Lecture
Van Marum Colloquium: Active surface adsorbate structure and complex materials exploration with Bayesian optimization
- Date
- Friday 24 March 2023
- Time
- Location
-
Gorlaeus Building
Einsteinweg 55
2333 CC Leiden - Room
- DM1.15
Abstract
Controlling the properties of organic/inorganic materials requires detailed knowledge of their molecular adsorption geometries. This is often unattainable, even with current state-of-the-art tools. Visualizing the structure of complex non-planar adsorbates with atomic force microscopy (AFM), for example, is challenging, and identifying it computationally is intractable with conventional structure search. To facilitate such complex search and characterization tasks, we have developed the Bayesian Optimization Structure Search (BOSS) [1] code [2]. BOSS uses active learning to strategically sample the parameter space of material-science tasks (experimental or computational). BOSS proposes new data acquisition points for maximum knowledge gain, balancing exploitation with exploration.
We couple BOSS to density-functional theory (DFT) calculations to study the adsorption of a camphor molecule on a copper surface. We identify 8 unique stable adsorbates [3]. By matching the stable structures to atomic force microscopy (AFM) images, we conclude that the experiments feature 3 different structures of chemisorbed camphor molecules. This is the first time that the atomic structure of bulky 3D adsorbates has been decoded [4]. Moving towards morphology analysis and prediction of organic charge transfer complexes (CTCs) on graphene (Gr), I will first show how we approximate an intercalated Gr/O/Ir(111) substrate by charged freestanding Gr without altering the properties of adsorbed molecules. Then I present preliminary results for BOSS structure searches of donor-acceptor pairs on intercalated Gr [5]. I will conclude the presentation of BOSS’ functionality with an experimental materials science example. We use BOSS to guide the extraction of lignin from birchwood and optimize the lignin properties for use in high value sustainable composite materials (e.g., carbon fibres, thermoplastics and three-dimensional printed objects) [6].
References
- Bayesian Inference of Atomistic Structure in Functional Materials, M. Todorović, M. U. Gutmann, J. Corander and P. Rinke, npj Comp. Mat. 5, 35 (2019)
- https://sites.utu.fi/boss/
- Detecting stable adsorbates of (1S)-camphor on Cu(111) with Bayesian optimization, J. Järvi, P. Rinke and M. Todorović, Beilstein J. Nanotechnol. 11, 1577 (2020)
- Integrating Bayesian Inference with Scanning Probe Experiments for Robust Identification of Surface Adsorbate Configurations, J. Järvi, B. Alldritt, O. Krejčí, M. Todorović, P. Liljeroth and P. Rinke, Adv. Funct. Mater. 31, 2010853 (2021)
- Efficient modeling of organic adsorbates on oxygen-intercalated graphene on Ir(111), J. Järvi, M. Todorović, and P. Rinke, Phys. Rev. B 105, 195304 (2022)
- Machine Learning Optimization of Lignin Properties in Green Biorefineries, J. Löfgren, D. Tarasov, T. Koitto, P. Rinke, M. Balakshin, M. Todorović, ACS Sustainable Chem. Eng. 10, 9469 (2022)