General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms ). Here we test the hypothesis that abiogenic H2 can be produced from rock-water reactions at 0°C in sufficient quantities to be able to support subglacial microbial activity.The ability of six silicate rock types samples from glacial catchments (gneiss, quartzite, shale, granite, nepheline-syenite, schist) to generate abiogenic H2 in water-rock reactions were tested. Calcite was used as a non-silicate control. The molar compositions of the starting materials are given in Supplementary Table S1. The rocks and mineral control were crushed to a range of different surface areas under an inert atmosphere, wetted with water, and the generation of H2 followed over time.All six silicate rock types produced H2 when crushed and wetted with water at 0°C (Figure 1). There was no detectable H2 generation in experiments with calcite (< 2.7 nmol H2 , with the free radicals formed through the shearing of surface mineral bonds during subglacial rock comminution. We note that free radicals were not detected in our crushed calcite, consistent with a lack of detectable H2 in these control experiments.Equation 1 For all three silicate rocks tested, rates of H2 generation were similar at 0°C and 10°C, but higher at 35°C (Figure 2). This temperature dependence may be due to the transformation of more stable SiO radicals to more reactive Si radicals at higher temperatures 15 .The amount of experimental H2 generated likely underestimated in situ abiogenic H2 subglacial production, for the following reasons. First, the shape of many of the H2 generation curves in Figures 2 and 3 suggest that production may continue for longer than the 120 hour experimental period. Second, it is likely that less stable surface radicals will be lost during and after dry crushing, prior to the addition of water. Third, H2 may be produced by additional water-rock interactions such as serpentinization 12,13 , although these are untested at subglacially relevant temperatures. Finally, additional H2 could have been released from fluid inclusions 18 during the initial dry crushing stage of our methods.Our experimental data suggests that the amount of H2 generated from subglacial rock comminution depends primarily on the rate of rock abrasion, the presence of liquid water, and We crudely estimated the catchment scale H2 generation by rock comminution for one of our study sites; the 600 km 2 Leverett Glacier, SW Greenland. We first used the regression equation in Figure 1c to calculate the potential H2 that could be generated from silica radicalwater reactions, using the measured mean molar silica surface area of suspended sediment from the catchment (Methods). We then scaled up our calculations using measurements of the annual suspended sediment flux from the glacier , and a diversity of methanogens 8,25 . Prior studies on subglac...

No abstract