Minors: Energy Studies, Bioengineering
Faculty Advisor: Yamil J. Colón, Department of Chemical and Biomolecular Engineering
Research Area: Energy Conversion and Efficiency

Diffusion of Methane and CO2 in CuBTC and IRMOF 1 (Spring 2024)

Methane and Carbon Dioxide are two of the most prominent greenhouse gasses in our atmosphere. We need to understand the chemistry of these two molecules in order to efficiently engineer a reduction in our emissions and the impact of our current emissions. Specifically, this project aims to understand the diffusivity of Methane and CO2 in CuBTC and IRMOF 1. These chemicals belong to a class of molecules called Metal Organic Frameworks (MOFs). MOFs are crystalline molecules known for their high surface area and pore volume. They are composed of organic linker molecules and metal nodes, offering a lot of variability. It is for these reasons that MOFs are heavily researched due to their capability for energy and gas storage (Introduction to Metal-Organic Frameworks, Yaghi, doi: 10.1021/cr300014x). There is a lot of research going into which MOFs are the most suitable for various tasks. CuBTC and IRMOF 1 in particular have been considered for their potential uses in high performance filtration, gas storage and flue gas scrubbing. Knowing the diffusivity for Methane and CO2 tells us exactly how well these materials work in emerging processes that are shaping the energy industry, such as natural gas management, gas separations, and carbon capture.

The objective of this project is to develop a model of the diffusivities of these gasses in the MOFs. This will be obtained by doing Molecular Dynamics simulations using a canonical ensemble in LAMMPS. To get the number of moles for these simulations we run Monte Carlo adsorption simulations. This information allows us to run the molecular dynamic simulations. These simulations will be used to generate a training data set for a machine learning model. This is used alongside an active learning protocol to select specific points to evaluate to make the model accurate with fewer points, and therefore it takes less time to simulate. We can then use these models to better understand how the materials interact with Methane and CO2 and how they could be used in the various processes that are shaping the energy industry today.

My role would be to code and run the Monte Carlo adsorption simulations, which would then be used to obtain the number of moles. Then I will use this information to run the Molecular dynamics diffusivity simulations, which are used to generate the training data set. I’d work under the graduate student Etinosa Osaro, using his active learning protocol to navigate the diffusion space. It would be similar to what Osaro has done for adsorption, but this time for diffusivity (Active learning for efficient navigation of multi-component gas adsorption landscapes in a MOF, Colón, doi: 10.1039/D3DD00106G). Then finally I’d analyze the diffusivity model to get useful information about the diffusivities of the gasses in the MOFs, and what this implies about the ways these materials can be used.

Final Report