Nonlinear polarization evolution in a polarization maintaining fiber?
Might sound crazy: we have forced the polarization state to evolve within a polarization maintaining (PM) fiber. But in fact a PM fiber is a birefringent one and, thus, any polarization state introduced into the fiber is divided into two waves traveling with different velocities. The two waves are conveniently called the "slow" and the "fast" one. These waves interfere at the output to form a new polarization state. This is a linear (non light intensity dependent) property of PM fibers. So what's new? Well, in our laboratory we are dealing with ultrashort pulses with extreme peak intensities. This intensities are so high that they start to influence the medium and through the medium they start to interact with each other (through cross- phase modulation and degenerate four-wave mixing). This interaction can further influence the output polarization state of light (a nonlinear interaction).
In our work we have chosen to eliminate the linear effects through the use of PM fiber segments rotated by 90 degrees with respect to each other. This way the "slow" and "fast" waves exchange roles in the consecutive segments. For small pulse energies the input polarization state is, therefore, preserved. For high energies, however, the nonlinear effects come into play and the polarization state starts to evolve.
Why bother? Well our funny device has a one major application.
Together with a polarizer, which blocks part of the output lighted
it can be used as a device with intensity dependent transmission.
And this is just the thing that is required for forcing fiber lasers
to work in pulsed regime. Yep. That what our new paper in
is about - a laser driven into pulsed regime with our new nonlinearities
polarization evolution scheme.