Determination Of Soil Heavy Metals Using X-ray Fluorescence

X-ray fluorescence spectrum (XRF) is a new type of fast quantitative analysis technology,and has been widely applied. Building the systemic analysis method of X-ray fluorescencespectrometric on farmland soil heavy metal will not only benefit the data collection andpreparation, but also on the promotion of spectrum technology in agricultural information for thefurther application in the field. This paper collected Beijing soil, Xinjiang clay and Heilongjiangsoil as the samples for detection, and studied the XRF spectrum of testing technology involvedthe optimal experiment condition, the chart processing and the method of mathematical model.The main work and progress are as follows:(1) Through the summary of the development situation on domestic and international X-rayfluorescence spectrum, the paper brief introduced the X-ray fluorescence spectrum researchdevelopment, the present situation. Through the further introduction on the principle of work ofX-ray fluorescence spectrum testing, this paper also show the methods of diagram processing,abnormal sample eliminating in the quantitative modeling process of XRF spectrum and themain parameter of evaluation indexes and validation process.(2) This paper studied the affect on the results of analysis by soil properties such as soiltypes, soil sample size, water content in the laboratory process. Through the single factorexperiment such as soil particle size, soil water content, pressure and testing time, this papershow that the soil samples treatment before testing could effectively reduce part matrix effect,improve the precision of soil testing instrument of heavy metals. And the optimum experimentalconditions for soil sample is dry moisture content (<5%), after a100mesh fineness, after0.3MPpressure production, testing time for120S.(3) The level of heavy metals, such as Pb, Zn, Cu, Cr levels in the collection of the Beijingbrown soil, Xinjiang clay soil and Heilongjiang soil samples were assayed by the chemicalmethod, and then build up the PLS model by sample X-ray fluorescence spectrum (XRF)analysis. Through the XRF spectra analysis of Pb, Zn, Cu, Cr, this paper select the optimalspectrum line interval for mining and determine the number of points in point. Compared withthe original spectrum diagram, baseline calibration methods comparing pretreatment effect, this paper chose the best preconditioning method, and build up the partial least squares (PLS)method on the different soil types of heavy metals in XRF quantitative analysis model. Theresults show that the X-ray fluorescence spectrum analysis method combining with the PLSmodel can better predict the Pb, Zn, Cu, Cr content of different types of soil.