1996A&A...307..237L

We have observed galactic λ3mm HCO⁺ absorption along 30 lines of sight toward compact extragalactic mm-wave continuum sources, using the Plateau de Bure Interferometer. Such absorption is surprisingly common, occurring approximately 30% as often as λ21cm absorption in H I. In fact, the mean free path for detection of absorption with optical depth of unity is the same in both species. The slope of the HCO⁺ optical depth and column density distribution functions is very similar to that of H I but the mean HCO⁺ linewidth is much smaller, 0.95km/s. A search for accompanying HCO⁺ emission finds only an occasional weak (0.05K) line, even when τ(HCO⁺)>1, so that quite accurate column densities may be derived from observations of only the lowest line. In comparing CO (emission) and HCO⁺ absorption, little consistency is at first apparent. However, we can now show that the strength of CO emission increases abruptly when N(HCO⁺)≃1-2x10¹²cm^–2. This is the same phenomenon observed in uv-absorption studies at N(H₂)≃4x10²⁰cm^–2 and is understandable in terms of the onset of dust- and self-shielding at extinctions below 1 magnitude in diffuse clouds or the outer regions of dark clouds. A similar turn on occurs in N(H₂CO) at the same values of N(HCO⁺). Comparison of CO and HCO⁺ emission and absorption can also be used to derive the physical conditions in the clouds; in general we find , if T_K≥10K. This is within but at the high end of the range for diffuse clouds studied optically and well below values appropriate to dark cloud cores. In a few cases, the presence of a relatively large electron fraction (X(e)≲1-3x10^–4) seems indicated, along with temperatures of 20-30K. Thus the excitation analysis is consistent with the existence of a CO turn on. Although it is somewhat surprising that relatively strong CO and (especially) ¹³CO emission is present when carbon is only partially recombined to and bound up in carbon monoxide, this is not atypical in the outer regions of dark clouds. We have also surveyed λ18cm OH absorption in six directions using the VLA. Unlike most other species, but in keeping with traditional notions of diffuse cloud chemistry, there is a remarkably uniform relationship between the OH and HCO⁺ column densities, even at low values. We find X(HCO⁺)=0.03-0.05X(OH) for 1x10¹²cm^–2≲N(OH)≲20x10¹²cm^–2; this is within 40% of values quoted for TMC-1, but at 100 times lower column density. Using relevant values of X(OH) observed in uv absorption spectra of diffuse clouds, X(OH)≃1-2x10^–7 at N(OH)=5x10¹³cm^–2 and E(B-V)=0.3mag, it follows that X(HCO⁺)≃3-6x10^–9 across a very broad range of extinction. Apparently, many diatomics and polyatomics form readily in diffuse clouds as long as there is any appreciable amount of H₂ formation, and well before the carbon conversion from C⁺ to CO is complete or CO emission is strong. Although the strong coupling between OH and HCO⁺ is characteristic of conventional diffuse cloud chemistry, such chemistry in general falls far short of reproducing the observed amounts of HCO⁺ and perhaps CO, even when OH can be explained. Although not understood at present, it is clearly the case that many molecules, including those formed on grains (i.e. H₂CO), appear with abundances characteristic of dark, cold, cloud cores even when A_V≲1mag.