Debonding mechanism of zirconia and lithium disilicate resin cemented to dentin

To evaluate debonding mechanism of zirconia and lithium disilicate cemented to dentin mimicking what could occur in a clinical setting. A null hypothesis of no difference in tensile bond strength between groups of zirconia and lithium disilicate cemented with resin cements was also tested. Zirconia rods (n = 100) were randomly assigned to two different surface treatment groups; air borne particle abrasion and hot etching by potassium hydrogen difluoride (KHF₂). Lithium disilicate rods (n = 50) were surface etched by hydrofluoric acid (HF). Five different dual cure resin cements were used for cementing rods to bovine dentin. Ten rods of each test group were cemented with each cement. Test specimens were thermocycled before tensile bond strength testing. Fracture morphology was visualized by light microscope. Mean surface roughness (Sa value) was calculated for randomly selected rods. Cohesive fracture in cement was the most frequent observed fracture morphology. Combination of adhesive and cohesive fractures were second most common. Fracture characterized as an adhesive between rod and cement was not observed for KHF₂ etched zirconia. Highest mean tensile bond strength was observed when cementing air borne particle abraded zirconia with Variolink Esthetic (Ivoclar Vivadent). All surface treatments resulted in Sa values that were significant different from each other. The number of cohesive cement fractures observed suggested that the cement was the weakest link in bonding of zirconia and lithium disilicate.