# Measuring spin direction by means of coherent light?

Let us assume that we have the ferromagnetic surface where all atoms have the same direction of spins. And we can interact with the single atom in such lattice. Is it possible to measure the direction of the atom's spin, by means of coherent light?

• I'm not sure that coherent light can be used on atomic scale, but I've seen it done for thin films. Dec 17, 2013 at 13:41

In the optics regime, every time a wave impinges on a surface it is modifying the angular momentum of an electron. Since the electron PE is usually comparable to the visible regime.

As for measurements with coherent waves. I don't think this is an easy task, though I think it is possible. I mention waves, because depending on the energy level of your electron it could require UV or higher energy waves. The light must be coherent to measure the angular momentum of an electron because the polarization of the wave determines how it will interact with the electrons. You would need to tune your wavelength to the material's electron shell that your interested in. If that electron shell has electrons that are only spinning in one direction, then I suppose yes, based on the polarization of the transmitted/reflected wave you could theoretically identify the spin state of the electron.

If you want to see a paper that determines spin states using linearly polarized light then please see attached. I personally think I would prefer to use only completely polarized right or left light. However, I am not an expert on the topic and the attached paper uses linear polarized light. Maybe they explore this topic in more detail...

• No, it is not violating Pauli Exculsion principle, because the spins are at DIFFERENT atom. So they can be the same technically. Dec 17, 2013 at 1:22
• It's mighty easier to use linearly polarized light. Then you can use rotation to measure magnetization. In turn rotation can be deduced fairly easily if you separate beams into orthogonal polarizations. Polarizing cube does this job well enough. It could probably be done with circular polarization but there are two problems. First - it will interact with measured magnetization. Second - technicality really but it's easier to make polarizing cube for linear polarization. You don't have to use fancy things like waveplates. Dec 17, 2013 at 13:48
• Thanks Sean. But attached article doesn't have any relevant points in there. They measured nuclear spin by gamma rays. Do you feel any differences? Dec 17, 2013 at 15:40