TITANS OF THE EARLY UNIVERSE The origin of the first supermassive - PowerPoint PPT Presentation

TITANS OF THE EARLY UNIVERSE The origin of the first supermassive black holes Tyrone E. Woods Monash Centre for Astrophysics July 18, 2018 With Alex Heger, Ralf Klessen, Lionel Haemmerle, and Daniel J. Whalen Tyrone E. Woods July 18, 2018 1 / 31 TITANS OF THE EARLY UNIVERSE

TITANS OF THE EARLY UNIVERSE The origin of the first supermassive black holes Tyrone E. Woods Institute of Gravitational Wave Astronomy University of Birmingham July 18, 2018 With Alex Heger, Ralf Klessen, Lionel Haemmerle, and Daniel J. Whalen Tyrone E. Woods July 18, 2018 2 / 31 TITANS OF THE EARLY UNIVERSE

No hot and luminous progenitor for most Type Ia supernovae Woods, Ghavamian, Excluded Progenitors of Tycho’s Supernova Badenes, and 38.5 1.4M ⊙ 1.2M ⊙ log 10 Bolometric Luminosity (erg/s) 38 1.0M ⊙ Gilfanov, Nature 0.8M ⊙ 3 0.6M ⊙ 37.5 2 Astronomy , 2017 0.51M ⊙ 37 4 1 36.5 See also, e.g., Woods 36 n = 1 35.5 c m & Gilfanov 2013, - 3 , d = 35 3 p c 2014, 2016 Johansson, 34.5 Permissible Models Woods et al., 2014, 34 10 1 2016 Effective Temperature (10 5 K) Tyrone E. Woods July 18, 2018 3 / 31 TITANS OF THE EARLY UNIVERSE

The Origin of High-redshift Quasars Tyrone E. Woods July 18, 2018 4 / 31 TITANS OF THE EARLY UNIVERSE

The Origin of High-redshift Quasars Truly massive (10 9 –10 10 M ⊙ ) quasars have been observed at redshift ∼ 7 (e.g., Mortlock + 2011, Wu + 2015). This is hard to explain: � M BH � t growth ∼ 0.1log 10 Gyr M seed Especially given pop III black holes “born starving” (Alvarez, Wise, & Abel 2009) Tyrone E. Woods July 18, 2018 5 / 31 TITANS OF THE EARLY UNIVERSE

The Origin of High-redshift Quasars Tyrone E. Woods July 18, 2018 6 / 31 TITANS OF THE EARLY UNIVERSE

The Origin of High-redshift Quasars See review by Mar Mezcua, 2017 Tyrone E. Woods July 18, 2018 7 / 31 TITANS OF THE EARLY UNIVERSE

Supermassive Stars – a little history How massive is supermassive? 10 4 –10 6 M ⊙ Initially hypothesized candidate for quasars Assumed to be formed “all at once” (monolithically) P gas Strongly radiation-dominated ( β = P tot << 1 ) : 4 P ∝ ρ 3 → polytrope, with index n = 3 Local adiabatic index Γ = 1 + 1 n ≈ 4 3 + β 6 Tyrone E. Woods July 18, 2018 8 / 31 TITANS OF THE EARLY UNIVERSE

Supermassive Stars – a little history Objects with Γ ≈ 4 / 3 are “trembling on the verge of instability” (Fowler 1964) Very small perturbation can trigger collapse! Chandrasekhar (1964) and others showed that there is a general relativistic correction to the critical pressure support needed: Γ ≈ 4 / 3 + 1.12 2 GM Rc 2 Criterion for instability: β 6 < 1.12 2 GM Rc 2 β ∝ M − 1 2 for β << 1 → ∼ 10 5 –10 6 M ⊙ stars will quickly collapse! Tyrone E. Woods July 18, 2018 9 / 31 TITANS OF THE EARLY UNIVERSE

KEPLER Stellar Evolution Code implicit Lagrangian hydrodynamics and stellar evolution (Weaver, Zimmerman, and Woosley 1978) solve conservation equations for mass, energy, and momentum in spherical symmetry equation of state allowing for general mixture of radiation, ions, and electrons of arbitrary degeneracy and relativity, as well as pair production include post-Newtonian correction to the acceleration due to gravity (e.g., Fuller + 1986) Tyrone E. Woods July 18, 2018 10 / 31 TITANS OF THE EARLY UNIVERSE

Monolithic Supermassive Stars Tyrone E. Woods July 18, 2018 11 / 31 TITANS OF THE EARLY UNIVERSE

Monolithic Supermassive Stars Tyrone E. Woods July 18, 2018 12 / 31 TITANS OF THE EARLY UNIVERSE

Supermassive Stars 10 Specific Volume [cm 3 g -1 ] 1 0.1 155000 M ⊙ 156000 M ⊙ 0.01 2000 3000 4000 5000 6000 7000 8000 9000 10000 Time [years] Tyrone E. Woods July 18, 2018 13 / 31 TITANS OF THE EARLY UNIVERSE

Supermassive Stars 10 7 10 6 CORE HYDROGEN DIRECT COLLAPSE Lifetime [Years] CORE HELIUM BURNING BURNING 10 5 10 4 10 3 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Mass [10 5 M O . ] Tyrone E. Woods July 18, 2018 14 / 31 TITANS OF THE EARLY UNIVERSE

How do you actually make a supermassive star? Option 1: Regan + , Nature Astro , 2017 Tyrone E. Woods July 18, 2018 15 / 31 TITANS OF THE EARLY UNIVERSE

How do you actually make a supermassive star? Option 2: high baryonic streaming velocities (Tanaka & Li 2014; Schauer et al., 2017; Hirano et al., 2017). Essentially an atomically-cooled halo with an assist? Option 3: really massive infall rates possible in high-z galaxy mergers? (Maier et al 2015) Option 4: Coalesence of a dense stellar cluster? (problems, see Latif et al., 2016) Tyrone E. Woods July 18, 2018 16 / 31 TITANS OF THE EARLY UNIVERSE

Initial Conditions Begin with a 10 M ⊙ , n = 3 polytrope with a central density of 10 − 3 gcm − 3 → a somewhat “puffy” protostar. Primordial composition, including deuterium and lithium. Consider accretion rates in the range 0.01 – 10 M ⊙ / yr, typical of atomically-cooled haloes Tyrone E. Woods July 18, 2018 17 / 31 TITANS OF THE EARLY UNIVERSE

Accreting Supermassive Stars Don’t Know how to Relax! Haemmerle, Woods, et al. (2017) Tyrone E. Woods July 18, 2018 18 / 31 TITANS OF THE EARLY UNIVERSE

The Onset of Nuclear-burning 10 6 10 5 Temperature [K] Density [g/cm -3 ] 10 4 10 9 10 3 10 2 10 8 10 1 10 0 10 7 10 -1 10 -2 10 6 10 -3 CNO Mass Fraction 10 -7 10 -8 10 -9 10 -10 10 0 10 1 10 2 10 3 10 4 10 5 Time [yr] Blue – 1 M ⊙ / yr Red – 10 M ⊙ / yr Tyrone E. Woods July 18, 2018 19 / 31 TITANS OF THE EARLY UNIVERSE

A Representative Case: 1 M ⊙ / yr Tyrone E. Woods July 18, 2018 20 / 31 TITANS OF THE EARLY UNIVERSE

Reminder: criterion for onset of instability Recall the Chandrasekhar general relativistic instability for supermassive stars: 4 P ∝ ρ 3 → polytrope, with index n = 3 Local adiabatic index Γ = 1 + 1 n ≈ 4 3 + β 6 Chandrasekhar (1964) and others showed that there is a general relativistic correction to the critical pressure support needed: Γ ≈ 4 / 3 + 1.12 2 GM Rc 2 Polytropic criterion for instability: β 6 < 1.12 2 GM Rc 2 Tyrone E. Woods July 18, 2018 21 / 31 TITANS OF THE EARLY UNIVERSE

When does collapse set in? 0.01 β /6, 10 5 M ⊙ 0.009 1.12(2GM)/Rc 2 , 10 5 M ⊙ β /6, 3.2 x 10 5 M ⊙ 0.008 1.12(2GM)/Rc 2 , 3.2 x 10 5 M ⊙ 0.007 β /6, 1.12(2GM)/Rc 2 0.006 0.005 0.004 0.003 0.002 0.001 0 10 2 10 3 10 4 10 5 10 6 Mass coordinate [M ⊙ ] Tyrone E. Woods July 18, 2018 22 / 31 TITANS OF THE EARLY UNIVERSE

The Most Massive Stars that Ever Lived!? Collapse after H-exhaustion 3 GR collapse during H-burning 2 final mass (10 5 M ⊙ ) 1.5 Hydrostatic limit for H-burning monolithic stars 1 0.8 0.6 Hydrostatic limit for He-burning monolithic stars 0.5 0.4 0.3 0.01 0.1 1 10 accretion rate (M ⊙ yr -1 ) � � ˙ � � × 10 5 M ⊙ M M SMS,final ≈ 0.83 log 10 + 2.48 M ⊙ yr − 1 Tyrone E. Woods July 18, 2018 23 / 31 TITANS OF THE EARLY UNIVERSE

The Most Massive Stars that Ever Lived!? Tyrone E. Woods July 18, 2018 24 / 31 TITANS OF THE EARLY UNIVERSE

Accreting Supermassive Stars with “realistic” accretion rates Tyrone E. Woods July 18, 2018 25 / 31 TITANS OF THE EARLY UNIVERSE

Accreting Supermassive Stars with “realistic” accretion rates Tyrone E. Woods July 18, 2018 26 / 31 TITANS OF THE EARLY UNIVERSE

Rotating Supermassive Stars ΩΓ -limit: known problem from “normal” Pop I massive star evolution v 2 � R eq = GM GM crit,1 eq → v crit,1 = R 2 R eq v 2 crit,2 = 2 π G ρ ( 1 − Γ Edd ) R 2 eq Tyrone E. Woods July 18, 2018 27 / 31 TITANS OF THE EARLY UNIVERSE

Rotating Supermassive Stars Haemmerle et al., 2017, submitted: Supermassive stars have to be slow rotators (v surf < 10 − 20%v crit,1 ). Supermassive star formation by accretion requires mechanisms efficient enough to remove most ( ≈ 99%) of the angular momentum from the accretion disc. Need to get rid of a lot of angular momentum somehow! Spiral arms in the disk? Magnetic braking? Tyrone E. Woods July 18, 2018 28 / 31 TITANS OF THE EARLY UNIVERSE

Tyrone E. Woods July 18, 2018 29 / 31 TITANS OF THE EARLY UNIVERSE

Conclusions Supermassive protostars accreting � 0.1 M ⊙ / yr collapse due to the GR while H-burning Final fate of (non-rotating) supermassive stars depends in a reliable way on accretion rate (variable rates qualitatively similar) Rotation rates of supermassive stars strongly constrained Even for non-rotating case, n = 3 polytrope poorly predicts moment of collapse... including when applied only to the core. www.tewoods-astro.com Tyrone E. Woods July 18, 2018 30 / 31 TITANS OF THE EARLY UNIVERSE

KEPLER Stellar Evolution Code d t = 4 π r 2 ∂ P − G rel m r + 4 π ∂ Q dv ∂ m r r 2 r ∂ m r d u ∂ ∂ � v − ∂ L � ( v r 2 ) + 4 π Q d t = − 4 π P + ε ∂ m r ∂ m r r ∂ m r rc 2 + 4 π P r 3 � � 1 + P ρ c 2 + 2 GM G rel = G m r c 2 Tyrone E. Woods July 18, 2018 31 / 31 TITANS OF THE EARLY UNIVERSE