Discussion Forum

An elliptical polarized state just means there is a phase shift between the field in different directions. So just measure the amplitude and phase of the reflected fields, say Ex and Ey, will give you what you want.

Thank you for the reply, Peng.

Since "An elliptical polarized state just means there is a phase shift between the field in different directions", so what I need to do is to measure the amplitudes and phases of the in-plane and out-of-plane components of the reflected field.

Under the default coordinate system, for p-polarized incidence, after Kerr reflection, the in-plane component would be Ex+Ez, and the out-of-plane component would be Ey. For s-polarized incidence, however, it is hard to tell the two components giving rise to the the Kerr rotation and ellipticity. Am I right?

best regards,
Xiaokun

Dear Xiaokun,

I do not really understand your concerns. Since I'm not working on this subject, please correct me if I am wrong.

As far as I know, after a wave is reflected by a magnetic-optic film, there will be two components. One is the "normal component". By normal it means it has the same polarization as the incident wave as s-polarized or p-polarized. The other component is the so-called "Kerr component", the direction of this component will depend on the magnetization direction. There will be a phase different between the normal component and the Kerr component. And this makes the total reflected field elliptical. So I guess the phase and amplitude difference of these two components are the qualities you need to measure, to calculate the ellipticity, right?

If so, you can always measure all components of the total reflected field (total field minus incident field, or use port) in a Cartesian coordinate, then do the math yourself to decomposite them into the normal component and the other component with respect to the direction of k. Do I understand correctly?

Dear Peng,

As far as I am concerned, it is totally correct for your description of the reflected light by a magnetized surface, being the known Magneto-optical Kerr Effect (MOKE).

What I was wondering all these days is that whether the COMSOL is able or not to give the information about the amplitude and phase difference of both normal component and Kerr component. And from your response, it seem to be positive to obtain them by measuring the total reflected field and then decomposing it into those two components. This is quite inspiring!

It's just, I am a rookie in comsol. I am still working on how to measuring the reflected field.. I could understand the "total field minus incident field" approach using the expression like "Ex-Ein" (total-incidence), right? However, what do you mean by "use ports"? Could you explain a little bit more?

Thank you so much!
Xiaokun

Dear Xiaokun,

To measure the amplitude and phase of both components, I actually have two ideas in my mind.

1) Using "scattered field" instead of "full field" solver. In this way, you can easily get the scattered field ( say, emw.relEx) conveniently, which is exactly "totalfield (emw.Ex) minus incident field (emw.Ebx)".

2) Using port. You can set two orthogonal ports ( e.g. with two different polarizations ) to absorb the reflected waves separately, and use S-parameter to calculate their amplitudes and phases.

Best regards.

Peng

Dear Peng,

The explanation couldn't be more explicit. I will try those two methods and if there is more question coming up, I'll turn to you for help in this thread. Thanks again!

Best,
Xiaokun

Dear, I am trying to simulate the magneto-optic Kerr effect, using the relative permittivity tensor, as it is known, such a tensor has in our case, components outside the diagonal, these components are dependent on the magneto-optical constant and also the field. However, when I am measuring the reflectance as a function of the normalized magnetic field, we observe that the comsol is not reading correctly the elements of the tensor, that is, the reflectance is constant even with the variation of the field. Can you help me?

In the figure found attached, Q is the magneto-optical constant, and m is the normalized magnetization, that is, it assumes values between 0 and 1.

Thanks.