Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Can PML match metal

Please login with a confirmed email address before reporting spam

Hi

I want to simulate a semi-infinite silver reflector, and a single Particle suspending over the reflector. Due to the scattering by the particle, the light can be coupled to the surface plasmon along the silver surface. In simulation, we have to use some boundary conditions or PML to absorbed the surface plasmon and other light scattered by the whole structure.

In my simulation, I make the whole structure surrounded by PML. But I found there are always some light reflected back by the PML. It seems the PML can not absorb the surface plasmon completely no matter how large the computational domain is.

So I want to know how one can model something like this, Is there anyone of you have succeeded in modelling metal involved structure like what I'm facing now?

see attachment for the structure



4 Replies Last Post May 3, 2011, 12:20 p.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 3, 2011, 5:58 a.m. EDT
In frequency domain I have used metals with PMLs a lot to absorb surface plasmons. Like you, I too was getting back-reflections. But then I set the width of the PMLs in properties window to lambdaS_xyz (xyz=application name) and it was all fine afterwards. I will suggest you to try it out. May be it will work for your case as well.
In frequency domain I have used metals with PMLs a lot to absorb surface plasmons. Like you, I too was getting back-reflections. But then I set the width of the PMLs in properties window to lambdaS_xyz (xyz=application name) and it was all fine afterwards. I will suggest you to try it out. May be it will work for your case as well.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 3, 2011, 8:22 a.m. EDT
Thanks

I have tried your suggestion, it is still not correct. I monitored the absorption by the particle, but the absorption is changing with the size of the computational domain. Theoretically, it should not depend on the size of computational domain, if we model it correctly.

did I set the domain not correctly?
the domain setting for the two domains next to the silver reflector is silver refractive index + PML

Thanks I have tried your suggestion, it is still not correct. I monitored the absorption by the particle, but the absorption is changing with the size of the computational domain. Theoretically, it should not depend on the size of computational domain, if we model it correctly. did I set the domain not correctly? the domain setting for the two domains next to the silver reflector is silver refractive index + PML

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 3, 2011, 10:29 a.m. EDT
This is crazy then. Yes, I think you have set it up perfect. Are you sure that you generate mesh elements of the same quality/density each time you change the computational field of view?
This is crazy then. Yes, I think you have set it up perfect. Are you sure that you generate mesh elements of the same quality/density each time you change the computational field of view?

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 3, 2011, 12:20 p.m. EDT
I think mesh is not a problem, I use very fine mesh. And even if it is problem of mesh, I think it is just some small errors numerically , but the absorption curves i got have different resonance frequencies, (the particle is metal too).

so I think the shift of the resonance of particle is mainly caused by the reflected surface plasmon from PML.


I think mesh is not a problem, I use very fine mesh. And even if it is problem of mesh, I think it is just some small errors numerically , but the absorption curves i got have different resonance frequencies, (the particle is metal too). so I think the shift of the resonance of particle is mainly caused by the reflected surface plasmon from PML.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.