PSI - Issue 41

Saveria Spiller et al. / Procedia Structural Integrity 41 (2022) 158–174 Saveria Spiller/ Structural Integrity Procedia 00 (2019) 000–000

167

10

tensile test. Instead, the specimen with minimum strain at break (1.87%) reached 660 MPa. The relative porosity was respectively 2.8% and 4.5%, remarkably different values. This result is therefore coherent with the ones aforementioned, confirming the relation between relative density (and relative porosity) and mechanical properties. Several printing parameters are proven to be affecting the porosity and the density of the printed parts. One of the most relevant, even though controversial, is layer thickness. For example, Kurose et al. (2020) printed and sintered several specimens of 316L with layer thickness varying from 0.1 to 0.3 mm, changing the layout on the printing platform. Specifically, specimens with both layer thicknesses were printed flat, on the side, or vertically on the platform, as depicted in Fig. 9. a) printing layout and b) tensile test curves from Kurose et al. (2020). The influence of both parameters on the mechanical properties was studied. The higher density was found to be about 92.9%, belonging to the specimen printed on the side on the platform, with a layer thickness of 0.1 mm. Analyzing the results, it was observed that the relative density of the sintered parts is indeed mainly dependent on the layer thickness rather than the printing layout. The specimens underwent tensile tests, which clearly showed that for each orientation the higher strength was reached by a specimen with 0.1 mm layer thickness (Fig. 9. a) printing layout and b) tensile test curves from Kurose et al. (2020)b). The authors explained that by decreasing the layer height it was possible to reach a better packing effect of the material, resulting in more dense powder particles after the debinding. Despite these results, the opposite effect was observed in other studies. For instance, Hassan et al. (2021) printed rectangular 316L specimens with a layer height of 0.3, 0.4, and 0.5 mm. To optimize the process, the printing speed was also varied, and the rectangular bars were printed both on the side and flat on the building platform. The authors stated that the porosity is more significantly dependent on the printing speed than on the layer height. Nevertheless, the best flexural properties were shown by specimens printed with a higher layer thickness. The same result, in terms of layer thickness, was reported by Godec et al. (2020). The work has already been mentioned previously, but it is interesting to underline that several layer heights were explored in the attempt of optimizing the green part strength, ranging from 0.12 to 0.28 mm. In the end, the optimal configuration involved a greater layer height of 0.3 mm. The reason produced by the authors is related to layer adhesion. Since the layer interfaces are generally weak areas, where possible deposition defects condensate and where the material continuity is interrupted, it can be beneficial to increase the layer thickness. In addition, it was also proven that higher layer thickness is less challenging to print, preserving the external appearance of the part.

Fig. 9. a) printing layout and b) tensile test curves from Kurose et al. (2020)

The porosity is strongly affected by the raster deposition strategy. The deposition pattern is an issue both with polymers and composite materials when dealing with FDM. Damon et al. (2019) tested different infill strategies on 316L specimens printed vertically and horizontally. The infill was rectilinear with diagonal, horizontal, and longitudinal rasters. Computer tomography was used to study the porosity inside the specimens, some of the photographs are shown in Fig. 10. CT scan pictures of the internal porosity of specimens printed with different raster angles. 0°, 45° and 90° (Damon et al., 2019). As expected, pores distribution, size, and morphology are dependent on the raster angle. Performing tensile tests, it was observed a major dependency of the mechanical properties from the porosity. The building orientation was proven to be less significant on the porosity of the part, as explained in the next paragraph. A UTS value above 500 MPa was obtained for the specimen printed flat-wise on the bed, with an