Exploring Molecules with Low Viscosity: Using Physics-Based Simulations and De Novo Design by Applying Reinforcement LearningClick to copy article linkArticle link copied!
- Nobuyuki N. Matsuzawa*Nobuyuki N. Matsuzawa*Email: [email protected]Engineering Division, Panasonic Industry Co., Ltd., 1006 Kadoma, Kadoma City, Osaka 571-8506, JapanMore by Nobuyuki N. Matsuzawa
- Hiroyuki MaeshimaHiroyuki MaeshimaEngineering Division, Panasonic Industry Co., Ltd., 1006 Kadoma, Kadoma City, Osaka 571-8506, JapanMore by Hiroyuki Maeshima
- Keisuke HayashiKeisuke HayashiEngineering Division, Panasonic Industry Co., Ltd., 1006 Kadoma, Kadoma City, Osaka 571-8506, JapanMore by Keisuke Hayashi
- Tatsuhito AndoTatsuhito AndoEngineering Division, Panasonic Industry Co., Ltd., 1006 Kadoma, Kadoma City, Osaka 571-8506, JapanMore by Tatsuhito Ando
- Mohammad Atif Faiz Afzal*Mohammad Atif Faiz Afzal*Email: [email protected]Schrödinger Inc., 101 SW Main Street, Suite 1300, Portland, Oregon 97204, United StatesMore by Mohammad Atif Faiz Afzal
- Kyle MarshallKyle MarshallSchrödinger Inc., 101 SW Main Street, Suite 1300, Portland, Oregon 97204, United StatesMore by Kyle Marshall
- Benjamin J. CosciaBenjamin J. CosciaSchrödinger Inc., 101 SW Main Street, Suite 1300, Portland, Oregon 97204, United StatesMore by Benjamin J. Coscia
- Andrea R. BrowningAndrea R. BrowningSchrödinger Inc., 101 SW Main Street, Suite 1300, Portland, Oregon 97204, United StatesMore by Andrea R. Browning
- Alexander GoldbergAlexander GoldbergSchrödinger Inc., 9868 Scranton Road, Suite 3200, San Diego, California 92121, United StatesMore by Alexander Goldberg
- Mathew D. HallsMathew D. HallsSchrödinger Inc., 9868 Scranton Road, Suite 3200, San Diego, California 92121, United StatesMore by Mathew D. Halls
- Karl LeswingKarl LeswingSchrödinger Inc., 120 West 45th Street, 17th Floor, New York, New York 10036, United StatesMore by Karl Leswing
- Mayank MisraMayank MisraSchrödinger Inc., 120 West 45th Street, 17th Floor, New York, New York 10036, United StatesMore by Mayank Misra
- Farhad RamezanghorbaniFarhad RamezanghorbaniSchrödinger Inc., 120 West 45th Street, 17th Floor, New York, New York 10036, United StatesMore by Farhad Ramezanghorbani
- Tsuguo MorisatoTsuguo MorisatoSchrödinger K.K., 13F Marunouchi Trust Tower North,1-8-1 Marunouchi Chiyoda-ku, Tokyo 100-0005, JapanMore by Tsuguo Morisato
Abstract
Molecules with viscosities lower than those of conventional organic solvents are highly sought after for applications in electrochemical devices such as batteries and capacitors. These molecules improve the electrical resistance of devices, enhancing their efficiency, especially at low temperatures. To identify new molecules with low viscosities, we conducted extensive molecular dynamics (MD) simulations on 10,000 molecules selected from the GDB-17 chemical structure database, specifically choosing molecules with fewer than 12 heavy atoms. Additionally, we performed density functional theory (DFT) calculations to determine the energies of the highest occupied molecular orbitals (HOMO) of these molecules as a surrogate for the oxidation potential. We used the data on viscosity and HOMO levels as training sets to develop machine-learning models that predict these properties. Using these models, we carried out molecular de novo design using the REINVENT method, a reinforcement-learning approach utilizing SMILES strings. This method aimed to identify molecules that minimize viscosity while maintaining sufficiently low HOMO levels for stability. The approach successfully identified new chemical structures with viscosities below 2 mPa·s and suitably low HOMO energies. We synthesized a novel compound from the top candidates and validated our predictions experimentally. The experimental results closely matched our predictions, demonstrating that combining physics-based simulations with reinforcement learning is an effective strategy for designing novel molecules with targeted properties.
Cited By
This article has not yet been cited by other publications.
Article Views
Altmetric
Citations
Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.
Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.
The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.