In the current study, the degree of conversion (DC) of bonding agents photoactivated using QTH or LED light-curing units (LCUs) was evaluated by Fourier Transform infrared spectroscopy with an attenuated total reflectance (ATR) device. Four LCUs were evaluated: one QTH (Optilux 501; Demetron Kerr) and three LEDs: Radii Cal (SDI), Elipar FreeLight 2 (3M ESPE) and Bluephase (Ivoclar Vivadent). Two etch-and-rinse (Scotchbond Multi-Purpose-SBMP and Single Bond 2-SB2) and two self-etch adhesives (Clearfil SE Bond-CSE, and Clearfil S3 Bond-CS3) were tested. For SBMP and CSE, the primer was not used. The irradiance and spectral emission of the LCUs were obtained with a radiometer and spectrometer. The materials were placed onto the ATR cell as thin films, the solvent was evaporated (when necessary) and photoactivation was carried out for 20 seconds. The DC (%) was evaluated after five minutes (n = 5). The data were statistically analyzed (p < 0.05). The irradiance for Optilux, Radii, FreeLight 2 and Bluephase was 760, 600, 1000 and 1100 mW.cm(-2), respectively. The wavelength of emission for Optilux was between 375 and 520 nm (peak at 496 nm), while for Radii, it was between 420 and 520 nm (peak at 467 nm). Freelight 2 presented an emission spectrum between 415 and 520 nm, and for Bluephase, it was between 410 and 530 nm, both having a peak at 454 nm. SB2 generally showed higher DC compared with the other bonding agents. When cured using the QTH unit, the DC results were SB2 = CS3 > CSE > SBMP; for all LEDs, the DC results showed SB2 > CSE > SBMP > CS3. For SB2, the highest DC was observed when the material was cured with Radii, while there were no significant differences among the other LCUs. CSE and CS3 showed higher DC when cured using the QTH unit, but similar results were observed among the LEDs. For SBMP, no significant differences among the LCUs were detected. In conclusion, the combination bonding agent vs curing unit had a significant effect on DC, mainly for the self-etch adhesives.