Sunscreen Filters Compared — UV Coverage, Reef Impact, and the Regulatory Gap
Complete UV filter comparison with CAS numbers, UVA/UVB absorption ranges, photostability ratings, coral reef toxicity data, regulatory status across EU/US/AU/JP, application rate data, and formulation stability matrices.
The FDA has not approved a new sunscreen filter since 1999 — and Americans are paying the price
European and Asian sunscreens use photostable, broad-spectrum UV filters like Tinosorb S, Tinosorb M, and Uvinul A Plus that were developed in the early 2000s. These filters provide superior UVA protection with better cosmetic elegance and photostability than the aging filters available in the US market. The FDA has not approved a single new sunscreen active ingredient since 1999 — a 27-year regulatory drought as of 2026.
The reason is not safety concern. It is process. The FDA treats sunscreen actives as over-the-counter drugs (not cosmetics, as the EU does), requiring a New Drug Application (NDA) process. The Sunscreen Innovation Act (2014) was supposed to expedite this; it did not. The result: American consumers either use older, less effective filters (avobenzone stabilized with octocrylene), order sunscreen from overseas, or rely solely on mineral filters — which are effective but cosmetically inferior for many skin tones.
This is not an abstract regulatory dispute. The US has the highest melanoma rates in the developed world (adjusted for latitude and skin type), and inadequate UVA protection from dated filters is a contributing factor. UVA penetrates deeper into the dermis, drives photoaging, and contributes to melanoma through mechanisms distinct from UVB-driven sunburn.
UV filter comparison — the complete table
| Filter Name | CAS Number | Type | UV Range (nm) | Peak Absorption (nm) | Photostable? | Max Concentration | Reef-Safe? | EU Status | US (FDA) Status | AU (TGA) Status | Japan Status |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Zinc oxide | 1314-13-2 | Mineral | 290-380 (broad) | ~360 | Yes | 25% (US/EU) | Yes (non-nano preferred) | Approved | Approved (Category I) | Approved | Approved |
| Titanium dioxide | 13463-67-7 | Mineral | 290-350 (UVB + UVA2) | ~320 | Yes | 25% (US); no limit (EU) | Likely (nano form debated) | Approved (not nano in sprays) | Approved (Category I) | Approved | Approved |
| Avobenzone (Parsol 1789) | 70356-09-1 | Chemical | 310-400 (UVA1) | 357 | No (loses 50-90% efficacy in 1 hr without stabilizer) | 3% (US); 5% (EU) | Unclear | Approved | Approved (Category I) | Approved | Approved |
| Octinoxate (Ethylhexyl methoxycinnamate) | 5466-77-3 | Chemical | 280-320 (UVB) | 311 | No (degrades + destabilizes avobenzone) | 7.5% (US); 10% (EU) | No (EC50 coral larvae: 50 ug/L) | Approved | Approved (Category I) | Approved | Approved |
| Oxybenzone (Benzophenone-3) | 131-57-7 | Chemical | 270-350 (UVB + UVA2) | 288, 325 | Yes | 6% (US); 6% (EU, under review) | No (EC50 coral: 10-50 ug/L; endocrine activity) | Approved (concentration under review) | Under GRASE review | Approved | Approved |
| Homosalate | 118-56-9 | Chemical | 295-315 (UVB) | 306 | Moderate | 15% (US); 10% (EU, lowered 2024) | Unclear | Approved (restricted 2024) | Under GRASE review | Approved | Approved |
| Octisalate (Ethylhexyl salicylate) | 118-60-5 | Chemical | 280-320 (UVB) | 307 | Good | 5% (US); 5% (EU) | Likely | Approved | Under GRASE review | Approved | Approved |
| Octocrylene | 6197-30-4 | Chemical | 290-360 (UVB + weak UVA2) | 303 | Yes (stabilizes avobenzone) | 10% (US); 10% (EU) | Under study (benzophenone degradation product concern) | Approved (under SCCS review) | Under GRASE review | Approved | Approved |
| Tinosorb S (Bemotrizinol, BEMT) | 187393-00-6 | Chemical | 280-400 (true broad spectrum) | 310, 343 | Yes (excellent) | — | 10% (EU) | Likely | Approved | NOT approved | Approved |
| Tinosorb M (Bisoctrizole, MBBT) | 103597-45-1 | Chemical (particulate) | 280-400 (true broad spectrum) | 305, 360 | Yes (excellent) | — | 10% (EU) | Likely | Approved | NOT approved | Approved |
| Mexoryl SX (Ecamsule) | 92761-26-7 | Chemical | 290-390 (UVA) | 345 | Yes | — | 10% (EU) | Likely | Approved | Single-product approval (L’Oreal Anthelios only) | Approved |
| Mexoryl XL (Drometrizole trisiloxane) | 155633-54-8 | Chemical | 290-400 (broad) | 303, 344 | Yes | — | 10% (EU) | Likely | Approved | NOT approved | Approved |
| Uvinul A Plus (Diethylamino hydroxybenzoyl hexyl benzoate) | 302776-68-7 | Chemical | 320-400 (UVA1) | 354 | Yes | — | 10% (EU) | Likely | Approved | NOT approved | Approved |
| Uvinul T 150 (Ethylhexyl triazone) | 88122-99-0 | Chemical | 280-320 (UVB) | 314 | Yes | — | 5% (EU) | Unknown | Approved | NOT approved | Approved |
| Amiloxate (Isoamyl p-methoxycinnamate) | 71617-10-2 | Chemical | 290-320 (UVB) | 310 | No | — | 10% (EU) | Unknown | Approved | NOT approved | Approved |
Filters in bold are modern photostable filters available in the EU, Australia, and Japan but NOT approved by the FDA. This represents the core regulatory gap.
SPF math — what the numbers actually mean
SPF measures UVB protection only. It is a ratio of the minimal erythemal dose (MED) with sunscreen to MED without sunscreen, tested at 2 mg/cm2 application rate.
| SPF Rating | UVB Blocked (%) | UVB Transmitted (%) | Relative Transmission vs SPF 15 | Cost per % UVB Blocked (diminishing returns) |
|---|---|---|---|---|
| 2 | 50.0 | 50.0 | 7.5x more than SPF 15 | Baseline |
| 15 | 93.3 | 6.7 | 1x (reference) | 1x |
| 30 | 96.7 | 3.3 | 0.5x SPF 15 transmission | Marginal improvement high |
| 50 | 98.0 | 2.0 | 0.3x | Diminishing |
| 70 | 98.6 | 1.4 | 0.21x | Strongly diminishing |
| 100 | 99.0 | 1.0 | 0.15x | Very low additional value |
The application gap destroys these numbers. SPF is tested at 2 mg/cm2. Actual consumer application averages 0.5-1.0 mg/cm2 — roughly 25-50% of the test amount. The relationship between application thickness and SPF is exponential, not linear:
| Actual Application (mg/cm2) | % of Test Amount | Effective SPF from Labeled SPF 30 | Effective SPF from Labeled SPF 50 |
|---|---|---|---|
| 2.0 (test standard) | 100% | 30 | 50 |
| 1.5 | 75% | ~17 | ~25 |
| 1.0 | 50% | ~7-10 | ~10-14 |
| 0.5 | 25% | ~3-5 | ~4-6 |
This means that labeled SPF 50 applied at half-thickness provides approximately SPF 10-14 — equivalent to labeled SPF 15 applied correctly. The practical conclusion: buying SPF 50 to compensate for under-application is a sound strategy. Alternatively, applying the correct amount of SPF 30 (one full teaspoon for the face and neck) is equally effective and more cost-efficient.
UVA protection systems — the global inconsistency
SPF measures UVB only. UVA rating systems vary by jurisdiction and are not directly comparable:
| Region | UVA System | Method | What It Tells You | Limitation |
|---|---|---|---|---|
| EU | UVA seal (circle logo) | In vivo UVAPF must be ≥1/3 of labeled SPF | Minimum UVA:UVB ratio guaranteed | Binary pass/fail; no gradation |
| Japan/Korea | PA+ to PA++++ | In vivo PPD (Persistent Pigment Darkening) | PA+ = PPD 2-3; PA++ = 4-7; PA+++ = 8-15; PA++++ = ≥16 | Graded and informative; PPD 16+ is excellent |
| UK | Boots Star Rating (1-5) | In vitro UVA:UVB absorbance ratio | 3 stars = minimum good; 5 stars = near-equal UVA/UVB protection | UK-specific; not widely used globally |
| USA | ”Broad Spectrum” label | In vitro critical wavelength ≥370 nm | Product passes a wavelength threshold test | Binary (yes/no); a product can pass with weak UVA protection; no quantitative information |
| Australia | ”Broad Spectrum” | AS/NZS 2604:2021 (in vitro + in vivo if claiming >SPF 50) | Similar to EU approach | Minimum standard, not graded |
The US problem is acute: “Broad Spectrum” on an American sunscreen tells the consumer almost nothing about UVA protection strength. A product can pass the critical wavelength test (λc ≥370 nm) with a UVA-PF of 3 and a labeled SPF of 50 — meaning the UVA:UVB protection ratio is 1:17. The same product would fail the EU’s 1/3 ratio requirement (which would demand UVA-PF ≥17). American consumers have no way to know how much UVA protection they are getting from the label alone.
Environmental impact comparison — reef toxicity data
Hawaii (Act 104, 2018), Palau, the US Virgin Islands, Key West, Bonaire, and Aruba have enacted or proposed sunscreen ingredient restrictions to protect coral reefs. The evidence base:
| Filter | CAS Number | Coral Larvae LC50 / EC50 | Coral Bleaching Threshold (lab) | Marine Bioaccumulation | Detected in Reef Water | Regulatory Bans |
|---|---|---|---|---|---|---|
| Oxybenzone (BP-3) | 131-57-7 | EC50: 8-340 ug/L (species dependent) | 10-50 ug/L induces bleaching markers | Moderate (BCF 57-195) | Yes (0.8-19.2 ug/L in Hawaii, US Virgin Islands) | Hawaii, Palau, USVI, Key West, Bonaire, Aruba |
| Octinoxate (OMC) | 5466-77-3 | EC50: 50-200 ug/L | 50-100 ug/L | Low-Moderate | Yes (0.2-7.4 ug/L) | Hawaii, Palau |
| Octocrylene | 6197-30-4 | LC50: >1000 ug/L (low acute toxicity) | Not well characterized | Degrades to benzophenone (detected in marine biota) | Yes (0.1-3.0 ug/L) | Under study; no current bans |
| Avobenzone | 70356-09-1 | LC50: >1000 ug/L | Not characterized | Low | Detected at low levels | No bans |
| Zinc oxide (non-nano) | 1314-13-2 | LC50: >1000 ug/L | No bleaching observed at environmental concentrations | Minimal (settles as insoluble particles) | Background levels | No bans; generally considered reef-safe |
| Zinc oxide (nano, <100nm) | 1314-13-2 | LC50: 10-100 ug/L (nano-specific toxicity) | Some evidence of oxidative stress in coral at high concentrations | Unknown (nano behavior in seawater poorly characterized) | Limited data | No specific nano bans |
| Titanium dioxide | 13463-67-7 | LC50: >1000 ug/L (bulk); variable (nano) | Photocatalytic ROS generation (nano form) | Minimal | Background levels | EU restricts nano TiO2 in spray sunscreens (inhalation risk, not reef) |
| Tinosorb S | 187393-00-6 | LC50: >1000 ug/L | No data showing bleaching | Low (high molecular weight, poor solubility) | Rarely detected | No bans |
| Tinosorb M | 103597-45-1 | LC50: >1000 ug/L | No data showing bleaching | Very low (particulate, insoluble) | Rarely detected | No bans |
Honest caveat: The relative contribution of sunscreen to coral reef decline versus ocean warming, acidification, agricultural runoff, and sedimentation is debated. Climate change is the dominant driver of coral bleaching globally. Sunscreen chemical stress is a localized, additive stressor most relevant at high-traffic reef sites (snorkeling areas, popular beaches) where concentrations can spike to ecologically relevant levels. Banning oxybenzone does not save reefs from thermal bleaching — but it removes an avoidable local stressor.
Formulation stability matrix
UV filters interact in formulation. Some combinations enhance stability; others cause degradation:
| Filter Combination | Stability Effect | Mechanism | Practical Implication |
|---|---|---|---|
| Avobenzone + Octocrylene | Stabilized | Octocrylene absorbs triplet-state avobenzone energy, preventing photolysis | Standard US formulation strategy; required for avobenzone-containing products |
| Avobenzone + Octinoxate | Destabilized | Octinoxate transfers energy to avobenzone, accelerating degradation of both | Should not be combined; some older products still use this |
| Avobenzone + Tinosorb S | Strongly stabilized | Tinosorb S quenches avobenzone triplet state more efficiently than octocrylene | EU/AU formulation advantage; unavailable in US |
| Zinc oxide + Chemical filters | Variable | ZnO can catalyze photodegradation of some organic filters (avobenzone, octinoxate) via surface ROS | Coated ZnO (silica or dimethicone coating) eliminates this; uncoated ZnO is problematic in hybrid formulations |
| Tinosorb S + Tinosorb M | Synergistic | Complementary absorption profiles; Tinosorb M’s particulate nature provides additional scattering | Gold-standard EU broad-spectrum combination |
| Titanium dioxide + Zinc oxide | Additive | Complementary UV ranges (TiO2: UVB-UVA2; ZnO: UVA-UVB broad) | Standard mineral-only formulation; combined coverage is excellent |
| Mexoryl SX + Mexoryl XL | Synergistic | SX (water-soluble, UVA) + XL (oil-soluble, broad) provide layered protection across both phases | L’Oreal Anthelios flagship technology |
Practical recommendations by use case
| Scenario | Recommended Approach | Key Filters | SPF Target | Reapplication |
|---|---|---|---|---|
| Daily commute (minimal sun) | Moisturizer with SPF or lightweight sunscreen | Zinc oxide or modern chemical filters | SPF 30 | Once in morning |
| Office work with window exposure | UVA-focused protection (glass blocks UVB, transmits UVA) | Avobenzone (stabilized), Tinosorb S, or zinc oxide | SPF 15-30 | Once |
| Beach/pool day | Water-resistant, high SPF, broad spectrum | Mineral (ZnO 20%+) or EU-formula with Tinosorb + avobenzone | SPF 50 | Every 2 hours and after water |
| Children (<6 months) | Shade and clothing; no sunscreen (FDA/AAP recommendation) | — | — | — |
| Children (>6 months) | Mineral-only formulation | Zinc oxide, titanium dioxide | SPF 30-50 | Every 2 hours |
| Reef/marine environment | Mineral sunscreen, non-nano preferred | Non-nano zinc oxide; avoid oxybenzone + octinoxate | SPF 30-50 | Every 2 hours |
| Sensitive/eczema-prone skin | Mineral-only, fragrance-free | Zinc oxide (coated) | SPF 30 | Every 2 hours |
| Dark skin tones (concern: white cast) | Tinted mineral or modern chemical filters | Tinted iron oxide + ZnO; or Tinosorb M (microfine, minimal cast) | SPF 30 | Every 2 hours |
Where the evidence is genuinely uncertain
Systemic absorption. An FDA study (Matta et al., 2019, 2020) found that oxybenzone, avobenzone, octocrylene, homosalate, octisalate, and octinoxate all exceed the 0.5 ng/mL blood concentration threshold that triggers additional safety testing — after application at recommended amounts. This does not mean these compounds are harmful at detected blood levels. It means the FDA’s own safety threshold was exceeded and further data is needed. EFSA and the SCCS (EU) have reviewed the same absorption data and maintained approval for most of these filters with concentration restrictions (homosalate lowered to 10% in the EU in 2024). The safety significance of systemic absorption at measured levels remains genuinely unresolved.
Nano vs non-nano mineral filters. Nano-sized zinc oxide and titanium dioxide (<100 nm) provide better cosmetic elegance (less white cast) but have different toxicological profiles than bulk particles. Intact skin appears to be an effective barrier — nano-particles do not penetrate past the stratum corneum in most studies. However, compromised skin (sunburn, eczema, cuts) may allow penetration. The EU requires nano-form labeling and restricts nano-TiO2 in spray products (inhalation concern). The FDA has no nano-specific regulation for sunscreens.
Long-term UVA filter adequacy. We have 60+ years of population-level data on the relationship between SPF (UVB protection) and skin cancer prevention. We have less than 20 years of widespread UVA-specific filter use. Whether modern UVA filters will produce the expected reduction in melanoma rates is plausible but not yet confirmed by long-term epidemiology. The mechanism is sound (UVA drives mutagenic cyclobutane pyrimidine dimers via photosensitization), but the population-level proof will take decades to accumulate.
The most defensible position: use sunscreen with both UVA and UVB protection, at SPF 30+, applied at adequate thickness (2 mg/cm2), reapplied every 2 hours during exposure. The specific filter choice matters less than consistent, adequate application. The best sunscreen is the one you will actually wear.