If I have a sample with phase mixture microstructure, can XPS determine Fermi energy for each phase present in the sample. Fermi energy EF is very important parameters in determining the structure and properties of materials.
If your sample is suficiently conducting, the Fermi-level is given by the back contact of your sample connected to the spectrometer. For non conducting samples you observe charging (often indicated by stränge photo Elektron peaks which look like a shark tooth). Charging will occure if the total photo current excited by your x-ray source can not be compensated by charges flowing from the back contact to the surface. For an inhomogenious sample all emissions at E Fermi will superpose, so no easy attribution to different areas is possible, as XPS is normally integrating over a larger area. E.g. a sample of Ag islands on Pd will show a strong emission from Pd d-levels at EF with no chance to see the Ag s-electron.
If you are shure your sample is sufficiently conducting for all constituents and there is no alloying between the phases, you can use the core levels of your sample to find the distance to EF, as the bindig energy is referenced to EF
In addition to what Christian wrote I think you should acquire the valence band spectra. Provided the sample is in good electrical contact and the spectrometer is well calibrated "0 eV" on the binding energy scale should correspond to the Fermi level. Then you can compare your spectra to the calculated density of states to get an idea about position of the Fermi level with respect to the electronic states.