Dresden 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 35: Organic molecules on inorganic substrates: Adsorption and growth II
O 35.6: Vortrag
Dienstag, 10. März 2026, 11:45–12:00, HSZ/0401
MAD-SURF: a general Machine-Learning Potential for molecular adsorption on metal surfaces — •Manuel González Lastre1, Joakim S. Jestilä2, Rubén Pérez1, and Adam S. Foster2 — 1Departamento de Física Teórica de la Materia Condensada, Universidad Autónoma de Madrid, E-28049 Spain — 2Department of Applied Physics, Aalto University, 00076, Espoo, Finland.
Predicting how organic molecules adsorb, assemble and interact on metal surfaces is central to surface chemistry, molecular electronics and scanning probe microscopy. Yet, the application of first-principles simulations to interfaces is hampered by the computational cost for evaluating the electronic structure for the large number of atoms typically involved. Here we present MAD-SURF, a general Machine-Learning Interatomic Potential tailored for molecular adsorption on metal surfaces. Trained on a broad dataset spanning diverse molecules, adsorption motifs, surfaces, molecular dynamics and non-covalent aggregates, MAD-SURF achieves near quantum-mechanical accuracy while enabling simulations that are orders of magnitude faster than Density Functional Theory. The model reliably reproduces energies, forces and adsorption geometries across different surfaces. We demonstrate its versatility by capturing experimentally observed systems, ranging from organic monolayers and polycyclic aggregates to flexible biomolecules and long-range metal surface reconstructions. By combining accuracy, speed, and generalizability, MAD-SURF offers a practical framework for accelerating atomistic simulations in surface science.
Keywords: machine-learning interatomic potentials; scanning probe microscopy; Density Functional Theory; atomistic modeling; surface chemistry