Wear and Friction Behavior of the Aerospace Bearing Steel M50 and a Nitrogen-Alloyed Stainless Steel Under Lubricated Sliding Conditions

Abstract

The material requirements on aircraft engine mainshaft bearings increase due to an elevated speed index (bearing bore diameter multiplied by rotational shaft speed) and slip ratios [1]. The formation of reaction layers on surfaces in mechanical contact is strongly affected by the tribological loading conditions, the materials used, the lubricant, and the service temperature. An appropriate reactivity between material and lubricant in tribological systems decreases wear and friction and increases the durability [2,3]. Goal of the paper is to compare wear and friction properties of standard aerospace bearing steel AMS 6491 (M50) with that of the high strength stainless steel grade AMS 5898. The nominal chemical compositions are 0.82C-4.1Cr-1V-4.2Mo (wt%) and 0.3C-0.4N-15.2Cr-1Mo (wt%), respectively. In order to characterize the material behavior under pure sliding conditions, ball on disc (BOD) experiments were performed with a contact pressure of 1GPa and a sliding speed of 10 cm/s at room temperature and at 150°C. As lubricant the jet engine oil Mobil Jet II was used. It is assumed that the reaction layer formation depends on the material composition and is also effected by the counterpart and the lubricant additives. Thus, the experiments were performed with two different ball materials. The first ball material was the same standard aerospace bearing steel (M50) as mentioned above and the second was a ceramic (Si3N4). The homogeneity and the distribution of the reaction layers as well as the wear rate were determined in the contact zone by means of optical profilometry, scanning electron microscopy (SEM) and Fourier transformation infrared spectroscopy (FTIR). The study showed that wear is significantly higher on the stainless steel grade compared to M50 (Fig. 1a and Fig. 1b). The dark areas in Fig. 2 are phosphorus rich regions on the wear track of M50. This reaction layer is mainly built up of phosphates, which result from the TCP lubricant additive. In the FT-IR spectra (Fig. 3) absorption bands at 1130 cm−1 (room temperature-BOD test) and at 1160 cm−1 (150°C-BOD test) are visible, which result from the P=O stretching [4]. The shift of the absorption band to lower wave numbers with decreasing test temperature is probably due to hydrogen bonding [5]. Contrary to AMS 6491 no measurable reaction layer was found on AMS 5898 after testing at room temperature (Fig. 1b). The friction coefficients of the two steels against Si3N4 balls determined in the BOD tests are compared in Fig. 4. AMS 5898 shows an abrupt increase of the friction coefficient after a sliding distance of 3.5 m. The reason for that is a material transfer of disc material to the ceramic ball as can be seen in Fig. 5. This transfer material causes a plowing of the disc and thus, an increased wear can be observed. The different Cr-contents and consequently oxide layers of AMS 5898 and AMS 6491, which react differently with TCP, might explain this behavior. AMS 5898 does not sufficiently react with TCP at room temperature to form a protective layer. Consequently, material transfer and increased wear occurs. In case of AMS 6491 an increase of the operating temperature cause a change of the reaction layer (see Fig. 3) and to an increase in the wear rate.