Abstract :
Summary form only given, as follows.It is, perhaps, not widely appreciated that the classical theory of cross-magnetic-field collisional transport breaks down when hD> r, for one or more species. Some examples where this occurs are: the edge of a tokamak plasma; plasmas around neutron stars; and nonneutral plasmas. This talk will review new theories and experiments carried out at UCSD that predict and measure collisional transport in this regime. When h~ > r, , collisional transport is dominated by long-range collisions with impact parameters on the order of hD or greater. Such collisions are completed neglected in the classical theory. In this regime the cross-field thermal conductivity is independent of the magnetic field, and is up to several hundred times greater than the classical prediction for the parameters of experiments performed here at UCSD. [IT]e st particle diffusion is also greatly enhanced over the classical theory, particularly so when motion along the magnetic field is rapid so that it can be averaged out, and the charges behave as rods of charge that ExB drift in the fields of the other charges. In this regime seminal work by Taylor and McNamara and Dawson and Okuda [2,3] showed that the diffusion scales as I/B, and is dominated by large-scale ??Dawson-Okuda vortices??. We have investigated the effect that small ExB velocity shears have on these vortices: the applied shear tends to rip the vortices apart, reducing the diffusion. This is analogous to the effect of velocity shear observed in turbulent tokamak plasmas, but here the transport can be calculated rigorously because fluctuations are collisional in origin, not turbulent. Preliminary experiments show factor of two agreement with the new theory for shear reduction of collisional diffusion.
