Scholarlink

<p>This paper presents a metasurface-based glucose sensor that integrates graphene with gold and silver nanostructures. The sensor features an optimized design, comprising rectangular silver resonators surrounded by concentric gold rings with precisely engineered dimensions. The optimized design demonstrates high sensitivity with peak value of 1000 GHzRIU⁻<sup>1</sup> within the frequency range of 0.3–0.7 THz and 1–1.3 THz, with a figure of merit of 22.222 RIU⁻<sup>1</sup>, a quality factor of 11.244, and a detection accuracy of 22.222. The sensor’s performance was evaluated by varying key parameters, including the chemical potential of graphene (0.1–0.9&#xa0;eV), angle of incidence (0–80°), and geometric dimensions. A 1D convolutional neural network model was developed to optimize the sensor design, achieving 100% accuracy in predicting sensor responses across various operating conditions. The proposed sensor offers a promising non-invasive solution for glucose monitoring, leveraging the tunable properties of graphene and the enhanced sensitivity of noble metal plasmonic, all within a compact and practical design.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-Sensitivity Glucose Detection Using a Terahertz Metasurface-Based Surface Plasmon Resonance Biosensor with Graphene and Plasmonic Nanostructures, Optimized by Machine Learning

  • 期刊论文

  • 发表日期：

    2025年2月15日

查看全文
  • Jacob Wekalao

摘要

This paper presents a metasurface-based glucose sensor that integrates graphene with gold and silver nanostructures. The sensor features an optimized design, comprising rectangular silver resonators surrounded by concentric gold rings with precisely engineered dimensions. The optimized design demonstrates high sensitivity with peak value of 1000 GHzRIU⁻1 within the frequency range of 0.3–0.7 THz and 1–1.3 THz, with a figure of merit of 22.222 RIU⁻1, a quality factor of 11.244, and a detection accuracy of 22.222. The sensor’s performance was evaluated by varying key parameters, including the chemical potential of graphene (0.1–0.9 eV), angle of incidence (0–80°), and geometric dimensions. A 1D convolutional neural network model was developed to optimize the sensor design, achieving 100% accuracy in predicting sensor responses across various operating conditions. The proposed sensor offers a promising non-invasive solution for glucose monitoring, leveraging the tunable properties of graphene and the enhanced sensitivity of noble metal plasmonic, all within a compact and practical design.