I have a closed packed 2D infinite layer of spherical Au nanoparticles. I want to set up a wave-optics (Maxwell's equations) model for it. To reduce the model to a unit cell, I have two choices:
1) A hexagonal unit cell with a nanoparticle in the middle (attachment 1)
2) A rectangular unit cell with 1/4th nanoparticles at each corner and 1 in the middle (attachment 2)
I am solving for the full-field. So for my boundary conditions, I have two ports (top: for excitation, bottom: transmission). To simulate an infinite array, I must implement Floquet periodic boundary conditions. I have done a couple of trials which have not yielded any satisfactory results. So here are some questions, I would like to reach out with:
1) With the hexagonal model, I have set up a simple model with port excitation at the top and PML (perfectly matched layer around), but did not get any useful results. On the otherhand, when I am doing the same for the rectangular domain (without any nanoparticle, just a medium), it works. Is there anything specific that I need to be careful about when I am setting up the unit cell hexagonal?
2) With the rectangular domain, when there is no nanoparticle or when the nanoparticle is in the centre, it works. But when the nanoparticle is positioned close to the walls, the solution becomes unrealistic. I am attaching the two cases. When the nanoparticle is closer than a certain distance from the wall, the solution becomes erroneous. I have read that one has to implement diffraction orders when there are "features" larger than the wavelength. How could I solve this problem?
3) Is anyone experienced with such models would like to share her/his expertise through collaboration?
Thank you for your kind help!
Maybe if you use adaptive mesh around hexagonal unit cell with a nanoparticle in the middle you will find good results.
Hi Moez,
Thank you for your response. Is there any specific reason why adaptive mesh works better in the hexagonal unit cell?
Hi Rituraj
The adaptive mesh offers a good resolution especially in the event of modification of the physical properties. Maybe if you change the cell size you will find better results.
Thank you. I have found the solution to this problem though different trials.
- Badges
- Science topic
- Similar topics
- Developmental Psychology
- Attachment