PSI - Issue 42

Mihaela Iordachescu et al. / Procedia Structural Integrity 42 (2022) 602–607 Mihaela Iordachescu et al./ Structural Integrity Procedia 00 (2019) 000 – 000

604

3

The microstructure features were determined by SEM on metallographic samples suitably polished and etched. The basic crystallographic phase found was the lath martensite, generated within the previous austenitic grains and grouped in blocks and packages of distinct orientation that could broaden the range of damage micro-mechanisms in aggressive environments. The decarburized zone observed close to the bar surface, with a thickness that varied between 10 and 40 µm, commonly develops in steels subjected to heat treatments at high temperatures or welding at high-speeding rate (Iordachescu D. et al., 2011), since these favor the reaction of atmospheric oxygen with the carbon adjacent to the bars surface. As a consequence of decarburization, the martensite is transformed into large ferrite grains. Fig. 1a and Fig. 1b show the steel microstructure in transverse and longitudinal sections of the bars together with the decarburized area and some oxide deposits embedded on the outer surface as a discontinuous and irregular coating. 2.2. Stress corrosion tests under slow strain rate tensile loading Fig. 1c shows the sketch of the flat tensile notched specimens designed for the stress corrosion cracking tests, in accordance with the research purpose of exploring the slow strain rate tensile behavior of the high-strength lath martensitic steel, as an experimental procedure to assess stress corrosion cracking sensitivity. Thus, the chosen configuration was that of a flat tensile specimen with threaded heads and a rectangular cross-section of 2.5 x 5.3 mm 2 in the testing area. This covers 23 of the 60 mm between the ends of the threaded heads and changes over the remaining length into a bar of circular section of 5 mm diameter, with the corresponding transitions. The specimens were extracted from the bars in such a way that their longitudinal axis was parallel to the bar axis and one of the smaller faces of the tested area coincided with the bar surface, in order to preserve their decarburized, surface layer. This face was notched with a machine tool in transverse direction to the traction axis, its depth varying in the tested specimens from 70 to 200 µm (Fig. 1d). In order to evaluate the steel sensitivity to stress corrosion cracking, the specimens were subjected to slow strain rate tensile load (SSRT-FIP) with the rectangular area immersed in a 20% aqueous solution of ammonium thiocyanate at 50°C, with the displacement of the testing machine actuator being set out to 0.0005 mm/min. Only one of the specimens was broken in the SSRT-FIP test, with the testing of the other three specimens being interrupted at distinct loads levels in order to reveal the evolution of the environmentally assisted cracking process with the testing time. Then, after being unmounted, these specimens were heat-tinted, at 250 0 C for 15 min, in order to mark the environmentally induced cracked surfaces before breaking in air of the remaining ligament by tensile loading the specimen up to fracture, at 0.05mm/min. 3. Experimental results 3.1. SSRT-FIP and fracture tests results Fig. 2 shows the evolution of the load as a function of time in the four SSRT – FIP tests, with separated plots for the phases carried out in the aggressive environment and in air. Since the loading process was carried out at constant machine cross-head displacement, all seven presented curves are equivalent to the load-displacement records of each testing phase, each one with its own scales . The four curves corresponding to the loading phases in the aggressive environment present a linear part with a slight load drop and an almost immediate recovery followed by the load increase (Fig. 2a, Fig. 2c, Fig. 2e and Fig. 2g). In the non-interrupted test case (P2 specimen), the load drop is hardly noticeable at the scales used in the graph (Fig. 2g) and is preceded by a slight non-linear bend before maximum load occurs. The next unloading of the specimen leads to its final breaking, with a double slope due to the process acceleration. The load drop prior to the maximum load of P2 specimen in the SSRT-FIP test was the reference adopted for the interruption of the assisted cracking tests of P3, P6 and P8 specimens. After unloading, these specimens were heat tinted and broken in air. According to Fig. 2b, Fig. 2d and Fig. 2f, the ruptures in air occurred after the maximum load level would have exceed those of in the previous SSRT-FIP tests by about 20%. However, the collapse of P8 required it to be largely deformed in plastic regime, while those of P3 and P6 were preceded by various small load pop-ins with hardly plastic deformation.