Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241

Contents lists available at ScienceDirect

Journal of the Mechanical Behavior of

Biomedical Materials
journal homepage: www.elsevier.com/locate/jmbbm

Force delivery of NiTi orthodontic arch wire at different magnitude of MARK

deflections and temperatures: A finite element study
⁎
M.F. Razalia, A.S. Mahmuda, , N. Mokhtarb
a
Nanofabrication and Functional Materials Research Group, School of Mechanical Engineering, Engineering Campus, Universiti Sains Malaysia, 14300 Penang, Malaysia
b
Craniofacial and Biomaterial Science Cluster, Advanced Medical and Dental Institute, Universiti Sains Malaysia, Bandar Putra Bertam, 13200 Kepala Batas, Penang,
Malaysia

A R T I C L E I N F O A B S T R A C T

Keywords: NiTi arch wires are used widely in orthodontic treatment due to its superelastic and biocompatibility properties.
Bending In brackets configuration, the force released from the arch wire is influenced by the sliding resistances developed
Leveling on the arch wire-bracket contact. This study investigated the evolution of the forces released by a rectangular
Binding NiTi arch wire towards possible intraoral temperature and deflection changes. A three dimensional finite ele-
Bracket friction
ment model was developed to measure the force-deflection behavior of superelastic arch wire. Finite element
Shape memory alloy
analysis was used to distinguish the martensite fraction and phase state of arch wire microstructure in relation to
the magnitude of wire deflection. The predicted tensile and bending results from the numerical model showed a
good agreement with the experimental results. As contact developed between the wire and bracket, binding
influenced the force-deflection curve by changing the martensitic transformation plateau into a slope. The arch
wire recovered from greater magnitude of deflection released lower force than one recovered from smaller
deflection. In contrast, it was observed that the plateau slope increased from 0.66 N/mm to 1.1 N/mm when the
temperature was increased from 26 °C to 46 °C.

1. Introduction properties of superelasticity, which permits recovery of very large strain

Orthodontic treatments involve the movement of tooth in various
directions and orientations at different stages. These stages are divided
into alignment and leveling, correction of molar relationship and space
closure, and finishing treatment. Movement of tooth is induced by force
generated from a pre-deformed wire hooked to the bracket on the tooth.
The design of bracket is supposed to allow the arch wire to slide at
negligible friction, thus tooth movement is achieved only by the
bending force generated by the pre-deformed wire. Most brackets are
made of stainless steel and arch wire is secured onto it by using ligation
ties. For alignment and leveling stage, orthodontist starts with round
light wire that can be bent easily and subsequently replaces it with a
thicker wire with rectangular shape (Kim et al., 2008). This sequence of
wires usage is important to gain light forces throughout the orthodontic
treatment, as tooth moves effectively at force below 1.0 N (Proffit et al.,
2014).
For decades, the use of superelastic nickel-titanium (NiTi) wire is
preferred against stainless steel wire during the early stage of ortho-
dontic treatment, with a goal of bringing the teeth into alignment
(Duerig et al., 2013). NiTi arch wires are chosen due to its unique
of approximately 6–8% without permanent deformation (Otsuka and
Wayman, 1999). This superelastic behavior is found to be suitable for
aligning and leveling treatment, in which it provides a constant mag-
nitude of force for effective tooth movement. As the arch wire slides to
recover its original straight shape, it pulls the teeth to a better position
(Singh, 2015). It is well documented that superelastic NiTi material
yields this large strain over a stress plateau by manifesting its stress-
induced martensitic (SIM) transformation (Jiang et al., 2009).
Even though NiTi arch wires are widely used, its real performance in
bracket configuration is still unclear since most of the force-deflection
studies focused on three point bending test (Kaphoor and
Sundareswaran, 2012; Varela et al., 2014). The force-deflection beha-
vior deduced from three point bending test is not directly transferable
to the oral clinical setting due to the absence of sliding resistance
consideration during the test. In order to obtain realistic clinical force,
Nucera et al. (2014) have included bracket engagement in their bending
test to consider the influence of sliding resistance towards the bending
behavior of NiTi arch wire. They found that the force released by the
0.36 mm round NiTi arch wire during recovery increased linearly from
3 mm to 1 mm, which completely defies the believe of constant and

⁎
Corresponding author.
E-mail address: abdus@usm.my (A.S. Mahmud).

http://dx.doi.org/10.1016/j.jmbbm.2017.09.021
Received 10 June 2016; Received in revised form 20 May 2017; Accepted 14 September 2017
Available online 18 September 2017
1751-6161/ © 2017 Elsevier Ltd. All rights reserved.
M.F. Razali et al.
continuous force delivery of NiTi arch wire in orthodontic treatment.
The sliding resistance in arch wire-bracket configuration was cate-
gorized by Kusy and Whitley (1999) into two components, classic
friction (FL) and binding (FBI). Classic friction refers to sliding resistance
created by elastomer ligatures when it drives the arch wire against the
base of the bracket slot. Meanwhile, binding is developed when the arch
wire is bent, with the magnitude of friction increases as the curvature of
the bend increases. Since arch wire slides through the slot of the ad-
jacent brackets during leveling treatment, a portion of the released
force (FR) from the NiTi arch wire is used up to overcome the generated
binding and classical friction at the wire-bracket interface (Baccetti
et al., 2009).
Researchers have developed various constitutive models of shape
memory alloy using different state variables to describe the phase
transition characteristics such as, reorientation of martensite, marten-
site volume fraction and energy dissipation. A few macro scales con-
stitutive models of shape memory alloy have been proposed in litera-
ture but feature only the superelastic (Auricchio and Taylor, 1997) or
both shape memory effect and superelastic behavior (Boyd and
Lagoudas, 1996). The superelasticity model by Auricchio and Taylor
(1997) is employed for the generation of UMAT/Nitinol subroutine
available in Abaqus FEA. In recent years, a few studies have utilized this
built-in subroutine and successfully predicted the true behavior of
shape memory alloy under its operating conditions (Pelton et al., 2013;
Robertson et al., 2015; Weafer et al., 2016).
In this study, a three dimensional numerical modeling of wire-
bracket system was developed to anticipate the bending behavior of
superelastic NiTi arch wire. The model adopted the concept introduced
by Auricchio and Taylor (1997) and a nonlinear finite element proce-
dure was incorporated into it. The objective of this work is to evaluate
the influence of binding friction towards force-deflection characteristics
of NiTi arch wire with respect to various magnitudes of deflection and
oral temperatures. The evolution of normal stress and martensite frac-
tion in the finite element model is subsequently analyzed to understand
the utilization of superelastic behavior during bending. This new
knowledge may assist orthodontist to quantify the true mechanical re-
sponse of NiTi arch wire in brackets configuration whilst improving
current installation practices.
2. Materials and methods
2.1. Experimental testing
In this study, three types of experimental analyses were carried out;
uniaxial tensile, modified three point bending and calorimetric test. The
uniaxial tensile test was conducted to quantify the mechanical beha-
viors parameters of NiTi wire for defining the superelastic behavior in
the material subroutine. Modified three point bending test was exe-
cuted by considering three brackets engagement during the test. The
result of bending test was used to validate the force-deflection curve
generated from the numerical model. The calorimetric test was done to
measure the transformation temperatures of the arch wire, in specific
the austenite finish (Af) temperature. The specimen used for all analyses
was superelastic NiTi arch wire with a rectangular cross section of
0.4 mm × 0.56 mm.
20 mm uniaxial specimens were cut from the straight end section of
the arch wire. The specimens were elongated to 1.7 mm at a displace-
ment rate of 1 mm/min in the direction of its length before unloaded at
similar rate. This test was conducted on an Instron model 3367 uni-
versal testing machine. The uniaxial test was carried out at three dif-
ferent temperatures of 26 °C, 36 °C and 46 °C. A small chamber was
used to control the environmental temperature of the specimen during
tensile test, with an accuracy of ± 1 °C. The desired temperature was
achieved by continuous flow of heated air through inlet and outlet
vents. The testing procedure and settings followed the standard of ISO
15841: Dentistry-Wires for use in Orthodontic. This uniaxial test was
235
Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241
Fig. 1. NiTi arch wire bent on modified three point bending setup.
repeated for three times with a new specimen used for each test. The
average value measured from the stress-strain curves at 26 °C was used
for the determination of material properties.
The modified three point bending test was conducted using the
same universal testing machine with a customized compressive loading
jig, as shown in Fig. 1. The lower load-cell capacity of 500 N was in-
stalled for this test to increase the sensitivity of load reading. The
clinical scale of three brackets apparatus was set up by setting the
distance between the brackets to be 7.5 mm, thus resembling the
average distance of adjacent teeth (Nucera et al., 2014). Three brackets
with 0.56 slot height and 2.80 mm slot width were used with respect to
its zero torque and angulation design. The brackets were glued to the
movable indenter and mounting base using cyanoacrylate adhesives.
No ligation tie was installed during the placement of wire inside the
bracket slot. Next, the specimen was deflected to 4 mm at 26 °C by
setting the indenter to move vertically downward and upward at a
displacement rate of 1 mm/min. This bending test was repeated for two
times, and the validation was done by comparing the experimental and
numerical result directly. In this study, the coupling of 0.4 mm ×
0.56 mm wire with the 0.56 mm-slot bracket reflects the common wire-
bracket combination used during the leveling treatment (Lombardo
et al., 2012). In fact, this 0.16 mm clearance is sufficient to permit
sliding of the arch wire along the bracket slot (Proffit et al., 2014).
The phase transformation temperatures of the specimens were de-
termined using TA-Q20 differential scanning calorimeter. The specimen
size was 7.86 mg and the heating and cooling rate was 10 °C/min. The
transformation temperatures were determined by the intersection of
tangent lines on the peaks as described in ISO 15841.
2.2. Constitutive model of shape memory alloy
To simulate the superelasticity of the NiTi arch wire, a built-in user
material subroutine (UMAT/Nitinol) in Abaqus 6.12.2 was employed.
This subroutine was established by Auricchio and Taylor (1997), based
on a generalized plasticity theory that decomposes strain components
into purely elastic and transformational strains. The change in linear
elasticity from the parent austenite phase to the stress-induced mar-
tensite phase was achieved by adopting a rule of mixtures. The change
in the stress level to induce phase transformation shifts linearly with
temperature. This model was chosen over other viable macro scales
constitutive models for its good agreement with experimental result
involving bending type deformation (Nematzadeh and Sadrnezhaad,
2012).
The material data used in the subroutine are tabulated in Table 1.
These data are generated from the uniaxial test of equiatomic NiTi,
deformed at room temperature.
M.F. Razali et al.
Table 1
User material data for constitutive model of superelastic NiTi arch wire (T = 26 °C).
Parameter Description
EA Austenite elasticity
(νA) Austenite Poisson's ratio
EM Martensite elasticity
(νM) Martensite Poisson's ratio
(εL) Transformation strain
(δσ/δT)L Loading
σSL Start of transformation loading
σEL End of transformation loading
T0 Reference temperature
(δσ/δT)U Unloading
σSU Start of transformation unloading
σEL End of transformation loading
σSCL Start of transformation stress in compression
2.3. Finite element simulation
In this study, the generated outputs from the numerical model were
compared against the experimental results. The comparisons were
executed to validate the numerical model capability in predicting the
deformation behavior of superelastic NiTi arch wire under tensile and
bending loads.
2.3.1. Tensile test
The 20 mm length uniaxial specimen was modeled by using a total
of 7884 linear hexahedral elements (C3D8). The load was applied at
one end by employing a displacement boundary condition while the
other end was set to be fixed. The testing parameters and conditions
were replicated from Section 2.1 to preserve the testing integrity.
Throughout the test, the total stress reacted on the direction of loading
was recorded at the fixed end of the specimen.
2.3.2. Bending test
The 30 mm length bending specimen was modeled by using a total
of 71,144 linear hexahedral elements with reduced integration
(C3D8R). The hour glassing effect of this reduced integration second
order eight-node element was controlled by the selection of stiffness
hourglass behavior (ABAQUS manual). The arch wire specimen was
assumed to be mechanically isotropic. A convergence analysis of force-
deflection response was performed to select an appropriate elements
size for the model. The bending model was considered to be converged
when the deviation of force between the numerical and experimental
results was less than 10%.
The bracket was designed into two parts, bottom and top part by
using bilinear rigid quadrilateral element (R3D4). Both parts have a
width of 2.80 mm and were separated by a distance of 0.56 mm, re-
presenting the real bracket slot dimension. The arch wire was as-
sembled into the brackets slot with no boundary condition defined to let
the wire to slide freely during the bending. Contact interaction between
arch wire and bracket was set as surface-to-surface contact with a
friction coefficient of 0.27. It has been reported in the literature that the
coefficient of friction between NiTi arch wire and stainless steel bracket
surface to be in the range of 0.23–0.30 (Whitley and Kusy, 2007).
As shown in Fig. 2, each bracket was assigned to its own reference
point (RP). Only the central bracket was set to move in y-direction at a
displacement rate of 1 mm/min, while the adjacent brackets were
constrained to be fixed in all directions (Ux = Uy = Uz = 0). The wire
was set to be deflected to 1, 2, 3 and 4 mm at three different tem-
peratures of 26 °C, 36 °C and 46 °C, respectively. The three tempera-
tures used in the present study portray the deformation behavior of the
wire for different cases; behavior at the time of installation (26 °C),
behavior inside the oral environment (36 °C) and behavior during ex-
posure to warm food intake (46 °C). The deflection ranges were chosen
based on norm clinical conditions encountered during the orthodontic
Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241
treatments (Parvizi and Rock, 2003). For comparison, the forces exerted
by the bent arch wire to the central bracket were collected from the
assigned reference point.
Value (unit)
44 (GPa) 3. Results
0.33
23 (GPa)
3.1. Thermal and mechanical deformation behavior
0.33
0.055
6.7 (MPa/°C) Fig. 3 displays the thermal martensite phase transformation beha-
377 (MPa) vior of the arch wire specimen. The transformation on cooling occurred
430 (MPa) in two stages preceded by the R-phase, thus shifted the martensite
26 (°C)
transformation temperature to further lower. The reverse transforma-
6.7 (MPa/°C)
200 (MPa) tion on heating occurred in a single stage of martensite to austenite
140 (MPa) with heat enthalpy of, Δh = 14.72 J/g. The transformation tempera-
452 (MPa) tures indicating the start and finish of the martensite phase transfor-
mation were measured at the intersection of tangent lines, as indicated
in the figure. For this specimen the austenite finish, Af is 16.13 °C and
martensite start, Ms is − 50.22 °C. This implies that at room tempera-
ture (~ 26 °C), the structure is at austenite state, thus tensile and
bending deformations at this temperature occur under superelastic
behavior.
Fig. 4 compares the stress-strain curve generated from the numerical
against the experimental uniaxial test conducted at three different
temperatures. For clarify purpose, only the most identical curve was
plotted in each plot for direct comparison with the numerical result.
The experimental data is indicated by the solid line and the numerical
data is indicated by the dashed line. The symbols labelled on Fig. 4a
characterized the material data used in Table 1. For all three test
temperatures, the numerical agreed well with the experimental. The
specimen exhibited good superelastic behavior as expressed by the
constant stress plateaus of about 6% strain, on both loading and un-
loading. This good agreement between numerical and experimental
results, in terms of loading and unloading plateau, transformation,
elastic and residual strains provided a good confidence for extended
modeling of deformation behavior of the alloy at different boundaries
and parameters conditions. Fig. 4d summarizes the temperature de-
pendent of stress-induced martensite transformation against tempera-
ture. The slope of stress increase against temperature was recorded to
be 6.7 MPa/°C, agreed well with the Classius-Clayperon relation
(Miyazaki and Otsuka, 1989). It was also acknowledged that a small
variation was registered in the value of critical stress for transforma-
tion, as indicated by the error bars inserted in the curves.
Fig. 5 displays the force-deflection plot of NiTi specimen upon
4.0 mm bending on three brackets system at 26 °C. The activation cycle
began with a short linear slope up to 0.3 mm of deflection, representing
the bending stiffness of the arch wire just before making a contact with
the top lateral edges of the bracket. Once this contact was established,
the wire became more constrained within the bracket slot, thus yielded
a steeper slope before reaching the plateau at about 1.1 mm of deflec-
tion length. The deactivation cycle started with a sudden force drop,
implying the onset of martensite transformation with regards to the
stress relief. Gradual force increment was observed throughout the wire
recovery from 3 mm to 1 mm. This is believed due to the weakening of
binding intensity as wire deflection reduces. The force-deflection curve
from the simulation, as indicated by the dashed line is in good agree-
ment with the experimental data. This provides a high level of con-
fidence on the numerical bending model for prediction of the force
performance of superelastic NiTi arch wire bent in a brackets config-
uration. Additionally, good consistency was established between the
experimental and the numerical curves, as indicated by the small dis-
crepancy of force magnitude (0.2 N) at 2.5 mm deflection.
The predicted force-deflection curve of NiTi specimen bent at dif-
ferent deflection magnitudes, and temperatures are shown in Fig. 6. The
highest magnitudes of activation and deactivation forces were recorded
at 4 mm and 1 mm, respectively. For a small deflection such as 1 mm,
the arch wire was deformed and recovered elastically. A short force
236
M.F. Razali et al. Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241

Fig. 2. Finite element model for the modified three point bending test with finer mesh near the contact interaction.

Fig. 3. Thermal transformation behavior of NiTi arch wire.

Fig. 5. Force-deflection curve of NiTi arch wire in a modified three point bending setup.

Fig. 4. Stress-strain curves of NiTi arch wire under uniaxial test at: (a) 26 °C, (b) 36 °C, (c) 46 °C and (d) critical stress for stress-induced martensitic phase transformation.

237
M.F. Razali et al.
Fig. 6. Numerical result of force-deflection behavior of NiTi arch wire bent at different deflection magnitudes and temperatures of: (a) 26 °C, (b) 36 °C, (c) 46 °C and (d) comparison of
activation and deactivation plateaus for 4 mm deflection group.
Fig. 7. Force gap of NiTi arch wire measured at different deflection magnitudes and
temperatures.
plateau of austenite to martensite transformation is observable only
during the recovery of arch wire from the 2 mm activation. Beyond
2 mm, the activation and deactivation of the wire lead to a positive and
negative gradient plateau, respectively. At a higher temperature, both
gradient plateaus delayed to a higher magnitude.
Fig. 6d compares the slope of the force plateaus upon 4 mm de-
flection at different temperatures. Based on the linearity of the gradient
plateaus in Figs. 6a, 6b and 6c, the slope is selected to be from 2 to
4 mm during the activation and from 3 to 1 mm during the deactiva-
tion, respectively. It is observed that the arch wire shared similar ac-
tivation rate of 1.0 N/mm along the selected deflection interval, re-
gardless of the applied wire temperature. This similar activation rate
corresponds to the identical increase of stress needed for martensitic
transformation and binding magnitude against the 10 °C temperature
elevation. In contrast, the deactivation occurred at dissimilar rates,
where the steepest force plateau of 1.1 N/mm occurred at 46 °C, fol-
lowed by 0.88 N/mm and 0.66 N/mm at 36 °C and 26 °C, respectively.
At 46 °C, the thermomechanical behavior of the wire increased the
238
Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241
stress needed for martensite to austenite transformation to a higher
extent, thus increasing the gap with the deactivation force. The large
gap between these two forces caused the deactivation rate to increase to
assist the wire recovery process.
It is interesting to note that the deactivation force varies with re-
spect to the maximum deflection applied on the arch wire. As shown in
Fig. 6a, the wire deflected to greater magnitude of deflection recorded
lower deactivation force than the one deflected to smaller magnitude.
This discrepancy in force magnitude formed a force gap (ΔF), which
continues for at least 0.5 mm before the two deactivation curves re-
merged. In current practice, orthodontist takes advantage of this force
gap in a reactivation process, by which arch wire is detached from a
bracket and subsequently reattached to improve the deactivation
magnitude.
Fig. 7 summarizes the force gap (ΔF) magnitude recorded at dif-
ferent set of deflections and temperatures. Each deflection group shows
a reduction in the magnitude of force gap as the temperature increases
from 26 °C to 46 °C. The highest reduction is observed on 1 mm group,
with the total difference in the force gap is about 0.73 N. Meanwhile,
the lowest reduction is logged on 3 mm group, with only 0.07 N of force
magnitude difference. This trend shows that the force gap reduces at
higher deflection group, and the change of temperature starts to be less
sensitive to the force gap magnitude.
3.2. Finite element results
Fig. 8 displays the normal stress distribution (S11) of the arch wire
during bending activation in 36 °C environment. Red and blue contour
expressed the elements under tension and compression, respectively.
The peak of compressive stress is much higher than the tensile stress
due to the greater compressive critical stress set for the model. The
M.F. Razali et al.
contour located at the bending curvature near to the bracket corner
characterizes the localized martensitic transformation. The maximum
value of tensile normal stress is observed on three hotspot regions,
specified to the outer elements of the wire that experienced the most
bending deformation.
Fig. 9 summarizes the maximum S11 magnitude measured during
wire bending in 36 °C environment. Along the 4 mm activation, the
maximum S11 was elevated from 455 to 562 MPa. It is interesting to
note that at 3 mm, the maximum S11 (551 MPa) exceeded the end
stress level of stress-induced martensitic transformation (σEL), which is
approximately 480 MPa (obtained from Fig. 6b). This stress discrepancy
accentuated that further wire activation to above 2 mm was accom-
panied by the elastic straining of martensite structure. Notice that the
deactivation from 4 to 2 mm corresponds to the combination of elastic
and martensitic transformation.
Fig. 10 shows the volume fraction of martensite structure (SDV21)
upon 4 mm activation at 36 °C. SDV21 is a nodal output of the sub-
routine, represents the volume fraction of austenite transformed to
martensite structure. The 0.0 and 1.0 magnitude denotes the zero and
fully transformed martensite region with respect to the bending de-
formation of the wire. It is seen that NiTi wire evolves in three stages: a
fully transformed martensite (M) zone located at the outer top portion
of the bent region, a partial transformation zone (A–M) that surrounded
the fully transformed martensite zone, and a single austenite (A) phase
Fig. 9. Maximum normal stress (S11) measured upon activation and deactivation of NiTi
arch wire at 36 °C.
239
Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241
Fig. 8. Normal stress distribution (S11) of NiTi arch
wire upon 4 mm activation at 36 °C.
located at the core and the remaining region of the wire.
4. Discussions
4.1. Bending behavior in bracket configuration
In this study, the bending behavior of NiTi arch wire at different
vertical deflections was simulated by three dimensional numerical
modeling. The binding generated at the arch wire-bracket contact was
considered to capture the influence of this sliding resistance component
towards bending behavior. The bending behavior was evaluated by
observing the force-deflection trend, martensite fraction and maximum
normal stress developed during vertical bracket displacement. The
combination of experimental and numerical visualization provides a
clear insight of how superelastic behavior of NiTi arch wires is deployed
during the orthodontic treatment.
Referring to Fig. 4, the arch wire elongated at temperature above Af
exhibited stress-strain curve in the form of superelastic behavior. The
linear elastic deformation at ~ 1% strain reflects the elastic yield of the
austenite structure. As the stress is further increased, the deformation
exhibited a stress plateau with positive gradient. At this stage, the wire
deformation involved microscopic phase change from austenite struc-
ture into martensite via SIM transformation. Surpassing this plateau,
the elongation proceeds with linear stress elevation, imitates the elastic
yielding of martensite structure. Upon unloading, this stress-induced
martensite variant becomes unstable and begins to reverse back to its
original phase structure, austenite at a lower stress level, thus forming a
stress hysteresis.
Similar deformation trend is expected to happen during bending.
Unfortunately, the activation and deactivation force measured from the
modified three point bending does not exhibit a similar plateau as de-
formed in three point bending (Wilkinson et al., 2002) and uniaxial test.
Instead, the plateau region exhibited positive and negative gradient
slope on the activation and deactivation of the wire, respectively. The
gradient plateau trend has been reported before by Naceur et al. (2014)
who predicted numerically the force-deflection behavior of NiTi arch
wire via bent deformation. They claimed that the gradient force pla-
teaus were formed due to the rise of stress needed for martensitic
transformation with respect to the accumulation of elastic strain en-
ergy.
However, in this present study, we believed that these gradient
force plateaus were triggered by the variation of binding friction
M.F. Razali et al.
Fig. 10. View cut of martensitic volume fraction (SDV21) evolution in arch wire upon 4 mm activation at 36 °C.
generated at the adjacent brackets such inferred by Baccetti et al.
(2009) and Nucera et al. (2014). Both authors did not provide the
binding–deflection data from the adjacent brackets to support their
theory due to the limitation of the experimental capabilities. It is sug-
gested that the activation force needed to bend and slide the arch wire
increases with respect to the rise of binding friction generated at the
wire-bracket contact. At the onset of wire recovery, the wire spring
back is used up to overcome the binding friction built up at the contact
regions, thus delaying the deactivation force. As the wire deflection
reduces further, this force is gradually elevated signifying the weak-
ening of the binding.
The variation of binding intensity over wire deflection at 36 °C is
shown in Fig. 11. This binding-deflection plot presents the sliding re-
sistance encountered by the wire following its continuous contact with
the bottom region of the bracket surfaces. The binding intensity was
obtained by multiplying the total magnitude of contact force (CFTM)
with the defined coefficient of friction. The linear increase of binding
from 0 to 1.1 mm was attributed to the elastic yielding of the austenite
Fig. 11. Prediction of binding friction of wire bent at 36 °C.
240
Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241
structure. Knowing that superelastic material to exert constant force,
the regular increase of binding from 1.1 mm and onwards signifies the
elevation of normal force at the contacting surfaces. This linear trend of
binding elevation over deflection agrees well with the experimental
work conducted by Murayama et al. (2013) and Reznikov et al. (2010).
4.2. Clinical perspectives
Even though bending of NiTi arch wire involves activation and
deactivation cycle, it is only the force measured during the deactivation
quantifies the magnitude of force released to the bracket (Elayyan et al.,
2010). It is important to understand the force–deflection behavior of
NiTi arch wire in bracket configuration so that orthodontist can cor-
relate the magnitude of deactivation force with current wire deflection.
The consensus of NiTi material delivering a constant force may no
longer be true for the case of bracket bending, due to the fact that the
wire delivers force in a gradient manner.
In the present study, the discussion only reflects the bending of
0.4 mm × 0.56 mm NiTi wire with 7.5 mm inter bracket setting. The
preliminary data here provides an early insight on how changes in wire
stiffness (owing to temperature changes) influenced the force magni-
tude of the wire. Besides the changes in the oral temperature, the or-
thodontic wire may also be subjected to bend with a smaller inter
bracket configuration. From an extended simulation work, if the same
wire is to be bent at 1.0 mm with smaller inter bracket distance, per say
7.0 mm, the activation and deactivation force was further increased
approximately by 23%. These force increments were highly related
with the increased in wire stiffness corresponded to the shortening of
wire length for active bending.
The trend of arch wire to release different deactivation force with
respect to the maximum deflection applied could be clinically relevant
for reactivation process. This reactivation process has been used by
orthodontist to elevate the amount of force released by the superelastic
NiTi wire that had previously been activated. For example; in the case
of tooth that has moved by 2 mm from its original 4 mm bracket dis-
placement, reactivating the same wire at 2 mm displacement may
trigger higher deactivation force than the one transmitted before. A
M.F. Razali et al.
data on how much this reactivation process improves the deactivation
force is necessary so that this process can be executed with precaution.
Since binding influences the deactivation forces, more study on the
force gap for different wire-bracket combination at various bending
configuration is a necessity.
During the orthodontic treatment period, it is a common practice for
orthodontist to gradually change the geometry of arch wire from round
to rectangular shape. This procedure is practiced to fill up the slot of the
bracket so that the teeth can move into predetermined position with
respect to the tip and torque prescribed on the bracket geometry. It is
interesting to highlight that for every loading condition executed in this
study, the combination of 0.4 mm × 0.56 mm rectangular wire with a
0.56 mm slot height bracket is ineffective to provide suitable deacti-
vation force (< 1.0 N) for tooth movement. Referring to Fig. 6b, the
rectangular wire may generate as much as 5.0 N force for a 1 mm
bracket displacement. This deactivation force is too high for human
tissue and may cause hyalinization of periodontal ligament, hence
discourages effective tooth movement (Proffit et al., 2014). Therefore,
the current practice of using thick rectangular wire at the end of le-
veling treatment should be revised to comply with the ideal force cri-
teria.
Apart from the light force criteria, effective tooth movement also
requires the force to be delivered in a constant magnitude throughout
the treatment period. On account of faster tooth movement over a
continuous application of force (Weiland, 2003), minimizing the slope
gradient is necessary to achieve shorter treatment duration. For this
reason, reduction on friction coefficient and normal force with which
the contacting surfaces are pressed together is compulsory to reduce the
sliding resistance, and hence its impact on the force-deflection trend.
Thus, due to the direct relation between wire stiffness and binding
(Tageldin et al., 2016), selecting a smaller wire might be helpful to
minimize the magnitude of normal force and to lessen the binding in-
tensity. Meanwhile, applying a special surface treatment on bracket and
arch wire surfaces may be a pleasant preference in reducing the friction
coefficient. Both options may be a decent area to be explored in the
future, to preserve light and constant force behavior of superelastic NiTi
arch wire in the bracket system.
The deactivation force of NiTi wire is highly dependent on tem-
perature. The force increases as the oral temperature increases. This
implies that intake of warm water by a patient leads to the stiffening of
NiTi wire, thus greater force is expected to be delivered to the mal-
occlusion. For this reason, if the deactivation force of a specific wire at
body temperature is estimated to be close to the 1.0 N, intake of warm
drinks should be avoided to evade pain. Additionally, NiTi wire also
shows greater variation in the magnitude of deactivation force in high
temperature environment.
5. Conclusions
The present study investigates the bending behavior of rectangular
NiTi arch wire at different magnitude of deflections and temperatures.
The following conclusions can be deduced from this study:
1. The wire subjected to greater deflection magnitude released lower
magnitude of deactivation force as compared to the one deflected at
smaller deflection, thus creating a force gap. The force gap value
reduced as the wire temperature and activation increased.
2. The variation of binding intensity at bracket contacts influenced the
force-deflection characteristics of NiTi wires by elevating and de-
laying the activation and deactivation force, respectively.
3. The wire exhibited greater activation and deactivation force at
higher bending temperature owing to the enhancement of wire
thermomechanical properties.
4. The finite element analysis demonstrated that the activation of NiTi
wires beyond 2 mm in bracket configuration was accompanied by
martensitic elastic deformation. The martensitic transformation of
241
Journal of the Mechanical Behavior of Biomedical Materials 77 (2018) 234–241
NiTi wire occurred locally, concentrated at the region of highly
deformed structure.
Acknowledgments
The authors are grateful for the financial support provided by
Universiti Sains Malaysia under the grant RUI 1001/PMEKANIK/
814244.
References
Auricchio, F., Taylor, R.L., 1997. Shape-memory alloys: modelling and numerical simu-
lations of the finite-strain superelastic behavior. Comput. Methods Appl. Mech. Eng.
143, 175–194.
Baccetti, T., Franchi, L., Camporesi, M., Defraia, E., Barbato, E., 2009. Forces produced by
different nonconventional bracket or ligature systems during alignment of apically
displaced teeth. Angle Orthod. 79 (3), 533–539.
Boyd, J.G., Lagoudas, D.C., 1996. A thermodynamical constitutive model for shape
memory materials. Part I. The monolithic shape memory alloy. Int. J. Plast. 12,
805–842.
Duerig, T.W., Melton, K.N., Stöckel, D., 2013. Engineering Aspects of Shape Memory
Alloys. Butterworth-Heinemann, London.
Elayyan, F., Silikas, N., Bearn, D., 2010. Mechanical properties of coated superelastic
archwires in conventional and self-ligating orthodontic brackets. Am. J. Orthod.
Dentofac. Orthop. 137, 213–217.
Jiang, F., Liu, Y., Yang, H., Li, L., Zheng, Y., 2009. Effect of ageing treatment on the
deformation behaviour of Ti–50.9 at% Ni. Acta Mater. 57, 4773–4781.
Kaphoor, A.A., Sundareswaran, S., 2012. Aesthetic nickel titanium wires–how much do
they deliver? Eur. J. Orthod. 34, 603–609.
Kim, T.-K., Kim, K.-D., Baek, S.-H., 2008. Comparison of frictional forces during the initial
leveling stage in various combinations of self-ligating brackets and archwires with a
custom-designed typodont system. Am. J. Orthod. Dentofac. Orthop. 133 (187-e15).
Kusy, R., Whitley, J., 1999. Influence of archwire and bracket dimensions on sliding
mechanics: derivations and determinations of the critical contact angles for binding.
Eur. J. Orthod. 21, 199–208.
Lombardo, L., Marafioti, M., Stefanoni, F., Mollica, F., Siciliani, G., 2012. Load deflection
characteristics and force level of nickel titanium initial archwires. Angle Orthod. 82,
507–521.
Miyazaki, S., Otsuka, K., 1989. Development of shape memory alloys. ISIJ Int. 29,
353–377.
Murayama, M., Namura, Y., Tamura, T., Iwai, H., Shimizu, N., 2013. Relationship be-
tween friction force and orthodontic force at the leveling stage using a coated wire. J.
Appl. Oral Sci. 21 (6), 554–559.
Naceur, I. Ben, Charfi, A., Bouraoui, T., Elleuch, K., 2014. Finite element modeling of
superelastic nickel-titanium orthodontic wires. J. Biomech. 47, 3630–3638.
Nematzadeh, F., Sadrnezhaad, S.K., 2012. Effects of material properties on mechanical
performance of Nitinol stent designed for femoral artery: finite element analysis. Sci.
Iran. 19, 1564–1571.
Nucera, R., Gatto, E., Borsellino, C., Aceto, P., Fabiano, F., Matarese, G., Perillo, L.,
Cordasco, G., 2014. Influence of bracket-slot design on the forces released by su-
perelastic nickel-titanium alignment wires in different deflection configurations.
Angle Orthod. 84, 541–547.
Otsuka, K., Wayman, C.M., 1999. Shape Memory Materials. Cambridge University Press,
Cambridge.
Parvizi, F., Rock, W.P., 2003. The load/deflection characteristics of thermally activated
orthodontic archwires. Eur. J. Orthod. 25, 417–421.
Pelton, A.R., Fino-Decker, J., Vien, L., Bonsignore, C., Saffari, P., Launey, M., Mitchell,
M.R., 2013. Rotary-bending fatigue characteristics of medical-grade Nitinol wire. J.
Mech. Behav. Biomed. Mater. 27, 19–32.
Proffit, W.R., Fields Jr, H.W., Sarver, D.M., 2014. Contemporary Orthodontics, fifth ed.
Mosby Press, St. Louis.
Reznikov, N., Har-Zion, G., Barkana, I., Abed, Y., Redlich, M., 2010. Measurement of
friction forces between stainless steel wires and “reduced-friction” self-ligating
brackets. Am. J. Orthod. Dentofac. Orthop. 138, 330–338.
Robertson, S.W., Launey, M., Shelley, O., Ong, I., Vien, L., Senthilnathan, K., Saffari, P.,
Schlegel, S., Pelton, A.R., 2015. A statistical approach to understand the role of in-
clusions on the fatigue resistance of superelastic Nitinol wire and tubing. J. Mech.
Behav. Biomed. Mater. 51, 119–131.
Singh, G., 2015. Textbook of Orthodontics, third ed. Jaypee Brothers Medical Publishers,
New Dehli.
Tageldin, H., de Llano Perula, M., Thevissen, P., Celis, J.-P., Willems, G., 2016. Resistance
to sliding in orthodontics: a systematic review. Jacobs J. Dent. Res. 3, 34.
Varela, J.C., Velo, M., Espinar, E., Llamas, J.M., Rúperez, E., Manero, J.M., Javier Gil, F.,
2014. Mechanical properties of a new thermoplastic polymer orthodontic archwire.
Mater. Sci. Eng. C Mater. Biol. Appl. 42, 1–6.
Weafer, F.M., Guo, Y., Bruzzi, M.S., 2016. The effect of crystallographic texture on stress-
induced martensitic transformation in NiTi: a computational analysis. J. Mech.
Behav. Biomed. Mater. 53, 210–217.
Weiland, F., 2003. Constant versus dissipating forces in orthodontics: the effect on initial
tooth movement and root resorption. Eur. J. Orthod. 25, 335–342.
Whitley, J.Q., Kusy, R.P., 2007. Influence of interbracket distances on the resistance to
sliding of orthodontic appliances. Am. J. Orthod. Dentofac. Orthop. 132, 360–372.
Wilkinson, P.D., Dysart, P.S., Hood, J.A.A., Herbison, G.P., 2002. Load-deflection char-
acteristics of superelastic nickel-titanium orthodontic wires. Am. J. Orthod. Dentofac.
Orthop. 121, 483–495.