Inelastic background subtraction from a set of angle-dependent XPS spectra using PCA and polynomial approximation

The contribution describes an application of the window principal component analysis (WPCA) to determine precisely the energy subregions in a set of angle-dependent XPS spectra where only the inelastic background exists. Localisation of the subregions enables to approximate the background with polinomial and to subtract the approximated background from the spectra. The reconstructed shapes of the lines obtained in this way were evaluated using several criteria. Application of the procedure to sets of simulated spectra proved that the optimal degrees of approximating polynomials ranged between the 2 and the 5 degrees, depending on the signal-to-noise ratio (SNR and on the structure of the spectra. (C) 1999 Elsevier Science Ltd. All rights reserved.