Wavelet-based interference correction for laser-induced breakdown spectroscopy

• Published in: Journal of analytical atomic spectrometry 2017-12, Vol.32 (12), p.241-246
• Main Authors: Guo, Y. M, Deng, L. M, Yang, X. Y, Li, J. M, Li, K. H, Zhu, Z. H, Guo, L. B, Li, X. Y, Lu, Y. F, Zeng, X. Y
• Format: Article
• Language: English
• Subjects: Chromium; Error correction; Interference; Laser induced breakdown spectroscopy; Low alloy steels; Manganese; Quantitative analysis; Regression analysis; Regression models; Root-mean-square errors; Scaling factors; Silicon; Spectra; Spectrum analysis; Titanium; Wavelet analysis; Wavelet transforms
• ISSN: 0267-9477; 1364-5544
• DOI: 10.1039/c7ja00204a

To minimize the impact of spectral interference on laser-induced breakdown spectroscopy (LIBS) quantitative analyses, an algorithm based on wavelet transform was developed for simultaneous correction of spectral interference and continuum background. The root-mean-square error of calibration (RMSEC) of the univariate regression model for the element of interest was applied to determine the wavelet function, decomposition level, and scaling factor α . When the interference-free analytical lines of the elements of interest cannot be directly obtained from the measured spectra, they can be extracted from the spectra with the developed method for quantitative analysis. This method was applied for LIBS analyses of chromium (Cr), silicon (Si), titanium (Ti), and manganese (Mn) with continuum backgrounds and spectral interference in low alloy steel samples. The root-mean-square errors of cross-validation (RMSECV) of elements Cr, Si, Ti, and Mn were 0.0295, 0.0140, 0.0183, and 0.0558 wt%, respectively. The results demonstrated that the developed algorithm contributed to accuracy improvement for LIBS quantitative analyses with the presence of spectral interference. A wavelet-based interference correction approach was proposed to improve the accuracy of LIBS quantitative analysis.