Page 198 - IJB-9-4
P. 198

International Journal of Bioprinting                    b-Ti21S TPMS FGPs produced by laser powder bed fusion



            Table 2. Geometrical details of the as-designed TPMS-FGPSs.
             L        r  CAD     Level constant   Ligament thickness   Pore size CAD   Ligament thickness   Pore size CAD_max
                       r
             (mm)     (−)        (t) (−)     CAD (mm)  a     (mm)  a         CAD_max (mm)  b   (mm)  b
             2.5      0.17       −0.41       0.72 ± 0.01     1.56 ± 0.18     0.73              1.74
                      0.34       −0.20       1.00 ± 0.01     1.38 ± 0.06     1.01              1.44
                      0.50        0.00       1.25 ± 0.01     1.15 ± 0.05     1.27              1.20
                      0.66        0.20       1.48 ± 0.04     0.92 ± 0.03     1.52              0.95

                      0.83        0.40       1.66 ± 0.16     0.64 ± 0.02     1.82              0.66
             4.0      0.17       −0.66       1.39 ± 0.04     1.99 ± 0.39     1.43              2.38
                      0.34       −0.32       1.63 ± 0.08     1.90 ± 0.22     1.71              2.12
                      0.50        0.00       1.84 ± 0.12     1.74 ± 0.14     1.96              1.88
                      0.66        0.32       2.04 ± 0.16     1.54 ± 0.10     2.20              1.64
                      0.83        0.66       2.10 ± 0.37     1.30 ± 0.06     2.47              1.36
            a  Data expressed as median ± MAD
            b  Data expressed as median + MAD

            parameters are characterized by means of 3D image   MAD (red dashed line in  Figure 2) is computed. For
            analysis software (ORS-Dragonfly) and are summarized in   illustrative purposes, the ligament thickness distribution
            Table 2. Ligament thickness and pore size are calculated   and color mapping associated with the measurement in
            by means of the wall thickness analysis method which   case of higher relative density for TPMS 2.5 are shown in
            permits to obtain the size distribution of the analyzed 3D   Figure 3. Median value equal to 1.66 (blue in the mapping
            elements. This method evaluates the local thickness of the   figure) results in an underestimate of the thickness, since
            ligament or pore, by fitting its volume with the maximum   major parts of the ligaments are around 1.80 (orange-red
            spheres at each location in the 3D structure [57,58] . In view   parts). The sum of median and MAD of 1.82 results are
            of the non-normal distribution of the values and the   closer to the real values. Since a reference dimension is
            measurement frequencies shown in  Figure 2 (the values   essential for comparing the as-designed specimen with
            are reported only for TPMS-FGPS 2.5 because of similar   the as-manufactured one, sum of median and MAD,
            results in the case of TPMS-FGPS 4), median and median   namely CAD_max in Table 2, of the pore size and ligament
            absolute deviation (MAD), rather than mean and standard   thickness of the different relative densities is used. Pore
            deviation, are used to define the pore size and ligament   size of the TPMS-FGPS 4.0 designed to have same relative
            thickness. Nevertheless, considering the size distribution   density levels of TPMS-FGPS 2.5, results in a lower pore
            of both pores and ligaments shown (Figure 2), smaller   size of 1.36 mm (lower relative density level of 0.17) too
            values are present with a frequency lower than 5% and only   large  for  the  osseointegration  requirement  where  a  pore
            the highest values are detected with frequency in the range   size in the range of 100–600 mm is necessary to optimize
            of 5%–50%. Therefore, in the case of relative densities   bone regeneration [25,44-46] . Correlations between relative
            with high ligament thickness (r  = 0.83,  Figure  2a) and   densities  (r ), ligament thickness (l ), and pore size (p )
                                                                                            t
                                      r
                                                                                                            s
                                                                        r
            high pore size (r  = 0.17,  Figure 2b), the median value   with the level constant “t” are evidenced in Figure 4, and
                          r
            (black dashed line) is poorly representative of the largest   the corresponding third-degree polynomial fitting are
            population of the distribution. The median and MAD   summarized in Equations VI–VIII and IX–XI for TPMS-
            values reported in Table 2 highlight that the MAD values   FGPS 2.5 and TPMS-FGPS 4.0, respectively .
                                                                                                  [47]
            increase when the size of pore or ligament increases.
                                                                                                       2
                                                                                                3
                                                                                      -17 2
            Under these circumstances, the higher number of low   ρ  = 0.500 + 0.799t + 7.080E t  + 0.038t   Adj R  = 1
                                                                r2.5
            values with a frequency inferior to 5% affects the median                                     (VI)
            value of the distribution that becomes not representative   l  = 1.266 + 1.258t + 0.051t  + 0.424t   Adj R  = 0.99981
                                                                                                    2
                                                                                             3
                                                                                     2
            of the real pore size and/or ligament thickness. To avoid   t2.5                             (VII)
            this inconsistency and to obtain a more representative
            dimensional value for the pore size or ligament thickness   p  = 1.197 − 1.196t + 0.017t  − 0.719t   Adj R  = 0.99987
                                                                                                    2
                                                                                      2
                                                                                             3
                                                                s2.5
            for the different relative densities, sum of median and                                     (VIII)
            Volume 9 Issue 4 (2023)                        190                          https://doi.org/10.18063/ijb.729
   193   194   195   196   197   198   199   200   201   202   203