PSI - Issue 19

Jean-Gabriel SEZGIN et al. / Procedia Structural Integrity 19 (2019) 249–258 Jean-Gabriel Sezgin et al./ Structural Integrity Procedia 00 (2019) 000 – 000

250

2

Gangloff and Somerday 2012; Hirth 1980). Hydrogen-related failure could be attributed to several mechanisms depending on factors such as the source of hydrogen, the metallurgy of the alloy, or the thermomechanical load. The principal mechanisms are the hydrogen-enhanced localized plasticity (HELP) (Birnbaum and Sofronis 1994), the hydrogen-enhanced decohesion (HEDE) (Oriani and Josephic 1974), the adsorption-induced dislocation emission (AIDE) (Lynch 2011), the hydrogen induced cracking (HIC) (Zapffe and Sims 1940), the defactant concept (Kirchheim 2012), and the hydrogen-enhanced strain-induced vacancies (HESIV) (Nagumo 2001). To optimize weight and performance of components, high-strength steels are pertinent. However, when an ultimate tensile strength (UTS) exceeds 1 GPa, the susceptibility of such steels to HE becomes significant (Bandyopadhyay, Kameda, and Mcmahon 1983; Chandler and Walter 1974, 1975; Gangloff 2003; Matsuoka et al. 2017; San Marchi and Somerday 2012; Sandoz 1972; Toplosky and Ritchie 1981; Walter and Chandler 1968; Yamabe et al. 2016). For low and medium-strength steels with the UTS of < 900 MPa, there exists an upper bound for the fatigue crack growth (FCG) acceleration ratio of ~30 (Matsuoka et al. 2017). In this case, the steels failed by quasi-cleavage (QC) and the hydrogen-induced successive crack growth (HISCG) model justified such a bounded acceleration. In contrast, SAE52100 (JIS-SUJ2) with the UTS of 1900 MPa presented intergranular (IG) facets without any upper bound for the FCG acceleration ratio (Yamabe et al. 2012). In this case, the IG failure was justified by the hydrogen-enhanced deformation twin model. These facts sustained an interest for the H-assisted FCG acceleration and related fracture surface morphology for steels with the UTS of around 1 GPa. The FCG rate is a parameter of great interest and some models have been proposed in the literature (Wei and Gangloff 1989). However, the effects of hydrogen on the FCG properties are multiple and have not been fully understood yet, as explained in (Wei and Simmons 1973; Nanninga et al. 2010). Some complex hydrogen effects on the FCG properties exist and therefore, the interested reader is invited to refer to comprehensive review such as (Nanninga et al. 2010; Petit et al. 1994). The fundamental mechanisms of the H-assisted FCG acceleration being not elucidated, the present paper adopts a phenomenological approach to model and predict the H-assisted FCG acceleration ratio. Literature exists about modelling aspects of such acceleration based on multiple mechanisms. The common models are the superposition model (Chen and Wei 1998; Landes and Wei 1969; Wei 2002; Wei and Gao 1983) and the process competition model proposed by Austen et al. (Amaro et al. 2014; Austen and Mcintyre 2014). The validity of these models in regard with the present context is discussed hereafter. The objective of this study was to investigate the effects of hydrogen on 17-4PH H1150 steel with the UTS nearly equal to 1 GPa under both monotonic and cyclic loading and characterize the frequency dependence of the H-assisted FCG acceleration. An interaction model considering both cycle- and time-dependent mechanisms was newly introduced and provided results in coherence with the experimental results.

Nomenclature f

testing frequency R stress ratio stress amplitude ( ) fatigue crack growth (FCG) rate resulting in type failure ( ) fatigue crack growth (FCG) acceleration ratio resulting in type failure 2. Material characterization

2.1. Metallurgical aspects

The alloy in interest was the 17-4PH H1150 stainless steel, composed of 0.04 C, 0.31 S, 0.87 Mn, 0.034 P, 0.004 S, 3.3 Cu, 4.24 Ni, 15.57 Cr, and 0.34 Nb (in mass %); the remainder was iron. This grade was obtained by application of a solution treatment at 1040°C during 1 h (water quenching) prior a precipitation hardening treatment at 620°C during 2 h and air cooling. Observations by means of scanning electron microscopy (SEM) / electron backscatter diffraction (EBSD) showed a martensitic microstructure with a residual austenite content lower than 4%.