H213-09
The Impact of High Salinity on Radiolytic H2 Yield and the Evolution of Subsurface Brine

Wednesday, 16 December 2020: 17:54
Virtual
Devan Nisson1, Oliver Warr2, Barbara Sherwood Lollar2, Thomas L Kieft3 and Tullis C Onstott1, (1)Princeton University, Department of Geosciences, Princeton, NJ, United States, (2)University of Toronto, Department of Earth Sciences, Toronto, ON, Canada, (3)New Mexico Institute of Technology, Department of Biology, Socorro, NM, United States
Abstract:
Radiolysis of water has been shown to be a source of H2 in the marine and terrestrial subsurface that supports chemolithoautotrophic microbial ecosystems. In addition to producing H2, the loss of water from radiolysis may influence the evolution of salinity in deep subsurface fluids. A brine recently discovered in the South African gold and uranium mine, Moab Khotsong at a depth of >2.9 km in the 3.0-2.8 Ga Witwatersrand Supergroup has high salinity (243 ppt), high abundance of H2 (~46%), and non-meteoric δ2H/δ18O values suggesting prolonged hydrogeologic isolation. To examine the effect of high salinity on radiolytic H2 yield and the potential contribution of water radiolysis to subsurface brine formation, the yields for H2 and radiolytic oxidants O2 and H2O2 were simulated as a function of Cl- concentrations. H2 yield increased from 0.44 molecules/100 eV with increasing salinity, with a resultant molecular yield of 0.48 molecules/100 eV for brine. O2 and H2O2 yields also increased for brine, from 0.29-0.34 and 5.7x10-3-6.9x10-3 molecules/100 eV, respectively. Using these results, we estimated the time required to form the Moab Khotsong brine by assuming a dosage rate of ~4x10-3 Gy/(yr-g rock), derived from prior 36Cl/Cl-based estimations of <100 ppm uranium content in the host rock, the brine stopping power (1.14 MeV/(g-cm-2)) and low host rock porosity (~1%). The annual yields for principal radiolytic products were ~70 nM/yr (H2), ~30 nM/yr (O2), ~10 nM/yr (H2O2), equivalent to a loss of ~80 nM/yr (H2O). In relation to lower dosage rates, these results suggest a 3x increase in H2O loss (nM/yr) for every 10x increase in uranium (ppm). If the initial subsurface fluid represented Archaean seawater (0.92 M Cl-) trapped in a closed system then the resulting brine (4M Cl-) would form within >400 myrs taking into account re-addition of H2O from abiotic CH4 production (~20 nM/yr). As such, a major contribution by seawater salinity is not required. If the precursor fluid were 2.0 Ga, low salinity, hydrothermal fluids (similar, but likely lower in starting salinity pre-radiolysis, to fluid inclusions found in fracture minerals of the Witwatersrand Supergroup (~0.4 M Cl-)), brine formation would occur within >900 myrs. These results indicate that water radiolysis is a major contributing factor to subsurface brine formation.