B.J. McLain | USGS Science

USGS1994

Methods of analysis by the U.S. Geological Survey National Water Quality Laboratory-Determination of antimony by automated-hydride atomic absorption spectrophotometry

The analysis of natural-water samples for antimony by automated-hydride atomic absorption spectrophotometry is described. Samples are prepared for analysis by addition of potassium and hydrochloric acid followed by an autoclave digestion. After the digestion, potassium iodide and sodium borohydride are added automatically. Antimony hydride (stibine) gas is generated, then swept into a heated quartz cell for determination of antimony by atomic absorption spectrophotometry. Precision and accuracy data are presented. Results obtained on standard reference water samples agree with means established by interlaboratory studies. Spike recoveries for actual samples range from 90 to 114 percent. Replicate analyses of water samples of varying matrices give relative standard deviations from 3 to 10 percent.

Open-File Report↗

USGS1993

Methods of analysis by the U.S. Geological Survey National Water Quality Laboratory-Determination of chromium in water by graphite furnace atomic absorption spectrophotometry

Graphite furnace atomic absorption spectrophotometry is a sensitive, precise, and accurate method for the determination of chromium in natural water samples. The detection limit for this analytical method is 0.4 microg/L with a working linear limit of 25.0 microg/L. The precision at the detection limit ranges from 20 to 57 percent relative standard deviation (RSD) with an improvement to 4.6 percent RSD for concentrations more than 3 microg/L. Accuracy of this method was determined for a variety of reference standards that was representative of the analytical range. The results were within the established standard deviations. Samples were spiked with known concentrations of chromium with recoveries ranging from 84 to 122 percent. In addition, a comparison of data between graphite furnace atomic absorption spectrophotometry and direct-current plasma atomic emission spectrometry resulted in suitable agreement between the two methods, with an average deviation of +/- 2.0 microg/L throughout the analytical range.

Open-File Report↗