Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 46: New Methods and Developments 3: Theory
O 46.8: Vortrag
Mittwoch, 7. September 2022, 16:45–17:00, H6
Designing Covalent Organic Frameworks Through Active Machine Learning — •Yuxuan Yao1,2, Christian Kunkel3, Karsten Reuter3, and Harald Oberhofer2 — 1Chair for Theoretical Chemistry and Catalysis Research Center, Technical University Munich — 2Fritz-Haber-Institut der Max-Planck-Gesellschaft — 3Chair for Theoretical Physics VII, University of Bayreuth
Covalent organic frameworks(COFs) are a class of materials with potential applications in many fields such as catalysis, sensing, or optoelectronics. It is well known that their design space is far too large to sample one by one. Focusing on their electronic properties, we modify an earlier active machine learning(AML) approach that explores the molecular design through the use of surrogate models for charge injection and transport descriptors. In this method, the Gaussian Process Regression(GPR) and AML are combined to train the molecular space. This way we ensure that only promising molecules or candidates that are very different from already explored ones have their descriptors evaluated on a comparatively expensive quantum mechanical level. Specifically, we modify molecular generation rules in order to produce three-fold rotational symmetric candidates molecules for use in COFs. In the future this approach can be generalized for any other symmetries, to potentially even allow for 3-dimensional network generation. The generation of a candidate space with well defined symmetries together with AML ensures a high efficiency in the detection of promising COFs with superior charge conduction properties and demonstrates the utility of this approach.