Chengappa et al. 2025 — Cu, Zn, tAs, Pb in 20 over-the-counter toothpastes by flame AAS, India
This Scientific Reports primary occurrence study quantifies four metals (copper, zinc, total arsenic, lead) by flame atomic absorption spectroscopy in twenty over-the-counter toothpastes purchased on Indian e-commerce platforms in 2024. All twenty samples contained measurable Cu, Zn, and As; five samples were below the lead reporting threshold. The highest single-product concentrations were Cu 0.0620 mg/L, Zn 7.4224 mg/L, As 0.5371 mg/L, and Pb 0.0544 mg/L. Comparing against the strictest jurisdiction the paper invokes — the EU SCCS recommendations of 0.1 mg/L As and 0.5 mg/L Pb in toothpaste — the paper concludes only one sample (the highest-As whitening toothpaste at 0.54 mg/L) exceeds the EU arsenic ceiling; the per-table enumeration shows three samples flagged as exceeding (two whitening and one kids’ toothpaste), and a stricter reading of the 0.1 mg/L limit against the per-sample As values would put roughly eleven of twenty above the ceiling. The paper-internal narrative/table/strict-reading mismatch is recorded in Verification notes. No samples exceed the cited US FDA, WHO, Health Canada, or India BIS 6356 limits, all of which are higher than the EU limit by a factor of 5–200 for arsenic and a factor of 10–40 for lead.
Key numbers
Sampling and aggregation: 20 toothpastes, each anonymised to a number; Table 2 of the source groups samples by toothpaste type with brand names disclosed. This page reports per-type aggregates only (no brand attribution per CLAUDE.md Part 12 strict reading). Concentrations are in mg/L of the digested toothpaste solution (1.0 g toothpaste digested in 5 mL HNO₃ + 50 mL deionised water, made up to 100 mL final volume).
Per-type metal concentration ranges (computed from Table 2, p. 5; ND = below the table’s reported LOD):
| Toothpaste type | n | Cu (mg/L) | Zn (mg/L) | tAs (mg/L) | Pb (mg/L) |
|---|---|---|---|---|---|
| Whitening | 2 | 0.009–0.016 | 0.060–0.98 | 0.24–0.54 | 0.029–0.035 |
| Herbal / ayurvedic / mint / lemon-mint | 8 | 0.0036–0.034 | 0.018–2.50 | ND–0.16 | ND in 3 of 8; non-ND 0.004–0.054 |
| Anti-bacterial / germ-fighting | 2 | 0.0032–0.062 | 0.042–2.30 | 0.10–0.18 | ND–0.023 |
| Kids | 1 | 0.006 | 0.034 | 0.27 | 0.030 |
| Anti-cavity | 1 | 0.038 | 0.045 | 0.16 | ND |
| Salt (active salt / natural salt) | 2 | 0.029 | 0.057–0.095 | 0.053–0.081 | 0.001–0.003 |
| Everyday-protection | 1 | 0.0056 | 0.032 | 0.16 | 0.019 |
| Calcium | 1 | 0.047 | 0.048 | 0.11 | ND |
| Cooling | 1 | 0.004 | 0.041 | 0.036 | 0.042 |
| Charcoal | 1 | 0.002 | 7.42 | 0.068 | 0.039 |
Pooled-sample marginal extremes (abstract and Results, pp. 1, 4; full-precision values):
- Cu maximum across all 20 samples: 0.0620 mg/L (anti-bacterial type).
- Zn maximum across all 20 samples: 7.4224 mg/L (charcoal type). Next-highest: 2.5002 mg/L (herbal type) and 1.8501 mg/L (herbal type).
- tAs maximum across all 20 samples: 0.5371 mg/L (whitening type). Next-highest: 0.2663 mg/L (kids type) and 0.2424 mg/L (whitening type). One sample (a herbal/ayurvedic/mint/lemon-mint type) was below the As LOD.
- Pb maximum across all 20 samples: 0.0544 mg/L (herbal type). Five samples were “below LOD” for Pb as reported in Table 2.
Limits of detection (LOD) reported by the source:
- Methods narrative (Results section, p. 4): As 0.05 mg/L, Pb 0.007 mg/L, Zn 0.003 mg/L, Cu 0.004 mg/L.
- Table 2 footnote (p. 5): As 0.001 mg/L, Pb 0.0005 mg/L, Zn 0.005 mg/L, Cu 0.002 mg/L.
- These two LOD sets are internally inconsistent; the Table 2 footnote values are consistent with the per-sample reported precision (some samples report e.g. Pb 0.001, 0.003, 0.004 mg/L which would be below the 0.007 narrative-stated LOD). See Verification notes.
Calibration characteristic concentrations (Methods, p. 4):
- As: linear-segmented; characteristic concentration 0.8011 mg/L; standards 10, 20, 30, 40, 50 mg/L.
- Pb: linear; characteristic concentration 0.1862 mg/L; standards 1, 3, 5, 7, 10 mg/L.
- Cu: linear; characteristic concentration 0.0764 mg/L; standards 1, 2, 3, 4, 5 mg/L.
- Zn: linear; characteristic concentration 0.0198 mg/L; standards 0.5, 1.0, 1.5 mg/L.
- All standards: RSD < 1.5%.
Regulatory comparison limits cited by the source (Introduction, pp. 2-3):
| Standard | tAs (mg/L) | Pb (mg/L) | Other metals (mg/L) |
|---|---|---|---|
| EU SCCS (EC 1223/2009; toothpaste) | 0.1 | 0.5 | ”complete exclusion; traces only” |
| US FDA (cosmetics) | 5 | 5 | 20 |
| WHO (cited for Pb) | — | 10 | — |
| Health Canada (cosmetics) | 5 | 20 | — |
| India BIS 6356 (toothpaste; the source cites both 2001 and 2017 BIS update years for this same specification) | 2 | not separately stated | 20 |
Exceedance counts the source reports (Results and Discussion, pp. 4-5):
- EU tAs (0.1 mg/L): the source’s narrative claims “one sample” exceeds; Table 2’s “comparison with regulatory limits” column flags three samples as exceeding (the two whitening samples 0.54 mg/L and 0.24 mg/L, and the kids’ sample 0.27 mg/L); a direct comparison of per-sample tAs values against 0.1 mg/L from Table 2 places ten of twenty strictly above 0.1 mg/L (eleven if “at or above 0.1” is used — Toothpaste 19 is exactly 0.10 mg/L). See Verification notes.
- EU Pb (0.5 mg/L): no samples exceed (maximum is 0.0544 mg/L, an order of magnitude below).
- US FDA Pb (5 mg/L), WHO Pb (10 mg/L), Health Canada Pb (20 mg/L), India BIS (As 2 mg/L, other metals 20 mg/L): no samples exceed any of these higher limits.
Per the source’s Discussion, the highest-tAs whitening sample (0.5371 mg/L) and the kids’-toothpaste (0.2663 mg/L) include natural plant extracts in their composition listing, which the authors flag as a plausible incoming-As route via incomplete purification of botanicals. The highest-Zn charcoal sample (7.4224 mg/L) is noted as not declaring zinc on its ingredient label, which the authors flag as a labelling-compliance concern. (Brand and ingredient lists for individual samples are in Table 1, p. 3 of the source; not reproduced here.)
Comparative concentration ranges from cited prior studies on toothpaste in other jurisdictions (Discussion, pp. 4-6) — listed here as the source reports them, not as endorsed values:
- Saudi Arabia (Salama 2015, n=4): As 0.0006–0.0269 mg/L; Pb 0.1856–0.6313 mg/L; Cu 0.5590–2.2988 mg/L.
- Saudi Arabia (Almukainzi et al. 2022, n=2): As 209.33–221.96 mg/L; Pb 75.86–78.31 mg/L; Cu 264–269.94 mg/L (these are 3-4 orders of magnitude higher than the present study; the source attributes the divergence to a different analytical platform — ICP-MS in those studies — and different digestion procedures).
- Bangladesh (Paul et al. 2019): As 0.027–0.637 mg/L; Pb 0.27–2.12 mg/L; Cu 2.78–13.1 mg/L.
- Nigeria (Orisakwe et al. 2016, n=35): Pb 4.514–23.575 mg/L.
- Nigeria (Odukudu et al. 2014): Pb 0.02 mg/L; Cu 0.23 mg/L; Zn 0.2555 mg/L.
- Malta (Vella & Attard 2019, n=9): Cu 0.73–3.68 mg/L; Zn 0.31–7.8 mg/L (with one sample at 2417 mg/L Zn).
- Iraq (Lawi et al. 2023, herbal toothpastes): Pb 1.00–12.05 mg/L; Zn 1.59–11.3 mg/L (with one sample at 402.34 mg/L Zn).
Methods (brief)
Toothpaste selection: 20 best-selling listings on Amazon.in, Flipkart.com, and Myntra.com (top three Indian e-commerce platforms by monthly visitors per Hendy 2024) by customer reviews, ratings, and online visibility. Sampling categories cover regular adult, whitening, herbal, and kids’ toothpastes. All samples stored away from direct sunlight under dry conditions and analysed within one week of purchase.
Digestion: 1.0 g toothpaste + 50 mL deionised water in a 100 mL beaker; 5 mL concentrated HNO₃ added; hotplate at 140 °C for 2 h under atmospheric pressure to a clear-mixture endpoint indicating complete digestion. Cool; centrifuge (rotor radius 10 cm, 5 min, 1368 ×g, room temperature); supernatant transferred to a 100 mL volumetric flask; diluted to 100 mL with deionised water. The digestion protocol is cited to Paul et al. 2019 (the Bangladesh comparator).
Quantification: Flame atomic absorption spectroscopy on a Thermo Scientific iCE 3000 Series AAS instrument. The paper does not separately report which lamp configurations or flame conditions were used per analyte.
Speciation: Total metals only. The paper reports “arsenic” without species separation; this page therefore records tAs per CLAUDE.md Part 14. Lead and copper and zinc are recorded as bare-element measurements.
Quality control: Reagent blanks (deionised water and nitric acid) analysed alongside samples; certified standard stock solutions used for calibration; glassware soaked in 10% HNO₃ overnight and rinsed with deionised water before use. Blinding: samples masked and numbered prior to analysis. No reference materials, recovery percentages, or spike-recovery values are reported. No replicate analyses (the authors explicitly note this absence as a limitation in the Limitations section, p. 6-7). No statistical analysis or uncertainty/confidence-interval reporting (the authors also note this absence as a limitation).
Limitations the authors state (Limitations, pp. 6-7): the four-metal scope omits other relevant cosmetic metals; convenience-sample selection from three e-commerce platforms by online popularity risks selection bias and may not represent the full Indian toothpaste market; calibration standard concentrations (1–50 mg/L depending on analyte) were higher than the trace-level concentrations observed in many samples, which the authors acknowledge can affect quantification accuracy at trace concentrations; no replicate analysis; no LOD validation against the test matrix; purely descriptive reporting without inferential statistics.
Implications
Certification (HMTc): Contributes Indian-market category-level occurrence data for the toothpaste (Cat 2 Row 11) evidence base. The dataset’s chief signal for threshold setting is on the upper-tail arsenic concentrations (eleven of twenty samples sit above the EU SCCS 0.1 mg/L As recommendation per a strict reading of Table 2; the source’s own narrative on the count is internally inconsistent, see Verification notes). Lead concentrations are all well below all cited regulatory limits, including the Washington State Toxic-Free Cosmetics Act 1000 ppb statutory limit when expressed on an equivalent basis (Pb max 0.0544 mg/L of digested 100 mL solution corresponds to roughly 5.44 µg per 1.0 g toothpaste, i.e. 5.44 ppm wet-weight, which exceeds 1000 ppb = 1 ppm; basis conversion from digested-solution mg/L to wet-weight ppm requires the digestion-dilution factor — recorded here as a flag for the synthesis pass, not asserted as authoritative). No threshold values are proposed on this page.
Courses: Useful as a regional-comparison teaching case for the Cat 2 toothpaste module, particularly the cross-jurisdictional regulatory spread (EU 0.1 mg/L As vs India BIS 2 mg/L As — a 20× gap on the same product category) and the in-jurisdiction analytical-platform sensitivity question (flame AAS vs ICP-MS produces concentration-range differences spanning 3-4 orders of magnitude across cited studies). Limited utility as a methods case study because no reference materials, no recovery data, no replicates, no statistical analysis, and no validated method-specific LODs are reported, and the paper carries an internal narrative-versus-table contradiction on exceedance counts.
App: Not directly applicable. The paper measures finished toothpaste products, which are outside the consumer-app scope (focussed on children’s foods and personal care).
Wiki pages this source may touch
- arsenic-total
- lead
- toothpaste
- eu-ec-1223-2009-cosmetics
- india-bis-6356-toothpaste
- washington-tfca-toothpaste-pb-1000ppb
Verification notes
- Brand firewall (Part 12 strict reading, 2026-05-17): the source’s Table 1 (p. 3) lists each of the 20 toothpastes with brand name and full ingredient declaration, and the source’s Table 2 (p. 5) attaches metal concentration values to brand names. This page aggregates exclusively to toothpaste-type level using the source’s own type column (whitening, herbal/ayurvedic/mint/lemon-mint, anti-bacterial/germ-fighting, kids, anti-cavity, salt, everyday-protection, calcium, cooling, charcoal). No brand names are reproduced. The source’s Table 1 brand identifiers are not reused for descriptive context anywhere on this page.
- Paper-internal contradiction on the EU arsenic exceedance count (recorded, not adjudicated):
- Abstract (p. 1) states: “the arsenic concentration of one sample exceeded the European Union (EU) standard.”
- Results body (p. 4) states: “the concentrations of these metals were within the permissible limits prescribed, except for toothpaste 1, which had an arsenic concentration of 0.53 mg/L, which exceeded the European Union standard.”
- Table 2 (p. 5) “comparison with regulatory limits” column flags three samples (the two whitening samples and the kids’ sample) as “Arsenic exceeds EU regulations.”
- Discussion (p. 5) restates the three-sample reading: “Two of the three samples with arsenic concentration exceeding EU limits were whitening toothpastes and the third sample was found to be a children’s toothpaste.”
- A direct strict-reading comparison of each per-sample As value against the EU SCCS 0.1 mg/L cited cap (Table 2 per-sample values) places ten of twenty samples strictly above 0.1 mg/L, with an eleventh sample (Toothpaste 19) exactly at 0.10 mg/L (eleven samples at “≥0.1”). This page reports all four readings (abstract “one”, Table 2 flagged “three”, strict “ten above / eleven at-or-above”) rather than asserting a single canonical count. The source’s “three flagged” interpretation appears to compare against an unstated higher threshold (perhaps 0.2 mg/L or “relatively high” cut-off) but this is not declared in the methods.
- Paper-internal contradiction on lead LOD (recorded, not adjudicated): Methods narrative (p. 4) reports Pb LOD 0.007 mg/L; Table 2 footnote (p. 5) reports Pb LOD 0.0005 mg/L. Per-sample Pb values include 0.001 and 0.003 mg/L reported as concrete (not below-LOD) values, which is consistent with the 0.0005 mg/L footnote LOD and inconsistent with the 0.007 mg/L narrative LOD. The “below LOD” notation in Table 2 is preserved as the source reports it; users of this dataset should treat “below LOD” entries cautiously given the dual-LOD problem.
- Paper-internal contradiction on India BIS year (recorded, not adjudicated): the source cites BIS 6356:2001 in the regulatory-limits paragraph (Introduction, p. 3) and “IS 6356-2021” in the reference list (ref 3); the BIS URL in the reference list points to an unspecified-year BIS page. This page records both citations and routes to a single india-bis-6356-toothpaste page (to be created via Karen’s Step 0 Lock workflow for regulation pages).
- Evidence tier: A-tier per CLAUDE.md Part 13 (peer-reviewed primary occurrence study, Scientific Reports), with the methodological caveats noted in Methods (brief) above — no replicates, no reference materials, no recovery data, calibration-standard mismatch with sample concentrations, no statistical analysis. The A-tier rating reflects the publication venue and peer review; downstream synthesis should weight this source against the methodological reservations.
- The source’s own systematic review (Chengappa et al. 2025, Frontiers in Dental Medicine 6:1543972, ref 30) is a separate paper that motivated the present study; it is not ingested here.
- Empty
ingredients: []andmatrices: [cosmetic-personal-care]are intentional: the paper measures metals as contaminants in a finished cosmetic product (toothpaste), not in a food-style ingredient. Thecosmetic-personal-carematrices vocabulary follows the established convention across sibling source pages (jitareanu2025, rbeida2023, balogun2024, massarsky2025, arshad2020). - The paper folder location (
babycare_04_Shampoo_Wipes_Sunscreen) is a manual-fetch curation choice; the paper itself covers OTC toothpastes broadly with one kids’-toothpaste sample among twenty. Routing to toothpaste (Cat 2) is appropriate; per-paper data supports both general toothpaste and the Cat 2 kids’-toothpaste row. - Audit subagent (2026-05-18) flagged the strict-reading As exceedance count as “approximately eleven” — verified against Table 2: exactly ten samples are strictly above 0.1 mg/L, with an eleventh sample at exactly 0.10 mg/L. Tightened to “ten of twenty strictly above 0.1 mg/L (eleven if at-or-above)” in Key numbers and in the contradiction-notes entry above.
- Audit subagent (2026-05-18) flagged the parenthetical attribution of the ND-As sample as confused (“anti-cavity/calcium type — actually the herbal-mint type”) — verified against Table 2: Toothpaste 16 (herbal/ayurvedic/mint/lemon-mint group) is the ND-As sample. Removed the self-correcting parenthetical; cleaned to “a herbal/ayurvedic/mint/lemon-mint type” in Key numbers.
- Audit subagent (2026-05-18) flagged the herbal-group Pb range “ND–0.054” as misleading because the non-ND floor is 0.004 mg/L (Toothpaste 9), not approaching ND — verified against Table 2: three of eight herbal samples are below the Pb LOD (Toothpastes 5, 13, 15) and five report concrete Pb values 0.004, 0.054, 0.024, 0.040, 0.009 mg/L. Corrected the per-type table cell to “ND in 3 of 8; non-ND 0.004–0.054”.
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.