The analysis of volcanic brines by freezing stage tandem LIBS-ICP-MS

Volcanic crater lakes pose unique hazards to the local population both during eruption and quiescence. The composition of crater lake fluids is therefore commonly monitored to aid hazard prediction and mitigation. Some of these fluids, such as at Kawah Ijen in Indonesia and Poas in Costa Rica, are hyper-acidic, concentrated brines (SO4-5 wt%, Cl - 2.5 wt%, Al - 5000 ppm). Analysis of such fluids requires extreme dilution due to the high total dissolved element load, which risks contamination and precipitate formation. To circumvent this undesired sample preparation and speed up the turn-around time on the analysis, we tested approaches which involved freezing the fluids to -30 degrees C and analysing them by simultaneous Laser Ablation-Inductively Coupled Plasma Quadrupole Mass Spectrometry (LA-ICP-QMS) and Laser-Induced Breakdown Spectroscopy (LIBS). LIBS is highly matrix sensitive, requiring matrix-matched reference materials. At the same time, LA-ICP-MS owes its success to being relatively insensitive to the matrix after accounting for differences in ablation yield. Analysing the fluids as a solid and sampling them by laser circumvents the usual dilution step. Fluid reference materials were prepared by adding the elements of interest to one of the crater lake fluid samples at known, but varying concentrations and by stepwise dilution of one of these fluids. Differences in ablation yield, resulting mainly from variations in the coupling of the laser with the sample, were accounted for by adding a known quantity of lithium as an internal standard. Working curves for both ICP-MS and LIBS were obtained for Al, Mg, Fe, Ca, K, Na, S and Cl, yielding relative trueness values for 4 crater lake samples of known composition of Al-7%, Ca-5%, Fe-1%, K < 1%, Mg-4%, Na-5%, S-3%, and Cl-6% for ICP-MS and Al-2%, Ca-9%, Fe-17%, K-10%, Mg-4%, Na < 1%, S-8%, and Cl-9% for LIBS. For these elements, ranging in concentration from around 400 mg/kg to 25 g/kg, LIBS yielded similar trueness results to ICP-MS. LIBS calibration without an internal standard produces acceptable results for Li only, allowing the cryo-LIBS to analyse the major elements while optimising the ICP-MS for trace-element analysis and obtaining a full suite of elements simultaneously without the need for sample dilution.