Materials Science and Engineering C 78 (2017) 682–689

Contents lists available at ScienceDirect

Materials Science and Engineering C

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

The effect of pH, fluoride and tribocorrosion on the surface properties of

dental archwires
Petra Močnik a, Tadeja Kosec a,⁎, Janez Kovač b, Milan Bizjak c
a
Slovenian National Building and Civil Engineering Institute, Dimičeva 12, SI - 1000 Ljubljana, Slovenia
b
Jozef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia
c
Faculty of Natural Sciences and Engineering, University of Ljubljana, Aškerčeva 12, SI-1000 Ljubljana, Slovenia

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

Article history: Nickel-titanium and stainless steel are the most commonly used alloys for orthodontic treatments. Even though
Received 8 November 2016 both are known to be resistant to corrosion, there are circumstances that can lead to undesired situations, like
Received in revised form 28 March 2017 localized types of corrosion attack, wear during sliding of an archwire though brackets and breakdowns due to
Accepted 6 April 2017
iatrogenic causes. The aim of this research was to analyse the influence of environmental effects on the corrosion
Available online 17 April 2017
and tribocorrosion properties of NiTi and stainless steel dental alloys. The effects of pH and fluorides on the elec-
Keywords:
trochemical properties were studied using the cyclic potentiodynamic technique. The migration of ions from the
Dental wires alloy into saliva during exposure to saliva with and without the presence of wear was analysed using ICP-MS
NiTi analyses. Auger spectroscopy was used to study the formation of a passive oxide layer on different dental alloys.
Stainless steel It was found that lowering the pH preferentially affects the corrosion susceptibility of NiTi alloys, whereas stain-
Tribocorrosion less steel dental archwires are prone to local types of corrosion. The NiTi alloy is not affected by smaller increases
Passive film of fluoride ions up to 0.024 M, while at 0.076 M (simulating the use of toothpaste) the properties are affected. A
Simulated saliva leaching test during wear-assisted corrosion showed that the concentrations of Ni ions released into the saliva
Ni release
exceeded the limit value of 0.5 μg/cm2/week. The oxide films on the NiTi and stainless steel alloys after the
tribocorrosion experiment were thicker than those exposed to saliva only.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction such as adopted mechanical testing, tribocorrosion testing and electro-

chemical testing.
Biocompatible metals have had a wide range of applications for a The corrosion performance of NiTi dental alloys has been investi-
number of years. However, their use is not only dependent on the bio- gated with respect to surface treatment methods in different studies
compatibility properties, but also on the requirements for their use [21–26]. Ar+, N+ and Nb+ implantation by plasma reduces the corro-
and applications in the human body. In general, the materials should sion current densities and produces surfaces with depleted Ni [23,27].
be biocompatible during the service life of their implantation or other Moreover, different pre-treatments affect the corrosion protection of
use. There are three main groups of biocompatible alloys in current such alloys to a large extent, e.g., a surface film produced by dry oxida-
use, i.e., Cr-rich stainless steels, Co\\Cr alloys and Ti alloys, among tion between 100 and 400 °C or boiling in water, which has the greatest
which NiTi has applications in dentistry because of its shape-memory effect on polished surfaces [22]. It was shown that electrochemical
properties. The research in this field, including electrochemical, me- polishing resulted in the highest TiO2:Ni ratio, with a minimal Ni pres-
chanical, microstructural, ion leaching and protein chelating, is exten- ence on the surface [26].
sive [1–12]. Concerning the tribocorrosion properties (tribocorrosion being the
In dental applications, stainless steel and NiTi alloy archwires are the simultaneous action of corrosion and wear) of NiTi alloys, studies are
most common; however, different Co\\Cr and improved NiTi alloys are very scarce [28]. There are only reports on tribological properties of
attracting more research interest [2,13–15]. The number of studies on NiTi alloy, without the presence of corrosive media [11,29,30]. However,
NiTi and stainless steel alloys is high [16], although there are many Ni release under corrosion with the simultaneous effect of mechanical
fewer for archwires [1–3,17–20]. The research on such demanding load in such environmental conditions has not yet been reported, even
shapes in the case of dental alloys results in experimental challenges, though it is very relevant in dental applications. As such, it has been in-
cluded in our study.
Since much of the reported research on dental materials describes
⁎ Corresponding author. the surface preparation and the state of the material as being detrimen-
E-mail address: tadeja.kosec@zag.si (T. Kosec). tal to the properties of such alloys when in use, it is important to focus

http://dx.doi.org/10.1016/j.msec.2017.04.050
0928-4931/© 2017 Elsevier B.V. All rights reserved.
 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689 683

on their findings and conclusions in order to study the properties of the the potentials are reported with respect to the saturated calomel elec-
archwires themselves. trode (SCE) scale. At least three measurements were performed to iden-
The aim of the present study was to study the properties of dental al- tify and minimise the errors in the electrochemical testing [32].
loys, i.e., NiTi dental archwires and stainless steel AISI 304 archwires, in
the conditions most commonly expected when the archwires are in use. 2.4. Tribo-electrochemical measurements
The effects of fluoride and pH were studied in simulated saliva. Also,
the concentration change of the dissolved ions during tribological wear A reciprocal tribometer was used for the tribocorrosion experiments
of the archwire and the properties of the oxide films before and after the (Tribotechnic, 2009, France).
wear were studied. Surface characterization by AES depth profiling was The normal loads in the tests were 1 N (with a contact pressure of
used for the chemical analysis of the surface oxide film at exposed and 750 MPa for the stainless steel and 300–480 MPa for the NiTi alloy),
worn surfaces in artificial saliva with and without the presence of fluo- with an average sliding speed of 0.33 mm/s and with a 6-mm diameter
ride ions. Al2O3 counter-body ball. The wear length was 2000 mm, which relates
to 1.7 h of wear and a total exposure time of 5 h. The metal samples
2. Materials and methods were immersed in a controlled volume of artificial saliva in order to
obtain information about the concentrations of the dissolved species
2.1. Preparation of the samples after the wear experiment.

Samples prepared from the two dental alloys wires (Dentsply GAC,
2.5. Exposure conditions and ICP/MS analysis
USA) were studied:

(1) Ni\\Ti alloy (51% Ni, Ti balance, denoted as NiTi archwires in this Dental archwires, cut length 0.4 cm, with an area 0.79 cm2 for the SS
study), Neo Sentalloy, rectangle shape, (0.457 × 0.635) mm. and 0.81 cm2 for the NiTi alloy were exposed separately in artificial sa-
(2) AISI 304 (denoted as SS archwires in this study) (69.5% Fe, liva at 37 °C for 21 days in 100 mL of simulated saliva solution. After
18.5% Cr, 9% Ni, 1% Mn, 0.75% Co, Si, C, P and S less than 1%), rect- 24 h, 48 h, 72 h, 1 week, 2 weeks and 3 weeks, 5 mL of the solution sam-
angle shape, (0.483 × 0.635) mm. ple from the middle of the flask was taken for the analysis.
For the tribocorrosion exposure, a 2–3 cm archwire was cut,
degreased and prepared for the tribocorrosion test. The exposed surface
Before the measurements, each sample was ultrasonically cleaned in
in the tribocorrosion cell was 0.12 cm2, while the volume of saliva in the
ethanol and dried. The surface roughness was 0.019 μm for NiTi wire
corrosion cell was 50 mL. The total exposure time during the
and 0.010 μm for stainless steel archwire.
tribocorrosion test was 5 h. The wear was conducted with 1 N force
Each time a new dental alloy was used. The samples (i.e., the work-
and a velocity of 0.33 mm/s, 153 cycles with a total length of
ing electrodes) for the electrochemical tests were cut from the archwire
1530 mm, and a 5-mm wear track. The tribocorrosion experiments
along the straight part to a full length of approximately 30 mm. The ex-
were performed at room temperature.
posed surface of the working electrode had an area of approximately
The total concentration of metal ions in the analysed samples was
1 cm2. The samples for the tribo-electrochemical tests were prepared
determined by using an inductively coupled plasma mass spectrometer
from the flat part (30 mm in length) of the archwire, and were glued
(ICP-MS), model 7700×, from Agilent Technologies (Tokyo, Japan). The
on a carbon plate with an electrical contact attached and the sides and
experimental working conditions for the ICP-MS were optimized for
contact area protected by an inert coating.
plasma robustness. The data was treated with Agilent ChemStation soft-
ware. The data processing was based on the peak area.
2.2. Solutions

The electrochemical measurements were performed in artificial sali- 2.6. AES analysis
va (designated “simulated saliva”) as a solution consisting of 0.60 g/L
NaCl, 0.72 g/L KCl, 0.22 g/L CaCl2·2 H2O, 0.68 g/L KH2PO4, 0.856 g/L Auger electron spectroscopy (AES) analyses were performed on the
Na2HPO4·12 H2O, 0.060 g/L KSCN, 1.5 g/L KHCO3 and 0.03 g/L citric PHI SAM 545 spectrometer. For the electron excitation a primary elec-
acid, pH = 6.5 [31]. tron beam of 3 keV and 1 μA, with a diameter of 40 μm, was used. During
Three different fluoride concentrations were studied, 0.014 M, the depth profiling the samples were sputtered using two symmetrical-
0.024 M and 0.076 M, simulating concentrations found in fluoride-con- ly inclined argon-ion beams of 3 keV over an area of 6 mm × 6 mm. The
taining mouthwashes (up to 0.6 g/L), the most common concentrations etching speed, measured for a multilayer structure of Ni/Cr with known
in corrosion studies (1.0 g/L) and in toothpastes (3.2 g/L). Fluoride was thickness, was about 2.0 nm/min. The atomic concentrations were cal-
added as NaF, p.a. purity, Fluke. culated by means of relative sensitivity factors given by the instrument
In order to study the effect of pH, lactic acid was added to the saliva producer.
in order to achieve pH 2.5 and pH 3.9 and represent inflammatory con-
ditions and eating habits, respectively. 3. Results and discussion

2.3. Electrochemical measurements 3.1. Electrochemical measurements

A three-electrode corrosion cell with a volume of 300 cm3 was used. 3.1.1. Effect of pH
Autolab PGStat 100, Floating version equipment was used, which was Cyclic potentiodynamic measurements were conducted on NiTi and
expanded with Nova software 1.10. stainless steel archwires in order to study the effect of pH in simulated
The electrochemical tests were performed as follows. A 2-hour stabi- saliva at 37 °C. As described in the Experimental section, after the steady
lization at open-circuit potential (OCP), with linear polarization mea- state was achieved, linear polarization measurements at a slow scan
surements at ± 20 mV vs. OCP at a scan rate of 0.1 mV/s. The cyclic rate were made, followed by cyclic potentiodynamic measurements.
polarization measurements were then performed starting from The effect of pH, simulating a normal pH environment (pH 6.5), diges-
−0.25 V vs. OCP, and progressing in the anodic direction up to +1.4 V tion situation (pH 3.9) and extremely low pH for inflammatory condi-
at a scan rate of 1 mV/s. The reverse potential was chosen at a potential tions (pH 2.5) are presented in Fig. 1 and Fig. 2, for the NiTi and SS
of 1.4 V vs. Eref or when a 0.1 mA/cm2 current density was reached. All archwires, respectively.
684 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689

Table 1
Electrochemical parameters of NiTi archwires in simulated saliva at various pH and 37 °C.

NiTi alloy at 37 °C Ecorr/mV jcorr/nA cm−2 Eb/V Erp/mV

Saliva, pH 6.5 −258 17.4 1.07 467

Saliva + lactic acid, pH 3.9 89 49.5 0.7 /
Saliva + lactic acid, pH 2.5 −375 823 0.3 −404

Table 2
Electrochemical parameters of SS archwires in simulated saliva at various pH and 37 °C.

SS at 37 °C Ecorr/mV jcorr/nA cm−2 Eb/V Erp/mV

Saliva, pH 6.5 −219 153 0.65 1.0

Saliva + lactic acid, pH 3.9 9.0 344 0.53 83
Saliva + lactic acid, pH 2.5 −172 108 0.2 16.4

Fig. 1. Cyclic potentiodynamic curves at 37 °C for NiTi alloy dental archwire in simulated
saliva at various pH values, scan rate 1 mV/s.
153 nA/cm2. The curve exhibits a passive region up to 0.65 V vs. SCE. A
sudden increase in the current density occurs and a positive hysteresis
The corrosion current density, jcorr for NiTi at pH 6.5 at 37 °C is is observed, showing the possible presence of a local type of attack for
relatively low at 17.4 nA cm−2 (Table 1). In the forward scan, a pseu- the SS archwires.
do-passive region is observed for the NiTi alloy in simulated saliva. At pH 3.9, the corrosion potential moves in a positive direction and
The pseudo-passive region has an approximate width of 1300 mV. The the corrosion current density increases to 344 nA/cm2. The passive re-
scan was reversed when 0.1 mA/cm2 was reached. The curve exhibits gion is narrower and is limited by the breakdown potential at 0.53 V.
a negative hysteresis in the reverse scan, indicating that NiTi at this A positive hysteresis is observed.
pH and body temperature is not subjected to a local type of corrosion. At the lowest pH of 2.5, the corrosion current density decreases to
The re-passivation potential (Erp = 467 mV) is more positive than the 108 nA/cm2. A pseudo-passive region is observed in the anodic part
Ecorr potential, which affirms the non-possibility of pitting formation. with higher current densities than in the case of the higher pHs. A re-
At pH 3.9, which resembles the conditions when eating and drinking verse scan results in a positive hysteresis, also showing that pitting or
acidic drinks, the corrosion potential is more positive, at 89 mV, and crevice corrosion can occur. The polarization resistance results (not
the corrosion current density increases to 49.5 nA/cm2. The pseudo-pas- shown) confirm that the Rp value at pH 6.5 is 1.71 MΩ/cm2, while at
sive region is narrower, approximately 600 mV. At 0.7 V vs. SCE, an pH 3.9 it is 0.87 MΩ/cm2 and 0.48 MΩ/cm2 at pH 2.5.
intensive breakdown is observed. The reverse scan results in a curve When comparing these results to those of the NiTi archwire, it is ob-
that exhibits positive hysteresis, showing the likelihood that a local served in the presented case that SS archwires have higher corrosion
type of corrosion attack is expected. (See Table 2.) current densities in a neutral environment, have a higher tendency for
At lower pH, 2.5, the corrosion current density is very high, a local type of attack and the pH environment does not have such a
823 nA/cm2. The current in the anodic region rises continuously as the high impact on the corrosion current density as in the case of NiTi
potential increases. The breakdown was observed at 0.3 V, making the archwires. At pH 3.9, NiTi has seven times lower corrosion current den-
passive region even narrower. Positive hysteresis shows smaller current sity than SS, while at pH 2.5, NiTi has eight times higher corrosion cur-
densities in the reverse directions. Also, Ecorr and Erp are very close, sug- rent density when compared to SS.
gesting the possibility of the formation of a localized corrosion attack.
Linear polarization measurements confirm this observation, with 3.1.2. Effect of fluoride
the Rp value being the highest for the NiTi archwire in pH 6.5, The cyclic potentiodynamic curves for the NiTi dental alloy at 37 °C
5.3 MΩ/cm2. At lower pH values, the Rp values decrease significantly, in simulated saliva at various concentrations of fluorides are presented
0.71 and 0.18 MΩ/cm2 at pH 3.9 and 2.5, respectively. This means that in Fig. 3.
the corrosion rate of the NiTi archwires increases by 7.5 times at Different fluoride concentrations were used for the testing in order
pH 3.9 and 30 times at pH 2.5. to simulate the use of fluoride in tooth-hygiene products: toothpaste di-
Cyclic potentiodynamic curves for the SS archwires at various pH are luted with the amount of saliva during 3 min of mouth washing: 0.6 g/L
presented in Fig. 2. The corrosion current density for SS at pH 6.5 is (0.014 M), concentration 1.0 g/L (0.024 M) was used as a most common

Fig. 2. Cyclic potentiodynamic curves at 37 °C for SS dental archwire in simulated saliva at Fig. 3. Cyclic potentiodynamic curves at 37 °C for the NiTi archwire in simulated saliva at
various pH values, scan rate 1 mV/s. various fluoride concentrations, scan rate 1 mV/s.
 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689 685

Table 3 Table 4
Electrochemical properties of NiTi archwires at various concentrations of fluorides. Electrochemical properties of SS archwires at various concentrations of fluorides.

NiTi alloy at 37 °C Ecorr/mV jcorr/nA cm−2 Eb/V Erp/mV SS at 37 °C Ecorr/mV jcorr/nA cm−2 Eb/V Erp/mV

Saliva −258 17.4 1.07 467 Saliva −219 153 0.65 1.0
Saliva + 0.014 M NaF −270 21.5 1.04 318 Saliva + 0.014 M NaF −198 156 0.54 −185
Saliva + 0.024 M NaF −257 19.1 1.01 469 Saliva + 0.024 M NaF −189 67.8 0.59 −176
Saliva + 0.076 M NaF −146 29.6 0.78 −264 Saliva + 0.076 M NaF −167 66.4 0.93 501

concentration in the corrosion studies of NiTi in an oral environment corrosion, as deduced from the positive hysteresis of the cyclic potenti-
[14,33,34] and concentration of 3.2 g/L (0.076 M) was used as a most odynamic curve. A similar observation was reported in the study of
common high concentration of fluoride ion in commercial toothpastes. Heravi et al. [35].
The corrosion potential, corrosion current density and breakdown In the presence of fluoride ions, the corrosion current density in the
potentials are very similar, as also shown in Table 3. The passive range passive region increases for both alloys, while in the case of the highest
in the anodic part of the curve consists of the two regions, region 1, fluoride concentration, like the one found in toothpastes of 0.076 M, the
where the current density slowly increases with the potential up to ap- fluoride concentration has a great effect on the NiTi alloy, narrowing the
proximately 0.6 V vs. SCE and region 2, where the current density in- passive range, and only in this case does it show susceptibility to local
crease is constant and not dependent on the fluoride concentrations. types of corrosion, as observed from the positive hysteresis.
Above 0.6 V, the current densities up to a breakdown potential are At this high fluoride concentration, the SS alloy behaves much better,
very similar and constant with the increase of the potential. The break- the passive range is increased and a positive hysteresis is not observed.
down potential for NiTi in the highest fluoride concentration is lower at Our study revealed that at high fluoride concentrations the SS alloy ex-
0.78 V and the reverse scan detected positive hysteresis. hibits better properties than the NiTi alloy.
All the curves up to the fluoride concentration of 0.024 M show neg- For Ti alloys it was shown that TiO2 is affected when the NaF concen-
ative hysteresis after the reverse of a scan, indicating that no local type tration exceeds 0.1 wt%, i.e., a 0.024 M concentration [36]. It was found
of corrosion attack is expected for NiTi archwires. Fluoride addition had that exceeding a concentration of 0.5 wt% for NaF in solution destroys
no great effect on the NiTi dental archwire corrosion properties in sim- pure titanium and its alloys as well [37]. Nakagawa et al. proved that
ulated saliva at 37 °C and pH 6.5 with the exception of a high fluoride the corrosion resistance of Ti is lost at a 0.1 wt% NaF concentration
concentration of 0.076 M. The polarization resistance values confirm and pH 4.2 or lower [37].
such an observation, for NiTi at pH 6.5 it is 5.32 MΩ/cm2, lower for It is well known that TiO2 oxide films decay following the reaction:
smaller fluoride additions, but it reduces to 1.6 MΩ/cm2 at highest stud- TiO2 + 6 F− + 4H+ ➔ TiF2− 6 + 2 H2O (1),
ied fluoride concentration. where the hexavalent titanium fluoride ion is highly soluble and dis-
The cyclic potentiodynamic curve for the SS archwires at 37 °C in solves away from the electrode [36].
simulated saliva and various concentrations of fluoride concentrations Another reaction pathway was reported by Nakagawa [37], where it
are presented in Fig. 4. The corrosion potential and corrosion current is claimed that the TiO2 oxide film dissolves in the presence of HF:
densities are very similar, as also shown in Table 4. The fluoride addi- Ti2O3 + 6 HF ➔ 2 TiF3 + 3 H2O (2),
tions have different effects to the passive properties as read from the an- TiO2 + 4 HF ➔ TiF4 + 2 H2O (3),
odic part of the curves. At the smaller additions of fluoride 0.014 M and TiO2 + 2 HF ➔ TiOF2 + H2O (4).
0.024 M, and for the solution without fluorides, the curves show posi- It was claimed by Kassab [38] that low concentrations of fluoride
tive hysteresis in the reverse scan, again showing the similarity with ions (0.02 M and 0.04 M NaF) in the case of NiTi dental archwires result-
the local type of corrosion attack. ed in local types of corrosion, which was not observed for pure Ti. It was
At the highest fluoride concentration the transpassive region is ob- also reported that at high concentrations, a general corrosion prevailed
served, with negative hysteresis and the smallest current densities in [38]. It was also claimed that in the presence of fluorides, pitting corro-
the anodic region. sion was observed for NiTi alloys [39]. The tendency for a local type of
The NiTi archwire dental alloy exhibits a very low corrosion current corrosion for the NiTi alloy was observed for the highest concentration
density (17 nA/cm2) in simulated saliva at 37 °C. When compared to the of fluorides (0.076 M). In the case of low fluoride concentrations, the
AISI 304 stainless steel archwire dental alloy, the corrosion current den- stainless steel archwire showed susceptibility to a local type of corro-
sity in this case is higher at 153 nA/cm2. Also, in contrast to the NiTi sion, as observed from the cyclic polarization curves.
alloy, the stainless steel alloy showed that it is prone to crevice
3.2. Tribocorrosion tests and leaching tests

Since dental alloys undergo different types of corrosion, it is very im-

portant to understand the mechanism of the total wear. Wear is the sum
of the mechanical, chemical and wear-induced corrosion effects during
rubbing, fretting etc., of the material in a corrosive environment. Wear
in a corrosive environment has an effect on the physical properties,
the state of the oxide film during and after the wear and the dissolution
of the constituent elements into the corrosive solution, e.g., a dental
archwire in saliva.
A simulation of the use of a wire in dental braces in a day would re-
sult in an approximate rubbing time of 100 min. A load of 1 N and a slid-
ing velocity of 0.33 mm/s on a reciprocating tribometer were chosen in
order to simulate the oral environment during 1 day of use.
A typical tribocorrosion experiment is presented in Fig. 5. The dental
archwire, stainless steel or NiTi, is first immersed in the saliva solution
Fig. 4. Cyclic potentiodynamic curve at 37 °C for the SS archwire in simulated saliva at for a determined time range, after which the rubbing starts. The current
various fluoride concentrations, scan rate 1 mV/s. density is measured, and is increased due to the damage of the oxide
686 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689

Fig. 5. Current measurements for NiTi and SS alloys during the tribocorrosion experiment a), image of NiTi b) and stainless steel c) wire after tribocorrosion test.

film. The higher the current, the higher the tendency for wear-assisted 41]. In controlled surfaces and environments, Ni ion release can be
corrosion is suspected. In the case of our experiment, the current fluctu- slow [41], but as we have shown in our study, wear of the NiTi in the tri-
ations for NiTi were higher than for SS, suggesting the possibility that bological contact releases very large amounts of Ni, not only in NiTi den-
the damage of the oxide film was more intensive and thus the ion tal alloys but also in stainless steel alloys.
leaching could be increased. At the end of the experiment, the current
density decreased again. After 5 h and re-establishing the steady state, 3.3. Passive film analysis
a sample for metal-ion release was taken.
The concentrations of the leached constituent elements were mea- An AES surface analysis was conducted in order to study the change
sured. The results of the exposure of the archwires to simulated saliva in the structure of the surface oxides on NiTi and SS dental archwires
at 37 °C for 3 weeks and after the tribocorrosion experiment, simulating when exposed to saliva during the tribological treatment. The AES
wear during 1 day of use of an orthodontic appliance, are presented in depth profiles were studied on the dental archwires outside and in the
Fig. 6. wear track.
During 21 days exposure of the AISI 304 and NiTi dental wires, the The thickness of passive films was obtained at the point where the
limits of 0.5 μg/cm2/week [40] for Ni ion release were not exceeded. value of oxygen concentration in AES depth profile drops on half of
During the first days of exposure, the concentration of Ni in the NiTi the initial value. The AES depth profiling for stainless steel archwires ex-
archwire was 0.48 μg/cm2/week, close to the limit of 0.5 μg/cm2/week, posed to saliva showed that the passive film on the archwire is 12 nm
but was not exceeded, as is clear from the results presented in Fig. 6. thick, while the thickness of the passive film on the wear track after
However, a significant increase in the ion release was detected after the tribological contact in the saliva is thicker, around 16 nm (Fig. 7).
1.7 h of sliding of an Al2O3 ball against the dental archwire, simulating The presence of Ca and C as constituents of the saliva was detected in
the wear process in the mouth. The limit for Ni release for the dental outermost parts of the oxide film in both cases.
wire after 1.7 h of sliding is 65 times too high for AISI 304 (concentration The highest Cr/Fe ratio was measured at a depth of 16 nm for the
of released Ni is 32.6 μg/cm2/week) and 134 times too high for NiTi stainless steel surface and 20 nm for the stainless steel in the wear
(concentration of released Ni is 69.4 μg/cm2/week) when taking the track. Ni reached an equilibrium at 16 and 20 nm for the stainless
value 0.5 μg/cm2/week into account, a value defined by the EC for Ni- steel at the surface and in the wear track, respectively, in relation to
containing products when in contact with the skin. Higher doses are the total thickness of the passive film.
allowed for biomaterials in contact with body fluids. The ratio of Cr/Fe was higher for the stainless steel wire on the sur-
The reported high biocompatibility of the NiTi alloy appears to be a face outside the wear track (Fig. 8), indicating that the newly grown
direct consequence of the low Ni ion release rate due to the buildup of oxide film in the wear track, although thicker, has a slightly lower con-
a TiO2 protective layer, which was shown by numerous studies [36,37, tent of Cr2O3.

Fig. 6. Concentrations of the measured ions, leached from the NiTi and SS archwires, exposed to simulated saliva at 37 °C for 3 weeks a) and c) and after tribocorrosion experiment b) and d)
(the dashed line represents the Ni limit according to [40]).
 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689 687

Fig. 7. AES depth profile on SS dental archwire exposed to artificial saliva outside a) and in the wear track b).

The depth profiles are shown in atomic concentration vs. depth. The
oxide layer thickness was estimated from the depth at which the oxy-
gen signal is half maximum. AES depth profiles of the as-received NiTi
archwire and the one exposed to simulated saliva show that the oxide
film consists of TiO2 and NiO, with TiO2 present at greater depths than
the NiO. The oxide film thickness is estimated to be 2.5 nm for as-re-
ceived and 2.0 nm for the NiTi archwire exposed to simulated saliva.
The AES depth profile on the surface shows similar oxide thickness
with a slightly different composition of the oxide film (Fig. 9a), which
is enriched with TiO2 at the surface of the oxide when compared to
the composition of the oxide film in the wear track, where also C and
H are present as impurities (Fig. 9b). The Ti/Ni ratio was followed before
etching, at 2 and at 4 min of sputtering time for different samples
(Table 5).
It is clear that all the specimens showed similar oxide layer thick-
nesses. Also, the AES data showed that there are no distinctive differ-
ences between the compositions of the oxide layers. The highest ratio
of Ti/Ni was found at the surface for the new NiTi archwire. The ratio
Fig. 8. Relative atomic concentration vs. depth for Cr/Fe ratio in the SS oxide layer.
Ti/Ni was slightly higher in the wear track (with or without fluorides)
The AES depth profiles of the Ni, Ti, O, C, Ca, P signals for the NiTi than on the samples immersed in saliva containing fluorides. With
archwire exposed to saliva, in the wear track after the tribocorrosion ex- these results we can conclude that the tribological contact in the saliva
periment and in the presence of 0.024 M fluorides are shown in Fig. 9. did not affect much the structure of the oxide film. The oxide film in

Fig. 9. AES profiles of NiTi archwire: outside wear track after exposure to simulated saliva a), after exposure to simulated saliva in the wear track b), in simulated saliva containing fluoride
ions outside the wear track c) and in simulated saliva containing fluoride ions in the wear track d).
688 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689

Table 5 References
Elemental ratio Ti/Ni from AES data on NiTi archwire at different exposure conditions.
[1] V. Katić, H. Otomačić Ćurković, D. Semenski, G. Baršić, K. Marušić, S. Špalj, Influence
Exposure condition Thickness Ti/Ni Ti/Ni Ti/Ni of surface layer on mechanical and corrosion properties of nickel-titanium ortho-
(0 min) (2 min) (4 min) dontic wires, Angle Orthod. 84 (6) (2014) 1041–1048.
[2] D. Mareci, K. Erar, I. Zetu, G. Bolat, C. Crimu, B. Istrate, C. Munteanu, M.N. Matei,
New - as received 2.5 nm 2.61 0.54 0.53
Comparative electrochemical behaviour of uncoated and coated NiTi for dental or-
Saliva - surface 2.0 nm 2.03 0.53 0.56
thodontic wires, Materiale Plastice 52 (2) (2015) 150–153.
Saliva - wear track 2.0 nm 1.74 0.51 0.55 [3] M. Mirjalili, M. Momeni, N. Ebrahimi, M.H. Moayed, Comparative study on corrosion
Saliva with fluoride - surface b2 nm 1.29 0.52 0.54 behaviour of Nitinol and stainless orthodontic wires in simulated saliva solution in
Saliva with fluoride - wear track N2 nm 2.49 1.51 1.49 presence of fluoride ions, Mater. Sci. Eng. C 33 (2013) 2084–2093.
[4] M. Pakshir, T. Bagheri, M.R. Kazemi, In vitro evaluation of the electrochemical behav-
iour of stainless steel and Ni-Ti orthodontic archwires at different temperatures, Eur.
J. Orthod. 35 (2013) 407–413.
[5] G. Rondelli, B. Vicentini, Localized corrosion behaviour in simulated human body
the NiTi alloys might be recovered by the preferential growth of TiO2 fluids of commercial Ni-Ti orthodontic wires, Biomaterials 20 (1999) 785–792.
due to its preferential thermodynamic stability over NiO, as already re- [6] G. Rondelli, Corrosion resistance tests on NiTi shape memory alloy, Biomaterials 17
ported in the literature (free energy of formation ΔG0 for TiO2– (1996) 2003–2008.
[7] N. Schiff, B. Grosgogeat, M. Lissac, F. Dalard, Influence of fluoridated mouthwashes
956 kJ/mol and for NiO −241 kJ/mol) [25]. However, we have shown on corrosion resistance of orthodontics wires, Biomaterials 25 (2004) 4535–4542.
that the wear process in saliva led to extensive Ni release in both the [8] A. Coda, S. Zilio, D. Norwich, F. Sczerzenie, Characterization of inclusions in VIM/VAR
NiTi and stainless steel dental wires. NiTi alloys, J. Mater. Eng. Perform. (2012) 2572–2577.
[9] J. Daems, J.P. Celis, G. Willems, Morphological characterization of as-received and in
vivo orthodontic stainless steel archwires, Eur. J. Orthod. (2009) 260–265.
4. Conclusions [10] F.J. Gil, E. Espinar, J.M. Llamas, J.M. Manero, M.P. Ginebra, Variation of the
superelastic properties and nickel release from original and reused NiTi orthodontic
archwires, J. Mech. Behav. Biomed. Mater. 6 (2012) 113–119.
Electrochemical (including tribocorrosion) experiments, a metal-ion
[11] B. Grosgogeat, E. Jablonska, J.M. Vernet, N. Jaffrezic, M. Lissac, L. Ponsonnet, Tribolog-
release study and AES surface analyses were performed on two different ical response of sterilized and un-sterilized orthodontic wires, Mater. Sci. Eng. C 26
dental alloys: NiTi and stainless steel. The aim was to study the effect of (2006) 267–272.
different environments on the state of the passive film and its structural [12] B. James, J. Foulds, L. Eiselstein, Failure analysis of NiTi wires used in medical appli-
cations, JFAPBC 5 (2005) 82–87.
and functional properties. [13] Y.S. Al Jabbari, Physico-mechanical properties and prosthodontic applications of Co-
It was found that the Cr dental alloys: a review of the literature, J Adv Prosthodont 6 (2014) 138–145.
[14] C. Zhang, S. Zhao, X. Sun, D. Sun, X. Sun, Corrosion of laser-welded NiTi shape mem-
• Corrosion susceptibility of NiTi dental archwires in simulated saliva is ory alloy and stainless steel composite wires with a copper interlayer upon expo-
increased with a decrease of the pH. sure to fluoride and mechanical stress, Corros. Sci. 82 (2014) 404–409.
[15] M.V. Alfonso, E. Espinar, J.M. Llamas, E. Ruperez, J.M. Manero, J.M. Barrera, E. Solano,
• Cyclic potentiodynamic scans for the AISI 304 alloy show susceptibil- F.J. Gil, Friction coefficients and wear rates of different orthodontic archwires in ar-
ity to pitting corrosion, indicated by the positive hysteresis, for all the tificial saliva, J. Mater. Sci. Mater. Med. 24 (2013) 1327–1332.
studied pHs. A pH decrease narrows the passive range and the corro- [16] M. Aziz-Kerrzo, K.G. Conroy, A.M. Fenelon, S.T. Farell, C.B. Breslin, Electrochemical
studies on the stability and corrosion resistance of titanium-based implant mate-
sion current densities do not vary as much as in the case of the NiTi rials, Biomaterials 22 (2001) 1531–1539.
alloy. [17] H.H. Huang, T.H. Lee, Electrochemical impedance spectroscopy study of Ti-6Al-4V
• The increase in the fluoride content in the simulated saliva does not alloy in artificial saliva with fluoride and/or bovine albumin, Dent. Mater. 21
(2005) 749–755.
affect the NiTi to a great extent; however, the highest studied concen-
[18] C.T. Kao, T.H. Huang, Variations in surface characteristics and corrosion behaviour of
tration showed that NiTi is susceptible to a local type of corrosion, as metal brackets and wires in different electrolyte solutions, Eur. J. Orthod. 32 (2010)
demonstrated by the positive hysteresis in the cyclic potentiodynamic 555–560.
[19] T.P. Chaturvedi, R.S. Dubey, Corrosion behaviour of titanium wires: an in vitro study,
curve.
Indian J. Dent. Res. 23–4 (2012) 479–483.
• An increasing fluoride concentration does not affect the properties of [20] C. Abalos, A. Paul, A. Mendoza, E. Solano, C. Palazon, F.J. Gil, Influence of soft drinks
the stainless steel AISI 304 alloy. It was shown that at smaller concen- with low pH on different Ni-Ti orthodontic archwire surface patterns, J. Mater. Eng.
trations the stainless steel dental archwire was susceptible to a local Perform. 22 (3) (2013) 759–766.
[21] I. Milošev, B. Kapun, The corrosion resistance of Nitinol alloy in simulated physiolog-
type of corrosion, while at the highest studied concentration of ical solutions part 1: the effect of surface preparation, Mater. Sci. Eng. C 32 (2012)
0.076 M fluorides, the passive region is established and no tendency 1087–1096.
for a local type of corrosion was observed. [22] I. Milošev, B. Kapun, The corrosion resistance of Nitinol alloy in simulated physiolog-
ical solutions part 2: the effect of surface treatment, Mater. Sci. Eng. C 32 (2012)
• Similar to previous findings in the literature, our study on the ion re- 1068–1077.
lease of stainless steel and NiTi dental archwires in simulated saliva [23] N. Shevchenko, M.-T. Pham, M.F. Maitz, Studies of surface modified NiTi alloy, Appl.
at 37 °C showed that Ni ion release from the dental alloys was Surf. Sci. 235 (2004) 126–131.
[24] L. Tan, W.C. Crone, Effects of methane plasma ion implantation on the microstruc-
below the limit of 0.5 μg/cm2/week. It was found that the concentra- ture and wear resistance of NiTi shape memory alloys, Thin Solid Films 472
tions of the leached ions (Ni and Ti in NiTi alloy and Ni, Cr, Mn, Fe in (2005) 282–290.
stainless steel) decreased with immersion time. [25] S. Trigwell, R.D. Hayden, K.F. Nelson, G. Selvaduray, Effects of surface treatment on
the surface chemistry of NiTi alloy for biomedical applications, Surf. Interface Anal.
• Our experiments in the presented study included ion leaching during
26 (1998) 483–489.
exposure to saliva with the simultaneous effect of mechanical load, [26] C. Yang, F. Chen, S. Chen, Anodization of the dental arch wires, Mater. Chem. Phys.
which simulated one day of wear when the archwires are used during 100 (2006) 268–274.
orthodontic treatment. It was shown that the values of Ni ion release [27] T. Zhao, Y. Li, Y. Xiang, X. Zhao, T. Zhang, Surface characteristics, nano-indentation
and corrosion behavior of Nb implanted NiTi alloy, Surf. Coat. Technol. 205 (2011)
exceeded the limit of 0.5 μg/cm2/week by a large factor; the release 4404–4410.
was 134 times higher than that allowed for NiTi dental archwire and [28] T. Kosec, P. Močnik, A. Legat, The tribocorrosion behaviour of NiTi alloy, Appl. Surf.
65 times higher than that allowed for stainless steel archwire. Sci. 288 (2014) 727–735.
[29] C. Zhang, Z.N. Farhat, Sliding wear of superelastic TiNi alloy, Wear 267 (2009)
• AES analyses showed that no major change in the composition of the 394–400.
surface oxide film on NiTi or stainless steel alloy AISI 304 was detected [30] W. Tillmann, S. Momeni, Influence of in-situ and postannealing technique on tribo-
after the tribological contact in simulated saliva. logical performance of NiTi SMA thin films, Surf. Coat. Technol. 276 (2015) 286–295.
[31] G.S. Duffo, E. Quezada Castillo, Development of an artificial saliva solution for study-
ing the corrosion behaviour of dental alloys, Corrosion 60–6 (2004) 594–600.
Acknowledgements [32] S.W. Tait, An Introduction to Electrochemical Corrosion Testing for Practicing Engi-
neers and Scientists, PairODocs Publications, Racine (WI, USA), 1994.
[33] T.H. Lee, T.K. Huang, S.Y. Lin, L.K. Chen, M.Y. Chou, H.H. Huang, Corrosion resistance
The work described in this paper was sponsored by the Slovenian of different nickel-titanium archwires in acidic fluoride-containing artificial saliva,
Research Agency, Grant No L3-5501 and PhD grant for P. Močnik. Angle Orthod. 80–3 (2010) 547–553.
 P. Močnik et al. / Materials Science and Engineering C 78 (2017) 682–689 689

[34] N. Schiff, B. Grosgogeat, M. Lissac, F. Dalard, Influence of fluoride content and pH on Janez Kovač (PhD in Electron Vacuum Technologies, Se-
the resistance of titanium and its alloys, Biomaterials 23 (2002) 1995–2002. nior Researcher at Jožef Stefan Institute) is a physicist
[35] F. Heravi, M.H. Moayed, N. Mokhber, Effect of fluoride on nickel-titanium and stain- and has been involved in characterization of materials for
less steel orthodontic archwires: an in-vitro study, J. Dent. 12 (2015) 49–59. 25 years at the Laboratory for surface and thin film analysis.
[36] H.H. Huang, Effect of fluoride and albumin concentration on the corrosion behaviour His research is devoted to reactions at surfaces, interfaces
of Ti-6Al-4V alloy, Biomaterials 24 (2003) 175–282. and in thin films by mass spectrometry and electron spec-
[37] M. Nakagawa, S. Matsuya, T. Shiraishi, M. Ohta, Effect of fluoride concentration and troscopy, plasma science, vacuum science and technology,
pH on corrosion behavior of titanium for dental use, J. Dent. Res. 78 (1999) development of surface and thin film analytical techniques
1568–1572. like ion sputter depth profiling. He specializes in XPS, AES
[38] J.E. Kassab, J.P. Gomes, Assessment of nickel titanium and beta titanium corrosion and ToF-SIMS spectroscopies.
resistance behaviour in fluoride and chloride environments, Angle Orthod. 83–5
(2013) 864–869.
[39] I. Ünal, S. Zor, H. Atapek, Influence of artificial saliva on NiTi orthodontic wires: a
study on the surface characterization, Mater. Sci. 47–6 (2012) 830–837.
[40] Directive EU, 2004/96/EC: Amending Council Directive 76/769/EEC as Regards
Prof. Milan Bizjak (PhD in Metalurgy, Full Professor) is a
Restrictions on the Marketing and Use of Nickel for Piercing Post Assemblies for
metalurgist and has been involved in synthesis and charac-
the Purpose of Adapting its Annex I to Technical Progress.
terization of materials for more than 30 years. His research
[41] M. Es-Souni, M. Es-Souni, H. Fischer-Brandies, Assessing the biocompatibility od NiTi
is mainly focused on the areas of powder metallurgy, rapid
shape memory alloys used for medical applications, Anal. Bioanal. Chem. 381 (2005)
solidification, composites, characterization and testing of
557–567.
materials and design of advanced materials. He specializes
in the design and manufacture of alloys, characterization of
Petra Močnik (BsC in Chemistry, PhD Student at Universi- the microstructure and link microstructure and properties.
ty of Ljubljana) is a chemist and has been involved in elec- He is involved in industrial projects, international collabora-
trochemical studies for 5 years. Currently she is engaged in tion in the frame of more than 15 international projects and
tribocorrosion processes. Her research work involves elec- networks, he is a member of a scientific program “Synthesis
trochemical and spectroscopic methods for the characteriza- and characterisation of materials” and holds national re-
tion of corrosion. She is preparing a doctoral thesis with the search projects.
title “Characterization of tribocorrosion processes on dental
alloys in different environments”.

Tadeja Kosec (PhD in Chemistry, Senior Research associ-

ate at Slovenian National Building and Civil Engineering
institute) is a chemist and has been involved in electrochem-
ical characterization of materials for 15 years. Her research is
devoted to electrochemical characterization of different mate-
rials, with special emphasis on copper alloys, nickel alloys,
steels and stainless steels. She specializes in passivation, corro-
sion, corrosion protection and inhibition. She is involved in
industrial projects, international collaboration on spent nuclear
fuel program and national projects Characterization of
Tribocorrosion processes and Tribocorrsoion procceses-from
theory to practice.