Efficiency of solvent extraction methods for the determination of methyl mercury in forest soils

Methyl mercury was determined by gas chromatography, microwave induced plasma, atomic emission spectrometry (GC-MIP-AES) using two different methods. One was based on extraction of mercury species into toluene, pre-concentration by evaporation and butylation of methyl mercury with a Grignard reagent followed by determination. With the other, methyl mercury was extracted into dichloromethane and back extracted into water followed by in situ ethylation, collection of ethylated mercury species on Tenax and determination. The accuracy of the entire procedure based on butylation was validated for the individual steps involved in the method. Methyl mercury added to various types of soil samples showed an overall average recovery of 87.5%. Reduced recovery was only caused by losses of methyl mercury during extraction into toluene and during pre-concentration by evaporation. The extraction of methyl mercury added to the soil was therefore quantitative. Since it is not possible to directly determine the extraction efficiency of incipient methyl mercury, the extraction efficiency of total mercury with an acidified solution containing CuSO4 and KBr was compared with high-pressure microwave acid digestion. The solvent extraction efficiency was 93%. For the IAEA 356 sediment certified reference material, mercury was less efficiently extracted and determined methyl mercury concentrations were below the certified value. Incomplete extraction could be explained by the presence of a large part of inorganic sulfides, as determined by x-ray absorption near-edge structure spectroscopy (XANES). Analyses of sediment reference material CRM 580 gave results in agreement with the certified value. The butylation method gave a detection limit for methyl mercury of 0.1 ng g(-1), calculated as three times the standard deviation for repeated analysis of soil samples. Lower values were obtained with the ethylation method. The precision, expressed as RSD for concentrations 20 times above the detection limit, was typically 5%.