Copyright © 2010 American Scientific Publishers
All rights reserved
Printed in the United States of America
Journal of
Nanoscience and Nanotechnology
Vol. 10, 4767–4772, 2010
Corrosion Behavior of Aluminum Doped Diamond-Like
Carbon Thin Films in NaCl Aqueous Solution
N. W. Khun and E. Liu∗
School of Mechanical and Aerospace Engineering, Nanyang Technological University,
50 Nanyang Avenue, Singapore 639798, Singapore
Aluminum doped diamond-like carbon (DLC:Al) thin films were deposited on n-Si(100) substrates
by co-sputtering a graphite target under a fixed DC power (650 W) and an aluminum target under
varying DC power (10–90 W) at room temperature. The structure, adhesion strength and surface
morphology of the DLC:Al films were characterized by X-ray photoelectron spectroscopy (XPS),
micro-scratch testing and atomic force
microscopy
(AFM), respectively.
The corrosion performance
Delivered
by Ingenta
to:
of the DLC:Al films was investigated
by
means
of
potentiodynamic
polarization
testing in a 0.6 M
Nanyang Technological University
NaCl aqueous solution. The results showed that the polarization resistance of the DLC:Al films
IP : 155.69.4.4
increased from about 18 to 30.7 k though the corrosion potentials of the films shifted to more
Mon,
01
Aug 2011 07:07:40
negative values with increased Al content in the films.
Keywords: DLC:Al Thin Film, Magnetron Sputtering, NaCl Solution, Corrosion, Potentiodynamic
Polarization.
Diamond-like carbon (DLC) films can have high hardness,
dielectric strength, biocompatibility and optical transparency as well as low wear and friction. These properties are strongly dependant on the fractions of sp3 and
sp2 hybridized carbon bonding in the films. DLC films
can also possess outstanding chemical properties in various environments. The corrosion resistance of DLC-coated
materials is closely related to the structure, surface morphology, thickness and defect density of the DLC films as
well as the interface condition between film and substrate.1
DLC films can be synthesized by different techniques
such as magnetron sputtering, laser ablation, cathodic
arc and various chemical vapor deposition (CVD)
processes.2–4 The impinging energy of sputtered carbon
species during film deposition is an important parameter
to get a dense packing of carbon atoms in DLC films.
However, a high impinging energy tends to induce a high
compressive stress due to the distortions of bond length
and bond angle of sp3 carbon bonds. Thus, a major constraint of DLC films is a poor adhesion of the films to
their substrates caused by high residual stresses. Some
efforts have been made to improve the adhesion of DLC
films by doping metals or non metals into the films.5–9
Zhang et al.10 reported that doping Al into DLC films
∗
Author to whom correspondence should be addressed.
J. Nanosci. Nanotechnol. 2010, Vol. 10, No. 7
drastically decreased the residual stress of the films. Liu
et al.11 reported that the adhesion strength of DLC films
had a significant influence on the effectiveness of corrosion protection. A sufficient adhesive strength of DLC
films can lessen undermining effects in corrosive media
and promote their corrosion resistance. However, due to
different electrochemical potentials between metals and
DLC films, introduction of metals into DLC films could
degrade the corrosion properties of the films in chemical
environments. Detailed mechanisms of Al incorporation
during magnetron sputtering deposition still needs to be
fully understood because the structures of DLC:Al films
are rather different from those of undoped DLC films.
In this paper, the effect of Al doping on the bonding
structure, surface morphology, adhesive strength and corrosion resistance of DLC:Al films deposited on n-Si substrates with DC magnetron co-sputtering was investigated
in terms of DC sputtering power applied to Al target.
2. EXPERIMENTAL DETAILS
Aluminum doped diamond like carbon (DLC:Al) thin
films were grown on highly conductive n-Si (100)
(1–5 × 10−3 cm) substrates by DC magnetron cosputtering deposition using a pure graphite target
(99.995% C) and a pure aluminum target (99.995% Al)
of 4 inch in diameter. The substrate rotation speed and
bias applied during all the film depositions were 33 rpm
1533-4880/2010/10/4767/006
doi:10.1166/jnn.2010.1682
4767
RESEARCH ARTICLE
1. INTRODUCTION
Khun and Liu
AI/C and O/C
circle of 1 cm in diameter. A saturated calomel reference
and −50 V, respectively. The Si substrates were cleaned
electrode (SCE) (244 mV vs. SHE at 25 C) and a platby ethanol in an ultrasonic container for 20 min followed
inum counter electrode were used during the polarization
by distilled water, and then dried with dry air. Prior to
tests.
the film deposition, the substrates were pre-sputtered with
Ar+ plasma at a substrate bias of −250 V for 20 min
to remove the oxide and contamination layers on the sur3. RESULTS AND DISCUSSION
faces. The DLC:Al film depositions were conducted at a
working pressure of 3.5 mTorr for 120 min. An Al interFigure 1(a) shows the Al/C and O/C atomic ratios in terms
layer between the DLC:Al films and the Si substrates was
of DC sputtering power applied to the Al target, where the
deposited by sputtering the same Al target with a fixed
O/C ratio increases from 0.06 to 0.70 and the Al/C ratio
DC power of 100 W for 1 min to promote the adhesion
linearly increases from 0 to 0.54 when the DC power is
of the films to the substrates and make an ohmic contact
increased from 10 to 90 W. An increase in surface oxybetween the films and substrates. Though all the deposigen percentage with increased Al content in the DLC:Al
tions were conducted with a fixed DC power of 650 W on
films results from a greater difference in electronegativities
the graphite target at room temperature, the maximum tembetween Al (∼1.61, pauling scale) and O (∼3.44) comperature reached during the film depositions was approxipared to the difference between Al and C (∼2.55). The
mately 62 C. The DC power applied to the Al target was
consistently higher O/C ratios than the Al/C ones indicate
varied from 10 to 90 W.
that Al on the surfaces of the DLC:Al films may be fully
10
The film chemical composition and bonding
configura-by Ingenta
Delivered
oxidized.to:
tion were investigated by means of X-ray
photoelectron
Nanyang Technological
TheUniversity
depth profile of the DLC:Al film deposited with
spectroscopy (XPS) (Kratos Axis Ultra) using passIPener: 155.69.4.4
20 W applied to the Al target shows that about 4.3 at.%
gies of 40 eV for C 1s, Al 2p and O 1s core
level01
spectra
Mon,
Aug 2011
07:07:40
oxygen
is incorporated into the film though a higher oxyand 160 eV for wide scans with a monochromatic Al K
gen content of about 12.11 at.% is observed at the initial
X-ray radiation (h = 1486.71 eV). XPS depth profiling
of the DLC:Al films was conducted using an etching rate
(a) 0.9
of 6.5 nm/min calibrated with a standard SiO2 film.
0.8
The bonding structure of the DLC:Al films was also
Al/C
0.7
investigated with confocal micro-Raman spectroscopy
O/C
0.6
(Renishaw RM 1000) using a He–Ne 632 nm laser.
The surface morphology of the DLC:Al films was
0.5
measured using scanning electron microscopy (SEM,
0.4
JEOL-JSM-5600LV) and atomic force microscopy (AFM,
0.3
Digital Instruments S-3000) and the surface roughness
0.2
(Ra ) was measured using AFM with five measurements on
0.1
each sample and an average roughness value being taken.
0
The adhesion strength of the DLC:Al films was mea40
90
10
20
55
70
sured with a micro-scratch tester (Shimadzu SST-101) havDC power on Al target (W)
ing a diamond stylus of 15 m in radius dragged down
(b) 110
the films under a progressive loading condition at room
temperature. The scan amplitude, frequency, scratch rate
90
and down speed for all the tests were set as 50 m, 30 Hz,
2 m/s and 2 m/s, respectively. Five measurements on
70
each sample were performed and an average value of the
critical load was taken.
50
O 1s
Potentiodynamic polarization measurements were carC 1s
ried out using a potentiostat/galvanostat station (EG&G
30
Al 2p
263A) having a three-electrode flat cell kit at a scan rate
Si 2p
of 0.8 mV/s at room temperature. The electrolyte used
10
for all the measurements was a deaerated and unstirred
0.6 M NaCl solution. For all the electrochemical mea–10
0
2000
4000
6000
8000
10000
surements, the DLC:Al film coated samples were cut into
Etching time (s)
2 cm × 2 cm square pieces and a gold layer was deposited
on the backsides of the Si substrates to make the testing
Fig. 1. (a) Al/C and O/C atomic ratios of DLC:Al films as a function
samples in good electrical connection during the polarizaof DC sputtering power applied to Al target and (b) depth profile of a
tion measurements. The testing area on the films was a
DLC:Al sample deposited with 20 W on Al target.
Elemental concentration (%)
RESEARCH ARTICLE
Corrosion Behavior of Aluminum Doped Diamond-Like Carbon Thin Films in NaCl Aqueous Solution
4768
J. Nanosci. Nanotechnol. 10, 4767–4772, 2010
Khun and Liu
Corrosion Behavior of Aluminum Doped Diamond-Like Carbon Thin Films in NaCl Aqueous Solution
1889
1768
1647
1526
1405
1284
1163
1042
921
70
55
40
(a)
20
10
0
80
79
78
77
76
75
74
73
72
71
70
Binding energy (eV)
(b) 160000
C 1s
Power (W):
(b)
Intensity (a.u.)
90
70
55
40
20
10
0
289
287
285
283
281
Binding energy (eV)
Fig. 2. XPS spectra of DLC:Al films: (a) Al 2p and (b) C 1s with
respect to DC power on Al target.
J. Nanosci. Nanotechnol. 10, 4767–4772, 2010
Fig. 4. Results determined from fitted Raman spectra shown in
Figure 3: (a) positions of G and D peaks and (b) FWHMs of D peaks
and ID /IG ratios with respect to DC sputtering power on Al target.
4769
RESEARCH ARTICLE
Intensity (a.u.)
800
Intensity (a.u.)
320000
stage till about 60 s due to exposure to air after venting
Power (W):
the deposition chamber as shown in Figure 1(b).
90
Figure 2(a) shows the XPS Al 2p spectra of the DLC:Al
70
films as a function of DC power on the Al target. The Al
2p peaks located at approximately 74.8 eV are attributed
55
to Al–O bonds.12 The XPS C 1s spectra of the DLC:Al
40
films versus the DC power on the Al target are shown
20
in Figure 2(b), where the C 1s peaks at approximately
10
284.8 eV shift to a slightly lower binding energy of about
284.6 eV with increased Al content, indicating that there
are more sp2 bonds in the films. For all the samples used
10000
in this study, there is no evidence of the formation of Al–C
bonds since no peaks are found at around 281.5 eV in all
the C 1s spectra. Therefore, it can be deduced that the
Raman shift (cm–1)
doped Al may exist as O bonded or pure elemental form
in the carbon matrix.10 From the fitted Al 2p peaks using
Fig. 3. Raman spectra of DLC:Al films.
Gaussian functions, it is found that the calculated area ratio
of the Al–O band over the Al–Al band is about 10.17 for
of the Raman spectra with increased Al content in the
the DLC:Al film deposited with 20 W on the Al
target. Anby Ingenta
Delivered
to:
films indicates a decrease in the level of disorder.13 The
Al–O/Al–Al ratio of about 16.92 is found
when
the
power
Nanyang Technological
University
Raman spectra of the DLC:Al films were deconvoluted
is increased to 90 W.
IP : 155.69.4.4
using a Gaussian function for G peaks and a Lorentzian
The Raman spectra of the DLC:Al films Mon,
deposited
with
01 Aug 2011
07:07:40
function for D peaks to investigate the effect of Al dopvarying DC power applied to the Al target from 10 to 90 W
ing on the bonding structure of the films. The increased
are shown in Figure 3, where the improved symmetry
Al in the films causes downshifts of both the G band
from 1538 to 1444 cm−1 and the D band from 1359 to
(a) 30000
1303 cm−1 as shown in Figure 4(a).14 The full-widths-atAl 2p
half-maximum (FWHMs) of the D bands decrease from
Power (W):
486 to 363 cm−1 with increased sputtering power on the
90
Al target (Fig. 4(b)), indicating an increase in ring order.15
Corrosion Behavior of Aluminum Doped Diamond-Like Carbon Thin Films in NaCl Aqueous Solution
Khun and Liu
360
(a)
340
Critical load (mN)
RESEARCH ARTICLE
Ra (nm)
An intensity ratio between D and G peaks, ID /IG , gives the
between the Al species and the C matrix that may degrade
information about carbon structure such as graphitic clusthe cohesive strength of the film.
tering and structure disorder.16 17 It is found that the ID /IG
The surface roughness (Ra ) of the DLC:Al films
ratios of the DLC:Al films increase from about 1.6 to 3.12
decreases from about 0.79 to 0.16 nm with increased sputwith increased DC power as shown in Figure 4(b).17 The
tering power from 10 to 90 W applied to the Al target
introduction of Al into the DLC films promotes the clusas shown in Figure 6(a), which cannot be correlated to
tering of the sp2 bonds, which in turn increases the ID /IG
the increased sp2 fraction with increased Al content in
ratio. The local increase in sp2 bonds in the amorphous
the films. A possible mechanism proposed by Corbella23
carbon matrix is caused by metal-induced graphitization,
and Peng27 is that a higher ion bombardment energy
which the metal species within the C matrix can act as a
catalyst to promote the formation of sp2 sites.18–23 In addition, the incorporation of Al into the DLC films can also
(a) 0.9
convert stress-induced sp3 bonds to sp2 bonds during the
0.8
film deposition.
0.7
From the micro-scratch tests, it is observed that the critical load of the DLC:Al films increases from about 248
0.6
to 343 mN when the sputtering power applied to the Al
0.5
target is increased from 10 to 90 W as shown in Figure 5.
0.4
It is well known that the adhesion strength of DLC films
Delivered by Ingenta to:
is apparently affected by the residual stress in the films. A
Nanyang Technological0.3University
higher sp2 /sp3 ratio can reduce the residual stress in a DLC
0.2
IP : 155.69.4.4
film as the sp2 bonds are shorter than the sp3 bonds.24–26
Mon, 01 Aug 2011 07:07:40
0.1
Therefore, an increased sp2 -bonded fraction with increased
0
Al content in the DLC films promotes the adhesion of the
0
20
40
60
80
100
films to the substrates.
DC power on Al target (W)
The insets of Figure 5 show the SEM micrographs of
(b)
the surface morphologies of the scratched DLC:Al films
deposited with 10 and 90 W applied to the Al target.
The observed brittle fracture of the DLC:Al film deposited
with 10 W may be due to stress-induced interfacial cracks.
One way of improving interfacial bond strength between
film and substrate would be to increase metal doping
level in the film in order to reduce the residual stress
in the film. The increased critical load of the DLC:Al
films with increased Al content in the films indicate an
improved interfacial bond strength between the films and
the substrates. However, a heavier Al incorporation into
the DLC:Al film deposited with 90 W applied to the
Al target causes a flaky feature of the scratched film at
the critical load, which is probably due to the interfaces
(c)
320
300
280
260
(b)
240
0
10
20
30
40
50
60
70
80
90
100
DC power on Al target (W)
Fig. 5. Critical loads of DLC:Al films as a function of DC power on
Al target. The insets show SEM micrographs of scratched DLC:Al films
deposited with (a) 10 and (b) 90 W.
4770
Fig. 6. (a) Surface roughness of DLC:Al films as a function of DC
power on Al target, and (b) and (c) AFM images of DLC:Al films
deposited with 10 and 90 W on Al target, respectively.
J. Nanosci. Nanotechnol. 10, 4767–4772, 2010
Khun and Liu
Corrosion Behavior of Aluminum Doped Diamond-Like Carbon Thin Films in NaCl Aqueous Solution
could smoothen a film surface by preferentially removing
materials from protruding regions on the film surface.
Smoothening of the DLC:Al films by this proposed mechanism is pronounced by increased bombardment energy
of heavier Al ions induced by the higher sputtering powers applied to the Al target during the film depositions.
Therefore, larger asperities are found on the surface of the
DLC:Al film deposited with 10 W (Fig. 6(b)) as well as
finer asperities are found on the surface of the DLC:Al
film deposited with 90 W (Fig. 6(c)).
From the potentiodynamic polarization curves of the
DLC:Al films obtained in the 0.6 M NaCl solution as
shown in Figure 7(a), the corrosion parameters such as
corrosion potential (Ecorr ) and current (Icorr ) are analyzed
using the Tafel technique. The polarization resistance (Rp )
values of the DLC:Al films are calculated from the anodic
(a ) and cathodic (c ) Tafel slopes and Icorr according to
the following formula:28
Table I. Results determined from potentiodynamic polarization curves
of DLC:Al films as shown in Figure 7(a).
DC power on
Al target (W)
10
20
40
55
70
90
Ecorr (mV)
Icorr (A)
Rp (k)
−232.7
−240.0
−242.3
−266.9
−284.4
−298.8
4.88
4.95
5.42
5.99
6.22
5.13
17.98
18.12
19.71
17.41
25.29
30.71
corroded area of a DLC:Al sample (90 W on Al target) after polarization
test.
J. Nanosci. Nanotechnol. 10, 4767–4772, 2010
2Al + 3H2 O → Al2 O3 + 6H+ + 6e−
4771
RESEARCH ARTICLE
The corrosion test results obtained are summarized in
Table I. The Ecorr of the DLC:Al films shifts from −232.7
to −298.8 mV versus SCE when the sputtering power on
the Al target is increased from 10 to 90 W. Due to the different electrochemical potentials between the Al species on
the film surface and the background C matrix, they behave
as tiny anodes and cathodes on the film surface, inducing anodic
Delivered by Ingenta
to:and cathodic current flows between them and
Rp = a × c /23 Icorr aNanyang
+ c (1)
resulting
in the dissolution of the films. It is found from
Technological
University
Table I that the Icorr of the DLC:Al films first increases
IP
:
155.69.4.4
with Rp in k, a and c in V /I-decade, and Icorr in A.
from07:07:40
4.88 to 6.22 A with increased power from 10 to
Mon, 01 Aug 2011
70 W and then turns to decrease to 5.13 A with further
increased power up to 90 W. The increased sp2 fraction in
(a)
the DLC:Al films with increased sputtering power applied
to the Al target, which is confirmed by the increased ID /IG
ratios (Fig. 4(b)), is one of the reasons for the increased
Icorr . In addition, the increase of the Icorr can also be
attributed to the doping of Al into the C matrix since the
Al species also degrade the sp3 -bonded cross-linking structure. Moreover, the resulted interfacial bonds between the
Al species and C matrix can be easily attacked by electrochemically active species, e.g., water molecules and Cl−
ions, in the electrolyte. Therefore, these effects become
pronounced when the Al content in the films is increased
by increasing the sputtering power on the Al target. However, the decreased Icorr of the DLC:Al film deposited with
90 W can be probably related to the development of an
(b)
oxide layer on the film surface with increased Al content in the film. A possible reason is that the oxide layer
would slow down the dissolution of the film by preventing
a direct access of the electrolyte to the active Al sites on
the film surface as well as retarding electrochemical dissolution reactions via hindering electron transfer process
through it.
From the XPS results shown in Figure 1(a), the increase
in the surface oxygen with increased sputtering power
on the Al target can be correlated to the growth of the
oxide layer. Besides, during the polarization measurements, refreshed Al sites produced by dissolving the oxide
layer on the film surface, as the applied potential is shifted
to higher positive values, react with water molecules to
Fig. 7. (a) Potentiodynamic polarization curves of DLC:Al films as a
form a new oxide layer through the following reaction:
function of DC power on Al target and (b) SEM micrograph showing the
Corrosion Behavior of Aluminum Doped Diamond-Like Carbon Thin Films in NaCl Aqueous Solution
RESEARCH ARTICLE
This process would replenish the dissolved oxide layer
through the anodic dissolution in the NaCl solution and
consequently make the dissolution of the film difficult during the measurement. The replenishing process becomes
more significant with increased Al content, resulting in
more difficult polarization processes. Therefore, the apparently increased Rp of the DLC:Al films from 17.98 to
30.71 k with increased sputtering power on the Al target
from 10 to 90 W confirms that the increased Al content in
the films makes the polarization processes more difficult
via the growth and replenishment of the oxide layers on
the film surfaces.
After the polarization tests, the surface morphologies of
the corroded DLC:Al films are characterized with SEM.
Figure 7(b) shows the corroded area of the DLC:Al film
deposited with 90 W applied to the Al target, in which
the film surface is covered with corrosion products probably produced from the reaction of the film with the
electrolyte.
Delivered by
Khun and Liu
References and Notes
1. Z. H. Liu, P. Lemoine, J. F. Zhao, D. M. Zhou, S. Mailley, E. T.
McAdams, P. Maguire, and J. McLaughlin, Diam. Relat. Mater.
7, 1059 (1998).
2. S. Aisenberg and R. Chabot, J. Appl. Phys. 42, 2953 (1971).
3. D. S. Whitmell and R. Williamson, Thin Solid Films 35, 255
(1976).
4. Y. Lifshitz, Diam. Relat. Mater. 5, 388 (1996).
5. T. I. T. Okpalugo, P. D. Maguire, A. A. Ogwu, and J. A. D.
McLaughlin, Diam. Relat. Mater. 13, 1549 (2004).
6. C. Corbella, E. Pascual, G. Oncins, C. Canal, J. L. Andujar, and
E. Bertran, Thin Solid Films 482, 293 (2005).
7. D. Y. Wang, Y. Y. Chang, C. L. Chang, and Y. W. Huang, Surf. Coat.
Tech. 200, 2175 (2005).
8. G. K. Burkat, T. Fujimura, V. Y. Dolmatov, E. A. Orlova, and M. V.
Veretennikova, Diam. Relat. Mater. 14, 1761 (2005).
9. I. G. Gonzalez, J. D. Jesus, D. A. Tryk, G. Morell, and C. R. Cabrera,
Diam. Relat. Mater. 15, 221 (2006).
10. S. Zhang, Y. Q. Fu, X. L. Bui, and H. J. Du, Inter. J. Nanosci. 3, 797
(2004).
11. C. L. Liu, D. P. Hu, J. Xu, D. Z. Yang, and M. Qi, Thin Solid Films
Ingenta
to:(2006).
496, 457
Nanyang Technological
University
12. P. Zhang,
B. K. Tay, G. Q. Yu, S. P. Lau, and Y. Q. Fu, Diam. Relat.
Mater. 13, 459 (2004).
IP : 155.69.4.4
4. CONCLUSIONS
13. C.
C. Chen and F. C. N. Hong, Appl. Surf. Sci. 242, 261 (2005).
Mon, 01 Aug 2011
07:07:40
14. S. Zhang, H. Du, S. E. Ong, K. N. Aung, H. C. Too, and X. Miao,
The effect of Al doping on the bonding structure, surface
Thin Solid Films 515, 66 (2006).
roughness, adhesion strength and corrosion resistance of
15. A. C. Ferrari and J. Robertson, Phys. Rev. B 61, 14095 (2000).
DLC:Al films was investigated in terms of DC sputtering
16. M. A. S. Oliveira, A. K. Vieira, and M. Massi, Diam. Relat. Mater.
power applied to Al target during film depositions. It was
12, 2136 (2003).
17. B. K. Tay and P. Zhang, Thin Solid Films 420–421, 177 (2002).
found that increased Al content in the DLC:Al films with
18. C. S. Lee, T. Y. Kim, K. R. Lee, and K. H. Yoon, Thin Solid Films
increased DC power on the Al target caused an increase in
447–448, 169 (2004).
adsorbed surface oxygen. Raman results revealed that the
19. H. Hofsass, H. Feldermann, R. Merk, M. Sebastian, and C. Ronning,
incorporation of Al in the DLC films promoted the forAppl. Phys. A 153–181, 66 (1998).
mation of sp2 bonds. Higher critical loads with increased
20. A. Oya and S. Otani, Carbon 17, 131 (1979).
sputtering power on the Al target were attributed to the
21. G. J. Qi, S. Zhang, T. T. Tang, J. F. Li, X. W. Sun, and X. T. Zeng,
Surf. Coat. Tech. 198, 300 (2005).
increased metal-induced sp2 bonds as well as the elemental
22. K. Bewilogua, R. Wittorf, H. Thomsen, and M. Weber, Thin Solid
Al dispersed in the films. It was found that the polarizaFilms 447–448, 142 (2004).
tion resistance of the DLC:Al films increased though their
23. C. Corbella, E. Bertran, M. C. Polo, E. Pascual, and J. L. Andujar,
corrosion potentials shifted to more negative values with
Diam. Relat. Mater. 16, 1828 (2007).
increased Al incorporation in the films.
24. S. Zhang, X. L. Bui, and Y. Fu, Surf. Coat. Tech. 167, 137 (2003).
25. J. P. Sullivan, T. A. Friedmann, and A. G. Baca, J. Electro. Mater.
26, 1021 (1997).
Acknowledgments: This work was supported by the
26. N. W. Khun, E. Liu, and H. W. Guo, Electroanalysis 20, 1851
research project (EWI-0601-IRIS-035-00) from the Envi(2008).
ronment and Water Industry Development Council (EWI),
27. X. L. Peng, Z. H. Barber, and T. W. Clyne, Surf. Coat. Tech. 138, 23
Singapore. N. W. Khun is grateful for the Ph.D. scholar(2001).
ship from the Nanyang Technological University (NTU),
28. B. Bhushan, Tribology and Mechanics of Magnetic Storage Devices,
2nd edn., Springer, New York (1996).
Singapore.
Received: 20 July 2008. Accepted: 20 January 2009.
4772
J. Nanosci. Nanotechnol. 10, 4767–4772, 2010