Amiri Qandashtani & Mohamadi Sani 2016 — As, Cd, Pb in 10 imported long-grain rice types on the Torbat-Heidarieh market, Iran

This survey reports total concentrations of arsenic (HGAAS, treated here as tAs absent any inorganic-species separation step), cadmium and lead in 210 packed long-grain (Indica) rice samples drawn from 10 imported rice types — three originating from India and seven from Pakistan — on the Torbat-Heidarieh retail market in 2015, plus a weekly-intake risk assessment under a 110 g/day per-capita rice-consumption assumption and a 60 kg adult-Iranian body-weight reference. Mean Pb concentrations across the 10-type panel exceeded the FAO/WHO Codex Alimentarius 0.2 mg/kg cereal limit in 9 of 10 types and exceeded the Iranian (ISIRI 12968) 0.15 mg/kg cereal limit in all 10 of 10 types; mean Cd exceeded the FAO/WHO 0.1 mg/kg cereal limit in 8 of 10 types; mean As exceeded the FAO/WHO 0.2 µg/g cereal limit in 4 of 10 types and exceeded the Iranian 0.15 µg/g limit in 5 of 10 types. The estimated weekly intake (EWI) of Pb computed from per-type means exceeded the JECFA Pb provisional tolerable weekly intake (PTWI, 25 µg/kg body weight per week) in 3 of 10 types; EWIs for Cd and As remained below their respective PTWI ceilings across the full panel.

Key numbers

Reference values cited by the paper for comparison:

PTWI reference values used for the risk assessment (µg/kg body weight per week, attributed by the paper to FAO/WHO Joint Expert Committee on Food Additives 2004 / FAO/WHO food standards programme):

Method validation parameters (Table 1, p. 2; values as printed in the paper, expressed in µg/g):

Per-type mean ± SD concentrations across the 10-type panel (Table 2, p. 3; µg/g on dry-weight basis):

Notes on the per-type SD structure: per-type SDs across the panel ranged from ±0.005 to ±0.017 for As (across the nine non-outlier types), ±0.008 to ±0.039 for Cd, and ±0.013 to ±0.223 for Pb. One Pakistan-origin type carries an As SD of ±0.20 in the published Table 2, a 10-fold outlier from the other nine As SDs and almost certainly a typographic error for ±0.020; see Verification notes.

Exceedance pattern (paper’s framing, Table 2 and prose, pp. 3-4):

• Mean Pb exceeded the FAO/WHO 0.2 mg/kg cereal ceiling in 9 of 10 types (3 of 3 Indian-origin and 6 of 7 Pakistani-origin; only one Pakistani-origin type at 0.199 mg/kg fell just below). Mean Pb exceeded the Iranian 0.15 mg/kg cereal ceiling in all 10 of 10 types. The three highest per-type Pb means (1.999, 1.961, 1.827 mg/kg) all fell in the same band and the prose identifies them by name as the top tier; under strict Part 12 reading the wiki page reports them as “the three highest-Pb types of the 10-type panel” without name attribution.
• Mean Cd exceeded the FAO/WHO 0.1 mg/kg cereal ceiling in 8 of 10 types; the two below-ceiling means (0.085 and 0.098 mg/kg) were both in the Pakistani-origin group. The three highest Cd means (0.385, 0.295, 0.242 mg/kg) span both origins. The paper does not cite an explicit Iranian Cd cereal ceiling for direct comparison.
• Mean As (HGAAS, treated as total) exceeded the FAO/WHO 0.2 µg/g cereal ceiling in 4 of 10 types (2 of 3 Indian-origin at 0.206 and 0.225 µg/g, and 2 of 7 Pakistani-origin at 0.243 and 0.314 µg/g). The panel mean of 0.161 µg/g fell below the FAO/WHO ceiling but above the Iranian 0.15 µg/g ceiling. Five of 10 types exceeded the Iranian 0.15 µg/g ceiling (3 Indian-origin and 2 Pakistani-origin).

Estimated weekly intake (EWI) across the 10-type panel (Table 3, p. 3; µg/kg body weight per week, computed as EWI = C_rice × (WI_rice / BW) with WI_rice = 770 g/week per capita (110 g/day × 7) and BW = 60 kg, per the Institute of Standards and Industrial Research of Iran 2010 assumption):

EWI exceedances vs PTWI:

• Pb EWI exceeded the 25 µg/kg bw/week PTWI in 3 of 10 types (1 Indian-origin at 25.66 and 2 Pakistani-origin at 25.54 and 25.17 µg/kg bw/week). An additional Indian-origin type fell at the PTWI ceiling itself (24 µg/kg bw/week, rounded as published in Table 3).
• Cd EWI was below the 7 µg/kg bw/week PTWI for all 10 types (panel maximum 4.60).
• As EWI was below the 15 µg/kg bw/week PTWI for all 10 types (panel maximum 4.03).

Methods (brief)

Sample collection. 210 packed long-grain (Indica) rice samples were drawn from 10 imported rice types available in retail stores in Torbat-Heidarieh city (Razavi Khorasan province, northeastern Iran) in 2015. Three types originated from India and seven from Pakistan; each retail pack weighed 15-20 kg. A 250 g aliquot was taken from each pack, sieved through 50 mesh (<0.30 mm), sealed in a plastic box, and stored at room temperature until analysis. The paper notes that production area was usually only vaguely described on the packaging (“India” or “Pakistan” without further geographic detail).

Sample preparation — arsenic. Rice samples were turned into ash with nitric acid and evaporated to dryness. The ash was then heated by a gradual temperature increase to 425 °C, re-dissolved in hydrochloric acid, and presented to the hydride generation atomic absorption spectrometer (HGAAS) for As determination at 193.7 nm absorption band, per CEN EN 14546:2005 (“Foodstuffs - Determination of trace elements - Determination of total arsenic by hydride generation atomic absorption spectrometry (HGAAS) after dry ashing”).

Sample preparation — cadmium and lead. Dry ashing per AOAC Official Method 999.11 (Jorhem 2000). Rice samples were dried at 105 °C for 48 h, then ashed at 450 °C with a gradual temperature ramp (≤50 °C/h). 6 M HCl (1+1) was added; the solution was evaporated to dryness; the residue was dissolved in 0.1 M HNO₃. The digested solution was analysed for Pb and Cd by flame and graphite furnace atomic absorption spectrometry on a Young Lin model 8020 (YL Instrument Co., Anyang, Republic of Korea). All samples were digested and analysed in triplicate. Concentrations were expressed as mg/kg on dry-weight basis (per Jorhem 2000).

Instrumentation and reagents. Chemicals and reagents from Merck (Darmstadt, Germany) at maximum purity; high-purity de-ionised water; certified reference material SRM 1568 (rice flour, National Institute of Standards and Technology, Gaithersburg, MD, USA). Quality control included reagent blanks, repeated samples, and the SRM 1568 CRM. Per Table 1 the As recovery was 93 % (certified 0.285 vs measured 0.251 µg/g); the Pb recovery was 95 % (certified 0.008 vs measured 0.0057 µg/g); the Cd recovery was 97 %, but the certified and measured values printed in Table 1 (118.4 and 111.55 µg/g) are inconsistent with the published NIST SRM 1568 / 1568a / 1568b rice-flour Cd certified values (which are on the order of 0.022 µg/g) — see Verification notes; this is a paper-internal anomaly, not a wiki-side transcription issue.

Speciation. Total As by HGAAS without arsenic-species separation; the wiki page uses tAs (total arsenic) rather than iAs per CLAUDE.md Part 14. Total Cd and total Pb by AAS. No mercury panel, no As speciation, no chromium speciation.

Statistical analysis. SPSS version 23. Duncan method for comparison of means. Sample t-test for comparison of mean values against available standards. Significance level p < 0.05.

Risk assessment. EWI = C_rice × (WI_rice / BW), with C_rice as the per-type mean (mg/kg dry weight), WI_rice as the weekly intake of rice = 110 g/day × 7 = 770 g/week per the ISIRI 2010 per-capita assumption, and BW = 60 kg average Iranian adult body weight. EWI values are reported in Table 3 in µg per kg body weight per week. PTWI reference values (15, 7, 25 µg/kg bw/week for As, Cd, Pb respectively) were attributed by the paper to FAO/WHO Joint Expert Committee on Food Additives 2004 / FAO/WHO food standards programme.

Implications

This source contributes Iranian retail-market occurrence data for total As, total Cd, and total Pb in imported long-grain (Indica) rice on a single-city (Torbat-Heidarieh) panel of 10 rice types in 2015, plus a worked single-commodity weekly-intake risk assessment under a defined 770 g/week per-capita and 60 kg body-weight Iranian adult assumption. Principal contributions to the wiki evidence pool:

• A Pb signal in imported long-grain rice with per-type means spanning 0.199-1.999 mg/kg across the 10-type panel and 9 of 10 type-means exceeding the Codex Alimentarius 0.2 mg/kg cereal ceiling (all 10 of 10 exceed the Iranian 0.15 mg/kg ceiling). The 3 highest type-means (1.999, 1.961, 1.827 mg/kg) cluster nearly an order of magnitude above the Codex ceiling. The Pb-in-rice band falls broadly within the range reported by the parallel Iranian cereal-market work pirsaheb2015-iran-kermanshah-cereals-metals (Kermanshah-market rice mean 1.35 mg/kg, 2014 sampling — Roya 2016 cites this paper as “Pirsaheb et al. 2016” per its print issue year), and the paper itself also cites prior Iranian work (Mirlohi et al. 2013; Naseri et al. 2015) in the same band.
• A Cd signal with per-type means spanning 0.085-0.385 mg/kg, 8 of 10 type-means exceeding the Codex 0.1 mg/kg cereal ceiling. The Cd mean range and exceedance frequency are higher than the Kermanshah-market rice Cd mean reported by the same parallel Iranian work (0.055 mg/kg, pirsaheb2015-iran-kermanshah-cereals-metals), and notably higher than several non-Iranian retail-market datasets the paper cites (Taiwan, South Korea, Sweden, Brazil, Nigeria — per-paper means in the 0.003-0.035 mg/kg band).
• A total-As (HGAAS, no species separation) signal with per-type means spanning 0.048-0.314 µg/g and 4 of 10 type-means exceeding the Codex 0.2 µg/g cereal ceiling. The panel mean of 0.161 µg/g sits just above the Iranian 0.15 µg/g ceiling.
• An origin-stratified contrast: the 3-type Indian-origin subgroup carries higher per-type Pb (range 1.430-1.999) and higher per-type Cd (range 0.221-0.385) means than the 7-type Pakistani-origin subgroup (Pb 0.199-1.961, Cd 0.085-0.295), though the Pakistani-origin subgroup contains the panel maximum Pb-mean and Cd-mean values. The paper does not attempt to attribute the origin contrast to specific cultivation, water, soil, or post-harvest pathways and explicitly notes that production area was vaguely described on the packaging.
• A 3-of-10 Pb-EWI-above-PTWI signal under the single-commodity 770 g/week assumption, including one Indian-origin type at 25.66 and two Pakistani-origin types at 25.54 and 25.17 µg/kg bw/week. The single-commodity Pb-EWI range for this Torbat-Heidarieh-market panel (2.87-25.66 µg/kg bw/week) brackets the single-EWI value reported by the parallel Kermanshah-market rice work (17.32 µg/kg bw/week, pirsaheb2015-iran-kermanshah-cereals-metals).
• A small-panel (n = 10 brands), single-city, single-year, single-cohort (adult 60 kg) sampling footprint with no As-species separation, no Hg panel, no Cr panel, and several paper-internal data-quality anomalies (see Verification notes). The B-tier classification reflects these limitations; the source should pool with other Iranian-market rice work for occurrence-distribution purposes rather than anchor any standalone characterisation.

Wiki pages this source may touch

• arsenic-total
• cadmium
• lead
• rice
• rice-bulk-grain

Verification notes

• Author-name handling and cite-key choice. The PDF byline reads “Amiri Qandashtani Roya & Mohamadi Sani Ali”; under Persian academic-name convention these resolve to first author Roya Amiri Qandashtani (family name “Amiri Qandashtani”, given name “Roya”) and corresponding author Ali Mohamadi Sani (family name “Mohamadi Sani”). The manual-fetch handle MFK_roya2016 was assigned from the PDF filename and uses the given name. The wiki cite-key roya2016-iran-torbat-heidarieh-rice-metals preserves the handle-alignment convention; the authors array carries the canonical “family given” order. No DOI or first-author-surname grep returned an existing wiki page (three identity checks all clean).
• Frontmatter discipline. All ingredient, product, metal, matrix, and jurisdiction slugs verified against the 2026-05-18 taxonomy snapshot (docs/gpt-collaboration/taxonomy-snapshot.md). Metals use the Part 14 abbreviation vocabulary (tAs, Cd, Pb). IN (India) and PK (Pakistan) jurisdictions reflect the origin countries of the rice types; IR (Iran) is the sampling-and-consumption jurisdiction.
• Speciation discipline (Part 14). HGAAS without an inorganic-species separation step measures total arsenic; the paper does not separate iAs from tAs. The metals: field accordingly uses tAs, not iAs. No mercury panel; no Cr panel.
• Brand firewall (Part 12, strict reading locked 2026-05-17). The paper names 10 commercial rice types — three of Indian origin (one a basmati cultivar designation, two commercial brand names) and seven of Pakistani origin (all commercial brand names). Under the strict 2026-05-17 reading of Part 12, no per-type brand or cultivar identifier was carried into the wiki page tables, narrative, or these verification notes. Per-type values are aggregated by country-of-origin subgroups (India n=3, Pakistan n=7) and the panel-level ranges and means are reported. The three highest-Pb types and the three highest-Cd types named in the paper’s prose are described in the wiki page as “the three highest-Pb types of the 10-type panel” / “the three highest-Cd types” without name attribution. Scientific-method vendor and material identifiers (YL Instrument Co. Young Lin model 8020, Merck reagents, NIST SRM 1568) are retained per Part 12 Exception 2.
• Wiki/HMTc firewall (Part 2). No threshold proposals, no consumer-audience advisories. Cross-paper comparisons in the Implications section are limited to descriptive band-placement against pirsaheb2015-iran-kermanshah-cereals-metals — the same paper Roya 2016 cites in its reference list as “Pirsaheb et al. 2016” (the wiki cite-key uses the online-publication year 2015 while Roya cites the print-issue year 2016; DOI 10.1080/19393210.2015.1099570 confirms the duality). Full cross-paper synthesis (e.g., updating the rice contamination_profile block) is the Part 9 workflow’s job and is not done in this ingest pass.
• Reference-value attribution. The paper cites the FAO/WHO Codex Alimentarius 0.2 mg/kg Pb and 0.2 µg/g As cereal ceilings (attributed to FAO/WHO 2004) and the 0.1 mg/kg Cd cereal ceiling (attributed to FAO/WHO 2004 and ISIRI 2010). The paper cites the Iranian national-standards 0.15 mg/kg Pb and 0.15 µg/g As cereal ceilings (attributed to ISIRI 12968, 1st ed., 2010). The paper does not explicitly cite an Iranian Cd cereal ceiling. These attributions are reported as the paper publishes them.
• Paper-internal data anomalies catalogued (none rise to “stop-condition” severity, all noted for the record).
  1. Table 1 Cd certified-value implausibility. Table 1 lists the SRM 1568 rice-flour Cd certified value as 118.4 µg/g and the measured value as 111.55 µg/g, computing a 97 % recovery. The published NIST SRM 1568 / 1568a / 1568b rice-flour Cd certified value is on the order of 0.022 µg/g (≈ 22 ng/g). The 118.4 / 111.55 values are very likely a typographic error in the paper (possibly the rows were mis-aligned with another CRM’s Cd, or possibly the values are ng/g and the column header is wrong). The 97 % recovery ratio (111.55 / 118.4 = 0.942 → 94.2 %, not 97 %) is also internally inconsistent. The wiki page reports the values verbatim as Table 1 publishes them and flags the anomaly here.
  2. Pb-range prose vs Table 2. The Results-and-discussion prose (p. 3) states “Pb levels in imported rice samples varied between 0.76-2.00 mg/kg.” Table 2 publishes per-type Pb means ranging from 0.199 ± 0.013 µg/g (one Pakistani-origin type) to 1.999 ± 0.182 µg/g (one Indian-origin type), so the 0.76 lower bound does not correspond to any per-type mean in Table 2. The 0.76 figure may refer to a within-type per-sample minimum that the paper does not otherwise tabulate, or may be a transcription error. The wiki page reports the per-type-mean range from Table 2 (0.199-1.999) and notes the discrepancy.
  3. Pakistani-origin As SD outlier. Table 2 shows one Pakistani-origin type’s As mean ± SD as 0.243 ± 0.20 µg/g. The other nine per-type As SDs across Table 2 range from ±0.005 to ±0.017; a SD of ±0.20 (82 % of the mean) is a 10-fold outlier from the rest of the table and is almost certainly a typographic error for ±0.020. The wiki page does not propagate this single SD; the aggregated panel range and per-origin ranges are unaffected by the SD typo.
  4. Indian-origin Pb EWI rounding. Table 3 publishes one Indian-origin type’s Pb EWI as “24” (no decimal). Computed from Table 2’s per-type Pb mean of 1.827 µg/g and the EWI formula (1.827 × 770 / 60 = 23.45), the value should be 23.45 µg/kg bw/week, not 24. Other Table 3 EWI values match the arithmetic to two decimal places (e.g., 1.999 × 770 / 60 = 25.66 ✓). The wiki page reports “24” as published in Table 3 and notes this minor rounding/transcription inconsistency.
• Basis. All per-type concentrations and EWI values are on a dry-weight basis as the paper describes (samples dried at 105 °C for 48 h prior to ashing). The EWI calculation applies these dry-weight concentrations against a 770 g/week consumption mass; the paper does not adjust for fresh-vs-cooked rice mass and the wiki page reports the EWI table as the paper publishes it without basis-conversion adjustment.
• Sampling structure. The paper reports 210 total packed samples from 10 types but does not publish the per-type sample count (whether 21 packs per type or some other distribution). Triplicate digestions per sample. The per-type means in Table 2 are accordingly the means across all per-type packs analysed.
• Data integrity. All values in the Key numbers section transcribed from Table 1 (CRM accuracy), Table 2 (per-type means ± SD, µg/g dry weight), Table 3 (EWI summary in µg/kg body weight per week), and the prose on pp. 2-4 (panel mean As 0.161 µg/g, exceedance counts, regulatory comparators). Per-origin aggregations and panel-level ranges/means computed from Table 2 directly; the panel-level Cd mean (~0.193) and panel-level Pb mean (~0.925) are not published by the paper and are reported here as derived aggregates. No paper-internal contradictions beyond the four anomalies catalogued above were identified.
• Audit-application notes (2026-06-01, Claude audit subagent, general-purpose, fresh context). Subagent verdict REVISE. Seven findings applied, zero findings rejected, one finding pushed back on with partial correction documented:
  • Check 1 ❌ #1 applied. Opening narrative Pb-exceedance count corrected from “6 of 10 types” to “9 of 10 types” against FAO/WHO 0.2 mg/kg and “10 of 10 types” against ISIRI 0.15 mg/kg. Verified against Table 2: only one Pakistani-origin type at 0.199 µg/g fell below 0.2 mg/kg; all 10 type-means exceeded 0.15 mg/kg. The Exceedance pattern bullet and the Implications first bullet were updated to match.
  • Check 1 ❌ #2 applied with corrected count. Opening narrative and Implications As-exceedance count corrected from “1 of 10 types” to “4 of 10 types” against FAO/WHO 0.2 µg/g. The audit subagent caught the wiki error and proposed “2 of 10” but the correct count from Table 2 is 4 of 10: two Indian-origin types (0.206, 0.225 µg/g) and two Pakistani-origin types (0.243, 0.314 µg/g). Verified independently against Table 2 before applying.
  • Check 1 ❌ #3 applied. Per-type As SD range in the Key numbers SD-structure paragraph corrected from “±0.005 to ±0.039” to “±0.005 to ±0.017 (across the nine non-outlier types)” — verified against Table 2 (max non-outlier As SD is the 1121-type at ±0.017; the ±0.039 figure was a Cd-SD transposition). The Verification-notes anomaly #3 paragraph already stated the correct range, so the body was the discrepant value.
  • Check 4 ❌ #1 applied. Verification-notes brand-firewall paragraph rewritten to remove the verbatim brand-name quote “(Del Mobarake, 1121 and Kimia brands)“. The redaction-explanation now describes the policy without re-naming the brands it strips, consistent with the pirsaheb2015-iran-kermanshah-cereals-metals precedent.
  • Check 4 ❌ #2 applied. Anomaly note 3 heading renamed from “Del-mobarake-origin As SD outlier” to “Pakistani-origin As SD outlier” (origin descriptor retained, brand name removed). Body unchanged (already used “one Pakistani-origin type” without naming).
  • Check 4 ❌ #3 applied. Anomaly note 4 heading renamed from “Kimia-origin Pb EWI rounding” to “Indian-origin Pb EWI rounding” (origin descriptor retained, brand name removed). Body unchanged.
  • Check 5 ⚠️ partial pushback with reframing. Subagent flagged the two cross-source comparisons to pirsaheb2015-iran-kermanshah-cereals-metals as wiki-side synthesis on the premise that “Roya 2016 does not cite Pirsaheb 2015.” Verified against Roya 2016’s reference list (p. 5): the paper cites “Pirsaheb M, Fattahi N, Sharafi K, Khamotian R, Atafar Z. 2016. … Food Addit Contam Part B. 9:15-20” — the same paper as wiki cite-key pirsaheb2015 (DOI 10.1080/19393210.2015.1099570; online publication year 2015, print issue year 2016). The wikilink target IS source-cited. However, the broader Part 2 point about quantitative comparative judgement (“a stronger single-commodity Pb-exposure signal than”) is fair and was softened to descriptive band-placement language (“brackets the single-EWI value reported by the parallel Kermanshah-market rice work”). The “consistent with prior Iranian work” phrasing was reframed to attribute the comparison to the source’s own reference list. The wikilinks were retained. The Wiki/HMTc firewall verification note was updated to acknowledge the cite-year duality.
  • Audit Checks 2 (slug vocabulary) and 3 (speciation and methods) were clean per the subagent’s report and required no changes.

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.