Britain Just Banned the Chemical in Your Gel Manicure. Here's What Europe Actually Decided.

On Saturday, 15 August 2026, a rule quietly came into force across England, Scotland and Wales. Manufacturers can no longer place gel nail products containing TPO on the British market. If you read about it at all, you probably read a headline saying gel manicures had been banned, possibly with something about infertility attached.

Almost none of that is right, and the parts that are right are stranger than the headline version.

Here’s the short version. Gel polish is not banned anywhere. One ingredient in it is: a photoinitiator called TPO. Europe removed it in September 2025, Britain has just started removing it, and the United States has done nothing at all. The reason Europe removed it has very little to do with anyone’s manicure.

What TPO actually does

TPO is trimethylbenzoyl diphenylphosphine oxide, CAS 75980-60-8. It’s a photoinitiator, which is a molecule that absorbs UV or LED light and cracks apart into free radicals, kicking off the chain reaction that turns a liquid gel into a hard film in thirty seconds under a lamp. That flash-cure is the entire value proposition of a gel manicure. No photoinitiator, no gel.

The important physical detail, which gets lost in coverage: TPO is consumed by the process. It’s the match, not the firewood. The Australian regulatory assessment cites a study that soaked a fully cured film in saline and recovered less than 0.2% of the TPO that went in. Once the lamp has done its work, most of what you applied has chemically transformed into something else.

Hold that thought, because it makes the next part odd.

What the EU did, and the mechanism that did it

On 1 September 2025, Commission Regulation (EU) 2025/877 added TPO to Annex II of the Cosmetics Regulation, the list of substances prohibited in cosmetic products. No transition period. No sell-through. EU salons went from legal to illegal overnight, and professional use counts, because a technician applying it to a client is “making available on the market.”

But the Commission didn’t sit down and decide gel polish was dangerous. What happened is more mechanical than that.

A 2024 delegated regulation, (EU) 2024/197, reclassified TPO as a category 1B reproductive toxicant under CLP, with effect from 1 September 2025. That effective date is why the cosmetics prohibition bit on exactly the same day. Article 15(2) of the Cosmetics Regulation says that once a substance carries a CMR 1A or 1B classification, it is automatically prohibited in cosmetics. The Commission states plainly that it “has no discretion to permit continued use unless a derogation is granted.” The ban wasn’t a decision. It was a consequence.

And the classification itself came from rats. Specifically, rats fed TPO by oral gavage in 28-day and 90-day studies at doses ranging from 50 up to 1,000 mg per kilo of body weight per day, which produced testicular atrophy and reduced sperm counts at the higher levels. What doesn’t exist is repeated-dose toxicity work by any route resembling how people meet this substance: no dermal studies, no inhalation studies, and no human reproductive data at all. The only human work is short patch testing on skin and nails, which found no irritation and no sign of sensitisation under those conditions.

The part that should bother you

Europe’s own Scientific Committee on Consumer Safety had already looked at TPO in nail products. In opinion SCCS/1528/14, adopted in March 2014, it concluded that TPO “is safe when used as a nail modelling product at a concentration of at maximum 5.0%,” while noting it’s a moderate skin sensitiser.

That wasn’t a hand-wave. The committee calculated a margin of safety of 1,515. Anything above 100 is conventionally considered acceptable. This one cleared the bar by a factor of fifteen.

The 2024 reclassification voided that opinion. Not because anyone found new evidence that manicures cause harm, but because under Article 15(2) a completed safety assessment stops applying the moment a CMR 1B classification lands, unless someone requests a fresh derogation.

Which brings us to the detail that says the most about how this happened. A derogation route existed. Any company could have submitted a scientific dossier arguing that a substance consumed during curing, applied to a keratin plate, with a margin of safety in four figures, deserved an exemption. The Commission’s own Q&A confirms what happened next: “Nobody submitted a derogation request for TPO before the adoption of Regulation (EU) 2025/877.”

The industry had a door and didn’t open it.

Britain took the same data and set a different clock

Great Britain went its own way on timing. SI 2026/23, the Cosmetic Products Regulation (EC) No 1223/2009 (Restriction of Chemical Substances) (Amendment and Transitional Provisions) Regulations 2026, sets out a staged removal:

15 August 2026: TPO products can no longer be placed on the GB market. This is the date that just passed.
14 February 2027: they can no longer be made available. Distributors stop supplying, technicians stop buying.
After that: technicians in GB may keep using stock they already legally bought, until it runs out.

So Britain gave itself roughly twelve months of lag plus a six-month sell-through window that the EU refused to give anyone. Northern Ireland, meanwhile, follows the EU under the Windsor Framework, which means it went dark on TPO back in September 2025. A nail technician in Belfast and one in Liverpool have been operating under different law for eleven months.

Australia looked at the same rats and shrugged

Here’s the calibration point. Australia’s industrial chemicals regulator assessed the identical animal dataset and concluded TPO warrants Category 2, “suspected of damaging fertility.” The EU’s harmonised classification process landed on Category 1B.

That one-category gap is the whole ballgame, because 1B triggers the automatic cosmetics prohibition and Category 2 doesn’t. Same rats, same doses, same papers, different judgment call, completely different outcome for every salon on the continent.

And America?

Nothing. TPO is legal in US nail products, for consumers and professionals alike. The FDA has no restriction on it and no proceeding underway. The Modernization of Cosmetics Regulation Act, which finally gave the FDA some real cosmetics authority, contains no mechanism resembling Article 15(2). There is no automatic CMR trigger in American law, which is exactly the structural gap we’ve written about in REACH vs TSCA.

So the same bottle of gel is contraband in Dublin, on a sell-through clock in Glasgow, and perfectly ordinary in Los Angeles.

What replaced it, and the uncomfortable footnote

EU formulas moved to TPO-L, ethyl trimethylbenzoyl phenylphosphinate, along with BAPO and a few other phosphine-family photoinitiators. If you’re buying gel now, “TPO-free” on the label mostly means TPO-L inside.

Is TPO-L safer? Nobody knows, and that’s not a rhetorical flourish. It’s lawful because it carries no CMR classification, which is a different thing from having been assessed. We could find no SCCS opinion on TPO-L for nail use, no harmonised CLP classification, and no proposal for one. Supplier literature tends to argue it’s the gentler molecule on structural grounds, which is a chemistry argument rather than a toxicology one.

Read that alongside what just happened and the shape of the problem gets clear. Europe removed the one photoinitiator in this class that had a completed safety opinion finding it safe, and replaced it with close structural relatives that have no opinion at all. You can think hazard-based substitution is the right default and still notice that this particular swap traded a known quantity for an unstudied one.

The honest verdict

TPO is banned in the EU and on a phased ban in Great Britain. Both are real, both are enforceable, and if you run a salon you need to act on them.

But the ban is not evidence that gel manicures damage fertility, and anyone telling you otherwise is reading a hazard classification as if it were a risk assessment. Those are different instruments doing different jobs. Hazard asks whether a substance can cause harm under any conditions. Risk asks whether it will, under the conditions you actually meet it. Rats swallowing gram-scale doses is the first question. A photoinitiator that’s largely gone by the time your topcoat is dry is the second.

Europe answered the first question and the second one came along for the ride. That’s the system working as designed. Whether it’s the system you’d design is a fair thing to argue about.

If what you actually want to know is whether your manicure is hurting you, TPO turns out to be a minor character. We’ve written that up separately in How Safe Is a Gel Manicure, Really?, where the ingredient with genuine human evidence behind it is one almost nobody has banned.