Molecular Simulations of Surfactant Interfaces

This thesis demonstrates the use of classical molecular simulations to provide a fundamental understanding of the molecular interactions in surfactant solutions that govern the macroscopic behavior of the system. Two directly related projects are highlighted as follows:

• Understanding the Cytotoxic Mechanisms of Aqueous Ionic Liquids Towards a Model Cell Membrane

• Predicting the Adsorption Isotherms of Nonionic Surfactants at the Air-Water Interface

In the former (comprising the majority of this thesis), molecular dynamics simulations of aqueous ionic liquids (ILs) are performed with a lipid bilayer as a model cell membrane. While ionic liquids have been the subject of extensive research in the past decade, the risks associated with their unknown toxicities have persisted as a major bottleneck in their commercialization. This work provides crucial molecular-level detail on the cytotoxic mechanisms of ionic liquids towards biological membranes that can help guide the rational design of novel, nontoxic ionic liquids. Both atomistic and coarse-grained simulations are used to investigate the molecular mechanistic pathways through which ILs can disrupt a model cell membrane, consistent with the experimental observations of a supported lipid bilayer in aqueous IL media.

In the latter, a newly-developed Monte Carlo scheme for simulations of surfactant solutions is presented. Specifically, this scheme is used to predict bulk concentrations and interfacial tensions at various coverages of surfactants at the air-water interface. Since the actual concentration regime of these systems of interest are typically very dilute (<< 10−5 mol. frac.), Monte Carlo simulations can, in principle, provide an advantage over molecular dynamics simulations, as insertion/deletion moves allow the system to reach concentrations below those limited by system size and within reasonable computational cost. In performing these simulations, state-of-the-art Monte Carlo moves are implemented in the Gibbs ensemble framework. These moves vastly improve sampling and help overcome the classical “insertion problem” of large molecules in a condensed medium. The future outlook of this methodology as well as its limitations are also discussed in detail.