Tribological performance of PEO-WC nanocomposite coating on Mg Alloys deposited by Plasma Electrolytic Oxidation

(1)

Tribological performance of PEO-WC nanocomposite coating on Mg Alloys deposited by Plasma Electrolytic Oxidation

Hadi NasiriVatan, Reza Ebrahimi-Kahrizsangi

ⁿ

, Masoud 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 18 July 2015 Received in revised form 6 February 2016

Accepted 22 February 2016 Available online 3 March 2016 Keywords:

PEO Wear

Nano-composite coating AZ31B Mg Alloys

a b s t r a c t

AZ31B Mg alloy specimens were coated by PEO technique in WC nano-powder containing phosphate based electrolyte under constant current densities. Wear rate and COF diagrams were plotted in order to investigation of WC nano-particles presence and it was found that nano-particle presence can decrease wear rate and COF signiﬁcantly down to 77% and 55%, respectively. Cross sectional imaging showed that coating thickness increase with coating time and in case of nano-particle containing coatings become more compact but thickness decreases slightly. Investigation of surface morphology indicated that in presence of nano-powder, porosity as well as mean diameter of pores decreased, but prolonging coating time even in presence of nano-powder, led to mean diameter of pores reached up to 8.34mm. Presence of nano-powders, also increased hardness of coatings considerably.

1. Introduction

Tungsten in carbide form has broad and important applications in industries as a refractory metal. According to proper hardness and high wear resistance of Tungsten Carbide (WC), it has vast applications as a hard material in metal working and drilling industries and also mining industries under high pressure and temperature[1–3].

Magnesium and Aluminum are usually used in weight-critical structural applications, for example vehicles and aerospace. Mg is about 35% lighter than Al, and both have similar melting point and strength. Mg has the drawback of limited formability, duo to its HCP structure, while Al has high formability as it is in FCC structure[4,5].

Mg alloys also has lower elastic module than Al, 40–45 GPa for Mg alloys in comparison to 69.6 for Al alloys. Mg alloys have promising applications in some number of areas in industries including automotive, aerospace, computer and IT through their low density and high specific strength and hardness. However their applications are strictly limited through their low thermal resistance. Up to now various techniques have been used to overcome these limitations of Mg alloys including anodizing[6], deposition of metallic resistant coatings[7], surface melting by laser beam[8], purification of cast leading to better wear and corrosion resistance[9]and modification

of surface composition by Radio-frequency plasma chemical vapor deposition (RF-CVD)[10]. Performing a surface protective process is suitable for Mg alloys[11].

Appling coatings in order to eliminate limitations of metal applications and to achieve maximum performance in them, is a way to simultaneously achieve protection against wear and keep mechanical behavior of base metal. Composite coatings duo to vast practical applications have signiﬁcant industrial importance. Pre- sence of nano-particles in composite coatings as hard and wear resistance materials are in high interests. Generally hard ceramic particles such as alumina, tungsten carbide and silicon carbide are used in order to increase composite coatings hardness[12].

Plasma Electrolytic Oxidation is a process similar to anodizing performed at high voltages, applied on light metals above dielec- tric breakdown of coating, in which a high crystallinity thick coating with desired properties is formed. This process leads to formation of plasma as micro discharges on metal surface and gas liberation. Short lifetime (about hundred ms to a few ms) micro discharges deﬁne chemical and thermal situations and play a signiﬁcant role in phase formation, composition and morphology of oxide coatings [13]. These sparks lead to forming plasma channels which in chemical reactions occur and cause to fabricate products (coating). Xin et al.[14]showed that type of current-AC or DC- applied for PEO coating process can alter the phase composition and consequently density, hardness and corrosion resistance of formed coating on Al alloy. Guangliang et al. [15]found that high and low current density lead to fabricate mainly

γ

^-Al2O3

and

α

^-Al2O3 respectively on Al alloy substrate. PEO is an envir- onmentally friendly and commodious process drawing attentions Contents lists available atScienceDirect

journal homepage:www.elsevier.com/locate/triboint

Tribology International

nCorresponding author at: Advanced Materials Research Center, Materials Engineering Department, Najafabad Branch, Islamic Azad University, Najafabad, Isfahan, Iran.

E-mail addresses:hadinasirivattan@gmail.com(H. NasiriVatan), rezaebrahimi@iaun.ac.ir(R. Ebrahimi-Kahrizsangi).

(2)

to itself since it is a new method of producing ceramic coatings with unique properties on valve metals, such as Al, Mg and Ti and their alloys[16]. Wear resistance, mechanical strength and electrical insulation of these metals and their alloys could be improved by PEO process[17]. Generally PEO is accepted as a prospective surface treatment for Mg alloys. Ceramic coatings could be formed in-situ in silicate based electrolytes, consisted of mostly MgO, Mg2SiO4 and MgAl2O4 [18]. These coatings have high hardness values and wear resistance comparing to uncoated substrate under proper conditions of coating process, and perform good electro- magnetic performance. Wheeler et al.[19]compared mechanical properties of coatings fabricated by hard anodizing and PEO process on 5056 AA indicating very better behavior of PEO coating.

Unfortunately they crack easily and fall of which leads to loosing protection of Mg alloys, since these oxide ceramic composite coatings show high illuminations and their expansion coefﬁcient differ signiﬁcantly from Mg alloys. It has reported that PEO coatings are porous in outer layers which is a drawback in tribological and wear performances. PEO is a complex process which reacts components of electrolyte after diffusion and electrophoresis through discharge channels and electrochemical oxidation on the metal surface. It has reported that electrolyte components parti- cipate in PEO process. Tong et al. introduced silicon oxide[20]and iron [21] particles into electrolyte and realized that they can embed in ceramic matrix and produce denser coatings.

Present research attempts to study effects of process para- meters on nano-composite coating properties, thickness and type.

It mainly focuses on study of wear properties surface morphology and composition of WC nano-composite coatings produced by PEO process. Wear resistance is studied through coefﬁcient of friction and wear rate.

2. Materials and methods

Samples used in this study were AZ31B prepared in square shape with thickness of 2 mm acquired from Bohler with composition presented in Table 1. Sample preparation included grinding with abrasive SiC papers in 5 meshes including 200, 800, 1200, 2500 and 4000. Deionized water and hot airﬂow were used to wash and dry samples, respectively. In order to connect anode to samples, they were drilled and screwed from smallest surface to attach the anode electrode. In order to obtain more reliable results statistically, according to repeatability of each process, for each series of experiment at least three samples were used.

Electrolyte composition used in PEO process was based on 4 g/lit sodium phosphate (Na3PO4) with 1.5 g/lit KOH and in 4 cases WC nanoparticles with diameter less than 80 nm (Fig. 1) were added in 5 g/lit concentration. Philips-CM120 TEM was used to determine nanoparticle size. A&D-GR202 balance with 50mg accuracy was used for weighing the samples. To achieve a uniform distribution of nanoparticles in fabricated coatings, suspension was stirred for 6 h on magnetic stirrer before coating process as well as during coating process.

PEO station consists of a 20 KW power supply, a rectiﬁer with maximum output voltage of 600 V and electrolyte container with water cooling system which maintained temperatures below 30°C. The current and potential difference between anode (sample) and cathode was measured by two separated avometers APPA-series 500 connected to a computer to plot potential-time diagram during coating process. During coating process, applied current density was maintained constant equal to 55.02 A/cm². Coating times were 300, 600, 900 and 1200 s. After coating

process, samples were rinsed with distilled water and dried in warm airﬂow (Table 2).

To investigate the morphology of free surface of coatings, cross section and composition of coatings, WEGA TESCAN SEM with RONTEC EDS analyzer and ZEISS SIGMA/VP SEM were used. ImageJ software was used to analyze SEM micrographs to calculate percentage of porosity, pore dimension and coating thickness; the error of this method is quietly related to statistical techniques so Table 1

Composition of AZ31B alloy used in this study as substrate of coating.

element Mg Zn Si Mn Fe Cu Al

Wt% Balance 1.00–1.15 Max 0.022

0.2–0.4 Max 0.003

0.00–

0.030

2.5–3.5

Fig. 1. TEM image of WC nanoparticles embedded into some coatings.

Table 2

Condition of PEO coating process in terms of coating time and presence of nano- powder in the electrolyte. Speciﬁcation of coatings fabricated by PEO process. Pore diameter, cross thickness and porosity percentage are evaluated by analyzing SEM image.

Sample code Nanopowder Time of coating deposition (min)

NP5 WC 5

NP10 WC 10

NP15 WC 15

NP20 WC 20

SS5 – 5

SS10 10

SS15 15

SS20 20

(3)

that according to calibration of the gauge of software, for example for each measurement in term of unit ‘mm’, the value of error would be 0.5mm. Furthermore QnIX8500 instrument was used to measure coating thickness so that for each sample, 18 measurement were performed. Surface roughness (Ra) of coatings was evaluated using Taylor Hobson Sutronic25-Stylus type surface

proﬁlometer with710 nm accuracy. This analysis was carried out 3 times for each sample and mean values were reported. In order to investigate wear resistance and coefﬁcient of friction of samples, pin on disk wear test was performed according to ASTM G-99 standard, applying 1.2 kgf and 0.5 kgf on pin for measuring COF and wear rate respectively with 5 mm round head pin made of WC-Co cermet in room temperature and ambience humidity.

Mean coefﬁcient of friction is obtained by averaging from all data of COF throughout the sliding pin on surface of coating after 1500 cm distance with 5 cm/s linear speed. Hardness of coatings was evaluated by SCTMC-MHV1000 with 0.5 HVN equipped by prismatic indenter according to E-384 ASTM standard.

3. Results and discussion

Fig. 2 Presents potential variations versus time for samples coated for 1200 s. Samples coated in shorter times showed very good sameness with these curves, so duo to high resemblance, only curves for samples coated for longest times with and without presence of nano-powder are presented here. Results are in agreement with previous studies[22]and show that samples coated in electrolyte containing nano-powder have 10 to 15 V lower sparking voltages in comparison to samples coated in electrolyte without nano-powder. In initial stages of sparking, the size of sparks is Fig. 2.Potential variations between anode (sample) and cathode (water cooling

stainless steel) in term of time during coating process for two samples coated for 1200 s.

Fig. 3.SEM images captured from free surface of coatings (a) NP5, (b) NP10, (c) NP15, (d) NP20, (e) SS5, (f) SS10, (g) SS15, (h) SS20.

(4)

smaller leading to create less porous coating with higher electrical resistance; therefore to maintain current density constant, more rate of voltage increase is demanded. But after about 100 s of coating process as growth of coatings takes place, sparks become larger leading to create more porous coating layers having less electrical resistance, hence less rate of voltage increase is required [23]. Nano-composite coating containing nano-powder has lower resistances in comparison to coating without nano-powder. This can be attributed to some reasons such as less porosity of nano- composite coatings, lower energy required for formation of coating in presence of nano-particles, etc. which will be discussed later. In addition, voltage-time diagrams of sample containing nano-powder are much more uniform and with lessﬂuctuations, showing more uniform formation of coating during PEO process.

Fig. 3 shows SEM images captured from free surface of coatings andFig. 4illustrates typically distribution of elements on surface of coating NP5 containing nano-powder. To study morphology of coatings with more details, SEM images from cross section of coatings are presented inFig. 5.Table 2represents weight, thickness, porosity and mean pore diameter of coatings. It can be observed that the structure of coatings become coarser by increasing coating time for both types of ordinary and nano-composite coatings. Addition of nano-powder decreases mean pore diameter from 8.42mm for samples without nano-powder to 7.19mm for samples with nano-powder and also decreases mean surface porosity from 10.59% for samples without nano-powder to 9.81% for samples with nano-powder. Introduction of nano-powders leads to more decrease of pore diameter (about 15%) than porosity percentage (less than 1%). If porosity percentage can be Fig. 4. EDS maps illustrating distribution of considered elements in free surface of coatings NP5. This image were obtained by EDS probe connected to SEM microscope.

Fig. 5.SEM images captured from cross section of coatings. (a) NP5, (b) NP10, (c) NP15, (d) NP20, (e) SS5, (f) SS10, (g) SS15, (h) SS20. Both types of close and open pores can be observed in these images.

(5)

estimated as Eq.(1):

P¼ ðn

π

^{d^2}Þ=4A100 ð1Þ

where in:

P¼porosity percentage;

n¼number of circular pores per unit area;

d¼mean pore diameter;

A¼unit area.

According to proportional of P to square of d and less decrease ofP than value should be decreased corresponding to decrease ofd, one can conclude that introduction of nano-powders leads to increase of number of pores in unit area. In presence of nano-powder, the outfalls Fig. 7.EDS map obtained from SEM image with high magniﬁcation from a single pore at free surface of sample NP15: (a) outside and (b) bottom of pore.

Fig. 6.EDS maps illustrating distribution of considered elements in cross section of coating NP5.

(6)

of pores become more closed, therefore porosity and mean pore diameter are decreased with addition of nano-powder, implying good agreement with similar works on nano-composite PEO coatings[24].

Additionally, all weight changes between before and after coating process, are positive for both coatings with and without nano-particles, indicating that coating formation rate is more than substrate dissolution rate [25]. As it can be realized from data presented in Table 2and as expected, increase of coating time leads to increase of weight change. In addition, investigation of coating thickness with probe shows that thickness increases with coating time.

As presented inFig. 4, all considered elements, even W, dis- tribute uniformly in coating, showing uniform distribution of nano-particles in coating and indicates no agglomeration has taken place during coating process[26]. Mean thicknesses measured by Image-J software fromFig. 5are listed inTable 2, again show that increase of coating time results in increase of thickness.

Fragments of coating dispatched from samples during metallo- graphy, in mount region implies ceramic nature of coatings. It seems that addition of nano-powder increases compactness of coatings and, in turn, decreases thickness of them. These results are in accordance with other ﬁndings in nano-composite PEO coatings[27]. Although increasing trend of thicknesses obtained from analyzing SEM images is almost same as that resulted from measurement by probe, there is considerable difference between thicknesses measured by probe and those obtained from analyzing cross section SEM images. This 1.8–2.7 times larger values of thicknesses measured by probe might be due to high porosity and unevenness of free surface of coatings which can affect on measured value of thickness.

Fig. 6 shows typically map of element distribution in cross section of coating NP5, and as it can be obviously observed all elements, including W, are uniformly distributed across coating.

Fig. 7 demonstrates a single pore from sample NP15 containing WC nano-powder, and apparently nano-particles are absorbed and trapped inside of pores, particularly in bottom of them so that walls of pores contain less nano-particles[28,29]. Somewhat open pores with links together can be seen in thisﬁgure.

Fig. 8shows variation of COF during wear test, andFig. 9shows typically SEM image of worn surface of coating NP5.Table 3pre- sents extracted data from wear test including wear rate, wear track width and COF as well as hardness and roughness of coatings.

Almost all samples without nano-powder reveal similar behavior and have close mean COF values. By embedding nano-powder into coatings, mean COF values begin to decrease [30–32]. As it is evident fromFig. 8, samples containing nano-powder show similar trend, except for sample NP5, which might be due to its very low

thickness leading to completely elimination of coating in early stages of wear test. Addition of nano-particles to coatings led to increase in hardness, 442 Hv0.1for coating fabricated in the most coating time. Surface composition of coatings with and without nano-powder does not differ much; however presence of nano- powder decreases wear rate and wear track width significantly. On the other hand presence of nano-powder, does not affect hardness considerably which can be attributed to unchanging the composition of coating, which consists of mainly MgSiO₄, MgO and Mg (OH)2and insufficient content of nano-powder forfilling pores to the desired level that can enhance the hardness of porous coating.

Presence of nano-powder in coatings decreases wear rate from 28.6510⁴ to 6.6710⁴mg m¹N¹ and wear track width from 1801mm to 491mm for coating created in the most coating time(Table 4).

Fig. 10shows images and element distribution in coatings after wear test. According to good wear behavior of NP20 sample, EDS analysis conﬁrms that owing to uniform distribution ofWover the cross of coating, even after removing portion of coating because of wear, still distribution of W is uniform leading to good wear resistance of nano-composite coatings. Also continuous tracks of wear and occurring no layer breakdown in worn samples, indicates very good cohesion and adhesion of coating to the substrate.

Fig. 8.Variations of coefﬁcient of friction during sliding the pin on free surface of coating for (a) samples with nano-powder and (b) samples without nano-powder.

Fig. 9.SEM micrographs of worn surface for sample NP5.

(7)

Table 3

Results extracted from wear and hardness test. Wear track width was measured from SEM image of worn samples. Much higher values of wear track width for ordinary coatings can be attributed to less strength of them.

Sample code

Weight before PEO (g)

Weight after PEO (g)

Weight difference (mg)

Average of hole diameter(lm)

Probe thickness(lm)

Surface Porosity percent

Cross thickness(lm)

Probe thickness Cross thickness

NP5 4.26116 4.27405 12.89 5.34 16.94 8.94 6.34 2.67

NP10 4.44741 4.46789 20.48 7.32 23.09 9.11 10.15 2.27

NP15 4.97646 5.00452 28.06 7.76 28.37 10.47 12.06 2.35

NP20 4.16794 4.20017 32.23 8.34 28.42 10.71 14.76 1.93

SS5 4.89875 4.91424 15.49 7.01 17.50 11.78 9.67 1.81

SS10 2.73822 2.75956 21.34 8.35 25.37 10.47 11.49 2.21

SS15 5.19725 5.22893 31.68 9.04 30.27 10.48 12.58 2.41

SS20 4.80496 4.84384 38.88 9.29 34.47 9.62 18.30 1.88

Table 4 Sample code

Average of friction

coefﬁcient70.005 Wear track width (lm) 70.5mm

Wear Rate (mg/(m N)) 10⁴70.005

Hardness (Hv0.1) 70.5HVN

Rughness(Ra)(lm) 710 nm

NP5 0.23 1165 9.71 232 0.99

NP10 0.19 1114 9.14 244 1.36

NP15 0.15 495 6.86 231 1.68

NP20 0.14 491 6.67 442 2.46

SS5 0.34 1318 22.94 220 1.14

SS10 0.31 1474 22.54 320 1.52

SS15 0.33 1534 24.29 231 1.95

SS20 0.31 1801 28.65 408 2.13

Fig. 10.EDS map obtained from SEM image of coatings after wear test illustrate element distribution in and around wear track: (d) NP20, (e) SS5, (f) SS10, (g) SS15, (h) SS20.

(8)

4. Conclusions

In this article, effect of coating time and presence of WC nano- powder in PEO nano-composite coatings was studied. Samples coated in suspension including nano-powder had 10–15 V less sparking voltage in comparison to samples without nano-powder.

Addition of nano-powder decreases mean pore diameter averagely 15% for samples with nano-powder comparing to to samples without nano-powder and also decreases averagely 6% mean surface porosity. Samples containing nano-powder were more compact comparing to ones without nano-powder. Distribution of elements over structure of coating was uniform and nano-particle absorbance was occurred inside pores, mainly at the bottom of them and walls of pores absorbed less nano-particles. No layer breakdown phenomena or non-uniform abrasion was observed in samples during wear test. By addition of nano-powder, mean value of COF was decreased averagely 45%. Agglomeration of nano- powders did not occur inside wear track. With introducing nano-particles into coating roughness did not change considerably but hardness increased with coating time up to 442 Hv_0.1.

References

[1]Reyes M, Neville A. Degradation mechanisms of Co-based alloy and WC metal–

matrix composites for drilling tools offshore. Wear 2003;255(7–12):1143–56.

[2]Prakash LJ. Application ofﬁne grained tungsten carbide based cemented car- bides. Int J Refract Met Hard Mater 1995;13(5):257–64.

[3]Fang ZZ, et al. Synthesis, sintering, and mechanical properties of nanocrystalline cemented tungsten carbide–a review. Int J Refract Met Hard Mater 2009;27(2):288–99.

[4]Agnew S. Wrought magnesium: a 21st century outlook. JOM 2004;56(5):20–1.

[5]Zhong XL, Wong WLE, Gupta M. Enhancing strength and ductility of magnesium by integrating it with aluminum nanoparticles. Acta Mater 2007;55 (18):6338–44.

[6]Salman SA, Okido M. 8-Anodization of magnesium (Mg) alloys to improve corrosion resistance. In: Song G-L, editor. Corrosion prevention of magnesium alloys. Woodhead Publishing; 2013. p. 197–231.

[7]Gu C, et al. High corrosion-resistance nanocrystalline Ni coating on AZ91D magnesium alloy. Surf Coat Technol 2006;200(18–19):5413–8.

[8]Dutta Majumdar J, et al. Effect of laser surface melting on corrosion and wear resistance of a commercial magnesium alloy. Mater Sci Eng: A 2003;361(1–

2):119–29.

[9]Lun Sin S, Elsayed A, Ravindran C. Inclusions in magnesium and its alloys: a review. Int Mater Rev 2013;58(7):419–36.

[10]KyziołK, Mędala M, KaczmarekŁ. Inﬂuence of plasmochemical modiﬁcation of Al–Cu–Mg alloys on surface structure and functional properties. Vacuum 2014;105:52–8.

[11]Kulekci M. Magnesium and its alloys applications in automotive industry. Int J Adv Manuf Technol 2008;39(9–10):851–65.

[12]Gnedenkov SV, et al. Protective properties of the nanocomposite coatings on Mg alloy. Solid State Phenom 2014;213:176–9.

[13]Matykina E, et al. Incorporation of zirconia into coatings formed by DC plasma electrolytic oxidation of aluminium in nanoparticle suspensions. Appl Surf Sci 2008;255(5, Part 2):2830–9.

[14]Xin S, et al. Inﬂuence of cathodic current on composition, structure and properties of Al2O3coatings on aluminum alloy prepared by micro-arc oxidation process. Thin Solid Films 2006;515(1):326–32.

[15]Guangliang Y, et al. The effects of current density on the phase composition and microstructure properties of micro-arc oxidation coating. J Alloy Compd 2002;345(1–2):196–200.

[16]Cao F-h, et al. Environmental friendly plasma electrolytic oxidation of AM60 magnesium alloy and its corrosion resistance. Trans Nonferr Met Soc China 2008;18(2):240–7.

[17]Minaev AN, et al. Composite coatings formed by plasma electrolytic oxidation.

Prot Met Phys Chem Surf 2011;47(7):840–9.

[18]Vladimirov BV, et al. Microarc oxidation of magnesium alloys: a review. Surf Eng Appl Electrochem 2014;50(3):195–232.

[19]Wheeler JM, Curran JA, Shrestha S. Microstructure and multi-scale mechanical behavior of hard anodized and plasma electrolytic oxidation (PEO) coatings on aluminum alloy 5052. Surf Coat Technol 2012;207:480–8.

[20]Tong H, Jin F, Shen L. Fabrication of porous ceramic coatings embedded nanograins on titanium alloys. J Korean Phys Soc 2005;46:S134–6.

[21]Jin F, et al. Improvement of surface porosity and properties of aluminaﬁlms by incorporation of Fe micrograins in micro-arc oxidation. Appl Surf Sci 2006;253(2):863–8.

[22] Stojadinović, S., et al., Synthesis and characterization of Al2O3/ZnO coatings formed by plasma electrolytic oxidation. Surface and Coatings Technology, (0).

[23]Guan Y-j, Xia Y. Correlation between discharging property and coatings microstructure during plasma electrolytic oxidation. Trans Nonferr Met Soc China 2006;16(5):1097–102.

[24]Lee KM, et al. Electrochemical response of ZrO2-incorporated oxide layer on AZ91 Mg alloy processed by plasma electrolytic oxidation. Surf Coat Technol 2011;205(13–14):3779–84.

[25]Li H, Sun Y, Zhang J. Effect of ZrO2particle on the performance of micro-arc oxidation coatings on Ti6Al4V. Appl Surf Sci 2015;342(0):183–90.

[26]Sreekanth D, Rameshbabu N. Development and characterization of MgO/

hydroxyapatite composite coating on AZ31 magnesium alloy by plasma electrolytic oxidation coupled with electrophoretic deposition. Mater Lett 2012;68(0):439–42.

[27]Rudnev VS, et al. Plasma-electrolytic formation, composition and catalytic activity of manganese oxide containing structures on titanium. Appl Surf Sci 2005;252(5):1211–20.

[28]Mu M, et al. One-step preparation of TiO2/MoS2composite coating on Ti6Al4V alloy by plasma electrolytic oxidation and its tribological properties. Surf Coat Technol 2013;214(0):124–30.

[29]Lim TS, Ryu HS, Hong S-H. Electrochemical corrosion properties of CeO2- containing coatings on AZ31 magnesium alloys prepared by plasma electrolytic oxidation. Corros Sci 2012;62(0):104–11.

[30]Cui S, et al. Corrosion resistance and wear resistance of plasma electrolytic oxidation coatings on metal matrix composites. Surf Coat Technol 2007;201 (9–11):5306–9.

[31]Srinivasan PB, Blawert C, Dietzel W. Dry sliding wear behaviour of plasma electrolytic oxidation coated AZ91 cast magnesium alloy. Wear 2009;266(11–

12):1241–7.

[32]Bala Srinivasan P, et al. Dry sliding wear behaviour of magnesium oxide and zirconium oxide plasma electrolytic oxidation coated magnesium alloy. Appl Surf Sci 2010;256(10):3265–73.