It is not a peak shift, but a change in the bond morphology of your material. Simply, you have more XPS signals from different bonds in the material due to chemical changes in its structure.
The pink line is the only one with "normal Cu in a salt" binding energy while black and red are exceptionally high. My guess would be that the atmosphere you applied for pink saturated the crystal surface in an electrically conducting manner while the treatments of black and red didn't which might have lead to partial (black) or complete (red) charging. Do you also observe shifts in the same direction for the other two elements in your perovskites?
Probably, there is no change in crystal structure but possibly there could be bonding effect. Do you observed the blue or red shift in peaks? See my question / answers and research papers on ResearchGate.