PSI - Issue 57

A. Radi et al. / Procedia Structural Integrity 57 (2024) 642–648 Achraf radi/ Structural Integrity Procedia 00 (2019) 000 – 000

644

3

and a homogenous charging of the sample diameter [12,13,15]. The hydrogen concentration was measured using Thermal Desorption Spectroscopy (TDS) on reference specimens (10mm*4mm*0.3mm) charged with the same conditions as fatigue specimens. TDS analysis revealed a higher hydrogen concentration for the HT4 state (C H =27wppm) compared to HT0 (C H =10wppm). The distribution of hydrogen concentration inside the material showed that around 70% of hydrogen was stored reversibly around precipitates (γ/γ’ interfaces) for both HT0 and HT4. These values are compared to pure nickel polycrystal with similar grain size and hydrogen charging with the same conditions where C H = 7wppm. Low cycle fatigue (LCF) tests were conducted at ambient temperature using an MTS 100kN machine on cylindrical specimens (gauge length of 15mm and diameter of 5mm). The fatigue tests involve subjecting a sample to a fully reversed sinusoidal waveform at a frequency of 0.05Hz, with a co ntrolled plastic strain amplitude (ε pa ) of 0.3% ±10% and a load ratio of R ε =-1. An extensometer with a sensitivity lower than 10 -5 was used to acquire accurate strain values. The tests were interrupted at various stages of the specimen’s lifetime, depending on the required surface observations, in order to track the evolution of the extrusion’s heights, thickness and inter -spacing of the dislocations slip bands based on AFM observations. The hysteresis loops are continuously recorded under strain-controlled conditions, which enable to access the cyclic behavior. Based on the analysis of Dickson et al [10], the cyclic stress amplitude (σ a ) can be decomposed into an effective stress (σ eff ) and back stress (X) using the stress-partitioning method. Cyclic stress-strain response is a fundamental material property and its understanding is important. In a two-phases alloy like Waspaloy, the cyclic response is associated with two different microstructure regimes (dislocations shearing and/or by-pass), depending on the precipitates size. As mentioned before, we will only present the results of the dislocation shearing precipitate in the weak coupling domain. The evolution of cyclic stress amplitude (σ a ) as a function of cycle number (N) is illustrated for two states ( figure 1 ): HT0 (d=10nm) and HT4 (d=30nm), with and without hydrogen. Both configurations exhibit similar behavior with or without hydrogen, characterized by an initial hardening phase followed by a slight softening and a saturation phase where the stress amplitude remains constant as a function of cyclic number. Both states (HT0 and HT4) show a softening of (σ a ) in presence of hydrogen with a decrease of (σ a ) of about 12% for HT0 and 6% for HT4 in the first cycle in presence of hydrogen. The softening of (σ a ) at the saturation cycle in presence of hydrogen depends strongly on the metallurgical state i.e precipitates size. For HT0, the H-induced softening observed at the cycle (N=1) decreases with the number of cycles going from 12% to less than 1% at the saturation cycle (N=100). On the other hand, for HT4, the softening of (σ a ) is equal to 6% for both first cycle and saturation cycle (N=80). The values of (σ a ) are reported in table 1 . To improve the understanding of the evolution of the flow stress of hydrogen-free and charged samples, hysteresis loops are analyzed to determine the internal stresses (effective stress (σ eff ) and the back stress (X). Based on the analysis of Dickson et al. [13], the cyclic stress amplitude (σ a ) can be decomposed into an effective stress (σ eff ) and back stress (X) using the stress partitioning method: = − and σ = 1 2 (σ − σ )+ 1 2 σ ∗ Where σ* is the thermal part of the flow stress and σ r denotes the reverse yield stress, which are measured from the experimental hysteresis loops with a plastic strain offset of (5 ± 2) x 10 -5 . The back stress (X) refers to the stress associated with a local strain process providing long-range interactions with mobile dislocations. The effective stress (σ eff ) is the stress required locally for a dislocation to move (short-range interactions, like friction stress, forest hardening and shearing precipitate interactions)[11]. The (σ eff ) and (X) stresses, with and without hydrogen, were calculated from the hysteresis loops. The hydrogen softening of the flow stress i.e stress amplitude in the first fatigue cycle is mainly supported by the softening of the effective stress for both precipitate states. The softening of (σ eff ) for HT0 and HT4 in presence of 3. Results 3.1. Cyclic behavior