PSI - Issue 66

Johannes L. Otto et al. / Procedia Structural Integrity 66 (2024) 256–264 Johannes L. Otto et al. / Structural Integrity Procedia 00 (2025) 000–000

263

8

acidification and hydrogen embrittlement, particularly in the case of martensite formation. The combined effect of these factors resulted in the formation of many small cracks. Considering that the EGC has a chloride ion concentration of 1000 ppm, it can be seen from Truman's work (1977) that for NaCl solutions, temperatures of 80°C and a pH of 3.5, the threshold for stress corrosion cracking for AISI 304 is just exceeded.

Fig. 8. Failed specimen at σ max = 400 MPa after N f = 7.7∙1 0

4 in synthetic exhaust gas condensate at 80°C and 0.5 bar with multiple crack initiation.

4. Conclusion and outlook In this study, the corrosion fatigue behaviour of diffusion-brazed butt joints using a nickel-based filler metal in various corrosive and non-corrosive environments was characterized. The mechanical properties of the brazed joint were high overall, with the ultimate tensile strength of the brazed specimens similar to that of the base material after comparable heat treatment, Otto et al. (2021). However, ductility was increased, leading to cyclic creep at higher amplitudes during the stress-controlled fatigue test. Due to the good corrosion resistance of the filler metal used, resulting from the adjusted Cr and Mo content, there was no extensive surface corrosion attack in the brazing seam. Nevertheless, all corrosive environments that contained chloride ions showed a significant decrease in fatigue life. However, it seems that they mainly influenced the crack initiation and the threshold for crack propagation but not the crack propagation, since the slope of the SN-curves did not depend on the test duration, which is usually observed in corrosion fatigue. The decrease was attributed to a combination of mechanical and corrosive mechanisms at the microscale, which is mostly determined by the presence of Cr-rich borides and their effect in the diffusion zone. The measurement of the electrochemical open circuit potential and its normalized course proved to be a valuable tool for assessing the condition of the passive layer and detecting microcracks at an early stage. In corrosive environments at elevated temperature of 80°C, there was a significant change in the failure mechanism with an increase in corrosion attack and multiple crack initiations all over the specimen surface, so that the brazing seam was no longer the weakest point, instead the austenitic base material was, due to stress corrosion cracking. This is remarkable, as it is a corrosive environment that is close to some industrial applications. To interpret the results, it must also be considered that the brazed joint in the study was one that solidified completely isothermally. In Otto et al. (2020), the effect of the non-isothermally solidified brazed joint under corrosion fatigue was shown for a brazed joint with a higher Cr and Mo content. In this case, a very brittle fracture behaviour occurs along the chain-like brittle phase structure in the centre of the brazing seam, significantly reducing the corrosion fatigue performance. In further investigations, the influence of martensite formation and Cr-depletion at the borides in the diffusion zone will be studied in detail, for which Kelvin probe force microscopy (KPFM) will be used to determine the surface potential. Residual stress states in the brazing seam needs to be considered and will be measured using X-ray diffraction (XRD), as these contribute to crack initiation and propagation. Furthermore, finite element methods (FEM) will be used to simulate the stress states on a microstructural level to better understand the influence of the borides.