Quantitative determination of the multicomponents with overlapping ultraviolet spectra using wavelet-packed transform...

Area of Science:

• Analytical Chemistry
• Spectrophotometry
• Chemometrics

Background:

• Simultaneous determination of multiple analytes in complex mixtures is challenging.
• Spectrophotometric methods often suffer from overlapping signals and noise.
• Existing methods like Partial Least Squares (PLS) have limitations in noise handling and prediction accuracy.

Purpose of the Study:

• To develop and validate a novel Wavelet Packet Transform-based Partial Least Squares (WPT-PLS) method.
• To achieve accurate simultaneous spectrophotometric determination of alpha-naphthylamine, p-nitroaniline, and benzidine.
• To enhance noise removal and prediction capabilities compared to traditional PLS.

Main Methods:

• Utilizing Wavelet Packet Transform (WPT) for signal decomposition and noise reduction.
• Employing the best-basis algorithm and thresholding for optimal noise removal.
• Applying Partial Least Squares (PLS) regression incorporating both spectral and concentration information.
• Optimizing WPT parameters: Db16 wavelet function and decomposition level 3.

Main Results:

• The WPT-PLS method demonstrated effective noise suppression and signal separation.
• Optimized WPT-PLS achieved a Relative Standard Error of Prediction (RSEP) of 2.23%.
• Standard PLS resulted in a higher RSEP of 2.71% for the same analytes.
• Experimental validation confirmed the method's success and superiority over PLS.

Conclusions:

• The WPT-PLS method is a powerful and effective tool for simultaneous spectrophotometric analysis.
• WPT significantly improves the performance of PLS by enhancing noise handling and prediction accuracy.
• This novel approach offers a promising alternative for complex mixture analysis in analytical chemistry.