I just add a EM circular polarized wave simulaton. The default state in the simulation is circular polarized wave, you can change the phase to zero to view a linear polarized wave. For the linear polarized wave, the phase for the electirc field and the phase for the magnetic field are the same. For circular polarized wave, the phase difference is 90 degree (When E field is maximum, B field is zero). Yes. You can produce a circular polarized wave from linear polarized wave. Let a linear polarized wave pass through a media in which the index of refractions in not uniform (i.e. the phase change of the E field in two perpendular direction are differnt). Adjust the thickness of the media so that the E field is 90 degree our of phase when the wave pass through the media. If we let the linear polarized wave pass through the above device it become a circular polarized wave. (Right or Left circular polarized wave depend on the phase angle is +90 or -90 degree). If we let the circular polarized wave pass another similar device again, the phase change becomes 180 (or 0) degree which is a circular polarized wave. Then we can use a linear polarizer to detect it polarization again. In principle, if we can have a device to determine the difference between 180 or 0 degree (may be by interference method), we can build a left/right circular polarizer. However, I am not sure if it can be done in practice for optics and quantum wave (because the wavelength is so small). I would said that it can be done, if we can build a device with dimension similar to the wavelength of the wave. But it can be done from theorical point of view. Correct me if someone know such device really exists!