Evaluating Leeb hardness as a reliable UCS predictor for stone masonry exposed to high temperatures: The influence of porosity across diverse carbonate lithologies

High temperatures significantly impact natural stone materials, altering their mechanical and physical properties. Traditional destructive methods for assessing these changes are often limited in some stone applications underscoring the importance of non-destructive techniques (NDTs) for accurate predictions. Leeb Hardness (HL) stands out as a promising NDT due to its simplicity, portability, and cost-effectiveness. This study investigates the effects of high-temperature exposure (300 °C and 600 °C) on HL, Ultrasonic Pulse Velocity (UPV), and Uniaxial Compressive Strength (UCS), and explores their interrelationships. While UPV has been previously studied as a UCS predictor under high-temperature conditions, this research evaluates HL’s potential as an alternative NDT, comparing its predictive capacity with UPV in the same context. Thirteen carbonate stones—including limestones, marbles, travertine, and breccia—were analysed due to their widespread application in commercial and heritage contexts. The study also highlights the critical influence of factors such as porosity, anisotropy, and grain size on these correlations. The findings provide a comprehensive framework for understanding the interplay between physical and mechanical properties in stones exposed to high temperatures, offering valuable insights into the characterization of stone materials in post-fire scenarios. Moreover, the reliability of NDTs for predicting UCS was demonstrated, with HL emerging as a more consistent predictor under diverse thermal conditions, reinforcing its applicability in high-temperature assessment of carbonate stones. Highlights : • Leeb Hardness (LH) outperforms UPV as a UCS predictor in thermally stressed stones. • Reference porosity is key for UCS estimations, influencing model accuracy. • Grain size affects LH-UCS correlations, while anisotropy impacts UPV-UCS models. • Temperature-specific models enhance UCS predictions across thermal conditions.