Postdoctoral Researcher in Machine Learning and Computational Chemistry

The University of Luxembourg is an international research university with a distinctly multilingual and interdisciplinary character. The University was founded in 2003 and counts more than 6,700 students and more than 2,000 employees from around the world. The University’s faculties and interdisciplinary centres focus on research in the areas of Computer Science and ICT Security, Materials Science, European and International Law, Finance and Financial Innovation, Education, Contemporary and Digital History. In addition, the University focuses on cross-disciplinary research in the areas of Data Modelling and Simulation as well as Health and System Biomedicine. Times Higher Education ranks the University of Luxembourg #3 worldwide for its “international outlook,” #20 in the Young University Ranking 2021 and among the top 250 universities worldwide.

The Faculty of Science, Technology and Medicine (FSTM) contributes multidisciplinary expertise in the fields of Mathematics, Physics, Engineering, Computer Science, Life Sciences and Medicine. Through its dual mission of teaching and research, the FSTM seeks to generate and disseminate knowledge and train new generations of responsible citizens, in order to better understand, explain and advance society and environment we live in.

Project

The project “PHysics- and data-driven multiscale modeling design of layered lead halide perovskiTe mAterials for Stable phoTovoltaICs” (PHANTASTIC) joins the efforts and competence of five scientific groups from the Université de Mons (Belgium), INSA of Rennes (France), Technion-Israel Institute of Technology (Israel), Technical University of Dresden (Germany), and the University of Luxemburg (Luxemburg). The global aim is to improve the long-term stability of 3D lead halide perovskites, the main limiting factor for the broad application of the material in industry.

The role of the TCP group, led by Prof. Alexandre Tkatchenko from the University of Luxemburg, is to develop reliable, efficient, and accurate machine learning force fields (MLFF) for molecular dynamics simulation of systems sizes ranging from ~1000 to 100.000 atoms while preserving the predictive power of first-principles modeling. Moreover, such simulations should correctly reproduce the formation and propagation of defects. This is a must-have requirement for any MLFF used to improve the stability of any material in general and 3D lead halide perovskites in particular. The novel equivariant MLFFs will be used with state-of-the-art training and sampling approaches, efficient DFT calculations, and recent developments in vdW methods to achieve this goal. These push the project to the frontiers of modern MLFF capabilities, making it challenging, engaging, and impactful.