Heavy Metals in Infant Clothing: Assessing Dermal Exposure Risks and Pathways for Sustainable Textile Policies

Xiong et al. 2025 — Heavy metals (9-analyte panel) in 33 infant/adult clothing samples with bioaccessibility, HQ/HI/CR and HaCaT cytotoxicity

This Toxics paper measured total and artificial-sweat-bioaccessible concentrations of nine heavy metals (As, Cd, Cr, Cu, Mn, Ni, Pb, Zn, Fe) by ICP-MS in 33 unwashed clothing samples purchased from online retail platforms, then computed ECHA-/US-EPA-formula dermal exposure, hazard quotient (HQ), hazard index (HI) and carcinogenic risk (CR) across five wear-scenario groups (adult-male T-shirt, adult-male underwear, adult-woman blouse, adult-woman underwear and infant under 1 year in a one-piece pajama). Infant-scenario HI exceeded the safe threshold (HI = 1.13) driven by Cd (HQ = 1.12), and infant Cr CR reached 4.35 × 10⁻⁵, approximately 27-fold higher than infant As CR. Artificial-sweat bioaccessibility was highest for Cr (37.78%) and Ni (28.51%); Pb and Fe were below detection in the sweat extract. HaCaT keratinocyte assays of sweat extracts produced dose-dependent cytotoxicity (22-59% viability reduction across T1-T5 groups) and ROS elevation (+130-131% in T2/T4; +17% in T1/T3/T5), with T2 the most pronounced cytotoxic group.

Key numbers

• Sample basis (Section 2.1, p. 3; Table 1, p. 3): 33 unwashed garment samples (1.0 g per analytical replicate). Sample-level fabric density 5-36 mg/cm² (Table 1). Sample pH 3.48-9.77 (Table 1, n=3 per sample, ISO 3071:2005). Materials include 100% Cot, 100% PA, 100% PET, 100% Ct, 100% Vs plus mixed-fiber blends (70% E / 30% PET, 60% Cot / 33% PET / 7% PA, etc.).
• Total heavy metal concentrations across the 33-sample set (Table 5, p. 7; mg/kg; ICP-MS):
  • As: mean 1.01, SD 0.30, max 1.87, min 0.53.
  • Cd: mean 0.16, SD 0.04, max 0.25, min 0.09.
  • Cr: mean 1.98, SD 1.12, max 4.32, min 0.23.
  • Cu: mean 11.30, SD 5.14, max 23.92, min 2.80.
  • Mn: mean 2.16, SD 0.60, max 3.61, min 1.24.
  • Ni: mean 1.51, SD 0.71, max 3.45, min 0.62.
  • Pb: mean 0.19, SD 0.09, max 0.47, min 0.07.
  • Zn: mean 13.83, SD 6.96, max 27.38, min 3.18.
  • Fe: mean 31.68, SD 17.79, max 83.85, min 4.84.
• OEKO-TEX Standard 100 limit values reported by the paper for cross-comparison (Table 6, p. 7; mg/kg): Baby wear (Class I) As 0.2, Cd 0.1, Cr 1.0, Cu 25, Ni 1.0, Pb 0.2 (Mn, Zn, Fe not specified by the standard). With-skin-contact (Class II) and without-skin-contact (Class III) and Accessories: As 1.0, Cd 0.1, Cr 2.0, Cu 50, Ni 4.0, Pb 1.0. The paper compares against this benchmark but states (Abstract, p. 1) that 80% of samples exceeded Class I limits for As (mean 1.01 mg/kg), Cd (max 0.25 mg/kg) and Cr (max 4.32 mg/kg); narrative on p. 7 reports five toxic metals (As, Cd, Cr, Pb and Ni) found to exceed the Class I (baby wear) limits.
• Artificial-sweat-bioaccessible concentrations (Table 9, p. 8; mg/kg; ISO 3160/2 artificial sweat, 36 °C, 24 h, 100 rpm):
  • As: mean 0.13, SD 0.17, max 0.16, min 0.07; bioaccessibility 12.91%.
  • Cd: mean 0.01, SD 0.01, max 0.02, min 0.01; bioaccessibility 6.33%.
  • Cr: mean 0.75, SD 0.55, max 1.33, min 0.06; bioaccessibility 37.78%.
  • Cu: mean 1.67, SD 0.73, max 3.18, min 0.63; bioaccessibility 14.78%.
  • Mn: mean 0.60, SD 0.54, max 1.68, min 0.06; bioaccessibility 27.63%.
  • Ni: mean 0.43, SD 0.18, max 0.64, min 0.19; bioaccessibility 28.51%.
  • Pb: below detection in sweat extract.
  • Zn: mean 2.77, SD 1.79, max 5.03, min 0.33; bioaccessibility 20.02%.
  • Fe: below detection in sweat extract.
• Bioaccessibility ranking (Section 3.3, p. 8): release order Zn > Cu > Cr > Mn > Ni > As > Cd; bioaccessibility percent ranking Cr > Ni > Mn > Zn > Cu > As > Cd. Pb and Fe below detection limits.
• Dermal-exposure equation (Section 2.3, p. 4): Exp_derm = C_cloth × 10⁻⁶ × d_cloth × A_skin × F_mig × F_contact × F_pen × T_contact × n / BW. Inputs (Table 2, p. 4): F_mig 0.005 (1/d), F_contact 1, F_pen 0.01 (As: 0.03), T_contact 1 d, n 1 event/d; body weight adult male 70 kg, adult woman 60 kg, infant under 1 year 6.98 kg; skin area adult male T-shirt 7120 cm² / underwear 1980 cm², adult woman blouse 6941 cm² / underwear 1723 cm², infant under 1 year one-piece pajamas 2754 cm².
• Reference doses RfD_dermal (Table 3, p. 5; mg·kg⁻¹·d⁻¹): As 3.00×10⁻⁴ (SF_dermal 1.50×10⁰), Cd 5.00×10⁻⁸, Cr 7.50×10⁻⁵ (SF_dermal 2.00×10⁻¹), Cu 4.00×10⁻², Mn 1.40×10⁻¹, Ni 8.00×10⁻⁴, Fe 7.00×10⁻¹, Zn 3.00×10⁻¹. Pb has no listed RfD_dermal in this paper’s Table 3.
• HQ values by wear scenario (Table 7, p. 8):
  • Adult-male T-shirt: As 9.22×10⁻⁴, Cd 2.88×10⁻¹, Cr 2.49×10⁻³, Cu 2.67×10⁻⁵, Mn 1.42×10⁻⁶, Ni 1.75×10⁻⁴, Pb 2.41×10⁻⁶, Zn 4.05×10⁻⁶, Fe 4.05×10⁻⁶; HI 2.91×10⁻¹.
  • Adult-male underwear: As 2.56×10⁻⁴, Cd 8.00×10⁻², Cr 6.94×10⁻⁴, Cu 7.42×10⁻⁶, Mn 3.96×10⁻⁷, Ni 4.86×10⁻⁵, Pb 6.69×10⁻⁷, Zn 1.13×10⁻⁶, Fe 1.12×10⁻⁶; HI 8.10×10⁻².
  • Adult-woman blouse: As 1.05×10⁻³, Cd 3.27×10⁻¹, Cr 2.84×10⁻³, Cu 3.04×10⁻⁵, Mn 1.62×10⁻⁶, Ni 1.99×10⁻⁴, Pb 2.74×10⁻⁶, Zn 4.61×10⁻⁶, Fe 4.60×10⁻⁶; HI 3.31×10⁻¹.
  • Adult-woman underwear: As 2.60×10⁻⁴, Cd 8.12×10⁻², Cr 7.04×10⁻⁴, Cu 7.53×10⁻⁶, Mn 4.02×10⁻⁷, Ni 4.94×10⁻⁵, Pb 6.79×10⁻⁷, Zn 1.14×10⁻⁶, Fe 1.14×10⁻⁶; HI 8.22×10⁻².
  • Infants under 1 year (one-piece pajamas, BW 6.98 kg): As 3.58×10⁻³, Cd 1.12×10⁰, Cr 9.68×10⁻³, Cu 1.04×10⁻⁴, Mn 5.53×10⁻⁶, Ni 6.78×10⁻⁴, Pb 9.33×10⁻⁶, Zn 1.57×10⁻⁵, Fe 8.27×10⁻⁷; HI 1.13×10⁰. Cd HQ > 1 and HI > 1 for the infant scenario.
• CR values by wear scenario (Table 8, p. 8; carcinogenic risk for As and Cr only; remaining analytes had no listed SF_dermal):
  • Adult-male T-shirt: As 4.15×10⁻⁷, Cr 1.12×10⁻⁵.
  • Adult-male underwear: As 1.15×10⁻⁷, Cr 3.12×10⁻⁶.
  • Adult-woman blouse: As 4.72×10⁻⁷, Cr 1.28×10⁻⁵.
  • Adult-woman underwear: As 1.17×10⁻⁷, Cr 3.17×10⁻⁶.
  • Infants under 1 year: As 1.61×10⁻⁶, Cr 4.35×10⁻⁵. Both within the paper’s acceptable range 10⁻⁴ to 10⁻⁶, with Cr roughly 27-fold higher than As in the infant scenario (Section 3.2, p. 7).
• HaCaT keratinocyte assays (Section 3.4-3.5, pp. 8-9):
  • CCK-8 cell viability after 24 h exposure to artificial-sweat clothing extract: time-dependent reduction in survival across treatment groups; T2 group most pronounced cytotoxic effect (cell survival markedly lower than all other test conditions); Abstract states 22-59% viability reduction across exposure conditions.
  • Cellular morphology (200× phase-contrast, Figure 2B): T3 and T4 groups exhibited loss of epithelial morphology, membrane blebbing and increased cell detachment, consistent with dose-dependent cytotoxic damage.
  • ROS production via DCFH-DA flow cytometry (Figure 3): T2 and T4 groups showed 130% and 131% fluorescence intensity vs control (100%); T1, T3 and T5 groups slight increase (117 ± 4.9%). Paper hypothesizes Cr content drives the ROS response.
• Material-color contamination pattern (Section 1, pp. 1-2; Section 3.2/3.5 commentary): black polyester consistently accumulates Cr and Ti; dark-colored textiles (black/gray/navy) carry higher Cr/Co levels than light-colored (pink/red) ones. Authors’ Introduction attributes additive sources as follows (p. 1-2; not directly quantified in this study): Sb serves as a flame-retardant catalyst in polyester; Cr and Cu act as mordants and metal-complex dyes (especially in black polyamide and colored cotton); Ti and Zn are used in moisture-wicking treatments. Dark-colored synthetic fabrics exhibited elevated metal loads (Abstract, p. 1).
• ICP-MS detection limits (Section 2.8, p. 6): As 0.006 µg/kg, Cd 0.003 µg/kg, Zn 0.003 µg/kg, Cu 0.003 µg/kg, Mn 0.002 µg/kg, Ni 0.005 µg/kg, Cr 0.02 µg/kg.

Methods (brief)

33 unwashed cost-effective clothing samples were randomly purchased from online market platforms, dried at 60 °C for 48 h, cut into 1 × 1 cm slices and weighed for fabric density (Table 1). pH measured per ISO 3071:2005 (1.0 g into 50 mL pure water, 2 cm × 2 cm pieces in triplicate, 2 h continuous shaking at 27 °C). For total-metal analysis, 0.5 g oven-dried cloth was microwave-digested with 10 mL concentrated HNO₃ (65% Suprapur grade, E. Merck, Darmstadt, Germany) in a Milestone Start D Microwave Digestion System (5 min ramp to 105 °C, 15 min at 180 °C, 20 min terminal at 200 °C). Cooled digests were 0.45 µm membrane-filtered, volumetrically adjusted to 25 mL with deionized water and cryogenically stored at −20 °C. Elemental quantification by ICP-MS with reagent blanks and certified reference material (NIST spinach leaves). Artificial sweat per ISO 3160/2 (lactic acid 0.1 wt%, urea 0.1 wt%, sodium chloride 0.5 wt% in 1 L deionized water); 1 g pre-treated textile in 20 mL freshly prepared sweat at 36 °C, 100 rpm shaking, 24 h, then 0.45 µm filtration (post 500 rpm, 10 min centrifugation) and ICP-MS of bioaccessible Cd, Cr, Cu, Mn, Ni, Zn, As (Pb and Fe below detection in sweat extract). HaCaT human skin keratinocytes in MEM (NEAA) basal medium with 10% FBS and 1% penicillin-streptomycin at 37 °C, 5% CO₂; 1 × 10⁴ cells/well in 96-well plates, 24 h adherence, then 24 h exposure to the sweat-extract clothing supernatant (0.22 µm filtered). Cell viability via CCK-8 (10 µL/well, 2 h, OD at 450 nm, Molecular Devices LLC, San Jose, CA). Inverted phase-contrast microscope (TS-100, Nikon, Tokyo) for morphology at 200×. Intracellular ROS via DCFH-DA flow cytometry (CyFlow Cube 6, Sysmex Partec, Nuremberg; ex/em 488/525 nm; 10,000 events). Reagent kit from Yfxbio Biotech Co., Ltd. (Nanjing, China). Statistics in GraphPad Prism v7.0 (GraphPad Software LLC, San Diego, CA); significance at p < 0.05. QA/QC: reagent blanks, blank controls, NIST spinach-leaves CRM (total digest), GB/T 17593-2013 reference. Stated limitations: only dermal exposure modeled (inhalation of textile microfibers not assessed); no socioeconomic-barrier analysis for low/middle-income-country populations; samples were unwashed (worst-case exposure scenario, not after-laundering); Cr measured as total Cr, not speciated for Cr-VI; As measured as total As, not speciated for inorganic As.

Implications

• Certification (HMTc): Direct occurrence and dermal-exposure evidence for an infant-textile / direct-skin-contact product category. The infant-scenario HI = 1.13 (Cd HQ = 1.12) and Cr-driven CR = 4.35 × 10⁻⁵ in the infant scenario provide quantitative anchors for the dermal-exposure pathway in the Infant and Child Durable Goods and Textiles HMTc category (Cat 10). Bioaccessibility fractions (Cr 37.78%, Ni 28.51%, Mn 27.63%, Zn 20.02%, Cu 14.78%, As 12.91%, Cd 6.33%) are useful for bioaccessibility-based standards work on textile-borne metals. The total-Cr / total-As speciation limitation should be noted in any downstream synthesis: the Cr-driven CR figure uses an SF_dermal value (2.00 × 10⁻¹) tied to Cr-VI but the measured analyte is total Cr, so the CR figure should be treated as conservative upper bound only, consistent with conventions Part 14 speciation discipline. The paper compares against OEKO-TEX Standard 100 (an industry private standard, not a government regulation) rather than against a public-record regulatory cap; the wiki regulations layer should not adopt OEKO-TEX as a regulation slug without an explicit decision.
• Courses: Useful teaching case for the dermal-exposure pathway in textiles, ECHA-formula dermal exposure modeling for vulnerable populations (infants under 1 year), bioaccessibility-vs-total-content reasoning, and HaCaT keratinocyte cytotoxicity as a complement to occurrence data.
• App: Indirect relevance — clothing is outside the primary food-ingredient app scope, but any future personal-care / direct-skin-contact app extension can reference this paper’s exposure equation and per-analyte HQ/HI/CR scenario table.
• Microbiome (if applicable): Not directly addressed by this paper. The ROS-mediated cytotoxicity mechanism in keratinocytes is consistent with broader Cr-skin-barrier literature cited by the authors (ref [41]) but the paper does not measure microbiome endpoints.

Wiki pages this source may touch

• arsenic-total
• cadmium
• chromium
• copper
• manganese
• nickel
• lead
• zinc
• iron
• infant-clothing

Verification notes

• 2026-05-17 fresh ingest (Claude Opus 4.7, autonomous v2.0 manual-fetch skill): NEW path; no prior wiki page existed for this DOI / raw_handle / cite_key per three identity checks.
• metals: [tAs, Cd, Cr, Cu, Mn, Ni, Pb, Zn, Fe] uses total-arsenic and total-chromium per Part 14 speciation discipline: the paper measures total As and total Cr by ICP-MS (Section 2.2, p. 4; Section 2.8 detection limits). Authors used Cr-VI-style SF_dermal = 2.00 × 10⁻¹ in the CR calculation; downstream readers should treat the resulting CR figure as conservative upper bound, not a Cr-VI-specific risk estimate.
• matrices: [textile] — adding textile as a new matrix bare-string vocabulary entry. No existing controlled-vocab term in system-prompt fits clothing/garment; cosmetics-source pages use [cosmetics] as analogous broad-string matrix, and textile is the parallel broad term for finished apparel. Recommend formal addition to the matrices controlled vocabulary at next vocabulary review.
• products: ["[[products/infant-clothing]]"] — paper measures infant clothing in the explicit infant scenario (one-piece pajamas, BW 6.98 kg, A_skin 2754 cm²) plus adult-male / adult-woman scenarios as comparison groups. The samples themselves are drawn from a single common pool of 33 garments and evaluated against both adult and infant scenarios using shared concentration data, so the infant-clothing product page is the primary routing destination; adult-clothing routing is not added because there is no adult-clothing product page and the adult scenarios are comparison framing rather than separately-sampled adult garments. The infant-clothing page is currently a scaffold (products_status: scaffold, literature_scope: thin); this source contributes the first quantitative occurrence + bioaccessibility + HQ/HI/CR evidence for the row.
• ingredients: [] — paper measures finished garments, not raw textile ingredients (cotton fibers, polyester yarn, dye constituents). No ingredient-level routing.
• jurisdictions: [CN] — samples drawn from Chinese online retail platforms (Yunnan / Kunming research base); methodology applies ECHA (EU) and US EPA dermal-exposure parameters but those are method-of-assessment frameworks, not the jurisdiction of the sampled product population.
• Pb and Fe below detection in artificial-sweat extract (Table 9 footnote ”—”); both are reported in total-content Table 5 but not in the bioaccessible-content Table 9. Treat this as detection-limit censoring, not as zero bioaccessibility.
• OEKO-TEX Standard 100 is referenced in the paper (Table 6) as the cross-comparison benchmark. It is an industry private standard (publisher: OEKO-TEX Association), not a government regulation, and is not added to wiki/regulations/ from this ingest per Part 10 (regulations are hard agency identifiers; private standards are not). If Karen later decides to track OEKO-TEX Class I/II/III/Accessories limits as a regulatory-context layer for textile products, that requires a separate decision and a separate rule_id slug.
• Brand firewall (Part 12): paper does not name brands; sample identification is by sample number, material composition and color only (Table 1). No brand attribution to delete.
• The paper’s narrative on p. 7 (“five toxic metals (As, Cd, Cr, Pb, and Ni) were found to exceed the stringent Class I limits”) differs slightly from the Abstract framing (“80% of samples exceeded OEKO-TEX Class I limits for As, Cd, and Cr”). Both statements are reproduced above as quoted; the difference appears to be the Abstract focusing on the top three exceedance metals by sample-count while the Results narrative enumerates all five Class-I-exceeding metals across the dataset. No internal-contradiction flag; the two statements are consistent if read as different aggregations of the same per-sample comparison.
• HQ values for the infant scenario reproduced from Table 7 column “Infants”; HI 1.13 × 10⁰ is the sum across the nine listed HQ values (verified arithmetically from the listed components within rounding tolerance). Cd HQ 1.12 × 10⁰ is the dominant driver.
• T1-T5 group identity is not specified in the body text I extracted; figures show six groups (Ctrl plus T1-T5) but the paper does not enumerate which of the 33 garments correspond to T1-T5 in the cell-exposure experiments. Treat the cytotoxicity findings as group-level effects of the sweat extracts, not as sample-specific attributions.
• Audit subagent (2026-05-17, fresh-context general-purpose) flagged that the wiki body had “black polyester consistently accumulates Cr and Cu”; verified against PDF p. 2 (“Synthetic fabrics exhibit distinct contamination patterns—black polyester consistently accumulates Cr and Ti”) — corrected to “Cr and Ti.” The earlier Cu attribution was a misread of the Introduction.
• Audit subagent (2026-05-17) flagged that the wiki body had “Ti and Sb appear in synthetic and flame-retardant treatments respectively (not quantified in this study)”; verified against PDF pp. 1-2 (paper distinguishes three additive roles: Sb as flame-retardant catalyst in polyester; Cr/Cu as mordants and metal-complex dyes especially in black polyamide and colored cotton; Ti/Zn as moisture-wicking treatments) — rewritten to preserve all three role attributions and to correctly assign Ti to moisture-wicking rather than to generic “synthetic treatments.”
• Audit subagent (2026-05-17) ⚠️-flagged products/infant-clothing as not in the taxonomy snapshot; verified against wiki/products/infant-clothing.md which exists with hmtc_category: 10, hmtc_row: 9, hmtc_subcategory_name: "Clothing (onesies, pajamas, outerwear)", products_status: scaffold, literature_scope: thin. Routing audit confirms infant-clothing,locked_hmtc_row,xiong2025-heavy-metals-infant-clothing,...,direct_evidence in product_source_routing_audit.csv with no entry in routing_unresolved.csv. Finding is a snapshot-staleness false positive; routing destination resolves cleanly.
• Audit subagent (2026-05-17) ⚠️-flagged matrices: [textile] as a new vocabulary entry; finding accepted as documented above — this Verification notes already declares textile as a proposed bare-string vocabulary addition awaiting formal review, parallel to the cosmetics precedent. No correction needed.

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.

Commit	Date	Description
ce3e07c	2026-05-28	activation | Vercel DATACITE env slots set, curators.md filled with founder entry + six scoped reviewer invitations, peer-review onboarding playbook drafted
51400b9	2026-05-28	audit-queue: gasparik2017-wild-boar-slovakia-metals audited-revised