Learning Forum

[you jo]

Hi, I'm trying to calculate the Purcell values in the bow-tie structure as below. however, the Purcell values are different as I modified the mesh values dx, dy, and dz. Especially, when I set the dz smaller than 1 nm, the Purcell value of the dipole in the Z direction shows an oscillating tendency vs the wavelength. The same result was for the box monitor and the automatic calculation. I want to know how to set the mesh value to obtain the right result without oscillation. Thank you for your attention.

If I set the mesh value as below, the result is different:

 

[Guilin Sun]

From your description, there are two issues:

1: Purcell factor changes with mesh size;

2: ripples using finer mesh.

Dipole is a special source that its magnitude decreases rapidly in space. To have max Purcell factor, it is recommended that the dipole should be located at the corresponding Yee cell for the specific field component (the source polarization). 

 

https://optics.ansys.com/hc/en-us/articles/360041612594-Purcell-factor-of-a-microdisk

 

For the ripples of the result when finer mesh is used, it is mainly due to PML reflection: finer mesh will lead thinner PML. You can 

1: use local override mesh around the gap.

2: increase simulation volume to make sure the backgroud mesh accuracy is not too high, such as mesh accuracy 6 or 8. 

3: use more PML layers.

4: use "Steep Angle PML" with more PML numbers.