Hello. I synthesized nickel iron selenide and measured XPS for analysis. I found something strange while I was analyzing the data. We don't know the cause, but compared to the previously reported papers, all the peaks were moving as much as -3eV. The values have already been corrected by C1s, so I can't think about what the problem is. I compared the main peak position of C1s with the standard value (285eV) and applied it to all peaks as much as the difference. Is there a problem with this method? Or could someone explain to me reasonable answer for me?
Dear Dr. Suncheol Kim ,
an effect similar to that found by you is present in the following papers and is related to the type (and the chemical-physical properties) of the compound you examined:
- Porous Nickel−Iron Selenide Nanosheets as Highly Efficient Electrocatalysts for Oxygen Evolution Reaction
Zhaoyang Wang, Jiantao Li, Xiaocong Tian, Xuanpeng Wang, Yang Yu, Kwadwo Asare Owusu, Liang He,*and Liqiang Mai
ACS Appl. Mater. Interfaces, 8, 19386−19392 (2016)
Available at: http://mai.group.whut.edu.cn/pub/2016/201611/P020161120370566435030.pdf
and at: https://pubs.acs.org/doi/pdf/10.1021/acsami.6b03392
- A nickel iron diselenide-derived efficient oxygen-evolution catalyst
Xiang Xu, Fang Song, and Xile Hub,
Nat. Commun. 7: 12324 (2016)
Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4980491/
- Overall electrochemical splitting of water at the heterogeneous interface of nickel and iron oxide
Suryanto, B.H.R., Wang, Y., Hocking, R.K. et al.
Nat. Commun. 10, 5599 (2019). https://doi.org/10.1038
Available at: https://www.nature.com/articles/s41467-019-13415-8
Best regards, Pierluigi Traverso.
Aside from the actual effects that Pierluigi Traverso discussed, please be aware that the C1s method is not as good as some authors pretend it is. See this project by G. Greczynski :
https://www.researchgate.net/project/Why-you-should-not-use-the-C-1s-peak-for-referencing-XPS-spectra
So please check whether the effect is also in the uncorrected data before reporting on it.
Well, if ALL the peaks in your spectra (Se, Fe, Ni) are shifted of the same COMMON amount with respect to expected values, that is about 3eV, the problem is hardly of chemical origin.
The only possible alternative causes I could address to are:
1) A trivial error in the referencing procedure... but I'm sure you have already double checked it.
2) a charging of your sample in the analysis region.
The latter point is not so strange. You may say that the same charging amount would have moved the C1s peak as well... True in general, but if you have measured the C1s peak just once through a wide scan before starting the campaign of high-resolution measurements in the Fe, Ni, Se core level ranges, , charge may have been accumulated during the wide scan.
To ascertain this, I recommend measuring again the reference peak, namely C1s in your case, AFTER the whole campaign of spectra collection has been completed.
Of course, the above said charging cannot occur, in case of metallic or semiconducting sample with a proper anchor to the sample holder. But it may happen in those cases in which the sample under nalysis is a film cotaing a non-conductive substrate, unless a proper electrical bridge is made with the same sample holder, usually by Ag paste.
Hope having given you some help.
If the measurements were made with an AlKa radiation source, carbon calibration is not the most convenient since the C 1s signal has a strong overlap with a selenium Auger signal (Se LMM).
Thank you all for your interest in my question.
I am sorry, but there was an error in my question. In fact, Se moved to a higher binding energy (BE) area.
To be more clear, I measured the XPS again.
We also measured and compared the XPS of Nickel iron hydroxide (NiFeOH) before adding Se. NiFeOH corresponds to the peak position of the previously reported Ni. However, if Se was added, Ni and Fe would still move to the lower BE area, whereas Se would move to the higher BE area. The values are approximately -4ev, +4ev, respectively. (Considering the oxidation states we expect)
I tried to check the oxidation states of the elements at each peak in a previously reported paper, but I couldn't find the answer anywhere.
For example,
(Ni's known peak is 852.6 (Ni0) / 854.6 (Ni2+) / 856.1 (Ni3+), Ni2p3/2) (The peak I got is ~850.x, I think it should be Ni2+)
Could you explain this phenomenon?
- Badges
- Science topic
- Similar topics
- Physics
- Condensed Matter Physics
- Solid State Physics
- PEAKS