Comparative evaluation of 3 microbond strength tests using 4 adhesive systems: Mechanical, finite element, and failure analysis.

STATEMENT OF PROBLEM: Bond strength (BS) values from in vitro studies are useful when dentists are selecting an adhesive system, but there is no ideal measuring method. PURPOSE: The purpose of this in vitro study was to investigate the influence of the evaluation method in the BS between dentin and composite resin. MATERIAL AND METHODS: Molars with exposed superficial dentin (N=240) were divided into 3 groups according to the test: microtensile (µTBS), microshear (µSBS), and micropush-out (µPBS). Each one was subdivided into 4 groups according to the adhesive system: total etch, 3- and 2-step; and self-etch, 2- and 1-step). For the µPBS test, a conical cavity was prepared and restored with composite resin. An occlusal slice (1.5 mm in thickness) was obtained from each tooth. For the µSBS test, a composite resin cylinder (1 mm in diameter) was built on the dentin surface of each tooth. For the µTBS test, a 2-increment composite resin cylinder was built on the dentin surface, and beams with a sectional area of 0.5 mm2 were obtained. Each subgroup was divided into 2 (n=10) as the specimens were tested after 7 days and 1 year of water storage. The specimens were submitted to load, and the failure recorded in units of megapascals. Original BS values from the µTBS and µSBS tests were normalized for the area from µPBS specimens. Original and normalized results were submitted to a 3-way ANOVA (α=.05). The correlation among mechanical results, stress distribution, and failure pattern was investigated. RESULTS: Significant differences (P<.05) were found among the adhesive systems and methods within both the original and normalized data but not between the storage times (P>.05). Within the 7 days of storage, the original BS values from µTBS were significantly higher (P<.001) than those from µPBS and µSBS. After 1 year, µSBS presented significantly lower results (P<.001). However, after the normalization for area, the BS values of the µTBS and µPBS tests were similar, and both were higher (P<.001) than that of µSBS in both storage times. In the µSBS and µTBS specimens, cohesive and adhesive failures were observed, whereas µPBS presented 100% of adhesive failures. The failure modes were compatible with the stress distribution. CONCLUSIONS: The storage time did not affect the results, but differences were found among the adhesives and methods. For comparisons of bond strength from tests with different bonding areas, the normalization for area seemed essential. The microshear bond test should not be used for bond strength evaluation, and the microtensile test needs improvement to enable reliable results regarding stress concentration and failure mode. The micropush-out test may be considered more reliable than the microtensile in the bond strength investigation, as demonstrated by the uniform stress concentration and adhesive failure pattern.

Crown fracture: Failure load, stress distribution, and fractographic analysis.

STATEMENT OF PROBLEM: The outcomes from load-to-failure tests may not be applicable to clinical situations. PURPOSE: The purpose of this study was to critically evaluate the efficacy of load-to-failure tests in the investigation of the fracture load and pattern of metal-free crowns. MATERIAL AND METHODS: Four groups were formed from 128 bovine roots restored with metal posts, resin cores, and feldspathic, leucite, or lithium disilicate ceramic systems or polymer crowns. Each group was divided into 4 (n=8) according to the cement: zinc phosphate, self-adhesive resin, autopolymerizing resin, and glass ionomer. Mean fracture loads from compressive tests were submitted to ANOVA and Tukey HSD test. Finite element and fractographic analyses were performed and associated with the fracture load and pattern. RESULTS: Significantly higher fracture load values were obtained for the lithium disilicate ceramic, but finite element and fractographic analyses showed that the cement effect could not be determined. The finite element analysis showed the cement likely affected the fracture pattern, confirmed that stresses in the cements were little affected by the crown materials, and found that the stressed conditions were lowest in the lithium disilicate compared with other crowns for all cement combinations. The stressed conditions in the crowns depended more on the adhesive properties than on the elastic modulus of the cement materials. The level of the stressed condition in the crowns at the occlusal surface was about the same or higher than along their cement interface, consistent with the fractography, which indicated fractures starting at the load point. Higher stress levels in the crowns corresponded with a lower number of catastrophic fractures, and higher stresses in the cements seemed to reduce the number of catastrophic fracture patterns. The highest stressed conditions occurred along the occlusal surface for crown materials with a low elastic modulus or in combination with adhesive cements. CONCLUSIONS: The method used was not appropriate either for investigating the crowns' fracture load and pattern or for stating the role of the cements within the crown-cement-tooth interaction.