Measuring interstellar gas-phase D/H ratios in the presence of H2

Abstract

Aims. To clarify the circumstances under which it is acceptable to approximate the interstellar gas-phase D/H ratio by D I/H I. Methods. Observed column densities of H I, D I, H 2 and HD are compared for six lines of sight having appreciable fractions of H 2 . Results. The overall fraction of deuterium in HD varies by a factor 3-4 but is found to be much smaller than the fraction of H in H 2 in all cases, implying that deuterium appears as D I and N(D I)/N(H I) exceeds the gas-phase D/H ratio in H 2 -bearing gas. Conclusions. Deuterium in H 2 -bearing gas contributes to the observed D I absorption and the D/H ratio should be inferred from N(D)/N(H) where N(D) = (N(D I) + N(HD)), N(H) = N(H I) + 2N(H 2 ): failure to do so biases the resulting D/H ratio upward, typically by 5%-15% in present data. Along sightlines with multiple kinematic components having different molecular fractions, fractionation can cause velocity differences between D I and H I profiles. Shifts between H 2 and HD velocity centroids may arise when the molecule-bearing gas has kinematic substructure reflecting regions of different ionization balance and HD/H 2 ratios.