Unexpected Far Field Behaviour

[Joshua Davis]

Hi,

I am trying to use SMUTHI to look at the effect of the incident angle of unpolarized light on the far field scattering intensity of a single sphere on a glass substrate. While this is very close to the SMUTHI “hello world” example I decided to simplify it even further and set both layers to be air so I could compare it with the analytical solutions offered by something like miepython. I would expect that at normal incidence the angular dependence of the scattering intensity to be identical to the analytical solution and that as the incident angle is changed the far field intensity angular dependence would be shifted by the same amount. While the analytical solution matches at normal incidence, the angular dependence shift is not consistent with the input angle. Would there be any numerical reason for this? I have been using the scattered_far_field and azimuthal_integral methods to calculate the scattered intensity.

I am also curious about the polar and azimuthal angle arguments for the scattered far field, I am used to the polar angle ranging from 0:2pi, it appears that for any value of the polar angle outside of 0:pi the scattered field calculations don’t appear to be working.

Hopefully the attached code illustrates my problem.

Best Regards,

Josh

[Amos Egel]

Hey Joshua,

thanks for the detailled and well prepared support request. 

In Smuthi, the polar angle refers to the angle between the z-axis and the scattered direction. That is why it is defined only between zero and pi. The azimuth angle is the angle that goes around the equator, so it is defined between zero and two pi.

I understand that you expect that the far field plots have a  maximum at the incident angle position (i.e., for an incident angle of 0.1*pi, you'd expect the scattered far field to peak at 0.1*pi).

That expectation is justified when plotting a slice of the far field, i.e., when you plot the scattered intensity as a function of polar angle for a fixed azimuth angle (namely, the incident azimuth angle). But the quantity that you plot is the azimuthal integral of the far field. Its maximum is not necessarily at the polar angle where the maximum of the 2D far field distribution is located.

Does this answer help to resolve the issue? Otherwise, don't hesitate to ask back.

Best regards, Amos

--
Smuthi project repository: https://gitlab.com/AmosEgel/smuthi
Online documentation https://smuthi.readthedocs.io/en/latest/
---
You received this message because you are subscribed to the Google Groups "Smuthi" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smuthi+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/smuthi/bd7de267-c23a-4cb3-a686-65f7e2ca3b84n%40googlegroups.com.