Complementary use of computer simulations and molecular-thermodynamic theory to model surfactant and solubilizate self-assembly

Abstract

Surfactants, or surface active agents, are used in many pharmaceutical, industrial, and environmental applications. Selection of the appropriate surfactant or mixture of surfactants for any given application is driven by the need to control bulk solution micellization and solubilization characteristics. The goal of this thesis has been to develop computer simulations and molecular-thermodynamic modeling approaches to predict these solution characteristics based on knowledge of surfactant and solubilizate chemical structure. The ability to make such predictions would give formulators in industry the ability to design and optimize surfactant formulations with a minimum of effort and expense. This thesis has explored the application of three theoretical approaches to model surfactant micellization and micellar solubilization. The first theoretical approach involves the use of computer simulations (CS) to obtain input parameters for molecular-thermodynamic (MT) modeling of surfactant micellization and micellar solubilization. This approach was motivated by the limitations inherent in computer simulations (the high computational expense of modeling self-assembly) and in MT modeling approaches (their restriction to structurally and chemically simple surfactants and solubilizates).

(cont.) A key input required for traditional MT modeling is the identification of the hydrated and the unhydrated portions (head and tail) of surfactants and solubilizates in a self-assembled micellar aggregate. By conducting simulations of surfactants and solubilizates at an oil/water interface (modeling the micelle core/water interface) or in a micellar environment, I have determined head and tail input parameters for simple and complex surfactants and solubilizates. This information has been successfully used as an input to MT modeling, and has been shown to extend the applicability of the traditional MT modeling approach to more complex surfactant and solubilizate systems than had been possible to date. A wide range of surfactant and solubilizate systems have been modeled with this approach, including ionic, zwitterionic, and nonionic surfactant/solubilizate systems. For each of the systems modeled, theoretical predictions were in reasonable agreement with the experimental data. A novel, alternative approach has also been developed to more accurately quantify the hydrophobic driving force for micelle formation by using atomistic molecular dynamics (MD) simulations to quantify the hydration changes that take place during micelle self-assembly.

(cont.) This new approach is referred to as the computer simulation/molecular-thermodynamic (CS-MT) model. In the CS-MT model, hydration information determined through computer simulation is used in a new MT model to quantify the hydrophobic effect, which is decomposed into two components: 9dehydr, the free-energy change associated with the dehydration of hydrophobic groups that accompanies aggregate self-assembly, and 9ydr, the change in hydration free energy experienced during aggregate self-assembly. The CS-MT model is formulated to allow the prediction of the free-energy change associated with the formation of aggregates of any shape and size after performing only two computer simulations if one of the surfactant/solubilizate in bulk water and the second of the surfactant/solubilizate in an aggregate of arbitrary shape and size. The CS-MT modeling approach has been validated by using it to model the formation of oil aggregates, the micellization behavior of nonionic surfactants in aqueous solution, and the micellization behavior of ionic and zwitterionic surfactants in aqueous solution. For each of the systems modeled, the CS-MT model predictions were in reasonable agreement with the experimental data, and in almost all cases were in better agreement with the experimental data than the predictions of the traditional MT model.

(cont.) The second theoretical approach explored in this thesis is the application of computer simulation free-energy (FE) methods to quantify the thermodynamics of mixed micelle formation. In this theoretical approach, referred to as the CS-FE/MT modeling approach, the traditional MT modeling approach, or experimental data, is first used to determine the free energy of formation of a pure (single) surfactant micelle. Subsequently, computer simulations are used to determine the free-energy change associated with alchemically changing the identity of individual surfactants present in the micelle to that of a second surfactant or solubilizate. This free-energy change, when added to the free energy of single surfactant micellization, yields the free energy associated with mixed micelle formation. The free energy of mixed micelle formation can then be used in the context of a thermodynamic description of the micellar solution to predict bulk solution properties such as the CMC and the equilibrium composition of the mixed micelle. The CS-FE/MT model has been used to model both binary surfactant micellization and micellar solubilization. The CS-FE/MT model was shown to be most accurate when the chemical structures of the mixed micelle components were similar and when small alchemical transformations were performed.

(cont.) The third theoretical approach explored in this thesis is the use of all-atomistic computer simulations to make direct predictions of surfactant solution properties. Although the computational expense associated with atomistic-level MD simulations restricts their use to the evaluation of a limited subset of surfactant solution properties, these simulations can provide significant insight into the structural characteristics of preformed surfactant aggregates and the self-assembly behavior of surfactant molecules over limited timescales. Simulation of monolayers of a homologous series of structurally complex fiuorosurfactants has been conducted in order to explore their behavior at a water/air interface and the origin of their ability to reduce surface tension. In addition, atomistic-level MD simulations have been conducted to study the self-assembly behavior of the triterpenoids asiatic acid (AA) and madecassic acid (MA) in aqueous solution. The computer simulation results were used to obtain information about: i) the kinetics of micelle formation, ii) the structural characteristics of the self-assembled micelles, and iii) micellization thermodynamics.

(cont.) This thesis presents a detailed, atomistic-level computer simulation and molecular-thermodynamic investigation of the micellar solution behavior of nonionic, zwitterionic, and ionic surfactants in aqueous solutions, as well as of the aqueous micellar solubilization of solubilizates by surfactants. It is hoped that the approaches developed in this thesis to use computer simulations and molecular-thermodynamic theory in a complementary way will not only extend our ability to make accurate predictions of surfactant solution behavior, but will also contribute to our fundamental knowledge of the solution behavior of surfactants and solubilizates. It is further hoped that this thesis will provide a solid foundation for future research in the area of surfactant science, and, more generally, that it will assist future researchers working to connect atomistic-level computer simulation methods with continuum thermodynamic models.