Are Cans Lined With Plastic? What's Actually Inside Your Food, Beer & Tuna Cans (2026)

Last updated: July 13, 2026

Why the lining exists at all

Start with the engineering, because it explains everything that follows. A food can is a metal pressure vessel that gets filled with a wet, salty, often acidic substance and then cooked, retort sterilisation heats the sealed can to around 120°C, and then left on a shelf for two or three years. Bare tinplate steel or bare aluminium cannot survive that. The metal corrodes, hydrogen builds up, the can swells, and long before any of that happens the food tastes metallic and picks up dissolved metal.

So the inside of the can is coated. The coating is a liquid polymer resin, sprayed onto the metal and cured in an oven into a continuous film about 5 to 12 micrometres thick, around a tenth the thickness of a human hair. It is applied to the inside and, usually, the outside too. Food cans are mostly three-piece welded tinplate steel; beverage cans are mostly two-piece drawn-and-ironed aluminium. Both get an internal coating. There is no meaningful category of “uncoated can” in a Western supermarket.

That is the honest core of the answer to the question in the title. Yes, cans are lined with plastic, all of them, by necessity. Anyone selling you a “plastic-free can” is selling you a glass jar.

What the lining is made of, the five chemistries

Here is where the last decade of change lives. Until roughly 2010, the default was an epoxy resin made by reacting BPA with epichlorohydrin, which produces BADGE, bisphenol A diglycidyl ether, the backbone of the classic gold-coloured can lining. Epoxy is genuinely excellent at the job: tough, flexible enough to survive the can being formed, and chemically resistant to almost any food.

Consumer pressure, then regulation, pushed the industry off it. The replacements are not one material but four families, each with its own weaknesses, which is precisely why no single one took over.

A fifth family, PVC-based (vinyl/organosol) coatings, also shows up in can testing. It is acid-resistant and flexible, but it is made from vinyl chloride (a known human carcinogen) and needs plasticisers, so it is the one replacement that is hard to call an upgrade. It is not among the four families the industry names as its standard non-BPA linings.

How much of the US canned-food supply actually moved off BPA?

Most of it, and this is one of the few consumer-chemistry stories where the pressure campaign genuinely worked. The Can Manufacturers Institute states plainly that “new linings…do not use bisphenol A (BPA) as a component of the can lining,” that nearly all (about 98%) of US food can production has moved to new linings, and that its 2019 Canned Food Market Basket Report found nearly 100% of sampled cans free of BPA. That is a trade body describing its own members, so treat it as a claim, not a measurement.

The independent numbers point the same way. The Center for Environmental Health, which had spent years publicly testing cans and naming brands, reported a steep decline between 2017 and 2019, with about 95% of cans testing free of BPA in its 2019 round. For comparison, the 2016 Buyer Beware coalition study, nearly 200 cans across the US and Canada, still found BPA-based epoxy in the majority of them. The shift in under a decade is real.

Two caveats worth carrying. First, the transition is a US food-can story. Imported cans, older stock, and some speciality products still carry epoxy. Second, and more important for this article: moving off BPA did not remove the plastic. It replaced one polymer film with a different polymer film. Which brings us to the substitution question.

The regrettable-substitution problem, stated precisely

You have read the shorthand a hundred times: BPA-free cans just use BPS instead, and BPS is just as bad. Half of that is right, and the half that is wrong matters.

The chemistry half is right. A systematic review in Environmental Health Perspectives compared BPA with its two commonest substitutes and found that BPS and BPF are of the same order of potency and the same mode of action as BPA, oestrogenic, anti-oestrogenic, androgenic and anti-androgenic. They are structural analogues, and they behave like it. Later in-vitro test batteries have reached the same conclusion across a wider panel of bisphenol analogues. Where BPS and BPF are used, thermal receipt paper, some polycarbonate replacements, calling the swap “regrettable” is fair.

The can-specific half is where the shorthand breaks down. The US can industry states explicitly that it does not intentionally add BPF, BPB or BPS to can linings, and the four families it names as replacements, acrylic, polyester, non-BPA epoxy, olefin polymers, are not bisphenol-S chemistry. So the honest position is this: the BPS-in-cans claim is not supported by what the can industry says it uses, and we are not going to repeat it as though it were.

That is not the same as saying the replacements are proven safe. The real problem is subtler and, in our view, more concerning: the replacements are less studied, and they bring their own migrants. Polyester linings have been shown to release cyclic polyester oligomers into food simulants and into infant food, compounds for which, as one 2023 analysis noted, no toxicological data exist and no migration limits have been set. Meanwhile, the Center for Environmental Health's own summary is blunt: adequate testing to establish that the new linings are safe for food has not been done, and only a couple of companies disclose which liner they moved to. We swapped a well-characterised bad actor for a set of poorly characterised unknowns. That is progress, but it is not a clean win, and you should hold it loosely. Our BPA-free deep dive and microplastics vs PFAS vs BPA explainer go further on why “free of X” labels tell you so little.

Do can linings actually shed microplastics? The honest answer

This is a microplastics site, so let us be exact about what is known and what is not, because the internet, including, in places, our own older coverage, has run ahead of the evidence here.

What is well established: chemical migration. BPA and BADGE demonstrably move from epoxy linings into food, at levels that show up in human urine within days. The Carwile trial at Harvard is the cleanest demonstration: 84 people, five days of one canned soup a day versus fresh soup, and a 1,221% rise in urinary BPA. That is not a modelling exercise or a lab simulant, it is a randomised crossover trial in people, and it is why the EFSA re-evaluation matters so much for canned food specifically.

What is plausible but unmeasured: particle release. A can lining is a polymer film in intimate contact with wet, sometimes acidic food, cooked at 120°C during sterilisation and then held for years. Every mechanism we know for microplastic release from food-contact polymers (heat, contact time, acidity, oil) is present. So the expectation that linings shed particles is entirely reasonable.

And yet: we could not find a single published study that counts microplastic particles released from a metal can lining into food. The 2025 systematic evidence map in npj Science of Food, the most complete survey of this question that exists, screening the literature and extracting 600 data entries from 103 eligible studies, is dominated by bottles (31% of entries), unspecified containers (19%), tea bags (12%) and cups (10%). Cans do not feature among the studied categories. The authors also note that most of the studies they included are of low quality. When you see a confident particle count for can linings online, check the citation: it usually leads back to a study of plastic containers or bottles, not cans.

The one direct measurement that does exist points down, not up. Karami and colleagues ran micro-Raman analysis on 20 brands of canned sardines and sprats from 13 countries and found no plastic particles at all in 16 of them, and 1–3 particles each in the other four. They concluded the health risk from micro- and mesoplastics in canned sardines was limited. That study only detected particles above ~149 µm, so it says nothing about the small stuff, but it is the best direct evidence we have, and it does not support the idea that a can is a particle firehose.

So the position we will defend: the chemical case against can linings is strong and specific; the particle case is theoretically sound and empirically empty. Anyone giving you a number of microplastics per can is making it up. We would rather tell you the evidence is thin than manufacture a statistic to match the headline.

Can type by can type, food, beer, tuna, tomatoes, cartons

The practical variation is not really between can types, it is between contents. Acid and fat and time are what pull chemistry out of a polymer. Here is how the common cases sort out.

For what is in the food rather than what the container is made of, see microplastics in canned food, microplastics in canned tuna and salmon, and microplastics in beer and wine.

Where the regulators actually stand

The two big food-safety agencies disagree, and you should know that rather than be told a consensus exists.

EFSA (Europe) moved decisively. Its 2023 re-evaluation of BPA, drawing on more than 800 studies published since 2013, set a tolerable daily intake of 0.2 nanograms per kilogram of body weight per day, roughly 20,000 times lower than the 4 µg/kg/day provisional figure it had used since 2015. The driver was immune effects: an increase in T-helper cells in the spleen, which EFSA linked to allergic lung inflammation and autoimmune conditions. Its conclusion was that consumers with average and high exposure, in all age groups, exceed the new TDI. The European Commission followed with Regulation (EU) 2024/3190, which bans BPA in food-contact varnishes and coatings; the general deadline for non-compliant coated articles is 20 July 2026, with an extension into 2028 for cans preserving fruit, vegetables and fishery products, the acidic, aggressive contents the alternatives handle worst.

The FDA has not moved. Its stated position remains that “the available information continues to support the safety of BPA for the currently approved uses in food containers and packaging,” based on a four-year review of over 300 studies completed in 2014. The FDA did remove BPA from baby bottles, sippy cups and infant formula packaging in 2012–13, but it is careful to say those amendments were based on abandonment of the use, not on safety. The agency has agreed to reconsider its BPA assessment; nothing has changed yet.

Germany's BfR has publicly disagreed with EFSA's methodology, and a published critique argued the evidence did not support lowering the TDI. This is a live scientific dispute, not a settled one. What is not in dispute: canned food is one of the largest dietary sources of BPA exposure, and the two agencies' positions on it are now separated by four orders of magnitude.

What to actually do about it

• Move acidic foods out of cans first. If you change one thing, change this. Tomatoes, tomato paste, citrus, pickled vegetables: buy them in glass jars. Glass is inert, there is no lining, and acidic contents are the highest-migration case for every lining chemistry including the new ones. Passata and chopped tomatoes in glass are widely available and barely cost more.
• Cartons are a real step down, but they are not plastic-free. A Tetra Pak-style carton is paperboard, aluminium foil and polyethylene, the layer touching your food is PE. What it does not have is a bisphenol epoxy or a 120°C retort cook. For milk, stock, coconut milk and passata, carton beats can. It just is not zero.
• Frozen beats canned for vegetables. Frozen produce skips the retort sterilisation and the years of shelf contact entirely, and the packaging is a cold-contact bag. Cold suppresses migration; it is the lowest-temperature link in the whole food chain.
• Dried beans are the cheapest swap in this article. A bag of dried chickpeas costs less than the cans it replaces, involves no lining at all, and, soaked overnight, simmered, frozen in portions, is genuinely no harder than opening a tin, just less instant.
• Prefer brands that disclose their lining. This is the lever with real leverage. Very few companies say which liner they moved to; a handful do, and they only do it because customers ask. If a brand will tell you it uses an acrylic or oleoresin lining, that is a meaningful signal, not because acrylic is proven safe, but because disclosure is the precondition for anyone ever finding out.
• Do not store food in an opened can. Once opened, the coating is exposed to air and the metal is exposed at the cut edge. Decant leftovers into glass. Two minutes, zero cost.

None of this requires you to fear a tin of chickpeas. The point of the ladder is that the top rungs (acidic food, oily food, long storage) are where the evidence actually is, and those are exactly the ones with easy substitutes. Everything else is a judgement call you can make calmly. The same logic runs through what “microwave safe” actually means and the wider microplastics in food hub.