PSI - Issue 42

Saveria Spiller et al. / Procedia Structural Integrity 42 (2022) 1239–1248 Saveria Spiller/ Structural Integrity Procedia 00 (2019) 000 – 000

1242

4

So far, most of the publications on MEAM are focused on the investigation of the building direction influence on the tensile properties of the specimens. Specimens are usually printed flat, on the edge, or upright on the platform, as shown in Fig. 2c. As expected, it was shown that if the loading direction is parallel to the building direction, the strength of the specimens is lower (Alkindi et al., 2021; Caminero et al., 2021; Henry et al., 2021; Kurose et al., 2020; Suwanpreecha et al., 2021). In the preparation of the G-code that the printer reads, several additional instructions must be provided, such as the nozzle temperature, the platform temperature, the flow rate multiplier, the cooling fan speed, and the print speed. The correct setting of these parameters is more challenging when using a high-filled polymer than a regular polymer such as PLA or ABS since they are more viscous and they have a different heat transmission capability. More details on the printing parameters can be found in (Gonzalez-Gutierrez et al., 2018; Loh et al., 2020; Spiller et al., 2022) 2.2. Debinding and sintering After the shaping phase, the green parts undergo debinding and sintering, common metallurgical processes. For example, MIM, and BJ entail the same procedure after the fabrication of the green parts. The debinding can be done in different ways depending mainly on the composition of the polymeric share in the base material. A thermal debinding can be sufficient (Sadaf et al. 2021), but it is more common to have a pre-step of solvent debinding to remove the main part of the binder, which is usually composed of waxes. A third possibility is the catalytic debinding, faster but more demanding in terms of equipment needed (Gonzalez-Gutierrez et al., 2018; Nurhudan et al., 2021). Once the binder is completely removed, the specimens are sintered. For stainless steel powder, the sintering cycle can reach 1360°C or more for a prolonged time, depending also on the powder particle size (Ramazani et al., 2022). Sintering is usually performed under a controlled atmosphere such as inert gas or reducing gas to prevent the formation of oxides and the consequences on the mechanical integrity and microstructural purity of the parts (German, 2010a; German, 2010b). As aforementioned, residual porosity is an issue in MEAMed parts, and it is highly dependent on the sintering procedure. The pores originate both in the printing phase and during the post-process. In the first case, the porosity is mainly dependent on the printing strategy, while in the second case it is due to a lack of densification during the sintering process. Higher sintering temperatures or longer times are helpful, even though these lead to another consequence, the enlargement of the grains (German, 2010b; Gong et al., 2019). It is possible to promote the sintering using metal powder with smaller average dimensions and regular shapes (German, 2010a; 2010b). During the sintering process, the specimens experience a volumetric reduction due to the agglomeration of the powder particles. Furthermore, the densification process entails the reduction of the voids left after the debinding, corresponding to areas with high polymer concentration. The result is that the parts show a linear shrinkage, proven to be higher along the Z-axes of the building plate (Gong et al., 2019; Kurose et al., 2020). Shrinkage has to be compensated in the design of the specimens. 3. Preliminary fabrication trials The purpose of the present work is to fabricate metal components using a commercial FDM printer and a polymeric filament filled with 88wt% of 316L SS powder. To preliminary explore the potentiality of the printing process, simple specimens were printed, starting with small 1×1×1cm 3 hollow cubes (see Fig. 3a). It was possible to restrict the list of the more influencing parameters to the following: nozzle temperature, layer thickness, print speed, cooling fan speed, and flow rate multiplier. Throughout the experimental campaign, a hardened steel nozzle with a 0.4 mm diameter was used since the standard brass nozzles are more sensitive to the abrasion provoked by the metal powder. Once a good appearance of the walls was achieved, the infill of the cubes was printed too (see Fig. 3b), with the following characteristics: rectilinear pattern, 100% density, and +/-45° raster angle. Some of the typical defects observed are shown in Fig.3: about the hollow cubes, the main problem was the adhesion of the layers, and it was mainly affected by the nozzle temperature (Fig. 3a1; 3a2). Once the infill was added, new problems appeared such as warping and other evident deformations, especially at the base of the solid cubes (Fig. 3b1; 3b2). In this case, it was necessary to print a brim to increase the adhesion to the platform. Being