Scholarlink

<p>The development of highly sensitive, label-free biosensing operating in THz frequency offers significant potential towards molecular screening and detection systems. This study presents a purely numerical investigation of a BaTiO<sub>3</sub>–graphene–Ag THz metasurface design for highly sensitive refractive-index analysis optimized using machine learning, with prospective relevance toward refractive-index-based SARS-CoV-2 screening and detection applications. The proposed structure uses high-permittivity BaTiO<sub>3</sub>, tunable plasmonic graphene and conductive Ag to increase the electromagnetic field concentration and sensitivity of the resonance within a multi-resonant structure supported by a SiO<sub>2</sub> substrate. Electromagnetic finite-element simulations have shown that a theoretical refractive-index sensitivity of 400&#xa0;GHz/RIU can be achieved and a very linear correlation exists between the resonance frequency and refractive index (R<sup>2</sup>= 0.98394). A tuning of about 30&#xa0;GHz of the resonance frequency was obtained across physiologically relevant changes of refractive index. Parametric studies also confirmed the ability to obtain large tunability by varying the chemical potential of graphene (from 0.1 to 0.9&#xa0;eV) and stability against geometrical changes of the resonators and incidence angles. The Machine-learning model integration provides excellent predictive accuracy (R<sup>2</sup> &gt; 0.999) for the design parameters, validating their efficiency in sensor optimization. This work is carried out exclusively at the simulation level, therefore, the analytical sample is described through its effective refractive index in the THz range. No biochemical functionalization, molecular binding assays, or experiments with biological or clinical samples were performed. The reported results provide a theoretical electromagnetic performance analysis and therefore future work will focus on prototype realization SARS-CoV-2 screening and early detection through target molecule and clinical validation of the diagnostic capabilities of the device.</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

Numerical Modeling and Machine-Learning Optimization of a BaTiO3–Graphene–Ag THz Metasurface for Refractive-Index Biosensing Towards SARS-CoV-2 Detection

  • 期刊论文

  • 发表日期：

    2026年6月16日

查看全文
  • K. Srisabarimani,
  • R. Dhivya,
  • T. Srihari,
  • P. Parthasarathy,
  • U. Arun Kumar

摘要

The development of highly sensitive, label-free biosensing operating in THz frequency offers significant potential towards molecular screening and detection systems. This study presents a purely numerical investigation of a BaTiO3–graphene–Ag THz metasurface design for highly sensitive refractive-index analysis optimized using machine learning, with prospective relevance toward refractive-index-based SARS-CoV-2 screening and detection applications. The proposed structure uses high-permittivity BaTiO3, tunable plasmonic graphene and conductive Ag to increase the electromagnetic field concentration and sensitivity of the resonance within a multi-resonant structure supported by a SiO2 substrate. Electromagnetic finite-element simulations have shown that a theoretical refractive-index sensitivity of 400 GHz/RIU can be achieved and a very linear correlation exists between the resonance frequency and refractive index (R2= 0.98394). A tuning of about 30 GHz of the resonance frequency was obtained across physiologically relevant changes of refractive index. Parametric studies also confirmed the ability to obtain large tunability by varying the chemical potential of graphene (from 0.1 to 0.9 eV) and stability against geometrical changes of the resonators and incidence angles. The Machine-learning model integration provides excellent predictive accuracy (R2 > 0.999) for the design parameters, validating their efficiency in sensor optimization. This work is carried out exclusively at the simulation level, therefore, the analytical sample is described through its effective refractive index in the THz range. No biochemical functionalization, molecular binding assays, or experiments with biological or clinical samples were performed. The reported results provide a theoretical electromagnetic performance analysis and therefore future work will focus on prototype realization SARS-CoV-2 screening and early detection through target molecule and clinical validation of the diagnostic capabilities of the device.