1

Introduction

The specific properties of shape memory alloys are related to a reversible solid-to-solid phase transformation, the so-called martensitic transformation . One of these properties is superelasticity , also called pseudoelasticity , which has allowed for the development of NiTi endodontic instruments. Root canal preparation instruments made of NiTi have been described to promote preparation of curved and narrow root canals whilst maintaining the original anatomy .

There has been considerable improvement in file design, manufacturing methods, and preparation techniques on rotary endodontic instruments made of NiTi alloys; however, intracanal fracture of instruments caused by flexural fatigue remains a primary concern in the practice of endodontics, especially for canals with severe curvatures . The mechanical performance of NiTi alloys is sensitive to their microstructures and associated thermomechanical history . Therefore, one of many promising solutions to improve fatigue resistance of rotary instruments is to optimize the microstructure of NiTi alloys through novel processing or new manufacturing technologies. Recently, a new NiTi wire termed M-Wire (Sportswire LLC, Langley, OK, USA) has been developed through a proprietary thermomechanical processing procedure. Initial data for this type of NiTi raw material suggests significantly improved fatigue resistance of endodontic rotary instruments in comparison with those made of conventional superelastic NiTi alloys .

According to Johnson et al. , M-Wire is composed of 508 Nitinol, which has undergone a processing method comprised of thermomechanically treating the raw wire under specific tensile stresses and temperatures. Alapati et al. carried out the first metallurgical characterization of M-Wire, having reported that, under certain processing conditions, M-Wire contained austenite, martensite and R-phase, whose relative proportions depended on the processing conditions. They also showed that M-Wire had higher transformation temperatures, compared with a conventionally treated NiTi wire employed for the manufacture of NiTi instruments. These results were confirmed in a more recent characterization study , which showed in addition that M-Wire had a lower apparent elastic modulus, as well as smaller transformation stress and mechanical hysteresis. According to the materials science literature , in NiTi alloys with approximately equiatomic composition austenite is the high-temperature phase (usually denoted as the β-phase). It is usually an ordered solid solution with the B2 structure and, upon cooling, it transforms into the martensite phase, with the B19′ structure. Depending on their previous thermomechanical history, in alloys with excess of Ni, an intermediate martensitic phase, called the R-phase, can form prior to the transformation of the austenite to B19′ martensite .

In the present work, tensile, three-point bending and rotating-bending fatigue tests were carried out to assess the mechanical properties of M-Wire compared with those of a conventional wire used for manufacturing rotary NiTi instruments. The non-deformed microstructure of the wires was examined by transmission electron microscopy, while the fracture surfaces of failed wires were examined by scanning electron microscopy. The null hypothesis was that M-Wire had the same physical and mechanical properties as a conventional NiTi wire.

2

Materials and methods

The conventional wire (NiTi) and the M-Wire (MW) were both provided by Dentsply Maillefer (Ballaigues, Switzerland). Both wires had 1.2 mm in diameter. Wires with 100 mm in length were tensile tested until rupture in an Instron 5580 testing machine (Instron, Norwood, MA, USA). The tests were performed at room temperature with a strain rate of 1.0 × 10 ⁻³ s ⁻¹ . Apparent modulus of elasticity ( E ), transformation stress from austenite to martensite ( σ _A-M ), ultimate tensile stress ( σ _UTS ) and total elongation ( e _t ) were determined as the average of three tests, using the analysis software Instron Series IX for Windows.

Vickers microhardness measurements were performed with a Leica Durimet II tester (Leica, Wetzlar, Germany) using a 100 gf load. Three 15 mm-long specimens of each wire were embedded in acrylic resin, ground to half their diameter and then polished with diamond paste. Twenty indentations were made for each type of wire.

Three-point bending tests were performed in triplicate in the device shown in Fig. 1 a. A force F was applied to the wire specimen resting between two supports 26 mm apart. In a three-point bending test, the bending moment ( M ) in N cm at L /2 is given by

where F is the force given in N and L the distance in cm between the two supporting points. When the wire is bent by applying a vertical displacement d , there is tension below the neutral axis and compression above it ( Fig. 1 b). The maximum vertical displacement was chosen to be 5 mm, so as to reach 4% of maximum tensile strain amplitude at surface of the specimen. This corresponds to approximately the maximum amount of deformation suffered by an endodontic instrument of size #25/0.06 at 3 mm from its tip, when introduced in a standard root canal, according to the equations of kinematics .

Apparatus (a) and geometry (b) of the three-point bending test.

Rotating-bending fatigue tests were carried out in the device shown in Fig. 2 . One end of the wire specimen was clamped to the axis of a direct current motor controlled by an adjustable power supply. The opposite end of the wire was connected to a tachometer. The average fatigue life was determined from n = 10 specimens of each type of wire tested at a tensile-strain amplitude of 6% at the outer surface of curvature. The outer radius of curvature R to give rise to this strain level was 12 mm, as can be calculated using Eq. (2) :

where ɛ is strain level, R is outer radius of curvature and d is diameter of the wire.

Rotating-bending fatigue bench.

To maintain the specimen temperatures at maximum 10 °C above room temperature, a pilot study using a thin thermocouple was performed to choose the appropriate rotation speed, which was set to 32 rpm.

All data were found to fit a normal distribution. The data obtained in the tensile tests, Vickers microhardness measurements and rotating bending fatigue tests were analyzed statistically by using one-way analysis of variance, and the comparison of means was conducted by using Tukey multiple comparison tests. The level of confidence was set at p < 0.05.

Three samples of the fatigue fractured wires were randomly selected and examined by scanning electron microscopy (SEM) (Jeol JSM 6360, Tokyo, Japan) to assess their surface characteristics. Non-deformed wire specimens were analyzed by transmission electron microscopy using a 200 kV T20 microscope (FEI, Hillsboro, OR, USA). The observed specimens were thin longitudinal and transverse sections of the wires prepared by focused ion beam (FIB) milling in a Helios 600 equipment (FEI, Hillsboro, OR, USA).