High proportions of inorganic arsenic in Laminaria digitata but not in Ascophyllum nodosum samples from Ireland

Overview

Ronan and colleagues measured total arsenic and inorganic arsenic (as As(V)) by ICP-MS and HPLC-ICP-MS across the different thallus parts of two wild Irish brown seaweeds: the kelp Laminaria digitata and the intertidal wrack Ascophyllum nodosum. Total arsenic was similar between species (36–131 mg/kg dry weight in L. digitata, 38–111 in A. nodosum), but the inorganic fraction diverged sharply: inorganic arsenic was always below 1% of total in A. nodosum (≤0.703 mg/kg DW) yet reached 2.2–87 mg/kg DW in L. digitata, rising along the thallus from the stipe to the decaying distal blades and exceeding 50% of total arsenic in the middle, distal, and decaying-distal blades. This establishes that high inorganic arsenic in edible seaweed is not confined to hijiki: kelp harvested from a clean site carried inorganic arsenic an order of magnitude above the French (3 mg/kg) and Australia–New Zealand (1 mg/kg) seaweed limits.

Key numbers

Concentrations in mg/kg dry weight (DW). Method LOD/LOQ: inorganic arsenic (As) 0.001/0.01; total As 0.002/0.02.

Laminaria digitata (kelp), n=5 thalli, 6 parts each:

• Total arsenic: 36–131 mg/kg DW across all parts (highest in distal blade 131 and decaying distal blade 129); mean of thalli 78 ± 24.
• Inorganic arsenic: 2.2–87 mg/kg DW, increasing from the stipe toward the distal parts; highest 87 mg/kg DW in decaying distal tips.
• Inorganic arsenic comprised >50% of total arsenic in most middle, distal, and decaying-distal blades, but <50% in stipe, meristem, and young blade.
• inorganic arsenic (iAs) in stipe/meristem/young blade was significantly lower than in middle/distal/decaying-distal blades (one-way ANOVA p=0.0009).
• Blade-only range (Table 1, stipe and meristem excluded, across five thalli): total As 59–114, inorganic As 30–62 mg/kg DW.

Ascophyllum nodosum (knotted wrack), n=25 thalli, annual segments:

• Total arsenic: 38–111 mg/kg DW, increasing with thallus-part age, highest (111) in 5-year-old sections; mean of thalli 64 ± 20.
• Inorganic arsenic: <1% of total throughout; 0.007–0.070 mg/kg DW across yearly sections, with two older (4- and 5-year) sections higher at 0.178 and 0.703 mg/kg DW.
• Total arsenic was significantly higher in L. digitata than A. nodosum (p=0.026); inorganic arsenic significantly higher in L. digitata (p<0.01).

Additional source-reported facts:

• The site (Galway Bay) is a routinely-monitored, relatively uncontaminated bay; seawater arsenic is typical of unpolluted Irish west-coast waters.
• Regulatory comparison: mean L. digitata inorganic arsenic in the young/middle/distal/decaying blades exceeded the French food limit (3 mg/kg DW) and the Australia–New Zealand limit (1 mg/kg) typically by an order of magnitude; A. nodosum inorganic arsenic always satisfied both. Applying the EU feed Directive 2003/100/EC basis (40 mg/kg total As at 12% moisture), mean total As in both species exceeded the limit (69 and 57 mg/kg respectively).
• The authors’ Table 1 (comparison with other studies) places this study’s L. digitata alongside Taylor & Jackson 2016 (New England: total As 50–106, iAs 2.8–20 mg/kg DW), García-Salgado 2012 (France: iAs 77 ± 3 mg/kg as As(V)), and others, confirming wide between-study variability in kelp inorganic arsenic.
• Context cited: EFSA (2014) estimated hijiki at 11 mg/kg iAs (lower than the 96 mg/kg of Rose 2007 and 117 mg/kg of Almela 2006); UK FSA (2010) and the Belgian Superior Health Council (2015) advise against hijiki consumption.

Methods (brief)

Whole thalli were collected at low tide (Spiddal, Galway Bay; 53.2409 N, −9.3095 W) in October 2015, rinsed, scraped of epiphytes, freeze-dried, and homogenised. Inorganic arsenic was determined as As(V): 0.100 g freeze-dried algae microwave-extracted (Mars 5) with 1% HNO3 + 2% H2O2 to oxidise As(III) to As(V), then HPLC-ICP-MS (Agilent 1260 HPLC + 8800 ICP-MS, Hamilton PRP-X100 anion-exchange column, 30 mM ammonium carbonate pH 8.5, germanium internal standard, O2 reaction gas, As measured as AsO+ at m/z 91). Total arsenic was determined after overnight concentrated-HNO3 pre-digestion and Ethos Up microwave digestion to 200 °C, by ICP-MS in MS/MS mode (m/z 91). Quality control used CRMs BCR-211 (rice; certified inorganic As) and ERM-CD200 (Fucus vesiculosus; certified total As): total-As recovery 90.4 ± 5.7% (BCR-211) and 96.4% (ERM-CD200); inorganic-As recovery 97.6% (BCR-211). The unequivocal determination of inorganic As as As(V) was confirmed by parallel HPLC-ICP-MS/ESI-MS showing no arsenosugar co-elution interference (acid extraction collapses arsenosugars to the aglycone-free sugar without generating inorganic As). All values are dry weight.

Implications

• Certification (HMTc): direct-evidence speciation data for the Category 6 seaweed-kelp-foods row (iAs/total arsenic (tAs) platform), and the most consequential finding so far for that row’s structure — kelp (Laminaria digitata), a primary component of “kombu”/sea-vegetable products, can itself carry inorganic arsenic up to 87 mg/kg DW (>50% of total in the blades), comparable to hijiki and an order of magnitude above the French and ANZ seaweed limits, even from a clean site. This argues that the within-row high-iAs concern is not hijiki-exclusive and that kelp/Laminaria warrants explicit treatment in the row’s clean/dirty structuring. The strong intra-thallus gradient (stipe low → decaying distal high) and species contrast (A. nodosum negligible) are processing/sourcing-relevant signals. All on the dry-weight, as-harvested basis.
• App: contributes to the seaweed ingredient contamination_profile for total and inorganic arsenic, with the species- and thallus-part-specific caveats (kelp blades high; knotted wrack low).

Update history

The five most recent substantive edits to this page, classified major (evidence or structure moved), correction (a published value or statement was wrong and has been fixed), or minor (narrative rewritten without changing the underlying evidence). Each description is derived from what the edit did to this page; the linked commit is the authoritative record, routine regeneration passes are excluded, and the full version history lives in git.