A Finite Element, Adaptive Mesh, Particle-In-Cell Code

72498S03-I Existing computer codes for modeling charged particles in three-dimensional geometries use finite-difference-analysis techniques and fixed, orthogonal, structured elements. The computational size of the problems of interest is extremely large, thereby severely limiting the types of problems that can be solved and the information that can be gained. A new approach that maintains accuracy while reducing computational cost is needed. This project will develop a finite element, particle-in-cell code with adaptive meshing, which could reduce the number of elements by two to three orders of magnitude while maintaining or improving the simulation accuracy. In Phase I, a finite element, adaptive mesh program, developed to model charged particles in static fields, will be modified to allow the solution of Maxwell¿s equations in the time domain. Key elements of the new computer code will be generated and compared for accuracy and efficiency against existing fixed mesh codes. Commercial Applications and Other Benefits as described by awardee: The new code should dramatically lower the computational cost of modeling charged particles in accelerators, microwave sources, and plasma devices. It would allow development of components and systems that cannot be designed with existing codes, thereby providing improved performance and reduced cost.