Concepcion, 2-6 March 2020
- (Hirano & Bromm 2017) During fragmentation, there are three outcomes
- Mergers, causing a burst of accretion
- Survival, creating a binary/multiple system
- Ejection, resulting in a low-mass metal-free star
- (Machida+ 2013) Magnetic fields cann reduce fragmentation, causing less stars with higher masses. However there hasn't been numerical convergence with increasing cells per Jeans length (Sur+ 2016).
- (e.g. Inayoshi+ 2020; Regan & Downes 2019) External radiation delays Pop III star formation to higher mass halos, which may cause very massive star formation or a more massive cluster of Pop III stars
- No big changes in chemistry and rates since First Stars V (2016)
- But the two big advances in the past four years
- Direct collapse black holes (DCBHs) in more massive halos (see Glover 2015 for chemistry)
- Magnetized collapses in non-ideal MHD (see Nakauchi+ 2019 for chemistry). Lithium is important at densities
$> 10^{10}$ cm$^{-3}$ because it becomes the dominant charge carrier
- (Glover & Abel 2008; Lique 2015) There are some uncertainties in the low-temperature H2 cooling rate
- (Coppola+ 2019) However those fits assume a 3:1 ortho/para ratio, this newer work includes an equilibrium ratio, which gives much more cooling at low temperatures (T < 1000 K)
- (Greif 2014) High-density opacity is still a problem because it is very computationally expensive.
- Small-scale feedback uncertainties
- Assumes equilibrium chemistry:
$T_{\rm rad} = T_{\rm gas}$ - Treatment of radiative feedback from the accretion luminosity needs improvement. How? See Jaura's talk later in the session
- Assumes equilibrium chemistry:
- Using AREPO to simulate (1 Mpc/h)$^3$ with sink formation at
$10^5$ cm$^{-3}$, forming several hundred of them. - Considering both streaming velocities and Lyman-Werner radiation feedback
- (Schauer+ 2019) Streaming velocities increase the minimum and typical halo mass that can cool and collapse, but there's no dependence on redshift
- (in prep) It's uncertain how the LW radiation and streaming velocities interplay; however the streaming velocities should decrease the SF and thus the LW background
- (Sugimura+ 2020) Using SFUMATO (Japanese AMR code; Matsumoto+ 2007), they studied the collapse, fragmentation, and formation of Pop III stars. Considering primordial chemistry and radiative transfer
- Formed a wide multiple system with a single massive (~60 Msun) star and a triple system, which are separated by ~0.5 pc.
- The binary separation is too large (in this one case) to result in a BBH merger within a Hubble time, i.e. a LIGO event.
- (Chon+ 2019) Magnetic fields may change this result
- How is typical high-mass Pop III star formation connected to rare high-mass Pop I star formation? The outcomes are the same, but the physics and chemistry during their formation may differ. This is the focus of their future studies (in prep).
- (e.g. Turk+ 2012; Liao+ 2019) Primordial magnetic fields are amplified through the Biermann battery, flux freezing, and dynamo before star formation
- (Sharda+ 2020) Fragmentation always occurs, regardless of the B-field strength. But the B-field favors less fragmentation and more massive stars
- Due to the late onset of radiation feedback and the absence of dust, primordial B-fields potentially impact the IMF more than the present-day case.
- (in prep) AREPO simulations of an isolated minihalo (BE mass of 2671 M$_\odot$) with primordial chemistry and sink particle formation at
$7 \times 10^{13}$ cm$^{-3}$ and radiation feedback (SimpleX method; Paardekooper+) - Evolve for ~20 kyr after sink particle formation
- Radiation feedback suppresses SF by about a factor of two (~500 to ~250 M$_\odot$) in total mass
- (Susa 2019) With a sink threshold of
$10^{15}$ cm$^{-3}$, there's little variation in the number of fragments between five different realizations - All of the work on the number of Pop III stars per halo is on the same trend:
$N \approx 3 (t / 1 yr)^{0.3}$ until radiative feedback suppresses star formation - Main conclusion: Pop III stars form in clusters with small numbers (~10).
- On BBHs: they expect a few binaries per minihalo that corresponds to ~10 mergers/yr/Gpc$^3$ (in prep)
- (Smith+ 2015) Pop2Prime simulations and their extended runs to z = 11.8 when the 2nd metal-enriched star forms (in prep).
- Fragmentation is suppressed if dust isn't considered in the cooling/chemistry model
- What metallicity is required for a collapse? Consider a system with the cooling time less than the dynamical time for gas within 0.1
$r_{\rm vir}$ .- Consider two cases: (1) existing species fractions from the simulation, and (2) total dissociation
- Finds that 0.01
$Z_\odot$ is required to induce cooling and collapse
- Dilution of SN ejecta with primordial gas is important to consider during the formation of metal-poor stars, however it is poorly understood.
- (in prep) Bayesian framework to fit abundance patterns, while considering mixing and dilution, in CEMP-no stars. Using Starfit (Heger & Woosley 2010) models.
- In a system of 200 EMP stars, they find that most of the stars favor massive stars between 10-25 (peaking at 15)
$M_\odot$ . Their constraints on the dilution mass pushes the progenitor masses down by a factor of 2. - In a simulation (1 Mpc/h with RT and streaming velocities; in prep) with all PISNe (200
$M_\odot$ ) they find a dilution mass of$3 \times 10^{6-8} M_\odot$ . So far, there are four metal-enriched stars with [Fe/H] between -1.7 and -4.0.
- In present-day SF, MHD and radiation pressure are the biggest sources of feedback, limiting the final mass, whereas photo-evaporation is a weak effect.
- As the metallicity decrases, the optical depth decreases, causing radiation pressure to be less important. Also photo-evaporation is weaker because it's more optically thick (dust absorption)
- Feedback doesn't set an upper mass limit. The core mass is more important.
- (Sana+ 2012) In the MW, ~75% of massive stars have close companions.
- (Tanaka & Omukai 2014) Protostellar disks are Toomre-Q unstable above [Z/H] = -5.
- SFE depends on the surface densites of the ICs and they decrase in low-Z environments.
- In future work, they'll include B-fields and ICs extracted from cosmological sims.
- Rotation will enhance mixing throughout the star because of sheer instabilities.
- More nitrogen migrates from the core to the envelope.
- When considering B-fields, this results in dynamo effects magnifying the B-field and allowing the star to be radiative and rotate like a solid-body.
- With dynamos, the star becomes hotter (more ionizing photons during reionization?) and experiences more mixing. Has much less nitrogen than the rotating, non-magnetic models.
- But what's the best model? How do will constrain this? Through astro-seismology (e.g. Eggenberger+ 2019).
- Future work needs to treat angular momentum transport within the star
- (Kobayashi+ 2019) For Type Ia SN, what is the ratio of single and double WD systems? WD-WD mergers can be sub-Chandrasekhar that doesn't produce enough Mn
- Because chemical enrichment is inhomogeneous, slow processes like r-process enhancement can happen later at low-metallicities.
- Not possible to reproduce the observed scatter in [Eu/Fe] at [Fe/H] = -2 to 0 from binary population synthesis. Lifetimes of NSMs are uncertain, however all they should have lifetimes with <0.1 Gyr.
- (Yong+ 2020) r-II star from SkyMapper may be modeled with a magneto-rotational SN from a 50
$M_\odot$ star
(Choplin) Rotation, explosion and nucleosynthesis in early massive stars and the abundances of metal-poor stars
- Because of the different types of EMP stars, the enrichment sources in the early Universe are diverse
- Jet-like explosions from fast rotators will result in special nucleosynthesis, long GRBs, and high entropy environments. Rotation boosts the s-process and heavy element abundances (> Sr)
- (in prep) Using a 2D relativistic hydro code, they find that the jet passes through the core reactivates a lot of the neutron capture reactions, populating the i-process region. No r-process.
- Jet-like explosion may be important sources of heavy elements in the Universe
- (Berg+ 2020) Using (sub)DLAs to probe chemical enrichment of the early universe, looking at overall metallicity and abundance ratios. They also compare to Local Group abundances.
- DLAs are similar to dSph in [Zn/Fe] and low alpha-elements and trace CGM environments
- (Cooke+ 2015, 2017; Berg+ 2016) [Z/H] < -2 DLAs have abundance ratios most similar to faint SNe or CCSNe.
- Only 32 metal-poor DLAs observed at high-res and sub-DLAs are 4-6 times less frequently observed.
- (Tanvir+ 2019) GRBs can be used to probe the UV escape fraction at high-z because they can measure the HI column densities. This reveals very low
$f_{\rm esc} < 0.005$ up to z=6. - (Chomock+ 2013) Intrinsic ISM lines can probe both the metallicity and HI column densities of the host galaxies, which can be compiled to help determine a redshift-dependent relation
- (Palmerio+ 2019) Long GRBs don't probe galaxies as one'd expect if it were to follow SF. This could be caused by a bias toward long GRBs happening more often in low-Z stars.
(Kirihara) Discrimination of heavy elements originating from Pop. III stars in z = 3 intergalactic medium
- (Ishiyama+ 2016) 8 Mpc/h N-body simulation with 2048$^3$, resolving halos down to
$2.4 \times 10^5 M_\odot$ . Using a SF and chemical enrichment model, exploring two extreme scenarios: all CCSNe and all PISNe. - They found that Pop III SNe can pre-enriched the vast majority of the IGM, whereas the same space in large-scale galaxy simulations (i.e. Illustris) only has enrichment near to the galaxies and not the voids.
- Pop III enrichment is dominant in regions with overdensities less than ~30.
- The median metallicity is between [Z/H] = -4.5 to -2, depending on redshift(?)
(Moriya) Searches for Population III pair-instability supernovae with upcoming near-infrared transient surveys
- In the 2020's there will be many NIR imagers: JWST, WFIRST (FoV ~ 0.3 deg$^2$, 2025 launch), ULTIMATE-Subaru (FoV ~ 0.05 deg$^2$, 2025 planned)
- If the Pop III IMF is top-heavy, there should be PISNe. How to search for them? They produce up to 50
$M_\odot$ of$^{56}Ni$ , whereas hypernovae produce up to 1$M_\odot$ . - (Moriya+ 2019) Using the light curves from Kasen+ (2011), they made predictions for WFIRST at z=6. >225
$M_\odot$ progenitors are detectable ~4 years after explsion (in the observer frame). Suggesting that we need one observation every 6 months for 5 years to search for events - (de Souza+ 2014) Provides a compilation of Pop III SFRs. Moriya+ (2019) uses them to make PISNe event rates, assuming a flat IMF. WFIRST will detect 0.6 per deg$^2$ per unit redshift out to z=6 (26.5 mag limit).
- With the FoV of the observatories, Subaru needs 100 nights over 5 years and WFIRST needs 25 exposures.
- These require spec follow-ups, where many other transients will be discovered (high-z superluminous SNe, AGNe, low-z NIR-bright transients). Therefore, we need to determine a good candidate selection with optical observations and host galaxy information.
- (Kinugawa+ 2014, 2016) Simulate a million Pop III binary evolution with a flat IMF over a Hubble time. Predict a merger rate of ~30 per year per Gpc$^3$, which is consistent with the LIGO events (10-100 per year per Gpc$^3$).
- They find that most mergers from Pop III stars come from systems with a total mass of 60
$M_\odot$ . For metal-enriched stars, the typical mass is ~10$M_\odot$ . - (Belczynski+ 2010) Solar metallicity stars lose much of their mass through winds and don't produce more massive remnants. Low-Z (< 0.1 solar) can retain more mass, though.
- If a Pop III star has
$< 50 M_\odot$ , it remains a blue giant and there's little mass transfer. Above that mass, the star becomes a red giant and can undergo mass transfer. - (Visbal+ 2015; Inayoshi+ 2016?) Only a ~1/3 reduction in Pop III SFR can result in a merger rate too low to be consistent with LIGO observations. But this isn't a problem because the LIGO constraints include BBHs from low-Z stars.
- Belczynski+ (2017) predicted that all Pop III BBHs merge in the early Universe. However, they used a modified low-Z model, whereas they treat all Pop III stars in the giant phase as red giants. Inconsistent with current Pop III models.
- (in prep) Fast merger times (< 1 Gyr) happen in BBHs with at least one with a spin >0.9. If both BHs have a<0.1 then the merger times are likely to be longer than the Hubble time.
- The mass distribution or z-dependence might distinguish between Pop I/II and III. DECIGO may detect BBH mergers at high-z.
- From the waveforms, there are many degeneracies between redshift, mass, mass ratio, spin, etc.
- (Sesana+ 2016) eLISA would first detect a BBH about a year before merger, and then predict the merger time/location that can then be observed with aLIGO afterwards.
- (Valiante+ in prep) Einstein telescope and LISA could distinguish between light and heavy BH seeds
- (Hartwig+ 2018) Predictions for mergers from massive BH seed formation, whereas the free parameters are the critical LW intensity and binarity fraction
- The problem with LISA is the data analysis because there'll be 1000s of sources. 3k Galactic binaries, 1-100 extragalactic massive binaries. The challenge will be to disentangle all of the signals.
- (Liu & Bromm 2020) GIZMO simulations: one zoom-in (
$10^{10} M_\odot$ at z=10) and a survey simulation of 4 Mpc/h- Primordial and metal-enriched star formation / feedback and BH seeding, growth, and feedback
- BH capture and inspiral model results in timescales between 1-10 Gyr
- GW event rates are sensitive to the inner slope of the DM density, increasing exponentially between ~0.7 and 1.5, because of binary hardening
- Local Pop III BBH merger rates are between 0.01 and 0.1 per year per Gpc$^{-3}$, comparable to the in-situ channel
- (Iwamoto+ 2005; Tominaga+ 2007) The amount of fallback is one of the most important mechanisms in determining the yields because the explosion mechanism is not well understood, so that fallback cannot be calculated from first principles.
- (Ishigaki+ 2018) Analyzing ~200 EMP stars to fit their abundance patterns with Pop III yield expectations (Umeda & Nomoto 2005; Tominaga+ 2007). Parameters: mass, explosion energy, mixing mass, ejecta fraction by mass, and hydrogen dilution mass.
- If the explosions are low-energy, supernova, or hypernova, the most likely progenitor masses are 13, 15, and 25
$M_\odot$ respectively. The CEMP stars are either best explained by 13 or 15$M_\odot$ models. - Most of EMP stars are best fit with $^{56}$Ni ejecta masses around 0.01-0.1
$M_\odot$ , where the CEMP stars are enriched by events producing <0.01$M_\odot$ . - Shortcomings: Cannot following detailed fallback and mixing; assumption of one SN per EMP star; metal dilution with pristine gas; connection to Pop III IMF is uncertain because of the unknown fraction of Pop III that don't produce metals
(Frebel) Characterizing the origin and properties of the halo r-process star population with data collected by the R-Process Alliance
- Chemical tagging with r-process stars is a promising avenue for tracing their birth halos.
- (Brauer+ 2018) Over 50% of r-II halo stars originate in small systems like Ret II. This was determined from a post-process analysis of the Caterpillar Simulations.
- R-process Alliance: 1500 snapshot spectra. Found ~30 new r-II stars, ~200 new r-I stars. They select stars based onthe [Ba/Eu] ratio (e.g. Cain+ 2020; Ezzeddine+ 2020).
- If NSM only makes low Sr, all excess needs to coem from SNe. If NSM could make more than what Re II NSM did, it's more complicated.
- Asteroseismology of MP stars. Precise abundances that are consistent with log(g).
- (Miglio+ 2016; Valentini+ 2019) Measured reliable ages and masses of red giants in GC M4 with 30% uncertainties.
- They will be targeting 68 new MP stars down to [Fe/H] = -3.5
- (Howes+ 2016) There are no CEMP stars between [Fe/H] = -2 and -3, and 30% of the [Fe/H] < -3 are CEMP. This is probably due to a selection effect.
- (Arentsen+ 2020; in prep) Metal-poor ([Fe/H] < -1) stars from the PRISTINE survey. Spec follow-up of ~8k stars and then find best fit.
- Discovered ~1400 new MP stars in the bulge and ~10 EMP stars. There are some CEMP stars in the sample.
- Cannot take the CEMP percentages at face value because there are selection effects, as very carbon-rich drops out of the selection region because the carbon spectral features decreases the flux in the filter. This only affects the extremely carbon-rich [C/Fe] > +2 stars.
- Molecular bands in 3D LTE: affects abundances in carbon and oxygen
- (Nordlander+ 2017) Beware of 3D NLTE effects on atoms
- (Nordlander+ 2019) UVES data for SMSS1605: CNO decrease, [Ni/Fe] ~ 0 and [C/O] ~ +0.5.
- [alpha/Fe] ~ 0.4 and no odd-Z detection
- Abudnances indicate enrichment from ~10
$M_\odot$ Pop III star
- (Reggaiani+ 2020; in prep) Using WISE and Spitzer wide-band filters to identify metal-poor stars in the LMC because of the CO molecular bands between 2-4(?) microns
- Then followed up in that sample to search for r-process enhanced stars. The [Ba/Fe] and [Eu/Fe] ratios are similar to MW halo r-stars.
- There are many more (by fraction) r-stars present in the LMC than the halo.
- This probably happened because the LMC has an alpha/Fe "knee" at lower iron metallicity, indicating that it was accreting pristine/metal-poor gas longer than the MW
- (Reggiani+ 2020; in prep) Using the same technique described in Schlaufman's talk to identify three MP stars in the inner bulge
- There is one outlier in [Mn/Fe] with a very high value
- The ratios suggest that the progenitors are CCSNe with "normal" masses and explosion energies. In the bulge progenitors, the SFR is high and the ISM is less homogeneous than the Galactic halo.
- (e.g. Minniti+ 2017; Palma 2019) 55 GCs in the bulge within 3.5 kpc of the center
- (Cescutti+ 2017) Metallicity distribution functions between the halo and bulge GCs, where the latter are more metal-rich. However the metal-rich bulge GCs could be older because of the faster chemical evolution
(del Mar Matas) Spectroscopic follow-up of metal poor candidates from the Pristine survey with Narval at TBL
- In the PRISTINE survey, they identified 15 interesting MP stars, and the most metal poor star is located in the thin disk.
- (Gonzalez-Hernandez+ 2020) The abundance pattern indicates that was enriched by a single Pop III star 21-27
$M_\odot$ with a low explosion energy and very little mixing and a lot of fallback
- (Mura+ in prep) It is very difficult to detect fluorine but it is a good indicator of its progenitors -- CCSNe, WR stars, and AGB stars
- In a CEMP-s/r star (HE 1305+0007), they find A(F) = +3.98, [F/Fe] = +???, [Fe/H] = -???.
- In a CEMP-s star (HE 1420-0551), they find A(F) = +3.93, [F/Fe] = +1.95, [Fe/H] = -2.53.
- This is inconsistent with theoretical models, which could suggest a different nucleo-channel (i-process?)
- The major of the 72 stars studied are on retrograde orbits with 6 having a likely origin from a dSph merger (outer halo orbit + Fe-poor + alpha-poor)
- Otherwise the sample is chemical homogeneous with standard halo abundances
- If they are sampling truly from the outer halo, then it looks very similar to the inner halo. OR is this probing the high-eccentricity tail of the Gaia Sausage?
- The mean Li abundance seems to decrease significantly only below [Fe/H] = -3.5, but there are only 3 stars there.
- The decrease is more milder in LRGB stars with respect to dwarfs.
(Kalari) Near-field cosmology with metal-poor stars: Births and deaths of stars in the Magellanic Clouds
- (Kalari+ 2018; Schneider+ 2019) The IMF deviates from Salpeter at lower metallicities (NGC 796 at 1/5 solar) and 30 Dor (1/2 solar).
- Observations of YSOs and pre-main MS stars are essential to observe to determine the IMF
- XRBs dominate over AGN at z>6. Possible high-z X-ray heating sources: XRBs, thermal emission from galaxies, BHs, DM annihilation, CRs, B-fields
- (in prep) Using a binary evo code (Binary_c) to estimate the number and properties of high-z XRBs, especially in metal-poor stars. Modeling host absorption with X-ray radiative transfer. Goal: source model population
- (Mirocha+ 2017; Cohen+ 2017) 21-cm signals, both the power spectrum and the global signal, are very sensitive to star formation/feedback modeling
- (Cohen+ 2019; Monsalve+ 2019) Trained an ANN with 30k models to emulate global signals
- Requires SF in minihalos, excludes inefficient heating
$f_X < 0.0042$ , favors soft X-rays$E_{\rm min} > 2.3$ keV, exclude high optical depth$\tau > 0.08$
- Requires SF in minihalos, excludes inefficient heating
- (Ota+ 2018) Observations suggest that QSOs don't live in biased environments. However, there could be some SF suppression from the QSO feedback
- (in prep) Zoom-in simulations (1 Mpc$^3$ within a ~60(?) Mpc$^3$ box) of QSO hosts, studying the suppression of nearby galaxy formation. They plant a radiation source into the QSO host.
- They find that hosts below
$10^9 M_\odot$ is significantly suppressed by a factor of 2-5 from photo-dissociation and ionization. The galaxy LF is slightly suppressed by ~2 above$M_{1500} = -14$ .
- (Chiou+ 2019) They find that streaming velocities can induce GC formation
- AREPO (2 Mpc)$^3$ simulation. They search for gas clouds that are displaced from their DM halos with a density higher than a critical value and find many
- These Streaming Induced Gas Objects (SIGOs) have similar characteristics as present-day globular clusters when passively evolved to
$z=0$
- Using Pop II and III SFRs from semi-analytic models and simulations (Hartwig+ 2018), they have Pop II SFR dominates at z < 12
- Binary_C model: Considers radiative feedback, binary fraction (Stacy+ 2012), eccentricities and periods (Oh+ 2015), and MS evolutionary models.
- At z=10, they find that 99% of minihalos have XRBs
- There are more energy deposited by LMXBs becuase they live longer
- Can probe the ISM/CGM through MgII, SiII, CII, OI, FeII, AlII (T ~ 10$^4$ K) and CIV, SiIV, NV (T ~ 10$^5$ K)
- (Bosman+ 2017) Searching for very high (z~7) metal absorption systems.
- (Becker+ 2019) White paper about searching for z>7 systems and the possibility of Pop III signatures
- (Suarez & Meiksin in prep) Using CLOUDY with circular absorption models with external radiation sources
- Studying CGM clouds that are
$10^6 M_\odot$ with 10-100 pc radii that are captured by high-res simulations
- Massive galaxies cannot provide all of the ionizing photons for reionization. Three main hypotheses: (1) more faint galaxies, (2) z>6 galaxies are producing more UV photons than expected, (3) the first galaxies could form earlier than expected
- They determine the ionizing luminosities w.r.t. redshift by calculating the SF history through SED fitting of many galaxies.
- (Laporte+ 2017; Hashimoto+ 2018) Detected [OIII] 88 micron line at z = 8-10
- (Katz+ 2019) In a (70 Mpc)^3 simulation, they found 3 objects similar to these sub-mm galaxies, suggesting that they're not rare
- Need ALMA to tightly constrain the SFRs and ages through emission line strengths
- The ages can also be constrained through dust emission.
- (Meyer+ submitted) Using the Ly-alpha emission line profile to estimate the UV escape fraction and finds that it could be >80%
- Intro
- Ultra-faint dwarfs (UFDs) are low luminosity (
$< 10^5 M_\odot$ ), M/L > 100, metal-poor ([Fe/H] = -4 -> -2), and old (12-13 Gyr) and reionization fossils - Expected properties from a galaxy inefficiently forming stars. The EMP stars in UFDs tend to be CEMPs. [Sr,Ba/Fe] are mostly unusually low, and rarely they are highly r-process enhanced. Sometimes s-process enrichment at high [Fe/H] by AGB stars
- New since FSV: discovery of r-process UFDs, 5 new UFDs discovered, doubled the number of UFD stars with hi-res spec, GAIA kinematics
- Ultra-faint dwarfs (UFDs) are low luminosity (
- Carina II. Associated with LMC.
- (Ji+ 2020) [alpha/Fe] declines but different for Mg and Ca. [Mg/Ca] declines by a factor of 6 with [Fe/H] or the history of the galaxy
- This can be explained by different enrichment sources. >20 Msun CCSNe, IMF-integrated yields, <15 Msun CCSNe, and/or Type Ia SNe. All of this points toward multiple / variable stellar populations in the UFD progenitor
- The UFDs with [Mg/Ca] vs. [Fe/H] negative slopes are LMC/SMC members
- Inhomogeneous chemical mixing in Ret II
- (Ji+ in prep) How to break degeneracies in [X/H] ratios? We can use Ret II in its r-process elements and possible NS merger enrichment source to break some degeneracies.
- [r/H] distribution traces the metal mixing, i.e. [Ba/H].
- [X/H] ~ 0.2 spread can be attributed to inhomogeneous mixing, but simulations sometimes show more (Emerick+ 2020)
- In Car II (s-process dominated), the [Ba/H] ~ 0.55 spread is larger because AGB winds are much less energetic and causes greater deviations within a protogalaxy
(Chiti) Detection of a spatially extended population of extremely metal-poor stars in the Tucana II ultra-faint dwarf galaxy
Embargoed.
- (Tarumi+ 2020) Can we determine the host galaxy and its progenitor properties through the [Eu/Fe] ratios from r-process enhancement?
- To test this, they run simulations of the formation of a UFD galaxy with controlled explosions in different sites: off-center (by a virial radius) and center to the halo
- In the central explosion, the UFD is highly enriched to [Eu/Fe] = +2 (similar to Ret II). In the off-center one, it's moderately enriched [Eu/Fe] = +1 (similar to Tuc III)
- The [Eu/H] ratio is affected by
- Numerator: travel distance between explosion and star formation
- Denominator: gas mass, similar among UFDs
- NS mergers allow for a large scatter in [r/H] within UFDs
- Small/large scatter in [r/H] is caused by short/long SF history
- (in prep) Using random IMF sampling of Pop III stars in the GAMETE (Salvadori+) semi-analytical model
- Using UFDs for observational constraints, the random sampling results in an increase by 3x in Pop III stars
- In Bootes I, they exclude tha tPop III stars have <0.8
$M_\odot$ ($m_{\rm char} = 0.35 M_\odot$ ) for a standard IMF - For Bootest I, Her, and Leo IV, they condclude that Pop III stars should have >0.8
$M_\odot$ or$m_{\rm char} > 1 M_\odot$
- (Yuan+ 2019? 2020?) Using self-organzing maps to identify dynamic groups in the Galactic halo.
- The dynamical group's stars have a common chemical enrichment origin and can be used to predict where EMP (and r-enhanced) stars exist
- UFDs are on a highly eccentric orbits and the inner halo may be populated by the remnants of earlier mergers of them
- Analyzing the simulations of Pallotini+ (2017) to study the satellite galaxies of a LBG at z=6.
- They have stellar masses between
$10^6$ and$10^9 M_\odot$ - SF in the satellites are mainly suppressed through Lyman-Werner feedback
- Their is no dependence on the distance from the main galaxy. The more massive (stellar) galaxies have long SF histories and old populations and grow through mergers. The lower mass galaxies have short SF histories, young populations, and are heavily affected by feedback.
- These galaxies (SFR > 2
$M_\odot$ /yr) should be detectable by JWST at z=6 within 3 hours. - Could the smallest galaxy be a high-z proto-GC? (e.g. Vanzella+ 2017)
- (Hosokawa+ 2013; Woods+ 2017; Haemmerle+ 2018) SMS with rapid (> 1
$M_\odot$ /yr) radiate nearly at Eddington, are 90% radiative, and can reach up to$\sim 3 \times 10^5 M_\odot/yr$ - SMS could be detected thorugh gravitational waves or ultra-long GRBs
- When the star is rotating, it's limited by the Omega-Gamma limit that accounts for angular momentum and radiation pressure
- (Haemmerle+ 2019) At very rapid accretion rates (~10
$M_\odot$ /yr), hydrostatic equilibrium is maintained through sound waves through the star. At higher masses, stars are well modeled by hylotropes (Begelman+)
- (Treister+ 2013) Stacking X-ray observations in the CANDELS fields at z>6 and found no excess in the galaxies. Derived upper limits to the cosmic BH mass density (see updated work by Ricarte & Natarajan 2018). Some explanations about the lack
- Active fraction of SMBHs is low (<20%)
- Most BH growth happens in lower mass and/or dustier, yet undetected, galaxies. However now this is ruled out.
- Fraction of low-z interlopers in the LBG samples are z>6 is higher than expected.
- (Buchner+ 2019) BH seeding rates continues even at low-z. SMBH merger rates peak z~2 at 100(?)/yr, which will be detectable by LISA
- (Treister+ 2011) Best way to determine the nature of BH seeds: to directly detect more "normal" SMBHs at z>6
- The Lynx Observatory will be able to detect this population
- (Ananna+ 2019; in prep) Semi-analytic model of reionization using a AGN population synthesis model (2019 paper), which is consistent with observations, showing that they contribute 10% to the ionizing budget.
- Performing rad-hydro simulations with PLUTO within the Bondi radius to study the circumnuclear disk around an IMBH
- Spherical coordinates:
$r = 5 \times 10^{3-5}$ AU - Gas inflow along the equatorial plane
- BH mass of
$10^4 M_\odot$ - Varies metallicity between [Z/H] = -3 -> -1 and inflow rates between 10 and 1,000 times the Eddington rate (
$10^{-4} M_\odot/yr$ )
- Spherical coordinates:
- The time-averaged accretion rate is 40x of Eddington in their fiducial model ([Z/H] = -2 and 100
$F_{\rm Edd}$ ). Accretion rates slightly increase with metallicity. With 1,000 $F_{\rm Edd}, 700x Eddington. - Key questions: How common is super-Eddington accretion? Is mass loss from the central slim disk negligible? Need to connect these simulations to smaller- and larger-scale simulations.
- (Chon & Omukai 2020) GADGET simulation of a cloud extracted from a cosmological simulation. Sink particle creation at
$2 \times 10^{16-17}$ cm$^{-3}$ and mergers. The accretion radius is ~12 AU and scales with particle mass. - At [Z/H] = -6, very little fragmentation occurs and SMSs form. At [Z/H] = -4, there is more fragmentation and SMSs form. At [Z/H] = -3, there is vigorous fragmentation, and a dense stellar cluster forms.
- Distribution at high-mass end reflects the large scale gas inflow. Massive stars preferentially grow in mass -> "super competitive accretion"
- (Latif+ 2014, 2015, 2016) Proto-SMS disk forms at high density (
$\gsim 10^{10}$ cm$^{-3}$), regardless of input physics such as H- opacity, magnetic fields, and subgrid turbulence models. - (Latif+ 2020) Studied four halos with varying spin parameters and found binary/multiple SMS candidates. Some clumps merge with each other, but some survive.
- Metal-free stars are unstable to radial pulsations, losing 8% of their mass during MS (Sonoi & Umeda 2012).
- They study this instability in very massive stars between 100-3000
$M_\odot$ between Z = 0 - 0.1$Z_\odot$ with MESA as a starting point in a perturbative analysis. - Stars are unstable in the early MS phase and red supergiant phase, where more mass loss is expected during that latter phase. During MS, they lose 10% and 20% of their mass for metal-free and metal-enriched cases, respectively. During the RSG phase, the respectively mass loss fraction is 10% and 100%.
- Thus IMBH formation isn't expected for the metal-enriched case, but above 300
$M_\odot$ and below [Z/H] = -2.5, it's possible and doesn't interrupt any mergers in a dense cluster.
- (Tagawa+ 2019) Using a 1D calculation combined with a semi-analytical model with a central star (1-100
$M_\odot$ ), collapsing gas, and surrounding stars (N-body). The core density ranges from$10^6$ to$10^{11}$ cm$^{-3}$ and the inflow rates between 0.01 and 1$M_\odot$ /yr. - For a high
$10^{11} cm$ ^{-3}$ core density, the star grows above the SMS rate and grows rapidly to a SMS. A factor of three lower in density results in a central star that contracts to the MS and expels the gas, limiting mass growth.
(Reinoso) Formation of massive black hole seeds following collapse and fragmentation of atomic cooling halos
- Simulations of a cluster of
$10^4 M_\odot$ stars and gas. The cluster is roughly spherical with a radius of 0.14 pc. They consider a range of parameters (number of stars, merger radii, etc) and run 336 direct N-body simulations. - Despite strong fragmentation, they still expect a massive 30-70
$M_\odot$ star to form in minihalos. In atomic cooling halos, this mass increases to 500-1700$M_\odot$ .