PSI - Issue 53

S. Leonardi et al. / Procedia Structural Integrity 53 (2024) 327–337

331

S. Leonardi et al. / Structural Integrity Procedia 00 (2023) 000–000

5

al. (2017); Du Plessis et al. (2020). Unlike periodic lattices, the architected metallic cellular materials of this work contain heterogeneous (i.e. disordered) pore features and thus prove attractive candidates for the investigation of the LPBF process-structure-property relationships. Since the main focus of this work is on topological defects, the key laser and scanning strategy parameters (notably the laser power P , laser speed v and the hatch distance h , i.e. the o ff set between two laser tracks) were kept constant for each powder employed. The latter are reported in Table 3 for each metallic powder, and were optimized to increase the metal density during a dedicated experimental campaign conducted prior to this study. Table 3 also reports, for each metallic powder, the type of used strategy and the number of contours used to 3D print the porous test samples. Alongside a di ff erent scanning strategy (incremental and stripe for the AlSi10Mg and the Inconel 625 parts respectively), the number of contours was varied between the two metallic powders. The latter was found to have a great impact on the 3D-printed pore topologies, as it will be shown in Section 3. It is also noted that the values of the laser power and scanning speed used for countours were di ff erent from those of the volume reported in Table 3. Notably, these were (P = 100W, v = 450mm / s) and (P = 300W, v = 730mm / s) for the Inconel 625 and the AlSi10Mg powders respectively.

Table 3. Main LPBF processing parameters for each powder employed. Laser Power P [W] Laser speed v [mm / sec]

Hatch distance h [ µ m]

Scanning strategy

No. of contours

Inconel 625 AlSi10Mg

200 300

900

120 130

stripe

2 1

1650

67° incremental

A total of six test samples were manufactured for this study, see Table 4. The following designation is used to denote the test sample: in sample M-X , M denotes the metallic powder (Al for AlSi10Mg, In for Inconel 625) and X the porosity content expressed in %. Each value of porosity corresponds to one of the porous architectures reported in Figure 2 and Table 1. As Table 4 shows, the random architectures at 30 and 40 % porosity were produced using both metallic powders, thereby allowing a pairwise comparison of the resulting topological pore features (Section 3).

Table 4. Summary of the porous test samples manufactured by LPBF for this study. Test sample designation Metallic powder

Porosity [%]

Numerical model

Al-20 Al-30 Al-40 In-30 In-40 In-60

20 30 40 30 40 60

N-20 N-30 N-40 N-30 N-40 N-60

AlSi10Mg

Inconel 625

2.3. Topology assessment by image post-process

This work aims at quantifying the topological defects induced by the LPBF process during fabrication of the random cellular architectures in Figure 2. In agreement with prior work, e.g. Liu et al. (2017), topological defects are hereinafter defined as geometrical mismatches between the as-defined defect-free numerical sample and the as manufactured imperfect test sample. More specifically, in this work we focus on two types of topological defects and namely: • the pore ellipticity : this is quantified by means of the pore aspect ratio. The latter is defined as the ratio of minor axis to the major axis. This parameter is equal to 1 for circular pores; • the pore oversizing / undersizing : this is quantified by means of the di ff erence between the actual, i.e. as − built , pore radius and its nominal value (i.e. as designed).