High-performance electrochemical sensor based on rGO/FeCo-MOF integrated with machine learning for simultaneous
Tien Dat Doan1, Duc Anh Nguyen2, Thi Hai Yen Pham1
1Institute of Chemistry, Vietnam Academy of Science and Technology (VAST), 18 Hoang Quoc Viet, Nghia Do Ward, Hanoi, 10000, Viet Nam.
Abstract:
The simultaneous monitoring of cadmium (Cd2+) and lead (Pb2+) ions in environmental waters is crucial due to their high toxicity, persistence, and bioaccumulation potential. Herein, we developed a novel electrochemical sensor by modifying a glassy carbon electrode (GCE) with a bimetallic iron-cobalt metal-organic framework (FeCo-MOF) and reduced graphene oxide (rGO) for the sensitive detection of Cd2+ and Pb2+. Solvothermally synthesized FeCo-MOF displays enhanced intrinsic conductivity and a larger surface area than the monometallic counterparts. The incorporation of rGO promoted efficient electron transfer and enlarged the electrochemically active surface area, while also serving as a protective layer to improve electrode stability. The synergistic effects between FeCo-MOF and rGO facilitate the preconcentration of Cd2+ and Pb2+ ions and enhance electron transfer, resulting in excellent analytical performance with high sensitivity and good repeatability. For individual determination, the rGO/FeCo-MOF/GCE sensor achieves linear ranges of 1-200 μg/L (limit of detection (LOD) = 0.62 μg/L) for Cd2+ and 0.1-50 μg/L (LOD = 0.06 μg/L) for Pb2+. For simultaneous determination, mutual interference between Cd2+ and Pb2+ was systematically evaluated using eight machine learning models. The Extra Trees regressor provides the best predictive performance, with a Mean Absolute Error (MAE) of 0.080, a Root Mean Squared Error (RMSE) of 0.198, and an R-squared (R2) of 0.997. The integration of the rGO/FeCo-MOF/GCE sensor with the inverse Extra Trees model enables accurate quantification of Cd2+ and Pb2+ in real environmental water samples with relative errors below 8%. These results demonstrate the robustness and potential of the sensor for practical environmental monitoring applications.

