Astron. Astrophys. 358, 1069-1076 (2000)

5. Diffuse cloud chemistry

Models of quiescent diffuse cloud gas-phase chemistry have their few possible successes - OH, CH, C₂, CN (see Table 2 and Federman et al. (1994)) - a couple of long-recognized failures (CH⁺, CO) and a host of new problems like C₂H, HCO⁺, H₂CO, and so on. The basic problem for the chemistry of trace species - how to get the ambient oxygen and carbon into molecular ions - is two-fold: atomic oxygen is only weakly ionized (by endothermic charge exchange with H⁺) and so does not participate in rapid ion-molecular hydrogen abstraction reactions (O OH⁺ + H), and the first ion of carbon unforgivably does not react rapidly with . At thermal speeds, radiative association is slow and at thermal temperatures the endothermic hydrogen abstraction reaction (C CH⁺ + H) cannot proceed.

The outstanding failure of conventional models of quiescent diffuse cloud chemistry to reproduce the observed amounts of CH⁺ (by far the largest discrepancy in Table 2) led originally to the idea of CH⁺ formation in interstellar hydrodynamic shocks (Crutcher, 1979; Elitzur & Watson, 1980), which was subsequently generalized to a magnetohydrodynamic shock (Draine & Katz, 1986) and serves as the basis for several other approaches.

The idea of models incorporating the interstellar magnetic field is to accelerate the dominant ionic species C⁺ and so drive the otherwise slow reactions of C⁺ and (among other things) without overproducing OH and O, as would occur if too much of the gas is heated (Falgarone et al., 1995; Federman et al., 1996). Bulk shock models fell from favor for nearly a decade, owing to several things; their prediction of velocity shifts, typically amounting to a few km s^-1 between molecular ions (CH⁺) and other species, which were seldom if ever observed; their seeming inability to produce N(CH⁺) in excess of per cloud; and the gradual emergence in the data of a correlation between N(CH⁺) and reddening (Gredel et al., 1993). But they have recently been rexamined and shown to produce, of all things, the observed correlation between OH and HCO⁺ (Flower & Pineau Des For^ets, 1998). Problems with predictions of unobserved kinematic differences between HCO⁺ and other species persist, however (Liszt & Lucas, 2000), in these models.

In an effort to solve the CH⁺ problem without the problematic aspects of large-scale interstellar shocks, several proposals have been made to drive the C⁺ + reaction in situ in diffuse gas (Falgarone et al., 1995; Hogerheijde et al., 1995; Federman et al., 1996; Joulain et al., 1998), principally by the dissipation of turbulent (magnetic) energy. If this can be done, relatively large amounts of CH and CH can be sustained, and HCO⁺ can form from O + CH as well as from the reactions of C⁺ + OH which dominate in quiescent cloud chemistries. It is claimed that the surprisingly high abundances we observe in diffuse gas can be explained in this way, although the similarity of dark cloud and diffuse gas relative abundance patterns seems a remarkable coincidence.

Most recently, Viti et al. (2000) have explored the possibility that C⁺ recombines on grains to form CH₄, which is then released into the gas phase and where its photodissociation products lead to enhanced abundances of simple hydrocarbons like C₂H, and to H₂CO, etc . Unfortunately, their results still do not produce sufficiently high abundances to explain our observations if the visual extinction is assumed to be as low as 1 mag.

The systematics of such turbulent chemistries remain to be explored but in the meantime we will produce a series of papers, beginning here, in which the systematics of several chemical groupings are exposed and compared. The next paper in this series will deal with cyanogen-bearing species such as HCN, HNC and CN. This will be followed by works discussing sulphur chemistry (CS, SO, SO₂, H₂S and HCS⁺), and NH₃ and CO.

© European Southern Observatory (ESO) 2000

Online publication: June 20, 2000
helpdesk.link@springer.de