5 Conclusions

ISO-CVF spectroscopy of four low-mass star-formation regions has been performed, and a total of 42 different sources were observed. The resulting low-resolution 5-16.5 $\mu$m spectra have been analysed, and a number of different conclusions drawn:

• The profile of the 9.7 $\mu$m silicate absorption feature varies somewhat from source to source, but usually shows a weak emissive shoulder at $\simeq$11.2 $\mu$m. Significantly broader profiles are observed in emission. A number of sources appear to show "composite'' profiles, with absorbing and emitting components superimposed.
• The presence of a long wavelength wing on the 15.2 $\mu$m CO₂ ice band in most of the sources suggests that a significant fraction of the CO₂ ice observed exists in a polar (H₂O-rich) phase.
• The strength of the unidentified 6.8 $\mu$m band is observed to correlate far more strongly with the 6.0 $\mu$m water ice band than with the 15.2 $\mu$m CO₂ ice band. This suggests, in a manner consistent with previous observations, that the carrier of this band is a strongly polar ice.
• Compared to the other 3 regions (which show similar spectral characteristics) the sources observed in Cha I are somewhat anomalous, showing comparatively little foreground absorption. However this is consistent with previous studies, which indicate that the observed Cha I sources represent an unusual cluster of YSOs which have formed at the edge of a dense cloud core.
• The $\rho$ Oph cloud appears to have a systematic under-abundance of water ice, relative to silicates, relative to both RCrA and Serpens.
• The CO₂:H₂O ice ratios towards the sources in $\rho$ Oph show considerably greater scatter than in the other regions, with values ranging from 0 to 0.4. This may be due to local conditions around the YSOs in $\rho$ Oph but more data is needed to constrain this problem further.

The most important limiting factor in this study is obviously the low spectral resolution of the CVF, which essentially prevents any detailed study of absorption profiles, particularly around the CO₂ ice feature. However the use of the CVF enabled the study of a large number of objects, many of which would not have been observable at higher spectral resolution due to the low S/N. Consequently we have obtained a unique data set, with a large sample size and broad wavelength coverage. This enabled simultaneous study of the dominant mid-IR spectral features along the line-of-sight to a large number of YSOs, in a manner not possible using other spectroscopic methods. Further detailed study of the ice absorption profiles will undoubtedly yield more information regarding the chemistry around these objects, and study of the silicate profiles may yield more information about the dust composition and the structure of the dust envelopes.

Acknowledgements

We thank Tom Greene for his work in planning the observations and for helpful comments. We thank Andy Longmore for useful comments about the manuscript and also thank Sylvain Bontemps for useful discussions. Part of this work formed part of the MPhys dissertation of RDA at the University of Edinburgh, and RDA also thanks the UKATC for funding a vacation studentship. Finally, we thank an anonymous referee for helpful advice which greatly improved the clarity of the paper.