Black holes: observations Lecture 4: BH coalescence Sergei Popov (SAI MSU)

2 Plan of the lecture 1.Hierarchical model of galaxy formation. 2.Gravitational wave rocket. 3.Black holes at large redshifts. 4.Coalescence of SMBHs. 5.BH coalescence in binaries. Reviews astro-ph/ Formation and early evolution of massive black holes astro-ph/ The Formation and Evolution of the First Massive Black Holes astro-ph/ Supermassive black hole mergers and cosmological structure formation

3 Structure growth in the universe Today the standard model of the structure formation is the hierarchical one. Numerical calculations of the evolution of the large-scale structure and single blocks reached a very high level of precision (arxiv: ).arxiv:

4 Large scale structure (Kravtsov et al.)

5 Large scale structure

6 Formation of clusters of galaxies (Kauffmann, Colberg, Diaferio, and White) tCDM LCDM 21x21 (Mpc/h) 3 35x35 (Mpc/h) 3 In the process of structure growth numerous coalescence of building blocks happen, each of these blocks can contain a BH. After a coalescence a new-formed BH slowly, due to dynamical friction, moves towards the center of the resulting structure. Formation of large galaxies is finished as z~2, after this no major mergers happen, only small satellites are captured by big galaxies.

7 Growth of clusters of galaxies

8 Minihalos and the first stars (Volker Bromm astro-ph/ )astro-ph/ Symbols indicate minihalos. Open symbols – Cooling is not effective enough. The critical line corresponds to equality between the cooling time and dynamical evolution time scale of a minihalo (free-fall time). This line separates dark halos and halos that can produce stars. In each minihalo a very small number of stars in formed.

9 The first stars and minihalos In the standard λCDM model the first massive BHs are formed at z>15 in minihalos with M> M O. These BHs produce the first miniquasars, which contribute to the reionization at z~ Such low mass of minihalos is explained by the role of molecular hydrogene (Tegmark et al. 1997). The first stars with masses solar and >260 solar masses produce BHs. A BH mass (in the case of the first stars) is typically >0.5 of the mass of a star.

10 The first stars (Yoshida et al. astro-ph/ )astro-ph/ Calculations have been done in the ΛCDM model. The picture is plotted for z=17. The size is 50 kpc. Stars are formed on the cross-sections of filaments (bright dots).

11 The problem of the existence of very massive BHs at high redshifts At redshifts ~7 already there are SMBHs with masses ~10 9 M 0. These redshifts correspond to the age of the universe 15 and to provide their rapid growth (note, that the accretion rate is limited by the Eddington rate).

12 BH mass growth (Madau astro-ph/ ) BH mass growth for different accretion efficiency. Halo mass functions at different z. These galaxies due to coalescence produce at z 0 =0.8 a Milky Way-like galaxy (10 12 solar masses, solid curves), or a slightly smaller one at z 0 =3.5 ( , dashed curves).

13 Mass growth lg М ~M 2 (Bondi) ~M (Eddington) lg t At first the mass is growing rapidly according to the Bondi formula. Then, when the Eddington limit is reached, the growth slows down. The so-called Salpeter time: the time in which the mass is doubled. Accretion and coalescence are both important for the mass growth. Now SMBHs in giant elliptical galaxies increase their masses mostly due to coalescence with satellites.

14 Mass growth, spin and activity Elliptical Disc (arXiv: ) Some time ago it was noticed, that radio emission from elliptical galaxies is stronger, than from disc galaxies. It was proposed that this can be related to faster rotation of central BHs in elliptical galaxies. Recent calculations (see the plot) demonstrated that it can be true. The reason is that the mass growth of BHs in ellipticals happen via more powerful episodes of accretion.

15 Evolution of SMBHs activity (Comastri astro-ph/ )astro-ph/ The plot shows the redshift distribution of AGNs detected by Chandra and XMM-Newton. The top histogram: all sources from the joint sample of Chandra and XMM. Red hatched region: sources identified in optics. Solid curve: results of modeling. The cutoff at high redshifts is not an artifact.

16 Evolution of quasars number (Combes astro-ph/ ) Very bright quasars are formed very early, and then their number is decreasing. For AGNs with low luminosities the evolution is not so pronounced, but anyway it is evident. For luminosities erg/s the maximum is at z~ , for erg/s – at z~2.

17 Mass and luminosity evolution (see Combes astro-ph/ ) Results of numerical simulations are shown (Merloni 2004). Lifetime grows with decreasing z.

18 Galactic cannibalism (C.Nipoti et al. astro-ph/ )astro-ph/ Results of calculations for the evolution of galaxies in the center of the cluster C On the left the present day (observed) configuration is shown. On the right – results of calculations for two models. Two variants of calcualtions differ by the way the DM particles are treated. The upper one seems to be more realistic. Dynamical friction is important.

19 Interacting galaxies (Hibbard, Barnes)

20 Milky Way and dark satellites (Madau astro-ph/ ) There is the so-called satellite problem: there are too many satellites in computer models of galaxy formation. The size of the shown region is 800 to 600 kpc (and 600 kpc deep). 110 million particles in this cube. Other models produce smaller numbers of satellites, but the problem is still there.

21 Gravitational wave rocket astro-ph/ How black holes get their kicks? In addition to energy and angular moment, gravitational waves also take away the linear momentum. So, the object formed via a coalescence gets a kick. The first estimate of the effect in the case of binaries was obtained in 1983 by Fitchett: Recently, this topic became very hot due to calculations in the framework of the hierarchical model. Continuosly new results appear to improve the formula above One of the first articles in the new wave was the paper by Favata et al. f(q)=q 2 (1-q)/(1+q) 5, f max =0.38

22 Favata et al. (2004) a/M=0.8, q=0.127 (rotation of the smaller BH is neglected) The velocity is high enough to escape from a not very massive halo, or to shake a central SMBH.

23 Maximum kick (Campanelli et al. gr-qc/ ) The velocity is strongly dependent on the relative orientation of BHs spins prior to coalescence.

24 Stellar density profile evolution on the case of two BHs (Combes astro-ph/ ) Flat profiles can be explained by an existence of the second BH.

25 Binary supermassive BHs (Rodriguez et al. astro-ph/ ) Galaxy Total mass: M 0 Distance between two BHs is 7.3 pc.

26 Examples of binary SMBHs 3С75 Abell 400

27 Neutron stars coalescence (Stephan Rosswog, visualisation: R. West)

28 Coalescence of BHs in binary systems Unfortunately, at the moment we do not know any systems with two compact objects, one of which is a BH. It is very difficult to identify a system with two BHs. However, models of the stellar evolution show, that such systems are quite natural result of binary evolution. Also, systems BH+NS can exist. Calculations show that systems BH+PSR should be relatively abundant (one system per several thousand PSRs). On one hand, systems with BHs are more rare than NS+NS systems, on the other hand, due to larger masses GW signal is much more powerful. So, coalescence of BHs can be observed from much larger distances. Probably, the first events to be registered by GW detectors like VIRGO and LIGO are coalescence in NS+BH systems. (see, for example, Lipunov et al

29 Last orbits of BHs astro-ph/ It is important to calculate in advance so-called waveforms. Otherwise, it is very difficult to identify the signal. Waveforms in the case of BH coalescence should be different from NS+NS coalescence.

30 Fall-down of matter onto a BH and GW emission See also gr-qc/ An Effective Search Method forgr-qc/ Gravitational Ringing of Black Holes In this paper the authors calculated a family of waveforms for effective search for gravitational ringing of BHs.

31 Registration of GW LIGO Detectors LIGO and VIRGO are ready and scientific runs are in progress. However, the sensitivity is not enough. Upgrade of detectors is forseen.

32 LISA The mission is approved. Launch in >2017. Unlike LIGO and VIRGO, which were build to observe coalescence of stellar mass BHs and NSs, LISA is sensitive an lower frequencies. The aim is to detect coalescence of SMBHs.

33 Main conclusions The first massive BHs are formed from the first massive stars at redshifts >15 in minihalos with masses about 10 6 М О. Halos (and BHs inside them) coalesce with each other in the process of hierarchical merging. Mass growth of BHs is due to accretion and coalescence. Already at z>6 there are SMBHs with masses ~10 9 М О. The GW rocket effect is important, especially early in the merging history, as at that time potentials were not so deep. Observations of GW signals are possible with detectors like VIRGO and LIGO (for stellar mass objects), and with LISA (in the case of SMBHs).