Chen et al. 2022 — Multi-element concentrations in Chinese peanuts by ICP-MS

This peer-reviewed primary study from the Shandong Academy of Agricultural Sciences measured 19 macro, micro and toxic elements in 66 peanut samples from six major Chinese peanut-producing provinces (Jilin, Liaoning, Hebei, Henan, Guangxi, Guangdong). The primary aim was geographical characterization via linear discriminant analysis (LDA); the secondary aim was a baseline toxic-element occurrence dataset for Chinese peanuts. After microwave digestion with HNO3 and H2O2, elements were quantified by ICP-MS (iCAP Q, Thermo) in kinetic-energy-discrimination mode against the GBW10013 (soybean) certified reference material (recoveries 91.4–109.1%; R² > 0.9993). All toxic-element concentrations were below the Chinese GB2762-2017 maximum levels (Cd 0.5 mg/kg, Cr 1 mg/kg in peanuts) and the Australia/New Zealand Food Standards Code limit for Cd (0.5 mg/kg, revised in 2009). Cadmium was the most prominent toxic metal because peanuts are geocarpic (fruits develop underground in direct soil contact), with the highest provincial mean in Liaoning (0.204 mg/kg) and the maximum observed individual sample also in Liaoning. The authors do not speciate arsenic (total As reported) and do not measure mercury.

Key numbers

Toxic element concentrations by province (mg/kg fresh weight, mean ± SD; Table 3, p. 16793; shared lowercase superscripts indicate groups not differing under Duncan’s multiple range test at p < 0.05; “ND” = not detected):

ANOVA across provinces: Cd p < 0.05 (regional differences significant); Pb p = 0.516, Cr p = 0.263, As p = 0.077 (not significant). Min/max Cd observed: 0.047 mg/kg (Hebei) and 0.204 mg/kg (Liaoning). Authors note Cd contamination in Chinese peanuts was higher than some earlier studies and is the main risk point because of the underground-fruit contact pattern.

Nickel concentrations (mg/kg fresh weight, Table 3): Jilin 3.08 ± 1.21 ab, Liaoning 7.53 ± 1.17 c, Henan 5.87 ± 6.11 bc, Hebei 1.34 ± 1.49 a, Guangxi 8.73 ± 1.77 c, Guangdong 8.33 ± 3.70 c. Ni was positively correlated with Mn (r = 0.908, p < 0.01) and Co (r = 0.900, p < 0.01) and negatively correlated with Mo (r = −0.858, p < 0.01).

Aluminum concentrations (mg/kg fresh weight, Table 3): Jilin 86.4 ± 29.9, Liaoning 113 ± 23, Henan 119 ± 93, Hebei 94.7 ± 31.2, Guangxi 75.5 ± 4.3, Guangdong 64.5 ± 30.3. No significant regional difference (p = 0.222).

Macro elements (mg/kg fresh weight, Table 3, abbreviated): K 6034–6743 (no significant province difference); Mg 1613–1765; Ca 436 (Guangxi, lowest, attributed to acidic red soil) to 664 (Jilin, highest); Na 6.32 (Liaoning) to 62.7 (Hebei).

Method validation (Table 2): LOD (mg/kg) — Al 0.036, As 0.0007, Cd 0.0001, Cr 0.001, Ni 0.005, Pb 0.003. LOQ (mg/kg) — Al 0.109, As 0.0021, Cd 0.0003, Cr 0.003, Ni 0.016, Pb 0.009. Recoveries vs GBW10013 soybean CRM — As 91.4%, Cd 100.0%, Cr 96.8%, Ni 103.0%, Pb 92.9%, Al 96.7%. R² ≥ 0.9993 for all calibrations.

Regulatory context cited by authors: China GB2762-2017 maximum level for Cd in peanuts = 0.5 mg/kg; Australia New Zealand Food Standards Code revised Cd peanut limit from 0.1 to 0.5 mg/kg in 2009; China GB2762-2017 ML for Cr in peanuts = 1 mg/kg. All sample means in this dataset are below these cited limits.

Classification model: Linear discriminant analysis on all 19 elements gave 97.0% correct original classification and 78.8% after leave-one-out cross-validation (Table 5).

Methods (brief)

Sampling: 66 peanut samples purchased in lots of at least 3 kg from acquisition stations near peanut-harvest land in six provinces; ground in a homogenizer (Midea MJ-BL80Y21) at 18,000 rpm for 5 min.

Digestion: 0.3 g sample weighed into tetrafluoroethylene digestion vessel, 6 mL HNO3 (Thermo Fisher Scientific, 67–70%, trace-metal grade) and 2 mL H2O2 (Sinopharm Chemical Reagent, 30%, analytical reagent) added, held overnight at room temperature; microwave digestion (MARS5, CEM, USA) ramped to 120 °C / 150 °C / 190 °C with 5- to 20-minute holds; brought to 25 mL final volume with ultrapure water (Milli-Q, 18 MΩ·cm).

Quantitation: ICP-MS (iCAP Q, Thermo, USA) in kinetic-energy-discrimination (KED) mode, 1550 W RF, argon plasma (99.999%), Ni / Pb / Cd / Cr / Al / As measured at isotopes 60Ni, 208Pb, 111Cd, 52Cr, 27Al, 75As (Table 1). Stock standards from Inorganic Ventures (Lakewood, NJ). Triplicate measurement per sample; values are means. Blanks subtracted; LODs computed as 3 × SD of blanks.

QA/QC: Certified reference material GBW10011 (soybean) from the Institute of Geophysical and Geochemical Exploration (China) used as the digestion CRM; agreement vs GBW10013 certified values (Table 2 in paper) used to validate accuracy and precision; recoveries 91.4–109.1%. Statistical analysis with SPSS 23.0 (one-way ANOVA, Duncan’s multiple range test, Pearson correlation, stepwise LDA with Wilks’ lambda).

Speciation and limitations: arsenic reported as total As (no inorganic/organic speciation); mercury not measured; chromium reported as total Cr (no Cr-VI speciation); values are on fresh-weight basis; per-province n ranges from 5 (Guangxi) to 19 (Henan), limiting statistical power for the southern provinces; the paper notes its own per-region n is small and recommends larger sample sets for future work.

Implications

• Certification: peanut-specific occurrence dataset for Cd, Pb, Cr, Ni, Al and total As from six Chinese growing regions, covering 66 samples on fresh-weight basis. All toxic-element means fall below the China GB2762-2017 MLs (Cd 0.5 mg/kg, Cr 1 mg/kg) and below the Australia/New Zealand 2009 Cd limit (0.5 mg/kg). Cadmium is the highest-priority toxic metal in the dataset (provincial means 0.047–0.204 mg/kg; max sample in Liaoning); Pb and total As are an order of magnitude or more lower. Aluminum sits at 64–119 mg/kg across provinces — substantially higher than the toxic metals in absolute terms, and a notable input for the peanuts row given the underground-fruit contact pattern. Nickel ranges 1.34–8.73 mg/kg across provinces with strong regional differentiation (Liaoning, Guangxi, Guangdong all > 7 mg/kg).
• Courses: useful case study for geographical-origin effects on heavy-metal uptake in legumes, particularly the demonstration that soil pH and type (acidic red soils in Guangdong and Guangxi vs sandy/gravel soils in northeastern provinces) drive Ca, Cd, and Ni partitioning, and that fertilizer regime (K, Mg in north-central crop-belt provinces) homogenizes some macro elements. The geocarpic-fruit biological mechanism (Section 3.2.3) is the teachable explanation for why peanuts accumulate Cd more readily than tree nuts.
• App: per-province Cd, Pb, total-As, Cr, Ni and Al ranges from this dataset can contribute to the ingredients/peanuts and products/peanuts contamination_profile blocks once Part 9 synthesis runs. Note that arsenic is total (not speciated), so inorganic-As contribution is unresolved from this paper alone.

Provenance notes

Open-access article distributed under Creative Commons Attribution-NonCommercial 3.0 Unported (CC BY-NC) per the title-page licensing statement. Royal Society of Chemistry publication; RSC Advances, 2022, 12, 16790–16797. DOI 10.1039/d2ra02148j confirmed against the article header. PDF retrieved into raw/Manual Fetch Kimi /condiment_papers/05_PB_Vanilla_Spices/ via the manual-fetch Kimi sweep.

Source-internal note: the paper’s CRM is GBW10011 (soybean, per Section 2.2 reagents/instrumentation) but the validation table (Table 2) and Section 2.4 reference GBW10013 as the assessment CRM. Both are soybean reference materials from the Institute of Geophysical and Geochemical Exploration of China; the methods text uses GBW10011 for digestion validation and GBW10013 for accuracy/precision assessment. The wiki preserves the GBW10013 reference because that is the CRM whose values are reported in the validation table.

Verification notes

• 2026-05-28 merge-enhance (Claude Opus 4.7, autonomous v2.0 manual-fetch skill). EXISTING path: DOI grep against wiki/sources/ matched this same page (chen2022-peanut-elements-china.md, DOI 10.1039/d2ra02148j). Prior page dated updated: 2026-05-14; full PDF re-read end-to-end (8 pages; Tables 1, 2, 3, 4 and 5 re-verified).
• Preserved from prior page: cite_key, evidence_tier (A), source_type (peer-reviewed), sample_n (66), publication, authors, year, DOI, metals list, ingredient and product slugs, matrices, jurisdictions, near_duplicates (empty).
• raw_handle updated from legacy manual-fetch-kimi (folder placeholder, not a unique handle) to MFK_determination-of-macro-micro-and-toxic-element-con per current MFK_-prefix convention (matches the v2.0 skill’s suggested handle and the sibling Manual Fetch Kimi naming pattern, e.g. MFK_determination-of-heavy-metals-concentration-in-tra).
• raw_path corrected from truncated "raw/Manual Fetch Kimi /condiment_papers/05_PB_Vanilla_Spices/Determination of macro, micro and toxic element content in peanut.pdf" to the actual filename "...peanuts from six provinces in China.pdf" (trailing-space convention in parent directory preserved).
• raw_sha256 added: ea2907a05664103ee24b16c2b4c1789b38ce98319f322bf2bd66b609b313354d (computed locally from the PDF).
• access_url added: https://doi.org/10.1039/d2ra02148j (DOI canonical resolver).
• no_doi_assigned: false added to match the current source-page schema.
• license corrected from "CC BY" to "CC BY-NC" per the explicit “Creative Commons Attribution-NonCommercial 3.0 Unported Licence” statement on the article PDF (page-1 left margin).
• sample_population expanded with per-province n breakdown (Jilin 10, Liaoning 10, Hebei 15, Henan 19, Guangxi 5, Guangdong 7), sampling protocol (≥3 kg lots from acquisition stations near harvest), and the fresh-weight basis. Prior version named only the provinces.
• Removed editorial Cd EU-ML comparison from Implications. The prior page asserted that the maximum observed Cd (0.204 mg/kg) “is well below EU ML (0.20 mg/kg for peanuts — note: EU ML coincides with highest observed here)“. This claim has two defects: (a) the paper does not cite the EU regulation, so the comparison is borderline-synthesis under Part 2; and (b) the math is wrong (0.204 mg/kg is above 0.20 mg/kg, not below). Removed; the Implications section now states only the regulatory comparisons the paper itself makes (China GB2762-2017 and Australia/NZ Food Standards Code).
• Removed HMTc-threshold-relevance phrasing from Implications. The prior page said Ni concentrations “relevant for HMT&C Ni standards,” which is a Part 2 wiki/HMTc firewall violation (the wiki reports what the literature says; it does not editorialize about HMTc threshold inputs on source pages). Removed; the Implications section now describes the Ni dataset on its own terms.
• Legacy heading ## Wiki pages updated on ingest removed. Replaced by the standard ## Provenance notes + ## Verification notes structure per current Part 6 / v2.0 template; downstream wiki pages this source routes to are derived from frontmatter by the routing layer, not maintained as a body-list.
• Key numbers section expanded to add per-province Duncan-group letter codes for Cd and Ni (preserved verbatim from Table 3), to add the LDA classification numbers (Table 5), to add per-province Al and Ca values (Table 3), and to add the source-cited regulatory limits (China GB2762-2017 Cd and Cr; Australia 2009 ANZ Food Standards Code Cd). The prior page collapsed the LOD/LOQ list and omitted the Duncan-grouping superscripts that the paper uses to distinguish statistically significant provincial differences.
• Methods (brief) section expanded to include digestion temperature ramp, RF power, isotope-per-element table, CRM source institution, ultrapure-water resistivity, statistical-software package, and the GBW10011 vs GBW10013 source-internal CRM discrepancy that prompts the note in Provenance.
• Numerical fidelity re-verified against the PDF tables: Cd by province (Table 3), Pb by province (Table 3), As by province (Table 3), Cr by province (Table 3), Ni by province (Table 3), Al by province (Table 3), LODs and LOQs (Table 2), CRM recoveries (Table 2), LDA correct-classification 97.0% original / 78.8% cross-validated (Table 5), ANOVA p-values for non-significant elements (Pb p=0.516, Cr p=0.263, As p=0.077, Al p=0.222) — all verified verbatim from the paper.
• Part 12 brand firewall: no brand names appear in source-page body. Method-section vendor and instrument names (Thermo Fisher Scientific HNO3, Sinopharm H2O2, Milli-Q ultrapure water, Midea MJ-BL80Y21 homogenizer, MARS5 CEM microwave, iCAP Q Thermo ICP-MS, Inorganic Ventures stock standards, SPSS 23.0 IBM) are scientific-method vendors permitted under the 2026-05-17 strict-reading Exception 2 (instruments, reagents, and statistical software named for reproducibility, not as brand attribution to contamination values).
• Part 2 wiki/HMTc firewall: the Implications section reports what the paper found (occurrence dataset, geographic differentiation, geocarpic biology) and the regulatory comparisons the paper itself draws. It does not propose HMTc thresholds, does not synthesize across other literature, and does not issue consumer risk advisories.

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.