Half-Shade Devices and Miniature Polarization Devices
This is a continuation from the previous tutorial - advances in detection and error correction for coherent optical communications - regular, irregular, and spatially coupled LDPC code designs.
Half-Shade Devices
It is sometimes necessary to measure accurately the azimuth of a beam of plane-polarized light, i.e., the angle the plane of vibration makes with a reference coordinate system.
This can be done most easily by using a polarizer as an analyzer and rotating it to the position where the field appears the darkest. The analyzer azimuth is then exactly \(90^\circ\) from the azimuth of the plane-polarized beam.
A more sensitive method is to use a photoelectric detector and offset on either side of the extinction position at angles where the intensities are equal. The average of these two angles is generally more accurate than the value measured directly, but care must be taken to keep the angles small so that asymmetries will not become important.
Before the advent of sensitive photoelectric detectors, the most accurate method of setting on a minimum was to use a half-shade device as the analyzer or in conjunction with the analyzer.
The device generally consisted of two polarizers having their axes inclined at an angle a to each other (angle fixed in some types and variable in others). As the device was rotated, one part of the field became darker while the other part became lighter. At the match position, both parts of the field appeared equally bright. The Jellett-Cornu prism, Lippich and Laurent half shades, Nakamura biplate, and Savart plate are examples of half-shade devices.
Ellipticity half-shade devices are useful for detecting very small amounts of ellipticity in a nominally plane-polarized beam and hence can indicate when a compensator has completely converted elliptically polarized light into plane-polarized light.
Two of these devices are the Bravais biplate and the Brace half-shade plate.
Miniature Polarization Devices
Polarization Devices for Optical Fibers
Single-mode optical fiber-type polarizers are important devices for optical fiber communication and fiber sensor systems. These polarizers have been made by a variety of techniques.
Polarizers have been made by bending or by tapering a birefringent fiber to induce differential attenuation in the orthogonal modes. In most cases a fiber was polished laterally and some device was placed in contact with the exposed guiding region of the fiber to couple out the unwanted polarization.
Bergh et al. used a birefringent crystal as the outcoupling device and obtained a high extinction ratio polarizer.
Optical fiber polarizers made with a metal film coated onto the polished area to eliminate the unwanted polarization state seem to be preferred because they are stable and rugged. The original version by Eickhoff used the thin cladding remaining after polishing as the buffer layer, but it had an insufficient extinction ratio.
Other designs using metal coatings were suggested by Gruchmann et al., and Hosaka et al. Feth and Chang used a fiber polished into its core to which a superstrate coated with a very thin metal layer was attached by an index-matching oil. Yu and Wu gave a theoretical analysis of metal-clad single-mode fiber-type polarizers. Dyott et al. made a metal-fiber polarizer from an etched D-shaped fiber coated with indium.
In the above approaches, either expensive components are used or the structure of the polarizer is complicated and fragile. Lee and Chen suggested a new way of fabricating high-quality metal-clad polarizers by polishing a fiber ~ 0.4 μm into its core and then overcoating it with a 265-nm MgF2 film as the buffer layer followed by a 100-nm Al film. Polarizers fabricated in this way had an average extinction ratio of 28 dB with a 2-dB insertion loss at a 0.63-μm wavelength or a 34-dB extinction ratio with a 3-dB insertion loss at 0.82 μm.
Other devices for optical fibers have also been designed. Ulrich and Johnson made a single-mode fiber-optical polarization rotator by mechanically twisting successive half-wave fiber sections in alternating directions; Hosaka et al.’s fiber circular polarizer was composed of a metal-coated fiber polarizer and a \(\lambda/4\) platelet fabricated on a birefringent fiber; polished-type couplers acting as polarizing beam splitters were made by Snyder and Stevenson.
Polarization Devices for Integrated Circuits
Small and highly efficient polarization devices are also needed for integrated circuits. Some such devices have been proposed and fabricated.
Uehara et al. made an optical waveguiding polarizer for optical fiber transmission out of a plate of calcite attached to borosilicate glass into which a three-dimensional high-index region had been formed by ion migration to act as the waveguide.
Mahlein deposited a multilayer dielectric film onto a glass superstrate which was then contacted to a planar waveguide to couple out the TM polarization. This paper contains a good description of the polarizer design as well as extensive references.
Suchoski et al. fabricated low-loss, high-extinction polarizers in LiNbO3 by proton exchange. Noe´ et al. achieved automatic endless polarization control with integrated optical Ti:LiNbO3 polarization transformers. This was a better method of matching polarization states between two superposed waves than techniques that had been used previously.
Finally, Baba et al. proposed making a polarizer for integrated circuits out of periodic metal-dielectric laminated layers (Lamipol structures). Their experiments with Al-SiO2 structures were encouraging.
The next tutorial introduces nondispersive prisms.