Tangential fretting wear tests were carried out on G20Mn5QT cast steel substrate and 316L cladding coating in ball/plane contact mode using a tangential fretting wear tester. The tangential fretting damage mechanism and damage evolution law under different displacement amplitudes (D=10, 20, 40 μm) at a fixed normal load of 30 N were discussed. The surface phases of the cast steel substrate and 316L cladding coating, the surface morphology of the wear zone, and the chemical elements of the wear scar were analyzed and characterized by using X-ray diffractometer (XRD), scanning electron microscope (SEM), white light interferometer, and electron spectrometer (EDS). The results show that after laser cladding of 316L coating, the Cr-containing hard phases such as Cr0.19Fe0.7Ni0.11 and the uniform metallographic structure generated during the cladding process increase the hardness of the coating surface by 14.3%; when the normal load Fn=30 N, with the increase of displacement amplitude, the micro-motion operation state of the cast steel substrate and the cladding coating gradually changes from the partial slip zone to the mixed zone to the complete slip zone, and the friction coefficient in the stable stage gradually increases, and the degree of wear damage gradually intensifies; in the partial slip zone, the damage mechanism of the cast steel substrate and the cladding coating is adhesive wear, and in the mixed zone and the complete slip zone, the damage mechanism is abrasive wear, delamination and oxidation wear; the damage degree of the 316L cladding coating is lighter than that of the cast steel substrate. In the mixed zone and the slip zone, the wear rate of the 316L cladding layer is reduced by about 4.26% compared with the G20Mn5QT cast steel substrate. 19.1%. Compared with the G20Mn5QT cast steel substrate, the 316L cladding coating exhibits higher anti-fretting wear performance.

As one of the key components connecting the bogie frame and the wheelset, the service stability of the axle box is an important factor affecting the safe operation of the train. The dynamic force between the locomotive wheels and rails causes the axle box to bear alternating loads. Under the action of the loads, the tight fitting surface between the bolts and the gaskets is prone to fretting wear, which seriously affects the service reliability and safety of the axle box. G20Mn5QT cast steel has become a common material for the manufacture of cast steel axle boxes due to its good plasticity and excellent resistance to brittle fracture [3]. At present, scholars mainly focus on the fracture toughness of G20Mn5QT cast steel and the fatigue performance of butt welds [4]. However, research on the friction, wear and protection of G20Mn5QT cast steel is relatively scarce, and reports on fretting wear and related protection are even rarer. In fact, the use of appropriate surface engineering technology can effectively improve the material’s anti-fretting wear performance. Laser cladding, as an advanced material surface modification technology, has the advantages of high laser beam energy, dense organization, high bonding strength between coating and substrate, large particle size and content variation compared with electroplating, thermal spraying and vapor deposition.

316L stainless steel powder (abbreviated as 316L) is a commonly used cladding material with the characteristics of low cost, strong corrosion resistance, high plastic toughness, good formability and weldability. Compared with the substrate, it shows excellent strength, toughness, wear resistance and corrosion resistance. Zhao Fangfang et al. prepared 316L cladding coating on the surface of 45 steel. The coating showed a fine and uniform microstructure and contained uniformly distributed hard points, which greatly improved the microhardness and wear resistance of the coating. Dong Hui et al. studied the effect of heat treatment temperature (650, 700, 750 and 800 °C) on the friction and wear properties of Ni/316L cladding coatings. Heat treatment at different temperatures can reduce the friction coefficient and wear rate of the coating, and the wear resistance of the coating corresponding to 750 °C heat treatment is the best. In addition, a denser and finer grain structure is produced during the laser-induced rapid cooling process, which makes the laser cladding 316L cladding coating show better wear resistance. Majumdar et al. studied the effect of different SiC contents (mass fraction of 5% and 20%) on the performance of 316L cladding coatings. The results showed that the microhardness of the cladding layer was increased by 125% and 400% respectively compared with the substrate, and the wear resistance of the cladding layer was also significantly improved. At present, the performance of 316L cladding coatings prepared on the surfaces of different materials has been studied at home and abroad. However, there is still a lack of research on the micro-tribological properties of 316L cladding coatings prepared on the surface of G20Mn5QT cast steel parts.

In this paper, 316L cladding coatings were prepared on the surface of G20Mn5QT cast steel parts by laser cladding technology, and the tangential micro-motion wear behavior of G20Mn5QT cast steel and its 316L cladding coatings was studied. The tangential micro-motion wear behavior of G20Mn5QT cast steel substrate and 316L cladding coatings was compared and analyzed from the aspects of dynamic response, wear mark micromorphology and tribochemical state of wear area, revealing the strengthening mechanism of 316L cladding coatings to improve the micro-motion wear performance of G20Mn5QT cast steel materials, providing theoretical support for the remanufacturing and repair of cast steel axle boxes, which has important theoretical significance and industrial application value for ensuring the safe service of axle boxes and extending the service life of axle boxes.

1 Experimental part
1.1 Experimental materials
Based on the actual assembly conditions of the shaft box mounting arm and the bolt gasket, the upper friction pair in this paper is Q355E steel ball, the ball sample diameter is 15 mm, and the surface roughness Ra=0.05 μm. The base material of the lower sample is G20Mn5QT cast steel, and the sample is processed into a block sample with a size of 20 mm×10mm×8 mm by wire cutting. Based on the assembly roughness requirements of the shaft box mounting surface and the bolt gasket mating surface, the sample surface is pre-polished with multi-grade sandpaper before the test to make the base material surface roughness (Ra) not exceed 0.1 μm, and then the sample is placed in anhydrous ethanol. After using ultrasonic cleaning to clean the impurities on the sample surface, the sample is blown dry and placed in a constant temperature drying oven for use. The main chemical composition and mechanical properties of the micro-motion friction pair material (Q355E/G20Mn5QT cast steel) are listed in Table 1 and Table 2, respectively.

The experimental preparation method of 316L laser cladding specimens is consistent with that of the substrate specimens. The laser cladding equipment is a RFL-C6000 (Wuhan Ruike Fiber Laser Technology Co., Ltd.) high-power fiber laser, equipped with 3 coaxial nozzles, a powder feeder, and a control cabinet. The shielding gas and carrier gas are both argon. The laser beam wavelength is 1 070 nm, the focal length is 11.6 mm, the scanning mode is linear scanning, and the overlap rate is 40%. The cladding substrate material is a G20Mn5QT cast steel plate with a size of 175 mm×150 mm×30 mm. The cladding material is 316L stainless steel powder prepared by gas physicochemical method, with a powder particle size of 53~150 μm. Before cladding, the plate is placed in ethanol for ultrasonic cleaning, and the powder is placed in a vacuum drying oven at 100 °C for 2 h. The uniformity of the cladding coating depends on factors such as powder feeding rate, laser power and scanning speed. Fan et al. prepared 316L cladding coating on the surface of G20Mn5QT cast steel. By comprehensively comparing the microhardness, tensile properties and wear rate of the cross section, the optimal cladding parameters were obtained: laser power of 2 300 W, scanning speed of 500 mm/min, and powder feeding rate of 14 g/min. The main chemical composition of 316L stainless steel powder is shown in Table 3, and the SEM photo of its morphology is shown in Figure 1.

1.2 Test equipment and methods
All fretting wear tests were conducted on a tangential fretting wear tester independently developed and built by the research team. Its structural principle is shown in Figure 2. During the test, the Q355E steel ball was the upper specimen, fixed on the force transmission arm installed on the upper specimen fixture, and the upper normal load was loaded by the weight. The G20Mn5QT cast steel and its 316L cladding coating were the lower specimens, fixed on the lower specimen fixture. The upper specimen ball fixture remained relatively still compared to the lower module, and the lower specimen block fixture was driven by the voice coil motor to perform tangential reciprocating motion. The motion displacement was transmitted to the computer by a high-precision grating displacement sensor to realize closed-loop control of the displacement. The test load was loaded by a calibrated counterweight, and the tangential force was collected in real time by a piezoelectric force sensor. The tangential displacement and friction force were transmitted to the host computer control software after denoising and filtering, and the Ft-D (friction force-displacement amplitude) curve displayed the fretting operation status in real time.
In order to explore the tangential fretting wear characteristics of G20Mn5QT cast steel and 316L cladding coating under different displacement amplitudes, the normal load Fn=30 N and displacement amplitudes D=10, 20 and 40 μm (referred to as small displacement, medium displacement and large displacement) were selected for tangential fretting wear tests. In order to reduce the test error, all fretting tests were strictly controlled at room temperature of 20~25 ℃. The relative humidity (RH) was controlled at 55%±10%, the reciprocating frequency (f) of the voice coil motor was 5 Hz, the duration was 4 000 s, and the total number of fretting test cycles was 2×104 times. To ensure the accuracy and reliability of the sample data, each group of tests was repeated at least 3 times under the same conditions. The main test parameters are listed in Table 4.

1.3 Characterization and analysis of micro-wear
In this experiment, a microhardness tester (KELITI-000ZB, China) was used to test the microhardness distribution of the 316L cladding coating from the surface to the core; an X-ray diffractometer (XRD, Rigaku Ultima IV, Japan) was used to analyze the physical structure of the G20Mn5QT cast steel substrate and the 316L cladding coating. The test voltage and current were 40 kV and 200 μA, respectively, the target material was Cu target, the diffraction peak half-height width (DS) = diffraction peak full width (SS) = 1°, the diffraction peak integrated intensity RS = 0.3 mm, the scanning rate was 2000 d/min, and the scanning range was 30°~100°. Scanning electron microscope (SEM, JEOL JSM-6610LV, Japan) and matching electron spectrometer (EDS, OXFORD X-MAX50 INCA-250, Japan) were used to observe the surface and cross-sectional morphology of the wear scar and analyze the chemical element composition of the wear scar; white light interferometer (3D profile, Bruker Contour GT, Germany) was used to collect the contour characteristic parameters of the wear scar, and the wear volume, wear area and maximum wear depth of the wear scar were obtained. The wear rate (W) calculation formula is as follows: See formula (1) in the figure, where W is the wear rate; V is the wear volume (mm3); Fn is the normal load (N); N is the number of cycles; D is the total distance of reciprocating micro-motion (m).

2 Results and analysis
2.1 Microstructure and hardness gradient of 316L cladding coating
Figure 3(a) shows the Vickers hardness distribution of 316L cladding coating along the depth direction. The average microhardness of the substrate (Substrate, Sub) is about 210HV0.2, the hardness of the heat affected zone (Heat affected zone, HAZ) is about 300HV0.2, and the hardness of the cladding layer (Cladding layer, CL) is about 240HV0.2, which is 14.3% higher than that of the cast steel substrate. Figure 3(b) shows the XRD spectrum of the phase analysis of the cast steel substrate and the surface of the 316L cladding coating sample. The main phase of the cast steel substrate is γ-Fe, while the main phase of the 316L cladding coating is Cr and Fe-Cr phase metal compounds (Cr0.19Fe0.7Ni0.11). The diffraction peak area shows that the content of Fe-Cr phase metal compounds is less than that of the original γ-Fe phase. The small amount of Cr0.19Fe0.7Ni0.11 new phase precipitated in the 316L cladding coating improves the hardness of the cladding layer. Figure 3(c) shows the metallographic structure of the 316L cladding coating. There are no significant defects such as holes and cracks on the surface of the cladding coating, and a variety of crystal structures such as slender columnar crystals and equiaxed crystals are generated in the cladding layer. The density and uniformity of this organizational structure make the coating hardness significantly improved compared to the substrate.

2.2 Fretting operation characteristics
2.2.1 Ft-D-N curve analysis
The friction force-displacement amplitude (Ft-D) curve can effectively reflect the real-time operation state of the fretting friction pair contact interface and the material response behavior, and is an important parameter for characterizing dynamics. Figure 4 shows the Ft-D-N curves of the G20Mn5QT cast steel substrate and the 316L cladding coating at different displacement amplitudes, and Figure 5 shows the schematic diagram of the starting position and movement direction of the Ft-D-N curve trajectory. As shown in Figure 4, when D=10 μm, the Ft-D curves of the substrate and the cladding layer evolve from the initial flat parallelogram to a straight line until the end of the cycle. The flat parallelogram at the beginning of the cycle is attributed to the protective effect of the oxide film or adsorption film on the friction pair surface, which promotes the relative sliding of the contact surface. With the increase of the number of cycles, the film on the friction pair surface is removed, the upper and lower samples are in direct contact, the friction force increases rapidly, the contact center is in a state of adhesion, and the Ft-D curve finally stabilizes to a straight line, and the micromotion runs in the partial slip regime (PSR). At this time, the friction deformation between the two contact interfaces is mainly coordinated by the elastic deformation of the material; when D=20 μm, the Ft-D curve of the substrate and the cladding layer gradually evolves from the initial flat parallelogram to an elliptical shape, the micromotion state of the interface is still in the center adhesion and edge microslip, the center adhesion area gradually decreases, the local area is already a three-body contact, but the contact between the two bodies still plays a large role, and the micromotion runs in the mixed regime
(Mixed regime, MR), at this time The friction deformation between the two contact interfaces is mainly coordinated by the elastic-plastic deformation of the material; similarly, when D=40 μm, the Ft-D curve of the substrate and the cladding layer The curves all present a regular parallelogram during the cycle. At this time, the micromotion runs in the slip regime (SR) [29-30], and the friction deformation between the two contact interfaces is mainly coordinated by the plastic deformation of the material. Therefore, under the condition of keeping the normal load unchanged, with the increase of displacement amplitude, the micromotion running state change trend of G20Mn5QT cast steel substrate and 316L cladding coating changes from partial slip zone and mixed zone to complete slip zone, and the Ft-D curve changes from a straight line at small displacement to an ellipse at medium displacement and a parallelogram at large displacement.

2.2.2 Friction coefficient curve
The friction coefficient curve can reflect the contact state of the friction pair in the micromotion wear process in real time and is an important dynamic response parameter for evaluating micromotion wear behavior. Figure 6 shows the time-varying friction coefficient curves of G20Mn5QT cast steel substrate and 316L cladding coating under three different displacement amplitudes. When D=10 μm, the time-varying curves of the friction coefficient of the substrate and the cladding layer can be divided into the running-in, rising and stable stages. When the micromotion starts to run, the friction coefficient is small due to the lubrication of the surface film, which is reflected in the Ft-D curve as a parallelogram. As the micromotion continues, the oxide film on the surface of the friction pair is damaged, and the upper and lower samples are in direct contact. The adhesion causes the friction coefficient to increase rapidly and reach the maximum value. The contact edge produces less wear debris due to microslip, but the amount of wear debris produced cannot change the contact state between the interfaces, and it is still the original two-body contact. After the rising stage, the friction coefficient has a slight fluctuation and remains stable. This stage is reflected in the Ft-D curve as a linear type. When stable, the friction coefficients of the substrate and the cladding layer are 0.411 and 0.406, respectively. When D=20 μm, the friction coefficient curve of the substrate goes through five stages: running-in, climbing, stabilization, decline and stabilization. After about 100 cycles, the initial surface film is removed, the direct contact area of ​​the contact pair surface increases, the friction coefficient increases, and the contact area is in a state of adhesion. With the further increase in the number of cycles, the relative slip between the contact interfaces gradually dominates, the formation speed of the third body layer accelerates, resulting in a decrease in the friction coefficient. Finally, the two-body contact and the three-body contact reach a new balance, and the friction coefficient stabilizes again. The friction coefficient curve of the cladding layer presents four typical stages: running-in, climbing, decline and stabilization. The removal of the initial surface film leads to an increase in the friction coefficient. After a certain number of cycles, the two-body contact transforms into a three-body contact and finally reaches a balance. The friction coefficient begins to drop sharply and finally reaches a stable state again. In this stage, the Ft-D curve presents a flat parallelogram at the beginning of operation, transforms into a straight line and then into an ellipse. In the stable stage, the friction coefficients of the substrate and the cladding layer are 0.612 and 0.647, respectively. When the displacement amplitude increases to D=40 μm, the friction coefficient curves of the substrate and the cladding layer show four stages: running-in, climbing, descending and stabilizing. The contact area of ​​the friction pair continues to increase, the wear debris generated at the interface deeply participates in the friction, and the friction coefficient shows an upward trend. With the increase of the micro-motion cycle, the third body layer formed at the contact body interface plays a bearing and lubricating role, and the friction coefficient shows a downward trend. When the generation and discharge of wear debris form a dynamic balance in the continuous process, the friction coefficient curve tends to be stable. At this time, the Ft-D curve is stable in a parallelogram. The friction coefficients of the substrate and the cladding layer in the stable stage are 0.675 and 0.728 respectively. It is worth noting that when the micro-motion runs in the slip zone, the friction coefficient of the cladding layer shows a further upward trend after the running-in and climbing stages. This may be because the wear debris formed by the 316L cladding coating during the micro-motion operation becomes hard particles that hinder the relative movement of the friction pair, which further increases the friction coefficient. By analyzing the micro-motion friction coefficient, it can be seen that the friction coefficient of the 316L cladding coating is higher than that of the G20Mn5QT cast steel substrate, which indicates that it does not have a good friction reduction effect. However, the research on the wear resistance of the 316L cladding coating must rely on a comprehensive evaluation of the detailed characterization of the wear volume and wear rate of the wear surface.

2.3 Wear morphology and wear mechanism analysis
Figures 7 and 8 show the SEM photos of the overall morphology of the wear marks of the G20Mn5QT cast steel substrate and the 316L cladding coating at different displacement amplitudes and the local damage characteristics at a constant normal load Fn=30 N, respectively. It can be seen that with the increase of the micro-motion displacement amplitude, the wear area of ​​the substrate and the cladding surface gradually increases. When the displacement amplitude D=10 μm, the damage morphology of the contact surface between the substrate and the cladding layer presents a typical micro-motion ring, the center of the contact surface is adhered and no obvious damage is observed, the surface grinding marks are clearly visible, and there is a small amount of wear debris accumulation in the edge micro-slip zone. The wear mechanism at this time is mainly adhesive wear. When the displacement amplitude D=20 μm, the damage morphology of the contact surface between the substrate and the cladding layer is elliptical, the substrate wear is more serious, flaky delamination and furrowing appear on the wear scar surface, there is loose accumulation of wear debris particles around, and slight plastic deformation can be observed in the wear scar area. At this time, the surface morphology of the wear scar of the cladding layer is similar to that of the substrate, the delamination of the cladding layer is relatively slight, no obvious plastic deformation is observed on the surface, and the degree of surface wear is lower than that of the substrate. The wear mechanism at this time is mainly adhesive wear, abrasive wear and delamination. When the displacement amplitude D=40 μm, the delamination and furrowing phenomenon of the substrate surface is serious, the wear scar damage is aggravated, and there are a large number of discharged wear debris particles on the edge. In contrast, the wear degree of the cladding layer is lower. The wear debris particles on the coating surface are repeatedly ground and compacted to cover the contact surface during the micro-motion operation. There is no obvious accumulation of wear debris particles on the edge of the wear mark. At this time, the wear mechanism is mainly abrasive wear and delamination. Comparing the surface wear morphology of the cast steel substrate and the 316L cladding coating, it can be seen that under the same micro-motion working conditions, the wear degree of the cladding layer surface is significantly lower. This is attributed to the Cr-containing hard phase generated during the laser cladding process. These hard phases have a significant effect on the improvement of the coating hardness. Due to the higher hardness of the cladding layer, it has better wear resistance and anti-stripping ability than the cast steel substrate, and the damage degree of the cladding layer during micro-motion operation is also less.

In order to explore the oxidation reaction of the wear mark surface during micro-motion operation and further explain its wear mechanism, EDS line scanning analysis was carried out on the cast steel substrate and 316L cladding coating. Figure 9 shows the EDS line scan of the O element in the substrate and cladding layer in three micro-motion operation areas under different displacement amplitudes. The ordinate is the count rate (Counts per second, cps), the abscissa is the scanning distance of the sample surface, and the scanning position is shown in Figures 7 and 8. It can be seen from Figure 9 that when the micro-motion is running in the partial slip zone, the O element count rate of the substrate and cladding layer in the unworn area has no obvious fluctuation, and slight oxidation occurs in the micro-slip area at the edge of the wear scar. When the micro-motion is running in the mixed zone and the slip zone, the O element count rate fluctuates greatly along the scanning direction, the O element content in the wear area increases, and the O element content in the middle of the wear scar is slightly lower than that at the edge. At this time, oxidation reaction occurs at the friction interface. In addition, the increase in the O element content in the unworn area is due to the fact that the larger displacement amplitude promotes the discharge of the third body layer, causing the wear debris to accumulate at the edge of the wear scar, and the fine wear debris particles at the edge are fully in contact with the O element, resulting in an increase in the O content. Therefore, when the micromotion runs in the mixed zone and the complete slip zone, the wear mechanism includes oxidation wear.

2.4 Analysis of wear scar morphology parameters and wear volume
Figure 10 shows the three-dimensional morphology information of the G20Mn5QT cast steel substrate and the 316L cladding coating under different displacement amplitudes. It can be seen from Figures 10(a) and (b) that when the normal load is constant, the surface wear area of ​​the cast steel substrate and the 316L cladding coating increases significantly with the increase of the micromotion displacement amplitude. When the micromotion runs in the partial slip zone, the substrate and the cladding layer are slightly damaged by the wear scar. The center of the wear scar has a slight indentation due to the normal load, and the edge has microslip and a small amount of wear debris accumulation. The maximum wear depths are 0.94 and 0.53 μm, respectively. When the micromotion is running in the mixed zone, the wear area of ​​the wear scar increases, and there is obvious accumulation of wear debris around it. At this time, the damage area of ​​the substrate is significantly larger than that of the cladding layer, and the wear depths are 4.26 and 1.26 μm, respectively. When the micromotion is running in the full slip zone, the wear scar is severely damaged, and there are pits on the surface of the wear scar due to flaking, and the surface and edge are scattered with wear debris particles. The damage area of ​​the substrate is larger than that of the cladding layer, and the wear depth is also significantly different, 19.05 and 5.75 μm, respectively. It is worth noting that when the micromotion is running in the full slip zone, a slight material accumulation phenomenon occurs on the surface of the cladding layer, which may be caused by the high hardness of the cladding layer, which causes the friction pair Q355E steel ball to fall off and peel off during the micromotion operation and be compacted on the surface. Figure 10(c) shows the wear volume and wear rate of the G20Mn5QT cast steel substrate and the 316L cladding coating. It can be seen from Figure 10(c) that when the micromotion is running in the partial slip zone, the wear of the substrate and the cladding layer is relatively slight. As the displacement amplitude increases, the wear volume of the substrate and the cladding layer also increases. When the micromotion is running in the mixed zone and the complete slip zone, the wear volume and wear rate of the cladding layer are significantly lower than that of the substrate, and the wear rate is reduced by about 4.26% and 19.1%, respectively. Through the analysis of the three-dimensional morphology and two-dimensional profile characteristics of the wear marks of G20Mn5QT cast steel and 316L cladding layer, it can be found that the damage degree of the cladding layer is significantly lower than that of the substrate. Although the 316L cladding layer did not show a better friction reduction effect than the substrate during the micromotion reciprocating process, its uniform organizational characteristics and high hardness make it have excellent wear resistance.

2.5 Cross-sectional morphology analysis
Figure 11 shows schematic diagrams of SEM photos of cross-sectional damage morphology of G20Mn5QT cast steel substrate and 316L cladding coating under different displacement amplitudes. As shown in Figure 11, under a fixed normal load, when the displacement amplitude is D=20 and 40 μm, obvious delamination and wear pits can be observed in the cross-sectional surface of the substrate wear scar, as shown in Figure 11(a), indicating that the existence of delamination causes the material to detach from the surface and form delamination pits, which aggravates the material damage. With the increase of displacement amplitude, the degree of substrate material delamination increases, the surface of the wear scar is scattered with delamination flaky wear debris, and the depth of the wear pit increases, as shown in Figure 11(b). Similar to the substrate, when the displacement amplitude is D=20 μm, delamination and wear pits can be observed in the wear scar profile, but the degree of material delamination is low at this time. When the displacement amplitude increases to D=40 μm, the profile morphology has obvious material delamination characteristics and the wear pit depth increases. The surface is scattered with wear debris layers, and the degree of material wear increases, as shown in Figure 11 (c) and (d). Through the analysis of the profile morphology of the wear area, it can be seen that with the continuous increase of the displacement amplitude, the material delamination characteristics of the cast steel substrate and the cladding layer become more obvious, and the wear pit depth gradually increases. Compared with the cast steel substrate, the 316L cladding coating has a lower degree of material delamination during the micro-wear process and the wear pit depth caused by delamination is shallower, indicating that the 316L cladding coating has better anti-stripping performance than the cast steel substrate.

2.6 Analysis of wear mechanism
Figure 12 shows the schematic diagram of the fretting wear damage of the G20Mn5QT cast steel substrate and 316L cladding coating in different fretting areas. When the fretting is in the partial slip zone, the G20Mn5QT cast steel substrate and cladding layer are slightly worn, the center of the wear mark is adhered, and a small amount of wear debris particles accumulate at the edge, as shown in Figure 12(a) and (a1). The wear mechanism is mainly adhesive wear. When the fretting is in the mixed zone and the slip zone, the surface of the substrate and cladding layer has obvious plowing and peeling phenomena, and there is loose accumulation of wear debris particles at the edge of the substrate wear mark, as shown in Figure 12(b); with the increase of displacement amplitude, the peeled wear debris particles are compacted and covered on the wear mark surface under the larger fretting displacement, and some of the discharged wear debris particles form wear debris particles accumulation at the edge of the wear mark, as shown in Figure 12(c). Due to the high hardness of the cladding layer surface, its wear degree is lower than that of the substrate. As shown in Figure 12(b1), there is a flaky wear layer formed by material peeling in the wear scar area, and there is less accumulation of wear debris particles at the edge. As the displacement amplitude increases, the flaky wear layer in the center of the cladding layer wear scar is compacted, and no obvious wear debris particles accumulate at the edge, as shown in Figure 12(c1). At this time, the wear mechanism is mainly abrasive wear, delamination and oxidation wear.

3 Conclusion
In this article, three groups of typical fretting displacement amplitudes in different areas were selected to discuss the anti-fretting damage characteristics of G20Mn5QT cast steel substrate and 316L cladding coating. The conclusions are as follows:
a. After laser cladding of 316L coating, the Cr-containing hard phases such as Cr0.19Fe0.7Ni0.11 generated during the cladding process and the uniform metallographic structure increase the hardness of the coating surface by 14.3%.
b. As the displacement amplitude increases, the fretting operating state of the G20Mn5QT cast steel matrix and cladding layer gradually changes from the partial slip zone to the complete slip zone, the material damage intensifies, and the friction coefficient in the stable stage gradually increases. In the mixing zone and slip zone, the wear rate of the 316L cladding layer is reduced by approximately 4.26% and 19.1% respectively compared with the G20Mn5QT cast steel substrate. The 316L cladding coating shows excellent anti-fretting wear performance.
c. When the fretting operation is in the partial slip zone, the center of the contact area is sticky and the edges are slightly slippery, and the wear mechanism is adhesive wear; when the fretting operation is in the mixed zone, the characteristics of mechanical delamination and abrasive wear gradually become apparent, and the depth of oxidation wear Affects the wear behavior, and the wear mechanism also includes adhesive wear; when fretting runs in the slip zone, the damage characteristics are mainly plowing and delamination, and severe friction and oxidation reactions occur at the contact interface. The wear mechanisms are delamination, abrasive wear and Oxidative wear.