Heavy Metal Index

Toxic chemicals in a child's world: an investigation into PVC plastic products

Source access

Publication
Greenpeace Research Laboratories, University of Exeter (UK); June 2001
Year
2001
Authors
Stringer, Ruth; Johnston, Paul; Erry, Bea
Source type
ngo report
License
copyrighted third party

Stringer, Johnston & Erry 2001 — Toxic chemicals in a child’s world: an investigation into PVC plastic products

This 22-page Greenpeace Research Laboratories technical report by Ruth Stringer, Paul Johnston, and Bea Erry of the University of Exeter, published June 2001, extends the line of NGO PVC-additive investigations that began with digangi1997-greenpeace-vinyl-children-products and the contemporaneous Greenpeace Stringer et al. 2000 children’s-toy phthalate survey (not in corpus). Where Di Gangi 1997 focused on toys and concentrated on lead and cadmium in PVC substrate, Stringer 2001 broadens the scope to 54 vinyl consumer products from 20 countries — toys and teethers, clothing and bibs, household furnishings, hygiene products, and food-contact items — and broadens the analyte panel to add bisphenol A and eight organotin compounds alongside phthalates, total tin, lead, and cadmium. STAT Analysis (Chicago, IL, USA) performed the phthalate/BPA/Pb/Cd/Sn analyses; GALAB (Geesthacht, Germany) performed organotin speciation on 14 samples per DIN EN 17025. The investigation found phthalates in all but one PVC sample (a Japanese teether, CW0141, which instead contained bisphenol A at 407 mg/kg), substrate cadmium ranging up to 230 mg/kg (a US drawer liner, CW0150) with two PVC products exceeding the EU EEC 1991 100 mg/kg cadmium maximum for PVC, substrate lead ranging up to 441 mg/kg in chlorinated PVC samples (a Chilean wallpaper, CW0129) and up to 5,220 mg/kg in a non-PVC US bib (CW0115), and organotin compounds in all but one of the 14 samples speciated — peaking at 51,338 µg/kg total organotins in a Canadian floor tile (CW0050), with the predominant organotin species being dibutyltin and dioctyltin. The report concludes that PVC additives “are still being widely used in products either specifically marketed for children, or for use in applications which could entail children experiencing significant exposure” and that “the only way to protect children from these harmful chemicals is to eliminate them from the products and hence from the home environment.” This page documents the heavy-metals findings (Pb, Cd, Sn, organotins); phthalate and bisphenol A findings are summarised for scope but the wiki’s HMI scope is the metals.

Key numbers

Sampling frame, screening, and analytical workflow (pp. 6–8)

  • Total products analysed: 54 consumer items from 20 countries; two products subdivided into two analytical units each (CW0006 bottle/syrup; CW0140 spoon-body/handle) yielding 56 analytical units total.
  • Chlorine-screen positive (assumed PVC): 45 of 56 units; chlorine-screen negative (non-PVC): 11.
  • Phthalate/BPA/Cd/Pb/Sn analyses: all 56 units at STAT Analysis, Chicago, IL, USA.
  • Organotin speciation: 14 samples at GALAB, Geesthacht, Germany; analysed for monobutyltin, dibutyltin, tributyltin, tetrabutyltin, monooctyltin, dioctyltin, tricyclohexyltin, triphenyltin per DIN EN 17025.
  • Limits of detection: phthalates 100 mg/kg; bisphenol A 100 mg/kg; cadmium 0.500 mg/kg; lead 0.500 mg/kg; tin 2.50 mg/kg; organotin individual species 0.3–0.4 µg/kg.

Cadmium in PVC substrate (Tables 1–5; mg/kg by weight; pp. 8–13)

  • Cadmium detected in 19 of 56 units; 18 of those 19 were PVC. The 19th was a Russian 100% cotton bib (CW0008) at 0.503 mg/kg (Greenpeace attributes to pigment in printed decoration).
  • Among PVC samples, cadmium concentrations ranged from 0.57 mg/kg (Thailand bath mat, CW0040) to 230 mg/kg (US plastic-coated paper drawer liner, CW0150).
  • Two PVC products exceeded the EU EEC 1991 100 mg/kg maximum for cadmium in PVC: drawer liner USA CW0150 at 230 mg/kg, and waterproof pants New Zealand CW0055 at 132 mg/kg.
  • Other notable Cd values: Brazilian car sunshade (CW0065) 3.47 mg/kg; USA stroller cover (CW0107) 1.09 mg/kg; Danish school-kit (CW0051) 6.09 mg/kg; Chilean wallpaper (CW0129) 6.33 mg/kg; Thai changing mat (CW0035) 35.7 mg/kg; Thai mattress cover (CW0037) 16.2 mg/kg; USA mattress pad (CW0106) 83.3 mg/kg; Hungarian cooler (CW0057) 69.9 mg/kg; Brazilian cooler (CW0062) 18.5 mg/kg; Mexican waterproof pants (CW0005) 1.15 mg/kg; Russian bib (CW0008, 100% cotton non-PVC) 0.503 mg/kg; Canadian bib (CW0048) 43.3 mg/kg; Japanese rain hat (CW0142) 1.18 mg/kg; USA shower curtain (CW0112) 3.69 mg/kg; Netherlands suckers (CW0098) 0.869 mg/kg. Spanish wallpaper (CW0104) Cd was below LOD at <0.500 mg/kg (Table 3).

Lead in PVC and non-PVC substrate (Tables 1–5; mg/kg; pp. 9, 12–14)

  • Lead detected in 26 of 56 units; 21 of those 26 were PVC. The five non-PVC lead-containing samples included the highest single dataset value.
  • Highest single lead value in the dataset: 5,220 mg/kg (US plastic-laminated cotton bib, CW0115, non-PVC). Greenpeace attributes to orange pigment in the brightly coloured product.
  • Highest lead value among PVC samples: 441 mg/kg (Chilean wallpaper, CW0129).
  • Other notable Pb values among chlorinated samples: Danish school-kit (CW0051) 35.40 mg/kg; Austrian inflatable swim ring (CW0060) 1.37 mg/kg; Brazilian car sunshade (CW0065) 3.27 mg/kg; USA stroller cover (CW0107) 1.03 mg/kg; New Zealand waterproof pants (CW0055) 15.7 mg/kg; Russian cotton bib (CW0008, non-PVC) 91.7 mg/kg; Canadian bib (CW0048) 0.761 mg/kg; USA shoes (CW0124) 0.726 mg/kg; USA hat (CW0125) 0.835 mg/kg; Japanese rain hat (CW0142) 179 mg/kg; Thai changing mat (CW0035) 74.5 mg/kg; USA changing mat (CW0105) <0.500 mg/kg; Philippines medicine dispenser (CW0079, non-PVC) 248 mg/kg; Italian spoon-handle (CW0140b) <0.500 mg/kg (note: handle PVC; main body non-PVC); Hungarian cooler (CW0057) 22.7 mg/kg; Brazilian cooler (CW0062) 179 mg/kg; Chilean place mat (CW0130) 0.845 mg/kg; USA shower curtain (CW0112) 17.4 mg/kg; USA carpet protector (CW0148) <0.500 mg/kg; Spanish wallpaper (CW0104) 1.69 mg/kg; USA drawer liner (CW0150) 15.1 mg/kg; USA inflatable sofa (CW0053) 32.9 mg/kg; USA lady’s handbag (CW0113) 22.4 mg/kg; Netherlands suckers (CW0098) 2.22 mg/kg.
  • Lowest detectable Pb among chlorinated PVC samples: 0.51 mg/kg (Denmark pencil case, CW0051; reported here as the school-kit, given as 35.40 mg/kg in Table 1 — corrected to source value: the introduction text states “lead concentrations ranged between 0.51 mg/kg in a pencil case from Denmark (CW0051)”; Table 1 shows 35.40 mg/kg for CW0051 schoolkit lead. Per source p.9 narrative the 0.51 mg/kg figure refers to pencil-case lead among the chlorinated samples; per Table 1 the value is 35.40 mg/kg for the same sample number. The discrepancy is internal to the report; flagged in Verification notes.)

Tin (total elemental, mg/kg; Tables 1–5; p. 9)

  • Tin detected in 17 PVC samples and 2 non-PVC samples (19 of 56 units total).
  • Lowest detected Sn: 3.45 mg/kg (USA waterproof pants, CW0128).
  • Highest Sn: 113 mg/kg (Chilean wallpaper, CW0129).
  • Other notable Sn values: Mexican waterproof pants (CW0005) 5.34 mg/kg; Brazilian car sunshade (CW0065) 5.55 mg/kg; USA bib CW0115 (non-PVC, plastic-laminated cotton) 83.7 mg/kg; USA shoes (CW0124) 9.83 mg/kg; USA hat (CW0125) 3.62 mg/kg; Thai bath mat (CW0040) 12.6 mg/kg; Canadian floor tile (CW0050) 76.5 mg/kg; USA carpet protector (CW0148) 2.50 mg/kg (boundary); Spanish wallpaper (CW0104) 1.69 mg/kg (note: <2.50 LOD); USA drawer liner (CW0150) <2.50 mg/kg; USA inflatable sofa (CW0053) <2.50 mg/kg; USA lady’s handbag (CW0113) 5.57 mg/kg; Thai changing mat (CW0035) 25.4 mg/kg; New Zealand changing mat (CW0056) 30.5 mg/kg; Netherlands changing-mat bag (CW0100) 3.51 mg/kg; USA changing mat (CW0110) 6.05 mg/kg; Mexican maize-syrup bottle (CW0006b) 28.5 mg/kg; Italian spoon-handle (CW0140b) 6.88 mg/kg; Hungarian cooler (CW0057) 13.3 mg/kg; UK high-chair cushion (CW0144) 111 mg/kg; Chilean place mat (CW0130) 3.71 mg/kg; Italian nasal aspirator (CW0138) <2.50 mg/kg.

Organotin speciation (µg/kg per species; Tables 1–5; pp. 9–10)

  • Organotins detected in 13 of 14 speciated samples (all except a non-PVC bib).
  • Lowest total organotins: 8.20 µg/kg (USA stroller cover, CW0107; main species monobutyltin 2.10, dibutyltin 2.20, monooctyltin 1.90, dioctyltin 1.60).
  • Highest total organotins: 51,388 µg/kg (Canadian floor tile, CW0050; main species monobutyltin 2,830, dibutyltin 13,600, monooctyltin 16,000, dioctyltin 18,700, tributyltin 258, tetrabutyltin <0.3). Note: the source narrative on p.9 gives the total as “51 338 ug/kg” while the p.11 narrative and Table 3 both give 51,388 µg/kg; the table value (51,388) is used here as the canonical analytical record (the p.9 figure is treated as a typographical error). Discrepancy flagged in Verification notes.
  • Other high-organotin samples: Chilean wallpaper (CW0129) 37,335 µg/kg total (monobutyltin 3,530, dibutyltin 15,500, monooctyltin 3,500, dioctyltin 14,000, tributyltin 805); USA bib CW0115 (non-PVC, plastic-laminated cotton) 12,671 µg/kg total (dibutyltin 11,000 — approaching 90% of organotin content; monobutyltin 1,600); USA bib CW0095 8,778.9 µg/kg total (dibutyltin 64.4, monooctyltin 2,860, dioctyltin 5,840); USA changing mat CW0110 4,372.7 µg/kg total (tributyltin 4,120 — i.e., over 4 mg/kg tributyltin in a children’s changing mat); USA floor tile CW0160 3,677.6 µg/kg total; USA lady’s handbag CW0113 3,045.6 µg/kg total (dibutyltin 2,250, dioctyltin 410, monooctyltin 278); Mexican maize-syrup bottle CW0006b 4,848.8 µg/kg total (predominantly octyltin; monooctyltin 819, dioctyltin 3,940, monobutyltin 19.9, dibutyltin 68.6, tributyltin 1.3); Italian nasal aspirator CW0138 266 µg/kg total (monobutyltin 13.7, dibutyltin 43.7, tributyltin 192, monooctyltin 4.3, dioctyltin 12.3).
  • Predominant organotin species across the speciated subset: dibutyltin (predominant in six samples), followed by dioctyltin (five samples), then tributyltin (two samples). Tricyclohexyltin and triphenyltin were not detected in any sample.
  • Tin mass-balance gap: Greenpeace notes that the sum of organotin species (as tin equivalents) does not account for the total tin found in any of the samples, implying additional organotin species not on the eight-compound panel or inorganic tin compounds.

Phthalates and bisphenol A (scope-context; not the HMI focus)

  • Phthalates detected in 44 of the 45 PVC units (one Japanese teether CW0141 was the lone PVC exception, containing bisphenol A instead). Total phthalate concentrations ranged from 103 mg/kg (Mexican maize-syrup bottle, CW0006b) to 378,590 mg/kg (≈ 37.9% by weight; Thai bath mat, CW0040), with most items containing >10% phthalates by weight.
  • DEHP predominated in 24 samples; DINP predominated in 18; BBP predominated in two; DBP and DEP not detected in any sample.
  • Bisphenol A detected only in two Japanese products: teether (CW0141) at 407 mg/kg and rain hat (CW0142) at 176 mg/kg.
  • US teethers CW0109 and CW0127 contained DINP at 208,000 mg/kg (≈ 20.8%) and 287,000 mg/kg (≈ 28.7%) respectively — both exceeding the 1998 CPSC voluntary phthalate-removal request and the 1999 EU mouthing-articles 0.1% DINP limit (EC 1999/815).
  • The squeeze ball from Japan (CW0143) contained only 114 mg/kg DINP, which Greenpeace flags as too low to plasticise the product — proposing it may be cross-contamination from another manufacturing batch.

Sample-by-sample heavy-metal exceedances of named regulatory limits

  • EU EEC 1991 (Council Directive 91/338) 100 mg/kg cadmium maximum in PVC products: exceeded by CW0150 (USA drawer liner, 230 mg/kg) and CW0055 (New Zealand waterproof pants, 132 mg/kg).
  • EC 1999/815 0.1% (1,000 mg/kg) phthalate (DINP, DEHP, DBP, DIDP, DNOP, BBP) limit for mouthing articles for children under 3 years: exceeded by US teether CW0109 (208,000 mg/kg DINP, ≈208× the limit) and US crib-rail teether CW0127 (287,000 mg/kg DINP, ≈287× the limit). The Japanese teether CW0141 was below the phthalate threshold (<100 mg/kg of each species) but contained 407 mg/kg bisphenol A.
  • No nationally binding limit for cadmium in non-PVC children’s products (e.g., the 0.503 mg/kg Cd in the cotton Russian bib CW0008 is below any limit, but Greenpeace flags pigment-decoration provenance as a contamination concern).
  • No nationally binding limit for tin or organotins in PVC children’s products at time of publication. Greenpeace notes the European Parliament (EP 2000) had called for substitution policies for soft PVC and phase-out of lead and cadmium stabilisers; organotin regulation was anticipated.

Methods (brief)

Sample collection. 54 PVC and PVC-suspect consumer products purchased in 20 countries between (date not specified in the source). Sample numbering convention: CW prefix + four-digit sample number (CW0005–CW0160; not consecutive).

Chlorine screening (Beilstein test or equivalent). Used to verify PVC vs non-PVC identity. 45 of 56 analytical units screened positive.

Phthalates, bisphenol A, cadmium, lead, tin (STAT Analysis, Chicago, IL, USA). Five phthalate esters (DEHP, BBP, DBP, DEP, DINP), bisphenol A, and three metals (Cd, Pb, Sn) measured. Phthalate LOD 100 mg/kg; BPA LOD 100 mg/kg; Cd LOD 0.500 mg/kg; Pb LOD 0.500 mg/kg; Sn LOD 2.50 mg/kg. The source notes “Analytical methods are described elsewhere (Di Gangi 1997, Harmon 2001)” without reproducing the methods locally.

Organotin speciation (GALAB, Geesthacht, Germany). 14 samples were analysed for eight organotin compounds — monobutyltin, dibutyltin, tributyltin, tetrabutyltin, monooctyltin, dioctyltin, tricyclohexyltin, triphenyltin — per the DIN EN 17025 analytical method. LODs were 0.3 µg/kg (per species) for the butyltin/cyclohexyltin/phenyltin series and 0.4 µg/kg for the octyltin series.

Study design. Cross-sectional consumer-product survey; one specimen per item; no replication or accelerated-aging sub-study (contrast with digangi1997-greenpeace-vinyl-children-products which included Pb extractability and four-week UV-A weathering sub-studies). No surface-wipe or migration testing reported in this 2001 report; total-substrate-content data only.

Implications

  • Regulatory baseline (EU PVC additives, pre-REACH). Stringer 2001 documents that as of 2001, two named PVC consumer products in commerce exceeded the EU EEC 1991 100 mg/kg cadmium limit for PVC (drawer liner, waterproof pants) and that organotin loadings into the 10⁴ µg/kg range were present in PVC consumer items including floor tiles, wallpapers, changing mats, and a non-PVC plastic-laminated cotton bib. The 2007 REACH Annex XVII restrictions on cadmium (entry 23) and organotins (entry 20) reflect the regulatory direction this Greenpeace report and its companions advocated. By the 2008 CPSIA Section 101 children’s-product lead substrate limit (currently 100 mg/kg), several of the Stringer 2001 lead-positive items (e.g., Chilean wallpaper 441 mg/kg, US plastic-laminated bib 5,220 mg/kg, Japanese rain hat 179 mg/kg, Philippines medicine dispenser 248 mg/kg) would exceed today’s federal US children’s-product Pb substrate ceiling.
  • HMTc audit implications (Part 2 firewall observed). The dataset adds 2001-vintage occurrence anchors for Pb, Cd, and Sn in PVC consumer products across 20 countries — extending the Di Gangi 1997 US-only window and providing the first NGO-laboratory organotin speciation in PVC children’s items. For HMTc certification of vinyl/PVC categories under the wiki’s product taxonomy (toys-substrate-materials, infant-clothing, bibs, diapers-and-components, pacifiers-and-sucking-teething-aids, etc.), the Stringer 2001 data are the legacy-baseline against which post-REACH and post-CPSIA progress is measured. This page does not propose HMTc thresholds.
  • App. Not directly relevant to ingredient contamination_profile data because no food-matrix occurrence values are reported (the maize syrup itself in CW0006a was below LOD for all parameters; the PVC bottle CW0006b carried the contamination). Potentially relevant to a future children’s-article-screening surface that uses material composition (PVC) and product category as a Pb/Cd/Sn/organotin contamination-likelihood predictor.
  • Courses. Useful primary document for the regulatory-history teaching module on the EU PVC Green Paper era (2000), the 1999 EC mouthing-articles phthalate ban, and the EEC 1991 cadmium-in-PVC restriction. Also a case study in NGO multi-country sampling design and the methodological argument for organotin speciation alongside total-tin determination.
  • Cross-reference. Stringer 2001 is the multi-country, broader-analyte-panel sibling of digangi1997-greenpeace-vinyl-children-products (US/Canada, Pb+Cd only, with extraction and aging sub-studies). The CPSC federal regulator response that took up the Di Gangi 1997 data (cpsc1997-pvc-children-products) predates Stringer 2001 and does not address the 2001 dataset specifically; no equivalent federal/EU regulatory-response document has been ingested.

Wiki pages this source may touch

Verification notes

  • Source identification. Stringer, R., Johnston, P. & Erry, B. “Toxic chemicals in a child’s world: an investigation into PVC plastic products.” Greenpeace Research Laboratories, University of Exeter, Prince of Wales Road, Exeter EX4 4PS, UK. June 2001. 22 pages. No journal venue; published as a Greenpeace technical report. The cover identifies Greenpeace as the publisher; the title page lists the three authors and the Exeter address.
  • DOI. None assigned. The report is grey literature published directly by Greenpeace Research Laboratories. No DataCite, Crossref, or ERIC identifier observed in the PDF. access_url left null pending verification against Greenpeace publication archives.
  • Author attribution. Ruth Stringer, Paul Johnston, and Bea Erry are all named on the title page. Author affiliation is Greenpeace Research Laboratories, University of Exeter. Stringer and Johnston also co-authored the cited Stringer & Johnston 2001 chlorine-industry overview (Kluwer Academic Publishers, ISBN 0-7923-6797-9). Stringer is also an author of the Stringer et al. 2000 children’s-toy phthalate study (Environmental Science and Pollution Research 7(1): 27–36), which is referenced but not in the wiki corpus.
  • Source-tier rationale. evidence_tier: B: per Part 13 grading, this is an NGO/grey-literature report rather than a peer-reviewed publication, but it carries (a) external accredited laboratory analysis at STAT Analysis (Chicago, IL, USA) for phthalates/BPA/Cd/Pb/Sn, (b) DIN EN 17025-compliant external accredited laboratory analysis at GALAB (Geesthacht, Germany) for eight organotin species, (c) explicit limits of detection per analyte, (d) consistent sample numbering and product identification across five data tables. B-tier (“good-quality observational studies, smaller cohorts, well-conducted government technical reports, regional regulator monitoring data”) is the appropriate fit for the same reason as digangi1997-greenpeace-vinyl-children-products (NGO authorship and advocacy framing are the quality-discounting factors; the laboratory data themselves are quantitatively defensible).
  • Source-type rationale. source_type: ngo-report: Greenpeace is an environmental advocacy NGO, not a peer-reviewed journal, government agency, or industry body. The Conclusions section makes explicit policy advocacy (“the only way to protect children from these harmful chemicals is to eliminate them from the products”), confirming the advocacy framing. The underlying laboratory data are nonetheless presented in a structured testing-report format compatible with the Cochrane corpus.
  • License rationale. copyrighted-third-party: Greenpeace Research Laboratories holds copyright. This page reproduces the report’s numerical findings under fair-use scholarly excerpting; the underlying source is not in the public domain.
  • Frontmatter products: field. Nine slugs selected from the 2026-05-18 taxonomy snapshot, all reflecting direct sample classes in Tables 1–5:
    • toys-substrate-materials (primary route — the report’s central methodological frame is total-content additive analysis of the PVC substrate of consumer articles across product types);
    • toys-squeeze (CW0033 Thai squeeze toy; CW0143 Japanese squeeze ball);
    • pacifiers-and-sucking-teething-aids (four teethers: CW0007 Russia non-PVC, CW0109 USA crib-rail teether, CW0127 USA teether, CW0141 Japan teether — the BPA outlier);
    • bibs (five bibs: CW0008 Russia non-PVC cotton, CW0012 Czech non-PVC, CW0048 Canada, CW0095 China, CW0115 USA non-PVC plastic-laminated cotton — the dataset’s lead and organotin peak among non-PVC items);
    • infant-clothing (rain hats CW0125 USA and CW0142 Japan; raincoats CW0135 France and CW0125 USA; CW0124 USA shoes/sandals — children’s PVC garments);
    • diapers-and-components (four pairs of waterproof pants/diaper covers: CW0005 Mexico, CW0010 Czech, CW0055 New Zealand — exceeds EU EEC 1991 100 mg/kg Cd limit at 132 mg/kg, CW0128 USA);
    • strollers (CW0107 USA stroller cover; CW0065 Brazilian car sunshade, used with strollers and car seats);
    • high-chairs-and-booster-seats (CW0144 UK high-chair cushion, ≈10% phthalates and 111 mg/kg tin);
    • toothbrushes (CW0078 Philippines toothbrush, non-PVC).
    • Test categories with no matching slug in the current taxonomy: bath mats; floor tiles; carpet protectors; wallpapers; drawer liners; shower curtains; inflatable sofas; suckers; lady’s handbags; changing mats; mattress covers; mattress pads; nasal aspirators; medicine dispensers; maize-syrup PVC bottles; PVC spoons/spoon handles; coolers; place mats; pencil cases/school kits; inflatable swim rings; rubber ducks. Per Phase 1 frontmatter discipline I do not invent slugs for these; the test results are documented in the page body but are not routed by products: until matching slugs exist. This matches the discipline applied on digangi1997-greenpeace-vinyl-children-products for the same family of dataset.
  • Frontmatter ingredients: [] is correct — no food ingredients are involved. The CW0006a Mexican maize syrup tested below LOD for every parameter and is reported as a negative control for migration from the CW0006b PVC bottle.
  • Frontmatter matrices: [] is correct — no food-matrix concentration values reported. All Cd, Pb, Sn results are reported as mg/kg by weight of the product substrate; organotin results are µg/kg by weight of the product substrate.
  • Frontmatter metals: [Pb, Cd, Sn] is correct — Pb, Cd, and total Sn are the three elemental analytes measured at STAT Analysis. The eight-species organotin panel at GALAB is captured in the body and in the Wiki pages this source may touch section as [[metals/organotins]]; “organotins” is not in the HMI metals abbreviation vocabulary (Pb/Cd/iAs/tAs/MeHg/tHg/Ni/Al/Cr/Cr-VI/Sn/Sb/U per Part 14), so the page-slug listing under organotins is the routing path.
  • Frontmatter jurisdictions lists all 20 sampling-country codes (AT Austria, BR Brazil, CA Canada, CL Chile, CN China, CZ Czech Republic, DK Denmark, FR France, HU Hungary, IT Italy, JP Japan, MX Mexico, NL Netherlands, NZ New Zealand, PH Philippines, RU Russia, ES Spain, TH Thailand, GB United Kingdom, US United States) plus EU for the regulatory framing (EEC 1991 Cd-in-PVC limit; EC 1999/815 phthalate ban for mouthing articles).
  • Frontmatter near_duplicates points to digangi1997-greenpeace-vinyl-children-products (the direct predecessor Greenpeace lead-and-cadmium-in-PVC report) and cpsc1997-pvc-children-products (the CPSC federal regulator response to the Di Gangi 1997 dataset, which Stringer 2001 cites as CPSC 1997). The Stringer et al. 2000 children’s-toy phthalate paper is referenced but not in the corpus; flagged for future discovery sweep. The Allsopp Greenpeace 2001 PVC-flooring report (“Poison underfoot”) and the Harmon 2001 Greenpeace USA “This Vinyl House” report are also referenced but not in the corpus.
  • Brand-firewall (Part 12, strict reading locked 2026-05-17). No retailer, manufacturer, or licensed-character brand names appear in the Key numbers section of this page. Stringer 2001 uses anonymised CW#### sample numbers throughout the data tables and refers to items by generic product type plus country of purchase (e.g., “a drawer liner from the USA (CW0150)” rather than naming the retailer or manufacturer). This anonymisation discipline distinguishes Stringer 2001 from Di Gangi 1997 (which named manufacturers, retailers, and licensed characters explicitly because the report itself was framed as a Proposition 65 enforcement document) and means the Part 12 Exception 1 (regulatory-event subject) frame is not engaged here.
  • Wiki/HMTc firewall (Part 2). The Implications section flags Stringer 2001’s regulatory comparators (EEC 1991 100 mg/kg Cd in PVC; EC 1999/815 0.1% phthalate limit for mouthing articles; post-2007 REACH Annex XVII directions for Cd and organotins) without proposing HMTc thresholds. No HMTc certification level is endorsed or critiqued.
  • Speciation flag. N/A for Pb and Cd (AAS-equivalent of acid-digested PVC yields total elemental Pb/Cd). Sn is reported as total elemental tin; organotin speciation is reported separately and captured in the [[metals/organotins]] page link. The tin-mass-balance gap (organotin species sum < total tin in every sample) is a published finding of the source itself, not an extraction or interpretation by this page.
  • Sample-size field. sample_n: 54 reflects the 54 distinct consumer products listed. The 56 analytical-units figure (with CW0006 and CW0140 subdivided) is noted in the sample_population paragraph and the Key numbers framing.
  • Regulation-page mapping. The report’s primary regulatory comparators are EEC 1991 Council Directive 91/338 (100 mg/kg Cd in PVC), EEC 1989 Council Directive 89/677 (Pb in paints; not PVC), EC 1999/815 (0.1% phthalates in mouthing articles for children under 3 years), and EC 2000 (Green Paper on PVC). None of these four EU instruments are in the current wiki/regulations/ slug list per the 2026-05-18 taxonomy snapshot. CPSC 1996, 1997, and 1998 are cited but the only related corpus entry is cpsc1997-pvc-children-products (the source-page version of the CPSC staff response; not a regulation page). DIN EN 17025 is the analytical-method standard for the organotin work and would not warrant a regulation page even if available. Flagged for future Part 10 regulation-page authoring rather than created speculatively from this single source.
  • Date and unit conventions. All Cd, Pb, and Sn substrate concentrations are reported in mg/kg by weight of the consumer-product material, equivalent to ppm or µg/g. Organotin species and total organotins are reported in µg/kg (ppb) by weight. Phthalate and BPA concentrations are reported in mg/kg. Sample IDs use the format CW#### (four-digit Greenpeace sample numbers). The DIN EN 17025 analytical method dates to (or before) 2001; the source does not specify an exact year for the standard.
  • Internal-source data-integrity flag (Key numbers, Lead section). The source narrative on p.9 states “Among the chlorinated samples, lead concentrations ranged between 0.51 mg/kg in a pencil case from Denmark (CW0051) and 441mg/kg in Chilean wallpaper (CW0129).” Table 1, however, lists CW0051 (described in Table 1 as “school kit (pencil case analysed)”) with a lead value of 35.40 mg/kg, not 0.51 mg/kg. Two possibilities: (a) the narrative-text “0.51 mg/kg” is a typographical error for the table value 35.40 mg/kg; (b) the narrative is reporting a sub-component analysis (e.g., the pencil case alone within a school-kit assembly) that is not reflected in the Table 1 entry. Either way, the per-sample Table 1 reading is the more reliable analytical record; the 0.51 mg/kg narrative figure is noted here for cross-reference but not used as the dataset minimum. The lead-range bound 0.51 mg/kg → 441 mg/kg among chlorinated PVC samples is therefore presented in the Key numbers section as 35.40 mg/kg → 441 mg/kg, with the narrative discrepancy noted.
  • Folder context vs paper scope. The PDF lives under _extracted_infantdurable_02_Strollers_Walkers_Swings/02_Strollers_Walkers_Swings/ in the Kimi corruption-issue raw tree, but the paper itself is a broad-scope multi-country survey of 54 PVC consumer products and is not specifically about strollers, walkers, or swings. The folder name reflects the Kimi-agent’s batch-organisation scheme during the May 2021 corruption-issue rescue, not the paper’s content. Within the dataset, the closest analogues to the folder name are: CW0107 USA stroller cover (routed under strollers); CW0065 Brazilian car sunshade (routed under strollers as a stroller/car-seat companion item); CW0109 USA crib-rail teether (routed under pacifiers-and-sucking-teething-aids, since teethers attached to crib rails are functionally pacifier/teething-aid format). No “walker” item appears in the dataset.
  • Raw integrity. raw_sha256 = abf61a50676c85667a805341aefc2cc7c7fe4dc5929650d5381f66f9ed8e04d2 confirmed against the PDF at the raw path.
  • Audit subagent findings applied (2026-06-01). Fresh-context Agent subagent audit returned verdict REVISE with 1 ❌ and 2 ⚠️ findings; verified each against the PDF:
    • ❌ Applied: CW0050 Canadian floor tile total organotins value is 51,388 µg/kg in Table 3 and on the p.11 narrative (“at 51 388 ug/kg, was not only over 10 times that of the US tile, but also the highest recorded in this survey”), while the p.9 narrative reads “51 338 ug/kg in a floor tile from Canada (CW0050)“. The wiki page took the table value (51,388) but did not previously flag the source-internal discrepancy. Now flagged in the Key numbers organotin bullet; the 51,388 µg/kg figure is retained as the canonical analytical record because (a) it agrees with Table 3 and (b) it agrees with the p.11 narrative; the p.9 “51 338” figure is treated as a typographical transposition (last two digits swapped) consistent with the table.
    • ⚠️ Applied: The Cd section had a leftover drafting TODO “Spanish wallpaper (CW0104) tag (note: …; the Spanish wallpaper actually with 0.500 mg/kg is none — verify against table).” Table 3 unambiguously reports CW0104 Cd as <0.500 mg/kg (non-detect). Resolved by replacing the parenthetical question with “Spanish wallpaper (CW0104) Cd was below LOD at <0.500 mg/kg (Table 3).” Netherlands suckers (CW0098) Cd 0.869 mg/kg added to the listing for completeness.
    • ⚠️ Applied: Sn section contained a typo “Italian Mexican syrup-bottle (CW0006b) 28.5 mg/kg” — CW0006b is the Mexican maize-syrup bottle (the Italian item is CW0140 spoon and CW0140b is the soft handle, listed separately). Corrected to “Mexican maize-syrup bottle (CW0006b) 28.5 mg/kg.”
    • Subagent self-corrected its initially-flagged ❌ on the “predominant organotin species” tally (six dibutyltin / five dioctyltin / two tributyltin) after re-reading source p.9 — count is correct as on the page.
    • Checks 2 (slug vocabulary), 3 (speciation/methods), 4 (Part 12 brand firewall), and 5 (Part 2 wiki/HMTc firewall) all returned ✅ clean.

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
c1aef382026-06-02audit-queue: hamid2021-bacterial-plant-biostimulants-review → audited-promote