Automatic determination of iron in geothermal fluids containing high dissolved sulfur-compounds using flow injection electrothermal atomic absorption spectrometry with an on-line microwave radiation precipitation–dissolution system Original Research Artic

A flow injection sample introduction in electrothermal atomic absorption spectrometry was used to determine iron in geothermal fluids containing high concentration of dissolved sulfate and sulfide anions. First, sulfide was precipitated as sulfur with a confluencing solution of hydrogen peroxide (10% v/v) on the walls of a knotted reactor exposed to microwave radiation, to ease the precipitation process. While the sulfate containing solution flows downstream and sequestered in a sampling arm for the sequential deposition of aliquots of sample on the graphite tube atomizer platform by means of positive air displacement, sulfur was dissolved with CCl4 and diverted to waste. This sequence was timed to synchronize with the spectrometer computer, which had been pre-programmed to introduce aliquots of Lu modifier, previous to each sample deposition, in order to minimize the sulfate interference, to improve integrated absorbance signal and the reproducibility of measurements. By using the less sensitive iron line at 296.7 nm, the linear range was from 12 to 280 μg l−1, with a characteristic slope of 0.00083, a characteristic mass of 104.8 pg and a detection limit of 72 pg (3.6 μg l−1). The precision of the method, obtained for 10 measurements, was studied as within-run and within-batch precision; the relative standard deviations obtained in both cases were less than 3%. The accuracy of geothermal fluids has been checked by recovery tests and comparison with an independent analytical method. Results obtained by the here proposed method correlate closely with those obtained by a tedious photometric procedure with o-phenanthroline. The sensitivity and the simplicity of this on-line procedure makes it attractive for routine determination of iron in waters involving large throughput of samples.