1: \documentclass[12pt]{article}
2: \usepackage[dvips]{graphics}
3: \usepackage{graphicx}
4: %\documentstyle[12pt]{article}
5: %% \documentstyle{article}
6: % \documentstyle[twocolumn]{article}
7: \setlength{\textwidth}{6.2in}
8: \setlength{\textheight}{9.0in}
9: \setlength{\oddsidemargin}{0.0in}
10: \setlength{\evensidemargin}{\oddsidemargin}
11: \setlength{\topmargin}{-0.7in}
12: % \linespread{1.5} % samo za latex2e
13: \title{Nonlocal black-hole thermodynamics and massive remnants}
14: \author{Hrvoje Nikoli\'c \\
15: Theoretical Physics Division, Rudjer Bo\v{s}kovi\'{c} Institute, \\
16: P.O.B. 180, HR-10002 Zagreb, Croatia \\
17: {\normalsize hrvoje@thphys.irb.hr} \\
18: \makebox[1in]{} \\
19: }
20: \date{\today}
21: %-----------------------------------------------------------------------
22: \begin{document}
23: \maketitle
24: \begin{abstract}
25: To alleviate the black-hole (BH) information problem,
26: we study a holographic-principle-inspired nonlocal model
27: of Hawking radiation in which radiated particles created at different
28: times all have the same temperature
29: corresponding to the instantaneous BH mass. Consequently,
30: the black hole loses mass not only by continuously radiating
31: new particles, but also by continuously warming previously radiated
32: particles. The conservation of energy implies that the radiation
33: stops when the mass of the black hole reaches the half of the initial
34: BH mass, leaving a massive BH remnant
35: with a mass much above the Planck scale.
36: \end{abstract}
37: \vspace*{0.5cm}
38: PACS: 04.70.Dy \newline
39: Keywords: Hawking Radiation; Black-Hole Remnant; Nonlocality
40: \vspace*{0.9cm}
41:
42: \noindent
43: There are two (not necessarily independent) problems related to the
44: information contained in black holes. The first one is to explain
45: why the information contained in the black hole is proportional
46: to its surface, rather than to its volume. The second one is to reconcile
47: the process of Hawking radiation with the principle of unitary
48: evolution, according to which information cannot be destroyed.
49:
50: A paradigm introduced as a theoretical
51: framework for dealing with the first problem is
52: the holographic principle \cite{hooft,sussk,bouso},
53: according to which the information
54: that can be stored inside the black hole is determined by its boundary
55: - the black-hole (BH) horizon.
56: One immediate consequence of the holographic principle is
57: {\em nonlocality} - in some way, the degrees of freedom inside the black hole
58: should know about the boundary. To explain the holographic principle,
59: one must go beyond local quantum field theory. There are indications
60: that string theory possesses certain nonlocal features
61: (see e.g. \cite{mald,seib,hor,nikolstring}), but the fact is
62: that there is not yet a general well-understood theory of
63: holography. Instead, the holographic principle often serves merely as a
64: guiding principle in construction of physical models.
65: %(see e.g. \cite{capl,nikolcosm1,nikolcosm2}).
66:
67: A possible solution of the second problem is a BH remnant
68: scenario, according to which the process of Hawking radiation
69: stops before the black hole evaporates completely, so that
70: all the BH information can be contained in the BH remnant.
71: A problem with the BH remnant scenario is to find a physical mechanism
72: that stops the Hawking radiation. If this mechanism is an effect of
73: quantum gravity, one generically expects that the significant
74: deviation from semiclassical gravity occurs at the Planck scale,
75: so that the mass and radius of the remnant are of the order of the
76: Planck mass and Planck distance, respectively. Such a light remnant that
77: can contain a huge amount of information is problematic
78: \cite{harv,gid,strom} because
79: light objects that can exist in a huge number of different states
80: are expected to be often produced in various physical processes,
81: which is not seen in nature. Thus, a {\em massive remnant}
82: \cite{gidrem} (i.e., a remnant with a mass much larger than
83: the Planck mass) seems to be a more attractive possibility, but the problem
84: is to find a physical mechanism that makes remnants so massive.
85: One possibility is that the light remnant cannot exist without
86: the Hawking radiation entangled with the BH interior \cite{nikolbh},
87: so that the total system - the black hole together
88: with its Hawking radiation - is not light at all.
89: However, it would be more appealing if the BH remnant itself
90: would remain massive.
91:
92: The purpose of the present work is to suggest a possible nonlocal
93: physical mechanism that could stop the process of
94: Hawking radiation much before
95: reaching the Planck scale, thus leaving a massive BH remnant that can store
96: the BH information. The mechanism we suggest
97: is inspired by the holographic principle and partially
98: by the suggestion \cite{gidnl} that there should
99: exist a nonlocal connection between the degrees of freedom in the
100: BH interior and those outside of the black hole.
101: The holographic principle suggests that the degrees of freedom
102: in the BH interior are determined by its boundary - the BH horizon.
103: On the other hand, if there is also a nonlocal relation between
104: some exterior degrees of freedom with those in the BH interior,
105: then these exterior degrees of freedom could also be determined
106: by the same BH boundary. However, it does not seem reasonable to
107: expect that {\em all} exterior degrees of freedom are determined
108: by a single BH boundary. (There may be a lot of black holes in the
109: universe, so one particular black hole cannot play a preferred
110: role for all exterior degrees of freedom.) Instead,
111: neglecting the interaction of the Hawking radiation with other
112: exterior degrees of freedom, we assume
113: that only the Hawking radiation radiated from this particular
114: black hole remains in a nonlocal contact with its boundary.
115: The horizon relevant to the process of Hawking radiation
116: is the apparent horizon \cite{viss1,viss2}, which, in the case of Hawking
117: radiation, is a time-dependent object. The time-dependent
118: temperature $T(t)$ associated with the time-dependent
119: apparent horizon is determined by the time-dependent BH mass $M(t)$,
120: through the relation \cite{hawk1,bd}
121: \begin{equation}\label{temp}
122: T=\frac{1}{8\pi M},
123: \end{equation}
124: where we use units $\hbar=c=G_{\rm N}=k_{\rm B}=1$.
125: In the standard semiclassical analysis of the process of
126: BH radiation, when particles
127: are radiated away from the black hole, then the temperature of
128: these particles does not change with subsequent BH evolution.
129: Instead, if these particles do not interact with other exterior degrees
130: of freedom, their temperature remains the same, despite the fact that
131: later BH temperature may change. Indeed, this is a consequence of
132: locality, according to which radiated particles cannot know
133: about possible later changes of the BH temperature.
134: However, if radiated particles remain in a nonlocal contact
135: with its source - the evolving apparent horizon -
136: then it seems reasonable to assume that this nonlocal contact
137: could manifest as a nonlocal thermodynamic system
138: in which {\em all} radiated
139: particles have the {\em same} temperature (\ref{temp}) given by the
140: instantaneous BH mass $M(t)$.
141: Thus, in addition to the standard backreaction of Hawking radiation
142: on the black hole (owing to which the BH mass
143: decreases, such that the total
144: energy measured by a distant observer is conserved),
145: there is an additional nonlocal backreaction of the horizon on
146: the previously radiatied particles,
147: owing to which the temperature of these radiated particles
148: increases.
149:
150: Now let us see how such a nonlocal backreaction leads to massive
151: BH remnants.
152: The energy of all radiated particles at the time $t$ is
153: \begin{equation}\label{e1}
154: E=N\epsilon ,
155: \end{equation}
156: where $N$ is the total number of radiated particles at the time $t$
157: and $\epsilon$ is the
158: average energy per particle. According to our assumption,
159: all radiated particles have the same
160: temperature $T(t)$ at a given time $t$,
161: so, for massless particles in a thermal equilibrium,
162: \begin{equation}\label{e2}
163: \epsilon=bT,
164: \end{equation}
165: where $b\sim 1$ is a dimensionless parameter that depends on the
166: spin of the particles.
167: Since the energy must be conserved, at each time $t$ there must be
168: \begin{equation}\label{cons}
169: M+N\epsilon=M_0,
170: \end{equation}
171: where $M_0$ is the initial BH mass corresponding to $N=0$.
172: Inserting (\ref{e2}) and (\ref{temp}) into (\ref{cons}), we obtain
173: \begin{equation}\label{qe}
174: N=\frac{(M_0-M)M}{b'} ,
175: \end{equation}
176: where $b'\equiv b/8\pi$.
177: Eq.~(\ref{qe}) can also be viewed as a quadratic equation for $M$,
178: with the general solution
179: \begin{equation}\label{qsol}
180: M=\frac{M_0}{2} \pm \sqrt{ \left( \frac{M_0}{2} \right)^2 -b'N } .
181: \end{equation}
182: The upper sign is the correct one consistent with the
183: requirement that $M=M_0$ when $N=0$. During the evolution
184: in time, $N$ increases while $M$ decreases, until the number
185: of radiated particles attains the critical value
186: \begin{equation}\label{Nc}
187: N_{\rm crit}=\frac{M_0^2}{4b'},
188: \end{equation}
189: which corresponds to the vanishing
190: square root in (\ref{qsol}). At this moment, the mass
191: of the black hole attains the critical value
192: \begin{equation}\label{Mc}
193: M_{\rm crit}=\frac{M_0}{2} .
194: \end{equation}
195: Can the mass further decrease after reaching this critical value?
196: Mathematically, it would be possible only by taking the lower
197: sign in (\ref{qsol}) at times after
198: the mass has reached the critical value. However,
199: in this case, after this critical moment of time,
200: $N$ should start to decrease.
201: This would correspond to an inverted process of Hawking radiation,
202: in which the outgoing Hawking particles suddenly reverse their direction
203: of motion, thus becoming ingoing particles that eventually become
204: destroyed when they approach the horizon.
205: A sudden inversion of the direction of motion does not seem to be
206: physical, so we conclude that
207: such a further decrease of mass is unphysical. In other words,
208: the critical mass (\ref{Mc}) is the {\em smallest possible mass}
209: of the black hole. The final state of BH radiation is
210: a massive remnant with the mass
211: equal to the half of the initial BH mass.
212: Clearly, such a massive remnant solves the problem of
213: unitary evolution without leading to overproduction
214: of light objects that can store a huge amount of information.
215: In addition, such massive BH remnants could be responsible
216: for the existence of dark matter in the universe.
217:
218: Now let us study the time evolution of such a radiating black hole.
219: The change of the radiation energy $E$ is equal
220: to the negative change of the BH mass $M$, i.e.
221: $dE=-dM$. Therefore, (\ref{e1}) implies
222: %The energy-conservation law (\ref{cons}) implies
223: \begin{equation}\label{e3}
224: -dM=\epsilon dN +N d\epsilon.
225: \end{equation}
226: Note that the second term on the right-hand side of (\ref{e3}) is
227: absent in the standard approach, because, in the
228: standard approach, one assumes that
229: the temperature of the fixed number of
230: radiated particles does not change, so that $d\epsilon =bdT=0$ in (\ref{e3}).
231: In our approach, from (\ref{e2}) and (\ref{temp}) we find
232: \begin{equation}\label{Ndif}
233: N d\epsilon =\frac{-b'N}{M^2}dM .
234: \end{equation}
235: The first term on the right-hand side of (\ref{e3})
236: can be calculated in the same
237: way as in the standard scenario. Assuming that the black hole
238: radiates as a perfect black body, we use the Stefan-Boltzmann law
239: \begin{equation}\label{SB}
240: \epsilon dN=\sigma A T^4 dt ,
241: \end{equation}
242: where $\sigma=\pi^2/60$ is the Stefan-Boltzmann constant
243: and $A=4\pi R^2=16\pi M^2$ is the BH surface. Thus,
244: (\ref{SB}) with (\ref{temp}) can be written as
245: \begin{equation}\label{SB1}
246: \epsilon dN=\frac{\sigma'}{M^2}dt ,
247: \end{equation}
248: where $\sigma'\equiv\sigma/256\pi^3$. Eq.~(\ref{SB1}) with (\ref{e2})
249: and (\ref{temp}) can be written as
250: the differential equation
251: \begin{equation}\label{Nt}
252: \frac{dN}{dt}=\frac{\sigma'}{b'}\frac{1}{M} .
253: \end{equation}
254: Similarly, by inserting (\ref{Ndif}) and (\ref{SB1}) into (\ref{e3}),
255: we obtain another differential equation
256: \begin{equation}\label{Mt}
257: \frac{dM}{dt}=\frac{-\sigma'}{M^2-b'N}.
258: \end{equation}
259: Eqs.~(\ref{Nt}) and (\ref{Mt}) represent a system of two
260: coupled equations that determine the functions $N(t)$ and
261: $M(t)$, with the initial conditions
262: $N(0)=0$ and $M(0)=M_0$. From (\ref{Nc}), (\ref{Mc}) and
263: (\ref{Mt}), we see that $dM/dt$ diverges at the critical values
264: of $N$ and $M$. For a numerical analysis of the system
265: of equations (\ref{Nt}) and (\ref{Mt}), it is convenient
266: to introduce the rescaled variables
267: \begin{equation}\label{scale}
268: \tau=\frac{\sigma'}{M_0^3}t, \;\;\;
269: m=\frac{M}{M_0}, \;\;\;
270: n=\frac{b'}{M_0^2}N .
271: \end{equation}
272: Eqs.~(\ref{Nt}) and (\ref{Mt}) then become
273: \begin{equation}\label{nmt}
274: \frac{dn}{d\tau}=\frac{1}{m} ,\;\;\;\;
275: \frac{dm}{d\tau}=\frac{-1}{m^2-n} ,
276: \end{equation}
277: with the initial conditions $n(0)=0$, $m(0)=1$,
278: which does not involve numerical constants that are not
279: of the order of unity.
280: Eqs.~(\ref{Nt}) and (\ref{Mt}) can also be solved analytically.
281: From (\ref{Nt}) and (\ref{Mt}) one obtains the differential equation
282: $dN/dM=(N-M^2/b')/M$, the solution of which is given by (\ref{qe}).
283: Thus, by inserting (\ref{qe}) into (\ref{Mt}),
284: we obtain a decoupled differential equation
285: \begin{equation}\label{Mtd}
286: \frac{dM}{dt}=\frac{-\sigma'}{2M^2-M_0 M} .
287: \end{equation}
288: This is easily integrated to give
289: \begin{equation}\label{Mtsol}
290: \frac{2M^3}{3}-\frac{M_0M}{2}-\frac{M_0^3}{6}=-\sigma't ,
291: \end{equation}
292: where the requirement $M(t=0)=M_0$ is incorporated.
293: This, together with (\ref{Mc}), determines the time needed
294: for the black hole to approach the critical mass:
295: \begin{equation}\label{tc}
296: t_{\rm crit}=\frac{5}{24}\frac{M_0^3}{\sigma'} .
297: \end{equation}
298:
299: Let us compare the results above with those in the standard
300: semiclassical scenario of BH radiation.
301: In this case, there is no second term on the right hand side of
302: (\ref{e3}), so (\ref{Mt}) is replaced by a simpler equation
303: $dM/dt=-\sigma'/M^2$. Thus, instead of (\ref{Mtsol}), we obtain
304: a simpler solution
305: \begin{equation}\label{Mts}
306: M=\sqrt[3]{M_0^3-3\sigma't}.
307: \end{equation}
308: The critical point at which $dM/dt$ diverges corresponds to
309: $M_{\rm crit}=0$ and $t_{\rm crit}=(1/3)M_0^3/\sigma'$.
310: The number of radiated particles satisfies (\ref{Nt}) which
311: is easily integrated to give
312: \begin{equation}\label{Nts}
313: N=\frac{M_0^2-(M_0^3-3\sigma't)^{2/3}}{2b'} ,
314: \end{equation}
315: leading to $N_{\rm crit}=M_0^2/2b'$. Comparing it with
316: (\ref{Nc}), we see that the number of particles produced
317: in the massive remnant scenario is equal to the half of that
318: in the standard scenario.
319: Combining (\ref{Mts}) with (\ref{Nts}) one obtains
320: \begin{equation}
321: M=\sqrt{ M_0^2 -2b'N } ,
322: \end{equation}
323: which is to be compared with (\ref{qsol}).
324: The time evolutions of the BH mass in the two scenarios
325: (Eqs.~(\ref{Mtsol}) and (\ref{Mts})) are compared in Fig.~1.
326: \begin{figure}[h]
327: %\includegraphics[width=3.5cm]{figfol1}
328: \includegraphics{figmbh}
329: \caption{\label{fig1}
330: The BH mass $M/M_0$ as a function of the rescaled time $\tau$.
331: The solid curve represents the massive remnant scenario, while the
332: dotted curve represents the standard scenario.}
333: \end{figure}
334:
335: As we have seen, the assumption that, at each time, all radiated
336: particles have the same temperature given by the instantaneous
337: BH mass leads to massive BH remnants. Such remnants offer
338: a solution to the problem of unitarity of BH evolution.
339: The main problem with such a scenario is to find an
340: independent justification of such an assumption
341: that clearly contradicts the principle of locality.
342: In particular, this assumption requires a preferred notion
343: of the time coordinate, but note that this preferred
344: time coordinate is the same preferred time coordinate as the one
345: needed to define particles in the standard description
346: of BH radiation \cite{bd}, which, in an ideal case,
347: can be identified with the Killing time.
348: (Note also that the definition of particles in
349: quantum field theory is intrinsically {\em nonlocal} \cite{bd},
350: even when formulated in terms of local particle currents \cite{nikcur}.)
351: We have argued that such a nonlocal thermodynamic behaviour
352: could be related to the holographic principle, but a
353: clear derivation of this relation is missing. Nevertheless,
354: the fact that such a simple assumption leads to such
355: a simple (and surprising!) solution to the problem of BH unitarity
356: leads us to suspect that this assumption could be on the right track.
357: Hopefully, the future research could reveal a more
358: compelling explanation of our assumption.
359:
360:
361: \section*{Acknowledgements}
362:
363: This work was supported by the Ministry of Science and Technology of the
364: Republic of Croatia.
365:
366: \begin{thebibliography}{99}
367:
368: \bibitem{hooft}
369: G. 't Hooft, gr-qc/9310026.
370:
371: \bibitem{sussk}
372: L. Susskind, J. Math. Phys. 36 (1995) 6377.
373:
374: \bibitem{bouso}
375: R. Bousso, Rev. Mod. Phys. 74 (2002) 825.
376:
377: %----------------
378: \bibitem{mald}
379: J. M. Maldacena, Adv. Theor. Math. Phys. 2 (1998) 231.
380:
381: \bibitem{seib}
382: N. Seiberg, hep-th/0601234.
383:
384: \bibitem{hor}
385: G. T. Horowitz, New J. Phys. 7 (2005) 201.
386:
387: \bibitem{nikolstring}
388: H. Nikoli\'c, hep-th/0605250.
389:
390: %-----------------
391: %\bibitem{capl}
392: %A. G. Cohen, D. B. Kaplan, A. E. Nelson,
393: %Phys. Rev. Lett. 82 (1999) 4971.
394: %
395: %\bibitem{nikolcosm1}
396: %B. Guberina, R. Horvat, H. Nikoli\'c,
397: %Phys. Rev. D72 (2005) 125011.
398: %
399: %\bibitem{nikolcosm2}
400: %B. Guberina, R. Horvat, H. Nikoli\'c,
401: %Phys. Lett. B636 (2006) 80.
402: %
403: %----------------
404: \bibitem{harv}
405: J.~A.~Harvey, A.~Strominger, hep-th/9209055.
406:
407: \bibitem{gid}
408: S.~B.~Giddings, hep-th/9412138.
409:
410: \bibitem{strom}
411: A.~Strominger, hep-th/9501071.
412:
413: %----------------
414: \bibitem{gidrem}
415: S.~B.~Giddings, Phys. Rev. D 46 (1992) 1347.
416:
417: \bibitem{nikolbh}
418: H.~Nikoli\'c, Int. J. Mod. Phys. D 14 (2005) 2257,
419: Honorable Mention of the Gravity Research Foundation 2005 Essay Competition.
420:
421: \bibitem{gidnl}
422: S.~B.~Giddings, hep-th/0605196.
423:
424: \bibitem{viss1}
425: M.~Visser, Phys.~Rev.~Lett.~80 (1998) 3436.
426:
427: \bibitem{viss2}
428: M.~Visser, Int.~J.~Mod.~Phys.~D 12 (2003) 649.
429:
430: \bibitem{hawk1}
431: S.~Hawking, Commun.~Math.~Phys.~43 (1975) 199.
432:
433: \bibitem{bd}
434: N.~D.~Birrell, P.~C.~W.~Davies, Quantum Fields in
435: Curved Space, Cambridge Press, NY, 1982.
436:
437: \bibitem{nikcur}
438: H. Nikoli\'c, Phys. Lett. B 527 (2002) 119; \\
439: H. Nikoli\'c, Int. J. Mod. Phys. D 12 (2003) 407; \\
440: H. Nikoli\'c, Gen. Rel. Grav. 37 (2005) 297.
441:
442: \end{thebibliography}
443:
444: \end{document}