Mukhi et al. 2022/2024 — As, V, Hg, Cd in 13 food and drug packaging materials, Karnataka, India

This source page covers both published versions of this F1000Research article (open peer review, three-version platform). Version 1 (Mukhi et al. 2022, DOI 10.12688/f1000research.121473.1, Marker-converted markdown at raw/markdown/FM_11109717/) and Version 3 (Mukhi et al. 2024, DOI 10.12688/f1000research.121473.3, PDF at raw/Papers Cube Manual Fetch/article (1) copy 2.pdf) are the same study with the same 13 samples and the same Table 3 concentration values. The substantive difference between versions is the permissible-limit reference used to compute exceedance counts: v1 stated Hg and Cd limits as ≤100 ppm (uncited); v3 revised these to ≤0.03 ppm (Hg) and ≤0.05 ppm (Cd) per Nordic Council 2015 Food Contact Materials guidance (TN2015-522). The result is that the same Table 3 data yields very different exceedance counts under v1 vs v3 limit references, and the abstract sentence changed accordingly between versions. Both readings are preserved below.

This study quantifies arsenic (As), vanadium (V), mercury (Hg), and cadmium (Cd) by ICP-OES in 13 types of food and drug packaging materials procured from a local market in Karnataka, India, with prior qualitative spot-test screening. Concentrations are reported in ppm of acid-digested extract. The headline finding under the v3 (Nordic Council 2015) limit reference is that all four metals exceeded permissible levels in the majority of samples, with arsenic and mercury exceeding in all 13 samples and cadmium exceeding in all 10 samples in which it was detected. Vanadium and cadmium were each found within or below the permissible limit in three samples. The authors interpret the findings as evidence that packaging itself is a non-trivial heavy-metal exposure pathway warranting stricter regulatory oversight, particularly for materials sourced from Indian supply chains.

Key numbers

Quantitative concentrations (Table 3, ICP-OES, ppm of digested extract)

Quantitative ranges as the paper reports them (Discussion, v3)

• Vanadium: 0.29 – 40.8 ppm (paper’s stated range, omits tissue at 0.097 ppm; counting tissue, the true detected range is 0.097 – 40.8 ppm).
• Arsenic: 1.7 – 236.2 ppm (paper’s stated range, omits medicinal blister packet at 0.54 ppm and cardboard at 0.83 ppm; counting all detections, the true range is 0.54 – 236.2 ppm).
• Mercury: 1.53 – 546 ppm.
• Cadmium: 2.2 – 337 ppm (paper’s stated range, omits medicinal closure at 0.28 ppm; counting all detections, the true range is 0.28 – 337 ppm).

Exceedance counts under v3 (Nordic Council 2015) limit reference

Counts are arithmetically identical under either the v3 prose values (Hg 0.003, Cd 0.005 ppm) or the v3 Table 3 header values (Hg 0.03, Cd 0.05 ppm) because every detected Hg and Cd concentration exceeds both candidate caps by ≥30×. Treat the prose values as authoritative.

• Vanadium > 0.01 ppm: 10 of 13 samples (paper’s stated count, treating tissue 0.097 ppm as “within limits”); strictly numerical exceedance is 11 of 13 because tissue (0.097 ppm) is 9.7× the 0.01 ppm cap. The paper’s text says vanadium was within limits in tissue, sachet, and cellophane; sachet and cellophane were ND, tissue was detected above the cap. This is an internal inconsistency in the paper.
• Arsenic > 0.002 ppm: 13 of 13 samples.
• Mercury > 0.003 ppm (prose) or > 0.03 ppm (Table header): 13 of 13 samples.
• Cadmium > 0.005 ppm (prose) or > 0.05 ppm (Table header): 10 of 13 detected samples (3 samples — aluminium can, plastic container, medicinal blister packet — were ND); all 10 detections exceeded the cap.

Exceedance counts under v1 (uncited 100 ppm Hg/Cd reference; SUPERSEDED)

• Vanadium > 0.01 ppm: 10 of 13 (same as v3).
• Arsenic > 0.002 ppm: 13 of 13 (same as v3).
• Mercury > 100 ppm: 2 of 13 (sachet 546 ppm, plastic bag 174 ppm).
• Cadmium > 100 ppm: 1 of 13 (sachet 337 ppm).

Qualitative screening (Table 2, spot-test presence/absence prior to ICP-OES)

• Vanadium positive: cardboard, medicinal closure, medicinal blister packet (3 of 13).
• Mercury positive: leak-proof bag, aluminium can, cellophane, sachet, plastic bag, medicinal closure, medicinal blister packet, aluminium bag (8 of 13).
• Arsenic positive: leak-proof bag, cellophane, sachet, cardboard, aluminium bag, aluminium box (6 of 13; the paper’s text says 5 but Table 2 shows 6).
• Cadmium positive: leak-proof bag, sachet, plastic bag, cardboard, aluminium bag (5 of 13).
• Qualitative detection in at least one metal: 10 of 13 samples (per abstract).

Instrument and quality assurance

• ICP-OES: Agilent Technologies; Expert software v7.100.6821.61355; firmware v2994.
• Multi-element standard for calibration: REICPCAL29A.
• Stated detection limits (v3 standard curves, Figure 10): V 0.000011 ppm; As 0.000156 ppm; Cd 0.000442 ppm; Hg <0.001 ppm.
• Digestion: USEPA 3050(B) acid digestion and USEPA 3051 microwave-assisted digestion. HCl + H₂SO₄ (80:20) for aluminium can, leak-proof bag; H₂SO₄ for plastic container; H₂SO₄ + HNO₃ for aseptic carton, sachet, medicinal closure, medicinal blister packet; HCl + HNO₃ (80:20) for cellophane, plastic bag, cardboard, aluminium bag, aluminium box, tissue. Digestion times 34–210 minutes (Table 1).
• Triplicate runs averaged.
• No blank control or recovery data reported (flagged by v1 reviewer Huang and acknowledged but not addressed by author response).
• LOQ not reported; v3 reviewer Senila explicitly criticised this; author response gave detection limits (above) but did not provide LOQ.

Methods (brief)

Thirteen packaging-material types (eleven food packaging, two drug packaging) were purchased from a local Mangalore market. After distilled-water rinsing and oven drying, 10 g of each was acid-digested using one of three protocols (microwave-assisted HCl/H₂SO₄; dehydrator-aid H₂SO₄ or H₂SO₄/HNO₃; acid digestion HCl/HNO₃ 80:20), filtered to 0.45 μm and then 0.25 μm, and quantified by ICP-OES (Agilent) against a REICPCAL29A multi-element standard. Qualitative screening preceded ICP-OES: spot test (sodium salicylate + phosphoric acid) for V, Aquasol detector strip for As, dithizone diphenylcarbazone cotton-swab test for Cd, and SenSafe dithizone strips (Industrial Test Systems, USA) for Hg. Triplicate runs for each test; averages reported. Institutional Ethics Committee approval obtained. Concentration values were compared against permissible limits stated as 0.01 (V), 0.002 (As), 0.03 (Hg, v3 only), 0.05 (Cd, v3 only) ppm referencing the European Council via Nordic Council guidance for authorities TN2015-522. Limits for Hg and Cd in v1 were stated as 100 ppm without citation; v3 revised these after peer-reviewer challenge.

Known limitations and methodological caveats:

• Extraction-versus-migration mismatch (post-publication reader comment, Davide Marchesi, Tetra Pak, 07 Jan 2026): The Nordic Council 2015 limits the authors cite are migration-test specific release limits (SRLs) — they bound the amount of metal that transfers from packaging into food under standardised migration testing, not the total metal in an acid extract of the packaging itself. Acid digestion of the packaging measures total metal content, which is necessarily ≥ migration. Comparing total-extraction concentrations against migration limits overstates the regulatory-compliance signal. Downstream synthesis should not propagate the v3 “exceeded permissible limits in all 13” claim without this caveat.
• Peer review trajectory: v1 (2022) was reviewed 1 approved-with-reservations + 1 not-approved (Huang, plant ionomics); v2 (2023) became 2 approved (Ganesapillai, Vellore IIT) + 1 not approved (Senila, ICP-OES specialist); v3 (2024) carries 2 approved + 1 not-approved status banner. Reviewer Senila later updated v3 to “Approved” per a 15 May 2024 editorial note. Net: among the three reviewers, two ultimately approved (Ganesapillai, Senila v3); one (Huang, v1 only) did not re-review and remains as the non-approval.
• Internal inconsistencies in the published paper: (1) The v3 Discussion states vanadium was “within the permissible limits in three samples (tissue, sachet, cellophane)” but Table 3 shows tissue at 0.097 ppm, which is 9.7× the stated 0.01 ppm cap. (2) The v3 Discussion gives ranges (V 0.29–40.8, As 1.7–236.2, Cd 2.2–337) that omit lower-end detected values present in Table 3. Treat the table values as authoritative for sample-level concentrations and ranges. (3) The v3 Table 3 column header gives the Hg permissible concentration as ≤0.03 ppm and Cd as ≤0.05 ppm, but the v3 Quantitative-analysis prose on the same page and the v3 Author Response 6 both give Hg ≤0.003 ppm and Cd ≤0.005 ppm (i.e., 10× more restrictive than the table header). Treat the prose values as authoritative because they are corroborated by the Author Response 6 explicit numbering; the table-header is a probable typo. Exceedance counts are unchanged under either value because every detected Hg/Cd concentration exceeds both candidates by ≥30×.
• No blank controls or recovery data. v1 reviewer Huang flagged this; author response did not address.
• Brand-name terminology in the original paper: The authors use “tetrapak” generically for aseptic carton packaging; a post-publication reader comment from Tetra Pak (07 Jan 2026) corrects this as a trademark-usage error. This wiki page uses “aseptic carton” in the body and notes the paper’s terminology in the sample_population field.

Implications

Occurrence data: Provides one of the only quantitative ICP-OES surveys of Indian-market food and drug packaging for V, As, Hg, and Cd, useful as occurrence-data input for any synthesis pass on packaging-borne contamination. Evidence tier B rather than A because of the missing blank/recovery QA, the unresolved migration-vs-extraction methodological objection, and the small sample-type-by-type n=1 design (each material type tested once, not multiple replicates of the same material from different lots).

Exposure pathway: Supports the broader literature view that food and drug packaging is a non-trivial heavy-metal exposure pathway. The specific quantitative claims should not be propagated downstream without the migration-vs-extraction caveat documented above.

Geographic specificity: Findings are specific to a single Mangalore market; do not generalise to all Indian-supply-chain packaging. The route from these data to country-level claims requires additional Indian-supply-chain surveys.

Microbiome: Not addressed.

Verification notes

• 2026-05-18 (audit subagent verdict REVISE; verified against v3 PDF page 9 prose and Author Response 6 p. 17 — applied): Subagent flagged the wiki’s v3 permissible-limit row (Hg ≤0.03, Cd ≤0.05) as inconsistent with the paper’s prose, which gives Hg ≤0.003 and Cd ≤0.005. Re-verified: v3 p. 9 Quantitative-analysis paragraph states “The maximum permitted quantity (ppm) as suggested by the European Council for Food Packaging Association (Nordic guidance for authorities) is; As – 0.002 ppm, V – 0.01 ppm, Hg – 0.003 ppm, Cd – 0.005 ppm.” v3 Author Response 6 (p. 17) corroborates the same values. v3 Table 3 column header reads ≤0.03 / ≤0.05 for Hg and Cd, which is a 10× discrepancy from the prose. Corrected: Table 3 in this wiki page now shows the prose values as authoritative and the table-header values as a flagged paper-internal inconsistency, and the inconsistency is documented in the Methods (brief) section alongside the V-tissue and range-omission flags. Exceedance counts unchanged because every Hg/Cd value exceeds both candidate caps.
• 2026-05-18 (audit subagent ⚠️ on matrices food-packaging/packaging-material; verified — finding noted, no edit): Subagent noted these are not in the taxonomy snapshot’s enumerated lists. Verified via grep -h '^matrices:' wiki/sources/*.md: both slugs appear repeatedly across the existing source-page corpus as established matrices for packaging-material studies. Per the audit-prompt note that “matrices vocabulary is its own controlled list,” these are corpus-established matrices for this paper’s scope. No defect; finding noted as a stale-snapshot artefact, no edit applied.
• 2026-05-18 (Claude session, merge-enhance from v3 PDF re-read): The prior page (updated 2026-05-14) used the v1 abstract’s exceedance counts (Hg 2 of 13, Cd 1 of 13) as the headline Key numbers but framed them as v1+v3-combined. v1 and v3 give different exceedance counts because v3 revised the Hg and Cd permissible-limit reference from an uncited 100 ppm value to the Nordic Council 2015 TN2015-522 values (0.03 ppm Hg, 0.05 ppm Cd). The prior page’s headline numbers were faithful to v1 but materially understated v3’s exceedance findings (v3: Hg 13/13, Cd 10/13 detected). Both readings are now preserved with their respective limit references.
  • Title field: kept v1 title as canonical title: to preserve cite-key alignment; added title_v3: for the v3-revised title.
  • DOI: kept doi: 10.12688/f1000research.121473.1 aligned with cite_key mukhi2022; added doi_latest: 10.12688/f1000research.121473.3 and updated access_url to resolve to the latest version.
  • Metals frontmatter: upgraded [As, V, Hg, Cd] → [tAs, V, tHg, Cd] to reflect ICP-OES total-element detection (non-negotiable speciation discipline per CLAUDE.md Part 14).
  • Evidence tier: lowered from A to B. ICP-OES quantitative with peer review supports A on instrument quality, but the missing blank/recovery controls, unresolved migration-vs-extraction methodological objection, n=1 per material type, and one persistent non-approval review collectively put this in B territory per Part 13 conventions.
  • Added raw_sha256 for the FM markdown (v1 source) and raw_path_v3 + raw_sha256_v3 for the PCMF PDF (v3 source).
  • sample_population rewritten to enumerate all 13 samples by type and to surface the “tetrapak” brand-term issue.
  • near_duplicates cleared — the v3 PDF is not a near-duplicate of v1; it is the same paper, later version. Both versions are documented in this single source page via the _v3 companion fields.
  • Added ## Key numbers Table 3 in full (Vanadium, tAs, tHg, Cd by sample, with both v1 and v3 permissible-limit rows) — the prior page summarised exceedance counts without the per-sample values, making downstream synthesis blind to which materials drove the headline numbers.
  • Added the paper-internal inconsistencies (V tissue 0.097 vs claimed “within limits”; As/V/Cd ranges in Discussion that omit lower-end detected values) — these are real defects in the published paper, not transcription errors, and downstream synthesis needs them flagged.
  • Added the v3 reader-comment caveat from Davide Marchesi (Tetra Pak, 07 Jan 2026) about extraction-vs-migration methodology — this materially affects how the exceedance claims should be propagated downstream and was not on the prior page.
  • Brand-firewall handling: replaced “tetrapaks” (a brand-name term, even if used generically in the paper) with “aseptic carton” in the body and Key numbers, noting the paper’s term in sample_population. SenSafe (Industrial Test Systems) and Aquasol kits retained in Methods as scientific-method vendor names per the Part 12 strict-reading Exception 2 (locked 2026-05-17).
  • Fixed broken wikilink [[supply-chain/packaging]] (no such page) → [[mitigation/packaging-and-storage]].
  • Added [[metals/vanadium]], [[metals/arsenic-total]], [[metals/mercury-total]] to “Wiki pages this source may touch” — prior page omitted V despite it being a primary measured metal, and did not use the speciation-explicit metal-page slugs.
  • ## Implications tightened to occurrence-data framing per Part 2 / audit Check 5; removed “South Asian markets should flag elevated As and V risk in the app model” forward-looking-prescription phrasing.
  • Section heading ## Wiki pages updated on ingest retained per current Part 6 template; Part 5b reminder: the model writes accurate frontmatter, the routing layer (and not this section) determines downstream fan-out.

Wiki pages this source may touch

• arsenic-total
• cadmium
• mercury-total
• vanadium
• packaging-and-storage

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.