PSI - Issue 39

Fabrizio Greco et al. / Procedia Structural Integrity 39 (2022) 638–648 Author name / Structural Integrity Procedia 00 (2019) 000–000

644

7

3. Results This section reports numerical results devoted to assessing the reliability and efficiency of the proposed diffuse interface (DIM) model. Specifically, the proposed DIM is applied to simulate the direct tensile test of the brick masonry couplet depicted in Figure 3-a. The couplet involves two units of length 100 mm, height 50 mm, and thickness 100 mm bonded with a mortar bed joint of 15 mm. Table 1 summarizes the mechanical properties of the brick units and mortar layer, which are consistent with experimental data reported in (Pluijm (1997)). Two steel plates of length 130 mm, height 65 mm, and thickness 130 mm embed the specimen, thus reproducing boundary conditions induced by the testing machine. Because of the symmetry of the geometry and boundary conditions, the numerical model is implemented involving only half of the specimen, imposing external constraints as depicted in Figure 3-b. From a mechanical point of view, under tensile actions, the couplet can fail because of the crisis of either brick mortar interfaces or mortar cracking. The proposed DIM approach can account for these distinct failure modes by adopting different cohesive properties for Brick-Mortar (B-M) and Mortar-Mortar (M-M) interfaces. Numerically, the crisis of the couplet originates in the weakest interface embedded into the model. Since it is beyond the scope of this study to examine the failure of the couplet because of the differences in B-M and M-M interfaces, only the failure of the couplet because of the crisis of the B-M interfaces is investigated, inserting cohesive elements only along brick mortar interfaces. The resulting numerical model configures a mechanical system made of two horizontal interfaces arranged in series (see Figure 3-b). As a consequence, there are 3 solutions for the current problem, which can be grouped into two families: the first comprises two solutions corresponding to the fracture of either the upper or bottom B-M interface; the second family contains the (only) solution that involves the simultaneous damaging of both interfaces. Of course, the first family contains stable solutions, whereas the second is the metastable one. To analyze the predictions of the proposed DIM without imperfections accurately, giving rise to stable and metastable solutions, two auxiliary models are considered (see Figure 3-c). The first (named SIM 1) has the same geometric configuration as the DIM but presents a single layer of cohesive elements on the upper B-M interface only. Such a model allows the computation of the stable solution. The second (denoted as SIM 2) comprises only a quarter of the specimen, with cohesive elements along with the upper B-M interface. Symmetry constraints permit reproducing the simultaneous damaging of both interfaces, thus enabling the computation of the metastable solution.

Figure 3. (a) A brick masonry couplet under direct tensile test; (b) a schematic of the proposed DIM model; (c) auxiliary models to analyze the fracture at Brick-Mortar (B-M) interfaces.