(107a) What Can Molecular Modeling and Machine Learning Teach Us about Ionic Liquid-Ionic Liquid Mixtures?

Ionic liquids are a class of non-aqueous solvents comprised entirely of ions. The asymmetry in structural motif of the ions and charge delocalization result into a large number of ionic liquids with melting points below room temperature. Over the last two decades, ionic liquids have drawn considerable attention from the research community due to the fact that many ionic liquids are practically non-volatile, thermally, chemically, and electrochemically stable. Additionally, a large number of cations and anions are available so that an ionic liquid with properties desirable for a given application or a process can be designed, at least in theory. A major theme of these investigation is on elucidating structure-property relationships of pure ionic liquids by a careful selection of the cation, anion, and/or substituents on the ions. At molecular-level, the extent to which such changes affect a delicate balance of electrostatic, hydrogen bonding, and van der Waals interactions will determine how different the macroscopic properties will be from those of the parent ionic liquid.

Our effort, over the past several years, has focused on understanding a connection between changes in the molecular-level interactions that occur in ionic liquid-ionic liquid mixtures and their impact on microscopic and macroscopic properties. In this presentation, we will discuss several ionic liquid mixtures in which the difference in the molar volume and hydrogen bonding ability of anions drive local structural reorganization of ions and long-range structural order in ionic liquid-ionic liquid mixtures, which is absent in pure ionic liquids. We will demonstrate that these structural transitions lead to varying gas dissolution mechanisms and a maximum in dielectric constant. Furthermore, we will showcase how machine learning can be leveraged to provide clues on which ionic liquid mixtures are likely to exhibit a maximum or a minimum in property such as ionic conductivity warranting a detailed molecular simulation-based investigation.