A total of 432 new 25-mm NiTi instruments (n = 12 per group) from JIZAI [Glider (size 13, 0.04 taper), JIZAI I (size 25, 0.04 taper), and JIZAI III (size 35, 0.04 taper)] and TruNatomy [Glider (size 17, 0.02 variable(v) taper), Prime (size 26, 0.04v taper), and Medium (size 36, 0.03v taper)] were tested for their torque/force generation and canal preparation ability (changes in volume, surface area, canal centering ratios, apical canal deviation and percentages of untouched area) in canals of extracted molar teeth. A further 42 instruments (n = 7, each) were used to compare the geometric designs and metallurgical characteristics. The overview of the experiment is outlined in Fig. 1.

The files were inspected with a digital microscope (VH-8000; Keyence, Osaka, Japan) at a magnification of ×20, and analysis was performed using ImageJ v1.50e software (Laboratory for Optical and Computational Instrumentation, Madison, WI, USA). The assessment included measuring the number of blades in the active part of the instrument and measuring the helix angle (the blades near the shaft in triplicate). Subsequently, the same instrument was examined under a scanning electron microscope (SEM) (JSM-7900 F; JEOL, Akishima, Japan) at 15 kV. Images were taken at ×100, ×150, and ×500 magnification to assess the blade geometry, cross-section, tip (active or non-active), surface finishing, and surface wear.

The EDS/SEM analysis (JSM-7900 F; JEOL) was performed on the surface of six instruments of each type at a working distance of 10 mm and a beam voltage of 15.0 kV.

The differential scanning calorimetry (DSC) analysis was conducted with three instruments of each brand. The working part of the instrument was cut into portions of 3 mm length and placed in an aluminum case, with a total weight of 20 mg. These cases were then placed in a DSC device (DSCe60; Shimadzu, Kyoto, Japan) together with a vacant aluminum case, which served as a reference. The experimental setup involved filling the chamber with argon gas and using liquid nitrogen as a coolant, with a cooling/heating rate of 0.33 °C/s. The thermal cycles were conducted by initially raising the temperature from 20 to 100 °C, then lowering it to − 100 °C to record the cooling chart, and then raising it to 100 °C to record the heating chart. The start and finish temperature of martensitic transformation (Ms and Mf, respectively), the R-phase transformation (Rs and Rf, respectively), and the reverse transformation (As and Af, respectively) were analyzed by determining the points of intersection between the maximum lambda gradient line and the baseline data.

This study was approved by the Dental Research Ethics Committee of Tokyo Medical and Dental University (number D2014-033-03). A total of 120 human mandibular molars, extracted for reasons unrelated to this study, were collected with informed consent from the donors. The teeth were stored in 100% relative humidity at 5 °C. After the distal root and crown were removed, the mesial root was measured and cut to a length of 12 ± 1 mm. It was then firmly fixed on a specially made-mold.

The teeth were scanned prior to preparation using a micro-CT device (inspeXio SMX-100CTPlus; Shimadzu) with specifications of 70 kV, 100 mA, a rotational pace of 0.5° through 360°, 1.0 mm thick aluminum filter, and a volumetric resolution of 0.03 mm. The images were visualized using three-dimensional imaging software (Amira 5.4.4; Visage Imaging, Berlin, Germany) with consistent parameters which resulted in 400–500 greyscale transaxial slices per sample. The buccolingual and mesiodistal diameters of the root canal at 5 mm prior to the apical foramen were measured. The cross-sectional configuration was determined by comparing the buccolingual and mesiodistal measurement ratios. The canals were classified as oval, long oval, or flattened, based on their aspect ratio of ≤ 2, 2 to 4, and ≥ 4, respectively [19].

The quantity of sample was identified using G*Power 3.1.7 software (Heinrich Heine Universität Düsseldorf, Düsseldorf, Germany) with an alpha level of 0.05, a beta power of 85%, and an effect size of 0.6, as guided by the outcomes of prior investigation [20]. This study utilized twelve samples for each group since a minimal sample size of 10 per group was specified. To ensure anatomical homogeneity, the samples were standardized according to criteria such as volume, surface area, degree of curvature (approximately 30˚), cross-sectional diameter (buccolingual and mesiodistal diameter of the root canal), internal morphology (oval canal), and canal length [21]. From this initial sampling, 72 mesial canals were sorted out and assigned to one of the six test groups (n = 12 each). There were no statistical differences in canal geometric criteria between the test groups (Table 1). The canals were subjected to random categorization (http://www.random.org/), resulting in three groups in accordance with the frequency of up-and-down movements at working length (1, 3, or 6 times) and further subdivided into two groups in accordance with the instrument system (JIZAI or TruNatomy).

An original automated canal instrumentation device [8, 22,23,24] was used (Fig. 1), which consisted of an endodontic motor (J Morita), a speed adjustable bench (MX2-500 N, Imada, Toyohashi, Japan) enabling up-and-down movement, and sensing apparatus for torque/force measurement. The device was structured to deliver an artificial up-and-down movement to simulate a manual shaping procedure (2-second descent followed by a 1-second ascent at 50 mm/min, regardless of the stress applied).

The same instrumentation protocol was used for each group. This included verifying the patency of the root canal and establishing the working length of 1 mm from the apical foramen with a 10 K file (Ready Steel; Dentsply Sirona). The canals were then instrumented using a single-length technique, with the JIZAI and TruNatomy instruments. The glide path preparation was performed with the JIZAI Glider (size 13, 0.04 taper) and the TruNatomy Glider (size 17, 0.02v taper). The first shaping was conducted with the JIZAI I (size 25, 0.04 taper) and the TruNatomy Prime (size 26, 0.04v taper), followed by the second shaping with the JIZAI III (size 35, 0.04 taper) and the TruNatomy Medium (size 36, 0.03v taper). The instruments were operated at 500 rpm and 1.5 N·cm. Each specimen was instrumented according to repeated up-and-down movements of 1, 3, or 6 times at the working length. The upward/downward force and torque values during the root canal preparation procedure were registered. The canal was disregarded from additional examination if an instrument fractured.

A total of 10 mL of 2.5% sodium hypochlorite (Takasugi Pharmaceutical, Fukuoka, Japan) was applied by gently moving a 30-G needle (Ultradent, South Jordan, Utah, USA). Additionally, a lubricant (RC-Prep: Premier Dental Products, Norristown, PA, USA) was utilized in the process. Three mL of 17% ethylenediaminetetraacetic acid (BSA Sakurai, Nagoya, Japan), 3.0 mL of 3% sodium hypochlorite, and 2.0 mL of saline solution (Otsuka Pharmaceutical, Tokushima, Japan) were applied following completion of the instrumentation. After the moisture in the canal was absorbed with paper points (Dentsply Sirona), the samples were rescanned, and micro-CT analysis was carried out as detailed previously.

The alteration in canal volume and surface area was determined by calculating the difference between the values measured before and after instrumentation (Fig. 2). The datasets prior to the instrumentation were overlayed onto datasets taken after instrumentation using the Affine Registration algorithm in the Amira 5.4.4 software (Visage Imaging). Four transaxial slides at one, three, five, and seven millimeters from the apical foramen were used to calculate the centering ratio: (the width of removed dentin in mesial direction − the width of removed dentin in distal direction)/ post-treatment canal width [20]. A better centering ability is indicated by scales that are close to 0. The untouched area was computed by: (number of static voxels × 100) / number of surface voxels [21]. Static voxels refer to the voxels that stayed in the same spot on the canal surface prior to as well as following the shaping procedure.

(a–c) Three-dimensional images of the mesiobuccal and mesiolingual canals before instrumentation (a), after instrumentation (b) and after superimposition (c) in a mandibular molar. (c) The black arrow indicates the uninstrumented canal area (in green), and the red area indicates the instrumented canal area. (d) The centroid line inside the root canal. (e) Illustrative image showing the shift of the central canal axis of both the mesiobuccal and mesiolingual canals of the mandibular molar. (f) Two-dimensional cross-sectional axial images before instrumentation, after instrumentation, and after superimposition. The black arrow indicates the canal before instrumentation (white) and canal after instrumentation (green) (f)

The apical canal deviation was calculated using the multi-dimensional coordinates of the x, y, and z axes. The geometric center of the apical canal axis was determined at 0.3 mm intervals, both pre- and post-canal preparation, over a 3 mm distance from the apex. The centroid line (Fig. 2e) was created by connecting the coordinates along the vertical axis and apical canal deviation (D) was calculated by the equation [25]:

$$}^}}\,}\,}-}} \right)^}}\,}\,}-}} \right)^}}\,}\,}-}} \right)^}}$$

Following fifteen minutes of ultrasonic rinsing, SEM photographs of the utilized files were taken at ×500 and ×1500 magnification to assess surface wear such as microcracks, blunt cutting edges, and disruption of the cutting edges. The evaluation was determined as described in a previous article [26] with adjustments as follows: Score 1, no wear on the active portion of the blade; Score 2, one to five areas of wear on the active portion of the blade; and Score 3, more than five areas of wear on the active portion of the blade [23].

The analysis was performed by two evaluators, with Cohen’s kappa intra-rater reliability of 0.8 and 0.79 and an inter-rater reliability of 0.79. The investigators were concealed from the instrumentation methodology.

All the statistical analysis was performed using SPSS software (v26.0; IBM, Armonk, NY, USA). Kolmogorov-Smirnov and Shapiro-Wilk tests were used to assess whether the data had a normally distributed population. The Kruskal-Wallis test with post-hoc Dunn’s test was conducted for statistical analysis of the morphometric values of the root canals before instrumentation, torque/force values, canal volume changes, and canal centering ratios. A significant value of 0.05 was set.