Modified iterative wavelets for background removal in laser-induced breakdown spectroscopy based on fiber laser ablation

Abstract

With the advantages of long-time operation and stable power, fiber lasers have been widely used in the industrial field and show great potential in laser-induced breakdown spectroscopy (LIBS). However, the bremsstrahlung background in LIBS based on fiber laser ablation (FL-LIBS) is much stronger than that in conventional LIBS. In this work, a modified method for removing the background in FL-LIBS was proposed. The spectral ranges were determined from the elemental contents in the reference sample for calibration and the simulated spectral curve of each element. The combinations of wavelet transform coefficients and the iterations were also optimized. The efficiency of the optimization process was improved by 2 orders of magnitude. The R² factors of the aluminum alloy and the micro-alloy steel were increased from 0.945 and 0.951 to 0.992 and 0.986, respectively. And the root mean square error of cross validation (RMSECV) of the aluminum alloy and the micro-alloy steel was decreased from 0.462 wt% and 1.66 wt% to 0.214 wt% and 1.31 wt%, respectively. Moreover, two conventional methods for background removal were used to compare with the proposed method, demonstrating that the proposed method had the highest accuracy. This work provides a convenient approach to realize accurate background removal and is expected to promote the development of FL-LIBS.