arXiv subjects

Floor van Donkelaar

Publications and source records attributed to Floor van Donkelaar.

9 recordsLinked to original sources

arXiv2026

Too shy to spin? Cosmic wallflowers as proto-globular clusters

We investigate the rotational properties of star-forming clusters at $z \sim 7.6$ in the high-resolution simulation MassiveBlackPS, focusing on two formation channels: clusters forming in galactic discs via gravitational instability and isolated circumgalactic systems, referred to as cosmic wallflowers, born out of cosmic filaments. Using stellar kinematics, we compare their rotational velocities, $v_{\rm rot}$, and rotational support, $v/σ$, to study whether formation environment leaves a clear dynamical imprint. We find a clear separation, wherein cosmic wallflowers systematically have lower rotational velocities and span a wide range in $v/σ$, whereas the identified disc clusters are strongly rotation-dominated and extend to higher $v_{\rm rot}$. When combined with stellar surface densities, a subset of the low-$v_{\rm rot}$ cosmic wallflowers lie surprisingly close to the observed globular cluster population in the Milky Way, whereas disc clusters remain offset. Within the cosmic wallflower population, we identify two regimes: lower-density, weakly rotating systems that overlap with these globular cluster properties, and denser, more rotationally supported systems that likely follow a different evolutionary pathway, possibly linking them to the origin of massive black hole seeds at high redshift. We further find that the gas content correlates with this behaviour, with gas-rich cosmic wallflowers preferentially occupying this low-rotation regime. This all suggests that environment and baryonic content together play a key role in setting the initial dynamical state and possible fate of clusters. In particular, weakly rotating, gas-rich cosmic wallflowers emerge as natural proto-globular cluster candidates, potentially evolving towards present-day systems through angular momentum loss and dynamical heating.

astro-ph.GA↗

arXiv2026

Cosmic wallflowers: the circumgalactic origins of isolated ultra-compact star clusters at $z>7$

The discovery of gravitationally lensed stellar clusters at high redshift with the James Webb Space Telescope (JWST) has revealed extremely compact, massive star-forming systems in galaxies at $z > 6$, providing a new window into early cluster formation. In this work, we investigate star cluster formation in the circumgalactic environments of gas-rich galaxies with stellar masses spanning between $\sim$$10^{8}$ - $10^{11}$ M$_{\odot}$ at $z > 7$, using the MassiveBlackPS cosmological hydrodynamical simulation with 2 pc resolution. We identify 55 baryon-dominated clusters forming outside galactic discs but within the virial radius of the primary halo. Star formation in these systems proceeds rapidly, reaching peak stellar surface densities above $10^{5}$ M$_{\odot}$ pc$^{-2}$, closely matching the compact clusters recently discovered by JWST in the lensed Cosmic Gems Arc at $z \approx 9.6$. Such extreme densities are a key pre-requisite to trigger runaway stellar collisions, indicating that a subset of our clusters would be a likely host of intermediate-mass black holes (IMBHs). We find that massive star clusters can form efficiently in the circumgalactic medium at early times through filament fragmentation, whereby high gas densities lead to rapid local collapse via a combination of thermal and gravitational instabilities. This formation pathway implies that some compact clusters formed in the quiet outskirts of forming galaxies rather than within their discs. Small variations in filament properties, including metallicity, density, and dark-matter content, influence the likelihood of a star cluster being able to form an IMBH seed. The formation of clusters in circumgalactic environments points to a potential evolutionary pathway connecting early off-disc clusters, present-day globular clusters, and the seeds of massive BHs.

astro-ph.GA↗

arXiv2025

Introducing the Phoebos simulation: galaxy properties at the dawn of galaxy formation

The James Webb Space Telescope (JWST) now allows us to observe galaxies at the end of cosmic dawn ($z \sim 10-15$) with unprecedented detail, revealing their morphologies, sizes, and internal structures. These observations offer crucial insights into the physical processes driving early galaxy formation. In this work, we introduce the Phoebos hydrodynamical cosmological simulation, a state-of-the-art 100 Mpc volume designed to study the formation and evolution of galaxies at the end of cosmic dawn and into the epoch of reionization. Phoebos includes a stellar feedback model that is intentionally weak, in order to address the high abundance of massive galaxies seen by JWST at early epochs. At variance with most large cosmological hydrodynamical simulations, we do not employ an effective equation of state model, instead our radiative cooling model allows us to capture the multi-phase nature of the gas inside and around galaxies. Phoebos reproduces key observables of early galaxy formation at $z \gtrsim 8$, including the stellar mass function and the stellar-to-halo mass relation. It also recovers the observed slope of the stellar size-to-mass relation and matches the specific star formation rate remarkably well. These results suggest that highly efficient star formation in the presence of only mild regulation from stellar feedback, drives early galaxy growth, supporting a scenario of rapid stellar mass assembly during cosmic dawn. There are indications in the cosmic star formation density that, at lower redshifts, Phoebos might overpredict the stellar mass within the systems, suggesting that a transition to a stronger stellar feedback may be necessary to reproduce later-time observations. These results highlight the potential of Phoebos to interpret JWST observations and to probe the evolving physical processes that shape galaxy formation.

astro-ph.GA↗

arXiv2025

Wandering intermediate-mass black holes in Milky Way-mass galaxies in cosmological simulations: myth or reality?

In this work, we address the following question: ``can we use the current cosmological simulations to identify intermediate-mass black holes (IMBHs) and quantify a putative population of wandering IMBHs?''. We compare wandering-IMBH counts in different simulations with different sub-grid methods and post-processing recipes, the ultimate goal being to aid future wandering-IMBH detection efforts. In particular, we examine simulations in which IMBHs are identified as BH seeds forming at high redshift and those in which they are identified using star clusters as proxies, which implicitly appeals to a stellar dynamical formation channel. In addition, we employ the extremely high-resolution cosmological hydrodynamical ``zoom-in'' simulation GigaEris with the star cluster proxies method to identify IMBHs. We find consistent counts of wandering high-redshift IMBHs across most of the different cosmological simulations employed so far in the literature, despite the different identification approaches, resulting in 5 to 18 wandering IMBHs per Milky Way-sized galaxy at $z \geq 3$. Nevertheless, we argue this is only coincidental, as a significant discrepancy arises when examining the formation sites and the mass ranges of the wandering IMBHs. Furthermore, we cannot determine how many of the IMBHs identified at high redshift in GigaEris will be wandering IMBHs at $z = 0$ as opposed to how many will accrete to the central supermassive BH, promoting its growth. All of this casts doubts on the ability of current cosmological simulations to inform observational searches for wandering IMBHs.

astro-ph.GA↗

arXiv2024

In-situ formation of star clusters at z > 7 via galactic disk fragmentation; shedding light on ultra-compact clusters and overmassive black holes seen by JWST

We investigate the nature of star formation in gas-rich galaxies at $z > 7$ forming in a markedly overdense region, in the whereabouts of a massive virialized halo already exceeding $10^{12}$ M$_{\odot}$. We find that not only the primary galaxy, but also the lower-mass companion galaxies rapidly develop massive self-gravitating compact gas disks, less than 500~pc in size, which undergo fragmentation by gravitational instability into very massive bound clumps. Star formation proceeds fast in the clumps, which quickly turn into compact star clusters with masses in the range $10^5$-$10^8$ M$_{\odot}$ and typical half-mass radii of a few pc, reaching characteristic densities above $10^5$ M$_{\odot}$ pc$^{-2}$. The properties of the clusters in the lowest-mass galaxy bear a striking resemblance to those recently discovered by the James Webb Space Telescope (JWST) in the lensed Cosmic Gems arc system at $z = 10.2$. We argue that, due to their extremely high stellar densities, intermediate-mass black holes (IMBHs) would form rapidly inside the clusters, which would then swiftly sink and merge on their way to the galactic nucleus, easily growing a $10^7$~M$_{\odot}$ supermassive black hole (SMBH). Due to the high fractional mass contribution of clusters to the stellar mass of the galaxies, in the range $20$-$40\%$, the central SMBH would comprise more than $10\%$ of the mass of its host galaxy, naturally explaining the overmassive SMBHs discovered by JWST at $z > 6$.

astro-ph.GA↗

arXiv2024

Exploring the fate of primordial discs in Milky Way-sized galaxies with the GigaEris simulation

Recent observations with JWST and ALMA have unveiled galaxies with regular discs at significantly higher redshifts than previously expected. This appears to be in contrast with constraints on the stellar populations of the Milky Way, suggesting that the bulk of the Galactic thin disc formed after $z=1$, and raises questions about the history, evolution, and survivability of primordial discs. Here, we use GigaEris, a state-of-the-art $N$-body, hydrodynamical, cosmological ``zoom-in'' simulation with a billion particles within the virial radius, to delve into the formation of the early kinematically cold discs (KCDs), defined by their ratio between the mean rotational velocity and the radial velocity dispersion, of a Milky Way-sized galaxy at redshifts $z\gtrsim 4$. Our analysis reveals a primarily inward migration pattern for disc stars formed at $z \gtrsim 6$, turning into a mix of inward and outward migration at later times. Stars migrating outwards undergo minimal kinematic heating, and might be identified as part of the thin disc forming at much later epochs. We find that approximately 76 per cent of all stars formed in the KCD at $z \sim 7$ become part of a pseudo-bulge by $z = 4.4$. This proportion decreases to below 10 per cent for KCD stars formed at $z \lesssim 5$. The inward migration of stars born in our KCDs at $z \gtrsim 4$ deviates from the expected inside-out formation scenario of thin discs at lower redshifts. Our results suggest a novel ``two-phase'' disc formation process, whereby the early disc transforms primarily into the pseudo-bulge within less than a billion years, whereas the present-day disc forms subsequently from higher-angular momentum material accreted at later times.

astro-ph.GA↗

arXiv2024

Stellar cluster formation in a Milky Way-sized galaxy at z>4 -- II. A hybrid formation scenario for the nuclear star cluster and its connection to the nuclear stellar ring

Nuclear star clusters (NSCs) are massive star clusters found in the innermost region of most galaxies. While recent studies suggest that low-mass NSCs in dwarf galaxies form largely out of the merger of globular clusters and NSCs in massive galaxies accumulate mass primarily through central star formation, the formation channel of the Milky Way's NSC is still uncertain. In this work, we use GigaEris, a high resolution N-body, hydrodynamical, cosmological ``zoom-in'' simulation, to investigate a possible formation path of the NSC in the progenitor of a Milky Way-sized galaxy, as well as its relation to the assembly and evolution of the galactic nuclear region. We study the possibility that bound, young, gas-rich, stellar clusters within a radius of 1.5 kpc of the main galaxy's centre at z>4 are the predecessors of the old, metal-poor stellar population of the Milky Way's NSC. We identify 47 systems which satisfy our criteria, with a total stellar mass of $10^{7.5}$ M$_{\odot}$. We demonstrate that both stellar cluster accretion and in-situ star formation will contribute to the formation of the NSC, providing evidence for a hybrid formation scenario for the first time in an N-body, hydrodynamical, cosmological ``zoom-in'' simulation. Additionally, we find that the gas required for in-situ star formation can originate from two pathways: gas-rich stellar clusters and gas influx driven by large-scale non-axisymmetric structures within the galaxy. This is partly supported by the presence of a stellar ring, resulting from gas dynamics, with properties similar to those of the Milky Way's nuclear stellar disc.

astro-ph.GA↗

arXiv2023

Stellar cluster formation in a Milky Way-sized galaxy at z>4 -- I. The proto-globular cluster population and the imposter amongst us

The formation history of globular clusters (GCs) at redshift $z > 4$ remains an unsolved problem. In this work, we use the cosmological, $N$-body hydrodynamical ``zoom-in'' simulation GigaEris to study the properties and formation of proto-GC candidates in the region surrounding the progenitor of a Milky Way-sized galaxy. The simulation employs a modern implementation of smoothed-particle hydrodynamics, including metal-line cooling and metal and thermal diffusion and allows to resolve systems at the scale of star clusters. We define proto-GC candidate systems as gravitationally bound stellar systems with baryonic mass fraction $F_{\rm b} \geq 0.75$ and stellar velocity dispersion $σ_{\star} < 20$ km s$^{-1}$. At $z=4.4$ we identify 9 systems which satisfy our criteria, all of which form between 10 kpc to 30 kpc from the centre of the main host. Their baryonic masses are in the range $10^5$- $10^7$ M$_{\odot}$. By the end of the simulation, they still have a relatively low stellar mass ($M_{\star} \sim 10^4$--$10^5$ M$_{\odot}$) and a metallicity ($-1.8 \lesssim {\rm [Fe/H]} \lesssim -0.8$) similar to the blue Galactic GCs. All of the identified systems except one appear to be associated with gas filaments accreting onto the main galaxy in the circum-galactic region, and formed at $z=5-4$. The exception is the oldest object, which appears to be a stripped compact dwarf galaxy that has interacted with the main halo between $z = 5.8$ and $z=5.2$ and has lost its entire dark matter content due to tidal mass loss.

astro-ph.GA↗

arXiv2021

From giant clumps to clouds -- II. The emergence of thick disc kinematics from the conditions of star formation in high redshift gas rich galaxies

High redshift disc galaxies are more gas rich, clumpier, and more turbulent than local Universe galaxies. This early era of galaxy formation imprints the distribution and kinematics of the stars that we observe today, but it is not yet well established how. In this work, we use simulations of isolated Milky Way-mass disc galaxies to study how kinematic properties of stars change when varying the gas fraction. This allows us to quantify the roles played by internal processes, e.g. gas turbulence and gravitational scattering off massive gas clumps, in establishing the observed stellar velocity dispersions and orbital eccentricities. We find that models with gas fractions $>20$ per cent feature a turbulent and clumpy interstellar medium (ISM), leading to zero-age stellar velocity dispersions $\sim 20-30~{\rm km\, s}^{-1}$ and high mean orbital eccentricities. Low eccentricities cannot arise from these physical conditions. For gas fractions below $20$ per cent, the ISM becomes less turbulent, with stellar velocity dispersions $<10~{\rm km\, s}^{-1}$, and nearly circular orbits for young stars. The turbulence present in gas-rich high redshift galaxies hence acts as a `barrier' against the formation of thin discs. We compare our findings to the Milky Way's age-velocity dispersion relation and argue that velocity dispersions imprinted already at star formation by the ISM contribute significantly at all times. Finally, we show that observed orbital eccentricities in the Milky Way's thick and thin discs can be explained entirely as imprints by the star-forming ISM, rather than by mergers or secular processes.

astro-ph.GA↗