Introduction
At present,
Orthodontic wires and arches are mainly made of metal alloys, such as NiTi shape memory alloy (SMA) and stainless steel (SS)
[1-5]. NiTi SMA wires were introduced
to orthodontics in 1972 by ANDREASEN.
The particular properties of superelasticity
(SE) and shape memory effect (SME), allow teeth to move under weak but constant
continuous forces over long treatment time and much larger displacement can be achieved
[4]. However, the stiffness of NiTi SMA wires is small, which can result in the loss of anchorage.
On the contrary, the high stiffness of SS wires can offer adequate anchorage
but its elasticity is low and can engender excessive orthodontic forces
preventing teeth from moving [6-8]. If NiTi SMA and
stainless steel wires were bonded together and used in orthodontic treatment,
the advantages of the two materials could be expressed by using SS and NiTi SMA wires as anchorage parts
and treatment parts respectively. This could greatly shorten the period of
orthodontic treatment and improve the quality.
However, it
is difficult to attach other metals to NiTi SMA
because of its microstructure and properties. NiTi
SMA is highly sensitive to changes in temperature and chemical composition when
compared to more common materials [9-11]. There are are
also several challenges in joining NiTi SMA to SS
successfully while maintaining the SE and SME of the NiTi
SMA and at the same time avoiding defects in the joint. In this study great
attention was paid to the procedures and weldling
methods for the homogeneous joint (NiTi-NiTi)
including electron beam welding, laser welding, friction welding, resistance
spot welding, resistance butt welding and brazing, etc [11-20]. So far, little
has been done to study welding of NiTi SMA and SS and
a reliable technique for joining NiTi SMA to other
materials has not been reported.
The present
work investigates the properties of a laser-brazed joint of NiTi
SMA and SS wires using silver-based filler metal and different brazing techniques
with particular attention to the heat affected zone (HAZ) of NiTi SMA. The purpose was to provide a theoretical and
experimental basis for manufacturing NiTi SMA components and structures in addition further development of
the application.
Methods and
Materials
Specimen
preparation
NiTi SMA
orthodontic wires (50mm×0.55mm×0.40mm) were obtained from the General Research
Institute for Non-ferrous Metals (Beijing, China), which consisted of 49.8 %
nickel and 50.2 % titanium with minimal amounts of trace elements (total of
carbon, oxygen and other trace elements <0.5 %). Its tensile strength and
elongation were 1213±52 MPa and 17±1% respectively. The
stainless steel orthodontic wires were obtained from 209 Longstone
Drive, Cherry
Hill, NJ, 08003, USA. The chemical composition of the
stainless steel (S32100) orthodontic wires (50 mm×0.55 mm×0.40mm) was 0.12 C,
1.0 Si, 2.0 Mn, 18±1 Cr, 9±1
Ni, 0.6±0.1 Ti, "d0.03 S, "d0.035 P, and the remainder Fe..
Mechanical properties at room temperature are 1247±56 MPa
ultimate tensile strength, 42±2% elongation. A silver-based filler metal was adapted
to braze NiTi SMA and SS wires with the composition
of 22% Cu, 18% Zn, 8% Sn and the balance Ag [21]. The
solidus and liquidus
temperature of the silver-based filler metal was 590! and 635! respectively.
Prior to
brazing, the brazing surfaces of base metals were polished with 600-grit SiC paper, and then ultrasonically cleaned in an acetone
bath. NiTi SMA and SS were brazed using a Nd:YAG laser welding machine
(JY-100, Laser Research Institute of Jilin Province, China) and the brazing heat input (Q=Pt)
was controlled by laser output power (P) and brazing time (t). The laser output
power and brazing time of 50 W/10 s, 60 W/15 s and 70 W/20 s were chosen for
this investigation. After brazing, no heat treatments were made on the joints. Twenty
specimens for each brazing parameter were prepared and subjected to tensile,
elasticity, bending testing and Vickers hardness measurements (Five for each
tests).
Fig.1.
Principle of mechanical property testing. A: Tensile strength test. B:
Elasticity test of the joint.
Tensile testing
Tensile
testing was conducted at room temperature with a universal testing machine
(AGS-10 kNG, Shimadzu, Kyoto, Japan) at a crosshead speed of 2 mm/min
and a gauge length of 6 mm, as shown in Fig.1 (A). The breaking stress (MPa) and percent elongation (%) were recorded. After
tensile testing, the fracture surfaces were observed using a scanning electron
microscope (JSM-5310, JEX Corp., Tokyo, Japan).
Elasticity
testing
The SE of NiTi SMA HAZ was investigated by stress-strain measurements
carried out at room temperature using the same electronic universal tester in
the same condition, as shown in Fig.1 (B). The gauge length of 3 mm was chosen
in this investigation on account of the width of NiTi
SMA HAZ was about 3 mm. The residual strain µ (%) of NiTi
SMA HAZ was recorded after loading and then unloading with the maximal strain µmax of 4%.
Bending
testing
The SME was
measured by the bending test to evaluate the shape recovery ratio (È) of NiTi SMA HAZ. Fig.2 shows a schematic diagram of the
bending test. The NiTi SMA side of joint specimen was
bent to 90° (¸) for 300 s, an angle (¸1) retained after unloading.
Then the specimen was put in boiling water (100!) for 5 s and ¸1 was recovered to ¸2. The shape
recovery ratio of NiTi SMA HAZ was determined by the
following formula:
.
Hardness measurements
Vickers
hardness measurements using a 200 g load
Fig.2. Schematic diagram of bending test.
and 30 s
dwell time were made across the two brazed metal interfaces to study the
modifications introduced by laser brazing. The hardness numbers were obtained
from two indentations in each specimen.
Statistical
analyses
All of the
test values were statistically analyzed by three-way ANOVA for analyzing three
factors, including the manufacturer, brazing parameters and the individual
element. Turkey’s test (±=0.05) was chosen as
the following multiple-comparison technique when necessary.
Results
The results
of the breaking stress (MPa) and percent elongation
(%) obtained from the tensile testing are depicted in Fig.3. The brazing
parameters have important influence on the mechanical properties of the
specimens. The breaking stress values of the specimens brazed at 60 W/15 t
reached 340±20 MPa and was not statistically
different (p>0.05) from those brazed at 70 W/20 t but was significantly
higher (p<0.01) than those brazed at 50 W/10 t (about 200±10 MPa). However, the elongation of the specimens increased
with an increase in the brazing heat input.
Fig.4
illustrates the stress-strain curves of NiTi SMA base
metal (no brazing) and NiTi SMA HAZs
of the joints brazed with different parameters at room temperature. As can be
seen, the behaviors for NiTi SMA base metal and NiTi SMA HAZs of the joints
brazed at 50 W/10 s and 60 W/15 s were not statistically different (p>0.05)
from each other. For these specimens, after a linear step, martensite
was stress induced at Ã=200±5 MPa. A flat
plateau up to a maximum strain of 4% was observed. Residual strains of µu=0.15% (no brazing), µu=0.37%
(50 W/10 s) and µu=0.51% (60 W/15 s) were
measured after unloading. On the other hand, for NiTi
SMA HAZ of the joint brazed at 70 W/20 s, the loading curve was not
statistically different (p>0.05) from that of NiTi
SMA base metal, but the unloading curve was significant different (p<0.01). The
residual strain reached 2.22% after unloading.
Fig.3. Breaking stress and percent elongation of the
specimens brazed with different brazing parameters. A: 50 W/10 s. B: 60 W/15 s.
C: 70 W/20 s.
Fig.4.
Stress-Strain curves of NiTi SMA base metal and HAZs.
Fig.5 shows
the bending test results of NiTi SMA base metal and NiTi SMA HAZs of the joints
brazed with different parameters. The results showed that the shape recovery
ratios (È) of all the specimens after brazing were significantly lower (p<0.01)
than that of NiTi SMA base metal. The shape recovery
ratio of NiTi SMA base metal was high up to 99.6% at
100!, while the È of NiTi SMA HAZ at the same temperature decreased with
increasing laser output power and brazing time, being 91.6% (50 W/10 s), 82.5%
(60 W/15 s) and 62.1% (70 W/20 s), respectively. When P and t were 50 W/10 s,
the shape recovery ratio of the HAZ reached 92.0% of the base metal recovery
ratio, and it sharply decreased to 62.3% of the base metal recovery ratio when P
and t were 70 W/20 s.
Fig.5.
Bending test results of NiTi SMA base metal and HAZs brazed with different brazing parameters. A: no
brazing. B: 50 W/10 s. C: 60 W/15 s. D: 70 W/20 s.
Fig.6.
Hardness profiles across joints.
The microhardness profiles across the brazing beam centerline
of the joints brazed with different parameters are presented in Fig.6 with y-axis, x-axis and “0” represented hardness values, distance
and the brazing seam centerline, respectively. The brazing seams showed the
lowest hardness values and the hardness values in both NiTi
SMA and SS HAZs increased with increasing distance
from the brazing seam centerline to the base metals. The laser output power and
brazing time had effects on the hardness values in both HAZs.
The increased P and t caused a decrease in hardness values in the HAZs. However, the hardness values in NiTi
SMA HAZ with increasing P and t presented more obvious changes compared with SS
HAZ.
Representative
fracture surfaces observed by SEM are shown in Fig.7. Fracture of the specimens
brazed at 50 W/10s occurred on the interface layer between NiTi
SMA and the filler metal. Fracture of the specimens brazed at 60 W/15 s mainly
occurred in the center of the brazing seam, as shown in Fig.7 (A). The fracture
surfaces displayed as-cast structures of the filler metal. When brazing heat
input was relatively high (at 70 W/20 s), fractures occurred in the NiTi SMA HAZ [Fig.7 (B)], and little ductile pores could be
seen on the fracture surfaces of the NiTi SMA base
metal appearing as typical equiaxial ductile pores [Fig.7
(C)].
Discussion
When
orthodontic wires are bowed and seated in orthodontic brackets attached to
teeth, fatigue fractures often occur after repeated loading and unloading in
modes similar to brittle fracture. As a result, the tensile strength and
flexural strength of the composite orthodontic wires made of NiTi SMA and SS are required to withstand this stress. In
this study, the high stiffness of SS and the SE and SME of NiTi
SMA were required, so the loss of SE and SME in NiTi
SMA HAZ after brazing must be minimal and the HAZ width of the two base metals
must be narrow.
The results
of tensile testing showed that the combined strength of the SS and the filler
metal were high, resulting in all fractures of the specimens occurring in the
center of the brazing seam; in the NiTi SMA HAZ; or at
the interface layer between NiTi SMA and the filler
metal. The stiffness in the SS HAZ was slightly influenced by the narrow width
of the SS HAZ of 1 mm. Since the width of SS HAZ was narrow (only 1 mm) (Fig.6),
attention was mainly focused on the changes of properties of NiTi SMA HAZ of the laser-brazed joint.
Newman et
al. [7, 8, 22] reported a maximum clinical load of
1.82 kg can be applied without preventing blood circulation in the periodontal
ligament. The minimal tensile strength of the specimens after brazing was about
200 MPa (about 4.49 kg), much bigger than 1.82 kg,
satisfying the clinical requirements. As a result, the main factors influencing
the clinical application of the composite archwires
were flexural strength and the loss of SE and SME of NiTi
SMA HAZ.
Fig.7. SEM
photographs after tensile testing. A: Fracture surface of the specimen brazed
at 60 W/15 s. B: Fracture surface of the specimen brazed at 70 W/20 s. B: Fracture
surface of the specimen of NiTi SMA base metal.
The SME of NiTi SMA HAZ brazed at 50 W/10 s was high (È=91.6%)
and the SE was similar to NiTi SMA base metal while
the width of NiTi SMA HAZ was narrow and only 2 mm
(Fig.6). However, when the laser-brazed joints were bent to more than 90°, 20
percent of the specimens fractured on the interface layer between NiTi SMA and the filler metal. The reason is that the
metallurgical bonding strength between NiTi SMA and
the filler metals was low and diffusion between NiTi
SMA and the filler metal hardly occurred due to the low brazing heat input. On
the other hand, when the laser-brazed joints brazed at 60 W/15 s and 70 W/20 s
were bent to more than 90°, fracture rarely occurred. But the SE of NiTi SMA HAZ brazed at 70 W/20 s was seriously influenced
and the SME loss was high (È=62.1%) while the width of NiTi
SMA HAZ was more than 3 mm. Also, the NiTi SMA HAZ
was seriously softened (Fig.6) and the elongation of the specimens
significantly increased (Fig.3). This was due to the damage of the corresponding
relation between the parent phase (B2) with orderly lattice structure and
twinning structure (B19’) caused by the martensitic
transformation of the NiTi SMA during laser brazing
heat cycles with fast heating and cooling rates.
SME and SE
(or transformation pseudo- elasticity) are always related to the thermo-elastic
martensitic transformation from the B2 parent phase
to the B19’ monoclinic phase in an approximate equiatomic
NiTi SMA [1]. One of the properties of the SME and SE
of Niti SMA is the reversibility of martensitic transformation in crystallography, i.e., the
corresponding relationship between the B2 parent phase and the B19 martensite must be maintained. The interfacial energy of
the coherent interface between B2 phase and B19’ martensite
of NiTi SMA is very low but its elastic distortional
potential is high due to the distortion on the interface. This maintains the
corresponding relation on the coherent interface. When NiTi
SMA and SS were laser-brazed, the brazing heat cycle was fast and the brazing
temperature was high. The constant growing of new phases made the elastic
distortional potential increase consistently by the thermal effect. The
corresponding relation between B2 phase and B19’ martensite
would be destroyed by the plastic deformation due to the increasing elastic
distortional potential beyond the yield limit of the parent phase [23]. As a
result, the corresponding relation between the parent phase and the martensite of NiTi SMA were
partially destroyed after NiTi SMA and SS wires were
laser-brazed at high brazing heat input, i.e., the SME and SE of NiTi SMA HAZ were partially destroyed. .
When NiTi SMA and SS were brazed at 60 W/15 s, the brazing heat
input was relatively low. As a result, the SE loss of NiTi
SMA HAZ was relatively low (Fig.5) but the SME was relatively high (È=82.5%)
while the width of NiTi SMA HAZ was relatively narrow
and about 2-2.5 mm (Fig.6). So the properties of NiTi
SMA HAZ could satisfy the clinical requirements.
Conclusion
This work
studied the properties of laser-brazed joint of NiTi
SMA and SS orthodontic wires using silver-based filler metal and different
brazing parameters. The tensile strength of the joint could reach 340±20 MPa while the loss of SE and SME of NiTi
SMA HAZ was relatively low by strictly controlling brazing heat input. The
properties of the composite archwires made of NiTi SMA and SS by laser brazing could satisfy the clinical
requirements and the composite archwires have good
prospects.
Acknowledgments
The authors
are grateful for the financial support from Jilin Province
Committee of Science and Technology of China (No. 20000518).
References
Otsuka K
and Wayman CM (eds.): ‘Shape Memory Materials’; 1998,
Cambridge University Press.
Andreasen
G. A clinical trial of alignment of teeth using 0.019 inch
thermal nitinol wire with transitional temperature
range between 31 and 45!.
Am J Orthod 1980; 78: 528-537.
Robert J.
Nikolai. Elastic responses to longitudinal torsion of
single-strand, rectangular, orthodontic archwire
segments. Dent Mater 1995; 11: 169-176.
Medawar L
El, Rocher P, Hornez J C,
et al. Electrochemical and cytocompatibility
assessment of NiTiNOL shape memory alloy for
orthodontic use. Biomolecular Engineering 2002; 19: 153-160.
Iijima M,
Ohno H, Kawashima, et al. Mechanical behavior at
different temperatures and stresses for superelastic
nickel-titanium orthodontic wires having different transformation temperatures.
Dent Mater 2002; 18: 88-93.
Bachman J. Torquing of stainless steel and nitinol
wires: A comparison of mechanical properties. Europ J Orthod
1983; 5:167-169.
Berkovitz
B KB. The structure of the periodontal ligament: an update. Eur J Orthod
1990; 12: 51-76.
Murrel EF, Yen E.H.K, Johnson RB. Vascular changes in the periodontal
ligament after removal of orthodontic forces. Am J Orthod
1996; 110: 280-286.
Beyer J, Besselink PA, Lindenhovius
JH. Shape memory
alloy’86 proceedings of the international symposium on shape memory alloys,
1986: 492-497.
Miyazaki S, et al. Fatigue and fracture of shape memory and superelasticity materials. Metal Physics Seminar 1980; 4-3:
111.
Gomez JM,
Mendez FJ, Urena A, et al. Transient liquid phase
(TLP) diffusion bonding of a copper based shape memory alloy using silver as
interlayer. Scripta Materialia
1997; 37(6): 861-867.
Tuissi A,
Besseghini S, Ranucci T, et
al. Effect of Nd-YAG laser welding on the functional
properties of the Ni-49.6at%Ti. Mater Sci Eng A 1999; 273-275: 813-817.
Hsu YH, Wang SK, Chen C. Effect
of CO2 laser welding on the shape memory and corrosion characteristics of TiNi alloy. Met Mater Trans, 2001; 32A: 569-576.
Ikai A, Kimura K, Tobush
H. TIG welding and shape memory effect of TiNi shape
memory alloy.
Journal of Intelligent Material Systems and Structures 1996; 7(6): 646-654.
Shinoda T,
Tsuchiya T, Takahashi H, et al. Functional characteristics of friction welded
near-equiatomic TiNi shape
memory alloy. Transactions of the Japan Welding Society 1991; 22(2): 30-36.
Hirose A,
Araki N. Welding TiNi shape memory alloys. Metals
1989; 59(8): 61-68.
Araki T,
Hirose A. Characteristics and fracture morphology of Ti-Ni type shape memory
alloy and its laser welded joint. Metals 1989; 38(428): 478-483.
Makita M,
Kimura K, Tobushi H, Lin PH. Transactions of Japan
Society of Mechanical Engineers (Part A) 1994; 60(579): 2603-2611.
Shinoda
T, Tsuchiya T, Takahashi H. Friction welding of shape memory alloy. Welding
International 1992; 6(1): 20.
Shinda T,
Owa T, Maguula V. Microstructural analysis of friction welded joints in TiNi alloys. Welding International 1999; 13(3): 24-29.
Sun DQ, Qiu XM, Zhu XC. Composite orthodontic arch
wires and manufacturing method in the oral cavity. Chinese patent,
01138750.5.
Newman GV.
Epoxy adhesives for orthodontic attachmants: progress
report. Am J Orthod 1965, 51: 901-912.
Wasilewski
R J. The ‘Yield’ behavior of stoichiometric
TiNi across the martensitic
transformation range. Scripta Metall 1971; 5-2: 131-136.
