skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Slanted-wall beam propagation.

Abstract

We present a new algorithm for wide-angle propagation through a general class of optical-waveguide structures defined by dielectric interfaces that may be slanted with respect to the direction of propagation. No regularity of the structure shapes is assumed, no coordinate transformations are employed, and the movement of each grid point between propagation steps is arbitrary within modest angular limitations. When used with an appropriate grid-generation algorithm, this method allows the modeling of an extremely wide variety of high-index-contrast waveguide structures, including meanders and tapers, with good phase accuracy and energy conservation.

Authors:
Publication Date:
Research Org.:
Sandia National Laboratories
Sponsoring Org.:
USDOE
OSTI Identifier:
909910
Report Number(s):
SAND2007-0132J
TRN: US200723%%268
DOE Contract Number:
AC04-94AL85000
Resource Type:
Journal Article
Resource Relation:
Journal Name: Proposed for publication in the IEEE Journal of Lightwave Technology.
Country of Publication:
United States
Language:
English
Subject:
36 MATERIALS SCIENCE; ACCURACY; ALGORITHMS; DIELECTRIC MATERIALS; ENERGY CONSERVATION; SIMULATION; TRANSFORMATIONS; WAVEGUIDES

Citation Formats

Hadley, G. Ronald. Slanted-wall beam propagation.. United States: N. p., 2007. Web. doi:10.1109/JLT.2007.901522.
Hadley, G. Ronald. Slanted-wall beam propagation.. United States. doi:10.1109/JLT.2007.901522.
Hadley, G. Ronald. Mon . "Slanted-wall beam propagation.". United States. doi:10.1109/JLT.2007.901522.
@article{osti_909910,
title = {Slanted-wall beam propagation.},
author = {Hadley, G. Ronald},
abstractNote = {We present a new algorithm for wide-angle propagation through a general class of optical-waveguide structures defined by dielectric interfaces that may be slanted with respect to the direction of propagation. No regularity of the structure shapes is assumed, no coordinate transformations are employed, and the movement of each grid point between propagation steps is arbitrary within modest angular limitations. When used with an appropriate grid-generation algorithm, this method allows the modeling of an extremely wide variety of high-index-contrast waveguide structures, including meanders and tapers, with good phase accuracy and energy conservation.},
doi = {10.1109/JLT.2007.901522},
journal = {Proposed for publication in the IEEE Journal of Lightwave Technology.},
number = ,
volume = ,
place = {United States},
year = {Mon Jan 01 00:00:00 EST 2007},
month = {Mon Jan 01 00:00:00 EST 2007}
}
  • Recently, a new algorithm for wide-angle beam propagation was reported that allowed grid points to move in an arbitrary fashion between propagation planes and was thus capable of modeling waveguides whose widths or centerlines varied with propagation distance. That algorithm was accurate and stable for TE polarization but unstable for wide-angle TM propagation. This deficiency has been found to result from an omission in one of the wide-angle terms in the derivation of the finite-difference equation and is remedied here, resulting in a complete algorithm accurate for both polarizations.
  • The efficiency of trapping ions in an electron-beam ion source (EBIS) is of primary importance for many applications requiring operations with externally produced ions: RIA breeders, ion sources, and traps. At the present time, the most popular method of ion injection is pulsed injection, when short bunches of ions get trapped in a longitudinal trap while traversing the trap region. Continuous trapping is a challenge for EBIS devices because mechanisms which reduce the longitudinal ion energy per charge in a trap (cooling with residual gas, energy exchange with other ions, and ionization) are not very effective, and accumulation of ionsmore » is slow. A possible approach to increase trapping efficiency is to slant the mirror at the end of the trap which is opposite to the injection end. A slanted mirror will convert longitudinal motion of ions into transverse motion, and, by reducing their longitudinal velocity, prevent these ions from escaping the trap on their way out. The trade-off for the increased trapping efficiency this way is an increase in the initial transverse energy of the accumulated ions. The slanted mirror can be realized if the ends of two adjacent electrodes, drift tubes, which act as an electrostatic mirror, are machined to produce a slanted gap, rather than an upright one. Applying different voltages to these electrodes will produce a slanted mirror. The results of two-dimensional (2D) and three-dimensional (3D) computer simulations of the ion injection into an EBIS are presented using simplified models of an EBIS with conical (2D simulations) and slanted (3D simulations) mirror electrodes.« less
  • Electron injection into a graphene sheet through a slanted armchair graphene nanoribbon (AGNR) is investigated. An incident mode, or subband, in the AGNR is valley-unpolarized. Our attention is on the valley-resolved nature of the injected electron beams and its connection to the incident mode. It is known for a normal injection that an incident mode will split symmetrically into two valley-resolved beams of equal intensity. We show, in contrast, that slanted injections result in asymmetric valley-resolved beam splitting. The most asymmetric beam splitting cases, when one of the valley-resolved beams has basically disappeared, are found and the condition derived. Thismore » is shown not due to trigonal warping because it holds even in the low incident energy regime, as long as collimation allows. These most asymmetric beam splitting cases occur at energies within an energy interval near and include the subband edge of an incident mode. The physical picture is best illustrated by a projection of the slanted AGNR subband states onto that of the 2D graphene sheet. It follows that the disappearing of a valley-resolved beam coincides with the situation that the group velocities of the projected states in the corresponding valley are in backward directions.« less
  • Level densities and their energy dependences for nuclei in the mass range of 47 {<=} A {<=} 59 were determined from the results obtained by measuring neutron-evaporation spectra in respective (p, n) reactions. The spectra of neutrons originating from the (p, n) reactions on {sup 47}Ti, {sup 48}Ti, {sup 49}Ti, {sup 53}Cr, {sup 54}Cr, {sup 57}Fe, and {sup 59}Co nuclei were measured in the proton-energy range of 7-11 MeV. These measurements were performed with the aid of a fast-neutron spectrometer by the time-of-flight method over the base of the EGP-15 pulsed tandem accelerator installed at the Institute for Physics andmore » Power Engineering (Obninsk, Russia). A high resolution of the spectrometer and its stability in the time of flight made it possible to identify reliably discrete low-lying levels along with the continuum part of neutron spectra. Our measured data were analyzed within the statistical equilibrium and preequilibrium models of nuclear reactions. The respective calculations were performed with the aid of the Hauser-Feshbach formalismof statistical theory supplemented with the generalized model of a superfluid nucleus, the back-shifted Fermi gas model, and the Gilbert-Cameron composite formula for nuclear level densities. Nuclear level densities for {sup 47}V, {sup 48}V, {sup 49}V, {sup 53}Mn, {sup 54}Mn, {sup 57}Co, and {sup 59}Ni and their energy dependences were determined. The results are discussed and compared with available experimental data and with recommendations of model-based systematics.« less
  • This paper considers an intense nonneutral charged particle beam propagating in the z-direction through a periodic focusing quadrupole magnetic field with transverse focusing force, -{kappa}{sub q}(s)[xe{sub x}-ye{sub y}], on the beam particles. Here, s={beta}{sub b}ct is the axial coordinate, ({gamma}{sub b}-1)m{sub b}c{sup 2} is the directed axial kinetic energy of the beam particles, q{sub b} and m{sub b} are the charge and rest mass, respectively, of a beam particle, and the oscillatory lattice coefficient satisfies {kappa}{sub q}(s+S)={kappa}{sub q}(s), where S is the axial periodicity length of the focusing field. The particle motion in the beam frame is assumed to bemore » nonrelativistic, and the Vlasov-Maxwell equations are employed to describe the collisionless nonlinear evolution of the distribution function f{sub b}(x,y,x{sup '},y{sup '},s) and the (normalized) self-field potential {psi}(x,y,s)=q{sub b}{phi}(x,y,s)/{gamma}{sub b}{sup 3}m{sub b}{beta}{sub b}{sup 2}c{sup 2} in the transverse laboratory-frame phase space (x,y,x{sup '},y{sup '}), assuming a thin beam with characteristic radius r{sub b}<<S. It is shown that collective processes and the nonlinear transverse beam dynamics can be fully simulated in a compact Paul trap configuration in which a long nonneutral plasma column (L>>r{sub p}) is confined axially by applied dc voltages V=const. on end cylinders at z={+-}L, and transverse confinement in the x-y plane is provided by segmented cylindrical electrodes (at radius r{sub w}) with applied oscillatory voltages {+-}V{sub 0}(t) over 90 deg. segments. Here, V{sub 0}(t+T)=V{sub 0}(t), where T=const. is the oscillation period, and the oscillatory quadrupole focusing force on a particle with charge q and mass m near the cylinder axis is -m{kappa}{sub q}(t)[xe{sub x}-ye{sub y}], where {kappa}{sub q}(t){identical_to}8qV{sub 0}(t)/{pi}mr{sub w}{sup 2}. This configuration offers the possibility of simulating intense beam propagation over large distances in a compact configuration which is stationary in the laboratory frame.« less