Rising environmental concerns about Ordinary Portland Cement (OPC), particularly its high carbon emissions (~ 1 ton CO2 per ton of cement), have accelerated the development of sustainable alternatives such as geopolymer concrete (GPC). This paper examines how the molarity of the alkaline activator (12 M and 14 M) and the curing temperature (40 °C, 60 °C, 80 °C) affect the mechanical, durability, and microstructural performance of GGBS-based geopolymer concrete. The compressive strength increased from approximately 31 MPa under 12 M molarity and 40 °C curing conditions to approximately 53 MPa at 14 M molarity and 80 °C curing, representing an overall strength enhancement of nearly 71%. It was found that compressive strength increased with higher molarity and temperature; at 28 days, flexural strength reached 6.5 MPa, and split tensile strength reached 5.1 MPa. Water absorption decreased by an average of 25.32%, indicating improved resistance to moisture ingress and suggesting enhanced durability. Microstructural analysis indicated morphological changes consistent with a denser geopolymer matrix at higher molarity and curing temperature. The observed trends suggest improved interfacial characteristics and matrix continuity; however, direct quantification of ITZ development and crack-resistance mechanisms was beyond the scope of the present study. In addition to experimental examination, machine learning algorithms, such as Artificial Neural Networks (ANNs), Gaussian Process Regression (GPR), Support Vector Machines (SVMs), and Random Forests (RFs), were developed to predict strength properties. The highest predictive accuracy (R2= 0.97–0.98) was observed in the GPR model, followed by ANN (R2 = 0.93–0.95); RF and SVM performed relatively poorly. SHAP analysis identified molarity as the most significant parameter (40–45%), followed by curing temperature and age. This study investigates the influence of alkaline activator molarity and curing temperature on the mechanical and durability behaviour of GGBS-based geopolymer concrete. SEM, XRD, and TGA analyses were performed to evaluate the microstructural development, phase formation, and thermal stability of the developed geopolymer concrete system. Research shows that the synergistic effect between molarity and temperature controls geopolymerization, ITZ densification, and overall performance. Moreover, the combination of experimental and data-driven solutions provides an exploratory predictive framework for investigating strength development and supporting data-driven evaluation of geopolymer concrete under the studied activation conditions, thereby fostering its use as a sustainable, high-performance alternative to traditional cement-based materials.

Experimental Insights and Machine Learning Prediction of Molarity and Temperature Effects on Geopolymer Concrete

期刊论文
发表日期：

2026年6月30日

查看全文

Dipak Patare,
Pravin Nemade

摘要

Rising environmental concerns about Ordinary Portland Cement (OPC), particularly its high carbon emissions (~ 1 ton CO₂ per ton of cement), have accelerated the development of sustainable alternatives such as geopolymer concrete (GPC). This paper examines how the molarity of the alkaline activator (12 M and 14 M) and the curing temperature (40 °C, 60 °C, 80 °C) affect the mechanical, durability, and microstructural performance of GGBS-based geopolymer concrete. The compressive strength increased from approximately 31 MPa under 12 M molarity and 40 °C curing conditions to approximately 53 MPa at 14 M molarity and 80 °C curing, representing an overall strength enhancement of nearly 71%. It was found that compressive strength increased with higher molarity and temperature; at 28 days, flexural strength reached 6.5 MPa, and split tensile strength reached 5.1 MPa. Water absorption decreased by an average of 25.32%, indicating improved resistance to moisture ingress and suggesting enhanced durability. Microstructural analysis indicated morphological changes consistent with a denser geopolymer matrix at higher molarity and curing temperature. The observed trends suggest improved interfacial characteristics and matrix continuity; however, direct quantification of ITZ development and crack-resistance mechanisms was beyond the scope of the present study. In addition to experimental examination, machine learning algorithms, such as Artificial Neural Networks (ANNs), Gaussian Process Regression (GPR), Support Vector Machines (SVMs), and Random Forests (RFs), were developed to predict strength properties. The highest predictive accuracy (R²= 0.97–0.98) was observed in the GPR model, followed by ANN (R² = 0.93–0.95); RF and SVM performed relatively poorly. SHAP analysis identified molarity as the most significant parameter (40–45%), followed by curing temperature and age. This study investigates the influence of alkaline activator molarity and curing temperature on the mechanical and durability behaviour of GGBS-based geopolymer concrete. SEM, XRD, and TGA analyses were performed to evaluate the microstructural development, phase formation, and thermal stability of the developed geopolymer concrete system. Research shows that the synergistic effect between molarity and temperature controls geopolymerization, ITZ densification, and overall performance. Moreover, the combination of experimental and data-driven solutions provides an exploratory predictive framework for investigating strength development and supporting data-driven evaluation of geopolymer concrete under the studied activation conditions, thereby fostering its use as a sustainable, high-performance alternative to traditional cement-based materials.