Optical coating design approaches based on the needle optimization technique

Lomonosov Moscow State University, Moskva, Moscow, Russia
Applied Optics (Impact Factor: 1.69). 03/2007; 46(5):704-10. DOI: 10.1364/AO.46.000704
Source: PubMed

ABSTRACT Design approaches for optical thin films that recognize the key role of a design's total optical thickness are presented. These approaches are based primarily on the needle optimization technique but also utilize other optimization procedures. Using the described design approaches, an optical coating engineer can obtain a set of theoretical designs with different combinations of principal design metrics (merit function value, number of layers, and total design optical thickness); this extends opportunities for choosing the most practical and manufacturable design. We also show that some design problems have multiple solutions with nearly the same combinations of principal design metrics.


Available from: Michael K. Trubetskov, Mar 09, 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper describes a systematic investigation on the use of the genetic algorithm (GA) to accomplish ultra-low reflective multilayer coating designs for optoelectronic device applications. The algorithm is implemented using LabVIEW as a programming tool. The effects of the genetic operators, such as the type of crossover and mutation, as well as algorithm parameters, such as population size and range of search space, on the convergence of design-solution were studied. Finally, the optimal design is obtained in terms of the thickness of each layer for the multilayer AR coating using optimized genetic operators and algorithm parameters. The program is successfully tested to design AR coating in NIR wavelength range to achieve average reflectivity (R) below 10−3 over the spectral bandwidth of 200 nm with different combinations of coating materials in the stack. The random-point crossover operator is found to exhibit a better convergence rate of the solution than single-point and double-point crossover. Periodically re-initializing the thickness value of a randomly selected layer from the stack effectively prevents the solution from becoming trapped in local minima and improves the convergence probability.
    Optics & Laser Technology 07/2015; 70. DOI:10.1016/j.optlastec.2014.12.022 · 1.65 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A new paradigm in the design of optical coatings connected with an outstanding computational efficiency of modern design techniques is discussed. Several other topics including pre-production error analysis, monitoring of coating production, and computational manufacturing of optical coatings are considered.
    IEICE Transactions on Electronics 10/2008; E91C(10). DOI:10.1093/ietele/e91-c.10.1622 · 0.39 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A combined-method based on local optimization and needle optimization is proposed to design broadband dispersion mirrors. The broadband dispersion mirrors are designed by using needle optimization combined with cross application of local optimization and changing high refractive index materials. We combine high efficiency of the local optimization and accuracy of needle optimization. This method can efficiently prevent optimizing process from being captured in a local minimum and obtain higher reflectivity and more smooth group delay dispersion (GDD) curves. The designed dispersion mirrors have a few numbers of layers, which do not have the thin layers that are not easily deposited.
    Optik - International Journal for Light and Electron Optics 02/2014; 125(3):939–941. DOI:10.1016/j.ijleo.2013.07.147 · 0.77 Impact Factor