Specific interaction theory versus Pitzer’s model in groundwaters and brines for checking equilibria/non-equilibria with calcite, gypsum, and halite: application to predict the evolution of solutions concentrated by evaporation in irrigated areas

Mass transfer between aquifers, vadose zone, including soils and waters may occur at equilibrium or out of equilibrium. Irrigating with low-quality waters can result in soil salinization and/or degradation of soil structure. Checking minerals/solutions equilibria from the chemical composition of solutions implies computing activities and Saturation Indexes (SI) of minerals. In semi-arid-to-arid areas, evaporation concentrates solutions and waters evolve in different geochemical pathways, mainly saline neutral path and alkaline path, separated by bifurcations. Strong non ideality of electrolyte solutions makes it difficult to compute accurately activities and SI. The objective of this paper is to compare Pitzer’s model and Specific Interaction Theory (SIT), both now incorporated in Phreeqc 3.0. Samples can be assigned to the saline neutral path with dominance of sulfate which is the majority and with dominance of chloride as the minority. Data were twofold: (i) groundwaters were sampled in an irrigated plain, in Lower Chéliff valley (Algeria), and analyzed, they cover the range from low to medium ionic strength; (ii) data from a saline system (Chott El Jerid, Tunisia) were taken from the literature to cover the range from medium to very high ionic strength, including brines. Data were processed with both models to check equilibria. Results opposing classical assumptions are obtained: (i) calcite does not form at equilibrium and requires a specific oversaturation ( SI ≃ 1.4 ), then relaxes to equilibrium. This is a general result that can be extended to many situations, where calcite forms, including sedimentation; (ii) gypsum, which is more soluble, forms at equilibrium; accordingly, the assumption of equilibrium at low temperature, i.e., in Earth’s surface conditions, holds for gypsum, but not for calcite; (iii) Pitzer’s model gives better results than SIT for calcite and gypsum, but SIT model gives better results for halite, while it is generally admitted that Pitzer’s model is better for I > 3 m .