Narukawa 2020 — Hijiki candidate reference material: total arsenic, inorganic arsenic (As(V)), and arsenosugars (inter-laboratory study)

Narukawa and colleagues (NMIJ/AIST, with the University of Graz) ran a two-laboratory inter-laboratory study to characterise a candidate hijiki seaweed certified reference material (NMIJ CRM 7405-b) for total arsenic and water-soluble arsenic species, validating methods against the existing certified hijiki CRM 7405-a. This is a metrology/CRM-development paper rather than a market survey, but it provides rigorously-quantified hijiki arsenic occurrence values on a dry-mass basis: the candidate hijiki material carried total arsenic 49.4 mg/kg and inorganic arsenic (measured as As(V)) 24.4 mg/kg dry weight (≈49% of total), and the control CRM 7405-a is certified at total arsenic 35.8 and As(V) 10.1 mg/kg. The values reinforce hijiki as a high-inorganic-arsenic seaweed and supply CRM-quality anchor points; because they are reference materials, not retail products, they are flagged below as method/anchor data rather than market-occurrence inputs.

Key numbers

Dry-mass basis (candidate-material moisture 7.84 ± 0.13%). As compounds in mg/kg as As.

Control CRM (NMIJ CRM 7405-a hijiki), certified values (k=2):

• Total arsenic: 35.8 ± 0.9 mg/kg
• As(V) (inorganic arsenic surrogate): 10.1 ± 0.5 mg/kg
• Inter-lab measured As(V): Lab 1 9.7 ± 0.1; Lab 2 10.1 ± 0.2 mg/kg (both within the certified value)

Candidate material (NMIJ CRM 7405-b hijiki), combined inter-laboratory values (Table 2; uncertainty k=2):

Additional source-reported facts:

• Inorganic arsenic in the candidate hijiki, measured as As(V), is 24.4 mg/kg dw — roughly 49% of the 49.4 mg/kg total arsenic; per-lab total-As was 49.2 ± 0.4 (Lab 1) and 49.6 ± 0.3 (Lab 2) mg/kg.
• Mass balance: water-soluble arsenic ≈53% and insoluble (residue) ≈47% of total As; sequential water extraction recovered the water-soluble fraction completely (As not detected in 2nd/3rd extractions); overall recovery ≈104%.
• The two laboratories agreed within measurement uncertainty for total As and all water-soluble species; relative combined standard uncertainties were 1.0% (total As), 1.3% (As(V)), and 1.7–1.8% (arsenosugars, DMA).
• Method LODs were sub-ng/g (e.g., As and As(V) by ICP-MS / HPLC-ICP-MS ~0.01–0.02 ng/g; DMA and arsenosugars ~0.07–0.08 ng/g).
• Arsenosugar certified values are not available for CRM 7405-a, which motivated the candidate-material work; NIST SRM 3232 Kelp Powder carries reference (not certified) arsenosugar values.

Methods (brief)

A ~22 kg hijiki (Hizikia fusiforme) batch collected from the west coast of Japan was washed, dried, freeze-pulverised, sieved (<100 µm), homogenised, gamma-sterilised, and bottled (640 bottles) as the candidate CRM 7405-b; NMIJ CRM 7405-a hijiki was the control. Two laboratories analysed both materials. Total arsenic was determined by ICP-MS (Agilent 7900 / 7500c) after microwave acid digestion (HNO3, with HF/H2O2 in Lab 2), with Ge or Rh internal standards. Water-soluble arsenic species were extracted with water (overnight rotation or 1 h ultrasonication) and separated by anion-exchange (Thermo AS14a), phenyl-hexyl reversed-phase, and ODS C18/C28 columns with HPLC-ICP-MS; arsenosugars were identified by UPLC-MS/MS (Waters Xevo TQD; SIM m/z 409 Sugar-SO4, 329 Sugar-OH). Quantification used SI-traceable NMIJ As(V) (CRM 7912-a) and DMA (CRM 7913-a) standards plus synthesised Sugar-OH/Sugar-SO4 standards. Moisture was determined at 105 °C for dry-mass correction; all values are dry mass. Combined property values were weighted means across the two laboratories with full uncertainty budgets.

Implications

• Certification (HMTc): supplies CRM-quality, dry-mass inorganic-arsenic anchor values for hijiki — the canonical high-iAs seaweed — within the Category 6 seaweed-kelp-foods row (iAs/tAs platform): inorganic arsenic (as As(V)) of 10.1 mg/kg (certified CRM 7405-a) and 24.4 mg/kg (candidate CRM 7405-b), at ≈49% of total in the candidate. These corroborate the hijiki high-iAs sub-treatment within the row’s clean/dirty structuring. Because both are reference materials rather than retail products, the occurrence-extraction pass should treat them as method/anchor points (e.g., evidence-fitness EF-3, or excluded from market-distribution pooling) rather than as market-survey occurrence values. All dry mass.
• Courses: a reference case on the CRM scarcity for arsenosugars and the metrology behind seaweed arsenic speciation.
• App: contributes hijiki inorganic-arsenic anchor values (as As(V)) to the seaweed ingredient contamination_profile, flagged as reference-material data.

Verification notes

• raw_handle MFK_narukawa2020 from the PDF filename; raw_path under “raw/Manual Fetch Kimi /June 8 Inorganic Arsenic Seaweed/“. No DOI was present in the extracted text; publication recorded as Analytical Sciences 2020, 36, 233–239 (The Japan Society for Analytical Chemistry); no_doi_assigned true, access_url null.
• Evidence tier A: peer-reviewed inter-laboratory metrology study with SI-traceable standards and full uncertainty budgets. Note this is a CRM-characterization, not a market survey — occurrence weight is limited to two hijiki reference materials (flagged in Implications for the extraction pass).
• Speciation: total As and inorganic As reported; the inorganic measure here is specifically As(V) (the dominant inorganic species in hijiki; CRM 7405-a is certified for As(V), not for the As(III)+As(V) sum). Recorded as iAs in frontmatter with the As(V)-specific basis stated in Key numbers; As(III) was not separately certified/quantified. Arsenosugars (Sugar-408/328/482/392) and DMA were measured and tabulated but NOT lifted to frontmatter — they are not HMI analyte-vocabulary tokens and recording any of them as iAs would be a speciation error.
• Units/basis preserved exactly (mg/kg dry mass; method LODs in ng/g). The candidate-material combined values (Table 2) are weighted means across the two labs; per-lab values are noted where relevant.
• sample_n=2 reflects the two hijiki reference materials with reported arsenic values (candidate 7405-b and control 7405-a); the inter-lab design used 18 sub-samples per lab (6 bottles × 3 sub-samples), described in the body.
• Brand firewall: not engaged. The materials are research reference materials (NMIJ/NIST CRMs); instrument and reference-material vendor names (Agilent, Thermo, Waters, NMIJ, NIST) are scientific-method reporting, permitted under Exception 2. No food brands.
• Jurisdiction JP (hijiki collected west coast of Japan; NMIJ/AIST). sampling_locations records the west-coast-of-Japan origin.
• matrices [edible-seaweed, macroalgae, dry-weight]: hijiki is a brown alga (Sargassum/Hizikia), not a kelp (Laminaria), so kelp is deliberately omitted here.

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.