MXene-metasurface plasmonic sensor for THz detection of Cu²⁺ and Mg²⁺ ions in water: Numerical investigation, machine learning assisted prediction and mining wastewater monitoring

Freshwater contamination by heavy metals and dissolved ionic contaminants has become a critical environmental and public health concern, particularly in mining regions where acid mine drainage, mineral processing activities, and wastewater discharge introduce hazardous metal contaminants into aquatic ecosystems. Ensuring sustainable water management and protecting groundwater resources in mining-affected regions require rapid, accurate, and field-deployable monitoring technologies for heavy metal contamination. Conventional detection methods often require expensive laboratory instrumentation and skilled personnel, limiting their suitability for rapid and on-site monitoring. To address these challenges, this study proposes a MXene-based terahertz (THz) metasurface plasmonic sensor for sensitive detection of Cu 2+ and Mg 2+ ions commonly found in mining wastewater and industrial effluents. The sensor consists of a periodic 4×4 array of unit cells incorporating Au/Ag elliptical resonators, a graphene-coated square resonator, and a MXene-functionalized square split-ring resonator on a SiO 2 substrate. Three-dimensional finite element simulations are performed in COMSOL Multiphysics to investigate resonance characteristics, electromagnetic field confinement, and tunability. A machine learning framework is also developed to rapidly predict spectral responses without repeated full-wave simulations. The proposed structure exhibits multiple sharp plasmonic resonances with stable performance up to a 70° incidence angle. Graphene chemical potential tuning from 0.1 to 0.9 eV enables dynamic spectral modulation. The sensor achieves maximum sensitivities of 0.966 THz/RIU for Cu 2+ and 0.945 THz/RIU for Mg 2+ , with figures of merit of 24.769 RIU −1 and 17.830 RIU −1 , respectively. The machine learning model attains R 2 > 0.9998 and MAPE <0.21%, demonstrating the potential of the proposed platform for real-time monitoring of mining wastewater, industrial discharge, and groundwater resources near mining sites.