I am trying to model Taylor-Aris dispersion in a pipe. The flow field is steady-state. I need to partition the initial concentration along the axis of the pipe so that there is a thin "disc" with uniform, dilute concentration somewhere in the pipe and zero outside that region. In more precise terms, I need to define a square wave along the z-coordinate while maintaining symmetry along r and theta. Has anyone done this?... or have an idea of how to start? It seems like it should be relatively simple but I've been working at it for a while

In case anyone is curious the trick is to define different cell types of the same material, then write a user subroutine to select a certain type and change the concentration.