Influence of Different Surface Pretreatments on Shear Bond Strength of an Adhesive Resin Cement to Various Zirconia Ceramics

The aim of this in vitro study was to assess the influence of surface pretreatment on shear bond strength (SBS) of an adhesive resin cement (G-CEM Link Force TM, GC Corporation, Tokyo, Japan) to three different yttria-stabilized tetragonal zirconia polycrystalline (Y-TZP) ceramics: (1) Copran Zirconia Monolith HT, COP; (2) Katana ML Zirconia, KAT; and (3) Metoxit Z-CAD HTL Zirconia, MET. In total, 45 cylinders (5 mm in diameter, 1 mm height) for each type of zirconia ceramic were prepared used a computer-aided design and computer-aided manufacturing (CAD/CAM) machine (software CEREC 4.2). Each type of zirconia was subdivided into three groups and each group received a different surface pretreatment; 15 samples were not conditioned as control (groups COP 1, KAT 1, MET 1), 15 samples were air-borne particle abraded with aluminum dioxide particles of 50-μm size at 0.3 MPa for 20 s (groups COP 2, KAT 2, MET 2), and 15 samples were hot-etched with a solution of hydrochloric acid and ferric chloride (groups COP 3, KAT 3, MET 3). After specimen fabrication, the adhesive cement–ceramic interface was analyzed using an SBS test. Subsequently, the adhesive remnant index (ARI) was measured. Data were submitted to statistical analysis. Air-borne particle abraded specimens showed the highest SBS values for COP and KAT groups. For MET, no significant difference was reported between air-borne particle abraded specimens and untreated controls. The lowest values were detected for acid-etched groups. A higher frequency of ARI = “1” and ARI = “2” was reported in control and air-borne particle abraded groups, whereas ARI = “3” was detected in hot-etched groups. No correlation was found between ARI score and shear bond strength. Air-borne particle abrasion is considered the best treatment for Zirconia Copran and Zirconia Katana ML, if it is followed by using dual-curing resin cement.


Introduction
In recent years, the use of aesthetic materials increased more and more in the field of dentistry. This might be the consequence of the development of new fabrication techniques and of the need to satisfy patients' desires. Considering the current restoration systems, computer-aided design and computer-aided manufacturing (CAD/CAM) ensures the production of tooth-colored restorations of both resin and ceramic materials in a short amount of time [1].
Composite resin block materials are available for CAD/CAM technology since 2000 [2]. These blocks are manufactured by industries with standardized criteria of elevated pressure and temperature Table 1. Characteristics of the zirconia ceramics tested in the present report, according to data published by manufacturers and literature sources.

Copran Zirconia Monolith HT
Zirconium dioxide (balance), yttrium oxide (5.15-5.55 In total, 45 cylinders of equal size (5 mm in diameter, 1 mm height) for each tested product were designed with CEREC software 4.2 platform. Samples of each type of zirconia were subdivided into three groups, and each one underwent a different surface treatment (no treatment, ABPA, or hot etching) as shown in the flow chart in Figure 1 and listed below.
After the surface treatments, a liquid primer (G-Multi PRIMER, GC Corporation, Tokyo, Japan) containing methacryloyloxydecyl dihydrogen phosphate (MDP) was applied to all the samples with a microbrush to guarantee the subsequent adhesion of the resin cement to the zirconia ceramics. With an air syringe, light air flow was applied for 5 s, which allowed the solvent (ethanol) to evaporate.
In order to standardize the procedures as much as possible, the same operator applied a dual-cure adhesive luting cement (G-CEM Link Force TM , GC Corporation, Tokyo, Japan), directly from the syringe inside a metallic annulus, with predetermined dimensions (external diameter 5 mm; internal diameter 3 mm; height 1 mm), which acted as a mold ( Figure 2). an air syringe, light air flow was applied for 5 s, which allowed the solvent (ethanol) to evaporate. In order to standardize the procedures as much as possible, the same operator applied a dualcure adhesive luting cement (G-CEM Link Force TM , GC Corporation, Tokyo, Japan), directly from the syringe inside a metallic annulus, with predetermined dimensions (external diameter 5 mm; internal diameter 3 mm; height 1 mm), which acted as a mold ( Figure 2).
The characteristics of the tested cementation system for zirconia are reported in Table 2.  After the elimination of the superficial excess cement with an Explorer 6, the samples were irradiated with light curing for 20 s using a light-emitting diode (LED) lamp (Elipar TM S10 LED Curing Light, 3M ESPE, Neuss, Germany) at a light intensity of 1200 mW/cm 2 , with a wavelength of 430-480 nm, standing with the ferrule at a distance of 1 mm from the cement [23]. The LED power a b c The characteristics of the tested cementation system for zirconia are reported in Table 2. After the elimination of the superficial excess cement with an Explorer 6, the samples were irradiated with light curing for 20 s using a light-emitting diode (LED) lamp (Elipar TM S10 LED Curing Light, 3M ESPE, Neuss, Germany) at a light intensity of 1200 mW/cm 2 , with a wavelength of 430-480 nm, standing with the ferrule at a distance of 1 mm from the cement [23]. The LED power was measured with the radiometer included in the instrument base. After polymerization, metallic annuli were removed, and zirconia discs were stored in saline for 7 d at a temperature of 37 • C [24].

Shear Bond Strength Testing
After storing, each specimen was placed in a universal testing machine (Instron Universal Testing Machine Model 3343, Instron Corporation, Norwood, MA, USA).
During the test, the zirconia specimens were subjected to a force applied with a steel tip, tangentially with respect to the cementation surface, until the failure of the cement [25], as shown in Figure 3. The crosshead speed was set to 1 mm/min [26]. In order to automatically record debonding force Materials 2020, 13, 652 6 of 13 (in newtons), the software Bluehill 2 (Instron Industrial Products, Grove City, Pennsylvania, PA, USA) was used. Data were converted into megapascals (MPa).

Shear Bond Strength Testing
After storing, each specimen was placed in a universal testing machine (Instron Universal Testing Machine Model 3343, Instron Corporation, Norwood, MA, USA).
During the test, the zirconia specimens were subjected to a force applied with a steel tip, tangentially with respect to the cementation surface, until the failure of the cement [25], as shown in Figure 3. The crosshead speed was set to 1 mm/min [26]. In order to automatically record debonding force (in newtons), the software Bluehill 2 (Instron Industrial Products, Grove City, Pennsylvania, PA, USA) was used. Data were converted into megapascals (MPa). After detachment, the external surface of each zirconia's sample, where adhesive was applied, was examined using an optical microscope to determine the proportion of residual adhesive using the adhesive remnant index (ARI) that evaluated the zirconia-cement interface. The ARI scale has a range between 0 and 3: (0) all the cement remained on the zirconia surface; (1) more than half of the cement remained on the zirconia surface; (2) less than half of the cement remained on the zirconia surface; (3) no adhesive remained on the zirconia surface [27].

Statistical Analysis
Statistical analysis was performed with R software (R version 3.1.3, R Development Core Team, R Foundation for Statistical Computing, Wien, Austria). Descriptive statistics, including the mean and the standard deviation, were calculated for each group. The normality of the data was calculated with the Kolmogorov-Smirnov test. Analysis of variance (ANOVA) was applied to establish whether significant differences in debond strength values existed among the groups. Tukey's test was used post hoc. Significance for all statistical tests was set at p < 0.05.
Finally, a frequency analysis of the ARI scores was carried out to assess the presence of differences between the residual adhesive indices of the different groups tested by means of the chisquare test. The level of significance for all tests was set at p < 0.05. b a After detachment, the external surface of each zirconia's sample, where adhesive was applied, was examined using an optical microscope to determine the proportion of residual adhesive using the adhesive remnant index (ARI) that evaluated the zirconia-cement interface. The ARI scale has a range between 0 and 3: (0) all the cement remained on the zirconia surface; (1) more than half of the cement remained on the zirconia surface; (2) less than half of the cement remained on the zirconia surface; (3) no adhesive remained on the zirconia surface [27].

Statistical Analysis
Statistical analysis was performed with R software (R version 3.1.3, R Development Core Team, R Foundation for Statistical Computing, Wien, Austria). Descriptive statistics, including the mean and the standard deviation, were calculated for each group. The normality of the data was calculated with the Kolmogorov-Smirnov test. Analysis of variance (ANOVA) was applied to establish whether significant differences in debond strength values existed among the groups. Tukey's test was used post hoc. Significance for all statistical tests was set at p < 0.05.
Finally, a frequency analysis of the ARI scores was carried out to assess the presence of differences between the residual adhesive indices of the different groups tested by means of the chi-square test. The level of significance for all tests was set at p < 0.05.

Results
Descriptive statistics are presented in Table 3. ANOVA showed the presence of significant differences among the various groups (p < 0.0001). According to post hoc Tukey testing, group COP 2 (Zirconia Copran Monolith HT air-particle abraded) and group KAT 2 (Zirconia Katana ML air-particle abraded) showed significantly higher bond strengths (p < 0.05), as represented in Figure 4. No significant difference was detected between them (p > 0.05). Groups KAT 1, MET 1, and MET 2 showed detachment forces not different from each other (p > 0.05) and significantly lower than groups COP 2 and KAT 2 (p < 0.05) but significantly higher than the other groups (p < 0.05). Groups COP 1, KAT 3, and MET 3 showed detachment forces that were not different from each other (p > 0.05) and significantly higher than group COP 3 (p < 0.05) but significantly lower than the other groups (p < 0.05). The lowest detachment forces were detected by group COP 3 (p < 0.05). Analyzing the effect of pretreatment, significantly higher forces were found in the groups (COP 2, KAT 2, MET 2) subjected to ABPA (p < 0.05), whereas significantly lower values were detected in the groups (COP 3, KAT 3, MET 3) subjected to pretreatment with hydrochloric acid and ferric chloride (p < 0.05). Intermediate values were detected in the control groups without pretreatment (COP 1, KAT 1, MET 1).  ARI scores are presented in Table 4. The chi-square test showed that the control and air-particle abraded groups showed a higher frequency of ARI = "1" and ARI = "2" scores, while the groups subjected to pretreatment with hydrochloric acid and ferric chloride showed a higher frequency of ARI scores = "3" (p < 0.05). As shown in Figure 5, no significant correlation (R 2 < 0.0687) was recorded between adhesion strength (MPa) and adhesive remnant index (ARI Score) in the three different conditions tested (control, ABPA, and hot etching).  ARI scores are presented in Table 4. The chi-square test showed that the control and air-particle abraded groups showed a higher frequency of ARI = "1" and ARI = "2" scores, while the groups subjected to pretreatment with hydrochloric acid and ferric chloride showed a higher frequency of ARI scores = "3" (p < 0.05). As shown in Figure 5, no significant correlation (R 2 < 0.0687) was recorded between adhesion strength (MPa) and adhesive remnant index (ARI Score) in the three different conditions tested (control, ABPA, and hot etching).

Discussion
Zirconia is an innovative material in the panorama of modern dentistry. It combines excellent aesthetic characteristics with many exceptional mechanical properties that make it suitable for the production of restorations in the posterior areas of the oral cavity [28].
In recent years, zirconia-based ceramics became the primary choice for the production of conventional fixed prostheses thanks to their aesthetic performance, their excellent mechanical properties, and their biocompatibility with the oral cavity [29]. The main problem that prevented the use of metal-free restorations is the ceramic having some micro-defects in its structure that over time tend to widen and eventually lead to fracture and failure of the prosthesis. This is true for almost all ceramic materials, but not for zirconia, which can overcome this problem by transforming the tetragonal zirconia grains into monoclinic zirconia grains (t → m), thus allowing the material to significantly increase its strength [30].
Conditio sine qua non to achieve complete clinical success and longevity of the restoration is the realization of a strong bond between the resin-based cement and the zirconium oxide restoration [12]. This is possible through the formation of both chemical bonds and micromechanical connections: the latter allow an increase in the surface roughness of the restoration, while the former allow obtaining a chemical adhesion between the two substrates [31].
Over the years, the pretreatment on zirconium oxide-based surfaces was extensively studied in order to make the bond with cement more effective and lasting. Due to the nature of the type of material, this research is very complex and still unfinished. Some authors showed that ABPA reduces

Discussion
Zirconia is an innovative material in the panorama of modern dentistry. It combines excellent aesthetic characteristics with many exceptional mechanical properties that make it suitable for the production of restorations in the posterior areas of the oral cavity [28].
In recent years, zirconia-based ceramics became the primary choice for the production of conventional fixed prostheses thanks to their aesthetic performance, their excellent mechanical properties, and their biocompatibility with the oral cavity [29]. The main problem that prevented the use of metal-free restorations is the ceramic having some micro-defects in its structure that over time tend to widen and eventually lead to fracture and failure of the prosthesis. This is true for almost all ceramic materials, but not for zirconia, which can overcome this problem by transforming the tetragonal zirconia grains into monoclinic zirconia grains (t → m), thus allowing the material to significantly increase its strength [30].
Conditio sine qua non to achieve complete clinical success and longevity of the restoration is the realization of a strong bond between the resin-based cement and the zirconium oxide restoration [12]. This is possible through the formation of both chemical bonds and micromechanical connections: the latter allow an increase in the surface roughness of the restoration, while the former allow obtaining a chemical adhesion between the two substrates [31].
Over the years, the pretreatment on zirconium oxide-based surfaces was extensively studied in order to make the bond with cement more effective and lasting. Due to the nature of the type of material, this research is very complex and still unfinished. Some authors showed that ABPA reduces the mechanical properties of zirconia [32], creating microfractures and alterations at the surface of the restoration. On the other hand, other research stated that the use of resinous cements containing specific phosphate monomers (10-MDP), combined with the ABPA procedure, seems to produce particularly effective results in guaranteeing stability to the restoration, as well as excellent retention, and minimizing the marginal gaps, thus avoiding the danger of infiltration and, therefore, the appearance of secondary caries [33]. Some authors used a solution based on hydrochloric acid heated to 100 • C to etch the metal arms of Maryland bridges [17], and a similar pretreatment was also proposed for zirconia ceramics. Hydrochloric acid is a highly corrosive substance capable of attacking metals such as iron, steel, and lead. However, the presence of hydrochloric acid alone is not capable of causing corrosion at the level of the zirconium surface, but the presence of oxidizing impurities such as the ferric ion (Fe 3+ ) or the copper ion (Cu 2+ ) leads to the breaking of the protective film of the zirconium oxide, allowing local corrosion and the creation of micro porosity. Hot etching, however, presents some limits; the vapors of the boiling solution must be aspirated by means of large hoods, and the manipulation of these caustic substances by the technician in the laboratory could constitute another important obstacle to the practical realization of this pre-treatment [34]. The acid etching of the restoration negatively influences the interaction between cement and zirconia. The ineffectiveness of this treatment method on zirconia is probably related to the physical properties of the material itself and to its high crystalline content [35]. Currently, ABPA is considered the best treatment for zirconia, which is followed by using auto-curing or dual-curing resin cements [36]. The efficacy of this technique was confirmed by both in vitro and clinical studies, reporting that ABPA with alumina particles at a moderate pressure followed by the application of an MDP-containing primer and of a composite resin luting cement provides long-term adhesion to zirconia ceramics [37][38][39].
According to results reported in literature, one of the purposes of this in vitro study was to confirm the main efficacy of ABPA, compared to hot etching and no pretreatment, on the shear bond strength of an adhesive resin cement to three recent zirconia ceramics. The first null hypothesis of the present report was partially rejected. As demonstrated, the adhesion values of the air-borne particle abraded zirconia were superior (COP 2-Zirconia Copran Monolith HT; KAT 2-Zirconia Katana ML) or equal (MET 2-Zirconia Metoxit Z-CAD HTL) to the untreated zirconia samples (groups COP 1, KAT 1, and MET 1, respectively) and the etched samples (groups COP 3, KAT 3, and MET 3 respectively).
It is of interest to point out that, only for Metoxit Z-CAD HTL, ABPA pretreatment caused an increase in strength bond values not statistically significant if compared to controls. Considering that values for untreated Metoxit Z-CAD HTL were even higher than those of Copran Monolith HT and Katana ML in the same condition, we can assert that the cementation system used in this report was effective as well. Therefore, the outcome regarding Metoxit Z-CAD HTL might be explained considering the chemical composition of this material or, more likely, its surface structure, which might be less responsive to ABPA.
With regard to the ABPA procedure, many studies in the literature described the elimination of abraded zirconia particles before proceeding with cementation. This might be performed by rinsing the samples with water or air/water spray [40,41] or through an ultrasonic bath containing distilled water or ethanol [22,41,42]. On the contrary, no rinsing of air-borne particle abraded zirconia was described by authors, taking into account a risk of alteration of the freshly air-abraded surface with a reduction of the bonding efficacy, despite other studies clearly showing that the cleaning method has little or no effect [43]. In the present report, rinsing was conducted for all the samples through an air/water vaporizer; this procedure seems to have caused no negative effects considering that air-borne particle abraded zirconia ceramics reported debonding forces significantly higher than untreated or hot-etched ones, except for Metoxit Z-CAD HTL, as previously discussed.
The second purpose of this study was to assess the adhesive remnant index (ARI) score in the three different conditions tested (control, ABPA, and hot etching) for each material. The adhesive remnant index (ARI) is a method of defining the adhesion failure site [44]. It is widely reported in orthodontics, even if some authors showed its use in conservative dentistry [45]. The second null hypothesis of the present report was partially rejected. The chi-square test showed that the control groups and the air-borne particle abraded groups showed greater residual adhesive indices (ARI) of "1" and "2", whereas specimens pretreated with hydrochloric acid and ferric chloride showed a higher prevalence of ARI = "3". However, a direct correlation between adhesion strength and adhesive remnant index score was not demonstrated.
Adhesion strength tests were extensively documented, and they were found to be useful to test new materials before clinical use [46]. The limitations of the present report are related to the fact that a limited number of zirconia samples were tested and that an experimental laboratory report cannot simulate intraoral conditions. Moreover, the materials were tested one week after bonding; thus, other storage times and temperatures can have a significant influence on the adhesion properties. Finally, the hydrolytic durability of the bonded specimens was not tested. Therefore, a direct extrapolation of the results of the present in vitro investigation to clinical practice must be carefully conducted. Further in vivo studies and randomized controlled trials (RCTs) are required in order to confirm the results of the present in vitro report.

Conclusions
The aim of this in vitro study was to assess the influence of surface pretreatment on shear bond strength of an adhesive resin cement (G-CEM Link Force TM, GC Corporation, Tokyo, Japan) to three different zirconia ceramics. Adhesion values of the air-borne particle abraded Zirconia Copran and Zirconia Katana ML were superior to the untreated zirconia, whereas they were equal considering air-borne particle abraded Zirconia Metoxit Z-CAD HTL. On the contrary, the hot-etched samples showed the lowest values.
Moreover, both the untreated and the air-borne particle abraded zirconia ceramics showed a higher frequency of ARI = 1 and ARI = 2, whereas hot-etched zirconia ceramics showed a higher frequency of ARI = 3. However, no direct correlation was found between ARI score and shear bond strength.