Pore solution chemistry of alkali-activated slag: a statistical and multi-factor analysis for predicting ion composition

Predicting pore solution chemistry in alkali-activated slag (AAS) is critical for understanding hydration mechanisms and durability. To address this, a comprehensive database was first compiled from literature and analyzed using dual sensitivity analysis (Pearson correlation and Grey Relational Analysis), which identified the Ca/Si ratio, Si/Al ratio, and water-to-solid ratio as the most influential factors governing key ion concentrations (OH-, Na, K, Al, Ca, Si). Based on these findings, targeted experiments were designed to generate a consistent dataset. A novel multi-factor nonlinear regression model was developed, successfully describing each ion concentration as a function of hydration time and the above ratios. The model demonstrates high accuracy, with R2 values exceeding 0.93 for OH-, Na, K, and Si, while most predictions are within ±15% of experimental measurements. This work establishes a robust, data-driven framework for predicting AAS pore solution chemistry, facilitating rational mix design of sustainable cementitious materials as essential input parameters.