Accurate quantitative determination of heavy metals in oily soil by laser induced breakdown spectroscopy (LIBS) combined with interval partial least squares (IPLS)

Abstract

Oily soil is harmful to the environment, and heavy metals therein may cause hazards to human health. A rapid detection method for heavy metals in oily soil is needed to provide accurate data support for in situ soil pollution assessment and restoration. In the present study, a laser-induced breakdown spectroscopy (LIBS) technique combined with interval partial least squares (IPLS) was proposed to determine Cu, Zn, Cr and Ni in oily soil samples. In order to obtain a good IPLS model, the full spectrum was divided into 10, 20, 30, 40, 50, 60, 70, 80 and 90 subintervals for IPLS model optimization. The results show that the 30th subinterval from an 80-interval case has the lowest RMSE and the best R². Compared to the PLS model using the full spectrum, the IPLS model with the 30th subinterval from the 80-interval case as the input variable has higher R²s and lower RMSEs for Cu, Zn, Cr and Ni. The R² was improved from 0.96 to 0.99, and the RMSE was reduced from 0.03 to 0.01. The calculation speed was increased by about 5 times. The LIBS technique coupled with the IPLS method can quickly and effectively realize the quantitative analysis of heavy metals in oily soil and can even be used for the analysis and control of soil pollution.