Page 505 - IJB-10-1
P. 505

International Journal of Bioprinting                          Macro and micro structure of a 3D-printed implant









































            Figure 3. (a) Load and boundary conditions for wedge-shaped fixation; (b) stress distribution for wedge-shaped fixation; (c) topology optimization of
            wedge-shaped structure; and (d) and loading and boundary conditions for bone plate fixation.

            analysis software for tetrahedral mesh generation using the   with diameters ranging between 15 μm and 45 μm).  The
                                                                                                         6
            free mesh approach (Figure 2).                     3D printing process involved utilizing a laser power of
                                                               200 W, a scanning rate of 0.6 m/s, and an exposure time
               Under the same load, boundary conditions, and
            material property settings in the previous optimization   of 125 s. By selectively scanning and melting the powder
                                                               with the laser, the component was fabricated once the
            simulation,  three fixation models were  simulated for   powder crystallized. The manufacturing accuracy and
            OWHTO: (i) solid wedge-shaped spacer (SWS), (ii)   layer thickness in this study were both set at 30 μm.
            optimal wedge-shaped spacer (OWS), and (iii) T plate (TP).
            The SWS and OWS analyses aimed to validate whether the   Upon printing process completion, the OWS underwent
            WS topology optimization could achieve the same strength   deburring and polishing using a magnetic polisher with
            as the SWS and compare the mechanical stability with the   stainless steel pins (Ø = 1 mm, L = 3 mm) at a speed of 2700
            commercially available traditional T plate fixation system.   rpm. The OWS was subsequently cleaned using ultrasonic
                                                                                      6
            The comparison criteria included overall displacement,   oscillations (Figure 5a and b).  Our 3D printing laboratory
            maximum equivalent stress in the plate, maximum    is certified by the ISO 13485 quality management system
            equivalent stress in the fixation screw, and maximum first   (Certificate Number: 1760.190828), ensuring compliance
            principal stress in the bone. For computational efficiency,   with necessary regulations and guaranteeing that safety
            the screw threads were simplified as cylindrical shapes   and quality standards are met for the implants. Following
            in all analyses because the focus was on comparing the   the  3D  printing  procedure,  the  OWS  underwent  acid
            HTO global mechanical responses under different fixation   etching to remove any residual sandblast particles, followed
                                                                                                          6
            systems.                                           by additional cleaning using ultrasonic oscillations. The
                                                               OWS in this study was designed using CAD software.
            2.4. OWS 3D printing                               The detailed dimensions of all OWS features in the CAD
            The OWS filled with YM lattices was manufactured using   software were defined as the actual dimensions. The wedge
            a metal 3D printer (AM250, Renishaw, Gloucestershire,   height (H), wedge angle (A), anterior-posterior length
            UK) with titanium alloy powder (Ti6Al4V ELI powder   (L), and screw hole diameters (S1, S2, and S3) the OWS


            Volume 10 Issue 1 (2024)                       497                          https://doi.org/10.36922/ijb.1584
   500   501   502   503   504   505   506   507   508   509   510