Grow p-type MoS2 on FeNC for CO2 Sensing in Complex Environments with Intelligent Recognition.

A wealth of theoretical studies demonstrates p-type MoS2 (p-MoS2) as a promising candidate for carbon dioxide (CO2) detection at room temperature. Its applications are retarded by issues associated with its practical chemical synthesis and sensing selectivity. Herein, a chemically tunable strategy is established for in situ growth of p-MoS2 with controlled thickness and n-/p-type transition on N- and Fe-enriched carbon (FeNC) nanosheets. The introduced sulfur vacancies (Svacs) enhance the sensitivity to CO2, and the modulated electron distribution suppresses surface oxygen ionization to improve sensing selectivity. The optimized p-type composites can detect CO2 fluctuation levels as low as 50 ppm at room temperature. Density functional theory (DFT) and grand canonical Monte Carlo (GCMC) simulations clarify the underlying mechanisms. A visualized machine learning (ML) model is developed using a hybrid ML strategy that generates regression surfaces from linear/nonlinear data. Through this model, a single sensor accurately discriminates CO2 from interfering and predicts its concentration and humidity with accuracies exceeding 95%. An intelligent sensing system capable of environmental monitoring and tracking exhaled CO2 is demonstrated. The measured fluctuations strongly correlate with physiological indicators, underscoring their potential for non-invasive health monitoring and medical diagnostics.