Efficiency in the sonocatalytic degradation of Rhodamine B dye catalyzed by TiO2 nanotubes as a function of initial concentration of dye, catalyst amount and power of ultrasonic irradiation at a frequency of 35 kHz was studied. A central composite design (CCD) was used for response surface modeling to evaluate the combined effects of these variables as well as to optimize the degradation efficiency of Rhodamine B. Satisfactory prediction based on a second-order model with high coefficient of determination (R-2) of 0.98 was achieved for the optimized sonocatalytic degradation process. Lastly, the significance and adequacy of the model were analyzed using analysis of variance (ANOVA). The optimal conditions for sonocatalytic degradation efficiency of Rhodamine B were found at 44.8 mg/L of Rhodamine B, 2.14 g/L of TiO2 nanotubes and ultrasonic power of 68.9W to achieve 94.6% dye removal under ultrasonic irradiation in 3 h. The significance of the findings at a confidence level of 95% was demonstrated. It was found that ultrasonic power, initial dye concentration and interaction between dye concentration and catalyst loading had important effects in the sonocatalytic degradation efficiency of the Rhodamine B by TiO2 nanotubes. (C) 2010 Elsevier B.V. All rights reserved.