For most
of this century, liquid helium was believed to completely wet all surfaces.
Because helium interacts so weakly with itself, it was thought that it
would interact more strongly with any surface, and therefore, that it would
wet any surface. Then in 1991 Cheng, Cole, Saam, and Treiner [3]
predicted that because the helium-cesium interaction was even weaker than
the helium-helium interaction, 4He would not wet cesium at low
temperatures. The weakness of the helium-cesium interaction is due chiefly
to the large size of the cesium atoms. The substrate potentials are shown
schematically in Figure 2 for both a weak-binding cesium substrate and
a typical strong-binding substrate. The potentials are composed of an attractive
part (typically van der Waals) and a repulsive part. Although the attractive
parts of both potentials are equal, the hard core repulsive part of the
cesium potential extends much further out from the surface resulting in
a much more weakly attractive total potential.
The CCST prediction sparked a number of experimental studies of helium adsorption on cesium. The first experiments were performed by Nacher and Dupont-Roc [8] who measured the flow of heat in helium films on a cesium surface. Their measurements indicated that cesium was indeed not wet by 4He at low temperatures. Mukherjee et al. [9] measured volumetric isotherms of helium on cesium plated graphite. Ketola et al. [10] used third sound in the superfluid helium films to probe the wetting of a cesium surface. These experiments all indicated that something unusual was occurring in the helium/cesium system.
The most complete set of data for the 4He/cesium system was measured by Rutledge and Taborek [4]. Using a quartz microbalance they were able to directly measure the mass of the adsorbed liquid helium film and made the first measurements of a prewetting phase diagram, showing a first order wetting transition at Tw ~ 2.0 K, and a prewetting critical point at Tpwc ~ 2.5 K.
The helium/cesium system is an extremely useful tool for the study of wetting phenomena for a number of reasons. First, it is simple: the substrate is a simple alkali metal and the interactions within the helium and between the helium and the substrate can be easily calculated. This allows for a direct comparison between experiment and theory. Second, it is a very clean system: At liquid helium temperatures, there are no unwanted impurities in the liquid to blur or distort the details of the wetting behavior. Third, the surface tensions which effect the wetting behavior can be easily manipulated experimentally so that a large range of behavior can be studied. The last and most important reason, which is probably a consequence of the first three, is that the helium/cesium system works. Using the helium/cesium system Rutledge and Taborek achieved their results in one year. By comparison, it took more than a dozen years before similar results were obtained with more traditional systems [11].