Loss Cone Shielding
Extracted source text · arXiv 2211.05858
\documentclass[fleqn,usenatbib]{mnras}
%\documentclass[a4paper,12pt]{article}
%\usepackage[bottom=25mm,top=25mm]{geometry}
%\usepackage{natbib}
\usepackage{framed}
\usepackage{graphicx,epsfig,psfig,multicol}
%\usepackage[normalem]{ulem}
\usepackage[dvipsnames]{xcolor}
\usepackage[]{inputenc,amssymb}
\usepackage{amsmath}
\usepackage{color}
\usepackage{epsfig}
\usepackage{epstopdf}
\usepackage{textcomp}
\usepackage{xcolor,cancel}
\usepackage{comment}
\usepackage{subcaption}
\usepackage[export]{adjustbox}% this is for telling the positioning of the figure eg. center ..
\usepackage{adjustbox}
\usepackage{placeins}% This is for using the float barrier ...
\usepackage{float}
\usepackage{morefloats}
\usepackage{lipsum}
\usepackage{mwe}
\usepackage{arydshln} %for dashed line
\definecolor{webgreen}{rgb}{0,.5,0}
\definecolor{webbrown}{rgb}{.6,0,0}
%\usepackage[pdfpagelabels]{hyperref}
%\begin{comment}
%\hypersetup{%
% colorlinks=true,hyperfootnotes=false,%
% breaklinks=true,%
% plainpages=false, bookmarksnumbered, bookmarksopen=true,
% bookmarksopenlevel=1,%
% urlcolor=webbrown, linkcolor=RoyalBlue, citecolor=webgreen,
% }
%\end{comment}
\usepackage[toc,page]{appendix}
\def \tc{\textcolor{magenta}}
\def \beq{\begin{equation}}
\def \eeq{\end{equation}}
\def \bea{\begin{eqnarray}}
\def \eea{\end{eqnarray}}
\newcommand{\OmgK}{\mbox{${\Omega}_{\rm K}$}}
\newcommand{\rmi}{\mbox{ ${\rm i}$ }}
\newcommand{\p}{\partial}
\newcommand{\rmd}{\mbox{${\rm d}$}}
\newcommand{\re}{\mbox{${\rm Re}$}}
\newcommand{\im}{\mbox{${\rm Im}$}}
\newcommand{\Msun}{\mbox{$ M_\odot $ }}
\newcommand{\cross}{\mbox{$\pmb{\times}$}}
\newcommand{\opunit}{\mbox{${\rm cm}^2/{\rm g}$}}
\newcommand{\Pmag}{\mbox{$P_{\rm m}$}}
\newcommand{\Prad}{\mbox{$P_{\rm r}$}}
\newcommand{\Pgas}{\mbox{$P_{\rm g}$}}
\newcommand{\kes}{\mbox{$\kappa_{\rm es}$}}
\newcommand{\mathL}{\mbox{$\mathcal{L}$}}
\newcommand{\mP}{\mbox{$m_{\rm p}$}}
\newcommand{\vK}{\mbox{$v_{\rm K}$}}
\newcommand{\mathF}{\mbox{$\mathcal{F}$}}
\newcommand{\sigSB}{\mbox{$\sigma_{\rm SB}$}}
\newcommand{\kB}{\mbox{$k_{\rm B}$}}
\newcommand{\qadv}{\mbox{$q_{\rm adv}^{-}$}}
\newcommand{\qrad}{\mbox{$q_{\rm rad}^{-}$}}
\newcommand{\lamP}{\mbox{$\lambda_{P} $}}
\newcommand{\Crad}{\mbox{$\mathcal{C}_{\rm rad}$}}
\newcommand{\Hs}{\mbox{$H_{\rm s}$}}
\newcommand{\rhos}{\mbox{$\rho_{\rm s}$}}
\newcommand{\rhocrit}{\mbox{$\rho_{\rm crit}$}}
\newcommand{\mucrit}{\mbox{$\mu_{\rm crit}$}}
\newcommand{\Ps}{\mbox{$P_{\rm s}$}}
\newcommand{\Tk}{\mbox{$T_{\rm k}$}}
\newcommand{\MdotEdd}{\mbox{$\dot{M}_{\rm Edd}$}}
\newcommand{\alT}{\mbox{$\alpha_{T} $}}
\newcommand{\Deltaa}{\mbox{$\Delta_{\rm a}$} }
\newcommand{\Deltab}{\mbox{$\Delta_{\rm b}$} }
\newcommand{\rhoa}{\mbox{$\rho_{\rm a}$}}
\newcommand{\rhob}{\mbox{$\rho_{\rm b}$}}
\newcommand{\alrho}{\mbox{$\alpha_{\rho} $}}
\newcommand{\zetaP}{\mbox{$\zeta_P$}}
\newcommand{\rcrit}{\mbox{$r_{\rm crit}$}}
\newcommand{\tf}{\mbox{$t_{\rm f}$}}
\newcommand{\tcrit}{\mbox{$t_{\rm crit}$}}
\newcommand{\rev}[1]{\textcolor{black}{#1}}
\def \cblue {\textcolor{blue}}
\def \cred {\textcolor{red}}
\def \ccyan {\textcolor{cyan}}
\def \cgreen {\textcolor{green}}
\def \cmag{\textcolor{magenta}}
%\usepackage{hyperref}
% \hypersetup{ colorlinks=true,
% linkcolor=blue,
% citecolor = blue,
% urlcolor=blue}
\usepackage[T1]{fontenc}
\DeclareRobustCommand{\VAN}[3]{#2}
\let\VANthebibliography\thebibliography
\def\thebibliography{\DeclareRobustCommand{\VAN}[3]{##3}\VANthebibliography}
%\setlength{\textwidth}{6.5in}
%\setlength{\oddsidemargin}{-6mm}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%% TITLE PAGE %%%%%%%%%%%%%%%%%%%
% Title of the paper, and the short title which is used in the headers.
% Keep the title short and informative.
\title[Loss Cone Shielding]{Loss Cone Shielding}
% The list of authors, and the short list which is used in the headers.
% If you need two or more lines of authors, add an extra line using \newauthor
\author[O. Teboul et al.]{
Odelia Teboul$^1$\thanks{E-mail: odelia.teboul1@mail.huji.ac.il},
Nicholas C. Stone$^1$, Jeremiah P. Ostriker$^2$
\\
% List of institutions
$^1$Racah Institute of Physics, The Hebrew University, 91904, Jerusalem, Israel
\\
$^2$Department of Astronomy, Columbia University, 550 West 120th St, New York, NY 10027, USA}
% These dates will be filled out by the publisher
\date{Accepted 2023 October 23. Received 2023 October 03; in original form 2022 November 17}
% Enter the current year, for the copyright statements etc.
\pubyear{2022}
\begin{document}
\twocolumn
\label{firstpage}
\pagerange{\pageref{firstpage}--\pageref{lastpage}}
\maketitle
\begin{abstract}
%A star wandering close enough to a massive black hole (MBH) can be ripped apart by the tidal forces of the black hole. The advent of wide-sky surveys at all wavelengths has quickly increased the number of tidal disruption events (TDEs) observed, and has revealed that i) observed TDE rates are lower than theoretical predictions and ii) E+A galaxies are significantly overrepresented. This overrepresentation further worsens the tension between observed and theoretically predicted TDEs for non-E+A galaxies. Classical loss cone theory focuses on the cumulative effect of many weak scatterings. However, a strong scattering can remove the star from the distribution before it could get tidally disrupted. Most stars undergoing TDEs come from within the radius of influence, {\it the densest environments of the universe}. In such environments, close encounters rare elsewhere become non-negligible. Hence, in {\it our revised loss cone theory version} we take into account classical two-body interactions as well as strong scattering, collisions, tidal captures, and study {\it under which conditions close encounters can shield the loss cone}. We i) analytically derive the impact of strong scattering and other close encounters ii) compute the loss cone dynamics evolution including both weak and strong encounters, and iii) derive analytical solutions to the Fokker-Planck equation with strong scattering. We find that i) TDE rates can be reduced to up to an order of magnitude ii) shielding reduces preferentially deeply plunging stars. We also show that stellar overdensities, one possible explanation for the E+A preference, can fail to increase TDE rates when taking into account strong scattering.
A star wandering close enough to a massive black hole (MBH) can be ripped apart by the tidal forces of the black hole. The advent of wide-field surveys at many wavelengths has quickly increased the number of tidal disruption events (TDEs) observed, and has revealed that i) observed TDE rates are lower than theoretical predictions and ii) E+A galaxies are significantly overrepresented. This overrepresentation further worsens the tension between observed and theoretically predicted TDEs for non-E+A galaxies. Classical loss cone theory focuses on the cumulative effect of many weak scatterings. However, a strong scattering can remove a star from the distribution before it can get tidally disrupted. Most stars undergoing TDEs come from within the radius of influence, the densest environments of the universe. In such environments, close encounters rare elsewhere become non-negligible. We revise the standard loss cone theory to take into account classical two-body interactions as well as strong scattering, collisions, tidal captures, and study under which conditions close encounters can shield the loss cone. We i) analytically derive the impact of strong scattering and other close encounters, ii) compute time-dependent loss cone dynamics including both weak and strong encounters, and iii) derive analytical solutions to the Fokker-Planck equation with strong scattering. We find that i) TDE rates can be reduced to up to an order of magnitude and ii) strong shielding preferentially reduces deeply plunging stars. We also show that stellar overdensities, one possible explanation for the E+A preference, can fail to increase TDE rates when taking into account strong scattering.
%Stars in galactic nuclei can be torn apart in tidal disruption events (TDEs) if two-body scatterings shift their orbits to a low angular momentum region, the ``loss cone.'' Loss cone refilling and TDE rate predictions are usually modeled in the diffusive limit of scattering. Here we show that neglected strong scatterings can play a major role in TDE rates if there is a population of steeply distributed (e.g. strongly segregated) stellar mass black holes. Strong scatterings will preferentially eject stars on nearly radial orbits, shielding the loss cone from its diffusive flux of stars and reducing TDE rates. By obtaining time-dependent solutions to the Fokker-Planck equation (both analytically and numerically), we find that for steep black hole distributions, the total TDE rate can be reduced by one order of magnitude, with a preferential reduction of deeply plunging TDE rates. Steeply cusped populations of stellar mass black holes offer a possible resolution to persistent discrepancies between observed (low) and predicted (high) TDE rates in low-mass galactic nuclei, and may also constrain some explanations for the post-starburst preference exhibited by TDE host galaxies.
\end{abstract}
% Select between one and six entries from the list of approved keywords.
% Don't make up new ones.
\begin{keywords}
transients: tidal disruption events – galaxies: kinematics and dynamics – stars: kinematics and dynamics – methods: analytical - methods: numerical
\end{keywords}
%\fontsize{12}{15}
%\selectfont
%\twocolumn
%\begin{comment}
\section{Introduction}
\label{sec:intro}
A star wandering too close to a massive black hole (MBH) can be ripped apart by the MBH's tidal forces. Roughly half of the gaseous debris from the disrupted star falls back onto the MBH, eventually circularizing into an accretion disk and powering a luminous flare \citep{Rees88, EvansKochanek89}.
Tidal disruption events (TDEs) have been theoretically predicted since the 1970s (e.g., \citealt{Hills75}) but were not observed until the advent of all-sky surveys \citep[beginning with the ROSAT X-ray all-sky survey;][]{Bade, Komossa}. The sample of observed TDEs has rapidly grown in the last decades and now spans all wavelengths, from the radio to gamma-rays. The last decade has seen a high rate of TDE discovery in wide-field optical surveys such as SDSS \citep{vanVelzen11}, Pan-STARRS \citep{Gezari, Chornock,Nicholl}, PTF \citep{Arcavi+14}, iPTF \citep{Blagorodnova, Hung}, ASASSN \citep{Holoien16, Hinkle}, and ZTF \citep{vanVelzen+19, Hammerstein22}. X-ray surveys are also returning to prominence as a driver of TDE discovery; at present, the SRG/eROSITA all-sky survey \citep{Sazonov} seems to be finding $\sim 20$ TDEs per year, comparable to the optical TDE discovery rate of ZTF. As time domain capabilities continue to expand, the observed sample of TDEs is expected to grow by orders of magnitude, with hundreds to thousands of new TDEs expected to be found in optical wavelengths by the LSST/Vera Rubin Observatory \citep{Ivezi, BricmanGomboc20}, %Athena observatory \citep{Barcons},
and in the near-ultraviolet by ULTRASAT \citep{Ultrasat}.
The opening of this new era for TDEs has allowed observers to empirically constrain the per-galaxy TDE rate to a range $\dot{N}_{\rm TDE} \sim 10^{-5} - 10^{-4}$ yr$^{-1}$~ gal$^{-1} $ \citep{Holoien16, vanVelzen18}, and revealed a surprising over-representation of these events in post-starburst (e.g. ``E+A'') galaxies \citep{Arcavi+14, French+16,French+17, Law, Graur, French20, Hammerstein21}. The latest estimations suggest that E + A galaxies are overrepresented by a factor of $ \approx 22 $ \citep{Hammerstein21}. %The observed TDE rates in non-E+A is therefore reduced by a factor of $\sim 3$ hampering reconciliation with the theoretical estimates of classical loss cone theory $\dot{N} \sim 10^{-4} - 10^{-3}$ yr$^{-1}.$ gal$^{-1} $ \citep{Wang}.
This further reduces the observationally inferred TDE rate in non-E+A galaxies, which is already in tension with theoretical (but empirically calibrated) rate estimates of $\dot{N} \sim 10^{-4} - 10^{-3}$ yr$^{-1}$~ gal$^{-1}$ (e.g., \citealt{WangMerritt04, StoneMetzger16}).
TDE rates are traditionally estimated from classical loss cone theory, which focuses on the cumulative effect of many weak scatterings. These scatterings are modeled as local and uncorrelated, and lead to effective diffusion coefficients describing the drift and diffusion of stellar populations through phase space \cite[see][for reviews]{Merritt13, Stone+20}. Strong, or small impact parameter scatterings are generally neglected, as they are largely outnumbered by weak scatterings. %weak scatterings %outnumber strong scattering by a factor .
%the net effect of strong scatterings on stellar diffusion is suppressed by a Coulomb logarithm $\ln \Lambda \sim \mathcal{O}(10-20)$ relative to the net effect of many weak scatterings.
However, strong scatterings can have a qualitatively different effect than weak ones on stellar populations. In particular, {\it a strong scattering can eject a star from the distribution} before it gets tidally disrupted. Most stars undergoing TDEs come from within the radius of influence, {\it the densest environments in the Universe}. Such an environment can facilitate ejection in strong scatterings, but it also increases the rate of close encounters that are rare elsewhere. Such close encounters can be divided into star-star interactions, including collisions and tidal captures, and star-stellar mass black hole (sBH) interactions, such as $\mu$-TDEs and tidal captures. In this paper we will revise the standard loss cone theory to take into account not simply weak two-body scatterings, but also destructive processes such as strong scattering ejections, collisions, and tidal captures, with the ultimate goal of determining under which conditions close encounters can {\it shield the loss cone}.
Close encounters are highly dependent on both the stellar and sBH slopes. An important prediction of stellar dynamics is the redistribution of arbitrary initial distributions of stars around MBHs over time into quasi-stationary states (QSSs). In the presence of one population of stars, a Bahcall-Wolf density cusp with stellar density $\rho \propto r^{-\gamma} $, will develop, with $\gamma=7/4$ \citep{BahcallWolf76}. If the system is composed of both stars and heavier objects, the heavier objects are expected to segregate towards the centre of the galactic nucleus and settle on a steeper cusp $ \gamma = 7/4 - 2$ while the light objects will have a weaker cusp $\gamma \approx 1.3 -1.5 $ \citep{BW77, PretoAmaroSeoane10, Amaro, Broggi}. In the strong segregation limit, when the heavier objects are relatively rare, the heavy objects have been predicted to settle to even steeper power-law slopes of $\gamma = 2 - 11/4 $ \citep{AlexanderHopman09}. \rev{Moreover for an adiabatically growing MBH, the stellar slope may assume steeper values of $\gamma \gtrsim 2$ \citep{Young}.}
We introduce the different kinds of close encounters that are relevant for our problem in \S \ref{sec:close}, and derive their analytical rates in \S \ref{sec:strong}. In \S \ref{sec:loss}, we solve the Fokker-Planck equation in angular momentum space in the presence of close encounters. In \ref{sec:analytical} we present our analytical time-dependent solutions of the Fokker-Planck equation with strong scattering. In \S \ref{sec:results} we compute the impact of close encounters on TDE rates and discuss our findings for the E+A preference. In \S \ref{sec:conclusions}, we discuss the implications of our work and briefly summarize it. %Useful analytic limits and test cases are given in Appendix \ref{app:analytics}.
\begin{comment}
\section{Introduction old }
Massive black holes (MBHs) appear to be ubiquitous in the nuclei of all galaxies above a certain mass threshold \citep{Soltan82, YuTremaine02, Heckman+04, GreeneHo07}. The presence of an MBH can dramatically alter the dynamics and evolution of their host galaxy \citep{FerrareseMerritt00, Tremaine+02, KormendyHo13}. One consequence of particular interest for transient astronomy is the creation of a ``loss cone'' in the phase space distribution of particles (e.g. stars and stellar-mass compact objects) orbiting the MBH \citep{FrankRees76, LightmanShapiro77, CohnKulsrud78}. In dense star clusters without central MBHs, there is no physical lower limit on the orbital angular momentum that can be achieved by a particle: zero-angular momentum orbits are permissible, and in principle may be long-lasting. The presence of a central MBH carves a conical gap in velocity space (at fixed radius), or equivalently places a non-zero lower limit on the angular momentum that particle orbits may have. Particles with less angular momentum suffer destructive interactions at pericenter. Single stars may be torn apart by tidal forces \citep{Hills75, Rees88}, giving rise to a tidal disruption event (TDE); binary stars may be separated by tides, creating hypervelocity stars \citep{Hills88}; compact objects too dense to be susceptible to tides may be captured by the MBH event horizon in a direct plunge, or alternatively may lose orbital energy to gravitational wave emission and begin an extreme-mass-ratio-inspiral (EMRI; \citealt{Ryan95, SigurdssonRees97}).
Arbitrary initial distributions of stars around MBHs will relax over time to quasi-stationary states (QSSs). Different dynamical processes drive this evolution but the most ubiquitous is two-body (sometimes ``non-resonant'') relaxation, the uncorrelated scattering of individual stars off each other in pairwise interactions. The usual theory of two-body relaxation in gravitational systems focuses on the cumulative effect of many weak, or diffusive, scatterings. These scatterings are modeled as local and uncorrelated, and lead to effective diffusion coefficients describing the drift and diffusion of stellar populations through phase space \cite[see][for a pedagogical review]{Merritt13}. Strong, or large-angle, scatterings are generally neglected, and this neglect is justifiable in some contexts: for example, one can show that the net effect of strong scatterings on stellar diffusion is suppressed by a Coulomb logarithm $\ln \Lambda \sim \mathcal{O}(10-20)$ relative to the net effect of many weak scatterings.
The simplest version of loss cone theory applies to stellar systems with spherical symmetry. In this case, one may orbit-average local diffusion coefficients in velocity space to compute fluxes of particles in the space of angular momentum $L$. The resulting QSS in $L$-space differs only logarithmically, by a factor $\approx \ln(L^2 / L_0^2)$, from the isotropic distribution that would otherwise emerge in the absence of a loss cone \citep{CohnKulsrud78}. When diffusion is slow, $L_0$ is the angular momentum of an orbit grazing the loss cone boundary, but when diffusion is fast, $L_0$ can be much smaller. The simple loss cone theory sketched above has been generalized to axisymmetric \citep{MagorrianTremaine99, VasilievMerritt13} and even triaxial star systems \citep{MerrittPoon04}, from Newtonian to general relativistic gravity \citep{Kesden12}, and some recent works have even examined how strong scatterings may alter the $L$ distribution of orbits entering the loss cone \citep{WeissbeinSari17}.
The predictions of loss cone theory are most commonly compared to observations in the case of TDEs\footnote{Current studies of hypervelocity stars in the Milky Way suffer from greater selection effects and false positives, making it hard to infer rates; EMRIs will not be detected until a low-frequency gravitational wave interferometer such as {\it LISA} is operating in space.} (see \citealt{Stone+20} for a recent review). While significant uncertainties exist, both with observational rate inferences and theoretical rate estimates, two observational surprises have emerged in the last few years. First, the observed rate of all TDEs appears to be lower than theoretically predicted estimates \citep{StoneMetzger16}. Two explanations include a steep TDE luminosity function (extending down to faint luminosities; \citealt{vanVelzen18}) and survey incompleteness due to dust extinction \citep{Roth+21}. This problem is worsened, however, by a second observational surprise: the relatively rare population of post-starburst galaxies ($\sim 1\%$ of all galaxies in the low-$z$ Universe) host an order unity fraction of all observed TDEs \citep{Arcavi+14, French+16, French+17}, suggesting that some unknown dynamical process is enhancing the rate of diffusion into the loss cone \citep{Stone+18, Madigan+18}.
Motivated by the first of these rate discrepancies, in this paper we develop a theory of loss cone evolution in the presence of {\it destructive} strong scatterings. One qualitative difference between weak and strong scatterings is that a sufficiently close encounter may permanently remove a star from the system it is bound to: either through physical destruction \citep{FrankRees76}, common at the smallest radii, or through ejection and escape \citep{LinTremaine80}, common at larger radii. Because two-body relaxation rates tend to be dominated by the heaviest surviving stellar species, we expect that this effect will be the strongest when a population of stellar-mass black holes (sBHs) is tightly bound to the central MBH. The sBHs may effectively ``shield'' the MBH from receiving stars that would normally enter its loss cone, as these stars would first have to survive many steps in their random walk through $L$-space in regions of high sBH density.
Although our primary motivation concerns the overall rate of TDEs in all galaxies, our investigation may also have implications for the apparent preference of observed TDEs for post-starburst hosts. As we will show, loss cone shielding effects depend sensitively on the slope of the stellar (and sBH) density profile $\rho(r)$, and in particular will only become notable for profiles steeper than the canonical QSS in energy space, the $\rho(r) \propto r^{-7/4}$ solution \citep{BahcallWolf76}. Such ``ultrasteep'' density profiles can emerge as the result of centrally concentrated star formation \citep{Generozov+18}, where a $\rho(r) \propto r^{-5/2}$ profile can be produced outside the region of star production/deposition. Quasi-stationary solutions for heavy objects (such as sBHs) may also emerge ($\rho(r) \propto r^{-\gamma}$, with $2 \le \gamma \le 11/4$) in the strong segregation limit\footnote{See however \citet{LinialSari22} for a different set of predictions on the outcome of strong segregation.} where light objects dominate the diffusion coefficients \citep{AlexanderHopman09}, an effect which has been seen in $N$-body simulations to produce $\gamma \approx 2$ power-law slopes \citep{PretoAmaroSeoane10}. Apart from theoretical predictions such as these, however, the strongest {\it observational} evidence for ultrasteep density profiles comes from post-starburst galaxies, where several of the nearest examples have been seen to have profiles with $\gamma \approx 2.25$ on scales of a few pc \citep{StonevanVelzen16, Li+22}. It is therefore possible that loss cone shielding may decrease TDE rates in post-starburst galaxies, posing an even greater observational puzzle.
We briefly introduce the types of close encounters that are relevant for our problem in \S \ref{sec:close}, and develop the formalism to estimate orbit-averaged loss rates in \S \ref{sec:strong}. In \S \ref{sec:loss}, we solve the Fokker-Planck equation in 1D ($L$-space) to determine how the classic Cohn-Kulsrud QSS is modified by the presence of strong scatterings and other close encounters. In \S \ref{sec:obs} we discuss the observational implications of these results, and we briefly summarize our work in \S \ref{sec:conclusions}. Useful analytic limits and test cases are given in Appendix \ref{app:analytics}.
\end{comment}
\section{Close Encounters}
\label{sec:close}
In a spherical galactic nucleus one can characterize the population of stars with an orbit-averaged distribution function \rev{ $f(\epsilon, J)$}, where $\epsilon$ is the orbital energy and $J$ the orbital angular momentum of stars. A variety of processes can remove stars from the distribution $f$. Most notably, a star of mass $M_\star$ and radius $R_\star$ will be destroyed by the central MBH (of mass $M_\bullet$) if it completes an orbit with galactocentric pericenter $R_{\rm p}$ less than the tidal radius,
\begin{equation}
R_{\rm t} = R_\star \left( \frac{M_\bullet}{M_\star}\right)^{1/3}. \label{eq:rt}
\end{equation}
A finite $R_{\rm t}$ (i.e. the presence of a central MBH) carves a conical gap in velocity space (at fixed radius), or equivalently places a non-zero lower limit on the angular momentum that particle orbits may have \citep{FrankRees76, LightmanShapiro77, CohnKulsrud78}. Particles with less angular momentum suffer destructive interactions at pericenter.
%When spherical symmetry applies, one may orbit-average local diffusion coefficients in velocity space to compute fluxes of particles in the space of angular momentum $L$. The resulting QSS in $L$-space differs only logarithmically, by a factor $\approx \ln(L^2 / L_0^2)$, from the isotropic distribution that would otherwise emerge in the absence of a loss cone \citep{CohnKulsrud78}. When diffusion is slow, $L_0$ is the angular momentum of an orbit grazing the loss cone boundary, but when diffusion is fast, $L_0$ can be much smaller. The simple loss cone theory sketched above has been generalized to axisymmetric \citep{MagorrianTremaine99, VasilievMerritt13} and even triaxial star systems \citep{MerrittPoon04}, from Newtonian to general relativistic gravity \citep{Kesden12}, and some recent works have even examined how strong scatterings may alter the $L$ distribution of orbits entering the loss cone \citep{WeissbeinSari17}.
However, before the typical star can get to an orbit with $R_{\rm p} \sim R_{\rm t}$, its pericenter must endure a long and potentially dangerous random walk through regions of high stellar and sBH density. During this random walk, sufficiently close encounters with the stellar/sBH population can eliminate our test star through a variety of different processes. Here we overview the relevant ones, roughly in order of increasing cross-section:
\begin{enumerate}
\item Direct physical collisions: star-star (or star-sBH) collisions with a pairwise pericenter $p<R_\star$ will often destroy the test star. At large distances $r$ from the MBH, where the velocity dispersion $\sigma \ll V_\star=\sqrt{2GM_\star/R_\star}$, the stellar surface escape speed, collisions with other stars can be coagulative rather than destructive\footnote{As we are focused on relatively small spatial scales, we neglect this possibility in this paper.}.
\item $\mu$-TDEs: slightly more distant encounters with a sBH ($p < r_{\rm t}$, where $r_{\rm t}$ is the same as $R_{\rm t}$ from Eq. \ref{eq:rt} but with the sBH mass $m_{\rm bh}$ substituted into Eq. \ref{eq:rt} for $M$), can destroy the test star through the action of the sBH's tides, in a scaled down version of a standard TDE. The observational signatures of these ``$\mu$-TDEs'' are poorly understood \citep{Perets+16}. However, the different dynamics and greatly reduced Eddington limit of an sBH (compared to an MBH) means that the resulting transient is unlikely to be mistaken for a standard TDE, and a star experiencing this fate can be safely removed from the distribution function.
\item Tidal captures: at slightly larger pericenter distances than the $\mu$-TDE radius, tides can excite mechanical oscillation modes in the test star. If this process converts enough orbital energy into mechanical energy, then an unbound parabolic/hyperbolic two-body orbit will capture into a bound binary \citep{Fabian+75, PressTeukolsky77, LeeOstriker86}. A tidal capture can only occur when the relative kinetic energy between the two objects is less than the binding energy of the test star. The long-term evolution of a tidal capture system is not entirely clear \citep{Generozov+18}. Some recent hydrodynamical simulations indicate that the captured star will be destroyed in a brief sequence of runaway partial (micro)-TDEs \citep{Kremer+22}, but even if this is not the case, a star that has become tightly bound to a sBH is unlikely to ever experience a TDE. If a tight binary's center-of-mass angular momentum starts falling to levels approaching the loss cone, then direct merger (i.e. $\mu$-TDE) becomes the most probable outcome \citep{Bradnick+17}, followed by tidal separation via the Hills mechanism. Star-star tidal captures are also possible, but again the most likely final result (if the binary is excited towards small $p$ values) is a merger, which would effectively remove one star from the distribution function.
\item Ejections: a strong (large-angle) scattering between the test star and another stellar-mass object can accelerate it above the local escape speed, $v_{\rm esc}$. When this happens, the star is removed from the system \citep{LinTremaine80} and can, under certain conditions, become a hypervelocity star \citep{YuTremaine03, OLearyLoeb08}.
%We note that some recent efforts have examined the role of strong scatterings in affecting the final parameters of loss cone orbits \citep{WeissbeinSari17, WeissbeinSari21}, but this is distinct from ejection and loss cone shielding.
\end{enumerate}
We will now calculate the different rates of the processes that can remove stars from the distribution. We will begin with an order of magnitude estimate in §\ref{sec:approx} and then present our analytical derivations for strong scattering in §\ref{sec:strong}, star-star interactions in §\ref{sec:star} and star-stellar mass black hole interactions in §\ref{sec:bh} .
\subsection{Approximate Description}
\label{sec:approx}
%$\rho_{\star} \propto r^{- \gamma_{\star}} $
%$\rho_{bh} \propto r^{- \gamma_{bh}} $
The rates of processes (i-iv) are governed by basically the same underlying physics, and can be calculated in the usual ``$n-\Sigma-v$'' way as
\begin{equation}
\dot{N} = n \pi p_{\rm c}^2 \left(1+ \frac{2G(M_\star+ \rev{M_2})}{p_{\rm c}v^2} \right) v_\infty,
\end{equation}
where $\dot{N}$ is the per-target rate of close encounters, $p_{\rm c}$ is the critical pericenter for the close encounter (e.g. direct collision, $\mu$-TDE, or tidal capture) to occur, $v_\infty$ is the relative velocity at infinity, $M_\star$ is the mass of the target star, \rev{ $M_{2}$} is the mass of the secondary (another star for star-star encounters or a sBH for TCs/$\mu$-TDEs), and $n$ is the number density of secondaries.
Far from the central MBH, the rates of these close encounters will be set by gravitational focusing, but close to the MBH, encounter rates will be determined by straight-line flyby trajectories. We will now approximate the rates of close encounters in these two regimes by making the simplified assumption that all stars have the same mass $M_\star$ and are distributed in space with a power-law $n_\star(r) \propto r^{-\gamma_\star}$, and likewise that all sBHs have a common mass $m_{\rm bh}$ and a power-law density profile $n_{\rm bh}(r) \propto r^{-\gamma_{\rm bh}}$. We further approximate the situation by assuming that all relative velocities are equal to the stellar velocity dispersion,
\begin{equation}
\sigma_\star = \sqrt{\frac{GM_{\bullet}}{r(1+\gamma_\star)}},
\end{equation}
sparing us the need to integrate over a relative velocity distribution. Under these simplifying assumptions, the (per-target) rates of star-star collisions, star-sBH $\mu$-TDEs, and star-sBH tidal captures scale, respectively, as
\begin{align}
\dot{N}_{\rm coll} \propto \label{eq:NDotCollTwiddle}
\begin{cases}
R_\star^2 r^{-\gamma_\star-1/2}, & r < r_{\rm coll}^{\rm tr} \\
R_\star M_\star r^{-\gamma_\star+1/2}, & r > r_{\rm coll}^{\rm tr}
\end{cases} \\
\dot{N}_{\rm \mu TDE} \propto \label{eq:NDotMuTDETwiddle}
\begin{cases}
R_\star^2 \alpha^{-2/3} r^{-\gamma_{\rm bh}-1/2}, & r < r_{\rm \mu TDE}^{\rm tr} \\
R_\star m_{\rm bh} (1+\alpha)\alpha^{-1/3} r^{-\gamma_{\rm bh}+1/2}, & r > r_{\rm \mu TDE}^{\rm tr}
\end{cases} \\
\dot{N}_{\rm TC} \propto \label{eq:NDotTCTwiddle}
\begin{cases}
\lambda^2 R_\star^2 \alpha^{-2/3} r^{-\gamma_{\rm bh}-1/2}, & r < r_{\rm TC}^{\rm tr} \\
\lambda R_\star m_{\rm bh} (1+\alpha) \alpha^{-1/3} r^{-\gamma_{\rm bh}+1/2}, & r > r_{\rm TC}^{\rm tr}.
\end{cases}
\end{align}
Here the transition radii between the focused and unfocused interaction regimes are
\begin{align}
r_{\rm coll}^{\rm tr} =& \frac{R_\star}{4(1+\gamma_\star)} \frac{M_{\bullet}}{M_\star} \approx 2\times 10^{-3}~{\rm pc}~M_6 r_\star m_\star^{-1} \label{eq:rtrans1} \\
r_{\rm \mu TDE}^{\rm tr} =& \frac{R_\star}{2(1+\gamma_{\rm bh})} \frac{\alpha^{2/3}}{1+\alpha} \frac{M_{\bullet}}{M_\star} \approx 8\times 10^{-4}~{\rm pc}~M_6 r_\star m_\star^{-1} \label{eq:rtrans2} \\
r_{\rm TC}^{\rm tr} =& \frac{\lambda R_\star}{2(1+\gamma_{\rm bh})} \frac{\alpha^{2/3}}{1+\alpha} \frac{M_{\bullet}}{M_\star} \approx 2\times 10^{-3}~{\rm pc}~M_6 r_\star m_\star^{-1}, \label{eq:rtrans3}
\end{align}
and we have further defined $\alpha = M_\star/m_{\rm bh}$ as the sBH-star mass ratio, and $\lambda \approx 2$ as the rough enhancement to the $\mu$-TDE cross-section for tidal captures\footnote{In reality, $\lambda \sim \mathcal{O}(1)$ depends weakly on the local properties of the background cluster \citep{LeeOstriker86}.}. In the approximate equalities above, we have taken $\gamma_\star = \gamma_{\rm bh} = 7/4$, $\lambda = 2$, and $\alpha = 0.1$ for concreteness, and used the shorthand variables $M_6 = M_\bullet / (10^6 M_\odot)$, $m_\star = M_\star / M_\odot$, and $r_\star = R_\star / R_\odot$.
It is important to note that tidal capture becomes impossible when the relative kinetic energy of the target and secondary becomes larger than the maximum amount of mechanical energy that can be injected into the target star's normal modes in a single pericenter passage. Crudely approximating this condition as $M_\star m_{\rm bh} (M_\star + m_{\rm bh})^{-1} \sigma^2/2 > u_\star GM_\star^2 / R_\star$, where $u_\star <1$ is a fudge factor representing the fraction of the star's binding energy that can be absorbed by mechanical oscillations, we find that tidal capture deactivates at radii below
\begin{equation}
r_{\rm TC} = \frac{R_\star}{2u_\star (1+\gamma_\star)} \frac{1}{1+\alpha} \frac{M}{M_\star}.
\end{equation}
Because $r_{\rm TC} \sim r_{\rm TC}^{\rm tr}\alpha^{-2/3}u_\star^{-1}$, and most likely $u_\star \sim \mathcal{O}(0.1)$, $r_{\rm TC}$ is generally greater than $r_{\rm TC}^{\rm tr}$ and tidal captures are always gravitationally focused.
We also want to understand the basic scaling laws describing rates of stellar ejection through strong scatterings. We approximate the specific impulse delivered to the target star in a strong scattering as $\Delta v \sim (G m_2 / p^2)(p / v_p)$, with $v_p$ the velocity at pericenter. Angular momentum conservation lets us rewrite this in terms of the velocity ($v_\infty$) and impact parameter ($b$) at infinity: $\Delta v \sim G m_2 / (b v_\infty)$. Thus the critical impact parameter that allows strong scatterings to eject stars will be $b_{\rm ej} = Gm_2 / (v \Delta v_{\rm ej}$), where $\Delta v_{\rm ej}$ is the critical specific impulse to eject a star. While it is tempting to equate $\Delta v_{\rm ej} = v_{\rm esc}(r) = \sqrt{2GM/r}$, we must remember that for stars on loss cone orbits, $1-e \ll 1$ and most of the orbit is spent traveling at speeds $v(r)\approx v_{\rm esc}(r)$. Therefore we approximate $\delta v_{\rm ej} \approx v_{\rm esc}(r) - v(r)$, or
\begin{equation}
\Delta v_{\rm ej} \approx \frac{1}{2\sqrt{2}} \sqrt{\frac{GMr}{a^2}}
\end{equation}
for a star on a highly eccentric orbit with $r \ll a$. We now see that the critical impact parameter for ejection is $b_{\rm ej} \sim 2\sqrt{2} a m_2 / M$. The local ejection rate from scatterers with number density $n$ will thus be $\dot{N}_{\rm ej} = n \pi b_{\rm ej}^2 v_\infty$, so that
\begin{equation}
\dot{N}_{\rm ej} \propto a^2 m_2^2 r^{-\gamma - 1/2}. \label{eq:NDotEjTwiddle}
\end{equation}
Here we have left $m_2$ and $\gamma$ general, as they may represent either star-star or star-sBH strong scatterings. In contrast to Eqs. \ref{eq:NDotCollTwiddle}-\ref{eq:NDotTCTwiddle}, there is no leading-order distinction between gravitationally focused or unfocused regimes for strong scatterings. Furthermore, there is no leading-order dependence on the mass $m_\star$ of the test star experiencing the scattering (although note that this equation implicitly assumes $m_\star \lesssim m_2$).
While Eqs. \ref{eq:NDotCollTwiddle}-\ref{eq:NDotTCTwiddle} and \ref{eq:NDotEjTwiddle} give local encounter rates $\dot{N}(r)$, we also want to compute their orbit-averaged versions, $\langle \dot{N} \rangle$. Because of our ultimate interest in the loss cone, we consider highly eccentric orbits with pericenter $p \ll a$, the stellar semimajor axis. At the order of magnitude level, we can approximate the orbit-averaged rate of some process at a given radius as $\langle \dot{N}(r) \rangle \sim \dot{N}(r) \times (r/a)^{3/2}$, and determine whether the rates are pericenter- or apocenter-dominated to get $\langle \dot{N} \rangle$.
For ejection in strong scatterings,
\begin{equation}
\langle \dot{N}_{\rm ej} \rangle \propto a^{1/2} p^{-\gamma + 1} m_2^2, \label{eq:NDotEjTwiddle2}
\end{equation}
so ejection rates will be pericenter-dominated as long as $\gamma > 1$, which we will assume for the remainder of this paper, \rev{as the shallowest profiles are $\gamma \sim$ 1.3 - 1.5 (e.g., \citealt{BW77}).} For other types of close encounters, encounter rates will likewise be pericenter-dominated at small radii $r< r^{\rm tr}$ so long as $\gamma >1$, but the situation at large radii $r > r^{\rm tr}$ is more complicated. From Eqs. \ref{eq:NDotCollTwiddle}-\ref{eq:NDotTCTwiddle}, we see that close encounter rates will always be pericenter-dominated when $\gamma > 2$, but when $1< \gamma < 2$, the situation depends on whether $p > p^{\rm tr}$, a transitional pericenter\footnote{If $\gamma=2$, then rates are pericenter-dominated for $p<r^{\rm tr}$, but each decade in radius contributes equivalently when $p>r^{\rm tr}$.}. More specifically, we find that when $1< \gamma < 2$, close encounter rates scale as
\begin{align}
\langle \dot{N}_{\rm ce}\rangle \propto \label{eq:NDotCETwiddleA}
\begin{cases}
p^{-\gamma+1}a^{-3/2}R_{\rm ce}^2, & p < p^{\rm tr}~{\rm or}~a<r^{\rm tr} \\
a^{-\gamma+1/2}R_{\rm ce}, & p > p^{\rm tr}~{\rm or}~p > r^{\rm tr}.
\end{cases}
\end{align}
Here we have approximated the transitional pericenter as
\begin{equation}
p^{\rm tr} \sim a \left( \frac{R_{\rm ce}M}{a(M_\star + m_2)} \right)^{1/(\gamma-1)} .
\end{equation}
Note that it \rev{is possible that} $p^{\rm tr}>a$; in this situation, rates are always pericenter-dominated, so even though $p^{\rm tr}$ no longer has a physical meaning, the inequality conditions in Eq. \ref{eq:NDotCETwiddleA} are still accurate.
Alternatively, when $\gamma >2$,
\begin{align}
\langle \dot{N}_{\rm ce}\rangle \propto \label{eq:NDotCETwiddleB}
\begin{cases}
p^{-\gamma+1}a^{-3/2}R_{\rm ce}^2, & p <r^{\rm tr} \\
p^{-\gamma+2}a^{-3/2}R_{\rm ce}, & a > r^{\rm tr}.
\end{cases}
\end{align}
Because of the many permutations of possibilities, we have not disaggregated collision, $\mu$-TDE, and tidal capture rates in the above formulae, instead referring to a ``close encounter radius'' $R_{\rm ce}$ which will be equal to $R_\star$, $r_{\rm t}$, or $r_{\rm TC}$ for these three processes, respectively. Likewise, we have denoted a general transition radius $r^{\rm tr}$ differentiating between focused and unfocused local encounter rates; the appropriate transitional radius should be chosen from among Eqs. \ref{eq:rtrans1}-\ref{eq:rtrans3}.
With these scaling relations in hand to provide physical intuition, we now proceed to more rigorously derive ejection and close encounter rates.
\subsection{Strong Scatterings}
\label{sec:strong}
In this section, we present our analytical derivations of orbit-averaged rates for strong scattering. We employ a stellar distribution function calculated under the assumptions that (i) the stellar density profile $\rho(r) = \rho_{\rm infl} (r/r_{\rm infl})^{-\gamma_\star}$, and (ii) the potential $\psi = GM_{\bullet} /r$ is everywhere Keplerian\footnote{Throughout the remainder of this paper, we adopt the usual stellar dynamics convention of positive-definite potentials and positive-definite energies for bound orbits.}. Under the assumption of isotropy, we apply an Eddington integral to $\rho$ to obtain
\begin{equation}
f_\star(\epsilon) = 8^{-1/2}\pi^{-3/2} \frac{\Gamma(\gamma_\star+1)}{\Gamma(\gamma_\star-1/2)} \frac{\rho_{\rm infl}}{\langle m_\star \rangle} \left( \frac{GM_{\bullet}}{r_{\rm infl}} \right)^{-\gamma_\star} \epsilon^{\gamma_\star-3/2} \label{eq:DF}
\end{equation}
where the (positive-definite) specific orbital energy is, for a given star at radius $r$ and velocity $v$, $\epsilon = \psi(r) - v^2/2$; $\langle m_\star \rangle$ is the average mass in the stellar population; and $r_{\rm infl}$ is the radius that encloses a mass in stars equal to the MBH mass.
\subsubsection{Equal mass scatterer}
We first consider strong scattering from an equal mass scatterer, e.g., from other stars. We will assume that all stars have the same mass $m_\star$ and that the distribution of velocities is isotropic at every point. Let us consider a test star whose velocity is $\bold{V}$. The probability for such a star to have an encounter during a time ${\rm d}t$ that will increase its velocity to $\bold{V + \delta v}$ is \citep{Henona}:
\begin{equation}
P = \frac{8 \pi G^2 m_\star^2 {\rm d}t}{ \delta v^5} {\rm d} \delta v_x {\rm d} \delta v_y {\rm d} \delta v_z \int_{v_0}^{\infty}f_\star (v ) v \,{\rm d}v\
\end{equation}
where $f_\star (v)$ is the distribution function of stars and $v_0$ is given by:
\begin{equation}
%\label{eq:conditionvo}
v_0= \frac{1}{ \delta v} \lvert \bold{V \cdot \delta v} + \delta v^2 \rvert .
\end{equation}
%\footnote{eq \ref{eq:conditionvo} assumes the same mass for the test star and the black hole, in our next iteration we will consider different masses }
Let $\theta $ be the angle between $\bold{V}$ and $\bold{\delta v}$, with $v_{\rm esc}$ the escape velocity at this point. The star will be ejected if :
%$$\lvert \bold{V + e} \rvert \geq v_{\rm esc} $$
\begin{equation}
\label{eq:conditionV1}
V^2 + \delta v^2 + 2 V \delta v \cos \theta \geq v_{\rm esc}^2.
\end{equation}
Replacing cartesian coordinates by spherical coordinates, the local ejection rate, i.e. the probability for one star to be ejected during a time ${\rm d}t$, is given by \citep{Henon}:
\begin{equation}
\dot{N}_{\rm ej} = \frac{32 \pi^2 G^2 m_\star^2 }{3 V ( v_{\rm esc}^2-V^2)^2} \int_{ \sqrt{v_{\rm esc}^2-V^2} }^{v_{\rm esc}} (v^2+ V^2-v_{\rm esc}^2)^{3/2} f_\star (v) v \,{\rm d}v .
\end{equation}
Assuming Keplerian motion and an escape velocity $v_{\rm esc}=\sqrt{2GM_\bullet/r}$, the local ejection rate becomes:
\begin{equation}
\label{eq:ejectionrate}
\dot{N}_{\rm ej}= \frac{ 2^{2 - \gamma_\star} \pi \rho_{\rm infl} a^2 m_\star V^{1 + 2\gamma_\star} }{( 1+\gamma_\star) M^2} \left(\frac{GM_\bullet }{r_{\rm infl}} \right)^{ - \gamma_\star}
\end{equation}
with $V$ the local Keplerian velocity of the test star and $a$ its semimajor axis.
As the relaxation time is much longer than an orbital period, the local ejection rate per star can then be orbit-averaged \rev{following Eq.\ref{eq:orbitAvg}} as:
\begin{align}
\label{eq:Orbitaverage}
\langle \dot{N}_{\rm ej} \rangle= &\frac{2^{2-\gamma_\star} \rho_{\rm infl} m_\star G^{1/2} a ^{3/2- \gamma_\star} r_{\rm infl}^{\gamma_\star} }{(\gamma_\star +1) M_\bullet ^{3/2} (1-e^2)^{\gamma_\star-1}}\\
&\times \int_{0}^{\pi} \frac{(1+e^2+ 2 e \cos \nu )^{1/2+\gamma_\star} }{ (1+ e \cos \nu )^2 } \,{\rm d}\nu \notag
\end{align}
with $\nu$ the true anomaly, and $e$ the eccentricity of the star.
The orbit-averaged ejection rate does not have a general closed form.
However, for relevant values of $\gamma_\star$ which are integers or half-integers, an analytical solution exists. Analytic solutions for some physically motivated values of $\gamma_\star$ are presented in Appendix \ref{app:analytics}.
\subsubsection{Unequal mass scatterer}
Let us now consider strong scattering from an unequal mass scatterer, i.e. stellar mass black holes with $ \alpha =m_\star/m_{\rm bh}$. The probability for a star with velocity $\bold{V}$ to have an encounter during a time ${\rm d}t$ that will increase its velocity to $\bold{V + \delta v}$ becomes:
\begin{equation}
P = \frac{8 \pi G^2 m_{\rm bh}^2 dt}{ \delta v^5} {\rm d} \delta v_x {\rm d} \delta v_y {\rm d} \delta v_z \int_{v_0}^{\infty}f_{\rm bh} (v ) v {\rm d}v,
\end{equation}
where $f_{\rm bh} (v)$ is the distribution function of stellar mass black holes (also assumed to reflect a power-law density profile, with $\rho_{\rm bh} \propto r^{-\gamma_{\rm bh}}$), and $v_0$ is given by:
\begin{equation}
%\label{eq:conditionvo}
v_0= \frac{1}{ \delta v} \lvert \bold{V \cdot \delta v} + \frac{m_\star+ m_{\rm bh}}{2 m_{\rm bh}} \delta v^2 \rvert
\end{equation}
The condition for the star to escape remains unchanged. Therefore, after some algebraic manipulations, the local ejection rate becomes :
\begin{equation}
\begin{split}
\dot{N}_{\rm ej,u} = 16 \pi^2 G^2 m_{\rm bh}^2 \left( \int_{v_1}^{v_2} I_A f_{\rm bh}(v) v {\rm d}v + \int_{v_2}^{v_3} I_B f_{\rm bh}(v) v {\rm d}v +\right. \\ \left. \int_{v_3}^{v_{\rm esc}} I_C f_{\rm bh}(v) v {\rm d}v \right) \label{eq:unequalEjectionsLocal}
\end{split}
\end{equation}
where
\begin{equation}
\begin{split}
I_{A} &=\frac{2\left[v^{ 2}-\alpha\left(v_{\rm esc}^{2}-V^{2}\right)\right]^{3 / 2}}{3 V\left(v_{\rm esc}^{2}-V^{2}\right)^{2}} \\
I_{B} &=\frac{2\left[v^{ 2}-\alpha\left(v_{\rm esc}^{2}-V^{2}\right)\right]^{3 / 2}+v \left[2 v^{ 2}-3 \alpha\left(v_{\rm esc}^{2}-V^{2}\right)\right]}{6 V\left(v_{\rm esc}^{2}-V^{2}\right)^{2}}\\
&-\frac{\left(2 v_{\rm esc}+V\right)} {6 V\left(v_{\rm esc}+V\right)^{2}}+ \frac{(1+\alpha)^{2}}{8 V\left(V+v\right) } \\
I_{C} &=\frac{3 v_{\rm esc}^{2}-V^{2}}{3\left(v_{\rm esc}^{2}-V^{2}\right)^{2}}+\frac{(1+\alpha)^{2}}{4\left(V^{2}-v^{2 }\right)}
\end{split}
\end{equation}
and the \rev{integration limits} are given by:
\begin{equation}
\begin{split}
v_1 &= \sqrt{ \alpha (v_{\rm esc}^2 - V^2)} \\
v_2 &= \frac{1}{2} [ (1+ \alpha) v_{\rm esc} - (1- \alpha) V ]\\
v_3 &= \frac{1}{2} [ (1+ \alpha) v_{\rm esc} + (1 + \alpha) V ]
\end{split}
\end{equation}
The local ejection rate per star can then be orbit averaged as in Eq. \ref{eq:orbitAvg}. This orbit-averaged ejection rate does not have a general closed form, although the local ejection rate (Eq. \ref{eq:unequalEjectionsLocal}) can be written in closed form for physically motivated values of $\gamma_{\rm bh}$.
\subsection{Star-star encounters: physical
\label{sec:star}
collisions and tidal captures}
Tidal captures and physical collisions between stars, although rare in most astrophysical environments, are non-negligible in the extreme densities of galactic nuclei. Assuming that all stars have the same mass $m_\star$, and that the test star has a velocity $\bold{V}$, the local collision rate is given by% Binney Tremaine eq (7.187) :
\begin{equation}
\dot{N}_{\rm coll} = \int {\rm d}^3 \bold{v} f(\bold{v}) \pi b^2 | \bold{V}- \bold{v} |
\end{equation}
with the gravitationally focused impact parameter $b^2 = r_{\rm coll}^2 + 4 G m r_{\rm coll}/V_0^2$ and $V_0= | \bold{V}- \bold{v} |$. To take into account both collisions and tidal captures from stars we can rewrite the formula as :
\begin{equation}
\dot{N}_{\star, \star} = 2\pi^2\int \displaylimits_0^{v_{\rm esc}} \int \displaylimits_0^{\sqrt{v_{\rm esc}^2-v_{\parallel}^2}}
\sqrt{(V-v_{\parallel})^2 + v_{\bot}^2 }\; b_{\rm max}^2 \; f_{\rm bh}(v) v_{\bot} {\rm d}v_{\bot} {\rm d}v_{\parallel}
\end{equation}
with $b_{\rm max}^2 = r_{\rm max}^2 + 4 G m r_{\rm max}/V_0^2$, where $r_{\rm max} = $ max$[r_{\rm coll}, r_{\rm capt} (V_0, m_{\star})]$ and $r_{\rm capt}=2R_{\rm t,\star}\approx 2R_\star$ is the approximate tidal capture radius we employ for a high-$\sigma$ environment \citep{Stone+17}.
This formula is only an approximation to a more rigorous calculation of the tidal capture radius, (see e.g., \citealt{LeeOstriker86}), but it will generally be correct to within $\sim 10 \%$ in a high-$\sigma$ galactic nucleus \citep{Generozov+18}.
Note that tidal capture cannot occur if the relative velocity $V_0 \gtrsim \eta v_\star = \eta \sqrt{Gm_\star / r_\star}$, where $\eta < 1$ is a dimensionless number of uncertain magnitude. When $V_0 \ll v_\star$, tidal capture can proceed cleanly through the linear excitation of normal modes in the star. However, when $V_0 \sim v_\star$, the oscillation modes reach nonlinear amplitudes and may begin to dissipate their energy rapidly through nonlinear mode-mode couplings or by steepening into shocks. Rapid dissipation of the mode mechanical energy can inflate the star and result in its destruction via a sequence of partial $\mu$-TDEs \citep{Kremer+22}. The fudge factor $\eta$ encodes the boundary between successful and unsuccessful captures; for simplicity, in the remainder of this paper we will take $\eta = 0.1$. Regardless of the true value of $\eta$, it is clear that at small galactocentric radii, stars are tidally disrupted rather than captured.
The orbit-averaged star-star collision/tidal capture rate $\langle \dot{N}_{\star, \star} \rangle$ are computed following standard orbit-averaging procedures (Eq. \ref{eq:orbitAvg}).
%Then, the orbit-averaged collision rate is:
%\begin{equation}
% \langle \dot{N}_{coll,TC} \rangle= \frac{(1-e^2)^{3/2}}{\pi} \int_{0}^{\pi} \frac{ \dot{N}_{coll} (a (1-e^2/ (1+ e \cos{\nu})) }{ (1+ e \cos \nu )^2 } \,d\nu
%\end{equation}
%with $\nu$ the true anomaly as before.
\subsection{Star-sBH encounters: $\mu$-TDEs and tidal captures }
\label{sec:bh}
A star wandering too close to a stellar mass black hole can be tidally captured or tidally disrupted. The characteristic tidal radius is $ r_{\rm t} = r_\star (m_{\rm bh} /m_\star )^{1/3} $, while the maximum initial pericentre distance resulting in tidal capture, $r_{capt }$, is approximated as before by $r_{\rm capt} = 2 r_{\rm t}$, with the same caveat that large $V_0$ are treated as $\mu$-TDEs rather than tidal captures.
%can be estimated by equating the hyperbolic orbital energy with the energy deposited in tides. This condition can be written as \citep{Generozov+18}:
%\begin{equation}
%\mu \frac{v_\infty^2 }{2} = \frac{G m_\star^2}{r_\star} \left( \frac{m_{bh}}{m_\star}\right)^2 \left( \frac{r_\star}{r_{capt}}\right)^6 T_2 (r_{capt }, m_\star/m_{bh})
%\end{equation}
%where $r_\star$ is the stellar radius, $\mu$ is the reduced mass and $T_2$ is the tidal coupling constant including only the dominant $\ell=2$ (quadrupolar) modes. For the closest encounters, non-linear effects need to be considered. Ivanov+2001 estimated the magnitudes of those non-linear effects for polytropic models. For typical stellar masses of $1 \Msun$ and $0.3 \Msun$, $r_{capt}\lesssim 2 r_t$, see Generozov Appendix A and B for a complete derivation.
\\
Considering a test star with velocity $\bold{V}$, its probability to be tidally captured or disrupted by one of the stellar mass black holes is given by:
\begin{equation}
\dot{N}_{\rm TC} = 2\pi^2\int \displaylimits_0^{v_{\rm esc}} \int \displaylimits_0^{\sqrt{v_{\rm esc}^2-v_{\parallel}^2}}
V_0\; r_o^2 \; [1+ 2 \frac{G (m_\star + m_{bh})}{r_o V_0^2 }] f_{\rm bh}(v) v_{\bot} {\rm d}v_{\bot} {\rm d}v_{\parallel} ,
\end{equation}
where
\begin {equation}
\begin{split}
V_0 & = \sqrt{(V-v_{\parallel})^2 + v_{\bot}^2 } \\
r_0 & = \max[r_{capt} (V_0,m_{\rm bh}/m_\star ) , r_{\rm t}]
\end{split}
\end {equation}
For practical evaluation, the tidal capture rate is orbit averaged following Eq. \ref{eq:orbitAvg}.
%The tidal capture rate can then be orbit averaged such as:
%\begin{equation}
% \langle \dot{ N}_{TC} \rangle= \frac{(1-e^2)^{3/2}}{\pi} \int_{0}^{\pi} \frac{ \dot{N}_{TC} (a (1-e^2/ (1+ e \cos{\nu})) }{ (1+ e \cos \nu )^2 } \,d\nu
%\end{equation}
%with $\nu$ the true anomaly.
\subsection{Comparison between the different close encounters}
\begin{figure}
\centering
\begin{subfigure}[b]{0.5\textwidth}
\centering
\includegraphics[width=\textwidth]{figs/Rate_equal_r.pdf}
\end{subfigure}
\hfill
\begin{subfigure}[b]{0.5\textwidth}
\centering
\includegraphics[width=\textwidth]{figs/Rate_unequal_r.pdf}
\end{subfigure}
\caption{{\it Top}: Average rates of close encounters between stars, as a function of the dimensionless test star pericenter $1-e$. The test star has semimajor axis $a=r_{\rm infl}$ and results are shown for different stellar slopes (color-coded). Dashed lines show strong scattering ejection rates; full lines show tidal captures and stellar collisions. Different colors corresponds to different slopes of the star $\gamma_{\star}$. {\it Bottom}: same as above, but now for unequal-mass encounters between stars and black holes of mass $m_{\rm bh}=30 M_\odot$. The slope of the star is fixed $\gamma_{\star}=3/2$ and the different colors correspond to different slopes of the stellar mass black holes $\gamma_{bh}$. In both panels, the inverse angular momentum relaxation time is plotted in black, $M_{\bullet} = 10^6 M_{\sun}$, $r_{\rm infl}= 0.761$ pc, and $m_{\star} = 1 M_{\sun}$.}
\label{fig:closeinfluence}
\end{figure}
Given the many close encounters at play in a galactic nucleus, it is useful to compare their relative rates. We perform this comparison in Fig. \ref{fig:closeinfluence}, which illustrates how strong scattering rates and other close encounter rates vary as one changes stellar eccentricity $e$, and various properties of the central cusp (e.g. $\gamma_\star$, $\gamma_{\rm bh}$). The rate of angular momentum relaxation,
\begin{equation}
\dot{N}_{\rm AM} \sim \mu(\epsilon) (1-e^2)
%\approx \frac{3 G^2 \langle m_\star^2 \rangle n_\star \ln\Lambda}{\sigma^3}(1-e^2) \label{eq:AMRelax}
\end{equation}
\rev{
is also plotted as a useful reference rate (here $\mu(\epsilon)$ is the diffusion coefficient as computed in Eq. \ref{eq:diffAvg}):}
%with $\mu(\epsilon)$, the diffusion coefficient as computed in Eq. \ref{eq:diffAvg} is also plotted as a useful reference rate
loss cone shielding is likely to only become important insofar as the rate of a destructive close encounter can become comparable to $\dot{N}_{\rm AM}$.
%Note that here $t_{\rm r}$ is the energy relaxation time, $\langle m_\star^2\rangle$ is the second moment of the stellar mass function, $n_\star$ is the stellar number density, the Coulomb logarithm $\ln \Lambda \approx \ln (0.4 M_\bullet/m_\star)$, and in general Eq. \ref{eq:AMRelax} is approximate\footnote{In Fig. \ref{fig:CERates}, we actually plot a more precise $\dot{N}_{\rm AM} = \mu/(1-e^2)$, where $\mu$ is an orbit-averaged angular momentum diffusion coefficient defined more precisely in \S \ref{sec:loss}.}. Notably, it is a single-species relaxation time, although it can be generalized straightforwardly to a multi-species cusp.
We see that in general, $\dot{N}_{\rm AM} \gg \dot{N}_{\rm ej}$ so long as the relevant power-law indices $\gamma \lesssim 2$. In contrast, $\gamma = 5/2$ leads to ejection rates at large $a$ that can exceed $\dot{N}_{\rm AM}$. At small $a$, ejection rates decline and can be exceeded by close encounter rates, but regardless of $\gamma$ and $a$, $\dot{N}_{\rm AM} \gg \dot{N}_{\rm ce}$ always.
To summarize, these simple rate comparisons show that (i) destructive close encounters will never effectively shield the loss cone; (ii) ejections in strong scatterings are likely to shield the loss cone for $a\sim r_{\rm infl}$ and $\gamma > 2$; (iii) at $a \ll r_{\rm infl}$, strong scatterings become subdominant to close encounters as a way to remove stars from the distribution function. Ultimately, however, these rate comparisons only provide an order of magnitude check of the competition between different physical processes, so we now move on to computing time-dependent solutions to the loss cone problem in the presence of strong scattering and close encounters.
\section{The Shielded Loss Cone}
\label{sec:loss}
\subsection{Loss Cone Theory}
The radial distribution of stars and compact objects evolves over time due to two-body relaxation. In the continuum limit, and assuming spherical symmetry, the stellar population is represented with a distribution function $ f(\epsilon,J)$, where $J$ is the specific angular momentum of a stellar orbit. For the near-radial orbits relevant for TDEs, angular momentum relaxation is much faster than energy relaxation and we will use a two-timescale argument: we separate $f(\epsilon,J) = f_\epsilon(\epsilon) f_j(J)$ and assume a frozen distribution of energy. %we follow the one-dimensional Fokker Planck equation in angular momentum space.
By assumption, stars are fixed in bins of orbital energy, but allowed to diffuse through angular momentum space in a random walk. This process is captured by the orbit-averaged Fokker-Planck equation \citep{MerrittWang05},
\begin{equation}
%\label{FK}
\frac{\partial f }{\partial \tau} = \frac{1}{4j}\frac{\partial}{\partial j } \left( j\frac{\partial f}{\partial j}\right) \label{eq:FP1}
\end{equation}
where $j \equiv J / J_{c}(\epsilon)=\mathcal{R}^{1 / 2}$ is a dimensionless angular momentum variable (normalized by the angular momentum of a circular orbit, $J_{\rm c}$) and $\tau \equiv \mu(\epsilon) t \approx t/t_{\rm r}$ is a dimensionless version of time $t$, with $\mu(\epsilon)$ the orbit-averaged diffusion coefficient at specific energy $\epsilon$:
\begin{equation}
\mu(\epsilon)=\frac{1}{P(\epsilon)} \oint \frac{d r}{v_{\rm r}} \lim _{\mathcal{R} \rightarrow 0} \frac{\left\langle(\Delta \mathcal{R})^{2}\right\rangle}{2 \mathcal{R}}. \label{eq:diffAvg}
\end{equation}
Here $P(\epsilon)$ is the orbital period of a radial orbit of energy $\epsilon$, $v_{\rm r}$ is the star's radial velocity, and the local diffusion coefficient $\langle (\Delta \mathcal{R} )^2\rangle$ is presented in Appendix \ref{app:diff}. Hence, $\tau \approx t / t_{\rm r}$, with $t_{\rm r}$, the energy relaxation time.
In the presence of strong scatterings and destructive close encounters, stars can be ejected or otherwise removed from the distribution, which we model by adding a sink term to the Fokker-Planck equation:
\begin{equation}
\frac{\partial f }{\partial \tau} = \frac{1}{4j}\frac{\partial}{\partial j } \left( j\frac{\partial f}{\partial j}\right) - \frac{\langle \dot{N}_{\rm ej} \rangle}{\mu (\epsilon)} f. \label{eq:FP2}
\end{equation}
Here $\langle \dot{N}_{\rm ej}\rangle $ is the orbit-averaged rate of close encounters, as derived in section \ref{sec:strong} for strong scattering and \rev{\ref{sec:star} for star-star encounters and \ref{sec:bh} for star-black hole encounters. }
The intial and inner boundary conditions depend on a dimensionless diffusivity parameter $q(\epsilon)=\mu(\epsilon)P(\epsilon)/j^2_{\rm lc}(\epsilon)$, which is defined in terms of the size of the loss cone in dimensionless angular momentum space ($j_{\rm lc} \approx \sqrt{ 2GM R_{\rm t}}$). The value of $q$ determines whether the loss cone evolves in the ``empty'' ($q \ll1$; stars immediately destroyed once $j\le j_{\rm lc}$) or ``full'' ($q \gg 1$; stars may move in and out of the loss cone multiple times per orbit) limits.
\begin{comment}
\rev{Following \citep{Lezhnin}, we use a Robin boundary condition which allow to interpolate between the regimes of empty ($q \ll1$) and full loss cone ($q \gg 1$):
\begin{equation}
f(j_{\rm lc}) = \frac{ \alpha j_{\rm lc}}{2} \left\frac{\partial f(j)}{\partial j}\right|_{j=j_{\rm lc}}
\end{equation}}
\rev{where
\begin{equation}
\alpha(q) \approx (q^2+ q^4)^{1/4}
\end{equation}
}
\rev{The outer boundary condition is of the Neumann type to prohibit flux of stars through j = 1:
\begin{equation}
\left. \frac{\partial f}{\partial j}\right|_{j=1}=0,
\end{equation}}
\end{comment}
%\begin{comment}
When $q(\epsilon) \ll 1$, %as an initial condition
one can assume an absorbing boundary condition at the loss cone, and a zero-flux boundary condition at $j=1$:
%$f(j,t=0)= 1$ for $ j_{\rm LC} < j < 1 $,
%with boundary conditions given by:
\begin{equation}
f\left(j \leq j_{\rm lc}, t\right)=0 ;\left.\quad \frac{\partial f}{\partial j}\right|_{j=1}=0.
\end{equation}
%where $j_{LC} = \sqrt{2GM r_t}/J_c(\epsilon)$ and $r_t$ the tidal disruption radius.
Conversely, when $q(\epsilon) \gg 1$, the distribution function does not go to zero until a much smaller value of dimensionless angular momentum, \rev{$j_0= j_{\rm lc}(\epsilon) \exp(-\alpha/2)$}, where
\begin{equation}
\rev{\alpha(q) \approx \left(q^2 + q^4 \right)^{1/4}}
\end{equation}
is an approximate flux variable that smoothly bridges the empty and full loss cone limits, \rev{which has been derived by returning to the local (non-orbit-averaged) Fokker–Planck equation and determining how f varies with radial phase assuming $f = 0$ at periapsis \citep{CohnKulsrud78, Merritt13}.} \footnote{\rev{ Here $\alpha$ stems from the equation without a sink term. Hence we checked the impact of a different $j_o$ on the solutions that we derived and found that, as long as $j_o \ll j_{lc}$, the impact was negligible.}} We thus have a more general set of boundary conditions that we use for all numerical solutions:
\begin{equation}
f\left(j \leq j_{\rm 0}, t\right)=0 ;\left.\quad \frac{\partial f}{\partial j}\right|_{j=1}=0.
\end{equation}
In both cases we take as an initial condition an isotropic distribution, $f(j)=1$, for $j_0 \le j \le 1$, and set $f(j)=0$ elsewhere.
%\end{comment}
%\rev{ When the scatterers have a shallow profile, the distribution function is close to the Cohn-Kulsrud solution logarithmic solution as shown in Fig. \ref{fig:FK} and the inner boundary condition does not change. For steep power law solution $\gamma=5/2$, the }and numerically that the presence of a sink term on the inner boundarIn the presence of strong scattering with a shallow profile, the solution also has a logarithmic form and the
The flux of stars that scatter into the loss cone per unit time and energy is given by:
\begin{equation}
\mathcal{F}(t ; \epsilon)= 2 \pi^2 \mu(\epsilon) P(\epsilon) J_{\rm c}^2(\epsilon) f_\epsilon(\epsilon) \left ( j\frac{\partial f_j(j,t) }{\partial j}\right)_{j= j_{\mathrm{lc}}}. \label{eq:flux}
\end{equation}
Then, the TDE rate is obtained by by integrating $\mathcal{F}(\epsilon)$ across many bins of energy $\epsilon$, such as :
\begin{equation}
\dot{N}_{\rm TDE} (t) = \int \mathcal{F}(t; \epsilon) {\rm d}\epsilon. \label{eq:totalRate}
\end{equation}
%We begin the integral at $\epsilon=\epsilon_{\rm infl}$ (as this is where our approximation of a Kepler potential breaks down, and in any case is usually around the peak of the flux curve), and integrate to a large value $\sim \epsilon_{\rm infl}$ (the exact upper limit has little impact on the results as the flux declines precipitously at high energies).
\subsection{Numerical Results}
%\label{sec:results}
We begin by solving the modified Fokker-Planck equation (Eq. \ref{eq:FP2}) numerically. An example of these time-dependent solutions is presented in Fig. \ref{fig:FK}, where we show the evolution of the angular momentum distribution (in a bin of fixed $\epsilon$), both with and without strong scatterings, \rev{in the diffusive regime ($q = 10^{-2}$)}. In the absence of strong scatterings, we see that we quickly achieve the logarithmic Cohn-Kulsrud profile \citep{CohnKulsrud78}. However, when strong scatterings are turned on, a range of outcomes are possible. For ($\gamma_\star = 3/2$ and $\gamma_{\rm bh}=2$), both stars and sBHs have a very small impact at early times and a moderate one at later times. \rev{The distribution function still has a very similar form to the logarithmic Cohn- Kulsrud solution, albeit with a small depletion almost evenly distributed across angular momentum. For $\gamma_{\rm bh} = 5/2$ (or an ultra-steep density star profiles $\gamma_\star=5/2$), the distribution function} can be radically modified relative to the standard Cohn-Kulsrud solution, with a major depletion of low-$j$ orbits. The ``cavity'' carved out of the distribution function at low $j$ in this case reflects the impact of strong scattering, which preferentially ejects stars on the most radial orbits (as it is generally the case that ejection rates for an individual star are pericenter-dominated, see e.g. Eq. \ref{eq:Orbitaverage}).
\begin{figure}
\centering
\begin{subfigure}[b]{0.5\textwidth}
\centering
\includegraphics[width=\textwidth]{figs/FKequal.pdf}
\end{subfigure}
\hfill
\begin{subfigure}[b]{0.5\textwidth}
\centering
\includegraphics[width=\textwidth]{figs/FKunequal.pdf}
\end{subfigure}
\caption{Numerical evolution of the distribution function $f (j;\epsilon)$ as a function of the dimensionless angular momentum $j$ at fixed energy $\epsilon$ corresponding to $a = r _{inf}$, shown for different snapshots in dimensionless time, $\tau$, for a Milky Way-like galaxy ($M=4\times 10^6 M_\odot$). The solid lines correspond to the evolution without strong scattering or other sink terms (Eq. \ref{eq:FP1}). \textit{Top}: strong scattering (Eq. \ref{eq:FP2}) from stars only; dashed lines show $\gamma_{\star} = 3/2$, dot-dashed lines $\gamma_{\star} = 5/2$. \textit{Bottom}: strong scattering from black holes; dashed lines show $\gamma_{\rm bh} = 2$, dot-dashed lines $\gamma_{\rm bh} = 5/2$. In this panel $\gamma_{\star} =3/2$, $N_{\rm bh}=10^3$, and $m_{\rm bh}= 30 M_{\odot}$. In both panels we see dramatic changes to the distribution $f$ for ultra-steep density cusps, but only minor changes for more shallow profiles.
}
\label{fig:FK}
\end{figure}
When strong scatterings are capable of strongly depleting the low-$j$ end of the distribution function, the result is a major reduction in loss cone flux (as $\mathcal{F} \propto (\partial f_j / \partial j)|_{\rm j=j_{\rm lc}}$). We further explore this effect by computing flux reduction for different black hole slopes and black hole mass\rev{,} applying Eq. \ref{eq:flux} to our numerical solutions, and plot the results in Fig. \ref{fig:Flux}. In this figure, we see that in all cases, there is relatively little evolution of the flux suppression factor $\mathcal{F}/\mathcal{F}_{\rm w}$ (here $\mathcal{F}_{\rm w}$ is the flux into the loss cone that would be achieved without the presence of strong scatterings). Suppression factors $\mathcal{F}/\mathcal{F}_{\rm w}$ are modest $\approx 1$ for $\gamma_{\rm bh}=2$, reach the factor-of-2 level for $\gamma_{\rm bh}=9/4$, and can produce multiple-order-of-magnitude drops in the TDE rate for steeper $\gamma_{\rm bh}$ values. As shown in Eq. \ref{eq:unequalEjectionsLocal}, we see stronger levels of flux suppression for larger values of $m_{\rm bh}$.
%In all these calculations, we assume a monochromatic mass function of stars (i.e. a single mass $m_\star$ for all stars in one run), and find higher degrees of flux suppression for lower values of $m_\star$. This effect {\it does not} come from preferential ejection of low-mass stars; both Eq. \ref{eq:NDotEjTwiddle2} and the more exact formulation of Eq. \ref{eq:unequalEjectionsLocal} show that while ejection rates $\propto m_{\rm bh}^2$, at fixed values of $m_{\rm bh}$, there is no leading-order dependence on $\alpha = m_\star / m_{\rm bh}$. The dependence of $\mathcal{F}/\mathcal{F}_{\rm w}$ on $m_\star$ comes instead from the role that the stellar mass function plays in setting relaxation times ($\sim \mu^{-1}$). The longer relaxation times produced by more bottom-heavy stellar mass functions give strong scatterings a longer window of opportunity to operate, and therefore result in a greater depletion of the distribution function at fixed $\tau$. This can be seen mathematically in Eq. \ref{eq:FP2}: larger (smaller) $\mu$ values decrease (increase) the magnitude of the sink term.
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/Flux1.pdf}}
\caption{Evolution of the flux of stars inside the loss cone with strong scattering ($\mathcal{F}$) divided by the flux of stars neglecting strong scattering ($\mathcal{F}_{\rm w}$) as a function of dimensionless time $\tau$ at fixed energy $\epsilon$ corresponding to $a = r _{inf}$, for a Milky Way like galaxy with $m_{\star} = 1 $, $\gamma_{\star} =3/2$, $N_{\rm bh}=10^3$. Different colors correspond to different stellar mass black hole density slopes $\gamma_{\rm bh}$ (see legends). Full lines show $m_{\rm bh}= 10 M_{\odot}$, dashed lines show $m_{\rm bh}= 30 M_{\odot}$, and dot-dashed lines $m_{\rm bh}= 50 M_{\odot}$ stars. }
\label{fig:Flux}
\end{figure}
\subsection{Analytical solutions of Fokker-Planck with strong scattering}
\label{sec:analytical}
The flux of stars inside the loss cone $\mathcal{F} (\epsilon ) $ is a sharply peaked function, and most of the stars come from energies close to the influence radius. As we have shown, in section \ref{sec:close}, the dominant close-encounter mechanism at influence radius is strong scattering. Therefore, we searched for an analytical solution to the Fokker-Planck equation with a sink term corresponding to strong scattering (Eq. \ref{eq:FP2}).
\subsubsection{Equal mass scatterer}
For an equal mass scatterer, we found an exact analytical form for the orbit-averaged ejection rate for all relevant integer or half-integer values of $\gamma_\star$. The formulas for $\gamma_\star=3/2, 5/2$ can be found in Appendix \ref{app:analytics}.\footnote{Formulas for $\gamma_\star=2$ are in close form with Hypergeometrics series} \rev{We assume a slow variation with respect to $\tau$ such as df/d$\tau \approx$ 0, and use the method of Frobenius to derive an analytic solution in angular momentum to the modified Fokker-Planck equation for these two power-law distributions.} For the shallower slope, $\gamma_\star= 3/2$, the solution is very close to the Cohn-Kulsrud profile, whereas for the steeper slope of $\gamma_\star= 5/2$, it has an exponential behaviour.
The solutions for $\gamma_\star= 3/2$ and $\gamma_\star= 5/2$ can be written explicitly as:
\begin{equation}
\begin{split}
f_{3/2}(j) & = ( 1 + 16 A j) ( a + b \ln{j}) - 32 A b j
\\
f_{5/2}(j) &= e^{\frac{- 4 \sqrt{2 A}}{\sqrt{j}}} j^{1/4} [a g_{-}(j) + b e^{\frac{8 \sqrt{2 A}}{\sqrt{j}}} g_{+} (j) ]
\label{eq:frobenius5.2}
\end{split}
\end{equation}
where $A = \langle \dot{N}_{ej} \rangle/ \mu (\epsilon)$ is a per-star sink term, $\langle \dot{N}_{ej} \rangle$ can be found in Appendix \ref{app:analytics}, \rev{$ g_{\pm} = 1 \pm \frac{\sqrt{j}}{32 \sqrt{ 2 A} } + 0.0022 \frac{j}{ A}$ }and two undetermined constants -- $b$ and $a$ -- need to be found. These solutions are valid at small $j$, which is the relevant parameter space for stars that could undergo a disruption. Their explicit time-dependence comes from the undetermined constants $a$ and $b$. In practice, $b$ is deterministically set as a function of $a$ using the absorbing boundary condition at $j=j_{o}$, so there is only one true time-dependent free parameter. We find that $a$ depends on the dimensionless time $\tau$ as $a \sim \tau^{-1/2}$.
Since the approximate time-scale for angular momentum relaxation to occur is $t_j(j)\sim j^2 t_{\rm r} $, our solution is accurate for $ j \lesssim j_{\rm CK} = \sqrt{\tau}$. In Fig. \ref{fig:analytical52equal}, the numerical solutions of the Fokker-Planck equation with strong scatterings (Eq. \ref{eq:FP2}) are compared to our analytical solution ($\gamma_\star=5/2$) at different dimensionless times. The agreement is excellent for values of $j \lesssim j_{\rm CK}$, the ``Cohn-Kulsrud'' angular momentum below which the distribution function has had time to relax into a QSS.
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/analytical.pdf}}
\caption{Evolution of the distribution function $f(j;\epsilon)$ as a function of the dimensionless angular momentum $j$ at fixed energy $\epsilon$ shown for different snapshots in dimensionless time, $\tau$ \rev{in the diffusive regime ($q=10^{-2}$)}. Here $\gamma_\star = 5/2$ and we consider a single-mass stellar population. The dashed curves show numerical solutions of the 1D Fokker-Planck equation and solid lines show the approximate analytic solution obtained from the method of Frobenius in Eq. \ref{eq:frobenius5.2}. The agreement is excellent for values of $j \lesssim j_{\rm CK}$, the ``Cohn-Kulsrud'' angular momentum below which the distribution function has had time to relax into a QSS.}
\label{fig:analytical52equal}
\end{figure}
\subsubsection{Unequal mass scatterer}
For an unequal mass scatterer, we found close form solutions for the local rates and derived approximate solutions for the orbit-averaged rates. We then repeat our application of the Frobenius method to find time-dependent solutions for physically motivated slopes. Specifically, we find solutions for the time evolution of the distribution function of stars scattering off a population of sBHs with power-law slopes of indices $\gamma_{\rm bh}=3/2 ,7/4 , 2, 9/4 $ and $5/2$. The resulting closed form solutions are as follows:
\begin{equation}
\begin{split}
f_{3/2,u}(j)&= ( 1 + 4 A j) (a + b \ln{j}) - 8 A j b
\\
f_{7/4,u}(j)&= (1 + 8 A \sqrt{j})^2 (a - \frac{b}{2} (2 \ln{16 A} + \ln{j}))
\\
- & 2 b (\gamma_{E} + 16 A (-1 + \gamma_{E}) \sqrt{j} + 32 A^2 (-3 + 2 \gamma_{E}) j)
\\
f_{2,u}(j) & = j^{-2 \sqrt{A}} (j^{4 \sqrt{A}} a +b)
\\
f_{9/4,u} & = e^{-8 \sqrt{A} j^{-1/4}} j^{1/8} \left( \frac{ \sqrt{\pi b }} {2 A^{1/4}} (-1 + \frac{ j^{1/4} }{ 64 A^{1/2}} ) \right.
\\
+ & \left. \frac{a}{ 2 \sqrt{2 \pi}} \cosh{(8 \sqrt{A} j^{-1/4} ) } \right)
\\
f_{5/2,u}(j)&= e^{\frac{-4 \sqrt{A}}{\sqrt{j} }} j^{1/4} [a h_{-}(j) + b e^ { \frac{8 \sqrt{A}}{\sqrt{j} }} h_{-} (j)]
\end{split}
\label{eq:frobeniusunequal}
\end{equation}
where $\gamma_E$ is Euler's constant and $h_{\pm} = 1 \pm \frac{\sqrt{j}}{32 \sqrt{A} } + 0.0044 \frac{j}{ A} $.
\\
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/analytic2.pdf}}
\caption{Evolution of the distribution function $f(j;\epsilon)$ as a function of the dimensionless angular momentum $j$ at fixed energy $\epsilon$, at dimensionless time, $\tau =0.1$ for different slopes of the black holes \rev{in the diffusive regime ($q=10^{-2}$)}. Here $\gamma_\star = 3/2$, the stellar mass black holes have a mass $m = 30 M_\odot$, and different colors correspond to different sBH slopes $\gamma_{\rm bh}$ (see labels in figure). The dashed curves show numerical solutions of the 1D Fokker-Planck equation and solid lines show the approximate analytic solution obtained from the method of Frobenius in Eq. \ref{eq:frobeniusunequal}. As in Fig. \ref{fig:analytical52equal}, the agreement is again excellent for values of $j \lesssim j_{\rm CK}$.}
\label{fig:analyticalunequal}
\end{figure}
As before, explicit time-dependence enters through two undetermined coefficients, $a$ and $b$, although application of the absorbing inner boundary condition allows us to eliminate one of these. We find that the remaining coefficient can be determined quite accurately with a matching procedure: we set the value of our solution, $f_{\gamma_\star, u}(x j_{\rm cK})$ equal to the value of the Cohn-Kulsrud distribution at $j=x j_{\rm CK}$. Here $x$ is a dimensionless number of order unity; we find that $x \sim 0.1$ yields a good match.
In Fig. \ref{fig:analyticalunequal}, we compare the analytical solutions we obtained for unequal-mass strong scatterings at various $\gamma_{\rm bh}$ with the numerical solutions to Eq. \ref{eq:FP2}. In all cases (analytic and numerical) the comparison is done at a fixed dimensionless time $\tau = 0.1 $. As can be seen, the agreement between analytical and numerical solutions is excellent for $ j \lesssim \sqrt{\tau}$.
\section{Impact on TDE rates}
\label{sec:results}
%We now compute TDE rates by integrating $\mathcal{F}(\epsilon)$ across many bins of energy $\epsilon$. Specifically, the total TDE rate
%The limits of integration go from tightly bound stars with energy $\epsilon_{\rm t}=GM/(2r_{\rm t})$ (in practice these contribute negligibly to the total TDE rates) up to stars orbiting near the influence radius $r_{\rm infl}$. Throughout this paper we have made the simplifying assumption of a Kepler potential $\psi=GM/r$; this is a reasonable approximation as in most galaxies, the majority (sometimes a large majority) of the loss cone flux comes from semimajor axes $a\le r_{\rm infl}$ \citep{StoneMetzger16}. For practical purposes,
\subsection{Rates}
As we have seen in Fig. \ref{fig:Flux}, the flux inside the loss cone reaches a quasi-stationary state after a time $t \sim 0.1 \: t_{r}$, with $t_r$, the relaxation time. Therefore, in this section we explore the impact on TDE rates of strong scattering after a time where the quasi-stationary state has been reached. To compute numerical fluxes we compute the diffusion coefficients $\mu (\epsilon)$ for a grid of energies. Equation \ref{eq:FP2} is then integrated forward for the appropriate dimensionless time at each $\epsilon$-value and the flux computed from equation \ref{eq:flux}. Finally, TDE rates are obtained by integrating across the energy bins, equation \ref{eq:totalRate}. To compute analytical fluxes, we use our solutions presented in Sec \ref{sec:analytical} for the appropriate dimensionless time at each $\epsilon$-value, then compute the fluxes with equation \ref{eq:flux} and finally integrate across multiple energy bins to obtain TDE rates, equation \ref{eq:totalRate}.
%Having analyzed in detail the evolution of $f(t,j;\epsilon)$, we now will integrate across many bins of orbital energy to determine the influence of strong scattering on overall TDE rates. Here we use the empirical scaling relation for cusp galaxies found in \citet{StoneMetzger16},
%\begin{equation}
% r_{\rm infl} = 11~{\rm pc} \left(\frac{M_\bullet}{10^8 M_\odot} \right)^{0.58},
%\end{equation}
%to determine a fiducial value of $r_{\rm infl}$ at each MBH mass. Because our analytic solutions for the time-dependent distribution function (modified by ejections due to strong scattering) agree so well with numerical solutions of Eq. \ref{eq:FP2}, we perform the integral in Eq. \ref{eq:totalRate} using these solutions (Eqs. \ref{eq:frobeniusunequal}) when producing further figures in this section, although we have verified these $\dot{N}_{\rm TDE}$ results by comparison to numerical solutions of Eq. \ref{eq:FP2}, evaluated and integrated across many bins of $\epsilon$.
%One important complication for practical evaluation of Eq. \ref{eq:totalRate} is that to integrate numerically across different bins of energy, it is necessary to have estimates for the derivative $\partial f/\partial j$ evaluated at $j=j_{\rm lc}$ in each energy bin (Eq. \ref{eq:flux}). However, both our numerical and our analytic solutions to the Fokker-Planck equation parametrize $f=f(\tau, j; \epsilon)$, where $\tau=t\mu(\epsilon)$ is an {\it energy-dependent} dimensionless time\footnote{Because the relaxation time is smaller at large $\epsilon$, a snapshot of a nuclear star cluster at fixed physical time $t$ will see small values of $\tau$ at $\epsilon \sim \epsilon_{\rm infl}$ and large values of $\tau$ at $\epsilon \gg \epsilon_{\rm infl}$.}. We therefore take care to calculate Eq. \ref{eq:totalRate} for an ensemble of $\partial f/\partial j$ values evaluated at fixed time $t$ rather than fixed $\tau$.
%\subsection{Rates}
In Fig. \ref{fig:rateSMBH}, we show energy-integrated TDE rates as a function of MBH mass both with and without the effects of strong scattering for cusp galaxies with radius of influence $ r_{\rm infl} = 11~{\rm pc} \left(\frac{M_\bullet}{10^8 M_\odot} \right)^{0.58}$ \citep{StoneMetzger16}.
We can note that our TDE rates without strong scattering shown in Fig. \ref{fig:rateSMBH} are a factor of a few higher than in previous loss cone calculations (e.g., \citealt{WangMerritt04, StoneMetzger16}). This effect is due to the inclusion of stellar mass black holes. Indeed, stellar mass black holes dominate the total relaxation rate and increase diffusion coefficients hence increasing the TDE rates by a factor of a few. This effect is further explored in Fig. \ref{fig:mbhMasses} where we show the influence of different sBH masses on TDE rates both with and without strong scattering.
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/Ratesmbh.pdf}}
\caption{Tidal disruption rates $\dot{N}_{\rm TDE}$ measured in units of stars per year with (colored curves) and without (black curve) strong scattering. Results are plotted against MBH mass $M_{\bullet}$, with the following parameters: $m_{\star} = 1 M_{\sun}$, $\gamma_{\star}= 3/2$, $m_{\rm bh} = 30 M_{\sun} $ and different sBH density profile slopes: in particular, we show $\gamma_{\rm bh}=2$, $\gamma_{\rm bh}=9/4$, and $\gamma_{\rm bh}=5/2$ as green, blue, and orange curves, respectively. Reductions in total TDE rates begin to become significant for $\gamma_{\rm bh} \ge 9/4$.}
\label{fig:rateSMBH}
\end{figure}
We see here that the effects of strong scattering are quite modest for $\gamma_{\rm bh}=2$ (reducing TDE rates by $\lesssim 50\%$), rise to a factor of 2 reduction for $\gamma=9/4$, and then become very large (at least an order of magnitude reduction) for $\gamma=5/2$. It is important to note, {\it that the TDE rates we obtain with $\gamma_{bh} =5/2$ are in agreement with the observed TDE rates}. In this example we have taken $m_{\rm bh}=30 M_\odot$. The impact of different SBH masses is explored in Fig. \ref{fig:mbhMasses}.
In Fig. \ref{fig:mbhMasses}, we show that in the absence of strong scattering, TDE rates increase by a factor $\approx 3$ as $m_{\rm bh}$ increases from $10M_\odot$ to $50 M_\odot$. In the presence of strong scattering, however, the TDE rate is substantially lower at all values of sBH mass and is relatively independent of $m_{\rm bh}$. Although larger $m_{\rm bh}$ values still increase diffusion coefficients, this is offset by the greater rate at which stars are ejected in close encounters (note that both effects scale $\propto m_{\rm bh}^2$).
In these calculations, we considered a broad range of sBH masses, from $m_{\rm bh} = 10 M_\odot - 50 M_\odot$. Here we are motivated partially by the ``initial mass function'' for sBHs formed via the deaths of massive stars. Depending on the metallicity of the stellar population, the maximum masses for sBHs formed in this way is likely to be $\approx 50 M_\odot$ \citep{SperaMapelli17}, although even with a favorable metallicity, the large majority of sBHs will have substantially lower masses\footnote{We note that this type of calculation is quite sensitive to uncertain prescriptions for line-driven wind mass loss \citep{Belczynski+10}.}. \rev{However, more recent simulations have shown that $30 M_{\sun}$ can be formed at stellar metallicity \citep{Bavera}. } In addition, once a population of sBHs is situated in a galactic center environment, it may be able to gain mass through mergers and accretion. Most notably, any episode of large-scale MBH accretion will grind down and capture many of the pre-existing sBHs into the MBH accretion disk \citep{SyerRees91}; once embedded in this disk, they can be processed to larger masses either via mergers \citep{Bellovary+16, Tagawa+20} or from direct gas accretion \citep{GilbaumStone22}. After the AGN episode ends, the population may relax into a more spherical configuration of the type we assume here \citep[although see][]{SzolgyenKocsis18}. Efficient sBH growth through merger may also occur during periods of time when a dense nuclear star cluster lacks a MBH \citep{AntoniniRasio16}, either early during its evolution or at later points if an MBH is temporarily ejected via gravitational wave recoil \citep{GualandrisMerritt08}.
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/mbh_r.pdf}}
\caption{Tidal disruption rates $\dot{N}_{\rm TDE}$ measured in units of stars per year with (colored curves) and without (black curve) strong scattering as a function of stellar black hole mass $m_{bh}$. Higher stellar mass black holes increase the diffusion coefficients and hence the TDE rates without strong scattering. However, with strong scattering there is a balance between the effect on the diffusion and the higher efficiency at rmoving of stars of heavier sBH. }
\label{fig:mbhMasses}
\end{figure}
%As can be seen in fig \ref{fig:rateSMBH}, strong scattering induce a significant reduction of the TDE rate at all SMBH masses. If the black hole slope is $\gamma=5/2 $, rates at all SMBH masses are reduced by $\sim 10 \percent $, allowing to reconcile theoretical rates with observed TDe rates. $\dot{N}_{Tde} \sim 10 \percent \dot{N}_{Tde} $
The rate suppression due to strong scatterings can depend strongly on the semimajor axis, or equivalently the energy $\epsilon$, of the stars in question. At low values of $\epsilon$, TDE rates can be suppressed by one or more orders of magnitude in the presence of high-mass and steeply cusped sBHs, but the suppression is often milder (especially at high $\gamma_{\rm bh}$) for the subset of stars on tightly bound, high-$\epsilon$ orbits. We have seen this numerically, and established it analytically in Eqs. \ref{eq:ejectionrate} and \ref{eq:Orbitaverage}. A consequence of this is that galactic nuclei can see a relative suppression of high-$\beta$ TDEs, which are overwhelmingly produced in the full loss cone (or ``pinhole'') regime where rates are suppressed most heavily by strong scatterings. Although the observational implications of high- vs low-$\beta$ TDEs are not yet clear, high-$\beta$ events may exhibit faster circularization \citep{Hayasaki+13, Bonnerot+16}, produce stronger thermal soft X-ray emission \citep{Dai+15}, produce stronger sub-relativistic outflows from the circularization process \citep{MetzgerStone16, LuBonnerot20}. These observational signatures could be less frequently produced in galaxies with large and steeply cusped populations of sBHs.
We quantify this effect in Fig. \ref{fig:pinhole}, which shows $f_{\rm pinhole}$, the fraction of TDEs occuring in the pinhole regime, as a function of MBH mass $M_\bullet$ (with and without strong scattering). Without strong scattering, the value of $f_{\rm pinhole}$ decreases with increasing $M_{\bullet}$ and drops steeply for $ M_{\bullet} \sim 10^8 M_{\sun}$, as expected for cusp galaxies. Interestingly, the influence of strong scattering modifies the pinhole fraction in two opposite directions, with the net effect depending on $\gamma_{\rm bh}$. The first effect is already explained in the preceding paragraph: at fixed $\tau$, loss cone flux is more greatly reduced (by strong scatterings) when $\epsilon$ is low than when $\epsilon$ is high. However, a countervailing force is visible in Fig. \ref{fig:FK}, where we see the mild but relevant time evolution of the suppression due to strong scatterings. The suppression factor $\mathcal{F}(\epsilon)/\mathcal{F}_{\rm w}(\epsilon)$ becomes smaller and smaller as $\tau$ grows, so when computing an energy-integrated $\dot{N}_{\rm TDE}$ rate, the bins of large $\epsilon$ (with larger $\tau$) will be further along the time evolution of their suppression curves. In Fig. \ref{fig:pinhole}, we see that for large $\gamma_{\rm bh}$, the first effect wins out and the pinhole fraction decreases due to ejections. But for $\gamma_{\rm bh}\lesssim 9/4$, the second effect wins out and the pinhole fraction sees little net change from strong scatterings (the net change that does exist is a very mild increase in $f_{\rm pinhole}$).
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/pinhole_r.pdf}}
\caption{The fraction, $f_{\rm pinhole}$, of all TDEs in a galaxy which fall into the pinhole regime of tidal disruption, plotted against MBH $M_{\bullet}$. The black curve shows this dependence without strong scattering, while the colored curves show the pinhole fraction in the presence of strong scattering (see labels in-panel). TDEs in the pinhole (or full loss cone) regime can access any $\beta$, while TDEs in the opposite, diffusive, regime, almost always have $\beta \approx 1$.}
\label{fig:pinhole}
\end{figure}
\subsection{E+A preference}
Observations have revealed that E+A and other post-starburst galaxies are significantly over-represented among TDE hosts \citep{Arcavi+14, French+16,French+17, Law, Graur, French20, Hammerstein21}. %The latest estimations find that E+A galaxies are overrepresented by a factor of $ \approx 22 $ \citep{Hammerstein21}.
A few solutions have been proposed to explain this preference: stellar overdensities \citep{Stone+18}, radial anisotropies \citep{Stone+18}, SMBH binaries (e.g., %\cite{MerrittWang05})
\citealt{Arcavi+14}) and accounting for a complete stellar mass function \citep{Bortolas}.
\\
In the framework of our revised loss cone theory, we have shown that steep slopes induce a strong reduction of TDE rates. Motivated by our findings, here we study the impact of stellar overdensities with strong scattering.
\\
E+A galaxies are post-starburst galaxies and hence, in this section, we consider only stars and no sBHs. If we assume that most of the central stars formed impulsively in a starburst, these ultra-steep stellar central density cusps will relax over time towards a steady state configuration: the Bahcall–Wolf cusp, with $\gamma_{\star} = 7/4$ \citep{BW}.
An ultra-steep stellar density profile will result in much shorter relaxation times, especially at small energies, and hence the relaxation to a BW cusp is quicker for steeper star profiles. Following \citet{Stone+18}, we define the Bahcall-Wolf radius $r_{\rm BW}$ as the radius where the post-starburst age $t$ equals the local energy relaxation time:
\begin{equation}
r_{\rm BW} =\left( \frac{M_{\bullet}^{3/2} \langle m_{\star} \rangle }{G^{1/2} \langle m^2_{\star}\rangle \rho_{inf} r_{inf}^{\gamma} \: t \: \ln \Lambda} \right) ^{\frac{1}{3/2- \gamma}}
\end{equation}
In Fig. \ref{fig:rbw}, we plot the Bahcall-Wolf radius $r_{\rm BW}$ divided by the influence radius for different MBH masses. %For times where $r_{\rm BW}/ r_{inf} \geq 0.01$, a broken power-law would be more accurate. Hence, for our analysis we do only consider times for which the assumption of a single power-law holds.
As $r_{\rm BW}$ expands, energy relaxation erodes the population of the most tightly bound stars, decreasing their density relative to the initial conditions. The density decrease at high $\epsilon$ reduces the efficiency of loss cone shielding, as it is the most tightly bound stars which are responsible for the strongest shielding effects. From Fig. \ref{fig:rbw}, we see that this cusp erosion is fastest for small MBHs and slowest for large MBHs. As we do not self-consistently model this cusp erosion in our calculations, we only compute TDE rate reductions at times where this effect would be minor.
%Indeed, when cusp erosion starts to be significant, ultra-steep stellar profiles will not be able to ``self-shield'' unless new stars are being continuously deposited at small radii (e.g. by the Hills mechanism).
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/rbw.pdf}}
\caption{Time evolution of the Bahcall-Wolf radius divided by the radius of influence for different MBH masses. Green, blue, orange, and purple lines correspond to $M_{\bullet}= 10^5 M_{\sun}$, $M_{\bullet}= 10^6 M_{\sun}$, $M_{\bullet}= 10^7 M_{\sun}$, and $M_{\bullet}= 10^8 M_{\sun}$, respectively. The dashed lines correspond to $\gamma_{\star}= 5/2 $ while full lines are for $\gamma_{\star}= 9/4 $. }
\label{fig:rbw}
\end{figure}
%In Fig. \ref{fig:ea}, we show TDE rates with strong scattering normalized by TDE rates without strong scattering, as a function of MBH masses. %As shown, in fig \ref{fig:rbw}, a MBH with $M_{\bullet}= 10^5 M_{\sun}$ requires a broken-power law and hence is not considered here.
%We neglect the smallest MBH masses where flattening of the inner profile may be most relevant. %At larger times, where the assumption holds only for larger MBH, $M_{\bullet}= 10^6 M_{\sun}$ is also not considered (dashed lines).
%We find that strong scatterings can reduce the TDE rate by up to an order of magnitude depending on the stellar slope and MBH mass, showing that ultrasteep stellar profiles may efficiently self-shield, reducing the otherwise high TDE rates they would produce.
%\rev{In Fig. \ref{fig:ea}, we present TDE rates that account for strong scattering, normalized by TDE rates that do not account for strong scattering. These rates are shown as a function of MBH masses for times and MBH masses where the erosion of the cusp can be neglected, as shown in Fig. \ref{fig:rbw}. We find that for the steep stellar profile: $\gamma \geq 9/4$ proposed as an explanation for the E+A preference, accounting for strong scattering can significantly reduce TDE rates by up to an order of magnitude, depending on the stellar slope and MBH mass. It can also be noted, that the effect is stronger for longer times since the starburst, as long as the time is less than the time required for cusp erosion. Our results suggest that ultra-steep stellar profiles can effectively self-shield, which makes them a less plausible explanation for the apparent preference of TDEs for E+A galaxies.}
\rev{In Fig. \ref{fig:ea}, we present TDE rates that consider the influence of strong scattering, normalized by TDE rates that do not account for strong scattering. These rates are depicted at two different times where the cusp erosion can be disregarded: $\frac{r_{BW}}{r_{inf}} \leq 10^{-3}$,} \footnote{\rev{This criterion derives from the fact that energies such as
$r \leq 10^{-3} r_{inf}$ contribute about 1\% of the TDE rate.}}
\rev{ (see Fig. \ref{fig:rbw}). We find that, for the steep stellar profile: $\gamma \geq 9/4$, accounting for strong scattering can reduce TDE rates by up to an order of magnitude. Additionally, it is worth noting that the effect is more pronounced for the dashed lines corresponding to times $t=100$ Myr and $t=150$ Myr (the cusp erosion is still negligible for the higher MBH masses). Our results suggest that ultra-steep stellar profiles can effectively self-shield, which makes them a less plausible explanation for the preference of TDEs for E+A galaxies.}
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/ea.pdf}}
\caption{Rates of TDEs with strong scattering divided by the rates of TDEs without strong scattering, shown as a function of the MBH mass. Green lines are for $\gamma_{\star}= 9/4 $, purple lines for $\gamma_{\star}= 5/2 $. Full lines are at $t= 10$ Myr (15 Myr for $\gamma_{\star}=9/4 )$, dashed lines are at $t= $100 Myr (150 Myr for $\gamma_{\star}=9/4 )$.}
\label{fig:ea}
\end{figure}
\begin{comment}
\subsection{Influence of the mass of the star}
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/Femass.pdf}}
\caption{Flux of stars inside the loss cone for different star masses with and without strong scattering, $\gamma_{\star}= 3/2 $, $m_{bh} = 30 M_{\sun} $, $\gamma_{bh}= 5/2 $ ,$ M_{\bullet} = 10^5 M_{\sun}$ }
\label{fig:Femass}
\end{figure}
As can be seen in fig \ref{fig:Femass}, classical loss cone theory predicts that the lighter the star, the more important the flux will be for every bin of energy. However, the presence of strong scattering completely modifies the picture making the heavier stars having more flux at radius of influence. Recently, it has been observed that stars with mass $m \sim 1-2 M_{\sun}$ were over represented among observed TDEs \citep{Mockler+22}. This observation doesn't reconcile with classical loss cone theory but could be a clear signature of the presence of strong scattering.
\\
\begin{figure}
\centerline{\includegraphics[width=90mm]{figs/peakFe.pdf}}
\caption{Ratio of the flux of stars inside the loss cone divided by the maximum of flux for this specific configuration , for different star masses with and without strong scattering, $\gamma_{\star}= 3/2 $, $m_{bh} = 30 M_{\sun} $, $\gamma_{bh}= 5/2 $ ,$ M_{\bullet} = 10^5 M_{\sun}$ }
\label{fig:peakFE}
\end{figure}
We can also note that the flux in the presence of strong scattering does not follow anymore a power-law as a function of energy. This is a key point, as in classical loss cone theory most of the flux is coming from the radius of influence. As can be seen, in fig\ref{fig:peakFE}, this effect is even stronger for $m_{\star} =0.3 M_{\sun}$.
\\
\subsection{Stellar Mass Function}
The TDE rate also depends on the present day mass function $d N_{\star} / d m_{\star}(\mathrm{PDMF})$ of stars. Therefore, we also compared the TDE rate obtained with monochromatic distributions to a more realistic Kroupa IMF :
$$
\chi_{\mathrm{Kro}}=\left.\frac{d N_{\star}}{d m_{\star}}\right|_{\text {Kro }}=\left\{\begin{array}{ll}
0.98\left(m_{\star} / M_{\odot}\right)^{-1.3} & , m_{\star}^{\min }<m_{\star}<0.5 M_{\odot} \\
2.4\left(m_{\star} / M_{\odot}\right)^{-2.3} & , 0.5 M_{\odot}<m_{\star}<m_{\star}^{\max } \\
0 & , \text { otherwise. }
\end{array}\right.
$$
where $m_{\star}^{\min }$ and $m_{\star}^{\max }$ have the same values as the Salpeter PDMF.
\\
As compared to a monochromatic $m_{\star}= 1 M_{\sun} $, TDE rates computed with a more realistic IMF are greater by a factor $\sim $ 1.6, making the tension between observed TDE rates and classical loss cone TDE rates even more important.
\end{comment}
%\section{Observational Implications}
%\label{sec:obs}
%{\bf NCS: place to discuss TDE rate suppression, effect on beta (TDE impact parameter) distribution, effect on beta (anisotropy parameter) in galactic nuclei, effect on TDE stellar mass distribution...what else?}
%Because the stellar ejection rate $\langle \dot{N}_{\rm ej} \rangle \propto \alpha^{-2}$, loss cone shielding depends sensitively on both the mass of the sBHs {\it and the mass of the stars being ejected}. In other words, for a fixed sBH mass distribution, the loss cone will be more effectively shielded for low-mass stars than for high-mass ones. This creates a new bias towards disruption of high-mass stars which, for sufficiently large $\gamma_{\rm bh}$ values, can greatly exceed the relatively weak biases present in standard loss cone theory. Absent strong scatterings, TDE rates scale $\dot{N}_{\rm TDE} \propto \ln(R_{\rm t})$ in the diffusive regime and $\dot{N}_{\rm TDE} \propto R_{\rm t}$. The combined TDE rate spanning both regimes will have a scaling that depends on the overall potential and density profile, but is often shallow, e.g. $\dot{N}_{\rm TDE} \propto R_{\rm t}^{1/4}$ in \citet{Macleod+12}. Furthermore, the tidal radius itself is a weak function of stellar mass, going $\propto M_\star^{0.47}$ on the lower main sequence or $\propto M_\star^{0.17}$ on the upper main sequence \citep{KippenhahnWeigert94}.
%The stronger mass-dependence in the rate suppression that we predict for steep sBH cusps may manifest itself observationally in TDE spectra. In particular, stars that burn via CNO-cycle reactions (i.e. those with $M_\star \gtrsim 1.3 M_\odot$) will, at any point in time, have the vast majority of their core CNO elements locked up in $^{14}$N. As a result, the late-time spectra from these events should see a depletion in C and O line emission, and a strengthening of N line emission \citep{Kochanek16}. Although obtaining abundance ratios from line measurements is challenging, estimates from the limited dataset available today do indicate a strong over-representation of CNO-burning stars among observed TDEs \citep{Mockler+22}.
\section{Discussion \& Conclusions}
\label{sec:conclusions}
In this paper, we have proposed a revised loss cone theory that takes into account both classical weak scattering and close encounters: strong scattering, collisions, tidal captures and $\mu$-TDEs. These close encounters can occur at non-negligible rates inside the radius of influence, %{\it the densest environments of the universe}.
where stellar densities can be higher than anywhere else in the universe. Below we summarize our key findings:
\\
\begin{enumerate}
\item For stars with semi-major axis $a \sim r_{\rm inf} $, i.e. the primary sources of TDEs, strong scattering rates are significantly larger than other close encounter rates. For very high orbital energies, collisions become the dominant close encounter. However, those energies have a very small contribution to TDE rates.
\item We derive time-dependent analytical solutions of the Fokker-Planck equation with strong scattering for %all
a broad range of physically motivated slopes. This family of analytic solutions, which we can write in closed form for all relevant integer, half-integer, and quarter-integer $\gamma$ values, is a generalization of the standard \cite{CohnKulsrud78} $f(j)$ distribution function in the presence of strong scatterings.
\item We find that for black hole slopes $\gamma_{\rm bh} \geq 9/4 $, the loss cone will be effectively shielded by strong scattering. TDE rates are reduced by a factor $\sim 2 $ for $\gamma_{\rm bh} = 9/4 $ and by a factor $\geq 10 $ for $ \gamma_{\rm bh} = 5/2$.
\item Strong scattering is more effective at removing stars with $a \sim r_{\rm inf}$ compared to stars with $a \ll r_{\rm inf}$. Thus, in addition to shielding the loss cone, strong scattering can also reduce the fraction of TDEs in the pinhole regime, and therefore the fraction of deeply plunging TDEs for $\gamma_{\rm bh} = 5/2$.
\item %Stellar overdensities, %the most
%\rev{one possible} explanation for the E+A preference among TDEs, is %less favoured when taking into account
%challenged by our revised loss cone theory. Indeed, in classical loss cone theory, ultra steep stellar slopes enhance the diffusion coefficients and, hence the TDE rates. However, we have shown that they also enhance the rates of strong scattering and hence reduce TDE rates. The net TDE enhancement %is therefore mild.
%\rev{induced by stellar overdensities is hence milder than previously assumed, making other possible explanations such as radial anisotropies (e.g \citep{Stone+18}) or a more complete mass function (e.g \citep{Bortolas}), more likely to explain the apparent preference of TDEs for E+A galaxies. }
\rev{Stellar overdensities have been proposed as one possible explanation for the observed preference of tidal disruption events (TDEs) in E+A galaxies. However, our revised loss cone theory challenges this idea. In classical loss cone theory, ultra-steep stellar slopes increase the diffusion coefficients and, thus, the TDE rates. However, we have shown that they also increase the rates of strong scattering, which can reduce TDE rates. As a result, the net enhancement of TDEs induced by stellar overdensities is milder than previously assumed. This finding makes other possible explanations, such as radial anisotropies (e.g., \citealt{Stone+18}) or a more complete mass function (e.g., \citealt{Bortolas}), more likely to explain the observed preference of TDEs for E+A galaxies.}
\end{enumerate}
%The key results of our paper, loss cone shielding in the strong scattering can shield is a conservative one. Indeed, we did consider only stars that underwent a strong enough scattering to remove them from the distribution.
In this paper, we have focused on the most dramatic qualitative difference between weak and strong scatterings: ejection. However, stars also have mildler but still strong encounters that can non-diffusively move them to new -- but still bound -- energies. %remove them from their initial energy but not from the distribution.
This effect will be the subject of a future work. The shielding effect we have studied depends on black hole slopes which have not been constrained observationally. $N$-body simulations have shown that sBHs %will
can settle to a slope of $\gamma_{\rm bh} \geq 2$ \citep{Amaro,Preto}, in agreement with some predictions of strong mass segregation \citep{AlexanderHopman09}. \rev{It can also be noted that, for an adiabatically growing MBH, the stellar slope may assume steeper values of $\gamma \gtrsim 2$ \citep{Young}.} Moreover, \cite{Generozov+18} showed that, in the presence of a source term that is continuously depositing new stars or sBHs at small radii, black holes will settle into a density profile with $ \gamma_{\rm bh} = 5/2$. Motivated by these studies, we considered different black hole slopes with $\gamma_{\rm bh} \leq 2.5 $.
\\
If steep profiles for black holes are common in galactic nuclei, this may explain persistent discrepancies between the high predicted rates of TDEs \citep{WangMerritt04} and the (relatively) low observed rates of TDEs in low-mass galactic nuclei. While the steep TDE luminosity function \citep{vanVelzen18} helps to resolve this rate discrepancy, it may be unable to resolve the entire breadth of the discrepancy if the occupation fraction of small MBHs ($\sim 10^5 M_\odot$) is high in dwarf galactic nuclei. Conversely, loss cone shielding may pose a challenge for the ``overdensity'' solution \citep{StoneMetzger16, StonevanVelzen16, Stone+18} to the observed post-starburst preference \citep{Arcavi+14, French+16, French+17,Hammerstein21} among the host galaxies of TDEs. While an ultra-steep stellar distribution can boost TDE rates up to a point, our work here shows that the gains from increasing $\gamma_\star$ beyond $9/4$ may be self-limiting, and that TDE rates may even drop as $\gamma_\star$ increases to very large values.
\\
It is worth mentioning that we considered a monochromatic distribution of stars throughout the paper. However, we did also compute the impact of a Kroupa IMF and found that it increased all TDE rates with and without strong scattering by a common factor of 1.6. %Therefore, the rates reduction hold with a Kroupa IMF.
%We also want to stress, that our condition to remove stars from the distribution through strong scattering is different than the criteria for hypervelocity stars.
It is also possible that shielding effects could be important for the production of hypervelocity stars in binary breakups \citep{Hills88}, as these are generally sourced from binaries with galactocentric semimajor axes $a \gg r_{\rm infl}$ \citep{Perets+07}.
\\
To conclude, we emphasize that we considered the 1D Fokker-Planck equation in angular momentum, neglecting energy evolution of the stellar (and sBH) distribution function. This may be important for steep density profiles that are not quasi-steady states. This, together with the effect of strong scattering that can non-diffusively move stars to higher energies will be the subject of a future work.
\begin{comment}
\section{Discussions \& Conclusions}
\label{sec:conclusions}
We have revisited the loss cone problem in spherical galactic nuclei, moving beyond the diffusive approximation normally employed to include the leading-order impact of strong scatterings: ejection. We have solved the 1D Fokker-Planck equation with the addition of a sink term that accounts for ejection of stars in strong scatterings (and also destruction of stars in other types of close encounters, such as direct collisions, $\mu$-TDEs, and tidal captures). Our primary motivation in this study was to investigate whether the MBH loss cone can be ``shielded'' from the standard, diffusive flux of stars that will become TDEs, and we have found that under some astrophysically plausible circumstances, the answer to this question is affirmative. Our primary conclusions are as follows:
\begin{enumerate}
\item Rates of ejection in strong scattering generally dominate rates of destructive close encounters when considering stars with semimajor axis $a \sim r_{\rm infl}$. Rates of destructive close encounters can dominate strong scatterings for $a \ll r_{\rm infl}$, but these tightly bound energies (a) are relatively unimportant for determining the total TDE rate and (b) have loss cones that are too large in $j$-space to be effectively shielded.
\item Ejection in strong scatterings will shield the loss cone effectively only for a population of scatterers with an ``ultrasteep'' density profile. If we parametrize the scatterers as having a power-law 3D density profile, $\rho(r) \propto r^{-\gamma}$, we require $gamma \gtrsim 9/4$ for loss cone shielding to be effective.
\item Loss cone shielding preferentially removes stars from the distribution function at values of $j \ll 1$, but that are still greater than the loss cone boundary $j_{\rm lc}$. This chokes off the diffusive flux of stars into the loss cone that would otherwise generate TDEs. We see this effect both by direct numerical solution of the modified 1D Fokker-Planck equation, but also in analytic, time-dependent solutions we derived with the method of Frobenius. This family of analytic solutions (which we can write in closed form for all relevant integer, half-integer, and quarter-integer $\gamma$ values) is a generalization of the standard \citet{CohnKulsrud78} $f(j)$ distribution function in the presence of strong scatterings.
\item Loss cone shielding can reduce the energy-integrated TDE rates in a galactic nucleus by a factor $\sim 2$ for $\gamma=9/4$ and by a factor $\sim 10$ for $\gamma=5/2$. The effect becomes stronger for scatterers of greater mass $m_{\rm bh}$, with ejection rates scaling $\propto m_{\rm bh}^2$.
%For fixed $m_{\rm bh}$ the per-star ejection rate is independent of test star mass, but the efficiency of loss cone shielding does depend indirectly on the stellar mass function via its impact on the relaxation time.
\item Because loss cone shielding is at its most effective for $a\sim r_{\rm infl}$ and always becomes ineffective for $a \ll r_{\rm infl}$, it reduces the fraction of TDEs in the pinhole regime, and therefore the fraction of deeply plunging TDEs.
\end{enumerate}
The broader astrophysical implications of our work depend on how common ultrasteep density profiles are in galactic nuclei (either for the bulk stellar population or for a subset of high-mass objects such as sBHs, as both scenarios are capable of producing a strong shielding effect). For the average stellar population, ultrasteep density profiles are strongly out of equilibrium in energy space \citep{BahcallWolf76}, although \citet{AlexanderHopman09} has shown that they may arise as a natural QSS for a heavy subspecies of sBHs under certain conditions. This analytic prediction has been confirmed by a limited number of $N$-body simulations \citep{PretoAmaroSeoane10}, although the assumptions of \citet{AlexanderHopman09} have also been challenged recently \citep{LinialSari22}. Ultrasteep density profiles may also arise as QSSs in the presence of a source term that is continuosly depositing new stars or sBHs at small radii; in the limit of continuous mass deposition and collisional evolution, this can produce a density profile with $\gamma=5/2$ \citep{Generozov+18}. Both {\it in situ} star formation (via multiple episodes of active galactic nucleus accretion disk fragmentation; \citealt{LevinBeloborodov03, GilbaumStone22}) or deposition of stars via the Hills mechanism \citep{Hills88, GouldQuillen03, Perets+07} could provide this source term. It is also possible that ultrasteep density profiles may exist as transient features following one-off episodes of nuclear star formation, as is suggested by observations of a $\gamma \approx 2.1-2.3$ cusp near the influence radii of nearby post-starburst galaxies \citep{StonevanVelzen16, Li+22}.
If ultrasteep cusps of massive objects are common in galactic nuclei, this may explain persistent discrepancies between the high predicted rates of TDEs \citep{StoneMetzger16} and the (relatively) low observed rates of TDEs in low-mass galactic nuclei. While the steep TDE luminosity function \citep{vanVelzen18} may help to resolve this rate discrepancy, it may also be unable to resolve the entire breadth of the discrepancy if the occupation fraction of small MBHs ($\sim 10^5 M_\odot$) is high in dwarf galactic nuclei. Conversely, loss cone shielding may pose a challenge for the ``overdensity'' solution \citep{StoneMetzger16, StonevanVelzen16, Stone+18} to the observed post-starburst preference \citep{Arcavi+14, French+16, French+17} among the host galaxies of TDEs. While an ultrasteep stellar distribution can boost TDE rates up to a point, our work here shows that the gains from increasing $\gamma_\star$ beyond $9/4$ may be self-limiting, and that TDE rates may even drop as $\gamma_\star$ increases to very large values.
Our work has various limitations which we hope to address in future studies. Most notably, we have worked in the 1D approximation, which neglects energy evolution of the stellar (and sBH) distribution function. This may be particularly important for ultrasteep density profiles that are not QSSs, as energy relaxation times can be quite short at small radii. We have considered at most 2-component cusps, but some aspects of strong scattering may change in the presence of a continuous mass spectrum. We have also limited ourselves to a purely Kepler potential for mathematical simplicity; as most TDEs originate from $a\lesssim r_{\rm infl}$, this is an acceptable approximation, but certainly can be improved upon (though likely at the cost of the analytic solutions we obtained). Obviously, we have only considered one manifestation of strong scattering -- terminal ejection of stars -- but in reality strong scattering will also non-diffusively shuffle stars through $\{\epsilon, j\}$-space as well. Our treatment of ejection is also somewhat approximate, as (given that the MBH is embedded in a deeper galactic potential) there is no exact criterion separating a permanently ejected star from one which may eventually return to the nucleus, provided it does not receive a strong torque near the apocenter of its galactic orbit.
Finally, we have only considered how strong scatterings may shield the loss cone of an MBH from TDEs, but other interesting astrophysical phenomena are governed by loss cone dynamics as well. We suspect that strong scatterings are unlikely to affect rates of extreme-mass ratio inspirals \citep{Ryan95, SigurdssonRees97}, as these are generally sourced from sBHs on very tightly bound orbits \citep{HopmanAlexander05}, but it is possible that shielding effects could be important for the production of hypervelocity stars in binary breakups \citep{Hills88}, as these are generally sourced from binaries with galactocentric semimajor axes $a \gg r_{\rm infl}$ \citep{Perets+07}.
\end{comment}
\section*{Acknowledgements}
The authors would like to thank Cole Miller, Eliot Quataert and Luca Broggi for helpful discussions. We also warmly thank the anonymous referee for their very useful comments and suggestions. OT gratefully acknowledges the support of the Einstein-Kaye scholarship and of the Israel Ministry of Science and Technology. This work was partially funded by the Israel Science Foundation (Individual Research Grant 2565/19) and the United States-Israel Binational Science Foundation (Grant No. 2019772).
\section*{DATA AVAILABILITY STATEMENT}
The data underlying this article will be shared on reasonable request to the corresponding author.
\bibliographystyle{mnras}
\bibliography{lc_refs}
\appendix
\onecolumn
\section{Orbit-averaged ejection rate}
\label{app:analytics}
Throughout this paper, we performed orbit averages of various interaction rates $X$ in the following way:
\begin{equation}
\langle X \rangle = \frac{2}{P} \int_0^\pi \dot{\nu}^{-1} X {\rm d}\nu.
\end{equation}
In the special case of a Kepler potential, this simplifies to
\begin{equation}
\langle X \rangle = \frac{2(1-e^2)^{3/2}}{\pi} \int_0^\pi \frac{X{\rm d}\nu}{(1+e\cos \nu)^2} . \label{eq:orbitAvg}
\end{equation}
We found that the orbit averaged ejection rate has an analytical form for $\gamma_\star$ values which are integers or half-integers. Hereafter are the analytic forms we found for interesting values of $\gamma$.
For an equal mass scatterer and $ \gamma_\star = 3/2$, we obtained:
\begin{equation}
\langle \dot{N}_{\rm ej} \rangle= \frac{2^{1/2} \pi \rho_{\rm infl} m G^{1/2} r_{\rm infl}^{3/2} (4 - 3 (1 - e^2)^{1/2})}{ (\gamma_\star+ 1)M ^{3/2} (1-e^2)^{1/2}} .
\end{equation}
\\
For an equal mass scatterer, with $\gamma_\star= 5/2$, we obtained:
\begin{equation}
\langle \dot{N}_{\rm ej} \rangle= \frac{2^{-1/2}\pi \rho_0 m G^{1/2} r_0^{5/2} (-4+ 12 e^2 + 5 (1-e^2)^{3/2})}{(\gamma_\star + 1) M ^{3/2} a (1-e^2)^{3/2}} .
\end{equation}
\section{Local diffusion coefficient }
\label{app:diff}
The local diffusion coefficient we evaluate in Eq. \ref{eq:diffAvg} is given by \citep{MagorrianTremaine99, WangMerritt04}:
\begin{equation}
\lim _{R \rightarrow 0} \frac{\left\langle(\Delta R)^{2}\right\rangle}{2 R}=\frac{32 \pi^{2} r^{2} G^{2}\left\langle m_{\star}^{2}\right\rangle \ln \Lambda}{3 J_{c}^{2}(\epsilon)}\left(3 I_{1 / 2}(\epsilon)-I_{3 / 2}(\epsilon)+2 I_{0}(\epsilon)\right)
\end{equation}
with:
\begin{equation}
I_{0}(\epsilon) \equiv \int_{0}^{\epsilon} f\left(\epsilon^{\prime}\right) \mathrm{d} \epsilon
\end{equation}
\\
and
\begin{equation}
I_{n / 2}(\epsilon) \equiv {[2(\psi(r)-\epsilon)]^{-n / 2} }
\int_{\epsilon}^{\psi(r)}\left[2\left(\psi(r)-\epsilon^{\prime}\right)\right]^{n / 2} f\left(\epsilon^{\prime}\right) \mathrm{d} \epsilon^{\prime}.
\end{equation}
%\section{Analytical solution to Fokker Planck equation %with strong scatterings}
%Using Frobenius method we derived an analytical solution for two others physically motivated slopes :
%For $\gamma=7/4$ the solution writes :
%\begin{equation}
% f_{7/4}(j)= (1 + 8 A \sqrt{j})^2 a -
% 2 (\gamma_{euler} + 16 A (-1 + \gamma_{euler}) \sqrt{j} +
% 32 A^2 (-3 + 2 \gamma_{euler}) j) b -
% 1/2 (1 + 8 A \sqrt{j})^2 b (2 \ln{16 A} + \ln{j})
%\end{equation}
\begin{comment}
\section{Matching Procedure}
\label{app:match}
\end{comment}
\end{document}
\begin{thebibliography}{}
\makeatletter
\relax
\def\mn@urlcharsother{\let\do\@makeother \do\$\do\&\do\#\do\^\do\_\do\%\do\~}
\def\mn@doi{\begingroup\mn@urlcharsother \@ifnextchar [ {\mn@doi@}
{\mn@doi@[]}}
\def\mn@doi@[#1]#2{\def\@tempa{#1}\ifx\@tempa\@empty \href
{http://dx.doi.org/#2} {doi:#2}\else \href {http://dx.doi.org/#2} {#1}\fi
\endgroup}
\def\mn@eprint#1#2{\mn@eprint@#1:#2::\@nil}
\def\mn@eprint@arXiv#1{\href {http://arxiv.org/abs/#1} {{\tt arXiv:#1}}}
\def\mn@eprint@dblp#1{\href {http://dblp.uni-trier.de/rec/bibtex/#1.xml}
{dblp:#1}}
\def\mn@eprint@#1:#2:#3:#4\@nil{\def\@tempa {#1}\def\@tempb {#2}\def\@tempc
{#3}\ifx \@tempc \@empty \let \@tempc \@tempb \let \@tempb \@tempa \fi \ifx
\@tempb \@empty \def\@tempb {arXiv}\fi \@ifundefined
{mn@eprint@\@tempb}{\@tempb:\@tempc}{\expandafter \expandafter \csname
mn@eprint@\@tempb\endcsname \expandafter{\@tempc}}}
\bibitem[\protect\citeauthoryear{{Alexander} \& {Hopman}}{{Alexander} \&
{Hopman}}{2009}]{AlexanderHopman09}
{Alexander} T., {Hopman} C., 2009, \mn@doi [\apj]
{10.1088/0004-637X/697/2/1861}, \href
{https://ui.adsabs.harvard.edu/abs/2009ApJ...697.1861A} {697, 1861}
\bibitem[\protect\citeauthoryear{{Amaro-Seoane} \& {Preto}}{{Amaro-Seoane} \&
{Preto}}{2011}]{Amaro}
{Amaro-Seoane} P., {Preto} M., 2011, \mn@doi [Classical and Quantum Gravity]
{10.1088/0264-9381/28/9/094017}, \href
{https://ui.adsabs.harvard.edu/abs/2011CQGra..28i4017A} {28, 094017}
\bibitem[\protect\citeauthoryear{{Antonini} \& {Rasio}}{{Antonini} \&
{Rasio}}{2016}]{AntoniniRasio16}
{Antonini} F., {Rasio} F.~A., 2016, \mn@doi [\apj]
{10.3847/0004-637X/831/2/187}, \href
{https://ui.adsabs.harvard.edu/abs/2016ApJ...831..187A} {831, 187}
\bibitem[\protect\citeauthoryear{{Arcavi} et~al.,}{{Arcavi}
et~al.}{2014}]{Arcavi+14}
{Arcavi} I., et~al., 2014, \mn@doi [\apj] {10.1088/0004-637X/793/1/38}, \href
{https://ui.adsabs.harvard.edu/abs/2014ApJ...793...38A} {793, 38}
\bibitem[\protect\citeauthoryear{{Bade}, {Komossa} \& {Dahlem}}{{Bade}
et~al.}{1996}]{Bade}
{Bade} N., {Komossa} S., {Dahlem} M., 1996, \aap, \href
{https://ui.adsabs.harvard.edu/abs/1996A&A...309L..35B} {309, L35}
\bibitem[\protect\citeauthoryear{{Bahcall} \& {Wolf}}{{Bahcall} \&
{Wolf}}{1976a}]{BahcallWolf76}
{Bahcall} J.~N., {Wolf} R.~A., 1976a, \mn@doi [\apj] {10.1086/154711}, \href
{https://ui.adsabs.harvard.edu/abs/1976ApJ...209..214B} {209, 214}
\bibitem[\protect\citeauthoryear{{Bahcall} \& {Wolf}}{{Bahcall} \&
{Wolf}}{1976b}]{BW}
{Bahcall} J.~N., {Wolf} R.~A., 1976b, \mn@doi [\apj] {10.1086/154711}, \href
{https://ui.adsabs.harvard.edu/abs/1976ApJ...209..214B} {209, 214}
\bibitem[\protect\citeauthoryear{{Bahcall} \& {Wolf}}{{Bahcall} \&
{Wolf}}{1977}]{BW77}
{Bahcall} J.~N., {Wolf} R.~A., 1977, \mn@doi [\apj] {10.1086/155534}, \href
{https://ui.adsabs.harvard.edu/abs/1977ApJ...216..883B} {216, 883}
\bibitem[\protect\citeauthoryear{{Bavera} et~al.,}{{Bavera}
et~al.}{2022}]{Bavera}
{Bavera} S.~S., et~al., 2022, \mn@doi [arXiv e-prints]
{10.48550/arXiv.2212.10924}, \href
{https://ui.adsabs.harvard.edu/abs/2022arXiv221210924B} {p. arXiv:2212.10924}
\bibitem[\protect\citeauthoryear{{Belczynski}, {Bulik}, {Fryer}, {Ruiter},
{Valsecchi}, {Vink} \& {Hurley}}{{Belczynski} et~al.}{2010}]{Belczynski+10}
{Belczynski} K., {Bulik} T., {Fryer} C.~L., {Ruiter} A., {Valsecchi} F.,
{Vink} J.~S., {Hurley} J.~R., 2010, \mn@doi [\apj]
{10.1088/0004-637X/714/2/1217}, \href
{https://ui.adsabs.harvard.edu/abs/2010ApJ...714.1217B} {714, 1217}
\bibitem[\protect\citeauthoryear{{Bellovary}, {Mac Low}, {McKernan} \&
{Ford}}{{Bellovary} et~al.}{2016}]{Bellovary+16}
{Bellovary} J.~M., {Mac Low} M.-M., {McKernan} B., {Ford} K.~E.~S., 2016,
\mn@doi [\apjl] {10.3847/2041-8205/819/2/L17}, \href
{https://ui.adsabs.harvard.edu/abs/2016ApJ...819L..17B} {819, L17}
\bibitem[\protect\citeauthoryear{{Ben-Ami} et~al.,}{{Ben-Ami}
et~al.}{2022}]{Ultrasat}
{Ben-Ami} S., et~al., 2022, arXiv e-prints, \href
{https://ui.adsabs.harvard.edu/abs/2022arXiv220800159B} {p. arXiv:2208.00159}
\bibitem[\protect\citeauthoryear{{Blagorodnova} et~al.,}{{Blagorodnova}
et~al.}{2019}]{Blagorodnova}
{Blagorodnova} N., et~al., 2019, \mn@doi [\apj] {10.3847/1538-4357/ab04b0},
\href {https://ui.adsabs.harvard.edu/abs/2019ApJ...873...92B} {873, 92}
\bibitem[\protect\citeauthoryear{{Bonnerot}, {Rossi}, {Lodato} \&
{Price}}{{Bonnerot} et~al.}{2016}]{Bonnerot+16}
{Bonnerot} C., {Rossi} E.~M., {Lodato} G., {Price} D.~J., 2016, \mn@doi
[\mnras] {10.1093/mnras/stv2411}, \href
{https://ui.adsabs.harvard.edu/abs/2016MNRAS.455.2253B} {455, 2253}
\bibitem[\protect\citeauthoryear{{Bortolas}}{{Bortolas}}{2022}]{Bortolas}
{Bortolas} E., 2022, \mn@doi [\mnras] {10.1093/mnras/stac262}, \href
{https://ui.adsabs.harvard.edu/abs/2022MNRAS.511.2885B} {511, 2885}
\bibitem[\protect\citeauthoryear{{Bradnick}, {Mandel} \& {Levin}}{{Bradnick}
et~al.}{2017}]{Bradnick+17}
{Bradnick} B., {Mandel} I., {Levin} Y., 2017, \mn@doi [\mnras]
{10.1093/mnras/stx1007}, \href
{https://ui.adsabs.harvard.edu/abs/2017MNRAS.469.2042B} {469, 2042}
\bibitem[\protect\citeauthoryear{{Bricman} \& {Gomboc}}{{Bricman} \&
{Gomboc}}{2020}]{BricmanGomboc20}
{Bricman} K., {Gomboc} A., 2020, \mn@doi [\apj] {10.3847/1538-4357/ab6989},
\href {https://ui.adsabs.harvard.edu/abs/2020ApJ...890...73B} {890, 73}
\bibitem[\protect\citeauthoryear{{Broggi}, {Bortolas}, {Bonetti}, {Sesana} \&
{Dotti}}{{Broggi} et~al.}{2022}]{Broggi}
{Broggi} L., {Bortolas} E., {Bonetti} M., {Sesana} A., {Dotti} M., 2022,
\mn@doi [\mnras] {10.1093/mnras/stac1453}, \href
{https://ui.adsabs.harvard.edu/abs/2022MNRAS.514.3270B} {514, 3270}
\bibitem[\protect\citeauthoryear{{Chornock} et~al.,}{{Chornock}
et~al.}{2014}]{Chornock}
{Chornock} R., et~al., 2014, \mn@doi [\apj] {10.1088/0004-637X/780/1/44},
\href {https://ui.adsabs.harvard.edu/abs/2014ApJ...780...44C} {780, 44}
\bibitem[\protect\citeauthoryear{{Cohn} \& {Kulsrud}}{{Cohn} \&
{Kulsrud}}{1978}]{CohnKulsrud78}
{Cohn} H., {Kulsrud} R.~M., 1978, \mn@doi [\apj] {10.1086/156685}, \href
{https://ui.adsabs.harvard.edu/abs/1978ApJ...226.1087C} {226, 1087}
\bibitem[\protect\citeauthoryear{{Dai}, {McKinney} \& {Miller}}{{Dai}
et~al.}{2015}]{Dai+15}
{Dai} L., {McKinney} J.~C., {Miller} M.~C., 2015, \mn@doi [\apjl]
{10.1088/2041-8205/812/2/L39}, \href
{https://ui.adsabs.harvard.edu/abs/2015ApJ...812L..39D} {812, L39}
\bibitem[\protect\citeauthoryear{{Evans} \& {Kochanek}}{{Evans} \&
{Kochanek}}{1989}]{EvansKochanek89}
{Evans} C.~R., {Kochanek} C.~S., 1989, \mn@doi [\apjl] {10.1086/185567},
\href {https://ui.adsabs.harvard.edu/abs/1989ApJ...346L..13E} {346, L13}
\bibitem[\protect\citeauthoryear{{Fabian}, {Pringle} \& {Rees}}{{Fabian}
et~al.}{1975}]{Fabian+75}
{Fabian} A.~C., {Pringle} J.~E., {Rees} M.~J., 1975, \mn@doi [\mnras]
{10.1093/mnras/172.1.15P}, \href
{https://ui.adsabs.harvard.edu/abs/1975MNRAS.172P..15F} {172, 15}
\bibitem[\protect\citeauthoryear{{Frank} \& {Rees}}{{Frank} \&
{Rees}}{1976}]{FrankRees76}
{Frank} J., {Rees} M.~J., 1976, \mn@doi [\mnras] {10.1093/mnras/176.3.633},
\href {https://ui.adsabs.harvard.edu/abs/1976MNRAS.176..633F} {176, 633}
\bibitem[\protect\citeauthoryear{{French}, {Arcavi} \& {Zabludoff}}{{French}
et~al.}{2016}]{French+16}
{French} K.~D., {Arcavi} I., {Zabludoff} A., 2016, \mn@doi [\apjl]
{10.3847/2041-8205/818/1/L21}, \href
{https://ui.adsabs.harvard.edu/abs/2016ApJ...818L..21F} {818, L21}
\bibitem[\protect\citeauthoryear{{French}, {Arcavi} \& {Zabludoff}}{{French}
et~al.}{2017}]{French+17}
{French} K.~D., {Arcavi} I., {Zabludoff} A., 2017, \mn@doi [\apj]
{10.3847/1538-4357/835/2/176}, \href
{https://ui.adsabs.harvard.edu/abs/2017ApJ...835..176F} {835, 176}
\bibitem[\protect\citeauthoryear{{French}, {Wevers}, {Law-Smith}, {Graur} \&
{Zabludoff}}{{French} et~al.}{2020}]{French20}
{French} K.~D., {Wevers} T., {Law-Smith} J., {Graur} O., {Zabludoff}
A.~I., 2020, \mn@doi [\ssr] {10.1007/s11214-020-00657-y}, \href
{https://ui.adsabs.harvard.edu/abs/2020SSRv..216...32F} {216, 32}
\bibitem[\protect\citeauthoryear{{Generozov}, {Stone}, {Metzger} \&
{Ostriker}}{{Generozov} et~al.}{2018}]{Generozov+18}
{Generozov} A., {Stone} N.~C., {Metzger} B.~D., {Ostriker} J.~P., 2018,
\mn@doi [\mnras] {10.1093/mnras/sty1262}, \href
{https://ui.adsabs.harvard.edu/abs/2018MNRAS.478.4030G} {478, 4030}
\bibitem[\protect\citeauthoryear{{Gezari} et~al.,}{{Gezari}
et~al.}{2012}]{Gezari}
{Gezari} S., et~al., 2012, \mn@doi [\nat] {10.1038/nature10990}, \href
{https://ui.adsabs.harvard.edu/abs/2012Natur.485..217G} {485, 217}
\bibitem[\protect\citeauthoryear{{Gilbaum} \& {Stone}}{{Gilbaum} \&
{Stone}}{2022}]{GilbaumStone22}
{Gilbaum} S., {Stone} N.~C., 2022, \mn@doi [\apj] {10.3847/1538-4357/ac4ded},
\href {https://ui.adsabs.harvard.edu/abs/2022ApJ...928..191G} {928, 191}
\bibitem[\protect\citeauthoryear{{Graur}, {French}, {Zahid}, {Guillochon},
{Mandel}, {Auchettl} \& {Zabludoff}}{{Graur} et~al.}{2018}]{Graur}
{Graur} O., {French} K.~D., {Zahid} H.~J., {Guillochon} J., {Mandel} K.~S.,
{Auchettl} K., {Zabludoff} A.~I., 2018, \mn@doi [\apj]
{10.3847/1538-4357/aaa3fd}, \href
{https://ui.adsabs.harvard.edu/abs/2018ApJ...853...39G} {853, 39}
\bibitem[\protect\citeauthoryear{{Gualandris} \& {Merritt}}{{Gualandris} \&
{Merritt}}{2008}]{GualandrisMerritt08}
{Gualandris} A., {Merritt} D., 2008, \mn@doi [\apj] {10.1086/586877}, \href
{https://ui.adsabs.harvard.edu/abs/2008ApJ...678..780G} {678, 780}
\bibitem[\protect\citeauthoryear{{Hammerstein} et~al.,}{{Hammerstein}
et~al.}{2021}]{Hammerstein21}
{Hammerstein} E., et~al., 2021, \mn@doi [\apjl] {10.3847/2041-8213/abdcb4},
\href {https://ui.adsabs.harvard.edu/abs/2021ApJ...908L..20H} {908, L20}
\bibitem[\protect\citeauthoryear{{Hammerstein} et~al.,}{{Hammerstein}
et~al.}{2022}]{Hammerstein22}
{Hammerstein} E., et~al., 2022, arXiv e-prints, \href
{https://ui.adsabs.harvard.edu/abs/2022arXiv220301461H} {p. arXiv:2203.01461}
\bibitem[\protect\citeauthoryear{{Hayasaki}, {Stone} \& {Loeb}}{{Hayasaki}
et~al.}{2013}]{Hayasaki+13}
{Hayasaki} K., {Stone} N., {Loeb} A., 2013, \mn@doi [\mnras]
{10.1093/mnras/stt871}, \href
{https://ui.adsabs.harvard.edu/abs/2013MNRAS.434..909H} {434, 909}
\bibitem[\protect\citeauthoryear{{H{\'e}non}}{{H{\'e}non}}{1960a}]{Henona}
{H{\'e}non} M., 1960a, Annales d'Astrophysique, \href
{https://ui.adsabs.harvard.edu/abs/1960AnAp...23..467H} {23, 467}
\bibitem[\protect\citeauthoryear{{H{\'e}non}}{{H{\'e}non}}{1960b}]{Henon}
{H{\'e}non} M., 1960b, Annales d'Astrophysique, \href
{https://ui.adsabs.harvard.edu/abs/1960AnAp...23..668H} {23, 668}
\bibitem[\protect\citeauthoryear{{Hills}}{{Hills}}{1975}]{Hills75}
{Hills} J.~G., 1975, \mn@doi [\nat] {10.1038/254295a0}, \href
{https://ui.adsabs.harvard.edu/abs/1975Natur.254..295H} {254, 295}
\bibitem[\protect\citeauthoryear{{Hills}}{{Hills}}{1988}]{Hills88}
{Hills} J.~G., 1988, \mn@doi [\nat] {10.1038/331687a0}, \href
{https://ui.adsabs.harvard.edu/abs/1988Natur.331..687H} {331, 687}
\bibitem[\protect\citeauthoryear{{Hinkle} et~al.,}{{Hinkle}
et~al.}{2021}]{Hinkle}
{Hinkle} J.~T., et~al., 2021, \mn@doi [\mnras] {10.1093/mnras/staa3170}, \href
{https://ui.adsabs.harvard.edu/abs/2021MNRAS.500.1673H} {500, 1673}
\bibitem[\protect\citeauthoryear{{Holoien} et~al.,}{{Holoien}
et~al.}{2016}]{Holoien16}
{Holoien} T.~W.~S., et~al., 2016, \mn@doi [\mnras] {10.1093/mnras/stv2486},
\href {https://ui.adsabs.harvard.edu/abs/2016MNRAS.455.2918H} {455, 2918}
\bibitem[\protect\citeauthoryear{{Hung} et~al.,}{{Hung} et~al.}{2017}]{Hung}
{Hung} T., et~al., 2017, \mn@doi [\apj] {10.3847/1538-4357/aa7337}, \href
{https://ui.adsabs.harvard.edu/abs/2017ApJ...842...29H} {842, 29}
\bibitem[\protect\citeauthoryear{{Ivezi{\'c}} et~al.,}{{Ivezi{\'c}}
et~al.}{2019}]{Ivezi}
{Ivezi{\'c}} {\v{Z}}., et~al., 2019, \mn@doi [\apj]
{10.3847/1538-4357/ab042c}, \href
{https://ui.adsabs.harvard.edu/abs/2019ApJ...873..111I} {873, 111}
\bibitem[\protect\citeauthoryear{{Komossa} \& {Bade}}{{Komossa} \&
{Bade}}{1999}]{Komossa}
{Komossa} S., {Bade} N., 1999, \aap, \href
{https://ui.adsabs.harvard.edu/abs/1999A&A...343..775K} {343, 775}
\bibitem[\protect\citeauthoryear{{Kremer}, {Lombardi}, {Lu}, {Piro} \&
{Rasio}}{{Kremer} et~al.}{2022}]{Kremer+22}
{Kremer} K., {Lombardi} J.~C., {Lu} W., {Piro} A.~L., {Rasio} F.~A.,
2022, \mn@doi [\apj] {10.3847/1538-4357/ac714f}, \href
{https://ui.adsabs.harvard.edu/abs/2022ApJ...933..203K} {933, 203}
\bibitem[\protect\citeauthoryear{{Law-Smith}, {Ramirez-Ruiz}, {Ellison} \&
{Foley}}{{Law-Smith} et~al.}{2017}]{Law}
{Law-Smith} J., {Ramirez-Ruiz} E., {Ellison} S.~L., {Foley} R.~J., 2017,
\mn@doi [\apj] {10.3847/1538-4357/aa94c7}, \href
{https://ui.adsabs.harvard.edu/abs/2017ApJ...850...22L} {850, 22}
\bibitem[\protect\citeauthoryear{{Lee} \& {Ostriker}}{{Lee} \&
{Ostriker}}{1986}]{LeeOstriker86}
{Lee} H.~M., {Ostriker} J.~P., 1986, \mn@doi [\apj] {10.1086/164674}, \href
{https://ui.adsabs.harvard.edu/abs/1986ApJ...310..176L} {310, 176}
\bibitem[\protect\citeauthoryear{{Lightman} \& {Shapiro}}{{Lightman} \&
{Shapiro}}{1977}]{LightmanShapiro77}
{Lightman} A.~P., {Shapiro} S.~L., 1977, \mn@doi [\apj] {10.1086/154925},
\href {https://ui.adsabs.harvard.edu/abs/1977ApJ...211..244L} {211, 244}
\bibitem[\protect\citeauthoryear{{Lin} \& {Tremaine}}{{Lin} \&
{Tremaine}}{1980}]{LinTremaine80}
{Lin} D.~N.~C., {Tremaine} S., 1980, \mn@doi [\apj] {10.1086/158513}, \href
{https://ui.adsabs.harvard.edu/abs/1980ApJ...242..789L} {242, 789}
\bibitem[\protect\citeauthoryear{{Lu} \& {Bonnerot}}{{Lu} \&
{Bonnerot}}{2020}]{LuBonnerot20}
{Lu} W., {Bonnerot} C., 2020, \mn@doi [\mnras] {10.1093/mnras/stz3405}, \href
{https://ui.adsabs.harvard.edu/abs/2020MNRAS.492..686L} {492, 686}
\bibitem[\protect\citeauthoryear{{Magorrian} \& {Tremaine}}{{Magorrian} \&
{Tremaine}}{1999}]{MagorrianTremaine99}
{Magorrian} J., {Tremaine} S., 1999, \mn@doi [\mnras]
{10.1046/j.1365-8711.1999.02853.x}, \href
{https://ui.adsabs.harvard.edu/abs/1999MNRAS.309..447M} {309, 447}
\bibitem[\protect\citeauthoryear{{Merritt}}{{Merritt}}{2013}]{Merritt13}
{Merritt} D., 2013, {Dynamics and Evolution of Galactic Nuclei}
\bibitem[\protect\citeauthoryear{{Merritt} \& {Wang}}{{Merritt} \&
{Wang}}{2005}]{MerrittWang05}
{Merritt} D., {Wang} J., 2005, \mn@doi [\apjl] {10.1086/429272}, \href
{https://ui.adsabs.harvard.edu/abs/2005ApJ...621L.101M} {621, L101}
\bibitem[\protect\citeauthoryear{{Metzger} \& {Stone}}{{Metzger} \&
{Stone}}{2016}]{MetzgerStone16}
{Metzger} B.~D., {Stone} N.~C., 2016, \mn@doi [\mnras]
{10.1093/mnras/stw1394}, \href
{https://ui.adsabs.harvard.edu/abs/2016MNRAS.461..948M} {461, 948}
\bibitem[\protect\citeauthoryear{{Nicholl} et~al.,}{{Nicholl}
et~al.}{2019}]{Nicholl}
{Nicholl} M., et~al., 2019, \mn@doi [\mnras] {10.1093/mnras/stz1837}, \href
{https://ui.adsabs.harvard.edu/abs/2019MNRAS.488.1878N} {488, 1878}
\bibitem[\protect\citeauthoryear{{O'Leary} \& {Loeb}}{{O'Leary} \&
{Loeb}}{2008}]{OLearyLoeb08}
{O'Leary} R.~M., {Loeb} A., 2008, \mn@doi [\mnras]
{10.1111/j.1365-2966.2007.12531.x}, \href
{https://ui.adsabs.harvard.edu/abs/2008MNRAS.383...86O} {383, 86}
\bibitem[\protect\citeauthoryear{{Perets}, {Hopman} \& {Alexander}}{{Perets}
et~al.}{2007}]{Perets+07}
{Perets} H.~B., {Hopman} C., {Alexander} T., 2007, \mn@doi [\apj]
{10.1086/510377}, \href
{https://ui.adsabs.harvard.edu/abs/2007ApJ...656..709P} {656, 709}
\bibitem[\protect\citeauthoryear{{Perets}, {Li}, {Lombardi} \&
{Milcarek}}{{Perets} et~al.}{2016}]{Perets+16}
{Perets} H.~B., {Li} Z., {Lombardi} James~C. J., {Milcarek} Stephen~R. J.,
2016, \mn@doi [\apj] {10.3847/0004-637X/823/2/113}, \href
{https://ui.adsabs.harvard.edu/abs/2016ApJ...823..113P} {823, 113}
\bibitem[\protect\citeauthoryear{{Press} \& {Teukolsky}}{{Press} \&
{Teukolsky}}{1977}]{PressTeukolsky77}
{Press} W.~H., {Teukolsky} S.~A., 1977, \mn@doi [\apj] {10.1086/155143},
\href {https://ui.adsabs.harvard.edu/abs/1977ApJ...213..183P} {213, 183}
\bibitem[\protect\citeauthoryear{{Preto} \& {Amaro-Seoane}}{{Preto} \&
{Amaro-Seoane}}{2010a}]{PretoAmaroSeoane10}
{Preto} M., {Amaro-Seoane} P., 2010a, \mn@doi [\apjl]
{10.1088/2041-8205/708/1/L42}, \href
{https://ui.adsabs.harvard.edu/abs/2010ApJ...708L..42P} {708, L42}
\bibitem[\protect\citeauthoryear{{Preto} \& {Amaro-Seoane}}{{Preto} \&
{Amaro-Seoane}}{2010b}]{Preto}
{Preto} M., {Amaro-Seoane} P., 2010b, \mn@doi [\apjl]
{10.1088/2041-8205/708/1/L42}, \href
{https://ui.adsabs.harvard.edu/abs/2010ApJ...708L..42P} {708, L42}
\bibitem[\protect\citeauthoryear{{Rees}}{{Rees}}{1988}]{Rees88}
{Rees} M.~J., 1988, \mn@doi [\nat] {10.1038/333523a0}, \href
{https://ui.adsabs.harvard.edu/abs/1988Natur.333..523R} {333, 523}
\bibitem[\protect\citeauthoryear{{Sazonov} et~al.,}{{Sazonov}
et~al.}{2021}]{Sazonov}
{Sazonov} S., et~al., 2021, \mn@doi [\mnras] {10.1093/mnras/stab2843}, \href
{https://ui.adsabs.harvard.edu/abs/2021MNRAS.508.3820S} {508, 3820}
\bibitem[\protect\citeauthoryear{{Spera} \& {Mapelli}}{{Spera} \&
{Mapelli}}{2017}]{SperaMapelli17}
{Spera} M., {Mapelli} M., 2017, \mn@doi [\mnras] {10.1093/mnras/stx1576},
\href {https://ui.adsabs.harvard.edu/abs/2017MNRAS.470.4739S} {470, 4739}
\bibitem[\protect\citeauthoryear{{Stone} \& {Metzger}}{{Stone} \&
{Metzger}}{2016}]{StoneMetzger16}
{Stone} N.~C., {Metzger} B.~D., 2016, \mn@doi [\mnras]
{10.1093/mnras/stv2281}, \href
{https://ui.adsabs.harvard.edu/abs/2016MNRAS.455..859S} {455, 859}
\bibitem[\protect\citeauthoryear{{Stone} \& {van Velzen}}{{Stone} \& {van
Velzen}}{2016}]{StonevanVelzen16}
{Stone} N.~C., {van Velzen} S., 2016, \mn@doi [\apjl]
{10.3847/2041-8205/825/1/L14}, \href
{https://ui.adsabs.harvard.edu/abs/2016ApJ...825L..14S} {825, L14}
\bibitem[\protect\citeauthoryear{{Stone}, {K{\"u}pper} \& {Ostriker}}{{Stone}
et~al.}{2017}]{Stone+17}
{Stone} N.~C., {K{\"u}pper} A. H.~W., {Ostriker} J.~P., 2017, \mn@doi
[\mnras] {10.1093/mnras/stx097}, \href
{https://ui.adsabs.harvard.edu/abs/2017MNRAS.467.4180S} {467, 4180}
\bibitem[\protect\citeauthoryear{{Stone}, {Generozov}, {Vasiliev} \&
{Metzger}}{{Stone} et~al.}{2018}]{Stone+18}
{Stone} N.~C., {Generozov} A., {Vasiliev} E., {Metzger} B.~D., 2018,
\mn@doi [\mnras] {10.1093/mnras/sty2045}, \href
{https://ui.adsabs.harvard.edu/abs/2018MNRAS.480.5060S} {480, 5060}
\bibitem[\protect\citeauthoryear{{Stone}, {Vasiliev}, {Kesden}, {Rossi},
{Perets} \& {Amaro-Seoane}}{{Stone} et~al.}{2020}]{Stone+20}
{Stone} N.~C., {Vasiliev} E., {Kesden} M., {Rossi} E.~M., {Perets} H.~B.,
{Amaro-Seoane} P., 2020, \mn@doi [\ssr] {10.1007/s11214-020-00651-4}, \href
{https://ui.adsabs.harvard.edu/abs/2020SSRv..216...35S} {216, 35}
\bibitem[\protect\citeauthoryear{{Syer}, {Clarke} \& {Rees}}{{Syer}
et~al.}{1991}]{SyerRees91}
{Syer} D., {Clarke} C.~J., {Rees} M.~J., 1991, \mn@doi [\mnras]
{10.1093/mnras/250.3.505}, \href
{https://ui.adsabs.harvard.edu/abs/1991MNRAS.250..505S} {250, 505}
\bibitem[\protect\citeauthoryear{{Sz{\"o}lgy{\'e}n} \&
{Kocsis}}{{Sz{\"o}lgy{\'e}n} \& {Kocsis}}{2018}]{SzolgyenKocsis18}
{Sz{\"o}lgy{\'e}n} {\'A}., {Kocsis} B., 2018, \mn@doi [\prl]
{10.1103/PhysRevLett.121.101101}, \href
{https://ui.adsabs.harvard.edu/abs/2018PhRvL.121j1101S} {121, 101101}
\bibitem[\protect\citeauthoryear{{Tagawa}, {Haiman} \& {Kocsis}}{{Tagawa}
et~al.}{2020}]{Tagawa+20}
{Tagawa} H., {Haiman} Z., {Kocsis} B., 2020, \mn@doi [\apj]
{10.3847/1538-4357/ab9b8c}, \href
{https://ui.adsabs.harvard.edu/abs/2020ApJ...898...25T} {898, 25}
\bibitem[\protect\citeauthoryear{{Wang} \& {Merritt}}{{Wang} \&
{Merritt}}{2004}]{WangMerritt04}
{Wang} J., {Merritt} D., 2004, \mn@doi [\apj] {10.1086/379767}, \href
{https://ui.adsabs.harvard.edu/abs/2004ApJ...600..149W} {600, 149}
\bibitem[\protect\citeauthoryear{{Young}}{{Young}}{1980}]{Young}
{Young} P., 1980, \mn@doi [\apj] {10.1086/158553}, \href
{https://ui.adsabs.harvard.edu/abs/1980ApJ...242.1232Y} {242, 1232}
\bibitem[\protect\citeauthoryear{{Yu} \& {Tremaine}}{{Yu} \&
{Tremaine}}{2003}]{YuTremaine03}
{Yu} Q., {Tremaine} S., 2003, \mn@doi [\apj] {10.1086/379546}, \href
{https://ui.adsabs.harvard.edu/abs/2003ApJ...599.1129Y} {599, 1129}
\bibitem[\protect\citeauthoryear{{van Velzen}}{{van
Velzen}}{2018}]{vanVelzen18}
{van Velzen} S., 2018, \mn@doi [\apj] {10.3847/1538-4357/aa998e}, \href
{https://ui.adsabs.harvard.edu/abs/2018ApJ...852...72V} {852, 72}
\bibitem[\protect\citeauthoryear{{van Velzen} et~al.,}{{van Velzen}
et~al.}{2011}]{vanVelzen11}
{van Velzen} S., et~al., 2011, \mn@doi [\apj] {10.1088/0004-637X/741/2/73},
\href {https://ui.adsabs.harvard.edu/abs/2011ApJ...741...73V} {741, 73}
\bibitem[\protect\citeauthoryear{{van Velzen}, {Stone}, {Metzger}, {Gezari},
{Brown} \& {Fruchter}}{{van Velzen} et~al.}{2019}]{vanVelzen+19}
{van Velzen} S., {Stone} N.~C., {Metzger} B.~D., {Gezari} S., {Brown}
T.~M., {Fruchter} A.~S., 2019, \mn@doi [\apj] {10.3847/1538-4357/ab1844},
\href {https://ui.adsabs.harvard.edu/abs/2019ApJ...878...82V} {878, 82}
\makeatother
\end{thebibliography}