13th International Conference on Fracture June 16–21, 2013, Beijing, China -3- parameters were selected: (U1) strike number of 34000 mm-2; static load of 50 N; revolution rate of 37 rpm, feed rate of 0.07 mm/rev; and (U2) strike number of 68000 mm-2; static load of 50 N; revolution rate of 18.5 rpm, feed rate of 0.07 mm/rev. The specimen processing conditions are listed in Table 1. 2.3. Materials characterization To examine the grain-refined layer and the nitriding layer, the cross-section of the all specimens were observed by an optical microscope (Olympus BN2). Before test, the samples were mechanically polished with the sandpapers (from grade 150 to grade 2000) and then a polish cloth with a liquid suspension of alumina. The saturated picric acid solution was used to etch the specimens in this test. X-ray diffraction (XRD) was employed to measure residual stress of all sample varieties, utilizing Cr-Kα radiation for a longer wavelength and deeper penetrating power. The test was carried out on a Rigaku XG-4026A1 using the classical sin2ψ method. Cu-Kα (on a Rigaku Rint-2000) was used to analyze the state of the surface layer nanostructure and the compound layer of nitrided samples to a depth of approximately 5 µm. The microhardness of the plastic deformation zone and nitriding layer was measured on a micro-Vickers hardness tester (MVK-E3, Akashi). The parameters used in the microhardness test were 50 gf and a duration 15 s. The surface roughness of the S45C specimens before and after the UNSM treatment was tested on a contact surface profiler (ULVAC DEKTAK3). To compare the fatigue properties of S45C with the two process methods, a rotating bending fatigue test was carried out on a dual-spindle rotating bending fatigue test machine under atmospheric conditions. The rotating frequency used was 52.5 Hz, and the stress ratio was -1. The failure fracturing of the samples was observed using scanning electron microscopy (SEM) (S-4700, Hitachi). Energy-dispersive X-ray spectroscopy (EDX), with the same machine, was used to detect the composition of any inclusions that induced the sub-surface fish-eye crack initiation. 3. Results and discussion The surface morphology of before and after UNSMed specimens was shown in Figure 2. It is obvious that UNSMed processing marks were produced on the specimen surface. As the processing principle, these marks show parallel and as the increase of processing number these marks become more obvious. For the parallel marks, the distance between each other was approximately 70 µm. This value equals to the processing parameter of the feed rate which is 0.07 mm/rev. During surface treatment, only the surface under the center of the tip can receive most severe plastic deformation as the reason of ball shape of the process tip. Comparing the surface morphology results of UNSM samples with different pre-surface hardness, it can be seen that for the softest surface sample (un-treated) the UNSM process marks show the most obvious after the process of 68000 mm-2. The surface roughness was test and shown in Figure 3. It can be found that the surface roughness (Ra) for the all UNSMed sample has an improvement and even there are a few of obvious UNSMed process marks on the UNSMed sample surface. As the increase of strike number, a little growth can be observed with the surface roughness. Combined with results of surface morphology it is easy to find that the UNSMed process marks produced during surface process were the main reason.
RkJQdWJsaXNoZXIy MjM0NDE=