doi: 10.1111/j.1460-2695.2006.01021.x
`
`Fatigue behaviour of nickel–titanium superelastic wires and
`endodontic instruments
`
`M. G. A. BAHIA 1, B. M. GONZALEZ 2 and V. T. L. BUONO 2
`1Department of Restoration Dentistry, 2Department of Metallurgical and Materials Engineering, Federal University of Minas Gerais, Belo Horizonte,
`MG, Brazil
`
`Received in final form 24 February 2006
`
`A B S T R A C T Endodontic files made of nickel–titanium (NiTi) superelastic wires can be employed in ro-
`tary techniques for cleaning and shaping curved root canals, suffering tensile–compressive
`strain cycles with maximum amplitudes between 3 and 5%. The aim of this work was to
`study the fatigue behaviour of this material under such high deformation conditions, using
`NiTi instruments and superelastic wires taken from their production line. One hundred
`load–unload tensile cycles in the superelastic regime (4% elongation) were applied to NiTi
`wires. New endodontic instruments were fatigue-tested simulating the geometrical con-
`ditions found in their clinical use. It was found that only small changes took place in the
`parameters describing the mechanical behaviour of the cycled wires. The measured aver-
`age number of cycles to failure varies inversely with the maximum tensile strain amplitude
`in the fatigue tests (r = 0.993).
`Keywords
`endodontic instruments; fatigue; mechanical behaviour; NiTi wires.
`D = file diameter (mm)
`D0 = file tip diameter (mm)
`D3 = file diameter at 3 mm from the tip (mm)
`L = distance from the file tip (mm)
`NCF = number of cycles to failure
`R = curvature radius of the artificial canal (mm)
`T = file taper
`εP = plastic strain at breakage (%)
`εT = tensile strain amplitude (%)
`σ A–M = transformation stress (MPa)
`σ UTS = ultimate tensile strength (MPa)
`
`N O M E N C L A T U R E
`
`I N T R O D U C T I O N
`
`The use of nickel–titanium (NiTi) superelastic wires to
`manufacture rotary endodontic files constitutes an im-
`portant development of the endodontic therapy, leading
`to the application of rotary techniques for the chemical–
`mechanical preparation of curved root canals. Regarding
`modern concepts, the final root canal shape has to allow
`adequate irrigation and close adaptation of the filling ma-
`terial during obturation. High lateral forces can develop in
`the preparation of curved canals due to the stiffness of the
`
`Correspondence: V. T. L. Buono. E-mail: vbuono@demet.ufmg.br
`
`endodontic instruments, influencing the amount of dentin
`removed from the canal walls. The resulting canal aberra-
`tions include ledges, zippings and perforations. NiTi ro-
`tary endodontic instruments were developed to overcome
`these inconveniences, because they are able to maintain
`the original canal shape without creating severe irregular-
`ities, particularly in narrow curved canals.1
`The main characteristic of the NiTi alloys used in the
`manufacture of rotary endodontic instruments is a large
`nonlinear elasticity that allows a recovery of up to 8%
`of the tensile strains upon unloading. This is called the
`superelastic effect and stems from the stress-induced for-
`mation of martensite taking place during loading, which,
`under appropriate conditions, accommodates the applied
`
`518
`
`c(cid:2) 2006 Blackwell Publishing Ltd. FatigueFractEngngMaterStruct 29, 518–523
`
`PGR2015-00019 - Ex. 1034
`US ENDODONTICS, LLC., Petitioner
`
`

`
`stress. In a stress–strain curve, formation of martensite
`starts at a given stress level, called the transformation
`stress, at which the curve departures from linearity. In this
`aspect, it is similar to the yield point of ordinary metallic
`materials, but the microscopic changes responsible for this
`deformation are reversible. When the stress is released,
`the martensitic phase transforms back into austenite and
`the material recovers its original dimensions and shape.
`Superelastic NiTi alloys are two to three times more flex-
`ible than stainless steel, the material employed to man-
`ufacture endodontic hand files. This is due to its low
`elastic modulus, but is also a consequence of the su-
`perelastic behaviour of NiTi, because the transformation
`stress remains approximately constant when the material
`is transforming to martensite. The restoration stress act-
`ing during unloading, when martensite transforms back
`to austenite, is also approximately constant and smaller
`than the transformation stress, giving rise to mechanical
`hysteresis during a load–unload cycle.
`In straight root canals, rotary endodontic files operate
`by cutting and removing organic tissue and debris, expe-
`riencing mostly frictional forces opposing their torsional
`motion. However, when the instrument rotates inside a
`curved root canal it is bent and thus submitted to tensile–
`compressive strain cycles in the region of the canal curva-
`ture, in addition to the torsional restraints. This alternat-
`ing strain cycling can give rise to failure of the instruments
`by fatigue.2,3 The strain levels attained by the endodontic
`files during this cyclic loading depend on the root canal
`and instrument geometries, being concentrated at the por-
`tion of the file positioned in the maximum curvature re-
`gion of the root canal. Among the two parameters gen-
`erally employed to define the root canal geometry, radius
`and angle of curvature, the former is the most meaning-
`ful, insofar as fatigue resistance of machine-driven NiTi
`instruments is concerned, because the tensile strain com-
`ponent is inversely proportional to this parameter.2 Fur-
`thermore, the importance of the geometrical factor in root
`canal shaping becomes even greater when multiple curva-
`tures are present. The high incidence of secondary cur-
`vatures in human lower molars (30%) and the fact that
`they occur predominantly in the apical third of the root
`
`FAT I G U E B E H AV I O U R O F N iT i S U P E R E L A S T I C W I R E S
`
`519
`
`canal, at a mean distance of 2.2 mm from the foramen,4
`demand that NiTi rotary endodontic instruments possess
`exceptional fatigue resistance at relatively high strain lev-
`els. When NiTi endodontic instruments are clinically em-
`ployed for preparing curved root canals using the rotary
`technique, cyclic bending strains with maximum tensile
`amplitudes in excess of 5% are developed.5
`Detailed knowledge of how such instruments behave un-
`der fatigue is thus fundamentally important to ensure that
`their clinical usage be safe. Therefore, the aim of this study
`was to obtain basic information on the mechanical be-
`haviour under cyclic loading of the NiTi wires employed
`in the manufacture of endodontic instruments.
`
`E X P E R I M E N TA L P R O C E D U R E
`
`The NiTi wires used in this work were provided by
`Dentsply-Maillefer (Baillagues, Switzerland) and were
`taken from the production line of ProFile rotary endodon-
`tic instruments just before the final machining step. The
`ProFile instruments used to evaluate the fatigue resistance
`were obtained from the regular suppliers, withdrawn from
`sealed boxes and sequentially numbered on the handle, us-
`ing a high-speed diamond bur. The cutting shaft of these
`instruments have a conical shape, starting at a non-cutting
`tip of small diameter and going up at increasing diame-
`ters along 16 mm, where the shaft becomes cylindrical.
`Circular grooves are made in this cylindrical shaft to in-
`dicate the size and taper (conicity) of the instrument. It
`is attached to the instrument’s handle, by means of which
`the file is set to rotate. An instrument size 25, 0.06 taper,
`usually denoted as 25/.06, is shown in Fig. 1, which is a
`low-magnification scanning electron microscopy (SEM)
`image composition (built by superposing two comple-
`mentary images). Six sets of instruments, each contain-
`ing 10 instruments of the same size and taper, in a total
`of 60 instruments, were analysed. Three sizes were em-
`ployed: 20, 25 and 30. The instrument size corresponds to
`the diameter of its tip in 10th of millimetres. The tapers
`analysed were 0.04 and 0.06, meaning that the diameter
`of the instruments cutting edge increased by 0.04 or 0.06
`mm in each millimetre of their length.
`
`Fig. 1 SEM image composition of an
`instrument size 25, 0.06 taper.
`
`c(cid:2) 2006 Blackwell Publishing Ltd. FatigueFractEngngMaterStruct 29, 518–523
`
`PGR2015-00019 - Ex. 1034
`US ENDODONTICS, LLC., Petitioner
`
`

`
`520
`
`M. G. A. BAHIA etal.
`
`Specimens of the NiTi wires in the as-received condi-
`tion, with 1.2 mm in diameter and 80 mm in length, were
`tensile-tested in an Instron universal testing machine. The
`transformation stress, σ A–M, the stress at maximum load
`(ultimate tensile strength), σ UTS and the plastic strain at
`breakage (total elongation), εP, were determined as the
`average values of three tests performed at room temper-
`ature and at a strain rate of 1.0 × 10−3 s−1, using an ex-
`tensometer. Cyclic loading tests were also carried out at
`room temperature, in similar NiTi specimens and at the
`same testing machine, at a strain rate of 1.0 × 10−2 s−1.
`The specimens were loaded in the superelastic regime to
`4% total elongation and unloaded to zero stress. This is a
`strain-controlled fatigue test, but unloading to zero strain
`would lead to plastic deformation of the specimens, be-
`cause the superelastic strain is not fully recovered after
`such a high straining level. This cycle was performed 100
`
`Fig. 2 Artificial canal with an instrument mounted.
`
`times and then the specimens were tensile-tested to failure
`at a strain rate of 1.0 × 10−3 s−1. As before, transforma-
`tion stress, stress at maximum load and plastic strain at
`breakage of the cycled wires were evaluated as the average
`of three tests.
`The fatigue resistance of the endodontic instruments was
`tested in a laboratory fatigue test bench, using a endodon-
`tic motor operating at 250 rpm and an artificial canal made
`with quenched H13 tool steel to avoid geometric changes
`due to wear. Figure 2 shows a file inserted in the artificial
`canal. Its radius and angle of curvature, 5 mm and 45◦, re-
`spectively, were chosen based on measurements of these
`parameters previously performed on curved mesial canals
`in mandibular human molars and curved buccal canals
`in maxillary molars.5 The parameter used to evaluate the
`fatigue resistance was the average number of cycles to fail-
`ure (NCF), calculated by multiplying the time to failure
`and the rotation speed. Fracture surfaces were analysed by
`SEM (Jeol JSM 6360).
`
`R E S U LT S A N D D I S C U S S I O N
`
`The average stress–strain curves obtained from three
`specimens of the NiTi wires in the as-received condition
`and from three specimens previously submitted to 100
`load–unload cycles are shown in Fig. 3. The stress peak at
`the beginning of the superelastic plateau corresponds to
`the nucleation of martensite variants in austenite, while
`the subsequent decrease in stress is associated with the
`propagation of these convenient orientated martensite
`variants.6 The mean values of σ A–M, σ UTS and εP, de-
`termined for the as-received and cycled wires, are shown
`in Table 1. Comparison of the values of these parame-
`ters indicates that only small changes in the mechanical
`
`Fig. 3 Average stress–strain curves of the
`NiTi wires in the as-received condition and
`after 100 load–unload cycles. The inset
`shows cycling curves.
`
`c(cid:2) 2006 Blackwell Publishing Ltd. FatigueFractEngngMaterStruct 29, 518–523
`
`PGR2015-00019 - Ex. 1034
`US ENDODONTICS, LLC., Petitioner
`
`

`
`FAT I G U E B E H AV I O U R O F N iT i S U P E R E L A S T I C W I R E S
`
`521
`
`Table 1 Mean values of the transformation stress, σ A–M, ultimate
`tensile strength, σ UTS and plastic strain at breakage, εP
`
`Condition
`
`Property
`
`As-received Cycled 100 times
`
`σ A–M (standard deviation), MPa
`σ UTS (standard deviation), MPa
`εP (standard deviation), %
`
`550 (7.5)
`1404 (7.0)
`11.2 (0.9)
`
`404 (21.3)
`1403 (18.1)
`12.4 (0.3)
`
`behaviour of the wires took place after 100 load–unload
`cycles up to 4% tensile strain. The ultimate tensile
`strength and the plastic strain at breakage are practically
`the same (the change in εP is of the order of the stan-
`dard deviation). In fact, the main effect of cyclic loading
`was decreasing σ A–M by about 26.5% and increasing the
`strain at the superelastic plateau by approximately 25%.
`The cycling behaviour illustrated by the inset in Fig. 3
`corroborates these observations, showing also that the un-
`recoverable strain increases, while the hysteresis loop de-
`creases, as the number of cycles is increased. Tolomeu
`et al.,7 comparing the mechanical properties of superelas-
`tic NiTi stents under monotonic and cyclic loading, found
`that stress changes due to cyclic loading took place mainly
`at the superelastic plateau, in agreement with the results
`found in the present work.
`The fracture surfaces of as-received and cycled wires
`tensile-tested to failure display the characteristic cup-and-
`cone fracture of ductile metals, with a peripheral shear
`area surrounding the fibrous central region. The transi-
`tion area of the fracture surface is illustrated in Fig. 4.
`The fibrous central region of the fracture surfaces con-
`tained dimples and slip lines, as expected for this type
`of failure. The presence of fatigue striations and numer-
`ous secondary cracks on the shear area of a previously
`cycled specimen can be observed in Fig. 4b. Similar fea-
`tures could not be detected on the shear area of wire
`specimens tensile-tested to failure in the as-received con-
`dition (Fig. 4a). The fatigue striations in the shear area
`of the fracture surfaces of the cycled wires indicate that
`the failure of these specimens in the tensile tests involved
`pre-existing fatigue cracks, developed during load–unload
`cycling.
`In the fatigue test device employed in this work, the in-
`struments were set to rotate at a fixed position, in such
`a way that the maximum curvature was always located at
`3 mm from the instrument’s tip. The knowledge of the
`instrument’s geometry allowed then the estimation of the
`tensile strain component at the surface of the instrument
`placed in region of maximum curvature, εT, which is given
`by
`εT = (cid:2) 2R
`D
`
`(cid:3)−1
`
`− 1
`
`Fig. 4 SEM images of the fractured surface of (a) an as-received
`and (b) a cycled wire tensile-tested to failure. Arrow in (b) indicates
`fatigue striations and secondary cracks.
`
`where R is the curvature radius of the artificial canal and
`D is the instrument’s diameter in the region of maximum
`curvature, which, in turn, is the sum of the diameter of
`its tip, D0, with the distance from the tip, L, times the
`instrument’s taper, T, that is, D = D0 + LT. The results of
`the fatigue tests of the endodontic files are summarized in
`Fig. 5 in terms of the NCF as a function of the instrument
`diameters at 3 mm from the tip, D3. It can be observed
`that the NCF decreases as D3, and thus the size and taper
`of the instruments, increase. Similar results were obtained
`by other authors in the studies of the fatigue behaviour of
`ProFile instruments.8–10
`A plot of NCF against the estimated values of εT is
`shown in Fig. 6. The measured average NCF varies in-
`versely with the maximum tensile strain amplitude in the
`fatigue tests (r = 0.993). The influence of the tensile strain
`component on the fatigue behaviour is in agreement with
`what is generally expected when high strain amplitudes are
`
`(1)
`
`c(cid:2) 2006 Blackwell Publishing Ltd. FatigueFractEngngMaterStruct 29, 518–523
`
`PGR2015-00019 - Ex. 1034
`US ENDODONTICS, LLC., Petitioner
`
`

`
`522
`
`M. G. A. BAHIA etal.
`
`Fig. 5 Average, NCF, as a function of instrument diameter at 3 mm
`from the tip, D3.
`
`Fig. 6 Change in the NCF with the estimated tensile strain
`component.
`
`Fig. 7 SEM images of the fractured surface of a 30/.06 ProFile
`instrument fatigue-tested to failure. Arrow in (a) marks the area
`enlarged in (b). Arrows in (b) indicate secondary cracks.
`
`employed. In other materials, such amplitudes would be
`far beyond the elastic regime. Although the expected fa-
`tigue lives of the tested instruments seem small (<103
`cycles), it is important to mention that they are usually
`employed to format only 10 curved root canals before
`being discarded. In a previous work, it was shown that
`the fatigue life consumption of these instruments during
`shaping 10 curved root canals corresponds to an average
`of 58% of their expected fatigue life.5
`The fracture surfaces of the instruments tested in fatigue
`showed the typical characteristics of fatigue failure at high
`stress levels, that is, a small peripheral shear area surround-
`ing a large fibrous central region, as illustrated in Fig. 7a.
`A higher magnification image of the shear area pointed
`out in Fig. 7a is presented in Fig. 7b, showing the pres-
`ence of fatigue striations and secondary cracks (arrowed).
`The common aspect revealed by SEM fractography was
`the presence of a high density of secondary fatigue cracks
`at the fracture surfaces of cycled wires tensile-tested to
`
`failure and of the endodontic instruments tested in fa-
`tigue. It is possible that the fast and multiple nucleation of
`secondary cracks, which can be associated with the large
`amount of martensite variant interfaces and twins gen-
`erated during the deformation cycle in the superelastic
`regime,11 is the dissipation mechanism responsible for the
`relatively high fatigue resistance of the superelastic NiTi
`alloys employed in endodontics.
`
`C O N C L U S I O N S
`
`Cyclic load–unload tensile deformation in the supere-
`lastic regime has little effect on the material’s mechani-
`cal properties. The strain-controlled fatigue tests of the
`analysed endodontic instruments demonstrated that they
`exhibit an exceptional fatigue resistance for the high level
`of tensile strains usually found in practice. This behaviour
`was associated to the dissipation of mechanical energy due
`
`c(cid:2) 2006 Blackwell Publishing Ltd. FatigueFractEngngMaterStruct 29, 518–523
`
`PGR2015-00019 - Ex. 1034
`US ENDODONTICS, LLC., Petitioner
`
`

`
`to nucleation and propagation of secondary fatigue cracks
`along martensite variants and twin borders formed during
`cyclic loading of superelastic NiTi.
`
`Acknowledgements
`
`This work was partially supported by Fundac¸ao de Am-
`paro a Pesquisa do Estado de Minas Gerais—FAPEMIG,
`Belo Horizonte, MG, Brazil, and Conselho Nacional
`de Desenvolvimento Cientifico e Tecnologico—CNPq,
`Bras´ılia, DF, Brazil.
`
`R E F E R E N C E S
`
`1 Walia, H., Brantley, W. A. and Gerstein, H. N. (1988) An initial
`investigation of the bending and torsional properties of Nitinol
`root canal files. J. Endod. 14, 346–351.
`2 Pruett, J., Clement, D. and Carnes, D. J. (1997) Cyclic fatigue
`testing of nickel–titanium endodontic instruments. J. Endod. 23,
`77–85.
`3 Melo, M. C. C., Bahia, M. G. A. and Buono, V. T. L. (2002)
`Fatigue resistance of engine-driven rotary nickel–titanium
`endodontic instruments. J. Endod. 25, 765–769.
`4 Cunninghan, C. J. and Senia, E. S. (1992) A three-dimensional
`
`FAT I G U E B E H AV I O U R O F N iT i S U P E R E L A S T I C W I R E S
`
`523
`
`study of canal curvatures in the mesial roots of mandibular
`molars. J. Endod. 18, 294–300.
`5 Bahia, M. G. A. and Buono, V. T. L. (2005) Decrease in the
`fatigue resistance of nickel–titanium rotary instruments after
`clinical use in curved root canals. Oral Surg. Oral Med. Oral
`Pathol. Oral Radiol. Endod. 100, 249–255.
`6 Huang, X. and Liu, Y. (2001) Effect of annealing on the
`transformation behavior and superelasticity of NiTi shape
`memory alloy. Scripta Mater. 45, 153–160.
`7 Tolomeu, D., Davidson, S. and Santinoranont, M. (2000) Cyclic
`properties of superelastic Nitinol: Design implications. In:
`Proceedings of the Second International Conference on Shape Memory
`and Superelastic Technologies, Pacific Grove, CA, USA, The
`International Organization on Shape Memory and Superelastic
`Technologies, pp. 471–476.
`8 Haikel, Y., Serfaty, R., Bateman, G., Senger, B. and Allemann,
`C. (1999) Dynamic and cyclic fatigue of engine-driven rotary
`nickel–titanium endodontic instruments. J. Endod. 25, 434–440.
`9 Yared, G. M., Bou Dagher F. E. and Machtou, P. (2001)
`Influence of rotational speed, torque and operators proficiency
`on ProFile failures. Int. Endod. J. 34, 47–53.
`10 Gambarini, G. (1999) Torsional and cyclic fatigue testing of
`ProFile NiTi rotary instruments. J. Evol. Dent. 2, 4–14.
`11 Horbogen, E. (2002) Some effects of martensitic
`transformation on fatigue resistance. Fatigue Fract. Eng. Mater.
`Struct. 25, 785–790.
`
`c(cid:2) 2006 Blackwell Publishing Ltd. FatigueFractEngngMaterStruct 29, 518–523
`
`PGR2015-00019 - Ex. 1034
`US ENDODONTICS, LLC., Petitioner