Page 130 - MSAM-2-3
P. 130

Materials Science in Additive Manufacturing                       Powder alteration caused by L-PBF process


























            Figure 9. D-values for line i=1.                   Figure 10. Location of powder samples collected along horizontal X-axis
                                                               and vertical Y-axis.
            are dragged further away. This result is also reported by
            Carrion et al.  Moreover, Pal et al.  reported that the
                                          [28]
                       [25]
            bulk density of the powder bed drops near the collector bin.
            It was also observed that near the dispenser, the D05, D50,
            and D10 values were slightly lower than those measured
            near the collector bin, whereas D90 and D95 showed
            an increase of 1.2 µm and 1.5 µm, respectively, near the
            dispenser . However, the decrease in particle diameters
                   [23]
            along the columns remains unjustified and is not discussed
            in the literature.
            3.2. Results of experiment 2: Evolution of PSD
            depending on position relative to a printed part
            3.2.1. Results of single contamination
            Figure 10 shows the locations of 6 powder samples collected
            along the horizontal X-axis, and 6  samples collected   Figure 11. Cumulative particle-size distribution horizontal line.
            along the vertical Y-axis. Figures 11 and 12 show that the
            cumulative  PSD  and  the  PSD  of  samples  positioned in   Table 3. Diameter D‑values variation along lines (toward
            negative abscissas x , x , x  (near the collector bin) are   right) and columns (toward bottom)
                                  −3
                               −2
                            −1
            greater than those located in the positive abscissas x , x , x
                                                          3
                                                       2
                                                     1
            (near the dispenser), indicating the same result as the first   Position  Δ D10 (µm)  Δ D50 (µm)  Δ D90 (µm)
            experiment, which demonstrated that the PSD in the   Line   i=1      -9.0%     -9.8%      -9.1%
            graph is displaced progressively left toward fine particles   i=2   -12.1%    -10.8%      -8.0%
            as the sample location moves rightward on the build         i=3      -6.3%     -9.6%     -10.9%
            plate. In addition, the diameter values D90, D50, and D10
            decrease by 5.1%, 12.1%, and 18%, respectively, toward      i=4     -14.1%    -14.6%     -12.2%
            the positive abscissas (Figure 13). Yet, the PSDs of samples   Column  j=1  -4.9%  -4.5%  -3.5%
            taken on the Y-axis are superposed, and their diameter      j=2      0.8%      1.6%       5.2%
            values remain approximately stable (Figure  14). As also    j=3      -5.6%     -4.8%      -3.9%
            shown in Figure 10, the powder bed area located below the   j=4     -10.0%     -9.5%      -6.9%
            printed cylinder shows a substantial amount of fine black
            particles as compared to the top area. The black particles
            are most likely spatters generated during printing of the   their impact on powder size along the Y-axis is negligible.
            solid cylinder and carried by the inert gas flowing from   Figure  10 also shows that spatters are not visible in the
            the top of the build plate toward the bottom . However,   powder bed along the X-axis. Changes to the PSD along
                                                [14]

            Volume 2 Issue 3 (2023)                         7                       https://doi.org/10.36922/msam.1781
   125   126   127   128   129   130   131   132   133   134   135