Transcript GW_Flaminio_R&D

Research and Development
for Gravitational Wave Detectors
Raffaele Flaminio
CNRS/LMA Lyon
Lisbon, 8 January 2008
1
Ground-based GW detectors

Focus on ground based laser interferometers
 Most sensitive detectors in operation
 LIGO, Virgo, GEO, TAMA,
 Some of the following applies to other kind of detectors (e.g. resonant detectors)
Lisbon, 8 January 2008
2
Present reach
 Gravitational collapses in the galaxy (or nearby ones)
 Test upper limits of known galactic pulsars (and look for unknown ones)
 Search for coalescing neutron stars up to a max distance of ~ 30 Mpc
Lisbon,
8 Januaryfor
2008merger of binary black holes to a max distance of ~150 Mpc
3
 Search
Toward GW astronomy

Present detectors will test upper limits

Even in the optimistic case rate
too low to start GW astronomy

Need to improve the
sensitivity

Increase the sensitivity
by 10  increase the
probed volume by 1000

Plans to improve the
present detectors
Lisbon, 8 January 2008
4
GW roadmap: time scale
´06 ´07 ´08 ´09 ´10 ´11 ´12 ´13 ´14 ´15 ´16 ´17 ´18 ´19 ´20 ´21 ´22
Virgo+
Virgo
GEO
Advanced Virgo
GEO HF
LIGO
Hanford
Livingston
LIGO+
Advanced LIGO
Launch
LISA
DS
E.T.

PCP
Transfer data
Construction Commissioning data
ET: Einstein Telescope
 Design study selected by the EU within FP7
Lisbon, 8 January 2008
5
GW roadmap: sensitivity scale
-19
10
rd
Virgo+ 2009
LIGO 2005
-20
10
Virgo Design
h(f) [1/sqrt(Hz)]
-21
10
(g)
(f)
(a) 3 Generation
(b)
LCGT 2005
AURIGA
(c) advanced LIGO
(d) advanced Virgo
(e) LIGO
(f) Virgo
(g) GEO600
GEO-HF
2009
-22
10
(e)
-23
10
DUAL Mo
(a)
(Quantum Limit)
(d)
Ad LIGO/Virgo NB
(b)
Advanced
LIGO/Virgo
(2014)
(c)
-24
10
Credit: M.Punturo
Einstein GW Telescope
-25
10
1
10
100
1000
10000
Frequency [Hz]
Lisbon, 8 January 2008
6
Present limitations ….
• Shot noise
-18
10
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
h(f) [1/sqrt(Hz)]
- Depends on quantum nature of light
- Decreases when more photons are used
-19
10
- Depends on the optical configuration adopted
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
1
10
100
1000
10000
Frequency [Hz]
Lisbon, 8 January 2008
7
…. and possible improvements
• Increase power stored in the
interferometer
-18
10
- increase laser power
- decrease optical losses
1) Mirrors heating and thermal lensing
- better thermal compensation
- decrease light absorption
-19
10
h(f) [1/sqrt(Hz)]
• But pay attention to:
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-20
10
(b)
-21
10
(d)
-22
2) Radiation pressure noise
- increase mirror mass
- optimize optical configuration
signal recycling
- use non classical light
light squeezing/quantum optics
10
(a)
(e)
(c)
-23
10
1
10
100
1000
10000
Frequency [Hz]
3) Non-linear coupling between the light
fields and the mirror suspensions
Lisbon, 8 January 2008
8
Present limitations …
• Mirror thermal noise
-18
10
h(f) [1/sqrt(Hz)]
- brownian motion
- due to temperature ….
- …plus any source of friction in the mirror
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-19
10
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
1
10
100
1000
10000
Frequency [Hz]
Lisbon, 8 January 2008
9
… and possible improvements
• Reduce friction in the mirrors
-18
10
• Friction in the coating
-19
10
h(f) [1/sqrt(Hz)]
- Main source of friction today
- Multi-layers SiO2/Ta2O5 used today
- Ta2O5 is the lossy material
look for new materials
materials science
- SiO2 layer lossier than raw material
improve deposition process
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-20
10
(b)
-21
10
(d)
• Friction in the substrate
- Best material so far: silica
- Avoid attaching anything to preserve
mechanical quality
- Move to electrostatic actuators avoiding
magnets attached to the mirror
Lisbon, 8 January 2008
-22
10
(a)
(e)
(c)
-23
10
1
10
100
1000
10000
Frequency [Hz]
10
Present limitations …
-18
• Pendulum thermal noise
10
h(f) [1/sqrt(Hz)]
- same kind of brownian motion
- due to temperature …
- … plus friction in the suspension wires
- or friction between the wires and the
mirror
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-19
10
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
1
10
100
1000
10000
Frequency [Hz]
Lisbon, 8 January 2008
11
… and possible improvements
-18
10
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-19
h(f) [1/sqrt(Hz)]
10
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
1
10
100
1000
Frequency [Hz]
• Decrease pendulum friction
Lisbon, 8 January 2008
- better suspensions wires (new materials)
- better wire clamping
- monolithic suspensions
12
- fused silica fibers
- silicate bonding
10000
Further reduction of thermal noise
• Thermal noise decrease as √T
- go to low temperatures
- friction vs temperature?
- depends on materials (materials science)
- look for optical materials with good
mechanical properties at low temperature
(silica not a good choice)
• Thermal lensing
Si
- due to laser power deposited in the mirror
- higher mirror thermal conductivity
lower thermal lensing
- higher wires thermal conductivity
heat extraction more efficient
• Silicon
- silicon a good candidate
- silicate bonding behavior at low T?
- thermal conductivity across bonding?
- on-going R&D
Lisbon, 8 January 2008
13
Cryogenics for GW detectors
COLD FINGER
• Need to cool large masses
• Vibration free cryogenics
• Soft thermal links
• Points of contact with underground detectors for rare events search
Lisbon, 8 January 2008
14
ILIAS: the STREGA joint research activity


Strong component within the ILIAS project
Goal: thermal noise reduction for GW detectors
 All the european groups working in thermal noise
reduction involved
 INFN (Ge, Fi, Fr, Le, Pd, Pi, Pg, Rm1, Rm2), CNRS
(LKB, ESPCI,LMA), Univ Glasgow, CNR (Trento),
Leiden, Jena, …
 All collaborations: Virgo, GEO, ROG, Auriga, MiniGRAIL

Ingredients:







Cryogenics suspensions
Cryogenics mirrors
Materials
Thermo-elastic studies
A key role for starting the ET design study
A lot more to do
But ILIAS ends in 2009 and support available
within ILIAS-NEXT very much reduced
Lisbon, 8 January 2008
15
Present limitations: seismic noise
-18
10
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-19
h(f) [1/sqrt(Hz)]
10
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
1
• Seismic noise
Lisbon, 8 January 2008
10
100
1000
10000
Frequency [Hz]
– residual transmission of seismic motion through the suspensions system
– 'relatively large' motion at very low frequency
→ need for a control system
→ control system noise
16
- sensitivity to weather conditions
Seismic noise: possible improvements
Better active isolation

 More sensitive accelerometers
 Very low-frequency tilt-meters
 Gryo-lasers
Softer springs ?

-18
10
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-19
h(f) [1/sqrt(Hz)]
10
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
Lisbon, 8 January 2008
1
10
17
100
1000
10000
Forthcoming limitations
Gravity gradient noise

 Limitation to existing infrastructure
 Will limit advanced detectors
-18
10
(a) Virgo Nominal sensitivity
(b) Seismic noise
(c) Pendulum thermal noise
(d) Mirror thermal noise
(e) Shot Noise
-19
h(f) [1/sqrt(Hz)]
10
Figure: M.Lorenzini
-20
10
(b)
-21
10
(d)
-22
10
(a)
(e)
(c)
-23
10
1
10
100
1000
10000
Frequency [Hz]
Lisbon, 8 January 2008
18
Improvement: go underground !

LISM: 20 m Fabry-Perot interferometer, R&D for LCGT,
moved from Mitaka (ground based) to Kamioka
(underground)
 Seismic noise much lower
 Operation easier
Lisbon, 8 January 2008
102 overall gain
103 at 4 Hz
19
Further improvements: spherical cavern
102 less seismic noise x 104 geometrical reduction
106 overall reduction (far from surface)
Reduction
factor
(Compression waves not included) Spherical Cavern
G.Cella
NN reduction of 104 @5 Hz
with a 20 m radius cave
5 Hz
10 Hz
20 Hz
40 Hz
Lisbon, 8 January 2008
20
Combination of improvements
Upper experimental hall
50-100 m tower to accommodate
long suspension for
low frequency goal
Ellipsoidal/spherical cave for
newtonian noise reduction
Credit: R.De Salvo
10 km tunnel
Lisbon, 8 January 2008
21
The ET concept

Need to improve sensitivity at low frequencies
 More physics is there

Present facilities limited by environmental disturbances
 Seismic noise
 Gravity gradients


ET Einstein Telescope
Concepts
Rüdiger, ‘85
 Underground
» Less seismic noise
» No wind
» Temperature stability




Cryogenic
30 km beam tube
100 m suspensions
Different geometry
» Triangle?
Lisbon, 8 January 2008
22
Conclusion

Present detectors are testing upper limits of GW predictions
 A few upgrades ready to be implemented (Virgo+, Enhanced LIGO)

Advanced detectors should see several events/month
 Sensitivity will profit from on-going R&D (e.g. coating thermal noise)
 Engineering needed (e.g. monolithic suspensions)

ET Einstein Telescope





Design study should start soon
R&D activity started within FP6 (STREGA)
Should continue within FP7
More investment needed
Points of contact with other fields of astroparticle physics
» Cryogenics
» Vibration isolation
» Underground operation

GW will participate to ILIAS-NEXT




GW networking
Networking with underground labs
A few small R&D activities
But more investment will be needed
Lisbon, 8 January 2008
23