A novel high-velocity DT pellet accelerator using MW gyrotron power is being proposed for refueling and active density control of large ITER-class reactors. Bench scale experiments are being planned at GA to test the concept using non-cryogenic solid propellant materials. However, materials such as naphthalene (C10H8) have high critical temperatures (T_crit = 748 K) and during the early hydrodynamic expansion of the propellant up to T_crit, the material is either a pure liquid or a two-phase liquid-vapor mixture. Whenever the saturated vapor pressure outruns the driving pressure, P_sat > P, the liquid can boil explosively into its vapor phase as additional thermal energy is applied by the gyrotron. Otherwise for P_sat < P, the pusher remains in the pure liquid phase up to T_crit. A theoretical formulation developed for both cases indicates that the pure liquid phase occurs when the gyrotron peak power exceeds 800 kW and projectile masses are above 0.17g, while two-phase flow occurs for lighter projectiles and lower powers. A newly designed waveguide/pellet launch tube with an RF window can withstand the peaked pressure of 16 MPa predicted after the first ~ 0.36 ms where the pellet reaches 175 m/s.

GA hosted a GYRO developers workshop during the week of January 15-19. Researchers from PPPL, LLNL, UCLA, UCSD, Univ. of Texas, as well as GA physicists, attended. The workshop focused on improvements to GYRO documentation and continued development of GYRO data analysis tools (synthetic diagnostics). Various presentations summarizing ongoing GYRO simulations of experiments were also given.

A new INCITE 2007 award on the Cray XT3 at ORNL has been given in support of a new 2007 SciDAC-SAP project (steady-state gyrokinetic transport simulations with GYRO). In addition, the DOE INCITE 2006 computer time award on the Cray X1E at ORNL for massively-parallel coupled ITG/TEM-ETG simulations using the GYRO code was renewed for a second year.

Joint GA-PPPL Highlight:

In collaboration with PPPL and LANL, the normal mode approach (NMA) code for studying feedback stabilization of the resistive wall mode in low rotation plasmas has been successfully extended to treat arbitrary up-down asymmetric tokamak equilibria. Results from application to cases simulating experiments show that the effect of feedback is expected to produce modifications to mode structures both at the plasma surface and on the resistive wall.

Recent studies of the effects of plasma elongation κ on ITG/TEM turbulence in GYRO simulations show that most of the elongation scaling for the transport originates from the gradient in κ, with a noticeably smaller contribution from κ directly. In previous studies using the Miller equilibrium model in which the κ gradient varies with κ, scans around the GA standard (STD) case showed the transport scales like 1/κ at fixed minor radius for the normalized GYRO diffusivities. In the recent studies, elongation scans over a broader range of parameters show that when the gradient in κ is fixed, the resulting scaling in κ is weak. While the κ-scaling for the ion energy transport is fairly robust, the scaling for the electron energy transport is stronger close to threshold. We also find that the κ-scaling for the electron energy transport increases (the exponent becomes more negative) as the safety factor decreases. Since the zonal flow amplitude depends on the safety factor, this suggests that the zonal flows might be contributing to the electron transport scaling with κ; for example, it is known that there is a nonlinear upshift in the critical gradient that is due to changes in the zonal flow behavior as κ is increased. This hypothesis is undergoing further investigation.