Machine learning-assisted nitrogen-doped carbon dots for Fe3+ detection in aqueous environments.

The concentration of iron ions is a crucial indicator for assessing water quality. In this study, nitrogen-doped carbon dots (NCDs) were synthesized using a microwave-assisted method with citric acid and urea as precursors, thereby establishing a fluorescence sensing platform for the detection of alkaline pH and Fe3+. During Fe3+ detection, the fluorescence intensity of NCDs was specifically quenched as the concentration of Fe3+ increased, demonstrating good linearity across the ranges of 1-10 µM and 10-100 µM, with a detection limit of 0.55 µM. By integrating smartphone-based image analysis, visual semi-quantitative detection of alkaline pH and Fe3+ was achieved. To enhance prediction accuracy across a broad concentration range, a machine learning model was introduced to develop a high-precision quantitative analysis method for Fe3+. The spiked recovery rates in actual water samples ranged from 99.26% to 101.14%, with relative standard deviations below 3%. This platform combines fluorescence sensing, smartphone imaging, and machine learning technologies, offering the advantages of simple operation and low cost, thus providing a novel strategy for the on-site rapid detection of Fe3+ in water environments.