Ho Jung PaikDepartment of Physics, University of Maryland

GWADW, May 26, 2015

(delivered by Krishna Venkateswara, U Wash)

New Low-Frequency GW Detectorwith Superconducting Instrumentation

LIGO-G1500670

2

Contents

Introduction to Superconducting Gravity Gradiometers (SGG)

SGG as a Gravitational-Wave Detector

Current and Future SGGs and Sensitivity

SOGRO: Design and Important Noise Sources

Sensitivity Goal

3

Introduction

4

Introduction to Superconducting Gravity Gradiometer (SGG)

2

jiij xx

SQUID output is proportional to the difference in displacement of the two masses

i=j : inline-componentij : cross-component

Gravity gradienttensor

Venkateswara

Venkateswara 5

Test of Newton’s Inverse-square law using SGG (1993)

0 zzyyxxIn free space,

Illustration of Test

Cryostat Schematic

2 limits on new Yukawa interaction

Poisson’s equation

6

Gravitational Wave Detection

Paik 7

SGG as a Gravitational Wave Detector? In the Newtonian limit, Riemann tensor becomes gravity

gradient. Gravity gradiometer measures Riemann tensor components. /2

00 jiijji xxR

L11 L12 L21 22L

I1 I2

ax

xxm1 m2

8

Current and Future SGGs

9

UM diagonal-component SGG (1993)

Test masses are mechanically suspended (fDM ~ 10 Hz).

Development was completed by early 1990s. 100 times better amplitude sensitivity than TOBA 20 years

earlier.

104 times better limit than TOBA in GW energy density.

3 x 10-11 Hz-1/2

TEST MASS 2

TEST MASS 1

SQUID

L t1

LS 3

R S P

LL 1

LL 2

LS 1

LS 2

R S S

R L

R B S

LB 2

LB 1

ILIS 2

IS 1

IS 2 - IS 1

IB 2

IB 1

R B P

L t2

(Moody et al., 2002)

Paik

10

Cross-component SGG (2011)

3-axis Cross-component SGG Cryostat Schematic

Gravity gradient sensitivityVenkateswara

1E = 10-9 s-2

GW strain sensitivity

3 x 10-10 Hz-1/2

Paik 11

New Superconducting tensor gravity gradiometer

More sensitive SGG is under development with NASA support.

Test masses are magnetically suspend (fDM ~ 0.01 Hz).

Very high sensitivity

Six test masses mounted a cube form a tensor gradiometer.

Test masses are levitated by a current induced along a tube.

12

SOGRO:Superconducting Omni-directional

Gravitational Radiation Observatory

Paik 13

Superconducting tensor gravitational wave detector

Each test mass has three degrees of freedom.

Combining six test masses, a tensor GW detector is formed.

By detecting all six components of the Riemann tensor, the source direction (, ) and wave polarization (h+, h) can be determined by a single detector. “Spherical” Antenna

SOGRO(Superconducting Omni-directional Gravitational Radiation Observatory)

Paik 14

Suspension of SOGRO

Go underground to reduce seismic and gravity gradient noise, as well as to be far away from moving objects.

Nodal support prevents odd harmonics from being excited.

Cable suspension gives fr < 103 Hz for three angular modes.

Active isolation unnecessary.

25-m pendulum gives fp = 0.1 Hz for two horizontal modes.

Provide passive isolation for high frequencies.

Problem: Platform is not rigid enough.

Triangulate with struts.

Alternative suspension: Optical rigid body Simpler cryogenics, larger baseline (up to 3 km?)

Paik 15

Magnetic levitation

Stanford (1976)

Field required to levitate 5-ton mass:

The biggest challenge:

To obtain symmetry, vertical DM resonance frequencies must also reduced to 0.01 Hz.

Employ “push-pull levitation” plus negative spring.

1 ton Al antenna wrapped with Nb-Ti sheet was levitated magnetically.

K). 4 at (Nb T 16.0 T 11.0

10

8.910510π422 ,

2

1

2/1372/1

0

0

2

TH

A

MgBMgA

B

c

Push-pull levitation

Superconducting negative spring

Paik 16

Parametric transducer

Near quantum-limited SQUIDs have 1/f noise below 100 kHz.

Signal needs to be upconverted to fp = 100 kHz by pumping at fp by using a parametric transducer.

An inductance bridge can be coupled naturally to a SQUID.

Physical Review D

Paik 17

Tuned capacitor-bridge transducer

Capacitor bridge coupled to a near quantum-limited SQUID thru S/C transformer.

LC resonance increases energy coupling by Qp .

Oscillator noise is rejected by the bridge balance.

Maintain precise bridge balance by feedback.

NBp

D

D

DBN Tk

Q

TkfE

2/1

2

221

1)(

Parametric gain:

Energy coupling constant: 2

2

M

QCE pp

mpmp ZZG /// 11

Paik 18

Achievable detector noise

pNBNBp

D

D

DBh nTkTk

Q

Tk

MLfS

,

11

8)(

2/1

2

22

42

Parameter SOGRO 1 SOGRO 2 Method Employed (SOGRO 1 /2)

Each mass M 5 ton 5 ton Nb square tube

Separation L 30 m 100 m Over “rigid” mounting platform

Antenna temp T 1.5 K 0.1 K Superfluid He / dilution refrigerator

DM frequency fD 0.01 Hz 0.01 Hz Magnetic levitation w/ negative spring

DM quality factor QD 108 109 Surface polished pure Nb

Signal frequency f 0.1-10 Hz 0.1-10 Hz Detector noise computed at 1 Hz

Pump frequency fp 50 kHz 50 kHz Tuned capacitor bridge transducer

Amplifier noise no. n 200 10 Near-quantum-limited SQUID

Detector noise Sh1/2(f ) 21020 Hz1/2 21021 Hz1/2 Two phase development

For CW signal with impedance-matched bridge transducer,

SOGRO can also be operated as tunable resonant detector.

D

B

Dp

D

D

NB

D

B

D

h Q

Tk

MLQ

Tk

Q

Tk

MLfS

8

8)(

3232

Paik 19

Seismic noise

Seismic noise of underground sites

Seismic noise of bedrock: 107 m Hz 1/2

Could be reduced to the required level by combining active isolation with CM rejection of the detector.

20-m pendulum with nodal support Passive isolation for f > 0.1 Hz.

107

108

109

1010

1011

1012

1013

1014

1015

1016

1017

1018

1019

Seismic background

Active isolation

_____________

Axis alignment and scale factor match

_____________

Error compensation

Detector sensitivity

_____________A

mp

litu

de

no

ise

leve

l (m

Hz

1/2 )

The Newtonian noise from seismic and atmospheric density fluctuations cannot be shielded.

GWs are transverse and cannot have longitudinal components whereas the Newtonian gradient does.

Could be distinguished in principle from near-field Newtonian gradients.

Paik 20

Newtonian gradient noise

.

)()()(

)()()(

)()()(

)( ,

)()(0

)()(0

000

)(

332313

232212

131211

ththth

ththth

ththth

th

thth

ththth NGGW

Paik 21

Sensitivity Goal

Paik 22

Sensitivity goals of SOGRO

Detector cooled to 1.5 K or 0.1 K and integrated with near-quantum-limited amplifiers.

Seismic noise rejected by 1012 by combining CMRR, passive & active isolation.

Newtonian noise rejected by 102~104 using tensor nature of the detector and external sensors.

SOGRO 1: T = 1.5 K, SOGRO 2: T = 0.1 K

Major challenges:Large-scale cryogenics.Mitigation of Newtonian noise.

Paik 23

Thank you!