What the three PFAS limits actually say

PFAS means per- and polyfluoroalkyl substances, a family of synthetic compounds used in packaging to resist grease and water. Article 5(5) of Regulation (EU) 2025/40 restricts them in food contact packaging. The Commission guidance for the Regulation confirms that from 12 August 2026 food-contact packaging shall not be placed on the market if it contains PFAS above the stated thresholds.

The three limits are not three versions of the same test. They sit at different orders of magnitude and are answered by different laboratory work.

Put them on one scale and the structure becomes obvious. 50 parts per million is 50,000 parts per billion, so the all-PFAS ceiling is two thousand times the individual compound limit. That is not generosity. It is an acknowledgement that the only practical way to bound every PFAS present, including compounds nobody has a standard for, is to measure fluorine rather than individual molecules. That asymmetry is what gives an untreated product a cheap route through.

The stepwise method, and how to run it on your own product

The Commission guidance describes a stepwise testing approach: quantification of total fluorine, then confirmation of organic fluorine, then a total oxidisable precursor assay. Read as a decision tree rather than a list of three tests you must all buy, it works like this.

Step one. Total fluorine. The laboratory measures all fluorine in the sample, organic and inorganic together. This is an upper bound on PFAS content, because every PFAS contains fluorine. If total fluorine comes back below the reporting limit the laboratory can support, no PFAS population large enough to breach the 50 ppm ceiling can be present, and the question is closed for that batch. It is the cheapest test in the sequence and the right first purchase for a product declared as having no coating, no lamination and no fluorochemical treatment.

Step two. Organic fluorine. Total fluorine includes inorganic fluorine, which is why the guidance’s second step exists. A plant material can carry inorganic fluorine from soil and water without any PFAS being present. If step one returns a figure high enough to matter, step two separates organic fluorine from the inorganic background before anyone concludes anything. Treating a total fluorine number as a PFAS number is the most common analytical error in this area, and it has failed products that were clean.

Step three. Total oxidisable precursor assay. The TOP assay oxidises precursor compounds into measurable perfluoroalkyl acids, catching substances a targeted panel would miss. It matters for treated and coated products. For an untreated product it is usually unnecessary unless a buyer’s protocol demands it.

The method you can run on your own numbers is this: decide which of the three thresholds your evidence actually addresses before you pay for a test. A targeted panel answers the 25 ppb and 250 ppb limits and says nothing reliable about the 50 ppm ceiling. Fluorine screening answers the ceiling and says nothing about individual compounds. Most buyers who think they have complete evidence have bought one of the two.

What an untreated leaf product has to show

Areca leaf tableware is made from the fallen leaf sheath of the areca palm, pressed to shape. Foliware India LLP declares the product as having no coating, no lamination and no fluorochemical treatment. That declaration is the starting point of the evidence file, not the end of it. A proportionate file for a product of this type contains four things.

The last column is the one buyers skip. A test report proves something about the material that went into the instrument. Tying it to what arrives on your pallet is a batch traceability question, and it belongs in the declaration of compliance rather than the laboratory report. The fields that carry that link are taken apart in the guide to the declaration of compliance, field by field.

Annex VII also requires the technical documentation to specify, for each plastic part representing 5 percent or more of the total weight of the packaging unit, the polymer used. On a product declared as containing no plastic part, the correct entry is an explicit statement that none is present. Silence is not the same thing, and an auditor will read it differently.

No stock exhaustion period, and what that means operationally

The Commission guidance is unambiguous: “the PPWR does not foresee a transitional period for the exhaustion of stocks. Therefore, food-contact packaging placed on the market after 12 August 2026 must comply with the PFAS limits.”

Placing on the market is the first making available in the Union, which for an import is tied to release for free circulation rather than to the sale date. If you are holding stock across that boundary, work through the position with your own customs broker and counsel rather than assuming either answer, because it turns on when each consignment was first made available and that is a records question specific to your entries.

For a buyer in Germany, France, the Netherlands, Sweden or Denmark, the practical consequence is that the evidence has to be in hand before the container clears, not before it is sold. Who has to hold it is set by the economic operator roles explained in the analysis of who carries the PPWR obligation on goods from India.

Where this gets more expensive than it should

Three situations turn a cheap evidence path into a costly one, and it is fairer to name them than to let you discover them at onboarding.

Your retailer has its own PFAS protocol. Several European grocery and foodservice groups specify a named targeted panel with their own reporting limits, often tighter than the Regulation. If your customer’s protocol is contractual, fluorine screening will not satisfy it and you will buy the targeted panel regardless. Ask your customer for their protocol before you commission anything.

You need per-batch rather than per-product evidence. A single screening report against a production period is proportionate for a stable process. A buyer who wants a certificate of analysis per consignment is buying a different, recurring cost, and that should be priced into the quotation rather than absorbed. Agree the frequency in the supply agreement.

You are combining suppliers. In a like-for-like tender across leaf, moulded pulp and coated board, the coated board line will need the full three-step cascade and the leaf line will not. Comparing on quoted unit cost while ignoring the difference in evidence cost gives you the wrong answer. The way to hold that comparison steady is set out in the guide to running a like-for-like tender between leaf, pulp and board suppliers.

The honest position for some buyers is that a European converter with an existing PFAS test history on file is faster to onboard than any overseas manufacturer, whatever the material. If your project has four weeks and no regulatory resource, that is the route to take.

Frequently asked questions

Related reading

• PPWR is in force: what changed for imported foodservice packaging
• Importer or manufacturer: who carries the PPWR obligation on goods from India
• The declaration of compliance, field by field
• Migration testing for plant fibre tableware: simulants, conditions and which lab

References