Scholarlink

<p>This work presents a terahertz (THz) glucose biosensor based on a graphene–gold hybrid metasurface that combines strong plasmonic confinement with electrical tunability. The sensor was numerically analysed using the finite element method, and its performance was evaluated over a refractive index range of 1.335–1.347 RIU, corresponding to glucose-induced variations in blood and interstitial fluid. The optimized structure achieved a maximum sensitivity of 1000&#xa0;GHz·RIU⁻<sup>1</sup>, with a constant full width at half maximum of 0.068 THz, yielding a figure of merit and detection accuracy of 14.706 RIU⁻<sup>1</sup>. The resonance frequency exhibited a linear dependence on both glucose concentration (R<sup>2</sup> = 1.00) and refractive index (R<sup>2</sup> = 0.85). Bayesian ridge regression was employed to model the relationship between resonance characteristics and sensing parameters, achieving high predictive accuracy with quantified uncertainty. The results demonstrate the potential of graphene–gold metasurfaces for high-performance, non-invasive THz glucose sensing.</p> Graphical abstract <p></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

Electrically tunable graphene–gold metasurface terahertz sensor for glucose detection using Bayesian ridge regression

  • 期刊论文

  • 发表日期：

    2026年5月6日

查看全文
  • Jacob Wekalao,
  • Hussein A. Elsayed,
  • Haifa A. Alqhtani,
  • Abdulkarem H. M. Almawgani,
  • Hussein S. Gumaih,
  • Yousif S. Adam,
  • Ahmed Mehaney,
  • Amuthakkannan Rajakannu

摘要

This work presents a terahertz (THz) glucose biosensor based on a graphene–gold hybrid metasurface that combines strong plasmonic confinement with electrical tunability. The sensor was numerically analysed using the finite element method, and its performance was evaluated over a refractive index range of 1.335–1.347 RIU, corresponding to glucose-induced variations in blood and interstitial fluid. The optimized structure achieved a maximum sensitivity of 1000 GHz·RIU⁻1, with a constant full width at half maximum of 0.068 THz, yielding a figure of merit and detection accuracy of 14.706 RIU⁻1. The resonance frequency exhibited a linear dependence on both glucose concentration (R2 = 1.00) and refractive index (R2 = 0.85). Bayesian ridge regression was employed to model the relationship between resonance characteristics and sensing parameters, achieving high predictive accuracy with quantified uncertainty. The results demonstrate the potential of graphene–gold metasurfaces for high-performance, non-invasive THz glucose sensing.

Graphical abstract