In this study a 2(4) factorial design was employed to investigate the supercritical CO2 devulcanization process of ground tire rubber from end of life tires performed using diphenyl disulfide (DD) as devulcanizing reagent. The aim of the experimental design was to investigate the influence on the process of temperature, pressure, amount of devulcanizing reagent, treatment time and their interactions. The crosslink density, sol fraction, gel fraction and sulfur content were chosen as experimental responses. Multiple linear regression was used for modeling the relationship between each response and the process variables. Reduced regression models were obtained for each response, considering only the significant variables and interactions. The predicted results from these reduced models showed good agreement with the experimental values. Temperature, amount of DD and DD-temperature interaction resulted the relevant parameters for the process. On the contrary, the influence of treatment time, pressure and all other interactions proved to be negligible. These results have an important outcome since this devulcanization process can be carried out in a short time and at relatively low pressure, with subsequent energy saving. (C) 2014 Elsevier B.V. All rights reserved.