RESUMO
Because stabilization of arsenite in water samples during transit and storage is troublesome, this work deals with a method to prevent this by on-site selective coprecipitation of arsenite with dibenzyldithiocarbamate and recovery of the coprecipitate by filtration through a 0.45-microm membrane filter. In the laboratory arsenic on the filter is quantitatively released by oxidation of arsenite to arsenate with H2O2 (6%) in alkaline medium (8 mmol L(-1) NaOH) at elevated temperature (85 degrees C) for 30 min followed by ultratrace determination by routine HGAFS and ICP-MS. It is shown that arsenate contamination of the coprecipitate is so low that arsenate concentrations three orders of magnitude higher than the arsenite concentration do not interfere; this is essential, because arsenate is usually the dominant arsenic species in water. Because significant preconcentration can be achieved in the solution obtained from the leached filter (normally a factor 20 but easily increased to 100) very low detection limits can be obtained (only limited by the purity of the materials and the cleanliness of working); a realistic limit of determination is 0.01 microg L(-1) arsenite. The procedure was used for the determination of arsenite in two ground waters from an ash depository site in the Salek valley (Slovenia). The matrix contained some elements at very high levels but this did not impair the efficiency of arsenite coprecipitation. The results obtained by use of HGAFS and ICP-MS were not significantly different at the 5% level for sub-microg L(-1) arsenite concentrations.
