EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Smart molecular design of NIR‐II organic fluorophores through self‐driven iterative evolution, deep learning, and fragment‐based assembly

Yu Zhang, Zhubin Hu, Xinyuan Wang, Zhenrong Sun, Yongye Liang iD, Cheng Zhong iD, Haitao Sun iD

Smart Molecules · 10.06.2026

DOI: 10.1002/smo2.70064 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Smart Molecules, online erschienen am 10.06.2026. Verfasst von Yu Zhang und 6 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Precision Spectroscopy School of Physics East China Normal University Shanghai China. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Near‐infrared II (NIR‐II, 1000–1700 nm) emissive molecules are highly valued for biomedical imaging, phototherapy, and optoelectronic applications due to their deep tissue penetration, reduced autofluorescence, and high signal‐to‐noise ratio. However, their rational design remains challenging, as conventional discovery relies heavily on labor‐intensive synthesis, limited quantum chemical calculations, and inefficient trial‐and‐error exploration. To overcome these limitations, an iterative AI‐driven molecular evolution strategy (AI4NIR‐II 1.0) is introduced that integrates time‐dependent density functional theory (TDDFT), transformer‐based predictive modeling, and generative molecular design. Starting from donor‐acceptor‐donor (D‐A‐D) and donor‐donor‐acceptor‐donor‐donor (D‐D‐A‐D‐D) fragment scaffolds, a predictive model for both absorption and emission properties was trained based on a high quality dataset containing ∼16,000 molecules annotated by the optimally‐tuned range‐separated LC‐ωHPBE* (OTRS) functional. A fine‐tuned generative model was subsequently incorporated into a self‐refining‐loop workflow that cycles through molecular generation, property screening, and dataset augmentation. The model achieves excellent predictive performance for both emission and absorption properties, that is mean absolute errors of 19 nm for emission peak wavelengths and of 11 nm for absorption peak wavelengths, and correlation coefficient ( R 2 ) exceeding 0.98 for wavelengths and oscillator strengths compared to OTRS‐TDDFT calculations. In addition, the resulting framework efficiently identifies promising NIR‐II candidates with accurate photophysical property predictions, and achieves speed improvements of three to four orders of magnitude over TDDFT calculations. Beyond accelerating NIR‐II fluorophore discovery, this self‐driven approach establishes a scalable and generalizable paradigm for NIR‐II molecular design, with applicability extending to optoelectronic materials and therapeutic compounds.

Schlagwörter

Stichwörter der Autor:innen

NIR‐IITDDFTdeep learningfragment‐based assemblymolecular evolution

Autor:innen und Institutionen

  • State Key Laboratory of Precision Spectroscopy School of Physics East China Normal University Shanghai China.
  • Department of Materials Science and Engineering Shenzhen Key Laboratory of Printed Organic Electronic Southern University of Science and Technology Shenzhen China.
  • Department of Chemistry Wuhan University Wuhan Hubei China.
  • Collaborative Innovation Center of Extreme Optics Shanxi University Taiyuan Shanxi China.

ORCID-Profile: Yongye Liang, Cheng Zhong, Haitao Sun

Zitieren

APA

Zhang, Y., Hu, Z., Wang, X., Sun, Z., Liang, Y., Zhong, C., & Sun, H. (2026). Smart molecular design of NIR‐II organic fluorophores through self‐driven iterative evolution, deep learning, and fragment‐based assembly. Smart Molecules. https://doi.org/10.1002/smo2.70064

Vancouver

Zhang Y, Hu Z, Wang X, Sun Z, Liang Y, Zhong C, et al. Smart molecular design of NIR‐II organic fluorophores through self‐driven iterative evolution, deep learning, and fragment‐based assembly. Smart Molecules. 2026. doi:10.1002/smo2.70064

BibTeX
@article{Zhang2026,
  author = {Zhang, Yu and Hu, Zhubin and Wang, Xinyuan and Sun, Zhenrong and Liang, Yongye and Zhong, Cheng and Sun, Haitao},
  title = {Smart molecular design of NIR‐II organic fluorophores through self‐driven iterative evolution, deep learning, and fragment‐based assembly},
  journal = {Smart Molecules},
  year = {2026},
  doi = {10.1002/smo2.70064}
}
Kostenlos starten →