thomas f krauss university of st. andrews, supa, school of ... · “room-temperature emission at...
TRANSCRIPT
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.1/60
Thomas F Krauss
University of St. Andrews, SUPA, School of Physics and Astronomy, St Andrews, UK
Integrated Nanophotonics: Issues and opportunities
Key collaborators: Pavia & Catania, Italy
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.2/60
St Andrews
Scotland’s First University
Thomas F Krauss
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.3/60
! Scotland, UK ! 18,000 people (1/3 students) ! 80km from Edinburgh ! 600 km from London
Glasgow Edinburgh
St Andrews Dundee
Aberdeen
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.4/60
St Andrews. The home of golf
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.5/60
LEO 1530 SEM/ Raith Elphy Plus!
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.6/60 CAIBE machine!
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.7/60
Photonic crystal waveguides
220 nm Si waveguide, airbridge or oxide clad epixnet nanostructuring platform: www.nanophotonics.eu
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.8/60
Silicon Photonics
What’s the problem silicon photonics is trying to solve ?
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.9/60
55 W-hour battery stores the energy of
1/2 a stick of dynamite.
If battery short-circuits, catastrophe is possible ...
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.10/60
fJ fJ pJ
Optical interconnects save energy
fJ fJ fJ ?
Optical Interconnect
Electrical Interconnect
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.11/60
On-chip Optical Interconnects
Use the photonics layer to shift data in a multicore architecture
Peter Kogge, DARPA study on Exascale Computing
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.12/60
Peter Kogge, DARPA study on Exascale Computing
Possible monolithic realisation
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.14/60
Optical interconnects
INTEL IBM
The idea is to use optical signals to distribute information on-chip, between processors.
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.15/60
fJ fJ pJ
fJ fJ fJ
Optical interconnects save energy
Optical Interconnect
Electrical Interconnect
External lightsource + modulator
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.16/60
External Lightsource + Modulator
!"
Requirements: Fast (ideally ps) => Carrier modulation =>Typical #n!10-4
=>Typical length ! cm
!" = !
k0!nL = ! # L = "2!n
$ 104"
Intel
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.18/60
Solution: Resonant enhancement. Effective optical path maintained. Electrical path reduced.
++ +
+
+ - - - - -
Volume of active carriers reduced.
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.19/60
Solution: Resonant enhancement. Effective optical path maintained. Electrical path reduced.
++
- -
Volume of active carriers reduced
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.20/60
Kotura ring: 30"m dia. = 100"m circumf. 50 fJ/bit 10 GHz bandwidth Tuning energy: > 100 fJ/bit Tolerances ?
OpEx 17, 22484 (2009)
State-of-the-Art
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.21/60
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.22/60
fJ fJ pJ
fJ fJ fJ
Optical interconnects save energy
Optical
Electrical
+ #$"
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.23/60
Photonic crystal waveguides
220 nm Si waveguide, airbridge or oxide clad
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.24/60
Slow Light Mechanism!
a!
In the slow light regime, one can imagine the mode taking a longer route - that’s why it takes more time, and why there is more light inside the structure.!
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.25/60
PhC MZI modulator
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.26/60
Thermally activated Mach-Zehnder
80 "m long PhC
L. O'Faolain et al., IEEE Photonics Journal 2, 404 (2010)
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.27/60
Dispersion curve
Switching performance
PhC – MZI Performance
L. O'Faolain et al., IEEE Photonics Journal 2, 404 (2010)
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.28/60
n i p
Electrical operation. Work in progress…..
Electrical operation
William Whelan-Curtin Kapil Debnath
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.29/60
Conclusion modulator
The slow light concept allows us to make small footprint, low driving power modulators with high bandwidth. They are of similar size, and therefore capacitance, as State-of-the-Art microring resonators, but offer far more bandwidth and do not need to be tuned.
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.30/60
fJ fJ pJ
fJ fJ fJ
2. Solution: Internal lightsource
Optical Interconnect
Electrical Interconnect
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.31/60
Defect luminescence
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.32/60
PL from SOITEC wafer
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.33/60
PL from SOITEC wafer + PhC Cavity
“Room-temperature emission at telecom wavelengths from silicon photonic crystal nanocavities”, R. Lo Savio et al., Appl. Phys. Lett. 2011
300x
300x enhancement !! Why 300x ? a) Purcell effect b) Extraction
efficiency
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.34/60
Nature News & Views, 1997
FP =3!3
4" 2QV
!rad =" nonrad
" radFP
+" nonrad
E. M. Purcell, Phys. Rev. 69, 37 (1946).
Purcell effect
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.35/60
Ti! f =2!h
f |H ' | i2"
Fermi’s Golden Rule
Matrix element Density of states
The Purcell factor relates to Fermi’s Golden Rule The transition between two quantum mechanical states is given by the product of the matrix element (derived from the Hamiltonian) and the density of final states. This transition probability is also called decay probability and is related to mean lifetime.
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.36/60
%,!"FP =
3!3
4" 2QV
Ti! f =2!h
f |H ' | i2"
%,!"
i
f
To enhance the interaction between a cavity and an emitter, they need to agree in emission wavelength and be in the same space -> Q/V
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.37/60
Fp =!
"cavity
"membrane
!cavity = 0.9
!membrane = 0.036
! = 300
Fp =12, !ext = 25
Extraction efficiency
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.38/60
The Purcell-factor makes defect emission “Room-temperatureable”
!rad =" nonrad
" radFP
+" nonrad
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.39/60
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.40/60
Using hydrogen treatment and well-designed designed photonic crystal cavities, we can achieve significant light emission directly from silicon. This is not yet sufficient for optical interconnects, but further improvements are possible. The output power is competitive with comparable III-V devices,. although not with III-V materials as such.
Conclusion light sources
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.41/60
2. Silicon Nanophotonics for Biosensors
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.42/60
The Lab on a chip (LoC) concept aims to realise biochemical analysis/synthesis in a miniaturised format.
Lab on a chip (LoC)
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.43/60
Surface Plasmon sensor!
Very high sensitivity -> Biacore Broad resonance
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.44/60
Silicon ring resonator as biodetector!
K. deVos, R. Baets et al., OPTICS EXPRESS 15, pp. 7610-7615 (2007).!
Lower sensitivity Narrow resonance -> Genalyte
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.45/60
#$"%&"
High sensitivity Narrow resonance -> ???
Di Falco, Krauss et al., APL 2009
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.46/60
Genalyte
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.47/60
Light source Spectrum Analyser
Chip in a lab
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.48/60
The possible integration of silicon light sources would lead to miniaturisation, large scale integration and simplicity.
Source
Transducer
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.49/60
Light source Spectrum Analyser
Lab on a chip
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.50/60
Autonomous sensors
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.51/60
2nd problem: Diffusion time
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.52/60
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.53/60
H Schmidt & AR Hawkins, Nature Photonics August 2011
Optofluidic laser
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.54/60
H Schmidt & AR Hawkins, Microfluidics & Nanofluidics (2008)
Fluorescence correlation spectroscopy (FCS), FRET,… Measure changes caused by molecular binding, not using the surface.
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.55/60
As it does not have volume, only surface, its entire structure is exposed to its environment and responds to any molecule that touches it. This makes it a good material for super-sensors capable of detecting single molecules of toxic gases.
Graphene
!"µ"Microphotonics S
t. A
ndre
ws
TF Krauss, Pavia March 2012 No.56/60
Key issues Biology 1. Lab on a chip – chip on a lab: Integration.
2. Move away from surface affinity biosensor. Novel integrated sensor architectures.