Changes in Crystal Phase, Morphology, and Flexural Strength of As-Sintered Translucent Monolithic Zirconia Ceramic Modiﬁed by Femtosecond Laser

: Conventional bonding technology suitable for silica-based ceramics is not applicable to zirconia, due to its polycrystalline phase composition, chemical stability, and acid corrosion resistance. The development of an effective treatment to improve its surface roughness and mechanical properties remains an unresolved problem. Therefore, to solve this problem, this in vitro study evaluated the changes in surface morphology and ﬂexural strength of translucent monolithic zirconia surfaces treated with femtosecond laser technology. As-sintered translucent zirconia specimens were subjected to airborne particle abrasion and femtosecond laser treatments, while control group specimens received no treatment. After treatment, the roughness and morphology of the treated zirconia surfaces were examined. The ﬂexural strength and X-ray diffraction of the treated specimens were measured and analyzed. Statistical inferential analysis included one-way analysis of variance at a set signiﬁcance level of 5%. The surface roughness after femtosecond laser treatment was signiﬁcantly improved when compared with the control group and the group that received the airborne particle abrasion treatment ( p < 0.05). In comparison with the airborne particle abrasion group, the ﬂexural strength of the group that received the femtosecond laser treatment was signiﬁcantly improved ( p < 0.05). The femtosecond laser approach using appropriate parameters enhanced the roughness of the zirconia without reducing its ﬂexural strength; therefore, this approach offers potential for the treatment of zirconia surfaces.


Introduction
Zirconia is highly popular in clinical use for dental restorations, due to its exceptional flexural strength, chemical resistance, and good aesthetics [1]. Depending on whether a glass-matrix phase is present or absent, or whether the material contains an organic matrix highly filled with ceramic particles, all-ceramic materials can be classified into three families: (1) glass-matrix ceramics, (2) polycrystalline ceramics, and (3) resin-matrix ceramics [2]. However, unlike other dental ceramic materials, zirconia has a polycrystalline phase composition without a glass phase composition, as well as good chemical stability and acid corrosion resistance. Owing to this, conventional bonding technology suitable for silica-based ceramics is not effective for zirconia. Thus, an important scientific problem in the use of zirconia materials is the effective treatment of zirconia to improve its surface roughness and mechanical properties [3].
Through the use of FSLs, the surfaces of zirconia ceramics could be effectively modified without inducing thermal or mechanical damage [17,18]. While limited research has been conducted on the surface pre-treatment of zirconia ceramics using laser technology, the ideal effect of zirconia modification by laser, in order to improve the bond strength, has not yet been achieved [19]. Therefore, the changes in surface morphology and flexural strength upon applying the FSL technique require further exploration.
This study aims to evaluate the effect of zirconia surface modification using the FSL method. The null hypothesis was that the application of the FSL would not affect the surface morphology and flexural strength of translucent monolithic zirconia ceramic material.

Zirconia Specimen Preparation
A total of 36 disc-shaped fully sintered translucent monolithic zirconia ceramic (UP-CERA, ST, Shenzhen, China) specimens (10 mm diameter × 2 mm thickness) and 36 rectangular fully sintered translucent monolithic zirconia specimens (25 mm long × 4 mm wide × 3 mm thick) were prepared from a dental zirconia blank (in which the content of Y 2 O 3 was 4.5-6%). A precision cutting machine was used to prepare the specimens, which were sintered using a programmable furnace according to the manufacturer's instructions. The temperature of the furnace was heated from room temperature, at a rate of 8 • C/min, to 1200 • C, then increased at a rate of 2 • C/min from 1200 • C to 1450 • C, maintained at 1450 • C for 2 h, and finally cooled at a rate of 10 • C/min to room temperature. Depending on the employed surface pre-treatment method, the 36 disc-shaped fully sintered zirconia specimens were randomly divided into three groups, with each group containing 12 specimens. Disc-shaped specimens were employed for observation of surface topography and assessment of roughness. Rectangular specimens were used for flexural strength testing. In order to avoid potential failure of specimens due to edge defects, a 45 • diagonal angle was introduced on the edges of all the rectangular specimens. Specimen preparation was carried out in accordance with the requirements of the ISO/CD 6872:2015 standard [20]. The calculation of the sample size of this study was carried out with reference to similar studies [21].

Zirconia Surface Treatment
The 72 sintered translucent zirconia specimens were classified into Groups I, II, and III, with each group containing 12 circular and 12 rectangular specimens. Group I was the control, with no treatment (NT) of the surface of the specimens. Group II underwent APA, whereby the surfaces of the specimens were air abraded (Renfert GmbH, Hilzingen, Germany) with 50 µm aluminum oxide (Al 2 O 3 ) particles at a perpendicular distance of 10 mm from the surface under 0.20 MPa of pressure for 20s. Group III underwent FSL treatment, whereby the surfaces of the specimens were microtextured with FSL (Amplitude Systemes, Tangerine laser head, Bordeaux, France) having a wavelength of 1030 nm, pulse width of 400 fs, repetition frequency of 200 kHz, peak power of 5 W, and single-pulse energy of 25 µJ. The FSL ablation was controlled using a three-axis numerically controlled laser galvanometer scanning system (175 mm lens focal length,~80 mm spot diameter, 2000 mm/s light spot scanning speed, 0.1 µm minimum step size along the z-axis, and 10 mm maximum step size; see Figure 1).

Surface Structure and Roughness Assessment
The surface roughness of the specimens was measured using a confocal threedimensional (3D) laser scanning microscope (VK-X210, Osaka, Japan). For each specimen, the surface roughness of three areas was measured under 200× magnification (n = 12/group). The 3D surface roughness parameter R a was calculated using VK Analyzer software. The acquisition R a (µm) value was derived from the average of three analyzed areas, using a simple average-type smoothing filter. The disc-shaped specimens of each group were sputter-coated with gold (108AUTO sputter coater, Cressington, Watford, Figure 1. Schematic of the FSL surface treatment system and laser trajectory during surface treatment of the specimen. The red line indicates the propagation path of the FSL in the system, and the blue line indicates the light from the surface treatment result received by the charge-coupled device, which monitored the real-time processing results as the laser processed the specimen surface.

Surface Structure and Roughness Assessment
The surface roughness of the specimens was measured using a confocal three-dimensional (3D) laser scanning microscope (VK-X210, Osaka, Japan). For each specimen, the surface roughness of three areas was measured under 200× magnification (n = 12/group). The 3D surface roughness parameter Ra was calculated using VK Analyzer software. The acquisition Ra (µm) value was derived from the average of three analyzed areas, using a simple average-type smoothing filter. The disc-shaped specimens of each group were sputter-coated with gold (108AUTO sputter coater, Cressington, Watford, UK), and the surface topographies of the specimens (n = 3/group) were observed using a scanning electron microscope (SEM, SU8010, Tokyo, Japan) under vacuum conditions and secondary electron detector of SE(UL), with 5.0 kV working voltage, at 500× and 10,000× magnifications.

X-Ray Diffraction Analysis (XRD)
The crystalline phases of the specimens were determined using an X-ray powder diffractometer (XRD) system (Bruker D8 ADVANCE, Karlsruhe, Germany), with the following measurement conditions: Cu K-α radiation, tube voltage of 40 kV, tube current of 40 mA, scanning angle range of 2θ = 25-60°, scanning speed of 17.7 s/step, and step size of 0.02°.

Flexural Strength Test (Three-Point Bending Test)
Flexural strength tests were conducted using a universal testing machine (5969R9273, Instron, Norwood, MA, USA) with 1 mm/min crosshead speed and 13 mm span, according to a three-point bending test. During a flexural strength test, the treated surface was placed in contact with the loading stylus (compressive loading zone). The upper limit of the load cell value applied was 2500 N [20]. The fracture loads of the specimens are expressed in Newtons, and their flexural strength was measured in megapascals, using the equation [20] = 3 /2 , where P is the fracture load (N), l is the span (mm), is the specimen width (mm), and b is the specimen height (mm).

Statistical Analysis
Descriptive statistics and inferential statistic measures were employed to analyze the flexural strength (MPa) and surface roughness (Ra). The mean and standard deviation were calculated, and the normality of the data distribution was confirmed by the

X-ray Diffraction Analysis (XRD)
The crystalline phases of the specimens were determined using an X-ray powder diffractometer (XRD) system (Bruker D8 ADVANCE, Karlsruhe, Germany), with the following measurement conditions: Cu K-α radiation, tube voltage of 40 kV, tube current of 40 mA, scanning angle range of 2θ = 25-60 • , scanning speed of 17.7 s/step, and step size of 0.02 • .

Flexural Strength Test (Three-Point Bending Test)
Flexural strength tests were conducted using a universal testing machine (5969R9273, Instron, Norwood, MA, USA) with 1 mm/min crosshead speed and 13 mm span, according to a three-point bending test. During a flexural strength test, the treated surface was placed in contact with the loading stylus (compressive loading zone). The upper limit of the load cell value applied was 2500 N [20]. The fracture loads of the specimens are expressed in Newtons, and their flexural strength was measured in megapascals, using the equation [20] σ = 3Pl/2ωb 2 , where P is the fracture load (N), l is the span (mm), ω is the specimen width (mm), and b is the specimen height (mm).

Statistical Analysis
Descriptive statistics and inferential statistic measures were employed to analyze the flexural strength (MPa) and surface roughness (Ra). The mean and standard deviation were calculated, and the normality of the data distribution was confirmed by the Shapiro-Wilk test. A one-way analysis of variance (ANOVA) and Least Significance Difference (LSD) post hoc tests were performed, in order to analyze the results among all groups (NT, APA, and FSL). All calculations were performed using SPSS 20 statistical software (SPSS Inc., Chicago, IL, USA). The significance level was set at α = 0.05.

Scanning Electron Microscopy Observations
The SEM images shown in Figure 2 present the surface topographies of the specimens that underwent the different surface treatment methods. Specifically, Figure 2a Figure 2b reveals the unit cell-like microstructure of the zirconia sample after heat treatment. Many grain boundaries can be observed, and the arrangement is dense. There are no voids between the grains. Figure 2c, which represents the APA group, shows that the surface was uniformly rough with sharp edges and corners. A few microcracks can be observed in the image shown in Figure 2d. Finally, the image of the FSL group, depicted in Figure 2e, shows that the surface was uniform and flat without deep grooves. Figure 2f shows an irregular structure with rounded edges and a few pores.

Scanning Electron Microscopy Observations
The SEM images shown in Figure 2 present the surface topographies of the s mens that underwent the different surface treatment methods. Specifically, Figure  represents the NT group, where Figure 2a shows that the surface was relatively flat Figure 2b reveals the unit cell-like microstructure of the zirconia sample after treatment. Many grain boundaries can be observed, and the arrangement is dense. T are no voids between the grains. Figure 2c, which represents the APA group, shows the surface was uniformly rough with sharp edges and corners. A few microcracks ca observed in the image shown in Figure 2d. Finally, the image of the FSL group, dep in Figure 2e, shows that the surface was uniform and flat without deep grooves. Figu shows an irregular structure with rounded edges and a few pores.

Surface Roughness Evaluation
The surface roughness results (measured by the Ra amplitude value) are record Table 1. One-way ANOVA analysis showed that there were statistical difference

Surface Roughness Evaluation
The surface roughness results (measured by the R a amplitude value) are recorded in Table 1. One-way ANOVA analysis showed that there were statistical differences between the three groups (p < 0.05; Figure 3a). Compared with the NT group (R a = 0.98 ± 0.18 µm) and the APA group (R a = 1.12 ± 0.28 µm), it was found that the FSL group achieved significantly superior roughness (R a = 1.42 ± 0.16 µm; Table 1). Therefore, the highest roughness of the zirconia surface was due to the FSL treatment (Figure 3a). tween the three groups (p < 0.05; Figure 3a). Compared with the NT group (Ra = 0.98 ± 0.18 µm) and the APA group (Ra = 1.12 ± 0.28 µm), it was found that the FSL group achieved significantly superior roughness (Ra = 1.42 ± 0.16 µm; Table 1). Therefore, the highest roughness of the zirconia surface was due to the FSL treatment (Figure 3a).   Table 2 shows the mean flexural strength values (MPa), along with the standard deviation, for the three groups. The one-way ANOVA analysis indicated a statistical difference among the three groups (p < 0.05; Figure 3b). No significant difference was found between FSL and NT or between APA and NT, but a significant difference was found between FSL and APA (Figure 3b). The effect of the FSL treatment was, therefore, found to be superior to that of the APA treatment (Figure 3b).   The mean difference is significant at the 0.05 level. NT: no treatment group; APA: airborne particle abrasion group; FSL: femtosecond laser group. Table 2 shows the mean flexural strength values (MPa), along with the standard deviation, for the three groups. The one-way ANOVA analysis indicated a statistical difference among the three groups (p < 0.05; Figure 3b). No significant difference was found between FSL and NT or between APA and NT, but a significant difference was found between FSL and APA (Figure 3b). The effect of the FSL treatment was, therefore, found to be superior to that of the APA treatment (Figure 3b). The mean difference is significant at the 0.05 level. NT: no treatment group; APA: airborne particle abrasion group; FSL: femtosecond laser group.

XRD
Through analysis of the XRD pattern ( Figure 4) on zirconia, we found that strong diffraction peaks of the tetragonal phase were identified in both the NT and APA groups. The zirconia surface with no treatment was completely tetragonal-phase (Figure 4a). Compared with the APA group, the diffraction peaks of the tetragonal phase for the NT group were dominant. The tetragonal phase ratio of the surface treated with APA was low, with a small amount of monoclinic phase (about 9.23%) and cubic phase (4.10%), as shown by the small peaks of monoclinic crystal and cubic phases ( Figure 4b); however, the XRD of the sample treated with FSL suggested that there may be an amorphous phase, as it was also possible to form salt crystals on the surface of the ceramic treated by FSL (Figure 4c).
Compared with the APA group, the diffraction peaks of the tetragonal phase f group were dominant. The tetragonal phase ratio of the surface treated with low, with a small amount of monoclinic phase (about 9.23%) and cubic phase ( shown by the small peaks of monoclinic crystal and cubic phases (Figure 4b); the XRD of the sample treated with FSL suggested that there may be an a phase, as it was also possible to form salt crystals on the surface of the ceramic FSL (Figure 4c).   NT (no treatment group), the zirconia surface for NT was completely tetragonal-phase; (b) for APA (airborne particle abrasion group), the tetragonal phase ratio of the surface was low, with a small amount of monoclinic and cubic phases; and (c) for FSL (femtosecond laser group), there may be an amorphous phase, as it was also possible to form salt crystals on zirconia surfaces treated with FSL.

Discussion
The purpose of this study was to evaluate the effect of femtosecond laser treatment on the surface roughness and flexural strength of translucent zirconia. Compared with the no treatment group, the surface roughness of translucent zirconia was significantly increased in the femtosecond laser group, while the flexural strength measured via three-point bending of zirconia modified by the femtosecond laser was found not to be significantly different from that of samples in the no treatment group. Thus, the null hypothesis of the study was partly rejected.
The bending strength and fatigue resistance of translucent zirconia are higher than those of glass ceramics [22]. Thus, translucent zirconia is a material that offers both mechanical strength and aesthetic performance, and its clinical applications are becoming more and more extensive [22]. However, translucent zirconia may be adversely affected by the use of high-pressure airborne abrasion, as microcracks and defects are generated, and flexural strength is reduced [23]. To date, limited studies have evaluated the effect of FSL modification on Y-TZP surfaces and, to the best of our knowledge, no existing study has evaluated the effect of FSL for treating the surface of translucent zirconia.
Studies have reported that the application of FSL forms groove-or pit-like structures on the surface of zirconia specimens, thereby increasing their surface roughness; however, this was accompanied by a reduction in flexural strength [24]. This approach may lead to sub-surface damage to the zirconia due to 50 µm alumina blasting, which may cause microcracks that limit the life of zirconia restorations [25]. In this study, we explored a specific parameter of FSL in the surface modification of translucent zirconia, in order to improve surface roughness without producing significant grooves or pits. Even though no statistical significance was found, the results in this study revealed that surface roughness was improved. As studies incorporating the FSL method continue to improve, FSL may become more suitable as a surface treatment for zirconia materials.
Although the use of APA with Al 2 O 3 particles is popular for the treatment of zirconia surfaces in order to increase the bond strength, studies have shown that the increased roughness imparted by this technique is accompanied by an increased fracture risk, thereby weakening the structure through the introduction of microcracks [26][27][28]. The results of our study demonstrated that FSL-treated translucent zirconia showed significantly higher mean R a values than those in the APA and NT groups, and there was no significant difference in the R a value between the APA and NT groups. These findings are consistent with those of the study of Inokoshi et al. [29], where the surface roughness of highly translucent Y-PSZ modified using Al 2 O 3 APA was not significantly enhanced, except for that of specimens comprising KATANA UTML (Kuraray Noritake, Japan). In contrast, FSL ablation of zirconia ceramic significantly enhanced surface roughness and improved the zirconia ceramic-resin bond strength, due to the presence of groove-like structures [30]. In our study, the translucent zirconia surface obtained following the FSL treatment presented uniform irregular structures without groove-like structures or pits, as confirmed in the SEM image shown in Figure 2e. Surface topography can be modified by femtosecond laser surface treatment and surface roughness of the zirconia can be increased; thus, the bond strength can be improved [31].
The main height parameters for evaluating surface roughness are average roughness (R a ), the root mean square of the height of each point of the contour (Rq), and the ten-point height of microscopic unevenness (R z ). The Ra value can represent the arithmetic mean deviation of the surface roughness profile amplitude parameter. Consequently, the R a value was used in this study to evaluate the surface roughness [32]. Compared with the control group, roughness in the APA group did not increase significantly, which may have been due to the particular cutting texture produced during the processing of the untreated translucent zirconia specimens, which yielded a certain roughness after sintering. The surfaces of the specimens were not polished in this experiment, in order to maintain the original surface morphologies of the final sintered translucent zirconia. On one hand, this allowed for simulation of the surface of a final sintered zirconia crown without polishing; on the other hand, when the surface of a zirconia specimen is highly polished, FSL irradiates a smooth surface, producing reflections, which may affect the treatment of the zirconia surface. In this study, the surfaces of the zirconia specimens were not polished, allowing the FSL to fully exert its plasma effect at a lower energy density. This may also contribute to the discrepancy in the roughness results between APA and FSL treatments reported in this study when compared with those in other studies. Various surface analysis systems, such as SEM and Atomic Force Microscopy (AFM), are useful for qualitative analysis, but three-dimensional microscopy was used for quantitative evaluation of the surface roughness variation.
The results of XRD showed that the monoclinic content of the APA group was increased, which indicated that aluminum airborne abrasion may lead to t-m phase transformation. At present, there is controversy about the influence of sandblasting on mechanical properties [33]. Some scholars have reported that t-m phase transformation can increase volume and produce protective residual compressive stress, thus preventing the further expansion of microcracks and, consequently, leading to enhanced mechanical strength. This is called the phase transformation toughening mechanism [34]. However, other scholars have indicated that small grain size (within 200 nm) can have a negative impact on the phase transformation and toughening mechanism, consequently reducing the mechanical strength of translucent ZrO 2 [35]. In this experiment, the XRD results indicated that the sintered translucent zirconia specimen appeared to be amorphized after the FSL treatment, which may have been caused by the phenomenon of "avalanche ionization" during interaction of the high-power and high-repetition FSL with the zirconia specimen surface, resulting in high-speed motion [36]. The hypothesis is that the plasma carries a certain element, which is deposited on the surface of the specimen to form a coating, resulting in amorphization [37]. It is also possible to form salt crystals on the surface of the ceramic; however, the interfacial topography between the "amorphized" zirconia layer after FSL and the substrate was not investigated, and further experimental verification is required.
Air particle abrasion may lead to sub-surface damage of the zirconia surface by 50 µm alumina, which may lead to microcracks that limit the life of zirconia restorations [38]. It has been reported that impact-induced defects were observed on zirconia surfaces modified by APA treatment; thus, the longevity of APA-treated zirconia ceramic prostheses may be shortened [39]. Consistent with previous studies, in our study, the surface treated with APA presented a number of microcracks and defects as revealed in Figure 2d, which may lead to a reduction in the flexural strength of the zirconia specimens. In contrast, in a study reported by Wang et al. [40], APA enhanced the flexural strength of zirconia, regardless of the particle size, air pressure, or blasting time. Song et al. [41] reported that flexural strength was significantly higher in the group of air-abraded zirconia specimens than that in the group without any treatment. The content of the monoclinic phase of the lower zirconia surface determines the mechanical behavior of the zirconia specimens, as this is where tensile stress is dispersed. Furthermore, this study indicated that APA of the inner surface of zirconia specimens, in order to improve their bonding performance, might also enhance their fracture strength [40]. In this experiment, we applied 50 µm Al 2 O 3 particles to treat the surface of the translucent zirconia specimen, in order to reduce tetragonal-monoclinic phase transformation.
There are two main methods for testing the bending strength of ceramics: uniaxial bending and biaxial bending. In uniaxial bending tests, a cuboid specimen is supported by two points and loaded vertically at one point (i.e., a three-point bending test) or two points (i.e., a four-point bending test). In a biaxial bending test, a thin disk is supported by a ring or three balls close to it, and a load is applied through a ball or a piston in its central area, or a smaller ring in its center. The above methods have been recognized in international standards [20]. Therefore, a three-point bending test was used to evaluate the used methods after surface treatment in this study. New possible methods for the mechanical analysis of materials have been reported as a future perspective for classic dynamometer systems, such as Dynamic Mechanical Analysis (DMA) and Brillouin's micro-spectroscopy [42]. Importantly, the flexural strength results presented herein were consistent with the observed SEM images. However, further investigations are required to evaluate the longterm stability of zirconia treated using different methods. It should also be noted that, in the experimental setup, line-patterning of zirconia surfaces was achieved; thus, tuning the FSL parameters should allow for independent variation of the pattern depth, overall roughness, and surface finish. More specifically, increasing both the fluence and the number of pulses will allow for deeper patterning, with the maximum achievable depth being 1 µm. However, increasing the number of pulses can have a detrimental effect on the quality of the lines produced, and surface damage can occur (e.g., intergranular cracking, open porosity, and nanodroplet formation), depending on the FSL parameters employed [43]. In future experiments, our research team will try to design an integrated processing device featuring a femtosecond laser. After the zirconia restorations are sintered, the dental technician can hold the working end of the device to modify the tissue surface of dental zirconia restorations, in order to achieve a clean, efficient, and damage-free effect.
One limitation of this study is that the long-term effects of FSL modification technology on the flexural strength of zirconia were not investigated. Therefore, an evaluation of resinzirconia bond strength and durability using different surface modification methods will be reported in a future article.

1.
Femtosecond laser technology offers potential for zirconia surface treatment.

2.
Through the employment of appropriate parameters, femtosecond laser treatment can be used to modify the surface of zirconia, in order to enhance its roughness without decreasing flexural strength. Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.