Glass formation of colloids confined in porous materials

Fluids that are confined within pores do show distinctively different properties than in the bulk: new phase transitions (such as wetting or layering) as well as drastic modifications in the phase behaviour (as suppressing or causing additional phase transitions) are observed. While these effects are reasonably well understood if the pores have a regular structure or geometry (as this is the case in zeolithes or carbon buckey tubes), the situation is more complex in disordered porous materials (such as aerogels, Vycor, or clay): here we are faced by a complex interplay between confinement, the random pore structure, and the connectivity of the pores. This particular behaviour of fluids confined in porous materials is of high technological and geophysical relevance: porous materials are widely used in the chemical, food and pharmaceutical industries for pollution control, mixture separation, and as catalysts. Further, since many rock and soil formations are porous, the topics of oil recovery, gas field technology, or removal of pollutants from ground water are important. As the design of such processes is at present largely empirical, a deeper understanding of these phenomena is not only of academic but also of technological relevance. In the present project we want to study the dynamic properties of colloids confined in porous materials and plan to put particular emphasis on the glass transitions. Our decision to investigate colloidal systems is based on several aspects: (i) most systems of technological relevance are colloids; (ii) the fact that effective interactions between colloidal particles can be easily `tailored` in the lab offers ideal and unprecedented possibilities for close cooperation between experiment and theory; and (iii) there is clear evidence that colloidal systems do show distinctively different properties than atomic or molecular systems which has led, for instance, in their phase behaviour to many unexpected results. The fact that the porous matrix does have a distinct influence on the (static) properties of the fluid makes us believe that this also holds for the dynamic behaviour. And indeed, investigations on well-defined geometries have indicated that this is the case. However, there is a lack of theoretical investigations on fluids that are confined in disordered porous matrices. While experiments date back to the early 1990s, first simulations have been performed only a few years ago and it was only last year that a theoretical concept has been put forward. This new technique makes theoretical investigations of these properties possible and will be the basis of our investigations. It is a combination of two theories that have proved reliable in their respective fields. On one side mode coupling theory, a very efficient tool to study dynamic behaviour and glass transitions in (bulk) fluids and on the other side the replica Ornstein-Zernike formalism to study the static properties of fluids confined in disordered porous materials. The merged concept thus covers both aspects of the project. With our work we want to provide a deeper insight how the presence of the matrix influences the dynamic behaviour, in particular the glass formation, of a colloidal fluid. The many new phenomena encountered in colloidal systems in combination with the drastic influence of the matrix on the static properties of the fluid represent nearly a guarantee that we will find many surprising effects in this widely unexplored topic. The close cooperation between theory and experiment that colloidal systems offer will help us to initiate collaborations with experimentalists, which puts forward the strong interdisciplinary character of the project.

Fluids that are confined in disordered porous materials do show distinctively different properties than in the bulk: new phase transitions (such as wetting or layering) emerge and substantial changes in the phase behaviour are observed. While such features are reasonably well understood in the case that the pores have a regular structure, the situation is entirely different when the pores are formed by a disordered matrix: here, the aforementioned phenomena are the result of a complex interplay between confinement, the random pore structure, and the connectivity of the pores. In addition, a realistic modelisation of such systems (as required in simulational and theoretical approaches) represents a formidable challenge In view of the fact that such systems are not only of purely academic interest, but do play a prominent role in many applied problems (ranging from technological applications, such as oil recovery, to biophysical problems, such as crowding in biological cells), a deeper understanding of the physical properties of fluids confined in disordered porous materials is of particular relevance. It was the dedicated aim of this proposal to contribute to yet open problems in this field. In the present project we have studied the dynamic properties of colloidal particles confined in disordered porous materials, putting particular emphasis on possible glass transition scenarios. Focusing on a suitable model system of colloids, we have carried out extensive computer simulations to study the dynamics of the system on the basis of the dynamic correlation functions. Indeed, the ensuing dynamic phase diagram shows a broad spectrum of glass transition scenarios which can be identified with our simulational approach. With a detailed geometrical analysis of the voids and traps that are accessible to the fluid particles, we could filter out down to the particle level those mechanisms that are responsible for the slowing down of the dynamics. From these investigations (which were complemented and confirmed on a qualitative level by theoretical approaches) we have gained a thorough insight into the complex mechanisms which govern the dynamics of this system. The implementation of a novel simulation technique allows us to investigate also the out-of-equilibrium properties of fluids confined in disordered porous materials: in this concept, the hydrodynamic interactions, induced by the microscopic solvent particles, are taken explicitly and faithfully into account. Thus a quantitative description of flow experiments with tracer particles through a disordered, porous matrix (as they are for instance of relevance in oil recovery scenarios) can now be modeled on the computer and makes a direct comparison with experimental investigations possible.