Multivariate calibration analysis of colorimetric mercury sensing using a molecular probe

Area of Science:

• Analytical Chemistry
• Environmental Science
• Materials Science

Background:

• Sensor selectivity is a major challenge, especially for chemical sensors facing interference.
• Multivariate analytical models offer a solution for quantifying analytes amidst complex mixtures.
• Ruthenium complexes show potential as chemical probes for ion detection.

Purpose of the Study:

• To develop a selective sensor for mercury ions in aqueous solutions.
• To assess the sensor's performance in the presence of common interfering metal ions (Cd2+, Pb2+, Cu2+, Zn2+).
• To optimize data analysis using multivariate methods and feature selection techniques.

Main Methods:

• Utilized a ruthenium bis-thiocyanate complex as a chemical probe for mercury ion detection.
• Employed partial least squares (PLS) for multivariate data analysis.
• Applied genetic algorithm and statistical feature selection (SFS) for data pre-treatment and refinement.

Main Results:

• Successfully detected mercury ions in aqueous solutions within the ppm range (0-2 mg/L).
• Demonstrated effective analysis in solutions containing high concentrations (19.5 mg/L) of interfering ions.
• Reduced the root mean square error of prediction from 10.22% to 6.27% using SFS.

Conclusions:

• The developed sensor system, combined with multivariate analysis and SFS, provides a selective and accurate method for mercury ion determination.
• Statistical feature selection significantly enhances sensor performance by minimizing prediction errors.
• This approach offers a robust solution for monitoring mercury contamination in environmental samples.