Uncertainty contributions to species specific isotope dilution analysis. Part 2. Determination of methylmercury by HPLC coupled with quadrupole and multicollector ICP-MS

Abstract

Species-specific isotope dilution analysis (IDA) has been used to determine the mass fraction of methylmercury in a fish muscle certified reference material, DORM-2. Two spike solutions were prepared and their mass fractions determined by coupling high-performance liquid chromatography (HPLC) with both quadrupole (Q) and multicollector (MC) inductively coupled plasma mass spectrometry (ICP-MS) and uncertainty budgets calculated. For the Q-ICP-MS determination the relative standard uncertainty of the spike was 4.1%. The major uncertainty contribution (62%) arose from the uncertainties associated with the relative isotopic abundances of Hg isotopes as given by IUPAC. The measured isotope amount ratios (200Hg∶199Hg and 205Tl∶203Tl) contributed 37% of the combined uncertainty. For the multicollector instrument the relative standard uncertainty of the spike solution mass fraction was 0.7%. In this case the uncertainty budget was dominated by the Hg isotopic uncertainties, with a contribution of 99%. The uncertainty contributions from the measured isotope amount ratios were reduced to 0.7% of the total when using MC-ICP-MS. The two spike solutions were separately used to determine the mass fraction of methylmercury in DORM-2 CRM by species-specific IDA, using HPLC coupled with quadrupole ICP-MS. The values found were 4.45 ± 0.90 µg g−1 and 4.25 ± 0.47 µg g−1 using spikes characterised by Q- and MC-ICP-MS, respectively (the uncertainties quoted are the expanded uncertainty, k = 2), compared with a certified value of 4.47 ± 0.32. The major uncertainty contribution for each determination was from the measured isotope amount ratios (70%) with lesser contributions from the uncertainties associated with the natural isotopic abundance of Hg and the spike mass fraction.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.