Si–Al Pitzer dataset: Consistent set of Pitzer activity model interaction parameters of Al and Si species, for modelling cements in saline systems with THEREDA

The Si and Al thermodynamic data and Pitzer interaction parameters (IPs) from THEREDA were re-evaluated to extend the applicability of the database for modeling cementitious systems from low to highly saline environments. This study involved reassessing the standard reference thermodynamic data for Si and Al aqueous species and relevant solids (quartz, gibbsite, boehmite) and producing a new consistent set of polythermal Pitzer interaction parameters for Si and Al species within the oceanic salt system Al–Si–Ca–Mg–Na–K–Cl–SO4–CO3–H2O. Experimental data done for a range of compositions and temperatures were used optimize the interaction parameters and their temperature coefficients. In the absence of data, estimation methods using analogous species or correlations with the SIT aqueous activity model parameters were applied. For Si in acidic to neutral pH regions, interaction parameters for Si(OH)4(aq) were determined from extensive binary and ternary experimental data. The formation constant for the first silicic acid hydrolysis and its temperature dependence were assessed, while the second hydrolysis species required estimation methods due to limited data. Interaction parameters with K+ were estimated, and the effect of Ca and Mg at high pH was accounted for using complexes between Si and alkaline earth metals. At high Si concentrations, polymeric species impact aqueous Si speciation. Due to limited data, selecting the silica tetramer (Si4O8(OH)44–) and its binary IPs sufficed to reproduce amorphous Si solubility data in NaCl solutions. For Al, standard thermodynamic properties of ions, hydrolyzed species, and hydroxide phases were derived using gibbsite solubility measurements. No temperature coefficients were required for interactions in the Al–Cl–H2O system, yielding similar results in ternary and quaternary systems. Complexes between aluminum and sulfate were accounted for by the β(2) interaction parameter. At intermediate pH, few measurements exist to derive IPs for hydrolyzed aluminum species, with only Cl– interactions derived. Al(OH)3(aq) IPs were assumed similar to Si(OH)4(aq). At high pH, extensive gibbsite solubility data enabled consistent IPs for the Na–K–Cl–OH–Al(OH)4 system. Potential IPs with Ca2+ and Mg2+ were not selected due to precipitation of hydrates in cementitious systems and the use complexes instead. Ternary IPs with SO42– and CO32– were estimated using OH– as an analog for Al(OH)4–. This provided satisfactory predictions from gibbsite solubility measurements in Na2CO3 and Na2SO4 solutions. No significant Si and Al interactions were expected in high pH cement system due to the low Si concentration maintained by cement phases like C–S–H or M–S–H precipitation. The updated Pitzer model for Si and Al aqueous speciation was compared with independent data on cement hydrate phases and solutions in moderate concentrations up to 85 °C, showing good performance. Additional testing against salt-rich systems and further investigations are recommended. The fitted, estimated, and fixed parameter values were documented, and the updated dataset was implemented in THEREDA (https://www.thereda.de/) for use in calculations.