Analysis of electron energy-loss spectra and images

Several methods are proposed for the elemental analysis of biological material with the use of electron energy-loss spectroscopy (EELS) and electron spectroscopic imaging (ESI) in a Zeiss EM902. For spectra, the Simplex-method was tested and compared to Egerton's 2- area method, the Steepest descent method and the Loglog method. The Simplex method and the Loglog method performed best. The Simplex method was then used for the analysis of biological specimens containing calcium and iron. The results of the analysis of a test-spectrum, an iron-spectrum and a calcium-spectrum taken with various widths of the r -region were compared. It was observed that calcium is a hard element for EELS-analysis because of the carbon K-edge just in front of its ionization-edge. Ca- and Fe-Bio-standards, which were co-embedded with the specimens, were used to obtain elemental concentrations in the specimen. In combination with the Kontron IBAS 2000, morphometrical image analysis was performed on cells and their reaction products. The influence of several acquisition factors was analyzed (objective-lens diaphragm diameter, image mode, number of integrations during image acquisition). Segmentation is applied using the cross-overs of the frrst derivative of the greyvalue distribution histogram of the image. Co-localization of several chemical elements could be demonstrated by comparing the net-intensity distribution images of the respective elements. In images, several methods were tested for the separation of the elementinformation from the continuum. The power-law method gives the best results. The influence of the width of the integration-region on the obtained relative concentration was tested. Since the power-law method did not overcome all problems encountered with image-analysis (Calcium), Multivariate statistical analysis (MSA) was exanrined as an alternative. The latter method performed very well and was able to analyze calcium images. MSA also reduces the influence of other sources of information in the images such as shading and radiation damage