Research Team of Professor Na Jong-geol at the Department of Chemical Engineering and Materials Science Publishes a...
Research Team of Professor Na Jong-geol at the Department of Chemical Engineering and Materials Science Publishes a Paper in Applied Energy
A joint research team at Ewha developed a solvent identification method to reduce greenhouse gas emissions easily and efficiently by applying deep learning technologies and chemical process simulations. Incorporating the findings of this study, a paper titled “Learning the properties of a water-lean amine solvent from carbon capture pilot experiments” was published in Applied Energy (SCIE), an academic journal within the top 3.85% of JCR in the engineering and chemical fields, on February 1 (Monday), 2021.
Professor Na Jong-geol and his joint research team from the Department of Chemical Engineering and Materials Science at ELTEC College of Engineering developed a methodology to estimate the properties of a solvent that can efficiently absorb CO2, integrating deep learning technologies and chemical simulations to minimize the need for experiments.
The Ewha research team developed a new water-lean amine solvent that can capture a large amount of CO2 using a lower amount of energy compared with existing solvents. In order to ensure the proper application of this newly developed solvent, it is necessary to develop an optimized process that is aligned with the thermodynamic and kinetic characteristics of the solvent. The research team applied a deep learning-based hybrid Bayesian inference technique to the pilot-scale tests, through which the team made it easier to infer the properties of a solvent using solely the pilot process with many uncertainties, and enabled the simultaneous implementation of a large-scale process optimization and analyses of the physiochemical properties of a solvent. This advancement is expected to reduce costs for processing greenhouse gases and lead to a large-scale reduction in CO2 emissions.
This study was jointly conducted by Professor Na Jong-geol (co-first author) at Ewha, senior researcher Lee Ung (corresponding author) at the Korea Institute of Science and Technology, and Dr. Kim Jeong-nam (co-first author), with the support of the “Carbon to X technology development project for the production of value-added chemicals” under the National Research Foundation of Korea (NRF).
Professor Na remarked, “The key of this study was to develop a methodology to simultaneously carry out the properties analysis and optimization processes, which were previously conducted in sequence,” and added, “This represents a successful case of a convergence study combining AI and domain knowledge to apply deep learning technology onto chemical process technology as a step toward achieving zero carbon emissions.” In particular, he emphasized, “This study is expected to greatly improve the speed of product development through highly advanced design techniques including AI-based solvents, materials, and catalysts,” voicing his aspirations to lead the field in the future.