PSI - Issue 68

Masayuki Arai et al. / Procedia Structural Integrity 68 (2025) 3–8 M. Arai et al. / Structural Integrity Procedia 00 (2025) 000–000

7

5

at room temperature. The load cell and crosshead displacements were recorded during the tensile tests using a data logger (midi LOGGER GL240, Graphtec Corporation), and crack propagation was continuously observed using a digital camera. 3.2. Test results and discussion Table 2 shows representative results of continuous observations during the tensile test of the CT specimen. For ℎ = 3.2, the M and F parts fracture as they are pulled apart along the joint interface. In contrast, in the FE analysis results, slip deformation is observed at the joint interface; however, the crack ultimately propagates to penetrate the base material at the joint root on the F side, causing the fracture. For ℎ = 9.6 mm, the M and F parts are caught at the kerakubi and pulled apart, causing the fracture. On the other hand, in the FE analysis results, although a crack occurs owing to stress concentration at the sickle-joint root of the F part, those parts slide along the sickle-joint interface and fracture as they are pulled apart. Therefore, for this case, the experimental and FE results are in agreement. Finally, for ℎ = 10.4 mm, slip deformation occurs along the sickle-joint interface, and the smallest cross-section of the joint and the kerakubi are torn off, causing fracture. In contrast, in the FE analysis results, the crack initiated at the root of the sickle joint in the F part propagates upward through the base material. By comparing the experimental results with the FE analysis results, we found that in some cases the interlocking structure is not significantly expressed in the test results. This is because differences in the radius shape at the corners of the 3D-printed CT test specimens, and differences reaching a few microns between the CAD data and the shape of the actual molded object, appear as differences in the stress concentration factor.

Table 2. Continuous observation results.

Fig. 5 shows the relationship between the sickle-joint height ( ℎ) and the fracture energy. The fracture energy was calculated by image processing of the area in the load–displacement relationship. In the figure, ● represents the test results and × represents the FE analysis results. The fracture energy in the test results increases with the sickle-joint height, reaching a maximum value at a height of approximately 6.0 mm then decreases. The FE analysis results show a similar tendency to that of the test results, although the FE analysis results are slightly higher. Therefore, when a sickle-joint structure is introduced into a brittle material, the material deforms significantly owing to the mutual sliding deformation at the sickle-joint interface and the elongation of the sickle joint, and a large fracture energy can be expected. However, the results of the tensile tests on the CT specimens produced by the stereolithography 3D printer do not fully agree with the FE analysis, because the stereolithography 3D printer was not able to fully reproduce the sickle-joint shape. Nevertheless, the FE analysis and tensile testing provide a perspective on improving the toughness by introducing the sickle-joint interlocking structure proposed in this study.