Depolarization of white light by a birefringent crystal II The Lyot depolarizer

An earlier paper [ S. Lu A. P. Loeber , J. Opt. Soc. Am.65, 248– 251 ( 1975)] investigated the depolarization of white light on passage through a birefringent crystal because of the unequal phase retardation of different wavelengths. Assuming the incident light to be completely polarized with the spectral-power distribution of a blackbody, the degree of polarization remaining after passage through various crystal thicknesses was determined. This paper extends the result of the earlier study to an analysis of the Lyot depolarizer. Earlier work by Billings [ J. Opt. Soc. Am.41, 966– 975 ( 1941)] considered the general case of two crystal retardation plates of different retardances and orientations to determine what requirements the plates must meet in order to act as a depolarizer. Billings assumed the intensity of the incident light to be independent of wavelength. Despite the different assumptions regarding the spectral-power distribution of the incident light, our investigation agrees with Billings’s conclusions that (1) the crystal retardances of both crystals must be large, (2) the crystal retardances cannot be equal, and (3) the angle between the crystal axes must be 45° for the combination of plates to act as a depolarizer. An additional conclusion by Billings, that the ratio of the two retardances must be an integral, is not indicated by our analysis. Results of our analysis are in qualitative agreement with the experimental work carried out by Billings on several modifications of the usual Lyot depolarizer configuration.