Lattice Boltzmann study of dissolution in porous media: Comparison of VOP with VOF-curved boundary coupling

In dissolution processes, during geometry evolution, the two-phase interface is continuously changed. For tracking the position of the interface, the method of combining VOF with curved boundary schemes was presented in this research. Since the combination would be rather complex, in addition to the original approach, a simplified approximate approach was also introduced. Further, for simulating heterogeneous dissolution in lattice Boltzmann (LB) framework, the volume of pixel (VOP) method has been widely used by researchers. Since the area of the solid-fluid interface is not captured by this method, the study of its validity was also presented in this research. For this purpose, simulations of calcium carbonate dissolution by hydrochloric acid in simple and complex porous media were performed. In addition to the VOP method, the volume of fluid (VOF) method in which the solid-fluid interface is tracked, was also applied for all cases of simulations. The results revealed that from a dissolution perspective, the performance of the VOF-curved boundary combination is essentially similar to that of the VOF-bounce back combination. However, the algorithm of bounce back was found to be more than 10% computationally efficient. Also, comparison of the outcomes from VOP and VOF indicated that VOP intrinsically overestimates the surface area of reactions by about 25-35%, which fictitiously leads to 20-30% higher reaction rates and nearly 20-30% less dissolution times. Hence, it was deduced that the temporal outcomes of VOP may not be valid. However, from a non-temporal perspective, the dissolution patterns of both methods were found to be essentially similar.

Automated summary

This research introduced a method combining VOF with curved boundary schemes for tracking the two-phase interface, and showed the application of the VOP method in a LB framework. The performance of VOF-curved boundary combination was found to be essentially similar to VOF-bounce back combination, with bounce back algorithm being more computationally efficient and VOP method overestimating reaction surface area leading to higher reaction rates and shorter dissolution times.