Mesoporous inorganic membranes, composed of materials such as silica and alumina and having pore sizes on the order of 2 to 50 nanometers, have significant potential for performing separations of mixtures of small molecules. Important examples include the removal of carbon dioxide from flue gases and the recovery of ethanol from fermentation broths. Chemists and materials scientists now have an amazing amount of control over the geometry and surface chemistry of the pores when synthesizing mesoporous inorganic membranes. However, actually knowing what pore geometry and chemistry to choose for a given separation remains an outstanding problem, partially due to the lack of appropriate models that connect detailed material structure to membrane flow rates.
In this project, a new modeling approach is proposed to capture the complex adsorption and flow mechanisms that take place inside the membrane pores during separation operations. The modeling, based dynamic mean field theory (DMFT), retains molecular-level detail while predicting membrane performance at the laboratory or industrial scale. A combined program of theoretical development, computer simulation, materials synthesis, and permeation measurement will develop DMFT into a tool that chemists and materials scientists can use to guide the manufacture of new membranes, and engineers can use to model the performance of industrial membrane units. A novel class of mesoporous membrane materials will also be developed as a key part of the project.
The PIs propose a collaborative theoretical/experimental research program that will transform the modeling of separations with mesoporous inorganic membranes through the further development and DMFT. DMFT has had an enormous impact on the closely related application of adsorption in mesoporous materials. With some further development to the dynamic aspects of the theory, DMFT msy also meet the challenge of predicting permeation through mesoporous membranes. The research team for this project spans molecular theory and modeling (Ford, Monson), membrane science and technology (Ford) and the materials science and engineering of mesoporous materials (Fan, Monson).
The intellectual merit of this proposal lies in two main objectives. The first is to extend and develop dynamic mean field theory (DMFT) for quantitatively accurate prediction of permeation of small molecules in mesoporous membranes. This is proposed to be accomplished by (i) establishing the method on lower-dimensional, geometrically simple pore models; and (ii) modifying the dynamics to quantitatively capture relevant transport mechanisms. The second major objective is to apply DMFT to model specific mesoporous membrane permeation experiments. This will be accomplished by (i) synthesizing a set of membranes with controlled pore size and geometry; (ii) comparing DMFT predictions to separation experiments on these membranes; and (iii) using DMFT to improve the accuracy of pore sizes obtained by permporometry, which uses co-permeation of a light gas and a condensable vapor to gain information about pore size.
If successful, the products of this research have the potential to advance the membrane industry in the U.S., especially as it is applied to the traditional and emerging fields of energy production. The proposed work may bridge the statistical thermodynamics, adsorption, and membrane communities while providing the membrane community with a new computational tool for predicting and interpreting permeation through mesoporous membranes. The PIs propose to integrate the research into the undergraduate and graduate curricula at UMass Amherst, through a popular undergraduate nanomaterials elective taught by one of the PIs (Fan) and a core graduate course in statistical thermodynamics taught by one of the other PIs (Ford or Monson). Existing outreach and recruitment programs at UMASS will also be leveraged.
