Several Fun Research Projectsat NAOJ

for the Future GW Detectors

LIGO Seminar @ CaltechAug. 8, 2006

Seiji KawamuraNational Astronomical Observatory of Japan

Picture: Sora Kawamura

National Astronomical Observatory of Japan (NAOJ)

TAMA300 is located on the NAOJ campus.

NAOJ is located in Tokyo.

Other research projectsat NAOJ

Displacement-noise free Interferometer RSE DECIGO MHz GW detection QND

Displacement-noise free Interferometer

Motivation Displacement noise: seismic Noise, thermal noise,

radiation pressure noise Cancel displacement noise shot noise limited sensitivity Increase laser power sensitivity improved indefinitely

Frequency

Sen

siti

vity

Displacement noise

Shot noise

Laser

PD

Diplacement noise

Cancel displacement noise

Increase laser power

Principle 1: GW and mirror motion interact with light differently

Difference outstanding forGW wavelength distance between masses

Mirror motionMirror motion

GW On propagation

On reflection

Light

Principle 2: Mirror motion can be cancelled by combining interferometer outputs

Increase # of mirrors Implement many

interferometers Take combination

outputs cancel mirror motion

Mirror motionLaser

PD

Mirror motion

Bi-directional MZ

Laser

PD

Example of DFI

Two 3-d bi-directional MZ

Take combination of 4 outputs

Mirror motion completely cancelled

GW signal remains (f 2)

Experiment (Ideal) One bi-directional MZ

GWMirror motion Laser

PD

GW Mirror motion

Extract

Experiment (Practical)

EOM used for GW and mirror motion

GWMirror motionLaser

PD

Laser

PD

～ ～Simulated mirror motion Simulated GW

Ideal 　　　　　　　　　　　　　　　　Practical

Results

Mirror motion 　　　　　　　　 GW signal

Mirror motion cancels out GW signal remains

Mirror motion to output

Difference

DifferenceGW signal to output

Next step Implement cavity to reduce the

effective frequency Demonstrate the cancellation of the BS

motion using two bi-directional MZ

References Kawamura and Chen, PRL, 93, (2004) 211103 Chen and Kawamura, PRL, 96 (2006) 231102 Chen, Pai, Somiya, Kawamura, Sato, Kokeyama, Ward,

submitted to PRL (gr-qc/0603054) Sato, Kawamura, Kokeyama, Ward, Chen, Pai, and So

miya, to be submitted to PRL

RSE

4m RSE

Supended mass RSE Miniature suspension system

Previous Accomplishments

Tuned RSE (w/o PRM) locked Detuned RSE (w/o PRM) locked Optical spring effect observed

– Miyakawa, Somiya, Heinzel, and Kawamura, Class. and Quantum Grav., 19 (2002) p.1555-1560

– Somiya, Beyersdorf, Arai, Sato, Kawamura, Miyakawa, Kawazoe, Sakata, Sekido, Mio, Appl. Opt. 44 (2005) pp. 3179-3191

Current Activity

Try new signal extraction method– Backup for Advanced LIGO– Baseline for LCGT

Lock RSE (w/ PRM)

New Signal Extraction Method

Signal MatrixDM

Port RF1 RF21

1

1

1

1

Phase Degrees of freedomLp Lm lp lm ls

Bright CRPM 0 - 0 0.0036 0 0.00261 0 0.0026 0 0.0013

Dark CRPM 90 - 0 0 0.001 00 1 0 0.00125 0

Bright AMPM 0 0 0.0017 0 0 0.730.0021 0 1 0.0006 0.51

Dark AMPM 90 0 0 0.001 0 00 0.00238 0 1 0

Pickoff AMPM 0 0-1.65 -0.0012

-0.00032 0 -1.3 01-0.0019 0

Black: Analytic resultsRed: Numerical simulation using “FINESSE”

lp

lsorthogonal

lp

lsdiagonal

Baseline Design for LCGT

Delocation

DMPort RF1 RF2

1 0

1 0

1 0

1 0

1 0

Phase Degrees of freedomLp Lm lp lm ls

Bright CRPM -1.3 - 0 0.00356 0 0.002561 0 -0.0026 -0.000062 -0.0013

Dark CRPM 90 - 0 0 0.001 00 1 0 0.0013 0

Bright AMPM -1.3 14.4 0.001 0 0 0-0.0017 0 1 0 0

Dark AMPM 90 -14.4 0 0.001 0 00 -0.0017 0 1 0

Pickoff AMPM -1.3 42.4 0.001 -0.001 0 010.00088 0 0 0

Option for LCGT- Could have potential advantages

Current Status

MZ locked only w/ PM FP Michelson locked Suspension system improved

DECIGO

What is DECIGO?Deci-hertz Interferometer Gravitational Wave Observatory - bridges the gap between LISA and terrestrial detectors. - could attain high sensitivity because of lower confusion noise.　

10-18

10-24

10-22

10-20

10-4 10410210010-2

Frequency [Hz]

Str

ain

[H

z-1/2]

LISA

DECIGO

Terrestrial Detectors

Confusion Noise

Pre-conceptual Design

Laser

Drag-free satellite

Arm cavity

Arm cavity

Drag-free satellite Drag-free satellite

FP-Michelson interferometerArm length: 1000 kmLaser power: 10 WLaser wavelength: 532 nmMirror diameter: 1 mMirror mass: 100 kgFinesse: 10Orbit and constellation: TBD

PD

PDKawamura, et al., CQG 23 (2006) S125-S131

Drag-free and FP Cavity

Local Sensor

Actuator

Displacement signal between the two Mirrors

Thruster Thruster

Displacement Signal between S/C and Mirror

Mirror

Requirements

[Practical force noise] 4x10-17 N/Hz per mirror

[Frequency Noise] @ 1 Hz First-stage stabilization ： 1 Hz/Hz Stabilization gain by common-mode

arm length ： 105

Common-mode rejection ratio ： 105

Science by DECIGO

NS+NS@z=1

BH+BH(1000Msun )

@z=1

Correlationfor 3 years

NS-NS (1.4+1.4Msun)•z<1 (SN>26: 7200/yr)•z<3 (SN>12: 32000/yr)•z<5 (SN>9: 47000/yr)

IMBH (1000+1000Msun)•z<1 (SN>6000)

Acceleration of Expansion of the Universe

NS-NS (z ～ 1)GW

DECIGO

Output

Expansion ＋ Acceleration?

Time

Str

ain

Template (No Acceleration)

Real Signal ? Phase Delay ～ 1sec (10 years)

Seto, Kawamura, Nakamura, PRL 87, 221103 (2001)

Roadmap for DECIGO

06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Design & Fabrication Observation

R&D Advanced R&D

PF1

DECIGO

Design & Fabrication Observation

PF2

DECIGO Pathfinder1

Local Sensor

Actuator Thruster

Objectives Drag-free system Cavity locking in space Modest sensitivity

at 0.1 – 1 Hz

DECIGO Pathfinder2

Laser

Drag-free satellite

Arm cavity

Arm cavity

Drag-free satellite Drag-free satellite

PD

PD

Objectives DECIGO with modest

specification Cavity locking between

two satellites Meaningful sensitivity

DECIGO Simulator

Clamp

Time

Ver

tica

l Po

siti

on

2m

1 sec

Release Hold Release Hold Release Hold

Objectives Continual free-fall environment Clamp release Modest sensitivity down to 0.1Hz Possibility of long arm

DECIGO Demonstrator

Satellite A Satellite B

Thruster Thruster

Actuator

Air-hockey table

Mirror A Mirror B

Local sensorLocal sensor

Objectives Lock acquisition

Budget and Working Group

Will submit a budget request ($18M for 6 years) this fall– R&D for DECIGO– PF1– w/ Pulsar Timing

DECIGO-WG: 120 members currently

MHz GW detection

Objectives and Scope

Detect GW at MHz Develop technologies for synchronous

recycling

GW Sources at MHz Inspiral of mini black holes GW from inflation period

Synchronous Recycling

Laser

Photo detector

BS

Recycling Mirror

Drever, 1983Dark Fringe

Response of Synchronous Recycling

Laser

Photo detector

BS

x

y

x

y

h

t

GWl GW 4 l

GW effect synchronously enhanced!

Plan for Table-top Experiment

l 75 cm

fGW 100 MHz

F 105

h 10-21 Hz-1/2

Integration for 1 year

h 10-26 Hz-1/2

Current Status

EOM

F 100

f 1

1 DOF to control

f 1f GW - f 1 - BW

Output

Locked w/ low Finesse

Noise Spectrum taken

Non 50:50Beam splitter

First Noise Spectrum

QND

Objectives and Scope Beat SQL using ponderomotive squeezing Use FP Michelson w/ super light mirrors

Plan1. Observe radiation pressure noise

2. Reduce radiation pressure noise w/ homodyne detection

3. Beat SQL

4. Expand the effective frequency range

Ponderomotive Squeezing Squeezed by phase change caused by reflection by free mass

Laser

Vacuum

Signal

Noise

Squeezing: frequency dependent

Cannot beat SQL w/ RF method

Caves, Walls & Milburn, Braginsky & Khalili, ...

Homodyne Detection

Local light

Laser

Homodyne phase

Signal

Noise

S/N can be improved by choosing appropriate homodyne phase!

Quantum Noise

0.1

1

10

100

1000

Sn

/ hS

QL

2

2 3 4 5 6 71

2 3 4 5 6 710

hSQL2

Shot Noise

Conventional(Arccot

Radiation pressure noise can be completely cancelled at one frequency

The frequency depends on homodyne phase

Strategy

Use super-light mirror Use high finesse

– Increase radiation pressure noise– Easier to detect

Use tuned IFO (no optical spring)

Design Parameter and Noise Estimate

Laser power 200 mW

Injected power into

the interferometer

120 mW

Finesse 7500

Front mirror mass 200 g

End mirror mass 23 mg

Diameter of the end mirror 3 mm

Thickness of the end mirror 1.5 mm

Beam radius on the end mirror 500 mm

Q factor of substrate 105

Loss angle of coating 4×10-4

Temperature 300 K

Length of the fiber 1 cm

Thickness of silica fiber 10 mm

Current Status

Homodyne detection method verified Vacuum tank ready Design of set-up complete Fiber of 10 m drawn successfully

The End