Fluid Phase Equilibria 337 (2013) 6–10 Contents lists available at SciVerse ScienceDirect Fluid Phase Equilibria j o ur nal homep age: www.elsevier.com/locate/fluid Comparison between three predictive methods for the calculation of polymer solubility parameters Ismael Díaz b, , Eduardo Díez a , Javier Camacho a , Salvador León b , Gabriel Ovejero a a Grupo de Catálisis y Procesos de Separación (CyPS), Departamento de Ingeniería Química, Facultad de C. Químicas, Universidad Complutense de Madrid, Avda. Complutense s/n, 28040 Madrid, Spain b Departamento de Ingeniería Química Industrial y del Medio Ambiente, Escuela Técnica Superior de Ingenieros Industriales, Universidad Politécnica de Madrid, C/José Gutiérrez Abascal 2, 28006 Madrid, Spain a r t i c l e i n f o Article history: Received 20 June 2012 Received in revised form 18 September 2012 Accepted 21 September 2012 Available online 28 September 2012 Keywords: Molecular dynamics COSMO-SAC Group contribution Polymer Bisphenol-A polycarbonate Ethylene-co-vinyl acetate Polyvinyl alcohol a b s t r a c t Solubility parameters (SP) of three polymers have been estimated and compared with both experimental values. The methods employed for the estimation are: traditional group contribution procedures (Fedors and van Krevelen), molecular dynamics simulation to calculate cohesive energy density and extension of the COSMO-SAC thermodynamic model to polymer mixtures. The aim of the paper is accurately pre- dicted polymer solubility parameters showing differences between methods. Selected polymers have been polyvinyl alcohol (PVA), ethylene-co-vinyl acetate (EVA) and bisphenol-A polycarbonate (PC). © 2012 Elsevier B.V. All rights reserved. 1. Introduction Polymer materials are extremely important in current way of life; adhesives, pipes, wrappings or multimedia devices are exam- ples of items that are mainly composed of polymer materials, widely produced all around the world [1]. Therefore, polymer processing has been an important research topic in the last decades. Many of the steps involved in polymer production processes are equilibrium staged steps, such as steam stripping or solvent devola- tization; this makes polymer–solvent compatibility a key aspect to accurately model these processes. In fact, the behaviour of the poly- mer is very important to be properly analyzed because, in many cases, it determines the design of the separation steps. However, polymer solution properties are quite different from conventional mixtures and they require using specific thermodynamics mod- els. Historically, the most widely employed model has been the well-known Flory–Huggins [2,3] model, based on the determina- tion of the so-called Flory–Huggins interaction parameter, 12 , that accounts for the compatibility of component 1 (solvent) and 2 (polymer). From the regular solution theory developed by Corresponding author. Tel.: +34 91 336 5341; fax: +34 91 394 4114. E-mail address: idiaz@quim.ucm.es (I. Díaz). Hildebrand [4,5] 12 can be related to the solubility parameters of both components by the simple expression: 12 = V 1 RT (ı 1 - ı 2 ) 2 (1) Solubility parameters can be calculated, following regular solu- tion theory, from the values of cohesive energy density, c (energy to separate molecules in the condensed phase): ı 1 = c = H 1,v - RT V 1 (2) where H 1,v is the vaporization energy. Later on, with Eq. (1), the value of 12 can be determined from the molar volume of the sol- vent (V 1 ) and the solubility parameter of the pure components. This implies that the value of the interaction parameter 12 of a polymer-containing mixture can be easily determined for a wide range of solvents if the value of the polymer solubility parameter is known. Many efforts have been made to develop both experimental techniques and prediction methodologies. Among the first ones, the most important techniques are swelling [6], inverse gas chromatog- raphy [7,8] or intrinsic viscosity [9] measurements. Among the prediction methods, van Krevelen [10] and Fedors [11,12] devel- oped both group contribution methods that are widely employed 0378-3812/$ see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.fluid.2012.09.028