Molecular hydrogen in galaxies

Abstract

This study aims to understand the key role played by molecular hydrogen in the evolution of galaxies, with a view to constraining its radial distribution in the Galaxy and the CO→H(_2) conversion factor α(_20).The star formation rate is shown to be correlated with the surface density of H(_2). A correlation between the molecular hydrogen fraction and the metallicity of a region allows the time evolution of H(_2) to be described. This leads to a modified 'Schmidt Law' of the SFR which explains quite naturally the production of galactic metallicity gradients and the constancy of the SFR in the absence of infall. A consistent closed model of the chemical evolution of the Galaxy is proposed to solve the G-dwarf problem, the stellar age-metallicity relation and the metallicity gradient, leading to the prediction of some initial amount of pre-disc processing of gas into visible and dark matter. It is found that a constant yield of metals is more appropriate than a yield proportional to metallicity. Possible time variations of the returned fraction, the dark matter fraction and the SFR are also studied. For consistency, we suggest that dark matter in the solar neighbourhood could be totally baryonic provided the Miller-Scalo IMF is modified at the lower end, that is, the dark matter resides in low mass stars or brown dwarfs. The production of metallicity gradients in spiral galaxies is shown to be a direct consequence of the radial variation of the total surface density of matter and the age of the disc. The role of molecular gas in the evolution of the Oort Cloud of comets is examined. It is shown that comet showers with a mean interval of ̴̱ 30My cannot be produced using perturbations of the Oort Cloud by known stars or molecular clouds. If there is indeed an apparent 30My periodicity in the terrestrial mass extinction and geological records, we argue that astronomically induced processes are unlikely to be the primary cause. Evidence is presented that the lifetime of the molecular gas phase is ≤ 2.lO(_8)y, and arguments, particularly from CO observations of the Virgo galaxy cluster, favouring longer lifetimes are shown to be not well founded. We suggest that the ICM in Virgo reduces the value of α(_20) as compared to isolated galaxies. From the above considerations, the radial distribution of in the Galaxy is derived and shown to agree in the inner Galaxy with that derived from ɤ-ray analysis. In the solar neighbourhood we find α(_20) = 2.5±0.5, and present evidence that α(_20) varies as a function of Galactocentric radius and from galaxy to galaxy.