Preliminary modeling of the ablation of a polystyrene (PS) shell predicts that a 500 m/s pellet containing densely packed dust grains easily crosses the q = 2 surface in DIII-D before complete ablation. Encapsulation of dust grains by PS shell pellets is being considered for delivering massive density increases to the plasma for the purpose of mitigating disruptions. Surface ablation proceeds by pyrolysis (thermal decomposition) of PS, which favors the production of its styrene monomer, C8H8. The sacrificial shell undergoes ablation while in flight through the hot plasma while shielding the interior dust grains from the plasma heat flux until the shell fully ablates. Once that happens, the bare “dust ball” is expected to undergo fragmentation and entrapment inside a radiatively cooled plasma, thereby allowing massive plasma densification. Experiments injecting small, 2 mm hollow shell pellets are being planned on DIII-D in order to compare the shell burnout distance with model predictions. With increasing ablation temperature downstream the monomer gas decomposes into smaller molecules, free radicals, and ions. This “cracking” process will be included in future gas dynamic modeling as it represents a significant heat sink in the ablation flow field, and tends to lower the pellet mass loss rate.

In collaboration between N. Gorelenkov and G. Kramer of PPPL and Alan Turnbull of GA, the public NOVA-K code package has been successfully ported to the GA workstation cluster. NOVA-K is an ideal MHD code with kinetic extensions and the package includes a q solver equilibrium code and a mapping code, as well as the CONT code. The public version is used extensively for calculations of ideally stable Alfven eigenmodes (AE modes) but can also be used to study kinetic effects on the unstable internal kink. The ported code was successfully tested for both the stable AE mode case and for an internal kink mode. Work is in progress to interface the stability code with the GA equilibrium codes EFIT and TOQ. This will enable more direct calculations of the AE mode spectrum of DIII-D discharges.

As part of our tightening of computer security, the internal DIII-D web site has been moved from <fusion.gat.com> to a new machine <diii-d.gat.com>. The significant growth in the geographical diversity of the DIII-D National Team has resulted in a concurrent growth in the usage of the DIII-D Web Site to communicate information. The recent tightening is intended to better protect this valuable resource from unauthorized access. In addition to the move to diii-d.gat.com, potential vulnerabilities have also been tested by external cyber security experts from CIAC (see http://www.ciac.org/ciac/index.html)

A 12 page article entitled “Supercomputing Boosts Fusion Research” featuring the GYRO code was just published in the Spring 2008 issue of the SciDAC Review magazine. The article describes the impact of gyrokinetic simulations on fusion research leading to ITER and can be downloaded in PDF form directly from http://www.scidacreview.org/0801/pdf/fusion.pdf, or accessed from http://www.scidacreview.org/0801/index.html.

Emily Belli gave an invited Sherwood talk summarizing her work with Jeff Candy on the new NEO code. NEO provides fast neoclassical equilibrium solutions to the drift kinetic equation with very detailed multi-species collision models. All local and nonlocal (large orbit) multi-species neoclassical flows are provided in real geometry (e.g. energy, plasma and impurity flows, bootstrap currents, poloidal and toroidal rotation). The paper compares the highly accurate numerical solutions with many previous neoclassical analytic results. Nonlocal deviations are very small except within the “potato radius” near the origin. NEO provides a considerable advance on the well known NCLASS code. NEO was partially funded by the Edge Simulation Laboratory project and the code will be released publicly. The results will be published soon in a peer-reviewed journal.