PPWR PFAS limits for food-contact packaging: what to prove

As of 4 November 2026, the PFAS limits in Article 5(5) of Regulation (EU) 2025/40 (the PPWR) have applied for nearly three months. Food-contact packaging that reaches any of the three limit values cannot be placed on the EU market, there is no sell-through period, and the EU still has no harmonised test method.

The legal limits come from the regulation and the Commission’s guidance notice. The testing and evidence advice below comes largely from law firms, consultancies and laboratories, some of which sell testing or compliance services, so treat their views as interpretation. This is information, not legal advice; check the official text before you rely on any reading.

The three limits in Article 5(5)

The restriction covers food-contact packaging only. It sets three separate limits, and a pack must stay below all of them. The Food Packaging Forum lists them as follows.

Limit	Value	Measured as	Polymeric PFAS
Any single PFAS	25 ppb (25 µg/kg)	Targeted analysis	Excluded from quantification
Sum of PFAS	250 ppb (250 µg/kg)	Sum of targeted analysis	Excluded from quantification
Total PFAS	50 ppm (50 mg/kg)	Total PFAS	Included

Our calculation: 50 mg/kg equals 50,000 µg/kg, and 50,000 ÷ 25 = 2,000, so the total-PFAS value is 2,000 times the single-substance value (50,000 ÷ 250 = 200 times the sum value). The three limits sit on very different scales because they answer different analytical questions.

The Food Packaging Forum notes that the limits apply to intentionally and non-intentionally added PFAS alike. They look at the concentration in the packaging, with migration into food outside the test. A supplier’s statement that no PFAS were added therefore settles little. Fieldfisher makes the same point in its practical compliance note: the Commission expects exceedances mostly where PFAS are added on purpose, but the legal test is presence.

What the limits cover

Compliance consultancy Regsurance reads the Commission FAQ as applying Article 5 to the whole packaging unit, so inks, varnishes, glues and adhesives count alongside the substrate.

The PFAS values apply to food-contact packaging. The same consultancy stresses that the general substance duties and the heavy-metal limit in Article 5 apply to all packaging from 12 August 2026. A coated paper wrap, a moulded fibre tray and a plastic tub with a fluorinated barrier are all in scope if they touch food; a shipping carton for an unrelated product is not.

Stock: what “placed on the market” means in practice

The Commission’s guidance, Commission Notice C/2026/3084, confirms that no general stock-exhaustion period exists for PFAS in food-contact packaging. Packaging Europe summarises the effect: packs placed on the market after 12 August 2026 must meet the limits, and packs placed before that date do not have to be withdrawn.

The date that matters is the first making available, not the production date. The Food Packaging Forum reports that packaging made before 12 August 2026 but first placed on the EU market afterwards must comply. A converter holding fluorinated-barrier paper in its own warehouse on that date has a problem; the same paper already supplied to a customer before that date is treated differently. The guidance also defines who counts as a manufacturer, producer or importer, and that role decides who carries the duty, so read those definitions against your own supply chain.

The Commission’s stepwise testing approach

The guidance admits there is no harmonised EU methodology for PFAS in food-contact packaging. It recommends a cascade instead, which the Food Packaging Forum and Packaging Europe both describe in the same order.

Step	Test	Result treated as compliant	Purpose
1	Total fluorine	Below 50 mg/kg	Screening of the whole sample
2	Organic versus inorganic fluorine, for example by pyrolysis-GC/MS	Organic fluorine below 50 mg/kg	Separates PFAS-type fluorine from fluoride salts
3	Direct total oxidisable precursor (TOP) analysis	Within 25 µg/kg and 250 µg/kg	Checks the targeted limits

Keller and Heckman’s Packaging Law blog adds that, on the evidence available to the Commission, samples passing the total fluorine test also pass steps 2 and 3.

Where total fluorine misleads

A step 1 result above 50 mg/kg does not show a breach. A peer-reviewed study on talc in polypropylene found that fluorine in mineral fillers raises the total fluorine signal and can produce false positives. That is the case for running step 2 before you reformulate anything.

In the other direction, Fieldfisher’s Belgian team cites the ECHA risk assessment committee (RAC) as saying that total fluorine methods cannot tell PFAS fluorine from other fluorine. RAC also found their quantification limits too high to screen reliably against the 25 µg/kg and 250 µg/kg values. Fieldfisher’s advice is to treat each result as an entry in a compliance file and not as a final answer.

Who has to do what

The PPWR puts the evidence burden on the manufacturer, with duties running up the chemical supply chain. The roles below follow the sources cited in this article.

Role	Obligation	Evidence to hold
Packaging manufacturer	Show PFAS compliance in technical documentation under Article 5(6) and Annex VII; the EU declaration of conformity covers Article 5(5) from 12 August 2026	Test results, supplier declarations, declaration of conformity
Manufacturer, importer or REACH downstream user of the chemicals	Where total fluorine exceeds 50 mg/kg, provide on request proof of how much fluorine is PFAS and how much is not (Article 5(5)(c))	Analytical proof of PFAS and non-PFAS fluorine
Market surveillance authorities	Enforce the limits; the Commission says it is working towards a harmonised approach across national authorities	Not applicable

The UK’s Food and Drink Federation guidance page lists Article 5(4) on heavy metals, Article 5(5) on PFAS and, where relevant, Article 11(1) on reusable packaging as the requirements covered by the conformity assessment and declaration of conformity. The Fieldfisher note reports that the second-edition FAQ says suppliers cannot refuse to hand over the relevant documentation to the manufacturer.

What is still open

The PPWR application date was not postponed, and the PFAS limits are untouched by the current simplification talks. Those talks concern a different duty: the Council set its negotiating position on 21 October 2026 on suspending the obligation to appoint an authorised representative for extended producer responsibility. That is a proposal heading into talks with Parliament, not adopted law.

Two practical gaps remain. Consultancy PSL reports that the second-edition FAQ confirms no list of PFAS substances will be published, so checking a supplier against a list does not meet Article 5(5). And with no harmonised method, two laboratories can return different answers on the same sample.

What to do now

List every food-contact pack and component you place on the EU market, including coatings, inks, varnishes and adhesives, and mark which ones use fluorinated chemistry or fluorinated polymers.
Ask each supplier for a written PFAS statement that refers to Article 5(5) and states the test basis, not only “PFAS-free” or “no intentionally added PFAS”.
Run step 1 total fluorine on representative samples of each material grade. Where a result exceeds 50 mg/kg, commission step 2 before changing the specification.
Where total fluorine exceeds 50 mg/kg, send a request under Article 5(5)(c) to your chemical suppliers for proof of PFAS versus non-PFAS fluorine.
Check the date each batch was first placed on the EU market. Stock first placed after 12 August 2026 must comply; treat anything undocumented as non-compliant until shown otherwise.
Add the PFAS evidence to your technical documentation and make sure the EU declaration of conformity references Article 5(5).
Agree with your laboratory which method and reporting limits it uses, and keep the reports with the batch records.
Read the Commission guidance notice and the second-edition FAQ (3 August 2026) directly, and take legal advice where your role as manufacturer, importer or producer is unclear.

Sources

Header image: illustration, AI-generated.