Microstructural and optimization studies on novel one-part geopolymer pastes by Box-Behnken response surface design method

Abstract

This paper reports the work on developing an optimized mix proportion of novel one-part geopolymer (OPG) binder produced by dry blending the solid aluminosilicate precursor and solid alkali source and then adding free water to the blended mix similar to the preparation of Ordinary Portland Cement (OPC). A three-level Box-Behnken Response Surface Method (RSM) design was used to study the properties of OPG mixes at fresh and hardened state and to test and develop the regression models. The Ground Granulated Blast Furnace Slag (GGBS) substitution, water to geopolymer solids (w/s) ratio, and the activator dosage were considered as the independent variables. The response target values were the flow value, initial and final setting time, and compressive strength. The multiple regression analysis with the quadratic polynomial model was used to fit the data, which offered an accurate and reliable match to the actual data. Scanning Electron Microscope (SEM), X-ray Diffraction (XRD), and Fourier Transform Infrared Spectroscopy (FTIR) were used to study changes in microstructure, mineral phase, and molecular bonding of OPG mixes, respectively. Based on the material characterization observation, the change in GGBS addition, w/s ratio, and activator dosage were discovered to have a considerable impact on both the fresh and hardened properties. The optimum mix proportion obtained was 51.39% GGBS substitution, 0.32 w/s, and 12.35% activator content, with 191 mm flow, 68.56 MPa of compressive strength, 59 and 191 mins of initial and final setting time, respectively. The target values obtained using the one-part geopolymer mix with 50% GGBS substitution, 0.3 w/s, and 12% activator content were in close agreement with the target values predicted by the optimized mix, confirming the efficiency of RSM in obtaining the optimum one-part geopolymer mix proportion.