In this paper, the effects of mechanical activation and roasting on treating titania slag process were systematically investigated. Several operating parameters as independent variables, namely activation time, temperature and roasting time, and synthetic rutile content as dependent variable, were optimized using response surface methodology (RSM) with a central composite design (CCD). A second-degree equation for independent and dependent variables was computed and used to create the response surface and contours. It showed that linear effect of activation time, temperature and roasting time, interaction effect between activation time and temperature, square effect of activation time showed influence on synthetic rutile content. According to results from analysis of variance (ANOVA), the value of the determination coefficient (R-2 = 0.9767) indicates that the model was a good fit that 97.67% of the variation could be explained well by the model. The value of the adjusted determination coefficient (adj.R-2 = 0.9558) was also very high to advocate for a high significance of the model. The predicted synthetic rutile content after process optimization was found to agree satisfactory with the experimental values. The optimum experiment parameters were found as follows: activation time of 170.42 min, temperature of 1149.28 degrees C, and roasting time of 20.79 min. The specific polymorphic phase transition of titania slag before and after combination mechanical activation and roasting process were obtained and characterized by X-ray Diffraction (XRD), respectively. (C) 2009 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.