Early History of Gravitational Wave Detectors

Transcription

1 Early History of Gravitational Wave Detectors Ho Jung Paik University of Maryland Gravitational Waves: New Frontier Seoul, Korea, Jan, 18-20, 2013

2 Disclaimer In this talk, I will report my own experience in the development of resonant-mass gravitational wave detectors and discuss its early history from the perspective of an insider. I will talk very little about laser interferometer detectors. For excellent, more comprehensive, reviews, see Peter R. Saulson, Physics of Gravitational Wave Detection: Resonant and Interferometric Detectors, in Proceedings of the 26th SLAC Summer Institute Gravity: From the Hubble Length to the Planck Length, ed. Lance Dixon, SLAC-R-538, pp (1998). Odylio D. Aguiar, The Past, Present and Future of the Resonant-Mass Gravitational Wave Detectors, Res. Astron. Astrophys. Vol. 11, 1 42 (2011). 18-Jan-13 Paik 2

3 From Newton to Einstein G mn = 8pT mn Newton s universal law of gravitation Instantaneous action at a distance 18-Jan-13 Paik Einstein s General Relativity Information carried at the speed of light Gravitational wave 3

4 Gravitational Waves c t 2 2 h mn 0, gmn mn hmn GW is also transverse But importance difference: Gravitational charge m > 0 Quadrupole wave w/ spin 2 Plus polarization Cross polarization 18-Jan-13 Paik 4

5 Does GW Couple to Masses? Gravitational wave is a ripple in space-time fabric itself. So, does it deposit energy to masses? 18-Jan-13 Paik 5

6 1 st Gravitational Wave Detector Joseph Weber ( ) Coincidence between two 1660-Hz antennas at Argonne and Maryland 18-Jan-13 Paik 6

7 Flurry of Room-Temp Detectors 1. Braginsky (Moscow State, 1972) Two 1640-Hz antennas (1300 kg) with capacitive transducer 2. Tyson (Bell Lab, ) and Douglas at (Rochester) Two 710-Hz antennas (3600 kg) with PZT 3. Drever (Glasgow, ), at Bristol and Reading Two 1000-Hz split-bars (300 kg) with PZT in between 4. Bramanti (Frascati, 1973) and Billing & Winkler (MPI) Two 1660-Hz antennas (1580 kg) with PZT 5. Garwin and Levine (IBM, 1973) Single 1695-Hz antenna (118 kg) with PZT 6. Hirakawa (Tokyo, 1973) Two 145-Hz square antennas (1400 kg) with cap transducer By 1973, they began to report null results. Sensitivities of these room-temp detectors were h ~ If Weber s events were GWs, the universe would evaporate into GWs in 10 6 yrs! 18-Jan-13 Paik 7

8 Birth of Cryogenic GW Detector In 1970, William Fairbank at Stanford University proposed a 3-mK antenna. I was a theory graduate student looking for a summer job and got recruited into his group. Steve Boughn and I persuaded Fairbank to set a milestone at 4.2 K before we go onto 3 mk. Steve was assigned to levitate the antenna magnetically. I was assigned to develop a superconducting inductive transducer. 18-Jan-13 Paik 8

9 Discovery of Resonant Transducer Fairbank s original suggestion: I 0 Levitate a superconducting test mass at one end of the antenna and measure the displacement of the antenna relative to it. Sensing Coil Superconducting Disk Levitation Coil Sensing Coil SQUID I discovered (1971): The persistent current in sensing circuit provides stiffness to the test mass. It the test mass (m) were resonant with the antenna (M), the amplitude would be gained by M/m. What happens? Two resonators beat. Energy is transferred from the antenna to the test mass: 1 2 Mω 0 2 x 2 = 1 2 mω 0 2 y 2 y x = M m 18-Jan-13 Paik 9

10 Principle of Resonant Transducer ω 0 ω + - ω B = ω + ω = ω 0 τ = T B 2 = π ω 0 M m m M Eventually, all the cryogenic resonant-mass detector groups adopted the resonant transducer. 18-Jan-13 Paik 10

11 Summer School at Lake Como The discovery was presented at International School of Physics Enrico Fermi, on Experimental Gravitation (Varenna, Italy, July, 1972). 18-Jan-13 Paik 11

12 MOUNTING FLANGE PROOF MASS DC SQUID AIR FILTER SUPERCONDUCTING COILS ANTENNA CIRCUIT BOARD 18-Jan-13 Paik 12 S/C Resonant Transducer , m m 2 0 d m A B L L md A B p m m Additional benefit: Nb diaphragm had very high Q (~10 7 ) Niobium Diaphragm

13 Cryogenic GW Detectors 1. Fairbank and Michelson (Stanford, ) 4-K 900-Hz antenna (4800 kg) w/ inductive transducer 2. Hamilton and Johnson (LSU, ), Allegro 4-K 900-Hz antenna (2300 kg) w/ inductive transducer 3. Pizzella (Rome Geneva, ), Explorer 2.6-K 900-Hz antenna (2300 kg) w/ capacitive-squid 4. Blair (W Australia, ), Niobe 5-K 700-Hz Nb antenna (1500 kg) w/ s/c cavity transducer 5. Hirakawa (Tokyo, 1995) 4-K 60-Hz torsion antenna (1200 kg) w/ capacitive transducer 6. Pizzella and Coccia (Rome, ), Nautilus 0.1-K 930-Hz antenna (2300 kg) w/ capacitive-squid 7. Cerdonio and Vitale (Padova, ), Auriga 0.2-K 930-Hz antenna (2300 kg) w/ capacitive-squid By late 1980 s, they began to report null results. These cryogenic detectors reached h ~ Jan-13 Paik 13

14 Stanford GW Detectors 4-K detector, mK detector, never completed 18-Jan-13 Paik 14

15 Other Cryogenic Detectors Allegro USA Auriga, Italy Niobe Australia Nautilus, italy Explorer Switzerland 18-Jan-13 Paik 15

16 GW Detector Theory 1. Gibbons and Hawking (1971): For bursts, one should be able to detect E s k B T ω a τ Q To achieve the best noise, maximize energy coupling: β = Energy coupled to the circuit, βe Total energy in the detector s βk B T ω τ a Q + C 2v T 2 1 τ 2. Braginsky (1974): The detector suffers a back-action noise. Quantum non-demolition to overcome the quantum limit. 18-Jan-13 Paik 16

17 3. Giffard (1976): GW Detector Theory Transducer modeled as linear two-port with v(t) and f(t) as input variables and I(t) and V(t) as output variables E n = 2k B T τ τ a + Z M S i ω τ + 2M Z 21 2 S e ω τ, β = Z 12 Z 21 Mω Z 22 For passive transducers impedance matched to amplifier, E n 2k BT τ τ a + k B T N βω a τ βω a τ = 2k BT βq a + 2k B T N, τ = 2 βω a For linear amplifiers, 2k B T N 2 a Standard quantum limit 18-Jan-13 Paik 17

18 Resonant-Mass Detector Condition to detect a GW pulse with strength h : π 2 a 2 Dh k T k T 2 2 S M 2 S B a B N Qa 2 Optimal strategy: S 2 S T a, EN 2kB N S T S Qa Large increases and reduces thermal Richard (1979): 3-mode detector: Insert a mass m i = between Mand m, T B 2 = π ω 0 M m i = π ω 0 S M m mm 1/4 noise. 18-Jan-13 Paik 18

19 Strain Sensitivity ([Hz] -1/2 ) 2-mode Transducer for Allegro 1.0E-17 Calculated Strain Spectrum of 4 Transducers 1.0E-18 LSU 2-mode m 1 = 1150 kg, Q 1 = 9 x 10 6 m 2 = 5.35 kg, Q 2 = 3 x 10 6 m 3 = kg, Q 3 = 3 x E E E-21 SQUID noise: 300 ħ 0 /k B (two-stage QD dc SQUID) f = 100 Hz, h = E Jan-13 Paik Frequency (Hz) 19 UMD 2-mode LSU 3-mode Current design UMD 3-mode

20 2-mode Transducer for Auriga Last stage: capacitor transducer tuned with inductance m 1 = 1150 kg, Q 1 = 6.4 x 10 6 m 2 = 3.5 kg, Q 2 = 8 x 10 5 m 3, eff = kg, Q 3 ~ 10 6 h < Hz -1/2 within f = 100 Hz SQUID noise: 1000 ħ 0 /k B (two-stage QD dc SQUID) 18-Jan-13 Paik 20

21 Network of Resonant Bars Allegro Explorer Auriga Nautilus IGEC Network Niobe 18-Jan-13 Paik 21

22 International Gravitational Event Collaboration (IGEC) ALLEGRO, AURIGA, EXPLORER, NAUTILUS, & NIOBE The search for bursts at resonance frequency ~ 900 Hz. Candidate events at SNR > 3-5 (background ~ 100/day). Data exchanged: peak amplitude, time of event and uncertainties. Threshold equivalent to ~0.02 M converted into a GW millisecond burst at a distance of 10 kpc. The accidental coincidence rate over 1-s interval (i.e., bandwidth of 1 Hz) was ~ few/week two-fold and ~few/century three-fold. No evidence for gravity wave bursts was found. 18-Jan-13 Paik 22

23 IGEC Coincidence Search P. Astone et al. Phys. Rev. D 68, (2003) Rate (y 1 ) The area above the blue curve is excluded with a coverage > 90% Search threshold h Upper limit on the rate of GW bursts from Galactic Center h ~ E ~ 0.02 M 10 kpc 18-Jan-13 Paik 23

24 Spherical Antenna 1. Forward (1971), Wagoner & Paik (1976): Full-sky coverage w/ uniform sensitivity By detecting its 5 quadrupole modes, source direction (, ) and polarization (h+, h) can be determined. Have larger cross section than a bar (by 70x) at the same frequency. 2. Johnson & Merkowitz (1993): 6 radial transducers on truncated icosahedral configuration maintains the degeneracy of the quadrupole moments. TIGA antenna 3. Cerdonio et al. (2001): A wideband detector could be constructed by a dual sphere. 18-Jan-13 Paik 24

25 TIGA USA Resonant Spheres MiniGrail The Netherlands Schenberg Brazil Sfera Italy 18-Jan-13 Paik 25

26 Moon as GW Detector? Weber (1972) sent a LaCoste- Romberg gravimeter to the Moon to search for low-frequency GWs. The gravimeter failed to operate. Paik and Venkateswara (2004) revisited the possibility of instrumenting the Moon as a spherical antenna. Advantages of the Moon: Very quiet seismically. Moonquakes could be identified and removed. The Moon is a very good resonator with Q ~ Jan-13 Paik 26

27 Moon as Spherical Detector A 10-kg s/c disk is levitated and its horizontal displacement is sensed in two directions with a superconducting circuit. I 0 Sensing Coil Superconducting Disk Sensing Coil SQUID Levitation Coil Transducer configurations: Energy summed over transducers Tetrahedral Icosahedral Paik-27

28 Overcoming the Limit by Light The sensitivity of acoustic detectors is limited by the mass and sound velocity: 2 π 2 Mω a 2 Dh 2 2k B T N h k BT N Mv 2 This limit can be overcome by connecting free test masses separated far away by a laser beam. 18-Jan-13 Paik 28

29 Laser Interferometer Detector 1. Gertsenshtein and Pustovoit (1962) Suggested Michelson interferometer as a superior detector 2. Forward and Miller (Hughes Lab, 1971) Built 8.5-m prototype interferometer with a single bounce 3. Weiss (MIT, 1972) Detailed design study of interferometer GW detector with noise and error analysis 4. Billing, Rüdiger and Winkler (Munich, 1979) Built delay-line interferometer with multiple reflection 5. Drever and Hugh (Glasgow, 1979) Built Fabri-Perot cavity with power and signal recycling This early work paved way to the construction and operation of TAMA, GEO, LIGO, and VIRGO. Now, Advanced LIGO, Advanced VIRGO, and KAGRA are under development. 18-Jan-13 Paik 29

30 Interferometer Hardware Fused silica mirror 6-W Nd:YAG laser 11-Aug-06 UKC Paik 30

31 Evolution of LIGO Sensitivity 18-Jan-13 Paik 31

32 TAMA Japan 300m Interferometer Detectors LIGO Louisiana 4000m Virgo Italy 3000m LIGO Washington 2000m & 4000m GEO Germany 600m 11-Aug-06 UKC Paik 32

33 Network of Interferometers LIGO GEO Virgo TAMA decompose detection locate the the confidence polarization sources of gravitational waves AIGO 18-Jan-13 Paik 33

34 Advanced LIGO Multiple Suspensions Active Seismic Sapphire Optics Higher Power Laser 18-Jan-13 Paik 34

35 Advanced LIGO Enhanced Systems laser suspension seismic isolation test mass Rate Improvement ~ 10 4 narrow band optical configuration 18-Jan-13 Paik 35

36 Gravitational Waves in Space LISA Three spacecraft form an equilateral triangle with sides 5 million km in length. 18-Jan-13 Paik 36

37 Conclusions Gravitational wave detection has been one of the greatest challenges in the past half a century. Many brilliant physicists have developed new instruments and pushed the frontiers of measurement technology. Sensitivity toward gravitational wave detection is improving on many fronts. Advanced LIGO, advanced VIRGO, and KAGRA under construction will be sensitive enough to detect GWs from > 100 Mpc. Improved upper limits are being set for all major sources -- binary inspirals, periodic sources, burst sources, and stochastic background. Hopefully, detections will be made soon!! 18-Jan-13 Paik 37