I have an XPS spectra, and want to focus on on the Ti2p area. I have a pristine sample, and another with oxygen vacancy? I want to show that there is formation of Ti(III) as charge compensation.
And how can I quantify them? I've attached the XPS spectra.
I know this will be a relatively simple analysis for some. I hope you can help me! Thanks!
In addition to the suggestions from R Gulyaev about experimental procedures I might add these notes about data analysis. They are to move your work from the realm of "first guess" to "publication-quality results".
* The spectra you show are smoothed curves. Do not peak fit the smoothed curves. Peak fit to the unsmoothed raw (noisy) data.
* You will want to remove the background properly. The best to use is a Shirley or Tougaard shape. The best method is to remove the background as part of the peak fitting process rather than before it.
* You will want to fit the Ti2+ and Ti3+ states as a fixed doublet with defined peak separations and peak height ratios.
The peak fitting is easy to do ... wrong. It is important to do it right. And do it on multiple spectra from multiple experiments (when you really are that interested in your system).
BTW, in case it is not clear, the term you want is peak fitting not deconvolution.
Thank you all for the informative answers! To answer some questions:
1. The TiO2-x powder I have is color blue, compared to the white TiO2 powder
2. The graph that obtained from the instrument center already came as this smooth graph.
3. I have read a paper where they subtracted both peaks, and used that information to reference the presence of Ti 3+.
I am to publish a paper, and this analysis is supplementary to my work! Thank you everyone!
> 3. I have read a paper where they subtracted both peaks, and used that information to reference the presence of Ti 3+.
This is one way to show changes that occur in the oxidation state of a metal. It is easier than peak fitting, and in some cases can give all the information you need. For example, now that you have the difference spectrum, why even bother to do peak fitting?
> I am to publish a paper, ...
As I have noted before, I have issues with specific parts to the approach that you are taking. In general terms, what I see so far are only "preliminary results". I would be glad to discuss my specific concerns about them not being "publication-ready". Feel free to contact me off-line when you are interested.
Dear Prof. Jeffrey,
Thank you for the valuable comments.
I have some doubts in XPS peak fitting analysis of my work.
Can you help me in this regards?
Yours,
Dennish Babu
How to plot XPS survey graph using Origin
https://youtu.be/t2xrywy5988
- Badges
- Science topic
- Similar topics
- Counseling Psychology
- Confusion