ExtMatModelML

Extending length and time scales of first principles material modelling via machine learning

Density functional theory (DFT) is an indispensable tool in materials research, but simulations of large systems and long-time dynamics can be prohibitively expensive. This project uses machine learning to construct potential energy surfaces from precomputed DFT energies and forces, making simulations with near-DFT accuracy accessible at a fraction of the cost.

The project focuses on molecular crystals, including temperature-driven effects and crystal screening, and on the dynamics of molecules on surfaces relevant to heterogeneous catalysis. It also aims to strengthen materials research in Croatia through collaboration and the education of young researchers.

Project details

  • Project code: UIP-2020-02-5675
  • Duration: 15 January 2021 – 14 January 2026
  • Principal investigator: Ivor Lončarić
  • Host institution: Ruđer Bošković Institute
  • Funder: Croatian Science Foundation, Installation Research Projects programme, 2020-02 call
  • Funding: €243,413.63
  • Keywords: density functional theory, molecular dynamics, machine learning, molecular crystals, catalysis
  • CroRIS: Project record

Project team

Research infrastructure

Hrvatski prijevod

Povećanje prostorne i vremenske skale modeliranja materijala iz prvih principa pomoću strojnog učenja

Projekt koristi strojno učenje ploha potencijalne energije kako bi povezao preciznost teorije funkcionala gustoće s modeliranjem većih sustava i dugotrajne dinamike. Istražuju se molekulski kristali, temperaturno inducirane pojave i dinamika molekula na površinama relevantna za heterogenu katalizu.

Related publications

1
Alfredo Serrano Jiménez, Alberto P Sanchez Muzas, Yaolong Zhang, Juraj Ovčar, Bin Jiang, Ivor Lončarić, J Iñaki Juaristi, Maite Alducin (2021). Photoinduced desorption dynamics of CO from Pd (111): a neural network approach. Journal of chemical theory and computation.
2
AS Muzas, Alfredo Serrano Jiménez, Juraj Ovčar, Ivor Lončarić, Maite Alducin, J Iñaki Juaristi (2022). Absence of isotope effects in the photo-induced desorption of CO from saturated Pd (111) at high laser fluence. Chemical Physics.
3
Ivor Lončarić, Maite Alducin, J Iñaki Juaristi (2022). O2 on Ag (110): A puzzle for exchange-correlation functionals. Chemical Physics.
4
Jasna Alić, Ivor Lončarić, Martin Etter, Mirta Rubčić, Zoran Štefanić, Marina Šekutor, Krunoslav Užarević, Tomislav Stolar (2024). Direct in situ measurement of polymorphic transition temperatures under thermo-mechanochemical conditions. Physical Chemistry Chemical Physics.
5
Lily M Hunnisett, Nicholas Francia, Jonas Nyman, Nathan S Abraham, Srinivasulu Aitipamula, Tamador Alkhidir, Mubarak Almehairbi, Andrea Anelli, Dylan M Anstine, John E Anthony, Others (2024). The seventh blind test of crystal structure prediction: structure ranking methods. Structural Science.
6
Bruno Mladineo, Ivor Lončarić (2024). Thermosalient phase transitions from machine learning interatomic potential. Crystal Growth & Design.
7
Bernhard Kretz, Ivor Lončarić (2025). Tuning Electronic Properties of Nanoporous Graphene. Inorganic Chemistry.
8
Luka Benić, Federico Grasselli, Chiheb Ben Mahmoud, Dino Novko, Ivor Lončarić (2025). Machine learning model for efficient nonthermal tuning of the charge density wave in monolayer NbSe2. Journal of Chemical Theory and Computation.
9
Shreya Sinha, Bruno Mladineo, Ivor Lončarić, Peter Saalfrank (2024). Multidimensional Neural Network Interatomic Potentials for CO on NaCl (100). The Journal of Physical Chemistry C.
10
Anastasiia Kholtobina, Ivor Lončarić (2025). Exploring elastic properties of molecular crystals with universal machine learning interatomic potentials. Materials & Design.
11
Bruno Mladineo, Teodoro Klaser, Martin Ende, Jasminka Popović, Ivor Lončarić, Željko Skoko (2025). Illustrating Extreme Negative Linear Compressibility in Thermosalient Molecular Crystals. Crystal Growth & Design.
12
Nina Girotto Erhardt, Aloı̈s Castellano, JP Alvarinhas Batista, Raffaello Bianco, Ivor Lončarić, Matthieu J Verstraete, Dino Novko (2025). Electron-mediated anharmonicity and its role in the Raman spectrum of graphene. npj Computational Materials.
13
Bruno Mladineo, J Iñaki Juaristi, Maite Alducin, Peter Saalfrank, Ivor Lončarić (2025). Photoinduced dynamics of CO on Ru (0001): Understanding experiments by simulations with all degrees of freedom. The Journal of Chemical Physics.
14
William P. Wood, Mihails Arhangelskis, Erika Bartůňková, Carlos E. S. Bernardes, A. Daniel Boese, Doris E. Braun, Dejan-Krešimir Bučar, Helena Butkiewicz, Ctirad Červinka, Bartolomeo Civalleri, Nicolas Couvrat, Erik De Ronde, Lorenzo Donà, Martin Dračínský, Dzmitry Firaha, Michal Fulem, Reynaldo Geronia II, Natalia Goncharova, Marlena Gryl, Johannes Hoja, Anna Hoser, Joanna Krzeszczakowska, Alexander List, Ivor Lončarić, Bruno Mladineo, Jonas Nyman, Edgar Olehovics, Mattia Raimondo, Ivo B. Rietveld, Rute I. S. Rodrigues, Luca Russo, Matteo Salvalaglio, Mafalda Sarraguça, Jiří Šnajdr, Vojtěch Štejfa, Guangxu Sun, Paul Tinnemans, Pamela S. Whitfield, Zhuocen Yang, Yizu Zhang, Sarah L. Price (2026). BEST-CSP Benchmark Study of Polymorphs I and II of Sulfamerazine and the Perils of Polytype Polymorphs. Crystal Growth & Design, 26, 476–493.
Lončarić, I. (2021). Modelling dynamics of molecules on surfaces. EUROMAT 2021.
Lončarić, I. (2021). Modelling Dynamics of CO at Metal Surfaces.
Lončarić, I. (2021). Modelling dynamics of CO at Metal Surfaces. Mini-symposium on Surface Science and Dynamics, Potsdam.
Lončarić, I. (2022). Modelling dynamics of CO on metal surfaces. Energy Dissipation at Interfaces: From Catalysis to Astrochemistry, Leiden.
Lončarić, I. (2022). Modeling ultrafast photoinduced dynamics of adsorbates on metals. CECAM Flagship Workshop, p. 9.
Šopar, Š. (2023). Machine learning of potentials for CsPbBr_xI_(3-x). Thesis, University of Zagreb.
Jurdana, J. (2023). Active learning of interatomic potentials for crystal structure prediction. Thesis, University of Zagreb.