Experimental characterisation of the friction coefficient of mortar bed joints in clay-brick masonry

Understanding the characteristics of material properties is an important aspect of the design of new, and assessment of existing, structures. Furthermore, the development of accurate statistical models of these properties is critical to a reliability-based assessment of structures. Examining the properties of masonry materials has been the subject of many publications, but few consider the friction behaviour of the brick-mortar interface under shear loading. This paper presents an extensive repetitive laboratory testing study where the residual friction responses of a range of extruded (perforated), pressed (frogged) and solid clay bricks have been observed under direct shear loading at various levels of pre-compressive normal stress. The sensitivity of this friction resistance to the interlocking mechanisms of extruded and pressed brick masonry, as well as the amount of normal stress, has been considered, facilitating the determination of a suitable probabilistic model of the friction coefficient. It was determined that the residual friction coefficient of clay-brick masonry can be accurately approximated with a Normal (Gaussian) distribution, and is comparably variable to the masonry compression strength, with a coefficient of variation equal to 0.14. Furthermore, the influence that pre-existing, tension-induced cracking has on the friction coefficient has been examined. It was found that tension-induced cracks do not significantly affect the residual friction coefficient. • Extensive experimental testing of unreinforced masonry behaviour under direct shear loading. • Probabilistic characterisation of the friction coefficient of the unit-mortar interface of unreinforced masonry. • Presentation of raw experimental data from 246 shear tests and 54 direct tension tests. • Experimental investigations exploring the influence of pre-existing tension cracks on the friction resistance under direct shear loading, as well as the difference in the directly measured residual friction coefficient and the initial friction coefficient determined in accordance with EN 1052–3 (2002).