I am attempting to study the charge density difference and local potential difference of the interface between wurtzite GaN and cubic diamond using DFT in VASP. I have optimized the bulk structures and then strained the GaN to fit the diamond lattice constant (diamond has a larger bulk constant) to create an interface. I am looking at both pristine and intercalated interfaces where some of the G is replaced by C (for the 0002 GaN on 111 diamond interface) or where some of the C is replaced by N (for the 000-2 GaN on 111 diamond interface). The intercalated interfaces are being studied because of some experimental results our collaborators obtained. My question is, what INCAR tags should be included for these interfaces?? I am doing constrained relaxation with a version of VASP I compiled that only allows for relaxation in the c-axis direction, a technique used when studying non-polar surfaces in a previous research project we did, and then using ISIF=3 during relaxation. I am not sure about dipole corrections however. The VASP manual states that IDIPOL is used for slab/charged cells, so I am assuming this isn't correct for interfaces. Are dipole corrections necessary and if so what is the correct way to go about including them. Thanks so much for any and all advice/suggestions.
I've added some images from VESTA of the interfaces for reference.