Determination of trace amounts of arsenic(III) and arsenic(V) in drinking water and arsenic(III) vapor in air by graphite-furnace atomic absorption spectrophotometry using 2,3-dimercaptopropane-1-sulfonate as a complexing agent

As(III), (12.5 - 200 ng) in drinking water (25 ml) reacted with 2,3-dimerca ptopropane-1-sulfonate (DMPS, 0.60 mg) in ammonium acetate buffer (2.0 mmol , pH 5.5) and formed a complex of As(III)-DMPS. The complex was selectively retained on two Sep-Pak C-18 cartridges in series, while As(V) could not b e retained. Each cartridge was eluted with methanol (2.00 ml). After additi on of Ni2+ (2.0 mg), a portion (20 mu l) was introduced into a graphite cuv ette and was atomized according to a temperature program. As(V) (up to 100 ng) in drinking water (25 ml) was pre-reduced to As(III) with L-cysteine. T he concentration of As(V) was determined by subtracting As(III) from the fi nal total concentration of As(III)+As(V). The method detection limits (3 si gma) for As(III) and As(V) were 0.11 mu g/l and 0.15 mu g/l, respectively; the calibration graphs were linear up to 8.00 mu g/l for As(III). Accuracie s of 97.6 - 99.3% were obtained for As(III) and As(V) in drinking water wit h a precision within 4.0%. The proposed method could be applied to the dete rmination of As(III) vapor in air with a detection limit of 2.8 ng.