Gravitational Waves: Sources and Detection
Nergis MavalvalaDepartment of Physics
Massachusetts Institute of Technology
Les Houches, August 2015
Lecture #1
Outline
Lecture 1
Gravitational wave (GW) basics
Sources and emission strength
Interferometric GW detectors
Conceptual ideas to actual realizations
Fundamental noise sources
First astrophysical searches
Lecture 2
Second‐generation detectors (currently under
construction)
Key technologies
Major challenges
Outline
Lecture 3
Third‐generation detectors
Quantum technologies
Squeezed states of light
Optomechanics (classical and quantum)
Gravity’s messenger
Understanding gravity
Newton (16th
century)
Universal law of gravitation
Worried about action at a
distance
Einstein (20th
century)
Gravity is a warpage of
space‐time
Matter tells spacetime how
to curve spacetime tells matter how to move
1 22
Gm mFr
4
8 GG Tc
Newton vs. Einstein
K. Glampedakis
GWs in linear gravity
For weak gravitational fields, flat spacetime (Minkowski metric) is perturbed (h)
GR field equations in vacuum reduce to a wave equation
Plane wave solution in weak gravity: transverse wave with two polarizations (+ and x)
Metric perturbation and strain
Minkowski metric in Cartesian coordinates
Linearized perturbation in transverse traceless gauge
Space‐time interval
Quadrupole formula
Derived by Einstein in 1918 by solving linearized field equations with a source term
Accurate for all sources as long as wavelength is much longer than source size R
Generalized source strength
GWs carry energy
The stress‐energy carried by GWs
Stress energy is not localized; a certain amount of stress‐ energy is contained in a region of the space which extends
over several wavelengths
The luminosity is the integral of the stress‐energy tensor over the area of that region of space
This gives
Astrophysical sources of GWs
GW luminosity
Sources need to be compact (R
≈
RSch
) and relativistic (v
≈
c)Lots of compact mass (neutron stars, black holes) Rapid acceleration (orbits, explosions, collisions)
Colliding compact starsBinary NS and BH
Supernovae
The big bangEarliest moments
The unknownLooking back in time
CMB 400 thousand
years
Now 13 billion
years
GWs 0 years
where
A bit of history
Gravitational radiation was first introduced by Einstein in 1916 in his seminal paper on General Relativity
In a subsequent paper in 1918 Einstein gave the first correct
formulation of gravitational waves
But he himself remained uncertain (not just of how immeasurably weak
they are, but of their very existence)
Submitted a retraction in a paper
with Rosen in 1937
Retracted the retraction after
discussion with Infeld and Robertson
Doubts and controversy finally subside after 1957
Experiment and observation have the final say (as usual) …
Gravitational waves ‐‐
the Evidence
PSR 1913 + 16
Two neutron stars orbiting each other at 0.0015c
Compact, dense, fast relativistic system
Emit GWs and lose energy
Used time of arrival of radio pulses to measure change in orbital period due to GW emission
Hulse & Taylor’s Binary Neutron Star System(discovered in 1974, Nobel prize in 1993)
Signal strength of binary pulsars
In our galaxy (21 thousand light years away, 8 kpc)
h ~ 1018
In the Virgo cluster of galaxies (50 million light
years away, 15 Mpc)
h ~ 1021
Typical binary pulsar at the end of its lifetime (100 million years from now)
M M
r
R
30
23
10 kg20 km200 Hz
10 m
MRfr
Effect on test particles
Place a pair of test particles at distances ±
x0
from the origin of a Cartesian coordinate system
Assume GW traveling in +z‐direction
Separation between particles given by
Similarly, for a pair of particles on the y‐axis
Key properties of GWs
Ripples in space‐time or propagating curvature perturbation
Propagate at speed of light
Stretch and squeeze the space transverse to
direction of propagation
Amplitude falls as 1/r
Two polarizations (+ and x)
Lowest allowed multipole is quadrupole
Emitted
by
compact rapid accelerating masses
Astrophysics with GWs vs. Light
Very different information, mostly mutually exclusive
Difficult to predict GW sources based on EM observations
Light GWAccelerating charge Accelerating mass
Images (pretty pictures) Waveforms (pretty sounds)
Absorbed, scattered, dispersed by matter
Very small interaction; matter is transparent
100 MHz and up 10 kHz and down
Gravitational wave detection using precision interferometry
Simple concept, challenging implementation
Make mirrors that are very still
Vibration isolation and thermal fluctuation control
Probe the mirror positions using laser light
Ultra-high precision optical measurement
Manipulate quantum fluctuations of the light
LaserLaser
GW detector at a glance
Optical cavities•
Mirrors facing each other
•
Builds up light powerLots of laser power P• Signal α
P
• Noise α10 W
20 kW
P
4 km
Mirrors hang as pendulums•
Quasi-free particles
•
Respond to passing GW•
Filter external force noise
Global network of detectors
GEO600 (HF)2011
Advanced LIGO Hanford 2015
Advanced LIGO Livingston 2015
Advanced Virgo2015 LIGO-India
2022
KAGRA2017
Why a global network?
Angular response is pure quadrupole
Nearly omni‐directional
Earth transparent to GW
Pinpoint sources in sky by triangulation
Localization depends strongly on
SNR and number of detectors
Large duty factor (fraction of time the network has high sensitivity)
Need five sites to get 4 detectors
operational ~85% of the time
Localization with LIGO and Virgo
Localization with LIGO, Virgo and India
10 kg Fused Silica25 cm diameter
10 cm thick
21
1810 4000
~ 10 meters
GWL h L
First phase of LIGO
Shot noiseSNR Power
Seismic noise
Thermal noise
End of lecture 1
We ended with the Initial LIGO noise curve. Tomorrow we will pick up where we left off, starting with limiting noise sources, and
moving on to Advanced LIGO.
Essential or helpful readings and viewings
General Relativity (GR) and Gravitational Waves (GWs)
Einstein (1916): Annalen
der
Physik
49, 769–822 (1916)
Schutz
(1984): Gravitational waves on the back of an envelope, Am. J. Phys. 52, 412
Cutler and Thorne (2002): http://arxiv.org/pdf/gr‐qc/0204090v1.pdf
Astrophysical sources and rates (2010): CQG 27, 173001
Caltech graduate course on GWs: Physics 237 http://elmer.tapir.caltech.edu/ph237/
GR meets precision laser interferometry
Weiss, RLE Report (1972): http://www.hep.vanderbilt.edu/BTeV/test‐
DocDB/0009/000949/001/Weiss_1972.pdf
Saulson book (1990): Fundamentals of interferometric gravitational wave detectors, Singapore Press
Fritschel/LSC (2015): “Advanced LIGO,”
Class. and Quant. Grav. 32, 074001 (http://lanl.arxiv.org/pdf/1411.4547.pdf)
Low frequency noises (<100 Hz)
Vibration isolation
Matichard review (2015):
http://authors.library.caltech.edu/56793/2/1407.6377v1.pdf
Mirror suspensions
Shapiro Ph.D. thesis (2014)
Squeezed film damping
Vitale et al. (2012)
Martynov Ph.D. thesis
Seismic noise regression
Driggers et al. PRD 86, 102001 (2012)
Thermal noise
Callen and Welton (1951): Phys. Rev. 83, 34
Saulson (1990): PRD 42, 2437
Harry, Bodiya and DeSalvo (2012): Optical coatings and thermal noise in precision measurement, Cambridge U. Press
Steinlechner et al. (2015): PRD 91, 042001
Yam et al. (2015): PRD 91, 042002
Quantum optics in GW detectors
Caves (1980): PRL 45, 75
Caves (1980): RMP 52, 341
Caves (1981): PRD 23, 1693
Kimble et al. (2001): PRD 65,
Schnabel et al. review (2010): Nat. Comm
McClelland et al. review (2011): Lasers and Photonics
LIGO Scientific Collaboration (2012): Nature Physics
LIGO Scientific Collaboration (2013): Nature Photonics
Optomechanics in GW detectors
Braginsky et al. (1967 and 1970): JETP 25, 30
and 31
Caves (1980): PRL 45, 75
Caves (1980): RMP 52, 341
Caves (1981): PRD 23, 1693
Buonanno and Chen (2002): PRD 65, 042001
Y. Chen review (2013): http://arxiv.org/pdf/1302.1924.pdf
Optomechanics experiments in GW‐land
Sheard et al. (2004): PRA 69, 051801
Corbitt et al. (2006): PRA 74, 021802
Miyakawa et al. (2006): PRD 74, 022001
Corbitt et al. (2007): PRL 98, 150802 and PRL 99,
LIGO Scientific Collaboration (2009): NJP
Neben et al. (2012): NJP 23, 1693
Buonanno and Chen (2002): PRD 65, 042001
Y. Chen review (2013): http://arxiv.org/pdf/1302.1924.pdf
Advanced Gravitational Wave Detectors
Nergis MavalvalaDepartment of Physics
Massachusetts Institute of Technology
Les Houches, August 2015
Lecture #2
First phase of LIGO
Shot noiseSNR Power
Seismic noise
Thermal noise
Limiting Noise Sources
Dissecting noise sources
Limiting noises
Seismic noise
Direct coupling
Newtonian noise
Thermal noise
Suspension Brownian
motion
Optical coatings
Mirror internal
Quantum noise
Shot noise
Radiation pressure noise
Other technical noises
Laser frequency
Laser intensity
Scattered light
Residual gas
Length and alignment control
systems
Magnetic actuation
Acoustic couplings
Nonlinear couplings (up‐
conversion …
Seismic noise
Mechanical oscillators (pendulums)
FROM
TO
Isolation~1/f2
x/F
Vibration isolation
Vibration isolation
Thermal noise
Fluctuation‐Dissipation Theorem
Response of any linear system in thermodynamic equilibrium
Brownian noise or Johnson noise
Fluctuating force spectrum is proportional to magnitude
of dissipation (mechanical loss)
HEAT BATH
Fluctuation-Dissipation TheoremCallen and Welton, 1951
LOSS
Motion
Menagerie of thermal noises
Mirror substrate
Suspension
Optical Coating
Anelasticity of materials
Pendulums are special
Pendulums can achieve much higher Q
than intrinsic material loss φ‐1
would imply
For vertical displacements, response due to elastic spring constant of wire stretching (with spring constant
)
For horizontal displacements, response due to gravitational restoring force (with spring constant
)
Only elastic spring constant has dissipative part, gravitational potential is lossless
The Quantum Noise Limit
Quantum Noise in an Interferometer
X1
X
X1
X
Laser
Radiation pressure noiseCoherent intracavity field + quantum fluctuations
fluctuating force mirror displacement
Shot noiseCoherent signal field + quantum fluctuations
fluctuating phase
First phases of LIGO (2000 to 2011)
Not Fast or Easy
Started in 2001…Many years and many technical noises later, we arrived at the design.
Sensitivity achieved during S6/VSR1 (2007 to 2010)
LIGO: H2
LIGO: L1
LIGO: H1
Virgo
GEOLIGO, GEO and Virgo share all data to form a global detector network.
Since 2006, roughly 2 years of network data have been collected.
The LIGO Scientific Collaboration includes over 50 Universities and about 1000 researchers.
LIGO listened… And had something to say
Astrophysics with first generation detectors
Journals include
Physical Review
Astrophysics Journal
Nature
Classical and Quantum
Gravity
New Journal of Physics
Topics include
Neutron star and black
hole coalescence
Gamma‐ray bursts
Known pulsars (e.g . Crab)
Unknown pulsars
Transient sources
(“bursts”)
Cosmological stochastic
background
Over 100 published results
No positive detections (yet)
The search for GRB070201
DM31
Abbott et al., Ap. J 681, 1419 (2008)Mazets et al., Ap. J 680, 545 (2008)Ofek et al., Ap. J 681, 1464 (2008)
25%50%75%90%
GRB 070201
Very luminous short duration, hard
gamma‐ray burst
Detected by Swift, Integral, others
Consistent with being in M31
Leading model for short GRBs:
binary merger involving a neutron star
Looked for a GW signal in LIGO
No plausible GW signal found
Can say with >99% confidence
that GRB070201 was NOT caused by a compact binary star merger in M31
Conclusion: it was most likely a Soft Gamma Repeater giant flare in M31
Why we didn’t hear anything yet?
Extrapolate NS‐NS inspirals to other galaxies weighted by blue‐light luminosity
Roughly 1 MW of blue‐light every 20 Mpc3
Events ~ Rate x Time x Detection Volume
?Estimated by population synthesis based on 5 know tight NS binaries Rate ~ 100/Myr
Second generation detectors
Advanced LIGO
Astrophysical motivations
Factor 10 better amplitude sensitivity
(Range)3
= rate
Factor 4 lower frequency bound
Use same infrastructure but replace detector
components with new designs
Expect to observe 1000x more galaxies by 2018
Facilities limits
Seismic noise: better isolation
10-24
10-23
10-22
10-21
10Hz 100Hz 1kHz 10kHz
Seis
mic
Strain1/√Hz
Thermal
Quantum
Each interferometer floats on tons of metal with hundreds of active control loops…
Active Isolation, 3 layersQuadruple Pendulum, 1Hz
Courtesy Matt Evans
Thermal noise: Less Loss
Strain1/√Hz
10-24
10-23
10-22
10-21
10Hz 100Hz 1kHz 10kHz
Thermal
Quantum
Quartz Suspension, Q ~ 600 MFused Silica Test Mass, 40 kg
It all ends in a 40 kg glass cylinder suspended by 400 μm glass fibers…
Courtesy Matt Evans
Shot noise: More power
10-24
10-23
10-22
10-21
10Hz 100Hz 1kHz 10kHz
Quantum
Strain1/√Hz
100 W input power5 kW on beam splitter750 kW in arm cavities
With nearly 1MW of circulating power, radiation pressure becomes a serious problem…
13um !!
Courtesy Matt Evans
More Power, Less Loss…Some woes!
NascentExcitation
MechanicalMode
RadiationPressure
PumpField
ScatteredField
(cavity gain)
High FinesseCavity
Courtesy Matt Evans
Instabilities from photon-phonon scatteringA mirror phonon can be absorbed
by the photon, increasing the photon energy
dampingThe photon can emit
the phonon, decreasing the photon energy acoustic instability
absorption emissionPhononPhonon
Damping Unstable oscillation
Advanced LIGO – here and now
Advanced LIGO noise budget
Advanced LIGO prepares for O1
Noise hunting
Advanced LIGO (2011…)
Radiation pressure noiseStronger measurement larger backaction
Shot noiseMore laser power
stronger measurement
Beyond Quantum Noise
Lecture 3
Quantum engineering
Quantum noise in an interferometer
X1
X
Laser
X1
X
Radiation pressure noiseCoherent intracavity field + quantum fluctuations
fluctuating force mirror displacement
Shot noiseCoherent signal field + quantum fluctuations
fluctuating phase
Quantum noise in an interferometer
X1
X
Laser
X1
X
Radiation pressure noiseCoherent intracavity field + quantum fluctuations
fluctuating force mirror displacement
Shot noiseCoherent signal field + quantum fluctuations
fluctuating phase
Squeezed state generation
How to squeeze photon states
Need to simultaneously amplify one quadrature and
de‐amplify the other
Create correlations between the quadratures
Simple idea nonlinear optical material where
refractive index depends on intensity of light illumination
Nonlinear optical interaction
a
a
a
a bb
Parametric oscillation Second harmonic generation
The output photon quadratures are
correlated
a
b
Parametric amplification
a
a
a
Squeezed light source
To Inter- ferometer
Squeezed state injection
Squeezing injection in LIGO
PowerRecycling
Mirror
H1 LASER
Anti-Symmetric Port
Arm Cavity (4 km)
V1
CONTROLLASER
PUMP LASER
SHG
Squeezed vacuum source
Output Photodiode
OMC
OPO
Vacuum Envelope
to squeezed light source: feed-back to PUMP laser frequency
for squeeze angle control
LIGO H1 Interferometer
Input Mode-Cleaner frequency shiftedcontrol beam
BeamSplitter
Arm Cavity (4 km)
FaradayIsolator
Output ModeCleaner
squeezed vacuum &frequency shifted control beam
(a) Coherent state of light
Quadrature Phase
In-Phase
(b) Vacuum state
In-Phase
(c) Squeezed vacuum state
In-Phase
OPO green pump beam
to squeezed light source:phase lock loop with PUMP laser
from H1 laser:phase lock loop with PUMP laser
from squeeze angle control photodiode: feed-back to PUMP
laser frequency
Squeeze Angle Control Photodiode
Quadrature Phase
Quadrature Phase
LIGO H1 Squeezed
Squeezing down to 150 Hz
2 dB (25%) improvement
LIGO Scientific Collaboration, Nature Photonics (2013)
Advanced LIGO with squeeze injection
Shot noise
Radiation pressure
McClelland, Mavalvala, Schnabel, and Chen, Lasers and Photonics Reviews (2011)Schnabel, Mavalvala, McClelland, and Lam, Nature Communication (2010)
Frequency dependent squeezing
Filter cavity
Frequently asked questions
If the GW changes the space‐time distance, doesn’t the wavelength of the light also change in the same way?
How do we know GWs
even exist? What if GR is wrong?
What are the best estimates and uncertainties for known and/or expected astrophysical sources?
Space detectors are so difficult and expensive. Why not just stick with terrestrial detectors?
Frequently asked questions
Why don’t we just make the interferometers longer?
How flat do the mirrors have to be? Does the technology exist to make a mirror surface as exquisite
as the interferometer path length changes of 10‐18
m?
If the mirrors are moving by many microns, how can we measure length changes of 10‐18
m?
Frequently asked questions
If thermal noise is such a big problem, why not cool the mirrors?
If Newtonian noise is such a big problem, why not measure and subtract it?
Isn’t it hopeless to do better than Advanced LIGO? After all, how can you circumvent the Heisenberg
Uncertainty limit?
Optomechanics in LIGO
Nergis MavalvalaDepartment of Physics
Massachusetts Institute of Technology
Les Houches, August 2015
Lecture #3
Squeezed state generation
Nonlinear optical interaction
Squeezed light source
To Inter- ferometer
Squeezed state injection
Squeezing injection in LIGO
PowerRecycling
Mirror
H1 LASER
Anti-Symmetric Port
Arm Cavity (4 km)
V1
CONTROLLASER
PUMP LASER
SHG
Squeezed vacuum source
Output Photodiode
OMC
OPO
Vacuum Envelope
to squeezed light source: feed-back to PUMP laser frequency
for squeeze angle control
LIGO H1 Interferometer
Input Mode-Cleaner frequency shiftedcontrol beam
BeamSplitter
Arm Cavity (4 km)
FaradayIsolator
Output ModeCleaner
squeezed vacuum &frequency shifted control beam
(a) Coherent state of light
Quadrature Phase
In-Phase
(b) Vacuum state
In-Phase
(c) Squeezed vacuum state
In-Phase
OPO green pump beam
to squeezed light source:phase lock loop with PUMP laser
from H1 laser:phase lock loop with PUMP laser
from squeeze angle control photodiode: feed-back to PUMP
laser frequency
Squeeze Angle Control Photodiode
Quadrature Phase
Quadrature Phase
LIGO H1 Squeezed
Squeezing down to 150 Hz
2 dB (25%) improvement
LIGO Scientific Collaboration, Nature Photonics (2013)
Advanced LIGO with squeeze injection
Shot noise
Radiation pressure
McClelland, Mavalvala, Schnabel, and Chen, Lasers and Photonics Reviews (2011)Schnabel, Mavalvala, McClelland, and Lam, Nature Communication (2010)
Frequency dependent squeezing
Filter cavity
Oelker Isogai et al. (2015)Extrapolation for Adv. LIGO:
16m filter cavity: factor of 2 reduction in shot noise (6dB),
25% reduction in radiation pressure noise (2 dB)
Optomechanics & Radiation Pressure
Radiation Pressure
When radiation pressure dominates
Techniques for improving gravitational wave detector sensitivity
Opportunities to study quantum effects in macroscopic systems
Observation of quantum radiation pressure
Generation of squeezed states of light
Quantum states preparation
Tools for quantum information science
Cavity Optomechanics Primer
Cavity optomechanics
F. Marquardt, Les Houches “Quantum Machines”
(2011)
Linearized approximation
Optomechanical Coupling
Laser-driven cavity mode at frequency ∆
Mechanical oscillator mode at frequency Ω
and
are experimental “knobs”
Coupled oscillators with coupling coefficient g(α, ∆)
Optomechanical Phenomena
On resonance (Δ
= 0)
Optical state phase shift
proportional to displacement
Blue detuning (Δ
= + Ω)
Two‐mode squeezing
Optical spring
Mechanical lasing / Parametric instabilities
Red detuning (Δ
= ‐
Ω)
State transfer between
photons and phonon
Optical damping
Cooling
Imbalance of Stokes and anti‐Stokes
Aspelmeyer, Kippenberg, and Marquardt, 1303.0733
Stokes scattering enhanced
Anti-Stokes scattering enhanced
Scattered photons have less energy
heating/amplification of mechanical mode
Scattered photons have more energy
cooling/damping of mechanical mode
Explosion of optomechanics experiments
PC zipper 1012
g
SiN3
membrane 108
g
Toroidal microcavity 1011
g
Micromirrors 106
g
12mm mirror 1 g
LIGO mirror 104
g
AFM cantilevers 108
g
WG-WGM 1011
g
Trampolines 107
g
NEMS 1011
g
Optomechanics in LIGO
The radiation pressure force couples the light field to mirror motion
Alter the dynamics of the mirror
Spring‐like forces optical trapping and acoustic instability
Viscous forces optical damping
Tune the frequency response of the GW detector
Manipulate the quantum noise
Quantum radiation pressure noise and the standard quantum
limit
Produce quantum states of the mirrors
Produced squeezed states of light
Clas
sica
lQ
uant
um
Classical forces in optomechanics
Detune optical field from cavity resonance
Change in mirror position changes intracavity power radiation pressure
exerts force on mirror
Time delay in cavity results in cavity
response doing work on mechanics
Optomechanical coupling in Adv. LIGO
750 kW in arm cavities gives large static force 3 mm displacement from equilibrium
Signal recycling cavity may be detuned to optimize interferometer response to GW sources
Detuned cavity equation of motion
Optomechanical rigidity gives harmonic restoring force with spring constant
Optomechanics in Adv. LIGO
SQL
Source‐specific tuning
23
10-24
10-23
10-22
10-21
10-20
10-19
Stra
in (1
/Hz)
1012 3 4 5 6 7 8 9
1022 3 4 5 6 7 8 9
1032 3 4 5 6 7 8 9
104
Frequency (Hz)
Hanford 4 km S6
Livingston 4 km S6
AdvLIGO, No Signal Recycling (early operation)
AdvLIGO, Zero Detuning (Low Power)
AdvLIGO, ZeroDetuning (High Power)
AdvLIGO, NS-NS optiimized AdvLIGO, High Frequency Detuning
Targeting different sources
24
101
102
103
10−2
4
10−2
3
10−2
2
Frequ
ency
[Hz]
Strain [1/Hz]
Adva
nced
LIGO
: Nom
inal
LMXB
SN
NS EOS
Binary Populations Tests of GR
IMBH
Pulsars
Multi-messenger (sky localization)
Courtesy
Matt Evans
Primordial GWB
Parametric instability
Damping and oscillation
Mirror phonon absorbed by photon decreases phonon
energy damping
Photon emitted by mirror phonon increases phonon energy acoustic instability
(unstable oscillation)
absorption
emission
Phonon
Phonon
High optical power, low mech. loss
NascentExcitation
MechanicalMode
RadiationPressure
PumpField
ScatteredField
(cavity gain)
High FinesseCavity
Courtesy Matt Evans
Acoustic instability in Adv. LIGO
12, , nnmmarmm GBQPR
mechanical modeQ-factor
power in arm cavity
spatial overlap optical-
mechanical mode
optical mode gain
optical mode linewidth
frequency match
Evans et al., PRL 114, 161102 (2015)
Approaching the quantum regime
Three scales on which we study quantum light‐mirror coupling
250 ng 1 g 10 kg
Gram‐scale mirrors
Experimental cavity setup
5 W
10%
90%
1 gram mirror
Optical fibers
Coil/magnet pairs for actuation (x5)
1 m
All‐optical trap
Blue detuned field gives optical spring (trap)
Red detuned field with 1/20 the power gives damping
Independently control restoring and damping forces
T. Corbitt et al., Phys. Rev. Lett 98, 150802 (2007)
Stable!
Stiff!
Eliminate frequency noise
Two identical cavities with 1 gram mirrors at the ends
Common‐mode rejection cancels out laser noise
T. Corbitt et al., Phys. Rev. A 73, 023801 (2006)
Optically trapped and cooled mirror
C. Wipf, T. Bodiya, et al. (March 2010)
1 gram mirror
Optical fibers
Teff
= 0.8 mK N = 35000
Optical dilution of thermal noise
Optical spring couples mechanics to “cold”
bath
Detailed balance:
Environment T ~ 300 K
Mechanics Teff
Laser T ~ 0 K
Kilogram scale mirrors
Thomas Corbitt Chris Wipf Daniel Sigg
Cooling the kilogram‐scale mirrors of Initial LIGO
LIGO Scientific Collaboration
Observations of QRPN and OM squeezing
Quantum radiation pressure and optomechanical squeezing observed
QRPN observed (2013)
Regal group, ColoradoNature Physics (2013)
Thermal noise
Rad
iatio
n pr
essu
re
Optomechanical squeezing (Caltech)
Painter group, CaltechNature (2013)
Optomechanical squeezing (Boulder)
Regal group, ColoradoPhys. Rev. X (2013)
Beyond Advanced LIGO
Advanced LIGO/Virgo Upgrades
Newtonian noise subtraction
Seismometer array
around test masses
2 to 3x reduction at few
Hz
Squeezed state injection
10 dB (3x) squeezing at
source
6 dB (2x) SNR improvement
realizable
Quantum noiseCoating Brownian noiseNominal aLIGOFrequency IndependentTotal Noise
Third generation detectors
Einstein TelescopeCosmic Explorer
Many options to consider…
10 to 40 km arm lengths
Underground caverns,
salt mines
Newtonian noise subtraction
factors of ~2 achieved
Mirrors cooled to 4 K
Vibrations worrisome
Silicon optics
200kg chunks
low mechanical loss
high thermal conductivity at low temp.
Xylophone
Low frequency cryogenic, low laser power
High frequency high laser power
Squeezed light injection
10 to 20 dB !
Cast of Characters > 900
Capturing the elusive wave…
Tests of general relativity
Directly observe ripples of
space‐time
Astrophysics
Directly observe the Black Holes, the Big
Bang, and objects beyond our current imagination
Fertile ground for quantum optics and optomechanics
at very macroscopic objects
40 kg