TESTING THE USE OF δ18O IN EXTRACTED WATER FROM VOLCANIC ASH, KAOLINITE, AND MICA AS A TOOL FOR PALEOCLIMATE AND PALEOALTIMETRY STUDIES: THE SEARCH FOR NEW STANDARDS AND METHODOLOGY

Recent experimental work by Nolan and Bindeman (2013) demonstrated that, even at room temperature, the δD of water in ash equilibrates with the changing δD of surrounding water, putting limitations on paleoclimate or paleoaltimetry if the δD of precipitation has changed since the ash was deposited. This same process may apply to secondary micas. In contrast, their work showed that the δ¹⁸O of water in glass did not change through time. Accurate analysis of the δ¹⁸O of extracted water from hydrated glass and mica samples is the purpose of our research. Although we have the means to determine relative changes in δ¹⁸O of extracted water by rapid thermal decomposition and pyrolysis using the TC/EA (High Temperature Conversion Elemental Analyzer) and a continuous flow mass spectrometer, there are some difficulties with this approach: water is bound as both OH^- and H₂O_mol in many samples, internal fractionations between total water and silicate O are poorly constrained, it is unknown if thermal decomposition induces internal isotopic exchange, and there is no known standard for these analyses.

We report initial results for δ¹⁸O_water of thermally decomposed biotite of known bulk δ¹⁸O. Micas contain all water as OH^- and these are currently also used as δD standards. Two separate biotite samples were homogenous in terms of their δ¹⁸O_water, yielding standard deviations of 0.9 ‰ (n=11) and 0.7 ‰ (n=30) respectively. In addition, we have analyzed experimentally synthesized hydrous rhyolite and dacite glasses (hydrated to ~2.4 wt.% H₂O at 800 °C) of known δD and δ¹⁸O_bulk. These samples contain nearly equal amounts of OH^- and H₂O_mol and are being tested as δ¹⁸O_water standards. The δ¹⁸O of the extracted water from the rhyolite and dacite glasses have standard deviations of 0.1 (n=2) and 0.6 (n=3) respectively. We are also testing kaolinite, brucite, and ash with variable total water and OH^-/H₂O_mol ratios from the 7.7 ka eruption of Mt. Mazama as possible standards for δ¹⁸O analyses on the TC/EA. Our second goal is to determine if the δ¹⁸O of extracted waters from micas and hydrated glasses can in principle have enough Δδ¹⁸O resolution to resolve diverse water δ¹⁸O values beyond errors to have a practical significance, which will improve the method of determining a region’s paleoclimate or paleoelevation.