Periodic Reporting for period 1 - FIRSTBHs (The formation of supermassive black holes in the early universe)
Reporting period: 2015-10-01 to 2017-09-30
Supermassive black holes of more than a billion solar masses have been observed when the Universe was less than a billion years old. As a reference, the entire mass in stars in the Milky Way is only about ten times larger, the black hole at the center of our galaxy is about 4 million solar masses, and the Universe is now almost 14 billion years old. Explaining the existence of the supermassive black holes at these early cosmic times is highly challenging, as they need to grow continuously during the chaotic early phases of the cosmos. Seed black holes with masses of 10,000-1,000,000 solar masses, as predicted by the “direct collapse” scenario, provide a promising pathway to explain their presence. In this project we simulated the formation and growth of these seeds, following the dynamics of the gas from cosmological scales down to scales comparable to the solar system. The gravitational collapse initially forms a rapidly accreting protostar, which subsequently moves to the main sequence and finally collapses to a massive black hole. In the first part of the project, we computed the properties of the forming supermassive stars and black holes. Subsequently, the simulations have been extended by incorporating ultraviolet and X-ray radiation from the black hole itself to assess its growth. This is a substantial step forward to understand the origin of high-redshift black holes.
This work hinged for the most part on numerical simulations: from cosmological simulations to idealized ones. In all cases the simulations contain a detailed chemical network. This is important because this study focuses on the first stars, forming from gas at primordial composition, i.e. as created by Big Bang nucleosynthesis. In this case the various species of Hydrogen (atomic, molecular, H-) dominate the energy budget. Many of the energy transitions are weak, and therefore a full chemical network is required in order to capture correctly the evolution of gas and its ability to collapse to the densities required for formation of stars. The simulations focused on mainly three questions, as follows.
(i) What are the properties of the supermassive stars that give rise to black hole seeds?
We performed simulations of the collapse of gas in primordial halos to unprecedented densities, down to scales below the typical stellar radius, in full cosmological context. Our results show that gas cools and fragments into clumps, but these clumps are expected to merge on short time-scales creating a supermassive star, or two, in a binary system.
(ii) What are the conditions under which supermassive stars and black hole seeds can form?
Direct collapse black holes form in pristine halos, i.e. at primordial chemical composition. In order to create a massive black hole seed, the host halo needs to be prevented from forming stars before the collapse happens. A crucial ingredient is the suppression of the formation of molecular hydrogen, conducive to extensive star formation. Suppression of molecular hydrogen can be accomplished by the presence of ultraviolet radiation that dissociates molecules. A key uncertainty is the escape fraction of ultraviolet radiation from the first galaxies. We have performed radiation-hydrodynamical simulations of first galaxies and found that the escape fraction depends on the propagation of the ionization front, and it can be as small as 3 per cent or as large as 100 per cent. We have also studied the effect of ionizing radiation, finding that it helps gas collapse, and of dust, identifying a critical upper limit that still allows for supermassive stars to form.
(iii) How do black hole seeds evolve?
We have worked on this using semi-analytical models, hydrodynamical simulations and radiation-hydrodynamical simulations. We have found that the halos where direct collapse black holes can form are rare, as they require very special conditions in their environment. We derived the number density of direct collapse black holes and investigate the discrepancies between different results in the literature. We are now completing the analysis of a radiation-hydrodynamical simulation where we study the growth of a black hole seed under the effect of feedback from stars and from its own power.
(i) What are the properties of the supermassive stars that give rise to black hole seeds?
We performed simulations of the collapse of gas in primordial halos to unprecedented densities, down to scales below the typical stellar radius, in full cosmological context. Our results show that gas cools and fragments into clumps, but these clumps are expected to merge on short time-scales creating a supermassive star, or two, in a binary system.
(ii) What are the conditions under which supermassive stars and black hole seeds can form?
Direct collapse black holes form in pristine halos, i.e. at primordial chemical composition. In order to create a massive black hole seed, the host halo needs to be prevented from forming stars before the collapse happens. A crucial ingredient is the suppression of the formation of molecular hydrogen, conducive to extensive star formation. Suppression of molecular hydrogen can be accomplished by the presence of ultraviolet radiation that dissociates molecules. A key uncertainty is the escape fraction of ultraviolet radiation from the first galaxies. We have performed radiation-hydrodynamical simulations of first galaxies and found that the escape fraction depends on the propagation of the ionization front, and it can be as small as 3 per cent or as large as 100 per cent. We have also studied the effect of ionizing radiation, finding that it helps gas collapse, and of dust, identifying a critical upper limit that still allows for supermassive stars to form.
(iii) How do black hole seeds evolve?
We have worked on this using semi-analytical models, hydrodynamical simulations and radiation-hydrodynamical simulations. We have found that the halos where direct collapse black holes can form are rare, as they require very special conditions in their environment. We derived the number density of direct collapse black holes and investigate the discrepancies between different results in the literature. We are now completing the analysis of a radiation-hydrodynamical simulation where we study the growth of a black hole seed under the effect of feedback from stars and from its own power.
This project went beyond the state of the art in many respects. First, the simulations include a full chemical network, and we added additional reactions and dust. Second, we used radiation-hydrodynamical simulations, which means that we treated photons directly, albeit in a simplified way. Radiation-hydrodynamical simulations are very expensive and only a limited number of groups have started using them. The techniques developed for this study are at the vanguard of the field, and can now be shared with other scientists to improve their own codes.