Stochastic modelling of fusion product transport and thermalization with nuclear elastic scattering

Monte Carlo methods are developed to model fusion product (fp) transport and thermalization with both Rutherford scattering and nuclear elastic scattering (NES) in high-temperature (T/sub i/, T/sub e/ greater than or equal to 50 keV), advanced-fuel (e.g. Cat-D,D-3He) plasmas. A discrete-event model is used to superimpose NES collisions on a Rutherford scattering model that contains the Spitzer coefficients of drag, velocity diffusion (VD), and pitch-angle scattering (PAS). The recoil ions from NES are thermalized using a generation-by-generation approach in which the knock-ons generated during fp slowing down become the source term of the next generation of superthermal ions. Variance-reduction techniques including drag enhancement, the exponential transform, source biasing, and angular scattering biasing are applied to increase the computational efficiency of the Monte Carlo simulations. Data for NES and Coulomb-nuclear interference (NI) cross sections is taken from the ENDL library at LLNL and bench-marked against NI data generated by R-Matrix methods at LANL