Laser-induced breakdown spectroscopy（LIBS） is a versatile tool for both qualitative and quantitative analysis.In this paper,LIBS combined with principal component analysis（PCA） and support vector machine（SVM） is applied to rock analysis.Fourteen emission lines including Fe,Mg,Ca,Al,Si,and Ti are selected as analysis lines.A good accuracy（91.38% for the real rock） is achieved by using SVM to analyze the spectroscopic peak area data which are processed by PCA.It can not only reduce the noise and dimensionality which contributes to improving the efficiency of the program,but also solve the problem of linear inseparability by combining PCA and SVM.By this method,the ability of LIBS to classify rock is validated.
Laser-Induced Breakdown Spectroscopy （LIBS） has been demonstrated to be an effective method for slag analysis. In order to better clarify the nature of the plasma generated from a slag sample, an Nd：YAG pulse laser at 1064 nm wavelength was used to ablate the slag sample in air. The temporal and spatial evolutions of plasma parameters, including emission intensity, electronic density and plasma temperature, have been studied. It is shown that the electron density and plasma temperature drop off rapidly with the delay time as a result of plasma expansion and cooling. It has been found that the electron density of the whole plasma is close to that of the center regions in the plasma. The results of the spatial distributions on the two-dimensional plane have shown that there is a big region with lower electron density values caused by the recombination process in the center of the plasma. The maximum of the plasma temperature takes place at the regions close to the target, and the border of the plasma front-head has higher plasma temperatures than that of the center part.
To reduce the influence of laser-induced breakdown spectroscopy （LIBS） experimental parameter fluctuations to quantitative analysis of slag components, a normalization method using integral intensity of plasma image was proposed and a series of experiments with slag samples were performed. Mg II 279.55 nm, Ca II 396.85 and Ca I 422.67 nm were selected as analytical lines, and analytical curves of reference mass fractions versus spectral line intensities were established. With the incre- ment of set threshold for edge extraction of plasma image, the determination coefficients and relative standard deviations of analytical curves were improved gradually and reached the optimmn values when the threshold was equal to 10 000. Comparing with the results without normalization and normalized by whole spectrum area, the relativity between spectral line intensity and mass fraction can be enhanced etfieiently after normalized by integral intensity of plasma image. The verification experiments with Ti alloy samples further confirmed the conclusions mentioned above.
