Silica-Based Infiltrations for Enhanced Zirconia-Resin Interface Toughness

Abstract

Novel silica-based infiltrations on the surface of zirconia have the potential to improve their bondability, allowing for the etching/silane adhesive bonding technique. Nonetheless, adhesively bonded joints are subject to mixed tensile and shear stresses when the restoration is in occlusal service. Thus, we aimed to investigate the effect of 2 novel silica-based infiltrations on the interfacial toughness of adhesively bonded zirconia using the Brazil nut method, which allows for controlled types of stresses to be applied at the interfaces. In total, 150 3Y-TZP (In-Ceram YZ; Vita) Brazil nuts were machined and randomly assigned to 3 groups: C, control (air abraded); SG, sol-gel silica infiltration; and GI, glass infiltration. SG specimens were immersed twice in silicic acid for 20 min and dried (100°C, 1 h). GI specimens were presintered (1,400°C, 1 h) before a glass powder slurry was applied to the intaglio surface. All specimens were then sintered (1,530°C, 2 h). Following adhesive bonding (Panavia F 2.0, Kuraray) and water storage (37°C) for 10 d, the Brazil nuts were subdivided into groups baseline and aged (40,000 thermal cycles between 5°C and 55°C, with a dwell time of 30 s). The Brazil nuts were subjected to axial-loading tests using various inclinations (precrack angle with load direction): Θ = 0°, 5°, 10°, 15°, or 25°, which define the stress type at the interface, from pure tension (0°) to increasing levels of shear. Under pure tension (0°), GI yielded superior interfacial fracture energy, SG and C were similar, and aging had no effect. Under predominantly shear stresses (25°), aging significantly decreased interfacial fracture energy of C and SG, while GI remained stable and was superior. The glass infiltration of the zirconia intaglio surface increases its adhesive bonding interfacial toughness. The sol-gel silica infiltration method requires improvement to obtain a homogeneous surface infiltration and an enhanced bond strength.

Keywords: bonding force; ceramics; fracture strength; resin cements; shear strength; tensile strength.

Figure 1.

Description of the study design: (A) Scanning electron microscopy images showing the microstructure of the zirconia used in this study, from left to right; as machined (InCeram YZ), the sol-gel–treated zirconia (surface view) with silica islands intermingled within zirconia grains and the glass-infiltrated zirconia, which was polished in a shallow angle to expose the underlying zirconia/glass graded layer. (B) Digital photograph of a Brazil nut specimen: i) side view of the 2 halves and ii) detail of the intaglio surface and precrack on one of the halves. (C) Schematic of the load-to-fracture test setup, where 2a is the crack length, R is the radius of the Brazil nut specimen, and Θ is the angle between the loading direction and crack orientation. At Θ = 0° and Θ = 25°, illustration of the direction of stresses applied to the bonded interface is shown.

Figure 2.

Plots of interfacial fracture energy (Gc) values calculated as a function of loading phase (Ψ), where Ψ = 0 represents specimens fractured in mode I (pure tensile) and increasing values of Ψ represent increasing levels of shear (mode II).

Figure 3.

Interfacial fracture test results: (A) Average interfacial fracture energy under pure tension (Θ = 0°) and predominantly shear (Θ = 25°) for each surface treatment and time. Uppercase letters indicate statistical differences among surface treatments, within stress type (Θ). Lowercase letters indicate statistical differences between baseline and aging within surface treatment and stress type. (B) Classification of failure type distribution in each test condition.

Figure 4.

SEM images of intaglio surfaces after interfacial fracture: (A) Illustration of the failure modes as classified in this study: I, crack propagates within the cement layer; II, crack propagates between the cement and ceramic at the same interface; III, crack starts at one interface and goes to the other across the cement thickness; and IV, crack kinks between the 2 interfaces. (B) For the sol-gel silica infiltration, the coating was not uniform. The high-magnification image shows silica islands where residual cement is still bonded after fracture.

Figure 5.

X-ray diffraction spectra for each group, with arrows pointing to the main monoclinic peak in the control and sol-gel groups after aging. Side panels show the monoclinic phase fraction (wt.%) calculated by the Rietvield refinement method.