Diffusive transport (no flow) with BC in PHAST

[SlippinJimmy]

Hi,

I have unsuccessfully been trying to model a simple diffusion problem in PHAST with either a Dirichlet (constant concentration), or flux BC (solute flux) without flow.

Although I have been able to set up diffusion only problems by defining a variable initial condition (IC) across a model domain, it does not seem to work with BC only. I have tried using both the FLUX_BC & SPECIFIED_HEAD type BC options (by setting flux to zero or constant heads across the domain) to no avail. I can easily get a solution if I define hydraulic flows, although I really want to model diffusion only in this case.

Has anyone tried doing this in PHAST? Is it a fundamental program limitation?

I would be rather grateful for any pointers!

[dlparkhurst]

Here is a modified version of your example. Sorry, it is not closer to your original. If you want diffusion only, you should use SPECIFIED_HEAD boundary with a "fixed" solution. The fixed solution requires the boundary cells to maintain the specified solution composition.

If you use FLUX_BOUNDARY, LEAKY, or SPECIFIED_HEAD with an "associated" solution, there will be no change in the domain if there is no flux to bring the associated solution into the domain.

[SlippinJimmy]

Thank-you Dr Parkhurst,

That works perfectly!

/Jimmy