How to calculate the green function of n number of dipoles placed randomly in photonic structures

HI,
We are trying to calculate the green function of n number of dipole source embedded randomly in fcc photonic crystal. We are attaching one file have look at it and please tell us how to proceed further. We have gone through one example that is already given in Knowledge exchange pcfcc.fsp (338.7 KB)
how many dipoles source we can use in the calculation.I have seen one example is given in your website
https://apps.lumerical.com/nanophotonic_applications_greens_function_and_ldos.html

:slightly_smiling_face:

Hello @physics,

I would suggest calculating the Green function 1 dipole at a time by creating a of dipole positions to sweep over. Consider this a systematic yet unbiased sampling technique, and the dipole cloud a completely random sampling technique. Having many sources will make your simulations less efficient, and I have less intuition about the results from such an approach. A priori I do not know how many dipoles you may need. This should be part of your convergence testing, starting with the results for say a 4x4, maybe try an 8x8 grid of dipoles to see if the results change significantly. I would suggest doing one dipole and one orientation at a time, so 4x4x3 simulations. You can get the Green function of n dipoles by coherent superposition of all the results. I would suggest modifying the scripts in the link that you provided. In addition to that page, I would refer you to Fluorescent Enhancment for more tips an this type of simulation, and to OLED symmetry for suggestions on how to reduce the number of simulations.

For further discussion see this KX post Computationally efficient Green's function calculation using quasi-normal modes. And this analysis of 3D FCC bandstructure.

I am not very experience with these methods, so I would suggest referring to L. Novotny and B. Hecht, Principles of Nano-Optics, Cambridge (2006).

Regards,

Hi @trobertson,
Thank you for the suggestions. I am trying to calculate the local density of states (LDOS) of the fcc photonic crystal sample. So it is possible to calculate LDOS with a single dipole source or do we require similar dipole clouds used in bandstructure calculations of FCC photonic crystal.