GIFSQ ID: FTN-0103
Food Fraud and Emerging Contaminants
Unit 1: What Food Fraud Is, and How It Works
Note
Unit 1: What Food Fraud Is, and How It Works
A question to start with
A shop sells a jar labelled pure honey for a high price. A second jar, half the price, carries the same label. Both look golden and thick, both pour slowly, both taste sweet. One of them is mostly honey from bees. The other is mostly syrup made from plants, with a little honey stirred in for flavour. Your eyes and your tongue cannot tell them apart with confidence, and the label on the cheap jar is lying. How would you prove it, and who is harmed when the lie is never found? This unit is about that lie: what food fraud is, the forms it takes, why people do it, and why it matters to everyone who eats, cooks, buys, or makes food.
Food fraud is the deliberate act of deceiving buyers about a food, for money. A person or a business changes a food, or changes what is said about it, so that it looks worth more than it is, or so that a cheaper food can be sold at a higher price. The key word is deliberate. A mistake in a factory, a misprinted label, or a batch that went wrong by accident is not fraud. Fraud is planned. Someone knows the food is not what the label claims, and sells it anyway.
The harm is wider than a lost few coins. Fraud steals money from buyers and from honest makers who cannot compete with a fake sold at a low price. It can also harm health, when a cheap substitute brings in something a person is allergic to, or when a food is diluted so much that it no longer feeds people the way they expect. And it damages trust: once buyers learn that olive oil, honey, or fish can be faked, they trust every jar and every fillet a little less.
The four main forms of fraud
Food fraud usually takes one of four forms, and it helps to learn them by name, because each form leaves different clues and needs a different test to catch it.
Substitution is the first form. A cheaper food is swapped in for a dearer one, in whole or in part. Horse meat sold as beef, a cheap white fish sold as cod, or a low grade oil sold as extra virgin olive oil are all substitution. The food looks right and often tastes close enough that nobody notices at the table. Substitution is the most common form of fraud in high value foods, because the gap between the price of the real food and the price of the fake is where the profit sits.
Dilution is the second form. A food is thinned with a cheaper liquid or filler, so that a smaller amount of the real food stretches further. Water added to milk, syrup added to honey, or cheap juice added to a named fruit juice are dilution. The food still contains some of the real thing, which makes dilution harder to spot than a full swap. A diluted food also feeds people less well: milk with added water has less protein and less calcium per glass, and a child who drinks it every day gets less nourishment than the label promises.
Misrepresentation is the third form. Nothing is added to the food and nothing is swapped, but the words about it are false. A fish caught in one sea is labelled as coming from another. A product made in a factory is labelled as farm made. A food grown with ordinary methods is labelled organic, or a food is given a false place name, so that buyers pay for a story that never happened. Misrepresentation is fraud of words, not of substance, and it is caught by paper trails and by tests that can show where a food really came from.
Counterfeiting is the fourth form. The packaging itself is copied: the brand name, the logo, the bottle, the seal. A fake bottle of a famous drink, or a fake pack of a well known spice brand, is sold as if it came from the real maker. Counterfeit food often comes with the other three forms as well, because the liquid or powder inside the copied pack is rarely the real product.
Questions to think about
- A spice seller mixes ground brick dust into chilli powder to add weight. Which form of fraud is that, and who is harmed first?
- A bottle says cold pressed olive oil, but the oil inside was pressed with heat and mixed with a cheaper oil. How many forms of fraud are in that one bottle?
- Why is dilution harder for a buyer to notice than a full substitution?
Think about the chilli powder question before you read on
Mixing a filler into a spice to add weight is dilution, and it can also be seen as adulteration, the general word for adding something foreign to a food. The buyer is harmed first, by paying spice prices for dust, and by eating something that was never meant to be food. If the filler is not a food at all, the health harm can be worse than the money harm, because brick dust carries minerals and grit that the body cannot use and that can damage teeth and the gut. The honest spice seller is harmed next, because the fraudster can undercut the true price.
Why fraud happens: the money gap
Fraud happens where three things meet: a food that buyers will pay a lot for, a cheaper substitute that is hard to tell apart, and a supply chain long enough that nobody checks every step. The profit comes from the gap between the two prices. If real saffron sells for many times the price of dyed corn silk, a fraudster who swaps a little of one for the other makes money on every gram. If the fake is never tested, the profit repeats on every sale.
High value foods are the usual targets. Olive oil, honey, saffron, vanilla, fish, meat, spices, fruit juice, coffee, and milk powder appear in fraud reports again and again around the world. These foods share three traits: they cost a lot per kilogram, they are sold in forms where the buyer cannot see the original plant or animal, and their quality is judged by taste, colour, or smell, which are easy to imitate and hard to measure at home.
Long supply chains give fraud room to hide. A fish may pass through a boat, a dock, a processor, a wholesaler, and a shop before it reaches a plate. At each step, someone handles the paperwork and the product, and a swap at any one step can travel onward under the original name. A short chain, where the buyer knows the maker and sees the food made, leaves far less room for fraud, which is one reason local and direct buying is often safer.
Price pressure also plays a part. When the price of a real food rises after a bad harvest, or when a buyer demands a price no honest maker can meet, the temptation grows. A buyer who insists on paying less than the cost of the real thing is, without knowing it, inviting a fake. Honest businesses feel this pressure too, and some give in. That is why fraud is not only a problem of criminals: it is a problem of markets that reward the cheapest price and never ask how that price was reached.
Adulteration, and the line between fraud and danger
Adulteration is the word used when something foreign is added to a food, or when a part of the food is removed. All dilution is adulteration, and some substitution is too. The word matters because food law in many places uses it, and because it covers the cases where the added thing is not a food at all.
Some adulterants are harmless fillers that only cheat the buyer out of money: starch added to a spice, or water added to juice. Others are dangerous. A dye added to make a spice look brighter can be a chemical never meant to be eaten. A cheap industrial chemical added to raise a protein test reading can poison people. The most harmful frauds in history have been of this kind, where the adulterant was chosen because it fooled a test, not because it was safe to eat.
This gives a useful line to remember: fraud is always a money crime, and sometimes it is also a safety crime. A fraud that only takes money is still serious, because it steals from buyers and honest makers. A fraud that adds a harmful substance is serious in a second way, because people can be hurt or killed by food they trusted. Food safety work treats both as matters to prevent, but the dangerous ones get the fastest response.
Who is harmed, and how
The first person harmed is the buyer, who pays for something they did not receive. The loss on one jar of honey is small, but across millions of jars the stolen money is large, and it is taken from ordinary shoppers.
The second person harmed is the honest maker. A beekeeper who keeps real bees, a farmer who grows real saffron, or a fisher who catches real cod cannot sell at the fraudster's price, because the real food costs more to produce. When fakes flood a market, honest makers lose sales, and some stop making the real food at all. The market then has less of the real thing and more of the fake, which is the opposite of what buyers wanted.
The third harm is to health. A person allergic to a nut can be harmed when a cheap nut is swapped into a food that claims to be nut free. A person who relies on milk for protein can be underfed when the milk is diluted. A baby fed on diluted formula can be harmed in ways that last. These harms are quiet and hard to trace back to one jar or one meal, which is why fraud is so hard to punish after the event.
The fourth harm is to trust. Food is one of the few things people put into their bodies every day without testing it first. That act rests on trust in labels, in brands, and in the people who sell food. Each fraud that comes to light weakens that trust a little, and buyers start to doubt foods that are honest too. Rebuilding trust takes far longer than breaking it.
The diagram shows the fraud profit chain. The steps point to the diagram: break any one link, by testing, by shortening the chain, or by refusing a price that cannot be honest, and the fraud stops paying.
Spotting fraud as a buyer
No buyer can test every food in a laboratory, but buyers can use several plain checks that catch many frauds. The first is price. A price far below the usual price for a dear food is a warning, not a bargain. Real saffron, real vanilla, and real extra virgin olive oil cost a certain amount to make, and no seller can sell them far below that cost and stay honest. If the price looks too good, ask how it is possible.
The second check is the label. A real label names the food, its origin, its ingredients, and the maker, in words that can be checked. A vague label, a label with no maker address, a label that avoids naming the exact fish or the exact oil, or a label with spelling mistakes, all deserve doubt. A label that claims a famous place name but gives no detail about the producer is a common sign of misrepresentation.
The third check is the seller. Buy from sellers who can say where the food came from, who made it, and who will answer if something is wrong. A seller who cannot answer those questions, or who changes the story, is a risk. Markets, shops, and online sellers all vary, and the ones who know their supply chain are the ones to trust.
The fourth check is the food itself, by the senses. Real honey thickens and crystallises over time; a honey that never crystallises and always pours like syrup deserves a question. Real saffron turns water a slow golden yellow and keeps its red threads; a fake colours the water at once and the threads fall apart. These sense checks are not proof, but they are good reasons to ask more questions or to buy from another seller.
Practical note: In a teaching kitchen or a market visit, compare two jars of honey, two bottles of oil, or two packs of a spice at very different prices. Read both labels line by line, note the origin, the maker, and the ingredients, and taste both if the setting allows it. Write down every difference you can find before you decide which one you trust, and keep the labels: a label is evidence, and a fraud case often starts with someone keeping a label that did not match the food inside.
Quick self-check
Question: A shop sells vanilla extract at one tenth of the price in every other shop. What is the most likely explanation?
Either the extract is diluted, or the vanilla flavour comes from a cheaper source while the label claims real vanilla. A price far below the cost of making the real food is a warning sign, and the buyer should ask the seller for the origin and the ingredients before buying.
Question: Why is misrepresentation fraud even when nothing is added to the food?
Because the buyer pays for a story as well as a substance: an origin, a method, or a place name. If the story is false, the buyer paid for something they did not receive, even though the food itself is edible and unchanged.
A worked example: counting the fraudster's profit
Fraud is a money crime, so the best way to see why it happens is to count the money. Take a simple example with round numbers chosen for practice, not real market prices. A seller buys a cheap syrup for 1 money unit per kilogram and real honey for 5 money units per kilogram. The seller mixes 1 kg of syrup into 4 kg of real honey and sells the 5 kg mixture as pure honey at 5 money units per kilogram.
Step one: work out what the mixture cost. The cost is 4 kg of honey at 5 units, which is 20 units, plus 1 kg of syrup at 1 unit, which is 1 unit. The total cost is 21 units for 5 kg.
Step two: work out what the mixture sells for. The seller charges 5 units per kilogram for 5 kg, so the income is 25 units.
Step three: work out the profit from the fraud. The income of 25 units minus the cost of 21 units is 4 units of extra profit on 5 kg, which is 0.8 units per kilogram.
Step four: scale it up. If the seller moves 1,000 kg a week, the extra profit is 800 units a week, or more than 40,000 units in a year, from one small trick that no buyer can taste. That is the money gap at work: a small difference per kilogram, repeated over a large volume, becomes a large reward, and the risk of being caught is the only thing standing against it.
The same arithmetic explains why testing matters. If testing catches the fraud once, the seller loses the batch, the customer, and the reputation, and the fine may be larger than a year of profit. A buyer who tests, and a market that punishes fraud, changes the arithmetic so that honesty pays better than cheating.
How fraud differs from a mistake
A factory can mislabel a batch by accident, a printer can swap two labels, or a new worker can add the wrong ingredient. These are mistakes, and they are dealt with by recalls, corrections, and better checks. Fraud is different in intent: someone meant to deceive, and someone gained money from the deception. Intent is hard to prove, which is why fraud cases often turn on patterns: the same wrong label appearing again and again, the same cheap substitute turning up in many batches, or records that were changed after the event.
This difference matters for food students because the fixes are different. A mistake is fixed by better training, clearer labels, and stronger checks. Fraud is fixed by those things too, but also by testing that looks for deception, by checking suppliers, by auditing records, and by making the penalty larger than the profit. A food business that only guards against mistakes will miss fraud, because fraud is designed to look like a normal product passing normal checks.
A useful test of intent is to ask who benefits and whether the error always runs in one direction. A random mistake sometimes makes a food better and sometimes worse, and it happens with different foods. A fraud always makes the food cheaper to produce and always benefits the seller, and it repeats with the same food and the same trick. When the error always points toward profit, suspicion is justified and testing should follow.
The shape of the whole subject
This subject follows the fraud from its roots to its defence. The next unit looks at famous cases and what each one taught the food world. Then comes detection: the laboratory methods that can tell a real food from a fake, from DNA to spectroscopy. After that, the subject turns to the contaminants that now sit beside fraud in the news: PFAS, the long lasting chemicals found in water, packaging, and food; microplastics and nanoplastics; and the older dangers of mycotoxins and heavy metals. The final unit builds a fraud defence: how a business finds its weak points, tests for them, and keeps its food honest. Each unit stands alone, but together they give one picture: fraud is a human act with a money motive, and it is beaten by knowledge, testing, and honest supply chains.
Try it yourself
Pick one high value food from your kitchen or a local shop: honey, olive oil, a spice, coffee, or fruit juice. Write a one page fraud profile for it: what is the real food, what would a fraudster swap in or dilute it with, what would the profit be, how could a buyer spot it, and what laboratory test would prove it. Use the four forms from this unit to name the fraud, and end with three questions you would ask the seller before you paid a high price. Compare your profile with a classmate who picked a different food: which food is easier to fake, and why?
Key ideas to carry forward
- Food fraud is deliberate deception about a food, done for money, and it is different from an accident or a mistake.
- The four main forms are substitution, dilution, misrepresentation, and counterfeiting.
- Fraud grows where a food is dear, a look-alike is cheap, and the supply chain is long and little checked.
- The harms are money lost by buyers, sales lost by honest makers, risks to health, and trust lost by the whole food system.
- Adulteration is adding something foreign to a food, and it can be a money crime, a safety crime, or both.
- A price far below the cost of the real food, a vague label, and a seller who cannot name the source are all warning signs.
- Fraud that always benefits the seller and repeats the same trick is different from a random mistake.
Glossary
Money gap
The difference between the price of the real food and the price of the cheap substitute, which is the fraudster's profit. Food tech example: the money gap between real honey and cheap syrup is what pays for honey fraud.
Profit motive
The money a person gains from an act, which in fraud is the reason the act happens. Food tech example: the profit motive for diluting milk is the money saved on every litre sold.
Adulteration
Adding something foreign to a food, or taking a valuable part out of it, so that the food is no longer what it claims to be. Food tech example: starch mixed into ground spice to add weight is adulteration.
Counterfeiting
Copying the packaging, brand, or seal of a real product so that a fake can be sold as the real thing. Food tech example: a copied bottle and label of a famous olive oil brand, filled with cheap oil, is counterfeiting.
Dilution
Thinning a food with a cheaper liquid or filler so that less of the real food is used. Food tech example: water added to milk, or syrup added to honey, is dilution.
Food fraud
Deliberate deception about a food, done for money, by changing the food or what is said about it. Food tech example: selling a cheap white fish under the name of a dearer fish is food fraud.
Misrepresentation
Saying false things about a food, such as a wrong origin, method, or place name, without changing the food itself. Food tech example: labelling ordinary farmed fish as wild caught is misrepresentation.
Substitution
Replacing a food, in whole or in part, with a cheaper look-alike. Food tech example: selling one kind of fish under the name of a dearer kind is substitution.
Supply chain
The series of steps a food passes through from farm or sea to the buyer: growing, catching, processing, packing, transport, and sale. Food tech example: a fish supply chain runs from the boat to the dock, the processor, the wholesaler, and the shop.
Traceability
The ability to follow a food back through its supply chain to its origin, using records at each step. Food tech example: a batch number on a jar of honey lets a buyer trace it back to the beekeeper.
High value food
A food that sells for a high price per kilogram, such as saffron, vanilla, honey, or olive oil. Food tech example: saffron is a high value food, so it is a common target for fraud.
Allergen
A substance in food that can cause an allergic reaction in some people, such as peanut, milk, egg, or certain nuts. Food tech example: a swapped nut that is not named on the label can harm a person with a nut allergy.
Composite sample
A sample made by mixing many small portions taken from different parts of a lot. Food tech example: scoops of spice from ten sacks are mixed to make one composite sample for testing.
Adulterant
The foreign substance added to a food in adulteration. Food tech example: a bright dye added to chilli powder to make it look stronger is an adulterant.
Label claim
A statement on a food label about what the food is, where it comes from, or how it was made. Food tech example: "cold pressed" on an olive oil bottle is a label claim that can be tested.
Origin
The place where a food was grown, caught, or made. Food tech example: the origin of a cheese may be the region and the farm where the milk came from.
References
- Spink, J. and Moyer, D. C., Defining the Public Health Threat of Food Fraud, 2011.
- Johnson, R., Food Fraud and Economically Motivated Adulteration of Food and Food Ingredients, 2014.
- Moore, J. C., Spink, J. and Lipp, M., Development and Application of a Database of Food Ingredient Fraud and Economically Motivated Adulteration, 2012.
- Everstine, K., Spink, J. and Kennedy, S., Economically Motivated Adulteration of Food and Food Ingredients, 2013.
- Manning, L., Food Fraud: Policy and Management, 2016.
Unit 2: Famous Fraud Cases and What They Taught
Note
Unit 2: Famous Fraud Cases and What They Taught
A question to start with
A mother buys a tin of milk powder for her baby. The label promises protein, and the baby drinks it every day. The powder looks right and dissolves right, and a simple protein test says the protein level is fine. Months later, the baby is ill, and doctors find the kidneys damaged. The powder was never short of protein on paper: a chemical had been added that fooled the protein test, while the real protein was missing. How could a test be fooled, and what does that case teach every food laboratory in the world? This unit looks at famous fraud cases, told in general terms, and pulls out the lesson each one left behind.
Famous cases matter because they changed the rules. Each one exposed a weak point that honest people had not seen: a test that could be fooled, a supply chain nobody checked, a label nobody verified. After each case, new tests were built, new checks were written into law, and businesses changed how they buy. Students who learn these cases learn where fraud hides, and that is more useful than memorising a list of names.
A note on how this unit is written: cases are described in general terms, what happened and what it taught, without inventing dates, amounts, or company names that cannot be verified. The lessons are the point, not the numbers.
The milk powder case: a test that could be fooled
The most taught fraud case of recent times concerns milk powder for babies, adulterated with a nitrogen rich industrial chemical. The standard protein test does not measure protein directly: it measures nitrogen, and then works out the protein from the nitrogen, because protein is the main nitrogen containing part of milk. The chemical added to the powder was rich in nitrogen but was not a protein, so the test reported a high protein value while the powder itself was low in real protein and carried a poison.
Babies fed on the powder were undernourished and many suffered kidney damage, because the chemical formed crystals in the kidneys. The fraud worked for a while because the test everyone trusted was the wrong test for this trick: it measured a stand in for protein, not protein itself.
The lesson changed food analysis around the world. First, a single test is never enough when money is at stake: a result must be checked by a second method that measures something different. Second, tests must be chosen with fraud in mind, not only with honest mistakes in mind. A method that works well for honest milk can fail badly when someone is trying to fool it. Third, the most vulnerable eaters need the strongest checks: foods for babies, for ill people, and for people who cannot choose for themselves deserve the tightest testing. After this case, laboratories added direct tests for the chemical itself, and buyers began to test milk powder for more than nitrogen.
The horse meat case: one meat sold as another
In another widely reported case, ready meals labelled as beef were found to contain horse meat. The meals were sold across several countries, through long supply chains, and the substitution was found only when DNA testing was used on the meat. The meat was not poisonous, and nobody was harmed by eating it, but the fraud was large, it crossed borders, and it showed that the paperwork said one thing while the food said another.
The lesson was about supply chains and paperwork. A ready meal maker may buy meat from a supplier, who bought it from a trader, who bought it from a processor, and at each step the paper says beef. If nobody tests the meat itself, a swap at any step travels all the way to the shop shelf. After this case, many food businesses started testing incoming meat with DNA methods, even when the supplier was trusted, and started to ask for shorter chains and for proof that matched the product, not only the paper.
The case also taught a lesson about trust and culture. Many buyers do not eat horse meat for reasons of culture, religion, or personal choice, and they were angry not because the meat was unsafe but because they had been misled about what they were eating. Fraud is not only about safety or money: it is about the right to know what is in your food and to choose what you eat.
The olive oil case: the oldest fraud still running
Olive oil fraud is one of the oldest frauds in the food trade, and it is still running today. Extra virgin olive oil sells at a high price, and cheaper oils, such as refined olive oil, or oils from other seeds, can be blended in without changing the look of the oil much. The blend is then sold under the extra virgin name, at the extra virgin price. Some frauds also misstate the origin, so that oil from one place is sold as oil from a famous region.
The lesson of olive oil is that fraud follows value and survives because the fake is hard to taste. Most buyers cannot tell a blend from a pure oil by taste alone, and the tests that can tell them apart need a laboratory. After many olive oil cases, testing grew sharper: chemists learned to read the fatty acid pattern of an oil like a fingerprint, and to test for markers that show whether an oil was refined with heat. The case also taught buyers to look for traceability, because an oil that can be traced back to a named grove and a named press is harder to fake than an oil that arrives with vague papers.
Olive oil also teaches that fraud can be a slow drain rather than a single event. A small amount of cheap oil in every bottle, over years, steals a large sum, and honest olive growers lose sales year after year. Stopping it needs steady testing, not a single campaign.
The honey case: syrup in the jar
Honey is a common fraud target because real honey is dear and syrups are cheap. The usual trick is to dilute honey with syrups made from plants, or to feed bees with sugar syrup so that the honey they make is thin and weak. The result still looks like honey and tastes sweet, and a simple sugar test may not catch it, because the sugars in the syrups are close to the sugars in honey.
The lesson of honey is that a test must match the trick. Early honey tests measured total sugar, which the fraud passed easily. Later tests used isotope analysis, which reads the tiny differences in the carbon atoms of the sugars, and can tell whether a sugar came from a bee working on flowers or from a factory making syrup from a crop plant. Honey also taught that fraud adapts: as one test improves, fraudsters try a new syrup that passes it, and laboratories must keep updating their methods. It is a race, not a single fix.
Honey fraud also harms beekeepers directly. A beekeeper with real hives spends a year tending bees and cannot sell honey at the price of syrup. When fake honey fills the shops, real beekeepers lose their living, and some stop keeping bees, which harms farming as well, because bees pollinate crops. The harm of a fake jar spreads far beyond the shop.
The fish case: the wrong name on the fillet
Fish fraud is common because a fillet has no head, no skin, and no shape to tell the buyer which fish it was. A cheap white fish can be sold as a dearer one, and a farmed fish can be sold as wild caught. The swap is often found by DNA testing, which can name the species from a small piece of flesh, and by checking the paperwork against the catch.
The lesson of fish is that the form of the food decides how easy fraud is. A whole fish is hard to fake, because the buyer can see it. A fillet, a fish cake, or a portion in a ready meal is easy to fake, because the buyer sees only white flesh. The more a food is processed and the further it travels from its original form, the easier it is to misname. This rule applies far beyond fish: ground spices, powders, oils, and juices are all easier to fake than the whole food they came from.
Fish fraud also teaches that origin claims need proof. A claim of wild caught or of a named sea is a claim about a journey, and it can only be checked by records and by tests that can show where a fish lived, such as isotope analysis of its tissues. A label alone is not proof.
The spice case: colour, weight, and danger
Spices have been faked for centuries, because they are dear, light, and sold as powders. The tricks include adding starch or ground shells to add weight, adding a dye to make a pale spice look bright, and selling one spice under the name of a dearer one. Some of the dyes used have been chemicals never meant to be eaten, and those cases moved from money fraud to safety danger.
The lesson of spices is that the cheapest adulterants can be the most harmful, because the fraudster picks whatever is cheap and looks right, without asking whether it is safe to eat. Spice testing now looks for dyes, for fillers, and for the chemical fingerprint of the real spice, and buyers are taught to buy whole spices and grind them at home when they can, because a whole spice is far harder to fake than a powder.
Spices also teach that fraud can hide in small amounts. A pinch of spice in a meal is small, so a small amount of adulterant in each pack is hard for a buyer to notice, but across millions of packs the profit is large. Testing must therefore be able to find small amounts, and a test that only finds large amounts of a fake will miss spice fraud entirely.
The diagram shows the lesson of each case. The steps point to the diagram: together the lessons form the first fraud defence, test the food, test it twice, and test it in the form the buyer actually eats.
Patterns across all the cases
Look across the cases and the same patterns repeat. First, the fraud always targets a food where the price gap is large. Second, the fraud always uses a substitute that passes the common test of the day. Third, the fraud always travels through a chain where one step is not checked. Fourth, the fraud is always found by a test that looks at the food in a new way: DNA for species, isotopes for origin, a fingerprint for composition. Fifth, the fraud always harms honest makers as well as buyers, and the honest makers are often the ones who push hardest for better testing after the event.
There is also a pattern in what happens after a case. Public attention rises, testing increases, new rules are written, and fraud falls for a while. Then attention fades, testing falls back, and fraud returns in a new form or with a new food. The work of fraud prevention is therefore never finished: it is a steady routine, not a one time campaign. A business that tests only after a scandal will always be one scandal behind.
Another pattern is that fraud is found by people who ask where the money is. Auditors who notice that a supplier sells below cost, inspectors who notice that a factory buys far less raw material than it sells as finished product, and buyers who notice a price that cannot be honest, all find fraud before a laboratory does. Numbers on paper, such as the amount of raw milk bought against the amount of cheese sold, can expose a fraud that no taste test could find. This is called a mass balance check, and it is one of the cheapest fraud tests there is.
Questions to think about
- Why did the milk powder fraud work even though the protein test was run correctly?
- A fishmonger sells whole fish and fillets. Which is easier to fake, and why does that matter for testing?
- After a famous fraud case, testing rises and then falls. What would keep testing steady instead?
Think about the milk powder question before you read on
The protein test measured nitrogen, not protein itself, and the added chemical was rich in nitrogen. The test was run correctly and still gave a wrong answer, because it measured a stand in for protein and the fraudster knew how to fool that stand in. The fix is a second test that measures protein, or the chemical itself, directly, so that a trick that fools one test is caught by the other.
What the cases teach a food student
A food student who knows these cases carries a set of working habits. Ask what a test really measures, not only what it is called. Ask whether a second, different test would give the same answer. Ask how long the supply chain is and which step nobody checks. Ask whether the price can be honest. Ask what the food looks like in the form the buyer eats, and whether that form hides the truth. These questions cost nothing, and they catch frauds that expensive instruments miss.
The cases also teach humility. Every generation of food scientists believed its tests were enough, and every generation was proved wrong by a new trick. The right attitude is to expect fraud, to test as if it is present, and to be pleased when testing finds nothing, rather than to assume that no news means no fraud. A laboratory that finds no fraud at all, year after year, may be testing well, or it may be testing the wrong things.
Practical note: Pick one of the cases in this unit and write a two page case file: what was faked, how the trick worked, which test found it, who was harmed, and what changed afterwards. Then add a line of your own: what test would you add today to catch the next version of the same trick? Keep your case file with the fraud profile from Unit 1, so that by the end of the subject you have a small library of frauds and defences.
Quick self-check
Question: Why is a mass balance check useful against fraud?
Because it compares what went into a factory with what came out. If a factory buys far less raw material than it sells as finished product, something was added or swapped, even if every finished pack passes its tests. The numbers expose what the product hides.
Question: A supplier offers beef at a price below the cost of raising cattle. What should the buyer do?
Refuse the price or test the meat before buying. A price below the cost of the real food cannot be honest, and the most likely explanation is substitution or dilution somewhere in the chain.
Try it yourself
Visit a market or a shop and look at three foods that are often faked: honey, olive oil, and a ground spice. For each one, read the label and note the price per 100 g, the origin, the maker, and any claim such as pure, extra virgin, or single origin. Then look for the cheapest and the dearest version of the same food. Write down the price gap and ask yourself: could the cheap one be honest at that price, and what would a fraudster gain by faking it? Bring your notes to class and compare gaps: which food has the largest gap, and does that match the foods that appear most often in fraud reports?
Key ideas to carry forward
- The milk powder case showed that a test which measures a stand in for a nutrient can be fooled, so a second, different test is needed.
- The horse meat case showed that paperwork can lie and long chains hide swaps, so the food itself must be tested.
- Olive oil fraud survives because the fake is hard to taste, so fingerprint tests matter more than tasting.
- Honey fraud adapts to each new test, so laboratories must keep updating their methods.
- Fish and spice fraud show that processed forms are easier to fake than whole foods, and that tests must find small amounts.
- Mass balance checks, comparing inputs with outputs, expose fraud cheaply.
- Fraud prevention is a steady routine, not a one time campaign after a scandal.
Glossary
Mass balance
A check that compares the amount of raw material bought with the amount of finished product sold. Food tech example: a cheese maker who buys little milk but sells much cheese fails a mass balance check.
DNA testing
A laboratory method that reads the genetic material of a food to name the species it came from. Food tech example: DNA testing of a fillet shows whether it is cod or a cheaper white fish.
Isotope analysis
A method that reads tiny differences in the atoms of a food to show where it came from or how it was made. Food tech example: isotope analysis of honey shows whether its sugars came from flowers or from a factory syrup.
Adulterant
The foreign substance added to a food in adulteration. Food tech example: a nitrogen rich chemical added to milk powder to fool a protein test is an adulterant.
Traceability
The ability to follow a food back through its supply chain to its origin, using records at each step. Food tech example: a batch number on a ready meal lets an inspector trace its meat back to the supplier.
Extra virgin olive oil
Olive oil pressed from olives without heat or chemicals, with a low level of free acidity, and sold as the highest grade of olive oil. Food tech example: a bottle labelled extra virgin should contain oil pressed only by mechanical means.
Refined oil
Oil that has been treated with heat or chemicals to remove flavour, colour, or acidity, and sold at a lower grade than a cold pressed oil. Food tech example: refined olive oil is cheaper than extra virgin oil and is often used to dilute it.
Fatty acid profile
The pattern of the different fatty acids in a fat or oil, used like a fingerprint to name the oil. Food tech example: the fatty acid profile of olive oil differs from that of seed oils, so a blend can be found.
Wild caught
Fish caught in the wild, as opposed to fish raised on a farm. Food tech example: a fillet labelled wild caught sells at a higher price than the same fish raised on a farm.
Single origin
A claim that a food comes from one named place or one producer. Food tech example: honey labelled single origin should come from one named region, not from a blend of many regions.
Ready meal
A prepared dish sold ready to heat and eat, often made from many ingredients through a long supply chain. Food tech example: a ready meal labelled beef may contain meat from several suppliers.
Fraudster
A person or business that carries out food fraud. Food tech example: a fraudster who dilutes honey with syrup gains money on every jar sold.
Protein test
A laboratory method that estimates protein, often by measuring nitrogen and working out the protein from it. Food tech example: the common protein test on milk measures nitrogen, so it can be fooled by a nitrogen rich chemical.
Stand in
A measurement used in place of a direct measurement, because it is easier or cheaper, such as nitrogen standing in for protein. Food tech example: nitrogen is a stand in for protein in the common milk test.
References
- Spink, J. and Moyer, D. C., Defining the Public Health Threat of Food Fraud, 2011.
- Moore, J. C., Spink, J. and Lipp, M., Development and Application of a Database of Food Ingredient Fraud and Economically Motivated Adulteration, 2012.
- Manning, L., Food Fraud: Policy and Management, 2016.
- Ellis, D. I., et al., Fingerprinting Food: Current Technologies for the Detection of Food Adulteration and Contamination, 2012.
- Woolfe, M. and Primrose, S., Food Forensics and Toxicology, 2004.
Unit 3: Detecting Fraud, DNA, Isotopes, Chromatography, and Spectroscopy
Note
Unit 3: Detecting Fraud, DNA, Isotopes, Chromatography, and Spectroscopy
A question to start with
Two jars of honey sit on a laboratory bench. Both are golden, both taste sweet, both pass a simple sugar test. One is pure honey from bees. The other is half syrup. The analyst must tell them apart without trusting the label, the taste, or the simple test. What tools can read the truth inside a food when the eyes and the tongue cannot? This unit is about those tools: the laboratory methods that name a species, trace an origin, separate a mixture, and read a chemical fingerprint.
Detection is the science of catching fraud, and it rests on one idea: every food carries marks of where it came from and what it is made of, and those marks survive processing in ways that a fraudster cannot easily erase. DNA survives in meat and fish. Isotopes survive in sugars and water. The pattern of fatty acids survives in oil. The job of the analyst is to read those marks and compare them with the marks of the real food.
No single method catches every fraud. A good laboratory uses several methods that look at different marks, so that a trick which fools one method is caught by another. This unit explains the four main families of methods, what each one is good at, and where each one falls short.
DNA methods: naming the species
DNA is the genetic material found in the cells of plants and animals, and it differs between species. A DNA test reads a short stretch of that material and matches it against a library of known species, so that a piece of meat or fish can be named with confidence. For fraud work, the common DNA methods are barcoding and polymerase chain reaction, known as PCR.
DNA barcoding reads a short standard stretch of DNA, the barcode, and compares it with a reference library. If the barcode matches cod, the fish is cod; if it matches a cheaper white fish, the label is wrong. Barcoding works well on raw and lightly cooked foods, and it can name the species in a fillet, a sausage, or a ready meal.
PCR is a method that copies a chosen stretch of DNA many times, so that even a tiny amount can be found. Real time PCR can also estimate how much of a species is present, which matters when a food is a mixture: a sausage that is mostly pork with a little beef can be shown to be a mixture, not a pure beef sausage. PCR can also find a species that should not be there at all, such as pork in a food labelled as free of pork, which matters for religious and allergy reasons as well as for fraud.
DNA methods have limits. Heat, pressure, and long cooking break DNA into short pieces, so a heavily processed food may give a weak signal or none at all. Refined oils, sugars, and syrups carry little or no DNA, so DNA cannot test honey or olive oil. And DNA names the species, not the origin: it can say a fish is cod, but not which sea it came from. For origin, the laboratory turns to isotopes.
Questions to think about
- A ready meal labelled beef gives a DNA result of pork and beef mixed. What does that prove, and what does it not prove?
- Why can DNA testing name the fish in a fillet but not test a bottle of olive oil?
- A sausage is heavily cooked and the DNA signal is weak. Which other method could help, and why?
Think about the ready meal question before you read on
The DNA result proves that pork is present in the meal, so the label claiming only beef is wrong. It does not prove who made the swap, or at which step of the supply chain it happened, and it does not by itself show how much pork is present unless a quantitative method was used. To answer those questions the laboratory would need a quantitative test and the supply chain records, so that the amount and the step can be traced.
Isotope analysis: reading where a food came from
Isotopes are forms of the same element with different weights, and their proportions in a food depend on where the food grew and how it was made. Carbon, hydrogen, oxygen, and nitrogen isotopes in a food carry a signature of the soil, the water, the climate, and the plant type. Isotope analysis reads that signature and compares it with the signature of the real food.
The classic use is honey. Sugars made by bees from flower nectar carry a carbon isotope pattern that differs from the pattern of sugars made in a factory from certain crop plants. When honey is diluted with such a syrup, the isotope pattern shifts, and the test can show that the honey is not pure, even when the total sugar looks right. The same idea works for fruit juice, for vanilla, and for alcohol in drinks.
Isotope analysis can also show origin. Water in a food carries the oxygen and hydrogen isotope pattern of the place where the food grew, because rain and ground water differ from place to place. A food claimed to come from one region can be checked against the isotope pattern of that region, and a mismatch is evidence that the claim is false. This is useful for foods sold under a place name, where the name itself is part of the price.
The limits of isotope work are cost and reference data. The instruments are dear, the work needs skill, and the result means little without a library of genuine samples from the claimed origin to compare against. A laboratory that has no genuine honey from a region cannot say whether a sample matches that region. Building those libraries is slow work, and it is one reason isotope testing is used for high value cases rather than for every jar on a shelf.
Chromatography: separating a mixture to see its parts
Chromatography is a family of methods that separate the parts of a mixture, so that each part can be seen and measured on its own. A sample is carried through a column or across a plate by a moving liquid or gas, and the parts travel at different speeds because they stick to the column or the plate in different ways. The result is a pattern of peaks, one for each part, and that pattern is a fingerprint of the food.
Gas chromatography is used for volatile parts, the small molecules that evaporate easily, such as flavour compounds and fatty acids. The fatty acid pattern of an oil, read by gas chromatography, is the fingerprint that exposes olive oil blended with cheaper oils: the blend has a pattern that pure olive oil never has. Liquid chromatography is used for larger and less volatile parts, such as sugars, dyes, preservatives, and the chemical markers of a spice. A liquid chromatography test can find a dye in chilli powder, or a sweetener in a juice that claims to be unsweetened.
Chromatography is often joined to mass spectrometry, a detector that weighs the parts as they leave the column and names them by their weight and by the fragments they break into. The joined method, known as LC-MS or GC-MS, is the workhorse of fraud detection: it can find a named adulterant at a very low level, and it can confirm a fingerprint with high confidence. A laboratory that finds a strange peak in a chromatogram can use mass spectrometry to name the substance behind it, which is how new frauds are often discovered.
The strength of chromatography is its detail and its low detection limit: it can find a small amount of a fake in a large amount of real food. Its weakness is that it needs a target: the analyst must know what to look for, or must read the whole pattern and spot what is unusual. A completely new adulterant, never seen before, may pass as an unknown peak until someone works out what it is.
Spectroscopy: reading light to read a food
Spectroscopy reads how a food absorbs, reflects, or scatters light, and uses that reading to judge what the food contains. Different molecules absorb light at different wavelengths, so the spectrum of a food, the pattern of light it absorbs across many wavelengths, is a fingerprint of its composition. The common methods are near infrared, known as NIR, mid infrared, known as FTIR, and Raman spectroscopy.
The great strength of spectroscopy is speed. A NIR instrument can scan a sample in seconds, without destroying it and often without any preparation, and a computer can compare the spectrum with a library of genuine samples and give an answer at once. This makes spectroscopy ideal for screening: a factory can scan every delivery of milk powder, every drum of oil, or every sack of spice, and send only the samples that look strange onward for slow, costly confirmation by chromatography or DNA.
Spectroscopy is also used to detect dilution and substitution in powders and liquids. The spectrum of pure honey differs from the spectrum of honey diluted with syrup. The spectrum of pure milk powder differs from the spectrum of milk powder with added starch. The differences are small, and a human eye cannot see them in the raw numbers, but a computer model trained on genuine samples can spot them.
The weakness of spectroscopy is that it is indirect: it does not name a molecule, it recognises a pattern. A good model needs many genuine samples to learn from, and a fraudster who finds a new trick may fool a model trained on old tricks. Spectroscopy is therefore a screening tool, not a final proof. A sample that fails a spectroscopic screen must be confirmed by a method that names the substance, and a sample that passes the screen is not proven genuine, only not suspicious in the ways the model knows.
How a laboratory puts the methods together
A good fraud laboratory works in layers. The first layer is screening: fast, cheap methods such as spectroscopy and simple chemical tests scan many samples and flag the strange ones. The second layer is confirmation: slow, precise methods such as chromatography with mass spectrometry, DNA, and isotope analysis prove what the strange sample really is. The third layer is the paper check: mass balance, supplier records, and traceability, which show where the fraud entered the chain.
This layered approach saves money and time. Screening every sample with the slowest method would be too costly, and screening with only the fast method would miss too much. The layers also protect against a single method being fooled: a fraud that passes a spectroscopic screen may fail a DNA test, and a fraud that passes a DNA test may fail an isotope test, because each method reads a different mark.
Sampling matters as much as the method. A laboratory test on a sample that does not represent the lot is worthless, however good the instrument. The sampling rules from food analysis apply here too: many increments, a composite sample, and a clear chain from the lot to the test portion. A fraudster who knows which sacks will be sampled can keep the tested sacks honest and sell the rest as fake, so sampling must be random and unannounced.
The diagram shows the layered detection chain. The steps point to the diagram: each layer catches what the last layer missed, and the paper check at the end shows where the fraud entered.
Limits, costs, and choosing the right test
Every detection method has a detection limit, the smallest amount of a fake it can reliably find. A method that finds 10 percent of a substitute will miss a fraud that adds only 2 percent, and a fraudster who knows the limit can stay below it. Laboratories therefore state the limit of each method and choose methods whose limits sit below the level that would make fraud profitable. A test that can find 1 percent of a cheap oil in olive oil is useful; a test that only finds 20 percent is not.
Cost also shapes the choice. DNA and isotope tests need skilled staff and dear instruments, so they are used for high value foods and for confirmation. Spectroscopy is cheap per sample once the instrument and the model are built, so it is used for screening many samples. A small business cannot run a full fraud laboratory, so it buys testing from a commercial laboratory, or it joins a testing scheme with other businesses, or it buys from suppliers who test and can show the results.
The choice of test also depends on the question. Is the question which species is in the meat? Use DNA. Is the question where the honey came from? Use isotopes. Is the question what was added to the oil? Use chromatography. Is the question whether this delivery looks like the last hundred deliveries? Use spectroscopy. A laboratory that starts from the question, rather than from the instrument it happens to own, gives better answers for less money.
Practical note: In a laboratory practical, run the same sample through two different checks, such as a simple sugar test and a careful reading of a label and a price, and note where the two checks agree and where they differ. The habit of using two methods and comparing the answers is the core skill of fraud detection, and it works with simple tools as well as with dear instruments.
Quick self-check
Question: Why is spectroscopy good for screening but not for final proof?
Because it recognises a pattern rather than naming a molecule. It can flag a strange sample in seconds, but a strange pattern must be confirmed by a method such as chromatography with mass spectrometry, which names the substance behind the pattern.
Question: A honey sample passes a total sugar test but fails an isotope test. What does that tell you?
That the total sugar is right but the source of the sugar is wrong. The honey was probably diluted with a syrup whose sugars match honey in amount but not in isotope pattern, so the sugar test was fooled and the isotope test caught the trick.
Try it yourself
Take three foods from your kitchen: a spice, a cooking oil, and a fruit juice. For each one, write down which detection method you would choose and why: DNA, isotope analysis, chromatography, or spectroscopy. Then write down what the method would find and what it could not find. For example, DNA can name the fish in a fillet but cannot test olive oil, because oil carries little DNA. Compare your choices with a classmate: where did you pick different methods for the same food, and which choice fits the question better?
Key ideas to carry forward
- Every food carries marks of its species, its origin, and its composition, and detection reads those marks.
- DNA methods name the species in meat and fish, and can find a species that should not be there.
- Isotope analysis reads where a food came from and whether its sugars came from a plant or a factory.
- Chromatography separates a mixture so that each part can be seen, measured, and named, often with mass spectrometry.
- Spectroscopy reads a light fingerprint fast and cheaply, and is ideal for screening many samples.
- A good laboratory works in layers: screen fast, confirm slowly, and check the paper trail.
- Every method has a detection limit, and the test must be chosen from the question, not from the instrument at hand.
Glossary
DNA barcoding
A method that reads a short standard stretch of DNA and matches it against a library to name the species. Food tech example: DNA barcoding of a fillet shows whether it is cod or a cheaper fish.
PCR
Polymerase chain reaction, a method that copies a chosen stretch of DNA many times so that even a tiny amount can be found. Food tech example: PCR finds a trace of pork in a food labelled free of pork.
Isotope
A form of an element with a different weight, whose proportions in a food depend on where the food grew and how it was made. Food tech example: carbon isotopes in honey show whether its sugars came from flowers or from a factory syrup.
Chromatography
A family of methods that separate the parts of a mixture so each part can be seen and measured. Food tech example: gas chromatography reads the fatty acid pattern of an oil to expose a blend.
Mass spectrometry
A detector that weighs molecules and names them by their weight and fragments, often joined to chromatography. Food tech example: LC-MS names the dye found in a spice sample.
Spectroscopy
Methods that read how a food absorbs or scatters light, giving a fingerprint of its composition. Food tech example: near infrared spectroscopy screens milk powder deliveries in seconds.
Near infrared
Light just beyond visible red light, used in NIR spectroscopy to scan foods quickly without destroying them. Food tech example: a NIR scan of a sack of flour checks its moisture and protein at once.
Detection limit
The smallest amount of a substance that a method can reliably find. Food tech example: a test with a detection limit of 1 percent finds 1 percent of cheap oil in olive oil, but misses 0.5 percent.
Screening
Fast, cheap testing of many samples to flag the strange ones for slower confirmation. Food tech example: spectroscopy screens every delivery, and only strange deliveries go on to DNA testing.
Confirmation
A slow, precise test that proves what a flagged sample really is. Food tech example: chromatography with mass spectrometry confirms a dye that spectroscopy flagged in chilli powder.
Fatty acid profile
The pattern of the different fatty acids in a fat or oil, used like a fingerprint to name the oil. Food tech example: the fatty acid profile of olive oil differs from that of seed oils, so a blend can be found.
Reference library
A collection of genuine samples and their test results, used to compare an unknown sample against. Food tech example: a DNA library of fish species lets a laboratory name an unknown fillet.
Real time PCR
A PCR method that watches the copying as it happens and can estimate how much of a species is present. Food tech example: real time PCR estimates how much beef and how much pork are in a mixed sausage.
Volatile compound
A small molecule that evaporates easily and carries flavour or aroma. Food tech example: gas chromatography reads the volatile compounds that give a spice its smell.
References
- Ellis, D. I., et al., Fingerprinting Food: Current Technologies for the Detection of Food Adulteration and Contamination, 2012.
- Woolfe, M. and Primrose, S., Food Forensics and Toxicology, 2004.
- Nielsen, S. S., Food Analysis, 5th edition, 2017.
- Downey, G., Food Authenticity and Traceability, 2003.
- Reid, L. M., O'Donnell, C. P. and Downey, G., Recent Technological Advances for the Determination of Food Authenticity, 2006.
Unit 4: PFAS, Where It Comes From, and Why It Matters
Note
Unit 4: PFAS, Where It Comes From, and Why It Matters
A question to start with
A non stick frying pan lets an egg slide off without oil. A takeaway box keeps grease from soaking through. A raincoat keeps you dry in a storm. All three can owe their trick to the same family of chemicals: PFAS. The same chemicals that keep grease and water out are now found in drinking water, in soil, in fish, and in food. They do not break down easily, in the environment or in the body. How did chemicals made to repel water and grease end up inside food, and what can food makers do about it? This unit explains PFAS in plain terms: what it is, where it comes from, how it reaches food, and how the food world is responding.
PFAS stands for per and polyfluoroalkyl substances, a large family of human made chemicals built around carbon and fluorine. The carbon fluorine bond is one of the strongest bonds in chemistry, and that strength is both the point and the problem. It makes PFAS resist heat, water, grease, and stains, which is why industry uses them. It also makes PFAS resist breaking down, so once they enter soil or water they stay there for a very long time. That is why they are often called forever chemicals.
There are thousands of PFAS chemicals, and they are not all the same. Some are long chain forms that build up in the body over years. Some are short chain forms that move more easily through soil and water. Some have been phased out in many places, and others have taken their place. For food students, the key point is not to memorise the names of each chemical: it is to understand the family, the routes into food, and the habits that reduce exposure.
Where PFAS comes from
PFAS was made for useful jobs, and it is still used in many of them. The best known uses are non stick cookware, grease resistant food packaging, stain resistant carpets and fabrics, waterproof clothing, and firefighting foams. It is also used in some industrial processes, such as making certain plastics and coatings, and in some paints and polishes.
The chemicals reach the environment in several ways. Factories that make or use PFAS can release them into air and water. Firefighting foam used in training or in a real fire can soak into soil and ground water. Waste sent to landfill can leak PFAS into the ground. Sewage sludge spread on fields as fertiliser can carry PFAS onto farmland. Once in soil or water, PFAS moves: it can travel in ground water far from the place it was released, and it can be taken up by crops and by animals that drink the water or eat the crops.
Food packaging is a direct route that food students should know well. Grease resistant paper and board, used for takeaway boxes, burger wrappers, popcorn bags, and bakery papers, has often been treated with PFAS so that grease does not soak through. The PFAS can move from the paper into the food, more so when the food is hot, greasy, and in contact with the paper for a long time. That is why the move to PFAS free packaging, covered in the Sustainable Food Packaging subject, matters for food safety as well as for the environment.
How PFAS reaches food
PFAS reaches food through water, soil, feed, and packaging, and each route matters for a different food.
Water is the largest route. Fish and shellfish living in water that carries PFAS take the chemicals into their flesh, and the chemicals build up over the life of the fish, so larger and older fish can carry more. Drinking water that carries PFAS brings the chemicals into people directly, and into food made with that water. Crops watered with contaminated water can take up PFAS through their roots, more so in leafy vegetables and in crops grown in contaminated soil.
Soil is the second route. Crops grown in soil that carries PFAS take up some of the chemicals, and animals grazing on contaminated land or eating contaminated feed take them up too. Meat, milk, and eggs from animals that lived on contaminated land can carry PFAS, and the chemicals can pass into milk and eggs. The amount depends on the level in the soil, the type of crop or animal, and the length of exposure.
Packaging and cookware are the third route, and the most direct one for a food business to control. Hot greasy food in a treated wrapper, food cooked in a worn non stick pan, or food stored in a treated container can pick up PFAS from the surface it touches. The amounts from one meal are small, but the chemicals build up in the body over years, so small amounts from many meals add up. That slow build up is what makes PFAS different from a poison that acts at once: the harm, where it occurs, comes from long exposure to small amounts.
The body clears PFAS slowly. Some forms stay in the body for years, stored in blood and in organs, and levels fall only slowly once exposure stops. This is why regulators around the world have moved to reduce exposure from all routes at once, rather than treating one food or one product in isolation.
Questions to think about
- A takeaway box keeps grease from soaking through. What chemical family might be doing that job, and what is the trade off?
- Fish from a river near an old industrial site carry more PFAS than fish from a clean lake. Why would larger, older fish carry more?
- Why is PFAS called a forever chemical, and what does that name tell you about cleaning it up?
Think about the takeaway box question before you read on
The grease resistance often comes from PFAS treated paper. The trade off is that the chemical that keeps grease out of the box can move into the food, more so when the food is hot and greasy and sits in the box for a while. The box is convenient and clean to hold, but the convenience carries a chemical that does not break down and that builds up in the body over years, which is why many food businesses are moving to PFAS free packaging.
Why PFAS matters for health
The health concern with PFAS is not a single dramatic poisoning, but a set of effects linked to long exposure. Research has linked high exposure to PFAS with effects on the liver, on cholesterol levels, on the immune system, on thyroid hormones, and on the growth and development of babies. The evidence is still growing, and scientists are careful about which effects are proven and which are only suggested, but the pattern across many studies is consistent enough that regulators have acted to reduce exposure.
Two points help students read the evidence fairly. First, the studies often look at people with high exposure, such as workers in a factory or people drinking heavily contaminated water, and the effects at low exposure are harder to prove. That does not mean low exposure is safe: it means the science is still catching up, and the prudent course is to reduce exposure where it can be reduced cheaply. Second, PFAS is a family, not one chemical, and different members of the family have different strengths of evidence. Long chain forms have the most evidence behind them, which is why they were phased out first in many places.
For food, the practical meaning is plain: food that carries PFAS adds to a person's total exposure, and total exposure is what matters. No single food is the whole story, and no single meal decides the outcome. The work of the food industry is to bring down the level in food where it can be brought down, by choosing cleaner water, cleaner land, cleaner feed, and cleaner packaging.
How PFAS is measured
PFAS is measured in food, water, and soil by liquid chromatography joined to mass spectrometry, known as LC-MS, the same family of methods met in Unit 3. The method can find PFAS at very low levels, measured in parts per trillion, which is a way of saying a few grains of the chemical in an Olympic swimming pool of water. Such low levels are needed because the chemicals matter even in tiny amounts when exposure lasts for years.
Measuring at parts per trillion is hard work. The laboratory must keep PFAS out of its own equipment, because PFAS is used in some laboratory plastics and coatings, and a dirty blank can give a false result. Samples must be taken in containers that do not add PFAS, and the laboratory must run blanks, samples with no PFAS, alongside the real samples to prove that the result came from the food and not from the laboratory. The cost of this care is one reason PFAS testing is done on a risk basis, testing the foods and the places most likely to carry it, rather than testing every food.
Results are reported for each PFAS chemical separately, and often as a sum of several forms, because people are exposed to the family as a mixture. A food may carry small amounts of several PFAS forms, and the sum is what matters for judging exposure. Students reading a PFAS report should look for which forms were tested, what the detection limit was, and whether the laboratory ran blanks: a report that names none of these is hard to trust.
What the food world is doing
The response to PFAS has four parts, and each part is work for food professionals.
First, stop adding it where it is not needed. Many food businesses have moved to PFAS free packaging, using grease resistant papers made with other coatings, or using packaging that does not need grease resistance at all. Cookware makers have developed non stick surfaces without PFAS, and fire services have moved to foams without PFAS for training. Each change removes one source.
Second, clean up water and soil where it is found. Water treatment can remove PFAS with activated carbon or with membrane filters, and contaminated soil can be treated or taken out of food production. This work is slow and costly, and it is done by governments and water companies rather than by single food businesses, but food businesses can help by testing their water and their land.
Third, test food on a risk basis. Foods from places known to carry PFAS, fish from certain waters, crops from certain soils, and milk from certain farms are tested more often, and foods that repeatedly test clean are tested less. This focuses the money where the risk is.
Fourth, set limits and advice. Regulators in many places have set limits for PFAS in drinking water and in certain foods, and have given advice on how often to eat fish from waters known to carry PFAS. The limits differ from place to place, because the science is still moving and because each place weighs the evidence in its own way. A food student should know the principle, that limits exist and are getting tighter, and should check the current rule in their own country rather than memorising one number from one place.
The diagram shows the routes of PFAS into food. The steps point to the diagram: each route can be reduced, and the work is to reduce all of them at once, because total exposure is what matters.
What a food business can do now
A food business does not need to wait for a new law to act on PFAS. The first step is to ask suppliers what packaging they use, and to switch to PFAS free packaging where it is available. The second step is to test the water used in the business, and to treat it if it carries PFAS. The third step is to know where ingredients come from, and to test ingredients from places known to carry PFAS more often. The fourth step is to keep records, so that if a food is found to carry PFAS, the source can be traced and fixed.
For a home cook, the steps are simpler: use PFAS free packaging where it is offered, replace worn non stick pans, and follow local advice on fish from local waters. No single step removes all exposure, and no step needs to be perfect to be worth doing. The goal is to bring total exposure down, step by step, while the larger work of cleaning water and soil goes on.
Practical note: Look at the food packaging in your kitchen or in a takeaway meal. Read the pack for any claim about grease resistance or PFAS free materials, and note the type of pack: paper, board, plastic, or coated. Hot greasy food in a grease resistant paper wrapper is the case where migration is highest, so ask what the wrapper is made of and whether a PFAS free version is offered. Write down what you find and compare it with a classmate: how many packs carry a PFAS free claim, and how many carry no claim at all?
Quick self-check
Question: Why does PFAS build up in the body over years?
Because the carbon fluorine bond resists breaking down, and the body clears PFAS slowly. Small amounts taken in meal after meal add up, because each new amount arrives before the last one has left the body.
Question: A food business switches to PFAS free packaging. Does that remove all PFAS from its food?
No. Packaging is one route, but PFAS can also reach food through water, soil, and feed. The switch removes the packaging route, which is a real gain, but the business should still test its water and know where its ingredients come from.
Try it yourself
List five foods you eat in a normal week. For each one, write down the likely PFAS routes: water, soil, feed, packaging, or cookware. Then write one practical step that would reduce the exposure for that food, such as choosing a different pack, a different source, or a different cooking pan. End with a short note on which route is hardest for a single person to control and why, and what that tells you about the need for action by water companies and regulators as well as by shoppers.
Key ideas to carry forward
- PFAS is a large family of human made chemicals built on the strong carbon fluorine bond, which makes them useful and very hard to break down.
- PFAS reaches food through water, soil, feed, packaging, and cookware, and the chemicals build up in the body over years.
- Fish, crops, meat, milk, and eggs can all carry PFAS when the water, soil, or feed is contaminated.
- Grease resistant food packaging is a direct route into food, more so for hot greasy food in long contact with the pack.
- PFAS is measured by LC-MS at very low levels, and the laboratory must guard against PFAS from its own equipment.
- The response is to stop adding PFAS where it is not needed, clean up water and soil, test on a risk basis, and set limits.
- Total exposure from all routes is what matters, so small reductions in many places add up.
Glossary
PFAS
Per and polyfluoroalkyl substances, a large family of human made chemicals built around carbon and fluorine, used to resist water, grease, and stains. Food tech example: PFAS treated paper keeps grease from soaking through a takeaway box.
Forever chemical
A common name for PFAS, because the chemicals break down very slowly in the environment and in the body. Food tech example: PFAS in soil can stay there for many years after it was released.
Carbon fluorine bond
The strong chemical bond between carbon and fluorine atoms, which gives PFAS its resistance to heat, water, and grease. Food tech example: the carbon fluorine bond is why a PFAS coating survives hot greasy food.
Long chain PFAS
PFAS chemicals with a longer carbon chain, which tend to build up more in the body. Food tech example: long chain PFAS were phased out first in many places because of their build up.
Short chain PFAS
PFAS chemicals with a shorter carbon chain, which move more easily through soil and water. Food tech example: short chain PFAS can travel further in ground water than long chain forms.
Migration
The movement of a chemical from packaging or cookware into food. Food tech example: PFAS migrates from a grease resistant wrapper into hot greasy food.
Parts per trillion
A very small unit of concentration, used for PFAS because the chemicals matter even in tiny amounts. Food tech example: PFAS in drinking water is measured in parts per trillion.
Activated carbon
A form of carbon with a large surface that traps chemicals from water, used in water treatment. Food tech example: activated carbon filters remove PFAS from drinking water.
Bioaccumulation
The build up of a chemical in a living thing over time, because intake is faster than clearance. Food tech example: PFAS bioaccumulates in fish, so older fish can carry more.
Grease resistant paper
Paper treated so that grease does not soak through it, often used for takeaway food. Food tech example: a burger wrapper made of grease resistant paper may carry PFAS unless it is labelled PFAS free.
Firefighting foam
Foam used to put out fuel fires, some forms of which have carried PFAS into soil and ground water. Food tech example: foam used in training at an airfield can contaminate nearby farmland.
Sewage sludge
The solid material left after treating waste water, sometimes spread on fields as fertiliser. Food tech example: sludge that carries PFAS can bring the chemicals onto farmland.
LC-MS
Liquid chromatography joined to mass spectrometry, the main method for measuring PFAS at very low levels. Food tech example: LC-MS measures PFAS in a fish sample in parts per trillion.
Blank
A test run with no sample, used to prove that a result came from the food and not from the laboratory. Food tech example: a blank run alongside a PFAS test shows whether the laboratory itself added PFAS.
References
- Cousins, I. T., et al., The High Persistence of PFAS, 2019.
- Sunderland, E. M., et al., A Review of the Pathways of Human Exposure to Poly and Perfluoroalkyl Substances, 2019.
- Domingo, J. L. and Nadal, M., Human Exposure to Per and Polyfluoroalkyl Substances through Food, 2019.
- Food and Agriculture Organization, Food Safety and PFAS, 2021.
- Buck, R. C., et al., Perfluoroalkyl and Polyfluoroalkyl Substances in the Environment, 2011.
Unit 5: Microplastics and Nanoplastics in Food
Note
Unit 5: Microplastics and Nanoplastics in Food
A question to start with
A bottle of water, a fillet of fish, a spoonful of sea salt, a cup of tea from a plastic teabag: each one can carry tiny pieces of plastic too small to see. The plastic did not start in the food. It started as a bag, a bottle, a fishing net, or a piece of packaging, and it broke into smaller and smaller pieces until it was small enough to be eaten without noticing. How small can plastic get, how does it reach food, and what do we know, and not know, about what it does inside the body? This unit explains microplastics and nanoplastics in plain terms.
A microplastic is a piece of plastic smaller than 5 millimetres, about the size of a sesame seed or smaller. A nanoplastic is far smaller, less than one micrometre, which is a thousand times smaller than a millimetre and too small to see with an ordinary microscope. The two sizes behave differently, and that difference matters: microplastics are mostly a problem of presence, while nanoplastics raise a problem of movement, because they may be small enough to pass into tissues.
Plastics are made of long chains of molecules called polymers, and the common ones in food are polyethylene, polypropylene, polystyrene, polyethylene terephthalate, known as PET, and polyvinyl chloride, known as PVC. Each plastic has its own shape, its own density, and its own way of breaking down, and each one leaves its own fingerprint when it is found in food.
How plastics become microplastics
Microplastics come from two routes. Primary microplastics are made small on purpose: tiny beads once used in face scrubs and toothpaste, pellets used as raw material in plastic factories, and fibres shed from synthetic clothes in washing machines. Many places have now banned plastic microbeads in cosmetics, but pellets and fibres still reach water through drains and waste.
Secondary microplastics are the larger share, and they are made by breaking down. A plastic bag in the sea, a bottle on a beach, a fishing net lost in the water, or a piece of packaging in a field is broken by sunlight, by waves, by heat, and by rubbing, into smaller and smaller pieces. Sunlight makes plastic brittle, waves snap it, and the pieces keep splitting until they are microplastics, and then nanoplastics. Unlike wood or paper, plastic does not rot away: it only gets smaller, and the smaller pieces are harder to collect and harder to see.
Plastic also reaches food through use. A plastic chopping board sheds tiny shavings when a knife cuts on it. A plastic teabag releases tiny pieces into hot water. A bottle of water stored for a long time, or shaken, or heated, can shed pieces from its walls and its cap. A food processed with plastic equipment, or packed in plastic film, can pick up pieces from the surfaces it touches. Each route adds a small amount, and the amounts add up across a whole diet.
How microplastics reach food
The main route into food is water. Rivers and seas carry microplastics from land, from waste water, and from fishing and shipping, and fish and shellfish living in that water take the pieces in. Shellfish such as mussels and oysters filter large volumes of water to feed, and they can hold microplastics in their gut and tissues. Fish can carry microplastics in the gut, and the gut is usually removed before eating, which lowers the amount a person eats, but shellfish are often eaten whole, gut and all, so the pieces are eaten too.
Salt is another route. Sea salt is made by evaporating sea water, and microplastics in that water can end up in the salt. The amounts are small, and salt is eaten in small amounts, so the total intake from salt is low, but the route shows how widely the pieces have spread.
Drinking water is a third route. Tap water and bottled water both carry microplastics, in amounts that vary with the source, the treatment, and the packaging. Bottled water often carries more than tap water, because the bottle itself is a source, and because the water sits in plastic for a long time. The amounts are still small, and water treatment can remove many of the larger pieces, but the smallest pieces are the hardest to filter out.
Food processing and packaging are a fourth route. Foods that are heated in plastic, stored in plastic, or cut on plastic pick up pieces from those surfaces. The amounts depend on the temperature, the time of contact, the age and wear of the plastic, and the type of food: hot, fatty, or acidic foods pick up more. This is a route that food businesses can control directly, by choosing the right materials and replacing worn plastic.
Questions to think about
- A mussel and a fish fillet are both eaten. Why does the mussel carry more microplastics to the plate?
- A plastic bottle is left in a hot car for a week. What would you expect to happen to the water inside, and why?
- Why are nanoplastics a different problem from microplastics, even though both are plastic?
Think about the mussel question before you read on
A mussel filters water to feed and holds the pieces in its gut, and it is eaten whole, gut and all, so the microplastics it filtered are eaten too. A fish carries microplastics mostly in its gut, and the gut is removed before the fillet is eaten, so the fillet carries far fewer pieces. The difference is not the water they lived in: it is which part of the animal reaches the plate.
What we know about health, and what we do not know
The health question for microplastics and nanoplastics is still open, and honest teaching must say so. What is known is that people eat, drink, and breathe microplastics every day, in amounts that are small but regular. What is not yet known is how much of that intake stays in the body, how much passes through, and what harm, if any, the pieces that stay can do.
Three concerns are studied most. The first is the plastic itself: a piece that lodges in the gut could cause irritation, and a nanoplastic small enough to pass into tissues could reach organs. The evidence for harm in people is still thin, because the studies are hard to do and the amounts are small, but laboratory studies on cells and animals show that very small pieces can cause stress in tissues, and that is a reason to keep studying and to reduce exposure where it is cheap to do so.
The second concern is the chemicals in the plastic. Plastics carry additives that make them flexible, coloured, or resistant to heat, and some of those additives can move out of the plastic into food. A microplastic in food can therefore carry chemicals as well as plastic, and the chemical may matter more than the piece itself. The third concern is the chemicals and microbes that stick to the surface of a microplastic in the environment: a piece that has floated in polluted water can carry pollutants on its surface, and those pollutants can be released in the gut.
The fair summary is this: microplastics are present in food, the amounts are small, the health effects in people are not yet proven, and the prudent course is to reduce the plastic that reaches food while the science continues. Panic is not justified by the evidence, and ignoring the problem is not justified either.
How microplastics are measured
Measuring microplastics is hard, because the pieces are small, they are everywhere, and the laboratory itself is full of plastic. The common methods are microscopy, spectroscopy, and thermal analysis, and each one fits a different size.
Microscopy counts and describes the pieces that can be seen, by shape, colour, and size: fibre, fragment, film, or bead. It is simple and direct, but it cannot name the plastic, and it cannot see nanoplastics at all.
Spectroscopy, often FTIR or Raman, reads the light fingerprint of a piece and names the plastic: polyethylene, polypropylene, polystyrene, and so on. It works well for microplastics down to a certain size, but it struggles with the smallest pieces, and it is slow when there are many pieces to read one by one.
Thermal analysis, such as pyrolysis joined to gas chromatography and mass spectrometry, heats a sample until the plastic breaks down and reads the gases it gives off, which name the plastic and estimate how much is present. It can handle very small pieces and can give a total amount, but it destroys the sample and loses the shape and size of each piece.
The newest methods try to find nanoplastics, using electron microscopy and new forms of spectroscopy, but the work is still young and the results are hard to compare between laboratories, because each laboratory uses different methods and different size limits. A student reading a microplastics study should always ask: what size range was measured, which method was used, and how did the laboratory keep its own plastic out of the sample? A study that does not answer those questions is hard to trust.
Contamination control is the heart of microplastics work. Laboratory staff wear cotton coats, not synthetic ones, they work under clean air, they use glass and metal tools instead of plastic, and they run blanks, samples with no food, to prove that the pieces found came from the food and not from the laboratory. Without that care, the laboratory measures itself.
The diagram shows the journey of plastic into food. The steps point to the diagram: the plastic only gets smaller, never disappears, so the work is to stop it entering water and food in the first place.
What food businesses and shoppers can do
The most effective action is to stop plastic reaching water and food at the source. That means less single use plastic, better waste collection, keeping plastic out of rivers and seas, and replacing plastic in food contact where another material works as well. A food business that cuts its plastic use cuts the pieces that reach its food, and it also cuts the waste its customers must deal with.
Inside the business, the steps are practical: replace worn plastic chopping boards, utensils, and containers, because worn plastic sheds more; avoid heating food in plastic where glass or metal can be used instead; store food in glass or metal for long storage; and choose packaging with less plastic, or with plastic that is less likely to shed. Each step is small, and together they lower the amount of plastic that touches food.
For a home cook, the steps are the same in miniature: use wood or glass boards, replace scratched plastic containers, do not heat food in plastic unless the container is made for it, and use a reusable bottle and bag. None of these steps removes all microplastics from a diet, because the pieces are already in water and in the sea, but each step lowers the amount added in the kitchen, which is the part a person can control.
Regulators are still deciding how to treat microplastics in food, because the science of harm is still young and the methods are still being standardised. The direction of travel is clear: limits and monitoring will grow as methods improve, and businesses that already measure and reduce their plastic will be ready when the rules arrive.
Practical note: Look at the plastic that touches food in your kitchen: chopping boards, containers, utensils, bottles, and wraps. Sort them into three groups: worn and scratched, in good condition, and rarely used. The worn group sheds the most and is the first to replace. Note the type of plastic on each item from its recycling mark, and write down which items touch hot food, because heat increases shedding. Keep the list and check it again in a month: which items have you replaced, and which ones still need a better material?
Quick self-check
Question: Why is a laboratory blank so important in microplastics work?
Because plastic is everywhere, including in the laboratory: in clothes, in tools, in the air. A blank run with no food shows how many pieces came from the laboratory itself, so that the pieces counted in the food can be corrected. Without a blank, the laboratory may be measuring itself.
Question: Why can spectroscopy name the plastic in a microplastic but microscopy cannot?
Microscopy only shows shape, colour, and size, which do not name the polymer. Spectroscopy reads the light fingerprint of the piece, and each polymer has its own fingerprint, so the plastic can be named.
Try it yourself
Choose one meal you eat often. List every place plastic touches that meal: the shop pack, the storage container, the chopping board, the cooking pan, the serving plate, the bottle you drink with. For each touch point, write one change that would reduce the plastic: a glass container, a wooden board, a metal bottle, a paper wrap. Then cook the meal once with as many of those changes as you can, and write down which changes were easy, which were hard, and which ones you would keep. Bring your notes to class and compare: which touch point did most people find hardest to change, and why?
Key ideas to carry forward
- A microplastic is a piece of plastic smaller than 5 millimetres, and a nanoplastic is far smaller, under one micrometre.
- Most microplastics in food are secondary, made by sunlight, waves, and wear breaking larger plastic into smaller pieces.
- Fish, shellfish, salt, drinking water, and food processing and packaging all bring microplastics to the plate.
- Shellfish eaten whole carry more microplastics than a fish fillet, because the gut is eaten too.
- The health effects in people are not yet proven, but laboratory studies give reason to reduce exposure while the science continues.
- Microplastics are measured by microscopy, spectroscopy, and thermal analysis, each fitting a different size, and blanks are essential.
- The most effective action is to stop plastic reaching water and food, and to replace worn plastic that touches food.
Glossary
Microplastic
A piece of plastic smaller than 5 millimetres. Food tech example: a fragment of a plastic bottle found in sea salt is a microplastic.
Nanoplastic
A piece of plastic smaller than one micrometre, too small to see with an ordinary microscope. Food tech example: a nanoplastic may be small enough to pass into tissues, which is why it is studied separately.
Polymer
A long chain of repeating molecules that makes up a plastic. Food tech example: polyethylene is the polymer in many plastic bags and bottles.
Polyethylene
A common plastic used for bags, bottles, and films, often marked PE. Food tech example: a polyethylene shopping bag can break down into microplastics in the sea.
Polypropylene
A common plastic used for tubs, caps, and fibres, often marked PP. Food tech example: a polypropylene yoghurt tub can shed pieces when it is scratched.
Polystyrene
A plastic used for foam packs and cups, often marked PS. Food tech example: a polystyrene foam tray can break into small fragments that reach food.
PET
Polyethylene terephthalate, the plastic used for most clear drinks bottles. Food tech example: a PET bottle can shed tiny pieces into the water it holds.
Primary microplastic
A microplastic made small on purpose, such as a pellet or a fibre. Food tech example: a fibre shed from a synthetic shirt in the wash is a primary microplastic.
Secondary microplastic
A microplastic made by breaking down a larger piece of plastic. Food tech example: a fragment of a bottle broken by sun and waves is a secondary microplastic.
Bioaccumulation
The build up of a substance in a living thing over time, because intake is faster than clearance. Food tech example: shellfish bioaccumulate microplastics because they filter large volumes of water.
FTIR
Fourier transform infrared spectroscopy, a method that reads the light fingerprint of a plastic to name it. Food tech example: FTIR names the polymer in a microplastic found in salt.
Raman spectroscopy
A method that reads scattered light from a sample to name its molecules, used for small microplastics. Food tech example: Raman spectroscopy can name a microplastic too small for FTIR.
Pyrolysis
Heating a sample until it breaks down, so that the gases it gives off can be read to name the plastic. Food tech example: pyrolysis joined to gas chromatography estimates how much plastic is in a sample.
Blank
A test run with no sample, used to prove that a result came from the food and not from the laboratory. Food tech example: a blank run alongside a microplastics test shows how many pieces came from the laboratory air.
References
- Thompson, R. C., et al., Lost at Sea: Where Is All the Plastic?, 2004.
- Cole, M., et al., Microplastics as Contaminants in the Marine Environment, 2011.
- Geyer, R., Jambeck, J. R. and Law, K. L., Production, Use, and Fate of All Plastics Ever Made, 2017.
- Food and Agriculture Organization, Microplastics in Fisheries and Aquaculture, 2017.
- Hartmann, N. B., et al., Microplastics as Vectors for Environmental Contaminants, 2017.
Unit 6: Mycotoxins and Heavy Metals
Note
Unit 6: Mycotoxins and Heavy Metals
A question to start with
A farmer stores maize in a damp barn through a wet season. The grain looks a little dull, and some of it smells musty, but most of it looks fine, so it is milled into flour and sold. A laboratory tests the flour and finds a poison made by mould, present in amounts too small to see or taste, but large enough to harm the liver over years of eating. The poison was never added by anyone: it grew in the grain while it sat in the damp. How do poisons grow in food without anyone adding them, and how do heavy metals, which are also never added on purpose, end up in the same foods? This unit is about the two oldest contaminants in food: mycotoxins, the poisons made by moulds, and heavy metals, the elements that come from soil, water, and industry.
Mycotoxins and heavy metals are different in kind, but they share three traits that make them hard to manage. Both are invisible in the food. Both build up their harm over long exposure rather than in a single meal. And both are controlled less by cleaning the food after the event than by preventing them from getting in, through good farming, good storage, and clean water and soil.
Mycotoxins: poisons made by moulds
A mycotoxin is a poison made by a mould, which is a type of fungus that grows on food. Not all moulds make mycotoxins, and not all mycotoxins are made on every food, but the ones that matter for food safety are well known, and they grow on the foods the world eats most: grains, nuts, dried fruit, coffee, cocoa, and spices.
The main mycotoxins to know are aflatoxins, made by Aspergillus moulds on maize, groundnuts, tree nuts, and spices; ochratoxin A, made by Aspergillus and Penicillium moulds on cereals, coffee, cocoa, dried fruit, and wine grapes; fumonisins, made by Fusarium moulds on maize; deoxynivalenol, also called vomitoxin, made by Fusarium moulds on wheat and barley; and patulin, made by Penicillium moulds on apples and apple products. Each one has its own mould, its own food, and its own health effect, but they share one habit: they grow when food is warm and damp, and they survive cooking, milling, and most processing.
Aflatoxins are the most studied, because they are the most harmful. They damage the liver, and long exposure is linked to liver cancer, more so in people who also carry a liver virus. They also slow the growth of children who eat contaminated food every day, because the poison harms the gut and the body cannot use food well. Aflatoxins are invisible, tasteless, and heat stable: cooking a contaminated grain does not remove them, and a loaf made from contaminated flour carries the poison into the bread.
Moulds need three things to make mycotoxins: a food they can grow on, warmth, and moisture. A grain stored damp, a nut stored warm, or a fruit left bruised and wet gives a mould all three. The mould grows as a visible fuzz, but the mycotoxin is a chemical, and it can spread into the food beyond the visible mould. Cutting off the mouldy part of a grain or a nut does not remove the poison, because the poison has already spread into the parts that look clean.
How mycotoxins are controlled
Control starts in the field. Crops that are damaged by insects or drought are more open to mould, so good farming, pest control, and harvesting at the right time all lower the risk. Drying is the next step, and it is the most important one: grain dried quickly to a low moisture after harvest gives mould no water to grow with, and grain kept dry in storage stays safe. Storage must be cool, dry, and ventilated, and the grain must be checked often, because a damp patch in a store can grow mould that spreads through the whole lot.
Sorting is a useful second line. Mouldy, broken, and discoloured grains carry more mycotoxin than sound grains, so removing them by hand, by colour sorting, or by density sorting lowers the level in the rest. Milling also helps in some cases, because the outer layers of a grain carry more mycotoxin than the inner part, and removing the bran can lower the level in the flour, though it also removes nutrients.
Testing is the final check. Mycotoxins are measured by liquid chromatography joined to mass spectrometry, and by rapid tests that can be used at a buying point, such as a test strip that shows whether a lot is above a limit. Because mycotoxins are unevenly spread, a few grains in a sack can carry most of the poison, so sampling must take many increments from many places, just as in fraud detection. A test on a single handful can miss the poison entirely, or can overstate it, depending on which handful was taken.
Limits for mycotoxins exist in many places, set for each mycotoxin and each food, and foods above the limit cannot be sold for that use. The limits differ from place to place, and a food student should check the current limit in their own country rather than memorising one number. The principle is the same everywhere: keep the level as low as can reasonably be achieved, and keep it below the limit that protects health.
Questions to think about
- A sack of groundnuts has a few mouldy nuts among many clean ones. Why is cutting out the mouldy nuts not enough to make the sack safe?
- Grain is harvested wet and dried slowly over a week. What is the risk, and what should the farmer have done instead?
- Why does a mycotoxin test need many increments from a lot, rather than one handful?
Think about the groundnut question before you read on
The mycotoxin is a chemical made by the mould, and it spreads into the nut beyond the visible mould. The nuts that look clean may already carry the poison, and the mouldy nuts may have shed poison onto the nuts around them. Removing the visibly mouldy nuts lowers the level, but it does not prove the rest is safe: only a test on a representative sample can show the true level in the sack.
Heavy metals: elements that do not go away
Heavy metals are elements, such as lead, cadmium, mercury, and arsenic, that are harmful in amounts above a low level. They are called heavy because they are dense, and they matter for food because they do not break down: once a heavy metal is in soil or water, it stays there, and once it is in a food chain, it moves from soil to plant to animal to person, and builds up over time.
Lead comes from old paint, old pipes, petrol that once carried lead, mining, and some industrial waste. It reaches food through soil, through water carried in old pipes, and through dust settling on crops. Lead harms the brain and the nervous system, and children are the most harmed, because their brains are still growing. There is no safe level of lead that scientists can name with confidence, so the aim is to keep it as low as possible.
Cadmium comes from mining, from phosphate fertilisers, from industrial waste, and from some batteries. It is taken up by crops, more so by leafy vegetables, root crops, and grains grown on soils that carry it, and it builds up in the kidney over a lifetime. Cadmium harms the kidney and the bones, and long exposure is the problem, not a single meal.
Mercury comes from mining, from burning coal, and from industrial waste, and it reaches the sea and rivers, where microbes turn it into methylmercury, a form that builds up in fish. Large, long lived fish that eat other fish carry the most methylmercury, because each fish in the chain adds its load to the next. Methylmercury harms the nervous system, and the advice in many places is for pregnant people and young children to limit how often they eat the largest fish.
Arsenic comes from rocks and soil, from mining, and from some pesticides used in the past. It reaches food through water and through soil, and rice is the best known case: rice grows in flooded fields, and the water carries arsenic into the grain. Arsenic in its inorganic form harms health over long exposure, and rice, which is eaten every day by many people, is watched closely. Cooking rice in plenty of water and draining it can lower the arsenic, because some of it moves into the cooking water.
How heavy metals are measured and controlled
Heavy metals are measured by atomic absorption and by mass spectrometry joined to plasma, methods that can find tiny amounts of an element in a food. The laboratory must keep its own metals out of the sample, because metals are in dust, in glassware, and in reagents, and a dirty blank can give a false result. As with PFAS and microplastics, the blank and the detection limit are part of the answer.
Control of heavy metals is mostly about the land, the water, and the feed. Soil that carries lead or cadmium should not grow food crops, or should grow crops that take up less of the metal. Water used for irrigation and for food making should be tested. Feed given to animals should be tested, because metals in feed pass into meat, milk, and eggs. Industrial waste must be kept out of water and soil, and old pipes and old paint must be replaced where they touch food or water.
Food processing can help in some cases. Washing and peeling vegetables removes metals carried in soil on the surface. Milling can remove the outer layers of a grain that carry more metal. Cooking rice in excess water and draining it lowers arsenic. But processing cannot remove a metal that is inside the food, and it cannot fix a contaminated field. The real control is prevention, at the farm and at the source of the metal.
Limits for heavy metals in food exist in many places, set for each metal and each food, and they are getting tighter as methods improve and as the science of low level harm grows. As with mycotoxins, a student should know the principle and check the current limit in their own country, rather than memorising a number that will change.
The diagram shows the two routes into food. The steps point to the diagram: both contaminants are invisible, both build up over years, and both are beaten by prevention rather than by cleaning the finished food.
A worked example: estimating exposure to a heavy metal
A limit on paper means little until it is turned into an amount a person actually eats. The method for doing that is called an exposure estimate, and it uses three numbers: the concentration of the metal in the food, the amount of the food eaten each day, and the body weight of the person. Work through it step by step with practice numbers, chosen to show the method, not real data for any food.
Step one: take the concentration. Suppose a grain carries 0.2 milligrams of a metal per kilogram of grain. Write that as 0.2 mg per kg.
Step two: take the daily intake of the food. Suppose a person eats 0.3 kg of that grain each day. Multiply the concentration by the intake: 0.2 mg per kg times 0.3 kg per day gives 0.06 mg per day. The kilograms cancel, leaving milligrams per day.
Step three: divide by body weight. Suppose the person weighs 60 kg. Divide 0.06 mg per day by 60 kg to get 0.001 mg per kg of body weight per day. That is the exposure: the amount of the metal per kilogram of body weight, each day, from this food alone.
Step four: compare with the tolerable intake. Toxicologists set a tolerable daily intake for a metal, which is the amount per kilogram of body weight per day that can be eaten over a lifetime without expected harm. If the exposure from this food is well below that value, the food alone is not a concern. If the exposure is close to or above it, the risk must be looked at: other foods may add more of the same metal, and children, who weigh less and eat more per kilogram of body weight, will have a higher exposure from the same food.
Two lessons follow from the arithmetic. First, a food eaten in large amounts every day matters more than a food eaten rarely, even if the rarely eaten food carries a higher concentration. Second, the same concentration is a bigger risk for a child than for an adult, because the divisor, body weight, is smaller. This is why limits are set per food and why advice about large fish and rice is aimed at children and pregnant people first.
Why these contaminants sit beside fraud
Mycotoxins and heavy metals are not fraud: nobody adds them on purpose to make money. They sit in this subject because they share the work of fraud detection: they are invisible, they need laboratory testing to find, they need representative sampling, and they are controlled by knowing the supply chain and testing on a risk basis. A food business that tests for fraud but ignores mycotoxins and metals is testing only half the danger.
They also sit here because fraud can hide them. A diluted food can carry a mycotoxin from a cheap, damp lot mixed into a clean lot. A substituted ingredient can bring a heavy metal from a place nobody checked. A fraud defence that only asks whether the species is right will miss a contaminant that came in with the substitute. The best defence tests for both: is the food what it claims to be, and is it clean?
Practical note: Look at the grains, nuts, dried fruit, and spices in your kitchen. Check the storage: are they dry, cool, sealed, and out of light? Is any pack damp, open, or past its date? Smell a handful of grain or nuts: a musty smell is a warning that mould has grown, and a musty lot should not be eaten or cooked with, because cooking does not remove mycotoxins. Write down what you find and what you would change about the storage.
Quick self-check
Question: Why does cooking not make mycotoxin contaminated grain safe?
Because mycotoxins are heat stable chemicals. Cooking kills the mould, but the poison the mould already made survives the heat and stays in the food. The only control is to stop the mould growing, by drying and storing the grain well.
Question: Why do large fish carry more methylmercury than small fish?
Because methylmercury builds up along the food chain. Each small fish carries a little, and a large fish that eats many small fish over a long life adds their loads together. The larger and older the fish, the more methylmercury it has gathered.
Try it yourself
Take a bag of rice, a bag of flour, and a pack of nuts from your kitchen or a shop. For each one, write a storage plan: where would you keep it, in what container, at what temperature, and how would you check it each month? Then write a testing plan: which mycotoxin or metal would you test for, how often, and what would you do if the result came back above the limit? Compare your plans with a classmate: which food has the highest risk in your plan, and does the storage plan match the risk?
Key ideas to carry forward
- Mycotoxins are poisons made by moulds on grains, nuts, dried fruit, coffee, cocoa, and spices, and they survive cooking.
- Aflatoxins, ochratoxin A, fumonisins, deoxynivalenol, and patulin are the main mycotoxins to know.
- Moulds need warmth and moisture, so quick drying and dry, cool storage are the main controls.
- Cutting off visible mould does not remove the poison, because it spreads into the food beyond the fuzz.
- Heavy metals such as lead, cadmium, mercury, and arsenic do not break down and build up along food chains and in the body.
- Large fish carry more methylmercury, and rice grown in flooded fields can carry arsenic.
- Both groups are controlled by prevention, clean land, clean water, clean feed, and good storage, plus testing on a risk basis with representative sampling.
Glossary
Exposure estimate
A calculation of how much of a substance a person takes in each day, from its concentration in food, the amount eaten, and body weight. Food tech example: an exposure estimate shows how much arsenic a person takes in from eating rice each day.
Tolerable daily intake
The amount of a substance per kilogram of body weight per day that can be taken in over a lifetime without expected harm. Food tech example: an exposure estimate is compared with the tolerable daily intake to judge whether a metal in food is a concern.
Mycotoxin
A poison made by a mould growing on food. Food tech example: aflatoxin made by mould on groundnuts is a mycotoxin.
Aflatoxin
A mycotoxin made by Aspergillus moulds on maize, groundnuts, tree nuts, and spices, which damages the liver. Food tech example: aflatoxin in groundnut flour survives baking into biscuits.
Ochratoxin A
A mycotoxin made by Aspergillus and Penicillium moulds on cereals, coffee, cocoa, and dried fruit. Food tech example: ochratoxin A is tested in coffee beans stored damp.
Fumonisin
A mycotoxin made by Fusarium moulds on maize. Food tech example: fumonisin in maize flour is a concern where maize is eaten every day.
Deoxynivalenol
A mycotoxin made by Fusarium moulds on wheat and barley, also called vomitoxin. Food tech example: deoxynivalenol is tested in wheat after a wet harvest.
Patulin
A mycotoxin made by Penicillium moulds on apples and apple products. Food tech example: patulin is tested in apple juice made from bruised apples.
Aspergillus
A group of moulds that includes the makers of aflatoxins and ochratoxin A. Food tech example: Aspergillus grows on groundnuts stored warm and damp.
Fusarium
A group of moulds that includes the makers of fumonisins and deoxynivalenol on cereals. Food tech example: Fusarium infects wheat in the field in wet weather.
Heat stable
Not destroyed by normal cooking temperatures. Food tech example: aflatoxin is heat stable, so baking does not remove it from flour.
Lead
A heavy metal that harms the brain and nervous system, with no safe level that scientists can name. Food tech example: lead from old pipes can reach food made with that water.
Cadmium
A heavy metal that builds up in the kidney over a lifetime, taken up by crops from soil. Food tech example: cadmium is tested in leafy vegetables grown on certain soils.
Methylmercury
A form of mercury made by microbes in water, which builds up in fish along the food chain. Food tech example: large fish that eat other fish carry the most methylmercury.
Inorganic arsenic
A harmful form of arsenic that reaches rice through flooded fields. Food tech example: cooking rice in plenty of water and draining it lowers inorganic arsenic.
Moisture content
The amount of water in a food, which decides whether mould can grow in storage. Food tech example: grain dried to a low moisture content stays safe in storage.
References
- International Agency for Research on Cancer, Aflatoxins, IARC Monographs, 2012.
- Pitt, J. I. and Hocking, A. D., Fungi and Food Spoilage, 3rd edition, 2009.
- Food and Agriculture Organization, Mycotoxin Prevention and Control in Food Grains, 2001.
- World Health Organization, Lead, Cadmium, Mercury and Arsenic in Food, 2011.
- Kabata-Pendias, A., Trace Elements in Soils and Plants, 4th edition, 2010.
Unit 7: Building a Fraud Defence, Vulnerability Assessment
Note
Unit 7: Building a Fraud Defence, Vulnerability Assessment
A question to start with
A small jam maker buys fruit from three suppliers, sugar from one, and jars from another. The jam sells well, and no customer has ever complained. One day a laboratory test finds that the jam contains a cheaper fruit than the label claims. The jam maker did not add it: one of the suppliers did, and the jam maker never checked. Where should the jam maker have looked, what should have been tested, and how could the fraud have been stopped before it reached the jar? This unit builds a fraud defence: a practical system any food business can use, built around a vulnerability assessment.
A fraud defence is a set of habits and checks that make fraud hard to do, easy to find, and unprofitable when it is found. It is not a single test or a single document. It is a routine that runs all year: know what you buy, know who you buy it from, know where the weak points are, test the weak points, and act when a test or a number looks wrong. The method for finding the weak points is called a vulnerability assessment, and it is the core of this unit.
The assessment asks three questions about each ingredient and each supplier: how much money would fraud make, how easy would it be to do, and how likely is it to be found? A food that is dear, easy to fake, and rarely tested is a high vulnerability. A food that is cheap, hard to fake, and tested often is a low vulnerability. The business then spends its testing money on the high vulnerabilities first.
Step one: map what you buy
The first step is to list everything the business buys, ingredient by ingredient, and to note for each one the price, the supplier, the country or region it comes from, and the form it arrives in. A jam maker lists fruit pulp, sugar, pectin, jars, and labels. A bakery lists flour, oil, eggs, milk powder, and spices. The list is the map of the business, and without it no defence can be built, because a business cannot guard what it has not listed.
For each item on the list, the business notes the price gap: the difference between the price of the real ingredient and the price of the cheapest look-alike. A large gap means a large temptation. It also notes the form: a whole food is hard to fake, a powder or a liquid is easy to fake, and a ready mixed ingredient is easier still, because the buyer cannot see its parts. It notes the chain: how many steps does the ingredient pass through before it arrives, and how many of those steps are checked by anyone?
This mapping often surprises people. A business that thought it bought ten ingredients finds it buys forty, including flavourings, colours, and processing aids that arrive in small amounts and are rarely tested. A business that thought it had one supplier for an ingredient finds it has three, because a buyer switched suppliers when prices rose and never told anyone. The map is the first defence: it makes the invisible visible.
Step two: score the vulnerability
The second step is to score each ingredient for vulnerability, using plain questions and a simple scale. The common method scores three factors: opportunity, motivation, and control. Opportunity is how easy the fraud would be: is a look-alike available, is the ingredient sold as a powder, is the chain long? Motivation is how much money it would make: is the price gap large, is the ingredient dear, is the market under price pressure? Control is how likely the fraud would be found: is the ingredient tested, is the supplier audited, is the paperwork checked?
Each factor is scored low, medium, or high, and the three scores together give an overall vulnerability for that ingredient. An ingredient that scores high on opportunity and motivation and low on control is the first priority for action. An ingredient that scores low on all three can be checked less often, which saves money for the high priorities.
The scoring is done by people who know the business: the buyer, the quality manager, the production manager, and the laboratory, sitting together and arguing honestly about each score. The argument matters more than the number, because it surfaces what each person knows: the buyer knows the price pressure, the production manager knows the form, the laboratory knows what can be tested. A score written by one person alone is weaker than a score argued over by the team.
The assessment is repeated regularly, at least once a year, and after any change: a new supplier, a new ingredient, a price rise, a fraud report in the news, or a failed test. A vulnerability that was low last year can be high this year, because prices move and fraudsters move with them.
Step three: choose the controls
For each high vulnerability ingredient, the business chooses controls: actions that make the fraud harder, easier to find, or less profitable. Controls fall into four groups.
Supplier controls come first. Buy from suppliers who can show where their food comes from, who test their own products, and who will share test results. Visit the supplier, or ask for an audit by a trusted body. Keep more than one supplier for a high risk ingredient where possible, so that a problem with one does not stop the business, and so that the two suppliers can be compared. Agree in writing what the ingredient must be, and what test will prove it.
Testing controls come second. Test the high vulnerability ingredients on arrival, using the methods from Unit 3, and test them on a plan that covers all suppliers and all seasons. Use screening for many samples and confirmation for strange ones. Keep the testing plan secret from suppliers where the law allows, or at least vary the timing, so that a fraudster cannot keep the tested deliveries honest and sell the rest as fake.
Paper controls come third. Run mass balance checks: compare the amount of raw material bought with the amount of finished product sold, and look for gaps. Check supplier paperwork against the product itself: does the delivery note match the batch, the date, and the amount? Look for paperwork that is always perfect, because real paperwork has small errors, and paperwork that is too perfect may have been written after the event.
People controls come fourth. Train staff to spot the warning signs: a price that cannot be honest, a supplier who changes their story, a delivery that arrives in different packaging, a product that tastes or looks different from usual. Give staff a way to report a worry without fear, and act on reports quickly. Many frauds are found by a worker who noticed something small and said so.
The diagram shows the fraud defence cycle. The steps point to the diagram: the cycle repeats, because prices move, suppliers change, and fraudsters adapt, so a defence built once and never reviewed soon stops working.
Step four: act on failures
A defence that finds a fraud and does nothing is worse than no defence, because it teaches the fraudster that testing has no teeth. Every failed test, every mass balance gap, and every credible report must lead to action: hold the affected product so that it cannot be sold, trace the ingredient back through the chain to find where the fraud entered, tell the supplier and ask for an explanation, and record what happened and what was done.
The action must be proportionate. A single failed test on one delivery may mean holding that delivery and testing more from the same supplier. A pattern of failures may mean stopping that supplier and finding another. A fraud that reaches customers may mean a recall, telling buyers what happened and what to do. The record of each action matters, because it shows the next auditor, the next customer, and the next regulator that the business takes fraud seriously and acts on what it finds.
The business should also share what it learns, where the law allows. Fraud intelligence, the knowledge of which foods are being faked and how, is more useful when it is shared between businesses, through trade groups and food safety networks. A fraud found by one business and shared with others can be stopped before it reaches them. A fraud found and kept secret can spread.
Enforcement, penalties, and reporting fraud
A fraud defence inside a business is only half the story. The other half is enforcement outside it: the inspectors, the laboratories, the courts, and the penalties that make fraud a bad business decision. Enforcement works in layers, and a food student should know what each layer does.
The first layer is inspection. Food inspectors visit factories, shops, and markets, check records, take samples, and look for the signs this subject has taught: prices that cannot be honest, papers that do not add up, and lots that cannot be traced. An inspector can hold a suspect lot, stop a line, and send samples to an official laboratory. The power to hold food on the spot is a strong tool, because a fraudster's profit depends on the food reaching the shelf.
The second layer is the laboratory result and the legal case. If the official laboratory confirms fraud, the business may face a fine, a ban on selling, a requirement to recall the food, or, in serious cases, prosecution in court. Penalties differ widely between places: some set fines as a fixed amount, others scale the fine to the profit made or the amount sold, and some add prison for fraud that harms health. The exact penalties are a matter for the law of the country, and a student should check the current rule in their own country rather than memorising a figure from another place.
The third layer is the market penalty, which is often larger than the legal one. A business caught in fraud loses customers, loses contracts, and loses the trust that took years to build. A buyer who finds fraud in one product will doubt every product from that seller. This is why honest businesses welcome enforcement: a fraudster who is caught and punished stops undercutting them on price.
The last part of enforcement is reporting. Fraud is often found by someone inside the chain: a worker who sees the trick, a driver who notices a swap, or a buyer who tests a sample. Safe reporting routes, where a worker can report without losing their job, are part of a healthy food system, and trade groups and food safety networks also collect fraud reports so that one business's discovery can warn the rest. A student who sees fraud and says nothing leaves the next buyer to be cheated in the same way.
Practical note: Look up the food enforcement body in your own country and write down three things: what it is called, what powers its inspectors hold, and where a member of the public reports suspected food fraud. Keep that note with your fraud defence: knowing who to call is part of the defence.
Keeping the defence alive
A fraud defence dies in three ways: it is written and then forgotten, it is starved of money, or it is run by one person who leaves. To keep it alive, the business should put the assessment on the calendar, give it a named owner, and give it a budget that survives a quiet year. A quiet year with no fraud found is a success, not a reason to cut testing: it may mean the defence is working, or it may mean the fraud has moved to a food nobody is testing.
The defence should also be tested itself. A business can run a blind test: send a known fake sample to its own laboratory, or ask a supplier to send a sample that the laboratory must judge without knowing its source. If the laboratory misses the fake, the defence has a hole, and the hole should be fixed before a real fraudster finds it.
Finally, the defence must stay honest about its limits. No defence catches every fraud, and no test is perfect. A business that claims its food is fraud proof is making a claim it cannot support. The honest claim is narrower and stronger: we know our weak points, we test them on a plan, we act on what we find, and we keep improving. That claim can be proved, and buyers can trust it.
Questions to think about
- A jam maker buys fruit pulp from a new supplier at a price far below every other supplier. What should the jam maker do before buying?
- Why should the timing of testing be varied, rather than testing the same delivery every month?
- A business finds no fraud for three years and cuts its testing budget. What is the risk in that decision?
Think about the new supplier question before you read on
The jam maker should ask how the low price is possible, ask for the origin and the test results, test a sample before buying in bulk, and buy a small trial amount first. A price far below the cost of the real food is a warning sign, and a new supplier with no track record deserves more testing, not less, because the business has no history to trust.
What a small business can do with little money
A small business cannot run a full fraud laboratory, but it can run a real defence with little money. The cheapest controls are the most useful: know your suppliers, buy from people you can visit, keep the chain short, refuse prices that cannot be honest, check the paperwork, run mass balance checks, and train staff to notice and report. Testing can be bought from a commercial laboratory for the highest risk ingredients only, and the cost of a few tests a year is small beside the cost of a fraud that reaches customers.
Joining a trade group or a testing scheme spreads the cost further: many small businesses testing together can afford a programme that none of them could afford alone, and they can share fraud intelligence. A small business that does these things is better defended than a large business that buys testing but never talks to its suppliers or checks its paperwork.
Practical note: Pick one food business you know, a bakery, a jam maker, a spice seller, or a restaurant, and write a one page vulnerability assessment for its three dearest ingredients. For each ingredient, score opportunity, motivation, and control as low, medium, or high, and name one control for each high score. End with a note on which ingredient you would test first and why. Keep the page with your fraud profile and your case file: together they are a starter fraud defence for a real business.
Quick self-check
Question: What three factors does a vulnerability assessment score?
Opportunity, how easy the fraud would be; motivation, how much money it would make; and control, how likely it would be found. An ingredient that scores high on opportunity and motivation and low on control is the first priority.
Question: Why is a mass balance check a cheap and useful control?
Because it uses numbers the business already has: what was bought and what was sold. A gap between the two shows that something was added or swapped, even when every finished pack passes its tests, and it costs nothing more than a careful look at the records.
Try it yourself
Choose a food you buy often, such as olive oil, honey, coffee, or a spice. Write a short vulnerability assessment for it as if you were the buyer for a small shop: what is the price gap, how easy is it to fake, how is it usually tested, and who sells it to you? Then write three controls you would use if you ran the shop: one supplier control, one testing control, and one paper control. Bring your assessment to class and compare it with a classmate who chose a different food: which food is the most vulnerable, and do your controls match the risk?
Key ideas to carry forward
- A fraud defence is a routine: map what you buy, score the vulnerability, choose controls, test, act, and review.
- Vulnerability is scored on opportunity, motivation, and control, and testing money goes to the high scores first.
- Controls cover suppliers, testing, paperwork, and people, and all four groups matter.
- Testing should be varied in timing, and strange results must be confirmed by a second method.
- Every failure must lead to action: hold, trace, fix, and record, or the defence teaches fraudsters that testing has no teeth.
- Mass balance checks and staff reports are cheap controls that catch frauds instruments miss.
- A small business can run a real defence with little money by keeping chains short, knowing suppliers, and testing only the highest risks.
Glossary
Enforcement
The work of inspectors, laboratories, and courts that finds fraud, stops suspect food, and punishes those caught. Food tech example: enforcement holds a suspect lot of honey and fines the seller who diluted it.
Whistleblowing
Reporting wrongdoing seen inside a business, through a route that protects the person who reports it. Food tech example: a worker who reports a swap of cheap oil into olive oil is whistleblowing.
Vulnerability assessment
A structured review that finds which ingredients and suppliers are most open to fraud, by scoring opportunity, motivation, and control. Food tech example: a jam maker scores fruit pulp as high vulnerability because it is dear, easy to dilute, and rarely tested.
Opportunity
How easy a fraud would be to carry out, such as whether a look-alike is available and whether the chain is long. Food tech example: a powder ingredient gives high opportunity, because a filler is easy to mix in.
Motivation
How much money a fraud would make, driven by the price gap between the real food and the fake. Food tech example: a large price gap between real saffron and dyed corn silk gives high motivation.
Control
How likely a fraud would be found, through testing, audits, paperwork checks, and staff reports. Food tech example: an ingredient tested on every delivery has strong control.
Mass balance
A check that compares the amount of raw material bought with the amount of finished product sold. Food tech example: a bakery that buys little flour but sells much bread fails a mass balance check.
Fraud defence
The set of habits and checks that make fraud hard to do, easy to find, and unprofitable. Food tech example: a spice seller who tests every delivery and knows every supplier has a fraud defence.
Supplier audit
A visit or a check by a trusted body to confirm that a supplier makes what it claims and keeps proper records. Food tech example: an audit of a honey supplier checks the hives, the extraction, and the sales records.
Blind test
A test where the laboratory does not know the source of the sample, used to check that the laboratory itself can find a fake. Food tech example: a known diluted honey sent blind to the laboratory tests whether the method works.
Fraud intelligence
Knowledge about which foods are being faked and how, shared between businesses and safety networks. Food tech example: a trade group warns its members that a certain spice is being faked with a new dye.
Recall
Removing a food from sale and telling buyers what to do, when the food is found to be wrong or unsafe. Food tech example: a ready meal found to contain the wrong meat may be recalled from shops.
Traceability
The ability to follow a food back through its supply chain to its origin, using records at each step. Food tech example: a batch number on a jar of jam lets a maker trace its fruit pulp to the supplier.
Processing aid
A substance used during making that is not meant to stay in the finished food, such as a filter aid or an antifoam. Food tech example: a processing aid used in juice making is listed in the business map even though it is not on the label.
Screening
Fast, cheap testing of many samples to flag the strange ones for slower confirmation. Food tech example: spectroscopy screens every delivery of spice, and only strange deliveries go on to chromatography.
Confirmation
A slow, precise test that proves what a flagged sample really is. Food tech example: DNA testing confirms a meat that spectroscopy flagged as strange.
References
- Spink, J. and Moyer, D. C., Defining the Public Health Threat of Food Fraud, 2011.
- Manning, L., Food Fraud: Policy and Management, 2016.
- Food and Agriculture Organization, Food Fraud, 2021.
- British Standards Institution, PAS 96, Defending Food and Drink, 2017.
- van Ruth, S. M., Huisman, W. and Luning, P. A., Food Fraud Vulnerability and Its Key Factors, 2017.
Unit 8: Food Authenticity and Traceability Systems
Note
Unit 8: Food Authenticity and Traceability Systems
A question to start with
A jar of honey arrives at a packing plant with papers saying it came from a named valley, from named beekeepers, in a named season. The honey tests pure. The price is fair. Yet the packing plant cannot show, step by step, how that jar travelled from the hive to the shelf. Is the honey authentic? The test says the product is genuine, but the story of the product has a gap, and a gap in the story is where fraud lives. This unit is about closing that gap: the systems that prove a food is what it claims to be, not by testing the food once, but by following it all the way.
Authenticity is a claim about identity: this food is the food named on the label, from the place named on the label, made in the way named on the label. Traceability is the record that backs the claim: the ability to follow a food forward to the buyer and back to the origin, one step at a time. Testing can prove what a sample is. Only traceability can prove where a whole lot came from, and only traceability can find every pack that must be recalled when something goes wrong.
This unit explains the main traceability systems: the one-step-back, one-step-forward rule, batch and lot codes, mass balance, chain of custody, and digital tools including blockchain. It also explains where each system works, where each fails, and how a food business builds a system that a fraudster cannot easily beat.
One step back, one step forward
The foundation of every traceability system is simple: every business in the chain must be able to name the business it bought from and the business it sold to. This is called one step back, one step forward. A miller knows which farms supplied the wheat and which bakeries bought the flour. A bakery knows which miller supplied the flour and which shops bought the bread. Put the steps together and the whole chain appears, from field to shelf.
The rule sounds easy, and for a short chain it is easy. A farm shop that sells its own jam needs only its own records. But a long chain is harder: a ready meal may hold twenty ingredients from ten countries, each with its own chain behind it. The meal maker must hold the records of every ingredient, and every supplier must hold the records behind them. When one link keeps poor records, the chain breaks at that link, and the food past that link cannot be traced.
Records must also be kept in a form that can be read quickly. Paper records in a filing cabinet can prove a chain in a week, but a recall needs the answer in hours. This is why the first improvement most businesses make is not a new technology but a tidy record: every delivery logged on the day it arrives, with the supplier name, the date, the lot code, and the amount, in one place where it can be found fast.
Questions to think about
- A bakery buys flour from three millers and mixes it in one silo. Can it trace a loaf back to one farm? Why or why not?
- Why does a recall need traceability records that can be read in hours, not weeks?
- A supplier keeps perfect records but stores them on paper in a locked office. What happens on a weekend when a recall starts?
Think about the mixed flour question before you read on
No, not to one farm. Once flour from three millers is mixed in one silo, the identity of each delivery is lost, and a loaf can only be traced back to the three millers as a group. To trace to one farm, the bakery would need to keep the deliveries separate, or to buy from one miller at a time, or to accept that the trace stops at the group. Mixing is honest and common, but it has a cost: the trace gets wider and less precise.
Batch codes, lot codes, and the identity of a lot
Traceability works on lots, not on single packs. A lot is a quantity of food made or packed under the same conditions, from the same materials, at the same time. Every pack in a lot shares one lot code, printed on the pack, and that code links the pack to the records of the lot: what went into it, when it was made, and where it went.
Lot size is a choice with real consequences. A small lot means a recall pulls few packs and a fraud is confined to a small amount of food, but small lots mean more codes, more records, and more work. A large lot is cheaper to run, but a problem in one ingredient can force a recall of a huge amount of food, and a fraud in one delivery can spread through the whole lot. Good businesses choose lot sizes that balance cost against risk, and they keep the high-risk ingredients in smaller, separate lots.
Lot codes also need to survive the chain. A code printed on a bulk sack is useless if the sack is emptied into a silo and the code is thrown away with the sack. The receiving business must record the code before the sack is opened, and link it to the new lot it joins. Every time a lot is split, mixed, or repacked, a new link must be written down. The chain is only as strong as its weakest link, and the weakest link is often a busy receiving dock where a code is not recorded.
Mass balance: the arithmetic of honesty
Mass balance is the simplest and cheapest fraud test in the whole subject, and it needs no laboratory. It compares what went into a business with what came out. A cheese maker who buys 10,000 litres of milk in a month cannot honestly sell cheese that needed 15,000 litres of milk, unless milk was also bought from somewhere else and recorded. A spice grinder who buys 1,000 kg of whole chillies cannot sell 1,500 kg of chilli powder. When the outputs are larger than the inputs, something was added, and the numbers say so even when the product passes every laboratory test.
Mass balance works because fraud usually adds bulk: water in milk, syrup in honey, cheap oil in olive oil, starch in spice. Each of those tricks makes the output bigger than the input, and the arithmetic catches it. The check needs three numbers for each material: the amount bought, the amount used, and the amount sold, with the normal losses of processing taken into account. A careful mass balance allows for real losses, such as water lost in drying or trimmings removed in cutting, so that honest businesses are not accused by mistake.
The check also works across a chain. An auditor can ask a supplier how much of a named ingredient it bought, and compare that with how much it sold as finished product to all its buyers. If the supplier sold more of the ingredient than it bought, some of its product was not what it claimed to be. This is why fraud is often found by auditors reading papers, not by analysts reading instruments, and why a business that keeps clean mass balance records is a hard target for fraud.
Practical note: Try a mass balance check on a simple food you make at home, such as jam. Weigh the fruit and sugar that go in, weigh the jam that comes out, and note the loss from evaporation. Then imagine a seller who claims to have made twice as much jam from the same fruit. The arithmetic shows the claim cannot be true, and no taste test is needed to prove it.
Chain of custody: who held the food, and when
Chain of custody is the record of every hand a food passed through, and every change made to it, from origin to sale. For a high value food, the chain of custody may include the farm, the collector, the processor, the packer, the shipper, and the seller, with a signed record at each handover. The purpose is to make substitution hard: a food that changes hands without a record can be swapped for a cheaper look-alike at that step, and nobody can prove where the swap happened.
Chain of custody matters most for foods whose value sits in a claim that cannot be tested in the product: a place name, a farming method, a breed, or a catch area. A test can show that a fish is cod, but no test can show that the cod was caught in the named sea, or by the named method. Only the chain of custody can support that claim, with records from the boat, the landing, and the processor that match each other.
The weak points of a chain of custody are the handovers: the moment a food leaves one business and enters another. Seals on containers, signed delivery notes, and matching lot codes at both ends all help, but they only work if both sides check. A seal that nobody inspects is decoration, and a delivery note that nobody reads is paper. The strongest chains are those where the buyer checks the seal and the code before accepting the delivery, and refuses a delivery whose papers do not match.
Digital traceability: from paper to shared records
Digital traceability replaces paper records with electronic records that can be searched, shared, and checked in minutes. The simplest form is a shared spreadsheet or a supplier portal where each business enters its deliveries and sales. More developed systems use barcodes or QR codes on each case, scanned at each step, so that the movement of a lot is recorded automatically as it travels.
Digital records bring two gains: speed and search. A recall that took days with paper can take hours with a database, because the system can find every pack from a lot at once, across all buyers. And a buyer can search a supplier's history in seconds, to check whether a supplier's sales match its purchases, which is a mass balance check done by computer.
Digital records also bring two risks. The first is that a digital record is only as honest as the person who entered it: a computer cannot tell whether the lot code typed in was the real one. The second is that different businesses use different systems, and the systems may not talk to each other. A chain in which each business keeps a perfect digital record in its own format, which no other business can read, is no better than a chain of paper. This is why the food world is working toward shared formats and shared rules for what a traceability record must hold, so that one business can read the record of the next.
Blockchain, and what it can and cannot do
Blockchain is a way of keeping shared digital records that many businesses can add to but none can quietly change. Each new record is linked to the one before it, and copies of the whole record are held by many parties, so that changing an old entry would mean changing every copy. For food, the idea is a shared ledger of a food's journey: each handover is entered once, seen by all, and cannot be rewritten later.
Blockchain can help traceability in real ways. It makes the record harder to alter after the event, which matters when a fraudster's trick is to rewrite the papers. It gives every party the same view of the chain, so that a buyer can check a claim without asking the seller for a favour. And it can carry a lot of detail, from lot codes to test results to temperature readings, in one place.
But blockchain cannot do the thing people most often expect of it: it cannot prove that the food in the box is the food in the record. A dishonest supplier can enter an honest record for a genuine lot, then ship a different lot in the same box. The ledger will be perfect and the food will be fake, because the ledger records what was typed, not what was shipped. This is the central limit of every digital system, and blockchain is no exception: the record is only as true as the step where the food meets the record, which is usually a person typing a code, scanning a box, or sealing a container.
Blockchain also costs money and effort: every party must join the same system, enter data in the same way, and keep the system running. For a short chain with trusted partners, a simple shared spreadsheet may give the same result for far less cost. The right question about blockchain is not whether the technology is impressive, but whether the chain has a problem that only a shared, unchangeable record can solve: many parties, low trust, and high value at stake. When those three are present, blockchain can be worth it. When they are not, simpler tools do the job.
The diagram shows one lot moving through the chain, with a record written at each step. If any step is missed, the trace stops there.
Building a traceability system that works
A traceability system that works starts with the food, not the software. The business first lists its ingredients, ranks them by fraud risk and by value, and builds the strongest trace for the riskiest few. A spice buyer traces its saffron and vanilla lot by lot; it may trace its salt by supplier only. Trying to trace everything to the same depth is costly and usually fails, while tracing the riskiest ingredients well gives most of the protection.
Next comes the discipline of the record: one person responsible for logging deliveries, one place where records live, one format for lot codes, and a rule that no delivery is accepted without its papers. Then comes the test: the business runs a mock recall, picks a lot at random, and times how long it takes to find every pack from that lot and every buyer who received it. A system that cannot answer in hours is not ready, however good it looks on paper.
The last step is to connect the trace to the fraud defence. Traceability records feed the mass balance check, the supplier audit, and the laboratory testing plan. A supplier whose papers do not add up gets more testing. A lot whose chain has a gap gets held until the gap is closed. In this way traceability is not a filing exercise but a working part of fraud defence: it tells the business where to look, and it proves the answer when the business finds something.
Quick self-check
Question: Why is mass balance a fraud test, when it uses no laboratory at all?
Because most fraud adds bulk, so the outputs become larger than the inputs. A business that sells more finished product than its raw materials could have made must have added something, and the arithmetic shows it even when the product passes its tests.
Question: A food is tracked on a blockchain from farm to shop. Can the buyer be sure the food is genuine?
No. The blockchain proves the record was not changed after it was entered, but it cannot prove that the food in the box matches the record. A dishonest supplier could enter a true record and ship a different lot. The record is only as true as the step where the food meets the record.
Try it yourself
Pick a food from your kitchen with a claim on its label: a place name, a farming claim, or a single origin. Write down every step you think it took from its origin to your kitchen, and every record that would be needed to prove each step. Then look for the weakest step: the handover where a swap would be easiest, and the record that would be hardest to check. Write down one change, such as a smaller lot size, a seal check on delivery, or a mass balance check on the supplier, that would make that step stronger.
Key ideas to carry forward
- Authenticity is a claim about identity, and traceability is the record that backs the claim.
- One step back, one step forward is the foundation: every business names its supplier and its buyer.
- Lot codes link each pack to the records of its lot, and lot size balances cost against the size of a recall.
- Mass balance compares inputs with outputs and catches fraud that adds bulk, with no laboratory needed.
- Chain of custody records every hand a food passed through, and matters most for claims no test can prove.
- Digital records bring speed and search, but they are only as honest as the person who entered them.
- Blockchain makes a shared record hard to change, but it cannot prove the food in the box matches the record.
- A working system starts with the riskiest ingredients, keeps one tidy record, and proves itself with a mock recall.
Glossary
Authenticity
The quality of a food being truly what its label claims it is, in kind, origin, and method. Food tech example: authenticity of olive oil means the oil is really olive oil from the named place, not a blend.
Traceability
The ability to follow a food forward to its buyers and back to its origin, one step at a time, using records at each step. Food tech example: a batch number on a jar of honey lets a buyer trace it back to the beekeeper.
One step back, one step forward
The rule that every business must be able to name its supplier and its buyer. Food tech example: a miller names the farms that supplied its wheat and the bakeries that bought its flour.
Lot
A quantity of food made or packed under the same conditions, from the same materials, at the same time, sharing one lot code. Food tech example: all jars of jam filled on one morning from one kettle form one lot.
Lot code
The printed code on a pack that links it to the records of its lot. Food tech example: the lot code on a spice jar lets an inspector find the batch, the supplier, and the test results.
Mass balance
A check that compares the amount of raw material bought with the amount of finished product sold. Food tech example: a cheese maker who buys little milk but sells much cheese fails a mass balance check.
Chain of custody
The record of every hand a food passed through, and every change made to it, from origin to sale. Food tech example: the chain of custody for a named catch shows the boat, the landing, and the processor in order.
Mock recall
A practice recall in which a business times how fast it can find every pack from one lot and every buyer who received it. Food tech example: a mock recall of one spice lot shows whether the records answer in hours or in days.
Blockchain
A way of keeping shared digital records that many parties can add to but none can quietly change, because each record links to the one before it. Food tech example: a blockchain ledger records each handover of a coffee lot from farm to roaster.
QR code
A printed square code that a phone or scanner can read to open a record or a web page. Food tech example: a QR code on a honey jar opens the traceability record for that lot.
Supplier portal
A shared online place where suppliers enter deliveries and buyers read them. Food tech example: a spice buyer checks a supplier's deliveries and test results in the supplier portal before ordering.
Handover
The moment a food passes from one business to another in the chain. Food tech example: the handover of a bulk oil delivery is checked by reading the seal and the lot code before acceptance.
Recall
The removal of a food from sale and from buyers because it is unsafe or wrongly described. Food tech example: a recall of one lot of milk powder pulls every pack with that lot code from shops.
Seal
A fastening on a container that shows whether it was opened in transit. Food tech example: a broken seal on a drum of oil tells the buyer the delivery must be checked before use.
Shared format
An agreed way of writing traceability records so that different businesses can read each other's records. Food tech example: a shared format for lot codes lets a packer read a farmer's delivery record without retyping it.
References
- Spink, J. and Moyer, D. C., Defining the Public Health Threat of Food Fraud, Journal of Food Science, 2011.
- Manning, L., Food Fraud: Policy and Management, 2016.
- Nielsen, S. S., Food Analysis, 5th edition, 2017.
- Downey, G., Food Authenticity and Traceability, 2003.
- Ellis, D. I., et al., Fingerprinting Food: Current Technologies for the Detection of Food Adulteration and Contamination, 2012.
Unit 9: Other Emerging Contaminants, Mineral Oils, Acrylamide, and Process Contaminants
Note
Unit 9: Other Emerging Contaminants, Mineral Oils, Acrylamide, and Process Contaminants
A question to start with
A loaf of bread, a box of breakfast cereal, and a bag of crisps sit on a shelf. None of them is adulterated. Nobody added anything to cheat the buyer. Yet each one may carry substances that were never meant to be in food: oils that moved from a printed box, a chemical that forms when potatoes are fried, and compounds that appear when food is heated, smoked, or refined. How do substances that nobody added on purpose end up in food, and what can the food world do about them? This unit is about those uninvited guests: the emerging contaminants that form during processing or slip in from packaging and the environment.
The word emerging does not mean new in the sense of just invented. It means newly noticed, newly measured, or newly understood. Many of these substances have been in food for years, but better instruments now find them at very low levels, and better science now explains what they do. The food world must then decide what to do: set limits, change a process, change a material, or wait for more evidence. That decision is harder than it looks, because the same process that creates a contaminant often creates the flavour, colour, or safety of the food itself.
This unit covers three groups: mineral oil hydrocarbons that migrate from packaging and processing, acrylamide that forms when starchy foods are cooked at high heat, and the wider family of process contaminants that form during heating, smoking, and refining. It explains where each one comes from, how it is measured, and what can be done to keep it low without losing the food.
Mineral oil hydrocarbons: oils that should not be in food
Mineral oil hydrocarbons, known as MOH, are mixtures of hydrocarbons that come from crude oil. They reach food in several ways. Printed paper and board packaging, especially recycled board, can carry mineral oils from printing inks and from the recycled fibres themselves. The oils move from the board into dry foods stored against it, such as rice, flour, pasta, and breakfast cereals. Lubricants used on food machinery, release agents, and coatings can also leave traces, and some foods pick up mineral oils from the environment, such as from exhaust or from jute sacks treated with oil.
MOH is not one substance but two main groups. The saturated group, MOSH, is made of straight and branched chains, and the aromatic group, MOAH, holds ring-shaped molecules. The two groups behave differently in the body, and the aromatic group is the one that raises more concern, because some of its members can damage genetic material. This is why the food world watches MOAH closely even at low levels, and why a test result must name which group was found, not just a total number.
Measuring MOH is hard work. The oils are mixtures of hundreds of similar molecules, and the laboratory must separate the two groups and measure each one, at very low levels, without adding contamination from its own instruments and solvents. A careless laboratory can contaminate its own sample with oil from a plastic cap or a solvent bottle, so MOH testing needs clean methods and blank samples that prove the laboratory itself added nothing.
Questions to think about
- A dry food is stored for months in a printed recycled board box. Which route would mineral oil take into the food, and why does storage time matter?
- Why must a laboratory report MOSH and MOAH separately, instead of one total mineral oil number?
- A factory changes its board supplier to one that uses a barrier layer. What should it test before and after the change, and why?
Think about the storage time question before you read on
Mineral oil moves slowly from the board into the food, through the air inside the pack and by direct contact, so a longer storage time lets more oil move. A dry food with a large surface area, such as cereal flakes or pasta, picks up more oil than a solid block, and warm storage speeds the movement. This is why long-stored dry foods in printed board are the classic case, and why a barrier layer or an inner bag can break the route.
How the food world keeps mineral oils low
The first defence against MOH is the pack. A barrier layer between the board and the food, such as a plastic or coated inner bag, or a board with a functional barrier, stops the oil from reaching the food. Virgin fibre board carries less mineral oil than recycled board, because the oil in recycled board comes from the inks and materials of its earlier life. Choosing virgin fibre for direct food contact, or using recycled board only with a proven barrier, is the standard answer for dry foods.
The second defence is the process. Food grade lubricants, chosen and used so that they cannot touch the food, replace general purpose oils on machinery near the line. Release agents and coatings are checked for mineral oil content. Jute sacks and transport materials are checked too, because a sack that carried oil can pass it to the food inside.
The third defence is testing and supplier control. Buyers of dry foods test high-risk products for MOH, set their own limits in buying specifications, and ask board suppliers for evidence that the board is low in mineral oil or carries an effective barrier. As with other emerging contaminants, the limit is a moving target: as instruments improve and evidence grows, the levels that buyers and regulators accept tend to fall, so a food business that builds low-MOH packaging now is ahead of the rule that may follow.
Acrylamide: the chemical that cooking creates
Acrylamide is a chemical that forms when starchy foods are cooked at high temperature, above about 120 degrees Celsius, in frying, baking, roasting, and toasting. It forms in a reaction between a natural amino acid called asparagine and sugars, in the browning reactions that give fried potatoes, baked bread crust, and roasted coffee their colour and flavour. Acrylamide is not added to food, and it cannot be removed once formed. It is a product of the cooking itself.
The foods that carry the most acrylamide are those that are starchy, cooked hot, and cooked until brown: crisps and chips, roast potatoes, bread crust, breakfast cereals, biscuits, and coffee. The amount depends on the raw material, the temperature, and the time. Potatoes stored cold build up more sugars, and those sugars make more acrylamide when fried. A darker fry makes more acrylamide than a lighter fry. A thin crisp, cooked right through, makes more than a thick chip with a soft centre.
Acrylamide matters because laboratory studies show it can damage genetic material and raise cancer risk in animals at high doses. The doses in food are far lower than the doses in those studies, so the risk to a person eating normal amounts is not settled by a single number. Regulators treat acrylamide as a substance to keep as low as reasonably achievable, which means the food world must reduce it where it can, without banning fried and baked foods outright.
Keeping acrylamide low without losing the food
Reducing acrylamide is a balancing act, because the same browning that makes acrylamide also makes flavour, colour, and crispness. The most useful lever is the raw material. Choosing potato varieties low in sugars, storing potatoes above the temperature that builds up sugars, and avoiding potatoes that have sprouted or gone green all cut acrylamide before cooking starts. Soaking cut potatoes in water before frying washes out some sugars and some asparagine, and lowers the acrylamide that forms.
The second lever is the cooking itself. Frying or baking to a light golden colour instead of a dark brown colour cuts acrylamide sharply, and lowering the temperature while giving a little more time can keep the texture while making less of the chemical. Thicker cuts make less acrylamide than thin cuts, because less of the food reaches the highest temperature. In bread, a lighter crust and a shorter bake at the end make less acrylamide, and in coffee the roast level sets the amount, with darker roasts carrying more.
The third lever is the recipe. Adding an enzyme that breaks down asparagine before cooking, known as asparaginase, is used in some baked and fried foods, and it can cut acrylamide without changing the flavour much. Changing the raising agent in biscuits, or adding a small amount of an acid ingredient, can also lower the amount formed. Each of these steps must be tested in the real product, because a change that cuts acrylamide but ruins the texture will not survive on a shelf.
Other process contaminants: heat, smoke, and refining
Acrylamide is the best known process contaminant, but it is not the only one. When fats and oils are heated and refined, compounds called 3-MCPD esters and glycidyl esters can form. They appear in refined vegetable oils and in foods made with them, such as infant formula, biscuits, and fried foods. The refining step that makes an oil clean, pale, and stable is the same step that can create these compounds, so refiners have had to change temperatures and methods to keep them low.
Smoking and direct drying can add polycyclic aromatic hydrocarbons, known as PAH, to foods. These form when fat drips onto a hot surface and the smoke carries the compounds back onto the food, or when food is dried over an open fire. Traditional smoked fish and meat, and foods dried over direct heat, carry the most PAH. The controls are the smoking method itself: indirect smoking, where the smoke is cooled and cleaned before it reaches the food, and keeping the food away from the flame, both cut PAH without losing the smoked character.
Furan is another process contaminant, formed when foods are heated in sealed containers, such as canned and jarred foods. It forms from sugars, amino acids, and fats during the heat treatment that makes the food safe. The amounts are low, and the control is again the process: gentler heating where safety allows, and recipes that form less furan. In each case the pattern is the same: the contaminant is born in a step that the food needs, so the answer is to adjust the step, not to remove it.
The diagram shows the order of control for process contaminants: start with the raw material, then storage, recipe, and process, and test at the end.
Measuring, limiting, and talking about process contaminants
Measuring a process contaminant means measuring a small amount of a chemical in a complex food, and the methods are the same family as fraud detection: chromatography joined to mass spectrometry, run by a laboratory that can prove its blank samples are clean. The results are reported against a benchmark or a limit, where one exists, and against the levels the business has seen in its own products over time. A single high result is a signal to check the batch, not a verdict on the whole product, and a trend over months tells more than one number.
Limits for process contaminants are set with care, because a limit that no producer can meet would remove foods from the shelf, while a limit set too high would change nothing. Regulators often use benchmark levels: values that most producers already meet, which push the worst performers to improve without punishing the rest. Food businesses also set their own internal limits, usually tighter than the public benchmark, so that a normal batch stays well inside the rule.
Talking about process contaminants is delicate. A headline that a food contains a chemical that can cause cancer in animals will frighten buyers, even when the amount in food is far below the doses in the studies. Honest communication states three things: the amount found, the amount in the studies, and what is being done to lower it. A food business that can show its own test results, its own trend over time, and the steps it has taken earns trust that a bare denial cannot buy.
Practical note: Fry two batches of potato chips at home from the same potatoes: one fried until light golden, the other until dark brown. Taste the difference, and note the colour difference. The darker batch carries more acrylamide, and the colour you can see is a rough guide to the chemical you cannot see. Light golden is the target in the factory too.
Quick self-check
Question: Why is acrylamide hard to remove from a food once it has formed?
Because it forms during cooking itself, in the browning reactions between asparagine and sugars. Once the food is cooked, the acrylamide is part of the food, so control must happen before and during cooking, not after it.
Question: A breakfast cereal tests high for MOSH but low for MOAH. What does the business do first?
It checks the packaging route first, because MOSH in a dry cereal stored in printed board points to migration from the pack. It would test the board, check whether a barrier is present, and trial a barrier layer or virgin fibre board, then retest the cereal.
Try it yourself
Take three foods from your kitchen that are fried, baked, or roasted: crisps, bread, and coffee are good choices. For each one, write down the process contaminant most likely to be present, the step that creates it, and one change in raw material, recipe, or cooking that would lower it. Then rank the three changes by how easy they would be in a real factory, and write down what would be lost, in flavour or texture, if the change went too far.
Key ideas to carry forward
- Emerging contaminants are often newly noticed or newly measured, not newly invented.
- Mineral oil hydrocarbons reach dry foods from printed and recycled board, and MOAH raises more concern than MOSH.
- Barriers, virgin fibre, and food grade lubricants are the main defences against mineral oils.
- Acrylamide forms when starchy foods cook hot and brown, from asparagine and sugars.
- Low sugar raw materials, lighter colour, and gentler cooking all cut acrylamide.
- Refining can create 3-MCPD and glycidyl esters, smoking can add PAH, and sealed heating can form furan.
- Every process contaminant is born in a step the food needs, so the answer is to adjust the step, not to remove it.
- Measure with chromatography and mass spectrometry, compare with benchmarks, and report the amount honestly.
Glossary
Mineral oil hydrocarbons
Mixtures of hydrocarbons from crude oil, known as MOH, that can reach food from packaging, lubricants, and the environment. Food tech example: MOH migrates from printed recycled board into breakfast cereal.
MOSH
Mineral oil saturated hydrocarbons, the straight and branched chain part of mineral oil. Food tech example: MOSH is the larger group found in a cereal tested for mineral oil migration.
MOAH
Mineral oil aromatic hydrocarbons, the ring-shaped part of mineral oil, watched closely because some members can damage genetic material. Food tech example: a low MOAH result matters more than a low total MOH result for a dry food.
Acrylamide
A chemical that forms when starchy foods cook at high temperature from asparagine and sugars. Food tech example: acrylamide forms in crisps fried until dark brown.
Asparagine
A natural amino acid in foods such as potatoes and cereals, one of the starting substances for acrylamide. Food tech example: potatoes low in asparagine make less acrylamide when fried.
Asparaginase
An enzyme that breaks down asparagine before cooking, used to cut acrylamide in baked and fried foods. Food tech example: asparaginase added to biscuit dough lowers the acrylamide in the baked biscuit.
Process contaminant
A substance formed during processing, such as heating, smoking, or refining, rather than added to the food. Food tech example: acrylamide is a process contaminant of frying and baking.
3-MCPD esters
Compounds that can form when vegetable oils are refined at high temperature. Food tech example: 3-MCPD esters are checked in refined oils used for infant formula.
Glycidyl esters
Compounds that can form during the refining of vegetable oils at high temperature. Food tech example: a refiner lowers its deodoriser temperature to cut glycidyl esters in its oil.
PAH
Polycyclic aromatic hydrocarbons, compounds that form when fat burns and smoke carries them onto food. Food tech example: PAH can be found in fish smoked directly over a fire.
Furan
A volatile compound that can form when food is heated in a sealed container such as a can or jar. Food tech example: furan is measured in canned foods after heat treatment.
Benchmark level
A reference value for a contaminant that most producers already meet, used to push the worst performers to improve. Food tech example: a crisp maker compares its acrylamide results with the benchmark level for crisps.
Barrier layer
A layer in packaging that blocks the movement of a substance from the pack into the food. Food tech example: an inner bag acts as a barrier layer that stops mineral oil moving from board into cereal.
Virgin fibre
Paper fibre used for the first time, not recycled, carrying less mineral oil than recycled fibre. Food tech example: virgin fibre board is chosen for direct contact with dry foods.
Blank sample
A sample with no food in it, run through the same test to prove the laboratory added no contamination of its own. Food tech example: a blank sample proves the MOH in a result came from the cereal, not from the laboratory.
References
- Nielsen, S. S., Food Analysis, 5th edition, 2017.
- Belitz, H. D., Grosch, W. and Schieberle, P., Food Chemistry, 4th edition, 2009.
- Stadler, R. H. and Lineback, D. R., Process-Induced Food Toxicants: Occurrence, Formation, Mitigation, and Health Risks, 2009.
- Food and Agriculture Organization of the United Nations and World Health Organization, Safety Evaluation of Certain Contaminants in Food, 2011.
- Woolfe, M. and Primrose, S., Food Forensics and Toxicology, 2004.
Unit 10: Food Fraud in Specific Commodities, Olive Oil, Honey, Spices, and Seafood
Note
Unit 10: Food Fraud in Specific Commodities, Olive Oil, Honey, Spices, and Seafood
A question to start with
Four foods sit on a shelf: a bottle of olive oil, a jar of honey, a tin of saffron, and a fillet of fish. Each one is expensive for its weight. Each one is easy to fake and hard to check by looking. Each one has been faked so often that whole testing methods were built for it alone. What makes these four foods the favourite targets of fraud, and what does each one teach about the fight against it? This unit goes commodity by commodity, because fraud is not one problem: it is a different problem in each food, with its own tricks, its own tests, and its own defences.
The four foods share three traits. They are high in value, so a small substitution earns real money. They are variable by nature, so a genuine sample can differ from the next, which gives a fraudster room to hide. And their quality is hard to judge by the senses, so a buyer cannot tell the fake from the real by taste or smell alone. A fraudster who understands those three traits can pick the trick that fits the food, and a defender who understands them can pick the test that fits the trick.
This unit stays in general terms: it describes the types of fraud and the types of test, not named companies or dated cases. The patterns are what matter, because the same patterns repeat in every market and every year.
Olive oil: the blend that tastes right
Olive oil fraud is the oldest in this unit, and it works because pure olive oil is dear and seed oils are cheap. The common trick is blending: a little cheap refined seed oil, or a refined olive oil, is mixed into a bottle sold as a higher grade. A second trick is mislabelling the grade: an oil that was refined with heat is sold as oil pressed without heat. A third trick is mislabelling the origin: oil from one place is sold as oil from a place whose name commands a higher price.
The defence starts with chemistry. The fatty acid pattern of olive oil differs from the pattern of seed oils, so gas chromatography can show a blend that the tongue cannot taste. Other markers show whether an oil was refined with heat, and whether it was made from olives or from another source. The defence also needs the paper: a mass balance check on the press, comparing the olives bought with the oil sold, and a traceability record back to the grove.
The buyer has a part to play too. A price far below the price of genuine oil is itself a signal, because no honest seller can sell below the cost of the olives. A buyer who buys the cheapest oil and expects the best grade is asking to be cheated. The honest trade in olive oil rests on buyers who pay a fair price, sellers who can show their papers, and laboratories that test the oil that the papers describe.
Questions to think about
- Why does a blend of olive oil and seed oil pass a taste test but fail a fatty acid test?
- A press buys olives from ten groves and sells oil under one place name. What does that do to the origin claim, and what should the label say instead?
- Why is a very low price itself a warning sign for olive oil, even before any test is run?
Think about the blend question before you read on
Because taste reads flavour, and a small amount of seed oil changes flavour only slightly, while the fatty acid pattern is a chemical fingerprint that changes in a way a laboratory can measure. The blend is built to fool the senses, so the defence must use a tool the senses cannot match.
Honey: syrup in the jar
Honey fraud works on the same principle as olive oil fraud: a dear natural product is diluted with a cheap sweetener. The common trick is to add syrup made from crops such as corn or rice, or from sugar beet or cane, to real honey. A second trick is to feed bees sugar syrup during the honey flow, so that the bees themselves store syrup in the comb. A third trick is to mislabel the origin or the flower source: honey from one place or one flower is sold as honey from a dearer place or flower.
Honey is hard to test because it is mostly sugar, and syrup is mostly sugar too. A simple sugar test cannot tell them apart. The tests that can are isotope analysis, which reads whether the sugars came from flowers or from a factory crop, and detailed sugar and marker analysis, which reads the small differences in the minor sugars and compounds that bees and flowers leave in real honey. No single test catches every syrup, so laboratories use several tests together, and they keep libraries of genuine honey from each origin to compare against.
The defence also needs beekeeper records. A beekeeper who can show how many hives were kept, how much honey each hive gave, and when it was harvested gives a buyer a mass balance to check. A seller who offers far more honey of one flower than the hives in that area could have made is selling something that cannot be true, however pure the sample in the laboratory.
Spices: colour, weight, and danger
Spice fraud is the most varied in this unit, because spices differ so much from each other. The tricks fall into three groups. The first is dilution with cheap bulk: starch, flour, or ground husks added to a ground spice to add weight. The second is substitution with a cheaper look-alike: a cheaper spice, or a part of the plant that carries little flavour, sold as the dear spice. The third is colour fraud: a dye added to make a dull spice look bright, or to make a cheap spice look like a dear one.
Saffron shows the pattern at its worst. It is the most expensive spice by weight, sold as thin threads, and it is faked with dyed threads of other plants, with added weight from moisture or sugar, and with colour from dyes. Chilli powder shows the danger: dyes added for colour have included substances that are not allowed in food at all, and that can harm health. Turmeric has been bulked with starch and brightened with dyes, and black pepper has been bulked with ground husks and seeds.
The tests for spices are microscopy, which shows the plant structures in a ground spice, and chromatography, which finds dyes, starch markers, and the flavour compounds that the real spice should carry. The defence also needs whole spice buying where possible: a whole chilli, a whole peppercorn, and a whole saffron thread are harder to fake than a powder, because the buyer can see what it is. Grinding at home, or buying from a grinder who tests, is the oldest defence of all.
Seafood: the wrong name on the fillet
Seafood fraud is mostly species substitution: a cheap fish is sold as a dear fish. Once a fish is filleted, skinned, and frozen, even an expert cannot always name it by looking, so the label is the only guide the buyer has, and the label can lie. The trick is profitable because the price gap between species is wide, and because a fillet of one white fish looks much like a fillet of another.
The test for seafood is DNA barcoding, which names the species from a small sample, even in a cooked or frozen fillet. It is the most reliable tool in this unit, because DNA names the species directly, and a cheap fish cannot carry the DNA of a dear fish. The limits are processing: heavy cooking and canning break DNA into short pieces, so a canned fish may need a shorter DNA test, and a heavily processed fish cake may carry too little DNA to read.
Seafood fraud also includes mislabelling the catch area, the farming method, and the weight. Fish can be sold as wild when it was farmed, as caught in one sea when it was caught in another, or with added water that is sold as fish. The weight trick is caught by weighing the fish after thawing and draining, and the catch claim is caught by chain of custody records from the boat to the shop, because no laboratory test can name the sea a fish came from.
Other foods at risk: milk, meat, coffee, and vanilla
Olive oil, honey, spices, and seafood are the classic targets, but they are not the only ones. Milk is diluted with water and rebuilt with cheap solids, and the old milk powder case showed that a test can be fooled when the fraudster knows which number the test reads. The defence is to test more than one number: protein, fat, and the markers that water and added solids change in different ways, so that a trick which fixes one number breaks another.
Meat fraud follows the same pattern as fish: a cheap meat sold as a dear meat, or a meat added to a product that claims to be free of it. The tests are DNA methods, and the paper check is the mass balance between animals bought and meat sold. The harm here is not only money: a meat that is not named on the label can break a religious rule or trigger an allergy, so meat fraud is also a safety and trust issue.
Coffee is faked by adding roasted grains, husks, or cheaper beans to dear beans, and by mislabelling the origin or the roast. The tests are microscopy of the grounds and chemical fingerprints of the bean, and the buyer can check by buying whole beans, which are harder to fake than ground coffee. Vanilla is faked by selling synthetic flavour as natural vanilla extract, and by diluting real extract. The test is the chemical pattern of the flavour compounds: real vanilla carries a family of related compounds, while synthetic flavour carries mainly one.
The lesson from these four extra foods is that the same four steps defend every commodity: name the trick, pick the test that reads it, check the paper, and check the price. A food student who can run those four steps for any food, from milk to vanilla, has the core skill of this whole subject.
The diagram shows the defence plan that fits all four commodities. The test changes with the food, but the steps stay the same.
What the four commodities teach together
The four foods teach the same lesson in four ways: the fraud fits the food, so the defence must fit the food too. A DNA test is the answer for fish and useless for olive oil, because oil carries little DNA. An isotope test is the answer for honey and useless for saffron, because the question is different. A buyer who applies one test to every food will miss most frauds, and a buyer who applies the right test to each food will catch most of them.
They also teach that price is information. In each food, a price far below the cost of the genuine article is a signal that should trigger testing, not a bargain to be celebrated. And they teach that the cheapest defence is often the paper defence: mass balance, traceability, and a supplier who can show where the food came from. A laboratory can prove a sample is fake, but the papers can prove that a whole lot cannot be genuine, and they can do it before the food is even tested.
The last lesson is about the buyer. In each food, the fraud survives because buyers want the dear food at a cheap price, and sellers who promise it find buyers who believe them. A market that pays a fair price for a genuine food, and that checks the claim, is a market where fraud earns little. A market that chases the lowest price is a market where fraud earns much, and no laboratory, however good, can fix that alone.
Practical note: Buy a small amount of a whole spice and the same spice ground, from the same shop, and compare the price per weight. Then look at the ground spice under a magnifying glass and note the colour and texture. Whole spices are harder to fake, and the price gap between whole and ground tells you how much room there is for a fraudster to earn.
Quick self-check
Question: Why is DNA the best test for fish but a poor test for olive oil?
Because a fish fillet carries the DNA of the fish it came from, so DNA names the species directly, while refined olive oil carries little or no DNA. The test must match the food, and for oil the matching test is the fatty acid pattern, not DNA.
Question: A honey seller offers far more of one flower honey than the hives in that area could make. What does that show, even if a sample tests pure?
That the mass balance fails: the amount sold is larger than the amount the hives could have made. A pure sample proves only that the sample is genuine, not that the whole lot is, and the numbers show the lot cannot all be what the label claims.
Try it yourself
Choose one of the four commodities and write a one-page buying plan for it: the trick most likely to be used, the test you would ask for, the paper check you would run, and the price below which you would refuse to buy. Then swap plans with a classmate who chose a different commodity, and check whether their test would work on your food. Where the tests differ, write down why the foods need different tests.
Key ideas to carry forward
- Olive oil is faked by blending and by mislabelling grade and origin, and fatty acid patterns expose the blend.
- Honey is faked by adding syrup and by mislabelling flower and origin, and isotope tests read the source of the sugars.
- Spices are faked by dilution, substitution, and dyes, and microscopy and chromatography expose them.
- Seafood is faked by species substitution, and DNA barcoding names the species directly.
- The fraud fits the food, so the defence must fit the food: one test cannot cover all four.
- Mass balance and traceability can prove a whole lot cannot be genuine, even when a sample tests pure.
- A price far below the cost of the real food is itself a warning sign.
- Whole spices and fair prices are simple defences that work in every market.
Glossary
Vanilla extract
A flavouring made by soaking vanilla pods in alcohol and water, carrying a family of flavour compounds from the pod. Food tech example: real vanilla extract carries many flavour compounds, while synthetic flavour carries mainly one.
Ground coffee
Coffee beans that have been ground into small pieces, easier to fake than whole beans because the bean cannot be seen. Food tech example: ground coffee can hide roasted grains that whole beans would show.
Blending
Mixing a cheap oil or ingredient into a dear one and selling the mixture as the dear one. Food tech example: seed oil blended into olive oil and sold as pure olive oil.
Fatty acid pattern
The pattern of the different fatty acids in a fat or oil, used like a fingerprint to name the oil. Food tech example: the fatty acid pattern of olive oil differs from that of seed oils, so a blend can be found.
Isotope analysis
A test that reads the proportions of isotopes in a food to show where it came from and how it was made. Food tech example: isotope analysis shows whether honey sugars came from flowers or from a factory syrup.
DNA barcoding
A method that reads a short standard stretch of DNA and matches it against a library to name the species. Food tech example: DNA barcoding of a fillet shows whether it is cod or a cheaper fish.
Microscopy
Looking at a sample under a microscope to see the plant or animal structures in it. Food tech example: microscopy of ground chilli shows starch grains that should not be there.
Dilution
Adding a cheap substance to a dear food to add bulk and lower its cost. Food tech example: syrup added to honey dilutes the real honey in the jar.
Substitution
Replacing one food with a cheaper look-alike and selling it under the dear food's name. Food tech example: a cheap white fish sold as a dear fish is substitution.
Saffron
The most expensive spice by weight, made from the dried threads of a flower, and a frequent target of fraud. Food tech example: saffron threads are faked with dyed threads of other plants.
Grade
A named quality level of a food, such as the grade of an olive oil, set by rules about how it was made. Food tech example: oil pressed without heat holds a higher grade than oil refined with heat.
Origin claim
A statement on a label about where a food came from. Food tech example: honey sold as coming from one valley carries an origin claim that traceability must support.
Catch area
The sea or water where a fish was caught, named on the label for some fish. Food tech example: a fish sold as caught in one sea carries a catch area claim that only chain of custody can prove.
Whole spice
A spice sold unground, in its natural form, such as a whole peppercorn or a whole chilli. Food tech example: whole black peppercorns are harder to fake than ground pepper.
Farmed fish
Fish raised in a farm rather than caught in the wild. Food tech example: farmed fish sold as wild fish is a common seafood fraud.
Thaw loss
The water lost when a frozen food thaws, used to check whether water was added to add weight. Food tech example: a fish fillet that loses much weight on thawing may have been sold with added water.
References
- Downey, G., Food Authenticity and Traceability, 2003.
- Ellis, D. I., et al., Fingerprinting Food: Current Technologies for the Detection of Food Adulteration and Contamination, 2012.
- Nielsen, S. S., Food Analysis, 5th edition, 2017.
- Woolfe, M. and Primrose, S., Food Forensics and Toxicology, 2004.
- Spink, J. and Moyer, D. C., Defining the Public Health Threat of Food Fraud, Journal of Food Science, 2011.