On a New Mechanism for Separating two Components in a Stationary Flow through Mesopores

When simulating the behavior of fluids in a stationary flow through mesopores we have observed a phenomenon that may prove useful in some cases as basis for separating fluid components. The scheme works at constant temperature which makes it energy efficient as are other schemes like (molecular) sieves or chromatography. Sieves rely on differences in molecular size and chromatography on different affinity of components to the solid material of the ‘packing’. The scheme presented here may sometimes complement the established techniques in that it is based on a different mechanism. The fluids to be separated can have the same molecular size and the same affinity to solid material they are in contact with. The only requirement for the scheme to work is that the miscibility behavior varies somewhat with pressure or density. From literature it is known that virtually any mixture reacts on strong variations of pressure. Even a mixture that behaves almost ideally at ambient pressure will show slight deviations from ideal miscibility when exposed to extreme pressure. The strong differences in pressure are not created by external means but by exploiting the spontaneous behavior of fluids in mesopores. If the experiment is designed correctly, strong pressure gradients show up in mesopores that are far beyond any gradient that could be established by technical means. Our simulations are carried out for situations where pressure inside the pores varies between a few hundred bar positive pressure and a few hundred bar negative pressure while the pressure in the gas phase outside the pores amounts to ca.170 mbar.

Progress in understanding fluids in mesopores

An important feature of porous materials is the adsorption hysteresis: the amount of an atomic or molecular species adsorbed from the gas phase is not only dependent on the gas pressure, but may depend in certain ranges of pressure on the history. Thus, the system may respond in different ways to identical experimental conditions which seems to contradict classical thermody-namics. While the phenomenon is known since about a century, it has not yet found a consistent theoretical description. In the pres-ent talk, we will-based on results of computer simulations-formulate rules that provide a consistent basis for the behavior of confined systems, or even for inhomogeneous systems in general. In other words, we present a new theory (confined thermodynamics) with its own definitions and rules. It will turn out, that hysteretic behavior does not impose a conceptual challenge any more, but follows in a natural way from these rules. The approach which is employed in the simulations is very akin to the density functional method. All quantities defined develop into the standard thermodynamic expressions when the density of amount becomes homogeneous.The second part of the talk is devoted to the potential for practical use. It turns out that the new theory does not only remove conceptual problems, but at the same time opens the route to a number of new states found in porous systems which may lead to im-proved applications. In particular we will focus on the possibility to drive a fluid in a pore into exotic states with negative pressure, provided one has full control over the phenomenon of adsorption hysteresis. Negative pressure states are in principal known since the time of Torricelli and they have been in the literature as experimentally accessible situations. Still, they have not been turned into practical usefulness which is likely to be caused by the notion of their metastability in macroscopic systems. However, fluids con-fined to nanopores have been proven to show reproducible behaviour. The present time appears to be suited for exploring the new ap-plications resting in fluid/pore systems: since about a decade material scientists have started to prepare pores with increasing accura-cy from an increasing variety of substances. On the other hand, the new theory presented in the first part of the talk provides the tool to drive a fluid/pore system reliably into any of the exotic states found within a hysteresis loop. Prospects of a few applications will be discussed.