Copolymers are important in the manufacture of new polymeric materials with specific characteristics. For linear polymers, thermodynamic models based on the thermodynamic perturbation theory are interesting, since this theory regards the association between monomers. In this work, cloud points of mixtures of copolymers (PEH, PEP, PEAA and PEVA) (PEP: poly(ethylene-co-propylene); PEAA: poly(ethylene-co-acrylic acid); PEH: poly(ethylene-co-1-hexene); PEVA: poly(ethylene-co-vinyl acetate)), a supercritical fluid (=C-2) and co-solvents (C-1, C-2, C-3, nC(4), =1C(4), =1C(6), AA, VA, He, N-2, CO2) (C-1: methane; C-2: ethane: =C-2: ethylene; C-3: propane: nC(4): n-butane; =1C(4): 1-butene; =1C(6): 1-hexene; AA: acrylic acid; VA: vinyl acetate; He: helium; N-2: nitrogen; CO2: carbon dioxide) were modeled using the PC-SAFT equation of state(Perturbed Chain-Statistical Associating Fluid Theory) with a one-type van der Waals mixing rule by fitting one single interaction parameter. Pure component parameters for the supercritical fluid and cosolvents were obtained by regression of vapor pressure and density data of saturated liquid, while pure component parameters for polymers that compose the copolymers were obtained by regression of pure liquid PVT data. Binary interaction and pure component parameter estimation was performed by using the modified maximum likelihood method. Relative deviations between the calculated and experimental cloud points show that the PC-SAFT model had an excellent performance. (C) 2009 Elsevier B.V. All rights reserved.