Extensions of the TEP Neutral Transport Methodology

Recent extensions of the Transmission and Escape Probability methodology and its implementation in the 2-D neutral transport code GTNEUT are presented. These extensions address the issues of anisotropy of the neutral distribution function at the interfaces and the non-uniformity of the first collision source in short mean free path regions. Comparisons with Monte Carlo for a number of model problems are discussed. The Transmission and Escape Probability (TEP) interface current balance method (1) has been developed and implemented into the 2-D neutral transport code GTNEUT to provide a fast and accurate calculation of neu- tral particle transport in the complex tokamak edge and divertor configuration. Tests of GTNEUT predictions against Monte Carlo calculations and experimental measurements in DIII-D have demonstrated (2-5) the accu- racy and computational efficiency of the TEP method for a wide range of conditions. However, calculations of detailed model problems (3,4) designed to test approximations in limiting cases have identified two main areas in which extensions in the original TEP methodology would be useful: 1) taking anisotropy into account in the calculation of first-flight transmission coefficients when the neutral mean free path (mfp) is much larger than the characteristic dimension of the computational region; and 2) taking into account that the escape of scattered or charge-exchanged neutrals is preferentially across the incident surface when the mfp is small compared to the characteristic dimension of the computational region. In this paper, we discuss recent extensions of the TEP methodology which address the above issues. The anisotropy of the neutral distribution function at the interfaces is taken into account by implementing Double P1 (DP1) and Double P2 (DP2) approximations. The preferential backscattering of scattered or charge-exchange neutrals across the incident surface is addressed by implement- ing an albedo-based condition to describe the fraction of the collided neutrals that are reflected back across the incident surface.