Time increasing rates of infiltration and reaction in porous media at the percolation thresholds

TL;DR

This study investigates how chemical reactions in porous media at percolation thresholds lead to increasing infiltration and reaction rates over time, revealing a complex interplay between fractal structures and reaction-diffusion dynamics.

Contribution

The paper introduces a scaling approach explaining the time increase of infiltration and reaction rates in porous media at critical percolation thresholds, linking exponents to fractal dimensions.

Findings

Initial subdiffusive infiltration matches fractal predictions.

Intermediate regime shows increasing infiltration and reaction rates.

Scaling relations depend on fractal and walk dimensions.

Abstract

The infiltration of a solute in a fractal porous medium is usually anomalous, but chemical reactions of the solute and that material may increase the porosity and affect the evolution of the infiltration. We study this problem in two- and three-dimensional lattices with randomly distributed porous sites at the critical percolation thresholds and with a border in contact with a reservoir of an aggressive solute. The solute infiltrates that medium by diffusion and the reactions with the impermeable sites produce new porous sites with a probability $r$, which is proportional to the ratio of reaction and diffusion rates at the scale of a lattice site. Numerical simulations for $r\ll1$ show initial subdiffusive scaling and long time Fickean scaling of the infiltrated volumes or areas, but with an intermediate regime with time increasing rates of infiltration and reaction. The anomalous exponent of the initial regime agrees with a relation previously applied to infinitely ramified fractals. We develop a scaling approach that explains the subsequent time increase of the infiltration rate, the dependence of this rate on $r$, and the crossover to the Fickean regime. The exponents of the scaling relations depend on the fractal dimensions of the critical percolation clusters and on the dimensions of random walks in those clusters. The time increase of the reaction rate is also justified by that reasoning. As $r$ decreases, there is an increase in the number of time decades of the intermediate regime, which suggests that the time increasing rates are more likely to be observed is slowly reacting systems.