Page 205 - IJB-10-4
P. 205

International Journal of Bioprinting                                   Horsetail-inspired lattice for bone use



































                    Figure 8. Graphical map of displacement (A through C) and stress (D through F) of 070r040t, 115r025t, and 175r040t samples.
            implies that the mode of deformation is more toward   Equation II, while that of the flange length is to a power
            bending. As  AR  decreases, the bending of the strut,   of three.
            compared with the element displacement at the extreme                 tr (  t − ) 3
                                                                                               2
            edge of the strut, is lower. Additionally, the displacement      I flange  =  inner  + mt    (XI)
                                                                                     12
            of the elements becomes aligned with the direction of
            compression and more non-symmetrical (in (001) plane)
            as  AR  decreases. The greater degree of strut bending   where r inner  =    and m is the flange mass, with r >>t.
            with increasing AR (e.g., r = 1.75 to 0.7) agrees with the   Notwithstanding the preceding observations, we
            conclusion of Deshpande et al. where bending-dominated   also observed that the flanged RVEs have approximately
            cellular architecture results in lower  E lattice 39  as evident   40% more surface area compared to that of the flangeless
            from Figure 7. With lower AR, the strut construction is   design. The additional surface area would be ideal for
            also less noticeable and, interestingly, exhibits behavior   bone scaffold design-related application since the surfaces
            closer to a parameter frame construction. It is apparent   would be required for cell adhesion and growth.
            that the lattices with lower AR exhibit a more pronounced   Figure 10 shows the stress distribution between the
            stretching response. The ability to enhance an inherently   flanged and flangeless design, both for big and small radii.
            less stiff BCC lattice to increase stiffness assists in latitude   From Figure 10, we are also able to observe that there are
            of mechanical performance tuning.                  comparatively lower numbers of spots on the RVEs with
            3.3. Performance comparison between flanged and    peak stresses for the flanged design than for the flangeless
            flangeless RVEs                                    design. Additionally, the respective peak stresses are
            We determined that t = 0.1625 for a flanged design would   also lower for the flanged design when compared to the
                                                               flangeless design. This can potentially be attributed to the
            result in a φ that is similar to that of t = 0.25 mm for a   presence of flanges as load transfer members and leads to
            flangeless design, with deviation of less than 1% in φ. The   better distribution of stresses.
            mechanical and geometrical performance of the two RVE
            sets is shown in Figure 9. We observed that the E lattice  is   3.4. Compression testing characteristics of
            greater for flangeless design as opposed to flanged, and this   lattice matrices
            observation is more apparent at bigger r. This observation   The stress–strain characteristics of the samples are
            is expected since, when comparing the simplified second   illustrated in Figure 11. Additionally, the elastic modulus of
            moment of area expression in Equation XI of that of  r   the matrices, E matrix , and the upper limit of the densification
            the flange, the effect of is power of four, as indicated in   strain, ε  and specific energy absorption are computed as
                                                                     d

            Volume 10 Issue 4 (2024)                       197                                doi: 10.36922/ijb.2326
   200   201   202   203   204   205   206   207   208   209   210