1:
2: \documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
3: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb,nofootinbib]{revtex4}
4:
5:
6: \usepackage{graphicx}% Include figure files
7:
8: \begin{document}
9:
10: \preprint{DF/IST-12.2003}
11:
12: \title{
13: Electromagnetic radiation from collisions at almost the
14: speed of light: an extremely relativistic charged
15: particle falling into a
16: Schwarzschild black hole
17: }
18:
19: \author{Vitor Cardoso}
20: \email{vcardoso@fisica.ist.utl.pt}
21: \affiliation{Centro Multidisciplinar de Astrof\'{\i}sica - CENTRA,
22: Departamento de F\'{\i}sica, Instituto Superior T\'ecnico,
23: Av. Rovisco Pais 1, 1049-001 Lisboa, Portugal,}
24:
25: \author{Jos\'e P. S. Lemos}
26: \email{lemos@physics.columbia.edu}
27: \affiliation{
28: Department of Physics, Columbia University, New York, NY 10027,
29: USA, \&
30: Centro
31: Multidisciplinar de Astrof\'{\i}sica - CENTRA, Departamento de
32: F\'{\i}sica, Instituto Superior T\'ecnico, Av. Rovisco Pais 1,
33: 1049-001 Lisboa, Portugal,}
34:
35: \author{Shijun Yoshida}
36: \email{yoshida@fisica.ist.utl.pt}
37: \affiliation{
38: {}Centro Multidisciplinar de Astrof\'{\i}sica - CENTRA,
39: Departamento de F\'{\i}sica, Instituto Superior T\'ecnico,
40: Av. Rovisco Pais 1, 1049-001 Lisboa, Portugal.}
41:
42:
43: \date{\today}
44:
45: \begin{abstract}
46: We investigate the electromagnetic radiation released during the high
47: energy collision of a charged point particle with a four-dimensional
48: Schwarzschild black hole. We show that the spectra is flat, and well
49: described by a classical calculation. We also compare the total
50: electromagnetic and gravitational energies emitted, and find that the
51: former is supressed in relation to the latter for very high energies.
52: These results could apply to the astrophysical world in the case
53: charged stars and small charged black holes are out there colliding
54: into large black holes, and to a very high energy collision experiment
55: in a four-dimensional world. In this latter scenario the calculation
56: is to be used for the moments just after the black hole formation,
57: when the collision of charged debris with the newly formed black hole
58: is certainly expected. Since the calculation is four-dimensional, it
59: does not directly apply to Tev-scale gravity black holes, as these
60: inhabit a world of six to eleven dimensions, although our results
61: should qualitatively hold when extrapolated with some care to higher
62: dimensions.
63: \end{abstract}
64:
65: \pacs{04.70.Bw, 04.30.Db}
66: %\keywords{Suggested keywords}
67:
68: \maketitle
69: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
70: \section{Introduction}
71: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
72:
73: In a previous paper we have investigated the gravitational radiation
74: emitted by a relativistic particle infalling into a four-dimensional
75: Schwarzschild black hole \cite{cardosolemos0}, and afterwards extended
76: it to the infall in a Kerr metric spacetime \cite{cardosolemos1}. The
77: interest of these studies lies in the fact they can describe the
78: gravitational radiation emitted in several collision phenomena, such
79: as the collision between small and massive black holes, between stars
80: and massive black holes, or between cosmic rays and small black holes,
81: to name a few. The possibility of detecting gravitational radiation
82: by the several operating antennae is now real, and a closer
83: understanding of these collisional processes is accordingly important.
84: Of course, to fully understand collisional processes, one has in
85: principle to go
86: beyond the infall of a particle and consider the collision between
87: two black holes with comparable masses, in which case one has to
88: take into account the full non-linearity and strong field regime of
89: Einstein's equations. Without resorting to numerical computations,
90: but instead using topological arguments, Hawking was able
91: to put upper limits on the gravitational radiation emitted in
92: a black hole - black hole collsion in four dimensions. This
93: calculation was then refined by D'Eath and Payne \cite{payne}.
94: Their calculation made use of the fact that, if one boosts the
95: Schwarzschild metric to high velocities, then it approaches the
96: Aichelburg-Sexl \cite{aichelburg} metric, which is a shock wave
97: spacetime describing the gravitational field of a massless particle.
98: It was found that the total efficiency of the shock
99: would be $\sim 16\%$ (see Yoshino and Nambu \cite{yoshino}
100: for the analysis in higher dimensions). In \cite{payne},
101: the inclusion of a second term in the news function
102: brings about a decrease in the efficiency from
103: $25\%$ (using only the first term) to $16\%$,
104: so one is entitled to ask what degree of confidence
105: do these results and techniques offer. For instance,
106: one might suspect the third term to lower
107: even more the efficiency. One needs to have other means
108: of computing this process, since D'Eath and Payne's
109: formalism is hard to pursue, due to its complexity. We have
110: recently \cite{cardosolemos0,cardosolemos1} proposed the ``point
111: particle paradigm'', which may well be a good candidate. Handling
112: high energy collisions of point particles with black holes is rather
113: simple, since the point particle may be looked at as a perturbation in
114: the black hole spacetime, and the full machinery to handle black hole
115: perturbations is well developed \cite{bhperturbation}. The
116: interesting thing about the high energy collision of point particles
117: with black holes is that in the limit that the mass $\mu$ of the
118: particle goes to the mass $M$ of the black hole, the result for the
119: total radiated gravitational energy agrees well with other predictions
120: and never violates the area theorem
121: \cite{cardosolemos0,cardosolemos1}. Moreover, one can use this method
122: to treat the collision of rotating holes, which does not seem feasible
123: using previous techniques.
124:
125: Up to now, most studies have concentrated on the emission of
126: gravitational radiation in the collision process, but if at least
127: one of the particles is charged, electromagnetic radiation is also
128: expected. Astrophysically this is perhaps less relevant for the time
129: being, since the particles one can consider, such as stars or small
130: black holes, are in principle discharged, although it is possible the
131: latter ones can carry some charge (see, e.g., \cite{malheiro}). On the
132: other hand, from the point of view of elementary black holes such a
133: study is certainly of interest: elementary black holes and
134: particles can easily carry electromagnetic charge, and the collision
135: between such objects can be put to operate, either by a machine in
136: some far future that can collide particles at center of mass energies
137: of the order of the four-dimensional Planck mass, $10^{16}\,$TeV, or
138: if there is any fundament in the TeV-scale gravity, by the near future
139: generation accelerators such as the LHC. The latter case has been
140: given a lot of consideration. The TeV-scale gravity \cite{hamed}
141: requires extra large dimensions (of the order of sub-millimeters or
142: smaller) in order to lower the higher-dimensional Planck mass to
143: energies of the order of TeVs. In these TeV-scale gravity models, the
144: gravitons are free to propagate in the higher dimensional spacetime,
145: whereas, due to experimental constraints, the standard model fields
146: live on a 3-brane, our universe. In this scenario, a particle
147: collider with a center of mass energy of the order of TeVs can
148: copiously produce higher dimensional small spherical type black holes,
149: with radius of the order of fermis or less, in a space with large
150: extra dimensions \cite{bhprod}, or even black branes if some of the
151: extra dimensions are large and other small (see \cite{cav} for a
152: review). Moreover, in the collision process the gravitons
153: escape to the extra dimensions and are therefore much harder, harder than
154: usual, to detect. On the other hand, all the electromagnetic radiation
155: emitted in such a collision can be detected.
156:
157: Thus it is important to know the electromagnetic spectrum and the
158: quantity of electromagnetic energy radiated in such an encounter. In
159: this paper we extend the previous calculations into to the
160: electromagnetic window and find the electromagnetic radiation emitted
161: by a highly relativistic electrically charged particle infalling into
162: a four-dimensional Schwarzschild black hole. This is a perturbation
163: calculation: we suppose a small charged particle falling into a
164: Schwarzschild black hole. Some former works that have dealt with the
165: phenomenon of electromagnetic radiation from a charged particle
166: falling from infinity into a Schwarzschild black hole are
167: \cite{zerilli1,ruffini72,ruffini74,chandra}. For example Ruffini
168: \cite{ruffini72} first calculated the electromagnetic energy spectra
169: and total radiated energy \cite{ruffini72}, for particles with low
170: Lorentz factors, $\gamma<<1$. Here we go into high $\gamma$s. We use
171: the point particle approximation and consider a charged point particle
172: colliding head-on at high velocity with a Schwarzschild black hole.
173: We show that the spectra is flat, and we also compare the total
174: electromagnetic and gravitational energies emitted, and find that the
175: former is supressed in relation to the latter for very high energies.
176: The numerical results extracted by us are in very good agreement with
177: Ruffini's results \cite{ruffini72}. We shall also see that there is a
178: classical calculation for this process that agrees extremely well with
179: our numerical results.
180:
181: Two comments are in order:
182: (i) Our calculation is for the collision of a small particle with a
183: black hole in a four-dimensional world. Therefore, in principle, it
184: could apply to the astrophysical world in the case charged stars and
185: charged black holes are out there, and to a very high energy collision
186: experiment in a four-dimensional world (without extra dimensions) and
187: with a fundamental Planck mass of $10^{16}\,$TeV. In this latter
188: scenario the calculation is to be used for the moments just after the
189: black hole formation, when the collision of charged debris with the
190: newly formed black hole is certainly expected. Furthermore, the
191: calculation can be extended to a black hole - black hole collision as
192: we have already argued \cite{cardosolemos0,cardosolemos1}, and, in
193: addition, it can give clues, although it does not aplly directly to the
194: usual collision process in a collider, i.e., the collision between one
195: non-charged particle with strong gravitational field (not necessarily
196: a black hole) and one charged particle.
197: These results may, however, serve as
198: a model to the electromagnetic radiation emitted in the initial phase
199: of a newly formed black hole, the stage
200: in which the black hole sheds its hair by emitting
201: gauge radiation, such as electromagnetic radiation.
202: (ii) Our calculation is four-dimensional. It does not apply to
203: Tev-scale gravity black holes, since these inhabit a world of six to
204: eleven dimensions, with some of the dimensions being large, others
205: perhaps being small. However, although there
206: are certainly some differences, qualitatively our results should
207: hold when extrapolated with some care to higher dimensions.
208:
209:
210:
211:
212: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
213: \noindent
214: \section{Basic Formalism}
215: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
216: In this study, we assume the charged particle and the emitted radiation
217: to be a small perturbation on the Schwarzschild spacetime, whose line
218: element is given by
219: %
220: \begin{eqnarray}
221: ds^2=
222: -e^\nu\,dt^2+e^\lambda\,dr^2+r^2(d\theta^2+\sin^2\theta\,d\varphi^2) \,,
223: \label{metric}
224: \end{eqnarray}
225: %
226: where $e^\nu=e^{-\lambda}=1-2M/r$, and $M$ stands for the mass of the
227: background spacetime. Because a complete prescription to deal with
228: this problem was given by Zerilli \cite{zerilli1}, here we briefly show
229: the essential parts of the formalism.
230: %
231: In virtue of the static property and spherical symmetry of the background
232: spacetime, perturbations can be decomposed by using vector harmonics
233: for the angular variables $\theta$ and $\varphi$, and the Fourier components
234: for the time variable $t$. After accomplishing a separation of variables,
235: perturbations are therefore characterized by
236: the harmonic indices $l$ and $m$, parity, and frequency $\omega$.
237: For the present case, where a particle
238: falls straight into a black hole along the $z$-axis or the $\theta=0$ line,
239: no axial parity perturbations are excited due to the symmetry of the motion.
240: Thus, only the polar parity perturbations are considered in this study.
241: According to Zerilli \cite{zerilli1}, our master equation for determining
242: the electromagnetic radiation is given by a single wave equation,
243: %
244: \begin{equation}
245: \frac{d^2 \tilde f_{lm}(\omega,r)}{ dr_*^2}+
246: \left[\omega^2-e^\nu\,\frac{l(l+1)}{ r^2}\right]\tilde f_{lm}(\omega,r)=
247: e^\nu \tilde S_{lm}\,,
248: \label{waveequation}
249: \end{equation}
250: where $r_*$ is the so-called tortoise coordinate, defined by
251: $dr/dr_*=e^{(\nu-\lambda)/2}$, or $r_*=r+2M\log(r/2M-1)$ in our case,
252: and $\tilde S_{lm}$ is the source term determined by the charge and
253: motion of the particle. In the case of the radially falling particle,
254: $\tilde S_{lm}$ is generally given by
255: %
256: \begin{equation}
257: \tilde S_{lm}=-2q\sqrt{l+\frac{1}{2}}\,
258: \frac{e^{i\omega T(r)}}{ r^2}\,\delta_{m0}\,,
259: \end{equation}
260: %
261: where $\delta_{ij}$ means the Kronecker delta, and $q$ denotes the
262: charge of the particle. Here, the function $T(r)$ is the coordinate
263: time of the particle parameterized by the radial position $r$ of the
264: particle. Since the charged particle is treated as a perturbation of
265: the background spacetime, the particle traces a radial geodesic of the
266: Schwarzschild geometry. Thus, the function $T(r)$ is determined by
267: (see e.g., Chandrasekhar \cite{chandra})
268: %
269: \begin{equation}
270: \frac{dT(r)}{ dr}=-\frac{e^{(\lambda-\nu)/2}}{\sqrt{1-\gamma^{-2}\,e^\nu}}\,,
271: \end{equation}
272: %
273: where $\gamma$ is an integral of motion, given by
274: $\gamma=(1-v_\infty^2)^{-1/2}$, where $v_\infty$ is the radial
275: velocity of the particle at spatial infinity.
276:
277: Once solutions $\tilde f_{lm}(\omega,r)$ of equation (\ref{waveequation}) are
278: calculated, time dependent functions $f_{lm}(t,r)$ can be obtained through
279: the inverse Fourier transformation,
280: %
281: \begin{equation}
282: f_{lm}(t,r)= \frac{1} {\sqrt{2\pi}}\int_{-\infty}^{\infty}e^{-i \omega t}
283: \tilde f_{lm}(\omega,r)d\omega\,.
284: \label{inversetransform}
285: \end{equation}
286: %
287: From the functions $f_{lm}(t,r)$, all the components of the field strength
288: $F_{\alpha\beta}$ can be derived except at the position of the charged
289: particle. For example, the $r$--$\theta$ component $F_{r\theta}$ is
290: given by
291: %
292: \begin{equation}
293: F_{r\theta}(t,r,\theta,\varphi)=e^{-\nu}\sum_{l,m}f_{lm}(t,r)
294: \frac{\partial Y_{lm}}{ \partial\theta}(\theta,\varphi)\,.
295: \end{equation}
296:
297: In order to obtain a unique solution of equation (\ref{waveequation}),
298: two boundary conditions must be specified, and physically acceptable
299: conditions are purely outgoing wave at spatial infinity and purely
300: incoming wave at the horizon, which are due to equation
301: (\ref{inversetransform}) given by
302: %
303: \begin{equation}
304: \tilde f_{lm}(r)\rightarrow \left\{
305: \begin{array}{ll}
306: A_{lm}^{\rm in}(\omega)\, e^{-i\omega r_*} & {\rm as}
307: \quad r_*\rightarrow -\infty\,,\\
308: A_{lm}^{\rm out}(\omega)\, e^{i\omega r_*} & {\rm as}
309: \quad r_*\rightarrow \infty \,.
310: \end{array}
311: \right.
312: \end{equation}
313: %
314: where $A_{lm}^{\rm out}(\omega)$ and $A_{lm}^{\rm in}(\omega)$ are not
315: dependent on $r$. Since we are interested only in solutions at
316: spatial infinity, what we have to do is to obtain $A_{lm}^{\rm
317: out}(\omega)$'s as functions of $\omega$ for our purpose of this
318: study. Adapting a standard Green's function technique, we can write
319: $A_{lm}^{\rm out}$ as the integral, given by
320: %
321: \begin{equation}
322: A_{lm}^{\rm out}(\omega)=\frac{1}{ 2i\omega C_l(\omega)}
323: \int_{2M}^{\infty}f_{{\rm L},l}
324: \tilde S_{lm}dr\,,
325: \end{equation}
326: %
327: where $f_{{\rm L},l}$ is a homogeneous solution of equation
328: (\ref{waveequation}) satisfying a boundary condition given by
329: %
330: \begin{equation}
331: f_{{\rm L},l}\rightarrow \left\{
332: \begin{array}{ll}
333: e^{-i\omega r_*} & {\rm as} \quad r_*\rightarrow -\infty\,,\\
334: B_l(\omega)\, e^{i\omega r_*}+C_l(\omega)\, e^{-i\omega r_*} &
335: {\rm as} \quad r_*\rightarrow \infty \,.
336: \end{array}
337: \right.
338: \label{boundary-conditions}
339: \end{equation}
340: %
341: The energy spectrum at spatial infinity is given by
342: %
343: \begin{equation}
344: \frac{dE}{ d\omega}=\sum_l\frac{dE_l }{d\omega}=
345: \sum_l\frac{l(l+1)}{ 2\pi}\,|A_{l0}^{\rm out}(\omega)|^2
346: \quad {\rm for} \quad \omega \ge 0\,,
347: \label{energyspectrumnum}
348: \end{equation}
349: %
350: In order to obtain numerical values of $A_{lm}^{\rm out}(\omega)$, we
351: start the integration of $f_{L,l}$ and $\int_{2M}^rf_{{\rm
352: L},l}\tilde{S}_{lm}dr'$ at $r=2M(1+10^{-6})$ by using a Runge-Kutta
353: method. Those functions are then integrated out to large values of
354: $r$. The integration is stopped if the absolute value of
355: $\int_{2M}^rf_{{\rm L},l}\tilde{S}_{lm}dr'$ converges within the
356: required numerical accuracy, and we simultaneously match $f_{L,l}$
357: with an asymptotic solution satisfying condition
358: (\ref{boundary-conditions}) at spatial infinity, which is given in
359: \cite{chandra}, to obtain a value of $C_l(\omega)$.
360:
361: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
362: \noindent
363: \section{Numerical Results}
364: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
365: %\vskip 1mm
366: \begin{figure}
367: \centerline{\includegraphics[width=8cm,height=8 cm]{f1.eps}}
368: \caption{
369: The electromagnetic energy spectra for the four lowest radiatable
370: multipoles, for a $\gamma=2$ particle falling from
371: infinity into a Schwarzschild black hole. The filled circles on the
372: curves of the energy spectra indicate the frequency of the fundamental
373: quasinormal mode associated with the corresponding harmonic index $l$.
374: }
375: \label{spectragraph}
376: \end{figure}
377:
378: \begin{figure}
379: \centerline{\includegraphics[width=8cm,height=8 cm]{f11.eps}}
380: \caption{
381: The electromagnetic energy spectra for the four lowest radiatable
382: multipoles, for a $\gamma \gg 10$ particle falling from
383: infinity into a Schwarzschild black hole. The filled circles on the
384: curves of the energy spectra indicate the frequency of the fundamental
385: quasinormal mode associated with the corresponding harmonic index $l$.
386: }
387: \label{spectragraphinfty}
388: \end{figure}
389: Following the numerical procedure just outlined, we have computed the
390: energy spectrum for several values of $\gamma$. In Figure 1 we show a
391: typical result of the energy spectrum, here for $\gamma=2$ and for the
392: four lowest radiatable multipoles. In Fig. 2 we show similar results
393: but for a particle with $\gamma \gg 10$.
394: A general feature of the energy
395: spectrum for high energy collisions is that it is flat up to some
396: critical frequency, after which it rapidly (exponentially) decreases
397: to zero. This was also verified for the gravitational energy spectrum
398: resulting from the high-energy collision of a black hole with a point
399: particle \cite{cardosolemos0,cardosolemos1}. This critical frequency is given,
400: in a very good approximation, by the frequency of the fundamental
401: quasinormal mode
402: associated with the corresponding harmonic index $l$. The corresponding
403: quasinormal frequencies are also indicated by a filled circle in Figures
404: 1 and 2.
405: The total energy $\Delta E$ radiated away is given by
406: \begin{equation}
407: \Delta E=\sum_{l=1}^\infty\Delta E_l\,,
408: \label{total-energy}
409: \end{equation}
410: where $\Delta E_l$ is the total radiated energy associated with
411: $l$-pole electromagnetic radiation, which is given by the area under
412: the curve of the energy spectrum $dE_l/d\omega$ (see Figure 1). For
413: the low $\gamma$ cases, the convergence of the infinite series
414: (\ref{total-energy}) is relatively good in the sense that a finite
415: summation of several $\Delta E_l$'s is sufficient to obtain a value of
416: $\Delta E$ with sufficient accuracy. However, the convergence of the
417: total radiated energy gets worse with increasing $\gamma$. This
418: contrasts with the analogous calculation of the gravitational energy
419: radiated in high energy collisions \cite{cardosolemos1} where a
420: rescaling was possible to show that $\Delta E \propto \gamma ^2$.
421: Here we shall see in the next section that the difficulty for the
422: electromagnetic case lies in the fact that $\Delta E \propto
423: \log{\gamma}$, for high $\gamma$'s. For a very fast particle,
424: $\gamma\sim\infty$, it is quite difficult to numerically obtain the
425: total radiated energy, since all the $l$-pole energy spectra
426: $dE_l/d\omega$ make a substantial contribution to the total radiated
427: energy. In this study, we calculated the total energy for several
428: values of $\gamma$ with $\gamma\le 10$. Some of these values are
429: shown in Table 1.
430:
431: %
432: \begin{table}
433: \caption{\label{tab:zfl} The total radiated electromagnetic
434: energy as function of $\gamma$ .}
435: \begin{ruledtabular}
436: \begin{tabular}{c|c}
437: $\gamma$ & $\Delta E\,M q^{-2}$ \\ \hline
438: 1 & $2.14\times 10^{-2}$ \\
439: 2 & $1.14\times 10^{-1}$ \\
440: 3 & $2.23\times 10^{-1}$ \\
441: 4 & $3.34\times 10^{-1}$ \\
442: 5 & $4.46\times 10^{-1}$ \\
443: \end{tabular}
444: \end{ruledtabular}
445: \end{table}
446: %\vskip 1mm
447: %
448: Previous works concerned with the electromagnetic
449: radiation from a charged particle falling from infinity into a Schwarzschild
450: black hole include for example Ruffini \cite{ruffini72}.
451: Ruffini first calculated the
452: electromagnetic energy spectra and total radiated energy \cite{ruffini72},
453: for low
454: $\gamma$ particles.
455: The numerical results extracted by us are in very good agreement with Ruffini's
456: results \cite{ruffini72}.
457: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
458: \noindent
459: \section{The Classical Calculation}
460: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
461: There is a classical calculation in electromagnetism that suggests
462: itself as a model to use in the high energy collision of charged particles:
463: the radiation emitted when a charge, with constant velocity $v_\infty$ is
464: suddenly decelerated to $v=0$. The deceleration is idealized as taking zero
465: seconds. This model calculation has been applied with great success to, for
466: example, beta decay \cite{chang49,jackson}.
467: The result for the energy spectrum per solid angle is \cite{chang49,jackson}
468: \begin{equation}
469: \left( \frac{d^2E}{ d\omega d\Omega}\right)_{\rm class}=
470: \frac{q^2v_\infty^2}{ 4\pi^2}
471: \frac{\sin^2\theta}{(1-v_\infty\cos\theta)^2}\,,
472: \end{equation}
473: %
474: or, integrating over solid angle,
475: %
476: \begin{equation}
477: \left(\frac{dE}{ d\omega}\right)_{\rm class}=
478: \frac{q^2}{ \pi}\left[\frac{1}{ v_\infty}
479: \log\left(\frac{1+v_\infty}{ 1-v_\infty}\right)-2\right]\,.
480: \label{dedw-flat}
481: \end{equation}
482: %
483: This formula has been tested with great accuracy experimentally. To get
484: the total energy, one has to integrate (\ref{dedw-flat}) over
485: frequencies, and a naive procedure would lead to infinities in the
486: total energy. To obtain reasonable results, one has to impose a
487: cutoff $\omega_c$ in the frequency depending on the particular problem
488: under consideration, and one obtains
489: \begin{equation}
490: \Delta E_{\rm class}=\frac{q^2}{ \pi}\left[\frac{1}{ v_\infty}
491: \log\left(\frac{1+v_\infty}{ 1-v_\infty}\right)-2\right]\omega_c\,.
492: \label{totEclass}
493: \end{equation}
494: How well do these classical formulas fit into our numerical results?
495: Very well indeed. To allow a more direct comparison, we shall first
496: decompose the energy spectrum (\ref{dedw-flat}) into spherical
497: harmonics, i.e., into the energy spectrum associated with each
498: $l$-pole radiation as follows:
499: \begin{eqnarray}
500: \left(\frac{dE}{ d\omega}\right)_{\rm class}
501: %&
502: =
503: %&
504: \sum_l
505: \left(\frac{dE_l}{ d\omega}\right)_{\rm class} \nonumber \\
506: %&
507: =
508: %&
509: \sum_l\frac{q^2}{ 4\pi^2l(l+1)}
510: \left|\int\frac{v_\infty\sin\theta}{ 1-v_\infty\cos\theta}
511: \frac{\partial Y_{l0}}{ \partial\theta}d\Omega \right|^2\,.
512: \end{eqnarray}
513: When we consider the limit of $v_\infty\rightarrow 1$, as shown in
514: equation (\ref{dedw-flat}), the total energy spectrum $\left(\frac{dE}{
515: d\omega}\right)_{\rm class}$ diverges. However, the energy spectrum due to
516: $l$-pole radiation $\left(\frac{dE_l}{ d\omega}\right)_{\rm class}$ converges
517: to a finite value, given by
518: \begin{equation}
519: \left(\frac{dE_l}{ d\omega}\right)_{\rm class}= q^2\,\frac{2l+1}{ \pi l(l+1)}\,.
520: \label{fit0}
521: \end{equation}
522: This yields for example $\frac{dE}{d\omega}_{l=1}=0.477 q^2$ and
523: $\frac{dE}{d\omega}_{l=2}=0.265 q^2$ as $\gamma \rightarrow
524: \infty$. This may be compared with the numerical values in Fig. 2. The
525: numerical agreement is excellent not only for these lower multipoles,
526: but also for higher ones, and one has therefore established
527: numerically that for high energy collisions the classical result
528: (\ref{dedw-flat}) for the energy spectrum is valid. Probably more
529: interesting is the total energy radiated. Since one already knows that
530: the classical result for $\frac{dE}{d\omega}$ is correct, let us now
531: try to predict the cutoff frequency by identifying $\Delta E$ in
532: equation (\ref{total-energy}) with (\ref{totEclass}). Making use of this
533: definition of $\omega_c$ and numerical values of $\Delta E$, we
534: evaluate the cutoff frequencies $\omega_c$ in (\ref{totEclass})
535: for several values of $\gamma$, and
536: list the values of $\omega_c$ in Table II. In Figure 3, we also show
537: the cutoff frequencies as a function of $\gamma$. It is found in this
538: figure that for relatively high vales of $\gamma$, $\omega_c$ increase
539: almost linearly with $\gamma$, and a good fit to our numerical values
540: is given by
541: %
542: \begin{equation}
543: M\omega_c=0.224+0.0598\,\gamma \,.
544: \end{equation}
545: %
546: In Figure 3, we have shown this linear function, and we can confirm that
547: this linear function is in good agreement with our numerical values for
548: high $\gamma$.
549: This means that a good approximation to the total electromagnetic energy
550: radiated is
551: \begin{equation}
552: \Delta E= \frac{q^2}{M \pi}\left[\frac{1}{ v_\infty}
553: \log\left(\frac{1+v_\infty}{ 1-v_\infty}\right)-2\right](0.224+0.0598\gamma)\,,
554: \label{eleapprox}
555: \end{equation}
556: or considering that $v_\infty \sim 1$ we have also
557: \begin{equation}
558: \Delta E= \frac{2q^2}{M \pi}\left[\log{2\gamma}-1 \right]
559: (0.224+0.0598 \gamma)\,,
560: \label{eleapprox2}
561: \end{equation}
562:
563:
564:
565:
566:
567: \bigskip
568: %\newpage
569: %\vskip 1mm
570: \begin{figure}
571: \centerline{\includegraphics[width=8cm,height=8 cm]{f2.eps}}
572: \caption{
573: The cutoff frequency, given as a function of $\gamma$.
574: The filled circles on the curve of the cutoff frequency indicate the cutoff
575: frequency calculated in this study. The linear function
576: $\omega_c=0.224+0.0598\,\gamma$ is also shown.
577: }
578: \label{cutofffrequencygraph}
579: \end{figure}
580: %\bigskip
581: %
582: \begin{table}
583: \caption{\label{tab:zfl2} The cutoff frequency $\omega _c$ as
584: a function of $\gamma$ .}
585: \begin{ruledtabular}
586: \begin{tabular}{c|c}
587: $\gamma$ & $\omega _c\,M $ \\ \hline
588: 1 & $\infty$ \\
589: 2 & $3.44\times 10^{-1}$ \\
590: 3 & $4.03\times 10^{-1}$ \\
591: 4 & $4.64\times 10^{-1}$ \\
592: 5 & $5.23\times 10^{-1}$ \\
593: \end{tabular}
594: \end{ruledtabular}
595: \end{table}
596: %\vskip 1mm
597:
598: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
599: \noindent
600: \section{Comparing the Amount of Gravitational and Electromagnetic
601: Energy radiated}
602: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
603: Previous studies on the high energy collision of point particles
604: with black holes led \cite{cardosolemos0,cardosolemos1} to the
605: following conclusion:
606: if a high energy point particle of mass $\mu$ collides head on with
607: a mass $M$ non-rotating black hole, the total amount of gravitational
608: energy radiated is
609: \begin{equation}
610: \Delta E_g=0.26 \frac{\mu ^2 \gamma ^2}{M}\,,
611: \label{vlemos}
612: \end{equation}
613: where $\gamma$ is again the Lorentz factor for the point particle.
614: Furthermore, it was also found that the spectra is flat and
615: also well described by a classical calculation (here classical
616: means again working on a flat background). Now, for high $\gamma$'s
617: (\ref{eleapprox2}) is
618: \begin{equation}
619: \Delta E \sim \frac{0.038 q^2}{M }\gamma \log{2\gamma}\,,\,\,\gamma
620: \rightarrow \infty\,.
621: \label{eleapprox3}
622: \end{equation}
623: So, we get for the ratio electromagnetic energy over gravitational energy,
624: \begin{equation}
625: \frac{\Delta E}{\Delta E_g} \sim 0.146 (\frac {q}{\mu})^2
626: \frac{\log{2\gamma}}{\gamma}\,,\,\,\gamma \rightarrow \infty\,.
627: \label{ratio}
628: \end{equation}
629: This means that electromagnetic energy is supressed in relation
630: to gravitational energy, for very high energies.
631:
632: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
633: \noindent
634: \section{Conclusions}
635: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
636: We have computed the electromagnetic spectrum and total energy
637: radiated during the high energy collision of a charged point particle
638: with a Schwarzschild black hole. Our results show that the
639: classical ``instantaneous collision'' calculation gives very good
640: results, in accordance with the full numerical ones. We have dealt
641: only with zero impact (i.e., head-on) collisions, but these results
642: are easily generalized to non head-on collisions. For example, one
643: expects that the classical results \cite{chang49,jackson} hold for
644: such cases, and therefore that the total energy decreases as one
645: increases the impact parameter. We stress that the
646: results presented here are valid for any particle charge $q$,
647: as long as the total effective stress energy of the particle is
648: small compared to the total energy content of the black hole.
649: Previous works \cite{cardosolemos0,cardosolemos1} have shown that the
650: high energy collision between two black holes of equal mass may be
651: well studied through the collision of a point particle with a
652: black hole, and then taking the limit of equal mass, although this is
653: formally not allowed. It is our belief that the same may be done
654: here. One needs however some other method to attack this problem,
655: to confirm or disprove this claim. The investigation carried
656: here can also be carried over to higher dimensions, a case which is of
657: more direct interest to TeV-scale gravity scenarios. Since Standard
658: Model fields inhabit a four-dimensional brane, what one would need to
659: generalize this construction would be to modify the induced
660: four-dimensional metric (\ref{metric}), as was done for example in
661: \cite{kanti}. The generalization is straightforward.
662:
663:
664: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
665: %\section*{Acknowledgements}
666: \noindent {\bf Acknowledgements -}
667: We thank George Smoot for asking, during the talk of one of us (VC) in
668: the Fourth International Workshop on New Worlds in Astroparticle
669: Physics (University of Algarve, Faro, Portugal), how the gravitational
670: radiation emitted in the process compares with the electromagnetic
671: radiation, from which this work resulted.
672: This work was partially funded by Funda\c c\~ao para a Ci\^encia e
673: Tecnologia (FCT) -- Portugal through project CERN POCTI/FNU/49521/2002.
674: VC acknowledges finantial support from FCT through PRAXIS XXI
675: programme.
676: JPSL acknowledges finantial support from ICCTI/FCT and thanks
677: Observat\'orio Nacional do Rio de Janeiro for hospitality.
678: SY acknowledges finantial support from FCT through project SAPIENS
679: 36280/99.
680:
681:
682:
683: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
684:
685: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
686:
687: \begin{thebibliography}{100}
688:
689: \bibitem{cardosolemos0} V. Cardoso and J. P. S. Lemos,
690: Phys. Lett. {\bf B 538}, 1 (2002).
691:
692: \bibitem{cardosolemos1} V. Cardoso and J. P. S. Lemos,
693: Gen. Rel. Gravitation {\bf 35}, 327-333 (2003);
694: Phys. Rev. D {\bf 67}, 084005 (2003).
695:
696: \bibitem{payne} P. D. D'Eath and P. N. Payne,
697: Phys. Rev. D {\bf 46}, 658 (1992);
698: Phys. Rev. D {\bf 46}, 675 (1992);
699: Phys. Rev. D {\bf 46}, 694 (1992);
700:
701: \bibitem{aichelburg} P. C. Aichelburg and R. U. Sexl,
702: Gen. Rel. Grav. {\bf 2}, 303 (1971).
703:
704: \bibitem{yoshino} H. Yoshino and Y. Nambu,
705: Phys. Rev. D {\bf 67}, 024009 (2003).
706:
707: \bibitem{bhperturbation} T. Regge and J. A. Wheeler,
708: Phys. Rev. {\bf 108}, 1063 (1957);
709: F. Zerilli, Phys. Rev. Lett. {\bf 24}, 737 (1970).
710: F. Zerilli, Phys. Rev. D {\bf 2}, 2141 (1970);
711: M. Davis, R. Ruffini, W. H. Press, and
712: R. H. Price, Phys. Rev. Lett. {\bf 27}, 1466 (1971);
713: S. A. Teukolsky, Astrophys. J. {\bf 185}, 635 (1973);
714: T. Nakamura and M. Sasaki, Phys. Lett. {\bf 89A}, 68 (1982);
715: M. Sasaki and T. Nakamura, Phys. Lett. {\bf 87A}, 85 (1981);
716: M. Sasaki and T. Nakamura,
717: Prog. Theor. Phys. {\bf 67}, 1788 (1982);
718:
719: \bibitem{malheiro} S. Ray, A. L. Espindola, M. Malheiro, J. P. S. Lemos
720: and V. T. Zanchin, Phys. Rev. D, in press (2003), astro-ph/0307262.
721:
722: \bibitem{hamed} N. Arkani-Hamed, S. Dimopoulos and G. Dvali,
723: Phys. Lett. {\bf B429}, 263 (1998); Phys. Rev. {\bf D59}, 086004 (1999).
724: I. Antoniadis, N. Arkani-Hamed, S. Dimopoulos and G. Dvali,
725: Phys. Lett. {\bf B436}, 257 (1998).
726:
727: \bibitem{bhprod} P. C. Argyres, S. Dimopoulos and J. March-Russell
728: Phys. Lett. {\bf B441}, 96 (1998);
729: S. Dimopoulos, G. Landsberg, Phys. Rev. Lett. {\bf 87}, 161602 (2001).
730:
731: \bibitem{cav} M. Cavaglia, Int. J. Mod. Phys.
732: {\bf A18}, 1843 (2003).
733:
734: \bibitem{zerilli1} F. Zerilli, Phys. Rev. {\bf D2}, 2141 (1970);
735: Phys. Rev. {\bf D9}, 860 (1974).
736:
737: \bibitem{ruffini72} R. Ruffini, Phys. Lett. {\bf 41B}, 334 (1972).
738:
739: \bibitem{ruffini74} M. Johnston, R. Ruffini, F. Zerilli,
740: Phys. Lett. {\bf 49B}, 185 (1974).
741:
742: \bibitem{chandra} S. Chandrasekhar, and S. Detweiler, Proc. R. Soc. London
743: {\bf A 344}, 441 (1975); S. Chandrasekhar, in
744: {\it The Mathematical Theory of Black Holes},
745: (Oxford University Press, New York, 1983).
746:
747: \bibitem{chang49} C. S. Wang Chang, and D. L. Falkoff,
748: Phys. Rev. {\bf 76}, 365 (1949).
749:
750: \bibitem{jackson} J. D. Jackson,
751: {\it Classical Electrodynamics}, (J. Wiley, New York 1975);
752: V. Cardoso, O. J. C. Dias and J. P. S. Lemos,
753: Phys. Rev. D {\bf }(in press), hep-th/0212168.
754:
755: \bibitem{kanti} P. Kanti and J. March-Russell,
756: Phys. Rev. D {\bf 67}, 104019 (2003).
757:
758: \end{thebibliography}
759: \end{document}
760: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
761:
762:
763:
764: