Astrophysical Problems andPrimordial Black Holes

A.D. DolgovNSU, Novosibirsk, 630090, Russia

ITEP, Moscow, 117218, Russia

Les Rencontres de Physique de la Valle d’Aoste

Results and Perspectives in Particle Physics La

Thuile, Aosta Valley, Italy

March 5-11, 2017

1

Recent astronomical observations inz ∼ 10 universe opened a box of sur-prises which are in striking conflictwith canonical expectations.The early universe is found to be denselypopulated with huge bright galaxies,super-heavy black holes (QSOs), su-pernovae (gamma-bursters), and is verydusty. All that simply did not haveenough time to be created.Similar problems are also discoveredin the present day universe.Solution in the standard theory is un-known.

2

All the mysteries are explained by amodel of very early heavy PBH for-mation with the predicted log-nomalmass-spectrum:

dN

dM= CM exp [−γ ln2(M/M0)],

AD, J. Silk, 1993;AD, M.Kawasaki, N.Kevlishvili, 2006(?)S.I.Blinnikov, AD, K.A.Postnov, 2014;SIB, AD, N.Poraiko, KAP 2016;AD, K.A.Postnov, 2017

3

Content (review of the puzzles andthe mechanism of solution):1. Discoveries in young hi-z universe.2. Puzzling features of the old con-temporary universe.3. GW discovery by LISA and emerg-ing problems.4. A model of massive PBH creation.5. Predictions and conclusion.

4

I. A brief review of high-z discoveries.

1. Several galaxies have been observedwith natural gravitational lens “tele-scopes. A few examples:a galaxy at z ≈ 9.6 which was createdat tU < 0.5 Gyr;a galaxy at z ≈ 11 has been detectedat tU ∼ 0.4 Gyr, three times more lu-minous in UV than other galaxies atz = 6− 8.

5

Not so young but extremely luminousgalaxy L = 3 · 1014L�; tU ∼ 1.3 Gyr.The galactic seeds, or embryonic blackholes, might be bigger than thoughtpossible. P. Eisenhardt: ”How do youget an elephant? One way is startwith a baby elephant.” The BH wasalready billions of M� , when our uni-verse was only a tenth of its presentage of 13.8 billion years. ”Anotherway to grow this big is to have goneon a sustained binge, consuming foodfaster than typically thought possible.”Low spin is necessary!

6

According to the paper ”Monsters inthe Dark”: density of galaxies atz ≈ 11 is 10−6 Mpc−3, an order ofmagnitude higher than estimated fromthe data at lower z.Origin of these galaxies is unclear.

7

2. Supermassive BH and/or QSO.About 40 quasars with z > 6 are al-ready known, each quasar containingBH with M ∼ 109M�.The maximum z is z = 7.085 i.e. thequasar was formed before the universereached 0.75 Gyr withL = 6.3 · 1013L�, M = 2 · 109M�,Similar situation with the others.The quasars are supposed to be su-permassive black holes and their for-mation in such short time by conven-tional mechanisms looks problematic.

8

Such black holes, when the Universewas less than one billion years old,present substantial challenges to the-ories of the formation and growth ofblack holes and the coevolution of blackholes and galaxies. Even the origin ofSMBH in contemporary universe dur-ing 14 Gyr is difficult to explain.Non-standard accretion physics andthe formation of massive seeds seemto be necessary. Neither of them isobserved in the present day universe.

9

Very recently another monster was dis-covered ”An ultraluminous quasar witha twelve billion solar mass black holeat redshift 6.30”. There is alreadya serious problem with formation oflighter and less luminous quasars whichis multifold deepened with this new”creature”. The new one with M ≈1010M� makes the formation abso-lutely impossible in the standard ap-proach.

10

3. Dust, supernovae, gamma-bursters...To make dust a long succession of pro-cesses is necessary: first, supernovaeexplode to deliver heavy elements intospace (metals), then metals cool andform molecules, and lastly moleculesmake macroscopic pieces of matter.Abundant dust is observed in severalearly galaxies, e.g. in HFLS3 at z =6.34 and in A1689-zD1 at z = 7.55.Catalogue of the observed dusty sourcesindicates that their number is an or-der of magnitude larger than predictedby the canonical theory.

11

Hence, prior to or simultaneously withthe QSO formation a rapid star for-mation should take place. These starsshould evolve to a large number of su-pernovae enriching interstellar spaceby metals through their explosionswhich later make molecules and dust.(We all are dust from SN explosions,but probably at much later time.) An-other possibility is a non-standard BBNdue to very high baryonic density, whichallows for formation of heavy elementsbeyond lithium.

12

Observations of high redshift gammaray bursters (GBR) also indicate ahigh abundance of supernova at largeredshifts. The highest redshift of theobserved GBR is 9.4 and there are afew more GBRs with smaller but stillhigh redshifts.The necessary star formation rate forexplanation of these early GBRs is atodds with the canonical star forma-tion theory.

13

II. Mysteries in the sky today and notso long ago.1. SMBH at the present day.Every large galaxy and some smallerones contain a central supermassiveBH whose masses are larger than 109M�in giant elliptical and compact lentic-ular galaxies and ∼ 106M� in spiralgalaxies like Milky Way. The origin ofthese superheavy BHs is not under-stood. These back holes are assumedto be created by matter accretion toa central seed. However, SMHBs areobserved in several small galaxies andeven in almost empty space, where isno material to make a SMBH.

14

The mass of BH is typically 0.1% ofthe mass of the stellar bulge of galaxybut some galaxies may have huge BH:e.g. NGC 1277 has the central BHof 1.7× 1010M�, or 60% of its bulgemass. This fact creates serious prob-lems for the standard scenario of for-mation of central supermassive BHsby accretion of matter in the centralpart of a galaxy. An inverted picturelooks more plausible, when first a su-permassive black hole was formed andattracted matter serving as seed forsubsequent galaxy formation.

15

More examples: F. Khan, K. Holley-Bockelmann, P. Berczik arXiv:1405.6425.Although supermassive black holescorrelate well with their host galaxies,there is an emerging view that out-liers exist. Henize 2-10, NGC 4889,and NGC1277 are examples of SMBHsat least an order of magnitude moremassive than their host galaxy sug-gests. The dynamical effects of suchultramassive central black holes is un-clear.

16

A recent discovery of an ultra-compactdwarf galaxy older than 10 Gyr, en-riched with metals, and probably witha massive black in its center seems tobe at odds with the standard modelJ. Strader, et al Astrophys. J. Lett.775, L6 (2013), arXiv:1307.7707.The dynamical mass is 2×108M� andR ∼ 24 pc - very high density.Chandra: variable central X-ray sourcewith LX ∼ 1038 erg/s, which maybe an AGN associated with a massiveblack hole or a low-mass X-ray binary.

17

”An evolutionary missing link? A modest-mass early-type galaxy hosting an over-sized nuclear black hole”, J. Th. vanLoon, A.E. Sansom, Xiv:1508.00698v1[astro-ph.GA] 4 Aug 2015.BH mass, MBH = (3.5±0.8)·108M�,

host galaxyMstars = 2.5+2.5−1.2·1010M�,

and accretion luminosity:LAGN = (5.3±0.4)·1045erg/s≈ 1012L�.The AGN is more prominent than ex-pected for a host galaxy of this mod-est size. The data are in tension withthe accepted picture in which this galaxywould recently have transformed froma star-forming disc galaxy into an early-type, passively evolving galaxy.

18

Very fresh publication: ”A Nearly NakedSupermassive Black Hole”, J. J. Con-don, J. Darling, Y. Y. Kovalev, L.Petrov, arXiv:1606.04067. A compactsymmetric radio source B3 1715+425is too bright (brightness temperature∼ 3× 1010 K at observing frequency7.6 GHz) and too luminous (1.4 GHzluminosity ∼ 1025 W/Hz) to be pow-ered by anything but a SMBH, but itshost galaxy is much smaller.

19

Globular clusters and massive BHs.Very recent news: BH withM ≈ 2000M� in the core of theglobular cluster 47 Tucanae.Origin unknown. Our prediction( AD,K. Postnov): if the parameters of themass distribution are chosen to fit theLIGO data and the density of SMBH,then the number of PBH with masses(2−3)×103M� is about 104−105 perone SMPBH with mass > 104M�.This density of IMBHs is sufficient toseed the formation of globular clus-ters in galaxies.It is assumed that all PBH withM > 104M� strongly accreted mat-ter and grew up to billion solar masses.

20

MACHOs. A similar or maybe evenconnected problem is related to thenature of MACHOs discovered throughgravitational microlensing by Machoand Eros groups. They are invisible(very weakly luminous or even non-luminous) objects with masses abouta half of the solar mass in the Galactichalo and in the center of the Galaxyand recently in the Andromeda (M31)galaxy. Their density is significantlygreater than the density expected fromthe known low luminosity stars.

21

MACHO group has announced 13 -17 microlensing events in the LargeMagellanic Cloud (LMC), a much largernumber than expected if MACHOswould be normal weakly shining stars.The fractional contribution of the den-sity of these compact ”lenses” withrespect to the dark matter density)was estimated to be in the range0.08 < f < 0.50 (95% CL)for 0.15M� <M < 0.9M�.

22

EROS (Experience pour la Recherched’Objets Sombres) collaboration hasplaced only an upper limit on the halofraction, f < 0.2 (95% CL) for theobjects in the specified aboveMACHO mass interval, while EROS2gives f < 0.1 in the mass range10−6M� <M < 1M�.

23

AGAPE collaboration, working on mi-crolensing in M31 (Andromeda) galaxy,finds the halo Macho fraction in therange 0.2 < f < 0.9, while MEGAgroup marginally conflicts with themwith an upper limit f < 0.3. Newerresults for EROS-2 and OGLE (Opti-cal Gravitational Lensing Experiment)in the direction of the Small Magel-lanic Cloud are: f < 0.1 at 95% confi-dence level for Machos with the mass10−2M� and f < 0.2 for Machos withthe mass ∼ 0.5M�.

24

Summary of limits on MACHOsMacho group: 0.08 < f < 0.50 (95%CL) for 0.15M� <M < 0.9M�;EROS: f < 0.2, 0.15M� <M < 0.9M�;EROS2:f < 0.1, 10−6M� <M <M�;AGAPE: 0.2 < f < 0.9,for 0.15M� <M < 0.9M�;EROS-2 and OGLE: f < 0.1 for M ∼10−2M� and f < 0.2 for ∼ 0.5M�.

25

Thus MACHOs for sure exist.Their density is comparable to the den-sity of the halo dark matter but theirnature is unknown.They could be brown dwarfs, deadstars, or primordial black holes.The first two options are in conflictwith the accepted theory of stellar evo-lution, if MACHOs were created inthe conventional way.

26

More mysteries:It was found that the BH masses areconcentrated in the narrow range(7.8± 1.2)M� (1006.2834)This result agrees with another paperwhere a peak around 8M�, a paucityof sources with masses below 5M�,and a sharp drop-off above 10M� areobserved, arXiv:1205.1805. These fea-tures are not explained in the stan-dard model of BH formation by stel-lar collapse.

27

3. Old stars in the Milky Way:

Employing thorium and uranium incomparison with each other and withseveral stable elements the age of metal-poor, halo star BD+17o 3248 was es-timated as 13.8± 4 Gyr.J.J. Cowan, C. Sneden, S. Burles, et alAp.J. 572 (2002) 861, astro-ph/0202429.

The age of inner halo of the Galaxy11.4± 0.7 Gyr, J. Kalirai, ”The Ageof the Milky Way Inner Halo” Nature486 (2012) 90, arXiv:1205.6802.

28

The age of a star in the galactic halo,HE 1523-0901, was estimated to beabout 13.2 Gyr. First time many dif-ferent chronometers, such as the U/Th,U/Ir, Th/Eu and Th/Os ratios to mea-sure the star age have been employed.

”Discovery of HE 1523-0901: A Stronglyr-Process Enhanced Metal-Poor Starwith Detected Uranium”, A. Frebe,N. Christlieb, J.E. Norris, C. ThomAstrophys.J. 660 (2007) L117; astro-ph/0703414.

29

Metal deficient high velocity subgiantin the solar neighborhood HD 140283has the age 14.46± 0.31 Gyr.H. E. Bond, E. P. Nelan, D. A. Van-denBerg, G. H. Schaefer, D. Harmer,Astrophys. J. Lett. 765, L12 (2013),arXiv:1302.3180.The central value exceeds the universeage by two standard deviations,if H = 67.3 and tU = 13.8;if H = 74, then tU = 12.5.

30

X. Dumusque, et al ”The Kepler-10Planetary System Revisited byHARPS-N: A Hot Rocky World anda Solid Neptune-Mass Planet”.arXiv:1405.7881; Ap J., 789, 154, (2014).

Very old planet, 10.6+1.5−1.3 Gyr.

(Age of the Earth: 4.54 Gyr.)A SN explosion must must precedeformation of this planet.

31

GW discovery by LIGO has proventhar GR works perfectly, existence ofBHs and GWs is established, but ”inmuch wisdom is much grief”, mostlycreated by GW150914.There are essentially three problemsin the standard theory:1. Origin of heavy BHs (∼ 30M�).2. Low spins of the coalescing BHs.3. Formation of BH binaries from theoriginal stellar binaries.

32

The first problem is a heavy BH ori-gin. Such BHs are believed to be cre-ated by massive star collapse, thougha convincing theory is still lacking.To form so heavy BHs, the progen-itors should have M > 100M� anda low metal abundance to avoid toomuch mass loss during the evolution.Such heavy stars might be present inyoung star-forming galaxies but theyare not yet observed in sufficiently highnumber.

33

Another problem is the low value ofthe BH spins in GW150914. It stronglyconstrains astrophysical BH formationfrom close binary systems. However,the dynamical formation of double mas-sive low-spin BHs in dense stellar clus-ters is not excluded. The second reli-able LIGO detection, GW151226,turned out to be closer to the stan-dard binary BH system.

34

Last but not the least, formation ofBH binaries. Stellar binaries wereformed from common interstellar gasclouds and are quite frequent in galax-ies. If BH is created through stel-lar collapse, a small non-sphericity re-sults in a huge velocity of the BH andthe binary is destroyed. BH forma-tion from PopIII stars and subsequentformation of BH binaries with(36 + 29)M� is analyzed in the liter-ature and is found to be negligible.

35

All these problems are solved if theobserved sources of GWs are the bi-naries of primordial black holes (PBH).Here a model of PBH formation ispresented which naturally reproducesthe puzzling properties of GW150914,the rate of binary BH merging eventsinferred from the first LIGO sciencerun, and provides seeds for early su-permassive BH formation.In addition, the mechanism explainsan avalanche of mysteries discoveredrecently and may provide all or a largefraction of cosmological DM

36

The model is based on the supersym-metric (Affleck-Dine) scenario for baryo-genesis modified by introduction of ageneral renormalizable coupling to theinflaton field, see below. It was sug-gested in 1993 (AD and J.Silk) anddiscussed in more details in severalour papers applied to an explanationof existence of the observed ”old” ob-jected in the young universe.As a byproduct it may predict abun-dant antimatter objects in the Galaxy.

37

Baryogenesis with SUSY condensate,Affleck and Dine (AD). SUSY pre-dicts existence of scalars with B 6= 0.Such bosons may condense along flatdirections of the quartic potential:

Uλ(χ) = λ|χ|4 (1− cos 4θ) ,

and of the mass term, m2χ2+m∗ 2χ∗ 2:

Um(χ) = m2|χ|2[1− cos (2θ+ 2α)] ,

where χ = |χ| exp (iθ) andm = |m|eα.If α 6= 0, C and CP are broken.In GUT SUSY baryonic number isnaturally non-conserved - non-invarianceof U(χ) w.r.t. phase rotation.

38

Initially (after inflation) χ is away fromorigin and when inflation is over startsto evolve down to equilibrium point,χ = 0, according to Newtonian me-chanics:

χ+ 3Hχ+U ′(χ) = 0.

Baryonic charge of χ:

Bχ = θ|χ|2

is analogous to mechanical angular mo-mentum. χ decays transferred bary-onic charge to that of quarks in B-conserving process. AD baryogenesiscould lead to baryon asymmetry of or-der of unity, much larger than 10−9.

39

If m 6= 0, the angular momentum, B,is generated by a different directionof the quartic and quadratic valleys atlow χ. If CP-odd phase α is small butnon-vanishing, both baryonic and an-tibaryonic regions are possible withdominance of one of them.Matter and antimatter domain mayexist but globally B 6= 0.

40

Affleck-Dine field χ with CW poten-tial coupled to inflaton Φ (AD andSilk; AD, Kawasaki, Kevlishvili):

U = g|χ|2(Φ− Φ1)2 + λ|χ|4 ln (|χ|2

σ2)

+λ1

(χ4 + h.c.

)+ (m2χ2 + h.c.).

Coupling to inflaton is the general renor-malizable one.

41

If the window to flat direction, whenΦ ≈ Φ1 is open only during a shortperiod, cosmologically small but pos-sibly astronomically large bubbles withhigh β could be created, occupying asmall fraction of the universe, whilethe rest of the universe has normalβ ≈ 6 · 10−10, created by small χ.Phase transition of 3/2 order.

42

This could lead lead to an early for-mation of compact stellar-type objectsand naturally to a comparable amountof anti-objects, such that the bulk ofbaryons and (equal) antibaryons arein the form of compact stellar-like ob-jects or PBH, plus the sub-dominantobserved homogeneous baryonic back-ground, the amount of antimatter maybe comparable or even larger than ofKNOWN baryons, but such “compact”(anti)baryonic objects would not con-tradict any existing observations.

43

The distributions of high baryon den-sity bubbles over length and mass havelog-normal form:

dN

dM= CM exp [−γ ln2(M/M0)]

where CM , γ, and M0 are constantparameters. The spectrum is prac-tically model independent, it is basi-cally determined by inflation.

44

NS-CFLHR MACHOEROS

ER II DF

FIRAS

WMAP

PBH

Figure 1: Constraints on PBH fraction in DM, f = ρPBH/ρDM, where the PBH massdistribution is taken as ρPBH(M) = M2dN/dM The existing constraints (extragalacticγ-rays from evaporation (HR), femtolensing of γ-ray bursts (F), neutron-star captureconstraints (NS-C), MACHO, EROS, OGLE microlensing (MACHO, EROS) survival ofstar cluster in Eridanus II (E), dynamical friction on halo objects (DF), and accretioneffects (WMAP, FIRAS)) The PBH distribution is shown for ADBD parameters µ =10−43 Mpc−1, M0 = γ + 0.1 × γ2 − 0.2 × γ3 with γ = 0.75 − 1.1 (red solid lines), andγ = 0.6− 0.9 (blue solid lines).

45

Conclusion1. A natural baryogenesis model leadsto abundant fomation of PBHs andcompact stellar-like objects in the earlyuniverse after the QCD phase transi-tion, t &&& 10−5 sec.2. These objects have log-normal massspectrum.3. Adjusting the spectrum parameteris possible to explain the peculiar fea-tures of the sources of GWs observedby LIGO.

46

4. The considered mechanism solvesthe numerous mysteries of z ∼ 10 uni-verse: abundant population of super-massive black holes, early createdgamma-bursters and supernovae, earlybright galaxies, and evolved chemistryincluding dust.5. There is persuasive data in favor ofthe inverted picture of galaxy forma-tion, when first a supermassive BHseeds are formed and later they ac-crete matter forming galaxies.6. An existence of supermassive blackholes observed in all large and somesmall galaxies and even in almost emptyenvironment is naturally explained.

47

7. ”Older than tU” stars may exist.8. Existence of high density invisible”stars” (machos) is understood.9. Explanation of origin of BHs with2000 M� in the core of globular clus-ter and the observed density of GCsis presented10. A noticeable fraction of dark mat-ter or all of it can be made of PBHs.

48

Concluson to conclusion: Large amountof astronomical data strongly demandabundant cosmological population ofPBH with wide mass spectrum. SuchPBH nicely explain the mysteries ac-cumulated during a few last years.

Testable predictions:A. Rate and masses of new GW events.B. Possible existence of antimatter inour neighborhood, even in the Galaxy.C. PBH with M = 2000−3000M� inthe cores of globular clusters.

49

THE END

50

The talk is based on the papers:S.Blinnikov, AD, N.Poraiko, K.Postnov,2016;AD, Beasts in Lambda-CDM Zoo, 2016;S. Blinnikov, AD, K.Postnov, 2014,Antimatter and antistars in the uni-verse and in the Galaxy;AD, Blinnikov, 2013, Stars and BlackHoles from the very Early Universe;C.Bambi, AD, 2007, Antimatter in theMilky Way;AD, M. Kawasaki, N. Kevlishvili, 2008,Inhomogeneous baryogenesis, cosmicantimatter, and dark matter;AD, J.Silk, 1992, Baryon isocurvaturefluctuations at small scales and bary-onic dark matter.

51