1 README.md Machine learning interatomic potentials for sulfur vacancy dynamics in single sheet MoS2 Flötotto, Aaron aut TU Ilmenau Technische Universität Ilmenau Flötotto Aaron 0009-0000-8230-5008 Spetzler, Benjamin aut CAU Kiel Christian-Albrechts-Universität zu Kiel Spetzler Benjamin 0000-0003-4255-5947 von Stackelberg, Rose aut TU Ilmenau Technische Universität Ilmenau von Stackelberg Rose 0009-0009-4120-5970 Ziegler, Martin aut CAU Kiel Christian-Albrechts-Universität zu Kiel Ziegler Martin 1019038101 Runge, Erich aut TU Ilmenau Technische Universität Ilmenau Runge Erich 0000-0002-2157-229X Dreßler, Christian aut TU Ilmenau Technische Universität Ilmenau Dreßler Christian 0000-0002-0145-1263 2026-01-07 2026-01-05 2025-12-15 10.48550/arXiv.2508.13790 Large-scale cooperative sulfur vacancy dynamics in two-dimensional MoS2 from machine learning interatomic potentials https://doi.org/10.48550/arXiv.2508.13790 en 36 GB This dataset contains two types of machine learning interatomic potentials (MLIPs) that have been trained to accurately describe the dynamics of sulfur vacancies in single sheet 2H-molybdenum disulfide (MoS2): (i) A graph neural network (GNN) that was fine-tuned for MoS2 starting with the MACE MP-0 foundation model using the MACE python package and (ii) a gaussian approximation potential that was trained on defective MoS2 structures using the Vienna ab-initio Simulation Package (VASP, version 6.4.0, https://vasp.at/). Both MLIPs are able to accurately describe the potential energy curves of the atomic jump that governs the sulfur vacancy dynamics. Benchmarking the MLIPs by running density functional theory calculations on equidistantly selected snapshots of MLIP MD trajectories yields force test errors of 42 meV/A and 74 meV/A for the GNN and GAP potentials, respectively. The test and training data set for the MACE model was obtained using the CP2K software package (version 2022.2, https://www.cp2k.org/). In addition to the trained MLIP models, this dataset also contains the corresponding training data sets. molecular dynamics defects in semiconductor transition-metal dichalcogenide 530 Carl-Zeiss-Stiftung Ilmenau School of Green Electronics P2022-00-135 https://ror.org/03ng4kg22 10.71758/refodat.68 2025-12-15T14:44:02.908Z 2026-01-07T14:20:59.413Z 2026-01-07T14:14:00.923Z f0a06eb220e6451195431a11ff00e234@tu-ilmenau.de {"identifier":"10.71758/refodat.68","type":"doi","additional":"","service":"Datacite","created":"2026-01-07T14:16:09.810Z","registered":"2026-01-07T14:16:11.779Z"} 8df7f559456b4421a7e488034161dba0@uni-jena.de