ARTICLE IN PRESS

Biomaterials 26 (2005) 22652272

www.elsevier.com/locate/biomaterials

Carbon plasma immersion ion implantation of

nickeltitanium shape memory alloys
R.W.Y. Poona, K.W.K. Yeungb, X.Y. Liua, P.K. Chua,, C.Y. Chunga,
W.W. Lub, K.M.C. Cheungb, D. Chanc
a
Department of Physics and Materials Science, City University of Hong Kong, Kowloon, Hong Kong
b
Department of Orthopaedics and Traumatology, The University of Hong Kong, Pokfulam, Hong Kong
c
Department of Biochemistry, The University of Hong Kong, Pokfulam, Hong Kong
Received 14 April 2004; accepted 20 July 2004
Available online 11 September 2004

Abstract

Nickeltitanium (NiTi) shape memory alloys possess super-elasticity in addition to the well-known shape memory effect and are
potentially suitable for orthopedic implants. However, a critical concern is the release of harmful Ni ions from the implants into the
living tissues. We propose to enhance the corrosion resistance and other surface and biological properties of NiTi using carbon
plasma immersion ion implantation and deposition (PIII&D). Our corrosion and simulated body uid tests indicate that either an
ion-mixed amorphous carbon coating fabricated by PIII&D or direct carbon PIII can drastically improve the corrosion resistance
and block the out-diffusion of Ni from the materials. Our tribological tests show that the treated surfaces are mechanically more
superior and cytotoxicity tests reveal that both sets of plasma-treated samples favor adhesion and proliferation of osteoblasts.
r 2004 Elsevier Ltd. All rights reserved.

Keywords: NiTi shape memory alloys; Orthopedic implants; Corrosion resistance; Mechanical properties; Osteoblasts

1. Introduction more prominent in stainless steels and titanium alloys

Nickeltitanium (NiTi) shape memory alloys possess
unique properties such as shape memory effects and
super-elasticity that most metallic biomaterials such as
stainless steels and titanium alloys do not have. More-
over, its elastic modulus [1] is closer to that of cortical
bones [2] as compared to that of other implantable
materials such as stainless steels [1,3] and titanium alloys
[3,4]. Favorable properties of NiTi have since been
investigated [57]. Several studies have indicated that
NiTi alloys exhibit good bio-compatibility in living
tissues [811], but a number of adverse effects have also
been reported. For instance, the osteogenesis process
and osteonectin synthesis activity have been found to be
Corresponding author. Tel: +852-27887724; fax: +852-27889549
or +852-27887830.
E-mail address: paul.chu@cityu.edu.hk (P.K. Chu).
compared to NiTi [12]. Other studies have revealed that
the cell death rate is more pronounced on NiTi [13] and
that its relatively poor corrosion resistance may lead to
an increase in cytotoxicity since deleterious materials
may be released from the substrate [14]. It is believed
that the noxious effects primarily stem from the leaching
of nickel ions [8,9] from the alloys causing toxic
reactions in humans, especially nickel hyper-sensitive
patients resulting in strong allergic reactions [15,16]. The
corrosion resistance of NiTi depends very much on
factors such as the surface morphology and homogene-
ity of the microstructures.
Carbon [17] and titanium carbide [18] layers have
been shown to possess excellent bio-compatibility under
some circumstances. Another advantage of these coat-
ings is that they also possess good mechanical strength.
Amorphous hydrogenated carbon lms have been
produced by electron cyclotron resonancemicrowave

0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2004.07.056
 ARTICLE IN PRESS
2266 R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272
plasma chemical vapor deposition (ECRMPCVD) [19],
plasma-enhanced chemical vapor deposition (PECVD)
[20], ltered cathodic vacuum arc (FCVA) deposition
[17], and radio-frequency (RF) glow discharge [21]. In
this work, we investigate the possibility of enhancing the
surface properties of NiTi shape memory alloys using
carbon plasma immersion ion implantation and deposi-
tion (PIII&D) [22]. The notion of using a surface lm to
combat corrosion is quite common but poor interfacial
properties may lead to premature lm delamination.
Implantation does not create a clear interfacial region
but surface wear and tear may be more severe compared
to a hard coating. Therefore, in this work, we studied
the efcacy of both approaches. In our rst experiment,
a graded C/NiTi interface was produced followed by the
deposition of an amorphous hydrogenated carbon lm
using RF C2H2 PIII&D. In our second set of experi-
ments, carbon was implanted into the specimens using
high-voltage glow discharge PIII in a C2H2 ambient.
Our objective is to produce a carbon lm or carbide
layer to enhance the surface anti-corrosion, mechanical,
as well as biological properties of NiTi shape memory
alloys.
2. Experimental methods
Circular NiTi bars with 50.8% Ni (SE508, Nitinol
Device Company, Fremont, USA) were cut into discs of
5 mm in diameter and 1 mm in thickness. They were
grinded, polished to a shiny surface texture, and then
ultrasonically cleaned with acetone and ethanol before
deposition or implantation was conducted in our plasma
immersion ion implanter. The deposition and implanta-
tion parameters are displayed in Table 1. Before
PIII&D, a glow discharge [23] was ignited in the same
C2H2 ambient at 30 kV for 20 min to produce a graded
carbide layer to enhance lm adhesion. The C2H2
plasma was sustained throughout the entire PIII&D
Table 1
Deposition/implantation parameters
Samples
#1
Description Control
RF
High voltage
Pulse width
Frequency
Duration
Base pressure
C2H2 working pressure
Dose
Vacuum annealing
process and so the dominant process here (Sample #2 in
Table 1) was deposition because the implantation duty
cycle was very low. On the other hand, during PIII alone
(Sample #3 in Table 1), the C2H2 plasma was ignited by
high-voltage glow discharge only when a high-voltage
pulse was applied to the sample. The predominant
process was thus ion implantation with very little
deposition because there was no plasma formation
between high-voltage pulses.
The elemental depth proles and Ti and C chemical
states were determined by X-ray photoelectron spectro-
scopy (XPS) (Physical electronics PHI 5802, Minnesota,
USA). Nano-indentation tests (MTS Nano Indenter XP,
USA) were conducted on ve areas to determine the
average hardness and Youngs modulus of the treated
and control samples. The electrochemical tests [24]
based on ASTM G5-94 (1999) and G61-86 (1998) were
performed by a potentiostat (VersaStat II EG&G, USA)
using a standard simulated body uid (SBF) at a pH
of 7.42 [25] and temperature of 3770.5 1C. The
ion concentrations in the SBF are shown in Table 2
[25]. A cyclic potential spanning between 400 and
+1600 mV was applied at a scanning rate of 600 mV/h.
Before the electrochemical tests, the medium was purged
with nitrogen for 1 h to remove dissolved oxygen and
nitrogen purging continued throughout the measure-
ments. The SBF taken from each sample after the
corrosion test was analyzed for Ni and Ti employing
inductively coupled plasma mass spectrometry (ICPMS)
(Perkin Elmer, PE SCIEX ELAN6100, USA).
In order to investigate the bioactivity of the control,
deposited, and implanted samples, osteoblasts isolated
from calvarial bones of 2-day-old mice that ubiquitously
expressed an enhanced green uorescent protein
(EGFP) were cultured in a Dulbeccos modied Eagles
medium (DMEM) (Invitrogen) supplemented with 10%
(v/v) fetal bovine serum (Biowest, France), antibiotics
(100 U/ml of penicillin and 100 mg/ml of streptomycin),
and 2 mM L-glutamine at 37 1C in an atmosphere of 5%
#2 #3
30 kV C2H2 40 kV C2H2
Plasma deposition Plasma implantation
500 W
30 kV 40 kV
100 ms 30 ms
100 Hz 200 Hz
90 min 90 min
1  10 5 Torr 1  10 5 Torr
3.8  10 4 Torr 2.0  10 3 Torr
4.5  1016 cm 2 5.5  1016 cm 2
1.0  10 5 Torr 600 1C for 5 h
 ARTICLE IN PRESS
R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272
Table 2
Ion concentration of SBF in comparison with human blood plasma [25].
Concentration (mmol)
Na+ K+ Ca2+
SBF 142.0 5.0 2.5
Blood plasma 142.0 5.0 2.5
CO2 and 95% air. The specimens (1 mm in thickness and
5 mm in diameter) were xed on the bottom of a 24-well
tissue culture plate (Falcon) using 1% (w/v) agarose. A
cell suspension consisting of 5000 cells was seeded onto
the surface of the untreated NiTi samples, deposited
NiTi specimens, implanted NiTi samples, and wells
without any metal discs serving as a control for normal
culturing condition. Cells were grown in 1 ml of medium
and changed every 2 days. Cell attachment was
examined after the rst day of culture, and cell
proliferation examined after 3 and 5 days of culture.
In our study, cells were allowed to reach conuence
during the examination period. To determine the cell
number, attached cells were released by digestion with
trypsinEDTA (Invitrogen) and counted using a hema-
tocytometer (Tiefe Depth Profondeur, Marienfeld,
Germany). Cell viability was assessed by staining with
0.2% Trypan blue (Sigma). The number of cells was
expressed as mean value7standard deviation (SD). The
data were analyzed by using unpaired two-sample t-test
in this study and the statistical analysis was performed
using the SPSS program (SPSS for Windows, Release
11.0.0, SPSS Inc., Chicago, USA)
Cell morphology was assessed using a uorescent
microscope (Axioplan 2, Carl Zeiss, Germany). The
attached living EGFP-expressing osteoblasts were vi-
sualized using a 450490 nm incident lter and the
uorescence images emitted at 510 nm captured using a
Sony DKS-ST5 digital camera.
3. Results
Fig. 1 shows the XPS depth proles obtained from the
untreated control, deposited, as-implanted, and im-
planted and annealed samples. The proles have been
plotted on a depth scale based on sputtering rates
calculated from a SiO2 reference under similar condi-
tions. Since it is well known that the sputtering rate
changes in the surface region and is different from that
of SiO2, the thicknesses of the carbon layer and
implanted zone shown in Fig. 1 are approximate,
but comparison among different samples is more valid.
Figs. 1c and d show that the annealing process broadens
the implanted layer. It is also obvious from the depth
2267
Mg2+ HCO3 Cl HPO24 SO24
1.5 4.2 148.5 1.0 0.5
1.5 27.0 103.0 1.0 0.5
proles that the Ni content is suppressed in the
implanted region.
The hardness results and Youngs modulus derived
from nano-indentation data are shown in Figs. 2ac. In
the control sample, both the hardness and Youngs
modulus are quite stable throughout the depth of
measurement. The hardness is 4.5 GPa while the
Youngs modulus is about 57 GPa. In the deposited
sample, the maximum hardness is 7 GPa at 50 nm from
the surface. It decreases very slowly and reaches a
constant value of 5.5 GPa at 165 nm. The Youngs
modulus, being about 55 GPa, is quite constant
throughout the depth of the measurement. The lower
hardness and Youngs modulus values in the rst 20 nm
of the deposited sample is probably due to the surface
substances such as moisture or oxides. In the implanted
and annealed sample, the maximum hardness is 9.5 GPa
at around 40 nm from the surface and gradually
diminishes to 4.5 GPa at 150 nm from the surface. The
Youngs modulus shows the maximum value of 110 GPa
at the topmost layer and then decreases gradually
achieving a rather constant value of 70 GPa between
100 and 150 nm from the surface. It should be noted that
these are higher than that of the control sample of
57 GPa. Our results show that the Youngs modulus of
the deposited carbon coating is 3.5% smaller than that
of the substrate, whereas the hardness is 2256% higher.
Hence, the carbon coating is mechanically stronger than
the substrate, or at least comparable to the latter. As to
the implanted and annealed sample, the hardness of
the treated layer is 11111% greater than that of the
substrate while the Youngs modulus is 9323% higher
throughout the depth of the measurement. Thus, the
carbide layer is also mechanically stronger than the
substrate.
Table 3 summarizes the essential ndings from our
electrochemical tests with Ecorr and Eb representing,
respectively, the corrosion potential and breakdown
potential as indicated by the initiation of pits. The
higher Ecorr and Eb suggest improvements in the
corrosion resistance. During our test, breakdown
occurred after 272 mV and the current increased
signicantly to 10 2 A in the control sample (#1). In
contrast, for the deposited (#2) and implanted and
annealed (#3) samples, the currents remained at
 ARTICLE IN PRESS
2268 R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272

NiTi Controlsample 30kV C2H2 deposition without annealing

70 C1s
100
O1s
60 Ti2p
Ni2p Ni2p
Atomic concentration (%)

Atomic concentration (%)

Ti2p 80
50 O1s
40 60

30
40
20

10 20

0 0
-10
0 10 20 30 40 50 60 70 0 50 100 150 200 250
(a) Depth (nm) (b) Depth (nm)

40kV C2H2 implantation without annealing 80 40kV C2H2 implantation with

70
65 600 C annealing
70
60 Atomic concentration (%)
55 C1s 60
Atomic concentration (%)

C1s
50 O1s O1s
45 Ti2p 50
Ti2p
40 Ni2p Ni2p
40
35
30 30
25
20 20
15
10
10
5 0
0
-5 -10
0 50 100 150 200 0 50 100 150 200
(c) Depth (nm) (d) Depth (nm)

Fig. 1. XPS depth proles acquired from the NiTi alloys: (a) #1 (untreated), (b) #2 (deposited without annealing), (c) #3 (implanted without
annealing), and (d) #3 (implanted and annealed).

relatively low levels even after 1600 mV had been
reached.
The SBF solutions after the electrochemical tests were
analyzed for Ni and Ti concentrations by ICPMS and
the results are summarized in Table 4. Our data show
that the amount of Ni leached from the control
(untreated) sample during the electrochemical test is
orders of magnitude higher than those detected from the
deposited and implanted samples. In fact, the amounts
of Ni detected from the SBF of both the deposited and
implanted and annealed samples are very close to the
instrumental detection limits. The ICPMS results
provide strong evidence that either PIII&D or PIII can
effectively block the out-diffusion of Ni from the NiTi
shape memory alloys to the SBF.
The cell morphology of the NiTi control (untreated),
NiTi after carbon deposition, and NiTi after carbon
implantation after 3 days of culturing are depicted in
Figs. 3(ac), respectively. It can be clearly seen that cells
are attached to and proliferate on all the samples. Fig. 4
shows that all the samples are well tolerated by the GFP
osteoblast cell line. After 3 days of culturing, the number
of cells was 7.3  104 (SD 1.2  104) on the NiTi control
sample, 20.6  104 (SD 3.0  104) on the carbon
deposited sample, and 12.5  104 (SD 1.2  104) on the
carbon implanted sample. After 5 days, the number was
13.3  104 (SD 2.3  104), 27.3  104 (SD 1.2  104), and
16.6  104 (SD 1.2  104) on the three samples, respec-
tively. After culturing for 3 and 5 days, the overall cell
proliferation on both the deposited and implanted
samples is much higher than that on the untreated NiTi
substrate. Dead cells cannot be found on any samples.
This reveals that the cyto-compatibility of the treated
samples is as good as NiTi [8,9]. The carbon deposited
samples are observed to be most favorable to cell
proliferation compared to others (po0:01).
 ARTICLE IN PRESS
R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272 2269

160 NiTi Control 24 Table 3

22 Essential results from the electrochemical tests
140
20
Young's modulus Samples
Young's modulus (GPa)

Nano-hardness (GPa)
120 Nano-hardness 18
16 #1 #2 #3
100
14
Ecorr (mV) 231 60 114
80 12
Eb (mV) 272 1145 1170
10
60
8
40 6
4
20 Table 4
2
0 0 Amount of Ni and Ti ions detected in SBF by ICPMS after
0 20 40 60 80 100 120 140 160 180 electrochemical tests
(a) Depth (nm)
Samples Ni content (mg/l) Ti content (mg/l)

30kV C2H2 deposition without annealing Ni Ti

80 20 #1 30.2324 0.1575
Young's modulus #2 0.0056 (ND) 0.0228
Nano-hardness
18
#3 0.0082 (ND) 0.057
16
Nano-hardness (GPa)
Young's modulus (GPa)

60
14 Note: mg/l=ppm; ND=not detected.
12
40 10
20% at about 100 nm. After that, it decreases gradually
8
to 0% at about 200 nm. This depth prole agrees with
6
20 the results of the high-resolution XPS analyses of the
4 Ti2p and C1s at different sputtered depths (not shown
2 here). Only a-carbon can be detected in the top 11 nm,
0 0 and from about 11 nm to 68 nm, the chemical shifts are
0 20 40 60 80 100 120 140 160 180 not signicant indicating that the Ti to C ratio remains
(b) Depth (nm) less than 0.15 indicative of a lm consisting of
predominantly carbon. At greater depths, chemical shifts
40kV C2H2 implantation with 600C annealing
160 24 representative of carbide emerge revealing unequivocally
150 22 the existence of a graded carbide interface between the
140 carbon lm and substrate. This graded interface may
130 Young's modulus 20
Young's modulus (GPa)

Nano-hardness (GPa)

Nano-hardness 18 enhance the lm adhesion and mitigate lm delamina-

120
110 16 tion. For the implanted and annealed sample, the shift
100
90 14 from a-carbon to carbide occurs at 11 nm, indicating
80 12 that an implanted zone has been formed. High-resolu-
70 10 tion Ti2p and C1 s spectra disclose the formation of
60
50 8 titanium carbide layer with graded carbon.
40 6 The suppression of the Ni content in the implanted
30 4
20 region after annealing is consistent with previous results
2
10 and can be attributed to the high afnity of titanium
0 0
0 20 40 60 80 100 120 140 160
towards carbon and oxygen under high-temperature
(c) Depth (nm) annealing [26]. Oxygen was co-implanted into the
samples as a result of the non-UHV (ultra-high vacuum)
Fig. 2. Nano-hardness and Youngs modulus proles acquired from implantation conditions in our PIII instrument [22]. The
the NiTi alloys: (a) #1, (b) #2, and (c) #3.
suppression of Ni in the implanted region is a benecial
phenomenon and if the carbide layer is inert and
mechanically strong enough, it may be a good candidate
4. Discussion for the Ni out-diffusion barrier.

4.1. XPS depth profiles 4.2. Nano-indentation

Fig. 1b shows that the carbon content is quite uniform Our results show that both the carbon coating
in the rst 60 nm and then drops rapidly from 90% to and implanted titanium carbide layer are mechanically
 ARTICLE IN PRESS
2270 R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272

Fig. 3. (a) Micrograph of Sample #1 after 3 days of cell culturing showing the EGFP-expressing mouse osteoblasts. Proliferation clusters are obvious
on the surface (Top: 100  magnication; Bottom: 200  magnication). (b) Sample #2. (c) Sample #3.

stronger than the substrate. Although the hardness
and Youngs modulus obtained in our experiments
are not as high as those claimed by Miyoshi et al.
in TixCy [27], mechanical enhancement has unam-
biguously been demonstrated. However, further studies
on the performance of the coating and the carbide layer
using the super-elastic and shape memory properties
have to be conducted to see their performance under
normal working conditions.
4.3. Corrosion resistance
The enhancement in the corrosion resistance of the
carbon lm may be attributed to the reduced electrical
 ARTICLE IN PRESS
R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272
30
NiTi-Untreated
NiTi-Deposited
Number of Cells (X10000)
25 NiTi-Implanted
Empty well (Control)
20
15
10
5
0
3 5
Day
Fig. 4. Cell proliferation versus number of days.
conductivity in comparison to NiTi alloys. The ex-
change of electrical charges at the sample surface is
reduced, consequently the extent of the electrochemical
corrosion [28]. Apart from this, carbon is inherently
chemically inert and titanium and carbon form very
strong and inert carbide bonds in the implanted samples.
This may play a role in its good corrosion resistance as
well. However, it should be noted that our electro-
chemical test is an accelerated corrosion test, and
although promising results have been obtained, more
work must be done to investigate the long-term effects.
We plan to conduct more extensive tests on their anti-
corrosion performance after the mechanical test using
the super-elasticity and shape memory effects to verify
the reliability of the treatments.
4.4. Cells test
The rate of proliferation can vary due to the
conditions of the substrate such as surface free energy,
surface stress, surface morphology, as well as interfacial
free energy [29]. Ponsonnet et al. [30] reported a low rate
of proliferation of human gingival broblasts on NiTi
samples with rough surfaces when compared to stainless
steels and Ti alloys with the same surface roughness.
Kapanen et al. [9] studied the transformation of growth
factor-b1 (TGF-b1) expression in ROS-17/2.8 osteosar-
coma cell line in different orthopedic implants with
different roughness and found that the NiTi with rough
surface promotes TGF-b1 secretion, and the same effect
has been found on some other rough titanium alloys as
well. It is known that this cytokine expression can
inhibit osteoblast proliferation and promote early
differentiation of osteoblast [31]. Es-Souni et al. [14]
also reported that the rough surface resulted in low cyto-
compatibility. It is likely that the surface roughness may
alter the corrosion resistance of the materials. If the
corrosion resistance is reduced, leaching of metal ions is
very possible thereby lowering the cyto-compatibility of
the substrate. The results reported in this paper suggest
2271
that carbon PIII&D is an effective method to improve
the surface anti-corrosion, mechanical, and biological
properties of NiTi, and we are conducting more work to
fathom the mechanism as well as other factors that may
impact the results.
5. Conclusion
Using PIII&D, we have deposited amorphous hydro-
genated carbon thin lms and implanted carbon into
NiTi substrates. The carbon coating formed on the
substrate has a graded carbide interface, thereby
strengthening lm adhesion. Titanium carbide is formed
in the implanted samples. Both sets of treated samples
exhibit better corrosion resistance compared to the
untreated control sample based on our corrosion test
and ICMPS analysis of the SBF. Mechanically, the
implanted and annealed carbide layer is stronger than
the NiTi substrate, while the deposited carbon coating is
a little stronger or at least comparable to the NiTi
substrate. Cell culture experiments reveal that the
treated samples are cyto-compatible. Thus, we have
demonstrated that PIII&D is effective in improving the
corrosion resistance, mechanical properties, and surface
biological properties of orthopedic NiTi shape memory
alloys.
Acknowledgments
We thank Mr. Wilson C.W. Chan for his expert
technical assistance. The work was jointly supported by
Hong Kong Research Grants Council (RGC) Compe-
titive Earmarked Research Grants (CERG) # CityU
1137/03E and # HKU 7283/00 M, a grant from the
Germany/Hong Kong Joint Research Scheme spon-
sored by the Research Grants Council of Hong Kong
and the German Academic Exchange Service Project
No. G_HK001/02, and City University of Hong Kong
Strategic Research Grant (SRG) # 7001447.
References
[1] Pelletier H, Muller D, Mille P, Grob JJ. Structural and
mechanical characterisation of boron and nitrogen implanted
NiTi shape memory alloy. Surf Coat Technol 2002;158159:
30917.
[2] Bayraktar HH, Morgan EF, Niebur GL, Morris GE, Wong EK,
Keaveny TM. Comparison of the elastic and yield properties of
human femoral trabecular and cortical bone tissue. J Biomech
2004;37(1):2735.
[3] Long M, Rack HJ. Titanium alloys in total joint replacementa
materials science perspective. Biomaterials 1998;19(18):162139.
[4] Ryhanen J. Biocompatibility evaluation of nickeltitanium shape
memory metal alloy. University of Oulu; 1999. p. 29.
 ARTICLE IN PRESS
2272 R.W.Y. Poon et al. / Biomaterials 26 (2005) 22652272
[5] Kujala S, Ryhanen J, Danilov A, Tuukkanen J. Effect of porosity
on the osteointegration and bone ingrowth of a weight-bearing
nickeltitanium bone graft substitute. Biomaterials 2003;
24:46917.
[6] Kujala S, Ryhanen J, Jamsa T, Danilov A, Saaranen J, Pramila
A, Tuukkanen J. Bone modeling controlled by a nickeltitanium
shape memory alloy intramedullary nail. Biomaterials 2002;23:
253543.
[7] Ryhanen J, Kallioinen M, Serlo W, Peramaki P, Junila J, Sandvik
P, Niemela E, Tuukkanen J. Bone healing and mineralization,
implant corrosion, and trace metals after nickeltitanium shape
memory metal intramedullary xation. J Biomed Mater Res
1999;47:47280.
[8] Kapanen A, Ilvesaro J, Danilov A, Ryhanen J, Lehenkari P,
Tuukkanen J. Behavior of nitinol in osteoblast-like ROS-17 cell
cultures. Biomaterials 2002;23:64550.
[9] Kapanen A, Kinnunen A, Ryhanen J, Tuukkanen J. TGF-[beta] 1
secretion of ROS-17/2.8 cultures on NiTi implant material.
Biomaterials 2002;23:33416.
[10] Kapanen A, Ryhanen J, Danilov A, Tuukanen J. Effect of nickel-
titanium shape memory metal alloy on bone formation. Bioma-
terials 2001;22:247580.
[11] Kujala S, Pajala A, Kallioinen M, Pramila A, Tuukkanen J,
Ryhanen J. Biocompatibility and strength properties of nitinol
shape memory alloy suture in rabbit tendon. Biomaterials
2004;25:3538.
[12] Berger-Gorbet M, Broxup B, Rivard C, Yahia LH. Biocompat-
ibility testing of NiTi screws using immunohistochemistry on
sections containing metallic implants. J Biomed Mater Res 1996;
32:2438.
[13] Jia M, Beatty MW, Reinhardt RA, Petro TM, Cohen DM, Maze
CR, Strom EA, Hoffman M. Nickel release from orthodontic
arch wires and cellular immune response to various nickel
concentrations. J Biomed Mater Res 1999;48:48895.
[14] Es-Souni M, Fischer-Brandies H. On the properties of two binary
NiTi shape memory alloys. Effects of surface nish on the
corrosion behavior and in vitro biocompatibility. Biomaterials
2002;23:288794.
[15] Lamster IB, Kalfus DI, Steigerwald PJ, Chasens AI. Rapid loss of
alveolar bone associated with nonprecious alloy crowns in two
patients with nickel hypersensitivity. J Periodont 1987;58:48692.
[16] Espana A, Alonso ML, Soria C, Guimaraens D, Ledo A. Chronic
urticaria after implantation of 2 nickel-containing dental pros-
theses in a nickel-allergic patient. Contact Dermat 1989;21:2046.
[17] Sheeja D, Tay BK, Nung LN. Feasibility of diamond-like carbon
coatings for orthopaedic applications. Diamond Relat Mater
2004;13(1):18490.
[18] Wu SK, LeeH CY, Lin C. A study of vacuum carburization of an
equiatomic TiNi shape memory alloy. Scripta Mater 1997;
37(6):83742.
[19] Jian SR, Fang TH, Chuu DS. Nanoindentation investigation of
amorphous hydrogenated carbon thin lms deposited by ECR-
MPCVD. J Non-Cryst Solids 2004;333(3):2915.
[20] Huran J, Hrubcin L, Kobzev AP, Liday J. Properties of
amorphous silicon carbide lms prepared by PECVD. Vacuum
1996;47(10):12235.
[21] Seth T, Cabarrocas PR. Plasma deposition of carbon lms at
room temperature from C2H2Ar mixtures: anodic vs. cathodic
lms. Thin Solid Films 2001;383(12):2169.
[22] Chu PK. Recent Developments and applications of plasma
immersion ion implantation (PIII). J Vac Sci Technol B 2004;
22(1):28996.
[23] Tian XB, Peng P, Chu PK. Enhancement of process efcacy using
seed plasma in pulsed high-voltage glow-discharge plasma
implantation. Phys Lett A 2002;303(1):1103.
[24] Starosvetsky D, Gotman I. Corrosion behavior of titanium
nitride coated NiTi shape memory surgical alloy. Biomaterials
2001;22(13):18539.
[25] Cho SB, Nakanishi K, Kokubo T, Soga N. Dependence of apatite
formation on silica gel on its structure: effect of heat treatment.
J Am Ceram Soc 1995;78:176974.
[26] Tian XB, Fu RKY, Wang LW, Chu PK. Oxygen-induced nickel
segregation in nitrogen plasma implanted AISI stainless steel.
Mater Sci Eng A 2004;316:2004.
[27] Miyoshi K, Pohlchuck B, Street KW, Zabinski JS, Sanders JH,
Voevodin AA, Wu RLC. Sliding wear and fretting wear of
diamond like carbon-based, functionally graded nanocomposite
coatings. Wear 1999;225-9(1):6573.
[28] Dorner-Reisel A, Schurer C, Irmer G, Muller E. Electrochemical
corrosion behaviour of uncoated and DLC coated medical grade
Co28Cr6Mo. Surf Coat Technol 2004;177-8:8307.
[29] Ponsonnet L, Reybier K, Jaffrezic N, Comte V, Lagneau C,
Lissac M, Martelet C. Relationship between surface properties
(roughness, wettability) of titanium and titanium alloys and cell
behavior. Mater Sci Eng C 2003;23:55160.
[30] Ponsonnet L, Comte V, Othmane A, Lagneau C, Charbonnier M,
Lissac M, Jaffrezic N. Effect of surface topography and chemistry
on adhesion, orientation and growth of broblasts on nickel
titanium substrates. Mater Sci Eng C 2002;21:15765.
[31] Bonewald L, Kester M, Schwartz Z, Swain L, Khare A, Johnson
T, Leach R, Boyan B. Effects of combining transformation
growth factor beta and 1,25-dihydroxyvitamin D3 on differentia-
tion of a human osteosarcoma (MG-63). J Biol Chem 1992;267:
89439.