The CPA (cubic-plus-association) equation of state is applied in this work to a wide range of systems of relevance to CO2 transport. Both phase equilibria and densities over extensive temperature and pressure ranges are considered. More specifically in this study we first evaluate CPA against density data for both CO2 and CO2–water and for vapor–liquid equilibrium for mixtures of CO2 with various compounds present in transport systems. In all of these cases we consider various possibilities for modeling CO2 (inert, self-associating using two-, three-, and four sites) and the possibility of cross-association with water. Finally, we evaluate the predictive performance of CPA for multicomponent CO2 mixtures in transport systems which also include water, methane, and H2S. The results are compared to both experimental data and selected other approaches from literature. The results for the multicomponent systems are predictions using parameters solely estimated from binary data. The target of this work is two-fold: to assess the performance of the model for mixtures of practical significance but also to identify the best modeling approach so that we can arrive to an “engineering approach” for applying CPA to acid gas mixtures. The overall conclusion is that CPA performs satisfactorily; the model in most cases correlates well binary data and predicts with good accuracy multicomponent vapor–liquid equilibria. Among the various approaches investigated, the best ones are when cross association of CO2 with water is accounted for or when CO2 is considered to be a self-associating molecule (with three or four sites). The final choice on the best approach requires investigating a much larger set of mixtures including also alcohols and glycols, which will be considered in future works.
Journal of Chemical and Engineering Data, 2014, Vol 59, Issue 10