Sensitivity factors are of course dependent on the source and I would highly recommend you to check the attachment with Average Matrix Relative Sensitivity Factors for XPS for different source (use the one with C1s as unity). These are based on the work published for XPS by M P Seah, I S Gilmore and S J Spencer in J. Electron Spectroscopy 120 93-111 (2001).

As ASF depends on the X-ray source and the element of concern, you can try demo version of CASA XPS for quantitative fitting of these things.

As above persons mentioned ASF is different for different sources.  I would prefer to take each component such as atomic cross section, attenuation length, transmission function of the analyzer , inelastic mean free path  to get the more precise elemental ratio.  As you mention you have different oxidation states , so one need to fit it with CASA XPS or fityk (free software) to get the elemental ratio. Bellow are two attached article for  atomic cross section and IEMF

As has been covered already you should use the sensitivity factors for the element in question (Ti in your example) and then curve fit the spectrum to get relative percentages of the total Ti concentration for the +4 and +3 states.

It is worth mentioning that where possible you should use the sensitivity factors supplied by the instrument manufacturer as these will often already have angualr corrections applied that are specific to the geometry of the system.

Equally, you have a choice of using Wagner factors or Scofield cross sections.

If your software outputs VAMAS format, then I would load that in to CASAXPS as it automatically contains the Transmission function etc that is required

If you have a Kratos XPS system, then you can use the Kratos modified Wager factors from www.casaxps.com\kratos for your quantification

Hallo Nisha as David wrote use the areas to obtain the relative percentages of the two oxidation states.  Is the sample homogene e.g. is there surface enrichment.  In which case you might be able to probe the depth using different energy photoelectron lines and/or by changing their energy with the excitation source. 

A little late to the discussion but for your problem you do not need to worry about the atomic sensitivity factor as the cross sections for photoemission are the same for the same element using the same photon energy (X-ray energy) and same orbital. The kinetic energies of T3+ and Ti4+ are close enough to have a negligible effect on the transmission function so the ratio of Ti4+ to 3+ in your case is simply 20:12, or 5:3. RSFs are needed for quantifying XPS of different elements since all they do is account for how likely it is that a photoelectron from a particular orbital and element is produced and detected. if you had an alloy of 50 %A and 50%B and the RSF of A and B were 1 and 0.5 respectively, then the areas of the XPS peaks would be 2A:1B. Dividing by the RSF would give you 2A:2B, or a 50:50 ratio as in the material.

Anjan

The RSF for any Sn state will be the same.  In its simplest form, you take the integrated peak area and divide by the RSF to get the corrected area.  This assumes the area has been corrected for transmission function of the analyser and also the RSFs are corrected for geometry and that a suitable background (eg shirley) has been applied.

There are a few examples on here that discuss this and equally a quick Google of "XPS quantification" will bring up more indepth resources from (for example) CasaXPS website

However, for your case it sounds as though you are more interested in the percentage of each Sn state?  That can be achieved by curve fitting.

However, Sn(II) and Sn(IV) are almost identical in their binding energy so separating them will be difficult and recording and subsequent analysis of the X-ray induced Auger lineshape will be required.

Hi Arely

You just need to use the value of 18.5 (I take it these are Scofield factors) if using the full doublet. Note the sum of the two components is the same as the total RSF.

If only using the area of the 2p3/2 peak for analysis then use the corresponding sensitivity factor for that peak

Dear All,  I think it would be better to try and keep questions as questions and not answers (well I would find it helpful),  although I can see that an answer throws up a related question:  The theme is rather large and can be very specific concerning species, sample type, source and detector system even software.  If you want to bring notice of other people to a thread a message can be sent.  Although late Andrews answer spelt it out very well (actually that was a message sorry). 

i am quite agree with prof. Andrew G Thomas