A Comparison Study on Wear Behaviors of Mo and Al 2 O 3-Mo Coatings from RT to 300 ◦ C

Mo and Al2O3-Mo coatings are fabricated on a low-carbon steel substrate using atmospheric plasma spraying. The microstructure and mechanical properties of two as-sprayed coatings, with a particular focus on the tribological behaviors from room temperature to 300 ◦C, are comparatively investigated in this study. Microstructural analysis of the coatings shows that the porosity of the Al2O3-Mo coating is higher than that of Mo coating. The addition of Al2O3 particles reduces the coating–substrate adhesion strength. The Al2O3-Mo coating, in comparison to the Mo coating, shows improved mechanical properties, such as hardness and wear resistance. The friction coefficients of both coatings increase with further increases in test temperatures. The friction coefficient of the Al2O3-Mo coating, tested above 100 ◦C, is lower than that of the Mo coating. The wear failure mechanisms of the two coatings are delamination, brittle fracture, oxidation and adhesion wear. In addition, local plastic deformation was also found in the Mo coating.


Introduction
Molybdenum (Mo) coatings are used to improve underlying materials' mechanical properties and to decrease their wear. They show excellent scuffing resistance to abrasive wear, as well as high corrosion resistance [1][2][3]. The wear resistance of hardened steel coated with a Mo coating could be enhanced by a factor of 2-18 [4]. Because of the excellent tribological properties, Mo coatings are widely used in automobile, aerospace, and pulp and paper industries to protect machine parts from wear and corrosion [5,6].
Thermal spraying is an advantageous deposition process that deposits a Mo layer on various substrate materials. Flame spraying is usually used to fabricate Mo coatings. However, it has a short service lifetime and low cohesive strength [7]. Plasma-sprayed Mo coatings have a low hardness of 300 HV, which results in very poor wear resistance [8]. To address these shortcomings, MoSi 2 , brass, Mo 2 C, NiCrBSi, and TiN were introduced into plasma-sprayed Mo coatings to improve wear resistance [5,[9][10][11]. However, the wear resistance of these Mo matrix composite coatings is also very limited. In other words, new Mo matrix composite coatings, which motivate the current work, should be developed.
The atmospheric plasma spraying (APS) technique endows an Al 2 O 3 coating with high hardness, outstanding corrosion and good wear resistance, and excellent chemical stability under critical application conditions [12]. Therefore, a plasma-sprayed Al 2 O 3 coating on metallic substrates has been widely used as a wear-resistant coating in engineering fields [13,14]. The close coefficient of thermal expansion (CTE) between Mo (6.7 × 10 −6 /K) and Al 2 O 3 (6.8 × 10 −6 /K) could reduce the thermal stress between the two phases boundaries. This leads to fewer micro cracks in the material. Meanwhile, it was reported that Al 2 O 3 /Mo composites, fabricated using powder metallurgy, have excellent toughness and wear resistance [15]. The above-mentioned studies suggested that the Al 2 O 3 -Mo coating could be a good alternative candidate in providing excellent wear resistance.
To reach a more efficient performance, modern engine pistons usually operate under higher temperatures. The temperature of the environment is an important factor in evaluating the tribological characteristics of a piston ring-coating material. However, in previous literature there is almost no research on the wear of plasma-sprayed Mo coatings integrated with Al 2 O 3 particles, especially at room temperature (RT) to 300 • C. In the present research, Mo coatings, with and without reinforced Al 2 O 3 dopant, were fabricated using APS, in an attempt to assess the microstructure, mechanical properties and wear behaviors (RT to 300 • C) of the two coatings. China. An argon-hydrogen mixture was used as the plasma-forming gas. The spraying parameters for Al 2 O 3 -Mo and Mo coatings are the same, and are listed in Table 1. Table 1. Atmospheric plasma spraying parameters.

Items Parameters
Ar flow rate 45 L/min H 2 flow rate 6.5 L/min Arc current 500 A Powder feeding rate 60~65 g/min Spraying distance 110 mm

Microstructures Characterization of Coatings
The phase compositions of the coatings were characterized via X-ray diffraction (XRD) with a Cu-Kα radiation of 0.154 nm in wavelength at a scan speed of 2 • /min (D8 Discover 2500, Bruker Corporation, Karlsruhe, Germany). A scanning electron microscope (SEM, JSM-6380LV, JEOL Ltd., Tokyo, Japan) equipped with an energy dispersive spectroscope (EDS, JEOL Ltd., Tokyo, Japan) was used to analyze the microstructure, morphologies of the coatings, and the wear tracks. The porosities of the coatings were evaluated by image analysis of SEM images (at a magnification of 500×) of the cross-section of the polished samples [16]. For each cross-section, 15 images were measured for average measurement values.

Mechanical Properties Test
Micro-hardness measurements of the coatings were determined on the polished cross-sections using a Digital Micro-Vickers Hardness Tester (HVS-1000, Jinan Liangong Testing Technology Corporation of China, Jinan, China) under a normal load of 50 g and a dwell time of 15 s. Measurements were carried out for each sample at least 20 times to obtain an average value. The adhesion strengths of the coatings were measured using a direct pull-off tensile method according to ASTM C633, using an Instron 3369 electronic universal testing machine [17]. Before the tests, both sides of the coating samples were bonded to the low-carbon steel by using epoxy resin (E-7, provided by Shanghai Research Institute of Synthetic Resins, Shanghai, China). The adhesion strength of the epoxy resin in the C633 test configuration was about 65 MPa. The tensile load was applied at a constant loading rate of 1.0 mm/min until the sample was broken. Adhesion strength can be calculated by dividing the maximum force by the surface zone of the samples. The reported adhesion strength was the mean value of 5 samples sprayed with the same parameters.

Friction and Wear Test
Friction and wear tests were carried out in a ball-on-disk contact configuration tribometer (HSR-2M, Lanzhou, China) under dry conditions. The commercially available steel ball; with a diameter of 5 mm, R a = 0.8 µm, and a hardness (HRC) of 48; was used as the upper specimen. The as-sprayed coating samples were used as the lower specimens. Prior to sliding tests, the as-sprayed coatings were mechanically ground using SiC emery papers to achieve the roughness (R a ) of 1.6 µm. The coatings surface slides against the steel ball in a linear back-and-forth sliding motion. The sliding tests were conducted under a normal load of 30 N, a sliding time of 60 min, a stroke length of 5 mm, a speed of 50 mm/s, and a stroke frequency of 5 Hz. The test temperatures were RT, 100 • C, 200 • C, and 300 • C. The friction coefficients were dynamically recorded by an online automatic measurement. The wear depths and widths as well as the wear volume loss of the coatings were analyzed using a three-dimensional surface profiler (NanoMap-500LS, AEP Technology, Inc., Santa Clara, CA, USA). The wear rate is calculated as: K = V/SF, where K is the volume wear rate (mm 3 /Nm), V is the wear volume (mm 3 ), S is the total sliding distance (m), and F is normal load (N).

Mechanical Properties Test
Micro-hardness measurements of the coatings were determined on the polished cross-sections using a Digital Micro-Vickers Hardness Tester (HVS-1000, Jinan Liangong Testing Technology Corporation of China, Jinan, China) under a normal load of 50 g and a dwell time of 15 s. Measurements were carried out for each sample at least 20 times to obtain an average value. The adhesion strengths of the coatings were measured using a direct pull-off tensile method according to ASTM C633, using an Instron 3369 electronic universal testing machine [17]. Before the tests, both sides of the coating samples were bonded to the low-carbon steel by using epoxy resin (E-7, provided by Shanghai Research Institute of Synthetic Resins, Shanghai, China). The adhesion strength of the epoxy resin in the C633 test configuration was about 65 MPa. The tensile load was applied at a constant loading rate of 1.0 mm/min until the sample was broken. Adhesion strength can be calculated by dividing the maximum force by the surface zone of the samples. The reported adhesion strength was the mean value of 5 samples sprayed with the same parameters.

Friction and Wear Test
Friction and wear tests were carried out in a ball-on-disk contact configuration tribometer (HSR-2M, Lanzhou, China) under dry conditions. The commercially available steel ball; with a diameter of 5 mm, Ra = 0.8 μm, and a hardness (HRC) of 48; was used as the upper specimen. The as-sprayed coating samples were used as the lower specimens. Prior to sliding tests, the as-sprayed coatings were mechanically ground using SiC emery papers to achieve the roughness (Ra) of 1.6 μm. The coatings surface slides against the steel ball in a linear back-and-forth sliding motion. The sliding tests were conducted under a normal load of 30 N, a sliding time of 60 min, a stroke length of 5 mm, a speed of 50 mm/s, and a stroke frequency of 5 Hz. The test temperatures were RT, 100 °C, 200 °C, and 300 °C. The friction coefficients were dynamically recorded by an online automatic measurement. The wear depths and widths as well as the wear volume loss of the coatings were analyzed using a three-dimensional surface profiler (NanoMap-500LS, AEP Technology, Inc., Santa Clara, CA, USA). The wear rate is calculated as: K = V/SF, where K is the volume wear rate (mm 3 /Nm), V is the wear volume (mm 3 ), S is the total sliding distance (m), and F is normal load (N).   XRD patterns of the raw powders, as-sprayed Mo, and Al 2 O 3 -Mo coatings, are shown in Figure 2. The Mo coating is composed of Mo and traces of MoO 3 phases, and the Al 2 O 3 -Mo coating mainly consisted of Mo, Al 2 O 3 and traces of MoO 3 phases. The existence of MoO 3 phase in the two coatings was caused by the oxidation of Mo particles during the APS process [18]. Metal oxides usually reside along the splat boundaries of as-sprayed Mo coatings, and could enhance the hardness and tribological properties of coatings [19].

Microstructure of As-Sprayed Coatings
Lubricants 2018, 6, x FOR PEER REVIEW 4 of 12 XRD patterns of the raw powders, as-sprayed Mo, and Al2O3-Mo coatings, are shown in Figure  2. The Mo coating is composed of Mo and traces of MoO3 phases, and the Al2O3-Mo coating mainly consisted of Mo, Al2O3 and traces of MoO3 phases. The existence of MoO3 phase in the two coatings was caused by the oxidation of Mo particles during the APS process [18]. Metal oxides usually reside along the splat boundaries of as-sprayed Mo coatings, and could enhance the hardness and tribological properties of coatings [19].  Figure 3 shows the surface and cross-section SEM morphologies of two kinds of coatings. As shown in Figure 3a,b, it is clear that the surfaces of both the Mo and Al2O3-Mo coatings consist of well-flattened splats, spherical features, insufficiently-flattened protuberances, and some pores, exhibiting the typical characteristics of plasma spray coatings. Some liquid droplets splashed when they were sprayed on the substrate surface. The splashed liquid particles subsequently changed to small spherical particles due to the effects of surface energy in the cool re-depositing process. Thus, the spherical particles were observed on the surface of the coatings. Figure 3c,d shows the cross-sectional morphologies of the Mo and Al2O3-Mo coatings, respectively. It is obvious that two coatings were stably bonded to the steel substrate without an interfacial decohesion, cracks, or pores. The typical splat and lamellar microstructure, some pores, and micro-cracks were also observed in the two as-sprayed coatings. The inherent characteristics of thermal sprayed technology inevitably result in the existence of microstructure defects in the coating [20]. The porosity proportion of the Mo and Al2O3-Mo coatings, estimated by the image analysis, are about (7.2 ± 0.3)% and (10.2 ± 0.5)%, respectively. This is attributed to the better spreadability of the melt Mo particles than the Al2O3 particles. The addition of Al2O3 particles increased the porosity of the as-sprayed Mo coating, which could influence the mechanical properties of the Mo coating, such as adhesion strengthen, hardness and wear resistance.  Figure 3 shows the surface and cross-section SEM morphologies of two kinds of coatings. As shown in Figure 3a,b, it is clear that the surfaces of both the Mo and Al 2 O 3 -Mo coatings consist of well-flattened splats, spherical features, insufficiently-flattened protuberances, and some pores, exhibiting the typical characteristics of plasma spray coatings. Some liquid droplets splashed when they were sprayed on the substrate surface. The splashed liquid particles subsequently changed to small spherical particles due to the effects of surface energy in the cool re-depositing process. Thus, the spherical particles were observed on the surface of the coatings. Figure 3c,d shows the cross-sectional morphologies of the Mo and Al 2 O 3 -Mo coatings, respectively. It is obvious that two coatings were stably bonded to the steel substrate without an interfacial decohesion, cracks, or pores. The typical splat and lamellar microstructure, some pores, and micro-cracks were also observed in the two as-sprayed coatings. The inherent characteristics of thermal sprayed technology inevitably result in the existence of microstructure defects in the coating [20]. The porosity proportion of the Mo and Al 2 O 3 -Mo coatings, estimated by the image analysis, are about (7.2 ± 0.3)% and (10.2 ± 0.5)%, respectively. This is attributed to the better spreadability of the melt Mo particles than the Al 2 O 3 particles. The addition of Al 2 O 3 particles increased the porosity of the as-sprayed Mo coating, which could influence the mechanical properties of the Mo coating, such as adhesion strengthen, hardness and wear resistance. XRD patterns of the raw powders, as-sprayed Mo, and Al2O3-Mo coatings, are shown in Figure  2. The Mo coating is composed of Mo and traces of MoO3 phases, and the Al2O3-Mo coating mainly consisted of Mo, Al2O3 and traces of MoO3 phases. The existence of MoO3 phase in the two coatings was caused by the oxidation of Mo particles during the APS process [18]. Metal oxides usually reside along the splat boundaries of as-sprayed Mo coatings, and could enhance the hardness and tribological properties of coatings [19].  Figure 3 shows the surface and cross-section SEM morphologies of two kinds of coatings. As shown in Figure 3a,b, it is clear that the surfaces of both the Mo and Al2O3-Mo coatings consist of well-flattened splats, spherical features, insufficiently-flattened protuberances, and some pores, exhibiting the typical characteristics of plasma spray coatings. Some liquid droplets splashed when they were sprayed on the substrate surface. The splashed liquid particles subsequently changed to small spherical particles due to the effects of surface energy in the cool re-depositing process. Thus, the spherical particles were observed on the surface of the coatings. Figure 3c,d shows the cross-sectional morphologies of the Mo and Al2O3-Mo coatings, respectively. It is obvious that two coatings were stably bonded to the steel substrate without an interfacial decohesion, cracks, or pores. The typical splat and lamellar microstructure, some pores, and micro-cracks were also observed in the two as-sprayed coatings. The inherent characteristics of thermal sprayed technology inevitably result in the existence of microstructure defects in the coating [20]. The porosity proportion of the Mo and Al2O3-Mo coatings, estimated by the image analysis, are about (7.2 ± 0.3)% and (10.2 ± 0.5)%, respectively. This is attributed to the better spreadability of the melt Mo particles than the Al2O3 particles. The addition of Al2O3 particles increased the porosity of the as-sprayed Mo coating, which could influence the mechanical properties of the Mo coating, such as adhesion strengthen, hardness and wear resistance.

Adhesion Strength and Micro Hardness of Coatings
The adhesion strengths of the coated specimens were measured, as shown in Figure 4. The adhesion strength of the as-sprayed Mo coating is about 43.8 ± 3.2 MPa, and that of Al2O3-Mo coating is approximately 33.6 ± 2.8 MPa. The average adhesion strength of pure Al2O3 coating, with substrate prepared by APS, is 26.9 ± 1.8 MPa [14]. It is obvious that Al2O3 decreased the adhesion strength of the as-sprayed Mo coating. After the adhesion strength test, the separated surfaces of the test samples revealed that the two kinds of coatings were not cohesively separated from the boundary of the splats but rather from the boundary between the coating and substrate. This shows that the adhesion strength among splats is better than that of the boundary between the coating and substrate. The metal Mo has a good bond performance with other materials [6]. When the Mo powders are used for spraying, the particles may be sufficiently melted and will strongly adhere to the rough surface of the substrate. For the Al2O3-Mo coating, the insufficiently-flattened Al2O3 particles had an adverse influence on the mechanical interlocking between splats or between the coating and substrate. Therefore, the addition of Al2O3 decreased the adhesion strength of Mo coating. The micro-hardness values of both coatings, measured using a standard rectangular pyramid diamond indenter (136 ± 0.5)°, are shown in Figure 5. The micro hardness of the Mo coating is about 3.12 ± 0.12 GPa (HV) and that of Al2O3-Mo coating is about 4.66 ± 0.16 GPa (HV). It is well known that Al2O3 ceramic is an attractive wear-resistant material because of its high hardness, and its elastic modulus (398.2 GPa [21]), which is much higher than that of Mo (310 GPa [22]). In general, the cracks, porosity, and poor adhesion between splats, decrease the micro hardness of the sprayed coatings [23]. However, as-sprayed pure Al2O3 coating has a hardness of about 13 GPa (HV) [14]. With the addition of 30 wt. % Al2O3 particles to the Mo coating, the micro-hardness of the as-sprayed Al2O3-Mo composite coating showed an increase of 49.5%, compared to that of the as-sprayed Mo coating.

Adhesion Strength and Micro Hardness of Coatings
The adhesion strengths of the coated specimens were measured, as shown in Figure 4. The adhesion strength of the as-sprayed Mo coating is about 43.8 ± 3.2 MPa, and that of Al 2 O 3 -Mo coating is approximately 33.6 ± 2.8 MPa. The average adhesion strength of pure Al 2 O 3 coating, with substrate prepared by APS, is 26.9 ± 1.8 MPa [14]. It is obvious that Al 2 O 3 decreased the adhesion strength of the as-sprayed Mo coating. After the adhesion strength test, the separated surfaces of the test samples revealed that the two kinds of coatings were not cohesively separated from the boundary of the splats but rather from the boundary between the coating and substrate. This shows that the adhesion strength among splats is better than that of the boundary between the coating and substrate. The metal Mo has a good bond performance with other materials [6]. When the Mo powders are used for spraying, the particles may be sufficiently melted and will strongly adhere to the rough surface of the substrate.

Adhesion Strength and Micro Hardness of Coatings
The adhesion strengths of the coated specimens were measured, as shown in Figure 4. The adhesion strength of the as-sprayed Mo coating is about 43.8 ± 3.2 MPa, and that of Al2O3-Mo coating is approximately 33.6 ± 2.8 MPa. The average adhesion strength of pure Al2O3 coating, with substrate prepared by APS, is 26.9 ± 1.8 MPa [14]. It is obvious that Al2O3 decreased the adhesion strength of the as-sprayed Mo coating. After the adhesion strength test, the separated surfaces of the test samples revealed that the two kinds of coatings were not cohesively separated from the boundary of the splats but rather from the boundary between the coating and substrate. This shows that the adhesion strength among splats is better than that of the boundary between the coating and substrate. The metal Mo has a good bond performance with other materials [6]. When the Mo powders are used for spraying, the particles may be sufficiently melted and will strongly adhere to the rough surface of the substrate. For the Al2O3-Mo coating, the insufficiently-flattened Al2O3 particles had an adverse influence on the mechanical interlocking between splats or between the coating and substrate. Therefore, the addition of Al2O3 decreased the adhesion strength of Mo coating. The micro-hardness values of both coatings, measured using a standard rectangular pyramid diamond indenter (136 ± 0.5)°, are shown in Figure 5. The micro hardness of the Mo coating is about 3.12 ± 0.12 GPa (HV) and that of Al2O3-Mo coating is about 4.66 ± 0.16 GPa (HV). It is well known that Al2O3 ceramic is an attractive wear-resistant material because of its high hardness, and its elastic modulus (398.2 GPa [21]), which is much higher than that of Mo (310 GPa [22]). In general, the cracks, porosity, and poor adhesion between splats, decrease the micro hardness of the sprayed coatings [23]. However, as-sprayed pure Al2O3 coating has a hardness of about 13 GPa (HV) [14]. With the addition of 30 wt. % Al2O3 particles to the Mo coating, the micro-hardness of the as-sprayed Al2O3-Mo composite coating showed an increase of 49.5%, compared to that of the as-sprayed Mo coating. The micro-hardness values of both coatings, measured using a standard rectangular pyramid diamond indenter (136 ± 0.5) • , are shown in Figure 5. The micro hardness of the Mo coating is about 3.12 ± 0.12 GPa (HV) and that of Al 2 O 3 -Mo coating is about 4.66 ± 0.16 GPa (HV). It is well known that Al 2 O 3 ceramic is an attractive wear-resistant material because of its high hardness, and its elastic modulus (398.2 GPa [21]), which is much higher than that of Mo (310 GPa [22]). In general, the cracks, porosity, and poor adhesion between splats, decrease the micro hardness of the sprayed coatings [23]. However, as-sprayed pure Al 2 O 3 coating has a hardness of about 13 GPa (HV) [14]. With the addition of 30 wt. % Al 2 O 3 particles to the Mo coating, the micro-hardness of the as-sprayed Al 2 O 3 -Mo composite coating showed an increase of 49.5%, compared to that of the as-sprayed Mo coating.

Friction Coefficient and Wear Rate of Coatings
The relation between the friction coefficient and time, and the average friction coefficients of both coatings at different test temperatures, are presented in Figure 6. It can be seen that the coefficients of friction-time curves of two coatings have similar trends. In other words, the friction coefficients are in a relative steady-state stage after a short running-in period. To reduce the measured deviation, the friction coefficient of each specimen was tested three times. As the wear test temperatures increased from RT to 300 °C, the average values of the friction coefficient for the Mo coating increased from 0.52 to 1.14, and that of the Al2O3-Mo coating increased from 0.75 to 1.05. Interestingly, compared to the Mo coating, the Al2O3-Mo coating showed higher average friction coefficients at RT. However, the Al2O3-Mo coating exhibits lower average friction coefficients at the other test temperatures, which indicated that the Al2O3 had a lubricating effect. The lubricating effect of the ceramics was also found in the as-sprayed ZrO2-MoSi2 coating, at a high temperature of 1100 °C [24].

Friction Coefficient and Wear Rate of Coatings
The relation between the friction coefficient and time, and the average friction coefficients of both coatings at different test temperatures, are presented in Figure 6. It can be seen that the coefficients of friction-time curves of two coatings have similar trends. In other words, the friction coefficients are in a relative steady-state stage after a short running-in period. To reduce the measured deviation, the friction coefficient of each specimen was tested three times. As the wear test temperatures increased from RT to 300 • C, the average values of the friction coefficient for the Mo coating increased from 0.52 to 1.14, and that of the Al 2 O 3 -Mo coating increased from 0.75 to 1.05. Interestingly, compared to the Mo coating, the Al 2 O 3 -Mo coating showed higher average friction coefficients at RT. However, the Al 2 O 3 -Mo coating exhibits lower average friction coefficients at the other test temperatures, which indicated that the Al 2 O 3 had a lubricating effect. The lubricating effect of the ceramics was also found in the as-sprayed ZrO 2 -MoSi 2 coating, at a high temperature of 1100 • C [24].

Friction Coefficient and Wear Rate of Coatings
The relation between the friction coefficient and time, and the average friction coefficients of both coatings at different test temperatures, are presented in Figure 6. It can be seen that the coefficients of friction-time curves of two coatings have similar trends. In other words, the friction coefficients are in a relative steady-state stage after a short running-in period. To reduce the measured deviation, the friction coefficient of each specimen was tested three times. As the wear test temperatures increased from RT to 300 °C, the average values of the friction coefficient for the Mo coating increased from 0.52 to 1.14, and that of the Al2O3-Mo coating increased from 0.75 to 1.05. Interestingly, compared to the Mo coating, the Al2O3-Mo coating showed higher average friction coefficients at RT. However, the Al2O3-Mo coating exhibits lower average friction coefficients at the other test temperatures, which indicated that the Al2O3 had a lubricating effect. The lubricating effect of the ceramics was also found in the as-sprayed ZrO2-MoSi2 coating, at a high temperature of 1100 °C [24].   Figure 7 shows the two-dimensional profiles of wear tracks for the Mo and Al 2 O 3 -Mo coatings at a test temperature of 300 • C. The measuring direction of the two-dimensional profile is vertical to the wear tracks. The wear profiles of the coated specimens for other test temperatures are not shown here for the sake of brevity. As shown in Figure 7, it can be seen that the width and depth of the scratches on the Mo coating are approximately 1750 µm and 110 µm, respectively. Whereas for the Al 2 O 3 -Mo coating, they are about 1250 µm and 56 µm, respectively. The profiles of these wear tracks show less wear volume for the Al 2 O 3 -reinforced Mo coating. The wear rates of both coatings at different test temperatures are shown in Figure 8. Their wear rate values at each test temperature are approximately 10 −5 mm 3 /Nm. It can be seen that the wear rates of both coatings increase with the increase in test temperatures. As shown in Figure 8, the wear rate of the as-sprayed Mo-Al 2 O 3 coating is less than that of the Mo coating when they were tested under the same conditions, especially in the test temperature of 100~300 • C. This indicates that the addition of 30 wt. % Al 2 O 3 enhances the wear resistance of Mo coating from RT to 300 • C.
Lubricants 2018, 6, x FOR PEER REVIEW 7 of 12 Figure 7 shows the two-dimensional profiles of wear tracks for the Mo and Al2O3-Mo coatings at a test temperature of 300 °C . The measuring direction of the two-dimensional profile is vertical to the wear tracks. The wear profiles of the coated specimens for other test temperatures are not shown here for the sake of brevity. As shown in Figure 7, it can be seen that the width and depth of the scratches on the Mo coating are approximately 1750 μm and 110 μm, respectively. Whereas for the Al2O3-Mo coating, they are about 1250 μm and 56 μm, respectively. The profiles of these wear tracks show less wear volume for the Al2O3-reinforced Mo coating. The wear rates of both coatings at different test temperatures are shown in Figure 8. Their wear rate values at each test temperature are approximately 10 −5 mm 3 /Nm. It can be seen that the wear rates of both coatings increase with the increase in test temperatures. As shown in Figure 8, the wear rate of the as-sprayed Mo-Al2O3 coating is less than that of the Mo coating when they were tested under the same conditions, especially in the test temperature of 100~300 °C . This indicates that the addition of 30 wt. % Al2O3 enhances the wear resistance of Mo coating from RT to 300 °C.   Figure 9 presents the SEM morphologies of the worn surface of the Mo coating after RT, 100 °C, 200 °C and 300 °C wear tests. A wear scar was commonly observed on the surface of plasma sprayed Mo coating, and the width and depth of the scar increased with the increase in test temperatures. As shown in Figure 9, splat delamination, brittle fracture and local plastic deformation were typically found in all specimens, regardless of wear test temperature. When a repeated load of 30 N was applied on the Mo coating surface, surface fatigue occurred in the coating. A crack easily initiated and gradually propagated below the surface, resulting in decohesion of the splat and delamination. The abrasive wear mechanism, perhaps, occurs at the boundary between splats, where the adhesion strength is weaker than that within the splat,  Figure 7 shows the two-dimensional profiles of wear tracks for the Mo and Al2O3-Mo coatings at a test temperature of 300 °C . The measuring direction of the two-dimensional profile is vertical to the wear tracks. The wear profiles of the coated specimens for other test temperatures are not shown here for the sake of brevity. As shown in Figure 7, it can be seen that the width and depth of the scratches on the Mo coating are approximately 1750 μm and 110 μm, respectively. Whereas for the Al2O3-Mo coating, they are about 1250 μm and 56 μm, respectively. The profiles of these wear tracks show less wear volume for the Al2O3-reinforced Mo coating. The wear rates of both coatings at different test temperatures are shown in Figure 8. Their wear rate values at each test temperature are approximately 10 −5 mm 3 /Nm. It can be seen that the wear rates of both coatings increase with the increase in test temperatures. As shown in Figure 8, the wear rate of the as-sprayed Mo-Al2O3 coating is less than that of the Mo coating when they were tested under the same conditions, especially in the test temperature of 100~300 °C . This indicates that the addition of 30 wt. % Al2O3 enhances the wear resistance of Mo coating from RT to 300 °C.   Figure 9 presents the SEM morphologies of the worn surface of the Mo coating after RT, 100 °C, 200 °C and 300 °C wear tests. A wear scar was commonly observed on the surface of plasma sprayed Mo coating, and the width and depth of the scar increased with the increase in test temperatures. As shown in Figure 9, splat delamination, brittle fracture and local plastic deformation were typically found in all specimens, regardless of wear test temperature. When a repeated load of 30 N was applied on the Mo coating surface, surface fatigue occurred in the coating. A crack easily initiated and gradually propagated below the surface, resulting in decohesion of the splat and delamination. The abrasive wear mechanism, perhaps, occurs at the boundary between splats, where the adhesion strength is weaker than that within the splat,  Figure 9 presents the SEM morphologies of the worn surface of the Mo coating after RT, 100 • C, 200 • C and 300 • C wear tests. A wear scar was commonly observed on the surface of plasma sprayed Mo coating, and the width and depth of the scar increased with the increase in test temperatures. As shown in Figure 9, splat delamination, brittle fracture and local plastic deformation were typically found in all specimens, regardless of wear test temperature. When a repeated load of 30 N was applied on the Mo coating surface, surface fatigue occurred in the coating. A crack easily initiated and gradually propagated below the surface, resulting in decohesion of the splat and delamination.

Wear Mechanisms of Coatings
The abrasive wear mechanism, perhaps, occurs at the boundary between splats, where the adhesion strength is weaker than that within the splat, especially in the coating microstructure by thermal spraying [25,26]. In addition, a load applied to the steel counterpart caused an abrasive effect on the Mo coating. Therefore, plastic deformation zones also usually formed.
Lubricants 2018, 6, x FOR PEER REVIEW 8 of 12 especially in the coating microstructure by thermal spraying [25,26]. In addition, a load applied to the steel counterpart caused an abrasive effect on the Mo coating. Therefore, plastic deformation zones also usually formed. In addition, delamination and brittle fracture of Mo splats usually occurred in the Mo coating. Some pores and cracks existed in the as-sprayed coating. These crack-initiation regions usually endure discontinuous strain or compressive stress during wear tests. Once the stress exceeds the fatigue limit of the Mo coating, during the cyclic contact process, cracks will be nucleated and propagated. Once the cracks connected and coalesced with one another, the particles pulled-out, fatigue delamination took place in these sites, and the brittle fracture was finally developed [25]. The inner cracks within the Mo splat caused the embrittlement of Mo material, which occasionally generated brittle fracture damage. Consequently, the contribution of Mo fracture resulted in degraded tribological property. Moreover, as displayed in Figure 9, splat delamination, brittle fracture and local plastic deformation occurred more and more frequently as wear test temperatures increased. In addition, delamination and brittle fracture of Mo splats usually occurred in the Mo coating. Some pores and cracks existed in the as-sprayed coating. These crack-initiation regions usually endure discontinuous strain or compressive stress during wear tests. Once the stress exceeds the fatigue limit of the Mo coating, during the cyclic contact process, cracks will be nucleated and propagated. Once the cracks connected and coalesced with one another, the particles pulled-out, fatigue delamination took place in these sites, and the brittle fracture was finally developed [25]. The inner cracks within the Mo splat caused the embrittlement of Mo material, which occasionally generated brittle fracture damage.
Consequently, the contribution of Mo fracture resulted in degraded tribological property. Moreover, as displayed in Figure 9, splat delamination, brittle fracture and local plastic deformation occurred more and more frequently as wear test temperatures increased. Figure 10 shows SEM micrographs of the worn surface of the Al 2 O 3 -Mo coating after different temperature wear tests. As seen in Figure 10a-d, the wear scars widths increase with increases in the test temperature. In comparison to the Mo coating, the worn surface of the Al 2 O 3 -Mo coating shows similar characteristics, such as delamination and brittle fracture. However, plastic deformation was hardly found on the worn surface. Under the same wear test conditions, the widths and depths of the wear scar for the Al 2 O 3 -Mo coating were smaller than those observed on the worn surface of the Mo coating, indicating the Al 2 O 3 -Mo coating had a superior wear resistance.
Lubricants 2018, 6, x FOR PEER REVIEW 9 of 12 Figure 10 shows SEM micrographs of the worn surface of the Al2O3-Mo coating after different temperature wear tests. As seen in Figure 10a-d, the wear scars widths increase with increases in the test temperature. In comparison to the Mo coating, the worn surface of the Al2O3-Mo coating shows similar characteristics, such as delamination and brittle fracture. However, plastic deformation was hardly found on the worn surface. Under the same wear test conditions, the widths and depths of the wear scar for the Al2O3-Mo coating were smaller than those observed on the worn surface of the Mo coating, indicating the Al2O3-Mo coating had a superior wear resistance. To further illustrate the wear mechanisms of the coatings, wear tracks were investigated with EDS. The results are listed in Table 2. These results show that the worn surface of the Mo splats of two coatings contained Fe, Mo, and O elements, indicating that oxidation occurred during the wear process. The reciprocating motion caused the increase in temperature on the contact surface, which lead to the oxidation. The presence of the Fe element indicates an adhesion wear between the coating and steel ball. Oxidation and adhesion wears are the main wear mechanisms of the two To further illustrate the wear mechanisms of the coatings, wear tracks were investigated with EDS. The results are listed in Table 2. These results show that the worn surface of the Mo splats of two coatings contained Fe, Mo, and O elements, indicating that oxidation occurred during the wear process. The reciprocating motion caused the increase in temperature on the contact surface, which lead to the oxidation. The presence of the Fe element indicates an adhesion wear between the coating and steel ball. Oxidation and adhesion wears are the main wear mechanisms of the two coatings. Based on the EDS analysis, the adhesion wear of the Mo coating was more severe compared to that of the Al 2 O 3 -Mo coating. In general, the friction and wear behaviors of a material depend on its microstructure, mechanical properties, and environmental conditions. During the wear test at RT, the metal oxide (MoO 3 ), which was formed in the APS process, played an important role in reducing the friction coefficient of the Mo coating. With an increase in the test temperature from RT to 300 • C, more severe decohesion and delamination of the splats appeared on the worn surface, causing the increase in surface roughness of the Mo coating. Meanwhile, the oxidation wear and adhesion wear increased, which resulted in a lot of adhesive metastases on the worn surface. Therefore, the friction coefficients increased with the further increases in test temperature. However, the worn surface of the Al 2 O 3 -Mo coating exhibited less signs of splats delamination, adhesion wear and lower roughness, causing the lower value of the friction coefficients, compared to those of the Mo coating.
The fracture and delamination of splats varied with the wear test temperature. This largely affected the wear rate of the coating. The fracture and delamination of splats became more and more severe (Figures 9 and 10), which resulted in an increase in the wear rates of the coatings. The hardness of a material also affects its wear properties. As mentioned above, strengthening with Al 2 O 3 can provide the Mo coating with high hardness ( Figure 5). The hardness of a steel counterpart is much higher than that of Mo coating. Thus, during dry sliding of the pure Mo coating, the steel counterpart will penetrate into the Mo coating. This results in a large removal of material due to plowing action. In the Al 2 O 3 -Mo coating, the contact interaction between the coating surface and the counterpart is mainly controlled by the strengthening Al 2 O 3 particles. During the wear test, load-bearing Al 2 O 3 particles protrude from the coating surface, which will reduce the direct contact area between the molybdenum part and steel ball and prevent a greater extent of wear.

Conclusions
In this work, the Mo and Al 2 O 3 -Mo coatings were fabricated by APS, using commercial Mo powders and mixed Al 2 O 3 -Mo powders respectively. The microstructure, mechanical properties, as well as the tribological properties from RT to 300 • C were comparably investigated. It is found that the porosity of the Mo coating (7.2 ± 0.4)% is much lower than that of the Al 2 O 3 -Mo coating (10.2 ± 0.5)%. The adhesion strength of Al 2 O 3 -Mo coating is about 33.6 ± 2.8 MPa, which is lower than the Mo coating adhesion strength of 43.8 ± 3.2 MPa. A hardness test showed that the addition of 30 wt. % Al 2 O 3 increased the Mo coating hardness by 49.5%. The Al 2 O 3 -added Mo plasma-sprayed coating shows a lower friction coefficient and better wear resistance compared to the Mo coating. The main wear mechanisms of the two coatings were delamination, brittle fracture, oxidation and adhesion wear, and local plastic deformation, which were especially found in the Mo coating. However, in real application, due to numerous circumstantial factors, the situation could be more complicated.
Therefore, in further research, it is necessary to investigate the optimization of the amount of Al 2 O 3 in order to maximize the mechanical properties under conditions that closely resemble real operation.
Author Contributions: J.Y. performed the experiments in addition to analyzing the data and writing the paper; Y.G. contributed the design of the experiments; Y.W. contributed the microstructure analysis of the samples; P.Z. have contributed with the revision of this paper; J.Q. conducted the analyzing of the experimental data.