Multivariate statistical analysis of spectrum lines and images

Recent developments in instrumentation and computing power have greatly improved the potential for quantitative imaging and analysis. A number of techniques are being explored for the purpose of analyzing these large data sets. Multivariate statistical analysis (MSA) provides a method for analyzing the raw data set as a whole. The information that can be extracted by MSA from a series of spectra is illustrated by an application to a TEM spectrum-line acquired with a Gatan Imaging Filter (GIF) at the Co-L edge for a phase boundary between the periclase- (CoO) and spinel- (Co{sub 3}O{sub 4}) structured phases of cobalt oxide. A series of 64 spectra, each of 512 channels, has been analyzed with MSA. The following MSA information is given: (1) logarithmic plot of the information content of the MSA-identified principle components of the series of spectra; (2) the spectrum line of the Co-L edge acquired with the GIF; (3) the first component of the variation; (4) the amplitude of the first component in each spectrum of raw data; (5) a second component; and (6) its amplitudes