Amplitudes from Coulomb to Kerr-Taub-NUT
William T. Emond, Yu-tin Huang, Uri Kol, Nathan Moynihan, Donal O'Connell
TL;DR
The paper builds a unified framework linking electric–magnetic duality, the Newman–Janis shift, and the double copy to a network of classical solutions ranging from Coulomb to Kerr–Taub–NUT, all encoded by simple three-point amplitudes. By working in linearised gravity and summing to all orders in spin, the authors derive compact, phase-encoded amplitudes for each node of an eight-solution cube and verify the classical double copy by matching impulses computed both from geodesic/electromagnetic methods and from on-shell amplitudes. They show that Kerr–Taub–NUT is obtained from Kerr via a gravitational duality, with mass and NUT charge rotating into each other, and that the corresponding amplitudes double-copy to reproduce the gravitational impulse from the gauge-theory result. The work provides strong evidence for the Weyl double copy and elucidates how asymptotic charges transform under these operations, offering a concrete, amplitude-based testbed for all-orders-in-spin and potential extensions to higher orders. The results pave the way for broader applications of amplitudes to classical gravity and highlight the deep role of symmetry operations in connecting disparate solutions.
Abstract
Electric-magnetic duality, the Newman-Janis shift, and the double copy all act by elementary operations on three-point amplitudes. At the same time, they generate a network of interesting classical solutions spanning from the Coulomb charge via the dyon to the Kerr-Taub-NUT spacetime. We identify the amplitudes corresponding to each of these solutions, working to all orders in spin, but to leading perturbative order. We confirm that the amplitudes double-copy when the solutions are related by the classical double copy. Along the way we show that the Kerr-Taub-NUT solution corresponds to a gravitational electric-magnetic duality rotation acting on the Kerr solution, again to all orders in spin, and demonstrate that the asymptotic charges also transform simply under our operations.
