Validated a priori calculation of tortuosity in porous materials including sandstone and limestone

An algorithm has been developed for the a priori calculation of tortuosity in a simulated void network, assuming a Markovian random walk process, with paths identified using the algorithms of Yen and Dijkstra, and path searching extended by successive pruning of the network graph. The void network is derived from the inverse modelling of percolation characteristics derived from mercury intrusion porosimetry. Experimental tortuosities have been determined for two porous limestones and one porous sandstone, by measuring the electric conductivity of inter-pore brine relative to that of the same quantity of bulk brine. A close match between simulation and experiment is obtained (R2=0.95). Tortuosities are also calculated for larger charged particles and for viscous transport. Further validation is provided in the form of a sensitivity analysis of tortuosity with respect to network connectivity. The new approach is particularly useful for the many materials that can be characterised by mercury porosimetry or porometry, but for which tortuosity cannot be measured directly. It has applications in a wide range of areas of current interest, such as oil and gas engineering, nuclear reactor core modelling, filtration, catalysis, ceramics, membranes and soil science.