A cryogenic scatterometer and scatter from particulate contaminants on mirrors.

The scattered light from particulate-contaminated mirrors was measured at visible and far-infrared wavelengths, and results were compared to that predicted by a theoretical model. In addition, a λ = 10.6 μm cryogenic scatterometer was constructed, and the temperature dependence of the scattered radiation was determined. The design and calibration of a λ = 10.6μm cryogenic scatterometer is presented. The BRDF is experimentally observed to have no measurable temperature dependence for the mirrors tested. Scatter data from oil-contaminated mirrors demonstrates some unexpected temperature dependencies, illustrating the necessity to know the temperature dependence of the samples, as well as the contaminants on them. Measurements from particulate-contaminated mirrors and their comparison with a modified Mie theory are presented. The method for cleaning the samples, counting and measuring particles, the measurement procedure, and the theoretical model employed to predict the contaminant scatter are discussed. Measurements and theoretical comparisons of λ = 0.6328μm scatter from distributions of polystyrene spheres on mirrors serve as a stepping stone for the irregularly-shaped dust contaminants. The theory predicts the forward scatter excellently, but is not as successful with the backscatter. Scatter measurements from dust-contaminated mirrors were performed at wavelengths of 0.6328μm and 10.6μm. The theory agrees excellently at the visible wavelength and is in good agreement in the far-infrared. The far-infrared measurements of the clean mirrors were, for angles beyond about 8° from specular, limited by the scatter from particulates and not the surface microroughness. Similar limitations can be experienced by other irregularities such as scratches, digs, and pinholes in coatings. It is shown that contaminant scatter dominance in the infrared requires only a few very small particles. So, the necessity to improve cleaning techniques is quite evident. Several effects of these findings are discussed, as well as other related topics: the cleanliness required for the scatter to be dominated by a mirror's surface microroughness, a modified specification for low-scatter infrared mirrors, incident angle invariance of clean and contaminated mirrors, the shape of the BRDF curves, and the relation between surface cleanliness level, clean room cleanliness class and BRDF