arXiv subjects

K. Maltsev

Publications and source records attributed to K. Maltsev.

3 recordsLinked to original sources

arXiv2025

New gravitational-wave data support a bimodal black-hole mass distribution

Detailed stellar evolution and supernova models yield a bimodal black-hole mass distribution with a narrow peak around 10 solar masses from stars within a narrow range of progenitor properties and a second broader peak starting around 20 solar masses from very massive progenitors. This bimodal black-hole mass distribution leads to a characteristic distribution of chirp masses of merging binary black holes, with two main peaks arising from the merger of two black holes where both come either from the low- or the high-mass peak and a smaller peak in between from the mixed merger of a low-mass and a high-mass black hole. We carry out a population synthesis study of binary black hole formation and compare the results to the observed chirp masses of gravitational-wave events. We find that only the bimodal black-hole mass prescription is able to reproduce the structure of peaks and gaps in the observed chirp-mass distribution, which is not matched by predictions from other remnant mass prescriptions in the literature.

astro-ph.HE↗

arXiv2025

Explodability criteria for the neutrino-driven supernova mechanism

Massive stars undergoing iron core-collapse at the end of their evolution terminate their lives either in successful or failed supernovae (SNe). The physics of core-collapse supernovae (CCSNe) is complex, and their understanding requires computationally expensive simulations. Using these to predict CCSN outcomes over large, densely sampled parameter spaces of SN progenitors, as is needed e.g. for population synthesis studies, is thus not feasible. To remedy this situation, we present explodability criteria that allow us to predict the final fates of stars by evaluating stellar structure variables at the onset of core-collapse. The criteria are calibrated to predictions of a semi-analytical SN model, evaluated over a set of $\sim$~3,900 heterogeneous stellar progenitors (single, binary-stripped and accretor stars). Over these, the criteria achieve an accuracy of >99\% agreement with the semi-analytical model. The criteria are tested on 29 state-of-the-art 3D CCSN simulation outcomes from two different groups. Furthermore, we find that all explodability proxies needed for our pre-SN structure-based criteria have two distinct peaks and intervening valleys as a function of the carbon-oxygen (CO) core mass $M_\mathrm{CO}$, which coincide with failed and successful SNe, respectively. The CO core masses of explodability peaks shift systematically with metallicity, $Z$, and with timing of hydrogen-rich envelope removal by binary mass transfer. With these, we identify critical values in $M_\mathrm{CO}$ that define windows over which black holes form by direct collapse and formulate a CCSN recipe based on $M_\mathrm{CO}$ and $Z$, applicable for rapid binary population synthesis and other studies. Our explodability formalism is consistent with observations of Type~IIP, IIb/Ib and Ic supernova progenitors and partially addresses the missing Red Supergiant Problem by direct black hole formation.

astro-ph.SR↗

arXiv2023

Scalable stellar evolution forecasting: Deep learning emulation vs. hierarchical nearest neighbor interpolation

Many astrophysical applications require efficient yet reliable forecasts of stellar evolution tracks. One example is population synthesis, which generates forward predictions of models for comparison with observations. The majority of state-of-the-art rapid population synthesis methods are based on analytic fitting formulae to stellar evolution tracks that are computationally cheap to sample statistically over a continuous parameter range. The computational costs of running detailed stellar evolution codes, such as MESA, over wide and densely sampled parameter grids are prohibitive, while stellar-age based interpolation in-between sparsely sampled grid points leads to intolerably large systematic prediction errors. In this work, we provide two solutions for automated interpolation methods that offer satisfactory trade-off points between cost-efficiency and accuracy. We construct a timescale-adapted evolutionary coordinate and use it in a two-step interpolation scheme that traces the evolution of stars from ZAMS all the way to the end of core helium burning while covering a mass range from ${0.65}$ to $300 \, \mathrm{M_\odot}$. The feedforward neural network regression model (first solution) that we train to predict stellar surface variables can make millions of predictions, sufficiently accurate over the entire parameter space, within tens of seconds on a 4-core CPU. The hierarchical nearest-neighbor interpolation algorithm (second solution) that we hard-code to the same end achieves even higher predictive accuracy, the same algorithm remains applicable to all stellar variables evolved over time, but it is two orders of magnitude slower. Our methodological framework is demonstrated to work on the MIST (Choi et al. 2016) data set. Finally, we discuss the prospective applications of these methods and provide guidelines for generalizing them to higher dimensional parameter spaces.

astro-ph.SR↗