Harnessing Hydrogen by Optimized Ion-Conductive Metal-Organic Frameworks

The development of fuel cell technology capable of efficiently converting hydrogen into electricity is considered a critical component in the transition away from CO2-emitting energy sources. Ion exchange membrane fuel cells are gathering considerable interest due to their low electricity usage. However, this technology's full commercial potential is held back by the performance-critical ion exchange membranes (IEM) not meeting the challenges of simultaneously achieving high ion conductivity, low manufacturing cost, and high robustness. Metal-organic frameworks (MOFs), a class of porous materials, exhibit channels that facilitate ion conduction, high structural stability, and chemical tunability, making MOFs a promising material for the next generation of IEMs. To exploit the full potential of MOFs, i.e., to rationally design MOFs, there is a knowledge need for understanding the underlying molecular mechanisms behind ion conduction inside MOFs. To satisfy this knowledge need, we will, in this project, develop a computational approach to model, understand, and optimize the ion conductivity in MOFs. We will develop from first principles machine-learning models capable of describing on the molecular scale the dynamics of ions in MOFs. From the simulated real-time dynamics of the ions, we will extract the conductivity of the ion in the MOF and validate it against experiments. By investigating a varied set of highly conductive MOFs, we aim to understand what makes a MOF a good ion conductor. By applying the molecular insight from the real-time dynamics in the MD simulations, we aim to discover high-conductivity MOFs with potential as IEMs in fuel cells.

The development of fuel cell technology capable of efficiently converting hydrogen into electricity is considered a critical component in the transition away from CO2-emitting energy sources. In this regard, ion exchange membrane fuel cells are gathering considerable interest due to their high electricity conversion rates. However, this technology's full commercial potential is held back by the performance-critical ion exchange membranes (IEM) failing to meet the challenges of achieving high ion conductivity, low manufacturing cost, and high robustness. Porous metal-organic frameworks (MOFs) exhibit channels that facilitate ion conduction, high structural stability, and chemical tunability, making MOFs a promising material for the next generation of IEMs. To exploit the full potential of MOFs, i.e., to rationally design MOFs, there is a knowledge need for understanding the underlying molecular mechanisms behind ion conduction inside MOFs. To satisfy this knowledge need, in H2O-MOF, we will develop a computational approach to model, understand, and optimize the ion conductivity in MOFs. Specifically, we will parameterize machine learning potentials for ions in MOFs on first-principles calculations to speed up molecular dynamics (MD) simulations by about three orders of magnitude. From the simulated real-time dynamics of the ions, we will extract the conductivity of the ion in the MOF from the diffusion coefficient through the Nernst-Einstein relation and validate it against experiments. By developing this modeling approach, we aim to describe ion conductivity in MOFs from first principles. By investigating a varied set of four highly conductive MOFs, we aim to understand what makes a MOF a good ion conductor. By applying the molecular insight from the real-time dynamics in the MD simulations, we aim to discover high-conductivity MOFs from first principles.