Material Classification Map Using Dual-Energy Method at Low-energy X-ray Spectrum: An Experimental and Monte-Carlo Simulation Study
Mahdi Ghafarzadeh,Mostafa Kabir,Mohammad Tavakoli Kejani
DOI: https://doi.org/10.1115/1.4065385
2024-04-23
Abstract:Abstract The objective of this article is to enhance a method for effectively discriminating light metallic materials from heavy ones at low X-ray energies. In this research, Monte Carlo simulations are employed to investigate the influential factors affecting material discrimination. Initially, for result validation, the experimental setup is fully simulated based on Monte Carlo method. The X-ray spectrum of 160 keV is simulated and then it is registered after interacting with step wedges made of iron, aluminum, graphite, and ABS at specific thicknesses, capturing the radiation flux at each step. The results are compared with the experimental findings obtained from a dual-layer detector, demonstrating excellent agreement. In practice, the dual-layer detector comprises a low-energy GOS detector, a copper filter, and a high-energy CsI(Tl) detector. The energy spectra of the registered X-rays on each layer of detectors are obtained using the Monte Carlo method. Materials with low, medium, and high atomic numbers are chosen for analysis. These materials are categorized into three groups: organic materials (comprising both light and heavy organic and biological substances), light metals, and heavy metals. Discrimination between materials is achieved independently of their thickness by utilizing a Material Classification Map (MCM) derived from a graph depicting the transmission ratio of low-energy X-ray photons versus the linear attenuation coefficient ratio for various materials with different atomic numbers. The results have been successfully validated through testing with various materials and thicknesses using both the experimental setup and Monte Carlo simulations.