scispace - formally typeset
Search or ask a question

High-Accuracy Finite-Difference Equations for Dielectric Waveguide Analysis II:

01 Jan 2002-
TL;DR: In this paper, the authors present a finite-difference equation for waveguide eigenmode solution valid for a dielectric corner, which is based on an expansion of the field components as powers of the radius at the corner.
Abstract: We present a discussion of the behavior of the electric and magnetic fields satisfying the two-dimensional Helmholtz equation for waveguides in the vicinity of a dielectric corner. Although certain components of the electric field have long been known to be infinite at the corner, it is shown that all components of the magnetic field are finite, and that finite-difference equations may be derived for these fields that satisfy correct boundary conditions at the corner. These finite-difference equations have been combined with those derived in the previous paper to form a full-vector waveguide solution algorithm of unprecedented accuracy. This algorithm is employed to provide highly accurate solutions for the fundamental modes of a previously studied standard rib waveguide structure. Indeed, with second-order accurate code it may not be worth the trouble to change the way corners are handled. However, for those intent on more accurate simulations, correct treatment of the corner points is essential. The history of the behavior of fields near a dielectric or metal wedge dates to the early work of Bouwcamp and Meixner (2), (3), who pointed out some of the peculiarities of this problem, including field components that diverge weakly at the corner as a small negative power of the distance from the corner (ra- dius). Meixner further published a solution formalism (3) based on an expansion of the field components as powers of the ra- dius. This expansion was later shown to be incorrect by An- derson and Solodukhov (4), and although papers continued to appear, almost ten years elapsed before a correct expansion was published. Makarov and Osipov (5) explained the reason for the problems with Meixner's series and correctly identified the missing terms as containing various powers of the logarithm of the radius. However, the latter authors did not treat a "gen- eral" corner, but considered only a single field component, and left their expansion in a very general form quite unsuitable for the derivation of finite-difference equations. This entire body of work, and in fact the existence of the problem itself, re- mained largely unknown to the optics community until 1992, when Sudbo (6) described the importance of the problem in the modeling of optical waveguide structures. Apparently unaware of Makarov and Osipov's work, he correctly pointed out the in- herent difficulties of including corner effects in any numerical algorithm for modeling waveguides. Since that time only a few authors have attempted to include corner effects in their mod- eling work (7), (8) and no one has attempted a thorough deriva- tion. This paper derives a finite-difference equation for waveguide eigenmode solution valid for a dielectric corner. This equation is ostensibly first-order accurate, and when combined with highly accurate difference equations in the interior regions and at simple planar boundaries (1), has resulted in a highly accurate full-vector waveguide eigenmode solver whose trun- cation error ranges from second to third order. This departure from the expected third-order behavior apparently results from the presence of infinite derivatives at the corner and will be discussed in more depth below. The present work may be viewed as an extension of the work of Makarov and Osipov (5), in that the standard solutions of the Helmholtz equation in a uniform region (i.e., Bessel functions and sines and cosines) are augmented by terms containing logarithms that appear similar to their solutions. However, here we consider the full vector
Citations
More filters
Journal ArticleDOI
TL;DR: This work develops a scheme based on a simple criterion taken from perturbation theory that is the only one that attains quadratic convergence with resolution for arbitrarily sloped interfaces and compares it with other published FDTD smoothing methods.
Abstract: Finite-difference time-domain (FDTD) methods suffer from reduced accuracy when modeling discontinuous dielectric materials, due to the inhererent discretization (pixelization). We show that accuracy can be significantly improved by using a subpixel smoothing of the dielectric function, but only if the smoothing scheme is properly designed. We develop such a scheme based on a simple criterion taken from perturbation theory and compare it with other published FDTD smoothing methods. In addition to consistently achieving the smallest errors, our scheme is the only one that attains quadratic convergence with resolution for arbitrarily sloped interfaces. Finally, we discuss additional difficulties that arise for sharp dielectric corners.

463 citations

Journal ArticleDOI
Jan Wiersig1
TL;DR: In this article, a boundary element method based on a Green function technique is introduced to compute resonances with intermediate lifetimes in quasi-two-dimensional dielectric cavities, and the efficiency of the method is demonstrated by calculating resonances in two coupled hexagonal cavities.
Abstract: A boundary element method based on a Green function technique is introduced to compute resonances with intermediate lifetimes in quasi-two-dimensional dielectric cavities. It can be applied to single or several optical resonators of arbitrary shape, including corners, for both TM and TE polarization. For cavities with symmetries a symmetry reduction is described. The existence of spurious solutions is discussed. The efficiency of the method is demonstrated by calculating resonances in two coupled hexagonal cavities.

251 citations

Journal ArticleDOI
TL;DR: Eigenvalue equations for solving full-vector modes of optical waveguides are formulated using Yee-mesh-based finite difference algorithms and incorporated with perfectly matched layer absorbing boundary conditions to calculate the complex propagation constants and the confinement losses of leaky waveguide modes.
Abstract: Eigenvalue equations for solving full-vector modes of optical waveguides are formulated using Yee-mesh-based finite difference algorithms and incorporated with perfectly matched layer absorbing boundary conditions. The established method is thus able to calculate the complex propagation constants and the confinement losses of leaky waveguide modes. Proper matching of dielectric interface conditions through the Taylor series expansion of the fields is adopted in the formulation to achieve high numerical accuracy. The method is applied to the study of the holey fiber with triangular lattice, the two-core holey fiber, and the air-guiding photonic crystal fiber.

113 citations

Journal ArticleDOI
TL;DR: The Flexible Local Approximation MEthods (FLAME) as mentioned in this paper is a special case of the Mehrstellen MEthod, which defines a set of local approximating functions (such as cylindrical or spherical harmonics, plane waves, harmonic polynomials, etc.) on a chosen grid stencil.

74 citations


Cites background or methods from "High-Accuracy Finite-Difference Equ..."

  • ...Harari–Turkel [53] and Singer–Turkel schemes [99] for the Helmholtz equation; the Hadley schemes [50, 51 ] for waveguide analysis; Cole schemes for wave propagation [26,27]; the Lambe–Luczak–Nehrbass schemes for the Helmholtz equation [68]....

    [...]

  • ...Hadley [50, 51 ] for electromagnetic wave propagation....

    [...]

Journal ArticleDOI
Shan Zhao1
TL;DR: This work overcomes the difficulty of the previous matched interface and boundary (MIB) method in dealing with interfaces with non-constant curvatures for optical waveguide analysis by avoiding the use of local cylindrical coordinates in the improved MIB method.

73 citations


Cites background from "High-Accuracy Finite-Difference Equ..."

  • ...The fourth order convergence can not be guaranteed if the interface is only C continuous, because certain corner singularity problems may occur [17,29,38]....

    [...]

References
More filters
Book
J. Meixner1
21 Mar 2018
TL;DR: In this article, the behavior of an electromagnetic field in the neighborhood of the common edge of angular dielectric or conducting regions is determined from the condition that the energy density must be integrable over any finite domain (the so-called edge condition).
Abstract: The behavior of an electromagnetic field in the neighborhood of the common edge of angular dielectric or conducting regions is determined from the condition that the energy density must be integrable over any finite domain (the so-called edge condition). Two cases are treated in detail 1) A region consisting of a conducting wedge and two different dielectric wedges with a common edge. 2) A region consisting of two different dielectric wedges with a common edge. It is also shown that near such edges, electrostatic and magnetostatic fields will exhibit the same behavior as the electromagnetic field.

424 citations

DOI
01 Feb 1988
TL;DR: In this paper, a simple accurate method, which automatically takes full account of the discontinuities in the normal electric field components across any arbitrary distribution of internal dielectric interfaces, is presented for the determination of polarised solutions of the Helmholtz wave equation.
Abstract: A simple accurate method, which automatically takes full account of the discontinuities in the normal electric field components across any arbitrary distribution of internal dielectric interfaces, is presented for the determination of polarised solutions of the Helmholtz wave equation. The application of the shifted inverse power iteration method to the resulting matrices, enables any required propagation eigenvalue (not necessarily the fundamanetal mode) to be determined, together with its corresponding electric field profile. It is found that the results, which are presented for various semiconductor rib waveguide structures, compare favourably with published vector finite element and scalar results.

271 citations

Journal ArticleDOI
TL;DR: In this article, a vector finite element method with the high-order mixed-interpolation-type triangular elements is described for the analysis of optical waveguiding problems, which is a combination of linear edge elements for transverse components of the electric or magnetic field and quadratic nodal elements for the axial one.
Abstract: A vector finite element method with the high-order mixed-interpolation-type triangular elements is described for the analysis of optical waveguiding problems. It is a combination of linear edge elements for transverse components of the electric or magnetic field and quadratic nodal elements for the axial one. The use of mixed-interpolation-type elements provides a direct solution for propagation constants and avoids spurious solutions. This approach can yield more accurate results compared with the conventional approach using the lowest order mixed-interpolation-type elements, namely, constant edge elements and linear nodal elements. The accuracy of this approach is investigated by calculating the propagation characteristics of optical rib waveguides. Results obtained for both E/sup x/ and E/sup y/ polarizations are validated using benchmark results produced by established methods. >

156 citations

Journal ArticleDOI
TL;DR: A review of the theory of field behavior near a dielectric wedge is presented in this paper, where the static results give an accurate account for the rate of growth of the fields in the singular cases, but some disagreement in nonsingular cases.
Abstract: A review of the theory of field behavior near a dielectric wedge is presented. An overview of the static results is given, and it is shown that the series of Meixner for the dynamic case in general is nonexisting. Numerical results for scattering by a dielectric cylinder of square cross section indicate that the static results give an accurate account for the rate of growth of the fields in the singular cases, but some disagreement in nonsingular cases.

137 citations