468015 Unraveling the Role of Pore Topology and Chemical Functionality on the Carbon Capture Performance of Metal-Organic Frameworks
The chemical functionality of MOFs has long been deemed key to enhance CO2 adsorption in MOFs. However, since the incorporation of functional group also alters MOF textural properties, conflicting information on the positive or negative impact of a chemical moiety on CO2 adsorption often arises. In addition, elucidating optimal functionalities experimentally has been difficult, because comprehensive sets of MOFs with identical structures but different functionality are not readily synthesizable or otherwise equally stable upon activation and testing. Similar issues are encountered in attempting to elucidate optimal pore topologies for CO2 adsorption. The above roadblocks can be circumvented using molecular simulation to evaluate a suitable, comprehensive, diverse set of “perfect MOFs” but the challenge is to efficiently create and evaluate such predictably large set in the computer.
In this work, we created in silico 426 MOFs carrying different simple chemical functionalities, based on 28 “parent” MOFs of 16 different topologies, using our recently developed automated MOF construction algorithm. Since CO2 adsorption is likely to be highly sensitive to Coulombic interactions, we explored different strategies to assign partial charges to MOF atoms, including a new building block-based strategy that we introduce in this work to be compatible with our MOF construction algorithm. Based on our simulations, we provide insights into how and under what conditions different chemical functionalities and topologies are likely to enhance CO2 adsorption. For instance, from our data analysis we find that at pressures relevant to post-combustion CO2 capture, alcohol and thiol functionalities tend to outperform other moieties often suggested to enhance CO2 adsorption such as amino functionalities. Additionally, we present potentially attractive MOF synthesis targets based on CO2 adsorption properties predicted during our computational screening.
See more of this Group/Topical: Computational Molecular Science and Engineering Forum