WorksPhysics and Astronomy
Helicity-squared scaling of post-eikonal phases for electromagnetic and gravitational waves in Ricci-flat spacetimes
Abstract
Spin-optical descriptions of electromagnetic and gravitational waves isolate finite-frequency corrections that are odd in the helicity. We study the complementary, helicity-even part of the first post-eikonal correction for scalar, electromagnetic, and linearized gravitational waves on Ricci-flat backgrounds. In a parallel-propagated null tetrad, the direct Weyl coupling of the gravitational field projects onto the circular transverse-traceless states as the multiple of the identity. The Ricci identity for the spin coefficient then rewrites this projection as a total derivative plus optical terms. The helicity-even phase of a spin- wave, measured relative to the scalar wave, then obeys a helicity-squared law, , up to endpoint terms. The Weyl coupling is what completes the scaling in the gravitational case. For weak-deflection scattering by a Schwarzschild black hole we evaluate the relevant integrals along the ray and obtain , with . The same coefficient follows independently, and without reference to any frame or gauge, from the Regge–Wheeler phase shifts combined with the asymptotics of the rotation functions . Direct numerical integration of the radial equations confirms it. At first order in the helicity-even phase difference vanishes, as required by the Born amplitudes. The helicity-odd spin-optical sector and the helicity-even post-eikonal sector are therefore organized by the first and second powers of the helicity, respectively.
Introduction
The high-frequency propagation of waves in curved spacetime has been studied since the geometrical-optics expansion of Isaacson, and the subject now includes systematic higher-order treatments of scalar, electromagnetic, and gravitational waves [1, 2, 3, 4, 5]. Spin optics and Berry geometry gave finite-frequency corrections a particularly clear form [6]. For circularly polarized electromagnetic and gravitational waves, polarization transport generates helicity-dependent corrections to the ray dynamics, among them the gravitational spin Hall effect [7, 8, 9, 10, 11, 12, 13, 14, 15]. These methods have been applied to Schwarzschild and Kerr, and the graviton Berry curvature has recently been obtained from quantum field theory [16, 17, 18].
The leading spin-optical correction is odd under helicity reversal, . It is not, however, the whole first post-eikonal correction. At the same order in the inverse frequency the transport equation for the amplitude contains three kinds of terms: the Berry term, linear in the polarization connection and odd in ; terms quadratic in the transverse derivatives of the polarization frame, which are even in ; and, for gravitational waves only, a direct algebraic coupling to the background Weyl tensor,
On a Ricci-flat background Maxwell’s equations have no term of this kind. Spin-optical treatments retain the first group of terms, and the helicity-even terms are usually left aside, either because they do not affect the ray trajectory at this order or because, taken separately, they depend on the choice of frame. Earlier work on polarized waves has accordingly focused on helicity-odd effects or on the general structure of the corrected field [4, 5, 19, 20], and beyond-eikonal corrections in Schwarzschild have been computed for scalar waves [21].
Two questions are left open. The first is how the helicity-even phase depends on the spin of the field. The second is whether the Weyl coupling gives gravitational waves a structure with no electromagnetic analogue at this order. Both bear on a structural statement that is well established for the helicity-odd sector. There the effective ray equations of gravitational waves have the same form as those derived from Maxwell’s equations [11], and fields of different spin differ only by a constant factor, the helicity, in the spin tensor [14]; methods developed for electromagnetic waves are carried over to gravitational waves by accounting for the difference in helicity [12]. Whether a comparable statement holds for the helicity-even sector, where the Weyl coupling is the only algebraic contact between the spin-2 field and the background curvature, has not been examined. To our knowledge, the spin dependence of the helicity-even post-eikonal phase has not been determined.
In this paper we show that the helicity-even sector obeys a helicity-squared law, and that the Weyl coupling is precisely what makes this law hold for gravity. Projected onto the circular transverse-traceless (TT) states, the Weyl operator is times the identity, so it neither converts nor splits the two gravitational helicities. In a parallel-propagated null tetrad, the Ricci identity for the spin coefficient rewrites as a total derivative plus products of optical scalars. The frame-derivative terms contain a connection part weighted by and an optical part weighted by . The two weights coincide for , and for the Weyl coupling supplies exactly their difference. The helicity-even phase of a spin- wave, measured relative to the scalar wave on the same congruence, is then times a common bulk integral plus endpoint terms. An analysis of the field strengths at the detector shows that the endpoint terms cancel, apart from local terms, so the law holds for the gauge-invariant phases. The helicity-odd sector is linear in , and the helicity-even sector is quadratic. In practical terms, the helicity-even phase of a gravitational wave follows from the electromagnetic one through the factor , without a separate evaluation of the Weyl term.
We then evaluate the helicity-even sector for weak-deflection scattering by a Schwarzschild black hole, in two independent ways. One calculation integrates the bulk quantities along the scattered ray. The other uses the Regge–Wheeler phase shifts and the asymptotics of the rotation functions , and involves no choice of frame or gauge. Both give
where is the impact parameter and the frequency. The first order in cancels, as required by the exact Born amplitudes [22], and direct numerical integration of the radial equations confirms the second-order coefficient.
The main contributions of this work are therefore the following: (i) the identity that relates the algebraic Weyl projection to the optical scalars of the congruence, and the helicity-squared law that follows from it; (ii) the endpoint analysis that extends the law to the gauge-invariant field strengths; and (iii) an explicit coefficient, obtained by two independent methods, which also serves as an analytic benchmark for numerical treatments of wave propagation in black-hole spacetimes. The effect itself is far below present observational sensitivity (Sec. VII); the interest of the result lies in the structure it reveals.
The paper is organized as follows. § 0.2 collects the post-eikonal equations for and the algebraic Weyl projection. Section III projects the first post-eikonal source in a parallel null tetrad. Section IV derives the helicity-squared law and treats the endpoint terms. § 0.5 evaluates the bulk integrals in Schwarzschild, and Sec. VI gives the partial-wave calculation. Section VII relates the result to spin optics and to wave-optics studies of lensing by rotating black holes, and Sec. VIII summarizes. Two appendixes contain the weak-deflection evaluation of the Weyl integral and the angular phase of .
We use signature and the curvature convention , for which the linearized Einstein equation takes the form ((II.1)).
Post-eikonal equations for
Consider a Ricci-flat background, . A massless scalar obeys . In Lorenz gauge the Maxwell potential obeys , which reduces to . For a traceless Lorenz-gauge metric perturbation, linearized gravity gives
[1, 3, 5]. The three cases may be written together as , with and .
We use the WKB ansatz
At leading order , so the rays are affinely parametrized null geodesics, . With the transport operator
the next two orders give
The gauge conditions must be imposed order by order. At leading order they make transverse (and traceless for ). At the next order they fix the longitudinal part of , for example for gravity. Let be a complex transverse null vector with , parallel transported along the rays. The circular states of helicity are
and the leading amplitude is . Here is real and obeys the flux law .
Algebraic Weyl projection
For , the matrix elements of between the circular TT states are
The off-diagonal elements vanish because the Weyl tensor is antisymmetric in each index pair. The diagonal elements coincide by pair symmetry, and is real. Therefore
The direct Weyl coupling produces neither conversion nor a splitting of the two helicities: it is helicity even. The scalar is invariant under . It is not invariant under a null rotation , which preserves all inner products of but mixes in components of the Weyl tensor along . Such a null rotation corresponds, at leading order, to a residual gauge transformation of . For this reason the integral of along a ray is not by itself an observable. The next sections show how it combines with the derivative terms.
Projected transport in a parallel null tetrad
Frame gradients
Complete to a null tetrad , with , parallel transported along the rays. Since , the tensor is symmetric. We use the projections
which correspond to the Newman–Penrose spin coefficients [23], up to signature-dependent signs. Here is real. The polarization connection is
a real covector with for a parallel frame. Using , and , the gradient of decomposes as
Writing , it follows that
Projected source
Projecting onto and using , we obtain
For helicity the sign of is reversed. The projections onto the opposite helicity vanish, because and . For these expressions reproduce the decomposition , of the gravitational TT source.
Phase and amplitude
Write the TT component of the first correction as . Since , the projection commutes with , and Eq. (II.4) gives
Since , the real part of corrects the phase and the imaginary part the amplitude. The amplitude correction is linear in and odd in the helicity. It is the field-level counterpart of the Berry term responsible for the spin-optical corrections [10, 11, 15]. The phase correction is helicity even.
The scalar term is common to all spins. It also dominates the approach to the caustic of the scattered congruence. We therefore consider the difference between the post-eikonal phases of the spin- and scalar waves on the same congruence. From Eqs. (III.4)–(III.8),
Helicity-squared law
Ricci identity
Differentiate along the ray and commute the derivatives. Using and , with , one finds
Tracelessness of the Weyl tensor, contracted with , gives . Hence
The algebraic projection of the Weyl tensor that enters the gravitational equation is thus a total derivative plus the same optical combination that appears in .
Two consequences follow at once. First, at order the products and are of second order. The integral of along the ray then reduces to the boundary term , which vanishes for scattering because decays at both ends. This is the structural reason why the nominal contribution of the Weyl term integrates to zero (Appendix A).
Second, inserting Eq. (IV.2) into Eq. (III.9) and integrating along the ray, we obtain
where
The last term vanishes both for Maxwell () and for linearized gravity (). Therefore
and, in particular,
Equation (IV.5) is the central structural result of this work. Without the Weyl coupling, the optical term in Eq. (III.9) would carry the coefficient , while the connection term carries . For the two coincide. For the direct Weyl coupling contributes exactly the missing times . The Weyl term is therefore not an independent helicity-even structure superimposed on the electromagnetic one: it is the term that makes the helicity-even sector of the gravitational wave scale as . The operator structure is familiar from . The linear term underlies spin optics, and the quadratic term is its helicity-even partner.
Frame dependence
The separate integrals in Eq. (IV.5) are defined in the parallel frame fixed on the incident screen. Under a local rotation , with , the field is unchanged, and so is the total phase. Under a parallel null rotation with constant , changes only at relative order (Appendix A). The same holds for : since , with and , the change is of the same relative order.
The TT phase is not itself a gauge-invariant quantity. The invariant observables are field strengths evaluated at the detector, and they differ from the TT phase by endpoint terms that do not accumulate along the ray. For Maxwell, the Newman–Penrose scalar follows directly from Eq. (II.2):
Endpoint terms for the gravitational amplitude
Equation Eq. (IV.7) shows that, for Maxwell, the phase of a field-strength component is the TT phase plus local terms, and that the local term cancels the boundary contribution in ; for an incident plane wave . We now carry out the analogous calculation for linearized gravity.
Use tetrad indices , , , , with denoting the contraction of a tensor with the tetrad vectors after differentiation. For helicity ,
The only nonvanishing component of is , and the TT component of is . In vacuum the linearized Weyl tensor equals the linearized Riemann tensor, so that, up to background-curvature terms of relative order ,
Each term follows from
with . Although and vanish, their derivatives need not, because the frame varies. In the first two terms of Eq. (IV.9) the term vanishes because ; in the bracket only the contraction survives, and . Hence
The third term is , since and . In the last term, , and
Collecting terms,
On a curved background the linearized Weyl tensor changes under a gauge transformation by . For transformations compatible with the WKB form, , this change is of relative order , so the phase in Eq. (IV.13) is gauge invariant at the order considered.
The structure is that of Eq. (IV.7), with the weight in front of . The terms and are local: they are evaluated at the detector, do not accumulate along the ray, and decay in the far zone. The term cancels the boundary contribution in Eq. (IV.6). This cancellation is exact. It does not rely on the decay of , which is guaranteed only at first order in : at second order, on the outgoing branch before the caustic, the right-hand side of Eq. (IV.1) is dominated by the approximately constant product , and grows with the distance from the lens. For both fields, therefore, the accumulated phase of the invariant field strength, measured relative to the scalar wave, is
up to local terms at the detector, and . The non-TT part of does not affect this result. The bulk calculation of § 0.5 and the partial-wave calculation of Sec. VI therefore evaluate the same invariant quantity.
Schwarzschild: bulk evaluation
From now on we use coordinates in units of , with and , and consider the weak-deflection regime . The turning point satisfies , so the leading coefficients below are the same whether expressed in or in .
Weyl integral
In the parallel frame fixed on the incident screen, the weak-deflection evaluation of Appendix A gives
The straight-line contribution at order is a total derivative and cancels between past and future infinity. The first nonvanishing term comes from the bending of the trajectory.
Connection integral
Commuting derivatives in the transport of Eq. (III.2) gives
Since and , the first term is of second order, and can be computed to along the unperturbed ray. Take , the ray through , and . In the static frame the vacuum Weyl tensor is determined by its electric part, , and
Integrating Eq. (V.2) from the incident screen, where ,
The identities , and used here have been checked numerically with the full vacuum Weyl tensor.
Result
Combining Eqs. (IV.5), (V.1) and (V.4),
The Weyl integral contributes, through the Ricci identity, one half of the electromagnetic coefficient; the connection integral contributes the other half. The gravitational coefficient is four times the electromagnetic one.
Schwarzschild: partial-wave evaluation
We now compute the same quantity from the scattering amplitudes. Nothing in this section depends on a choice of frame, tetrad, or gauge.
Constraint at first order in
In the static linearized field , with , the Born amplitude is determined by . For a massless scalar , independent of the angle. For the Maxwell field between states of equal helicity. For gravitons the helicity-preserving amplitude carries [22, 24]. Hence
where . Since , the difference is a polynomial in . It is a contact term in impact-parameter space. At first order in there is therefore no spin-dependent phase at , to all orders in .
Phase shifts and angular functions
In Schwarzschild spacetime the helicity-preserving amplitude takes the standard form [25, 26, 27]
The phase shifts are those of the Regge–Wheeler equation , with
For the axial and polar phase shifts are related by Chandrasekhar’s transformation [28]. Their relative phase is of order , far beyond the order considered here, so helicity reversal does not enter.
We count orders in at fixed , with , and keep all terms of first order in . These are the terms .
Radial part. Relative to , the spin-dependent part of is of order . Its effect on the phase shift is therefore given exactly, at order , by first-order WKB perturbation:
with
where is the root near unity. We obtained the coefficient numerically, to ten significant figures, by Richardson extrapolation. Higher-order WKB corrections and the Langer term are spin independent at this order.
Angular part. The functions obey an equation whose potential differs from that of by (Appendix B). In the oscillatory region , their phase differs from that of by
Stationary phase
For attractive scattering the dominant branch of Eq. (VI.2) is , and the stationary point is fixed by . At fixed scattering angle, and to first order in the spin-dependent pieces, the displacement of the stationary point does not affect the phase. Therefore
with the null deflection angle
The terms of first order in cancel, , as required by Eq. (VI.1). At second order,
in exact agreement with Eq. (V.5). The factor enters here through the coefficient of the Regge–Wheeler potential and through the angular potential .
Equation (VI.7) holds at first order in for all . Its values, computed with the exact deflection angle and the exact integral , are listed in Table I. They approach with a correction of order ; a Richardson extrapolation of the last two entries gives .
| 10 | 1.369419408 | 0.5903957876 | −3.019852 |
| 20 | 1.142697847 | 0.2361359954 | −1.739966 |
| 40 | 1.064444142 | 0.1081040765 | −1.412052 |
| 80 | 1.030766135 | 0.0519284942 | −1.285951 |
| 160 | 1.015046466 | 0.0254708775 | −1.229990 |
| 400 | 1.005940974 | 0.0100743045 | −1.198394 |
| 800 | 1.002957804 | 0.0050184915 | −1.188171 |
| 1600 | 1.001475754 | 0.0025046124 | −1.183116 |
| 3200 | 1.000737093 | 0.0012511518 | −1.180602 |
Table I. Invariant post-eikonal phase difference in Schwarzschild, Eq. (VI.7), evaluated with the exact deflection angle and the exact integral . The last column is , whose limit is
Numerical checks
We integrated the Regge–Wheeler equations for and numerically. Each integration starts inside the centrifugal barrier with the growing WKB solution and continues to , where the phase is read off with the local wave number. The resulting phase-shift differences are compared with Eq. (VI.4) in Table II. The relative differences are between and . At the second-order part of Eq. (VI.4) accounts for about of the total, so it is confirmed to better than . The angular phase (VI.6) was checked against the zeros of and relative to those of , for and , , with agreement at the level of . The Python scripts used for these checks and for Table I were written with the assistance of an AI tool (see the Acknowledgments). They use standard libraries (SciPy and mpmath) and are provided upon request.
| numerical | Eq. (VI.4) | relative difference | |||
| 10 | 200 | 20.05 | |||
| 10 | 400 | 40.05 | |||
| 20 | 400 | 20.02 |
Table II. Phase-shift difference from direct integration of the Regge–Wheeler equations, compared with the first-order WKB expression, Eq. (VI.4).
Discussion
Helicity-odd and helicity-even sectors
The first post-eikonal correction separates cleanly under helicity reversal. The helicity-odd part, carried by , is linear in the polarization connection and in . It underlies the spin Hall effect [10, 11, 15, 17]. The helicity-even part, measured relative to the scalar wave, is quadratic in for both the electromagnetic and the gravitational field, Eq. (IV.5). The direct Weyl coupling does not produce a separate gravitational structure. Through the Ricci identity it is precisely what upgrades the optical term from to . A description of the gravitational wave obtained by weighting the electromagnetic helicity-even sector with is therefore complete at this order, whereas a weighting linear in would miss exactly the contribution of the Weyl term.
The coefficient in Eq. (I.2) has a simple origin in each calculation. Along the ray, half of it comes from the Weyl integral and half from the transverse twist of the parallel frame. In partial waves it is the difference between the second-order radial term, , and the kinematic term produced by the angular functions at the second-order deflection, .
Relation to partial-wave and wave-optics studies
Partial-wave treatments of scalar, electromagnetic and gravitational scattering by black holes are well developed [25, 26, 27]. Wave-optics studies of gravitational lensing by rotating black holes have recently been carried out with the Teukolsky equation. In particular, Kubota et al. [29] solved the Teukolsky equation for gravitational waves emitted by a binary and lensed by a Kerr black hole, in the regime where the wavelength is comparable to the gravitational radius. They found a helicity-dependent, small-period oscillation in the power spectrum of the forward amplification factor, which grows with the black-hole spin for prograde configurations. That work addresses the full wave problem numerically, in a regime where is of order unity, and its central effect distinguishes the two helicities. The present work addresses a different question. We work in the high-frequency, weak-deflection regime, isolate the helicity-even correction at first post-eikonal order, and compare fields of different spin on the same background. Our result is an analytic coefficient that depends on and is the same for both helicities. Low-frequency comparisons between Teukolsky scattering and scattering amplitudes [24] are also complementary to ours.
Two further lines of work deserve comment. Beyond-geometric-optics corrections to gravitational lensing have been studied for gravitational waves at linear order in the lens potential [20] and, for scalar waves, up to second order in [21]. In the scalar case the first correction appears at order and grows with the distance from the lens. This is the behavior of the common term , which cancels in . The spin-dependent integrands considered here decay as , and their integrals converge. The absence of a spin-dependent phase at order is consistent with the linear-order analysis of Ref. [20].
In the scattering-amplitude literature, the bending angle of massless particles has been computed at one loop, and it contains spin-dependent terms of the form [30, 31]. These terms originate in quantum loops, are linear in , and are obtained for at first order in . The effect found here is of a different nature. It is a classical wave effect, proportional to . In particle variables, with , it contributes to the deflection angle only at order , beyond the truncation of Refs. [30, 31]. At the classical level the helicity-preserving eikonal phase of general relativity is universal at second post-Minkowskian order [32]. Finally, a helicity-even response of gravitons has recently been identified in a quantum-geometric gradient expansion [33]. That effect is sourced by the Ricci tensor and vanishes in vacuum, whereas the effect described here arises entirely from the Weyl curvature.
Magnitude of the effect
For a gravitational wave, Eq. (I.2) gives . With , a wave of frequency passing a black hole of has . At this gives rad, and at , rad. For a black hole of the corresponding values are and rad. At the higher-order terms in increase these numbers by a factor of about 2.6 (Table I), which does not change the conclusion. For cosmological lenses the relevant impact parameters are of the order of the Einstein radius, , and the effect is entirely negligible. Larger values would require , where the post-eikonal expansion no longer applies and the full wave problem must be solved [29]. The helicity-even phase difference is therefore far below the phase resolution of present and planned detectors. It also enters the observed frequency-dependent phase together with the spin-independent term , which is larger and grows with the distance from the lens [21].
The interest of the result is thus structural rather than observational. Equation (IV.5) fixes how the helicity-even part of the first post-eikonal correction depends on the spin of the field, and it complements the helicity-odd spin-optical sector. The coefficient , together with the vanishing of the first-order term, also provides an analytic benchmark for numerical treatments of wave propagation in black-hole spacetimes, such as Teukolsky or wave-optics lensing codes, in the high-frequency regime.
Kerr
The derivation of Sec. IV uses only the vacuum Einstein equations and a parallel tetrad, so Eq. (IV.5) holds in Kerr spacetime as well. In the weak-deflection expansion the rotation parameter first enters the Weyl integral at order (Appendix A). Gravitomagnetic corrections to are of relative order , so they also affect only at order . The leading coefficient is therefore expected to be unchanged for . An invariant confirmation, and the rotation-dependent terms at order , require an invariant evaluation, for example from the Teukolsky equation with spin-weighted spheroidal harmonics. This is left for future work.
Conclusions
We have studied the helicity-even part of the first post-eikonal correction for scalar, electromagnetic, and linearized gravitational waves on Ricci-flat backgrounds. The direct Weyl coupling acts on the circular TT states as times the identity, so it neither converts nor splits the helicities. In a parallel null tetrad the Ricci identity for rewrites as a total derivative plus the optical combination . With this identity the helicity-even phase of a spin- wave, measured relative to the scalar wave, equals times a common bulk integral plus endpoint terms; for the invariant field strengths the endpoint terms cancel, apart from local terms at the detector. In the gravitational case the Weyl coupling supplies the part of the scaling that the frame-derivative terms alone would miss.
For weak-deflection scattering by a Schwarzschild black hole the helicity-even phase difference vanishes at first order in and at second order is . We obtained this coefficient in two ways: by integrating the bulk quantities along the ray, and from the Regge–Wheeler phase shifts together with the asymptotics of . The second route involves no frame or gauge choice, and direct numerical integration of the radial equations confirms it. The spin-optical and post-eikonal corrections of electromagnetic and gravitational waves are therefore organized by the first and second powers of the helicity. Extending the invariant evaluation to the rotation-dependent terms in Kerr spacetime remains open.
Acknowledgments
The author wishes to express sincere gratitude to the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Fundação Cearense de Apoio ao Desenvolvimento Científico e Tecnológico (FUNCAP) for their valuable support through grants Project UNI-00210-00230.01.00/23 (FUNCAP), 303438/2026-6 (CNPq/PQ) and 420854/2025-8 (CNPq).
Use of AI tools.—This work made use of Claude (Anthropic, model claude-opus-5-5), accessed via the Claude desktop application in September 2026. Based on author prompts, the tool was used to assist with analytical derivations, numerical checks via Python scripts, and drafting and editing the manuscript. All analytical results were independently verified by the author, and all numerical outputs were reproduced by rerunning the scripts. The author takes full responsibility for the content of this paper.
Data Availability Statement
The code used in this work is available from the corresponding author upon reasonable request.
Conflict of Interest Statement
The author declares that there is no conflict of interest regarding the publication of this paper.
Weak-deflection evaluation of the Weyl integral
We work in units of and normalize the energy of the reference null geodesic to unity. For Schwarzschild, and for the equatorial Kerr problem with , the radial equation reads , with
Here is the impact parameter in units of , and the Schwarzschild case is . The turning point , defined by , satisfies . With ,
so does not enter at this order. The affine distance from the turning point is
In the parallel frame fixed on the incident screen, the quantity obeys . Denote by its value at the turning point. Then on the incoming and outgoing branches, and
Adding both branches at the same radius cancels the terms linear in :
Since , its contribution enters at . More precisely, a parallel null rotation changes the integral by exactly , with . Expanding, . The leading coefficient,
vanishes because, along the straight line ,
This is the explicit form of the boundary term in Eq. (IV.2). At the next order,
which gives Eq. (V.1).
A numerical evaluation of the same integral for the equatorial Kerr problem with , with determined from the turning radius and both branches integrated to infinity, gives and for and . These values converge to , consistently with the absence of at this order.
Angular phase of
The rotation functions with equal indices can be written in terms of Jacobi polynomials,
and the Hilb-type asymptotics of Jacobi polynomials [34] give at leading order, with for every . The spin dependence appears at the next order. The function obeys
which reduces to the Legendre case for . In the oscillatory region the WKB phase difference is
The integration constant is fixed by the small-angle expansion , which corresponds to . Corrections are of relative order and . In the scattering problem these correspond to order .
References
References
- [1]R. A. Isaacson, Gravitational radiation in the limit of high frequency. i. the linear approximation and geometrical optics, Physical Review 166, 1263 (1968).DOI
- [2]R. A. Isaacson, Gravitational radiation in the limit of high frequency. ii. nonlinear terms and the effective stress tensor, Physical Review 166, 1272 (1968).DOI
- [3]S. R. Dolan, Geometrical optics for scalar, electromagnetic and gravitational waves on curved spacetime, International Journal of Modern Physics D 27, 1843010 (2018).DOI
- [4]S. R. Dolan, Higher-order geometrical optics for electromagnetic waves on a curved spacetime (2018), arXiv:1801.02273 [gr-qc].arxiv.org/abs/1801.02273
- [5]A. I. Harte, Gravitational lensing beyond geometric optics: I. formalism and observables, General Relativity and Gravitation 51, 14 (2019).DOI
- [6]M. V. Berry, Quantal phase factors accompanying adiabatic changes, Proceedings of the Royal Society of London A 392, 45 (1984).
- [7]V. P. Frolov and A. A. Shoom, Spinoptics in a stationary spacetime, Physical Review D 84, 044026 (2011).DOI
- [8]N. Yamamoto, Spin hall effect of gravitational waves, Physical Review D 98, 061701 (2018).DOI
- [9]V. P. Frolov, Maxwell equations in a curved spacetime: Spin optics approximation, Physical Review D 102, 084013 (2020).DOI
- [10]M. A. Oancea, J. Joudioux, I. Y. Dodin, D. E. Ruiz, C. F. Paganini, and L. Andersson, Gravitational spin hall effect of light, Physical Review D 102, 024075 (2020).DOI
- [11]L. Andersson, J. Joudioux, M. A. Oancea, and A. Raj, Propagation of polarized gravitational waves, Physical Review D 103, 044053 (2021).DOI
- [12]P. K. Dahal, Spin optics for gravitational waves, Astronomy 1, 271 (2022).DOI
- [13]A. I. Harte and M. A. Oancea, Spin hall effects and the localization of massless spinning particles, Physical Review D 105, 104061 (2022).DOI
- [14]L. Andersson and M. A. Oancea, Spin hall effects in the sky, Classical and Quantum Gravity 40, 154002 (2023).DOI
- [15]V. P. Frolov and A. A. Shoom, Gravitational spinoptics in a curved space-time, Journal of Cosmology and Astroparticle Physics 2024 (10), 039.arxiv.org/abs/2406.17905
- [16]V. P. Frolov, Spinoptics in the schwarzschild spacetime, Journal of Cosmology and Astroparticle Physics 2024 (12), 051.arxiv.org/abs/2410.19413
- [17]V. P. Frolov and A. Koek, Spinoptics in the kerr spacetime: Polarized wave scattering, Physical Review D 111, 104081 (2025).DOI
- [18]R. Ito, K. Mameda, and N. Yamamoto, Spin hall effect and berry curvature of gravitons from quantum field theory (2026), preprint, arXiv:2605.19817 [gr-qc].arxiv.org/abs/2605.19817
- [19]P. K. Dahal, Rayleigh propagation in Kerr spacetime (2023), arXiv:2301.08250 [gr-qc].DOI
- [20]G. Cusin and M. Lagos, Gravitational wave propagation beyond geometric optics, Physical Review D 101, 044041 (2020).DOI
- [21]E. Bruyère and C. Pitrou, Gravitational lensing beyond the eikonal approximation (2026), arXiv:2601.10239 [gr-qc].arxiv.org/abs/2601.10239
- [22]W. K. D. Logi and S. J. Kovács, Gravitational scattering of zero-rest-mass plane waves, Physical Review D 16, 237 (1977).
- [23]E. T. Newman and R. Penrose, An approach to gravitational radiation by a method of spin coefficients, Journal of Mathematical Physics 3, 566 (1962).
- [24]Y. F. Bautista, A. Guevara, C. Kavanagh, and J. Vines, Scattering in black hole backgrounds and higher-spin amplitudes. Part I, Journal of High Energy Physics 2023, 136 (2023).arxiv.org/abs/2107.10179
- [25]J. A. H. Futterman, F. A. Handler, and R. A. Matzner, Scattering from Black Holes (Cambridge University Press, Cambridge, 1988).
- [26]S. R. Dolan, Scattering and absorption of gravitational plane waves by rotating black holes, Classical and Quantum Gravity 25, 235002 (2008).DOI
- [27]L. C. B. Crispino, S. R. Dolan, and E. S. Oliveira, Electromagnetic wave scattering by Schwarzschild black holes, Physical Review Letters 102, 231103 (2009).DOI
- [28]S. Chandrasekhar, The Mathematical Theory of Black Holes (Clarendon Press, Oxford, 1983).
- [29]K. Ichiro Kubota, S. Arai, H. Motohashi, and S. Mukohyama, Spin wave optics for gravitational waves lensed by a Kerr black hole, Physical Review D 110, 124011 (2024).DOI
- [30]N. E. J. Bjerrum-Bohr, J. F. Donoghue, B. R. Holstein, L. Planté, and P. Vanhove, Bending of light in quantum gravity, Physical Review Letters 114, 061301 (2015).
- [31]N. E. J. Bjerrum-Bohr, J. F. Donoghue, B. R. Holstein, L. Planté, and P. Vanhove, Light-like scattering in quantum gravity, Journal of High Energy Physics 2016, 117 (2016).
- [32]M. A. Huber, A. Brandhuber, S. D. Angelis, and G. Travaglini, Eikonal phase matrix, deflection angle, and time delay in effective field theories of gravity, Physical Review D 102, 046014 (2020).DOI
- [33]M. Mehraeen, Quantum geometry of gravitons (2026), arXiv:2609.15237.arxiv.org/abs/2609.15237
- [34]G. Szegő, Orthogonal Polynomials, American Mathematical Society Colloquium Publications, Vol. 23 (American Mathematical Society, Providence, RI, 1939).