File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1021/jacs.3c07766
- Scopus: eid_2-s2.0-85176495017
- Find via
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Article: Encoding Hole–Particle Information in the Multi-Channel MolOrbImage for Machine-Learned Excited-State Energies of Large Photofunctional Materials
Title | Encoding Hole–Particle Information in the Multi-Channel MolOrbImage for Machine-Learned Excited-State Energies of Large Photofunctional Materials |
---|---|
Authors | |
Issue Date | 24-Oct-2023 |
Publisher | American Chemical Society |
Citation | Journal of the American Chemical Society, 2023, v. 145, n. 44, p. 24098-24107 How to Cite? |
Abstract | We present a novel class of one-electron multi-channel molecular orbital images (MolOrbImages) designed for the prediction of excited-state energetics in conjunction with the state-of-the-art VGG-type machine-learning architecture. By representing hole and particle states in the excitation process as channels of MolOrbImages, the revised VGG model achieves excellent prediction accuracy for both low-lying singlet and triplet states, with mean absolute errors (MAEs) of <0.08 and <0.1 eV for QM9 molecules and large photofunctional materials with up to 560 atoms, respectively. Remarkably, the model demonstrates exceptional performance (MAE < 1 kcal/mol) for the T1 state of QM9 molecules, making it a non-system-specific model that approaches chemical accuracy. The general rules attained, for instance, the improved performance with well-defined MO energies and the reduced overfitting concern via the inclusion of physically insightful hole–particle information, provide invaluable guidelines for the further design of orbital-based descriptors targeting molecular excited states. |
Persistent Identifier | http://hdl.handle.net/10722/348331 |
ISSN | 2023 Impact Factor: 14.4 2023 SCImago Journal Rankings: 5.489 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Chen, Ziyong | - |
dc.contributor.author | Yam, Vivian Wing-Wah | - |
dc.date.accessioned | 2024-10-08T00:31:41Z | - |
dc.date.available | 2024-10-08T00:31:41Z | - |
dc.date.issued | 2023-10-24 | - |
dc.identifier.citation | Journal of the American Chemical Society, 2023, v. 145, n. 44, p. 24098-24107 | - |
dc.identifier.issn | 0002-7863 | - |
dc.identifier.uri | http://hdl.handle.net/10722/348331 | - |
dc.description.abstract | <p>We present a novel class of one-electron multi-channel molecular orbital images (MolOrbImages) designed for the prediction of excited-state energetics in conjunction with the state-of-the-art VGG-type machine-learning architecture. By representing hole and particle states in the excitation process as channels of MolOrbImages, the revised VGG model achieves excellent prediction accuracy for both low-lying singlet and triplet states, with mean absolute errors (MAEs) of <0.08 and <0.1 eV for QM9 molecules and large photofunctional materials with up to 560 atoms, respectively. Remarkably, the model demonstrates exceptional performance (MAE < 1 kcal/mol) for the T<sub>1</sub> state of QM9 molecules, making it a non-system-specific model that approaches chemical accuracy. The general rules attained, for instance, the improved performance with well-defined MO energies and the reduced overfitting concern via the inclusion of physically insightful hole–particle information, provide invaluable guidelines for the further design of orbital-based descriptors targeting molecular excited states.<br></p> | - |
dc.language | eng | - |
dc.publisher | American Chemical Society | - |
dc.relation.ispartof | Journal of the American Chemical Society | - |
dc.title | Encoding Hole–Particle Information in the Multi-Channel MolOrbImage for Machine-Learned Excited-State Energies of Large Photofunctional Materials | - |
dc.type | Article | - |
dc.identifier.doi | 10.1021/jacs.3c07766 | - |
dc.identifier.scopus | eid_2-s2.0-85176495017 | - |
dc.identifier.volume | 145 | - |
dc.identifier.issue | 44 | - |
dc.identifier.spage | 24098 | - |
dc.identifier.epage | 24107 | - |
dc.identifier.eissn | 1520-5126 | - |
dc.identifier.issnl | 0002-7863 | - |