Polarization measurements for atomic and molecular emissions in the vacuum ultraviolet

The various factors involved in making measurements of the polarization of weak atomic and molecular emissions in the vacuum ultraviolet are discussed. Materials in current use are evaluated with particular emphasis on the most important sources of systematic error. For measurements of circular polarization these involve possible misalignment of the retarder as well as uncertainty in the phase shift itself. The circular polarization of Lyman alpha is considered as a particular example. For measurements of the linear polarization fraction, the need for selection of devices with high polarizance is stressed, together with care in choice of materials and operating environment to minimize the risk of temporal drift of device characteristics. Errors relating to the actual detection of optical signals are also considered.