**2. Materials and Methods**

In this experimental work, the workpiece material was an AISI 52100 bearing steel (EN 100Cr6) ring. This steel material is widely used for the manufacture of bearing rings. The rings were heat-treated in order to obtain the desired hardness of 61 ± 1 HRC. AISI 52100 bearing steel material has a phase composition consisting of martensitic structure (95.3%), carbides (4.6%), and neglected retained austenite (<1%). The material properties of this steel are given in Table 1.

**Table 1.** Material properties of AISI 52100.


Surface integrity induced by three kinds of finishing processes was studied via SEM investigations and residual stress measurements. Hard turning tests were conducted on a high-precision machine, a prototype lathe positioned in an air-conditioned room, and developed for machining polish-mirror surfaces and hard steel [7,26]. The cubic boron nitrite (cBN) inserts and the machining process parameters used for the hard turning experiments have been precisely detailed elsewhere [7]. As the preconized surface roughness target from the bearing manufacturer, the bearing rings were finished by the Mcrorectif company, Saint-Étienne, France to a preferable target roughness average Ra of approximately 0.2 μm for grinding and 0.05 μm for sequential grinding and honing. The process parameters and roughness are controlled parameters. The roughness average (Ra) defined by ISO 4287 was measured using a scanning white-light interferometer ZygoTM NewView 200 (Zygo Corporation, Middlefield, CT, USA). As can be seen, by applying honing after grinding, the surface roughness becomes smooth. The surface roughness of the precision hard-turned specimen was investigated in previous studies [27,28]. Indeed, a very low surface roughness value Ra = 0.1 μm is obtained in precision hard turning. Sequential grinding and

honing processes allow us to achieve very low surface roughness Ra ≈ 0.05 with respect to precision hard turning.

The specimens were prepared following standard metallographic techniques, as shown in Figure 2. Preliminary specimens were cut using an abrasive cutter and the Struers Discotom 5 machine (Struers, Ballerup, Denmark). The sectioned specimens were then hot mounted in Struers PolyFast conductive resin. Mounted specimens were then ground using 180-grit paper, followed by 320-, 600-, 800-, and 1200-grit papers in a Struers TegraPol-31 system and polished with diamonds. The polished surfaces were cleaned using an ultrasonic bath and then etched with 2% Nital solution for approximately 10 s. Microstructural investigations were carried out using the environmental scanning electron microscope XL30 ESEM-FEG (Philips Electron Optics, Eindhoven, The Netherlands) equipped with an energy-dispersive X-ray (EDX) analyzer.

**Figure 2.** (**a**) Bearing ring finished by precision hard turning; (**b**) sectioned specimen; (**c**) mounted, polished, and etched specimen.

The bearing rings were investigated regarding in-depth residual stresses. These residual stresses were measured by the X-ray diffraction method using a PRECIX robotic system (Figure 3a). The measurements were performed using Cr-Kα radiation diffracted at 2θ = 156◦ in the atomic plan {2 1 1} of the steel. Thus, these conditions give access to the strain localized at a depth of approximately 6 μm according to EN-15305 [29]. As shown in Figure 3b, the residual stresses were measured along the circumferential direction (σc) and tangential direction (σt). Both machined surface and in-depth residual stresses were measured. In the case of in-depth residual stresses, successive layers of material were removed by using chemical etching and electropolishing to avoid the reintroduction of residual stress. To evaluate residual stresses during the RCF test, the measurements were carried out on the bearing raceway, as shown in Figure 3c, thanks to a lead mask, which adapts the spot size of the X-ray beam with the raceway width.

The rolling contact fatigue tests were performed on a twin-disc testing machine (Figure 4) available at CETIM Senlis, France. The contact is made between two discs, one being cylindrical and the other one being crowned with a crown radius of 17.5 mm, as shown in Figure 5. The two contacting discs have the same external radius of 35 mm and width of 14 mm. Tests were carried out under pure rolling conditions (slide-to-roll ratio = 0%), entrainment speed of 11 m/s, and lubricated with MobilGear 629oil injection. The applied normal load is 1100 daN, which is equivalent to the Hertzian contact pressure of 4.5 GPa. For each finished ring, two tests were performed under the same conditions. The twin-disc testing machine is equipped with sensors at the proximity of each disc to detect spalling occurrence. Thus, the test is stopped until spalling forms or when reaching 10 million cycles.

**Figure 3.** (**a**) Setup for the residual stress measurements; (**b**) residual stress measurement directions; (**c**) limiting device of X-ray beam with the raceway width.

**Figure 5.** Drawing of ring specimens.
