Related Experiment Video
Updated: Aug 2, 2025
09:30
Bacterial Detection & Identification Using Electrochemical Sensors
Published on: April 23, 2013
28.4K
Tree-based machine learning models assisted fluorescent sensor array for detection of metal ions based on silver
Xihang Chen1, Jinming Xu1, Huangmei Zhou1
1State Key Laboratory of Precision Spectroscopy, East China Normal University, No.500, Dongchuan Rd., Shanghai 200241, China.
Summary
A new fluorescent sensor using silver nanoclusters efficiently detects six heavy metal ions. Machine learning models accurately identified ions in mixtures and river water, offering a promising solution for pollution monitoring.
Area of Science:
- Environmental Chemistry
- Nanotechnology
- Analytical Chemistry
Background:
- Heavy metal pollution poses significant risks to environmental and human health.
- Accurate detection and differentiation of heavy metal ions are crucial for effective remediation.
- Existing methods may lack efficiency or specificity in complex environmental matrices.
Purpose of the Study:
- To develop a novel fluorescent sensor array for detecting and differentiating six specific metal ions.
- To utilize machine learning for analyzing fluorescence variations and identifying metal ions.
- To validate the sensor's performance in complex samples like river water.
Main Methods:
- Preparation of a single silver nanocluster (Ag NCs) probe using poly(methyl vinyl ether-alt-maleic acid) (PMVEM) as a ligand.
- Exploitation of differential fluorescence properties of Ag NCs in water and dimethyl sulfoxide (DMSO).
- Application of tree-based machine learning models (e.g., K-Nearest Neighbors Classifier) for pattern recognition and classification.
- Integration of concentration models within a pipeline optimization framework for quantitative analysis.
Main Results:
- The Ag NCs probe exhibited characteristic fluorescence variations upon interaction with different metal ions.
- The K-Nearest Neighbors Classifier combined with linear discriminants achieved high classification accuracy.
- Accurate identification of metal ions was demonstrated in binary mixtures and real river water samples.
- Six unique concentration regression models were developed for quantitative analysis of each metal ion.
Conclusions:
- The developed fluorescent sensor array provides an effective platform for heavy metal ion detection.
- The integration of machine learning enhances the accuracy and reliability of the analytical framework.
- This approach offers a promising and sensitive method for monitoring heavy metal pollution in environmental samples.
Keywords:
Fluorescent sensor arrayHeavy metal ion detectionMachine learningSilver nanoclusterTree-based pipeline optimization tool
