Abstract
The origin of the correlation between the effective spins ($\chi_{\rm eff}$) and mass ratios ($q$) of LIGO-Virgo-KAGRA's binary black holes (BBHs) remains a central question in gravitational-wave astronomy. This work investigates the potential $\chi_{\rm eff}-q$ correlation by accounting for the recent identification of two distinct BBH subpopulations. We find that the previously reported correlation significantly weakens or disappears for the low-mass subpopulation when a second $\chi_{\rm eff}$ distribution is introduced for high-mass events. Bayesian evidence strongly favors two separate $\chi_{\rm eff}$ distributions over a single mass-ratio-dependent distribution ($\ln\mathcal{B} > 4.2$). The first subpopulation aligns with first-generation BBHs, while the second exhibits characteristics consistent with hierarchical mergers in Active Galactic Nucleus (AGN) disks.
Introduction
In the study of Binary Black Holes (BBHs), two parameters are particularly diagnostic of formation history: the mass ratio ($q$) and the effective inspiral spin ($\chi_{\rm eff}$). Previous analyses of GWTC-2 and GWTC-3 data reported a curious anti-correlation where binaries with more unequal masses tended to have higher spins. This study investigates whether this correlation is a fundamental physical law of BBH evolution or a statistical artifact resulting from the superposition of two distinct formation channels: isolated binary evolution and dynamical environments like AGN disks or globular clusters.
Methodology
The authors applied Hierarchical Bayesian Inference to 69 BBH events from the GWTC-3 catalog, selecting events with a False Alarm Rate (FAR) of less than 1 per year. They compared three primary model architectures:
- Base Model: A single population where spin distribution parameters vary linearly with the mass ratio.
- Mixture Model: A population split into two distinct groups based on a fixed mixing fraction.
- Transition Model: A population that shifts from one type to another at a specific "transition mass" ($m_t$).
To ensure robustness, the primary mass distribution was modeled using both parametric "PowerLaw+Peak" (PP) and flexible "PowerLaw+Spline" (PS) approaches. Model preference was determined using Bayes factors ($\ln \mathcal{B}$).
Key Results and Statistical Evidence
The analysis reveals that the $\chi_{\rm eff}-q$ correlation observed in the total population is likely a result of superposition. When the high-mass subpopulation is modeled separately, the correlation in the low-mass group vanishes.
As shown in the table above, models incorporating two separate spin distributions are strongly favored over the single correlated model, with Bayes factors reaching as high as $\ln \mathcal{B} = 6.7$ for the Transition(PS) model.
Subpopulation Characteristics
The data identifies two clear "families" of black holes:
- Group 1 (First-Generation): Characterized by a narrow spin distribution peaking at $\sim 0.05$ and primary masses mostly below $40 M_\odot$. This group shows a sharp decline in mass beyond $40 M_\odot$, consistent with the predicted upper mass gap from pair-instability supernovae.
- Group 2 (Hierarchical Mergers): Characterized by a broad spin distribution peaking at $\mu_{\chi,2} \sim 0.4$ and primary masses starting around $30 M_\odot$.
The high spin and mass signatures of the second group align with theoretical predictions for hierarchical mergers occurring within the dense, gas-rich environments of AGN disks. The transition between these two populations is inferred to occur at a primary mass of approximately $49 M_\odot$.
Discussion and Future Outlook
While the evidence for a dual-population structure is statistically strong, the authors note that the current sample size of high-spin events remains small (approximately 11 events). This limitation makes it difficult to definitively distinguish between AGN-disk environments and other dynamical settings like stellar clusters, or to rule out isolated binary pathways such as stable mass transfer.
The upcoming release of data from LIGO-Virgo-KAGRA's fourth observing run (O4) is expected to significantly increase the event catalog. This larger dataset will be essential for confirming the hierarchical merger hypothesis and refining our understanding of how different astrophysical environments shape the black hole populations we observe today.
Summary updated