C sp2/sp3 hybridisations in carbon nanomaterials – XPS and (X)AES study
B. Lesiak ,L. Kövér, J. Tóth, J. Zemek, P. Jiricek, A. Kromka, N. Rangam
Applied Surface Science (2018) 452, 223–231
Shape of the C KLL (KVV) Auger spectrum provides a measure of C sp2/sp3 hybridisations, alternative to C1s spectra fitting. Due to a smaller kinetic energy of C KLL electrons than C 1s photoelectrons the investigated information depths are attributed to lower or higher surface sensitivity, respectively. Shape of the KLL Auger spectrum of carbon nanostructures reflects density of electronic states (DOS) and contains contributions of sp2 (graphite) and/or sp3 (diamond) hybridisations, whereas for oxygen and hydrogen containing nanostructures this shape reflects chemical effects. C sp2/sp3 content is evaluated from parameter D, which is defined as an energy difference between the maximum and minimum of the first-derivative C KLL spectrum, where dependence of parameter D on C sp2/sp3 hybridisations is assumed to be linear between the D values of graphite and diamond. Derived values of parameter D and therefore C sp2/sp3 hybridisations were found to be influenced by procedure of smoothing the Auger spectrum and D values used for reference materials with pure sp3 and pure sp2 hybridisations. Purpose of this work was to estimate reliability of C sp2/sp3 hybridisations derived from parameters D determined for a set of carbon nanomaterials and study the chemical and morphological effects on the measured parameter D values. Presence of an inhomogeneous distribution of hybridisation as a function of depth from the surface was identified mainly in graphene oxides in contrast to graphite and reduced graphene oxide. The largest influence on parameter D and then evaluated C sp2/sp3 content resulted from oxygen and hydrogen at the surface and applied smoothing procedure in contrary to structural properties of carbon nanomaterials (crystallinity, grain size). Values of parameter D for C sp3 and C sp2 hybridisations, i.e. 13.2 eV and 23.1 eV, respectively, are recommended to be used for linear interpolation proposed by Lascovich et al.