Effect of porous structures in zeolite/geopolymer composites on flow fields using GPU accelerated volumetric lattice Boltzmann method*

Nuclear waste containing radionuclei such as cesium (Cs) and strontium (Sr) isotopes are hazardous to the environment. Composites with porous structures are usually used to immobilize Cs/Sr by uptake from radioactive effluent streams. For the development of nuclear waste materials devoted to Cs/Sr absorption, it is important to understand the effect of porous structures in composites on flow fields. In this work, we used volumetric lattice Boltzmann method (VLBM) to solve complex flow in porous structures. A MATLAB code was developed to reconstruct 3-D porous structures from images generated by scanning electron microscope. GPU parallel computing is combined with VLBM to get higher computational efficiency. Validation of this method against the analytical solutions in a 3-D straight pipe and numerical solutions for a simple porous structure using a finite element software capability developed from the Idaho National Laboratory's Multiphysics Object-Oriented Simulation Environment shows good agreement. By comparing the flow in pure geopolymers and zeolite/geopolymer composites, it is expected to glean insights into the effect of porous structures on flow fields which could enhance the development of advanced nuclear waste form materials with better Cs/Sr absorption performance.