Rose Bengal (C.I. name is Acid Red 94) was irradiated with UV light in the presence of hydrogen peroxide. The photoinduced decolorization of the dye was monitored spectrophotometrically. The apparent rate of clecolorization was calculated from the observed absorption data and was found to be pseudo first order. A systematic study of the effect of dye concentration and H2O2 concentration on the kinetics of dye clecolorization was also carried out. Dye clecolorization increased with increasing H2O2 concentration and decreasing dye concentration. The maximum dye clecolorization was determined as 90% with 0.005 mM dye at optimum 0.042 M H2O2 and pH 6.6. Additionally, the effect on clecolorization of this dye in the presence of some additives (ions) was also investigated. It was seen that sulphite caused a maximum effect on % decolorization of the dye solution. A plausible explanation involving the probable radical initiated mechanism was given to explain the dye clecolorization. The experimental data was also optimized using the response surface methodology (RSM). According to ANOVA results, the proposed model can be used to navigate the design space. It was found that the response of Rose Bengal degradation is very sensitive to the independent factors of dye concentration, H2O2 concentration, pH and reaction time. The proposed model for D-optimal design fitted very well with the experimental data with R-2 and R-adj(2) correlation coefficients of 0.85 and 0.80, respectively. (C) 2008 Elsevier B.V. All rights reserved.