Laboratory calibration of chemical volcanic gas sampling techniques using an artificial fumarole

We have constructed an artificial laboratory fumarole to calibrate the most common chemical volcanic gas sampling techniques and obtain a quantitative measure of their efficacy. We have also developed and tested a new rugged and portable venturi spray gas sampler. The venturi sampler reproduced the output gas composition most accurately, followed by the Giggenbach bottles, filter packs, and lastly alkaline traps. Passive alkaline traps, however, did better than filter packs when sampling more concentrated fumarole gases. Under ideal conditions, the accuracy of the Giggenbach bottles was identical to the venturi sampler, although there was slightly more scatter. The Giggenbach sampler was more susceptible to problems with condensation on the input train even in a laboratory setting, and this technique was only effective in relatively concentrated gas streams. Filter packs are also effective, but extreme care must be exercised to maintain strong undersaturation with respect to the acid gas. If strong undersaturation (high pH) is not maintained, the filter packs return erroneously low S/Cl and S/F ratios. Use of a pH indicator is an effective way of avoiding this problem. The passive alkaline traps also under-sample sulfur, resulting in low reported S/Cl and S/F ratios. It appears that the overall sampling efficiency of all techniques was not strongly affected by oxygen fugacity over the limited range tested. When detecting sulfate and sulfite simultaneously, we found no difference in total sulfur before and after oxidation. This suggests that all sulfur from the gas regardless of oxidation state was absorbed as sulfite or sulfate and/or was quickly oxidized in solution. This conclusion is supported by IC HS− reference samples.