Learning Forum

[buchinskii.av]

Hello everyone! I need in your help. I try to get physical modes in FDTD port and use PML BCs for this goal. But from time to time I get same modes and the only thing they differ in is the sign of neff. Can anyone explain why this happens when adding PML? I know that it is reffer to negative phase velocity but I don't understand why it is when PML BCs turned on. Please, help me! Thank you!

[Guilin Sun]

Apology for the late reply due to some reasons.

When PML bCs are involved, the modes solved can propagate along positive and negative axis, and all are physical. So the negative neff means the mode travels backward. This is to ensure the modes keep to be lossy for non-gain materials.  If you do not force this here 

you may get a mode with artificial gain. This is because, the FDE solver or mode source solves beta^2 internally. To convert beta^2 to neff, we need to choose which root of the complex sheet . This can lead to beta with real part positive but imaginary part negative (gain), or real part negative but imaginary part positive (backward propagation), even for a passive waveguide. Ideally it has both real part and imaginary part positive, which is notmal and for most solutions.