2020
DOI: 10.1038/s41524-020-00388-2
|View full text |Cite
|
Sign up to set email alerts
|

Machine learning for accelerating the discovery of high-performance donor/acceptor pairs in non-fullerene organic solar cells

Abstract: Integrating artificial intelligence (AI) and computer science together with current approaches in material synthesis and optimization will act as an effective approach for speeding up the discovery of high-performance photoactive materials in organic solar cells (OSCs). Yet, like model selection in statistics, the choice of appropriate machine learning (ML) algorithms plays a vital role in the process of new material discovery in databases. In this study, we constructed five common algorithms, and introduced 5… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

1
103
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
8

Relationship

0
8

Authors

Journals

citations
Cited by 110 publications
(104 citation statements)
references
References 52 publications
1
103
0
Order By: Relevance
“…However, this prediction was based on a limited dataset (19 training data and 6 test data), which is too small to deliver convincing statistics ( Fathollahi and Sajady, 2018 ). Similarly poor performance was also observed in the prediction of experimental power conversion efficiency of organic solar cell devices, where RF was identified to be the best performing model, yet with a low r value of 0.70 ( Wu et al., 2020 ).…”
Section: Resultsmentioning
confidence: 80%
“…However, this prediction was based on a limited dataset (19 training data and 6 test data), which is too small to deliver convincing statistics ( Fathollahi and Sajady, 2018 ). Similarly poor performance was also observed in the prediction of experimental power conversion efficiency of organic solar cell devices, where RF was identified to be the best performing model, yet with a low r value of 0.70 ( Wu et al., 2020 ).…”
Section: Resultsmentioning
confidence: 80%
“…119 (b) Workflow employed by Wu et al for building, applying and evaluating ML methods to identify and synthesize highperforming molecular candidates for OPV applications. 131…”
Section: View Article Onlinementioning
confidence: 99%
“…The resulting database was then exploited by artificial neural networks and random forest methods to find potentially high-performing choices for polymer:fullerene OPV devices 119. (b) Workflow employed by Wu et al for building, applying and evaluating ML methods to identify and synthesize highperforming molecular candidates for OPV applications 131. Images reprinted (adapted) with permission from ref 119.…”
mentioning
confidence: 99%
See 1 more Smart Citation
“…DL methods have also been exploited in predicting ground- and excited state properties for thousands of organic molecules, where the accuracy for small molecules can be even superior to QM ab initio methods [ 78 ]. Recent advances in the use of machine learning and computational chemistry methods to study organic photovoltaics are discussed in other works [ 80 , 81 , 82 ].…”
Section: Materials Discoverymentioning
confidence: 99%