Blank sample denoising algorithm (BSDA): An effective spectral noise reduction in water sample LIBS detection.

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Environmental Science

Background:

• Laser-induced breakdown spectroscopy (LIBS) offers advantages for elemental analysis in water.
• Water matrix effects often degrade LIBS spectral signal quality (SNR, stability).
• Weak spectral signals can be mistaken for noise, limiting detection capabilities.

Purpose of the Study:

• To develop a robust denoising algorithm for LIBS spectral data from water samples.
• To improve signal-to-noise ratio and detection limits for elemental analysis in water.
• To establish a practical data processing workflow for enhanced LIBS analysis.

Main Methods:

• Construction of a blank sample spectral database using deionized water.
• Implementation of a data processing workflow including blank sample screening, internal standard correction, blank sample correction, and spectral smoothing.
• Application of the algorithm to LIBS spectral data of Na, Mg, Ca, K, Sr, and Li in marine water samples.

Main Results:

• Significant improvement in spectral quality, with SNR and detection limits enhanced by at least one order of magnitude.
• Up to 36-fold increase in signal intensity for Li and a 25.2-fold decrease in detection limit for K.
• Effective extraction of previously unobservable weak spectral peaks, demonstrating enhanced sensitivity.

Conclusions:

• The proposed denoising algorithm and data processing workflow effectively enhance LIBS analysis of water samples.
• The method is simple, practical, and universally applicable to various water LIBS detection technologies.
• This approach shows significant potential for improving elemental analysis in diverse aquatic environments.