Phase Equilibria Thermodynamics of Water and Ionic Liquids
Ionic Liquids (ILs) have emerged as potential substitutes for common organic solvents for industrial applications. ILs present many interesting thermodynamic and transport properties that suggest their utilization in carbon capture, catalysis or energy storage technologies. These properties can be tuned by means of modifying the constituent ions or by varying the concentration of the impurities present in the system. Many experimental and computational studies have pointed to the fact that ILs are hygroscopic and that the presence of water can have a dramatic effect on IL properties such as density, viscosity or surface tension. In this work, Gibbs Ensemble Monte Carlo simulations were used to calculate absorption isotherms of water in three imidazolium-based Ionic Liquids: ([C4MIM][PF6]), ([C4MIM][TF2N]) and ([C4MIM][Cl]). An systematic evaluation of several modified water models and various IL force fields was conducted. The main conclusion of the study is that, to consistently model water absorption, force fields that include effects such as polarization are likely necessary.
Development of the Discrete Fractional Component Monte Carlo Method
In the Discrete Fractional Component Monte Carlo method, an expanded ensemble is constructed using a parameter that modulates the intermolecular interactions of a single molecule with the rest of the system. Changes to are preformed discretely, allowing the measurement of thermodynamic properties only when a particle is in a fully coupled state. The set of values are specified by the user, offering flexibility of controlling the gradual insertion and deletion of molecules. This method is combined with a molecular fragment sampling scheme that allows efficient sampling of molecules with intramolecular degrees of freedom. Finally, the technique is used to estimate the transfer and solvation free energies of a test system comprising hexane and water.
Development of Cassandra, an open-source Monte Carlo code
Perhaps the biggest hurdle in Monte Carlo molecular simulations is the lack of reusable software. As opposed to molecular dynamics (MD) or quantum mechanics (QM) methods, Monte Carlo codes are not widely available, easy to use nor routinely used by a large community. This is not surprising. MD and QM methods are essentially concerned with solving a single equation (i.e. Newton or Schroedinger equation), whereas MC techniques require the user to use system-dependent moves to guarantee correct configurational sampling. For instance, polydisperse polymeric systems will require a connectivity-altering and concerted rotation MC moves typical phase equilibria calculation will require a destruction or creation move to attain equality of chemical potentials. It is this diversity of moves that makes it hard to create a general purpose MC code capable of simulating any chemical system. To address this problem, the open-source MC code Cassandra has been under active development over the past ten years at the University of Notre Dame. Some of the goals of this project include the development and implementation of advanced sampling methods in order to simulate molecules with complex topologies, the design of a reusable software to ease the implementation of new methods and the creation of tools to make the code easy to use.