Transcript GEO-HF

Harald Lück
ILIAS Meeting
for the GEO team
Palma, October 2005
What is GEO-HF ?
 A research and upgrade program instead of the
name of a new detector
 sequential upgrades of GEO600 (similar to LIGO+
/ VIRGO+ )
 prototype research to prepare these upgrades
 support transit of 3rd generation Lab research to
detector subsystems or detector configuration
Optical Layout
interferometer with
„dual recycling“
modecleaner
12W Laser
detector
GEO600 Noise Sources
(Design values)
GEO600 Theoretical Noise Budget
-18
10
Seismic
Suspension TN
Substrate TN
Coating TN
Thermorefractive
Shot 200Hz
Total
-19
10
-20
ASD [h/ Hz]
10
-21
10
-22
10
-23
10
v2.0
-24
10
10
1
10
2
3
Freq. [Hz]
10
GEO600 Noise Sources
(Design values)
GEO600 Theoretical Noise Budget
-18
10
Seismic
Suspension TN
Substrate TN
Coating TN
Thermorefractive
Shot 1 kHz
Total
-19
10
-20
ASD [h/ Hz]
10
-21
10
-22
10
-23
10
v2.0
-24
10
1
10
2
10
3
Freq. [Hz]
10
GEO HF - Motivation
 Provide scientifically interesting data with the GEO
instrument until 2014
– optimized at low frequencies for network analysis or
– optimized for high frequency sources
 Perform developments and tests towards third
generation detectors
– technologies, materials and optical schemes
 Maintain capability to do high-fidelity
measurements
– Discover and analyze additional noise sources
– keep team of experts in GEO collaboration
→ upgrade GEO600 and work on prototypes
Shot Noise
GEO Theoretical Noise Budget
-18
10
Seismic
Coating TN
Thermorefractive
Shot 1 kHz
Shot 200Hz
Shot 500Hz
-19
10
-20
ASD [h/ Hz]
10
-21
10
-22
10
-23
10
-24
10
1
10
2
10
3
Freq. [Hz]
10
High Frequency Sources
Possible way forward for
‘GEO-HF’
GEO Theoretical Noise Budget
– increase circulating power
– improve thermal
compensation scheme
– optimize signal recycling
bandwidth
-22
ASD [h/ Hz]
10
 use squeezed light to
reduce shot noise
 change main mirrors
-23
10
Seismic
Coating TN
Thermorefractive
Shot 1 kHz
-24
10
10
 improvements without
changing mirrors
1
10
2
3
Freq. [Hz]
10
– reduce coating thermal
noise if possible
– further increase circulating
power and / or reduce
signal recycling bandwidth
Changes
interferometer with
„dual recycling“
modecleaner
Laser
increase laser power
40W / 200W
detector
Nd:YAG 200W Laser System
200W
12W
0.8W
Injection-Locked Single Frequency Power
300
200
200
Slope: 30 %
150
100
50
0
400
500
600
700
800
900
1000
1100
1200
Pump Power [W]
230
Output Power [W]
Output Power [W]
250
150
P = 195 W
100
50
2
225 W; M =1,45
Output Power [W]
225
0
00:00
2
218 W; M =1,3
220
2
215 W; M =1,2
215
2
213 W; M =1,14
210
205
930
940
950
Pump Power [W]
960
970
00:10
00:20
00:30
00:40
Time [hh:mm]
 continuous single-frequency
operation >8h
4 Rod Nd:YVO4
Amplifier
39W output power
11W seed / 115W pump
2/8/2 mm Nd:YVO4 rod
GEO 600 style
11W seed laser
Changes
• mid-arm pumping
• new gate valves
interferometer with
„dual recycling“
modecleaner
Laser
reduce Finesse of MC to
keep peak intensity
constant
detector
Changes
• (reduce absorption)
• thermal compensation
interferometer with
„dual recycling“
modecleaner
Laser
detector
Squeezed light
600 m
PowerRecycling
mirror
600 m
Laser
Signal-recycling mirror
Faraday Rotator
Squeezed state
Photo diode
Rotated Squeezing Ellipse
Table Top Setup
Sqeezing Enhanced
Signal/Noise
Optional Changes
replace main mirrors
interferometer with
„dual recycling“
modecleaner
Laser
detector
Further GEO-HF
optional changes
 based on thermal noise
considerations we consider
replacing the end mirrors with
new mirrors
 better coatings (mechanical and
optical losses) if available
 possibly switch to silicon/
sapphire if thermal noise
performance is better than in
fused silica
Coated fused silica mirror for GEO600
~18cm diameter
Thermal Noise for GEO Main
Optics
GEO-HF noise levels based on best possible coating losses
-22
10
h [Hz
-1/2
]
full coating, silica
doped coating, silica
aperiodic coating, silica
silica substrate
thermorefractive BS
-23
10
-24
10
2
10
3
10
Frequency [Hz]
4
10
Ongoing lab work on
coatings
 Considerable work in LIGO Scientific Collaboration
and elsewhere on studies of coating loss
 Results suggest:
– Ta2O5 is dominant source of coating dissipation
– doping Ta2O5 with TiO2 can reduce dissipation by ~factor 2
 Further, studies by Pinto et al suggest by using multilayers of ‘non-standard’ periodicity fraction of lossy
high index material may be reduced?
Silicon
 Strong interest in cryogenic silicon for use in ‘3rd
generation detectors’

Laboratory studies ongoing on:
future detectors
fabrication of, and dissipation
in, silicon suspension elements

intrinsic dissipation in bulk
silicon

fabrication and dissipation of
monolithic silicon pendulums

e.g. GEO-HF, EGO
Silicon suspension technology.
Thermal noise options
-22
h [Hz-1/2]
10
full coating, silica
doped coating, silica
aperiodic coating, silica
silica substrate
aperiodic coating, sapphire
sapphire substrate
aperiodic coating, silicon
silicon substrate
10
10
-23
-24
10
1
10
2
3
Frequency [Hz]
10
10
4
Prototype Work within
GEO-HF
 Glasgow (in operation):
– Advanced Configuration Prototype
– Cryogenic Materials Test System
 Hannover (planned):
– test subsystem before installation in GEO
 active seismic isolation system
 digital control
 fast mirror installation tools and techniques
– test 3rd Generation techniques and configurations
GEO-HF Summary
 We plan to operate a flexible GW detector in a worldwide
network until advanced IFOs come online.
 Sequential upgrades to improve high frequency sensitivity
will be made.
– higher circulating power
– lower coating thermal noise ( ? )
– squeezed light injection

GEO collaboration will operate prototypes to prepare
these upgrades and
 to support the transit of 3rd generation Lab research to the
detector subsystems or detector configuration level.
 Timeline: starting upgrading after extended data taking
2007/2008
 Proposal for ~3.5 M€ approved by MPG
Results for silicon at room temperature
Measured loss factors for two samples of bulk silicon
1.4E-07
The doped [111] sample
typically
showed
lower
dissipation, though whether
this was due to the crystalline
orientation of the sample, the
dopant, or some other reason,
is as yet unknown.
1.2E-07
[100]
Loss
1.0E-07
8.0E-08
6.0E-08
4.0E-08
2.0E-08
[111]
0.0E+00
30
35
40
45
50
55
60
Frequency (kHz)
 Lowest loss obtained so far = (9.6 +/- 0.3) x 10-9
 Comparable with the lowest loss factors measured at room
temperature
(consistent with results by Mitrofanov et al from earlier times)
 (Cryogenic measurements in preparation)
Controlling the SR bandwidth
Bandwidth  Finesse of the SR
cavity  Reflektivity of MSR
Heater
Signalrecycling-Etalon
(r=150mm, d=75mm)
Reflectivity is controlled by
temperature
Diffraction Gratings
1st order Littrow
2nd order Littrow
All-Reflected Cavities
University of Jena
Littrow order
1st
2nd
Finesse
1580
400
diff. efficiency
99.62%
0.58%
loss
0.2%
0.4%

Title: The GEO-HF Project

Abstract: The GEO 600 gravitational wave detector uses advanced
technologies like signal recycling and monolithic fused-silica suspensions to
achieve a sensitivity close to the km scale LIGO and VIRGO detectors. As
soon as the design sensitivity of GEO600 is reached the detector will be
operated as part of the worldwide network to acquire data of scientific interest.
The limited infrastructure at the GEO site does not allow for a major upgrade of
the detector. Hence the GEO collaboration decided to improve the sensitivity of
the GEO detector by small sequential upgrades some of which will be tested in
prototypes first. The development, test and installation of these upgrades are
named "The GEO-HF Project" and this contribution will give a status report on
this project.
contents

GEO600
–
–
–

description
current sensitivity
limits
GEO HF Detector
–
–
–
motivation
sources
ways forward




increase circulating power and power handling capability
change optics if coating technology allows for lower coating thermal noise
use squeezed light
GEO HF Prototypes
–
–
–
–
test subsystems / optical layout before installation
test installation procedures for fast installation at site
allow for high displacement sensitive measurements
test and develop third generation materials, optical readout schemes, control
schemes
Improvements to GEO
sensitivity possible by
changing to silicon mirrors? –
possible option.
substrate, fp, cavities silicon, silica bs
cavities
bs
doped coating, total
coating, cavities
coating, bs
total (fp substrate + fp coating)
thermorefractive
-22
h [/Hz]
10
-23
10
-24
10
1
10
2
3
10
10
Frequency [Hz]
4
10
Improvements to GEO
sensitivity possible by
changing to silicon mirrors? –
possible option.
substrate, fp, cavities silicon, silica bs
cavities
bs
doped coating, total
coating, cavities
coating, bs
total (fp substrate + fp coating)
-22
h [/Hz]
10
-23
10
-24
10
1
10
2
3
10
10
Frequency [Hz]
4
10
GEO-HF noise levels based on best possible coating losses
-21
10
full coating, silica
doped coating, silica
aperiodic coating, silica
silica substrate
aperiodic coating, sapphire
sapphire substrate
aperiodic coating, silicon
silicon substrate
-22
h [Hz-1/2]
10
-23
10
-24
10
2
10
3
10
Frequency [Hz]
4
10
GEO-HF noise levels based on best possible coating losses
full coating, silica
doped coating, silica
aperiodic coating, silica
silica substrate
aperiodic coating, sapphire
sapphire substrate
aperiodic coating, silicon
silicon substrate
-22
h [Hz-1/2]
10
-23
10
-24
10
2
10
3
10
Frequency [Hz]
4
10
Current estimate of thermal
noise limited sensitivity of
GEO
10
-21
substrate, loss = 1e-7
standard coating
Total (coating + substrate)
-22
h [/Hz]
10
10
10
-23
-24
10
1
10
2
10
3
Frequency [Hz]

Uses:
– Substrate loss of 1 x 10-7 (Suprasil mirrors)
– Current SiO2/Ta2O5 coatings
10
4
Squeezing vs
Quadrature
Phase squeezing
Squeezing at 45°
Amplitude squeezing
R. Schnabel et al., Class. Quantum
Grav. 21, S1045 (2004)]
Frequency
dependent
Squeezing
at -45°
squeezing,
(single filter cavity)
Estimated GEO600 thermal
noise – high Q value
10
-21
substrate, loss =1e-7
substrate, loss =5e-9
-22
Limit set by
coating
thermal noise
alone
h [/Hz]
10
10
10
-23
-24
10


1
10
2
10
3
10
4
Frequency [Hz]
Assume:
– Loss in Suprasil = 5 x 10-9
In this case we would then be limited by coating thermal noise  strong interest in
reducing coating dissipation
Possible improvements to
thermal noise limited
sensitivity of GEO for better
coatings
-21
10
Thermal noise from:
-22
substrate, loss = 5e-9
standard coating
Standard coating+substrate
Coating using doped Ta2O5
Doped coating+substrate
Reduced coating noise
using Non-Standard Periodicity
(NSP) ?????– Pinto et al
NSP coating+substrate
h [/Hz]
10
-23
10
-24
10
1
10
2
10
3
10
Frequency [Hz]
44
10
Studies of coating loss ongoing – (see poster by G. Cagnoli)
GEO-HF noise levels based on best possible coating losses
full coating, silica
doped coating, silica
aperiodic coating, silica
silica substrate
aperiodic coating, sapphire
sapphire substrate
aperiodic coating, silicon
silicon substrate
-22
h [Hz-1/2]
10
-23
10
-24
10
2
10
3
10
Frequency [Hz]
4
10
10
10
-18
GEO Simulated Noise –
high Q
Seismic
Suspension TN
-19
Substrate TN
Coating TN
[h*Hz-1/2]
10
10
10
10
10
Thermorefractive
-20
Shot
Total
-21
-22
-23
-24
10
1
2
10
Freq. [Hz]
10
3
Studies of silicon as a test mass
substrate
 Preliminary room T measurements
made of mechanical dissipation of
bulk silicon samples suspended
on silk thread or wire loops
– Internal resonant modes of the
samples excited; decay of mode
amplitude measured
Clamp
Suspension
thread/wire
Test mass
To high voltage
Excitation plate
(behind mass)
Schematic diagram of front view of suspended
test mass.

Silicon samples cut along different
crystal axes, [111] and [100]. The [111]
sample was boron-doped.
Dissipation of two silicon samples of
identical
geometry,
supplied
by
collaborators in Stanford, was measured
over a range of frequencies.
Shot Noise in SR IFOs
GEO Theoretical Noise Budget
-22
ASD [h/ Hz]
10
Seismic
Coating TN
Thermorefractive
Shot 1 kHz
Shot 200Hz
Shot 500Hz
-23
10
-24
10
10
1
10
2
3
Freq. [Hz]
10
GEO600 sensitivity evolution
h(t) [Hz-1/2]
Jan 02
10
-17
10
-18
10
-19
Aug 02 (S1)
Jan 04 (S3)
Aug 04
Feb 05 (S4)
10
-20
10
-21
Sept 05
10
2
Frequency [Hz]
10
3