Response Surface Method for Parameter Optimization of Soils Stabilized with an Alternative Binder

This paper examines the interplay and optimization of influencing parameters in expansive soil stabilization using ground granulated blast furnace slag (GGBFS), which is an industrial byproduct. In this study, bentonite-sand (B-S) mixtures of two different proportions were prepared to simulate soils of distinct degrees of expansiveness and mixed with different percentages of GGBFS and molding water content and cured for 0, 14, and 28 days with specimen combinations based on Design of Experiments (DoE). The collective effect of parameters such as binder content, molding water content, and curing time was studied and a mathematical correlation was developed using the response surface methodology (RSM) based on a face-centered central composite design for optimal response of unconfined compressive strength (UCS). Statistical significance of the developed model was established by ANOVA. The results indicated that among the chosen variables, curing time had a higher contribution to UCS of GGBFS-treated soils. Optimized GGBFS content for B-S mixtures with lower sand content was found to be 20.69%, which yielded a UCS of 539.5 kPa, and the B-S mixture with a higher sand content demanded a higher binder content (24.89%) to achieve a maximum UCS of 592.8 kPa. The highest UCS was attained for both B-S mixtures when the mixtures were molded with a water content higher than the respective optimum moisture contents (OMCs). The modified free swell index (MFSI) of the treated soils was also evaluated to assess the influence of GGBFS on the swelling potential and revealed that stabilization with GGBFS significantly reduced the MFSI. The findings of this study will significantly aid in decision-making regarding the optimization of GGBFS content with reference to molding water content and curing period with relevance to expansive soil.