Evaluation of the Calibration-Free and Multivariate Method for Quantitative Analysis in Laser-Induced Breakdown Spectroscopy

The calibration-free laser-induced breakdown spectroscopy (CF–LIBS) method is used to obtain the concentration of the constituents of samples because it overcomes the limitation of matrix-matched standards in the calibration curve method of quantification. However, there are often doubts that remain about the efficiency of the CF–LIBS method. Hence, in the present work, different certified reference materials (CRMs) of plants and soil were employed to check the capabilities of the CF–LIBS method. If the emission lines of an element are missing in the LIBS spectra, its contribution in the CF–LIBS result will be missing as well, which leads to incorrect quantification. Therefore, in order to overcome this problem in CF–LIBS, instead of only determining the elemental concentrations, an additional step to calculate the concentration ratio of all elements with respect to the concentration of a major element was added. The calculated concentration ratios for different elements are more accurate than the elemental concentration obtained by CF–LIBS. Along with the CF–LIBS method, the partial least square regression (PLSR) approach was also applied for the prediction of the concentration.