On the Rapid Generation of Complete XRF Spectra for Material Analysis from Fundamental Parameters

Abstract

X-ray Fluorescence (XRF) analysis technology is used widely to detect and measure elemental compositions of target samples. The MCNP code developed by LANL can be utilized to simulate and generate the XRF spectrum of any sample with various elemental compositions. However, one shortcoming of MCNP code is that it takes quite a lot of time (in hours or longer) to generate one XRF spectrum with reasonable statistical precision; the other shortcoming is that MCNP code cannot produce L shell spectrum accurately. In this paper, a new computation model based on the Sherman equation (i.e., Fundamental Parameters, FP) is proposed to overcome the drawbacks of the MCNP code. The most important feature of this model is to achieve a full and accurate generation of spectral information of each element in a target material very rapidly (in seconds or less), including both K and L shell spectral peaks. Furtherly, it is demonstrated that the simulated data by this new mode match the experimental data very well. It proves that the proposed model can be a better alternative of MCNP code in the application of generation the XRF spectra of many materials, in terms of speed and accuracy. The proposed model can perform the simulation of XRF spectra in situ both fast and accurately, which is essential for real-time calculation of chemical composition by use of X-ray spectrometer, especially for those trace elements in target materials.