.png){kind=icon}
.png){kind=icon}
Return
Graphene ribbon-based terahertz metamaterial for multi-toxic-gas detection: a comprehensive theoretical framework with machine learning and adaptive sensing
.png){kind=icon}
OA
AI
DOI：10.1016/j.sbsr.2026.101059.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We present a comprehensive theoretical and numerical framework for the simultaneous detection of multiple toxic gases (SO 2 , NO 2 , CO, NH 3 , H 2 S) using a graphene ribbon-based terahertz metamaterial absorber. The structure is designed to exhibit multiple resonant modes that exactly match the characteristic rotational-vibrational fingerprints of the target gases in the 0.5–6 THz range. A rigorous multi-physics model is developed, combining quantum-mechanical molecular response via Voigt profiles, a multi-resonance Lorentzian metamaterial model, competitive Langmuir-Hill adsorption isotherms, Maxwell-Garnett effective medium theory for gas-metamaterial composites, and noise analysis for limit of detection (LOD) determination. Closed-form expressions for sensitivity, selectivity, response time, and cross-sensitivity are derived. Numerical simulations predict sub-ppm LODs (0.8 ppm for SO 2 , 0.6 ppm for NO 2 ), response times below 0.5 s, and selectivity ratios exceeding 20:1 against common interferents such as H 2 O and CO 2 . A random forest classifier achieves 98.2% accuracy A random forest classifier achieves 98.2% accuracy on synthetic spectral datasets, with a detailed analysis of accuracy versus signal-to-noise ratio. Environmental stability is analyzed over temperature (250–350 K), humidity (0–100% RH), and long-term operation (3 years) using theoretical degradation models. Adaptive sensing and sensor fusion algorithms are shown to improve measurement efficiency and accuracy. The theoretical predictions are presented with clear discussion of model limitations, including effective medium theory applicability, surface adsorption effects, and the need for experimental validation. This work establishes a solid theoretical foundation for the development of low-cost, real-time, multi-gas sensors for environmental monitoring, industrial safety, and healthcare.
Keywords:
Terahertz spectroscopy
Graphene metamaterials
Multi-gas detection
Sulfur dioxide
Nitrogen dioxide
Carbon monoxide
Ammonia
Hydrogen sulfide
Machine learning
Limit of detection
Environmental monitoring
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.9
Papers:
736
Citations:
2.6K
Organization
View more organizations