Purpose: To evaluate the Knoop hardness (KH) and the bond strength (BS) at the tooth/restoration interface of conventional methacrylate- (Filtek Supreme) and silorane-based (Filtek P90) composites photoactivated by different methods using an LED Freelight 2. Materials and Methods: Bond strength was tested in a universal testing machine by the "push-out" test in restored cavities measuring 2 × 1.5 × 2 mm with a C-factor of 2.2, prepared in 60 bovine teeth. To restore the cavities, the respective adhesive system of each composite was used (Single Bond 2 and P90 system adhesives). The composites were photoactivated by 3 different methods: continuous light: 40 s at 1000 mW/cm2; soft-start: 10 s at 150 mW/cm2 + 38 s at 1000 mW/cm2; pulse delay: 5 s at 150 mW/cm2, followed by a 3-min wait (without photoactivation) and 39 s at 1000 mW/cm2. Before the push-out test was performed, the KH was analyzed at the top and bottom of the restorations. Data were statistically anaylzed using ANOVA and Tukey's test. Results: The photoactivation methods produced no differences in BS or KH in the same composite, while Filtek P90 (28.0 MPa) showed higher BS values than Filtek Supreme (22.3 MPa) and a lower KH. Conclusion: The composite Filtek P90 was capable of increasing bond strength, but presented lower Knoop hardness. Palabras clave: resin composite, photoactivation, bond strength, silorane, methacrylate