PSI - Issue 66

Karolina Głowacka et al. / Procedia Structural Integrity 66 (2024) 108 – 121 Author name / Structural Integrity Procedia 00 (2025) 000–000

119

12

The first set of drawings, i.e., Fig. 3, corresponds to the three-point bending of a sample 2 mm high, 15 mm wide, and with a support spacing of 60 mm. This situation reflects pure bending – almost no shear, but still, stresses/strains from the loading pin and supports were unavoidable. During the loading pin displacement at a constant speed, the force increased linearly until sudden failure. The sample underwent uniform deformation in an arc until time ‘1’. The DIC analysis photo shows that at the place where failure was to occur, the strain along the fibers was 2.2%, while the remaining strains were insignificant. In the photo at a time ‘2’, one can observe a ‘frame’ captured after the maximum force was obtained, i.e., after the sudden breakage of the fibers in the middle of the sample, from the bottom of the sample, i.e., from the tension side. At that time, the shape no longer resembled an arc but two halves of a linear shape. Only the area near the loading pin remained in an arc shape. Although the further analysis was not significant from the point of view of the material's mechanical strength (the element no longer transmitted significant force), loading of the sample was continued, which resulted in the observation of increasingly visible broken fibers torn from the matrix. The next set of photos, i.e., Fig. 4, represents three-point bending samples with a height of 20 mm, width of 25 mm, and support spacing of 60 mm. This is one of the methods of conducting shear strength tests [17] because the damaging shear stresses, in this case, are shear stresses, not normal stresses. In this case, the first part of the bending test of the sample (until failure, moment ‘1’) was predictable, i.e., the force increased linearly during the displacement of the loading pin. However, when the maximum force was obtained, the destruction of the sample was not noticeable to the unarmed eye. Additionally, only a slight decrease in the value can be observed on the loading force graph, and the view of the sample did not change at all. However, when observing the analysis using DIC, it was visible that significant shear strains appeared on the sample, especially in the middle of the sample height, where the maximum shear stresses occur (resulting from physics). Horizontal red lines are also visible at different sample heights, indicating very high shear strain occurring between individual layers. Then, shortly after the force dropped, it began to increase again, but this time not as a result of bending but compression along the height of the sample. At that time, numerous cracks between the layers of the sample began to be visible, which confirmed that the first destruction came from delamination occurring as a result of the shear stresses. As a result of compression, the value of the force increased, even significantly above the maximum force that led to delamination. Only after exceeding a specific limit, marked with a circle, was the sample already so delaminating that the individual layers slipped between each other, and the value of the loading force dropped. The behavior of the sample was observed until complete destruction, marked with the moment ‘2’. Another set of photos (Fig. 5) represents four-point bending samples with a height of 10 mm, a width of 15 mm, and a support spacing of 240 mm external and 160 mm internal. In this case, at the moment of failure, a noticeable delamination occurred at the mid-height of the sample. Referring to the bending graph, it is worth noting that after reaching the maximum force (‘1’), the force value first decreased slightly. The failure was imperceptible to the naked eye (although with the use of DIC, a shear strain of 1.8° could be observed, which could be expected to destroy the material soon), after which the force decreased rapidly. The sample underwent a noticeable delamination (‘2’). The incomplete decrease in force can be attributed to the partial bridging of the fibers in the composite. Then, during further loading, the force continued to increase, behaving as if it was not one damaged sample with a height of 10 mm but two stacked on top of each other with a height of about 5 mm. After the destruction of these "halves," another drop was observed in the force graph, while the "halves" themselves were subject to further delamination ('3'). In the case of the next set of photos (Fig. 6) representing a four-point bending sample with a height of 20 mm, a width of 25 mm, and a support spacing of 240 mm outside and 160 mm inside, the behavior was observed analogous to the previous four-point bending. However, since the initial height was twice as large, the sample split into two parts more times, which can also be deduced from the bending graph. The most interesting case was the case of a three-point bending sample with a height of 10 mm, width of 15 mm, and support span of 120 mm. In this case, both types of destruction were observed (translaminar and interlaminar fractures), which makes it difficult to clearly determine whether, with such a ratio of normal to shear stresses ( σ max / τ max ≈ 24), normal or shear stresses (strains) led to destruction. Ultimately, three samples were destroyed as a result of delamination, one by translaminar cracking, while in the case of one, both modes of fracture co-occurred. Different types of cracks probably resulted from the uneven distribution of fibers in the material - when the sample was more homogeneous, there were no significantly less durable areas; therefore, destruction with this ratio of normal to shear