Determination of methylmercury in natural water samples by steam distillation and gas chromatography atomic fluorescence spectrometry

Steam distillation was evaluated as a technique for the separation of methylmercury from natural water samples prior to quantification by GC-atomic fluorescence spectrometry. Recoveries of methylmercury chloride spikes ranged from similar to 100% in a wide variety of natural freshwater and estuarine samples to 80% in seawater. The addition of ammonium pyrrolidine dithiocarbamate (APDC) was found to improve the recovery of methylmercury chloride spikes in MilliQ water from 73 to 89% and in seawater from 80 to 85%. Codistillation of inorganic mercury was eliminated by addition of APDC to the samples. Precision in MilliQ water was 2.0% RSD at 0.2 ng L-1 CH3HgCl (n = 10) and 1.6% RSD at 2.0 ng L-1 CH3HgCl (n = 10). The limit of detection for the method was 0.024 ng.L-1 (3 sigma) for a 50 mL sample. The steam distillation procedure was tested for and found free of measurable artifactual formation of methylmercury. The method was compared to nitrogen-assisted distillation and found to give comparable results with the added advantage of handling sample sizes up to 100 mL. Compared to previously employed separation procedures, steam distillation offers the advantages of robustness and a considerably increased sample throughput (at least four samples per hour) without any compromise in analytical performance.