Arxiv:2104.00074V2 [Gr-Qc] 25 Jul 2021

Ringdown of Charged Compact Objects using Membrane Paradigm

Mostaﬁzur Rahman1, ∗ and Arpan Bhattacharyya1, †

1Indian Institute of Technology, Gandhinagar, Gujarat-382355, India Although black holes are an integral part of the standard model of astrophysics and cosmol- ogy, their existence poses some serious fundamental problems. In recent years, several horizonless compact object models were proposed to address those issues. As the gravitational wave detectors started to observe more and more merger events with a large signal-to-noise ratio, gravitational wave spectroscopy could hold the key to uncover the existence of these objects. This is because the late time ringdown signals of horizonless compact objects diﬀer from that of the black holes. In this paper, we study the ringdown properties of charged compact objects and compare them with those obtained in the black hole scenario. Since the internal structure and the equation of state of these compact objects are largely unknown, we employ membrane paradigm to obtain appropriate boundary conditions for the perturbations of these objects. This model can describe the ringdown properties of a large variety of compact objects.

I. INTRODUCTION quantum mechanical description of gravity. Several horizonless dark exotic compact object (ECO) models were proposed to address some of those issues in recent years. The black holes are defined as the region of spacetime It has been conjectured that [28–32], one can describe where the gravity is so strong that even light can not black holes as ensembles of micro-states represented by escape their crushing grip. They are widely accepted as smooth, horizonless geometries without closed time-like dark astrophysical objects, which are the end stage of curves, termed as fuzzballs. Sometimes, it is possible to dying stars with a mass roughly greater than three solar find such horizonless solutions, with the same charges as masses [1,2]. The idea of black holes is interesting for a black hole from supergravity for eg. [33–35]. In the several reasons. The physics of gravitational collapse and gravastar model, the event horizon is replaced by a thin the formation of black holes are well understood. More- shell that connects a de Sitter interior core with an ex- over, the spacetime outside the black hole is perfectly reg- ternal Schwarzschild geometry [36–38]. They can arise ular as curvature singularities are hidden inside its event as a result of phase transition near what could have been horizon and thus do not hinder measurements outside the position of the event horizon. They can also origi- the black hole region [3,4]. Furthermore, it has been es- nate through quantum field theoretical processes, which tablished that black holes in general relativity (GR) are replace the horizon with a quantum phase boundary [39– stable against external perturbations [5–10]. They are 41]. Axion or axion like dark energy models predict the also the simplest astrophysical objects. “No-hair” theo- existance of dark compact objects, called the boson stars rem says that stationary black holes in GR can be char- which can form naturally from gravitational collapse [42– acterized by their mass, charge and angular momentum 44]. [11–13]. These properties of black holes make them an ideal laboratory to probe physics in strong gravity regime It is expected that the observational signatures of the [14–16]. Recent discovery of gravitational waves (GWs) ECOs will differ in general from those of the black holes. and black hole shadow shows progress in this direction In order to study those signatures, we need to quan- [17–19]. tify the characteristics of these objects. Since the de- Nevertheless, the theory of black holes poses some serious tails about their internal structure are largely unknown, questions to the conventional view on nature. Although a model-independent way has been proposed by address- arXiv:2104.00074v2 [gr-qc] 25 Jul 2021 the exterior of these objects is free from pathologies, the ing the following questions: how compact and dark are same can not be said about their interior. These objects these objects are [45, 46]? These parameters allow us shelter singularities and Cauchy horizon (for charged and to pragmatically quantify the difference between ECO rotating black holes ) in their interior where classical and black hole spacetime. From the observational per- GR breaks down [20–25]. Moreover, Hawking showed spective, gravitational wave spectroscopy could be a very that these objects can emit radiations using semi-classical powerful tool to distinguish these objects from black analysis [26]. However, complete evaporation of the black holes [46–50]. Merger of compact object binaries are hole breaks the unitary property of quantum mechanics among the most important astrophysical processes that as it can turn a pure state into a mixed state[27]. The res- give rise to the GWs. It is widely believed that the post- olution of these problems is essential to understand the merger remnant eventually settles down to a stationary state by emitting GWs through a process called ringdown. These waves not only contains information about the nature of the final object but also about the underly- ∗ mostafi[email protected] ing theory of gravity [14–16, 51–53]. It has been noted in † [email protected] 2 the literature that the ringdown properties of ECOs gen- boundary conditions of charged compacts objects. In erally differ from that of the black holes. It is well-known Section IV and Section V, we present the ringdown prop- that the ringdown signal of a black hole dies down at later erties and QNMs of these objects. We discuss the quasi- times [54–57]. However, one obtains modulated repeti- normal modes of the compact object in the limit Q → M tive signals from the compact object, so-called the grav- in Section VI. A discussion about the detectability of itational wave echoes [45–50, 58]. The repetition time is gravitation wave echoes is given Section VII. Finally, usually governed by the compactness parameter, whereas with some remarks, we conclude our paper. the modulation depends on the reflection coefficient of the object. In fact, recently some authors claimed that Notation and Convention: Throughout the paper, pos- they have found observational evidence of GW echoes in itive signature convention (−, +, +, +) is used. Greek the LIGO data [59–63]. However, these claims are still indices are used to denote four-dimensional tensors, shrouded in controversy [64–67]. whereas lower case roman indices are used to denote three-dimensional tensors. The fundamental constants One of the most important steps to study the ring- are assumed as ~ = c = G = 1. down properties of ECOs is to set up appropriate boundary conditions. In [68, 69] for objects like fuzzballs and gravastars, the authors used gauge/gravity duality and membrane paradigm to study it. In black hole membrane II. MEMBRANE PARADIGM FOR PERTURBED CHARGED COMPACT OBJECTS paradigm, the event horizon is viewed as 2+1 dimensional viscous fluid membrane [70–73]. This formulation gives us an intuitive understanding of many of the properties The black hole membrane paradigm provides an intu- of black holes. In [46], the authors used the membrane itive way to understand the dynamics of a black hole. In paradigm to obtain the boundary conditions for neutral the membrane paradigm, we consider a fictitious 2 + 1 ECOs in a model-independent way. Ref. [74] done the dimensional time-like surface just outside the black hole same for rotating compact objects. event horizon H, so-called the stretched horizon S [70– In this paper, we study the ringdown properties of 73]. It can be thought of as the world tube of a fam- charged compact objects using the membrane paradigm. ily of fudicial observers (FIDO) which are at rest with Motivation for considering charged black holes stems respect to the black hole. We denote the four veloci- µ from the fact that they provide us with a variety of in- ties of the FIDOs by u . Thus, the time-like congru- µ teresting phenomenological scenarios, e.g., in the con- ence u is the generator of the stretched horizon. The µ text of standard model [75], in astrophysical context [76]. outward-pointing unit normal to S is denoted by n . The It is well known that electrically charged astrophysical energy-momentum tensor τab of some matter sources at objects get neutralized rather quickly by accreting ion- stretched horizon produces a jump in the extrinsic cur- a c ized plasma and other quantum mechanical processes vature K b = h bna;c in accordance to Israel junction like Schwinger’s pair production and Hawking radiation condition [86–88], as a consequence of electron’s large charge-to-mass ratio 21 τab = K hab − Kab (1) e/me ≈ 10 [77, 78]. Nevertheless, certain minicharged dark matter models can evade the fast neutralization process since the charge of these minicharged particles is where, K is the trace of the extrinsic curvature Kab and K = K+ − K− denotes its discontinuity across the only a fraction of the electrons’ charge [77]. Another ab ab ab interesting possibility is the magnetically charged black stretched horizon. The metric tensor on S is represented holes [75, 78] and compact objects like magnetars [76]. by hab = gab − nanb. In the membrane paradigm, the It has been discussed in the literature that these objects energy-momentum tensor is chosen to contain all the in- are relatively long-lived than their electrically charged formation about the interior [88]. Hence the following counterpart (see [78] for further discussion). Since the condition is satisfied [46, 88], construction discussed here is quite generic, it can be ap- − plied to a variety of compact object models [28–32, 36– Kab = 0 , τab = Khab − Kab (2) 38, 42–44, 79–82] and could potentially shed some light to the observational signatures of some modified gravity where, we have omitted the prefix in K+ for brevity. theories e.g. the brane-world models [53, 83–85]. This ab When the stretched horizon coincides with H, it behaves construction is also interesting from the phenomenolog- like a 2 + 1 dimensional dissipative fluid membrane with ical point of view since it gives an intuitive understand- its energy-momentum tensor given by the following rela- ing of the interactions of electromagnetic fields with the tion [46, 88–90] charged membrane.

The paper is organized as follows: In Section II, we τab = ρuaub + (p − ζΘ)γab − 2ησab . (3) brieﬂy discuss the membrane paradigm. In Section III, we show how this paradigm leads to the appropriate Here ρ and p represents the density and pressure of the ﬂuid while ζ and η denotes the bulk and shear viscosity. 3

Θ is the expansion, σab is the shear tensor, ua is the 3- Q ωBH velocity of the fluid and γab = hab + uaub is the spacelike 0.0 0.373673895745753-0.088969864662434 i cross section of S. For a black hole in general relativity, ρ 0.4 0.378117882893637-0.089365820010727 i and p depend on the spacetime geometry whereas the pa- 0.6 0.385757219512767-0.089607106118930 i rameters ζ and η take the following value ζ ≡ −1/16π BH 0.8 0.400503177075279-0.088609031463709 i and ηBH ≡ 1/16π respectively. In this paper, we are interested in studying the ringdown TABLE I: The quasinormal frequencies of a properties of a charged compact objects. Hence we con- Reissner-Nordströmblack hole in the fundamental l = 2 sider that the fiducial observers are located at the surface mode. of a charged compact object. The geometry outside the object is described by a Reissner-Nordströmspace-time [3,4, 91, 92], and making use of the diffeomorphism symmetry of the system, we find that the only non-vanishing components dr2 of δg in the odd parity sector (in Regge-Wheeler gauge) ds2 = −f(r) dt2 + + r2(dθ2 + sin2 θ dφ2) , (4) µν f(r) are [7],

iωt which is a solution of the Einstein-Maxwell equation δgtφ = h0(r) e sin θ ∂θYl0 (8) iωt δgrφ = h1(r) e sin θ ∂θYl0 Q G = 8πT ,F = − dt ∧ dr (5) µν µν r2 and their symmetric counterparts. Here, we set the 1 g azimuth quantum number m to zero without losing T = F F α − µν F F γδ ,F µν = 0 . (6) µν 4π µα ν 4 γδ ;ν any generality. Similarly, the only independent component of the perturbed field tensor is δFθφ. Substitut- ing the expressions for δg , δg and δF in Einstein- 2 2 tφ rφ θφ Here, the function f(r) = 1−2M/r+Q /r has two zeros Maxwell equation Eq. (5), it can be shown that the func- p 2 2 p 2 2 at re = M + M − Q and rc = M − M − Q , tions h0 and h1 are related by the following expression, corresponding to the location of its event and Cauchy h0 = (if(r)/ω)∂r(h1f(r)) [7]. Redefining the function horizon. The surface of the compact object is situated at h1(r) = (2iωrU(r))/µf(r) and δFθφ = l(l + 1)H(r), we R = re(1 + ), where is a measure of the compactness obtain a pair for coupled differential equations for U(r) of the object. In the limit → 0, the stretched horizon and H(r) where µ2 = l(l + 1) − 2[91]. The equations for coincides with the event horizon. Using the condition U(r) and H(r) can be decoupled if we use the following given in Eq. (2) for the spacetime mentioned above, we substitution [91], obtain the following expression for ρ and p at the surface √ of the compact object [46], Z1 = q1H(r) + −q1q2U(r) √ (9) Z2 = − −q1q2H(r) + q1U(r) pf(R) 2f(R) + f 0(R)R ρ(R) = − , p(R) = p (7) 4πR 16πR f(R) where the expresssion for q1 and q2 can be written as follows,

Note that the pressure term diverges, whereas the density p 2 2 2 q1 = 3M + 9M + 4 (l + l − 2) Q vanishes at the black hole limit → 0. (10) p 2 2 2 q2 = 3M − 9M + 4 (l + l − 2) Q The odd parity (axial) perturbation equation for a III. PERTURBATION OF CHARGED charged compact object takes the following form [91] COMPACT OBJECTS 2 d Zi 2 2 + ω − Vi(r) Zi = 0 (11) In this section, we discuss governing equations and dr∗ the boundary conditions for the perturbations of charged where dr∗ = dr/f(r) is the tortoise co-ordinate, compact objects. The perturbation equations for Reissner-Nordströmblack holes were obtained by sev- f(r) 4Q2 q V (r) = l(l + 1) + − j (12) eral authors using different methods[7, 93, 94]. Here, we i r2 r2 r mainly adopt the method described in [7]. and i, j = 1, 2 (i 6= j). Note that, in the Q → 0 limit, per- We consider the first-order perturbations δgµν and turbation equation for Z2 corresponds to Regge-Wheeler δFµν in the background quantities gµν and Fµν respec- equation [5, 91]. tively so that the perturbed metric and field tensor be- comesg ˜µν = gµν + δgµν and F˜µν = Fµν + δFµν . By To study these compact objects’ ringdown properties, expanding δgµν and δFµν in terms of tensor harmonics we need to impose the boundary conditions at infinity 4 and the surface of the object. In the case of a black hole, radially freely falling observers (FFO) who√ satisfies the quasinormal modes are defined as the solution of Eq. (11) following equation of motion, dr/dt = −f 1 − f. To with the following boundary condition that there are them, the fiducial observers sitting on the surface of the only incoming waves at the event horizon outgoing waves compact object (FIDO) are moving outward with veloc- at the infinity. The quasinormal modes of a Reissner- ity [71] Nordströmblack hole in fundamental l = 2 mode are presented in Table. I for different values of charge param- 1 dr p v = −√ = 1 − f(R) , (16) eter Q. The outgoing boundary condition remains true f dτ r=R for compact objects. However, the boundary conditions at the surface of the object get modified according to the where τ is the time measured by the FIDO at S (see Sec- reflectivity and compactness of the object. In this pa- tion A). Note that the fields measured by FFOs are fi- per, we rely on the membrane paradigm to obtain generic nite. However, FIDO measurements are Lorentz boosted boundary conditions on the compact object’s surface S. as follows [71], Details of the calculation are presented in Section A 1. r r r r Here, we briefly outline the necessary steps to obtain the EFIDO = EFFO ,BFIDO = BFFO , boundary conditions. Up to the first order of perturba- θ θ φ tion in the axial sector, the normal vector to the stretched EFIDO = γ EFFO − vBFFO , horizon nµ remains unaltered. This indicates that the Bφ = γ Bφ − vEθ , (17) only non-vanishing components of the extrinsic curvature FIDO FFO FFO are δK and δK and their symmetric counterparts (see φ φ 03 23 E = γ E + vBθ , Section A 1 for further discussion). Moreover, the metric FIDO FFO FFO component δgtφ imparts rotation to the compact object. Bθ = γ Bθ + vEφ . Consequently, the fluid at the membrane now has a ve- FIDO FFO FFO φ locity component δu along the azimuth direction while √ √ its other velocity components remain unchanged. By re- where γ = 1/ 1 − v2 = 1/ f is the Lorentz factor. Now φ k placing the expression for δK03, δK23 and δu in Eq. (2) ~ θ φ consider the following term, (ˆn × BFIDO) = −BFIDO. and Eq. (3), we obtain the boundary condition for U(r) For a compact enough object, f(r) is small on the surface at the surface of the compact object as given by the fol- object, hence we can expand v as v = 1−(1/2)f +O(f 2). lowing expression (see Section A 1) ~ k θ Under such consideration, (ˆn × BFIDO) becomes µQf 0 k U (r∗) = H(r) θ φ P (ˆn × BFIDO) = −BFIDO r∗(R) h h θ φ i 3 2 2 ! 2 f r µ + 2 r − 6M + 4Q iω = γ (1 − (1/2)f)EFFO − v(1 + (1/2)f)BFFO + O(f ) − + U (r) 2rPh 16πη θ φ f θ φ 3 r=R = γ E − vB − γ E + B + O(f 2 ) (13) FFO FFO 2 FFO FFO 2 2 θ θ where µ = l(l + 1) − 2, Ph = r(r − 3M) + 2Q and R is = EFIDO − ECor the radius of the compact object. (18) In order to obtain the boundary condition for H(r), we θ θ φ 3 where, E = γ(f/2) E + B + O(f 2 ) is the notice that the membrane paradigm predicts the exis- Cor FFO FFO µ correction term. Here, in the first line we have used the tence of surface 4-current js along the surface of the ob- 2 µ following relation v(1 + (1/2)f + O(f )) ≈ 1. Similarly, ject [71, 88]. The surface current js is related to the field ~ k φ tensor F˜µν by the following expression, we can show that (ˆn×BFIDO) can be written as follows,

1 (ˆn × B~ k )φ = Bθ = Eφ − Eφ (19) jµ = F˜µν n . (14) FIDO FIDO FIDO Cor s 4π ν φ φ θ 3 where, E = γ(f/2) E − B + O(f 2 ). The above expression can be written in terms of the elec- Cor FFO FFO tric and magnetic field observed by the FIDOs as follows, Eq. (18) and Eq. (19) can be written in a more compact form as follows, ⊥ 0 EFIDO = 4πjs = 4πσe (15) ~ k ~ k ~ k k A A EFIDO = (ˆn × BFIDO) + ECor . (20) (B~ ) = 4π(~js × nˆ) . FIDO √ Eq. (15) tells that the normal component of the electric Note that, since γ = 1/ f, the main contribution for ~ k field and the tangential component magnetic field is dis- (ˆn × BFIDO) comes from the first term of Eq. (18) and continuous at the membrane due to the presence of sur- Eq. (19) for a compact enough object. Moreover, in the ~ k face charge and current, respectively, following the laws black hole limit f → 0, ECor vanishes identically. The of electrodynamics. To determine the boundary condi- physical reasoning behind this is as follows, in the black tion on the compact object’s surface, we consider a set of hole limit FIDOs are moving outward at near the speed 5 of light (see Eq. (16)) with respect to any timelike ob- its compactness. Moreover, in the black hole limit, this server and thus they see the tangential component of the term vanishes identically and ρs becomes ρs = ρSBH ≡ electromagnetic fields as incoming waves [71, 88]. For 4π. By redefining the resistivity as ρS = ρs − ρCor, we axial perturbation, the only non-vanishing component of can express the above equation as follows, E~ k is E~ φ FIDO FIDO −4πF˜rφn ρ = r . (23) S σ δuφ Eφ = (ˆn × B~ k )φ + Eφ e FIDO FIDO Cor (21) ˜rφ φ Substituting the expression for F˜rφ, n , σ and δuφ in = −4πF nr + ECor . r e Eq. (23), we obtain the boundary condition for H(r) (or, equivalently for δF ) at the surface of the compact ob- where we have used Eq. (15) in the second line. Moreover, θφ ject as follows, the surface current on the stretched horizon along the az- φ φ imuth direction can be written as follows, j = σeδu . " 2 0 2Q f(r)H (r) The expression for resistivity ρs can be obtained via H (r∗) = − Ohm’s law, r (R) rPh(r) ∗ (24) # µQf(r) iQω (4π − ρS) φ ˜rφ E −4πF nr + U (r) + 2 FIDO (22) Ph(r) 32π ηµr ρs = φ = φ + ρCor , r=R j σeδu Using the transformations outlined in Eq. (9), we can φ φ where ρCor = ECor/j . Note that, the expression for ρCor rewrite the boundary conditions Eq. (13) and Eq. (24) in depends on parameters of the compact object as well as terms of Z1 and Z2 which can be expressed as follows,

2 2 2 " 2 2 2Q f(r) 6Mq1 + µ (l(l + 1)r + 6M)r − 2q2r + 8Q ! dZ1 1 iQ ω 4πµ r − q1(ρS − 4π) = − + Z1 dr D 16π2ηr rP ∗ r∗(R) h 1 2 2 ! # 1 − 2 iq1Qω q1 (4π − ρS) + 4πµ r π rηµl(l + 1)q1Qf(r) − 2 − Z2 D 16µrπ η Ph r=R (25)

" 2 2 ! dZ2 1 iµQω Q (4π − ρS) − πq1r 8µl(l + 1)q1Qr f(r) = − − Z1 dr 8rD π2η P ∗ r∗(R) h 2 2 2 2 2 1 f(r) 2µ Q l(l + 1)r − 18Mr + 2q2r + 8Q + 3Mq1 r(l(l + 1)r − 6M) + 4Q − (26) D rPh ! # iω q 6πMr + Q2ρ − 4πQ2 + 4πµ2Q2r + 1 S Z 16π2ηr 2 r=R

2 2 where D = 6Mq1 + 8µ Q . Note that, when the surface IV. RINGDOWN OF CHARGED COMPACT of the compact object coincides with the radius of the OBJECTS photon sphere rph, the term Ph vanishes and the coef- ﬁcient of Z1 and Z2 diverges in Eq. (25) and Eq. (26). This implies Z1 = Z2 = 0 in this limit regardless the In this section, we discuss the ringdown properties of value of the parameters Q, η, ρS. The same is also true charged compact objects. It is quite well-known that when shear viscosity η of the compact object vanishes. when an astrophysical object gets perturbed, it emits gravitational waves to settle down to a stationary state. According to perturbation theory in static spacetime, the ringdown phase can be described in terms of the superposition of damped sinusoidal waves of the following form P −iωnlt h(t) = nl Anl e Ynl(Ω) where n is the overtone 6

Q=0 Q=0.5 0.6 BH 0.6 BH

-8 ϵ= 10 , ηr=0.5 -8 ϵ= 10 , ηr=0.5, ρr=0.5 -8 0.4 ϵ= 10 , ηr=0.9 0.4 -8 ϵ= 10 , ηr=0.9, ρr=0.5 -4 ϵ= 10 , ηr=0.5 -8 ] ] ϵ= 10 , ηr=0.5, ρr=0.9 0.2 0.2 -4

10M 10 , 0.5, 0.5 10M ϵ= ηr= ρr= = = ,r t, ,r t, [ 0.0 [ 0.0 2 2   Z Z

-0.2 -0.2

-0.4 -0.4

0 50 100 150 200 0 50 100 150 200 t t -8 ϵ= 10 , η=η BH, ρS=ρSBH Q=0.5, η=η BH, ρS=ρSBH

0.6 Q=0.0 0.6 ϵ/ϵ ph=0.1

Q=0.4 ϵ/ϵ ph=0.5

0.4 0.4 Q=0.8 ϵ/ϵ ph=0.99 ] ]

0.2 0.2 10M 10M = = ,r t, ,r t, [ 0.0 [ 0.0 2 2   Z Z

-0.2 -0.2

-0.4 -0.4

0 50 100 150 200 0 50 100 150 200 t t

FIG. 1. A comparative study of the ringdown properties of differnt charged compact objects with different values charge Q, compactness parameter , shear viscosity η and resistivity ρS is presented for l = 2 mode. Here, ηr ≡ 1−η/ηBH, ρr ≡ 1−ρS /ρSBH and ph is the value of compactness parameter when the surface of the compact object coincides with its photon sphere. In the top panels, the ringdown waveforms of highly compact ( . 0.0001) neutral (in the left) and charged (in the right) objects are presented. In this scenario, GW echoes are present. In each of these plots, the black curve represents the ringdown properties of black holes. The bottom left panel shows the ringdown properties for different values of Q, whereas the bottom right panel depicts the same for relatively less compact objects (ph > & 0.1). As evident, the GW echoes are absent for higher values of .

number, Ynl(Ω) is the generalized Legendre polynomial initial conditions, and ωnl are the complex frequencies known as the quasinormal modes (QNMs) whose real part ω denotes the 2 R Zˆ (0, r ) = 0 , ∂ Zˆ (0, r ) = e−(r∗−7) . (27) frequency of the oscillation. In contrast, the imaginary i ∗ t i ∗ part ωI denotes the damping rate [95–98]. These QNMs can be obtained by solving the perturbation equations In Fig. 1, we plot the ringdown waveforms for different of the form of Eq. (11) with appropriate boundary con- compact objects with different values of charge Q, com- ditions. In particular, there are only incoming waves at pactness , shear viscosity η and resistivity ρS. In the the event horizon in the black hole scenario. Moreover, subsequent discussions, we adopt the nomenclature pro- these QNMs are characterized by the mass (M), charge vided in [45, 99] to describe objects with different values (Q) and angular momentum (a) of the black hole in ac- of compactness parameter. Objects with compactness cordance to the no-hair theorem [11–13]. However, as we parameter 0.01 < < ph are termed as Ultra Compact will see in the absence of the event horizon, these modes objects (UCOs) whereas objects with < 0.01 are dubbed also depend on the reflection coefficient and the object’s as Clean-photon sphere objects (ClePhOs) where ph is compactness. In order to study the time domain behavior the value of the compactness parameter when the radius these perturbed objects, we use an inverse Fourier trans- of the object R = re(1+) coincides with the radius of the unstable photon orbit r . The top panel of Fig. 1 shows formation −iωZi(ω, r) → ∂tZî(t, r∗) in Eq. (11), Eq. (25) ph and Eq. (26) and solve these equations with the following the ringdown waveforms of neutral (left) and charged (right) ClePhOs for different values of ηr ≡ 1−η/ηBH and ρr ≡ 1−ρS/ρSBH. In each of these plots, the black curves represent the ringdown of a black hole. As evident, the 7

ϵ= ϵ 150 0.01 ph ϵ=0.2ϵ Q=0 ph ϵ=0.8ϵ Q=0.5 ph Q=0.8 Q=0.99 20 100 echo echo t t

10 50

10 0 1. × 10-8 1. × 10-6 0.0001 0.01 1 0 0 0.2 0.4 0.6 0.8 1 ϵ ϵ / ph Q

FIG. 2. The variation of echo time techo/M as a function of normalized compactness parameter /ph (left panel) and charge Q (right panel) is plotted. In each of these plots, the solid black line represents the typical decay time scale of Reissner-Nordström black holes, td = 1/ωI ≈ 10M. As evident, the echo time decreases with the increase of . The QNMs of the object get modified if td > techo. Clearly, this criteria is satisfied only for the objects with & 0.05ph. ringdown signals of ClePhOs are identical to those of the left panel of Fig. 1, we present the ringdown waveforms black holes in the beginning. However, we obtain repeti- of UCOs (0.01 < < ph). In this scenario, the grav- tive signals in the form of gravitational wave echoes from itational wave echoes are absent. This is due to fact ClePhOs at later times whereas the signals die down for that with the increase of , the size of the cavity be- the black holes. Moreover, the time separation techo (so tween the surface of the compact object and the max- (i) called the “echo time”) between two consecutive signal ima of the perturbation potential Vmax (located at rph in decreases with the increase of . the geometrical-optics approximation) decreases. These cavities are inefficient to effectively trap high frequency The phenomenon can be explained through follow- 2 (i) ing arguments[45, 46, 51, 58]. In geometrical-optics ap- radiations (ω ≈ Vmax). Here, the reflection coefficient proximation, QNMs are interpreted in terms of photons of the compact object plays a crucial role. The reflection coefficient of the object is defined through the asymptotic trapped in unstable photon orbit rph where the perturbation potential has its maxima. The real part of the behavior of Zi(r∗) as follows [46], frequency is related to the angular velocity of the pho- −iωr∗ (i) iωr∗ tons at rph. In contrast, the imaginary part is related to Zi(r∗) ∼ e + R e , r∗ → ∞ . (29) the decay time scale td of null geodesics at the unstable photon orbit [100]. Thus, the radiations generated at rph This simply means that incoming waves from infinity are will reach the surface of the compact object in time. partially reflected back from the compact object’s surface. From now on, we denote R(2) by R for the sake of 2 rph dr simplicity. We obtain the reflection coefficients |R| by ∆t = solving the perturbation equations Eq. (11) with bound- ˆ f(r) R ary conditions Eq. (25) and Eq. (26) using Numerov al- 1 1 + ε gorithm [101]. = re(ph − ) − log[ ] + log[ ] , 2κe ph 2κc ph + ε (28) In the top left panel of Fig. 3, we present the reflec- where ph = (rph −re)/re and ε = (re −rc)/re. Part of ra- tion coefficient as a function of frequency for different diations get reflected from the surface of object and give values of . Here we can see that highly compact objects rise to the gravitational wave echoes in time techo = 2∆t with η = ηBH, ρS = ρSBH and smaller values of Q acts [45, 51, 61]. In the left panel of Fig. 2, we plot the as a perfect absorber of high-frequency waves. However, variation of techo with for different values of Q. As for less compact objects, this is no longer true. With evident, techo decreases with the increase of which is the increase of , the value of the reflection coefficient of in accordance to the behavior of the gravitational wave high-frequency waves increases. Moreover, as the surface echoes presented in the top panels of Fig. 1. The effect of the compact object coincides with the photon sphere of charge on the ringdown properties of ClePhOs is pre- radius rph, the object behaves like a perfect reflector re- sented in the bottom left panel of Fig. 1. As we can gardless of its frequency (in this limit, the coefficient of see that the techo increases with the increase of Q. This Z1 and Z2 in Eq. (25) and Eq. (26) vanishes which re- phenomena can be explained by following the same argu- sults in reflecting boundary conditions Z1 = Z2 = 0). 2 (i) ment described above. The plot of techo versus Q in the Hence the high-frequency waves (ω ≈ Vmax) can not be right panel of Fig. 2 confirms the same. In the bottom effectively trapped inside the cavity between the surface 8

Q=0.5, l=2, η=η BH, ρS=ρ SBH ϵ=0.1, l=2, η=η BH, ρS=ρ SBH

1.0 1.0

0.8 0.8

ϵ=0.2ϵ ph 0.6 0.6 Q=0.0

2 ϵ=0.4ϵ ph 2 Q=0.4 | ℛ | ℛ ϵ=0.6ϵ ph Q=0.6 ϵ=0.8ϵ ph 0.4 0.4 Q=0.8 ϵ=0.999ϵ ph

0.2 0.2

0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 ω ω

ϵ=0.5ϵph ,Q=0.5, l=2, ρS=ρ SBH ϵ=0.5ϵph ,Q=0.5, l=2, η=η BH

1.0 1.0

0.8 0.8

ηr=0.0

0.6 ηr=0.2 0.6 ρr=0.0 2 2

ηr=0.4 ρr=0.4 | ℛ | ℛ

ηr=0.6 ρr=0.8 0.4 =0.8 0.4 ηr ρr=1.0

ηr=1.0

0.2 0.2

0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 ω ω

FIG. 3. The variation of reflection coefficient |R|2 as a function of ω for different values of compactness parameter (top left), charge Q (top right), shear viscosity ηr ≡ 1 − η/ηBH (bottom left) and resistivity ρr ≡ 1 − ρS /ρSBH (bottom right) is presented. Here, ph is the value of compactness parameter when the surface of the compact object coincides with its photon sphere. of the compact object and the maxima of the perturba- over, from the bottom right panel of Fig. 3, we can see (i) tion potential Vmax which explains the absence of echo that the reflection coefficient has a very weak dependence in this limit. In the top right panel of Fig. 3, we present on ρr ≡ 1−ρS/ρSBH for η = ηBH. Thus the amplitude of the reflection coefficient as a function of frequency for the signal in Fig. 1 remains practically unaltered with the different charge values. As evident, a neutral compact change of ρr. In order to understand this phenomena, we object with η = ηBH and ρS = ρSBH acts a perfect ab- calculate the reflection coefficient of the high frequency sorber of high frequency waves. However, in the presence waves (Mω 1). In this limit, we can neglect the poten- of the charge parameter, the reflection coefficient of the tial term Vi(r) in Eq. (11). Similarly, we only retain the compact object is non-vanishing. terms that are proportional to ω in Eq. (25) and Eq. (26). Moreover, for the sake of simplicity, we consider compact From the top panel of Fig. 1, we find that the ampli- objects with small charge parameter (Q M). Under tude of the ringdown signal increases with the increase such consideration, the reflection coefficient turns out to of the parameter ηr ≡ 1 − η/ηBH. The phenomenon can be be understood by looking at the reflection coefficient’s dependence on η as presented in the bottom left panel r 1 − η/η 2 8Q (1 − η/η ) of Fig. 3. The object behaves like a perfect absorber of |R|2 = BH − BH 1 + η/η 3 (1 + η/η )2 high-frequency waves for smaller values of ηr (η ≈ ηBH). BH BH 2 2 But with the increase of ηr, the value of |R| increases. 2Q 3(ρS − ρSBH)(1 − η/ηBH) + (16πR)η/ηBH + 3 When the value of ηr approaches unity (i.e. for η → 0), 9πR (1 + η/ηBH) the object becomes a perfect reflector (this is due to the + O(Q3) . fact that in the limit η → 0, the coefficients of the Z1 and Z diverges in Eq. (25) and Eq. (26) and thus we obtain (30) 2 Note that, in the Q → 0 limit, the expression for re- reflecting boundary conditions Z1 = Z2 = 0). This explains the dependence of the ringdown signal on η. More- flection coefficient exactly coincides with [46]. Even in the presence of a small charge Q, the dominant contri- 9

η=η BH,ρ S=ρ SBH, l=2 η=η BH,ρ S=ρ SBH, l=2 1.20 = Q=0 Q 0 2.5

= Q=0.4 1.15 Q 0.4

= Q=0.6 Q 0.6 2.0 1.10 Q=0.8 Q=0.8 BH BH I

R 1.5 1.05 / ω / ω I R ω ω

1.00 1.0

0.95 0.5

0.90 0.001 0.010 0.100 1 0.001 0.010 0.100 1 ϵ ϵ ϵ ϵ / ph / ph

BH BH FIG. 4. The variation of the quantities ωR/ωR (left panel) and ωI/ωI (right panel) which represents the ratio of the fundamental quasinormal mode (l = 2 mode) of a charged compact object with those obtained for charged black holes with respect to the compactness parameter for diﬀerent values of charge Q is presented. The shear viscosity and resistivity is taken as η = ηBH = 1/16π and ρS = ρSBH = 4π. As evident, for smaller values of the compactness parameter ( . 0.01), the quasinormal modes of the compact object remains comparable to the black hole case. However, for less compact objects ( & 0.01), quasinormal modes get modiﬁed.

bution in the reflection coefficient comes from the first that the solutions at r0 are a superposition of incoming term. The contribution of ρS comes only in the order and outgoing waves. Similarly, near the asymptotic in- 2 O(Q ). Moreover, note that, in the limit η → ηBH, the finity, we use the following power series. reflection coefficient is independent of ρS. This explains N (i) the behaviour in Fig. 3. X bn Z (r) = ekrrp , (32) i rn n=0 where, p = iω(1/κ − 1/κ ). Note that, there exist two V. QUASINORMAL NORMAL SPECTRA OF e c CHARGED COMPACT OBJECTS independent solutions at the infinity corresponding to k = ±iω. This give us the behavior of Zi at large values of r = rinf. We obtain the quasinormal modes by The quasinormal modes of the object are the eigenval- numerically integrating Eq. (11) from r0 to rinf and then ues of Eq. (11) with the folowing boundary conditions comparing them with Eq. (32) with QNM boundary con- that there are only outgoing modes (Zi ≈ exp(iωr∗)) at dition that there are only outgoing solution at rinf. the asymptotic region r∗ → ∞ while it satisfies Eq. (25) In Fig. 4, we present the variation of the real (left and Eq. (26) at the surface of the star. Here, we employ panel) and imaginary part (right panel) of fundamen- the so-called “shooting method” to obtain the quasinor- tal quasinormal mode (l = 2 mode) as a function of the mal spectrum (see [98, 102, 103] for a detailed discussion compactness parameter for different values of electric of this method). The basic idea behind this method is charge Q. Here, we scale the QNMs of the compact ob- to integrate the perturbation equation numerically from ject with η = ηBH and ρS = ρSBH relative to the value a point r0 = R(1 + δ) (where, δ << 1) close to the sur- obtained in black hole scenario (the QNMs of Reissner- face of the compact object to infinity, where we impose Nordströmblack holes are presented in Table. I). As the outgoing boundary condition. In order to find the evident from the figure, the QNM spectra of the Cle- behaviour of the perturbation function at r0, we use the PhOs (corresponding to . 0.01) do not differ much following ansatz. from that of the black holes. However, for less compact N objects ( & 0.01), we observe significant modification in iω iω X − 2κ 2κ (i) n the spectrum. The phenomena can be explained through Zi(r) = (r − re) e (r − rc) c an (r − R) , (31) n=0 causality arguments. As discussed in the previous section, in the geometrical-optics approximation, the insta- where, κe and κc denote the surface gravity at the posi- bility time scale td of photon trajectories at the photon tion of the event, and the Cauchy horizon, respectively sphere governs the damping rate of the prompt ringdown and N is an arbitrary finite integer. Using the perturba- signal. For Reissner-Nordströmblack holes, the typical tion equations Eq. (11) along with the boundary condi- value of td is ≈ 10M. On the other hand, starting from tions Eq. (25) and Eq. (26), we can write the coefficients the photon orbit, null geodesics completes a round trip (i) (i) an in terms of a0 . The boundary conditions dictate between the photon sphere and the surface of the com- 10

ϵ=0.001, ρS =ρ SBH , l =2 ϵ=0.001, ρS =ρ SBH , l =2 1.2 1.2 Q=0

Q=0.4 1.15 1. Q=0.6

Q=0.8 1.1 0.8 BH BH R I / ω / ω I R Q=0 ω ω 1.05 0.6 Q=0.4

Q=0.6 1 0.4 Q=0.8

0.95 -7 0.2 1. × 10 0.00001 0.001 0.1 1 1. × 10-7 0.00001 0.001 0.1 1 1-η/ηBH 1-η/ηBH

BH BH FIG. 5. The variation of the quantities ωR/ωR (left panel) and ωI/ωI (right panel) which represents the ratio of the fundamental quasinormal mode (l = 2 mode) of a charged compact object with those obtained for charged black holes with respect to ηr ≡ 1 − η/ηBH is plotted for diﬀerent values of charge Q. The compactness parameter and the resistivity of the object is taken as = 0.001 and ρS = ρSBH = 4π.

BH BH pact object in time techo (see Eq. (28)). Thus, the bound- The dependence of ωR/ωR (left panel) and ωI/ωI ary will have a impact on the QNM spectrum only when (right panel) on the parameter ρr = 1 − ρS/ρSBH is pre- techo < td [46]. From Fig. 2, we can see that the techo sented in Fig. 6 for different values of charge Q. As is larger than 10M (represented by black horizontal line) expected, for neutral compact objects, this parameter for highly compact objects. Thus, the boundary does not is irrelevant as can be seen from the figure. However, modify the QNM spectrum for these objects. However, for charged objects, the QNM spectrum get modified for for the less compact objects, ( & 0.01), we can easily higher values of ρr. For real part of the QNM, the mod- check that techo < td and hence the boundary conditions ification is less than 0.5%. However, the imaginary part do modify the spectrum significantly. gets significant modification in the limit ρr → 1. In general, the ringdown waveform of a compact object can be described as the addition of a prompt ringdown signal (primary pulse) and a series of echo pulses [48,104]. VI. QUASINORMAL MODE IN THE NEAR The pulses are separated by time techo (or equivalently EXTREMAL LIMIT by a phase factor iωtecho). If the time separations between the pulses are larger than its width (this happens for highly compact objects, see Fig. 2), there is no inter- In this section, we discuss about the quasinor- ference, and the total ringdown signal appears to be a mal modes of the Reissner-Nordströmblack holes and sum of these individual pulses, as can be seen in the top charged compact objects in the near etremal limit Q → panel of Fig. 1 [48]. In this scenario, the dominant QNM M. Following [25, 105, 106], we use the following coordi- is practically indistinguishable from the black hole QNM nate transformation, [48, 104]. However, for a less compact object (techo is small) where the separation between the echoes is smaller p τ than the pulse width, there is no distinct echo pulse. Due r → Q + ε% , M → Q2 + ε2B2 , t → . (33) ε to inference between the echo signals, the total ECO signal becomes a single damped sinusoid with a frequency different from that of a black hole [48]. Here, B represents the deviation from the extremity, and ε is a small parameter. Under this transformation, the In Fig. 5, we present the variation of the quantities line element Eq. (4) can be rewritten as follows, BH BH ωR/ωR (left panel) and ωI/ωI (right panel) as a function of the parameter ηr = 1 − η/ηBH for different values 1 ds2 = −f(%) dτ 2+ d%2+Q2 (dθ2+sin2 θ dφ2) , (34) of charge Q. Here, we consider a object with compactness f(%) = 0.0001 and resistivity ρS = ρSBH. As evident from Fig. 5, when ηr . 0.01, the QNM spectum of the compact where f(%) = (%2 −B2)/Q2. The event horizon is located object remains practically indistinguishable from that of at % = B. We consider that the surface of the compact the black hole. However, we observe a significant modifi- object is located at %R = B(1+). By using the transfor- cation in the spectrum when the value of ηr approaches mation Eq. (33) and introducing the tortoise coordinate to unity. %∗ = log[(%−B)/(%+B)]/2κ, we rewrite the perturbation 11

ϵ= 0.001, η=η , l=2 ϵ= 0.001, η=η BH, l=2 BH 1.010 1.1 Q=0

Q=0.4 1.0 1.005 Q=0.6

Q=0.8 0.9 BH BH R I 1.000 / ω / ω

I = R Q 0 ω ω 0.8 Q=0.4

0.995 Q=0.6 0.7 Q=0.8

0.990 0.6 0.001 0.005 0.010 0.050 0.100 0.500 1 0.001 0.005 0.010 0.050 0.100 0.500 1 1-ρ /ρ S SBH 1-ρS/ρSBH

BH BH FIG. 6. The variation of the quantities ωR/ωR (left panel) and ωI/ωI (right panel) which represents the ratio of the fundamental quasinormal mode (l = 2 mode) of a charged compact object with those obtained for charged black holes with respect to ρr ≡ 1 − ρS /ρSBH is shown for diﬀerent values of charge Q. The compactness parameter and the shear viscosity of the object is taken as = 0.001 and η = ηBH = 1/16π.

equations Eq. (11) for Z1 and Z2 as follows, the solution becomes

2 in −iw%∗ out iw%∗ d Zi(%∗) 2 Zi(x) = Ai e + Bi e (38) 2 + w − Vi(%∗) Zi(%∗) = 0 , d%∗ (35) in out 2 where, A = AiCi and B = BiDi, and σi(σi + 1)κ i i Vi(%∗) = sinh2 (κ% ) ∗ − iw 2aibi Ci = 2 2κ (1 + ) , 2 c where, w = ω/ε, κ = B/Q , σ1 = (l+1), and σ2 = (l−1). (39) 2 iw 2(1 + ai − c)(1 + bi − c) Here, we consider terms upto O(ε ). Moreover, we can Di = (−2) 2κ 1 + . write Eq. (35) as, 2 − c

d2Z (x) 3x dZ (x) Here, we used the following property of the hypergeo- x(x − 1) i + (1 − ) i dx2 2 dx metric function F1(a, b, c; z) = 1 + abz/c. Similarly, near (36) the asymptotic boundary % → ∞ (% → 0, x → 1), the w2 σ (σ + 1) ∗ + − i i Z (x) = 0 solution becomes 4κ2x 4(1 − x) i in − σi Ai − iw 2 Z (x) = (1 − x) 2 x 2κ F (a , b , c; 1) by introducing a new variable x = 1/ cosh (κ%∗). The i C 1 i i solution of this equation can be written in terms of hy- i out Bi iw pergeometric functions as follows, + (−x) 2κ F (1 + a − c, 1 + b − c, 2 − c; 1) . D 1 i i i − σi − iw (40) Z (x) = (1 − x) 2 A x 2κ F (a , b , c; x) i i 1 i i Demanding that there is only outgoing solution at the (37) asymptotic boundary, the coeﬃcient of Ain should van- iw i + Bi (−x) 2κ F1(1 + ai − c, 1 + bi − c, 2 − c; x) ish. Making use of the following property of hypergeo- metric function, where, Γ(c)Γ(c − a − b) F (a, b, c; 1) = , iw σ 1 iw σ iw 1 (41) a = − − i + , b = − − i , c = 1 − . Γ(c − a)Γ(c − b) i 2κ 2 2 i 2κ 2 κ we obtain the following conditions, c−ai = −n or c−bi = Similarly, we can write the boundary conditions Eq. (25) −n, where n is an integer. Substituting the values of ai, and Eq. (26) in terms of the variable x. In this section, bi, and c, we obtain the expression for quasinormal modes we are interested in ﬁnding the quasinormal spectrum as of highly compact objects for which the surface of the i compact object is located at %∗(%R) ≈ log[]/2κ 0 wn = −i(n + σi + 1)κ (42) (xR ≈ 2). In this scenario, the exact expression of the boundary conditions is not needed to obtain the quasi- Note that, Ref. [25, 105, 106] found the same expres- normal modes. Near the surface of the compact object, sion of quasinormal modes for near extremal Reissner- 12

Nordstr¨omblack holes 1. Thus, even in Q → M limit, 1 the quasinormal spectrum of a highly compact object re- Q=0.0 mains the same as the Reissner-Nordstr¨omblack holes, 0.8 which reassures our previous results. Q=0.4 Q=0.8

BH 0.6 SNR VII. DETECTABILITY OF GW ECHOES / 0.4 SNR

One interesting feature about the perturbation of 0.2 Reissner-Nordströmgeometry is that Maxwell’s mode H(r) appears in the quasinormal spectrum along with 0 the gravitational modes U(r)[7, 93, 94]. As discussed 0. 0.2 0.4 0.6 0.8 1 in Section III, the perturbation functions H(r) and U(r) 1-η/ηBH form a system of coupled differential equations (which are FIG. 7. The SNR for echo signal relative to prompt ringdown related to master function Z1(r) and Z2(r) via Eq. (9)). is plotted as function of ηr ≡ 1 − η/ηBH for different values Thus, excitation of one mode will inevitably excite the −8 of Q. Here, we take = 10 and ρS = ρSBH. The black other. horizontal line in the plot represents the detection threshold, The energy flux at infinity in the Maxwell and gravita- corresponding to SNR= 10. We assumed that the SNR for 2 2 tional sector is proportional to |H(r)| and |U(r)| re- prompt ringdown is 24.4, similar to event GW150914. spectively for each multipole [77]. Hence, we calculate gravitational perturbation function U(r) using Eq. (9), Eq. (11), Eq. (25) and Eq. (26) and discuss about the rel- for which detection is possible, gets widened in presence ative strength of echo signal as compared to the prompt of charge parameter (ηr & 0.7 for Q = 0.4 and Q = 0.8). ringdown signal of black holes. For this purpose, we However, the SNR has a very weak dependence on the subdivide the ringdown waveform of compact objects as compactness parameter and resistivity. U(t) = U BH(t) + U echo(t), where U BH(t) represents the prompt ringdown signal whereas U echo(t) is the echo signal [107]. Here, we adopt matched filter method to calculate the signal to noise ratio (SNR), the expression of VIII. CONCLUSION which can be represented as follows [108–110] s ∞ |U˜(f)|2 The rapid developments of GW detectors made it pos- SNR = 4 df , (43) sible to detect more and more events with a large signal- ˆ0 Sn(f) to-noise ratio, and we are hopeful that GW spectroscopy where, U˜(f) = ∞ exp(2πift)U(t)dt is the Fourier will eventually enable us to probe physics in most ex- −∞ treme conditions, from the strong gravity regime [14– transformation of´ the strain U(t) and Sn(f) is the one- sided power spectral density of the detector. For our 16] to the very early universe [116, 117]. It also pave study, we scale the ringdown signal to have the same am- the way to test different predictions of GR (e.g. the no-hair theorem)[118] and to find quantum mechanical plitude as GW150914 [111, 112]. We obtain Sn(f) from the estimated noise curve during O1 [113, 114]. The re- effects near the horizon [47, 119–121]. Recently, it has sult is presented in Fig. 7 where we plotted the relative been noted that GW spectroscopy can also help us de- SNR of the echo signal to prompt ringdown as a function tect the existence of horizon as the ringdown properties of η for different values of Q. Here, we choose η = 10−8 of black holes generally differs from horizonless compact r objects [45–51, 58]. This can be seen from Fig. 1 where and ρS = ρSBH. In the plot, the black horizontal line represents the threshold of detection, corresponding to we have compared the ringdown properties of charged SNR= 10. Moreover, we have assumed that the SNR for non-rotating compact objects and black holes. We ob- the prompt ringdown is 24.4 similar to event GW150914 tain a repetitive signal for (at least!) extremely com- [109, 115]. As can be seen from the plot, for an uncharged pact objects at later times, which is in contrast to the configuration, LIGO detectors will discover GW echo for black hole scenario where there are decaying power-law η 0.8 (η 0.2η ). However, the parameter range tails [54–57]. The perturbation theory very accurately r & . BH describes this ringdown phase. However, first, we have to set up an appropriate boundary to study the response of these objects to external perturbations, which is ex- 1 Ref. [25, 105, 106] calculates the quasinormal mode of near extremely tricky since the internal structure of these objects tremal Reissner-Nordström black holes under scalar field pertur- are largely unknown [48–50]. Following [46], we have em- bation where the perturbation equation has the same form as ployed a membrane paradigm to obtain the quasinormal Eq. (35) with σi = l (see [106]). spectrum of these objects in a model-independent way. 13

The paradigm tells us that the surface of the compact tively. However, in the limit Q → M, the quasinormal object essentially behaves like a viscous fluid membrane modes of highly compact objects become independent to an outside observer [70–73]. We have shown that the of the shear viscosity and resistivity. The quasinormal perturbations of these objects in the odd party sector modes become purely imaginary in this scenario, simi- depend on the compactness , the shear viscosity η and lar to the near-extremal Reissner-Nordströmblack holes resistivity ρS of the membrane along with its mass and [25, 105, 106]. charge. Our study shows that both ringdown waveform and Moreover, we have shown that both the compactness quasinormal modes carry information about the nature parameter and charge play a significant role in determin- of the object. In particular, we have shown that gravi- ing the reflection coefficient of the object. A object with tational wave echoes are present in the case of ClePhOs η = ηBH and ρS = ρSBH behaves like a perfect absorber of high frequency waves for smaller values of and Q. (objects with compactness parameter . 0.01). How- ever, the echo time t decreases with the increase of However, with the increase of these two parameters, the echo reflection coefficient of the object is non-vanishing. In . Finally, for UCOs (objects with 0.01 . . ph), either the echos are absent, or the waveform is a super- particular, when the surface of the compact object co- position of prompt ringdown and the first echo. The incides with the photon sphere radius, it behaves like a echo time increases with the increase of charge parame- perfect reflector. This behaviour is also true in the limit ter Q. However, the effect is most significant in the limit η → 0. Q ≈ M. The shear viscosity η of the membrane controls We discuss the detectability of the echo signal in Sec- the modulation of the echos. The amplitude of the ring- tion VII. Our study reveals that the LIGO detectors will down signal decreases with the increase of η. Lastly, the detect the echo signal from an uncharged compact object resistivity ρ has a small effect on the ringdown signal. S for the parameter range η ∈ (0.8, ηBH). Interestingly, the The phenomena can be explained by studying the optical parameter range for which the detection is possible in- properties of the charged membrane, where the electro- creases for a charged compact object. The SNR has a magnetic perturbation can be treated as the incidence of weak dependence on the compactness parameter and the an electromagnetic field on a charged sphere. When the resistivity. electromagnetic field passes through a medium, it inter- acts with the internal charge and current density. Hence, It is straight forward to extend the calculation in it give rise to complex frequency-dependent optical con- even parity sector. However, in this scenario, the non- ~ k ~ θ stants, like complex refractive index N = n + ik, skin vanishing component of EFIDO is EFIDO (see Eq. (21) and depth δ, attenuation coefficient α, and complex conduc- Eq. (22)). Moreover, there will be a surface current along tivity function σ [122]. Here, n and k represent the or- θ direction, jθ as the membrane has a velocity compo- dinary refractive index and extinction coefficient, respec- nent δuθ along this direction (see the Appendix of [46]). tively. Both of these parameters depend on the complex Furthermore, as discussed in [46], the ringdown proper- conductivity function. Moreover, for normal incidence ties will depend on an additional parameter namely on of the electromagnetic wave, the reflection coefficient be- its the bulk viscosity ζ in this scenario. comes |R|2 = ((n − 1)2 + k2)/((n + 1)2 + k2) which also depends on the conductivity. In our opinion, an analog- Since our analysis is quite generic, it can describe the ical situation can be found for the membrane paradigm ringdown properties of a large variety exotic compact ob- too. However, solving Maxwell’s equation in our case is jects [28–32, 36–38, 42–44, 79–81]. A possible extension quite cumbersome, and for the moment, we left it for of our study is to analyze the ringdown properties of mag- future work to make a more concrete connection. netic black holes [75] and compact objects like Magnetars [76]. It is well known that electrically charged astrophys- For larger values of the compactness parameter, the ical objects get neutralized rather quickly by accreting QNMs could be a distinguisher between a black hole and ionized plasma and other quantum mechanical processes a compact object. This can be seen from Fig. 4. For like Schwinger’s pair production and Hawking radiation small values of compactness ( . 0.01) and charge pa- [77, 78]. However, magnetic black holes are relatively rameter (Q), the quasinormal spectrum of these objects long-lived (see [78] for further discussion). Moreover, is practically indistinguishable from those charged black in certain parameter spaces, the magnetic field near the holes. The interior of the object will have a significant horizon could be quite large so that it can restore elec- effect on the QNM spectrum if techo is less than the decay troweak symmetry. Another possible extension is to con- time scale td of a photon at the photon orbit [46] which sider the effect of spin. However, these are beyond the is not the case in the above-mentioned scenario. How- scope of this paper. ever, with the increase of compactness parameter, techo becomes less than td and we notice a modification in the QNM spectrum. Similarly, we show that the QNM spectrum gets modified as the value of the shear viscosity η and resistivity ρ shifts away from ηBH and ρSBH respec- 14

ACKNOWLEDGEMENTS as follows, 1 K = − pf(r)f 0(r) We thank Kabir Chakravarti for useful discus- tt 2 (A3) sion. M.R is supported by the postdoctoral fel- Kφφ p Kθθ = = r f(r) . lowship (MIS/IITGN/PD-SCH/201415-006) by IIT- sin2 θ Gandhinagar. A.B. is supported by Start Up Research Grant (SRG/2020/001380) by Department of Science & Replacing these expressions in Eq. (2) and Eq. (3), we Technology Science and Engineering Research Board (In- obtained the expression for density and pressure as given dia). in Eq. (7).

1. First Boundary condition at the surface of the Appendix A: Boundary condition at the surface of object the object

Upon perturbation, the only non-vanishing compo- We assume that at each point of the timelike stretched nents of δgµν in the axial sector are δgtφ and δgrφ as horizon S; there is a ﬁducial observer (FIDO) moving given by Eq. (8). As discussed in Section III, upto the with four-velocity U µ. The observer is at rest with re- ﬁrst order of perturbations, the components of nµ respect to the compact object and making the measure- mains unchanged. Thus the non vanishing components ments in his local Cartesian coordinate system. of the δKµν becomes

p 1 eˆ = ∂ = α−1∂ , eˆ = f∂ δK = sin θe−iωtpf (h0 + iωh ) ∂ P (cos θ) t τ t r r tφ 2 0 1 θ l 1 1 (A1) eˆθ = ∂θ , eˆφ = ∂φ , 1 −itωp r r sin θ δKθφ = h1(r) sin(θ)e f 2 cot(θ)∂θPl(cos θ) 2 √ where, α = f is called the lapse function. The normal + l(l + 1)Pl(cos θ) . µ µ to the strechted horizon n is a spacelike i.e. nµn = 1. For the spacetime given by Eq. (4), the normal to S and (A4) the four velocities of the FIDO can be represented as follows, whereas the 4-velocity of the FIDO takes the following form

√ 1 µ 1 φ nµ = (0, f, 0, 0) ,U µ = (√ , 0, 0, 0) . (A2) U = (√ , 0, 0, δu ) . (A5) f f Substituting Eq. (A4) and Eq. (8) in Eq. (2), we obtain Using Eq. (A2), we can obtain the expression for extrinsic the expression for nonvanishing component of the energy a c curvature from the relation K b = h bna;c which is given momentum tensor of the membrane as follows,

−iωt 0 0 sin(θ)e ∂θPl(cos θ)(rh0(r)f (r) + f(r)(−irωh1(r) − rh0(r) + 4h0(r))) δτtφ = 16πrpf(r) (A6) −itω h1(r) sin(θ)e (2 cot(θ)∂θPl(cos θ) + l(l + 1)Pl(cos θ)) δτθφ = − . 16πpf(r)

Again by substituting Eq. (A5) and Eq. (8) in Eq. (3), rameters as follows, we obtain another expression for the energy-momentum 2 φp tensor of the membrane in terms of dissipative ﬂuid pa- δτtφ = sin(θ) − R δu f(R) sin(θ)(P + ρ) −itω (A7) − ρh0(r)e ∂θPl(cos θ)

2 2 φ δτθφ = η −R sin θ ∂θδu . 15

By equating tφ component of Eq. (A6) and Eq. (A7), we Replacing Eq. (A8) in θφ component of Eq. (A7) and obtain the expression for δuφ as equating it with the θφ component of Eq. (A6), we obtain the following expression,

−h (r)f 0(r) + f(r)(h0 (r) + iωh (r)) h (r) 0 0 1 = − 1 (A9) f(r)(rf 0(r) − 2f(r)) 16πηr

By substituting h0 = (if(r)/ω)∂r(h1f(r)) and redeﬁning e−iωt∂ P (cos θ)(−h f 0 + f (h0 + iωh )) the function h1(r) = (2iωrU(r))/µf(r) and δFθφ = l(l + δuφ = θ l √ 0 0 1 (A8) 1)H(r), we obtain the boundary condition for U(r ) r sin(θ) f (rf 0 − 2f) ∗

2 2 ! 0 µQf(r)H (r) f(r) r µ + 2 r − 6M + 4Q iω U (r∗) = − + U (r) (A10) P (r) 2rP (r) 16πη r∗(R) h h r=R

Note that, this equation coincides with Eq. (13).

[1] R. Narayan, “Black holes in astrophysics,” New J. Astronomische Nachrichten 296 no. 1, (1975) 45–46, Phys. 7 (2005) 199, arXiv:gr-qc/0506078. https://onlinelibrary.wiley.com/doi/pdf/10.1002/asna.19752960110. [2] A. Mueller, “Experimental Evidence of Black Holes,” https://onlinelibrary.wiley.com/doi/abs/10. PoS P 2GC (2006) 017, arXiv:astro-ph/0701228. 1002/asna.19752960110. [3] R. M. Wald, General Relativity. Chicago Univ. Pr., [12] W. Israel, “Event horizons in static vacuum Chicago, USA, 1984. space-times,” Phys. Rev. 164 (Dec, 1967) 1776–1779. [4] E. Poisson, A Relativist’s Toolkit: The Mathematics of https: Black-Hole Mechanics. Cambridge University Press, //link.aps.org/doi/10.1103/PhysRev.164.1776. 12, 2009. [13] B. Carter, “Axisymmetric black hole has only two [5] T. Regge and J. A. Wheeler, “Stability of a degrees of freedom,” Phys. Rev. Lett. 26 (Feb, 1971) schwarzschild singularity,” Phys. Rev. 108 (Nov, 1957) 331–333. https: 1063–1069. https: //link.aps.org/doi/10.1103/PhysRevLett.26.331. //link.aps.org/doi/10.1103/PhysRev.108.1063. [14] E. Berti et al., “Testing General Relativity with [6] S. Klainerman and J. Szeftel, “Global Nonlinear Present and Future Astrophysical Observations,” Stability of Schwarzschild Spacetime under Polarized Class. Quant. Grav. 32 (2015) 243001, Perturbations,” arXiv:1711.07597 [gr-qc]. arXiv:1501.07274 [gr-qc]. [7] F. J. Zerilli, “Perturbation analysis for gravitational [15] LIGO Scientific, Virgo Collaboration, B. P. Abbott and electromagnetic radiation in a reissner-nordström et al., “Tests of general relativity with GW150914,” geometry,” Phys. Rev. D 9 (Feb, 1974) 860–868. Phys. Rev. Lett. 116 no. 22, (2016) 221101, https: arXiv:1602.03841 [gr-qc]. [Erratum: Phys.Rev.Lett. //link.aps.org/doi/10.1103/PhysRevD.9.860. 121, 129902 (2018)]. [8] B. S. Kay and R. M. Wald, “Linear stability of [16] N. Yunes, K. Yagi, and F. Pretorius, “Theoretical schwarzschild under perturbations which are Physics Implications of the Binary Black-Hole Mergers non-vanishing on the bifurcation 2-sphere,” Classical GW150914 and GW151226,” Phys. Rev. D 94 no. 8, and Quantum Gravity 4 no. 4, (Jul, 1987) 893–898. (2016) 084002, arXiv:1603.08955 [gr-qc]. https://doi.org/10.1088/0264-9381/4/4/022. [17] LIGO Scientific, Virgo Collaboration, B. P. Abbott [9] M. Dafermos and I. Rodnianski, “A Proof of the et al., “Observation of Gravitational Waves from a uniform boundedness of solutions to the wave equation Binary Black Hole Merger,” Phys. Rev. Lett. 116 on slowly rotating Kerr backgrounds,” Invent. Math. no. 6, (2016) 061102, arXiv:1602.03837 [gr-qc]. 185 (2011) 467–559, arXiv:0805.4309 [gr-qc]. [18] LIGO Scientific, Virgo Collaboration, B. P. Abbott [10] M. Casals and L. F. Longo Micchi, “Spectroscopy of et al., “Tests of General Relativity with the Binary extremal and near-extremal Kerr black holes,” Phys. Black Hole Signals from the LIGO-Virgo Catalog Rev. D 99 no. 8, (2019) 084047, arXiv:1901.04586 GWTC-1,” Phys. Rev. D 100 no. 10, (2019) 104036, [gr-qc]. arXiv:1903.04467 [gr-qc]. [11] I. J. Tredcr, “C.w. misner, k.s. thorne, j.a. wheeler: [19] Event Horizon Telescope Collaboration, Gravitation. w.h. freeman and company limited, K. Akiyama et al., “First M87 Event Horizon reading (england) 1973 — xxvi + 1279 seiten, preis £ Telescope Results. I. The Shadow of the Supermassive 19.20 (clothbound); £ 8.60 (paperbound),” Black Hole,” Astrophys. J. Lett. 875 (2019) L1, 16

arXiv:1906.11238 [astro-ph.GA]. and Condensate Stars,” J. Phys. Conf. Ser. 314 [20] R. Penrose, Singularities of spacetime., pp. 217–243. (2011) 012010, arXiv:1107.5086 [gr-qc]. 1978. [41] H. Kawai and Y. Yokokura, “A Model of Black Hole [21] H. Reall, “Viewpoint: A possible failure of Evaporation and 4D Weyl Anomaly,” Universe 3 determinism in general relativity,” Physics 11 (2018) . no. 2, (2017) 51, arXiv:1701.03455 [hep-th]. [22] M. Dafermos, “The Interior of charged black holes and [42] R. Brito, V. Cardoso, and H. Okawa, “Accretion of the problem of uniqueness in general relativity,” dark matter by stars,” Phys. Rev. Lett. 115 no. 11, Commun. Pure Appl. Math. 58 (2005) 0445–0504, (2015) 111301, arXiv:1508.04773 [gr-qc]. arXiv:gr-qc/0307013. [43] S. L. Liebling and C. Palenzuela, “Dynamical Boson [23] V. Cardoso, J. a. L. Costa, K. Destounis, P. Hintz, and Stars,” Living Rev. Rel. 15 (2012) 6, arXiv:1202.5809 A. Jansen, “Quasinormal modes and Strong Cosmic [gr-qc]. Censorship,” Phys. Rev. Lett. 120 no. 3, (2018) [44] E. Seidel and W.-M. Suen, “Formation of solitonic 031103, arXiv:1711.10502 [gr-qc]. stars through gravitational cooling,” Phys. Rev. Lett. [24] O. J. C. Dias, F. C. Eperon, H. S. Reall, and J. E. 72 (1994) 2516–2519, arXiv:gr-qc/9309015. Santos, “Strong cosmic censorship in de Sitter space,” [45] V. Cardoso and P. Pani, “Testing the nature of dark Phys. Rev. D 97 no. 10, (2018) 104060, compact objects: a status report,” Living Rev. Rel. 22 arXiv:1801.09694 [gr-qc]. no. 1, (2019) 4, arXiv:1904.05363 [gr-qc]. [25] M. Rahman, S. Chakraborty, S. SenGupta, and A. A. [46] E. Maggio, L. Buoninfante, A. Mazumdar, and Sen, “Fate of Strong Cosmic Censorship Conjecture in P. Pani, “How does a dark compact object Presence of Higher Spacetime Dimensions,” JHEP 03 ringdown?,” Phys. Rev. D 102 no. 6, (2020) 064053, (2019) 178, arXiv:1811.08538 [gr-qc]. arXiv:2006.14628 [gr-qc]. [26] V. P. Frolov and I. D. Novikov, Black hole physics : [47] V. Cardoso, E. Franzin, and P. Pani, “Is the basic concepts and new developments. 1998. gravitational-wave ringdown a probe of the event [27] S. D. Mathur, “The Information paradox: A horizon?,” Phys. Rev. Lett. 116 no. 17, (2016) 171101, Pedagogical introduction,” Class. Quant. Grav. 26 arXiv:1602.07309 [gr-qc]. [Erratum: Phys.Rev.Lett. (2009) 224001, arXiv:0909.1038 [hep-th]. 117, 089902 (2016)]. [28] S. D. Mathur, “The Fuzzball proposal for black holes: [48] Z. Mark, A. Zimmerman, S. M. Du, and Y. Chen, “A An Elementary review,” Fortsch. Phys. 53 (2005) recipe for echoes from exotic compact objects,” Phys. 793–827, arXiv:hep-th/0502050. Rev. D 96 no. 8, (2017) 084002, arXiv:1706.06155 [29] I. Bena and N. P. Warner, “Black holes, black rings [gr-qc]. and their microstates,” Lect. Notes Phys. 755 (2008) [49] L. F. L. Micchi and C. Chirenti, “Spicing up the recipe 1–92, arXiv:hep-th/0701216. for echoes from exotic compact objects: orbital [30] I. Bena and N. P. Warner, “Resolving the Structure of diﬀerences and corrections in rotating backgrounds,” Black Holes: Philosophizing with a Hammer,” Phys. Rev. D 101 no. 8, (2020) 084010, arXiv:1311.4538 [hep-th]. arXiv:1912.05419 [gr-qc]. [31] K. Skenderis and M. Taylor, “The fuzzball proposal for [50] L. F. Longo Micchi, N. Afshordi, and C. Chirenti, black holes,” Phys. Rept. 467 (2008) 117–171, “How loud are echoes from exotic compact objects?,” arXiv:0804.0552 [hep-th]. Phys. Rev. D 103 no. 4, (2021) 044028, [32] S. D. Mathur, “Fuzzballs and the information paradox: arXiv:2010.14578 [gr-qc]. A Summary and conjectures,” arXiv:0810.4525 [51] V. Cardoso, S. Hopper, C. F. B. Macedo, [hep-th]. C. Palenzuela, and P. Pani, “Gravitational-wave [33] V. S. Rychkov, “D1-D5 black hole microstate counting signatures of exotic compact objects and of quantum from supergravity,” JHEP 01 (2006) 063, corrections at the horizon scale,” Phys. Rev. D 94 arXiv:hep-th/0512053. no. 8, (2016) 084031, arXiv:1608.08637 [gr-qc]. [34] O. Lunin and S. D. Mathur, “Metric of the multiply [52] N. Yunes and X. Siemens, “Gravitational-Wave Tests wound rotating string,” Nucl. Phys. B 610 (2001) of General Relativity with Ground-Based Detectors 49–76, arXiv:hep-th/0105136. and Pulsar Timing-Arrays,” Living Rev. Rel. 16 [35] S. D. Mathur, A. Saxena, and Y. K. Srivastava, (2013) 9, arXiv:1304.3473 [gr-qc]. “Constructing ‘hair’ for the three charge hole,” Nucl. [53] R. Dey, S. Chakraborty, and N. Afshordi, “Echoes Phys. B 680 (2004) 415–449, arXiv:hep-th/0311092. from braneworld black holes,” Phys. Rev. D 101 [36] G. Chapline, E. Hohlfeld, R. B. Laughlin, and D. I. no. 10, (2020) 104014, arXiv:2001.01301 [gr-qc]. Santiago, “Quantum phase transitions and the [54] E. W. Leaver, “Spectral decomposition of the breakdown of classical general relativity,” Int. J. Mod. perturbation response of the schwarzschild geometry,” Phys. A 18 (2003) 3587–3590, arXiv:gr-qc/0012094. Phys. Rev. D 34 (Jul, 1986) 384–408. https: [37] P. O. Mazur and E. Mottola, “Gravitational vacuum //link.aps.org/doi/10.1103/PhysRevD.34.384. condensate stars,” Proc. Nat. Acad. Sci. 101 (2004) [55] C. Gundlach, R. H. Price, and J. Pullin, “Late-time 9545–9550, arXiv:gr-qc/0407075. behavior of stellar collapse and explosions. i. linearized [38] M. Visser and D. L. Wiltshire, “Stable gravastars: An perturbations,” Phys. Rev. D 49 (Jan, 1994) 883–889. Alternative to black holes?,” Class. Quant. Grav. 21 https: (2004) 1135–1152, arXiv:gr-qc/0310107. //link.aps.org/doi/10.1103/PhysRevD.49.883. [39] E. Mottola and R. Vaulin, “Macroscopic Eﬀects of the [56] C. Gundlach, R. H. Price, and J. Pullin, “Late-time Quantum Trace Anomaly,” Phys. Rev. D 74 (2006) behavior of stellar collapse and explosions. ii. nonlinear 064004, arXiv:gr-qc/0604051. evolution,” Phys. Rev. D 49 (Jan, 1994) 890–899. [40] E. Mottola, “New Horizons in Gravity: Dark Energy https: 17

//link.aps.org/doi/10.1103/PhysRevD.49.890. 345–383. [57] S. Hod, “Radiative tail of realistic rotating [73] R. H. Price and K. S. Thorne, “Membrane viewpoint gravitational collapse,” Phys. Rev. Lett. 84 (Jan, 2000) on black holes: Properties and evolution of the 10–13. https: stretched horizon,” Phys. Rev. D 33 (Feb, 1986) //link.aps.org/doi/10.1103/PhysRevLett.84.10. 915–941. https: [58] V. Cardoso and P. Pani, “Tests for the existence of //link.aps.org/doi/10.1103/PhysRevD.33.915. black holes through gravitational wave echoes,” Nature [74] B. Chen, Q. Wang, and Y. Chen, “Tidal response and Astron. 1 no. 9, (2017) 586–591, arXiv:1709.01525 near-horizon boundary conditions for spinning exotic [gr-qc]. compact objects,” Phys. Rev. D 103 no. 10, (2021) [59] R. S. Conklin, B. Holdom, and J. Ren, “Gravitational 104054, arXiv:2012.10842 [gr-qc]. wave echoes through new windows,” Phys. Rev. D 98 [75] J. Maldacena, “Comments on magnetic black holes,” no. 4, (2018) 044021, arXiv:1712.06517 [gr-qc]. arXiv:2004.06084 [hep-th]. [60] Q. Wang and N. Afshordi, “Black hole echology: The [76] S. Mereghetti, “The strongest cosmic magnets: Soft observer’s manual,” Phys. Rev. D 97 (Jun, 2018) Gamma-ray Repeaters and Anomalous X-ray Pulsars,” 124044. https: Astron. Astrophys. Rev. 15 (2008) 225–287, //link.aps.org/doi/10.1103/PhysRevD.97.124044. arXiv:0804.0250 [astro-ph]. [61] J. Abedi, H. Dykaar, and N. Afshordi, “Echoes from [77] V. Cardoso, C. F. B. Macedo, P. Pani, and V. Ferrari, the Abyss: Tentative evidence for Planck-scale “Black holes and gravitational waves in models of structure at black hole horizons,” Phys. Rev. D 96 minicharged dark matter,” JCAP 05 (2016) 054, no. 8, (2017) 082004, arXiv:1612.00266 [gr-qc]. arXiv:1604.07845 [hep-ph]. [Erratum: JCAP 04, [62] B. Holdom, “Not quite black holes at LIGO,” Phys. E01 (2020)]. Rev. D 101 no. 6, (2020) 064063, arXiv:1909.11801 [78] D. Ghosh, A. Thalapillil, and F. Ullah, “Astrophysical [gr-qc]. hints for magnetic black holes,” Phys. Rev. D 103 [63] R. S. Conklin and B. Holdom, “Gravitational wave no. 2, (2021) 023006, arXiv:2009.03363 [hep-ph]. echo spectra,” Phys. Rev. D 100 no. 12, (2019) [79] E. G. Gimon and P. Horava, “Astrophysical violations 124030, arXiv:1905.09370 [gr-qc]. of the Kerr bound as a possible signature of string [64] K. W. Tsang, A. Ghosh, A. Samajdar, theory,” Phys. Lett. B 672 (2009) 299–302, K. Chatziioannou, S. Mastrogiovanni, M. Agathos, and arXiv:0706.2873 [hep-th]. C. Van Den Broeck, “A morphology-independent [80] L. Buoninfante and A. Mazumdar, “Nonlocal star as a search for gravitational wave echoes in data from the blackhole mimicker,” Phys. Rev. D 100 no. 2, (2019) first and second observing runs of Advanced LIGO and 024031, arXiv:1903.01542 [gr-qc]. Advanced Virgo,” Phys. Rev. D 101 no. 6, (2020) [81] T. Damour and S. N. Solodukhin, “Wormholes as 064012, arXiv:1906.11168 [gr-qc]. black hole foils,” Phys. Rev. D 76 (2007) 024016, [65] F. Salemi, E. Milotti, G. A. Prodi, G. Vedovato, arXiv:0704.2667 [gr-qc]. C. Lazzaro, S. Tiwari, S. Vinciguerra, M. Drago, and [82] P. Dutta Roy, S. Aneesh, and S. Kar, “Revisiting a S. Klimenko, “Wider look at the gravitational-wave family of wormholes: geometry, matter, scalar transients from GWTC-1 using an unmodeled quasinormal modes and echoes,” Eur. Phys. J. C 80 reconstruction method,” Phys. Rev. D 100 no. 4, no. 9, (2020) 850, arXiv:1910.08746 [gr-qc]. (2019) 042003, arXiv:1905.09260 [gr-qc]. [83] N. Dadhich, R. Maartens, P. Papadopoulos, and [66] N. Uchikata, H. Nakano, T. Narikawa, N. Sago, V. Rezania, “Black holes on the brane,” Phys. Lett. B H. Tagoshi, and T. Tanaka, “Searching for black hole 487 (2000) 1–6, arXiv:hep-th/0003061. echoes from the LIGO-Virgo Catalog GWTC-1,” Phys. [84] R. Dey, S. Biswas, and S. Chakraborty, “Ergoregion Rev. D 100 no. 6, (2019) 062006, arXiv:1906.00838 instability and echoes for braneworld black holes: [gr-qc]. Scalar, electromagnetic and gravitational [67] J. Westerweck, A. Nielsen, O. Fischer-Birnholtz, perturbations,” arXiv:2010.07966 [gr-qc]. M. Cabero, C. Capano, T. Dent, B. Krishnan, [85] S. Chakraborty, S. Datta, and S. Sau, “Tidal heating G. Meadors, and A. H. Nitz, “Low significance of of black holes and exotic compact objects on the evidence for black hole echoes in gravitational wave brane,” arXiv:2103.12430 [gr-qc]. data,” Phys. Rev. D 97 no. 12, (2018) 124037, [86] W. Israel, “Singular hypersurfaces and thin shells in arXiv:1712.09966 [gr-qc]. general relativity,” Nuovo Cimento B Serie 44 no. 1, [68] N. Oshita, Q. Wang, and N. Afshordi, “On Reflectivity (July, 1966) 1–14. of Quantum Black Hole Horizons,” JCAP 04 (2020) [87] G. Darmois, Les équationsde la gravitation 016, arXiv:1905.00464 [hep-th]. einsteinienne. Gauthier-Villars, 1927. [69] Q. Wang, N. Oshita, and N. Afshordi, “Echoes from http://eudml.org/doc/192556. Quantum Black Holes,” Phys. Rev. D 101 no. 2, [88] M. K. Parikh and F. Wilczek, “An action for black (2020) 024031, arXiv:1905.00446 [gr-qc]. hole membranes,” Phys. Rev. D 58 (Aug, 1998) [70] T. Damour, “Surface Effects in Black-Hole Physics,” in 064011. https: Marcel Grossmann Meeting: General Relativity, p. 587. //link.aps.org/doi/10.1103/PhysRevD.58.064011. Jan., 1982. [89] S. Chatterjee, M. Parikh, and S. Sarkar, “The Black [71] K. S. Thorne, R. H. Price, and D. A. MacDonald, Hole Membrane Paradigm in f(R) Gravity,” Class. Black holes: The membrane paradigm. 1986. Quant. Grav. 29 (2012) 035014, arXiv:1012.6040 [72] D. MacDonald and K. S. Thorne, “Black-hole [hep-th]. electrodynamics - an absolute-space/universal-time [90] T. Jacobson, A. Mohd, and S. Sarkar, “Membrane formulation,” Mon. Not. Roy. Astron. Soc. 198 (1982) paradigm for einstein-gauss-bonnet gravity,” Phys. 18

Rev. D 95 (Mar, 2017) 064036. https: Wu, “Spontaneous Pair Production in //link.aps.org/doi/10.1103/PhysRevD.95.064036. Reissner-Nordstrom Black Holes,” Phys. Rev. D 85 [91] S. Chandrasekhar, The mathematical theory of black (2012) 124041, arXiv:1202.3224 [hep-th]. holes. 1985. [107] Y.-T. Wang, Z.-P. Li, J. Zhang, S.-Y. Zhou, and Y.-S. [92] M. Hobson, G. Efstathiou, and A. Lasenby, General Piao, “Are gravitational wave ringdown echoes always relativity: An introduction for physicists. 10, 2006. equal-interval?,” Eur. Phys. J. C 78 no. 6, (2018) 482, [93] V. Moncrief, “Gauge-invariant perturbations of arXiv:1802.02003 [gr-qc]. Reissner-Nordstrom black holes,” Phys. Rev. D 12 [108] B. Allen, W. G. Anderson, P. R. Brady, D. A. Brown, (1975) 1526–1537. and J. D. E. Creighton, “FINDCHIRP: An Algorithm [94] S. Chandrasekhar and B. C. Xanthopoulos, “On the for detection of gravitational waves from inspiraling metric perturbations of the reissner-nordströmblack compact binaries,” Phys. Rev. D 85 (2012) 122006, hole,” Proceedings of the Royal Society of London. arXiv:gr-qc/0509116. Series A, Mathematical and Physical Sciences 367 [109] LIGO Scientific, Virgo Collaboration, B. P. Abbott no. 1728, (1979) 1–14. et al., “GW150914: First results from the search for http://www.jstor.org/stable/79860. binary black hole coalescence with Advanced LIGO,” [95] C. V. Vishveshwara, “Scattering of Gravitational Phys. Rev. D 93 no. 12, (2016) 122003, Radiation by a Schwarzschild Black-hole,” Nature 227 arXiv:1602.03839 [gr-qc]. (1970) 936–938. [110] A. Testa and P. Pani, “Analytical template for [96] S. Chandrasekhar and S. L. Detweiler, “The gravitational-wave echoes: Signal characterization and quasi-normal modes of the Schwarzschild black hole,” prospects of detection with current and future Proc. Roy. Soc. Lond. A 344 (1975) 441–452. interferometers,” Phys. Rev. D 98 (Aug, 2018) 044018. [97] K. D. Kokkotas and B. G. Schmidt, “Quasinormal https: modes of stars and black holes,” Living Rev. Rel. 2 //link.aps.org/doi/10.1103/PhysRevD.98.044018. (1999) 2, arXiv:gr-qc/9909058. [111] LIGO Scientific, Virgo Collaboration, T. D. Abbott [98] E. Berti, V. Cardoso, and A. O. Starinets, et al., “Improved analysis of GW150914 using a fully “Quasinormal modes of black holes and black branes,” spin-precessing waveform Model,” Phys. Rev. X 6 Class. Quant. Grav. 26 (2009) 163001, no. 4, (2016) 041014, arXiv:1606.01210 [gr-qc]. arXiv:0905.2975 [gr-qc]. [112] LIGO Scientific and V. Collaboration. (GW150914). [99] V. Cardoso and P. Pani, “The observational evidence [113] LIGO GW Open Science Center. for horizons: from echoes to precision (H1-GDS-CALIB-STRAIN). gravitational-wave physics,” arXiv:1707.03021 [114] LIGO Scientific and V. Collaboration. [gr-qc]. (gw-openscience.org). [100] V. Cardoso, A. S. Miranda, E. Berti, H. Witek, and [115] LIGO Scientific and V. Collaboration. (SNR). V. T. Zanchin, “Geodesic stability, Lyapunov [116] L. B. Bethke, Exploring the Early Universe with exponents and quasinormal modes,” Phys. Rev. D 79 Gravitational Waves. PhD thesis, Imperial Coll., (2009) 064016, arXiv:0812.1806 [hep-th]. London, 2014. [101] F. Caruso and V. Oguri, “Numerov umerical method [117] L. A. Boyle, Gravitational waves and the early applied to the Schrödingerequation,” Rev. Bras. Ens. universe. PhD thesis, Princeton U., 2006. Fis. 36 no. 2, (2014) [118] V. Cardoso and L. Gualtieri, “Testing the black hole http://www.scielo.br/pdf/rbef/v36n2/10.pdf, ‘no-hair’ hypothesis,” Class. Quant. Grav. 33 no. 17, arXiv:1403.7092 [quant-ph]. (2016) 174001, arXiv:1607.03133 [gr-qc]. [102] P. Pani, “Advanced Methods in Black-Hole [119] E. Barausse, V. Cardoso, and P. Pani, “Can Perturbation Theory,” Int. J. Mod. Phys. A 28 (2013) environmental effects spoil precision gravitational-wave 1340018, arXiv:1305.6759 [gr-qc]. astrophysics?,” Phys. Rev. D 89 no. 10, (2014) 104059, [103] R. A. Konoplya and A. Zhidenko, “Quasinormal arXiv:1404.7149 [gr-qc]. modes of black holes: From astrophysics to string [120] R. Dey and N. Afshordi, “Echoes in the Kerr/CFT theory,” Rev. Mod. Phys. 83 (2011) 793–836, correspondence,” Phys. Rev. D 102 no. 12, (2020) arXiv:1102.4014 [gr-qc]. 126006, arXiv:2009.09027 [hep-th]. [104] P. Bueno, P. A. Cano, F. Goelen, T. Hertog, and [121] S. Basak, P. D. Roy, and S. Kar, “A new model with B. Vercnocke, “Echoes of Kerr-like wormholes,” Phys. solitary waves: solution, stability and quasinormal Rev. D 97 no. 2, (2018) 024040, arXiv:1711.00391 modes,” arXiv:2012.10967 [hep-th]. [gr-qc]. [122] C. Kittel, Introduction to Solid State Physics. Wiley, [105] Y.-W. Kim, Y. S. Myung, and Y.-J. Park, 8 ed., 2004. http://www.amazon.com/ “Quasinormal modes and hidden conformal symmetry Introduction-Solid-Physics-Charles-Kittel/dp/ in the Reissner-Nordströmblack hole,” Eur. Phys. J. 047141526X/ref=dp_ob_title_bk. C 73 (2013) 2440, arXiv:1205.3701 [hep-th]. [106] C.-M. Chen, S. P. Kim, I.-C. Lin, J.-R. Sun, and M.-F.