Citation: Ya-Jun Yin, Li-Fang Yin, Yi Zhao, Xin Xu, Yu-Qi Xia, Jing-Jing Zhao, Jia-Qi Bai, Guang-Jun Nan, Ji-Fen Wang, Lu-Yi Zou. Data-driven generation and efficient screening of MR-TADF materials[J]. Chinese Chemical Letters, ;2026, 37(9): 112633. doi: 10.1016/j.cclet.2026.112633 shu

Data-driven generation and efficient screening of MR-TADF materials

Ya-Jun Yin ^{a, b, 1} ,
Li-Fang Yin ^{a, 1} ,
Yi Zhao ^{a, 1} ,
Xin Xu ^a ,
Yu-Qi Xia ^a ,
Jing-Jing Zhao ^a ,
Jia-Qi Bai ^a ,
Guang-Jun Nan ^{c, *,} ,
Ji-Fen Wang ^b ,
Lu-Yi Zou ^{a, *,}

a.
Institute of Theoretical Chemistry, College of Chemistry, Jilin University, Changchun 130023, China
b.
School of Resources and Environmental Engineering, Shanghai Polytechnic University, Shanghai Engineering Research Center of Advanced Thermal Functional Materials, Shanghai 201209, China
c.
Department of Physics, Zhejiang Normal University, Jinhua 321004, China

gjnan@zjnu.edu.cn

zouly@jlu.edu.cn

Received Date: 28 November 2025
Revised Date: 5 March 2026
Accepted Date: 16 March 2026
Available Online: 17 March 2026

Figures(12)

The pursuit of high color purity in next-generation displays has spurred intense interest in multiple-resonance thermally activated delayed fluorescence (MR-TADF) emitters, which combine narrowband emission with high exciton utilization efficiency. However, the intricate relationship between molecular structure and photophysical properties renders rational design highly challenging. Herein, we present an integrated computational framework that unifies automated data extraction, de novo molecular generation and multi-property prediction for the targeted discovery of MR-TADF materials. Our approach leverages large-language-model-based literature mining and chemical structure recognition to construct a curated dataset (MR585), which guides a Transformer-based variational autoencoder for generative molecular design. Machine learning models trained on this dataset achieve test-set R² values of 0.692 for predicting the singlet-triplet energy gap (ΔE_ST) and 0.788 for full width at half maximum (FWHM). From the generated candidates, we identify 15 structurally novel emitters with ΔE_ST values of 0.13–0.33 eV and narrow FWHMs of 26.2–46.6 nm. Theoretical analysis further reveals that two sulphur-containing BCzBN derivatives exhibit enhanced spin-orbit coupling, accelerated reverse intersystem crossing and reduced efficiency roll-off compared to their parent structures. This work establishes a comprehensive AI-driven platform that bridges automated knowledge extraction with generative chemistry, offering an efficient and scalable pathway towards high-performance narrowband emitters.
- MR-TADF,
- Machine learning,
- Transformer-VAE,
- Singlet-triplet energy gap,
- Full width at half maximum
1. [1]
  A. Farokhi, S. Lipinski, L.M. Cavinato, et al., Chem. Soc. Rev. 54 (2025) 266–340. doi: 10.1039/d3cs01102j
2. [2]
  S. Diesing, L. Zhang, E. Zysman-Colman, I.D.W. Samuel, Nature 627 (2024) 747–753. doi: 10.1038/s41586-024-07149-x
3. [3]
  T. He, Z. Zhang, L. Yin, et al., Chem. Res. Chin. Univ. 41 (2025) 1133–1143. doi: 10.1007/s40242-025-5163-0
4. [4]
  J.G. Yang, X. Feng, N. Li, et al., Sci. Adv. 9 (2023) eadh0198. doi: 10.1126/sciadv.adh0198
5. [5]
  T.F. He, A.M. Ren, Y.N. Chen, et al., Inorg. Chem. 59 (2020) 12039–12053. doi: 10.1021/acs.inorgchem.0c00980
6. [6]
  M. Liu, B. Hou, Y. Li, Y. Pan, B. Yang, Comput. Theor. Chem. 1231 (2024) 114414. doi: 10.1016/j.comptc.2023.114414
7. [7]
  Y. Takeda, Acc. Chem. Res. 57 (2024) 2219–2232. doi: 10.1021/acs.accounts.4c00353
8. [8]
  T.F. He, A.M. Ren, G.H. Li, et al., J. Phys. Chem. Lett. 12 (2021) 2232–2244. doi: 10.1021/acs.jpclett.1c00119
9. [9]
  Y.L. Zhang, T.F. He, Z.K. Zhao, et al., Inorg. Chem. 62 (2023) 7753–7763. doi: 10.1021/acs.inorgchem.3c00383
10. [10]
  J.M. Ha, S.H. Hur, A. Pathak, J.E. Jeong, H.Y. Woo, NPG. Asia Mater. 13 (2021) 53. doi: 10.1038/s41427-021-00318-8
11. [11]
  Z.K. Zhao, T.F. He, Q. Gao, et al., Inorg. Chem. 63 (2024) 17435–17448. doi: 10.1021/acs.inorgchem.4c01657
12. [12]
  T. Hatakeyama, K. Shiren, K. Nakajima, et al., Adv. Mater. 28 (2016) 2777–2781. doi: 10.1002/adma.201505491
13. [13]
  Z. Yang, S. Li, L. Hua, et al., Chem. Sci. 16 (2025) 3904–3915. doi: 10.1039/d4sc08708a
14. [14]
  M.X. Du, J.O. Zhou, X.F. Luo, L. Duan, Moore More 1 (2025) 79–98. doi: 10.1109/mwc.2025.3600040
15. [15]
  K.R. Naveen, H.I. Yang, J.H. Kwon, Commun. Chem. 5 (2022) 149. doi: 10.1038/s42004-022-00766-5
16. [16]
  G. Ricci, A. Landi, J.C. Sancho-García, Y. Olivier, Adv. Opt. Mater. 13 (2025) 2402765. doi: 10.1002/adom.202402765
17. [17]
  B.H. Jhun, Y. Park, H.S. Kim, et al., Nat. Commun. 16 (2025) 392. doi: 10.1038/s41467-024-55620-0
18. [18]
  J.M. Jin, C. Shi, W.C. Chen, Y. Huo, Chem. Commun. 61 (2025) 10731–10746. doi: 10.1039/d5cc02070k
19. [19]
  K. Shizu, H. Kaji, Commun. Chem. 5 (2022) 53. doi: 10.1038/s42004-022-00668-6
20. [20]
  D. Hall, J.C. Sancho-García, A. Pershin, et al., J. Phys. Chem. A 127 (2023) 4743–4757. doi: 10.1021/acs.jpca.2c08201
21. [21]
  J.M. Dos Santos, D. Hall, B. Basumatary, et al., Chem. Rev. 124 (2024) 13736–14110. doi: 10.1021/acs.chemrev.3c00755
22. [22]
  A. Wu, Q. Ye, X. Zhuang, et al., Precis. Chem. 1 (2023) 57–68. doi: 10.1021/prechem.3c00005
23. [23]
  Y. Chen, W. Yan, Z. Wang, J. Wu, X. Xu, J. Chem. Theory. Comput. 20 (2024) 9500–9511. doi: 10.1021/acs.jctc.4c01151
24. [24]
  Z. Wang, W. Zhang, M. Jiang, et al., J. Phys. Chem. Lett. 15 (2024) 12501–12512. doi: 10.1021/acs.jpclett.4c03214
25. [25]
  H.S. Kim, H.J. Cheon, S.H. Lee, et al., Sci. Adv. 11 (2025) eadr1326. doi: 10.1126/sciadv.adr1326
26. [26]
  H. Shi, Y. Shi, Z. Liang, et al., Chem. Engin. J. 494 (2024) 153150. doi: 10.1016/j.cej.2024.153150
27. [27]
  Z. Chen, J. Song, L. Hu, et al., Chin. Chem. Lett. 37 (2026) 111967. doi: 10.1016/j.cclet.2025.111967
28. [28]
  Y. Yin, L. Yin, Y. Zhao, et al., Chem. Res. Chin. Univ. 41 (2025) 1173–1185. doi: 10.1007/s40242-025-5175-9
29. [29]
  S.Y. Zhang, Y. Zhao, L. Wang, Z.P. Liu, Y. Tang, CCS Chem. 7 (2025) 3493–3506. doi: 10.31635/ccschem.025.202405037
30. [30]
  Y. Li, B. Zhang, A. Ren, et al., Chem. Engin. J. 501 (2024) 157676. doi: 10.1016/j.cej.2024.157676
31. [31]
  D. Huang, J.M. Cole, Sci. Data 11 (2024) 80. doi: 10.1038/s41597-023-02897-3
32. [32]
  J. Arús-Pous, S.V. Johansson, O. Prykhodko, et al., J. Cheminform. 11 (2019) 71. doi: 10.1186/s13321-019-0393-0
33. [33]
  D. Weininger, J. Chem. Inf. Comput. Sci. 28 (1988) 31–36. doi: 10.1021/ci00057a005
34. [34]
  K. Rajan, A. Zielesny, C. Steinbeck, J. Cheminformatics 12 (2020) 65. doi: 10.1186/s13321-020-00469-w
35. [35]
  D. Rogers, M. Hahn, J. Chem. Inf. Model. 50 (2010) 742–754. doi: 10.1021/ci100050t
36. [36]
  H. Shi, Y. Shi, G. Zeng, et al., Chem. Engin. J. 525 (2025) 170146. doi: 10.1016/j.cej.2025.170146
37. [37]
  F. Chen, L. Zhao, X. Wang, et al., Sci. China Chem. 64 (2021) 547–551. doi: 10.1007/s11426-020-9944-1
38. [38]
  A. Vaswani, N. Shazeer, N. Parmar, et al., in: Proceedings of the 31st international conference on neural information processing systems, Curran Associates Inc., New York, USA, 2017, pp. 6000–6010.
39. [39]
  R. Gómez-Bombarelli, J.N. Wei, D. Duvenaud, et al., ACS Cent. Sci. 4 (2018) 268–276. doi: 10.1021/acscentsci.7b00572
40. [40]
  M.H.S. Segler, T. Kogej, C. Tyrchan, M.P. Waller, ACS Cent. Sci. 4 (2018) 120–131. doi: 10.1021/acscentsci.7b00512
41. [41]
  D.P. Kingma, J. Ba, in: Proceedings of 3rd international conference on learning representations (ICLR 2015), San Diego, CA, USA, 2015.
42. [42]
  S. Jasial, Y. Hu, M. Vogt, J. Bajorath, F1000Res. 5 (2016) 591. doi: 10.12688/f1000research.8357.1
43. [43]
  W.S. Torgerson, Psychometrika 17 (1952) 401–419. doi: 10.1007/BF02288916
44. [44]
  L. Breiman, Mach. Learn. 45 (2001) 5–32. doi: 10.1023/A:1010933404324
45. [45]
  T. Chen, C. Guestrin, in: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, Association for Computing Machinery, New York, USA, 2016, pp. 785–794.
46. [46]
  J.H. Friedman, Ann. Stat. 29 (2001) 1189–1232. doi: 10.1214/aos/1013203450
47. [47]
  P. Geurts, D. Ernst, L. Wehenkel, Mach. Learn. 63 (2006) 3–42. doi: 10.1007/s10994-006-6226-1
48. [48]
  J. Bergstra, R. Bardenet, Y. Bengio, B. Kégl, Advances in Neural Information Processing Systems, J. Shawe-Taylor, R. Zemel, P. Bartlett, F. Pereira, K.Q. Weinberger (Eds.) Eds, Curran Associates, Inc., 2011.
49. [49]
  D.R. Jones, M. Schonlau, W.J. Welch, J. Glob. Optim. 13 (1998) 455–492. doi: 10.1023/A:1008306431147
50. [50]
  X.F. Luo, H.X. Ni, H.L. Ma, et al., Adv. Opt. Mater. 10 (2022) 2102513. doi: 10.1002/adom.202102513
51. [51]
  Q. Li, Y. Wu, X. Wang, et al., Chem. Eur. J. 28 (2022) e202104214. doi: 10.1002/chem.202104214
52. [52]
  C. Jiang, Y. Nie, C. Cao, et al., Adv. Sci. 13 (2026) e12796. doi: 10.1002/advs.202512796
53. [53]
  Z. Huang, H. Xie, J. Miao, et al., J. Am. Chem. Soc. 145 (2023) 12550–12560. doi: 10.1021/jacs.3c01267
54. [54]
  T.F. He, Y.N. Chen, X.L. Hao, et al., J. Phys. Chem. Lett. 17 (2026) 205–213. doi: 10.1021/acs.jpclett.5c03741
55. [55]
  L. Yin, Y. Zhao, Q. Gao, et al., Chem. Sci. 17 (2026) 2599–2615. doi: 10.1039/d5sc06741c
1. [1]
  Zhiyuan Chen , Jinyu Song , Lai Hu , Peng Xu , Zhengyi Sun , Xiao-Chun Hang , Hongjun Zhu , Senqiang Zhu , Rui Liu . Structure-aware machine learning for predicting photophysical properties of MR-TADF materials. Chinese Chemical Letters, 2026, 37(7): 111967-. doi: 10.1016/j.cclet.2025.111967
2. [2]
  Lu Li , Jianing Shen , Qinkun Xiao , Chaozheng He , Jinzhou Zheng , Chaoqin Chu , Chen Chen . Stable crystal structure prediction using machine learning-based formation energy and empirical potential function. Chinese Chemical Letters, 2025, 36(11): 110421-. doi: 10.1016/j.cclet.2024.110421
3. [3]
  Qingyun Hu , Wei Wang , Junyuan Lu , He Zhu , Qi Liu , Yang Ren , Hong Wang , Jian Hui . High-throughput screening of high energy density LiMn_1-xFe_xPO₄ via active learning. Chinese Chemical Letters, 2025, 36(2): 110344-. doi: 10.1016/j.cclet.2024.110344
4. [4]
  Yuting Wu , Haifeng Lv , Xiaojun Wu . Design of two-dimensional porous covalent organic framework semiconductors for visible-light-driven overall water splitting: A theoretical perspective. Chinese Journal of Structural Chemistry, 2024, 43(11): 100375-100375. doi: 10.1016/j.cjsc.2024.100375
5. [5]
  Honglin Chen , Rupeng Wang , Zixiang He , Shih-Hsin Ho . Data-driven insights into nonradical activation mechanisms for biochar inverse design: A synergistic approach using DFT and machine learning with meta-analysis. Chinese Chemical Letters, 2026, 37(2): 111372-. doi: 10.1016/j.cclet.2025.111372
6. [6]
  Zonglin Li , Shihua Zou , Zining Wang , Georgeta Postole , Liang Hu , Hongying Zhao . Machine learning in electrochemical oxidation process: A mini-review. Chinese Chemical Letters, 2025, 36(8): 110526-. doi: 10.1016/j.cclet.2024.110526
7. [7]
  Yunzhe Zheng , Si Sun , Jiali Liu , Qingyu Zhao , Heng Zhang , Jing Zhang , Peng Zhou , Zhaokun Xiong , Chuan-Shu He , Bo Lai . Application of machine learning for material prediction and design in the environmental remediation. Chinese Chemical Letters, 2025, 36(9): 110722-. doi: 10.1016/j.cclet.2024.110722
8. [8]
  Yajun Mao , Huchuan Yan , Keteng Li , Cui Lai , Xing Fan , Dengsheng Ma , Guangming Zeng , Lei Qin . Machine learning in electrocatalytic ammonia synthesis: A mini review on catalyst discovery and optimization. Chinese Chemical Letters, 2026, 37(9): 112440-. doi: 10.1016/j.cclet.2026.112440
9. [9]
  Zixing Xu , Ruiying Chen , Chuanming Hao , Qionghong Xie , Chunhui Deng , Nianrong Sun . Peptidome data-driven comprehensive individualized monitoring of membranous nephropathy with machine learning. Chinese Chemical Letters, 2024, 35(5): 108975-. doi: 10.1016/j.cclet.2023.108975
10. [10]
  Han-Bin Liu , Xiaoyu Cheng , Zhou Guo , Juan Yang , Fuwen Tan , Donghui Lan , Jian-Ping Tan , Bing Yi , Weixin Zhai , Qing-Hui Guo . CrownBind-IA: A machine learning model predicting binding constants between crown ethers and alkali metal ions. Chinese Chemical Letters, 2025, 36(12): 111149-. doi: 10.1016/j.cclet.2025.111149
11. [11]
  Xu He , Wenjie Gao , Jinglei Xu , Zhanjun Cheng , Wenchao Peng , Beibei Yan , Guanyi Chen , Ning Li . Machine learning-assisted construction of C=O and pyridinic N active sites in sludge-based catalysts. Chinese Chemical Letters, 2025, 36(12): 111019-. doi: 10.1016/j.cclet.2025.111019
12. [12]
  Shuang Li , Penghui Yuan , Xinyi Zhang , Meiru Liu , Dezhi Yang , Linglei Kong , Li Zhang , Yang Lu , Guanhua Du . Revolutionizing sepsis therapy: Machine learning-driven co-crystallization reveals emodin's therapeutic potential. Chinese Chemical Letters, 2026, 37(2): 111289-. doi: 10.1016/j.cclet.2025.111289
13. [13]
  Na Qin , Wenxin Guo , Fangxiu Li , Houfeng Zhang , Hong Liu , Chang Zhang , Lipiao Bao , Lei Liu , Muneerah Alomar , Siqi Zhao , Jian Zhang , Xing Lu . Recent advances in machine learning-driven discovery of alloy electrocatalysts for hydrogen evolution reaction. Chinese Chemical Letters, 2026, 37(3): 112021-. doi: 10.1016/j.cclet.2025.112021
14. [14]
  Dongshi Feng , Jiangdong Dai , Zhi Zhu , Pengwei Huo , Yongsheng Yan , Chunxiang Li . Molecularly imprinted electrochemical sensor arrays combined with machine learning for simultaneous determination of three neonicotinoid insecticides. Chinese Chemical Letters, 2026, 37(5): 111789-. doi: 10.1016/j.cclet.2025.111789
15. [15]
  Jieyu Liu , Junze Zhang , Haigang Deng , Shuoao Wang , Xingxing Jiang , Li Wang , Changhong Wang . Understanding the activity origin of Pd-anchored single-atom alloy catalysts for NO-to-NH₃ conversion by DFT studies and machine learning. Chinese Chemical Letters, 2025, 36(12): 110656-. doi: 10.1016/j.cclet.2024.110656
16. [16]
  Hao Chen , Haiyuan Liao , Qi Zhou , Yang Liu , Guojun Liu , Yuan Yao . Electronegativity-oriented coordination regulation of main-group metal single-atom catalysts for oxygen reduction to H₂O₂: A combined study of first-principles and machine learning. Chinese Chemical Letters, 2026, 37(3): 110711-. doi: 10.1016/j.cclet.2024.110711
17. [17]
  Yicheng Li , Qian Liu , Tianhao Li , Hao Bi , Zhurui Shen . Recent achievements in rare earth modified metal oxides for environmental and energy applications: A review. Chinese Chemical Letters, 2025, 36(9): 110698-. doi: 10.1016/j.cclet.2024.110698
18. [18]
  Xi Xue , Kai Chen , Hanyu Sun , Xiangying Liu , Xue Liu , Shize Li , Jingjie Yan , Yu Peng , Mohammad S. Mubarak , Ahmed Al-Harrasi , Hai-Yu Hu , Yafeng Deng , Xiandao Pan , Xiaojian Wang . Bidirectional Chemical Intelligent Net: A unified deep learning–based framework for predicting chemical reaction. Chinese Chemical Letters, 2025, 36(11): 110968-. doi: 10.1016/j.cclet.2025.110968
19. [19]
  Chengde Wang , Liping Huang , Shanshan Wang , Lihao Wu , Yi Wang , Jun Dong . A distinction of gliomas at cellular and tissue level by surface-enhanced Raman scattering spectroscopy. Chinese Chemical Letters, 2024, 35(5): 109383-. doi: 10.1016/j.cclet.2023.109383
20. [20]
  Xiaoxiao Wang , Bolun Wang , Fenfen Ji , Jie Yan , Jiacheng Fang , Doudou Zhang , Ji Xu , Jing Ji , Xinran Hao , Hemi Luan , Yanjun Hong , Shulan Qiu , Min Li , Zhu Yang , Wenlan Liu , Xiaodong Cai , Zongwei Cai . Discovery of plasma biomarkers for Parkinson’s disease diagnoses based on metabolomics and lipidomics. Chinese Chemical Letters, 2024, 35(11): 109653-. doi: 10.1016/j.cclet.2024.109653

Get Citation

PDF

XML

Metrics

PDF Downloads(0)
Abstract views(10)
HTML views(2)

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142