Machine Learning to Adaptively Predict Gold Nanorod Sizes on Different Substrates
Source abstract
Abstract
Correlating a nanoparticle’s morphology with its optical properties is essential and is achieved by a combination of electron microscopy and optical spectroscopy. Machine learning has gained attention for enhancing in situ measurements and enabling inverse nanoparticle design. However, new training data for each specific condition are often required when testing data differ from training data. We propose a method to adapt existing training data for predicting the size of gold nanorods (AuNRs) on different substrates. This method is based on simulated spectra of AuNRs on glass and indium tin oxide-coated glass (ITO), adapting the resonance energy between substrates. Using the adapted data, we train a decision tree regressor to predict AuNR sizes on ITO and test it with experimental data on ITO. This correction achieves comparable accuracy in predicting AuNR length to a decision tree trained directly on ITO. In addition, we apply the correction method to predict AuNR sizes on Al 2 O 3, despite the lack of extensive training data, leading to an improvement in length prediction as well. Our analysis reveals that length prediction is more sensitive to the change in the resonance energy, suggesting that substrate differences mostly affect the length prediction. Overall, adapting training data enables real-time size determination across various environments without additional training data.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Materials Science · Gold and Silver Nanoparticles Synthesis and Applications · Machine Learning and ELM
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Physics
University of Illinois Urbana−Champaign
Department of Chemistry
Department of Chemical and Biomolecular Engineering
Beckman Institute for Advanced Science and Technology
Department of Electrical and Computer Engineering
Funding
Army Research Office · BioXFEL Science and Technology Center · National Science Foundation
References (74)
Show source-provided references
1. Bohren C. F. · 1988
2. 10.1021/j100155a00910.1021/j100155a009 ↗
3. Kreibig U. · 199510.1007/978-3-662-09109-8 ↗
4. 10.1021/acsnano.1c0139910.1021/acsnano.1c01399 ↗
5. 10.1021/jz200702m10.1021/jz200702m ↗
6. 10.1021/jp201030910.1021/jp2010309 ↗
7. 10.1039/D1NR08130F10.1039/D1NR08130F ↗
8. 10.1021/jp051684610.1021/jp0516846 ↗
9. 10.1038/nchem.147810.1038/nchem.1478 ↗
10. 10.1039/C4CS00131A10.1039/C4CS00131A ↗
11. 10.1038/s41566-018-0246-910.1038/s41566-018-0246-9 ↗
12. 10.1126/sciadv.aar420610.1126/sciadv.aar4206 ↗
13. 10.1515/nanoph-2019-047410.1515/nanoph-2019-0474 ↗
14. 10.1021/acsphotonics.7b0137710.1021/acsphotonics.7b01377 ↗
15. 10.1002/aisy.20230012110.1002/aisy.202300121 ↗
16. 10.1038/s41586-018-0337-210.1038/s41586-018-0337-2 ↗
17. 10.1021/acs.nanolett.9b0409010.1021/acs.nanolett.9b04090 ↗
18. 10.1021/acsnano.2c0573110.1021/acsnano.2c05731 ↗
19. 10.1021/acs.chemrev.2c0006110.1021/acs.chemrev.2c00061 ↗
20. 10.1038/s41565-022-01284-010.1038/s41565-022-01284-0 ↗
21. 10.1063/5.016996110.1063/5.0169961 ↗
22. 10.1073/pnas.231567012110.1073/pnas.2315670121 ↗
23. 10.1002/smll.20220765810.1002/smll.202207658 ↗
24. 10.1073/pnas.221140611910.1073/pnas.2211406119 ↗
25. 10.1021/acsnano.3c0551010.1021/acsnano.3c05510 ↗
26. 10.1002/adma.20200162610.1002/adma.202001626 ↗
27. 10.1038/s41578-021-00337-510.1038/s41578-021-00337-5 ↗
28. 10.1038/s41524-021-00520-w10.1038/s41524-021-00520-w ↗
29. 10.1126/sciadv.abj550510.1126/sciadv.abj5505 ↗
30. 10.1021/acs.jpcc.2c0757810.1021/acs.jpcc.2c07578 ↗
31. 10.1021/acs.chemmater.3c0275110.1021/acs.chemmater.3c02751 ↗
32. 10.1039/C9NR03450A10.1039/C9NR03450A ↗
33. 10.1021/acs.jpcc.0c1068010.1021/acs.jpcc.0c10680 ↗
34. 10.1039/D2NH00146B10.1039/D2NH00146B ↗
35. 10.1021/acs.jpcc.3c0593810.1021/acs.jpcc.3c05938 ↗
36. 10.1021/acs.jpcc.4c0245910.1021/acs.jpcc.4c02459 ↗
37. 10.1021/acsnano.4c0975310.1021/acsnano.4c09753 ↗
38. 10.1021/acs.jpclett.2c0302810.1021/acs.jpclett.2c03028 ↗
39. 10.1021/acs.nanolett.4c0126110.1021/acs.nanolett.4c01261 ↗
40. 10.1021/acs.jpcc.1c0393710.1021/acs.jpcc.1c03937 ↗
41. 10.1021/jp990183f10.1021/jp990183f ↗
42. 10.1021/jp050728810.1021/jp0507288 ↗
43. 10.1039/b714759g10.1039/b714759g ↗
44. Quiñonero-Candela J. · 2009
45. 10.1038/s41524-023-01012-910.1038/s41524-023-01012-9 ↗
46. 10.1109/TKDE.2022.317812810.1109/TKDE.2022.3178128 ↗
47. 10.1109/TKDE.2009.19110.1109/TKDE.2009.191 ↗
48. 10.1039/D3DD00162H10.1039/D3DD00162H ↗
49. 10.1186/s40537-019-0197-010.1186/s40537-019-0197-0 ↗
50. 10.48550/arXiv.1211.486010.48550/arXiv.1211.4860 ↗
51. 10.1126/sciadv.150098810.1126/sciadv.1500988 ↗
52. 10.1021/acs.jpca.0c0319010.1021/acs.jpca.0c03190 ↗
53. 10.1021/jp047316410.1021/jp0473164 ↗
54. 10.1063/1.220321210.1063/1.2203212 ↗
55. 10.1103/PhysRevB.86.23514710.1103/PhysRevB.86.235147 ↗
56. 10.1021/la047488s10.1021/la047488s ↗
57. 10.1021/la203451p10.1021/la203451p ↗
58. 10.1002/smll.20150244910.1002/smll.201502449 ↗
59. 10.1021/acsnano.0c0677110.1021/acsnano.0c06771 ↗
60. 10.1038/s41570-022-00391-910.1038/s41570-022-00391-9 ↗
61. 10.1002/adma.20210650610.1002/adma.202106506 ↗
62. Breiman L. · 1984
63. Géron A. · 2022
64. 10.1088/2399-1984/ab9c3b10.1088/2399-1984/ab9c3b ↗
65. Pedregosa F. · J. Mach. Learn. Res. · 2011 · 12 · 2825
66. 10.1063/5.007862110.1063/5.0078621 ↗
67. 10.1021/acs.accounts.0c0087210.1021/acs.accounts.0c00872 ↗
68. 10.1021/acs.jpcc.8b0635010.1021/acs.jpcc.8b06350 ↗
69. 10.1021/acs.nanolett.0c0022310.1021/acs.nanolett.0c00223 ↗
70. 10.1021/jp971656q10.1021/jp971656q ↗
71. 10.1007/s11468-010-9130-210.1007/s11468-010-9130-2 ↗
72. 10.1021/jp991764810.1021/jp9917648 ↗
73. 10.1021/nl073042v10.1021/nl073042v ↗
74. 10.1021/jp101272w10.1021/jp101272w ↗