Influence of Fly Ash, Alcofine, Alkaline Molarity, Curing Duration and Machine Learning Predictions on Geopolymer Concrete

Abstract

The study is motivated from the perspective of developing an eco-friendly and effective alternative to cement-based materials. One of the contributions of this research lies in the assessment of three different concentrations of sodium hydroxide (8M, 12M, and 16M) along with the use of Activated Low-Calcium Fly Ash (ALF) as a supplementary cementitious material. Another contribution is the use of predictive modelling on experimental data using Decision Tree regression for estimating the compressive strength. The experimental results showed that the workability, compressive strength, split tensile strength, and flexural strength of GPMC improved with the increase of fly ash content from 325 to 400 kg/m³. Also, the best mechanical performance was recorded at the highest molarity of 16M and 28 days of curing. Analysing the stress and strain showed typical elastic behaviour with gradual softening after peak stress which is a characteristic of geopolymer-based materials. Testing for water absorption showed that increase in fly ash content resulted in lower porosity which indicates increased density and long-term durability. In addition, the predictive modelling approach also reached a very high coefficient of determination, validating the model’s robustness and the relationship between the input parameters and the compressive strength value. As a conclusion, the research emphasises the capabilities of optimising the mix design parameters to enhance the mechanical properties of the geopolymer concrete, making it more eco-friendly when compared to traditional Portland cement based systems.