Determination of Lead and Cadmium in Synthetic and Natural Hair Dyes in Morocco Using Differential Pulse Polarography

Benabbes et al. 2021 — Lead and cadmium in 16 commercial hair dyes on the Moroccan market by differential pulse polarography

This open-access Portugaliae Electrochimica Acta study measured lead and cadmium in 16 hair dye products (13 synthetic direct dyes, 3 natural henna-based dyes) purchased from Moroccan markets. Samples were digested with HNO₃/H₂O₂ in a closed-vessel microwave system (Anton Paar Multiwave PRO) and analyzed by differential pulse polarography (DPP) on a Metrohm 797 VA Computrace with a dropping mercury electrode. The validated method had LODs of 0.080 ppm Pb and 0.43 ppm Cd. Across the 16 products the authors found Pb concentrations ranging from below LOD to 3617.02 ppm (median 1196.17 ppm) and Cd from below LOD to 459.57 ppm (mean 55.55 ppm); 10 of 16 samples exceeded an applicable Pb acceptance limit and 9 of 16 exceeded an applicable Cd acceptance limit when compared against Health Canada (10 ppm Pb / 3 ppm Cd) and the German federal cosmetic-impurity limits (20 ppm Pb / 5 ppm Cd). The highest Pb and Cd values both occurred in the same sample, a natural henna-based dark brown dye of UAE origin. Evidence tier is B: peer-reviewed open-access journal, validated analytical method with reported LOD/LOQ/linearity/recovery/precision, but n=1 per individual product (16 distinct products, not 16 replicates of one), brand and origin coding only (no compositional or pigment analysis to attribute the source of contamination), and no analytical-blank or certified-reference-material recovery values reported.

Key numbers

All concentrations are reported in ppm (mg/kg) in the as-sold hair-dye product, as printed in the source. The standard-deviation column in Table 3 reports ± 0.001 for every Pb value and ± 0.0001 for every Cd value — this is the method-precision SD (Table 2 reports precision as CV 0.72% for Pb and 0.93% for Cd at the calibration concentrations), not a between-replicate SD for the individual product sample. The source states each product was analyzed by standard addition with three additions per sample.

Analytical method validation (Table 2 of the source)

Analyte	Linearity (r)	Accuracy (recovery ± RSD)	Precision (CV %)	LOD (ppm)	LOQ (ppm)
Pb	0.992	100.33 ± 4.9 %	0.72	0.080	0.243
Cd	0.999	102.60 ± 0.87 %	0.93	0.43	1.31

Note: the abstract states the correlation coefficient ranged “from 0.992 to 0.999” and Table 2 shows r = 0.992 for Pb and r = 0.999 for Cd. The Results narrative inverts this attribution (“correlation coefficient values (r) of 0.992 and 0.999 for Cd and Pb, respectively”); Table 2’s per-analyte assignment is the authoritative reading and is what is transcribed here. The paper-internal inconsistency is flagged in Verification notes.

Per-product Pb and Cd values (Table 3 of the source, in ppm)

Products are reported by the authors with anonymized brand codes (Brand-1 through Brand-6) and shade descriptors only. Country of origin is the country listed by the source for each product. SHD = Synthetic Hair Dye; NHD = Natural Hair Dye (henna-based per the discussion).

Product code	Shade	Type	Origin	Pb (ppm)	Cd (ppm)	Retail price (€)
1a	Very light blonde	SHD	France	1324.20 ± 0.001	54.79 ± 0.0001	11.50
1b	Light ash blonde	SHD	France	1042.65 ± 0.001	56.87 ± 0.0001	11.50
1c	Brown	SHD	France	1106.19 ± 0.001	53.09 ± 0.0001	11.50
1d	Dark ash blonde	SHD	France	960.69 ± 0.001	52.40 ± 0.0001	11.50
2a	Light brown	SHD	USA	< LOD	113.20 ± 0.0001	8.30
2b	Dark blonde	SHD	USA	< LOD	105.26 ± 0.0001	10.65
3a	Medium intense violet brown	SHD	Italy	1490.19 ± 0.001	< LOD	12.16
3b	Light warm blonde	SHD	Italy	448.43 ± 0.001	53.81 ± 0.0001	12.16
3c	Very light blonde	SHD	Italy	1574.07 ± 0.001	55.55 ± 0.0001	12.16
4a	Brown	NHD	UAE	676.32 ± 0.001	< LOD	2
4b	Maroon	NHD	UAE	1743.55 ± 0.001	< LOD	2
4c	Dark brown	NHD	UAE	3617.02 ± 0.001	459.57 ± 0.0001	2
5a	Medium brown	SHD	Germany	1196.17 ± 0.001	57.41 ± 0.0001	6.79
5b	Light brown	SHD	Germany	1414.63 ± 0.001	< LOD	7.95
5c	Medium brown (natural)	SHD	USA	< LOD	< LOD	9.24
6	Brown	SHD	France	585.93 ± 0.001	46.87 ± 0.0001	6.89

Headline panel statistics

• Pb (n=16): median 1196.17 ppm [Q1 = 960.69; Q3 = 1490.19]; range LOD to 3617.02 ppm. Thirteen of 16 samples had Pb above LOD; the three < LOD samples were products 2a, 2b, and 5c. Of the 13 detected, 13 fell in the range 448.43 to 3617.02 ppm (the source describes this as “extremely high lead level”). Maximum: product 4c (natural henna-based dark brown, UAE origin), 3617.02 ppm.
• Cd (n=16): mean 55.55 ppm [Q1 = 53.54; Q3 = 81.33]; range LOD to 459.57 ppm. Eleven of 16 samples had Cd above LOD; the five < LOD samples were products 3a, 4a, 4b, 5b, and 5c. Maximum: product 4c (natural henna-based dark brown, UAE origin), 459.57 ppm — the same product that carries the panel-maximum Pb.
• Both Pb and Cd below LOD: only product 5c (medium brown natural shade, USA origin).
• Pearson correlation between metal content and retail price: r = −0.399 for Pb and r = −0.354 for Cd; the authors interpret this as higher-priced products tending to carry lower heavy-metal loads in this panel.

Regulatory benchmarks the source compares against (Discussion, p. 42)

The source explicitly compares its 16 measured values against four external regulatory or guidance benchmarks that the authors identify. These are the comparison values from the source’s own narrative; the regulatory citations are reproduced from the source and have not been independently re-verified here.

Authority	Benchmark Pb	Benchmark Cd	Source-reported context
Health Canada (heavy-metal-impurity guidance)	10 ppm	3 ppm	Cited as “technically avoidable when exceeded” thresholds; source ref 10.
Germany (federal cosmetic-impurity limits)	20 ppm	5 ppm	Cited as “technically avoidable above” thresholds for toothpastes and other cosmetics; source refs 11-12.
US FDA cosmetic guidance	10 ppm Pb (cosmetic lip products / externally applied)	Not set for hair dyes	Source ref 13; the source explicitly states the FDA has not set acceptable levels of Pb or Cd in hair dyes.
United Kingdom	”1% of lead is still admitted in hair dyes” (i.e., 10,000 ppm)	Not specified	Source ref 14; the discussion notes this as a markedly looser regional standard.

Per the source’s discussion: “ten products have rates outside the acceptance limit for Pb and nine for Cd.” The source does not specify which of the four benchmark values it counted against, but applying the strictest cited thresholds (Health Canada 10 ppm Pb / 3 ppm Cd) to Table 3, 13 of 16 samples exceed 10 ppm Pb (all 13 with detectable Pb) and 11 of 16 samples exceed 3 ppm Cd (all 11 with detectable Cd) — both higher than the source’s text-stated counts. The source’s “ten / nine” exceedance counts likely reference the German thresholds (20 ppm Pb / 5 ppm Cd), against which 13 of 16 samples still exceed Pb and 11 of 16 still exceed Cd; the discrepancy is left unresolved in the source. Flagged in Verification notes.

Methods

Sample selection. Sixteen commercial hair dye products were purchased from markets in Morocco (the source does not name a specific city beyond “Moroccan market”). Thirteen samples were classified by the authors as “direct (temporary or semi-permanent) dyes”; three samples were classified as natural henna-based dyes. Products spanned six brand families (anonymized as Brand-1 through Brand-6) from five countries of origin (France, USA, Italy, UAE, Germany). Sample collection dates and lot numbers are not reported. The authors do not document product purchase prices for compositional comparison purposes but include retail price per product in Table 3 for the correlation analysis.

Sample preparation and digestion. Closed-vessel microwave digestion. A 200 mg sample was placed in a Teflon vessel and reacted with 4 mL of 65% trace-metal-free nitric acid (HNO₃, suprapure grade) for 4 hours at room temperature. After that pre-digestion step, 1 mL of 30% hydrogen peroxide (H₂O₂) was added and the sealed vessels were heated in a Multiwave PRO microwave system (Anton Paar) at 230 °C for 15 minutes. After cooling, digests were transferred to sealed polypropylene tubes. A method blank was treated through the same protocol with each batch.

Instrumentation. Differential pulse polarography (DPP) on a 797 VA Computrace (Metrohm, Switzerland — the source says “Germany”; the manufacturer is Swiss-headquartered with a German subsidiary). The cell used a dropping mercury electrode (DME) working electrode, a platinum counter electrode, and an Ag/AgCl/(3 M KCl) reference electrode. The supporting electrolyte was 1 M KCl. Working volume: 20 mL distilled water + 1 mL supporting electrolyte + 100 µL of the acid-digested sample. Pure mercury (99.9%) for the DME and pure nitrogen (99.999%) for deoxygenation/purge (180 s pre-bubble, also between additions). Electrode drop time 0.4 s; potential scan −0.2 V to −0.7 V; pulse amplitude 50 mV.

Quantitation strategy. Standard addition with three sequential 100 µL additions of standard solution per sample. Calibration standards were prepared from 1000 mg/L Pb stock (Pb(NO₃)₂) and 1000 mg/L Cd stock (CdCl₂) in double-distilled water. Calibration ranges: 10–60 ppm for Pb; 40–640 ppm for Cd.

Method validation. Validation followed ICH guidelines (source ref 9). Parameters validated: linearity (r = 0.992 for Pb, 0.999 for Cd per Table 2); accuracy by spike-recovery on the sample matrix (100.33 ± 4.9% for Pb, 102.60 ± 0.87% for Cd); precision expressed as CV (0.72% for Pb, 0.93% for Cd); LOD = 3 σ/S (0.080 ppm Pb, 0.43 ppm Cd); LOQ = 10 σ/S (0.243 ppm Pb, 1.31 ppm Cd). Accuracy was determined by standard addition on the finished product spiked with a known amount of each metal because no analyte-free hair-dye matrix was available.

Method-quality reporting gaps. The paper does not report: (a) certified reference material recoveries (only spike-recovery on the product matrix); (b) method-blank values or procedural-blank contamination checks; (c) number of replicate digestions per product (the polarographic Table 3 SDs of ± 0.001 for every Pb value and ± 0.0001 for every Cd value clearly are not between-replicate SDs and instead appear to be the method-precision SDs reported in Table 2 applied uniformly across all sample rows); (d) source of the standards (e.g., NIST-traceable, single-element vs multi-element); (e) instrument intermediate precision across the three calibration days; (f) drift correction protocol across the 16-sample run sequence. The use of a dropping mercury electrode rather than the modern hanging-mercury-drop or solid-electrode equivalents is unusual for a 2021 cosmetic-metals study but is internally consistent with the validation data the authors report.

Implications

Certification (HMTc). The paper contributes occurrence data for two HMTc-relevant adult personal-care product rows: rinse-off / direct / semi-permanent hair dyes and permanent oxidative hair dyes (the source’s “direct dyes” classification suggests the bulk of samples fall in the rinse-off / semi-permanent category, but the brand-and-shade descriptors are consistent with the kinds of products that other surveys classify as permanent oxidative dyes, so allocation between rows may need to be revisited at the per-product level once the Cat 2 / cosmetic taxonomy lock distinguishes by dye chemistry rather than by source-provided label). With n=1 per product (16 products, not 16 replicates), the standalone weight of any single Pb or Cd value is limited; the paper is best treated as one regional contributor (Morocco market, 2021) that pools with other hair-dye occurrence studies for per-row marginal distributions. The natural henna-based dyes from UAE origin (products 4a-4c) carry the panel-maximum Pb and Cd, consistent with the source’s text observation that the highest values were in a natural henna-based dye; per the source’s discussion this echoes prior findings by Ozbek et al. (2016) and Jallad et al. (2008) of higher Pb in natural-henna versus synthetic hair dyes.

Courses. Useful as a worked example of differential-pulse-polarography validation under ICH guidelines on a personal-care matrix, and as a teachable case study of regulatory benchmark heterogeneity (Health Canada 10 / 3 ppm vs Germany 20 / 5 ppm vs UK 1% Pb vs FDA “no acceptable level set for hair dyes”) for regulatory-affairs courses. The negative price-versus-contamination correlation (r ≈ −0.40 for Pb, −0.35 for Cd) is a useful but underpowered observation at n=16 to discuss in supply-chain or QA courses; the authors offer two non-mutually-exclusive hypotheses (higher-priced producers invest more in QC; production technology differs between price tiers).

App. Relevant for the consumer-facing app’s adult personal-care scoring layer once that layer covers hair-dye products. The source reports per-product values keyed to anonymized brand codes; the values feed product-row occurrence distributions, not brand-level scoring (consistent with Part 12 brand firewall).

Microbiome. Not addressed in the paper.

Wiki pages this source may touch

• lead
• cadmium
• hair-dye-rinse-off
• permanent-hair-dyes-tints

Verification notes

• 2026-05-29 fresh ingest (Claude Opus 4.7, autonomous v2.0 manual-fetch skill). New page; no prior wiki revision. DOI grep (10.4152/pea.202101037), raw_handle grep (MFK_lead-and-cadmium-in-synthetic-food-colors-and-hair), and benabbes2021 cite-key grep against wiki/sources/ all returned zero matches before ingest.
• Folder placement is misleading. The source PDF is filed under raw/Manual Fetch Kimi /condiment_papers/05_PB_Vanilla_Spices/ despite being a hair-dye / cosmetics paper, not a spices / condiments paper. The Kimi-generated filename (“Lead and Cadmium in Synthetic Food Colors and Hair Dyes.pdf”) additionally describes the paper as covering food colors, but the paper’s actual title and content are about synthetic and natural hair dyes only — no food colors are measured. Frontmatter products: and matrices: reflect the actual paper content (hair dyes), not the folder placement.
• Brand-firewall compliance (Part 12 strict, 2026-05-17 lock). The source anonymizes the tested brands as “Brand-1” through “Brand-6” with shade descriptors only; no commercial brand names are printed in the source’s Table 1 or Table 3. No brand attribution had to be stripped from this ingest. Product codes (e.g., “1a”, “4c”) are the source’s own anonymization codes, not wiki-side brand identifiers.
• Wiki/HMTc firewall (Part 2). No synthesis-across-papers claims, no HMTc threshold proposals, no consumer risk advisories appear in this page. The source’s own discussion references to prior literature (Ozbek 2016 natural-henna-vs-synthetic, Jallad 2008 henna paint-on-tattooing, Hussein 2015 Iraq hair dyes, Sani 2016 Nigeria cosmetics) are reported as attributions to the source’s own narrative, not as cross-source wiki synthesis.
• Paper-internal r-value attribution inconsistency. Table 2 reports r = 0.992 for Pb and r = 0.999 for Cd. The Results narrative (“Method validation” subsection, p. 40) states “correlation coefficient values (r) of 0.992 and 0.999 for Cd and Pb, respectively” — which inverts Table 2. The abstract (“the method was linear with a correlation coefficient value (r) that ranged from 0.992 to 0.999”) is consistent with either reading. Table 2 is treated as authoritative in this transcription.
• Paper-internal exceedance-count inconsistency. The discussion states “ten products have rates outside the acceptance limit for Pb and nine for Cd” without specifying which regulatory threshold was applied. Tabulating Table 3 against the four benchmarks the source cites: 13 of 16 samples exceed 10 ppm Pb (Health Canada), 13 of 16 exceed 20 ppm Pb (Germany), 13 of 16 exceed FDA’s 10 ppm Pb guidance, and 0 of 16 exceed the UK’s 1% (10,000 ppm) Pb allowance. For Cd: 11 of 16 exceed 3 ppm (Health Canada) and 11 of 16 exceed 5 ppm (Germany). The source’s “ten / nine” counts do not match any of these cleanly; the closest interpretation might be that the authors used a different (unstated) acceptance reference, or that “ten” / “nine” refer to a subset (e.g., synthetic-only samples) rather than the full panel. Documented here for the synthesis pass; the wiki transcribes both the source’s text-stated counts and the threshold-by-threshold recount.
• Speciation. Differential pulse polarography on acid-digested samples measures total elemental Pb and Cd. Reported as Pb and Cd (total) per the speciation convention (Part 14). The source itself uses “Pb” and “Cd” without speciation qualifier, consistent with this interpretation.
• Method localization detail. The source identifies the 797 VA Computrace instrument as “(Metrohm, Germany).” Metrohm is headquartered in Switzerland (Herisau) with a major German subsidiary in Filderstadt; the source’s “Germany” likely refers to the German distributor. This is a minor source-internal locator note that does not affect the analytical interpretation; transcribed as printed elsewhere in the Methods section with a parenthetical correction in this verification note.
• Henna identification. The discussion identifies the natural dyes as henna-based (Lawsonia inermis), referencing prior literature on henna-driven lead contamination (refs 15-16). The source does not analyze the henna powder separately from the dye preparation. The three natural hair dye products are reported as finished products (4a, 4b, 4c) rather than as the henna ingredient. Wiki ingredients: is empty because the current taxonomy does not include a henna or lawsonia-inermis ingredient slug; this is a possible future-page proposal (with the literature threshold question deferred to Karen / the routing-unresolved log if a second henna-specific source lands).
• License. Portugaliae Electrochimica Acta operates as an open-access journal published by the Sociedade Portuguesa de Electroquímica; the specific Creative Commons license is not printed on the article PDF. Marked as license: open-access pending verification against the journal’s policy page.
• DOI verified. DOI 10.4152/pea.202101037 printed on the first page of the article header.
• Authors verified against PDF byline. M. Benabbes (corresponding, benabbesmajda@gmail.com), M. Alami Chentoufi, B. Mojemmi, H. Benzeid, H. A Touré, A. Cheikh, Y. Rahali, M. O.B. Idrissi, M. Draoui, M. Bouatia. Institutional affiliations: (a) Laboratory of Analytical Chemistry, Faculty of Medicine and Pharmacy, Mohammed V University, Rabat, Morocco; (b) Faculty of Pharmacy, Abulcasis University, Rabat, Morocco; (c) Team of Formulation and Quality Control of Health Products, Faculty of Medicine and Pharmacy, Mohammed V University, Rabat, Morocco.
• Publication / acceptance dates. Received 5 June 2018; accepted 5 August 2020; published in 2021 issue (Vol. 39, No. 1, pp. 37-44). The two-year review-to-acceptance gap is unusual but is reproduced as the source prints it.

Ingest log

• 2021 (publication): Portugaliae Electrochimica Acta 39(1):37-44. Received 2018-06-05, accepted 2020-08-05.
• 2026-05-29 fresh ingest (Claude Opus 4.7, autonomous v2.0 manual-fetch skill): NEW path. Three identity checks against wiki/sources/ (DOI grep, raw_handle grep, cite-key grep) returned zero matches. PDF read in full (8 pages: abstract, introduction, materials & methods, results including Tables 1-3, discussion, conclusion, references). Source page written, routing audit queued.

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
12d44c4	2026-05-29	audit: eaglessmith2016-freshwater-fish-mercury-western-us-canada [revised]