Data scaling for quantitative imaging XPS

Abstract

XPS spectrum image data sets acquired on laboratory instruments have inherently poor signal/noise, and require the use of multivariate analytical techniques to avoid prohibitively long acquisition times. However, when procedures that order the data by variance are used, the data set must be scaled beforehand, since it has a Poisson distribution. Different scaling methods may be used, but their effectiveness in allowing a separation of the chemical information from the noise is critical if loss of information is to be avoided. The performance of three methods, square root, root mean square and optimal scaling, has been compared, and their effectiveness for quantitative photoelectron spectromicroscopy is discussed.

High‐resolution XPS spectromicroscopy

“…In addition, the more systematic use of presently existing efficient noise reduction procedure and image treatments would make chemical state identification possible within a typical 20 nm image pixel. One of the important future challenges is quantitative XPS spectromicroscopy with XPEEM, as performed already with other parallel imaging instruments; [20,22] beyond the preliminary work, [39] this will require a full characterization of the instrument and the use of eventcounting detectors. However, other challenges are closer to be achieved until this becomes a reality, for example, the advent of the first complete spectromicroscopic laboratory photoemission instrument allowing simultaneously the spectral imaging of corelevel electrons and of valence electrons in real and reciprocal space; this is already within reach since the spectral imaging of valence electron in reciprocal space [41] was already demonstrated with HeI-excited spectroscopic PEEM.…”

Developments in numerical treatments for large data sets of XPS images

“…Studies have shown that data scaling can improve significantly noise reduction by PCA . The whole point of the outlier filter is to move the data towards a Poisson distribution so that root mean scaling is used, allowing PCA to separate the chemical information from the noise.…”

Imaging XPS for industrial applications

“…Prior to the development of delay-line detectors, XPS imaging did not provide any quantitative information [9]. Since such developments, over the last decade or so, development of traceable and quantitative analysis methods for XPS image analysis have been reported [9,[17][18][19]. However, XPS spectrum image data sets acquired using standard laboratory spectrometers tend to have inherently poor signal-to-noise, and therefore require the use of multivariate analytical techniques to achieve data scaling and avoid prohibitively long acquisition times [19][20][21][22][23][24].…”

“…Through the generation of such spectra-at-pixel, standard spectroscopic data processing techniques can be applied for each spectrum. Problematically, the short acquisition times lead to very poor signal to noise ratios for a spectrum from a single pixel, however, the large datasets generated are suitable for multivariate analysis which can be used to gain significant improvement in signal to noise and further information on the underlying surface chemistry [7,19,20,22,24]; the examples which follow highlight such analysis.…”