1: 

2: 

3: \documentclass[preprint,aps,floats,subeqn,showpacs]{revtex4}

4: \usepackage{epsfig}

5: % Uncomment next two lines for A4 paper size, comment for letter size

6: %\addtolength{\textheight}{17.6mm}

7: 

8: %%%%% number equations by section %%%%%%%%

9: %\makeatletter

10: %\@addtoreset{equation}{section}

11: %\makeatother

12: 

13: % Command Definitions

14: 

15: \def\real{I\negthinspace R}

16: \def\zed{Z\hskip -3mm Z }

17: \def\half{\textstyle{1\over2}}

18: \def\quarter{\textstyle{1\over4}}

19: \def\sech{\,{\rm sech}\,}

20: \def\ie{{\it i.e.,}}

21: \newcommand{\be}{\begin{equation}}

22: \newcommand{\ee}{\end{equation}}

23: \newcommand{\bea}{\begin{eqnarray}}

24: \newcommand{\eea}{\end{eqnarray}}

25: \newcommand{\bml}{\begin{mathletters}}

26: \newcommand{\eml}{\end{mathletters}}

27: \newcommand{\re}[1]{(\ref{#1})}

28: %\newcommand{\bml}{\begin{subequations}}

29: %\newcommand{\eml}{\end{subequations}}

30: %

31: \def\aprle{\buildrel < \over {_{\sim}}}

32: \def\aprge{\buildrel > \over {_{\sim}}}

33: 

34: \begin{document}

35: 

36: \tighten

37: 

38: \preprint{IUB-TH-041}

39: \draft

40: 

41: %\twocolumn[\hsize\textwidth\columnwidth\hsize\csname @twocolumnfalse\endcsname

42: 

43: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

44: 

45: %\wideabs{    % Uncomment this line for two-column output

46: 

47: \title{ Spherically symmetric Yang-Mills solutions  in

48: a $(4+n)$- dimensional space-time}

49: \renewcommand{\thefootnote}{\fnsymbol{footnote}}

50: \author{ Yves Brihaye\footnote{Yves.Brihaye@umh.ac.be}}

51: \affiliation{Facult\'e des Sciences, Universit\'e de Mons-Hainaut,

52: 7000 Mons, Belgium}

53: \author{Fabien Clement}

54: \affiliation{Facult\'e des Sciences, Universit\'e de Mons-Hainaut,

55: 7000 Mons, Belgium}

56: \author{Betti Hartmann\footnote{b.hartmann@iu-bremen.de}}

57: \affiliation{School of Engineering and Sciences, International University Bremen (IUB),

58: 28725 Bremen, Germany}

59: 

60: \date{\today}

61: \setlength{\footnotesep}{0.5\footnotesep}

62: 

63: 

64: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

65: \begin{abstract}

66: We consider the Einstein-Yang-Mills Lagrangian

67: in a $(4+n)$-dimensional space-time.

68: Assuming the matter and metric fields to be 

69: independent of the $n$

70: extra coordinates, a spherical symmetric Ansatz for

71: the fields leads to a set of

72: coupled ordinary differential equations. We find that

73: for $n > 1$ only solutions with either  one non-zero Higgs field or

74: with all Higgs fields constant exist.

75: We construct the

76: analytic solutions which fulfill this conditions

77: for arbitrary $n$, namely the Einstein-Maxwell-dilaton solutions.

78: 

79: We also present generic solutions of the

80: effective

81: 4-dimensional Einstein-Yang-Mills-Higgs-dilaton model, which

82: possesses $n$ Higgs triplets coupled in a specific way to

83: $n$ independent dilaton fields. These solutions are the abelian Einstein-Maxwell-

84: dilaton solutions and analytic non-abelian solutions, which have

85: diverging Higgs fields. In addition, we construct numerically 

86: asymptotically flat and finite energy solutions for $n=2$. 

87: \end{abstract}

88: 

89: \pacs{04.20.Jb, 04.40.Nr, 04.50.+h, 11.10.Kk }

90: \maketitle

91: 

92: 

93: \section{Introduction}

94: In an attempt to unify electrodynamics and general relativity,

95: Kaluza introduced an extra - a fifth dimension \cite{kaluza} and assumed all

96: fields to be independent of the extra dimension.

97: Klein \cite{klein} followed this idea, however, he assumed the fifth dimension

98: to be compactified on a circle of Planck length. The resulting theory

99: describes $4$-dimensional Einstein gravity plus Maxwell's equations. 

100: One of the new fields appearing in this model

101: is the dilaton, a scalar companion of the metric tensor.

102: In an analogue way, this field arises in the

103: low energy effective action of superstring theories and is

104: associated with the classical scale invariance of these models \cite{maeda}.

105: 

106: When studying spherically symmetric solutions in higher dimensional systems,

107: two possible approaches seem possible: a) to assume the solutions

108: to be spherically symmetric in the full $d$ dimensions or b) to assume

109: the solutions to be spherically symmetric only in 4 dimensions. The solutions

110: obtained in the first approach are surely important at very high energies, i.e.

111: very early stages of the universe. The second approach, which assumes the extra dimensions

112: to be ``spectator'' is of importance for physics in the universe today.

113: 

114: Volkov recently followed the idea of extra dimensions and

115: constructed a $(4+1)$ dimensional Einstein-Yang-Mills

116: (EYM) system \cite{volkov}. It turned out that the EYM particles

117: are completely destroyed by gravity, but that, however, particle-like solutions

118: (so-called ``EYM vortices'') exist if one assumes the fields to be independent of the extra coordinate.

119: The system then reduces to an effective $4$-dimensional

120: Einstein-Yang-Mills-Higgs-dilaton (EYMHD) model, which was studied in

121: detail in \cite{bh1}. Through the dimensional reduction, one Higgs triplet

122: and a dilaton field appears in the model.

123: 

124: In this paper, we study the system of \cite{volkov} in $(4+n)$ dimensions, assuming all

125: fields to be independent of the $n$ extra coordinates.

126: We give the model and equations of motion in Section II. The solutions are

127: spherically symmetric in 4 dimensions, while the extra dimensions

128: (which are ``spectator'') are associated with a Ricci flat manifold.

129: It turns out that in $n > 1$ dimensions only solutions

130: with one non-zero Higgs field or with all Higgs fields constant exist. 

131: We give the analytic solutions which fulfill this condition 

132: for generic $n$, namely the abelian Einstein-Maxwell-dilaton (EMD).

133: In Section III, we present the 4-dimensional

134: effective Einstein-Yang-Mills-Higgs-dilaton (EYMHD) model, which has

135: $n$ independent Higgs and dilaton fields. Since in this model, no constraint

136: on the Higgs fields arises, we construct generic solutions.

137: We give the analytic solutions available, namely the abelian 

138: Einstein-Maxwell-dilaton (EMD) solutions and ``non-abelian'' solutions with

139: diverging Higgs fields.

140: Finally, we present our numerical solutions of the effective

141: model for $n=2$.

142: We give our conclusions in Section IV.

143: 

144: 

145: 

146: 

147: 

148: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

149: \section{The $(4+n)$-dimensional Einstein-Yang-Mills model}

150: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

151: 

152: The Einstein-Yang-Mills Lagrangian

153: in $d=(4+n)$ dimensions is

154: given by:

155: 

156: \begin{equation}

157: \label{action}

158:   S = \int \Biggl(

159:     \frac{1}{16 \pi G_{(4+n)}} R   - \frac{1}{4 e^2}F^a_{M N}F^{a M N}

160:   \Biggr) \sqrt{g^{(4+n)}} d^{(4+n)} x

161: \end{equation}

162: with the SU(2) Yang-Mills field strengths

163: $F^a_{M N} = \partial_M A^a_N -

164:  \partial_N A^a_M + \epsilon_{a b c}  A^b_M A^c_N$

165: , the gauge index

166:  $a=1,2,3$  and the space-time index

167:  $M=0,...,(4+n)-1$. $G_{(4+n)}$ and $e$ denote

168: respectively the $(4+n)$-dimensional Newton's constant and the coupling

169: constant of the gauge field theory. $G_{(4+n)}$ is related to the Planck mass

170: $M_{pl}$ by $G_{(4+n)}=M_{pl}^{-(2+n)}$ and $e^2$ has the dimension of $[{\rm length}]^n$.

171: In the following, we denote the coordinates $x_{(3+k)}$ by $y_k$ with 

172: $k=1,...,n$.

173: 

174: If both the matter functions and the metric functions

175: are independent on $y_k$, the fields can be

176: parametrized as follows:

177: \begin{equation}

178: g^{(4+n)}_{MN}dx^M dx^N = 

179: e^{-\Xi}g^{(4)}_{\mu\nu}dx^{\mu}dx^{\nu}

180: +\sum_{k=1}^n e^{2\zeta_k} (dy^k)^2 

181: \ , \ \mu , \nu=0, 1, 2, 3 

182: \end{equation}

183: with

184: \begin{equation}

185: \Xi=\sum_{k=1}^n \zeta_k 

186: \end{equation}

187: and

188: \begin{equation}

189: A_M^{a}dx^M=A_{\mu}^a dx^{\mu}+ \sum_{k=1}^n \Phi_k^a dy^k   \  . \

190: \end{equation}

191: $g^{(4)}$ is the $4$-dimensional metric tensor 

192: and the $\zeta_{j}$ and  $\Phi_j^a$, $j=1,...,n$,  

193: play the

194: role of  dilatons and Higgs fields, respectively. 

195:  

196: Note that in the case $n=1$, the above parametrization coincides

197: with the one in \cite{volkov}.

198: 

199: 

200: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

201: \subsection{Spherically symmetric Ansatz}

202: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

203: 

204: For the metric the spherically symmetric Ansatz

205: in Schwarzschild-like coordinates reads \cite{weinberg}:

206: \begin{equation}

207: ds^{2}=g^{(4)}_{\mu\nu}dx^{\mu}dx^{\nu}=

208: -A^{2}(r)N(r)dt^2+N^{-1}(r)dr^2+r^2 d\theta^2+r^2\sin^2\theta

209: d^2\varphi

210: \label{metric}

211: \ , \end{equation}

212: with

213: \begin{equation}

214: N(r)=1-\frac{2m(r)}{r}

215: \ . \end{equation}

216: In these coordinates, $m(\infty)$ denotes the (dimensionful) mass of

217: the field configuration.\

218: 

219: For the gauge and Higgs fields, we use the purely magnetic hedgehog ansatz

220: \cite{thooft} :

221: \begin{equation}

222: {A_r}^a={A_t}^a=0

223: \ , \end{equation}

224: \begin{equation}

225: {A_{\theta}}^a= (1-K(r)) {e_{\varphi}}^a

226: \ , \ \ \ \

227: {A_{\varphi}}^a=- (1-K(r))\sin\theta {e_{\theta}}^a

228: \ , \end{equation}

229: \begin{equation}

230: \label{higgsansatz}

231: {\Phi}^a_{j}=c_j v H_j(r) {e_r}^a \ \ , \ \ j=1,...,n \ ,

232: \end{equation}

233: where $v$ is a mass scale, while $c_j$ are dimensionless constants

234: determining the vacuum expectation values of the Higgs fields 

235: $\langle \Phi_j \rangle = c_j v$.

236: In absence of a Higgs potential these have to be set by hand.

237: Finally, the dilatons are scalar fields depending only on $r$~:

238: \begin{equation}

239: \zeta_j =\zeta_j(r) \ \ , \ \ k = 1,...,n 

240: \ . \end{equation}

241: 

242: 

243: 

244: 

245: With these conventions, the non-vanishing components of the 

246: $(4+n)$-dimensional

247: energy momentum tensor read:

248: \begin{eqnarray}

249: & T_0^0 &= - e^{\Xi}

250: \left(\sum_{k=1}^n (A_k  + C_k ) + B + D\right) \ , \nonumber \\

251: & T_1^1 &= - e^{\Xi}

252: \left(\sum_{k=1}^n (-A_k  + C_k ) - B + D\right) \ , \nonumber \\

253: & T_2^2 &= T_3^3 = - e^{\Xi}

254: \left(\sum_{k=1}^n A_k  - D\right) \ ,  \nonumber \\

255: & T_j^j &= - e^{\Xi}

256: \left(-2 (A_j+C_j) 

257: + \sum_{k=1}^n (A_k + C_k) + B + D\right) \ , \ j=1,...,n \ ,

258: \nonumber \\

259: & T_j^k &=  e^{\Xi} e^{-(\zeta_j + \zeta_k)} \left(\frac{N}{2} H'_j H'_k

260:                                        + \frac{K^2}{x^2} H_j H_k\right)

261: \ \ , \ \ j \neq k  \ \ , \

262: \end{eqnarray}

263:  where we use the abbreviations

264:  \be

265:  A_j = \frac{1}{2} e^{-2 \zeta_j} N (H_j')^2 \ \ , \ \

266:  C_j = e^{-2 \zeta_j} K^2 H_j^2 \frac{1}{x^2} \ \  \ \

267:  \ee

268:  and

269:  \be

270:   B =  e^{\Xi} \frac{1}{x^2} N (K')^2 \ \ , \ \

271:   D =  e^{\Xi} \frac{1}{2 x^4}  (K^2 - 1)^2

272:  \ee

273:  The radial variable  $x=e v r$ was introduced and

274:  the prime denotes the  derivative with respect to $x$. 

275: 

276: \subsection{Equations of motion}

277: 

278: Defining the coupling constant $\alpha = v \sqrt{G_{(4+n)}}$ and the 

279: mass function $\mu=evm$ 

280: we obtain two first order equations for the two metric functions

281: \be

282: \label{mu}

283: \mu' = \alpha^2 x^2\left(\sum_{k=1}^n (A_k+C_k) 

284: + B + D\right) + \frac{1}{2}N x^2 \Theta  \ ,

285: \ee

286: \be

287: \label{A}

288:    A' = \alpha^2 A x \left(\sum_{k=1}^n e^{-2\zeta_k} (H_k')^2 

289:    + 2 e^{\Xi} \frac{K'^2}{x^2}\right) + x A \Theta

290: \ee

291: with 

292: \be

293:   \Theta \equiv \frac{3}{4}\left(\sum_{k=1}^n (\zeta_k')^2 

294: + \frac{2}{3} \sum_{k > k'} \zeta_{k}' \zeta_{k'}' \right)  \ .

295: \ee

296: The components of the Einstein equations related to the extra dimensions lead

297: (after suitable linear combinations) to equations for the

298: $n$ dilaton fields~:

299: \begin{equation}

300: \label{zeta}

301: (x^2 A N \zeta_j ')'

302: = \alpha^2 A x^2 e^{-\Xi}

303: \left[ \frac{2}{n+2}\sum_{M=0}^{d-1} T_M^M -2 T_j^j\right]  \  \ , \ \ j=1,...,n \ .

304: \end{equation}

305: The field equations for the gauge field and the

306: $n$ Higgs fields read respectively

307: \be

308: \label{k}

309:    (e^{\Xi}ANK')'

310:    = A\left( \frac{1}{x^2}e^{\Xi}K(K^2-1)

311:        + \sum_{k=1}^n e^{-2 \zeta_k} K H_k^2\right)  \ ,

312: \ee

313: \be

314: \label{h}

315:    (e^{-2\zeta_j}x^2 A N H_j')' = 2 A e^{-2 \zeta_j} K^2 H_j \ \ , \ \

316:    j=1,...,n   \ .

317: \ee

318: Finally, since the off-diagonal components of the

319: Einstein tenor vanish, we obtain an extra constraint on the 

320: fields from the $jk$-components of the energy-momentum tensor:

321: \begin{equation}

322: \label{constraint}

323: e^{\Xi} e^{-(\zeta_j + \zeta_k)} \left(\frac{N}{2} H'_j H'_k

324:                                        + \frac{K^2}{x^2} H_j H_k\right)=0

325: \end{equation}

326: Clearly, these equations are symmetric under the simultaneous

327: exchange $\zeta_{j'},  H_{j'} \leftrightarrow \zeta_j, H_j$. Further, they

328: are invariant under the rescaling 

329: \be

330:    H_{j} \rightarrow {\lambda}^{-1} H_j \ \ , \ \

331:    x \rightarrow \lambda x \ \ , \ \

332:    \alpha \rightarrow \lambda^2 \alpha  \ , \ \ \ \lambda \in \real  \ .

333: \ee

334: 

335: The constraint (\ref{constraint}) will only be fulfilled

336: for specific cases. 

337: With our parametrisation, there are two possibilities: 

338: (i) only {\it one} non-zero Higgs field and arbitrary gauge field $K(x)$

339: or (ii) constant Higgs fields and $K(x)=0$.

340: 

341: Since we are interested in generic solutions,

342: we go back to the 4-dimensional effective action motivated by this model

343: in which, of course, off-diagonal terms of the energy-momentum tensor

344: don't appear. Before we discuss this effective model, we give the

345: solutions with constant Higgs fields available in the ``full'' model, namely

346: the Einstein-Maxwell-dilaton (EMD) solutions.

347: 

348: Let us remark that there are several possibilities of modification

349: of the ``full'' model considered here, 

350: which could eventually lead to solutions with generic Higgs and gauge fields, 

351: namely we could

352: 

353:  (i) add a non-diagonal term of the form $f_{56}dx_5 dx_6$

354: in the metric, this leads to one extra equation for the $f_{56}$

355: function, but it has to be checked that this function stays regular,

356: 

357: (ii) start with an SU(N) gauge group and choose the different Higgs

358: fields in orthogonal SU(2) subalgebras of the Lie algebra of SU(N).

359: 

360: 

361: 

362: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

363: \subsection{Einstein-Maxwell-Dilaton (EMD) solutions}

364: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

365: Assuming the matter fields to be constant:

366: \begin{equation}

367: K(x)=0 \ , \ H_j(x)=c_j \ , \ j=1,...,n

368: \end{equation}

369: and 

370: \begin{equation}

371: \zeta_j(x)=\zeta(x) \ \ \ \forall \  j  \ ,

372: \end{equation}

373: we find that the above equations (\ref{mu}), (\ref{A}), (\ref{zeta})

374: admit exact solutions which are related to the Einstein-Maxwell-Dilaton (EMD) solutions 

375: \cite{maeda}. Here, we will only discuss the extremal case, which corresponds to

376: a solution with horizon at the origin. 

377: 

378: The extremal EMD solutions have unit magnetic charge and mass

379: \begin{equation}

380: \label{massEMD}

381: \frac{\mu_{\infty}}{\alpha^2}=\frac{1}{\alpha} \sqrt{\frac{n+2}{2(n+1)}} \ .

382: \end{equation}

383: The value of the metric component $N(x)$ at the origin $x=0$ reads:

384: \begin{equation}

385: \label{NEMD}

386: N_{\small EMD}(0)=\left(\frac{n}{2(n+1)}\right)^2

387: \end{equation}

388: and the dilaton field is:

389: \begin{equation}

390: \zeta_{\small EMD}(x)=\frac{1}{n+1} \ln \left(1-\frac{X_-}{X}\right)

391: \end{equation}

392: with

393: \begin{equation}

394: X_-=\left(\frac{2(n+1)}{n+2}\right)^{1/4} \ \ , \ \ \frac{x}{\alpha}=X

395: \left(1-\frac{X_-}{X}\right)^{n/(2n+2)}  \ .

396: \end{equation}

397: 

398: 

399: 

400: 

401: 

402: 

403: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

404: \section{The effective $4$-dimensional Einstein-Yang-Mills-Higgs-Dilaton (EYMHD) model}

405: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

406: As in the 5-dimensional case \cite{volkov} (in our notation $n=1$) the 

407: equations given in the previous section (apart from the constraint

408: (\ref{constraint})) can equally well be derived 

409: from an effective 4-dimensional

410: Einstein-Yang-Mills-Higgs-Dilaton (EYMHD) Lagrangian. 

411: The Lagrangian density for the matter fields then reads:

412: \begin{eqnarray}

413: \label{effeL}

414:    L_M =

415: &-& \frac{1}{4} e^{2 \kappa \Gamma }F^a_{\mu \nu} F^{a, \mu \nu}

416: - \sum_{k=1}^n 

417: \frac{1}{2} e^{-4 \kappa \Psi_k} D_{\mu} \Phi_1^a D^{\mu} \Phi_1^a

418: \nonumber \\

419: &-& \frac{1}{2} \left(\sum_{k=1}^n \partial_{\mu} \Psi_k  \partial^{\mu} \Psi_k               

420:         +\frac{2}{3} \sum_{k > k'} \partial_{\mu} \Psi_k  \partial^{\mu} 

421: \Psi_{k'}\right)

422: \end{eqnarray}

423: with

424: \begin{equation}

425: \Gamma=\sum_{k=1}^n \Psi_k  \ .

426: \end{equation}

427: The kinetic part in the dilaton fields could be diagonalized, however,

428: we find it more convenient to leave it in the form above which reveals

429: the symmetry $\Psi_j, \Phi_j \leftrightarrow \Psi_{j'}, \Phi_{j'}$.

430: 

431: The Lagrangian (\ref{effeL}) is then coupled minimally to Einstein gravity

432: according to the full action

433: \begin{eqnarray}

434: \label{action4}

435:               S &=& S_G + S_M \nonumber \\

436:               &=& \int \sqrt{-g^{(4)}} \left(L_G + L_M \right)  d^4 x

437: \end{eqnarray}

438: where $L_G = R/(16 \pi G_4)$, $R$ is the Ricci scalar

439: and $G_4$ is the $4$-dimensional Newton's constant.

440: 

441: Note that the dilaton fields are coupled by an independent coupling

442: constant $\kappa$ to the gauge and Higgs fields. In this respect, the

443: dilatons here are treated as independent scalar fields, while in

444: the action (\ref{action}) they appear as parts of the metric tensor.

445:  

446: After the rescaling

447: \begin{equation}

448:           \Psi_j  =  v \psi_j \ \ , \ \ \kappa = \frac{\gamma}{v} \ \ , \ \

449: \alpha=v\sqrt{G_4}

450: \end{equation}

451: the resulting set of equations only depends on the coupling constants

452: $\alpha$ and $\gamma$. We refrain from giving the explicit form of the

453: equations here, but refer the reader to \cite{bh1} for the case $n=1$.

454: 

455: Note that the equations (\ref{mu}), (\ref{A}), (\ref{zeta}), (\ref{k}), (\ref{h})

456: become equivalent to the field equations associated to (\ref{action4})

457: by using the same Ans\"atze for the $4$-dimensional metric, the gauge and

458: Higgs fields, but by identifying 

459: \begin{equation}

460: \zeta_j=2\gamma \psi_j=2\kappa \Psi_j \ \ , \ \ \alpha^2=3\gamma^2  \ . 

461: \end{equation} 

462: Remarkably, this identification turns out to be independent on $n$.

463: 

464: This model now has solutions for which 

465: all $n$ Higgs fields can be non-constant. First, we will present the

466: 4-dimensional effective

467: counterparts of the abelian Einstein-Maxwell-dilaton solutions. Then, we will

468: present analytic non-abelian solutions with non-constant but

469: diverging Higgs fields, which are not available in the full model

470: since they don't fulfill the constraint (\ref{constraint}). Finally, we will give

471: the asymptotically flat, finite energy solutions for $n=2$, which we construct

472: numerically.  

473: 

474: 

475: 

476: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

477: \subsection{Einstein-Maxwell-Dilaton (EMD) solutions}

478: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

479: 

480: Assuming the matter fields to be constant:

481: \begin{equation}

482: K(x)=0 \ , \ H_k(x)=c_k \ , \ k=1,...,n

483: \end{equation}

484: and 

485: \begin{equation}

486: \psi_k(x)=\psi(x) \ \ \ \forall \  k   \ ,

487: \end{equation}

488: we find that the equations associated to the effective action (\ref{action4})

489: admit exact solutions which are related to the Einstein-Maxwell-Dilaton (EMD) solutions 

490: \cite{maeda}. As in the previous section, 

491: we will only discuss the extremal case. Note that now, of course, the solution

492: depends on both $\alpha$ and $\gamma$. 

493: 

494: The extremal EMD have unit magnetic charge and mass

495: \begin{equation}

496: \frac{\mu_{\infty}}{\alpha^2}=

497: \frac{1}{\sqrt{\alpha^2 + \tilde{\gamma}^2}} \ , 

498: \end{equation}

499: where 

500: \begin{equation}

501: \tilde{\gamma}^2=\frac{3n}{2+n} \gamma^2  \ .

502: \end{equation}

503: The value of the metric component $N(x)$ at the origin $x=0$ reads:

504: \begin{equation}

505: N_{\small EMD}(0)=\left(\frac{\tilde{\gamma}^2}{\alpha^2 + \tilde{\gamma}^2}

506: \right)^2

507: \end{equation}

508: and the dilaton field is:

509: \begin{equation}

510: \psi_{\small EMD}(x)=\left(\frac{3}{2+n}\right)\frac{1}{\alpha^2 + \tilde{\gamma}^2}  

511: \ln \left(1-\frac{X_-}{X}\right)

512: \end{equation}

513: with

514: \begin{equation}

515: X_-=\left(\frac{\alpha^2+\tilde{\gamma}^2}{\alpha^2

516: }\right)^{1/4} \ \ , \ \ \frac{x}{\alpha}=X  

517: \left(1-\frac{X_-}{X}\right)^{\tilde{\gamma}^2/(\alpha^2+\tilde{\gamma}^2) }

518: \end{equation}

519: 

520: 

521: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

522: \subsection{Analytic, non-abelian solutions}

523: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

524: 

525: Very similar to the $n=1$ case \cite{volkov} non-abelian, 

526: analytic solutions exist in this system. If we assume the limit

527: $\alpha^2=3\gamma^2$ (which, of course, is the most interesting one),

528: we find solutions of the form:

529: \begin{eqnarray}

530: \label{ads}

531: & & N(x)=c_1 \ \ , \ \ A = x^{(1+c_2)} \ \ , \ \  \psi_1(x)=\psi_2(x)

532: =...\equiv\psi(x)=\frac{4\alpha}{\sqrt{3}n} \ln\left(\frac{x}{c_3}\right) \nonumber \\

533: &  & H_1(x)=H_2(x)=...=H(x)=\sqrt{c_4} \left(\frac{x}{c_3}\right)^{2/n} 

534: \ \ \ , \ \  K(x)=\sqrt{q}

535: \end{eqnarray}

536: where $c_1$, $c_2$, $c_3$ and $c_4$ are constants depending only on $n$, $q$

537: and $\alpha$

538: and are given by:

539: \begin{eqnarray}

540: & &c_1= \frac{-2q^2n^3(n + 1) + qn^3(2n^2 + 5n + 6) - n^4}{8

541: qn(n + 1) + 4(2-n)}\ \ , 

542: \ \ c_2= \frac{2}{n}(1+2\alpha^2 c_4)\ \ , \nonumber \\

543: & & c_3^2=\frac{ - 4 q^3 \alpha^2 (n + 1) 

544: + 2 q^2 \alpha^2 (2 n^2 + 7n + 8) + 4q

545: \alpha^2( - n^2 - 3n - 3) + 2\alpha^2 n}{2 q^2 n( - n^2 - 3n - 2) +

546: q (2 n^4 + 5 n^3 + 6 n^2 + 2n - 4) - n^3} \ \ , \nonumber \\

547: & & c_4  = \frac{1-q}{n c^2_3}  \ , 

548: \end{eqnarray}

549: and $q$ satisfies the equation

550: \begin{equation}

551: q^3(2n^4 + 4n^3 + 2n^2) + q^2(- 2  n^5 - 3 n^4 

552: + n^3 + 10 n^2 + 8  n) + q( 3 n^4 + 2  n^3 +

553:  3  n^2 + 6 n + 8) - n^3 + n^2 - 2n = 0 \ .

554: \end{equation}

555: Solving this condition for $q$ numerically for various

556: values of $n$, we find one real positive solution for $n=1$, $2$, $3$ and

557: three positive solutions for $n=4$, $5$, $6$, $7$. However, when

558: more than one solution for $q$ are available, it turns out that only one of

559: the solutions is acceptable since for the others some of the remaining

560: parameters (e.g. $c_4$ or $c_3$) become negative.

561: In Table 1, we give the acceptable solutions and their numerical values

562: for the parameters $c_1$, $c_2$, $c_3$, $c_4$:

563: \begin{center}

564: {\bf Table 1}: Values of $q$ and constants 

565: $c_1$, $c_2$, $c_3$, $c_4$ for different $n$

566: \\

567: \begin{tabular}{|l|l|l|l|l|l|}

568: \hline

569: $n$ & $K(x)=\sqrt{q}$ & $c_1$ & $c_2$  & $c_3^2/\alpha^2$

570: &$\alpha^2 c_4$   \\

571: \hline 

572: $1$    & $0.293$  & $0.016$  & $5.246$ & $1.126$ & $0.811$\\

573: $2$    & $0.294$  & $0.064$  & $1.702$  & $1.301$ & $0.351$\\

574: $3$    & $0.280$  & $0.102$  & $0.982$  & $1.297$ & $0.237$\\

575: $4$    & no real solution & & & &\\

576: $5$    & $0.249$  & $0.149$  & $0.495$  & $1.581$ & $0.118$\\

577: $6$    & $0.236$  & $0.151$  & $0.554$  & $0.476$ & $0.331$\\

578: $7$    & $0.225$  & $0.086$  & $2.402$  & $0.036$ & $3.704$ \\

579: \hline

580: \end{tabular}

581: \end{center}

582: 

583: 

584: 

585: 

586: 

587: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

588: \subsection{Numerical results for $n=2$}

589: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

590: The $4$-dimensional

591: effective theory for $n=1$ was studied in \cite{bh1}.

592: 

593: 

594: Here, we will put the emphasis on the case $n=2$ and again

595: discuss the solutions in the limit $\alpha^2=3\gamma^2$.

596: In this model, two Higgs 

597: fields naturally occur. Theories involving two Higgs fields are also

598: interesting from the viewpoint of the supersymmetric extension of

599: the Standard model.

600: 

601: The case $n=2$ involves two parameters

602: only, namely $\alpha$ and $\rho\equiv c_2/c_1$.  

603: As discussed previously in this paper, it is sufficient to consider

604: $\rho \geq 1$. 

605: 

606: The solutions can then be characterized by their mass~:

607: \be

608:    M = \frac{1}{\sqrt{c_1^2 + c_2^2}}\frac{\mu(\infty)}{\alpha^2}  \ ,

609: \ee

610: where the first factor is extracted in such a way that the mass of the

611: solution in the $\alpha=0$ limit is normalized to $1$, corresponding to the BPS 

612: monopole.

613: 

614: The values $\psi_{1,2}(0)$ will appear to be useful 

615: to further characterize the solutions. 

616: The deviation of the solutions from the flat space solutions can be

617: ``measured'' by the values of $A(x)$ at the origin, $A(0)$,

618:  and by the minimum of $N(x)$, say $N_m$. Since $A(x)$ always increases

619: monotonically, $A(0)$ also represents the minimum of the function.

620: 

621: In the following, we will discuss our numerical results \cite{foot1} for

622: two different and fixed values of $\rho$ and varying $\alpha$.

623: 

624: We again discuss the solutions in the limit of $\alpha^2=3\gamma^2$.

625: 

626: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

627: \subsubsection{Boundary conditions}

628: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

629: 

630: We will study globally regular, asymptotically flat 

631: solutions of the system above. This implies the following boundary conditions:

632: \begin{equation}

633: K(0)=1 \ , \ \ H_j(0)=0 \ , \ \ \partial_{x}\psi_j|_{x=0}=0 \ ,

634: \ \ \mu(0)=0

635: \  \label{bc1}

636: \end{equation}

637: at the origin and 

638:  \begin{equation}

639: K(\infty)=0 \ , \ \ H_j(\infty)=c_j \ ,

640: \ \ \psi_j(\infty)=0 \ , \ \ A(\infty)=1

641: \ , \label{bc2} \end{equation}

642: at infinity. We assume $c_j\neq 0$ in order for the solutions to have

643: a magnetic charge.

644: Using the symmetries discussed above, we see that we can

645: set ${\rm min} \{c_j, j=1,...,n\} = c_1 = 1$ without loosing generality

646:  and study

647: the equations with varying $\alpha$ and $c_2, c_3, ..., c_n$ $ > 1$.

648: Of course, in the case $c_1=c_2=...=c_n = 1$ the equations are symmetric

649: and we expect $H_1(x) = H_2(x)=...H_n(x)\equiv H(x)$ , 

650: $\psi_1(x) = \psi_2(x)=...=\psi_n(x)\equiv \psi(x)$.

651: 

652: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

653: \subsubsection{$\rho=1$}

654: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

655: 

656: In this case, the two Higgs functions $H_1(x)$ and $H_2(x)$

657: are equal. This, of course, implies immediately that also 

658: $\psi_1(x)=\psi_2(x)$.

659: 

660: In the flat limit ($\alpha =0$) the solution is the  BPS monopole \cite{ps,bogo}

661: with $A(x)=N(x)=1$, $\psi_1(x) = \psi_2(x) = 0$ and the 

662: matter functions $K(x)$, $H_1(x)= H_2(x)$ have the well 

663: known BPS profiles.

664: 

665: Increasing $\alpha$, our numerical results reveal

666: that the  solution gets progressively deformed by gravity.

667: The solutions form a branch on which the

668: mass $M$ dimnishes with increasing $\alpha$.

669: The same holds true for the values $A(0)$, $N_m$.

670: 

671: The function $N(x)$ indeed develops a minimum which becomes deeper

672: while gravity increases. At the same time the dilaton functions

673: $\psi_{1,2}(x)$ are non trivial for $\alpha\neq 0$. The value of $\psi_{1,2}(x=0)$

674: is negative and decreases with increasing $\alpha$.

675: 

676: Our results are illustrated in Fig.~\ref{fig1}.

677: The figure also demonstrates that the branch of gravitating solutions

678: does not exist for arbitrarily large values of the parameter $\alpha$.

679: Indeed, we find that solutions cease to exist for $\alpha > \alpha^{(1)}_{max}$

680: and $\alpha^{(1)}_{max}$ thus constitutes a maximal value of $\alpha$. We find that

681: this critical value depends on the parameter $\rho$ and is given here by:

682:  \be

683:        \alpha^{(1)}_{max} \approx 0.75   {\rm \ \ for \ \ } \rho = 1 \ \ \ . \ \ \

684:  \ee

685: 

686: Our numerical analysis further 

687: reveals that for $\alpha \leq \alpha^{(1)}_{max}$  another branch

688: of solution exists. 

689: The mass of the solutions on this second branch is higher

690: than the mass of the corresponding solutions on

691: the first branch. This is demonstrated in Fig.~\ref{fig2}.

692: 

693: Our numerical results further suggest as illustrated by the figures

694: that the solutions on the 

695: second branch stop to exist at some local minimal value 

696: $\alpha = \alpha^{(1)}_{min}$.

697: 

698: To be more precise, the further evolution of the 

699: branches is involved~: indeed several small branches 

700: exist for $\alpha$ $\epsilon$ $[\alpha_{min}^{(k)}, \alpha_{max}^{(k+1)}]$

701: with a smaller and smaller extend in $\alpha$.

702: Nevertheless, it appears clearly from the figures 

703: that for $\alpha \rightarrow \alpha_{cr}$, where $\alpha_{cr} : = 

704: \alpha^{(k)}_{min}=\alpha^{(k+1)}_{max}$ for some $k$,

705: the value $A(0)$ tends to zero, while the values

706: $\vert \psi_{1,2}(0) \vert$

707: increase considerably and likely become infinite.

708: Note, however, that the additional branches are not visible in the mass plot

709: (see Fig.\ref{fig2}), because the numerical values are very close to

710: those of the second branch.

711: The existence of several branches was also noticed for the $n=1$ system in \cite{volkov}

712: and in another Einstein-Yang-Mills model in 5 dimensions \cite{bcht}.

713:  

714:  

715: We find numericallly:

716: \begin{equation}

717: \alpha^{(1)}_{min}\approx 0.240 \ \ , \ \ \alpha^{(2)}_{max}\approx 0.294 

718: \ \ , \ \ \alpha_{cr}\approx 

719: 0.285  

720: \end{equation}

721: 

722: At the same time, the value $N_m$ stays strictly positive and tends

723: to the value given by (\ref{ads}) in the critical limit.

724: We find numerically that $N_m\approx 0.065$ for $\alpha\approx \alpha_{cr}$, 

725: while $c_1\approx 0.064$ for $n=2$. 

726: The function $N(x)$ 

727: reaches this minimum at a value of $x$, which tends to zero for 

728: $\alpha \rightarrow \alpha_{cr}$. At the same time, the function $N(x)$

729: becomes nearly flat on a plateau surrounding this minimum. 

730: 

731: The function $K(x)$ starts to develop oscillations around the

732: value $\sqrt{q}=0.294$. Processing on the branches, the number of oscillations

733: increases and becomes infinite in the critical limit. 

734: 

735: Very similar to what was observed in \cite{weinberg, volkov}, the

736: Yang-Mills domain walls get disconnected from the outside world

737: by an infinitely long throat in the strong gravity limit. 

738: 

739: 

740: 

741: %%%%%%%%%%%%%%%%%%%%%%%%

742: \subsubsection{$\rho=2$}

743: %%%%%%%%%%%%%%%%%%%%%%%%

744: 

745: In order to understand the influence of two Higgs fields with different expectation

746: values, i.e. $\rho\neq 1$, on the domain of existence 

747: of the solutions, we studied in detail the case $\rho=2$.

748: 

749: Varying $\alpha$, the existence of several branches of solutions is 

750: qualitatively very similar to the case $\rho=1$. The numerical analysis, however, 

751: reveals that for $\rho > 1$, the solutions exist on a smaller interval

752: of the coupling constant $\alpha$ in comparison to the case $\rho=1$.

753: Namely, we find:

754: \begin{equation}

755: \alpha^{(1)}_{max}\approx 0.496 \ , \ \alpha^{(1)}_{min}\approx 0.150 \ , 

756: \  \alpha^{(2)}_{max}=0.185 \ , \

757:  \alpha_{cr}\approx 0.179 \ \ \

758: {\rm for} \ \ \rho=2 \ . 

759: \end{equation}

760: In the critical limit for $\alpha\approx \alpha_{cr}$, we find

761: \begin{equation}

762: M\approx 2.05 \ \ , \ \ N_m\approx 0.068

763: \end{equation}

764: One of the apparent differences is that the two dilaton fields are non-equal. The values 

765: of $\psi_1(0)$ and $\psi_2(0)$ are superposed on Fig.~\ref{fig3}.

766: The difference $\psi_2(0)-\psi_1(0)$ is also shown and indicates

767: that in the critical limit, the two dilaton profiles remain significantly

768: different close to the origin. 

769: 

770: The profiles of the function $K(x)$, $N(x)$, $\psi_1(x)$ and $\psi_2(x)$

771: are shown in Fig.~\ref{fig4} for $\alpha=0.179$. This solution is already close

772: to the critical solution. Note that the function $K(x)$ starts to develop

773: oscillations completely analogue to the case $\rho=1$.

774: 

775: 

776: The numerical results suggest that in the critical limit a 

777: solution of the analytic, non-abelian type described in the previous

778: subsection is reached. 

779: The question how this solution looks like for $\rho\neq 1$ will be addressed in

780: a future publication.

781: 

782: 

783: 

784: \section{Conclusions}

785: Both string theories \cite{pol} as well as

786: so-called ``brane worlds'' \cite{brane}, which assume the Standard

787: model fields to be confined on a 

788: 3-brane (that is embedded in a higher dimensional space-time)

789: have enhanced the idea  that space-time

790: possesses more than four dimensions. In the former, the extra dimensions

791: are compactified

792: on a scale of the Planck length, while in the latter they are non-compact.

793: Non-perturbative, classical solutions of 

794: field theory models certainly play a major role in these theories.

795: 

796: 

797: It is therefore natural to investigate the classical solutions of

798: higher dimensional  Einstein-Yang-Mills theory.

799: In this paper, we have investigated Einstein-Yang-Mills

800: theories in 1+3+n dimensions, in which the matter and

801: metric fields are chosen to be independent of the extra $n$ coordinates.

802: Dimensional reduction then leads to $n$ dilaton fields

803: coupled individually to $n$ Higgs fields. These $n$ Higgs fields

804: can have independent vacuum expectation values.

805: 

806: The corresponding spherically symmetric equations admit several types

807: of solutions: (a) vacuum solutions, (b) embedded abelian Einstein-Maxwell-dilaton

808: solutions, (c) non-abelian solutions with diverging Higgs fields and

809: (d) fully non-abelian magnetically charged solutions.

810: The vacuum solutions (a) are trivial, while (b) and (c) were

811: constructed analytically for the case when all Higgs fields have

812: identical expectation values. The solutions of  type (d) 

813: had to be constructed 

814: numerically. 

815: 

816: We believe that type (b) and (c) solutions can also

817: be constructed analytically for generic values of the Higgs's

818: expectation values and we plan to reconsider this problem in a 

819: future publication.

820: 

821: It would also be interesting to 

822: study this model

823: for larger gauge groups or to adopt a different

824: compactification scheme of the codimension space, which here was chosen

825: to be $S^1 \times S^1 \times \dots S^1$. Following the investigation

826: for $n=1$ \cite{bbh1}, the model including a

827: cosmological constant is presently under investigation. 

828: In \cite{bbh1} the introduction of a cosmological

829: constant in the $(4+1)$-dimensional model led to a Liouville type potential

830: in the effective $4$-dimensional theory.

831: 

832: 

833: 

834: {\bf Acknowledgements}

835: Y. B. gratefully acknowledges the Belgian F.N.R.S. for financial support.

836: We thank E. Radu and D. H. Tchrakian for useful discussions.

837: 

838: \begin{thebibliography}{99}

839: \bibitem{kaluza}

840: T. Kaluza, Sitzungsber. Preuss. Akad. Wiss. Berlin (1921).

841: \bibitem{klein}

842: O. Klein, Z. Phys. {\bf 37} (1926), 895.

843: \bibitem{maeda}

844: G. Gibbons and K. Maeda, Nucl. Phys. {\bf B298} (1988), 741;

845: D. Garfinkle, G. Horowitz and A. Strominger, Phys. Rev. {\bf D43} (1991), 371.

846: \bibitem{volkov}

847: M. S. Volkov, Phys. Lett. {\bf B524} (2002), 369.

848: \bibitem{bh1} Y. Brihaye and B. Hartmann, Phys. Lett. {\bf B 534} (2002), 137.

849: \bibitem{weinberg}

850: K. Lee, V.P. Nair and E.J. Weinberg,

851:  Phys. Rev. {\bf D45} (1992), 2751;\\

852:  P. Breitenlohner, P. Forgacs and D. Maison,

853:  Nucl. Phys. {\bf B383} (1992), 357;\\

854:  P. Breitenlohner, P. Forgacs and D. Maison,

855:  Nucl. Phys. {\bf B442} (1995), 126.

856: \bibitem{thooft}

857:  G. `t Hooft, 

858: %Magnetic monopoles in unified gauge theories, 

859:  Nucl.~Phys.~ {\bf B79} (1974), 276;\\

860:  A.~M. Polyakov, 

861: %Particle spectrum in quantum field theory,

862:  JETP Lett. {\bf 20} (1974), 194.

863: 

864: 

865: \bibitem{foot1}

866: To integrate the equations, we used the differential equation

867: solver COLSYS which involves a Newton-Raphson method \cite{colsys}.

868: \bibitem{colsys} 

869: U. Ascher, J. Christiansen and R. D. Russell, Math. Comput. {\bf 33} 

870: (1979), 659; ACM Trans. Math. Softw. {\bf 7} (1981), 209.

871: \bibitem{ps} M. K. Prasad and C. M. Sommerfield, Phys. Rev. Lett. {\bf 35} (1975), 760.

872: \bibitem{bogo}

873: B. Bogomol'nyi, Sov. J. Nucl. Phys. {\bf 24} (1976), 449.

874: \bibitem{bcht}

875: Y. Brihaye, A. Chakrabarti, B. Hartmann and D. H. Tchrakian,

876: Phys. Lett. {\bf B 561} (2003), 161

877: \bibitem{pol} see e.g. J. Polchinski, {\it String Theory}, Cambridge University press (1998).

878: \bibitem{brane} V. A. Rubakov and M. E. Shaposhnikov, Phys. Lett. {\bf 125 B} (1983), 136; {\bf 125 B} (1983), 139;

879: G. Davli and M. Shifman, Phys. Lett. {\bf B 396} (1997), 64; {\bf 407} (1997), 452;

880: I. Antoniadis, Phys. Lett. {\bf B 246} (1990), 377; 

881: N. Arkani-Hamed, S. Dimopoulos and G. Dvali, Phys. Lett. {\bf B 429} (1998), 263;

882: I. Antoniadis, N. Arkani-Hamed, S. Dimopoulos and G. Dvali, Phys. Lett. 

883: {\bf B 436} (1998), 257; L. Randall and R. Sundrum, Phys. Rev. Lett. {\bf 83} (1999), 3370; {\bf 83} (1999), 4690. 

884: \bibitem{bbh1} B. Hartmann, Y. Brihaye and B. Bertrand, Phys. Lett. {\bf B 570} (2003), 137.

885: \end{thebibliography}

886: 

887: 

888: 

889:  \newpage

890: \begin{figure}

891: \centering

892: \epsfysize=20cm

893: \mbox{\epsffile{bch1.eps}}

894: \caption{\label{fig1} The values $A(0)$, $N_m$, $-\psi(0)$ are shown

895: as functions of $\alpha$ for $\rho=1$. The indices ``1'',

896: ``2'', ``3'', ``4'', respectively correspond to the 1., 2., 3. and 4. 

897: branch of solutions (see also Fig.~\ref{fig2}). }

898: \end{figure}

899: 

900:  \newpage

901: \begin{figure}

902: \centering

903: \epsfysize=20cm

904: \mbox{\epsffile{bch2.eps}}

905: \caption{\label{fig2} The mass $M$ of the solutions is shown

906: as function of $\alpha$ for $\rho=1$ and $\rho=2$. ``1'', ``2'' and ``3'', respectively

907: denote the 1., 2. and 3. branch of solutions. }

908: \end{figure}

909: 

910:  \newpage

911: \begin{figure}

912: \centering

913: \epsfysize=20cm

914: \mbox{\epsffile{bch3.eps}}

915: \caption{\label{fig3} The values of $-\psi_1(0)$ and $-\psi_2(0)$ as

916:  well as the difference $\psi_2(0)-\psi_1(0)$ are shown

917: as functions of $\alpha$ for $\rho=2$. ``1'', ``2'', ``3'' and ``4'', respectively

918: denote the 1., 2., 3. and 4. branch of solutions.}

919: \end{figure}

920: 

921:  \newpage

922: \begin{figure}

923: \centering

924: \epsfysize=20cm

925: \mbox{\epsffile{bch4.eps}}

926: \caption{\label{fig4} 

927: The profiles of the functions  $\psi_1(x)$, $\psi_2(x)$, $K(x)$, $A(x)$ and $N(x)$  are shown

928: for $\rho=2$ and $\alpha$ close to the critical value $\alpha_{cr}\approx 0.179$.}

929: \end{figure}

930: \end{document}

931: 

932: 

933: