Abstract

Binary optical elements are being investigated to improve the performance of many optical systems. The critical issue in the design of many of the systems is the diffraction efficiency one can theoretically expect to achieve. The current models of the diffraction efficiency of binary optical elements can be separated into two regimes. The first regime, valid when the periodicities of the element are all much greater than the incident wavelength, is modeled by scalar diffraction. The second regime, based on the electromagnetic theory of diffraction, is theoretically valid for all wavelength to period ratios but is computationally limited to periodicities on the order of a wavelength. Unfortunately, a void is left between the two regimes where practical models exist. This paper presents a non-rigorous model that attempts to fill the void of predicting diffraction efficiencies for moderate wavelength to period ratios. Results of the model are compared to both scalar and electromagnetic results in the limit of large and small wavelength to period ratios.

© 1990 Optical Society of America