70

Chapter 4

Corrosion and sliding wear behavior of plasma sprayed

nanoceramic coatings for biomedical applications

4.1 INTRODUCTION

Presently ceramics are considered as a better alternative for metallic based

implants, as they offer a number of advantages over metals and alloys. Besides being

biologically inert, ceramics produce very little wear debris and further the

compositions and hence the microstructure can be tailored to match the material

properties to that of the natural bone. The articulating surfaces on hip, knee and

shoulder implants are currently being fabricated using either of the two types of

ceramics, alumina or zirconia, that are scratch-resistant and significantly harder than

the metal combinations (Hulbert et al; 1993, Jerome Chevalier et al; 2004). These two

ceramic materials can also be used on both the ball as well as the socket components

of an implant. One of the disadvantages of using ceramics as implants is the material's

limited lifespan. Even a small crack or porosity may result in catastrophic failure of

ceramic materials and once several batches of implants made of zirconia were

withdrawn due to manufacturing defects (Charles et al; 1995).In addition, ceramic

implants made of zirconia undergo slow degradation during long term implantation in

the human body (Anis Paul et al; 2011, Santos et al; 2004).With orthopedic

procedures increasingly being performed on younger patients, surgeons are eagerly

awaiting for researchers to develop technological breakthroughs that would extend the

life of an implant.

In order to develop an implant with high wear resistance and superior fracture

toughness, one can profitably try to make use of the advantageous properties of both

the ceramics as well as that of metals, as the former exhibits high wear resistance,

while the latter is more ductile and mimics the natural bone. This can be achieved by

the development of ceramic coatings on metallic substrate which will be more suitable

for the manufacturing of new high wear resistant implants. Apart from the

enhancement of wear resistance, ceramic coatings are also found to improve the

osseointegration and hence the service period of the implant.

71

The works of Kabacoff et al and Gell et al showed that the nanostructured

Al2O3-13TiO2 coatings made on some of the mechanical parts such as periscope

piston rod etc used in navy and submarine gave rise to better indentation crack

resistance, adhesion strength, spallation and wear resistance (Gell et al; 2001

Kabacoff et al; 2002). With regard to biomaterials, nano features on the surface

results in superior bone adhesion and strength, as it mimics the porous bone surface. It

is well documented that nanoceramic coatings made using HAP, CNT and TiO2

powders and surfaces tailored with nano topography using different techniques exhibit

excellent biocompatibility (Gadow et al; 2010, Anup kumar keshri et al; 2009,

Webster et al; 2001). It is important to note that the same composites of alumina-

titania namely viz; Al2O3-13TiO2 has the same effect in the field of bioimplants too.

The more recent and significant work of Soo et al, performed on the femoral head

made out of Al2O3-13TiO2 clearly demonstrated that this composition possesses

improved mechanical properties such as hardness, fracture toughness and wear

resistance (Soo Whon Lee et al; 2003). The above conclusions were arrived at by

varying the composition of TiO2.

Though extensive research has been carried out on alumina-titania composite

coatings for other applications, only few attempts have been made on the

development of wear resistant nano Al2O3-13TiO2 based coatings especially using

plasma spray technique for biomedical applications. The work carried out by Richard

et al on commercially pure titanium (Cp titanium) and Ti-13Nb-13Zr (Ti1313)

biomedical alloy, demonstrated that there is an enhancement in fretting wear

resistance of both the above materials, when coated with nano Al2O3-13TiO2 and

ZrO2 powders using plasma spray technique (Richard et al; 2009).

The major drawbacks of plasma spray coating are their inherent porosity and

cracks formed on the coatings. In addition, in thermal barrier coatings, the mismatch

in thermal expansion coefficient between the substrate and coating leads to frequent

failure of the ceramic coatings. Geis et al have observed that the adhesion of the

coatings can be increased considerably by modifying the surface texture or having

suitable bond coat (Geis et al; 2004). NiCrAlY alloy is the commonly used bond coat

for alumina/ zirconia coatings on super alloys. A bond coat is selected in such a way

that the bond coat has low thermal mismatch with the substrate. It is important to note

72

that a suitable bond coating for the ceramics such as alumina and zirconia coated on

biomedical alloys such as titanium alloys, Mg and its alloys as well as other alloys

have not been tried so far. Recent studies have shown that a bilayered coating

consisting of ZrN/Zr on AZ91 Mg alloy exhibited superior corrosion behavior

(YunchangXina et al; 2009). Chang et al have deposited Al2O3 on ZrO2 coated AISI

440C substrate and observed improvement in wear and corrosion resistance when

compared with that of the bare substrate (Chang et al; 2004). Bang-Yen Chow et al,

have performed HAP coating on Ti-6Al-4V alloy using ZrO2 as an intermediate layer

and obtained remarkable improvement in the adhesion strength of the coating (Bang-

Yen Chow et al; 2002) .

The above studies prompted us to make an attempt to deposit Al2O3-13TiO2

and ZrO2 nanoceramic coatings on the Ti-13Nb-13Zr alloy using plasma spray

technique and evaluate their wear and corrosion properties in simulated body

conditions. Further, an effort was made to develop a bilayered coating using both the

above mentioned powders. This chapter describes procedures adopted to develop

thick ceramic coatings and their corrosion and wear behaviors.

4.2 EXPERIMENTAL TECHNIQUES

Atmospheric plasma spraying equipment mainly consists of two electrodes

namely cathode and the anode. Tungsten filament acts as cathode while copper as

anode. When a primary gas especially, argon is allowed to pass between the above

two electrodes, they get ionized and transform into plasma. The plasma generated will

have a high temperature of around 10,000º K to 15,000º K. When the powder is fed

into the plasma flame, it absorbs the heat energy from the plasma flame and gets

melted and finally accelerated towards the substrate thereby forming a coating.

4.2.1 FEED POWDER

Reconstituted nanostructured alumina-titania (Al2O3-13TiO2,), which will be

referred as AT hereafter and 7wt% yttria stabilized zirconia (YSZ) powders were

procured from Inframat Corporation, USA. The procured AT powder consisted of

trace amount of additives such as ZrO2 and CeO2 which were incorporated into the

powder so as to reduce the sintering temperature and enhance the densification

73

process. The composition and the size of the powders used are given in Tables 4.1 and

4.2. The morphologies of both the nano structured AT and 7 wt% YSZ powders are

shown in figure 4.1(a, b). All the coatings were performed on the Ti-13Nb-13Zr alloy

specimens of dimensions of 35 x 20 x 4 mm3.

Table 4.1 Specification of Al2O3-13 wt%TiO2 powder

Al2O3 : TiO2 wt ratio 87 : 13

CeO2 content, wt % 6 – 8 %

ZrO2 content, wt % 8 – 10 %

Powder grain size (ave.) 50 - 500 nm

Agglomerated powder size (ave.) 30 µm

Tap density 2.0 g/cm3

Table 4.2 Specification of 7wt% ZrO2 powder

Y2O3 : ZrO2 wt ratio 7 : 93

Agglomerated powder size 15-150 µm

Tap density 1.4-1.7 g/cm3

(a) (b)

Figure 4.1 SEM Images of (a) AT powder (b) 7 wt% YSZ powder

4.2.2 COATING DEPOSITION

Coating of the reconstituted nanostructured YSZ, AT powders were carried

out using 9 MB Metco Plasma spray system (80 kW) at constant carrier gas flow rate.

74

The process parameters were varied by changing the current from 450 to 600 A and

the voltage from 50 to 55 V. The optimized parameters which yielded thick coating

without spallation that were obtained after repeated trials are given in the Table 4.3.

Bilayered coatings were performed using these optimized parameters, that is, initially

zirconia was coated using the parameters furnished in column A in Table 4.3 and over

this coating, Al2O3-13%TiO2 was coated using the parameters given in the Column B

in Table 4.3.

Table 4.3 Optimized plasma spray parameters used for plasma spraying AT and

YSZ powders

Parameters YSZ coating (A) AT coating(B)

Plasma Current (A) 450 600

Plasma Voltage (V) 55 50

Ar gas flow pressure(NLPM) 42 42

H2 gas flow pressure(NLPM) 8 9

Carrier gas flow (psi) 58 58

Number of spray passes (No) 8 4

Spray distance(cm) 20 20

4.2.3 POWDERS AND COATING CHARACTERIZATION

The morphologies of the starting powders and the coatings were observed by

JEOL JSM-6360 scanning electron microscope (SEM). For cross-sectional SEM

studies, the samples were mounted using bakelite powder followed by grinding and

polishing with emery papers of different grit sizes, ranging from 120 µm to 1600 µm

and mirror finished by 1 µm diamond paste. Phase analysis of the starting powders

and the coatings were performed using Philips 3121 X-ray diffractrometer with CuKα

radiation. The current and voltage were set at 40kV and 20 mA respectively and the

data were collected in the 2θ range from 10ο to 90

ο in a step scan mode with a step of

2ο/min.The phase analysis of the coatings was also carried out using Raman

spectrometer (DILOR-JOBIN-YVON-SPEX). Microhardness measurements were

performed on the cross-section (Y-Z) of the coated samples using micro Vickers

hardness tester. A load of 200g was applied for 15 s and the microhardness values

were measured at five different places across the cross-section and its standard

75

deviation is reported. The thicknesses of the plasma sprayed AT, YSZ and BL-1 are

~200 m, ~200 m and ~200 m respectively. A bilayered coating with 400 m

(200m of YSZ and 200m of AT) was also plasma sprayed (BL-2). The average

surface roughness of the coatings was measured using Talysurf FTS 50 profilometer

(Taylor Hobson Make). Porosity measurements were carried out on the SEM images

using MATERIAL-PRO Software attached with optical microscope. The system is

used to obtain a digitized image of the object. The total area captured by the objective

of the microscope can be measured using MATERIAL-PRO software. Hence the area

covered by the pores is separately measured and the porosity is determined.

4.2.4 ELECTROCHEMICAL CORROSION BEHAVIOR OF COATINGS

Potentiodynamic polarization experiments were carried out on the bare

substrate Ti-13Nb-13Zr alloy (1cm x 1cm) as well as the plasma sprayed coatings in

simulated body fluid (Hank‟s solution) to evaluate their corrosion resistance. All the

potential measurements were made with reference to a saturated calomel electrode

(SCE). A Platinum foil was used as counter electrode and an electrochemical interface

(Gill AC, ACM make) was used for conducting the experiments. For the

electrochemical measurements, the substrate was placed in a Teflon holder, with a 6

mm diameter window and exposed to the solution. Open circuit potential (OCP)–time

measurements were carried out for an hour to achieve a steady open-circuit potential,

which was measured as the corrosion potential. On attaining a constant potential,

potentiodynamic polarization was started from an initial potential of 250 mV below

the OCP. The scan rate used was 0.166 mVs-1

as per ASTM F2129 standards. The

experiments were repeated thrice to check the repeatability. Electrochemical

Impedance Spectroscopy (EIS) measurements were performed with the same

experimental setup used for potentiodynamic polarization studies. The frequency

response analyzer and potentiostat were driven by Z plot software. Impedance

measurements were carried out with a frequency sweep ranging from 10,000 HZ to 1

HZ. The software enables to obtain the best fit from the acquired data, which in turn

led to smooth and reliable curves.

76

4.2.5 SCRATCH TEST

Scratch test was performed using a commercial micro scratch tester (DUCOM,

India) to evaluate the scratch resistance of the coatings following the procedure

adopted by Sung Park et al (Sung Park et al; 2008). A spherical Rockwell C diamond

stylus of 200 µm radius was used to produce the scratch. The test was carried out in

ramp loading mode with the load varying from 20N- 200 N. The loading was varied

in steps of 2 N/mm and the stylus scanned the coating surface perpendicularly at a

speed of 0.5 mm/s. The total length of the scratch scar was 10 mm.

4.2.6 WEAR TEST

The sliding wear test was carried out using reciprocating wear tester TR-285M

machine (DUCOM, India). A ball made of Al2O3 with the diameter of 5.2 mm and

the coated substrate of dimension 35x20x4mm3 was used as a flat in this study. The

wear testing experiments were performed as mentioned in Table 2.3. The parameters

of wear testing are shown in Table 4.4.

Table 4.4 parameters for wear testing

Load applied 10 N

Frequency 2 Hz

Temperature 37º C

Sample dimensions 25mmx20mmx4 mm

Diameter of Alumina ball 5.2 mm

No. of Cycles 105 cycles

Environment Hank‟s solution

4.3 RESULTS AND DISCUSSION

4.3.1 MICROHARDNESS AND MICROSTRUCTURE OF PLASMA SPRAYED

COATINGS

4.3.1.1 MICROHARDNESS OF THE PLASMA SPRAY COATINGS

The micro Vickers hardness values for all the coatings were found to be

substantially higher when compared to that of the uncoated Ti-13Nb-13Zr alloy. The

77

BL coating exhibited significantly higher microhardness value (1096 ± 10 HV) which

is ~ 1.3 times that of AT (820 ± 22 HV) and ~ 1.8 times that of YSZ coatings (617 ±

25 HV). Figure 4.2 shows the variations in hardness of bilayered coating.The increase

in the microhardness of the BL coating is due to the densely packed AT layer over the

ZrO2 coating. Inspite of the fact that same processing parameters were used to coat

AT on YSZ, unlike in single AT coating which consisted of bimodal structure and

some porosity (Figure 4.3), in the present case the top AT coating was found to be

completely melted and adhered well with the ZrO2 layer. Thus the hardness of the BL

coating was considerably higher than the AT or YSZ coatings. The hardness values

measured across the cross section along X- direction are given in Table 4.5.

Table 4.5 Hardness of the plasma sprayed coatings

Coating Hardness (HV)0.2 Hardness (GPa)

Ti-13Nb-13Zr alloy 196 ± 3 1.96± 0.03

ZrO2 617 ± 25 6.17± 0.25

Al2O3-13TiO2 820 ± 22 8.20± 0.22

Bilayered 1096 ± 10 10.96± 0.10

0

200

400

600

800

1000

1200

20 40 60 150 200 Substrate

Depth in micron

Ha

rdn

es

s (

Hv

)

Fig 4.2 Hardness variation of bilayered coating

78

4.3.1.2 MICROSTRUCTURAL ANALYSIS OF AT COATINGS

The SEM images of the surface of the plasma sprayed AT coatings and EDAX

spectra of the same are shown in Figure 4.4(a-c). The SEM micrographs of AT

coating consisted of fully melted (FM), some partially melted (PM) and few unmelted

(UM) particles (Figure 4.4(a)). The porosity in the AT coating as calculated from the

SEM image is found to be 0.6%. The SEM micrographs of AT coatings reveal the

presence of the two regions, one with melted particles of micron size (FM, PM) and

some unmelted of nano size (UM). This kind of microstructure of the coatings with

two distinct regions is often referred to as a bimodal microstructure by several authors

(Eun Pil Song et al; 2006, Kabacoff et al; 2002, Youwang et al; 2000, Ramachandran

et al;1998, Venkataraman et al; 2006). The unmelted region consisted of smaller

particles mostly in the nanosize range (250nm-500 nm) (Figure 4.4(b)). The formation

of this kind of bimodal microstructure (melted and unmelted particles) is attributed to

the fact that the heat transfer within the agglomerated particles is lower compared to

the dense feed stock. With the exposure to the plasma flame, the outer shell of the

agglomerated nano particles is melted and rapidly solidifies upon falling on the

substrate to form micron sized zones, while the inner core will retain its nanosized

structure (Eun Pil Song et al; 2006).The resoldified regions will thus consist of fully

melted alumina with titanium ions and unsoldified regions will have unmelted

alumina (Jordon et al; 2001). The volume percentage of the unmelted particles present

in the AT coating was found to be 56% as obtained from the clemex vision software

attached with optical microscope.

The EDAX spectral analysis of the coatings at various regions is given in

Table 4.6. From the EDAX spectra, it is clearly evident that both the partially melted

(Region 1) and completed melted (Region 2) regions consist of more or less equal

concentrations of alumina and titania. (Figure 4.4(c)). In addition, from the high

resolution SEM image of the partially melted region (Figure 4.5), two kinds of

microstructures were observed i.e liquid phase sintered region and a solid phase

sintered region. Dongsheng Wang et al have made similar observations on coating

these ceramic powders on mild steel substrate. (Dongsheng Wang et al; 2009). As the

melting point of TiO2 is lower compared to that of Al2O3, the existence of the liquid

phase sintered region can be attributed to the selective melting of TiO2 nanoparticles

79

during plasma spraying, whereas, the solid phase sintered regions would possess

nanoparticles of Al2O3-TiO2 in unmelted state during plasma spraying.

4.3.1.3 MICROSTRUCURAL ANALYSIS OF YSZ COATINGS

In order to obtain a dense YSZ coating, the voltage was varied from 50 A to

60 A and the current value was so chosen such that it (Table 4.3) did not result in

spallation of the coating. The parameter that yielded dense coating without any

spallation was considered to be the optimized values. The SEM images of YSZ coated

surface revealed the presence two kinds of structures, one is fully melted splats and

the other structure appears to be poorly consolidated by fine particles. In addition,

pores (3.5%) and cracks (Figure 4.6) were also observed. A similar microstructure

was observed by Lin et al and the presence of poorly consolidated particles in the

coating is attributed to the limited growth of the particles due to the rapid melting of

nanostructured powders. Thus the lower momentum of these droplets with fine grains

results in poorly consolidated structure on deposition (Lin et al; 2003).

It is important to mention that it is easier to spray YSZ as the specific heat

capacity (450 J/Kg) and latent heat capacity (750 J/Kg K) of the YSZ are much lower

than that of the AT whose values are 750J/Kg K (Specific heat capacity) and 1500

J/kg K (latent heat) . However, the YSZ coatings consisted of larger number of poorly

consolidated particles, because the carrier gas flow rate was kept constant in both the

cases. The presence of voids in YSZ can be due to its high specific mass. As the

momentum of YSZ is high due to its higher specific mass (6000 kg/m3) than AT

(4000 kg/m3), it does not stay for longer period in the plasma resulting in improper

melting and hence leading to higher void content in the coating.

The SEM micrographs (Figure 4.7) of the BL coating exhibited completely

different microstructures when compared to YSZ and AT coatings. The surface

morphology of the BL coating showed the presence of large number of fully melted

splats with very less porosity (0.01%). In contrast to the AT coating which consisted

of both melted and unmelted particles, in the BL coating, a complete melting of

alumina particles was observed (Sathish et al; 2011). At this juncture it is not clear as

to why the alumina particles were completely melted in the BL coatings when

compared to pure AT coating. In addition, it is clearly evident that AT coating

80

exhibits good adhesion with the already deposited YSZ layer when compared to direct

coating of AT over the bare substrate as there were no clear interface observed

between the substrate and YSZ or YSZ and AT coatings in the SEM micrographs

(Figure 4.8). The variation in the microstructures of the BL and AT coatings is due to

the difference in the surface temperatures. The surface of the preheated YSZ should

have been high than the preheated substrate, while forming BL coating. The high

temperature of the preheated YSZ coating is attributed to its low thermal conductivity.

Thus the BL coating was completely melted on the YSZ surface and this has resulted

in superior adhesion of the same. Similar to what have been observed in the present

study, Sarikaya also have reported the formation of layered coating due to higher

surface temperature leading to higher microhardness and reduced porosity of the

plasma sprayed coatings (Sarikaya, 2005).

The average surface roughness of plasma sprayed coatings was in the range of

6 to 10 µm, while, the bilayer coating exhibited the lowest surface roughness of about

6 µm when compared to AT and YSZ coatings.

Figure 4.3 SEM image of the plasma sprayed AT coating

Unmelted

Fully melted

Partially melted

81

Figure 4.4 SEM images of surface of plasma sprayed AT coating (a) bimodal

nature (b) unmelted region showing nanosize particles (c) a typical EDAX

spectra of plasma sprayed AT coating.

Table 4.6 EDAX data recorded on the surface of plasma sprayed AT coating

Point on the

SEM image

Element wt% ( %Error)

Oxygen Aluminium Titanium

1 (PM) 39.50 (0.61) 51.63 (0.25) 8.87 (0.26)

2 (FM) 37.33 (0.40) 51.89 (0.26) 10.78 (0.29)

3 (PM) 36.85 (0.39) 51.85 (0.26) 11.29 (0.30)

4 (FM) 42.09 (0.46) 45.14 (0.25) 12.77 (0.36)

82

Figure 4.5 High magnification SEM image of partially melted region of AT

coating

Figure 4.6 SEM image of the surface of plasma sprayed YSZ coating

83

Figure 4.7 Surface morphology of plasma sprayed BL coating

Figure 4.8 Cross sectional SEM image of BL coating

Al2 O3 - 13 TIO2

YSZ

84

(a) (b)

Figure 4.9 Cross sectional morphology of plasma sprayed (a) AT coating (b) YSZ

coating

4.3.2 PHASE ANALYSIS BY XRD AND RAMAN SPECTROSCOPY

As plasma spraying of YSZ and AT powders results in various phase

transformations, both XRD and Raman spectroscopic analysis were carried out.

Raman spectroscopy was taken as complementary as it gives hints regarding the

presence of some non stoichiometric phases along with other peaks which are not

identified in the XRD. Figure 4.10 shows the XRD patterns of the nanostructured

YSZ powder and plasma sprayed YSZ coating. The XRD patterns of both the YSZ

powder and the coating were found to be similar. All the peaks observed were

corresponding to only tetragonal zirconia phase and peaks corresponding to cubic

zirconia were absent in XRD analysis. On the other hand, the Raman spectra of

plasma sprayed YSZ coating revealed the presence of cubic zirconia around 247 cm-1

in addition to the peaks at 149 cm-1

(b1g), 191cm-1

, 322 cm-1

(b1g), 466 cm-1

(eg) and

639 cm-1

(eg) corresponding to tetragonal zirconia (Ghosh et al; 2006, Purohit et al;

2006) (Figure 4.12(a)). The crystallite sizes of the YSZ powder as calculated from

Scherrer formula were 26 nm for tetragonal zirconia.

85

The XRD patterns of AT powders, plasma sprayed AT and BL coatings are

shown in Figure 4.11. The XRD pattern of AT powder showed diffraction peaks

corresponding to α-Al2O3, - Al2O3, brookite, anatase and rutile TiO2 (Figure 4.11(a)).

Additional peaks corresponding to CeO2, ZrO2 and baddeleyite TiO2 were also seen

and similar observation has been reported by Sanchez et al (Sanchez et al; 2008). The

average crystallite size calculated from Scherrer formula for AT powder were 52 nm

and 64 nm respectively for α-Al2O3 and rutile TiO2 phases.

Apart from the- Al2O3 and γ- Al2O3, the XRD analysis of the AT coating

revealed the presence of Al2TiO5 (Figure 4.11(b)). Peaks corresponding to Al2TiO5

phase in the AT coating could have resulted from the reaction between Al2O3 and

TiO2 in the plasma flame. The γ- Al2O3 phase observed in the XRD pattern of AT

powder was not noticed in plasma sprayed AT coating and instead a broad hump was

noticed around the same region. The broadened XRD peak in the coating indicates

nucleation of amorphous phases. The presence of γ- phase in the coated surface is due

to the complete melting of the starting powder and also due to the fact that γ-Al2O3

nucleates in preference to α-Al2O3 during rapid solidification of liquid droplets

(Xinhua Lin et al; 2003, Mcpherson et al; 1980, Dubourg eta al; 2007). It has been

reported by several authors that plasma spraying of AT powder results in the

formation of phases such as of γ-Al2O3, α- Al2O3 and very few peaks corresponding to

brookite and rutile TiO2 (Chang-sheng Zhai et al; 2005, Dongsheng Wang et al; 2009,

Sanchez et al; 2008, Ibrahim et al; 2010).As there were no peaks corresponding to

neither brookite nor rutile TiO2 in the XRD of the AT coating, Raman spectroscopy

was performed. Raman analysis did not show the presence of any bands

corresponding to brookite phase, however, it revealed the presence of rutile TiO2 at

263, 438 and 604 cm-1

(Figure 4.12(b)). The crystallite size calculated from Scherrer

formula was approximately 55 nm for α-Al2O3 and 31 nm for γ-Al2O3. This clearly

indicates that there is not much of a grain growth during plasma spraying of AT

powder.

Contradictory to the AT coatings, the XRD pattern (Figure 4.11(c)) of the BL

coating clearly revealed the presence of brookite TiO2 in addition to α-Al2O3, -

86

Al2O3, Al2TiO5 phases and the crystallite size was measured to be 11 nm, 13 nm, 47

nm and 2 nm respectively for α-Al2O3, -Al2O3, Al2TiO5 and brookite TiO2 . It is

believed that brookite being a metastable phase could have resulted from rapid

quenching of the splats.

Raman analysis was performed for the BL coating as the XRD generally do

not clearly distinguish the presence of tetragonal and cubic zirconia due to line

broadening, However, Raman spectra of BL coating (Figure 4.12(c)) showed only one

peak at 639 cm-1

corresponding to tetragonal zirconia and two peaks corresponding to

the mineral phase of corundum-ruby were at 871 cm-1

and 986 cm-1

(Tomaszek et al;

2004).

Figure 4.10 XRD patterns of (a) YSZ powder and (b) YSZ powder and (b) YSZ coating

87

0

5000

10000

15000

20000

0 200 400 600 800 1000

Raman shift (cm-1)

Inte

nsity

(a.

u.)

0

5000

10000

15000

20000

0 200 400 600 800 1000

Raman shift (cm-1)

Inte

nsity

(a.

u.)

Figure 4.11 XRD patterns of (a) AT powder, (b) AT coating and (c) bilayer

coating

t- Tetrogonal

c- Cubic

(a)

t t

c t t

- baddeleyite ZrO2

88

0

1000

2000

3000

4000

5000

6000

7000

8000

0 200 400 600 800 1000

Raman shift (cm-1)

Inte

nsity

(a.u

.)

0

1000

2000

3000

4000

5000

6000

7000

8000

0 200 400 600 800 1000

Raman shift (cm-1)

Inte

nsity

(a.u

.)

Figure 4.12 Raman spectra of plasma sprayed (a) YSZ coating (b) AT coating

(c) bilayered plasma sprayed coating

4.3.3 POTENTIODYNAMIC ANODIC POLARIZATION STUDIES

The corrosion behavior of the coated and uncoated substrates is discussed

using open circuit potential and polarization curves. Figure 4.13(a-e) shows the

potentiodynamic polarization plots obtained for the bare substrate and the plasma

sprayed coatings in Hank‟s solution. A significant shift in OCP towards the noble

direction and very less corrosion rate was observed for the plasma sprayed BL

coatings (Figure 4.13 (e)) when compared to the AT and YSZ coatings. In addition,

bilayered coating (BL-2) exhibited a stable passivation up to the maximum scanning

potential. The icorr (corrosion current density) was the lowest for the BL coatings (BL-

(b)

(c)

R

R R

R-Rutile

0

2000

4000

6000

8000

10000

12000

0 300 600 900

Raman shift (cm-1)

Inte

nsity

(a.u

.)

0

2000

4000

6000

8000

10000

12000

0 300 600 900

Raman shift (cm-1)

Inte

nsity

(a.u

.)

0

1000

2000

3000

4000

5000

6000

7000

8000

0 200 400 600 800 1000

Raman shift (cm-1)

Inte

nsity

(a.u

.)

0

1000

2000

3000

4000

5000

6000

7000

8000

0 200 400 600 800 1000

Raman shift (cm-1)

Inte

nsity

(a.u

.) t-tetrogonal zirconia

M- Mineral phase of

corundum Ruby

t

M M

89

1 = 0.025 µAcm-2

and BL-2 = 0.07 nAcm-2

) when compared to that of the AT (1.2

µAcm-2

), YSZ (4.3 µAcm-2

) coatings and the substrate (4.49 µAcm-2

). Similarly, the

corrosion rates were 1.42x10-6

mm yr-1

, 0.0005 mm yr-1

, 0.023 mm yr-1

, 0.086 mm yr-

1 and 0.089 mm yr

-1 for BL-2, BL-1, AT, YSZ and the substrate respectively. From

the above results it is obvious that the BL coating exhibits strong passivating nature

and highest corrosion resistance and amongst all, BL-2 exhibited the highest corrosion

resistance, when compared to the other coatings. The high corrosion resistance of the

BL coating can be explained based on their microstructural features. The micrograph

of the BL coatings consisted of large amount of melted particles and lower porosity

when compared to the AT and YSZ coatings which possessed two different structures

and high porosity. The lower corrosion resistance of AT and YSZ coatings may be

attributed to the presence of high porosity (0.6%) in AT and the presence of poorly

consolidated splats and cracks in YSZ. The equivalent circuits used to fit the Electro

Chemical Impedance Spectroscopy (EIS) parameters along with the schematic of the

microstructures are shown in Figure 4.14(a-c). We presume that a very thin dense

interface layer would have been formed in between these two layers which would

have enhanced the corrosion resistance (Figure 4.14(c)).

The Nyquist plot of bare substrate showed two time constants. The first time

constant is attributed to the resistance between the electrolyte and very thin oxide film

present on the titanium alloy substrate and the second time constant should be

attributed to the substrate alloy. The very low corrosion resistance offered by the

substrate is due to the poor quality of the oxide film. The Nyquist plot of YSZ plasma

sprayed coating shows two time constants and Warburg behaviour. The Warburg

behaviour may be ascribed to the resistance offered due to the corrosion products

blocking the pores/cracks present in the coating. AT plasma sprayed coating also

showed Nyquist plot similar to that of YSZ coating. The Nyquist plot of BL-2 coating

exhibited two time constants and no Warburg component was observed (Figure 4.13

inset). The absence of Warburg component might be due to the presence of lower

porosity in the BL coating.

90

Figure 4.13 Potentiodynamic polarization plots of (a) bare substrate, (b) AT

plasma sprayed coating, (c) YSZ plasma sprayed coating and (d) bilayered

plasma sprayed coating (BL-1) (e) bilayered plasma sprayed coating (BL-2)

(inset Nyquist plot of bilayered coating)

Figure 4.14 Schematic diagrams representing the typical defects in the

microstructure of plasma sprayed YSZ, AT and BL coatings responsible for the

corrosion behaviour along with their equivalent circuits.

4.3.4 SCRATCH TEST RESULTS

The resistance to scratch and abrasion of the coating was tested using

scratch tester. The scratch tracks of the YSZ, AT and BL coatings are shown in Figure

4.15 (a-c).Amongst all the coatings, the scratch groove of YSZ coating was much

deeper and having a larger width of about 25µm. In addition, minute cracks and some

amount of deformation were also observed on the edges of the scratched area (Figure

91

4.15(a)). In the case of AT coating, slight damages were noticed around the grooves

(Figure 4.15(b)) indicating that the cause of failure is due to ductile fracture. On the

other hand, the width of the scratch grooves was found to be very small in the BL

coating (10 µ m) indicating higher scratch resistance of this coating when compared

with others (Figure 4.15(c)).The least scratch resistance of YSZ coating is mainly due

to the presence of poorly consolidated splats, pores and lower hardness, while the

higher scratch resistance of AT coating may be attributed to the presence of nanosized

alumina particles embedded in the fully melted splats. The highest scratch resistance

of BL coating is due to the higher hardness and less porosity of the coating.

(a) (b)

(c)

Figure 4.15 Scratch tracks of nanostructured (a) YSZ coating (b) AT coating

(c) BL coating

Deep groove

92

4.3.5 WEAR TEST RESULTS

The wear behavior of the three coated substrates are explained by using wear

vs time plot (Figure 4.16), wear rate measurement and three dimensional wear plots

(Figure 4.17 (a-c)). The friction coefficients measured during wear testing and wear

rate measured using mass loss calculations are given in Table 4.7. Figure 4.15 clearly

reveals that for both AT and BL coatings, the wear was initially high and later on, the

wear rate became steady and remained constant till the completion of the experiment

which lasted for 105 cycles. However, for YSZ coating, the wear depth increased

linearly with time, which has led to more amount of material removal. The easier

removal of the coating is obviously due to the presence of poorly consolidated splats

and pores present in the coating (Sathish et al; 2011).

Table 4.7 Wear test results

Coating Wear rate (mm3/N m)

Coefficient of

friction

Un coated Ti-13Nb-13Zr 4.43x10-6

0.74

Nano AT 1.48x10-6

0.48

Nano YSZ 3.7x10-6

0.70

Nano BL 7.4X10-9

0.50

93

Time (hrs)

Wea

r (μ

m)

Figure 4.16 Wear Vs Time

The three dimensional plots depict a real time evolution of coefficient of

friction as a function of the number of cycles vs. displacement and enables one to

understand the wear mechanism operating with respect to the number of cycles. It can

be observed from the figure 4.17 (a-c) that the wear loops are parallelograms for most

of the cycles indicating that the gross slip regime is maintained for all the coatings

(Tian et al; 2008). However, for YSZ coating, there has been an abrupt increase in the

area of the wear loops after 40,000 cycles which indicates that the cracks and poorly

consolidated splats have led to the damage of the coating which in turn has further

enhanced the wear to the maximum extent. Thus, from the above observations, it is

evident that amongst all the coatings, the BL coating exhibited the highest wear

resistance and the YSZ coating offered the least resistance to wear. These features are

in agreement with what has been so far observed and discussed with regard to the

studies made on scratch resistance.

Nano YSZ coating

Nano AT coating

Nano BL coating

94

(a)

(b)

95

(c)

Figure 4.17 Three dimensional plots (a) BL coating (b) AT Coating (c) YSZ

Coating.

4.3.6 WORN SURFACE MORPHOLOGY

In order to identify and understand the wear mechanisms for all the three

coatings, worn surfaces were analyzed using SEM equipped with EDS and figure 4.18

(a-c) shows the worn surface morphologies of the AT, YSZ and BL coatings. The

worn surface of the AT coating is observed to be smooth and large amount of

particles were retained even after 105 cycles. The particles observed in the worn

surface of the AT coating were confirmed as alumina by EDS analysis (Figure 4.18

(a)). The high wear resistance of the nanocoatings is due to the presence of unmelted

(few nanometers in size) and melted alumina particles in the microstructure. It is

important to note that this bimodal nature present in the coating leads to high

resistance to the crack propagation (Figure 4.4). Several workers have quoted the

same reason for the increase in the wear resistance for this kind of bimodal structure

(Rico et al; 2009, Sofiane Guessasma et al; 2006, Normand et al; 2000).Contrary to

the above, for YSZ coating, almost the entire coating was worn out after 105cycles

96

and the worn surface was observed to be quite rough. The higher magnification of the

worn surface of the YSZ coated sample revealed the absence of the coating and also

the presence of grooves and plastic deformations on the substrate surface (Figure

4.18(b)). The EDS analysis across the worn surface confirmed the presence of the Ti,

Nb and Zr corresponding to the base material indicating that the coating has been

completely removed because of the poorly consolidated particles, cracks and loosely

bound splats present in the coating under the reciprocatory loading conditions.

In contrast to the above, the worn surface of the BL coating is smooth and

large amount of particles are retained after 105cycles of wear testing (Figure 4.18(c)).

Also the width (1.6mm) of the wear track was very much smaller than that of the

diameter (5.2 mm) of the alumina ball used for wear testing indicating that this

coating is offering higher wear resistance. During wear test, the particles from the

Al2O3 ball are found to be smeared on the coated surface. Further, even visual

observation of the ball showed the wearing of ball. Moreover at higher magnification,

some hard particles were observed across the worn surface and the EDS analysis of

this particle exhibited peaks corresponding to Al and oxygen, confirming the presence

of Al2O3 particles in the coated surface. Since, the bilayered coating has higher

hardness, the alumina ball was worn out considerably after wear test and this confirms

that the alumina particles present on the worn out surface were from the ball.

The high wear resistance of BL coating consisting of brookite phase can also

be explained based on the difference in the available slip systems between the BL and

AT coatings. Brookite phase (Orthorhombic) of TiO2 consists of five slip systems

whereas, the anatase and rutile (tetragonal) phases of titania posses only two slip

systems. The existence of higher number of slip system leads to greater ductility and

in turn high wear resistance and hence the wear resistance of BL coating is higher due

to the presence of brookite phase.

These studies clearly bring out the fact that amongst all the coatings, the wear

rate was minimal for the bilayered coating.

The major contributing factors for high wear resistance of BL coating are high

scratch resistance and high hardness. As stated earlier, the high hardness of BL

coating is due to the presence of fully melted particles consisting of few nano particles

97

and the formation of few stable α (corundum) phase regions and α phase is well

known for its high hardness. Further, it should be noted that even at the time of wear

testing, certain phase transformations occur which will also lead to further increase in

the wear resistance of coated substrates. In the present work, during the process of

wear testing of BL coating, the melted unstable γ Al2O3 on absorption of frictional

heat transforms to stable α Al2O3 and increases the hardness of coating thereby

improving the wear resistance and the above has been reported by several workers

(Chong-gui Li et al;2010,Berriche et al;2000). In our study also, the XRD clearly

revealed the presence of α (corundum) phases in the BL coating and thus we conclude

that the above factors have led to the substantial improvement in the wear resistance

It is important to mention that the wear behavior of AT (bimodal structure)

and BL (completely melted particles with few nano particles) attained using

agglomerated nanoparticles was superior than the wear behavior of the fully melted

AT coating obtained using spraying micron sized powders (Jordan et al; 2001). The

low wear resistance of conventional AT coating is attributed to the poor adhesion of

the coating to the substrate. In the present work, the increase in the wear resistance of

the AT coating is due to the bimodal structure as discussed earlier and the reason for

the highest wear resistance of the BL coating is because of the complete melting of

AT which has led to higher adhesion with YSZ layer which was initially coated on

the bare substrate.

98

(a)

99

(b)

Worn surface

100

(c)

Figure 4.18 Worn surface morphologies of nanostructured (a) AT coating

(b) YSZ coating and (c) BL coating

101

4.4 CONCLUSIONS

This work perhaps is the first report of the microstructure, hardness, scratch

and wear behavior of the bilayered coating that was performed on the titanium alloy

in connection with biomedical implants.

Phase analysis of the coated and uncoated substrates carried out using XRD

and Raman spectroscopy clearly explained the phase transformations occurring during

plasma spraying. The formation of rutile TiO2 in AT coating and tetragonal ZrO2

in BL coating was observed using Raman analysis.

The phase analysis of the bilayered coating revealed the presence of more

amount of stable (corundum) phase resulting in substantial improvement in the

hardness, corrosion, scratch and the wear resistance.

The microstructural analysis of Al2O3-13TiO2 coating showed the formation of

bimodal microstructure, whereas, the microstructure of the bilayered coating revealed

the presence of large amount of melted particles with less porosity compared to a

mono layer AT coating.

The vast improvement in the corrosion resistance of the BL coating is mainly

due to the complete melting of particles and also due to better adhesion and less

porosity. In addition, the EIS measurements of all the three coatings were described

by the equivalent circuit and the results obtained from these studies further supported

the results of the polarization studies. From the equivalent circuits developed from the

Nyquist plot and the fitted Rp and Qdl values, it was clearly evident that BL coatings

possessed higher passivation and less corrosion. In addition, from the absence of

Warburg impedance, it was evident that there was minimal porosity in the BL

coatings.

It is observed that the corrosion resistance of the bilayered coating is higher

when compared to that of both the Al2O3-13TiO2 and the ZrO2 coatings. The decrease

in the corrosion resistance of Al2O3-13TiO2 coating when compared to the bilayered

coating is mainly due to the presence of high porosity and loosely bound unmelted

alumina particles present in the coating. On the other hand, the presence of poorly

102

consolidated splats and cracks in the ZrO2 coatings have led to the decrease in its

corrosion resistance. Thus, it is obvious that the microstructure of the coatings play a

significant role in the corrosion characteristics of the materials.

A similar trend was also observed in the wear performance of these coated

substrates. The bilayered coating exhibited two hundred and five hundred fold

increase in the wear resistances when compared with that of the nanostructured

Al2O3-13TiO2 and ZrO2 coatings. Further, these results corroborate well with the

results obtained using the scratch testing.

There has been a substantial improvement in the scratch resistance of

bilayered coating when compared with that of the monolayer coatings which was

obviously seen by the decrease in the scratch width of the BL coating.

Based on this study, we recommend bilayered coating for titanium alloys to be

used for biomedical applications after subjecting it to in vivo testings. Also our

detailed studies make us to suggest that the newly developed bilayered coating

obtained using plasma spray technique can be explored for other industrial

applications which require higher wear and oxidation resistance.