Wear and Corrosion Resistance of Al0.5CoCrCuFeNi High-Entropy Alloy Coating Deposited on AZ91D Magnesium Alloy by Laser Cladding

In order to improve the wear and corrosion resistance of an AZ91D magnesium alloy substrate, an Al0.5CoCrCuFeNi high-entropy alloy coating was successfully prepared on an AZ91D magnesium alloy surface by laser cladding using mixed elemental powders. Optical microscopy (OM), scanning electron microscopy (SEM), and X-ray diffraction were used to characterize the microstructure of the coating. The wear resistance and corrosion resistance of the coating were evaluated by dry sliding wear and potentiodynamic polarization curve test methods, respectively. The results show that the coating was composed of a simple FCC solid solution phase with a microhardness about 3.7 times higher than that of the AZ91D matrix and even higher than that of the same high-entropy alloy prepared by an arc melting method. The coating had better wear resistance than the AZ91D matrix, and the wear rate was about 2.5 times lower than that of the AZ91D matrix. Moreover, the main wear mechanisms of the coating and the AZ91D matrix were different. The former was abrasive wear and the latter was adhesive wear. The corrosion resistance of the coating was also better than that of the AZ91D matrix because the corrosion potential of the former was more positive and the corrosion current was smaller.


Introduction
High-entropy alloys are a new kind of alloy with excellent properties such as good wear resistance, excellent corrosion resistance, excellent oxidation resistance, low electrical conductivity, low thermal conductivity, and low coefficient of thermal expansion; they were invented by Yeh in 1995 [1][2][3][4][5][6][7]. They are composed of five or more elements in the same or an approximately equal molar ratio and have simple BCC and/or FCC solid solution phases. Their microstructure and properties are different from those of traditional alloys such as Fe-and Ni-based alloys or intermetallic compounds such as Ti-Al, Ni-Al, and Fe-Al compounds; this is because of the high entropy effect, severe lattice distortion effect, sluggish diffusion effect, and cocktail effect of the former [5,7].
It is well known that magnesium and its alloys have been widely used in many fields such as the automotive, communication, and aerospace industries, but their poor wear resistance and corrosion resistance hinder their application in those situations where these properties are required. To solve this problem, a method called laser cladding has been developed to improve the wear and corrosion resistance of magnesium and its alloys, and numerous literature reports have confirmed this result [33,34]. Based on the excellent wear resistance and corrosion resistance of high-entropy alloys [1][2][3][4][5][6][7] and on the basis of an author's previous research [32], we decided to continue the preparation of new wear-resistant and corrosion-resistant high-entropy alloy coatings by laser cladding on an AZ91D magnesium alloy substrate, which is widely used. To our best knowledge, no studies have reported an Al 0.5 CoCrCuFeNi high-entropy alloy coating fabricated by laser cladding on an AZ91D magnesium alloy substrate. The reason for studying the Al 0.5 CoCrCuFeNi alloy is that it has a simple FCC solid solution phase and exhibits excellent wear resistance due to its large work-hardening capacity [35].
In this paper, mixed powders of Cu, Ni, Al, Co, Cr, and Fe were used to fabricate an Al 0.5 CoCrCuFeNi high-entropy alloy coating on AZ91D magnesium alloys by laser cladding. The microstructure, wear behavior, and corrosion behavior of the coating are shown in detail.

Experimental
Laser cladding of mixed powders of Cu, Ni, Al, Co, Cr, and Fe was undertaken on an AZ91D magnesium alloy (Guangdong Dongguan Jubao Magnesium Alloy Material Co., Ltd, Dongguan, China). The sample size used in the laser cladding was 100 mm × 50 mm × 10 mm. The powders (Hebei Xingtai Nangong Zhongzhou Alloy Material Co., Ltd, Xingtai, China) had particle sizes of 48-75 µm and purity of 99 wt %. The different element powders were mixed by ball milling (DECO-PBM-V-0.4L, Hunan Changsha Deco Instrument Equipment Co., Ltd, Changsha, China) according to the nominal composition of the Al 0.5 CoCrCuFeNi high-entropy alloy. The mixed powder from the ball milling was preset on the surface of the AZ91D magnesium alloy using 4 vol % PVA solution. The thickness of the preset layer was about 0.6 mm. The laser cladding experiment was completed under the protection of 99.999% high-purity argon by using a TR050 type CO 2 high-power laser. The optimized laser cladding parameters were as follows: laser power was 3000 W, laser scanning speed was 10 mm/s, laser spot diameter was 4 mm, and laser spot overlap rate was 25%.
The phase structure and microstructure of the coating after laser cladding were identified and observed using an X-ray diffractometer (X' Pert PRO), an optical microscope (Axiovert 200MAT), and a scanning microscope (Quanta 400) with an energy spectrum, respectively. The microhardness of the coating cross section was measured using a microhardness tester (MICROMET 3). The loading force was 100 grams and the loading time was 15 s.
The wear resistance of the coating was evaluated by the dry sliding wear method, in which the size of the sample used for testing was 10 mm × 10 mm × 10 mm, and the friction pair used for matching was bearing steel (AISI52100, HV 0.1 700). Figure 1 shows a schematic illustration of the block-on-ring sliding wear tester. The parameters for the dry sliding wear were as follows: the applied load was 98 N, the dry sliding speed was 0.4187 m/s, the dry sliding wear time was 75 mins, and the Entropy 2018, 20, 915 3 of 10 dry sliding wear distance was 1884 m. The AZ91D magnesium alloy was selected as the experimental material for the dry sliding wear comparison. For each experimental datapoint, the average value of three experimental results was taken as the final data. Wear weight loss was measured using an electronic analytical balance (Bartorius BS110) with an accuracy of 0.1 mg.
The corrosion resistance of the coating was evaluated by testing the corrosion potential polarization curve of a 3.5 wt % sodium chloride solution. A saturated calomel electrode (SCE) was used as the reference electrode and platinum was used as the counter electrode. The test was performed from −2.5 V to 1.5 V with a scanning speed of 1 mV/s. The surface morphologies of the worn and corroded specimens were observed using a scanning microscope (Quanta 400) with an energy spectrum.
3 75 mins, and the dry sliding wear distance was 1884 m. The AZ91D magnesium alloy was selected as the experimental material for the dry sliding wear comparison. For each experimental datapoint, the average value of three experimental results was taken as the final data. Wear weight loss was measured using an electronic analytical balance (Bartorius BS110) with an accuracy of 0.1 mg.
The corrosion resistance of the coating was evaluated by testing the corrosion potential polarization curve of a 3.5 wt % sodium chloride solution. A saturated calomel electrode (SCE) was used as the reference electrode and platinum was used as the counter electrode. The test was performed from −2.5 V to 1.5 V with a scanning speed of 1 mV/s. The surface morphologies of the worn and corroded specimens were observed using a scanning microscope (Quanta 400) with an energy spectrum.  Figure 2a shows an optical image of the cross section of the coating/AZ91D substrate sample after laser cladding. It can be seen from Figure 2a that there were basically no defects such as microcracks or pores in the coating. The interface between coating and substrate indicates that they were well combined. In addition, small pieces of AZ91D magnesium alloy substrate were also found in the coating (Figure 2a). This is believed to be due to some partially melted Mg being detached from the substrate and becoming trapped within the rapidly solidifying coating. Figure 2b shows the scanning electron microscopy morphology of position A in Figure 2a. It can be seen from Figure 2b that the coating has typical dendrite microstructure characteristics, and the chemical compositions of the dendrite (DR) and inter-dendrite (ID) are given in Table 1. As shown in Table 1, copper segregates significantly in the inter-dendrite region. In other words, local segregation of copper occurred in the Al0.5CoCrCuFeNi high-entropy alloy coating prepared by laser cladding. These results are similar to the microstructure characteristics of Al0.5CoCrCuFeNi high-entropy alloys prepared by arc melting in the literature [35][36][37][38]. The difference is that the dendrite microstructure of the laser cladding coating is fine due to the rapid heating, melting, and rapid solidification of the preset layer during laser cladding.   Figure 2a shows an optical image of the cross section of the coating/AZ91D substrate sample after laser cladding. It can be seen from Figure 2a that there were basically no defects such as microcracks or pores in the coating. The interface between coating and substrate indicates that they were well combined. In addition, small pieces of AZ91D magnesium alloy substrate were also found in the coating (Figure 2a). This is believed to be due to some partially melted Mg being detached from the substrate and becoming trapped within the rapidly solidifying coating. Figure 2b shows the scanning electron microscopy morphology of position A in Figure 2a. It can be seen from Figure 2b that the coating has typical dendrite microstructure characteristics, and the chemical compositions of the dendrite (DR) and inter-dendrite (ID) are given in Table 1. As shown in Table 1, copper segregates significantly in the inter-dendrite region. In other words, local segregation of copper occurred in the Al 0.5 CoCrCuFeNi high-entropy alloy coating prepared by laser cladding. These results are similar to the microstructure characteristics of Al 0.5 CoCrCuFeNi high-entropy alloys prepared by arc melting in the literature [35][36][37][38]. The difference is that the dendrite microstructure of the laser cladding coating is fine due to the rapid heating, melting, and rapid solidification of the preset layer during laser cladding. The copper segregation in the Al 0.5 CoCrCuFeNi high-entropy alloy can be explained by the mixing enthalpies [39] of different atomic pairs in Table 2. As shown in Table 2, all of the mixing enthalpies of copper with iron, chromium, cobalt, and nickel are positive. This means that copper has a low affinity for these atoms and is easily repelled by them. In the process of laser cladding, different elements in the Al 0.5 CoCrCuFeNi high-entropy alloy were mixed evenly due to the convection and agitation of the laser cladding pool. However, during the cooling and solidification stage, copper was excluded from the dendrites due to the low affinity of copper atoms with iron, chromium, cobalt, and nickel; thus, copper segregation occurred at the inter-dendrites. The copper segregation in the Al0.5CoCrCuFeNi high-entropy alloy can be explained by the mixing enthalpies [39] of different atomic pairs in Table 2. As shown in Table 2, all of the mixing enthalpies of copper with iron, chromium, cobalt, and nickel are positive. This means that copper has a low affinity for these atoms and is easily repelled by them. In the process of laser cladding, different elements in the Al0.5CoCrCuFeNi high-entropy alloy were mixed evenly due to the convection and agitation of the laser cladding pool. However, during the cooling and solidification stage, copper was excluded from the dendrites due to the low affinity of copper atoms with iron, chromium, cobalt, and nickel; thus, copper segregation occurred at the inter-dendrites.  Figure 3 presents the XRD patterns of the Al0.5CoCrCuFeNi high-entropy alloy coating prepared by laser cladding. The analytical result confirmed that the phase was a simple FCC solid solution in the laser-clad coating. This result proved that the Al0.5CoCrCuFeNi high-entropy alloy coating was successfully fabricated by laser cladding on the AZ91D magnesium alloy based on the mixed powders of Cu, Ni, Al, Co, Cr, and Fe of Al0.5CoCrCuFeNi and the optimized laser cladding parameters. This result is consistent with the results reported in the literature [35][36][37][38]. Thus, for the laser-clad Al0.5CoCrCuFeNi high-entropy alloy coating, both the dendrites and Cu-rich interdendrites were of one simple FCC phase (Figure 1b). The formation of a simple solid solution phase rather than intermetallic compounds is mainly attributed to the significant lowering of free energy by the high enthalpy of mixing [40]. Figure 4 shows the microhardness of the laser-clad Al0.5CoCrCuFeNi high-entropy alloy coating. As can be seen from Figure 4, the microhardness of the coating was HV0.1365-about 3.7 times of the AZ91D matrix (HV0.198) and higher than that of the Al0.5CoCrCuFeNi high-entropy alloy (HV5233) prepared by arc melting [35][36][37]. The high microhardness of the laser-clad coating was attributed to the solid solution strengthening of different alloy elements and fine grain strengthening of the fine dendritic structure during the cooling process. Obviously, the high microhardness of the coating is more beneficial to increasing the wear resistance of the coating.   Figure 3 presents the XRD patterns of the Al 0.5 CoCrCuFeNi high-entropy alloy coating prepared by laser cladding. The analytical result confirmed that the phase was a simple FCC solid solution in the laser-clad coating. This result proved that the Al 0.5 CoCrCuFeNi high-entropy alloy coating was successfully fabricated by laser cladding on the AZ91D magnesium alloy based on the mixed powders of Cu, Ni, Al, Co, Cr, and Fe of Al 0.5 CoCrCuFeNi and the optimized laser cladding parameters. This result is consistent with the results reported in the literature [35][36][37][38]. Thus, for the laser-clad Al 0.5 CoCrCuFeNi high-entropy alloy coating, both the dendrites and Cu-rich inter-dendrites were of one simple FCC phase (Figure 1b). The formation of a simple solid solution phase rather than intermetallic compounds is mainly attributed to the significant lowering of free energy by the high enthalpy of mixing [40].    Figure 4 shows the microhardness of the laser-clad Al 0.5 CoCrCuFeNi high-entropy alloy coating. As can be seen from Figure 4, the microhardness of the coating was HV 0.1 365-about 3.7 times of the AZ91D matrix (HV 0.1 98) and higher than that of the Al 0.5 CoCrCuFeNi high-entropy alloy (HV 5 233) prepared by arc melting [35][36][37]. The high microhardness of the laser-clad coating was attributed to the solid solution strengthening of different alloy elements and fine grain strengthening of the fine dendritic structure during the cooling process. Obviously, the high microhardness of the coating is more beneficial to increasing the wear resistance of the coating.   Figure 5 shows the dry sliding wear weight loss of different samples. According to Figure 5, the wear resistance of the coating was better than that of the AZ91D matrix, and the wear weight loss was about 2.5 times smaller than that of the AZ91D matrix. This is in accordance with their corresponding microhardness values (Figure 4). In other words, the hardness is higher and the wearability is better. Figure 6 shows the surface morphologies of different samples after dry sliding wear. In Figure 6, both of them present obvious groove characteristics. However, the wear mechanisms of the two samples were also significantly different. The coating showed obvious abrasive wear characteristics, and the AZ91D matrix showed obvious adhesive wear characteristics. Some elongated pockmarks and microcracks (Figure 6a) can be found on the worn surface of the AZ91D matrix. Table 3 shows the EDS results of different positions of different worn samples. It can be seen from Table 3 that AZ91D matrix position 1 and coating position 1 are each matrix materials, although each has a small amount of oxidation due to friction heating. Both AZ91D matrix position 2 and coating position 2 are iron oxide particles, but there are a lot of fine particles on the worn surface of the coating, while the worn surface of the AZ91D matrix is much cleaner.   Figure 5 shows the dry sliding wear weight loss of different samples. According to Figure 5, the wear resistance of the coating was better than that of the AZ91D matrix, and the wear weight loss was about 2.5 times smaller than that of the AZ91D matrix. This is in accordance with their corresponding microhardness values (Figure 4). In other words, the hardness is higher and the wearability is better. Figure 6 shows the surface morphologies of different samples after dry sliding wear. In Figure 6, both of them present obvious groove characteristics. However, the wear mechanisms of the two samples were also significantly different. The coating showed obvious abrasive wear characteristics, and the AZ91D matrix showed obvious adhesive wear characteristics. Some elongated pockmarks and microcracks (Figure 6a) can be found on the worn surface of the AZ91D matrix. Table 3 shows the EDS results of different positions of different worn samples. It can be seen from Table 3 that AZ91D matrix position 1 and coating position 1 are each matrix materials, although each has a small amount of oxidation due to friction heating. Both AZ91D matrix position 2 and coating position 2 are iron oxide particles, but there are a lot of fine particles on the worn surface of the coating, while the worn surface of the AZ91D matrix is much cleaner.

Wear Properties
The causes of the above phenomena should be related to the hardness of the respective matrix materials. Figure 4 shows that the microhardness of the coating was about HV0.1365, while the microhardness of the AZ91D matrix was only HV0.198. In addition, the coating is composed of Al0.5CoCrCuFeNi high-entropy alloy, and Al0.5CoCrCuFeNi high-entropy alloy has superior workhardening characteristics [35], but the AZ91D matrix does not have the same excellent workhardening capability. As a result, the microhardness of the coating continued to increase when dry sliding wear was applied to the bearing steel friction pair and may even approach or exceed the hardness of the bearing steel friction pair; it would then shear off the steel ring material, and iron wear debris would be smeared over the coating surface. With the extension of the dry sliding wear time, the iron wear debris would be oxidized due to heat generated by friction. Therefore, iron oxide abrasive particles will be used as an abrasive cutting the Al0.5CoCrCuFeNi high-entropy alloy coating. In contrast, the AZ91D matrix did not have this process and only the soft AZ91D matrix was transferred to the hard bearing steel surface, so its wear rate (30.38 μg/s) was much higher than that (11.93 μg/s) of the coating.   Figure 7 and Table 4 show the potentiodynamic polarization curves and corrosion properties, respectively, of different samples in 3.5 wt % sodium chloride solution. According to the corrosion electrochemical behavior of the two samples and the lack of a passivating region in Figure 7, both the AZ91D matrix and the coating are active dissolved materials. Therefore, using the principle of "the smaller the corrosion current and the higher the corrosion potential, the better the corrosion  The causes of the above phenomena should be related to the hardness of the respective matrix materials. Figure 4 shows that the microhardness of the coating was about HV 0.1 365, while the microhardness of the AZ91D matrix was only HV 0.1 98. In addition, the coating is composed of Al 0.5 CoCrCuFeNi high-entropy alloy, and Al 0.5 CoCrCuFeNi high-entropy alloy has superior work-hardening characteristics [35], but the AZ91D matrix does not have the same excellent work-hardening capability. As a result, the microhardness of the coating continued to increase when dry sliding wear was applied to the bearing steel friction pair and may even approach or exceed the hardness of the bearing steel friction pair; it would then shear off the steel ring material, and iron wear debris would be smeared over the coating surface. With the extension of the dry sliding wear time, the iron wear debris would be oxidized due to heat generated by friction. Therefore, iron oxide abrasive particles will be used as an abrasive cutting the Al 0.5 CoCrCuFeNi high-entropy alloy coating. In contrast, the AZ91D matrix did not have this process and only the soft AZ91D matrix was transferred to the hard bearing steel surface, so its wear rate (30.38 µg/s) was much higher than that (11.93 µg/s) of the coating.  Table 4 show the potentiodynamic polarization curves and corrosion properties, respectively, of different samples in 3.5 wt % sodium chloride solution. According to the corrosion electrochemical behavior of the two samples and the lack of a passivating region in Figure 7, both the AZ91D matrix and the coating are active dissolved materials. Therefore, using the principle of "the smaller the corrosion current and the higher the corrosion potential, the better the corrosion resistance of the material" to evaluate the corrosion resistance of the active dissolved material, it can be seen that the corrosion resistance of the Al 0.5 CoCrCuFeNi high-entropy alloy coating prepared by laser cladding was better than that of the AZ91D matrix. This is reflected by the fact that the former had a significantly higher corrosion potential (E corr = −0.998 V) than did the latter (E corr = −1.46 V), Entropy 2018, 20, 915 7 of 10 and the former also exhibited a lower corrosion current (i corr = 1.60 × 10 −4 A/cm 2 ) than did the latter (i corr = 6.20 × 10 −4 A/cm 2 ). 7 resistance of the material" to evaluate the corrosion resistance of the active dissolved material, it can be seen that the corrosion resistance of the Al0.5CoCrCuFeNi high-entropy alloy coating prepared by laser cladding was better than that of the AZ91D matrix. This is reflected by the fact that the former had a significantly higher corrosion potential (Ecorr = −0.998 V) than did the latter (Ecorr = −1.46 V), and the former also exhibited a lower corrosion current (icorr = 1.60  10 -4 A/cm 2 ) than did the latter (icorr = 6.20  10 -4 A/cm 2 ).   Figure 8 shows the corroded surfaces of two samples. It can be seen from Figure 8a that the magnesium oxide film generated on the AZ91D matrix cannot provide effective protection due to its loose and porous properties and the extremely low equilibrium potential (−2.37 V) of magnesium in the AZ91D matrix ( Table 3) Therefore, the corrosion of the AZ91D matrix was severe (Table 4 and Figure 8a). On the other hand, the corrosion attack on the Al0.5CoCrCuFeNi high-entropy alloy coating specimen was much less acute. Figure 8b reveals that the corroded surface has a grainy appearance, indicating that corrosion attacks may mainly occur on dendrite boundaries. It is considered that the presence of Cr, Al, and Ni in the Al0.5CoCrCuFeNi high-entropy alloy coating prepared by laser cladding could form dense chromium oxides, aluminum oxides, and nickel oxides at the surface, and this could be an important factor that contributes to the relatively high corrosion resistance of the Al0.5CoCrCuFeNi high-entropy alloy coating specimen.   Figure 8 shows the corroded surfaces of two samples. It can be seen from Figure 8a that the magnesium oxide film generated on the AZ91D matrix cannot provide effective protection due to its loose and porous properties and the extremely low equilibrium potential (−2.37 V) of magnesium in the AZ91D matrix ( Table 3) (a) AZ91D matrix specimen (b) Laser-clad specimen It should be pointed out that the composition of the high-entropy Al0.5CoCrCuFeNi selected in this paper is not superior to that of other high-entropy compositions (for example, the one studied in Reference [32]). In this paper, only the high-entropy Al0.5CoCrCuFeNi composition is chosen because of its good corrosion resistance.

Corrosion Properties
As for how to choose the composition of the coating to obtain high corrosion resistance, we think that the composition of the high-entropy alloy should be based mainly on whether the same Therefore, the corrosion of the AZ91D matrix was severe (Table 4 and Figure 8a).