Monte Carlo Simulations for Phase Equilibria and Software Development

Phase equilibria thermodynamics has a central role in understanding many processes relevant to chemical engineering. The main motivation of this dissertation is the development, implementation and application of molecular simulation methods to predict phase equilibria thermodynamics. Molecular simulations allow researchers to study matter at the microscopic level and make predictions of macroscopic properties at a wide range of conditions. Important advances have occurred over the past fifty years regarding molecular simulation methodologies. However, some systems offer exceptional challenges to standard simulation algorithms. Thus, there is a strong incentive to develop, implement and apply state-of-the-art molecular simulation techniques to tackle these hard problems.

First, the theory underlying classical Monte Carlo simulations as applied to molecular physics is reviewed. Then, specific algorithms to simulate molecules with highly coupled internal degrees of freedom are discussed. Additionally, the molecular models that represent inter- and intramolecular interactions are introduced. These methods and models are applied to the simulation the vapor liquid equilibria of water and ionic liquids. It is shown that electronic polarization should be included in phase equilibria calculations when simulating this type of systems. Next, a novel Monte Carlo method that aims to mitigate pathologies found in standard phase equilibria techniques is presented. The method is applied several systems to prove its validity. Finally, algorithms that were implemented as part of the development efforts of an open source Monte Carlo code are overviewed and validation tests are presented for each of the implemented techniques.