1 -1/29- Applications and Sustainability Functionality in Nanophotonics Daniel Erni General and Theoretical Electrical Engineering (ATE), Faculty of Engineering, University of Duisburg-Essen, and CENIDE Center for Nanointegration Duisburg-Essen, University of Duisburg-Essen, D-47048 Duisburg The Interface Problem How is a functional nanophotonic device accessed by its environment ? Large scale differences. How to bridge the gap between the nano and the micro/macro? -2/29-
16
Embed
Applications and Sustainability Functionality in Nanophotonics€¦ · in Nanophotonics Daniel Erni General and Theoretical Electrical Engineering (ATE), Faculty of Engineering, University
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
-1/29-
Applications and Sustainability �–
Functionality in Nanophotonics
Daniel Erni
General and Theoretical Electrical Engineering (ATE), Faculty of Engineering, University of Duisburg-Essen, and CENIDE �– Center for Nanointegration Duisburg-Essen, University of Duisburg-Essen, D-47048 Duisburg
The Interface Problem How is a functional nanophotonic device accessed by its environment ?
Large scale differences.
How to bridge the gap between the nano and
the micro/macro?
-2/29-
2
The Functionality-vs-Volume Problem
Functionality is provided by optical signal processing within a nano volume (respective a sub-wavelength volume).
Open question: How does the complexity of the functionality scales with decreasing volume?
How is nanoscopic functionality implemented and exploited?
Is there a degradation of functionality for decreasing nanophotonic device volumes ?
Functionality ( ) confined to a single site (volume).
Functionality ( ) encoded into an anomaly (defect).
Functionality ( ) dispersed over the structure (dilution).
How is the nanophotonic functionality actually provided ?
There are 3 typical paradigms of implementation characterized by the structural length scale relative to the operating wavelength .
Ldevice
Lunit cell Lmicro structure
-4/29-
3
Agenda Squeezing light into the nanoscale.
On tight light guiding.
Light confinement to metal surfaces: «Plasmonics».
Optical nanoantennas: «Nantennas».
Photonic crystal devices.
Electromagnetic / optical metamaterials.
Few concluding remarks.
«On the implementation of functionality in nanophotonic device design»
-5/29-
Squeezing Light Into Small Scales
Example: Tight light guiding for dense optical integration.
The very first task: light confinement
-6/29-
4
-7/29-
straight wave guide
R = 200 µm
= 1.3 µm / n = 0.1
R = 50 µm
R = 10 µm
On Tight Light Guiding I
20 µm
(A) Electronic chip
100 µm
(B) Photonic chip (conventional)
1250 couplers/cm2 (R = 200 µm)
5.7·108 transistors/cm2 (8-core Itanium, 32 nm)
456�‘000 : 1 Minimal radius of curvature fully determines the integration density.
Comparing integration densities
On Tight Light Guiding II Photonic wires
Rib waveguide
2D-MMP: T = 6%
Simulation: X. Cui Fabrication: F. Robin (ETH Zürich)
2D-MMP: T = 99%
Photonic wire
Strong horizontal light guiding.
conventional light guiding.
X. Cui, Ch. Hafner et al., Opt. Expr., 14(10), pp. 4351, 2006. X. Cui, Ch. Hafner, F. Robin, D. Erni, et al., Proc. SPIE vol.
6617, pp. 66170D-1-11, June 2007.
5 µm
5 µm
1550nm
1550nm
InGaAsP/InP
T < �– 4dB
Via Evolution Strategies (ES)
-8/29-
5
-9/29-
On Tight Light Guiding III
Dielectric waveguides:
Light confinenment is the solution of a boundary value problem (cf. total internal reflection at a boundary interface), which translates into an eigenvalue problem.
Light confinement increases with increasing refractive index contrast.
Tight light guiding needs new confinement respective new guiding mechanisms.
A first conclusion
(1) Metallic boundaries: Plasmonics.
(2) Alternative mechanism: Defect waveguiding in
photonic crystals.
Light Confinement to a Metal Surface
(1) «Field-driven» plasma resonance:
oscillating field (light)
Oscillating carriers
Surface Plasmons
Charge carriers (electrons) have mass and thus inertia. Resonant system between
electric field electrons. The light field is «glued»
to the (lossy) metal surface.
(2) Dispersion relation:
SPP: surface plasmon polariton («glued», i.e. guided Zenneck wave)
SP: surface plasmon (guided slow wave up to localized resonance). plasmon freq. / P : SP = P
1+
-10/29-
6
T = 77.2%
Simulation
Transmission in the «Nano» I Plasmonic light guiding
(2) Metallic groove waveguide:
S. I. Bozhevolnyi, et al., Nature, 440, pp. 508-511, March, 2006.
(1) Metallic slot waveguide: L. Liu, et al., Opt. Express, 13(17),
pp. 6645-6650, Aug. 15, 2005.
Measurement
100 nm 20 nm Metall
Metal
-11/29-
Transmission in the «Nano» II Plasmonic band-stop filter
Planar crystal: strong periodic 2D perturbation. No propagation states allowed within the PBG. Introduction of a defect confined field states.
Line defect «encodes» a channel waveguide.
-20/29-
11
Photonic Crystals II Optimal design of a PhC waveguide bend Wrestling around even with simple device designs
modeling
end-fire spectra
P. Strasser, D. Erni, et al., J. Opt. Soc. Am. A., vol. 25, no. 1, pp. 67, Jan. 2008.
Lossy 2D model (FEM) for the hole-type PhC waveguide bend.
Optimization of the bending area in 2D.
Verification in 3D (FDTD). Fabrication in InP/InGaAsP; end-fire characterization. Transmission: �– 8 dB �– 3 dB , bandwidth doubled.
A: upper single-mode region
425 nm
-21/29-
-22/29-
1
2
92.7 %
90.7 %
Photonic Crystals III Filtering T-junction (diplexer)
J. Smajic, Ch. Hafner, and D. Erni, Opt. Express, vol. 11, no. 6, pp. 567-571, March 24, 2003.
E. Moreno, D. Erni, and Ch. Hafner, Phys. Rev. E, vol. 66, no. 3, pp. 036618-1-12, Sept. 27, 2002.
Size: 7.5 µm × 5.0 µm (@ = 1.55 µm). Smallest diplexer topology at that time.
Si rod in air / a = 575 nm
Si rod in air a = 575 nm
12
K. Rauscher, D. Erni, W. Bächtold, OWTNM 2005, April 8-9, Grenoble, France, 2005. P. M. Nellen, P. Strasser, V. Callegari, R. Wüest, D. Erni, and F. Robin, Microelectronic
Engineering (MEE), vol. 85, no. 5-8, pp. 1244-1247, 2007. Photonic Crystals IV Compact functional devices (1) Power splitter:
J. Sun, et al, Nature, vol. 493, pp. 195-199, Jan. 2013.
-28/29-
15
Thanks. Further Information:
www.ate.uni-due.de
Check our site on «publications»
-29/29-
Appendix
16
-A1-
Photonic Crystals II Is the PhC scheme apt for functional device design?
PhC T-junction: (maximal BW)
J. Smajic, Ch. Hafner, and D. Erni, J. Opt. Soc. Am. A, vol. 21, no. 11, pp. 2223-2232. Nov. 2004.
Evolutionary algorithms Sensitivity-based gradient search There are enough degrees of
freedom in the small PhC lattice volume to implement functionality !
(a) binary:
(b) continuous:
Si rod in air a = 1 µm
Active PhC Devices Organic PhC laser
1st order 2nd order
lasing 494 nm (TM)
PL spectrum (pulsed pump at 355 nm)
R. Harbers, P. Strasser, D. Caimi, R. F. Mahrt, N. Moll, D. Erni, W. Bächtold, B. J. Offrein, and U. Scherf, J. Opt. A: Pure Appl. Opt., vol. 8, S273-S277, 2006.