Different ceramic surface cleaning methods have been suggested after the acid conditioning. The aim was to evaluate the effect of different protocols used to remove the remaining hydrofluoric acid on the shear bond strength (SBS) between lithium disilicate and resin cement. Forty-four specimens of lithium disilicate (IPS e.max Press) were divided in 4 groups (n=11): group C (control, no treatment); group HF+S (5% hydrofluoric acid + silane); group HF+US+S (5% hydrofluoric acid + ultrasound cleaning + silane); group HF+PH+S (5% hydrofluoric acid + 37% phosphoric acid + silane). Scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) were performed to characterize the surface morphology. The SBS test was performed on the resin/ceramic interface, and the failure mode was characterized. SBS values were submitted to 1-way ANOVA and the Tukey test (?=.05). The relation between surface treatment and failure modes was analyzed using the chi-squared test (?=.05). The surface treatment type interfered in the shear strength (p<.001) and higher SBS values were observed for the groups HF+US+S (17.87 MPa) and HF+PH+S (16.37 MPa). The surface treatment did not influence the failure mode (p=.713). No fluorsilicate salts were observed after ultrasound cleaning. The utilization of ultrasound cleaning was an effective procedure to remove remaining fluorsilicate salts, promoting the highest SBS values.