Learning Forum

[Debasish Biswasray]

Hi all,

I am using 'Lumerical FDTD' to simulate a structure involving metallic spheres. I observed that when the separation between spheres is decreased, the peak extinction cross-section gradually redshifts. But when the separation was less than 0.5 nm, the absorption started blueshifting. This blueshifting for smaller separations is quantum mechanically explained through the tunneling of charges.

I wonder how this tunneling can be explained through classical FDTD simulations. I want to know whether there is any analogy between quantum mechanics and Lumerical FDTD simulations. Otherwise, can this blueshifting be explained through classical FDTD?

Please help me understand this.

[Afroditi Petropoulou]

Dear Debasish,

Since you mentioned a separation between the spheres of less than 0.5nm, could you please tell me what is the mesh size that you have used? If the mesh is not fine enough, maybe the two spheres are touching each other. In this case the result that you get is probably a numerical error.

Considering the simulation is correct (the mesh grid is fine enough) the blueshift could be possibly observed due to very strong coupling between the spheres. The tunneling effect though cannot be simulated with FDTD.

Best Regards,

Afroditi

[Debasish Biswasray]

Thank you for your reply, Afroditi.

I used a mess size of 0.25 x 0.25 x 0.25 nm^3 to resolve the smaller separations. I hope the resolution is fine.

When the nanospheres are getting closer (from 25 nm to 0.5 nm), the peak extinction shifts red (which is a sign of plasmon coupling, according to earlier reports). When the separation is varied from 0.5 nm to 0 nm, the peak gradually blueshifts, which means the coupling between the spheres is reducing. I want to understand: is it true that for such smaller distances, the coupling decreases? If it decreases, then why?

[Afroditi Petropoulou]

Hello Debasish,

I would suggest as a first step to run a convergence test. You can gradually decrease the mesh size and check if the results for small separation (<0.5nm) are consistent. 

I cannot tell for sure why you observe the blueshift as I don't know exactly the simulation setup. But the tunneling effect cannot be considered the cause for this.

Best Regards,

Afroditi