Study of Early-Age Hydration, Mechanical Properties Development, and Microstructure Evolution of Manufactured Sand Concrete Mixed with Granite Stone Powder.

This study explored the potential of granite stone powder (GSP) as a supplementary cementitious material (SCM). The 72 h early hydration process stages of GSP-mixed slurry were analyzed in depth, and the mechanical properties of manufactured sand concrete (MSC) mixed with GSP were investigated. Physical phase types, morphological characteristics, and pore structure evolution were investigated using an X-ray diffractometer, scanning electron microscope, and mercury intrusion approach (MIP). Atomic force microscopy was used to show the interface transition zone between aggregate and slurry in phase images, height images, and 3D images, allowing quantification of ITZ and slurry by calculating the roughness. Gray entropy analysis was used to evaluate the significance of the effect of pore size distribution parameters on mechanical strength, and the GSP-content-mechanical-strength gray model GM (1, 1) was established to predict mechanical strength. The results indicate that, compared with the reference group, the GSP cement slurry system exhibited a delayed hydration process acceleration rate, with a 1.04% increase in cumulative heat of hydration observed in the 5% test group and an 11.05% decrease in the 15% test group. Incorporating GSP in MSC led to decreased mechanical properties at all ages, with significant decay observed when incorporation ranged from 10% to 15%. Although the type of hydration products remained unchanged, there was a decrease in the number of C-S-H gels and gel pores, while large pores increased, resulting in increased porosity and roughness of the interface transition zone and slurry. Large pores (>1000 nm) were found to have the greatest influence on mechanical strength, with gray correlation above 0.86. The GM (1, 1) model yielded accurate predictions, showing good agreement with measured data and thus it can be identified as belonging to a high-precision prediction model category. These findings provide theoretical support and a reference for applying GSP as an SCM, laying the groundwork for data-based specification development.