Wire preparation
Rectangular stainless steel orthodontic archwires, 0.43 × 0.64 mm (3M Unitek, USA), were cut into pieces with a distal-end cutter (Hu-Friedy, USA) for further sample preparation. A curved shape was bent following the Ovoid Arch Form OrthoForm™ III (3M Unitek, USA) with an Adams Plier (678-320-U5 Hu-Friedy, USA). The length of the pieces was measured with an electronic digital calliper to ±0.1 mm accuracy. The total sample size of 120 pieces, 60 pieces per wire shape type, was determined by a multiple comparisons Power Analysis (Tukey-Kramer) test. These archwires were randomly divided into four groups, three experimental groups and one control group. Each group contained 30 pieces of wires, including 15 bent wires and 15 straight wires (24 mm). Additionally, one short straight wire (10 mm) was included for each group to study surface characteristics under a scanning electron microscope.
Immersing experiment
A total of 1 L of artificial saliva was prepared (0.400 g/L NaCl, 0.400 g/L KCl, 0.795 g/L CaCl2, 0.010 g/L Na2S9H2O, 1.000 g/L CH4N2O, 0.789 g/L KH2PO4) and divided into three portions equally, which were calibrated to pH 5.6, 6.6, and 7.6, respectively. Experimental groups were marked as groups 1, 2, and 3 in the same order as pH values. The wires were immersed in 20 mL artificial saliva and placed in the incubator (SANYO Electric Biomedical Co., Japan) at 37 °C for 4 weeks. The solution was refreshed every week to ensure a constant pH. The wires of group 4 were exposed to air at room temperature, served as the control group.
Surface roughness test
Surface roughness test was carried out by a calibrated contact profilometer (RTD-200 Portable Surface Roughness tester, New Star, Australia). Mean surface roughness values (Ra-values, μm) are defined as the arithmetic mean of the absolute departures of the roughness profile from the mean line. The software of profilometer was set at a Gauss roughness filter type with a cut-off length of 0.25 mm and an assessment length of 0.75 mm. The measurement was performed by a diamond point (φ = 5 μm) moving along the wire with a traversing speed of 0.135 mm/s at a pressing force of 4 mN. Prior to surface roughness test, the wires were taken out of the immersion solution and air-dried on a paper towel. Each wire was measured at three different positions. To ensure the reliability of measurements, these positions were indicated by lines as shown in Fig. 2a. The mean of three Ra-values per wire was used for statistical analysis.
Four-point flexural test
In this research, the flexural test was performed by a Z2.5 ZwickiLine Universal Testing Machine (Zwick/Roell, Germany) with a customized four-point appliance (Fig. 2b), which consists of two loading pins and two supporting pins. A 500-N maximal force, 2-mV/V sensitivity load cell (XforceP, Zwick/Roell, Germany), containing two loading pins, was installed in the machine with a crosshead speed of 1 mm/min. The load span, distance between both loading pins, was set at 10 mm. Loading pins deformed the wire at a speed of 1 mm/min until a threshold of 25 N was reached (Fig. 2). The load-deflection data obtained from the test were plotted as force-deflection curves, and the flexural Young’s modulus was calculated by the computer software testXpert II (Zwick/Roell, Germany) (Fig. 3).
Scanning electron microscopy
The surface of 10-mm straight wires was observed by SEM apparatus (G2 Pro, Phenom-World BV, The Netherlands) on its 0.43-mm side with a ×2000 magnification. Three places on each sample were randomly chosen for image scanning. The SEM software (Phenom Pro Suite, Phenom-World BV, The Netherlands) was used to quantity the surface morphology by reconstructing a 3D surface, from which mean surface roughness values (Ra) and mean surface depth values (Rz) were calculated. Three different regions of each image were selected for 3D reconstruction and calculation. Ra is the arithmetic mean of the absolute departures of the roughness profile from the mean line. Rz is the roughness depth, calculated as the sum of the height of the highest profile peak from the mean line and the depth of the deepest profile valley from the mean line.
Statistical analyses
All data were analysed by SPSS 22 (IBM® SPSS®, USA). A two-way analysis of variance (ANOVA) test was carried out for surface roughness (Ra-values), flexural Young’s moduli, and maximum force values, followed by post hoc multiple comparison tests. A significance level of p = 0.05 was used.