Astron. Astrophys. 347, L19-L22 (1999)

4. Discussion

In view of the wealth of information in the spectra, a detailed analysis will be presented in a later paper. The following discussion is aimed only at establishing the chemical nature of the material producing the µm band.

Of the molecules expected to occur or that have been identified in grain mantles, only two have absorption features at µm: HDO and SO₂. For solid SO₂, the feature is a weak combination band, but the fundamental asymmetric stretching mode of the molecule occurs at µm (Sandford & Allamandola 1993; Boogert et al. 1997). Boogert et al. (1997) looked for the signature of solid SO₂ at 7.55 µm toward W33A and NGC7538 IRS9. Their results show that solid SO₂ appears to be present toward W33A with a column density of cm^-2, but only an upper limit of cm^-2 could be placed toward NGC7538 IRS9. Using optical depth spectra a column density, N (molecules/cm^-2), of the absorber can be estimated using , where is the optical depth at wavenumber (cm^-1), and A (cm/molecule) is the integrated band absorbance (band strength) (e.g. Allamandola et al. 1992). For the SO₂ combination band at µm @ cm/molecule (Sandford & Allamandola 1993), which for our spectra leads to a column density of solid SO₂ of cm^-2 for W33A, and cm^-2 for NGC7538 IRS9. These values are one and two orders of magnitude larger than those estimated by Boogert et al.. Solid SO₂ can therefore not be responsible for the absorption feature at µm observed toward these objects.

If the 4.1 µm absorption features observed towards W33A and NGC7538 IRS9 are due to solid HDO in the grain mantles, then it is possible to estimate the ratio of solid HDO to solid H₂O column densities, , in those lines-of-sight. For the 4.1 µm HDO stretch mode, cm/molecule (Ikawa & Maeda 1968), hence using the optical depth spectra in Fig. 1, we obtain: cm^-2 for W33A, and cm^-2 for NGC7538 IRS9. The solid H₂O column densities for these two objects are rather uncertain. The 3.08 µm water-ice band is saturated in the case of W33A (), and has a peak optical depth for NGC7538 IRS9 (Willner et al. 1982). Taking cm/molecule for the 3.08 µm water-ice band (d'Hendecourt & Allamandola 1986), and a typical full-width at half-maximum of cm^-1 (Duley & Smith 1995; Teixeira & Emerson 1999), an estimate of the water-ice column density can be obtained using an approximation of the expression in the previous paragraph: . This results in cm^-2 for W33A, and cm^-2 for NGC7538 IRS9. The water-ice column density can also be obtained from the 6.0 µm OH bend feature (Tielens & Allamandola 1987; Schutte et al. 1996b). The estimates from the 6.0 µm band are cm^-2 for W33A (Allamandola et al. 1992), and cm^-2 for NGC7538 IRS9 (Schutte et al. 1996b). We conclude that the solid HDO-to-H₂O ratio is in the range of to for W33A, and for NGC7538 IRS9.

The apparent lack of detection of a feature at µm toward AFGL2136 does not preclude an enhancement in the deuteration in the mantles along that line-of-sight. Integration of the optical depth spectrum in Fig. 1 in the same wavelength range as for the other two objects, results in cm^-2. Because of the uncertainties in the way the spectrum was assembled, and in the determination of the baseline (Sec. 3), this must be regarded as an upper limit. The estimated water-ice column density for AFGL2136 is cm^-2 (Schutte et al. 1996a). This leads to a ratio , which is within the range of values estimated for W33A.

The ratios show an enhancement in the deuteration of water in the grain mantles towards at least 2 of the 3 observed objects, of a factor of to relative to the cosmic [D]/[H] ratio. These results agree with the lower predictions of the chemical models of dense clouds which include deuterium chemistry and gas-grain interaction (cf. Sec. 1). Moreover, the solid HDO-to-H₂O ratios towards W33A and AFGL2136 are of the same order as the gas phase HDO-to-H₂O ratios observed towards "hot cores" (; Jacq et al. 1990; Helmich et al. 1996), and only up to a factor of 60 higher towards NGC7538 IRS9. Our HDO/H2O ratios are somewhat higher than the found in outgassed material (presumably from cometary ices) in Comet C/1996 B2 (Hyakutake) (Bockelée-Morvan et al. 1998), in the sense expected if cometary material has a solar nebula as well as an interstellar component. It would be also interesting to investigate if the HDO/H2O ratio correlates with grain (or gas) temperature, but as we cannot estimate the temperature of the HDO containing grains, nor locate them in any particular gas region along the line-of-sight, this is not possible at present. In a future work we will report on the correlations of the amount of HDO with other observed and derived properties of these lines-of-sight, with a view to elucidating the conditions which are favourable for large abundances of HDO ice.

Our results provide support for the assumption that the origin of high levels of deuterium fractionation in "hot cores" is evaporation of the deuterated species from the grain mantles. These observations provide the first evidence for the presence in the grain mantles of the link that has been missing between observations and models of "hot core" and dark cloud chemistry: deuteration in grain mantles.

© European Southern Observatory (ESO) 1999

Online publication: June 30, 1999
helpdesk.link@springer.de