Major sources of uncertainties in the analysis of methylmercury using gold amalgamation sampling

Abstract

In an effort to improve the accuracy of measurement techniques for methylmercury (MeHg), the possible experimental bias in its analysis by the gold amalgamation method was investigated with special reference to gaseous elemental mercury (GEM). To this end, a series of calibration experiments were conducted using liquid-phase standards of MeHg prepared at four concentrations (2, 4, 10, and 20 ng mu L-1) at each of four injection volumes (between 0.2 and 2 mu L). Calibration results of MeHg were then examined mainly in terms of two calibration approaches: a fixed standard concentration (FSC) method (variable volumetric injection of standards prepared at a given concentration) and fixed standard volume (FSV) method (injection of multiple standards with varying concentrations at a given volume). The calibration results of MeHg, when evaluated against those of GEM, indicate that the FSC-based calibration of the former is unreliable due to the deterioration of linearity with increasing loading volumes of liquid standard. Thus it is recommended that the FSV approach should be used as the liquid sample loading method to improve the robustness and to minimize the calibration bias of MeHg.