• Journal of the Korean earth science society
• /
• v.34 no.7
• /
• pp.681-692
• /
• 2013

We aimed to evaluate the effectiveness of the Giggenbach bottle method and develop the related pretreatment and analytical methods using artificial fumarolic gases. The artificial fumarolic gases were generated by mixing $CO_2$, CO, $H_2S$, $SO_2$, $H_2$, and $CH_4$ gas streams with a $N_2$ stream sparged through an acidic medium containing HCl and HF, with their compositions varied by adjusting the gas flow rates. The resultant fumarolic gases were collected into an evacuated bottle partially filled with a NaOH absorption solution. While non-condensible gases such as CO, $H_2S$, and $CH_4$ accumulated in the headspace of the bottle, acidic components including $CO_2$, $SO_2$, HCl, and HF that were dissolved into the alkaline solution. Like other acidic components, $H_2S$ also dissolved into the solution, but it reacted with dissolved $Cd^{2+}$ to precipitate as CdS when $Cd(CH_3COO)_2$ was added. The non-condensible gases were analyzed on a gas chromatography. Then, CdS precipitates were separated from the alkaline solution by filtration, and they were pretreated with $H_2O_2$ to oxidize CdS-bound sulfide into sulfate. In addition, a portion of the solution was also pretreated with $H_2O_2$ to oxidize sulfite to sulfate. Following the pretreatment, the resultant samples were analyzed for $SO_4^{2-}$, $Cl^-$ and $F^-$ on an ion chromatography. In the meanwhile, dissolved $CO_2$ was analyzed on a total organic carbon-inorganic carbon analyzer without such pretreatment. According to our experimental results, the measured concentrations of the fumarolic gases were shown to be proportional to the gas flow rates, indicating that the Giggenbach bottle method is adequate for monitoring volcanic gas. The pretreatment and analytical methods employed in this study may also enhance the accuracy and reproducibility of the Giggenbach bottle method.