LOW-LIGHT-LOSS INTERFERENCE POLARIZER

When light passes through polarizing devices, a considerable fraction of the light energy is usually lost. The authors of the paper under review*) describe a device that makes it possible to reduce these losses to a significant extent and to achieve a very nearly complete separation of the light beam into two mutually perpendicular polarized components. This thereby creates the possibility of simultaneously measuring the intensities of both components and, consequently, of determining the degree of polarization as the result of a single measurement.

The device is an ordinary pile of alternating layers with high and low refractive indices, positioned relative to the primary beam at Brewster’s angle. Its distinctive feature is that the layer thicknesses are chosen in such a way that the rays reflected from successive interfaces reinforce one another by interference. This makes it possible substantially to increase the reflection coefficient of the pile and to make it close to unity.

In practice such a device can be made by successively depositing the layers on the base of a three-sided equilateral right-angled glass prism. As the material for forming the layers, the authors recommend zinc sulfide and cryolite. After the layers have been deposited, the pile is covered with a glass prism identical to the prism serving as the base (Fig. 1).

The authors fabricated a pile with an area of 16 cm², consisting of 10 layers. Measurements showed that the degree of polarization in the transmitted beam was 99.9%, and in the reflected beam about 99%; moreover, the reflection coefficient was close to 0.95. The total intensity of both beams reached 84% of the intensity of the unfiltered radiation, the losses (16%) being attributable mainly to the surfaces of the glass prism. The incomplete polarization in the reflected beam is a consequence of the inexact fulfillment of Brewster’s conditions both at the boundaries of the glass prism with the pile and as a result of the nonparallelism

) H. Schröder and R. Schläfer, Zeits. f. Naturforschung 4a*, 576 (1948).

of the light beam. The authors note that, when the stack is illuminated with white light, the polarized components emerge from the stack slightly colored.

If the stack is used not for measuring polarization but for the purpose of producing polarized radiation, then, with the aid of the device shown in Fig. 2, it is possible to obtain practically a twofold increase in the polarized light flux (without increasing its intensity). The idea of the device is extremely simple and needs no explanation.

Fig. 1.

Fig. 2.

The authors do not touch upon the technology of depositing the layers that form the stack.

V. Yur’ev