Polarizers

For low-power applications, Polaroid filters (polarizing sheets, sheet polarizers) are often used. These are also called dichroic. They consist of a special doped plastic sheet (a polymer materials), which has been stretched in one direction, such that the polymer chains are more or less aligned along one axis. Light a polarization direction along the chains is strongly absorbed, whereas the absorption is weak for light with a polarization direction perpendicular to these. The polymer sheet is usually mounted in some solid holder, which may have marks indicating the direction of polarization for maximum transmission.

Similar sheets are used for polarizing glasses (also called polarized glasses). In case of sunglasses, only vertically polarized light is transmitted. This reduces glare from water surfaces, for example, as horizontally polarized light experiences stronger reflection at such surfaces. In case of polarizing glasses for 3D viewing, one eye gets the vertical polarization and the other eye the horizontal one. In this way, a 3D display can transmit separate images for the eyes.

Polymer sheet polarizers can be made quite large, and usually they are quite cheap.

A more modern type of absorptive polarizer is based on silver or copper nanoparticles embedded in a thin glass plate. These glass polarizers are much more expensive and not available in very large sizes, but offer a substantial better performance in terms of polarization extinction ratio. Laminated versions are mechanically more robust and cause lower wavefront distortions.

Absorptive filters can handle only quite limited optical powers (have a low optical damage threshold), because the absorbed power is converted to heat, and the sheet can easily be damaged by overheating it. In case of simple polymer sheet polarizers, an optical intensity of only 1 W/cm² may already be critical.

There are reflective polarizing films, which largely transmit light with a certain linear polarization direction while reflecting (not absorbing) most of the light. Such films are widely used for liquid crystal displays, which need polarized illumination. While at least half of the generated light would normally be lost at the entrance polarizer, that part can at least be partly “recycled” if it is reflected towards the light source and thereby at least partly converted into the right polarization direction.

Much higher optical powers can be handled by polarizers where light with the “rejected” polarization state is not absorbed but only sent to some other direction. (If it needs to absorbed later on, a beam dump can stand much higher powers than an optical element.) The most common type of polarizing beam splitters exploit birefringence of a transparent crystalline material such as quartz (SiO₂), calcite (CaCO₃), yttrium vanadate (YVO₄), beta barium borate (BBO) or magnesium fluoride (MgF₂). Often, two pieces of such material with different orientations of the optical axis are cemented together (or joined with a small air space). The device is often mounted in a polymer housing, which may also contain a beam dump for light with the rejected polarization direction.

Different physical principles of birefringent polarizers are used:

• In some types of polarizers, such as the Nicol prism, the Glan–Thompson prism, the Glan–Taylor prism and the Glan–Foucault prism, total internal reflection occurs for one polarization state, but not for the other, so that one obtains completely different beam directions for these outputs.
• Other types of polarizers, such as the Wollaston prism, the Nomarski prism, the Rochon prism and the Sénarmont prism, only exploit somewhat different refraction angles due to birefringence, and not any reflection. Here, the two output beams are less strongly separated.

Such birefringent crystalline polarizers differ in various respects:

• For true polarizing beam splitters such as the Wollaston prism, both output beams are completely polarized. This is not the case for some other designs, such as the Glan–Thompson prism, the Glan–Foucault prism and the Glan–Taylor prism (even though these can be more or less optimized in that respect). Note that for many applications only one output is used, so that this aspect is not always relevant.
• For some designs (e.g., the Glan–Taylor prism, the Glan–Foucault prism, the Rochon prism and the Sénarmont prism), one of the involved rays is not deflected, i.e., continues to propagate after the polarizers in the same direction as the incident beam.
• Some designs work properly only for a narrow range of incidence angles, whereas others have a wider angular acceptance range. For example, the Glan–laser prism is a variant of the Glan–Taylor prism which works only in a narrow range of angles (which is usually no problem when working with low-divergence laser beams), but has lower optical losses and a higher optical damage threshold.
• Those designs using a cement rather than an air gap tend to have lower optical damage threshold.
• Many polarizers have anti-reflection coatings, which function well only in a limited wavelength range.
• Some crystal materials such as BBO allow for operation at particularly short wavelengths in the ultraviolet spectral region, whereas others are well suited for infrared light.

The reason for the use of many different designs of birefringent polarizer designs is that different applications can have quite different requirements on the polarizers, and no design can meet all requirements.

There are different kinds of thin-film polarizers. Thin-film plate polarizers (Figure 3) consist of a dielectric coating on some glass substrate. (Note that the substrate does not need to be birefringent.) For non-normal incidence (in a certain range of incidence angles), the reflectivity of the coating can be strongly polarization-dependent. It is possible to have the “rejected” beam at a deflection angle of 90°, which is often convenient. However, many thin-film plate polarizers are operated at Brewster's angle, so that no anti-reflection coating is required on one side.

There are polarizing cube beam splitters where the dielectric coating is applied to one 45° prism and another 45° prism is cemented to the coating, such that overall one obtains a cube.

Note that they can work only in a limited wavelength range, since the interference effects in the multilayer coating are of course wavelength-dependent. However, operation in a range of few hundred nanometers is possible.

An advantage of thin-film polarizers is that they can be made with rather large dimensions, which is more difficult with crystalline (birefringent) polarizers.

The article on thin-film polarizers gives more details.

Wire grid polarizers are made by fabricating very narrow (sub-wavelength) metal stripes on a glass substrate (using a lithographic technique) or in a free-standing arrangement (for longer wavelengths). Such devices reflect s-polarized light, while p-polarized light is transmitted. They can be used at very high average power levels.

Here you can submit questions and comments. As far as they get accepted by the author, they will appear above this paragraph together with the author’s answer. The author will decide on acceptance based on certain criteria. Essentially, the issue must be of sufficiently broad interest.

Please do not enter personal data here; we would otherwise delete it soon. (See also our privacy declaration.) If you wish to receive personal feedback or consultancy from the author, please contact him e.g. via e-mail.