Sadeghi et al. 2021 — Heavy metals in tahini from the Tehran retail market (DPASV)

Sadeghi and coauthors surveyed cadmium, lead, copper, and zinc concentrations in 37 commercial tahini brands (n=74 duplicate samples) purchased from Tehran retail markets, using differential-pulse anodic-stripping voltammetry (DPASV) on a 746 VA trace analyzer with a hanging mercury drop electrode. Across all 37 brands, mean concentrations were 0.015 ± 0.01 µg/100g for lead and 0.012 ± 0.007 µg/100g for cadmium — equivalent to 0.15 ppb Pb and 0.12 ppb Cd on a per-kg basis, both well below the Codex CXS 193-1995 maximum level of 0.01 mg/kg (10 ppb) for tahini. Mean zinc was 3.79 ± 2.43 µg/100g and mean copper was 1.48 ± 0.08 µg/100g (the abstract conclusion transposes the Cd and Cu mean values; see Verification notes for the cross-check against Table 1 and Figure 1). The authors conclude that Tehran-market tahini is a high-value food meeting nutritional needs without toxic concern from lead or cadmium at the concentrations they measured, while recommending continued attention to soil-uptake risks in sesame cultivation given the crop’s known sensitivity to soil contamination.

Key numbers

Brand-level concentrations of Zn, Cu, Pb, and Cd in 32 unflavored (“simple”) and 5 flavored tahini brands (µg/100g, mean ± SD across triplicate experiments per brand; Table 1, pp. 42–43):

Selected per-brand extremes from Table 1:

• Highest Zn: brand 14 at 7.99 ± 0.10 µg/100g (unflavored).
• Lowest Zn: brand 9 at 0.27 ± 0.17 µg/100g (unflavored).
• Highest Cu: brand 12 at 2.79 ± 1.09 µg/100g (unflavored).
• Lowest Cu: brand 6 at 0.16 ± 0.02 µg/100g (unflavored).
• Highest Pb: brand 4 at 0.048 ± 0.004 µg/100g (unflavored).
• Lowest Pb: brand 15 at 0.001 ± 0.001 µg/100g (unflavored).
• Highest Cd: brand 23 at 0.028 ± 0.007 µg/100g (unflavored).
• Lowest Cd: brand 2 at 0.003 ± 0.001 µg/100g (unflavored).

Unit conversions for cross-paper comparison (1 µg/100g = 10 µg/kg = 10 ppb on a wet/as-sold basis):

• Mean Pb across 37 brands: 0.015 µg/100g = 0.15 ppb.
• Mean Cd across 37 brands: 0.012 µg/100g = 0.12 ppb.
• Mean Zn across 37 brands: 3.79 µg/100g = 37.9 ppb.
• Mean Cu across 37 brands: 1.48 µg/100g = 14.8 ppb.
• Maximum Pb across 37 brands: 0.048 µg/100g = 0.48 ppb (brand 4).
• Maximum Cd across 37 brands: 0.028 µg/100g = 0.28 ppb (brand 23).

The basis (wet vs dry) is not explicitly stated; sample preparation involved drying followed by 450 °C muffle-furnace ashing and acid redissolution, but the reported concentrations are expressed per gram of original sample mass (2 g weighed before drying per Section 2.4), so values are read as as-sold/wet weight here. Tahini’s intrinsic moisture content is low (~1–2 %), so the wet/dry distinction is small for this matrix.

Regulatory anchors cited by the authors (Section 3, pp. 42–43):

• Codex CXS 193-1995 maximum level for Zn in tahini: 4 mg/kg (= 400 µg/100g). All 37 brands well below (overall mean 3.79 µg/100g; maximum 7.99 µg/100g = 0.0799 mg/kg, P < 0.001).
• Codex CXS 193-1995 maximum level for Cu in tahini: 2 mg/kg (= 200 µg/100g). All 37 brands well below (overall mean 1.48 µg/100g; maximum 2.79 µg/100g = 0.0279 mg/kg, P < 0.001).
• Codex CXS 193-1995 maximum level for Pb and Cd in tahini: 0.01 mg/kg (= 1 µg/100g = 10 ppb). All 37 brands below (maximum Pb 0.048 µg/100g = 0.48 ppb; maximum Cd 0.028 µg/100g = 0.28 ppb), with P < 0.001.

Subgroup contrasts (Section 3, p. 43): Zn, Cd, and Pb showed no significant difference between simple and flavored brands (P = 0.802 for Zn, P = 0.954 for Cd, P = 0.64 for Pb). Cu was significantly higher in unflavored (“simple”) brands than in flavored brands (P = 0.022) — the authors attribute this to dilution by added flavoring matrices in the flavored subset rather than to a contamination-source difference. Brand-name was not significantly associated with element concentration (P > 0.05 for all four elements).

Methods (brief)

Tahini samples were collected from local retail markets in Tehran, Iran. In total, 37 commercial brands were purchased in duplicate (n = 74), comprising 32 unflavored (“simple”) and 5 flavored brands (flavored variants prepared by addition of coffee, chocolate, date, vanilla, and cocoa). Sample size was justified using Eq. 1 with σ = 2.02, α = 0.05, β = 0.2, µ₀ = 4.61, µ₁ = 3.79.

Sample preparation (Section 2.4): for each analysis, 2 g of sample was weighed, heated until drying, cooled to ambient temperature, treated with 10 mL of 65 % w/v HNO₃ until a clear solution was obtained, heated again, then transferred to a muffle furnace at 450 °C for 24 h. The cooled ash was washed with 65 % w/v HNO₃ and filtered through Whatman paper no. 1; the final solution was made up to 25 mL with HNO₃ and used for analysis.

Instrumental analysis (Section 2.5): differential-pulse anodic-stripping voltammetry (DPASV) on a Metrohm 746 VA Trace Analyzer (Metrohm, Switzerland) equipped with a three-electrode cell — hanging mercury drop electrode (HMDE) as working, axillary platinum electrode as auxiliary, Ag/AgCl as reference. Deposition potential −1.20 V; deposition time 30–90 s; scan rate 6 mV/s; pulse amplitude 50 mV; final potential +0.3 V. Reagents (Section 2.3): tartaric acid, CH₃COONa, Pb(NO₃)₂, Cd(NO₃)₂, Cu(NO₃)₂, Zn(NO₃)₂, and HNO₃ were analytical grade (Merck, Germany). Stock HNO₃ 1 g/L 0.005 M. Supporting electrolyte: 0.2 M acetate buffer at pH 4.7 containing 0.2 M tartaric acid. Calibration via standard-addition method — 500 µL sample added to the polarographic cell with 10 mL supporting electrolyte recorded as blank, then 100 µL of 1 g/L Zn/Cu/Pb/Cd standard added and the polarogram recorded. Standards prepared at 1 g/L for Zn, Cu, Pb, and Cd at +0.3 V final potential.

Statistical analysis (Section 2.6): SPSS v.16, one-way ANOVA at p = 0.05, all experiments performed in triplicate.

Limitations stated or evident: (1) no certified reference material recovery, LOD, or LOQ figures are reported (Sections 2.5 and 2.6); (2) DPASV with HMDE is sensitive for trace Pb, Cd, Cu, and Zn but is not the modern gold standard (ICP-MS) used in the comparable tahini characterisation of Potortì et al. 2026; (3) the abstract conclusion text reports Cd and Cu mean values transposed relative to Table 1 (see Verification notes); (4) no Hg, Ni, As, or other HMI-tracked analytes were measured; (5) the basis (wet vs dry) of the per-gram concentrations is not explicitly stated, though the procedure indicates per-2 g-original-sample reporting; (6) single-city, single-purchase-round design limits temporal and supply-chain generalisability.

Implications

Certification: in this 37-brand Tehran-market dataset, Pb and Cd in tahini sit roughly 20–80× below the Codex CXS 193-1995 maximum level of 0.01 mg/kg (10 ppb) — mean Pb 0.15 ppb, mean Cd 0.12 ppb, maximum Pb 0.48 ppb, maximum Cd 0.28 ppb. These are roughly two to three orders of magnitude lower than the Pb and Cd values reported by Potortì et al. 2026 in seven Italian-market tahini brands (Potortì Pb LOQ < 0.01 mg/kg in six brands but 0.50 mg/kg in one Turkey-origin brand; Cd LOQ < 0.01 mg/kg in five brands but 0.35 mg/kg in T-1). The two papers therefore disagree on the upper tail of the tahini contamination distribution by orders of magnitude, with no obvious cross-paper bridge for the methodology, basis convention, or sampling round. The paper itself supports no HMT&C threshold tightening beyond Codex 0.01 mg/kg.

Courses: a teaching example of why method choice and reference-material reporting matter for cross-paper comparison. The DPASV/HMDE method here gives values 100–1000× lower than ICP-MS values for the same matrix in Potortì 2026, and the absence of CRM recovery or LOD/LOQ reporting in the Sadeghi methods section prevents a direct method-bias adjudication. Also a useful illustration of paper-internal transcription error (the Cd/Cu transposition in the abstract conclusion was caught only by cross-checking Table 1 and Figure 1).

App: occurrence data for Pb (0.001–0.048 µg/100g across 37 brands), Cd (0.003–0.028 µg/100g across 37 brands), Cu (0.16–2.79 µg/100g), and Zn (0.27–7.99 µg/100g) in tahini from the Tehran retail market in 2021. For category occurrence modelling, hold these values separately from the Italian-market Potortì 2026 distribution rather than pooling — the order-of-magnitude offset is not reconcilable from the published methods alone.

Wiki pages updated on ingest

• lead
• cadmium
• sesame
• tahini
• nut-seed-butters-other
• codex-cxs-193-1995-tin-canned-foods

Verification notes

Source written 2026-06-02 from raw/manual-fetch/Kimi_Agent_Download Corruption Issue/condiments2_papers/03_Condiments/06_Sadeghi_tahini_heavy_metals_2021.pdf (SHA256 26c2cc6e27c2bfeb1ce1c5a804495ad34e122c67714081fc41ad1147d3106e0b). DOI 10.30502/jhhhm.2021.288521.1034 verified from the journal masthead on page 1. License is CC BY-NC 4.0 per the masthead statement.

Paper-internal Cd/Cu abstract transposition: the abstract conclusion (p. 40) states “Concentration of Zn, Cd, Pb, and Cu was calculated as 3.79 ±2.43, 1.48 ±0.08, 0.015 ±0.01, and 0.012 ±0.007 µg/100 g, respectively.” Cross-check against Table 1 (pp. 42–43) and Figure 1 (p. 43): Table 1 Cd values range from 0.003 to 0.028 µg/100g across all 37 brands, with a typical sample around 0.01 µg/100g, and Table 1 Cu values range from 0.16 to 2.79 µg/100g with most samples between 0.5 and 2.5 µg/100g. Figure 1’s bar chart shows simple-Cu ~1.5, flavored-Cu ~0.6, with Pb and Cd both visually near zero. The mean values 1.48 ± 0.08 and 0.012 ± 0.007 therefore correspond to Cu and Cd respectively, not Cd and Cu as the abstract states. The Discussion text on p. 43 — “but higher amount of Cu in non-flavored samples was detected compared to flavored samples (P = 0.022)” — also confirms Cu is the ~1.5 µg/100g element. This wiki page reports the correct attribution (Cu mean 1.48, Cd mean 0.012) in the Key numbers and Implications sections, and notes the abstract transposition explicitly in the tier_rationale and here so downstream synthesis or app ingest does not propagate the abstract’s mis-labelled values.

Basis convention: the paper reports concentrations per 2 g of original (pre-drying) sample mass, so values are read as as-sold/wet weight here. Tahini’s intrinsic moisture content is low (~1–2 %), so the wet/dry distinction is minor for this matrix. No certified reference material recovery, LOD, or LOQ figures are reported in Sections 2.5 or 2.6.

Cross-paper comparison: Sadeghi 2021 Pb and Cd values are roughly two to three orders of magnitude below those reported by Potortì et al. 2026 in seven Italian-market tahini brands. Both papers cite EU/Codex limits in the same mg/kg basis and both report Pb and Cd below those limits for the bulk of their samples (Potortì’s T-1 exceeds both limits substantially; Sadeghi reports no exceedances). No paper-side methodological bridge resolves the order-of-magnitude gap; the wiki carries both as separate distributions and flags the discrepancy for any future synthesis pass.

Audit subagent (2026-06-02) flagged the Implications/Certification subsection’s category-level inferences — specifically the phrases “favours treating tahini as a generally low-Pb/Cd commodity in well-sourced markets” and “the Potortì high outlier (T-1) confirms that origin/supplier screening matters” — as drifting from source-page reporting into Part 9 synthesis / Part 2 policy territory. Verified: the audit’s reading is correct; those sentences extrapolate from a single survey + a cross-paper comparison to a category-level supplier-screening lesson, which is the synthesis workflow’s job, not a source page’s. Revised the paragraph to drop the generalising sentences and keep only the cross-paper offset observation and the no-HMT&C-tightening boundary statement.

Brand firewall: per Part 12, no brand names are disclosed in the source — samples are coded 1–37 in Table 1. No Part 12 redaction was needed. The brand-by-brand range and extremes are reported here as code numbers only.

Ingredient slug note: [[ingredients/sesame]] and [[ingredients/tahini]] are declared as routing destinations matching the precedent set by Potortì et al. 2026. Neither page currently exists in wiki/ingredients/; both are tolerated by the routing audit and will be auto-stubbed when ingredient frequency crosses the ≥freq-2 threshold (this is the second tahini-specific paper to declare them, so they should auto-stub on the next create-missing-pages.mjs run).

Regulation slug note: the paper cites “Codex” maximum levels for Zn (4 mg/kg), Cu (2 mg/kg), and Pb/Cd (both 0.01 mg/kg) in tahini. The closest existing regulation page in the current taxonomy is codex-cxs-193-1995-tin-canned-foods (the Codex General Standard for Contaminants and Toxins in Food and Feed, CXS 193-1995, which contains the metal MLs across food categories). The taxonomy snapshot does not yet contain a tahini- or sesame-specific Codex MLs slug; the existing CXS 193 page is the correct routing target for these limits.

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.