Numerical solution to Vlasov equation: the 1D code. E. Fijalkow.

PROGRAM SUMMARY
Title of program: vl1dper
Catalogue identifier: ADJQ
Ref. in CPC: 116(1999)329
Operating system: Unix
Number of lines in distributed program, including test data, etc: 597
Programming language used: Fortran
Computer: Cray C 90

Nature of physical problem:
The numerical solution of the Vlasov Poisson system is commonly used to follow the evolution of plasmas. For years the most used method has been particles simulation (PIC codes). Advantages of the method are its capability to run with a small number of particles, its accurate treatment of advection and the absence of a need for velocity space meshes. The major inconvenience of that method is its inefficiency to present the behaviour of high velocity particles in near Maxwellian plasmas. These kind of plasmas are those encountered in plasma acceleration problems - as laser, plasma acceleration, fast electrons beams and so on. The advantage of the present code is its ability to treat with the same accuracy all the points of the phase-space.

Method of solution
The code is based on time splitting and the flux balance method [1]. The initial Vlasov equation is split into two equations, one for the evolution of the space variable, the second for the velocity variable. Both equations are solved iteratively by use of the Flux Balance method. The Poisson equation is solved by a method of spline interpolation [2] both to solve the potential and for, from the potential, the electric field.

Typical running time
For the testrun the time is 4.98 sec. That means 0.153 mu sec/grid point/time step on a HP 9000, 180 MHz computer.

References

 [1] E. Fijalkow, A Numerical Solution to Vlasov Equation, Comp. Phys.   
     Commun. 116 (1999) 319.                                             
 [2] G. Knorr, B. Joyce and A.J. Marcus, J. Comp. Phys. 38, 227-236      
     (1980).