E distribution in nanohle array

Dear all,

I am trying to simulate a phonon polariton Thz nano hole array resonator. The reference of basic idea can be found in this paper: https://link.springer.com/article/10.1007/s00339-007-4084-8. The attached are my FDTD file and modified script calculate the volume averaged field enhancement in the paper.
FTIR CAVITY.fsp (465.9 KB) usr_integrate_poynting2.lsf (1.8 KB)
The material property of SiC is based on Lorentz model, which can be found here: https://iopscience.iop.org/article/10.1088/1464-4258/9/9/S07.
The plot I got is in reasonable agreement with the paper, but the E field distribution always has some ‘hotspots’ which looks like the artefact.

This can be more apparent when I tried to simulate our own structure, which consists of a dielectric layer on top of a phonon polariton substrate.
BaF2.fsp (1.9 MB)

Can anybody helps me deal with this problem? Thanks!


Hello @rx8,

These are due to the rectilinear grid used by FDTD, which makes it difficult to resolve curved surfaces:

You can reduce these by increasing the mesh density in this region with a mesh override object. When calculating results like the volume averaged field enhancement, you should use convergence testing to determine if the mesh is fine enough. Alternatively, you could use the DGTD solver, which uses a finite element mesh that is much better at resolving curved surfaces.

I hope this helps. Let me know if you have any questions.

I see. Thank you very much for your help!

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.