The original selenium polarizer has been found to suffer from two defects: (1) the degree of polarization is not sufficiently high, and (2) the necessary dimensions become unwieldy. Both defects are overcome by resorting to two reflections from parallel selenium mirrors. The efficacy of this procedure is proved by a study of the dichroism of beryl. A discussion of selenium mirrors produced by vacuum distillation follows.
A. H. PFUND, "Improved Infra-Red Polarizer1," J. Opt. Soc. Am. 37, 558-559 (1947)