The early evolution of protostellar disks with metallicities in the $Z=1.0-0.01~Z_\odot$ range was studied with a particular emphasis on the strength of gravitational instability and the nature of protostellar accretion in low-metallicity systems. Numerical hydrodynamics simulations in the thin-disk limit were employed that feature separate gas and dust temperatures, and disk mass-loading from the infalling parental cloud cores. Models with cloud cores of similar initial mass and rotation pattern, but distinct metallicity were considered to distinguish the effect of metallicity from that of initial conditions. The early stages of disk evolution in low-metallicity models are characterized by vigorous gravitational instability and fragmentation. Disk instability is sustained by continual mass-loading from the collapsing core. The time period that is covered by this unstable stage is much shorter in the $Z=0.01~Z_\odot$ models as compared to their higher metallicity counterparts thanks to the higher mass infall rates caused by higher gas temperatures (that decouple from lower dust temperatures) in the inner parts of collapsing cores. Protostellar accretion rates are highly variable in the low-metallicity models reflecting a highly dynamical nature of the corresponding protostellar disks. The low-metallicity systems feature short, but energetic episodes of mass accretion caused by infall of inward-migrating gaseous clumps that form via gravitational fragmentation of protostellar disks. These bursts seem to be more numerous and last longer in the $Z=0.1~Z_\odot$ models in comparison to the $Z=0.01~Z_\odot$ case. Variable protostellar accretion with episodic bursts is not a particular feature of solar metallicity disks. It is also inherent to gravitationally unstable disks with metallicities up to 100 times lower than solar.

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.

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 $Z=1~Z_{\odot}$, for instance, the SFE increases from 0.02 to 0.3 with $\Sigma$ from $10$ to $300~{\rm M_{\odot} {\rm pc^{-2}}}$. 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 HII 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 is reduced by a factor of $\sim 0.3$ for the cases with $Z=10^{-2}~Z_{\odot}$ 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.

Aims: We explore the long-term evolution of young protoplanetary disks with different approaches to computing the thermal structure determined by various cooling and heating processes in the disk and its surroundings. Methods: Numerical hydrodynamics simulations in the thin-disk limit were complemented with three thermal evolution schemes: a simplified $\beta$-cooling approach with and without irradiation, in which the rate of disk cooling is proportional to the local dynamical time, a fiducial model with equal dust and gas temperatures calculated taking viscous heating, irradiation, and radiative cooling into account, and also a more sophisticated approach allowing decoupled dust and gas temperatures. Results: We found that the gas temperature may significantly exceed that of dust in the outer regions of young disks thanks to additional compressional heating caused by the infalling envelope material in the early stages of disk evolution and slow collisional exchange of energy between gas and dust in low-density disk regions. The outer envelope however shows an inverse trend with the gas temperatures dropping below that of dust. The global disk evolution is only weakly sensitive to temperature decoupling. Nevertheless, separate dust and gas temperatures may affect the chemical composition, dust evolution, and disk mass estimates. Constant-$\beta$ models without stellar and background irradiation fail to reproduce the disk evolution with more sophisticated thermal schemes because of intrinsically variable nature of the $\beta$-parameter. Constant-$\beta$ models with irradiation can better match the dynamical and thermal evolution, but the agreement is still incomplete. Conclusions: Models allowing separate dust and gas temperatures are needed when emphasis is placed on the chemical or dust evolution in protoplanetary disks, particularly in sub-solar metallicity environments.

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.

Direct collapse black hole (DCBH) formation with mass $\gtrsim 10^{5}~M_{\odot}$ is a promising scenario for the origin of high-redshift supermassive black holes. It has usually been supposed that the DCBH can only form in the primordial gas since the metal enrichment enhances the cooling ability and causes the fragmentation into smaller pieces. What actually happens in such an environment, however, has not been explored in detail. Here, we study the impact of the metal enrichment on the clouds, conducting hydrodynamical simulations to follow the cloud evolution in cases with different degree of the metal enrichment $Z/Z_{\odot}=10^{-6}-10^{-3}$. Below $Z/Z_{\odot}=10^{-6}$, metallicity has no effect and supermassive stars form along with a small number of low-mass stars. With more metallicity $Z/Z_{\odot} \gtrsim 5 \times 10^{-6}$, although the dust cooling indeed promotes fragmentation of the cloud core and produces about a few thousand low-mass stars, the accreting flow preferentially feeds the gas to the central massive stars, which grows supermassive as in the primordial case. We term this formation mode as the {\it super competitive accretion}, where only the central few stars grow supermassive while a large number of other stars are competing for the gas reservoir. Once the metallicity exceeds $10^{-3}~Z_{\odot}$ and metal-line cooling becomes operative, the central star cannot grow supermassive due to lowered accretion rate. Supermassive star formation by the super competitive accretion opens up a new window for seed BHs, which relaxes the condition on metallicity and enhances the seed BH abundance.

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.

Magnetic fields play such roles in star formation as the angular momentum transport in star-forming clouds, thereby controlling circumstellar disc formation and even binary star formation efficiency. The coupling between the magnetic field and gas is determined by the ionization degree in the gas. Here, we calculate the thermal and chemical evolution of the primordial gas by solving chemical reaction network where all the reactions are reversed. We find that at ~ 10^14-10^18 /cm^3, the ionization degree becomes 100-1000 times higher than the previous results due to the lithium ionization by thermal photons trapped in the cloud, which has been omitted so far. We construct the minimal chemical network which can reproduce correctly the ionization degree as well as the thermal evolution by extracting 36 reactions among 13 species. Using the obtained ionization degree, we evaluate the magnetic field diffusivity. We find that the field dissipation can be neglected for global fields coherent over > a tenth of the cloud size as long as the field is not so strong as to prohibit the collapse. With magnetic fields strong enough for ambipolar diffusion heating to be significant, the magnetic pressure effects to slow down the collapse and to reduce the compressional heating become more important, and the temperature actually becomes lower than in the no-field case.

Supermassive stars (SMSs) with mass $\sim10^{5}~\rm{M}_{\odot}$ are promising candidates for the origin of supermassive black holes observed at redshift $\gtrsim6$. They are supposed to form as a result of rapid accretion of primordial gas, although it can be obstructed by the time variation caused by circum-stellar disc fragmentation due to gravitational instability. To assess the occurrence of fragmentation, we study the structure of marginally gravitationally unstable accretion discs, by using a steady one-dimensional thin disc model with detailed treatment of chemical and thermal processes. Motivated by two SMS formation scenarios, i.e., those with strong ultraviolet radiation background or with large velocity difference between the baryon and the dark matter, we consider two types of flows, i.e., atomic and molecular flows, respectively, for a wide range of the central stellar mass $10-10^5~\rm{M}_{\odot}$ and the accretion rate $10^{-3}-1~\rm{M}_{\odot}~\rm{yr}^{-1}$. In the case of a mostly atomic gas flowing to the disc outer boundary, the fragmentation condition is expressed as the accretion rate being higher than the critical value of $10^{-1}~\rm{M}_{\odot}~\rm{yr}^{-1}$ regardless of the central stellar mass. On the other hand, in the case of molecular flows, there is a critical disc radius outside of which the disc becomes unstable. Those conditions appears to be marginally satisfied according to numerical simulations, suggesting that disc fragmentation can be common during SMS formation.

With the recent discovery of a dozen dusty star-forming galaxies and around 30 quasars at z>5 that are hyper-luminous in the infrared ($\mu$$L_{\rm IR}>10^{13}$ L$_{\odot}$, where $\mu$ is a lensing magnification factor), the possibility has opened up for SPICA, the proposed ESA M5 mid-/far-infrared mission, to extend its spectroscopic studies toward the epoch of reionization and beyond. In this paper, we examine the feasibility and scientific potential of such observations with SPICA's far-infrared spectrometer SAFARI, which will probe a spectral range (35-230 $\mu$m) that will be unexplored by ALMA and JWST. Our simulations show that SAFARI is capable of delivering good-quality spectra for hyper-luminous infrared galaxies (HyLIRGs) at z=5-10, allowing us to sample spectral features in the rest-frame mid-infrared and to investigate a host of key scientific issues, such as the relative importance of star formation versus AGN, the hardness of the radiation field, the level of chemical enrichment, and the properties of the molecular gas. From a broader perspective, SAFARI offers the potential to open up a new frontier in the study of the early Universe, providing access to uniquely powerful spectral features for probing first-generation objects, such as the key cooling lines of low-metallicity or metal-free forming galaxies (fine-structure and H2 lines) and emission features of solid compounds freshly synthesized by Population III supernovae. Ultimately, SAFARI's ability to explore the high-redshift Universe will be determined by the availability of sufficiently bright targets (whether intrinsically luminous or gravitationally lensed). With its launch expected around 2030, SPICA is ideally positioned to take full advantage of upcoming wide-field surveys such as LSST, SKA, Euclid, and WFIRST, which are likely to provide extraordinary targets for SAFARI.

Discovery of high-redshift (z > 6) supermassive black holes (BHs) may indicate that the rapid (or super-Eddington) gas accretion has aided their quick growth. Here, we study such rapid accretion of the primordial gas on to intermediate-mass (10^2 - 10^5 M_sun) BHs under anisotropic radiation feedback. We perform two-dimensional radiation hydrodynamics simulations that solve the flow structure across the Bondi radius, from far outside of the Bondi radius down to a central part which is larger than a circum-BH accretion disc. The radiation from the unresolved circum-BH disc is analytically modeled considering self-shadowing effect. We show that the flow settles into a steady state, where the flow structure consists of two distinct parts: (1) bipolar ionized outflowing regions, where the gas is pushed outward by thermal gas pressure and super-Eddington radiation pressure, and (2) an equatorial neutral inflowing region, where the gas falls toward the central BH without affected by radiation feedback. The resulting accretion rate is much higher than that in the case of isotropic radiation, far exceeding the Eddington-limited rate to reach a value slightly lower than the Bondi one. The opening angle of the equatorial inflowing region is determined by the luminosity and directional dependence of the central radiation. We find that photoevaporation from its surfaces set the critical opening angle of about ten degrees below which the accretion to the BH is quenched. We suggest that the shadowing effect allows even stellar-remnant BHs to grow rapidly enough to become high-redshift supermassive BHs.

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.

Shocks may have been prevalent in the early Universe, associated with virialization and supernova explosions, etc. Here, we study thermal evolution and fragmentation of shock-compressed clouds, by using a one-zone model with detailed thermal and chemical processes. We explore a large range of initial density (1-1e5 /cm^3), metallicity (0-1e-2 Z_sun), UV strength (0-500 times Galactic value), and cosmic microwave background temperature (10 and 30 K). Shock-compressed clouds contract isobarically via atomic and molecular line cooling, until self-gravitating clumps are formed by fragmentation. If the metals are only in the gas-phase, the clump mass is higher than ~ 3 M_sun in any conditions we studied. Although in some cases with a metallicity higher than ~ 1e-3 Z_sun, re-fragmentation of a clump is caused by metal-line cooling, this fragment mass is higher than ~ 30 M_sun. On the other hand, if about half the mass of metals is condensed in dust grains, as in the Galactic interstellar medium, dust cooling triggers re-fragmentation of a clump into sub-solar mass pieces, for metallicities higher than ~ 1e-5 Z_sun. Therefore, the presence of dust is essential in low-mass (< M_sun) star formation from a shock-compressed cloud.

Dust enables low-mass stars to form from low-metallicity gas by inducing fragmentation of clouds via the cooling by its thermal emission. Dust may, however, be evacuated from star-forming clouds due to radiation force from massive stars. We here study the condition for the dust evacuation by comparing the dust evacuation time with the time of cloud destruction due to either expansion of HII regions or supernovae. The cloud destruction time has weak dependence on the cloud radius, while the dust evacuation time becomes shorter for a cloud with the smaller radius. The dust evacuation thus occurs in compact star-forming clouds whose column density is $N_{\rm H} \simeq 10^{24} - 10^{26} ~{\rm cm^{-2}}$. The critical halo mass above which the dust evacuation occurs becomes lower for higher formation redshift, e.g., $\sim 10^{9}~M_{\odot}$ at redshift $z \sim 3$ and $\sim 10^{7}~M_{\odot}$ at $z \sim 9$. In addition, metallicity of the gas should be less than $\sim 10^{-2} ~ Z_{\odot}$. Otherwise the dust attenuation reduces the radiation force significantly. From the dust-evacuated gas, massive stars are likely to form even with metallicity above $\sim 10^{-5}~Z_{\odot}$, the critical value for low-mass star formation due to the dust cooling. This can explain the dearth of ultra-metal poor stars with the metallicity lower than $\sim 10^{-4}~Z_{\odot}$.

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.

We consider whether the maximum mass of first stars is imposed by the protostellar spin, i.e., by the so-called $\Omega\Gamma$-limit, which requires the sum of the radiation and centrifugal forces at the stellar surface be smaller than the inward pull of the gravity. Once the accreting protostar reaches such a marginal state, the star cannot spin up more and is not allowed to accrete more gas with inward angular momentum flux. So far, however, the effect of stellar radiation on the structure of the accretion disk has not been properly taken into account in discussing the effect of $\Omega\Gamma$-limit on the first star formation. Here, we obtain a series of the steady accretion-disk solutions considering such effect and find solutions without net angular momentum influx to the stars with arbitrary rotation rates, in addition to those with finite angular momentum flows. The accretion of positive angular momentum flows pushes the star beyond the $\Omega\Gamma$-limit, which is allowed only with the external pressure provided by the circumstellar disk. On the other hand, the accretion with no net angular momentum influx does not result in the spin-up of the star. Thus, the existence of the solution with no net angular momentum influx indicates that the protostars can keep growing in mass by accretion even after they reach the $\Omega\Gamma$-limit.

Fragmentation of filaments into dense cores is thought to be an important step in forming stars. The bar-mode instability of spherically collapsing cores found in previous linear analysis invokes a possibility of re-fragmentation of the cores due to their ellipsoidal (prolate or oblate) deformation. To investigate this possibility, here we perform three-dimensional self-gravitational hydrodynamics simulations that follow all the way from filament fragmentation to subsequent core collapse. We assume the gas is polytropic with index \gamma, which determines the stability of the bar-mode. For the case that the fragmentation of isolated hydrostatic filaments is triggered by the most unstable fragmentation mode, we find the bar mode grows as collapse proceeds if \gamma < 1.1, in agreement with the linear analysis. However, it takes more than ten orders-of-magnitude increase in the central density for the distortion to become non-linear. In addition to this fiducial case, we also study non-fiducial ones such as the fragmentation is triggered by a fragmentation mode with a longer wavelength and it occurs during radial collapse of filaments and find the distortion rapidly grows. In most of astrophysical applications, the effective polytropic index of collapsing gas exceeds 1.1 before ten orders-of-magnitude increase in the central density. Thus, supposing the fiducial case of filament fragmentation, re-fragmentation of dense cores would not be likely and their final mass would be determined when the filaments fragment.

Direct collapse black holes (DCBHs) are excellent candidates as seeds of supermassive black holes (SMBHs) observed at $z \gsim 6$. The formation of a DCBH requires a strong external radiation field to suppress $\rm H_2$ formation and cooling in a collapsing gas cloud. Such strong field is not easily achieved by first stars or normal star-forming galaxies. Here we investigate a scenario in which the previously-formed DCBH can provide the necessary radiation field for the formation of additional ones. Using one-zone model and the simulated DCBH Spectral Energy Distributions (SEDs) filtered through absorbing gas initially having column density $N_{\rm H}$, we derive the critical field intensity, $J_{\rm LW}^{\rm crit}$, to suppress $\rm H_2$ formation and cooling. For the SED model with $N_{\rm H}=1.3\times10^{25}$ cm$^{-2}$, $8.0\times10^{24}$ cm$^{-2}$ and $5.0\times10^{24}$ cm$^{-2}$, we obtain $J_{\rm LW}^{\rm crit}\approx22$, 35 and 54, all much smaller than the critical field intensity for normal star-forming galaxies ($J_{\rm LW}^{\rm crit}\simgt 1000$). X-ray photons from previously-formed DCBHs build up a high-$z$ X-ray background (XRB) that may boost the $J_{\rm LW}^{\rm crit}$. However, we find that in the three SED models $J_{\rm LW}^{\rm crit}$ only increases to $\approx80$, 170 and 390 respectively even when $\dt{\rho}_\bullet$ reaches the maximum value allowed by the present-day XRB level ($0.22, 0.034, 0.006~M_\odot$yr$^{-1}$Mpc$^{-3}$), still much smaller than the galactic value. Although considering the XRB from first galaxies may further increase $J_{\rm LW}^{\rm crit}$, we conclude that our investigation supports a scenario in which DCBH may be more abundant than predicted by models only including galaxies as external radiation sources.

Supermassive stars (SMS; ~ 10^5 M_sun) formed from metal-free gas in the early Universe attract attention as progenitors of supermassive black holes observed at high redshifts. To form SMSs by accretion, central protostars must accrete at as high rates as ~ 0.1-1 M_sun/yr. Such protostars have very extended structures with bloated envelopes, like super-giant stars, and are called super-giant protostars (SGPSs). Under the assumption of hydrostatic equilibrium, SGPSs have density inverted layers, where the luminosity becomes locally super-Eddington, near the surface. If the envelope matter is allowed to flow out, however, a stellar wind could be launched and hinder the accretion growth of SGPSs before reaching the supermassive regime. We examine whether radiation-driven winds are launched from SGPSs by constructing steady and spherically symmetric wind solutions. We find that the wind velocity does not reach the escape velocity in any case considered. This is because once the temperature falls below ~ 10^4 K, the opacity plummet drastically owing to the recombination of hydrogen and the acceleration ceases suddenly. This indicates that, in realistic non-steady cases, even if outflows are launched from the surface of SGPSs, they would fall back again. Such a "wind" does not result in net mass loss and does not prevent the growth of SGPSs. In conclusion, SGPSs will grow to SMSs and eventually collapse to massive BHs of ~ 10^5 M_sun, as long as the rapid accretion is maintained.

We study the impact of star-forming mini-haloes, and the Initial Mass Function (IMF) of Population III (Pop III) stars, on the Galactic halo Metallicity Distribution Function (MDF) and on the properties of C-enhanced and C-normal stars at [Fe/H]<-3. For our investigation we use a data-constrained merger tree model for the Milky Way formation, which has been improved to self-consistently describe the physical processes regulating star-formation in mini-haloes, including the poor sampling of the Pop III IMF. We find that only when star-forming mini-haloes are included the low-Fe tail of the MDF is correctly reproduced, showing a plateau that is built up by C-enhanced metal-poor (CEMP) stars imprinted by primordial faint supernovae. The incomplete sampling of the Pop III IMF in inefficiently star-forming mini-haloes (< $10^{-3}$ $M_\odot$/yr) strongly limits the formation of Pair Instability Supernovae (PISNe), with progenitor masses $m_{\rm popIII}$=[140-260] $M_\odot$, even when a flat Pop III IMF is assumed. Second-generation stars formed in environments polluted at >50% level by PISNe are thus extremely rare, corresponding to $\approx$ 0.25% of the total stellar population at [Fe/H]<-2, which is consistent with recent observations. The low-Fe tail of the MDF strongly depends on the Pop III IMF shape and mass range. Given the current statistics, we find that a flat Pop III IMF model with $m_{\rm popIII}$=[10-300] $M_\odot$ is disfavoured by observations. We present testable predictions for Pop III stars extending down to lower masses, with $m_{\rm popIII}$=[0.1-300] $M_\odot$.

The formation of supermassive stars (SMSs) via rapid mass accretion and their direct collapse into black holes (BHs) is a promising pathway for sowing seeds of supermassive BHs in the early universe. We calculate the evolution of rapidly accreting SMSs by solving the stellar structure equations including nuclear burning as well as general relativistic (GR) effects up to the onset of the collapse. We find that such SMSs have less concentrated structure than fully-convective counterpart, which is often postulated for non-accreting ones. This effect stabilizes the stars against GR instability even above the classical upper mass limit $\gtrsim 10^5~M_\odot$ derived for the fully-convective stars. The accreting SMS begins to collapse at the higher mass with the higher accretion rate. The collapse occurs when the nuclear fuel is exhausted only for cases with $\dot M \lesssim 0.1~M_\odot~{\rm yr}^{-1}$. With $\dot{M} \simeq 0.3 - 1~M_\odot~{\rm yr}^{-1}$, the star becomes GR-unstable during the helium-burning stage at $M \simeq 2 - 3.5~\times 10^5~M_\odot$. In an extreme case with $10~M_\odot~{\rm yr}^{-1}$, the star does not collapse until the mass reaches $\simeq 8.0\times 10^5~M_\odot$, where it is still in the hydrogen-burning stage. We expect that BHs with roughly the same mass will be left behind after the collapse in all the cases.

We present coupled stellar evolution (SE) and 3D radiation-hydrodynamic (RHD) simulations of the evolution of primordial protostars, their immediate environment, and the dynamic accretion history under the influence of stellar ionizing and dissociating UV feedback. Our coupled SE-RHD calculations result in a wide diversity of final stellar masses covering 10 Msun $\lesssim M_* \lesssim$ 1000 Msun. The formation of very massive ($\gtrsim$ 250 Msun) stars is possible under weak UV feedback, whereas ordinary massive (a few x 10 Msun) stars form when UV feedback can efficiently halt the accretion. This may explain the peculiar abundance pattern of a Galactic metal-poor star recently reported by Aoki et al. (2014), possibly the observational signature of very massive precursor primordial stars. Weak UV feedback occurs in cases of variable accretion, in particular when repeated short accretion bursts temporarily exceed 0.01 Msun/yr, causing the protostar to inflate. In the bloated state, the protostar has low surface temperature and UV feedback is suppressed until the star eventually contracts, on a thermal adjustment timescale, to create an HII region. If the delay time between successive accretion bursts is sufficiently short, the protostar remains bloated for extended periods, initiating at most only short periods of UV feedback. Disk fragmentation does not necessarily reduce the final stellar mass. Quite the contrary, we find that disk fragmentation enhances episodic accretion as many fragments migrate inward and are accreted onto the star, thus allowing continued stellar mass growth under conditions of intermittent UV feedback. This trend becomes more prominent as we improve the resolution of our simulations. We argue that simulations with significantly higher resolution than reported previously are needed to derive accurate gas mass accretion rates onto primordial protostars.

Observations of quasars at $ z > 6$ suggest the presence of black holes with a few times $\rm 10^9 ~M_{\odot}$. Numerous models have been proposed to explain their existence including the direct collapse which provides massive seeds of $\rm 10^5~M_{\odot}$. The isothermal direct collapse requires a strong Lyman-Werner flux to quench $\rm H_2$ formation in massive primordial halos. In this study, we explore the impact of trace amounts of metals and dust enrichment. We perform three dimensional cosmological simulations for two halos of $\rm > 10^7~M_{\odot}$ with $\rm Z/Z_{\odot}= 10^{-4}-10^{-6}$ illuminated by an intense Lyman Werner flux of $\rm J_{21}=10^5$. Our results show that initially the collapse proceeds isothermally with $\rm T \sim 8000$ K but dust cooling becomes effective at densities of $\rm 10^{8}-10^{12} ~cm^{-3}$ and brings the gas temperature down to a few 100-1000 K for $\rm Z/Z_{\odot} \geq 10^{-6}$. No gravitationally bound clumps are found in $\rm Z/Z_{\odot} \leq 10^{-5}$ cases by the end of our simulations in contrast to the case with $\rm Z/Z_{\odot} = 10^{-4}$. Large inflow rates of $\rm \geq 0.1~M_{\odot}/yr$ are observed for $\rm Z/Z_{\odot} \leq 10^{-5}$ similar to a zero-metallicity case while for $\rm Z/Z_{\odot} = 10^{-4}$ the inflow rate starts to decline earlier due to the dust cooling and fragmentation. For given large inflow rates a central star of $\rm \sim 10^4~M_{\odot}$ may form for $\rm Z/Z_{\odot} \leq 10^{-5}$.

The growth of the first super massive black holes (SMBHs) at z > 6 is still a major challenge for theoretical models. If it starts from black hole (BH) remnants of Population III stars (light seeds with mass ~ 100 Msun) it requires super-Eddington accretion. An alternative route is to start from heavy seeds formed by the direct collapse of gas onto a ~ 10^5 Msun BH. Here we investigate the relative role of light and heavy seeds as BH progenitors of the first SMBHs. We use the cosmological, data constrained semi-analytic model GAMETE/QSOdust to simulate several independent merger histories of z > 6 quasars. Using physically motivated prescriptions to form light and heavy seeds in the progenitor galaxies, we find that the formation of a few heavy seeds (between 3 and 30 in our reference model) enables the Eddington-limited growth of SMBHs at z > 6. This conclusion depends sensitively on the interplay between chemical, radiative and mechanical feedback effects, which easily erase the conditions that allow the suppression of gas cooling in the low metallicity gas (Z < Zcr and JLW > Jcr). We find that heavy seeds can not form if dust cooling triggers gas fragmentation above a critical dust-to-gas mass ratio (D > Dcr). In addition, the relative importance of light and heavy seeds depends on the adopted mass range for light seeds, as this dramatically affects the history of cold gas along the merger tree, by both SN and AGN-driven winds.

The formation of SMSs is a potential pathway to seed SMBHs in the early universe. A critical issue for forming SMSs is stellar UV feedback, which may limit the stellar mass growth via accretion. In this paper we study the evolution of an accreting SMS and its UV emissivity under conditions of realistic variable accretion from a self-gravitating circumstellar disc. First we conduct a 2D hydrodynamical simulation to follow the long-term protostellar accretion until the stellar mass exceeds $10^4~M_\odot$. The disc fragments due to gravitational instability, creating a number of small clumps that rapidly migrate inward to fall onto the star. The resulting accretion history is thus highly time-dependent: short episodic accretion bursts are followed by longer, relative quiescent phases. We show that the circumstellar disc for the so-called direct collapse model is more unstable and generates greater variability over shorter timescales than normal Pop III cases. We conduct a post-process stellar evolution calculation using the obtained accretion history. Our results show that, regardless of the strong variability of the accretion rates, the stellar radius monotonically increases with almost constant effective temperature at $T_{\rm eff} \simeq 5000$ K as the stellar mass increases. The resulting UV feedback is too weak to hinder mass accretion due to the low flux of stellar UV photons, thus verifying our implicit assumption of no stellar feedback during the hydrodynamic simulations. The insensitivity of stellar evolution to variable accretion is attributed to the fact that typical timescales of variability, $\lesssim 10^3$ years, are too short to affect the stellar structure. We argue that this evolution will continue until the SMS eventually collapses to produce a massive black hole by the general relativistic instability after the stellar mass reaches $\gtrsim 10^5~M_\odot$.

We investigate the role of the H_2^+ channel on H_2 molecule formation during the collapse of primordial gas clouds immersed in strong radiation fields which are assumed to have the shape of a diluted black-body spectra with temperature T_rad. Since the photodissociation rate of H_2^+ depends on its level population, we take full account of the vibrationally-resolved H_2^+ kinetics. We find that in clouds under soft but intense radiation fields with spectral temperature T_rad < 7000 K, the H_2^+ channel is the dominant H_2 formation process. On the other hand, for harder spectra with T_rad > 7000 K, the H^- channel takes over H_2^+ in the production of molecular hydrogen. We calculate the critical radiation intensity needed for supermassive star formation by direct collapse and examine its dependence on the H_2^+ level population. Under the assumption of local thermodynamic equilibrium (LTE) level population, the critical intensity is underestimated by a factor of a few for soft spectra with T_rad < 7000 K. For harder spectra, the value of the critical intensity is not affected by the level population of H_2^+. This result justifies previous estimates of the critical intensity assuming LTE populations since radiation sources like young and/or metal-poor galaxies are predicted to have rather hard spectra.

We perform a large set of cosmological simulations of early structure formation and follow the formation and evolution of 1540 star-forming gas clouds to derive the mass distribution of primordial stars. The star formation in our cosmological simulations is characterized by two distinct populations, the so-called Population III.1 stars and primordial stars formed under the influence of far ultraviolet (FUV) radiation (Population III.2D stars). In this work, we determine the stellar masses by using the dependences on the physical properties of star-forming cloud and/or the external photodissociating intensity from nearby primordial stars, which are derived from the results of two-dimensional radiation hydrodynamic simulations of protostellar feedback. The characteristic mass of the Pop III stars is found to be a few hundred solar masses at z ~ 25, and it gradually shifts to lower masses with decreasing redshift. At high redshifts z > 20, about half of the star-forming gas clouds are exposed to intense FUV radiation and thus give birth to massive Pop III.2D stars. However, the local FUV radiation by nearby Pop III stars becomes weaker at lower redshifts, when typical Pop III stars have smaller masses and the mean physical separation between the stars becomes large owing to cosmic expansion. Therefore, at z < 20, a large fraction of the primordial gas clouds host Pop III.1 stars. At z =< 15, the Pop III.1 stars are formed in relatively cool gas clouds due to efficient radiative cooling by H_2 and HD molecules; such stars have masses of a few x 10 Msun. Since the stellar evolution and the final fate are determined by the stellar mass, Pop III stars formed at different epochs play different roles in the early universe.

We study dissipation process of magnetic fields in the metallicity range $0-1 Z_{\odot}$ for contracting prestellar cloud cores. By solving non-equilibrium chemistry for important charged species including charged grains, we evaluate the drift velocity of the magnetic-field lines with respect to the gas. We find that the magnetic flux dissipates in the density range $10^{12}{\rm cm^{-3}} \lesssim n_{\rm H} \lesssim 10^{17}{\rm cm^{-3}}$ for the solar-metallicity case at the scale of the core, which is assumed to be the Jeans scale. The dissipation density range becomes narrower for lower metallicity. The magnetic field is always frozen to the gas below metallicity $\lesssim 10^{-7}-10^{-6}Z_\odot$, depending on the ionization rate by cosmic rays and/or radioactivity. With the same metallicity, the dissipation density range becomes wider for lower ionization rate. The presence of such a dissipative regime is expected to cause various dynamical phenomena in protostellar evolution such as the suppression of jet/outflow launching and fragmentation of the circumstellar disks depending on the metallicity.

We examine the linear stability and nonlinear growth of the thermal instability in isobarically contracting gas with various metallicities and FUV field strengths. When the H2 cooling is suppressed by FUV fields (G_0>10^-3) or the metallicity is high enough (Z/Zs>10^-3), the interstellar medium is thermally unstable in the temperature range 100-7000 K owing to the cooling by CII and OI fine-structure lines. In this case, a bi-phasic medium with a bimodal density probability distribution is formed as a consequence of the thermal instability. The characteristic scales of the thermal instability become smaller with increasing metallicity. Comparisons of the nonlinear simulations with different resolution indicates that the maximum scale of the thermal instability should be resolved with more than 60 cells to follow runaway cooling driven by the thermal instability. Under sufficiently weak FUV fields and with low metallcity, the density range of the thermal instability shrinks owing to dominance of H2 cooling. As the FUV intensity is reduced, bi-phasic structure becomes less remarkable and disappears eventually. Our basic results suggest that, in early galaxies, i) the thermal instability has little effect for the medium with Z/Zs<10^-4, ii) fragmentation by the thermal instability could determine mass spectrum of star clusters for 10^-4<Z/Zs<0.04, and iii) thermally bistable turbulent interstellar medium like our galaxy becomes ubiquitous for Z/Zs>0.04, although the threshold metallicity depends on the conditions such as thermal pressure, FUV strength and redshift.

We investigate the condition for the formation of low-mass second-generation stars in the early universe. It has been proposed that gas cooling by dust thermal emission can trigger fragmentation of a low-metallicity star-forming gas cloud. In order to determine the critical condition in which dust cooling induces the formation of low-mass stars, we follow the thermal evolution of a collapsing cloud by a one-zone semi-analytic collapse model. Earlier studies assume the dust amount in the local universe, where all refractory elements are depleted onto grains, and/or assume the constant dust amount during gas collapse. In this paper, we employ the models of dust formation and destruction in early supernovae to derive the realistic dust compositions and size distributions for multiple species as the initial conditions of our collapse calculations. We also follow accretion of heavy elements in the gas phase onto dust grains, i.e., grain growth, during gas contraction. We find that grain growth well alters the fragmentation property of the clouds, and that this still does not approach to the value in the local universe. The critical conditions can be written by the gas metallicity Zcr and the initial depletion efficiency fdep,0 of gas-phase metal onto grains, or dust-to-metal mass ratio, as (Zcr/10^{-5.5} Zsun) = (fdep,0/0.18)^{-0.44} with small scatters in the range of Zcr = [0.06--3.2]x10^{-5} Zsun. We also show that the initial dust composition and size distribution are important to determine Zcr.

We studied the gravitational collapse of a warm (~8000 K) primordial-gas cloud as a candidate progenitor for a supermassive star (SMS; >10^5 Msun) using a three-dimensional hydrodynamical simulation, including all the relevant cooling processes of both H_2 and H, which can potentially induce cloud fragmentation. This is the first simulation of this kind to resolve protostar formation. We find that the cloud undergoes runaway collapse without a major episode of fragmentation. Although the H_2 fraction jumps by a large factor via the three-body reaction at ~10^-13 g/cm^3, its cooling remains inefficient due to the optical thickness, and the temperature remains >3000 K. When the central core of the cloud becomes opaque to continuum radiation at ~10^-8 g/cm^3, a hydrostatic protostar with ~0.2 Msun is formed. The protostar grows to the mass ~1 Msun and the radius ~2 AU within ~1 yr via rapid accretion of dense filamentary flows. With high accretion rate ~2 Msun/yr, the protostar is expected to turn into a SMS within its lifetime, eventually collapsing to a seed for the supermassive black hole observed in the early Universe at z~7.

We investigate the origin of carbon-enhanced metal-poor (CEMP) stars starting from the recently discovered $\rm [Fe/H]<-7.1$ star SMSS J031300 (Keller et al. 2014). We show that the elemental abundances observed on the surface of SMSS J031300 can be well fit by the yields of faint, metal free, supernovae. Using properly calibrated faint supernova explosion models, we study, for the first time, the formation of dust grains in such carbon-rich, iron-poor supernova ejecta. Calculations are performed assuming both unmixed and uniformly mixed ejecta and taking into account the partial destruction by the supernova reverse shock. We find that, due to the paucity of refractory elements beside carbon, amorphous carbon is the only grain species to form, with carbon condensation efficiencies that range between (0.15-0.84), resulting in dust yields in the range (0.025-2.25)M$_{\odot}$. We follow the collapse and fragmentation of a star forming cloud enriched by the products of these faint supernova explosions and we explore the role played by fine structure line cooling and dust cooling. We show that even if grain growth during the collapse has a minor effect of the dust-to-gas ratio, due to C depletion into CO molecules at an early stage of the collapse, the formation of CEMP low-mass stars, such as SMSS J031300, could be triggered by dust cooling and fragmentation. A comparison between model predictions and observations of a sample of C-normal and C-rich metal-poor stars supports the idea that a single common pathway may be responsible for the formation of the first low-mass stars.

It has been proposed that supermassive black holes (SMBHs) are originated from direct-collapse black holes (DCBHs) that are formed at z gtrsim 10 in the primordial gas in the case that H2 cooling is suppressed by strong external radiation. In this work, we study the critical specific intensity J^crit required for DCBH formation for various radiation spectral shapes by a series of one-zone calculations of a collapsing primordial- gas cloud. We calculate the critical specific intensity at the Lyman-Werner (LW) bands J^crit_LW,21 (in units of 10^-21 erg s^-1 Hz^-1 sr^-1 cm^-2) for realistic spectra of metal-poor galaxies. We find J^crit is not sensitive to the age or metallicity for the constant star formation galaxies with J^crit_LW,21 = 1300-1400, while J^crit decreases as galaxies become older or more metal-enriched for the instantaneous starburst galaxies. However, such dependence for the instantaneous starburst galaxies is weak for the young or extremely metal-poor galaxies: J^crit_LW,21 = 1000-1400 for the young galaxies and J^crit_LW,21 approx 1400 for the extremely metal-poor galaxies. The typical value of J^crit for the realistic spectra is higher than those expected in the literature, which affects the estimated DCBH number density n_DCBH. By extrapolating the result of Dijkstra, Ferrara and Mesinger, we obtain n_DCBH sim 10^-10 cMpc^-3 at z = 10, although there is still large uncertainty in this estimation.

HD dominates the cooling of primordial clouds with enhanced ionization, e.g. shock-heated clouds in structure formation or supernova remnants, relic HII regions of Pop III stars, and clouds with cosmic-ray (CR) irradiation. There, the temperature decreases to several 10 K and the characteristic stellar mass decreases to $\sim 10\ {\rm M}_{\odot}$, in contrast with first stars formed from undisturbed pristine clouds ($\sim 100\ {\rm M}_{\odot}$). However, without CR irradiation, even weak far ultra-violet (FUV) irradiation suppresses HD formation/cooling. Here, we examine conditions for HD cooling in primordial clouds including both FUV and CR feedback. At the beginning of collapse, the shock-compressed gas cools with its density increasing, while the relic HII region gas cools at a constant density. Moreover, shocks tend to occur in denser environments than HII regions. Owing to the higher column density and the more effective shielding, the critical FUV intensity for HD cooling in a shock-compressed gas becomes $\sim 10$ times higher than in relic HII regions. Consequently, in the shock-compressed gas, the critical FUV intensity exceeds the background level for most of the redshift we consider ($6 \lesssim z \lesssim 15$), while in relic HII regions, HD cooling becomes effective after the CR intensity increases enough at $z \lesssim 10$. Our result suggests that less massive ($\sim 10\ {\rm M}_{\odot}$) Pop III stars may be more common than previously considered and could be the dominant population of Pop III stars.

We perform a large set of radiation hydrodynamics simulations of primordial star formation in a fully cosmological context. Our statistical sample of 100 First Stars show that the first generation of stars have a wide mass distribution M_popIII = 10 ~ 1000 M_sun. We first run cosmological simulations to generate a set of primordial star-forming gas clouds. We then follow protostar formation in each gas cloud and the subsequent protostellar evolution until the gas mass accretion onto the protostar is halted by stellar radiative feedback. The accretion rates differ significantly among the primordial gas clouds which largely determine the final stellar masses. For low accretion rates the growth of a protostar is self-regulated by radiative feedback effects and the final mass is limited to several tens of solar masses. At high accretion rates the protostar's outer envelope continues to expand and the effective surface temperature remains low; such protostars do not exert strong radiative feedback and can grow in excess to one hundred solar masses. The obtained wide mass range suggests that the first stars play a variety of roles in the early universe, by triggering both core-collapse supernovae and pair-instability supernovae as well as by leaving stellar mass black holes. We find certain correlations between the final stellar mass and the physical properties of the star-forming cloud. These correlations can be used to estimate the mass of the first star from the properties of the parent cloud or of the host halo, without following the detailed protostellar evolution.

Dust grains in low-metallicity star-forming regions may be responsible for the formation of the first low-mass stars. The minimal conditions to activate dust-induced fragmentation require the gas to be pre-enriched above a critical dust-to-gas mass ratio Dcr=[2.6--6.3]x10^-9 with the spread reflecting the dependence on the grain properties. The recently discovered Galactic halo star SDSS J102915+172927 has a stellar mass of 0.8 Msun and a metallicity of Z=4.5x10^-5 Zsun and represents an optimal candidate for the dust-induced low-mass star formation. Indeed, for the two most plausible Population III supernova progenitors, with 20 Msun and 35 Msun, the critical dust-to-gas mass ratio can be overcome provided that at least 0.4 Msun of dust condenses in the ejecta, allowing for moderate destruction by the reverse shock. Here we show that even if dust formation in the first supernovae is less efficient or strong dust destruction does occur, grain growth during the collapse of the parent gas cloud is sufficiently rapid to activate dust cooling and likely fragmentation into low-mass and long-lived stars. Silicates and magnetite grains can experience significant grain growth in the density range 10^9 /cc < nH<10^12 /cc by accreting gas-phase species (SiO, SiO2, and Fe) until their gas-phase abundance drops to zero, reaching condensation efficiencies =1. The corresponding increase in the dust-to-gas mass ratio allows dust-induced cooling and fragmentation to be activated at 10^12 /cc < nH < 10^14 /cc, before the collapsing cloud becomes optically thick to continuum radiation. We show that for all the initial conditions that apply to the parent cloud of SDSS J102915+172927, dust-driven fragmentation is able to account for the formation of the star.

Fragmentation of protostellar disks controls the growth of protostars and plays a key role in determining the final mass of newborn stars. In this paper, we investigate the structure and gravitational stability of the protostellar disks in the full metallicity range between zero and the solar value. Using the mass-accretion rates evaluated from the thermal evolution in the preceding collapse phase of the pre-stellar cores, we calculate disk structures and their evolution in the framework of the standard steady disks. Overall, with higher metallicity, more efficient cooling results in the lower accretion rate and lower temperature inside the disk: at zero metallicity, the accretion rate is ~ 1e-3Msun/yr and the disk temperature is ~ 1000 K, while at solar metallicity, ~ 1e-6Msun/yr and 10 K. Despite the large difference in these values, the zero- and solar-metallicity disks have similar stability properties: the Toomre parameter for the gravitational stability, which can be written using the ratio of temperatures in the disk and in the envelope as Q ~ (T_disk/T_env)^3/2, is > 1, i.e., marginally stable. At intermediate metallicities of 1e-5--1e-3Zsun, however, the disks are found to be strongly unstable with Q ~ 0.1--1 since dust cooling, which is effective only in the disks due to their high density (> 1e10 cm^-3), makes the temperature in the disks lower than that in the envelopes. This indicates that masses of the individual stars formed as a result of the protostellar disk fragmentation can be significantly smaller than their parent core in this metallicity range. The typical stellar mass in this case would be a few Msun, which is consistent with the observationally suggested mass-scale of extremely metal-poor stars.

Supermassive stars (SMSs) forming via very rapid mass accretion (Mdot >~ 0.1 Msun/yr) could be precursors of supermassive black holes observed beyond redshift of about 6. Extending our previous work, we here study the evolution of primordial stars growing under such rapid mass accretion until the stellar mass reaches 10^{4 - 5} Msun. Our stellar evolution calculations show that a star becomes supermassive while passing through the "supergiant protostar" stage, whereby the star has a very bloated envelope and a contracting inner core. The stellar radius increases monotonically with the stellar mass, until =~ 100 AU for M_* >~ 10^4 Msun, after which the star begins to slowly contract. Because of the large radius the effective temperature is always less than 10^4 K during rapid accretion. The accreting material is thus almost completely transparent to the stellar radiation. Only for M_* >~ 10^5 Msun can stellar UV feedback operate and disturb the mass accretion flow. We also examine the pulsation stability of accreting SMSs, showing that the pulsation-driven mass loss does not prevent stellar mass growth. Observational signatures of bloated SMSs should be detectable with future observational facilities such as the James Webb Space Telescope. Our results predict that an inner core of the accreting SMS should suffer from the general relativistic instability soon after the stellar mass exceeds 10^5 Msun. An extremely massive black hole should form after the collapse of the inner core.

Photoevaporation by stellar ionizing radiation is believed to play an important role in the dispersal of disks around young stars. The mass loss model for dust-free disks developed by Hollenbach et al. is currently regarded as a conventional one and has been used in a wide variety of studies. However, the rate in this model was derived by the crude so-called 1+1D approximation of ionizing radiation transfer, which assumes that diffuse radiation propagates in a direction vertical to the disk. In this study, we revisit the photoevaporation of dust-free disks by solving the 2D axisymmetric radiative transfer for steady-state disks. Unlike that solved by the conventional model, we determine that direct stellar radiation is more important than the diffuse field at the disk surface. The radial density distribution at the ionization boundary is represented by the single power-law with an index -3/2 in contrast to the conventional double power-law. For this distribution, the photoevaporation rate from the entire disk can be written as a function of the ionizing photon emissivity, Phi_EUV, from the central star and the disk outer radius, r_d, as follows: Mdot_PE = 5.4 x 10^-5 x (Phi_EUV/10^49 sec^-1)^1/2 x (r_d/1000 AU)^1/2 Msun/yr. This new rate depends on the outer disk radius rather than on the gravitational radius as in the conventional model, caused by the enhanced contribution to the mass loss from the outer disk annuli. In addition, we discuss its applications to present-day as well as primordial star formation.

Supermassive stars (SMSs; M>10^5 Msun) and their remnant black holes are promising progenitors for supermassive black holes (SMBHs) observed in the early universe at z>7. It has been postulated that SMSs forms through very rapid mass accretion onto a protostar at a high rate exceeding 0.01 Msun/yr. According to recent studies, such rapidly accreting protostars evolve into "supergiant protostars", i.e. protostars consisting of a bloated envelope and a contracting core, similar to giant star. However, like massive stars as well as giant stars, both of which are known to be pulsationally unstable, supergiant protostars may also be also unstable to launch strong pulsation-driven outflows. If this is the case, the stellar growth via accretion will be hindered by the mass loss. We here study the pulsational stability of the supergiant protostars in the mass range M<10^3 Msun through the method of the linear perturbation analysis. We find that the supergiant protostars with M>600 Msun and very high accretion rate Mdot>1.0 Msun/yr are unstable due to the kappa mechanism. The pulsation is excited in the He^+ ionization layer in the envelope. Even under a conservative assumption that all the pulsation energy is converted into the kinetic energy of the outflows, the mass-loss rate is ~10^-3 Msun/yr, which is lower than the accretion rate by more than two orders of magnitude. We thus conclude that the supergiant protostars should grow stably via rapid accretion at least in the mass range we studied. As long as the rapid accretion is maintained in the later stage, protostars will become SMSs, which eventually produce seeds for the high-z SMBHs.

The first stars in the universe ionized the ambient primordial gas through various feedback processes. "Second-generation" primordial stars potentially form from this disturbed gas after its recombination. In this Letter, we study the late formation stage of such second-generation stars, where a large amount of gas accretes onto the protostar and the final stellar mass is determined when the accretion terminates. We directly compute the complex interplay between the accretion flow and stellar ultraviolet (UV) radiation, performing radiation-hydrodynamic simulations coupled with stellar evolution calculations. Because of more efficient H2 and HD cooling in the pre-stellar stage, the accretion rates onto the star are ten times lower than in the case of the formation of the first stars. The lower accretion rates and envelope density result in the occurrence of an expanding bipolar HII region at a lower protostellar mass M_* \simeq 10Msun, which blows out the circumstellar material, thereby quenching the mass supply from the envelope to the accretion disk. At the same time the disk loses mass due to photoevaporation by the growing star. In our fiducial case the stellar UV feedback terminates mass accretion onto the star at M_* \simeq 17Msun. Although the derived masses of the second-generation primordial stars are systematically lower than those of the first generation, the difference is within a factor of only a few. Our results suggest a new scenario, whereby the majority of the primordial stars are born as massive stars with tens of solar masses, regardless of their generations.

We combine the output of hydrodynamical simulations of Population III star cluster formation with stellar evolution models, and calculate the evolution of protostars experiencing variable mass accretion rates due to interactions within a massive disk. We find that the primordial protostars are extended 'fluffy' objects for the bulk of their pre-main-sequence lifetimes. Accretion luminosity feedback from such objects is high, but as shown in previous work, has a minimal effect on the star cluster. The extended radii of the protostars, combined with the observation of close encounters in the simulations, suggests that mergers will occur in such systems. Furthermore, mass transfer between close protostellar binaries with extended radii could lead to massive tight binaries, which are a possible progenitor of gamma ray bursts.

We study effects of the fully ionized initial state, or pre-ionization, on the subsequent thermal evolution of low-metallicity clouds under various intensities of the external far-ultraviolet(FUV) and cosmic-ray(CR) fields. The pre-ionization significantly affects the thermal and dynamical evolution of metal-free clouds without FUV/CRs by way of efficient HD formation. On the other hand, the pre-ionization effect on the thermal evolution is limited in very low-density regime for more metal-enriched clouds ([Z/H] >~ -4) or those under modest FUV (>10^{-3}) or CR field (>0.1 of the present-day Galactic disk levels). In any case, for >10^8 cm^{-3}, neither the initial ionization state nor the irradiating FUV strength affect the thermal evolution. The dust cooling is an important mechanism for making sub-solar mass fragments in low-metallicity gas. Since this fragmentation occurs at the temperature minimum by the dust cooling at >10^{10} cm^{-3}, this process is not vulnerable either to initial ionization state or external radiation.

The relative importance of metals and dust grains in the formation of the first low-mass stars has been a subject of debate. The recently discovered Galactic halo star SDSS J102915+172927 (Caffau et al. 2011) has a mass less than 0.8 Msun and a metallicity of Z = 4.5 10^{-5} Zsun. We investigate the origin and properties of this star by reconstructing the physical conditions in its birth cloud. We show that the observed elemental abundance trend of SDSS J102915+172927 can be well fitted by the yields of core-collapse supernovae with metal-free progenitors of 20 Msun and 35 Msun. Using these selected supernova explosion models, we compute the corresponding dust yields and the resulting dust depletion factor taking into account the partial destruction by the supernova reverse shock. We then follow the collapse and fragmentation of a star forming cloud enriched by the products of these SN explosions at the observed metallicity of SDSS J102915+172927. We find that [0.05 - 0.1] Msun mass fragments, which then lead to the formation of low-mass stars, can occur provided that the mass fraction of dust grains in the birth cloud exceeds 0.01 of the total mass of metals and dust. This, in turn, requires that at least 0.4 Msun of dust condense in the first supernovae, allowing for moderate destruction by the reverse shock. If dust formation in the first supernovae is less efficient or strong dust destruction does occur, the thermal evolution of the SDSS J102915+172927 birth cloud is dominated by molecular cooling, and only > 8 Msun fragments can form. We conclude that the observed properties of SDSS J102915+172927 support the suggestion that dust must have condensed in the ejecta of the first supernovae and played a fundamental role in the formation of the first low-mass stars.

Direct collapse of supermassive stars (SMSs) is a possible pathway for generating supermassive black holes in the early universe. It is expected that an SMS could form via very rapid mass accretion with Mdot ~ 0.1 - 1 Msun/yr during the gravitational collapse of an atomic-cooling primordial gas cloud. In this paper we study how stars would evolve under such extreme rapid mass accretion, focusing on the early evolution until the stellar mass reaches 1000 Msun. To this end we numerically calculate the detailed interior structure of accreting stars with primordial element abundances. Our results show that for accretion rates higher than 0.01 Msun/yr, stellar evolution is qualitatively different from that expected at lower rates. While accreting at these high rates the star always has a radius exceeding 100 Rsun, which increases monotonically with the stellar mass. The mass-radius relation for stellar masses exceeding ~ 100 Msun follows the same track with R_* \propto M_*^0.5 in all cases with accretion rates > 0.01 Msun/yr; at a stellar mass of 1000 Msun the radius is about 7000 Rsun (~= 30 AU). With higher accretion rates the onset of hydrogen burning is shifted towards higher stellar masses. In particular, for accretion rates exceeding Mdot > 0.1 Msun/yr, there is no significant hydrogen burning even after 1000 Msun have accreted onto the protostar. Such "supergiant" protostars have effective temperatures as low as Teff ~= 5000 K throughout their evolution and because they hardly emit ionizing photons, they do not create an HII region or significantly heat their immediate surroundings. Thus, radiative feedback is unable to hinder the growth of rapidly accreting stars to masses in excess of 1000 Msun, as long as material is accreted at rates Mdot > 0.01 Msun/yr.

We propose a new scenario for supermassive star (SMS;>10^5Msun) formation in shocked regions of colliding cold accretion flows near the centers of first galaxies. Recent numerical simulations indicate that assembly of a typical first galaxy with virial temperature (~10^4K) proceeds via cold and dense flows penetrating deep to the center, where the supersonic streams collide each other to develop a hot and dense (~10^4K, ~10^3/cc) shocked gas. The post-shock layer first cools by efficient Ly alpha emission and contracts isobarically until 8000K. Whether the layer continues the isobaric contraction depends on the density at this moment: if the density is high enough for collisionally exciting H2 rovibrational levels (>10^4/cc), enhanced H2 collisional dissociation suppresses the gas to cool further. In this case, the layer fragments into massive (>10^5Msun) clouds, which collapse isothermally (~8000K) by the Ly alpha cooling without subsequent fragmentation. As an outcome, SMSs are expected to form and evolve eventually to seeds of supermassive black holes (SMBH). By calculating thermal evolution of the post-shock gas, we delimit the range of post-shock conditions for the SMS formation, which can be expressed as: T>6000K/(n/10^4/cc) for n<10^4/cc and T>5000-6000K for n>10^4/cc, depending somewhat on initial ionization degree. We found that metal enrichment does not affect the above condition for metallicity below 10^-3Zsun if metals are in the gas phase, while condensation of several percent of metals into dust decreases this critical value of metallicity by an order of magnitude. Unlike the previously proposed scenario for SMS formation, which postulates extremely strong ultraviolet radiation to quench H2 cooling, our scenario naturally explains the SMBH seed formation in the assembly process of the first galaxies, even without such a strong radiation.

The first stars fundamentally transformed the early universe by emitting the first light and by producing the first heavy elements. These effects were predetermined by the mass distribution of the first stars, which is thought to have been fixed by a complex interplay of gas accretion and protostellar radiation. We performed radiation-hydrodynamics simulations that followed the growth of a primordial protostar through to the early stages as a star with thermo-nuclear burning. The circumstellar accretion disk was evaporated by ultraviolet radiation from the star when its mass was 43 times that of the Sun. Such massive primordial stars, in contrast to the often postulated extremely massive stars, may help explain the fact that there are no signatures of the pair-instability supernovae in abundance patterns of metal-poor stars in our galaxy.

We explore the minimal conditions which enable the formation of metal-enriched solar and sub-solar mass stars. We find that in the absence of dust grains, gas fragmentation occurs at densities nH ~ [10^4-10^5]cm^{-3} when the metallicity exceeds Z ~ 10^{-4} Zsun. The resulting fragmentation masses are > 10 Msun. The inclusion of Fe and Si cooling does not affect the thermal evolution as this is dominated by molecular cooling even for metallicities as large as Z = 10^{-2} Zsun. The presence of dust is the key driver for the formation of low-mass stars. We focus on three representative core-collapse supernova (SN) progenitors, and consider the effects of reverse shocks of increasing strength: these reduce the depletion factors, fdep = Mdust/(Mdust+Mmet), alter the shape of the grain size distribution function and modify the relative abundances of grain species and of metal species in the gas phase. We find that the lowest metallicity at which fragmentation occurs is Z=10^{-6} Zsun for gas pre-enriched by the explosion of a 20 Msun primordial SN (fdep > 0.22) and/or by a 35 Msun, Z=10^{-4} Zsun SN (fdep > 0.26); it is ~ 1 dex larger, when the gas is pre-enriched by a Z = 10^{-4} Zsun, 20 Msun SN (fdep > 0.04). Cloud fragmentation depends on the depletion factor and it is suppressed when the reverse shock leads to a too large destruction of dust grains. These features are all consistent with the existence of a minimum dust-to-gas ratio, Dcr, above which fragmentation is activated. We derive a simple analytic expression for Dcr which, for grain composition and properties explored in the present study, reads Dcr = [2.6 - 6.3] x 10^{-9}. When the dust-to-gas ratio of star forming clouds exceeds this value, the fragmentation masses range between 0.01 Msun and 1 Msun, thus enabling the formation of the first low-mass stars.

We study effects of external ionization by cosmic rays (CRs) and X-rays on the thermal evolution of primordial clouds under strong far-ultraviolet (FUV) radiation. A strong FUV radiation dissociates H2 and quenches its cooling. Even in such an environment, a massive cloud with Tvir>10^4 K can contract isothermally at 8000 K by Lyman alpha cooling. This cloud collapses monolithically without fragmentation, and a supermassive star (>10^5 Msun) is believed to form at the center, which eventually evolves to a supermassive black hole (SMBH). However, candidates of FUV sources, including star-forming galaxies, are probably sources of strong CRs and X-rays, as well. We find that the external ionization promotes H2 production and elevates the threshold FUV intensity Jcr needed for the SMBH formation for CR energy density U_CR>10^-14 erg/cm^3 or X-ray intensity J_X>10^-24 erg/s/cm^2/sr/Hz at 1 keV. The critical FUV flux increases in proportion to U_CR^{1/2} (J_X^{1/2}) in the high CR (X-ray, respectively) limit. With the same value of FUV intensity at the Lyman limit (13.6 eV), the H^- photodissociation rate, with threshold of 0.755 eV, increases and the H2 molecules decrease with decreasing effective temperature of the FUV sources T*. The lower value of T* thus results in the lower critical FUV flux at 13.6 eV. Using an empirical relation between intensities of FUV and CR/X-ray from nearby star-forming galaxies, we find that external ionization effect remarkably enhances the critical FUV flux for sources with T* as high as 10^5 K and composed of stars with <100 Msun to a level that is not realized in any halo. This indicates that, to induce SMBH formation, the FUV sources must be either Pop II/I galaxies with low brightness temperature, Pop III galaxies with a very top-heavy IMF, or Pop III galaxies too young to harbor sources of CR/X-ray, e.g., supernova remnants or high-mass X-ray binaries.

The collapse of dense cores with different metallicities is studied by hydrodynamical calculations coupled with detailed chemical and radiative processes. For this purpose, we construct a simple chemical network with non-equilibrium reactions among 15 chemical species, which reproduces the abundance of important molecular coolants by more detailed network very well. The evolution is followed until the formation of a hydrostatic protostar at the center. In a lower-metallicity gas cloud, the temperature during the collapse remains high owing to less efficient cooling. Using the temperature evolution at the center as a function the density, we discuss the possibility of fragmentation during the dust-cooling phase. The critical metallicity for the fragmentation is 10^{-5}Z_sun assuming moderate elongation of the cloud cores at the onset of this phase. From the density and velocity distributions at the time of protostar formation, we evaluate the mass accretion rate in the subsequent accretion phase. Using these accretion rates, we also calculate the evolution of the protostars under the assumption of stationary accretion flow. Finally, we discuss possible suppression of fragmentation by heating of the ambient gas by protostellar radiation, which is considered important in the contemporary star formation. We argue that it is negligible for <10^{-2}Zsun.

Mass accretion onto (proto-)stars at high accretion rates > 10^-4 M_sun/yr is expected in massive star formation. We study the evolution of massive protostars at such high rates by numerically solving the stellar structure equations. In this paper we examine the evolution via disk accretion. We consider a limiting case of "cold" disk accretion, whereby most of the stellar photosphere can radiate freely with negligible backwarming from the accretion flow, and the accreting material settles onto the star with the same specific entropy as the photosphere. We compare our results to the calculated evolution via spherically symmetric accretion, the opposite limit, whereby the material accreting onto the star contains the entropy produced in the accretion shock front. We examine how different accretion geometries affect the evolution of massive protostars. For cold disk accretion at 10^-3 M_sun/yr the radius of a protostar is initially small, about a few R_sun. After several solar masses have accreted, the protostar begins to bloat up and for M \simeq 10 M_sun the stellar radius attains its maximum of 30 - 400 R_sun. The large radius about 100 R_sun is also a feature of spherically symmetric accretion at the same accreted mass and accretion rate. Hence, expansion to a large radius is a robust feature of accreting massive protostars. At later times the protostar eventually begins to contract and reaches the Zero-Age Main-Sequence (ZAMS) for M \simeq 30 M_sun, independent of the accretion geometry. For accretion rates exceeding several 10^-3 M_sun/yr the protostar never contracts to the ZAMS. The very large radius of several 100s R_sun results in a low effective temperature and low UV luminosity of the protostar. Such bloated protostars could well explain the existence of bright high-mass protostellar objects, which lack detectable HII regions.

We investigate the effects of the Cosmic Microwave Background (CMB) radiation field on the collapse of prestellar clouds. Using a semi-analytic model to follow the thermal evolution of clouds with varying initial metallicities and dust contents at different redshifts, we study self-consistently the response of the mean Jeans mass at cloud fragmentation to metal line-cooling, dust-cooling and the CMB. In the absence of dust grains, at redshifts z < 10 moderate characteristic masses (of 10s of Msun) are formed when the metallicity is 10^{-4} Zsun < Z < 10^{-2.5} Zsun; at higher metallicities, the CMB inhibits fragmentation and only very large masses (of ~ 100s of Msun) are formed. These effects become even more dramatic at z > 10 and the fragmentation mass scales are always > 100s of Msun, independent of the initial metallicity. When dust grains are present, sub-solar mass fragments are formed at any redshift for metallicities Z > 10^{-6} Zsun because dust-cooling remains relatively insensitive to the presence of the CMB. When Z > 10^{-3} Zsun, heating of dust grains by the CMB at z > 5 favors the formation of larger masses, which become super-solar when Z > 10^{-2} Zsun and z > 10. Finally, we discuss the implications of our result for the interpretation of the observed abundance patterns of very metal-poor stars in the galactic halo.

Star formation in relic HII regions of the first stars is investigated using magneto-hydrodynamical simulations with a nested grid method that covers 10 orders of magnitude in spatial scale and 20 orders of magnitude in density contrast. Due to larger fraction of H_2 and HD molecules, its prestellar thermal evolution is considerably different from that in the first star formation. Reflecting the difference, two hydrostatic cores appear in a nested manner: a protostar is enclosed by a transient hydrostatic core, which appears during the prestellar collapse. If the initial natal core rotates fast at a rate with rotational to gravitational energy ratio \beta_0 > 0.01-0.1, the transient hydrostatic core fragments to \sim 10 M_\odot sub-cores at density \sim 10^9 cm^-3. With smaller rotation energy, fragmentation occurs at higher density while a single protostar forms without fragmentation if rotation is extremely slow with \beta_0 < 10^-6 -10^-5. If magnetic field is present, these threshold values of \beta_0 is boosted owing to angular momentum transport by the magnetic breaking. Magnetic field also drives the protostellar outflows. With strong magnetic field, two distinct outflows are observed: The slower one emanates from the transient hydrostatic core, while the faster one from the protostar. These flows may affect the final stellar mass by ejecting some of masses in the initial core, and also may play some role in driving and maintenance of interstellar turbulence in young galaxies.

We study the evolution of supernova remnants in a low-metallicity medium $Z/Z_{\odot} = 10^{-4}$ -- $10^{-2}$ in the early universe, using one-dimensional hydrodynamics with non-equilibrium chemistry. Once a post-shock layer is able to cool radiatively, a dense shell forms behind the shock. If this shell becomes gravitationally unstable and fragments into pieces, next-generation stars are expected to form from these fragments. To explore the possibility of this triggered star formation, we apply a linear perturbation analysis of an expanding shell to our results and constrain the parameter range of ambient density, explosion energy, and metallicity where fragmentation of the shell occurs. For the explosion energy of $10^{51}{\rm ergs} (10^{52}{\rm ergs})$, the shell fragmentation occurs for ambient densities higher than $\gtrsim 10^{2} {\rm cm^{-3}}$ (10 ${\rm cm^{-3}}$, respectively). This condition depends little on the metallicity in the ranges we examined. We find that the mode of star formation triggered occurs only in massive ($\gtrsim 10^{8}M_{\odot}$) haloes.

The fragmentation process in collapsing clouds with various metallicities is studied using three-dimensional nested-grid hydrodynamics. Initial clouds are specified by three parameters: cloud metallicity, initial rotation energy and initial cloud shape. For different combinations of these parameters, we calculate 480 models in total and study cloud evolution, fragmentation conditions, orbital separation and binary frequency. For the cloud to fragment during collapse, the initial angular momentum must be higher than a threshold value, which decreases with decreasing metallicity. Although the exact fragmentation conditions depend also on the initial cloud shape, this dependence is only modest. Our results indicate a higher binary frequency in lower-metallicity gas. In particular, with the same median rotation parameter as in the solar neighbourhood, a majority of stars are born as members of binary/multiple systems for < 10^-4 Z_sun. With initial mass < 0.1 M_sun, if fragments are ejected in embryo from the host clouds by multi-body interaction, they evolve to substellar-mass objects. This provides a formation channel for low-mass stars in zero- or low-metallicity environments.

Dust grains coagulate into larger aggregates in dense gas. This changes their size distribution and possibly affects the thermal evolution of star-forming clouds. We here investigate dust coagulation in collapsing pre-stellar cores with different metallicities by considering the thermal motions of grains. We show that coagulation does occur even at low metallicity $\sim 10^{-6} \Zsun$. However, we also find (i) that the H$_2$ formation rate on dust grains is reduced only after the majority of H$_2$ is formed; and (ii) that the dust opacity is modified only after the core becomes optically thick. Therefore, we conclude that the effects of dust coagulation can safely be neglected in discussing the temperature evolution of the pre-stellar cores for any metallicity as long as the grain motions are thermal.

The final mass of a newborn star is set at the epoch when the mass accretion onto the star is terminated. We study the evolution of accreting protostars and the limits of accretion in low metallicity environments. Accretion rates onto protostars are estimated via the temperature evolution of prestellar cores with different metallicities. The derived rates increase with decreasing metallicity, from 10^-6 M_sun/yr at Z = Z_sun to 10^-3 M_sun/yr at Z = 0. With the derived accretion rates, the protostellar evolution is numerically calculated. We find that, at lower metallicity, the protostar has a larger radius and reaches the zero-age main-sequence (ZAMS) at higher stellar mass. Using this protostellar evolution, we evaluate the upper stellar mass limit where the mass accretion is hindered by radiative feedback. We consider the effects of radiation pressure exerted on the accreting envelope, and expansion of the HII region. The mass accretion is finally terminated by radiation pressure on dust grains in the envelope for Z > 10^-3 Z_sun and by the expanding HII region for lower metallicty. The mass limit from these effects increases with decreasing metallicity from 10 M_sun at Z = Z_sun to about 300 M_sun at Z = 10^-6 Z_sun. The termination of accretion occurs after the central star arrives at the ZAMS at all metallicities, which allows us to neglect protostellar evolution effects in discussing the upper mass limit by stellar feedback. The fragmentation induced by line cooling in low-metallicity clouds yields prestellar cores with masses large enough that the final stellar mass is set by the feedback effects.

The nature of the first generation of stars in the Universe remains largely unknown. Observations imply the existence of massive primordial stars early in the history of the universe, and the standard theory for the growth of cosmic structure predicts that structures grow hierarchically through gravitational instability. We have developed an ab initio computer simulation of the formation of primordial stars that follows the relevant atomic and molecular processes in a primordial gas in an expanding universe. The results show that primeval density fluctuations left over from the Big Bang can drive the formation of a tiny protostar with a mass of just one percent that of the Sun. The protostar is a seed for the subsequent formation of a massive primordial star.

Primordial gas in protogalactic dark matter (DM) halos with virial temperatures Tvir > 10^4 K begins to cool and condense via atomic hydrogen. Provided this gas is irradiated by a strong ultraviolet (UV) flux and remains free of H2 and other molecules, it has been proposed that the halo with Tvir ~10^4 K may avoid fragmentation, and lead to the rapid formation of a supermassive black hole (SMBH) as massive as M=10^5-10^6 Msun. This ``head--start'' would help explain the presence of SMBHs with inferred masses of several x 10^9 Msun, powering the bright quasars discovered in the Sloan Digital Sky Survey at redshift z>~6. However, high-redshift DM halos with Tvir~10^4K are likely already enriched with at least trace amounts of metals and dust produced by prior star-formation in their progenitors. Here we study the thermal and chemical evolution of low-metallicity gas exposed to extremely strong UV radiation fields. Our results, obtained in one-zone models, suggest that gas fragmentation is inevitable above a critical metallicity, whose value is between Zcr~3x10^{-4} Zsun (in the absence of dust) and as low as Zcr~ 5 x 10^{-6} Zsun (with a dust-to-gas mass ratio of about 0.01 Z/Zsun). We propose that when the metallicity exceeds these critical values, dense clusters of low--mass stars may form at the halo nucleus. Relatively massive stars in such a cluster can then rapidly coalesce into a single more massive object, which may produce an intermediate-mass BH remnant with a mass up to M <~10^2-10^3 Msun.

Formation of massive stars by accretion requires a high accretion rate of > 10^-4 M_sun/yr to overcome the radiation pressure barrier of the forming stars. Here, we study evolution of protostars accreting at such high rates, by solving the structure of the central star and the inner accreting envelope simultaneously. The protostellar evolution is followed starting from small initial cores until their arrival at the stage of the Zero-Age Main Sequence (ZAMS) stars. An emphasis is put on evolutionary features different from those with a low accretion rate of 10^-5 M_sun/yr, which is presumed in the standard scenario for low-mass star formation. With the high accretion rate of 10^-3 M_sun/yr, the protostellar radius becomes very large and exceeds 100 R_sun. It is not until the stellar mass reaches 40 M_sun that hydrogen burning begins and the protostar reaches the ZAMS phase, and this ZAMS arrival mass increases with the accretion rate. At a very high accretion rate of > 3 x 10^-3 M_sun/yr, the total luminosity of the protostar becomes so high that the resultant radiation pressure inhibits the growth of the protostars under steady accretion before reaching the ZAMS stage. Therefore, the evolution under the critical accretion rate 3 x 10^-3 M_sun/yr gives the upper mass limit of possible pre-main-sequence stars at 60 M_sun. The upper mass limit of MS stars is also set by the radiation pressure onto the dusty envelope under the same accretion rate at 250 M_sun. We also propose that the central source enshrouded in the Orion KL/BN nebula has effective temperature and luminosity consistent with our model, and is a possible candidate for such protostars growing under the high accretion rate. (abridged)

If a significant fraction of metals is in dust, star-forming cores with metallicity higher than a critical value ~10^{-6}-10^{-5}Z_sun are able to fragment by dust cooling, thereby producing low-mass cores. Despite being above the critical metallicity, a metallicity range is found to exist around 10^{-5}-10^{-4}Z_sun where low-mass fragmentation is prohibited. In this range, three-body H_2 formation starts at low (~100K) temperature and thus the resulting heating causes a dramatic temperature jump, which makes the central part of the star-forming core transiently hydrostatic and thus highly spherical. With little elongation, the core does not experience fragmentation in the subsequent dust-cooling phase. The minimum fragmentation mass is set by the Jeans mass just before the H_2 formation heating, and its value can be as high as ~10M_sun. For metallicity higher than ~10^{-4}Z_sun, H_2 formation is almost completed by the dust-surface reaction before the onset of the three-body reaction, and low-mass star formation becomes possible. This mechanism might explain the higher characteristic mass of metal-poor stars than in the solar neighborhood presumed from the statistics of carbon-enhanced stars.

The fragmentation process of primordial-gas cores during prestellar collapse is studied using three-dimensional nested-grid hydrodynamics. Starting from the initial central number density of n \sim10^3 cm^-3, we follow the evolution of rotating spherical cores up to the stellar density n \simeq 10^{22} cm^-3. An initial condition of the cores is specified by three parameters: the ratios of the rotation and thermal energies to the gravitational energy (\beta_0, and \alpha_0, respectively), and the amplitude of the bar-mode density perturbation (A_\phi). Cores with rotation \beta_0 > 10^{-6} are found to fragment during the collapse. The fragmentation condition hardly depends on either the initial thermal energy \alpha_0 or amplitude of bar-mode perturbation A_\phi. Since the critical rotation parameter for fragmentation is lower than that expected in the first star formation, binaries or multiples are also common for the first stars.

We use cosmological hydrodynamic simulations with unprecedented resolution to study the formation of primordial stars in an ionized gas at high redshifts. Our approach includes all the relevant atomic and molecular physics to follow the thermal evolution of a prestellar gas cloud to very high densities of ~10^{18} cm^{-3}. We locate a star-forming gas cloud within a reionized region in our cosmological simulation. The first run-away collapse is triggered when the gas cloud's mass is ~40 Msun. We show that the cloud core remains stable against chemo-thermal instability and also against gravitational deformation throughout its evolution. Consequently, a single proto-stellar seed is formed, which accretes the surrounding hot gas at the rate ~10^{-3} Msun/year. We carry out proto-stellar evolution calculations using the inferred accretion rate. The resulting mass of the star when it reaches the zero-age main sequence is M_ZAMS ~40 Msun. We argue that, since the obtained M_ZAMS is as large as the mass of the collapsing parent cloud, the final stellar mass should be close to this value. Such massive, rather than exceptionally massive, primordial stars are expected to cause early chemical enrichment of the Universe by exploding as black hole-forming super/hypernovae, and may also be progenitors of high redshift gamma-ray bursts. The elemental abundance patterns of recently discovered hyper metal-poor stars suggest that they might have been born from the interstellar medium that was metal-enriched by supernovae of these massive primordial stars.

We evaluate the broadband afterglow emission of very high redshift gamma-ray bursts (GRBs) using standard relativistic blastwave models with both forward and reverse shock components. For a broad range of parameters, a generic property for GRBs at redshifts $z \sim$ 5--30 is that the emission peaks in the millimeter to far-infrared bands with milli-Jansky flux levels, first at a few hours after the burst due to the reverse shock, and then again for several days afterwards with somewhat lower flux due to the forward shock. The radio, submillimeter and infrared continuum emission should be readily detectable out to $z \ga 30$ by the Atacama Large Millimeter Array (ALMA), Extended Very Large Array (EVLA), Square Kilometer Array (SKA) and other facilities. For relatively bright bursts, spectroscopic measurements of molecular and atomic absorption lines due to ambient protostellar gas may be possible. Utilizing models of primordial protostellar clouds, we show that under certain conditions, appreciable absorption may be caused by HD rotational transitions even in metal-free environments. After sufficient metal enrichment, absorption from CO rotational transitions and [OI] fine-structure transitions can also become strong. With appropriate observing strategies in combination with optical telescopes, ALMA and/or SKA may be able to detect such lines, offering a unique probe of physical conditions in individual Pop III and early Pop II star forming regions. We also remark on potential near-infrared absorption features due to electronic transitions of H$_2$.

First protostellar cores are young stellar objects in the earliest evolutionary stage. They are hydrostatic objects formed soon after the central portions of star-forming cores become optically thick to dust emission. We consider their characteristics in the emitted radiation, and discuss their evolution with increasing mass of the cores. Particular attention is paid to detailed radiative and chemical processes in the postshock relaxation layer located at the surface of the core, where the majority of radiation is emitted. Most of the radiation is originally emitted in the dust continuum in mid-infrared wavelength (~10-30 micron), which reprocessed to far-infrared with ~100-200 micron. Although some fraction (~0.1) of the radiation energy is emitted in the H2O lines at the accretion shock, most is absorbed and reemitted in the dust continuum in the envelope. The H2O lines account for at most ~1/100 of the observed luminosity. If a cavity is present in the envelope due to outflow or rotation, the dust and H2O line emission in the mid-infrared wavelength from the shock can be observed directly, or as a reflection nebula. Among forthcoming observational facillities, SPICA is the most suitable for detecting either direct or processed radiation from first-core objects.

We examine observational characteristics of multi-phase turbulent flows in the diffuse interstellar medium (ISM) using a synthetic radiation field of atomic and molecular lines. We consider the multi-phase ISM which is formed by thermal instability under the irradiation of UV photons with moderate visual extinction $A_V\sim 1$. Radiation field maps of C$^{+}$, C$^0$, and CO line emissions were generated by calculating the non-local thermodynamic equilibrium (nonLTE) level populations from the results of high resolution hydrodynamic simulations of diffuse ISM models. By analyzing synthetic radiation field of carbon lines of [\ion{C}{2}] 158 $\mu$m, [\ion{C}{1}] $^3P_2-^3P_1$ (809 GHz), $^3P_1-^3P_0$ (492 GHz), and CO rotational transitions, we found a high ratio between the lines of high- and low-excitation energies in the diffuse multi-phase interstellar medium. This shows that simultaneous observations of the lines of warm- and cold-gas tracers will be useful in examining the thermal structure, and hence the origin of diffuse interstellar clouds.

We study the formation of the first generation of stars in the standard cold dark matter model, using a very high-resolution hydordynamic simulations. Our simulation achieves a dynamic range of 10^{10} in length scale. With accurate treatment of atomic and molecular physics, it allows us to study the chemo-thermal evolution of primordial gas clouds to densities up to n = 10^{16}/cc without assuming any a priori equation of state; a six orders of magnitudes improvement over previous three-dimensional calculations. All the relevant atomic and molecular cooling and heating processes, including cooling by collision-induced continuum emission, are implemented. For calculating optically thick H2 cooling at high densities, we use the Sobolev method. To examine possible gas fragmentation owing to thermal instability, we compute explicitly the growth rate of isobaric perturbations. We show that the cloud core does not fragment in either the low-density or high-density regimes. We also show that the core remains stable against gravitational deformation and fragmentation. We obtain an accurate gas mass accretion rate within a 10 Msun innermost region around the protostar. The protostar is accreting the surrounding hot gas at a rate of 0.001-0.01 Msun/yr. From these findings we conclude that primordial stars formed in early minihalos are massive. We carry out proto-stellar evolution calculations using the obtained accretion rate. The resulting mass of the first star is M_ZAMS = 60-100 Msun, with the exact mass dependent on the actual accretion rate.

The protostellar jets driven by the formation of the first stars are studied by using MHD nested grid simulations. Starting from a slowly rotating spherical cloud of 5.1 times 10^4 Msun permeated by a uniform magnetic field, we follow the evolution from the central number density n = 10^3 cm^-3 to n simeq 10^23 cm^-3. We calculate four models that differ in initial magnetic field strengths and angular velocities. In all models, protostars of simeq 10^-3 Msun are formed at n simeq 10^22 cm^-3 in accordance with one-dimensional calculations. By this epoch, the magnetic flux density is amplified by 10 orders of magnitude from the initial value. Consequently, the formed protostar possesses the magnetic field of \sim 10^6 G that is much larger than the flux density of the present counterparts, reflecting the fact that the dissipation of a magnetic field is ineffective in primordial gas clouds. If the initial magnetic field B > 10^-9 (n/10^3 cm^-3)^2/3 G, the protostellar jet is launched and its velocities reaches \sim 70 km s^-1 by the time the protostellar mass becomes (4-6) times 10^-3 Msun, and a fraction (3-10%) of the accreting matter is blown off from the central region. Owing to the interaction of these ejecta with surrounding matter, expanding bow shocks are created at both heads of the jet. If this jet continues to sweep out the surrounding gas that otherwise accretes onto the central star or circumstellar disk, the final mass of the first star can be substantially reduced. In addition, dense post-shock regions behind the bow shocks are expected to promote the chemical reactions (formation of H_2 and HD), and this provides possible environments for subsequent low-mass star formation in the early universe.

The thermal and fragmentation properties of star-forming clouds have important consequences on the corresponding characteristic stellar mass. The initial composition of the gas within these clouds is a record of the nucleosynthetic products of previous stellar generations. In this paper we present a model for the evolution of star-forming clouds enriched by metals and dust from the first supernovae, resulting from the explosions of metal-free progenitors with masses in the range 12 - 30 Msun and 140 - 260 Msun. Using a self-consistent approach, we show that: (i) metals depleted onto dust grains play a fundamental role, enabling fragmentation to solar or sub-solar mass scales already at metallicities Zcr = 10^{-6} Zsun; (ii) even at metallicities as high as 10^{-2} Zsun, metals diffused in the gas-phase lead to fragment mass scales which are ~ 100 Msun; (iii) C atoms are strongly depleted onto amorphous carbon grains and CO molecules so that CII plays a minor role in gas cooling, leaving OI as the main gas-phase cooling agent in low-metallicity clouds. These conclusions hold independently of the assumed supernova progenitors and suggest that the onset of low-mass star formation is conditioned to the presence of dust in the parent clouds.

Dynamical collapse and fragmentation of low-metallicity cloud cores is studied using three-dimensional hydrodynamical calculations, with particular attention devoted whether the cores fragment in the dust-cooling phase or not. The cores become elongated in this phase, being unstable to non-spherical perturbation due to the sudden temperature decrease. In the metallicity range of 10^{-6}-10^{-5}Z_sun, cores with an initial axis ratio >2 reach a critical value of the axis ratio (>30) and fragment into multiple small clumps. This provides a possible mechanism to produce low-mass stars in ultra-metal-poor environments.

We study the evolution of gas in HII regions around the first stars after the death of the exciting stars. If the first star in a small halo dies without supernova (SN), subsequent star formation is possible in the same halo. We thus investigate the effect of ionization of the gas on subsequent star formation within small halos in the early universe using one-dimensional hydrodynamics with spherical symmetry along with non-equilibrium primordial gas chemistry. We find that the enhanced electron fraction facilitates the formation of molecular hydrogen at the cores of these halos. The low temperature circumstances produced by the H_2 cooling is suitable for HD formation and the resultant cooling further drops the temperature below 100 K. Consequently, low-mass stars with primordial abundances can form even in a small halo. After accreting the interstellar metals, these stars might resemble low-mass ultra metal-poor stars discovered in the present Galactic halo.

We study formation of molecules in primordial prestellar clumps and evaluate the line luminosities to assess detectability by next-generation facilities. If the initial H_2 fraction is sufficiently high, HD becomes an important coolant in the clumps. The luminosity from such HD cooling clumps is lower than that from H_2 cooling ones because of lower temperature (<100K). As for Li reactions, we include the three-body LiH formation approximately. The Li molecular fraction remains very low (<10^{-3}) throughout the evolution owing to the high dissociative reaction rate of LiH +H -> Li + H_2. LiH does not become an important coolant in any density range. The luminous emission lines from the prestellar cores include H_2 rovibrational lines: 1-0 Q(1), 1-0 O(3), 1-0 O(5), and pure rotational lines: 0-0 S(3), 0-0 S(4), 0-0 S(5). The next-generation facilities SPICA and JWST are able to detect H_2 emission in a large pre-galactic cloud that forms metal-free stars at a high rate of \sim 10^3 M_s/yr at redshift z<10. We also derive an analytical expression for the luminosity that reproduces the numerical results.

The thermal and chemical evolution of star-forming clouds is studied for different gas metallicities, Z, using the model of Omukai (2000), updated to include deuterium chemistry and the effects of cosmic microwave background (CMB) radiation. HD-line cooling dominates the thermal balance of clouds when Z \~ 10^{-5}-10^{-3} Z_sun and density ~10^{5} cm^{-3}. Early on, CMB radiation prevents the gas temperature to fall below T_CMB, although this hardly alters the cloud thermal evolution in low-metallicity gas. From the derived temperature evolution, we assess cloud/core fragmentation as a function of metallicity from linear perturbation theory, which requires that the core elongation E := (b-a)/a > E_NL ~ 1, where a (b) is the short (long) core axis length. The fragment mass is given by the thermal Jeans mass at E = E_NL. Given these assumptions and the initial (gaussian) distribution of E we compute the fragment mass distribution as a function of metallicity. We find that: (i) For Z=0, all fragments are very massive, > 10^{3}M_sun, consistently with previous studies; (ii) for Z>10^{-6} Z_sun a few clumps go through an additional high density (> 10^{10} cm^{-3}) fragmentation phase driven by dust-cooling, leading to low-mass fragments; (iii) The mass fraction in low-mass fragments is initially very small, but at Z ~ 10^{-5}Z_sun it becomes dominant and continues to grow as Z is increased; (iv) as a result of the two fragmentation modes, a bimodal mass distribution emerges in 0.01 < Z/Z_sun < 0.1. (v) For > 0.1Z_sun, the two peaks merge into a singly-peaked mass function which might be regarded as the precursor of the ordinary Salpeter-like IMF.

Molecular hydrogen line radiation emitted in formation events of first-generation stars are evaluated in a discussion of its detectability by future observational facilities. H_2 luminosity evolution from the onset of prestellar collapse until the formation of a \sim 100 M_{\odot} protostar is followed. Calculations are extended not only to the early phase of the runaway collapse but also to the later phase of accretion, whose observational features have not been studied before. Contrary to the runaway collapse phase, where the pure-rotational lines are always dominant, in the accretion phase rovibrational line emission becomes prominent. The maximum luminosity is also attained in the accretion phase for strong emission lines. The peak intensity of the strongest rovibrational line reaches \sim 10^{-29} (W/m^2), corresponding to the flux density of 10^{-5} (\mu Jy), for a source at the typical redshift of first-generation star formation, 1+z=20. Although the redshifted rovibrational H_2 emission from such an epoch falls in the wavelength range of the next-generation infrared satellite, Space Infrared Telescope for Cosmology and Astrophysics, for exceeding the detection threshold 10^7 such protostars are required to reach the maximum luminosity simultaneously in a pregalactic cloud. It is improbable that this condition is satisfied in a realistic scenario of early structure formation.

The initial mass function (IMF) of metal-free stars that form in the initial starburst of massive (virial temperatures >10^4K) metal-free protogalaxies is studied. In particular, we focus on the effect of H2 photodissociation by pre-existing stars on the fragmentation mass scale, presumedly determined by the Jeans mass at the end of the initial free-fall phase, i.e., at the so-called ``loitering phase,'' characterized by the local temperature minimum. Photodissociation diminishes the Jeans mass at the loitering phase, thereby reducing the fragmentation mass scale of primordial clouds. Thus, in a given cloud, far ultraviolet (FUV) radiation from the first star, which is supposedly very massive (about 10^3Msun), reduces the mass scale for subsequent fragmentation. Through a series of similar processes the IMF for metal-free stars is established. If FUV radiation exceeds a threshold level, the star-forming clumps collapse solely through atomic cooling. Correspondingly, the fragmentation scale drops discontinuously from a few time 10Msun to sub-solar scales. In compact clouds (>1.6kpc for clouds of gas mass 10^8Msun), this level of radiation field is attained, and sub-solar mass stars are formed even in a metal-free environment. Consequently, the IMF becomes bi-modal, with peaks at a few tenths of Msun and a few times 10Msun. The high-mass portion of the IMF is found to be a very steep function of the stellar mass, xi_high(m) being proportinal to m^{-5}. Therefore, the typical mass scale of metal-free stars is significantly smaller than that of the very first stars. Also we study the thermal instability in collapsing primordial prestellar clumps, and discuss why the thermal instability occuring during the three-body H2 formation does not appear to manifest itself in causing further fragmentation of such clumps.

We study the H2 cooling emission of forming galaxies, and discuss their observability using the future infrared facility SAFIR. Forming galaxies with mass >10^11 Msun emit most of their gravitational energy liberated by contraction in molecular hydrogen line radiation, although a large part of thermal energy at virialization is radiated away by the H Ly alpha emission. For more massive objects, the degree of heating due to dissipation of kinetic energy is so great that the temperature does not drop below 10^4 K and the gravitational energy is emitted mainly by the Ly alpha emission. Therefore, the total H2 luminosity attains the peak value of about 10^42 ergs/s for forming galaxies whose total mass 10^11 Msun. If these sources are situated at redshift z=8, they can be detected by rotational lines of 0-0S(3) at 9.7 micron and 0-0S(1) at 17 micron by SAFIR. An efficient way to find such H2 emitters is to look at the Ly alpha emitters, since the brightest H2 emitters are also luminous in the Ly alpha emission.

The earliest stars to form in the Universe were the first sources of light, heat and metals after the Big Bang. The products of their evolution will have had a profound impact on subsequent generations of stars. Recent studies of primordial star formation have shown that, in the absence of metals (elements heavier than helium), the formation of stars with masses 100 times that of the Sun would have been strongly favoured, and that low-mass stars could not have formed before a minimum level of metal enrichment had been reached. The value of this minimum level is very uncertain, but is likely to be between 10^{-6} and 10^{-4} that of the Sun. Here we show that the recent discovery of the most iron-poor star known indicates the presence of dust in extremely low-metallicity gas, and that this dust is crucial for the formation of lower-mass second-generation stars that could survive until today. The dust provides a pathway for cooling the gas that leads to fragmentation of the precursor molecular cloud into smaller clumps, which become the lower-mass stars.

The process of star formation from metal-free gas is investigated by following the evolution of accreting protostars with emphasis on the properties of massive objects. The main aim is to establish the physical processes that determine the upper mass limit of the first stars. Although the consensus is that massive stars were commonly formed in the first cosmic structures, our calculations show that their actual formation depends sensitively on the mass accretion rate and its time variation. Even in the rather idealized case in which star formation is mainly determined by dot{M}acc, the characteristic mass scale of the first stars is rather uncertain. We find that there is a critical mass accretion rate dot{M}crit = 4 10^{-3} Msun/yr that separates solutions with dot{M}acc< dot{M}crit in which objects with mass >> 100 Msun can form, provided there is sufficient matter in the parent clouds, from others (dot{M}acc > dot{M}crit) where the maximum mass limit decreases as dot{M}acc increases. In the latter case, the protostellar luminosity reaches the Eddington limit before the onset of hydrogen burning at the center via the CN-cycle. This phase is followed by a rapid and dramatic expansion of the radius, possibly leading to reversal of the accretion flow when the stellar mass is about 100Msun. (abridged)

Recent studies suggest that the initial mass function (IMF) of the first stars was likely to be extremely top-heavy, unlike what is observed at present. We propose a scenario to generate fragmentation to lower masses once the first massive stars have formed and derive constraints on the primordial IMF. We estimate the mass fraction of pair-unstable supernovae, shown to be the dominant sources of the first heavy elements. These metals enrich the gas up to about $10^{-5}$ solar metallicity, when a transition to efficient cooling-driven fragmentation occurs producing 1 solar mass clumps. We argue that the remaining fraction of the first stars ends up in 100 solar mass VMBHs (Very Massive Black Holes). We obtain constraints on the fraction of first stars that contribute to the initial metal enrichment and the transition redshift for primordial IMF away from a top-heavy one, by making various assumptions about the fate of these VMBHs at late times. We conclude with a discussion of several hitherto unexplored implications of a high-mass dominated star formation mode in the early Universe.

We analytically investigate the formation of an HII region in the accreting envelope of a newborn star. Special care is taken to examine the role of ionizing radiation force. This effect modifies velocity and density distributions and thereby affects the expansion of the HII region. As a result, the upper limit of the stellar mass imposed by the growth of an HII region around a forming star is increased by a larger factor than the previous estimate. In particular, for a star forming out of metal-free gas, this mechanism does not impose a firm upper limit on its mass.

We investigate the formation by accretion of massive primordial protostars in the range 10 to 300 Msun. The high accretion rate used in the models (4.4 x 10^{-3} Msun/yr) causes the structure and evolution to differ significantly from those of both present-day protostars and primordial zero-age main sequence stars. After an initial expansion of the radius (for < 12 Msun), the protostar undergoes an extended phase of contraction (up to 60 Msun). The stellar surface is not visible throughout most of the main accretion phase, since a photosphere is formed in the infalling envelope. Also, significant nuclear burning does not take place until a protostellar mass of about 80 Msun. As the interior luminosity approaches the Eddington luminosity, the protostellar radius rapidly expands, reaching a maximum around 100 Msun. Changes in the ionization of the surface layers induce a secondary phase of contraction, followed by a final swelling due to radiation pressure when the stellar mass reaches about 300 Msun. This expansion is likely to signal the end of the main accretion phase, thus setting an upper limit to the protostellar mass formed in these conditions.

Thermal and chemical evolution of primordial gas clouds irradiated with far-ultraviolet (FUV; < 13.6 eV) radiation is investigated. In clouds irradiated by intense FUV radiation, sufficient hydrogen molecules to be important for cooling are never formed. However, even without molecular hydrogen, if the clouds are massive enough, they start collapsing via atomic hydrogen line cooling. Such clouds continue to collapse almost isothermally owing to successive cooling by H^{-} free-bound emission up to the number density of 10^{16} cm^{-3}. Inside the clouds, the Jeans mass eventually falls well below a solar mass. This indicates that hydrogen molecules are dispensable for low-mass primordial star formation, provided fragmentation of the clouds occurs at sufficiently high density.

The thermal and chemical evolution of gravitationally collapsing protostellar clouds is investigated, focusing attention on their dependence on metallicity. Calculations are carried out for a range of metallicities spanning the local interstellar value to zero. During the time when clouds are transparent to continuous radiation, the temperatures are higher for those with lower metallicity, reflecting lower radiative ability. However, once the clouds become opaque, in the course of the adiabatic contraction of the transient cores, their evolutionary trajectories in the density-temperature plane converge to a unique curve that is determined by only physical constants. The trajectories coincide with each other thereafter. Consequently, the size of the stellar core at the formation is the same regardless of the gas composition of the parent cloud.

We study the H2 photodissociation regions around OB stars in primordial gas clouds whose virial temperatures are between a few hundred and a few thousand Kelvin. In such small objects, a single O star can photodissociate a mass equal to that of the cloud itself. As a result, the clouds deplete their molecular coolant and cannot cool in a free-fall time, and subsequent star formation is totally quenched. This indicates that stars do not form efficiently in small objects and that these objects contribute little to the reionization of the universe.

We investigate the thermal and dynamical evolution of primordial gas clouds in the universe after decoupling. Comparing the time-scale of dynamical evolution with that of fragmentation, we can estimate the typical fragmentation scale. We propose the following scenario of the formation process of first luminous objects consisting of large number stars. First, by pancake collapse of the overdensity regions in the expanding universe or collision between clouds in potential wells, quasi-plane shocks form. If the shock-heated temperature is higher than about $10^4$ K, the postshock gas cools down to several hundred K by \Hbun line cooling, and the shock-compressed layer fragments into filamentary clouds. The filamentary cloud collapses dynamically once more and fragments into cloud cores. Finally, a primordial star forms in a cloud core. We show that the minimum mass of the first star is essentially determined by the Chandrasekhar mass. Also, we investigate the dynamical collapse of cloud cores by numerical simulation and show that the evolution paths of the central regions of the cores depend only very weakly on the total core mass. After mass accretion, a massive star may be formed in a core, since the estimated mass accretion rate is very large. In such a case, it may be possible for many massive stars form almost simultaneously. Then the clouds can be luminous objects. On the other hand, if the shock-heated temperature is lower, effective star formation is delayed significantly.

The evolution of collapsing metal free protostellar clouds is investigated for various masses and initial conditions. We perform hydrodynamical calculations for spherically symmetric clouds taking account of radiative transfer of the molecular hydrogen lines and the continuum, as well as of chemistry of the molecular hydrogen. The collapse is found to proceed almost self-similarly like Larson-Penston similarity solution. In the course of the collapse, efficient three-body processes transform atomic hydrogen in an inner region of $\sim 1 M_{\sun}$ entirely into molecular form. However, hydrogen in the outer part remains totally atomic although there is an intervening transitional layer of several solar masses, where hydrogen is in partially molecular form. No opaque transient core is formed although clouds become optically thick to H$_{2}$ collision-induced absorption continuum, since H$_{2}$ dissociation follows successively. When the central part of the cloud reaches stellar densities ($\sim 10^{-2} {\rm g cm^{-3}}$), a very small hydrostatic core ($\sim 5 \times 10^{-3} M_{\sun}$) is formed and subsequently grows in mass as the ambient gas accretes onto it. The mass accretion rate is estimated to be $3.7 \times 10^{-2} M_{\sun} {\rm yr^{-1}} (M_{\ast}/M_{\sun})^{-0.37}$, where $M_{\ast}$ is instantaneous mass of the central core, by using a similarity solution which reproduces the evolution of the cloud before the core formation.

The contraction processes of metal-free molecular clouds of starlike mass (or cloud cores) are investigated. We calculate radiative transfer of the H_2 lines and examine quasi-static contraction with radiative cooling. Comparing two time-scales, the free-fall time t_ff and the time-scale of quasi-static contraction t_qsc (nearly equal to t_cool, the cooling time) of these cores, we find that the ratio of the two time-scales t_ff/t_qsc, i.e., the efficiency of cooling, becomes larger with contraction even under the existence of cold and opaque envelopes. In particular, for fragments of primordial filamentary clouds, for which t_ff is nearly equal to t_qsc at the fragmentation epoch, they collapse dynamically in the free-fall time-scale. This efficiency of cooling is unique to line cooling.