EcoBlendNet: A physics-informed neural network for optimizing supplementary material replacement to reduce the carbon footprint during cement hydration

Abstract The addition of supplementary cementitious materials (SCMs) to cement triggers a complex cement hydration system characterized by intricate mineral admixture interactions. This work develops EcoBlendNet, a novel physics-informed neural network (PINN), to analyze carbon emissions during SCMs-enhanced cement hydration. EcoBlendNet integrates experimental data and the chemo-physical aspects of cement hydration in a heated cement paste, accurately predicting concrete maturity and compressive strength by capturing early-age temperature rises for various mixing blends, including Portland cement, cement-fly ash blends, and cement-slag blends. SCMs effectively reduce temperature rises without compromising strength development. The work illustrates a statistical method to efficiently leverage limited SCMs resources for mitigating environmental impacts in concrete construction. Quantitative analysis reveal that replacing 45–80% of cement with industrial fly ash and slag can reduce CO2 emissions by 60–80% during cement hydration. The validated EcoBlendNet offers a precise and interpretable tool for eco-friendly mixing blend selection, harmonizing with experimental methods. It’s adaptable to different material properties and mineral admixtures, thereby promoting eco-friendly concrete construction.