Luca Schaufelberger¹, J. Terence Blaskovits¹, Ruben Laplaza^1,2, Clemence Corminboeuf^1,2*, Kjell Jorner^3*

1 Ecole polytechnique fédérale de Lausanne (EPFL), Institute of Chemical Sciences and Engineering, Lausanne, Switzerland, CH-1015

2 National Center for Competence in Research – Catalysis (NCCR-Catalysis), Ecole polytechnique fédérale de Lausanne (EPFL), Lausanne, Switzerland, CH-1015

3 ETH Zürich, Institute of Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 1, Zürich, Switzerland, CH-8093

* Corresponding authors emails: clemence.corminboeuf@epfl.ch, kjell.jorner@chem.ethz.ch

DOI10.24435/materialscloud:yn-vz [version v1]

Publication date: Jul 04, 2024

How to cite this record

Luca Schaufelberger, J. Terence Blaskovits, Ruben Laplaza, Clemence Corminboeuf, Kjell Jorner, Inverse design of singlet fission materials with uncertainty-controlled genetic optimization, Materials Cloud Archive 2024.104 (2024), https://doi.org/10.24435/materialscloud:yn-vz

Description

Singlet fission has shown potential for boosting the power conversion efficiency of solar cells, but the scarcity of suitable molecular materials hinders its implementation. We introduce an uncertainty-controlled genetic algorithm (ucGA) based on ensemble machine learning predictions from different molecular representations that concurrently optimizes excited state energies, synthesizability, and singlet exciton size for the discovery of singlet fission materials. We show that uncertainty in the model predictions can control how far the genetic optimization moves away from previously known molecules. Running the ucGA in an exploitative setup performs local optimization on variations of known singlet fission scaffolds, such as acenes. In an explorative mode, hitherto unknown candidates displaying excellent excited state properties for singlet fission are generated. We suggest a class of heteroatom-rich mesoionic compounds as acceptors for charge-transfer mediated singlet fission. When included in larger conjugated donor-acceptor systems, these units exhibit strong localization of the triplet state, favorable diradicaloid character and suitable triplet energies for exciton injection into semiconductor solar cells. As the proposed candidates are composed of fragments from synthesized molecules, they are likely synthetically accessible.

Materials Cloud sections using this data

Files

File name	Size	Description
1_Fragment_Pool.zip MD5md5:fcd6eb4ef724671660b2af83596e5150	1.9 MiB	Folder containing reFORMED (cores and substituents) as well as the uncurated fragments.
2_External_Test_Set.zip MD5md5:956a72f6b44a2698cd86b4cb166cce11	417.0 MiB	Folder containing the TD-DFT calculations of the external test set.
3_Singlet_Fission_Candidates_pruned_adiabatic.zip MD5md5:95c09f3eebb57c492553296b30f50bbb	9.9 MiB	Folder containing the TD-DFT calculations of the (pruned) top candidates from the ucGA, as shown in Figure 7.
4_Screening_Based_on_Structure_Property_Relationships.zip MD5md5:031ef970e2c55bbb2395e645bc72dd75	7.0 MiB	Folder containing the TD-DFT calculations of top candidates from the ML screening, as shown in Figure 8.
5_Diradical.zip MD5md5:211f5e766c4a1ce4bbe8e1c9f4402394	5.0 KiB	Folder containing the xyz files for the diradical analysis.
Data_FORMED.csv MD5md5:749166b2fb38215deeed85b6d2130949	115.5 MiB	CSV file containing the tabulated properties for the FORMED database, including SMILES.
READ_ME.txt MD5md5:76264694c3905b28e8f3e1085e8a3abc	695 Bytes	READ_ME

License

Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Keywords

singlet fission machine learning uncertainty quantification inverse design genetic algorithm