5.1 The carbon isotope balance in our experiments
The δ13C signatures of methane vary around a value of –55‰ (V-PDB), with FT18-4 as the one outlier with a δ13C of –44‰ (Fig. 2). All methane samples have depleted δ13C compared to the starting CO (–28‰). The isotopic difference between CH4 and the starting CO of ~ 25‰ likely reflects13C/12C fractionation during methane synthesis. Kueter et al., (2019) determined the equilibrium13C/12C fractionations between CO, CO2 and CH4 at various temperatures between 1200 °C and 300 °C. At 300 °C, the equilibrium13C/12C fractionation between CO and CH4 is –3.1‰, meaning that CH4 should be enriched in 13C relative to CO by 3.1‰. This is at odds with our observation of a ~25‰ to ~40‰ depletion in13C/12C in CH4relative to the initial reactant CO. Instead methane δ13C values are broadly consistent with an approach to13C/12C equilibrium between methane and CO2. This implies that the synthesis of CO2 from CO oxidation via reaction 4 was nearly quantitative so that CO2 had nearly the same13C/12C as the initial CO reactant (–28‰). Subsequent 13C/12C equilibration between CH4 and this CO2would yield methane with a 25‰ 13C depletion relative to CO2 at 250 °C, and roughly 40 ‰ at 150 °C (Kueter et al., 2019), corresponding to predicted methane δ13C values of ~ –53‰ and −68‰ at 250 and 150 °C, respectively. This data supports near-complete conversion of CO to CO2 and complete equilibration between CH4 and CO2. While imperfect, the magnitude of the shift in δ13C from the initial CO reactant values to the product methane are generally consistent with these CH4-CO2 equilibrium values. In this context, the scatter in the δ13C values, and imperfect correlation with T (Fig. 2a) would reflect various degrees of departure from full isotopic equilibrium and/or conversion of CO to CO2. The FT18-4 outlier would reflect the furthest removal from CH4 -CO2 equilibrium, and/or incomplete conversion of CO to CO2, consistent with this experiment being conducted at the lowest temperature. The alternative interpretation is that the CH4δ13C values reflect a kinetic carbon isotope effect involving the reduction of CO. In this case the correlation with temperature, albeit imperfect, would perhaps reflect a reservoir effect with a single kinetic fractionation factor, or a temperature-dependent kinetic fractionation factor.
5.2 The hydrogen isotope balance in our experiments
The δD signatures of methane range from –610 to –580‰ V-SMOW. If methane is in D/H equilibrium with experimental water (–119‰ V-SMOW), the methane δD would fall between –250 and –350‰ for experiments conducted at temperatures between 250 and 130°C (Horibe and Craig, 1995). Thus, the observed δD signatures of methane of roughly –600‰ reflect unambiguous D/H disequilibrium with water. During FTT synthesis, methane molecules form via a catalytic surface reaction of sequential hydrogen additions, presumably sourced directly from H2. It is expected, therefore, that it is the D/H ratio of H2 that is a dominant factor in controlling the D/H of methane in our experiments, not that of water. No D/H measurements of H2 are available in this study, but at equilibrium, CH4 should be ~ 540‰ higher in δD than coexisting H2 at ~ 200°C (Horibe and Craig, 1995). In lieu of D/H for H2 in our experiments, we may make use of three FTT experimental studies that have reported D/H of both of CH4 and H2 to infer the likely extent of isotopic equilibrium between these species. All of these studies report D/H ratios in CH4 and H2 are within 150 ‰ of each other (Fu et al., 2007; McCollom et al., 2010; Taran et al., 2010), far from the ~ 500‰ difference expected at equilibrium. This suggests that abiotic methane synthesized in the laboratory is not in isotopic equilibrium with H2. We can explore this further if we assume H2 formed in D/H equilibrium with water in our experiments. In this case δD for H2 is predicted to be approximately −560‰ at 200oC, in equilibrium with the liquid water δD of -119‰ (Horibe and Craig, 1995; Rolston et al., 1976). Since the product methane has δD values of approximately –600‰ at this T, it appears that the CH4 has D/H similar to that of the reactant H2, as in previous experiments. Under strict isotopic equilibrium between methane and an infinite pool of H2, CH4 should be ~ 540‰ higher in δD than coexisting H2 at ~ 200°C. Similar D/H for CH4 and H2 would suggest that either the methane consumed most or all of the H2 available locally, or that there is a D/H kinetic fractionation that offsets the equilibrium fractionation by nearly 500‰. Overall, we conclude that a combination of equilibrium fractionation between H2O and product H2, followed by synthesis of methane with similar D/H than reactant H2 explains the methane δD values of approximately –600‰ obtained here.
The predicted range in δD values relative to VSMOW for H2 formed in equilibrium with the effectively infinite reservoir of liquid H2O from ~ 170 to 250 °C is −588 to −513‰, respectively, defining a slope of 0.44‰/°C over this temperature range. This is similar to the observed positive correlation between methane δD values with T in our experiments, with a slope of ~0.38‰/°C, lending further support for the inheritance of D/H from H2 equilibrated with water, with the exclusion of the 130°C outlier FT18-4 (Fig. 2). This interpretation of inheritance of D/H from H2 convolved with a kinetic isotope effect is consistent with previous similar published experiments in which the D/H values of CH4 and H2 are within a few tens of permil of one another (e.g., McCollom et al., 2010).