PSI - Issue 53

S. Leonardi et al. / Procedia Structural Integrity 53 (2024) 327–337 S. Leonardi et al. / Structural Integrity Procedia 00 (2023) 000–000

330

4

the number of pores (i.e. 200 voids) is here motivated by a two-fold reason. On one hand, a high number of pores is necessary in order to ensure the representativity of the volume element (see Hooshmand-Ahoor et al. (2022); Zerhouni et al. (2019); Tarantino et al. (2019)). On the other hand, increasing this value at a fixed target porosity results into pores with both smaller diameter and interpore distance. The latter prove in turn di ffi cult to manufacture with high geometrical accuracy by LPBF process, as it will be shown hereinafter.

Table 1. Topological features of the porous architectures reported in Figure 2 and investigated in this study Model designation Porosity [%] Total no. of pores No. of internal pores Internal porosity* [%]

Porosity type

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

20 30 40 60

200 200 200 200

185 182 181 179

16.9 27.5 35.8 51.1

Equisized Equisized Equisized

Polydispersed

* The internal porosity is calculated as the ratio of the surface of the internal pores to the whole area of the squared cell.

2.2. LPBF manufacturing

The random porous materials of this work were fabricated by means of LPBF additive manufacturing technology. This process, also known as Selective Laser Melting consists in building a 3D metallic component in a layer-by layer fashion, where each layer of fine metal powder is selectively melted using a high-power laser (Herzog et al. (2016)). This additive manufacturing process is today used with a variety of metallic powders to build 3D parts of any geometrical complexity (DebRoy et al. (2018); Herzog et al. (2016); Zheng et al. (2017)). In this work, the test samples were additively manufactured using a SLM HL 125 machine (SLM Solutions GmbH) equipped with a Laser Fiber IPG Yb:YAG 500W 1070nm, with spot size of 73 µ m. The printer has a build plate with plane dimensions 123 × 123mm and was operated under controlled atmosphere during the entire production process. Porous test samples of dimensions 12 . 5 × 12 . 5 × 10mm 3 were produced for this study using two di ff erent metallic gas-atomized commercial powders, namely a Nickel- and an Aluminum-alloy powders. The former goes under the commercial name Inconel 625 and was purchased from Oerlikon, the latter has trade name AlSi 10 Mg and was purchased from TEKNA. Their particle granulometry is similar and reported values from the data sheet provided by the manufacturers are respectively D10 = 21 µ m, D50 = 35 µ m, D90 = 58 µ m for Inconel 625 (De Terris et al. (2021)) and D10 = 27 µ m, D50 = 41 µ m, D90 = 560 µ m for AlSi10Mg. For the sake of clarity, D10, D50, and D90 denote respectively the particle size below which 10 %, 50 % and 9 % of the powder material falls. The powder chemical composition is reported in Table 2.

Table 2. Chemical composition of the two metallic powders used to fabricate the test samples, namely Inconel 625 and AlSi10Mg.

Inconel 625

Al

Ti

Si

S

P

O Nb + Ta N Mo Mn Fe

Cu Cr

Co C Ni

0.12

0.11 0.02 0.001

< 0.001 0.02 3.6

0.01 8.9

< 0.01 0.3 0.01 21.2

< 0.01 0.03 Balance

AlSi10Mg

Si

Fe Cu Mn Mg

Ni

Zn

Pb

Sn

Ti

C H N O OtherAl

9.0-11.0

< 0.5

< 0.03

< 0.4 0.2-0.45

< 0.05

< 0.1

< 0.05

< 0.05

< 0.15

< 0.05

< 0.01

< 0.01

< 0.1

< 0.15 Balance

Prior work shows that the quality of the 3D-printed parts - both in terms of geometrical accuracy and densification of the metal matrix - depend on numerous parameters, and notably on the environment (e.g. gaseous atmosphere), the powder granulometry as well as the laser and scanning strategy. A systematic investigation of these parameters on the quality of homogeneous parts manufactured by LPBF process is reported, e.g., in De Terris et al. (2021); Traore et al. (2022) for Inconel 625 and in Zhao et al. (2022); Bagherifard et al. (2018) for AlSi10Mg powders. The role of the LPBF process parameters is exacerbated in cellular structures due to the complex internal geometry, see Liu et