1997A&A...327..453C

We review here the main direct or indirect ways to detect the possible presence of large amounts of cold molecular hydrogen in the outer parts of disk galaxies, an hypothesis that we have recently developed. Direct ways range from H₂ absorption in the UV domain to detection of the radio hyperfine structure: the ortho-H₂ molecule has an hyperfine, or ultrafine, structure in its fundamental state, due to the coupling between the rotation-induced magnetic moment, and the nuclear spin. This gives rise to 2 magnetic dipole transitions, at the wavelengths of 0.5 and 5.5km. Indirect ways are essentially the detection of the HD and LiH transitions, and in some environments like clusters of galaxies, more heavy trace molecules such as CO. We discuss from this point of view the recent discovery by COBE/FIRAS of a very cold Galactic dust component (4-7K) which could correspond to a dominating gas mass component of the ISM, if interpreted as standard dust emission. Some of the proposed means could be applied to the well-known molecular clouds, to bring some new light to the problem of the H₂/CO conversion ratio.