UNIFYING ROLE OF RADIAL ELECTRIC FIELD SHEAR IN THE CONFINEMENT TRENDS OF TRANSITIONLESS REGIMES IN TFTR

Turbulence suppression by radial electric field shear (E r) is shown to be important in the enhanced confinement of TFTR supershot plasmas. Simulations of supershot ion temperature profiles are performed using an existing parameterization of transport due to toroidal ion temperature gradient modes, extended to include suppression by Er shear. New spectroscopic measurements of Er differ significantly from prior neoclassical estimates. Supershot temperature profiles appear to be consistent with a criterion describing near- complete turbulence suppression by intrinsically generated E r shear. Helium spoiling and xenon puffing experiments are simulated to illustrate the role of E r shear in the confinement changes observed. The original model, developed prior to the availability of poloidal rotation measurements, employed a neoclassical calculation of the intrinsically generated flows driven by density and temperature gradients, using the model of Ref. (4 ) to parameterize toroidal ITG modes. New poloidal rotation measurements in TFTR (5 ), however, indicate a significant discrepancy with standard neoclassical results in both supershots and reverse shear plasmas. The E r shearing rate derived from spectroscopically measured quantities is typically a factor of two larger than that derived using neoclassical (6 ) poloidal flows, with a profile peaked more closely to the magnetic axis. When the toroidal rotation is large, the measured and neoclassical E r profiles are qualitatively similar, masking the difference. However, TFTR supershot cases with balanced tangential neutral beam injection reveal a significant discrepancy between neoclassical and measured poloidal velocities, resulting in very different Er profiles. Nevertheless, when the measured poloidal rotation, rather than the neoclassical calculation, is used, the simulated core ion temperatures remain consistent with the same overall turbulence suppression criterion. While the previous turbulence suppression criterion appears to be preserved using the new measurements, the threshold condition for the suppression to occur has to be adjusted to compensate for the increased shearing rate. This adjustment is within the accuracy of the numerical simulations which form the basis for the criterion (7 ). The dependence of this approach on