-
Time-domain topology optimization of power dissipation in dispersive dielectric and plasmonic nanostructures
Authors:
Johannes Gedeon,
Izzatjon Allayarov,
Antonio Calà Lesina,
Emadeldeen Hassan
Abstract:
We present a density-based topology optimization scheme for locally optimizing the electric power dissipation in nanostructures made of lossy dispersive materials. By using the complex-conjugate pole-residue (CCPR) model, we can accurately model any linear materials' dispersion without limiting to specific material classes. We incorporate the CCPR model via auxiliary differential equations (ADE) i…
▽ More
We present a density-based topology optimization scheme for locally optimizing the electric power dissipation in nanostructures made of lossy dispersive materials. By using the complex-conjugate pole-residue (CCPR) model, we can accurately model any linear materials' dispersion without limiting to specific material classes. We incorporate the CCPR model via auxiliary differential equations (ADE) into Maxwell's equations in the time domain, and formulate a gradient-based topology optimization problem to optimize the dissipation over a broad spectrum of frequencies. To estimate the objective function gradient, we use the adjoint field method, and explain the discretization and integration of the adjoint system into the finite-difference time-domain (FDTD) framework. Our method is demonstrated using the example of topology optimized spherical nanoparticles made of Gold and Silicon with an enhanced absorption efficiency in the visible-ultraviolet spectral range. In this context, a detailed analysis of the challenges of topology optimization of plasmonic materials associated with a density-based approach is given.
△ Less
Submitted 8 July, 2024;
originally announced July 2024.
-
Free-form inverse design of arbitrary dispersive materials in nanophotonics
Authors:
Johannes Gedeon,
Emadeldeen Hassan,
Antonio Calà Lesina
Abstract:
In the last decades nanostructures have unlocked myriads of functionalities in nanophotonics by engineering light-matter interaction beyond what is possible with conventional bulk optics. The space of parameters available for design is practically unlimited due to the large variety of optical materials and geometries that can be realized by nanofabrication techniques. Thus, computational approache…
▽ More
In the last decades nanostructures have unlocked myriads of functionalities in nanophotonics by engineering light-matter interaction beyond what is possible with conventional bulk optics. The space of parameters available for design is practically unlimited due to the large variety of optical materials and geometries that can be realized by nanofabrication techniques. Thus, computational approaches are necessary to efficiently search for optimal solutions. In this paper, we enable the free-form inverse design in 3D of linear optical materials with arbitrary dispersion and anisotropy. This is achieved by implementing the adjoint method based on the complex-conjugate pole-residue pair model within a parallel finite-difference time-domain solver, suitable for high-performance computing systems. Our method is tested on the canonical nanophotonic problem of field enhancement in a gap region. The obtained free-form designs of metallic and dielectric materials satisfy the fundamental curiosity of how optimized nanostructures look like in 3D. Unconventional free-form designs revealed by our method, although may be challenging or unfeasible with current technology, bring new insight into how light interacts with nanostructures, and could provide new ideas to inspire forward design.
△ Less
Submitted 29 April, 2023;
originally announced May 2023.
-
Machine learning the derivative discontinuity of density-functional theory
Authors:
Johannes Gedeon,
Jonathan Schmidt,
Matthew J. P. Hodgson,
Jack Wetherell,
Carlos L. Benavides-Riveros,
Miguel A. L. Marques
Abstract:
Machine learning is a powerful tool to design accurate, highly non-local, exchange-correlation functionals for density functional theory. So far, most of those machine learned functionals are trained for systems with an integer number of particles. As such, they are unable to reproduce some crucial and fundamental aspects, such as the explicit dependency of the functionals on the particle number o…
▽ More
Machine learning is a powerful tool to design accurate, highly non-local, exchange-correlation functionals for density functional theory. So far, most of those machine learned functionals are trained for systems with an integer number of particles. As such, they are unable to reproduce some crucial and fundamental aspects, such as the explicit dependency of the functionals on the particle number or the infamous derivative discontinuity at integer particle numbers. Here we propose a solution to these problems by training a neural network as the universal functional of density-functional theory that (i) depends explicitly on the number of particles with a piece-wise linearity between the integer numbers and (ii) reproduces the derivative discontinuity of the exchange-correlation energy. This is achieved by using an ensemble formalism, a training set containing fractional densities, and an explicitly discontinuous formulation.
△ Less
Submitted 30 June, 2021;
originally announced June 2021.
