Synchronous Extraction and Quantitative Speciation of Arsenic and Chromium in Sediments by High-Performance Liquid Chromatography – Inductively Coupled Plasma – Mass Spectrometry (HPLC-ICP-MS)

The optimum chromatographic separation conditions, polyatomic ions interference correction, and extraction parameters were investigated for simultaneous extraction and quantification of As(III), dimethylarsinic acid (DMA), monomethylarsonic acid (MMA), As(V) and Cr(VI) with excellent resolution. The ammonium salts (NH4H2PO4 and NH4NO3) were examined as mobile phases at different pH values by altering the composition during the chromatographic run in gradient elution mode using high-performance liquid chromatography – inductively coupled plasma – mass spectrometry (HPLC-ICP-MS). The 10 mM and 70 mM NH4NO3 composition at pH 8.7 enabled sufficient baseline separation of the target As and Cr species at a total run time of 18 minutes. The identified polyatomic ion interferences were corrected by an inter-element method using mathematical equations. The optimum extraction conditions containing a mixture of 0.375 M (NH4)2HPO4 and 50 mM EDTA at neutral pH as an extraction buffer, extraction temperature of 150 °C and volume of 15 mL were attained by a factorial design technique. The lowest concentrations of the selected elemental species that can be reliably detected by the procedure were in the range from 0.064 to 0.682 ng g−1. The accuracy of the developed procedure was demonstrated by achieving percentage recoveries of 96.8% for As(V) and 97.7% for Cr(VI) using San Joaquin Soil standard reference material 2709a. Proposed method was validated by achieving percentage recoveries from 84.9 to 104% for As species and Cr(VI) using spiking and recovery studies at 3x and 25x the limit of quantification values of the target species. The selected As and Cr species in real sediment were monitored using the optimized analytical method. The differences in concentrations of Cr(VI), As(III), MMA and As(V) between the seasonal state of the river were statistically significant (p ˂0.05) at a 95% confidence level as determined by analysis of variance (ANOVA).