Food Contact Ink Migration

Illustrative case study. Company names and identifying details are fictional. Technical details reflect real industry practice and current regulation.

Food-contact ink — the printing on packaging — can migrate into food. Photoinitiators (used in UV-cured inks), plasticizers, and mineral oils don’t stay on the package — they migrate through substrates, around edges, and via set-off (transfer from printed outer to unprinted inner in the reel). When a baby food brand found benzophenone (a photoinitiator) in its product, the ink — not the food — was the contaminant.

A baby food brand with an ink problem

PureStart Baby Food (fictional) produced 100,000 pouches of baby food a day in a 50-employee facility — fruit and vegetable purees, hot-filled into stand-up pouches (PET/Al/PE laminate), printed with UV-cured inks (vibrant colors, fast drying). The ink specification — “food-packaging grade” — didn’t specify photoinitiator limits. Baby food — the most sensitive consumer group — deserved better.

The benzophenone that migrated

Routine testing — expanded after an EU alert on photoinitiators — found benzophenone at 0.8 mg/kg in the baby food. Benzophenone — a photoinitiator in UV inks — is a possible endocrine disruptor (EFSA: TDI 0.03 mg/kg bw/day). For a 10 kg infant eating 200g of product, the exposure (0.16 mg) exceeded the TDI (0.3 mg? No — 0.03 × 10 = 0.3 mg — the exposure was below TDI, but detectable benzophenone in baby food is unacceptable regardless).

The migration path — set-off. The pouches — printed on the outside, stored as flat reels (printed side contacting food-contact side) — transferred benzophenone from ink to inner layer during storage. The hot-fill (85°C) then drove the migrant into the food. The aluminum layer — a functional barrier — was bypassed by set-off (the contaminant was already on the food-contact side before filling).

The ink supplier — certified “food-packaging grade” — used benzophenone as a photoinitiator (standard for UV inks). The certification didn’t prohibit it — because regulations (at the time) didn’t limit it in packaging inks. The gap was regulatory, not just supplier.

What actually caused it

1. Set-off migration. Reel storage — printed side contacting food-contact side — transferred ink components to the inner layer. The functional barrier (aluminum) was irrelevant — the contaminant bypassed it via set-off.

2. Ink specification gap. “Food-packaging grade” didn’t exclude benzophenone — the spec was vague, the regulation was silent. Baby food — requiring the highest purity — got standard ink.

3. Hot-fill driving migration. The 85°C fill temperature accelerated migration — heat drives diffusion. The process (hot-fill) interacted with the packaging (set-off contamination) to maximize transfer.

What changed on the floor

Immediate: all pouches with UV inks were quarantined — tested for benzophenone (and other photoinitiators), rejected if positive. Affected baby food (500,000 pouches) was recalled — voluntary, precautionary (exposure below TDI, but zero tolerance for baby food).

Within 30 days: the ink specification was rewritten — water-based inks only (no UV-cured, no photoinitiators), benzophenone-free certification required, migration testing (every new ink, before approval). Reel storage was modified — interleaving (barrier paper between printed and food-contact sides) to prevent set-off.

Within 90 days: the company switched to shrink-sleeve labels (applied after filling — no ink-to-food contact, no set-off risk). The packaging became unprinted (plain laminate) + sleeve (printed, removable) — eliminating the migration path entirely. The industry was notified — the company shared its findings with the baby food trade association.

What the numbers showed after

Twelve months post-correction: zero photoinitiator detections in 300 finished-product lots, zero in packaging migration tests. The shrink-sleeve switch — initially seen as expensive ($0.05/pouch) — enabled marketing (“ink never touches your baby’s food”) that increased sales 15%. The recall ($1M) was painful — but the brand trust (transparency, voluntary action) recovered within 6 months.

What you’d do Monday morning

Check your packaging inks — UV-cured or water-based? If UV, ask for the photoinitiator list — benzophenone, ITX, 4-methylbenzophenone. If present, assess the migration risk (especially for baby food, fatty foods, hot-fill).

Then check your reel storage — is the printed side contacting the food-contact side? If yes, you have set-off risk — regardless of barrier layers. Interleave or rethink the printing (sleeves, labels after filling).

Yeah, but actually — baby food with benzophenone — even below the TDI — is unacceptable. Not because the science says it’s harmful at that level, but because parents trust baby food to be pure — and detectable industrial chemicals violate that trust. Zero isn’t always scientifically necessary — but for baby food, it’s commercially and ethically necessary. Formulate your packaging specs for trust, not just compliance. The regulation sets the floor; parent expectations set the ceiling.