Simulation of phase transformations in soft matter

Soft condensed matter systems such as polymers, micellar solutions, colloidal suspensions, liquid crystals and gels are of great importance both from a fundamental as well as technological point of view. Such systems often display an exceptionally rich and intricate phase behavior originating in a delicate balance between entropic effects and weak interaction energies. Under suitable conditions, soft matter can self-organize into novel structures that can be tailored by tuning the interactions between the constituents of the material. In the present project we will use computer simulations to study the formation of newly discovered crystalline clustering phases that may occur in polymeric colloids as well as structural phase transformations in nanocrystals. Since these processes involve rare nucleation events, special numerical techniques are required. Here we will use the transition path sampling method, based on a statistical mechanics of trajectories, for this purpose. An important part of the project will consist in developing and implementing an enhanced way to generate trajectories tailored for the simulation of nucleation phenomena and a new method to calculate nucleation rates. We expect this project to increase our understanding of the nucleation of phase transitions in soft matter systems and nanocrystals and to result in a new transition path sampling technique generally applicable to diffusive rare events.

Soft condensed matter systems such as polymers, colloidal suspensions, liquid crystals and gels are of great importance both from a fundamental as well as technological point of view. These materials can self-assemble into a rich variety of structures driven by a delicate balance between weak interactions and entropic effects. To understand the mechanism behind such a rich phase behavior with the help of computer simulations was the central goal of this project. Computer simulations of phase transformations and self-assembly processes occurring in soft-matter are computationally very challenging, because they are characterized by widely different time scales. This so called rare event problem arises from energetic barriers, comparable to high mountain passes on an alpine landscape, which can slow down the motion of the system dramatically. In this project we have developed improved computer simulation algorithms to address this rare event issue and carry out the computer simulation of phase transitions in soft and nanoscale matter on high performance computers. From such simulations one can obtain detailed information on the transformation mechanism with atomistic resolution, providing the information required to understand, and eventually control the behavior of these systems. Applying these computer simulation methods, we have studied the crystallization of supercooled liquid mixtures cooled below their freezing point. We discovered that the crystallization rate is much slower than that of pure liquids, possibly due to the occurrence of icosahedral structures that are incommensurate with the preferred structure of the crystalline state. Using the same algorithms we have also studied the microscopic mechanism for the demixing of a liquid mixture that occurs at low temperatures. We find that the transition proceeds via an intermediate state that differs in density from both the initial as well as the final state. Another focus of the project was on the behavior of water inside narrow pores. In such nanoscale confinement, which can be realized with carbon nanotubes, water has unusual physical properties, which hold promise for a wide range of biomimetic and nanotechnological applications. Based on our simulations, we predict that at low-temperatures interactions of water molecules located in different pores lead to a transition to an antiferroelectric state. This effect should also influence proton conduction of these membranes, which are of technological interest as materials for hydrogen fuel cells. To elucidate the microscopic origin of friction we also simulated a monolayer of particles on a regular substrate and driven by an external force, and explained the rich dynamical behavior observed in recent experiment.