Modeling and optimization of fly ash–slag-based geopolymer using response surface method and its application in soft soil stabilization

Abstract

Quantitatively optimizing critical factors for geopolymer production and explaining the interaction effects between each factor are significant for engineering applications. A three-level Box–Behnken design of response surface methodology was applied to optimize the fly ash–slag-based geopolymer paste (17 experimental mixture) and the main factors selected for the investigation were alkali equivalent, activator modulus, and slag replacement ratio to achieve maximum compressive strength. The results were fitted with the quadratic polynomial equation using multiple regression analysis and the model provided an accurate and reliable fit to the factual data. Afterward, this study investigates the use of optimized paste as a sustainable stabilizer (grouting reinforcement method) for improving the mechanical performance of soft soil in Hangzhou, China. By studying the compressive strength of stabilized soil with various stabilizer content (8%–14%), curing age (0–28 days), and moisture content (30%–60%), the effects of these preparation parameters were evaluated. Moreover, the quasi-water–cement ratio was introduced to predict the stabilized soil’s strength development and a corresponding empirical formula (correlation coefficient of 0.98) was proposed. The changes in microstructure, mineral phase, and molecule bonds were investigated using the XRD, FTIR, and FESEM, respectively. The results reveal that the decrease in the initial moisture content and increase in geopolymer inclusion have several improvement effects on curing strength. The geopolymer gel structure gradually formed, binding soil particles together with the new hydration product formation after curing, and the final stabilized soil was presented to have a more compact and strong microstructure.