Previous Page  |  Table of Contents  |  Next Page

J. Ilja Siepmann, Fellow

Molecular Simulations of Phase Equilibria and Development of Transferable Force Fields and Efficient Monte Carlo Algorithms

Detailed investigations of retention processes in gas-liquid and reversed-phase liquid chromatography and of phase equilibria and microscopic structures in supercritical fluid extraction systems and micellar surfactant systems are the main focus of this project. Accurate predictions of fluid phase equilibria and other thermophysical properties of complex chemical systems are of great fundamental and practical importance.

Configurational-bias Monte Carlo simulations in the Gibbs ensemble are currently employed to study the structures of stationary phases and to predict the partitioning of sample molecules between stationary liquid phase and mobile gas phase. As a first step, the stationary phase is modeled as a bulk liquid (neglecting the contributions arising from the liquid/solid-support and liquid/vapor interfaces) in thermodynamic contact with an interacting carrier gas (i.e., not an ideal gas). Three parameters are varied in these simulations: temperature, stationary phase composition, and solute. Simulations are carried out for squalane, Kovats B-87, carbowax (polyethylene-oxide), and dimethyl or methylphenyl silicone polymers. The homologous series of normal alkanes, primary alcohols, symmetric ethers, and alkyl benzenes, and a variety of branched alkanes are employed as solute molecules. This diversity of solutes allows a detailed investigation of the influence of shape, flexibility and functionality on the retention processes. The concept of enthalpy-entropy compensation is being investigated for these groups of solutes.

Another aspect of this endeavor to learn about chromatographic retention processes is directed at elucidating the effects of interfacial adsorption on retention. In particular, simulations are being carried out for systems where the bulk-liquid stationary phase is replaced by a system containing explicit liquid/vapor interface. The relative ratio of bulk liquid to interfacial liquid (related to the phase ratio and the column packing) can be varied to determine the contribution of interfacial adsorption to retention. Furthermore, these researchers are studying stationary systems that contain explicit liquid/solid-support and liquid/vapor interfaces.

Separations of multicomponent mixtures with supercritical solvents utilize differences in volatility (the driving force in ordinary distillations) and differences in the specific solute-solvent interactions (the driving force in ordinary liquid extractions). While supercritical fluids show solvency strength close to liquids, their transport properties are gas-like. As a first step, these researchers have started with an investigation of binary systems consisting of ethane (or ethylene) as the supercritical solvent and a high-molecular-weight alkane as the solute and liquid phase. The next step is studying the solvation of linear alkanes in supercritical CO2. Interest lies in how well these researchers' united-atom model (without partial charges) performs in combination with a simple-point-charge model for CO2. Thereafter, these researchers are studying ternary systems consisting of the supercritical solvent, a co-solvent (or entrainer), and the solute. In many applications, a small quantity of an entrainer is added to the supercritical solvent in order to enhance solvent power and selectivity. The changes caused by entrainers are much more marked in supercritical fluid extractions than in liquid extractions. A typical test case would be the solubility of benzoic acid in ethane with polar acetone as entrainer. Analysis of the microscopic structure of the supercritical phase should provide knowledge on the mechanism of the solubility enhancement.

To answer fundamental questions on surfactant behavior, these researchers are examining the self-organization and solubilization thermodynamics of hydrofluorocarbons in alkanes or perfluoroalkanes and of sodium salts of carboxylic acids in water. The former are examples of functionally primitive surfactants, while the later are prototypical ionic surfactants.A further project involves gaining accurate knowledge of phase equilibria and other thermophysical properties of complex fluid mixtures of enormous fundamental and practical importance. The success of molecular simulation in predicting thermophysical properties and in advancing an understanding of the relationship betwen molecular architecture and macroscopic observables, depends on the availability of efficient simulation algorithms and accurate force fields. For a long time, the available simulation techniques and the computer power were limiting progress. However, over the last ten years, enormous advances have been made in simulation methods and it is now becoming evident that attemtion should be shifted to developing sufficiently accurate force fields.

A further project involves gaining accurate knowledge of phase equilibria and other thermophysical properties of complex fluid mixtures of enormous fundamental and practical importance. The success of molecular simulation in predicting thermophysical properties and in advancing an understanding of the relationship betwen molecular architecture and macroscopic observables, depends on the availability of efficient simulation algorithms and accurate force fields. For a long time, the available simulation techniques and the computer power were limiting progress. However, over the last ten years, enormous advances have been made in simulation methods and it is now becoming evident that attemtion should be shifted to developing sufficiently accurate force fields.

99/87

99/133

99/134

99/179

99/180

The goal of the proposed research is to design two levels of transferable force fields for a large spectrum of organic molecules and water. The first level, called TraPPE (Transferable Potentials for Phase Equilibria), employs the united-atom representation and simple Lennard-Jones and Coulombic terms. In the second level, called TraPPE-pol (polarizable), all atoms are modeled explicitly, and both the van der Waals and electrostatic interactions can respond to changes in the environment. Whereas the first level is designed for simplicity and computational efficiency with good accuracy, the second level is aimed solely at the highest possible level of accuracy and transferability. These force fields allow these researchers to study single and multi-component phase equilibria, excess properties of mixing, and transport properties. The transferable force fields encompass linear, branched, and cyclic alkanes, alkenes, alkynes, alcohols, ethers, ketones, carboxylic acids, alkylbenzenes, perfluorinated alkanes, and last, but not least, water.

Another project is currently working on the development of three novel Monte Carlo algorithms. Adiabatic nuclear and electronic sampling Monte Carlo (ANES-MC) allows for efficient Monte Carlo sampling of polarizable force fields in the canonical, isobaric-isothermal, and Gibbs ensembles. The ANES-MC algorithm treats the electronic motion as special degrees of freedom, optimized on-the-fly, to a low-temperature thermostat in close analogy to the Car-Parrinello molecular dynamics scheme. A special configurational -bias Monte Carlo algorithm, called CBMC-(H2O)n, is currently being developed. This algorithm replaces a cluster of water molecules with a larger solute molecule and facilitates the simulation of phase equilibria involving aqueous phases. Finally, these researchers are currently developing a general version of fixed-endpoint configurational-bias Monte Carlo FE-CBMC that allows for efficient conformational relaxation of the interior of polymers with arbitrary complexity modeled by standard molecular mechanics force fields.

This information is available in alternative formats upon request by individuals with disabilities. Please send email to alt-format@msi.umn.edu or call 612-624-0528.

HOME | QUESTIONS | FEEDBACK
Employment | Events | Links | People | Programs | Publications | Support | Welcome