This result contradicts the XPS paradigm, in which BE of core level peaks is defined by the type and the nature of chemical bonds. Noteworthy, peaks due to two other chemical states of C atoms in AdC, O=C–O and C–O, shift in the same manner from higher to lower BE, as the probe is moved from Al to Au foil. Moreover, for the intermediate probe placements a double C–C/C–H peak is observed. 1, the binding energy of the C–C/C–H peak that dominates the C 1s spectra of AdC, with the envelope characteristic of adsorbed AdC, depends strongly on which sample area is probed and varies from 286.6 eV for the data recorded from the Al foil (bottom spectrum) to 285.0 eV for AdC/Au (top).
The spectra are recorded from the area 0.3 × 0.7 mm 2 that is sequentially moved from the Al part (position 1) to the Au part (position 9) with the step of 0.1 mm. 14 are 2.1 and 3.2 nm for Al and Au substrates, respectively.
The AdC layer thickness estimated from the attenuation of the substrate signal using the electron mean free path values reported in Ref. Here, we present direct evidence, which should terminally disqualify the charge referencing method based on the C 1s peak of AdC, and refute the deep-rooted notion that the same chemical state gives rise to peaks at well-defined BE values.Ĭommonly available Al and Au foils with AdC layers resulting from prolonged air exposure are set in contact and mounted together on the sample holder. The same rigid BE shift is then applied to all sample signals, hence assuming that the correction is independent of the electron kinetic energy.
The method is temptingly simple and requires no other effort than recording the C 1s peak of AdC and setting the C–C/C–H component at the BE arbitrary chosen from the range 284.6 to 285.2 eV, as recommended by the ISO charge referencing guide 13. Here, by far the most common charge referencing method is the one that relies on the adventitious carbon (AdC) contamination that occurs on essentially all samples analyzed by XPS 3, 12, 13-which accounts for the extreme popularity of this referencing technique. The situation is, however, cumbersome for insulators, which dominate in XPS analyses. Such samples typically exhibit a clear cut-off in the density of states at the Fermi level (so-called Fermi edge, FE), which serves as a natural zero on the BE scale 11. The availability of an internal energy reference, in general, does not present a big challenge for conducting materials in electrical contact to the spectrometer. In order to distinguish peak shifts caused by charging from those due to chemistry, an internal reference level is necessary to establish. This, however, does not solve the referencing problem as one can never a priori assume that the surface is neutral during an XPS measurement. To neutralize the negative charge loss and to enable spectra acquisition from poorly-conducting samples, low-energy electrons 10 or a combination of electrons and ions (supplied by the so-called flood gun) are used. If that does not take place, the surface charges positively, which effectively lowers kinetic energy of emitted photoelectrons due to the Coulomb interaction and, in consequence, results in an uncontrolled shift of spectral peaks towards higher BE values. The charge neutrality condition requires that the loss of negative charge from the surface region (the consequence of the photoelectric effect) is compensated with sufficiently high rate by electrons from the sample bulk, the substrate, or the surrounding environment. The primary reason for this is the possibility of positive charge accumulation in the sample surface region ( surface charging) 9.
This does not, however, guarantee that the BE of energy level of the sample of interest for study (different from the calibration set) is correctly reproduced. For the latter to be reliable, the spectrometer has to be correctly calibrated 8. The chemical state identification is conventionally done by comparing the extracted binding energy (BE) values to compound reference data bases such as the NIST XPS 7. However, significant fraction of these numerous XPS papers contains data that have been wrongly interpreted due to the lack of skills, experience, or knowledge 5, but also because an improper referencing method was employed 6.
The tremendous growth of the XPS technique is driven by the possibility of chemical state identification 2, 3, enabled nearly 60 years ago by the first observation of S 2p peak splitting in the XPS spectrum of sodium thiosulfate 4, caused by the fact that S atoms in Na 2S 2O 3 are present in two distinctly different chemical environments. With more than 12,000 papers published annually, the value of XPS in materials science can hardly be overestimated 1.