Low-temperature superficial chemical changes and post-entrapment effects alter CO2 budget estimation in vapor bubbles of glass inclusions

Abstract:

Quantifying the CO₂ budget in glass inclusions containing a shrinkage bubble has become an important topic with the development of methodology based on microthermometry and micro-Raman measurements [1]. It is possible to determine CO₂ in the shrinkage bubble and to calculate the bulk inclusion+bubble CO₂ content, but should the methodology be refined for natural glass inclusion samples in a volcanic arc context ? We attempt to use the method for quantifying the major gas output (CO₂, H₂O, S, Cl) and understanding the evolution of gas mixture from melt and vapor at San Cristóbal volcano in the Central American Volcanic Arc (CAVA) by taking into account for the first time volatile contents of < 15 - 35 µm glass inclusions by using high spatial resolution/sensitivity NanoSIMS 50L ion microprobe. The presented approach focus on 16 - 527 ppm CO₂ content in uncorrected glass inclusions with the presence of a shrinkage bubble and the added CO₂ content could range between 550 - 1630 ppm. H₂O in the bubble (1-X µm) was quantified with a Horiba Jobin Yvon HR800 micro-Raman spectrometer instrument by developing a curve spectrum correction method to isolate and put into evidence the Raman-Water-Liquid band from the glass-water band background, and by estimating their mass and volume with 1 µm imagery precision. Inside the glass inclusions, NanoSIMS multi-elements intensity imagery allowed to visualize the region where the intersected bubble is found and it should be avoided when programming NanoSIMS spots on the inclusion surface. Carbonates peaks were also detected along the vapor bubble walls by Raman and may be associated with unstable C-phases with a variety of low-°T minerals detected by SEM [2]. The presence of such low °T mineral phases must represent important low °T superficial and hydrothermal effects and may alter the CO₂ budget estimation of primary melts at San Cristóbal.

[1] Moore et al., 2015, Am. Mineral. 100, 806-823.

[2] Kamenetsky et al., 2001, EPSL. 184,  685-702