What PVA actually is, and what the evidence does not settle

Polyvinyl alcohol, CAS 9002-89-5, is not made from a monomer called vinyl alcohol. It is made by polymerising vinyl acetate and then hydrolysing the result, swapping acetate groups for hydroxyl groups until the chain is as water-loving as the manufacturer wants it1. How far that reaction runs is the degree of hydrolysis, and it governs most of what the material does. The FAO and WHO expert committee assessment puts it plainly: the physical characteristics and the specific functional uses depend on the degree of polymerisation and the degree of hydrolysis, and the material is classed as partially or fully hydrolysed1.

The same assessment records two further things worth holding on to: polyvinyl alcohol is soluble in water, and it is not known to occur as a natural product1. So it is a synthetic polymer built from a petrochemical feedstock, which is what people mean when they call it a plastic, and it is water-soluble, which is what almost nothing else called a plastic in a kitchen is. Nearly every argument that follows is a disagreement about which of those two facts should govern.

The film does one job. It holds a measured dose together, dry, until water arrives, and brands describe it in those terms whenever they describe it at all. Reckitt lists polyvinyl alcohol in the ingredient entry for Finish Powerball Quantum with the stated function “Dissolvable Film”2. Earth Breeze lists it on its ingredients page under two functions, “Film former” and “Structuring Agent”, each described as holding the cleaning ingredients until dissolved in water3. Dropps names the polymer as PVOH on the product page itself4.

That is the whole of the case for it, and it is a real one. A pre-dosed unit needs something holding the dose, and the alternatives are a scoop and a box, a pressed tablet, or a sheet. Every brand that has taken the film out has done one of those three instead.

This is the sentence the subject turns on. Dissolving is a physical change: the chains come apart from each other and disperse among water molecules, and the polymer is still a polymer, now invisible. Biodegrading is a biological one: micro-organisms break the chains down and mineralise the carbon, at which point it stops being a polymer at all. A film disappearing in a dishwasher is evidence of the first and none at all of the second.

European law keeps the two apart on purpose. Under the REACH restriction on intentionally added microplastics, degradation must be proved with named test methods whose pass criterion is percentage mineralisation, that is, degradation all the way down to carbon dioxide, or a degradation half-life in a simulation test5. Solubility is proved separately, by a different appendix, and the Commission’s explanatory guide defines it as the maximum amount of a material that can be dissolved in water, distinguishing it from dispersibility5. The two words are not interchangeable in the regulation, even where they are in the marketing.

This is where the evidence thins, and where both sides sound more certain than their own documents.

In 2021 Charles Rolsky and Varun Kelkar published an estimate that around 77 per cent of the polyvinyl alcohol reaching US wastewater treatment from laundry and dishwasher pods passes through intact, putting national emissions at roughly 8,100 metric tonnes a year6. It is worth being exact about what that paper is, because it is quoted as though it were a measurement campaign. It is not: it combines a review of published degradation figures with a consumer survey of 527 respondents and water-use data, and models the outcome6. It was funded in part by Blueland, a brand selling pods without the film, and the authors record that the funder had no role in design, analysis or the decision to publish6. The paper states its own central limitation too, that little research exists monitoring the biodegradability of polyvinyl alcohol in the natural environment6.

In January 2023 Blueland and the Plastic Pollution Coalition, joined by fifteen other organisations, petitioned the EPA to order testing of PVA under the Toxic Substances Control Act and to move it on the Safer Chemical Ingredients List from a green circle to a grey square meanwhile7. The marks carry weight: a green circle means the chemical has been verified to be of low concern on experimental and modelled data, a grey square that it may not be acceptable in products seeking the label8.

The EPA refused, in a letter signed on 21/04/20239. Its reasoning is the useful part. The agency held that the petition had not shown existing information to be insufficient; that the supporting data “relied on a study estimating the persistence of PVA and several studies on the health and environmental impacts of microplastics, rather than the soluble PVA used in Safer Choice-certified products”; that only certain PVA structures are permitted in certified products; and that it had identified data not discussed in the petition giving “a clear weight of evidence” that those structures meet the standard9. It did not publish that data with the letter, which is the part a reader is entitled to find unsatisfying.

The public counter-evidence comes largely from the other interested party. In 2024 a team of Procter and Gamble scientists published simulation testing under OECD 303A reporting average removal of 97.4 per cent for a detergent-grade polymer, with no evidence of it sticking to sludge solids, so the removal was biological rather than the polymer being carried off with the sludge10. The paper states that no outside funding supported the work and declares no conflicts, and every author is employed by the company that sells Tide10. Procter and Gamble also publishes a technical review of the 2021 estimate, criticising it for pooling grades of polymer that are not the detergent grade, for generating no new measurements and for a modelling error, and reporting a panel of eight reviewers convened by the firm SciPinion who scored the paper 3.9 out of 10 for credibility, seven of them concluding that detergent-grade PVA should pass biodegradability criteria11.

Set the two side by side and the shape of the problem shows. The estimate that started the argument was part-funded by a brand selling pods without the film; the measurements answering it come from employees of a company selling pods with it. Both disclose, both are on the record, and neither is field measurement from a working treatment plant.

Under the two regulatory definitions that actually exist, no. The reasons have nothing to do with whether it is harmful.

The EU restriction on intentionally added microplastics has applied since 17/10/2023 and catches solid synthetic polymers contained in or coating particles12. It excludes four categories, one of them polymers with a solubility greater than 2 g/L proved by the prescribed test method5. The Commission’s explanatory guide is asked in as many words whether water-soluble synthetic polymers such as PVA fall in scope, and answers that polymers meeting the solubility criteria are outside it5. Being outside a restriction is a statement about a definition rather than a finding of safety: the restriction targets particles, and a dissolved polymer is not one.

California’s definition, adopted for drinking water in June 2020, comes at it from the other direction and lands in the same place. Microplastics in drinking water are solid polymeric materials in particles with at least three dimensions between 1 nm and 5,000 µm13. There is no solubility clause because none is needed: once the film has dissolved there is no particle left to count.

So the accurate answer is a dull one. Before the wash the film is a solid synthetic polymer; after it, a dissolved one, and what matters is what the treatment works and the river do next. Whether microplastic is the right label is a debate about vocabulary, and it settles nothing about the water.

Names first. On an ingredient list the polymer appears as polyvinyl alcohol, PVOH or PVA, or in a functions column as dissolvable film, film former or structuring agent, with CAS 9002-89-5 beside it23. If any of those is on the list, the product contains it.

Then read the list rather than the bullets, because the two do not always agree. Cleancult’s dishwasher tablets page describes the wrap as “a plastic free dissolving film”, and the ingredient list on the same page ends “Polyvinyl Alcohol Film”, both present when the page was read on 14/08/202614. Earth Breeze’s product page for detergent sheets promises no plastic waste and names no polymer, while the brand’s own ingredients page lists polyvinyl alcohol as a film former and a structuring agent315. Neither page is inaccurate on its own terms, and a shopper who reads one comes away with a different picture from a shopper who reads the other.

• Find the full list before concluding anything. If the product page has none, look for an ingredients page, a SmartLabel entry or a safety data sheet.
• Read the list against the claim. Where a marketing line and an ingredient list disagree, the list is the document with a legal duty behind it.
• Do not read the outer carton as an answer. The question is what is wrapped round the dose, not what is wrapped round the box.
• Treat a sheet like a pod until the list says otherwise, because the polymer can be the matrix holding a sheet together rather than a wrapper round a dose3.

One structural thing helps in the US. The Cleaning Product Right to Know Act has required online ingredient disclosure since 01/01/2020 and label disclosure since 01/01/2021, so a brand selling into California must publish a list somewhere, even when the somewhere is not the product page16. That is why a SmartLabel entry or an ingredients hub often carries what the product page does not.

We compare three narrow things, each checkable from a page in about a minute: whether the product contains the film, whether the page says so, and whether the film arrives home as packaging. We do not rule on whether polyvinyl alcohol is harmful, because the published evidence supports no verdict in either direction, and inventing one would be the same failure as inventing a cleaning result for a product nobody has washed anything with.

The third of those three is the one that needs defending, so here is the defence. A film wrapped round every individual dose is a wrapper: manufactured, shipped and disposed of with the product at one unit per use, and a disposal route down a drain rather than into a bin does not stop it being the pack. A per-unit film therefore counts against a product on packaging, whatever the outer carton is made of. That rule costs a paper-boxed pod exactly what it costs a plastic-wrapped one, and it applies to the brand that owns this site on the same terms as to everyone else.

What would change our mind is specific: field measurements of polyvinyl alcohol in effluent and receiving waters at working treatment plants, published with the method, by people with no commercial stake in the answer. Until that exists, this page describes an argument rather than resolving it, and the tables go on comparing what a label says.