Acar 2012 — Cd, Pb, Cu, Fe, Zn in Turkish dietary vegetable oils and olives by ETAAS/FAAS

This study determined Cd, Pb, Cu, Fe, and Zn in 53 edible vegetable oil samples (soybean, sunflower, flower, nut, corn, olive) and 70 table olive samples (olive-1, black, green, and black/green crushed-with-seeds preparations) purchased from supermarkets in Ankara, Turkey. Cd, Pb, and Cu were determined by electrothermal atomic absorption spectrometry (ETAAS) using a novel Sc + Ir + NH4H2PO4 chemical modifier mixture; Fe and Zn were determined by flame atomic absorption spectrometry (FAAS). All samples were prepared by microwave-assisted nitric acid/hydrogen peroxide digestion. Across vegetable oils, Cd ranged 31–58 µg/kg, Pb 0.04–0.11 mg/kg, Cu 0.04–0.13 mg/kg, Fe 0.96–2.14 mg/kg, and Zn 1.01–1.74 mg/kg. Across table olive preparations, Cd ranged 43–108 µg/kg, Pb 0.11–0.32 mg/kg, Cu 2.17–3.80 mg/kg, Fe 10.6–18.1 mg/kg, and Zn 2.33–4.44 mg/kg. The author reports Cd in olive samples as higher than several literature values cited but in agreement with the EC Commission Regulation 1881/2006 maximum contaminant levels and Turkish food codex regulations. The paper measures total Pb and total Cd (no speciation); arsenic and mercury are not measured.

Note on evidence tier: assigned B. Single-author study with reasonable analytical rigor (SRM 1577b bovine liver recovery 96–101%, RSDs 3.4–6.2%, microwave digestion, well-described modifier optimisation), but per-oil-type n is small (n=8–12), the geographic frame is restricted to Ankara market supply, and the panel omits arsenic and mercury (the two HMI-priority analytes most regulated in food safety contexts). Total Cd and total Pb are reported without speciation, which is appropriate for these elements in foods.

Key numbers

All values are wet weight basis. Cadmium is reported in µg/kg throughout (Table 3 footnote b); Pb, Cu, Fe, Zn are in mg/kg. Reported as mean ± standard deviation (95% CI) over min–max, n = five replicate measurements per sample (Table 3).

Vegetable oils (Table 3, page 390):

Table olive preparations (Table 3, page 390):

Across-preparation ranges (Table 4 “This study” column, page 391):

• Cd in oils: 31–58 µg/kg (= 0.031–0.058 mg/kg); Cd in olives: 0.04–0.11 mg/kg.
• Pb in oils: 0.04–0.11 mg/kg; Pb in olives: 0.11–0.32 mg/kg.
• Cu in oils: 0.04–0.13 mg/kg; Cu in olives: 2.17–3.80 mg/kg.
• Fe in oils: 0.96–2.14 mg/kg; Fe in olives: 10.6–18.1 mg/kg.
• Zn in oils: 1.01–1.74 mg/kg; Zn in olives: 2.33–4.44 mg/kg.

Comparison with cited literature ranges for Pb in oils (Table 4): 0.009–0.0074 (Zhu et al. 2011), 0–0.0074 (Pehlivan et al. 2008), 0–0.03 (Mendil et al. 2009), 0.074–0.104 (Bakkali et al. 2009); the maximum 0.11 mg/kg reported here exceeds Bakkali’s upper bound. For Cd in oils the literature ranges cited (Mendil 0.09–4.57, Zhu 2.64–8.43, Pehlivan 0.2–4.5, Bakkali 0.0013–0.0123 — all reported as mg/kg in Table 4) span more than four orders of magnitude; the present study’s 31–58 µg/kg (0.031–0.058 mg/kg) sits within that broad envelope but on the higher side compared with Bakkali and Pehlivan. The author notes Cd in oils is higher than literature values but consistent with the EC 1881/2006 maximum allowed Cd for vegetable oil-containing foods.

Recovery (Table 2, SRM 1577b bovine liver and oil-sample spike, page 388):

• Cd: certified 0.50 ± 0.03 mg/kg, found 0.48 ± 0.02, recovery 96%. Oil-sample 1.0 µg/L spike → 2.40 ± 0.08 (98%); 2.0 µg/L spike → 3.41 ± 0.13 (99%).
• Cu: certified 160 ± 8 mg/kg, found 158 ± 7, 99%.
• Pb: certified 0.129 ± 0.004 mg/kg, found 0.13 ± 0.01, 101%.
• Fe: certified 184 ± 15 mg/kg, found 181 ± 9, 98%.
• Zn: certified 127 ± 16 mg/kg, found 124 ± 6, 98%.

Limits of detection (LOD, 3σb/m on 20 blanks): Cd 0.1 µg/L, Cu 0.6 µg/L, Pb 0.9 µg/L, Fe 15.0 µg/L, Zn 12.0 µg/L. Characteristic masses (m0, ETAAS, integrated absorbance 0.0044): Cd 0.8 pg, Pb 8.3 pg, Cu 19 pg. RSDs of analytes in samples were below 7% (range 3.4–6.2%).

Regulatory benchmarks cited by the author (page 390):

• Maximum permissible values in Turkish standards (TS 2812:1991; TS 3606:1997) and Turkish food codex regulations (TKB 2002) for vegetable oils and refined liquid oils: 1.5 mg/kg Fe, 0.1 mg/kg Cu, 0.1 mg/kg Pb (no Cd value cited for vegetable oils in this set).
• TKB 2002 cap for olive oil and non-refined vegetable oil: Cu 0.1 mg/kg, Fe 3.0 mg/kg; Cu 0.4 mg/kg and Fe 5.0 mg/kg cited as alternative refining-state-dependent caps.
• Kowalewska et al. 2005 / Zhu et al. 2011 “approved contents of some metals in oils from national and international requirements” as cited by the author: 0.05 mg/kg Cd, 1–1.5 mg/kg Fe, 0.1 mg/kg Cu, 0.1 mg/kg Pb.
• IOOC 2003 / Economic European Communities 2006 maximum recommended levels cited by the author: 0.3 mg/kg Fe, 0.1 mg/kg Cu and Pb for oils and olives, 0.2 mg/kg Cd for soybeans, and 0.050 mg/kg Cd for vegetables, pine nuts, etc.
• Turkish local food standards for table olives (TS 774:2003) cited as the safe-limit benchmark for olive samples; no numerical value is reproduced in the paper.
• EVM 2003 safe upper level for Cu intake: 10 mg for a 60 kg adult.
• Birick and Basoglu 2006: maximum allowed Fe concentration 150 mg/kg.
• JECFA 1999 PTDI cited by author: Fe 48 mg, Cu 3 mg, Pb 214 µg, Zn 60 mg per 60 kg adult.

The author concludes that intake of Cd, Pb, Cu, Fe, and Zn from daily consumption of these Turkish oils and olives may not be a risk for human health, and that the various metal levels are below the safe limits specified for table olives by Turkish local food standards (TS 774:2003).

Methods (brief)

ETAAS for Cd, Pb, Cu: Hitachi Model 180/78 graphite furnace with Zeeman background correction; pyrolytically coated graphite tubes (P/N 190-6007) with graphite platforms (P/N 190-6008); peak-area (integrated absorbance) mode. Element lamps: Cd 10 mA / 228.8 nm; Cu 10 mA / 324.8 nm; Pb 10 mA / 283.3 nm. Argon (99.98% v/v) carrier gas. Heating program (Table 1): pyrolysis 900 °C (Cd), 1,300 °C (Cu), 1,250 °C (Pb); atomization 1,500 °C (Cd), 2,700 °C (Cu), 2,000 °C (Pb); ramp/hold and cleaning steps as specified.

Chemical modifier (novel mixture, the paper’s analytical contribution): Sc + Ir + NH4H2PO4 mixture — 2.0 g/L Sc + 0.4 g/L Ir + 5.0 g/L NH4H2PO4. Modifier solutions prepared from Sc2O3 (Merck, 99.99% pure) dissolved in HNO3, IrO2 (Merck) dissolved in HNO3, and NH4H2PO4 (Merck, 99.99% pure). The mixture stabilizes Cd, Pb, and Cu to higher pyrolysis temperatures and was selected after comparison with Sc-only and Sc + Ir variants and with previously reported modifiers including W-Rh, W-Ir, W-Ru, and Sc + Pd + NH4NO3 systems.

FAAS for Fe, Zn: Hitachi Model 180/80 flame AAS, air-acetylene; Fe 248.3 nm, Zn 213.9 nm; element lamps Fe 12.5 mA, Zn 10 mA. Slit width 1.3 nm for all lamps.

Sample digestion: microwave-assisted closed-vessel digestion (Milestone Ethos 1600, MLS Ethos, Italy). Triplicate 1.00–2.00 g oil, 1.00–2.00 g olive, and 0.50–1.00 g SRM 1577b bovine liver portions were weighed into separate Teflon digestion vessels with 2 mL ultra-pure water, 6 mL HNO3 (65% m/m), and 2 mL H2O2 (30% m/m). Vessels stood at ambient temperature for 4 h with gentle swirl, then digested at 160 °C, 800 W maximum power, for 20–25 min. Digest transferred to 25 mL volumetric flasks with 1% v/v HNO3 wash. Blank solutions prepared with the same procedure to check reagent contamination. Standard-addition method used to verify digestion validity: known amounts of analyte added to the second and third vessels of each triplicate.

Auto-sampler solution: 0.1% v/v HNO3 + 0.1% v/v Triton X-100 to disperse sample solutions onto the platform. Calibration: aqueous standard solutions diluted from 1,000 mg/L BDH stock standards in 0.1% v/v nitric acid, freshly prepared before use. Calibration dynamic ranges: Cd 0.3–6 µg/L, Pb 3.0–80 µg/L, Cu 2.0–80 µg/L, Fe 0.05–6.0 mg/L, Zn 0.04–1.0 mg/L. Correlation coefficients (R²) > 0.99 for all analytes. LOQ = lower limit of dynamic range = 10σb/m.

Sample sizes: 53 oil samples (8 soybean, 12 sunflower, 8 flower, 8 nut, 8 corn, 9 olive) and 70 table olive samples (20 olive-1, 15 black, 12 green, 10 black crushed with seeds, 13 green crushed with seeds), all purchased from various supermarkets in Ankara on different dates and originating from various plants and regions of Turkey. For olives crushed with seeds, the seeds were separated by hand or crushed, and the flesh was homogenized with a stainless blender to assess possible contamination from the olive seed.

No speciation: this paper reports total Pb, total Cd, and total Cu (and total Fe, total Zn). Arsenic and mercury are not measured.

Basis: wet weight throughout (page 387, section 3.4: “All metal concentrations found in samples were on a wet weight basis as mg kg−1 except cadmium” — Cd reported in µg/kg). Figure 2’s y-axis label “Concentration (mg/kg)” for the Cd panel is inconsistent with Table 3’s footnote-b µg/kg units and with Table 4’s Cd Olive “This study” 0.04–0.11 range (mg/kg-equivalent of 40–110 µg/kg); the µg/kg reading from Table 3 is the consistent unit basis used throughout this page.

Implications

• Certification (HMTc): contributes Turkish-market occurrence data for Pb and Cd in dietary vegetable oils and table olives, n = 53 oils and 70 olives across 11 distinct product subtypes. Pb values in oils (0.04–0.11 mg/kg) reach but do not clearly exceed the 0.10 mg/kg Pb cap the author cites from Turkish standards and IOOC/EC requirements; one oil-category maximum (sunflower 0.11 mg/kg, soybean 0.10 mg/kg, corn 0.10 mg/kg, olive 0.10 mg/kg) sits at or just above that line. Pb values in table olives (0.11–0.32 mg/kg) materially exceed the 0.10 mg/kg oil cap and are presented in the paper against a Turkish table-olive safety standard (TS 774:2003) that the paper does not numerically reproduce. Cd in oils (31–58 µg/kg = 0.031–0.058 mg/kg) sits below the 0.2 mg/kg TKB 2002 cap for soybean oil and pine nuts cited by the author and at or near the 0.05 mg/kg IOOC/EC Cd value the author also cites for oils.

• Courses: useful as a worked example of (1) the Sc + Ir + NH4H2PO4 modifier system for stabilizing Cd, Pb, and Cu in oil/olive matrices in ETAAS, (2) the standard-addition method’s use as a digestion-validation check, and (3) the basis convention that table olives are reported in mg/kg wet weight while Cd values are routinely reported in µg/kg even in the same table.

• App: oil-side Pb and Cd inputs are in the low end of the range that ingredient-list-driven estimators care about; the table-olive Pb range (0.11–0.32 mg/kg) is the more contaminant-relevant signal from this paper and should be carried into per-serving estimators for olive-as-ingredient and table-olive-as-snack use cases. Table-olive Cd (43–108 µg/kg) is the second most contaminant-relevant signal. Treat as one of several Turkish-market data points; combine with Pehlivan et al. 2008 (n=17 oils), Karatasli 2018 (n=26 table olives, Mediterranean Turkey), and other regional studies before drawing geographic conclusions.

Wiki pages this source may touch

• olive-oil
• olives
• sunflower-oil
• corn-oil
• soybean-oil
• vegetable-oils
• olive-oil
• cooking-oils-other
• non-root-vegetables
• lead
• cadmium
• copper
• iron
• zinc

Verification notes

• 2026-06-01 Fresh ingest from Manual Fetch Kimi /June 1/acar2012 (1).pdf. Three identity checks (DOI 10.3989/gya.047512 grep, raw_handle MFK_acar2012-1 grep, cite-key acar2012* grep) all returned no existing pages.
• “Flower oil” (n=8 in Table 3) is the author’s label for one of the six oil categories sampled; the paper does not specify the seed/source of this “flower” oil (Spanish abstract uses “aceite de flores”). Likely candidates include safflower or another flower-seed oil, but the paper does not resolve this. Frontmatter omits a “flower-oil”/“safflower-oil” ingredient slug because the identity is not confirmed in the source; values are reported under the author’s label only.
• “Nut oil” (n=8 in Table 3) is similarly a generic author label; the paper does not specify whether this is hazelnut, walnut, or another nut-source oil. Frontmatter omits a specific nut-oil ingredient slug because the identity is not confirmed in the source.
• Cd is reported in µg/kg in Table 3 (footnote b) for both oils and olives, and converts to the 0.04–0.11 mg/kg Table 4 “This study” range for olives via 40–110 µg/kg = 0.040–0.110 mg/kg. Figure 2’s Cd-panel y-axis “Concentration (mg/kg)” is an apparent figure-label typo (would imply 50–110 mg/kg Cd in olives, four orders of magnitude above the Table 3/Table 4 consistent reading); page follows Table 3/Table 4 as the consistent unit basis.
• The author measures total Pb and total Cd (no speciation); the page uses Pb and Cd in metals: accordingly.
• products/non-root-vegetables is used as the table-olive routing destination following the same convention as karatasli2018-radionuclide-heavy-metal-turkey-olives; a products/table-olives provisional scaffold would be the natural long-term routing destination for table-olive papers in this folder and others (Step 0 Lock candidate, flagged for Karen). Not creating a new product page from this session per task hard constraints.
• The paper cites several Turkish (TS 2812:1991, TS 3606:1997, TKB 2002, TS 774:2003), Turkish food codex, and international (IOOC 2003, EC Commission Regulation 1881/2006, JECFA 1999, EVM 2003) regulatory benchmarks. The wiki has neither Turkish food codex (TKB/TS) regulation pages for vegetable oils and olives nor an EC 1881/2006 vegetable-oil page yet, so author-cited limits are preserved in prose without wikilinks rather than asserted as cross-page facts.
• Author affiliation: Orhan Acar, Gazi University, Atatürk Occupational High School, 06750, Akyurt, Ankara, Turkey. Corresponding author: oacar@gazi.edu.tr.
• Article metadata: Grasas y Aceites 63(4), 383–393, October–December 2012, ISSN 0017-3495, DOI 10.3989/gya.047512. Grasas y Aceites publishes under CC BY 3.0 ES (verified per journal license policy current at publication date); license set accordingly.
• 2026-06-01 Audit subagent (Claude general-purpose, fresh context) verdict REVISE; one ❌ and two ⚠️ findings. ❌ verified against PDF page 390 and applied: the wiki had misattributed the Kowalewska et al. 2005 / Zhu et al. 2011 “approved contents” set (0.05 Cd, 1–1.5 Fe, 0.1 Cu, 0.1 Pb mg/kg) to IOOC/EC; the author’s actual IOOC 2003 / EC 2006 citation is 0.3 mg/kg Fe, 0.1 mg/kg Cu and Pb, 0.2 mg/kg Cd for soybeans, and 0.050 mg/kg Cd for vegetables and pine nuts. Corrected to two distinct bullets. The prior page’s separate “TKB 2002 maximum allowed Cd: 0.2 mg/kg for soybean oil, pine nuts, etc.; 0.050 mg/kg for vegetables” bullet was also misattributed (these are IOOC/EC values, not TKB 2002) and has been removed; the values are preserved in the new IOOC/EC bullet.
• 2026-06-01 Audit ⚠️ on Pb-in-oils literature attribution verified against PDF Table 4 and applied: Zhu et al. 2011 (0.009–0.0074 mg/kg) and Pehlivan et al. 2008 (0–0.0074 mg/kg) are separate Table 4 entries, not co-attributed. Split into two entries in the comparison sentence.
• 2026-06-01 Audit ⚠️ on matrices vocabulary noted as advisory only: bare-string matrices slugs (olive-oil, sunflower-oil, corn-oil, soybean-oil, edible-oil, table-olives) are not in the taxonomy-snapshot four-list (which covers ingredients/products/metals/regulations only) but are consistent with prior-ingest convention (pehlivan2008-vegetable-oils-turkey-icp-aes, karatasli2018-radionuclide-heavy-metal-turkey-olives); no change required.
• 2026-06-01 Audit confirmed all 55 numeric cells in Table 3 (oils and olives), all SRM 1577b recoveries, all LODs, characteristic masses, calibration ranges, RSDs, ETAAS heating-program temperatures, modifier composition, and microwave digestion conditions are faithful to the source. Checks 2 (slug vocabulary), 3 (speciation/methods), 4 (brand firewall — only Exception 2 scientific vendor names), and 5 (wiki/HMTc firewall) all clean.

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.