Bending characteristics of nitinol wire

I. Lopez,* J. Goldberg, ** and C. J. Burstone* Fnrmingrm,

Corn

ntil recently only a limited group of alloys have been used for orthodontic
wires, namely, gold-caper, austenitic stainless steel, and chromium-cobalt-nickle (CrCo-Ni) based materials. In general, the last two groups have similar mechanical properties
although processing, stress-relieving, heat-treating, and minor alloying adjustments allow
for small product differences. On the other hand, the gold-based wires have a modulus of
elasticity approximately one half that of the other two groups. Unfortunately, the yield
strength of gold alloys is less by approximately the same factor, which gives them a ratio
of yield strength to modulus of elasticity that is nearly the same as that of stainless steel
and Cr-Co-Ni alloys, Since springback or resilience is proportional to this ratio, all three
categories of alloys are approximately the same in regard to this clinical characteristic.
When used in orthodontic appliances, wires with low moduli of elasticity in combination with high resilience aid in delivering clinically desirable low continuous forces and
increased working time. It is these benefits, coupled with technologic advances in metallurgy and wire processing, which have promulgated the introduction of several lowstiffness, high-springback orthodontic wires. Braided wires use traditional stainless steel
but achieve their improved properties through unique design of the wires cross section. A
second approach to obtaining high-deflection wires is through the use of novel alloys,
the most notable being nitinol, a nickel-titanium alloy.2
Nitinol was developed by William F. Buehle? in the early 1960s. The original alloy
contained 55 percent nickel and 45 percent titanium, which resulted in a one-to-one
stoichrometric ratio of these elements. The most unique feature of this NiTi intermetallic
compound is the memory phenomenon, which is a result of temperature-induced
crystallographic transformations. Andreasen suggested that these shape changes might
be used by the orthodontist to apply forces. This memory principle is not used clinically,
although it would appear plausible.
The commercially available nitinol orthodontic wire contains 1.6 percent cobalt to
modify the transition temperature and mechanical properties. Even without the memory effect, the unusually low modulus of elasticity of 4.8 x lo6 p.s.i. and high resilience offer desirable features to the orthodontist. Using a clinical model, Andreasen and
Barrett demonstrated that nitinol had a lower stiffness than stainless steel and could be
deflected further without permanent deformation when tied into malaligned brackets.
Andreasen and Morrow have evaluated the bending characteristics and spring rate of
From the University of Connecticut Health Center, School of Dental Medicine.
This study was supported in part by Research Grant DE03953 from the National Institute of Dental
Research, National Institutes of Health. Bethesda. Md.
*Department of Orthodontics.
** Department of Restorative Dentistry.
00X!-9416/79/050569+07$00.70/0

1979 The C. V. Mosby Co.

569

straight sections of nitinol wire and have demonstrated it5 LIW in Class I, Cla\
II.
and
Class III malocclusions.
Although nitinol cannot be formed over a sharp edge, tirjt- and second-order hcnds
can be placed in the wire. Because of its unusual metallurgic structure, howcvcr. it cannot
be assumed, a priori. that bent-wire appliances have the same properties as unbent wires.
Indeed, even stainless steel demonstrates differences in propertics after a bend is placed.
depending on the direction of subsequent loading. Placement of permanent bends in wires
and activation in the oppositr direction is clinically important in all orthodontic techniques, since straight wires must be modified for exact finishing detail and to deliver the
necessary forces during treatment.
The purpose of this study was to characterize further the bending characteristics of
nitinol wire. In addition to straight-wire sections, standardized bends were also evaluated.
Finally, the time dependence of these measurements was considered.
Materials and methods
Nitinol* and a standard austenitic stainless steel wire,? each 0.18 inch in diameter,
were evaluated in three different modes. Permanent deformation versus reflection characteristics of straight-wire sections were measured with a Tinius Olsen stiffness testerz by
the procedure outlined in the new ADA specification No. 32 for orthodontic wires. Three
samples of each wire were tested. A 0.25 inch span length was used to minimize the
length of wire being fed into the testing span during bending. This mode of testing is
illustrated schematically in Fig. 1, A.
A second series of bending tests was made on straight sections, but instead of the
instantaneous loading conditions used in the procedure referred to above, samples were
loaded until 60 degrees of deflection had occurred and were maintained for 0, 1, 5, 20,40,
or 60 minutes and then released. A Tinius Olsen tester was again used, and permanent
deformation was measured for three different samples at each of the above-listed times,
for a total of eighteen samples per material. This experiment was repeated, but the same
three samples were reused at each of the indicated times, for a total of three samples per
material.
The third set of tests considered the permanent deformation versus deflection characteristics of nitinol and stainless steel wire after permanent bending. Samples were first
given a permanent bend of 35 degrees. Care was taken to bend in only one direction, and a
template was used to ensure reproducibility. The samples were then gripped in the Tinius
Olsen stiffness tester with the initial permanent bend 0.5 mm. from the edge of the vise.
Standard test procedures were then followed, with the force being applied in a direction
opposite that of the initial bend, as shown in Fig. 1, B. In another series of tests,
schematically represented in Fig. 1. C, the wires were first permanently deformed to 90
degrees (P,) and then returned to the 35 degree permanent bend (PZ). Samples were then
tested in the same direction as the last bend. PZ. In Fig. 1. F indicates the direction of
loading used during the test.
*Unitek Corporation, Monrovia, Calif.
Wandard Permachrome, Unitek Corporation, Monrovia, Calif.
STinius Olsen Testing Machine Corporation, Philadelphia, Pa,

Volume 15
Number 5

Bending characteristicsof nitinol wire 571

TEST
GRIPS

*
\
w

1 \

\i
\

P=35
\\

Fig. 1. The three modes of testing. P indicates the direction of bending of the configuration before
loading, which is represented by dashed lines. In C, P, and P, show change of loading direction to
produce preload configuration. F shows the direction of loading used during the test.

Results
Test results for straight-wire sections are shown in Fig. 2, where permanent deformation is plotted as a function of deflection for the stainless steel and nitinol wires. The
ability of the nickel-titanium alloy to undergo greater elastic deflections is clearly demonstrated. At any given activation, the nitinol experiences less permanent set. For wires
0.018 inch in diameter, the nitinol sustained approximately 23 degrees of activation and
stainless steel 13 degrees before any permanent deformation was evident. These results
are a further documentation of the flexibility and elasticity of nitinol described in earlier
works.2, 5x6
The behavior of these wires after the introduction of permanent bends is characterized
in Fig. 3. The data from Fig. 2 have been replotted for comparison. The most notable
feature of this figure is the large change in nitinol when tested in different modes. With
respect to bending of straight sections (Fig. 1, A), nitinol demonstrates superior elastic
properties as compared to stainless steel. However, when NiTi alloy is tested in a direction
opposite to apermanent bend (Fig. 1, B), there is a considerable loss of elastic behavior.
In fact, for activations of less than 40 degrees, nitinol exhibits greater permanent defor-

DEFLECTION, DEGREES
Fig. 2. Permanent deformation versus deflection for straight-wire sections, mode A in Fig. 1.
g

70-

ii
0

60 -

STAINLESS

E
g50-

DEFLECTION, DEGREES
Fig. 3. Permanent deformation versus deflection for the three modes of testing illustrated in Fig. 1

m&ion than stainless steel, their relative performance switching for activations larger than
50 degrees. When the nitinol wire was bent to 90 degrees and returned to 35 degrees prior
to testing (Fig. 1, C), it again demonstrated the desirable low permanent deformation,
although its performance was not comparable to the straight-wire test.
The magnitude of change associated with testing of stainless steel in different modes
was not nearly as large as that of the NiTi alloy. The changes which did occur, however.
document the desirability of owrbmdzng stainless steel. Even with small activations of 10
to 35 degrees, stainless steel in straight and overbent sections showed less permanent
deformation than the samples tested in a direction opposite the last permanent bend.
The time dependence of nitinol and stainless steel is illustrated in Fig. 4, where

Volume 75
Number 5

Bending characteristics

./

7-

TiEsEDe -

of nitinol

wire

573

- - 4

SAMPLES
STAINLESS
STEEL

- I- ,;DEpENGNT
- - - -E
SAMPLES

.1I
NITINOL

,
0

IO

20

30

40

50

60

TIME, MINUTES
Fig. 4. Permanent deformation versus length of time samples held at a GO-degreebend, x-x-x represents data obtained with three independent samples at each time interval. .-. -. represents data
obtained with the same three samples reused at each time interval. Range of values obtained at each
data point is indicated.

permanent deformation is plotted as a function of the length of time the samples were
maintained at a 60-degree bend. Two curves are shown for each material. The lines
labeled independent samples resulted from three replications at times of 0, 1, 5, 20,
40, and 60 minutes, new specimens being used for each test. Each line labeled reused
samples was obtained with a total of three samples, each one being retested at the time
intervals indicated.
The unexpected result was the time dependency of nitinol, even when independent
samples were used. The nitinol experienced approximately 5 degrees of additionalpermanent set over 60 minutes, most of which occurred within the first 5 minutes. A one-way

Fig. 5. Photograph showing nitinol on top and stainless steel wire directly underneath. A 35-degree
permanent bend was placed in each wire (testing mode 6). Each wire was then deflected 20 degrees in
a direction opposite the original permanent bend and released. The nitinol wire experienced 5 degrees
more permanent deformation.

analysis of variance of the nitinol independent samples indicated a statistically significant

difference (p < 0.05) between the various test times. A regression analysis of the same
data resulted in the equation
D =

IO + 2.02v

where D is permanent deformation in degrees and t is time in minutes. The average

permanent deformation of three nitinol samples tested for a period of 48 hours was 21
degrees, which compared favorably with the calculated value of 19.9 degrees.
As one might anticipate, the recycled samples experienced even larger increases in
permanent deformation, as the cold-working was apparently cumulative in its effect.
Discussion
The superior elastic behavior of nitinol wire as compared to conventional orthodontic
stainless steel wire has been clearly verified with straighr-wire sections. Cantilever bending of such specimens of nitinol up to 30 degrees caused insignificant permanent deformation, while a comparable deflection of the stainless steel resulted in a permanent set of
approximately 10 degrees.
Comparable superiority cannot be expected, however, when the wires experience
more elaborate manipulation. When permanently deformed and activated in an opposire
direction, nirinol actually undergoes more permanent deformation than stainless steel for
activations of less than 40 degrees. An example is shown in Fig. 5. Clinically, if it is
necessary to deform nitinol wire permanently, it should be overbent and permanently
deformed in the direction that the appliance will ultimately be activated. This technique is
similar to that recommended for stainless steel. Even with this technique, however, the
clinician must recognize that nitinol, although superior to stainless steel. does not have the
degree of superiority demonstrated by comparisons of straight unbent wire sections.
Nitinol appears to experience an unexpected time-dependent relaxation phenomenon.
Increases in permanent deformation were small when independent samples were evaluated
for periods of up to 60 minutes. However, the tests conducted for 48 hours indicated that
the relaxation phenomenon continues for at least this time period. Furthermore, samples
with a history of prior permanent bends appear to experience this relaxation effect to an

Volume 75
Number 5

Bending characteristics of nitinol wire

575

even greater extent. For the test conditions used in the present study, the permanent
deformation of the reused samples doubled in just 60 minutes, from 10 degrees to 20
degrees (Fig. 4). This would suggest that repeated bending of the wire should be avoided.
If springback (maximum elastic deflection) rather than permanent deformation is used for
comparison, nitinol exhibits 2.5 times the springback of stainless steel wire when loading
is instantaneous. However, if the load is maintained over 48 hours, which is more
applicable to the clinical situation, the superiority in springback is reduced to a factor of
2.0. Further research is needed to fully characterize the extent and cause of the timedependent behavior.
The unorthodox mechanical behavior of nitinol is associated with the unusual metallurgic structure of this nickel-titanium intermetallic compound. This same structure is also
responsible for the memory effect and the favorable modulus of elasticity. As with any
new dental material, the clinician must recognize its attributes and deficiencies in order to
use it to full advantage in the clinical setting. Bending, time dependency, and the cumulative effects of cold working can all have deleterious effects on the elastic characteristics of
nitinol. With proper appreciation and understanding of these manipulative variables,
however, the orthodontist can fully exploit the advantageous features of this new alloy.
REFERENCES
I. Burstone, C. J.. Baldwin, J. J., and Lawless, D. T.: The application of continuous forces to orthodontics,
Angle Orthod. 31: I-14, 1961.
2. Andreasen, G. F., and Hilleman, T. B.: An evaluation of 55 cobalt substituted nitinol wire for use in
orthodontics. J. Am. Dent. Assoc. 82: 1373- 1375, 1971.
3. Buehler, W. J., and Wang, F. E.: A summary of recent research on the nitinol alloys and their potential
application in ocean engineering, Ocean Eng. 1: 105- 120, 1968.
4. Andreasen, G. F.. and Brady, P. R.: A use hypothesis for 55 nitinol wire for orthodontics, Angle Orthod. 42:
172- 177, 1972.
5. Andreasen, G. F., and Barrett, R. D.: An evaluation of cobalt-substituted nitinol wire in orthodontics, AM. J.
ORTHOD. 63: 462-469,

1973.

6. Andreasen, G. F.. and Morrow, R. E.: Laboratory and clinical analysesof nitinol wjire, AM. J. ORTHOD. 73:
142-151, 1978.
7. Council on Dental Materials and Devices: American Dental Association specification No. 32 for orthodontic
wires not containing precious metals, J. Am. Dent. Assoc. 95: 1169- 1171, 1977.