384948 A Molecular Dynamics Study Assessing the Accuracy of the Generalized Amber Force Field to Predict the Thermophysical Properties of 20 Ionic Liquids

Monday, November 17, 2014
Galleria Exhibit Hall (Hilton Atlanta)
Kayla Sprenger, Vance Jaeger and Jim Pfaendtner, Chemical Engineering, University of Washington, Seattle, WA

Ionic liquids (ILs) are organic salts that typically have melting points below 100°C. These salts are known as “designer” solvents, because small changes can be made to the structure (i.e. cation or anion) or chemical composition of the mixture that lead to large changes in the physical properties of the liquid; properties such as melting point, solvation power, and viscosity can all vary greatly with such subtle changes. Due to their low viscosities and negligible vapor pressures, as well as exceptional thermal and chemical stabilities, ILs have come to play important roles in many fields, such as electrochemistry, separation science, catalysis, organic synthesis, and biocatalysis and biomass conversion.1 Despite the many advances ILs have led to in these areas of research, there is still a great need for novel quantitative approaches that can easily, quickly, and accurately predict the transport and thermodynamic properties of these unique solvents, with knowledge solely of their chemical structures. If this can be done, it would allow for many more ILs to be explored for various applications by decreasing the time to sift through and test the multitudes of possible cation/anion combinations.

This work critically assesses the ability of the Generalized Amber Force Field, or GAFF, to predict various properties of ILs. An entirely automated method has been created that allows for quick parameterization of these nonaqueous solvents. Specifically, thermodynamic properties of density and heat capacity and transport properties of self-diffusivity and viscosity have been computed, using classical Molecular Dynamics, for twenty industrially-relevant ILs. Results show that this simple force field is, in nearly every case, very adept at predicting all the computed properties of the ILs studied. Parallel simulations with Tip3p water have been done for comparison. This poster also presents a simple extension of the automated method in order to trivially enable simulation of biomass or biomolecules in the presence of ILs.


[1] Sambasivarao and Acevedo. J. Chem. Theory Comput. 5, 1038–1050 (2009).

Extended Abstract: File Not Uploaded