Rif

Sa

b

a

ARRAA

KZPECBC

1

tcm(rrfifioctaa(

h0

Applied Surface Science 317 (2014) 198–209

Contents lists available at ScienceDirect

Applied Surface Science

journa l h om epa ge: www.elsev ier .com/ locate /apsusc

ole of electrolyte composition on structural, morphological andn-vitro biological properties of plasma electrolytic oxidation filmsormed on zirconium

andhyarani Ma, Prasadrao Tb, Rameshbabu Na,∗

Department of Metallurgical and Materials Engineering, National Institute of Technology, Tiruchirappalli 620015, Tamilnadu, IndiaDepartment of Physics, Koneru Lakshmaiah University, Vaddeswaram, Guntur 522502, Andhra Pradesh, India

r t i c l e i n f o

rticle history:eceived 15 April 2014eceived in revised form 12 August 2014ccepted 13 August 2014vailable online 27 August 2014

eywords:irconium implantlasma electrolytic oxidationlectrolyte chemistry

a b s t r a c t

Development of oxide films on metallic implants with a good combination of corrosion resistance, bioac-tivity and cell adhesion can greatly improve its biocompatibility and functionality. Thus, the present workis aimed to fabricate oxide films on metallic Zr by plasma electrolytic oxidation (PEO) in methodicallyvaried concentrations of phosphate, silicate and KOH based electrolyte systems using a pulsed DC powersource. The oxide films fabricated on Zr are characterized for its phase composition, surface morphology,chemical composition, roughness, wettability, surface energy, corrosion resistance, apatite forming abil-ity and osteoblast cell adhesion. Uniform films with thickness varying from 6 to 11 �m are formed. XRDpatterns of all the PEO films showed the predominance of monoclinic zirconia phase. The film formed inphosphate + KOH electrolyte showed superior corrosion resistance, which can be ascribed to its pore free

orrosionioactivityell adhesion

morphology. The films formed in silicate electrolyte showed higher apatite forming ability with goodcell adhesion and spreading over its surface which is attributed to its superior surface roughness andwettability characteristics. Among the five different electrolyte systems employed in the present study,the PEO film formed in an electrolyte system with phosphate + silicate + KOH showed optimum corrosionresistance, apatite forming ability and biocompatibility.

. Introduction

Excellent biocompatibility, high mechanical strength and frac-ure toughness, reasonably good corrosion resistance, low thermalonductivity, together with low elastic modulus (92 GPa) and lowagnetic susceptibility (−13.8 × 10−6 cm3/mol) make zirconium

Zr) one of the best implant material in orthopedic and dentalestoration fields [1,2]. The biocompatibility and high corrosionesistance of Zr are ascribed due to the formation of a native oxidelm on its surface. However, this native oxide (zirconia, ZrO2)lm is categorized as bio-inert [3] that restricts the formationf chemical bonds with bone tissue during implantation, whichould be a drawback since, an early integration between bioma-erial and bone is advantageous for most implant applications. In

ddition, this native oxide layer is very thin, at most 2–5 nm [4]nd can be lost soon due to wear, when Zr is used in high loadbody weight) bearing implant applications [5], e.g., total hip and

∗ Corresponding author. Tel.: +91 431 2503464; fax: +91 431 2500133.E-mail addresses: rameshrohith@gmail.com, nrb@nitt.edu (R. N).

ttp://dx.doi.org/10.1016/j.apsusc.2014.08.081169-4332/© 2014 Elsevier B.V. All rights reserved.

© 2014 Elsevier B.V. All rights reserved.

knee arthroplasties. Further, pitting corrosion on Zr surface hasbeen observed at lower potentials in simulated body fluid (SBF)conditions [6]. This consequence in an increase in corrosion rateand in turn may reduce the biological performance and servicetime of the Zr implant. As a result, appropriate surface treatmentis required to improve the bioactivity and corrosion resistance ofZr, without sacrificing its biocompatibility.

In this theme, various surface modification techniques, e.g.,physical or chemical vapor deposition, sol–gel process, plasmaspray coating, thermal oxidation, anodization and plasma elec-trolytic oxidation (PEO) have been adopted in recent years toimprove surface properties of Zr [4,7–11]. Of these, PEO also termedas micro arc oxidation (MAO) or spark plasma anodization (SPA)is an emerging technique to develop firmly adherent, crystalline,porous, relatively rough and thick oxide coatings on Zr and othervalve metals in environmental friendly alkaline based electrolytes.The oxide layers developed by PEO offer a unique combination ofwear and corrosion resistance of aluminum, titanium and mag-nesium alloys [12–14]. Further, the morphology and thickness of

the oxide films formed during PEO can be controlled over a widerange by changing both intrinsic parameters (electrolyte concen-tration, composition and pH) and extrinsic parameters (electrical

ce Science 317 (2014) 198–209 199

pTiacgita

bmioacb(eaocsWohateAracomttatfaastbb

rbproista

2

2

2ta

Fig. 1. Variations of cell voltage with PEO process time for a period of 360 s observedduring the formation of S1–S5 films.

Table 1Composition and concentration of electrolytes used in the PEO film formation andthe identification codes for the films.

Sl. no. Electrolyte composition Electrolyteconcentration

Samplecode

1 Na3PO4·12H2O 13 g/L S12 Na3PO4·12H2O + KOH 10 g/L + 3 g/L S23 Na2SiO3·9H2O 13 g/L S34 Na2SiO3·9H2O + KOH 10 g/L + 3 g/L S45 Na3PO4·12H2O + Na2SiO3·9H2O

+ KOH5 g/L + 5 g/L + 3 g/L S5

S. M et al. / Applied Surfa

arameters, electrolyte temperature and treatment time) [15].herein, electrolyte composition and concentration play a key rolen obtaining the desired coatings of special phase compositionnd microstructure [14,16]. Each electrolyte introduces differentations and anions into the electrolyte solution which consequentlyets incorporated during the oxide film growth, thereby stronglynfluencing the resultant coating characteristics such as, nature ofhe pore and its size distribution, the phases present in the coatingnd its corrosion, wear and biological properties [13,14,16].

Among the wide range of electrolytes, silicate and phosphateased electrolytes were commonly used to form oxide films onagnesium and titanium, especially, when these implants are spec-

fied for biomedical applications [17,18]. Cheng et al., done seriesf works for developing wear resistant oxide films on Zircaloy 2nd Zircaloy 4 in silicate and phosphate electrolytes [19–22]. PEOoatings formed in silicate based electrolytes are usually a com-ination of monoclinic zirconia (m-ZrO2) and tetragonal zirconiat-ZrO2), and t-ZrO2 increases with increase in concentration of thelectrolyte, suggesting that silicon species stabilize t-ZrO2 [21,22],nd the coatings formed in phosphate based electrolyte comprisedf m-ZrO2 [19,20,22]. Zho et al., produced 11–15 �m thick PEOoatings on a Zr–1Nb alloy in phosphate electrolyte and the corro-ion protection of these coatings in 0.5 M Li–OH was studied [23].

ang et al. [5] and Chen et al. [24] produced thin oxide coatingsf 6–15 �m thick on Zr–2.5Nb alloy in silicate electrolyte usingigh frequency DC power source. The coatings comprised of highmount of m-ZrO2 and provide better wear and corrosion resis-ance than a commercial autoclaved black oxide coating. Matykinat al. reported a 100 �m thick oxide coating on Zr alloy (Zirlo) underC conditions in an alkaline silicate electrolyte and its corrosionesistance in H2SO4 environment was studied [25]. Studies werelso focused toward understanding the PEO process, luminescenceharacteristics and growth kinetics during the oxide film formationn Zr in AC [26] and DC current regime [6,27]. Although the surfaceodification of Zr by PEO in silicate and phosphate based elec-

rolytes are widely investigated, the studies were mainly focusedoward understanding the growth kinetics of oxide film formationnd phase transformation. Further, the earlier works were concen-rated on improving the wear and corrosion resistance propertiesor nuclear applications. However, no literature at present is avail-ble regarding the potential use of PEO coated oxide films in silicatend phosphate based electrolytes for biomedical applications. Theurface properties such as, wettability, roughness, corrosion resis-ance in physiological environment, apatite forming ability andiocompatibility of the oxide films formed in silicate and phosphateased electrolytes remain unexplored.

The effect of PEO treatment time on the composition and cor-osion resistance of the oxide films formed on Zr in phosphateased electrolyte has been reported by the authors’ group [28]. Theresent work is aimed at evaluating the in-vitro biological (cor-osion resistance, bioactivity, cell adhesion) characteristics of thexide films formed on Zr by PEO treatment in methodically var-ed concentrations of phosphate, silicate and KOH based electrolyteystems and to recommend the suitable electrolyte system for PEOreatment on Zr to have optimum properties for orthopedic implantpplications.

. Experimental

.1. Formation of oxide films on Zr

Commercial purity Zr (>99.5 wt%) with coupon dimensions of0 mm × 15 mm × 1.5 mm were used for the present study. Prior,o PEO treatment, the coupons were polished with abrasive papers,nd then cleaned with acetone and deionized water in an ultrasonic

bath. A DC power supply unit (with a maximum output currentof 15 A and a maximum peak voltage of 900 V) was employed tocarry out the PEO process. The Zr coupons were then treated in fivedifferent electrolyte solutions containing methodologically var-ied concentrations of tri-sodium ortho phosphate (Na3PO4·12H2O,Merck India Pvt. Ltd.), sodium meta silicate (Na2SiO3·9H2O, MerckIndia Pvt. Ltd) and potassium hydroxide (KOH, Merck India Pvt. Ltd)for 6 min at a constant current of 1 A corresponding to 150 mA/cm2

current density at the workpiece. The applied duty cycle and fre-quency was 95% and 50 Hz, respectively. The observed variation involtage during the PEO process time of 6 min at a constant currentof 1 A is reported in Fig. 1. To control the bath temperature closeto the room temperature, the electrolyte bath was water cooledduring the process thereby avoiding thermally driven growth pro-cess. To ensure uniform electrolyte concentration and dissipationof heat generated during the process, the electrolyte solution waskept under continuous stirring by a digital magnetic stirrer (Q 20Amodel, REMI make, India). The breakdown voltage was recordedin all electrolyte systems by a careful physical observation of theappearance of the micro sparks on the anodic surface. The break-down voltage was recorded in triplicate and the average value wasreported in the present study. The final voltages observed at the endof 6 min in all electrolyte systems were also recorded. After PEO pro-cess, the treated samples were cleaned with deionized water andair dried at room temperature. The composition and concentrationof five different electrolytes used in the present study with theiridentification codes are presented in Table 1. The PEO treated Zrsamples are further referred to with these identification codes and

the untreated Zr is referred as “S”.

2 ce Scie

2

RrabDte

d

waimetUtaZ2Twitccdswo

E

wes

2

sdbw[csepswsbwTbrat

R

00 S. M et al. / Applied Surfa

.2. Physico-chemical characterization of PEO treated Zr samples

The phases present in the oxide films was characterized by aigaku Ultima III X-ray diffractometer (40 kV, 30 mA) with a Cu K�

adiation over a 2� range from 20◦ to 70◦ with a step size of 0.05◦

nd a scan speed of 1◦ min−1. The formed phases were identifiedy matching relevant data from the Joint Committee on Powderiffraction Standards (JCPDS) cards. The crystallite size d (nm) of

he formed ZrO2 phases can be determined by using Scherrer’squation [29]:

(nm) = 0.9�

cos �(1)

here � (nm) was the X-ray wavelength, � (rad) the diffractionngle and (rad) the full width of the peak at half maximumntensity (FWHM). The surface morphology, film thickness and ele-

ental composition of all the films were examined using scanninglectron microscope (SEM, Model: Hitachi–S3000 N) equipped withhe energy-dispersive X-ray spectroscopy (EDS, Thermo ScientificltraDry). Before SEM analysis, PEO treated samples were sput-

ered with a thin gold layer to make the surfaces conductive. Theverage surface roughness (Ra) of untreated Zr and PEO treatedr samples was measured using a surface profilometer (Surtronic5, Taylor-Hobson, Precision, UK) with an accuracy of 0.01 �m.he contact angle (�) of Zr substrate and PEO treated Zr samplesas measured by the sessile drop method with the Easy DROP

nstrument (DSA100, KRUSS, Germany) using distilled water ashe contacting solvent. The drop image was captured by a videoamera and an image analysis system was used to calculate theontact angle based on the average of left and right angles of eachrop. Measurements were obtained at ten different locations on theample surface and the average contact angle value of all samplesas reported. The surface energy (Es) has been calculated from the

btained contact angles using the equation [30]:

S = Evl cos � (2)

here Evl is the surface energy between water and air under ambi-nt condition, (i.e., 72.8 mJ/m2 at 20 ◦C) for pure water and � is thetatic contact angle.

.3. Electrochemical corrosion study

To evaluate the in-vitro corrosion properties of untreated Zrubstrate and the PEO fabricated ZrO2 films, Tafel and potentio-ynamic polarization (PDP) tests were conducted under simulatedody fluid (SBF) condition (7.4 pH and 37 ◦C). The SBF test mediumas prepared following the procedure suggested by Kokubo et al.

31]. The Tafel and PDP plots were obtained using a computerontrolled ACM Gill AC (ACM Instruments, Cumbria, UK) corro-ion testing unit. A three electrode cell, with saturated calomellectrode as reference electrode, sample as working electrode andlatinum foil as counter electrode, was employed in the presenttudy. During the test, the sample with an exposed area of 0.5 cm2

as kept in contact with the test solution. Prior to these tests, all theamples were immersed in the test solution for 4 h to attain a sta-le open circuit electrode potential (OCP). The OCP measurementsere conducted for every 5 min during the immersion time of 4 h.

he corrosion current density and corrosion rate was determinedy Tafel extrapolation method from the Tafel plots, over a potentialange of ±200 mV (Tafel region) with reference to OCP employing

scan rate of 0.166 mV/s. The polarization resistance (Rp) of all the

est samples is calculated using Stern–Geary equation [32]:

p = ˇa × ˇc

2.303jcorr(ˇa + ˇc)(3)

nce 317 (2014) 198–209

where ˇa and ˇc are slopes of anodic and cathodic Tafel plots,respectively, and jcorr is the corrosion current density. From theobtained Rp values, the protection efficiency (PE) of all the oxidefilms was calculated according to the formula [33]:

Protection efficiency (%)

= Rp (PEO treated Zr) − Rp (untreated Zr)Rp (PEO treated Zr)

× (1 0 0) (4)

Further, the passivation behavior of the samples in SBF mediumwas studied over a potential range of −500 mV to 3000 mV vs OCPat a scan rate of 100 mV/min by performing potentiodynamic polar-ization test.

2.4. In-vitro bioactivity study

The apatite forming ability of the untreated Zr substrate andoxide films was evaluated by immersing the samples in KokuboSBF solution [31]. This solution has an ion concentration similarto human blood plasma and has been widely used to estimate thein-vitro behavior of biomaterials [34,35]. The solution was bufferedto the pH of 7.4 using tris (hydroxymethyl) amino methane andconcentrated HCl at 36.5 ◦C. Each sample was immersed in a plasticvial containing 62 mL of SBF solution and was kept under staticconditions inside a biological thermostat at 37 ◦C. The ratio of thesample’s surface area (in mm2) to volume of SBF solution (in mL)was set equal to 10 [31]. The SBF solution was refreshed every 24 h,and has been kept stable and colorless without any deposit duringthe test period to sustain the ionic concentration, so that the lack ofions would not slow down the apatite formation. After immersingfor 14 days, the samples were taken out from the SBF, gently washedwith deionized water, dried in air and stored in vacuum desiccatorand characterized by SEM for its surface morphology and EDS forelemental composition.

2.5. Cell adhesion test

In order to study the biocompatibility of oxide films, an in-vitrocell adhesion test was performed on all PEO treated samples anduntreated Zr using well-characterized human osteoblast cell line,human osteosarcoma cells (HOS, National Centre for cell science,Pune, India). All samples were autoclaved for sterilization purposesat 121 ◦C for 20 min before testing. HOS cells seeded on condi-tioned material (glass cover slip) was used as a test control. Thesecells were sub cultured and seeded at a density of 1 × 104 cells/cm2

on all the samples in minimum essential medium supplementedwith 10% fetal bovine serum under standard cell culture conditions(sterile chamber maintained at 37 ◦C and a humidified environ-ment: 5% CO2/95% air). After 48 h, samples were rinsed three timeswith phosphate buffered saline to remove the non adherent cells,while adherent cells were fixed with 2.5% glutaraldehyde. Sampleswere then dehydrated through graded concentration of alcohol fol-lowed by critical point drying. Dried samples were sputter-coatedwith gold, and were subjected to SEM examination to evaluate theadhesion of HOS cells.

3. Results and discussion

3.1. Voltage-time (V–T) response during PEO process

During the PEO treatment, the voltage response with processtime was recorded. The V–T response curves of the PEO process of

Zr in the five different electrolyte systems at a current density of150 mA/cm2 for 6 min are shown in Fig. 1. V–T curves in differentelectrolytes exhibit two characteristic regions, namely anodic oxi-dation region (up to breakdown voltage) and PEO region (above

S. M et al. / Applied Surface Science 317 (2014) 198–209 201

Table 2The pH and conductivity values of five different electrolyte systems with characteristic voltage values observed during PEO process and surface roughness and energy of theformed films.

Sample code pH � (mS/cm) Vb (±5 V) (Tb) (±2 s) Vf (±1 V) Surface roughness, Ra (�m)* Surface energy (mJ/m2)**

S1 11.92 9.85 239 14 487 0.60 ± 0.05 57.8 ± 0.9S2 12.39 15.56 215 12 463 0.48 ± 0.03 5.33 ± 0.5S3 12.27 12.26 242 12 500 0.72 ± 0.09 72.46 ± 1.2S4 12.55 18.21 230 12 462 0.70 ± 0.07 72.16 ± 1.1S5 12.46 16.54 234 14 471 0.65 ± 0.06 62.40 ± 0.7

iffere five d

bwibrlvwfaoP(sosIbuTriidotpZvtttltb2itdovclKVsaTr

ttw

where I 1 1 1m

, I(1 1 1)m and I(1 0 1)t are intensities of mono-clinic peak at 28.2 ◦C, 31.5 ◦C and intensity of tetragonal peakat 30.2 ◦C, respectively. The volume fraction of m-ZrO2 phaseto t-ZrO2 phase in all oxide films is calculated and reported

* Standard deviations calculated considering 4 different regions for each of two d** Standard deviations calculated by considering water contact angles obtained at

reakdown voltage). In the initial stages, a thin barrier film formith no apparent sparks on the anodic surface. As the film grown

n thickness during the anodic process, the process voltage needs toe increased till the breakdown voltage of the initially grown bar-ier film is reached [13], and as a result the process voltage increasesinearly at a very high rate of 15–20 V/s as shown in Fig. 1. When theoltage reached Vb, numerous micro-sparks were observed on thehole sample surface which indicates the start of the PEO process

rom there the chemical reactions between the plasma componentsnd the Zr substrate in a discharge channel take place [28]. Basedn the difference in the rise in the voltage and sparking behavior,EO region in Fig. 1 is divided into 3 stages, namely, dynamic PEOsparking) stage, near steady state PEO (arcing) stage and steadytate PEO (beyond critical voltage, Vc) region. From Fig. 1 it wasbserved that the variation of slope of the V–T curves shows atrong dependence on their respective conductivity of electrolyte.n addition to V–T response curves, the process parameters, namelyreakdown voltage (Vb), breakdown time (Tb), final voltage (Vf) val-es are also influenced by the respective electrolyte conductivity.he pH, conductivity, Vb, Tb and Vf values corresponding to theespective electrolyte system in the present study were reportedn Table 2. Among the process parameters, Tb could reflect the easen the formation of dielectric barrier layer and the occurrence ofischarge phenomena. Lower the Tb, higher the ease of formationf barrier layer [13]. Therefore, in the present study, all the elec-rolytes showed the feasibility of forming a barrier layer in shorteriods of time. Although, Tb of the initial anodic oxide film onr in all electrolytes was found to be in the range of 12–14 s, theoltage required for its breakdown is different with different elec-rolytes. It can be observed from Fig. 1 that in phosphate electrolyte,he Vb of 239 V (S1) is lower than the voltage for the silicate elec-rolyte (242 V for S3), shows that the PEO process could start atower process voltage and form fewer defects. Further, the addi-ion of KOH to both silicate and phosphate electrolytes reduces thereakdown voltage drastically from 239 V (S1) to 215 V (S2) and42 V (S3) to 230 V (S4), respectively. This can be attributed to the

ncrease in the electrolyte conductivity with KOH addition. Higherhe electrolyte conductivity more is the primary electron currentensity, which subsequently results in reaching the critical valuef the secondary electron current density at comparatively loweroltages leading to film breakdown [13]. The Vb in mixed electrolyteontaining phosphate + silicate + KOH is found to be 234 V (S5) wasower than the voltages observed in individual electrolytes withoutOH. From Fig. 1, at a process time of 5 min, all the samples showedc, beyond which the anodic voltage maintained a comparativelytable value (steady-state PEO region). The final voltages observedt the end of 6 min in five different electrolytes were reported inable 2. The different Vf values revealed that the plasma chemicaleactions at the substrate/electrolyte interface were rather diverse.

Thus, the physical properties of the oxide film may vary dueo the variations in structural and morphological characteristics ofhe oxide films formed in different electrolyte systems. Hence, itas valuable to characterize the formed films for its structural and

nt coupons related to the same film.ifferent locations for each of two different coupons related to the same film.

morphological features in order to investigate the effect of processvoltages observed in different electrolytes.

3.2. Phase analysis of oxide films

XRD analysis was performed to investigate the phase forma-tion and transformation of the oxide film formed in differentelectrolytes. OriginPro8 software was used to draw XRD graphsfrom the obtained data and the diffraction intensities were nor-malized by the intensity of the strongest diffraction peak obtainedin the films. The XRD patterns of S1–S5 oxide films and untreated Zrare shown in Fig. 2. All the oxide films show the presence of mono-clinic zirconia (m-ZrO2, JCPDS card no 37-1484) phase with a traceof tetragonal zirconia (t-ZrO2, JCPDS card no 42-1164) phase. Fur-ther, intensity of t-ZrO2 peaks (indexed by symbol ‘*’, Fig. 2) changeswith respect to the electrolyte. In the quantitative phase analysisof oxide films, the method proposed by Toraya et al. was used todetermine the relative proportions of monoclinic and tetragonalphases [36]:

Vm = 1.311Xm

1 + 0.311Xm(5)

where Xm is the mass fraction of monoclinic phase, which can becalculated using Garvie and Nicholson’s formula [37]:

Xm =I(

1 1 1)

m+ I(1 1 1)m

I(

1 1 1)

m+ I(1 1 1)m + I(1 0 1)t

(6)

( )

Fig. 2. XRD patterns of substrate S, and PEO treated S1–S5 oxide films.

202 S. M et al. / Applied Surface Science 317 (2014) 198–209

Table 3Phase composition, crystallite size and intensity values of highest intense

(1 1 1

)m

, (2 0 0)m and (1 0 1)t planes obtained from XRD patterns of PEO fabricated S1–S5 films. M

refers to monoclinic zirconia, T refers to tetragonal zirconia.

Sample Phasespresent

Volume % (±0.3) Crystallite size (nm) (±0.5) Intensity (a.u.) of M peaks

M T(

1 1 1)

m(1 0 1)t

(1 1 1

)(2 0 0)

S1 M, T 94 6 32.7 22.4 100 55S2 M, T 94 6 34.5 22.3 100 56S3 M, T 91 9 28.9 21.7 79 100

25.31.

ipSvstepyets9

m(zb

TfrnZfvfS

rfo1ctsoscaAo[bfrpfomradtt

S4 M, T 91 9

S5 M, T 95 5

n Table 3. The volume fraction of m-ZrO2 phase to t-ZrO2hase is found to be 94% for S1 and S2 oxide films, 91% for3 and S4 oxide films and 95% for S5 oxide film. The reduction inolume % of m-ZrO2 to t-ZrO2 for S3 and S4 samples proves that Sitabilizes the t-ZrO2 phase at low temperatures. The formation of-ZrO2 is due to the fact that the instant temperature at the plasmalectrolytic discharge region is up to 103 ◦C to 106 ◦C during the PEOrocess [11]. Further, the higher voltages (∼500 V, S3 sample) canield higher localized temperature in the PEO discharge channels,nhancing the transformation of m-ZrO2 to t-ZrO2 because the lat-er is a high temperature phase. Thus, all the PEO treated sampleshowed t-ZrO2 phase and the maximum vol% of t-ZrO2 obtained is%, when the films are grown in silicate containing electrolyte.

Although the films formed in different electrolytes are having-ZrO2 in major, orientation change of m-ZrO2 from

(1 1 1

)to

2 0 0) is observed. From XRD patterns, a quantitative characteri-ation of the prominence of the observed orientation was definedy using specific intensity values of

(1 1 1

)m

and (2 0 0)m planes.

able 3 shows the relative intensities for(

1 1 1)

and (2 0 0) dif-racting planes of m-ZrO2 for all oxide films. It was found that theesultant intensity ratios of

(1 1 1

)and (2 0 0) for all the films are

ot in accordance with the standard JCPDS card no 37-1484 of m-rO2 where, the relative intensity value of 100 for

(1 1 1

)and 21

or (2 0 0) planes, resulting in the intensity ratio of 4.8, which isery less in the fabricated oxide films, suggests that the oxide filmsormed by PEO in the present study are preferentially oriented. The1, S2, S4 and S5 oxide films had a

(1 1 1

)m

orientation and has

elative intensity values of 100 for(

1 1 1)

mand 55, 56, 67 and 56

or (2 0 0)m planes, respectively. In contrast, the S3 film had a (2 0 0)rientation and has relative intensity value of 79 for

(1 1 1

)m

and00 for (2 0 0) planes. The results reveal that when the silicate iononcentration in the electrolyte increases, the growth of oxide filmshrough (2 0 0) plane increases and the oxide film formed in higherilicate containing electrolyte (S3 film) grown completely via (2 0 0)rientation. Thus, the presence of silicate in the electrolyte hastrongly changed crystal orientation during oxide film growth pro-ess. Though, many researchers have fabricated oxide films on Zrnd its alloys in silicate containing electrolyte by PEO using eitherC or DC power source, the growth of m-ZrO2 films toward (2 0 0)rientation was not observed in their studies [6,19–27]. Wang et al.5] fabricated m-ZrO2 films on Zr–2.5Nb alloy in silicate electrolytey PEO at a constant duty cycle of 80%, voltage of 400 V and in arequency range of 1000 Hz to 4000 Hz. In contrast to the aboveeports, all the films have shown strong orientation toward (2 0 0)lane irrespective of the applied frequencies. In addition, the pre-erred growth of m-ZrO2 films toward (2 0 0) orientation was alsobserved during the fabrication of ZrO2 thin films on Si substrate byetallo-organic chemical vapor deposition by Jouli et al. [38]. They

eported that the observed oxide film growth orientation might be

ssociated with thermodynamic phenomena that were involveduring the process. Upon careful observation of all the experimen-al conditions that were employed in previous PEO studies and inhe present study, it can be deduced that the orientation changes

4 20.1 100 673 19.7 100 56

in oxide films on Zr in silicate containing electrolyte is probablyassociated with the effect of duty cycle. Although, the concen-tration of electrolyte and PEO electrical parameters employed isdifferent compared to the parameters maintained in existing lit-erature, all the authors have worked at low duty cycles (26–30%).Thus, the high duty cycle that was employed in the present study(95%) and in the study of Wang et al. (80%) [5] had significantlychanged the growth of oxide films toward (2 0 0) orientation. Thisis in accordance with the plasma arc discharge characteristics in thethermal history of the already grown oxide film. It should be notedthat although the electrolyte kept at constant room temperature,the presence of high intense sparks at discharge channels at highvoltages provides higher thermal energy and increases the local-ized temperature at the substrate. In general, at higher duty cyclesthe localized heat increases with increasing PEO oxidation timeresulted in increased surface mobility which might be contribut-ing to surface atom rearrangement. Thus, at a pulse duty of 95%,at low voltages observed during the process, the oxide films had a(

1 1 1)

orientation, and the oxide films formed at higher voltagesgrown toward (2 0 0) orientation.

In addition to oxide phases, diffraction peaks that correspond toZr substrate (JCPDS card no. 05-0665) were also found from the XRDpatterns shown in Fig. 2. It can be seen from Fig. 2 that the intensityof (0 0 2) and (1 0 1) planes corresponding to Zr substrate (highestintensity peaks of Zr, indexed by the symbol ‘•’, Fig. 2) is very lowfor S1 and S2 films, where as their intensity is high for S3 and S4films and is again found to be low for S5 film. This can be attributedto the high thickness of the oxide films in phosphate based elec-trolytes. It can also be ascribed to the less porous structure thatcould prevent the incident X-ray beam reaching the substrate. Thecrystallite size of m-ZrO2 and t-ZrO2 phases in all films were eval-uated from the FWHM values of

(1 1 1

)m

and (1 0 1)t reflections(Fig. 2) by using Eq. (1) and the obtained values are reported inTable 3. It can be observed from Table 3 that crystallite size is higherfor m-ZrO2 phase and lower for t-ZrO2 phase, irrespective of elec-trolyte. Further, crystallite size of both m-ZrO2 and t-ZrO2 phasesformed in phosphate containing electrolytes (S1 and S2) are foundto be higher than formed in silicate containing electrolytes (S3 andS4). The variations in crystallite size of m-ZrO2 and t-ZrO2 phasesand its orientation clearly indicate that the film growth process andits thermal history are greatly influenced by the composition of theelectrolytes.

3.3. Surface morphology and composition of oxide films

The surface morphology of the S1–S5 oxide films are shown inFig. 3. All the films showed significantly different morphologicalfeatures depending on the discharge characteristics during the filmgrowth process. S1 and S2 oxide films formed in phosphate elec-

trolyte showed a smooth surface with very fine pore morphology.Further, the addition of KOH to phosphate electrolyte completelyreduced the porosity and thus highly dense oxide films are formed.The SEM micrographs of S1 film shows the presence of fine

S. M et al. / Applied Surface Science 317 (2014) 198–209 203

ted S1

dtpntfmabcrtrapoaattSo

wmse

Fig. 3. SEM surface micrographs of PEO fabrica

ischarge channels appearing as circular pores distributed all overhe surface having an approximate diameter of 2–3 �m and a fewores of 5–10 �m diameter is observed. Whereas, for S2 film theumber of discharge channels increases while their size decreaseso 2 �m or less. Thus, the addition of KOH to phosphate electrolyteormed dense oxide film with fine pore morphology. The fine pore

orphology in S2 film can be attributed to the easy film formationnd lower breakdown voltages as interpreted in Section 3.1. It cane seen from Fig. 3 that the morphology of oxide films changedompletely in silicate electrolyte. S3 and S4 films showed bothough and smooth regions covering the surface randomly. Further,he addition of KOH to the silicate electrolyte decreased the rougheregions as shown in Fig. 3. Thus, the morphology of the oxide filmsppears to be rougher with the silicate electrolyte system com-ared to the phosphate electrolyte system. Further, EDS analysisf the S3 oxide film shows higher silicon content (22 wt%) in roughnd porous regions and lower silicon content (4.5 wt%) in smoothnd dense regions. The surface morphology of S5 film is similar tohat of S2 film with fine discharge channels surrounding all overhe surface. Notably, porous structure that was observed in S3 and4 films is not apparent in S5 film. Though porosity variations arebserved, none of the films showed micro cracks on its surfaces.

The elemental composition and thickness of the formed films

ere studied and the surface EDS spectrum and cross-sectionalicrographs of the oxide films are presented in Fig. 4. Cross-

ectional images show that all the oxide films formed in differentlectrolytes in the present study are highly dense and uniform in

–S5 films in different electrolyte compositions.

nature. The thickness values obtained at the end of 6 min of treat-ment time is found to be 9.5 ± 0.7 �m, 6.3 ± 0.4 �m, 7 ± 1.1 �m,5.7 ± 0.9 �m and 6.9 ± 0.8 �m for S1–S5 films, respectively. All thefilms grow at a very low rate, indicating a low current efficiency.The thickness of oxide films formed in phosphate electrolyte are1–2 �m higher than the oxide films formed in silicate containingelectrolyte. Further, it was observed that addition of KOH addedto electrolyte reduces the thickness of the oxide films by 2–3 �m,irrespective of the base electrolyte. The higher thickness of S1 andS3 oxide films is due to the presence of higher spark voltages dur-ing PEO treatment. Higher voltages cause sparks with high levelof energy which produces molten materials with high temperaturethat is easier to erupt and form a thick oxide film. The S5 film formedin mixed electrolyte system containing phosphate + silicate + KOH,is having a thickness of 6.9 ± 0.8 �m higher than S2 and S4 andlower than S1 and S3 films.

The EDS spectrum of oxide films (Fig. 4) revealed the existenceof zirconium, oxygen and electrolyte borne elements. The elemen-tal compositions obtained from the spectra are reported in Table 4.It can be observed from Fig. 4 and Table 4 that the electrolyte borneelement P is incorporated into S1, S2 and S5 oxide films and the Pcontent tends to decrease with a decrease in phosphate concentra-tion in the electrolyte. Similarly, S3 and S4 films show the presence

of Si, and the content of Si tends to decrease with a decrease in sili-cate concentration in the electrolyte. Further, the S5 film formedin mixed electrolyte shows the presence of both Si and P in itsspectrum. This is in agreement with the reported fact that higher

204 S. M et al. / Applied Surface Science 317 (2014) 198–209

Fig. 4. SEM-EDS spectra and cross-sectional SEM micrographs obtained from

Table 4Element concentration measured by SEM-EDS at the surfaces of PEO fabricated filmsshown in Fig. 3.

Sample Element (at.%)

Zr O P Si K Na

S1 29.96 61.14 6.73 0 0 2.17S2 35.02 58.34 4.63 0 00.75 1.26S3 24.42 61.10 0 10.84 0 3.63

trmsoieml

3

wvmiS0fistt

S4 27.80 59.80 0 9.02 1.27 2.11S5 37.38 52.72 4.52 3.51 00.70 1.17

he concentration of phosphate/silicate in the electrolyte solutionesults in PEO discharges, which can sinter electrolyte borne ele-ents into the film [13]. Though the SEM-EDS spectra show a

ignificant amount of electrolyte borne elements like P and Si inxide films in respective electrolytes, no separate P and Si contain-ng phases were found in the XRD patterns (Fig. 2) indicating thatither the electrolyte borne elements might be doped into the ZrO2atrix or the formed phases, if any, might be less than the detection

imit of XRD.

.4. Surface roughness measurement of PEO fabricated oxide films

The average surface roughness (Ra) of PEO fabricated oxide filmsere studied using surface roughness gauge and their obtained

alues are reported in Table 2. Ra of Zr substrate prior to PEO treat-ent was 0.4 ± 0.05 �m (not shown in Table 2). The PEO treatment

ncreases the surface roughness and the roughness (Ra) of S1, S2,3, S4 and S5 oxide films is 0.60 ± 0.05, 0.48 ± 0.03, 0.72 ± 0.09,.70 ± 0.07 and 0.65 ± 0.06, respectively. The roughness values of

lms formed in silicate containing electrolyte (S3 and S4) areignificantly higher than that of films formed in phosphate con-aining electrolytes (S1 and S5) which can also be evident fromheir respective SEM surface micrographs shown in Fig. 3 and

the secondary electron (SE) signals of PEO treated S1–S5 oxide films.

discussed in Section 3.3. Further, the addition of KOH to base elec-trolytes reduced the surface roughness of the films, irrespectiveof the electrolyte. From Table 2, it is clear that S3 sample exhibitssignificantly increased surface roughness due to the presence ofhigh surface porosity of the films shown in Fig. 3, in comparisonto other films. Thus, electrolyte composition has a strong effecton surface roughness of the PEO treated samples. It was reportedin literature that implant surface having 0.3 to 2 �m roughness isexpected to improve bone responses favorably [30]. Thus, in thepresent study, all the films are expected to exhibit good responsewith osteoblast cells.

3.5. Wettability of PEO fabricated oxide films

The wettability of untreated Zr and PEO fabricated films wasevaluated by contact angle method. Fig. 5 shows the contact anglevalues and respective water droplet images of S, S1, S2, S3, S4and S5 samples. The contact angles of PEO films are 37.4 ± 0.9◦,85.8 ± 0.5◦, 5.5 ± 1.2◦, 7.6 ± 1.1◦ and 31 ± 0.7◦ for S1, S2, S3, S4 andS5 samples, and 60.5 ± 2.6◦ for untreated Zr indicating that all thePEO fabricated films are hydrophilic. Except S2, all the PEO treatedfilms shows good wettability than untreated Zr. A decrease in wett-ability of S2 film compared to other oxide films is be due to thepresence of low surface roughness (0.48 ± 0.03 �m) and pore freemorphology of the film (shown in Fig. 3) with the addition of KOHto phosphate electrolyte. Thus, the electrolyte composition, notablymodified the wettability of the oxide films formed. The high wett-ability showed by S1, S3, S4 and S5 films with Zr substrate is apositive indicator of apatite growth in-vitro and good cell attach-ment and spreading in-vivo [39]. The surface energy of untreatedZr and PEO formed films was calculated using the Eq. (2) and the

obtained values were reported in Table 2. The surface energy of PEOfilms are S1 57.8 ± 1, S2 5.33 ± 0.4, S3 72.46 ± 1.2, S4 72.16 ± 1.1, S562.40 ± 0.8 mJ/m2 and for Zr substrate is 35.8 ± 2.5 mJ/m2. The S3and S4 films which are having higher surface wettability are found

S. M et al. / Applied Surface Scie

FZ

tloriiofptu

3

3

ee[msaTiwspsfndriaif4sfuffi−−

ig. 5. Contact angle values and respective water droplet images on the surfaces ofr substrate and PEO treated Zr films in different electrolytes.

o have higher surface energy and S2 and untreated Zr are havingower surface energy. This is due to the fact that the surface energyf any material is directly proportional to its wettability. It has beeneported in literature, the surface free energy plays an essential rolen guiding the first events occurring at the biomaterial/biologicalnterface (such as interaction and adsorption of water and proteinsn implant surfaces) and in general, cell adhesion on implant sur-ace is proportional to their surface free energy [40]. Thus, in theresent study, the initial cell attachment and spreading is expectedo be high on S3 and S4 films compared to the other films andntreated Zr.

.6. In-vitro electrochemical corrosion characteristics

.6.1. Open circuit potential (OCP) measurementThe thermodynamic tendency of an implant material to undergo

lectrochemical reactions within the body fluid environment isssential in order to understand their stability in the human body28]. One simple way to study the resistance offered by the implant

aterial in body fluids is to monitor its OCP as a function of immer-ion time. The higher the OCP of the material, the lower is itsffinity to participate in the electrochemical corrosion reactions.hus, OCP of formed oxide films in different electrolytes was stud-ed by immersing in SBF medium for 4 h, and the obtained results

ere compared with that of Zr substrate. OCP curves of Zr sub-trate and PEO treated S1–S5 films tested for 4 h in SBF medium arerovided in Fig. 6(a) and the potential values attained at an immer-ion period of 4 h are reported in Table 5. It can be clearly seenrom Fig. 6(a) that OCP values of all oxide films shifted toward theoble direction compared to Zr substrate which indicates thermo-ynamic stability of oxide films in taking part in electrochemicaleactions. Thus, the oxide films formed by PEO on Zr substratenhibited the cathodic reaction of water reduction [13]. Further,ll oxide films showed small instability in the initial 60 min ofmmersion time irrespective of the electrolyte employed for filmormation, and then maintained almost stable OCP values up to

h. The instability in the initial period of all oxide films repre-ents increased activity due to the defective behavior of the newlyormed layer at the metal/oxide interface, and the steady OCP val-es attained at long exposure times (4 h) might be attributed to the

ormation of a passive film on its surface. The OCP values of oxidelms obtained after 4 h immersion period in SBF are −241 V, −86 V,291 V, −282 V and −254 V for S1, S2, S3, S4 and S5 samples, and491 V for Zr substrate. Among all oxide films, OCP of S2 film shifted

nce 317 (2014) 198–209 205

toward more noble direction and showed almost a constant higherOCP value over the entire period of immersion time, which can beattributed due to the presence of pore free morphology (shownin Fig. 3) with a lower surface roughness (Section 3.4). Thus, theaddition of KOH to phosphate electrolyte significantly improvedthe thermodynamic stability of the S2 film in SBF medium. On theother hand, S1, S3, S4 and S5 films showed almost similar OCP val-ues lower than the S2 film and higher than the S (Zr substrate).Thus, from the OCP results, the thermodynamic tendency of PEOfilms toward the electrochemical oxidation reaction increases inthe order of S2 < S1 < S5 < S4 < S3 < S and among all the samples, S2film has shown more noble behavior toward the anodic reactions.

3.6.2. Tafel extrapolation methodThe Tafel extrapolation method is used to find out the corro-

sion current density (jcorr) from cathodic and anodic slopes (ˇc

and ˇa). The Tafel plots are obtained over a potential range of±200 mV with reference to the stable OCP and the derived kineticparameter values are reported in Table 5. From the reportedvalues it can be observed that after PEO treatment, the jcorr

values of oxide films are significantly lower than that of the Zrsubstrate (6.2 × 10−2 �A/cm2). Among the films, the jcorr valueof S2 film was lower at a value of 2.8 × 10−4 �A/cm2 and jcorr

value of S3 film was higher at a value of 2.89 × 10−3 �A/cm2.The polarization resistance (Rp) of oxide films and Zrsubstrate were calculated using Eq. (3) and the obtained val-ues are reported in Table 5. In accordance with the jcorr values, alloxide films showed higher polarization resistance in 7.4 pH SBFmedium. Further, Rp values of S1, S3, S4 and S5 films are found tobe two orders of magnitude higher than Zr substrate. S2 film offershigher Rp value of 1719.7 × 102 k� cm2, which is three orders ofmagnitude higher than that of untreated Zr whose Rp value is558.9 k� cm2. Thus, the S2 oxide film with pore free morphologyacts as a barrier toward the extremely corrosive Cl− ions thatare there in the SBF solution thereby enhancing the corrosionresistance of Zr. It is notable that although the films formed insilicate containing electrolyte S3 and S4 are having higher porosityas depicted from the surface SEM morphology shown in Fig. 3 anddiscussed in Section 3.3, they offered higher corrosion resistancethan Zr substrate and slightly lower than the other oxide filmswith lower porosity. From the obtained Rp values, the protectionefficiency (PE) of all oxide films and Zr substrate were calculatedusing Eq. (4) and are reported in Table 5. PE of oxide films isfound to be 98, 99.7, 95.3, 96.2 and 96.4% for S1–S5, respectively.Thus, all oxide films showed good protection efficiency towardsaturated SBF ions and by considering the jcorr, Rp, corrosion rateand passivation behavior of substrate and PEO oxide films, theelectrochemical resistance toward corrosion is found to increasein the order of S < S3 < S4 < S5 < S1 < S2.

3.6.3. Potentiodynamic polarization studiesThe passivation behavior of the untreated Zr and the PEO fab-

ricated oxide films in a 7.4 pH SBF environment were studied byconducting PDP test over a potential range of −500 mV to 3000 mVvs SCE. Various passivation parameters, namely active potential,passive potential, breakdown potential and passivation range oftest samples can be obtained using this PDP curve in order toanalyze their protective nature toward the test medium by apply-ing a wide range of potential. The PDP curves of the S, S1, S2,S3, S4 and S5 samples are displayed in Fig. 6(b). From Fig. 6(b)it can be observed that the Zr substrate (S) shows four regionswithin the applied potential range, namely, active region, passive

region, breakdown region and repassivation region. In the begin-ning of anodic potential up to 30 mV, the current density increaseslinearly, then Zr undergo passivation by forming a passive layeron its surface until 450 mV. At 450 mV, there is a sudden rise

206 S. M et al. / Applied Surface Science 317 (2014) 198–209

) curve

ibtuuddoc

Fa

Fig. 6. Open circuit potential (a) and potentiodynamic polarization (b

n the current density in a short range of potential which mighte due to the breakdown of the formed passive film and oncehe potential reached 758 mV, repassivation takes place where aniform current density is maintained up to 2500 mV. Thus, Zrndergoes pitting potential at 450 mV with the protection interval,

efined as the difference Epitt − Ecorr is 956 mV and pitting interval,efined as the difference Erepass − Epitt is 308 mV. All the PEO formedxide films showed improved corrosion resistance with lowesturrent density and the highest corrosion potential compared to

ig. 7. SEM micrographs on the surfaces of untreated Zr and PEO treated S1–S5 samplesll samples. S5′ represents higher magnification image showing the fine network structu

s of untreated Zr and PEO formed S1–S5 films in 7.4 pH SBF medium.

Zr substrate. Thus, the formed oxide films are less vulnerable tocorrosion in the SBF environment. Additionally, among all oxidefilms, S1, S2, S4 and S5 films exhibited excellent pitting corrosionresistance over a potential range of present experimental con-ditions. In contrast, S3 film shows a similar trend with the Zr

substrate. It can be seen from Fig. 6(b) that S3 film is active in thebeginning of anodic potential and undergoes pitting corrosion at351 mV with a protection interval of 648 mV. When the polariza-tion potential is over 554 mV, the repassivation takes place at the

soaked in SBF for 14 days. Apatite particle formation can be seen on the surface ofre of apatite particle formed on S5 film.

S. M et al. / Applied Surface Science 317 (2014) 198–209 207

Table 5The thermodynamic and kinetic parameter values for substrate (S) and PEO coated S1–S5 samples obtained from open circuit potential and potentiodynamic polarizationtests.

Sample OCP (mV) vs SCE ˇa (mV) |ˇc| (mV) jcorr (�A/cm2) Corrosion rate (mm/year) Rp (k� cm2) Protectionefficiency (%)

S −491 164 146 6.2 × 10−2 13.5 × 10−2 558.9 –S1 −241 198 133 1.23 × 10−3 1.63 × 10−4 287.8 × 102 98S2 −86 283 110 2.8 × 10−4 3.81 × 10−5 1719.7 × 102 99.7S3 −291 222 116 2.89 × 10−3 4.53 × 10−4 118.1 × 102 95.3S4 −282 224 114 2.22 × 10−3 7.61 × 10−4 149.1 × 102 96.2S5 −254 210 119 2.11 × 10−3 6.39 × 10−4 157.0 × 102 96.4

Table 6Elemental concentration of untreated Zr and PEO fabricated S1–S5 films after soaking in SBF for 14 days as shown in Fig. 7.

Sample Element (at.%)

ZrL OK KK SiK CaK PK ClK NaK MgK CK

S 10.25 54.65 0.05 – 14.30 9.55 1.16 1.70 0.92 7.42S1 4.16 42.84 0.09 – 13.05 8.52 5.74 5.06 0.79 19.75S2 4.99 50.70 0.12 – 15.76 10.02 1.10 1.52 1.03 14.76

.95

.74

.69

fiAuacfiraSofia3i

3

t

S3 4.77 33.57 0.08 1.30 11S4 6.35 44.45 0.17 0.78 13S5 4.27 43.57 0.07 1.48 11

lm/Zr substrate interface under the porous oxide film of S3 sample.lthough, S3 film undergoes pitting attack similar to untreated Zr, itndergoes repassivation within short intervals of applied potentialnd the repassivation potential of S3 film improves about 200 mVompared to Zr substrate. The formation of pitting corrosion of S3lm is strongly dependent on its surface morphology and surfaceoughness (shown in Fig. 3 and reported in Table 2). More porousnd rough the surface is, more will be the exposed area towardBF, thus corrosive ions from SBF can enter inside, and reacts withxide film thereby forming pits on its surface. Among all the oxidelms, S2 film showed higher corrosion resistance with lower jcorr

nd noble Ecorr values without undergoing pitting potential within000 mV, thus, denser films with less pore morphology are helpful

n improving the corrosion resistance.

.7. Bioactivity of PEO fabricated ZrO2 films

Generally, an outstanding implant surface should have proper-ies such as superior bioactivity and biocompatibility. In particular

Fig. 8. Surface SEM micrographs showing human osteosarcoma (HOS) cells adhesio

7.25 10.29 9.81 0.99 19.998.46 2.75 2.65 0.59 20.067.30 1.21 1.75 0.73 27.93

for dental and orthopedic applications, the implant surface shouldbe capable to speed up new bone formation and bond with bonesteadily and firmly at the whole period of implantation [41]. Inorder to study the apatite forming ability of PEO films and Zr sub-strate, they are immersed in 7.4 pH SBF medium for 14 days and thesurfaces were examined with SEM-EDS to check the mineralizationof apatite layer on its surface and its elemental composition. Fig. 7shows the SEM surface morphology of Zr substrate and PEO oxidefilms after immersion in SBF for 14 days and their elemental com-position is reported in Table 6. The appearance of higher content ofCa, P, Mg, Na and K and the decrease in the at% for Zr (Table 6) indi-cating that the apatite formed on all oxide films is significantly thickin nature. It can be seen from Fig. 7 that spherical like deposits wereformed on untreated Zr and EDS results shows that these depositswere rich in Ca and P elements with Ca/P ratio of 1.50 which is

low compared to Ca/P ratio of stochiometric hydroxyapatite (1.67).Further, Fig. 7 shows the Ca–P layer formed on Zr substrate is notuniform. Thus, the apatite forming ability of untreated Zr in SBF’senvironment is found to be poor. Contrary, spheres like deposits

n and spreading over PEO treated S1–S5 films and untreated Zr (S) after 48 h.

2 ce Scie

aisharthccrttbsrfp

3

pooSiwogatatr

pata

4

oatoienWisrcpafswtWme

[

[

[

[

[

[

[

[

[

[

[

[

[

08 S. M et al. / Applied Surfa

re observed on all PEO oxide films (Fig. 7) and the whole surfaces covered by a uniform layer indicating that the PEO films showuperior bioactivity. EDS results of all PEO films (Table 6) show aigh content of Ca and P elements with Ca/P ratio between 1.53nd 1.65, which is nearly equal to Ca/P ratio of 1.67 for stochiomet-ic hydroxyapatite. Ca/P ratio in oxide films is found to be lowerhan 1.67, suggested that the formed apatite is a calcium deficientydroxyapatite. The mineral component of bone is essentially aalcium deficient hydroxyapatite with a Ca/P ratio <1.67, which isompositionally different, but, structurally similar to stoichiomet-ic hydroxyapatite [42]. It was reported in the literature [28] thathe biomimetic apatites have comparatively low Ca/P ratio owingo the lack of Ca2+ ions in the apatite crystal as the Ca2+ ions coulde substituted by K+, Na+ and Mg2+ ions that are present in SBFolution. Moreover, the calcium deficient hydroxyapatite with Ca/Patios ranging from 1.67 to 1.33 originate from the loss of Ca2+ ionsrom the unit cell have varying biodegradable characteristics andlay an important role in bone remodeling [42].

.8. Cell adhesion characteristics of oxide films

The cell adhesion test with HOS cells on PEO treated S1–S5 sam-les and untreated Zr after 48 h of culture (Fig. 8) shows that allxide films and untreated Zr were able to support the attachmentf cells. However, the HOS cell adhesion and spreading on S3 and4 samples was significantly higher than that of S1 and S2 samples,ndicating that Si incorporated oxide films having porous structure

as capable to promote cell spreading better than P incorporatedxide films with pore free morphology. Further, the SEM micro-raph (Fig. 8) of S5 film showed that more HOS cells were attachednd spread well on its surface. Thus, the relatively rough and wet-able nature of S5 film offered favorable sites for cell attachmentnd spreading. It thus further supports the hypothesis that elec-rolyte composition and concentration clearly played importantoles in the HOS cell responses.

Thus, surface modification of Zr in mixed electrolyte havinghosphate, silicate and KOH with good wettability and considerablemount of surface roughness augments HOS cell response in addi-ion to excellent pitting corrosion resistance and apatite formingbility.

. Conclusions

Electrolyte composition was found to have a substantial rolen PEO voltage responses such as the breakdown voltage (Vb)nd final voltage (Vf) which in turn had a strong influence onhe structural and morphological characteristics of the formedxide films. Surface porosity and roughness are found to be lown oxide films formed in phosphate electrolyte than in silicatelectrolyte. Further, the addition of KOH to base electrolyte sig-ificantly reduced the porosity and roughness of the oxide films.ettability and surface energy are higher for the films formed

n silicate containing electrolyte which are related to their higherurface porosity and roughness. All PEO oxide films showed supe-ior pitting corrosion resistance and higher apatite forming abilityompared to untreated Zr. The film formed in phosphate and phos-hate + KOH electrolyte exhibited highest electrochemical stabilitynd least corrosion current density, which is ascribed to its poreree morphology. In contrary, the films formed in silicate andilicate + KOH electrolyte showed higher apatite forming abilityith good HOS cell adhesion and spreading, which is attributed

o the superior wettability and surface energy of their surfaces.hereas, the oxide film formed in the phosphate + silicate + KOHixed electrolyte was found to be optimum one with higher surface

nergy and roughness, good wettability and bioactivity, resistance

[

nce 317 (2014) 198–209

toward pitting corrosion and superior osteoblast cell attachmentand spreading.

Acknowledgements

The authors would like to acknowledge the facilities procuredthrough the grants received from the Department of Scienceand Technology, New Delhi (SR/S3/ME/0024/2011, dated 3rdJuly 2012) and Department of Biotechnology, New Delhi (BT/PR-1731/MED/32/99/2008, dated 19th August 2009).

References

[1] S. Pilathadka, D. Vahalova, T. Vosahlo, The zirconia: a new dental ceramic mate-rial. An overview, Prague Med. Rep. 108 (2007) 5–12.

[2] T. Vagkopoulou, S.O. Koutayas, P. Koidis, J.R. Strub, Zirconia in dentistry PartI—Discovering the nature of an upcoming bioceramic, Eur. J. Esthet. Dent. 4(2009) 1–23.

[3] Y.T. Liu, T.M. Lee, T.Z. Lui, Enhanced osteoblastic cell response on zirconia bybio-inspired surface modification, Colloids Surf., B: Biointerfaces 106 (2013)37–45.

[4] A.G. Sanchez, W. Schreiner, G. Duffo, S. Cere, Surface characterization ofanodized zirconium for biomedical applications, Appl. Surf. Sci. 257 (2011)6397–6405.

[5] L. Wang, X. Hu, X. Nie, Deposition and properties of zirconia coatings on a zir-conium alloy produced by pulsed DC plasma electrolytic oxidation, Surf. Coat.Technol. 221 (2013) 150–157.

[6] W. Simka, M. Sowa, R.P. Socha, A. Maciej, J. Michalska, Anodic oxidation ofzirconium in silicate solutions, Electrochim. Acta 104 (2013) 518–525.

[7] U. Leushake, T. Krell, U. Schulz, M. Peters, W.A. Kaysser, B.H. Rabin, Microstruc-ture and phase stability of EB-PVD alumina and alumina/zirconia for thermalbarrier coating applications, Surf. Coat. Technol. 94–95 (1997) 131–136.

[8] R. Brenier, J. Mugnier, E. Mirica, XPS study of amorphous zirconium oxide filmsprepared by sol–gel, Appl. Surf. Sci. 143 (1999) 85–91.

[9] G. Shanmugavelayutham, S. Yano, A. Kobayashi, Microstructural characteriza-tion and properties of ZrO2/Al2O3 thermal barrier coatings by gas tunnel-typeplasma spraying, Vacuum 80 (2006) 1336–1340.

10] F.Y. Zhou, B.L. Wang, K.J. Qiu, H.F. Li, L. Li, Y.F. Zheng, Y. Han, In vitro corrosionbehavior and cellular response of thermally oxidized Zr–3Sn alloy, Appl. Surf.Sci. 265 (2013) 878–888.

11] M.D. Klapkiv, N.Y. Povstyana, H.M. Nykyforchyn, Production of conversionoxide-ceramic coatings on zirconium and titanium alloys, Mater. Sci. 42 (2006)277–286.

12] L.R. Krishna, A.S. Purnima, N.P. Wasekar, G. Sundararajan, Kinetics and proper-ties of micro arc oxidation coatings deposited on commercial Al alloys, Metall.Mater. Trans. A 38 (2007) 370–378.

13] K. Venkateswarlu, N. Rameshbabu, D. Sreekanth, A.C. Bose, V. Muthupandi, N.K.Babu, S. Subramanian, Role of electrolyte additives on in-vitro electrochemicalbehavior of micro arc oxidized titania films on Cp Ti, Appl. Surf. Sci. 258 (2012)6853–6863.

14] L.R. Krishna, G. Poshal, G. Sundararajan, Influence of electrolyte chemistry onmorphology and corrosion resistance of micro arc oxidation coatings depositedon magnesium, Metall. Mater. Trans. A 41 (2010) 3499–3508.

15] J.S. Santos, F.T. Strixino, E.C. Pereira, The influence of experimental conditionson the morphology and phase composition of Nb doped ZrO2 films preparedby spark anodization, Corros. Sci. 73 (2013) 99–106.

16] A.B. Ragov, A.I. Slonova, I.V. Mironov, The influence of homogeneous electrolytecomposition on micro plasma synthesis and characteristics of Fe containingcoatings on A1050 alloy, Appl. Surf. Sci. 287 (2013) 22–29.

17] L. Wang, L. Shi, J. Chen, Z. Shi, L. Ren, Y. Wang, Biocompatibility of Si incorporatedTiO2 film prepared by micro arc oxidation, Mater. Lett. 116 (2014) 35–38.

18] X. Qiu, P. Wan, L. Tan, X. Fan, K. Yang, Preliminary research on a novel bioac-tive silicon doped calcium phosphate coating on an AZ31 magnesium alloy viaelectrodeposition, Mater. Sci. Eng., C 36 (2014) 65–76.

19] Y. Cheng, J. Cao, Z. Peng, Q. Wang, E. Matykina, P. Skeldon, G.E. Thompson,Wear-resistant coatings formed on zircaloy-2 by plasma electrolytic oxidationin sodium aluminate electrolytes, Electrochim. Acta 116 (2014) 453–466.

20] Y. Cheng, F. Wu, J. Dong, X. Wu, Z. Xue, E. Matykina, P. Skeldon, G.E. Thompson,Comparison of plasma electrolytic oxidation of zirconium alloy in silicate andaluminate based electrolytes and wear properties of the resulting coatings,Electrochim. Acta 85 (2012) 25–332.

21] Y. Cheng, F. Wu, E. Matykina, P. Skeldon, G.E. Thompson, The influences ofmicro discharge types and silicate on the morphologies and phase compositionsof plasma electrolytic oxidation coatings on zircaloy-2, Corros. Sci. 59 (2012)307–315.

22] Y. Cheng, W.U. Fan, Plasma electrolytic oxidation of zircaloy-4 alloy with DC

regime and properties of coatings, Trans. Nonferrous Met. Soc. China 22 (2012)1638–1646.

23] Z. Zou, W. Xue, X. Jia, J. Du, R. Wang, L. Weng, Effect of voltage on propertiesof microarc oxidation films prepared in phosphate electrolyte on Zr–1Nb alloy,Surf. Coat. Technol. 222 (2013) 62–67.

ce Scie

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[on the biological properties of MAO formed ZrO2, Colloids Surf., B: Biointerfaces

S. M et al. / Applied Surfa

24] Y. Chen, X. Nie, D.O. Northwood, Investigation of plasma electrolytic oxida-tion coatings on Zr–2.5Nb alloy using high temperature/pressure autoclave andtribological tests, Surf. Coat. Technol. 205 (2010) 1774–1782.

25] E. Matykina, R. Arrabal, P. Skeldon, G.E. Thompson, P. Wang, P. Wood, Plasmaelectrolytic oxidation of a zirconium alloy under AC conditions, Surf. Coat.Technol. 204 (2010) 2142–2151.

26] S. Cengiz, Y. Gencer, The characterization of the oxide based coating synthesizedon pure zirconium by plasma electrolytic oxidation, Surf. Coat. Technol. 242(2014) 132–140.

27] S. Stojadinovic, R. Vasilic, M. Petkovic, I. Belca, B. Kasalica, M. Peric, L. Zekovic,Luminescence during the anodization of zirconium, Electrochim. Acta 79 (2012)133–140.

28] M. Sandhyarani, N. Rameshbabu, K. Venkateswarlu, D. Sreekanth, Ch. Subrah-manyam, Surface morphology, corrosion resistance and in-vitro bioactivity ofP containing ZrO2 films formed on Zr by plasma electrolytic oxidation, J. AlloysCompd. 553 (2013) 324–332.

29] L.C. Ann, S. Mahmud, S.K.M. Bakhori, A. Sirelkhatim, D. Mohamad, H. Hasan,A. Seeni, R.A. Rahman, Effect of surface modification and UVA photoactivationon antibacterial bioactivity of zinc oxide powder, Appl. Surf. Sci. 292 (2014)405–412.

30] B.R. Sunil, A.A. Kumar, T.S. Sampathkumar, U. Chakkingal, Role of biomineral-ization on the degradation of fine grained AZ31 magnesium alloy processed bygroove pressing, Mater. Sci. Eng., C 33 (2013) 1607–1615.

31] T. Kokubo, H. Takadamo, How useful is SBF in predicting in-vivo bone bioactiv-

ity, Biomaterials 27 (2006) 2907–2915.

32] A.S. Toloei, V. Stoilov, D.O. Northwood, A new approach to combating corrosionof metallic materials, Appl. Surf. Sci. 284 (2013) 242–247.

33] A. Madhankumar, N. Rajendran, A promising copolymer of p-phenylendiamineand o-aminophenol: chemical and electrochemical synthesis, characterization

[

nce 317 (2014) 198–209 209

and its corrosion protection aspect on mild steel, Synth. Met. 162 (2012)176–185.

34] T. Kokubo, H. Kushitani, S. Sakka, T. Kitsugi, T. Yamamuro, Solutions able toreproduce in-vivo surface structure changes in bioactive glass-ceramic A–W, J.Biomed. Mater. Res. 24 (1990) 721–734.

35] F. Wang, L. Shi, W.X. He, D. Han, Y. Yan, Z.Y. Niu, S.G. Shi, Bioinspiredmicro/nano fabrication of dental implant-bone interface, Appl. Surf. Sci. 265(2013) 480–488.

36] H. Toraya, M. Yoshimura, S. Shigeyuki, Calibration curve for quantitative analy-sis of the monoclinic-tetragonal ZrO2 systems by X-ray diffraction, J. Am. Ceram.Soc. 67 (1984) 119–121.

37] R.C. Garvie, P.S. Nicholson, Phase analysis in zirconia systems, J. Am. Ceram.Soc. 55 (1972) 303–305.

38] M. Jouili, M. Andrieux, I. Gallet, N. Prudhomme, V. Ji, Texture and residual stressanalysis by XRD on metastable tetragonal zirconia films obtained my MOCVD,ECS Trans. 25 (2009) 413–420.

39] S.A. Pauline, N. Rajendran, Effect of Sr on the bioactivity and corrosion resistanceof nanoporous niobium oxide coating for orthopedic applications, Mater. Sci.Eng., C 36 (2014) 194–205.

40] X. Chen, A. Nouri, Y. Li, J. Lin, H.D. Peter, C. Wen, Effect of surface roughness ofTi, Zr, and TiZr on apatite precipitation from simulated body fluid, Biotechnol.Bioeng. 101 (2008) 378–387.

41] Z. Zhang, K. Wang, C. Bai, X. Li, X. Dang, C. Zhang, The influence of UV irradiation

89 (2012) 40–47.42] S.P. Victor, C.P. Sharma, Poly methacrylic acid modified CDHA nanocomposites

as potential pH responsive drug delivery vehicles, Colloids Surf., B: Biointer-faces 108 (2013) 219–228.