Heavy Metal Index

A Feasibility Study Investigating Action Limits for Certain Heavy Metal Impurities in Cosmetic Products

21 min read

Source access

Publication
UK Office for Product Safety and Standards (OPSS), BEIS/OPSS Contract Ref FM21190; report prepared by LGC
Year
2023
Authors
Axford I; Hill S; Bird G; Singh S
Source type
government report
License
uk open government licence

OPSS / Axford et al. 2023 — UK market survey of Sb, tAs, Cd, Pb, tHg in 91 cosmetics; international regulatory feasibility review

This report is a UK government feasibility study commissioned by the Office for Product Safety and Standards (OPSS) and executed by LGC under Contract FM21190. It combines (1) a comprehensive international review of regulatory limits and non-regulatory guidance values for technically-unavoidable heavy-metal impurities (Sb, As, Cd, Pb, Hg) in cosmetic products, (2) identification and verification of a fit-for-purpose analytical method based on ISO 21392:2021 (microwave-assisted HNO₃/HCl ± HF digestion + ICP-MS), and (3) a small UK-market survey of 91 cosmetic products covering 10 product types across high-street and online retailers. The report concludes that the German BVL guidance values (Sb <0.5, As <0.5, Cd <0.1, Pb <2.0 general / <0.5 toothpaste / <5.0 theatre-make-up, Hg <0.1 mg/kg) can be applied to the UK market with minimal impact at the 90 % compliance threshold the BVL itself uses, with eye shadow as the principal exception (3 of 9 eye-shadow samples exceeded the BVL 2 mg/kg Pb general-cosmetics limit; max Pb 8.8 mg/kg). Regulatory framing rests on UK EU-retained law Regulation (EC) No 1223/2009 Article 17 (technically-unavoidable trace impurities permitted in good manufacturing practice) read with Article 3 (safety) and Article 8 (GMP).

Key numbers

Regulatory / guidance limits compiled by the literature review (mg/kg unless stated)

  • China — Technical Safety Standard for Cosmetics 2015 (Table 2, p. 7): As ≤2, Cd ≤5, Pb ≤10, Hg ≤1; general (whole-product limits, not impurity-only).
  • Thailand — single regulatory cadmium limit Cd 3 mg/kg in cosmetic products (Section 2.2 and Table 10, p. 21).
  • USA — 21 CFR §700.13 mercury preservative: Hg ≤1 mg/kg general cosmetics, ≤65 mg/kg eye-area preservative; FDA 2016 recommended industry maximum Pb ≤10 mg/kg in cosmetic lip products and externally-applied cosmetics; FDA colour-additive heavy-metal limits (typical: As ≤3 ppm, Pb ≤20 ppm, Hg ≤1 ppm on colorant-as-ingredient basis) (p. 7-8, p. 11).
  • Germany BVL — non-regulatory guidance, technically-avoidable impurity limits in cosmetics (Table 3, p. 9; Table 16f, p. 40):
    • General cosmetics: Sb <0.5, As <0.5, Cd <0.1, Pb <2.0, Hg <0.1 mg/kg
    • Toothpaste: Pb <0.5 mg/kg
    • Theatre/carnival make-up: As <2.5, Pb <5.0 mg/kg
  • ASEAN — guidance limits for heavy-metal impurities in cosmetics (Table 4, p. 10): As <5, Cd <5 (Thailand regulatory 3 mg/kg overrides), Pb <20, Hg <1 mg/kg.
  • Health Canada — guidance on heavy-metal impurities in cosmetics (Table 5, p. 10): action thresholds at Sb >5, As >3, Cd >3 mg/kg.
  • ICCR (International Cooperation on Cosmetics Regulation) — recommended target levels: Pb ≤10 mg/kg in finished cosmetic products excluding oral-cavity products (ICCR 2013); Hg ≤1 mg/kg as total mercury in finished cosmetic products (ICCR 2016) (p. 7-8).
  • KvW (Netherlands) — 1980 guidance values for technically-unavoidable heavy-metal impurities (Table 8, p. 16): Sb <5, As <5, Cd <5, Pb <20, Hg <1 mg/kg.

Method LODs / LOQs

  • Hepp et al. (FDA, 2014) — microwave HNO₃/HF digestion + ICP-MS (Table 6, p. 13): LOD As 0.048, Cd 0.018, Pb 0.0084, Hg 0.0010 mg/kg; LOQ As 0.16, Cd 0.058, Pb 0.028, Hg 0.0032 mg/kg.
  • Perkin Elmer application note — microwave HNO₃/HF digestion + ICP-MS, estimated detection levels in solid cosmetics (Table 7, p. 15): As 0.014, Cd 0.0069, Pb 0.0096, Hg 0.037 mg/kg.
  • ASEAN method (GF-AAS / FIAS-AAS) reported recoveries from spiked cosmetic creams: As 84-86 %, Cd 66-71 %, Pb 85-99 %, Hg 95-108 % (p. 12). Determination limits As <2.5, Cd <1, Pb <10, Hg <0.5 µg/g.
  • ISO 21392:2021 inter-laboratory trial (10 labs, 7 cosmetic samples, Pb): relative repeatability 7.5-15.5 %, relative reproducibility 15-80 % (Section 2.4, p. 15).

Reference data from prior published surveys cited by OPSS (selected, all in mg/kg unless stated)

  • JRC EUR 24886 EN (2010-2011 European market): 113 lip products (81 lipsticks + 32 lip glosses) from 34 brands across 12 EU Member States; 30 of 81 lipsticks (37 %) >1 mg/kg Pb; 1 of 32 lip glosses (3 %) >1 mg/kg Pb; max 3.75 mg/kg Pb in a red lipstick (p. 16-17).
  • BVL 2019 monitoring — lipsticks WITH glitter (Table 9a, p. 20; n analysed varies by metal): Sb n=139, mean 0.096, median 0.050, max 0.410, BVL guidance 0.5, 0 samples exceeded; As n=135, mean 0.124, median 0.063, max 0.507, BVL guidance 0.5, 1 sample exceeded; Cd n=168, mean 0.022, median 0.025, max 0.030, BVL guidance 0.1, 0 exceeded; Pb n=169, mean 0.401, median 0.250, max 6.130, BVL guidance 2.0, 3 exceeded; Hg n=152, mean 0.028, median 0.013, max 0.428, BVL guidance 0.1, 2 exceeded.
  • BVL 2019 monitoring — lipsticks WITHOUT glitter (Table 9b, p. 20): Sb n=103, mean 0.587, median 0.125, max 24.2, BVL 0.5, 8 exceeded; As n=100, mean 0.070, median 0.019, max 0.997, BVL 0.5, 1 exceeded; Cd n=103, mean 0.013, median 0.005, max 0.014, BVL 0.1, 0 exceeded; Pb n=103, mean 0.330, median 0.190, max 7.610, BVL 2.0, 1 exceeded; Hg n=108, mean 0.033, median 0.033, max 0.431, BVL 0.1, 6 exceeded. The mean (0.587)/median (0.125) divergence and max (24.2 mg/kg) for Sb in lipsticks-without-glitter indicate a heavily right-skewed distribution driven by a small number of high outliers.

LGC UK-market survey (n = 91, 9 product groups analysed; makeup powders verified but not analysed) — analytical detail

Method: ISO 21392:2021-derived microwave acid digestion (Multiwave Go, Anton Paar; mouthwash on Discover SP-D, CEM); two digestion programmes (Short 30-min ramp + 30-min hold at 200 °C; Long ramp/hold at 160/180/200 °C totalling 125 min) selected by matrix; acid mix HNO₃ 8 mL + HCl 1 mL + HF 0.5 mL on Long for lipstick/lip gloss/lip liner/mascara/eye powders/eye liners/face powders/liquid foundation, HNO₃ 5 mL + HCl 1 mL + H₂O₂ 1 mL on Short for mouthwash, HNO₃ 8 mL + HCl 1 mL + HF 0.5 mL on Short for toothpaste and sun protection (Table 13, p. 29). Sample mass 0.1 ± 0.01 g, final volume 50 mL water (mouthwash 20 mL), analytical dilution 2-fold (mouthwash 5-fold), total dilution factor 1000-fold for all matrices (Table 14, p. 30). Detection: Agilent 7700x / 7900 ICP-MS, He collision mode for ⁷⁵As, no-gas mode for ¹¹¹Cd, ¹²¹Sb, ²⁰²Hg, ²⁰⁸Pb; internal standards ¹¹⁵In, ¹²⁸Te, ¹⁵⁹Tb, ¹⁷⁵Lu; ⁹⁵Mo monitored to correct ¹¹¹Cd interferences (Table 15, p. 30). All samples in duplicate; for 12 samples where one replicate was <LOQ and the other above, the worst-case (above-LOQ) value is reported.

LGC UK-market survey — per-product-group summary vs BVL guidance (Tables 16a-16e, p. 38-39; mg/kg)

  • Sb (Table 16a, p. 38; BVL general guidance 0.5 mg/kg): lipstick LOQ 0.096, 7/9 <LOQ, 0 >BVL, max 0.292. Lip gloss LOQ 0.050, 9/9 <LOQ, 0 >BVL, max 0.050. Toothpaste LOQ 0.031, 6/9 <LOQ, 0 >BVL, max 0.040. Mouthwash LOQ 0.029, 6/6 <LOQ, 0 >BVL, max 0.029. Lip liner LOQ 0.018, 5/10 <LOQ, 2 >BVL, max 0.665. Sun protection LOQ 0.047, 9/9 <LOQ, 0 >BVL, max 0.047. Eye shadow LOQ 0.028, 1/9 <LOQ, 2 >BVL, max 1.053. Mascara LOQ 0.015, 8/9 <LOQ, 0 >BVL, max 0.074. Eye liner LOQ 0.011, 3/9 <LOQ, 0 >BVL, max 0.194. Foundation LOQ 0.015, 4/9 <LOQ, 1 >BVL, max 0.585.
  • As (Table 16b, p. 38; BVL general 0.5 mg/kg): lipstick LOQ 0.023, 1/9 <LOQ, 0 >BVL, max 0.283. Lip gloss LOQ 0.039, 6/9 <LOQ, 0 >BVL, max 0.090. Toothpaste LOQ 0.101, 8/9 <LOQ, 0 >BVL, max 0.140. Mouthwash LOQ 0.007, 6/6 <LOQ, 0 >BVL, max 0.007. Lip liner LOQ 0.060, 1/10 <LOQ, 1 >BVL, max 0.505. Sun protection LOQ 0.087, 9/9 <LOQ, 0 >BVL, max 0.087. Eye shadow LOQ 0.060, 0/9 <LOQ, 3 >BVL, max 1.620. Mascara LOQ 0.048, 2/9 <LOQ, 0 >BVL, max 0.255. Eye liner LOQ 0.060, 1/9 <LOQ, 0 >BVL, max 0.418. Foundation LOQ 0.045, 0/9 <LOQ, 0 >BVL, max 0.144.
  • Cd (Table 16c, p. 39; BVL general 0.1 mg/kg): no samples exceeded the BVL limit in any product group. Lipstick LOQ 0.003, max 0.015. Lip gloss LOQ 0.001, max 0.012. Toothpaste LOQ 0.005, max 0.046. Mouthwash LOQ 0.001, max 0.001. Lip liner LOQ 0.002, max 0.026. Sun protection LOQ 0.002, max 0.014. Eye shadow LOQ 0.004, max 0.045. Mascara LOQ 0.002, max 0.006. Eye liner LOQ 0.001, max 0.031. Foundation LOQ 0.001, max 0.018.
  • Pb (Table 16d, p. 39; BVL general 2.0 mg/kg, toothpaste 0.5 mg/kg, theatre 5.0 mg/kg): lipstick LOQ 0.012, 0 >BVL (general), max 1.447. Lip gloss LOQ 0.009, 0 >BVL, max 0.267. Toothpaste LOQ 0.026, 3/9 >BVL (toothpaste 0.5 mg/kg threshold), max 0.639. Mouthwash LOQ 0.007, 0 >BVL, max 0.050. Lip liner LOQ 0.009, 3/10 >BVL, max 2.545. Sun protection LOQ 0.168, 0 >BVL, max 0.248. Eye shadow LOQ 0.129, 6/9 >BVL, max 8.769. Mascara LOQ 0.011, 0 >BVL, max 0.859. Eye liner LOQ 0.014, 1/9 >BVL, max 3.780. Foundation LOQ 0.008, 0 >BVL, max 0.764. Across the full 91-sample set: 13/91 samples exceeded Pb 2 mg/kg, 4/91 exceeded Pb 5 mg/kg, 0/91 exceeded Pb 10 mg/kg.
  • Hg (Table 16e, p. 39 — note: the source mis-titles Table 16e as “Pb” but its column structure (LOQ values <0.002-<0.073 mg/kg vs the actual Pb LOQs <0.007-<0.168 mg/kg in Table 16d) and the BVL 0.1 mg/kg threshold for Hg general cosmetics, together with the discussion’s Hg-eye-shadow finding, confirm Table 16e is Hg; see Verification notes): lipstick LOQ 0.010, 0 >BVL, max 0.058. Lip gloss LOQ 0.073, 1/9 >BVL, max 0.101. Toothpaste LOQ 0.014, 0 >BVL, max 0.025. Mouthwash LOQ 0.002, 0 >BVL, max 0.002. Lip liner LOQ 0.003, 0 >BVL, max 0.071. Sun protection LOQ 0.008, 0 >BVL, max 0.008. Eye shadow LOQ 0.005, 3/9 >BVL, max 0.519. Mascara LOQ 0.005, 0 >BVL, max 0.009. Eye liner LOQ 0.006, 0 >BVL, max 0.050. Foundation LOQ 0.015, 1/9 >BVL, max 0.585 — but this entry is identical to the Sb-Foundation row in Table 16a (LOQ 0.015, 1 >BVL, max 0.585) and contradicts the Section 4.5 Hg-by-product chart (p. 37), which shows all 9 Foundation Hg values well below 0.1 mg/kg. The Foundation Hg row in Table 16e appears to be a copy-paste error from Table 16a Sb in the published report; see Verification notes.

LGC UK-market survey — headline category-level finding

Eye shadow drove the exceedances for every metal except cadmium: 1.053 mg/kg Sb (red eye shadow), 1.620 mg/kg As (purple eye shadow), 8.769 mg/kg Pb (mid-price red-purple eye shadow), 0.519 mg/kg Hg (one of two eye-shadow samples with detectable Hg, low-price red-purple). Lip liner was the second-most-contaminated category for Sb, As, Pb. Toothpaste exceeded the BVL toothpaste-specific Pb threshold (0.5 mg/kg) in 3 of 9 samples while remaining well below the general-cosmetics 2 mg/kg threshold. Highly-coloured products (orange-red, brown, red eye shadows and lip liners) were associated with the largest Pb and Sb signals, with the commonly-listed reddish pigment being CI 77491 (iron oxide) (Discussion, p. 41-42). Price and place-of-purchase (high-street vs online) did not appear to correlate with metal levels.

Discussion conclusion (Section 2.6 and Section 5)

OPSS concludes that the BVL guidance values, applied with the BVL’s own 90 % compliance criterion, could be applied to the UK market with minimal impact on the basis of this 91-sample exploratory survey (Section 5, p. 42). The report flags that the small sample size and exploratory scope require further confirmatory work with larger market coverage, and recommends extending the scope to chromium (Cr) and other elements detected in the semi-quantitative screen.

Evidence Fitness

This is a UK government feasibility study with a primary 91-sample analytical survey, full method validation against ISO 21392:2021, and an A-tier comprehensive international regulatory review. The analytical lab (LGC) operates under ISO 17034 and ISO 17043 accreditations for the relevant work. Evidence the report supports:

  • Reconstructable per-product-group occurrence dataset for Sb, tAs, Cd, Pb, tHg in 9 of 10 declared cosmetic groups on the UK market in 2022 (per-sample numerical results in Appendix 2 with the printed report; this page captures the group-level summary tables 16a-16e and the discussion findings).
  • Methods-fitness evidence: ISO 21392:2021 with HF acid adaptation is fit-for-purpose for the UK cosmetic matrix, including refractory SiO₂/TiO₂-containing samples.
  • Regulatory-context evidence: comprehensive 2023 snapshot of cosmetic heavy-metal regulations and non-regulatory guidance across UK/EU, US, China, Thailand, Japan, Canada, Germany, ASEAN, and ICCR.
  • Eye-shadow Pb finding (6/9 > BVL 2 mg/kg; max 8.8 mg/kg; no sample > 10 mg/kg FDA recommended industry max) is a direct UK-market observation pertinent to any UK or downstream HMT&C threshold discussion on adult eye-makeup category.

The report does not support exposure modelling of children’s cosmetic use; sample selection covered 10 adult-cosmetic categories not children’s products. The 91-sample size is too small to support percentile-distribution claims by product group (OPSS itself flags this as exploratory). The single-replicate worst-case reporting for 12 marginal cases means the per-product values for those entries are slight over-estimates rather than true duplicate means.

Methods (brief)

Sample selection by OPSS-defined purposive sampling: low, mid and high price ranges, ~70/30 high-street/online split, range of colours/shades, 10 cosmetic-product groups risk-rated by SiO₂/TiO₂ content and ingestion/eye/inhalation pathway (Table 1, p. 3); ratings 1 (lip and oral products) and 2 (eye and sun-protection products) were prioritised, makeup powders verified but not analysed due to time/budget constraints.

Sample preparation: matrix-specific homogenisation (lipstick heated 85 °C/1 h; lip gloss stirred; lip liner cut open with stainless-steel scalpel + dry-ice pestle-and-mortar pulverisation; mascara/eye liner disassembled with sharp tweezers; face powders crushed with pestle and mortar). Sample mass 0.1 ± 0.01 g.

Digestion: microwave acid digestion per ISO 21392:2021, two programmes selected by matrix (Tables 12-13, p. 29). Multiwave Go (Anton Paar, UK) Teflon vessels for all matrices except mouthwash (Discover SP-D, CEM Corporation, UK, with quartz vessels). Acids trace-element grade (SpA, Romil, UK).

Method validation: 8 reagent blanks + 1 duplicate spiked sample + QC material (lipstick + lip-gloss proficiency-testing materials from LGC Standards under ISO 17043, used at consensus values) or NIST SRM 2709a San Joaquin Soil (selected for its high SiO₂/TiO₂ content). Mouthwash validated by triplicate spiked samples only (no fit-for-purpose QC available). Acceptance criteria: linearity >0.99, internal-standard recovery 75-125 %, IQC standard recovery 80-120 %, triplicate-spike recovery 70-150 %, QC/CRM recovery 75-125 % (excluding mouthwash), duplicate RSD ≤20 %. Verification passed for every matrix.

Analytical determination: Agilent 7700x or 7900 collision/reaction-cell ICP-MS, helium collision mode for ⁷⁵As, no-gas mode for ¹¹¹Cd, ¹²¹Sb, ²⁰²Hg, ²⁰⁸Pb; Pb and Hg reported as sum of isotopes; ⁹⁵Mo correction for ¹¹¹Cd; ¹¹⁵In/¹²⁸Te/¹⁵⁹Tb/¹⁷⁵Lu internal standards (Table 15, p. 30). External calibration with ISO 17034-accredited standards (Romil, UK); IQC from independent supplier batch. Semi-quantitative scan identified Ba, Co, Cr, Cu, Li, Mo, Ni, Sn, Tl, V, Zr in Appendix 3 (p. 53); Zn additionally mentioned in Section 3.5.4 (p. 31) as an interference-source element observed during method verification.

Statistical handling: duplicate measurements averaged for graphical presentation; for the 12 instances where one replicate was <LOQ and the other above, the above-LOQ value reported as worst-case.

Speciation status: total elemental concentrations only. Arsenic measured as ⁷⁵As (total As, recorded as tAs). Mercury measured as ²⁰²Hg by ICP-MS (total Hg, recorded as tHg). No inorganic/methyl speciation. Chromium-hexavalent vs total chromium is NOT distinguished in the semi-quantitative Cr data (Cr, not Cr-VI).

Limitations OPSS acknowledges: small sample size (9-10 per product group), no analysis of makeup powders, exploratory scope, single-laboratory measurements, reliance on proficiency-testing-derived QC consensus values rather than certified reference materials (no matrix-matched cosmetic CRMs exist), reproducibility 15-80 % for Pb across labs in the ISO inter-laboratory trial (i.e., the method works for fit-for-purpose decisions but inter-lab comparisons require care).

Implications

  • Certification: this is the highest-quality publicly-available UK-government regulatory feasibility evidence on heavy-metal occurrence in adult-market cosmetics as of 2023. Use as the regulatory-context anchor for any Cat 13 (Cosmetics — Leave-on) row that maps to the categories surveyed: eye-makeup, makeup-foundation-powders-blush, sun-suntan-products. Use as the regulatory-context anchor for Cat 20 (Oral Care) rows mouthwash-oral-rinse and any future adult-toothpaste row. The eye-shadow Pb finding (6/9 > 2 mg/kg, max 8.8 mg/kg) is the strongest direct-evidence signal in the corpus for Cat 13 Row 7 (eye-makeup).
  • Courses: regulatory-literacy module reference. The 20× range between EU Pb 0.5 ppm (colour-additive level) and Health Canada Pb 10 mg/kg (impurity-action level), and the ICCR 1 ppm Hg / 10 ppm Pb policy-driven target levels, illustrate that “cosmetic heavy-metal limits” are not a single number but a layered set of impurity, colour-additive, ingredient, and finished-product caps that resolve differently by jurisdiction.
  • App: not directly relevant to the food-app contamination-profile workflow; this is a cosmetics paper.

Wiki pages this source may touch

  • eye-makeup — direct evidence: 27 UK-market samples (9 mascara + 9 eye liner + 9 eye shadow); eye shadow drives Pb, Sb, As exceedances; Pb max 8.8 mg/kg.
  • makeup-foundation-powders-blush — direct evidence: 9 UK-market foundation-liquid/cream samples; 1 sample > BVL 0.5 mg/kg Sb; max Pb 0.764 mg/kg.
  • sun-suntan-products — direct evidence: 9 UK-market sun-protection samples; all <BVL guidance for every analyte; max Pb 0.248 mg/kg.
  • mouthwash-oral-rinse — direct evidence: 6 UK-market mouthwash samples; all <BVL guidance for every analyte; max Pb 0.050 mg/kg.
  • toothpaste — direct evidence: 9 UK-market toothpaste samples; 3/9 exceed the BVL toothpaste-specific Pb 0.5 mg/kg threshold while remaining below the BVL general-cosmetics 2 mg/kg threshold; max Pb 0.639 mg/kg.
  • lead — UK-market lead occurrence data across 10 cosmetic product groups; 13/91 samples > 2 mg/kg, 4/91 > 5 mg/kg, 0/91 > 10 mg/kg.
  • cadmium — UK-market cadmium occurrence; no samples exceeded BVL 0.1 mg/kg in any of 91 samples; max 0.046 mg/kg in toothpaste.
  • arsenic-total — UK-market total As occurrence; eye-shadow drove exceedances, max 1.62 mg/kg.
  • mercury-total — UK-market total Hg occurrence; eye-shadow drove exceedances, max 0.519 mg/kg.
  • antimony — UK-market antimony occurrence; eye-shadow and lip-liner drove exceedances, max 1.053 mg/kg.

Verification notes

  • Cite-key choice. Report is an OPSS commissioning of LGC under BEIS/OPSS Contract Ref FM21190; published March 2023 on the UK gov.uk site under OPSS, not under LGC’s own publications. Lead author Ian Axford is the LGC analyst, not the issuing authority. Cite-key uses the issuing-authority convention (opss2023-) consistent with other agency-issued documents in the corpus; the LGC author team is preserved in the authors field for citation accuracy.
  • Table 16e mis-titled in the source. The source PDF (p. 39) titles Table 16e as “Summary of Heavy Metal Content for Pb in Comparison to BVL (German) Guidance Values”; this duplicates the Table 16d title. The column structure and values (LOQs <0.002 to <0.073 mg/kg consistent with Hg-method LOQs reported in Section 3, the 0.001-0.519 mg/kg max-content range, the 0.1 mg/kg BVL threshold, and the 3/9 eye-shadow exceedances) match the Hg data shown in the Section 4.5 chart (p. 37) and the Section 5 discussion’s Hg-eye-shadow finding. Table 16e is the Hg summary. This page reports Table 16e as the Hg data accordingly; the title in the source is a typographical error in the published report. This is the kind of source-side error worth surfacing if OPSS is ever re-contacted for a revised release.
  • Speciation. The paper measures total elemental concentrations only (ICP-MS detection of ⁷⁵As, ²⁰²Hg). metals: [Sb, tAs, Cd, Pb, tHg] records the speciation correctly. No inorganic-As / methyl-Hg / Cr-VI distinction is available.
  • Routing — lip products (lipstick, lip gloss, lip liner) not declared. The OPSS sample set includes 28 adult lip products (9 lipstick + 9 lip gloss + 10 lip liner), with the lip-liner subcategory accounting for several BVL exceedances (2/10 >Sb, 1/10 >As, 3/10 >Pb, including the 2.545 mg/kg Pb max). However, the wiki’s products/lipstick.md is RETIRED (Cat 2 redirect target only; HMTc Cat 2 covers ages 0-5 and adult lipstick is reserved for a future Cat 13 row that has not yet been Step-0-locked), and no adult lip-makeup / lip-products slug exists in the current Cat 13 scaffold (Cat 13 rows scaffolded so far: Row 3 face-neck-leave-on, Row 5 makeup-foundation-powders-blush, Row 6 makeup-body-paints-bases-fixatives, Row 7 eye-makeup, Row 10 sun-suntan-products; no row for lip products is published). To prevent silent mis-routing into a retired children’s-cosmetic page, the 28 lip-product results are not declared in products: for this ingest. When Karen builds the Cat 13 lip-products Step 0 Lock and scaffolds the corresponding page, this source page should have that slug added to its products: array and the routing audit re-run; this paper is among the highest-quality direct-evidence sources available for that future row.
  • Routing — toothpaste page is currently Cat 2. products/toothpaste.md carries hmtc_category: 2, hmtc_row: 11 (Children’s toothpaste row in the Cat 2 Step 0 Lock). The OPSS toothpaste sample set is adult UK toothpastes purchased from high-street/online retailers, not children’s-marketed toothpaste. Routing to products/toothpaste is the only existing routing destination; this is the same single-page-currently-covers-both-children-and-adult pattern that the wiki currently uses for toothpaste. When Karen builds a Cat 20 adult-toothpaste row separate from Cat 2, the products array on this page may need updating. Flagged for Karen’s awareness.
  • Sample folder placement. This paper was filed under raw/Manual Fetch Kimi /Children Personal Care Papers/babycare_03_Powder_Lotion_Oil_Diaper/. The paper is NOT a children’s-cosmetic paper — it is a UK adult-cosmetics market survey commissioned under the UK Cosmetic Products Regulation (EC) No 1223/2009 regulatory feasibility programme, with samples drawn from adult high-street retailers (Boots/Superdrug shelves) without explicit child-cosmetic stratification. The folder placement is incidental (it appears Karen filed it under the broadest cosmetics-papers folder available). Routing here treats it as adult Cat 13 / Cat 20 evidence per its actual content.
  • Brand firewall. The published report includes a numbered sample list (Appendix 1, p. 44-47) that names each sample’s price, retail channel (high-street / online), colour, and an anonymised sample code (22Bxxxx / 22Cxxxx) but does NOT publish brand names attached to those codes. The report’s category-level summaries (Tables 16a-16e and Discussion) report findings by product group with no brand attribution. This page reports the data as the source does — category-level only, sample codes only. No brand attribution. Method-vendor names that ARE in this paper (Agilent 7700x / 7900 ICP-MS, Anton Paar Multiwave Go, CEM Discover SP-D, PerkinElmer PinAAcle reference, LGC Standards UK proficiency-testing materials under ISO 17043, Romil ISO 17034 calibration standards, NIST SRM 2709a San Joaquin Soil reference material) are retained per the 2026-05-17 Part 12 Exception 2.
  • Wiki/HMTc firewall (Part 2). This source describes the BVL 90 % compliance criterion as a policy choice by Germany’s BVL, not as a Cochrane-style consensus. The OPSS finding that “BVL guidance values can be applied to the UK market with minimal impact” is reported as the report’s own conclusion, not as a wiki recommendation. The eye-shadow Pb 2 mg/kg vs theatre-make-up 5 mg/kg gap, and the suggestion that “eye shadows may need a different guidance limit,” is reported as OPSS’s discussion rather than re-stated as wiki policy.
  • Jurisdictions. The report reviews regulations from UK, EU, US (federal FDA), China, Thailand, Japan, Canada, Germany (BVL non-regulatory), ASEAN block, South Korea (ICCR member), and Brazil (ICCR member). The jurisdictions: array captures the principal-discussed eight (UK, EU, US, CN, JP, CA, DE, TH); KR and BR appear in the ICCR-member listing but the report does not present KR- or BR-specific limits.
  • DOI / access URL. No DOI; this is a UK government report not assigned a DOI. The access_url is the gov.uk publication page for the report; the actual PDF URL on the asset CDN rotates and is not stable enough to record.
  • Audit subagent application (2026-05-17). Fresh-context audit subagent reviewed against the PDF; verdict REVISE. Findings applied: (a) corrected Table 16e Hg LOQ range in Key numbers from “<0.005-<0.073” to “<0.002-<0.073” (mouthwash LOQ <0.002 was the true minimum, not <0.005); (b) added a flag that the Foundation Hg row in Table 16e (LOQ 0.015, 1 >BVL, max 0.585) is identical to the Sb-Foundation row in Table 16a and contradicts the Section 4.5 chart on p. 37 — the row appears to be a source-side copy-paste error in the OPSS report; (c) completed the semi-quantitative-scan element enumeration to match Appendix 3 (Ba, Co, Cr, Cu, Li, Mo, Ni, Sn, Tl, V, Zr) rather than the truncated list previously written; (d) softened “policy-actionable” to “strongest direct-evidence signal” in Implications to keep the wording in the literature-reporting register rather than the policy register; (e) removed unsupported method-vendor name examples (NRCC, JMP/SAS) from this brand-firewall note since neither is mentioned in the OPSS paper itself. Findings rejected as false positives: the audit flagged four products: slugs (eye-makeup, makeup-foundation-powders-blush, sun-suntan-products, mouthwash-oral-rinse) as absent from the 2026-05-17 taxonomy snapshot; verified that all four exist as live pages under wiki/products/ and the routing audit run on this source resolved all five product targets cleanly (0 unresolved). The taxonomy snapshot is stale relative to the Cat 13 / Cat 20 scaffolding committed earlier on 2026-05-17; the snapshot needs refreshing, not the source page. Net verdict: revised, retained.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.

CommitDateDescription
ce3e07c2026-05-28activation | Vercel DATACITE env slots set, curators.md filled with founder entry + six scoped reviewer invitations, peer-review onboarding playbook drafted
51400b92026-05-28audit-queue: gasparik2017-wild-boar-slovakia-metals audited-revised