How to excite edge modes in a ribbon waveguide using dipole source?


in many papers,researchers say with a dipole point source
place over the ribbon waveguides in the z-direction
excite the edge modes that propagated in the x-direction.

Can you tell how can I do this?


converging simulation about graphene plasmonic

Hi Dezyani,

Probably you just need to place a dipole source close enough to the ribbon and use a mode expansion monitor to see what modes of the waveguide are excited.


for the first try, I simulated surface plasmon waveguide switch example .But my result is not true.I do not know dipole polarization direction, choose magnetic or electric dipole and simulation time.
simulation file:graphene_waveguide_switch.fsp (727.9 KB)



Hi Dezyani,

The polarization direction depends on the mode you want to excite. Also, it might be tricky to find the right position of the dipole to effectively excite the desired modes. You can try using a cloud of dipoles above the ribbon with different orientations and positions.

I think that part of the challenge is that the fields are very localized near the graphene sheet. You mentioned that this is done in some papers. Maybe we can look at one of those references and try to reproduce the result there.


I think this paper is good:



Because the width of graphene is under 50nm,only edge mode propagate along this


Hi Dezyani,

This paper was discussed earlier in this post:

I used the same simulation file from this post and replaced the mode source by a dipole source: Plasmonic bandpass filter_modFG_dipole.fsp (558.0 KB). The simulation now takes longer to achieve a small enough auto shutoff level. Even though the auto shutoff is not triggered within the original simulation time, you can see that part of the light from the dipole is coupled to the edge modes in the graphene nanoribbons:

To get more accurate results you probably need to run the simulation for a longer time. However, I wanted to point out that unless you are interested in the generation of the edge mode itself, it is probably more practical to use the mode source. When the dipole source is used more stuff happens in the simulation region because part of the radiation is not coupled to the ribbons and is scattered out.


Because of I could not find transmission spectra like the paper result.
please show me transmission in your simulation with a dipole.



Hi Dezyani,

I wouldn’t expect that using the dipole would help to get better results. If I understood the paper correctly, the reflection and transmission are normalized to the power through port 1, where they are looking at the surface plasmon edge mode. Therefore, it sounds like they simply use the dipole to generate the mode but all the analysis and normalization is done with respect to the surface plasmon edge mode.

For the analysis in the case of a dipole source I would use mode expansion monitors at the positions where you want to consider the input and output (ports 1 and 2 in the paper). The mode expansion monitor allows you to find the amount of power propagating forward and backward in a given mode.