Evaluation of Polarization Dependence of a Spectral Radiance Calibration System in UV and VUV regions

We have been attempting to establish an absolute scale of spectral radiance in ultraviolet and vacuum ultraviolet regions by using synchrotron radiation as a primary standard light source and also attempting to calibrate an under-test light source by comparing the under-test source with the primary standard light source. A calibration system consisting of a focusing mirror, a monochromator and a detector was introduced for the calibration. If the efficiency of the calibration system were constant, the calibration could be done only by comparing the detector signals obtained from measuring each of the light sources. However the efficiency of the calibration system differs with each of the light sources because the primary and under-test light sources differ in terms of polarization state, and the transmittance or reflectance of the optical element depends on incident light polarization and the efficiency of the calibration system, and hence also depends on incident light polarization. Consequently, determining the polarization dependence of the calibration system is necessary for the calibration, as is comparing the detector signals. Furthermore, if the polarization dependence of the focusing mirror or the grating of the monochromator in the calibration system is not uniform, it is necessary to make a map showing the polarization dependence of the calibration system versus the vertical and horizontal angles at which the radiation of each light source enters the calibration system. We evaluated the polarization dependence of the calibration system on the vertical angle and evaluated the uncertainty of the spectral radiance calibration due to the angular distribution