Hot cores are dense, small, warm clumps of gas found close to young massive stars (see, e.g., reviews by Millar 1993; Walmsley & Schilke 1993 and more recently Kurtz et al. 2000). They contain abundances of molecules that appear somewhat anomalous when compared with those in the much larger and cooler interstellar molecular clouds, being richer, for example, in saturated hydrocarbons. Hot core chemistry arises generally as a consequence of high mass star formation, although it has also been observed towards the low-mass protostar IRAS 16293 (Ceccarelli et al. 2000). In the case of massive star formation, the newly formed star heats dust in its vicinity, and the warmed dust then evaporates the icy mantles that were deposited in the pre-stellar phase. Thus, the molecules released into the gas trace directly the physical conditions during the cloud collapse that led to star formation.
There have been many studies of the chemical evolution of the molecules from the evaporated ices (e.g. Brown et al. 1988; Caselli et al. 1993; Charnley 1997; Millar et al. 1997; Cesaroni et al. 1998; Hatchell et al. 1998a; Millar & Hatchell 1998; Viti & Williams 1999). These have shown that observed hot cores must have ages limited to 104 to 105 years. Therefore, the study of hot cores gives insight into events close in time as well as space to the formation of the star. For most models it is assumed that ice is evaporated from the dust instantaneously, immediately after the star has formed. However, Viti & Williams (1999) noted that the star will not attain its maximum luminosity instantaneously, but over a finite period that may be comparable with the normal age of a hot core. The dust would in this case be warmed up over a comparable period, so that initially only weakly bound molecules such as CO and N2 would be released from the ice while more strongly bound molecules (e.g. H2O) would be released later. In this view, the chemical composition of material entering the gas phase changes with time. Therefore, the contraction time of high mass stars to the main sequence (a poorly known quantity; Hanson 1998) might possibly be inferred from the time dependent chemical composition of hot cores.
However, this is not the only possible generalization of the commonly
accepted picture of hot core evolution. In addition to their
luminosities, young stars generate winds that may strongly influence
the stellar environments. It is well known that low-mass pre-zero age
main sequence (ZAMS) stars have strong winds (Bachiller 1996). Given
the present lack of knowledge about high mass stars at a similar stage
of evolution, we assume that they also may have winds with mechanical
luminosities of the order of 10 
(Shepherd & Churchwell
1996). This view is supported by observations that show the existence
of outflows around stars without UCHII regions (Reid et al. 1995;
Shepherd & Churchwell 1996; Molinari et al. 1998; Cesaroni et al. 1999). We therefore consider the possible consequences of winds
from the star impacting on the material that will become the hot core,
and address the following question: if wind-generated shocks desorb
molecules from ices, are there any observational tracers of this
effect? We shall assume that the shock may arrive on the core gas
before, during or after the thermal evaporation process driven by the
stellar radiation.
Although the stellar wind is likely to have a very high velocity (much larger than 100 km s-1) as it leaves the star, the perturbation it creates in the dense hot core travels with a much smaller velocity. In this paper, we consider the effects of relatively slow C-type shocks, with velocities around 10 to 20 km s-1. Shocks of much higher speed would be easily traceable through SiO emission (Lada et al. 1978) from sputtered grains. Shocks of much smaller speeds than 10 km s-1 are unlikely to make a significant contribution to the chemistry in hot cores as the postshock temperatures would be too low to affect the chemistry significantly. We are making the first detailed study of the possible effects of shocks that may leave subtle evidence of their existence within hot cores. If this initial study indicates that shock tracers can be identified, then future studies will have to develop a detailed treatment of the shock passage. We consider only the effect of the passage of a single shock. It is possible to envisage that the wind may vary in such a way as to generate a succession of shocks. However, a steady wind acting on a smooth core will send only a single shock through it, and this is the system that we explore in this initial calculation. Multiple shocks are beyond the scope of the present calculation.
In Sect. 2 we describe the details of the models evaluated, and in Sects. 3, 4, and 5 we give results for several different classes of model. Section 6 contains our conclusions from this first detailed study of this complex problem. These conclusions are quite straightforward, and are given briefly here. Firstly, if all species more massive than helium are completely frozen-out so that none of them remains in the gas phase and hydrogenation occurs on surfaces, then we can find no chemical signature of the passage of a shock through the hot core, regardless of when the shock arrives in the core's evolution. Secondly, if - as observations suggest - there is always a small residual amount of matter containing elements more massive than helium in the gas phase, then the following ratios of molecular abundances are specific indicators of shock passage: NS/CS; SO/CS; and HCO/H2CO. Thirdly, these ratios are useful indicators both in cases where the warming of the dust by stellar radiation takes place instantaneously and where it occurs over a finite period.
These clear conclusions highlight, however, the present paucity of suitable high resolution observations in lines of molecules that are appropriate for testing the shock hypothesis. The over-riding conclusion is, therefore, that further specific observations still need to be made if we are to assess whether dynamical as well as radiative effects are likely to be important within hot cores.
Copyright ESO 2001