Laboratory of Computational Science and Modeling COSMO
The macroscopic behaviour of materials is determined by the chemical and physical interactions between atoms, averaged over thermal and quantum mechanical fluctuations. At the laboratory of Computational Science and Modelling we simulate matter at the atomic scale, to be able to predict and optimise the properties of materials and molecules starting from fundamental physical principles.
To do so, we need to develop computer simulation methods that are both efficient and accurate, relying on an interdisciplinary expertise that combines understanding of statistical physics, quantum chemistry and materials engineering, with the mathematical and computational tools of data science and machine learning. We then apply these techniques to address materials science problems that are relevant for pressing societal issues – from the metallurgy of lightweight alloys for sustainable transportation, to the determination of the structure and behaviour of hydrogen-bonded materials such as pharmaceutical compounds and bio-inspired polymers.
You can learn more on what we do, and what methods we use, on the Research page, or on the News Archive. If you are interested in joining us, have a look at the Jobs page for open positions and application instructions.
|Head of the laboratory:|
Prof. Michele Ceriotti
tel: +41 (0)21 69 32939
Ms. Isabel Nzazi
tel: +41 (0)21 69 32925
|EPFL STI IMX COSMO|
MXG 338 (Bâtiment MXG)
Seminar on machine-learning interatomic potentials by Miguel Caro
Dr. Miguel Caro (Aalto University) will give a virtual talk on 'Simulating carbon materials with machine learning interatomic potentials', on Thursday 3 June at 10:30. Zoom details will be communicated separately.
Dr. Timothy Moore gives a seminar on Coarse-Graining
Dr. Timothy Moore, from the Glotzer group at the University of Michigan, will give a seminar on "Coarse-Graining for Biomolecular Self-Assembly" on Friday, March 26, at 15:00. Zoom details will be circulated by e-mail.
Structure and properties of amorphous silicon from machine learning
A combination of machine learning models predicting the stability as well as the electronic properties of materials with the accuracy of quantum mechanics allowed to understand the structural transitions that silicon undergoes when compressed to tens of GPa. The study, performed by an international team from Oxford, Cambridge, the US Naval Research Laboratory and Ohio University, as well as EPFL's Laboratory of Computational Science and Modeling, has been published in Nature.