⚛️ high-energy theory

Landau singularity of the K3 maximal cut in black-hole scattering at fifth post-Minkowskian order

This paper identifies a Landau singularity in the K3 maximal cut of fifth-order post-Minkowskian black-hole scattering that causes a divergence at 94% the speed of light, and demonstrates how requiring the cancellation of this divergence fixes the gravitational wave memory contribution while deriving key coefficients via maximal cuts and boundary limits without full integration-by-parts reduction.

Original authors: Matthew D. Schwartz

Published 2026-10-28

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Matthew D. Schwartz

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

When two black holes spiral toward each other, they do not simply fall together; they dance through a warping of space and time so intense that it sends ripples across the universe. These ripples, known as gravitational waves, carry the story of the collision to detectors on Earth. To read that story accurately, scientists must build precise mathematical models of how these massive objects interact. For decades, physicists have used a method called the post-Minkowskian expansion to describe this interaction, breaking the complex gravity of the encounter into a series of simpler steps. Each step adds a layer of detail, much like adding more terms to a long equation, allowing them to predict the path of the black holes with increasing accuracy. As we prepare for the next generation of detectors, which will hear these cosmic whispers with unprecedented clarity, the need for these calculations has reached a new peak, requiring a level of precision that pushes the boundaries of current mathematics.

In this latest work, a physicist at Harvard University has tackled a specific, stubborn problem that appeared at the fifth step of this calculation. At this level of detail, which involves four loops of complex integration, the mathematics describing the scattering of black holes suddenly breaks down. When the black holes pass each other at a speed roughly 94 percent of the speed of light, a specific part of the calculation explodes into infinity. This is not a physical reality; black holes do not behave strangely at that specific speed. Instead, it is a mathematical artifact, a ghost in the machine that must be removed for the theory to make sense. The source of this ghost was traced to the geometry of a complex surface known as a K3 surface, a shape that appears in the equations governing the interaction. The researcher found that at this critical speed, the surface develops a sharp point, or a "node," that sits directly on the path the calculation must follow. This creates a pinch, forcing the numbers to blow up.

The paper demonstrates that this singularity is a specific type of mathematical obstruction known as a Landau singularity. It is not caused by a real physical process happening in the universe, but rather by the way the mathematical tools are arranged. Because the universe does not actually produce infinite forces at this speed, the infinite result must be canceled out by another part of the calculation. The only part of the theory capable of providing this cancellation is the "memory region," a component of the calculation that accounts for the permanent displacement left behind by the gravitational waves. By carefully analyzing the geometry of the K3 surface and the path of the calculation, the author showed that the singularity lies exactly where the integration path touches the sharp point of the surface. This confirms that the divergence is a second-type singularity, a pinch occurring at infinite momentum that has no counterpart in the real, classical world.

Having identified the source of the problem, the study then turned to the solution. The cancellation of this infinity depends on how the gravitational waves are routed through the calculation, a choice that was previously made by convention rather than derived from first principles. The author evaluated the memory-region integrals using a specific routing prescription that had been proposed in earlier work. By deriving the exact mathematical form of these integrals, the study confirmed that this routing successfully cancels the infinity, leaving a finite, physical result. This process fixed a crucial constant in the theory, known as $c_M$ , to the value of one. This constant is essential for predicting the scattering angle of the black holes, which in turn helps determine the waveform of the gravitational waves they emit.

The work also extended beyond the immediate problem of the singularity. The researcher mapped out the mathematical landscape of the entire calculation, identifying four distinct geometric shapes that govern the behavior of the black holes at this order of precision. These shapes include the K3 surface and more complex structures known as Calabi-Yau threefolds. The study provided a complete description of how the mathematical functions associated with these shapes behave, including how they change when the black holes are in a bound orbit, spiraling inward rather than scattering apart. This continuity is vital for connecting the theory of scattering black holes to the theory of merging ones, which is what we actually observe in the sky.

Furthermore, the paper established a set of exact relationships between the different parts of the calculation, reducing the number of independent pieces that need to be computed. It evaluated twelve specific integrals at a single speed, providing the necessary starting data to solve the entire system. These results were obtained without simplifying the full four-loop calculation, relying instead on the properties of the maximal cuts and the boundaries of the integration. The study also explored what happens if a different mathematical prescription is used for the gravitational waves. It found that using a different, more standard approach would leave the infinity uncanceled, leading to a physically impossible result. This suggests that the specific routing used to cancel the singularity is not just a convenient choice, but a necessary feature of the theory to ensure it remains finite and physical.

Ultimately, this paper provides a deeper understanding of the mathematical structure underlying the collision of black holes. It shows that the strange infinities appearing in high-precision calculations are not random errors but are tied to the specific geometry of the equations. By proving that these infinities must cancel out and showing exactly how they do so, the work removes a significant barrier to calculating the gravitational waves from merging black holes with the precision required for future observations. The findings confirm that the theory is consistent and provide the tools needed to push these calculations even further, ensuring that when the next generation of detectors listens to the universe, the models they use to interpret the sound are as sharp and clear as the signals themselves.

Technical Summary: Landau singularity of the K3 maximal cut in black-hole scattering at fifth post-Minkowskian order

Problem Statement
The calculation of gravitational waveforms from merging black holes requires precise analytic solutions to the two-body problem in General Relativity. The post-Minkowskian (PM) expansion, which expands the scattering amplitude in powers of Newton's constant $G$ while remaining exact in velocity, has reached fifth order (5PM), corresponding to four-loop integrals. At this order, specifically in the second order of the self-force expansion (5PM-2SF), a divergence appears in the potential-region contribution to the deflection angle when the relative Lorentz factor $\gamma = 3$ (approximately 94% of the speed of light). This divergence is known to arise from periods of a K3 surface (specifically the Apéry–Fermi K3) which possesses a node at this velocity. While previous work established that this unphysical divergence must be canceled by contributions from the "memory region" (where two gravitons are radiative), the analytic origin of the singularity within the K3 integrals and the specific routing of graviton propagators required to fix the remaining free constant ( $c_M$ ) in the memory contribution were not fully understood.

Methodology
The author employs techniques developed for Feynman integrals, adapted to the worldline formalism of the PM expansion, to analyze the analytic structure of the 5PM-2SF integrals without performing a full integration-by-parts (IBP) reduction of the four-loop integrand.

Maximal Cuts and Geometry: The study focuses on the maximal cuts of the integral families. The planar K3 sector (topology 40) is mapped onto the Apéry–Fermi K3 surface using the loop-by-loop Baikov representation. The integration contour is identified as a torus $T^2$ .
Landau Analysis: The author investigates whether the singularity at $\gamma=3$ is a Landau singularity. They examine the Landau equations in Feynman parameters for subtopologies and faces of the Newton polytope to identify first-type singularities (pinches at finite momentum).
Second-Type Singularity: Recognizing that the singularity does not appear in the standard Feynman parameter analysis for finite momenta, the author analyzes the loop-by-loop Baikov representation to identify a second-type singularity (a pinch at infinite loop momentum).
Periods and Differential Equations: The torus period of the maximal cut is expanded as a series of Apéry numbers. The author proves that this period satisfies the Apéry differential operator ( $L_A$ ) and derive closed-form expressions for the Frobenius coefficients of the periods for all four non-polylogarithmic geometries involved (two K3 surfaces and two Calabi-Yau threefolds).
Auxiliary-Mass Flow: The author generalizes the AMFlow algorithm to "PMflow" to compute response matrices for the dissipative sector. This allows them to find linear relations among master integrals and determine boundary constants at a specific kinematic point ( $\gamma = 5/4$ ) using Feynman propagators.
Memory Region Evaluation: The memory-region boundary integrals are evaluated in the frequency representation using the "retarded routing" prescription (averaged with time-reversed routing) to determine the constant $c_M$ .

Key Contributions and Results

Identification of the Landau Singularity: The paper demonstrates that the divergence at $\gamma = 3$ is a second-type Landau singularity of the maximal cut of the K3 sector. The node of the Apéry–Fermi K3 surface lies exactly on the integration torus of the maximal cut. This results in a "pinch" at infinite loop momentum. By the Coleman–Norton theorem, this singularity corresponds to no real classical process, confirming its unphysical nature and the necessity of cancellation.
Routing Independence of the Singularity: The analysis shows that the position of the singularity is determined by the Landau locus of the cut surface and is independent of the $i0$ prescription (routing) of the radiative gravitons. Consequently, the singularity itself does not dictate the choice of routing for the memory region.
Analytic Determination of $c_M$ : Using the frequency representation of the memory-region integrals with the specific routing prescription from Driesse et al. (retarded propagators directed toward the three-graviton vertex, averaged under time reversal), the author derives the Laurent coefficients of the boundary integrals $I^{(M)}_1$ and $I^{(M)}_3$ in closed form. They analytically confirm that this routing yields $c_M = 1$ , fixing the memory contribution up to the single constant required to cancel the divergence.
Dissipative Sector Analysis: The author provides the first-stage master-integral counts for the four dissipative families at 5PM-2SF. For the smallest family ( $PX_2$ ), they identify ten exact linear relations among master integrals using the auxiliary-mass flow response matrices. They evaluate twelve boundary master integrals at $\gamma = 5/4$ with Feynman propagators, providing explicit closed forms for their pole and finite parts.
Continuation to Bound Orbits: The Frobenius bases of the four Picard–Fuchs operators (governing the K3 and Calabi–Yau periods) are continued from the scattering region ( $\gamma > 1$ ) to bound-orbit energies ( $\gamma < 1$ ). The author computes the monodromy matrices relating continuations along different paths and determine the local exponents of the Apéry sector at the threshold ( $\gamma = 1$ ).
Closed Formulas: Closed formulas are derived for the Frobenius coefficients of the Calabi–Yau threefold period (topology 37) and the torus period of the Apéry K3 surface.

Significance
The paper clarifies the analytic origin of the $\gamma=3$ divergence in 5PM black-hole scattering, establishing it as a second-type Landau singularity of the K3 maximal cut. This geometric insight confirms that the divergence is an artifact of the integration contour and must cancel against the memory region. While the singularity does not uniquely select a routing prescription from first principles, the author analytically verifies that the specific routing proposed in previous literature leads to a consistent cancellation with $c_M=1$ . The work extends the toolkit for high-order PM calculations by applying maximal cut analysis and auxiliary-mass flows to dissipative integrals, providing boundary data and relations necessary for future full 5PM-2SF calculations involving radiation reaction. The continuation of these periods to bound kinematics provides the necessary analytic structure for mapping scattering results to effective-one-body models for inspiraling binaries.

More like this