30 Dec 2024
Recently, the team led by Associate Professor, Dr. Lifeng Ding, from the Department of Chemistry and Materials Science, School of Science, at Xi’an Jiaotong-Liverpool University (XJTLU) published significant research findings in the prestigious journal <<Separation and Purification Technology>>.
Dr. Xiaoyu Wu is the lead author of the study and supervised by Dr Lifeng Ding from the School of Science at XJTLU. The study introduces the innovative "Pore+" concept, which assigns unique "fingerprints" to porous materials (e.g., metal-organic frameworks), dramatically enhancing the accuracy and interpretability of machine learning models in predicting the adsorption and separation performance of porous materials.
AI-Powered Materials Science: Pore+ as a Material Feature Engineering
In this research, the team developed an enhanced descriptor system, "Pore+," to design material features for AI models. Pore+ precisely captures the geometric structure and chemical heterogeneity of porous material pore spaces. Compared with traditional descriptors, Pore+ excels in handling complex datasets, improving predictive performance, and enabling feature analysis. This groundbreaking approach signifies the deep integration of AI technologies with materials science, paving a novel path for designing high-performance adsorption and separation materials.
"Our study demonstrates the critical role of advanced feature engineering in designing new materials with specific functionalities. Pore+ not only empowers AI models to assist in developing novel porous materials but also offers new perspectives for addressing global environmental and energy challenges." Dr Lifeng Ding explained.
Dr. Xiaoyu Wu, the lead author of this study, successfully completed his Ph.D. at Xi’an Jiaotong-Liverpool University and has joined the National University of Singapore as a postdoctoral researcher. Under the guidance of Dr. Lifeng Ding, Dr. Wu has gained extensive experience across multiple fields, including molecular dynamics, large-scale Monte Carlo methods, density functional theory, and machine learning, showcasing the team's international influence in cultivating interdisciplinary talent.
About the Team and Recent Advances
As the principal investigator of the team, Dr. Lifeng Ding has built a solid research foundation in materials design and computational modelling. Before joining XJTLU, he served as an EU Marie Curie Fellow at the University of Surrey, UK, focusing on computational studies of materials for energy storage and carbon capture. Recently, his team has received two grants from the National Natural Science Foundation of China (NSFC), laying a strong foundation for the development and application of novel materials.
Ph.D. Recruitment: Join Us in Shaping the Future
The team is dedicated to advancing the integration of AI technologies with materials science, exploring cutting-edge interdisciplinary research at the intersection of chemistry, physics, computational science, materials chemistry, and mathematics. Multiple Ph.D. scholarships are now available, and outstanding candidates with backgrounds in chemistry, materials science, or AI are encouraged to apply. Join this vibrant research group, participate in pioneering projects, and contribute to solving global technological challenges.
Contact Information
Dr. Lifeng Ding (Lifeng.Ding@xjtlu.edu.cn)
Research Paper Link
https://doi.org/10.1016/j.seppur.2024.130933
30 Dec 2024