The role of major mergers in triggering super-Eddington accretion

arXiv:2606.04081 · astro-ph.GA, astro-ph.CO, astro-ph.HE · Submitted 2026-06-02 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: We're looking at the paper: "The role of major mergers in triggering super-Eddington accretion".

Jocelyn: Here's the gist of the paper. The gist: A major merger between two relatively low-mass halos at high redshift can trigger episodes of sustained super-Eddington accretion, but this requires the complete switching off of black hole feedback for substantial growth in these systems.

Vera: Context and Motivation JWST observations have revealed black holes with masses of several million solar masses already at z > 8, challenging our understanding of their growth mechanisms The role of major mergers in triggering super-Eddington accretion.

Jocelyn: In this work, we investigate whether a major merger between two relatively low-mass halos (Mhalo ∼ 109 M⊙) at high redshift can trigger episodes of sustained SE accretion, with particular focus on the role of BH feedback Since the advent of the James Webb Space Telescope (JWST), the presence of supermassive black holes (SMBHs) at very high redshift has challenged previous models of BH formation and accretion Simulation Setup and Accretion Modeling The simulations were performed with the code gizmo (Hopkins 2015), which employs the same domain decomposition and N-body algorithms as its predecessor gadget (Springel 2005) Accretion is modelled using the Bondi-Hoyle-Lyttleton (BHL; Hoyle & Lyttleton 1939; Bondi & Hoyle 1944) formula The accretion rate is computed by kernel-averaging the individual contributions from the gas particles within this kernel The BH kernel is defined as the region enclosing its 96 nearest gas neighbours Black Hole Feedback Mechanisms The code implements a physically motivated prescription for three accretion regimes classified by the Eddington ratio, λ = M˙BH/M˙Edd These three regimes are – λ < 1.25 × 10−3: advection-dominated accretion flow (ADAF), – 1.25 × 10−3 ≤ λ < 1: sub-Eddington radiatively efficient accretion, – λ ≥ 1: SE accretion The prescriptions are particularly relevant to investigate the interplay between the SMBH and the host galaxy in cosmological zoom-in simulations Results on Merger Impact and Feedback For the relatively low-mass halos studied here, our feedback prescription efficiently suppresses gas accretion, preventing substantial BH growth We find that, although the merger drives gas inflows towards the central regions, this is not sufficient to trigger sustained SE accretion Post-merger SE accretion episodes are observed only when BH feedback is entirely switched off Amongst the feedback channels considered, kinetic feedback is the primary mechanism regulating BH growth Moreover, the only significant SE accretion episodes occur immediately after BH seeding, while the merger itself does not produce a substantial enhancement of the accretion rate In order to isolate the role of BH feedback in our simulations, we performed a simulation in which both kinetic and radiative feedback were disabled When no BH feedback is considered, we observe a significant growth in BH mass, particularly for the BH with an initial mass of 104 M⊙, which increases to ∼ 2 × 105 M⊙ following a substantial SE accretion episode occurring at the time of seeding The results presented in Sections 4.3 and 4.4 indicate that BH feedback is the primary mechanism regulating gas accretion onto BHs in these systems Even when extending the simulations to z = 8 to capture possible delayed effects of the merger, we do not find sustained SE accretion over long time-scales The merger itself does trigger gas inflows towards the central regions and increases the gas density, as indicated by the sharp rise in stellar mass at z ∼ 11 in Figure 2 However, this is not sufficient to produce a sustained enhancement in the accretion rate, for two reasons First, the time-scale required to build up a high-density gas reservoir around the BH is not short, and in such low-mass haloes even a single SN explosion can disperse the gas before sustained accretion is established Second, in the other runs, radiation pressure and kinetic feedback partially counteract the compression of the gas, preventing the central density around the BH from reaching the values required for strong accretion The results presented in Sections 4.3 and 4.4 indicate that BH feedback is the primary mechanism regulating gas accretion onto BHs in these systems In particular, the strength of the kinetic feedback plays a crucial role When the kinetic component is neglected

see panels (c) and (f) in Figure 6: , accretion rates increase and persist over longer time-scales

∼103 kyr, see panel (c) Figure 7: , and BHs grow significantly, reaching masses of 105−6 M⊙ When kinetic feedback is included, the initial SE phase is rapidly quenched and BH growth remains strongly suppressed The behaviour observed in runs including kinetic feedback is directly linked to the adopted prescription Future Directions Another important result of our simulations is that the only significant SE accretion episodes occur immediately after BH seeding If fEdd ≫ fc, then ϕ ∼ ϕMAD (see Equations 8–9), so β becomes a function of just the spin a; for a = 0.7, β ≃ 138, whereas for a = 0.2, β ≃ 7 A simple energetic argument further illustrates the strength of this feedback For very high accretion rates (fEdd >> fc), the jet efficiency becomes primarily a function of the BH spin, ηjet(a) (see Equations 8, 9, 10, and 11) Under these assumptions, the injected energy can easily exceed the binding energy of the host halo An SE burst with λ ∼ 10–100, lasting even a few kyr, can therefore inject energy comparable to the binding energy of the halo into the gas surrounding the BH, potentially ejecting the gas from the halo The only exception remains Run4

panel (d) of Figure 4: , in which all forms of feedback are neglected however, in this case SE accretion is stopped by SN explosions The merger itself does trigger gas inflows towards the central regions and increases the gas density, as indicated by the sharp rise in stellar mass at z ∼ 11 in Figure 2 However, this is not sufficient to produce a sustained enhancement in the accretion rate, for two reasons First, the time-scale required to build up a high-density gas reservoir around the BH is not short, and in such low-mass haloes even a single SN explosion can disperse the gas before sustained accretion is established Second, in the other runs, radiation pressure and kinetic feedback partially counteract the compression of the gas, preventing the central density around the BH from reaching the values required for strong accretion The behaviour observed in runs including kinetic feedback is directly linked to the adopted prescription Although physically motivated, as it is based on the Blandford-Znajek mechanism (Blandford & Znajek 1977), it still presents some limitations In particular, while we included a variable magnetic flux (Equation 8), the BH spin was kept fixed This parameter is crucial for determining the efficiency of kinetic feedback (see Equation 10) A low-spin BH undergoing SE accretion may spin down through the extraction of rotational energy, potentially leading to an initial phase more similar to the case without kinetic feedback, or even to hyper-Eddington accretion (Inayoshi et al.

Vera: 2016) The behaviour observed in runs including kinetic feedback is directly linked to the adopted prescription Although physically motivated, as it is based on the Blandford-Znajek mechanism (Blandford & Znajek 1977), it still presents some limitations In particular, while we included a variable magnetic flux (Equation 8), the BH spin was kept fixed This parameter is crucial for determining the efficiency of kinetic feedback (see Equation 10) A low-spin BH undergoing SE accretion may spin down through the extraction of rotational energy, potentially leading to an initial phase more similar to the case without kinetic feedback, or even to hyper-Eddington accretion (Inayoshi et al.

Title and authors: Vera: So, we’re looking at this paper titled "The role of major mergers in triggering super-Eddington accretion." It’s about how merging two relatively small halos can cause black holes to start feeding gas way faster than the Eddington limit.

Jocelyn: Right, it seems like they’re testing whether that merger event is enough to kick a black hole into an episode of super-Eddington accretion, focusing on those low-mass halos at high redshift.

Subrahmanyan: From my side, I'm interested in the idea of triggering sustained super-Eddington accretion because that would be a significant mechanism for growth in these early systems.

Vera: Exactly. The paper investigates if a major merger between two halos with masses around ten nine M at high redshift can actually drive those episodes, and it really focuses on the role of black hole feedback there <ref:2606.04081#pg1>.

Jocelyn: And they use a simulation setup based on gizmo, which is pretty standard for these kinds of cosmological zoom-in simulations.

Subrahmanyan: The methodology seems to be using the Bondi-Hoyle-Lyttleton formula to model accretion rates, computed by kernel-averaging the gas particles within a region defined by the black hole’s ninety-six nearest neighbors <ref:2606.04081#pg1>.

Vera: And then they introduce these three accretion regimes based on the Eddington ratio lambda, lambda = BH / Edd, which are advection-dominated accretion flow, sub-Eddington radiatively efficient accretion, and super-Eddington accretion where lambda is greater than one.

Jocelyn: That feedback prescription is pretty physically motivated, classifying the flow based on how fast it's growing compared to the Eddington limit.

Subrahmanyan: The simulations then test these regimes by looking at the interplay between the black hole and its host galaxy in cosmological zoom-in simulations, which is where I see most of my interest lie.

The paper's summary: Vera: So, we’ve seen how the merger drives gas in, but now we need to talk about what that actually means for sustained growth when you look at the whole picture of this paper on "The role of major mergers in triggering super-Eddington accretion."

Jocelyn: Basically, they’re saying that even if you have a big collision between two relatively small halos, it isn't automatic that the black hole will just start eating gas at a super-Eddington rate for a long time.

Subrahmanyan: That’s the main point from their theoretical side, right? They’re focusing on those low-mass halos at high redshift, which is where we expect to see these early seeds of growth.

Vera: Exactly. They found that the merger does increase the gas density toward the center—you see a sharp rise in stellar mass at redshift z equals eleven in their figures—but that compression isn't enough to keep the accretion rate high and steady over time.

Jocelyn: They pointed out two main reasons why it doesn't sustain itself: first, the time required for that gas reservoir to build up around the black hole is quite long, and even in these small halos, a single supernova can blow that gas away before it ever gets stable.

Subrahmanyan: And the second reason is that in other simulation runs where you include feedback mechanisms like radiation pressure or kinetic feedback, those forces actually fight against the compression of the gas around the black hole.

Vera: So what this means for us observing these systems is that we can't just look at a merger and immediately expect to see a long period of super-Eddington growth; you need to see if those gas reservoirs can actually stay dense enough.

Jocelyn: They did show that the only way they got sustained super-Eddington accretion was when they completely turned off all black hole feedback in the simulation, which is a big caveat for what we see in reality.

Subrahmanyan: That highlights how critical that feedback prescription is; it's not just a footnote, it's the regulator of whether you get growth or not.

Vera: It suggests that for these early black holes to grow fast, they need a very specific environment where the gas can be held in place, and without that specific mechanism—the feedback—it’s just a burst then nothing.

Jocelyn: It also shows us that those initial super-Eddington phases are really just fleeting events, happening right after the black hole is first seeded, not something that lasts for millions of years.

Subrahmanyan: So looking at the big picture of structure formation, it means the merger is a trigger for gas flow, but it’s not a guarantee for sustained high accretion without these specific physical limitations being overcome.

Vera: And this leads us to think about what happens if we change those feedback rules in our models. We can test how much stronger the gas needs to be before that suppression kicks in.

Jocelyn: If you want to settle whether these mergers are the main driver, we need follow-up observations that can measure that gas density profile closer to those high redshift epochs.

The paper's improvements: Vera: So, we've covered the paper's main findings on mergers and feedback regulating super-Eddington accretion, but now we look at what they suggest to fix their simulation setup or improve their understanding.

Jocelyn: They propose a few ways to make these simulations more realistic so the results don't just depend on the specific rules they chose for those feedback mechanisms.

Subrahmanyan: It seems like one big direction is incorporating physically motivated kinetic feedback models, like jets or winds, across all accretion regimes instead of just picking one prescription based on the Eddington ratio lambda.

Vera: That makes sense because it moves away from a purely mathematical classification to something that actually mimics how energy is injected into the surrounding gas flow during an accretion event.

Jocelyn: They also suggest refining how they calculate the accretion rate itself, moving beyond just the Bondi-Hoyle-Lyttleton formula by using a kernel that resolves the actual size of a jet at its base instead of just assuming a fixed volume.

Subrahmanyan: That would allow the AI to model how kinetic feedback actually depletes or ejects mass from the accretion flow more accurately, which is something they felt was missing in their current setup.

Vera: From a structure formation standpoint, that refinement could help us understand if those turbulence-driven environments in low-mass halos really dictate the growth history of the black hole.

Jocelyn: Another improvement they touch on involves making the coupling between radiation and chemistry more robust by explicitly accounting for AGN Compton heating in the chemical network, using specific temperatures for soft and hard X-rays.

Subrahmanyan: That would let us model how those radiative processes actually influence the thermal state of gas reservoirs around black holes, which is a bit more complex than just tracking mass.

Vera: So these changes are focused on getting the physics of the interaction between the black hole and its environment tighter, which should give us better constraints on whether mergers are actually powerful enough to sustain super-Eddington growth in those early times.

Jocelyn: And if the kinetic feedback is indeed the key regulator, then any future time-domain observations that show sustained accretion will be a strong signal that those jets or winds are actively shaping the growth of the system.

Conclusion: Vera: So, to wrap up on "The role of major mergers in triggering super-Eddington accretion," they're telling us that sustained super-Eddington growth in these early black holes requires completely turning off black hole feedback mechanisms.

Jocelyn: Right, so the merger alone isn't enough to keep things going; you need that feedback mechanism switched off for those long-term episodes to happen.

Subrahmanyan: It means that the physics of gas accretion onto a black hole is more about the balance between inflow and outflow than just a simple gravitational trigger from a collision.

Vera: Exactly, and they found that kinetic feedback is really what dictates whether you see sustained growth or if it just fizzles out quickly.

Jocelyn: It’s a pretty strong finding because it connects the merger event directly to the physical processes governing how gas behaves near the black hole's event horizon.

Subrahmanyan: If we want to find these objects at redshift z equals eight, we need observations that can tell us what kind of feedback is actually active in those dense environments.

Vera: That makes sense because our current observational constraints are often limited by how well we model that feedback, and this paper points out exactly where the uncertainty lies.

Jocelyn: And they suggest that if we see a sustained accretion episode, it’s not just a black hole problem; it’s a galaxy-scale problem involving how the surrounding gas is compressed and heated.

Subrahmanyan: It really emphasizes that for cosmology, structure formation isn't just about the initial density fluctuations; it's about these localized feedback loops that determine the final mass of the dark matter halo.

Vera: So, we’re looking at how mergers set up an initial inflow, but then some physics has to take over to keep that inflow from immediately dispersing.

Jocelyn: It reminds me of what we’ll be looking at next when we discuss those time-domain signals, because if accretion is sustained, there should be a measurable signature in the radio or optical light.

Subrahmanyan: And that's where I'll want to see observations that can constrain the gas density profile around these merging halos; that would really confirm or kill their model for how fast they grow.

Riccardo Caleno, Tommaso Zana, Raffaella Schneider, Alessandro Lupi, Pedro R. Capelo, Lucio Mayer, Alessandro Trinca, Rosa Valiante, Marta Volonteri

Dipartimento di Fisica, Sapienza University of Rome Department of Physics, INAF Osservatorio Astronomico di Roma, INFN Sezione Roma1 Department of Physics, Sapienza University of Rome Sapienza School for Advanced Studies Como Lake Center for Astrophysics DiSAT Università degli Studi dell’Insubria INFN Sezione di Milano-Bicocca Institute for Astronomy University of Edinburgh Institute for Astronomy Sorbonne Université CNRS Institut d’Astrophysique de Paris

astro-ph.GA, astro-ph.CO, astro-ph.HE

Submitted: 2026-06-02

Updated: 2026-10-21

Comments: 12+2 pages, 8 figures; accepted for publication in Astronomy & Astrophysics (A&A)

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 73/100

The gist: The gist: A major merger between two relatively low-mass halos at high redshift can trigger episodes of sustained super-Eddington accretion, but this requires the complete switching off of black hole

Key concepts

Super-Eddington Accretion (SE): This occurs when a black hole accretes matter at a rate significantly higher than the Eddington limit. In this paper, it is modeled as an accretion regime where the mass inflow rate exceeds the theoretical maximum sustainable rate for that black hole's mass.
Black Hole Feedback: This refers to physical processes where energy or momentum from the black hole (like jets or winds) pushes back against the surrounding gas. The paper found that this feedback is crucial; without it, BHs grow significantly, but with it, growth is strongly suppressed.
Bondi-Hoyle-Lyttleton (BHL) Accretion: This formula is used to model how gas particles fall onto the black hole. It calculates the accretion rate based on the gravitational pull of the black hole and the density of nearby gas, effectively describing how matter flows in from its surroundings.
Kinetic Feedback: This specific type of feedback involves physical outflows like jets or winds that move gas away from or disrupt it around a black hole. The simulation results show that kinetic feedback is the primary mechanism regulating gas accretion and preventing runaway growth.

Terminology

Accretion Flows — 1× in this paper · explained in 1 episode

Summary

The gist: A major merger between two relatively low-mass halos at high redshift can trigger episodes of sustained super-Eddington accretion, but this requires the complete switching off of black hole feedback for substantial growth in these systems.

Context and Motivation

JWST observations have revealed black holes with masses of several million solar masses already at z > 8, challenging our understanding of their growth mechanisms The role of major mergers in triggering super-Eddington accretion. In this work, we investigate whether a major merger between two relatively low-mass halos (Mhalo ∼ 109 M⊙) at high redshift can trigger episodes of sustained SE accretion, with particular focus on the role of BH feedback Since the advent of the James Webb Space Telescope (JWST), the presence of supermassive black holes (SMBHs) at very high redshift has challenged previous models of BH formation and accretion

Simulation Setup and Accretion Modeling

The simulations were performed with the code gizmo (Hopkins 2015), which employs the same domain decomposition and N-body algorithms as its predecessor gadget (Springel 2005) Accretion is modelled using the Bondi-Hoyle-Lyttleton (BHL; Hoyle & Lyttleton 1939; Bondi & Hoyle 1944) formula The accretion rate is computed by kernel-averaging the individual contributions from the gas particles within this kernel The BH kernel is defined as the region enclosing its 96 nearest gas neighbours

Black Hole Feedback Mechanisms

The code implements a physically motivated prescription for three accretion regimes classified by the Eddington ratio, λ = M˙BH/M˙Edd These three regimes are – λ < 1.25 × 10−3: advection-dominated accretion flow (ADAF), – 1.25 × 10−3 ≤ λ < 1: sub-Eddington radiatively efficient accretion, – λ ≥ 1: SE accretion The prescriptions are particularly relevant to investigate the interplay between the SMBH and the host galaxy in cosmological zoom-in simulations

Results on Merger Impact and Feedback

For the relatively low-mass halos studied here, our feedback prescription efficiently suppresses gas accretion, preventing substantial BH growth We find that, although the merger drives gas inflows towards the central regions, this is not sufficient to trigger sustained SE accretion Post-merger SE accretion episodes are observed only when BH feedback is entirely switched off Amongst the feedback channels considered, kinetic feedback is the primary mechanism regulating BH growth Moreover, the only significant SE accretion episodes occur immediately after BH seeding, while the merger itself does not produce a substantial enhancement of the accretion rate In order to isolate the role of BH feedback in our simulations, we performed a simulation in which both kinetic and radiative feedback were disabled When no BH feedback is considered, we observe a significant growth in BH mass, particularly for the BH with an initial mass of 104 M⊙, which increases to ∼ 2 × 105 M⊙ following a substantial SE accretion episode occurring at the time of seeding The results presented in Sections 4.3 and 4.4 indicate that BH feedback is the primary mechanism regulating gas accretion onto BHs in these systems

Conclusion on Sustained Accretion

Even when extending the simulations to z = 8 to capture possible delayed effects of the merger, we do not find sustained SE accretion over long time-scales The merger itself does trigger gas inflows towards the central regions and increases the gas density, as indicated by the sharp rise in stellar mass at z ∼ 11 in Figure 2 However, this is not sufficient to produce a sustained enhancement in the accretion rate, for two reasons First, the time-scale required to build up a high-density gas reservoir around the BH is not short, and in such low-mass haloes even a single SN explosion can disperse the gas before sustained accretion is established Second, in the other runs, radiation pressure and kinetic feedback partially counteract the compression of the gas, preventing the central density around the BH from reaching the values required for strong accretion The results presented in Sections 4.3 and 4.4 indicate that BH feedback is the primary mechanism regulating gas accretion onto BHs in these systems In particular, the strength of the kinetic feedback plays a crucial role When the kinetic component is neglected [see panels (c) and (f) in Figure 6], accretion rates increase and persist over longer time-scales [∼103 kyr, see panel (c) Figure 7], and BHs grow significantly, reaching masses of 105−6 M⊙ When kinetic feedback is included, the initial SE phase is rapidly quenched and BH growth remains strongly suppressed The behaviour observed in runs including kinetic feedback is directly linked to the adopted prescription

Future Directions

Another important result of our simulations is that the only significant SE accretion episodes occur immediately after BH seeding If fEdd ≫ fc, then ϕ ∼ ϕMAD (see Equations 8–9), so β becomes a function of just the spin a; for a = 0.7, β ≃ 138, whereas for a = 0.2, β ≃ 7 A simple energetic argument further illustrates the strength of this feedback For very high accretion rates (fEdd >> fc), the jet efficiency becomes primarily a function of the BH spin, ηjet(a) (see Equations 8, 9, 10, and 11) Under these assumptions, the injected energy can easily exceed the binding energy of the host halo An SE burst with λ ∼ 10–100, lasting even a few kyr, can therefore inject energy comparable to the binding energy of the halo into the gas surrounding the BH, potentially ejecting the gas from the halo The only exception remains Run4[panel (d) of Figure 4], in which all forms of feedback are neglected however, in this case SE accretion is stopped by SN explosions The merger itself does trigger gas inflows towards the central regions and increases the gas density, as indicated by the sharp rise in stellar mass at z ∼ 11 in Figure 2 However, this is not sufficient to produce a sustained enhancement in the accretion rate, for two reasons First, the time-scale required to build up a high-density gas reservoir around the BH is not short, and in such low-mass haloes even a single SN explosion can disperse the gas before sustained accretion is established Second, in the other runs, radiation pressure and kinetic feedback partially counteract the compression of the gas, preventing the central density around the BH from reaching the values required for strong accretion The behaviour observed in runs including kinetic feedback is directly linked to the adopted prescription Although physically motivated, as it is based on the Blandford-Znajek mechanism (Blandford & Znajek 1977), it still presents some limitations In particular, while we included a variable magnetic flux (Equation 8), the BH spin was kept fixed This parameter is crucial for determining the efficiency of kinetic feedback (see Equation 10) A low-spin BH undergoing SE accretion may spin down through the extraction of rotational energy, potentially leading to an initial phase more similar to the case without kinetic feedback, or even to hyper-Eddington accretion (Inayoshi et al. 2016) The behaviour observed in runs including kinetic feedback is directly linked to the adopted prescription Although physically motivated, as it is based on the Blandford-Znajek mechanism (Blandford & Znajek 1977), it still presents some limitations In particular, while we included a variable magnetic flux (Equation 8), the BH spin was kept fixed This parameter is crucial for determining the efficiency of kinetic feedback (see Equation 10) A low-spin BH undergoing SE accretion may spin down through the extraction of rotational energy, potentially leading to an initial phase more similar to the case without kinetic feedback, or even to hyper-Eddington accretion (Inayoshi et al.

Improvements for AI systems

Super-Eddington accretion modeling in simulations can be improved by incorporating physically motivated kinetic feedback models, such as jets or winds, across all accretion regimes to better regulate black hole growth. This allows AI systems to more accurately predict whether a major merger will trigger sustained super-Eddington accretion episodes based on the observed interplay between BH feedback and gas inflows.
The simulation framework can be enhanced by implementing a more realistic description of chaotic cold accretion, angular momentum transport, or mass fluxes within the black hole kernel instead of relying solely on the BHL prescription. This refinement would allow AI systems to model how turbulent environments typical of low-mass haloes influence black hole growth histories more accurately.
The accretion rate calculation can be made more physically accurate by adopting a kernel that resolves the size of a jet at its base rather than assuming a fixed volume, as suggested by Takeo et al. 2020. This would allow AI systems to better capture the impact of kinetic feedback on gas depletion and mass loss from the accretion flow.
The coupling between radiation and chemistry can be more robust by explicitly accounting for AGN Compton heating in the chemical network, assuming specific temperatures for soft and hard X-rays as described in Appendix A. This enables AI systems to model how radiative processes influence the thermal state of gas reservoirs around black holes more accurately.