Journal of Alloys and Compounds

(1)

Structural, tribological and electrochemical behavior of SiC nanocomposite oxide coatings fabricated by plasma electrolytic oxidation (PEO) on AZ31 magnesium alloy

H. Nasiri Vatan, R. Ebrahimi-kahrizsangi

^*

, M. Kasiri-asgarani

Advanced Materials Research Center, Materials Engineering Department, Najafabad Branch, Islamic Azad University, Najafabad, Isfahan, Iran

a r t i c l e i n f o

Article history:

Received 2 November 2015 Received in revised form 9 May 2016

Accepted 10 May 2016 Available online 12 May 2016

Keywords:

Plasma electrolytic oxidation Nanocomposite

Hardness Wear Corrosion

a b s t r a c t

This study investigated the microstructural, tribological and electrochemical properties of oxide layers containing silica (SiC) nanoparticles produced by plasma electrolytic oxidation (PEO) process on AZ31 magnesium alloy. For this purpose, PEO process was carried out in different current densities in aluminate-silicate and phosphate based electrolytes with and without SiC nanoparticles. Microstructure and composition of the coatings was studied by scanning electron microscope (SEM) equipped with energy dispersive spectroscope (EDS) and X-ray diffraction (XRD). Tribological properties of the uncoated AZ31 Mg alloy and PEO-synthesized oxide coatings were evaluated using a reciprocating ball-on-disk tribometer. The electrochemical behavior of the coated specimens was investigated by potentiodynamic polarization test and electrochemical impedance spectroscopy (EIS) measurements in 3.5 wt%

NaCl solution. The coatings produced from aluminate-silicate electrolyte possess high hardness and provide a low wear rate and friction coefﬁcient than those obtained from phosphate electrolyte.

Furthermore, the SiC-containing coatings registered much lower friction coefﬁcient and wear rate than the un-coated AZ31 Mg alloy and the SiC-free oxide coatings, exhibiting excellent self lubricating property. The coatings obtained from the aluminate-silicate electrolyte showed better corrosion resistance due to their compacter microstructure and higher chemical stability.

1. Introduction

Magnesium and its alloys have potentially been used in a wide range of light-weight engineering applications of aerospace, auto- motive, electro-communication, biomedical, computers, etc. owing to their superior characteristics such as high strength/weight ratio, good castability, good ductility, high dimensional stability, etc.[1,2].

However, magnesium and its alloys have poor corrosion resistance because of their high chemical activity and high negative electric potential[3,4]and also low wear resistance[5].

Therefore, a lot of investigations have been going on to improve the aforementioned properties by different surface treatments.

There are different coating technologies to coat magnesium and its alloys including phosphating, chromating, plating, organic coating, gas-phase deposition processes, anodization, etc. On the other hand, every method has its own advantages and disadvantage.

Among them, plasma electrolytic oxidation (PEO) is an environmentally friendly technique for surface treatment of metallic materials in order to enhance the corrosion and wear resistance by producing a relatively thick, dense and hard coating on light weight metals, such as magnesium, titanium, aluminum and their alloys [6e8]. PEO method has some advantages over the other surface treatments including excellent adhesion of the oxide coating to the substrate, single-step processing, relatively low cost, ease of con- trolling and environmentally friendly processing [9,10]. These characteristic of the PEO coatings are mainly dependent on the process parameters, chemical composition of the substrate and the electrolyte used[11].

Investigations have shown that the PEO coatings on magnesium alloys are mainly composed of magnesium oxide along with some other electrolyte-born phases such as MgAl₂O₄, Mg₂SiO₄, Mg3(PO4)2, etc.[12e15]. During recent years, a few attempts have been done to fabricate nanocomposite coatings by addition of appropriate nanopowders into the PEO electrolyte in order to enhance the ﬁnal properties of the coatings such as wear and corrosion resistance, thermal conductivity, etc.[16e21]. Moreover,

*Corresponding author.

E-mail address:[email protected](R. Ebrahimi-kahrizsangi).

Contents lists available atScienceDirect

Journal of Alloys and Compounds

j o u rn a l h o m e p a g e :h t t p : / / w w w . e l s e v i e r . c o m / l o c a t e / j a l c o m

(2)

limited studies have been focused on the effect of the nanoparticles addition on the mechanical, tribological, and electrochemical properties of the PEO coatings produced in different current densities and electrolytes.

Therefore, in the present study, SiC-reinforced nanocomposite oxide layers were formed on AZ31 Mg alloy through plasma electrolytic oxidation in aluminate/silicate and phosphate-based alkaline electrolytes at different current densities in order to obtain possible high hardness, anti-corrosion and anti-wear coatings. The resultant coatings were analyzed by scanning electron microscopy (SEM), mechanical, tribological, and electrochemical measurements.

2. Experimental procedure

2.1. Materials and preparation of the coatings

AZ31 Mg alloy was used as the substrate in this work. Square samples with size of 10 mm10 mm5 mm were cut from the AZ31 Mg alloy ingot and then were ground and polished with SiC abrasive papers (up to mesh 1200) to obtain an average surface

roughness (R_a) of about 0.2mm. In order to obtain a stable current transfer through the specimens and the electrolyte, a threaded hole with a diameter of 3 mm was created on one side of the samples.

The PEO processes were conducted on a 20 kW (maximum voltage of 700 V and maximum current of 3 A) DC power source in aluminate-silicate and phosphate-based electrolytes. The aluminate-silicate electrolyte was prepared using NaAlO2(2 g/L), Na₂SiO₃(2 g/L) and KOH (1.5 g/L) in distilled water; and phosphate electrolyte was prepared using Na3PO4(4 g/L) and KOH (1.5 g/L) in distilled water. SiC nanopowder (5 g/L) was also added into the electrolytes for producing the SiC-containing nanocomposite coatings on AZ31 Mg alloy. SiCﬁne nanoparticle was received from Plasmachem (Germany) with the average particle size of about 30 nm (Fig. 1). After the PEO treatment, the samples were rinsed in ethanol and distilled water, and then dried in air at room temperature. The parameters of the PEO process, such as electrolyte composition, current density and treatment time, are listed in Table 1.

2.2. Characterization of the coatings 2.2.1. Microstructure study

The surface morphology and cross-section of the PEO coatings were observed using a scanning electron microscope (SEM, SIGMA/

VP). The elemental composition of the oxide coatings was analyzed using energy dispersive X-ray spectrometer (EDS) attached to the SEM. XRD experiments were performed on a Siemens D-500 diffractometer with Cu Karadiation in the range of 20e90with a step width of 0.02. The average percentage of the open surface porosity and average diameters of the pores (pore size) of the oxide layers was measured based on the SEM micrographs using imageJ software.

2.2.2. Micro-hardness and tribological measurements of the PEO coatings

Micro-hardness of the un-coated AZ31 Mg alloy and all PEO- coated samples were measured by an MH-5 hardness tester with a Vickers indenter at a load of 10 g and for a loading duration time of 5 s. The friction and wear characteristics of the PEO-synthesized coatings were evaluated in a ball-on-disc conﬁguration against stainless steel balls of Ø 3 mm using a reciprocal-sliding UMT-2MT tribometer under dry sliding conditions. Wear rates of the coatings can be calculated from the following equation:

u

¼ V

ðLNÞ (1)

whereuis wear rate, V is wear volume, L is sliding distance and N is load.

2.2.3. Electrochemical behavior

The electrochemical behavior of the samples was examined using Solarton 1250 potentiostat and a conventional three- Fig. 1.TEM image of the SiC nanoparticles used in PEO process.

Table 1

Speciﬁcation of the electrolytes and PEO process used to coat samples.

Sample number Electrolyte SiC nanopowder Conductivity mS cm¹ pH Current density (mA/cm²) Treatment time (min)

1 Aluminate-silicate No 49.5 13.2 11.5 20

2 No 48.2 13.1 23 20

3 Yes 50.9 13.1 11.5 20

4 Yes 49.3 13.2 23 20

5 Phosphate No 38.7 13.2 27.5 20

6 No 39.4 13.1 55 20

7 Yes 40.1 13.1 27.5 20

8 Yes 39.2 13.1 55 20

H. Nasiri Vatan et al. / Journal of Alloys and Compounds 683 (2016) 241e255 242

(3)

Fig. 2.Voltage-time responses for PEO process of AZ31 Mg alloy substrate in alkaline aluminate-silicate electrolyte in the absence (a,b) and presence (c,d) of SiC nanoparticles in current densities of 11.5 (a,c) and 23 mA/cm²(b,d); phosphate-based electrolyte in the absence (e,f) and presence (g,h) of SiC nanoparticles in current densities of 27.5 (e,g) and 55 mA/cm²(f,h).

(4)

Fig. 3.SEM surface morphology of the PEO coatings grown in alkaline aluminate-silicate electrolyte in the absence (a,b) and presence (c,d) of SiC nanoparticles in current densities of 11.5 (a,c) and 23 mA/cm²(b,d); phosphate-based electrolyte in the absence (e,f) and presence (g,h) of SiC nanoparticles in current densities of 27.5 (e,g) and 55 mA/cm²(f,h).

(5)

electrode cell, employing a saturated calomel electrode (SCE) as the reference electrode and a platinum grid as the counter electrode.

The PEO-coated samples were used as the working electrode, the test solution was 3.5 wt% NaCl, and all of the corrosion tests were carried out at room temperature. After 30 min of immersion in NaCl solution (to reach a relatively stable condition), electrochemical impedance spectroscopy (EIS) measurements were conducted with a frequency range of 10 mHze1 MHz and a 10 mV peak-to-peak AC excitation. The Potentiodynamic polarization tests were conducted at a scan rate of 1 mV/s from0.30 V toþ1.0 V with respect to open circuit potential.

3. Results and discussion

3.1. Voltage-time curves for PEO treatment

The changes of the voltage during the PEO process in different electrolytes and conditions are shown inFig. 2. These curves can be characterized by four stages in all electrolytes. During thefirst stage (50e60 s), the dissolution of the substrate is accompanied by the formation of a thin barrier layer with an oxygen evolution on the surface of the anode, and the voltage increases linearly with time in a constant rate. In this stage, no apparent sparks were found on the substrate surface. When the voltage reaches a critical value, sparking occurs. The critical voltage, defined as the breakdown voltage, is responsible for the formation of micro sparks on the surface of the sample and has a strong dependence on the electrolyte composition and conductivity[22,23]. In this stage (stage II), a large number of uniformly distributed brightfine sparks were appeared and moved quickly across the surface of the sample surface. The rate of increase of voltage during the second stage decreases in all electrolytes. In the third stage, the lifetime and size of the micro sparks on the surface of the sample increase, and the slope of the curve becomes smaller than the second stage. Then, the process enters the stage four and a relatively steady voltage was established on the sample surface. In the fourth stage, the appearance of the micro sparks was gradually changed; their population decreased but their size and lifetime increase, their color was also changed from white to orange and red. The breakdown andfinal voltages of V-t curves of the samples treated in the phosphate electrolyte are relatively higher than those processed at aluminate-silicate electrolyte. Moreover, thefinal voltages obtained in the V-t curves of the samples processed in higher current densities is higher than that recorded in lower current densities in both electrolytes. Visual observations during the PEO process confirmed that the discharges become steadier in the higher current density, which is in good agreement with previous investigations[24]. The voltage responses during the PEO process were practically the same in the electrolyte without and with SiC nanoparticles. This means that the addition of SiC nanoparticles did not significantly affect the variation of oxidation potential during the PEO process.

The pH values in Table 1 indicate that all electrolytes are approximately at the same level of alkalinity. However, the conductivity of the aluminate-silicate electrolyte shows higher value compared to the phosphate electrolyte. This, in turn, leads to applying higher voltages for breaking down the resistance of the dielectric layer and also for starting the sparking phenomena on the surface of the substrate in the phosphate-based electrolyte[25].

The relationship between the breakdown voltage and the electrolyte conductivity has been described by a theoretical model proposed by Ikonopisov[26]:

V_BD¼a_Bþb_Blog 1

k

(2) where, VBis the breakdown voltage, aBis a constant for a given metal, bBis a constant for a given electrolyte, and k is the electrolyte conductivity. This equation shows that there is a reverse relationship between the breakdown potential of the PEO process and the conductivity of the electrolyte, which is in good consistency with the results obtained in this paper.

3.2. Structure and composition of the coatings

The morphologies of the top surface of the PEO coatings produced in different conditions are shown inFig. 3. As can be seen, all the PEOfilms have relatively high porosity and some microcracks on the surface. It has been verified that the formation of the pores is mainly due to the molten oxide and gas bubbles thrown out of the discharge channels, while the high thermal stresses are the driving force for initiating cracks on the coating surface[27]. The appearance of the repeated and concentrated sintering on the surface of the oxide coatings produced in aluminate-silicate electrolyte may be related to the characteristics of the micro-sparks during the PEO process[28]. The SEM micrographs of the these coatings suggest outwardflow of material along a discharge channel through the underlying coating to the coating surface, and then spreading over the coating surface and solidification[16].

The relative percentage of porosities and average size of pores on the surface of the PEO coatings measured based on the SEM images are summarized inTable 2. By observing the surface morphologies of the PEO coatings (Fig. 3) and corresponding analysis of pores (Table 2), it can be noted that in the case of the samples treated in aluminate-silicate electrolyte there is an increasing amount of porosity and pore size on the surface with increasing of current density. Liang et al.[29]and Li et al.[30]studied the effect of applied current density on the properties of PEO coatings on magnesium alloy and found that higher current densities result in intensiﬁed sparking discharges, which in turn lead to enlarged pore size on the surface of the coatings. Bayati et al.[31,32]showed that when a structural defect (such as a pore) forms by an electrical discharge, it is a more susceptible place for next electron ava- lanches because of its lower breakdown voltage in comparison with other areas of the surface which are not porous[33]. In contrast, for the coatings produced in phosphate-based electrolyte with increasing current density from 27.5 to 55 mA/cm², the pore size and amount of porosity both decrease. In this case it seems that another mechanism is prevailing. The high electric current passes the cell generates high temperatures at the avalanche regions and this generated heat is high enough to sinter and melt of the growing layer. Because of these phenomena, some of the pores will be vanished[34].

The small variation in average pore size and corresponding porosity percent demonstrate that the addition of SiC nanoparticles to the base electrolytes has not signiﬁcant inﬂuence on the pore morphology. This is mainly due to this fact that the surface morphology (pore size and density) is mainly dependent on the electrical conductivity of the electrolyte, which is constant for the

Table 2

Porosity percent and average pore size of the samples treated in different conditions.

Sample number^a 1 2 3 4 5 6 7 8

Pore size (mm) 7.57 11.32 7.83 10.56 10.22 9.29 5.13 3.93 Porosity percent (%) 22.45 23.17 18.17 19.45 23.07 9.62 9.16 6.36

aTo see the samples’number please refer toTable 1.

(6)

Fig. 4.SEM micrographs (backscattered electrons) of cross-section of AZ31 magnesium alloy after PEO treatment in alkaline aluminate-silicate electrolyte in the absence (a,b) and presence (c,d) of SiC nanoparticles in current densities of 11.5 (a,c) and 23 mA/cm²(b,d); phosphate-based electrolyte in the absence (e,f) and presence (g,h) of SiC nanoparticles in current densities of 27.5 (e,g) and 55 mA/cm²(f,h).

(7)

SiC-containing and SiC-free electrolytes (seeTable 1).

The SEM micrographs of the cross-section features of the PEO coatings formed in different conditions are presented inFig. 4. As can be seen, the thickness of the coatings changes with the electrolyte type and current density. It is evident that the coating thickness increases with increasing current density, which is in good agreement with previous investigations[35]. Furthermore, the coatings obtained in phosphate-based electrolyte are thicker than those obtained in aluminate-silicate electrolyte.

It is veriﬁed that the more powerful sparks results in the formation of a thick layer on the surface of the substrate, as an equation suggested by Sundararajan et al.[36]:

V_p¼

p

4

d²_p dc

2

¼

p

8

d²_pd_c (3)

According to this model, the volume of the material (Vp) deposited on the coating surface by a single discharge channel depends on the discharge channel diameter (dc) and pancake diameter (dp). So, by considering both of Eqs.(2) and (3), it can be concluded that the lower electrical conductivity of the phosphate- based electrolyte leads to appearance of bigger and more powerful sparks on the surface and subsequently higher thickness will be obtained. The authors’visual observations also conﬁrm that during the PEO process of the samples treated in phosphate electrolyte, the sparks were bigger and stronger, while in the aluminate-silicate electrolyte, more uniform distribution andﬁner size of the sparks were seen on the surface.

Chemical composition of the PEO coatings obtained from EDS spectrum are presented inTable 3. It is clear that the chemical compositions of the coatings are signiﬁcantly dependent on the composition of electrolyte. EDS analysis reveal the presence of Mg and O as the major elements of the coatings. Other elements from the substrate and electrolyte including Al, P, and Si were also found.

As can be seen, a higher current densities lead to an increase of Si:Mg, P:Mg and Al:Mg ratios.

According to the chemical and electrochemical bases, the following formation mechanism is proposed. These reactions can take place under cathodic and anodic conditions in the vicinity of the substrate. Previous investigations have shown that the major phases produced in aluminate and silicate based electrolytes are Mg2SiO4, MgO and MgAl2O4 [2,37] while MgO, MgAl2O4 and Mg₃(PO₄)₂are the main components in phosphate-based electrolyte[25,38]. During theﬁrst stages of the PEO process, Mg trans- forms into Mg^2þion and then reacts with other ions present in the discharge channels such as O², SiO32, PO43, OHand AlO², which are obtained from the components of the electrolyte. MgO is formed by outward migration of Mg^2þions from the anode (substrate) into the discharge channels and inward migration of O² ions form the electrolyte into the discharge channels because of the presence of high electricﬁelds between the two poles of the cell [35,39].

Mg/Mg^þ²þ2e (4)

Mg²^þþO²/MgO (5)

Magnesium orthosilicate can be formed in two ways, according to eqs.(6)e(8) [35,40,41]:

SiO322e/SiO2þ1/2O2 (6)

2MgOþSiO2/Mg2SiO4 (7)

2Mg^2þþ2SiO32/Mg2SiO4þSiO2 (8) MgAl2O4can also be formed via the following reactions[39]:

Table 3

Elemental composition of the PEO-coated samples determined from the EDS analysis.

Electrolyte Current density SiC nanopowder Mg (wt%) O (wt%) Al (wt%) Si(wt%) P (wt%) C (wt%)

Aluminate-silicate 11.5 ✕ 6.74 85.64 4.29 3.34 e e

23 ✕ 7.37 83.92 4.81 3.9 e e

11.5 ✓ 41.2 23.6 13.2 23.7 e 17.5

23 ✓ 22.8 21.1 13.1 26.1 e 16.9

Phosphate 27.5 ✕ 40.8 27.4 0.9 9.6 12.1 9.2

55 ✕ 39.6 24.0 1.4 11.8 13.6 9.6

27.5 ✓ 33.2 28.8 0.8 16.1 12.6 8.5

55 ✓ 36.3 22.3 2.1 17.9 14.7 6.7

Fig. 5.XRD patterns of the PEO-coated samples in different processing conditions.

(8)

Fig. 6.Vickers hardness of the oxide coatings obtained in different electrolytes and current densities.

Fig. 7.The relationship between the friction coefﬁcients and testing time of various samples under dry sliding conditions.

(9)

AlO₂þ2H₂O/Al(OH)₄ (9) 4Al(OH)44e/4Al(OH)3þ2H2OþO2 (10)

2Al(OH)3/Al2O3þ3H2O (11)

Mg²^þþ2Al(OH)₄/MgAl₂O₄þ4H₂O (12)

Al2O3þMgO/MgAl2O4 (13)

Mg²^þþAl₂O₃þ2OH/MgAl₂O₄þH₂O (14) Mg3(PO4)2is produced via inward migration of PO43ions from the electrolyte into the discharge channels and outward migration of Mg²^þ ions from the metallic substrate toward the discharge channels under high temperatures, according to the following equation[35,37]:

3Mg²^þþ2PO43/Mg3(PO4)2 (15)

Fig. 5shows the XRD patterns of the PEO coatings produced in SiC-containing aluminate-silicate and phosphate electrolytes.

These patterns indicate the presence of MgO, MgAl2O4and Mg2SiO4

phases for coatings produced in aluminate-silicate electrolyte, and phases of MgO and Mg₃(PO₄)₂ for coatings obtained from phosphate-based electrolyte. The phase investigation results conﬁrm the reaction and growth mechanisms that were discussed above.

3.3. Hardness and wear properties of the PEO coatings

The microhardness of the Mg alloy substrate and PEO coatings prepared in different electrolytes and current densities are shown inFig. 6. The hardness of the bare AZ31 Mg alloy is only about 64 HV. It can be seen that the microhardness of the oxide coatings is two to eight times higher than that of the magnesium alloy substrate. Hardness of the coatings mainly depends on two factors:

phase structure and compactness. Hardness measurements show that the micro hardness of the PEO-synthesized oxide layers in aluminate-silicate electrolyte is more than those obtained in phosphate-based electrolyte. This is because of the fact that the

micro-hardness of spinel Mg₂SiO₄is higher than that of MgO and Mg3(PO4)2 [28,35,42]. Furthermore, the microhardness of the samples processed in the electrolytes with SiC nanoparticles is higher than that of the coatings obtained from the electrolytes without SiC nanoparticles under the same processing parameters.

It has been shown that the nanoparticles couldﬁll the pores or be embedded in the oxide coatings, leading to an increase the hardness of the coatings[43].

The evolution of the friction coefﬁcients versus sliding time for the PEO coatings is depicted inFig. 7. In addition, the friction co- efﬁcient of the uncoated substrate is also presented for comparison.

As can be seen, compared with the uncoated substrate, all of the Fig. 8.Wear rates obtained by pin-on-disc tests for the uncoated substrate and the PEO-Coated samples.

Fig. 9.SEM micrograph of wear track of the bare Mg alloy.

(10)

Fig. 10. Wear track of PEO-synthesized coatings obtained in alkaline aluminate-silicate electrolyte in the absence (a,b) and presence (c,d) of SiC nanoparticles in current densities of 11.5 (a,c) and 23 mA/cm²(b,d); phosphate-based electrolyte in the absence (e,f) and presence (g,h) of SiC nanoparticles in current densities of 27.5 (e,g) and 55 mA/cm²(f,h).

(11)

PEO-coated samples register lower friction coefﬁcients under dry sliding conditions. It is also observed, there exist non-steady wear period in the wear process of the un-coated AZ31 magnesium alloy.

The addition of SiC nanoparticles to the electrolyte results in obtaining nanocomposite coatings with lower and more stable friction coefﬁcients than those prepared in SiC-free electrolyte. This may be due to the“rolling effect”made by SiC nanoparticles on the surface of the PEO coatings which acted as lubricants during the

sliding tests[44].

The wear rates of the PEO-coated samples were calculated, and results are shown inFig. 8. It can be seen that the wear rates of the SiC-containing PEO coatings are much smaller than the ordinary oxide coatings. Since the nanocomposite coatings get rolling friction and presence of the SiC nanoparticles increase the hardness of the coatings[45]. In the case of the wear mechanism of the nanocomposite coatings, SiC nanoparticles lubricate as balls between

Fig. 11.Potentiodynamic polarization curves of bare Mg alloy and PEO-coated samples in 3.5 wt% NaCl aqueous solution at room temperature and open to air.

(12)

the contact areas and hence change frictional form from sliding to rolling. This phenomenon results in the decrease of friction coef- ﬁcient and also wear rate.

The differences in the friction behavior and the wear rate of the PEO coatings in comparison with bare alloy have been further investigated by SEM micrographs of wear tracks, as shown inFig. 9 and Fig. 10. It can be seen that the wear track of the un-coated magnesium alloy is deeper and wider than those of the PEO- synthesized coatings under the same wear conditions. There are wide ploughs and grooves on the worn surface of the bare magnesium alloy, indicating severe wear occurred on this sample. In- vestigations have shown that these characteristics of the worn surface represent the abrasive wear[46,47]. For the oxide coatings without SiC nanoparticles, relatively slight grooves and ploughs are seen and the widths of wear tracks become narrower and smother than the bare magnesium alloy because of their high hardness and load-bearing capacity.

Fig. 10also shows the SEM morphologies of worn surfaces of the SiC-containing PEO coatings. The entire worn surfaces are quite smooth and no evidence of signiﬁcant detachment of the coatings can be found. These observations indicate that only slight wear damage occurred on the surface of the nanocomposite coatings during sliding test.

It has been shown that the wear resistance of a sample has much to do with its surface hardness [18]. SiC is a kind of ultra hard particles and its hardness can be reach to 3000 HV. When these nano-sized ceramic particles embedded into the coatings, even in a very low volume fraction, they can play a protective role in enhancing the wear resistance. These SiC nanoparticles distributed on the surface of the oxide coatings and then involved in the wear process during the sliding test. These particles reduce signiﬁcantly the frictional shear stress between the coating and steel balls and leads to the decrease of the frictional coefﬁcient and wear rate of the PEO coatings.

3.4. Electrochemical behavior of the PEO coatings 3.4.1. Potentiodynamic polarization

The protectiveness of the PEO coatings were evaluated through the potentiodynamic polarization tests in 3.5 wt% NaCl solution, and the obtained curves are shown in Fig. 11. Corrosion-related parameters like corrosion current density (icorr), corrosion potential (Ecorr), anodic and cathodic Tafel slopes (baand bc) were directly derived from the polarization curves, and are summarized in Table 4. Polarization resistance (Rp) was also calculated by Stern- Geary equation[48]:

Rp¼ b_ab_c

2:303icorrðbaþ bcÞ (16)

When the un-coated substrate is exposed to a corrosive medium (like NaCl solution), a loose and porous Mg(OH)₂layer is quickly

forms on its surface. Increase in the anodic potential results in ac- celeration of Clions migration into the porous layer of Mg(OH)₂ and subsequently promotes the dissolution of the substrate. In contrast, in the case of the PEO-coated specimens, the oxide layer acts as a barrier layer against the transportation of the corrosive ions and preventing them from reaching to the substrate. There- fore, the PEO-treated samples have lower corrosion current density and higher polarization resistance in compared with the bare magnesium alloy substrate. Furthermore, it is evident that the samples with nanocomposite coatings have even lower corrosion current densities than that of the coatings prepared in the SiC-free electrolyte. With the increase in the current density of the PEO process, the corrosion current density tended to be lower.

The higher corrosion resistance of the coatings prepared in aluminate-silicate electrolyte can be attributed to two factors. The ﬁrst factor is the more compactness of the coatings processed in aluminate-silicate electrolyte in comparison with ones produced in phosphate electrolyte, which can prevent the diffusion of the corrosive ions to substrate. The big size of the pores on the surface of the coatings prepared in phosphate-based electrolyte has adverse effect on the corrosion resistance. Second factor is the difference between the chemical compositions of the coatings obtained from two different electrolytes. Based on the EDS analysis results presented inTable 3and the possible reactions can be occur during PEO process (Eqs.(4)e(15)), the coatings produced in aluminate- silicate electrolyte are mainly composed of forsterite (Mg2SiO4) that is chemically stable in NaCl solution, which is in consistence with the results reported by Liang et al.[38]. Besides, it has been reported that the Mg3(PO4)2phase presented in the PEO coatings produced in phosphate-based electrolyte dissolves in aqueous solutions and so is not stable[49,50].

The different polarization behavior of the samples coated in electrolytes with and without presence of SiC nanoparticles can be interpreted as follow: there are many micro pores and micro cracks on the surface of the PEO coatings that are easy paths for trans- ferring of corrosive intermediates (like Cl ions) into the inner layers of the coatings. This leads to increasing of polarization current density. In the case of the nanocomposite coatings, because of the blocking effect of the SiC nanoparticles (which deposited in the pores), transfer of Clcorrosive ions were held back and the increase of corrosion current density was suppressed during polarization measurements. Furthermore, SiC nanoparticles have poor electrical conductivity and hence they do not take part in electrochemical processes but rather cause hindrance[18]. Other reasons for better anti-corrosion behavior of the samples treated higher current densities are their thicker PEO coatings, as shown inFig. 4.

3.4.2. EIS

In order to investigate the barrier properties of the oxide layer/

solution interface, EIS spectra of the PEO-coated specimens were recorded at open circuit potential (OCP) over the wide frequency

Table 4

Corrosion-related parameters and corrosion rate (milli-inches per year: MPY) obtained from potentiodynamic polarization curves ofFig. 10.

Electrolyte Current density SiC nanopowder Ecorr(mV) Icorr(mA) ba(V/decade) bc(V/decade) Rp(U.cm²) Corrosion rate (MPY)

Bare e e 1505 32.74 85 321 175 150.7

Aluminate-silicate 11.5 ✕ 1485 1.84 190 107 3171 8.5

23 ✕ 1515 0.89 235 192 10122 1.98

11.5 ✓ 1420 1.26 48 241 2708 5.8

23 ✓ 1470 0.13 33 64 13956 0.6

Phosphate 27.5 ✕ 1477 3.87 116 174 1533 17.8

55 ✕ 1360 3.27 169 297 2808 15.0

27.5 ✓ 1489 1.80 112 158 3104 8.3

55 ✓ 1452 0.28 12 63 3047 1.3

(13)

range of six decades. Fig. 12 shows the impedance spectra as Niquest plots (imaginary part of Z vs. real part of Z) of bare and PEO- coated AZ31 Mg alloy samples after immersion in NaCl aqueous solution at room temperature. The EIS results seem to conﬁrm clearly the potentiodynamic polarization results. Because MgAl₂O₄ and Mg2SiO4have higher stability in neutral environments than MgO, the presence of these two phases in samples treated in aluminate-silicate electrolyte will be beneﬁcial to improving the

corrosion resistance. In addition, thicker coatings can also contribute to better corrosion resistance for the samples prepared in higher current densities.

4. Conclusions

Surface modiﬁcation of magnesium alloy AZ31 was performed by PEO process in aluminate-silicate and phosphate-based

Fig. 12.Niquest plots of uncoated and PEO-coated magnesium alloy AZ31 in 3.5 wt% NaCl solution at room temperature and open to air after an immersion time of 1 h for stabilization.

(14)

electrolytes with and without addition of SiC nanoparticles. Ac- cording to the above results and discussions, it can be concluded that both aluminate-silicate and phosphate-based electrolytes are beneﬁcial for improvement of the wear and corrosion resistance of the magnesium alloy substrate but the effectiveness of the coatings produced in aluminate-silicate electrolyte was superior to those obtained in phosphate-based electrolyte, which is mainly due to their higher hardness, relatively dense structure, and also more chemically stable nature of the coatings formed in aluminate- silicate based electrolyte in corrosive medium.

PEO coatings formed in SiC-containing electrolytes have higher microhardness than those formed in nanoparticle-free electrolytes.

Under dry sliding conditions, the wear rate of the former are lower than that of the latter. Furthermore, the nanocomposite coatings exhibit a lower friction coefﬁcient than those obtained in the electrolyte without SiC nanoparticles. The reason is that the SiC nanoparticles work as balls between the contact areas and then turn sliding friction to rolling friction. Due to the blocking effect of the SiC nanoparticles, the nanocomposite coatings have much better corrosion resistance than the simple PEO coatings such as corrosion current density decreases and impedance also increases in 3.5 wt% NaCl solution. Apart from pore plugging, the poor electrical conductivity of the SiC nanoparticles helps to enhancing anti- corrosion properties.

References

[1] M. Laleh, A. Sabour Rouhaghdam, T. Shahrabi, A. Shanghi, Effect of alumina sol addition to micro-arc oxidation electrolyte on the properties of MAO coatings formed on magnesium alloy AZ91D, J. Alloys Compd. 496 (2010) 548e552.

[2] S.L. Aktug, S. Durdu, I. Kutbay, M. Usta, Effect of Na2SiO3.5H2O concentration on microstructure and mechanical properties of plasma electrolytic oxide coatings on AZ31 Mg alloy produced by twin roll casting, Ceram. Int.http://dx.

doi.org/10.1016/j.ceramint.2015.09.056.

[3] C. Blawert, W. Dietzel, E. Ghali, G. Song, Anodizing treatments for magnesium alloys and their effect on corrosion resistance in various environments, Adv.

Eng. Mater 8 (2006) 511e533.

[4] Y.K. Lee, K. Lee, T. Jung, Study on microarc oxidation of AZ31B magnesium alloy in alkaline metal silicate solution, Electrochem. Commun. 10 (2008) 1716e1719.

[5] N. Yamauchi, K. Demizu, N. Ueda, N.K. Cuong, T. Sone, Y. Hirose, Friction and wear of DLC ﬁlms on magnesium alloy, Surf. Coat. Technol. 193 (2005) 277e282.

[6] X. Li, X. Liu, B.L. Luan, Corrosion and wear properties of PEO coatings formed on AM60B alloy in NaAlO2electrolytes, Appl. Surf. Sci. 257 (2011) 9135e9141.

[7] A. Dey, R.U. Rani, H.K. Thota, A.K. Sharma, P. Bandyopadhyay, A.K. Mukhopadhyay, Microstructural, corrosion and nanomechanical behaviour of ceramic coatings developed on magnesium AZ31 alloy by micro arc oxidation, Ceram. Int. 39 (2013) 3313e3320.

[8] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, S.J. Dowey, Plasma electrolysis for surface engineering, Surf. Coat. Technol. 122 (1999) 73e93.

[9] P. Gupta, G. Tenhundfeld, E.O. Daigle, D. Ryabkov, Electrolytic plasma technology: science and engineering-An overview, Surf. Coat. Technol. 201 (2007) 8746e8760.

[10] S. Sarbishei, M.A. Faghihi Sani, M.R. Mohammadi, Study plasma electrolytic oxidation process and characterization of coatings formed in an alumina nanoparticle suspension, Vacuum 108 (2014) 12e19.

[11] V. Raj, M. Mubarak Ali, Formation of ceramic alumina nanocomposite coatings on aluminium for enhanced corrosion resistance, J. Mater. Proces. Technol.

209 (2009) 5341e5352.

[12] D. Sreekanth, N. Rameshbabu, K. Venkateswarlu, Effect of various additives on morphology and corrosion behavior of ceramic coatings developed on AZ31 magnesium alloy by plasma electrolytic oxidation, Ceram. Int. 38 (2012) 4607e4615.

[13] L.R. Krishna, G. Poshal, G. Sundararajan, Inﬂuence of electrolyte chemistry on morphology and corrosion resistance of micro arc oxidation coatings deposited on magnesium, Metall. Mater. Trans. A 41 (2010) 3499e3508.

[14] Q. Cai, L. Wang, B. Wei, Q. Liu, Electrochemical performance of microarc oxidationﬁlms formed on AZ91D magnesium alloy in silicate and phosphate electrolytes, Surf. Coat. Technol. 200 (2006) 3727e3733.

[15] R. Arrabal, E. Matykina, P. Skeldon, G.E. Thompson, A. Pardo, Transport of species during plasma electrolytic oxidation of WE43-T6 magnesium alloy, J. Electrochem. Soc. 155 (2008) C101eC111.

[16] R. Arrabal, E. Matykina, F. Viejo, P. Skeldon, G.E. Thompson, M.C. Merino, AC plasma electrolytic oxidation of magnesium with zirconia nanoparticles, Appl.

Surf. Sci. 254 (2008) 6937e6942.

[17] M. Aliofkhazraei, A. Sabour Rouhaghdam, T. Shahrabi, Abrasive wear behaviour of Si3N4/TiO2nanocomposite coatings fabricated by plasma electrolytic oxidation, Surf. Coat. Technol. 205 (2010) S41eS46.

[18] L. Song-mei, Y. Xiu-mei, L. Jian-hua, Y. Mei, W. Liang, Y. Kang, Microstructure and abrasive wear behaviour of anodizing compositeﬁlms containing SiC nanoparticles on Ti6Al4V alloy, J. Cent. South Univ. 21 (2014) 4415e4423.

[19] M. Ming, Z. Xin-jian, X. Qian, L. Jun, H. Xiao-di, Preparation and tribological properties of self-lubricating TiO2/graphite composite coating on Ti6Al4V alloy, Appl. Surf. Sci. 258 (22) (2012) 8570e8576.

[20] S.V. Gnedenkov, S.L. Sinebryukhov, D.V. Mashtalyar, I.M. Imshinetskiy, A.S. Gnedenkov, A.V. Samokhin, Y.V. Tsvetkov, Protective composite coatings obtained by plasma electrolytic oxidation on magnesium alloy MA8, Vacuum 120 (2015) 107e114.

[21] S.V. Gnedenkov, S.L. Sinebryukhov, D.V. Mashtalyar, I.M. Imshinetskiy, A.V. Samokhin, Yu.V. Tsvetkov, Fabrication of coatings on the surface of magnesium alloy by plasma electrolytic oxidation using ZrO2 and SiO2 nanoparticles, J. Nanomater. 2015 (2015). Article ID 154298.

[22] M. Laleh, F. Kargar, A. Sabour Rouhaghdam, Formation of a compact oxide layer on AZ91D magnesium alloy by microarc oxidation via addition of cerium chloride into the MAO electrolyte, J. Coat. Technol. Res. 8 (6) (2011) 765e771.

[23] J.M. Albella, I. Montero, J.M. Martınez-Duart, A theory of avalanche breakdown during anodic oxidation, Electrochim. Acta 32 (1987) 255e258.

[24] Z. Li, Y. Yuan, X. Jing, Effect of current density on the structure, composition and corrosion resistance of plasma electrolytic oxidation coatings on MgeLi alloy, J. Alloys Compd. 541 (2012) 380e391.

[25] A. Ghasemi, N. Scharnagl, C. Blawert, W. Dietzel, K.U. Kainer, Inﬂuence of electrolyte constituents on corrosion behaviour of PEO coatings on magnesium alloys, Surf. Eng. 26 (2010) 321e327.

[26] S. Ikonopisov, Theory of electrical breakdown during formation of barrier anodicﬁlms, Electrochim. Acta 22 (1977) 1077e1082.

[27] H.F. Guo, M.Z. An, Growth of ceramic coatings on AZ91D magnesium alloys by micro-arc oxidation in aluminateeﬂuoride solutions and evaluation of corrosion resistance, Appl. Surf. Sci. 246 (2005) 229e238.

[28] J. Liang, L. Hu, J. Hao, Characterization of microarc oxidation coatings formed on AM60B magnesium alloy in silicate and phosphate electrolytes, Appl. Surf.

Sci. 253 (2007) 4490e4496.

[29] J. Liang, P.B. Srinivasan, C. Blawert, M. St€ormer, W. Dietzel, Effect of current density on the microstructure and corrosion behaviour of plasma electrolytic oxidation treated AM50 magnesium alloy, Appl. Surf. Sci. 255 (2009) 4212e4218.

[30] Z. Li, Y. Yuan, X. Jing, Effect of current density on the structure, composition and corrosion resistance of plasma electrolytic oxidation coatings on MgeLi alloy, J. Alloys Compd. 541 (2012) 380e391.

[31] M.R. Bayati, F. Golestani-Fard, A.Z. Moshfegh, How photocatalytic activity of the MAO-grown TiO2 nano/micro-porousﬁlms is inﬂuenced by growth parameters? Appl. Surf. Sci. 256 (2010) 4253e4259.

[32] F. Samanipour, M.R. Bayati, F. Golestani-Fard, H.R. Zargar, T. Troczynski, A.R. Mirhabibi, An innovative technique to simply fabricate ZrO2eHAeTiO2 nanostructured layers, Colloids Surfaces B Biointerfaces 86 (2011) 14e20.

[33] A. Rapacz-Kmita, A. Slosarczyk, Z. Paszkiewicz, Mechanical properties of HApeZrO2composites, J. Eur. Ceram. Soc. 26 (2006) 1481e1488.

[34] V. Shojaei-Rad, M.R. Bayati, F. Golestani-Fard, H.R. Zargar, J. Javadpour, Fabrication of ZrO2eAl2O3 hybrid nano-porous layers through micro arc oxidation process, Mater. Lett. 65 (2011) 1835e1838.

[35] S. Durdu, M. Usta, Characterization and mechanical properties of coatings on magnesium by micro arc oxidation, Appl. Surf. Sci. 261 (2012) 774e782.

[36] G. Sundararajan, L. Rama Krishna, Mechanisms underlying the formation of thick alumina coatings through the MAO coating technology, Surf. Coat.

Technol. 167 (2003) 269e277.

[37] S. Durdu, A. Aytac, M. Usta, Characterization and corrosion behavior of ceramic coating on magnesium by micro-arc oxidation, J. Alloys Compd. 509 (2011) 8601e8606.

[38] J. Liang, P. Bala Srinivasan, C. Blawert, M. St€ormer, W. Dietzel, Electrochemical corrosion behaviour of plasma electrolytic oxidation coatings on AM50 magnesium alloy formed in silicate and phosphate based electrolytes, Elec- trochim. Acta 54 (2009) 3842e3850.

[39] H.F. Guo, M.Z. An, H.B. Huo, S. Xu, L.J. Wu, Microstructure characteristic of ceramic coatings fabricated on magnesium alloys by micro-arc oxidation in alkaline silicate solutions, Appl. Surf. Sci. 252 (2006) 7911e7916.

[40] S.G. Xin, L.X. Song, R.G. Zhao, X.F. Xu, Composition and thermal properties of the coating containing mullite and alumina, Mater. Chem. Phys. 97 (2006) 132e136.

[41] S.V. Gnedenkov, O.A. Khrisanfova, A.G. Zavidnaya, S.L. Sinebryukhov, V.S. Egorkin, M.V. Nistratova, A. Yerokhin, A. Matthews, PEO coatings obtained on an MgeMn type alloy under unipolar and bipolar modes in silicate- containing electrolytes, Surf. Coat. Technol. 204 (2010) 2316e2322.

[42] Y. Wang, D. Wei, J. Yu, S. Di, Effects of Al2O3nano-additive on performance of micro-arc oxidation coatings formed on AZ91D Mg alloy, J. Mater. Sci. Technol.

30 (10) (2014) 984e990.

[43] M. Aliofkhazraei, A.S. Rouhaghdam, Wear and coating removal mechanism of alumina/titania nanocomposite layer fabricated by plasma electrolysis, Surf.

Coat. Technol. 205 (2011) S57eS62.

[44] M. Mu, J. Liang, X. Zhou, Q. Xiao, One-step preparation of TiO2/MoS2com- posite coating on Ti6Al4V alloy by plasma electrolytic oxidation and its tribological properties, Surf. Coat. Technol. 214 (2013) 124e130.

(15)

[45] S. Li, M. Zhu, J. Liu, M. Yu, L. Wu, J. Zhang, H. Liang, Enhanced tribological behavior of anodicﬁlms containing SiC and PTFE nanoparticles on Ti6Al4V alloy, Appl. Surf. Sci. 316 (2014) 28e35.

[46] A.K. Mondal, B.S.S. Chandra, S. Kumar, Wear behaviour of AE42þ20% safﬁl Mg- MMC, Tribol. Int. 40 (2007) 290e296.

[47] D. Zhang, Y. Gou, Y. Liu, X. Guo, A composite anodizing coating containing superﬁne Al2O3particles on AZ31 magnesium alloy, Surf. Coat. Technol. 236 (2013) 52e57.

[48] M. Stern, A.L. Geary, A theoretical analysis of the shape of polarization curves, J. Electrochem. Soc. 104 (1957) 56e63.

[49] H. Fleisch, R.G.G. Russell, S. Bisaz, J.D. Termine, A.S. Posner, Inﬂuence of py- rophosphate on the transformation of amorphous to crystalline calcium phosphate, Calc. Tiss. Res. 2 (1) (1968) 49e59.

[50] Y. Mori, A. Koshi, J. Liao, H. Asoh, S. Ono, Characteristics and corrosion resistance of plasma electrolytic oxidation coatings on AZ31B Mg alloy formed in phosphateesilicate mixture electrolytes, Corros. Sci. 88 (2014) 254e262.