Search

In this work liquid–liquid equilibrium (LLE) data for aqueous two–phase system (ATPS) containing polyethylene glycol 6000 (PEG–6000) and Na 2 CO 3 were measured at T = (293.15, 303.15 and 313.15) K. Our results showed that increasing temperature expands the two–phase region; meanwhile large two–phase region was seen in this system which makes it applicable for industrial purposes. Experimental data were modeled using UNIQUAC, Extended UNIQUAC (E–UNIQUAC) and e–NRTL models to obtain binary interaction parameters. The model parameters were obtained using Shuffled Complex Evolution (SCE) optimization algorithm. The results showed that all studied models were in good agreement with the experimental data. Average deviation (%Δ w ) of e–NRTL, E–UNIQUAC and UNIQUAC models were 1.26, 1.37 and 3.46%, respectively. Finally in order to check the predictability of the E–UNIQUAC model, liquid–liquid equilibrium in systems containing PEG with different molecular mass (1000, 3350, 4000 and 8000) and Na 2 CO 3 were studied; It was found that liquid–liquid equilibrium data can be predicted satisfactorily using adjusted parameters in this work. [ABSTRACT FROM AUTHOR]

In this work the phase equilibrium of an aqueous two phase system (ATPS) containing polypropylene glycol (PPG, molecular weight = 425 kg·mol−1) and NaClO4 was investigated at atmospheric pressure and at 288.15 and 298.15 K. Two phase regions and composition of phases were determined. Our results show that as the temperature increases, the two-phase region expands. Also, the extended UNIQUAC (E-UNIQUAC) equation was used to correlate the equilibrium data. To reduce the number of adjustable parameters, ATPSs composed of PEG and PPG were collected from the literature and simultaneously correlated using the E-UNIQUAC model. Also, the effect of temperature on the liquid–liquid equilibrium (LLE) was considered by using temperature-dependent parameters. In the modeling, two different scenarios were supposed. In the first, polymer and salt were treated as solutes (Case A), while in the second, the pseudo-solvent approach was considered (Case B). The results showed good agreement with experimental data in both cases. The average absolute deviation of the model using Case B was about 0.2% and that for Case A was about 2% in the ATPS composed of PEG. Meanwhile, the reported errors in the ATPS containing PPG for Case A and Case B were almost equal. [ABSTRACT FROM AUTHOR]