Machine learning property prediction for organic photovoltaic devices
Source abstract
Abstract
Abstract Organic photovoltaic (OPV) materials are promising candidates for cheap, printable solar cells. However, there are a very large number of potential donors and acceptors, making selection of the best materials difficult. Here, we show that machine-learning approaches can leverage computationally expensive DFT calculations to estimate important OPV materials properties quickly and accurately. We generate quantitative relationships between simple and interpretable chemical signature and one-hot descriptors and OPV power conversion efficiency (PCE), open circuit potential (Voc), short circuit density (Jsc), highest occupied molecular orbital (HOMO) energy, lowest unoccupied molecular orbital (LUMO) energy, and the HOMO–LUMO gap. The most robust and predictive models could predict PCE (computed by DFT) with a standard error of ±0.5 for percentage PCE for both the training and test set. This model is useful for pre-screening potential donor and acceptor materials for OPV applications, accelerating design of these devices for green energy applications.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Organic Electronics and Photovoltaics · Conducting polymers and applications · Machine Learning in Materials Science
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (65)
Show source-provided references
1. Abdulrazzaq, O. A., Saini, V., Bourdo, S., Dervishi, E. & Biris, A. S. Organic solar cells: a review of materials, limitations, and possibilities for improvement. Part. Sci. Technol. 31, 427–442 (2013).10.1080/02726351.2013.769470 ↗
2. Cui, Y. et al. Over 16% efficiency organic photovoltaic cells enabled by a chlorinated acceptor with increased open-circuit voltages. Nat. Commun. 10, 2515 (2019).10.1038/s41467-019-10351-5 ↗
3. Mosconi, E., Amat, A., Nazeeruddin, M. K., Grätzel, M. & De Angelis, F. First-principles modeling of mixed halide organometal perovskites for photovoltaic applications. J. Phys. Chem. C 117, 13902–13913 (2013).10.1021/jp4048659 ↗
4. Janković, V. & Vukmirović, N. Dynamics of exciton formation and relaxation in photoexcited semiconductors. Phys. Rev. B 92, 235208 (2015).10.1103/PhysRevB.92.235208 ↗
5. Mikhnenko, O. V., Blom, P. W. & Nguyen, T.-Q. Exciton diffusion in organic semiconductors. Energy Environ. Sci. 8, 1867–1888 (2015).10.1039/C5EE00925A ↗
6. Coropceanu, V. et al. Charge transport in organic semiconductors. Chem. Rev. 107, 926–952 (2007).10.1021/cr050140x ↗
7. Proctor, C. M., Kuik, M. & Nguyen, T.-Q. Charge carrier recombination in organic solar cells. Prog. Polym. Sci. 38, 1941–1960 (2013).10.1016/j.progpolymsci.2013.08.008 ↗
8. Ran, N. A. et al. Charge generation and recombination in an organic solar cell with low energetic offsets. Adv. Energy Mater. 8, 1701073 (2018).10.1002/aenm.201701073 ↗
9. Winkler, D. A. & Le, T. C. Performance of deep and shallow neural networks, the universal approximation theorem, activity cliffs, and QSAR. Mol. Inform. 36, 1600118 (2017).10.1002/minf.201600118 ↗
10. Mesta, M., Chang, J. H., Shil, S., Thygesen, K. S. & García-Lastra, J. M. A protocol for fast prediction of electronic and optical properties of donor-acceptor polymers using density functional theory and tight-binding method. J. Phys. Chem. A. 123, 4980–4989 (2019).10.1021/acs.jpca.9b02391 ↗
11. Altman, N. S. An introduction to kernel and nearest-neighbor nonparametric regression. Am. Stat. 46, 175–185 (1992).10.1080/00031305.1992.10475879 ↗
12. Rupp, M. Machine learning for quantum mechanics in a nutshell. Int. J. Quant. Chem. 115, 1058–1073 (2015).10.1002/qua.24954 ↗
13. Vu, K. et al. Understanding kernel ridge regression: common behaviors from simple functions to density functionals. Int. J. Quant. Chem. 115, 1115–1128 (2015).10.1002/qua.24939 ↗
14. Padula, D., Simpson, J. D. & Troisi, A. Combining electronic and structural features in machine learning models to predict organic solar cells properties. Mater. Horiz. 6, 343–349 (2019).10.1039/C8MH01135D ↗
15. Breiman, L. Random forests. Mach. Learn. 45, 5–32 (2001).10.1023/A:1010933404324 ↗
16. Svetnik, V. et al. Random forest: a classification and regression tool for compound classification and QSAR modeling. J. Chem. Inf. Comput. Sci. 43, 1947–1958 (2003).10.1021/ci034160g ↗
17. Guelman, L. Gradient boosting trees for auto insurance loss cost modeling and prediction. Expert Syst. Appl. 39, 3659–3667 (2012).10.1016/j.eswa.2011.09.058 ↗
18. Sahu, H., Rao, W., Troisi, A. & Ma, H. Toward predicting efficiency of organic solar cells via machine learning and improved descriptors. Adv. Energy Mater. 8, 1801032 (2018).10.1002/aenm.201801032 ↗
19. Alexander, D. L., Tropsha, A. & Winkler, D. A. Beware of R 2: simple, unambiguous assessment of the prediction accuracy of QSAR and QSPR models. J. Chem. Inf. Model 55, 1316–1322 (2015).10.1021/acs.jcim.5b00206 ↗
20. Pereira, F. et al. Machine learning methods to predict density functional theory B3LYP energies of HOMO and LUMO orbitals. J. Chem. Inf. Model 57, 11–21 (2016).10.1021/acs.jcim.6b00340 ↗
21. Cortes, C. & Vapnik, V. Support-vector networks. Mach. Learn. 20, 273–297 (1995).
22. Lopez, S. A. et al. The Harvard organic photovoltaic dataset. Sci. Data 3, 160086 (2016).10.1038/sdata.2016.86 ↗
23. Scharber, M. C. et al. Design rules for donors in bulk‐heterojunction solar cells—towards 10% energy‐conversion efficiency. Adv. Mater. 18, 789–794 (2006).10.1002/adma.200501717 ↗
24. lharbi, F. et al. An efficient descriptor model for designing materials for solar cells. npj Comput. Mater. 1, 15003 (2015).10.1038/npjcompumats.2015.3 ↗
25. Pyzer-Knapp, E. O., Simm, G. N. & Guzik, A. A Bayesian approach to calibrating high-throughput virtual screening results and application to organic photovoltaic materials. Mater. Horiz. 3, 226–233 (2016).10.1039/C5MH00282F ↗
26. Lopez, S. A., Sanchez-Lengeling, B., de Goes Soares, J. & Aspuru-Guzik, A. Design principles and top non-fullerene acceptor candidates for organic photovoltaics. Joule 1, 857–870 (2017).10.1016/j.joule.2017.10.006 ↗
27. Burden, F. & Winkler, D. Optimal sparse descriptor selection for QSAR using Bayesian methods. QSAR Comb. Sci. 28, 645–653 (2009).10.1002/qsar.200810173 ↗
28. Burden, F. R. & Winkler, D. A. An optimal self‐pruning neural network and nonlinear descriptor selection in QSAR. QSAR Comb. Sci. 28, 1092–1097 (2009).10.1002/qsar.200810202 ↗
29. Winkler, D. A. & Burden, F. R. Bayesian neural nets for modeling in drug discovery. Drug Discov. Today. BIOSILICO 2, 104–111 (2004).10.1016/S1741-8364(04)02393-5 ↗
30. Katritzky, A. R. et al. Quantitative correlation of physical and chemical properties with chemical structure: utility for prediction. Chem. Rev. 110, 5714–5789 (2010).10.1021/cr900238d ↗
31. Wold, S., Eriksson, L. & Clementi, S. in Chemometric Methods in Molecular Design (ed. van de Waterbeemd, H.) 309–338 (Wiley, Weinheim, 1995).10.1002/9783527615452.ch5 ↗
32. Tropsha, A., Gramatica, P. & Gombar, V. K. The importance of being earnest: validation is the absolute essential for successful application and interpretation of QSPR models. QSAR Comb. Sci. 22, 69–77 (2003).10.1002/qsar.200390007 ↗
33. Nowak-Król, A. et al. Modulation of band gap and p-versus n-semiconductor character of ADA dyes by core and acceptor group variation. Org. Chem. Front. 3, 545–555 (2016).10.1039/C6QO00046K ↗
34. Fujita, T. & Winkler, D. A. Understanding the roles of the “two QSARs”. J. Chem. Inf. Model. 56, 269–274 (2016).10.1021/acs.jcim.5b00229 ↗
35. Johansson, U., Sönströd, C., Norinder, U. & Boström, H. Trade-off between accuracy and interpretability for predictive in silico modeling. Future Med. Chem. 3, 647–663 (2011).10.4155/fmc.11.23 ↗
36. Salzner, U. & Kiziltepe, T. Theoretical analysis of substituent effects on building blocks of conducting polymers: 3,4’-substituted bithiophenes. J. Org. Chem. 64, 764–769 (1999).10.1021/jo9813944 ↗
37. Luponosov, Y. N. et al. Effects of electron-withdrawing group and electron-donating core combinations on physical properties and photovoltaic performance in D-π-A star-shaped small molecules. Org. Electron. 32, 157–168 (2016).10.1016/j.orgel.2016.02.027 ↗
38. Golbraikh, A. Molecular dataset diversity indices and their applications to comparison of chemical databases and QSAR analysis. J. Chem. Inf. Comput. Sci. 40, 414–425 (2000).10.1021/ci990437u ↗
39. Becke, A. Density-functional thermochemistry: the role of extract exchange. J. Chem. Phys. 98, 648–645 (1993).
40. Becke, A. D. Density-functional exchange-energy approximation with correct asymptotic behavior. Phys. Rev. A 38, 3098–3100 (1988).10.1103/PhysRevA.38.3098 ↗
41. Weigend, F. & Ahlrichs, R. Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: design and assessment of accuracy. Phys. Chem. Chem. Phys. 7, 3297–3305 (2005).10.1039/b508541a ↗
42. Ma, J., Li, S. & Jiang, Y. A time-dependent DFT study on band gaps and effective conjugation lengths of polyacetylene, polyphenylene, polypentafulvene, polycyclopentadiene, polypyrrole, polyfuran, polysilole, polyphosphole, and polythiophene. Macromolecules 35, 1109–1115 (2002).10.1021/ma011279m ↗
43. Churchwell, C. J. et al. The signature molecular descriptor: 3. Inverse-quantitative structure–activity relationship of ICAM-1 inhibitory peptides. J. Mol. Graph. Model. 22, 263–273 (2004).10.1016/j.jmgm.2003.10.002 ↗
44. Faulon, J.-L., Churchwell, C. J. & Visco, D. P. The signature molecular descriptor. 2. Enumerating molecules from their extended valence sequences. J. Chem. Inf. Comput. Sci. 43, 721–734 (2003).10.1021/ci020346o ↗
45. Faulon, J.-L., Visco, D. P. & Pophale, R. S. The signature molecular descriptor. 1. Using extended valence sequences in QSAR and QSPR studies. J. Chem. Inf. Comput. Sci. 43, 707–720 (2003).10.1021/ci020345w ↗
46. Carbonell, P., Carlsson, L. & Faulon, J.-L. Stereo signature molecular descriptor. J. Chem. Inf. Model. 53, 887–897 (2013).10.1021/ci300584r ↗
47. O’Boyle, N. M. et al. Open Babel: an open chemical toolbox. J. Cheminformatics 3, 33 (2011).10.1186/1758-2946-3-33 ↗
48. Topliss, J. G. & Costello, R. J. Chance correlations in structure-activity studies using multiple regression analysis. J. Med. Chem. 15, 1066–1068 (1972).10.1021/jm00280a017 ↗
49. MacKay, D. Bayesian framework for backpropagation networks. Neural Comput. 4, 448–472 (1992).10.1162/neco.1992.4.3.448 ↗
50. Lucic, B., Amic, D. & Trinajstic, N. Nonlinear multivariate regression outperforms several concisely designed neural networks on three QSPR data sets. J. Chem. Inf. Comput. Sci. 40, 403–413 (2000).10.1021/ci990061k ↗
51. Burden, F. & Winkler, D. in Artificial Neural Networks: Methods and Applications (ed. Livingstone, D. J.) 23–42 (Humana Press, 2009).10.1007/978-1-60327-101-1_3 ↗
52. Neal, R. M. in Bayesian Learning for Neural Networks. 29–53 (Springer, New York, 1996).10.1007/978-1-4612-0745-0_2 ↗
53. Gauraha, N. Introduction to the LASSO. Resonance 23, 439–464 (2018).10.1007/s12045-018-0635-x ↗
54. Tibshirani, R. Regression shrinkage and selection via the lasso. J. R. Stat. Soc. B 58, 267–288 (1996).10.1111/j.2517-6161.1996.tb02080.x ↗
55. Feiler, C. et al. In silico screening of modulators of magnesium dissolution. Corros. Sci. 163, 108245 (2019).10.1016/j.corsci.2019.108245 ↗
56. Manallack, D. T., Burden, F. R. & Winkler, D. A. Modelling inhalational anaesthetics using Bayesian feature selection and QSAR modelling methods. ChemMedChem 5, 1318–1323 (2010).10.1002/cmdc.201000056 ↗
57. Mikulskis, P., Alexander, M. R. & Winkler, D. A. Towards Interpretable machine learning models for materials discovery. Adv. Intell. Syst. 1, 1900045 (2019).10.1002/aisy.201900045 ↗
58. Rasi Ghaemi, S. et al. High-throughput assessment and modeling of a polymer library regulating human dental pulp-derived stem cell behavior. ACS Appl. Mater. Interfaces 10, 38739–38748 (2018).10.1021/acsami.8b12473 ↗
59. Burden, F. R. & Winkler, D. A. Robust QSAR models using Bayesian regularized neural networks. J. Med. Chem. 42, 3183–3187 (1999).10.1021/jm980697n ↗
60. Burden, F. R. & Winkler, D. A. New QSAR methods applied to structure−activity mapping and combinatorial chemistry. J. Chem. Inf. Comput. Sci. 39, 236–242 (1999).10.1021/ci980070d ↗
61. Winkler, D. A. & Burden, F. R. Robust QSAR models from novel descriptors and Bayesian regularised neural networks. Mol. Simul. 24, 243–258 (2000).10.1080/08927020008022374 ↗
62. LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436–444 (2015).10.1038/nature14539 ↗
63. Glorot, X., Bordes, A. & Bengio, Y. in Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics. 315–323 (Fort Lauderdale, FL, USA, 2011).
64. Okut, H. in Artificial Neural Networks—Models and Applications (ed. Rosa, J. L. G.) 27–48 (IntechOPen, London, 2016).
65. Perez-Rodriguez, P. & Gianola, D. brnn: Bayesian Regularization for Feed-Forward Neural Networks. https://CRAN.R-project.org/package=brnn (2020).