A simple approach to measuring thick organic films using the XPS inelastic background

Abstract

In this work, we explore the possibility of finding a simple mathematical function that describes XPS background intensities as a function of energy and depth of emission. We describe the shape of the global structure of the background in XPS survey spectra for organic overlayer films with thicknesses greater than~10 nm. In this situation, the shape of XPS backgrounds becomes statistical, and the detailed structure of, for example, energy-loss functions and elastic-scattering cross sections can be disregarded. Using Irganox 1010 as a model organic overlayer on silicon dioxide, gold, and silver substrates, a consistent description of the XPS background can be found, and overlayer thicknesses greater than 50 nm can be measured in the case of gold substrates.Furthermore, by using a suitable subset of these spectra, it is possible to separate the influence of overlayer thickness, electron kinetic energy, and intensity on the global shape of the spectrum. We show that it is possible to fully describe the shape of calibrated XPS survey spectra with reasonable accuracy for samples that have organic overlayers using only binding energy and sensitivity factor data. The same functions are shown to provide a reasonable description of the XPS spectra of silicon with an oxide overlayer.

Intensity calibration and sensitivity factors for XPS instruments with monochromatic Ag Lα and Al Kα sources

“…To connect the Al Kα transmission function to near constant transmission at higher energies, we employ the inelastic background from PMMA acquired on the same instrument. It has been found previously that the inelastic background of pure organic materials can be closely described, to less than 5% deviation, by an exponential function, A .exp( E / E 1 ) . Where A is a constant that changes only at the kinetic energy of peaks in the spectrum, E is the kinetic energy, and E 1 is a characteristic energy.…”

“…The change in background intensity at each peak is in proportion to the composition of the material and the sensitivity factors of those peaks . Figure A shows such a fit to the region above 2000 eV kinetic energy, where the intensity step at the O 1s peak is 1.17 times that of the C 1s peak, since the oxygen to carbon ratio is 0.4 and the AMRSF of O 1s is 2.93.…”

“…The best fit provided a characteristic energy of E 1 = 1240 eV. This is somewhat higher than the 730 eV found for Al Kα radiation, but this is to be expected given the smaller contribution of elastic scattering for both carbon and oxygen at higher electron kinetic energies . In the case of no elastic scattering, the background intensity would be nearly constant in intensity, and the characteristic energy would tend to infinity.…”

“…The increase in information depth in comparison with peak intensities is by a factor which is approximately proportional to the electron kinetic energy. At approximately 1400 eV kinetic energy and in organic materials that factor increase in information depth is in the region of 3 . By extrapolation, this implies that the Ag Lα source has approximately five times the information depth of the Al Kα source if inelastic background intensities can be analysed.…”

“…In theory, it is possible to analyse these backgrounds and obtain the overlayer thickness. However, currently, the only method of analysing the complete wide‐scan spectrum of a substrate/overlayer combination is valid only for Al Kα sources and organic overlayers . Nevertheless, this method was applied to the 15 nm film and it returned an equivalent organic thickness of 28 nm.…”