61st Annual Report on Research 2016

The separation of alkane isomers is a very important process in the petroleum industry. For instance, cryogenic distillation is often used for the separation of hexane isomers to increase the octane ratings in gasoline. Similarly, nanoporous materials, such as e.g. zeolitic materials, can also be used in separation applications. Here we focus on recently synthesized nanoporous materials known as Metal-Organic Frameworks (MOFs) and Covalent Organic Frameworks (COFs). These materials are increasingly studied for their potential applications in gas storage, separation and catalysis. MOFs are very promising materials since their pore connectivity, structure and dimensions can be controlled by varying the linkers, ligands and metals used when synthesizing the material. COFs differ from MOFs as the organic building blocks of COFs are held together by covalent bonds rather than by metal ions in MOFs. This makes COFs especially promising for practical applications as lightweight materials for gas storage and separation. Both MOFs and COFs can be potentially used for octane boosting, if e.g. alkanes with a lower research octane number (RON) are strongly adsorbed while alkanes with a higher RON are weakly adsorbed. The aim of this research is to develop a new molecular simulation method leading to a complete thermodynamic analysis of the adsorption of alkane mixtures in MOFs and COFs. The new method relies on determining the density of states, and thus the partition function, of adsorbed alkanes through the combination of state-of-the-art efficient sampling methods. In practice, this can be achieved through the use of a method recently developed in our research group, the Expanded Wang-Landau simulations, combined with Configurational Bias Monte Carlo methods. The specific advantage of the new method is the following. Once the partition function for the sorbates is known, the formalism of statistical mechanics leads to a direct evaluation of all thermodynamic properties of the adsorbed phase, including e.g. the entropy and the free energy of desorption. In particular, the desorption free energy is a key property, since it corresponds to the minimum isothermal work for regenerating the adsorbent. This property is especially useful in comparing the performance of adsorbents, since most of the operating costs are associated with degassing the adsorbent in preparation for the next cycle.

In the first year of the grant, we have carried out two main developments of the Expanded Wang-Landau simulation methods. The first development has consisted in extending the EWL method to flexible molecules that compose the adsorbed phases covered in this grant. This was achieved by combining the EWL method with Configurational Bias Monte Carlo schemes, that allow to sample the conformational changes undergone by flexible molecules, such as alkane chains, upon mixing and adsorption. This leads to an accurate evaluation of the partition function of the adsorbed phases, which, in turn, provides access, through the formalism of statistical mechanics, to key properties, like the free energy of desorption. This opens the door to a full understanding of the effect of the MOF on molecular selectivity during mixture adsorption, and of the trade-off between the selectivity and the operating costs to regenerate the adsorbent. The second key development carried out in the first year of the grant is the extension of the EWL method beyond classical systems to quantum models. We focused here on a quantum model that is very popular in materials science, i.e. systems modeled within a tight-binding approach. This leads to a considerably more flexible approach, that can incorporate directly quantum effects on the thermodynamic properties of adsorption.

The developments carried out in the first year of the grant have allowed us to extend the EWL method to flexible molecules, such as linear and branched alkanes, and to increase considerably the type of model that can used in conjunction with the EWL method, which now range from classical force fields to quantum tight-binding systems. The plan for the second year of the grant is to apply the EWL method to develop a full picture for the molecular selectivity of metal-organic frameworks during the adsorption of alkane mixtures, to characterize the operating costs through the evaluation of the desorption free energy and, on this basis, to screen different metal- and covalent-organic frameworks for such applications. In addition, the transport coefficients for alkane mixtures in MOFs and COFs will be evaluated to analyze the suitability of these nanoporous materials to serve as membranes for gas separation.