Molecular Simulations of Reverse Micelles

Reverse micelles formed in apolar media via surfactant self-assembly have applications in protein separation and drug delivery, in addition to providing an aqueous chemistry platform for nanomaterial synthesis and enzymatic reactions. Self-assembly systems composed of lipid species in particular have attracted much attention due to their biocompatibility, and importance in food and pharmaceutical industry. Despite this many fundamental aspects remain poorly understood. Reverse micellisation is sensitive to e.g. the presence of moisture, solvent quality and surfactant structure. The aim of this thesis is to investigate how the aforementioned control factors influence aggregate morphology, as well as, the aggregation propensity and mechanism of lipid surfactants. This is carried out through the use of both all-atom molecular dynamics simulations, as well as, simulations combining a simplified coarse-grained solvent representation with fully atomistic solutes. The simulations show that in the absence of moisture weakly polar surfactants form small aggregates through step-wise aggregation mechanism while more polar lipids can cooperatively aggregate into large reverse micelles. Small amounts of water can trigger cooperative aggregation in weakly aggregating surfactants also, and modify the shape of aggregated structures. Increasing solvent polarity, on the other hand, favours step-wise aggregation by attenuating intersurfactant interactions. Finally, the results of this thesis suggest that the micromorphology of the simulated reverse micelles is sensitive to the simulation model employed, in particular to the tail/solvent mixing energetics and the implicit polarization of dipolar groups. The findings outlined above can be used to better interpret experimental observations, and facilitate understanding and control of reverse micellar systems. In addition, the hybrid resolution simulations of this thesis present an advancement in simulation methodology by expanding coarse-grained solvent representation to lipid species, and by explicitly accounting for electrostatics.