(221c) Simulations of Reactive Transport with the Openfoam Toolkit

In this poster I will present a sample of results from recent numerical simulations of reactive transport in various pore structures. The poster will focus on the numerical methods used in the calculations, and in particular on the customized libraries that manage the evolution of the unstructured mesh surrounding the dissolving surfaces.

Results will be presented showing that large mesh motions are possible without any topological changes to the mesh, if the relaxation of the mesh points is well controlled. At present, topological changes must be carried out externally, using the standard OpenFOAM meshing code snappyHexMesh. A mesh of the evolving surfaces is written to disk and used as an input to the meshing algorithm. Dynamic remeshing with topological changes is possible, but the capabilities are limited in comparison with snappyHexMesh.

We present some benchmark comparisons which show the advantages of a body-fitted mesh. Results for a dissolving cylinder were found to be much more rapidly convergent than calculations on a structured mesh using (for example) embedded boundary conditions.

The code is robust, failing in only two quite predictable ways. First when the skewness of the mesh becomes too high (in excess of 75 degrees); this can be solved by remeshing. Second, as the cusp at the tail becomes sharper it introduces an instability in the mesh relaxation. Further work is needed to resolve this issue but the problem can be reduced in severity by smoothing the surface mesh using a convex hull algorithm to delete interior points.

This work was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Chemical Sciences, Geosciences, and Biosciences Division under Award Numbers DE-FG02-98ER14853 and DE-SC0018676