INTRODUCTION

Orthodontic mechanics using closing loops have two distinct advantages over sliding mechanics: there is no friction and pure translation (bodily movement) is theoretically possible, provided that a moment to force ratio (M:F) of approximately 10:1 can be maintained.1

Many studies have investigated the characteristics of the T-loop design, usually made from titanium molybdenum alloy (TMA),2–5 but to date very little has been published on T-loops made from nickel-titanium (NiTi). These wire alloys have the advantages of a shape-memory effect combined with superelasticity, together with excellent corrosion resistance and biocompatibility.6 Initial studies by Kum et al7 have shown that non-preactivated T-closing loops made of NiTi failed to achieve an optimum M:F for bodily tooth translation, with average values below 6:1. This was improved by adding preactivation bends into the NiTi T-loops, with M:F values greater than 10:1 being achieved.8

It is well documented that NiTi wires are temperature sensitive, with smaller forces produced at lower temperatures.69–11 This is due to the change in the crystal structure from the austenitic to the martensitic form as the temperature is decreased611 and is known as thermoelastic martensitic transformation.12 Austenitic to martensitic transformation can also occur as a result of stress application, otherwise known as stress-induced martensitic (SIM). There is thus a strong interrelationship between temperature and stress, and, as stated by Meling and Ødegaard,13 “a decrease in temperature is equivalent to an increase in stress.”

There is evidence that the effect of short-term temperature change on the bending stiffness of super-elastic NiTi wires is dependent on whether the wire is in the activation or deactivation phase.13 The deactivation phase is of clinical interest, and it appears that a transient drop in temperature has less enduring influence on the force produced than a transient rise in temperature, where the effect is longer lasting. Intraoral temperature can fluctuate quite markedly, with values between 5°C and 58°C being recorded in the region of the upper incisor teeth,1415 although, on average, the median temperature fluctuates between 33°C and 37°C for approximately 80% of the time.15

To our knowledge, no data is available on the effect of temperature on the forces, moments, and moment to force ratios produced by NiTi T-loops. The aim of this study was, therefore, to investigate these parameters in both NiTi and TMA T-loops at temperatures ranging from 10°C to 50°C in 10° increments.

MATERIALS AND METHODS

The method used in this study is largely based on a method previously described.7 Briefly, the top arm of a custom-built stand (Figure1) supported an analogue micrometer (range: 0–25 mm, ±0.001 mm) with a nonrotating shaft (Mitutoyo, Kawasaki, Kanagawa, Japan) to which was attached a force transducer (Entran, model ELJ-IM-O-0.2 N; Entran Sensors and Electronics, Fairfield, NJ). An aluminum extension was attached to the lower end of the force transducer, to which was bonded a stainless steel twin bracket (Victory series miniature twin stainless steel bracket with 0° torque and 0° angulation; 3M Unitek, Monrovia, Calif) such that the upper end of a mounted test wire was in line with the center axis of the load cell. The middle arm of the stand held a moment transducer (Sensotec, model 34/0911-14; Sensotec, Columbus, Ohio) onto which an aluminum cap was fixed. A second bracket was bonded to this cap such that it was precisely centered over the pivotal axis of the moment transducer and perfectly in line with the first bracket. The arm was positioned in such a way that the brackets were separated by a distance of 10 mm. A third arm on the stand was used to mount a third bracket at a distance of 6 mm from the second bracket and served as an anchor point to prevent the moment transducer from being overloaded.

View Figure

• Download as Powerpoint
• Download Figure

The T-loops were formed from straight lengths of 0.018″ × 0.025″ Japanese superelastic NiTi (Neo Sentalloy, F-100; GAC, Bohemia, NY) and 0.017″ × 0.025″ TMA (Ormco, Orange, Calif) wires. The T-loop dimensions were as follows: height 8.45 mm, width 10.45 mm, radii 1 mm. The T-loop shape was set in the NiTi wires by forming and clamping the wire around stainless steel pins on a template and heating them at 510°C for 9 minutes in a crown furnace (Ivoclar programat P90, Schaan, Liechtenstein). Preactivation bends of 15° and 30° for each loop arm were formed by clamping the legs at the required angle on the template prior to heating. Four samples of each preactivation configuration were tested for both NiTi and TMA wires (N = 24).

The T-loop samples were positioned in the brackets such that the loop was positioned equidistant between the moment and force transducer brackets (centered position) and held in place with O rings (Alastic; 3M Unitek). The upper and lower ends of the sample wire were secured with Gurin locks (0.018–0.022″; 3M Unitek) to prevent the wire slipping through the brackets during activation.

The custom stand was placed in a thermostatically controlled insulated oven (model 76J-1; RFL Industries Inc, Boonton, NJ) that housed a small electrically driven fan to distribute the heat. Cooling was obtained by adding dry ice to the oven until 10°C was reached and maintained for a minimum of 10 minutes. Ambient air temperature was measured by two digital thermocouple thermometers, and the average reading of both was recorded. Once temperature had been maintained, the wire T-loop sample was opened to a maximum of 8 mm and back to 0 mm in 1-mm increments. The moments and forces were recorded at each increment opening. The temperature was then increased to 20°C using the oven elements (coarse control), and a 60-W bulb was placed in the oven (fine control). The T-loop specimen was again opened and closed in 1-mm increments, with force and moment values recorded. This process was repeated for temperatures 30°C, 40°C, and 50°C, and for all samples of both NiTi and TMA wires at preactivations of 0°, 15°, and 30°. Temperature fluctuations did not exceed 0.5°C from the desired temperature.

The transducers were calibrated at each temperature by suspending known weights in situ directly from the bracket mounted on the force transducer, or from removable light aluminum arms at a known distance from either side of the moment transducer. In both instances, a direct linear relationship was produced for each temperature. Linear regression equations were calculated (Excel 2000; Microsoft, Redmond, Wash), which were used to determine the output readings in grams or gram.mm for the force and moment transducers, respectively.

The following assumptions were made for this study:

• — Heat setting of the closing loop shape into the NiTi wire had negligible effect on the elastic properties of the wire. In other words, a complete single phase material was achieved after loop setting with minimal alteration in the transitional temperature range.

• — The distance between brackets was representative of average clinical cases.16

• — Activation and deactivation of the T-loops were limited to the elastic properties of the wire materials.

• — Clamping of a Guerin lock (0.018–0.022″) to the samples had negligible effect on the wire material.

• — Forces and moments that were generated in one plane had negligible effect on those generated out of plane.17

• — Thermal expansion and contraction of the test apparatus was identical for all wire samples tested.

• — The stiffness of an archwire is affected by both Young's modulus of elasticity of the wire material and the cross-section moment of inertia. The NiTi wire was larger than the TMA, producing an increased cross-sectional area, although this is offset by the lower modulus of elasticity.

Mixed-model analysis (Stata, version 9; Stata Corporation, College Station, Tex) was used to detect statistical differences between the mean results of material type (Japanese NiTi and TMA), preactivation (0°, 15° and 30°), and temperature (10°C, 20°C, 30°C, 40°C, and 50°C) for the force, moment, and M:F produced during activation and deactivation. Mixed models are an extension of analysis of covariance that cope with lack of independence in the data. In this study, P < .01 was considered statistically significant. The standard error was adjusted for different correlations between and within experiments. The use of a cubic polynomial function was used because it is a biconcave curve that best approximates the hysteresis section of the curve and allows the use of a linear model in the statistical analysis.

RESULTS

The effect of temperature on the force production of NiTi and TMA loops at 0°, 15°, and 30° preactivation are shown in Figure 2a–c, respectively. The TMA activation and deactivation curves are linear, as expected, while the NiTi curves show hysteresis on the deactivation stage as well as a flatter curve. The force produced by the NiTi samples increased as the temperature increased, and this was significant for all degrees of preactivation (P = .000). Statistically, temperature affected the 15° and 30° preactivated TMA samples, but not the 0° preactivated samples. The force values produced by the NiTi samples differed significantly from the equivalent TMA samples for each temperature (P = .000) except for the 0° and 15° preactivation samples at 40°C.

View Figure

• Download as Powerpoint
• Download Figure

The effect of temperature on the moments produced is shown in Figure 3. A pattern similar to that of the force production is noted, with TMA exhibiting linear curves and the NiTi wires the characteristic flattened curve with hysteresis. The moments increased as the temperature increased, which was significant (P = .000) for all NiTi samples across all temperatures, except for the 15° preactivation samples at 10°C and 20°C. Temperature did not influence the moments produced by TMA to any significant degree. The moments produced by the NiTi loops at each temperature significantly differed from the equivalent TMA loop at the same temperature, except for the 0° preactivation loops at 40°C and 50°C.

View Figure

• Download as Powerpoint
• Download Figure

Moment to force ratios were calculated mathematically from the forces and moments and are illustrated in Figure 4. Temperature did not affect the M:F as much as either the force or the moments, and was not significant except for the 30° preactivation TMA loop at 40°C and 50°C and the 0° preactivation NiTi loops. There was variation in how the materials compared to each other at each temperature, with significant differences being recorded for the 15° preactivation loops at temperatures above 20°C and the 30° preactivation loops at 50°C.

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

Temperature had a significant effect on the force produced by the NiTi T-loops, with lower forces accompanying each 10°C drop in temperature. These results mirror previous findings for straight lengths of NiTi wires undergoing first-order displacement bends.6913 The optimal force for tooth movement depends on a complexity of factors including the type of tooth movement desired, the geometry and size of the object being moved, and the existing biological constraints, such as bone density, cellular activity rates, periodontal health, and alveolar bone levels.

Light forces (generally regarded as less than 200 g) can produce an adequate biological response in the periodontal tissue, and heavier forces are associated with hyalinization, undermining bone resorption, and are associated with root resorption.18 As an example, from Figure 2 it can be seen that a deactivation opening of 4 mm will produce a load (force) of less than 50 g at 10°C in all NiTi loops, whereas the force at the same opening is greater than 200 g at 50°C. This force fluctuation in a single wire at different temperatures can have a significant impact on the force levels experienced at the periodontal level. Research by Moore et al15 has shown that the intraoral temperature remained within the 33–37°C range for 79% of the time, higher temperatures were experienced for only 1% of the time, and cooler temperatures for 20% of the time. The lower dental arch also appears to experience a wider temperature fluctuation compared to the upper arch.14 Thus, cooling is of greater importance than heating when considering force delivery from a NiTi wire, particularly for lower archwires.

Temperature influenced the moments produced by the NiTi loops in a similar fashion to the force values, with lower moments at lower temperatures. Previous research has indicated that the 15° preactivated loop produced a smaller moment compared to the 0° loop up to an activation of 3 mm, after which the curves crossed and the 15° preactivated wire produced a higher moment before reversing again on deactivation.8

In this study, the peak moments produced by the 0° preactivation were higher than the 15° preactivation over the whole activation and deactivation range, and the 30° preactivation T-loops had moments slightly higher than the 0° preactivation loops (Figure 3a–c). There are two possible explanations for this phenomenon: firstly, the moment recorded at the second bracket is influenced by binary bracket geometries, with factors such as interbracket distance, degree of activation, and position of the loop influencing the final value.19 Secondly, with preactivation bends incorporated at the occlusal level of the loop, leg crossover can complicate the neutral reference position when the loop is positioned into the test apparatus. Care was taken to ensure that the neutral position was obtained prior to the start of each experimental procedure, but other uncontrollable factors such as wire creep may have influenced the resting neutral position at the different temperatures. Creep is defined as the time-dependant plastic strain of a material under a static load or constant stress.20

The overall M:F ratio produced by the NiTi loops was largely unaffected by temperature, indicating that the drop in force value per temperature change was proportionally similar to the drop in the moment value, thus maintaining an overall constant M:F ratio. Some fluctuation is visible with the M:F lines of the 15° and 30° preactivation wires at 10°C. This can be ascribed to low force readings when the loop approaches the neutral position at the low temperature.

The TMA wires in this study acted as the comparator for the NiTi T-loops, and the results show that temperature had minimal influence on the force, moment, or M:F ratio of these T-loops. We have previously shown that increasing preactivation bends in TMA and NiTi wires leads to resultant increases in moments.8 Though this trend is followed in the current NiTi sample, the distinction is less clear for the TMA sample, including a moments reading for the 0° preactivation wire that is greater than zero (Figure 3). This can be attributed to two main causes, namely, the effect of repeat open-and-closing cycling and the effect of temperature. Error determination of repeat open-and-closing cycling (at a constant temperature) revealed that the initial moments reading was higher by between 5% and 10% of the average remaining repeat values. This error was minimised in the study by starting at different temperatures within sample groups. Although temperature did not have a significant effect on the TMA loops, there was some variation in the forces and moments that can purely be ascribed to temperature fluctuation. This averaged 31 g (±29 g) for force and 49 g·mm (±104 g·mm) for moments. Both of these effects would serve to diffuse the initial values recorded for TMA and mask the moment effect produced by the preactivation bend.

Some of the negative readings obtained in this experimental study reflect the more erratic behavior of the readings as the loop activation approaches zero. This is possibly due to slight movement of the loops when engaging it to the apparatus, accurately obtaining the neutral position of the loops with preactivation bends, and slippage, despite steps being taken to minimize these effects.

The distance between brackets was set at 10 mm and 6 mm, the former representing the average calculated maximum distance between brackets on a second premolar and a canine in the instance where a first premolar tooth has been extracted. This distance may vary according to the size of the brackets used as well as the size of the particular individual's teeth. The latter distance reflects a representative distance between two adjacent brackets on the anterior teeth.

It should be noted that the experimental data obtained in this study for Neo Sentalloy NiTi wires is a combination of both temperature- and stress-induced martensitic-austenitic transformation during loop activation, and it is difficult to distinguish the contribution of each for any particular temperature and activation setting.

CONCLUSIONS

• Temperature has a significant effect on both the forces and moments produced by a T-loop made from NiTi, with lower values recorded as the temperature decreases.

• The M:F remained relatively constant at the different temperatures, indicating that the reduction of the force and moment values occurred in a proportionally similar fashion.

• Temperature had minimal influence on the forces, moments, and M:F of TMA T-loops.