On the Precision of EDS Analysis with Sample Tilted at 70° (original) (raw)

Measurements of relative x-ray line intensities and their application to a single standard procedure for quantitative x-ray microanalysis

Journal of Applied Physics, 2006

The orientation problem in polycrystalline cubic materials has been simplified, using fundamental relationships, so that the determination of a quantitative interrelationship between the various Bragg peak intensities is no longer a formidable task. This is demonstrated with a cubic CuBe Co alloy having a fiber texture, and a conventional focusing diffractometer. Because data are required which extend over a larger range in d spacings, extinction, thermal, and static atomic displacements must be included into the analysis of intensities. The displacement terms and the extinction parameters may be of primary interest or used as a correction. Seventeen diffraction peaks are used in the example. These must be internally consistent with a crystallite orientation function, the cubic symmetry of the sample, extinction effects influencing the two strongest peaks, and attenuation due to atomic displacements. Tabulated coefficients are presented which greatly reduce the task of calculating the orientation function. Acorrection is given for instrumental smearing which should be considered for stronger textures than the intermediate case examined or for intermediate textures and nonfocusing instrumental conditions.

A flexible and accurate quantification algorithm for electron probe X-ray microanalysis based on thin-film element yields

Spectrochimica Acta Part B: Atomic Spectroscopy, 2003

Quantitative analysis by means of electron probe X-ray microanalysis (EPXMA) of low Z materials such as silicate glasses can be hampered by the fact that ice or other contaminants build up on the Si(Li) detector beryllium window or (in the case of a windowless detector) on the Si(Li) crystal itself. These layers act as an additional absorber in front of the detector crystal, decreasing the detection efficiency at low energies (-5 keV). Since the layer thickness gradually changes with time, also the detector efficiency in the low energy region is not constant. Using the normal ZAF approach to quantification of EPXMA data is cumbersome in these conditions, because spectra from reference materials and from unknown samples must be acquired within a fairly short period of time in order to avoid the effect of the change in efficiency. To avoid this problem, an alternative approach to quantification of EPXMA data is proposed, following a philosophy often employed in quantitative analysis of X-ray fluorescence (XRF) and protoninduced X-ray emission (PIXE) data. This approach is based on the (experimental) determination of thin-film element yields, rather than starting from infinitely thick and single element calibration standards. These thin-film sensitivity coefficients can also be interpolated to allow quantification of elements for which no suitable standards are available. The change in detector efficiency can be monitored by collecting an X-ray spectrum of one multi-element glass standard. This information is used to adapt the previously determined thin-film sensitivity coefficients to the actual detector efficiency conditions valid on the day that the experiments were carried out. The main advantage of this method is that spectra collected from the standards and from the unknown samples should not be acquired within a short period of time. This new approach is evaluated for glass and metal matrices and is compared with a standard ZAF method. ᮊ

A profile-fitting procedure for analysis of broadened X-ray diffraction peaks. II. Application and discussion of the methodology

Journal of Applied Crystallography, 1988

A convolutive profile-fitting procedure is described for analysing X-ray diffraction peak profiles broadened by microstructural factors (crystallite size and lattice disorder). The method requires, in a first stage, an accurate determination of the instrumental function, which is subsequently convoluted with a parametric function adjusted to fit the diffraction profile intensities of the specimen investigated. In the calibration of the instrument function throughout the angular range 20-145 ° in 20, 58 peaks of a well crystallized a-quartz specimen are examined. Provision is made to include in the instrument function an exponential function containing an angle-dependent asymmetry parameter. In the present methodology, a pseudo-Voigt function is suggested to obtain the shape factors (integral breadth, peak width at half maximum, Gaussian content) that contain useful information related to the microstructural properties in the frame of the so-called 'simplified methods' of line broadening analysis. Furthermore, if the optimized profile functions in the Fourier analysis of the broadened peaks are used directly, many relationships follow straightforwardly in the real-space domain. In this way it is easy to apply the formalisms currently used to derive physical information (e.g. the Warren & Averbach and Vogel, Haase & Hosemann methods).

Quantitative X-ray energy dispersive analysis with the transmission electron microscope

X-Ray Spectrometry, 1975

A comprehensive study of the interfacing of an X-ray energy dispersive spectrometer to a transmission electron microscope is presented. Optimum detector location, sample tilt, collimation and accelerating voltage are discussed. Quantitative elemental analysis is obtained for thin film specimens using corrections for spectrometer response, absorption and background. Peak spectra are sorted using a method of profile fitting based on nonlinear simplex minimization. Using a simple analysis scheme, the peak data is reduced to elemental composition with an accuracy and precision on the order of 1%. Examples from Fe-Ni and Ho-Co alloys are given.