Selective medium reactions for the 'arsenic(III)', 'arsenic(V)', dimethylarsonic acid and monomethylarsonic acid determination in waters by hydride generation on-line electrothermal atomic absorption spectrometry with in situ preconcentration on Zr-coated graphite tubes

Methods were developed for the trace determination of total arsenic, 'As(III)', 'As(V)', dimethylarsonic acid (DMAA) and monomethylarsonic acid (MMAA) in sea and hot-spring waters by hydride generation/trapping and atomization in a graphite furnace using coated graphite tubes. Selective medium reactions were used to generate selectively hydrides of each arsenic species: 1. reduction with NaBH4 in thioglycollic acid medium (total arsenic); 2. reduction with NaBH4 in hydrochloric acid medium (inorganic arsenic,'As(III)' and 'As(V)'); 3. reduction with NaBH4 in citric acid/sodium hydroxide buffer ('As(III)'); and 4. reduction with NaBH4 in acetic acid ('As(III)' and DMAA). The 'As(V)' and MMAA can be evaluated by difference. Iridium, tungsten and zirconium-coated graphite tubes were investigated for the in situ preconcentration of arsenic hydrides. Trapping onto Zr-coated graphite tubes offers the best analytical performances, with detection limits of 56 ng l(-1) for total arsenic (first selective medium reaction); 50 ng l(-1) for inorganic arsenic and 'As(III)' and DMAA (second and fourth selective medium reaction), and 80 ng l(-1) for 'As(III)' (third selective medium reaction). (C) 1998 Elsevier Science B.V. All rights reserved.