Page 436 - IJB-10-3
P. 436

International Journal of Bioprinting                                  Different modeling of porous scaffolds






































            Figure 1. (A) Filling structure in the spine. (B) Thickening structure in the leg bone. (C) Minimal surface units with thickening and filling. (D) Resulting
            units under the two modeling strategies. Abbreviations: D, Diamond; F, Fill; G, Gyroid; IW-P, I-graph-wrapped package; P, Primitive; T, Thicken.


            mechanical performance while retaining their continuous   porous scaffolds. For the filled structure, the ContourPlot3D
            transition advantages.                             command in Mathematica software was employed, where
               Hence, four commonly used minimal surfaces—     the formula was input, and the offset position of the surface
            Primitive (P), Gyroid (G), Diamond (D), and I-graph-  was adjusted by controlling the parameter C to achieve a
            wrapped package (IW-P)—were selected as the basic units   filled porous scaffold with a porosity of 60%. As for the
            of minimal porous scaffolds in  this study. Maintaining   thickened structure, the surface is imported into Wrap
            consistent unit size, surface offset, and porosity, two   software for surface thickening to obtain the thickened
            modeling strategies, filling and thickening, were employed   porous scaffolds (as shown in Figure 2). The samples were
            on these surfaces (see Figure 1C) to create porous scaffold   fabricated using a selective laser melting device (Dimetal
            units (see  Figure 1D). Scanning electron microscopy,   280, Tridonic Additive Technology Co., Ltd., Canton,
            micro-computed tomography  (micro-CT),  and other   China) with a powder composition of Ti6Al4V (TC4)
            equipment were used to observe the macro/micro     alloy. The printing parameters were as follows: laser power
            morphologies of the additive-manufactured porous TC4   of 175 W, powder layer thickness of 30 μm, laser spot size
            alloy scaffolds. Compression tests were carried out to   of 180 μm, and scanning speed of 1000 mm/s. The scan
            investigate the deformation mechanism and mechanical   direction was rotated by 67° between consecutive layers.
            performance variations of the scaffolds. The effect of filling   After printing was completed, the scaffolds were cut from
            and thickening strategies on the permeability of the scaffold   the substrate and cleaned in ultrasonic baths with ethanol
            was investigated by using the descending head method and   for further use.
            finite element analysis.
                                                               2.2. Macro- and microscopic characterization
            2. Scaffold fabrication and                        of scaffolds
            performance characterization                       The scaffold porosity was determined using the drainage
                                                               method,  with the calculation formula presented in
                                                                      28
            2.1. Scaffold design and fabrication               Equation I. In the equation,  P represents the sample’s
            The mathematical functions from Table 1 were imported   porosity (%), m  is the mass (g) of the sample immersed
                                                                            w
            into Mathematica software to generate surfaces and fill   in pure water, m  is the mass (g) of the scaffold, V  is the
                                                                            d                           d
            Volume 10 Issue 3 (2024)                       428                                doi: 10.36922/ijb.2565
   431   432   433   434   435   436   437   438   439   440   441