No abstract

No abstract

No abstract

We present experimental evidence that mercury forms an insulating phase in films condensed on glass substrates at liquid-helium temperatures. This insulating phase is metastable and it exists in films only 0 up to some critical thickness d, =60 A. Our results cannot be explained by the assumption that the films consist of isolated islands. We believe that similar insulating phases exist also in quench-condensed lead films and maybe in some other metals. As far as we know the possibility that a disordered system of metallic atoms may form an insulator has never been considered.A standard way to produce disordered metallic films is vapor deposition at cryogenic temperatures.In most quench-condensed films the onset of the electric conductivity corresponds to a very low film thickness of d -10 A. It is a direct consequence of the low atom mobility. However, there exist a few metals which have no finite conductivity if the film is thinner than 60 -80 A. The first observation of this unusual behavior was reported for mercury films more than fifty years ago. ' These films were deposited on glass substrates at T=20 K (the same results were obtained later at liquid-helium temperatures. ' A similar value of d, has also been observed for the conductivity onset in lead films. ' The usual explanation of the high d, values is that these films consist of metallic grains with vacuum gaps between them. An increase in the film thickness above d, closes the gaps between the grains and makes the film electrically continuous. This model was proposed, in one of the first papers on this topic, and it seems to be commonly accepted now. We want to propose here a different explanation of this phenomenon. We suggest that these metals form insulating amorphous phases when condensed at low temperatures. The most probable explanation is that the low film density is responsible for the insulating behavior. These insulating phases are metastable and an increase in the film thickness causes a transition to a more dense conducting state. The mechanism of the transition should be similar to that observed in some quench-condensed amorphous metals when the film growth leads to a transition to crystalline phases. This transition can be jumpwise ' or continuous ' depending on the film material (see also Ref. 11 and references therein). Some experimental results presented in this paper support our idea. Films were prepared and studied in a stainless-steel evaporation chamber, which was completely immersed in liquid helium. A substrate and a quartz thickness monitor were placed in the upper part onto a block with a small helium container separated from the main bath.The glass substrate was indium sealed to the opening of the container. The bottom surface of the substrate faced the vacuum of the chamber, while the top one was in contact with helium in the container. We were able to heat the substrate up to the room temperature while the body of the evaporator remained in liquid helium. This baking procedure was used to clean the substrate, taking adva...

No abstract