Background subtraction III: The application of REELS data to background removal in AES and XPS

An analysis is provided of the use of reflected electron energy loss spectroscopy (REELS) data for the removal, by deconvolution, of the inelastically scattered electron background in Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) measurements in order to deduce the peak area intensities. Traditionally, the REELS data have either been used directly or the background intensity in the REELS data have been scaled by dividing by a factor A with respect to the elastic peak intensities, prior to the deconvolution. The scaling leads to an increase in the measured peak area approximately proportional to A. Measurements of the angular variation of the REELS data for scattering angles θ in the range 133°⩽θ⩽153° show for the first time that the inelastically scattered intensity can vary by a factor of 2.5, for an angular change of 16°, with respect to the elastic peak background intensity. This angular variation has structure which relates directly to that of the differential elastic scattering cross-section. A simple relation for A, for losses greater than 80 eV, is thus found involving the differential elastic scattering cross-section and the transport cross-section. This leads to a result for A, for our instrument, which has a minimum of 0.9 at Rb (37) and a maximum of 1.6 at Ta (73), consistent with experimental data. For losses less than 80 eV, the scaling changes exponentially with energy from 1/A at high values of energy loss to 1 at zero energy loss with a characteristic decay energy of around 26 eV. Using these results, an angle-averaged set of REELS data may be generated for spectral deconvolution in AES and XPS.