pith. machine review for the scientific record. sign in

arxiv: 2605.09517 · v1 · submitted 2026-05-10 · 🌌 astro-ph.HE · astro-ph.GA· astro-ph.SR

Recognition: 2 theorem links

· Lean Theorem

A low viscosity relatively thick twisted disk in a supermassive binary black hole as a potential model of OJ 287

Pavel B. Ivanov, Viacheslav V. Zhuravlev

Pith reviewed 2026-05-12 04:38 UTC · model grok-4.3

classification 🌌 astro-ph.HE astro-ph.GAastro-ph.SR
keywords twisted accretion disksupermassive binary black holeOJ 287Lense-Thirring precessiondisk crossingsrelative thicknessviscosity parameter
0
0 comments X

The pith

A twisted disk with relative thickness around 0.1 around the primary black hole in OJ 287 produces only two crossings with the secondary's orbit per period.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper models twisted accretion disks orbiting the more massive black hole in a supermassive binary using orbital parameters from the precessing massive model of the blazar OJ 287. For thin low-viscosity disks the model produces a twisting spiral wave near a resonance between the secondary's forcing frequency and the Lense-Thirring frequency of disk rings, resulting in multiple orbit crossings per period that contradict the model. When the relative thickness reaches δ ≳ 0.1 the typical inclination of disk material inside the binary orbit falls below the orbital inclination, so the secondary crosses the disk only twice per orbital period. The authors propose that extra heating from the secondary-disk collisions can sustain this thickness while the disk remains in its quasi-stationary precessing state.

Core claim

In the frame precessing with the Lense-Thirring frequency the twisted disk relaxes to a quasi-stationary configuration. For relative thickness δ ≳ 0.1 the inclination of the disk within the binary orbit is smaller than the orbital inclination, producing exactly two crossings of the secondary's orbit with the disk per orbital period, which is consistent with the precessing massive model of OJ 287 when additional heating from collisions maintains the adopted thickness.

What carries the argument

The quasi-stationary twisted disk in the precessing frame whose shape is set by the interplay between the relative thickness δ = h/r and the viscosity parameter α, with an additional resonance mechanism that generates a twisting spiral wave when α is small and δ is small.

If this is right

  • Only two crossings per orbital period occur once δ ≳ 0.1.
  • A twisting spiral wave appears near resonance in thin low-viscosity disks and produces multiple crossings that would invalidate the precessing massive model.
  • Collision heating offers a route to maintain δ ~ 0.1 without destroying the precession or the chosen orbital elements.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same thickness-dependent crossing behavior could apply to other periodic blazars suspected of hosting supermassive binaries.
  • Varying the collision-heating efficiency in simulations would test whether the required δ ~ 0.1 is reachable for OJ 287 parameters.
  • The resonance spiral wave may alter angular-momentum transport in thin disks of other binary systems even if it is ruled out for OJ 287.

Load-bearing premise

Additional heating from secondary-disk collisions can sustain a relative thickness of about 0.1 while preserving both the quasi-stationary precessing state and the adopted orbital parameters of the precessing massive model.

What would settle it

Hydrodynamical simulations that include the collision heating and show either that δ cannot be held near 0.1 or that the precessing state is disrupted, or observational timing of OJ 287 outbursts that reveals more than two disk crossings per orbital period.

Figures

Figures reproduced from arXiv: 2605.09517 by Pavel B. Ivanov, Viacheslav V. Zhuravlev.

Figure 1
Figure 1. Figure 1: FIG. 1. Top panel: The solid curve represents Ω [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. The dependency of [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Top panel. We show the time dependencies of [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Top panel. The dependencies [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. The absolute value of the wave-like part of [PITH_FULL_IMAGE:figures/full_fig_p017_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. The dependencies [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Solid curves represent the same numerical values of [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Top panel. The dashed (red) curve shows the height of t [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. Three-dimensional images of the twisted disk repres [PITH_FULL_IMAGE:figures/full_fig_p020_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. The dependencies of [PITH_FULL_IMAGE:figures/full_fig_p021_10.png] view at source ↗
read the original abstract

In this Paper we consider twisted accretion disks in supermassive binary black hole by analytical and numerical means. It is assumed that the disk orbiting around the more massive rotating component and that the disk rings are inclined with respect to the orbital plane. We use orbital parameters of the binary often employed in the precessing massive (PM) model of the well-known blazar OJ 287. Unlike our previous investigation of a similar problem, here we consider disks with both small and relatively large relative thicknesses $\delta=h/r$, where $h$ is the disk's height at a typical radius $r$, as well as a range of values of the viscosity parameter, $\alpha$, including the cases when $\alpha \lesssim \delta$. Similar to our previous results, we find that the twisted disk relaxes to a quasi-stationary state in the frame precessing with the Lense-Thirring frequency of the orbit. However, its shape is qualitatively different from that corresponding to the case of $\delta=10^{-3}$ and $\alpha=0.1$ considered in our previous work. In a disk with $\delta=10^{-3}$ but $\alpha \le 2\cdot 10^{-2}$, we find the new effect of generation of a twisting spiral wave near the resonance between a forcing frequency associated with the presence of the secondary and the Lense-Thirring frequency of a particular disk ring defined in the precessing frame. We propose an analytic theory of it, which is in a good agreement with our numerical results. This effect leads to multiple crossings of the orbit with the disk per one orbital period, which contradicts the PM model. When $\delta \gtrsim 0.1$, a typical disk's inclination within the orbit of the binary is smaller than that of the orbit which results in only two crossings of the orbit with the disk per one orbital period. We suggest that the additional heating of the disk gas by the secondary-disk collisions may result in $\delta \sim 0.1$.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript examines twisted accretion disks around the primary in a supermassive binary black hole system using orbital parameters from the precessing massive (PM) model of OJ 287. Analytic and numerical methods are applied to disks with varying relative thickness δ = h/r and viscosity α. For δ = 10^{-3} and α ≲ 2×10^{-2}, a resonant spiral wave is generated near the Lense-Thirring resonance, producing multiple disk-orbit crossings per orbital period. For δ ≳ 0.1 the disk relaxes to a quasi-stationary precessing state with inclination smaller than the binary orbit, yielding only two crossings per period. The authors propose that secondary-disk collisions provide additional heating sufficient to maintain δ ∼ 0.1.

Significance. If the self-consistency of δ ∼ 0.1 can be demonstrated, the work identifies a resonance mechanism that would invalidate the PM model for thin, low-viscosity disks and supplies a possible route to restore consistency via a thicker disk. The analytic treatment of the spiral wave and its agreement with numerics constitute a concrete, falsifiable prediction for the number of crossings as a function of α and δ.

major comments (2)
  1. [Abstract] Abstract and final paragraph: the statement that “additional heating of the disk gas by the secondary-disk collisions may result in δ ∼ 0.1” is presented without any energy-balance estimate, comparison of collision heating rate to radiative or viscous cooling, or solution for the resulting equilibrium thickness. Because the two-crossing regime (and therefore compatibility with the PM model) is stated to require δ ≳ 0.1, this omission leaves the central claim without demonstrated self-consistency.
  2. [Analytic theory section] The analytic theory of the spiral wave is asserted to agree with the numerical results, yet no derivation of the resonance condition, dispersion relation, or error analysis is supplied. Without these steps it is impossible to verify that the multiple-crossing outcome for α ≲ δ is robust rather than an artifact of the adopted numerical setup or boundary conditions.
minor comments (1)
  1. [Introduction] Notation for the Lense-Thirring frequency in the precessing frame should be defined explicitly once, with a clear distinction from the orbital frequency.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful and constructive review of our manuscript. The comments highlight important points regarding self-consistency and the presentation of the analytic theory. We address each major comment below and will revise the manuscript accordingly to strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract] Abstract and final paragraph: the statement that “additional heating of the disk gas by the secondary-disk collisions may result in δ ∼ 0.1” is presented without any energy-balance estimate, comparison of collision heating rate to radiative or viscous cooling, or solution for the resulting equilibrium thickness. Because the two-crossing regime (and therefore compatibility with the PM model) is stated to require δ ≳ 0.1, this omission leaves the central claim without demonstrated self-consistency.

    Authors: We agree that the suggestion for δ ∼ 0.1 would benefit from a quantitative estimate to demonstrate plausibility. The statement was intended as a qualitative indication that collisions could supply the necessary heating, rather than a complete equilibrium solution. In the revised manuscript we will add a simple order-of-magnitude comparison of the collision heating rate to radiative and viscous cooling, using the orbital parameters of the PM model, to show that δ ∼ 0.1 lies within the range that can be maintained. This will be placed in the discussion section and will not alter the main numerical and analytic results. revision: yes

  2. Referee: [Analytic theory section] The analytic theory of the spiral wave is asserted to agree with the numerical results, yet no derivation of the resonance condition, dispersion relation, or error analysis is supplied. Without these steps it is impossible to verify that the multiple-crossing outcome for α ≲ δ is robust rather than an artifact of the adopted numerical setup or boundary conditions.

    Authors: The resonance condition follows from matching the Lense-Thirring frequency of a disk ring (in the frame precessing with the binary) to the forcing frequency set by the secondary’s orbital motion. We acknowledge that the manuscript presents only the resulting condition and the numerical agreement without the intermediate dispersion relation or a systematic error analysis. In the revision we will expand the analytic section to derive the dispersion relation for the twisted-disk waves, provide the explicit resonance condition, and include a brief comparison of predicted versus measured wave amplitudes together with tests at different resolutions and boundary conditions to confirm robustness. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation chain is self-contained

full rationale

The paper adopts orbital parameters from the external PM model of OJ 287 as fixed inputs and then performs independent numerical simulations plus a new analytic theory for the twisted-disk evolution under varying δ and α. The reported relaxation to a quasi-stationary precessing state, the generation of the resonant spiral wave for small δ, and the reduction to two orbital crossings for δ ≳ 0.1 are direct outputs of these calculations rather than re-statements of fitted quantities or prior self-citations. The brief suggestion that secondary-disk collisions “may result in δ ∼ 0.1” is presented as an unquantified hypothesis and is not used as a load-bearing step in any derivation; the core results on disk shape and crossing number stand without it. No self-definitional loops, fitted-input predictions, or uniqueness theorems imported from the authors’ own prior work appear in the chain.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The model rests on standard thin-disk approximations, Lense-Thirring precession, and orbital parameters taken from the PM model of OJ 287; no new entities are postulated.

free parameters (2)
  • viscosity parameter α
    Varied across a range including values ≲ δ; not fitted to new data but explored to reveal the spiral-wave regime.
  • relative thickness δ
    Explored at 10^{-3} and ≳ 0.1; the value δ ∼ 0.1 is suggested rather than derived from first principles.
axioms (2)
  • domain assumption Disk rings remain inclined with respect to the binary orbital plane and relax to a quasi-stationary twisted state in the frame precessing at the Lense-Thirring frequency.
    Invoked throughout the setup and used to interpret both analytic and numerical results.
  • domain assumption Orbital parameters of the binary are those commonly employed in the precessing massive (PM) model of OJ 287.
    Adopted as input without re-derivation.

pith-pipeline@v0.9.0 · 5696 in / 1557 out tokens · 51704 ms · 2026-05-12T04:38:48.024815+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

Reference graph

Works this paper leans on

69 extracted references · 69 canonical work pages · 1 internal anchor

  1. [1]

    and references therein), our estimates show that δ could be of the order of 0 . 1. Therefore, the requirement that the disk should be relatively thick to have only two black hole-disk collisio ns per one orbital period may be consistent with the thick disk models, since the disks may quite significantly increa se their height due to additional gas heating ...

    work page

  2. [2]

    The twisted disk model Following [6], for a numerical work, we use a time dependent, fully relat ivistic semi-analytic twisted disk model proposed in [17]. In this model, the black hole rotational parameter is formally small, and the flat disk model formally coincides with the Novikov-Thorne extension of the Shakura-Suny aev α − model for a Schwarzschild b...

    work page

  3. [3]

    (7) Note that we neglect the terms proportional to Ω 2(r) assuming that they give a small contribution after averaging over several periods of Einstein precession of the binary

    to its r.h.s., we obtain δ2 4 GM r d dr [ ˆW,r (Ω b LT − Ω E − iα Ω k) ] = (Ω LT − Ω b LT − Ω 1) ˆW + Ω 1βb. (7) Note that we neglect the terms proportional to Ω 2(r) assuming that they give a small contribution after averaging over several periods of Einstein precession of the binary. We also s tress that it is assumed in eq. (

    work page

  4. [4]

    It is convenient to represent eq

    and all analytic estimates below that Ω 1(r) does not depend on time [29]. It is convenient to represent eq. ( 7) using a set of redefined quantities determining its coefficients. We in troduce Ω c ≡ Ω ceiΨ = α Ω k + i(Ω b LT − Ω E) (8) and Ω s = δ 2 Ω k. Using these variables eq. ( 7) can be rewritten in a more compact form r2Ω 2 s d dr [ Ω − 1 c ˆW,r ] = i...

    work page

  5. [5]

    can adequately describe the shape of Ω 1 the expression ( 10) can be brought to a very simple form with the help of eqns ( 1), ( 3): ∆Ω ≈ qΩ b K + Ω b LT − 2χ ( rg r ) 3/ 2 Ω K. (11) Note that formally the same equation can be obtained in the fully relat ivistic, low viscosity α ≪ 1 and slow rotation |χ | ≪ 1 limit from eqns (59) and (60) of [17] instead ...

    work page

  6. [6]

    is reduced to the simple form describing a formally inviscid stationary twisted disk around a Kerr black hole in the post-N ewtonian approximation, see [15] and also [17]. When χ > 0 it has a simple general solution in terms of Bessel functions J ν (y): ˆW = y3/ 5 ˆW∗ [ 2− 3/ 5Γ ( 2 5 ) J− 3/ 5(y) + CrelJ3/ 5(y) ] , y = 6 5 √( 32 3 χ δ2 )( r rg ) − 5/ 4 ,...

    work page

  7. [7]

    Since we assume that there is no influence on the disk’s dyna mics from the side of large radii, the part of the solution corresponding to an ingoing wave should be set to zero

    that the solution far from the binary’s orbit is a linear combination of ingoing and outgoing waves proportional to ei(Ω b LT t± 2 3 |z|3/ 2), where plus (minus) corresponds to ingoing (outgoing) solution. Since we assume that there is no influence on the disk’s dyna mics from the side of large radii, the part of the solution corresponding to an ingoing wa...

    work page

  8. [8]

    we can obtain a useful approximate expression for the location of the resonance rr ≈ 33 ( 2. 5χ ∗ (1. 5q∗ + χ ∗ ) ) 1/ 3 rg, (20) where we introduce χ ∗ = χ/ 0. 5, q∗ = q/ 7. 5 ·10− 3, and we set a = 60 rg and Ω b LT ≈ 10− 2χ Ω b K corresponding to the binary orbital parameters used in this study. In [6] it was shown that when viscosity is large, one can ...

    work page

  9. [9]

    This solution can be sm oothly matched to the equilibrium solution (18)

    in the vicinity of the resonance by decomposing the coefficients entering this equat ion in Taylor series in ( r − rr)/r r which is expressed in terms of the so-called Scorer functions [32]. This solution can be sm oothly matched to the equilibrium solution (18). It was also shown that this analytic result is in good agreement with the numerical simulations...

    work page

  10. [10]

    ( 9) near the resonance Let us consider a region close to rr, where x = (r − rr)/r r ≪ 1

    An approximate solution of eq. ( 9) near the resonance Let us consider a region close to rr, where x = (r − rr)/r r ≪ 1. In this region we can assume that Ω c, Ω s and Ω 1 are equal to their values at rr, and approximate ∆Ω as ∆Ω = Ω resx, (22) where Ω res does not depend on r. In what follows, we assume that Ω res is positive, which is valid when χ > 0. ...

    work page

  11. [11]

    ( 9), it was assumed that |x|= |r − rr|/r r is small

    A WKBJ analysis of the wave-like contribution In the above discussion of the wave-like contribution to the solution of eq. ( 9), it was assumed that |x|= |r − rr|/r r is small. At larger scales, it can be matched to a WKBJ solution of eq. ( 9), which describes free stationary (in the precessing frame) waves with wavelengths much shorter than r. These wave...

    work page

  12. [12]

    5 ) 1/ 2 . We can therefore use the known asymptotic expressions for Bess el functions in the formal limit y → 0 to estimate fp, ˆW ≈ ˆW∗ [ 1 + Brelχ 3/ 5δ− 6/ 5 ( r rg ) − 3/ 2] , where Brel = 227/ 10 51/ 532/ 5Γ(3 / 5) Crel, (47) which finally gives fp ≈ 3 2 Brelχ 3/ 5δ− 6/ 5rg 3/ 2r− 5/ 2 p 1 + Brelχ 3/ 5δ− 6/ 5(rg/r p)3/ 2 . (48) As we have mentioned a...

    work page

  13. [13]

    This results in two linear algebraic equations for C1, 2, which are readily solved

    and ( 49) are satisfied at r = rp and r = ra, respectively. This results in two linear algebraic equations for C1, 2, which are readily solved. We compare this solution with the results of numerical simulations below, see Fig. 7 in Section IV. It is seen that typical analytic values of β in the region rp < r < r a are larger than the numerical values by a ...

    work page

  14. [14]

    We check that this is also t he case for an extended time span corresponding to the relaxation to the quasi-stationary state

    As seen from this Figure, the curves describing the ( st) model are practically periodic, with the period corresponding to th e Lense-Thirring precession of the binary. We check that this is also t he case for an extended time span corresponding to the relaxation to the quasi-stationary state. The curves describ ing the ( dn) model exhibit the same period...

    work page

  15. [15]

    As seen in these panels, when r < r r ≈ 30rg, there is an excess of the amplitude, β , in comparison with the ’equilibrium’ solution described by eq

    The top and bottom panels correspond to α = 2 ·10− 2 and 2 ·10− 4, respectively. As seen in these panels, when r < r r ≈ 30rg, there is an excess of the amplitude, β , in comparison with the ’equilibrium’ solution described by eq. ( 18) at r ∼ rr. This excess is due to the effect of launching the ingoing wave of twist at th e resonance discussed in Section...

    work page

  16. [16]

    separately for the thin and relatively thick disks. For the thin disks, it is seen from, e.g., Fig ( 4) that the main contribution to the radial gradient of W within the orbit appears to be determined by the wave-like part ˆWwave at the radii smaller than rr (we neglect the strong oscillations of β at r > r r in the low viscosity ( dn) case, which do not ...

    work page

  17. [17]

    2 ) They computed the evolution of the disk till only ∼ 40Ω − 1 b , which is much smaller than the expected timescale of relaxation to th e quasi-stationary state

    The authors assumed that both black holes are non-rotating. 2 ) They computed the evolution of the disk till only ∼ 40Ω − 1 b , which is much smaller than the expected timescale of relaxation to th e quasi-stationary state. 3) They assumed that the orbit is initially perpendicular to the disk plane. Thus , it is important to extend the results of [43] to ...

    work page

  18. [18]

    Sillanpaa, S

    A. Sillanpaa, S. Haarala, M. J. Valtonen, B. Sundelius, and G . G. Byrd, ApJ 325, 628 (1988)

    work page 1988

  19. [19]

    H. J. Lehto and M. J. Valtonen, ApJ 460, 207 (1996)

    work page 1996

  20. [20]

    M. J. Valtonen, H. J. Lehto, K. Nilsson, J. Heidt, L. O. Takalo , A. Sillanp¨ a¨ a, C. Villforth, M. Kidger, G. Poyner, T. Pursimo, S. Zola, J. H. Wu, X. Zhou, K. Sadakane, M. Drozdz, D. Koziel, D. Marchev, W. Ogloza, C. Porowski, M. Siwak, G. Stachowski, M. Winiarski, V. P. Hentunen, M. Nis sinen, A. Liakos, and S. Dogru, Nature 452, 851 (2008) , arXiv:0...

    work page arXiv 2008

  21. [21]

    Komossa, D

    S. Komossa, D. Grupe, A. Kraus, M. A. Gurwell, Z. Haiman, F. K. Liu, A. Tchekhovskoy, L. C. Gallo, M. Berton, R. Blandford, J. L. G´ omez, and A. G. Gonzalez, MNRAS 522, L84 (2023) , arXiv:2302.11646 [astro-ph.HE]

    work page arXiv 2023

  22. [22]

    M. J. Valtonen, S. Zola, A. Gopakumar, A. L¨ ahteenm¨ aki, M. Tornikoski, L. Dey, A. C. Gupta, T. Pursimo, E. Knudstrup, J. L. Gomez, R. Hudec, M. Jel ´ ınek, J.ˇStrobl, A. V. Berdyugin, S. Ciprini, D. E. Reichart, V. V. Kou prianov, K. Matsumoto, M. Drozdz, M. Mugrauer, A. Sadun, M. Zejmo, A. Sillanp¨ a¨ a, H . J. Lehto, K. Nilsson, R. Imazawa, and M. Ue...

    work page arXiv 2023

  23. [23]

    P. B. Ivanov and V. V. Zhuravlev, MNRAS 528, 337 (2024) , arXiv:2401.01317 [astro-ph.GA]

    work page arXiv 2024

  24. [24]

    D. N. C. Lin and J. Papaloizou, ApJ 309, 846 (1986)

    work page 1986

  25. [25]

    Goldreich and S

    P. Goldreich and S. Tremaine, ApJ 233, 857 (1979) . 25

    work page 1979

  26. [26]

    N. I. Shakura, AZh 49, 921 (1972)

    work page 1972

  27. [27]

    N. I. Shakura and R. A. Sunyaev, A&A 24, 337 (1973)

    work page 1973

  28. [28]

    P. B. Ivanov, I. V. Igumenshchev, and I. D. Novikov, ApJ 507, 131 (1998)

    work page 1998

  29. [29]

    M. A. Abramowicz, B. Czerny, J. P. Lasota, and E. Szuszkiewic z, ApJ 332, 646 (1988)

    work page 1988

  30. [30]

    Blaes, Y.-F

    O. Blaes, Y.-F. Jiang, J.-P. Lasota, and G. Lipunova, Space Sci. Rev. 221, 120 (2025) , arXiv:2505.04402 [astro-ph.HE]

    work page arXiv 2025

  31. [31]

    J. C. B. Papaloizou and J. E. Pringle, MNRAS 202, 1181 (1983)

    work page 1983

  32. [32]

    P. B. Ivanov and A. F. Illarionov, MNRAS 285, 394 (1997)

    work page 1997

  33. [33]

    Demianski and P

    M. Demianski and P. B. Ivanov, A&A 324, 829 (1997)

    work page 1997

  34. [34]

    V. V. Zhuravlev and P. B. Ivanov, MNRAS 415, 2122 (2011) , arXiv:1103.5739 [astro-ph.HE]

    work page arXiv 2011

  35. [35]

    I. D. Novikov and K. S. Thorne, in Black Holes (Les Astres Occlus), edited by C. Dewitt and B. S. Dewitt (1973) pp. 343–450

    work page 1973

  36. [36]

    Morales Teixeira, P

    D. Morales Teixeira, P. C. Fragile, V. V. Zhuravlev, and P. B. Ivanov, ApJ 796, 103 (2014) , arXiv:1406.5514 [astro-ph.HE]

    work page arXiv 2014

  37. [37]

    V. V. Zhuravlev, P. B. Ivanov, P. C. Fragile, and D. Morales Te ixeira, ApJ 796, 104 (2014) , arXiv:1406.5515 [astro-ph.HE]

    work page arXiv 2014

  38. [38]

    P. B. Ivanov, V. V. Zhuravlev, and J. C. B. Papaloizou, MNRAS 481, 3470 (2018) , arXiv:1809.02585 [astro-ph.HE]

    work page arXiv 2018

  39. [39]

    D. N. Page and K. S. Thorne, ApJ 191, 499 (1974)

    work page 1974

  40. [40]

    J. J. Zanazzi and D. Lai, MNRAS 473, 603 (2018)

    work page 2018

  41. [41]

    We remind that the latter dependency is caused by our require ment that the disk gas in the vicinity of the crossing points is removed from the system and does not contribute to the aver aged torque exerted by the binary on the disk

    work page

  42. [42]

    See, however, the additional discussion of effects caused by Ω 2 below

    work page

  43. [43]

    J. D. Larwood, R. P. Nelson, J. C. B. Papaloizou, and C. Terque m, MNRAS 282, 597 (1996) , arXiv:astro-ph/9604013 [astro-ph]

    work page arXiv 1996

  44. [44]

    J. D. Larwood and J. C. B. Papaloizou, MNRAS 285, 288 (1997) , arXiv:astro-ph/9609145 [astro-ph]

    work page arXiv 1997

  45. [45]

    It may be shown that, in the lin ear approach, a twisted disk consists of ellipses with a small eccentricity

    We note that the auxiliary variable A has a simple physical meaning. It may be shown that, in the lin ear approach, a twisted disk consists of ellipses with a small eccentricity . A value of the eccentricity is an odd function of the disk hei ght. A determines this value as well as the orientation of their maj or axes; see [15] and [16]

    work page

  46. [46]

    [6] used ˆΩ 1 to represent the time independent ’average’ value of Ω 1

    work page

  47. [47]

    It corresponds to the requirement that α < χ − 2/ 5δ4/ 5, see [15]

    work page

  48. [48]

    (37) of [15] contains a misprint, the facto r y (denoted as y2 in [15]) should be in the power 3 / 5, while it has the power 3 / 2 in [15]

    We remind that eq. (37) of [15] contains a misprint, the facto r y (denoted as y2 in [15]) should be in the power 3 / 5, while it has the power 3 / 2 in [15]

    work page

  49. [49]

    We remind that these functions are solutions of an inhomogen eous Airy equation

    work page

  50. [50]

    Note a misprint in their eq

    [6] used Ω ∗ instead of Ω res. Note a misprint in their eq. (58), where the sign minus shoul d be absent in the front of Ω ∗ . This misprint does not propagate further in the text

    work page

  51. [51]

    Olver, D

    F. Olver, D. Lozier, R. Boisvert, and C. Clark, NIST Handbook of Mathematical Functions (Cambridge Univer sity Press, Cambridge, 2010)

    work page 2010

  52. [52]

    Note that when Ω res < 0 this condition is reversed

    work page

  53. [53]

    However, we believe that using a different relativistic mode l won’t change our qualitative conclusions

    We note that this constant is expected to be model dependent. However, we believe that using a different relativistic mode l won’t change our qualitative conclusions

    work page

  54. [54]

    It is interesting to note that there is another well pronounc ed peak in the spectrum of the ( dn) model corresponding to two times smaller period π/ (2Ω b E )

    work page

  55. [55]

    P. C. Fragile and O. M. Blaes, ApJ 687, 757 (2008) , arXiv:0807.2453 [astro-ph]

    work page arXiv 2008

  56. [56]

    G. I. Ogilvie and H. N. Latter, MNRAS 433, 2403 (2013) , arXiv:1303.0263 [astro-ph.SR]

    work page arXiv 2013

  57. [57]

    L. E. Held and G. I. Ogilvie, MNRAS 546, stag245 (2026) , arXiv:2511.18655 [astro-ph.SR]

    work page arXiv 2026

  58. [58]

    P. B. Ivanov, A. G. Polnarev, and P. Saha, Monthly Notices of the Royal Astronomical Society 358, 1361–1378 (2005)

    work page 2005

  59. [59]

    This is related to a relatively fast orbital evol ution of the binary black hole due to the emission of gravitat ional waves

    We follow [6], who argued that, in the case of OJ 287, these col lisions do not result in the formation of a gap or some cavity in the disk. This is related to a relatively fast orbital evol ution of the binary black hole due to the emission of gravitat ional waves. Thus, Σ represents some typical disk density at the ra dii order of the binary semi-major axis

    work page

  60. [60]

    S. M. Ressler, L. Combi, B. Ripperda, and X. Li, ApJ 993, L22 (2025) , arXiv:2509.18241 [astro-ph.HE]

    work page arXiv 2025

  61. [61]

    P. B. Ivanov, E. V. Mikheeva, V. N. Lukash, A. M. Malinovsky, S . V. Chernov, A. S. Andrianov, V. I. Kostenko, and S. F. Likhachev, Physics Uspekhi 62, 423 (2019)

    work page 2019

  62. [62]

    M. J. Valtonen, L. Dey, A. Gopakumar, S. Zola, S. Komossa, T. P ursimo, J. L. Gomez, R. Hudec, H. Jermak, and A. V. Berdyugin, Galaxies 10, 1 (2021) , arXiv:2112.10102 [astro-ph.HE]

    work page arXiv 2021

  63. [63]

    N. I. Shakura and R. A. Sunyaev, MNRAS 175, 613 (1976)

    work page 1976

  64. [64]

    J. Ross, H. N. Latter, and M. Tehranchi, MNRAS 468, 2401 (2017) , arXiv:1703.00211 [astro-ph.HE]

    work page arXiv 2017

  65. [65]

    Indeed, si nce the evolution timescale is so short, a number of such systems observed in the present time is expected to be quite s mall

    Note that such a short timescale of orbital evolution presen ts a difficulty for a theory of formation of systems with the parameters appropriate for the PM model of OJ 287. Indeed, si nce the evolution timescale is so short, a number of such systems observed in the present time is expected to be quite s mall

    work page

  66. [66]

    This makes it different from the known radial oscillations of the disk tilt discovered by [15], which are suppressed already at α ∼ δ

    It is interesting to note that such a spiral wave survives unt il the rather high viscosity α ≫ δ. This makes it different from the known radial oscillations of the disk tilt discovered by [15], which are suppressed already at α ∼ δ

    work page

  67. [67]

    Kley and R

    W. Kley and R. P. Nelson, ARA&A 50, 211 (2012) , arXiv:1203.1184 [astro-ph.EP]

    work page arXiv 2012

  68. [68]

    and Ogilvie, Gordon and Tanaka, Hidekazu , month = mar, year =

    S. Paardekooper, R. Dong, P. Duffell, J. Fung, F. S. Masset, G. Ogilvie, and H. Tanaka, in Protostars and Planets VII , Astronomical Society of the Pacific Conference Series, Vol. 534, edited by S. Inutsuka, Y. Aikawa, T. Muto, K. Tomida, and M. Tamura (2023) p. 685, arXiv:2203.09595 [astro-ph.EP]

    work page arXiv 2023

  69. [69]

    The Bardeen-Petterson effect in accreting supermassive black-hole binaries: disc breaking and critical obliquity

    R. Nealon, E. Ragusa, D. Gerosa, G. Rosotti, and R. Barbieri, MNRAS 509, 5608 (2022) , arXiv:2111.08065 [astro-ph.HE]

    work page internal anchor Pith review Pith/arXiv arXiv 2022