Summary:
This is a proposal to the Department of Energy to construct and operate a modest experiment to test the hypothesis that a flowing liquid metal shell or first wall can stabilize magnetohydrodynamic instabilities in a magnetic fusion energy reactor. The proposal has two objectives:
(1) First, it will test the hypothesis that a two conducting shell system, with one shell rotating poloidally with respect to the other stationary shell can stabilize the resistive wall mode.
(2) Second, it will evaluate to what extent a flowing liquid metal wall behaves like a moving solid conducting wall by replacing a moving solid copper conducting shell used in (1) with a flowing liquid sodium wall.

A crucial issue for the attractiveness of a number of toroidal confinement concepts is the ability to stabilizelong wavelength external kink modes. The tokamak, ST and RFP configurations rely on the existence of a perfectly conducting wall at the plasma boundary to stabilize external MHD modes. Stabilization of the beta-driven kink in advanced tokamaks allows one to exceed the obtainable beta limits by a substantial fraction over the beta limit of the no-wall configuration. This leverage is amplified in ST configurations where the presence of a perfectly conducting wall is predicted to increase the beta limit by a factor of 2. In an RFP, a number of external modes are predicted to be unstable which makes the presence of a conducting wall a necessity for stable operation.

While perfect conducting walls can provide robust stabilization, a finite amount of conductivity in the wall causes the stabilizing wall eddy currents to decay away. On the resistive time scale of the wall flux can leak through the wall and provide a mechanism by which the plasma can access the free-energy available to the external kink. This mode, the resistive wall mode (RWM) grows on the wall time scale. In the absence of plasma flows, if the plasma is unstable without a wall, there will always be either RWM or an external kink; the plasma is always unstable. Theory suggests that if two conducting walls are used and if one of the walls is moving with respect to the other wall, the RWM can be stabilized.

Recently, there has been a resurgence of interest in using a liquid lithium as a first wall material in magnetic fusion devices. A flowing liquid lithium first wall has the potential of not only removing particles and heat from the reactor, but also of stabilizing the RWM if used in conjunction with a second conducting wall. Both of these conjectures need testing. In the case of heat removal, it is not clear that the thermal diffusivity of liquid lithium is sufficiently high to remove power at achievable flow velocities without excessive surface heating. In the case of stabilization of MHD modes, it is not clear that the turbulent flow is equivalent to a moving solid conductor since eddies in the flow can potentially move magnetic flux and result in an anomalously high resistivity.

We propose to investigate the stabilization of the resistive wall mode using a flowing liquid metal wall in a linear screw pinch geometry. A linear geometry (as opposed to a toroidal geometry), makes the experiment simpler and cheaper to build, and makes it possible to begin with a moving conducting wall constructed out of solid conductor, rather than liquid sodium. The experiment will consist of a 10 cm radius, 100 cm length plasma column in a 1 kG solenoidal field. Up to 19 kA of plasma current will be imposed by utilizing an array of 19 plasma guns developed for helicity injection on MST. During the first phase of the experiment, the cylindrical plasma column will be surrounded by two close fitting, 1 to 2 mm thick, cylindrical copper shells (the outer one which can be spun in the poloidal direction at approximately 100 Hz (60 m/s), driven by motor).

During the second phase, the outer conductor will be replaced by a flowing liquid sodium wall (1 to 2 cm thick with flows of 20 m/s). The liquid wall may act different from the solid wall due to the presence of flow driven instabilities and subsequent turbulence in the wall. These instabilities may produce anomalous transport of current in the layer which can ultimately affect the rotational stabilization process. Demonstration of stabilization and comparison between the flowing liquid metal and the moving solid wall will be the two principal scientific outcomes of this work.