Rüstemli & Tuncay 2026 — Mineral and Heavy Metal Content in Canned Tuna by Packaging Medium (Van Province, Türkiye)

Rüstemli and Tuncay measured 15 elements (Mg, Al, P, K, Ca, Cr, Mn, Fe, Cu, Zn, As, Se, Sn, Hg, Pb) plus Co, Ni, Sr, and Ba in 105 canned tuna samples from the Turkish market, with three equal-size groups (n=35 each) of water-packed, oil-packed, and sauce-packed product. Samples were drawn from seven domestic Turkish brands sold in Van Province in September–November 2022. Analysis was by ICP-MS (Agilent 7800) after microwave-assisted nitric acid digestion, with method accuracy verified against DORM-4 Fish Protein certified reference material (recoveries 95–102%). Arsenic was reported as total arsenic only — no inorganic/organic speciation was performed — and mercury was reported as total mercury with no methylmercury speciation. The headline finding is that oil-packed product carried substantially higher concentrations of nearly every element measured (essential and toxic) than the water- and sauce-packed product, with mercury exceeding the Turkish Food Codex 1.0 mg/kg limit in 5.71% of samples (concentrated in oil-packed) and total arsenic exceeding the FAO/WHO 1.0 mg/kg guideline in 74–100% of samples across packaging types.

Key numbers

Sample design: n=105 total (n=35 water-packed, n=35 oil-packed, n=35 sauce-packed); seven domestic Turkish brands, multiple production lots per brand; sample size determined by power analysis (α=0.05, power=0.80, effect size=0.5, ≥30/group). Reporting basis: mg/kg wet weight throughout.

Essential macroelements (Table 1, mean ± SD mg/kg wet weight; range in parentheses; superscripts a/b/ab denote Duncan post-hoc grouping at p<0.05):

Trace elements (Table 2, mean ± SD mg/kg wet weight; range in parentheses):

Potentially toxic elements with regulatory comparators (Table 3, mean ± SD mg/kg wet weight; range in parentheses):

Regulatory exceedances:

• Mercury (tHg) exceeded the Turkish Food Codex (TFC 2023) 1.0 mg/kg limit in six samples (5.71% of all samples): one sauce-packed sample at 1.28 mg/kg (2.86% of sauce-packed) and five oil-packed samples at 1.05, 1.11, 1.20, 1.35, and 2.68 mg/kg (14.29% of oil-packed). Three of those six samples (the two highest oil-packed at 1.35 and 2.68 mg/kg, plus the 1.28 mg/kg sauce-packed sample) also exceeded the Codex Alimentarius Commission 1.2 mg/kg limit — 2.86% of all samples.
• Lead (Pb) and tin (Sn): all 105 samples complied with the 0.30 mg/kg Pb and 200 mg/kg Sn maximum residue limits.
• Total arsenic exceeded the FAO/WHO (2015) 1.0 mg/kg guideline in 80% of sauce-packed, 100% of oil-packed, and 74.29% of water-packed samples. The overall tAs range was 0.31–7.08 mg/kg (overall mean 1.49 mg/kg); the highest single value was 7.08 mg/kg in oil-packed tuna.

Magnitude of the packaging-medium effect (text, comparing oil-packed to water-packed means): Mg ~3.4× higher, P ~4.3× higher, Fe ~4.4× higher, Hg ~4.0× higher in oil-packed than water-packed product. Oil-packed concentrations were significantly higher than water- and sauce-packed for most analytes at p<0.05.

Estimated daily intake (EDI) and target hazard quotient (THQ), Table 4 — worst-case using oil-packed mean concentrations, IR = 0.033 kg/day, BW = 70 kg:

Total arsenic THQ exceeded the non-carcinogenic risk threshold (THQ > 1); mercury THQ approached the threshold; lead THQ was well below.

Methods (brief)

Cross-sectional sampling, September–November 2022, Van Province, Türkiye; ethical approval from Van Yüzüncü Yıl University Non-Interventional Clinical Research Ethics Committee (Decision No. 2022/11-03, dated 3 November 2022). Sample preparation followed AOAC Official Method 999.10 (AOAC 2005). Water-packed tuna (tuna + liquid) was homogenized in a high-speed blender (IKA T25 Digital Ultra-Turrax, Germany) for 2 min at 10,000 rpm and oven-dried at 105±2 °C in pre-cleaned porcelain crucibles for ~18–24 h to constant weight. Oil-packed and sauce-packed samples were freeze-dried at −40 °C for 18 h (Sanyo MDF U32865, Japan) to facilitate homogenization and digestion, then ground in a ceramic mortar and pestle to prevent metal contamination. Homogenized powders were stored at −20 °C in acid-washed polypropylene containers until digestion.

Microwave-assisted acid digestion was performed in an ETHOS EASY microwave system (Milestone, Italy) using 65% suprapur HNO₃ (Merck 100456, Germany) and ultrapure Type I water (18.2 MΩ·cm, Milli-Q). Approximately 200±5 mg of dried or freeze-dried sample was weighed into TFM digestion vessels, 7 mL concentrated HNO₃ and 1 mL ultrapure water were added, a 30 min pre-digestion step at room temperature was followed by a three-step ramp to 200 °C. Cooled digests were brought to 15 mL with ultrapure water and analyzed by ICP-MS within 48 h. Quality control included procedural blanks (three per batch of 10–12 samples), DORM-4 Fish Protein certified reference material (NRCC) analyzed in triplicate with each batch, and matrix spike recovery tests on ~10% of randomly selected samples. CRM recoveries were 95–102% for all analyzed elements.

ICP-MS was performed on an Agilent 7800 (Agilent Technologies, USA) at the Central Research Laboratory of Bayburt University, Türkiye. Samples were introduced via a MicroMist concentric nebulizer with a Scott-type double-pass spray chamber cooled to 2 °C. Analyses were performed in helium collision/no-gas mode to reduce polyatomic interferences. Measurements were performed in triplicate with online internal standards (⁷²Ge, ¹¹⁵In, ²⁰⁹Bi at 10 µg/L) to compensate for instrumental drift, matrix effects, and signal fluctuations. Calibration standards were prepared daily from 1000 mg/L Merck CertiPUR® stocks in 2% (v/v) HNO₃, with six-point plus blank curves showing R² > 0.999; mid-range check standards yielded recoveries of 90–110%. LOD and LOQ were calculated as three and ten times the standard deviation of procedural blanks (n=10). Concentrations below the LOQ were reported as <LOQ; values below the LOD were assigned LOD/2 for statistical analyses. Statistical analysis used the Shapiro–Wilk normality test followed by one-way ANOVA with Duncan’s multiple range test for post-hoc pairwise comparisons (p<0.05) in IBM SPSS Statistics 25.

EDI was computed as C × IR / BW where C is the mean metal concentration (mg/kg wet weight), IR is daily canned tuna ingestion rate (0.033 kg/day), and BW is adult body weight (70 kg). THQ was computed as EDI / RfD using EPA oral reference doses. EDI/THQ were calculated for the worst-case (oil-packed) mean concentrations of Hg, As, and Pb only.

Limitations

The authors explicitly note that total arsenic was measured without speciation; because inorganic arsenic is primarily responsible for arsenic toxicity while organic arsenic species (predominant in fish muscle) are far less harmful, the calculated arsenic THQ of 4.78 substantially overestimates actual non-carcinogenic risk. The 100% exceedance of the FAO/WHO 1.0 mg/kg guideline in oil-packed product likely reflects the same overestimation, since fish-muscle arsenic is dominated by arsenobetaine. Mercury was reported as total mercury with no methylmercury speciation; the authors treat tHg as a proxy for MeHg in the risk calculation, which is the EPA-fish-advisory convention but adds uncertainty. Sampling was geographically anchored in Van Province (eastern Türkiye) over a concentrated three-month window in 2022, limiting generalizability to other Turkish regions, other times of year, and other production lots. Brand identities were not disclosed, so brand-level performance cannot be evaluated from this paper (which is consistent with the wiki’s brand firewall). Bioavailability from the different packaging media was not experimentally determined, so dietary exposure inferences from raw concentrations may overstate or understate actual systemic exposure. The EDI/THQ used a single mean body weight (70 kg) and a single ingestion rate (0.033 kg/day) rather than distributional inputs.

Implications

Adds Turkish-market occurrence data for total arsenic, total mercury, lead, tin, aluminum, chromium, iron, manganese, copper, zinc, cobalt, nickel, barium, and strontium in canned tuna stratified by packaging medium (water-packed, oil-packed, sauce-packed) at n=35 per stratum. The dataset supports the existing canned-fish/canned-tuna occurrence corpus and provides a Turkish counterpart to existing Ecuadorian (Aguilar-Miranda 2024), Brazilian (De Lima 2021), Mexican (Rodriguez-Mendivil 2019), Italian (Pappalardo 2017), Serbian (Novakov 2017), and 36-country (Ulusoy 2023) datasets. The packaging-medium stratification — with oil-packed product showing systematically higher concentrations of most analytes — is the principal new contribution to the occurrence record and is relevant to any canned-tuna product-category page that wants to disaggregate by packaging medium.

Speciation discipline: the paper measured total arsenic and total mercury; downstream synthesis on iAs or MeHg cells should not draw on Rüstemli & Tuncay 2026 occurrence values without explicit speciation correction or qualification. The total arsenic exceedances are a marker of total-As burden, not directly of inorganic-As risk.

Wiki pages this source may touch

• canned-tuna
• fish
• seafood
• canned-fish
• canned-seafood
• mercury-total
• arsenic-total
• lead
• tin
• aluminum
• chromium
• iron
• manganese
• copper
• zinc
• cobalt
• nickel
• barium
• magnesium

Verification notes

The published paper contains an internal inconsistency on the page reporting the arsenic exceedance proportions. The Results / Regulatory compliance paragraph (page 5) states that arsenic concentrations exceeded the FAO/WHO guideline in “80% of oil-packed, 80% of sauce-packed, and 74.29% of water-packed samples.” The Discussion (page 10) and Figure 2 (page 6) instead report 80% sauce-packed, 100% oil-packed, and 74.29% (≈75% per the figure label) water-packed. The abstract is consistent with the Discussion/Figure 2 wording, stating that arsenic was “frequently detected across all packaging types (74–100% of samples).” This source page reports the Discussion/Figure 2/abstract-consistent values (80/100/74.29%), which match the visualization in Figure 2. The page-5 “80% of oil-packed” figure appears to be a typographical error in the Results text.

Audit subagent (2026-06-04, fresh-context general-purpose Agent) flagged that the Chromium (Cr) oil-packed row in Table 1 was reported here as “0.60 (0.37–2.10)” with no SD; re-reading PDF page 4 confirmed the SD 0.34 is in fact printed in the cell (“0.60 ± 0.34”). The initial transcription dropped the ”± 0.34” portion of the cell during the first PDF read. The cell was corrected to “0.60 ± 0.34 (0.37–2.10) ᵃ” against PDF page 4 Table 1.

Mercury is reported throughout as total mercury (tHg) with no methylmercury speciation, per the authors’ methods description. Arsenic is reported throughout as total arsenic (tAs) with no inorganic/total speciation, as the authors explicitly state in the Discussion. The metals frontmatter and key-numbers text use tHg and tAs accordingly, per the wiki’s non-substitution rule.

Brand identities were withheld by the authors for ethical reasons; no brand names appear in the paper or on this page. The seven domestic Turkish brands are referenced only as a count, which is consistent with both the source’s ethical framing and the wiki’s Part 12 brand firewall.

The CRM is referred to in the abstract as “DORM-4 Fish Protein” and in the references as “DORM-4 Fish protein certified reference material (Willie et al. 2012)”; the methods text describes it as “DORM-4 Fish Protein Certified Reference Material (NRCC).” All three are the same NRCC material; the page uses “DORM-4 Fish Protein” per the abstract.

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.