Can we expect similar behavior among CuNiTi 35°C wires?

ABSTRACT Objective: This paper aims to verify the thermodynamic, mechanical and chemical properties of CuNiTi 35ºC commercial wires. Methods: Forty pre-contoured copper-nickel-titanium thermodynamic 0.017 x 0.025-in archwires with an Af temperature of 35°C were used. Eight wires from five different manufacturers (American Orthodontics® [G1], Eurodonto® [G2], Morelli® [G3], Ormco® [G4] and Orthometric® [G5]) underwent cross-sectional dimension measurements, tensile tests, SEM-EDS and differential scanning calorimetry (DSC) tests. Parametric tests (One-way ANOVA and Tukey post-test) were used, with a significance level of 5%, and Pearson’s correlation coefficient test was performed between the Af and chemical elements of the wires. All sample tests and statistical analyses were double-blinded. Results: All wires presented standard dimensions (0.017 x 0.025-in) and superelastic behavior, with mean plateau forces of: G1 = 36.49N; G2 = 27.34N; G3 = 19.24 N; G4 = 37.54 N; and G5 = 17.87N. The Af means were: G1 = 29.40°C, G2 = 29.13°C and G3 = 31.43°C, with p>0.05 relative to each other. G4 (32.77°C) and G5 (35.17°C) presented statistically significant differences between each other and among the other groups. All samples presented Ni, Ti, Cu and Al in different concentrations. Conclusions: The chemical concentration of the elements that compose the alloy significantly influenced the thermodynamic and mechanical properties.


INTRODUCTION
Thermodynamic wires are wires that undergo changes in their crystallographic arrangement depending on the relationship between the temperature they are exposed to and their transition temperatures. These inherent temperatures can be manipulated by heat treatments or by atomic substitution, e. g. by replacing part of the nickel or titanium concentration of a nickel-titanium (Ni-Ti) alloy by copper (Cu), resulting in CuNiTi, a tertiary alloy. [1][2][3][4][5][6] The addition of Cu to NiTi alloys also reduces stress and temperature hysteresis, giving more stability to the superelastic characteristics. [1][2][3][4][5][6] Wires manufactured with this alloy have been marketed by brands in an attempt to show which wire characteristics are responsible for the particular behavior among brands of the same cross-sectional diameter.

MATERIAL AND METHODS
The sample consisted of forty 0.017 x 0.025-in pre-contoured CuNiTi wires, with austenitic finish (Af) temperature of 35°C, divided into five groups (G1 to G5), according to their commercial brand, for double-blinded tests ( Table 1).

CROSS-SECTIONAL MEASUREMENTS
The cross-sectional measurements of the wires were performed using a digital caliper with an accuracy of 0.001 mm (Starret, USA). Five wires randomly selected from each manufacturer were cleaned along with the caliper claws with alcohol, and each one was measured for both height and width at five different points of each archwire. 10

DIFFERENTIAL SCANNING CALORIMETRY (DSC)
To define the temperature transition range (TTR) of the wires, samples were taken from the straightest portion of each archwire. The samples were cut with orthodontic pliers into lengths of approximately 3 mm and weights of approximately 3.5 mg using a precision electronic scale with an accuracy of 10 μg. 11,12 Each specimen was cleaned with alcohol, dried and placed in a covered and sealed aluminum crucible for a DSC test on a Netzsch Polyma DSC 214 instrument (Selb, Germany). An atmosphere of nitrogen gas at 50ml/min filled the heating chamber and an empty aluminum crucible was the inert reference. 13,14 The temperature range of the test was from 60°C to - 40°C The samples did not reach the plastic deformation limit, as the deformation carried out was 8% of their initial length. 3,15,16 Samples damaged by crushing or sliding caused by the mechanical grips were discarded. 4 The whole test was per- controller; e. hot air blower at minimum speed and distance of 30cm from the sample. B) View of the clamp system of the tensile test with controlled temperature: a. sample attached to the clamps; b. thermometer of the temperature controller placed behind the sample; c. clamp system; d. load cell; e. styrofoam thermal box surrounding the entire clamp system. The superelasticity rate (SE rate) was calculated using the ratio between the elastic modulus of the two lines generated in the deactivation curves of the tested wires (E2/E1). The first elastic modulus (E1) was obtained from the straight line of the superelastic clinical plateau, and the second (E2) was obtained from the first three points of the deactivation curve of the load/deflection graph. Wires presented superelastic behavior when their second SE rate was higher than 8. 5,17

ENERGY DISPERSIVE SPECTROSCOPY (SEM/EDS)
The microstructure of the wires was observed by SEM/EDS in a T-330 A JEOL-JSM microscope (Toronto Surplus & Scientific Inc., Canada) 18 to determine the superficial chemical composition of the wires in each group and the phases (elements or associations) that compose them. A fractographic image was obtained and the chemical composition was automatically determined in percentages.

STATISTICAL ANALYSIS
Statistical analyses of the data were performed by comparing their normal distribution using the Kolmogorov-Smirnov test.

CROSS-SECTIONAL MEASUREMENTS
The cross-sectional dimensions were consistent with the information provided by the manufacturers (Table 2).

DIFFERENTIAL SCANNING CALORIMETRY (DSC)
The DSC tests showed that almost all wires presented an Af below that reported by the manufacturers, except group 5:  (Table 3).   (Table 4).

ENERGY DISPERSIVE SPECTROSCOPY (SEM/EDS)
The percentage of titanium (Ti), nickel (Ni), copper (Cu) and aluminum (Al) surface concentrations for the archwires of each manufacturer is given in Figure 2. The Pearson correlation test performed between Af and the chemical elements was not significant (p > 0.05).   The novelty of this study lies in its questioning of what variable in the alloy is responsible for the diverse mechanical behavior of the wires that should theoretically express the same patterns (Fig 3).
In order to understand any discrepancies between the wires of different manufacturers analyzed in this study, height

15
The results showed that the Cu concentration was 4.57% in G5, which was the lowest of all. This low concentration was not expected to present satisfactory results in relation to Af compared to the copper concentrations in the other brands. However, the Al concentration was 4.05%, which was much higher than the other brands, except for G2. This suggest that the association of these chemical elements may have been responsible for the excellent thermodynamic behavior of this wire.
The Pearson correlation test showed that, although weak, there was a negative correlation between Af and Cu (ρ = -0.408), and a positive correlation between Af and Al (ρ = 0.168). This could mean that a higher concentration of Cu in the alloy tends to decrease Af, while the higher concentration of Al in the alloy tends to increase Af, corroborating with the interpretations obtained from the results and other reports. 7,8,9,25 The Eurodonto ® COBRE NiTi wires (G2) showed similar concentrations of Cu and Al (4.83% and 4.35%, respectively); however, their thermal behavior was lower (Af = 29.13°C). In addition, it showed an equiatomic ratio between nickel and titanium (Ni 42.38% and Ti 42.62%), and a plateau force of 27.3N. The fact that there is a balanced ratio between nickel and titanium in these wires already guarantees stability in the thermal and mechanical behavior of this material, 5,24 and no specific treatment is required to prevent decomposition at other phases however, with a Cu concentration equal to 6.32% and a smaller quantity of Al (0.81%). The fact that the Af temperature in this group is practically 7°C below the mean body temperature suggests that the crystalline structure of this metal alloy reached its fully stabilized austenitic phase when inserted into the intraoral environment. 3,5,9,22,25 In the case of this study, this occurred when subjected to a uniaxial traction test with a con- probably the reasons that this sample expressed high plateau forces and its Af temperature was below 35°C.
Among the results found in this study, the one not expected was the inclusion of aluminum in the alloy, since the manufacturers did not include it in the description of the compositions, naming all wires of this kind as a tertiary alloy -CuNiTi. Nowadays, NiTi and Cu-based alloys are the most studied shape memory alloys (SMAs) in this class of materials, due to their unique thermomechanical behavior. However, despite the preference for NiTi, which is currently the most used SMA, Cu-based SMAs are emerging as potential substitutes, mainly in metallurgical studies for diverse uses, in particular, superelastic Cu 17 Al 11 Mn 4 that exhibits mechanical properties that are similar to those of NiTi, but less expensive. 25 Although the literature has shown that binary and tertiary alloys containing Cu and Al in their composition have exhibited excellent thermal and electrical properties, as well as shape memory and superelasticity, 5,25 this information has not been reported in orthodontic wires yet.
Despite the differences found in this study, all the wires evaluated can be used clinically, however, the thermodynamic properties must be considered, in order to enhance the mechanical potential of the wires and optimize the clinical outcomes. Wires with a higher Af temperature demand minor activations to reach the superelastic plateau at deactivation when exposed to body temperature during clinical use, while under the same environmental conditions, those wires with a lower Af temperature need higher activations to express the superelastic plateau at deactivation.