Determination of arsenic in petroleum refinery streams by electrothermal atomic absorption spectrometry after multivariate optimization based on Doehlert design

This paper reports the development of a methodology for the determination of arsenic in petroleum refinery aqueous streams containing large amounts of unknown volatile organic compounds, employing electrothermal atomic absorption spectrometry with polarized Zeeman-effect background correction. In order to make the procedure applicable, the influence of chemical modification and the drying step was examined. Also, pyrolysis and atomization temperatures and the amount of nitric acid added to the sample were optimized using a multivariate approach based on Doehlert matrix. Obtained results indicate that, in this kind of sample, arsenic must be determined by standard addition procedure with a careful control of the drying step temperature and ramp pattern. In order to evaluate the accuracy of the procedure, a test was performed in six spiked samples of petroleum refinery aqueous streams and the relative errors verified in the analysis of such samples (added As between 12.5 and 190 μg l−1) ranged from −7.2 to +16.7%. The detection limit and the relative standard deviation were also calculated and the values are 68 pg and 7.5% (at 12.5 μg l−1 level), respectively.