Ziarati et al. 2019 — Toxic metals in Iranian and Italian flavoured olive oil

This study measured Pb, Cd, Ni, and As in 480 commercial olive oil samples purchased across three consecutive seasons from markets in Lombardy (Italy) and Tehran (Iran), comparing non-flavoured (virgin, extra-virgin) and flavoured (fungi, aroma vegetable, pepper) varieties. Table 3 reports higher Iranian Pb, Cd, and As combined means than Italian samples and a lower Iranian Ni combined mean, although the narrative text states that Iranian Ni was higher; treat the country-level Ni comparison as source-internal contradictory. Fungi-flavoured olive oil exhibited the highest Italian Pb and Cd concentrations, while pepper-flavoured olive oil showed the lowest heavy metal levels across the Italian flavour categories.

Key numbers

Italian olive oil (mean ± SE, µg/g):

• Pepper-flavoured: Pb 0.984 ± 0.091, Cd not detectable, Ni not detectable, As 0.00020 ± 0.00001
• Fungi-flavoured: Pb 12.3301 ± 0.112, Cd 1.004 ± 0.0123, Ni 10.231 ± 0.0143, As 0.00040 ± 0.00001
• Vegetable-flavoured: Pb 10.762 ± 0.088, Cd 0.0876 ± 0.004, Ni 12.349 ± 0.128, As 0.00030 ± 0.00001
• Non-flavoured: Pb 10.111 ± 0.034, Cd 0.0967 ± 0.0054, Ni 14.180 ± 0.011, As 0.00040 ± 0.00001

Iranian olive oil series (mean ± SE ranges across four anonymized series, µg/g):

• Pb: 11.209 ± 0.054 to 18.783 ± 0.016
• Cd: 2.001 ± 0.0022 to 4.181 ± 0.042
• Ni: 9.786 ± 0.0102 to 14.444 ± 0.283
• As: 0.0030 ± 0.0001 to 0.0050 ± 0.0002

Combined means (mean ± SE, µg/g):

• Italian: Pb 8.546 ± 0.043, Cd 0.396 ± 0.031, Ni 12.253 ± 0.078, As 0.00020 ± 0.00001
• Iranian: Pb 14.944 ± 0.045, Cd 3.362 ± 0.032, Ni 9.579 ± 0.065, As 0.0039 ± 0.0002
• Cd in Iranian samples was approximately ten times higher than in Italian samples (p < 0.005)
• Ni caveat: Table 3 reports Iranian Ni (9.579 µg/g) below Italian Ni (12.253 µg/g), but the surrounding narrative says Iranian Ni was higher than Italian Ni; preserve the tabulated values and do not use the country-level Ni direction without source review.

Storage effect (combined samples, µg/g):

• Pb: 14.221 ± 0.028 at 1 month, 15.128 ± 0.033 at 3 months, 15.867 ± 0.081 at 6 months, and 27.111 ± 0.102 at 12 months
• Cd: 2.989 ± 0.032 at 1 month, 3.156 ± 0.022 at 3 months, 4.001 ± 0.005 at 6 months, and 4.989 ± 0.011 at 12 months

Note: µg/g is equivalent to mg/kg. The reported Pb concentrations and several Cd values are orders of magnitude above the paper’s own regulatory discussion. The 0.1 mg/kg Codex/USDA limits cited in the introduction are stated for Pb and As; the paper also states that fungi-flavoured Cd exceeded a FAO/WHO permissible level but does not tabulate a Cd-specific vegetable-oil limit. The source should be treated as high-caution occurrence evidence.

Methods (brief)

ICP-AES (Varian Vista) for Cd, Pb, and Ni after dilute HNO3/H2O2 extraction (3.0 g oil + 1 mL 10% HNO3 + 0.5 mL 30% H2O2, shaken, held at 60 °C for 1.5 h, centrifuged, diluted to 25 mL). Arsenic work used ICP-MS (Agilent 7500c) and chromatographic separation of As species against As(III), As(V), DMA, and MMA standards; total-As digestion used HNO3/H2O2 microwave digestion (CEM Mars5). Samples were purchased across autumn 2016, winter 2017, and spring 2017 from recognized markets in Lombardy (Italy) and Tehran (Iran). Three replicates per sample; certified reference material used for QC (recoveries >95%). Reported values are dry weight. LOD established as 3× standard deviation of procedural blanks.

Implications

• Certification (HMTc): Provides source-specific occurrence evidence for Pb, Cd, Ni, and As in flavoured and non-flavoured olive oils from two market geographies. Because the reported Pb and Cd values are extremely high relative to the paper’s own regulatory discussion, use as high-caution occurrence evidence rather than as a direct calibration input.

• Courses: Storage time as a driver of Pb and Cd accumulation (12-month values roughly double 1-month values in this dataset) is a practical supply-chain case study for QA training.

• App: Adds contextual evidence that flavoured oils may not share the same occurrence profile as plain olive oil; do not use the extreme values as contamination-profile calibration inputs until independently replicated or cross-audited.

Caution: The source’s quantitative values are unusually high relative to the paper’s own regulatory discussion. They should be used to indicate source-reported direction-of-effect (fungi-flavoured > pepper-flavoured in the Italian subset; Iranian combined Pb/Cd/As > Italian combined Pb/Cd/As) rather than as calibration inputs until independently replicated. The country-level Ni direction is internally inconsistent between Table 3 and the narrative.

Wiki pages this source may touch

• olive-oil
• olive-oil
• lead
• cadmium
• nickel
• arsenic-total

Verification notes

• 2026-05-17 Codex merge-enhance: identity checked against DOI 10.2478/ata-2019-0012 and the manual-fetch PDF; corrected legacy raw_handle/truncated path, rerouted from products/cooking-oils-other to products/olive-oil, corrected Italian Table 1 transcriptions (non-flavoured Pb/As and fungi vs vegetable Ni), aggregated the four Iranian anonymized series into ranges to avoid brand/series-level ranking, and changed legacy headings to current source-page headings.
• Strict Part 12 cleanup: source reports four anonymized Iranian series; the wiki page now reports the range across the series rather than listing series-by-series values with a highest-series callout.
• Arsenic methods include speciation standards and separation, but the result tables report a single As concentration; frontmatter therefore retains tAs pending cross-audit.
• edible-oil, olive-oil, and flavoured-olive-oil are source-specific matrix descriptors used for routing/context; ingredient and product declarations use existing wiki slugs.
• 2026-05-18 Cross-vendor audit (Codex) made the source-internal Ni contradiction explicit: Table 3 reports Iranian Ni lower than Italian Ni (9.579 vs 12.253 µg/g), while the narrative says Iranian Ni was higher. Also tightened the regulatory note so Codex/USDA 0.1 mg/kg is not over-attributed to Cd.
• 2026-05-25 Claude cross-audit (fresh-context subagent): PROMOTE verdict. All Table 1/2/3/4 values verified against pp. 66-67; methods description verified against pp. 65-66; brand-firewall (Iranian series aggregation) and HMTc-firewall (no threshold proposals) verified clean. Applied finding: restored SE values on the Italian/Iranian combined-means line for completeness (was reporting only the mean from Table 3). Leave-as-noted findings: Iranian series-4 Ni SE rounding (±0.283 vs paper’s ±0.2831, trivial), pre-bath 50 Hz × 60 s mixer detail not transcribed (methods minutiae), CRM vendor (Alpha-Line, Chem Tech Analytical) not named on page (incomplete but not erroneous), flavoured-olive-oil matrix descriptor remains non-canonical pending system-level routing-layer review.

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.