The position of the peaks is way off. I'd guess that by about +3 eV.

This is likely due to mis-calibrated spectrometer Fermi energy. You can calibrate it by scanning a clean sample of gold (usually Au 4f) or graphite (C 1s) and compare the peak position to the literature (NIST's XPS database is extremely useful). Charging could also play a role here but I haven't seen this to cause such severe shifts.

As for your question - is correcting the shift of the binding energy important? It depends on what you are trying to get from your data. Sometimes I'm interested in the relative distance between peaks or a particular peak shape and then the exact absolute positions aren't very important.

However, in this case the offset is so large that it may be challenging to properly identify the oxidation/chemical states of the elements. Moreover, if you ever want to publish these results - as a reviewer I would never let this pass with the energies being off so much.

C1s peak for TiC appears at ~282.9eV and C1s for C-C bond at ~284.5eV. Most probably, the most intense band comes from the graphite base and may be taken as reference. But then there should not be so much shift for the carbide peak.

@Mainak Roy: Thank you for your answer. Sorry , there was a typo error in my question. My material is not TiC it is Ti3SiC2. 

You may learn about peak positions in databases such as the NIST or www.lasurface.com. Check the availability of data for your compounds there! Good luck, Dirk

The position of the peaks is way off. I'd guess that by about +3 eV.

This is likely due to mis-calibrated spectrometer Fermi energy. You can calibrate it by scanning a clean sample of gold (usually Au 4f) or graphite (C 1s) and compare the peak position to the literature (NIST's XPS database is extremely useful). Charging could also play a role here but I haven't seen this to cause such severe shifts.

As for your question - is correcting the shift of the binding energy important? It depends on what you are trying to get from your data. Sometimes I'm interested in the relative distance between peaks or a particular peak shape and then the exact absolute positions aren't very important.

However, in this case the offset is so large that it may be challenging to properly identify the oxidation/chemical states of the elements. Moreover, if you ever want to publish these results - as a reviewer I would never let this pass with the energies being off so much.

I agree with Lior. Your peak position are quite shifted ( +3eV). Previously, la surface website was suggested. I would also suggest xpssimplified.com - where you can find basic B.E. value for components depending also on it's chemical structure.

Most of all... you have to be careful when doing the C-shift - check also that your Ti peak is in the position it should be. It is difficult for me to tell you which C peak to consider - but I think either the one you see at 288.3 or 293.4eV. Finally, it would be better if you can make a literature search for titanium silicon carbide and any XPS evaluation performed on it.

Mahesh, 

I feel, this huge shift can be due to the surface charge effect, which generally appear for high band gap samples. Which kind of samples are you using I mean powder or solid films? how you are mounting your sample to the sample holder I mean by carbon tape or molly clamp?

Usually, almost on any kind of sample is present the contamination of aliphatic carbon at the BE = 285.0 eV. However, for the start it is necessary to know if your X-ray source (Al or Mg?) is monochromatic. If not, you can have the contribution to all photoelectron peaks from X-ray satellites.

Since your material is electrically conductive, prior to shifting any of the binding energies, check to determine if the Fermi Energy is at 0 eV.  This is the most accurate and reproducible method for setting accurate binding energies.  You could also evaporate a small amount of Au on the surface and use the Au 4f binding energy to set all other values.  Be sure the Au layer is extremely thin as to not mask the underlying material.

 Shifting the carbon can create major interpretation issues as your sample may have multiple carbon chemistries including CO and CO2 along with Ti-C and Si-C.  

What are the binding energies of the Silicon 2p and 2s?

Your sample is Ti3SiC2 and doesn't contain oxygen. I don't understand why you have an oxygen peak (O1S at 539 eV ( intense) and 535eV ( less intense and broad)). Contamination !!!

I am not sure that your material is electrically conductive. It seems that it is a kind of ceramic.

What kind of material did you analyze ? A bulk material or a thin film deposited on a substrate. If it is a thin film on a ceramic substrate, it is possible that you have a charge up problem due to the electrical conduction of your sample. You can use a reference (evaporatin of Au on a small part of the surface) as suggested by David.   

Ti3SiC2 is a MAX Phase material. We are working with Ti2AlC MAX phase powder from Kanthal AB at present project. Ti2AlC powder is not very conductive and there is surface contamination issue as well. In the attachment, you can see spectra from Ti, Al, C and O from the as-received Ti2AlC powder and their binding energies. The XPS spectra are obtained using Theta Probe XPS.

Referencing your spectra by setting C 1s peak at 285.0 eV (or any other arbitrary chosen value) is highly unreliable.

See here:

https://www.researchgate.net/project/think-twice-before-you-use-C-1s-as-a-binding-energy-reference-in-your-XPS-work?_sg=oAzj2jcA8GF9q-IJIva6OW1fNa6Rq6XR2UGAZlVFoRXa98lZr1wmpJkwfXaRAouJsGOLcdICcH8V1ltnD9qLV4UqQMyAP3ayI9vp