Multi-element quantitative analysis of soils by laser induced breakdown spectroscopy (LIBS) coupled with univariate and multivariate regression methods

摘要: It is crucial to make comprehensive assessments of soil under economic and efficiency requirements in order guide the rational use resources, including identification attributes, control pollutants, management nutrients. Laser induced breakdown spectroscopy (LIBS) excellent for such applications due its unique advantages simple preparation, rapid measurement, multiple-element analysis. An analysis Si, Al, Mg, Ca, Na, K, Mn, Ba, Ti, Cr, Cu, Sr P different standard soils using LIBS reported here comparison quantitative results a univariate regression method (calibration curve) two multivariate methods (partial least squares (PLSR) support vector (SVR)). As result, correlation coefficients (R2) Na elements were all greater than 0.90, while calibration curves Ti presented poor linear performances with low R2 values below 0.90. The robustness SVR model was superior PLSR better prediction ability lower relative deviation (RSD) both training data test data, whilst opposite observed predicted set used as external verification. errors given by except K those SVR. Al within 10%, other between 10% 20%, 45% Cu. therefore meaningful propose reference among regression, nonlinear multi-element samples complex matrix conditions.