# Sunscreen UV Filters — Master Synthesis ## Spectra & Photodegradation of All Globally-Approved Active Ingredients **Compiled:** 2026-05-07 **Methodology:** 6 parallel deep-research subagents per `/workspace/overview/research-agent-instructions.md`. ~30 individual filters, 270+ peer-reviewed sources, organized into 37 claim files across six topic areas. > All numerical claims (λmax, ε, photodegradation rate) trace to a primary peer-reviewed source listed in the per-area `sources.md`. Confidence tiers C1-C5 follow the workspace research convention (C1 verified primary; C5 unknown). Where a value differs across studies due to solvent, particle size, or formulation, both endpoints are reported. --- ## 1. The 30 Filters at a Glance ``` 280 320 360 400nm | UVB | UVA-II | UVA-I | Visible | | | | | | | ─── ORGANIC, UVB-DOMINANT ──────────────────────────────────────────────── PABA ████ peak 283 Octinoxate (OMC) █████ peak 308–311 ⚠ photolabile Octisalate (EHS) ███ peak 305–307 Homosalate (HMS) ████ peak 306–309 Cinoxate █████ peak 308 Padimate O █████ peak 311 Octocrylene ████ peak 303 (→ benzophenone Downs 2021) 4-MBC (Enzacamene) ████ peak 300 ⚠ EU ban May 2026 Polysilicone-15 █████ peak 312 Octyl Triazone (EHT) █████ peak 314 ⭐ ε≈135,000 highest of any Ensulizole (PBSA) ████ peak 302 (water-sol.) Trolamine Salicylate ███ peak 298 ─── ORGANIC, UVA OR DUAL ───────────────────────────────────────────────── Avobenzone (BMDBM) ███████ peak 357 ⚠⚠ photolabile Meradimate █████ peak 336 ⚠ ¹O₂ sensitiser DHHB (Uvinul A Plus) ███████ peak 354 ⭐ photostable Bisoctrizole / MBBT (Tinosorb M) █████ █████████ dual 305 + 360 ⭐ broadest Iscotrizinol (Uvasorb HEB) █████ peak 310 ─── ORGANIC, BROAD-SPECTRUM / NEXT-GEN ─────────────────────────────────── Bemotrizinol (Tinosorb S) █████ ███████ dual 310 + 340 ⭐ Ecamsule (Mexoryl SX) █████ peak 345 Drometrizole Trisiloxane (Mex XL) █████ ██████ dual 303 + 344 Mexoryl 400 (MCE) █████ peak 385 ⭐ Tinosorb A2B (TBPT) █████ peak 310 + scatter Bisdisulizole (Neo Heliopan AP) █████ peak 335 (water-sol.) ─── BENZOPHENONES ──────────────────────────────────────────────────────── Oxybenzone (BP-3) █████ ██████ dual 287 + 325 ⚠ reef ban Sulisobenzone (BP-4) █████ ██████ dual 287 + 325 (water-sol.) Dioxybenzone (BP-8) █████ ██████ dual 282 + 325–352 Mexenone (BP-10) █████ █████ dual 287 + 325 (obsolete) Benzophenone-1 (BP-1) █████ █████ dual 287 + 325 (metabolite) ─── INORGANIC SEMICONDUCTORS ───────────────────────────────────────────── Zinc Oxide (ZnO) ████████████████ Eg=3.37 eV; cutoff 368 nm TiO₂ rutile ███████████████████ Eg=3.00 eV; cutoff 413 nm TiO₂ anatase ████████████████ Eg=3.20 eV; cutoff 387 nm (avoided in cosmetics) 280 320 360 400nm ``` --- ## 2. Master Filter Reference Table | INCI / Common name | CAS | MW | λmax (nm) | ε (M⁻¹cm⁻¹) | E(1%, 1cm) | Photostab | FDA % | EU % | Detail file | |---|---|---|---|---|---|---|---|---|---| | **UVB-dominant** | | | | | | | | | | | Octinoxate (EHMC, OMC) | 5466-77-3 | 290.4 | 308–311 | 23,300–25,000 | 826 | poor | 7.5 | 10 | `uvb-organic/c001` | | Octisalate (EHS) | 118-60-5 | 250.3 | 305–307 | 4,800 | 192 | excellent | 5 | 5 | `uvb-organic/c002` | | Homosalate (HMS) | 118-56-9 | 262.3 | 306–309 | 5,000 | 191 | excellent | 15 | 7.34 face only | `uvb-organic/c003` | | Octocrylene (OCR) | 6197-30-4 | 361.5 | 303 | 12,000 | 332 | good (BP issue) | 10 | 10 | `uvb-organic/c004` | | Cinoxate | 104-28-9 | 250.3 | 308 | 20,650 | 825 | poor | 3 | — | `uvb-organic/c005` | | PABA | 150-13-0 | 137.1 | 283 | 13,500 | 985 | mod. | 15 | banned | `uvb-organic/c006` | | Padimate O | 21245-02-3 | 277.4 | 311 | 28,400 | 1,024 | good | 8 | 8 | `uvb-organic/c006` | | Ensulizole (PBSA) | 27503-81-7 | 274.3 | 302 (H₂O) | 25,000 | 911 | excellent | 4 | 8 | `uvb-organic/c007` | | Polysilicone-15 (Parsol SLX) | 207574-74-1 | ~6,000 | 312 | per-chr ~6,500 | 175 | excellent | — | 10 | `uvb-organic/c008` | | Octyl Triazone (EHT, Uvinul T 150) | 88122-99-0 | 823.1 | 314 | **135,000** | 1,500 | best | — | 5 | `uvb-organic/c009` | | 4-Methylbenzylidene Camphor (4-MBC) | 36861-47-9 | 254.4 | 300 | 24,500 | 963 | good | — | banned May 2026 | `uvb-organic/c010` | | Trolamine Salicylate | 2174-16-5 | 287.3 | ~298 | 3,000–3,600 | — | very good | 12 (proposed not GRASE) | — | `benzophenones/c004` | | **UVA / dual** | | | | | | | | | | | Avobenzone (BMDBM) | 70356-09-1 | 310.4 | **357** (enol) | 35,000 (MeOH) | 1,130 | **POOR** | 3 | 5 | `uva-organic/c001` | | Meradimate | 134-09-8 | 275.4 | 336 | 5,000 | 180 | mod (¹O₂ sensitiser) | 5 | — | `uva-organic/c002` | | DHHB (Uvinul A Plus) | 302776-68-7 | 397.5 | 354 | 39,000 (EtOH) | 980 | excellent | TEA pending | 10 | `uva-organic/c003` | | Bisoctrizole / MBBT (Tinosorb M) | 103597-45-1 | 658.9 | 305 + 360 | 52,000 / 46,000 | 700 | excellent | TEA pending | 10 | `uva-organic/c004` | | Iscotrizinol (Uvasorb HEB) | 154702-15-5 | 766.0 | 310 | 110,000 | 1,435 | excellent | TEA pending | 10 | `uva-organic/c005` | | **Broad-spectrum / Mexoryl** | | | | | | | | | | | Bemotrizinol (Tinosorb S, BEMT) | 187393-00-6 | 627.8 | 310 + 340 | 50,000 @340 | 790 | 98.4% / 50 MED | **proposed GRASE 6%** Dec-2025 | 10 | `broad-spectrum/c001` | | Ecamsule (Mexoryl SX) | 92761-26-7 | 562.7 | 345 | 20,000 (19-47k range) | 360 | 94–99% / 25 MED | NDA only (Anthelios SX) | 10 | `broad-spectrum/c002` | | Drometrizole Trisiloxane (Mexoryl XL) | 155633-54-8 | 501.9 | 303 + 344 | 17–18,000 @344 | 340 | ≥97% / 50 MED | TEA pending | 15 (highest of any) | `broad-spectrum/c003` | | Mexoryl 400 (MCE) | 1419401-88-9 | 322.4 | **385** | 63,000 | 1,956 | "100%" sponsor | — | 3 (Reg 2020/1684) | `broad-spectrum/c004` | | Tinosorb A2B (TBPT) | 31274-51-8 | 537.7 | 310 + scatter | 80,000 (THF) | — | excellent | — | 10 (nano + non-nano) | `broad-spectrum/c005` | | Bisdisulizole (Neo Heliopan AP) | 180898-37-7 | 674.6 | 335 (H₂O) | 52,000 | 770–800 | >95% | — | 10 | `broad-spectrum/c006` | | **Benzophenones** | | | | | | | | | | | Oxybenzone (BP-3) | 131-57-7 | 228.3 | 287 + 325 | 15,000–21,000 | — | good (ESIPT) | 6 | 2.2% body / 6% face | `benzophenones/c001` | | Sulisobenzone (BP-4) | 4065-45-6 | 308.3 | 286 + 325 | 14,000–16,000 | — | good | 10 | 5 | `benzophenones/c002` | | Dioxybenzone (BP-8) | 131-53-3 | 244.2 | 282 + 325–352 | 15,000 | — | good | 3 | — | `benzophenones/c003` | | Mexenone (BP-10) | 1641-17-4 | 242.3 | 287 + 325 | ~14,000 (est) | — | good (analog) | — | — (obsolete) | `benzophenones/c006` | | Benzophenone-1 (BP-1) | 131-56-6 | 214.2 | 287 + 325 | 16,000 | — | good | — | — (metabolite) | `benzophenones/c005` | | **Inorganic** | | | | | | | | | | | Zinc Oxide (ZnO) | 1314-13-2 | 81.4 | cutoff 368 nm (Eg=3.37 eV) | semiconductor | — | n/a (catalyses others) | 25 (GRASE) | 25 | `inorganic/c001` | | Titanium Dioxide (TiO₂ rutile) | 13463-67-7 | 79.9 | cutoff 387–413 nm | semiconductor | — | n/a (catalyses others) | 25 (GRASE) | 25 | `inorganic/c002` | --- ## 3. Photodegradation Reference (key kinetic values) > **Always state the solvent / vehicle / lamp**: the same molecule can be photolabile in one solvent and photostable in another. Avobenzone is the textbook example. ### 3.1 Photolabile filters (significant loss under standard solar dose) | Filter | Quantitative loss | Conditions | Photoproducts | Source | |---|---|---|---|---| | **Avobenzone** | ~36% loss / 1 h sunlight | neat, alone | 14 species: arylglyoxals, benzils, benzoic acid derivatives, dibenzoylmethane radical | Schwack & Rudolph 1995 *J Photochem Photobiol B* 28:229; FDA citation | | Avobenzone (4%) alone | ~77% loss / 25 MED | film, no stabiliser | as above | Bonda 2008 | | **Avobenzone (cyclohexane)** | "appreciable" | aprotic solvent | photoiso. + degradation | Mturi & Martincigh 2008 *J Photochem Photobiol A* 200:410 | | Avobenzone (methanol) | "essentially photostable" | protic — H-bonding stabilises enol | none significant | Mturi & Martincigh 2008 | | **Octinoxate (OMC)** | ~10% loss / 35 J/cm² UVA | EtOH | trans → cis isomer, dimers | Tarras-Wahlberg et al. 1999 *J Invest Dermatol* 113:547 | | **Octinoxate + Avobenzone** | both filters degraded; ESR-detectable radicals | film | [2+2] / Paterno-Büchi cycloadduct | Sayre, Dowdy, Gerwig, Shields, Lloyd 2005 *Photochem Photobiol Sci* 4:699 | | **Cinoxate** | analogous to OMC (cinnamate class) | — | trans→cis, dimers | inferred (C3) | | PABA | photoallergenic; mutagenic photoproducts | aqueous/EtOH | 4-aminobenzaldehyde + radicals | Knowland et al. 1993 *Mutat Res* 304:39 | | **Octocrylene → Benzophenone** | retro-aldol over shelf life: avg 39 mg/kg fresh, 75 mg/kg accelerated, max 435 mg/kg | commercial sunscreens | benzophenone | Downs et al. 2021 *Chem Res Toxicol* 34:1046 | ### 3.2 Photostable filters (>95% retained at typical solar exposures) | Filter | Quantitative retention | Mechanism | Source | |---|---|---|---| | Octisalate (EHS) | flat absorbance ≥50 MED | ESIPT; weak chromophore | Couteau 2007 | | Homosalate (HMS) | flat absorbance ≥50 MED | ESIPT | Couteau 2007 | | Ethylhexyl Triazone (EHT) | >99% / 50 MED | rigid triazine; low Φ_loss | Berset 2005; BASF data | | Polysilicone-15 (Parsol SLX) | >99% | polymer immobilisation; siloxane backbone | DSM data | | Ensulizole (PBSA) | photostable chromophore | water-soluble, but Φ(¹O₂) = 0.10 = ROS generator | Couteau 2007 | | 4-MBC | mostly photostable | benzylidene-camphor TICT | Couteau 2007 | | **DHHB (Uvinul A Plus)** | >95% / 25 MED | ESIPT (intramolecular H-bond → enol) | Herzog 2002+ BASF | | **Bisoctrizole (Tinosorb M)** | 95–99% | ESIPT in two hydroxyphenyl-benzotriazole arms + particulate | Chatelain & Gabard 2001 | | **Bemotrizinol (Tinosorb S)** | 98.4% / 50 MED | ESIPT-like via methoxyphenyl-triazine | Chatelain & Gabard 2001 *Photochem Photobiol* 74:401 | | Iscotrizinol (Uvasorb HEB) | ~10% loss / 25 h sun | rigid triazine | Berset 2005 | | Mexoryl SX (Ecamsule) | 94–99% / 25 MED | reversible camphor isomerisation | Seité et al. 2000 | | Mexoryl XL | ≥97% / 50 MED | ESIPT in 2-(2H-benzotriazol-2-yl)-phenol | L'Oréal data | | Mexoryl 400 (MCE) | "100% intrinsic" | sponsor data only — primary photophysics paper not yet located | L'Oréal data; SCCS/1605/19 | | TBPT (Tinosorb A2B) | excellent (no characterised photoproducts) | sub-ps internal conversion | Naumov 2023 | | Bisdisulizole | >95% / standard solar | rigid bis-benzimidazole | Symrise data | | Oxybenzone (BP-3) | ~80% / 1 h sun; Φ_loss 10⁻⁴–10⁻³ | ESIPT (k≈10¹² s⁻¹); **but generates BP-1** | Cuderman & Heath 2007 | ### 3.3 Inorganic — the filter is photostable but **catalyses degradation of others** | Filter | ROS generation | Mechanism | Mitigation | |---|---|---|---| | Anatase TiO₂ | 5–10× more •OH than rutile; high Φ(O₂•⁻), Φ(•OH) | bandgap exciton → e⁻/h⁺ → ROS | **Avoid anatase in cosmetics** | | Rutile TiO₂ | low | indirect bandgap; longer recombination | use coated grade only | | ZnO | intermediate | ionic dissolution adds Zn²⁺ + ROS | coat (silica/dimethicone) — 60–80% suppression | | Mn-doped rutile (Optisol) | reduced >90% | dopant traps e⁻/h⁺ | Wakefield et al. 2004 | | **Effect on AVB** | uncoated nano-TiO₂ bleaches avobenzone | direct ROS attack | always use coated TiO₂ + AVB | --- ## 4. Photostabilization Combinations (real products are mixtures) ### Hierarchy of stabilisers for avobenzone | Stabiliser | Mechanism | AVB residual: with vs without (representative) | Caveats | |---|---|---|---| | **Octinoxate (OMC)** | DESTABILISER ([2+2] cycloaddition with BMDBM diketo triplet) | both filters lost faster | **avoid this combination entirely** | | **Octocrylene (OC)** | Triplet–triplet energy transfer (kq ~10⁹ M⁻¹s⁻¹) | 23% → 90% / 25 MED | OC degrades to benzophenone (Downs 2021) | | **Bemotrizinol (Tinosorb S)** | TTET, broadband photostable | 30% → 90% / 30 MED | universal stabiliser; not US-approved (until pending order) | | **Bisoctrizole (Tinosorb M)** | optical filtering + TTET | 70–85% retention | particulate; not US-approved | | **DHHB (Uvinul A Plus)** | TTET; co-stabilises | substantial improvement | not US-approved | | **DEHN (Corapan TQ)** | TTET via naphthalate ³ππ*; Φ_Δ=0.44 (generates ¹O₂) | <30% → >80% / 25 MED | pair with salicylates to quench ¹O₂ (Yagi 2019) | | **Polyester-8 (Polycrylene)** | grafted cyanodiphenyl-acrylate (OC chromophore on polymer) | substantial; less than free OC at equal mass | no skin penetration | | **DESM (Oxynex ST)** | TTET + ¹O₂ quench + phenolic radical scavenge | >85% / 25 MED | manufacturer data | | **Solastay S1 (Methoxycrylene)** | singlet-state quencher (different mechanism!) | 50.6% AVB / 120 min sun (best-in-class commercial-SPF50) | proprietary; less common | | α-Tocopherol / Vitamin E | radical scavenger | AVB+VitE 1:2: ~85% vs ~60% alone | small contribution; common antioxidant | ### Triplet energy ladder (the photophysics of why these work) ``` ~70 kcal/mol | BMDBM enol S1 — fast ESIPT (productive UV-A absorption) ~60–66 | BMDBM diketo T1 (n,π*) — the photoreactive species | ↓ all triplet quenchers attack here ~58 | DEHN T1 (³ππ* naphthalate) ~55 | OC ≈ 4-MBC ≈ MBBT T1 ~50–55 | Bemotrizinol T1, DESM T1 ~50 | OMC (cinnamate) T1 ← but instead of quenching, undergoes [2+2] ``` Quenching requires the stabiliser's T1 to lie ≥2-3 kcal/mol below BMDBM diketo T1. --- ## 5. Solvent / Vehicle Effects (the most under-cited variable) Avobenzone in particular has dramatic solvent-dependent λmax and photostability: | Solvent | λmax enol (nm) | Photodegradation outcome | |---|---|---| | Cyclohexane | 355 | appreciable degradation | | Ethyl acetate | 356 | photoiso. + photodegradation | | **Methanol** | 358 | **essentially photostable** (protic stabilises enol) | | DMSO | 363 | photoisomerisation only (O₂-dependent), no degradation | (Mturi & Martincigh 2008, *J Photochem Photobiol A* 200:410.) **Vehicle ranking for unstabilised avobenzone (worst → best):** mineral oil ≈ light paraffins ≈ anhydrous waxes < cyclohexane ≈ DMSO < ethyl acetate < C12-C15 alkyl benzoate ≈ dicaprylyl carbonate < ethanol < methanol/water (protic) ≈ encapsulated (β-cyclodextrin, lipid microparticles). --- ## 6. Regulatory Matrix (May 2026) | Filter | FDA US | EU Annex VI | Australia TGA | Japan | Korea | Reef bans | |---|---|---|---|---|---|---| | Octinoxate (OMC) | 7.5% | 10% | 10% | 20% | 7.5% | **HI / Palau / Thailand / USVI** | | Octisalate | 5% | 5% | 10% | 10% | 5% | — | | Homosalate | 15% | 7.34% face only | 10% | 10% | 10% | — | | Octocrylene | 10% | 10% | 10% | 10% | 10% | **Palau / USVI** | | PABA | 15% (rare) | **banned** (2009) | banned | banned | banned | — | | Padimate O | 8% | 8% | 8% | 10% | 8% | — | | Octyl Triazone (EHT) | — | 5% | 5% | 5% | 5% | — | | Polysilicone-15 | — | 10% | 10% | 10% | 10% | — | | 4-MBC | — | **banned May 2026** | 4% (review) | — | — | — | | Ensulizole (PBSA) | 4% | 8% | 4% | 3% | 4% | — | | Cinoxate | 3% | — | 6% | 5% | — | — | | Avobenzone | 3% | 5% | 5% | 10% | 5% | — | | Meradimate | 5% | — | 5% | — | 5% | — | | DHHB | TEA pending | 10% | 10% | 10% | 10% | — | | Bisoctrizole (Tinosorb M) | TEA pending | 10% | 10% | 10% | 10% | — | | Iscotrizinol | TEA pending | 10% | — | 10% | 10% | — | | Bemotrizinol (Tinosorb S) | **proposed GRASE 6% Dec 2025** | 10% | 10% | 10% | 10% | — | | Ecamsule (Mexoryl SX) | NDA (Anthelios SX 2006 only) | 10% | 10% | 10% | 10% | — | | Drometrizole Trisiloxane (Mexoryl XL) | TEA pending | 15% (highest cap) | 15% | 15% | 15% | — | | Mexoryl 400 (MCE) | — | 3% (Reg 2020/1684) | — | — | — | — | | Tinosorb A2B (TBPT) | — | 10% (nano + non-nano) | 10% | — | 10% | — | | Bisdisulizole | — | 10% | 10% | — | — | — | | Oxybenzone (BP-3) | 6% (insufficient data for GRASE) | **2.2% body / 6% face** | 10% | 5% | 5% | **HI / FL Keys / Palau / USVI / Aruba / Bonaire** | | Sulisobenzone (BP-4) | 10% (insufficient data) | 5% | 10% | 10% | 5% | — | | Dioxybenzone (BP-8) | 3% (insufficient data) | — | 3% | — | — | — | | Trolamine Salicylate | 12% (proposed **NOT GRASE**) | — | 12% | — | — | — | | Zinc Oxide | 25% (GRASE) | 25% | 25% | quasi-drug | listed | — | | Titanium Dioxide | 25% (GRASE) | 25% | 25% | quasi-drug | listed | — | **Notable trajectories:** - The Dec 2025 proposed FDA order on bemotrizinol (6% GRASE) is the first new TEA approval in decades — may break the logjam for DHHB, MBBT, iscotrizinol, drometrizole trisiloxane, ethylhexyl triazone. - Trolamine salicylate is proposed by FDA as **not GRASE** in sunscreens (2021 Proposed Order). - 4-MBC is banned in EU effective May 2026 for endocrine disruption. - Mexoryl 400 (approved EU via Reg 2020/1684, broadly commercialized 2022–2024) is the newest filter — λmax 385 nm closes the previously-uncovered ultra-long UVA1 band. --- ## 7. Cross-cutting Findings ### 7.1 The structure–photostability–regulatory triangle The "next-generation" filters (Tinosorb M/S, DHHB, Mexoryl SX/XL/400, EHT, iscotrizinol, TBPT) share three properties: 1. **High molecular weight** (>500 Da typically) → no skin penetration → no measurable plasma absorption 2. **Photostable chromophores** (ESIPT, rigid triazines, particulates) → useful in real products without aggressive stabilisation 3. **EU/Australia/Japan approved but US TEA-pending** (until Dec 2025 bemotrizinol order) The legacy FDA filter set (octinoxate, oxybenzone, octocrylene, avobenzone) is the inverse: small MW (<400 Da), measurable plasma absorption (Matta JAMA 2019/2020 — all exceeded the FDA 0.5 ng/mL threshold), and either photolabile or with secondary safety concerns (BP-3 endocrine + reef; OC → benzophenone; OMC photolability + endocrine). ### 7.2 The avobenzone-octocrylene-benzophenone trilemma US sunscreens needing UVA-I protection have, in practice, only one approved filter: **avobenzone**. Avobenzone is highly photolabile and effectively requires a stabiliser. The most widely-used stabiliser is **octocrylene**. Octocrylene undergoes retro-aldol decomposition to benzophenone over shelf life (Downs 2021): fresh products contain 6–186 mg/kg benzophenone, post-aging up to 435 mg/kg. **Benzophenone is an IARC Group 2B possible carcinogen.** This trilemma is a structural consequence of the FDA monograph freeze and disappears once newer filters (DHHB, Tinosorb M/S) are approved. ### 7.3 ESIPT is the dominant photostability mechanism A surprising structural unifier: most photostable filters share the same photophysical mechanism — **excited-state intramolecular proton transfer (ESIPT)** in <1 picosecond: - Benzophenones (BP-3/4/8) — 2-OH ↔ C=O hydrogen bond - Tinosorb S / M — methoxyphenyl-triazine and hydroxyphenyl-benzotriazole - DHHB — diethylamino-hydroxybenzoyl - Mexoryl XL — 2-benzotriazolylphenol Avobenzone is the conspicuous exception: it relies on enol↔keto tautomerism *across* a six-membered intramolecular H-bond, but the **diketo form is photoreactive** rather than dissipating to ground state. This is why avobenzone alone fails and ESIPT-class filters succeed. ### 7.4 Inorganic vs organic is a false dichotomy Marketing splits "physical/mineral" from "chemical" filters. Physically: - All filters work primarily by **absorption**. Cole, Shyr & Ou-Yang 2016 showed only 4–5% UV reflection from ZnO/TiO₂ — absorption dominates 10:1. - Inorganics absorb via semiconductor electronic-band-gap transitions; organics via π→π* / n→π* HOMO-LUMO transitions. Both are absorption. - The actually meaningful distinctions are: skin penetration (inorganic ≪ organic small-MW), photocatalysis (inorganic catalyses adjacent organic decomposition; organic does not), and aesthetic feel (whitening — driven by inorganic refractive index, n=2.7 for rutile vs ~1.5 for organics in oil). ### 7.5 Spectral coverage gaps in practice No single filter covers 280–400 nm uniformly. Modern SPF50+ formulae stack 4–7 filters spread across: - **UVB workhorse**: octinoxate / octocrylene / EHT / homosalate - **UVA-II bridge**: BP-3 / ecamsule / bemotrizinol - **UVA-I anchor**: avobenzone / DHHB / Mexoryl 400 / MBBT - **Particulate / inorganic**: ZnO and/or TiO₂ + TBPT The deep UVA-I band (370–400 nm) — implicated in photoaging and pigmentary disorders — was poorly covered until Mexoryl 400 (λmax 385 nm) was approved in EU via Reg 2020/1684 (commercial rollout 2022–2024). ZnO and TBPT (with its scattering tail) help but partially. This band remains the weakest spot in current US-approved formulae specifically because Mexoryl 400 has no FDA path. --- ## 8. Methodology / Provenance ### Research process - Six parallel deep-research subagents (run 2026-05-07) using WebSearch + WebFetch on PubMed, Google Scholar, ScienceDirect, ACS, Wiley, RSC, SCCS opinion documents, FDA Federal Register, JAMA, manufacturer technical bulletins - Each agent followed the workspace `research-agent-instructions.md` claim format with C1–C5 confidence tiers - Each numerical claim cites a primary peer-reviewed source with DOI - Data conflicts flagged (e.g. ε for ecamsule cited as 19,000–47,000 across studies) rather than averaged ### File index ``` sunscreen-filters/ ├── index.md # project dashboard ├── MASTER-SYNTHESIS.md # this file └── research/ ├── uvb-organic/ (10 claim files + summary + sources) │ c001 Octinoxate · c002 Octisalate · c003 Homosalate · c004 Octocrylene │ c005 Cinoxate · c006 PABA/Padimate O · c007 Ensulizole · c008 Polysilicone-15 │ c009 EHT · c010 4-MBC ├── uva-organic/ (5 claim files + summary + sources) │ c001 Avobenzone · c002 Meradimate · c003 DHHB · c004 Bisoctrizole · c005 Iscotrizinol ├── broad-spectrum/ (6 claim files + summary + sources) │ c001 Bemotrizinol · c002 Ecamsule · c003 Drometrizole Trisiloxane │ c004 Mexoryl 400 · c005 Tinosorb A2B · c006 Bisdisulizole ├── benzophenones/ (6 claim files + summary + sources) │ c001 Oxybenzone · c002 Sulisobenzone · c003 Dioxybenzone │ c004 Trolamine salicylate · c005 BP-1 · c006 Mexenone ├── inorganic/ (2 claim files + summary + sources) │ c001 Zinc Oxide · c002 Titanium Dioxide └── photostab/ (8 claim files + summary + sources) c001 OMC+AVB incompatibility · c002 OC stabilisation · c003 Polycrylene c004 DEHN · c005 Tinosorb S as stabiliser · c006 Tinosorb M c007 DESM/Oxynex ST · c008 Formulation factors ``` ### Aggregated data gaps (C5 — "couldn't find" items across all areas) Compiled from per-area C5 lists: 1. **Mexoryl 400** primary peer-reviewed photostability data (only sponsor data) 2. **TBPT (Tinosorb A2B)** singlet-oxygen quantum yield numerics 3. **Iscotrizinol** Φ_T and Φ_Δ values (no primary measurement located) 4. **Bisoctrizole** primary peer-reviewed ε (most data manufacturer-derived) 5. **Ecamsule** ε at 345 nm — disagreement spans 19,000–47,000 across studies 6. **Cinoxate** modern photodegradation kinetics 7. **Polysilicone-15** Φd and Φ(E/Z) measurements 8. **Mexenone (BP-10)** modern ε, Φ_loss, photoproducts 9. **Avobenzone** individual photoproduct-channel quantum yields 10. **Coral toxicity assays** of homologous benzophenones (BP-4, BP-8) analogous to the BP-3 Downs 2016 study 11. **All filters** — long-term chronic dermal exposure studies for nano-coated forms 12. **Damaged skin penetration** — explicitly flagged unknown by SCCS for inorganics 13. **Triplet energies and rate constants** for several photostabiliser pairs (Polycrylene, DESM, MBBT) — inferred from chromophore analogy rather than directly measured ### Confidence tier breakdown across the corpus - **Identity / regulatory status**: predominantly C1 (primary regulatory documents, EU Annex VI, FDA monograph) - **λmax values**: C1 in solvent-specified literature; mixed when comparing across studies (different solvents → different values) - **ε / E(1%, 1cm)**: C1–C2; substantial disagreement for some filters (ecamsule) - **Photodegradation kinetics**: C1 for legacy filters with extensive study; C2 (manufacturer data) for newer filters - **Photostabilisation quantitative claims**: C1 for mechanism (Sayre 2005, Schwack 1995, Chatelain 2001, Lhiaubet-Vallet 2010, Mturi 2008, Downs 2021); C2 for absolute % residuals (manufacturer-affiliated) - **Inorganic photocatalysis ROS quantitation**: C1 (Hirakawa & Nosaka 2002, Wakefield 2004); C4 for some Egerton extinction values (graph-digitised) --- *End of master synthesis. See per-area `_summary.md` for category-level overviews and `claims/cNNN-*.md` for per-filter detail with full citation lists.*