Spins of Black Holes in X-ray Binaries and the Tension with the Gravitational Wave Measurements

TL;DR

The paper tackles the apparent discrepancy between low natal BH spins inferred from gravitational-wave mergers ($a_*\,\sim\,0.1$--$0.2$, with $\,\sim\,90\%$ at $a_*\,\lesssim\,0.6$) and the high spins reported for BHs in X-ray binaries, particularly those with high-mass donors ($a_*\,\sim\,0.8$--1.0). It surveys the spin-diagnosis landscape: GW-based inferences for BBHs, and EM-based spectral and timing methods for XRBs (reflection, continuum-fitting, QPOs, jets, and polarization), highlighting substantial modeling-systematic uncertainties in the EM approaches. The authors argue that the GW picture of predominantly low spins is robust, while EM results may be biased high by disk-physics assumptions (e.g., reflection decomposition, disk atmosphere, warm coronae, plunging-region stress). They propose a pathway toward reconciliation via more realistic magnetized-disk models that remain stable in $0.05\lesssim L/L_{ m Edd}\lesssim 1$, along with plausible accretion-spin-up scenarios and consideration of distinct binary populations, followed by targeted future observations to refine the spin census. Overall, the work underscores the need for improved disk theories and coordinated multimessenger efforts to connect BH formation, accretion physics, and relativistic gravity.

Abstract

We review current challenges in understanding the values and origin of the spins of black holes in binaries. Thanks to recent advances in astrophysical instrumentation, the spins can now be measured using both gravitational waves emitted by merging black holes and electromagnetic radiation from accreting X-ray binaries containing black holes. A key finding of the gravitational-wave observatories is that premerger black holes in binaries have low spin values, with an average dimensionless spin parameter of $a_*\sim$0.1--0.2, with 90\% having $a_*\lesssim 0.6$. This implies that the natal spins of black holes are generally low, and the angular momentum transport in massive stars is efficient. On the other hand, most of the published spins in X-ray binaries are very high. In particular, this is the case for binaries with high-mass donors (potential progenitors of mergers), where their published spins range from 0.8 to 1.0. At the same time, their short lifetimes prevent significant spin-up by accretion. Those with low-mass donors could be spun-up to $a_*\gtrsim 0.7$ by accretion only if the donor initial masses were more than several solar masses, which remains unproven. However, the existing methods of spin measurements suffer from significant systematic errors. The method relying on relativistic X-ray line broadening is based on the separation of the observed spectra into incident and reflected ones, which is highly uncertain. The method relying on spectral fitting of accretion disk continua uses models that predict the disk to be highly unstable, while stability is observed. Improved stable models predict lower spins. The published spin measurements in X-ray binaries are uncertain. The spins of the binaries with high-mass donors may be low, while those with low-mass donors have a broader spin distribution, ranging from low to high.

Figures (8)

• Figure 1: Distribution of the individual values of the lower (denoted by $a_A$) and higher ($a_B$) spins of the components of premerger BH binaries. Adapted from LVK25b. See Section \ref{['GW']} for discussion.
• Figure 2: An example of an evolutionary scenario leading to the formation of a merging BBH system that involves a mass ratio reversal and tidal spin-up of the second-born BH. Based on the example described by Olejak24. See Section \ref{['GW']} for a description of the evolution states. The initial eccentricity at the zero-age main sequence (ZAMS) is $e=0.40$, but the orbit is circularized quickly already in RLOF I.
• Figure 3: The NICER (black crosses) and NuSTAR A and B (blue and red crosses, respectively) spectra of GX 339--4 fitted by the relativistic disk model slimbh (magenta), Compton upscattering of the disk emission (green) and its reflection from the disk (red), with the total spectrum shown by the blue curve. The spectrum below a few keV is attenuated by the Galactic absorption, as shown by the black curve. We see that the Comptonization spectrum (green), which is both emitted toward the observer and incident on the disk (thus giving rise to the reflection component), is strongly curved and very different from a power law. Adapted from Zdziarski25a.
• Figure 4: The vertical structure of a puffy disk around a non-rotating BH at $\dot M_{\rm accr}\approx 0.6 \dot M_{\rm Edd}$ averaged over time and azimuth. The color map shows the density, and the dashed white curve gives the disk scale height, defined by $\left[\int \rho(r,z) z^2{\rm d}z/\int\rho(r,z){\rm d}z\right]^{1/2}$. The dotted green-to-yellow curves give the contours of the gas temperature, $kT_{\rm e}$ in keV. The solid white curves show the $z$-integrated optical depth for $\tau_{\rm T} = 1$ (the photosphere), 10, 100, and 300. We see the presence of a surface layer with $\tau_{\rm T}\sim 10$ at the temperature of $kT_{\rm e}\sim 1$ keV, which can be interpreted as a warm corona. We note that the flow does not experience any noticeable change at the ISCO radius, $6\, G M/c^2$. Based on Lancova19.
• Figure 5: The radio/X-ray correlation of the luminosities in BH LMXBs in the hard state, using the compilation by Bahramian22. The black points show the data for MAXI J1659--152, GRO J1655--40, XTE J1720--318, IGR J17177--3656, MAXI J1836--194, GS 1354-64, and XTE J1650--500.