Comparison of models for calculation of the thermodynamic properties of NH 3 -CO 2 -H 2 O mixture

Couple of models have been developed to calculate thermodynamic properties of NH3-CO2-H2O systems. These models are typically an equation of state for the vapor phase and an activity coefficient model for the liquid phase (Que & Chen, 2011). The activity coefficient models can be divided into three groups based on previous studies, Pitzer model, electrolyte Non Random Two Liquid (e-NRTL) model and extended UNIQUAC model. Que & Chen (2011) deem the e-NRTL model the model the most suitable for process modelling and simulations since it requires only binary interaction parameters and makes use of mole fraction concentration scale consistently for both the short range local compositions interactions and the long range Debey-Huckel expression. Darde (2011) compared the built in e-NRTL model from Aspen Plus to an upgraded version of the extended UNIQUAC model developed by Thomsen et al. (1996). His findings were that the extended UNIQUAC model is significantly more accurate than the e-NRTL model from Aspen. He does mention that if the binary interaction parameters were better fitted to experimental data for NH3-CO2-H2O mixture, the e-NRTL model might become more competitive with the extended UNIQUAC model. Since then the e-NRTL model has been modified in this way by couple of authors, included Que & Chen (2011) and Niu et al. (2013). Both of their adjusted models have then been used by other authors for process modelling, for example Zhang & Guo (2014) used the model with adjusted parameters from Niu et al. (2013) and Liu et al. (2015) used the modified model from Que & Chen (2011). In this paper the extended UNIQUAC model is compared with the e-NRTL thermodynamic model that is built into the most recent version of Aspen Plus, and two modified e-NRTL models; the one developed by Que and Chen (2011) and a new fit. This is done to confirm if the modified models can reach similar accuracy as the extended UNIQUAC model and how much more accurate they are compared to the built in model in Aspen Plus.