To evaluate the effect of the particle size of sandblasting and the composition of the resin cement on the microtensile bond strength (MTBS) to zirconia.
Forty zirconia blocks (Cercon, Dentsply) were polished and randomly treated as follows: Group 1 (NT): no treatment; Group 2 (APA-I): airborne particle abrasion (Cobra, Renfert) using 25-µm aluminium-oxide (Al(2)O(3))-particles; Group 3 (APA-II): APA with 50-µm Al(2)O(3)-particles; and Group 4 (APA-III): APA using 110-µm Al(2)O(3)-particles. Ceramic blocks were duplicated in composite resin. Samples of each pretreatment group were randomly divided into two subgroups depending on the resin cement used for bonding the composite disks to the treated zirconia surfaces. Subgroup 1 (PAN), which was a 10-MDP-containing luting system, used Clearfil Ceramic Primer plus Panavia F 2.0 (Kuraray) and Subgroup 2 (BIF) used Bifix SE (VOCO) self-adhesive cement. After 24h, bonded specimens were cut into 1±0.1mm(2) sticks. MTBS values were obtained using a universal testing machine (cross-head speed=0.5mm/min). Failure modes were recorded and the interfacial morphology of the debonded microbars was SEM-assessed. Two-way ANOVA, Student-Newman-Keuls tests, and the step-wise linear regression analysis were performed with the MTBS being the dependent variable (p<0.05).
Despite the sandblasting granulometry, PAN bonded to air-abraded surfaces attained the highest MTBS and frequently showed mixed fractures. BIF recorded no significant differences in MTBS depending on the conditioning method, and registered the highest rates of premature and adhesive failures.
The 10-MDP-containing luting system seems to be the most suitable to bond zirconium-oxide ceramic, mainly after sandblasting.