GW190517: Rapid Spin BBH Merger

• GW190517 is a binary-black-hole merger uniquely defined by confidently large spins in both components, distinguishing it from typical events.
• Multi-pipeline detections with network SNRs around 10–11 and advanced waveform analyses confirm its astrophysical significance and precise parameter estimation.
• Population inference and formation-channel studies show GW190517 anchors a rapid-spin subpopulation, impacting models of high-angular-momentum formation and new-physics constraints.

GW190517, also designated GW190517_055101, is a binary-black-hole merger observed on 2019-05-17 05:51:01 UTC and reported as a high-confidence astrophysical event in the LIGO-Virgo catalogs. In early catalog treatments it appeared as a stellar-mass binary black hole with moderate mass ratio and small positive effective inspiral spin, but later spin-focused reanalyses elevated it to a distinctive status: within the GWTC-3 sample, it is the only event whose individual-component spins are both confidently large, and it therefore anchors inferences of a distinct rapidly spinning subpopulation of merging binary black holes (Abbott et al., 2020, Collaboration et al., 2021, Hussain et al., 2024).

1. Detection, catalog placement, and data analysis

GW190517 was detected by multiple search pipelines with network signal-to-noise ratios of order $10$–$11$. In GWTC-2, GstLAL reported a network matched-filter signal-to-noise ratio of $10.2$ with false-alarm rate $0.5\,{\rm yr}^{-1}$, PyCBC recovered the event with network SNR $\approx 9.8$ and FAR $\approx 0.8\,{\rm yr}^{-1}$, and cWB found a coincident excess with network SNR $\approx 11.0$ and FAR $\approx 0.3\,{\rm yr}^{-1}$. In GWTC-2.1, the event retained high significance, with GPS time $1241135698.0$ s $(\pm 0.004\,{\rm s})$, network SNR $11$0 in MBTA, $11$1 in GstLAL, and $11$2 in PyCBC, and astrophysical probability $11$3 in all pipelines (Abbott et al., 2020, Collaboration et al., 2021).

The event was classified as a binary-black-hole merger. GWTC-2.1 states that binary-black-hole candidates are those for which both component masses exceed $11$4, and GW190517 satisfies this criterion with negligible probability of belonging to neutron-star-containing classes (Collaboration et al., 2021).

The catalog analyses also document the evolution of the data-treatment pipeline. GWTC-2 used aligned-spin and precessing waveform families, including higher-mode models such as SEOBNRv4PHM and NRSur7dq4, and adopted higher-mode runs as fiducial because the Jensen-Shannon divergence between the $11$5 and higher-mode results exceeded $11$6 bit for GW190517’s mass ratio and $11$7 (Abbott et al., 2020). GWTC-2.1 reprocessed the strain data offline with best-available calibration models, subtracted known instrumental lines, applied the same data-quality flags as GWTC-2, and performed Bayesian parameter estimation with waveform models including spin precession and higher harmonics in the LALInference/LALSuite framework (Collaboration et al., 2021).

2. Source parameters and the transition from a catalog event to a spin outlier

Published parameter estimates for GW190517 differ across catalog releases and reanalyses. This is explicit in the literature, where different waveform families, priors, and data products are used. The principal published values are summarized below.

Analysis	Quoted parameters	Distinctive point
GWTC-2 fiducial higher-mode analysis	$11$8, $11$9, $10.2$0, $10.2$1, $10.2$2, $10.2$3 Gpc, $10.2$4	Higher-mode results adopted as fiducial (Abbott et al., 2020)
GWTC-2.1	$10.2$5, $10.2$6, $10.2$7, $10.2$8, $10.2$9, $0.5\,{\rm yr}^{-1}$0, $0.5\,{\rm yr}^{-1}$1 Mpc, $0.5\,{\rm yr}^{-1}$2	Reprocessed strain data and updated PE (Collaboration et al., 2021)
IMRPhenomXPHM spin reanalysis	$0.5\,{\rm yr}^{-1}$3, $0.5\,{\rm yr}^{-1}$4, with 90% highest-posterior-density intervals roughly $0.5\,{\rm yr}^{-1}$5, $0.5\,{\rm yr}^{-1}$6	Only event with both component spins confidently large (Hussain et al., 2024)

A central distinction in the published interpretation of GW190517 is between $0.5\,{\rm yr}^{-1}$7 and the individual spin magnitudes. GWTC-2 reported

$0.5\,{\rm yr}^{-1}$8

and found $0.5\,{\rm yr}^{-1}$9, while GWTC-2.1 quoted $\approx 9.8$0 and $\approx 9.8$1 (Abbott et al., 2020, Collaboration et al., 2021). Later work instead emphasized the two-dimensional $\approx 9.8$2 plane and found GW190517 far from the bulk of the catalog because both component spins are large and comparatively well measured (Hussain et al., 2024).

This contrast is important for interpretation. In catalog-level summaries based on total mass, mass ratio, and $\approx 9.8$3, GW190517 was described as lying within the range of “ordinary” stellar-mass BBHs and as consistent with the broader O3a population (Collaboration et al., 2021). In analyses that resolve individual spin magnitudes, it becomes a statistical outlier.

3. Hierarchical population inference and the rapidly spinning subpopulation

A dedicated reanalysis of GWTC-3 modeled the joint spin-magnitude distribution as a two-component mixture of truncated Gaussians on $\approx 9.8$4,

$\approx 9.8$5

where $\approx 9.8$6 denotes the dominant slowly spinning mode and $\approx 9.8$7 the subdominant rapidly spinning mode (Hussain et al., 2024).

From the full 69-event fit, the inferred mixture fraction is

$\approx 9.8$8

so $\approx 9.8$9 of binaries lie in the rapid-spin mode. The dominant mode has $\approx 0.8\,{\rm yr}^{-1}$0, diagonal dispersions $\approx 0.8\,{\rm yr}^{-1}$1, and a mild anticorrelation $\approx 0.8\,{\rm yr}^{-1}$2. The rapid mode has $\approx 0.8\,{\rm yr}^{-1}$3, diagonal dispersions $\approx 0.8\,{\rm yr}^{-1}$4, and $\approx 0.8\,{\rm yr}^{-1}$5 consistent with $\approx 0.8\,{\rm yr}^{-1}$6 (Hussain et al., 2024).

Within this model, GW190517 is the event that anchors the high-spin component. Averaging over the hyper-posterior yields $\approx 0.8\,{\rm yr}^{-1}$7, corresponding to $\approx 0.8\,{\rm yr}^{-1}$8 odds in favor of the rapid-spin mode, and

$\approx 0.8\,{\rm yr}^{-1}$9

The same analysis reports support for two distinct subpopulations with greater than $\approx 11.0$0 credibility and states explicitly that the hint of a rapidly spinning subpopulation hinges on GW190517 (Hussain et al., 2024).

The population result also refines the structure of the low-spin mode. In the dominant mode, the $\approx 11.0$1 highest-posterior-density interval for $\approx 11.0$2 is roughly $\approx 11.0$3, with Bayes factor against $\approx 11.0$4 of $\approx 11.0$5. A single-component fit also hints at weak anticorrelation, with median $\approx 11.0$6 and $\approx 11.0$7, but cannot capture the separate high-spin peak (Hussain et al., 2024). A common misconception is therefore that the result is merely a statement about broad spin variance; the actual claim is more specific, namely a bimodal structure in the $\approx 11.0$8 plane.

4. Formation-channel interpretations

Several formation-channel studies identify GW190517 as a candidate for channels that produce high, approximately aligned spins. In a Bayesian comparison among isolated-binary channels, the chemically homogeneous evolution (CHE) prior is strongly favored over both common-envelope evolution (CEE) and stable mass transfer (SMT). For GW190517_055101, Qin et al. report $\approx 11.0$9 and $\approx 0.3\,{\rm yr}^{-1}$0. After applying prior-odds corrections based on local-rate ratios, the corresponding logarithmic odds are $\approx 0.3\,{\rm yr}^{-1}$1 and $\approx 0.3\,{\rm yr}^{-1}$2, implying $\approx 0.3\,{\rm yr}^{-1}$3 (Qin et al., 2022).

Under the CHE prior, the inferred source properties shift toward the expected channel morphology: chirp mass remains near $\approx 0.3\,{\rm yr}^{-1}$4, the mass ratio peaks sharply at $\approx 0.3\,{\rm yr}^{-1}$5, and $\approx 0.3\,{\rm yr}^{-1}$6 and $\approx 0.3\,{\rm yr}^{-1}$7 both peak near $\approx 0.3\,{\rm yr}^{-1}$8, excluding zero and unity (Qin et al., 2022). This is consistent with the later population-level statement that isolated chemically homogeneous evolution at low metallicity naturally accommodates a system in which both progenitors retain high angular momentum and collapse into rapidly spinning black holes with roughly aligned spins (Hussain et al., 2024).

The same high-angular-momentum picture appears in long-gamma-ray-burst modeling. Bavera et al. define

$\approx 0.3\,{\rm yr}^{-1}$9

and find $1241135698.0$0 for GW190517, corresponding to an $1241135698.0$1 probability that the binary formed in association with a luminous long gamma-ray burst (Bavera et al., 2021). Their broader model links tidal spin-up in close binaries to both spinning BBH mergers and luminous LGRBs.

A distinct interpretation emerges in the 166-event three-component spin-mixture analysis of Bartos et al., which fixes component shapes for standard, hierarchical-merger, and accretion channels and infers only their mixing fractions. That study finds strong evidence, $1241135698.0$2, for an accretion-origin high-spin subpopulation with $1241135698.0$3 and $1241135698.0$4 credible interval $1241135698.0$5. GW190517 is among the top five accretion candidates and is color-coded at $1241135698.0$6, with $1241135698.0$7, $1241135698.0$8, $1241135698.0$9, and $(\pm 0.004\,{\rm s})$0 (Bartos et al., 10 May 2026).

Across these studies, hierarchical-merger explanations are comparatively disfavored for this event. The spin-population analysis states that GW190517’s pair of large, well-measured $(\pm 0.004\,{\rm s})$1’s is inconsistent with the tail of the slow-spin population and is also disfavored in a hierarchical $(\pm 0.004\,{\rm s})$2–$(\pm 0.004\,{\rm s})$3 merger channel, which would produce a single large-spin component rather than two (Hussain et al., 2024). The accretion-mixture analysis similarly states that the hierarchical-merger prediction of $(\pm 0.004\,{\rm s})$4 is decisively disfavored as the location of the high-spin subpopulation, whereas the accretion peak at $(\pm 0.004\,{\rm s})$5 is preferred (Bartos et al., 10 May 2026). This suggests that, within current models, GW190517 is more naturally associated with channels that spin up both components.

5. Recoil kicks, remnant retention, and dynamical environments

Fragione and Loeb use GW190517 as a case study for recoil kicks and remnant retention. In their summary of the event, the measured parameters are $(\pm 0.004\,{\rm s})$6, $(\pm 0.004\,{\rm s})$7, $(\pm 0.004\,{\rm s})$8, $(\pm 0.004\,{\rm s})$9, and final spin $11$00 (Fragione et al., 2020).

Their recoil model adopts

$11$01

with the coefficients and spin decompositions given explicitly in their implementation of Eqs. (1)–(3). For GW190517 they draw $11$02 realizations of $11$03 under an isotropic-spin prior with $11$04 uniform in $11$05, discarding draws whose computed $11$06 or $11$07 lie outside the $11$08 LVC credible intervals (Fragione et al., 2020).

The resulting recoil distribution depends strongly on $11$09. For $11$10, the kick distribution peaks at $11$11 with median $11$12; for $11$13, peak $11$14, median $11$15; for $11$16, peak $11$17, median $11$18; and for $11$19, peak $11$20, median $11$21. Each $11$22 is well approximated by a log-normal with $11$23–$11$24 (Fragione et al., 2020).

The environmental implication is that only clusters with $11$25 can retain GW190517’s remnant unless progenitor spins were very low, $11$26; if $11$27, only the deepest potentials with $11$28 suffice (Fragione et al., 2020). The same study argues that matching both $11$29 and $11$30 under an isotropic single-generation prior requires $11$31, while a second-generation progenitor with $11$32 offers an alternative route. This does not overturn the two-high-spin interpretation from later work, but it shows that dynamical scenarios with deep escape potentials remain relevant in recoil-based modeling.

6. Constraints on ultralight bosons and dark photons

Because GW190517 contains a rapidly spinning black hole in the $11$33–$11$34 range, it has been used to test superradiant spin-down by ultralight fields. For scalar bosons, the operative condition is

$11$35

with $11$36 the Kerr horizon angular velocity and $11$37 the gravitational fine-structure constant (Ng et al., 2020).

Ng et al. use GWTC-2 measurements $11$38 and $11$39 and argue that, if a scalar boson of mass $11$40 in the range $11$41–$11$42 existed with negligible self-interaction and black-hole lifetime $11$43 Myr, the primary would have been spun down to $11$44 well before merger. For GW190517, the entire $11$45 credible region of $11$46 lies in the forbidden region above $11$47 for those $11$48, and the posterior probability of a boson in that band drops below $11$49 (Ng et al., 2020). The abstract of that work states that the statistical evidence is mostly driven by GW190412 and GW190517.

An analogous argument has been made for vector superradiance. Using the primary black hole in GW190517, Bhaumik et al. derive an excluded dark-photon mass window

$11$50

assuming a timescale of a few million years from the time of binary formation to merger. Their conservative evaluation uses the lower edge of the $11$51 GW190517 contour, $11$52 and $11$53, together with the superradiant condition and the growth-rate estimate for the dominant $11$54 mode (Ghosh et al., 2021).

These exclusions, however, are model-dependent. Xie and Huang study quartic self-interactions $11$55 and show that self-interactions introduce a new lower critical cloud mass, causing the superradiant growth to terminate earlier. For GW190517’s primary, they state that the usual excluded scalar-boson band near $11$56 is entirely reopened if

$11$57

equivalently $11$58 (Xie et al., 13 Mar 2025). In that regime, the cloud never grows large enough to spin the hole down outside the $11$59 LIGO posterior. A central caveat in interpreting GW190517 as a superradiance constraint is therefore whether the boson can be treated as effectively non-self-interacting.

GW190517 has thus become important in two distinct senses. Empirically, it is a secure BBH detection whose early catalog characterization placed it among heavy but otherwise ordinary O3a mergers. Methodologically, it is the event that most directly supports the inference of a rapidly spinning BBH subpopulation and the event that most sharply tests high-angular-momentum formation channels and superradiant new-physics scenarios. The literature converges on the point that its significance arises not from exceptional mass alone, but from the combination of near-equal masses with large inferred component spins (Hussain et al., 2024, Qin et al., 2022, Bartos et al., 10 May 2026).