Hi Vladimir

Polarization means that the electric field of your coherent light beam (laser) is oscillating in a particular stable direction. Using different wave plates will change the polarity. It depends on what type you are using. Please look at the link below and you will find the answer to your question

http://www.rp-photonics.com/polarization_of_laser_emission.html

Regards, Radhwan

plase look at the link

http://www.photonics.com/Article.aspx?AID=32949

All properties of laser light are the result amplification and resonator selection. Properties of laser optical scheme cannot be completely independent on polarization. Due to competing character of establishing of laser generation, "strongest" survive, and weaker die. Than, if we attempt to obtain unpolarized laser light,  "strongest" polarization will, nevertheless, establish oneself. Therefore, it is worthwhile to pre-determine the polarization, and often small difference in gain or losses for two independent poarizations is enough to obtain polarization ration 100:1. But it is rather difficult to obtain better than 500:1 due to depolarization in optics. 

• As Aleksandr wrote the polarization of a laser often has to due with the gain competition. Often an element such an intracavity brewster window, or the cut of a solid state gain medium,  will increase the loss for one polarization so that its gain lies below the threshold. Even without  such polarization selecting elements small birefringencies due for example to stress in mirror mounts can select a given polarization axis. In "unpolarized" HeNe, consecutive longitudinal modes usually have orthogonal linear polarizations due to this birefringence and as the cavity length changes the modes drift under the gain curves, causing a fluctuation in the polarization of the output beam.  

To my knouledge, the beam light emerges polarized from laser when this device uses a polarizer.  Otherwise, no!

Laser when generated are un-polarized . This unpolarized laser is polarized either linearly or circularly by passing it through polarising quartz crystals. ..

For polarization selection in laser produced beamone uses incidence at Brewster's angle on transparent plate. The polarization component which is parallel to an incident plane (p-component) passes entirely into plate without reflection. In contrast, the component with the polarization  which is perpendicular to an incident plane experiences reflection. There is a mirror behind the plate, which returns light to the active medium.

So, a plate weakens s-component four times before it returns to an active medium. P-component retains their intensity at Brewster's angle refraction. Multiple eliminations of s-component provide almost polarized light.

The laser in principle is un-polarized light. The intra-cavity  electro-optic devices are used to generate Q-sw pulses in solid state lasers with linear or circular polarizations. The Brewster angle of gas laser tubes can induce partial polarization too. A single mode laser is not necessarily polarized too. In fact the modes under the gain profile are not inherently polarized by themselves and the birefringence and the intra- cavity optics may  create polarization for some purposes.

The stimulated emission process preserves the light polarization. The laser light is made of stimulated photons, showing the same frequency, momentum and polarization. Of course, the active gain properties and the parameters of the resonant cavity must match each other to support the predominant stimulated emission process, building up the lasing mode. The cavity provides a tight selection of the wavelength, momentum and polarization of sustained photons inside the cavitiy, determining the prevailing stimulated photons population and definitely the characteristic of the laser light.

The polarization state of a laser depends on the anisotropic properties of either the gain medium or of the resonator providing amplification of the stimulated emission. In gas lasers (isotropic medium)  without Brewster components small anisotropies of the mirrors can indeed favor an alternate orthogonal polarisation state for consecutive longitudinal cavity modes as already stated by Michael Bell for the compact HeNe lasers. However Brewster-ended gas cells ensure a stable P - polarised emission in gas lasers. For solid - state lasers employing isotropic gain medium such as NdYag glass the use of an intra - cavity Brewster plate is necessary to obtain a polarizer emission, unless a non - planar ring oscillator design (NPRO), which brings the small amount of anisotropy,  is implemented in the case of monolithic resonators. The situation is different for anisotropic (birefringent gain matrix) solid-state gain for which 2 natural orthogonal polarizations (pi or sigma pol) usually exist, as e.g. NdYLF corresponding to 2 slightly different wavelengths. Without any intra - cavity Brewster plate to favor the pi or sigma polarizer emission, the one with the largest gain cross-section will prevail.

Reminder: Stimulated emission does not preserve the polarization of light because the coherence is not defined in terms of polarization. The polarization is the property of resonator conversely it is not the coherence property of light !

Vladimir, stimulated photons have same polarization as stimulating ones, so as long as your "first photon" leaves in cavity, it will stimulate all the others to be emitted in the same polarization. If no polarization selection optics is placed within your laser the polarization will be random. In fact, many simple HeNe lasers will start with random polarization, and you can switch between those by knocking the laser lightly!!! near one of the cavity mirrors.

You can check the this  effect  examining the  input beam into the amplifier and the output beam that emerges from the amplifier. For unpolarized input beam , the output  is most likely unpolarized and for polarized coherent input, the output resembles to be amplified as a polarized coherent beam.

As I understand that stimulated emission process will lead to the generation of a second photon having exactly same characteristics (phase, frequency, polarization etc.)as that of the stimulating photon. Therefore, if the input beam is unpolarized it may lead to unpolarized output if it is not polarized further by the cavity/resonator.

It also does not contradict with the coherence properties of laser radiation.

Laser light emerges polarized when a polarizer is added to the device, e. g.,  taking advantage of  the Brewster angle in a thin film polarizer placed at the window of the laser.  I think this procedure was the first used,  in one face of a synthetic (solid state) Rubi-laser.

The laser beam that traverses  the Brewster angle window is  partially polarized.

Polarization of a partial polarized light  can be improved by adding a seguency of such (Brewster) filmes in front of the light beam.  

Yes and at the expense of the  reduction in intensity.

Of course,  according the Malus law.

( and I think the experimental physicists can offer better alternatives )

I guess by using a wave plate, say HWP before of the polarizer in the beam's path and then tune both WP and the polarizer as the polarizer may well be offset s'times would reduce your losses in beam's intensity.

OK !  

CIRCULARLY POLARIZED LASER LIGHT IS OBTAINED WHEN ITS LINEARLY POLARIZED LIGHT CROSSES A 1/4 WAVE PLATE .