PSI - Issue 34

T. Silva et al. / Procedia Structural Integrity 34 (2021) 45–50

47 3

Author name / Structural Integrity Procedia 00 (2019) 000–000

Fig. 2. Metallographic samples of chemically etched (nital 2%) AMed maraging steel, in perpendicular-to-build direction (a) and CMed maraging steel (b); electrolytically etched (oxalic acid at 6V for 50s) maraging steel in perpendicular-to-build (c) and parallel-to-build (d) directions.

(CMed) material (cast, vacuum remelted) was equally considered, and identical specimens were machined. Chemical analysis of both AMed and CMed steels was performed using spark emission spectroscopy (as shown in table 1) and its compliance with the standard was verified. Samples polishing and micrograph analysis enabled relative density cal culation (99.7 %) and through the application of electrolytical etching the observation of macroscopic AM structures (such as melt pool and laser trace) was possible. No major defects were observed and the processing conditions were confirmed (refer to figure 2). The chemical etching reveals the prevailing plate-like martensite in the AMed sample when compared with the CMed, which may contribute to a reduced ductility and micro crack formation tendency of the former (Yokota and Lai, 1974). From the rectangular prismatic geometries specimens were manufactured towards thermal properties identification. Thermal expansion was determined using a Linesis DIL L75PT push-rod dilatometer whereas laser flash analysis (us ing Linesis LFA 1000) was employed for determining thermal conductivity and specific heat. The fully detailed exper imental procedure is shown in Silva (2021). Compression tests were performed in perpendicular- and parallel-to-build direction. For that, ø4x4 mm cylinders were machined from both cylindrical and small square prismatic specimens. These specimens were submitted to compression in distinct strain-rate conditions (from quasi-static to 6000 s − 1 ). Tensile specimens with smooth and notched geometry were obtained from the large square prismatic geometries, al lowing for fracture strain estimation under distinct (positive) stress triaxialities. In addition, the hat-shaped geometries resulted in double-notched specimens, intended for multiaxial compression tests. These are subdivided into 3 types of specimen, di ff ering in their notch configuration, resulting in a mixed state of (i) compression / shear, (ii) shear and (iii) tension / shear, when submitted to a simple compressive load. As explained by Silva et al. (2021a) and Silva et al. (2021b) both tensile and double notched specimens, in conjunction with the quasi-static compression specimens en abled the characterization of the maraging steel in di ff erentiated loading conditions as well as a wide range of stress triaxiality and lode angle parameter, allowing for state-of-stress sensitive modelling, which was conducted using a numerical / experimental approach.

3. Results and discussion

3.1. Physical testing

Figures 3a and 3b display the dilatometric curves for each metalurgical condition of the 18Ni300 maraging steel. The non-uniform expansion reveals the occurrence of phase changes, which according to Carvalho et al. (2013) can be perceived by the material contractions, highlighted in figures 3c and 3d. The first contraction corresponds to the precipitation of intermetallic compounds. Higher precipitation seems to occur in the conventionally manufactured maraging steel, which might be related with the (marginally) higher Ti and Co content (Dossett and Totten, 2014). Precipitation is followed by another (larger) contraction, which is related with the start of austenite formation (marten site reversion), and happens in two distinct phases (Kro´ l et al., 2020). During cooling, the large material expansion that starts around 200 ◦ C (refer to figures 3a and 3b) sets the start of austenite to martensite transformation. Whereas in the CMed material, precipitation and 1st step martensite reversion are markedly separate, these two e ff ects seem to take place almost simultaneously in the AMed maraging steel. Also important to note is the magnitude of both 1st and 2nd martensite reversion stages, which is higher for the CMed maraging steel. This seems to be in agreement with the