High temperature tribological and electrochemical behavior of heat-treated Al2O3-base hybrid coating sprayed by HVOF thermal spray technique

In the present study, the hybrid coating was developed by HVOF over the cast iron substrate. The hybrid coating material comprises Al2O3 (80% w/w) enriched with graphite, MoS2, and fumed silica in w/w of 5%, 10%, and 5% respectively. Microstructure and phase composition of prepared coated samples were analyzed by SEM-EDS, XRD and Raman Spectroscopy. The coated samples were heat-treated and their mechanical, electrochemical and tribological behavior was compared with the as-coated samples. There was a considerable change observed in the micro-hardness and anti-corrosion properties of the coated material after heat treatment. The residual stresses in the heat-treated sample decreased compared to the as-coated sample. The high-temperature tribological investigation was carried out in non-lubricated conditions at three different temperatures 30 °C, 150 °C, and 300 °C with a constant load of 40 N and variable sliding speed. The coefficient of friction and wear rate was calculated and it was found that there was a decrement up to 36% in the wear rate of the heat-treated samples compared to the as-coated sample. The SEM morphologies of the wear track showed the presence of small cracks, adhesion, abrasion and smear regions due to plastic deformation.


Introduction
The conventional materials used for advanced industrial applications have to experience extreme operational conditions. The recent advancement in materials resulted in more substantial bulk properties. The undesirable gradual degradation of the surface due to adverse operating conditions proves to be extremely dangerous for the performance and life of the material. Surface engineering plays a remarkable role in producing highperformance surfaces with enhanced working life and performance. Several surface modification techniques are available for improving the performance of a conventional material [1]. Surface coating is one of the most widely used techniques for surface engineering. Using surface coating, one can enhance the performance of material by deposition of desirable material and achieving the required performance [2,3]. Various methods of surface coatings like thermal spray [4], physical and chemical vapor deposition [5], sol-gel [6], and electro-deposition [7] are available, having specific applications. The tribological applications favor the thermal spray coating due to the high deposition rate and thick and dense coating over various substrate materials [8].
The high density, defect-free coating is preferred for engineering applications. The HVOF thermal spray coating can produce a very low porosity coating of a wide range of coating materials over the substrate up to a thickness of several hundred microns [9]. The flame generated by the fuel in the presence of oxygen carries the coating material in a molten or semi-molten state at a high velocity towards the substrate surface. The high thickness of the coating can be deposited at a swift rate. The deposition has a high density with less porosity resulting in a wear resistance surface with superior mechanical and chemical performance [3].
The researchers have used the HVOF thermal spray coating for the deposition of several coating materials over a wide range of substrates. Recent studies have involved the use of a composite coating to achieve multiple advantages [10][11][12][13]. WC-CoCr deposited using HVOF thermal spray showed the combined effect of the metallic and ceramic coating resulting in corrosion, erosion and wear resistance coating [14,15]. NiCrWMoCuCBFe coating deposited over 316 L SS resulted in higher corrosion resistance. The coating tested for cavitation erosion did not have traces of surface damage, cracks or pores [16]. AlCoCrFeNiTi 0.5 coated over austenitic stainless steel using HVOF thermal spray showed homogenous, BCC phase coating [17]. The metal oxides like TiO 2 [18], Cr 2 O 3 [19], Al 2 O 3 [20] can be deposited over the substrate using HVOF thermal spray resulted in hightemperature corrosion and wear resistance [21][22][23]. Al 2 O 3 coating showed tested in KCl/NaCl environment showed no significant traces of scaling/corrosion upto a temperature of ≈650°C due to the development of KAlO 2 [24][25][26].
In the present research, the hybrid coating was developed on the cast iron substrate. To enhance the tribological performance of cast iron for IC engine applications. The hybrid coating material was prepared with the combination of conventional coating material i.e., Alumina and MoS 2 blended with some environmental waste like fumed silica and carbonized rice husk. The silica fumes have high hardness and very low density resulting in reduced weight of the coating over the substrate. The MoS 2 and graphite gives self-lubricating characteristics to the coating resulting in a lower coefficient of friction. Figure 1 shows the flow chart describing the research carried out.

Preparation of Al 2 O 3 -C-MoS 2 -SiO 2 coating
The cast iron plate having a thickness of 10 mm and cross-section 50×50 mm 2 was used as the substrate material. The casted plate was grit blasted to remove the contamination. The surface was then degreased with acetone and baked at 100°C. Al 2 O 3 powder was used as the base novel hybrid coating material. Tao et al observed that the composition of Al 2 O 3 more than 50% decreased the porosity and resulted in dense coating [27]. Dong et al observed with composite coating with a higher percentage of Al 2 O 3 showed reduced residual stresses and improved toughness [28]. Gupta and Kumar studied the various compositions of Al 2 O 3 -based composite coating. The coating with 3%-6% reinforcement showed enhanced mechanical and erosion characteristics [29]. The MoS 2 , graphite and fumed silica in w/w % of 10%, 5% and 5%, respectively were used as reinforcement to enrich the base coating material. The coating material has been described in table 1. The rice husk is the non-edible environmental waste left after extracting rice from the plant. The graphite powder was produced by carbonized rice husk. The rice husk was carbonized in the furnace in a controlled environment and ball milled to reduce the particle size.
The hybrid coating material was laid over the surface using the HVOF thermal spray technique. The parameters of HVOF coating are given in table 2.

Heat treatment of the as-coated sample
The molten/semi-molten coating material having high kinetic energy hits the substrate. The heat and kinetic energy available with the coating material resulted in residual stresses in the coated samples. These stresses can be relieved by the heat treatment processes. The heat treatment temperature and soaking time depend on the substrate and coating [30,31]. The sample was heat-treated at 550°C for 3 h as the stress-relieving temperature for cast iron ranges from 530°C-580°C and the soaking time is kept around 3 to 3.5 h [32]. The Al 2 O 3 coating heat-treated at 500°C to 700°C for 2 h showed improved characteristics [31]. So the sample was heat-treated in the furnace at 550°C and shocked for 3 h. The sample was cooled in the furnace. The change in morphology and various properties was quantified.

Coating characterization
The coated samples were investigated for morphology and phase analysis. The coated surface was polished using SiC paper of 2000 grit size and 0.5 μm diamond grinding. The microstructure was observed under the optical microscope. The SEM+EDS (JSM IT500) analysis was performed at different magnifications for analyzing the surface of the coated sample.
The phase of coating material was investigated by x-ray diffraction (XRD) (D8, Bruker, USA). The coated sample was placed under the XRD investigator at a 2θ angle range between 10°to 80°. The various phases of the coated sample before and after heat treatment were analyzed by Raman spectroscopy (Renishaw plc UK). The thermal spray-coated specimen may generate residual stresses during the coating operation. This residual stress governs the coating quality and its performance. The residual stresses in the coating were determined by the x-ray method using the residual stress analyzer (PULSTEC μ-X360n). The coated sample was placed under the x-ray and the residual stresses were calculated. The microhardness of the diamond polished samples was measured on Viker's scale under 300gf load with a dwell time of 20 s at 5 different points on the coated samples using the microhardness tester (HM 2000S, Fischer, Germany). The mean value was considered as the average microhardness of the specimen.

Electrochemical and tribological investigation
The electrochemical analysis was carried out at atmospheric temperature using Autolab PG-STAT 302 N, Metrohm, Switzerland electrochemical workstation. The counter electrode of Platinum and a reference electrode of silver/silver chloride. The 100 mm 2 areas of the coated sample were exposed to the H 2 SO 4 bath of pH 6. The voltage was increased gradually to check the breakdown potential. The tribological investigation was carried out on the reciprocating tribometer. Before the tribological investigation, the coated surface was polished using SiC paper of 2000 grit size and 0.5 μm diamond grinding. The cylindrical counter body (pin) of ASTM A36 steel having a diameter of 10 mm and length of 20 mm was used as the counter body. The load of 40 N was applied normally to the pin. The experiment was carried out at three different speeds of 0.16 m s −1 , 0.20 m s −1 and 0.24 m s −1 with a stroke length of 10 mm. The test sample temperature was also varied to 30°C, 150°C and 300°C for the high-temperature tribological analysis. The test run was carried out for a 1500 mm sliding distance. The surfaces were cleaned after tribological investigation and the wear scar was examined under the SEM.

Coating characterization
The coated samples were investigated for the surface characterization under the scanning electron microscope at different resolutions. The SEM microgram shown in figure 2 has a versatile phase of the coated surface. The splats of molten-coated material were observed primarily. There were some voids/pores present also observed along with a few semi-molten particles of Al 2 O 3 . This versatile phase may be observed due to the higher coating thickness and the pining action of molten and semi-molten coating material. The fine grains were observed on the surface. The higher flow rate of carrier gas resulted in quenching of the coated surface; hence the fine grains were formed. The EDS analysis showed the various elements present on the coated surface. More than 80% of the composition was of Al and O as the primary coating material was Al 2 O 3 . The C, Si and Mo were less in composition than Al and O. The composition of Mo was around 6.5% followed by C was around 5%. The Si was also observed as it was added in the form of fumed silica in the coating powder.
The microstructure of the coated sample before and after heat treatment has been shown in figure 2. The microscopic image of the sample was studied by software ImageJ to measure the grain size and the porosity. The number of grains and pores of the heat-treated sample was compared with the as-coated samples. The grains for the as-coated were course, some porosity was observed over the surface. After heat treatment, the sample was observed again as shown in figure 3(b), the was a grain were courser and the porosity has reduced. The porosity of as coated sample was measured as 3.917% which was reduced to 2.734% resulting in an even denser coating. This resulted in better surface properties of heat-treated samples.
Raman spectroscopy was performed on the as-coated and heat-treated samples have been shown in figure 4. For the as-coated samples, the Raman bands corresponding to αAl 2 O 3 showed multiple peaks at the Raman shift of ≈350 and 635. The peaks at Raman shift of ≈840 and 930 correspond to the γ Al 2 O 3 phase. The weak peak at ≈1550 and 1640 Raman shifts correspond to G and D bands of graphite, respectively. There are peaks observed at Raman shift ≈810 and 550 & 1110 corresponding to MoS 2 and fumed silica, respectively.
In the heat-treated sample, a sharp peak was observed at a Raman shift of ≈780 corresponding to the γ Al 2 O 3 phase. Another sharp peak was observed at ≈1330 corresponding to fumed silica.
The peaks for the x-ray Diffraction as-coated samples and the heat-treated sample almost overlap as shown in figure 5. The sharper peaks are observed in the case of the heat-treated sample as it is more crystalline than the as-coated sample. The peaks corresponding to the α-Al 2 O 3 phase for the as-coated sample have been observed at a 2θ angle of ≈26.1, 28.4 53.2 68.1 are broader whereas, in the heat-treated sample, almost all the peaks overlap are sharper. A peak has been observed 2θ angle of ≈30.3 corresponding to γ Al 2 O 3 , the peak at 28.3 angles has been absent which is the indication of conversion of α Al 2 O 3 phase to γ Al 2 O 3 . The other peaks for γ Al 2 O 3 have been observed at a 2θ angle of ≈55.8 and 63.7 for both as-coated and heat-treated samples. A scattered peak corresponding to fumed silica has been observed at 2θ angle of ≈27.8. The sharp peak at 2θ angle of ≈42.4 shows graphite phase whereas at 2θ angle of ≈44.7 corresponds to MoS 2 .
The micro hardness was measured at five different points on the coated and heat-treated samples. Starting from the center of the samples to 50 μm and 100 μm on both sides. The microhardness for the as-coated and heat-treated samples was measured as 960.98 HV and 874.18 HV, with a standard deviation of 299.29 HV and 171.15 HV respectively. The deviation in the microhardness values was ≈1.75 times the mean value for the ascoated samples, whereas the heat-treated samples showed almost similar values when tested as different locations, as shown in figure 6. This resulted from a more uniform grain structure and dense coating after the heat treatment. The heat treatment resulted in better intermolecular cohesion and reduced porosity.
The residual stress was measured using the x-ray method. The x-rays reflected from the work surface were observed. The material was considered to have inter-planar spacing(d) of 1.170 Å. The Debye ring was obtained having a 2D and 3D structure with red, yellow, and blue colors. Red color signifies the maximum (high) residual stress concentration subsequently yellow lower and blue lowest residual stress concentration. The residual stress was determined by calculating the strain from the diffraction peak. The abscissa (X-axis) represents the Cosine of the azimuth angle of the Debye-Sherrer ring and the Y-axis represents the strain(ε α 1). The inclination of the high-stress concentration line was measured for the calculation of residual stress. The distortion ring was not scattered, indicating uniform residual stress throughout the targeted area as shown in figure 7. The calculated value of residual stress at the as-coated specimen was 33 MPa and for the heat-treated sample, it was −16 MPa. The value of residual stress for the heat-treated sample was lesser in magnitude as compared to the as-coated sample. Moreover, the negative residual stress indicates higher bond strength between the coating and the substrate material, resulting in enhanced stability coating.

Tribological investigation of coating
The tribological investigation was carried out for both as-coated and heat-treated samples in non-lubricated conditions. The test was carried out at constant load, different sliding speeds and elevated temperatures.

Study of coefficient of friction
The average value of COF was calculated by dividing the athematic sum of total COF by the total number of data points at which the COF was obtained. The coefficient of friction (COF) had a higher and unstable value at the start of the tribological test. This is due to the contact of fresh asperities of the tribo surfaces. After some cycles the COF values were stable, so the comparison of the performance of various coating was done after neglecting the COF for the first 200 mm of the run. FOR BETTER COMPARISON, the COF vs sliding speed for various experiments have been plotted in a single graph shown in figure 8(a).
The COF for the heat-treated sample had lower values as compared to the as-coated samples as shown in figure 6. The wear track of as-coated samples had traces of adhesive wear, spalling pit and cracks. The spalling pits were not observed in the case of heat-treated samples as shown in figures 9(a) and (b). The heat-treated samples had lower porosity and dense coating resulting in an intact surface. There was a decrease in COF by ≈23% in the case of the heat-treated sample compared to the as-coated sample.
The COF was also affected by the sliding speed but the effect was less as compared to temperature. The plastic deformation of the surface was greater at higher speeds. The unmelted particles that come out from the surface result in the formation of grooves and craters. This resulted in enhanced COF values. The COF was highest in the case of as-coated samples when tested at 0.24 m s −1 and 300°C. The higher speed resulted in the withdrawal of  unmelted particles from the coated surface. This, when combined with the surface porosity, created vacancies on the surface. The higher sliding speed and elevated temperature resulted in slacking of splats and resulted in higher values of COF. The phenomenon continued as the sliding distance increased and resulted in a continuous increase in COF. The heat-treated samples had refined grains. This resulted in less surface porosity and denser coating. Such slacking was there due to the removal of some of the loose particles but the amount was very less compared to the as-coated sample.   The wear rate is inversely proportional to the surface hardness. In the case of the as-coated sample, the wear rate was observed higher than in the heat-treated samples. However, the mean hardness of the as-coated sample was greater. During the tribological investigations, the pores and unmelted particles can be plowed out due to smearing at a greater sliding speed figure 9. The heat-treated sample had lower porosity and a highly dense coating, which resulted in a reduced wear rate. The smearing resulted in the delamination and abrasion at higher sliding speed and temperature, which were the dominating wear mechanisms. The wear rate increased significantly at higher temperatures and sliding speeds. The temperature had more influence on the wear rate compared to sliding speed as shown in figure 10.
The wear track was studied under the SEM for investigation of the wear mechanism as shown in figure 9. There was traces of spalling pits, adhesive wear and cracks on the wear track of as-coated sample under the room temperature tribological study ( figure 9(a)). The heat-treated sample, when examined under similar tribological conditions, has a more intact surface, the smearing of the surface was observed due to high load resulting in plastic deformation of the surface ( figure 9(b)). At elevated temperatures, the as-coated sample had plastic deformation due to smearing of the surface along with the micron size wear debris scattered throughout the wear track as shown in figure 9(c). The wear debris is the cause of the brittleness of the surface. This condition became even worse when the sliding speed increased. The plowing off of the surface and porosity resulted in wider cracks and craters as shown in figure 9(e). These craters resulted in an accelerated wear rate and higher COF values. The heat-treated sample had lesser brittleness of coating hence there was plastic deformation and delamination of the wear track ( figure 7(d)). The abrasive wear and cracks produced due to localized adhesion were observed at higher sliding speed as shown in figure 9(f).

Electrochemical behavior of the coating
The potentiodynamic polarization test was carried out at room temperature at a scanning rate of 1 mV s −1 to examine the electrochemical inertness of the as-coated and heat-treated samples shown in figure 11. The ascoated and heat-treated samples both showed superior resistance to the corrosive environment. The corrosion current density was very low and there was a wide passive region observed which indicated the corrosion resistance of coated samples.
The corrosion current in the case of heat-treated samples and as-coated samples were 25×10 7 A cm −2 and 32×10 7 A cm −2 respectively, whereas the brake down voltage for both was −0.35 V and −0.41 V respectively. The superior values of corrosion current and break-down voltage were due to the dense surface produced after heat treatment. The heat treatment has also reduced the crystalline defects such as grain boundaries, precipitation and segregation. There are favorable sites for the adverse chemical environment to attack. The as- coated sample also had reduced defect and homogenous structure which was further improved by the heat treatment.

Conclusions
• Alumina-based hybrid coating deposited over the cast iron substrate showed the dense amorphous phase of the coated layer.
• The heat treatment of coating was done which resulted in a reduction in microhardness by ≈10%. Still, there was a minor variation in the microhardness sample when tested at different locations compared to the ascoated sample. • The residual stresses in the heat-treated sample had a lesser magnitude compared to the as-coated sample. Moreover, it was compressive compared to the tensile in the case of an as-coated sample resulting in a stable coating.
• The tribological investigation shows the reduction in COF by ≈23% and wears rate by ≈36% for the heattreated sample as compared to the as-coated sample.  • The coating showed excellent corrosion resistance. The chemical stability was further increased by heat treatment.

Data availability statement
The data that support the findings of this study are available upon reasonable request from the authors.