It is possible, yes!

Without knowing any details of your sample or MOKE setup, I can offer some basic advice:

1) I would suggest to use lock-in detection with your MOKE setup. You can pick up a really, really small Kerr signal (like one from a 2nm sample) by modulating the polarization of your incoming beam. A much cheaper way, one that we use in our lab, is to chop (either electrically or physically) your incoming beam and lock onto that frequency. This method is not as sensitive as modulating the polarization of your beam, though.

2) Another way to get the most Kerr response out of your setup is to play with the angle of incidence. For polar MOKE, use normal incidence. For the others, the angle of incidence which gives you the greatest MOKE signal will depend on the index of refraction of your sample/substrate. Another idea (if you are doing longitudinal or transverse MOKE) is to angle your sample such that the angle of incidence is the Brewster's angle of your substrate. I've never tried this personally, but it was an idea my adviser came up with.

This paper helped me understand how parameters like this effect the Kerr response: http://scitation.aip.org/content/aip/journal/jap/74/11/10.1063/1.355081

Hope I could help.

Tatiana

It is possible, yes!

Without knowing any details of your sample or MOKE setup, I can offer some basic advice:

1) I would suggest to use lock-in detection with your MOKE setup. You can pick up a really, really small Kerr signal (like one from a 2nm sample) by modulating the polarization of your incoming beam. A much cheaper way, one that we use in our lab, is to chop (either electrically or physically) your incoming beam and lock onto that frequency. This method is not as sensitive as modulating the polarization of your beam, though.

2) Another way to get the most Kerr response out of your setup is to play with the angle of incidence. For polar MOKE, use normal incidence. For the others, the angle of incidence which gives you the greatest MOKE signal will depend on the index of refraction of your sample/substrate. Another idea (if you are doing longitudinal or transverse MOKE) is to angle your sample such that the angle of incidence is the Brewster's angle of your substrate. I've never tried this personally, but it was an idea my adviser came up with.

This paper helped me understand how parameters like this effect the Kerr response: http://scitation.aip.org/content/aip/journal/jap/74/11/10.1063/1.355081

Hope I could help.

Tatiana

I agree with Tatiana. We have used MOKE with the lock-in technique routinely for magnetic characterization down to ca. 2 atomic monolayers, or 0.4 nm, and this is basically limited by the reduction in the Curie temperature of the samples which makes them paramagnetic. We use a photoelastic modulator for this purpose, but it is possible to do it also with a modulable semiconductor laser.

I agree to the proposals above. However, in some cases is it even possible to detect the MOKE signal of 2 atomic layers without modulating the polarization or amplitude, e.g. for Co/Cu (DOI: 10.1063/1.360884). The most important thing is to use a stabilized laser source and polarizers with superior extinction ratio (~10^-6).

Generally, you get higher signals from polar MOKE than from in-plane MOKE. However, this is usually given by the material, the substrate and the thickness and cannot be changed easily.

In the papers below we have used the simple MOKE configuration to measure the hysteresis loops of a few atomic layers of Co/Cu(100) and Co/Cu(1 1 17). In addition, we used the lock-in technique to modulate the magnetic field and detect the ac-susceptibility of the ultrathin films, even above the Curie temperature.

https://aip.scitation.org/doi/10.1063/1.355653

https://journals.aps.org/prb/abstract/10.1103/PhysRevB.50.16074

Best regards,

Stephan