Experimental Reporting in XPS #
Reporting the methodology of your data acquisition and analysis is highly important in any publication, be it a research paper, thesis or internal report.
The inclusion of experimental parameters and conditions allow the reader to make an informed decision to the data presented and highlight any possible issues which may have been overlooked (e.g. analysis induced damage).
An example of a poor description can be seen below:
X-ray photoelectron spectroscopy (XPS) analysis was conducted on a <insert system name> with a monochromatized Al X-ray source.
Poor descriptions are far to commonly encountered in journal articles. We often know nothing about X-ray power, the data step size, pass energy for analysis, was charge compensation used, how the data was calibrated or how analysed – what software, sensitivity factors employed etc.
A better example would be something more comprehensive such as that below written for a Kratos Axis Ultra DLD spectrometer.
It is evident from the this experimental that a lot more information is provided. The reader can now make an informed decision on the data based on having knowledge of the acquisition parameters, the pass energy which would affect any peak fitting performed for example. Additionally, we have an idea of the software and methods used for quantification also.
X-ray Photoelectron Spectroscopy (XPS) analysis was performed using a Kratos Axis Ultra DLD (Kratos, Manchester, UK) using a monochromatic Al Ka X-ray source (10 mA x 12 kV = 120 W). The samples were mounted using double sided scotch tape placed on glass slide and attached using copper clips to the sample bar, the glass slide ensured the samples were floated from the spectrometer. The base pressure of the spectrometer was 5×10-10 Torr, rising to ca. 1×10-9 Torr during analysis.
High resolution spectra were collected at a pass energy of 40 eV, with a step size of 0.1 eV over a typical energy window of 20 – 30 eV, depending on the element. Survey spectra were recoded at a pass energy of 150 eV, using a step size of 1 eV over the energy range -5 – 1350 eV. All data was collected in the Hybrid mode using a slot aperture resulting in an analysis area of 700 x 300 microns. The system resolution at 40 eV pass energy is 0.68 eV, as measured by the FWHM of the Ag(3d5/2) peak for sputter cleaned metallic silver.
Charge compensation was performed using low energy electrons coupled with a magnetic immersion lens. The spectra were then subsequently calibrated to their corresponding C(1s) line of adventitious carbon taken to be 284.8 eV.
Data was analysed using CasaXPS (v2.3.25) using Kratos RSF’s which are modified Wagner factors. All data was transmission corrected in the analysis software and analysed using a Shirley type background. Where performed, fitting was achieved using lineshapes derived from bulk standards recorded on the same spectrometer using the LA peak shape in CasaXPS,
A 2023 study led by Dr. George Major and Prof. Matthew Linford of Brigham Young University, USA – found that across a series of high quality journals, only 66% of publications reported the spectrometer used in data acquisition, only 40% reported the X-ray source, and only 10% reported the pass energy used for data collection!
The methodology they used to assess good reporting was as follows:
- Whether raw data, or only synthetic peaks, were shown.
- Whether the XPS data were fit.
- Whether backgrounds were shown under the XPS fits.
- The synthetic peak shape(s) used in any fits.
- The background(s) used in any fits.
- Whether the XPS data were in the main text or supplemental information.
- The spectrometer used.
- The XPS fitting software used.
- The pass energy used for data acquisition.
- The x-ray source used.
- The spot size used
Guidance from George Major, B. Maxwell Clark, Kevin Cayabyab, Nathan Engel, Christopher Easton, Jan Čechal, Don Baer, Jeff Terry, and Matthew Linford #
#
Instrument/experimental parameters: #
- The manufacturer and model of the spectrometer.
- The x-ray anode, or source and x-ray energy, e.g., Al Kα with 1486.6 eV, include beamline and synchrotron, if applicable.
- Whether the source is monochromatized or non-monochromatized.
- The type of energy analyzer, e.g., concentric hemispherical analyzer.
- The x-ray power, beam size, and any beam rastering.
- The geometry of the measurement, i.e., the angle between the sample plane and the entrance to the electron energy analyzer and the angle with the x-ray source. Alternatively, the system geometry may be described with respect to the sample normal.
- The lens mode of the analyzer (wide angle, angle-resolved, focused spot).
- The pressure during the analysis or the instrument base pressure.
- The energy resolution (pass energy) of the survey and narrow scans and the number of scans taken.
- The dwell times or scan time and step sizes of the spectra.
- Whether (and how) charge compensation was employed, and if the spectra were energy corrected, e.g., by referencing to the C 1s peak.
- How the sample was mounted, e.g., whether it was electrically isolated from the instrument or grounded to it.
- Anything else that is relevant or unique to the experiment, e.g., if the sample was tilted for angle-resolved XPS, in situ preparation of the sample, the use of near-ambient pressure XPS, information about the polarization vector of the incident light, etc.
- If a sample damage test was conducted.
If peak fitting was performed: #
- The background selected.
- The types of synthetic peaks used, e.g., Voigt, Gaussian, Lorentzian, or a pseudo-Voigt (GLS or GLP) functions.
- Constraints placed on the synthetic peak shapes.
- The software used for peak fitting.
- The source of any sensitivity factors used or the procedure used to determine them.
They recommend that published XPS spectra show the following information and/or contain the following formatting. Some of these recommendations only apply to fitted spectra. #
- In plotted spectra, binding energy should increase to the left, as per convention.
- The raw data should be shown, not some processed, e.g., smoothed, form of it.
- The background chosen should be shown, and it should be present under the entire peak envelope (over reasonable limits). In general, the background should not be subtracted from the spectrum.
- The sum of the synthetic peaks should be present so that it can be compared to the raw data.
- The individual, synthetic peaks in the fit should be shown, unless there is justification to combine them.
- The residuals to the fit should be shown. This is an important, graphical way of assessing the quality of a fit. The residuals reveal where a fit matches and does not match the raw data.
Major, George H., et al. “Insufficient reporting of x-ray photoelectron spectroscopy instrumental and peak fitting parameters (metadata) in the scientific literature.” Journal of Vacuum Science & Technology A 41.4 (2023).
