The sunscreen aisle has developed a binary: mineral sunscreens (zinc oxide, titanium dioxide) marketed as “clean,” “natural,” and safe for sensitive skin and babies; chemical sunscreens (avobenzone, oxybenzone, octinoxate, octocrylene, homosalate, octisalate) positioned as cosmetically elegant but increasingly shadowed by safety concerns. The clean-beauty industry has amplified the mineral-as-safe narrative, and a set of FDA- sponsored studies has given it scientific language to borrow from.

The underlying questions are worth separating out. How does each filter type actually work? What does the systemic absorption data show — and what does it not show? Is one type meaningfully better at preventing skin damage? And are the reef-safety bans a reliable guide to human safety?

How Each Filter Type Works

The most persistent misconception about mineral sunscreens is that they work purely by reflecting UV radiation off the skin surface like a physical shield. This is a simplification that has been repeated in marketing for decades and is not supported by the physics.

Both mineral and chemical filters primarily work by absorbing UV radiation and dissipating it as heat. Zinc oxide and titanium dioxide do scatter some UV light, particularly at higher concentrations, but absorption is the dominant mechanism for both filter types. The practical distinction is:

• Mineral filters (zinc oxide, titanium dioxide) sit on top of the skin, are not significantly absorbed into the bloodstream, provide broad-spectrum coverage (zinc oxide covers both UVA and UVB well; titanium dioxide is stronger in UVB), and are photostable — they do not degrade significantly in sunlight. The trade-off is cosmetic: they leave a white cast, particularly on darker skin tones, and can feel heavy.
• Chemical filters (organic UV absorbers) are absorbed into the upper layers of the skin where they absorb UV and release the energy as heat. They are cosmetically elegant — invisible, lightweight, easy to formulate — but some are photounstable (avobenzone degrades in sunlight unless stabilized) and some absorb systemically. Coverage varies by ingredient: oxybenzone covers both UVA and UVB; avobenzone covers UVA well; most others primarily address UVB.

The Systemic Absorption Studies

The legitimate scientific basis for the mineral-as-safer framing comes from two randomized clinical trials conducted by FDA researchers and published in JAMA.

The 2019 Matta et al. pilot study tested four active ingredients — avobenzone, oxybenzone, octocrylene, and ecamsule — applied under maximal-use conditions (four applications per day to 75% of body surface area). All four exceeded the FDA's threshold of 0.5 ng/mL plasma concentration above which nonclinical safety data are required before a GRASE determination can be made. Oxybenzone in particular reached plasma concentrations well above the threshold within the first day.

The 2020 follow-up study extended the testing to six ingredients (adding homosalate, octisalate, and octinoxate) across multiple formulations, again under maximal-use conditions. All six showed systemic absorption above the threshold. Importantly, the studies measured blood concentration, not harm. Both papers included an explicit note that the findings should not discourage sunscreen use, as the risks of UV-induced skin damage and skin cancer are well established.

The FDA's position, set out in its proposed administrative order on OTC sunscreens, is that only zinc oxide and titanium dioxide currently have sufficient data to be classified as Generally Recognized as Safe and Effective (GRASE). This does not mean the chemical filters are unsafe — it means the FDA has asked for more safety data, which has not yet been provided. The GRASE framework applies a precautionary standard; it is not a finding of harm.

What Absorption Does and Doesn’t Mean

The systemic absorption data is real. The leap from “absorbs into the bloodstream” to “disrupts your hormones” is not.

The endocrine-disruption concern centers primarily on oxybenzone, which has shown weak estrogenic activity in some in vitro and rodent studies. The American Academy of Dermatology reviewed this evidence and concluded that there is no clinical or epidemiological evidence of endocrine disruption in humans at doses encountered through sunscreen use, even accounting for systemic absorption. The plasma concentrations reached in the Matta studies, while above the FDA's precautionary threshold, are orders of magnitude below levels associated with hormonal effects in animal models.

The honest summary: absorption is documented, harm is not. The FDA wants more data before granting GRASE status. That is a regulatory precaution, not a safety finding.

The Reef-Safe Claim

Hawaii's SB2571, signed into law in 2018, banned the sale of sunscreens containing oxybenzone and octinoxate, citing laboratory evidence that these compounds can bleach coral and disrupt larval development at high concentrations. The US Virgin Islands and Key West enacted similar bans.

The environmental evidence has been disputed. The concentrations required to produce coral-bleaching effects in laboratory studies are orders of magnitude higher than those measured in reef environments, and sunscreen runoff accounts for a small fraction of the stressors on coral reefs compared to ocean warming and acidification. The National Oceanic and Atmospheric Administration (NOAA) identifies climate change as the primary driver of coral bleaching, not sunscreen chemicals.

The reef-safe framing has crossed from environmental marketing into an implied human safety signal — if it is bad for the coral reef, it must be bad for you. The evidence does not support that inference. Ecological toxicity at laboratory concentrations and human clinical harm at use concentrations are different questions.

What Actually Determines Real-World Protection

The evidence on photoprotection is consistent on one point that the filter-type debate tends to obscure: application behavior is the primary driver of real-world SPF, not filter chemistry.

SPF ratings are tested under laboratory conditions using 2 mg/cm² of product — roughly a full teaspoon for the face alone, and a shot glass worth for the body. Most people apply 25–50% of that amount. Underapplication reduces effective SPF non-linearly: applying half the recommended amount of an SPF 50 product does not produce SPF 25 protection; it produces roughly SPF 7.

Chemical sunscreens, because they are cosmetically lighter and easier to blend, tend to be applied more generously and reapplied more consistently. A mineral SPF 50 that produces a visible white cast and feels heavy will be applied sparingly and often not reapplied — which negates any theoretical filter advantage. A chemical SPF 30 applied correctly and reapplied every two hours will provide substantially better real-world protection than a mineral SPF 50 applied once at half dose.

The AAD's guidance is consistent on this: the best sunscreen is the one you will actually use correctly and consistently. That is a formulation preference question, not a filter-type safety question. On this front, your existing piece on SPF label claims covers the gap between rated and real-world protection in more detail.

The US Filter Gap

One genuine issue the mineral-vs-chemical framing obscures is that the US sunscreen market is working with a limited and outdated filter toolkit. European and Asian consumers have access to newer, more photostable chemical filters — including tinosorb S, tinosorb M, and mexoryl SX — that provide excellent broad-spectrum coverage without the stability problems of avobenzone.

These filters have not been approved for use in the US because of a regulatory backlog under the old OTC monograph system, which required an act of Congress to reform. The FDA's time-and-extent applications process has been slow. The approval of bemotrizinol in 2025 — covered in our separate piece on the FDA's first new sunscreen ingredient in 20 years — represents the first step in closing this gap. It does not change the current evidence on the existing approved filters.

The Bottom Line

For most people, the practical guidance is straightforward: choose the formulation you will use consistently and apply generously. If you prefer mineral for cosmetic or precautionary reasons, zinc oxide provides excellent broad-spectrum coverage and is well-tolerated on sensitive skin. If you prefer chemical for the finish and ease of application, current evidence does not support avoiding it.

The one population where the precautionary argument carries more weight is infants under six months, for whom the FDA recommends mineral-only formulations (or shade and clothing) given their higher skin surface-to-body- weight ratio and developing systems. For everyone else, the existing evidence does not support the “mineral is safe, chemical is dangerous” binary.

References & Further Reading

• Matta, M. K., et al. (2019). Effect of Sunscreen Application Under Maximal Use Conditions on Plasma Concentration of Sunscreen Active Ingredients: A Randomized Clinical Trial. JAMA, 321(21), 2082–2091.
• Matta, M. K., et al. (2020). Effect of Sunscreen Application on Plasma Concentration of Sunscreen Active Ingredients: A Randomized Clinical Trial. JAMA, 323(3), 256–267.
• US Food and Drug Administration. (2021, September 24). Proposed Administrative Order: Sunscreen Drug Products for Over-the-Counter Human Use (OTC000008).
• US Food and Drug Administration. Should You Put Sunscreen on Infants? Not Usually.
• US Food and Drug Administration. Sunscreen Drug Products: Time and Extent Applications and Other Regulatory Approaches.
• American Academy of Dermatology. Sunscreen FAQs.
• National Oceanic and Atmospheric Administration. Coral Reef Threats.
• Hawaii Legislature. (2018). SB2571 CD1 — Relating to Sunscreen.
• Environmental Working Group. EWG's Guide to Sunscreens.