Chemicals in Sunscreen: What's Actually In Your Tube and How to Choose

When the FDAFood and Drug Administration ran a clinical trial on those names in 2020, every one of the six chemical filters tested crossed the agency's own threshold for ordering further safety studies within hours of a single application. 6 of 6 chemical filters tested in Matta 2020 exceeded the FDA's 0.5 ng/mL further-study threshold on day 1 — JAMA, n=48 of the active ingredients in chemical sunscreen passed into the bloodstream from one day's use. Some were still circulating three weeks later Matta et al. 2020. No mineral filter has ever crossed that threshold in any study, because mineral filters don't penetrate the skin in the first place.

This article is about which set of chemicals you choose, not whether to wear sunscreen. UVUltraviolet radiation is an IARCInternational Agency for Research on Cancer Group 1 carcinogen, it causes skin cancer, and protection works. The argument is narrower: chemical filters absorb into the body and persist; mineral filters sit inert on top of the skin; both protect equivalently against UV; the price gap is small and shrinking. It's part of the endocrine-disruptor story this site keeps coming back to, and the eso-friendly position is the same one the FDA proposed in 2019 — only zinc oxide and titanium dioxide have the safety data to clear Category I.

Sunscreen actives split into two classes. Chemical filtersOrganic-molecule UV absorbers — they take in ultraviolet photons, undergo molecular excitation, and re-emit the energy as small amounts of heat. They sit inside the skin to do their job. absorb UV inside the skin and are the dominant active ingredient in most US, UK and EU products. Mineral filtersInorganic UV reflectors — physical barriers that scatter and reflect UV photons rather than absorbing them. They sit on top of the skin. reflect UV from the surface. The FDA's 2019 monograph review classified 12 of 16 chemical filters reviewed by the FDA were placed in Category III — 'insufficient data' — in the 2019 proposed rule — 84 FR 6204 chemical filters as Category III ('insufficient data') and only zinc oxide and titanium dioxide as Category I — generally recognised as safe and effective up to 25% concentration US FDA 2019.

The active-ingredients box on the back label is the only thing that matters. Front-of-pack words like 'natural,' 'reef-safe,' 'mineral-based' and 'clean' have no regulatory definition in any market — a 'mineral-based' sunscreen can legally contain chemical filters, and 'reef-safe' has been ruled misleading by both the FTCUS Federal Trade Commission and the UK ASAAdvertising Standards Authority. The back-label active list cannot lie.

Two things in that table do most of the work. The first is that the chemical filters are everywhere — oxybenzone alone appears in roughly 2,500 US sunscreens and continues to show up in lipsticks, foundations, BB creams, hairsprays and nail polishes for non-SPF reasons (it stabilises colour and fragrance). Octocrylene is in around 2,999 octocrylene-containing SPF products marketed in the US in 2019, per Downs et al. 2021. The second is the regulatory split: of the sixteen chemical filters in the FDA's 1999 monograph, only the two minerals cleared the agency's safety bar. Nine of the twelve Category III filters have been on US shelves for over forty years on the strength of pre-1972 GRASE assumptions and not much else.

Forty-one years after the FDA approved the first chemical sunscreens for over-the-counter sale, the agency ran the first proper pharmacokinetic trial of how much of them ends up in the bloodstream. Matta and colleagues published a pilot RCTRandomised Controlled Trial in JAMAJournal of the American Medical Association in 2019 (n=24, four filters), then a larger follow-up in 2020 (n=48, six filters across four product types). The headline result: every chemical filter tested crossed the FDA's own further-study threshold within hours of a single application Matta et al. 2020.

Here's the sequence in the 2020 study. Forty-eight participants arrived at a phase-1 clinical unit, had a baseline blood draw, and were randomised to one of four sunscreen formulations — a lotion, an aerosol spray, a non-aerosol spray and a pump spray. A trained applicator coated 75% of each participant's body — face, arms, legs, torso — at 2 mg per square centimetre, the dose the FDA's labelling assumes consumers actually use. Day 1 was a single application; days 2 to 4 were four applications a day. Blood was drawn at fixed intervals across 21 days, more than thirty samples per participant. Two hours after the first application, every one of the six tested filters had a geometric-mean plasma concentration above 0.5 ng/mL. Oxybenzone reached 258× ng/mL on the lotion arm; avobenzone reached 7.1 ng/mL; octocrylene, homosalate, octinoxate and octisalate were all above the threshold Matta et al. 2020.

The 0.5 ng/mL threshold itself deserves a careful sentence, because it is widely misframed. It is not a 'safety limit.' It comes from the FDA's November 2016 nonbinding guidance for industry on maximal-usage trials of topical drug products — the level above which the agency considers additional non-clinical toxicology studies necessary before it can stand behind a product's safety. Verbatim from the guidance: 'non-clinical toxicity studies may be waived if an adequately conducted human PK MUsT results in a steady-state blood level of <0.5 ng/mL.' Crossing the threshold doesn't mean a chemical is dangerous. It means the regulator's own evidentiary bar requires more data — data that for most of these filters does not yet exist.

The persistence pattern is where the filters separate. Most cleared the bloodstream within a week to ten days after the dosing stopped. Two didn't. Oxybenzone was still above the threshold in 96% of participants three weeks after a single application phase — Matta 2020 of participants at day 21, and homosalate was above it in 55%. Octocrylene, avobenzone, octinoxate and octisalate were all back below by day 7 to 14. The difference matters because the 'sunscreen day' picture in your head — beach, holiday, occasional summer use — is not the picture in the typical consumer's bloodstream, which now reflects daily SPF moisturisers and chemical-filter cosmetics worn year-round.

Beyond absorption, each chemical filter has its own evidence trail. The class shares an estrogenic signal in laboratory and animal work, but the specific signals — and the specific concerns — differ between filters. The FDA's 2019 review covered them as a group. The EU's Scientific Committee on Consumer Safety has been reviewing them one by one and reducing concentration limits.

Oxybenzone (benzophenone-3, BP-3) is the most-studied filter and gets its own profile here. It carries the heaviest evidence load: 96% of NHANES urine samples carry it Calafat et al. 2008, it crosses the placenta and appears in breast milk Krause et al. 2018, in vitro and rodent assays show estrogenic and anti-androgenic activity, and it is restricted in baby bottles and capped in cosmetics across the EU. It is also the filter most-cited in the coral-reef story, though the field-evidence quality control could be stronger than it has been. Of all the chemical filters in the active-ingredients box, this is the one you most want to leave on the shelf.

Octinoxate (octyl methoxycinnamate, OMC) is the workhorse UVB absorber and the second filter banned alongside oxybenzone in Hawaii. The estrogenic signal here is older than the absorption story — Schlumpf and colleagues showed in 2001 that OMC produced a uterotrophic response in immature rats at oral doses overlapping with consumer biomonitoring ranges, in the first systematic UV-filter estrogenicity screen Schlumpf et al. 2001. A larger 2008 review confirmed multi-generational endocrine effects in rat developmental studies for the closely related camphor-based filters, with measurable LOAELLowest Observed Adverse Effect Level — the lowest dose at which a study finds a measurable adverse effect on a tested endpoint.s as low as 7 mg/kg/day Schlumpf et al. 2008. A 2024 Mediterranean field study found that octinoxate at concentrations as low as 30 ng/Lthe lowest of three environmentally relevant Mediterranean concentrations tested in Posidonia oceanica seagrass — García-Márquez & Agawin 2024 damaged the seagrass Posidonia oceanicaA Mediterranean seagrass that forms underwater meadows critical to reef-equivalent ecosystems. Sometimes called Neptune grass., suppressing chlorophyll, primary production and nitrogen fixation García-Márquez and Agawin 2024.

Homosalate (HMS) is a UVB absorber that the EU's SCCSScientific Committee on Consumer Safety reviewed in 2020 and concluded was not safe at the previously permitted 10% concentration. Regulation 2022/1176 reduced it to a maximum of 7.34% in face products only — body products have to use other filters — with the cap applying from 1 July 2025. SCCS Opinion SCCS/1622/20 on homosalate, finalised 24-25 June 2020. The reasoning was a MoSMargin of Safety calculation: at 10% body-product use, the realistic-exposure modelling didn't leave enough headroom over no-effect levels in animal studies.

Avobenzone (butyl methoxydibenzoylmethane, BMDM) is the dominant UVA absorber in chemical formulations, and it is intrinsically photo-unstable. Without a photostabiliser it loses 50–90% of its UV-A absorbance after one hour of UV exposure — which is why almost every chemical sunscreen pairs it with octocrylene. In one published comparison, 4% avobenzone retained only 23% of its photostability after 25 minimal-erythema-dose exposures alone; adding 3.6% octocrylene pushed retention to 90%. The inconvenient corollary: the filter doing the photostabiliser job is itself a Category III ingredient with its own degradation problem.

Octocrylene is the avobenzone partner, and the chemistry of that pairing has a wrinkle nobody mentions on the label. Downs and colleagues bought 17 commercial sunscreens (9 EU, 8 US), tested for benzophenoneA breakdown product of octocrylene generated by retro-aldol degradation on the shelf. IARC Group 2B (possibly carcinogenic to humans, 2013)., and found it in every octocrylene-containing product. Recently-purchased products averaged 39 mg/kg benzophenone (range 6–186); after six weeks of FDA-accelerated stability aging, the average rose to 75× mg/kg with a maximum of 435 mg/kg. The single octocrylene-free product they tested was always non-detect. The mechanism is a retro-aldol condensation that runs slowly at shelf temperature: octocrylene degrades into benzophenone, which the IARCInternational Agency for Research on Cancer classified as a Group 2B (possible) human carcinogen in 2013, and which Downs estimated could be up to 70% dermally absorbed Downs et al. 2021. That is to say: the SPF 50 lotion in your medicine cabinet from last summer is, on average, a more concerning chemical product than the same lotion was when you bought it.

Across the chemical class, biomonitoring picks up the cumulative picture better than any single filter. Krause and colleagues' 2012 review pooled the available human data and reported UV-filter detection in 96% of US urine samples and 85% of Swiss breast milk samples Krause et al. 2012; Schlumpf's 2010 milk study traced that detection most strongly to cosmetic-product use of 4-methylbenzylidene camphor and octocrylene, in 85.2% of the 54 mother-child pairs tested Schlumpf et al. 2010. The placental and breast-milk transfer evidence is exactly the kind of finding that turns a daily-cosmetic exposure into a multi-generational one.

Mineral sunscreens use zinc oxide or titanium dioxide as the active ingredient — sometimes both. Both are GRASEGenerally Recognised As Safe and Effective-classified by the FDA at concentrations up to 25%. The safety case rests on two studies that asked the obvious question: do the particles actually penetrate intact human skin? The answers are remarkably consistent, including for the nanoparticle-sized formulations that worried regulators in the early 2010s.

Three sunscreen formulations were tested on human volunteers in 2009 — including one with 20-nm-coated titanium dioxide. After two hours of exposure (and up to 48 hours under occlusion), nuclear-microscopy quantification found zinc and titanium only at the very surface of the stratum corneum and in the openings of pilosebaceous follicles. The viable epidermis and deeper layers were devoid of both metals, even after 48 hours of occluded exposure Filipe et al. 2009. The FDA's own National Center for Toxicological Research extended the question in a minipig model — closer to human skin than rodents — and found the same answer: no significant dermal penetration of nano- or submicron TiO2 from sunscreen formulations applied repeatedly over 22 days Sadrieh et al. 2010. A small number of isolated TiO2 particles appeared in the dermis but with no consistent pattern, attributed by the authors to incidental contamination rather than penetration.

The frequent counter-argument — but isn't titanium dioxide banned in EU food now? — needs a precise answer. Yes, but for a different reason and a different exposure route. EFSAEuropean Food Safety Authority's 2021 opinion concluded that nano-titanium-dioxide as the food additive E171 could not be ruled out for genotoxicity from oral exposure, and Regulation (EU) 2022/63 banned it as a food additive from 7 August 2022 EFSA 2021. That ban applies to food, not to sunscreen. The mechanism of concern (oral absorption + gut interaction with nanoparticles) is unrelated to the topical question, and the 2021 SCCS review of TiO2 in cosmetics — including sunscreen — reached a different conclusion: safe up to 25% for non-spray products, with restrictions on spray and powder formats. Conflating the two is one of the most common factual errors in popular sunscreen articles. SCCS Opinion SCCS/1583/17 on titanium dioxide nano in cosmetics; subsequent assessments have re-affirmed the topical safety position.

The remaining argument against mineral has historically been cosmetic — the white cast on darker skin and the heavier feel against the face. That gap has closed. Tinted mineral formulations now match a full range of skin tones, and modern non-nano zinc-oxide dispersions sit substantially closer to invisible than the chalky pastes of fifteen years ago. The 'mineral feels worse' objection in 2026 is mostly a hangover from the 2010s formulations, not a structural feature of the chemistry.

Often not — and this is one of the cleanest pieces of consumer dermatology evidence available. SPF testing is conducted at 2 mg of sunscreen per square centimetre of skin. Real-world application thickness in published behavioural studies is between 0.39 and 1.0 mg per square centimetre — between a fifth and half the lab dose Petersen and Wulf 2014. The relationship between thickness and delivered SPF is not linear; it is roughly exponential.

An in vivo measurement of the relationship in 2007 found that applying 1 mg/cm² (half the recommended amount) makes the SPF fall to roughly the square root of the label number; applying 0.5 mg/cm² (a quarter) makes it fall to the fourth root Faurschou and Wulf 2007. In practical terms: a labelled SPF 50 product applied at half-thickness delivers somewhere around SPF 7. Applied at quarter-thickness, it delivers closer to SPF 3. That is an order of magnitude less protection than the front of the bottle promised. The fix is mechanical — apply more, reapply, and don't trust a single morning application to last all day — rather than chemical.

There's also a UVA-versus-UVB asymmetry that the SPF number doesn't capture at all. UVAUltraviolet A, longer-wavelength UV penetrating to the dermis drives skin ageing and indirect DNA damage; UVB drives sunburn. SPF measures UVB protection. Look for a UVA-PF rating, a UVA circle logo (EU), four- or five-star UVA ratings (UK Boots system), or PA+++ to PA++++ (Asian markets) — not just the SPF figure on the front.

The regulatory picture splits along the same line oxybenzone's regulatory picture does. The EU has moved on the cosmetics side, slowly. The US has moved on the local-cosmetics side and stalled federally. The Pacific island and Caribbean jurisdictions have moved on the environmental side. No major Western market has banned chemical filters outright — every restriction is concentration-capped or product-category-specific. The simplest read of the regulatory map is that nine of twelve Category III filters remain on US shelves under interim status while reviewers slowly work through them one at a time.

On the EU side, Commission Regulation (EU) 2022/1176, adopted 7 July 2022, amended Annex VI of the Cosmetics Regulation 1223/2009. The headline restrictions: oxybenzone capped at 6% in face products and 2.2% in body products from 28 July 2023, octocrylene reduced to 10% (9% in propellant sprays), and homosalate restricted to 7.34% in face products only with the cap applying from 1 July 2025. The reductions were driven by SCCS opinions concluding the previous limits did not give enough margin of safety under realistic-use exposure modelling — even where the human-clinical evidence base remained inconclusive. The regulator's line: the lab signal is real, the clinical consequence is not pinned down, and the cap comes down anyway.

On the US federal side, the FDA's February 2019 Proposed Rule (84 FR 6204) classified twelve chemical filters as Category III, two (PABAPara-aminobenzoic acid — an early UV-B filter mostly removed from sunscreens because of high allergic-contact dermatitis rates. and trolamine salicylate) as Category II (non-GRASE), and only zinc oxide and titanium dioxide as Category I. Under the CARES ActCoronavirus Aid, Relief, and Economic Security Act 2020 — the legislation that rolled the proposed rule into a deemed-final-order interim status pending further FDA review. of 2020, the proposed rule was rolled into a September 2021 administrative order, which has not been finalised as of April 2026. Existing chemical-filter sunscreens remain legally marketable in the United States under deemed-final-order interim status — which means American shelves still carry products built around twelve chemicals the agency itself classified as 'insufficient data' over seven years ago.

The Pacific and Caribbean jurisdictions read the science earlier and acted faster. Hawaii's Act 104 of 2018 banned in-state sale of sunscreens containing oxybenzone or octinoxate from 1 January 2021. An attempt to add avobenzone and octocrylene via SB 132 in 2021 passed the Hawaii Senate but died in House committee. Maui County went further in 2022 with Ordinance 5306, banning all non-mineral filters at the county level. The US Virgin Islands' Act 8185 was the first jurisdiction globally to add octocrylene to an oxybenzone-octinoxate ban, with the sale prohibition taking effect 30 March 2020. Palau's Responsible Tourism Education Act, in force from 1 January 2020, listed ten banned ingredients including every chemical filter the FDA had questioned and the four parabens widely used in cosmetics formulation.

The simplest decision rule is also the most defensible. Read the active-ingredients box on the back of the bottle. If 'zinc oxide' or 'titanium dioxide' (or both) is the only active listed, the product is mineral — switch into it. If any of the twelve Category III filters appear, the product is chemical — and there is now a mineral alternative on the same shelf at a comparable price for almost every sunscreen format. The Switch is straightforward. No part of this requires you to memorise a list of fifteen chemical names — the box does the work for you.

What this article does not say is 'stop wearing sunscreen.' UV protection prevents skin cancer, and the dermatology profession has been right to keep saying so loudly. The argument is narrower: if a chemical filter and a mineral filter give you the same UV protection, but the chemical filter is in your bloodstream above the FDA's own further-study threshold for three weeks after a single application — and the mineral filter sits inert on top of your skin — the precautionary calculus favours the mineral one, and the cost of the switch is essentially zero. The Matta authors wrote in both 2019 and 2020 that 'these findings do not indicate that individuals should refrain from the use of sunscreen' — and that hedge is the right one. The choice is between two things that work, not between sunscreen and no sunscreen.

Mineral filters are the eso-friendly answer here, and there is unusually clean regulatory consensus to support that: zinc oxide and titanium dioxide are the only two sunscreen actives the FDA has classified as Category I, the only filters permitted in Maui County and Palau, and the only ones with primary-source human in vivo evidence of non-penetration. Look for products where the back-label active-ingredients box lists one or both — and only one or both. Be wary of products marketed as 'mineral-based' or 'natural' that bury chemical filters lower in the active-ingredients box; the marketing copy is unregulated, the back-label box is not.

Eso World does not currently sell sunscreen — that's a planned future category, not a current one. The recommendation here is about anyone's mineral product, not a specific brand. The same back-label test the rest of this site applies to oral care, tap water, and endocrine disruptors more broadly applies here exactly: read the box, choose what stays out of your bloodstream, and don't trust the front-of-pack story to do the work the back-of-pack data is doing.

Forty-one years after the first chemical sunscreens cleared FDA approval, the agency that approved them ran the first proper pharmacokinetic trial of how much of those chemicals end up in the bloodstream from a single day's use. The answer was: all six of them, fast, and for one of them, weeks. The proposed rule that would settle the question has been pending since 2019. The EU has moved its caps twice since then on filters the SCCS could not declare safe. The mineral alternatives — already on the shelf, made by the same companies, in the same aisles, at almost the same price — were the answer the FDA's own Category I bar pointed at six years ago. The aisle has not yet caught up to the monograph review. The label can.