Comparison of Microanalysis Using Energy-Dispersive X-Ray Spectroscopy and Electron Energy Loss Spectroscopy

Paterson, James H (1988) Comparison of Microanalysis Using Energy-Dispersive X-Ray Spectroscopy and Electron Energy Loss Spectroscopy. PhD thesis, University of Glasgow.

Full text available as:
Download (10MB) | Preview


The work presented in this thesis is concerned with the development and comparison of two techniques for microanalysis in the electron microscope. These are energy-dispersive x-ray spectroscopy (EDX) and electron energy loss spectroscopy (EELS). Particular emphasis has been placed on light element analysis, as these elements have become accessible to EDX analysis with the recent advent of windowless x-ray detectors. The interest here is confined to analysis without recourse to standard specimens of known composition. Standardless analysis requires a theoretical knowedge of the processes which give rise to the various features in both types of spectrum. Chapter 2 outlines the formalism of the relevant basic theory, and describes how expressions are obtained for the cross sections relevant to inelastic scattering of electrons by interactions with inner-shell electrons in the specimen, and to the x-ray production which is associated with such scattering. The signals in EELS and EDX spectra which arise due to inner shell scattering occur at energy losses and photon energies respectively which are characteristic of the atoms in the specimen. Other processes, which contribute to non-characteristic backgrounds in both cases, are described briefly. The work for this thesis was carried out using three microscopes: two similar scanning transmission electron microscopes (STEMs), one of which was equipped with a windowless x-ray detector, and one transmission electron microscope (TEN). Chapter 3 gives a description of one of the STEMs, and goes on to outline the differences between the two STEMs. The TEM is then briefly described. The remainder of the chapter discusses the detectors and spectral acquisition systems fitted to each microscope. The results obtained on EDX are presented in chapters 4, 5 and 6. Chapter 4 starts by discussing general EDX analysis techniques, and then goes on to detail a series on measurements of the ratios of characteristic signals to the local background in the EDX spectrum. The results obtained using all three microscopes are then used as the basis of a parameterisation of the ionisation cross section for the K-shell. The parameterisation allows the prediction of this quantity within an accuracy of ~20% over a range of elements with 14 and a range of accelerating voltages from 80keV to 200k. eV. Chapter 5 details the investigation of L-shell ionisation cross sections. This involved measuring the ratio of the K-shell/L-shell cross sections over a wide range of elements, and using the results of the previous chapter to deduce the L-shell cross sections. These measurements required the detector efficiency to be carefully considered over the entire energy range of an EDX spectrum. The accuracy with which the L-shell cross sections could be determined was limited by uncertainties in the values in the literature for the relevant fluorescence yields. Nevertheless, the results generally suggested that the model used for the K-shell may be applied also to the L-shell. Chapter 6 gives the results of the analyses of a number of compounds of light elements, and shows that many difficulties exist in the determination of quantitative information on these elemnts by EDX. EELS analysis procedures are discussed in chapter 7. Conventional procedures, and an alternative technique proposed by Steele et al. (1985), are detailed. The latter approach involves the inclusion in the function which is fitted to the spectrum of a scaled theoretical cross section. Fitting may then be carried out both before and after the characteristic edge onset. This removes the need for the background to be extrapolated beneath the edge, as conventional background fitting requires. Results of the application of the new procedure are given in chapter 8. Firstly, it is used to re-analyse EELS data which had previously been analysed conventionally. These data were recorded simultaneously with equivalent EDX data. The results showed that the use of the new technique leads to some improvement in the correlation between concentration ratios determined using EELS and EDX. A second analysis, of TiB2 end CrB2 failed to produce any conclusive results. Finally, chapter 9 discusses the implications of the results obtained in this work, and suggests some ways in which the accuracy of each of the two techniques might be improved.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Analytical chemistry
Date of Award: 1988
Depositing User: Enlighten Team
Unique ID: glathesis:1988-77624
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 14 Jan 2020 11:53
Last Modified: 14 Jan 2020 11:53

Actions (login required)

View Item View Item


Downloads per month over past year