Generation of polarization-squeezed beams

[James2018]

TL;DR

Polarization-squeezed beams can be generating by interfering amplitude-squeezed beams of light with orthogonal polarizations on polarizing beam splitters

Stokes operator squeezed continuous variable polarization states​

https://arxiv.org/pdf/quant-ph/0208103.pdf

Polarization squeezing and continuous-variable polarization entanglement​

https://arxiv.org/ftp/quant-ph/papers/0108/0108098.pdf

It is shown that a light beam formed by interference of two orthogonally-polarized quadrature-squeezed beams exhibits squeezing in some of the Stokes parameters. Passage of such a primary polarization-squeezed beam through suitable optical components generates a pair of polarization-entangled light beams with the nature of a two-mode squeezed state.

I just found out from these sources that beams of light that are squeezed with less uncertainty in amplitude or phase than a coherent state, generated in nonlinear crystals, when overlapped on a polarizing beam splitter generate a polarization-squeezed beam.

I thought interference between orthogonal polarizations was not possible?

"The two constituent orthogonal linearly polarized states of natural light cannot interfere to form a readily observable interference pattern, even if rotated into alignment (because they are incoherent)."
https://en.wikipedia.org/wiki/Fresnel–Arago_laws

And what exactly does a polarization-squeezed beam mean?

 

[Cthugha]

I thought interference between orthogonal polarizations was not possible?

"The two constituent orthogonal linearly polarized states of natural light cannot interfere to form a readily observable interference pattern, even if rotated into alignment (because they are incoherent)."
https://en.wikipedia.org/wiki/Fresnel–Arago_laws

That statement is a bit too brief and uses a rather limited definition of what interference actually means. You will indeed not get a spatial or temporal modulation pattern of the intensity (which is usually called interference pattern) for orthogonally polarized beams.

Still, the superposition principle applies and by superposing two beams of orthogonal polarization, you will of course as a result get beams with a different polarization. For example, if you start with horizontally and vertically polarized beams of the same amplitude, you may get diagonal or circular polarization. The whole paper you refer to focuses on these polarization properties of the resulting beams.

And what exactly does a polarization-squeezed beam mean?

The definition is inside the paper you cited:
Light is said to be polarization squeezed, according to the definition in Sec. I, when the variance in one or more of the Stokes parameters is smaller than the coherent-state value.
The three Stokes parameters correspond to the degree of polarization in the linear, diagonal and circular basis, respectively.

As you need a horizontal and a vertical beam of exactly the same amplitude and phase to create an exactly diagonally polarized beam, it makes sense that the amplitude and phase noise of the initial beams also play a role in the properties of the polarization of the resulting beam. If the amplitude of one beam is larger than the other, the beam polarization will not be exactly diagonal, but slightly more horizontal or vertical. If the phase is slightly off, it will become slightly circular. Accordingly, you can just check, how small the polarization fluctuations can become when using two coherent beams. If you can beat these fluctuations in any of the parameters, you have polarization squeezing.

 

[James2018]

Well I guess that answers my question, thank you.