###Machine learning assisted modelling and discovery of materials for low-carbon hydrogen production
Project ID: 2228bd1157 (You will need this ID for your application)
Research Theme: Energy and Decarbonisation
UCL Lead department: Chemical Engineering
Lead Supervisor: Ozgur Yazaydin
Project Summary:
For hydrogen to become a viable zero or low-carbon energy carrier in a decarbonized energy system, shortcomings in its production, storage and transportation need to be addressed. Currently, most of the hydrogen comes from steam reforming of natural gas, a chemical process that releases large amounts of CO2. To produce low-carbon hydrogen in conjunction with steam reforming of natural gas, the by-product CO2 must be separated for indefinite storage or for use in chemical manufacture. Furthermore, hydrogen must be transported in large quantities over long distances, however, as hydrogen has a low energy density by volume, the cost of transportation is high. Using ammonia in liquid form as a carbon-free hydrogen carrier for transport over long distances would have lower costs than transporting it as pure hydrogen. For the end user, however, ammonia must then be converted back to hydrogen and the by-product nitrogen will need to be separated. This project aims to employ machine learning assisted molecular modelling to develop and discover materials that can separate CO2, nitrogen and other impurities from hydrogen. More specifically, the project will be part of an overarching aim to build a software pipeline that will take advantage of advances in machine learning approaches and accelerate different stages of modelling of nanostructured materials for separation processes that are related to applications in the areas of energy and the environment. The ideal candidate should be enthusiastic about making a contribution towards achieving a sustainable and low-carbon energy future. They should have a background in chemical engineering, physics, chemistry, materials science or another related discipline, and have strong fundamentals in mathematics and good computational skills. By the end of the project, it is expected that the candidate will have become an expert in computer programming, materials modelling at the molecular level and different machine learning approaches.