Evaluation of Shear Bond Strength of Composite to Porcelain According to Surface Treatment

This study evaluated the shear bond strength of porcelain/composite using 40 metal + porcelain + composite cylindrical specimens divided into 4 groups, according to porcelain surface treatment: 1) no treatment, 2) mechanical retentions performed with diamond burs, 3) etching with phosphoric acid+silane, and 4) etching with hydrofluoric acid+silane. After being stored in distilled water at room temperature for one week, the specimens were submitted to a shear force (load) and the data were analyzed statistically (ANOVA). The means (in Mpa) of the groups were: 4.71 (group 1); 4.81 (group 2); 11.76 (group 3); 11.07 (group 4). There were no statistically significant differences between groups 1 and 2 and between groups 3 and 4.


INTRODUCTION
Metal-ceramic crowns are a restorative treatment alternative which combines metal resistance and porcelain esthetics, tending to last for many years if made according to specific technical rules.Factors such as impact and fatigue, occlusal forces, and incompatibility between metal and porcelain physical properties may result in porcelain fracture, frequently of a cohesive nature (1)(2)(3).Due to the inherently brittle nature of ceramic restorative materials, failure under intraoral conditions of metal-ceramic restorations is not uncommon (4).The majority (65%) of the failures have been observed in the anterior region, whereas 35% were in the posterior region.Sixty percent of the failures occurred at the labial, 27% at the buccal, 5% at the incisal, and 8% at the occlusal regions.These fractures were mainly in the maxilla (75%), predominantly at the labial surface (5).Thus, it is necessary to assess the possible causes of fracture, so that the most suitable treatment can be recommended.Depending on the extension of the area to be restored, cost and time available may range from the making of a new prosthe-sis, facet (6), overcasting (7) to a composite repair (8)(9)(10).
According to Kato et al. (11), the bond strength to porcelain was predominantly influenced by the type of conditioning agent rather than the type of luting agent.
Usually, when performing repair with composite on porcelain, the surface must be prepared with burs, sandblasted with aluminum oxide, etched and silaned.The increase in porcelain surface roughness by means of diamond burs and discs contributes to the repair's adhesive resistance, and the use of acids, by creating microretentions, improves adhesion, producing cohesive failures if used with silane agents (12).Research shows that sandblasting with 50-µm aluminum oxide is a better method for preparing the surface than bur performed retentions (13).Nevertheless, sole sandblasting is not sufficient to improve the bond of composite-porcelain interface (4,14).
In one of the first studies on porcelain repair with composite, utilizing a silane agent, it was verified that this technique might be a clinical solution to fractured Braz Dent J (2003) 14(2): 132-135 porcelain repairs (15).Silanization improved repair resistance when hydrofluoric acid was previously utilized (16).However, there is no consensus on the best porcelain treatment method to be applied in different clinical situations.
Thus, aiming at further clarifying this issue, the present study evaluated the porcelain-composite bond strength, employing several methods for surface treatment associated with an adhesive system prior to resin application.

MATERIAL AND METHODS
Forty metal + porcelain + composite specimens were made using a stainless steel matrix, comprised of a metallic base over which a 22.0-mm high X 30-mm diameter cylinder was adapted, presenting a central perforation of 6.5-mm deep by 6.0-mm diameter surface, determining specimen dimensions.The opposite surface had a 5.0-mm diameter screw, utilized for specimen removal following condensation and stabilization of the set during shear tests.The matrix had spacing discs of 6.0 mm diameter by 1.5 mm height, which adapted to the central perforation, allowing thickness standardization of the metal (2 discs; 3.0 mm height X 6.0 mm diameter), porcelain (1 disc; 1.5 mm height X 6.0 mm diameter) and composite (no disc; 1.5 mm height X 6.0 mm diameter).The matrix and spacing disc are shown in Figure 1.
To obtain the wax pattern, 2 discs were positioned in the interior of the matrix, corresponding to porcelain and composite portions.Fused wax was poured and, after cooling, removed with the aid of the screw.The patterns were cast in Ni-Cr alloy (Durabond Universal, São Paulo, SP, Brazil) and polished with aluminum oxide abrasive burs at low rotation until their complete fitting in the interior of the matrix.One of the surfaces was sandblasted with 50-µm aluminum oxide particles (Trijato Odonto Larcon, Maringá, PR, Brazil), after which the metallic discs were cleaned and degreased in pressurized water steam (Vaporeto, Polti, Italy), for 1 min.After drying, these surfaces were kept free from manual contact.Conventional application and opaque firing in paste, in two layers (Opaquer, IPS Classic, Ivoclar), and specimen porcelain was done according to manufacturer instructions.Figure 2 shows the manufacturing phase.
The ceramic surface was made uniform by abra-   The specimens were subsequently stored in distilled water, at room temperature, for 7 days, and then submitted to shear tests in a universal assay machine (Kratos, Dinamômetros Kratos Ltda, São Paulo, SP, Brazil) with a 0.5-mm/min speed.The active tip was applied on the composite-porcelain interface.For the tests, the specimens were placed in the matrix, exposing only the composite portion (Figure 3).
The data obtained were recorded and submitted to analysis of variance (ANOVA) (p<0.05).Individual comparisons were made with the Tukey test (p<0.05).

RESULTS
The means and standard deviations for the 4 groups are reported in Table 1.By comparing the results of groups 1 and 2, one observes that the performance of mechanical retentions on the porcelain surface added little to increase the bond strength of composite with porcelain.Individual comparisons performed by the Tukey test showed statistically significant differences when using conditioning with either phosphoric (group 3) or hydrofluoric acid (group 4), associated to silane agent and adhesive in the comparison to the control group (use of adhesive only) and to group 2 (mechanical retention).However, the differences be-tween the types of acids were not significant, nor were the differences between groups 1 and 2 statistically significant.

DISCUSSION
The clinical success of porcelain repair systems is almost entirely dependent on the integrity of the bond between porcelain and composite resin.The bond is achieved either by chemical or mechanical methods.Etching procedures have been used to facilitate bonding between porcelain and resin, creating a porous surface on the porcelain that leads to a retentive bond (4).
The combination of sandblasting with acids on porcelain silanization has provided the best results.When failure involves only the porcelain, the acidconditioning is effective in increasing micromechanical retention (9); but, it was observed that silane priming considerably improved bond strength to the porcelain of the composite material (17).
Hydrofluoric acid is not necessary to obtain resistant adhesion between composite and porcelain if the silane agent, which reacts with OH groups on the porcelain surface, is used.Silane agents appear to be the essential components for a porcelain repair procedure, by modifying its surface structure, rendering it more reactive with the composite, enabling chemical adhesion in both inorganic/porcelain and organic/composite surfaces.Theoretically, the silane bond should provide a stable repair of the fractured porcelain (18).Lacy et al. (12) observed that when silane was not applied the composite bond strength to porcelain was relatively weak, regardless of the type of surface treatment, with the failure occurring at the interface.The values were higher with silane application after etching.Thus, the findings suggests that bonding to porcelain is mostly chemical and not mechanical (19).The results of the present study cooborate this.Leibrock et al. (20) did not observe statistically significant differences in bond strength following thermocycling and mechanical overloading among the systems which used phosphoric or hydrofluoric acid.Owing to its chemically aggressive nature, hydrofluoric acid should be avoided and substituted by phosphoric acid, which, although aggressive, is not as noxious to tissues but produces the same effects if associated with the silane agent.It is not possible to state whether the lower values found in this study are compatible with the clinical success of a composite repair, but quite probably, the application of silane is capable of contributing to the longevity of this sort of procedure.

Figure 1 .
Figure 1.Stainless steel matrix used for making the specimens.A) Base.B) Cylinder.C) Screw.D) Frontal perforation.E) Spacing disc.

Figure 2 .
Figure 2. Manufacturing phase.F) Wax pattern.G) Cast pattern.H) Metal + porcelain.Figure3.Matrix with the specimen in place for test execution.
sion with a 2214 KG Sorensen diamond bur prior to surface treatment.The specimens were randomly divided into 4 groups of 10 specimens each.Group 1:

Table 1 .
Shear resistance (in MPa) of the 4 analyzed groups.