Sulfate attack and chloride-induced corrosion are recognized as crucial factors leading to the deterioration and failure of concrete structures. Chloride ingress triggers steel corrosion, while sulfate attack alters the microstructure of concrete matrix. When the concrete is subjected to a poorly mineralized or acid solution, calcium leaching happens and has also a negative effect on concrete durability. In harsh natural environments, concrete is vulnerable to the combined sulfate-chloride attacks, often accompanied by calcium leaching. The mechanism of coupled ionic attack differs substantially from that of individual attacks. This paper was to propose a numerical model on the interaction between sulfate attack, chloride attack and calcium leaching. The ionic transport patterns and hydration product distributions under coupled degradation were quantitively analyzed by the proposed model, thereby providing new insights into the deterioration mechanisms in concrete under the simultaneous sulfate-chloride attack and calcium leaching.

A comprehensive transport model was established based on the accelerating effect of aggressive ions on the calcium solid-liquid equilibrium curve. The impact of calcium leaching on the ionic binding effect was also taken into accounts. The chemical reactions between multiple ionic species and cementitious matrix were incorporated based on the transport model. A reaction source term intuitively reflecting the desorption effect of sulfate ions on chloride ions was also introduced. A transport-chemo model was further developed. Based on the volumetric changes induced by expansive hydration products and calcium leaching coupled with the damage effect caused by expansive cracking in the matrix, a time-dependent diffusivity model was established. Finally, an integrated coupled degradation model was proposed, encompassing transport, chemical reactions, and time-dependent diffusivity. The proposed model was validated through third-party experiments according to sulfate concentration distribution and chloride ion concentration distribution. This model is capable to predict ionic transport and product distribution based on given initial material parameters and external ion concentrations. Some indexes such as ionic penetration depth, time-varying diffusion coefficient and product content could be applied to monitor the deterioration process of concrete.

Under combined sulfate-chloride attack, there are two distinct regions in the concrete, i.e., a sulfate-rich zone and a chloride-rich zone. The diffusion of free sulfate ions is restricted, so that the increase in total sulfate content peak at a concrete depth of 3 mm is attributed to the precipitation of ettringite (AFt). As an intermediate product of AFt, the content of newly produced gypsum remains low. In contrast, chloride ions progressively penetrate deeper, resulting in the accumulation of chlorine hydrate products such as Friedel's salt, in which a peak gradually moves inward. The production of Friedel's salt falls between AFt and gypsum in terms of concentration.

The concrete surface is primarily affected via calcium leaching, leading to a more than fourfold increase in the diffusion coefficient. As the influence of calcium leaching diminishes with increasing concrete depth, there is a rapid reduction in porosity. A distinct trough in the porosity curve emerges at approximately 3 mm, precisely aligning with the sulfate-rich zone. The mutual counteraction of internal pore filling effect and damage effect due to expansive cracks results in the time-dependent diffusion coefficient close to the initial value.

Simultaneous exposure to sulfate and chloride attacks may provide a short-term mitigation of either sulfate or chloride attack individually. Chloride ions have a more significant inhibitory effect on sulfate attack under these combined conditions. For calcium leaching, sulfate attack can become more severe and concentration peaks of ettringite, and gypsum appear near the surface. Neglecting calcium leaching inhibits the precipitation of ettringite, thus causing the peak content of ettringite to decrease from 201 mol/m³ to 23 mol/m³, which is only 1/10 of that for calcium leaching.

Increasing external chloride concentrations amplify a suppressive impact on sulfate attack. Apart from the environment factor, a higher initial calcium aluminate content in concrete corresponds to a more pronounced sulfate attack in concrete matrix. The concentration of hydrate products increases linearly with the extension of time.

The proposed model had the interactions among various factors and provided a rational prediction of the coupled degradation process. Concrete had distinct sulfate and chloride zones. The sulfate penetration depth remained with the deposition of AFt near the surface. In contrast, the chloride penetration depth gradually increased with a broader distribution of generated Friedel's salt. The trough of porosity distribution corresponded to sulfate-rich zone at depth of 3 mm, in which the diffusion coefficient was approximate to the initial value. The pore filling effect was offset by the damage effect caused by expansive cracks. Combined attack could mitigate sulfate or chloride attack in the short term. Chloride ions had a more pronounced inhibitory effect on sulfate attack. Calcium leaching led to less ettringite generation but a wider distribution in concrete. Based on the proposed model, a parameter analysis of environmental or material parameters was performed, thus having insights to service life prediction of concrete under chloride-sulfate attack.