Experimental investigation of the macroscopic flow of He II due to an oscillating grid in the zero temperature limit

A systematic experimental investigation of the macroscopic flow properties of extremely pure He II in the zero temperature limit is reported, covering the pressure range 0.3<P<24.8 bar . The flow is generated by electrostatically driven oscillations of a thin, tightly stretched, circular, square-mesh nickel grid. With growing amplitude of oscillation, the flow changes character at a first critical threshold from pure inviscid superflow past a submerged body of hydrodynamically enhanced mass, to a flow regime that is believed to involve a boundary layer composed of quantized vortex loops. Here the oscillatory motion of the grid acquires strongly nonlinear features. These include double-valued (reentrant) resonance curves and a decrease in the resonant frequency with increasing drive amplitude, but without any appreciable increase in damping. On further increase of the drive level, a second critical threshold is attained: here, the resonant frequency reaches a stable value, the response amplitude almost stops growing, but the linewidth increases. Finally, the flow acquires the character of fully developed classical turbulence, characterized by a square-root dependence of flow velocity on the driving force. Additional flow features attributable to the presence of remanent vorticity are observed and discussed.

The nucleation of vorticity by ions in superfluid 4 He II. Theory of the effect of dissolved 3 He

The theory developed in Part I is extended to include a discussion of nucleation by negative ions in the presence of dissolved 3 He at a concentration such that at a low temperature the negative ion bubble is likely to have adsorbed on its surface either one or two 3 He atoms. It is argued that the adsorbed 3 He atom can change the nucleation rate for two reasons: the atom can modify the perturbation applied to the helium at the surface of the ion; and it can act as a source of energy. The second of these effects is explored in some detail. It is shown that the 3 He atom is probably less strongly bound to the ion than it would be to the core of a vortex line; furthermore the atom adsorbed onto the surface of the ion can exist in a number of excited states (Shikin states), which are thermally populated even at quite low temperatures. Therefore, when nucleation of a vortex takes place, the 3 He atom might move from the ion surface to the core of the vortex or simply from one Shikin state to another of lower energy; in either case there is a release of energy. The existence of this energy release means, first, that nucleation becomes energetically possible at a reduced ionic velocity and, secondly, that the energy barrier opposing nucleation is reduced in size. Therefore the critical velocity for vortex nucleation is reduced, and, for a given supercritical velocity, the rate of nucleation is increased. Addition of a second 3 He atom would have a similar effect. Numerical estimates suggest that the energy release may be sufficient to explain the recent experimental results of Nancolas, Bowley & McClintock ( Phil. Trans. R. Soc. Lond . A 313, 537 (1985)), without the need to assume a significantly modified perturbation. But further experiments are required to check the detailed predictions of the theory.