Ammar et al. 2023 — Element migration from traditional and pressure-cooker aluminum cookware into cooked rice-and-beef meals and EU 10/2011 food simulants, Saudi Arabia

The authors quantified migration of Al, Fe, As, Cd, and Pb from four Saudi-market aluminum cooking pots — two traditional pots (manufacturer codes AC, AS) and two aluminum pressure cookers (codes ACP, APP) — into actual cooked food (rice + beef + acidifying and seasoning ingredients) and into Regulation (EU) No 10/2011 food simulant B (3% w/v acetic acid). Pot composition was characterised by spark-OES; food and simulant samples were analysed by ICP-MS after microwave digestion (HNO₃ + H₂O₂). SEM-EDS and XPS were applied in parallel and were demonstrated, by the authors, to be unsuitable for quantifying these migration loads because the migrated concentrations fall below the EDS and XPS detection limits (≈0.1 wt.%, i.e. ≈1000 µg/g) — only ICP-MS resolved them. The experimental design isolates four migration drivers: food acidity (tomato sauce + black lemon vs plain), spice load, water source (tap vs mineral), and extended cooking time (2 h boil only vs 2 h boil + 8 h at 70 °C), with the extended-time condition then repeated across all four pots to compare traditional vs pressure-cooker construction.

Key numbers

Migration values are reported in µg/g (≡ mg/kg) of cooked food or simulant, as means of three replicates with standard deviations. Table 6 values represent migrated mass after subtracting the elemental contribution of the raw ingredients measured in Table 5; Table 7 simulant values are direct measurements against a 3% acetic acid reference sample R0 with no subtraction.

Cookware composition (Table 4, Spectromaxx OES, wt.%). All four pots are >98.5 wt.% Al; the pressure cookers (ACP, APP) contain slightly different minor-element distributions than the traditional pots:

• Al: AC 99.24, AS 99.60, ACP 98.59, APP 99.53
• Fe: AC 0.3782, AS 0.2483, ACP 0.6036, APP 0.2889
• Mn: AC 0.0219, AS 0.0012, ACP 0.6528, APP 0.0144 (ACP elevated)
• Ni: AC 0.0029, AS 0.0008, ACP 0.0010, APP 0.0011
• Cd: AC 0.0006, AS 0.0005, ACP 0.0006, APP 0.0005
• Pb: <0.0065 across all four (below LOD)
• Sn: <0.0005 across all four
• Other detected at sub-percent levels: Mg, P, Cr, Si, Ti, Sr, Zr, Cu, Zn, V, Sb, B, Co, Ca.

Raw-ingredient baseline (Table 5, ICP-MS, µg/g, mean ± SD of n=3). These are pre-cooking values for the ingredients used in the food experiments and serve as the subtractor for Table 6:

• Rice: Al 15.20 ± 1.017; Fe 10.94 ± 0.597; As 0.0892 ± 0.0064; Cd 0.0313 ± 0.0027; Pb 0.1261 ± 0.0029
• Dry lemon (black lemon): Al 58.61; Fe 84.06; As 0.0262; Cd 0.0137; Pb 0.2101
• Clove: Al 83.48; Fe 79.39; As 0.0173; Cd 0.0062; Pb 0.4919
• Black pepper: Al 134.4; Fe 131.6; As 0.0436; Cd 0.0169; Pb 0.2819
• Cardamom: Al 48.22; Fe 52.62; As 0.2691; Cd 0.0716; Pb 0.6583
• Kabsa spices: Al 110.2; Fe 192.2; As 0.0696; Cd 0.0374; Pb 0.3814
• Cinnamon: Al 55.53; Fe 40.17; As 0.2284; Cd 0.0543; Pb 1.1688
• Beef (meat): Al 11.64; Fe 24.12; As 0.0071; Cd 0.0092; Pb 0.0360
• Tomato sauce: Al 16.10; Fe 15.32; As 0.0150; Cd 0.0203; Pb 0.0158
• Table salt: Al 21.55; Fe 14.54; As 0.1330; Cd 0.1663; Pb 0.3698

Migration into cooked food (Table 6, ICP-MS, µg/g, mean ± SD of n=3 — values after subtracting Table 5 raw-ingredient contributions).

• AC-1 (rice + beef + tap water + oil; 2 h boil; pH 8.83 → 8.33): Al 80.17 ± 5.997; Fe 6.493 ± 0.494; As 0.0053 ± 0.0007; Cd 0.0005 ± 0.00004; Pb 0.0056 ± 0.00084
• AC-2 (AC-1 + tomato sauce + black lemon; pH 6.48 → 6.20): Al 133.7 ± 8.501; Fe 8.922 ± 0.676; As 0.0172 ± 0.0014; Cd 0.0008 ± 0.000092; Pb 0.0087 ± 0.00092
• AC-3 (AC-2 + salt + Kabsa spices + cinnamon + cloves + cardamom + black pepper; pH 6.72 → 6.17): Al 129.4 ± 6.277; Fe 9.296 ± 0.830; As 0.0113 ± 0.0011; Cd 0.0002 ± 0.000019; Pb 0.0078 ± 0.00089
• AC-4 (AC-3 prepared with bottled mineral water in place of tap water; pH 6.80 → 6.72): Al 130.1 ± 9.304; Fe 11.98 ± 0.942; As 0.0279 ± 0.0026; Cd 0.0001 ± 0.000012; Pb 0.0084 ± 0.00091
• AC-5 (AC-3 with 8 h hold at 70 °C after the 2 h boil; pH 6.60 → 6.48): Al 157.9 ± 8.722; Fe 14.45 ± 1.157; As 0.0265 ± 0.0012; Cd 0.0010 ± 0.00013; Pb 0.0039 ± 0.00049
• AS-5 (AC-5 conditions, manufacturer-2 traditional pot; pH 6.70 → 6.32): Al 145.4 ± 7.697; Fe 9.171 ± 0.721; As 0.0267 ± 0.0031; Cd 0.0007 ± 0.00011; Pb 0.0054 ± 0.00078
• ACP-5 (AC-5 conditions, manufacturer-3 pressure cooker; pH 6.53 → 6.30): Al 252.7 ± 7.793; Fe 17.22 ± 1.316; As 0.0292 ± 0.0021; Cd 0.0002 ± 0.00003; Pb 0.0034 ± 0.00053
• APP-5 (AC-5 conditions, manufacturer-4 pressure cooker; pH 6.41 → 6.10): Al 221.0 ± 6.085; Fe 9.401 ± 0.320; As 0.0182 ± 0.0017; Cd 0.0009 ± 0.00014; Pb 0.0025 ± 0.00037

Migration into Regulation (EU) No 10/2011 food simulant B (Table 7, ICP-MS, µg/g, mean ± SD of n=3). Simulant is 3% w/v acetic acid; reference R0 is the simulant pre-contact (Al 0.0067, Fe 0.0363, As 0.0014, Cd 0.00009, Pb 0.0009 µg/g). OM4 conditions: 100 °C / 1 h. OM5 conditions: 100 °C / 2 h (the EU-defined worst case applied here).

• AC-FS OM4: Al 11.75 ± 0.293; Fe 0.9413; As 0.0072; Cd 0.00032; Pb 0.0011
• AS-FS OM4: Al 9.148 ± 0.164; Fe 0.4725; As 0.0029; Cd 0.00007; Pb 0.0006
• ACP-FS OM4: Al 61.76 ± 1.693; Fe 0.2143; As 0.0102; Cd 0.00005; Pb 0.0005
• APP-FS OM4: Al 35.05 ± 0.914; Fe 1.064; As 0.0586; Cd 0.00005; Pb 0.0010
• AC-FS OM5: Al 121.7 ± 4.629; Fe 5.397; As 0.0110; Cd 0.00007; Pb 0.0013
• AS-FS OM5: Al 117.6 ± 2.035; Fe 4.050; As 0.0175; Cd 0.00018; Pb 0.0017
• ACP-FS OM5: Al 249.6 ± 3.354; Fe 9.868; As 0.0151; Cd 0.00008; Pb 0.0016
• APP-FS OM5: Al 211.8 ± 5.586; Fe 5.843; As 0.0149; Cd 0.00011; Pb 0.0021

Estimated daily intake (Table 9, mg/day). The authors compute EDI from Table 6 µg/g values assuming a 100 g/day consumption portion of the cooked food, and compare against FAO/WHO acceptable daily intakes cited as Al 21 mg/day, Fe 18 mg/day, As 0.030 mg/day, Cd 0.058 mg/day, Pb 0.252 mg/day (sources Table 9 §3.6, refs [50–52]):

• AC-1: Al 9.327; Fe 2.417; As 0.0046; Cd 0.0019; Pb 0.0079 — all below limits.
• AC-2: Al 14.73; Fe 2.782; As 0.0056; Cd 0.0019; Pb 0.0079 — all below.
• AC-3: Al 14.44; Fe 3.029; As 0.0052; Cd 0.0020; Pb 0.0087 — all below.
• AC-4: Al 14.52; Fe 3.298; As 0.0069; Cd 0.0020; Pb 0.0088 — all below.
• AC-5: Al 17.29; Fe 3.544; As 0.0067; Cd 0.0021; Pb 0.0084 — all below (Al close to limit at 82% of 21 mg/day).
• AS-5: Al 16.05; Fe 3.017; As 0.0068; Cd 0.0021; Pb 0.0085 — all below.
• ACP-5: Al 26.78 (exceeds 21 mg/day); Fe 3.821; As 0.0070; Cd 0.0020; Pb 0.0083.
• APP-5: Al 23.61 (exceeds 21 mg/day); Fe 3.040; As 0.0059; Cd 0.0021; Pb 0.0082.

Pressure-cooker pots used under the AC-5 extended-cooking regime (ACP-5, APP-5) push estimated daily aluminum intake above the FAO/WHO 21 mg/day acceptable level at a 100 g/day consumption assumption; all other cooked-food conditions remain below. Fe, As, Cd, and Pb remain well below their respective limits in every condition.

Methods (brief)

Cookware-composition characterisation. A Spectromaxx Q4-Tasman benchtop spark-OES (Bruker, Germany), calibrated with the “Al-10-M” pure-aluminum / low-alloy program against a certified reference aluminum sample, analysed four 3×3 cm specimens per pot (cut from bottom and side positions with an angle grinder), each in triplicate (12 tests/pot, averaged).

Cooked-food analysis. Cooked-food samples (210–260 mg per replicate) were digested in a Milestone START D microwave system (Leutkirch im Allgau, Germany) in 3 mL HNO₃ (70% analytical grade) + 1 mL H₂O₂ (30% w/v) under a three-stage programme: 10 min at 800 W / 100 °C; 10 min at 800 W / 130 °C; 15 min at 800 W / 180 °C; then 30 min forced ventilation cool-down. Digests were transferred quantitatively to 100 mL volumetric flasks, made up with ASTM Type II purified water (Thermo Barnstead LED, 0.055 µS/cm), and analysed on an Agilent ICP-MS 7800 (Model G8421A) under operating conditions: RF 1550 W, sample depth 8.0 mm, micro-mist nebuliser, carrier gas 1.05 L/min, make-up 0.15 L/min, plasma 15 L/min, He gas 5 mL/min, dynamic reaction cell active, energy discrimination 3 V, 3 points/peak, 3 repetitions, integration 0.3 s. Multi-element calibration standards (10 µg/mL, Agilent Technologies) at three concentrations; method verified by spiked samples in triplicate.

Food-simulant analysis. Food Simulant B prepared per Regulation (EU) No 10/2011 (3% w/v acetic acid). OM4 test: 100 °C / 1 h. OM5 test (the EU-defined worst-case for foods up to 121 °C): 100 °C / 2 h. Same ICP-MS configuration; no digestion required for the simulants. Reference R0 (uncontacted simulant) carried in every batch.

Raw-ingredient analysis. Rice, meat, lemon, cloves, black pepper, cardamom, Kabsa spices, cinnamon, tomato sauce, and table salt were digested and analysed identically to cooked food (triplicate).

SEM-EDS and XPS (qualitative). SEM-EDS (Apreo Lovac FEGSEM, Thermo Fisher, with OCTANE PRO EDS, 200× and 500× imaging) and XPS (Thermo Fisher) were applied to dried, ground food samples (≈20 g, dried at 60 °C overnight, ground 10 min by mortar-and-pestle). For the ACP-5 condition only, neither technique detected Al, Fe, As, Cd, or Pb because all five fall below the EDS detection limit (≈0.1 wt.%, ≈1000 µg/g, refs [42–44]) and the XPS detection limit (0.1–1 at.%, refs [45–47]). Only C and O were detected by both techniques. The authors present this as a methodological finding: EDS and XPS are not appropriate for the trace concentrations at issue.

Speciation note. The paper reports total As and total Pb (and total Cd, Fe, Al) only. No As speciation (iAs vs tAs) was performed; no Hg measurement was attempted. The single-acid HNO₃ + H₂O₂ digestion and ICP-MS quantification protocol delivers total-metal values without redox-state preservation. Implications for inorganic-arsenic interpretation are noted under Verification notes.

Implications

Certification: Direct evidence on aluminum and minor-element migration from food-contact aluminum cookware (traditional pots and pressure cookers) into a rice-and-beef-based cooked-meal matrix and into Regulation (EU) No 10/2011 simulant B. Data inform the metal-and-alloy cookware row of the food-contact category. The dominant finding is that aluminum migration is sensitive to four operational variables — food acidity, cooking duration, cooking temperature, and pot construction (traditional vs pressure cooker) — and that the pressure-cooker × extended-cooking combination is what pushes estimated daily aluminum intake above the FAO/WHO 21 mg/day level at the authors’ 100 g/day consumption assumption. As, Cd, and Pb migration is minor across all conditions tested; Fe migration tracks aluminum but at one to two orders of magnitude lower mass.

Courses: Usable as a teaching case on (i) the four-driver structure of metal migration from food-contact materials (acidity, time, temperature, construction); (ii) the food-simulant vs actual-food validation logic of Regulation (EU) No 10/2011 (the OM5 condition reproduces the ACP-5 aluminum-leaching ordering despite the simpler matrix); (iii) the detection-limit mismatch between SEM-EDS / XPS and ICP-MS that the paper makes explicit and that recurs in cookware-leaching literature; and (iv) why pressure-cooker construction matters mechanically — the elevated system pressure increases ionic migration rate at the metal–food interface.

App: The aluminum contamination_profile for the metal-and-alloy cookware product row receives a direct-evidence contribution. The cookware-route attribution should be flagged for users selecting pressure cookers for extended-time preparation of mildly acidic foods; the same pressure cookers used for short cooking on neutral foods do not produce the same outcome. Ingredient-page aluminum cells for rice, beef, tomato, salt, cinnamon, black pepper, and the spices group should treat Table 6 values as cookware-driven migration, not as intrinsic-ingredient occurrence; Table 5 values are the intrinsic-ingredient measurements and should be the input to the ingredient-page contamination_profile blocks for the spices in particular (cinnamon Pb 1.17 µg/g, cardamom Pb 0.66 µg/g, clove Pb 0.49 µg/g, black pepper Pb 0.28 µg/g — these are the unbiased baseline before cookware contact).

Verification notes

• Brand-firewall (Part 12): The four pot manufacturers are referred to throughout the paper only as “manufacturer (1),” “manufacturer (2),” “manufacturer (3),” and “manufacturer (4)” with the codes AC, AS, ACP, APP attached. No manufacturer brand name, store identifier, or retail trade name is disclosed in the paper or the supplementary materials referenced (Tables S1–S4 are technical specifications, not brand identities). Retained as published.
• Methods-vendor exception (Part 12 Exception 2): Spectromaxx Q4-Tasman (Bruker), Milestone START D microwave digestor, Agilent ICP-MS 7800 (Model G8421A), Apreo Lovac FEGSEM (Thermo Fisher), OCTANE PRO EDS, Thermo Fisher XPS, Barnstead LED water purification, Panreac Applichem reagents — all are methods-side vendor identifications. Retained.
• Speciation note: The paper reports total As only. The arsenic values in Tables 5–7 are total arsenic; they cannot be propagated as inorganic arsenic (iAs) without independent speciation. Wiki pages that consume these values for As must label them as tAs, not iAs, and any iAs-based regulatory comparison (EU 2015/1006, FDA Closer-to-Zero rice action levels) requires the speciation caveat.
• Subtraction-of-baseline arithmetic (Table 6): The authors state that Table 6 values are the migration component after subtracting the raw-ingredient contributions measured in Table 5. The wiki preserves their published values without re-deriving the subtraction; readers needing the unsubtracted cooked-food totals would have to add the relevant Table 5 mass-weighted ingredient contributions back in. This subtraction is the standard cookware-attribution method; it is not a hidden assumption.
• Estimated daily intake (Table 9): The EDI calculation assumes a 100 g/day consumption portion. This is a single-population, single-portion assumption; the paper does not stratify by body weight or by realistic per-population intake. The “exceeds FAO/WHO 21 mg/day” finding for ACP-5 and APP-5 is correct under that assumption; readers applying it to other populations or portions should re-do the arithmetic from Table 6 values. The wiki preserves the published EDI figures and labels the 100 g/day assumption explicitly.
• FAO/WHO reference values quoted by the paper: Al 21 mg/day, Fe 18 mg/day, As 0.030 mg/day, Cd 0.058 mg/day, Pb 0.252 mg/day. The aluminum reference corresponds approximately to the JECFA PTWI of 2 mg/kg bw/week (≈ 1 mg/kg bw/day at 70 kg = 70 mg/week ≈ 10 mg/day) — the paper’s 21 mg/day figure is higher than the most recent JECFA derivation, but it is the figure cited by the authors. The wiki preserves the paper’s published threshold without re-deriving it; downstream synthesis pages should reconcile against efsa-aluminium-twi (EFSA TWI 1 mg/kg bw/week, which is roughly an order of magnitude stricter than the paper’s quoted 21 mg/day).
• SEM-EDS and XPS detection-limit framing: The paper’s statement that EDS detection is ~0.1 wt.% (≈1000 µg/g) and XPS detection is 0.1–1 at.% is published as a methodological finding. The conclusion that EDS and XPS are not suitable for quantifying these migration loads is the authors’ own; the wiki reports it as the source’s methodological note rather than as cross-source synthesis.
• Matrices vocabulary: cooked-rice and cooked-meat are corpus-convention matrices labels reflecting that the food experiments are 350 g rice + 500 g beef per condition (both ingredients are present in every condition); food-contact-materials covers the pot-composition characterisation; food-simulants covers the OM4 and OM5 Regulation (EU) No 10/2011 test arm.
• Ingredients-frontmatter scope: Lemon (the source uses “black lemon,” i.e. dried lime), cloves, and cardamom appear in Table 5 (raw-ingredient elemental data) but are not declared in the ingredients: frontmatter because no lemon, cloves, or cardamom page exists yet; their Table 5 values are preserved in the body and surfaced for any future freq-driven ingredient stub. The [[ingredients/spices]] declaration covers the spices-group context; specific spices that do have pages (cinnamon, black pepper) are declared individually.
• Folder placement note: This PDF was filed by the upstream sorter into 02_Honey_Syrups_Vinegar/ under the Kimi corruption-issue staging directory. The paper itself is about aluminum-cookware migration; “vinegar” appears only as acetic acid in the EU 10/2011 simulant. The folder location is preserved (immutable per Part 4) but is not load-bearing for routing.
• Funding-source disclosure: Saudi Standards, Metrology, and Quality Organization (SASO), General Department of Research and Studies, grant number 12-8-8. Reported.
• Data-availability disclosure: The paper states the data are available from the corresponding author on request and are not publicly available due to contract conditions with SASO. The values transcribed here are those published in Tables 4, 5, 6, 7, and 9 of the paper.
• Evidence tier: B-tier. Single-laboratory study with proper triplicate replication and food-simulant validation. The cookware-composition arm uses a CRM, but the ICP-MS food arm uses multi-element calibration standards verified by spiked samples (no certified food-matrix CRM reported). The 100 g/day EDI assumption is a single-population framing. The pot manufacturers are anonymised, which is appropriate for the wiki’s brand-firewall posture but limits cross-corpus reconciliation against other pot-leaching datasets.
• Audit subagent (2026-06-02, general-purpose fresh context) verdict REVISE; 0 ❌ findings, 2 ⚠️ concerns. Both concerns verified against repo state and found to be false positives, no content changes made: (1) Auditor flagged [[ingredients/salt]] as missing from the taxonomy snapshot — verified against wiki/ingredients/salt.md, which exists; the auditor’s snapshot was stale (auditor self-noted the possibility). The routing audit confirms 0 unresolved entries for this source. (2) Auditor flagged the four matrices labels (cooked-rice, cooked-meat, food-contact-materials, food-simulants) as not in the canonical matrices vocabulary — verified that the corpus-convention precedent for these labels is established in at least five prior source pages including bassioni2012-aluminum-foil-meat-cooking-uae (cooked-meat, food-contact-materials), dordevic2019-aluminum-foil-culinary-baking-czech, carey2015-rice-arsenic-percolating-cooking, efsa2015-nickel-food-water, and joyce2016-bush-meat-cooking-ghana. The routing audit accepts these labels with 0 malformed entries. The matrices-vocabulary note above already documents the corpus-convention status. Numerical fidelity (Tables 4, 5, 6, 7, 9), speciation discipline, brand-firewall posture, and Part 2 wiki/HMTc-firewall posture all audited clean. Effective post-verification verdict: PROMOTE.

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.