We have obtained low-resolution |(λ /Δ λ ~ 160)| 3-µm spectra of a selection of field stars lying behind the Taurus dark cloud. Several of these stars have abnormal extinction laws, which are best characterized by |AK≤ 0.27E(J - K)| and |AK≤ 5.4E(J - K)|. We find a very good correlation between AV and |τ3.05| (the H2O ice feature), confirming a previous result, and between AV and |τ3.45| (the long-wavelength wing), yielding the ratios |τ 3.05/AV = 0.059± 0.003| and |τ 3.45/AV = 0.0077± 0.0005|. The threshold extinctions for the H2O ice band and the long-wavelength wing are found to be the same, within the uncertainties, at |A_v₀| = 2.6. The fact that these thresholds are the same means that the long-wavelength wing cannot be a result of a highly refractory hydrocarbon residue. The lack of substructure in the long- wavelength wing also argues against a hydrocarbon origin for the wing. By interpreting the H2O ice feature with a model employing silicate and graphite grain cores that follow a power-law size distribution, we find that the ice band observations can only be fitted by models in which either (i) only the largest grains have significant ice mantles, or (ii) there are two grain populations throughout the cloud, one of ‘bare’ grains and one of grains that have a constant ice mantle thickness of 0.35±0.05 µm. In the latter model, we estimate the ratio of bare to mantled grains to be |~ 3 × 10⁴:1|. Both models, combined with the observed |τ3.05/AV| ratio, imply that 8–9 per cent of the oxygen is depleted on to the grains as H2O ice mantles. We discuss a number of mechanisms for the production of the H2O ice and long-wavelength wing thresholds, and conclude that photo-desorption is the most likely means of prevention of H2O ice mantle formation near the cloud edges. Photo-desorption, combined with a clumpy cloud structure, also naturally explains the bare/mantled grain distribution, as UV photons penetrating into the cloud prevent ice mantle formation everywhere except within dense clumps.
No takes yet. Share an insight, caveat, or question.
Smith et al. (1993) studied this question.