1997A&A...326..963B

We present detailed observations of the 1.2mm continuum and the CO line emission in the exceptional starburst/active galaxy NGC 3079 with the IRAM 30-m telescope. The 1.2mm thermal dust emission is much less centrally concentrated than the CO or radio continuum. Roughly 10% of the total 1.2mm flux comes from the central 11" as opposed to ∼40% for the CO or radio continuum. We find no evidence for cool dust in the center of NGC 3079. A dust temperature of 32K fits the data well. We then determine N(H₂)/I_CO(2–1)≃3x10¹⁹cm^–2.(K.km/s)^–1, roughly an order of magnitude below most estimates. In the disk, cool dust is present. For T_dust=17K, the cross-section (or absorption coefficient) we derive for dust in the atomic medium is very close to the value of Draine & Lee (1984ApJ...285...89D). The combination of the errors in the HI column density and the 1.2mm thermal dust emission are probably less than a factor 2 in these and other recent observations. Unless the dust temperature has been severely overestimated, the dependence of the dust opacity on wavelength (λ) is closer to λ^–2 than λ^–1.5 between 100 µm and 1.2mm. Based on this cross-section, we estimate an average conversion factor of N(H₂)/I_CO(2–1)≃1-2x10²⁰ cm^–2.(K.km/s)^–1 in the disk. This is within the range of estimates for our galactic disk, although lower than early estimates. Given the very low N(H₂)/ I_CO(2–1) value found for the nuclear region, we compare it with ratios derived from radiative transfer equations and ¹³CO measurements. These estimates are in good agreement with our dust-based N(H₂)/ I_CO factor. It appears unlikely that grain emission is enhanced (with respect to that in the atomic medium) through grain-grain agglomeration or through condensation of molecules onto grains. The dust mantles may be partially evaporated in the center of NGC 3079. The dust temperature may allow evaporation of CO-rich mantles and C-shocks are another plausible mechanism capable of disrupting mantles while leaving CO molecules intact. The effect would be to raise the N(H₂)/I_CO(2–1) ratio slightly and to increase the gas-phase CO abundance.