Simulating only a subsection of a larger lattice

[Chels]

Hello,

I am trying to investigate the optical absorption spectra of iron impurities in quartz, the problem is the impurities are only trace amounts in a larger crystal, specifically the iron replaces one silicon. So using the periodic options in CP2K I need to simulate quite a lot of the regular quartz in order to have a decent ratio between the Silicon and Iron.

I'm not particularly interested in the excitation energies of all the different silicon atoms, just the iron and the oxygens immediately adjacent.

I am currently using an edited version of the TDDFT example here:

https://manual.cp2k.org/trunk/methods/properties/optical/tddft.html

I am wondering if maybe I could perform a geometry optimisation, export/save the potentials from that and then run the energy sim and use the EXTERNAL_POTENTIAL in DFT with only the atoms I want to see the excitation energies of? Does the DFT EXTERNAL_POTENTIAL get used when applying TDDFT?? Is that even what EXTERNAL_POTENTIAL is meant for??

This is all further complicated by the fact that I am working with fairly limited computing power so if I am able to reduce the number of atoms simulated that will be a huge help!!

Thank you if anyone can help!!