Our current research focuses on development and application of machine learning methods to model interatomic interactions, including the development of many-body atomic descriptors. Topics include carbon-based nanostructures and nanomaterials, metal nanoparticles, catalysis and X-ray spectroscopy.

This page is currently under construction. You can visit our Publication List for an up-to-date account of our published work. You can also browse the list below to find out more about different topics of our past research (we are hoping to update this list as soon as time allows).

Research topics

Experiment-driven computational simulation of materialsMay 22, 2024This is a synopsis of the paper by T. Zarrouk, R. Ibragimova, A.P. Bartók and M.A. Caro, “Experiment-Driven Atomistic Materials Modeling: A Case Study Combining X-Ray Photoelectron Spectroscopy and Machine Learning Potentials to Infer the Structure of Oxygen-Rich Amorphous Carbon”, J. Am. Chem. Soc., DOI:10.1021/jacs.4c01897, available as an Open Access PDF from the publisher’s website. … [...] Read more...

Representing databases of materials and molecules in two dimensionsMay 17, 2024The research highlighted in this post is part of the following paper: “Cluster-based multidimensional scaling embedding tool for data visualization”, by P. Hernández-León and M.A. Caro, Phys. Scr. 99, 066004 (2024). Available Open Access . The figures in this article are reproduced from that publication under the CC-BY 4.0 license. One of the consequences of the emergence … [...] Read more...

Looking for stable iron nanoparticlesJune 26, 2023This research is published in R. Jana and M.A. Caro, “Searching for iron nanoparticles with a general-purpose Gaussian approximation potential”, Phys. Rev. B 107, 245421 (2023) [also available from the arXiv]. Reprinted figures are copyright (c) 2023 of the American Physical Society. The video is copyright (c) 2023 of M.A. Caro. In the field of … [...] Read more...

Understanding the structure of amorphous carbon and silicon with machine learning atomistic modelingMarch 6, 2023When we think about semiconductors the first ones to come to mind are silicon, germanium and the III-Vs (GaAs, InP, AlN, GaN, etc.), in their crystalline forms. In fact, the degree of crystallinity in these materials often dictates the quality of the devices that can be made with them. As an example, dislocation densities as … [...] Read more...

Automated X-ray photoelectron spectroscopy (XPS) prediction for carbon-based materials: combining DFT, GW and machine learningJuly 13, 2022The details of this work are now published (open access ) in Chem. Mater. and our automated prediction tool is available from nanocarbon.fi/xps. Many popular experimental methods for determining the structure of materials rely on the periodic repetition of atomic arrangements present in crystals. A common example is X-ray diffraction. For amorphous materials, the lack … [...] Read more...

Structure and properties of nanoporous carbonsJanuary 4, 2022Nanoporous carbons are an emerging class of materials with important applications in energy storage. In particular, the ability of graphitic carbons to intercalate ions is exploited in commercial Li-ion batteries where the anode is (typically) made of graphite and Li ions become electrostatically bound to the carbon host as the battery is charged: electrons are … [...] Read more...

Adding Van der Waals corrections to GAP force fieldsAugust 7, 2021This post is about the following paper (and the figures are reproduced from it and are copyright of the American Physical Society): “Machine learning force fields based on local parametrization of dispersion interactions: Application to the phase diagram of C60“, by Heikki Muhli, Xi Chen, Albert P. Bartók, Patricia Hernández-León, Gábor Csányi, Tapio Ala-Nissila, and … [...] Read more...

SUMO HPC fat node for GAP trainingFebruary 10, 2021We are very excited to welcome our new dedicated state-of-the-art “fat node” (jargon for “high memory server”) SUMO, which we will use primarily to train machine learning GAP potentials. These potentials use lots of structural and energy/forces atomic data, and thus require large amounts of RAM. Our SUMO-I machine (we’re being optimistic that there will … [...] Read more...

Structure and growth mechanism of amorphous carbonNovember 3, 2020Our latest work (of a long list) on amorphous carbon simulation, titled “Machine learning driven simulated deposition of carbon films: From low-density to diamondlike amorphous carbon” just appeared in Physical Review B (https://doi.org/10.1103/PhysRevB.102.174201, check also on arXiv if you don’t have an APS subscription). This paper is the product of a lot of work, spanning … [...] Read more...

Computational discovery of new piezoelectric materialsFebruary 7, 2020The availability of new piezoelectric materials compatible with silicon chip integration for micro-electromechanical systems (MEMS) application is a highly attractive prospect. These new materials will help to bridge the gap between mechanical and electronic devices, making MEMS increasingly small and efficient. AlN is today’s industry’s standard and research is intensifying worldwide on AlN derivatives such as ScAlN. By alloying AlN with Sc, the crystal lattice is locally distorted due to the phase competition between the rock-salt ScN and wurtzite AlN structures, resulting in a progressive transition of AlN from wurtzite into a hexagonal-layered structure as the amount of Sc dopant atoms increases. This, in turn, induces an enhancement of the piezoelectric coefficients of ScAlN up to 50% Sc content (see figure). [...] Read more...