Open Access

ARTICLE

ZnO/ZnS sensor with broadband visible response for flexible polyethylene terephthalate substrates combined with artificial intelligence analysis

X. Y. Chen^a,b, Y. H. Cai^a, Y. S. Chen^c, S. J. Huang^b, M. H. Li^d, Y. H. Li^e, C. H. Lin^c, H. Chen^a,*

a Department of Applied Materials and Optoelectronic Engineering, National Chi Nan University, Puli 54561, Taiwan ROC
b Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei 10607, Taiwan ROC
c Department of Photonics, National Cheng Kung University, Tainan 70101, Taiwan ROC
d Department of Optoelectronic Engineering, National Formosa University, Yunlin 63201, Taiwan ROC
e AI Research Center, Hon Hai Research Institute, 11494, Taipei, Taiwan ROC

* Corresponding Author:

Chalcogenide Letters 2025, 22(9), 777-785. https://doi.org/10.15251/CL.2025.229.777

Received 29 June 2025; Accepted 07 September 2025;

Abstract

This study focuses on the development of zinc oxide (ZnO)/zinc sulfide (ZnS) core-shell structures on flexible polyethylene terephthalate (PET) substrates for enhanced light sensing. PET offers high elasticity, optical transparency, and chemical resistance, making it ideal for wearable optoelectronics. By optimizing the vulcanization process, a uniform ZnS shell is formed on the exposed regions of ZnO nanorods (NRs), significantly enhancing ZnO-based sensor’s sensitivity to visible light, especially red light (peak wavelength at 630 nm). Structural and spectral analyses confirm the successful formation of the ZnO/ZnS heterostructure, improved charge separation, and broadened light response. To improve data processing and classification accuracy, a one-dimensional convolutional neural network (1D-CNN) is applied to analyze the time-series signals from the sensor. The model achieves 100% training accuracy and nearly perfect performance on the test set, as shown in the confusion matrices. This demonstrates strong generalization and stable classification across different light conditions. The integration of nanomaterial engineering and AI-assisted analysis highlights a promising strategy for future development of intelligent, flexible, and high-performance optical sensing systems.

---

Keywords

---