Abstract
It has long been suggested that the Cauchy horizon of dynamical black holes is subject to a weak null singularity, under the mass inflation scenario. We study in spherical symmetry the Einstein–Maxwell–Klein–Gordon equations and while we do not directly show mass inflation, we obtain a “mass inflation/ridigity” dichotomy. More precisely, we prove assuming (sufficiently slow) decay of the charged scalar field on the event horizon, that the Cauchy horizon emanating from time-like infinity \(\mathcal {CH}_{i^+}\) can be partitioned as \(\mathcal {CH}_{i^+}= {\mathcal {D}} \cup {\mathcal {S}}\) for two (possibly empty) disjoint connected sets \({\mathcal {D}}\) and \({\mathcal {S}}\) such that
-
\({\mathcal {D}}\) (the dynamical set) is a future set on which the Hawking mass blows up (mass inflation scenario).
-
\({\mathcal {S}}\) (the static set) is a past set isometric to a Reissner–Nordström Cauchy horizon i.e. the radiation is zero on \({\mathcal {S}}\).
As a consequence of this result, we prove that the entire Cauchy horizon \(\mathcal {CH}_{i^+}\) is globally \(\underline{C^2-{ inextendible}}\), extending a previous local result established by the author. To this end, we establish a novel classification of Cauchy horizons into three types: dynamical (\({\mathcal {S}}=\emptyset \)), static (\({\mathcal {D}}=\emptyset \)) or mixed. As a side benefit, we prove that there exists a trapped neighborhood of the Cauchy horizon, thus the apparent horizon cannot cross the Cauchy horizon, which is a result of independent interest. Our main motivation is to prove the \(C^2\) Strong Cosmic Censorship Conjecture for a realistic model of spherical collapse in which charged matter emulates the repulsive role of angular momentum. In our case, this model is the Einstein–Maxwell–Klein–Gordon system on space-times with one asymptotically flat end. As a consequence of the \(C^2\)-inextendibility of the Cauchy horizon, we prove the following statements, in spherical symmetry:
-
1.
Two-ended asymptotically flat space-times are \(C^2\)-future-inextendible i.e. \(C^2\) Strong Cosmic Censorship is true for Einstein–Maxwell–Klein–Gordon, assuming the decay of the scalar field on the event horizon at the expected rate.
-
2.
In the one-ended case, under the same assumptions, the Cauchy horizon emanating from time-like infinity is \(C^2\)-inextendible. This result suppresses the main obstruction to \(C^2\) Strong Cosmic Censorship in spherical collapse.
The remaining obstruction in the one-ended case is associated to “locally naked” singularities emanating from the center of symmetry, a phenomenon which is also related to the Weak Cosmic Censorship Conjecture.
Similar content being viewed by others
Notes
In the two-ended case, no additional assumption is required. In the one-ended case, we obtain the result assuming additionally the absence of “locally naked singularity” emanating from the center of symmetry, a slightly stronger statement than the Weak Cosmic Censorship Conjecture.
This statement does not prove that the metric is \(C^1\)-inextendible but does give the insight that a breakdown occurs already at the \(C^1\) level.
This is consequence of the Raychaudhuri equation and the dominant energy condition, as the quantity \(\frac{4|\partial _u r|}{\Omega ^2}\) is monotonic c.f. (A.7).
In fact, the space-time is \(C^2\) extendible and the Hawking mass finite if \(f_R\) decays exponentially at a sufficiently fast rate. This phenomenon explains why in the cosmological setting, mass inflation is not expected for a certain range of parameters c.f. [12].
Essentially, such data decay weakly and are non-oscillating, so do not obey the asymptotics of Conjecture 1.7 (non-generic behavior).
e.g. they are important for the \(C^0\)-inextendibility result of [26].
In fact, the original theorem of [47] only requires \(s>\frac{1}{2}\), but we assume \(s>\frac{3}{4}\) to simplify the proofs in the present paper.
Additionally, if if \(s>1\), we prove in [47] that \((M,g,\phi ,F)\) admits a continuous extension to \(\mathcal {CH}_{i^+}\).
Of course, if \(\mathcal {CH}_{i^+}\) is the ingoing Cauchy horizon of the one-ended case, or \(\mathcal {CH}_{i^+_1}\) in the two-ended case, we can chose \(\tau =u\) and \(\varsigma =v\).
By this, we mean that \(\rho ^{-1}\) and \(\varpi ^{-1}\) extend continuously to 0 on \(\mathcal {CH}_{i^+}\).
Otherwise, we already know that \(\mathcal {LB} \cap \{u \le u_s\} \subset {\mathcal {T}}\) (recall that \(\mathcal {LB}\) is defined as the future of a space-like trapped curve \(\gamma \), c.f. [47]).
We are using, implicitly, a standard local well-posedness argument for characteristic initial data entirely inside the space-time, i.e. whose closure is disjoint from the boundary of the Penrose diagram.
Precisely Lemma 10.3, in which it is assumed that the rectangle has finite volume. The finiteness of the volume is later obtained by a different argument, using a monotonicity property specific to the uncharged and massless considered in [32].
Making point-wise lower bounds assumptions on the scalar field on the event horizon, and exploiting a special monotonicity property which is specific to the uncharged and massless case.
As noticed in [16], the monotonicity of the Hawking mass in the uncharged and massless case allows to propagate the mass blow-up to the entire Cauchy horizon, which implies that condition (5.2) is violated everywhere on \(\mathcal {CH}_{i^+}\) for the space-times under consideration. This technique does not, however, survive when the field is massive and/or charged.
This additional assumption is required to obtain a lower bound on r, which is not necessarily valid as one approaches \(\{(u_{\infty }(\mathcal {CH}_{i^+}),v=+\infty )\}\) the end-point towards which r may (or may not) approach 0, c.f. [28].
In particular R is of finite space-time volume, but as we shall see, the stronger assumption \(R \subset {\mathcal {T}}\) turns out to be necessary to our method.
Recall that, in retrospect, Lemma 5.8 was concerned by Cauchy horizon of mixed type: in this case, \(\phi \) and Q were necessarily bounded.
This is because we require estimates, even weak ones, just to prove that \(\nu \) has a non-zero limit on \(\mathcal {CH}_{i^+}\).
The proof of [32], which does not use the equations, can be exactly reproduced in our setting.
Note that Ori considers Cauchy horizons of mixed type in [37].
Note that while the estimates of [47] concern the scalar field case, they can be immediately transposed to the null dust case with no further work, as the Einstein equations are the same, except with fewer terms and moreover the propagation of both dust clouds \(f_R\) and \(f_L\) is trivial.
References
Angelopoulos, Y., Aretakis, S., Gajic, D.: A vector field approach to almost-sharp decay for the wave equation on spherically symmetric, stationary spacetimes. Ann. PDE, volume 4, Article number: 15 (2018)
Angelopoulos, Y., Aretakis, S.: Dejan Gajic Late-time asymptotics for the wave equation on spherically symmetric, stationary spacetimes. Adv. Math. 323, 529–621 (2018)
Burko, L.: Gaurav Khanna Universality of massive scalar field late-time tails in black-hole spacetimes. Phys. Rev. D 70, 044018 (2004)
Burq, N., Planchon, F., Stalker, J., Tahvildar-Zadeh, A.S.: Strichartz estimates for wave and Schroedingerequations with potentials of critical decay. Indiana Univ. Math. J. 53(6), 1667–1682 (2004)
Burq, N., Planchon, F., Stalker, J., Tahvildar-Zadeh, A.S.: Strichartz estimates for the wave and Schroedinger equations with the inverse-square potential. J. Funct. Anal. 203, 519–549 (2003)
Choquet-Bruhat, Y.: Theoreme d’existence pour certains systemes d’equations aux derivees partielles non linaires. Acta Math. 88, 141–225 (1952)
Christodoulou, D.: The formation of black holes and singularities in spherically symmetric gravitational collapse. Commun. Pure Appl. Math. 44(3), 339–373 (1991)
Christodoulou, D.: Bounded variation solutions of the spherically symmetric Einstein-scalar field equations. Commun. Pure Appl. Math. 46, 1131–1220 (1993)
Christodoulou, D.: On the global initial value problem and the issue of singularities. Class. Quantum Grav. 16, A23–A35 (1999)
Christodoulou, D.: The instability of naked singularities in the gravitational collapse of a scalar field. Ann. Math. (2) 149(1), 183–217 (1999)
Christodoulou, Demetrios.: The formation of black holes in general relativity. European Mathematical Society (2009)
Costa, J.L., Girão, P.M., Natário, J., Silva, J.D.: On the global uniqueness for the Einstein–Maxwell-scalar field system with a cosmological constant Part 3: mass inflation and extendibility of the solutions. Ann PDE 3, 8 (2017)
Dafermos, M.: Stability and instability of the Cauchy horizon for the spherically symmetric Einstein–Maxwell-scalar field equations. Ann. Math. 158, 875–928 (2003)
Dafermos, M.: The interior of charged black holes and the problem of uniqueness in general relativity. Commun. Pure Appl. Math. 58, 0445–0504 (2005)
Dafermos, M.: Spherically symmetric spacetimes with a trapped surface. Class. Quantum Grav. 22(11) (2005)
Dafermos, M.: Black holes without spacelike singularities. Commun. Math. Phys. 332, 729–757 (2014)
Dafermos, M., Luk, J.: The interior of dynamical vacuum black holes I: the \(C^0\)-stability of the Kerr Cauchy horizon. (2017). arXiv:1710.01772
Dafermos, M., Rendall, A.: Inextendibility of expanding cosmological models with symmetry. Class. Quantum Grav. 22(23), L143–L147 (2005)
Dafermos, M.: Igor Rodnianski A proof of Price’s law for the collapse of a self-gravitating scalar field. Invent. math. 162, 381–457 (2005)
Donninger, R.: Joachim Krieger a vector field method on the distorted Fourier side and decay for wave equations with potentials. Mem. Am. Math. Soc. (2015)
Gajic, D., Luk, J.: The interior of dynamical extremal black holes in spherical symmetry. Pure Appl. Anal. 1(2), 263–326 (2019)
Hiscock, W.: Evolution of the interior of a charged black hole. Phys. Lett. A 83(3), 110–112 (1981)
Hod, S., Piran, T.: Late-time evolution of charged gravitational collapse and decay of charged scalar hair-II. Phys. Rev. D 58, 024018 (1998)
Hod, S., Piran, T.: Mass-inflation in dynamical gravitational collapse of a charged scalar-field. Phys. Rev. Lett. 81, 1554–1557 (1998)
Kehle, C.: Yakov Shlapentokh-Rothman a scattering theory for linear waves on the interior of Reissner–Nordström black holes. Ann. Henri Poincaré 20 (2019)
Kehle, C., Van de Moortel, M.: Continuous extendibility and non-linear scattering on the interior of dynamical black holes in the presence of matter, in preparation
Kehle, C., Van de Moortel, M.: The null contraction singularity at the Cauchy horizon of dynamical black holes, in preparation
Kommemi, J.: The global structure of spherically symmetric charged scalar field spacetimes. Commun. Math. Phys. 323, 35 (2013). https://doi.org/10.1007/s00220-013-1759-1
Koyama, H., Tomimatsu, A.: Asymptotic power-law tails of massive scalar fields in Reissner–Nordström background. Phys. Rev. D 63, 064032 (2001)
Konoplya, R., Zhidenko, A.: A massive charged scalar field in the Kerr–Newman background I: quasi-normal modes, late-time tails and stability. Phys. Rev. D 88, 024054 (2013)
Luk, J.: Weak null singularities in general relativity. J. AMS 31, 1–63 (2018)
Luk, J.: Sung-Jin Oh Strong Cosmic Censorship in Spherical Symmetry for two-ended Asymptotically Flat Initial Data I. The Interior of the Black Hole Region, Preprint (2017)
Luk, J.: Sung-Jin Oh Strong Cosmic Censorship in Spherical Symmetry for two-ended Asymptotically Flat Initial Data II. The Exterior of the Black Hole Region, Preprint (2017)
McNamara, J.: Instability of black hole inner horizons. Proc. R. Soc. Lon. A 358, 499–517 (1978)
Moschidis, G.: The Einstein–Null dust system in spherical symmetry with an inner mirror: structure of the maximal development and Cauchy stability., Preprint, arXiv:1704.08685, (2017)
Oren, Y., Piran, T.: On the collapse of charged scalar fields. Phys. Rev. D 68, 044013 (2003)
Ori, A.: Inner structure of a charged black hole: an exact mass-inflation solution. Phys. Rev. Lett. 67(7), 789–792 (1991)
Penrose, R.: Structure of space-time. In: Penrose, R., DeWitt, C.M. (eds.) Battelle Rencontres, p. 22. W. A. Benjamin, New York (1968)
Penrose, R.: Gravitational collapse: the role of general relativity. Riv. Nuovo Cim. 1, 252–276 (1969)
Planchon, F., Stalker, J., Tahvildar–Zadeh, A.S.: Dispersive estimate for the wave equation with the inverse-square potential. Discrete Contin. Dyn. Syst. 9(6), 1387–1400 (2003)
Planchon, F., Stalker, J., Tahvildar-Zadeh, A.S.: \(L^p\) estimates for the wave equation with the inverse-square potential. Discrete Contin. Dyn. Syst. 9(2), 427–442 (2003)
Poisson, E., Israel, W.: Internal structure of black holes. Phys. Rev. D 63, 1663–1666 (1989)
Poisson, E., Israel, W.: Inner-horizon instability and mass inflation in black holes. Phys. Rev. Lett. 67(7), 789–792 (1991)
Price, R.: Nonspherical perturbations of relativistic gravitational collapse. I. Scalar and gravitational perturbations. Phys. Rev. D (3) 5, 2419–2438 (1972)
Sbierski, J.: Characterisation of the energy of Gaussian beams on Lorentzian manifolds: with applicationsto black hole spacetimes. Anal. PDE 8(6), 1379–1420 (2015)
Sbierski, J.: The \(C^0\)-inextendibility of the Schwarzschild spacetime and the spacelike diameter in Lorentzian geometry. J. Differ. Geom. 108(2), 319–378 (2018)
Van de Moortel, M.: Stability and instability of the sub-extremal Reissner–Nordström black hole interior for the Einstein–Maxwell–Klein–Gordon equations in spherical symmetry. Commun. Math. Phys. 360(1), 103–168 (2018)
Van de Moortel, M.: Decay of weakly charged solutions for the spherically symmetric Maxwell—Charged-Scalar-Field equations on a Reissner–Nordström exterior space-time, preprint, 2018. arXiv:1804.04297
Van de Moortel , M.: Charged scalar fields on Black Hole space-times, PhD Thesis, University of Cambridge (2018)
Van de Moortel, M.: The breakdown of weak null singularities inside black holes, preprint. arXiv:1912.10890 (2019)
Acknowledgements
I am very grateful to Jonathan Luk for suggesting this problem and for fruitful discussions. I would like to thank Sung-Jin Oh for this interest in this problem, and for insightful discussions. I am grateful to Mihalis Dafermos and Jonathan Luk for useful comments on the manuscript. Most of this work was completed while I was a Ph.D. student at the University of Cambridge, and a visiting graduate student at Stanford University, whose hospitality I gratefully acknowledge. Finally, I would like to thanks two anonymous reviewers for helpful suggestions which helped clarify some aspects of the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by P. Chrusciel
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix A Construction of a Cauchy Horizon of Mixed Type for the Einstein–Null–Dust Model
Appendix A Construction of a Cauchy Horizon of Mixed Type for the Einstein–Null–Dust Model
In this appendix, we construct an example of a Cauchy horizon of mixed type, following Definition 3.3. In the second part of our development, we prove the blow up of the Hawking mass for mixed and dynamical type Cauchy horizons in the Einstein–Maxwell-null-dust model, following Poisson and Israel [43] andFootnote 26 Ori [37], the first instances of the mass inflation scenario in the literature. Notice that their model is very elementary, as the dust clouds are simply transported linearly in the null directions, and only interact indirectly, via the metric: thus, the “scattering theory” is trivial. Additionally, such a model does not allow for one-ended regular solutions, unlike the charged/massive scalar field model. In this appendix we sketch a mathematical proof of mass inflation, using the methods of [47]; this is also an opportunity to illustrate the terminology of the present paper in the simpler setting of dust, where little analysis is required and monotonicity suffices. The system (1.7), (1.8), (1.9), (1.10), (1.11), (1.12), (1.13) can be expressed in spherical symmetry as
Proposition A.1
We consider the maximal development of smooth, spherically symmetric and admissible (in the sense of Definition 2.6) two-ended initial data \((M={\mathcal {Q}}^+ \times _r {\mathcal {S}}^2,g_{\mu \nu }, \phi ,F_{\mu \nu })\) satisfying the Einstein–Maxwell-null-dust system. Suppose that Assumptions 1, 2 and 3 of Theorem 3.1 are satisfied, and we also work under the same (U, v) gauge choice. We also choose \(U_{max} \in {\mathbb {R}}\) such that \( [0,U_{max}]\times \{v=+\infty \} \subset \mathcal {CH}_{i^+}\) in the Penrose diagram \({\mathcal {Q}}^+\). We consider data \(f_L^0\) on the event horizon \({\mathcal {H}}^+= \{0\} \times [v_0,+\infty )\) and \(f_R^0\) on an ingoing cone \(\underline{C}_{v_0}= [0,U_{max}]\times \{v_0\}\); we assume that both \(f_R^0\) and \(f_L^0\) are smooth in the (U, v) coordinate system. We also make assumptions analogous to Assumptions 4 and 5 of Theorem 3.1 i.e. for some \(s>\frac{1}{2}\) and some \(C>0\):
Then by continuity of \(f^0_R\) on \([0,U_{max})\), there are three possibilities:
-
(a)
For all \(0<U_s<U_{max}\), there exists \(0 \le U \le U_s\) such that \(f^0_R(U) \ne 0\). We call this the dynamical case.
-
(b)
For all \(U\in [0,U_{max})\), \(f^0_R(U) = 0\). We call this the static case.
-
(c)
There exists \(U_T \in (0,U_{max})\) such that for all \(U\in [0,U_{T}]\), \(f^0_R(U) = 0\), and such that for every \(\epsilon >0\), there exists \(U\in (U_{T},U_{T}+\epsilon ]\) such that \(f^0_R(U) \ne 0\). We call this the mixed case.
Then we have, in the three different cases:
-
(a)
In the dynamical case, r extends to \(\mathcal {CH}_{i^+}\) as function \(r_{CH}\) which is strictly decreasing on \([0,U_{max})\).
-
(b)
In the static case, r extends to \(\mathcal {CH}_{i^+}\) as a constant \(r_{-}>0\).
-
(c)
In the mixed case, r extends to \(\mathcal {CH}_{i^+}\) as a function \(r_{CH}\), which is constant on \([0,U_T]\), and strictly decreasing on \((U_T,U_{max} )\).
If we additionally assume that \(f_L^0(v)=C \cdot v^{-s}+o(v^{-s})\), for some \(s>\frac{1}{2}\), where v is defined by gauge (3.1), then we have
-
(a)
In the dynamical case, the Hawking mass \(\rho \) blows up on \(\mathcal {CH}_{i^+}\cap [0,U_{max}] \).
-
(b)
In the static case, the Hawking mass \(\rho \) is constant on \(\mathcal {CH}_{i^+}\cap [0,U_{max}] \).
-
(c)
In the mixed case, the Hawking mass \(\rho \) is constant on \(\mathcal {CH}_{i^+}\cap [0,U_{T}] \) and blows up on \(\mathcal {CH}_{i^+}\cap (U_T,U_{max}]\).
Remark A.1
Note that \(\mathcal {CH}_{i^+}\ne \emptyset \), applying (an easier version of) the argument of [47], which proves the stability of the Reissner–Nordström Cauchy horizon for the Einstein–Maxwell–Klein–Gordon model in spherical symmetry.
Proof
We first prove that the possibilities reduce to the three cases a, b, c. Define the set \(E:=\{U\in [0,U_{max}), f_R^0(U)\ne 0\}\). If \(E=\emptyset \) then we are in case b. If \(E \ne \emptyset \), define \(U_T=\inf E\). There are two possibilities: either \(U_T=0\) or \(U_T>0\). If \(U_T=0\) then we are in case a by definition of E. If \(U_T>0\) then for all \(0 \le U < U_T\), \(f_R^0(U)= 0\). By continuity of \(f_R^0\) we have also \(f_R^0(U_T)= 0\). Again by definition of E, we are therefore in case c.
We work in the gauge \(\kappa _{|{\mathcal {H}}^+} \equiv 1\) and we denote \(\varphi _R(u)= r(u,v) f_R(u,v)\), \(\varphi _L(v)= r(u,v) f_L(u,v)\) by (A.5), (A.6).
By the stability estimatesFootnote 27 of [47] (see Sect. 4), one can show that for some sub-extremal parameters \(0<|e|<M\),
holds on some space-like curve \(\gamma \) terminating at \(i^+\), for v large enough. We will then establish estimates for all \(U \in [0,U_{max}]\): combining (A.1) and (A.2), we get \( \partial _U \partial _v \log (r \Omega ^2) = \frac{ \Omega ^{2}}{2r^{2}} \cdot (1- \frac{ 3 e^{2}}{r^{2}})\). In view of \(|e| > r_-(M,e)\) and the monotonicity of r (which implies that \(r(U,v)\le r(U_{\gamma }(v),v) \) in our region of interest), we have \( \partial _U \partial _v \log (r \Omega ^2)(U,v) <0\) for all \(U \in [0,U_{max}]\). Integrating in U from \(\gamma \), assuming v large enough (in particular \(v\ge v_{\gamma }(U_s)\)), we have, for all \(U \in [0,U_{max}]\):
where the last inequality follows from v being large enough. This proves that for all \(U \in [0,U_{max}]\)
Using (A.10) with \(\partial _U \partial _v \log (r \Omega ^2) = \frac{ \Omega ^{2}}{2r^{2}} \cdot (1- \frac{ 3 e^{2}}{r^{2}})\) and the fact that \( C(M,e)^{-1} \lesssim |1- \frac{ 3 e^{2}}{r^{2}}| \lesssim C(M,e)\) by monotonicity of r and the fact that r is lower bounded, we get an improved estimate, still for all \(U \in [0,U_{max}]\):
Now, we can write (A.1), since \(\nu \le 0\), as \(\partial _v( r|\nu |) =\Omega ^2 \cdot (1- \frac{e^2}{r^2})\). Since r is decreasing in U, \((1- \frac{e^2}{r^2(U,v)})<0\) for all \(U \in [0,U_{max}]\), as \(r<r_-(M,e)<|e|\). Hence \(v \rightarrow r\nu (U,v)\) is strictly decreasing for all fixed U, thus has a limit \(r\nu _{\mathcal {CH}_{i^+}}(U) \in {\mathbb {R}}_-\). Moreover, we establish, using (A.1), the following estimate for all \(U \in [0,U_{max}]\):
Now, writing (A.1) as \(\partial _U(- r\lambda ) =\Omega ^2 \cdot (1- \frac{e^2}{r^2})\), we get decay for \(\lambda \) so we can extend r to \(\mathcal {CH}_{i^+}\) into a function \(r_{\mathcal {CH}_{i^+}} >0\). Therefore, for any fixed \(U<U_{max}\), \(v \rightarrow \nu (U,v)\) has a limit as \(v\rightarrow +\infty \) that we consistently denote \(\nu _{\mathcal {CH}_{i^+}}(U)\).
Since r is lower bounded in this region, we have \(\partial _U (\kappa ^{-1})(U,v) \gtrsim C \cdot \varphi _R^2(U) \cdot e^{|K_-|(M,e) v}\) hence, integrating from \(\gamma \):
Then, there are our three possibilities, starting with the easiest:
-
1.
Dynamical case: then for all \(U\in [0,U_{max})\), \(\int _{0}^U(f^0_R)^2(U')dU' > 0\). Indeed by (a), for all \(U\in [0,U_{max})\), there exists \(U'<U\) with \((f^0_R(U'))^2 > 0\), hence by continuity there exists \(\epsilon >0\) such that \((f^0_R(U'))^2 > 0\) on \([U',U'+\epsilon ]\), which implies \(\int _{0}^U(f^0_R)^2(U')dU' > 0\) . In view of the limit \(U_{\gamma }(v) \rightarrow 0\) as \(v\rightarrow +\infty \) (recall that \(\gamma \) terminates at \(i^+\)) then \(\lim _{v \rightarrow +\infty } \int _{U_{\gamma (v)}}^U (f_R^0(U'))^2dU' >0\) for all \(U\in [0,U_{max})\), hence \(\kappa ^{-1}(U,v) \rightarrow +\infty \). If \( \nu _{CH}(U)=0\) for some \(0<U < U_{max}\), then (A.12) is contradicted (after dividing both sides of (A.12) by \(r \Omega ^2\)). Hence \(\nu _{\mathcal {CH}_{i^+}}(U)<0\) for all \(U\in [0,U_{max}]\) thus \(r_{\mathcal {CH}_{i^+}}\) is strictly decreasing on \([0,U_{max})\) as desired . Moreover \( \kappa ^{-1}(U,v) \sim \frac{4|\nu _l|(U)}{\Omega ^2(U,v)}\) by (A.12). If additionally we have a lower bound \(f_L^0(v)=C \cdot v^{-s}+o(v^{-s})\) then by the stability estimates of [47], \(-\lambda \sim C \cdot v^{-2s}\), at least for U small enough and in the future of the curve \(\gamma \). Since \(2\rho -r= \kappa ^{-1} \cdot (-\lambda )\), \(\rho (U,v) \rightarrow +\infty \) as \( v \rightarrow +\infty \), for small U. A forciciori, \(\varpi (U,v) \rightarrow +\infty \) for small U, and by the monotonicity of (A.3), \(\varpi (U,v) \rightarrow +\infty \) for all \(U\in [0,U_{max})\). Hence, for all \(U\in [0,U_{max})\), \(\rho (U,v) \rightarrow +\infty \) as \( v \rightarrow +\infty \).
-
2.
Static case: for all \(U\in [0,U_{max})\), \(f^0_R(U) = 0\). Then \(\kappa (U,v) = 1\) as \(\partial _U (\kappa ^{-1})=0\). By (A.12), it means that \(\nu _{\mathcal {CH}_{i^+}}(U)=0\) for all \(U\in [0,U_{max})\) hence \(r_{\mathcal {CH}_{i^+}}\) is constant and by (A.3) \(\varpi _{\mathcal {CH}_{i^+}}\) is constant. Their values are \( r_{\mathcal {CH}_{i^+}} \equiv r_-(M,e)>0\) and \( \varpi _{\mathcal {CH}_{i^+}} \equiv M >0\) by the stability estimates of [47]. This implies that \(\rho \) is constant on \(\mathcal {CH}_{i^+}\): \(\rho (U,v) \rightarrow M-\frac{e^2}{r_-(M,e)}\) as \(v \rightarrow +\infty \) for all \(U\in [0,U_{max})\).
-
3.
Mixed case: \(\int _{0}^{U_T}(f^0_R)^2(U')dU' = 0\) but \(\int _{U_T}^{U}(f^0_R)^2(U')dU' > 0\) for all \(U_T<U<U_{max}\). Indeed, the above statement follows from the same logic as for the dynamical case, see earlier discussion. Then similarly, \(r_{\mathcal {CH}_{i^+}}(U)= r_-(M,e)\), \(\varpi _{\mathcal {CH}_{i^+}}(U)= M\), and \(\rho _{\mathcal {CH}_{i^+}}(U)= M-\frac{e^2}{r_-(M,e)}\) for all \(U \in [0,U_T]\). Yet, \(\kappa ^{-1}(U,v) \rightarrow +\infty \) for all \(U \in (U_T,U_{max})\), as \(\int _{0}^U (f_R^0(U'))^2dU' >0\). Thus \(\nu _{\mathcal {CH}_{i^+}}(U)<0\) for all \(U \in (U_T,U_{max})\) and \(r_{\mathcal {CH}_{i^+}}\) is strictly decreasing on \((U_T,U_{max})\). If additionally we have a lower bound \(f_L^0(v)=C \cdot v^{-s}+o(v^{-s})\), then we have the blow up of \(\rho \), as in the dynamical case.
\(\square \)
Remark A.2
Notice that the proof was considerably easier for the uncharged dust than for a charged scalar field, as we used in this section some special monotonicity properties that are not exploitable in the charged case. In contrast, in the earlier sections, we proved estimates that were harder to obtain but also more robust as they do not rely on monotonicity.
Rights and permissions
About this article
Cite this article
Van de Moortel, M. Mass Inflation and the \(C^2\)-inextendibility of Spherically Symmetric Charged Scalar Field Dynamical Black Holes. Commun. Math. Phys. 382, 1263–1341 (2021). https://doi.org/10.1007/s00220-020-03923-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00220-020-03923-w