Hi guys,
I am a complete beginner in XPS curve-fitting and deconvolution. Does anyone have tips for determining the number of peaks to fit?
Dear Dr. Thao Nguyen ,
below are some brief suggestions:
- Start peak fit to the unsmoothed data;
- Remove the background as part of the peak fitting process rather than before it;
- During the fitting of different element states, be carefull to peak separations and peak height ratios (you can consult specific handbooks);
- Try to give also a logical sense, both from the chemical and physical data (number of peaks for curve fitting) point of view. The individual contributions of the hypothesized compounds should be closely related to the type of experimentation performed.
Best regards and good work, Pierluigi Traverso
Dear Thao,
There are so many software's are available now a days, both on line and offline.4
Choose one for your samples.
Wishing you good luck.
Thanks you all for your precious advices !
Dear Thao, as Pierluigi said, you should know the possible composition of your sample. For that, You can run EDSX analysis.
Good luck.
Thank you all for your informative advice.
Sir Pierluigi Traverso , would you mind recommending some good handbooks for beginners? Thanks
Dear Dr. Thao Nguyen ,
I suggest you the following books; the first two related to the measurement technique while the third is a handbook consisting of 3 parts that illustrate, for different specific compounds, important data that can be useful in the interpretation of your spectra:
1) An Introductionto Surface Analysisby XPS and AES
By J.F. Watts and J. Wolstenholme
John Wiley & Sons Ltd (2003)
ISBN 0-470 84712 3
2) Surface Analysis by Auger and X-Ray Photolelectron Spectroscopy
By D. Briggs and J. T. Grant
Published by IM Pubblications and Surface Spectra Limited (2003)
ISBN 1-901 01904 7
3) Handbook of Monochromatic XPS Spectra
A) The Element and Native Oxides
B) Polymers and Polymers Damaged by X Ray
C) Semiconductors
By B.V. Crist
John Wiley & Sons Ltd (2000)
ISBN 0-471 49265 5 (V1), ISBN 0-471 49267 1 (V2), ISBN 0-471 49266 5 (V3)
…. and finally, three very interesting web sites too, available at:
- http://www.xpsfitting.com/
- https://www.nanolabtechnologies.com/instrument-xps-esca/
- https://xpssimplified.com/resources.php
Good work and my best regards, Pierluigi Traverso
Fitting peaks is best learned from an experienced advisor . Start with a known composition and look in the spektra displayed in the literatur (PHI- Handbook, NIST Data base etc.). For each peak associated with a chemical shift their should be a component in the binding partner. contaminations like surface oxides should also represented in the binding partner. Spin orbit splits should always be taken with the appropriate intensity ratio and fixed energy separation and handeld at a first step with fixed fwhm. For unmonochromatized XPS first look for the X-ray satellites. If shake ups or multiplett-splitting is expected mark the energy regions. Check if overlapping paeks from other components are present (e.g. C1s and Ru3d). For identification of the chemical state often the Ager parameter is very useful, so record also the Augers in your spectra. For an example you may look into:
Meier, U.; Pettenkofer, C.
Morphology of the Si-ZnO interface
AppliedSurfaceScience. 252, iss. 4,(2005). 1139-1146
or
A Popp, C Pettenkofer
Epitaxial CuGaSe2 Thin Films− Removal of Cu2− xSe Secondary Phase From Film Surface
physica status solidi (b) 2017,254 (11)
Hi Thao,
You can use the XPSPEAK fit software to perform fitting of your experimental spectrum:
https://xpspeak.software.informer.com/4.1/
While performing the peak fitting of the experimental spectra, make sure each fitted peak has physical significance, which you can accesses from the literature for your particular sample.
Hope this is informative.
Thanks
Suhail
You can easily determine the related peaks of your sample using Software, XPSPeak4.1. It is free to access.
Charging correction:
There can be charging in the samples; which can be corrected by referencing Carbon C1s at 284.6 eV.
The peaks of carbon: C1s ref1 is recorded in the beginning of the scans, and C1s ref2 at the end of the scans.
Check that the peak positions of these two peaks is same (plus/minus ~0.2 eV).
Take average of the two peaks positions C1s ref1 and C1s ref2.
Subtract 284.6 eV from the average value to calculate the difference.
This difference gives the peak shift due to charging.
Subtract the the 'difference value' from the BE values of all the element's data.
After subtraction the peaks will move towards lower BE side.
This completes the process of charging correction.
You may then proceed for analysis.
Thi Phuong Thao Nguyen Deconvolution of a composite peak into its individual peaks plays an important role in the interpretation of many types of graphs including XRD, XPS, FTIR, and PL etc. In this video, I have discussed how to deconvolute simple combined peaks, composite peaks and how to correct missing data in a given peak with the help of deconvolution in OriginLab. In the case you want to further ask about it, please do comment on the specific video, I'll respond to it shortly. I have provided the practice files (OriginLab) here. Thanks
https://youtu.be/WX_X3XlbKRA
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix