Microstructural alterations during HPT tests on annealed 52100 bearing steel

Abstract

White Etching Areas – WEAs have been associated with premature and catastrophic damage to rolling bearings. Severe and localized plastic deformation close to the contact surface has been suggested as one of the main drivers for WEAs formation and subsequent evolution. Similar conditions can be created by Severe Plastic Deformation (SPD) processes such as High‐Pressure Torsion (HPT). HPT combines high compressive and torsional stresses to produce grain refinement on a nano scale. HPT tests are quick to perform and thus could provide an easier way to investigate the formation of white etching areas or white structures compared to both rolling contact fatigue tests and an analysis of bearings from service. Therefore, HPT has been used to study white structures in a 52100‐bearing steel in this thesis. Due to the limitations on specimen hardness of the HPT equipment, AISI 52100 bearing rollers in their as received quenched and tempered service condition were annealed to reduce their hardness from 800 HV to 180±10 HV prior to HPT testing. After annealing, disc samples 10 mm in diameter and 1 mm thick were HPT processed under a range of pressures and number of turns. White Etching Structures ‐ WES were found to form in two different regions in the disc samples namely on the top surface of the sample which had been in contact with the upper anvil (designated Top White Etching Areas – TWEAs) and subsurface 240 – 360 µm from the bottom surface of the disc (designated Subsurface White Etching Areas – SWEAs). A morphological characterization was carried out on the HPT produced samples by Macroscopy (MA) and Light Optical Microscopy (LOM) and an Image Processing Program (IPP) was utilised to evaluate the %WES in relation to the pressure and strain applied during the HPT tests. Micro and nano hardness measurements were taken of the WES and the matrix surrounding them. This was then followed by a detailed characterisation conducted using SEM / SEI / BSI to obtain information on the structure, size and shape of WES. SEM / EBSD /EDS were also utilised to evaluate features such as texture, phase distribution and chemical composition not only of the WES, but also of the material surrounding them. TWEAs covered the top surfaces of the discs as many white structures, some isolated, other more interconnected, that seemed to follow the circular tracks on the sample; they were restricted to depths of less than 100 µm from the top surface. SWEAs varied in size and shape with radial distance, from spots with ill‐defined borders (called cloud SWEAs) to well defined solid structures (called solid SWEAs). Both TWEAs and SWEAs shared a similar morphology that consisted of a fibrous / layered structure that appeared to follow patterns caused by the plastic deformation. Their hardness was approximately 24 – 46 % higher than the deformed ferritic matrix. EBSD analysis showed TWEAs and SWEAs in dark contrast in IQ images and as black regions in IPF, AP and KAM maps suggesting poor or even a total lack of indexed diffraction patterns from these are as resulting from a nanocrystalline structure. No recrystallized structures were found in the white structures. EDS analysis indicated that carbon and chromium were dragged from the carbides and dissolved in the nanocrystalline arrangement present in the TWEAs and SWEAs. EDS also revealed inclusions of aluminium with sulphur and manganese that seemed to have suffered a similar dissolution process to that observed for C and Cr. The origin of the fibres / layered structure is related to the interaction between neighbouring regions that flow under different conditions due to high strains. In the case of TWEAs, sliding of the asperities on the top anvil against the sample created heterogeneous flow conditions in small neighbouring regions and instabilities between them. For the SWEAs, the presence of discontinuities such as carbide clusters, pearlite colonies or non‐metallic inclusions generated vortices that promoted instabilities in the ferritic matrix. Some of the SWEAs located at the edge of the samples could have arisen from other effects, such as the lateral friction on the vertical faces of the cavities in the anvils, flash formation due to the quasi‐constrained condition of the HPT tests and the presence of a Dead Metal Zone (DMZ) at the corners of the samples. Some cracks were observed close to or inside the white structures and were caused by the differences in response of these harder areas with respect to the surrounding softer matrix during the deformation imposed by the HPT test. However, not all white structures were associated with cracks, suggesting that cracks are not necessarily a precursor to white structures. Some of the features, such as the fibrous structure, changes in carbide distribution and high hardness seen in HPT samples are similar to those observed in the WEAs located in failed rolling element bearings. This suggests that a similar phenomenon could be occurring in the vicinities of inclusions located subsurface in rolling bearings and lead to the formation of WEAs.

More information

Identifiers

ORCID for Ling Wang: orcid.org/0000-0002-2894-6784

Catalogue record

Export record

Altmetrics