Kurniawati et al. 2021 — Heavy metals in Jakarta staple foods by neutron activation analysis

This conference-proceedings paper reports cobalt (Co), total chromium (Cr), total mercury (tHg), and zinc (Zn) concentrations in 14 staple food commodities purchased from five traditional markets across the five administrative regions of Jakarta, Indonesia, using neutron activation analysis (NAA). Lead, cadmium, and inorganic arsenic — the priority analytes for most food-safety regulation — were not measured. Several commodities reported Hg concentrations within or above Indonesia’s SNI 7387:2009 maximum permissible limit (MPL) of 0.03–0.5 mg/kg for Hg, with crackers (kerupuk) recording the highest single concentration at 0.25 mg/kg wet weight. The Health Risk Index (HRI) calculated by the authors using Indonesian SUSENAS consumption data placed rice Hg at HRI = 2.1 (above the safe threshold of 1.0), driven primarily by the very high daily rice consumption rate of 240 g/person/day rather than by per-gram Hg concentration. All other food-metal combinations in the study returned HRI < 1.

Key numbers

Concentration ranges across all commodities (wet weight, mg/kg, Table 3)

• Co: 0.01 (papaya) to 0.15 (crackers) mg/kg
• Cr: 0.02 (egg, water spinach) to 0.49 (meat) mg/kg
• tHg: 0.02 (meat) to 0.25 (crackers) mg/kg
• Zn: 1.87 (papaya) to 46.2 (soybean) mg/kg

Full commodity table — Co, Cr, tHg, Zn (mean ± SD, wet basis, mg/kg, Table 3)

Water content (% wet weight, Table 2)

soybean 9.81 ± 0.42; mung bean 9.37 ± 2.12; rice 11.2 ± 1.00; tempeh 67.6 ± 7.23; tofu 81.0 ± 0.97; papaya 89.9 ± 3.51; crackers 4.26 ± 0.68; water spinach 94.6 ± 1.70; spinach 93.4 ± 1.36; carrot 88.1 ± 5.17; egg 78.0 ± 0.73; mackerel 74.1 ± 2.39; chicken 76.1 ± 2.18; meat 75.3 ± 1.39. Authors note these are consistent with USDA composition data.

Reference oral doses used in HRI math (Table 1, mg/kg·day)

• Zn — 0.3 (FAO/WHO 2013)
• Cr — 1.5 (FAO/WHO 2014)
• Co — 0.043 (Chauhan & Chauhan 2014)
• Hg — 0.0003 (USEPA IRIS 2008)

Indonesian SUSENAS-based consumption rates used in DIR (Table 5, kg/day)

Rice 0.2400; Tempeh 0.0220; Tofu 0.0220; Egg 0.0186; Spinach 0.0104; Water spinach 0.0119; Chicken 0.0118; Papaya 0.0076; Carrot 0.0027; Mackerel 0.0020; Meat 0.0014; Crackers 0.0006; Mung bean 0.0004; Soybean 0.0001. Adult body weight assumed 55.9 kg (Orisakwe et al. 2012). Children 0–6 y exposure = 1/3 of adult per Guerra et al. 2012.

Maximum permissible limits the paper compares against (Table 4, mg/kg)

• Cr — 0.5–2.0 (China GB 2762-2012)
• Zn — 50–250 (Vietnam Technical Regulations on Mycotoxin and Heavy Metals MRLs)
• Co — 1.5 (Carmen Valadez-Vega et al. 2011, cited as a regulatory benchmark by the authors)
• Hg — 0.03–0.5 (Indonesia SNI 7387:2009, Badan Standardisasi Nasional)

Cr, Zn, and Co concentrations were below their respective MPLs across all commodities. Hg concentrations in several commodities (rice 0.15; soybean 0.22; mung bean 0.19; tempeh 0.14; crackers 0.25; mackerel 0.18) fell within or above the 0.03 mg/kg lower bound of the SNI Hg range; the soybean, crackers, and mackerel values all exceed the 0.03 mg/kg lower bound but remain below the 0.5 mg/kg upper bound.

Health Risk Index values (Table 6)

Reported HRI ≥ 1 (i.e., not safe per the authors’ own threshold):

• Rice Hg HRI = 2.1 (Table 6 reports 2.1E+00)

All other HRI cells in Table 6 are < 1. Selected non-rice values: tempeh Hg 0.18; tofu Hg 0.060; spinach Hg 0.045; egg Hg 0.048; rice Zn 0.32; tempeh Zn 0.013; tofu Zn 0.016; spinach Zn 0.0063; rice Co 0.0091; meat Hg 0.0016; crackers Hg 0.0083.

Methods (brief)

Analytical method: neutron activation analysis (NAA) using the rabbit-system facility of the G.A. Siwabessy multipurpose reactor at Serpong, Indonesia (operator: BATAN, the National Nuclear Energy Agency). Thermal neutron flux 10¹³ n·cm⁻²·s⁻¹. Samples were counted on a CANBERRA HPGe gamma spectrometer (~60,000 s acquisition time) after one-month decay; Genie 2000 software handled spectrum analysis. NAA reports total elemental content with no speciation; the page therefore reports tHg (total mercury) and total Cr, not MeHg or Cr-VI.

Standards: 100 µL ICP multi-element standard (E. Merck) dried and sealed in polyethylene vials served as the calibration standard.

Quality control: SRM NIST 1567a Wheat Flour and SRM NIST 8414 Bovine Muscle Powder were analyzed alongside samples. Reported recoveries 103–104%; reported coefficient of variance (CV) 4.06–8.42%. Replicate counts per commodity, sample sizes per market, and per-element LODs are not reported in the published paper.

Health Risk Index: HRI = DIR/RfD, where DIR = (C_metal × Q_intake)/body weight. Adult body weight 55.9 kg; children’s intake set at 1/3 of adult. HRI threshold for safety set at < 1 per the authors’ citation of Khan et al. 2008 and WHO/FAO 2013.

Key methodological scope limitation for the wiki: NAA in this study measured only Co, Cr, Hg, and Zn. Pb, Cd, and iAs — the three contaminants most central to food-safety regulation and HMTc certification — are not in the panel.

Implications

Certification: Indonesian-market tHg occurrence data for a defined panel of staple foods and proteins. The rice tHg HRI = 2.1 finding is a single-study signal that the HMTc rice-row literature evidence base for Hg can incorporate; HMTc certification thresholds for Hg in rice are not derivable from this single source. Pb and Cd were not measured, so the source contributes nothing to the priority-analyte evidence base for these matrices.

Courses: Useful teaching case for how high-volume staple consumption (rice at 240 g/day in Indonesia) can drive an HRI above 1 even at moderate per-gram concentrations. Also a worked example of NAA as an alternative to ICP-MS in settings with reactor access but limited mass spectrometry infrastructure, including the consequent loss of speciation (no iAs vs tAs, no MeHg vs tHg, no Cr-VI vs Cr).

App: tHg, Co, Cr, and Zn data for Indonesian-market staple foods, with the caveats that (a) replicate counts per commodity are unreported, (b) no Pb/Cd/iAs data, and (c) all values are wet basis as marketed.

Verification notes

Enhancement run 2026-05-18 (cite-key existed before 2026-05-14 with several defects). Defects corrected against the source PDF:

• Expanded metals: from [Cr, tHg] to [Cr, tHg, Co, Zn] — Co and Zn are explicitly measured for all 14 commodities and reported in Table 3 alongside Cr and Hg. The prior frontmatter omitted half the studied panel.
• Corrected ingredients/soybean (invented slug, not in taxonomy) → ingredients/soy (current taxonomy slug). Added [[ingredients/fish]], [[ingredients/chicken]], [[ingredients/meat]] since the paper measures mackerel, chicken, and meat as distinct commodities.
• Expanded products: from [fresh-fish] to [fresh-fish, snacks-crackers-biscuits] — the paper measures kerupuk (Indonesian crackers) as a distinct commodity with the second-highest Hg concentration in the panel (0.25 mg/kg), which is product-grade information. Both target pages exist on disk.
• Replaced matrices: [staple-food, rice, vegetables, legumes, meat, fish, egg] (mostly non-vocabulary) with [rice, legumes, leafy-vegetable, root-vegetable, fruit, fish, egg, meat, poultry, snack-cracker] aligned with the system-prompt matrix vocabulary and reflecting all 14 commodities. Free-form matrix labels are accepted per the routing-layer convention.
• Replaced the Cr-min citation in concentration ranges — Table 3 shows Cr = 0.02 in both egg and water spinach; the prior page named only egg.
• Corrected the tHg-min citation — Table 3 shows tHg = 0.02 in meat (papaya is 0.03); the prior page incorrectly attributed the tHg minimum to papaya.
• Expanded ## Key numbers to include the full Table 3 commodity matrix (all 14 rows, all four metals), the Table 2 water-content list, the Table 1 RfDs, the Table 5 SUSENAS consumption rates, the Table 4 MPLs, and an explicit Table 6 HRI summary so the page no longer relies on selective commodity excerpts. Prior key-numbers excerpts for rice/soybean/spinach/mackerel/crackers all matched the source and were preserved.
• Removed the cross-source synthesis claim in the ## Implications Certification line that said the rice Hg HRI result “is at odds with the general consensus that rice Hg is low risk globally.” Per Part 2 the source page reports what this one paper found, not how it relates to other literature. The replaced text now frames the rice Hg HRI as a single-study signal that the HMTc rice-row Hg evidence base can incorporate, without proposing a threshold or comparing to consensus.
• Replaced the legacy heading ## Wiki pages updated on ingest with the current ## Wiki pages this source may touch per the 2026-05-08 schema convention.
• Added access_url: https://doi.org/10.1063/5.0053376 (DOI resolves to AIP’s free-access landing page; the PDF on page 1 carries a FREE access tag; license string remains unknown because the AIP Conference Proceedings copyright/license terms are not printed on the article PDF).
• Tightened sample_population to enumerate the five Jakarta markets named in the paper (Cilincing, Tomang, Klender, Mampang, Pasar Senen) and to flag the unreported per-commodity replicate counts.

Paper-internal inconsistency flagged (not a routing or wiki defect): the Conclusions section asserts that “the foods from Jakarta are safe for consumption” and the discussion text claims the HRI calculation for all commodities shows values < 1. Table 6, however, prints rice tHg HRI as 2.1E+00 (i.e., 2.1, above the authors’ own < 1 safe threshold) and recomputing DIR/RfD from the paper’s own inputs (rice tHg 0.15 mg/kg × 0.24 kg/day ÷ 55.9 kg ÷ 0.0003 mg/kg·day) returns 2.14. The Table 6 value is internally consistent with the methodology; the Conclusions and discussion overstate safety. The wiki page reports the data as the authors printed them and flags the inconsistency here.

Preserved fields: cite_key, raw_handle, raw_path, source_type, evidence_tier, license, near_duplicates, sample_n, jurisdictions: [ID].

Wiki pages this source may touch

• chromium
• mercury-total
• cobalt
• zinc
• rice
• soy
• spinach
• carrot
• eggs
• fish
• chicken
• meat
• fresh-fish
• snacks-crackers-biscuits

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.