2021 Annual Meeting

(62c) From Data to Process Design: Modeling Azeotropic Separation of High Global Warming Potential Hydrofluorocarbon Refrigerants Using Ionic Liquids

Authors

Dowling, A. - Presenter, University of Notre Dame
Maginn, E. - Presenter, University of Notre Dame
Garciadiego, A., University of Notre Dame
Befort, B., University of Notre Dame
Franco, G., University of Notre Dame
Hydrofluorocarbon (HFC) mixtures are commonly used as refrigerants. Unfortunately, some pure HFCs and HFC mixtures exhibit high global warming potentials (GWPs) and are being phased out as part of the 2016 Kigali agreement [1]. For example, R-410a and R-407c contain a significant fraction of R-125 (50% and 25%, respectively), an HFC with a GWP 3500 times that of CO2. There are approximately 100 million kilograms of HFC-32 (a low GWP HFC) in global circulation as a component of R-410a, worth approximately half a billion dollars. HFC mixtures form azeotropes, which makes it impractical to separate by conventional methods. While it is desirable to separate and reuse or recycle HFC components from existing refrigerant mixtures, current azeotropic separation technologies are infeasible or not practical for these mixtures.

Ionic liquids are known as "designer solvents" because of the wide variety of cations and anions that can be paired to form ILs with different properties. Previous studies have shown ILs can exhibit high HFC solubility and desirable solvent properties [2, 3, 4, 5, 6]. Shiflett and colleagues [7] performed AspenPlus simulations showing that ILs can function as entrainers for the separation of HFC azeotropic mixtures. While experimentation proved the feasibility of IL use for azeotropic HFC separation, this technology must be applicable for various HFC refrigerant systems, which requires consideration of millions of potential IL cation/anion pairs and corresponding process designs. This necessitates leveraging computational tools, especially accurate thermophysical property prediction methods, and rigorous process optimization models, to rapidly screen and optimize candidate ILs and process designs.

In this talk, we describe how we model ternary phase behavior of hydrofluorocarbon (HFC) and ionic liquid (IL) mixtures from binary and ternary data. We developed a process design framework based on open-source IDAES-PSE Python modeling framework and Pyomo to analyze an IL's feasibility as a separating agent using computational and experimental data. Here, we answer the question given a mixture of HFCs, which ionic liquid is ideal for the separation and the process design? We have estimated ternary diagrams from binary data with a 10% error in the compositions. We have also been able to calculate the phase behavior of six different ILs qualitatively. We make a direct comparison between the Peng-Robinson cubic equation of state (EoS) and previously published ternary diagrams predicted from the more sophisticated soft-SAFT EoS [8]. We verify our framework by calibrating with soon-to-be-published first-of-a-kind ternary data. Additionally, we qualitatively explore the feasibility of new ILs as separation agents. We have created a framework that allows us to carry out process design from binary and ternary data. The framework is fast enough to rapidly screen possible entrainers for HFC separation in a matter of seconds.

References

[1] United Nations Environment Programme. Ozone Secretariat. (2006). Handbook for the Montreal protocol on substances that deplete the ozone layer. UNEP/Earthprint.

[2] Plechkova, N. V., & Seddon, K. R. (2008). Applications of ionic liquids in the chemical industry. Chemical Society Reviews, 37(1), 123-150.

[3] Chávez-Islas, L. M., Vasquez-Medrano, R., & Flores-Tlacuahuac, A. (2011). Optimal molecular design of ionic liquids for high-purity bioethanol production. Industrial & Engineering Chemistry Research, 50(9), 5153-5168.

[4] Shiflett, M. B. and Yokozeki, A. (2006) Vapor- liquid- liquid equilibria of hydroorocarbons 1-butyl-3-methylimidazolium hexauorophosphate". Journal of Chemical & Engineering Data 51.5 1931-1939.

[5] Strasser D. (2007). Photoelectron spectrum of isolated ion-pairs in ionic liquid vapor. The Journal of Physical Chemistry 111(17) 3191-3195.

[6] Rita, A. C. Morais, Harders, A. N., Baca, K. R. , Olsen, G. M. , Befort, B, Dowling, A. W., Maginn, E. J. ,& Shiflett, M. B.(2020) Industrial & Engineering Chemistry Research 59(40), 18222-18235

[7] Shiflett, M. B. and Yokozeki, A. (2006). Chimica Oggi - Chemistry Today, 24(2).

[8] S. Asensio-Delgado, D. Jovell, G. Zarca, A. Urtiaga, and F. Llovell. ( 2020). Thermodynamic and process modeling of the recovery of R410a compounds with ionic liquids. International Journal of Refrigeration, 118:365-375,