Stefaniak et al. 2014 — Dissolution of Ni, Be, Cr in artificial sweat for dermal bioaccessibility
This Environmental Science: Processes & Impacts paper compares reactant-assisted digestion (“total” or “soluble” mass) vs artificial-sweat extraction (bioaccessible mass) for three metal sensitizers - beryllium metal, nickel metal, and chromium carbide (Cr3C2) - and uses speciated isotope dilution mass spectrometry (SIDMS) to determine Cr-VI vs Cr-III speciation in artificial sweat over time. The work establishes that reactant-assisted digestion overstates bioaccessible (dermal) exposure to all three metals, and that Cr-VI added to artificial sweat decays into Cr-III over hours.
Key numbers
- Bioaccessibility ratios (Table 2, fraction dissolved in sweat / fraction in chemical digestion):
- Be: sweat 0.15 ± 0.01% vs digestion 0.29 ± 0.02% → 1/2 ratio
- Ni: sweat 0.0013 ± 0.0003% vs digestion 0.14 ± 0.01% → 1/108 ratio
- Cr-total: sweat 0.0024 ± 0.0008% vs digestion 6.08 ± 0.31% → 1/2500 ratio
- Artificial sweat composition (Harvey et al. recipe): 61 constituents; highest-concentration components Na+ 30.7 mM, Cl- 23.2 mM, lactic acid 14.0 mM, urea 10.0 mM, K+ 6.1 mM, ammonium 5.3 mM; pH adjusted to 5.3 with NaOH. Skin pH range generally 4.2-6.1 (median 5.3).
- Sweat extraction conditions: 0.050 L artificial sweat (calculated from average adult whole-body sweat output 1.08 L/h × 5% body-area fraction for hands × 1 hour); 0.05 g powder per sample; powders sealed between two filters and immersed in artificial sweat using the static dissolution technique; triplicate samples at 33, 36, 45 °C across 7 time points spanning ~12 hours.
- Dissolution kinetics (Fig. 3, p. 345-346): Be and Cr both biphasic — initial rapid phase half-time 4-6 h for Be and 3-5 h for Cr, both followed by a slower long-term phase with half-times on the order of hundreds of days. Ni dissolution linear over time with calculated half-times on the order of several hundred days.
- On total-mass basis (per Results section): Be <0.35%, Ni <0.003%, Cr <0.004% of total powder dissolved in artificial sweat over the test duration.
- Cr speciation in sweat (Fig. 4 / Results): Cr-VI added as potassium dichromate at 2 or 10 µg/mL decayed over 420 minutes; decay accompanied by Cr-III formation in the sweat model (speciated isotope dilution MS using 53Cr-VI / 50Cr-III).
- Powder characterization (Fig. 2): Be metal particle size 3.2 µm (geometric mean, GSD 1.8); BET surface area 5.6 ± 0.02 m²/g; density 1.86 ± 0.02 g/cm³. Cr3C2 cluster size 2.6 µm (GSD 1.6); surface area 1.04 ± 0.02 m²/g; density 6.97 ± 0.19 g/cm³. Ni metal cluster size 4.0 µm (GSD 1.5); surface area 7.04 ± 0.18 m²/g; density 9.46 ± 0.29 g/cm³.
Methods (brief)
Three high-purity metal sensitizer powders characterized by Auger spectroscopy (surface composition 50-75 Å outer layer), XRD (crystalline composition), TEM/SEM (morphology, particle size), helium pycnometry (density), N2 BET (specific surface area). Reactant-assisted digestion protocols: Be 0.01 M HCl (Profumo et al. 2008); Ni 0.1 M ammonium citrate/citric acid pH 4.4 sequential digestion (Zatka et al.); Cr-total NIOSH 7303 concentrated HNO3/HCl at 95 °C. Sweat extraction: artificial sweat (Harvey et al. recipe) at 33, 36, 45 °C, triplicate samples at 7 time points across 12 h. Metals quantified by ICP-OES. Cr speciation by SIDMS using 53Cr-VI/50Cr-III isotope dilution with IC-ICP-DRC-MS. Limitations stated: static (not flow-through) extraction may underestimate sweat-removal kinetics; chosen powder masses (0.05 g) exceed realistic dermal loadings to ensure analytical detection limits; static contact time of 12 h represents long workshift exposure.
Implications
- Certification (HMTc): Direct methodological evidence that “total Cr” measurements digested by acids massively overstate dermally bioaccessible Cr (1/2500 ratio); for Ni and Be the overstatement is also large (1/108 and 1/2 respectively). Important for any HMI threshold work on children’s products where total-Cr measurements are the primary occurrence data: dermal-route risk modeling needs to apply a bioaccessibility correction or accept conservative overestimation. The Cr-VI → Cr-III decay in sweat is relevant to Part 14’s discipline around Cr-VI vs Cr discipline: skin-surface Cr-VI does not remain Cr-VI for long.
- Courses: Teaching reference for the distinction between “total,” “soluble,” and “bioaccessible” metal in occupational/consumer dermal exposure assessment. Pairs with the OEHHA and ATSDR dermal-absorption guidance.
- App: Not relevant to ingredient contamination_profile (no food-matrix data).
Wiki pages updated on ingest
Verification notes
- “artificial-sweat” matrix is not in the standard matrices vocabulary; flagging for Part 10 review. Used as bare string.
- This paper is methodology + materials-science occurrence, not finished-product surveillance. The 1/2500 Cr bioaccessibility ratio is for chromium carbide powder; ratios for Cr-VI in cosmetic pigments may differ and would need source-specific measurement.
- Cr-VI half-life in artificial sweat (“hours, decayed over 420 min” per Results) is a non-trivial finding for the Cr-VI vs total-Cr discipline in dermal exposure assessment.
- 2026-05-17 merge-enhance pass: corrected sweat-volume calculation unit description (1.08 L/h whole-body output × 5% body-area fraction for hands, not L/h/m²); renamed “Wiki pages this source may touch” to template-standard “Wiki pages updated on ingest”; added replicate-sample structure note (triplicate samples at each temperature).
- 2026-05-17 audit-application pass (fresh-context subagent REVISE verdict, 2 concerns applied, 0 rejected): restored potassium (K+ 6.1 mM) to the artificial-sweat highest-concentration components list (verified PDF p. 344); split conflated initial-rapid-phase half-times so Be is 4-6 h and Cr is 3-5 h separately (verified PDF p. 345). Soft-flags on Implications framing and limitations gloss reviewed and left as-is: both stay within Part 2 bounds (no HMTc threshold proposal, no consumer risk advisory) and within faithful source rendering.
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.