1: \documentstyle[12pt]{article}

2: \input epsf

3: \newcommand{\sect}[1]{\section{#1}\setcounter{equation}{0}}

4: \def\theequation{\thesection.\arabic{equation}}

5: \setlength{\textheight}{7.5in}

6: \setlength{\topmargin}{0in}

7: \arraycolsep 2pt

8: %\usepackage{amsfonts}

9: \begin{document}

10: \date{\today}

11: %--------------------------------------------------------------------

12: \def\be{\begin{equation}}

13: \def\ee{\end{equation}}

14: \def\bea{\begin{eqnarray}}

15: \def\eea{\end{eqnarray}}

16: \def\a{\alpha}

17: \def\b{\beta}

18: \def\N{{\mathcal{N}}}

19: \def\t{{\triangle}}

20: \def\g{{\gamma}}

21: \def\l{{\lambda}}

22: \def\A{{\cal A}}

23: 

24: 

25: %==================================================================

26: 

27: 

28: \newpage

29: \bigskip

30: \hskip 3.7in\vbox{\baselineskip12pt

31: \hbox{NSF-ITP-01-72}}

32: 

33: \bigskip\bigskip

34: 

35: \centerline{\large \bf  On the thermodynamic instability of  LST}

36: 

37: 

38: 

39: \bigskip\bigskip

40: 

41: \centerline{{\bf

42: Alex Buchel\footnote{buchel@itp.ucsb.edu}}}

43: 

44: \bigskip

45: \centerline{Institute for Theoretical Physics}

46: \centerline{University of California}

47: \centerline{Santa Barbara, CA\ \ 93106-4030, U.S.A.}

48: 

49: 

50: 

51: \begin{abstract}

52: \baselineskip=16pt

53: The high energy thermodynamics of Little String Theory (LST) is known

54: to be unstable. An unresolved question is whether the corresponding

55: instability in LST holographic dual is of stringy or supergravity

56: origin. We study UV thermodynamics of a large metric deformation of

57: the LST dual realized (in the extremal case) by type IIB fivebranes

58: wrapping a two-sphere of a resolved conifold, and demonstrate that the

59: resulting black hole has negative specific heat.  This explicitly

60: shows that the high energy thermodynamic instability of the LST

61: holographic dual is of the supergravity origin.

62:   

63: 

64: 

65: 

66: 

67: 

68: 

69: \end{abstract}

70: \newpage

71: \setcounter{footnote}{0}

72: 

73: 

74: \baselineskip=17pt

75: %==================================================================

76: 

77: \section{Introduction}

78: Little string theory (LST) is  a  non-local  theory 

79: defined on the world volume of  NS5-branes in the limit of vanishing 

80: string coupling $g_{str}\to 0$ where the string scale 

81: $\a'$ is kept fixed \cite{brs,s9705} (for a review see \cite{a9911}).

82: This theory does not include dynamical gravity and in the IR 

83: flows to six-dimensional Yang-Mills theory. Nonetheless,

84: LST is quite different from a local field theory: 

85: it exhibits T-duality in toroidal compactifications and a 

86: Hagedorn density of states at very high energy --- both 

87: the intrinsic attributes

88: of the ``standard'' string theories. 

89: The latter property in particular implies that the 

90: statistical mechanics of LST breaks down at a finite 

91: temperature, known as the Hagedorn temperature.

92: The purpose of this paper is to better understand the

93: ultraviolet thermodynamics of the LST.

94: 

95: A classical thermodynamics of the LST on the world volume 

96: of flat NS5 branes can be easily deduced from its 

97: holographic dual, realized as a near horizon geometry 

98: of non-extremal NS5 branes \cite{ms9710}. It was shown in \cite{ms9710} that 

99: as  the number $N$ of the five-branes is large, $N\gg 1$, 

100: and the energy $\mu$ above 

101: extremality in string unites satisfies $\mu\gg N$, the 

102: background geometry is smooth with small curvatures everywhere; 

103: in addition, the string perturbation theory is also good 

104: as the dilaton is bounded from above by its value 

105: at the horizon, ${N\over \mu}$. Ignoring loop/stringy 

106: corrections, one finds that the Hawking temperature of the 

107: resulting black hole is independent of the energy $\mu$

108: and coincides precisely with the Hagedorn temperature $T_H$ of the 

109: LST 

110: \be

111: \b_H={1\over T_H}=2\pi\sqrt{N\a'}\,.

112: \label{th}

113: \ee

114: Since the can tune the energy and the temperature of the 

115: system independently, its equation of state (in the 

116: ultraviolet) is 

117: \be

118: S=\b_H E\,,

119: \label{seh}

120: \ee

121: which leads to an exponential growth of the density of 

122: states

123: \be

124: \rho(E)\sim e^{\b_H E}\,.

125: \label{ph}

126: \ee

127: One-loop corrections to the Hagedorn density of states of LST were 

128: studied in \cite{ho00,br00,ks00}. The finite energy corrections 

129: to the density of states (\ref{ph}) were  argued to be of the 

130: form 

131: \be

132: \rho(E)\sim E^{\a} e^{\b_H E}\biggl(1+O\left({1\over E}\right)\biggr)\,.

133: \label{pf}

134: \ee

135: The sign of $\a$ in (\ref{pf}) is of uttermost importance as it determines 

136: the stability of the thermal ensemble representing LST at 

137: high energy. The explicit calculation of \cite{ks00} indicated 

138: that $\alpha$ is negative implying the negative specific heat 

139: and thus the thermal instability of the system. 

140: According to authors of \cite{ks00} the instability in question is of stringy 

141: origin. It is represented by a (massless at tree level) 

142: string mode that winds once around the Euclidean time direction,

143: but is supposed to become tachyonic at one-loop level.     

144: 

145: An alternative explanation of the instability of the LST  

146: at high energy has been advocated in \cite{r0104}. 

147: Following the conjecture of \cite{gm09,gm11} that 

148: thermodynamic instabilities 

149: in field theories should correspond to classical instabilities of the 

150: dual spacetime geometry, Rangamani proposed \cite{r0104} that the 

151: supergravity dual to LST at Hagedorn 

152: temperature suffers from a Gregory-Laflamme (GL) \cite{gl9301,gl9404}  

153: like instability, thereby causing the thermal ensemble to be unstable. 

154: Though he did not manage to explicitly demonstrate that the metric 

155: fluctuations about the background of interest have indeed a zero 

156: frequency mode, that is capable of explaining the origin of the 

157: instability, Rangamani conjectured that the required mode 

158: is the one responsible for the GL instability in the 

159: near-extremal NS5-branes claimed in \cite{re0104}, that would survive 

160: the decoupling limit for the finite temperature LST.

161:  

162: In this paper we prove the proposal of \cite{r0104} and present 

163: yet another explicit example advocating the general philosophy 

164: of \cite{gm09,gm11}. We do not perform the stability 

165: analysis near the background representing 

166: finite temperature LST realized on  flat NS5 branes 

167: as suggested in \cite{r0104}. 

168: Rather, we study large metric deformations and 

169: show that the resulting black hole geometries 

170: have negative specific heat at the {\it classical} level.

171: Recall that in the flat case the temperature of the LST is (classically)

172: independent of the energy. By changing the 

173: theory in the IR (say wrapping NS5 branes on a 2-cycle, so that the  

174: theory is four dimensional macroscopically), while preserving 

175: the UV characteristics (having a 2-cycle of a finite size),

176: we generically expect that temperature should become energy dependent. 

177: This must definitely be the case, if the deformation 

178: induces a  phase transition in the IR, so that 

179: the deformed geometry is simply singular (and thus does not make sense) 

180: for sufficiently small energy. 

181: Then, by studying the classical thermodynamics of 

182: the deformed background at  high energy 

183: we should be able to deduce $\a$ in (\ref{pf}).

184: Specifically, we propose to study the stability 

185: issue of LST at high energy from the thermodynamics 

186: of large number of type IIB NS5 branes wrapping a two-cycle 

187: of a resolved conifold. The relevant extremal solution

188: was discussed in \cite{chv1,chv2,mn0008}. It has $\N=1$  SUSY 

189: in four dimensions and exhibits 

190: a naked singularity in the IR associated with the 

191: chiral symmetry breaking of the dual gauge theory at zero 

192: temperature. The nonextremal deformation of this background 

193: was constructed in \cite{bf0103}. It was argued there that the 

194: naked singularity (at the extremality) will be hidden beyond the black 

195: hole horizon for sufficiently large energy away from the extremality.

196: 

197: The paper is organized as follows. In the next section 

198: we describe the gravitational background representing 

199: a large number of type IIB NS5 branes wrapping a 2-cycle in the

200: resolved conifold geometry at finite temperature. We study 

201:  thermodynamics of this black hole solution, and 

202: show that at high energy the black hole has a negative specific 

203: heat. We analytically compute $\a$ in the  

204: density of states expression (\ref{pf}) for the discussed 

205: deformation. We briefly conclude in section 3. 

206: Technical details can be found in the appendix.

207: 

208: 

209: \section{High energy thermodynamics of $\N=1$ deformed LST}

210: In the previous section addressing the origin 

211: of thermodynamic instability of LST at high energy, 

212: we proposed to study  large metric deformation of its holographic dual 

213: realized  by wrapping NS5 branes on a two-sphere of a resolved conifold. 

214: The motivation for choosing this particular background 

215: came from the fact that the corresponding extremal 

216: solution had a naked singularity in the IR associated with 

217: the zero temperature chiral symmetry breaking phase transition 

218: of the dual gauge theory. As the result, the Hawking 

219: temperature of its nonextremal deformation should be energy 

220: dependent, and so the classical analysis should be enough 

221: to extract $\a$ in the ``corrected'' entropy-energy relation 

222: \be

223: S=\b_H E+\a \log E +O\left(1/E\right)\,.

224: \label{corr}

225: \ee  

226: 

227: We proceed by recalling the  gravitational 

228: background holographically dual to the $\N=1$

229: deformed LST at finite temperature,

230: originally constructed in \cite{bf0103}. From now on 

231: we set the string scale $\a'=1$. 

232: The ten dimensional string frame metric, NS-NS 3-form $H$, 

233: and the dilaton $\phi$

234: are given by\footnote{$\t,\ f(\t)$ used here are related 

235: to $c_1(r),\ h(r),\ a$ of \cite{bf0103} 

236: as follows  $\t\equiv \t_1(r)$, 

237: $f(\t)=c_1^8(r) \t_1(r) h(r)/A$, $N=a^2$.} 

238: \bea

239: ds^2_{str}&=&\t^2 dt^2-d\bar{x}^2-N\biggl[ f^2 (d[\ln\t])^2+

240: {h\over 4}\left(d\theta_1+\sin^2\theta_1 d\phi_1^2\right)\cr

241: &&+{1\over 4}\left(d\theta_2+\sin^2\theta_2 d\phi_2^2\right)

242: +{1\over 4}\left(d\psi+\sum_{i=1}^2 \cos\theta_i d\phi_i\right)^2\ \biggr]\,,\cr

243: \cr

244: H&=&{N\over 4}\biggl[\left(d\psi+\sum_{i=1}^2 \cos\theta_i d\phi_i

245: \right)\wedge\left(\sin\theta_1 d\theta_1\wedge d\phi_1-\sin\theta_2 

246: d\theta_2\wedge d\phi_2\right)\biggr]\,,\cr

247: \cr

248: e^{-2\phi}&=&C {f\over h \t }\,,

249: \label{m10}

250: \eea

251: where $N$ is the number of NS5 branes, $C$  is an integration 

252: constant related to the asymptotic  

253: string coupling,  and $\t$ is a 

254: new  radial coordinate defined in such a way that 

255: $\t=0_+$ determines the black hole horizon, and $\t\to 1_-$ is the 

256: ultraviolet asymptotic. 

257: 

258: From eq.~(5.56) of \cite{bf0103}, $f$ and $h$ satisfy 

259: second-order Toda-like system of differential equations 

260: \bea

261: &&\left[\ln{f}\right]''=4 f^2\left(1+{2\over h}-{1\over h^2}\right)\,,

262: \label{f}

263: \eea

264: \bea

265: &&\left[\ln h\right]''=4 f^2\left({2\over h}-{2\over h^2}\right)\,,

266: \label{h}

267: \eea 

268: supplemented by the first-order constraint 

269: \bea

270: 0&=&\left[{f^2 \over h^2}\right]'\left[h^2\right]'+

271: {f^2 \over h^2}(h')^2+2 h^2 \left(\left[{f\over h}

272: \right]'\right)^2-2 f^2 h^2\cr

273: &&-8 {f^4 \over h^2}(h^2+2 h -1)\,.

274: \label{consi}

275: \eea

276: In eqs.~(\ref{f}),~(\ref{h}),~(\ref{consi}) 

277: the derivatives are with respect to 

278: \be

279: y\equiv \ln \t\,,\qquad y\in(-\infty,0)\,.

280: \label{y}

281: \ee

282: As explained in \cite{bf0103}, to have a regular Schwarzschild

283: horizon we must have 

284: \bea

285: &&f= e^y\left(f_0+\sum_{n=1}^{\infty}f_n e^{2y n}\right)\,,\cr

286: &&h= h_0+\sum_{n=1}^{\infty}h_n e^{2y n}\,,\qquad  {\rm as}\ y\to -\infty\,,

287: \label{hor}

288: \eea

289: for some positive constants\footnote{Only $h_0$ is an 

290: independent parameter,  $f_0$ should be fixed by the 

291: boundary conditions \cite{bf0103}.} $f_0,h_0$. In the UV region we 

292: have 

293: \bea

294: &&h\to +\infty\,,\qquad  {\rm as}\ y\to 0_{-}\,,\cr

295: &&f\to +\infty\,,\qquad 

296: {\rm as}\ h\to +\infty\,.

297: \label{ext}

298: \eea

299: Furthermore, it was argued in \cite{bf0103} that both 

300: the IR and the UV asymptotics are compatible provided  $h_0>1$,

301: where the inequality incorporates the physics of chiral symmetry 

302: restoration at finite temperature of the dual gauge theory.

303: 

304: 

305: We now study the thermodynamic properties of the black hole

306: (\ref{m10}) keeping the dilaton at the horizon $g_h$ fixed, for 

307: large $h_0\gg 1$. The first condition amounts to choosing 

308: \be

309: C={h_0\over g_h^2 f_0}\,,

310: \label{Cdef}

311: \ee

312: and as we will see insures (for $g_h\ll 1$) 

313: that the dilaton is everywhere small.

314: The second condition is the high energy limit of the thermodynamics.

315: We don't know how to solve (\ref{f}), (\ref{h}) in general. 

316: However, the $h_0\to \infty$ asymptotic of the solution can be 

317: extracted with some work. We find\footnote{See appendix for the 

318: details.}

319: \bea

320: h(y)&=&-2 y -2 \log[-\sinh y]+h_0+O(1)\,,\cr

321: \cr

322: f(y)&=&-{1\over 2\sinh y} \left(1-{1\over h(y)}+o\left({1\over h(y)}\right)

323: \right)\,.

324: \label{solution}

325: \eea

326: With (\ref{solution}) we see that the dilaton is indeed bounded 

327: by its value at the horizon, also 

328: \be

329: f(y) e^{-y}\bigg|_{y\to -\infty}\to f_0=1-1/h_0+o(1/h_0)\,. 

330: \label{fff}

331: \ee

332: From the  metric of (\ref{m10}) we can read off the Hawking temperature

333: by identifying the periodicity of its  Euclidean time direction 

334: with the inverse temperature  

335: \be

336: T^{-1}\equiv\b=2\pi N^{1/2} f_0=\b_H \left(1-{1\over h_0}+o\left({1\over 

337: h_0}\right)\right)\,.

338: \label{temp}

339: \ee 

340: Next, we compute the Bekenstein-Hawking entropy of the geometry 

341: (\ref{m10}). We find the 8-dimensional area of the event horizon $\A_8$

342: of the black hole to be

343: \be

344: \A_8=2 g_h^{-2} N^{5/2} \pi^3 h_0 V_3\,,

345: \label{area}

346: \ee

347: where $V_3$ is the 3-dimensional volume. The entropy of the 

348: black hole is then

349: \be

350: S={\A_8\over 4 G_N}={N^{5/2} h_0 V_3\over 16 \pi^3 g_h^4}\,.

351: \label{entropy}

352: \ee

353: From the ordinary statistical mechanics we know that the 

354: energy is $dE=T dS$, so from (\ref{temp}), (\ref{entropy})

355: we find the entropy-energy relation for our particular 

356: deformation of LST

357: \be

358: S(E)=\b_H E+\alpha\log E+o(\log E)\,,

359: \label{enen}

360: \ee 

361: with 

362: \be

363: \a=-{N^{5/2} V_3\over 16\pi^3 g_h^4}\,.

364: \label{compal}

365: \ee

366: Note that as in the flat LST case \cite{ks00}, $\alpha$ is 

367: and extensive quantity --- it is proportional to the noncompact 

368:  factor  $V_3$ of the fivebrane world-volume. 

369: Since $\a$ in (\ref{compal}) is negative, 

370: the black hole (\ref{m10}) has negative specific heat and 

371: is thus thermodynamically unstable. 

372: 

373: 

374: \section{Conclusion}    

375: The purpose of this paper was to argue that the instability 

376: of the ultraviolet LST thermodynamics in its holographic 

377: dual  is  of a  supergravity origin, as suggested in \cite{r0104}.

378: To do so, we identified a large metric deformation of the 

379: gravity dual to LST where the (conjectured) threshold instability 

380: of the flat LST dual should be enhanced to classical instability 

381: of the nonextremal generalization of the deformed background.

382: We argued that this should happen if the deformation in the 

383: (extremal case) induces a phase transition in the IR, 

384: but does not affect the UV physics. In our example,

385: this large metric deformation is realized by a large number 

386: of NS5 branes wrapping a 2-cycle of a resolved conifold.  

387: By studying the UV thermodynamics of this background 

388: we explicitly showed that resulting black hole has a negative 

389: specific heat.

390:  

391: The nonextremal geometry of interest has small curvatures 

392: and small perturbative string loop corrections if both 

393: $1/N$ and $g_h$ are small. Thus our classical analysis 

394: is reliable.  

395: 

396: 

397: 

398: 

399: 

400: 

401:    

402:  

403: 

404: 

405: 

406: 

407: 

408: \section*{Acknowledgments}

409: We wish to thank    Anton Kapustin, Igor Klebanov

410: and Arkady Tseytlin

411: for valuable discussions. This work  was supported in 

412: part by NSF grants PHY97-22022 and PHY99-07949. 

413: 

414: 

415: \section*{Appendix}

416: In what follows we solve (\ref{f}), (\ref{h}) subject to the boundary 

417: conditions (\ref{hor}), (\ref{ext}) in the limit $h_0\to \infty$.

418: Note that from (\ref{f}), (\ref{h}) both 

419: $f$ and $h$ are rapidly increasing functions, so in the 

420: zero order approximation in the limit $h_0\to \infty$, we can set

421: $1/h=0$ in (\ref{f}). This equation has then a general solution 

422: \be

423: f\approx f^{(0)}=-{C_1\over 2 \sinh(C_1 y+C_2)}\,,

424: \label{solf0}

425: \ee

426: where $C_i$ are integration constants.

427: From the UV asymptotic we need $f\to\infty$ as $y\to 0_-$. 

428: This fixes $C_2=0$. Furthermore, the asymptotic at the horizon 

429: fixes $C_1=1$ so that  

430: \be

431: f^{(0)}=-{1\over 2 \sinh y}\,.

432: \label{ff0}

433: \ee

434: Note that  satisfying the asymptotics 

435: uniquely fixes $f_0$,  

436: \be

437: f_0^{(0)}=1\,.

438: \label{f00}

439: \ee 

440: In the leading order eq.~(\ref{h}) reads

441: \be

442: [\ln h^{(0)}]''={2\over h^{(0)} \sinh^2 y}\,.

443: \label{hh0}

444: \ee

445: Above equation is difficult to solve analytically. 

446: We claim however that in the limit $h_0\to \infty$

447: \be

448: [h^{(0)}]'' h^{(0)}\gg ([h^{(0)}]')^2\,,

449: \label{ineq}

450: \ee 

451: uniformly for all $y$. In approximation (\ref{ineq}), eq.~(\ref{hh0})

452: simplifies 

453: \be

454: [h^{(0)}]''={2\over  \sinh^2 y}\,.

455: \label{hsimp}

456: \ee

457: The general solution of (\ref{hsimp}) is 

458: \be

459: h^{(0)}=C_1 y-2 \log[-\sinh y]+C_2\,,

460: \label{h01}

461: \ee

462: where $C_i$ are integration constants. To satisfy 

463: asymptotic at the horizon we must choose $C_1=-2$ and $C_2=h_0-2 \ln 2$

464: so that 

465: \be

466: h^{(0)}=-2 y-2 \log[-\sinh y]+h_0-2\ln 2\,.

467: \label{hof}

468: \ee

469: We can now go back to (\ref{ineq}) to check the self-consistency 

470: of the approximation. 

471: Indeed, we find that ${[h^{(0)}]'' h^{(0)}\over  ([h^{(0)}]')^2}$

472: with $h^{(0)}$ given by (\ref{hof}) has a global minimum 

473: at $y=\left(-1/h_0+O(h_0^{-2})\right)$, with the value at the minimum 

474: $\left(h_0/2-\log h_0+ O(1)\right)\to \infty$ as $h_0\to \infty$.

475: 

476: To study the leading correction to the Hagedorn 

477: thermodynamics we actually need to know the first correction 

478: to (\ref{ff0}). We take the following ansatz for the leading 

479: correction

480: \be

481: f=-{1\over 2 \sinh y} \left(1+{\gamma\over h^{(0)}}+o({1\over h^{(0)}})

482: \right)\,,

483: \label{1st}

484: \ee

485: where $h^{(0)}$ is the zero order solution (\ref{hor}).

486: Substituting (\ref{1st}) into (\ref{f}), using eq.~(\ref{hsimp})

487: and the approximation (\ref{ineq}), we  find $\gamma=-1$. 

488: In a similar fashion we can compute the leading correction 

489: to $h^{(0)}$ and see that it is a fixed constant.  

490: From (\ref{1st}) it follows that 

491: \be

492: f_0=1-{1\over h_0}+o\left({1\over h_0}\right)\,.

493: \label{1stf0}

494: \ee

495: 

496: Since our conclusion about UV thermodynamic instability of LST 

497: hinges on the fact that $\g<0$, we also did numerical 

498: analysis to confirm (\ref{1stf0}). In the remaining of this

499: section we  describe  them and present the results. 

500: First, we rewrite (\ref{f}) and (\ref{h}) in terms of 

501: \bea

502: &&f_1(t)\equiv {e^y\over f(y)}\,,\cr

503: &&f_2(t)\equiv {1\over h(y)}\,,

504: \label{red}

505: \eea

506: where we introduced new variable 

507: \be

508: t\equiv e^{2 y},\qquad t\in [0,1]\,.

509: \label{tdef}

510: \ee

511: We find 

512: \bea

513: 0&=&t f_1(t) [f_1(t)]''+f_1(t) [f_1(t)]'-t ([f_1(t)]')^2+1+2 f_2(t)

514: -f_2(t)^2\,,\cr

515: \cr

516: 0&=&t f_1(t)^2 f_2(t) [f_2(t)]''+f_1(t)^2 f_2(t) [f_2(t)]'-t f_1(t)^2 

517: ([f_2(t)]')^2\cr 

518: &&+2 f_2(t)^3-2 f_2(t)^4\,,

519: \label{eq}

520: \eea

521: where all the derivatives are with respect to $t$.

522: Near  $t=0$ we have power series expansion

523: \bea

524: f_1=\sum_{k=0}^{\infty} \a_k t^k\,,\qquad f_2=\sum_{k=0}^{\infty} 

525: \b_k t^k\,,

526: \label{t0}

527: \eea

528: with 

529: \bea

530: &&\a_0={1\over f_0},\qquad \a_1=-{f_0 (h_0^2+2 h_0-1)\over h_0^2},\qquad 

531: \a_2=\cdots\,,\cr

532: &&\b_0={1\over h_0},\qquad \b_1=-{2 f_0^2 (h_0-1)\over h_0^3},\qquad 

533: \b_2=\cdots\,,

534: \label{init}

535: \eea

536: while at $t\to 1_-$ we have boundary condition

537: \be

538: 0<f_1(t)\ll f_2(t)\to 0\,.

539: \label{bound}

540: \ee

541: 

542: 

543: \begin{figure}

544: \centering

545: \epsfxsize=2.5in

546: \hspace*{0in}\vspace*{.2in}   

547: \epsffile{gr.ps}   

548: \caption{Numerical determination of $f_0(h_0)$.

549: Dots represent values of $\log[1-f_0(h_0)]$  evaluated  

550: as detailed in the appendix section.  

551: The slope is predicted to be $(-1)$ from (\ref{1stf0}).}

552: \end{figure}

553: 

554: 

555: 

556: 

557: In practice we integrated (\ref{eq}) from $t=\delta t=10^{-4}$

558: with initial conditions, determined by ${f_0,h_0}$, specified 

559: by the first six terms in the expansion (\ref{t0}). 

560: Notice that is if a set  $\{f_1(t,[f_0,h_0]), f_2(t,[f_0,h_0])\}$

561: is a solution to (\ref{eq}), then so does 

562: $\{f_1(t\l,[f_0/\sqrt{\l},h_0]), f_2(t\l,[f_0/\sqrt{\l},h_0])\}$ 

563: for arbitrary $\l$. The latter in particular implies 

564: that if $f_1(t=1,[f_0,h_0])=0$ (as it should be for the 

565: boundary condition (\ref{bound})), than 

566: $f_1(t=\l,[f_0/\sqrt{\l},h_0])=0$. This suggested a 

567: practical ``trick'' of fixing $f_0$ in terms of $h_0$ from 

568: the integration: for a fixed $h_0$ we take $f_0=1$ and find 

569: (via numerical integration) $t^*\equiv t^*(h_0)$ such that 

570: $f_1(t^*(h_0),[1,h_0])=0$; then, above considerations guarantee that 

571: $f_1(1,[\sqrt{t^*(h_0)},h_0])=0$. That is 

572: \be

573: f_0(h_0)=\sqrt{t^*(h_0)}\,.

574: \label{f0h0}

575: \ee

576: The results of the numerical computations are presented in Fig.~1,

577: as a plot of $\log [1-\sqrt{t^*(h_0)}]$ versus $\log(h_0)$. 

578: According to (\ref{1stf0}) we expect the large $\log(h_0)$ 

579: slope to be  $(-1)$. We see that this is indeed the case.

580: 

581: 

582: 

583: 

584: \newpage

585: \begin{thebibliography}{99}

586: 

587: \bibitem{brs} M.~Berkooz, M.~Rozali and  N.~Seiberg,

588: ``Matrix Description of M-theory on $T^4$ and $T^5$,'' 

589: {\it Phys.Lett.} {\bf B408} (1997) 105, hep-th/9704089.

590: 

591: 

592: 

593: \bibitem{s9705} N.~Seiberg, "Matrix Description of M-theory on $T^5$  

594: and $T^5/Z_2$," {\it Phys.Lett.} {\bf B408} 

595: (1997) 98, hep-th/9705221.

596: 

597: 

598: \bibitem{a9911} O.~Aharony, "A brief review of little string 

599: theories," {\it Class.Quant.Grav.}

600: {\bf 17}, (2000) 929, hep-th/9911147. 

601: 

602: \bibitem{ms9710} J.~M.~Maldacena and A.~Strominger, 

603: "Semiclassical decay of near extremal fivebranes,"

604: {\it JHEP} {\bf 9712}, (1997) 008, hep-th/9710014.

605: 

606: \bibitem{ho00} T.~Harmark and N.~A.~Obers,

607: "Hagedorn Behaviour of Little String Theory from

608: String Corrections to NS5-Branes," 

609: {\it Phys.Lett.} {\bf B485} 

610: (2000) 285, hep-th/0005021.

611: 

612: \bibitem{br00} M.~Berkooz and M.~Rozali, 

613: "Near Hagedorn Dynamics of NS Fivebranes, or A New

614: Universality Class of Coiled Strings,"

615: {\it JHEP} {\bf 0005}, (2000) 040, hep-th/0005047.

616: 

617: \bibitem{ks00} D.~Kutasov and D.~A.~Sahakyan,

618: "Comments on the Thermodynamics of Little String

619: Theory," 

620: {\it JHEP} {\bf 0102}, (2001) 021, hep-th/0012258.

621: 

622: \bibitem{r0104} M.~Rangamani, 

623: "Little String Thermodynamics," 

624: {\it JHEP} {\bf 0106} (2001) 042,

625: hep-th/0104125. 

626: 

627: \bibitem{gm09} S.~S.~Gubser and  I.~Mitra, 

628: "Instability of charged black holes in anti-de Sitter space,"

629: hep-th/0009126.

630: 

631: \bibitem{gm11} S.~S.~Gubser and  I.~Mitra, 

632: "The evolution of unstable black holes in anti-de Sitter

633: space," hep-th/0011127.

634: 

635: \bibitem{gl9301} R.~Gregory and  R.~Laflamme, 

636: "Black Strings and p-Branes are Unstable,"

637: {\it 

638: Phys.Rev.Lett.} {\bf 70} (1993) 2837, hep-th/9301052.

639: 

640: 

641: \bibitem{gl9404} R.~Gregory and  R.~Laflamme,

642: "The Instability of Charged Black Strings and p-Branes,"

643: {\it Nucl.Phys.} {\bf  B428} (1994) 399, hep-th/9404071.

644: 

645: 

646: \bibitem{re0104} H.~S.~Reall, 

647: "Classical and Thermodynamic Stability of Black Branes," 

648: hep-th/0104071.

649: 

650: \bibitem{chv1} Ali H.~Chamseddine and  Mikhail S.~Volkov,

651: "Non-Abelian BPS Monopoles in N=4 Gauged Supergravity,"

652: {\it Phys.\ Rev.\ Lett.} {\bf 79} (1997) 3343, hep-th/9707176.

653: 

654: \bibitem{chv2} Ali H.~Chamseddine and  Mikhail S.~Volkov,

655: "Non-Abelian Solitons in N=4 Gauged Supergravity and Leading Order

656: String Theory,"

657: {\it Phys.\ Rev.\ } {\bf D57} (1998) 6242, hep-th/9711181.

658: 

659: 

660: \bibitem{mn0008} J.~Maldacena and C.~Nunez, "Towards the large 

661: N limit of pure N=1 super Yang Mills," {\it Phys.\ Rev.\ Lett.\ } {\bf 86} 

662: (2001) 588, hep-th/0008001.

663: 

664: 

665: \bibitem{bf0103} A.~Buchel and  A.~Frey, "Comments on 

666: supergravity dual of pure N=1 Super

667: Yang Mills theory with unbroken chiral symmetry," hep-th/0103022.

668: 

669: 

670: \end{thebibliography}

671: 

672: 

673: 

674: 

675: \end{document}

676: 

677: 

678: 

679: 

680: 

681: 

682: 

683: 

684: 

685: 

686: 

687: 

688: 

689: 

690: 

691: 

692: 

693: 

694: 

695: 

696: 

697: 

698: 

699: 

700: