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.
Periodic Nanoparticle on Substrate
Posted Nov 4, 2016, 10:44 a.m. EDT 2 Replies
Please login with a confirmed email address before reporting spam
Hi,
I have a question about the 'Scatterer on Substrate' Model. From what I understand the model currently
works in the following way:
1) Calculate the full field solution without the nanoparticle over an infinite substrate by using fresnels equations and floquet boundary conditions.
2) Use this solution as the input background field with the nanoparticle present giving you the scattered field.
3) Use PML's to absorb this scattered field and calculate the cross sections.
Essentially the above simulates only a single nanoparticle on a substrate as the PML in the second interface absorbs any outgoing fields. If I want to simulate a periodic array of nanoparticles how could this be done? I tried simply adding floquet boundary conditions to the second interface but this throws back an error of "Failed to find destination boundaries or the destination selection is empty".
I then tried calculating both in full field with periodic conditions. When i subtract them from each other (emw2-emw) in the results section it appears to give what I think is the scattered field but I'm not sure if this approach is valid. In addition if it is, am i able to get the absorption and scattering cross sections from this? Do i somehow need a third interface with (emw2-emw) as the background field, but not actually solve anything?
Thanks for your help.
I have a question about the 'Scatterer on Substrate' Model. From what I understand the model currently
works in the following way:
1) Calculate the full field solution without the nanoparticle over an infinite substrate by using fresnels equations and floquet boundary conditions.
2) Use this solution as the input background field with the nanoparticle present giving you the scattered field.
3) Use PML's to absorb this scattered field and calculate the cross sections.
Essentially the above simulates only a single nanoparticle on a substrate as the PML in the second interface absorbs any outgoing fields. If I want to simulate a periodic array of nanoparticles how could this be done? I tried simply adding floquet boundary conditions to the second interface but this throws back an error of "Failed to find destination boundaries or the destination selection is empty".
I then tried calculating both in full field with periodic conditions. When i subtract them from each other (emw2-emw) in the results section it appears to give what I think is the scattered field but I'm not sure if this approach is valid. In addition if it is, am i able to get the absorption and scattering cross sections from this? Do i somehow need a third interface with (emw2-emw) as the background field, but not actually solve anything?
Thanks for your help.
2 Replies Last Post Nov 17, 2016, 7:05 a.m. EST
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Josh,
Your first approach(simply adding floquet boundary conditions to the second interface) seems to be right.
And the error should be something simple, I also got similar ones. I suggest you to try fixing the error in your first approach.
Regards,
Eugene
Your first approach(simply adding floquet boundary conditions to the second interface) seems to be right.
And the error should be something simple, I also got similar ones. I suggest you to try fixing the error in your first approach.
Regards,
Eugene
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
1) To simulate a periodic lattice you should do one of the following:
A) Change the BCs on the sides to FLOQUET type BCs. This must be done in pairs since the field is mapped from on side to the other, hence the error you see.
B) Change the BCs on the sides to PECs/PMCs (which should be which depends on the polarization of the incoming wave).
The advantage of approach (A) is that it remains valid for any angle of incidence. The approach (B) is only valid at normal incidence, however it is both conceptually and computationally simpler (e.g. the model will run faster/require less memory).
2) To calculate the scattering cross sections your should integrate the relative poynting vector over the particle surface (ewfd.relPoavx*nx + ewfd.relPoavy*ny + ewfd.relPoavz*nz). To calculate the absorption cross section your should integrate the energy dissipated in the volume of the particle (ewfd.Qh).
A) Change the BCs on the sides to FLOQUET type BCs. This must be done in pairs since the field is mapped from on side to the other, hence the error you see.
B) Change the BCs on the sides to PECs/PMCs (which should be which depends on the polarization of the incoming wave).
The advantage of approach (A) is that it remains valid for any angle of incidence. The approach (B) is only valid at normal incidence, however it is both conceptually and computationally simpler (e.g. the model will run faster/require less memory).
2) To calculate the scattering cross sections your should integrate the relative poynting vector over the particle surface (ewfd.relPoavx*nx + ewfd.relPoavy*ny + ewfd.relPoavz*nz). To calculate the absorption cross section your should integrate the energy dissipated in the volume of the particle (ewfd.Qh).