If you have C-O and C=O, that should also be components of the O1s. However, it may happen that the envelope looks like it's a single component while it's actually not. In this case here:

Article Graphite Mediated Oxidation of Coronene Adsorbates: A UHV Study

the spectrum also looked like a single wide peak, but it was clear due to the fact that it was wider than the spectrum of equally treated HOPG that several species had to be in the mix there. Only upon heating which caused the central component to shrink (see figure S4, supplement) earlier than the others, the multi-component nature became more apparent to the eye.

In your case you may have bad luck since I assume baking your paper is not an option, so a component resolution of the whole peak may not be seriously possible.

Jürgen Weippert

Hi,

Thank for your input, I take a good look at O 1s and compare with heat-treated carbon paper. I found that heat-treated carbon paper show obvious C=O peak which the position is far from C-O peak. Therefore the O 1s that I show here could not have fitted it. So I think I over fitted C 1s peak. Only C=C, C-C, C-O, and pi-pi peak would exist in C 1s.

Thank you.

I would as firstly why is your C-C peak that broad? To a first approximation one would assume a similar FWHM for the C-C, C-O and C=O components.

From the FHWM of your O1s signal I would suspect there are at least two components. Any significant concentration of carbonyl containing species however I may expect to also show pi-pi* structure above ~ 534 eV.

David J. Morgan Good point about the FWHMs. Verifying that is indeed the first step.

Kavin Teenakul Have you ever taken reference spectra of HOPG or another graphitic sample without sp3 and maybe also a sample with exactly one well-defined sp3 carbon species? Then the width of both the asymmetric peak as well as the sp3 components would be secured.

Jürgen Weippert

Hi,

I unfortunately do not have sample that is highly sp2 or sp3.

Hi David J. Morgan,

In the single O 1s peak that I show above, the peak is very symmetric, I could not justify a hidden peak besides that single peak.

However, I do have heat-treated carbon paper with O 1s peak here. Heat-treatment is done at 500C 3hr in air. So new oxygen functional groups should be found.

The peak labelled C-O have FWHM of 2.3 and C=O is 1.7.

The FWHM of C-O and C=O should be similar as well is that right? So I think I am missing something?

The peak in the UT sample may look symmetric, but that doesn't mean there's not 2 peaks in there - however to fit them you need to be able to justify them being there by other means (e.g. vibrational technique etc). You could have species with very similar binding energies which would mean the peak is broadened.

Id also say your peaks are fitted with a very long tail, a Voigt function is OK, but these look almost fully Lorentzian which I certainly wouldn't expect.

Going back to your original C1s spectrum, there looks to be C-O and C=O functions based on the raw data intensity (not the fit), both of these have different O 1s binding energies, so if they are present, then the O1s spectrum has to be different (as in the HT spectrum)