Computational design of electrically conductive metal–organic frameworks: exploring strategies to construct charge transport pathways

Abstract

Metal–organic frameworks, or MOFs, are crystalline nanomaterials synthesized from metal ions or clusters and organic ligands that participate in coordination bonds with each other to form multi-dimensional frameworks. MOFs are well-known for their ultrahigh porosity and chemical tunability, which has made them promising candidate materials for a variety of adsorption-based applications ranging from gas storage and separation to chemical sensing, catalysis, drug delivery, and many more. More recently, it has been shown that electrical conductivity can be induced in MOFs, which have long been firmly regarded as electrical insulators. Such discovery has led to their successful utilization in electronic or electrochemical application fields such as supercapacitors, field-effect transistors, electrocatalysis, chemresistive sensing, etc. One could expect that chemical tunability and high porosity of MOFs would allow these materials to achieve outstanding and unprecedented performances in these new application fields. Nonetheless, progress is stunted by the limited number of electrically conductive MOFs discovered to date, and hence continued discovery of new and improved electrically conductive MOFs is crucial. To this end, this dissertation reports on the computational design of new electrically conductive MOFs by exploring strategies to construct long-range charge transport (CT) pathways. First, it is demonstrated that rational modifications of previously insulating MOFs can newly induce electrical conductivity. As a proof-of-concept, sequential linker installation and guest molecule intercalation are performed in appropriately chosen pre-existing MOFs for the completion of long-range CT pathways. In these systems, it is additionally shown that spatial and energetic matching between the constituents of proposed CT pathways is essential, and to achieve this, metal and/or linker substitution are additionally performed. Next, electrically conductive MOFs with significant framework flexibility are designed for the first time. Newly designed MOFs uniquely exhibit tunable CT properties, characterized by the gradual transition from 1D to 2D CT and significant enhancement in band dispersion as the MOFs traverse through the allowed structural phases. Subsequently, topologically guided construction of 3D π-d conjugated MOFs is presented. In the newly constructed MOFs, high porosity is secured while π-d conjugation between framework components is also effectively retained for long-range CT. Such MOFs are expected to be highly suitable for applications where both high conductivity and porosity are required. Lastly, a deep learning model is developed for the prediction of electronic structures in MOFs with an aim to easily and quickly determine the presence of CT pathways. The model is trained to predict the density of states (DOS) of MOFs, which subsequently allows for the ascertainment of Fermi level and band gap of the material, as well as the identification of framework components that majorly contribute to frontier bands for CT. Based on all the information predicted by the deep learning model, presence of long-range CT pathways can be successfully determined. All in all, design approaches and computational methods newly proposed and developed in this dissertation will significantly expedite the discovery of new electrically conductive MOFs for various electrical and electrochemical applications.