Study on LOD of Trace Elements by XRF Analysis Using BP & Adaboost and PLS Methods

Abstract:

Poisonous elements such as Cd, Hg, Pb, As, Zn, Cr, Ni, Cu etc. are commonly observed in polluted soil and hard to be removed by soil microbes. It is of significant importance to identify these poisonous elements in-situ and accurately both in qualitative and quantitative sense. In order to determine the Limit of Detection (LOD) for trace elements (e.g. Cadmium) in polluted soil samples based on Energy Dispersion X-ray fluorescence (ED-XRF) spectroscopy, approximately 60 national standard soil samples were collected and measured by an XRF equipment. The authors firstly utilize the Method Detection Limit (MDL) algorithm to calculate the LOD of trace elements, and then develop a new model called Back Propagation Adaboost (BP & Adaboost) classification to determine the LOD based on a presumed tolerance error (e.g. 5%). Furthermore, the Multivariate- Partial Linear Squares Regression (M-PLSR) method is applied to regress the data and validate the LOD values. In this paper, the authors make a detailed comparison between the BP algorithm and the BP & Adaboost classification algorithm under different presumed detection limits, and it is found that the detection results achieved the best qualitative prediction of Cd element (i.e. whether it exists in soil) based on the BP & Adaboost algorithm. The experimental results indicate that the BP & Adaboost algorithm is the most effective method to determine and decrease the LOD of trace element (such as Cd) in soil. The advantages are: It combines the classification effects of several weak classifiers, and determines that the LOD of element Cd is 0.5mg/kg with prediction error rate of 5%. Compared with the traditional methods like MDL, it is proved that the BP & Adaboost algorithm is appropriate to be used in the terms of prediction accuracy. It is recommended that the BP & Adaboost classification method shall be used for material analysis on XRF spectroscopy.

• Periodical: Advances in Science and Technology (Volume 105)
• Pages: 99-109
• DOI: https://doi.org/10.4028/www.scientific.net/AST.105.99
• Online since: April 2021
• Authors: Wan Qi Yang, Xin Lu, Fu Sheng Li*, Yan Chun Zhao
• Keywords: Adaboost, Back Propagation, Limit of Detection, Material Analysis, X-Ray Fluorescence

References

[1] M. García-Carmona, A. Romero-Freire, M. Sierra Aragón, et al, Evaluation of Remediation Techniques in Soils Affected by Residual Contamination with Heavy Metals and Arsenic, Journal of Environmental Management. 191 (2017) 228-236.

DOI: 10.1016/j.jenvman.2016.12.041

[2] A.G.F. Boim, L.C.A. Melo, F.N. Moreno, et al, Bioconcentration factors and the risk concentrations of potentially toxic elements in garden soils, Journal of Environmental Management. 170(2016) 21-27.

DOI: 10.1016/j.jenvman.2016.01.006

[3] Q.L. Yang, Z.Y. Li, X.N. Lu, et al, A Review of Soil Heavy Metal Pollution from Industrial and Agricultural Regions in China: Pollution and Risk Assessment, The Science of the Total Environment. 642(2018) 690-700.

DOI: 10.1016/j.scitotenv.2018.06.068

[4] X.L. Jia, B.F. Hu, B. Marchant, et al, A Methodological Framework for Identifying Potential Sources of Soil Heavy Metal Pollution Based on Machine Learning: A Case Study in the Yangtze Delta, China. Environmental Pollution. 250(2019) 601-609.

DOI: 10.1016/j.envpol.2019.04.047

[5] Z.H. Pan, K.Y. Wang, L. Wang, et al, Risk Assessment System Based on WebGIS for Heavy Metal Pollution in Farmland Soils in China, Sustainability (Basel, Switzerland). 9(2017) 1846.

DOI: 10.3390/su9101846

[6] H.F. Wang, Q.M. Wu, W.Y. Hu, et al, Using Multi-medium Factors Analysis to Assess Heavy Metal Health Risks along the Yangtze River in Nanjing, Southeast China. Environmental Pollution. 243(2018) 1047-1056.

DOI: 10.1016/j.envpol.2018.09.036

[7] V. Rackwitz, M. Ostermann, U. Panne, et al, Performance of μ-XRF with SEM/EDS for Trace Analysis on the Example of RoHS Relevant Elements-Measurement, Optimisation and Prediction of the Detection Limits, JAAS (Journal of Analytical Atomic Spectrometry). 28(2013) 1466-1475.

DOI: 10.1039/c3ja50064k

[8] S. Ershadi, A. Shayanfar, Are LOD and LOQ Reliable Parameters for Sensitivity Evaluation of Spectroscopic Methods?, Journal of AOAC International. 101(2018) 1212-1213.

DOI: 10.5740/jaoacint.17-0363

[9] C. Mcgladdery, D.C. Weindorf, S. Chakraborty, et al, Elemental assessment of vegetation via portable X-ray fluorescence (PXRF) spectrometry. Journal of Environmental Management. 210(2018) 210-225.

DOI: 10.1016/j.jenvman.2018.01.003

[10] S. Carter, A.S. Fisher, M.W. Hinds, et al, Atomic spectrometry update. Review of advances in the analysis of metals, chemicals and materials. Journal of Analytical Atomic Spectrometry. 28(2013) 1814-1869.

DOI: 10.1039/c3ja90051g

[11] K. Nakano, N. Chihiro, O. Kazunori, et al, Depth Elemental Imaging of Forensic Samples by Confocal Micro-XRF Method, Analytical Chemistry. 83(2011) 3477-3483.

DOI: 10.1021/ac1033177

[12] L. Herreros-Chavez, M.L. Cervera, and A. Morales-Rubio, Direct Determination by Portable ED-XRF of Mineral Profile in Cocoa Powder Samples, Food Chemistry. 278(2013) 373-379.

DOI: 10.1016/j.foodchem.2018.11.065

[13] X.P. Wang, F. Zhang, J.L. Ding, et al, Estimation of soil salt content (SSC) in the Ebinur Lake Wetland National Nature Reserve (ELWNNR), Northwest China, based on a Bootstrap-BP neural network model and optimal spectral indices. The Science of the total environment. 615(2018) 918-930.

DOI: 10.1016/j.scitotenv.2017.10.025

[14] N. Loris, B. Sheryl, L. Alessandra, Double committee adaboost. Journal of King Saud- University-Science. 25(2013) 29-37.

[15] S. Al‐Shidhani, N.U. Rehman, F. Mabood, et al, Quantification of Incensole in Three Boswellia Species by NIR Spectroscopy Coupled with PLSR and Cross‐Validation by HPLC, Phytochemical Analysis. 29(2018) 300-307.

DOI: 10.1002/pca.2743

[16] G. Guven, H. Samkar, Examination of Dimension Reduction Performances of PLSR and PCR Techniques in Data with Multicollinearity. Iranian Journal of Science and Technology, Transactions A: Science. 43(2019) 969-978.

DOI: 10.1007/s40995-018-0565-1

[17] W.D. Li, X. Yang, H. Li, et al, Hybrid Forecasting Approach Based on GRNN Neural Network and SVR Machine for Electricity Demand Forecasting, Energies. 10(2017) 44.

DOI: 10.3390/en10010044

[18] C. Zhen, K. Dohyeon, K. Hee-joung, Brain image segmentation based on improved BP-adaboost neural network. Society of Photo-Optical Instrumentation Engineers(SPIE). 111050(2019) 110500Q–110500Q-7.