To address the problem of the self-absorption of laser-induced breakdown spectroscopy affecting the quantitative detection accuracy of bauxite, a self-absorption correction method based on the plasma electron temperature and electron density is proposed, and a double self-absorption correction model is established to realize the self-absorption correction of the emission spectra of the main elements in bauxite.

A laser-induced breakdown spectroscopy (LIBS) diagram is shown in Fig. 1. LIBS is used to excite the spectrum of bauxite elements. Before data acquisition, the surface of the sample is cleaned using a laser, and the cleaning number is set to five. The ablation acquisition spectra are obtained at nine different positions arranged in a matrix form on the bauxite sample. After five cleaning cycles, ablation is performed at each position. After excluding the abnormal data, the average value of the remaining data is used as the final spectral data value of the sample. The first correction is completed using the internal reference lines and electron temperature, and the second correction is completed using the spectral line broadening theory and electron density.

The plasma emission spectrum of the bauxite sample covers the region from 190 nm to 980 nm (Fig.2). A Boltzmann curve is drawn using the spectral intensity and parameters of each element to obtain the plasma electron temperature. The self-absorption coefficient of the spectral line is then calculated. Boltzmann diagrams of Al, Si, Fe, and Ti before correction are shown in Fig 3. Al I 396.16 nm, Fe I 396.119 nm, Si I 288.158 nm, and Ti I 498.173 nm are used as the internal reference lines to correct the spectral lines. The Boltzmann diagram after the first correction is shown in Fig.4. Taking the corrected spectral peak as the highest point, the actual spectral line broadening in the case of self-absorption can be obtained via Lorentz fitting. At the same time, the electron density is calculated by using the Hα line, and the Hα line spectrum of bauxite is shown in Fig.5. The self-absorption coefficient is calculated using the electron density and actual spectral line broadening, and the spectral lines are corrected for a second time. The Boltzmann diagram is drawn using the intensity after the second correction. The Boltzmann diagram after the second correction is shown in Fig.6. The quantitative results for the main elements in the bauxite are listed in Table 2. As shown in Fig.6, after the second correction, the Boltzmann fitting coefficient of each element in the sample is significantly improved, and most of the data points are distributed on the fitting line. As shown in Table 2, after the first correction, the accuracies for the mass fractions of Al, Si, Fe, and Ti increase by 5.21%, 5.94%, 11.28%, and 5.62%, respectively. Simultaneously, the accuracies of the Al, Si, Fe, and Ti mass fractions after the second correction are further improved by 3.19%, 4.26%, 4.49%, and 3.37%, respectively.

In this study, a self-absorption correction method for bauxite based on the plasma electron temperature and electron density is studied, and the accuracy of LIBS detection is improved. The LIBS self-absorption phenomenon of Al, Si, Fe, and Ti in the bauxite samples is analyzed, and a correction model is established. The experimental results show that the Boltzmann fitting coefficient of each element in the sample significantly improves after two corrections. This analytical method can improve the quantitative detection accuracy of CF-LIBS to a certain extent without using standard samples.