which sets productivity; and draw-off factor, the weight of extract drawn divided by the weight of roast coffee in the column. Finer grinds extract faster and shorten the cycle, but they also raise the pressure drop across the bed until the flow cannot be sustained. Larger plants need coarser grinds, because a big bed must support itself against compression and collapse, and fines drift and block the voids.

Check your answer

Why must factory extract be 15 to 25% solids when a cup is only 1 to 3%? Because every kilogram of water in the extract has to be removed later by drying, so a dilute extract would cost far too much to dry.

Try it yourself: Take two identical spoonfuls of ground coffee. Pour the same amount of hot water through one in a single pass and through the other in three small pours, letting each pour drain before the next. Taste both, diluted to the same final volume. The staged one usually tastes stronger, which is the same idea as a battery, on a kitchen scale.

31.2 What comes out of the coffee: yield and composition

Yield is the dry weight of soluble solids in the extract, as a percentage of the roast coffee used. The book warns that a yield figure misleads unless you know what it refers to: roast coffee or green coffee, dry basis or as is, with moisture counted or not, and whether process losses are included. Green yield, based on the original green coffee, is lower than roast yield by the roasting loss factor.

A medium roast arabica coffee has, on a dry basis, roughly the composition in the table below. The caramelised and condensation products line is a figure by difference: what is left when everything measured directly is subtracted from 100%.

Table 31.1: Approximate typical composition of a medium roast arabica coffee, percent on a dry basis

ComponentPercent
Polysaccharides32.0
Caramelised and condensation products23.2 (figure by difference)
Coffee oil18.0
Protein10.0
Minerals, as oxide ash5.0
Pectins3.0
Chlorogenic acids2.5
Lignins2.0
Aliphatic acids1.5
Caffeine1.4
Quinic acid0.7
Trigonelline0.4
Reducing sugars0.3
Volatile compounds, except acids0.1 (additional, not counted in the total)
Total100.0

Robusta differs mainly in less coffee oil, down by about 6%, more caffeine, up by 1.0% or more, and more chlorogenic acid, up by 1 to 2%. Its soluble solids run about 3% higher than arabica at 100 °C and about 6% higher at higher temperatures.

How much dissolves depends on temperature. Household extractions give 18 to 25% yield from dry medium roast arabica, and exhaustive extraction at 100 °C gives about 28 to 30%; robusta runs 2 to 3% higher in each case. In a household brew, quinic and aliphatic acids extract at 100%, alkaloids and chlorogenic acids at 85 to 100%, volatiles at 40 to 100%, protein at 15 to 20% and melanoidins at 20 to 25%, while only about 1.5% of the coffee oil comes out. Oil and lignin stay in the grounds almost untouched.

Above 100 °C, under pressure, protein, polysaccharides and melanoidins that would not dissolve at 100 °C start to come into solution, so yield rises with temperature. Thaler reached 53% yield from a medium roast Colombian arabica and 58% from a medium roast Angolan robusta, and Kjaergaard and Andresen reported a maximum of 60% for a roasted robusta. That extra yield is not free flavour: caffeine and mineral matter, already fully extracted at low yield, fall as a percentage of the dry solids as yield rises. Acidity rises a little with yield: pH 5.1 at 30% yield and pH 4.8 at 50% yield. Cooling of extracts is routine, because holding extracts hot, say above 60 °C for several hours, spoils them, and volatile distillates and coffee oil need low temperatures and headspace oxygen below 4%.

Check your answer

If two extracts come from the same coffee but one has a 30% yield and the other a 50% yield, do they contain the same things? They contain essentially the same components, but in different proportions: the high yield extract has more polysaccharide and melanoidin material, less caffeine and mineral matter per gram of solids, and it is slightly more acid, pH 4.8 against 5.1.

31.3 Balances, rates and the yield against output trade-off

A factory needs a mass balance: what goes in must come out, as coffee, water or extract. For one cycle of a percolation battery, the book sets the balance over a single equivalent column. A worked example shows how the numbers lock together: with a water to coffee ratio of 3 to 1, a yield of 45% on roast coffee, and a column volume to coffee weight ratio of 2.63, the extract concentration works out at 22.5%, and the draw-off factor is 2.00. Each unit weight of coffee is actually exposed to much more water than 3 to 1, because it meets the water passing through every column on stream.

Rate is a different question from balance: how fast do the solids come out? Two resistances can limit the rate: diffusion inside the particle, as solutes travel through the cellular matrix to the surface, and the liquid film at the particle surface, set by how fast the liquid flows past. The Biot number weighs the two. In practice, raising the draw-off factor raises yield, and raising feed water temperature, say up to 180 °C, raises yield too. The fresh coffee stage needs liquid at 70 to 100 °C so the grounds wet properly, and volatiles are best pulled out before the coffee meets water much above 100 °C.

Productivity is kilograms of soluble coffee solids per hour, at the required yield and at an economic extract concentration. The trade-off is sharp, from a 2.7 m³ continuous countercurrent screw extractor: yield approaches a maximum of 57% on dry roast coffee as residence time grows, and 90 minutes is typical for about 45% yield with a 100 to 175 to 175 °C profile. At 400 kg per hour of roast coffee the yield is 42% and productivity is 168 kg per hour of dry solubles; at 600 kg per hour the yield falls to 30% but productivity rises slightly to 180 kg per hour. With percolation batteries the key is cycle time: shorter cycles mean more coffee per hour, but faster liquor flow means higher pressure drops, and the bed can choke as fines migrate and the bed squeezes shut. Hard feed water is softened first to stop scale: water at 400 ppm as calcium carbonate gives about 184 ppm sodium, and so 700 to 900 ppm in the finished instant coffee.

Check your answer

A plant manager raises throughput from 400 to 600 kg of roast coffee per hour. Does output of soluble solids rise in proportion? No. Yield falls from 42% to 30%, so soluble output rises only from 168 to 180 kg per hour.

31.4 Saving the aroma: how volatile compounds are handled

Roast coffee holds over 700 volatile compounds, but only about 0.1% of its weight in total, and some important ones are present at parts per billion. Many are carried in the coffee oil itself. Water is a poor carrier for the non-polar ones that prefer oil, and heat and air destroy them fast. So factories handle the volatiles separately: strip them out early, condense them carefully, and put them back at the end.

Steam is the favoured carrier for pre-stripping: roast coffee, often already inside a percolation column, is steamed so the volatiles distil off and condense with the steam, and the column then goes on stream in the normal way. Extracts can be stripped the same way, by partial evaporation and condensation. Stripping efficiency depends on the relative volatility of each compound against water, taken as 1: below 1 for acetic acid, 120 for acetaldehyde. Evaporating 80% of the water would remove virtually all the acetaldehyde but leave 31% of the acetic acid behind. In practice the amount evaporated for volatile recovery ranges from a few percent to about 70%.

Vapours must be condensed as completely as possible, usually with chilled water in a tubular condenser, though liquid ammonia and dry ice work better. Non-condensible gases carry volatiles away, so they must be kept as low as possible. Two stage condensation is also used: a first cut at about 90 °C is discarded, while the cut condensed at 20 °C, about 5 to 20% of the original vapour, is kept.

Arabica roast coffee holds up to about 18% coffee oil by weight, and much of its aroma lives in that oil. Oil can be pressed out of roast coffee or spent grounds mechanically and purified. Coffee oil, plain or fortified with steam distillates, is plated onto the finished instant coffee in small amounts, of the order of 1.0% or less, usually as the jars are filled. Gases released during grinding can be trapped as an aroma frost and added the same way, and volatiles can be packed into small capsules that float to the surface of the cup and release aroma slowly. Compounds not taken from coffee itself are not used in 100% pure coffee extracts.

31.5 Spray drying: turning extract into powder in seconds

Drying takes the extract to a low moisture content of 2 to 5%. Two commercial routes dominate: spray drying, where water evaporates into hot air, and freeze drying, where frozen water sublimes under vacuum. Drum drying is hardly used now. A spray dried powder must be free flowing, with particles about 200 to 300 µm across on average, and it must dissolve fast: a heaped spoon should dispense about 2 to 3 g for a single cup of about 170 ml. Spray dried coffee has been agglomerated into granules since about 1968, less to speed dissolving than for looks, since granules resemble roast and ground coffee.

Spray drying has four steps: disperse the liquid as droplets, contact the spray with hot air, evaporate the water while the air cools, and separate the dried product from the air. For coffee, contact is essentially cocurrent, spray and air moving in the same direction, in a tall, narrow tower. The favoured spraying device is the centrifugal pressure nozzle, with an internal swirl chamber that throws liquid out as a hollow cone shaped sheet that breaks into droplets. Pressures are of the order of 690 kPa, or 100 psig, and much higher. Flow follows Q = K times the square root of P divided by density, and the ratio Q over the square root of density is called the flow number. Spinning discs, common in milk drying, give fine sprays around 80 µm and need wide chambers, so they suit milk better than coffee, where droplets are around 300 µm.

Drying runs in two periods: a constant rate period, when water evaporates freely from the droplet surface, and a falling rate period, when the inside cannot feed water to the surface fast enough. Droplets leave the nozzle at about 40 m per second and slow to a free fall velocity of about 1.7 m per second for a 400 µm particle in air at 260 °C. A humidity chart example in the book heats air from 20 °C to 220 °C and cools it to 100 °C, giving a wet bulb temperature of 54 °C and an outlet dew point of about 43 °C, below which moisture condenses in ducts.

Keeping the volatiles in is the hard part. The selective diffusion idea, worked out by Thijssen and colleagues, explains how a drying droplet traps them: as the droplet surface dries, the diffusion coefficient of a volatile compound falls far faster than that of water, so a skin forms that lets water out but holds volatiles in. The maths uses the Fourier number, Fo = D times t divided by r squared: retention of 0.90 or better needs a Fourier number below 0.01, while 0.37 retention corresponds to 0.05, and losses stop when the surface water concentration falls to about 25% by weight. Near the nozzle, droplets can balloon, and circulation inside a fast moving droplet can carry volatiles to the surface. An oil phase in the extract, say up to 0.1% by weight on a solids basis, helps retention.

The solids effect is dramatic in model work: acetone in maltose solution retained only 4% when spray dried from 20% solids, but 68% from 55% solids. Retention also rose with outlet air temperature from 75 to 140 °C, less so with feed temperature from 20 to 80 °C, and fell with rising outlet air humidity. In real coffee the gains are smaller, because the non-diffusive losses near the nozzle play a large part. Reducing sugars can react with free amino groups in Maillard browning, and a check of reducing sugar before and after drying shows whether it has happened. Drying takes only seconds, which helps, but powder resting on hot surfaces at the outlet can still be damaged.

Coffee extract in, through centrifugal pressure nozzles at the top

↓

Hot air in, cocurrent with the spray, droplets dry as they fall

↓

Dried powder falls out at the base of the tower

↓

Outlet air carries fines to an external cyclone for separation

Figure 31.2: A spray drier for coffee extract, with twin centrifugal pressure nozzles and primary separation at the base of the tower. Drawn from Clarke, Coffee, Volume 2: Technology, Ch. 6.

Check your answer

Why does a factory concentrate its extract before spray drying? Two reasons: drying water is expensive, and volatile retention rises sharply with solids concentration, from 4% retention at 20% solids to 68% at 55% solids in the model work.

31.6 Freeze drying and pre-concentration: cold routes to the same powder

Freeze drying starts by freezing the extract, then subliming the ice under high vacuum with controlled heat. Coffee extracts are frozen to around minus 40 °C, often after foaming with inert gas, then granulated while frozen and loaded into trays. Commercial freeze drying of coffee began about 1965, and it has stayed successful where many other freeze dried foods faded.

The physics sets strict limits. At minus 24 °C, with 77% coffee solids, the remaining 23% water is unfreezable, equal to 0.30 kg per kilogram of solids: bound water that will not form ice. The frozen matrix must stay rigid, or it collapses as pores open. Coffee extracts are rigid enough at minus 24 °C, and minus 30 °C gives a safety margin, because their polysaccharides stiffen the matrix; fruit juices, rich in simple sugars, may need minus 54 °C. A typical operating vacuum is 0.3 mmHg, matching ice at minus 30 °C, and the condenser runs near 0.03 mmHg and minus 50 °C. Subliming ice needs latent heat of the order of 685 cal per gram. Ice conducts heat well, about 2.2 W per metre per kelvin, but dried coffee conducts about one hundredth as well, so a dried layer can limit the rate. Volatile retention is generally good: slow freezing retains more, and retention rises as the volatile compound gets bigger. It is not related to boiling point or relative volatility.

Drying water is expensive, and freeze drying most of all, so extracts are usually pre-concentrated first. Evaporation uses film evaporators, with residence times of seconds. Viscosity climbs steeply after about 42% soluble solids, and concentrations above that get hard to reach, though a wiped film evaporator is claimed to reach about 70%. Three effects are considered the maximum suitable for coffee extracts. Freeze concentration starts from a 25% extract, which begins to freeze at about minus 2 °C. Cooling to minus 4 °C leaves a liquid at about 40% solids, with ice making up 50% of the original weight of water. The ice must be separated with as little adhering liquid as possible, and large ice crystals are preferred because they carry less liquid. Volatile loss is minimal, a marked advantage over evaporation, and about 40% solids is the practical limit, set by viscosity. Energy use is comparable to single effect evaporation but not to multi-effect, and capital costs run about 2.5 times those of single stage evaporation on 1977 costing. Reverse osmosis removes water through membranes without heat, but the book notes this route is probably much less used than the other two for coffee extracts.

Check your answer

Why freeze an extract at minus 24 °C rather than colder? At minus 24 °C all the water that can freeze in a coffee extract has frozen, and the matrix is rigid enough not to collapse. Going colder costs refrigeration for no gain in frozen water, though minus 30 °C is used as a safety margin.

31.7 Fines, agglomeration and the dust and fire risk

Spray nozzles make a range of droplet sizes, so the powder has a range of particle sizes too. In practice a fraction is taken out as fines. A cut-off is chosen, for example only 5% at 200 µm and below left in the product. Primary separation can happen inside the drier itself: outlet air leaves through a duct at right angles to the tower, and the coarse product falls by gravity from the cone shaped base. The air then carries the fines to an external cyclone, chosen to catch, say, 99% at 20 µm. How much powder may be emitted to the air is set by local or national law.

Agglomeration builds granules from spray dried powder, averaging about 1400 µm, by rewet methods, using steam or water droplets, or by straight-through methods that cause cohesion during drying. Instant coffee agglomerates through thermoplasticity at higher surface moisture, and the granules must then be dried back to normal instant moisture, often in a fluidised bed. Granules are generally darker than powders.

Quality control covers bulk density, moisture by vacuum oven (ISO 3726) or Karl Fischer titration, colour by reflectance meter, and screen analysis (ISO 7532), from which the size at 50% cumulative amount gives average particle size. Insoluble matter is measured by ISO 7534: traces below 1000 ppm at sizes above 100 µm. Flavour is checked by trained panels, with gas chromatography as back-up. Powder handling carries a real hazard: instant coffee and chicory powders can self-heat and ignite, and spent grounds have a high heat content. Plant design must treat dust as a fire risk.

31.8 Packing roast coffee: the carbon dioxide problem

Roast coffee keeps breathing after roasting. The gases it releases are mainly carbon dioxide, about 87% by volume, with 7.3% carbon monoxide and 5.3% nitrogen. Whole beans hold about 2 to 5 ml of CO₂ at normal temperature and pressure per gram of roast coffee, and probably more, with darker roasts and robusta holding more than lighter roasts and arabica. The gas is held under several atmospheres of pressure inside the bean. Grinding releases much of it at once: Heiss and Radtke put it at 45% released within 5 minutes of grinding for a fine ground coffee, and Barbara measured 30% lost within 5 minutes at an average particle size of 1000 µm and 70% at 500 µm.

Pack that coffee straight into a sealed pack and the gas has nowhere to go: pressure builds until the pack bursts or the lid blows on opening. Leave the pack open and oxygen gets in and stales the coffee. Radtke measured release from whole beans over up to 2500 hours to reach near equilibrium. Decaffeinated coffees release more gas, and faster. The figures are not exact totals, because pressure building up in the closed measuring system, reaching about 600 mmHg, holds back complete release.

Table 31.3: Release of carbon dioxide from roast whole bean coffee, ml at normal temperature and pressure per 100 g

Time from start of measurement (hours)Coffee 1Coffee 2Coffee 1, decaffeinatedCoffee 2, decaffeinated
00000
200180200275220
400250265340275
1400405
1800400450
2500378

Ground coffee releases far faster than whole beans: a drip grind sample packed one hour after grinding still gave off an average of 121 ml of CO₂ per 100 g after 40 days. Five answers exist for a closed pack: a container strong enough to hold the pressure, which is impractical; a bag permeable to CO₂, or fitted with a one-way valve; packing under vacuum, though even a full vacuum may not be enough; a very large headspace, also impractical; or holding the roast coffee before packing until much of the gas has gone.

Stability is the other half. Staling comes from oxygen acting on the roast coffee, faster at higher temperature and higher moisture, plus a slow loss and unbalancing of volatiles. Whole beans in an air pack change little for 10 to 12 days, show a flavour difference in the brew after about 40 days, and reach the limit of acceptability after about 70 days. Packed under vacuum with low oxygen, significant change takes about 180 days, and the coffee stays fully acceptable for at least 18 months. Each 10 °C rise in temperature cuts shelf life by 50%, and each 10 °C fall doubles it, which is why refrigerator storage helps.

Ground coffee stales faster, so low oxygen packing is needed. At 0.5% initial oxygen by volume, in a closed pack stored around 21 °C with moisture below about 4%, shelf life is about 6 months to high quality, 12 to 17 months to medium quality, and 20 to 25 months to low quality. At 1% oxygen the figures are 4, 9 to 17 and 14 to 20 months. In an air pack at 21% oxygen, medium quality lasts only 10 to 15 days. A reference sample at zero oxygen stored at minus 20 °C showed virtually no change after 20 months. Oxygen is also picked up at grinding and can sit absorbed in the coffee, so a vacuum pack's measured oxygen can rise for several days after packing, then fall again. Oxygen is lowered by pulling a high vacuum just after filling, or by flushing with inert gas. The book regards oxygen by weight as the better measure: a maximum around 20 µg per gram of coffee matches 1% oxygen in a high vacuum pack. An old 1947 figure for the oxygen needed to stale roast coffee is 14 ml per pound, or 40 µg per gram.

Pack types follow from this. Whole beans can go in simple bags for short periods, but long life needs a gas tight laminate, a one-way valve such as the Goglio or Hesser valve, which lifts at a set pressure to release excess CO₂, and flushing with nitrogen or CO₂. Ground coffee packs in three forms. Cans, the oldest type, are tin-plated steel with lacquer, impermeable to water, air and volatiles; the popular United States size for 1 lb holds about 1080 ml. A can must withstand a vacuum of about 28.5 inHg against 30 inHg barometric pressure, to leave less than 1% oxygen, and the coffee must be degassed first so later CO₂ evolution only restores the pressure to about atmospheric. Hard packs are flexible laminates that collapse onto the coffee under high vacuum to form a brick: a typical laminate is PETP at 12 µm, aluminium at 12 µm and LDPE at 70 µm, inside a cardboard box. The pressure must not rise past about 735 mbar, or the brick goes soft, which consumers read as staleness even when the coffee is fine. Degassing times for hard packs run from 1 to 30 hours, and a medium roast, coarse grind needs about 8 hours. Soft or pillow packs use no vacuum and rely on inert gas flushing, with good seals and no pinholes.

Roast coffee releases CO₂: 2 to 5 ml per gram in whole beans

↓

Degas in bulk, or vent through a one-way valve

↓

Pack at low oxygen: vacuum, or inert gas flush

↓

Keep oxygen near 20 µg per gram and moisture below about 4%

Figure 31.3: The roast coffee packing problem in four steps: gas out, oxygen out, then seal. Drawn from Clarke, Coffee, Volume 2: Technology, Ch. 7.

Check your answer

A hard pack goes soft on the shelf. Has the coffee gone stale? Not necessarily. Softness means pressure has risen inside, from CO₂ evolution or air leaking in. The book notes softness does not imply deterioration, but consumers often read it that way, which is why the pressure limit of about 735 mbar matters.

31.9 Packing instant coffee and the machines that do it

Instant coffee has no CO₂ problem, but it has two of its own: oxygen spoils its aromatics, and moisture makes it cake. At about 7 to 8% moisture it cakes into a pasty or solid mass, even from air of quite low relative humidity. Clinton's work on sealed jars is instructive: agglomerated granules at 4.3% moisture and 4% oxygen, and freeze dried coffee at 2% moisture and 2% oxygen, both stayed acceptable at 18 months, though the freeze dried coffee lost acceptability markedly in the first few months and then levelled off. The lesson is to keep in-jar oxygen down to levels comparable with roast and ground coffee. Even so, simple spray dried coffee held below 4 to 5% moisture keeps its original quality for at least two years in an air pack at 21% oxygen. Instant coffee was once packed entirely in tins; around 1960 glass jars took over, sealed with a waxed paper or metal foil diaphragm and a screw cap.

Packing plant is specialised. Degassing plant holds roast coffee in vented tote bins, sometimes warmed to about 30 °C, or applies vacuum, then inert gas, then a second evacuation, and finally flushes with inert gas to just above atmospheric pressure. Soft pack machines form, fill, seal and flush at 70 to 120 packs per minute. Instant coffee jar fillers run up to 360 jars per minute for small sizes, filling by volume with an adjustable vacuum to hit weight, and gas packing the jars to a set oxygen level; some spray coffee oil onto the instant coffee just before filling. Weight control matters twice over: the law sets strict rules, and the product is expensive, so the maker wants minimum give-away above the legal weight.

31.10 Waste: what the grounds are worth

Instant coffee manufacture leaves a large residue. Roast ground beans are treated with water at high temperature and pressure, and the insoluble grounds are blown to silos at 75 to 80% moisture, then screw pressed to a cake of about 50% moisture. Only the coarsest grounds press well: fine material stays in the press effluent at about 2 to 7% by weight. A Pennwalt process feeds that effluent to a continuous centrifuge with a claimed solids recovery of 85 to 95%. The book gives a rule of thumb: 1 tonne of green coffee releases 480 kg of spent grounds, assuming 20% loss on roasting and 40% extraction efficiency. Three soluble coffee plants in Brazil together produced 33,600 kg of dry spent grounds a day by 1977, double their output ten years earlier.

Blair's analysis, the fullest the book cites, shows spent grounds as a fibrous, high energy material.

Table 31.4: Composition of spent coffee grounds, after Blair, percent unless stated

ComponentValue
Moisture7 to 8
Crude protein10 to 12
Crude fibre35 to 44
Ether extract22 to 27
Nitrogen-free extractives13 to 18
Ash0.25 to 1.0
Gross energy6930 kcal per kg dry matter
Starch equivalent, ruminants75

The protein figure misleads: roasting cuts the biological value of the protein, and the soluble protein has already been extracted. The oil, 9 to 20%, is rich in palmitic acid at 34.2% of the lipid and linoleic acid at 37.0%, with more than 50% unsaturated fatty acids and an iodine value of 85 to 93, which implies a real risk of rancidity, partly offset by natural antioxidants. The high unsaponifiable matter, 9 to 12%, would be a problem for making edible oil.

Two uses are economically viable, and both need the residue processed near the soluble coffee plant, because hauling so much water is wasteful. As feed, the pressed grounds are dried to 10 to 12% moisture in a rotary drier and sold, as in the Cherco product launched in 1969, for ruminant feeds at up to 7.5%, with the slightly bitter taste limiting inclusion to below 10% in complete feeds. About 40,000 tonnes a year were being eaten in animal feeds in Western Europe after 1970. As fuel, the pressed grounds are burnt as the base load in refractory cells, with gas or oil added to hold peak steam demand. They burn virtually smoke free to a low ash residue. The catch is moisture: at 75% moisture the calorific value falls to 3300 kcal. The book's rule is to dry grounds to around 33% for efficient fuel use, and never below 30% with high heating temperatures, because of the fire risk.

Primary processing waste belongs to the same story, in brief. Green coffee is only 50 to 55% of the dry matter of the ripe cherry, and 18% on a fresh weight basis, so nearly half the cherry becomes by-product. Dry processing leaves husks, which burn at about 14 MJ per kg and can be charred to charcoal at 29.4 MJ per kg; husks can also be fed to steers or dairy cows at up to 20% of the ration. Wet processing leaves pulp and large volumes of wastewater: pulping waters carry a chemical oxygen demand of 13,900 to 28,000 ppm and a five day biological oxygen demand of 1800 to 2920 ppm, while fermentation waters carry a chemical oxygen demand of 3000 to 10,000 ppm and a biological oxygen demand of 1250 to 2200 ppm. Water use can be cut from a traditional 91,000 litres per tonne of clean coffee to 13,650 litres with recirculation. Biological filters can cut biological oxygen demand by more than 94% at a loading near 0.1 kg per cubic metre per day, though a traditional filter for a factory making 1 tonne of clean coffee a day needs about 400 m² of land and 500 to 800 m³ of packing.

Practical note: If you run a small roastery, the book's packing numbers translate directly. Grind releases gas fast, so grind close to packing time, degas ground coffee for hours, not minutes, before sealing a hard pack, and judge a soft pack by its seal and its oxygen, not by how firm it feels. For instant coffee at home, keep the jar sealed and dry: moisture above 7% cakes it, and no amount of careful brewing fixes a caked jar.

Key ideas to carry forward

  • Factory extraction is multi-stage and countercurrent, giving 15 to 25% solids, because drying a 1% home brew would cost too much water removal.
  • Feed water up to 180 °C raises yield, but water at 175 °C needs 120 psig overpressure, and finer grinds choke the bed with pressure drop.
  • Household brews yield 18 to 25% and exhaustive extraction at 100 °C gives 28 to 30%; at higher temperatures yields reach 53 to 60%, but caffeine and minerals fall as a share of the solids and pH falls from 5.1 to 4.8.
  • Yield and output trade against each other: 400 kg per hour gives 42% yield and 168 kg per hour of solubles, while 600 kg per hour gives 30% yield and 180 kg per hour.
  • Volatiles are stripped early with steam or evaporation, condensed carefully, and plated back at about 1% oil or less; retention in spray drying rises sharply with feed solids.
  • Spray drying makes 200 to 300 µm powder at 2 to 5% moisture in seconds; freeze drying freezes extract to about minus 40 °C and sublimes ice at about 0.3 mmHg.
  • Evaporation stalls above about 42% solids through viscosity; freeze concentration reaches about 40% solids at minus 4 °C with little volatile loss.
  • Roast coffee releases CO₂, 2 to 5 ml per gram, and packs need degassing, vacuum or inert gas, and a one-way valve or a permeable bag; ground coffee at 0.5% oxygen keeps 6 months at high quality against 10 to 15 days in air.
  • Instant coffee cakes at 7 to 8% moisture and needs low oxygen and low moisture in the jar; jar fillers run up to 360 jars per minute.
  • One tonne of green coffee leaves 480 kg of spent grounds, which burn as fuel when dried to about 33% moisture or feed ruminants at up to 7.5% after drying to 10 to 12% moisture.

Glossary

Percolation battery

A row of five to eight columns of roast and ground coffee through which hot water flows countercurrently, with extract drawn off from the fresh column. Food tech example: a battery fed with water up to 180 °C under pressure.

Yield

The dry weight of soluble solids in an extract as a percentage of the roast coffee used, with the basis stated. Food tech example: a household brew at 18 to 25% yield against exhaustive extraction at 28 to 30%.

Selective diffusion

The idea that a drying droplet forms a skin that lets water out fast while holding volatile compounds in, because their diffusion falls faster than that of water. Food tech example: spray drying a 55% solids feed to retain 68% of acetone in model work, against 4% from a 20% feed.

Fourier number

A dimensionless number, diffusion coefficient times time divided by radius squared, used to predict volatile retention in drying. Food tech example: retention of 0.90 or better needs a Fourier number below 0.01.

Centrifugal pressure nozzle

A spray nozzle with an internal swirl chamber that throws liquid out as a hollow cone of droplets. Food tech example: a nozzle running near 690 kPa to make droplets around 300 µm.

Agglomeration

Joining spray dried powder particles into granules averaging about 1400 µm, by wetting and redrying. Food tech example: agglomerated granules that look like roast and ground coffee.

Freeze drying

Drying by freezing a liquid and subliming the ice under high vacuum with controlled heat. Food tech example: coffee extract frozen to about minus 40 °C and dried at about 0.3 mmHg.

Degassing

Letting roast coffee release its carbon dioxide before packing, in bulk bins, sometimes under vacuum with inert gas. Food tech example: a medium roast coarse grind degassed for about 8 hours before hard packing.

Caking

Instant coffee setting into a pasty or solid mass when it picks up moisture, at about 7 to 8% moisture. Food tech example: an instant coffee jar left open in damp air turning into a solid lump.

Spent coffee grounds

The insoluble residue left after extracting roast coffee for instant coffee, pressed from 75 to 80% moisture to about 50%. Food tech example: spent grounds dried to about 33% moisture for boiler fuel.

References

  • Clarke, R. J. and Macrae, R. (eds). Coffee, Volume 2: Technology. Elsevier Applied Science, London, 1987. Chapters 5, Extraction; 6, Drying; 7, Packing of Roast and Instant Coffee; 9, Waste Products.
  • Clarke, R. J. Extraction. In Clarke, R. J. and Macrae, R. (eds), Coffee, Volume 2: Technology, Chapter 5.
  • Clarke, R. J. Drying. In Clarke, R. J. and Macrae, R. (eds), Coffee, Volume 2: Technology, Chapter 6.
  • Clarke, R. J. Packing of Roast and Instant Coffee. In Clarke, R. J. and Macrae, R. (eds), Coffee, Volume 2: Technology, Chapter 7.
  • Adams, M. R. and Dougan, J. Waste Products. In Clarke, R. J. and Macrae, R. (eds), Coffee, Volume 2: Technology, Chapter 9.

Unit 32: Decaffeination and Brewing the Cup

Note

Unit 32: Decaffeination and Brewing the Cup

Two questions sit behind this unit, and they pull in opposite directions. The first: how do you take the caffeine out of coffee and leave the coffee behind? The second: how do you get the coffee out of the bean and into the cup, and why does the same coffee taste different from five different brewers? The book answers the first with three industrial routes, solvent, water and supercritical carbon dioxide, and the second with a careful review of brewing science plus one researcher's own experiments on five machines, tasted by twelve experts.

Questions to think about

  • If caffeine is only 1 to 2% of the bean, why is it so hard to remove without damaging flavour?
  • Why does a decaffeination plant wet its beans to 40% moisture or more before it starts?
  • Why does a finer grind not always give a stronger cup?
  • Why did the gravity filter brewer, the simplest machine tested, give the most balanced cup?

32.1 Why take the caffeine out, and why do it before roasting

Many of the physiological effects of a cup of coffee come from its caffeine. Arabica holds about 1% caffeine and robusta about 2%. A cup of regular coffee holds on average 85 mg of caffeine, with a range of 50 to 150 mg depending on preparation, blend and cup size. A cup of instant coffee holds about half that, with a reported range of 40 to 110 mg.

Decaffeination began with Dr Ludwig Roselius and Dr Karl Wimmer of the Kaffee-Handels-Aktien-Gesellschaft in Bremen, Germany, in a German patent dated 1905. Their company sold a caffeine-free coffee under the name Kaffee Hag. Roselius took the business to the United States around 1912. During the First World War the American Kaffee Hag business was expropriated, passed to a Mr Gund, and later sold to the Kellogg Company. After the war Roselius started the Sanka Coffee Corporation, first importing decaffeinated green coffee from Germany and roasting it in New York. In 1927 he made a deal with General Foods, which financed the Sanka factory in Brooklyn, New York, and in 1932 he sold his interest to General Foods. In December 1937 General Foods bought the American Kaffee Hag Company from Kellogg, gaining complete control of decaffeinated coffee in the United States and Canada at that time. About 1955 the Nestle Corporation and others introduced decaffeinated coffee in the United States, and in the 1970s General Foods bought the Kaffee Hag Company in Germany back from the heirs of Roselius.

Commercial decaffeination is mostly done on green coffee beans before roasting, to keep flavour and aroma problems small. The desirable qualities of the drink develop during roasting, so doing the chemical work first, on the green bean, loses less of what roasting will create. The two most common techniques are solvent decaffeination and water decaffeination, and a third, supercritical carbon dioxide, is treated separately because its plant is so different. The book also notes patent literature on removing caffeine from roasted coffee extract.

Demand was moving fast when the book was written. Per person consumption of undecaffeinated coffee in the United States had fallen steadily since 1970, while per person consumption of decaffeinated coffee had risen 153% in 12 years. The ratio of undecaffeinated to decaffeinated cups per person fell in that period from 16 to 1 down to 4 to 1. The drive, in the book's words, was for high quality decaffeinated coffees produced in an economical and safe environment.

Check your answer

Why decaffeinate green beans instead of roasted beans? Because the flavour and aroma of coffee develop during roasting. Working on the green bean keeps the later roasting step intact, so fewer aroma and flavour changes and losses occur.

32.2 Solvent decaffeination: wet beans, the right solvent, and a clean strip

The most direct approach is to contact pre-wetted green coffee with a suitable solvent, in batch or continuous equipment. Many solvents have been suggested. The book lists seven criteria for choosing one: safety, cost of solvent, caffeine solubility, caffeine specificity, ease of solvent removal and recovery, toxicity and chemical reactivity, and environmental effects.

History moved through three solvents. Benzene was used first, then replaced by trichloroethylene when it became available at reasonable prices, because it was non-flammable and less toxic. Then many makers switched to methylene chloride, or dichloromethane, which has a higher caffeine solubility and a lower boiling point, so it is easier to remove from the beans. Trichloroethylene had been under investigation by the United States Food and Drug Administration since 1976, and the only chlorinated solvent used either directly or indirectly by the time of the book was methylene chloride. Besides methylene chloride, the only known commercial solvents were ethyl acetate, coffee oil or other triglycerides, and supercritical carbon dioxide.

Water is the surprise ingredient. Unlike most solvent extraction, decaffeinating beans needs a lot of water added: a minimum water content of about 20% is required, and below 20% only limited decaffeination happens, no matter how much solvent is used. As much as 55%, or saturation, is used. One explanation is that water softens and opens the cell structure so caffeine can diffuse out. The book calls a second explanation more likely: caffeine bonds to chlorogenic acid and its potassium salt, potassium chlorogenate, a complex essentially insoluble in the non-polar solvents commonly used, and water frees it.

A typical semi-continuous process runs in a battery of columns. Fresh green beans are steamed for half an hour, reaching 16 to 18% moisture, then pre-wetted to above 40% moisture. The wetted beans are extracted countercurrently by solvent at 50 to 120 °C. The column from which most caffeine has been removed, 95 to 98% decaffeinated, is isolated, drained of solvent and steam stripped to remove all residual solvent. The decaffeinated beans are discharged and dried at once. Caffeine rich solvent leaving the fresh side goes for caffeine recovery and solvent regeneration, and the recycled solvent must be saturated with water first, or it will dry out the pre-wetted beans.

Typical operating conditions from United States Patent 3,671,263 are: steaming 30 minutes at 230 °F, pre-wetting at 150 °F to 42% water by weight, caffeine extraction with methylene chloride at 150 °F for 10 hours, using 4 lb of methylene chloride per pound of green coffee beans, and steam stripping for 1.5 hours.

How the solvent is cleaned depends on its boiling point. A volatile solvent is simply evaporated, usually in multi-effect or other low utility evaporators, leaving a caffeine rich sludge to refine. A high boiling solvent needs a water wash in a countercurrent liquid-liquid extractor, or a high vacuum sublimator, both to regenerate the expensive solvent and to recover the valuable caffeine. All waste streams must be very low in solvent content, to meet environmental and legal requirements.

Green beans steamed 30 minutes, to 16 to 18% moisture

↓

Pre-wet to above 40% moisture

↓

Countercurrent solvent extraction at 50 to 120 °C, to 95 to 98% decaffeinated

↓

Steam strip residual solvent, dry beans at once

↓

Caffeine rich solvent to recovery and refining, clean solvent recycled

Figure 32.1: Solvent decaffeination of green coffee in a battery of columns. Drawn from Katz, Coffee, Volume 2: Technology, Ch. 3.

32.3 Water decaffeination: washing caffeine out with coffee's own extract

Water decaffeination was invented in 1941 by Berry and Walters of the General Foods Corporation. The idea is simple and neat: wash caffeine out of the beans with water, but use water that already holds everything else the bean would lose. The extraction water contains essentially equilibrium quantities of the non-caffeine soluble solids, about 15% solids other than caffeine, and no, or very little, caffeine. That way the water takes caffeine and leaves the rest, instead of stripping more than 20% of the bean's weight in soluble solids.

The claimed advantages are higher extraction rates, no extraction of water-insoluble waxes by solvent, purer caffeine in recovery, and less heat treatment of the beans because there is no solvent stripping step and no contact between solvent and beans. Pre-steaming and pre-wetting also disappear, because the water extract supplies all the water needed to free the caffeine inside the bean.

The plant is again a battery of columns, with green beans extracted countercurrently by the water extract. Dry beans absorb water preferentially, so the solids concentration can climb as high as 30%. A bean residence of about 8 hours removes about 98% of the original caffeine. The decaffeinated wet beans hold about 53% moisture. Adhering solubles are washed off with fresh water, and that wash water goes back into the caffeine-free extract to bring it to its starting level of about 15% solids. The washed beans are air dried, then sold as decaffeinated beans or processed into decaffeinated roast coffee or decaffeinated instant coffee.

The caffeine rich extract, about 0.5% caffeine by weight, must be recycled after its caffeine is removed, to keep the solids balance. Caffeine is usually removed by contact with a specific solvent, though newer patents use solid adsorbents. The extract is centrifuged first to remove suspended solids such as coffee chaff, then contacted with an immiscible solvent such as methylene chloride. A rotary disc contactor works best, cutting caffeine in the green coffee extract to below 0.05% by weight. The extract rises through the dense methylene chloride at around 80 °C, or 180 °F, which gives higher caffeine distribution and fewer emulsion problems. Note: 80 °C is 176 °F, not 180 °F, and the book itself prints 180 °F. A solvent to extract ratio of about 4 to 8 removes 99% or more of the caffeine. After contact, dissolved solvent must be stripped from the lean extract in a distillation column before it touches beans again. The overheads come off at the azeotropic composition, 98.5% by weight for methylene chloride, unless excess steam is used.

Adsorbents had a poor record: in the past they altered the quality of the final coffee and were not economic. A later Coffex AG process claims to fix this with activated carbon preloaded with other coffee extract substances, or with substitutes of similar molecular size, especially carbohydrates such as cane sugar, so the charcoal takes up as few extracted substances as possible. The claim is that apart from caffeine, practically no other coffee components are removed.

Green beans into a battery of columns

↓

Extracted countercurrently for about 8 hours by caffeine-free extract at about 15% solids

↓

Decaffeinated wet beans at about 53% moisture, washed and air dried

↓

Caffeine rich extract, about 0.5% caffeine, decaffeinated by solvent contact and recycled

Figure 32.2: Water decaffeination, where the beans are washed with coffee extract that holds everything except caffeine. Drawn from Katz, Coffee, Volume 2: Technology, Ch. 3.

Check your answer

Why use coffee extract instead of plain water to wash the beans? Plain water would dissolve all the soluble solids, more than 20% of the bean's weight. Extract already saturated with those solids can only take up caffeine, so the bean keeps its other soluble material.

32.4 Supercritical carbon dioxide: an inert gas that behaves like a solvent

Supercritical carbon dioxide decaffeination was commercialised by HAG-GF in Germany, based on the 1970 patent of Studiengesellschaft Kohle of Mulheim, Germany. Its great advantage is the use of an inert gas compressed to pressures high enough to dissolve caffeine, in a closed cycle. Its great disadvantage is the same high pressure: costly equipment and batch processing.

Bean preparation matches the other solvent routes: mechanical cleaning to remove dust and chaff, then steaming and wetting to raise moisture to a 30 to 50% range. Pre-moistened beans go into the extractor vessel while solid adsorbent goes into the adsorption vessel. Carbon dioxide is loaded in and a circulation pump starts. A heat exchanger raises temperature and so raises pressure to the extraction level. As carbon dioxide flows through the vessels, caffeine leaves the moist beans, and caffeine rich carbon dioxide passes to the adsorber, where caffeine sticks to activated carbon. Caffeine-free carbon dioxide then starts a new cycle. The average caffeine concentration in the carbon dioxide is low, so large quantities must be circulated, and equipment is designed to keep pressure drop small.

The patent proposes moist carbon dioxide between 40 °C and 80 °C at pressures between 120 and 180 bar, with caffeine removed from the carbon dioxide by lowering temperature or pressure to precipitate it, or to aid adsorption. At the end, carbon dioxide is unloaded to a holding vessel, beans and adsorbent are discharged, and the moist decaffeinated beans are dried to about 10% water in a vacuum or hot air drier. Further processing matches any other coffee bean.

Product quality is claimed to be good for two reasons: carbon dioxide is chemically stable and inert, so it does not react with coffee constituents, and it is highly selective for caffeine, avoiding losses of non-caffeine solids. Quality is claimed to be comparable to a regular undecaffeinated coffee, because aroma and flavour precursors are not lost during the process.

Table 32.2: The three decaffeination routes side by side

RouteWhat does the workKey conditions from the bookMain strengthMain weakness
SolventMethylene chloride, ethyl acetate, coffee oil or triglyceridesBeans wetted above 40% moisture; extraction 50 to 120 °C; 95 to 98% caffeine removedDirect, well proven, batch or continuousResidual solvent must be stripped; waste streams must stay very low in solvent
WaterCaffeine-free coffee extract at about 15% solidsBean residence about 8 hours; about 98% removed; beans leave at about 53% moistureNo solvent touches the beans; no pre-steaming or pre-wetting neededCaffeine must still be taken out of the extract, usually with solvent
Supercritical carbon dioxideCarbon dioxide at high pressure, caffeine caught on activated carbonBeans at 30 to 50% moisture; 40 to 80 °C; 120 to 180 bar; beans dried to about 10% waterInert and highly selective for caffeineCostly high pressure equipment; batch processing

Try it yourself: Compare the three routes as a buyer. If you want no solvent near the beans, water or carbon dioxide wins. If you want the cheapest proven plant, solvent wins. If you want the most selective removal with no chemical reaction, carbon dioxide wins. Every route trades one cost for another, and the book gives no single winner.

32.5 Decaffeinating roast coffee and extract, and refining the caffeine

Decaffeinating roast coffee extract has real attractions, mostly in processing and equipment costs, especially where roasted extract feeds instant coffee production. The catch is that the extract is already a finished product with good flavour and aroma, so every step must preserve those qualities. The usual first step is to strip the aroma and collect it by standard methods, as in Unit 31, section 31.4. The hard problem is the loss of trace flavour components to the contacting solvent. Two recent approaches in United States patents tackle it: use an inert, very specific solvent such as liquid or supercritical carbon dioxide to contact the extract, or recover the trace components by reflux and recycle them back to the extract, even keeping their level in the solvent high enough to stop extraction from the roasted extract. After decaffeination, processing matches undecaffeinated extract, with the stripped aromas added back.

The caffeine taken out is itself a valuable product, used as a food or drug additive. Its two main sources are coffee, as a by-product of decaffeination, and synthesis from urea and chloroacetic acid. Price follows mainly the cost of synthesis and demand. To refine coffee sourced caffeine economically, the feed to refining must be about 70% pure or better, and most decaffeination routes deliver caffeine at 70 to 85% purity.

Shuman's refining process, invented in 1948, combines purifying steps to remove coffee waxes and oils and caffeine-like water soluble materials that would darken the crystals: pure caffeine forms white, needle shaped crystals, while impure caffeine is dark and coarsely granular. The unrefined caffeine arrives as a water solution or a dried cake. Any residual solvent is steam distilled off, and the caffeine sludge is dissolved with recycled mother liquor from a later centrifugation. Activated carbon, recycled from a finishing step, removes most colour. The slurry is filtered, and the spent carbon is discarded, reactivated, or stripped with acetic acid to recover its caffeine. At this stage the caffeine is 80 to 90% pure. Crystallisation and centrifugation split the stream: crystals go for a second purification, and the mother liquor is contacted with pure solvent to recover essentially all its remaining caffeine, with aqueous alkali helping the separation. Fresh activated carbon is added to the redissolved crystals, and a second crystallisation and centrifugation gives a crystal cake that is air dried and broken to yield white United States Pharmacopeia grade caffeine.

Check your answer

Why strip the aroma before decaffeinating a roasted extract? Because the extract already holds the finished flavour, and the solvent would carry trace flavour compounds away with the caffeine. Stripping first protects them, and they are added back after decaffeination.

32.6 Brewing as solid-liquid extraction: three stages

Brewing is solid-liquid extraction, the same family of operation as factory extraction in Unit 31, but at home scale and in a single stage. Ground roast coffee meets a liquid, usually hot water, and soluble constituents pass into the liquid to form the brew. The book, following Clo's thesis summary, describes it in three stages.

Initial state. The soluble material is spread more or less evenly through the solid. Solvent soaks into the matrix and dissolves the soluble part. If this happens fast, a clear state exists at time zero: the solvent inside the solid is very concentrated, and the solvent outside is near its starting concentration.

Transfer period. Mass moves almost at once from the concentrated solution inside the solids towards the dilute solution at the surface. In fresh products such as fruit or vegetables, early changes to the product can shape the first part of the transfer, but the main transport is molecular diffusion of the solute in the solvent.

Final state. When concentrations inside and outside the solid match, equilibrium is reached. The enriched solvent is called the extract, and the insoluble part, depleted of solute but swollen with imbibed solvent, is called the raffinate. A ternary diagram, a triangle with insoluble fraction, soluble fraction and solvent at its three corners, can show the composition of a mixture at any instant as a single point inside the triangle.

That is the tidy theory. Practice adds grind, temperature, time, water chemistry and the machine itself, and the rest of this unit follows the book through those variables.

32.7 How people brew: manual methods, household machines and catering equipment

Manual techniques. Filtration, or percolation, puts ground coffee in a filter of paper, cloth, porcelain or another material and passes boiling water through it. Heiss and Radtke recommend wetting the bed first, and the book notes that this method depends more than others on each drinker's knack. In the United States this is the drip pot, in the United Kingdom the filter cone. Infusion puts ground coffee in boiling water in a vessel, leaves it 5 to 10 minutes, then decants through a screen or filter: the open pot in the United States, the jug in the United Kingdom. Boiling, or decoction, is the same but the coffee stays heated throughout, for a shorter time: the saucepan method in the United Kingdom. Turkish coffee is a special decoction: very finely ground coffee, mixed with powdered sugar, heated strongly in special equipment, with many recipes in use.

Household equipment. The book divides the many domestic brewers into three families. Filtration under gravity brings water to the boil, forces it by a slight excess pressure over ground coffee in a filter, and extracts continuously until the water is used up. Variants include filtration under pressure and filtration with recirculation, the electric or stove-top percolator. Continuous water circulation keeps the ground coffee suspended in extract while boiling water enters from a second chamber, and when vapour production stops and cooling begins, the relative vacuum pulls the brew through a filter: the vacuum maker in the United States, the Cona in the United Kingdom. Vaporisation under pressure suits highly roasted, finely ground coffee and makes a very concentrated drink, espresso and ristretto in Italian use.

Catering equipment. Urns work by gravity filtration for 60 or more cups in one percolation, holding the brew at drinking temperature in a double walled shell filled with heated water, with a tap at the base for near continuous supply. Airline equipment also filters, under a slight overpressure, making a litre and a half of brew every 3 minutes. Vending machines, developed over 30 years, serve a cup at any hour with milk and sugar as required. Many use instant coffee, but others use roast and ground coffee in single portions in a paper filter, and elaborate recent machines run the whole chain from grinding the beans to throwing out the grounds. Espresso machines, working by vaporisation under pressure, spread from Italy across Europe and then beyond, and multi-head versions serve coffee bars and restaurants.

Raw materials. Only two matter for the brew: ground roast coffee and the extracting water. Roast development is the main control on the sensory quality of the brew, so roaster type, temperature, time, pressure and water injection all need strict control, and fast roasting has changed much of what was known about roasting mechanisms. Water quality, especially hardness, matters for the organoleptic and visual properties of the brew, and the mineral salts and organic constituents of water have been studied in depth. The book also mentions pre-treatments of green beans such as dewaxing and gamma irradiation, studied as ways to improve the roast and the cup.

Three household brewing families: three alternatives, not three steps

Family 1: Filtration under gravity
Water drips through coffee in a filter, brew collects below; preparation can last up to about 15 minutes

Variants of this family: filtration under pressure, about 100 seconds; filtration with recirculation, brew circulated again and again for as long as the drinker leaves it on

Family 2: Continuous water circulation
Boiling water rises to the coffee, then vacuum pulls the brew down; contact time chosen by the drinker
Family 3: Vaporisation under pressure
Hot water and vapour push through fine, dark roast coffee in about 100 seconds

Figure 32.3: Three household brewing families. Filtration under pressure and filtration with recirculation are variants of gravity filtration, not families of their own. Drawn from Pictet, Coffee, Volume 2: Technology, Ch. 8.

32.8 What is in the cup: yield, composition and the senses

Extraction yield, or exhaustion, is the percentage of the solid matter of the roast coffee that passes into solution while the brew is made. It normally runs between 15 and 35%, depending on species and variety, roast level, grind and technique. The brew itself holds 1 to 3% soluble solids, and nearly 10% of the dissolved substances stay behind in the liquid held by the exhausted grounds. Most researchers using defined equipment report 9 to 28% yield as a function of grind, and Ara and Thaler report up to 40% from practical extraction near 100 °C. Part of the spread comes from the many different analytical methods used to measure dry matter, and no single reference method has won general use.

Vitzthum's average composition of a coffee brew, as a percentage of dry matter, is given in the table below. The book warns that these figures carry no statement of how far they move with brewing conditions.

Table 32.3: Chemical composition of a typical coffee brew, percent on a dry basis, after Vitzthum

ConstituentPercent of dry matter
Polysaccharides24.1
Chlorogenic acids15.0
Total ash14.1
Presumed proteins6.0
Caffeine4.8
Carboxylic acids3.0
Trigonelline1.6
Monosaccharides1 to 2

Individual compounds behave differently. Aroma compounds in the brew match those in the roast coffee in kind, but not in amount, and it is hard to tell whether the gap comes from selective dissolving or selective vaporisation. Streuli's view is that the smell of roast coffee comes above all from its lipid components and the aromatic principles tied to them, which explains why those principles transfer incompletely to the brew. The share of the bean's caffeine that reaches the cup depends markedly on how much coffee is used for the water: the more coffee per unit of water, the lower the caffeine content as a share of total dry matter. Trigonelline and its breakdown product nicotinic acid depend almost entirely on roast level, so measuring them in the brew can show how dark the roast was. Total ash, especially potassium, is often proposed as a reference for estimating yield, because it dissolves well and measures reliably.

Acidity is one of the few sensory qualities that can be tied to analysis with confidence. The pH of a brew is in practice not affected by extraction yield, because the other dissolved constituents buffer it strongly. Total acidity, the base needed to neutralise a given amount of extract, rises with longer brewing time and with higher brewing temperature. Chlorogenic acids dissolve readily, and Polish workers built a method on them to estimate brew strength and so the yield of the brewing just done. Free radicals in the brew, measured by magnetic resonance, rise with extraction temperature.

The senses are harder to pin down. The dominant sensory characteristics are bitterness, acidity and aroma in general, with astringency, caramel and phenolic notes as secondary characteristics. Van Roekel adds paper notes, harshness, rancidity, saltiness, and cereal and sweet almond notes. Acidity and astringency can be tied to cyclic and non-cyclic acids, but bitterness cannot be blamed on any single group of compounds. The methylxanthines, caffeine chief among them, seem less important than expected, because the complexes they form with phenolic acids are less active in taste than the free compound. Phenolic acids, almost entirely depsides of quinic acid, track astringency more than bitterness. Body depends on a combination of sensory characteristics, each strengthening the others.

Holding a brew hot also changes it: heating promotes vaporisation of aroma compounds and their chemical degradation, and changes the non-volatile compounds too, with chlorogenic acids mainly involved and a perceptible rise in pH. Several authors report a clear fall in quality, though Pangbourn confirmed only slight changes after 3 hours of heating. Reheating has been studied in full as well.

The Coffee Brewing Institute's brewing control chart ties the two numbers a brewer controls, formula and extraction, together. It plots formula in gallons per pound against extraction in ounces per pound. Two conversions from the book make it usable: an extraction of 3.50 ounces per pound equals 21.9%, and a formula of 2.5 United States gallons per pound equals a water to coffee ratio of 20.9, or 47.8 g of coffee per litre of water. The chart itself does not judge blend quality: it only shows whether the brew sits in the expected range of strength and extraction.

Check your answer

Two brews have different extraction yields but the same pH. Is that a mistake? No. The book states that brew pH is in practice not affected by extraction yield, because the dissolved constituents buffer it. Total acidity, not pH, is the measure that rises with brewing time and temperature.

32.9 Five brewers put to the test: Pictet's own research

The second half of Chapter 8 reports Pictet's own laboratory work, presented at the 8th ASIC Colloquium. It had two aims: to define, for commercial machines each standing for one technique, how the main working parameters shape the physicochemical and sensory properties of a home brew; and to find, with a panel of twelve expert tasters, which sensory characteristics dominate each taster's judgement, using a multi-dimensional mathematical analysis.

Five machines were chosen as the most representative of the techniques in section 32.7. Other machines were tested but gave less reproducible results and were left out. More than 40 brews were made on each type. The coffee was a good quality blend, mainly arabica from Central America and Africa, at a medium roast chosen to develop the best taste and aromatic qualities. Water to coffee ratio and grind were varied as well as the machine.

The five machines. Continuous water circulation, the Cona or vacuum maker, pushes boiling water up to the coffee in an upper vessel for a contact time the drinker chooses, then cooling creates a vacuum that pulls the brew down through a filter. Filtration by gravity, the automatic filter brewer, boils water, drives it above the filter, and lets it percolate through the coffee into a receiver below; preparation can last up to some 15 minutes. Filtration by gravity under pressure fixes the filter so extraction runs under a slight excess pressure, cutting total time to about 100 seconds. Filtration with recirculation, the electric percolator, circulates the liquid through the coffee again and again by a thermosyphonic system, for as long as the drinker leaves it on. Vaporisation under pressure, the household espresso type, pushes water slightly above its boiling point through the coffee bed, cutting brewing to about 100 seconds at maximum liquid volume.

Soluble solids and yield. Soluble solids concentration falls steadily as the water to coffee ratio rises, for every machine, though the curves differ in slope. Two machines cannot go below a certain water volume: continuous water circulation and filtration by gravity. Gravity filtration gives the most concentrated extracts, through its long brewing time, and vaporisation under pressure matches it, through optimum contact between coffee and water. Filtration under pressure gives relatively dilute extracts, mainly because its contact time is very short, possibly too short. For the two machines whose time the drinker controls, the trial time of about 6 minutes at the highest water amount was too short to get a concentrated extract. Yield follows the same pattern. Time of processing matters most: the gap between slow gravity filtration and fast filtration under pressure is small at low water amounts and grows as water rises. Grind was varied from 650 to 1200 µm average particle diameter. Its effect is strong where brewing is long but contact is short, as in gravity filtration, and weak where contact is repeated, as in recirculation, where yield barely changes with grind at any water ratio. In short, fine grinding helps most when the water passes through only once.

Water retention by the grounds depends mainly on the coffee itself, its surface area and pore size, and very little on the machine: about two parts of water to one part of coffee for a coarse grind. As a share of the water used, retained water falls as the water amount rises.

Aromatic content. Volatile content was measured by evaporating each brew and reading the ultraviolet absorption of the condensed volatiles: an indicative measure only, of relative aromatic richness as a share of the starting material's aroma. Total transfer is never reached, because much aroma stays in the spent grounds, in the lipid fraction, or in the extract the grounds absorb. The two methods with long contact between coffee and liquid, continuous water circulation and filtration with recirculation, give brews particularly poor in aromatic principles, and the book blames vaporisation loss. Grind acts the same way it acts on solids: its effect on aroma depends on the machine, and recirculation loses volatiles at every grind.

Individual constituents. Brews were freeze dried and analysed for their main constituents. Easily extracted substances fall slightly in share as yield rises: total reducing sugars, liberated by acid hydrolysis, and total ash, though ash only for a fine grind. The rest rise moderately with yield, behaving as less soluble substances: presumed proteins, chlorogenic acids, caffeine and direct reducing sugars. Coarse grinding changes little, apart from the ash behaviour already noted, though reducing sugars fall perceptibly and protein rises. Standard deviations sometimes reach plus or minus 5%, so the book urges reading the general trend, not the fine detail.

What the tasters found. Each brew was scored on 14 sensory criteria: five for aroma, intensity, balance, freshness, heaviness and woodiness, and nine for taste, body, balance, sharpness, bitterness, acidity, astringency, hardness, cereal note and earthy note. Scores were ratioed against a reference brew to make relative taste profiles. At low yield, from a small water volume, differences from the reference are mostly negative, except the cereal note, which consumers generally like little. As water volume and yield rise, the differences shrink, but preference does not reverse until the maximum usable water volume is reached, when balance and freshness of aroma and natural acidity improve. The cereal note persists, typical of filtration under pressure.

Canonical analysis projects the 14 attributes onto two principal axes carrying nearly 80% of the information: the first axis alone carries 61.6%, the second 17.9%. The first axis expresses the fullness of the coffee's natural characteristics, fed most by aroma intensity and balance and by taste body and balance; the cereal note points the opposite way. The second axis is bipolar: freshness and fineness on the positive side, bitterness, acidity and harshness, generally from a higher roast, on the negative side. Aroma intensity tracks taste body, aroma freshness tracks acidity, and aroma heaviness groups with taste astringency, bitterness, sharpness and harshness. Stepwise discriminant analysis cuts the 14 attributes to five that together describe each brew almost completely, in order: taste body or fullness, taste balance, natural acidity, aroma intensity, and aroma balance. Of these, acidity gives the most basic discrimination on its own.

For a given technique, raising the water volume, and so the yield, moves the brew along the positive direction of the first axis, strengthening fullness and balance, and usually along the positive direction of the second axis too, improving quality and freshness. Making the grind finer moves the brew in the negative direction of the second axis, a perceptible loss of those two qualities. The machines rank clearly in the panel's judgement. Continuous water circulation gives brews of the highest quality and freshness, but oddly lacking in the full body the general quality of a brew needs. Vaporisation under pressure gives brews strongly marked by the coffee's innate character, with bitterness and harshness dominant, but fineness and freshness absent. Filtration under pressure and filtration with recirculation give medium quality brews with no dominant character, clearly below the others. Filtration by gravity gives the most balanced extracts, joining natural acidity and bitterness with aromatic intensity and fullness of taste.

Table 32.4: The five brewing techniques compared in Pictet's experiments

TechniqueHow it worksTimeWhat the panel found
Continuous water circulation, Cona or vacuum makerBoiling water rises to the coffee, then vacuum pulls the brew downChosen by the drinkerHighest quality and freshness, but lacking body
Filtration by gravity, automatic filter brewerWater drips through coffee in a filterUp to about 15 minutesMost balanced, with acidity, bitterness, aroma intensity and fullness
Filtration by gravity under pressureFilter fixed so extraction runs under slight excess pressureAbout 100 secondsMedium quality, no dominant character; dilute extracts
Filtration with recirculation, percolatorBrew circulates through the coffee again and againChosen by the drinkerMedium quality; poor in aromatics through vaporisation loss
Vaporisation under pressure, household espressoWater above boiling point pushed through fine, dark roast coffeeAbout 100 secondsStrong innate character, bitterness and harshness dominant, fineness and freshness absent

What Pictet's five brewers taught: four findings, side by side

Finding 1: More water, higher yield
Fuller and more balanced brew
Finding 2: Finer grind
Higher yield in a filter brewer, but less freshness
Finding 3: Long contact, recirculation
Aroma lost by vaporisation
Finding 4: Gravity filtration
The most balanced cup of the five

Figure 32.4: What Pictet's five brewers taught: water, grind and contact time pull the cup in different directions. Drawn from Pictet, Coffee, Volume 2: Technology, Ch. 8.

Check your answer

Which brewer gave the best cup in Pictet's panel tests, and why? Filtration by gravity, the automatic filter brewer. Its long brewing time and single pass of water gave the most concentrated extracts and the most balanced cup, joining natural acidity and bitterness with aromatic intensity and fullness of taste. The fastest brewers were too short in contact, and the recirculating brewers lost aroma to vaporisation.

Try it yourself: Brew the same coffee twice in a filter brewer, once with your normal water amount and once with a third more water, keeping the coffee weight the same. Taste both side by side and note which is fuller and which is fresher. Then brew the same coffee in a percolator left running for ten minutes and compare its smell with the filter cup. You are repeating Pictet's two clearest findings in your own kitchen.

32.10 What the book leaves open

Pictet ends with a candid summary. The review of the literature shows the physics of brewing is well established: the stages of solid-liquid extraction, the passage of solutes from solid to liquid and their diffusion outwards, are known from experiment, and mathematical models match practical trials well. What is missing is the detail for individual constituents: caffeine can be followed with certainty, but for complex substances such as carbohydrates and non-caffeinic nitrogenous compounds, the effects of dissolving cannot be separated from the changes caused by the extraction conditions, especially temperature. That field, the book says, has been insufficiently explored and should be studied in the years to come.

Sensory evaluation has become relatively easy for a defined, reproducible brew: trained tasters can score aroma and taste qualities against a reference or on their own. The hard part is tying a sensory attribute to one chemical constituent or a small group of them. Some simple links between aromatic notes and defined volatile compounds are already noted, but fundamental research over several years is still needed before organoleptic properties can be matched to chemical composition beyond doubt.

Practical note: For a home brewer, the book's findings give a short working list. Wet the bed before the main pour. Use the water to coffee ratio on the brewing control chart as a starting point, 47.8 g of coffee per litre of water at a formula of 2.5 gallons per pound. Grind finer when the water passes through once, and do not expect finer grinding to help much in a percolator. Do not leave a brew circulating or held hot for long: aroma leaves by vaporisation, and total acidity climbs with time and temperature. If the cup tastes thin, add water to raise yield, not heat.

Key ideas to carry forward

  • Arabica holds about 1% caffeine and robusta about 2%; a regular cup averages 85 mg, with a range of 50 to 150 mg, and instant holds about half.
  • Decaffeination is done on green beans before roasting, so roasting can still build flavour and aroma afterwards.
  • Solvent decaffeination needs beans wetted above 40% moisture, works at 50 to 120 °C, removes 95 to 98% of caffeine, and the solvent must be stripped and recycled with waste streams kept very low in solvent.
  • Water decaffeination washes beans for about 8 hours with caffeine-free extract at about 15% solids, removes about 98% of caffeine, and leaves beans at about 53% moisture.
  • Supercritical carbon dioxide works at 40 to 80 °C and 120 to 180 bar on beans at 30 to 50% moisture, catches caffeine on activated carbon, and dries beans to about 10% water; it is inert and selective but needs costly batch plant.
  • Caffeine for refining must be about 70% pure or better; Shuman's 1948 process brings it to 80 to 90% after carbon treatment and ends with white, needle shaped United States Pharmacopeia grade crystals.
  • Brewing is solid-liquid extraction in three stages, initial, transfer and final, ending at equilibrium between extract and raffinate.
  • Home brew yields run 15 to 35%, with brews at 1 to 3% solids, and Vitzthum's brew is 24.1% polysaccharides, 15.0% chlorogenic acids and 14.1% ash on a dry basis.
  • Brew pH is buffered and does not move with yield, while total acidity rises with brewing time and temperature.
  • In Pictet's tests on five machines with twelve tasters, gravity filtration gave the most balanced cup, continuous circulation the freshest but thinnest, espresso the strongest but harshest, and pressure filtration and recirculation the middling cups.
  • Five sensory attributes describe a brew almost completely: taste body, taste balance, natural acidity, aroma intensity and aroma balance, with acidity discriminating most.

Glossary

Decaffeination

Removing caffeine from coffee, usually from green beans before roasting, by solvent, water or supercritical carbon dioxide. Food tech example: solvent extraction removing 95 to 98% of the caffeine from pre-wetted green beans.

Methylene chloride

Dichloromethane, the chlorinated solvent used commercially for decaffeination, chosen for high caffeine solubility and a low boiling point. Food tech example: extraction at 150 °F for 10 hours with 4 lb of solvent per pound of beans.

Water decaffeination

Decaffeination with water that already holds the bean's non-caffeine soluble solids, so only caffeine is taken out. Food tech example: beans extracted for about 8 hours by caffeine-free extract at about 15% solids.

Supercritical carbon dioxide

Carbon dioxide compressed to pressures where it dissolves caffeine like a solvent, used here at 120 to 180 bar and 40 to 80 °C. Food tech example: caffeine caught on activated carbon while carbon dioxide circulates in a closed cycle.

Caffeine refining

Purifying crude caffeine from decaffeination to white, needle shaped crystals of United States Pharmacopeia grade. Food tech example: Shuman's process using activated carbon and two crystallisations to reach 80 to 90% purity after carbon treatment.

Solid-liquid extraction

Passing a liquid through a solid so soluble constituents move into the liquid, in three stages: initial, transfer and final. Food tech example: hot water passing through ground roast coffee to make a brew.

Extract

The enriched liquid that leaves a solid-liquid extraction, holding the dissolved solids. Food tech example: a coffee brew holding 1 to 3% soluble solids.

Raffinate

The insoluble solid left after extraction, depleted of solute but swollen with solvent. Food tech example: spent coffee grounds holding about two parts of water to one part of coffee after brewing.

Extraction yield

The percentage of the solid matter of the roast coffee that passes into solution during brewing. Food tech example: a home brew yield of 15 to 35%, with most studies reporting 9 to 28%.

Filtration under gravity

Brewing by letting hot water drip through ground coffee in a filter under its own weight. Food tech example: the automatic filter brewer that gave the most balanced cup in Pictet's tests.

Continuous water circulation

Brewing in which boiling water rises to the coffee and the brew is then pulled down by vacuum, as in the Cona or vacuum maker. Food tech example: the freshest but thinnest brews in Pictet's tests.

Vaporisation under pressure

Brewing by pushing water above its boiling point through fine, dark roast coffee under slight overpressure, as in household espresso. Food tech example: brewing in about 100 seconds with bitterness and harshness dominant.

Brewing control chart

A chart relating formula, in gallons per pound, to extraction, in ounces per pound, for judging brew strength and yield. Food tech example: 3.50 ounces per pound equal to 21.9% extraction, and 2.5 gallons per pound equal to 47.8 g of coffee per litre.

Body

The fullness of a coffee brew in the mouth, one of the five sensory attributes that together describe a brew almost completely. Food tech example: gravity filtration giving full body where continuous circulation gives thin body.

References

  • Clarke, R. J. and Macrae, R. (eds). Coffee, Volume 2: Technology. Elsevier Applied Science, London, 1987. Chapter 3, Decaffeination of Coffee; Chapter 8, Home and Catering Brewing of Coffee.
  • Katz, S. N. Decaffeination of Coffee. In Clarke, R. J. and Macrae, R. (eds), Coffee, Volume 2: Technology, Chapter 3.
  • Pictet, G. Home and Catering Brewing of Coffee. In Clarke, R. J. and Macrae, R. (eds), Coffee, Volume 2: Technology, Chapter 8.