Page 460 - IJB-9-4
P. 460

International Journal of Bioprinting                   β-Ti21S auxetic FGPs produced by laser powder bed fusion


            residual stresses . Furthermore, other manufacturing   and a volume energy density between 40 and 90 J/mm .
                                                                                                            3
                         [43]
            imperfections can occur during the printing process,   A 45° alternate scan strategy was used. Five samples for
            namely, the variation of the cross-section and the strut   each auxetic geometry were printed horizontally on the
            waviness, which modify the final mechanical response of   longer side to permit a better printability of the inclined
            the lattice structure .                            struts (Figure 2).
                           [44]
              Very few authors have investigated the manufacturability   A pre-alloyed plasma atomized  β-Ti21S alloy (GKN
            and mechanical properties of cellular structures made   Hoeganaes Corporation, Cinnaminson, NJ, USA) with a
            by  β-Ti. A  recent study evaluated the printability of   powder size distribution of 25 – 60 µm was selected. The
            simple cubic cells in β-Ti21S alloy underlying a suitable   chemical composition is shown in Table 2.
            manufacturing quality for strut thickness above 0.5 mm .
                                                        [19]
                                                               2.2. Metrological and material characterizations
              The aim of the present work is to evaluate the
            manufacturability of two different auxetic FGPSs with aspect   As-manufactured samples were characterized by means of
            ratio equal to 1.5 and angle θ of 15° and 25° with a relative   2D and 3D metrological characterizations. In detail, SEM
            density gradient of 0.34 – 0.49 – 0.66 and of 0.40 – 0.58 –   inspections of the lateral and top sample surfaces were
            0.75, respectively. 2D metrological characterization by   used to conduct the 2D dimensional analysis of the strut
            scanning electron microscopy (SEM) and 3D metrological   and pore size. The size of 10 pores and 10 struts for each
            characterization by X-ray micro-CT (µ-CT) imaging were   level of density were measured using a 2D image analysis
            carried out and compared. Preliminary investigation of the
            mechanical properties and comparison with analytical and   Table 1. Geometrical details of the designed auxetic FGPSs
            numerical homogenization analyses was also conducted.
                                                                θ (°)   ρ  CAD (‑)   Strut thickness    Pore size
                                                                         r
            2. Materials and experimental procedures                                  CAD (mm)       CAD (mm)
                                                               15         0.34         1.17±0.02     1.12±0.47
            2.1. Specimen design and preparation
                                                                          0.49         1.47±0.03     0.98±0.39
            Two different auxetic FGPSs were designed by means            0.66         1.78±0.10     0.78±0.32
            of nTopology software, and the geometrical details are   25   0.40         1.20±0.02     1.00±0.40
            summarized in  Figure  1. Each relative density level was
            characterized by a height of 3-unit cells and a solid base    0.58         1.51±0.06     0.81±0.31
            with a thickness of 5 mm was added at the bottom of the       0.75         1.80±0.24     0.63±0.27
            structure (Figure 1). The highest relative density level is   CAD: Computer-aided design, FGPSs: Functionally graded porous
            designed to improve osseointegration thanks to pore size   structures
            smaller than 800 µm, while the lowest density permits to
            decrease the elastic modulus close to that of the cancellous   Table 2. The chemical composition of β‑Ti21S (wt. %)
            bone.
                                                                Element  Mo    Al    Nb    Si   O    Ni    Fe
              In the auxetic FGPS with θ = 25° the highest density level   Weight %  14.6  2.8  2.8  0.3  0.11  0.004  Bal.
            becomes too dense leading to the loss of auxetic geometry
            (Figure 1C). All CAD parameters were characterized by
            means of 3D image analysis software (ORS-Dragonfly) and
            are summarized in Table 1.
              Strut thickness and pore size were calculated by means
            of the wall thickness analysis method which permits to
            obtain the size distribution of the analyzed 3D elements.
            This method evaluates the local thickness of the 3D
            object, namely, strut or pore, by fitting its volume with the
            maximum spheres at each location in the 3D structure [26,45] .
            According to this method, a high pores size deviation is
            expected, due to the small spheres fitting the pore near
            the corners. The different FGPSs were printed by means
            of a LPBF machine model MYSINT100 (SISMA SPA,      Figure 2. As-manufactured functionally graded porous structures lying
            Piovene Rocchette, Italy) on a platform of 100 mm in Ar   down on the longer side with schematical representation to highlight the
            atmosphere, with a laser spot of 55 µm, a power of 200W   building direction.


            Volume 9 Issue 4 (2023)                        452                          https://doi.org/10.18063/ijb.728
   455   456   457   458   459   460   461   462   463   464   465