Gas separation in microporous materials: A computational study of the influence of structural features on the selectivity
Final Report Abstract
Computational chemistry methods at different levels of theory were used to obtain new insights into the adsorption of gas molecules in ordered microporous materials, with a focus on zeolites and metal-organic frameworks (MOFs). These materials hold much promise as adsorbents for use in gas separation applications, both in processes that are already employed on an industrial scale, e.g. in the separation of hydrocarbon mixtures, and in novel applications, e.g. for the removal of carbon dioxide from exhaust gases. The emphasis of this project was placed on the influence of pore topology and specific interaction sites (extra-framework cations, coordinatively unsaturated metal sites, functional groups) on the affinity towards different guest molecules of interest. A systematic understanding of these factors should be very helpful for the design and improvement of adsorbents for separation applications. The first part of the project concentrated on the influence of framework topology on the CO2/N2 separation properties of all-silica zeolites. Purely siliceous systems were chosen as suitable model systems to single out the impact of the topology, keeping the framework composition constant. Force-field based grand-canonical Monte Carlo (GCMC) simulations were the method of choice to predict macroscopic (adsorption isotherms, selectivities) and microscopic (adsorption sites) properties of interest. In an initial set of 18 zeolite frameworks, an analysis of the adsorption sites revealed that the strongest interaction with CO2 is reached in systems with narrow cages that are accessible through eight-ring windows. The crystallochemical concept of natural tilings was employed to identify other zeolite frameworks with similar structural features, and additional simulations were performed for these systems. High CO2/N2 selectivities and reasonably high CO2 working capacities were reached in a number of these systems. The direct relationship between the presence of certain building units and the high affinity towards CO2 shows how the natural tiling can be used to find promising candidate materials for a given task. The remaining parts of the projects relied to a large extent on density-functional theory calculations, in most cases including a semiempirical dispersion correction (DFT-D). In particular, the adsorption of small molecules (CO2, N2, CH4) and saturated and unsaturated C2 and C3 hydrocarbons at the extra-framework cations of chabazite-type systems was studied. The DFT-D calculations provided insights into the dominant interactions, and allowed for predictions of cation types that should have the most beneficial impact on the separation properties. Another series of studies, which were carried out in the framework of different collaborations, investigated the adsorption of small molecules at the coordinatively unsaturated metal sites of MOFs. For example, the development of a DFT-derived potential model to describe the interaction between alkenes and unsaturated Cu sites in the MOF Cu3(1,3,5-benzene-tricarboxylate)2 led to a greatly improved prediction of alkene adsorption isotherms when compared to experiment. In the final part of the project, DFT-D calculations were used to gain insights into the influence of the substituent attached to the organic linker on the relative affinity for CO2 over H2 in a zeolitic imidazolate framework (ZIF). Nitro- and aldehyde-functionalised systems were identified as particularly promising materials, primarily due to the increased electrostatic interaction between the CO2 molecule and the substituents. It was the key goal of this project to enhance the systematic understanding of structural factors on the affinity of adsorbents towards small guest molecules. Thus, the main lines of research concentrated on comparative studies of idealised model systems, rather than attempting to mimick experimentally accessible systems as closely as possible. As these systematic insights are expected to exhibit a reasonable degree of transferability, the key findings of this project should aid the rational development (or, ultimately, design) of novel microporous adsorbents for gas separations.
Publications
- Influence of Zeolite Topology on CO2/N2 Separation Behavior: Force-Field Simulations using a DFT-Derived Charge Model, J. Phys. Chem. C 2012, 116, 26449-26463
M. Fischer, R. G. Bell
- Identifying Promising Zeolite Frameworks for Separation Applications: A Building-Block-Based Approach, J. Phys. Chem. C 2013, 117, 17099-17110
M. Fischer, R. G. Bell
- Cation-exchanged SAPO-34 for adsorption-based hydrocarbon separations: Predictions from dispersion-corrected DFT calculations, Phys. Chem. Chem. Phys. 2014
M. Fischer, R. G. Bell
(See online at https://doi.org/10.1039/c4cp01049c) - Interaction of hydrogen and carbon dioxide with sod-type zeolitic imidazolate frameworks: a periodic DFT-D study, CrystEngComm 2014, 16, 1934-1949
M. Fischer, R. G. Bell
(See online at https://doi.org/10.1039/c3ce42209g)
