A comparison of different geometrical models in calculating physicochemical properties of quaternary systems

Abstract

In this paper, the geometric model has been extended to estimate the physicochemical properties of quarternary systems based on the data of its six sub-binary systems. It has been pointed out that the asymmetrical model is not suitable for the calculation of the systems higher than three components due to the difficulty of selecting the asymmetrical component. Though this difficulty will disappear in the symmetrical model, another trouble will be introduced instead. When two of “ n ” components are identical, this symmetrical model cannot be reduced to a “ n − 1 ” component system, of course, which is unreasonable. Based on these facts, Chou et al. have proposed a “new generation geometric model” that can overcome all these defects mentioned above. The correctness of this new generation geometric model has already been proved through a theoretical analysis and partially demonstrated by some ternary examples. However, it has never been tested in high order practical systems due to the lack of data. Recently, some volume and viscosity data in the Propan-2-ol + methylacetate + dichloromethane + n -pentane quaternary system have been found that can be used to judge which model would be better. The calculation results show that in the viscosity calculation our model is the best, while in the volume calculation our model is still good, when the calculation error is considered. These facts prove that the new generation geometrical model can work well in a higher order system.