Microstructure and properties of the low-power-laser clad coatings on magnesium alloy with different amount of rare earth addition

Applied Surface Science 353 (2015) 405–413

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Microstructure and properties of the low-power-laser clad coatings on magnesium alloy with different amount of rare earth addition Rundong Zhu, Zhiyong Li ∗ , Xiaoxi Li, Qi Sun The Welding Research Center, College of Material Science and Engineering, North University of China, Tai Yuan 030051, PR China

a r t i c l e

i n f o

Article history: Received 27 January 2015 Received in revised form 17 May 2015 Accepted 12 June 2015 Available online 27 June 2015 Keywords: Laser clad AZ91D magnesium alloy Rare earth Microstructure Properties

a b s t r a c t Due to the low-melting-point and high evaporation rate of magnesium at elevated temperature, high power laser clad coating on magnesium always causes subsidence and deterioration in the surface. Low power laser can reduce the evaporation effect while brings problems such as decreased thickness, incomplete fusion and unsatisfied performance. Therefore, low power laser with selected parameters was used in our research work to obtain Al–Cu coatings with Y2 O3 addition on AZ91D magnesium alloy. The addition of Y2 O3 obviously increases thickness of the coating and improves the melting efficiency. Furthermore, the effect of Y2 O3 addition on the microstructure of laser clad Al–Cu coatings was investigated by scanning electron microscopy. The energy-dispersive spectrometer (EDS) and X-ray diffractometer (XRD) were used to examine the elemental and phase compositions of the coatings. The properties were investigated by micro-hardness test, dry wear test and electrochemical corrosion. It was found that the addition of Y2 O3 refined the microstructure. The micro-hardness, abrasion resistance and corrosion resistance of the coatings was greatly improved compared with the magnesium matrix, especially for the Al–Cu coating with Y2 O3 addition. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Due to its advantageous physical properties such as high specific strength, low specific weight, excellent damping characteristics, large thermal conductivity, good machinability and electromagnetic shielding characteristics, magnesium alloy is widely used for automation, aerospace, communication and electronics device [1,2]. However, magnesium alloys also have some disadvantageous surface and chemical properties like poor corrosion and wear resistance, high chemical reactivity and poor creep resistance, so that their extensive use is rather limited in many other applications [3,4]. To improve their surface and chemical properties as micro-hardness, abrasion resistance and corrosion resistance, many surface modification techniques, such as conversion coating, gas phase deposition process, anodizing and organic coating have been used. Sebastien Pommiers et al. [5] had tried to form chromium conversion coatings on the surface of magnesium alloy but due to the toxicity of hexavalent chromium the process has strong limitation. Pichel et al. [6] had obtained coating with TiN on AM60 magnesium alloy by physical vapor deposition but in such

∗ Corresponding author. Tel.: +86 13753115250. E-mail addresses: [email protected], lzhy [email protected] (Z. Li). http://dx.doi.org/10.1016/j.apsusc.2015.06.071 0169-4332/© 2015 Elsevier B.V. All rights reserved.

case the manufacturing parameters are difficult to be controlled. Li Wei-ping et al. [7] fabricated Mg2 SiO4 films on AZ91D magnesium alloy by non-sparking anodization process but the composition of the electrolyte in this process is too complicated and therefore it takes a long time to form the film. Consequently all the above processes turn out to be highly expensive and have a rather low efficiency. Laser cladding is an effective material processing method that produces a surface layer with many advantages. It has already been applied to metals such as carbon steel and Ni-based super alloy to improve their properties. However, the application of laser cladding to magnesium alloy cannot be so successful because of its high evaporation rate at elevated temperature. When the laser beam is focused on the surface of the materials, the energy is transferred to the matrix producing large amount of magnesium fume in a short time. The evaporation leads to bad formation of the coating and increases the dilution rate of the matrix. Furthermore, the properties of the clad coating are not satisfying even though researchers have tried different parameters for the laser cladding of magnesium alloy. According to our previous work [8], it turns out that when the laser energy is too high, it may cause subsidence due to evaporation of the matrix, even when the speed is suitably chosen. With low power laser, the energy transmitted to the matrix is limited and the solidification rate of the coating is faster, which can protect

406

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

Table 1 Chemical composition (wt.%) of AZ91D magnesium alloy. Chemical composition

Mg

Al

Zn

Mn

Si

Be

Others

Relative amount (%)

89.97

8.99

0.71

0.25

0.048

0.0071

≤0.002

the matrix from evaporation. However, when the power is too low, the coating always undergoes an incomplete fusion and it attains a limited thickness. It is critical to get coatings with satisfying formation and properties for the laser cladding of magnesium alloy. Therefore, in this paper we try to use a laser, which has lower power than that used for other research work [9–11], in the attempt to clad magnesium alloys with an acceptable formation of the coating and a suitable thickness by adding different amount of Y2 O3 . With strong chemical activities and large atomic radius, rare earth can react easily with many elements, playing a favorable role in alloys [12]. Rare earths are widely used in metallurgy. Furthermore, the rare earth oxides are unpurified particles in magnesium alloys which are liable to absorb more laser energy than the uniform alloys. In this research work, the rare earth oxide added into the Al–Cu powders help to form laser clad coatings on magnesium alloy with increased thickness. In recent years, rare earths are gradually introduced to modify the surface properties of engineering components by flame spraying, laser cladding and electric plating [13–15]. Sharma et al. [16] had added different amount of La2 O3 into flame sprayed Ni based coatings to study its effects on the microstructure, hardness and abrasive wear behavior of the coatings. The result showed that the optimal addition of La2 O3 was 1.2 wt.%. Li Ming-Xi et al. [17] studied the effect of Y2 O3 on microstructure of laser clad cobalt-based alloy coating on Ni-based super-alloy. Both the fine and short dendritic microstructure was found and the columnar to equiaxed transition occurred when Y2 O3 was added. Wang Hong-Yu et al. [18] had added CeO2 into the NiCoCrAlY coatings to study the effects of CeO2 on microstructure and properties of the coatings. The results show that with the addition of CeO2 the growth pattern of interface grain was changed. The microstructure was refined and the microhardness, the thermal shock resistance were improved. However, few reports about the rare-earth oxides applied to magnesium alloy surface modification are available in the published literatures. According to our previous experiments in which Y2 O3 , La2 O3 and CeO2 are added into cladding powders which are put on AZ91D magnesium alloy, the laser clad coatings with Y2 O3 addition showed to have the best formation. According Wu Yu-Rong’s research [19], La2 O3 and CeO2 are light rare earth oxide while Y2 O3 belongs to heavy rare earth oxide. Heavy rare earth oxides are always better than light rare earth oxides for improving the mechanical properties of the magnesium alloy. Therefore, Y2 O3 was selected to be the additive in the laser cladding powders in our research work. We study the effects of Y2 O3 on the microstructure, micro-hardness, wear and corrosion properties of the Al–Cu alloy coatings of AZ91D magnesium alloy, which could offer an experimental and theoretical basis for a promising application of rare-earth oxide on magnesium alloy.

2. Experimental procedures 2.1. Preparation of samples The matrix used in the study is AZ91D magnesium alloy with dimension of 15 mm × 15 mm × 4 mm and the primary powders are Al (98.0% purity) and Cu (98.0% purity) (the mass ratio of Al:Cu is 7:3). The chemical composition of the AZ91D magnesium alloy is shown in Table 1. The plates were polished with metallographic sand paper, washed with alcohol and dried in air in order to

produce a smooth surface without contaminants. The rare earth oxide Y2 O3 was added into the primary powders with 0 wt.%, 0.4 wt.%, 0.8 wt.%, 1.2 wt.% and 2.0 wt.% respectively. As shown in Table 2, Five groups of cladding powders were prepared in which the Al–Cu alloy powders and Y2 O3 were mixed with adhesives. After having uniformly milled, the mixed powders were preplaced onto the AZ91D magnesium alloy plate with a thickness of 0.8 mm. In order to make the results reliable and reproducible, enough specimens for different groups were prepared for comparing the performance and observing their microstructure. 2.2. Preparation of the coatings A 0.4KW pulsed Nd:YAG laser system was used for cladding. After a series of tests, the optimal experimental parameters were selected as given in Table 3. Argon (99.999% purity) was used for shielding the melting region from being oxidized. As the laser beam moves, the metals are deposited on the matrix, forming the coatings (unmodified coating: the coating without Y2 O3 . Y2 O3 -modified coating: the coating with Y2 O3 ). 2.3. Analysis on microstructure, the elemental and phase compositions of the coatings The microstructure of the coatings with and without Y2 O3 was investigated with a SU-1500 model scanning electron microscope. The elemental composition of the coatings was identified by a Thermo type energy disperse spectroscopy. The D/max-RB model X-ray diffractometer was used to examine the phase compositions of the coatings. 2.4. Investigation of the properties of the coatings The micro-hardness was measured by a DHV-1000 vickers hardness tester with load of 1.96 N and loading time of 15 s. The micro-hardness was measured every 0.1 mm along the vertical direction of the transverse cross section. Wear resistance of the coatings was evaluated with a UMT-3 reciprocating type sliding wear tester. The load is 3 N and the duration is 10 min. The ball of the friction pair is GCr15 with the radius of 2 mm. The electrochemical corrosion of coatings was measured by a ZF-3 type constant potential rectifier. 3. Results and analysis 3.1. Microstructure The micrographs of the coatings at low magnification (30×) are shown in Fig. 1 where the average thickness of coatings is in the range of 0.5–0.8 mm. The serrated morphology at bonding zone Table 2 Specimen groups and Chemical composition (wt.%) of coatings. Specimen group

Composition

1 2 3 4 5

Al–Cu + 0 wt.% Y2 O3 Al–Cu + 0.4 wt.% Y2 O3 Al–Cu + 0.8 wt.% Y2 O3 Al–Cu + 1.2 wt.% Y2 O3 Al–Cu + 2.0 wt.% Y2 O3

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

407

Table 3 The parameters of laser cladding. Power (W)

Velocity (mm/s)

Frequency (HZ)

Pulse width (ms)

Radius of beam spot (mm)

240

2

13

3.8

0.3

indicates that there is an excellent metallurgical bonding between the coatings and the matrix. There are no cracks in the coatings but only few pores. As shown in Fig. 1, with the increase of Y2 O3 addition, the thickness of the coating increases correspondingly. The coating with 1.2% Y2 O3 addition is about 0.8 mm and the coating without Y2 O3 addition is only 0.5 mm. As described in Fig. 2, when the laser focuses on the cladding powders, the redundant energy is lost mainly by two ways: it is either transferred to the matrix or carried away by the metallic vapor. Rare earths, which have a higher absorption rate of the laser energy, can reduce the evaporation of both the cladding powders and the matrix magnesium. Therefore, the clad coatings with addition of Y2 O3 are thicker than

the coatings without Y2 O3 addition. On the other hand, the former can improve the weldability of deposited metal [20], which also contributes to the formation of the deposited bead. The magnification of the pore in the coatings was showed in Fig. 1(f). Some solid particles with irregular shape exist in the pore which was caused by the incomplete fusion of laser clad powders. Fig. 3 shows that the coatings have a strong metallic bond to the matrix, which indicates that the selected cladding powders adhere well to the AZ91D magnesium alloy. According to Fig. 3, the coatings can be divided into the alloy layer (AL), the bonding zone (BZ) and the heat affected zone (HAZ). Typical microstructure of fusion zone near the matrix side is afterbirth-like crystal, arborescent crystal

Fig. 1. Micrographs of the cross-section of unmodified and Y2 O3 -modified coatings.

408

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

Fig. 2. The model of the laser energy flowing.

and isometric crystal. The arborescent crystal preferentially grows along the direction of temperature diffusion. In addition, some agglomerated particles with finer crystal are formed. To make clear the effects of Y2 O3 addition on the microstructure and elemental distribution of the coatings, both the transverse cross section observed at 1000 times magnification as well as the EDS analysis results are shown in Fig. 4 and Table 4 respectively. The coating without Y2 O3 shows lenticular nuclei and

the average diameter of them is about 1.5 ␮m (Fig. 4(a)). Compared with the coating without Y2 O3 , the microstructure of the Y2 O3 /Al–Cu coatings appears to have a different dendritic eutectic (Fig. 4(b)–(e)). The dendritic eutectic distributes uniformly among the dendrites, which is beneficial to micro-hardness, wear resistance and corrosion resistance of the coatings [21]. With Y2 O3 addition, the morphology of the crystal changes from lenticular nucleus into either coarse dendritic crystal or fringe-crystal, or dispersed particles. In Fig. 4(b) (0.4% Y2 O3 ), it is shown that the coarse dendritic crystals distribute themselves among the primary dendrite crystals and, as a consequence, coarse particles are formed in the microstructure. As shown in Fig. 4(c) (0.8% Y2 O3 ) and Fig. 4(d) (1.2% Y2 O3 ), the crystal texture appears to be a fringe-crystal without coarse particles. As it is known, Y2 O3 has higher melting point than that of Al and Cu and therefore the particles containing the rare earth solidify earlier than Al and Cu. During the process of dendrite growth, those pre-solidifying particles supply sufficient nucleating centers for dendrites and prevent the dendrite from coarsening. EDS analysis indicates that the main elements of the coating without Y2 O3 are Al, Cu, Mg. The amount of Al is up to 73.27 wt.%, which is similar to the precursor mixed alloy powders. By contrast it is found that the crystal texture is uniform and the dendrite growth

Fig. 3. Micrograph of bonding between the bottom portion of the clad coatings and the matrix.

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

409

Fig. 4. Microstructure of unmodified and Y2 O3 -modified coatings.

direction is regular just when the addition of Y2 O3 is 0.8 wt.%. There are big solid particles in the coatings with 0.4 wt.% and 2.0 wt.% Y2 O3 . Such particles reduce the uniformity and compactness of the microstructure, which in turn would damage the wear resistance and corrosion resistance of the coatings. Through the EDS analysis, it is known that the chemical element percentage in dendrites is different. When the high energy laser beam focused on the

coating, the heat would melt both the coating and part of the matrix. The melted matrix would rise into the melting-pool, leading to the dilution of the coating. According to some relevant experiments found in the literature [22–25], the effects of the addition of Y2 O3 on microstructure of laser clad coatings can be observed. During the process of laser cladding, most of the Y2 O3 were dissolved and decomposed into Y

Table 4 Elements composition of the microstructure analyzed by EDS. Y2 O3 addition amount, wt.% position elements

Al Cu Mg Y Others

0.0 (Fig. 4a)

0.4 (Fig. 4b)

A

A 72.21 12.23 11.82 0.24 2.50

73.27 12.59 12.33 0 1.81

0.8 (Fig. 4c)

1.2 (Fig. 4d)

2.0 (Fig. 4e)

B

A

A

A

B

70.78 12.96 13.73 0.36 2.17

74.72 10.54 12.26 0.42 1.86

72.33 12.27 12.15 0.50 1.95

72.64 13.27 11.93 0.76 1.40

72.46 14.36 10.29 0.62 2.27

410

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

Fig. 6. Micro-hardness distribution across the cross-section of coatings.

In which  is the wavelength of X-ray and is a constant for a certain diffraction condition, d is the inter-planar spacing of HKL crystal plane and  is the diffraction angle. The increase of diffraction angle  would result in the decrease of inter-planar distance d as given by the following relationship: d= Fig. 5. X-ray diffraction spectra of laser clad coatings.

due to intense metallurgical reactions, which can be proved by EDS analysis results in Table 4. As is well known, Y belongs to surfaceactive elements with rather large atomic radius. It means that Y can react easily with other elements to form stable compounds [26]. During the solidification process of the melting-pool, Y in the clad coatings tends to distribute in the liquid–solid interface or at the grain boundary, where the atomic arrangement is irregular. In this way the rare earth increases the tendency of composition undercooling and increases the nucleation rate. Thus, the growth of grain would be greatly suppressed and the size of grains is smaller [27]. Some non-dissolved Y2 O3 may act as heterogeneous nucleation. In addition, Y can react with O, N, etc. elements to form some high melting point compounds, which lower the melting point of the cladding powder and increase the fluidity of the melting pool. Moreover, the slag together with formed compounds may ascend toward the surface of the clad layer through the molten deposit. It is worthy pointing out that all factors mentioned above contribute to the refinement of the microstructure, as well as to both the purification and the compactness of the clad layer.

3.2. The X-ray diffraction spectra of the coatings The XRD spectra of the laser clad coatings are shown in Fig. 5. As can be seen in Fig. 5, the strong reflection occurs in the angle ranges of 30◦ –50◦ . The primary phases Mg32 Al47 Cu7 , Mg2 Cu6 Al5 , Al2 CuMg and Al12 Mg17 are with strong intensities. Other phases, such as MgAl2 O4 , Y2 O3 and Al4 MgY show weaker intensities. By comparing the spectra of the coatings, it is noticed that the addition of Y2 O3 results in the formation of such compounds as Al4 MgY and MgAl2 O4 . In addition, it is also noticed that the diffraction angle  of the coatings with Y2 O3 is larger than that of the coating without it. Indeed according to Bragg’s law: 2d sin  = 

(1)



a

(2)

H 2 + K 2 + L2

Since it is observed that the addition of Y2 O3 can increase the diffraction angle, it can be inferred that the inter-planar distance and the lattice constant is decreased. This fact obviously implies that the microstructure of the coatings is refined and the secondary dendrite spacing is also reduced. 3.3. Micro-hardness In Fig. 6 it is shown that the micro-hardness of the laser clad coatings is higher than that of the matrix, which is about 65 HV. For the coatings with Y2 O3 addition, the middle zone of the coatings has higher micro-hardness than the surface and the bonding zone. However, the coating without Y2 O3 has a relatively uniform distribution of micro-hardness. For the bonding zone, the substrate diffused toward the interface and the micro-hardness decreases. But when the laser beam is focused on the surface of coating, it is found that the non uniform temperature field had a negative influence on the micro-hardness of the surface. The coatings produced with the addition of Y2 O3 possess higher micro-hardness than that of the coating without Y2 O3 . According to the microstructure and EDS results, the increase of micro-hardness in the coatings with Y2 O3 addition is mainly caused by both the presence of Al–Cu–Y2 O3 , and the Al–Cu–Y dendritic structure, as well as by the finer microstructure. Average micro-hardness of the coating with 0.8 wt.% Y2 O3 is nearly the same as the coating with 2.0 wt.% Y2 O3 , i.e., 310 HV. The coating with 0.4 wt.% Y2 O3 possesses lower micro-hardness about 280 HV. The coating with 1.2 wt.% Y2 O3 has the highest average micro-hardness about 367 HV, which is 6–7 times higher than that of the AZ91D magnesium matrix. 3.4. Room temperature wear behaviors The friction coefficient is an important indication of abrasive resistance. The average friction coefficients are shown in Fig. 7a and the transient friction coefficients versus time are shown in Fig. 7b–f for both unmodified and Y2 O3 -modified coatings. Fig. 7a shows that the friction coefficient decreases with the increase of Y2 O3 . This

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

411

Fig. 7. A comparison of the friction coefficient between the unmodified and Y2 O3 -modified coatings.

fact shows that the addition of Y2 O3 has a positive effect on wear resistance. The maximum average friction coefficient is 0.313 for 0 wt.% Y2 O3 and the minimum average friction coefficient is 0.21 for 2.0 wt.% Y2 O3 . As shown in Fig. 7b–e, the friction coefficient is lowest at the beginning of every experiment. As the time goes, the friction coefficient increases rapidly. The sudden increase of friction coefficient may be caused by the increase of abrasive dust at the surface. When the addition of Y2 O3 is 0.4 wt.%, 0.8 wt.% and 1.2 wt.%, the friction coefficient increases gradually with time. The friction coefficient of the coatings with 0 wt.% and 2.0 wt.% Y2 O3 fluctuates around a certain value. Firstly, it is normally assumed that the crystal structure has effects on the friction coefficient and on the wear property of material. Buckley [28] showed that low friction coefficient occurred only for those crystal structures that deform exclusively in the basal plane. For the familiar crystalline structures such as bodycentered cubic (bcc), face-centered cubic (fcc) and hexagonal close-packed (hcp), the friction coefficient normally follows the order of hcp < fcc < bcc . As is known, Al and Cu belong to the face-centered cubic structure and Y and Mg have hexagonal

close-packed structure. Therefore, the average friction coefficient of Y2 O3 -modified coating is smaller than that of unmodified coating. Secondly, both the refinement of microstructure and the decreasing of porosity in coatings decrease the friction coefficient of the coatings [29]. Thirdly, the addition of Y2 O3 would refine grain and increase the grain boundary, which is beneficial to the microhardness of the coatings. In turn the micro-hardness is the most important factor which connected with the wear resistance of material. Therefore, the decrease of the friction coefficient of the coatings with Y2 O3 is certainly caused by the uniform distribution of the rare earth element in the coatings. 3.5. Corrosion In Table 5, we compare both the corrosion potential and the corrosion current density of the unmodified coating, Y2 O3 -modified coating and AZ91D magnesium alloy matrix. The results show that the corrosion potential of the laser clad coatings is higher than that of the matrix and the corrosion current density is lower than the latter, which indicates that the corrosion resistance of AZ91D

412

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

Fig. 8. Polarization curve of the unmodified and Y2 O3 -modified coatings.

magnesium alloy is improved by laser cladding process. The corrosion current density of coatings with Y2 O3 addition declines. When the additive amount of Y2 O3 is 0.8 wt.%, the corrosion potential is −1327 mV, which is 22 mV more positive than the unmodified coating. The corrosion current density of the coating with 0.8 wt.% Y2 O3 addition is 0.00072 mA/cm2 , which is about 20 times lower than that of the unmodified coating. The corrosion current density of the coating with 0.4 wt.% Y2 O3 is 0.0033 mA/cm2 , which is very close to that of the coating with 2.0 wt.% Y2 O3 addition. The coating with 1.2 wt.% Y2 O3 has the smallest corrosion current density, which is about 0.00036 mA/cm2 . As it is well known, rare earth oxide can act as a barrier for the diffusion of the corrosion products which helps to improve the corrosion resistance [30]. Besides the rare earth oxide also hinders water molecule from permeating into and hinders gas from diffusing outward the corrosion boundary, preventing corrosion reaction of the substrate materials. Therefore, the rare earth improves the corrosion resistance of the Y2 O3 -modified specimens. Polarization curves of the specimens in 3.5% NaCl solution are given in Fig. 8. There is an obvious difference of polarization resistance between the coated specimens and the AZ91D magnesium alloy matrix. The corrosion potential of the Y2 O3 -modified coatings shifts from −1500 mV to −1800 mV while the corresponding corrosion current density changes only a little, indicating that the corrosion of the Y2 O3 -modified specimens is in the stage of passivation. In this way it reduced the corrosion current density and improved the corrosion resistance as well. However, the unmodified coating specimen and the AZ91D magnesium alloy matrix are still corroded seriously at the moment. Furthermore,

Table 5 Corrosion parameters of different specimens in 3.5% NaCl solution. Specimen group

Corrosion potential (mV)

Corrosion current density (mA/cm2 )

AZ91D matrix 0 wt.% Y2 O3 0.4 wt.% Y2 O3 0.8 wt.% Y2 O3 1.2 wt.% Y2 O3 2.0 wt.% Y2 O3

−1485 −1355 −1356 −1327 −1359 −1325

0.03 0.014 0.0033 0.00072 0.00036 0.0032

the cathode and anode corrosion reactions of the coated specimen are restrained to a certain extent and obviously the anode reaction is even more restrained, as it is also proved by Xu’s [31] former research work. The data of corrosion potential shifts from −1485 mV of the substrate specimen to −1325 mV of the modified coating with 2.0 wt.% Y2 O3 . There is no doubt that the corrosion resistance of the coated specimens is improved, for the Y2 O3 -modified coatings is modified using the same elements, their corrosion potential would not change significantly. Obviously the effect of different amount Y2 O3 on the corrosion current density of the coatings is similar to what we have mentioned above. 4. Conclusions 1. A low-power-laser was applied to deposit the Al–Cu alloy coatings with different additive amount of Y2 O3 on AZ91D magnesium alloy matrix successfully. The metallurgical bond without cracks was achieved at the interfaces between the matrix and the coatings. The Y2 O3 -modified coatings are thicker than the unmodified Al–Cu coating. 2. The microstructure of the coatings is uniform and smooth. With the addition of Y2 O3 , the grains of the coatings are refined and the EDS analysis indicates that the dilution rate of the coatings declines. The primary phases in the coatings are Mg32 Al47 Cu7 , Mg2 Cu6 Al5 , Al2 CuMg and Al12 Mg17 . The Al4 MgY and MgAl2 O4 phases can be found in Y2 O3 -modified coatings. 3. The micro-hardness of the coatings improves obviously compared with the AZ91D magnesium alloy matrix. When the additive amount of Y2 O3 is 1.2 wt.%, the coating has the highest average micro-hardness value, i.e., 376 HV, which is 6 times higher than that of the matrix. 4. The friction experiments show that with the increase of Y2 O3 addition, the average friction coefficient of the coatings declines and wear resistance of the coatings is improved. 5. The corrosion resistance of the Y2 O3 -modified specimens is better than that of the unmodified coating. The corrosion current density of the coating with 1.2 wt.% Y2 O3 is 0.00036 mA/mm2 , which is two orders lower than the Al–Cu alloy coating. The coating with 2.0 wt.% Y2 O3 possesses the highest corrosion potential, i.e., −1325 mV, which is 160 mV higher than that of the matrix.

R. Zhu et al. / Applied Surface Science 353 (2015) 405–413

Acknowledgements The work described in this paper was financially supported by the Research Project supported by Shanxi Scholarship Council of China (Project No. 2012-69), the Science and Technology Project for the Researchers Overseas in Shanxi Province (2013) and the Natural Science Foundation of Shanxi Province of China (2012011021-1). The authors are grateful to Yingqiao Zhang and Yun Ye for offering help in scanning experiment. The authors also thank professor Mercurio for the help in language. The Technology Project for the Researchers Overseas in Shanxi Province (Project No. 2013-07). The Natural Science Foundation of Shanxi Province of China (Project No. 2012011021-1). References [1] S. Ignat, P. Sallamand, D. Grevey, Magnesium alloys laser (Nd:YAG) cladding and alloying with side injection of aluminium powder, Appl. Surf. Sci. 225 (2004) 124–134. [2] J. Yao, G.P. Sun, H.Y. Wang, S.Q. Jia, S.S. Jia, Laser cladding of AZ91D magnesium alloys with Al + Si + Al2 O3 , J. Alloys Comp. 407 (2006) 201–207. [3] J.E. Gray, B. Luan, Protective coatings on magnesium and its alloys – a critical review, J. Alloys Comp. 336 (2002) 88–113. [4] G. Abbas, L. Li, U. Ghazanfar, et al., Effect of high power diode laser surface melting on wear resistance of magnesium alloy, Wear 260 (2006) 175– 180. [5] S. Pommiers, J. Frayret, et al., Alternative conversion coatings to chromate for the protection of magnesium alloys, Corros. Sci. 84 (2014) 135–146. [6] M. Pichel., N. Candela, R. Barea, Deposition of TiN coatings using ACPVD technique on AM60 alloy, Boletin de la Sociedad de Ceramica y Vidrio 52 (2013) 118–126. [7] W.P. Li, W. Li, L.Q. Zhu, Non-sparking anodization process of AZ91D magnesium alloy under low AC voltage, Mater. Sci. Eng. B 178 (2013) 417–424. [8] R.D. Zhu, Z.Y. Li, X.X. Li, Q. Sun, Simulation and experimental verification of laser cladding temperature field for Al–Cu alloy on AZ91D magnesium alloy surface, Surf. Technol. 6 (2014) 84–89. [9] A. Fabre, J.E. Masse, Friction behavior of laser cladding magnesium alloy against AISI 52100 steel, Tribol. Int. 46 (2012) 247–253. [10] Y. Yang, H. Wu, Improving the wear resistance of AZ91D magnesium alloys by laser cladding with Al–Si powders, Mater. Lett. 63 (2009) 19–21. [11] T.M. Yue, Y.P. Su, Laser cladding of SiC reinforced Zr65 Al7.5 Ni10 Cu17.5 amorphous coating on magnesium substrate, Appl. Surf. Sci. 255 (2008) 1692– 1698. [12] K.L. Wang, Q.B. Zhang, M.L. Sun, Rare earth elements modification of laser-clad nickel-based alloy coatings, Appl. Surf. Sci. 174 (2001) 191–200.

413

[13] H. Zhang, Y. Zou, Z.D. Zou, Microstructures and properties of low-chromium high corrosion-resistant TiC-VC reinforced Fe-based laser cladding layer, J. Alloys Comp. 622 (2015) 62–68. [14] Z.Y. Zhang, B. Liang, H.J. Guo, Interface microstructure and tribological properties of flame spraying NiCr/La2 O3 coatings, J. Therm. Spray Technol. 23 (2014) 1404–1412. [15] J.J. Hu, S. Han, G.X. Sun, Effect of strain rate on tensile properties of electric brush-plated nanocrystalline copper, Mater. Sci. Eng. A 618 (2014) 621–628. [16] S.P. Sharma, D.K. Dwivedi, P.K. Jain, Effect of La2 O3 addition on the microstructure, hardness and abrasive wear behavior of flame sprayed Ni based coatings, Wear 267 (2009) 853–859. [17] M.X. Li, Y.Z. He, X.M. Yuan, Effect of nano-Y2 O3 on microstructure of laser cladding cobalt-based alloy coatings, Appl. Surf. Sci. 252 (2006) 2882–2887. [18] H.Y. Wang, D.W. Zuo, X.F. Li, Effects of CeO2 nanoparticles on microstructure and properties of laser cladded NiCoCrAlY coatings, J. Rare Earth 28 (2010) 246–250. [19] Y.R. Wu, Theory Simulation for Thermodynamic and Solid Solution Properties of Rare Earth Magnesium Alloys, Hunan University, Changsha, 2007. [20] P.Q. Xu, X.H. Tang, Y. Shun, Effect of Y2 O3 addition on microstructure of Nibased alloy+Y2 O3 /substrate laser clad, J. Mater. Process. Technol. 208 (2008) 549–555. [21] Q.L. Deng, A.N. Xie, Z.J. Ge, J.L. Song, Influence of rare-earth La2 O3 on the properties of laser cladding on 316L stainless steel coatings, in: Proceeding of the ICSFT, September 2006. [22] B.N. Liang, Z.Y. Zhang, Z.P. Wang, B.M. Chen, Rare earth effect on the microstructure and tribological properties of FeNiCr coatings, Rare Met. 29 (2010) 270–275. [23] J. Liu, H.M. Deng, H.Y. Cao, Influence of rare-earth element doping on structure and optical properties of BiFeO3 thin films fabricated by pulsed laser deposition, Appl. Surf. Sci. 307 (2014) 543–547. [24] S. Gavras, M. Easton, M.A. Gibson, Microstructure and property evaluation of high-pressure die-cast Mg–La-rare earth alloy, J. Alloys Comp. 597 (2014) 21–29. [25] S.F. Chen, S.Y. Liu, Y. Wang, Microstructure and properties of HVOF-spray NiCrAlY coatings modified by rare-earth, J. Therm. Spray Technol. 23 (2014) 809–817. [26] W.B. Liu, B.L. Liu, J. Li, Effects of characteristics of Y2 O3 commercial powders on structural and optical properties of Nd:YAG ceramics, Ceram. Int. 40 (2014) 11399–11404. [27] M.C. Wang, W.C. Wu, X.Y. Mao, Modification by addition of Al–Y alloy on cast iron surface by laser, J. Chin. Rare Earth Soc. 10 (1992) 234–238. [28] D.H. Buckley, Surface Effects in Adhesion, Friction, Wear and Lubrication Tribology Series, vol. 5, Elsevier, Amsterdam, 1981, pp. 315. [29] J.N. Li, C.Z. Chen, C.F. Zhang, Effect of nano-CeO2 on microstructure properties of TiC/TiN + nTi(CN) reinforced composite coating, Bull. Mater. Sci. 35 (2012) 399–404. [30] Y. Zhang, Y. Xu, D.R. Zhou, et al., Effect of rare earth Ce on corrosion resistance for AZ91 magnesium alloy, J. Harbin Inst. Technol. 34 (2002) 376. [31] Y. Xu, X. Chen, Z.S. Lu, et al., Corrosion characteristic and protection technique for Mg alloy, Corros. Sci. 33 (2001) 753.