1:  %%%%%%%%%%%%%%%%% End of the XXX.LANL.GOV header %%%%%%%%%%%%%%%%%%%%%%

2: 

3: %  Written in LaTeX, macro JHEP, available at http://jhep.sissa.it/   %

4: % options: hyper, nohyper, 11pt, 12pt, draft (comments on margin),... %

5: 

6: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

7: 

8: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

9: 

10: \documentclass[nohyper,11pt,letterpaper]{JHEP3}

11: \usepackage{epsfig}

12: \usepackage{graphicx}

13: %                                                           letterpaper

14: %      also legalpaper,executivepaper,a4paper,a5paper,b5paper,landscape

15: 

16: 

17: 

18: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

19: 

20: %                    FUNNY MATH FONTS                                      %

21: 

22: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

23: 

24: 

25: 

26: %\newfont{\frak}{eufm10 scaled 1200}

27: \newcommand{\mfrak}[1]{\mbox{\frak #1}}

28: %\newfont{\Bbb}{msbm10 scaled 1200}     %instead of eusb10

29: %newcommand{\mathbb}[1]{\mbox{\Bbb #1}}

30: \DeclareSymbolFont{AMSa}{U}{msa}{m}{n}

31: \DeclareSymbolFont{AMSb}{U}{msb}{m}{n}

32: \let\Box\relax

33: \DeclareMathSymbol{\Box}{\mathord}{AMSa}{"03}

34: \def\IZ{{\mathbb Z}}

35: \def\IR{{\mathbb R}}

36: \def\IC{{\mathbb C}}

37: \def\IQ{{\mathbb Q}}

38: 

39: 

40: 

41: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

42: 

43: %                    GENERAL PURPOSE DEFINITIONS                           %

44: 

45: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

46: 

47: \def \eqn#1#2{\begin{equation}#2\label{#1}\end{equation}}

48: \def \rut{2/5 transformation}

49: \def \abs#1{\left\vert#1\right\vert}

50: \def \Rut{{\mathbb G}}             %name of the group

51: \def \lpl{L_{planck}}

52: \def \lst{L_{string}}

53: \def \ham{\mathcal H}

54: \def \Lag{\mathcal L}

55: \def \action{\mathcal A}

56: \def\hacek{\accent20}                           % originally \v used by \nu

57: 

58:                                                 % for writing Horava...

59: 

60: 

61: 

62: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

63: 

64: %                      TITLE PAGE                                          %

65: 

66: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

67: 

68: \title{Vacuum Bubble in an Inhomogeneous Cosmology}

69: 

70: \author{W. Fischler, S. Paban, M. \v{Z}ani\'{c}\\

71: Department of Physics \\ University of Texas \\ 1 University Station, C1608 \\ Austin,

72: TX 78712 \\

73: E-mail: \email{fischler@physics.utexas.edu, paban@physics.utexas.edu, marija@mail.utexas.edu}}

74: 

75: \author{C. Krishnan \\ International Solvay Institutes \\ Physique 

76: Th\'{e}orique et Math\'{e}matique \\ ULB C.P. 231, Univerisit\'{e} Libre 

77: de Bruxelles, \\ B-1050, Bruxelles, Belgium \\

78: E-mail: \email{Chethan.Krishnan@ulb.ac.be}}

79: 

80: \abstract{We study the propagation of bubbles of new vacuum in a radially

81: inhomogeneous Lema\^{i}tre-Tolman-Bondi background that

82: includes a cosmological constant.  This exemplifies   

83: the classical evolution of a tunneling bubble through a metastable state 

84: with 

85: curvature inhomogeneities, and  

86: will be relevant in the context of the 

87: Landscape. We demand that the matter profile in the LTB background 

88: satisfy the weak energy condition. For sample profiles 

89: that satisfy this restriction, 

90: we find that the evolution of the 

91: bubble (in terms of the physically relevant

92: coordinates intrinsic to the shell) is largely unaffected by 

93: the prsence of local inhomogeneities. Our setup 

94: should also be a useful toy model for capturing 

95: the effects of ambient inhomogeneities on an inflating region. 

96: }

97: 

98: \keywords{Classical Theories of Gravity, Cosmology of Theories Beyond the 

99: SM}

100: 

101: \received{???????? ?st, 2000} \accepted{???????? ?th, 1998}

102: \preprint{\\UTTG-10-07\\ULB-TH/07-31}

103: %\preprint{\\ULB-TH/07-31}

104: 

105: \begin{document}

106: 

107: 

108: 

109: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

110: 

111: %          Table of contents automatic !!!                                 %

112: 

113: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

114: 

115: 

116: 

117: \section{\bf Introduction}

118: 

119: Although bubble propagation in homogeneous cosmological backgrounds

120: has been amply studied \cite{Israel:1966rt}-\cite{Lake:1984pn}, bubble 

121: propagation in

122: inhomogeneous backgrounds is virtually unexplored 

123: \cite{Khakshournia:2002jr}. 

124: %This issue, however, might be relevant in the context of the Landscape. 

125: Since tunneling is of enormous importance \cite{Vilenkin:1983xq, 

126: Hawking:2006ur} in populating the string Landscape \cite{Kachru:2003aw, 

127: Susskind:2003kw, Aguirre:2007an, Chang}, we expect that the evolution of a 

128: new Universe through 

129: an old one, is worthy of study.

130: 

131: One reason for the lack of emphasis on inhomogeneous 

132: backgrounds is the cosmic no-hair theorem \cite{Gibbons:1977mu, 

133: Hawking:1981fz, Starobinsky:1982mr}.

134: This theorem states that the evolution of an expanding universe that 

135: contains a cosmological constant will eventually be dominated by 

136: vacuum energy, and will asymptote to a de Sitter space. It will 

137: imply in particular, that inhomogeneous backgrounds containing a 

138: cosmological constant (like the ones constructed on Landscape vacua) will 

139: evolve toward 

140: `ironing out' such inhomogeneities, and therefore it will make sense to 

141: restrict the study of bubble propagation to homogeneous

142: backgrounds. 

143: %But to the best of our knowledge, the cosmic no hair theorem 

144: %has not been 

145: %unambiguously proven, nor have its hypotheses been clearly stated, thus 

146: %warranting a re-appraisal of the problem. 

147: But fluctuations can still clump locally even in an asymptotically de 

148: Sitter space, and therefore the question of 

149: evolution of vacuum bubbles in an inhomogeneous spacetime is still 

150: relevant. In the context of eternal inflation and the string landscape, we 

151: have to face up to the possibility that the Universe that we inhabit is a 

152: result of bubble evolution through an inhomogeneous relic Universe. In 

153: principle this could give rise to experimental signatures in the form of 

154: CMB fluctuations, about which we comment in the final section. 

155: 

156: One particular work that stands out in this context is that of Wald 

157: \cite{Wald:1983ky}, who did a comprehensive study of of 

158: the late time evolution of anisotropic but homogeneous spaces with a 

159: cosmological constant. In this restricted setting, he was able to 

160: define the conditions for the cosmic no hair theorem to be satisfied by 

161: taking advantage of the Bianchi classification of homogeneous, 

162: anisotropic metrics.  Wald studied the time evolution of such metrics 

163: when the energy momentum tensor is the sum of two components: a 

164: cosmological constant and a term that satisfies the dominant  and strong 

165: energy conditions. He found that with the exception of the 

166: Bianchi IX, the other models will always asymptote to a de Sitter space, 

167: within a time scale $\sqrt{\frac{3}{\Lambda}}$. In

168: the case of the Bianchi IX models, the future behavior depends on the 

169: relative sizes of the cosmological constant and the spatial-curvature. 

170: Only when the latter exceeds the cosmological constant will the time 

171: evolution {\em not} be asymptotically de Sitter.

172: 

173: 

174:   It is often claimed that inflation solves  the horizon and flatness 

175: problems but,  as already remarked in its

176: early years, these claims are based on the assumption that the  

177: pre-inflationary state of the universe can be described by a homogeneous 

178: and isotropic Friedmann-Robertson-Walker (FRW) space. Under these strong assumptions, inflation does indeed reduce considerably the amount of fine tuning  necessary to explain the current experimental measurements. In general, however,  one expects a high degree of inhomogeneity in the initial conditions, with some regions of the universe expanding while others contract.  Holographic cosmology  might be the only exception to this rule because it naturally produces a homogeneous and almost flat space-time   \cite{Banks:2001px, Banks:2003ta, Banks:2004cw}.

179:   

180: 

181: The 

182: question of whether an inflating region can

183: continue to inflate if the ambient region is inhomogeneous, has been 

184: explored by several authors for different 

185: pre-inflationary backgrounds.

186: In a set of papers  

187: \cite{Goldwirth:1989pr,Goldwirth:1991rj,Goldwirth:1989vz}, Goldwirth and Piran sought to answer the question of whether large inhomogeneity in the very early universe could prevent it from entering a 

188: period of inflation. In their work,

189: inflation was not driven by a cosmological constant but by an inflaton field. Their analysis is numerical and restricted to an inhomogeneous but isotropic and  closed-universe background. They found that a large initial inhomogeneity could indeed suppress the onset of inflation, and that  the inflaton field must have a suitable value over a region of several horizon sizes in order for inflation to start.  This last finding raises the issue of an acausal initial condition. 

190: Other numerical simulations disagree with these results claiming that inhomogeneous backgrounds will have enough inflation to explain current observations \cite{Kurki-Suonio:1993fg,Berera:2000xz}. 

191: Analytical computations that use the long wave-length approximation seem to confirm that large inhomogeneities of the spatial curvature prevent the onset of inflation \cite{Deruelle:1994pa, Iguchi:1996jy, Iguchi:1996rh}. 

192: 

193: In this work, we will not be able to conclusively settle this issue once

194: and for all. Instead, our goal here is to study a particular problem that

195: we believe captures some of the same physics by playing local curvature

196: effects against the might of the cosmological constant. The problem we

197: consider is the classical evolution of a vacuum bubble in an

198: inhomogeneous background, containing a cosmological constant and dust.

199: %The lesson from the analyses quoted above is that spatial curvature 

200: %effects can overcome the might of the cosmological constant. In this 

201: %paper we will study the evolution of a bubble in a background that 

202: %contains both a cosmological constant and dust. 

203: This background, which we 

204: will refer to as the Lema\^{i}tre-Tolman-Bondi (LTB)  space-time  

205: \footnote{ We slightly generalize it from the original form to allow for 

206: a positive cosmological constant.}, will be the simplest possible 

207: inhomogeneous space, spherically symmetric and with only one center 

208: \cite{Lemaitre:1933gd}-\cite{Bondi:1947av}. The bubble will be created in 

209: a region of space that is expanding but will encounter through its 

210: evolution regions of varying curvature. In a nutshell, the problem is 

211: that the critical size for bubble nucleation depends locally on the scale 

212: factor. As this factor changes as a function of the position of the 

213: shell, it is possible for a supercritical bubble to become subcritical  

214: \footnote {

215: A bubble that can become subcritical in the future is a fluctuation, not 

216: a phase transition. }. Our analysis will be confined to the evolution of 

217: the bubble, it will not address the interesting issue of tunneling 

218: probabilities in inhomogeneous backgrounds. This latter issue is very 

219: complicated, some attempts to address it have

220:  been made in \cite{Abbott:1987xq}. 

221: 

222: Before embarking on describing the details, we emphasize that we are 

223: {\em not} 

224: dealing with the most general inhomogeneous bubble evolution imaginable. 

225: There are three major assumptions we make to render the problem tractable, 

226: and we list them below. The results of this paper should therefore be 

227: regarded as exploratory, rather than conclusive.

228: \begin{itemize}

229: \item The first assumption we make is that the ambient spacetime into 

230: which the bubble evolves is the radially inhomogeneous LTB form. This 

231: allows us to work with an exact solution of Einstein's equation, but the 

232: restricted form of the metric and matter in this solution implies that we 

233: are not dealing with the most general radially inhomogeneous cosmology 

234: admissible. But as we explain in more detail in the next section, this 

235: still leaves us with a great deal of flexibility in choosing the 

236: inhomogeneity profiles. On the inside of course, we are dealing with a 

237: vacuum bubble, 

238: so we take the matter there to be only a cosmological constant without 

239: loss of generality.

240: \item The second assumption that we make is that the evolution of the new 

241: phase can be adequately captured using a thin-wall approximation. In 

242: particular this means that we can deal with all the transient phenomena 

243: between the two regions using a thin layer. The more general scenario 

244: would be to model the matter leakage from one region 

245: to the other using a gradual profile, but we believe the 

246: thin-wall captures the phenomena involved to a first 

247: approximation.  

248: \item The above two assumptions together give us some traction. Once we 

249: make them, the 

250: problem that we need to solve reduces to that of 

251: solving the Israel junction conditions for a thin shell. But the junction 

252: conditions do 

253: not fully fix the evolution of the shell until we specify an equation of 

254: state on the bubble-wall. In the purely general relativistic context that 

255: we are working in, these equations of state are not fixed by the 

256: dynamics. So what we do 

257: here is to take the first step in this direction by considering a few 

258: equations of state of the form $p= w \rho$, for a range of values of $w$. 

259: We make some comments elsewhere on how one might try to fix this equation of 

260: state using field-theoretic arguments, but the full solution of this issue 

261: in the complicated background that we are working in, is likely to be 

262: difficult. In any event, as a start, we settle here for an 

263: exploration of these equations of state, leaving a more thorough analysis 

264: for future.

265: 

266: \end{itemize}

267: 

268: We investigate a few different scenarios using our numerical formalism. 

269: The first is the homogeneous limit of LTB, namely 

270: Robertson-Walker cosmologies with or without matter. We reproduce and 

271: extend known results in the literature using our methods, and offer 

272: some 

273: comments and new perspectives. Particular attention is paid to the case of 

274: bubble-propagation in an FRW universe with a Big Crunch in the future. We 

275: find that there are new ways (details are in a later section) 

276: in which the bubble can evolve in this case: it can even recollapse. The 

277: final scenario we consider is an LTB space that is truly 

278: inhomogeneous. In our examples, we pick our inhomogeneity profiles to 

279: have a fluctuation that is as sharp as possible while still respecting 

280: the weak energy condition. Surprisingly, 

281: we find that the qualitative bubble 

282: evolution (as seen in the physically relevant coordinates intrinsic to the 

283: shell or in the inside coordinates) is essentially the same with or 

284: without the fluctuation. This suggests that the energy conditions work 

285: towards the smoothening out of exotic bubble evolution patterns.

286: 

287: In the next section we shall describe the LTB model that represents the background on which the bubbles will propagate. In section 3, we will review the propagation of bubbles in a homogeneous background. We will consider the already known case where the outside background is simply a de Sitter space, either flat or closed, and the novel case where matter is added to the cosmological constant. In this latter case, we shall explore the interesting situation of a bubble moving in a background that evolves from expanding to collapsing. 

288: In section 4 we shall study the truly inhomogeneous 

289: LTB background. 

290: Our results, future directions and cautionary remarks are put together in

291: the Conclusion.

292: We have relegated to the Appendix, an explanation of 

293: the known formalism to study the propagation of bubbles in general relativity.

294: 

295: 

296: \section{\bf Setup: The Bubble and the Background} 

297: %\section{\bf Description of the Background on which the Bubble Propagates }

298: 

299: A propagating bubble divides the space-time into three regions: outside, shell and inside. 

300: To give a proper description of this space-time in general relativity we 

301: will make use of the junction conditions, first presented by Israel  

302: \cite{Israel:1966rt} and extensively used since then by many authors. In particular, we will follow the implementation of this formalism developed by Berezin, Kuzmin and Tkachev \cite{Berezin:1987bc}.

303: 

304: The model we will study consists of a bubble of true vacuum propagating on 

305: a metastable state whose energy-momentum tensor contains a higher 

306: cosmological 

307: constant  and dust. 

308: The bubble is assumed to be a thin-shell, with a perfect-fluid energy-momentum tensor whose equation of state will be allowed to vary. The true vacuum will be assumed homogeneous and isotropic while the outside background will be assumed to be spherically symmetric about one point in space. 

309: The line element of the \underline{outside region} \cite{Peebles:1967} is: 

310: 

311: \eqn{metric_b}{ds^2=d{t}^2 - \frac{[a(t,r) + r 

312: a'(t,r)]^2}{1-\frac{r^2}{R^2(r)}} dr^2 - a^2(t,r) r^2 d\Omega_2}

313: The expansion parameter $a(t,r)$ is both a function of time and of the radial coordinate $r$. 

314: Partial derivatives with respect to the radial coordinate will be represented by a prime. The function

315: $R(r)$ is an arbitrary function of the radial coordinate only and will be taken to be positive everywhere.

316: We will study the evolution of the bubble for different choices of  $R(r)$. It can be seen from these equations that we can choose a function 

317: $R(r)$ and

318: then integrate for $a(t,r)$ at each $r$. Then we can use this $a(t,r)$ to

319: {\em

320: define} the $d(t,r)$ in (\ref{EOMB}), with the caveat that the

321: choice of $R(r)$ has to be made in such a way that this energy density

322: needs to be positive.  

323: Therefore, the description of the spacetime is 

324: essentially complete (apart from fixing certain initial conditions which 

325: we will get to in a minute), once we specify the function $R(r)$. Because 

326: it is our prejudice that curvature will interfere with the expansion

327: of the bubble when its value is comparable to the cosmological constant,

328: we will choose $R(r)$ such that it can locally overwhelm the effect of 

329: the cosmological constant. 

330: 

331: 

332:  

333: The equations of motion for the expansion factor and the dust density $d(t,r)$, in units where $8 \pi G=c=1$, are:

334: 

335: \begin{eqnarray} 

336: \Big(\frac{\partial_t a(t,r)}{a(t,r)}\Big)^2 + \frac {1}{a^2(t,r) R^2(r)} 

337: & = &  \frac{A} {a^3(t,r)} + \frac{\Lambda_{out}}{3}  \label{EOM}\\

338: \frac{1}{3} d(t,r) a^2(t,r) ( a(t,r) + r a'(t,r)) & = & A  \\ \nonumber \label{EOMB}

339: \end{eqnarray}

340: $A$ is a constant. These equations reduce to the familiar Friedmann-Robertson-Walker equations in

341: the limit where $R(r)$ is independent of the radial coordinate. Besides choosing $R(r)$ the solution to the equations of motion will involve the choice of  the initial condition $a(t_0,r)$. For simplicity, we will 

342: choose this function to be independent of the radial coordinate $r$.

343: 

344: The results that will be presented in the forthcoming sections 

345: correspond

346: to different choices

347: of the function $R(r)$. 

348: %The plot of the scale factor as a function of $t$

349: %and $r$ for an illustrative choice of $R(r)$ is given in

350: %figure \ref{scale-factor}.

351:  For different

352: choices of $R(r)$, the collapse-profile can be completely different. One

353: way to understand this is to notice that the Friedmann-type

354: equation here can be thought of as an energy equation for a particle

355: of zero total energy in a potential

356: \eqn{potential}{V(a)=-\frac{A}{a}-\frac{\Lambda_{out}

357: a^2}{3}+\frac{1}{R^2}}

358: For those values of $r$ for which the maximum of this potential is above

359: zero, if the universe

360: starts at a big-bang, it will recollapse, but for other values of $r$

361: there is no recollapse.

362: 

363: 

364: The \underline{bubble} will be assumed to be a thin-shell (described 

365: by a hypersurface $\Sigma$) and spherically symmetric. Its energy momentum 

366: tensor will be assumed to be of a perfect-fluid type: 

367: $ {S_\tau}^\tau= \sigma, {S_\theta}^\theta={S_\phi}^\phi= -P$, $P=w 

368: \sigma$  and with the metric:

369: 

370: \eqn{metricbubble1}{{ ds^2|}_{\Sigma}=  d \tau^2  - \rho^2(\tau) 

371: d\Omega_2 

372: }

373: The \underline{interior of the bubble} will be assumed homogeneous and described by 

374: 

375: \eqn{metricbubble2}{ ds^2 = dT^2 - b^2(T) \left( \frac{dz^2}{1+z^2} + z^2 

376: d\Omega_2 \right) }

377: with the following equation of motion for the scale factor:

378: 

379: \eqn{FRWbubble}{ \left( \frac{db}{dT}  \right)^2 = 

380: \left( \frac{\Lambda_{in}}{3} \right) b^2(T) + 1}

381: We are assuming the inner space-time of the bubble to be open as  derived from a tunneling process \cite{Coleman:1980aw}.

382: 

383: Given this setup, our aim is to understand the evolution of the bubble. 

384: The dynamics of the bubble is governed by the Israel junction conditions, 

385: the details of which are relegated to an appendix. The coupled 

386: differential equations that govern shell evolution turn out to be 

387: (\ref{sigmaequation}), (\ref{Ton}), or in a more explicit form, 

388: (\ref{eomforshell1}), (\ref{eomforshell2}).

389: 

390: %The Israel conditions connect these  three portions of the space-time. 

391: %The first condition requires that the metric be a continuous function at 

392: %the junction. In terms of the coordinates of the outside(inside) region 

393: %the position of the hypersurface $\Sigma$ will be given respectively by 

394: %$r=x(t)$ and $z=Z(T)$. The continuity of the 

395: %metric at $\Sigma$ requires:

396: 

397: %\begin{eqnarray}

398: %& & \rho(\tau) = a( t, x(t)) x(t) = b(T) Z(T)  \label{metriccont1}  \\  

399: %\nonumber  \\  

400: %& & d\tau = \sqrt{1 - \frac{b^2(T)}{1+Z^2(T)} {\left(\frac{d Z}{dT} 

401: %\right )}^2}   \,\,\,  dT  \label{metriccont2} \\  \nonumber \\

402: %& & dt  = \frac{\sqrt{ 1 - x^2/R^2}}{\sqrt{ - \dot{x}^2 { ( a x),}^2_x  + 

403: %1 - x^2/R^2}} \,\,\,  d\tau \label{changevariables}

404: %\end{eqnarray}

405: %where  $ { (a x),}_x \equiv {(a'(t,r) r + a(t,r)) |}_{r=x(t)}$. The 

406: %remaining junction conditions are, as derived in \cite{Berezin:1987bc} 

407: %and in units where $8 \pi G =1$:

408: 

409: %\begin{eqnarray}

410: %\sigma & = &  2 \,  [ {K_2}^2] \label{jc1}  \label{sigma}  \\ \nonumber 

411: %\\ 

412: %- P          & = &   [ { K_0}^0] + [{K_2}^2]   \label{P}  \\  \nonumber 

413: %\\

414: %\frac{d \sigma}{d \tau} & = &  - \frac{2}{\rho}  \frac{d \rho}{d \tau} ( 

415: %\sigma+P) - [{T_0}^n] \label{sigmaev}

416: %\end{eqnarray}

417: %$[{K_i}^j]$ represents the discontinuity of the outer curvature tensor 

418: %and $[{T_0}^n]$ the discontinuity of this component of the energy 

419: %momentum tensor across $\Sigma$. 

420: %The general expression for the outside curvature tensor was derived in 

421: %\cite{Berezin:1987bc}. Instead of duplicating its

422: %derivation we will write the corresponding expression for the inside and 

423: %outside metric.

424: 

425: %\begin{itemize}

426: %\item{Outside Metric}

427: 

428: %\begin{eqnarray}

429: % {K_2}^2 & = & - \frac{\gamma_{out}}{\rho} ( \dot{\rho}^2 - 

430: %\Delta_{out})^{1/2}     \hspace{5ex}     \mbox{where}  \hspace{5ex}  

431: %\Delta_{out}= -1 + \rho^2 \left( \frac{\Lambda_{out}}{3} + \frac{A}{a^3} 

432: %\right) \nonumber \\ \nonumber \\

433: %{K_0}^0  & = & -  \frac{\gamma_{out}}{\sqrt{ \dot{\rho}^2 - 

434: %\Delta_{out}}} \left( \ddot{\rho}  + \rho \left( - \frac{ 

435: %\Lambda_{out}}{3}+ \frac{ A}{ 2 a^3} \right)  + \frac { \rho}{2}  \,\, 

436: %\frac{ d(t,r) \,\, \dot{x}^2  \,\, {(a x),}_x^2}{( - \dot{x}^2 {( a 

437: %x),}_x^2 + 1 - x^2/R^2)}\right) \nonumber \\ \nonumber \\

438: %{T_0}^n & = & -  \frac{\dot{x} \,\,  { (a x),}_x }{( - \dot{x}^2 {( a 

439: %x),}_x^2 + 1 - x^2/R^2)}  d(t,x) \,\, \sqrt{ 1- x^2/R^2}  \end{eqnarray}

440: 

441: %where $ { (a x),}_x =  {a'(t,r) r + a(t,r) |}_{r=x(t)}$. 

442: %$\gamma_{out}=+1$ when the radius of the outside region is increasing

443: %in the outward direction normal to the surface $\Sigma$, and 

444: %$\gamma_{out}=-1$ 

445: %if the radius is decreasing.

446: 

447: %\item{Inside Metric}

448: %\begin{eqnarray}

449: % {K_2}^2 & = & - \frac{\gamma_{in}}{\rho} ( \dot{\rho}^2 - 

450: %\Delta_{in})^{1/2}     \hspace{5ex}   \mbox{where} \hspace{5ex}    

451: %\Delta_{in}= -1 + {\rho}^2 \frac{\Lambda_{in}}{3}  \nonumber \\ \nonumber 

452: %\\

453: %{K_0}^0 & = & -  \frac{\gamma_{in}}{\sqrt{ \dot{\rho}^2 - \Delta_{in}}} 

454: %\left( \ddot{\rho}  - \rho  \frac{ \Lambda_{in}}{3} \right)  \nonumber \\  

455: %\nonumber \\

456: %{T_0}^n & = &  0 

457: %\end{eqnarray}

458: % $\gamma_{in}=+1$ when the radius of the inside region is increasing

459: %in the outward direction normal to the surface $\Sigma$, and 

460: %$\gamma_{in}=-1$ 

461: %if the radius is decreasing.

462: 

463: %\end{itemize}

464: %The equations (\ref{sigma})-(\ref{sigmaev}) are not independent, 

465: %the latter is an integrability condition of the first two. 

466: 

467: As always in general relativity, along with the dynamical equations, we 

468: also need an equation of state to fully specify the dynamics. 

469: Determining the equation of state between $\sigma$ and $P$ on the 

470: shell from first principles is a difficult problem that requires a field 

471: theoretic model for matter on either side. Since the LTB metric for a 

472: generic choice of the curvature profile $R(r)$ is known only numerically, 

473: this is doomed from the start. In the absence of a detailed understanding 

474: of this phase transition we will confine our study to perfect-fluid 

475: shells, $P = w \, \sigma$, and explore various values of $w$. 

476: 

477: It turns out that in the absence 

478: of dust on the outside, the shell-evolution equations can be written 

479: entirely 

480: in terms of 

481: $\rho(\tau)$ because (\ref{Ton}) becomes trivial, but 

482: this is no longer possible when dust is included. Indeed, equation 

483: (\ref{sigmaequation}) can be written as (reproduced from the appendix):

484: \begin{eqnarray}

485: \dot{\rho}^2 & = & -1 + B^2 \rho^2,  \label{EOMR}\\ \nonumber \\

486: B^2 & = & \frac{\Lambda_{in}}{3} + \left(  \frac{\sigma}{4} +  

487: \frac{1}{\sigma} \left( \frac{\Lambda_{out}-\Lambda_{in}}{3} + 

488: \frac{A}{a^3} \right) \right)^2. 

489: \end{eqnarray}

490: In general, to fix the evolution of $B(\tau)$ we need 

491: (\ref{Ton}), along with a knowledge of the scale factor through numerical 

492: solution of (\ref{EOM}). 

493: 

494: We also get constraints on the parameters and the allowed phase space 

495: from the positivity of energy, in particular, the condition that $\sigma$ 

496: be positive. The explicit relations that we will use later can again be 

497: found in the appendix.

498: 

499: \section{\bf Bubble Expansion in a Homogeneous Background}

500: 

501: The study of the bubble evolution can be done in different sets of coordinates. When dust is present, the problem of interest is most conveniently analyzed in the coordinates of the outside background.  To familiarize ourselves with these coordinates  we will devote this section to analyze

502: a simpler problem. The simplification will come from restricting the outside background to be homogenous

503: ($R(r) = \mbox{constant}$). First, we will redo the well studied motion of a vacuum shell in a cosmological constant

504: background. Then we will add a dust energy density to the cosmological constant. 

505: 

506: 

507: \subsection{Homogeneous Background without Matter}

508: 

509: The Israel junction conditions (\ref{sigmaequation})-(\ref{Ton}) tell 

510: us that 

511: it is consistent to have a vacuum shell (surface energy is constant and $w=-1$) in a transition that 

512: separates two de Sitter spaces with different values of the cosmological constant. Since this problem

513: has been solved many times in the literature we will only quote the results \cite{Berezin:1987bc} . 

514: 

515: In terms of the intrinsic bubble coordinates, the motion of the shell is given by:

516: 

517: $$ \rho(\tau) = \frac{1}{B} \cosh {B \tau} $$

518: where 

519: 

520: $$ B^2 =  \frac{\Lambda_{in}}{3} + \left( \frac{\sigma}{4} + \frac{\Lambda_{out}-\Lambda_{in}}{3 \sigma} \right) ^2 $$

521: regardless of the curvature of the outside background. The effect of this curvature only becomes manifest

522: when we express the motion of the bubble in terms of the outside coordinates. 

523: In these coordinates the evolution of the position of the bubble is given by:

524: 

525: \eqn{xevol}{ \frac{d x}{dt} = \frac{- (1- \frac{x^2}{R^2}) \sqrt{(\frac{\Lambda}{3} - \frac{1}{a^2 R^2})} \pm \sqrt{(B^2 a^2 x^2 -1)(B^2-\frac{\Lambda}{3})(1- \frac{x^2}{ R^2})}}{ a^2 x (B^2-\frac{1}{a^2 R^2})} }

526: where the  further assumption that $\Lambda_{in}=0$ and $\Lambda_{out}=\Lambda$ has been made. The evolution of the scale factor is given by the well known expression

527: 

528: $$ \Big(\frac{da/dt}{a}\Big)^2 + \frac{1}{a^2 R^2} = 

529: \frac{\Lambda}{3}  \label{FRWk}$$

530: For  times $ t >> \frac {1}{H} = \sqrt{3/\Lambda}$,  (\ref{xevol}) reveals the following asymptotic behavior :

531: 

532: \begin{eqnarray}

533: a(t) & \rightarrow &   \frac{1}{2 H  R} \,\, e^{H t} \hspace{4ex}  \mbox{when} \hspace{2ex}  a(0)=  \frac{1}{ H R}  \\ \nonumber \\

534: x(t)  &  \rightarrow &   R \sin \left( \frac{\mbox{constant}}{R} \right) - O( e^{- Ht}) \hspace{4ex}  \mbox{if}  \hspace{1ex}  R>0 \hspace{1ex} \mbox{ and finite}  \label{x1} \\ \nonumber\\ 

535: x(t) & \rightarrow &    \mbox{constant}+ O(e^{-Ht})  \hspace{6ex}  \mbox{if}  \hspace{1ex}  R \rightarrow \infty  \label{x2} \\  \nonumber \\

536: \frac{dx}{dt} & \rightarrow & \frac{1}{a} \,\, \sqrt{1 - \frac{x^2}{R^2}} 

537: \, \, 

538: \frac{ \frac{\sigma}{4}  - \frac{\Lambda}{ 3 \sigma}}{\frac{\sigma}{4}  + \frac{\Lambda}{ 3 \sigma}} \label{sol}

539: \end{eqnarray}

540: 

541: The value of the constants in (\ref{x1}) and ( \ref{x2}) is very dependent on initial conditions as can be seen in Figure \ref{figure_x_vac_nom}. Also (\ref{sol})  corroborates the well known fact \cite{Coleman:1980aw} that bubbles propagate asymptotically in time at the speed of light in the thin wall limit of $\Lambda \ll 3 \sigma $.

542: 

543:  A space with positive curvature only makes sense if $a(0) \geq \frac{1}{RH} $. Once the initial value exceeds this critical value the space expands forever. The effect of the curvature becomes

544: eventually negligible. Thus it is not surprising that the evolution of the bubble is similar regardless of the curvature. 

545: There is a difference in the possible asymptotic value that $x$ can take; in the positive curvature case

546: $x$ always has to remain below $R$. The rate of

547: the expansion is asymptotically given by $B$ in the shell coordinates and by $H$ in the

548: outside coordinates. 

549: In terms of the inside coordinates the motion of the bubble is given by:

550: 

551: $$ Z(T) = \frac{1}{B} \sqrt{1 + B^2 ( T - T_0)^2}$$

552: This expression can be derived from the metric matching condition at the 

553: shell (\ref{mink-match1}) and 

554: assuming that $b(T_0)=1$, which is a good assumption when $\Lambda_{in}=0$. $T_0$ is the inside time at which the bubble is created. For late times the bubble propagates at the speed of light as expected \cite{Coleman:1980aw}. Ultimately, we are always interested in the motion of the bubble as viewed by the observer inside the bubble. The moral of this example is that the motion, as viewed by the inside observer,  is the same regardless of the amount of curvature outside the bubble. 

555: 

556: \begin{figure}

557: \centering

558: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

559: \centering

560: \includegraphics[width=6cm]{figure_x_vac_nom.jpg}

561: \end{minipage}%

562: \begin{minipage}[b]{.55\textwidth}

563: \centering

564: \includegraphics[width=6cm]{figure_x_diffR_vac_nom.jpg}

565: \end{minipage}\\[-10pt]

566: \hspace{-1.4cm}\begin{minipage}[t]{.4\textwidth}

567: \caption{

568: Time evolution of the bubble in  the outside coordinates, $x[t]$, for several initial sizes,

569: $x_{init}=20,55,90$. The background is assumed homogeneous ($R=100$) and without matter.

570: The scales on the axes depend on the parameter values we have chosen and 

571: are therefore not terribly important. 

572: }\label{figure_x_vac_nom}

573: \end{minipage}%

574: \hspace{1.8cm} \begin{minipage}[t]{.4\textwidth}

575: \caption{

576: Time evolution of the bubble in the outside coordinates, $x[t]$, in homogeneous backgrounds without matter.

577: Colors correspond to different choices of R; $R=100$ red, $R=150$ green, $R=\infty $ blue.

578: } \label{figure_x_diffR_vac_nom}

579: \end{minipage}%

580: \end{figure}

581: 

582: \begin{figure}[htp]

583: \centering

584: \includegraphics [width = 12 cm]{figure_rho_diffR_vac_nom.jpg}

585: \caption{Time evolution of the bubble in the coordinates of the bubble, $\rho[\tau]$, in homogeneous backgrounds without matter 

586: (same as in Figure \ref{figure_x_diffR_vac_nom}).}

587: \label{figure_rho_diffR_vac_nom}

588: \end{figure} 

589: 

590: In our equations above (and in the ones that follow) we have chosen the

591: natural units $c=8\pi G=1$. Since we are dealing only with classical

592: evolutions, $\hbar$ never comes up, so we have the freedom to pick one

593: more unit. In the numerical evolution that we will undertake later, this

594: means that there is rescaling of the coordinates (by a dimensionful

595: parameter) that we are still free to do. 

596: %We need to make sure that  the scales involved do not cross

597: %the Planck scale because that would invalidate our classical analysis.

598: %We will always choose our various independent parameters

599: %(like $\Lambda$, $\sigma$, $A$) to be within  a few orders of magnitude

600: %of each other, to ensure that their relative scales are not Planckian. 

601: We will choose the various parameters (like $\Lambda$, $\sigma$, $A$) 

602: within a few orders of magnitude: the actual values themselves are not too 

603: relevant because of the

604: above-mentioned freedom in choosing the scale for the coordinates.

605: Physically we expect  the cosmological constant to

606: reflect the scale corresponding to the minimum of some effective scalar

607: potential, whereas the bubble stress-tensor should reflect  the

608: potential barrier between the two minima. 

609: %So in principle it is not

610: %inconceivable that for a sufficiently near-zero cosmological constant,

611: %the barrier height be many, many orders of magnitude bigger than the

612: %cosmological constant. 

613: For a generic potential, we expect that the

614: scales involved will be not too different in order of magnitudes. As far 

615: as the evolution of the shell goes, the only place where we

616: could have trans-Planckian effects becoming significant is when there is a

617: singularity in the spacetime itself. In particular the fact that the

618: shell surface is an actual physical wall, and {\em not} a surface at

619: conformal

620: infinity (like a horizon), means that we do not have to worry about finite

621: energy modes observed far away being red-shifted trans-Planckian modes

622: \cite{Farhi:1986ty, farhi2, fischler1, fischler2}.

623: 

624: 

625: The effects of curvature on the bubble propagation, as seen by the outside observer,  are illustrated in Figures \ref{figure_x_vac_nom},

626:  \ref{figure_x_diffR_vac_nom} and \ref{figure_rho_diffR_vac_nom}. To simulate these evolutions we have chosen the following values for the parameters, in units of $8 \pi G=1$:

627: \begin{eqnarray}

628:  \Lambda & =  &  3 \times 10^{-5}   \nonumber \\

629:  a_{init} & = & 5  \nonumber \\

630: \gamma_{out} & = & \gamma_{in}=+1 \nonumber \\

631: \sigma_{init} & = &10^{-3} 

632: \label{initial-conditions}

633: \end{eqnarray}

634: These values  for the parameters can not be chosen independently since they have to satisfy the condition  (\ref{xi_constraint}) that  yields:

635: 

636: \eqn{sigma_constr} {\sigma \leq 2 \sqrt { \frac {\Lambda} {3} } = 6.3 \times 10^{-3}}

637: Also from the equation (\ref{xevol}) the lower and the upper bound on $x$ are:

638: 

639: \eqn{x_range} { \frac {1}{aB} \leq x \leq R }

640: We have chosen the initial size of the bubble in accordance with this range.

641: 

642: In particular, Figure \ref{figure_x_vac_nom} shows the evolution of the bubble in

643: the outside coordinates for various initial sizes. Note that the maximum size of the bubble

644: in these coordinates corresponds to the limit $x=R$. 

645: 

646: Figure \ref{figure_x_diffR_vac_nom} shows the effects of curvature on bubble propagation

647: by comparing evolutions of bubbles with the same initial size $x_{init}$, but in different

648: curvature background. We see that more curvature corresponds to a slower evolution of the bubble 

649: in the outside coordinates; however, the corresponding bubble evolution $ \rho[\tau]$ 

650: in the bubble coordinates is exactly the same, irrespective of the amount of background curvature,

651: and it is shown in Figure \ref{figure_rho_diffR_vac_nom}.

652: 

653: \begin{figure}[htp]

654: \centering

655: \includegraphics [width = 12 cm]{figure_sigma_fixedR100_nom}

656: \caption{Time evolution of the surface energy density $\sigma [t]$ on the bubble. The background is assumed 

657: homogeneous ($R=100$) and without matter.

658: Colors correspond to different equations of state; $w=1/3$ red, $w=0$ yellow, $w=-1/3$ green, $w=-1$ blue. }

659: \label{figure_sigma_fixedR100_nom}

660: \end{figure} 

661: 

662: \begin{figure}

663: \centering

664: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

665: \centering

666: \includegraphics[width=6cm]{figure_rho_fixedR100_nom}

667: \end{minipage}%

668: \begin{minipage}[b]{.55\textwidth}

669: \centering

670: \includegraphics[width=6cm]{figure_x_fixedR100_nom}

671: \end{minipage}\\[-10pt]

672: \hspace{-1.2cm}\begin{minipage}[t]{.4\textwidth}

673: \caption{

674: Time evolution of the bubble in the coordinates of the bubble, $\rho[\tau]$.

675: The background is assumed homogeneous ($R=100$) and without matter.

676: Colors correspond to different equations of state; $w=1/3$ red, $w=0$ yellow, $w=-1/3$ green, $w=-1$ blue.

677: }\label{figure_rho_fixedR100_nom}

678: \end{minipage}%

679: \hspace{2cm} \begin{minipage}[t]{.4\textwidth}

680: \caption{

681: Time evolution of the bubble in the outside coordinates, $x[t]$. The background is assumed

682: homogeneous ($R=100$) and without matter. $x_{init}=20$.

683: Colors correspond to different equations of state; $w=1/3$ red, $w=0$ yellow, $w=-1/3$ green, $w=-1$ blue.

684: } \label{figure_x_fixedR100_nom}

685: \end{minipage}%

686: \end{figure}

687: 

688: In addition, one might wonder what would happen if we did not assume the 

689: bubble to be

690: a vacuum bubble, namely if $w\neq -1$. Then the evolution of the energy 

691: density $\sigma$ on the

692: bubble is governed by (\ref{eomforshell2}). In the absence of dust on the 

693: outside the solution to this equation is:

694: 

695: \eqn{sigmawnot}{\sigma = \frac{ \zeta}{\rho^{2(1+w)}} } 

696: where $\xi$ is a constant. When substituting this solution on 

697: (\ref{EOMR}) we obtain the following equation for $\rho(\tau)$

698: 

699: \eqn{rhoevolwnot}{ \dot{\rho}^2 = -1 +  \left( \frac{ \zeta}{ 4 \rho^{(1+2 w)}} + \frac{\Lambda}{3 \zeta}  \, \rho^{3+2w} \right) ^2 } 

700: For $\rho$  large, the solution to this equation takes the form:

701: 

702: \eqn{rhownot}{ \rho(\tau) \sim \frac{\rho(\tau_0)}{ \left[ 1 - (1+ 2 w) \frac{\Lambda}{3 \zeta} \rho(\tau_0) ^{1 +  2 w} ( \tau-\tau_0) \right] ^{1/(1+2 w)} } }

703: This analysis of the asymptotic behavior is also captured in the results plotted   in Figure \ref{figure_sigma_fixedR100_nom}, and Figure \ref{figure_rho_fixedR100_nom}.

704: 

705: Figure \ref{figure_x_fixedR100_nom} depicts the corresponding bubble evolution in the outside coordinates, $x[t]$. 

706: 

707: {\em In summary}, in this section we have considered a closed FRW 

708: background that will eventually asymptote to a de Sitter space.

709: The outside curvature doesn't affect the evolution of the vacuum bubble 

710: when it is described in terms of the shell coordinates, and therefore,

711: nor in terms of the inside coordinates. The 

712: curvature only makes a difference when the evolution is studied in terms 

713: of the outside coordinates. In these coordinates, the effect of a bigger 

714: curvature is to slow down the evolution of the comoving bubble 

715: coordinate. The asymptotic value of this coordinate depends on the 

716: initial condition but is always smaller that the curvature radius.  When 

717: the equation of state of the shell is not that of vacuum, the evolution 

718: in the shell coordinates is much faster than that of vacuum but is also 

719: independent of the outside curvature. When described from the point of 

720: view of the outside coordinates, this evolution is faster than the 

721: corresponding one for the vacuum shell, but is qualitatively similar. 

722: 

723: 

724: \subsection{Homogeneous Background with Matter}

725: 

726: To study the evolution of the bubble in a space that might contract we have to modify the outside

727: energy density. One option is to introduce matter. In this instance the evolution of the scale parameter

728: is given by

729: 

730: $$ \Big(\frac{da}{dt}\Big)^2 +\frac{1}{R^2} - \frac{\Lambda}{3} a^2- 

731: \frac{A}{a}=0 

732: $$

733: 

734: % \begin{figure}

735: % \includegraphics{POTE}

736: % \caption{This figure represents different forms of the potential according to the three possible values of

737: % A. The blue curve indicates $A>\sqrt{\frac{ 4k^3}{9 \Lambda}$, the green curve corresponds to $A=

738: % \sqrt{\frac{ 4k^3}{9 \Lambda}$ and the red one $A<\sqrt{\frac{ 4k^3}{9 \Lambda}}$}

739: % \label{Potential}

740: % \end{figure}

741: 

742: 

743: If  $ A \leq \frac{1}{R^3} \sqrt{\frac{4}{9 \Lambda}}$ and $a(0)$ is small enough the expansion of the universe will reverse into contraction. With this new form of outside energy density, however, it is no longer consistent with the matching conditions to have a constant surface energy density on the bubble, not even when $\sigma+P=0$. The evolution will be given by the equations:

744: 

745: \begin{eqnarray}

746: \frac{dx}{dt} & = &{\frac{ - \left( 1 - \frac{x^2}{ R^2} \right) 

747: \frac{da/dt}{a} \pm \sqrt{ \left( 1- \frac{x^2}{ R^2} \right) (a^2 

748: B^2 x^2 

749: -1) \left(B^2 - \left(\frac{\Lambda}{3} + \frac{A}{a^3}\right) \right)} }{x a^2 ( B^2- \frac{1}{a^2 R^2})} \label{hom_x} } \\ \nonumber \\ 

750: \frac{d \sigma}{dt} & = & - 2 \left( \frac{da/dt}{a} + 

751: \frac{dx/dt}{x} \right)( \sigma + P ) + 

752: \gamma_{out}\frac{dx/dt}{\sqrt{- 

753: (dx/dt)^2 a^2 + 1 - \frac{x^2}{ R^2}}} \frac{3 A}{a^2} \label{hom_sigma} 

754: \\ 

755: \nonumber \\

756: B & = &  \frac{\sigma}{4} + \frac{1}{ \sigma} \left(\frac{\Lambda}{3}+ \frac{A}{a^3} \right) \label{hom_B}

757: \end{eqnarray}

758: As in the previous section in order to solve these equations an assumption about the equation of state on the bubble was needed. We did make the assumption that the shell is made of a perfect fluid with equation of state $ P = w \, \sigma$, and explored several values for $w$.

759: For the dust energy to be comparable to the cosmological constant at early times, we choose

760: \eqn{matter}{ A  = 10^{-4}}

761: in Planck units. The critical curvature which is the minimal curvature needed for the space to 

762: turn around and eventually collapse is given by:

763: 

764: \eqn{R_cr} {R_{cr} = \left( \frac {9 \Lambda A^2}{4} \right) ^{-1/6} = 107}

765: The position of the maximum of the potential $V[a]$ is given by:

766: 

767: \eqn{a_max} {a_{max}= \left( \frac{3A}{2\Lambda} \right)^{1/3} = 1.7}

768: In our simulations we choose the initial value of the scale factor $a$ to be $a_{init}=a(0) = 1$.

769: 

770: The outside background geometry will be completely specified once we choose

771: the amount of curvature. We will consider several cases.

772: 

773: 

774: \subsubsection {$R > R_{cr}$}

775: 

776: 

777: The universe will always expand. If at the instant the bubble is created $a(0)$ is to the left of max of the potential (\ref{potential}), then the universe will experience a period of slower growth until it eventually goes over the max and the expansion becomes dominated by the cosmological constant. 

778: 

779: We will now investigate bubble propagation on such background. We are again assuming

780: that the inside region has zero energy density, and that

781: $\gamma_{out}=\gamma_{in}=+1$. In this case, the condition (\ref{xi_constraint}) yields:

782: 

783: \eqn{sigma_constr} {\sigma \leq 2 \sqrt { \frac {\Lambda} {3} + \frac {A} {a^3} }}

784: This constraint will be the strongest as $a\rightarrow \infty $, when, as is the case without matter, it reduces to:

785: 

786: \eqn{sigma_lower} { \sigma \leq 6.3 \times 10^{-3}}

787: Once again, we choose:

788: 

789: \eqn{sigma_init} {\sigma_{init} = 10^{-3}}

790: where $\sigma_{init}$ is given in Planck units.

791: From the equation (\ref{hom_x}) the lower and the upper bound on $x$ are:

792: 

793: \eqn{x_range} { \frac {1}{aB} \leq x \leq R }

794: 

795: We will choose the initial size of the bubble in accordance with this range

796: and compare the evolution for several equations of state on the surface of the bubble.

797: Decreasing $R$ (and therefore increasing the curvature of the space) will have the effect

798: of slowing down the evolution, as long as $R > R_{cr}$.

799: 

800: \begin{figure}

801: \centering

802: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

803: \centering

804: \includegraphics[width=6cm]{figure_x_fixed_500}

805: \end{minipage}%

806: \begin{minipage}[b]{.55\textwidth}

807: \centering

808: \includegraphics[width=6cm]{figure_rho_fixed_500}

809: \end{minipage}\\[-10pt]

810: \hspace{-1.2cm}\begin{minipage}[t]{.4\textwidth}

811: \caption{Time evolution of  the bubble in the outside coordinates,

812: $x[t]$, for several initial sizes, $x_{init}=10,50,100$. The background

813: contains homogeneous curvature ($R=500$) and matter

814: ($A=10^{-4}$). Colors correspond to different equations of state;

815: $w=1/3$

816: red, $w=0$ yellow, $w=-1/3$ green, $w=-1$ blue.}\label{figure_x_fixed_500}

817: \end{minipage}%

818: \hspace{1.2cm} \begin{minipage}[t]{.4\textwidth}

819: \caption{Time evolution of the bubble in the bubble coordinates,

820: $\rho[t]$, for the same background conditions as in

821: Figure \ref{figure_x_fixed_500}.

822: Colors correspond to different equations of state; $w=1/3$ red, $w=0$

823: yellow, $w=-1/3$ green, $w=-1$ blue.

824: } \label{figure_rho_fixed_500}

825: \end{minipage}%

826: \end{figure}

827: 

828: First let us consider the case with very little curvature, $R=500$. Figure 

829: \ref{figure_x_fixed_500} shows the time evolution of bubbles of several

830: initial sizes in the outside coordinates. In most cases bubbles grow without

831: reaching the upper bound, $x=R$. Smaller $w$ slows down the expansion rate, and in particular,

832: the choice of $w=-1$ soon leads to a contracting bubble in the outside coordinates.

833: Depending on the initial size of the bubble, $w=-1$ bubbles either reach

834: the lower bound, $x=\frac{1}{aB}$, which results in the breakdown of the simulation,

835: or eventually stabilize and stop contracting.

836: 

837: 

838: Time evolution in the coordinates on the bubble (Figure \ref{figure_rho_fixed_500})

839: reveals behavior similar to the case without matter. Namely, for the bubbles that do survive,

840: the evolution is slower than in the case without matter, but the overall qualitative behavior 

841: stays the same.

842: 

843: \begin{figure}

844: \centering

845: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

846: \centering

847: \includegraphics[width=6cm]{figure_xfull_fixed_120}

848: \end{minipage}%

849: \begin{minipage}[b]{.55\textwidth}

850: \centering

851: \includegraphics[width=6cm]{figure_rhofull_fixed_120}

852: \end{minipage}\\[-10pt]

853: \hspace{-1.2cm}\begin{minipage}[t]{.4\textwidth}

854: \caption{

855: Time evolution of the bubble in the outside coordinates, $x[t]$, for several initial sizes,

856: $x_{init}=10,50,100$. The background contains

857: homogeneous curvature ($R=120$) and matter ($A=10^{-4}$).

858: Colors correspond to different equations of state; $w=1/3$ red, $w=0$ yellow, $w=-1/3$ green, $w=-1$ blue.

859: }\label{figure_x_fixed_120}

860: \end{minipage}%

861: \hspace{1.2cm} \begin{minipage}[t]{.4\textwidth}

862: \caption{

863: Time evolution of the bubble in the bubble coordinates, $\rho[t]$, for the same background conditions as in

864: Figure \ref{figure_x_fixed_120}.

865: Colors correspond to different equations of state; $w=1/3$ red, $w=0$ yellow, $w=-1/3$ green, $w=-1$ blue.

866: } \label{figure_rho_fixed_120}

867: \end{minipage}%

868: \end{figure}

869: 

870: Next we turn to the case with more curvature, but still such that $R > R_{cr}$.

871: From Figure \ref{figure_x_fixed_120} we see that in most cases bubbles again grow, but now

872: they asymptote to the upper bound, $x=R$. 

873: 

874: The corresponding bubble evolution in the coordinates on the bubble is shown in

875: Figure \ref{figure_rho_fixed_120}. The overall effect of more curvature is a further

876: slowdown in the time evolution of the bubbles.

877: 

878: {\em In summary}, in this section we have considered a closed FRW background that contains dust as well as a cosmological constant but that will eventually asymptote to a de Sitter space. In this case it is no longer consistent to have vacuum shells separating de interior and exterior region \footnote{The point of this section is to consider bubble creation for times sufficiently small for the dust density to make a difference. For long times the dust density will redshift and this background will become indistinguishable from the previous one.}. We have studied the motion  for different values of $w$. In shell coordinates, for $w=-1$, we have found that some bubbles do eventually become subcritical or stabilize and stop contracting. 

879: 

880: 

881: \subsubsection {$R < R_{cr}$}

882: 

883: \begin{figure}[htp]

884: \centering

885: \includegraphics [width = 10 cm]{figure_a_crunch_100}

886: \caption{Time evolution of the scale factor $a[t]$ for $R=100$. ($\Lambda =3 \times 10^{-5}, A=10^{-4}$.)}

887: \label{figure_a_crunch_100}

888: \end{figure}

889: 

890: For curvature greater than the critical value (i.e. $R < R_{cr}$) the space will

891: eventually collapse. To investigate this case we choose $R=100$. 

892: The evolution of the scale factor is shown in Figure \ref{figure_a_crunch_100}.

893: 

894: In the outside coordinates, the time evolution of most bubbles will once again

895: lead to the upper bound, $x=R$ (Figure \ref{figure_x_crunch_cont_100}). 

896: Bubbles with $w=-1$ contract and hit the lower bound $x=\frac{1}{aB}$, 

897: which stops their further evolution.

898: 

899: Figure \ref{figure_rho_crunch_cont_100} reveals corresponding behavior in the

900: coordinates on the bubble. Bubbles which do not hit the bound $x=\frac{1}{aB}$

901: eventually collapse along with the collapse of the space itself.

902: 

903: \begin{figure}

904: \centering

905: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

906: \centering

907: \includegraphics[width=6cm]{figure_x_crunch_two_edited}

908: \end{minipage}%

909: \begin{minipage}[b]{.55\textwidth}

910: \centering

911: \includegraphics[width=6cm]{figure_rho_crunch_two_edited}

912: \end{minipage}\\[-10pt]

913: \hspace{-1.4cm}\begin{minipage}[t]{.4\textwidth}

914: \caption{

915: Time evolution of the bubble in the outside coordinates, $x[t]$, for crunching background (Figure \ref{figure_a_crunch_100}).

916: $x_{init}=10$. $R=100$. Colors correspond to different equations of state: $w=1/3$ red, $w=0$ blue.

917: }\label{figure_x_crunch_cont_100}

918: \end{minipage}%

919: \hspace{1.8cm} \begin{minipage}[t]{.4\textwidth}

920: \caption{

921: Time evolution of the bubble in the bubble coordinates, $\rho[t]$, for crunching background

922: (Figure \ref{figure_a_crunch_100}). $x_{init}=10$. $R=100$. Colors correspond to different equations of state; $w=1/3$ red, $w=0$ blue.

923: } \label{figure_rho_crunch_cont_100}

924: \end{minipage}%

925: \end{figure}

926: 

927: {\em In summary}, in this section we have considered a closed FRW background that contains dust and a cosmological constant and that will eventually contract. As in the previous case it is not longer consistent to have vacuum shells separating the inside and outside region. Depending on the initial size, bubbles with smaller value of $w$ become subcritical and disappear while the others continue to grow and eventually collapse along with the space itself. 

928: 

929: \section{\bf Bubble Expansion in Inhomogeneous Backgrounds}

930: 

931: \subsection{Generating a Curvature Profile}

932: 

933: In order to generate an inhomogeneous curvature profile, we choose:

934: 

935: \eqn{R(r)}{R(r)=(\alpha r + \beta R_{cr})\left( 1\pm \frac{1}{\gamma + (\delta -r)^2} \right)}

936: in the definition of the LTB metric. 

937: Choosing $\beta >1$ will allow us to generate a sharp drop in the function $R(r)$, for

938: the cases where the minus sign is chosen. The position and the width of the extremum

939: will be regulated by the parameter $\delta$, and the depth/height by $\gamma$. 

940: %Since we do not want our space to crunch anywhere, we

941: %will require $R(r) > R_{cr}$ for all $r$. 

942: 

943: Furthermore, we want to make sure to satisfy the weak energy condition, namely

944: that the matter density stays positive:

945: 

946: \eqn{d(t,r)}{d(t,r)=\frac{3A}{a^2(t,r) (a(t,r) + r a'(t,r))}}

947: Since $A$ is positive definite, this gives us a condition:

948: 

949: $$a(t,r) + r a'(t,r)>0$$

950: This means that $a(r)$ should nowhere fall faster than $1/r$.

951: 

952: It should be noted that (\ref{R(r)}) incorporates an approximation to a 

953: delta 

954: function in the curvature profile. Since a constant value of $R(r)$ can be 

955: re-absorbed by a variable redefinition in the metric, we are interested 

956: in the physics associated with a spatially varying $R(r)$. In regions 

957: of spacetime where $R(r)$ is roughly constant, we expect that the 

958: evolution 

959: of the bubble would be as in the FRW case\footnote{A crucial ingredient 

960: in coming to this conclusion is the fact that the LTB scale factor can 

961: be solved as an {\em ordinary} differential equation in time $t$, point by 

962: point in the spatial grid. See Section 

963: 2.}. In the next 

964: sub-section, two representative examples which capture the essential 

965: features of these gradient effects are presented.

966: 

967: \subsection{Examples}

968: 

969: 

970: \begin{figure}[htp]

971: \centering

972: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

973: \centering

974: \includegraphics[width=6cm]{a_prof}

975: \end{minipage}%

976: \begin{minipage}[b]{.55\textwidth}

977: \centering

978: \includegraphics[width=6cm]{figure_R_prof_m}

979: \end{minipage}\\[-10pt]

980: \hspace{-1cm}\begin{minipage}[t]{.4\textwidth}

981: \caption{

982: The position dependent scale factor $a(r,t)$ of the the LTB spacetime, for 

983: the curvature profile depicted in red in Figure \ref{figure_R_prof_m}.

984: }\label{a_prof}

985: \end{minipage}%

986: \hspace{2cm} \begin{minipage}[t]{.4\textwidth}

987: \caption{

988: $R(r)$ works as an inhomogeneity profile. Two choices are shown 

989: here, the red curve results in an inhomogeneous spacetime.

990: } \label{figure_R_prof_m}

991: \end{minipage}%

992: \end{figure}

993: 

994: \begin{figure}[htp]

995: \centering

996: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

997: \centering

998: \includegraphics[width=6cm]{figure_x_prof_m}

999: \end{minipage}%

1000: \begin{minipage}[b]{.55\textwidth}

1001: \centering

1002: \includegraphics[width=6cm]{figure_rho_prof_m}

1003: \end{minipage}\\[-10pt]

1004: \hspace{-1cm}\begin{minipage}[t]{.4\textwidth}

1005: \caption{

1006: Evolution of bubbles in outside coordinates, $x(t)$, for the two curvature 

1007: profiles shown in Figure \ref{figure_R_prof_m}.

1008: }\label{figure_x_prof_m}

1009: \end{minipage}%

1010: \hspace{2.0cm} \begin{minipage}[t]{.4\textwidth}

1011: \caption{

1012: Evolution of bubbles in bubble coordinates, $\rho(\tau)$, for the two 

1013: curvature profiles shown in Figure \ref{figure_R_prof_m}.

1014: } \label{figure_rho_prof_m}

1015: \end{minipage}%

1016: \end{figure}

1017: 

1018: As an example, we first choose the minus sign in $R(r)$, with $\alpha =0.1$, $\beta = 3$ and $\delta = 20$.

1019: In order for $R$ to stay above $R_{cr}$, in this case we need $\gamma > 1.5$.

1020: In addition, numerical simulations show that to satisfy the weak energy condition,

1021: $\gamma \geq 2.9$, and since we'd like to investigate the sharpest possible profile,

1022: we choose $\gamma = 2.9$.

1023: 

1024: This choice of the function $R(r)$ generates a profile in the evolution

1025: of the scale factor $a(r,t)$, as shown in Figure \ref{a_prof}. We will compare evolution

1026: of a bubble in such background with evolution in a curvature background 

1027: without a fluctuation in the profile. The two choices of $R(r)$ are shown 

1028: in Figure \ref{figure_R_prof_m}.

1029: The resulting bubble evolutions are shown in Figures \ref{figure_x_prof_m} and \ref{figure_rho_prof_m}.

1030: We see that even though in the outside coordinates bubble evolution is greatly affected by the curvature profile,

1031: such effect is absent for the bubble evolution in the coordinates on the bubble.

1032: 

1033: For comparison, we will also choose $R(r)$ with the plus sign, and

1034: $\alpha =0.1$, $\beta = 2$,$\gamma=0.7$ and $\delta = 20$, depicted in  Figure \ref{figure_R_prof_m_up}.

1035: This results in a profile for $a(r,t)$ shown in Figure \ref{a_prof_up}. 

1036: The corresponding bubble evolution

1037: is shown in Figures \ref{figure_x_prof_m_up} and \ref{figure_rho_prof_m_up}.

1038: Again, we find that the fluctuation does not qualitatively alter the 

1039: bubble evolution.

1040: 

1041: \begin{figure}[htp]

1042: \centering

1043: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

1044: \centering

1045: \includegraphics[width=6cm]{a_prof_up}

1046: \end{minipage}%

1047: \begin{minipage}[b]{.55\textwidth}

1048: \centering

1049: \includegraphics[width=6cm]{figure_R_prof_m_up}

1050: \end{minipage}\\[-10pt]

1051: \hspace{-1cm}\begin{minipage}[t]{.4\textwidth}

1052: \caption{

1053: $a(r,t)$ for curvature profile depicted in red in Figure 

1054: \ref{figure_R_prof_m_up}

1055: }\label{a_prof_up}

1056: \end{minipage}%

1057: \hspace{2cm} \begin{minipage}[t]{.4\textwidth}

1058: \caption{

1059: Second example of an inhomogeneous background: Two choices of $R(r)$ are 

1060: shown.

1061: } \label{figure_R_prof_m_up}

1062: \end{minipage}%

1063: \end{figure}

1064: 

1065: \begin{figure}[htp]

1066: \centering

1067: \hspace{-0.8cm}\begin{minipage}[b]{.5\textwidth}

1068: \centering

1069: \includegraphics[width=6cm]{figure_x_prof_m_up}

1070: \end{minipage}%

1071: \begin{minipage}[b]{.55\textwidth}

1072: \centering

1073: \includegraphics[width=6cm]{figure_rho_prof_m_up}

1074: \end{minipage}\\[-10pt]

1075: \hspace{-1cm}\begin{minipage}[t]{.4\textwidth}

1076: \caption{

1077: Evolution of bubbles in the outside coordinates, $x(t)$, for the two 

1078: curvature 

1079: profiles shown in Figure \ref{figure_R_prof_m_up}.

1080: }\label{figure_x_prof_m_up}

1081: \end{minipage}%

1082: \hspace{2cm} \begin{minipage}[t]{.4\textwidth}

1083: \caption{

1084: Evolution of bubbles in the shell coordinates, $\rho(\tau)$, for the two 

1085: curvature profiles shown in Figure \ref{figure_R_prof_m_up}.

1086: } \label{figure_rho_prof_m_up}

1087: \end{minipage}%

1088: \end{figure}

1089: 

1090: \section{\bf Conclusions}

1091: We have examined the evolution of bubbles of true vacuum in several

1092: backgrounds (that always include a cosmological constant) to assess the

1093: effect of curvature on the propagation. 

1094: %The effect of curvature does seem

1095: %to qualitatively change the bubble evolution when looked at in terms of

1096: %the outside coordinates, but it does not seem important when the

1097: %evolution is studied in terms of the more physically relevant coordinates

1098: %on the shell or its interior. 

1099: The bubble evolution changes qualitatively when looked at in terms of the 

1100: outside coordinates, but when studied in terms of the more physically 

1101: relevant coordinates on the shell or its interior, the local curvature is 

1102: largely irrelevant. In the presence of 

1103: matter, we observe that 

1104: some bubbles, with an equation of state close to $w=-1$, disappear. Those 

1105: that survive do behave qualitatively the same. 

1106: 

1107: The explanation for this insensitivity to inhomogeneities is to

1108: be found in the weak energy condition: our background spacetime is

1109: consistent only when (\ref{d(t,r)}) is positive, which restricts the

1110: sharpness of the inhomogeneity profiles that are allowed. This could of

1111: course be a restriction arising from the ans\"atz for the background that

1112: we are working with. But it should be mentioned that in the work of  Wald

1113: \cite{Wald:1983ky} that was mentioned in the introduction, the

1114: energy conditions played a crucial role in the proof of the cosmic no

1115: hair theorem for anisotropic (but homogeneous) cosmologies.

1116: 

1117: We view this work as an attempt to test the requirements to arrive at a 

1118: universe like ours without making the simplifying assumptions of starting 

1119: with an FRW metric.  There are many directions in which the work here can 

1120: be extended. We have

1121: examined only a few inhomogeneity profiles here, a more thorough

1122: understanding of the evolution curves in more generic situations is

1123: certainly of interest. For all the various initial conditions and 

1124: collapse/inhomogeneity profiles that are allowed, the question of bubble 

1125: evolution should be tractable with our 

1126: machinery\footnote{Though such a scan would admittedly be much more 

1127: numerically challenging than the few cases we have undertaken here.}. 

1128: If we think of our setup as a toy model for inflation, it will 

1129: definitely be of interest to study the {\em genericity} of inflation in 

1130: the space of possible inhomogeneity profiles.  

1131: Another important direction to follow is to evaluate the tunneling 

1132: amplitudes for inhomogeneous ambient spaces. This is a notoriously 

1133: difficult problem on which any enlightenment will be welcome. 

1134: 

1135: Another problem that is of interest in this context is whether there are

1136: signatures left on the new (inside) Universe due to the propagation of the

1137: bubble through the inhomogeneous ambience. The bubble can be thought of

1138: as a Casimir cavity with a moving wall and the cavity radiation (the

1139: analogue of the CMB in this context) will carry imprints of the external

1140: inhomogeneities. Because of the spherical symmetry, this is effectively a

1141: two-dimensional QFT problem, with moving mirror boundary conditions

1142: \cite{Birrell}.

1143: 

1144: 

1145: 

1146: \section{Acknowledgments}

1147: 

1148: CK would like to thank Frank Ferrari for useful discussions, and the 

1149: audience at the Joint ULB-VUB-KUL-Solvay seminar for feedback on a talk 

1150: based on this paper. The work of CK is supported in 

1151: part by IISN - Belgium (convention 4.4505.86), by the Belgian National 

1152: Lottery, by the European Commission FP6 RTN programme MRTN-CT-2004-005104 

1153: in which CK is associated with V. U. Brussel, and by the Belgian Federal 

1154: Science Policy Office through the Interuniversity Attraction Pole P5/27. 

1155: 

1156: The work of W. Fischler, S. Paban and M. \v{Z}ani\'{c} is

1157: partially supported by the National Science Foundation under Grant No. PHY-0455649.

1158:  

1159: \section{\bf Appendix}

1160: 

1161: In this appendix, we will derive the coupled differential equations that

1162: form the starting point for our numerical shell evolution plots. These

1163: equations are a direct consequence of Israel's junctions

1164: conditions\footnote{A pedagogical

1165: derivation of the junction conditions based on a distributional approach

1166: to Einstein's equations

1167: can be found in \cite{Poisson(2004)}.

1168: }, which relate the discontinuity in the extrinsic curvature {\em across}

1169: the shell surface to the energy-momentum layer {\em on} the shell. The

1170: general expressions for these equations for the case

1171: of spherically symmetric junctions

1172: %(see eqns.(\ref{sigma1}-\ref{conservation}))

1173: can be found in,

1174: e.g., \cite{Berezin:1987bc}. Here we discuss a few of the ingredients

1175: that go into our specific problem by doing some illustrative computations

1176: {\em ab initio}. The aim is to stress some issues like choices of signs

1177: and coordinates which are often not adequately addressed in the

1178: litearture.

1179: 

1180: We will work with the mostly minus $\{+,-,-,- \}$ metric,

1181: and the signs on the energy-momentum tensor for the ideal fluid are fixed

1182: by

1183: \eqn{energy}{T_{\alpha}^{\ \beta}=(\epsilon+p)u_{\alpha}

1184: u^{\beta}-p\delta_{\alpha}^{\ \beta}.}

1185: Einstein's equation takes the form $G_{\alpha \beta}=T_{\alpha

1186: \beta}$, in natural units ($8\pi G=c=1$). Later, when we work with

1187: quantities defined on the shell, we will have analogous sign conventions

1188: for them as well. Throughout, Latin indices denote

1189: 3-dimensional objects defined on the

1190: shell hypersurface, Greek indices stand for 4-dimensional

1191: quantities, and semi-colon is shorthand for the covariant derivative. We

1192: will also need

1193: $e^\alpha_a\equiv \frac{\partial x^\alpha}{\partial y^b}$, which  are

1194: projectors

1195: (pull-backs) that can be used to project a 4-quantity on to the 3-surface.

1196: 

1197: The first of Israel's two conditions says that the metric induced on the

1198: shell from the bulk 4-metrics on either side should match, and be equal

1199: to the 3-metric on the shell.

1200: The assumption of spherical

1201: symmetry restricts the form of

1202: the intrinsic metric on the shell to the form,

1203: \eqn{shellmetric}{{ds_3}^2=d\tau^2-{\rho(\tau)}^2d\Omega_2,}

1204: where $\tau$ is the only independent coordinate on the

1205: shell, which we take to be the shell proper time. So looked at from the

1206: outside, on the shell, we can parameterize the

1207: coordinates as

1208: $r=x(\tau)$ and $t=t(\tau)$, and from the inside,

1209: $T=T(\tau)$ and $z=z(\tau)$. Since the metric from either side on the 

1210: shell

1211: should agree with the 3-metric on the shell,

1212: %(the

1213: %first junction condition),

1214: we get two conditions from the inside

1215: (open-FRW),

1216: \eqn{mink-match1}{b(T)z=\rho(\tau), \ \Big(\frac{dT}{d\tau}\Big)^2=

1217: 1+\frac{b(T)^2}{1+z^2}\Big(\frac{dz}{d\tau}\Big)^2,}

1218: and two from the outside (LTB),

1219: \eqn{peeb-match1}{a\big(t,x\big)x=\rho(\tau), \

1220: \Big(\frac{dt}{d\tau}\Big)^2=1+\frac{{\big(a(t,x)x\big),_{x}}^2}{1-x^2/R(x)^2}\Big(\frac{dx}{d\tau}\Big)^2.}

1221: All the variables in these equations are thought of as functions of 

1222: $\tau$.

1223: 

1224: Now we turn to the second junction conditions, which determine the

1225: dynamics of the shell. These are expressed in

1226: terms of the extrinsic curvature:

1227: \eqn{defineK}{K_{ab}=n_{\alpha;\beta} \ e^\alpha_a \ e^\beta_b.}

1228: Here $n_\alpha$ is the outward normal to the

1229: surface under question (since we are working with a closed shell, there

1230: is no ambiguity in making this choice). There

1231: are many different choices and sign

1232: conventions for the extrinsic curvature in the literature, we have made

1233: the above definitions so that the extrinsic curvature of a

1234: 2-sphere is positive. In particular, this differs from

1235: \cite{Berezin:1987bc},

1236: but is in agreement with most of the other references/reviews on the 

1237: subject \cite{Poisson(2004), Lake2, 

1238: LagunaCastillo:1986je, Barrabes:1991ng}. With 

1239: these prescriptions, the

1240: second

1241: junction condition takes the form

1242: \eqn{irael-g}{[K_{ab}]-h_{ab}[K]=S_{ab},}

1243: with $h_{ab}$ denoting the shell metric, which in our case is

1244: (\ref{shellmetric}). The 3-tensor $S_{ab}$ stands for the surface

1245: energy-momentum tensor, and we will assume it to have a perfect-fluid form

1246: analogous to (\ref{energy}). Square brackets stand for discontinuities

1247: across the shell: $[K]=K_{out}-K_{in}$.

1248: 

1249: It can be shown \cite{Berezin:1987bc}, that the junctions conditions imply

1250: a conservation

1251: law of

1252: the form

1253: \eqn{conserve}{{S_a^{\ b}}_{; b}+[e^\alpha_a T^{\beta}_{\alpha}

1254: n_\beta]=0,}

1255: where $T^{\beta}_{\alpha(out/in)}$ is the bulk energy-momentum tensor on 

1256: either

1257: side.

1258: The advantage of this equation is that it is first order, and one can

1259: exchange\footnote{This is analogous to the fact that the covariant

1260: energy conservation equation is an integrability condition for Einstein's

1261: equations.} one of

1262: the second order equations arising from the

1263: junction conditions with (\ref{conserve}). This is indeed what we will

1264: do.

1265: 

1266: %The shell coordinates $y^a$ are $(\tau, \theta, \phi)$ from either side.

1267: 

1268: Because of spherical symmetry and the form of

1269: the

1270: three metric, the extrinsic curvature $K_a^{\ b}$ has

1271: independent components

1272: $K_\tau^{\ \tau}$

1273: and $K_\theta^{\ \theta}=K_\phi^{\ \phi}$, while the surface energy tensor

1274: $S_a^{\ b}$

1275: contains $S_\tau^{\ \tau}\equiv \sigma$ and

1276: $S_\theta^{\ \theta}=S_\phi^{\ \phi}\equiv -P$. So the

1277: independent relations that

1278: arise from the second junction conditions are

1279: \begin{eqnarray}\label{sigmaequation}

1280: -\frac{\sigma}{2}&=&[K_\theta^\theta],\\

1281: P&=&[K_\tau^\tau]+[K_\theta^\theta]\label{pequation}.

1282: \end{eqnarray}

1283: Both $\sigma$ and $P$ are purely functions of

1284: $\tau$ by spherical symmetry. The conservation law (\ref{conserve}) takes

1285: the form,

1286: \eqn{Ton}{\frac{d\sigma}{d\tau}+\frac{2}{\rho}\frac{d\rho}{d\tau}(\sigma+P)

1287: +[T_\tau^n]=0, }

1288: where $[T_\tau^n] \equiv [e^\alpha_\tau T^{\beta}_{\alpha}

1289: n_\beta]$. The evolution of the shell is completely determined by

1290: (\ref{sigmaequation}) and (\ref{Ton}), so our task then is to write down

1291: these differential equations explicitly so that we can proceed with the

1292: numerics.

1293: 

1294: We have different

1295: sets of coordinates in each of the three regions, but not all

1296: of these coordinates can be explicitly written in terms of the others. So 

1297: we will 

1298: write

1299: the matching conditions in terms of the LTB

1300: coordinates. The LTB metric is known only numerically

1301: (see (\ref{EOM}), (\ref{EOMB})), so there

1302: is no hope of writing it in terms of the other coordinates, thereby 

1303: making this

1304: choice inevitable.

1305: Since the evolution is best

1306: understood in the shell coordinates, once we have the evolution curves in

1307: LTB coordinates, we will translate

1308: them numerically to $\rho(\tau)$.

1309: %Our task

1310: %then, is to write the above equations entirely in terms of the Peebles

1311: %coordinates.

1312: 

1313: We start with (\ref{sigmaequation}). 

1314: To calculate the extrinsic curvatures, we need the normal vectors in

1315: the corresponding coordinates. We will start with the LTB side where

1316: the coordinates are $x^\alpha=(t,x,\theta,\phi)$. The projectors are:

1317: \begin{eqnarray}

1318: u^\alpha\equiv

1319: e^\alpha_\tau=&&\hspace{-0.1in}\Big(\frac{dt}{d\tau},\frac{dx}{d\tau},0,0\Big),

1320: \nonumber \\

1321: e^\alpha_\theta=\big(0,0,1,0\big), && \ \

1322: e^\alpha_\phi=\big(0,0,0,1\big). \nonumber

1323: \end{eqnarray}

1324: Since $u^{\alpha}$ is the 4-velocity of the bubble, the normal $n_{\beta}$

1325: will be determined (upto a sign) by the two conditions $u^\alpha

1326: n_\alpha=0$ and $n^\alpha n_\alpha=-1$, where raising and lowering are

1327: done with the LTB metric. The second condition arises because our

1328: shell is timelike: a timelike shell is defined by a spacelike

1329: normal. A simple calculation yields,

1330: \eqn{peeble-normal}{{n_\alpha}=\gamma_{out}\left(\frac{-(ax),_{x}{\dot

1331: x}}{\sqrt{(1-x^2/R^2)}}

1332: %{\dot t}^2-{(ax),_{x}}^2{\dot

1333: %x}^2}}

1334: ,\frac{(ax),_{x}{\dot t}}

1335: {\sqrt{(1-x^2/R^2)}}

1336: %{\dot t}^2-{(ax),_{x}}^2{\dot x}^2}}

1337: ,0,0\right)}

1338: where the dots are with respect to $\tau$. 

1339: %(note that in the main text, 

1340: %dots stand for $t$-derivatives). 

1341: The

1342: choice of sign for the normal is encoded in $\gamma_{out}$, and is fixed

1343: by the condition that the normal should point from the inside to the

1344: outside. For an expanding shell, this

1345: means that $\gamma_{out}=+1$. For a collapsing shell $\gamma_{out}=

1346: -1$. 

1347: 

1348: Using these, the $K^\theta_\theta$ on the LTB side turns out to be,

1349: \eqn{Ktheta+}{ K^{\theta}_{\theta (out)}

1350: =h^{\theta\theta}n_{\theta;\theta}=\frac{1}{\rho^2}\Big(n_{\theta,\theta}-

1351: \Gamma^\alpha_{\theta\theta}n_{\alpha}\Big)

1352: =\gamma_{out}\Big(\frac{(ax),_t(ax),_x {\dot x}+(1-x^2/R^2){\dot 

1353: t}}{\rho

1354: \sqrt{(1-x^2/R^2)}}\Big).}

1355: %{\dot t}^2-{(ax),_{x}}^2{\dot x}^2}}.

1356: We have used (\ref{peeb-match1}) to do some of the simplifications, 

1357: and $(ax),_t$ stands for $(x \partial_t a)$ because at this stage $x$ and 

1358: $t$ are unrelated variables: when we eliminate $\tau$ to write $x$ as a 

1359: function of $t$, this will no longer be the case. 

1360: Repeating the above calculation for 

1361: $K_\theta^\theta$ on the

1362: inside (open-FRW), with coordinates $x^{\alpha}=(T, z, \theta, \phi)$, we

1363: have

1364: \eqn{mink-K}{K_{\theta (in)}^\theta=\gamma_{in}\Big(\frac{z b

1365: \frac{db}{dT}\dot z

1366: +(1+z^2) \dot T

1367: }{\rho\sqrt{1+z^2}}\Big).}

1368: Again, $\gamma_{in}=+1$ when the radius of the inside region is 

1369: increasing; otherwise, $\gamma_{in}=-1$. 

1370: %As long as the transformations relating the

1371: %various coordinate systems are monotonic, the notion of an

1372: %expanding/contracting shell is an invariant one, and this is the case we

1373: %will consider.

1374: The junction condition becomes,

1375: \eqn{mixed}{\gamma_{out}\Big(\frac{(ax),_t(ax),_x {\dot

1376: x}+(1-x^2/R^2){\dot

1377: t}}{\rho

1378: \sqrt{(1-x^2/R^2)}}%{\dot t}^2-{(ax),_{x}}^2{\dot x}^2}}

1379: \Big)-

1380: \gamma_{in}\Big(\frac{z b \frac{db}{dT}\dot z

1381: +(1+z^2) \dot T

1382: }{\rho\sqrt{1+z^2}}\Big)=-\frac{\sigma}{2}.}

1383: By using (\ref{peeb-match1}), it is possible to rewrite this

1384: equation in a more standard form as:

1385: \eqn{standard-junc}{\gamma_{out}\sqrt{{\dot

1386: \rho}^2-\Delta_{out}}-\gamma_{in}\sqrt{{\dot

1387: \rho}^2-\Delta_{in}}=-\frac{\sigma\rho}{2},}

1388: where $\Delta_{out}=-(1-x^2/R^2)+{(ax),_t}^2$, and

1389: $\Delta_{in}=-(1+z^2)+z^2(db/dT)^2$. The quickest way to demonstrate this 

1390: is to start with the final expressions. For instance, for the LTB piece, 

1391: we can expand $\dot\rho$ as $(ax)_{,t}\dot t+(ax)_{,x}\dot x$, use the 

1392: relation

1393: \eqn{crutch}{1-\frac{x^2}{R^2}=\frac{(ax)_{,x}^2\dot x^2}{\dot t^2-1} 

1394: }

1395: once, assemble a perfect square from the pieces, and then use 

1396: (\ref{crutch}) again to end up with the first piece in (\ref{mixed}). 

1397: Equation (\ref{crutch}) here is just a rewriting of (\ref{peeb-match1}). 

1398: An 

1399: analogous transformation can be done for the open-FRW part of the 

1400: equation.

1401: 

1402: The advantage of the form (\ref{standard-junc}) is that now we can use the 

1403: LTB

1404: equation of motion (resp. the open-FRW equation of motion) to simplify the

1405: two terms further to write

1406: \begin{eqnarray}

1407: &&\Delta_{out}=-1+\Big(\frac{A}{a^3}+\frac{\Lambda_{out}}{3}\Big)\rho^2,

1408: \label{delta+}\\

1409: &&\Delta_{in}=-1+\frac{\Lambda_{in}}{3}\rho^2\label{delta-}.

1410: \end{eqnarray}

1411: 

1412: Now, we take up the task of writing these equations purely in the LTB

1413: coordinates, for otherwise, despite being correct, they will be of no

1414: practical value in computations because of the mixing of coordinates.

1415: Using (\ref{delta+}) and (\ref{delta-}), we can

1416: write (\ref{standard-junc}) as

1417: \eqn{rhodot}{{\dot \rho}^2=\rho^2 B^2-1,}where

1418: \eqn{B}{B^2=\frac{\Lambda_{in}}{3}+\Big(\frac{\sigma}{4}+\frac{1}{\sigma}

1419: \Big( \frac{\Lambda_{out}-\Lambda_{in}}{3}+\frac{A}{^3}\Big)\Big)^2.

1420: %\frac{1}{9\sigma^2}\Big(\frac{3\sigma^2}{4}+\Lambda+\frac{3A}{a^3}\Big)

1421: }

1422: This form is useful because we just have to focus on

1423: ${\dot \rho}$ and

1424: $\rho$ because everything else is already manifestly in

1425: LTB coordinates.

1426: %Since $\rho=ax$, the task really is to look express ${\dot \rho}$

1427: %in terms of $x$ and $t$.

1428: The idea is that instead of choosing

1429: $\rho(\tau)$ as the curve for the

1430: bubble-evolution,

1431: %\footnote{actually each point on the evolution-``curve" is

1432: %a sphere because

1433: %the $(\theta, \phi)$ coordinates go along for the ride.}

1434: we want to

1435: parameterize it as $x(t)$. Since $x$ and $t$ are dependent

1436: variables {\em{on}} the shell, this is legitimate. So we write

1437: $\rho=ax$, and rewrite ${\dot \rho}$ as

1438: \eqn{rhodot-xt}{{\dot \rho}=\frac{\Big(x\partial_t

1439: a+(ax),_{x}\frac{dx}{dt}\Big)}{\sqrt{1-\frac{{(ax),_x}^2}{1-x^2/R^2}

1440: \Big(\frac{dx}{dt}\Big)^2}}.}

1441: Using this in (\ref{rhodot}), and using the LTB equation of motion for

1442: $\partial_t a$, we end up with,

1443: \eqn{eomforshell1}{\frac{dx}{dt}=\frac{\left\{\begin{array}{cc}\hspace{-2.25in}-x(1-\frac{x^2}{R^2})\Big(\frac{A}{a}+\frac{\Lambda

1444: a^2}{3}-

1445: \frac{1}{R^2}\Big)^{1/2} \pm\\

1446: \hspace{1in}{\pm}x\Big\{(1-\frac{x^2}{R^2})(a^2x^2B^2-1)\Big(a^2B^2-\frac{A}{a}-\frac{\Lambda

1447: a^2}{3}\Big)\Big\}^{1/2} \end{array}\right\}}

1448: {(ax),_{x}(a^2x^2B^2-x^2/R^2)}}

1449: This is the first of the two shell evolution equations (the other being

1450: (\ref{Ton})) in a form that is

1451: directly applicable

1452: for numerical simulations. The  sign on the square root comes from

1453: the square root of (\ref{rhodot}) and has to be chosen according to

1454: whether

1455: the shell is expanding or contracting.

1456: 

1457: The second of the coupled shell-equations is easily translated into the

1458: LTB coordinates as well. On the LTB side, matter takes the form

1459: dust + cosmological constant (\ref{EOM}). So, using (\ref{peeble-normal}),

1460: \eqn{Ttn+}{T^n_{\tau (out)}\equiv e^t_\tau T^t_t n_t+e^x_\tau T^x_x n_x

1461: =-\frac{\gamma_{out}(ax)_{,x}\frac{dx}{dt}d(x,t)\sqrt{1-x^2/R^2}}

1462: {1-x^2/R^2-(ax)_{,x}^2(dx/dt)^2}.}

1463: On the inside, the only matter is the vacuum energy, which gives rise to

1464: $T_{\tau (in)}^n=0$. These along with (\ref{peeb-match1}) can be used to

1465: write (\ref{Ton}) in the form

1466: \eqn{eomforshell2}{\frac{d\sigma}{dt}+\frac{\Big(x\partial_t

1467: a+(ax),_{x}\frac{dx}{dt}\Big)}{ax}(\sigma+P)-\frac{\gamma_{out}(ax)_{,x}\frac{dx}{dt}d(x,t)}

1468: {\sqrt{1-\frac{x^2}{R^2}-(ax)_{,x}^2(dx/dt)^2}}=0.}

1469: Together with an equation of state relating $\sigma$ to $P$, and the

1470: LTB

1471: equation of motion (\ref{EOM}), 

1472: %to get a numerical handle on the LTB scale factor,

1473: (\ref{eomforshell1}, \ref{eomforshell2}) comprise the starting point

1474: for the

1475: various special cases we study in the main text of this paper.

1476: 

1477: When solving for shell dynamics, we should demand that the

1478: positive energy condition be satisfied, that is

1479: \eqn{pec}{ \sigma > 0. }

1480: Using the formulas derived here (in particular

1481: (\ref{standard-junc})), we can write :

1482: \eqn{deltas} {\Delta_{out} - \Delta_{in} = \frac{\rho^2 \sigma^2}{4} +

1483: \gamma_{out} \rho \sigma ( \dot\rho^2 - \Delta_{out} )^{1/2}, }

1484: which together with (\ref{pec}) implies:

1485: $$ \Delta_{out} - \Delta_{in}  >  \frac{\rho^2 \sigma^2}{4} \hspace{4ex}  

1486: \mbox{if}  \hspace{4ex} \gamma_{out} = +1 $$

1487: $$ \Delta_{out} - \Delta_{in}  <  \frac{\rho^2 \sigma^2}{4} \hspace{4ex}  

1488: \mbox{if}  \hspace{4ex} \gamma_{out} = -1 $$

1489: Furthermore, defining:

1490: 

1491: \eqn{def_xi} {\xi \equiv \frac {4 (\Delta_{out} - \Delta_{in} )} {\rho^2 

1492: \sigma^2} }

1493: and using (\ref{standard-junc}), the following relationship holds:

1494: 

1495: \eqn{xi_constraint} { \gamma_{in} \mid \xi + 1 \mid  - \gamma_{out} \mid 

1496: \xi - 1 \mid = 2 }

1497: 

1498: Depending on the outer and inner geometry, and given the energy densities

1499: both inside and outside the bubble, there will only be a certain range of 

1500: possible

1501: values for the energy density $\sigma$ on the surface of the bubble 

1502: consistent with (\ref{xi_constraint}). We will use this to 

1503: constrain the value of $\sigma$ in the various cases addressed in this 

1504: paper.

1505: 

1506: 

1507: % ==========================================================================

1508: %

1509: 

1510: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1511: %                      REFERENCES                            %

1512: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1513: \newpage

1514: \begin{thebibliography}{19}        %here 19 is the widest mark...

1515: %-----Type it \bibitem[how it is marked]{how we call it}Authors, hep-th/...

1516: %-----Citations are then made by \cite{how we call it} in text ------------

1517: %-----\bibitem without [how it is denoted] is numbered 1,2,3....

1518: 

1519:   %\cite{Israel:1966rt}

1520: \bibitem{Israel:1966rt}

1521:   W.~Israel,

1522:    %``Singular Hypersurfaces And Thin Shells In General Relativity,''

1523:   Nuovo Cim.\ B {\bf 44S10}, 1 (1966)

1524:   [Erratum-ibid.\ B {\bf 48}, 463 (1967\ NUCIA,B44,1.1966)].

1525:   %%CITATION = NUCIA,B44S10,1;%%

1526: 

1527: 

1528: 

1529: %\cite{Berezin:1987bc}

1530: \bibitem{Berezin:1987bc}

1531:   V.~A.~Berezin, V.~A.~Kuzmin and I.~I.~Tkachev,

1532:    %``Dynamics Of Bubbles In General Relativity,''

1533:   Phys.\ Rev.\ D {\bf 36}, 2919 (1987).

1534:   %%CITATION = PHRVA,D36,2919;%%

1535: 

1536: %\cite{Lake:1984pn}

1537: \bibitem{Lake:1984pn}

1538:   K.~Lake,

1539:   %``Equation Of Motion For Bubble Boundaries,''

1540:   Phys.\ Rev.\  D {\bf 29} (1984) 1861.

1541:   %%CITATION = PHRVA,D29,1861;%%

1542: 

1543: %\cite{Khakshournia:2002jr}

1544: \bibitem{Khakshournia:2002jr}

1545:   S.~Khakshournia and R.~Mansouri,

1546:   %``Formation of cosmological mass condensation within a FRW universe: exact

1547:   %general relativistic solutions,''

1548:   Phys.\ Rev.\  D {\bf 65}, 027302 (2002)

1549:   [arXiv:gr-qc/0307023].

1550:   %%CITATION = PHRVA,D65,027302;%%

1551: 

1552: %\cite{Vilenkin:1983xq}

1553: \bibitem{Vilenkin:1983xq}

1554:   A.~Vilenkin,

1555:   %``The Birth Of Inflationary Universes,''

1556:   Phys.\ Rev.\  D {\bf 27}, 2848 (1983).

1557:   %%CITATION = PHRVA,D27,2848;%%

1558: 

1559: %\cite{Hawking:2006ur}

1560: \bibitem{Hawking:2006ur}

1561:   S.~W.~Hawking and T.~Hertog,

1562:   %``Populating the landscape: A top down approach,''

1563:   Phys.\ Rev.\  D {\bf 73}, 123527 (2006)

1564:   [arXiv:hep-th/0602091].

1565:   %%CITATION = PHRVA,D73,123527;%%

1566: 

1567: %\cite{Kachru:2003aw}

1568: \bibitem{Kachru:2003aw}

1569:   S.~Kachru, R.~Kallosh, A.~Linde and S.~P.~Trivedi,

1570:   %``De Sitter vacua in string theory,''

1571:   Phys.\ Rev.\  D {\bf 68}, 046005 (2003)

1572:   [arXiv:hep-th/0301240].

1573:   %%CITATION = PHRVA,D68,046005;%%

1574: 

1575: %\cite{Susskind:2003kw}

1576: \bibitem{Susskind:2003kw}

1577:   L.~Susskind,

1578:   %``The anthropic landscape of string theory,''

1579:   arXiv:hep-th/0302219.

1580:   %%CITATION = HEP-TH/0302219;%%

1581: 

1582: %\cite{Aguirre:2007an}

1583: \bibitem{Aguirre:2007an}

1584:   A.~Aguirre, M.~C.~Johnson and A.~Shomer,

1585:   %``Towards observable signatures of other bubble universes,''

1586:   Phys.\ Rev.\  D {\bf 76}, 063509 (2007)

1587:   [arXiv:0704.3473 [hep-th]].

1588:   %%CITATION = PHRVA,D76,063509;%%

1589: 

1590: 

1591: %\cite{Chang:2007eq}

1592: \bibitem{Chang}

1593:   S.~Chang, M.~Kleban and T.~S.~Levi,

1594:   %``When Worlds Collide,''

1595:   arXiv:0712.2261 [hep-th].

1596:   %%CITATION = ARXIV:0712.2261;%%

1597: 

1598: %\cite{Gibbons:1977mu}

1599: \bibitem{Gibbons:1977mu}

1600:   G.~W.~Gibbons and S.~W.~Hawking,

1601:   %``Cosmological Event Horizons, Thermodynamics, And Particle Creation,''

1602:   Phys.\ Rev.\  D {\bf 15}, 2738 (1977).

1603:   %%CITATION = PHRVA,D15,2738;%%

1604: 

1605: %\cite{Hawking:1981fz}

1606: \bibitem{Hawking:1981fz}

1607:   S.~W.~Hawking and I.~G.~Moss,

1608:   %``Supercooled Phase Transitions In The Very Early Universe,''

1609:   Phys.\ Lett.\  B {\bf 110}, 35 (1982).

1610:   %%CITATION = PHLTA,B110,35;%%

1611: 

1612: %\cite{Starobinsky:1982mr}

1613: \bibitem{Starobinsky:1982mr}

1614:   A.~A.~Starobinsky,

1615:   %``Isotropization of arbitrary cosmological expansion given an effective

1616:   %cosmological constant,''

1617:   JETP Lett.\  {\bf 37}, 66 (1983).

1618:   %%CITATION = JTPLA,37,66;%%

1619:   

1620: %\cite{Wald:1983ky}

1621: \bibitem{Wald:1983ky}

1622:   R.~W.~Wald,

1623:   %``Asymptotic Behavior Of Homogeneous Cosmological Models In The Presence Of A

1624:   %Positive Cosmological Constant,''

1625:   Phys.\ Rev.\ D {\bf 28}, 2118 (1983).

1626:   %%CITATION = PHRVA,D28,2118;%%

1627:   

1628:     %\cite{Banks:2001px}

1629: \bibitem{Banks:2001px}

1630:   T.~Banks and W.~Fischler,

1631:   %``An holographic cosmology,''

1632:   arXiv:hep-th/0111142.

1633:   %%CITATION = HEP-TH/0111142;%%

1634:   

1635:   %\cite{Banks:2003ta}

1636: \bibitem{Banks:2003ta}

1637:   T.~Banks and W.~Fischler,

1638:   %``Holographic cosmology 3.0,''

1639:   Phys.\ Scripta {\bf T117}, 56 (2005)

1640:   [arXiv:hep-th/0310288].

1641:   %%CITATION = PHSTB,T117,56;%%

1642:   

1643: %\cite{Banks:2004cw}

1644: \bibitem{Banks:2004cw}

1645:   T.~Banks, W.~Fischler and L.~Mannelli,

1646:   %``Microscopic quantum mechanics of the p = rho universe,''

1647:   Phys.\ Rev.\  D {\bf 71}, 123514 (2005)

1648:   [arXiv:hep-th/0408076].

1649:   %%CITATION = PHRVA,D71,123514;%%

1650: 

1651: %\cite{Goldwirth:1991rj}

1652: \bibitem{Goldwirth:1991rj}

1653:   D.~S.~Goldwirth and T.~Piran,

1654:   %``Initial conditions for inflation,''

1655:   Phys.\ Rept.\  {\bf 214}, 223 (1992).

1656:   %%CITATION = PRPLC,214,223;%%

1657: 

1658: %\cite{Goldwirth:1989pr}

1659: \bibitem{Goldwirth:1989pr}

1660:   D.~S.~Goldwirth and T.~Piran,

1661:    %``Inhomogeneity And The Onset Of Inflation,''

1662:   Phys.\ Rev.\ Lett.\  {\bf 64}, 2852 (1990).

1663:   %%CITATION = PRLTA,64,2852;%%

1664:   

1665:   %\cite{Goldwirth:1989vz}

1666: \bibitem{Goldwirth:1989vz}

1667:   D.~S.~Goldwirth and T.~Piran,

1668:    %``Spherical Inhomogeneous Cosmologies And Inflation. 1. Numerical Methods,''

1669:   Phys.\ Rev.\ D {\bf 40}, 3263 (1989).

1670:   %%CITATION = PHRVA,D40,3263;%%

1671:   

1672:   %\cite{Deruelle:1994pa}

1673: \bibitem{Deruelle:1994pa}

1674:   N.~Deruelle and D.~S.~Goldwirth,

1675:   %``Conditions for inflation in an initially inhomogeneous universe,''

1676:   Phys.\ Rev.\  D {\bf 51}, 1563 (1995)

1677:   [arXiv:gr-qc/9409056].

1678:   %%CITATION = PHRVA,D51,1563;%%

1679:   

1680: %\cite{Iguchi:1996jy}

1681: \bibitem{Iguchi:1996jy}

1682:   O.~Iguchi, H.~Ishihara and J.~Soda,

1683:   %``Inhomogeneity of spatial curvature for inflation,''

1684:   Phys.\ Rev.\  D {\bf 55}, 3337 (1997)

1685:   [arXiv:gr-qc/9606012].

1686:   %%CITATION = PHRVA,D55,3337;%%

1687:   

1688:   %\cite{Iguchi:1996rh}

1689: \bibitem{Iguchi:1996rh}

1690:   O.~Iguchi and H.~Ishihara,

1691:   %``Onset of inflation in inhomogeneous cosmology,''

1692:   Phys.\ Rev.\  D {\bf 56}, 3216 (1997)

1693:   [arXiv:gr-qc/9611047].

1694:   %%CITATION = PHRVA,D56,3216;%%

1695:   

1696:   %\cite{Lemaitre:1933gd}

1697: \bibitem{Lemaitre:1933gd}

1698:   G.~Lemaitre,

1699:   %``The expanding universe,''

1700:   Gen.\ Rel.\ Grav.\  {\bf 29}, 641 (1997)

1701:   [Annales Soc.\ Sci.\ Brux.\ Ser.\ I Sci.\ Math.\ Astron.\ Phys.\  A {\bf 53}, 51 (1933)].

1702:   %%CITATION = ASSBA,A53,51;%%

1703:   

1704:   %\cite{Tolman:1934za}

1705: \bibitem{Tolman:1934za}

1706:   R.~C.~Tolman,

1707:   %``Effect of inhomogeneity on cosmological models,''

1708:   Proc.\ Nat.\ Acad.\ Sci.\  {\bf 20}, 169 (1934).

1709:   %%CITATION = PNASA,20,169;%%

1710:   

1711:   %\cite{Bondi:1947av}

1712: \bibitem{Bondi:1947av}

1713:   H.~Bondi,

1714:   %``Spherically symmetrical models in general relativity,''

1715:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 107}, 410 (1947).

1716:   %%CITATION = MNRAA,107,410;%%

1717:   

1718: %\cite{Kurki-Suonio:1993fg}

1719: \bibitem{Kurki-Suonio:1993fg}

1720:   H.~Kurki-Suonio, P.~Laguna and R.~A.~Matzner,

1721:   %``Inhomogeneous inflation: Numerical evolution,''

1722:   Phys.\ Rev.\ D {\bf 48}, 3611 (1993)

1723:   [arXiv:astro-ph/9306009].

1724:   %%CITATION = ASTRO-PH 9306009;%%

1725: 

1726: %\cite{Berera:2000xz}

1727: \bibitem{Berera:2000xz}

1728:   A.~Berera and C.~Gordon,

1729:   %``Inflationary initial conditions consistent with causality,''

1730:   Phys.\ Rev.\ D {\bf 63}, 063505 (2001)

1731:   [arXiv:hep-ph/0010280].

1732:   %%CITATION = HEP-PH 0010280;%%

1733:   

1734:   

1735:   

1736:   %\cite{Abbott:1987xq}

1737: \bibitem{Abbott:1987xq}

1738:   L.~F.~Abbott, D.~Harari and Q.~H.~Park,

1739:   %``VACUUM DECAY IN CURVED BACKGROUNDS,''

1740:   Class.\ Quant.\ Grav.\  {\bf 4}, L201 (1987).

1741:   %%CITATION = CQGRD,4,L201;%%

1742:   

1743:   

1744:   

1745: 

1746: 

1747: %\cite{Peebles:1967}

1748: \bibitem{Peebles:1967}

1749:   P.~J.~E.~Peebles,

1750:   %``The Gravitational Instability of the Universe,''

1751:   Astrophys.\ J.\  {\bf 147}, 859 (1967).

1752:   %%CITATION = ASJOA,147,859;%%

1753: 

1754:   

1755:   %\cite{Coleman:1980aw}

1756: \bibitem{Coleman:1980aw}

1757:   S.~R.~Coleman and F.~De Luccia,

1758:   %``Gravitational Effects On And Of Vacuum Decay,''

1759:   Phys.\ Rev.\ D {\bf 21}, 3305 (1980).

1760:   %%CITATION = PHRVA,D21,3305;%%

1761:   

1762:   

1763:   

1764:   %\cite{Farhi:1986ty}

1765: \bibitem{Farhi:1986ty}

1766:   E.~Farhi and A.~H.~Guth,

1767:   %``AN OBSTACLE TO CREATING A UNIVERSE IN THE LABORATORY,''

1768:   Phys.\ Lett.\ B {\bf 183}, 149 (1987).

1769:   %%CITATION = PHLTA,B183,149;%%

1770: 

1771: \bibitem{farhi2}

1772:   E.~Farhi, A.~H.~Guth and J.~Guven,

1773:   %``IS IT POSSIBLE TO CREATE A UNIVERSE IN THE LABORATORY BY QUANTUM

1774:   %TUNNELING?,''

1775:   Nucl.\ Phys.\ B {\bf 339}, 417 (1990).

1776:   %%CITATION = NUPHA,B339,417;%%

1777: 

1778: \bibitem{fischler1}

1779:   W.~Fischler, D.~Morgan and J.~Polchinski,

1780:   %``QUANTUM NUCLEATION OF FALSE VACUUM BUBBLES,''

1781:   Phys.\ Rev.\ D {\bf 41}, 2638 (1990).

1782:   %%CITATION = PHRVA,D41,2638;%%

1783: 

1784: \bibitem{fischler2}

1785:   W.~Fischler, D.~Morgan and J.~Polchinski,

1786:   %``QUANTIZATION OF FALSE VACUUM BUBBLES: A HAMILTONIAN TREATMENT OF

1787:   %GRAVITATIONAL TUNNELING,''

1788:   Phys.\ Rev.\ D {\bf 42}, 4042 (1990).

1789:   %%CITATION = PHRVA,D42,4042;%%

1790:   

1791: \bibitem{Poisson(2004)} E. Poisson, ``A

1792: Relativist's Toolkit : The Mathematics of Black-hole Mechanics",

1793: Cambridge, UK: Cambridge University Press, 2004

1794: 

1795: \bibitem{Lake2} K. Lake, in {\it 8th 

1796: Brazilian School of Cosmology and Gravitation}, edited by M. Novello

1797: (World Scientific, Singapore, 1988), pp. 1-82.

1798: 

1799: %\cite{LagunaCastillo:1986je}

1800: \bibitem{LagunaCastillo:1986je}

1801:   P.~Laguna-Castillo and R.~A.~Matzner,

1802:   %``Inflation And Bubbles In General Relativity,''

1803:   Phys.\ Rev.\  D {\bf 34}, 2913 (1986).

1804:   %%CITATION = PHRVA,D34,2913;%%

1805: 

1806: %\cite{Barrabes:1991ng}

1807: \bibitem{Barrabes:1991ng}

1808:   C.~Barrabes and W.~Israel,

1809:   %``Thin shells in general relativity and cosmology: The Lightlike limit,''

1810:   Phys.\ Rev.\  D {\bf 43}, 1129 (1991).

1811:   %%CITATION = PHRVA,D43,1129;%%

1812:   

1813: \bibitem{Birrell}

1814: N.~D.~Birrell and P.~C.~W.~Davies, ``Quantum Fields in Curved Space,"

1815: Cambridge (1999). 

1816:   

1817: \end{thebibliography}

1818: \end{document}

1819: