Avobenzone: The UVA Filter That Fades in Sunlight — and Why Sunscreens Still Depend on It

What avobenzone is

Avobenzone is the name on the label for butyl methoxydibenzoylmethane (CAS 70356-09-1), an oil-soluble dibenzoylmethane derivative also known in older literature by the trade name Parsol 1789. It absorbs across the whole UVA band, 320–400 nm, with a peak at about 357 nm — that is, it covers both UVA2 (320–340 nm) and the longer UVA1 wavelengths (340–400 nm) that penetrate deepest into skin and drive photoageing. That breadth is the whole point of the ingredient, and it is why a sunscreen's SPF number tells you almost nothing about it: SPF is a UVB measure, so a product can score a high SPF from its UVB filters while its UVA cover rests almost entirely on avobenzone.

The chemistry explains both the usefulness and the fragility. In its resting state avobenzone sits mostly as an enol — a form held together by an internal hydrogen bond — and it is the enol that absorbs UVA. Absorbing UV energy can flip it into a keto form that does not absorb UVA, and from there the excited molecule can break apart instead of shedding the energy as heat. A filter that is supposed to spend all day in sunlight is therefore, in its unprotected state, consumed by the very radiation it is meant to block. Everything else in this article — the stabilisers, the reapplication advice, the regulator's caution — follows from that one awkward fact.

Tip: Judge a chemical sunscreen by its UVA story, not its SPF. Look for a broad-spectrum claim and for avobenzone (or another named UVA filter) in the actives list; a high SPF with no UVA filter is a sunburn product, not a complete sun product.

Why formulators use it

Formulators use avobenzone because, for decades in the United States, there was no good chemical alternative for long-wave UVA. A study characterising UVA protection in broad-spectrum products noted that only two ingredients approved in the US — avobenzone and zinc oxide — provide true broad-spectrum protection at wavelengths above 360 nm, and that model formulas containing 3% avobenzone gave superior attenuation above 360 nm compared with formulas containing 5% titanium dioxide. In other words, avobenzone is not a marginal extra: in a chemical sunscreen it is usually the UVA half of the product. It is also invisible on skin, with no white cast, and as an oil-soluble liquid it blends easily into light lotions, sprays and daily moisturisers — the formats people actually reapply, which matters more than any laboratory number.

That dependence is also why the June 2026 US addition of bemotrizinol to the sunscreen monograph matters (see the regulators section below): it is the first new chemical filter in the American monograph in decades, and the first real challenger to avobenzone's long hold on the American UVA job. Until bemotrizinol products actually reach shelves in number, however, a chemical sunscreen that claims broad-spectrum cover in the US is, in practice, an avobenzone sunscreen with a supporting cast.

Where you will find it

• Chemical sunscreens — lotions, sprays and sticks, where it is usually paired with UVB filters such as homosalate, octisalate or octocrylene
• Daily SPF moisturisers, primers and foundations — where a small amount of UVA protection is added to a product whose main job is not sun protection, and where the single morning application is exactly the use pattern photostability worries about most
• Lip balms and lip products with SPF — where it appears at the lower concentrations used for lip products, and where a little of it can also be swallowed
• Leave-on skin and hair products with UV claims — where it protects the product surface and the skin or hair beneath from light damage

Related: Octinoxate: The UVB Filter Behind the Reef-Sunscreen Bans — the UVB filter that, in film studies, degrades alongside avobenzone rather than stabilising it, which is why the two together are a flag, not a reassurance.

The concern: a filter that fades, absorbs, and occasionally sensitises

The photostability evidence comes in two layers: what happens to avobenzone alone, and what happens to real products on real skin. The laboratory layer is unambiguous. Avobenzone's photodegradation has been studied since Chatelain and Gabard's 2001 work showing that the broadband filter bis-ethylhexyloxyphenol methoxyphenyl triazine (Tinosorb S) stabilises it, and later reviews summarise the unstabilised picture as a loss of roughly half or more of its UVA absorbance within an hour of simulated solar radiation. The product layer is messier, which is the honest version. In a 2007 study of seven commercial sunscreens exposed to natural sunlight and to ultraviolet lamps, three products were unstable after 90 minutes of natural sun, with UVA absorbance indices between 0.41 and 0.76 — that is, they kept only 41% to 76% of their UVA absorbance — while three others stayed photostable, with UVA indices between 0.85 and 0.99. The difference between those two groups is formulation, not the ingredient alone: the same filter, stabilised well, holds; stabilised badly, it fades.

Two interactions make the fading worse. First, avobenzone and octinoxate degrade together: a film study using electron spin resonance found concomitant photolysis of avobenzone and octinoxate that predominated over the expected isomerisation, with free radicals detected that persisted even after exposure had ended. Second, avobenzone's breakdown products are not inert. A study of its photodegradation under simulated solar radiation notes that photoproducts such as arylglyoxals and benzils have been associated with photoallergic and cytotoxic reactions. That does not mean every avobenzone sunscreen produces them in meaningful amounts on skin — the study was on the filter in solution and in formulations under laboratory irradiation — but it gives the photoallergy reports below a plausible mechanism, rather than leaving them as unexplained anecdotes.

The stabiliser story is the reassuring half of the same evidence, and it should be stated plainly. The UVA-protection study that found 3% avobenzone superior above 360 nm also found that judicious selection of actives and stabilising agents maintained avobenzone's photostability over four hours of UV exposure, and that small losses of avobenzone (under 20%) did not significantly reduce a product's protective effects as measured by SPF and by UVA protection factor on human skin. In the US the classic stabiliser is octocrylene, which quenches avobenzone's excited triplet state; in Europe, bemotrizinol (Tinosorb S) does the same job, as Chatelain and Gabard showed in 2001. Antioxidants have been studied in the same role: in the photodegradation study above, vitamin E, vitamin C and ubiquinone each improved avobenzone's photostability, with ubiquinone also raising SPF in formulations. The practical point for a shopper is that "contains avobenzone" is not a complete description of a sunscreen. A stabilised avobenzone formula and an unstabilised one are, for UVA purposes, different products wearing the same active name — and the ingredient list, which does name the stabiliser when it is another UV filter, is the only place the difference shows.

The absorption story is smaller for avobenzone than for its neighbours, and that matters for proportion. In the randomised trial by Matta and colleagues, published in JAMA in 2020, 48 healthy adults applied commercial sunscreens to 75% of their body surface at 2 mg per square centimetre, once on the first day and four times daily on the following days. Avobenzone was the trial's primary endpoint, and its geometric mean maximum plasma concentrations were 7.1 ng/mL from lotion, 3.5 ng/mL from aerosol spray, 3.5 ng/mL from nonaerosol spray and 3.3 ng/mL from pump spray — all above the FDA's 0.5 ng/mL threshold that triggers further safety testing, and passed on day one after a single application. For scale, those numbers sit far below oxybenzone's 258.1 and 180.1 ng/mL in the same trial, and below homosalate's 13.9 to 23.1 ng/mL. The threshold, again, is a trigger for more data, not a safety limit, and the authors were explicit that their findings do not indicate that individuals should refrain from using sunscreen. A 2025 toxicology review adds the other half of the exposure picture: clinical data indicate low percutaneous absorption of avobenzone in humans, at or below 0.59% of the applied dose, with limited excretion in urine and four tentative metabolites identified.

The allergy story splits by who is counted, and the split explains why two honest sources can sound so different. A 2024 narrative review for clinicians reports contact sensitivity and positive patch and photopatch tests for the dibenzoylmethanes from Germany, England, France, Italy and the USA, making avobenzone the second most common sunscreen allergen after benzophenone-3 (oxybenzone) — but it also notes that allergic reactions to sunscreens were relatively uncommon compared with their widespread use over four decades. A 2025 retrospective of dermal safety studies, covering about 39,000 participants in repeat patch testing, found that 99.5% experienced no sensitisation, that the seven reactions seen were attributed to a preservative rather than to the UV filters, and that no photoallergy or phototoxicity reactions were observed. Both can be true: patch-test clinics count people who were already suspected of allergy, so their rates run high; clinical safety studies count general volunteers, so theirs run low. The 2025 review sits between them: avobenzone generally did not cause dermal irritation or sensitisation, but indications of photoallergy have been reported in clinical case studies. For a person with no history of sunscreen reaction, allergy is a minor risk. For a person whose rash appears only where sunscreen met sunlight, it is the first ingredient to test.

Related: Homosalate: The UVB Filter Europe Cut Back to Face Products Only — the filter most often paired with avobenzone in American sunscreens, whose own absorption numbers ran higher in the same FDA trial.

What regulators and reviewers say

Europe has the simplest position: avobenzone is a permitted UV filter under Regulation (EC) No 1223/2009, Annex VI, at a maximum of 5% in the ready-to-use product, and it has been in use there since the late 1970s. The United States is more complicated, and the complication is procedural rather than toxicological. Under the deemed final order created by the CARES Act, sunscreens containing 16 active ingredients, avobenzone included, are treated as generally recognised as safe and effective under the 1999 monograph. The FDA's 2021 proposed order (OTC000008), which restates the 2019 proposed rule, proposes the opposite for avobenzone and eleven other chemical filters: not generally recognised as safe and effective, because additional data are needed to show that they are. The FDA's own questions-and-answers page stresses what that proposal is not: it does not represent a conclusion that those ingredients are unsafe, but a request for additional information in light of changed conditions, including substantially increased sunscreen usage. Only two ingredients from that proposal have been finalised so far — PABA and trolamine salicylate, removed by a 2026 final order — and the FDA says it will address the other proposals, avobenzone included, in a future order or orders. In the meantime the American ceiling for avobenzone in a sunscreen remains 3%, as the 2025 toxicology review states for the United States and Canada, and as marketed Drug Facts panels showing "Avobenzone 3%" confirm.

Australia, which regulates sunscreens as therapeutic goods, landed in 2025 on the reassuring side for this ingredient. The Therapeutic Goods Administration's safety review of seven active sunscreen ingredients judged five of them low risk and appropriate for use, avobenzone among them: its margin of safety, calculated with the Australian Sunscreen Exposure Model, was above 100 at current maximum concentrations, and the review concluded these ingredients are unlikely to cause any significant systemic toxicity when used in therapeutic sunscreens. The same review recommended tighter controls only for homosalate and oxybenzone. It also stated its limitations plainly, and they apply to avobenzone too: it depended on published assessment reports rather than raw study data, it did not consider consumer products other than sunscreens, it did not consider metabolites or impurities, and on hormones it said the available information on avobenzone and four other filters indicates potential endocrine effects, but the data are not adequate to derive a conclusion as to causality in humans.

The 2025 comprehensive toxicology review of avobenzone, published in Critical Reviews in Toxicology, goes further in the reassuring direction — and its authorship should be read alongside its conclusions. Several authors were affiliated with the Personal Care Products Council, with dsm-firmenich, or with Procter and Gamble, alongside academic toxicologists. With that noted, its findings are specific and checkable: no evidence of effects on the oestrogen, androgen or thyroid systems; no indication of genotoxicity in vitro or in vivo; no formal two-year carcinogenicity study, but a 90-day dietary study in rats that showed no increase in hyperplasia in any tissue and no evidence of cytotoxicity; a no-observed-adverse-effect level of 450 mg per kilogram per day from a rat dietary subchronic study; and, on conservative exposure assumptions, margins of exposure above 100 at the permitted maximum in the United States and Canada. Its conclusion is that avobenzone is unlikely to pose a risk to human health when used in sunscreen products at permitted maxima. The advocacy database EWG Skin Deep (labelled here as an advocacy source, not a regulator) sits on the cautious side of the same evidence in a different way: its product listings rate avobenzone low for cancer, developmental and reproductive toxicity, allergy, endocrine disruption and irritation, moderate for use restrictions, and flag contamination concerns naming degradation-related compounds. EWG's profile is a hazard ranking, not a regulator's risk assessment, and the difference between those two things — hazard in a dish versus risk at a real dose — is the whole debate in miniature.

The debate

The position this site takes follows the regulators rather than the slogans: avobenzone at sunscreen concentrations is not shown to harm people, its fading in sunlight is real and formulation-dependent rather than a reason to fear the molecule, and the hormone question is genuinely unresolved in the regulators' framing — "not adequate to conclude causality" — even though the newest review found no signal. If you use a chemical sunscreen, the more useful questions are whether it names a stabiliser, whether you reapply it, and whether you apply enough of it — not whether one ingredient on its label appears in a headline. And if you want the photostability question to disappear entirely, mineral sunscreen answers it outright.

Who should avoid it, or ask first

• People with a diagnosed sunscreen allergy, or a rash that appears only where sunscreen met sunlight — avobenzone is the second most common sunscreen allergen in patch-test literature after oxybenzone, and photopatch testing, not an ordinary patch test, is what catches photoallergy
• People who spend long hours in sun and reapply rarely — the fading problem bites hardest on a single morning application left to face a full day outdoors; a photostable or mineral formula, or disciplined reapplication, is the direct answer
• People whose sunscreen also lists octinoxate — the pair is known to degrade together in film studies, so a formula listing both is a reason to prefer a stabilised or mineral alternative, not a reason to assume the SPF on the front still holds at midday
• People who want the simplest systemic-exposure profile — mineral actives sit on the skin's surface, so the Matta absorption numbers simply do not arise
• Anyone who wants the newest American option — bemotrizinol joined the US monograph in June 2026, so products listing it are worth watching for as the first photostable chemical alternative to avobenzone in the US market

How to spot avobenzone on a label

• Avobenzone — the name on US Drug Facts panels, where it appears as an active ingredient with its percentage, up to 3%
• Butyl Methoxydibenzoylmethane — the same ingredient in INCI lists in Europe, Australia and most of the world, where no percentage is printed and its position in the list hints at the amount
• Parsol 1789 — a trade name for the same compound, still seen in older literature and some markets
• Its usual stabilisers — Octocrylene (the classic US stabiliser), Bemotrizinol, Bis-Ethylhexyloxyphenol Methoxyphenyl Triazine (Tinosorb S), and Methylene Bis-Benzotriazolyl Tetramethylbutylphenol (Tinosorb M) in products sold outside the US; a formula that names avobenzone with none of these is the one to question
• Its usual companions — Homosalate, Octisalate and Octocrylene (UVB partners in the same blend); Octinoxate in the same list is the pairing film studies show degrading together
• The broad-spectrum trap — a broad-spectrum claim certifies UVA cover at the time of testing, not at hour four on a beach; the reapplication interval on the pack, not the claim on the front, is the guard against a faded filter

Alternatives: the same sun protection, without this filter

The simplest alternative removes the entire debate rather than one ingredient from it: a mineral sunscreen whose only active ingredients are zinc oxide and/or titanium dioxide. Mineral filters scatter and absorb UV on the skin's surface, they carry the strongest regulatory safety standing of any sunscreen actives, and they avoid both the absorption question and the photostability question. The honest trade-offs are the ones our sunscreen guide works through: a possible white cast, especially on deeper skin tones, a heavier feel than a chemical lotion, and the need to apply generously, since a thin, invisible layer of mineral sunscreen protects far less than its label SPF promises.

Two practical notes close the loop. First, the chemical alternative has changed recently: bemotrizinol, a photostable broad-spectrum filter long used outside the US, was added to the American monograph in June 2026, so American shoppers may soon see it listed as an active in its own right rather than only as a stabiliser in European formulas. Second, as with every free-from claim elsewhere on this site, a product whose only virtue is what it leaves out should still be judged on what it put in instead: an avobenzone-free sunscreen with weak UVA cover has not improved your situation by dropping one filter. Read the whole actives list, check the broad-spectrum claim, and let the bottle you will actually reapply be the final judge.

Checked against Matta et al. 2020, the TGA safety review of seven sunscreen ingredients (2025), Norman et al. 2025, the FDA questions and answers on over-the-counter sunscreen, Regulation (EC) No 1223/2009 Annex VI, and the EWG Skin Deep database (an advocacy source) · Last updated 2026-11-01 · How we assess ingredients

Read more

Octinoxate: The UVB Filter Behind the Reef-Sunscreen BansHomosalate: The UVB Filter Europe Cut Back to Face Products OnlyZinc Oxide: Broad-Spectrum Mineral Protection, the White-Cast Trade-Off, and Who It Suits