> Thank you all very much. > > Thank Postnikov answer my basic questions. and I am very sorry that I did > not describe the problem clearly. > Take another example to describe the problem, > step 1, calculate DOS the free O2 molecule, > step 2, put that free O2 molecule and surface together but make sure the > distance is long enough that there is no interaction between them. step > 3, > compare step 1 O2 2p orbital with the step 2 O2 2p orbital (the Fermi > level > of them is shifted to zero). > I expected the figure should be the same (completely o), but they are > different. The whole figure is translated on the x axis. It is like two > functions, when y > 0, y=x and y=x-2. The figures of the two functions is > the same, but they are different in the axis. Although as Postnikov said > that the fermi level has much useful meaning, when we shift the Fermi > level > to zero, we usually take "E-E_F" as x axis (E is energies for which DOS is > calculated, E_F is Fermi energy ). > If the case that I met exists , we cannot just choose "E-E_F" as x axis, > so > how can we shift the PDOS correctly ? > Thank you very much again
Sorry, this explanation does not make thinks any clearer. Before you talked about charged O2-, now you talk about surface. I come back to my first question: what, in fact, are you doing? Anyway. If you compare a) isolated molecule in a box and b) the same molecule in a much bigger box, which also includes a slab, the Fermi energies will be very different, because these two are very different systems. However, if you are sure that the molecule is far away and does not "directly" interact with the surface (no hybridization, no charge leakage) - its PDOS, in principle, must be "almost" the same as of isolated molecule. Technically, they might be slightly different because the presence of slab, even at a distance, will affect electrostatics. Mind that the regimes "molecule" and "slab", in cases a) and b) do treat Madelung terms differently. Best regards Andrei Postnikov
