Determination of trace levels of dimethylsulfoxide (DMSO) in seawater and rainwater

A procedure for the trace analysis of dimethylsulfoxide (DMSO) in seawater and other aqueous solutions has been developed. The method is based on reduction of DMSO to dimethyl sulfide (DMS) by titanium trichloride (TiCl3) and the subsequent measurement of DMS by gas chromatography. As little as 1 picomole of DMSO could be detected by the method yielding a detection limit of 1 nM for a 1 ml sample. Precision for most natural water samples with 5–20 nM DMSO was better than 10%. Water samples stored in the dark and either frozen (−20°C) or acidified (0.05 M HCl; 4°C) showed no significant losses or contamination over a 3 week period. The reduction reaction was selective for DMSO, in that other sulfur compounds including methionine, homocysteine, cysteine, glutathione, dimethyl sulfone, methionine sulfoxide, dimethylsulfoniopropionate (DMSP) and S-methylmethionine did not yield DMS. However, a required basification step in the procedure caused conversion of DMSP to DMS, therefore parallel samples without TiCl3 were used to correct for dissolved DMSP or DMS in the solution analyzed. Tests showed that DMSP was the only compound which yielded DMS during the basification step. Significant analytical blanks were observed if precautions were not taken to minimize air contamination of glassware, reagents and the analyte solution. Detailed information on how to minimize blanks is given. Using the method we measured seawater DMSO concentrations ranging from 1 to 13 nM and rainwater DMSO concentrations ranging from 1 to 26 nM. A preliminary incubation experiment with water from Mobile Bay, Alabama indicated that added DMSO (50 nM) was consumed over a 5 day period. The method should be useful in studies of DMSO and DMS biogeochemistry.