LAUSR

The bonding performance of a glycerol phosphate dimethacrylate (GPDM)-based, two-step, self-etch (SE) adhesive was experimentally compared to that of 10-methacryloyloxydecyl dihydrogen phosphate (MDP)-based universal adhesives in different application modes for enamel bonding. Microtensile bond strength (μTBS) for adhesives bonded to enamel was measured initially (24 h water storage) and after 10 000 thermocycles plus water storage for 30 days. A GPDM-based, two-bottle, two-step, self-etch adhesive (Optibond Versa, OV) and three one-bottle MDP-based universal adhesives, one self-etching (Tetric N Bond Universal, TNBU) and two with etch-and-rinse (E&R) processing (Single Bond Universal (SBU); Clearfil Universal Bond Quick (CUBQ)), were tested. Scanning electron microscopy (SEM) evaluated nanoleakage at the bonding interfaces. A profilometer determined roughnesses of enamel surfaces after phosphoric acid etching, OV priming, or TNBU conditioning. SEM observed the corresponding surface morphology. NMR and X-ray photoelectron spectroscopy (XPS) characterized chemical bonding in hydroxyapatites (HAps) conditioned with the adhesives. Etch-and-rinse samples had significantly stronger bonding than self-etch samples (p < 0.05) irrespective of aging. The μTBS values for initial and aged OV were significantly higher than those of TNBU (p < 0.05). Aging did not significantly decrease μTBS for any sample except TNBU (p < 0.05), but it significantly aggravated nanoleakage. Etch-and-rinse processing resulted in less nanoleakage than self-etching; the OV samples leaked less than TNBU, both before and after aging. Phosphoric acid etching achieved the highest enamel surface roughness, followed by OV primer. Ca–O–P bonds in hydroxyapatite conditioned with TNBU, SBU, and CUBQ were confirmed by NMR, which showed similar results to XPS observations of conditioned hydroxyapatite powders except OV primer. The GPDM-based, two-step, self-etch adhesive can provide higher micromechanical retention potential, bond strength, and durability than the MDP-based universal adhesive in self-etch mode but lower performance than the MDP-based universal adhesive in etch-and-rinse mode. None of the tested adhesives could avoid nanoleakage after aging.