EU PFHxA Limits Begin: 25 ppb Is an Impurity Ceiling

What is the signal?

The legal basis is Commission Regulation (EU) 2024/2462 of 19 September 2024. It added entry 79 to Annex XVII of REACH (Regulation (EC) No 1907/2006), restricting the use and the placing on the market of PFHxA (undecafluorohexanoic acid), its salts, and PFHxA-related substances. The text on EUR-Lex states that from 10 October 2026 these shall not be placed on the market, or used, at a concentration equal to or greater than 25 ppb for the sum of PFHxA and its salts, or 1,000 ppb for the sum of PFHxA-related substances, measured in homogeneous material. Cosmetic products fall into that October 2026 group, and the recitals describe the 24-month deferral granted to cosmetics.

So what changed this week is not the rule but the end of its transition. The regulation was published two years ago, and the transition period expired on 10 October 2026. Reporting from Personal Care Insights and HPC Magazine carries the same point: the 25 ppb and 1,000 ppb limits apply to cosmetics from 10 October.

Read the nature of the threshold first. 25 ppb is on the order of 25 micrograms per kilogram. Personal Care Insights reported that PFHxA is not intentionally added as a primary cosmetic ingredient but appears as a chemical impurity, or as a breakdown product of other intentionally added PFAS. The same coverage noted that because the thresholds sit at trace level, companies may need to check raw materials for impurities and not only for intentionally added ingredients.

The signal from the U.S. runs in a different direction. In its congressionally mandated report under MoCRA, 'Report on the Use of PFAS in Cosmetic Products and Associated Risks', the FDA said it lacked the data to reach definitive safety determinations for PFAS used in cosmetics. Based on cosmetic product listing data, 51 PFAS are used in 1,744 formulations, and the agency focused on the 25 most frequently used, representing roughly 96% of PFAS intentionally added to cosmetic products. Its conclusion was that toxicological data for a majority of these substances are incomplete or unavailable.

Let me set the limits of this reading. The 25 ppb and 1,000 ppb figures, the 10 October 2026 date and the homogeneous-material basis were confirmed as sentences of the regulation itself in the EUR-Lex results, but the listing of cosmetic products among the covered items was confirmed from another language version of the same regulation rather than the English text. The explanation of how existing stock is exempt comes from trade reporting, not from the regulation. The FDA's 51 substances and 1,744 formulations are a tally of listing data submitted by companies, which may differ from actual use, and 'insufficient data' means neither safe nor unsafe. The two sets of numbers belong to different regimes and cannot be added or compared. For a formula bound for Europe, though, the conclusion converges: the market that has put a number on the limit, not the one that has deferred, sets the working ceiling.

1. "We did not add it" is not an answer to 25 ppb

When a banned substance list grows, the usual response is to remove that substance from the formula and re-check the ingredient list. This limit does not end there, because what it measures is not the intentionally added ingredient but the concentration inside the finished material. A trace amount that arrived with a raw material, without ever being added on purpose, is enough to stop the product going to market.

The unit of verification therefore moves down from the formula to the raw material. The work is to see which materials in the formula passed through a process using fluorinated chemistry, or through a supply chain that uses fluorinated surfactants or water-repellent treatments. Surface-treated pigments, silicone film formers, water-repellency aids and some processing aids are the points usually raised. Which materials actually apply can only be established from supplier documentation, so building a questionnaire for suppliers is faster than assembling a list from assumptions.

Read the unit of the limit alongside it. The regulation measures concentration in homogeneous material. That is not an average taken across a diluted finished product, and in multi-layer or multi-phase products, or products with packaging components, which part counts as the unit of measurement decides the result. This interpretation belongs on the list of questions for your testing laboratory and local regulatory contact.

• Whether suppliers confirmed in writing, per material, whether fluorinated chemistry was used
• Whether pigment surface treatments, film formers and processing aids are inside the question's scope
• Whether the unit of measurement was checked as homogeneous material rather than the finished product
• Whether a retest and alternative-material route is defined for results near the limit

2. Stock placed before 10 October and shipments after it are different

According to Personal Care Insights, the restriction does not apply to products placed on the EU market before 10 October 2026. That includes products already supplied to a distributor or retailer and products already imported into the EU. Products still held by the manufacturer, or not yet imported, must meet the new limits before entering the market.

In practice this distinction surfaces as a scheduling question rather than an inventory one. Even within one production lot, the volume that cleared customs in early October and the volume shipping in November fall under different rules. So the first thing to establish is less 'does our product contain PFHxA' than 'when does the next shipment enter'.

Where responsibility sits also shifts with the trading structure. In Europe the party placing a cosmetic on the market is the Responsible Person, and the exporting manufacturer supplies the documentation that party requires. If the contract defines that supply as reaching only the ingredient list and the safety assessment, it leaves open who pays for testing when a request arrives at impurity level.

• Whether the next EU shipping and customs clearance dates are written down per item
• Whether volumes already in market are separated from volumes still on hand
• Whether the documentation scope the Responsible Person will request was checked in the contract
• Whether the cost and owner of impurity testing are written into the trading terms

3. INCI names alone will not find the substance

The first practical obstacle is searching. Regulation names substances through chemical nomenclature and CAS numbers; a cosmetic ingredient list uses INCI names. Coverage of this restriction pointed out that the difficulty of linking INCI names with regulatory substances makes screening harder. Scanning an ingredient list by eye to pick out the substance is not a method you can rely on.

There are still clues. Reporting noted that ingredients whose names contain 'perfluorohexyl' can serve as an identification cue. In its conclusion table, the FDA report grouped the substances and flagged one, perfluorohexylethyl triethoxysilane, as a potential concern at the highest concentration used in body lotion. That names containing 'perfluorohexyl' appear in both documents makes them a usable starting point for screening.

A starting point is all it is. Routes of entry as an impurity do not show up in a name. Verification has to run on two tracks: picking out fluorinated ingredients from the ingredient list, and asking raw material suppliers about fluorinated processes even for formulas where no such ingredient appears. The first can be done in house and the second needs supplier cooperation, so their lead times differ.

• Whether ingredients containing 'perfluoro' or 'fluoro' were pulled from the list first
• Whether a route exists to convert INCI names into CAS numbers for checking
• Whether supplier checks were also run on formulas with no fluorinated ingredient
• Whether the time suppliers take to reply is reflected in the development schedule

4. The U.S. answered the same question with "not enough data"

In the same period the U.S. conclusion was a reservation rather than a number. The FDA said it had insufficient data to reach definitive safety determinations for PFAS in cosmetics, citing limited data on use levels, the absence of dermal and oral absorption data, and a lack of mechanistic information. In the FDA's announcement the Commissioner said toxicological data for most PFAS are incomplete or unavailable, leaving significant uncertainty about consumer safety. The agency says it will keep monitoring emerging data.

Reading the two regimes together simplifies the development decision. One has set a number, the other has deferred judgement. A formula shared across both markets has to follow the one with a number. That the U.S. has no prohibition on an ingredient is not grounds for using it, and conversely a formula meeting the European limit creates no problem in the U.S.

If the product is going to several markets, the choice splits in two here: separate formulas per market, or one formula built to the strictest limit. Separate formulas only have to meet each market's own rule, but you end up with two sets of material codes, production lots and test records. A single formula keeps management simple at the cost of narrowing the materials you may use. The smaller the volume, the more often the single formula wins.

• Whether the choice between a shared and a split formula was made at the planning stage
• Whether the added material codes and test records of a split formula were costed
• Whether alternatives exist for materials dropped to meet the strictest limit
• Whether a market without a rule was written down as a market that permits it

5. The next stage is the universal PFAS restriction

This measure is not a ban on PFAS as a whole. Reporting notes that the EU has been explicit that this is not a blanket ban on all PFAS but targets one specific group of substances and defined uses. The proposed universal restriction on the whole PFAS group is being assessed separately by ECHA.

What matters for cosmetics is the direction of that assessment. According to Personal Care Insights, SEAC's draft opinion on the universal restriction concluded that no derogation for cosmetics is justified, with a final opinion expected by the end of 2026. An opinion is not the restriction itself, but it suggests that even if another transition period follows, the direction is already set.

That changes how to judge the present task. Clearing only this 25 ppb limit may mean running the same review again in a few years. Writing down now what fluorinated materials actually do in the formula, by contrast, is work that carries over to the next stage. Record the function each such material carries — water repellency, wear, a smooth feel — and check in advance whether a replacement route exists for each function.

There is no need to rush the replacement. A substitution that fails to hold the function returns as returns and redevelopment. But a formula that has an answer written down to 'what breaks if this material is removed' and one that does not will take very different amounts of time when the rules move again.

• Whether the function each fluorinated material carries is written down per formula
• Whether a replacement route was checked, even in one line, for each function
• Whether the stability and sensory testing a substitution would require was estimated
• Whether the timetable of the universal restriction is marked on the development roadmap

The first question for the brief

Instead of writing 'for export to Europe', add a date. Put in one line the day of your next EU shipment and which items go with it, and the range of items to check now narrows immediately. Volumes already in market and volumes still on hand follow different rules even when they are the same product.

The second thing to write is one column beside the raw material list. For every material in the formula, mark yes or no against 'has the supplier confirmed in writing whether a fluorinated process was used'. The cells still marked no are your risk list. The absence of a fluorinated ingredient from the list does not fill that column.

The last is one line on function. If you are using a fluorinated material, write down whether its job in this product is water repellency, wear or texture, and record the replacement route in the same cell. While the universal PFAS restriction is under discussion, that single line decides whether redevelopment comes sooner or later.