In sheath simulations in moment_kinetics we have two wall boundaries at z = +-L/2. We prescribe a source in the volume of the domain to fuel the ions. It may be useful to prescribe a source which forces the solution towards a specific function in a small range of the plasma. For example
S_krook[F_imposed,F] = nu * ( F_imposed - F),
with F_imposed = a Maxwellian with a specified and fixed density, parallel flow and thermal speed, and nu a frequency that may (or should) depend on spatial coordinates. For example, we might use the source only near z = 0.
This is potentially desirable in simulations with the self-collision operator C[F,F], because it is plausible that we require that there is no source term proportional to the null space of C[F,F] for a steady state to be reached. F = F_imposed would be a solution for which C = S_krook = 0.
@johnomotani Does such a source term already exist in the moment kinetics code?
In sheath simulations in moment_kinetics we have two wall boundaries at z = +-L/2. We prescribe a source in the volume of the domain to fuel the ions. It may be useful to prescribe a source which forces the solution towards a specific function in a small range of the plasma. For example
S_krook[F_imposed,F] = nu * ( F_imposed - F),
with F_imposed = a Maxwellian with a specified and fixed density, parallel flow and thermal speed, and nu a frequency that may (or should) depend on spatial coordinates. For example, we might use the source only near z = 0.
This is potentially desirable in simulations with the self-collision operator C[F,F], because it is plausible that we require that there is no source term proportional to the null space of C[F,F] for a steady state to be reached. F = F_imposed would be a solution for which C = S_krook = 0.
@johnomotani Does such a source term already exist in the moment kinetics code?