A coupled physical–chemical model for mass transfer considering damage evolution: Sulfate ions transport in a cement-based material as an example

Abstract

Damage induced in mass transfer process is a ubiquitous phenomenon in nature and living systems. A coupled physical–chemical modeling framework is proposed to clarify the mass transfer in damaged media. Such a theoretical framework shows that the mass transfer problem is no longer linear but non-linear, due to the mass transfer behavior is influenced not only by transport driving mechanisms but also by physical consumption mechanisms, chemical consumption mechanisms, and damage evolution. Based on the theoretical framework presented in this paper, one can choose specific mechanisms in each category, different mass transfer theories can be constructed to meet specific practical applications requirements. To verify the correctness and validity of this theoretical framework, we choose the theoretical model of sulfate ion transport in cement-based materials. This is because sulfate ions are the main cause to induce the corrosion cracks in concrete in the marine environment. The following mechanisms are considered: concentration gradient driven transfer, chemical reaction, continued hydration, filling effect, and expansion damage. 3D nonlinear partial differential equations with variable coefficients are derived for the system. The numerical results show that the chemical reaction has the dual characteristics of self-stagnation and acceleration of sulfate ion transport. The chemical reaction products consume the free ions and fill the pore structure, which reflects the self-stagnation of mass transfer, while the expansion force generated by the chemical products leads to the initiation of corrosion crack, which reflects the acceleration of mass transfer. The competitive outcome of the two mechanisms is acceleration feature as the dominant one. Finally, the proposed model efficacy is validated by comparing the numerical results with previously reported experimental data.