Effect of creep feed grinding on surface integrity and fatigue life of Ni3Al based superalloy IC10

Shuiqi ZHANG, Zhongxue YANG, Ruisong JIANG, Qiho JIN,Qing ZHANG, Wenhu WANG

a The Key Laboratory of Advanced High Temperature Structural Materials, AECC Beijing Institute of Aeronautical Materials, Beijing 100095, China

b School of Aeronautics and Astronautics, Sichuan University, Chengdu 610065, China

c School of Construction Machinery, Chang’an University, Xi’an 710064, China

d School of Mechanical Engineering, Northwestern Polytechnical University, Xi’an 710072, China

KEYWORDS Creep feed grinding;Fatigue life;Grinding parameters;IC10;Surface integrity

Abstract Ni3Al-based superalloy IC10 is widely used in high temperature components of aeroengines because of its superior mechanical properties. In this paper, the creep feed grinding properties of IC10 were investigated experimentally.The effects of grinding parameters on the grinding forces and temperature were examined.Moreover, the influences of surface roughness and hardening on the high-cycle fatigue life of IC10 specimens were studied.To control the creep feed grinding parameters and enhance the fatigue life of IC10 components,the experimental results were summarized to offer a useful reference point.It is concluded that,the grinding depth is the most important factor which influencing the grinding forces and temperature;the surface roughness is the main and unfavorable factor on the fatigue life of IC10, while the surface hardening has a positive influence on the fatigue life; to obtain a better surface quality and improve the fatigue life of IC 10, the recommended grinding parameter domain involves wheel speed ∈[15, 20] m/s, feed rate ∈[150, 200]mm/min, and grinding depth ∈[0.4, 0.5] mm.

1. Introduction

Ni3Al-based superalloy are a diverse group of materials,which considered to be near-future candidates for advanced high temperature structural materials in aerospace applications,e.g. turbine blades, disks, and cases, due to their superior mechanical properties at high temperatures,1,2IC10 superalloy is a new kind of Ni3Al-based superalloy, and will be applied in the guide vane of the turbine, due to the relatively low density, high melting point, and excellent oxidation and corrosion resistance at high temperatures.3,4However, IC10 has been considered as typical difficult-to-cut materials due to the high strengths at elevated temperatures,high work hardening,and low thermal diffusivity.Creep feed grinding process is usually used for machining the turbine blade root with a rather low feed speeds and a large cut depths.5,6Although substantial efforts were carried out to understand the creep feed grinding process and the finished surface integrity for machining Ni3Al-based superalloy, a variety of problem still demanded to be solved before industrial application.

To promote the engineering application and gain more indepth understanding about the creep feed grinding Ni3Albased superalloy, multitudes of efforts and attentions have been paid already from both academic and engineering aspects. Most of them can be categorized into two groups:(i) the machining properties, including grinding parameters,forces and temperature, (ii) the grinding surface integrity and fatigue life.

For the IC10 machining properties aspect, as a representative of Ni3Al-based intermetallics, IC10 is typical difficult-tomachine materials because of its wear resistance, and is hard to machine with the conventional machining methods such as turning or milling. Currently, most research focuses on the material preparation process,7,8microstructure evolution behavior,8and the mechanical properties of IC10 superalloys,9 few efforts have focused on the machining technology, thus limiting the engineering applications of this kind of materials.In aspect of grinding similar materials,Yao et al.10conducted an experimental study, which found that single alumina (SA)wheels are more suitable for Inconel 718 grinding, and that better surface integrity can be obtained withap=5 μm,vw=16 m/min, andvs=5 m/s as grinding parameters.Sunarto11studied the grinding abilities of ultrafinepolycrystalline cBN (PcBN-U) wheels and a representative conventional monocrystalline cBN wheel. They found that the grinding forces decreased by 20%-30% using PcBN-U wheels when grinding Nimonic 80A. Furthermore, the PcBN-U grit is suitable for applications with high dimensional accuracy in creep feed profile grinding for superalloys,because it causes less profile wear, and hence better form retention.Ding et al.12applied the brazed cBN abrasive wheel in the creep feed grinding of the K424 nickel-based superalloy. The grinding temperature decreased to 100°C with corresponding cooling conditions and a burn-free and crack-free ground surface was achieved. Gu et al.2found that microcrystalline alumina (MA) abrasive wheels showed even better grinding performance in the aspects of grinding forces, force ratios,and specific grinding energy in the creep feed grinding of the single crystal superalloy DD6. Miao et al.13compared the grindability of wrought GH4169, equiaxial cast K403, directionally solidified cast DZ408 and single crystal DD6 superalloys by brown alumina wheels in the creep feed grinding process, with regarding to the grinding force, grinding ratio,wheel wear and surface integrity. The results shown that the mechanical properties of workpiece has the significant influence on the grindability of nickel-based superalloys in terms of the grinding force due to the large friction between the alumina abrasive grains and workpiece material. The wheel clogging was found in grinding DD6. The thermal softening behavior increased with increasing material removal rate,and the affected layer achieved to 40 μm. Ding et al.14reviewed the monolayer electroplated and brazed cubic boron nitride(CBN)superabrasive wheels for grinding metallic materials,the fabrication techniques and mechanisms of wheel were reported.The applications of monolayer CBN wheel for grinding steels, titanium alloys, and nickel-based superalloys are also provided.

Grinding surface properties have a significant influence on the fatigue life of the components. Regarding the effect of the grinding surface integrity on the fatigue life of superalloys,Sinha et al.15studied the surface burn during the grinding of Inconel 718, and discovered that it can be ground more efficiently using alumina grinding wheels compared to the SiC grinding wheel, since the chemical reaction between the SiC grits and the workpiece material at a higher grinding temperature may result in severe attrition wear.Bhaduri et al.16investigated the ultrasonic assisted creep feed grinding of Inconel 718. Their results indicated that the use of ultrasonic assisted operation typically resulted in the reduction of grinding forces and associated workpiece surface roughness.Ultrasonic vibration generally leads to an increase in the number of active cutting points on the wheel, which subsequently results in a decrease of corresponding average workpiece surface roughness. Chen et al.17tested the creep feed grinding of the directionally solidified superalloy DZ4, revealing that the force ratio changed rapidly from 3.7 to 4.6 as grinding burns occurred. This might be attributed to the fact that the grinding-induced high temperature softened the matrix material and led to a higher probability for the wheel to be clogged.Cai et al.18carried out experimental studies on the grinding surface properties of the single crystal superalloy DD5. It was shown that the wheel speed has the most influence on surface roughness, and the optimal process parameters are the grinding wheel linear speed being 30 m/s, the grinding depth 20 μm,and the feed rate 0.4 m/min.With these process parameters,a plastic deformation layer of about 2 μm appears on the grinding subsurface, and a work hardening layer of about 0.5 μm appears between the grinding surface and the plastic deformation layer.Huang and Ren19studied the effect of surface integrity on the high-cycle and low-cycle fatigue life of the nickel-based superalloy GH33A at room temperature and 550°C, respectively. Doremus et al.20evaluated the influence of initial residual stresses surrounding scratches or dent type surface anomalies on the fatigue crack growth of Inconel 718 alloy.Galatolo and Fanteria21studied the influence of turning parameters on the high-temperature fatigue performance of Inconel 718. They found that residual stresses dominated the fatigue life.Sun et al.22proved that the specimens with larger compressive residual stress and thinner metamorphic layers have stronger abilities to resist fatigue crack initiation. Li et al.23found that surface roughness has predominant and negative effects on the failure life of Inconel 718 specimens when it is higher than 0.4 μm. However, when the surface roughness is shorter than 0.3 μm, the surface microhardening rate has predominant and positive effects on the fatigue life.Ding et al.24provided a model to predict the surface topography in grinding process by a textured monolayer CBN wheels.The surface topology of textured monolayer CBN wheel was reconstructed and the wheel topology evolution on the undeformed chip thickness nonuniformity was determined. The ground surface roughness was improved with a continuous reducing undeformed chip thickness nonuniformity. The reconstructed surface topology of wheel can be used to predict the workpiece topography in different stages of the grinding process.

Despite the large number of studies on the grinding and fatigue life of nickel-based superalloys mentioned above, few researchers have examined those of Ni3Al-based superalloys.However, different microstructures between these two types of superalloys lead to different mechanical properties. Consequently, their grinding performance will be different. To our best knowledge, only Zhu et al.25investigated the creep feed grinding forces and surface roughness of the Ni3Al-based superalloy IC10.However,to meet the industrial applications,a deeper understanding of creep feed grinding surface quality and fatigue performance are needed. Hence, this study aims to achieve a better understanding of creep feed grinding performance of the IC10 superalloy. The surface integrity and its influence on the fatigue life were investigated.The organization of the paper is as follows: Section 2 discusses the detailed conditions of the grinding trials.In Section 3,the experimental results are presented and discussed. Finally, conclusions and future work are provided in Section 4.

2. Experimental procedure

2.1. Materials and specimens

The IC10 Ni3Al-based superalloy was prepared by Beijing Institute of Aeronautical Materials, and the nominal chemical composition (wt%) is showed in Table 1. The normative heat treatment of IC10 included the solid solution process(1180°C for 2 h, 1265°C for 2 h, + air cooling) and the aging treatment (1050°C for 4 h). In order to intensify the solid solution process, the IC 10 material was heated at 1180°C for 2 h to redissolve the phase with lower melting firstly. Then, the IC 10 material was heat at 1265°C for 2 h to dissolve the excess phase into the solid solution sufficiently. The microstructure was examined before and after the heat treatment, respectively (see Fig. 1). The microstructure of the casting superalloy IC10 is dominated by γ’ phase,γ+γ’ eutectics phase, and a small amount of carbide precipitated phase. The γ’ phase has an irregular shape (see Fig. 1(a)). However, after the heat treatment, the residual γ+γ’eutectics phase is basically eliminated, and the γ’ phase has a regular shape (see Fig. 1(b)).

Two kinds of specimens were used in this study. One was for investigating the grinding properties of IC10,and the other for testing the fatigue life. The grinding specimens are cuboid(29 mm×15 mm×15 mm). The CAD model of the fatigue specimens with a thickness of 4 mm is shown in Fig. 2. To avoid crack initiation at the edges of the specimens, all the edges were subjected to chamfering ofR=0.1 mm.Moreover,the unexamined surfaces and edges were polished toRa<0.1 μm.

Fig. 1 Microstructure of IC10 superalloy.

Table 1 Nominal chemical composites of superalloy IC10.

2.2. Experimental setup

The experiments were carried out on a CHEVALIER FSGB818CNC creep feed grinder (see Fig. 3(a)). Down grinding was conducted in this study. A white alumina (WA) and pink fused alumina(PA)grains mixed grinding wheel was employed to grind the IC 10 superalloy. The WA and PA with a grain size of 80 meshes were bonded by resinoid bond,and the mixing ratio of WA and PA was 1:1. The grinding wheel with a diameter of 300 mm was shaped by the Fenghang grinding wheel manufacturing Co. Ltd at Handan city. To keep the grinding wheel surface topography consistent during each experiment,the wheel was dressed after each grinding process.The schematic diagram of the grinding process is shown in Fig. 3(b). The grinding forces were recorded by the dynamometer, and the grinding temperature was tested by semi-artificial thermocouple. The schematic diagram of the handcrafted artificial thermocouple is shown in Fig. 3(c).

Since the fatigue specimen is too thin to be fixed with a conventional clamp,a magnetic fixture was used to hold the specimen during the creep feed grinding process, as shown in (see Fig. 4).

Three-factor five-level experiments were conducted to investigate the influence of the grinding parameters on the grinding surface integrity of the IC10, as shown in Table 2.

Fig. 3 Experimental setup for Creep feed grinding IC10 superalloy.

Fig. 4 Fixture of fatigue specimen.

Table 2 Design of experimental parameters.

2.3. Measurement

The grinding force was measured online by a Kistler 9257B piezoelectric dynamometer, coupled with a Kistler 5080A multi-channel charge amplifier and computer data acquisition software. Meanwhile, the grinding temperature was measured with a semi-artificial thermocouple,the diameter of constantan wire was 0.6 mm. The cooling liquid (20% emulsion and 80%water) was used in the grinding process. After the grinding tests, the surface roughness was measured by a surface roughness tester (T620A) with evaluation and a cut-off length of 0.8 mm. Measurement of each point was repeated three times and the average values were reported. The surface microhardness measurements were conducted by a Vickers microhardness tester with a load of 25 g. In addition, the surface topography was observed by a scanning electron microscope(SEM).

Fig. 5 Fatigue test.

2.4. Fatigue test

The fatigue specimens were machined with the same grinding parameters in Table 2 to obtain the corresponding surface integrity. The fatigue cycles of different specimens were obtained by a QBG-25KN-3 type fatigue test machine with a 0 stress ratio tensile test. The maximum stress reached 300 MPa (see Fig. 5).

3. Results and discussion

3.1. Grinding forces

The grinding forces have a considerable influence on surface roughness, residual stresses, and some other grinding-induced defects.26Hence,it is important to achieve a deep understanding of grinding forces (including normal forceFnand tangential forceFt) with different process parameters. As the Kistler force measuring platform is settled, grinding forces can be obtained by decomposing the measured vertical force (Fv)and the horizontal force (Fh) according to Eqs. (1)-(3).11

whereFnis the normal force,Ftis the tangential force,Fvis the vertical force,Fhis the horizontal force,θ is the angle between the vertical force and the normal force acting on the grinding zone, which can be expressed as,

Fig. 6 Influence of process parameters on grinding forces.

whereapis the grinding depth, anddsthe grinding wheel diameter.

In order to investigate the influence of the process parameters on the grinding forces, the grinding forces were measured based on the experimental parameters in Table 2. Each group of the measurement was repeated five times. The mean value and standard deviation of the grinding force were analyzed.The ratio between the standard deviation and the mean value usually less than 7.11%, hence, the tested results were valid.The variation of the grinding forces with grinding parameters are presented in Fig.6.From Fig.6(a),it can be seen that bothFnandFtdecreased with the increase of the wheel speed from 10 m/s to 30 m/s,by 7.5%and 28.7%,respectively.The mechanism is that as the wheel speed increases,the undeformed chip thickness of the single abrasive grain decreases,resulting in the decrease of the grinding forces. It is noticed that with the increase of the workpiece feed rate,the normal force increased by 32.7%, and the tangential force by 35.5%. With the increase of the grinding depth,both the normal and tangential forces increased because of the increase of the undeformed chip thickness.Fnincreased by 47.3%, andFtis 47.2%.

Moreover, based on the above analysis, it can be seen that the grinding depth has the most significant influence on the grinding forces, followed by the workpiece feed rate and the wheel speed. For the creep feed grinding of superalloy IC10,the normal force is ranged between 380 and 570 N, and the tangential force between 190 and 350 N, much larger than those in the conventional grinding of superalloys.

3.2. Grinding temperature

The grinding temperature was also tested based on the experimental parameters in Table 2. The mean value of grinding temperature was counted according to the results of five times repeat testes. The error ratio between the standard deviation and the mean value usually less than 25.85%. Compared with the test results of grinding forces,the error ratio was relatively high.It is difficult to measure grinding temperature accurately,hence, the test results were receivable. The influence of the grinding process parameters on the grinding force are demonstrated in Fig.7.The influence of the wheel speed on the grinding temperature is showed in Fig. 7(a). It can be seen that as the wheel speed increased from 10 m/s to 30 m/s, the grinding temperature increased from 840°C to 1100°C. The mechanism is that with the increase of the wheel speed, the number of abrasive grains participating in the grinding increases,resulting in the increase of the grinding temperature. On the other hand, as the feed rate of the workpiece increased, the grinding temperature increased slightly while the feed rate was lower than 150 mm/min. With the further increase of the feed rate, the grinding temperature decreased slightly (Fig. 7(b)). Fig. 7(c) shows the effect of the grinding depth on the grinding temperature. As the grinding depth increased, the grinding temperature showed an observably upward trend.When the grinding depth increased from 0.3 mm to 0.7 mm,the grinding temperature rose from 420℃to 1150°C. The main mechanisms can be summarized as follows: 1) the increase in the grinding depth leads to an increase in the chip deformation resistance, resulting in the increase of friction; 2)the increase in the grinding depth leads to a significant increase in the grinding arc length,which makes it more difficult for the coolant to enter the grinding zone.Hence,the grinding temperature increases rapidly.

According to Fig. 7, the grinding temperature usually achieved at 800-1000°C, the semi-artificial thermocouple was employed to measure the grinding temperature, the cooling liquid was employed in the grinding process and the burn defect was not founded at the grinding surface.Compared with the recent reports,6the test results of grinding temperature were relatively higher.Considering the diameter of constantan wire was 0.6 mm,the grinding temperature of grinding surface was recorded at the tip of the constantan weir,shown in Fig.3(c). Hence, the temperature record region was nearly as a point, the thermoelectric potential was recorded immediately before the thermal diffusion or boiling effect,the tested results of grinding temperature were relatively high. In addition, the contact arc of the creep feeding grinding is much larger than plane grinding, the cooling liquid is difficult to inject the contact arc, and the cooling performance of the liquid is deficient consequentially. Meanwhile, the temperature of grinding surface was up to 700-900°C for the creep feed grinding of turbine blade root (Inconel 718) at last finishing grinding process with a grinding depth of 0.5 mm.6Liu et al.27provided an in-situ infrared temperature-measurement method for creep feed grinding process, and the temperature of grinding surface achieved 1000°C in grinding martensitic stainless steel.

Fig. 7 Influence of process parameters on grinding temperatures.

Furthermore, the grinding temperature usually achieved at 800-1000°C, there was no burn defect at grinding surface in grinding IC 10. Because the IC 10 is applied in the guide vane of the turbine, and the operating temperature is high than 1200℃.8The normative heat treatment of IC10 included the solid solution process (1180°C for 2 h, 1265°C for 2 h, +air cooling), the heat treatment reinforced the property of IC 10. Hence, the heat of grinding can’t cause burn defect at grinding surface.

3.3. Surface roughness

The ground surface roughness is commonly used to characterize the surface finishing, and have crucial effect to the fatigue life of shaped components.24Surface roughness is the most important factor for the evaluation of the machined surface quality. Five specimens were prepared for each group of process parameters in Table 2, and the surface roughness of each specimen was tested three times, then, the mean values and standard deviations of the grinding surface roughness were counted. The ratio between the standard deviation and the mean value usually less than 17.08%. Fig. 8 shows the effects of the creep feed grinding parameters on the surface roughness, while Fig. 9 presents the 3D topography of the grinding surface. It can be seen that the surface roughness varied from 0.3 μm to 0.75 μm. In addition, the surface roughness decreased with the increase of the wheel speed when it was lower than 20 m/s. After that, the surface roughness increased rapidly with the increase of the wheel speed. The mechanisms are summarized as:1)the undeformed chip thickness decreases with the increase of the wheel speed when it is lower than 20 m/s, leading to slight surface roughness; however, 2) when the wheel speed is higher than 20 m/s, the grinding temperature increases rapidly in the grinding zone,resulting in the adhesive wear of the grinding wheel. Hence, the surface roughness increased (Fig. 8(a)). Furthermore, it can be seen that with the increase of the feed rate and the grinding depth,the surface roughness increased.It increased particularly rapidly when the feed rate is higher than 200 mm/min. The surface roughness increased linearly with the increase of the grinding depth(Fig.8(b),Fig.8(c)).From Fig.9,it can also be seen that both the feed rate and the grinding depth have a significant influence on the surface roughness.

According to references 19 and 23, surface roughness has predominant and negative effects on the fatigue life of superalloys. To obtain a better surface quality, the following creep feed grinding parameters are recommended:wheel speed varying from 10 m/s to 20 m/s, feed rate lower than 200 mm/min,and cutting depth shorter than 0.5 mm.

Fig. 8 Influence of process parameters on surface roughness.

Fig. 9 3D topography of surface with different grinding parameters.

Fig. 10 Effects of grinding parameters on surface microhardening.

3.4. Surface microhardening

Surface and subsurface microhardness have significant influence on the fatigue life of superalloys. During the deep creep grinding process, the microhardening mechanism of surface layer is a combination of plastic deformation and high temperature softening of superalloy IC10.

Fig. 10(a) displays the effect of the wheel speed on the microhardness of the surface.It can be seen that the workpiece suffered a critical surface microhardening. The microhardness gradually decreased along the grinding depth direction until it reached the substrate microhardness. Furthermore, the degree of hardening gradually decreased with the increase of the wheel speed. The corresponding surface microhardnesses with 10 m/s, 20 m/s and 30 m/s are 490HV, 429HV and 420HV respectively, while the substrate microhardness of IC10 is 404HV. Hence, the microhardening rates are 21.3%, 6.2%,and 4.0% respectively. According to the Fig. 10(b), it seems that the feed rate has little influence on the surface microhardness. The corresponding surface microhardnesses at the feed rates of 100 mm/min, 200 mm/min, and 300 mm/min were 420HV, 416HV, and 414HV respectively, the microhardening rates were 4.0%, 3.0%, and 2.5%, respectively. Fig. 10(c)shows the effect of the grinding depth on the microhardness of the grinding surface. As the grinding depth increased, the degree of microhardening gradually increased. When the grinding depths were 0.3 mm, 0.5 mm and 0.7 mm, the corresponding surface microhardnesses were 450HV, 470HV,520HV, and the microhardening rates were 11.4%, 16.3%,28.7%, respectively.

3.5. Fatigue test results

Based on the recommended creep feed grinding parameters in Section 2.3, an orthogonal experiment was designed to reveal the relationship between the surface integrity and the fatigue life of superalloy IC10. The grinding parameters and experimental results are shown in Table 3.

According to the data in Table 3, the relationship between the fatigue lifeNfand the surface integrity parameters can be expressed with a multiple linear regression equation:

whereNfis the fatigue life of the test samples,Rais the surface roughness,his the microhardness depth, andNis the mocrohardness rate.

Moreover, the influences of the surface roughness, the microhardness depth, and the microhardness rate on the fatigue life are shown in Fig. 11.

From Eq.(4)and Fig.11(a),it can be seen that the surface roughness has a dominantly negative influence on the fatigue life of the IC10 specimen. The mechanism is that the higher surface roughness raised the effective stress concentration coefficient.19The deeper the wave valley of the surface roughness or the smaller the curvature of the micronotch root radius,the larger the coefficient of the effective stress concentration, as shown in Fig. 12.

From Eq. (1) and Fig. 11(b) & (c), it can be seen that the fatigue life magnified with the hardening depth and hardening rate increased,both the hardening depth and hardening rate of the microhardness have positive influences on the fatigue life.Mechanism can be explained that the grinding force leads plastic deformation at the subsurface, as shown in Fig. 13. The mechanical properties of the plastic deformation layer can be quantized by the microhardness depth and rate, as shown in Table 3. The plastic deformation layer prevented the protruding of dislocation lines. Hence, the initiation and propagation of the fatigue cracks were delayed at the machined surface. In addition,the microhardness of the surface was larger than that of the substrate, the crack occurred under the surface. Hence,when the hardening depth and hardening rate increasing, larger tensile stress or more cycles were needed to expand the fatigue cracks.

Table 3 Results of fatigue life test.

Fig. 11 Effects of surface integrity on fatigue life.

Moreover,the fracture surfaces of two fatigue samples were investigated,as shown in Fig.14.The surface roughness of the samples I and II were 0.94 μm and 0.44 μm, and the corresponding fatigue lives were 1.25×106 and 9.69×106,respectively. From Fig. 14(a) and 14(b), it can be seen that both samples underwent single source fatigue. Typical fatigue fractures include fatigue cracks, fatigue striations, dimples and brittle failures. Furthermore, the fatigue cracks initiated from the grinding surface. Compared with sample II, the surface quality of sample I was worse. More defects were found on the surface of sample I,as shown in Fig.14(c)and 14(d).Poor surface quality is often accompanied with processing defects(such as scratches, crush damage, tool mark gaps, uneven grinding lines), which can easily cause stress concentration,increasing the probability of fatigue cracking and reducing the fatigue life of the workpiece.Therefore,controlling the surface quality,particularly,the surface roughness of the grinding surface, is significant method for improving the fatigue life of grinding workpieces.

Fig. 12 Micrograph of surface roughness.

Fig. 13 Microstructure of grinding surface (vs=30 m/s,vw=250 mm/min, ap=0.5 mm).

4. Conclusions and scope of future work

In this study,the creep feed grinding properties of Ni3Al-based superalloy IC10 were studied experimentally.A serial of experiments were carried out to analyze the influences of the process parameters on the grinding forces, temperature, and surface qualities. Moreover, the influence of the surface qualities on the fatigue life was investigated. The following main conclusions of this study can be drawn:

1) The creep feed grinding forces decrease with the increase of the wheel speed, while increase as the feed rate and the grinding depth increase, particularly the latter.

2) The grinding depth has greater influences on the temperature compared with the wheel speed and feed rate.The grinding temperature rose by almost 3 times when the grinding depth increased from 0.3 mm to 0.7 mm.

3) A relatively good surface quality can be obtained when the wheel speed is between 10-20 m/s,the feed rate lower than 200 mm/min, and the grinding depth shorter than 0.5 mm.

4) The surface roughness has a dominant but unfavorable influence on the fatigue life of the IC10 workpiece because of stress concentration. On the contrary, the microhardness depth and rate have a favorable influence on the fatigue life, since the case hardening can prevent crack initiation and propagation.

5) To obtain a good finished machining surface and an improved fatigue life for IC10 workpieces,the following creep feed grinding parameters are recommended: the wheel speed∈[15,20] m/s, the feed rate ∈[150, 200]mm/min, and the grinding depth ∈[0.4, 0.5] mm. The corresponding surface and fatigue life are demonstrate:the surface roughness ∈[0.60, 0.65] μm, the microhardening rates ∈[10.4%, 14.2%], the depth of hardened layer ∈[120, 150] μm, and the fatigue life is about ∈[3.58×106, 4.61×106]

However,more work still needs to be done before achieving a deeper understanding of the grindability and engineering applications for this new kind of material which shows such good performance.The material removal mechanism,the chip formation, and the residual stress model will be studied in the near future.

Fig.14 Comparisons of the fatigue fracture(for sample I,vs=10 m/s,vw=200 mm/min,ap=0.5 mm,Ra=0.94 μm,and for sample II, vs=20 m/s, vw=100 mm/min, ap=0.5 mm, Ra=0.44 μm).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This study was supported by NSAF (No. U1830122) and the National Natural Science Foundation of China (No.51775443).