A Sn-Ta-O-doped vertical graphene electrochemical sensor based on a machine learning prediction model for monitoring...

The growing diversity of beverages intensifies demand for on-site heavy metal detection to ensure safety. Multi-heteroatom co-doped graphene electrodes exhibit superior electrocatalytic activity and stability over traditional carbon-based electrodes, enabling their application in machine learning (ML)-assisted real-time Cd²⁺ detection. An industrial manufacturing process combining physical and chemical vapor deposition was employed to fabricate novel tin(Sn)‑tantalum(Ta)‑oxygen(O)-doped vertical graphene (VG) (STO-VG) electrodes with uniform surface properties. STO-VG sensors combined traditional linear regression with ML prediction models for real-time Cd²⁺ detection in samples. The sensor exhibits a wide detection range (0.1-200 μM), low detection limit (S/N = 3; 1.80 nM), and excellent long-term stability. Furthermore, the sensor demonstrated excellent recovery (95.6 %-105.2 %) and reliability for the real-time monitoring of Cd²⁺ in various beverages. This study provides a stable STO-VG sensor and an efficient (ML)-based strategy for the on-site real-time determination of Cd²⁺.