In this study, the dual- and self-curing potential of self-adhesive resin cements (SARCs) as thin, clinically-relevant cement films was investigated. The SARCs tested were: BisCem (BSC; Bisco), Maxcem Elite (MXE; Kerr), RelyX Unicem clicker (UNI; 3M ESPE), seT capsule (SET; SDI), and SmartCem 2 (SC2; Dentsply Caulk). The conventional cement RelyX ARC (3M ESPE) was tested as a reference. The degree of conversion (DC) as a function of time was evaluated by real-time Fourier transform infrared spectroscopy with an attenuated total reflectance (ATR) device. The cements were either photoactivated for 40 seconds (dual-cure mode) or not photoactivated (self-cure mode). The cement film thickness was 50 ± 10 μm. The DC (%) was evaluated 1, 5, 10, 15, 20, 25, and 30 minutes after placing the cement on the ATR cell. Data for DC as a function of time were analyzed by two-way repeated measures analysis of variance (ANOVA). DC values at 30 minutes for the self- and dual-cure modes were submitted to one-way ANOVA. Post hoc comparisons were performed using the Student-Newman-Keuls test (p<0.05). The rate and the extent of conversion were lower for the SARCs compared with the conventional cement. Means ± standard deviations (SD) for the dual-cure mode at 30 minutes were: 75 ± 5 (ARC)a, 73 ± 8 (SET)a, 61 ± 4 (MXE)b, 51 ± 9 (BSC)c, 51 ± 4 (UNI)c, and 48 ± 3 (SC2)c, while in the self-cure mode means and SD were 62 ± 6 (ARC)a, 54 ± 3 (MXE)b, 40 ± 6 (SC2)c, 35 ± 2 (UNI)c, 35 ± 3 (SET)c, and 11 ± 3 (BSC)d. The DC for the dual-cure mode was generally higher than the self-cure, irrespective of the time. Discrepancies in DC between the dual- and self-cure modes from 11% to 79% were observed. In conclusion, SARCs may present slower rate of polymerization and lower final DC than conventional resin cements, in either the dual- or self-cure mode.