Masha Baryakhtar

Home , Black hole bomb, Pulsar timing array, Stellar black hole

GRAVITATIONAL WAVES: FROM DETECTION TO NEW PHYSICS SEARCHES

Masha Baryakhtar

New York University

Lecture 1 June 29. 2020 Today

• Noise at LIGO (ﬁnishing yesterday’s lecture)

• Pulsar Timing for Gravitational Waves and Dark Matter

• Introduction to Black hole superradiance

2 Gravitational Wave Signals

Gravitational wave strain L ⇠ L Advanced LIGO

Advanced LIGO and VIRGO already made several discoveries

Goal to reach target sensitivity in the next years Advanced VIRGO 3 Gravitational Wave Noise Advanced LIGO sensitivity

LIGO-T1800044-v5

4 Noise in Interferometers

• What are the limiting noise sources?

• Seismic noise, requires passive and active isolation

• Quantum nature of light: shot noise and radiation pressure

5 Signal in Interferometers

•A gravitational wave arriving at the interferometer will change the relative time travel in the two arms, changing the interference pattern Measuring path-length Signal in Interferometers • A interferometer detector translates GW into light power (transducer) •A gravitational wave arriving at the• interferometerIf we would detect changeswill change of 1 wavelength the (10-6) we would be limited to relative time travel in the two arms, 10changing-11, considering the interference the total effective pattern arm-length (100 km) • •Converts gravitational waves to changeOur ability in light to detect power GW at is thetherefore detector our ability to detect changes in light power

• Power for a interferometer will be given by •The power at the output is

• The key to reach a sensitivity of 10-21 sensitivity is to resolve the path- length difference in a tiny fraction, ie. 10-10

ICE 02/07/18 GW detection - M. Nofrarias 19 Measuring path-length Signal in Interferometers • A interferometer detector translates GW into light power (transducer) •A gravitational wave arriving at the• interferometerIf we would detect changeswill change of 1 wavelength the (10-6) we would be limited to relative time travel in the two arms, 10changing-11, considering the interference the total effective pattern arm-length (100 km) • •Converts gravitational waves to changeOur ability in light to detect power GW at is thetherefore detector our ability to detect changes in light power

• Power for a interferometer will be given by •The power at the output is

• The key to reach a sensitivity of 10-21 sensitivity is to resolve the path- length• differenceAdding multiple in a tiny fraction,`bounces’ ie. increases10-10 the effective length of the arms, giving a phase difference at the output of

ICE 02/07/18 GW detection2N - M.L Nofrarias2πc 19 Δφ = h c λ Signal in Interferometers

•A gravitational wave arriving at the interferometer will change the relative time travel in the two arms, changing the interference pattern •Converts gravitational waves to change in light power at the detector

• The wavelength of the LIGO laser is 1064 nm and the arms are 4km • N ~100 2NL 2πc Δφ = h c λ • A change in phase of order 1 (light to dark) then corresponds to seeing strains of 10^-12 Signal in Interferometers

• The wavelength of the LIGO laser is 1064 nm and the arms are 4km • N ~100 2NL 2πc Δφ = h c λ • A change in phase of order 1 (light to dark) then corresponds to seeing strains of 10^-12 • To see a difference in power at the level of distinguishing phases of 1 part in 10^10, need 10^20 photons (within measurement time)! Shot Noise:

•Changes in the number of photons hitting the detector can looks like the power is changing, i.e. the phase difference is changing.

• If on average the power is a kW, we have10^20 photons in 0.01 sec to measure 100 Hz frueqnecies • Then the ﬂucuations in photon number are poissonian, i.e. dN ~10^10 photons Shot Noise:

•Changes in the number of photons hitting the detector can looks like the power is changing, i.e. the phase difference is changing.

• If on average the power is a kW, we Shot noisehave10^20 photons in 0.01 sec to measure 100 Hz frueqnecies • Then the flucuations in photon number are • Since we arepoissonian, measuring i.e. power dN ~10^10 fluctuations photons at the output, these fluctuations• Theseare indistinguishable looks like changes from in mirror arm lengthdisplacements of

• And we are using mirror displacements to measure GW as the fractional length change in one arm

• So brightness ﬂuctuations are interpreted as equivalent gravitational wave noise

ICE 02/07/18 GW detection - M. Nofrarias 22 Turn up the power? • Increasing length of the interferometer increases signal but not noise. decreasing wavelength can help Shot noise Shot noise • Clearly increasing power gets rid of this issue

• And we are using• mirrorAnd wedisplacements are using mirror to measure displacements GW as theto measure fractional GW as the fractional length change in onelength arm change in one arm

• So brightness ﬂuctuations• So brightness are interpreted ﬂuctuations as equivalent are interpreted gravitational as equivalent gravitational wave noise wave noise

ICE 02/07/18 ICE 02/07/18 GW detection - M. Nofrarias GW detection - M. Nofrarias 22 22 Turn up the power? • Increasing length of the interferometer increases signal but not noise. decreasing wavelength can help Shot noise Shot noise • Clearly increasing power gets rid of this issue

• Since we are measuring• Since power we are fluctuations measuring atpower the output, fluctuations these at the output, these fluctuations are indistinguishablefluctuations are from indistinguishable mirror displacements from mirror displacements• However, increasing power increases the force on the mirror • For a perfect mirror, the radiation pressure

• So brightness ﬂuctuations• So brightness are interpreted ﬂuctuations as equivalent are interpreted gravitational as equivalent gravitational wave noise wave noise

• And we are using• mirrorAnd wedisplacements are using mirror to measure displacements GW as theto measure fractional GW force as the fractionalis equal to the power length change in onelength arm change in one arm • Fluctuations in the power give fluctuations • So brightness fluctuations• So brightness are interpreted fluctuations as equivalent are interpreted gravitational as equivalent in gravitational the force, and thus fluctuations in the wave noise wave noise Radiation pressure mirror position

• The ﬂuctuating force turns into a displacement in the test mass

ICE 02/07/18 ICE 02/07/18 GW detection - M. Nofrarias GW detection - M. Nofrarias 22 22

• which can be expressed, as in the previous case, as an equivalent gravitational wave noise

• Radiation pressure and shot noise are competing effects, what would be

the noise if we minimise this two contributions, hrp(f,P) = hshot(f,P)

ICE 02/07/18 GW detection - M. Nofrarias 24 Noise in Interferometers

• As power is increased, tradeoff between shot noise and radiation pressure noise in interferometers • Attempting to measure system more precisely (more photons) eventually causes a backreaction and disturbs the system you’re trying to measure • `Standard quantum limit’

16 Noise in Interferometers

From original LIGO proposal (`89):

17 Gravitational Wave Signals Advanced LIGO sensitivity

LIGO-T1800044-v5

18 Today

• Noise at LIGO (ﬁnishing yesterday’s lecture)

• Pulsar Timing for Gravitational Waves and Dark Matter

• Introduction to Black hole superradiance

19 Pulsar Timing Arrays

Credit: B. Saxton (NRAO/AUI/NSF) 20 Pulsar Timing Arrays

•Pulsars: highly magnetized rapidly rotating neutron stars with a coherent source of radio waves. Millisecond pulsars are the special class of pulsars with a stable rotational period of ~ 1-10 ms and stable pulse frequency.

•The stability (although not very well understood) of these millisecond pulsars can be used to make a galaxy-wide network of precise clocks, by measuring the arrival time of the pulses Vela Pulsar Cambridge University Lucky Imaging Group

21 Pulsar Timing Arrays

• Parkes Pulsar Timing Array (PPTA) observing 25 pulsars • North American Nanohertz Observatory for Gravitational Waves (NANOGrav) observating 45 pulsars • European Pulsar Timing Array (EPTA) observing 42 pulsars

• International Pulsar Timing Array (IPTA) collaboration of all three: new data release consists of 65 pulsars in 2019

22 Pulsar Timing Arrays

• Astronomers know of a few thousand neutron stars, a subset of those pulsars, with varying levels of stability • Pulsar J1909-3744 has been observed Parkes Radio Telescope for 11 years.

• During this time all 115,836,854,515 rotations can be ﬁt • The rotational period of this star is known to 15 decimal places • One of the most accurate clocks in the universe!

23 Pulsar Timing Arrays

•Pulsar timing arrays can detect low frequency gravitational waves (10−9 − 10−7 Hz), limited by observation time. Earth- pulsar systems act as a huge arm length.

24 Pulsar Timing Arrays

25 Pulsar Timing Arrays

26 Pulsar Timing Arrays

27 Pulsar Timing Arrays

•PTAs could detect the stochastic signals from supermassive black holes in the early stage of inspirals, as well as other low-frequency signals •Also measure properties of solar system

28 PTA searches for scalar dark matter

•As we know (and you learned from Francesco), we know there is dark matter in the galaxy. •One candidate is a very light scalar ﬁeld, that interacts only through gravity. •Typically this is an impossible exercise in Pulsardetection Timing Array, artists conception

Image credit: David J. Champion

29 PTA searches for scalar dark matter

•However, if the scalar ﬁeld is very light, it can cause changes in the ﬂuctuations of the metric in our galaxy which may be observable

Image credit: David J. Champion

30 PTA searches for scalar dark matter

•However, if the scalar ﬁeld is very light, it can cause changes in the ﬂuctuations of the metric in our galaxy which may be observable

Image credit: David J. Champion

31 PTA searches for scalar dark matter

•Blackboard

Pulsar Timing Array, artists conception

32 PTA searches for scalar dark matter

33 PTA searches for scalar dark matter

34 Today

• Noise at LIGO (ﬁnishing yesterday’s lecture)

• Pulsar Timing for Gravitational Waves and Dark Matter

• Introduction to Black hole superradiance

35 Superradiance and Black Holes

How to Extract Energy from Black Holes and Discover New Particles Outline

• Superradiance and rotating BHs • Gravitational Atoms • Signs of New Particles • Black Hole Spindown • GW signals

37 Motivation • Ultralight scalar particles often found in theories beyond the Standard Model

• E.g. the QCD axion solves the `strong-CP’ problem

• As already discussed, ultralight scalars can make up the DM

30 M⦿ black hole radius 108 106 104 102 1 10-2 10-4 10-6

106 CAST

SN1987A Telescopes 108 SN1987A ADMX 1010 HAYSTAC

1012 ALP dark matter 1014 QCD axion 1016 GWs, 1018 BH spindown axion 38 Planck scale scale Planck 10-14 10-12 10-10 10-8 10-6 10-4 10-2 1 Astrophysical Black Holes and Ultralight Particles

• Black holes in our universe provide nature’s laboratories to search for light particles

• Set a typical length scale, and are a huge source of energy black hole (30 M☉) Amplitude• Sensitive to QCD axions with GUT- to Planck-scale decay constants f a 100 km Amplitude

compton t t for a 103km wavelength 10-12 eV −1 x particle: kHz t frequency 39 Superradiance: gaining from dissipation part I

• A an object scattering off a rotating cylinder can increase in angular momentum and energy.

• Effect depends on dissipation, necessary to change the velocity

Ball scattering off cylinder with lossy surface slows down

40 Superradiance: gaining from dissipation part I

• A an object scattering off a rotating cylinder can increase in angular momentum and energy.

• Effect depends on dissipation, necessary to change the velocity

Ball scattering off rapidly rotating cylinder with lossy surface speeds up!

41 Superradiance: gaining from dissipation part I

• A wave scattering off a rotating object can increase in amplitude by extracting angular momentum and energy.

• Growth proportional to probability of absorption when rotating object is at rest: dissipation necessary to increase wave amplitude Superradiance condition: Angular velocity of wave slower than angular velocity of BH horizon, ⌦ < ⌦ a BH Zel’dovich; Starobinskii; Misner 42 Superradiance Gravitational wave ampliﬁed when scattering from a rapidly • Arotating wave scattering black hole off a rotating object can increase in amplitude by extracting angular momentum and energy.

• Growth proportional to probability of absorption when rotating object is at rest: dissipation necessary to change the wave amplitude Superradiance condition: Angular velocity of wave slower than angular velocity of BH horizon,

⌦a < ⌦BH

NumericalZel’dovich; GR simulation Starobinskii; by Will Misner East 43 Superradiance Particles/waves trapped in orbit around the BH repeat this process continuously

Press & Teulkosky

“Black hole bomb” exponential instability when surround BH by a mirror

Kinematic, not resonant condition

Superradiance condition: Angular velocity of wave slower than angular velocity of BH horizon,

⌦a < ⌦BH

44 Superradiance Particles/waves trapped in orbit around the BH repeat this process continuously

Press & Teulkosky

“Black hole bomb”: exponential instability when surround BH by a mirror

Kinematic, not resonant condition

Superradiance condition: Angular velocity of wave slower than angular velocity of BH horizon,

⌦a < ⌦BH

45 Superradiance

• −1 Particles/waves trapped μa near the BH repeat this process continuously t

• For a massive particle, e.g. axion, gravitational potential barrier provides trapping

G M μ V(r) = − N BH a r • For high superradiance rates, compton wavelength should be rg comparable to black hole radius:

11 −1 6 10 eV r ≲ μa 3km ⇥ g a ⇠ µa [Zouros & Eardley’79; Damour et al ’76; Detweiler’80; Gaina et al ’78] [Arvanitaki, Dimopoulos, Dubovsky, Kaloper, March-Russell 2009; Arvanitaki, Dubovsky 2010] 46 Gravitational Atoms

Axion Gravitational Atoms

G M μ V(r) = − N BH a r n = 1, ℓ = 0, m = 0 n = 2, ℓ = 1, m = 1 n = 3, ℓ = 2, m = 2

Gravitational potential similar to hydrogen atom

`Fine structure constant` Radius Occupation number 2 n 75 80 α ≡ GNMBHμa ≡ rgμa rc ≃ ∼ 4 − 400rg N ∼ 10 − 10 α μa

47 Gravitational Atoms

Axion Gravitational Atoms

G M μ V(r) = − N BH a r n = 1, ℓ = 0, m = 0 n = 2, ℓ = 1, m = 1 n = 3, ℓ = 2, m = 2

Gravitational potential similar to hydrogen atom

`Fine structure constant` Radius Occupation number 2 n 75 80 α ≡ GNMBHμa ≡ rgμa rc ≃ ∼ 4 − 400rg N ∼ 10 − 10 α μa Boundary conditions at horizon give imaginary frequency: exponential growth for rapidly rotating black holes α2 E ≃ μ 1 − + iΓsr ( 2n2 ) 48 Superradiance: a stellar black hole history

A black hole is born with spin a* = 0.95, M = 40 M⦿. 1.0 ℓ = 1 ℓ = 2 BH rotates quickly 0.8 enough to ℓ = 3 BH lightcrossing time ⌧ SR

* superradiate a > 100 0.2 msec (60 km) 0.6 Spin ⌧ BH BH ⌦

Hole Particle wavelength 0.4 < a ⌦ 1 msec (300 km) Black 0.2 t

-13 ma=6*10 eV 0.0 20 40 60 80 100 120 140

Black Hole Mass Mü 49

H L Superradiance: a stellar black hole history

BH spins down and fastest-growing level is formed Cloud radius Once BH angular velocity matches that of the level, growth stops 6 msec (2000 km) 1.0

0.8 BH lightcrossing time * a 0.2 msec (60 km) 0.6 ~ 1 year (~180 superradiance times) Spin

Hole Particle wavelength 0.4 no longer satisﬁes l=1 SR condition 1 msec (300 km) Black 0.2 t

-13 ma=6*10 eV 0.0 20 40 60 80 100 120 140

Black Hole Mass Mü 50

H L Superradiance: a stellar black hole history

~1078 particles Cloud can carry up to a few percent of the black hole mass: huge energy density 1.0 black hole angular 0.8 momentum * a decreases by{

0.6 1078 ℏ ~ 1 year Spin

Hole 0.4 no longer satisﬁes l=1 SR condition Black 0.2

-13 ma=6*10 eV 0.0 20 40 60 80 100 120 140

Black Hole Mass Mü 51

H L Superradiance: a stellar black hole history

Annihilations to GWs deplete ﬁrst level Gravitational waves can be observed in LIGO continuous wave searches 1.0 Signals fall into ongoing searches for 0.8 gravitational waves

* from asymmetric a rotating neutron 0.6 ~ 1 year

Spin stars Up to thousands of Hole 0.4 no longer satisﬁes l=1 SR condition observable signals annihilation signal lasts 3000 years above current LIGO Black upper limits — lack of 0.2 observation disfavors a

-13 range of axion masses ma=6*10 eV 0.0 20 40 60 80 100 120 140

Black Hole Mass Mü 52

H L