1 Introduction

The circumstellar environment of young stars is difficult to study, as in its early phases a young accreting and evolving star can be surrounded by sufficient gas and dust to cause a large extinction. However, at visible and near-IR wavelengths this continuum extinction is smooth and contains little information on the nature of the surrounding gas and dust. The re-radiated energy is emitted mainly at wavelengths longer than 2 $\mu$m. Molecular vibrational-rotational gas-phase and solid-state transitions occur in the mid-IR, and a number of different molecules have been detected towards YSOs (see the review by van Dishoeck & Hogerheijde 1999). Spectroscopy of YSOs has also confirmed the presence of a number of molecular ices, such as ices of water, CO₂ and CH₃OH (e.g. Gibb et al. 2000; Boogert et al. 2000). In addition we also see both absorption and emission from refractory solids such as silicates and PAHs, and infrared continuum emission. This continuum emission results from the reprocessing of stellar radiation by the dense dust envelopes of the YSO, and this infrared excess is often used as a criterion for the identification of YSOs in photometric surveys.

Unfortunately, studies from the ground are constrained to work within atmospheric windows. In some cases, such as CO₂, telluric absorption by the molecule at 15.2 $\mu$m makes its study in astronomical objects impossible. Furthermore, the sensitivity of observations in the mid-IR is greatly reduced because telescope and atmospheric thermal emission greatly increase the photon shot noise. Consequently, ground-based studies longward of K band are generally restricted to observing relatively bright objects. The only solution to these problems is to move to an airborne or space platform, where even a modest sized telescope can outperform ground-based systems.

Mid-IR spectra of young stars at high resolution were successfully taken with the Infrared Space Observatory (ISO, see Kessler et al. 1996 for an overview of the ISO mission) by various investigators (e.g. W33A by Gibb et al. 2000; Elias 29 by Boogert et al. 2000). In an attempt to push mid-IR observations to even higher levels of sensitivity we made use of ISOCAM (Cesarsky et al. 1996) and the Circular Variable Filters (CVF) to image 42 YSO candidates in four well-known low-mass star formation regions: RCrA, $\rho$ Ophiuchi, Serpens and Chamaeleon I. The high sensitivity and good sample size was achieved by using the low CVF resolution and by observing dense clusters of YSOs, so that ISOCAM's imaging capability allowed simultaneous spectra to be obtained. It should be noted that the low spectral resolving power of the CVF ($\lambda$/ $\Delta\lambda\simeq40$) essentially prohibits the detection of gas-phase spectral features; this is a study of the molecular ices and dust that lie around these young stars. However it is possible to draw a number of conclusions about the physics and chemistry around these YSOs, with information about their composition, thermal behaviour and evolution readily obtainable from such spectroscopic observations.