Diffusional isotope fractionation of singly and doubly substituted isotopologues of H2, N2 and O2 during air-water gas transfer

Abstract

Air-water gas transfer largely influences the geochemical and biogeochemical cycles of essential atmospheric components (e.g. O2 and CO2), in which gas molecular diffusion in water is recognized as the rate limiting step. Isotope compositions in these gas molecules are useful tools to quantify this mass transfer process, in which diffusional isotope fractionation factors (i.e. αdiff) are the key intrinsic parameters. These αdiffs are often determined by gas transfer experiments with large uncertainties because the roughness of water surface can affect the interpretation of experimental data. In this study, molecular dynamic simulations were employed to investigate directly the diffusional isotope fractionation for singly and doubly substituted isotopologues of H2, N2, and O2. The results show that diffusional isotope fractionation factors are dependent on both the molecular mass and moment of inertia, which is consistent with previous findings for polyatomic molecules rather than for monoatomic ones. When comparing with the kinetic isotope fractionation (i.e. αk) determined by gas transfer experiments, I found that αk is likely close to (αdiff)1/2 within errors (i.e. αk = (αdiff)1/2), rather than to (αdiff)2/3 that has often been employed to calculate αdiff using αk in literature (i.e. αk = (αdiff)2/3). If this is the case, the results further indicate that the nuclear quantum effect is not significant when αdiff is of interest. With these findings, I determined the isotope fractionation relationship θ for different O2 isotopologues to be 0.5100 ± 0.0002 and 1.9535 ± 0.0013 respectively for 17θdiff (≡ln17αdiff/ln18αdiff) and 36θdiff (≡ln36αdiff/ln18αdiff) as an example.