Physics‐Guided Transformer Enables Efficient Inverse Design of Terahertz Metasurfaces for Glucose Sensing
Small · 22.08.2026
DOI: 10.1002/smll.75213 ↗Kurz gefasst
Fachartikel in der Zeitschrift Small, online erschienen am 22.08.2026. Verfasst von Yafeng Hao und 13 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Widegap Semiconductor Optoelectronic Materials and Technologies, North University of China, Taiyuan, China. Die Arbeit nennt 52 Literaturangaben. Gefördert von: National Natural Science Foundation of China, General Project of China Postdoctoral Science Foundation, Ningbo Science and Technology Program.
Abstract (Originalsprache)
Inverse design of terahertz (THz) metasurfaces via algorithmic models has become a mainstream trend and finds wide application in THz biosensing devices. However, traditional inverse design methods without optimization often rely solely on mathematical correlations, lacking physical interpretability and resulting in outcomes that deviate from practical requirements. This work proposes a Physics‐Guided Transformer (PGT) model tailored for the inverse design of THz biosensing metasurfaces. By incorporating an additional physics‐guided spectral attention (PGS‐Attention) module, the proposed model effectively captures global electromagnetic coupling effects and long‐range dependencies between THz spectral sequences and metasurface units, significantly enhancing physical interpretability. As a result, the computation time of the PGT model is 6.38 × 10 6 times faster than that of full‐wave simulations. Furthermore, label‐free biosensing experiments are conducted on the inversely designed metasurface using glucose solutions with 0.1‐0.5 g/mL. Experimental results demonstrate that the electromagnetically induced transparency (EIT) window generated at 0.96 THz achieves a sensitivity of 280 GHz/RIU. This study provides a comprehensive technical route for physically interpretable inverse design models of THz metasurfaces and offers a solution to the black‐box dilemma in artificial intelligence (AI)‐driven micro/nano‐optics design.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- State Key Laboratory of Widegap Semiconductor Optoelectronic Materials and Technologies, North University of China, Taiyuan, China.
- College of Electronic Information Engineering & Hebei Key Laboratory of Digital Medical Engineering, Hebei University, Baoding, China.
- Forehope Electronic (Ningbo) Co., Ltd., Ningbo, China.
- School of Mechanical Engineering and Mechanics, Eastern Institute of Technology, Ningbo, China.
ORCID-Profile: Bing Han
Förderung
- National Natural Science Foundation of China
- General Project of China Postdoctoral Science Foundation
- Ningbo Science and Technology Program
Zitieren
APA
Hao, Y., Dang, L., Huang, Y., Yao, S., Niu, W., Ji, Y., Chang, Y., Tian, H., Lei, C., Zhong, L., Xu, Y., Liang, T., Han, B., & Li, T. (2026). Physics‐Guided Transformer Enables Efficient Inverse Design of Terahertz Metasurfaces for Glucose Sensing. Small. https://doi.org/10.1002/smll.75213
Vancouver
Hao Y, Dang L, Huang Y, Yao S, Niu W, Ji Y, et al. Physics‐Guided Transformer Enables Efficient Inverse Design of Terahertz Metasurfaces for Glucose Sensing. Small. 2026. doi:10.1002/smll.75213
BibTeX
@article{Hao2026,
author = {Hao, Yafeng and Dang, Lizhi and Huang, Yujie and Yao, Shuaishuai and Niu, Wenyu and Ji, Yukun and Chang, Yaning and Tian, Huiyou and Lei, Cheng and Zhong, Lei and Xu, Yupeng and Liang, Ting and Han, Bing and Li, Tengteng},
title = {Physics‐Guided Transformer Enables Efficient Inverse Design of Terahertz Metasurfaces for Glucose Sensing},
journal = {Small},
year = {2026},
doi = {10.1002/smll.75213}
}
EUVIMED