This paper consists essentially of two parts. The first part pertains to the comparison of different roughness evolution models to experimental results for Mo/Si multilayer coated optics. The authors found excellent agreement with the roughness modeling for a multilayer coating on a relatively rough substrate, with 0.34 nm rms, whereas the agreement was not as good when a relatively smooth 0.11 nm rms substrate was coated. They attribute the deviation between experiment and theory for the smooth substrate to the “instrument signature,” which I interpret as something like the quality of the collimation of the incoming beam. They are correct that this collimation will have a more dramatic effect for the sample with a rougher substrate and rougher interfaces than for the smoother sample.
The second essential part of this paper is their claim that 442 nm light can be used to characterize the surface finish of optics, even though light of this high of a wavelength is known to not capture higher spatial frequency roughness, an important parameter in determining scattering in EUV optical systems. I found this part to be the most interesting and most practical component of the paper. They not only assume that the roughness exhibits a fractal-like behavior and postulate the extension of the inverse power law to higher spatial frequencies, but they provide experimental evidence to support the claim that it can be extended. It would clearly be a benefit in the development of suitable condenser optics, as well in the cost of ownership in the production of these optics, if 442 nm light could be used for metrology as opposed to atomic force microscopy or actinic (i.e., 13.5 nm) light. The authors did use optics that were likely polished under similar process conditions, and I wonder if perhaps this study should be extended in the future to test the extendibility of the 442 nm light metrology technique to higher frequencies for optics prepared using different polishing procedures, perhaps those provided by different manufacturers. In general, this paper should definitely be of interest to those in the EUV optics community.
You must log in to add comments.