Transfer of the Unsaturated Helium II Film
Proceedings of the Royal Society A
The critical transfer rate of the unsaturated helium n film has been measured on surfaces of glass and german silver by a heat conduction method. It is found that a considerable reduction of the transfer rate occurs when the vapour pressure over the film is decreased only slightly below the saturation value. At a given percentage of the saturation pressure, there exists a critical temperature above which film flow will not take place. This critical tem perature is shown to be sharply defined,
... d to decrease with decreasing percentage satura tion. At the full saturation pressure the transfer rate is markedly different on the various surfaces used, but as the vapour pressure over the film decreases the flow rate tends to become the same for all surfaces. The critical temperature for onset of superfluidity is also independent of the substrate. The temperature dependence of the transfer rate is different from that for the saturated film, but very similar to the variation found in the flow of liquid through channels of width less than that of the saturated film. I n t r o d u c t io n Following the extensive work on helium n films in equilibrium with the saturated vapour, Brown & Mendelssohn (1947) investigated film transport under unsaturated conditions. The results obtained were of a qualitative nature only, but showed th a t helium films in contact with the unsaturated gas phase could still be super fluid. Their method, however, involved difficulty in interpretation due to the uncertainty of the state of the helium in the superleaks used by them. Superleaks were also used by Long & Meyer (1950 in a more detailed investigation. These authors allowed helium to flow through a superleak from a pressure below the saturation value either into a vacuum (their method I), or to a pressure only slightly lower than the original (method II). The results obtained with the two methods were completely different. Whereas with method I, the onset of super fluidity was suppressed to lower temperatures with decreasing percentage satura tion, method I I yielded superfluidity at all temperatures in the helium II range even for films as thin as 1^ statistical layers. The authors gave reasons why the results obtained with method II should be accepted as the correct ones and also why the helium in the superleak itself should be in the form of a film. These discrepancies induced us to re-examine the whole question of super fluidity of unsaturated films, using a method which did not rely on superleaks.