Shear Bond Strength of Repaired CAD/CAM Resin-Based Composite Materials Submitted to Er:YAG Laser Treatments at Different Powers

: This study aimed to investigate the effects of different powers of Er:YAG laser irradiation on the shear bond strength (SBS) of repaired CAD/CAM resin-based composite materials. A total of 180 CAD/CAM resin-based composite specimens (5 × 5 × 2 mm) were obtained (Shofu Block HC—SB; Grandio Blocs—GB). They were allocated into six groups according to surface pretreatment methods: no surface pretreatment (control), hydroﬂuoric acid (HF), diamond bur and 3 W, 5 W, and 7 W Er:YAG lasers (20 Hz) (n = 15). Silane and universal adhesive were applied. The repair procedure was completed with nano-ceramic composite resin (Ceram-X Sphere TEC-One). The samples were thermocycled for 10,000 cycles (5–55 ◦ C). The SBS was evaluated with a universal test machine (1 mm/min). A 3D optic proﬁlometer was used to assess the surface topography. Statistical analysis was performed with a two-way ANOVA and Bonferroni tests ( p < 0.05). For SB samples, HF and diamond bur caused signiﬁcantly higher SBSs than 3 W and 7 W lasers, while for GB samples, they led to a signiﬁcantly higher SBS than all laser treatments. For SB samples, the 5 W laser led to the highest SBS, while for GB samples, the 7 W laser caused the highest SBS ( p < 0.05). For both blocks, adhesive failure was more common for the 3 W laser, and a decrease in adhesive failures and an increase in mixed failures were observed with increasing laser irradiation. The 3D optic proﬁlometer revealed that smoother surfaces were obtained with the 3 W laser than other laser irradiation at different powers. Pretreatment with increasing Er:YAG laser powers led to similar bond strengths to hydroﬂuoric-acid-and diamond-bur-treated CAD/CAM resin-based composite materials. A more powerful laser treatment is required to provide a higher bond strength for restorative materials containing a higher inorganic ceramic content.


Introduction
With the rise in popularity of computer-aided design and computer-aided manufacturing (CAD/CAM), recent technology offers stable esthetic materials, minor adaptation to restorations and less time-consuming manufacturing of restorations [1]. In recent years, CAD/CAM resin-based composite materials have been preferred since they are a more easily machinable material compared to ceramic restorative materials. In addition, these materials have been shown to be less abrasive to opposing teeth [2]. They contain a polymeric matrix and reinforcing fillers with less brittleness and hardness. Their elastic moduli were found to be similar to those of enamel and dentin. In addition, they have higher levels of inorganic content and a higher degree of conversion of the resin matrix than conventional composites. These materials also may be advantageous in terms of intraoral reparability [3,4].
Defective dental restorations are frequently encountered in general dental practice. Repairing is preferred due to the less costly procedure which also preserves the tooth tissue [5]. An important advantage is that repairing a partially defective restoration prolongs the tooth retention time [6]. Removal of indirect restorations itself is usually not possible without harmful effects for the tooth or to the restoration [7]. Repair is a more low-cost procedure and preserves the tooth tissue better compared to replacement of the restoration [8].
The defective surface of CAD/CAM resin-based composites should be pretreated before the application of repair materials [9]. Various surface pretreatment methods such as chemical etching with hydrofluoric acid (HF), tribochemical silica coatings, air abrasion, roughening with diamond burs or combinations of any of these techniques are employed to roughen the surface to obtain a better adhesion [7,9,10]. Recently, an alternative approach involves the use of lasers such as Er:YAG lasers [5], Nd:YAG lasers [11] and femtosecond lasers [12] for the surface pretreatment of indirect restorations. Laser methods are considered to be more conservative than other mechanical surface pretreatment methods [13]. Er:YAG laser treatment is considered highly promising because its wavelength corresponds to the absorption peak of water. This method can create microretentive surfaces and increase the bond strength of dental restorative materials [14]. However, so far, there is no consensus as to the best surface pretreatment method for the optimum repair strength of CAD/CAM materials. There are limited studies investigating the shear bond strength (SBS) of repaired CAD/CAM resin-based composite materials using Er:YAG laser irradiation at different powers and comparing its effectiveness to other surface pretreatment methods in the literature. Therefore, the aim of this study was to investigate the effects of different irradiation powers of Er:YAG lasers on the SBS of repaired CAD/CAM resin-based composite materials.
The null research hypothesis was that there would be no difference in the SBS of repaired CAD/CAM resin-based composite materials after Er:YAG laser treatment at different irradiation powers and other different surface pretreatments.

Specimen Preparation
The composition, brands and batch numbers of all the tested materials in this study are shown in Table 1.
Two different types of prefabricated CAD/CAM resin-based composite materials were used: Grandio Blocs (Voco GmbH, Cuxhoven, Germany) (GB) and Shofu Block HC (Shofu Inc., Kyoto, Japan) (SB). A total of 180 CAD/CAM resin-based composite specimens were prepared (5 × 5 × 2 mm) with a low-speed cutting saw under water cooling (Mecatome T180 Presi Gmbh, Hagen, Germany) (N = 90). CAD/CAM specimen thicknesses were confirmed to be 2.0 ± 0.1 mm with a digital micrometer (Asimeto, Digital Caliper, SBSimpson, Ontario, Canada). Then, the specimens were immersed in a selfcured acrylic resin (Imicryl, Konya, Turkey) and the repair surface remained uncovered. The specimens were sequentially polished with a polishing device (Minitech 233, Presi Grenoble, France) under water cooling by using #600, #800, #1000 and #1200 grit silicon carbide abrasive papers for 10 s in order to obtain standardized and smooth surfaces [15]. Any debris formed due to silicon carbide paper polishing was cleaned ultrasonically and air-dried. According to the surface pretreatment methods, each CAD/CAM specimen was randomly allocated into six groups (n = 15). A single operator completed all the surface pretreatment procedures.
No surface pretreatment (control): No surface pretreatment was applied. This group served as a control.
Hydrofluoric acid (HF): Hydrofluoric acid (9%) (Ultradent Porcelain Etch, Ultradent Products Inc., South Jordan, UT, USA) was used on the surfaces for 60 s and then rinsed with water for 20 s according to the manufacturer's instructions. The specimens were air-dried [16]. Diamond bur: The surfaces were roughened for 5 s using a coarse fissure (green band, 125-150 µm) diamond bur (Le Blond A&M Instruments, College Station, TX, USA) under water cooling at 200,000 rpm with a high-speed hand piece through unidirectional movements [17]. The bur was changed after every five specimens. The specimens were cleaned by water spray rinsing for 30 s, followed by air-drying.
Laser irradiation (3 W): An Er:YAG laser (Fotona; AT Fidelis, Ljubljana, Slovenia) was used (wavelength: 2940 nm, power setting: 150 mJ, repetition rate: 20 Hz) at an average power output of 3 W. A non-contact hand piece (H02) was applied perpendicular to the surface of the specimen with a cylindrical sapphire optical fiber tip (1.3 mm in diameter and 8 mm in length). The surfaces were irradiated with sweeping movements under water cooling for 20 s at a distance of 1 mm (super short pulse mode) and air-dried for 20 s.
Laser irradiation (5 W): An Er:YAG laser (Fotona; AT Fidelis, Ljubljana, Slovenia) was used (wavelength: 2940 nm, power setting: 250 mJ, repetition rate: 20 Hz) at an average power output of 5 W. A non-contact hand piece was applied as mentioned above.
Laser irradiation (7 W): An Er:YAG laser (Fotona; AT Fidelis, Ljubljana, Slovenia) was used (wavelength: 2940 nm, power setting: 350 mJ, repetition rate: 20 Hz) at an average power output of 7 W. A non-contact hand piece was applied as mentioned above.
After surface pretreatment procedures, the silane agent was applied (Monobond Plus, Ivoclar Vivadent, AG, Lichtenstein) for 60 s using a microbrush and allowed to air-dry for 30 s in accordance with the manufacturer's instructions. After that, a universal adhesive system (3M Single Bond Universal Adhesive, 3M ESPE, St. Paul, MN, USA) was used for 20 s with a microbrush and polymerized with a light-emitting diode light curing unit (LED LCU) (Valo, Ultradent, South Jordan, UT, USA) (irradiance of 1000 mW/cm 2 ) for 10 s in accordance with the manufacturer's instructions.
The surfaces were repaired with a nano-ceramic composite resin (Ceram X SphereTEC One, Dentsply, New York, NY, USA) using cylinder-shaped teflon molds (diameter: 2.4 mm, height: 2 mm). Using the mold, all composite resins were placed and polymerized with an LED LCU (Valo, Ultradent, South Jordan, UT, USA) (1000 mW/cm 2 ) for 10 s according to the manufacturer's instructions. A radiometer (SDI LED, Radiometer, SDI, Australia) was used to check the light intensity. All specimens were kept in distilled water at 37 • C for 24 h.

SBS Test
Notch-edged SBSs of all specimens were determined via a universal test machine (Shimadzu Corp., Kyoto, Japan) until fracture (crosshead speed: 1 mm/min) [18]. The SBSs (MPa) were calculated by dividing the maximum force recorded at fracture by the bonded surface area.

Failure Mode Analysis
After fracture, failure mode analyses were performed using a stereomicroscope at 15× magnification. An 'adhesive failure' mode was noted if failure occurred along the junction of the CAD/CAM material and the composite resin. An 'cohesive failure' mode was considered if the fracture occurred in the CAD/CAM material (restorative material) or composite resin (repair material). A 'mix failure' mode was considered if the fracture occurred in both the CAD/CAM material (restorative material) and/or the composite resins (repair material) and along the junction of these materials.

Statistical Analysis
Based on the literature, the specimen size was determined with a power analysis [12]. In the current study, a minimum of 15 specimens were required for a medium effect size (d = 0.50), with 95% power and a 5% type 1 error rate.
A software program (SPSS 22.0 Windows, SPSS Inc., Chicago, IL, USA) was used for statistical analyses. For the mean SBS data, the normality of variables was first analyzed using Shapiro-Wilk test and the homogeneity of variances was analyzed using Levene's test. The data were normally distributed. To compare the group differences, a two-way ANOVA test was conducted. Bonferroni's test was used for all pairwise comparisons. Statistical significance was detected at a confidence level of 0.05 in all analyses.

Surface Topography Analysis
For each group, one specimen was subjected to surface pretreatment to evaluate the three-dimensional (3D) surface topography with an optic profilometer (Nanomap 1000 WLI, AEP Technology, Saratoga, CA, USA). The scan range was adjusted to 232 mm, the vertical dynamic range was adjusted to 500 µm and the stylus loading force was set to 12 mg. A color scale and graphics were used for interpretation of the images. Different values are represented with different colors. Negative values indicate troughs while positive values indicate peaks. Table 2 represents the mean SBS values and standard deviations of all tested groups. A Bonferroni test showed that when comparing the surface pretreatment methods, for both CAD/CAM materials, the groups treated with HF and bur had significantly higher SBSs than the ones with no treatment (control) (p < 0.05). In addition, when compared to the group with no treatment, for SB, the groups treated with a 5 W laser showed significantly higher SBSs; whereas, for GB, the groups treated with a 7 W laser revealed significantly higher SBSs (p < 0.05).

SBS Analysis
When comparing the CAD/CAM materials, GB showed significantly higher SBSs than SB for non-treated samples and those treated with HF and a 7 W laser (p < 0.05). SB showed significantly higher SBSs than GB for samples treated with bur and 3 W and 5 W lasers (p < 0.05).
For SB, the groups treated with HF and bur revealed significantly higher SBSs than the ones treated with 3 W and 7 W lasers, while for GB, the groups treated with HF and bur showed significantly higher SBSs than the ones treated with 3 W, 5 W and 7 W lasers (p < 0.05). For SB, the groups treated with a 5 W laser exhibited similar SBSs to the ones treated with HF (p > 0.05).
For SB, the groups treated with bur showed significantly higher SBSs than the ones treated with other surface pretreatments, while for GB, the groups treated with HF showed significantly higher SBSs than the ones treated with other surface pretreatments (p < 0.05).
With regard to the Er:YAG laser groups, for SB, the groups treated with a 5 W laser showed significantly higher SBS values, while for GB, the groups treated with a 7 W laser showed significantly higher SBS values (p < 0.05). Figure 1 and Table 3 show the failure modes after SBS testing for all tested groups. For SB, the predominant mode of failure was adhesive-type failure, except for the specimens treated with bur. Non-pretreated specimens exhibited the highest frequency of adhesive-type failure while the specimens treated with bur had the lowest frequency of adhesive-type failure. The specimens treated with bur had the highest rate cohesive failure in restorative materials (60%). The specimens treated with a 7 W laser showed adhesive-type (46.66%) and mixed-type failures in the repair material (46.66%) at equal rates. Cohesive failure in the repair material was observed for specimens treated with bur. Mixed-type failure in the restorative material was observed for the specimens treated with HF and a 5 W laser.

Failure Mode Analysis
For GB, the predominant modes of failure were adhesive-type and mixed-type failure in the repair material. The specimens treated with 3 W and 5 W lasers showed the highest rate of adhesive-type failures, while the specimens treated with bur and HF exhibited the highest rate of mixed-type failure in the repair material. Cohesive failure in the repair material was observed for specimens treated with HF, while cohesive failure in the restorative material was found for non-pretreated specimens. Mixed-type failure in the restorative material was observed for the specimens treated with HF and bur.

Surface Topography Analysis
For both CAD/CAM resin-based composite materials, the non-treated groups (control) presented smooth surfaces (Figure 2A,a,G,g). The groups treated with HF showed surface irregularities with peaks ( Figure 2B,b,H,h), while the groups treated with bur showed deep parallel valleys ( Figure 2C,c,I,i). In regard to the Er:YAG laser groups, for both CAD/CAM resin-based composite materials, the groups treated with a 3 W laser exhibited a less rough surface ( Figure 2D,d,J,j). Greater microporosities were observed with 5 W laser treatment for SB ( Figure 2E,e), while micro-scratches were seen for GB ( Figure 2K,k). For GB, the 7 W laser caused more irregularities with peaks and valley ( Figure 2L,l).

Discussion
In the present study, the effects of different powers of Er:YAG laser irradiation and other different surface pretreatments on the SBS of repaired CAD/CAM resin-based composite materials were investigated. The null hypothesis, which proposed that there would be no difference in SBSs for repaired CAD/CAM resin-based composite materials after Er:YAG laser irradiation at different powers and other different surface pretreatment methods, was rejected in accordance with the results of the present study.
In clinical practice, for CAD/CAM resin-based composite restorations, fractures may occur during the milling process, along with internal defects of the block or parafunctional habits [19]. If premature occlusal contacts are present, the formation of cracks may be observed within the block structure and may lead to fracture and failures in the restoration. Thus, intraoral repair procedures could be needed [10]. Aging is a result of exposure of dental materials to different temperatures, mechanical forces and chemical changes in the oral environment [19]. These aging processes may cause alteration of the restorative material and will also affect the SBS of repaired restorative materials [20]. While evaluating dental materials in in vitro conditions, the thermocycling test method is assumed to be a reliable and important method for simulating intraoral conditions [21]. In the present study, after a repair procedure, to investigate the long-term stability of the adhesion, the specimens were submitted to 10,000 thermocycles (5-55 • C) for aging, which corresponds to one year of clinical functioning [22]. In addition, the aging procedure was not performed before the repair procedure because we intended to examine the shear bond strength of freshly repaired CAD/CAM restorations, since, in clinical practice, fractures can occur on new restorations right after cementation due to premature contacts, etc.
The repair procedure includes surface pretreatment of the restoration and silane and adhesive system applications [10]. However, there is not any consensus on the most efficient protocol for a successful repair yet. In the present study, two different CAD/CAM resin-based composite materials (Shofu Block HC (SB) and Grandio Blocs (GB)) were used. They have ceramic fillers within a dimethacrylate matrix. However, the ceramic filler contents and filler types of these materials differ. In addition, a nanoceramic composite resin was used as the repair material since this material has a similar structure to these CAD/CAM materials.
Adhesion of indirect restorative materials occurs through chemical bonding on the restoration surface and micromechanical interlocking [23]. Universal adhesive systems have a multi-purpose formulation that may adhere to metal, ceramic or composite restorative materials [24]. Inorganic components of the CAD/CAM blocks could be primed with silane or phosphoric monomers of this adhesive system [25]. Silane agents form a chemical bond between inorganic and organic surfaces, increase the wettability of the surfaces and thereby promote adhesion [7]. Yoshihara et al. [26] reported that separate silane application was more effective than incorporating a silane agent into universal adhesives. Therefore, in this study, a universal adhesive system and separate silane application were employed.
The intention of the surface pretreatment strategies is to provide a higher surface energy for a better wettability and adequate surface roughness [27]. HF is an influential agent and is often preferred in ceramic repairs [28]; it enhances micro-retention and reveals hydroxyl groups, which favors chemical bonding with monomers [29]. Gupta et al. compared 10% HF, 37% phosphoric acid, 7% maleic acid and 30% citric acid pretreatments on aged composite resins. They reported that HF treatment led to the highest bond strength [30]. Diamond burs are used as a surface pretreatment method for the repair of nanoceramic materials to produce mechanical retentions that are filled with the adhesive system [31]. Stasser et al. indicated that a strong roughness could be achieved by grinding with a coarse-grained bur [32]. In addition, Ataol et al., who investigated the effects of different surface pretreatments (Er/Cr:YSGG laser, HF etching and alumina blasting combined with silane) on the repair bond strength of ceramic CAD/CAM materials, indicated that the use of HF etching followed by silane application led to higher bond strength values [33]. In the present study, for both CAD/CAM materials, hydroflouric acid and bur caused significantly higher SBSs than intreated surfaces. This is in line with the results of 3D optic profilometer images, which show that peaks and valleys obtained with HF and bur are consistent with the bond strength values. The results obtained are opposite to those ofŞişmanoglu et al. They evaluated the repaired bond strength of different CAD/CAM resin-based composites with different surface pretreatment methods (HF etching, tribochemical silica coating and air abrasion) and reported that SB treated with HF showed a similar bond strength to the samples with no surface pretreatment [16]. The difference in results could be due to the absence of a separate silane application and the different repair material they used in their study (Filtek Ultimate Universal).
HF etching is one of the most preferred surface pretreatment techniques for repair restorations [34]. Strasser et al. indicated that HF etching increases the surface roughness and surface energy, thereby improving adhesion by reacting with the glass component in CAD/CAM resin-based composite blocks [32]. Hence, the effect of HF is mainly caused by the filler particle composition of the restorative material. Both CAD/CAM resin-based composites used in this study contain inorganic glass fillers. GB has a nanohybrid-based structure with an 86% inorganic content, while SB has a lower inorganic filler ratio (61%). In the present study, for GB, significantly higher bond strength values were determined when the blocks were treated with HF, while in SB, the highest bond strength results were observed when the blocks were treated with a bur. These results are in line with the comparison of the CAD/CAM resin-based materials. For bur-treated groups, SB showed higher bond strength values, while for HF-treated groups, GB showed higher bond strengths. These findings could be attributed to the higher glass content in GB. In addition, SB contains relatively large and spherical fillers, as well as nano-fillers. Diamond bur can remove these particles from the surface and cause roughness, leading to holes [35].
Laser pretreatment uses a laser beam for surface roughening with ablation and melting. However, during laser treatment, considerable local temperature rises can be harmful to the surfaces of restorative materials; therefore, appropriate irradiation parameters should be chosen [13]. Thus, in this study, three different laser powers (3 W/150 mJ, 5 W/250 mJ and 7 W/350 mJ) were used at a 20 Hz frequency for 20 s [5,36,37]. Erdemir et al., who evaluated the effect of different surface pretreatment methods (tribochemical silica coating, HF etching, 6 W Er:YAG laser and diamond bur) on the repaired bond strength of lithiumdisilicate-reinforced CAD/CAM ceramic materials, reported that 6 W laser irradiation produced lower bond strengths than diamond bur [5]. In the present study, for SB, HF and bur caused significantly higher SBSs than 3 W and 7 W lasers, while for GB, these surface pretreatment methods led to significantly higher SBSs than all laser treatments. In addition, in the present study, when comparing laser surface pretreatments at different powers, different findings in SBSs were obtained due to the differences in composition of the CAD/CAM blocks. With regard to the laser treatments at different powers, for SB, 5 W-laser-treated groups showed significantly higher SBSs, while for GB, 7 W-laser-treated samples exhibited significantly higher SBSs than laser treatments at different powers. When comparing CAD/CAM resin-based materials, for 3 W and 5 W laser groups, SB exhibited higher bond strengths, while for 7 W laser groups, GB showed higher bond strengths. These results could be clarified by the fact that a more powerful laser treatment could effectively roughen the surface of GB due to its higher inorganic ceramic content. This is in line with the results of 3D optic profilometer images, which showed greater microporosities with the 5 W laser for SB and with the 7 W laser for GB. In addition, in the current study, for both CAD/CAM materials, smoother surface appearances were obtained with the 3 W laser than other Er:YAG laser irradiations at different powers, consistent with the lower bond strength values.
An adhesive failure mode is indicative of a lower bond strength, while the mixed mode failure is frequently correlated with a higher bond strength [38]. In this study, stereomicroscope observations highlighted that while adhesive failure was more common for both blocks treated with the 3 W laser, a decrease in adhesive failures and an increase in mixed failures were observed with increasing laser irradiation. More cohesive failures can be attributed to the repair strength being closer to the fracture strength of the restorative materials used [39,40]. For SB treated with bur, higher amounts of cohesive failures were observed, which is consistent with the higher repair bond strengths. In addition, for GB treated with HF, mixed failures in addition to cohesive failures were detected, indicating that acid application could increase surface adhesion and provide locking.
Regarding the limitations of this study, the Er:YAG laser was used at a 20 Hz repetition rate for 20 s to pretreat the surfaces. Therefore, future studies should investigate the impact of various pulse frequencies, durations and output powers of the Er:YAG laser on the bond strength of repaired CAD/CAM materials.

Conclusions
Despite the limitations of this study, it can be concluded that pretreatment with increasing Er:YAG laser powers led to similar bond strengths to hydrofluoric-acid-and diamond-bur-treated CAD/CAM resin-based composite materials. In addition, a more powerful laser treatment is required to provide a higher bond strength for restorative materials containing higher inorganic ceramic contents. The results of 3D optic profilometer images confirmed the bond strength values. Further in vitro and in vivo studies are essential to confirm the obtained results. Data Availability Statement: Data is unavailable due to privacy.