PSI - Issue 41

Fabio Distefano et al. / Procedia Structural Integrity 41 (2022) 470–485 Author name / Structural Integrity Procedia 00 (2019) 000–000

477

8

Where D1, D2, D3, D4 and D5 are material constants. For the G7 unit cells, in order to have at least three elements along the strut diameter, the mesh size adopted was 0.06 mm for cells with strut diameter of 0.2 mm, and 0.16 mm for cells with strut diameter 0.5 mm. For the IWP unit cells, the surface was meshed with elements of 0.15 mm size, while three elements were guaranteed along cells thickness. The cells were modelled with first order hexahedral elements, while the plates with quad elements. Boundary conditions aim at replicating the compression tests. An imposed velocity of 1 mm/min along y axis was applied on the superior plate and the inferior plate was constrained in all directions as shown in Fig. 4. In addition, on the superior plate was applied a further constraint to avoid translation in x and z directions, and rotation in all directions.

3. Results 3.1. Morphological analysis

SEM observations of the structures were performed before carrying out the experimental tests. The matching among designed and actual geometric parameters was the focus of the morphological analysis. Thus, a comparison between actual and designed cell diameters and face sizes was evaluated. The results are summarised in Table 8.

Table 8. Cell geometric parameters percentage variation between real and designed values G7_0.05 G7_0.1 G7_0.2 D designed [mm] 2.90 2.10 1.40 D real [mm] 2.89 2.06 1.42 D variation [%] -0.4 -1.9 1.4 A designed [mm] 1.70 1.14 0.65 A real [mm] 1.17 0.83 0.48 A variation [%] -31.18 -27.19 -26.15

The designed values of cell diameter are higher than the actual ones for all the configurations, with negligible percentage variation values observed for all the relative densities. Designed face size values are higher than real ones for all the configurations. Discrepancies were observed for all the relative densities and the percentage variation decrease while increasing the relative density. However, the real face sizes are greater than the minimum value of 600 µm necessary to fulfil the osteointegration process (Caliogna et al., 2020) for the specimens G7_0.05 and G7_0.1. 3.2. Compression tests results Compression tests results are reported in Fig. 5 in terms of stress-strain curves. Stress was obtained as the ratio between the axial load and the cross section of lattice specimens, whereas strain was evaluated from the crosshead displacement. The stress-strain curves resulting from compression tests are in accordance with data present in literature, following the same path of other microlattice structures (Al-Ketan et al., 2018). The first stage represents the initial linear elastic region, followed by a slight slope variation up to a maximum stress value, which was evaluated as the compressive strength σ c of the lattice structure. In the second stage, due to the failure of some unit cells within the lattice, a sudden load fall appears, followed by a negligible plateau region, with stress fluctuations. In the third stage, the lattice densification caused by the contact of collapsed cells leads to a stress increase. In Table 9 are reported the parameters obtained from the compression tests. Crush strength was evaluated as the mean stress in the second stage of the curve before the stress increase. The TEA was evaluated as the area under the load-displacement curve obtained during compression tests. In order to ensure results comparability, the TEA was calculated, for all specimens, up to a strain equal to 12%. SEA was calculated as the ratio between the TEA and the microlattice structure density.