Applied Surface Science 258 (2012) 1929–1934

Contents lists available at ScienceDirect

Applied Surface Science

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

Effects of TiN coating on the corrosion of nanostructured Ti–30Ta–xZr alloys for

dental implants
Won-Gi Kim a , Han-Cheol Choe b,c,∗
a
Department of Dental Technology, Daegu Health College, Daegu, Republic of Korea
b
Department of Dental Materials, Research Center of Nano-Interface Activation for Biomaterials, School of Dentistry, Chosun University, Gwangju, Republic of Korea
c
Research Center for Oral Disease Regulation of the Aged, School of Dentistry, Chosun University, Gwangju, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Electrochemical characteristics of a titanium nitride (TiN)-coated/nanotube-formed Ti–Ta–Zr alloy

Available online 27 July 2011 for biomaterials have been researched by using the magnetic sputter and electrochemical methods.
Ti–30Ta–xZr (x = 3, 7 and 15 wt%) alloys were prepared by arc melting and heat treated for 24 h at 1000 ◦ C
Keywords: in an argon atmosphere and then water quenching. The formation of oxide nanotubes was achieved by
Nanotube anodizing a Ti–30Ta–xZr alloy in H3 PO4 electrolytes containing small amounts of fluoride ions at room
TiN coating
temperature. Anodization was carried out using a scanning potentiostat, and all experiments were con-
Corrosion behaviour
ducted at room temperature. The microstructure and morphology of nanotube arrays were characterized
Biomaterials
by optical microscopy (OM), field emission scanning electron microscopy (FE-SEM) and X-ray diffraction
(XRD). The TiN coatings were obtained by the radio-frequency (RF) magnetron sputtering technique.
The depositions were performed from pure Ti targets on Ti–30Ta–xZr alloys substrates. The corrosion
properties of the specimens were examined using potentiodynamic test in a 0.9% NaCl solution by using
potentiostat. The microstructures of Ti–30Ta–xZr alloys were changed from an equiaxed to a needle-like
structure with increasing Zr content. The interspace between the nanotubes was approximately 20, 80
and 200 nm for Zr contents of 3, 7 and 15 wt%, respectively. The corrosion resistance of the TiN-coated
on the anodized Ti–30Ta–xZr alloys was higher than that of the untreated Ti alloys, indicating a better
protective effect.
© 2011 Elsevier B.V. All rights reserved.

1. Introduction
Titanium (Ti) and its alloys are widely used as a dental implant
material in clinical dentistry and as an orthopedic implant mate-
rial due to their good mechanical property, corrosion resistance,
and biocompatibility. However, there are certain disadvantages
such as poor osteointegration properties, high elastic modulus, and
low corrosive-wear resistance [1–4]. In recent years, the forma-
tion of self-organized pores has been achieved on various valve
metals such as Zr, Hf, Nb, and Ta including including Ti [5–10]. Self-
organized porous TiO2 has been produced in fluoride-containing
electrolytes such as HF electrolytes, chromic acid–HF mixtures,
(NH4)2 SO4 –NH4 F mixtures, H2 SO4 –HF mixtures and H3 PO4 –NaF
mixtures [11–15]. Typically in these experiments, pores of a diam-
eter of approx. 100 nm and a pore length of approx. 500 nm was
reported. Nanotube titanium oxide formation on the Ti or Ti alloy
surface is important to improve cell adhesion and proliferation in
clinical use. It should be possible to control the nanotube size and
morphology for biomedical implant use by controlling the applied
voltage, alloying element, current density, anodization time, and
electrolytes.
The titanium nitride (TiN) coatings are used widely in many
dental and industry fields due to their high hardness, good wear
resistance, good adhesion, excellent corrosion resistance, and low
friction coefficient [16–18]. However, there still remains a need for
improvement in the bone adhesion or in the mechanical properties
of the Ti alloy to fabricate dental implants [19]. Many techniques
such as physical vapor deposition (PVD), chemical vapor deposition
(CVD), ion-beam deposition, and radio-frequency (RF) magnetron
sputtering have been used to produce TiN films [20,21].
In this study, in order to improve the corrosion resistance of
bio-implant for biocompatibility, we investigated the electrochem-
ical behaviors of a TiN-coated Ti–30Ta–xZr alloy before and after
nanotube formation in 0.9% NaCl solution using electrochemical
methods.

∗ Corresponding author. Present address: Department of Dental Materials &

2. Experimental details
Research Center of Nano-Interface Activation for Biomaterials, School of Dentistry,
Chosun University, Gwangju, Korea. Tel.: +82 62 230 6896; fax: +82 62 230 6896. Ternary alloys were prepared from Ti (Grade 4, G&S Tita-
E-mail address: hcchoe@chosun.ac.kr (H.-C. Choe). nium, USA), Ta (99.95% purity, Kurt J. Lesker Company, USA),

0169-4332/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.apsusc.2011.06.151
1930 W.-G. Kim, H.-C. Choe / Applied Surface Science 258 (2012) 1929–1934

Fig. 1. OM and FE-SEM images of the Ti–30Ta–xZr alloys (200×): (a) Ti–30Ta–3Zr; (b) Ti–30Ta–7Zr; (c) Ti–30Ta–15Zr.

and Zr (99.95% purity, Kurt J. Lesker Company, USA), having Zr pure) and N2 (99.9995% pure) gases was controlled by a mass
contents 3, 7, and 15 wt%. Each alloy was melted six times to flow controller. The mixture of gas was maintained by infu-
improve chemical homogeneity using a vacuum arc melting fur- sion of Ar and N2 gas at a ratio of 1:7 (infusion volume; Ar:
nace. Heat treatment was subsequently performed at 1000 ◦ C 5 sccm, N2 : 35 sccm) (sccm stands for standard cc/min) for 5 min.
for 24 h in an argon atmosphere for homogenization of the Pre-sputtering and deposition times were 20 and 60 min, respec-
microstructure, followed by water quenching. Specimens for elec- tively.
trochemical treatment were prepared by standard sequential Electrochemical potentiodynamic polarization studies were
metallographic polishing with emery papers and then given a carried out in 0.9% NaCl solution at 36.5 ± 1 ◦ C using a potentiostat
polish with 0.3 ␮m Al2 O3 slurry. All polished specimens were ultra- (EG&G Co., PARSTAT 2273, USA). A conventional three-electrode
sonically cleaned and degreased in acetone. Microstructures of system with high-density graphite as counter electrode and
the alloys were examined by optical microscopy (OM, Olympus saturated calomel electrode (SCE) as reference was used. The prepa-
BX60MF, Japan) and field emission scanning electron microscopy ration of the sample was as described above. The sample edges were
(FE-SEM, Hitachi 4800, Japan). The specimens for the OM and carefully covered with epoxy to avoid the possible crevice attack.
FE-SEM analysis were etched in Keller’s solution consisting of The electrolyte was deaerated using high-purity Ar gas for 30 min
2 mL HF, 3 mL HCl, 5 mL HNO3 , and 190 mL H2 O. X-ray diffrac- before starting the experiment. Deaeration was continued at a uni-
tion (XRD) was employed to identify the phases and crystallinity form rate during the experiment. The potentiodynamic polarization
in the Ti–30Ta–xZr alloys, using a diffractometer (X’pert Pro test with a scan rate of 1.67 mV s−1 was carried out from −1500 mV
MPD, Philips, Netherlands) and Cu K␣ radiation. The chemi- to 2000 mV. The Tafel extrapolation was followed to determine the
cal composition of the nanotube were characterized by X-ray corrosion parameters; based on a software-based approximation.
photoelectron spectroscopy (XPS, Kratos Axis Ultra DLD Multi- The all surface morphologies were observed by FE-SEM with high
Technique). resolution.
Electrochemical experiments were carried out with a conven-
tional three-electrode configuration having a platinum counter
electrode, a saturated calomel reference electrode, and an anode
working electrode. Before anodization, the samples were pol-
ished by standard ANSI silicon carbide papers of different
grades ranging from 100 to 2000 polished, ultrasonically cleaned
in deionized water and dried in flowing nitrogen. The sam-
ple was embedded with epoxy resin, leaving a square surface
area of 10 mm2 exposed to the anodizing electrolyte, which
was 1.0 M H3 PO4 containing NaF. Anodization treatments were
carried out using a scanning potentiostat (Model 362, EG&G,
USA).
TiN coating on the cleaned Ti–30Ta–xZr alloys surface were
performed by means of RF magnetron sputtering (CX-600S,
COMDEL, Korea) system. Pure Ti (99.998%) was used for the
coating. Before coating, the working chamber was evacuated to
1.0 × 10−6 Torr and kept at room temperature for 2 h. The RF
power for Ti deposition was 100 W, and the working pressure
was kept at 1.0 × 10−3 Torr. During sputtering of TiN, a reac-
tive gas mixture of argon (Ar) and nitrogen (N2 ) was used for
Fig. 2. XRD patterns of Ti–30Ta–xZr alloys: (a) Ti–30Ta–3Zr; (b) Ti–30Ta–7Zr; (c)
deposition onto the metal alloy plate. The feed of Ar (99.9999% Ti–30Ta–15Zr.
 W.-G. Kim, H.-C. Choe / Applied Surface Science 258 (2012) 1929–1934 1931

Fig. 3. FE-SEM images of top-views, bottom-views, and cross-section for nanotube formed on the Ti–30Ta–xZr alloys: (a), (d), (g) Ti–30Ta–3Zr; (b), (e), (h) Ti–30Ta–7Zr; (c),
(f), (i) Ti–30Ta–15Zr.

3. Results and discussion

Fig. 1 shows the microstructures of the Ti–30Ta–xZr alloys with

different Zr contents (3, 7, and 15 wt%). The microstructures of
the Ti–30Ta–3Zr and Ti–30Ta–7Zr alloys had two constituents:
an equiaxed structure and a needle-like structure. These con-
stituents changed with increasing Zr content, and an almost
entirely needle-like structure was observed in the Ti–30Ta–15Zr
alloy. Consequently, the microstructures of Ti–30Ta–xZr alloys
were strongly dependent on the amount of Zr incorporated in the
atomic structure of Ti–30Ta alloy. This could be related that Zr ele-
ment played role to increase the martensite transformation and
increased the recrystallization temperatures of the Ti–30Ta–xZr
alloys with increasing the Zr content [22].

Fig. 5. XPS profiles of nanotube formed on the surface of Ti–30Ta–xZr alloys: (a)
Fig. 4. Variations of nanotube interspace, diameter, and length with Zr content. Ti–30Ta–3Zr; (b) Ti–30Ta–15Zr.
1932 W.-G. Kim, H.-C. Choe / Applied Surface Science 258 (2012) 1929–1934

Table 1
Electrochemical parameters of Ti–30Ta–xZr alloys from anodic polarization curves: corrosion potential (Ecorr ), current density at 300 mV (I300 mV ), corrosion current density
(Icorr ), and passivation current density (Ipass ).

Samples Ecorr (mV) I300 mV (A/cm2 ) Icorr (A/cm2 ) Ipass (A/cm2 )

Non-treated Ti–30Ta–3Zr −490 1.214 × 10−5 5.170 × 10−7 8.722 × 10−6

surface Ti–30Ta–7Zr −440 9.205 × 10−6 3.584 × 10−7 5.391 × 10−6
Ti–30Ta–15Zr −430 5,242 × 10−6 1,339 × 10−7 4.082 × 10−6
Nanotube-formed Ti–30Ta–3Zr −1340 7.846 × 10−5 1.453 × 10−5 1.497 × 10−4
Ti–30Ta–7Zr −1290 5.905 × 10−5 3.311 × 10−5 3.296 × 10−5
Ti–30Ta–15Zr −1150 2.354 × 10−5 9.660 × 10−6 3.857 × 10−5
TiN-coated/nanotube-formed Ti–30Ta–3Zr −1320 3.319 × 10−5 6.171 × 10−6 1.072 × 10−4
Ti–30Ta–7Zr −1120 3.996 × 10−6 1.942 × 10−6 5.337 × 10−6
Ti–30Ta–15Zr −620 1.951 × 10−6 2.323 × 10−6 2.515 × 10−6

Peaks on the XRD patterns were interpreted with the aid of
published JCPDS information for powder standards (International
Centre for Diffraction Data, USA). XRD patterns of the homogenized
Ti–30Ta–xZr alloys are summarized in Fig. 2, in which peaks for
the martensite ␣ , ␣ and metastable ␤ phase are labeled. Fedotov
et al. [23] reported that, in Ti–xTa system, the quenched alloys have
the hexagonal (␣ ) martnesite structure with Ta content up to 26%;
the orthorhombic (␣ ) martensite with Ta content in the range of
26–52%; and the single metastable BCC ␤ structure when Ta con-
tent is larger than 65%. It can be seen that phase transformation
of quenched ternary Ti–30Ta–xZr alloys are sensitive to Zr content.
Although Zr is not considered a strong ␤-stabilizing element in Ti
alloys, its addition caused displacement of the martensite trans-
formation temperatures, a phenomenon usually attributed to the
content of ␤-stabilizing elements [24]. It has been suggested that
the ␣ (1 0 2) peak increased, whereas, the ␤ (2 1 1) peak decreased
with increasing Zr content. Therefore, the ␤ → ␣ transformation
progressed gradually with increasing Zr content due to Zr displace-
ment [25]. In Fig. 2, each diffraction peak is shifted to a lower angle
with increasing Zr content.
Fig. 3a–c shows a typical top-view FE-SEM images of nan-
otube layer prepared by the anodization of Ti alloy at 10 V in 1 M
H3 PO4 + 0.8 wt% NaF for 2 h. The nanotubes have an inner average
diameter from about 160 to 200 nm, with a tube wall thickness
of about 30 nm. The presence of Zr in the Ti–30Ta alloy leads to
formation of oxide nanotubes perpendicular to the substrate and
separated from each other by a gap (interspace) [26]. For the Zr
content of 3 and 7 wt% (Fig. 3a and b), the interspace between
the nanotubes was very small (about 40 nm). As the Zr content
increased 15 wt%, the interspace between the nanotubes increased
to about 200 nm (Fig. 3c). The diameter of the small nanotube
increased with Zr content, whereas the diameter of the large nan-
otube decreased with Zr content. Also, the interspace between
nanotubes predominantly increased with Zr content. The previous
study confirmed that ZrO2 had a role in the formation of nanotubes
on the surface [22]. We can assume this phenomena as follows,
firstly, nanotubes nucleate at the abundance of Ti content on the
metal oxide film [MO2 ] due to difference of bonding energy of Ti,
Ta and Zr oxides, secondly, small nantubes nucleate at the inter-
space which is not attacked area by F− ion with abundance of Ta
and Zr content on the metal oxide film [MO2 ] [27]. From the present
research on Ti–30Ta–xZr alloys, zirconium titanate nanotubes show
a higher structural flexibility than titanium oxide nanotubes. Thus
the diameter and depth of the nanotubes can be varied by changing

Fig. 6. FE-SEM images of TiN-coated layer on nanotubular surface by RF magnetron sputtering: (a) top-view; (b) cross-section.
 W.-G. Kim, H.-C. Choe / Applied Surface Science 258 (2012) 1929–1934 1933

the alloy composition. Fig. 3d–f shows bottom-view FE-SEM images

of nanotube layer formed on Ti–30Ta–xZr alloys. The nanotube
diameters on Ti–30Ta–3Zr and Ti–30Ta–7Zr show an irregular dis-
tribution while on the Ti–30Ta–15Zr, the oxide layer consists of
two distinct diameter tubes – larger tubes are surrounded by sev-
eral smaller tubes. From the present results, one can deduce that a
suitable amount of alloying elements is required for the nanotube
formation on two-size scales, and that the Zr content critically
affects the diameters in such arrangements [10]. Fig. 3g–i shows
cross-section FE-SEM images of the samples. A cross section was
taken from mechanically scratched samples, where some pieces
had flaked off and were upside down. The image clearly shows that
the nanotubes are open on the top, but closed on the bottom. As the
Zr content increased from 3 to 15 wt%, the average length of the
nanotubes was 2.0 ␮m, 4.0 ␮m, and 6.8 ␮m, respectively (Fig. 4).
This phenomenon that the formation of higher aspect ratio nano-
tubes will be ordered by increasing Zr content due to more latter
the equilibrium with the chemical dissolution rate than Ti.
Fig. 5 shows the XPS wide scan spectrum of nanotube formed
on the surface of Ti–30Ta–xZr alloy. The XPS spectrum indicates
the differences in chemical properties of the anodic oxide surface.
Regarding the spectrum peaks about the chemical composition of
the anodic TiO2 surface, it was reported that the O 1s originating
from the TiO2 layer has a typical binding energy of 530.3 eV, while
the binding energies of organic oxygen-containing species and the
oxides of several other metals are generally in the range 532–534 eV
[28]. The nanotube arrays are predominately titanium, zirconium,
tantalum, and oxygen, with phosphorous and fluoride species due
to solution incorporation in the nanotube. These results can be
explained that the nanotube surface is composed of TiO2 , Ta2 O5
and ZrO2 , the concerning the values of binding energy for these
peaks are in good agreement with energy reference from nanotube
layer of Ti–30Ta–xZr alloys. We can assume that contamination is
the likely source for the carbon found in the samples.
Fig. 6 shows the TiN-coated surface of the Ti–30Ta–15Zr alloy.
The surface had a golden yellow or light yellow color. At high mag-
nification, the TiN layer showed a ring-like shape at the top of
the nanotube (Fig. 5a). The cross-sectional of TiN layer formed on
the nanotube was uniformly coated with 300 nm thickness. We
observed low amplitude regular dunes, covering whole surface
(Fig. 5b). The TiN layer was found to form and regenerate the TiO2
film at the surface interface between the bones and the implant
surface as follows [28–29]:

TiN + O2 → TiO2 + 1/2 N2 G = −139 kcal/mol

Fig. 7 shows the anodic polarization curves obtained for

(a) non-treated surface, (b) nanotube formed, and (c) TiN-
coated/nanotube-formed on the Ti–30Ta–xZr alloy. Corrosion
potential (Ecorr ) was −490 mV for the Ti–30Ta–3Zr and −430 mV
for the Ti–30Ta–15Zr. As the Zr content increased, Ecorr of alloy
decreased due to Zr oxide formation. But nanotube-formed and
TiN-coated/nanotube-alloys showed the range of −1150 mV and
−620 mV. Ecorr of nanotube-formed and TiN-coated/nanotube-
formed alloys predominately decreased due to formation of
nanotube on the surface [8]. Even though Ecorr of nanotube formed
alloy decreased, passive region showed longer than that of non-
treated alloy. It is due to the presence of a passive film, a few nm
thickness, such as TiO2 , Ta2 O5 and ZrO2 , which formed rapidly
on the specimen surface [22]. The passivation current density
(Ipass ) of the nanotube formed or TiN-coated/nanotube-formed Fig. 7. Anodic polarization curves of Ti–30Ta–xZr alloys after the potentiodynamic
alloy was less noble than the non-treated surface. The corrosion test in 0.9% NaCl solution at 36.5 ± 1 ◦ C: (a) non-treated; (b) nanotube formed; (c)
current density (Icorr ) of the TiN-coated/nanotube-formed alloy is TiN-coated/nanotube-formed.
significantly higher than that of the nanotube-fromed alloy. The
TiN-coated/nanotube-formed alloy significantly decreased the cor-
rosion rate of the non-treated surface. The corrosion parameters
1934 W.-G. Kim, H.-C. Choe / Applied Surface Science 258 (2012) 1929–1934

obtained from the potentiodynamic polarization measurements, [3] Y.L. Zhou, M. Niinomi, T. Akahori, Mater. Sci. Eng. A 384 (2004) 92.
including Ecorr , I300 mV , Icorr and Ipass are summarized in Table 1. [4] A. Kar, K.S. Raja, M. Misra, Surf. Coat. Technol. 201 (2006) 3723.
[5] W.G. Kim, H.C. Choe, Y.M. Ko, W.A. Brantley, Thin Solid Films 517 (2009) 5033.
[6] X.J. Feng, J.M. Macak, S.P. Albu, P. Schmuki, Acta Biomater. 4 (2008) 318.
4. Conclusions [7] S.H. Jang, H.C. Choe, Y.M. Ko, W.A. Brantley, Thin Solid Films 517 (2009) 5038.
[8] V.S. Saji, H.C. Choe, Corros. Sci. 51 (2009) 1658.
[9] Y.H. Jeong, K. Lee, H.C. Choe, Y.M. Ko, W.A. Brantley, Thin Solid Films 517 (2009)
The following main conclusions can be drawn from the present 5365.
investigation: [10] H. Tsuchiya, T. Akaki, J. Nakata, D. Terada, N. Tsuji, Y. Koizumi, Y. Minamino, P.
Schmuki, S. Fujimoto, Corros. Sci. 51 (2009) 1528.
[11] V. Zwilling, M. Aucouturier, E. Darque-Ceretti, Electrochim. Acta 45 (1999) 921.
(1) Microstructures were changed from ␤ phase to ␣ and ␣ phase, [12] H. Tsuchiya, J.M. Macak, A. Ghicov, L. Taveira, P. Schmuki, Corros. Sci. 47 (2005)
and changed from an equiaxed to a needle-like structure with 3324.
increasing Zr content. [13] Y.H. Jeong, W.G. Kim, G.H. Park, H.C. Choe, Y.M. Ko, Trans. Nonferrous Met. Soc.
China 19 (2009) 852.
(2) The nanotubes had a mean inner diameter of about 160–200 nm
[14] K. Lee, W.G. Kim, J.Y. Cho, S.W. Eun, H.C. Choe, Trans. Nonferrous Met. Soc. China
with a tube-wall thickness of about 30 nm. The interspace 19 (2009) 857.
between the nanotubes of Ti–30Ta–3, 7 and 15Zr showed [15] W.G. Kim, H.C. Choe, Trans. Nonferrous Met. Soc. China 19 (2009) 1005.
[16] S. Rudenja, C. Leygraf, J. Pan, P. Kulu, E. Taliments, V. Mikli, Surf. Coat. Technol.
approximately 25 nm, 80 nm, and 200 nm, respectively.
114 (1999) 129.
(3) TiN film at the tip of nanotube was nucleated and grew to direc- [17] T. Okabe, T. Hero, Cells Mater. 5 (1995) 211.
tion of nanotube growth with 300 nm thickness. [18] H. Brauner, Surf. Coat. Technol. 62 (1993) 618.
(4) The TiN-coated/nanotube-formed Ti–30Ta–15Zr alloy exhib- [19] P.R. Mezger, N.H.J. Creugers, J. Dent. 20 (1992) 342.
[20] H. Altun, H. Sinici, Mater. Charact. 59 (2008) 266.
ited very low passivation current density than that of [21] S. Shimadaa, T.Y. Takadaa, J. Tsujino, Surf. Coat. Technol. 199 (2005) 72.
non-treated surface alloy. [22] W.G. Kim, H.C. Choe, W.M. Ko, J. Korean Phys. Soc. 54 (2009) 1036.
[23] S.G. Fedotov, T.V. Chelidez, Kovneristyy S Yu.K., V.V. Sanadze, Phys. Met. Metall.
62 (1986) 109.
Acknowledgement [24] D.Q. Martins, M.E.P. Souza, S.A. Souza, D.C. Andrade, C.M.A. Freire, R. Caram, J.
Alloys Compd. 478 (2009) 111.
This study was supported by NRF: No. R13-2008-010-00000-0. [25] A.V. Dobromyslove, V.A. Elkin, Scripta Mater. 44 (2001) 905.
[26] J.M. Macak, H. Tsuchiya, A. Ghicov, K. Yasuda, R. Hahn, S. Bauer, P. Schmuki,
Curr. Opin. Solid State Mater. Sci. 11 (2007) 3.
References [27] H.C. Choe, Y.H. Jeong, W.A. Brantley, J. Nanosci. Nanotechnol. 10 (2010) 4684.
[28] C. Massaro, P. Rotolo, F. De Riccardis, E. Milella, A. Napoli, M. Wieland, M. Textor,
[1] M. Niinomi, D. Kuroda, K. Fukunaga, M. Morinaga, Y. Kato, T. Yashiro, A. Suzuki, N.D. Spencer, D.M. Brunette, J. Mater. Sci. Mater. Med. 13 (2002) 535.
Mater. Sci. Eng. A 263 (1999) 193. [29] D.M. Brunette, P. Tengvall, M. Textor, P. Thomsen, Titanium in Medicine,
[2] X. Wang, Y. Li, J. Lin, P.D. Hodgson, C. Wen, Dent. Mater. 24 (2008) 1549. Springer, Berlin, 2001.