RESUMEN
A method has been developed to simulate the effects of scattered light on the image quality of optical systems. The coherent model is based on geometrical optics to take account of wavefront aberrations caused by lenses, applies finite-element calculation to solve Maxwell's equations around small scattering structures such as edges of diffractive surface zones, and uses scalar diffraction for free-space light propagation. The implementation is discussed in detail, and the operation is demonstrated on diffractive intraocular lenses. Point spread and modulation transfer functions are evaluated for an axial object point, taking account of scattered light as a function of slant angle and round radius of diffractive zone edges. Results show that, at a distance of ±200 Airy radius (i.e., ±2.1∘) from the axis, scattered irradiance is about 5 times more than without considering edge effects. Optimum round radius was found to be 7% of the step height, which agrees with simple geometrical optical estimations.
RESUMEN
For the simulation of relief-type diffractive surfaces, an efficient method has been developed and described. Based on zone decomposition, our approach maps the transmitted wavefront by ray tracing, while point spread function/modulation transfer function (PSF/MTF) plots are calculated by scalar diffraction, taking light diffracted into multiple orders into account inherently. Using a parametric user-defined surface, our solution makes the analysis and optimization of diffractive lenses possible directly inside optical design software. Implementation was carried out in ZEMAX in the form of a swift Windows Dynamic Link Library extension using an approximative, non-iterative algorithm. The average computation time increments relative to standard built-in surfaces are 38% and 21% for PSF and MTF calculations, respectively. Application of our method is illustrated by the analysis of diffractive intraocular lenses. For validation, numerical results were compared with analytical formulae and industry-standard measurements. The ray-tracing error caused by our approximation proved to be less than ${7} \cdot {{10}^{ - 6}}$7â 10-6 wavelength, and the difference from the theoretical MTF calculations is 1%-2%. The RMS difference of the simulated-measured through-focus MTF values at 50 lp/mm is 0.031, equaling ${2}\sigma $2σ measuring accuracy.