Ashraf 2012 — Heavy metals in seven vegetable oil varieties consumed in Saudi Arabia

This study determined concentrations of seven elements (Cu, Zn, Fe, Mn, Cd, Pb, and total As) in 161 commercially available edible vegetable oil samples from Saudi hypermarkets using graphite furnace atomic absorption spectrometry (GF-AAS) after microwave digestion, verified against NIST SRM 1577b (bovine liver). The author concludes that estimated daily and weekly dietary intakes from 25 g of oil per day (the WHO GEMS/Food maximum) remain well below the JECFA provisional tolerable weekly intakes (PTWIs) for Cd, Pb, and inorganic As, posing no significant health risk to the Saudi adult population at the analytical concentrations reported.

Key numbers

All concentrations are wet weight, µg/g, except cadmium (ng/g). Per-oil-type means (X ± SD) and ranges are from Table 1.

Arsenic (total tAs): mean range across the seven oil types 0.011–0.018 µg/g. Sesame and rapeseed oils highest (both 0.018 ± 0.001 µg/g); corn oil lowest (0.011 ± 0.002 µg/g). All-sample range 0.005–0.023 µg/g. GF-AAS detection limit listed in Table 6 as 0.121 µg/g — the reported sample means all fall below this LOD, so the As values are best read as upper-bound estimates rather than confirmed quantification.

Cadmium: mean range 2.36–6.34 ng/g. Corn oil highest (6.34 ± 0.51 ng/g); olive oil lowest (2.36 ± 0.11 ng/g); sesame (5.78 ± 0.21 ng/g) and rapeseed (5.71 ± 0.11 ng/g) also elevated. All-sample range 1.98–7.16 ng/g. GF-AAS detection limit listed in Table 6 as 0.092 µg/g (= 92 ng/g) — the reported sample means are all well below this LOD, so the Cd values are also best read as upper-bound estimates. Calibration against NIST SRM 1577b was performed at the certified value of 0.5 µg/g Cd (Table 6), well above the sample concentration range.

Lead (Pb): corn oil and olive oil were below detection limit (BDL); the remaining five oils had detectable Pb. Means: sunflower 0.011 ± 0.001, soybean 0.014 ± 0.001, peanut 0.013 ± 0.001, sesame 0.017 ± 0.002, rapeseed 0.011 ± 0.001 µg/g. Sesame oil highest detected. GF-AAS detection limit listed in Table 6 as 0.098 µg/g — the detected Pb means all fall below this LOD, so the Pb values are likewise best read as upper-bound estimates near or below the LOD.

Copper (Cu): mean range 0.035–0.286 µg/g. Olive oil highest (0.286 ± 0.016); corn oil lowest (0.035 ± 0.001). LOD 0.011 µg/g.

Zinc (Zn): mean range 0.955–3.10 µg/g per Table 1; the discussion text on page 2 incorrectly states that soybean oil had the lowest Zn — Table 1 shows soybean at 1.36 ± 0.02 µg/g and sesame at 0.955 ± 0.071 µg/g, so sesame is the actual minimum. Corn oil highest (3.10 ± 0.290). LOD 0.010 µg/g.

Iron (Fe): mean range 17.3–57.8 µg/g. Olive oil highest (57.8 ± 3.8); corn oil lowest (17.3 ± 0.8). LOD 0.017 µg/g.

Manganese (Mn): mean range 0.130–0.586 µg/g per Table 1 main concentration data; reported as 0.178–0.586 µg/g in the discussion text (the text omits soybean 0.130 µg/g as the actual minimum). Peanut oil highest (0.586 ± 0.033); soybean oil lowest in the table (0.130 ± 0.012). LOD 0.019 µg/g.

Estimated weekly dietary intake (EWDI) vs JECFA PTWI (for 70 kg adult, 175 g oil/week; Table 3):

• Cd: 0.006–0.016 µg/kg bw/week vs PTWI of 7 µg/kg bw — less than 0.3% of PTWI.
• Pb: 0.028–0.043 µg/kg bw/week vs PTWI of 25 µg/kg bw — less than 0.2% of PTWI. The Pb PTWI of 25 µg/kg bw is the value the source cites; JECFA withdrew the Pb PTWI in 2010 on the basis that no level of Pb exposure can be considered safe, so the “fraction-of-PTWI” framing carries the historical caveat.
• As (total): 0.028–0.045 µg/kg bw/week vs the JECFA PTWI of 15 µg/kg bw for inorganic As — less than 0.3% of the inorganic-As PTWI. The source measures total As, not iAs, so the comparison overstates the safety margin to whatever extent the As in these oils is inorganic.

Methods

GF-AAS (Shimadzu AA-6200 with HGA graphite furnace, ASC 6100 autosampler), argon inert gas. Microwave digestion: ~1 g sample + 5.0 mL HNO₃ (suprapure, E. Merck) + 3.0 mL H₂O₂, ramp program 2 min/250 W, 2 min/0 W, 6 min/250 W, 5 min/400 W, 8 min/550 W, 8 min vent; diluted to 10.0 mL with doubly distilled water. Sample procurement: 32 corn, 28 sunflower, 21 soybean, 19 sesame, 17 rapeseed, 17 peanut, and 27 olive oil samples purchased from Saudi hypermarkets during 2011–2012, refrigerated until analysis, run before stated expiry dates. Analyses performed in triplicate. Recovery experiments: standard additions of 5–20 µg/g per element, recoveries reported as >95% (Table 5). Certified reference material: NIST SRM 1577b (bovine liver) — measured values agreed with certified values within 1–3% across Fe (184 vs 182 ± 14 µg/g), Zn (127 vs 126 ± 7), Mn (10.5 vs 10.3 ± 0.7), Pb (0.129 vs 0.126 ± 0.02), Cd (0.5 vs 0.48 ± 0.04), Cu (160 vs 157 ± 9), and As (0.05 vs 0.05 ± 0.01) (Table 6). All concentrations reported on a wet-weight basis.

Analytical limitation: the GF-AAS detection limits in Table 6 (Pb 0.098 µg/g, As 0.121 µg/g, Cd 0.092 µg/g) are at or above the sample concentrations reported in Table 1 for Pb, As, and Cd. The CRM-recovery work was performed at certified-value concentrations one to three orders of magnitude above the sample range. Quantitative confidence in the trace-Pb, trace-As, and trace-Cd numbers in this paper should be treated as low; the values are most defensible as upper-bound estimates near the instrument floor rather than as precise concentrations.

Speciation: total arsenic only. No iAs/tAs partition. No Hg measured.

Implications

Certification: Vegetable oils from the Saudi market show very low Cd (≤ 6.4 ng/g = 6.4 ppb) and near-detection-limit Pb (BDL to 0.017 µg/g = 17 ppb). These are useful international comparison points for HMT&C ingredient profiles for culinary oils, though the LOD-driven analytical limitation flagged above means the values should be weighted lower than papers using ICP-MS or other methods with lower LODs.

Courses: Illustrates the role of GF-AAS methodology and the importance of matching analytical LODs to expected concentrations in edible oil matrices — a key quality-control point for commodity-level heavy metal screening. Also a worked example of an EDDI/EWDI dietary-intake calculation against JECFA PTWI guideline values.

App: Cd data in the 2–7 ng/g range and Pb data in the BDL to 17 ppb range for common culinary oils provide reference points for the vegetable-oil ingredient cluster (corn, sunflower, soybean, peanut, sesame, olive, rapeseed). The LOD limitations argue for weighting these values lower than ICP-MS-derived measurements when synthesizing the contamination profile.

Wiki pages this source may touch

• olive-oil
• corn-oil
• sunflower-oil
• soybean-oil
• peanut-oil
• sesame-oil
• rapeseed-oil
• olive-oil
• cooking-oils-other
• cadmium
• lead
• arsenic-total
• copper
• zinc
• iron
• manganese

Verification notes

• 2026-05-26 merge-enhance (Claude Opus 4.7, manual-fetch-kimi re-ingest): replaced placeholder raw_handle: manual-fetch-kimi with the canonical MFK_levels-of-selected-heavy-metals-in-varieties-of-ve handle; fixed truncated raw_path (was Vegetable Oils Co.pdf, actual file is Vegetable Oils Collected from Saudi Arabia.pdf); populated products: ["[[products/olive-oil]]", "[[products/cooking-oils-other]]"] — the food-paper-with-empty-products defect flagged in routing_malformed.csv.
• 2026-05-26 factual correction: prior page asserted “corn oil, olive oil, and rapeseed oil were below detection limit (BDL)” for Pb. Verified against Table 1 of the source — rapeseed oil Pb is 0.011 ± 0.001 µg/g (not BDL); only corn oil and olive oil are BDL. Corrected. The corresponding row in Table 3 (EWDI Pb) confirms the BDL pattern: NA values appear only for corn and olive columns.
• 2026-05-26 expanded Key numbers to include Cu, Zn, Fe, Mn means and ranges (all reported in Table 1 and previously omitted from the page body even though all seven elements are in metals:). Added Pb PTWI-withdrawal caveat and the tAs-vs-iAs comparison caveat on the EWDI-vs-PTWI framing.
• 2026-05-26 modernized Methods section with per-oil sample counts, microwave digestion ramp program, and the CRM-recovery-vs-sample-concentration analytical-limitation paragraph (replacing the earlier speculative “typographical unit inconsistency” framing of the Cd LOD — the unit µg/g in Table 6 is internally consistent with how the calibration was performed and is not a typo).
• 2026-05-26 modernized “Wiki pages updated on ingest” heading to “Wiki pages this source may touch” per current source-page convention; added arsenic-total, copper, zinc, iron, manganese to the touch list to match the expanded frontmatter coverage.
• publication: "Asian Journal of Chemistry (Uncorrected Proof)" retained from prior page. The PDF carries an “Uncorrected Proof” watermark but does not state the journal title or volume/issue on any of its five pages. The publication field is therefore unverified against the source itself; flagging for downstream cross-vendor audit confirmation rather than overwriting without evidence.
• doi: null / no_doi_assigned: true retained — the source PDF contains no DOI string.
• 2026-05-26 fresh-context audit (general-purpose subagent) — verdict PROMOTE. All five checks (numerical fidelity, slug vocabulary, speciation/methods, Part 12 brand firewall, Part 2 wiki/HMTc firewall) returned ✅ on substantive findings. Two ⚠️ concerns reviewed: (1) edible-oil matrices term — confirmed in corpus use across 12 sibling source pages alongside vegetable-oil, accepted vocabulary, no change; (2) Zn page-2-text vs Table-1 minimum inconsistency — applied: Zn key-number line now flags the source-internal contradiction analogously to the existing Mn note. Rapeseed-Pb-BDL correction from the merge-enhance pass was independently re-verified against Table 1 by the subagent.

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.