AbstractThe objective of this study was to develop a technique for collecting and analyzing volatile arsenic species including arsine (AsH3), methyl arsine (CH3AsH2), dimethyl arsine (CH3)2AsH, and trimethyl arsine (CH3)3As emitted from paddy soil. The four volatile arsenic species were captured on AgNO3‐impregnated sand then eluted and quantified. The AgNO3‐impregnated sand was synthesized by crystallizing AgNO3 on 0.5‐1 mm id. sandy particles at 70 oC. The volatile arsenic species were generated by hydride a mixture containing sodium arsenous (NaAsO2), disodium methyl arsenate ((CH3AsO(ONa)2), sodium cacodylate ((CH3)2AsO(ONa)), and trimethyl arsine oxide ((CH3)3AsO) using NaBH4. The formed volatile arsenic species were purged with an Ar stream then retained on a trap packed with the AgNO3‐impregnated sand followed by elution with 2 mL of 2.5% nitric acid and 0.5 mL of 30% H2O2 at 70oC prior to total analysis by ICPMS and speciation analysis by ion exchange chromatography ‐ ICPMS. The total analysis of arsenic revealed that AgNO3‐impregnated sand can efficiently retain the four volatile arsenic with 91‐102% recovery within a range of 20‐2000 ngAs as total amounts. The detection limit for total arsenic species through the trap was 6.5 ngAs. For a trap packed with 300 mg of AgNO3‐impregnated sand, the capture capacity was greater than 2000 ngAs. The speciation of arsenic eluted from the trap was performed by separation on a Hamilton PRP‐X100 anion‐exchange column with gradient elution using 100 mM (NH4)2CO3 mobile phase followed by ICPMS detection. A pot experiment was conducted to evaluate the efficiency of the purge and trapping method for naturally formed volatile arsenic species emitted from flooded paddy soil. Preliminary results showed that methyl arsenic, trimethyl arsenic and arsine were emitted from the studied paddy soil in which arsine contributes 90% of arsenic emission. This method is showing promise to be applied for the emission of volatile arsenic analysis from paddy soil as well as other solid types of samples.
Read full abstract