Tidal heating and torquing of the primary black hole in eccentric-orbit, non-spinning extreme-mass-ratio inspirals to 22PN order

TL;DR

This paper computes high-order post-Newtonian expansions of energy and angular momentum fluxes onto a nonspinning black hole in eccentric EMRIs, including tidal heating and torquing effects, to improve gravitational wave modeling.

Contribution

It provides the first 22PN order analytic expansion of horizon fluxes for eccentric EMRIs, incorporating eccentricity series and resummed functions, enhancing accuracy of dissipation calculations.

Findings

Horizon absorption is less convergent than infinity flux expansion.

Useful results are obtained for orbits with p=10 and eccentricity up to 0.5.

Total dissipation known to 19PN when combined with infinity flux results.

Abstract

We calculate the high-order post-Newtonian (PN) expansion of the energy and angular momentum fluxes onto the horizon of a nonspinning black hole primary in eccentric-orbit extreme-mass-ratio inspirals. The first-order black hole perturbation theory calculation uses \textsc{Mathematica} and makes an analytic expansion of the Regge-Wheeler-Zerilli functions using the Mano-Suzuki-Takasugi formalism. The horizon absorption, or tidal heating and torquing, is calculated to 18PN relative to the leading horizon flux (i.e., 22PN order relative to the leading quadrupole flux at infinity). Each PN term is a function of eccentricity $e$ and is calculated as a series to $e^{10}$. A second expansion, to 10PN horizon-relative order (or 14PN relative to the flux at infinity), is computed deeper in eccentricity to $e^{20}$. A number of resummed closed-form functions are found for the low PN terms in the series. Using a separate Teukolsky perturbation code, numerical comparisons are made to test how accurate the PN expansion is when extended to a close $p=10$ orbit. We find that the horizon absorption expansion is not as convergent as a previously computed infinity-side flux expansion. However, given that the horizon absorption is suppressed by 4PN, useful results can be obtained even with an orbit as tight as this for $e \le 1/2$. Combining the present results with our earlier expansion of the fluxes to infinity makes the knowledge of the total dissipation known to 19PN for eccentric-orbit nonspinning EMRIs.