Search

Supermassive stars (SMSs) and heavy seed black holes, as their remnants, are promising candidates for Supermassive Black Hole (SMBH) progenitors, especially for ones observed in the early universe $ z\simeq 8.5-10$ by recent JWST observations. Expected cradles of SMSs are the atomic cooling halos ($M_{\rm halo}\simeq 10^7~{\rm M_\odot}$), where "cold accretion" emerges and possibly forms SMSs. We perform a suit of cosmological radiation hydrodynamics simulations and investigate star formation after the emergence of cold accretion, solving radiative feedback from stars inside the halo. We follow the mass growth of the protostars for $\sim 3~{\rm Myr}$, resolving the gas inflow down to $\sim 0.1~{\rm pc}$ scales. We discover that, after cold accretion emerges, multiple SMSs of $m_\star \gtrsim 10^5{\rm M_\odot}$ form at the halo centre with the accretion rates maintained at $\dot{m}_\star \simeq 0.04~{\rm M_\odot}~{\rm yr^{-1}}$ for $\lesssim 3~{\rm Myr}$. Cold accretion supplies gas at a rate of $\dot{M}_{\rm gas}\gtrsim 0.01-0.1~{\rm M_\odot}~{\rm yr^{-1}}$ from outside the halo virial radius to the central gas disc. Gravitational torques from spiral arms transport gas further inward, which feeds the SMSs. Radiative feedback from stars suppresses H$_2$ cooling and disc fragmentation, while photoevaporation is prevented by a dense envelope, which attenuates ionising radiation. Our results suggest that cold accretion can bring efficient BH mass growth after seed formation in the later universe. Moreover, cold accretion and gas migration inside the central disc increase the mass concentration and provide a promising formation site for the extremely compact stellar clusters observed by JWST., Comment: 19 pages, 15 figures, accepted by MNRAS

We present an analytic description of the spherically symmetric gravitational collapse of radiatively cooling gas clouds. The approach is based on developing the "one-zone" density-temperature relationship of the gas into a full dynamical model. We convert this density-temperature relationship into a barotropic equation of state, which we use to calculate the density and velocity profiles of the gas. From these quantities we calculate the time-dependent mass accretion rate onto the center of the cloud. The approach clarifies the mechanism by which radiative cooling induces gravitational instability. In particular, we distinguish the rapid, quasi-equilibrium contraction of a cooling gas core to high central densities from the legitimate instability this contraction establishes in the envelope. We develop a refined criterion for the mass scale of this instability, based only on the chemical-thermal evolution in the core. We explicate our model in the context of a primordial mini-halo cooled by molecular hydrogen, and then provide two further examples, a delayed collapse with hydrogen deuteride cooling and the collapse of an atomic cooling halo. In all three cases, our results agree well with full hydrodynamical treatments., Comment: 11+3 pages, 14+3 figures

JWST has brought us new insights into Cosmic Dawn with tentative detection of the unique signatures of metal-free Population III (Pop III) stars, such as strong HeII emission, extremely blue UV spectrum, and enhanced nitrogen abundance. Self-consistent theoretical predictions of the formation rates, sites, and masses of Pop III stars are crucial for interpreting the observations, but are challenging due to complex physical processes operating over the large range of length scales involved. One solution is to combine analytical models for the small-scale star formation process with cosmological simulations that capture the large-scale physics such as structure formation, radiation backgrounds, and baryon-dark matter streaming motion that regulate the conditions of Pop III star formation. We build an analytical model to predict the final masses of Pop III stars/clusters from the properties of star-forming clouds, based on the key results of small-scale star formation simulations and stellar evolution models. Our model for the first time considers the interplay between feedback and fragmentation and covers different modes of Pop III star formation ranging from ordinary small ($\sim 10-2000\ \rm M_\odot$) clusters in molecular-cooling clouds to massive ($\gtrsim 10^{4}\ \rm M_\odot$) clusters containing supermassive ($\sim 10^{4}-3\times 10^{5}\ \rm M_\odot$) stars under violent collapse of atomic-cooling clouds. As an example, the model is applied to the Pop III star-forming clouds in the progenitors of typical haloes hosting high-$z$ luminous quasars, which shows that formation of Pop III massive clusters is common ($\sim 20-70\%$) in such biased ($\sim4\sigma$) regions, and the resulting heavy black hole seeds from supermassive stars can account for a significant fraction of observed luminous ($\gtrsim 10^{46}\ \rm erg\ s^{-1}$) quasars at $z\sim 6$., Comment: 15+8 pages, 9 + 14 figures, accepted for publication in MNRAS

In the context of planet formation, growth from micron-sized grains to kilometer-sized planetesimals is a crucial question. Since the dust growth rate depends on the amount of dust, realizing planet formation scenarios based on dust growth is challenging in environments with low metallicity, i.e. less dust. We investigate dust growth during disk evolution, particularly focusing on the relationship with metallicity. We perform two-dimensional thin-disk hydrodynamic simulations to track the disk evolution over 300 kyr from its formation. The dust motion is solved separately from the gas motion, with its distribution changing due to drag forces from the gas. Dust size growth is also accounted for, with the magnitude of the drag force varying according to the dust size. We employ three models with metallicities of 1.0, 0.1, and 0.01 ${\rm Z}_{\odot}$, i.e. dust-to-gas mass ratios of 10$^{-2}$, 10$^{-3}$, and 10$^{-4}$, respectively. In the disks with the metallicities $\ge0.1$ ${\rm Z}_{\odot}$, the dust radii reach cm sizes, consistent with estimations from the dust growth timescale. Conversely, for the metallicity of 0.01 ${\rm Z}_{\odot}$, the maximum dust size is only 10$^{-2}$ cm, with almost no growth observed across the entire disk scale ($\sim$100 au). At the metallicities $\ge0.1$ ${\rm Z}_{\odot}$, the decoupling between grown dust and gas leads to non-uniform dust-to-gas mass ratios. However, deviations from the canonical value of this ratio have no impact on the gravitational instability of the disk. The formation of dust rings is confirmed in the innermost part of the disk ($\sim$10--30 au). The dust rings where the dust-to-gas mass ratio is enhanced, and the Stokes number reaches $\sim$0.1, are suitable environments for the streaming instability. We conjecture that planetesimal formation occurs through the streaming instability in these dust rings., Comment: 13 pages, 12 figures, submitted to A&A

The drag force experienced by astronomical objects moving through gaseous media (gas dynamical friction) plays a crucial role in their orbital evolution. Ostriker (1999) derived a formula for gas dynamical friction by linear analysis, and its validity has been confirmed through subsequent numerical simulations. However, the effect of gas accretion onto the objects on the dynamical friction is yet to be understood. In this study, we investigate the Mach number dependence of dynamical friction considering gas accretion through three-dimensional nested-grid simulations. We find that the net frictional force, determined by the sum of the gravitational force exerted by surrounding gas and momentum flux transferred by accreting gas, is independent of the resolution of simulations. Only the gas outside the Bondi-Hoyle-Lyttleton radius contributes to dynamical friction, because the gas inside this radius is eventually absorbed by the central object and returns the momentum obtained through the gravitational interaction with it. In the subsonic case, the front-back asymmetry induced by gas accretion leads to larger dynamical friction than predicted by the linear theory. Conversely, in the slightly supersonic case with the Mach number between 1 and 1.5, the nonlinear effect leads to a modification of the density distribution in a way reducing the dynamical friction compared with the linear theory. At a higher Mach number, the modification becomes insignificant and the dynamical friction can be estimated with the linear theory. We also provide a fitting formula for dynamical friction based on our simulations, which can be used in a variety of applications., Comment: 16 pages, 12 figures, 1 Tabel. Accepted for publication in ApJ

The stellar initial mass function (IMF) in the early universe is essential to understand the formation of ancient galaxies. To this end, we conduct a series of long-term radiation hydrodynamic simulations following star cluster formation, varying the metallicity from $Z/Z_\odot = 10^{-4}$ to $1$. We particularly consider the effects of protostellar radiative feedback, which modify the exact shape of the IMF and determine the star formation efficiency (SFE), i.e. the ratio between the mass in stars and the initial gas mass in the parental cloud. Our results show that the IMF changes from a Salpeter-type to a top-heavy function as the metallicity decreases. When $Z/Z_\odot \lesssim 10^{-2}$, the IMF becomes log-flat and distinct from a Salpeter-like IMF. Stellar feedback is effective in shaping both the low- and high-mass ends of the IMF. Heating of dust grains by stellar radiation suppresses small-scale fragmentation and reduces the number of low-mass stars with $M_* \lesssim 1~M_\odot$ at all metallicities. The ionizing radiation hinders the growth of massive stars, steepening the slope of the IMF at the high-mass end. The resulting feedback is more effective at lower metallicity, and star formation is regulated by stellar radiative feedback, with the SFE decreasing with decreasing metallicity. We suggest that the unexpectedly large number of UV-bright galaxies at $z>10$ reported by JWST observations can be explained by considering star cluster formation at $Z/Z_\odot \sim 10^{-2}$ or $10^{-3}$, where the IMF is top-heavy, but the SFE is not too low due to stellar feedback., Comment: 22 pages, 19 figures, submitted to MNRAS

We study the formation of Pop III stars by performing radiation hydrodynamics simulations for three different initial clouds extracted from cosmological hydrodynamics simulations. Starting from the cloud collapse stage, we follow the growth of protostars by accretion for $\sim 10^5$ yr until the radiative feedback from the protostars suppresses the accretion and the stellar properties are nearly fixed. We find that the Pop III stars form in massive and wide binaries/small-multiple stellar systems, with masses $>30\,M_\odot$ and separations $>2000$ au. We also find that the properties of the final stellar system correlate with those of the initial clouds: the total mass increases with the cloud-scale accretion rate, and the angular momentum of the binary orbit matches that of the initial cloud. While the total mass of the system in our simulations is consistent with our previous single-star formation simulations, individual masses are lower due to mass sharing, suggesting potential modification in the extent of feedback from Pop III stars in the subsequent evolution of the Universe. We also identify such systems as mini-binaries embedded in a wider outer multiple-star system, which could evolve into progenitors for observed gravitational wave events., Comment: 30 pages, 20 figures, published in ApJ

Direct imaging observations of planets revealed that wide-orbit ($>10$ au) giant planets exist even around subsolar-metallicity host stars and do not require metal-rich environments for their formation. A possible formation mechanism of wide-orbit giant planets in subsolar-metallicity environments is the gravitational fragmentation of massive protoplanetary discs. Here, we follow the long-term evolution of the disc for 1 Myr after its formation, which is comparable to disc lifetime, by way of a two-dimensional thin-disc hydrodynamic simulation with the metallicity of 0.1 ${\rm Z}_{\odot}$. We find a giant protoplanet that survives until the end of the simulation. The protoplanet is formed by the merger of two gaseous clumps at $\sim$0.5 Myr after disc formation, and then it orbits $\sim$200 au from the host star for $\sim$0.5 Myr. The protoplanet's mass is $\sim$10 ${\rm M}_{\rm J}$ at birth and gradually decreases to 1 ${\rm M}_{\rm J}$ due to the tidal effect from the host star. The result provides the minimum mass of 1 ${\rm M}_{\rm J}$ for protoplanets formed by gravitational instability in a subsolar-metallicity disc. We anticipate that the mass of a protoplanet experiencing reduced mass loss thanks to the protoplanetary contraction in higher resolution simulations can increase to $\sim$10 ${\rm M}_{\rm J}$. We argue that the disc gravitational fragmentation would be a promising pathway to form wide-orbit giant planets with masses of $\ge1$ ${\rm M}_{\rm J}$ in subsolar-metallicity environments., Comment: 11 pages, 8 figures, submitted to MNRAS

Young massive clusters (YMCs) are the most massive star clusters forming in nearby galaxies and are thought to be a young analogue to the globular clusters. Understanding the formation process of YMCs leads to looking into very efficient star formation in high-redshift galaxies suggested by recent JWST observations. We investigate possible observational signatures of their formation stage, particularly when the mass of a cluster is increasing via accretion from a natal molecular cloud. To this end, we study the broad-band continuum emission from ionized gas and dust enshrouding YMCs, whose formation is followed by recent radiation-hydrodynamics simulations. We perform post-process radiative transfer calculations using simulation snapshots and find characteristic spectral features at radio and far-infrared frequencies. We show that a striking feature is long-lasting, strong free-free emission from a $\sim$ 10pc-scale HII region with a large emission measure of $\gtrsim 10^7 \mathrm{cm}^{-6} \ \mathrm{pc}$, corresponding to the mean electron density of $\gtrsim 10^3~\mathrm{cm}^{-3}$. There is a turnover feature below $\sim$ 10 GHz, a signature of the optically-thick free-free emission, often found in Galactic ultra-compact HII regions. These features come from the peculiar YMC formation process, where the cluster's gravity effectively traps photoionized gas for a long duration and enables continuous star formation within the cluster. Such large and dense HII regions show distinct distribution on the density-size diagram, apart from the standard sequence of Galactic HII regions. This is consistent with the observational trend inferred for extragalactic HII regions associated with YMCs., Comment: 12 pages, 10 figures, accepted for publication in MNRAS

Massive close binary stars with extremely small separations have been observed, and they are possible progenitors of gravitational-wave sources. The evolution of massive binaries in the protostellar accretion stage is key to understanding their formation process. We, therefore, investigate how close the protostars, consisting of a high-density core and a vast low-density envelope, can approach each other but not coalesce. To investigate the coalescence conditions, we conduct smoothed particle hydrodynamics simulations following the evolution of equal-mass binaries with different initial separations. Since Population (Pop) I and III protostars have similar interior structures, we adopt a specific Pop~III model with the mass and radius of $7.75\;M_{\odot}$ and $61.1\;R_{\odot}$ obtained by the stellar evolution calculations. Our results show that the binary separation decreases due to the transport of the orbital angular momentum to spin angular momentum. If the initial separation is less than about 80 per~cent of the sum of the protostellar radius, the binary coalesces in a time shorter than the tidal lock timescale. The mass loss up to the merging is $\lesssim 3$ per~cent. After coalescence, the star rotates rapidly, and its interior structure is independent of the initial separation. We conclude that there must be some orbital shrinking mechanism after the protostars contract to enter the zero-age main-sequence stage., Comment: 17 pages, 13 figures, Accepted for Publication in ApJ

Direct collapse of supermassive stars is a possible pathway to form supermassive black hole seeds at high redshifts. Whereas previous three-dimensional (3D) simulations demonstrate that supermassive stars form via rapid mass accretion, those resolving the stellar interior have been limited. We here report 3D radiation-hydrodynamic (RHD) simulations following the evolution of rapidly accreting protostars resolving the stellar interior. We use an adaptive mesh refinement code with our newly developed RHD solver employing an explicit M1 closure method. We follow the early evolution until the stellar mass reaches $\sim 10~M_\odot$ from two different initial configurations of spherical and turbulent clouds. We demonstrate that, in both the cases, a swollen protostar whose radius is $100\mathrm{-}1000~R_\odot$ appears, as predicted by the stellar evolution calculations. Its effective temperature remains a few thousand Kelvin, and the radiative feedback by ionizing photons is too weak to disturb the accretion flow up to the epoch examined in this work. In the turbulent case, the protostar rotates rapidly at more than 0.4 times the Keplerian velocity owing to the angular momentum provided by the initial turbulence. The protostar approximates an oblate spheroid, and its equatorial radius is more than twice the polar radius. Our results suggest that we need to consider the rapid stellar rotation to elucidate the realistic 3D protostellar evolution in the supermassive star formation., Comment: 9 pages, 5 figures, accepted to ApJ

We investigate the first emergence of the so-called cold accretion, the accretion flows deeply penetrating a halo, in the early universe with cosmological N-body/SPH simulations. We study the structure of the accretion flow and its evolution within small halos with $\lesssim 10^8~{\rm M}_\odot$ with sufficiently high spatial resolutions down to $\sim 1 \ {\rm pc}$ scale. While previous studies only follow the evolution for a short period after the primordial cloud collapse, we follow the long-term evolution until the cold accretion first appears, employing the sink particle method. We show that the cold accretion emerges when the halo mass exceeds $\sim 2.2\times 10^7 \ {\rm M}_\odot\left\{\left(1+z\right)/15 \right\}^{-3/2}$, the ${\it minimum}$ halo masses above which the accretion flow penetrates halos. We further continue simulations to study whether the cold accretion provides the dense shock waves, which have been proposed to give birth to supermassive stars (SMSs). We find that the accretion flow eventually hits a compact disc near the halo centre, creating dense shocks over a wide area of the disc surface. The resulting post-shock gas becomes dense and hot enough with its mass comparable to the Jeans mass $M_{\rm J}\sim 10^{4-5} \ {\rm M}_\odot$, a sufficient amount to induce the gravitational collapse, leading to the SMS formation., Comment: 17 pages, 17 figures, accepted for publication in MNRAS

Cosmic metallicity evolution possibly creates the diversity of star formation modes at different epochs. Gravitational fragmentation of circumstellar discs provides an important formation channel of multiple star systems, including close binaries. We here study the nature of disc fragmentation, systematically performing a suite of two-dimensional radiation-hydrodynamic simulations, in a broad range of metallicities, from the primordial to the solar values. In particular, we follow relatively long-term disc evolution over 15 kyr after the disc formation, incorporating the effect of heating by the protostellar irradiation. Our results show that the disc fragmentation occurs at all metallicities $1$--$0$ $Z_{\odot}$, yielding self-gravitating clumps. Physical properties of the clumps, such as their number and mass distributions, change with the metallicity due to different gas thermal evolution. For instance, the number of clumps is the largest for the intermediate metallicity range of $10^{-2}$--$10^{-5}$ $Z_{\odot}$, where the dust cooling is effective exclusively in a dense part of the disc and causes the fragmentation of spiral arms. The disc fragmentation is more modest for $1$--$0.1$ $Z_{\odot}$ thanks to the disc stabilization by the stellar irradiation. Such metallicity dependence agrees with the observed trend that the close binary fraction increases with decreasing metallicity in the range of $1$--$10^{-3}$ $Z_{\odot}$., Comment: 17 pages, 16 figures, submitted to MNRAS

We study the effects of the magnetic fields on the formation of massive, self-gravitationally bound cores (MBCs) in high-speed cloud-cloud collisions (CCCs). Extending our previous work (Sakre et al. 2021), we perform magnetohydrodynamic simulations following the high-speed (20 - 40 km s$^{-1}$) collisions between two magnetized (4 $\mu$G initially), turbulent clouds of different sizes in the range of 7 - 20 pc. We show that a magnetic field effect hinders the core growth, particularly after a short-duration collision during which cores cannot get highly bound. In such a case, a shocked region created by the collision rapidly expands to the ambient medium owing to the enhanced magnetic pressure, resulting in the destruction of the highly unbound cores and suppression of gas accretion to massive cores. This negative effect on the MBC formation is a phenomenon not seen in the past hydrodynamic simulations of similar CCC models. Together with our previous work, we conclude that the magnetic fields provide the two competing effects on the MBC formation in CCC; while they promote the mass accumulation into cores during a collision, they operate to destroy cores or hinder the core growth after the collision. The duration of collision determines which effect prevails, providing the maximum collision speed for the MBC formation with given colliding clouds. Our results agree with the observed trend among CCC samples in the corresponding column density range; clouds with higher relative velocity require higher column density for the formation of massive stars (Enokiya et al. 2021)., Comment: Submitted to MNRAS journal

We investigate the formation of merging binary black holes (BHs) through isolated binary evolution, performing binary population synthesis calculations covering an unprecedentedly wide metallicity range of Population (Pop) I, II, III, and extremely metal-poor (EMP) binary stars. We find that the predicted merger rate density and primary BH mass ($m_1$) distribution are consistent with the gravitational wave (GW) observations. Notably, Pop III and EMP ($< 10^{-2}$ $Z_\odot$) binary stars yield most of the pair instability (PI) mass gap events with $m_1 = 65$--$130$ $M_\odot$. Pop III binary stars contribute more to the PI mass gap events with increasing redshift, and all the PI mass gap events have the Pop III origin at redshifts $\gtrsim 8$. Our result can be assessed by future GW observations in the following two points. First, there are no binary BHs with $m_1=100$--$130$ $M_\odot$ in our result, and thus the $m_1$ distribution should suddenly drop in the range of $m_1=100$--$130$ $M_\odot$. Second, the PI mass gap event rate should increase toward higher redshift up to $\sim 11$, since those events mainly originate from the Pop III binary stars. We find that the following three assumptions are needed to reproduce the current GW observations: a top-heavy stellar initial mass function and the presence of close binary stars for Pop III and EMP binary stars, and inefficient convective overshoot in the main-sequence phase of stellar evolution. Without any of the above, the number of PI mass gap events becomes too low to reproduce current GW observations., Comment: 26 pages, 13 figures, 2 tables, ApJ accepted. BSEEMP, a binary population synthesis code used in this paper, is published on github: https://github.com/atrtnkw/bseemp

We study the early growth of massive seed black holes (BHs) via accretion in protogalactic nuclei where the stellar bulge component is assembled, performing axisymmetric two-dimensional radiation hydrodynamical simulations. We find that when a seed BH with $M_\bullet \sim 10^5~M_\odot$ is embedded in dense metal-poor gas ($Z=0.01~Z_\odot$) with a density of $\gtrsim 100~{\rm cm}^{-3}$ and bulge stars with a total mass of $M_\star \gtrsim 100~M_\bullet$, a massive gaseous disk feeds the BH efficiently at rates of $\gtrsim 0.3-1~M_\odot~{\rm yr}^{-1}$ and the BH mass increases nearly tenfold within $\sim 2$ Myr. This rapid accretion phase lasts until a good fraction of the gas bounded within the bulge accretes onto the BH, although the feeding rate is regulated owing to strong outflows driven by ionizing radiation emitted from the accreting BH. The transient growing mode can be triggered for seed BHs formed in massive dark-matter halos with masses of $\gtrsim 10^9~M_\odot$ at $z\sim 15-20$ (the virial temperature is $T_{\rm vir}\simeq 10^5~{\rm K}$). The host halos are heavier and rarer than those of typical first galaxies, but are more likely to end up in quasar hosts by $z\simeq 6$. This mechanism naturally yields a mass ratio of $M_\bullet/M_\star >0.01$ higher than the value seen in the local universe and the existence of such overmassive BHs provides us a unique opportunity of detecting highly accreting seed BHs at $z\sim 15$ with AB magnitude of $m_{\rm AB} \sim26 - 29$ mag at $2~\mu{\rm m}$ (rest-frame 10 eV) by the upcoming observations by the James Webb Space Telescope and Nancy Grace Roman Space Telescope., Comment: 28 pages, 11 figures, accepted for publication to ApJ. Movies of the simulation are available at https://www.youtube.com/playlist?list=PLqDrKvxmu0l4fyJ74u5ZYyQ8PmwuNZA32

Combining numerical simulations and analytical modeling, we investigate whether close binary systems form by the effect of magnetic braking. Using magnetohydrodynamics simulations, we calculate the cloud evolution with a sink, for which we do not resolve the binary system or binary orbital motion to realize long-term time integration. Then, we analytically estimate the binary separation using the accreted mass and angular momentum obtained from the simulation. In unmagnetized clouds, wide binary systems with separations of >100 au form, in which the binary separation continues to increase during the main accretion phase. In contrast, close binary systems with separations of <100 au can form in magnetized clouds. Since the efficiency of magnetic braking strongly depends on both the strength and configuration of the magnetic field, they also affect the formation conditions of a close binary. In addition, the protostellar outflow has a negative impact on close binary formation, especially when the rotation axis of the prestellar cloud is aligned with the global magnetic field. The outflow interrupts the accretion of gas with small angular momentum, which is expelled from the cloud, while gas with large angular momentum preferentially falls from the side of the outflow onto the binary system and widens the binary separation. This study shows that a cloud with a magnetic field that is not parallel to the rotation axis is a favorable environment for the formation of close binary systems., Comment: Accepted for publication in MNRAS

The detections of gravitational waves (GW) by LIGO/Virgo collaborations provide various possibilities to physics and astronomy. We are quite sure that GW observations will develop a lot both in precision and in number owing to the continuous works for the improvement of detectors, including the expectation to the newly joined detector, KAGRA, and the planned detector, LIGO-India. In this occasion, we review the fundamental outcomes and prospects of gravitational wave physics and astronomy. We survey the development focusing on representative sources of gravitational waves: binary black holes, binary neutron stars, and supernovae. We also summarize the role of gravitational wave observations as a probe of new physics., Comment: 89pages, Prog. Theor. Exp. Phys. accepted

Recent high-resolution simulations demonstrate that disks around primordial protostars easily fragment in the accretion phase before the protostars accrete less than a solar mass. To understand why the gravitational instability generally causes the fragmentation so early, we develop a one-dimensional (1D) non-steady model of the circumstellar disk that takes the mass supply from an accretion envelope into account. We also compare the model results to a three-dimensional (3D) numerical simulation performed with a code employing the adaptive mesh refinement. Our model shows that the self-gravitating disk, through which the Toomre $Q$ parameter is nearly constant at $Q \sim 1$, gradually spreads as the disk is fed by the gas infalling from the envelope. We further find that the accretion rate onto the star is an order of magnitude smaller than the mass supply rate onto the disk. This discrepancy makes the disk more massive than the protostar in an early evolutionary stage. Most of the infalling gas is used to extend the outer part of the self-gravitating disk rather than transferred inward toward the star through the disk. We find that similar evolution also occurs in the 3D simulation, where the disk becomes three times more massive than the star before the first fragmentation occurs. Our 1D disk model well explains the evolution of the disk-to-star mass ratio observed in the simulation. We argue that the formation of such a massive disk leads to the early disk fragmentation., Comment: 22 pages, 14 figures, accepted for publication in ApJ

Supermassive stars (SMSs) with $\sim10^{4-5}~\mathrm{M}_{\odot}$ are candidate objects for the origin of supermassive black holes observed at redshift $z$>6. They are supposed to form in primordial-gas clouds that provide the central stars with gas at a high accretion rate, but their growth may be terminated in the middle due to the stellar ionizing radiation if the accretion is intermittent and its quiescent periods are longer than the Kelvin-Helmholtz (KH) timescales at the stellar surfaces. In this paper, we examine the role of the ionizing radiation feedback based on the accretion history in two possible SMS-forming clouds extracted from cosmological simulations, following their evolution with vertically-integrated two-dimensional hydrodynamic simulations with detailed thermal and chemical models. The consistent treatment of the gas thermal evolution is crucial for obtaining the realistic accretion history, as we demonstrate by performing an additional run with a barotropic equation of state, in which the fluctuation of the accretion rate is artificially suppressed. We find that although the accretion becomes intermittent due to the formation of spiral arms and clumps in gravitationally unstable disks, the quiescent periods are always shorter than the KH timescales, implying that SMSs can form without affected by the ionizing radiation., Comment: 14 pages, 11 figures, accepted for publication in MNRAS

The evolution of protostellar outflows is investigated under different mass accretion rates in the range $\sim10^{-5}-10^{-2} {\rm M}_\odot$ yr$^{-1}$ with three-dimensional magnetohydrodynamic simulations. A powerful outflow always appears in strongly magnetized clouds with $B_0 \gtrsim B_{\rm 0, cr}$ $=10^{-4} (M_{\rm cl}/100 {\rm M}_\odot){\rm G}$, where $M_{\rm cl}$ is the cloud mass. When a cloud has a weaker magnetic field, the outflow does not evolve promptly with a high mass accretion rate. In some cases with moderate magnetic fields $B_0$ slightly smaller than $B_{\rm 0,cr}$, the outflow growth is suppressed or delayed until the infalling envelope dissipates and the ram pressure around the protostellar system is significantly reduced. In such an environment, the outflow begins to grow and reaches a large distance only during the late accretion phase. On the other hand, the protostellar outflow fails to evolve and is finally collapsed by the strong ram pressure when a massive $(\gtrsim 100 {\rm M}_\odot)$ initial cloud is weakly magnetized with $B_0 \lesssim 100 \mu {\rm G}$. The failed outflow creates a toroidal structure that is supported by magnetic pressure and encloses the protostar and disk system. Our results indicate that high-mass stars form only in strongly magnetized clouds, if all high-mass protostars possess a clear outflow. If we would observe either very weak or no outflow around evolved protostars, it means that strong magnetic fields are not necessarily required for high-mass star formation. In any case, we can constrain the high-mass star formation process from observations of outflows., Comment: Accepted for publication in MNRAS

We perform the first three-dimensional radiation hydrodynamical simulations that investigate the growth of intermediate-mass BHs (IMBHs) embedded in massive self-gravitating, dusty nuclear accretion disks. We explore the dependence of mass accretion efficiency on the gas metallicity $Z$ and mass injection at super-Eddington accretion rates from the outer galactic disk $\dot{M}_{\rm in}$, and find that the central BH can be fed at rates exceeding the Eddington rate only when the dusty disk becomes sufficiently optically thick to ionizing radiation. In this case, mass outflows from the disk owing to photoevaporation is suppressed and thus a large fraction ($\gtrsim 40\%$) of the mass injection rate can feed the central BH. The conditions are expressed as $\dot{M}_{\rm in} > 2.2\times 10^{-1}~M_\odot ~{\rm yr}^{-1} (1+Z/10^{-2}~Z_\odot)^{-1}(c_{\rm s}/10~{\rm km~s}^{-1})$, where $c_{\rm s}$ is the sound speed in the gaseous disk. With increasing numerical resolution, vigorous disk fragmentation reduces the disk surface density and dynamical heating by formed clumps makes the disk thickness higher. As a result, the photoevaorative mass-loss rate rises and thus the critical injection rate increases for fixed metallicity. This process enables super-Eddington growth of BHs until the BH mass reaches $M_{\rm BH} \sim 10^{7-8}~M_\odot$, depending on the properties of the host dark-matter halo and metal-enrichment history. In the assembly of protogalaxies, seed BHs that form in overdense regions with a mass variance of 3-4$\sigma$ at $z\sim 15-20$ are able to undergo short periods of their rapid growth and transits into the Eddington-limited growth phase afterwards to be supermassive BHs observed at $z>6-7$., Comment: 23 pages, 10 figures, submitted to ApJ

The direct collapse (DC) is a promising mechanism that provides massive seed black holes (BHs) with $\sim 10^{5}~M_{\odot}$ in the early universe. To study a long-term accretion growth of a DCBH thus formed, we perform cosmological radiation-hydrodynamics simulations, extending our previous work where we investigated its formation stage. With a high spatial resolution down below the Bondi radius, we show that the accretion rate onto the BH is far below the Eddington value. Such slow mass growth is partly because of the strong radiative feedback from the accreting BH. Moreover, we find that the BH has a large velocity of $\sim 100~{\rm km~s^{-1}}$ relative to the gas after it falls into the first galaxy, which substantially reduces the accretion rate. The latter effect stems from the fact that the DCBHs form in metal-free environments typically at $\sim 1~$kpc from the galaxy. The BH accelerates as it approaches the galactic center due to the gravity. The relative velocity never damps after that, and the BH does not settle down to the galactic center but continues to wander around it. An analytic estimate predicts that the DCBH formation within $\sim 100$~pc around the galactic center is necessary to decelerate the BH with dynamical friction before $z=7$. Since metal enrichment with $Z \sim 10^{-5} - 10^{-3}~Z_\odot$ is expected in such a case, the formation of DCBHs in the metal-poor environments is preferable for the subsequent rapid growth., Comment: 13 pages, 8 figures, submitted to MNRAS

We study star cluster formation in various environments with different metallicities and column densities by performing a suite of three-dimensional radiation hydrodynamics simulations. We find that the photoionization feedback from massive stars controls the star formation efficiency (SFE) in a star-forming cloud, and its impact sensitively depends on the gas metallicity $Z$ and initial cloud surface density $\Sigma$. At $Z=1~Z_{\odot}$, SFE increases as a power law from 0.03 at $\Sigma = 10~M_{\odot}{\rm pc^{-2}}$ to 0.3 at $\Sigma = 300~M_{\odot}{\rm pc^{-2}}$. In low-metallicity cases $10^{-2}- 10^{-1} Z_{\odot}$, star clusters form from atomic warm gases because the molecule formation time is not short enough with respect to the cooling or dynamical time. In addition, the whole cloud is disrupted more easily by expanding H{\sc ii} bubbles which have higher temperature owing to less efficient cooling. With smaller dust attenuation, the ionizing radiation feedback from nearby massive stars is stronger and terminate star formation in dense clumps. These effects result in inefficient star formation in low-metallicity environments: the SFE drops by a factor of $\sim 3$ at $Z=10^{-2}~Z_{\odot}$ compared to the results for $Z=1~Z_{\odot}$, regardless of $\Sigma$. Newborn star clusters are also gravitationally less bound. We further develop a new semi-analytical model that can reproduce the simulation results well, particularly the observed dependencies of the SFEs on the cloud surface densities and metallicities., Comment: 17 pages, 13 figures, accepted for publication in MNRAS

The star formation in molecular clouds is inefficient. The ionizing EUV radiation ($h \nu \geq 13.6$ eV) from young clusters has been considered as a primary feedback effect to limit the star formation efficiency (SFE). We here focus on effects of the stellar FUV radiation (6 eV $\leq h \nu \leq$ 13.6 eV) during the cloud disruption stage. The FUV radiation may further reduce the SFE via photoelectric heating, and it also affects the chemical states of the gas that is not converted to stars ("cloud remnants") via photodissociation of molecules. We have developed a one-dimensional semi-analytic model which follows the evolution of both the thermal and chemical structure of a photodissociation region (PDR) during the dynamical expansion of an HII region. We investigate how the FUV feedback limits the SFE, supposing that the star formation is quenched in the PDR where the temperature is above a threshold value (e.g., 100K). Our model predicts that the FUV feedback contributes to reduce the SFEs for the massive ($M_{\rm cl} \gtrsim 10^5 M_{\odot}$) clouds with the low surface densities ($\Sigma_{\rm cl} \lesssim 100$ M$_{\odot}$pc$^{-2}$). Moreover, we show that a large part of the H$_2$ molecular gas contained in the cloud remnants should be "CO-dark" under the FUV feedback for a wide range of cloud properties. Therefore, the dispersed molecular clouds are potential factories of the CO-dark gas, which returns into the cycle of the interstellar medium., Comment: 16 pages, 10 figures, published in MNRAS

We study the formation of very metal-poor stars under protostellar radiative feedback effect. We use cosmological simulations to identify low-mass dark matter halos and star-forming gas clouds within them. We then follow protostar formation and the subsequent long-term mass accretion phase of over one million years using two-dimensional radiation-hydrodynamics simulations. We show that the critical physical process that sets the final mass is formation and expansion of a bipolar HII region. The process is similar to the formation of massive primordial stars, but radiation pressure exerted on dust grains also contributes to halting the accretion flow in the low-metallicity case. We find that the net feedback effect in the case with metallicity $Z = 10^{-2}~Z_{\odot}$ is stronger than in the case with $Z \sim 1~Z_{\odot}$. With decreasing metallicity, the radiation pressure effect becomes weaker, but photoionization heating of the circumstellar gas is more efficient owing to the reduced dust attenuation. In the case with $Z = 10^{-2}~Z_{\odot}$, the central star grows as massive as 200 solar-masses, similarly to the case of primordial star formation. We conclude that metal-poor stars with a few hundred solar masses can be formed by gas accretion despite the strong radiative feedback., Comment: 18 pages, 13 figures, accepted for publication in MNRAS

We review the current status of knowledge concerning the early phases of star formation during cosmic dawn. This includes the first generations of stars forming in the lowest mass dark matter halos in which cooling and condensation of gas with primordial composition is possible at very high redshift ($z > 20$), namely metal-free Population III stars, and the first generation of massive black holes forming at such early epochs, the so-called black hole seeds. The formation of black hole seeds as end states of the collapse of Population III stars, or via direct collapse scenarios, is discussed. In particular, special emphasis is given to the physics of supermassive stars as potential precursors of direct collapse black holes, in light of recent results of stellar evolution models, and of numerical simulations of the early stages of galaxy formation. Furthermore, we discuss the role of the cosmic radiation produced by the early generation of stars and black holes at high redshift in the process of reionization.

Coalescence of intermediate-mass black holes (IMBHs) as a result of the migration toward galactic centers via dynamical friction may contribute to the formation of supermassive BHs. Here we reinvestigate the gaseous dynamical friction, which was claimed to be inefficient with radiative feedback from BHs in literature, by performing 3D radiation-hydrodynamics simulations that solve the flow structure in the vicinity of BHs. We consider a $10^4~M_\odot$ BH moving at the velocity $V_{\rm flow}$ through the homogeneous medium with metallicity $Z$ in the range of $0-0.1~Z_\odot$ and density $n_{\infty}$. We show that, if $n_{\infty} \lesssim 10^{6}~{\rm cm^{-3}}$ and $V_{\rm flow} \lesssim 60~{\rm km~s^{-1}}$, the BH is accelerated forward because of the gravitational pull from a dense shell ahead of an ionized bubble around the BH, regardless of the value of $Z$. If $n_{\infty} \gtrsim 10^{6}~{\rm cm^{-3}}$, however, our simulation shows the opposite result. The ionized bubble and associating shell temporarily appear, but immediately go downstream with significant ram pressure of the flow. They eventually converge into a massive downstream wake, which gravitationally drags the BH backward. The BH decelerates over the timescale of $\sim 0.01$~Myr, much shorter than the dynamical timescale in galactic disks. Our results suggest that IMBHs that encounter the dense clouds rapidly migrate toward galactic centers, where they possibly coalescence with others., Comment: 12 pages, 11 figures, submitted to MNRAS

We study the formation of massive Population III binary stars using a newly developed radiation hydrodynamics code with the adaptive mesh refinement and adaptive ray-tracing methods. We follow the evolution of a typical primordial star-forming cloud obtained from a cosmological hydrodynamics simulation. Several protostars form as a result of disk fragmentation and grow in mass by the gas accretion, which is finally quenched by the radiation feedback from the protostars. Our code enables us, for the first time, to consider the feedback by both the ionizing and dissociating radiation from the multiple protostars, which is essential for self-consistently determining their final masses. At the final step of the simulation, we observe a very wide ($\gtrsim 10^4\,\mathrm{au}$) binary stellar system consisting of $60$ and $70\,M_\odot$ stars. One of the member stars also has two smaller mass ($10\,M_\odot$) companion stars orbiting at $200$ and $800\,\mathrm{au}$, making up a mini-triplet system. Our results suggest that massive binary or multiple systems are common among Population III stars., Comment: 6 pages, 3 figures, published in ApJL, movie available at https://www.youtube.com/watch?v=794O0yGWGp0

One of the major goals of next-generation space-borne and ground-based telescopes is to detect and characterize the first galaxies that were in place in the first few hundred million years after the big bang. We study the spectral energy distribution (SED) of the first galaxies and discuss the prospects for detection and identification. We consider very young star-forming galaxies at $z=15$ and incorporate the contribution from pre-main-sequence (PMS) stars. Unlike in the present-day galaxies, primordial protostars are not embedded in dusty gas clouds, and hence the light from them can be visible at a wide range of wavelengths. We use MESA code to follow the PMS evolution and use the BT-Settl model to calculate the SED of individual PMS stars. We show that PMS stars contribute to boost the flux in the mid-infrared, and that the galaxy SED at very early evolutionary phases is overall redder than at later phases. The infrared flux contribution is comparable to that caused by emission lines powered by massive stars. We argue that the contribution from PMS stars is important for characterizing young galaxies in the early Universe and also for target selection with future deep galaxy surveys., Comment: 5 pages, 4 figures, published online in MNRAS letters

The disk fragmentation is a possible process leading to the formation of Population III stellar binary systems. However, numerical simulations show diverse fates of the fragments; some evolve into stable binaries and others merge away with a central star. To clarify the physics behind such diversity, we perform a series of three dimensional hydrodynamics simulations in a controlled manner. We insert a point particle mimicking a fragment in a self-gravitating disk, where the initial mass and position are free parameters, and follow the orbital evolution for several tens of orbits. The results show great diversity even with such simple experiments. Some particles shortly merge away after migrating inward, but others survive as the migration stalls with the gap-opening in the disk. We find that our results are well interpreted postulating that the orbital angular momentum is extracted by (i) the gravitational torque from the disk spiral structure, and (ii) tidal disruption of a gravitationally-bound envelope around the particle. Our analytic evaluations show the processes (i) and (ii) are effective in an outer and inner part of the disk respectively. There is a window of the gap-opening in the middle, if the envelope mass is sufficiently large. These all agree with our numerical results. We further show that the binaries, which appear for the "survival" cases, gradually expand while accreting the disk gas. Our theoretical framework is freely scalable to be applied for the present-day star and planet formation., Comment: 16 pages, 14 figures, published in MNRAS

Following our recent work devoted to the effect of accretion on the pre-main-sequence evolution of low-mass stars, we perform a detailed analysis of episodic excursions of low-mass protostars in the Hertzsprung-Russell (H-R) diagram triggered by strong mass accretion bursts typical of FU Orionis-type objects (FUors). These excursions reveal themselves as sharp increases in the stellar total luminosity and/or effective temperature of the protostar and can last from hundreds to a few thousands of years, depending on the burst strength and characteristics of the protostar. During the excursions, low-mass protostars occupy the same part of the H-R diagram as young intermediate-mass protostars in the quiescent phase of accretion. Moreover, the time spent by low-mass protostars in these regions is on average a factor of several longer than that spent by the intermediate-mass stars in quiescence. During the excursions, low-mass protostars pass close to the position of most known FUors in the H-R diagram, but owing to intrinsic ambiguity the model stellar evolutionary tracks are unreliable in determining the FUor properties. We find that the photospheric luminosity in the outburst state may dominate the accretion luminosity already after a few years after the onset of the outburst, meaning that the mass accretion rates of known FUors inferred from the bolometric luminosity may be systematically overestimated, especially in the fading phase., Comment: 15 pages, 12 figures

The super-Eddington accretion onto intermediate seed BHs is a potential formation mode of supermassive black holes exceeding $10^9~M_\odot$ in the early universe. We here investigate how such rapid accretion may occur with finite amounts of heavy elements contained in the gas and dust. In our 1D radiation-hydrodynamics simulations, the radiative transfer is solved for both the direct UV lights emitted by an accretion disk and the diffuse IR lights thermally emitted by dust grains. Our results show that the radiative force by the IR lights causes a strong feedback to regulate the mass accretion. The resulting mean accretion rate is lower with the higher metallicity, and there is the critical metallicity $Z \sim 10^{-2}~Z_\odot$, above which the super-Eddington accretion is prevented by the radiation pressure of the IR lights. With this taken into account, we examine if the dusty super-Eddington accretion occurs in young galaxies using a simple model. We show that a sufficient number of galaxies at $z \gtrsim 10$ can be such potential sites if BHs accrete the cold dense gas with $T \sim 10^2$ K, approximately the thermal equilibrium value at $Z = 10^{-2}~Z_\odot$. We argue that the efficiency of the BH growth via the rapid accretion depends on the metallicity, and that the metallicity slightly lower than $10^{-2}~Z_\odot$ provides a chance for the most efficient growth., Comment: 15 pages, 9 figures. Accepted for publication in MNRAS. Eq. (14) and (21) were corrected

In recent years, the discovery of massive quasars at z~7 has provided a striking challenge to our understanding of the origin and growth of supermassive black holes in the early Universe. Mounting observational and theoretical evidence indicates the viability of massive seeds, formed by the collapse of supermassive stars, as a progenitor model for such early, massive accreting black holes. Although considerable progress has been made in our theoretical understanding, many questions remain regarding how (and how often) such objects may form, how they live and die, and how next generation observatories may yield new insight into the origin of these primordial titans. This review focusses on our present understanding of this remarkable formation scenario, based on discussions held at the Monash Prato Centre from November 20--24, 2017, during the workshop "Titans of the Early Universe: The Origin of the First Supermassive Black Holes.", Comment: Solicited review article (accepted) for Publications of the Astronomical Society of Australia. 38 pages, 15 figures, 1 table

The formation of supermassive stars (SMSs) is a possible pathway to seed supermassive black holes in the early universe. This chapter summarizes recent theoretical efforts to understand their evolution, highlighting effects of very rapid accretion at the rates of $\dot{M}_* \gtrsim 0.1~$M$_\odot$ yr$^{-1}$. Stellar evolution calculations predict that such an accreting SMS has a characteristic feature, i.e., a very large radius that monotonically increases as the stellar mass increases. The radius exceeds $7000$ R$_\odot$ (or $30$ AU) after the star accretes the gas of more than $10^3$ M$_\odot$. We show that the emergence of the "supergiant protostar" stage is a key for the formation of SMSs, because resulting radiative feedback against the accretion flow is substantially weakened during this stage. We also show that these SMSs end their lives while the accretion continues, and that their final fates vary with different accretion rates., Comment: Preprint of the Chapter "Evolution and Final Fates of Rapidly Accreting Supermassive Stars", to be published in the review volume "Formation of the First Black Holes", Latif M. and Schleicher D.R.G., eds., World Scientific Publishing Company, 2018, pp. [see https://www.worldscientific.com/worldscibooks/10.1142/10652]

We perform a suite of radiation hydrodynamics simulations of photoevaporating disks with varying the metallicity in a wide range of $10^{-3} \, Z_\odot \leq Z \leq 10^{0.5} \, Z_\odot $. We follow the disk evolution for over $\sim 5000$ years by solving hydrodynamics, radiative transfer, and non-equilibrium chemistry. Our chemistry model is updated from the first paper of this series by adding X-ray ionization and heating. We study the metallicity dependence of the disk photoevaporation rate and examine the importance of X-ray radiation. In the fiducial case with solar metallicity, including the X-ray effects does not significantly increase the photoevaporation rate when compared to the case with ultra-violet (UV) radiation only. At sub-solar metallicities in the range of $Z \gtrsim 10^{-1.5} \, Z_\odot $, the photoevaporation rate increases as metallicity decreases owing to the reduced opacity of the disk medium. The result is consistent with the observational trend that disk lifetimes are shorter in low metallicity environments. Contrastingly, the photoevaporation rate decreases at even lower metallicities of $Z \lesssim 10^{-1.5} \, Z_\odot $, because dust-gas collisional cooling remains efficient compared to far UV photoelectric heating whose efficiency depends on metallicity. The net cooling in the interior of the disk suppresses the photoevaporation. However, adding X-ray radiation significantly increases the photoevaporation rate, especially at $Z \sim 10^{-2}\, Z_\odot$. Although the X-ray radiation itself does not drive strong photoevaporative flows, X-rays penetrate deep into the neutral region in the disk, increase the ionization degree there, and reduce positive charges of grains. Consequently, the effect of photoelectric heating by far UV radiation is strengthened by the X-rays and enhances the disk photoevaporation., Comment: 13 pages, 8 figures

We present the first simulations of the formation and feedback of massive stars which account for radiation forces as well as photoionization feedback (along with protostellar outflows). In two different accretion scenarios modeled, we determine the relative strength of these feedback components and derive the size of the reservoir from which the forming stars gained their masses. We find that protostellar outflows alone limit the stellar mass growth only in an accretion scenario with a finite mass reservoir; when including accretion and ram pressure from large scales (> 0.1 pc), protostellar outflows do not limit stellar mass growth at all. Photoionization and HII regions dominate the feedback ladder only at later times and only on large scales. Specifically, photoionization yields a broadening of the bipolar outflow cavities and a reduction of the gravitational infall momentum by about 50 percent, but does not limit the stellar mass accretion. On the other hand, we find radiation forces restrain the gravitational infall toward the circumstellar disk, impact the gravito-centrifugal equilibrium at the outer edge of the disk, and eventually shut down stellar accretion completely. The most massive star formed in the simulations accreted 95 solar masses before disk destruction; this mass was drawn-in from an accretion reservoir of approximately 240 solar masses and 0.24 pc in radius. In the regime of very massive stars, the final mass of these stars is controlled by their own radiation force feedback., Comment: 22 pages, accepted for publication at A&A

We report ALMA Cycle 3 observations in CO isotopes toward a dense core, MC27/L1521F in Taurus, which is considered to be at an early stage of multiple star formation in a turbulent environment. Although most of the high-density parts of this core are considered to be as cold as $\sim$10 K, high-angular resolution ($\sim$20 au) observations in $^{12}$CO ($J$ = 3--2) revealed complex warm ($>$15--60 K) filamentary/clumpy structures with the sizes from a few tens of au to $\sim$1,000 au. The interferometric observations of $^{13}$CO and C$^{18}$O show that the densest part with arc-like morphologies associated with the previously identified protostar and condensations are slightly redshifted from the systemic velocity of the core. We suggest that the warm CO clouds may be consequences of shock heating induced by interactions among the different density/velocity components that originated from the turbulent motions in the core. However, such a small-scale and fast turbulent motion does not correspond to a simple extension of the line-width-size relation (i.e., Larson'{}s law), and thus the actual origin remains to be studied. The high-angular resolution CO observations are expected to be essential in detecting small-scale turbulent motions in dense cores and to investigate protostar formation therein., Comment: 17 pages, 6 figures, accepted for publication in ApJ

We conduct a theoretical study of the formation of massive stars over a wide range of metallicities from 1e-5 to 1Zsun and evaluate the star formation efficiencies (SFEs) from prestellar cloud cores taking into account multiple feedback processes. Unlike for simple spherical accretion, in the case of disk accretion feedback processes do not set upper limits on stellar masses. At solar metallicity, launching of magneto-centrifugally-driven outflows is the dominant feedback process to set SFEs, while radiation pressure, which has been regarded to be pivotal, has only minor contribution even in the formation of over-100Msun stars. Photoevaporation becomes significant in over-20Msun star formation at low metallicities of <1e-2Zsun, where dust absorption of ionizing photons is inefficient. We conclude that if initial prestellar core properties are similar, then massive stars are rarer in extremely metal-poor environments of 1e-5 - 1e-3Zsun. Our results give new insight into the high-mass end of the initial mass function and its potential variation with galactic and cosmological environments., Comment: 13 pages, 9 figures, accepted for publication in The Astrophysical Journal

Accretion on to seed black holes (BHs) is believed to play a crucial role in formation of supermassive BHs observed at high-redshift (z>6). Here, we investigate the combined effect of gas angular momentum and radiation feedback on the accretion flow, by performing 2D axially symmetric radiation hydrodynamics simulations that solve the flow structure across the Bondi radius and the outer part of the accretion disc simultaneously. The accreting gas with finite angular momentum forms a rotationally-supported disc inside the Bondi radius, where the accretion proceeds by the angular momentum transport due to assumed alpha-type viscosity. We find that the interplay of radiation and angular momentum significantly suppresses accretion even if the radiative feedback is weakened in an equatorial shadowing region. The accretion rate is O(alpha)\sim O(0.01-0.1) times the Bondi value, where alpha is the viscosity parameter. By developing an analytical model, we show that such a great reduction of the accretion rate persists unless the angular momentum is so small that the corresponding centrifugal radius is \lesssim 0.04 times the Bondi radius. We argue that BHs are hard to grow quickly via rapid mass accretion considering the angular momentum barrier presented in this paper., Comment: 17 pages, 12 figures, published in MNRAS

The driving mechanism of massive outflows observed in high-mass star-forming regions is investigated using three-dimensional magnetohydrodynamics (MHD) and protostellar evolution calculations. In our previous paper, we showed that the mass outflow rate depends strongly on the mass accretion rate onto the circumstellar disk around a high-mass protostar, and massive outflows may be driven by the magnetic effect in high-mass star-forming cores. In the present study, in order to verify that the MHD disk wind is the primary driving mechanism of massive outflows, we quantitatively compare outflow properties obtained through simulations and observations. Since the outflows obtained through simulations are slightly younger than those obtained through observations, the time-integrated quantities of outflow mass, momentum, and kinetic energy are slightly smaller than those obtained through observations. On the other hand, time-derivative quantities of mass ejection rate, outflow momentum flux, and kinetic luminosity obtained through simulations are in very good agreement with those obtained through observations. This indicates that the MHD disk wind greatly contributes to the massive outflow driving from high-mass protostars, and the magnetic field might significantly control the high-mass star formation process., Comment: 16 pages, 8 figures

Formation of supermassive stars (SMSs) with mass ~10^4 Msun is a promising pathway to seed the formation of supermassive black holes in the early universe. The so-called direct-collapse (DC) model postulates that such an SMS forms in a hot gas cloud irradiated by a nearby star-forming galaxy. We study the DC SMS formation in a fully cosmological context using three-dimensional radiation hydrodynamics simulations. We initialize our simulations using the outputs of the cosmological simulation of Chon et al. (2016), where two DC gas clouds are identified. The long-term evolution over a hundred thousand years is followed from the formation of embryo protostars through their growth to SMSs. We show that the strength of the tidal force by a nearby galaxy determines the multiplicity of the formed stars and affects the protostellar growth. In one case, where a collapsing cloud is significantly stretched by strong tidal force, multiple star-disk systems are formed via filament fragmentation. Small-scale fragmentation occurs in each circumstellar disk, and more than 10 stars with masses of a few times 10^3 Msun are finally formed. Interestingly, about a half of them are found as massive binary stars. In the other case, the gas cloud collapses nearly spherically under a relatively weak tidal field, and a single star-disk system is formed. Only a few SMSs with masses ~ 10^4 Msun are found already after evolution of a hundred thousand years, and the SMSs are expected to grow further by gas accretion and to leave massive blackholes at the end of their lives., Comment: 19 pages, 20 figures, Submitted to MNRAS

We investigate the upper stellar mass limit set by radiative feedback by the forming star with various accretion rates and metallicities. To this end, we numerically solve the structures of both a protostar and its surrounding accretion envelope assuming a spherical symmetric and steady flow. The optical depth of the dust cocoon, a dusty part of the accretion envelope, differs among the direct light from the stellar photosphere and the diffuse light re-emitted as dust thermal emission. As a result, varying the metallicity qualitatively changes the way that the radiative feedback suppresses the accretion flow. With a fixed accretion rate of $10^{-3}M_{\odot} {\rm yr^{-1}}$, the both direct and diffuse lights jointly operate to prevent the mass accretion at $Z \gtrsim 10^{-1}Z_{\odot}$. At $Z \lesssim 10^{-1}Z_{\odot}$, the diffuse light is no longer effective, and the direct light solely limits the mass accretion. At $Z \lesssim 10^{-3}Z_{\odot}$, the HII region formation plays an important role in terminating the accretion. The resultant upper mass limit increases with decreasing metallicity, from a few $\times~10~M_\odot$ to $\sim 10^3~M_\odot$ over $Z = 1Z_{\odot} - 10^{-4}~Z_\odot$. We also illustrate how the radiation spectrum of massive star-forming cores changes with decreasing metallicity. First, the peak wavelength of the spectrum, which is located around $30 \mu {\rm m}$ at $1Z_{\odot}$, shifts to $< 3 \mu {\rm m}$ at $Z \lesssim 0.1 Z_\odot$. Second, a characteristic feature at $10 \mu \rm{m}$ due to the amorphous silicate band appears as a dip at $1Z_{\odot}$, but changes to a bump at $Z \lesssim 0.1 Z_{\odot}$. Using these spectral signatures, we can search massive accreting protostars in nearby low-metallicity environments with up-coming observations., Comment: 21 pages, 20 figures, accepted for publication in MNRAS

The origin of super-massive black holes in the early universe remains poorly understood.Gravitational collapse of a massive primordial gas cloud is a promising initial process,but theoretical studies have difficulty growing the black hole fast enough.We report numerical simulations of early black hole formation starting from realistic cosmological conditions.Supersonic gas motions left over from the Big Bang prevent early gas cloud formation until rapid gas condensation is triggered in a proto-galactic halo. A protostar is formed in the dense, turbulent gas cloud, and it grows by sporadic mass accretion until it acquires 34,000 solar masses.The massive star ends its life with a catastrophic collapse to leave a black hole -- a promising seed for the formation of a monstrous black hole., Comment: Published in Science, combined with updated SOM, additional images and movies are available at http://www-utap.phys.s.u-tokyo.ac.jp/naoki.yoshida/Blackhole/0929e.html

We report ALMA observations in 0.87 mm continuum and $^{12}$CO ($J$ = 3--2) toward a very low-luminosity ($<$0.1 $L_{\odot}$) protostar, which is deeply embedded in one of the densest core MC27/L1521F, in Taurus with an indication of multiple star formation in a highly dynamical environment. The beam size corresponds to $\sim$20 AU, and we have clearly detected blueshifted/redshifted gas in $^{12}$CO associated with the protostar. The spatial/velocity distributions of the gas show there is a rotating disk with a size scale of $\sim$10 AU, a disk mass of $\sim$10$^{-4}$ $M_{\odot}$ and a central stellar mass of $\sim$0.2 $M_{\odot}$. The observed disk seems to be detached from the surrounding dense gas, although it is still embedded at the center of the core whose density is $\sim$10$^{6}$ cm$^{-3}$. The current low-outflow activity and the very low luminosity indicate that the mass accretion rate onto the protostar is extremely low in spite of a very early stage of star formation. We may be witnessing the final stage of the formation of $\sim$0.2 $M_{\odot}$ protostar. However, we cannot explain the observed low luminosity with the standard pre-main-sequence evolutionary track unless we assume cold accretion with an extremely small initial radius of the protostar ($\sim$0.65 $R_\odot$). These facts may challenge our current understanding of the low mass star formation, in particular the mass accretion process onto the protostar and the circumstellar disk., Comment: 9 pages, 4 figures, accepted for publication in The Astrophysical Journal

Protoplanetary disks are thought to have lifetimes of several million years in the solar neighborhood, but recent observations suggest that the disk lifetimes are shorter in a low metallicity environment. We perform a suite of radiation hydrodynamics simulations of photoevaporation of protoplanetary disks to study the disk structure and its long-term evolution of $\sim 10000$ years, and the metallicity dependence of mass-loss rate. Our simulations follow hydrodynamics, extreme and far ultra-violet radiative transfer, and non-equilibrium chemistry in a self-consistent manner. Dust grain temperatures are also calculated consistently by solving the radiative transfer of the stellar irradiation and grain (re-)emission. We vary the disk gas metallicity over a wide range of $10^{-4}~ Z_\odot \leq Z \leq 10 ~Z_\odot$. The photoevaporation rate is lower with higher metallicity in the range of $10^{-1} \,Z_\odot \lesssim Z \lesssim 10 \,Z_\odot$, because dust shielding effectively prevents far-ultra violet (FUV) photons from penetrating into and heating the dense regions of the disk. The photoevaporation rate sharply declines at even lower metallicities in $10^{-2} \,Z_\odot \lesssim Z \lesssim 10^{-1}\,Z_\odot$, because FUV photoelectric heating becomes less effective than dust-gas collisional cooling. The temperature in the neutral region decreases, and photoevaporative flows are excited only in an outer region of the disk. At $10^{-4}\,Z_\odot \leq Z \lesssim 10^{-2}\,Z_\odot$, HI photoionization heating acts as a dominant gas heating process and drives photoevaporative flows with roughly a constant rate. The typical disk lifetime is shorter at $Z=0.3~Z_\odot$ than at $Z = Z_\odot$, being consistent with recent observations of the extreme outer galaxy., Comment: 25 pages, 17 figures, to appear in ApJ

The pre-main-sequence evolution of low-mass stars and brown dwarfs is studied numerically starting from the formation of a protostellar/proto-brown dwarf seed and taking into account the mass accretion onto the central object during the initial several Myr of evolution. The stellar evolution was computed using the STELLAR evolution code developed by Yorke & Bodenheimer with recent modifications by Hosokawa et al. The mass accretion rates were taken from numerical hydrodynamics models of Vorobyov & Basu computing the circumstellar disk evolution starting from the gravitational collapse of pre-stellar cloud cores of various mass and angular momentum. The resulting stellar evolution tracks were compared with the isochrones and isomasses calculated using non-accreting models. We find that mass accretion in the initial several Myr of protostellar evolution can have a strong effect on the subsequent evolution of young stars and brown dwarfs. The disagreement between accreting and non-accreting models in terms of the total stellar luminosity L_st, stellar radius R_st and effective temperature T_eff depends on the thermal efficiency of accretion, i.e., on the fraction of accretion energy absorbed by the central object. The largest mismatch is found for the cold accretion case, in which essentially all accretion energy is radiated away. The relative deviations in L_st and R_st in this case can reach 50% for 1.0-Myr-old objects and remain notable even for 10-Myr-old objects. In the hot and hybrid accretion cases, in which a constant fraction of accretion energy is absorbed, the disagreement between accreting and non-accreting models becomes less pronounced, but still remains notable for 1.0-Myr-old objects. These disagreements may lead to the wrong age estimate for objects of (sub-)solar mass when using the isochrones based on non-accreting models (abridged)., Comment: 17 pages, 9 figures, accepted for publication by Astronomy & Astrophysics