I am new to XPS technique. Lately I have been trying to analyze some XPS peaks of my samples. How I do it is: First I plot the "narrow" scan of an element (peak) in Origin, then using Multiple peak fitting I try to fit multiple peaks (into the original single peak) which I assume might be contributing to a shift or distortion in the peak. I take the values of "centre of the peak" and FWHM of the peaks being fitted from NIST XPS library and look for a good match. If the Chi-square is decent I take the bonds (contributing peaks) which I have assumed initially to be present in my sample.
Am I doing it right, if not then what's the the correct way to do it? My technique is a bit time consuming and I am not very sure of the contributing peaks. Also it would be really nice if someone can suggest any free software which helps in easy analysis of XPS data. Thanks in advance.
I agree with Lior - the best you can do is measure some simple samples (I can recomment to start with a sputter-cleaned metal like Copper, Nickel, or a noble metal) and try to model the core levels of this element. With less noble metals you may do the next step - try to measure and analyse the data without sputter cleaning, for those more reactive elements, you will recognize metal as well as oxide contributions in the core levels of the metal, and in addition also oxygen 1s and carbon 1s peaks. so fitting those spectra will be more demanding. The next step would be a multilemental sample, and so on ... But again, even for me (I should mention that I started with XPS in the early 90ies ...) quantitative fitting and modelling of XPS spectra is time consuming, and there is no better way to learn than just practising, practising, practising!
Good luck, Dirk
I have recently performed XPS analysis of Ca doped LaInO3 sample. Can anyone tell me the right procedure to get quantitative analysis. From literature, some papers reported that atomic fraction is calculated by area intensity divided by atomic sensitivity factor of specific element to that of summation of area intensity divided by atomic sensitivity factor of all elements present in the composition.
Handbook of X-ray photoelectron spectroscopy by J.F. Moulder
Now confusion is that should i consider the area of all splitted core level peaks (i.e. e.g., 3d5/2 and 3d3/2 for La) or else?
Some papers reports that quantitative analysis can be performed using intensity area with consideration of atomic photoemmision crossection.
http://www.sciencedirect.com/science/article/pii/S0039602809002386
I am confused.
Please advice.
Dear Kapil,
of course you have to devide each peak INDIVIDUALLY by its sensitivity factor (photoemission cross section). In your case, probably the In 3d5/2, the La 3d5/2 and the O1s. Then you have to calculate the sum of those intensities, and divide the individual peak (corrected by the sensitivity factor) with this sum. Ending up, you will get the at-% composition for each element!
In cases where core level splitting is small, usually the sum of the core levels is tabulated. Hope this helps, Dirk
Dear Dirk
Thanks for the quick reply. Now for In 3d, the peak splittings takes place due to spin orbit coupling i.e., 3d3/2 and 3d5/2. In case of In element, should i consider the sum of area of both 3d5/2 and 3d3/2 peaks? If so, then only atomic scattering factor of 3d5/2 for In is given but not of 3d3/2. Same is the case for La 3d core level.
Please advice.
Dear Kapil Sood,
Do you have this book in soft form
"Handbook of X-ray photoelectron spectroscopy by J.F. Moulder"
thanks and regards
Nasir
Dear Srujan,
There is another question similar to yours. It discuss the width of your peaks, which is one of the things to include in your fitting model. You may find some additional info in the following discussion:
https://www.researchgate.net/post/How_to_set_the_FWHM_full_width_at_half_maxima_for_deconvoluting_the_XPS_spectra
The FWHM is dependent upon the instrumentation, where the Gaussian energy contribution from an hemispherical analyzer (DE) depends on pass energy (Ep), slit width (w) and the mean radius of the hemissphere (r) plays a role when estimating the theoretical resolution contribution of the spectrum. See for example equation 2 in G. Beamson and D. Briggs, High Resolution XPS of Organic Polymers, John Wiley & Sons (1992) where DE=Epw/(2r).
It also depends on the light source, which for simplicity could be estimated to have a Gaussian distribution. For non monochromatic sources the energy width is the life time of the transition, e.g. about 0.8 eV for Al K alpha. If using a monochromated source the crystal of the mono is determining the resolution. Investigating Ag 3d (lifetime of 310 meV) can then give you an idea of the resoltuion. The resolution should also be specified by the manufacturer.
A Gaussian temperature contribution should also be taken into account. For room temperature it is about 90 meV and at 77 K it is around 23 meV. See for example Hüfners Photoelectron Spectroscopy, Springer.
For Lorentizian lifetime contributions to the spectra i can recommend:
Papp, J.L.C.a.T., Widths of the atomic K--‐N7 levels. Atomic Data and Nuclear Data Tables, 2001. 77: p. 1--‐56.
To calculate the Voight contribution for this one can use the equation from this paper:
Olivero, J.J. and R.L. Longbothum, Empirical fits to voigt line--‐width --‐ brief review. Journal of Quantitative Spectroscopy & Radiative Transfer, 1977. 17(2): p. 233--‐236.
where DE_Voight=((DE_Gauss)^3/2+(DE_Lifitime)^3/2)^2/3
This will give you a theoretical estimation of what to expect under ideal conditions not taking for example vibrational broadening into account.
D. Briggs and M.P, Seah, Practical surface analysis also gives an insight in some of the above John Wiley & Sons (1983)
The above method was used for example in Eriksson, Review of Scientific Instruments, 85(2014)075119 to calculate the contributions from the instrumentation for a Ag 3d peak.
If you are using CasaXPS there are some films within the area of your question. see for example: https://www.youtube.com/watch?v=-nCchGGAxGE
Best regards,
John
- Badges
- Science method
- Similar topics
- Methodology
- Crystallography
- X-ray Diffraction