Tinggi et al. 2025 — Heavy metals in commercial spices and herbs, Queensland Australia

This study analyzed eight heavy metals (Al, As, Cd, Cr, Pb, Hg, Ni, Sr) in 69 dried commercial spice and herb samples purchased from Queensland, Australia markets, representing seven commodity types (basil, chilli, cinnamon, paprika, black pepper, sesame seeds, turmeric) imported from multiple countries. A separate analysis compared 13 conventional and 7 organic turmeric products. All metals were measured by ICP-MS (Agilent 8800 Triple Quad) following microwave-assisted acid digestion (CEM MARS 6), validated against certified reference materials. The study found wide variation in heavy metal content across commodities and origins, with lead in cinnamon from Vietnam (7.0 mg/kg) and India (2.5 mg/kg) representing the most concerning findings. Estimated dietary exposures from typical Australian spice consumption (1.9 g/day) were below all EFSA and WHO health-based guidance values for the metals assessed. Organic turmeric showed significantly higher Cr and Ni than conventional turmeric.

Key numbers

Table 4 — Metal concentrations across all seven commodities (mean ± SD; range in parentheses), mg/kg:

Co-occurring extremes in one Vietnamese cinnamon sample: Pb 7.0 mg/kg, Al 920 mg/kg, Cr 4.0 mg/kg, Ni 2.0 mg/kg (text, page 6–7).

Table 5 — Conventional vs organic turmeric (mean ± SD; range; mg/kg):

Authors’ summary: organic turmeric had higher mean values for every element measured, but the difference reached statistical significance only for Cr (P < 0.05) and Ni (P < 0.01); for Al, As, Pb, and Sr the differences were not statistically significant (Table 5 caption; text, page 7).

Table 6 — Estimated dietary exposure (1.9 g/day spice intake, 70 kg adult; concentration in µg/g equivalent to mg/kg):

*µg/kg bw/week. tHg excluded from the exposure calculation because >75% of samples fell below the 0.005 mg/kg LOR (Table 6 footnote).

Highest single-sample concentrations (text, pages 4–7 and Table 4 ranges):

• Pb 7.0 mg/kg — cinnamon from Vietnam; 2.5 mg/kg — cinnamon from India (Supplementary Table 1)
• Al 920 mg/kg — cinnamon (same sample as the 7.0 mg/kg Pb cinnamon)
• Cd 3.7 mg/kg — cinnamon (single outlier)
• Cr 4.0 mg/kg — cinnamon from Vietnam (same sample as the Pb/Al extremes)
• tAs 1.7 mg/kg — single black pepper sample (mean for black pepper 0.18 mg/kg)
• Ni 4.8 mg/kg — black pepper
• Sr 290 mg/kg — basil (n=2 only)
• tHg 0.059 mg/kg — maximum across all samples; ~75% of samples <LOR (0.005 mg/kg)

Methods (brief)

ICP-MS (Agilent 8800 ICP-MS Triple Quad, Agilent Technologies, Tokyo, Japan) with platinum cones, Agilent SPS 4 autosampler and Integrated Sample Introduction System (ISIS), operating conditions in Table 2 (RF power 1,550 W; argon flow 15 L/min; nebuliser pump 0.1 rps; three replicates per sample; scandium, rhodium, and iridium as internal standards). Microwave-assisted digestion (CEM MARS 6, USA) of 0.3–0.5 g dried sample with 5 mL 69% high-purity nitric acid (Seastar Chemicals, Canada) in sealed Teflon vessels using a three-step programme (400 W ramp 7 min to 85 °C, hold 5 min; 800 W ramp 10 min to 110 °C, hold 10 min; 1,600 W ramp 7 min to 165 °C, hold 10 min). Digested solutions diluted to 40 mL with high-purity deionised water (Aqua Cure, Burscough, UK) before analysis. Calibration curves linear 0.1–100 µg/L (r ≥ 0.9997).

Quality control: certified reference materials ERM CE-278K (mussel tissue, n=8 replicates) and NIST SRM 1547 (peach leaves, n=5 replicates) processed alongside samples; recoveries: As 97.2/105.7%, Cd 94.6/95.6%, Cr 73.5/—, Pb 93.5/90.8%, tHg 92.0/82.8%, Ni 90.1/80.4%, Sr 88.9/103.4%; Al recovery 68.5% (matrix-effect-attenuated in the high-Al SRM). LOD: Al 0.04, tAs 0.001, Cd 0.001, Cr 0.006, Pb 0.002, tHg 0.001, Ni 0.003, Sr 0.003 mg/kg. LOR (10× SD of blanks): Al 0.1, tAs 0.005, Cd 0.005, Cr 0.01, Pb 0.005, tHg 0.005, Ni 0.01, Sr 0.01 mg/kg. Below-LOR values substituted with LOR for statistical analysis. Laboratory participates in the National Measurement Institute (NMI) Australia proficiency-testing programme for trace metals in food.

Statistical analysis: descriptive statistics and Mann–Whitney non-parametric tests in IBM SPSS 29.0.2; boxplots in R 4.0.4.

Speciation: total arsenic (tAs) and total mercury (tHg) only; no inorganic-arsenic or methylmercury speciation reported. The exposure calculation for arsenic conservatively benchmarks tAs against the EFSA inorganic-arsenic BMDL01 (Table 6 footnote), treating all measured As as if it were iAs.

Implications

Certification: Cinnamon from Vietnam (Pb 7.0 mg/kg) and India (Pb 2.5 mg/kg) exceeds the EU Commission Regulation 1881/2006 (as amended by 1317/2021) maximum of 2.0 mg/kg Pb in dried bark spices, cited by the authors on page 2; both exceed the Australian FSANZ maximum of 0.1 mg/kg cited on page 6–7. The single Vietnamese cinnamon sample carrying co-occurring Pb 7.0 mg/kg, Al 920 mg/kg, Cr 4.0 mg/kg, and Ni 2.0 mg/kg is consistent with the paper’s discussion of leaching from stainless-steel processing equipment as a multi-metal contamination pathway, in addition to soil/plant uptake.

Courses: Useful as a case study in (1) the spread of contamination across imported-spice supply chains; (2) the limits of “organic” as a heavy-metal-safety signal — organic turmeric in this study had higher mean values for every metal, with Cr and Ni reaching statistical significance; (3) ICP-MS method validation with explicit QC reporting (CRMs, recoveries, LOD/LOR, proficiency-test participation).

App: Provides Australian-market occurrence data for cinnamon (Pb up to 7.0 mg/kg, Cd up to 3.7 mg/kg, Al up to 920 mg/kg), black pepper (tAs up to 1.7 mg/kg, Ni up to 4.8 mg/kg), basil (Al up to 400 mg/kg, Sr up to 290 mg/kg, n=2), and turmeric (Pb up to 0.90 mg/kg conventional, 0.45 mg/kg organic) as contamination_profile inputs for the corresponding ingredient pages.

Verification notes

• 2026-05-28 (Claude Opus 4.7, v2.0 manual-fetch merge-enhance — Phase 3 audit application): fresh-context audit subagent (Phase 2) returned REVISE with one ❌ and two ⚠️ items. All three verified independently against the PDF and taxonomy snapshot, and all three applied:
  • ❌ “aluminium” in the Wiki pages section used the British spelling, which is not in docs/gpt-collaboration/taxonomy-snapshot.md (snapshot lists aluminum, and the on-disk file is wiki/metals/aluminum.md). Corrected to [[metals/aluminum]].
  • ⚠️ “ID” (Indonesia) appeared in jurisdictions: but Table 1 (page 3) does not list Indonesia as a country of origin for any of the 69 commercial samples — Indonesia is mentioned only on page 9 as a passing prior-study comparator (“cinnamon from Indonesia contributes about 94% of TWI”). Removed ID. Remaining jurisdictions re-verified against Table 1: AU (retail), IN, CN, LK, VN, BD, MY, EG, FJ, ES.
  • ⚠️ tier_rationale claimed recoveries “89–106% except Al at 68.5%” but Table 3 (page 5) shows Cr ERM-CE-278K recovery is 73.5%, below the stated lower bound. Rewrote to honestly report the actual recovery range (73.5–105.7%) with both sub-89% elements (Al 68.5%, Cr 73.5%) named, while preserving the paper’s own statement (page 4) that recoveries were within the certified-value uncertainty ranges except for Al.
• 2026-05-28 (Claude Opus 4.7, v2.0 manual-fetch merge-enhance — Phase 1): page was originally created 2026-05-14 in the kimi-cond05 batch ingest. Phase 1 enhancements applied:
  • Set per-PDF raw_handle: MFK_heavy-metal-analysis-in-commercial-spices-and-herb (was the generic batch handle manual-fetch-kimi).
  • Restored full raw_path with the trailing-space folder name (“Manual Fetch Kimi ”) matching the on-disk filename “Heavy metal analysis in commercial spices and herbs by ICP-MS and estimated dietary exposure.pdf” (prior path truncated the filename mid-title at “ICP-MS and.pdf”).
  • Added raw_sha256 (d05e1c20…75c0c), access_url (https://doi.org/10.20517/jeea.2025.10), and tier_rationale documenting the A-tier placement (full ICP-MS QC, two CRMs, LOD/LOR for all elements, NMI proficiency-test participation).
  • Expanded jurisdictions to include all country-of-origin codes named in Table 1 (was missing EG, ID, FJ, ES; AU is the retail jurisdiction).
  • Added sampling_locations: [Queensland] (retail-market jurisdiction).
  • Refined license from “CC BY” to “CC BY 4.0” per the licence statement on the title page.
  • Expanded sample_population to give the per-commodity n breakdown explicitly (was a one-line summary).
  • Expanded the Key numbers section from a bullet-list extract to the full Table 4 reproduction (all eight elements across seven commodities + total), the full Table 5 conventional/organic turmeric comparison with Mann–Whitney significance, and the full Table 6 dietary-exposure calculation (was missing the µg/g, µg/day, and HBGV-value columns).
  • Expanded Methods (brief) to include the full ICP-MS operating conditions (Table 2), the three-step microwave-digestion programme, both CRMs with per-element recoveries, LOD and LOR values for all eight elements, and the statistical software versions.
  • Added an explicit speciation line: total arsenic and total mercury only; no iAs or MeHg speciation. The exposure calculation conservatively benchmarks tAs against the EFSA iAs BMDL01.
  • Rewrote the Implications > Certification paragraph to remove the prior “HMT&C sourcing specifications for cinnamon should include Pb and Al screening, with Vietnam and India as elevated-risk origins” sentence — that is a forward-looking HMT&C-policy recommendation, which per CLAUDE.md Part 2 belongs on HMT&C certification pages rather than on a single-source page. The factual content (Pb 7.0/2.5 mg/kg in Vietnamese/Indian cinnamon vs EU 2.0 and FSANZ 0.1 mg/kg limits; co-occurring Pb/Al/Cr/Ni extremes consistent with stainless-steel leaching pathway discussed in the paper) is preserved because the paper itself makes those comparisons.
  • Replaced the legacy ## Wiki pages updated on ingest heading with the current template ## Wiki pages this source may touch (the legacy heading is flagged by docs/gpt-collaboration/verification-checklist.md).
• DOI (10.20517/jeea.2025.10), title, authors (Tinggi U; Farrell M; Porter J; Mitchell S; Jurd S), year (2025), publication (J. Environ. Expo. Assess.), licence (CC BY 4.0), and metals list (Al, tAs, Cd, Cr, Pb, tHg, Ni; Sr also measured but not currently in the metals taxonomy) verified against page 1 citation block and Tables 2–6.
• All Table 4 / Table 5 / Table 6 values reproduced exactly from the source PDF; no rounding, no unit conversion (mg/kg in source = mg/kg in wiki; µg/g in Table 6 = mg/kg).
• Brand firewall (Part 12): the source does not name any brands of spice or herb; commodity × country-of-origin breakdowns are reproduced without brand attribution. The methods section names instrument vendors (Agilent, CEM, Seastar Chemicals, Aqua Cure, IBM SPSS) and reference-material suppliers (ERM, NIST), which is preserved per the 2026-05-17 scientific-method vendor-name exception.

Wiki pages this source may touch

• herbs-and-spices
• turmeric
• cinnamon
• black-pepper
• spices
• lead
• cadmium
• arsenic
• mercury
• nickel
• aluminum
• chromium

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.