PSI - Issue 23

Mattias Calmunger et al. / Procedia Structural Integrity 23 (2019) 215–220 M. Calmunger et al. / Structural Integrity Procedia 00 (2019) 000–000

216

2

of AM components. The PBF-L technology has already been successfully applied in commercial combustor compo nents for industrial gas turbines, where it has been used for repairing of burners to simplify the repair and decrease costs compared to conventional techniques, see Andersson et al. (2017). With the use of AM, the performance of such components can also be optimized by allowing for more design freedom with, for example, more complex internal cooling channels. Combustor components in a gas turbine are exposed to high temperatures and large mechanical loads. To withstand the loading conditions, materials with satisfying properties for this working environment have to be used. Combustor components are therefore advantageously manufactured by ductile nickel-based superalloys, due to their oxidation resistance and strength at high temperature. In addition, due to more cycle-driven running profiles of industrial gas turbines, the components will be exposed to more fatigue loading, which has to be considered during the design. However, the fatigue performance of AM gas turbine materials are not yet completely known. This study focuses on the fracture appearance on specimens tested in low-cycle fatigue (LCF) at room temperature. Especially, the dendritic structures present in the microstructure, and how it influences the fracture, was studied. Cylindrical specimens were prepared from an AM adapted nickel-based superalloy Hastelloy X. The specimens where manufactured by PBF-L in two di ff erent building directions: 0 ◦ and 90 ◦ , where the angles refer to the angle between the longitudinal axis of the specimen and the building platform. The specimens were built in an EOS M290 machine; the thickness of the build layers were 40 µ m. Prior to testing, the specimens were machined to remove the influence of the surface roughness, no additional treatment was performed before testing. LCF testing was performed in strain control on 20 specimens (as reported by Lindstro¨m et al. (2018)); here, only a subset of the specimens was used as given in Tab. 1. The specimens where run either until rupture or until 60 % load drop and pulled to fracture. Table 1. Testing parameters for the specimens used in the current work; R ε = ε min /ε max , where ε min and ε max are the minimum and maximum total strains respectively. Building direction Temperature, ◦ C R ε ∆ ε , % 0 ◦ Room temperature 0 1.3 90 ◦ Room temperature 0 1.3 The fracture surfaces from the specimens tested as described in Tab. 1 were studied in a scanning electron micro scope (SEM). After the crack initiation point had been identified in stereomicroscopy, the fracture surfaces were cut through the initiation point and polished, as illustrated in Fig. 1. This allowed for a two-step investigation; i) inter esting features were identified on the fracture surface (view II in Fig. 1) and ii) the specimen was rotated so that the adjacent, underlying microstructure could be studied (view III in Fig. 1). In addition, a piece of the specimen far away from the fracture surface was cut and polished as outlined in Fig. 1 (view I). The crack initiation point was marked to be able to correlate the building direction of the 0 ◦ specimen to the crack propagation direction. The scanning electron microscopy (SEM) techniques electron channelling contrast imaging (ECCI) and electron backscatter di ff raction (EBSD) were used to characterize the deformed microstructure. For the ECCI and EBSD investigations, a Hitachi SU-70 field emission SEM were used. ECCI was performed at 10 kV acceleration voltages and a working distance of 7 mm, using a solid state 4-quadrant backscatter detector. EBSD was performed using a Oxford electron backscatter di ff raction detector. EBSD maps were measured at 15 kV acceleration voltage and a working distance of 25 mm and the EBSD-maps were produced using a step size of 0.1 µ m. The software CHANNEL 5 was used for the microstructure evaluation. After EBSD mapping, the sample was electro-etched (10 vol.% oxalic acid and 10 V), which revealed the melt pool boundaries using a light optical microscope (LOM). Prior to etching it was necessary to briefly polish the specimen with 0.04 µ m colloidal silica suspension, since areas contaminated during EBSD would otherwise not etch. After etching, a series of SEM images were taken in the same region as studied in EBSD, allowing the melt pool boundaries to be overlayed on the EBSD map. 2. Experiments