23rd International Colloquium Tribology - January 2022 263 Study of the early stages of subsurface cracks and microstructural alterations in 100Cr6 under hydrogen and RCF influence Fernando José López-Uruñuela Tekniker, Basque Research and Technology Alliance (BRTA), C/ Iñaki Goenaga, 5, 20600 Eibar, Spain Department of Mechanical Engineering, University of the Basque Country (UPV/ EHU), Bilbao, Spain Corresponding author: fernando.lopez@tekniker.es Beatriz Fernandez-Diaz Tekniker, Basque Research and Technology Alliance (BRTA), C/ Iñaki Goenaga, 5, 20600 Eibar, Spain Bihotz Pinedo Tekniker, Basque Research and Technology Alliance (BRTA), C/ Iñaki Goenaga, 5, 20600 Eibar, Spain Josu Aguirrebeitia Department of Mechanical Engineering, University of the Basque Country (UPV/ EHU), Bilbao, Spain 1. Introduction WEC and WSF have been studied for the last 20 years. This premature failure affects wind turbine gearbox bearings among other mechanical components from other industrial sectors. Hydrogen is one of the main drivers triggering subsurface cracks and consequently WEC. Nevertheless, it has not been established so far how hydrogen acts when it comes to steel embrittlement in this specific case. According to literature, hydrogen can act by the following mechanisms: HELP, HEDE and HESIV [1]. In the present study, the authors put effort into trying to understand which HE mechanism plays a key role in pores and micro-cracks initiation. Transgranular and intergranular crack morphologies were examined in order to shed light on how hydrogen enhances steel embrittlement. 2. Experimental methodology Disc-on-disc tribometer configuration with spall lubrication was used. The specimens were 20 and 40 mm in outer diameter and 3 mm of contact line was used. The tested lubricant was a poly-α-olefine synthetic oil (ISO VG320). Premature subsurface cracks and WEC were reproduced through a transient condition test to simulate the actual wind turbine gearbox bearing working conditions. The maximum calculated Hertzian stress did not exceed 1,9 GPa and the maximum SRR was set to 45% while the mean value was 21% SRR. For hydrogen uptake, the specimen was put into a solution (sulfuric acid (H2SO4) (0,1 mol/ l) solution and potassium thiocyanate (KSCN) (1 g/ l)) under a −1,2 V potential (this potential was determined by making a previous polarization curve) for 4 h. This method ensures cathodic protection of the discs so that possible corrosion is prevented. After testing specimens were analysed using optical microscopy and SEM. The corresponding metallographic preparations were carried out prior to the analyses. These included cutting, grinding, and polishing. 3. Results and discussion The specimens not precharged did not show any failure. It is of the authors opinion the specimens were not subjected to enough cycles to generate subsurface fatigue. In the case of the precharged specimens, two scenarios can be distinguished, those specimens that were precharged twice, once at the beginning of the test and once in the middle of the test and those that were precharged just once, before testing. Specimens precharged twice were tested until the spalling occurs while the test with specimens precharged once was stopped before spalling. WECs and cracks without microstructural alteration were found at subsurface level in the spalled discs. Some of the cracks were connected to the surface while others were oriented parallel to the raceway without reaching the surface. WEAs formed because of crack face rubbing were found [2]. Likewise, WEAs whose formation was not possible through this mechanism were identified. Crack face rubbing in these cases was not feasible because of the lack of an adjacent crack. Nevertheless, these microstructural alterations were surrounded by micro-cracks. It is hypothesized that regions of material where micro-cracks develop are more prone to trigger 264 23rd International Colloquium Tribology - January 2022 Study of the early stages of subsurface cracks and microstructural alterations in 100Cr6 under hydrogen and RCF influence plastic deformation since the matrix has been weakened. This, in conjunction with hydrogen embrittlement favours WEAs formation [3]. The belief in the need for the existence of a crack prior to the formation of WEAs comes from what was found in the test where premature stages of WEC formation were studied. In this test, where no spalling developed, numerous subsurface micro-cracks appeared. These were sometimes accompanied by small volumes of WEA. No isolated WEA was found without adjacent or surrounding cracks. These results suggest that under the influence of hydrogen and RCF conditions WEC occurs because of the presence of subsurface cracks. Figure 1: WEA surrounded by micro-cracks. Through the detailed analysis of the cross-sections, we tried to interpret how hydrogen affects the generation of subsurface cracks and microstructural alterations. The modes of hydrogen embrittlement commonly presented in the literature seem to have a synergistic effect on the generation of this failure mode. On the one hand, HEDE favours the presence of pores that eventually coalesce with cycling to generate cracks [4]. On the other hand, HELP favours the movement of dislocations which in turn promote plastic deformation phenomena in regions of material under high stress [5]. One of the ways to find out how hydrogen acts in the formation of cracks is to study the morphology of the cracks, i.e., to analyse whether they are transgranular or intergranular cracks. In the present study, both transgranular cracks and WEAs were found. Although the presence of hydrogen has generally been associated with intergranular fracture, the results shown here prove that under certain conditions (RCF) and HE transgranular fracture is promoted. Figure 2: Transgranular WEA crossing grain boundary. 4. Conclusions The following conclusions can be drawn from the present study: • Under the influence of RCF and HE cracks tend to initiate in the subsurface. • With cycling, microstructural alterations develop in regions where micro-cracks were formed. • Hydrogen acts according to the two most proposed mechanisms in literature: HELP and HEDE. • Cracks are formed by the coalescence of pores generated in regions of maximum shear stress. • With cycling, both transgranular cracks accompanied by WEAs and transgranular WEAs surrounded by cracks develop. References [1] Hussein A, Krom AHM, Dey P, Sunnardianto GK, Moultos OA, Walters CL. The effect of hydrogen content and yield strength on the distribution of hydrogen in steel: a diffusion coupled micromechanical FEM study. Acta Mater 2021; 209: 116799. doi: 10.1016/ j.actamat.2021.116799. [2] Lai J, Stadler K. Investigation on the mechanisms of white etching crack (WEC) formation in rolling contact fatigue and identification of a root cause for bearing premature failure. Wear 2016; 364- 365: 244-56. doi: 10.1016/ J.WEAR.2016.08.001. [3] Kürten D, Khader I, Raga R, Casajús P, Winzer N, Kailer A, et al. Hydrogen assisted rolling contact fatigue due to lubricant degradation and formation of white etching areas. Eng Fail Anal 2019; 99: 330- 42. doi: 10.1016/ J.ENGFAILANAL.2019.02.030. [4] Spille J, Wranik J, Barteldes S, Mayer J, Schwedt A, Zürcher M, et al. A study on the initiation processes of white etching cracks (WECs) in AISI 52100 23rd International Colloquium Tribology - January 2022 265 Study of the early stages of subsurface cracks and microstructural alterations in 100Cr6 under hydrogen and RCF influence bearing steel. Wear 2021: 203864. doi: 10.1016/ j. wear.2021.203864. [5] Connolly M, Martin M, Bradley P, Lauria D, Slifka A, Amaro R, et al. In situ high energy X-ray diffraction measurement of strain and dislocation density ahead of crack tips grown in hydrogen. Acta Mater 2019; 180: 272-86. doi: 10.1016/ j.actamat.2019.09.020.