🔭 astrophysics

Massive stellar cannibals: How stellar mergers drive mass-loss in extremely massive stars

This paper demonstrates that stellar mergers involving extremely massive stars and smaller inspiraling companions drive significant mass loss (approximately 10–30% of the system's mass) by converting orbital energy into thermal energy, a critical process that must be accounted for when modeling the formation and evolution of supermassive stars.

Original authors: J. Roman-Garza, T. Fragos, C. Charbonnel, L. Ramírez-Galeano, M. Kruckow, E. Farag

Published 2026-02-03

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: J. Roman-Garza, T. Fragos, C. Charbonnel, L. Ramírez-Galeano, M. Kruckow, E. Farag

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Cosmic Gluttons

Imagine a dense star cluster as a crowded dance floor. In the center, the biggest dancers (stars) tend to sink to the middle, while smaller ones stay on the edges. Occasionally, two stars get too close and crash into each other.

This paper focuses on a specific type of crash: a giant star (called an "Extremely Massive Star" or EMS, weighing thousands of times more than our Sun) swallowing a smaller star (like a 10 to 70-sun star).

The big question the authors asked is: When the giant eats the smaller star, does it get bigger, or does it spit some of itself back out?

The Setup: A 1D "Virtual Lab"

Simulating a star crash in 3D is like trying to predict the exact path of every single drop of water in a tsunami; it's incredibly hard and takes too much computer power.

Instead, the authors used a clever shortcut. They used a 1D model (think of it as looking at a star as a series of nested onion layers rather than a 3D ball). They adapted a famous star-simulation tool called MESA to act like a "stellar crash test dummy." They programmed it to simulate a smaller star spiraling inward into the giant star, like a satellite falling into a planet's atmosphere.

The Process: The "Hot Air" Effect

Here is what happens during the crash, step-by-step:

The Spiral: The smaller star doesn't just hit the giant; it spirals down into the giant's outer layers (the "envelope").
The Friction: As the smaller star plows through the giant's gas, it creates massive friction (drag). This is like rubbing your hands together quickly to generate heat.
The Heat Injection: This friction turns the smaller star's orbital energy into heat. This heat gets dumped into the giant star's outer layers.
The Pulsation (The "Breathing"): The giant star is already very unstable (it's barely holding itself together). When you dump a huge amount of heat into it, it doesn't just sit there; it starts to pulsate.
- Analogy: Imagine a very loose, over-inflated balloon. If you suddenly blow hot air into it, it doesn't just get bigger; it starts to wobble and shake violently.
The Ejection: The shaking gets so violent that the outer layers of the giant star are thrown off into space. The star literally "coughs up" some of its own mass.

The Results: How Much Mass is Lost?

The authors ran simulations with different sizes of "eaters" (1,000, 3,000, and 5,000 solar masses) and different sizes of "food" (10 to 70 solar masses).

The Outcome: The giant star loses a significant amount of mass. Depending on the size of the crash, the star ejects between 1% and 30% of the total system's mass.
The Mechanism: Most of the energy from the crash isn't used to make the star hotter or denser; it is used almost entirely to blow mass away.
Why? The authors explain that these giant stars are like "marginal" structures. Their internal physics is so close to the edge of collapse that a little push (the heat from the crash) is enough to make them unstable and fling material outward.

The "Recipe" for Future Scientists

The paper provides a "recipe" (a set of rules) for other scientists who simulate star clusters.

The Rule: If you want to know if two stars will merge or just bounce off each other, don't just look at their size. Look at their orbit.
The Threshold: If the smaller star is on a very tight, circular path, it will merge. If it's on a wide, elliptical path, it might just graze the giant and fly away (a "scattering" event).
The Correction: Previous models assumed these giant stars wouldn't lose much mass during a crash. This paper says, "Actually, they lose a lot." If you ignore this mass loss, your models of how black holes form will be wrong.

The Bottom Line

When a massive star "cannibalizes" a smaller one, it doesn't just grow bigger. The process is so violent and energetic that the giant star actually shrinks by blowing off a chunk of its own skin.

This is a crucial piece of the puzzle for understanding how the universe creates the most massive black holes and how the chemical makeup of the early universe was polluted by these stellar giants. The authors conclude that this "mass loss" is a non-negotiable part of the story; you can't understand these cosmic giants without accounting for the fact that they are messy eaters who spill their food everywhere.

Technical Summary: Massive Stellar Cannibals: How Stellar Mergers Drive Mass-Loss in Extremely Massive Stars

Problem Statement
The formation of extremely massive stars (EMSs, $10^3 \le M < 10^4 M_\odot$ ) and supermassive stars (SMSs, $M \ge 10^4 M_\odot$ ) in dense, low-metallicity stellar environments is theorized to occur via stellar mergers. These objects are critical candidates for intermediate-mass black hole progenitors and chemical polluters in the early universe. However, a self-consistent theory for their dynamical formation requires a precise understanding of mass accretion and loss processes. While previous studies have investigated collision-induced mass loss for lower-mass stars ( $1\text{--}50 M_\odot$ ) or provided semi-analytical prescriptions for EMSs, there has been a lack of hydrodynamic frameworks specifically addressing the merger-induced mass loss in the $>1000 M_\odot$ regime. The central challenge lies in modeling the complex interplay between orbital energy dissipation, hydrodynamic drag, and the structural response of an EMS, which is marginally bound and radiation-pressure dominated.

Methodology
The authors employed a 1D hydrodynamic framework adapted from the StellarInspiral1D code, integrated as an external module with the Modules for Experiments in Stellar Astrophysics (MESA) code.

Stellar Models: EMS models of $1000$, $3000$, and $5000 M_\odot$ were constructed via high cold-gas accretion rates (following Ramírez-Galeano et al. 2025) at a metallicity of $Z=10^{-4}$ . These models were relaxed using a Newtonian hydrodynamic solver followed by a Riemann solver to ensure numerical stability before interaction.
Simulation Setup: The simulations modeled the inspiral of companions with masses $10\text{--}70 M_\odot$ (mass ratios $q < 0.1$ ) into the EMS envelopes. The orbital evolution was governed by gravitational and hydrodynamic drag forces (following Ostriker 1999; Binney & Tremaine 2008), with the work done by drag deposited as thermal energy in the EMS envelope.
Hydrodynamics: The study utilized MESA's implicit Riemann solver for inviscid fluids to resolve contact discontinuities and shocks without artificial viscosity, which could otherwise dampen kinetic energy and suppress mass loss.
Termination Criteria: Simulations halted when the companion filled its Roche lobe or exceeded the companion's virial temperature, defining the "merger event."

Key Contributions and Results

Mechanism of Mass Loss: The study demonstrates that as a companion spirals into an EMS, the deposited orbital energy drives radial pulsations in the EMS envelope. These pulsations accelerate the stellar surface; when the surface velocity exceeds the local escape velocity, mass is ejected.
Quantification of Mass Loss:
- The total ejected mass is estimated to be $\sim 10\text{--}30\%$ of the system's total mass ( $M_{EMS} + M_{comp}$ ), depending on the companion mass.
- For a $1000 M_\odot$ EMS merging with a $10 M_\odot$ companion, the total escaping mass is estimated at $\sim 5 M_\odot$ .
- For a $1000 M_\odot$ EMS merging with a $70 M_\odot$ companion, the unbound mass reaches $\gtrsim 100 M_\odot$ , with a theoretical upper limit of $\sim 300 M_\odot$ based on energy conservation.
- The mass-loss rates scale with companion mass: $\sim 10^{-3}\text{--}10^{-1} M_\odot \text{ yr}^{-1}$ for low-mass companions and $\gtrsim 10^{-1}\text{--}10 M_\odot \text{ yr}^{-1}$ for high-mass companions.
Energy Budget: The results indicate that the majority of the orbital energy deposited by the inspiral is utilized to eject mass rather than being radiated away. This efficiency is attributed to the EMS structure being barely bound, with a mean adiabatic exponent $\gamma \approx 4/3$ , rendering them dynamically unstable to perturbations.
Comparison with Analytical Models: The hydrodynamic results predict higher mass loss than the semi-analytical prescription by Ramírez-Galeano et al. (2025), particularly for low-mass companions where the analytical model predicted negligible loss. The discrepancy arises because the analytical model assumes no mass loss if the injected energy is lower than the local binding energy, whereas the hydrodynamic simulations show that pulsations can still drive ejection.
Orbital Outcomes: Using a static EMS profile approximation for eccentric orbits, the authors classified interaction outcomes (mergers vs. scatterings). They found that for initially unbound orbits ( $e > 1$ ), a merger requires a semi-latus rectum $\ell \lesssim 0.1 R_{EMS}$ . For bound orbits, mergers are guaranteed unless external perturbations intervene.

Significance and Claims
The paper concludes that merger-induced mass loss is a non-negligible process for the formation and evolution of extremely massive stars. The findings suggest that previous models may have underestimated the mass loss during the dynamical assembly of EMSs and SMSs. By providing a prescription derived from hydrodynamic simulations, the authors offer a method to estimate mass loss and interaction outcomes (merger vs. scattering) that can be implemented in larger-scale dynamical simulations of stellar clusters. The study emphasizes that the efficiency of mass ejection is intrinsically linked to the marginally bound nature of these stars, a factor that must be accounted for when predicting the final masses of these objects and their potential roles as progenitors of intermediate-mass black holes or chemical polluters throughout cosmic history.