Parts | Days | Selection | Search | Updates | Downloads | Help
DS: Fachverband Dünne Schichten
DS 14: Poster
DS 14.53: Poster
Tuesday, April 2, 2019, 17:00–20:00, Poster E
Finding and understanding surface structures with SAMPLE — •Lukas Hörmann, Andreas Jeindl, Alexander T. Egger, and Oliver T. Hofmann — Institute of Solid State Physics, NAWI Graz, Graz University of Technology, Austria
Even if the physical properties of an organic semiconductor are ever so promising in the bulk phase, they may drastically change upon adsorption on the surface. Specifically, surfaces can induce the formation of polymorphs with worse, or under the right conditions, also greatly improved properties. Normally, the exponential growth of possible polymorphs with system size prohibits rigorous computational studies, that could explore the full configurational and thermodynamic search space. Thus, we use SAMPLE [1,2], which employs machine learning to suitably fit a physical energy model and therewith efficiently calculate the adsorption energies of an exhaustive set of coarse grained polymorphs.
We showcase the capabilities of this approach for monolayers of molecules with very different interactions on coinage metals. With SAMPLE we not only find the best polymorphs, but also defects and other local minima. Ab-initio thermodynamics allows us to also consider temperature effects and create phase diagrams. Our unique combination of a physically inspired energy model and statistical learning enables us to gain insight into the molecular interactions on the surface. This allows us to not only tell which polymorph forms, but also which interactions are the reasons for the formation of specific structures.
[1] Hörmann et al., arXiv:1811.11702
[2] Scherbela et al., Phys. Rev. Materials 2, 043803