Hybrid graphene MXene WS₂ terahertz metasurface detects chikungunya

Researchers have developed a computational design for a terahertz metasurface biosensor that integrates graphene, MXene, tungsten disulfide, gold, and copper into a multilayer structure. This sensor detects refractive-index changes linked to chikungunya virus infection. The study, published in Results in Optics by G. Arunachalam, A.R. Kalaiarasi, Giri.G. Hallur, and V. Parthasarathy, reports a predicted sensitivity of 907 GHz per refractive-index unit. The design also incorporates an XGBoost machine-learning model that predicts sensor behavior with over 99.8% accuracy.

Chikungunya virus, transmitted by Aedes mosquitoes, causes severe joint pain and lacks approved treatments or vaccines. Current diagnostic methods face challenges, especially in decentralized settings. The proposed sensor leverages the terahertz band, which interacts with biomolecules without ionizing damage, enabling label-free detection.

The sensor's architecture features a square unit cell with resonators coated in tungsten disulfide, enhancing light–matter interaction. MXene-functionalized concentric rings and an outer gold ring improve electromagnetic coupling and plasmonic interactions. A graphene sheet acts as a broadband absorber, while copper elements enhance conductivity. The structure is supported by a silicon dioxide substrate, ensuring mechanical stability.

Fabrication aligns with established microfabrication techniques, involving polymer-assisted graphene transfer and deposition of metal resonators. The graphene surface can be functionalized with chikungunya-specific molecules for biosensing. Electromagnetic modeling shows a resonance at 0.355 THz, with significant modulation of transmission contrast through graphene's chemical potential.

The sensor maintains high transmittance across various angles, crucial for practical deployment. Geometric optimization enhances resonance strength, achieving a sensitivity of 907 GHz/RIU. The machine-learning model reduces computational costs, enabling rapid design space exploration without compromising accuracy.

The work is a computational proof-of-concept, detecting changes in the dielectric environment rather than specific viral particles. Future steps include device fabrication, biofunctionalization, and experimental validation to develop a practical tool for rapid, label-free diagnostics.