Improved Simultaneous Quantification Of Trace Heavy Metals With Total-reflection X-ray Fluorescence Analysis

Since simultaneous quantitative analysis of trace heavy metals is urgently required by dynamic environmental monitoring,an improved simultaneous quantification method is studied for trace heavy metal monitoring based on total-reflection X-ray fluorescence(TXRF)analysis.Investigating related technologies,developing the TXRF instrument,analyzing influence factors to measurement performance,designing the quantitative analysis method,and evaluating the overall performance are carried out in order during the research.The main achievements of this research are as follows.(1)An integrated TXRF analysis system is designed and constructed for environmental trace heavy metal monitoring applications for the first time in China,where instrument architecture,core components,mechanical structure,control circuits and application software are developed.Its element measuring range is enlarged by utilizing optional W-anode and Ag-anode X-ray tube.A computer-assisted glancing angle controller is introduced to optimize the glancing angle for total reflection of the primary beam.It is demonstrated that the sample carrier can be tilted around a horizontal axis in a stepwise manner at a resolution of 0.03°,energy resolution of TXRF spectrum is better than 145 eV,and simultaneous analysis of K,Ca,Cr,Mn,Fe,Co,Ni,Cu,Zn and Pb can be achieved.(2)Based on the TXRF analysis platform,the basic measurement process of water samples with heavy metals is designed.Then the internal and external factors affecting the measurement performance are analyzed.Spectral interference,as a key factor in TXRF,is studied in detail,of which the sources include multiple elements of samples,Si,Ni and Pb inside the platform,Ar in the atmosphere,scattering of the primary beam,and escape peaks from detector artifacts.Analysis results show that the main problems which should be solved during quantitative analysis are spectral noise,background drift and spectral overlap.(3)An adaptive TXRF spectral preprocessing method adopts an improved wavelet threshold denoising algorithm for spectral smoothing,and adopts an adaptive wavelet analysis algorithm for background correction.Especially,an intelligent multi-element quantification method is proposed to solve the spectral overlap problem.In a bran-new spectral decomposition framework,collected characteristic curves of existing elements in the whole spectral scope are used to approximate measured TXRF spectra.Moreover,considering the benefits of particle swarm optimization(PSO)in solving combinatorial optimization problems,the benefits of radial basis function network(RBFN)in approximating non-linear function and performing fast computation,and the benefits of simulated annealing(SA)in finding the local optimal result,a novel hybrid algorithm,PSO-RBFN-SA,is designed to solve the curve fitting problem.It is demonstrated that the adaptive TXRF spectral preprocessing method obtains high peak-to-background ratio and avoids over-fitting or under-fitting of background curves,while the intelligent multi-element quantification method is inherently scalable and obtains both high measurement accuracy and computational efficiency.(4)Performance evaluation of the TXRF analysis platform and the related quantification method shows that simultaneous quantification of trace heavy metals is achieved.The detection limit of Cr,Mn,Fe,Co,Ni,Cu and Zn is below 0.5 μg/L,while the detection limit of Pb is below 4 μg/L.Testing samples in concentration of 4 mg/L,the measurement repeatability of Cr,Mn,Fe,Co,Ni,Cu and Zn is better than 4%,while the measurement repeatability of Pb is better than 10%.Testing samples in concentration of 1 mg/L or above,the measurement error of Cr,Mn,Fe,Co,Ni,Cu,Zn and Pb is less than 10%.The results of practical industrial wastewater monitoring verify that TXRF quantitative analysis can be used as an alternative reliable method for heavy metal monitoring,and has promising future in diverse applications.