One process which provides significant cooling involves collisions between molecules. When two molecules collide, they convert some of their thermal (kinetic) energy into a form of potential energy. The energy can be stored in the molecule either by simple rotation or by internal vibration or even by lifting one or more electrons into a "higher" less bound orbit around the atoms in the molecule. This energy can be later released by the emissions of a photon of a particular energy that is characteristic of these molecular species. Photons that escape the cloud carry this energy with them, thus helping to cool the cloud. Atoms and molecules are considered to be good coolants if (1) they readily emit photons following a collision and (2) they are present in large enough quantities that a significant number of photons are emitted. In this way the collapse of an interstellar cloud is tied to the chemical composition of that cloud.

Hydrogen and helium are, by far, the most abundant elements in interstellar clouds. However, these elements are very poor coolants because they cannot be collisionally induced to emit photons at the low gas temperatures characteristic of molecular clouds. Two decades of theoretical studies have consistently predicted that a large fraction of the total cooling is borne by a few other atoms and molecules, notably gaseous water (HO), carbon monoxide (CO), molecular oxygen (O), and atomic carbon (C).

Recently two members of the SWAS team, David Neufeld and Gary Melnick, along with Stephen Lepp, re-investigated the cooling in the interstellar medium. Figure 1 shows an example of their results presenting cooling curves (cooling as a function of the density of the most abundant molecule, molecular hydrogen H) computed for the cold environment expected in interstellar clouds prior to the formation of a star.

This figure demonstrates that at lower densities and temperatures CO and O are the dominant coolants, but at high densities HO along with a host of other molecules, under the heading ``other molecules'', add up their contributions to dominate the cooling. However, this situation changes dramatically when the temperature is raised.

Figure 2 demonstrates these changes in a plot showing the density and temperature regimes where different molecules dominate the cooling. As in Figure 1, CO is the dominant coolant at low densities and temperatures, but at higher densities and temperatures water becomes the principle coolant. These conditions can be expected to occur when interstellar clouds collapse. Thus the theory predicts that during the process in which every star has formed, the collapsing cloud should pass through a phase in which water is the dominant coolant.

The amount of cooling scales with the number of molecules present, or the abundance. Water is believed to be a good coolant at high temperatures both because models of the chemistry predict that its abundance will undergo a dramatic increase at higher temperatures and a large number of transitions are capable of being excited and are therefore available to cool the gas (Figure 3).

However, we currently have little or no information on the abundance of either water or molecular oxygen because the photons of these important species are blocked by the Earth's atmosphere. In contrast, the CO molecule, which can be easily observed from ground based observatories, has been confirmed as an important coolant of the interstellar medium.

One of the primary goals of the SWAS mission is to establish the means by which these clouds cool as they collapse to form stars and planets. Using SWAS, astronomers will observe the emission from HO, HO, O, CO, and C. All of these species are important coolants in the varied phases of the interstellar medium and therefore are central to the evolution of molecular clouds and the process of star formation.