Page 441 - IJB-10-3
P. 441

International Journal of Bioprinting                                  Different modeling of porous scaffolds






































            Figure 6. Stress–strain curves for different structures. (A) P-T and P-F scaffolds. (B) G-T and G-F scaffolds. (C) D-T and D-F scaffolds. (D) IW-P-T and IW-P-F
            scaffolds and three stages of the compression process. Abbreviations: D, Diamond; F, Fill; G, Gyroid; IW-P, I-graph-wrapped package; P, Primitive; T, Thicken.


            However, the P structure exhibits different results from the   to scaffold fracture along the diagonal.  The G, D, and IW-
                                                                                             46
            other three structures, as it is a more concentrated material   P-filled scaffolds correspond to previous research results,
            with larger pillar sizes, enabling it to bear stress more   with  failure  occurring  along  the  diagonal  shear  band.
                                                                                                            47
            effectively in the vertical direction. 43          It is worth noting that the fracture band of the D minimal
               Figure 7 shows the deformation behavior of the scaffolds   surface-thickened scaffold is almost perpendicular to the
            at 5%, 10%, and 30% strain during the compression process.   loading direction of the scaffold. After compression, the
            All scaffolds fracture through the formation of extrusion   scaffold shows a barrel-shaped contour, and this expansion
            bands, but the position and angle of fracture faces vary   increases the lateral resistance of the structure, enhancing
            for different units and modeling strategies. The results are   its load-bearing capacity in the later stages of compression.
            consistent  with  the  team’s  earlier  research, 44,45   indicating   Overall, the  surface  thickening method  reduces  the
            that the fracture in scaffolds is most likely to occur at   distance between units in the scaffold, improves the overall
            stress concentration points. In the P minimal surface-  load uniformity, and enhances the deformability of the
            filled scaffold, the pillar sizes at the unit connections are   entire sample during the compression process. This is
            the smallest and oriented horizontally. Thus, the maximum   reflected in the significant reduction in stress fluctuations
            stress is in the horizontal direction, and the fracture face   in the stress–strain curve for the thickened  scaffold.
            is also horizontal. When the pressure exceeds the load-  Compared to the filled scaffold, the thickened scaffold can
            bearing capacity, some pillar structures may fail, leading   more effectively distribute the compression load, avoiding
            to collapse and fragmentation, and forming a serrated   sudden and severe failure.
            platform and pillar fracture.                         Table 4 shows the mechanical performance of the

               The thickening strategy has an impact on the fracture   porous  scaffolds.  It  can  be  observed that  the  ultimate
            faces of the scaffolds. The P minimal surface-thickened   strength of the scaffolds varies between 135.78 and 250.90
            scaffold exhibits a different pattern, with shear band failure   MPa, and the modulus ranges from 3.03 to 4.57 GPa. The
            along the diagonal. The possible reason is that surface   compressive properties of human bone are influenced by
            thickening disperses the material, reducing its resistance   age and location, with Young’s modulus and yield strength
            in the vertical direction, and internal defects can also lead   in the range of 0.3–30 GPa and 33–193 MPa, respectively.
                                                                                                            48

            Volume 10 Issue 3 (2024)                       433                                doi: 10.36922/ijb.2565
   436   437   438   439   440   441   442   443   444   445   446