Such damage is particularly evident in organic polymers containing unsaturated bonds and halogens (such as PVC and PVA) and catalytic materials with high valence oxidation states. The figure below show the changes in peak shape with prolonged analysis of a modified polyacrylamide. Note how the peak at 286.6 eV changes between consecutive scans (time ca.5 min between acquisitions).

In order to check for this damage we can run individual sweeps (for example by using a repeat sequence function) and then combining the data in post processing.

The figure below shows the reduction of an Au(III) salt where the  spectra were recorded in ca. 60 second intervals, clearly the concentration of the lower Au species increases due to reduction of the Au(III) species at ca. 87 eV. It can be very beneficial to set up an analysis so that a single sweep of a potentially sensitive region is recorded before the analysis of any further regions/surveys so that a comparison may be made.

The analyst may consider the following points

• Record reducible elements first and again at end to judge reduction
• Consider multiple measurement areas
• Consider reduced X-ray power (may affect charge neutralisation in some instances)
• Optimise charge neutralisation parameters on a replicate sample (or a clear area of the sample away from the analysis spot)

Polymer degradation has been studied greatly by Beams on and Briggs (Beamson, G., and Briggs, D. (1992) High Resolution XPS of Organic Polymers: The Scienta ESCA300 Database, Wiley-Blackwell), however deviations from the behaviour of polymers in Beamson and Briggs noted by HarwellXPS staff prompted an in-depth reinvestigation of these materials. We see that both x-rays and dual neutralisation can have a damaging effect on materials.

Dual neutraliser systems, however, are noted to impart a greater degree of reduction than electron flood guns alone.

Typically, maximising signal intensity is achieved by finding the analysis spot for a sample (i.e. moving the sample vertically until the sample finds the cross section of the incoming X-ray beam and the analysis column). This process may be performed manually or automatically on modern spectrometers, though in the case of ceria-based samples – one ought consider optimising this parameter manually and with any X-ray illumination ceased – to prevent unwanted sample reduction prior to even a single spectral acquisition. It should be noted that relying on a visual height optimisation will require good alignment between the optimal camera and the analysis position. In order to assess the degree of reduction for a standard sample (pure CeO2); reduction profiles were recorded for samples having undergone the automated sample position process vs manual sample process.

The below figure reports the degree of reduction following X-ray irradiation at various X-ray powers and the desirable impact of utilising a low power source may be seen in the low Ce(III) contents obtained from using a 45 W X-ray source. Furthermore, if we look at the rate of reduction as a function of power, we see that using a low power source minimises the % reduction per minute. If it proves possible to obtain appropriate signal:noise spectra using a low power source, it ought be concluded that this represents the preferable experimental set-up when analysing ceria-based materials. This observation should be of particular importance when considering samples analysed by high flux density sources (e.g. synchrotron radiation), in which the potential for high levels of rapid reduction exists.

XPS analysis of soft materials remains a challenging task, with sample degradation presenting itself across a wide range of different materials. Correct protocols when performing the experimental spectral acquisition will ensure minimal proliferation of errors in terms of scientific understanding during subsequent data treatments. Furthermore, XPS spectra of titanium butoxide will provide a valuable reference for the materials understanding of alkoxide based metal oxide and mixed-metal oxide functional systems.

Following a series of measurements it was observed that the Ti 2p region was indicating the presence of a small population of Ti³⁺ states, while the sample appeared to have degraded in colour.

In order to further investigate the relationship between analysis conditions and induced chemical changes to the sample, the analysis was repeated at a higher temperature of -60°C in order to increase the degree of reduction (figure below) and more readily observe differences due to experimental parameter settings. At this higher temperature, we see an exaggerated reduction of the Ti(IV) towards Ti(III) and may observe more clearly the changing molecular chemistry.

1. Morgan, David J., and Sharukaa Uthayasekaran. “Revisiting degradation in the XPS analysis of polymers.” Surface and Interface Analysis 55.6-7 (2023): 556-563. Read it online here.
2. McLaren, Rachel L., Gareth R. Owen, and David J. Morgan. “Analysis induced reduction of a polyelectrolyte.” Results in Surfaces and Interfaces 6 (2022): 100032. Read it online here.
3. Isaacs, Mark A., et al. “XPS surface analysis of ceria-based materials: Experimental methods and considerations.” Applied Surface Science Advances 18 (2023): 100469. Read it online here.
4. Isaacs, Mark A. “Low temperature XPS of sensitive molecules: Titanium butoxide photoelectron spectra.” Applied Surface Science Advances 18 (2023): 100467. Read it online here