Abstract:

Natural stones are the widely used building materials in civil engineering and architectural projects. Surface abrasion resistance determination of natural stones is one of the important issues considered in flooring projects. This mechanical property is determined by performing standardized laboratory tests. In recent years, however, there has been considerable interest in its determination by means of indirect methods using the physical, mechanical, and mineralogical properties of stone materials. As a new approach, the present work was undertaken to investigate the potential usability of chemical compositions of true marbles as a guide to predict their Wide Wheel Abrasion (WWA) and Böhme Abrasion (BA) test values. By performing multiple linear regression analyses, statistically significant regression models with high prediction accuracies were derived (p-values < 0.05; R2 ═ 0.98­0.99). In the established prediction models MgO, SiO2, CaO, and (CaO/SiO2) ratio were determined as the most significant predictor variables. These models may be updated by inclusion of data from further studies on other types of marbles.

Keywords: