T R A C K       P A P E R
ISSN:2394-3661 | Crossref DOI | SJIF: 5.138 | PIF: 3.854

International Journal of Engineering and Applied Sciences

(An ISO 9001:2008 Certified Online and Print Journal)

Element Analysis with Fundamental Parameters using an XRF Spectrum Analysis MATLAB Algorithm

( Volume 3 Issue 3,March 2016 ) OPEN ACCESS

Girish Balasubramanian, Senthil Arumugam Muthukumaraswamy


This paper proposes a MATLAB algorithm which can perform X-ray Fluorescence (XRF) spectrum analysis using a type of calibration method known as fundamental parameters (FP). This calibration method is unique because it uses the theoretical relationship between measured X-ray intensities and concentrations of the elements in a given sample. This is different from an empirical approach which uses known sample spectra and known composition and results to obtain results. The fundamental parameter approach is more complex compared to the empirical approach as there are different factors involved in the calculation of these parameters. In this paper, An XRF spectrometer (X-MET8000 by Oxford Instruments) was used for performing experimental gathering of sample spectra. Sample spectra for various pure elements and alloys were obtained and used as an input to the proposed MATLAB algorithm. The MATLAB algorithm uses the FP approach to qualitatively and quantitatively identify the elements present in the tested spectra. The proposed algorithm was able to successfully identify the elements present in the sample as well the elemental composition by means of the FP approach. This algorithm was also compared with results from another similar algorithm that was used to identify alloys by empirical means. The results obtained which include the elemental composition and the present elements is then compared with standardised known test samples of various pure elements and alloys for verification as a measure of accuracy and validity of the MATLAB algorithm which was found to be accurate to within 0.6 percentage points. The working speed of the MATLAB algorithm was also tested experimentally and was observed to be able to process 11-82 samples per second depending on the mix of the sample and type of analysis.

Paper Statistics:

Total View : 492 | Downloads : 483 | Page No: 4-10 |

Cite this Article:
Click here to get all Styles of Citation using DOI of the article.