Simultaneous determination of inorganic mercury, methylmercury, and total mercury concentrations in cryogenic fresh-frozen and freeze-dried biological reference materials.

Two speciated isotope dilution (SID) approaches consisting of a single-spike (SS) method and a double-spike (DS) method including a reaction/transformation model for the correction of inadvertent transformations affecting mercury species were compared in terms of accuracy, method performance, and robustness for the simultaneous determination of methylmercury (MeHg), inorganic mercury (iHg), and total mercury (HgT) concentrations in five biological Standard Reference Materials (SRMs). The SRMs consisted of oyster and mussel tissue materials displaying different mercury species concentration levels and different textural/matrix properties including freeze-dried (FD) materials (SRMs 1566b, 2976, and 2977) and cryogenically prepared and stored fresh-frozen (FF) materials (SRMs 1974a, 1974b). Each sample was spiked with (201)iHg (Oak Ridge National Laboratory, ORNL) and Me(202)Hg (Institute for Reference Materials and Measurements. IRMM-670) solutions and analyzed using alkaline microwave digestion, ethylation, and gas chromatography inductively coupled plasma mass spectrometry (GC/ICP-MS). The results obtained by the SS-SID method suggested that FF and FD materials are not always commutable for the simultaneous determination of iHg, MeHg, and HgT, due to potential transformation reactions resulting probably from the methodology and/or from the textural/matrix properties of the materials. These transformations can occasionally significantly affect mercury species concentration results obtained by SS-SID, depending on the species investigated and the materials considered. The results obtained by the DS-SID method indicated that the two classes of materials were commutable. The simultaneous and corrected concentrations of iHg, MeHg, and HgT obtained by this technique were not found to be statistically different form the certified and reference concentration together with their expanded uncertainty budgets for the five SRMs investigated, exemplifying the robustness, the accuracy, and the improved commutability of this method compared to SS-SID measurements.