Detection of multiple metal ions in water with a fluorescence sensor based on carbon quantum dots assisted by stepwise prediction and machine learning

• Published in: Environmental chemistry letters Vol. 20; no. 6; pp. 3415 - 3420
• Main Authors: Liu, Yuying, Chen, Jiao, Xu, Zijun, Liu, Hao, Yuan, Tingting, Wang, Xiyuan, Wei, Jianjie, Shi, Qingdong
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.12.2022; Springer Nature B.V
• Subjects: Accuracy; Analytical Chemistry; Carbon; Deionization; Earth and Environmental Science; Ecotoxicology; Environment; Environmental Chemistry; Fluorescence; Geochemistry; Heavy metals; human health; Ions; Iron; Lakes; Learning algorithms; Machine learning; Metal concentrations; Metal ions; Metals; Original Paper; Pollution; prediction; Quantum dots; Receptors; Sensor arrays; Sensors; Water analysis; Water sampling; Carbon quantum dots; Fluorescence sensor array; Stepwise prediction; Metal ions
• ISSN: 1610-3653; 1610-3661
• DOI: 10.1007/s10311-022-01475-0

Pollution by heavy metals is threatening the environment and human health, yet there is a lack of a rapid methods to detect multiple metal ions. Here, we built a fluorescence sensor array based on carbon quantum dots to detect Cr 6+ , Fe 3+ , Fe 2+ , and Hg 2+ in environmental samples. We added xylenol orange as the receptor to construct the sensor array under pH regulation. We also designed a SX-model by combining stepwise prediction and machine learning to assist the fluorescence sensor array in detecting single and mixed heavy metal ions in deionized water and real samples. Results show that the sensor array detects four heavy metal ions within a concentration range of 1–50 μM with an accuracy of 95%, and the sensor identifies binary mixed samples with an accuracy of 95%. In addition, metal ions occurring in 144 lake water samples were discriminated with 100% accuracy. Overall, the SX-model-assisted fluorescence sensor array is an efficient method for detecting heavy metal ions in environmental samples.