1: \documentstyle[12pt]{article}

2: \input epsf.sty

3: \topmargin -.5cm

4: \textheight 21cm

5: \oddsidemargin -.125cm

6: \textwidth 16cm

7: 

8: \def\ZZZ{{\hbox{ Z\kern-1.6mm Z}}}

9: \newcommand{\bet}{\sigma}

10: \newcommand{\bbb}{{\bar b}}

11: \newcommand{\gam}{\tau}

12: \newcommand{\eps}{\epsilon}

13: \newcommand{\ra}{\rangle}

14: \newcommand{\la}{\langle}

15: \newcommand{\T}{\chi_{T}(k)}

16: \newcommand{\Tm}{\chi_{T}(k')}

17: \newcommand{\Cn}{{\cal C}_n}

18: \newcommand{\vp}{\varphi}

19: \newcommand{\ve}{\varepsilon}

20: \newcommand{\tl}{\wt\lambda}

21: \newcommand{\dt}{(\vec \nabla T)^2}

22: \newcommand{\hp}{{\wh\Phi}}

23: \newcommand{\hq}{{\wh Q_B}}

24: \newcommand{\he}{{\wh\eta_0}}

25: \newcommand{\ha}{{\wh{A}}}

26: \newcommand{\lllb}{\Bigl\langle\Bigl\langle}

27: \newcommand{\rrrb}{\Bigr\rangle\Bigr\rangle}

28: \newcommand{\tf}{\wt f}

29: \newcommand{\vv} {\bar v}

30: \newcommand{\uu} {\bar u}

31: \newcommand{\K}{{\rm K_1}}

32: \newcommand{\Kt}{{\rm \widetilde K_1}}

33: 

34: \newcommand{\B}{b'}

35: \newcommand{\C}{c'}

36: \newcommand{\bB}{\bar b'}

37: \newcommand{\bC}{\bar c'}

38: \newcommand{\Bu}{B_{\vec u}}

39: \newcommand{\VV}{{\cal V}}

40: \newcommand{\BB}{{\cal B}}

41: \newcommand{\II}{{\cal I}}

42: \newcommand{\GG}{{\cal G}}

43: \newcommand{\AAA}{{\cal A}}

44: \newcommand{\FF}{{\cal F}}

45: \newcommand{\HH}{{\cal H}}

46: \newcommand{\MM}{{\cal M}}

47: \newcommand{\CC}{{\cal C}}

48: \newcommand{\OO}{{\cal O}}

49: \newcommand{\QQ}{{\cal Q}}

50: \newcommand{\PP}{{\cal P}}

51: \newcommand{\EE}{{\cal E}}

52: \newcommand{\LL}{{\cal L}}

53: \newcommand{\lll}{\langle\langle}

54: \newcommand{\rrr}{\rangle\rangle}

55: \newcommand{\square}{\Box}

56: \newcommand{\half}{{1\over 2}}

57: \newcommand{\wt}{\widetilde}

58: \newcommand{\wh}{\widehat}

59: \newcommand{\wc}{\check}

60: \newcommand{\wb}{\bar}

61: \newcommand{\bd}{\bar{\rm D}}

62: \newcommand{\RR}{{\cal R}}

63: \newcommand{\NN}{{\cal N}}

64: \newcommand{\TT}{{\cal T}}

65: \newcommand{\bg}{\bar g}

66: \newcommand{\bb}{\bar b}

67: \newcommand{\bG}{\bar G}

68: \newcommand{\bT}{\bar \Theta}

69: \newcommand{\SSS}{{\cal S}}

70: \newcommand{\tlx}{\left(\tilde \lambda ; X^0(0) \right)}

71: \newcommand{\al}{\alpha}

72: \newcommand{\omk}{\omega^{(n)}_{\vec k}}

73: \newcommand{\onk}{\omega^{(N)}_{\vec k_\perp}}

74: 

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

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

77: \newcommand{\ben}{\begin{eqnarray}\displaystyle}

78: \newcommand{\een}{\end{eqnarray}}

79: \newcommand{\refb}[1]{(\ref{#1})}

80: \newcommand{\p}{\partial}

81: \newcommand{\sectiono}[1]{\section{#1}\setcounter{equation}{0}}

82: \renewcommand{\theequation}{\thesection.\arabic{equation}}

83: %\renewcommand{\theequation}{\arabic{equation}}

84: \def\one{{\hbox{ 1\kern-.8mm l}}}

85: \def\zero{{\hbox{ 0\kern-1.5mm 0}}}

86: 

87: \begin{document}

88: {}~

89: {}~

90: \hfill\vbox{\hbox{hep-th/0410103}%\hbox{MRI-P-020401}

91: }\break

92: 

93: \vskip .6cm

94: 

95: 

96: \centerline{{\Large \bf Tachyon Dynamics in Open String

97: Theory}\footnote{Based on lectures given at the 2003 and 2004 ICTP Spring

98: School, TASI 2003, 2003 Summer School on Strings, Gravity and 

99: Cosmology at Vancouver, 2003 IPM String

100: School at Anzali, Iran, 2003 ICTP Latin American School at Sao

101: Paolo, 2004 Nordic meeting at Groningen and 2004 Onassis Foundation 

102: lecture at Crete.}}

103: 

104: 

105: \medskip

106: 

107: \vspace*{4.0ex}

108: 

109: \centerline{\large \rm

110: Ashoke Sen}

111: 

112: \vspace*{4.0ex}

113: 

114: 

115: \centerline{\large \it Harish-Chandra Research

116: Institute}

117: 

118: \centerline{\large \it  Chhatnag Road, Jhusi,

119: Allahabad 211019, INDIA}

120: 

121: \centerline{E-mail: ashoke.sen@cern.ch,

122: sen@mri.ernet.in}

123: 

124: \vspace*{5.0ex}

125: 

126: \centerline{\bf Abstract} \bigskip

127: 

128: In this review we describe our current understanding of the properties of

129: open string tachyons on an unstable D-brane or brane-antibrane system in

130: string theory. The various string theoretic methods used for this study

131: include techniques of two dimensional conformal field theory, open string

132: field theory, boundary string field theory, non-commutative solitons etc.

133: We also describe various attempts to understand these results using field

134: theoretic methods. These field theory models include toy models like

135: singular potential models and $p$-adic string theory, as well as more

136: realistic version of the tachyon effective action based on

137: Dirac-Born-Infeld type action. Finally we study closed string background

138: produced by the `decaying' unstable D-branes, both in the critical string

139: theory and in the two dimensional string theory, and describe the open

140: string completeness conjecture that emerges out of this study. According

141: to this conjecture the quantum dynamics of an unstable D-brane system is

142: described by an internally consistent quantum open string field theory

143: without any need to couple the system to closed strings. Each such system

144: can be regarded as a part of the `hologram' describing the full string

145: theory.

146: 

147: \vfill \eject

148: 

149: \baselineskip=16pt

150: 

151: \tableofcontents

152: 

153: 

154: \sectiono{Introduction} \label{s1}

155: 

156: This introductory section is divided into two parts. In section

157: \ref{smot} we give a brief motivation for studying the tachyon dynamics in

158: string theory. Section \ref{sorg} summarizes the organisation of the

159: paper.

160: 

161: \subsection{Motivation} \label{smot}

162: 

163: Historically, a tachyon was defined as a particle that travels faster than

164: light. Using the relativistic relation $v=p/\sqrt{p^2 + m^2}$ between

165: the velocity $v$, the spatial

166: momentum

167: $p$ and mass $m$ of a particle we see that

168: for real $p$ a

169: tachyon must have negative mass$^2$. Clearly neither of these

170: descriptions

171: makes a convincing case for the tachyon.

172: 

173: Quantum field theories offer a much better insight into the role of

174: tachyons. For this consider a scalar field $\phi$ with conventional

175: kinetic term, and a potential

176: $V(\phi)$ which has an extremum at the origin. If we carry out

177: perturbative quantization of the scalar field by expanding the potential

178: around $\phi=0$, and ignore the cubic and higher order terms in the

179: action, we find a particle like state with mass$^2=V''(0)$. For $V''(0)$

180: positive this describes a particle with positive mass$^2$. But for

181: $V''(0)<0$ we have a particle with negative mass$^2$, {\it i.e.} a

182: tachyon!

183: 

184: In this case however the existence of the tachyon has a clear

185: physical interpretation. For $V''(0)<0$, the potential $V(\phi)$

186: has a maximum at the origin, and hence a small displacement of

187: $\phi$ away from the origin will make it grow exponentially in

188: time. Thus perturbation theory, in which we treat the cubic and

189: higher order terms in the potential to be small, breaks down. From

190: this point of view we see that the existence of a tachyon in a

191: quantum field theory is associated with an instability of the

192: system which causes a breakdown of the perturbation theory. This

193: interpretaion also suggests a natural remedy of the problem. We

194: simply need to expand the potential around a new point in the

195: field space where it has a minimum, and carry out perturbative

196: quantization of the theory around this point. This in turn will

197: give a particle with positive mass$^2$ in the spectrum.

198: 

199: Unlike quantum field theories which provide a second

200: quantized description of a particle, conventional formulation of string

201: theory uses a first quantized formalism. In this formulation the spectrum

202: of single `particle' states in the theory are obtained by quantizing the

203: vibrational modes of a single string. Each such state is characterized by

204: its

205: energy $E$ and momentum $p$ besides  other quantum numbers, and

206: occasionally one finds states for which $E^2-p^2<0$. Since $E^2-p^2$ is

207: identified as the mass$^2$ of a particle, these states correspond to

208: particles of negative mass$^2$, {\it i.e.} tachyons.

209: 

210: The simplest

211: example

212: of such a tachyon appears in the $(25+1)$ dimensional bosonic string

213: theory. This theory has closed strings as its fundamental excitations, and

214: the lowest mass$^2$ state of this theory turns out to be tachyonic.

215: One might suspect that this tachyon

216: may have the same origin as in a quantum field theory, {\it i.e.} we may

217: be carrying out perturbation expansion around an unstable point, and that

218: the tachyon may be removed once we expand the theory about a stable

219: minimum of the potential.

220: Unfortunately, the first quantized description of string theory

221: does not allow us

222: to test this hypothesis. In particular, whether the closed string tachyon

223: potential  in the

224: bosonic string theory has a stable minimum still remains an unsolved

225: problem, and many people believe that this theory is inconsistent due to

226: the presence of the tachyon in its spectrum.

227: Fortunately various versions of superstring theories, defined in (9+1)

228: dimensions, have tachyon free closed string

229: spectrum. These theories are the starting points of most

230: attempts at constructing a unified theory of nature.

231: 

232: Besides closed strings, some string theories also contain open

233: string excitations with appropriate boundary conditions at the two

234: ends of the string. According to our current understanding, open

235: string excitations exist only when we consider a theory in the

236: presence of soliton like configurations known as

237: D-branes\cite{9510017,9611050,0007170}. Conversely, inclusion of open 

238: string

239: states in the spectrum implies that we are quantizing the theory

240: in the presence of a D-brane. To be more specific, a D-$p$-brane is a

241: $p$-dimensional extended object, and in the presence of such a

242: brane lying along a $p$-dimensional hypersurface $S$, the theory

243: contains open string excitations whose ends are forced to move

244: along the surface $S$. In the presence of $N$ D-branes (not

245: necessarily of the same kind) the spectrum contains $N^2$

246: different types of open string, with each end lying on one of the

247: $N$ D-branes. The physical interpretation of these open string

248: states is that they represent quantum excitations of the system of

249: D-branes.

250: 

251: It turns out that in some cases the spectrum of open string states

252: on a system of D-$p$-branes also contains tachyon. This happens

253: for example on D-$p$-branes in bosonic string theory for any $p$,

254: and D-$p$-branes in type IIA / IIB superstring theories for odd /

255: even values of $p$. Again, from our experience in quantum field

256: theory one would guess that the existence of the open string

257: tachyons represents an instability of the D-brane system whose

258: quantum excitations they describe. The natural question that

259: arises then is: is there a stable minimum of the tachyon potential

260: around which we can quantize the theory and get sensible results?

261: 

262: Although our understanding of this subject is still not complete,

263: last several years have seen much progress in answering this question.

264: These notes are designed to primarily review the main developments in this

265: subject.

266: 

267: \subsection{Organisation of the review} \label{sorg}

268: 

269: This review is organized as follows. In section \ref{s2} we give a

270: summary of the main results reviewed in this article.

271: In sections \ref{s3} - \ref{snc} we analyze

272: time independent classical solutions involving the open string tachyon 

273: using various techniques. Section \ref{s3} uses

274: the correspondence between two dimensional conformal field

275: theories and classical solutions of the equations of motion in

276: open string field theory. Section \ref{s4} is based on direct analysis of 

277: the equations

278: of motion of open string field theory. In sections \ref{s7} and

279: \ref{snc} we discuss application of the methods of boundary string

280: field theory and non-commutative field theory respectively. In section 

281: \ref{s10} we construct and analyze the

282: properties of time dependent solutions involving the tachyon. In

283: section \ref{s5.3} we describe an effective field theory which

284: reproduces qualitatively some of the results on time independent

285: and time dependent classical solutions involving the tachyon.

286: Section \ref{s5} is devoted to the discussion of other toy models,

287: {\it e.g.} field theories with singular potential and $p$-adic

288: string theory, which exhibit some of the features of the static

289: solutions involving the open string tachyon. In section \ref{s9}

290: we study the effect of closed string emission from the time

291: dependent rolling tachyon background on an unstable D-brane. In

292: section \ref{s11} we apply the methods discussed in this review to

293: study the dynamics and decay of an unstable D0-brane in two

294: dimensional string theory, and compare these results with exact

295: description of the system using large $N$ matrix models. Finally

296: in section \ref{sopenclosed} we propose an open string

297: completeness conjecture and generalized holographic principle which

298: explain some of the results of sections \ref{s9} and \ref{s11}.

299: 

300: Throughout this paper we work in the units:

301:  \be \label{econv1}

302: \hbar = c =\alpha' = 1\, .

303:  \ee

304: Thus in this unit the fundamental string tension is $(2\pi)^{-1}$.

305: Also our convention for the space-time metric will be

306: $\eta_{\mu\nu}={\rm diag}(-1,1,\ldots 1)$.

307: 

308: Before concluding this section we would like to caution the reader that 

309: this review 

310: does not cover all aspects of tachyon condensation. For example we do not 

311: address open string tachyon condensation on D$p$-D$p'$ brane system or 

312: branes at angles\cite{9704006,9703217}. We also do not review various 

313: attempts to 

314: find possible cosmological applications of the open string 

315: tachyon\cite{9812483,0105204,0105203,0107058,0204008}; nor do we 

316: address issues 

317: involving 

318: closed 

319: string tachyon condensation\cite{0108075}. We refer the reader to the 

320: original papers and their citations in spires database for learning these 

321: subjects.

322: 

323: Finally we would like to draw the readers' attention to many other

324: reviews where different aspects of tachyon condensation have been

325: discussed. A partial list includes 

326: refs.\cite{9908126,9908144,0005029,0102076,0102085,0106195,

327: 0301094,0311017,

328: 0405064}. For some early studies in open string tachyon dynamics, 

329: see \cite{bar1,bar2,bar3}.

330: 

331: 

332: \sectiono{Review of Main Results}

333: \label{s2}

334: 

335: In this section we summarize the main results reviewed in this article. 

336: The derivation of these results will be discussed in the rest of this 

337: article.

338: 

339: \subsection{Static solutions in superstring theory} \label{s2.1}

340: 

341: We begin our discussion by reviewing the properties of D-branes in

342: type IIA and IIB superstring theories. D$p$-branes are by definition

343: $p$-dimensional extended objects on which fundamental open strings can

344: end. It is well known\cite{dai,leigh,9407031} that type IIA/IIB

345: string theory contains BPS D$p$-branes for even / odd

346: $p$,

347: and that these D-branes carry Ramond-Ramond (RR) charges\cite{9510017}.

348: These

349: D-branes are oriented, and have definite mass per unit $p$-volume known as

350: tension. The tension of a BPS D$p$-brane in type IIA/IIB string theory is

351: given by:

352:  \be \label{e2.1}

353: \TT_p = (2\pi)^{-p} \, g_s^{-1}\, ,

354:  \ee

355: where $g_s$ is the closed string coupling constant. The BPS D-branes are

356: stable, and all the open string modes living on such a brane have

357: mass$^2\ge 0$. Since these branes are

358: oriented, given a specific BPS D$p$-brane, we shall call a D$p$-brane

359: with opposite orientation an

360: anti-D$p$-brane, or a $\bd p$-brane. The D0-brane in type IIA string

361: theory also has an anti-particle known as $\bd$0-brane, but we cannot

362: describe it as a D0-brane with reversed orientation.

363: 

364: Although a BPS D$p$-brane does not have a negative mass$^2$ (tachyonic) 

365: mode, if we consider a

366: coincident BPS D$p$-brane - $\bd p$-brane pair, then the open string

367: stretched

368: from the brane to the anti-brane (or vice-versa) has a tachyonic

369: mode\cite{9403040,9511194,9604091,9604156,9612215}. This is due to the 

370: fact 

371: that 

372: the GSO projection rule for these open 

373: strings is opposite of that for open strings whose both ends lie on the 

374: brane (or the anti-brane). As a result the ground state in the 

375: Neveu-Schwarz (NS) sector, which is normally removed from the spectrum by 

376: GSO projection, now becomes part of the spectrum, giving rise to a 

377: tachyonic mode. Altogether there are 

378: two tachyonic modes in the spectrum, -- one from the open string 

379: stretched from the brane to the anti-brane and the other from the open 

380: string stretched from the anti-brane to the brane.

381: The mass$^2$ of each of these tachyonic modes is given by 

382:  \be \label{e2.3}

383: m^2 = -{1\over 2}\, .

384:  \ee

385: 

386: 

387: Besides the stable BPS D$p$-branes, type II string theories also

388: contain in their spectrum unstable, non-BPS

389: D-branes\cite{9805019,9806155,9808141,9809111,9901014}.

390: The simplest way to define these D-branes in IIA/IIB string theory is to

391: begin with a coincident BPS D$p$ -- $\bd p$-brane pair in type IIB/IIA

392: string theory, and then take an orbifold of the theory by $(-1)^{F_L}$,

393: where ${F_L}$ denotes the contribution to the space-time fermion

394: number from the left-moving sector of the world-sheet. 

395: Since the RR fields are odd

396: under $(-1)^{F_L}$, all the RR fields of type IIB/IIA theory

397: are

398: projected out by the $(-1)^{F_L}$ projection. The twisted sector states

399: then give us back the  RR fields of type IIA/IIB theory. Since

400: $(-1)^{F_L}$ reverses the sign of the RR charge, it

401: takes a BPS D$p$-brane to a $\bd p$-brane and vice versa. As a result its 

402: action on the

403: open string states on a D$p$-$\bd p$-brane system is to

404: conjugate the Chan-Paton

405: factor by the exchange operator

406: $\sigma_1$. Thus modding out the D$p$ - $\bd p$-brane by $(-1)^{F_L}$

407: removes

408: all open string states with Chan-Paton factor $\sigma_2$ and $\sigma_3$

409: since these

410: anti-commute with $\sigma_1$, but keeps the open string states with 

411: Chan-Paton factors $I$ and $\sigma_1$. This gives us a non-BPS 

412: D$p$-brane\cite{9812031}.

413: 

414: The non-BPS D-branes

415: have precisely those dimensions which BPS D-branes do not have. Thus

416: type IIA string theory has non-BPS D$p$-branes for odd $p$ and type IIB

417: string theory has non-BPS D$p$-branes for even $p$. These branes are

418: unoriented and carry a fixed mass per unit $p$-volume, given by

419:  \be \label{e2.2}

420: \wt\TT_p = \sqrt 2\, (2\pi)^{-p} \, g_s^{-1}\, .

421:  \ee

422: The most important feature that distinguishes the non-BPS D-branes from

423: BPS D-branes is that the spectrum of open strings on a non-BPS D-brane

424: contains a single mode of negative mass$^2$ besides infinite number of

425: other modes of

426: mass$^2\ge 0$. This tachyonic mode can be identified as a particular 

427: linear combination of the two tachyons living on the original 

428: brane-antibrane pair that survives the $(-1)^{F_L}$ projection, and has 

429: the same mass$^2$ as given in \refb{e2.3}.

430: Another important feature that distinguishes a BPS D$p$-brane from a

431: non-BPS D$p$-brane is that unlike a BPS D$p$-brane which is charged

432: under the RR $(p+1)$-form gauge field of string theory, a

433: non-BPS D-brane is neutral under these gauge fields.

434: Various other properties of

435: non-BPS D-branes have been reviewed in \cite{9904207,9905006,9908126}.

436: 

437: Our main goal will be to understand the dynamics of these tachyonic modes. 

438: This however is not a simple task.

439: The dynamics of open strings living on a D$p$-brane is described by a

440: $(p+1)$ dimensional (string) field theory, defined such that the free

441: field quantization of the field theory reproduces the spectrum of open

442: strings on the D$p$-brane, and the S-matrix elements computed from this

443: field theory reproduce the S-matrix elements of open string theory on the

444: D-brane. On a non-BPS D-brane the existence of a single scalar tachyonic

445: mode shows that the corresponding open string field theory must contain a

446: real scalar field $T$ with mass$^2=-1/2$, whereas the same reasoning shows

447: that open string field theory associated with a coincident

448: brane-anti-brane system must contain two real scalar fields, or

449: equivalently one complex scalar field $T$ of mass$^2=-1/2$. However these

450: fields have non-trivial coupling to all the infinite number of other

451: fields in open string field theory, and hence one cannot study the

452: dynamics of these tachyonic modes in isolation. Furthermore since the

453: $|mass^2|$ of the tachyonic modes is of the same order of magnitude as

454: that of the other heavy modes of the string, one cannot work with a simple

455: {\it low energy} effective action obtained by integrating out the other

456: heavy

457: modes of the string. This is what makes the analysis of the tachyon

458: dynamics non-trivial. Nevertheless, it is convenient to state the results

459: of the analysis in terms of an effective action

460: $S_{eff}(T, \ldots)$ obtained by

461: formally integrating out all the positive mass$^2$ fields.

462: This is what we shall

463: do.\footnote{At this stage we would like to remind the reader that

464: our analysis will

465: be only at the level of classical open string field theory,

466: and hence integrating

467: out the heavy fields simply amounts to eliminating them by their equations

468: of motion.

469: } Here $\ldots$ stands for all the massless bosonic fields, which in

470: the case of

471: non-BPS D$p$-branes include one gauge field and $(9-p)$

472: scalar fields associated

473: with the transverse coordinates.

474: For D$p$-$\bd p$ brane pair the massless fields

475: consist of two $U(1)$ gauge fields and $2(9-p)$ transverse scalar fields.

476: 

477: First we shall state two properties of $S_{eff}(T,\ldots)$ which are trivially

478: derived from the analysis of the tree level S-matrix:

479:  \begin{enumerate}

480: \item For a non-BPS D-brane the tachyon effective action has a $Z_2$

481: symmetry under $T\to -T$, wheras for a brane-anti-brane system the tachyon

482: effective action has a phase symmetry under $T\to e^{i\alpha}T$.

483: 

484: \item Let $V(T)$ denote the tachyon effective potential, defined such that for

485: space-time independent field configuration,

486: and with all the massless fields set to

487: zero, the tachyon effective action $S_{eff}$

488: has

489: the form:

490:  \be \label{e2.4}

491: -\int d^{p+1} x \, V(T)\, .

492:  \ee

493: In that case $V(T)$ has a maximum at $T=0$. This is a straightforward

494: consequence of the fact that the mass$^2$ of the field $T$ is

495: given by $V''(T=0)$, and this is known to be negative. We shall

496: choose the additive constant in $V(T)$ such that $V(0)=0$.

497: \end{enumerate}

498: The question that we shall be most interested in is whether $V(T)$ has a

499: (local) minimum, and if it does, then how does the theory behave around this

500: minimum? The answer to this question is summarized in the following three

501: `conjectures'

502: \cite{9805019,9805170,9807138,

503: 9808141,9911116,9812031}:\footnote{Although

504: initially these properties were

505: conjectured, by now

506: there is sufficient evidence for these conjectures

507: so that one can refer to them as

508: results rather than conjectures.}

509:  \begin{enumerate}

510: \item $V(T)$ does have a pair of

511: global minima at $T=\pm T_0$ for the non-BPS D-brane, and a one parameter

512: ($\alpha$) family of

513: global minima at $T=T_0 e^{i\alpha}$ for the brane-antibrane

514: system. At this minimum

515: the

516: tension of the original D-brane configuration

517: is exactly canceled by the negative

518: contribution of the potential $V(T)$. Thus

519:  \be \label{e2.5}

520: V(T_0) + \EE_p = 0\, ,

521:  \ee

522: where

523:  \be \label{e2.6}

524: \EE_p = \cases{\wt \TT_p \quad \hbox{for non-BPS D$p$-brane} \cr

525: 2\TT_p \quad \hbox{for D$p$-$\bd p$ brane pair} }\, .

526:  \ee

527: Thus the total energy density vanishes at the minimum of the

528: tachyon potential. This

529: has been illustrated in Fig.\ref{f1}.

530: 

531:  \begin{figure}[!ht]

532:  \begin{center}

533: \leavevmode

534: \epsfysize=5cm

535: \epsfbox{fig3.eps}

536: \end{center}

537: \caption{The tachyon potential on an unstable D-brane in

538: superstring theories. The tachyon potential on a

539: brane-antibrane system is obtained by revolving this diagram about the 

540: vertical axis.

541: } \label{f1}

542: \end{figure}

543: 

544: 

545: \item Since the total energy density vanishes at $T=T_0$, and

546: furthermore, neither

547: the non-BPS D-brane nor the brane-antibrane system carries any RR charge,

548: it is

549: natural to conjecture that the configuration $T=T_0$ describes the vacuum

550: without any

551: D-brane.

552: This in turn implies that in perturbation theory we should not get any

553: physical

554: open string states by quantizing the theory

555: around the minimum of the potential, since open string states live only on

556: D-branes. This is counterintuitive, since in conventional

557: field theories the

558: number

559: of perturbative physical states do not change as we go from

560: one extremum of the

561: potential to another extremum.

562: 

563:  \begin{figure}[!ht]

564:  \begin{center}

565: \leavevmode

566: \epsfysize=5cm

567: \epsfbox{fig4.eps}

568: \end{center}

569: \caption{The kink solution on a non-BPS D-brane.} \label{f2}

570: \end{figure}

571: 

572: 

573: \item Although there are no perturbative physical states

574: around the minimum of the

575: potential, the equations of motion derived from the tachyon effective action

576: $S_{eff}(T,\ldots)$ does have non-trivial time independent classical

577: solutions. It is conjectured that these

578: solutions represent lower dimensional D-branes. Some examples are

579: given

580: below:

581:  \begin{enumerate}

582: \item The tachyon effective

583: action on a non-BPS D$p$-brane admits a classical kink

584: solution as shown in Fig.\ref{f2}. This solution depends on

585: only one of the spatial

586: coordinates, labeled by $x^p$ in the figure, such that $T$ approaches $T_0$ as

587: $x^p\to\infty$ and

588: $-T_0$ as $x^p\to -\infty$, and interpolates between these two

589: values around $x^p=0$.

590: Since the total energy density vanishes for $T=\pm T_0$, we

591: see that for the above

592: configuration the energy density is concentrated around a

593: $(p-1)$ dimensional subspace $x^p=0$. This kink

594: solution describes a BPS

595: D-$(p-1)$-brane in the same theory\cite{9812031,9812135}.

596: 

597: \item There is a similar

598: solution on a brane-antibrane system, where the imaginary

599: part of the tachyon field

600: is set to zero, and the real part takes the form given in

601: Fig.\ref{f2}. This is not a

602: stable solution, but describes a non-BPS D-$(p-1)$-brane

603: in the same theory\cite{9805019,9808141}.

604: \item Since the tachyon field $T$

605: on a D$p$-$\bd p$-brane system is a complex field,

606: one can also construct a vortex solution where $T$ is a

607: function of two of the

608: spatial coordinates (say $x^{p-1}$ and $x^p$) and takes the form:

609:  \be \label{e2.7}

610: T = T_0 \, f(\rho) e^{i\theta}\, ,

611:  \ee

612: where

613:  \be \label{e2.8}

614: \rho = \sqrt{(x^{p-1})^2 + (x^p)^2}, \qquad \theta = \tan^{-1}(x^p/x^{p-1})\,,

615:  \ee

616: are the polar coordinates on the $x^{p-1}$-$x^p$ plane and the function 

617: $f(\rho)$ has the property:

618:  \be \label{e2.9}

619: f(\infty)=1, \qquad f(0)=0\, .

620:  \ee

621: Thus the potential energy associated with the solution vanishes as

622: $\rho\to\infty$. Besides the tachyon the solution also

623: contains an accompanying

624: background gauge field which makes

625: the covariant derivative of the tachyon fall off

626: sufficiently fast for large $\rho$ so

627: that the net energy density is concentrated

628: around the $\rho=0$ region. This gives a

629: codimension two soliton solution. This solution describes a BPS

630: D-$(p-2)$-brane in the same

631: theory\cite{9808141,0003124}.

632: 

633: \item If we take a coincident pair of non-BPS D-branes,

634: then the D-brane effective field theory

635: around $T=0$ contains a U(2) gauge field,

636: and there are four tachyon states represented by a $2\times 2$ hermitian

637: matrix

638: valued scalar field transforming in the

639: adjoint representation of this gauge group.

640: The $(ij)$ component of the matrix represents the tachyon in the open

641: string sector beginning on the $i$-th D-brane and ending on the $j$-th

642: D-brane. A family of minima of the tachyon potential can be found by

643: beginning with the configuration $T=T_0\pmatrix{1 & 0 \cr 0 & -1}$ which

644: represents the tachyon on the first D-brane at its minimum $T_0$ and the

645: tachyon on the second D-brane at its minimum $-T_0$, and then making an

646: SU(2) rotation. This gives a family of  minima of the form $T=T_0\, \hat

647: n.\vec \sigma$, where $\hat n$ is a unit vector and $\sigma_i$ are the

648: Pauli matrices.   At any of these minima of the tachyon potential

649: the SU(2) part of the gauge group is broken to U(1) by the vacuum

650: expectation value of

651: the tachyon. 

652: 

653: This theory contains a 't Hooft - Polyakov monopole

654: solution\cite{thooft,polyakov} which depends on three of the spatial 

655: coordinates $\vec x$, and

656: for which

657: the asymptotic form of the tachyon and the SU(2) gauge field strengths

658: $F^a_{\mu\nu}$

659: are given by:

660:  \be \label{ethooft}

661: T(\vec x) \simeq T_0 \, {\vec\sigma.\vec x \over |\vec x|}\, , \qquad

662: F^a_{ij}(\vec

663: x) \simeq

664: \epsilon^{aij} {x^a\over |\vec x|^3}\, .

665:  \ee

666: The energy density of this solution is concentrated around $\vec x=0$ and

667: hence

668: this gives

669: a codimension 3 brane. This solution describes

670: a BPS D-$(p-3)$-brane in the same

671: theory\cite{9812135,0003124}.

672: 

673: \item If we consider a system of two D$p$-branes and two $\bd p$-branes, all

674: along the same plane, then the

675: D-brane world-volume theory has an $U(2)\times U(2)$

676: gauge field, and a $2\times 2$

677: matrix valued complex tachyon field $T$, transforming in

678: the (2,2) representation of the gauge group. The $(ij)$ component of the

679: matrix represents the tachyon field coming from the open string with ends

680: on

681: the $i$-th D-brane and the $j$-th $\bd$-brane. In this case the minimum

682: of the tachyon potential where the 11 component of the tachyon

683: takes value $T_0 e^{i\alpha}$ and the 22 component of the tachyon takes

684: value $T_0 e^{i\beta}$ corresponds to $T=T_0\pmatrix{e^{i\alpha} & 0\cr 0

685: & e^{i\beta}}$. A family of minima may now be found by making arbitrary

686: U(2) rotations from the left and the right. This gives $T=T_0\, U$ with

687: $U$ being an arbitrary $U(2)$ matrix.

688: 

689: Let $A_\mu^{(1)}$ and $A_\mu^{(2)}$

690: denote the gauge fields in the two SU(2) gauge groups. Then we can construct a

691: codimension 4 brane solution where the fields depend on four of the spatial

692: coordinates, and have the asymptotic behaviour:

693:  \be \label{e2.10}

694: T \simeq T_0 \, U(x^{p-3}, x^{p-2}, x^{p-1}, x^p), \qquad A_\mu^{(1)}

695: \simeq {i} \p_\mu U U^{-1}, \qquad  A_\mu^{(2)}\simeq 0 \, ,

696:  \ee

697: where $U$ is an SU(2)

698: matrix valued function of four spatial coordinates, corresponding to the

699: identity map

700: (winding number one map) from

701: the surface $S^3$ at spatial infinity to the SU(2) group

702: manifold. This describes a

703: BPS D-$(p-4)$-brane in the same theory \cite{9808141,0003124}.

704: 

705: \end{enumerate}

706: 

707: Quite generally if we begin

708: with sufficient number of non-BPS D9-branes in type IIA

709: string theory, or D9-$\bd 9$-branes in type IIB string theory, we can

710: describe any lower dimensional D-brane as classical solution in this

711: open string field theory \cite{9810188,9812135,0003124}. This has led to a

712: classification of

713: D-branes using

714: a branch of mathematics  known as 

715: K-theory\cite{9810188,9812135,9812226,9901042,

716: 9902160,9904153,

717: 9905034,

718: 9907140,

719: 9908091,9908121,

720: 9910109,9912279,0001143,

721: 0002023,0006223,

722: 0007175,0010007,

723: 0012164,0108100,

724: 0111151,

725: 0111169,0212059,

726: 0304018}.

727: 

728: \end{enumerate}

729: 

730: \subsection{Time dependent solutions in superstring theory} \label{s2.2}

731: 

732: So far we  have only discussed time independent solutions of the tachyon

733: equations of

734: motion. One could also ask questions about time dependent solutions. In

735: particular,  given that the tachyon potential on a non-BPS D$p$-brane or a

736: D$p$-$\bd p$ pair has the form given in

737: Fig.\ref{f1}, one

738: could ask:  what happens if we displace the tachyon from the maximum of

739: the potential

740: and let it  roll down towards its minimum?\footnote{For simplicity in this

741: section we shall only describe spatially homogeneous time dependent

742: solutions, but more general solutions which depend on both space and time

743: coordinates can also be studied\cite{0207105,0212248}.} If $T$ had been an

744: ordinary scalar field

745: then the answer  is simple: the tachyon field $T$ will simply oscillate

746: about the

747: minimum $T$ of the  potential, and in the absence of any dissipative force

748: (as is the

749: case at the classical level) the oscillation will continue for ever.

750: The energy density $T_{00}$ will remain constant during this oscillation,

751: but other

752: components of the energy-momentum  tensor, {\it e.g.} the pressure

753: $p(x^0)$, defined through $T_{ij}=p(x^0)\, \delta_{ij}$ for $1\le i,j \le

754: p$,

755: will oscillate

756: about their average value.

757: However for the

758: case of the string theory tachyon  the answer is different and somewhat

759: surprising\cite{0203211,0203265}.

760: It turns out that for the rolling tachyon solution on an unstable

761: D-brane the  energy density on the brane remains constant as in

762: the case of

763: a usual scalar field, but the pressure, instead of oscillating about an average

764: value, goes to zero asymptotically. More precisely, the non-zero

765: components of $T_{\mu\nu}$ take the form\footnote{The energy momentum

766: tensor

767: $T_{\mu\nu}$ is confined to the plane of the original D-brane, and

768: hence all expressions for

769: $T_{\mu\nu}$ are accompanied by a $\delta$-function in the transverse

770: coordinates which we shall denote by $\delta(x_\perp)$.

771: This factor may occasionally be

772: omitted for brevity. Also, only the

773: components of the stress tensor along the world-volume of the brane are

774: non-zero, {\it i.e.} $T_{\mu\nu}\ne 0$ only for $0\le \mu, \nu \le p$.}

775:  \ben \label{e2.11}

776: && T_{00}=\EE \, \delta(x_\perp), \qquad T_{ij} = p(x^0) \, \delta_{ij},

777: \qquad 1\le i,j\le p\nonumber \\

778: && p(x^0) = -\EE_p \, \tf(x^0)\, \delta(x_\perp)\, ,

779:  \een

780: where $\EE$ is a constant labelling the energy density on the

781: brane, $\EE_p$ is given by \refb{e2.6}, $\delta(x_\perp)$ denotes a

782: delta-function in the coordinates transverse to the brane and the function

783: $\tf(x^0)$ vanishes as $x^0\to \infty$. In order to give the 

784: precise form of $\wt f(x^0)$ we need to consider two

785: different

786: cases:

787:  \begin{enumerate}

788: \item $\EE\le \EE_p$: In this case we  can label the solution by

789: a parameter

790: $\tl$ defined through the relation:

791:  \be \label{e2.12}

792: T_{00} = {\EE_p} \, \cos^2(\pi \tl) \, \delta(x_\perp).

793:  \ee

794: $T_{00}$ includes the contribution from the tension of the D-brane(s) as

795: well as the tachyon kinetic and potential energy.

796: Since the total energy density available to the system is less than $\EE_p$,

797: -- the energy density at the maximum of the tachyon potential describing the

798: original brane configuration, -- at some instant of time during its

799: motion the tachyon is expected to come to rest at some point away from

800: the maximum of the

801: potential. We can  choose this instant of time as

802: $x^0=0$. The function $\tf(x^0)$ in this case takes the form:

803:  \be \label{e2.13}

804: \tf(x^0) = {1\over 1 + e^{\sqrt 2 x^0} \sin^2(\tl\pi)} + {1 \over

805: 1 + e^{-\sqrt 2 x^0} \sin^2(\tl\pi)} - 1\, .

806:  \ee

807: {}From this we see that as $x^0\to \infty$, $\tf(x^0)\to 0$. Thus the

808: pressure

809: vanishes asymptotically.

810: 

811: Note that for $\tl={1\over 2}$ both $T_{00}$ and $p(x^0)$ vanish

812: identically. Thus this solution has the natural interpretation as the

813: tachyon being placed at the minimum of its potential. The solution for

814: $\tl={1\over 2}+\epsilon$ is identical to the one at $\tl={1\over

815: 2}-\epsilon$; thus the inequivalent set of solutions are obtained by

816: restricting $\tl$ to the range $[-{1\over 2}, {1\over 2}]$.

817: 

818: \item $\EE\ge \EE_p$: In this case we can label the solutions by

819: a parameter

820: $\tl$ defined through the relation:

821:  \be \label{e2.14}

822: T_{00} = {\EE_p} \cosh^2(\pi\tl) \, \delta(x_\perp)\, .

823:  \ee

824: Since the total energy density available to the system is

825: larger than $\EE_p$,

826: at some instant of time during its

827: motion the tachyon is expected to pass the point $T=0$ where the potential

828: has a maximum.

829: We can choose our initial condition such that at

830: $x^0=0$ the tachyon is at the maximum of the potential and has a non-zero

831: velocity. The function $\tf(x^0)$ in this case takes the

832: form:\footnote{This result can be trusted only for $|\tl|\le

833: \sinh^{-1}1$.}

834:  \be \label{e2.15}

835: \tf(x^0) = {1\over 1 + e^{\sqrt 2 x^0} \sinh^2(\tl\pi)} + {1 \over

836: 1 + e^{-\sqrt 2 x^0} \sinh^2(\tl\pi)} - 1\, .

837:  \ee

838: Since as $x^0\to \infty$, $\tf(x^0)\to 0$, the pressure

839: vanishes asymptotically.

840: 

841: \end{enumerate}

842: 

843: The energy momentum tensor $T_{\mu\nu}$ given above is computed by 

844: studying the coupling of the D-brane to the graviton coming from the 

845: closed string sector of the theory. Besides the graviton, there are other 

846: massless 

847: states in superstring theory, and a D-brane typically couples to these

848: massless fields as well. We can in particular consider the sources $Q$ and 

849: $J^{(p)}_{\mu_1\ldots

850: \mu_p}$  produced by the D-brane for 

851: the 

852: dilaton

853: $\Phi_D$ and RR $p$-form gauge fields $C^{(p)}_{\mu_1\ldots

854: \mu_p}$ respectively. It turns out that as the

855: tachyon rolls down on a non-BPS D-$p$ brane or a D$p$-$\bd

856: p$-brane pair stretched along the $(x^1,x^2,\ldots x^p)$

857: hyperplane, it produces a source for the dilaton field of the

858: form:

859:  \be \label{edilcharge}

860: Q(x^0) = \EE_p \wt f(x^0)\, \delta(x_\perp)\, ,

861:  \ee

862: where $\wt f(x^0)$ is the same function as defined in \refb{e2.13} and

863: \refb{e2.15}.

864: Furthermore a rolling tachyon on a non-BPS D-$p$-brane produces

865: an RR $p$-form source of the form\cite{0204143}:

866:  \be \label{errcharge}

867: J^{(p)}_{1\ldots p} \propto \sin(\tl\pi) \left[ {e^{x^0/\sqrt 2} \over 1 +

868: \sin^2(\tl\pi) e^{\sqrt 2 x^0} } - {e^{-x^0/\sqrt 2} \over 1 +

869: \sin^2(\tl\pi) e^{-\sqrt 2 x^0} } \right] \, \delta(x_\perp)\, ,

870:  \ee

871: for the case $\EE\leq \EE_p$, and

872:  \be \label{errchargep}

873: J^{(p)}_{1\ldots p} \propto \sinh(\tl\pi) \left[ {e^{x^0/\sqrt 2} \over 1

874: +

875: \sinh^2(\tl\pi) e^{\sqrt 2 x^0} } + {e^{-x^0/\sqrt 2} \over 1 +

876: \sinh^2(\tl\pi) e^{-\sqrt 2 x^0} }\right] \, \delta(x_\perp)\, ,

877:  \ee

878: for the case $\EE\geq\EE_p$. The sources for other massless fields

879: vanish

880: for this solution.

881: 

882: 

883: The assertion that around

884: the tachyon vacuum there are no physical open string states implies

885: that there is no small oscillation of finite frequency around the minimum

886: of the

887: tachyon potential. The lack of oscillation in the pressure is consistent

888: with this

889: result.

890: However the existence of classical solutions with arbitrarily

891: small energy density (which can be achieved by taking $\tl$ close to 1/2 in

892: \refb{e2.12}) indicates that quantization of open string field

893: theory around

894: the tachyon vacuum does give rise to non-trivial quantum states which in the

895: semi-classical limit are described by the solutions that we have found.

896: 

897: 

898: \subsection{Static and time dependent solutions in bosonic string theory}

899: \label{s2.3}

900: 

901: Bosonic string theory in (25+1) dimensions has D$p$-branes for all

902: integers $p\le

903: 25$ with tension\cite{polbook}

904:  \be \label{e2.16}

905: \TT_p = g_s^{-1} \, (2\pi)^{-p}\, ,

906:  \ee

907: where $g_s$ as usual denotes the closed string coupling constant and we

908: are using

909: $\alpha'=1$ unit.

910: The spectrum of open strings on each of these D-branes

911: contains a

912: single

913: tachyonic state with mass$^2=-1$, besides infinite number of other states of

914: mass$^2\ge 0$. Thus among the infinite number of fields appearing in the

915: string

916: field theory on a D$p$-brane, there is a scalar field $T$ with negative

917: mass$^2$.

918: If as in the case of superstring theory we denote by $S_{eff}(T,\ldots)$ the

919: effective action obtained by integrating out the fields with positive

920: mass$^2$, and by

921: $V(T)$ the effective potential for the tachyon obtained by restricting to

922: space-time

923: independent field configurations and setting the massless fields to zero,

924: then

925: $V(T)$ will have a maximum at $T=0$. Thus we can again ask: does the

926: potential

927: $V(T)$ have a (local) minimum, and if it does, how does the open string

928: field theory

929: behave around this minimum?

930: 

931: Before we go on to answer these questions, let us recall that

932: bosonic string theory also has a tachyon in the closed string

933: sector, and hence the theory as it stands is inconsistent. Thus

934: one might wonder why we should be interested in studying

935: D$p$-branes in bosonic string theory in the first place. The

936: reason for this is simply that 1) although closed string tachyons

937: make the quantum open string field theory inconsistent due to

938: appearance of closed strings in open string loop diagrams,

939: classical open string field theory is not directly affected by the

940: closed string tachyon, and 2) the classical tachyon dynamics on a

941: bosonic D$p$-brane has many features in common with that on a

942: non-BPS D-brane or a brane-antibrane pair in superstring theory,

943: and yet it is simpler to study than the corresponding problem in

944: superstring theory. Thus studying tachyon dynamics on a bosonic

945: D-brane gives us valuable insight into the more relevant problem

946: in superstring theory.

947: 

948: We now summarise the three conjectures describing the static properties of the

949: tachyon effective action on a bosonic D$p$-brane\cite{9902105,9911116}:

950: 

951:  \begin{figure}[!ht]

952:  \begin{center}

953: \leavevmode

954: \epsfysize=5cm

955: \epsfbox{fig5.eps}

956: \end{center}

957: \caption{The tachyon effective potential on a D$p$-brane in bosonic

958: string theory.

959: } \label{f3}

960: \end{figure}

961: 

962: 

963:  \begin{enumerate}

964: 

965: \item The tachyon effective potential $V(T)$ has a {\it local} minimum at

966: some value $T=T_0$,

967: and at this minimum the tension $\TT_p$ of the original D-brane is

968: exactly canceled

969: by the negative value $V(T_0)$ of the potential. Thus

970:  \be \label{e2.17}

971: V(T_0) + \TT_p = 0\, .

972:  \ee

973: The form of the potential has been shown in Fig.\ref{f3}. Note that unlike

974: in the case of superstring theory, in this case the tachyon potential does

975: not have a global minimum.

976: 

977: \item Since the total energy density vanishes at $T=T_0$, it is

978: natural to identify the configuration $T=T_0$ as the vacuum without any

979: D-brane.

980: This in turn implies that there are no physical perturbative open strings

981: states

982: around the minimum of the potential, since open string states live only on

983: D-branes.

984: 

985:  \begin{figure}[!ht]

986:  \begin{center}

987: \leavevmode

988: \epsfysize=5cm

989: \epsfbox{fig6.eps}

990: \end{center}

991: \caption{The lump solution on a D$p$-brane in bosonic string theory.}

992: \label{f4}

993: \end{figure}

994: 

995: \item Although there are no perturbative physical states around the

996: minimum of the

997: potential, the equations of motion derived from the tachyon effective action

998: $S_{eff}(T,\ldots)$ does have non-trivial time independent classical

999: {\it lump} solutions of various codimensions. A codimension $q$ lump

1000: solution on a

1001: D$p$-brane, for which $T$ depends on $q$ of the spatial coordinates and

1002: approaches

1003: $T_0$ as any one of these  $q$ coordinates goes to infinity,

1004: represents a

1005: D-$(p-q)$-brane of bosonic string theory. An example of a

1006: codimension 1

1007: lump solution has been shown in Fig.\ref{f4}.

1008: 

1009: \end{enumerate}

1010: 

1011: This summarises the properties of time independent solutions,

1012: but one can also ask

1013: about time dependent solutions. In particular we can ask: what happens if we

1014: displace the tachyon from the maximum of its potential and let it roll?

1015: Unlike in

1016: the case of superstrings, in this case the potential (shown in

1017: Fig.\ref{f3}) is not

1018: symmetric around $T=0$,

1019: and hence we expect different behaviour depending on whether we have

1020: displaced the

1021: tachyon to the left (away from the local minimum) or right (towards the

1022: local

1023: minimum). As in the case of superstring theory, the energy density

1024: on the brane remains

1025: constant

1026: during the motion, but the pressure along the brane evolves in time:

1027:  \be \label{e2.18}

1028: p(x^0) = - \TT_p \, \tf(x^0)\, \delta(x_\perp) \, .

1029:  \ee

1030: In order to specify the form of $\tf(x^0)$ we consider two cases

1031: separately.

1032: 

1033:  \begin{enumerate}

1034: 

1035: \item $T_{00} = \EE\delta(x_\perp)$, $\EE\leq \TT_p$: In this case we can

1036: parametrize $T_{00}$ as:

1037:  \be \label{e2.19}

1038: T_{00} = {\TT_p} \, \cos^2(\pi\tl) \, \delta(x_\perp)\, ,

1039:  \ee

1040: and choose the origin of the time coordinate $x^0$ such that at $x^0=0$ 

1041: the tachyon has zero

1042: velocity and is displaced from $T=0$ by a certain amount

1043: determined by the

1044: parameter $\tl$.

1045: Then

1046: the function $\tf(x^0)$ appearing in \refb{e2.18} is given

1047: by\cite{0203211,0203265}:

1048:  \be \label{e2.20}

1049: \tf(x^0)={1\over 1 + e^{x^0} \sin(\tl\pi)} + {1 \over

1050: 1 + e^{-x^0} \sin(\tl\pi)} - 1\, .

1051:  \ee

1052: Note that $\pm \tl$ gives the same $T_{00}$ but different $\tf(x^0)$. This

1053: is due to

1054: the

1055: fact that positive sign of $\tl$ corresponds to displacing the tachyon

1056: towards the

1057: local minimum of the potential, whereas negative value of $\tl$

1058: corresponds to

1059: displacing $T$ towards the direction in which the potential is unbounded

1060: from

1061: below. As we can see from \refb{e2.20}, for positive $\tl$ the function

1062: $\tf(x^0)$

1063: approaches zero as $x^0\to\infty$, showing that the system evolves to a

1064: pressureless

1065: gas.

1066: In particular, for $\tl={1\over 2}$,

1067:  \be \label{elhalf}

1068: \wt f(x^0) = 0\, .

1069:  \ee

1070: Thus $T_{\mu\nu}$ vanishes identically, and we can identify this solution

1071: to be the one where the tachyon is placed at the local minimum of the

1072: potential. On the other hand, for negative $\tl$, $\tf(x^0)$ blows up at

1073:  \be \label{e2.21}

1074: x^0 = \ln{1\over |\sin(\tl\pi)|} \equiv t_c\, .

1075:  \ee

1076: This shows that if we displace the tachyon towards the direction in which

1077: the

1078: potential is unbounded from below, the system hits a singularity at a

1079: {\it finite

1080: time.}

1081: 

1082: 

1083: \item $T_{00}  = \EE\delta(x_\perp)$, $\EE\geq \TT_p$: In this case we

1084: can parametrize $T_{00}$ as:

1085:  \be \label{e2.22}

1086: T_{00} = {\TT_p} \, \cosh^2(\pi\tl) \, \delta(x_\perp)\, .

1087:  \ee

1088: Then for an appropriate choice of the origin of the time coordinate $x^0$

1089: the function $\tf(x^0)$ appearing in \refb{e2.18} is given

1090: by\cite{0203211,0203265}:

1091:  \be \label{e2.23}

1092: \tf(x^0) = {1\over 1 + e^{x^0} \sinh(\tl\pi)} + {1 \over

1093: 1 - e^{-x^0} \sinh(\tl\pi)} - 1\, .

1094:  \ee

1095: This equation is expected to be

1096: valid only for $|\tl|\le \sinh^{-1} 1$.

1097: Again we see that $\pm \tl$ gives the same $T_{00}$ but different

1098: $\tf(x^0)$. Positive

1099: sign of

1100: $\tl$

1101: corresponds to pushing the tachyon towards the

1102: local minimum of the potential, whereas negative value of $\tl$

1103: corresponds to

1104: pushing $T$ towards the direction in which the potential is unbounded from

1105: below. For positive $\tl$ the function $\tf(x^0)$

1106: approaches zero as $x^0\to\infty$, showing that the system evolves to a

1107: pressureless

1108: gas. On the other hand, for negative $\tl$, $\tf(x^0)$ blows up at

1109:  \be \label{e2.24}

1110: x^0 = \ln{1\over |\sinh(\tl\pi)|}\, .

1111:  \ee

1112: This again shows that if we displace the tachyon towards the direction in

1113: which the

1114: potential is unbounded from below, the system hits a singularity at a finite

1115: time.

1116: 

1117: \end{enumerate}

1118: 

1119: Bosonic string theory also has a massless dilaton field and we can

1120: define the dilaton charge density as the source that couples to this

1121: field. As in the case of superstring theory, a rolling tachyon on

1122: a D-$p$-brane of bosonic string theory produces a source for the

1123: dilaton field

1124:  \be \label{ebosdil}

1125: Q = \TT_p \, \wt f(x^0)\, \delta(x_\perp)\, ,

1126:  \ee

1127: with $\wt f(x^0)$ given by eq.\refb{e2.20} or \refb{e2.23}.

1128: 

1129: \subsection{Coupling to closed strings and the open string completeness 

1130: conjecture} \label{ssummary3}

1131: 

1132: So far we have discussed the dynamics of the open string tachyon at the

1133: purely classical level, and have ignored the coupling of the

1134: D-brane to closed strings. Since D-branes act as sources for various 

1135: closed

1136: string fields, a time dependent open string field

1137: configuration such as the rolling tachyon solution acts as a time

1138: dependent source for closed string fields, and produces closed string

1139: radiation. This can be computed using the standard techniques. For

1140: unstable Dp-branes with all $p$ directions wrapped on circles, one 

1141: finds that the total energy carried by 

1142: the closed string radiation is infinite\cite{0303139,0304192}. However 

1143: since the initial D$p$-brane has 

1144: finite energy it is appropriate to regulate this divergence by putting an 

1145: upper cut-off on the energy of the emitted closed string. A natural choice 

1146: of this cut-off is the initial

1147: energy

1148: of the D-brane. In that case one finds that

1149: 

1150: \begin{enumerate}

1151: \item All the energy of the D-brane is radiated away into closed strings

1152: even though any single closed string mode carries a small ($\sim g_s$)

1153: fraction of the D-brane energy.

1154: 

1155: \item Most of the energy is carried by closed strings of mass $\sim

1156: 1/g_s$.

1157: 

1158: \item The typical momentum

1159: carried by these closed strings along directions transverse to the

1160: D-brane is of

1161: order $\sqrt{1/g_s}$, and the typical winding charge carried by these

1162: strings

1163: along directions tangential to the D-brane is also of order

1164: $\sqrt{1/g_s}$.

1165: 

1166: \end{enumerate}

1167: 

1168: {}From the first result one would tend to conclude that the effect of 

1169: closed string

1170: emission should invalidate the classical open string results on the

1171: rolling tachyon system discussed earlier. There are however some

1172: surprising coincidences:

1173: \begin{enumerate}

1174: \item The tree level open string analysis tell us that the final system

1175: associated with the rolling tachyon configuration has

1176: zero pressure. On the other hand closed string emission results tell us

1177: that the final closed strings have momentum/mass and winding/mass ratio of

1178: order $\sqrt{g_s}$ and hence pressure/ energy density ratio of order

1179: $g_s$. In the $g_s\to 0$

1180: limit this vanishes. Thus it appears that the classical open string

1181: analysis correctly predicts the equation of state of the final system of

1182: closed strings into which the system decays.

1183: 

1184: \item The tree level open string analysis tells us that the final system 

1185: has zero

1186: dilaton charge. By analysing the properties of the closed string radiation

1187: produced by the decaying D-brane one finds that these closed strings also

1188: carry zero dilaton charge. Thus the classical open string analysis

1189: correctly captures the properties of the final state closed strings

1190: produced during the D-brane decay.

1191: 

1192: \end{enumerate}

1193: 

1194: These results (together with some generalizations which will be 

1195: discussed briefly in section \ref{sph}) suggest that the classical open 

1196: string theory already

1197: knows about the properties of the final state closed strings produced by

1198: the decay of the D-brane\cite{0305011,0306137}. This can be formally

1199: stated as an open string completeness

1200: conjecture according to which the complete dynamics

1201: of a

1202: D-brane is captured by the quantum open string theory without any need to

1203: explicitly consider the coupling of the system to closed

1204: strings.\footnote{Previously this was called the open-closed string

1205: duality conjecture\cite{0306137}. However since there are many different

1206: kinds of open-closed string duality conjecture, we find the name open

1207: string completeness conjecture more appropriate. In fact the proposed

1208: conjecture is not a statement of equivalence between the open and closed

1209: string description since the closed string theory could have many more

1210: states which are not accessible to the open string theory.} Closed strings

1211: provide

1212: a dual description of the system. This does not imply that any arbitrary

1213: state in string theory can be described in terms of open string theory on

1214: an unstable D-brane, but does imply that all the quantum states required 

1215: to

1216: describe the dynamics of a given D-brane

1217: are contained in the open string theory associated with that D-brane.

1218: 

1219: At the level of critical string theory one cannot prove this conjecture.

1220: However it turns out that this conjecture has a simple realization in a

1221: non-critical two dimensional string theory. This theory has two equivalent

1222: descriptions: 1) as a regular string theory in a somewhat complicated

1223: background\cite{DAVID,DISKAW} in which the world-sheet dynamics of the

1224: fundamental string is described by the direct sum of a free scalar field

1225: theory and

1226: the Liouville theory with central charge 25, and 2) as a theory of free

1227: non-relativistic

1228: fermions moving under a shifted inverted harmonic oscillator

1229: potential $-{1\over 2} \, q^2 + {1\over g_s}$\cite{GROMIL,BKZ,GINZIN}.

1230: Although in the free fermion description the potential is unbounded from

1231: below, the ground state of the system has all the negative energy states

1232: filled, and hence the second quantized theory is well defined. The map

1233: between these two theories is also known. In particular the closed string

1234: states

1235: in the first

1236: desciption are related to the quanta of the scalar field obtained by

1237: bosonizing the

1238: second quantized fermion field in the second

1239: description\cite{DASJEV,SENWAD,GROSSKLEB}.

1240: 

1241: In the regular string theory description the theory also has an unstable

1242: D0-brane with a tachyonic mode\cite{0101152}. The classical properties of

1243: this tachyon

1244: are identical to those discussed in section \ref{s2.3} in the context of

1245: critical bosonic string theory. In particular one

1246: can construct time dependent solution describing the rolling of the

1247: tachyon away from the maximum of the potential. Upon taking into account

1248: possible closed string emission effects one finds that as in the case of

1249: critical string theory, the D0-brane decays completely into closed

1250: strings\cite{0305159}.

1251: 

1252: By examining the coherent closed string field configuration produced in

1253: the D0-brane decay, and translating this into the fermionic description

1254: using the known relation between the closed string fields and the

1255: bosonized fermion, one discovers that the radiation produced by `D0--brane

1256: decay' precisely corresponds to a single fermion excitation in the theory.

1257: This suggests that the D0-brane in the first description should be 

1258: identified as the single fermion

1259: excitation in the second

1260: description of the theory\cite{0304224,0305159,0305194}. Thus its

1261: dynamics is described by that of

1262: a single particle moving under the inverted harmonic oscillator potential 

1263: with a lower-cutoff on the energy at the fermi level due to Pauli

1264: exclusion principle.

1265: 

1266: Given that the dynamics of a D0-brane in the first description is 

1267: described by an open string

1268: theory, and that in the second description a D0-brane is identified with 

1269: single 

1270: fermion excitation, we can conclude that the open 

1271: string theory for the D0-brane must be equivalent to

1272: the single particle

1273: mechanics with potential $-{1\over 2}q^2 + {1\over 

1274: g_s}$, 

1275: with an additional 

1276: constraint $E\ge 0$.

1277: A consistency check of this proposal is that the

1278: second derivative of the inverted harmonic oscillator potential at the

1279: maximum precisely matches the negative mass$^2$ of the open string tachyon

1280: living on the D0-brane\cite{0305159}.

1281: This `open string theory' clearly

1282: has the ability to describe the complete dynamics of the D0-brane 

1283: \i.e.\ the single fermion 

1284: excitations. It is possible but not necessary to describe the system in

1285: terms of the closed string field, \i.e.\ the scalar field obtained by 

1286: bosonizing the second quantized

1287: fermion field. This is in complete accordance 

1288: with the open 

1289: string

1290: completeness

1291: conjecture proposed earlier in the context of critical string theory.

1292: 

1293: 

1294: \sectiono{Conformal Field Theory Methods} \label{s3}

1295: 

1296: In this section we shall analyze time independent solutions involving the 

1297: open string tachyon using the

1298: well known correspondence between classical solutions of

1299: equations of

1300: motion of string theory, and two dimensional (super-)conformal field

1301: theories

1302: (CFT).

1303: A D-brane configuration in a space-time background is

1304: associated

1305: with a two dimensional conformal field theory on an infinite

1306: strip (which can be conformally mapped to a disk or

1307: the upper

1308: half plane) describing propagation of open string excitations on the 

1309: D-brane. Such conformal field theories are known as boundary conformal

1310: field

1311: theories (BCFT) since they are defined on surfaces with

1312: boundaries. The

1313: space-time background in which the D-brane lives determines the bulk

1314: component of

1315: the CFT, and associated with a particular D-brane configuration we have

1316: specific

1317: conformally invariant boundary conditions / interactions involving 

1318: various fields of this

1319: CFT.  Thus for example for a D$p$-brane in flat space-time we have

1320: Neumann boundary condition on the $(p+1)$ coordinate fields tangential to 

1321: the

1322: D-brane world-volume and Dirichlet boundary condition on the coordinate

1323: fields transverse to the D-brane. Different classical solutions

1324: in the open string field theory describing the dynamics of a D-brane are

1325: associated

1326: with different conformally invariant boundary interactions in this

1327: BCFT.

1328: More specifically, if we add to the original world-sheet action a boundary

1329: term

1330:  \be \label{ex21}

1331: \int dt \, V(t) \, ,

1332:  \ee

1333: where $t$ is a parameter labelling the boundary of the world-sheet and $V$

1334: is a boundary vertex operator in the world-sheet theory, then for a

1335: generic $V$ the conformal invariance of the theory is broken. But

1336: for

1337: every $V$ for which we have a (super-)conformal field theory, there is an

1338: associated solution of the classical open string field equations.

1339: Thus we can

1340: construct solutions of equations of motion of open string field theory by

1341: constructing appropriate conformally invariant boundary

1342: interactions in

1343: the BCFT

1344: describing the original D-brane configuration. This is the approach we

1345: shall take in

1346: this section.

1347: 

1348: In the rest of the section we shall outline the logical steps

1349: based on this approach which lead to the results on time

1350: independent solutions described in section \ref{s2}.

1351: 

1352: \subsection{Bosonic string theory} \label{s3.1}

1353: 

1354: We begin with a space-filling D25-brane of bosonic string theory. We 

1355: shall show

1356: that as stated in the third conjecture in section \ref{s2.3}, we can 

1357: regard the D24-brane as a codimension 1 lump solution of the

1358: tachyon

1359: effective action

1360: on the D25-brane.

1361: This is done

1362: in two steps:

1363:  \begin{enumerate}

1364: 

1365: \item First we find the conformally invariant BCFT associated with the

1366: tachyon

1367: lump

1368: solution on a D25-brane. This is done by

1369: finding a series of marginal deformations that connects the $T=0$

1370: configuration

1371: on the D25-brane to the tachyon lump solution.

1372: \item Next we show that this BCFT is identical to that describing a

1373: D-24-brane.

1374: This is done by following what happens to the original BCFT describing

1375: the D25-brane

1376: under this series of marginal

1377: deformations.

1378: \end{enumerate}

1379: Thus we first need to find a series of marginal deformations connecting the

1380: $T=0$ configuration to the tachyon lump solution on the D-25-brane. This

1381: is done as

1382: follows:

1383:  \begin{enumerate}

1384: \item Let us choose a specific direction $x^{25}$ on which the lump

1385: solution will eventually depend. For

1386: simplicity

1387: of notation we shall define $x\equiv x^{25}$. We first compactify $x$ on a

1388: circle of radius $R$. The resulting world-sheet theory is conformally

1389: invariant for every $R$.

1390: This configuration has energy per unit 24-volume given by:

1391:  \be \label{e3.1}

1392: 2\pi R \TT_{25} = R \TT_{24}\, .

1393:  \ee

1394: In deriving \refb{e3.1} we have used \refb{e2.16}.

1395: 

1396: \item At $R=1$ the {\it boundary operator} $\cos X$ becomes exactly marginal

1397: \cite{9402113,9404008,9811237,9902105}.

1398: A simple way to see this is as follows. For $R=1$ the bulk CFT has an

1399: enhanced

1400: $SU(2)_L\times SU(2)_R$ symmetry. The $SU(2)_{L,R}$ currents

1401: $J^a_{L,R}$ ($1\le a\le

1402: 3$) are:

1403:  \be \label{eb3}

1404: J^3_L = i \bar\p X_L, \quad J^3_R = i \p X_R, \quad J^1_{L,R} = \cos (2

1405: X_{L,R}) \quad

1406: J^2_{L,R}

1407: = \sin(2 X_{L,R})\, ,

1408:  \ee

1409: where $X_L$ and $X_R$ denote the left and right moving components

1410: of $X$ respectively:\footnote{In our convention left and right

1411: refers to the anti-holomorphic and holomorphic components of the

1412: field respectively.}

1413:  \be \label{eb4}

1414: X = X_L + X_R\, .

1415:  \ee

1416: For $\alpha'=1$ the fields $X_L$ and $X_R$ are normalized so that

1417:  \be \label{exnorm}

1418: \p X_R(z) \p X_R(w) \simeq -{1\over 2(z-w)^2}\, , \qquad \bar\p

1419: X_L(\bar z) \bar\p X_L(\bar w) \simeq -{1\over 2(\bar z-\bar

1420: w)^2}\, .

1421:  \ee

1422: For definiteness let us take the open string world-sheet to be the

1423: upper half plane with the real axis as its boundary. The Neumann

1424: boundary condition on $X$ then corresponds to:

1425:  \be \label{eb1}

1426: X_L=X_R \quad \to \quad J^a_L=J^a_R\, \quad \hbox{for $1\le a\le 3$}

1427:  \ee

1428: on the real axis.

1429: Using eqs.\refb{eb3}-\refb{eb1}

1430: the boundary operator $\cos X$ can be regarded as the

1431: restriction of $J^1_{L}$ (or

1432: $J^1_R$) to the real axis.

1433: Due to SU(2) invariance of the CFT we can now describe

1434: the theory in terms of a new free scalar field $\phi$, related to $X$ by

1435: an

1436: SU(2) rotation, so that

1437:  \be \label{eb2}

1438: J^1_L = i \bar\p \phi_L, \qquad J^1_R = i \p\phi_R\, .

1439:  \ee

1440: Thus in terms of $\phi$ the boundary operator $\cos

1441: X=\cos(2X_L)=\cos(2X_R)$ is proportional to the restriction of

1442: $i\bar\partial\phi_L$ (or $i\p\phi_R$) at the boundary. This is

1443: manifestly an exactly marginal operator, as it corresponds to

1444: switching on a Wilson line along $\phi$.

1445: 

1446:  \begin{figure}[!ht]

1447:  \begin{center}

1448: \leavevmode

1449: \epsfysize=5cm

1450: \epsfbox{fig2.eps}

1451: \end{center}

1452: \caption{The plot of $T(x) = -\al \cos x$ for $\al=.5$. This looks like a

1453: lump

1454: centered around $x=0$, but the height, being equal to $\al$, is

1455: arbitrary.}

1456: \label{f5}

1457: \end{figure}

1458: 

1459: Due to exact marginality of the operator $\cos X$, we can

1460: switch on a conformally invariant

1461: perturbation of the form:

1462:  \be \label{e3.2}

1463: - \al \int dt \cos (X(t)) = -i\alpha \int dt \, \bar\p

1464: \phi_L\, ,

1465:  \ee

1466: where $\al$ is an arbitrary constant and $t$ denotes a parameter labelling 

1467: the boundary of the world-sheet.

1468: From the

1469: target space view-point switching on a perturbation proportional to

1470: $-\cos X$

1471: amounts to giving the tachyon field a vev proportional to $-\cos x$.

1472: This in turn

1473: can be interpreted as the creation of a lump centered at $x=0$ (see

1474: Fig.\ref{f5}). At this

1475: stage however the amplitude $\al$ is arbitrary, and hence the lump has

1476: arbitrary

1477: height. Since the boundary perturbation \refb{e3.2} is marginal, the

1478: energy of

1479: the configuration stays constant during this deformation at its initial

1480: value at

1481: $R=1$, $\al=0$. Using \refb{e3.1} we get the energy per unit 24-volume to

1482: be

1483: $\TT_{24}$, {\it i.e.} the energy density of a D-24-brane!

1484: 

1485: 

1486: 

1487:  \begin{figure}[!ht]

1488:  \begin{center}

1489: \leavevmode

1490: \epsfysize=5cm

1491: \epsfbox{fig1.eps}

1492: \end{center}

1493: \caption{Marginal flow in the $R-\al$ plane.

1494: } \label{f6}

1495: \end{figure}

1496: 

1497: \item Since we are interested in constructing a lump solution at 

1498: $R=\infty$ we

1499: need to

1500: now take the radius back to

1501: infinity. However for a generic $\al$, as soon as we switch on a radius

1502: deformation,

1503: the boundary operator $\cos {X\over R}$ develops a one point function

1504: \cite{9902105}:

1505:  \be \label{e3.3}

1506: \left\la \cos {X(0)\over R} \right\ra_{R;\al} \propto (R-1) \sin(2\pi \al) 

1507: \, ,

1508:  \ee

1509: for $R\simeq 1$.

1510: This

1511: indicates that

1512: the configuration fails to satisfy the open string field equations and 

1513: hence no longer describes a BCFT.\footnote{This is related to the fact 

1514: that near $\alpha=0$ the operator $\cos(X/R)$ has dimension $1/R^2$ and 

1515: hence deformation by $\alpha\int dt\, \cos(X(t)/R)$ does not give a BCFT 

1516: for generic $\alpha$ and $R$.}  However if $\al=0$ or $1/2$, then the

1517: one point function vanishes, not only for $R\simeq 1$ but for all values

1518: of $R$

1519: \cite{9902105}. Thus for these values of $\al$ we can get a BCFT for

1520: arbitrary

1521: $R$. For

1522: $\al=0$ the resulting configuration is a D-25-brane, whereas for

1523: $\al=1/2$ we can

1524: identify the configuration as a tachyon lump solution on a D-25-brane for

1525: any value of $R\ge 1$.

1526: \end{enumerate}

1527: The motion in the $R-\al$ plane as we follow the three step

1528: process

1529: has been

1530: shown in Fig.\ref{f6}.

1531: 

1532: It now remains to show that the BCFT constructed this way with

1533: $\al=1/2$, $R=\infty$ describes a D-24-brane. For this we need to

1534: follow the fate of the BCFT under the three step

1535: deformation that takes us from $(\alpha=0, R=\infty)$ to $(\alpha= 

1536: 1/2, R=\infty)$. In the first step, involving reduction of $R$ from

1537: $\infty$ to 1, the D25-brane remains a D25-brane. The second

1538: step,-- switching on the perturbation \refb{e3.2}, -- does

1539: introduce non-trivial boundary interaction. It follows from the

1540: result of \cite{9402113,9404008,9811237,9902105} that at $R=1$ the

1541: BCFT at $\al=1/2$ corresponds to putting a Dirichlet boundary

1542: condition on the coordinate field $X$. A simple way to see this is

1543: as follows. Since the perturbation $\int dt \cos(X(t))$ is

1544: proportional to $\int dt J^1_L(t)$, the effect of this

1545: perturbation on any closed string vertex operator in the interior

1546: of the world-sheet will be felt as a rotation by $2\pi\al$ in the 

1547: $SU(2)_L$

1548: group about the 1-axis.\footnote{This explains the periodicity of 

1549: \refb{e3.3}

1550: under $\al\to \al +1$.} For $\al=1/2$ the angle of rotation is

1551: precisely $\pi$ and hence

1552: it changes $X_L$ to $-X_L$ in any closed string vertex operator

1553: inserted in the bulk. By redefining $-X_L$ as $X_L$ we can ensure

1554: that the closed string vertex operators remain unchanged, but as a

1555: result of this redefinition the boundary condition on $X$ changes

1556: from Neumann to Dirichlet:

1557:  \be \label{efx1}

1558: X_L = - X_R\, .

1559:  \ee

1560: Thus

1561: we can conclude that when probed by closed

1562: strings, the

1563: perturbed BCFT at $\al=1/2$ behaves as if we have Dirichlet boundary

1564: condition on

1565: $X$.

1566: 

1567: Since all other fields $X^\mu$ for $0\le\mu\le 24$ remain unaffected by

1568: this

1569: deformation, we see that the

1570: BCFT at $R=1$, $\al=1/2$ indeed describes a D-24-brane with its transverse

1571: direction compactified on a circle of radius 1. The subsequent fate of

1572: the

1573: BCFT under the radius deformation that takes us from $(\alpha=1/2, R=1)$ 

1574: to $(\alpha=1/2, R=\infty)$ then follows the fate of a D24-brane

1575: under such a

1576: deformation, {\it i.e.} the D24-brane remains a D24 brane as the radius

1577: changes. Thus

1578: the final BCFT at $R=\infty$ describes a D24-brane in non-compact

1579: space-time.

1580: 

1581: This establishes that a lump solution on a D-25-brane describes a

1582: D-24-brane. Note

1583: that the argument goes through irrespective of the boundary condition on

1584: the

1585: coordinates $X^1,\ldots X^{24}$; thus the same analysis shows that a

1586: codimension 1

1587: lump on a D$p$-brane describes a D-$(p-1)$ brane for any value of $p\ge

1588: 1$. Repeating this

1589: procedure $q$-times

1590: we can also establish that a codimension $q$ lump on a D$p$-brane

1591: describes a

1592: D-$(p-q)$-brane.

1593: 

1594: This establishes the third conjecture of section \ref{s2.3}. This in turn

1595: indirectly

1596: proves conjectures 1 and 2 as well. To see how conjecture 1 follows

1597: from

1598: conjecture 3, we note that D24-brane, and hence the lump solution, has a

1599: finite

1600: energy per unit 24-volume. This

1601: means that the lump solution must have vanishing energy density as

1602: $x^{25}\to\pm

1603: \infty$, since otherwise we would get infinite energy per unit 24-volume by

1604: integrating the energy density in the $x^{25}$ direction.

1605: Thus if $T_0$ denotes the value to which $T$ approaches as

1606: $x^{25}\to\pm\infty$,

1607: then the total energy density must vanish at $T=T_0$. Furthermore $T_0$

1608: must be a

1609: local extremum of the potential in order for the tachyon equation of

1610: motion to be

1611: satisfied as $x^{25}\to\pm\infty$. This shows

1612: the existence of a local extremum of the potential where the total energy

1613: density

1614: vanishes, as stated in conjecture 1.

1615: 

1616: To see how the second conjecture arises, note that D24-brane and hence the 

1617: lump solution supports

1618: open strings

1619: with ends moving on the $x^{25}=0$ plane. This means that if we go far 

1620: away from

1621: the lump

1622: solution in the $x^{25}$ direction, then there are no physical open string

1623: excitations in this region. Since the tachyon field configuration in this

1624: region is

1625: by definition the $T=T_0$ configuration, we arrive at the second

1626: conjecture that

1627: around $T=T_0$ there are no physical open string excitations.

1628: 

1629: Finally we note that our analysis leading to the BCFT associated with the

1630: tachyon

1631: lump solution is somewhat indirect. One could ask if in the diagram shown

1632: in

1633: Fig.\ref{f6} it is possible to go from $\al=0$ to $\al=1/2$ at any value

1634: of $R>1$

1635: directly, without following the circuitous route of first going down to

1636: $R=1$ and

1637: coming back to the desired value of $R$ {\it after switching on the 

1638: $\alpha$-deformation}. It turn out that it is possible to do this, but 

1639: not via marginal

1640: deformation. For a generic value of $R$, we need to perturb the BCFT

1641: describing

1642: D25-brane by an operator

1643:  \be \label{e3.4}

1644: -\al \int dt \, \cos (X(t) / R) \, ,

1645:  \ee

1646: which has dimension $R^{-2}$ and hence is a relevant operator for $R>1$.

1647: It is known

1648: that under this relevant perturbation the original BCFT describing the

1649: D25-brane

1650: flows into another BCFT corresponding to putting Dirichlet boundary

1651: condition on

1652: $X$ coordinate \cite{9406125,0003101}. In other words, although for a

1653: generic $\al$ the

1654: perturbation

1655: \refb{e3.4} breaks conformal invarinance, for a specific value of $\al$

1656: corresponding to the infrared fixed point, \refb{e3.4} describes a new BCFT

1657: describing the D-24-brane.\footnote{Of course this will, as usual, also

1658: induce flow in various other coupling constants labelling the boundary

1659: interactions. The precise description of these flows will depend on the

1660: renormalization scheme, and we could choose a suitable renormalization

1661: scheme where the other coupling constants do not flow. In space-time

1662: language, this amounts to {\it integrating out} the other fields.} By the

1663: usual correspondence between equations of motion

1664: in string theory and two dimensional BCFT, we would then conclude that

1665: open string

1666: field theory on a D-25-brane compactified on a circle of radius $R>1$ has

1667: a

1668: classical solution describing a D24-brane. This classical solution can be

1669: identified

1670: as the lump.

1671: 

1672: The operator $-\cos(X/R)$ looks ill defined in the $R\to\infty$ limit, but

1673: the

1674: correct procedure is to expand this in powers of $X/R$ and keep the leading

1675: term. This amounts to perturbing the D25-brane BCFT by an operator

1676: proportional to

1677:  \be \label{e3.5}

1678: \int dt X(t)^2\, .

1679:  \ee

1680: In the infrared this perturbation takes us to the BCFT of a D24-brane,

1681: localized at

1682: $x=0$. This result is useful in the study of tachyon

1683: condensation in

1684: boundary string field

1685: theory\cite{0009103,0009148,0010108}, and will be made use of in section

1686: \ref{s7}.

1687: 

1688: \subsection{Superstring theory} \label{s3.3}

1689: 

1690: In this section we shall generalize the analysis of section \ref{s3.1} to

1691: superstring theory showing that various classical solutions involving the

1692: tachyon on D9-$\bar{\rm D}9$-brane pair of IIB or non-BPS D9-brane of IIA

1693: represent lower dimensional D$p$-branes. For definiteness we shall

1694: illustrate in detail the representation of a non-BPS D8-brane of IIB as a

1695: kink solution on a D9-$\bar{\rm D}9$-brane pair\cite{9808141} (case 3(b)

1696: in section \ref{s2.1}), and then

1697: briefly

1698: comment on the other cases.

1699: 

1700: The tachyon state on a D9-$\bar{\rm D}9$-brane system comes from open

1701: strings with one leg on the D9-brane and the other leg on the $\bar{\rm

1702: D}9$-brane. Thus the corresponding vertex operator will carry an

1703: off-diagonal Chan-Paton factor which we can take to be the Pauli matrix

1704: $\sigma_1$ or $\sigma_2$. This gives rise to two real tachyon fields $T_1$

1705: and $T_2$ on the world-volume of this D-brane system, which can be

1706: combined into a complex tachyon $T=T_1+i T_2$. Thus in this convention the

1707: coefficients of $\sigma_1$ and $\sigma_2$ represent the real and

1708: imaginary parts respectively of the complex

1709: tachyon field.

1710: We shall show that a tachyonic kink involving the real part $T_1$ of $T$,

1711: with the imaginary part set to zero, represents a non-BPS D8-brane.

1712: 

1713: The vertex operator of the tachyon $T_1$ carrying

1714: momentum $k$ in the $-1$ picture\cite{FMS1,FMS} is given by:\footnote{A

1715: brief review of bosonization of superconformal ghosts, picture changing 

1716: and  physical open string vertex operators in superstring

1717: theory is given at the beginning of section \ref{s4.4}.}

1718:  \be

1719: \label{efh2} V_{-1}(k) = c \, e^{-\phi_g} \, e^{ik\cdot X} \otimes

1720: \sigma_1\, ,

1721:   \ee

1722: where $\phi_g$ is the bosonic field arising out of bosonization of

1723: the $ \beta$-$\gamma$ ghost system\cite{FMS}. As usual this vertex

1724: operator is inserted at the boundary of the world sheet. The

1725: on-shell condition is

1726:  \be \label{efh3} k^2 =

1727: {1\over 2}\, ,

1728:   \ee

1729: showing that the corresponding state has mass$^2=-{1\over 2}$. The

1730: same vertex operator in the $0$ picture is

1731:  \be \label{efh4} V_0 = -

1732: \sqrt 2\, c \, k\cdot\psi \, e^{ik\cdot X} \otimes \sigma_1\, ,

1733:   \ee

1734: where $\psi^\mu$ is boundary value of the world-sheet

1735: superpartners $\psi^\mu_L$ or $\psi^\mu_R$ of $X^\mu$. On the

1736: boundary $\psi^\mu_L$ and $\psi^\mu_R$ are equal. The fields

1737: $\phi_g$ and $\psi^\mu$ are normalized so that their left and

1738: right-moving components satisfy the operator product expansion:

1739:  \ben \label{eopephipsi}

1740: && \p \phi_{gR}(z) \p \phi_{gR}(w) \simeq -{1\over (z-w)^2}\, ,

1741: \qquad \bar\p \phi_{gL}(\bar z) \bar\p \phi_{gL}(\bar w) \simeq

1742: -{1\over (\bar

1743: z-\bar w)^2}\, , \nonumber \\

1744: && \psi^\mu_R(z) \psi^\nu_R(w) \simeq {\eta^{\mu\nu}\over z-w}, \qquad 

1745: \psi^\mu_L(\bar z)

1746: \psi^\nu_L(\bar w) \simeq {\eta^{\mu\nu}\over \bar z-\bar w}\, .

1747:  \een

1748: 

1749: In order to show that the kink solution involving this tachyon represents

1750: a

1751: non-BPS D8-brane,

1752: we proceed exactly as in the case of bosonic string theory, \i.e. we

1753: first find the BCFT associated with the tachyonic

1754: kink solution on the D9-$\bar{\rm D}9$-brane pair, and then show that this

1755: BCFT is identical to that of a non-BPS D8-brane. In order to find the

1756: BCFT associated with the tachyonic

1757: kink, we need to identify a series of steps which take us

1758: from the $T=0$ configuration to the tachyonic kink configuration. This is

1759: done as follows\cite{9808141}:

1760:  \begin{enumerate}

1761: \item

1762: We first compactify the $x\equiv x^9$ direction into a circle of radius

1763: $R$. We would like to take the radius to an appropriate critical value

1764: (analog of $R=1$ in the bosonic case) where we can create a kink solution

1765: via a marginal boundary deformation. However,

1766: in order to create a single kink on a circle, we need to have a

1767: configuration where the tachyon is anti-periodic along the circle.

1768: This can be achieved by switching on

1769: half a unit of Wilson line along the circle associated with one of the

1770: branes, since the tachyon field is charged under this gauge field. This is

1771: a boundary marginal deformation.

1772: 

1773:  \begin{figure}[!ht]

1774:  \begin{center}

1775: \leavevmode

1776: \epsfysize=5cm

1777: \epsfbox{figfh1.eps}

1778: \end{center}

1779: \caption{The plot of $T_1(x)/\alpha$ vs. $x/R$ in the range $(-\pi,

1780: \pi)$.

1781: } \label{fh1}

1782: \end{figure}

1783: 

1784: \item

1785: In the presence of this Wilson line,

1786: a configuration of the form:

1787:  \be \label{efh5}

1788: T_1(x) = \sqrt 2\,  \al\, \sin{x\over 2R}\, ,

1789:  \ee

1790: is an allowed configuration and

1791: has the shape of a kink solution (see Fig.\ref{fh1}).

1792: Here the normalization factor of $\sqrt 2$ has been chosen for

1793: convenience.

1794: The corresponding vertex operators in the $-1$ and the 0 pictures are

1795: given by, respectively,

1796:  \be \label{efh6}

1797: V_{-1} = c e^{-\phi_g} \sin{X\over 2R} \otimes \sigma_1\, ,

1798:  \ee

1799: and

1800:  \be \label{efh7}

1801: V_{0} = {i\over \sqrt 2 R} c \, \psi \, \cos{X\over 2R} \otimes

1802: \sigma_1\, ,

1803:  \ee

1804: where $X\equiv X^9$, $\psi\equiv \psi^9$.

1805: Thus switching on a background tachyon of the form \refb{efh5}

1806: corresponds to adding a world sheet perturbation:

1807:  \be \label{efh8}

1808: i\, {\al\over R} \, \int dt \, \psi \, \cos{X\over 2R} \otimes

1809: \sigma_1\, .

1810:  \ee

1811: We want to find a critical value of $R$ for which \refb{efh8} represents

1812: an exactly marginal deformation so that \refb{efh5} describes a solution 

1813: to 

1814: the classical open string field equations for any $\alpha$.

1815: The perturbing operator in \refb{efh8} has dimension ${1\over 4

1816: R^2}+{1\over 2}$.

1817: At $R=1/\sqrt 2$ this operator has dimension 1. Furthermore, one can show

1818: that at this radius it becomes

1819: an exactly marginal operator\cite{9808141,0108102,0108238}. To see this,

1820: note that at this radius

1821: we can fermionize the space-time boson $X$ and then rebosonise the

1822: fermions as follows:

1823:  \be \label{efh9}

1824: X=X_L + X_R\, ,

1825:  \ee

1826:  \be \label{efh10}

1827: e^{\pm i\sqrt 2 X_L} \sim {1\over \sqrt 2} \, (\xi_L \pm i \eta_L)\, ,

1828: \qquad

1829: e^{\pm i\sqrt 2 X_R} \sim {1\over \sqrt 2} \, (\xi_R \pm i \eta_R)\, ,

1830:  \ee

1831:  \be \label{efh11}

1832: e^{\pm i\sqrt 2 \phi_{L\atop R}}\sim {1\over \sqrt 2} \, (\xi_{L\atop R} 

1833: \pm 

1834: i

1835: \psi_{L\atop R}),

1836: \qquad

1837: e^{\pm i\sqrt 2 \phi'_{L\atop R}}\sim {1\over \sqrt 2} \, (\eta_{L\atop R} 

1838: \pm i

1839: \psi_{L\atop R})\,

1840: ,

1841:  \ee

1842:  \ben \label{efh11.5}

1843: && \psi_R \xi_R = \sqrt 2 \p\phi_R, \quad \eta_R \xi_R = \sqrt 2

1844: \p X_R,

1845: \quad \psi_R \eta_R = \sqrt 2 \p \phi'_R, \nonumber \\

1846: && \psi_L \xi_L = \sqrt 2 \bar\p\phi_L, \quad \eta_L \xi_L = \sqrt

1847: 2 \bar\p X_L, \quad \psi_L \eta_L = \sqrt 2 \bar \p \phi'_L,

1848:  \een

1849: where $\xi_L$, $\eta_L$ are left-chiral Majorana fermions,

1850: $\xi_R$, $\eta_R$ are right-chiral Majorana fermions,

1851: $\phi=\phi_L+\phi_R$, $\phi'=\phi'_L+\phi'_R$ are free bosons and

1852: $\sim$ in \refb{efh10}, \refb{efh11} denotes equality up to

1853: cocycle factors.\footnote{For a discussion of the cocycle

1854: factors in this case, see ref.\cite{0003124}.} $\xi$, $\eta$, $\phi$, 

1855: $\phi'$ are

1856: normalized so that

1857:  \ben \label{exinorm}

1858: && \xi_R(z) \xi_R(w) \simeq {1\over z-w} \simeq \eta_R(z)\eta_R(w),

1859: \qquad \xi_L(\bar z) \xi_L(\bar w) \simeq {1\over \bar z-\bar w}

1860: \simeq \eta_L(\bar z)\eta_L(\bar w)\, , \nonumber \\

1861: && \p\phi_R(z) \p\phi_R(w) \simeq - {1\over 2(z-w)^2}, \qquad

1862: \bar\p\phi_L(\bar z) \bar\p\phi_L(\bar w) \simeq - {1\over 2(\bar 

1863: z-\bar w)^2}, \nonumber \\

1864: && \p\phi'_R(z) \p\phi'_R(w) \simeq - {1\over 2(z-w)^2}, \qquad

1865: \bar\p\phi'_L(\bar z) \bar\p\phi'_L(\bar w) \simeq - {1\over 2(\bar

1866: z-\bar w)^2} \, .

1867:  \een

1868: Eq.\refb{efh10} defines $\xi_{L,R}$ and $\eta_{L,R}$ in terms of

1869: $X_{L,R}$, and eq.\refb{efh11} defines $\phi_{L,R}$ and

1870: $\phi'_{L,R}$ in terms of $\xi_{L,R}$, $\eta_{L,R}$ and

1871: $\psi_{L,R}$. \refb{efh11.5} follows from \refb{efh10},

1872: \refb{efh11} and the usual rules for bosonization. The boundary

1873: conditions on various fields, following from Neumann boundary

1874: condition on $X$ and $\psi$, and eqs.\refb{efh10} -

1875: \refb{efh11.5}, are:

1876:  \be \label{efh12}

1877: X_L=X_R, \quad \psi_L=\psi_R, \quad \xi_L=\xi_R, \quad \eta_L=\eta_R,

1878: \quad \phi_L=\phi_R, \quad \phi'_L=\phi'_R\, .

1879:  \ee

1880: Using eqs.\refb{efh9}-\refb{efh12} the boundary perturbation

1881: \refb{efh8} takes the form:

1882:  \be \label{efh13}

1883: i\, \sqrt 2 \, {\al } \,  \int \, dt \, \bar\p\phi_L(t) \otimes

1884: \sigma_1\, .

1885:  \ee

1886: This corresponds to switching on a Wilson line along the $\phi$ direction 

1887: and is clearly an exactly marginal deformation. Thus for any

1888: value of the coefficient $\al$ in \refb{efh13} (and hence in \refb{efh8})

1889: we get a boundary

1890: CFT.

1891: 

1892: \item In order to create a kink solution in the non-compact

1893: theory. we need to take the radius $R$ back to infinity after

1894: switching on the deformation \refb{efh8}. However here we

1895: encounter an obstruction; for generic $\alpha$ the boundary vertex 

1896: operator \refb{efh8}

1897: develops a one point function on the upper half

1898: plane\cite{9808141} for $R\ne 1/\sqrt 2$:

1899:  \be \label{efh14}

1900: \la\psi\cos{X\over 2R}(0) \otimes \sigma_1\ra_{UHP} \propto (R-1)

1901: \sin(2\pi\al)\, .

1902:  \ee

1903: This is a reflection of the fact that for $R\ne 1/\sqrt 2$ the operator

1904: $\psi\cos(X/2R)$ is no longer marginal. This shows that in order

1905: to take the radius back to $\infty$, $\al$ must be fixed at 0 or

1906: $1/2$. $\alpha=0$ gives us back the original D-brane system,

1907: whereas $\alpha=1/2$ gives the kink solution.

1908: \end{enumerate}

1909: 

1910: To summarize, in order to construct the tachyon kink solution on a 

1911: D9-$\bd9$ brane pair we first compactify the $x$ direction on a circle of 

1912: radius $1/\sqrt 2$ and swich on half unit of Wilson line along $x$ on one 

1913: of the branes, then switch on the deformation \refb{efh8} and take 

1914: $\alpha$ to 1/2, and finally take $R$ back to infinity. This gives the 

1915: construction of the tachyonic kink configuration 

1916: on a D9-$\bar{\rm D}9$-brane pair as a BCFT. It now remains to show

1917: that this BCFT actually describes a non-BPS D8-brane, {\it i.e.}

1918: the effect of this deformation is to change the Neumann boundary

1919: condition on $X$ and $\psi_L$, $\psi_R$ to Dirichlet boundary

1920: condition. This is done by noting that \refb{efh13} is the contour

1921: integral of an anti-holomorphic current, and hence the effect of this on

1922: a correlation function can be studied by deforming the contour and

1923: picking up residues from various operators. Using eqs.\refb{efh9} - 

1924: \refb{efh11} it is easy to see that for

1925: $\alpha=1/2$, the exponential of \refb{efh13} transforms the left-moving 

1926: fermions $\xi_L$,

1927: $\eta_L$ and $\psi_L$ as:

1928:  \be \label{efh16}

1929: \xi_L\to-\xi_L, \qquad \eta_L\to \eta_L, \qquad \psi_L\to -\psi_L\, .

1930:  \ee

1931: In terms of the original fields, this induces a transformation

1932: $\bar \p X_L\to   - \bar\p X_L$, $\psi_L\to -\psi_L$ on every vertex 

1933: operator inserted

1934: in the interior of the world-sheet. The right-moving world-sheet

1935: fields are not affected. By a redefinition $X_L\to -X_L$, $\psi_L\to 

1936: -\psi_L$, we can leave the vertex

1937: operators unchanged, but this changes the boundary condition on $X$ and

1938: $\psi$ from Neumann to Dirichlet:

1939:  \be \label{efh17}

1940: \bar\p X_L=-\p X_R, \qquad \psi_L=-\psi_R\, .

1941:  \ee

1942: This clearly shows that switching on the deformation \refb{efh13} with

1943: $\alpha={1\over 2}$

1944: corresponds to creating a D8-brane transverse to the circle of radius 

1945: $R=1/\sqrt 2$. Taking the

1946: radius back to infinity leaves the D8-brane unchanged.

1947: 

1948: 

1949: The analysis can be generalized in many different ways. First of all,

1950: since the

1951: boundary condition on $X^0,\ldots X^8$ played no role in the analysis, we

1952: can choose them to be anything that we like. This establishes in general

1953: that a kink solution on a D$p$-$\bar{\rm D}p$-brane pair represents a

1954: non-BPS D-$(p-1)$-brane.

1955: The analysis showing that a kink solution on a non-BPS D-$p$-brane

1956: represents a BPS D-$(p-1)$-brane is essentially identical. We begin with

1957: the kink solution on the BPS D$p$-$\bd p$-brane pair in type IIB/IIA 

1958: theory

1959: for $p$ odd/even and mod

1960: this out by $(-1)^{F_L}$. This converts type IIB/IIA theory to type

1961: IIA/IIB

1962: theory, the

1963: D$p$-$\bd p$-brane pair to a non-BPS

1964: D-$p$-brane, and the non-BPS D-$(p-1)$-brane to a BPS

1965: D-$(p-1)$-brane.\footnote{One can show\cite{9810188,9812031} that the 

1966: tachyon state as well as all other open string states on a non-BPS

1967: D-brane carrying Chan-Paton factor $\sigma_1$ are odd under $(-1)^{F_L}$. 

1968: Thus these modes are projected out after

1969: modding out the theory by $(-1)^{F_L}$, and we get a tachyon free BPS

1970: D-brane. \label{fn13}} This shows that the kink solution on a non-BPS 

1971: D-$p$-brane

1972: can be identified as a BPS D-$(p-1)$-brane.

1973: Finally the analysis showing that a

1974: codimension $k$

1975: soliton on a D$p$-$\bar{\rm D}p$-brane pair or a non-BPS D$p$ brane

1976: produces a D-$(p-k)$-brane can be done by compactifying $k$ of the

1977: coordinates tangential to the original D-brane on a torus $T^k$ of

1978: appropriate radii, switching on a marginal deformation that creates the

1979: codimension $k$ soliton, and finally proving that the BCFT obtained at the

1980: end of this marginal deformation is a D-$(p-k)$-brane\cite{0003124}.

1981: 

1982: 

1983: 

1984: \subsection{Analysis of the boundary state} \label{s3.4}

1985: 

1986: Given a boundary CFT describing a D-brane system, we can associate

1987: with it a boundary state

1988: $|\BB\ra$\cite{callan1,callan2,9604091,9707068,9912275}. This is a {\it

1989: closed string state} of ghost number 3, and has the following

1990: property. Given any closed string state $|V\ra$ and the associated

1991: vertex operator $V$, the BPZ inner product $\la\BB|V\ra$ is given

1992: by the one point function of $V$ inserted at the centre of a unit

1993: disk $D$, the boundary condition on $\p D$ being the one

1994: associated with the particular boundary CFT under consideration:

1995:  \be \label{efg1}

1996: \la\BB|V\ra \propto \la V(0)\ra_D\, .

1997:  \ee

1998: Note that in order to find the contribution to the boundary state at

1999: oscillator level $(N,N)$, we only need to compute the inner product of the

2000: boundary state with closed string states of level $(N,N)$. This in turn

2001: requires computation of one point function on the disk of closed string

2002: vertex operators of level $(N,N)$.

2003: 

2004: {}From the definition \refb{efg1} it is clear that the boundary state

2005: contains

2006: information about what kind of source for the closed string states is

2007: produced by the D-brane system under consideration. This has been made

2008: more

2009: precise in appendix \ref{appb}. There it has been shown that if the

2010: boundary state $|\BB\ra$ associated with a D-brane in bosonic string 

2011: theory has an expansion of the form:

2012:  \be \label{efg5.5}

2013: |\BB\ra =

2014: \int {d^{26} k\over (2\pi)^{26}} [\wt F(k) +  (\wt A_{\mu\nu}(k)+ \wt

2015: C_{\mu\nu}(k))

2016: \alpha^\mu_{-1}

2017: \bar\alpha^\nu_{-1}

2018: + \wt B(k) (b_{-1} \bar c_{-1}  + \bar b_{-1} c_{-1}) + \ldots ] (c_0

2019: +\bar

2020: c_0)

2021: c_1

2022: \bar c_1 |k\ra\, ,

2023:  \ee

2024: where $\wt F$, $\wt A_{\mu\nu}=\wt A_{\nu\mu}$, $\wt

2025: C_{\mu\nu}=-\wt C_{\nu\mu}$, $\wt B$ etc. are fixed functions,

2026: $\alpha^\mu_{-n}$, $\bar\alpha^\mu_{-n}$ are oscillators of

2027: $X^\mu$, and $b_{-n}$, $c_{-n}$, $\bar b_{-n}$, $\bar c_{-n}$ are

2028: ghost oscillators, then the energy momentum tensor $T_{\mu\nu}(x)$, 

2029: defined as the source for the graviton field, is given by

2030:  \be \label{efg9a}

2031: T_{\mu\nu}(x) \propto (A_{\mu\nu}(x) + \eta_{\mu\nu} B(x) )\, .

2032:  \ee

2033: We shall choose the normalization of $|\BB\ra$ in such a way that the

2034: above equation takes the form:

2035:  \be \label{efg9}

2036: T_{\mu\nu}(x) = {1\over 2} \, (A_{\mu\nu}(x) + \eta_{\mu\nu} B(x) )\, .

2037:  \ee

2038: 

2039: 

2040: Let us apply this to the specific D-brane system studied in

2041: section \ref{s3.1}, namely D-$p$-brane wrapped on a circle of

2042: radius 1. We take the directions transverse to the brane to be

2043: $x^1,\ldots x^{25-p}$, and the spatial directions along the brane

2044: to be $x^{26-p},\ldots x^{25}$. The boundary state of the initial

2045: D-$p$-brane without any perturbation is given

2046: by\cite{callan1,callan2,9604091,9707068,9912275}:

2047:  \be \label{efg10}

2048: |\BB\ra = \TT_p \,  |\BB\ra_{c=1} \otimes |\BB\ra_{c=25} \otimes

2049: |\BB\ra_{ghost}\, ,

2050:  \ee

2051: where $\TT_p$ is the tension of the D$p$-brane as given in \refb{e2.16},

2052: $|\BB\ra_{c=1}$ denotes the boundary state associated with the

2053: $X^{25}\equiv X$ direction, $|\BB\ra_{c=25}$ denotes the boundary state

2054: associated with the other 25 directions $X^0,\ldots X^{24}$, and

2055: $|\BB\ra_{ghost}$ denotes the

2056: boundary state associated with the ghost direction. We have:

2057:  \be \label{efg11}

2058: |\BB\ra_{c=25} =

2059: \int {d^{25-p} k_\perp\over (2\pi)^{25-p}} \,

2060: \exp\left(\sum_{\mu,\nu=0}^{25}\sum_{n=1}^\infty

2061: {1\over

2062: n} \eta_{\mu\nu} \, (-1)^{d_\mu} \, \alpha^\mu_{-n} \bar

2063: \alpha^\nu_{-n} \right)

2064: |k_\parallel=0,

2065: k_\perp\ra\, ,

2066:  \ee

2067:  \be \label{efg12}

2068: |\BB\ra_{ghost} = \exp\left(-\sum_{n=1}^\infty (\bar b_{-n} c_{-n} +

2069: b_{-n}

2070: \bar c_{-n})

2071: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\, ,

2072:  \ee

2073: and,

2074:  \be \label{efg13}

2075: |\BB\ra_{c=1} = \sum_{m=-\infty}^\infty \exp\left(-\sum_{n=1}^\infty

2076: {1\over

2077: n} \alpha_{-n} \bar

2078: \alpha_{-n} \right) |k=0, w=m\ra\, ,

2079:  \ee

2080: where $d_\mu=1$ for Neumann directions and 0 for Dirichlet directions,

2081: $k_\parallel$ denotes momentum

2082: along the D-$p$-brane in directions other than $x^{25}$, $k_\perp$ denotes

2083: momentum transverse to the D-$p$-brane,  $\alpha_n$, $\bar \alpha_n$

2084: without any superscript denote the

2085: $X^{25}\equiv X$ oscillators, and $k$ and $w$ denote the momentum and

2086: winding number respectively

2087: along the

2088: circle along $x^{25}$. Let us denote by $x^M$ the coordinates other

2089: than $x^{25}\equiv x$ along the D-$p$-brane world-volume, and by $x^m$

2090: the

2091: coordinates transverse to the D-$p$-brane world-volume. We shall choose

2092: $M$ to run over the values 0 and

2093: $(26-p),\ldots, 24$ and $m$ to run over the values $1,\ldots, (25-p)$.

2094: Expanding the boundary state in powers of the various oscillators, and

2095: comparing this expansion with \refb{efg5.5}, we get the following

2096: non-zero components of $A_{\mu\nu}$ and $B$:

2097:  \be \label{efg14}

2098: A_{xx} = - \TT_p \, \delta(x_\perp) \, , \quad A_{MN} = -

2099: \TT_p \, \eta_{MN} \, \delta(x_\perp) \, , \quad A_{mn}

2100: = \TT_p \,

2101: \delta_{mn} \, \delta(x_\perp) \, ,

2102: \quad B = - \TT_p \, \delta(x_\perp)\, .

2103:  \ee

2104: and hence

2105:  \be \label{efg15}

2106: T_{xx} = - \TT_p \, \delta(x_\perp) \, , \qquad T_{MN} = - \TT_p \,

2107: \eta_{MN} \, \delta(x_\perp)\, , \qquad

2108: T_{mn} = 0.

2109:  \ee

2110: This is the energy-momentum tensor associated with a D-25-brane.

2111: 

2112: We shall now study the change in the boundary state under the deformation

2113: of the boundary CFT by the marginal operator 

2114: \refb{e3.2}\cite{9402113,9811237}. Using the

2115: boundary condition \refb{eb1} we can rewrite \refb{e3.2} as

2116:  \be \label{egf16}

2117: - \al \int dt \cos (2X_L(t)) \, dt = - \al \int dt J^1_L(t)\, ,

2118:  \ee

2119: Adding such a perturbation at the boundary effectively rotates the

2120: left-moving world-sheet component of the boundary state by an angle

2121: $2\pi\al$ about the 1-axis. In particular for $\al={1\over 2}$, the 

2122: effect of

2123: this perturbation  is a rotation by $\pi$ about the 1-axis which changes

2124: $J^3_L$ to $-J^3_L$, {\it i.e.} $\alpha_n$ to $-\alpha_n$ in the exponent

2125: of \refb{efg13}, and also

2126: converts the winding number $w=m$ to momentum $k=m$ along $x^{25}$.

2127: Thus $|\BB\ra_{c=1}$ is transformed to:

2128:  \be \label{etras}

2129: \sum_{m=-\infty}^\infty \exp\left( \sum_{n=1}^\infty {1\over n} \al_{-n}

2130: \bar\al_{-n}\right) |k=m, w=0\ra\, .

2131:  \ee

2132: The other components of $|\BB\ra$ given in \refb{efg11}, \refb{efg12}

2133: remain

2134: unchanged.

2135: The

2136: result

2137: is precisely the boundary state associated with a D-$(p-1)$-brane, and the

2138: corresponding $T_{\mu\nu}$ computed using \refb{efg5.5}, \refb{efg9}

2139: precisely reproduces the energy-momentum tensor of a D-$(p-1)$-brane

2140: situated

2141: at $x=0$. We shall shortly derive this as a special case of a more general

2142: result.

2143: 

2144: For a general $\al$ the effect of this rotation on the boundary state is

2145: somewhat complicated but can nevertheless be done by expressing the

2146: boundary state in a suitable basis\cite{9811237}. This analysis

2147: gives\cite{9402113,9811237,0203211}

2148:  \ben \label{efg16.5}

2149: |\BB\ra_{c=1} &=& |k=0\ra + \sum_{n=1}^\infty \sin^n(\pi\al) \Big(

2150: |k=n\ra + |k=-n\ra\Big) - \cos(2\pi\al) \alpha_{-1}

2151: \bar\alpha_{-1} |k=0\ra

2152: \nonumber \\

2153: && + \sum_{n=1}^\infty \sin^n(\pi\al) \alpha_{-1}

2154: \bar\alpha_{-1}\Big( |k=n\ra + |k=-n\ra\Big) + \ldots \, ,

2155:  \een

2156: where $\ldots$ denote terms with oscillator level higher than

2157: (1,1) and terms with winding modes. A more detailed discussion of

2158: the higher level zero winding number terms will be given in

2159: section \ref{s10} (see eq.\refb{ebs1} and discussion below this equation). 

2160: Combining \refb{efg16.5} with

2161: \refb{efg11}, \refb{efg12}, and using eqs.\refb{efg5.5},

2162: \refb{efg10}, we get

2163:  \ben \label{efg18}

2164: && \wt B(k) = (2\pi)^{p+1} \, \TT_p \, \left[-\delta(k) -

2165: \sum_{n=1}^\infty \Big(

2166: \delta(k-n) +

2167: \delta(k+n)

2168: \Big)  sin^n(\al\pi)\right] \, \delta(k_\parallel) \, , \nonumber \\

2169: && \wt A_{xx}(k) = \left[-(2\pi)^{p+1} \, \TT_p \, \Big(1 +

2170: \cos(2\pi\al)\Big) \delta(k) \, \delta(k_\parallel) -

2171: \wt B(k)

2172: \right] \, ,

2173: \nonumber \\

2174: && \wt A_{xM} = 0, \qquad \wt A_{MN} = \wt B(k) \, \eta_{MN}\, ,

2175: \quad \wt A_{xm} = 0, \quad \wt A_{mn} =-\wt B(k)\, \delta_{mn},

2176: \quad \wt A_{mM}=0 \, . \nonumber \\

2177:  \een

2178: The Fourier transform of these equations give:

2179:  \ben \label{efg20}

2180: && B = -\TT_p\, f(x)\, \delta(x_\perp)\, , \quad A_{xx} = -\TT_p\,

2181: g(x) \, \delta(x_\perp), \quad A_{MN} =

2182: - \TT_p\, f(x) \, \eta_{MN}\, \delta(x_\perp), \nonumber \\

2183: &&  A_{xM} = 0, \quad A_{xm} = 0, \quad A_{mn} = \TT_p\, f(x) \,

2184: \delta_{mn}\, \delta(x_\perp)\,  , \quad A_{mM} = 0,

2185:  \een

2186: where

2187:  \ben \label{efg21}

2188: f(x) &=& 1 + \sum_{n=1}^\infty \, \sin^n(\al\pi) \, (e^{inx} +

2189: e^{-inx})

2190: = {1\over 1 - e^{ix} \sin(\al\pi)}

2191: + {1\over 1 - e^{-ix} \sin(\al\pi)} - 1\, , \nonumber \\ g(x) &=& \Big(1 +

2192: \cos(2\pi\al)\Big) - f(x)\, .

2193:  \een

2194: In arriving at the right hand side of eqs.\refb{efg21} we have performed

2195: the sum over $n$, using the fact that it is a convergent sum for

2196: $|\sin(\pi\al)| < 1$. Using eq.\refb{efg9} the energy momentum

2197: tensor

2198: $T_{\mu\nu}$ is now given by:

2199:  \ben \label{efg22}

2200: && T_{xx} = -  \TT_{p} \cos^2(\pi\al) \, \delta(x_\perp), \quad

2201: T_{MN} = -\TT_{p}

2202: \,    f(x) \,

2203: \eta_{MN} \,\delta(x_\perp) \nonumber \\

2204: && T_{xM} = 0, \quad T_{xm} = 0, \quad T_{mM}=0, \quad T_{mn}=0\,

2205: \nonumber \\

2206: && \qquad \qquad \hbox{for} \quad

2207: M,N =0, \, (26-p), \ldots

2208: 24, \quad m,n=1,\ldots (25-p)\, .

2209:  \een

2210: This gives the energy momentum tensor associated with the boundary CFT of

2211: section \ref{s3.1} for arbitrary value of $\al$. Note that $T_{xx}$ is $x$ 

2212: independent. This is a consequence of the conservation law $\p_x T_{xx} + 

2213: \eta^{MN}\p_M T_{Nx}=0$.

2214: 

2215: Using \refb{edchargeapp}

2216: we can also see that the function

2217: $\wt B(k)$ measures the source of the dilaton field $\wt\phi(k)$. This

2218: suggests that we define the dilaton charge density to be

2219:  \be \label{edcharge}

2220: Q(x) = -  B(x) = \TT_{p} \, f(x)\, \delta(x_\perp)\, .

2221:  \ee

2222: The overall normalization of $Q$ is a matter of convention.

2223: 

2224: An interesting limit is the $\al\to {1\over 2}$ limit. For $x\ne

2225: 2n\pi$ with integer $n$, both $f(x)$ and $g(x)$ can be seen to

2226: vanish in this limit. On the other hand for any $\al< {1\over 2}$,

2227: we can compute $\int_{-\pi}^{\pi} f(x) dx$ by a contour integral,

2228: and the answer turns out to be $2\pi$. Thus we would conclude that

2229: as $\al\to {1\over 2}$, $f(x)$ approaches a delta function

2230: concentrated at $x=0$ (and hence also at $2n\pi$). Hence in this

2231: limit,

2232:  \be \label{efg23}

2233: T_{xx} = T_{xM} = T_{xm}=T_{mn}=T_{mM}= 0, \quad T_{MN} = -2\,

2234: \pi\, \TT_{p}\, \delta(x_\perp) \, \, \eta_{MN}

2235: \sum_{n=-\infty}^\infty \delta(x-2n\pi)  \, .

2236:  \ee

2237: This is precisely the energy momentum tensor of a D-$(p-1)$-brane

2238: situated on a circle at $x=0$, since $2\pi \, \TT_p$ is the

2239: D-$(p-1)$-brane tension $\TT_{p-1}$.

2240: 

2241: Following the same logic as the one given for the $\alpha\to{1\over

2242: 2}$ limit, we can see

2243: that for $\al\to -{1\over 2}$ we again get a

2244: D-$(p-1)$-brane situated on a circle, but this time at $x=\pi$ instead of 

2245: at

2246: $x=0$. Thus \refb{efg23} is now replaced by:

2247:  \be \label{ealphamh}

2248: T_{xx} = T_{xM} = T_{xm}=T_{mn}=T_{mM}= 0, \quad T_{MN} = -2\,

2249: \pi\, \TT_{p}\, \delta(x_\perp)  \, \eta_{MN} \,

2250: \sum_{n=-\infty}^\infty \delta\left(x-(2n+1)\pi\right) \, .

2251:  \ee

2252: 

2253: 

2254: 

2255: 

2256: A similar analysis can be carried out for the superstring theory as 

2257: well\cite{9903123}.

2258: Instead of going through the details of the analysis, we quote here the

2259: final answer\cite{0203265}. For the deformed boundary CFT described in

2260: section

2261: \ref{s3.3}, the energy-momentum tensor is given by:

2262:  \ben \label{efg24}

2263: && T_{xx} = -  \EE_p\, 

2264: \cos^2(\pi\al) \, \delta(x_\perp), \quad T_{MN} = -\EE_{p}

2265: \,    f(x) \, \delta(x_\perp) \, \eta_{MN}, \nonumber \\ &&

2266: T_{xM} =

2267: T_{xm} =

2268: T_{mM} =

2269: T_{mn} =0,

2270:  \een

2271: where $x^m$ denote directions transverse to the D-brane, $x^M$

2272: denote directions (other than $x$) tangential to the D-brane, $\EE_p$

2273: denotes the tension of the original brane

2274: system ($\wt \TT_p$ for the non-BPS D$p$-brane and $2\TT_p$ for the

2275: brane-antibrane system), and

2276:  \be \label{efg25}

2277: f(x) = {1\over 1 - e^{i\sqrt 2 x} \sin^2(\al\pi)}

2278: + {1\over 1 - e^{-i\sqrt 2 x} \sin^2(\al\pi)} - 1\, .

2279:  \ee

2280: The dilaton charge density is given by:

2281:  \be \label{esupdil}

2282: Q(x) = \EE_p \, f(x) \, \delta(x_\perp)\, .

2283:  \ee

2284: As $\al\to 0$, we get back the energy-momentum tensor of the original

2285: D9-brane system. On the other hand, as $\al\to {1\over 2}$, $f(x)$

2286: approaches sum of delta functions concentrated at $x=2\pi n/ \sqrt 2$ for 

2287: integer $n$. Since the $x$ coordinate is compactified on a circle of 

2288: radius $1/\sqrt 2$, the

2289: resulting $T_{\mu\nu}$ reduces to that of a D-$(p-1)$-brane situated at

2290: $x=0$.

2291: 

2292: Finally, a kink solution on a non-BPS D-$p$-brane (but not on a

2293: brane-antibrane pair) also produces a source for the Ramond-Ramond

2294: $p$-form gauge field, given by\cite{0204143}:

2295:  \be \label{erreuclid}

2296: Q^{(p)}_{M_1\ldots M_p} \propto \epsilon_{M_1\ldots M_p} \,

2297: \sin(\alpha\pi) \, \left[ {e^{ix/\sqrt{2}} \over 1 -

2298: \sin^2(\alpha\pi) e^{\sqrt{2} ix}} + {e^{-ix/\sqrt{2}} \over 1 -

2299: \sin^2(\alpha\pi) e^{-\sqrt{2} ix}} \right]\, \delta(x_\perp)\, .

2300:  \ee

2301: This result can be derived from the Ramond-Ramond component of the

2302: boundary state. In the $\alpha\to {1\over 2}$ limit \refb{erreuclid} 

2303: correctly 

2304: reproduces the RR charge of the D-$(p-1)$-brane located at $x=2\pi n/\sqrt 

2305: 2$.

2306: 

2307: Discussion on various other aspects of conformal field theory methods 

2308: reviewed in this section can be found in 

2309: refs.\cite{9902160,9903123,9903139,0001066,

2310: 0002061,0003110,

2311: 0005114,

2312: 0006122,0009090,

2313: 0009252,0101211,0102192,

2314: 0401003}.

2315: 

2316: \sectiono{Open String Field Theory} \label{s4}

2317: 

2318: Although the results on tachyon dynamics on a D-brane were stated in

2319: section \ref{s2} in terms of the effective action obtained by formally

2320: integrating out the heavy fields, in general it is difficult to do this in

2321: practice. Conformal field theory methods described in the last section

2322: provide an indirect way of constructing solutions of the classical

2323: equations of motion without knowing the effective action. But if we want a

2324: more direct construction of the classical solutions, we need to

2325: explicitly take into account the coupling of the

2326: tachyon to infinite number of other fields associated with massive open

2327: string states.

2328: The formalism

2329: that allows us to tackle this problem head on  is string field theory, --

2330: a field theory with

2331: infinite number of fields. 

2332: This will be the topic of discussion of the present section.

2333: We begin by reviewing the formulation of first

2334: quantized open bosonic string theory, followed by a review of bosonic open

2335: string field theory. We then show that this string field theory can be

2336: used to test the various conjectures about the tachyon effective field

2337: theory. At the end we briefly discuss the case of superstring field

2338: theory. An excellent and much more detailed review of string field theory 

2339: with

2340: application to the

2341: problem of tachyon condensation can be found in \cite{0311017}.

2342: 

2343: \subsection{First quantized open bosonic string theory} \label{s4.0}

2344: 

2345: For a given space-time background which is a solution of the classical

2346: equations of motion in string theory, we have a two dimensional CFT. This

2347: CFT is a direct sum of two CFT's, the matter CFT of central charge 26, and

2348: the ghost CFT of central charge $-26$. The matter CFT depends on the

2349: choice of the space-time background, but the ghost CFT is universal and is

2350: described by anti-commuting fields $b$, $\bar b$, $c$, $\bar c$ of

2351: conformal dimensions (2,0), (0,2), (1,0) and (0,1) respectively.

2352: Physically this conformal field theory describes the propagation of

2353: closed string in this space-time background.  A

2354: D-brane in this space-time background is in one to one correspondence to a

2355: two dimensional conformal field theory on the upper half plane (or unit

2356: disk) with specific conformally invariant boundary condition on the real

2357: axis (unit circle).  This conformal field theory describes the

2358: propagation of an open string living on the D-brane in this

2359: space-time background. The boundary conditions on the matter fields

2360: depend on

2361: the specific D-brane that we are considering, but those on the ghost

2362: fields

2363: are universal, and take the form:

2364:  \be \label{e4.1}

2365: b=\bar b, \qquad c =

2366: \bar c\, ,

2367:  \ee

2368: on the real axis. This gives rise to the mode expansion:

2369:  \be \label{e4.2}

2370: b = \sum_n b_n z^{-n-2}, \qquad c = \sum_n c_n

2371: z^{-n+1}, \qquad \bar b = \sum_n b_n \bar z^{-n-2}, \qquad \bar c =

2372: \sum_n c_n \bar z^{-n+1}\, ,

2373:  \ee

2374: where $z$ denotes the complex coordinate labelling the upper

2375: half-plane. We shall denote by $t$ the coordinate

2376: labelling the real axis.

2377: The SL(2,R) invariant vacuum $|0\ra$ of

2378: the open string state space satisfies:

2379:  \be \label{e4.3}

2380: b_n|0\ra =0

2381: \quad \hbox{for} \quad n\ge -1, \qquad c_n|0\ra =0 \quad \hbox{for} \quad

2382: n\ge 2\, .

2383:  \ee

2384: 

2385: 

2386: Let us denote by $\HH$ the vector space of states in the combined

2387: matter-ghost BCFT,

2388: obtained by acting on $|0\ra$ the ghost oscillators $b_{-n}$ ($n\ge 2$),

2389: $c_{-n}$

2390: ($n\ge -1$) and the matter vertex operators. By the usual state-operator

2391: correspondence

2392: in BCFT, for every state $|\phi\ra$ in $\HH$, there is a unique local

2393: {\it boundary}

2394: vertex operator $\phi(t)$ such that

2395:  \be \label{e4.4}

2396: \phi(0)|0\ra = |\phi\ra\, .

2397:  \ee

2398: The states in $\HH$ can be classified by their ghost numbers, defined

2399: through the

2400: following rules:

2401:  \begin{enumerate}

2402: \item $b$, $\bar b$ have ghost number $-1$.

2403: \item $c$, $\bar c$ have ghost number 1.

2404: \item All matter operators have ghost number 0.

2405: \item The SL(2,R) invariant vacuum $|0\ra$ has ghost number 0.

2406: \end{enumerate}

2407: We define by $\HH_n$ the subspace of $\HH$ containing states of ghost number

2408: $n$.

2409: 

2410: Given a pair of states $|A\ra,|B\ra\in \HH$, and

2411: the associated vertex operators $A(t)$ and $B(t)$, we define the BPZ

2412: inner product

2413: between the states as:

2414:  \be \label{e4.12}

2415: \la A | B\ra = \la I\circ A(0) B(0)\ra_{UHP}\, ,

2416:  \ee

2417: where $\la \cdot\ra_{UHP}$ denotes the correlation function of the BCFT

2418: on the upper

2419: half plane, $f\circ A(t)$ for any function $f(z)$ denotes the

2420: conformal

2421: transform of

2422: $A(t)$ under the map $f$, and

2423:  \be \label{e4.13}

2424: I(z) = - 1/z\, .

2425:  \ee

2426: Thus for example if $A(t)$ is a primary operator of weight $h$ then 

2427: $f\circ A(t) = \left( f'(t)\right)^h A(f(t))$.

2428: It is a well known property of the correlation function of the matter

2429: ghost

2430: BCFT on

2431: the upper half plane that the correlator is non-zero only if the total

2432: ghost number

2433: of all the operators add up to three. Thus $\la A|B\ra$ is non-zero only

2434: if the

2435: ghost

2436: numbers of $|A\ra$ and $|B\ra$ add up to three.

2437: 

2438:  \begin{figure}[!ht]

2439:  \begin{center}

2440: \leavevmode

2441: \epsfysize=5cm

2442: \epsfbox{fig10.eps}

2443: \end{center}

2444:  \caption{Images of the upper half unit disk under the maps $g_1$ and $g_2$.

2445: The solid line gets mapped to the solid line and the dashed line

2446: gets mapped to the dashed line.} \label{f7}

2447:  \end{figure}

2448: By a conformal transformation $w=(1+iz) / (1-iz)$ that takes the upper

2449: half plane to the unit

2450: disk, we

2451: can reexpress \refb{e4.12} as

2452:  \be \label{e4.14}

2453: \la A | B\ra = \la g_2\circ A(0) g_1\circ B(0)\ra_{D}\, ,

2454:  \ee

2455: where $\la \cdot \ra_D$ denotes correlation function on a unit disk, and,

2456:  \be \label{e4.15}

2457: g_1(z) = {1+iz \over 1-iz}, \qquad g_2(z) = - g_1(z)\, .

2458:  \ee

2459: An intuitive understanding of the maps $g_1$ and $g_2$ may be obtained by

2460: looking at

2461: the image of the upper half unit disk under these maps. This has been

2462: shown in

2463: Fig.\ref{f7}.

2464: 

2465: Physical open string states on the D-brane are states in

2466: $\HH_1$

2467: satisfying the following criteria:

2468:  \begin{enumerate}

2469: \item The state $|\phi\ra$ must be BRST invariant:

2470:  \be \label{e4.5}

2471: Q_B|\phi\ra = 0\, ,

2472:  \ee

2473: where

2474:  \be \label{e4.6}

2475: Q_B = {1\over 2\pi i} \left[ \ointop c(z) T_m(z) dz + \ointop b(z) c\p

2476: c(z) dz\right]\, ,

2477:  \ee

2478: is the BRST charge carrying ghost number 1 and  satisfying

2479:  \be \label{e4.7}

2480: (Q_B)^2=0\, .

2481:  \ee

2482: Here $T_m(z)=\sum_n L^{(m)}_n z^{-n-2}$ stand for the $zz$ component of

2483: the world-sheet stress

2484: tensor of the matter part of

2485: the BCFT

2486: describing the D-brane. $\ointop$ denotes a contour around the origin.

2487: \item Two states $|\phi\ra$ and $|\phi'\ra$ are considered equivalent

2488: if they differ by a state of the form $Q_B|\Lambda\ra$:

2489:  \be \label{e4.8}

2490: |\phi\ra \equiv |\phi\ra + Q_B|\Lambda\ra\, ,

2491:  \ee

2492: for any state $|\Lambda\ra\in \HH_0$.

2493: \end{enumerate}

2494: Thus physical states are in one to one correspondence with the elements of

2495: BRST

2496: cohomology in $\HH_1$.

2497: 

2498: In the first quntized formulation there is also a well defined

2499: prescription, known

2500: as Polyakov prescription, for computing tree and loop amplitudes

2501: involving physical

2502: open string states as external lines.

2503: 

2504: \subsection{Formulation of open bosonic string field theory} \label{s4.1}

2505: 

2506: The open string field theory describing the dynamics of a D-brane is, by

2507: definition,

2508: a field theory satisfying the following two criteria:

2509:  \begin{enumerate}

2510: \item Gauge inequivalent solutions of the linearized equations of

2511: motion are in

2512: one to one

2513: correspondence with the physical states of the open string.

2514: \item The S-matrix computed using the Feynman rules of the string field

2515: theory

2516: reproduces the S-matrix computed using Polyalov prescription to all

2517: orders in the

2518: perturbation theory.

2519: \end{enumerate}

2520: If we are interested in studying only the classical properties of string

2521: field

2522: theory, we can relax the second constraint a bit by requiring that the

2523: S-matrix

2524: elements agree only at the tree level. Nevertheless for the open bosonic

2525: string

2526: field theory that we shall be describing \cite{osft}, the

2527: agreement has

2528: been

2529: verified to

2530: all

2531: orders in perturbation theory \cite{GIDDMARW}.

2532: 

2533: The first step in the construction of string field theory will be to

2534: decide what

2535: corresponds to a general off-shell string field configuration. Usually

2536: when one goes

2537: from the first to second quantized formulation, the wave-functions /

2538: states of the

2539: first quantized theory become the field configurations of the second

2540: quantized

2541: theory. However, a generic off-shell field configuration in the second

2542: quantized theory

2543: does not satisfy the physical state condition.\footnote{For example

2544: second quantization of a non-relativistic Schrodinger problem

2545: describing a particle of mass $m$ moving under a potential $V$ in

2546: three

2547: dimensions is described by the action $\int \, dt \, \int \, d^3 x \,

2548: \Psi^*\left( i {\p\Psi\over \p t} +{\hbar^2\over 2m} \vec\nabla^2\Psi

2549: - V\Psi\right)$. A general off-shell field configuration $\Psi(\vec

2550: x,t)$ does not satisfy the Schrodinger equation. Rather, Schrodinger

2551: equation appears as the classical equation of motion for $\Psi$

2552: derived from this action.}  This condition comes as the linearized

2553: equation

2554: of motion of the field theory. In the same spirit we should expect that

2555: the a generic off-shell string field configuration should correspond to a

2556: state in

2557: the BCFT without the restriction of BRST invariance, and the physical state

2558: condition, {\it i.e.} the BRST invariance of the state,

2559: should emerge as the linearized equation of motion of the string field

2560: theory.

2561: This however still does not uniquely fix the space of string field

2562: configurations,

2563: since we can, for example take this to be the whole of $\HH$, or $\HH_1$,

2564: or even a

2565: subspace of $\HH_1$ that contains at least one representative from

2566: each BRST cohomology class. It turn out that the simplest form of open

2567: string field theory

2568: is obtained by taking a general off-shell string field configuration to

2569: be a state

2570: in $\HH_1$, {\it i.e.} a state $|\Phi\ra$ in $\HH$ of ghost number 1

2571: \cite{osft}.

2572: 

2573: Since we are attempting to construct a string {\it field theory}, one

2574: might wonder

2575: in what sense a state $|\Phi\ra$ in $\HH_1$ describes a field

2576: configuration. To see this

2577: we need to

2578: choose a basis of states $|\chi_{1,\alpha}\ra$ in $\HH_1$.

2579: Then we can expand $|\Phi\ra$ as

2580:  \be \label{e4.9}

2581: |\Phi\ra = \sum_\alpha \phi_\alpha |\chi_{1,\alpha}\ra\, .

2582:  \ee

2583: Specifying $|\Phi\ra$ is equivalent to specifying the coefficients

2584: $\phi_\alpha$. Thus the set of numbers $\{\phi_\alpha\}$ labels a given

2585: string field

2586: configuration.

2587: Let us for example consider

2588: the case of a

2589: D$p$-brane in (25+1) dimensional Minkowski space.

2590: We define Fock

2591: vacuum states

2592: $|k\ra$ labelled by $(p+1)$ dimensional momentum $k$ along the D$p$-brane

2593: as:

2594:  \be \label{e4.40}

2595: |k\ra = e^{ik.X(0)}|0\ra\, .

2596:  \ee

2597: A generic state carrying momentum $k$ will be created by a set of

2598: oscillators acting on the state

2599: \refb{e4.40}.

2600: Thus in this case the index

2601: $\alpha$ in \refb{e4.9}

2602: includes the $p+1$ dimensional continuous momentum index $\{k^\mu\}$

2603: along the

2604: D$p$-brane, and a discrete

2605: index $r$ which originate from various oscillators, and runs over

2606: infinite number of

2607: values. Hence we can write:

2608:  \be \label{e4.10}

2609: \{ \phi_\alpha \} \to \{ \phi_{\{k^\mu\},r} \} \equiv \{ \phi_r(k^0,

2610: \ldots

2611: k^p)\}\, , \qquad \{|\chi_{1,\alpha}\ra\} \to \{|\chi_{1,r}(k)\ra\}, 

2612: \qquad \sum_\alpha\to \sum_r \, \int\, {d^{p+1}k\over

2613: (2\pi)^{p+1}}\, .

2614:  \ee

2615: In other words the string field configuration is labelled by infinite

2616: number of

2617: functions $\{\phi_r(k^0,\ldots k^p)\}$. The Fourier transforms

2618:  \be \label{efoursft}

2619: \wt\phi_r(x^0,\ldots

2620: x^p) \equiv \int \, {d^{p+1}k\over

2621: (2\pi)^{p+1}} \, e^{ik.x} \, \phi_r(k^0,\ldots k^p)

2622:  \ee

2623: give rise

2624: to infinite number of

2625: fields in

2626: $(p+1)$ dimensions. Thus we see that the configuration space of string

2627: field theory

2628: is indeed labelled by infinite number of fields.

2629: 

2630: For later use we shall now choose some specific normalization convention

2631: for the open string states in this theory.

2632: We normalize $|k\ra$ as

2633:  \be \label{e4.41}

2634: \la k| c_{-1} c_0 c_1 | k\ra = (2\pi)^{p+1} \delta(k+k')\, .

2635:  \ee

2636: This gives

2637:  \be \label{e4.42}

2638: \la 0| c_{-1} c_0 c_1 | 0\ra = (2\pi)^{p+1} \delta(0) = V_{p+1}\, ,

2639:  \ee

2640: where $V_{p+1}$ denotes the total volume of the space-time occupied by

2641: the D-brane.

2642: In arriving at the right hand side of \refb{e4.42} we have used the usual

2643: interpretation of delta function of momentum at zero argument as the

2644: space-time volume. (This can be seen more explicitly by putting the

2645: system in a

2646: periodic box.) This suggests that given two states $|A\ra$ and $|B\ra$,

2647: both

2648: carrying zero momentum, it is useful to define a modified inner product:

2649:  \be \label{e4.43}

2650: \la A | B\ra' = {1\over V_{p+1}} \, \la A | B\ra\, .

2651:  \ee

2652: In this convention

2653:  \be \label{ecnormp}

2654: \la 0| c_{-1} c_0 c_1 |0\ra' = 1\, .

2655:  \ee

2656: 

2657: 

2658: Let us now turn to the task of

2659: constructing an action for the string field theory. Given a string field

2660: configuration

2661: $|\Phi\ra$, the action $S(|\Phi\ra)$ should give a number. The action

2662: proposed in

2663: \cite{osft} is:

2664:  \be \label{e4.11}

2665: S = -{1\over g^2} \, \left[ {1\over 2} \la \Phi | Q_B|\Phi\ra + {1\over

2666: 3} \la

2667: \Phi|\Phi* \Phi\ra \right]\, .

2668:  \ee

2669: Here $g$ is a constant, known as the open string coupling constant.

2670: This

2671: is to be

2672: distinguished from the closed string coupling constant $g_s$ introduced

2673: earlier in

2674: eq.\refb{e2.16} for example. The precise relation between $g$ and $g_s$

2675: depends on

2676: which D-brane system we are considering and will be discussed later. In

2677: general $g_s\propto g^2$ with the constant of proportionality depending on

2678: the brane system on which we are formulating the open string field theory.

2679: $Q_B$

2680: is the BRST charge defined in \refb{e4.6}. The BPZ inner product $\la

2681: \cdot|\cdot\ra$ has

2682: been defined earler in eqs.\refb{e4.12}, \refb{e4.14}. The only

2683: operation appearing on the right hand side of \refb{e4.11} that has

2684: not been defined so far is the $*$-product

2685: $|A*B\ra$ \cite{osft} for $|A\ra, |B\ra \in \HH$. We shall define this

2686: now.

2687: 

2688:  \begin{figure}[!ht]

2689:  \begin{center}

2690: \leavevmode

2691: \epsfysize=5cm

2692: \epsfbox{fig9.eps}

2693: \end{center}

2694: \caption{The images of the upper half unit disk under the maps $h_1$,

2695: $h_2$ and

2696: $h_3$.

2697: } \label{f8}

2698: \end{figure}

2699: One can show that the BPZ inner

2700: product is non-degenerate. Hence $|A*B\ra$ is completely specified if

2701: we

2702: specify the

2703: inner product $\la C|A*B\ra$ for any state $|C\ra$. This is given as

2704: follows:

2705:  \be \label{e4.16}

2706: \la C|A*B\ra = \la h_1\circ C(0) h_2\circ A(0) h_3\circ B(0)\ra_D\, ,

2707:  \ee

2708: where

2709:  \be \label{e4.17}

2710: h_1(z) = \left({1+iz \over 1-iz}\right)^{2/3}, \qquad

2711: h_2(z) = e^{-2\pi i/3} h_1(z), \qquad h_3(z) = e^{-4\pi i/3} h_1(z)\, .

2712:  \ee

2713: The $\circ$ operation and $\la\cdot|\cdot \ra$ have been defined

2714: around

2715: eqs.\refb{e4.12} - \refb{e4.13}. $\la\cdot\ra_D$ denotes correlation 

2716: function on the unit disk as usual.

2717: The images of the upper half unit disk under the maps $h_1$, $h_2$ and

2718: $h_3$ are

2719: shown in Fig.\ref{f8}.

2720: 

2721: Using this definition of the $*$-product we can compute the action given in

2722: \refb{e4.11} for any string field configuration $|\Psi\ra$. It turns out

2723: that $Q_B$,

2724: $*$ and $\la\cdot|\cdot\ra$ satisfy some important identities:

2725:  \begin{enumerate}

2726: \item $Q_B$ is nilpotent:

2727:  \be \label{e4.18}

2728: (Q_B)^2 = 0\, .

2729:  \ee

2730: \item $Q_B$ can be `integrated by parts':

2731:  \be \label{e4.19}

2732: \la Q_B A|B\ra = -(-1)^{n_A} \la A |Q_B|B\ra\, ,

2733:  \ee

2734: where $n_A$ denotes the ghost number of the state $|A\ra$.

2735: \item $Q_B$ distributes over the $*$-product:

2736:  \be \label{e4.20}

2737: Q_B|A*B\ra = (Q_B|A\ra)*|B\ra + (-1)^{n_A} |A\ra * (Q_B|B\ra)\, .

2738:  \ee

2739: \item The BPZ inner product is symmetric:

2740:  \be \label{e4.21}

2741: \la A|B\ra = \la B|A\ra\, .

2742:  \ee

2743: \item The quantity $\la A|B*C\ra$ is cyclic:

2744:  \be \label{e4.22}

2745: \la A|B*C\ra = \la C|A*B\ra\, .

2746:  \ee

2747: \item The $*$-product is associative:

2748:  \be \label{e4.23}

2749: (|A\ra * |B\ra)*|C\ra = |A\ra * (|B\ra * |C\ra)\, .

2750:  \ee

2751: \end{enumerate}

2752: These identities can be proved by using the general properties of

2753: the

2754: matter-ghost BCFT, without restricting ourselves to any specific

2755: choice

2756: of the

2757: matter BCFT. We however do need to use the fact that the matter BCFT has

2758: total

2759: central charge

2760: 26.

2761: 

2762: Using these identities one can show that the action \refb{e4.11} is

2763: invariant under

2764: an infinitesimal gauge transformation:

2765:  \be \label{e4.24}

2766: \delta|\Phi\ra = Q_B|\Lambda\ra +|\Phi\ra * |\Lambda - |\Lambda\ra *

2767: |\Phi\ra\, ,

2768:  \ee

2769: where the infinitesimal gauge transformation parameter $|\Lambda\ra$ is

2770: an arbitrary

2771: state in $\HH_0$. More specifically, if $\{|\chi_{0,s}\ra\}$ denote a set

2772: of basis

2773: states in $\HH_0$, and if we expand $|\Lambda\ra$ as

2774:  \be \label{e4.25}

2775: |\Lambda\ra = \sum_s \lambda_s |\chi_{0,s}\ra\, ,

2776:  \ee

2777: then the coefficients of expansion $\lambda_s$ are infinitesimal and

2778: represent the

2779: gauge transformation parameters.\footnote{Again by regarding the sum over

2780: $s$ as a sum over a discrete index and integration over the continuous

2781: momentum index we can regard $\{\lambda_s\}$ as a set of functions of the

2782: momentum along the D-brane, or by taking their Fourier transform, a set

2783: of functions of the coordinates along the D-brane world-volume.} The

2784: variation of the action $S$ under

2785: the transformation \refb{e4.24} vanishes to first order in $\lambda_s$.

2786: 

2787: The equations of motion obtained by requiring $\delta S=0$ under

2788: arbitrary variation

2789: $\delta |\Phi\ra$ to first order in $\delta|\Phi\ra$, gives

2790:  \be \label{e4.26}

2791: Q_B|\Phi\ra + |\Phi\ra * |\Phi\ra =0\, .

2792:  \ee

2793: Thus at the linearized level the equations of motion take the form:

2794:  \be \label{e4.27}

2795: Q_B|\Phi\ra = 0\, .

2796:  \ee

2797: This agrees with the physical state condition \refb{e4.5}. Furthermore,

2798: at the

2799: linearized level the gauge transformation \refb{e4.24} takes the form:

2800:  \be \label{e4.28}

2801: \delta |\Phi\ra = Q_B|\Lambda\ra\, .

2802:  \ee

2803: Thus equivalence under linearized gauge transformation reproduces the

2804: equivalence

2805: relation \refb{e4.8} of the first quantized theory. This shows that gauge

2806: inequivalent solutions of the linearized equations of motion of string

2807: theory are in

2808: one to one correspondence to the physical states of the first quantized

2809: theory. This

2810: is one of the requirements that the string field theory must satisfy.

2811: 

2812: It can be shown that the other requirement, that the S-matrix elements

2813: involving

2814: physical external states computed using the Feynman rules of string field

2815: theory

2816: reproduce the S-matrix elements computed using Polyakov prescription, is

2817: also

2818: satisfied by the open string field theory described here

2819: \cite{osft,giddings,GIDDMARW}. The

2820: computation in

2821: string field theory requires a gauge fixing. The most commonly used gauge

2822: is the

2823: Siegel gauge, where we require:

2824:  \be \label{e4.29}

2825: b_0 | \Phi\ra =0\, .

2826:  \ee

2827: We shall make use of this gauge condition later.

2828: 

2829: \subsection{Reformulation of the tachyon condensation conjectures in 

2830: string field theory} \label{s4.2}

2831: 

2832: We shall now reformulate the three conjectures about the tachyon

2833: potential on a

2834: bosonic D-brane in the language of string field theory. Since the

2835: conjectures

2836: involve properties of classical solutions of open string field

2837: theory

2838: (translationally invariant vacuum solution and lump solutions) we

2839: shall begin

2840: by reviewing certain properties of classical solutions in open string field

2841: theory. Let $|\Phi_{cl}\ra$ denote a specific solution of the

2842: classical

2843: equations of

2844: motion \refb{e4.26}:

2845:  \be \label{e4.30}

2846: Q_B|\Phi_{cl}\ra + |\Phi_{cl}\ra * |\Phi_{cl}\ra =0\, .

2847:  \ee

2848: If we want to study string field theory around this classical

2849: solution, it is convenient to define shifted field $|\Psi\ra$ as

2850:  \be \label{e4.31}

2851: |\Psi\ra = |\Phi\ra - |\Phi_{cl}\ra\, ,

2852:  \ee

2853: and rewrite the original action as

2854:  \be \label{e4.32}

2855: S(|\Phi\ra) = S(|\Phi_{cl}\ra) + \wt S(|\Psi\ra)\, ,

2856:  \ee

2857: where

2858:  \be \label{e4.33}

2859: \wt S(|\Psi\ra) = -{1\over g^2} \left[ {1\over 2} \la \Psi| \QQ|\Psi\ra

2860: +{1\over 3} \la \Psi|\Psi * \Psi\ra \right]\, ,

2861:  \ee

2862: with

2863:  \be \label{e4.34}

2864: \QQ|A\ra \equiv Q_B|A\ra + |\Phi_{cl}\ra * |A\ra - (-1)^{n_A} |A\ra *

2865: |\Phi_{cl}\ra\, ,

2866:  \ee

2867: for any state $|A\ra\in \HH$. One can show that as long as

2868: $|\Phi_{cl}\ra$

2869: satisfies

2870: the equation of motion \refb{e4.30}, all the identities satisfied by

2871: $Q_B$,

2872: $*$-product and BPZ inner product hold with $Q_B$ replaced by $\QQ$.

2873: This, in turn shows that $\wt S(|\Psi\ra)$ is invariant under a gauge

2874: transformation:

2875:  \be \label{e4.35}

2876: \delta|\Psi\ra = \QQ|\Lambda\ra + |\Psi\ra * |\Lambda\ra - |\Lambda\ra

2877: * |\Psi\ra \, .

2878:  \ee

2879: {}From the structure of the action \refb{e4.33} and the gauge

2880: transformation law

2881: \refb{e4.35} it follows that the spectrum of perturbative physical open

2882: string

2883: states around the solution $|\Phi_{cl}\ra$, obtained by finding the

2884: gauge

2885: inequivalent

2886: solutions of the linearized equations of motion derived from $\wt

2887: S(|\Psi\ra)$, is

2888: in one to one correspondence with the cohomology of $\QQ$. In other words,

2889: they are given by eqs.\refb{e4.5}, \refb{e4.8} with $Q_B$ replaced by

2890: $\QQ$.

2891: 

2892: We are now in a position to restate the tachyon condensation conjectures

2893: in the

2894: language of string field theory. For simplicity we shall restrict our

2895: discussion to

2896: static D$p$-branes in flat (25+1) dimensional space-time, but many of the

2897: results

2898: hold for D-branes in more general space-time background.

2899:  \begin{enumerate}

2900: \item There is a translationally invariant solution $|\Phi_0\ra$ of the

2901: string field

2902: theory equations of motion:

2903:  \be \label{e4.36}

2904: Q_B|\Phi_0\ra + |\Phi_0\ra * |\Phi_0\ra =0\, ,

2905:  \ee

2906: such that

2907:  \be \label{e4.37}

2908: -{1\over V_{p+1}} \, S(|\Phi_0\ra) + \TT_p = 0\, .

2909:  \ee

2910: Here $V_{p+1}$ is the volume of the D-brane world-volume. Since for a space-time

2911: independent solution the value of the action is given by $-V_{p+1}$

2912: multiplied by

2913: the value of the potential at $|\Phi_0\ra$, \refb{e4.37} is a restatement

2914: of

2915: \refb{e2.17}.

2916: \item Associated with the solution $|\Phi_0\ra$ we have a nilpotent

2917: operator $\QQ$

2918: defined through eq.\refb{e4.34}

2919:  \be \label{e4.38}

2920: \QQ|A\ra \equiv Q_B|A\ra + |\Phi_0\ra * |A\ra - (-1)^{n_A} |A\ra *

2921: |\Phi_0\ra\, .

2922:  \ee

2923: Cohomology of $\QQ$ represents the spectrum of physical open string states 

2924: around

2925: $|\Phi_0\ra$. Since we do not expect any physical open string state around 

2926: the

2927: tachyon vacuum

2928: solution, the cohomology of $\QQ$ must be trivial.

2929: \item $\forall q$ such that $0<q\le p$, there should be a solution

2930: $|\Phi^q\ra$

2931: of the equations of motion which depends on $q$ of the spatial

2932: coordinates and

2933: represents a D-$(p-q)$-brane. The requirement that the energy per unit

2934: $(p-q)$-volume

2935: of this solution agrees with the tension of the D-$(p-q)$-brane gives:

2936:  \be \label{e4.39}

2937: S(|\Phi^q\ra) - S(|\Phi_0\ra) = - V_{p-q+1} \TT_{p-q}\, .

2938:  \ee

2939: Note that in computing the energy of the solution we take the zero of the

2940: energy to

2941: be at the tachyon vacuum solution, since by the first conjecture this

2942: represents the

2943: vacuum without any D-brane. Note also that the solution $|\Phi^0\ra$,

2944: which is

2945: supposed to describe the original D$p$-brane, is to be identified as the

2946: trivial

2947: solution ($|\Phi^0\ra = 0$). Thus $S(|\Phi^0\ra)=0$, and for $q=0$ 

2948: eq.\refb{e4.39} reduces to \refb{e4.37}.

2949: 

2950: \end{enumerate}

2951: 

2952: \subsection{Verification of the first conjecture}

2953: 

2954: In this subsection we shall discuss verification of

2955: eq.\refb{e4.37}.

2956: For this we need to study the component form of the action. Expanding the

2957: string

2958: field $|\Phi\ra$ in a basis as in \refb{e4.9} and substituting

2959: it in the

2960: expression for the action \refb{e4.11}, we get

2961:  \be \label{e4.44}

2962: S(|\Phi\ra) = -{1\over g^2} \left[{1\over 2} \AAA_{\alpha  \beta}

2963: \phi_\alpha \phi_\beta + {1\over 3} \CC_{\alpha\beta\gamma}

2964: \phi_\alpha \phi_\beta \phi_\gamma\right]

2965:   \ee

2966:  \begin{eqnarray} \label{e4.45}

2967: \AAA_{\alpha \beta} &=& \la

2968: \chi_{1,\alpha} | Q_B| \chi_{1,\beta}\ra \nonumber

2969: \\ \cr

2970: \CC_{\alpha\beta\gamma} &=&

2971: \la h_1\circ \chi_{1,\alpha}(0) \, h_2\circ \chi_{1,\beta}(0) \,

2972: h_3\circ \chi_{1,\gamma}(0)\ra_D \, .

2973: \end{eqnarray}

2974: As discussed before, for a D$p$-brane in flat space-time, the

2975: label $\alpha$ can be split into a pair of labels, -- a discrete

2976: label $r$ and a continuous momentum label $\{k^\mu\}$ along

2977: directions tangential to the D$p$-brane. Thus $\{\phi_\alpha\}$

2978: can be regarded as a set of functions $\{\phi_r(k)\}$, and

2979: $\sum_\alpha$ in the action can be replaced by $\sum_r \int

2980: d^{p+1} k$. The basis states $\{|\chi_{1,\alpha}\ra\}$ can be

2981: thought of as the set $\{|\chi_{1,r}(k)\ra\}$ with

2982: $|\chi_{1,r}(k)\ra$ being a state built on the Fock vacuum $|k\ra$

2983: by the action of various oscillators.

2984: 

2985: Analysis of the tachyon vacuum solution is simplified due to the fact

2986: that the

2987: solution is translationally invariant. In momentum space this allows

2988: us to

2989: write:

2990:  \be \label{e4.46}

2991: \phi_r(k) = \phi_r \, (2\pi)^{p+1} \delta^{(p+1)}(k) \, .

2992:  \ee

2993: Thus the expansion \refb{e4.9} can be rerwitten as

2994:  \be \label{e4.47}

2995: |\Phi\ra = \sum_r \phi_r |\chi_{1,r}\ra\, ,

2996:  \ee

2997: where $\{ |\chi_{1,r}\ra\}\equiv \{|\chi_{1,r}(k=0)\ra\}$ is a

2998: basis of zero momentum states in $\HH_1$. The component form of

2999: the action restricted to this subspace is given by,

3000:  \be \label{e4.48}

3001: S(|\Phi\ra)= -{1\over g^2} \Big[{1\over 2} \AAA_{rs}

3002: \phi_r

3003: \phi_s + {1\over 3} \CC_{rst} \phi_r \phi_s

3004: \phi_t\Big]\, ,

3005:  \ee

3006: where $\AAA_{rs}$ and $\CC_{rst}$ are defined in the same way as in

3007: \refb{e4.45},

3008: with

3009: the state $\chi_{1,\alpha}$ etc. replaced by $\chi_{1,r}$.

3010: Note however that since $|\chi_{1,s}\ra$ carries zero momentum, both

3011: $\AAA_{rs}$ and

3012: $\CC_{rst}$ carry explicit factors of $V_{p+1}$ due to the normalization

3013: condition \refb{e4.42}, and hence it will be more

3014: convenient for our analysis to define new coefficients $A_{rs}$ and

3015: $C_{rst}$ by

3016: removing this volume factor:

3017:  \be \label{e4.49}

3018: \AAA_{rs} = V_{p+1} A_{rs}, \qquad \CC_{rst} = V_{p+1} C_{rst}\, ,

3019:  \ee

3020:  \be \label{e4.49a}

3021:  A_{rs} = \la \chi_{1,r}|Q_B|\chi_{1,s}\ra', \qquad C_{rst} = \la 

3022: \chi_{1,r}|\chi_{1,s}*\chi_{1,t}\ra'\, ,

3023:  \ee

3024: where $\la \cdot\ra'$ has been defined in \refb{e4.43}. 

3025: The action \refb{e4.48} now may be written as

3026:  \be \label{e4.50}

3027: S(|\Phi\ra) = -{1\over g^2} \, V_{p+1} \, \VV(|\Phi\ra)\, ,

3028:  \ee

3029: where

3030:  \ben \label{e4.51}

3031: \VV(|\Phi\ra) &=& {1\over 2} A_{rs} \phi_r \phi_s + {1\over 3} C_{rst}

3032: \phi_r \phi_s

3033: \phi_t \nonumber \\

3034: &=& {1\over 2} \la \Phi| Q_B|\Phi\ra' + {1\over 3}

3035: \la\Phi|\Phi*\Phi\ra'\, .

3036:  \een

3037: Conjecture

3038: 1, given in \refb{e4.37} can now be rewritten as

3039:  \be \label{e4.52}

3040: {1\over g^2} \VV(|\Phi_0\ra) + \TT_p = 0\, .

3041:  \ee

3042: We can bring this into a more suggestive form by expressing $\TT_p$ in

3043: terms of

3044: $g^2$. An expression for $\TT_p$ in terms of closed string coupling

3045: constant $g_s$

3046: has been given in \refb{e2.16}, but we would like to express this in

3047: terms of the

3048: open string coupling constant $g$. This can be done in many ways.

3049: One way of doing this is to examine the open string field theory

3050: action carefully to determine the inertial mass per unit volume of

3051: the D-$p$-brane and identify this with $\TT_p$. This analysis yields the

3052: relation\cite{9911116}:

3053:  \be \label{e4.53}

3054: \TT_p = {1\over 2\pi^2 g^2}\, .

3055:  \ee

3056: Substituting this into \refb{e4.52} we get

3057:  \be \label{e4.54}

3058: 2\pi^2 \VV(|\Phi_0) + 1 = 0\, .

3059:  \ee

3060: 

3061: Thus our task is to begin with the $\VV(|\Phi\ra)$ given in \refb{e4.51},

3062: find a

3063: local minimum of this expression by varying the various coefficients

3064: $\phi_r$, and

3065: show that the value of $\VV(|\Phi\ra)$ at the local minimum satisfies

3066: \refb{e4.54}.

3067: Several simplifications occur in this computation which allows us to

3068: restrict

3069: $|\Phi\ra$ to a subspace smaller than the space of all zero momentum

3070: states in

3071: $\HH_1$.

3072: First of all it turns out that the structure of $S(|\Phi\ra)$ allows a

3073: consistent

3074: truncation of the action where we restrict

3075: $|\Phi\ra$ to linear combination of states created from the vacuum

3076: $|0\ra$ by the

3077: action of the ghost oscillators and matter {\it Virasoro generators}

3078: \cite{9911116},

3079: instead of

3080: letting $|\Phi\ra$ be an arbitrary linear combination of matter and ghost

3081: oscillators acting on the vacuum.\footnote{Consistent truncation of the

3082: action means

3083: that if we restrict $|\Phi\ra$ to this subspace, then the equations of

3084: motion

3085: associated with the components of $|\Phi\ra$ outside this subspace are

3086: automatically

3087: satisfied.} We shall call the subspace generated by these

3088: states the universal subspace.\footnote{Once we restrict $|\Phi\ra$ to the

3089: universal subspace, all the conformal field theory correlation functions

3090: which go into the

3091: computation of $\VV(|\Phi\ra)$ are independent of the specific choice of

3092: the conformal field theory used for this

3093: computation. This shows that once we have established \refb{e4.54} for

3094: some unstable D-brane in some closed string background, it proves the

3095: first conjecture for any D-brane in any closed string background in the

3096: bosonic string theory\cite{9911116}. A similar argument also works for

3097: the superstring theory.} Second, the $\VV(|\Phi\ra)$

3098: has a $Z_2$ symmetry known as the twist invarince, under which

3099:  \be \label{e4.55}

3100: |\Phi\ra \to (-1)^{L_0+1} |\Phi\ra\, ,

3101:  \ee

3102: where $L_n$'s denote the total Virasoro generators of the matter-ghost

3103: BCFT. If we

3104: denote by `twist'  the $(-1)^{L_0+1}$ eigenvalue of a state, then this

3105: $Z_2$

3106: symmetry allows us to restrict $|\Phi\ra$ to twist even sector in our

3107: search for the

3108: tachyon vacuum solution. Finally, due to the gauge invariance of the

3109: action, we can

3110: impose a gauge condition.

3111: One can show that in the twist even sector the Siegel gauge

3112: \refb{e4.29} is a good

3113: choice of

3114: gauge around the point $|\Phi\ra=0$ \cite{9912249}. Thus to begin with, we

3115: could

3116: look for a

3117: solution $|\Phi_0\ra$ in the Siegel gauge, and after we have obtained the

3118: solution,

3119: verify that

3120: the Siegel gauge is still a good choice of gauge near the solution

3121: $|\Phi_0\ra$. We shall denote by 

3122: $\wt\HH_1$ the restricted subspace of $\HH_1$ satisfying all

3123: these requirements.\footnote{In the Siegel gauge the action has an

3124: SU(1,1) invariance\cite{siegel-zwie,0010190} which allows us to further

3125: restrict $|\Phi\ra$ to

3126: SU(1,1) singlet

3127: subspace\cite{0010190}, but this has not so far been used effectively in

3128: simplifying the

3129: analysis. Nevertheless, once the solution has been found, one can explicitly

3130: check that the solution is an SU(1,1) singlet\cite{0010190}.}

3131: 

3132: Let us define the level of a state $|s\ra$ to be the difference

3133: between the $L_0$ eigenvalue $h$ of $|s\ra$ and the $L_0$

3134: eigenvalue of the state $c_1|0\ra$ representing the zero momentum

3135: tachyon. Since the latter state has $L_0$ eigenvalue $-1$, the

3136: level of $|s\ra$ is given by $(1+h)$. Thus, for example,

3137: $c_1|0\ra$ has level 0,  $c_{-1}|0\ra$ and $c_1 L^{(m)}_{-2}|0\ra$

3138: has level 2 etc. Using this definition we can partially order the zero 

3139: momentum

3140: basis states in the order of increasing level. Thus for example in

3141: $\wt\HH_1$, $|\Phi\ra$ can be expanded as:

3142:  \be \label{e4.56}

3143: |\Phi\ra = \phi_0 c_1 |0\ra + \phi_1 c_{-1} |0\ra + \phi_2

3144: L^{(m)}_{-2} c_1 |0\ra + \cdots\, ,

3145:   \ee

3146: where $\cdots$ involves states of level 4 and higher. Let us now

3147: make a drastic approximation where we set all the coefficients

3148: other than that of level zero state $c_1|0\ra$ to zero.

3149: Substituting this into \refb{e4.51} we get

3150:  \be \label{e4.57}

3151: \VV(\phi_0) = {1\over 2} \, \phi_0^2 \, \, \la 0 |c_{-1} Q_B

3152: c_1|0\ra' +{1\over 3} \, \phi_0^3 \, \, \la h_1\circ c(0) h_2\circ

3153: c(0) h_3\circ c(0)\ra'_D \, .

3154:   \ee

3155: The relevant correlation functions can be easily evaluated and

3156: give

3157:  \be \label{ehalf}

3158:   \VV(\phi_0) = -{1\over 2} \, \phi_0^2 + {1\over 3} \,

3159: \left({3\sqrt 3 \over 4}\right)^3 \, \phi_0^3\, .

3160:   \ee

3161: This has a local minimum at $\phi_0=(4/3\sqrt 3)^3$, and at this

3162: minimum,

3163:  \be \label{e4.58}

3164:   2\pi^2 \VV(\phi_0) = -(2\pi^2)

3165: (4/3\sqrt 3)^6 / 6 \simeq -.684 \, .

3166:   \ee

3167: This is about $68\%$ of the conjectured answer $-1$ given in \refb{e4.54}.

3168: 

3169: This is the beginning of a systematic approximation scheme known as the

3170: level

3171: truncation\cite{kost-sam1,kost-sam2,kost-pot1,9912249,0002237,0211012}. We 

3172: define the

3173: level of a coefficient $\phi_r$ to be

3174: the level of

3175: the state $|\chi_{1,r}\ra$ that it multiplies in the expansion of

3176: $|\Phi\ra$.

3177: We now define a level $(M,N)$ approximation to $\VV(|\Phi\ra)$

3178: as follows:

3179:  \begin{enumerate}

3180: \item Keep all fields $\phi_r$ of level $\le M$.

3181: 

3182: \item Keep all terms in the action for which the sum of the levels of all

3183: fields in

3184: that term is $\le N$.

3185: Thus for example at level (2,4) we shall include interaction terms of the

3186: form

3187: 0-0-0, 0-0-2, 0-2-2 but ignore interaction terms of the form 2-2-2.

3188: 

3189: \item This gives an expression for $\VV(|\Phi\ra)$ involving finite

3190: number of fields

3191: and finite number of terms.

3192: We find a (local) minimum of this $\VV(|\Phi\ra)$ and evaluate the value

3193: of

3194: $\VV(|\Phi\ra)$ at this minimum.

3195: 

3196: \end{enumerate}

3197: This defines the level $(M,N)$ approximation to $\VV(|\Phi_0\ra)$.

3198: In order for this to be a sensible approximation scheme, we need to ensure

3199: that the answer converges as we

3200: increase the values of $M,N$.

3201: In actual practice

3202: this method converges quite rapidly. For example the value of

3203: $-2\pi^2\VV(|\Phi_0\ra)$ in level $(L,2L)$ approximation increases

3204: monotonically

3205: towards 1 as we increase the value of $L$ up to $L=10$, reaching the

3206: value .9991 at

3207: $L=10$ \cite{0002237}. However beyond level 12 the value of

3208: $-2\pi^2\VV(|\Phi_0\ra)$

3209: overshoots the

3210: expected value 1, and continiue to increase with $L$ till about level 18

3211: approximation\cite{0211012}. Nevertheless analysis of the tachyon

3212: potential obtained by integrating out all fields other than $\phi_0$

3213: shows that $-2\pi^2\VV(|\Phi_0\ra)$ eventually

3214: turns back

3215: and approaches the expected value 1 from above \cite{0208149,0211012}.

3216: 

3217: Since the solution is constructed in the Siegel gauge, we need to verify

3218: that

3219: Siegel gauge is a valid gauge choice for this solution. Operationally

3220: what this

3221: amounts to is the following. In arriving at the solution, we have made

3222: sure that the

3223: variation $\delta S$ of the action under a variation $\delta|\Phi\ra$

3224: vanishes

3225: to first order in $\delta|\Phi\ra$ around the solution,

3226: {\it provided $\delta|\Phi\ra$ satisfies the Siegel gauge condition

3227: $b_0\delta|\Phi\ra=0$.} In order to check that the solution

3228: satisfies the

3229: full string field theory equations of motion we need to make sure that

3230: $\delta S$

3231: vanishes to first order in $\delta|\Phi\ra$ even if $\delta|\Phi\ra$ does

3232: not

3233: satisfy the Siegel gauge condition. To check this, we can simply take the

3234: first

3235: order variation of $\VV(|\Phi\ra)$ with respect to components of

3236: $|\Phi\ra$ which

3237: violate Siegel gauge condition and verify that these derivatives vanish

3238: when

3239: evaluated in the background of the solution $|\Phi_0\ra$ found using the

3240: level

3241: truncation scheme. This has been checked explicitly in

3242: refs.\cite{0009105,0211012}.

3243: 

3244: \subsection{Verification of the second and third conjectures} \label{s4.3}

3245: 

3246: We shall now briefly discuss the verification of the second and the third

3247: conjectures. Of these  the analysis of the third conjecture,

3248: eq.\refb{e4.39},

3249: proceeds in a way very

3250: similar to that of the  first 

3251: conjecture\cite{0003031,0005036,0008053,0008101}. The main difference is 

3252: that since the

3253: solution depends on $q$ of the spatial coordinates, we can no longer

3254: restrict

3255: $|\Phi\ra$ to be in the zero momentum sector for finding the solution

3256: $|\Phi^q\ra$;

3257: instead we must allow $|\Phi\ra$ to carry momentum along these $q$

3258: directions. The

3259: analysis can be simplified by compactifying the $q$ directions along

3260: which we want

3261: the lump to form. This makes the momenta in these directions discrete,

3262: and as a

3263: result $|\Phi\ra$ can still be expanded in the discrete basis.

3264: 

3265: The explicit construction of the solution now proceeds via a modified

3266: level

3267: truncation scheme where the level of a state, given by $(L_0+1)$,

3268: includes not only

3269: the oscillator contribution but also the contribution to $L_0$ due to the

3270: momentum

3271: along the

3272: compact directions. The level $(M,N)$ approximation is defined exactly as

3273: before,

3274: and for finite $(M,N)$ we still have a finite number of variables with

3275: respect to

3276: which we need to extremize the action. For $q=1$ the result converges

3277: fast

3278: towards

3279: the expected value as we increase the level of approximation

3280: \cite{0005036}. In particular, the tension of the soliton becomes

3281: independent of the radius of compactification, as is expected of a D-brane

3282: whose transverse direction is compactified. The convergence is also

3283: reasonably good for $q=2$

3284: \cite{0008053,0008101}. Today the best available results for the tension

3285: of the

3286: codimension $q$

3287: lump solution for $q=1$ and $q=2$ differ from their conjectured values

3288: by $1\%$

3289: \cite{0005036} and

3290: $13\%$ \cite{0008053,0008101} respectively.

3291: For larger values of $q$ the number of fields below a given level

3292: increases rapidly,

3293: slowing down the

3294: convergence. The number of fields also increases rapidly as we increase

3295: the radius of

3296: compactification since states carrying different momenta become closely

3297: spaced in

3298: level. Nevertheless the analysis has been done for different radii, and

3299: the tension

3300: of the resulting lump solution has been shown to be independent of the

3301: radius to a

3302: very good approximation as is expected of a D-brane.

3303: 

3304: All the lump solutions described above have been constructed in the Siegel 

3305: gauge. For the codimension one lump, the validity of the Siegel gauge 

3306: choice has been tested in \cite{0101014}. 

3307: 

3308: For verifying conjecture 2, we need to check that $\QQ$ defined in

3309: eq.\refb{e4.38} has trivial cohomology. This can be done as follows

3310: \cite{0103085}:

3311:  \begin{enumerate}

3312: \item Take the best available value of

3313: $|\Phi_0\ra$ obtained using the level truncation scheme and construct

3314: $\QQ$ from there using eq.\refb{e4.38}.

3315: \item Construct solutions of $\QQ|A\ra=0$ by taking

3316: $|A\ra$ to be arbitrary linear combinations of states up to a certain

3317: level.

3318: \item Show that for every such $|A\ra$, there is a state $|B\ra$

3319: such that $|A\ra=\QQ|B\ra$.

3320: \end{enumerate}

3321: There is however a further

3322: complication due to the fact that $|\Phi_0\ra$ obtained in the level

3323: truncation scheme is only an approximate solution of the equations of

3324: motion, and as a result $\QQ$ defined in

3325: eq.\refb{e4.38} does not square to zero exactly when we use this

3326: approximate value

3327: of $|\Phi_0\ra$. Hence a state $|A\ra = \QQ\, |B\ra$ does not

3328: satisfy $\QQ|A\ra=0$

3329: exactly. We can circumvent this problem by using an

3330: approximate rather than an exact analysis of the 

3331: $\QQ$-cohomology\cite{0103085}.

3332: Given a state $|A\ra$ satisfying

3333: $\QQ|A\ra=0$, we check if there is a state $|B\ra$ such that the

3334: ratio of the norm of $(|A\ra-\QQ|B\ra)$ to the norm of $|A\ra$ can be made

3335: {\it small}. Of course there is no natural norm in the space of the

3336: string field since the BPZ inner product of ghost number 1 state

3337: with itself vanishes by ghost charge conservation, but we could use

3338: several different artificial norms ({\it e.g.} by explicitly inserting a 

3339: factor of $c_0$ in the BPZ inner product), and

3340: check if the final conclusion is sensitive to the choice of the norm.

3341: It was found in \cite{0103085} that

3342: if we

3343: carry out this analysis in a subspace which includes string states up

3344: to a given

3345: oscillator level

3346: $L$, and carrying momentum $k$ with $|k^2|

3347: \le L$,

3348: then all $\QQ$ closed states are also $\QQ$-exact to within 1\%

3349: accuracy. This gives numerical evidence for the absence of the physical

3350: open string states around the tachyon vacuum.

3351: 

3352: A somewhat different approach to this problem has been suggested in 

3353: \cite{0309164}.

3354: Various other aspects of tachyon condensation in

3355: bosonic open string field

3356: theory have been discussed in

3357: refs.\cite{0001201,0002117,0003031,0006240,0007235,0007153,

3358: 0008033,0010034,

3359: 0010247,

3360: 0011238,0101162,0103103,

3361: 0105024,0105156,0105246,

3362: 0105264,

3363: 0106068,0107046,

3364: 0109182,0201095,

3365: 0201159,0202133,0205275,0302182,0304261,

3366: 0311115,

3367: 0403200,0406023,

3368: 0409249}. 

3369: 

3370: \subsection{Superstring field theory} \label{s4.4}

3371: 

3372: In this subsection we shall briefly discuss the use of superstring field

3373: theory in verifying the conjectured properties of the tachyon potential in

3374: superstring theory. For this discussion we shall use the

3375: Berkovits version of superstring field 

3376: theory\cite{9503099,9912121,0105230,0109100}. An

3377: alternative approach to this problem based on various cubic 

3378: versions of

3379: superstring field theory\cite{inverse,AREF1,AREF2,pty} has been proposed 

3380: in

3381: \cite{0004112,0011117,0107197},

3382: but we shall not discuss it here.

3383: 

3384: We shall begin by reviewing the formulation of first quantized

3385: open superstring  theory on a non-BPS

3386: D-brane\cite{9806155,9808141,9809111} in a convention which will

3387: facilitate the formulation of superstring field

3388: theory\cite{9503099,9912121,0002211}. We restrict ourselves to the

3389: Neveu-Schwarz (NS) sector of the theory, since only the bosonic

3390: fields arising in this sector are involved in the construction of

3391: the classical solutions describing the tachyon vacuum and various

3392: lower dimensional D-branes. In the first quantized formulation the

3393: bulk world-sheet theory is given by a $c=15$ superconformal field

3394: theory together with a set of anticommuting ghosts $b$, $c$, $\bar

3395: b$, $\bar c$ and commuting ghosts $\beta$, $\gamma$, $\bar\beta$,

3396: $\bar\gamma$. The fields $\beta,\gamma$ can be replaced by a pair

3397: of fermions $\xi,\eta$, and a scalar $\phi_g$ through the

3398: relations\cite{FMS}

3399:  \be \label{ei1}

3400: \beta=\p\xi e^{-\phi_g}, \qquad \gamma =\eta e^{\phi_g}\, .

3401:  \ee

3402: $\xi$ and $\eta$ have dimensions (0,0) and (1,0) respectively,

3403: whereas $\phi_g$ is a chiral scalar field with background charge

3404: so that $\la e^{q\phi_g(0)}\ra_D$ is non-zero only for $q=-2$.

3405: There are similar relations involving the anti-holomorphic fields.

3406: The fields $\xi$, $\eta$ and $\phi_g$ are normalized such that

3407:  \be \label{eghostnorm}

3408:  \xi(z) \eta(w) \simeq {1\over z-w}\, , \qquad \p \phi_g(z)

3409:  \p\phi_g(w) \simeq -{1\over (z-w)^2}\, ,

3410:  \ee

3411: with a similar relation among the anti-holomorphic components.

3412: Since the open string vertex operators will involve the boundary

3413: values of various fields, and since on the boundary the

3414: holomorphic and the anti-holomorphic ghost fields are set equal,

3415: we shall not need to refer to the anti-holomorphic ghost fields

3416: explicitly. The ghost number ($n_g$) and the picture number

3417: ($n_p$) assignments of various fields are defined as follows:

3418:  \ben \label{ei2}

3419: && b:\quad n_g=-1, n_p=0, \qquad c:\quad n_g=1, n_p=0 \, , \nonumber \\

3420: && e^{q\phi_g}:\quad n_g=0, n_p=q\, , \nonumber \\

3421: && \xi:\quad n_g=-1, n_p=1, \qquad \eta:\quad n_g=1, n_p=-1\, . \nonumber

3422: \\

3423:  \een

3424: The matter fields as well as the SL(2,R) invariant vacuum carry zero

3425: ghost and picture number.

3426: The GSO operator is given by:

3427:  \be \label{egso}

3428: (-1)^F (-1)^q\, ,

3429:  \ee

3430: where $F$ denotes the world-sheet fermion number of the matter

3431: fields, and $q$ denotes the $\phi_g$ momentum.

3432: 

3433: The physical open string states are states of ghost number 1, with

3434: each physical state having different representations in different

3435: picture numbers\cite{FMS}. Furthermore, they are required to satisfy

3436: the following conditions:

3437:  \begin{enumerate}

3438: \item $|\phi\ra$ is BRST invariant:

3439:  \be \label{esuperbrst}

3440: Q_B|\phi\ra = 0\, ,

3441:  \ee

3442: where $Q_B$ is the BRST charge defined as

3443:  \be \label{eqb4}

3444: Q_B = \oint dz \, j_B(z) = \oint dz \Bigl\{  c \bigl( T_m +

3445: T_{\xi\eta} + T_{\phi_g}) + c \partial c b +\eta \,e^{\phi_g} \, G_m

3446: - \eta\p \eta e^{2\phi_g} b \Bigr\}\, ,

3447:  \ee

3448:  \be \label{eb5}

3449: T_{\xi\eta}=\p\xi\,\eta, \quad T_{\phi_g}=-{1\over 2} \p\phi_g \p

3450: \phi_g -\p^2\phi_g \, .

3451:  \ee

3452: $T_m$ is the matter stress tensor and

3453: $G_m$ is the matter

3454: superconformal generator.

3455: \item Two states $|\phi\ra$ and $|\phi'\ra$ are considered to be

3456: equivalent

3457: if they differ by $Q_B|\Lambda\ra$ for some state $|\Lambda\ra$:

3458:  \be \label{esuperequiv}

3459: |\phi\ra \equiv |\phi\ra + Q_B|\Lambda\ra\, .

3460:  \ee

3461: \item $|\phi\ra$ can be either GSO odd or GSO even. The vertex

3462: operators of GSO even states are accompanied by Chan-Paton factors $I$

3463: ($I$ being the $2\times 2$ identity matrix), whereas vertex operators

3464: of GSO odd states are accompanied by Chan-Paton factor $\sigma_1$. This 

3465: rule is inherited from the parent brane-antibrane system for which the GSO 

3466: even states carry Chan-Paton facors $I$ and $\sigma_3$ and GSO odd states 

3467: carry Chan-Paton factors $\sigma_1$ and $\sigma_2$. The $(-1)^{F_L}$ 

3468: projection removes the states carrying Chan-Paton factors $\sigma_2$ and 

3469: $\sigma_3$.

3470: 

3471: \item 

3472: \label{smallp} 

3473: The field $\xi$ appears in the vertex operators for physical

3474: states only through its derivatives. In terms of the state $|\phi\ra$

3475: it means that $|\phi\ra$ is annihilated by $\eta_0$, where

3476: $\eta_n$, $\xi_n$ denote the modes of the fields $\eta$ and $\xi$

3477: defined

3478: through the

3479: expansion

3480:  \be \label{eetaxiexp}

3481: \eta(z) = \sum_n \eta_n z^{-n-1}, \qquad \xi(z) = \sum_n \xi_n

3482: z^{-n}\, .

3483:  \ee

3484: \end{enumerate}

3485: 

3486: This condition given in item \ref{smallp} above gives what is

3487: known as the small picture representation of the physical 

3488: states\cite{9108021}. For the formulation of

3489: the superstring field theory it is more convenient to use the big

3490: picture\cite{9503099,9912121} where given a state $|\phi\ra$

3491: satisfying the conditions given above, we use the state

3492: $\xi_0|\phi\ra$ to represent the same physical

3493: state.\footnote{Consequently a physical state in the big picture

3494: is annihilated by $Q_B\eta_0$ instead of $Q_B$.} In terms of

3495: vertex operators this corresponds to multiplying the vertex

3496: operator in the small picture representation by $\xi$. Thus for

3497: example a representation of the on-shell tachyon vertex operator

3498: carrying momentum $k$ in the small picture is $c e^{-\phi_g}

3499: e^{ik.X} \otimes \sigma_1$. The same vertex operator in the big

3500: picture will be given by $\xi\, c e^{-\phi_g} e^{ik.X} \otimes

3501: \sigma_1$. According to our convention given in \refb{ei2} the

3502: small picture representation is in picture number $-1$ whereas the

3503: big picture representation is in picture number 0.

3504: 

3505: Let us now turn to the construction of open superstring field theory.

3506: As in the case of bosonic string theory,

3507: a general open string field configuration $\wh \Phi$ is represented by

3508: a state

3509: in the world-sheet boundary conformal field theory which do not

3510: satisfy all the requirements of a physical state. It turns out that

3511: the choice of a general off-shell field configuration in superstring

3512: field theory is as follows. In the NS sector it contains two

3513: components, a GSO even component $\Phi_+$

3514: accompanied by a Chan-Paton factor $I$ and a GSO odd component $\Phi_-$

3515: accompanied by the Chan-Paton factor $\sigma_1$.

3516: Both $\Phi_+$ and $\Phi_-$ are required to have picture number 0 and

3517: ghost number 0.

3518: Thus we can

3519: write:\footnote{We shall adopt the convention that fields or operators

3520: with internal

3521: CP factors

3522: included are denoted by symbols with a

3523: hat on them, and fields or operators without internal

3524: CP factors included are

3525: denoted by symbols without a hat.}

3526:  \be

3527: \label{efsf}

3528: \hp = \Phi_+ \otimes I  + \Phi_- \otimes \sigma_1\, .

3529:  \ee

3530: 

3531: The string field theory action to be given below will involve

3532: calculating correlation functions inolving the vertex operators

3533: $\Phi_\pm$. In manipulating these correlation functions, we need

3534: to keep in mind that the string field {\it components} in the NS

3535: sector are always grassman even, whereas the open string vertex

3536: operators which these components multiply in the expression of

3537: $\Phi_\pm$ may be grassman even or grassman odd depending on the

3538: world-sheet fermion number carried by the vertex operator. In

3539: particular the fermion $\psi^\mu$ associated with the matter BCFT,

3540: and $b$, $c$, $\xi$, $\eta$ and $e^{q\phi_g}$ for odd $q$ are

3541: grassman odd fields. Using eqs.\refb{ei2}, \refb{egso} one can

3542: then show that $\Phi_+$ is grassman even whereas $\Phi_-$ is

3543: grassman odd. In particular, the zero momentum tachyon vertex

3544: operator $\xi c e^{-\phi_g}\otimes \sigma_1$ is GSO odd and

3545: grassman odd. Note the extra factor of $\xi$ in the tachyon vertex

3546: operator compared to the conventions used in section \ref{s3}.

3547: This is due to the fact that we are using the big picture.

3548: 

3549: The BRST charge $\wh Q_B$

3550: acting on the string field and the operator $\wh\eta_0$ is defined by:

3551:  \be \label{ei3}

3552: \wh Q_B = Q_B\otimes \sigma_3, \qquad

3553: \wh \eta_0 = \eta_0\otimes \sigma_3\, .

3554:  \ee

3555: Also, given a set of open string vertex operators $\ha_1,\ldots \ha_n$

3556: on a non-BPS D$p$-brane, we define:

3557:  \be \label{e200}

3558: \lll \wh A_1\ldots \wh A_n \rrr = Tr \Bigl\langle f^{(n)}_1 \circ

3559: \ha_1(0)\cdots

3560: f^{(n)}_n\circ \ha_n(0)\Bigr\rangle_D \, ,

3561:  \ee

3562: where the trace is over the internal CP matrices, and

3563:  \be \label{esk3}

3564: f^{(n)}_k(z) = e^{2\pi i (k-1)\over n} \Big({1+iz\over 1-iz}

3565: \Big)^{2/n}\,\quad  \hbox{for} \quad n\geq 1 .

3566:  \ee

3567: $\la \cdot\ra_D$ denotes correlation function on a unit disk as usual.

3568: 

3569: In terms of these quantities the open superstring field theory action on a

3570: non-BPS D-brane can be

3571: written as\cite{0001084,0002211},

3572:  \be \label{e00}

3573: S={1\over 4g^2} \lllb (e^{-\hp} \hq e^{\hp})(e^{-\hp}\he e^\hp) -

3574: \int_0^1 ds (e^{-s\hp}\p_s e^{s\hp})\{ (e^{-s\hp}\hq e^{s\hp}),

3575: (e^{-s\hp}\he e^{s\hp})\}\rrrb\, ,

3576:  \ee

3577: where we have divided the overall normalization by a factor

3578: of two in order to compensate for the trace operation on the

3579: internal matrices. $s$ in the second term is just an integration 

3580: parameter. By expanding the various exponentials in a Taylor series 

3581: expansion, and explicitly carrying out the $s$ integral for each term one 

3582: can represent the action as a power series expansion in the string field. 

3583: This is given in eq.\refb{e3a}.

3584: 

3585: \refb{e00} can be shown to 

3586: be invariant under the infinitesimal gauge transformation\cite{0001084}

3587:  \be \label{egtrsa}

3588: \delta e^{\hp} = (\hq \wh\Omega) e^{\hp} + e^{\hp}(\he\wh\Omega') \, ,

3589:  \ee

3590: where the infinitesimal gauge transformation parameters $\wh\Omega$ and

3591: $\wh\Omega'$ are states

3592: with $(n_g,n_p)$ values $(-1,0)$ and $(-1,1)$

3593: respectively. The internal CP indices carried by the gauge  parameters

3594: are as follows

3595:  \be

3596: \label{ingp}

3597: \wh\Omega = \Omega_+ \otimes \sigma_3  + \Omega_- \otimes i\sigma_2\,,

3598: \qquad \wh\Omega' = \Omega'_+ \otimes \sigma_3  + \Omega'_- \otimes

3599: i\sigma_2\, .

3600:  \ee

3601: The GSO even $\Omega_+$, $\Omega'_+$

3602: are

3603: Grassmann odd, while the GSO odd $\Omega_-$, $\Omega'_-$ are Grassmann

3604: even.

3605: The proof of invariance of \refb{e00} under \refb{egtrsa} can be carried

3606: out by

3607: straightforward algebraic manipulations\cite{0001084,0002211}.

3608: 

3609: An analysis similar to that in the case of bosonic string theory shows

3610: that with the normalization we have used here, the tension of the non-BPS

3611: D$p$-brane is given by:

3612:  \be \label{ec1}

3613: \TT_p = {1\over 2 \pi^2 g^2}\, .

3614:  \ee

3615: Thus in order to prove the first conjecture about the tachyon potential we

3616: need to show the existence of a solution $|\hp_0\ra$ of the equations of

3617: motion derived from the string field theory action, such that

3618:  \be \label{epreds}

3619: S(|\hp_0\ra) = V_{p+1} \, {1\over 2 \pi^2 g^2}\, ,

3620:  \ee

3621: where $V_{p+1}$ is the volume of the D$p$-brane world-volume.

3622: 

3623: The calculation proceeds as in the case of bosonic string theory using the

3624: level truncation scheme\cite{0001084,0002211}. For this we need to first

3625: expand the

3626: action \refb{e00} in a power series expansion in the string field. This

3627: gives

3628:  \be \label{e3a}

3629: S = {1\over 2 g^2} \sum_{M,N=0}^\infty {1 \over (M+N+2)!} {M+N \choose N}

3630: (-1)^N \left\la\left\la  (\hq \hp) \hp^M (\he\hp)\hp^N

3631: \right\ra\right\ra\, .

3632:  \ee

3633: As in the case of bosonic string theory, for finding the tachyon

3634: vacuum solution we can restrict the string field to the universal

3635: subspace created from the SL(2,R) invariant vacuum by the action

3636: of the ghost oscillators and matter super-Virasoro generators. The

3637: string field theory action in this universal subspace is invariant

3638: under a $Z_2$ twist symmetry, which, acting on a vertex operator

3639: of conformal weight $h$, has an eigenvalue given by $(-1)^{h+1}$

3640: for even $2h$, and $(-1)^{h+{1\over 2}}$ for odd $2h$. Using this

3641: symmetry, and the fact that the zero momentum tachyon vertex

3642: operator $\xi c e^{-\phi_g}\otimes \sigma_1$ is twist even, we can

3643: restrict the string field to twist even sector for finding the

3644: tachyon vacuum solution.  Finally, we choose the gauge conditions

3645:  \be \label{egauge}

3646: b_0\hp=0, \qquad \xi_0\hp=0\, ,

3647:  \ee

3648: for fixing the gauge symmetries generated by $\wh\Omega$ and $\wh\Omega'$

3649: respectively.

3650: 

3651: We can now proceed to find the tachyon vacuum solution using the level

3652: truncation scheme as in the case of bosonic string theory. Since the zero

3653: momentum tachyon state has conformal weight $-{1\over 2}$, we define the

3654: level of the state (and the coefficient multiplying it) to be ${1\over

3655: 2}+h$ where $h$ is the conformal weight of the state. It turns out that

3656: although the action \refb{e3a} has infinite

3657: number of terms, up to a given level only a finite number of terms

3658: contribute due to various charge conservation\cite{0002211}.

3659: This allows us to express the action to a given level of approximation as

3660: a finite order polynomial in a finite number of string fields.

3661: The resulting action is then extremized with respect to the component

3662: fields to find the vacuum solution $|\hp_0\ra$.

3663: 

3664: The result for $S(|\hp_0\ra)$ converges rapidly to the expected

3665: value \refb{epreds}. At level (0,0) approximation, where we keep

3666: only the zero momentum tachyon vertex operator $\xi c

3667: e^{-\phi_g}\otimes \sigma_1$, the action contains two terms,

3668: proportional to $t^2$ and $t^4$ respectively, where $t$ is the

3669: coefficient of $\xi c e^{-\phi_g}\otimes \sigma_1$ in the

3670: expansion of $\hp$. By minimizing the action with respect to $t$,

3671: we get $\pi^2/16$ (about 60\%) times the expected

3672: answer\cite{0001084}. At level $(3/2, 3)$ we get about 85\% of the

3673: conjectured answer\cite{0002211} and at level (2,4) we get 89\% of

3674: the conjectured answer\cite{0003220,0004015}.

3675: 

3676:  \begin{figure}[!ht]

3677: \leavevmode

3678:  \begin{center}

3679: \epsfbox{combo.eps}

3680: \end{center}

3681: \caption[]{\small The tachyon potential $v(t)=V(t)/\TT_p$ in level (3/2,

3682: 3) approximation (solid line).

3683: For reference we also show

3684: the

3685: zeroeth order potential (dashed line).} \label{fber1}

3686: \end{figure}

3687: In this case we can also try to integrate out all the fields other

3688: than the tachyon $t$ labelling the coefficient of $\xi c

3689: e^{-\phi_g}\otimes \sigma_1$ and obtain a tachyon effective

3690: potential. To level (3/2, 3), this can be done analytically. The

3691: shape of the tachyon potential at level (0,0) and level (3/2, 3)

3692: has been shown in Fig.\ref{fber1}. {}From this we can clearly see

3693: the emergence of the double well shape of the potential.

3694: 

3695: In principle, the conjecture describing a codimension one D-brane as a

3696: kink solution on

3697: the non-BPS D$p$-brane can be analyzed following a procedure similar to

3698: that in the case of bosonic open string theory. Similarly the verification

3699: of the absence of physical open string states around the tachyon vacuum

3700: can also be carried out in a manner similar to that in the case of bosonic

3701: string theory. However in practice, due to technical complexities

3702: involved in these

3703: analyses, not much work has been done in these directions.

3704: 

3705: Various other aspects of tachyon condensation in open superstring field

3706: theory have been discussed in 

3707: \cite{0104230,0204155,0209186,0305103,0008127}.

3708: 

3709: \subsection{Vacuum String Field Theory}

3710: 

3711: The analysis of section \ref{s4.2} shows that it should be possible to

3712: reformulate string field theory by expanding the action around the tachyon 

3713: vacuum

3714: solution $|\Phi_0\ra$. The resulting string field theory action should

3715: have the

3716: same form as the original string field theory action with cubic

3717: interaction, but

3718: in the quadratic term the BRST operator $Q_B$ is replaced by another

3719: operator $\QQ$:

3720:  \be \label{evv0}

3721: \wt S(|\Psi\ra) = -{1\over g^2} \left[ {1\over 2} \la \Psi| \QQ|\Psi\ra

3722: +{1\over 3} \la \Psi|\Psi * \Psi\ra \right]\, .

3723:  \ee

3724: $|\Psi\ra$ as usual is a state of ghost number 1 of the first

3725: quantized open string living on the original D-brane. For future

3726: reference we shall call the boundary conformal field theory

3727: associated with this D-brane BCFT$_0$. In order to determine the

3728: precise form of $\QQ$, we need to know the analytic form of the

3729: solution $|\Phi_0\ra$ describing the tachyon vacuum. Given that no

3730: analytic expression for $|\Phi_0\ra$ is known at present, we could

3731: ask if it is possible to {\it guess} a form for $\QQ$ which satisfies

3732: all the conditions and conjectures described in section

3733: \ref{s4.2}. A general form of $\QQ$ proposed in

3734: \cite{0012251,0106010,0108150,0111129} is:

3735:  \be \label{evv1}

3736: \QQ = \sum_{n=0}^\infty a_n \left(  c_n +  (-)^n \, c_{-n} \right) \, ,

3737:  \ee

3738: where $a_n$ are some coefficients and $c_n$ are the usual ghost

3739: oscillators. The important point to note is that $\QQ$ is independent of

3740: the matter part of the BCFT. For any choice of the $a_n$'s, this $\QQ$

3741: can be shown to satisfy

3742: the conditions \refb{e4.18} - \refb{e4.20} with $Q_B$ replaced by $\QQ$,

3743: as is required for the gauge

3744: invariance of the theory. Further it has vanishing cohomology in the Fock

3745: space. To see this note that if $a_0\ne 0$, then given any solution

3746: $|\psi\ra$ of the equation $\QQ|\psi\ra=0$ we

3747: have $|\psi\ra = \QQ (a_0)^{-1} b_0 |\psi\ra$. If $a_0=0$, but $a_m\ne 0$

3748: for some $m$, we can write $|\psi\ra =

3749: {1\over 2a_m} \QQ ((-1)^m b_m - b_{-m})|\psi\ra$.

3750: Hence $|\psi\ra$ is $\QQ$

3751: trivial.

3752: 

3753: It now remains to find the classical solutions predicted by the first and

3754: the third conjecture in this field theory.

3755: Clearly the tachyon vacuum solution corresponds to the

3756: configuration $|\Psi\ra=0$.

3757: Thus the non-trivial solution we need to look for are those describing

3758: various D-branes, including the original D-brane associated with BCFT$_0$.

3759: In order to find them, we

3760: look for classical solutions of the form:

3761:  \be \label{eo3}

3762: \Psi = \Psi_g \otimes \Psi_m\, ,

3763:  \ee

3764: where $\Psi_g$

3765: denotes a state obtained by acting with the ghost

3766: oscillators on the SL(2,R) invariant vacuum of the ghost

3767: sector of BCFT$_0$, and

3768: $\Psi_m$  is a

3769: state obtained by acting with matter vertex operators on the SL(2,R)

3770: invariant

3771: vacuum of the matter sector of BCFT$_0$.

3772: If we

3773: denote by

3774: $*^g$ and $*^m$ the star product in the ghost and matter sector

3775: respectively, the equations of motion

3776:  \be \label{eo2}

3777: \QQ \Psi + \Psi * \Psi = 0\,

3778:  \ee

3779: factorize as

3780:  \be \label{eo4}

3781: \QQ \Psi_g = - K \Psi_g *^g \Psi_g \,,

3782:  \ee

3783: and

3784:  \be \label{eo5}

3785: \Psi_m = K^{-1} \, \Psi_m *^m \Psi_m\, ,

3786:  \ee

3787: where $K$ is an arbitrary constant that can be changed by scaling $\psi_g$ 

3788: and $\psi_m$ in opposite directions.

3789: We further assume that the ghost part $\Psi_g$ is universal for all

3790: D-$p$-brane solutions. Under this assumption the ratio of energies

3791: associated with two different D-brane solutions, with matter parts

3792: $\Psi_m'$ and $\Psi_m$ respectively, is given by:

3793:  \be \label{eo7}

3794: {\EE'\over\EE} = {\wt S(\Psi_g\otimes \Psi_m')\over \wt S(\Psi_g\otimes 

3795: \Psi_m)} = {\langle \Psi_m' | \Psi_m'\rangle_m \over \langle \Psi_m

3796: |

3797: \Psi_m\rangle_m} \, ,

3798:  \ee

3799: with $\langle \cdot| \cdot\rangle_m$ denoting BPZ inner product in

3800: the matter BCFT. Thus the ghost part drops out of this calculation.

3801: The equation \refb{eo5} for the mattter part tells us that $|\Psi_m\ra$ is

3802: a projector under the $*$-product in the matter sector up to a constant of

3803: proportionality.

3804:  \begin{figure}[!ht]

3805: \leavevmode

3806:  \begin{center}

3807: \epsfxsize 14cm

3808: \epsfbox{fvsft.eps}

3809: \end{center}

3810: \caption[]{\small The geometry used for defining the right hand side of

3811: eq.\refb{eo8}. The thin part of the

3812: real line has boundary conditions / interactions relevant to BCFT$_0$,

3813: and the thick part of the real line has boundary conditions /

3814: interactions relevant to BCFT.}

3815: \label{fvsft}

3816: \end{figure}

3817: 

3818: 

3819: 

3820: General methods for constructing such projectors have been

3821: developed in

3822: \cite{0008252,0102112,0105058,0105059,0105168,0106036,0202151}.

3823: Let BCFT denote some boundary conformal field theory with the same

3824: bulk world-sheet action as BCFT$_0$, {\it i.e.} both BCFT and

3825: BCFT$_0$ represent D-branes in the same space-time background.

3826: Consider now a state $|\Psi_m^{BCFT}\ra$ in the matter part of

3827: BCFT$_0$, defined through the relation:\footnote{Note that once we have 

3828: chosen the reference BCFT$_0$, the string field $|\Psi\ra$ is always a 

3829: state in  BCFT$_0$ even if it describes a D-brane 

3830: associated with another BCFT.}

3831:  \be \label{eo8}

3832: \la \Psi_m^{BCFT} | \phi\ra = \la f\circ \phi(0)\ra'_{BCFT}

3833:  \ee

3834: for any state $|\phi\ra$ in the matter part of BCFT$_0$.

3835: Here $f(\xi)$ is the conformal map

3836:  \be \label{eo9}

3837: f(\xi) = \tan^{-1}\xi\, ,

3838:  \ee

3839: and $\la\cdot\ra'_{BCFT}$ denotes correlation function of the matter 

3840: theory on the

3841: upper half plane, with the boundary condition associated with

3842: BCFT$_0$ in the range $-{\pi\over 4} -\epsilon \le x \le{\pi\over

3843: 4}+\epsilon$ and the boundary condition / interaction associated

3844: with some other boundary conformal field theory BCFT in the range

3845: ${\pi\over 4}+\epsilon<x<\infty$ and $-\infty< x < -{\pi\over

3846: 4}-\epsilon$. Here $\epsilon$ is a small positive number which

3847: should be taken to zero at the end. The geometry has been shown in

3848: Fig.\ref{fvsft} with the thin part of the real line having

3849: boundary conditions / interactions relevant to BCFT$_0$, and the

3850: thick part of the real line having boundary conditions /

3851: interactions relevant to BCFT. Note that since $f(0)=0$, the

3852: vertex operator $\phi$ is inserted at the origin where we have

3853: boundary condition associated with BCFT$_0$. This is consistent

3854: with the fact that $|\phi\ra$ is a state in BCFT$_0$. It was shown

3855: in \cite{0105168} that $|\Psi_m\ra$ defined in \refb{eo8} satisfy

3856: the projector equation \refb{eo5}, with a constant $K$ that is

3857: independent of the choice of BCFT. Furthermore, if we denote by

3858: $|\Psi_m^{BCFT'}\ra$ a state defined through eq.\refb{eo8} with

3859: BCFT replaced by another boundary conformal field theory BCFT$'$

3860: (with the same bulk CFT as BCFT or BCFT$_0$), one can

3861: show that the ratio

3862:  \be \label{eo11}

3863: {\langle \Psi_m^{BCFT} | \Psi_m^{BCFT}\rangle \over \langle

3864: \Psi_m^{BCFT'} | \Psi_m^{BCFT'}\rangle} \, ,

3865:   \ee

3866: is equal to the ratio of tensions of the D-branes associated with

3867: BCFT and BCFT$'$ respectively\cite{0105168}. \refb{eo7} then

3868: suggests that we identify

3869:  \be \label{eo12}

3870: |\Psi_g\ra \times |\Psi_m^{BCFT}\ra 

3871:  \ee

3872: as the classical solution in vacuum string field theory describing

3873: the D-brane associated with BCFT. In particular the D-brane

3874: associated with BCFT$_0$ is described by the state

3875: $|\Psi_g\ra \times |\Psi_m^{BCFT_0}\ra$ where in the computation of 

3876: $|\Psi_m^{BCFT_0}\ra$ the correlation

3877: function on the right hand side of \refb{eo8} is calculated with

3878: the boundary condition associated with BCFT$_0$ along the whole

3879: real axis.

3880: 

3881: The conformal map $f$ defined in \refb{eo9} has the property

3882: that it maps the point $i$

3883: to $\infty$, and the points $\pm 1$ to $\pm{\pi\over 4}$. If we

3884: think of the open string to be situated along the unit semi-circle

3885: on the upper half plane, then the point $i$ is the mid-point of

3886: the open string, and the map $f$ sends the mid-point to $\infty$

3887: which is a point on the boundary of the world-sheet. 

3888: It turns out that this is the important property that makes the state 

3889: $|\Psi_m^{BCFT}\ra$ into a projector\cite{0202151}.

3890: One can

3891: construct other conformal transformations which map the mid-point of the 

3892: open string 

3893: to the boundary, and these have

3894: also been used to construct projectors of the $*$-algebra using

3895: formula similar to \refb{eo8} \cite{0202151}. It is generally

3896: believed that different projectors associated with different

3897: conformal maps but same BCFT give gauge equivalent solutions, so

3898: that we have one inequivalent solution for a given BCFT.

3899: 

3900: Note that in the discussion so far we did not have to know anything about

3901: the coefficients $a_n$ in \refb{evv1}. It turns out that various

3902: consistency conditions leads to a unique choice of $\QQ$ up to a constant

3903: of proportionality:

3904:  \be \label{evv5}

3905: \QQ = \gamma c(i) \, ,

3906:  \ee

3907: where $\gamma$ is a constant. The coefficient $\gamma$ can be fixed by

3908: requiring that the classical solutions described above not only reproduces

3909: the ratios of various D-brane tensions, but also give the overall

3910: normalization of the tension correctly. Unfortunately this leads to a

3911: singular coefficient $\gamma$\cite{0111129}. This indicates that vacuum

3912: string

3913: field theory described by the kinetic term \refb{evv5} must be related to

3914: the original open string field theory expanded around

3915: the tachyon vacuum by a singular field redefinition\cite{0111129}. A

3916: complete

3917: understanding of this regularization procedure remains a challenge as of

3918: today, although there has been quite a lot of

3919: progress\cite{0204012,0310264}.

3920: 

3921: Various other aspects of vacuum string field theory have been discussed in 

3922: \cite{0105184,0110124,0110136,

3923: 0111034,0111069,0111087,0111092,

3924: 0111153,0111281,

3925: 0112169,0112202,

3926: 0112214,

3927: 0112231,0201015,0201060,

3928: 0201136,0201149,

3929: 0201177,

3930: 0201197,0201229,

3931: 0202087,0202139,0203071,0203188,0203227,

3932: 0204031,0204138,0204172,0204233,

3933: 0206208,0207001,0207044,

3934: 0208009,0208067,0209186,0211286,

3935: 0212055,0212335,

3936: 0301079,

3937: 0302151,0304270,0305010,

3938: 0311198,0403031,0403283,

3939: 0404154,0409063}.

3940: 

3941: \sectiono{Boundary String Field Theory} \label{s7}

3942: 

3943: So far we have discussed two different approaches to studying the

3944: problem of tachyon condensation, -- the approach based

3945: on the correspondence between two dimensional conformal field theories and

3946: classical solutions of string field theory equations of motion, and the 

3947: direct

3948: approach based on the analysis of classical equations of motion of open

3949: string field theory. Although it is generally believed that these two

3950: approaches are equivalent, this equivalence is not manifest. In

3951: particular there is no known procedure for finding an explicit

3952: solution of the string field theory equations of motion associated with a

3953: given two dimensional conformal field theory (or vice versa). In this

3954: section we shall discuss a

3955: different version of string field theory which makes the relationship

3956: between these two approaches more

3957: explicit\cite{9208027,9210065,9303067,9303143,9311177}. This version of

3958: string field

3959: theory has been given the name boundary string field theory (BSFT). We

3960: shall restrict our discussion to bosonic string theory only; for boundary

3961: string field theory associated with superstring theory see

3962: \cite{0010108,0103089,0106231}.

3963: 

3964: As in the case of the cubic string field theory, the open string field in

3965: boundary string field theory is a state $|\Phi\ra$ of ghost number 1 of

3966: the first quantized open string. We

3967: can associate with it a boundary vertex operator $\Phi$ of ghost number 1.

3968: Let $\{\chi_{1,\alpha}\}$ denote a complete set of

3969: vertex operators of ghost number 1, so that we can expand $\Phi$ as

3970:  \be \label{eeb2}

3971: \Phi = \sum_\alpha \vp_\alpha \chi_{1,\alpha}\, .

3972:  \ee

3973: $\{\vp_\alpha\}$ are the dynamical variables of the string field

3974: theory.\footnote{Presumably these $\vp_\alpha$'s are related to the 

3975: $\phi_\alpha$'s

3976: appearing in eq.\refb{e4.9} by a complicated field redefinition.} We also 

3977: define

3978:  \be \label{eeb0a}

3979: |V\ra = b_{-1} |\Phi\ra\, ,

3980:  \ee

3981: and let $V$ be the vertex operator associated with $|V\ra$. Clearly $V$

3982: has ghost number 0. The boundary string field theory action

3983: $S_B$ is a function of the variables $\{\vp_\alpha\}$ given by the 

3984: solution to the equation\cite{9208027}

3985:  \be \label{eeb3}

3986: {\p S_B\over \p \vp_\alpha} = - {\TT_p\over 2} \, \int

3987: {d\theta\over

3988: 2\pi} \, \int {d\theta'\over 2\pi} \, \la \chi_{1,\alpha}(\theta)\, \{

3989: Q_B,

3990: \Phi(\theta')\}\ra_V\, .

3991:  \ee

3992: Here

3993: $\TT_p$ is the tension

3994: of the

3995: D-$p$-brane,

3996: $Q_B$ is the usual BRST charge defined as the contour integral of

3997: the BRST currents $j_B(z)$, $\bar j_B(\bar z)$,

3998: and

3999: $\la \cdot \ra_V$ denotes the correlation function in the two

4000: dimensional

4001: field theory on a unit disk,

4002: described by the world-sheet action:

4003:  \be \label{eeb4}

4004: s_{Bulk} + \int_0^{2\pi} \, {d\theta\over 2 \pi} \, V(\theta)\, .

4005:  \ee

4006: Here the angle $\theta$ parametrizes the boundary of the disk, and

4007: $s_{Bulk}$ denotes the bulk world-sheet action involving matter and ghost

4008: fields. In computing $\la \cdot \ra_V$ we must use unnormalized

4009: correlation functions with the convention that in the absence of any

4010: boundary deformation,

4011:  \be \label{eghno}

4012: -{1\over 2} \, \la c(P) \p c(P) \p^2 c(P)\ra_{V=0}  = V_{p+1}\, ,

4013:  \ee

4014: where $P$ is any point on the disk and $V_{p+1}$ is the volume of 

4015: space-time along the D-$p$-brane whose

4016: dynamics we are describing.

4017: Note that although in general the conformal invariance is broken by the

4018: boundary interaction term for the action

4019: \refb{eeb4}, the theory in the bulk is still conformally invariant, and

4020: hence $j_B(z)$, $\bar j_B(\bar z)$ in the bulk are well defined. This can 

4021: be used to define

4022: $Q_B$ appearing in \refb{eeb3}. The overall normalization constant in

4023: \refb{eeb3} is fixed by

4024: comparing the open string amplitudes computed from the action

4025: \refb{eeb3} with those computed using the world-sheet

4026: theory\cite{9210065} or in cubic open string field

4027: theory\cite{0009191}.

4028: 

4029: This gives the BSFT action for a given string field configuration

4030: $|\Phi\ra$. The expression for the action simplifies for special class of 

4031: $|\Phi\ra$ of the

4032: form\footnote{At this stage we cannot claim that \refb{espp1} corresponds 

4033: to a consistent truncation. However the string field configuration 

4034: \refb{een14} 

4035: that will be of interest to us does correspond to a consistent 

4036: truncation.}

4037:  \be \label{espp1}

4038: \Phi = c V_m\, ,

4039:  \ee

4040: where $V_m$ is a linear combination of primary operators in the matter

4041: BCFT. Let

4042: $\{V_{mj}\}$ denote a complete set of primary vertex operators in the

4043: matter part

4044: of

4045: the BCFT, so that we can expand $V_m$ as 

4046:  \be \label{evmpara}

4047: V_m = \sum_j \lambda_j

4048: V_{mj}\, .

4049:  \ee  

4050: If $\Delta_j$ denotes the conformal weight of $V_{mj}$ then

4051:  \be \label{exxxx1}

4052: \{Q_B, \Phi(\theta)\} =- \sum_j (\Delta_j-1)

4053: \lambda_j

4054: V_{mj}(\theta) c\partial c(\theta)\, .

4055:  \ee

4056: \refb{eeb3} now gives:

4057:  \be \label{espp1aa}

4058: {\p S_B \over \p \lambda_i} =-{\TT_p\over 2} \, \sum _j (\Delta_j-1) \,

4059: \lambda_j

4060: \, \int

4061: {d\theta\over

4062: 2\pi} \, \int {d\theta'\over 2\pi} \, \la\, c(\theta) \,

4063: V_{mi}(\theta)\,

4064: c(\theta')\, \p c(\theta')\, V_{mj}(\theta')\, \ra_V \, .

4065:  \ee

4066: This equation however is not covariant under a change in the coordinate

4067: system $\lambda_i$ labelling the two dimensional quantum field

4068: theories\footnote{Such a change in

4069: the coordinate system may be induced by a

4070: change in the

4071: renormalization scheme in this quantum field theory.} 

4072: and holds only in

4073: a

4074: special coordinate system in which the

4075: relation \refb{exxxx1} continues to hold even in the deformed theory 

4076: described by the action \refb{eeb4}.

4077: A covariant version of this equation which is independent of the way we 

4078: choose the parameters $\vec\lambda$ labelling $V_m$ takes the 

4079: form\cite{9303143}

4080:  \be \label{espp2}

4081: {\p S_B \over \p \lambda_i} \propto \sum_j\, \beta^j(\vec\lambda)

4082: G_{ij}(\vec

4083: \lambda)\, ,

4084:  \ee

4085: where $\beta^j$ is the beta-function determining the renormalization group

4086: flow of $\lambda^j$ and $G_{ij}$ is the Zamolodchikov metric in the space

4087: of two dimensional theories deformed by boundary vertex operators in

4088: the matter sector. For the parametrization of $V_m$ given in 

4089: \refb{evmpara} and a suitable renormalization scheme we have

4090:  \be \label{echeck}

4091: \beta^j(\vec\lambda)\propto (\Delta_j-1)\lambda_j, \qquad

4092: \la c(\theta) V_{mi}(\theta) c(\theta') \p c(\theta') V_{mj}(\theta')\ra_V 

4093: \propto G_{ij}(\vec\lambda)\, ,

4094:  \ee 

4095: and 

4096: \refb{espp2}

4097: agrees with \refb{espp1aa}. Eq.\refb{espp2} makes it clear that a

4098: conformally

4099: invariant

4100: two dimensional field theory in the matter sector, for which

4101: $\beta^j$'s

4102: vanish, corresponds to a solution of the BSFT equations of motion.

4103: 

4104: There is however a difficulty in

4105: defining the action using \refb{eeb3} for string field configurations

4106: associated with vertex operators of dimension $>1$ since they give rise to

4107: non-renormalizable world-sheet field theory.

4108: For this reason a systematic quantization procedure for this field

4109: theory has not yet been developed.

4110: Fortunately in our analysis

4111: of tachyon condensation using classical BSFT we shall not need to deal

4112: with such

4113: operators. In particular in order to study classical solutions in BSFT

4114: involving the tachyon field, we shall focus on the string field

4115: configurations $|\Phi\ra$ of the form

4116:  \be \label{eeb6}

4117: \Phi = c T(X)

4118:  \ee

4119: where $T(x)$ is some function which has the interpretation of

4120: being the tachyon profile in BSFT. In this case $V=T(X)$ and

4121:  \be \label{elapl}

4122: \{Q_B, \Phi\} = \{ Q_B, c T(X)\} = c\, \p c \, (\square_X+1)T(X)\, ,

4123:  \ee

4124: where

4125:  \be \label{edefbox}

4126: \square_x = \eta^{\mu\nu} {\p\over \p x^\mu} {\p\over \p x^\nu}\, .

4127:  \ee

4128: 

4129: First consider the case of constant tachyon $T=a$. In this case

4130: $\Phi(\theta)=a\, c(\theta)$

4131: and $V(\theta) = a$ (multiplied by the

4132: the identity operator $I$ in the world-sheet theory).

4133: This gives

4134:  \be \label{eeb7}

4135: \int_0^{2\pi} \, {d\theta\over 2 \pi} \, V(\theta) = a\, .

4136:  \ee

4137: Thus the world-sheet correlation functions are

4138: the same as the ones in the original theory except for a factor of

4139: $e^{-a}$. This in particular shows that the right hand side of \refb{eeb3}

4140: vanishes for $a\to\infty$ and hence in this limit we have a solution of

4141: the equations of motion. In order to find the value of the action as

4142: $a\to\infty$ we note that, using $\{Q_B, c(z)\} = c \p c(z)$, \refb{eeb3} 

4143: gives

4144:  \be \label{eeb8}

4145: {\p S_B \over \p a} =  -{\TT_p\over 2} \, a \, e^{-a} \, \int

4146: {d\theta\over

4147: 2\pi} \, \int {d\theta'\over 2\pi} \, \la c(\theta)\, c\p

4148: c(\theta')\ra\, ,

4149:  \ee

4150: where the correlation function on the right hand side now has to be

4151: calculated in the original theory.

4152: Using

4153:  \be \label{eeb9}

4154: \la c(\theta)\, c\p

4155: c(\theta')\ra = - V_{p+1} \, |e^{i\theta} - e^{i\theta'}|^2 = - 4

4156: V_{p+1} \, \sin^2{\theta -

4157: \theta'\over 2} \, ,

4158:  \ee

4159: we get

4160:  \be \label{eeb10}

4161: {\p S_B \over \p a} = \TT_p\, V_{p+1} \, a \, e^{-a}\, .

4162:  \ee

4163: This gives\cite{0009103,0009148}

4164:  \be \label{eeb11}

4165: S_B(a) = - V_{p+1} \, \TT_p \, (a+1) \, e^{-a}\, ,

4166:  \ee

4167: up to an additive constant.

4168: Since $-S_B / V_{p+1}$ can be identified as the potential, we see that the

4169: tachyon potential as a function of $a$ is given by

4170:  \be \label{eeb12}

4171: \VV(a) = \TT_p \, (a+1) \, e^{-a}\, .

4172:  \ee

4173: Thus

4174: the difference between the values of the potential at the maximum $a=0$

4175: and the local minimum at $a=\infty$ is given by $\TT_p$,

4176: in accordance with the conjecture 1.

4177: 

4178: In order to see how a D-$q$-brane may be obtained as a

4179: solution of the BSFT equations of motion on a D-$p$-brane,

4180: we consider a

4181: more general tachyon profile\cite{0009148}

4182:  \be \label{eeb13}

4183: T = a + \sum_{i=q+1}^p u_i (x^i)^2 \, .

4184:  \ee

4185: This corresponds to a boundary deformation of the form \refb{eeb4} with

4186:  \be \label{een14}

4187: V = a + \sum_{i=q+1}^p u_i

4188: (X^i)^2\, .

4189:  \ee

4190: Since this term is quadratic in the fields, the world-sheet theory is

4191: still exactly solvable. In particular on the unit

4192: disk the

4193: normalized two point function of the $X^i$'s in the presence

4194: of this boundary deformation is

4195: given by\cite{9210065}:

4196:  \ben \label{e2pta}

4197: \GG_{ij}\left(\theta, \theta'\right)  &\equiv& \la X^i(

4198: \theta) X^j(\theta') c\p c\p^2

4199: c(P)\ra_V /

4200: \la c\p c\p^2 c(P) \ra_V \nonumber \\

4201: &=& \left[-2\ln\Big|1 -

4202: e^{i(\theta'-\theta)}\Big| +

4203: {1\over 2

4204: u_i} - 4

4205: u_i \sum_{k=1}^\infty {1\over k (k+2u_i)} \,

4206: \cos\left(k(\theta-\theta')\right)\right]\, \delta_{ij}\, . \nonumber \\

4207:  \een

4208: Here $P$ is any point on the disk. The only role of the $c\p c\p^2c$

4209: factor in this

4210: expression is to soak up the ghost zero modes.

4211: Defining

4212:  \be \label{enorma}

4213: :(X^i(\theta))^2: = \lim_{\theta'\to \theta} \left(

4214: X^i(\theta')

4215: X^i(\theta) +

4216: 2\, \ln

4217: \Big| 1 -

4218: e^{i(\theta'-\theta)}\Big |\right)

4219:  \ee

4220: we get

4221:  \be \label{enormvev-}

4222: \la :(X^i(\theta))^2: c\p c\p^2 c(P)\ra_V / \la c\p c\p^2 c(P) \ra_V

4223: = {1\over 2

4224: u_i} - 4

4225: u_i \sum_{k=1}^\infty {1\over k (k+2u_i)} \, .

4226:  \ee

4227: This may be rewritten as\cite{9210065}

4228:  \be \label{enormvev}

4229: \la :(X^i(\theta))^2: c\p c\p^2 c(P)\ra_V / \la c\p c\p^2 c(P) \ra_V

4230: = -

4231: {d\over du_i} \ln

4232: Z_1(2u_i)\, ,

4233:  \ee

4234: where

4235:  \be \label{eeb15a}

4236: Z_1(v) = \sqrt{v} e^{\gamma v} \Gamma(v)  \, .

4237:  \ee

4238: $\gamma$ is the Euler number.

4239: The

4240: partition

4241: function $\la  c\p c\p^2 c(P)\ra_V$ on the disk in the presence of

4242: the boundary

4243: deformation $\int{d\theta\over 2\pi} V(\theta)$ with $V(\theta)$

4244: given by \refb{een14} is now obtained by solving the equation:

4245:  \be \label{epdiff}

4246: {\p \la  c\p c\p^2 c(P) \ra_V \over \p u_i} = -\int {d\theta\over 2\pi}

4247: \, \la :(X^i(\theta))^2:  c\p

4248: c\p^2 c(P)\ra_V = \la  c\p c\p^2 c(P) \ra_V \, {d\over du_i} \ln Z_1(2u_i)

4249: \, .

4250:  \ee

4251: A solution to this equation

4252: is given by

4253:  \be \label{epart}

4254: -{1\over 2} \, \la  c\p c\p^2 c(P)\ra_V = e^{-a} \, V_{q+1} \,

4255: (2\pi)^{(p-q)/2}  \,

4256: \prod_{i=q+1}^p Z_1(2u_i)\, ,

4257:  \ee

4258: where

4259: $V_{q+1}$ denotes the volume of the $(q+1)$ dimensional space-time

4260: spanned by $x^0$, $x^1$, $\ldots$ $x^q$.

4261: Note that the differential equation \refb{epdiff} determines

4262: only the $u_i$ dependence of the right hand side of

4263: \refb{epart}, and does not determine its dependence on $a$ or

4264: the overall normalization. The $a$ dependence is fixed trivially by

4265: noting that its presence

4266: gives rise to an additive constant $a$ in the world-sheet action and

4267: hence a multiplicative factor of $e^{-a}$ in any unnormalized

4268: correlation function.

4269: The constant part of the normalization factor is determined by

4270: requiring that the answer reduces to $V_{p+1}$ for $a=u_i=0$ in

4271: accordance with \refb{eghno}. This is however somewhat subtle since the 

4272: right hand side of \refb{epart} contains a factor of $V_{q+1}$ instead of 

4273: $V_{p+1}$. The reason for this

4274: is

4275: that for non-zero $u_i$'s the integration over the zero modes $\int d x^i$ 

4276: for $i=q+1,\ldots p$

4277: is replaced by:

4278:  \be \label{eeb16}

4279: \int d x^i e^{-u_i (x^i)^2} = \sqrt{\pi/u_i}\, .

4280:  \ee

4281: Thus in the $u_i\to 0$ limit, we should make the replacement

4282:  \be \label{ereplacem}

4283: V_{q+1}

4284: \prod_{i=q+1}^p \sqrt{\pi/u_i}\quad\longrightarrow\quad V_{p+1}\, .

4285:  \ee

4286:  With this

4287: prescription

4288: the right hand side of \refb{epart} reduces to $V_{p+1}$ for $a\to 

4289: 0$, $u_i\to 0$ in

4290: accordance

4291: with \refb{eghno}.

4292: 

4293: In order to calculate

4294: $S_B(a,\vec u)$ using \refb{eeb3} for the field configuration

4295: \refb{eeb13} we need to compute $\{Q_B\, , \,

4296: c(\theta')(X^i(\theta'))^2\}$. This is

4297: done using \refb{elapl}. Since

4298: $\square_X (X^i)^2=2$, we get

4299:  \be \label{ethird}

4300: \{Q_B\, , \,

4301: c(\theta')(X^i(\theta'))^2\} =  c\p c(\theta') \left(2 +

4302: (X^i(\theta'))^2\right)\, .

4303:  \ee

4304: Thus \refb{eeb3} gives

4305:  \be \label{ebnew1}

4306: {\p S_B \over \p u_i} = -{\TT_p\over 2} \, \int

4307: {d\theta\over

4308: 2\pi} \, \int {d\theta'\over 2\pi} \, \left \la c(\theta)

4309: (X^i(\theta))^2 \, c\p

4310: c(\theta') \left(

4311: a + \sum_{i=q+1}^p \, u_i \left(2+(X^i(\theta'))^2\right)\right)

4312: \right\ra_V \, .

4313:  \ee

4314: Also

4315:  \be \label{ebnew2}

4316: {\p S_B \over \p a} = -{\TT_p\over 2} \, \int

4317: {d\theta\over

4318: 2\pi} \, \int {d\theta'\over 2\pi} \, \left\la c(\theta) \, c\p

4319: c(\theta')  \left(

4320: a + \sum_{i=q+1}^p \, u_i \left(2+(X^i(\theta'))^2\right)\right)

4321: \right\ra_V\, .

4322:  \ee

4323: The right hand side of \refb{ebnew2} can be easily evaluated using

4324: eqs.\refb{enormvev}, \refb{epart}. For example, we have

4325:  \be \label{eextran1}

4326: \la c(\theta) c\p c(\theta')\ra_V = 2\, \sin^2\left({\theta - \theta'\over

4327: 2}\right) \, \la c \p c \p^2 c(P)\ra_V\, ,

4328:  \ee

4329:  \be \label{eextran2}

4330: \left\la c(\theta) c\p c(\theta') \left( X^i(\theta')\right)^2\right\ra_V

4331: = 2\,

4332: \sin^2\left({\theta - \theta'\over

4333: 2}\right) \, \la c \p c \p^2 c(P)\ra_V \left( -{d\over d u_i}\ln Z_1(2

4334: u_i)\right)\, .

4335:  \ee

4336: The correlation function appearing in \refb{ebnew1} can also be evaluated

4337: with the help of Wick's theorem

4338: using the propagator \refb{e2pta} for the $X^i$-fields. The

4339: result for $S_B$, obtained after integrating eqs.\refb{ebnew1},

4340: \refb{ebnew2} is given by\cite{9210065}

4341:  \be \label{eeb15}

4342: S_B(a, \vec u) = -\TT_p \, V_{q+1} \, (2\pi)^{(p-q)/2}\, \left( a + 1 +

4343: 2\sum_i u_i - \sum_i

4344: u_i

4345: {\p\over

4346: \p

4347: u_i}\right) e^{-a} \prod_i Z_1(2 u_i)

4348: \, .

4349:  \ee

4350: 

4351: In order to find a solution of the equations of motion with non-vanishing

4352: $\{u_i\}$, we first solve for $\p S_B/\p a=0$. This gives:

4353:  \be \label{eeb17}

4354: a = \left(-2 \sum u_i + \sum u_i {\p\over \p u_i}\right) \ln Z_1(2

4355: u_i) \,

4356: .

4357:  \ee

4358: Substituting this back into \refb{eeb15} gives

4359:  \be \label{eeb18}

4360: S_B = -\TT_p \, V_{q+1} \, (2\pi)^{(p-q)/2} \, \prod_i \, \exp\left[ 2 u_i

4361: - u_i {\p\over \p

4362: u_i} \ln Z_1(2 u_i) + \ln Z_1(2 u_i) \right] \, .

4363:  \ee

4364: Using the definition of $Z_1(v)$ given in \refb{eeb15a} one can show

4365: that $-S_B$ is a monotone decreasing function of $u_i$ and

4366: hence has minima at $u_i=\infty$. Using Stirling's formula one gets, for

4367: large $u$,

4368:  \be \label{estir}

4369: \ln Z_1(2u) \simeq 2u \ln(2u) - 2u + 2\gamma u + \ln\sqrt{2\pi}\, .

4370:  \ee

4371: This gives, at $u_i=\infty$,

4372:  \be \label{eeb19}

4373: S_B = -\TT_p \, V_{q+1} \, (2\pi)^{p-q} \, .

4374:  \ee

4375: This describes a $q$-dimensional brane along $x^1,\ldots x^q$ with tension

4376:  \be \label{eeb20}

4377: \TT_p \,(2\pi)^{p-q} \, .

4378:  \ee

4379: This is precisely the tension of a D-$q$-brane. Thus we see that BSFT on a

4380: D-$p$-brane

4381: contains as a classical solution D-$q$-branes with $q\le p$, in accordance

4382: with the third conjecture.

4383: 

4384: Various other aspects of open string tachyon condensation in boundary 

4385: string field theory have been discussed in 

4386: \cite{0010021,0010218,

4387: 0011009,0011002,0011033,0012089,

4388: 0012198,0012210,0101002,

4389: 0101087,0102063,

4390: 0103021,0103056,0103079,

4391: 0104099,

4392: 0104143,0104164,

4393: 0105076,0105098,

4394: 0105227,0105238,

4395: 0105245,0107098,0109032,

4396: 0109187,0203185,

4397: 0206025,0206102,

4398: 0211180,

4399: 0301179,

4400: 0302125,0308123,

4401: 0404237,0408157}.

4402: 

4403: \sectiono{Non-commutative Solitons} \label{snc}

4404: 

4405: In closed string theory constant anti-symmetric gauge field configuration

4406: $B_{\mu\nu}$ describes a pure gauge configuration. However this has

4407: non-trivial effect on the dynamics of a D-brane situated in such a

4408: background. For simplicity let us consider the case where only the spatial

4409: components of the $B_{\mu\nu}$ field are non-zero, and the closed

4410: string metric is the flat Minkowski metric. It was shown in

4411: \cite{9711165,9903205,9908142,0006071} that if $S(\{\phi_r\})$ denotes the 

4412: complete tree

4413: level open string field theory action on a D-brane in the absence of

4414: anti-symmetric tensor field background, with $\phi_r$

4415: denoting various

4416: components of string fields, then the open string field theory action in

4417: the presence of the anti-symmetric tensor field background is obtained by

4418: replacing

4419: in $S(\{\phi_r\})$ the original closed string metric $\eta_{\mu\nu}$

4420: by a new `open string metric' $G_{o\mu\nu}$, and

4421: all ordinary products between two field combinations $A$ and $B$ by

4422: the star product:\footnote{In order to make sense of this prescription

4423: we need to know the precise order in which we should arrange the open

4424: string fields before replacing the ordinary product by star product.

4425: The correct prescription is to begin with a system of $N$ coincident

4426: D-branes so that each field is replaced by an $N\times N$ matrix

4427: valued field and ordinary products of fields get replaced by trace

4428: over product of these matrix valued fields in a given cyclic order. In

4429: order to

4430: represent the effect of a background anti-symmetric field

4431: configuration in terms of non-commutative field theory, we arrange the

4432: fields in the same cyclic order even for $N=1$ and then replace the

4433: ordinary product by star product.}

4434:  \be \label{estar1}

4435: A * B \equiv \exp\left( {i\over 2} \theta^{\mu\nu} \p_{x^\mu}

4436: \p_{y^\nu}\right)

4437: A(x)

4438: B(y)\bigg|_{x=y}\, .

4439:  \ee

4440: The anti-symmetric tensor $\theta^{\mu\nu}$

4441: and the symmetric open string metric

4442: $G_o^{\mu\nu}$ along the D-brane

4443: world-volume are

4444: given by the equation:

4445:  \be \label{estar2}

4446: G_o^{\mu\nu} + {1\over 2\pi} \, \theta^{\mu\nu} = \left((\eta +

4447: B)^{-1}\right)_{\mu\nu} \, .

4448:  \ee

4449: Here $B_{\mu\nu}$ on the right hand side is the pull-back of the

4450: anti-symmetric tensor field on the D-brane world-volume. There is also

4451: a change in the effective string coupling constant\cite{9908142}

4452: causing a change in the overall normalization of the action.

4453: We shall determine this indirectly through eq.\refb{eceq} below.

4454: 

4455: A simplification occurs in the limit of large $B_{\mu\nu}$.

4456: For

4457: simplicity let us consider the case where only the $1-2$ component of

4458: $B_{\mu\nu}$ is non-zero, with $x^1$ and $x^2$ being directions tangential

4459: to the D-brane, and we take $B_{12}\equiv B$ to

4460: be large. In this limit:

4461:  \be \label{estar3}

4462: G_o^{11} = G_o^{22} \simeq {1\over B^2}, \qquad G_o^{12} =

4463: G_o^{21}=0,

4464: \qquad

4465: \theta^{12} = -\theta^{21} \simeq -{2\pi\over B}\, .

4466:  \ee

4467: Note that in the $B\to\infty$ limit $G_o^{ij}/\theta^{12}$ vanishes for

4468: $i,j=1,2$. Since after replacing $\eta^{\mu\nu}$ by $G_o^{\mu\nu}$

4469: in $S(\{\phi_r\})$

4470: all explicit derivatives with respect to the $x^1$

4471: and $x^2$

4472: coordinates in

4473: $S(\{\phi_r\})$

4474: are contracted with the metric $G_o^{ij}$, we see that such derivative

4475: terms may be neglected in the large $B$ limit. If we focus on classical

4476: solutions of the effective action which depend only on these two

4477: coordinates, then we can simply drop all explicit derivative terms

4478: from the effective action, keeping only the implicit derivatives coming

4479: through the definition of the star-product \refb{estar1}. In other words

4480: if $-\int d^{p+1} x \, V(\{\phi_r\})$ denotes the part of the original

4481: action

4482: $S(\{\phi_r\})$ without any

4483: derivative

4484: terms,

4485: then in order to look for solutions which depend only on $x^1$ and

4486: $x^2$ in the presence of a strong anti-symmetric tensor field in the

4487: 1-2 plane, we can work with

4488: the string field theory action:

4489:  \be \label{estar4}

4490: -C\int d^{p+1} x \, V_*(\{\phi_r\})\, ,

4491:  \ee

4492: where the subscript $*$ in $V$ denotes that all the products of fields

4493: inside $V$ are to be interpreted as star product, and the constant

4494: $C$ takes into account the change in the overall normalization of the

4495: action due to the $\sqrt{-\det G_o}$ term that will multiply the

4496: Lagrangian density, and the overall change in the normalization of the

4497: action due to the change in the string coupling constant. The

4498: equations of motion now are:\footnote{In order to determine how the

4499: fields are arranged inside $(\p_sV)_*$, we begin with a term in 

4500: $V_*(\{\phi_r\})$ and, using the cyclicity of the star product under the 

4501: integral, bring the $\phi_s$ factor with respect to which we are 

4502: differentiating to the extreme left. The derivative of this term with 

4503: respect to $\phi_s$ is then given by removing the $\phi_s$ factor from the 

4504: string, leaving the rest of the terms in the same order.}

4505:  \be \label{estar5}

4506: (\p_s V)_*(\{\phi_r\}) = 0\, .

4507:  \ee

4508: For later convenience we shall choose the string field $\{\phi_r\}$

4509: such

4510: that $\{\phi_r=0\}$

4511: denotes the tachyon vacuum, and $V(\{\phi_r\})$ vanishes at

4512: $\{\phi_r=0\}$. This convention is different from the one used in

4513: section \ref{s4} for example, where $\{\phi_r=0\}$ describes the

4514: original D-brane on which the string field theory is formulated.

4515: 

4516: A general method for constructing soliton solutions to

4517: eq.\refb{estar5} was developed in \cite{0003160}. Suppose

4518: $\{\phi^{(0)}_r\}$ denotes a translationally invariant

4519: configuration corresponding to a local extremum of the tachyon

4520: potential:

4521:  \be \label{estar6}

4522: \p_s V(\{\phi^{(0)}_r\}) = 0\, .

4523:  \ee

4524: In particular we can choose $\{\phi^{(0)}_r\}$ to be the configuration

4525: describing the original D-brane on which open string field theory is

4526: defined.

4527: Now suppose $f(x^1,x^2)$ denotes a function such that

4528:  \be \label{estar7}

4529: f * f = f\, .

4530:  \ee

4531: Then for

4532:  \be \label{estar8}

4533: \phi_r(x) = \phi^{(0)}_r \, f(x^1,x^2)\, ,

4534:  \ee

4535: we have

4536:  \be \label{epsvstar}

4537: (\p_sV)_*(\{\phi_r\}) = \p_s V(\phi^{(0)}_r) \, f(x^1, x^2) = 0\, .

4538:  \ee

4539: Thus \refb{estar8}

4540: is a solution of \refb{estar5}. For this solution

4541:  \be \label{estar}

4542: V_*(\{\phi_r(x)\}) = V(\{\phi^{(0)}_r\}) \, f(x^1,x^2)\, .

4543:  \ee

4544: Thus from eq.\refb{estar4} we see that the energy per unit

4545: $(p-2)$-volume associated

4546: with this solution, obtained by integrating the energy density

4547: $CV_*(\{\phi_r(x)\})$ over $x^1$ and $x^2$, is:

4548:  \be \label{estar9}

4549: C \, V(\{\phi^{(0)}_r\}) \, \int \, d

4550: x^1\, dx^2 \, f(x^1, x^2) \, .

4551:  \ee

4552: 

4553: This allows us to construct space dependent solutions by starting with a

4554: translationally invariant solution. We shall now show that if we take

4555: $\phi^{(0)}_r$ to be the translationally invariant solution

4556: describing

4557: the original D-brane, then for a suitable choice of $f$

4558: satisfying \refb{estar7}, \refb{estar8}

4559: represents a soliton solution describing a codimension 2

4560: D-brane\cite{0005006,0005031,0006071}.

4561: In this case $V(\{\phi^{(0)}_r\})$

4562: denotes the tension

4563: $\TT_p$ of

4564: the

4565: original D-brane on which we have formulated the string field theory

4566: in the absence of any background anti-symmetric tensor field. Also due

4567: to eq.\refb{estar4}

4568: $CV(\{\phi^{(0)}_r\})$ should describe the tension of the D-brane in

4569: the presence of the background $B$-field. Since in the presence of the

4570: $B$-field the tension of a D-brane gets multipled by a factor of

4571: $\sqrt{-\det(\eta+B)}$ which in the present example takes the value

4572: $\sqrt{1+B^2}\simeq B$ for large $B$, we have

4573:  \be \label{eceq}

4574: C\, V(\{\phi^{(0)}_r\}) = B\, \TT_p \, .

4575:  \ee

4576: Thus \refb{estar9}

4577: takes the form:

4578:  \be \label{estar10}

4579: B\, \TT_{p} \, \int \, d

4580: x^1\, dx^2 \, f(x^1, x^2) \, .

4581:  \ee

4582: This is

4583: the energy per unit $(p-2)$-volume of the solution \refb{estar8}.

4584: 

4585: An $f$ satisfying \refb{estar7} is given by\cite{0003160}:

4586:  \be \label{estar11}

4587: f(x^1, x^2) = 2\exp\left(-{B\over 2\pi} \left((x^1)^2 + (x^2)^2\right)

4588: \right)\, .

4589:  \ee

4590: This 

4591: can be 

4592: checked by rewriting \refb{estar7} in

4593: terms of Fourier transform of $f$, and noting that if $\wt A(k)$, $\wt 

4594: B(k)$ and $\wt{A*B}(k)$ denote the Fourier transforms of $A(x)$, $B(x)$ 

4595: and 

4596: $A*B(x)$ respectively, then eq.\refb{estar1} takes the form

4597:  \be \label{estarfourier}

4598:  \wt{A*B}(k) = \int {d^{p+1}q\over (2\pi)^{p+1}} \, \exp\left( -{i\over 

4599: 2} \, \Theta^{\mu\nu} q_\mu (k-q)_\nu\right) \wt A(q) \wt B(k-q)\, .

4600:  \ee

4601: In 

4602: this case

4603:  \be \label{etar12}

4604: \int d x^1 dx^2 f(x^1, x^2) = {4\pi^2\over B}\, ,

4605:  \ee

4606: and hence the tension of the codimension 2 solution, as given by

4607: \refb{estar10}, is

4608:  \be \label{estar13}

4609: 4\pi^2 \, \TT_p \, .

4610:  \ee

4611: This is precisely the tension of a D-$(p-2)$-brane. This shows

4612: that we can identify the non-commutative soliton given in

4613: eqs.\refb{estar8}, \refb{estar11} with a codimension two D-brane,

4614: in accordance with conjecture 3.\footnote{Note that this analysis

4615: does not provide a verification of conjecture 1, but given

4616: conjecture 1, it provides a verification of conjecture 3.}

4617: 

4618: 

4619: Eq.\refb{estar7} for $f$ is the requirement that $f$ is a projection

4620: operator under the star product. The $f$ given in \refb{estar11}

4621: describes a rank one projector\cite{0003160}. In general, if we take

4622: $f$ to be a rank $n$ projector, it describes a configuration of $n$

4623: D-$(p-2)$-branes. By making appropriate choice of $f$, one can

4624: construct multi-soliton

4625: solutions describing multiple D-branes located at arbitrary points in the

4626: transverse space\cite{0003160,0103256}. We shall not discuss this

4627: construction,

4628: and refer

4629: the interested reader to the original literature.

4630: 

4631: Various other aspects of non-commutative tachyon condensation have been 

4632: discussed in 

4633: \cite{9912274,0007226,0008013,0008023,0008214,

4634: 0009038,0009142,0010016,

4635: 0010028,0010058,0010060,0010101,0011094,

4636: 0012081,0007078,0007217,

4637: 0008064,0009002,

4638: 0009030,0009101,

4639: 0011079,0011090,

4640: 0011223,0012217,

4641: 0101001,

4642: 0101125,

4643: 0101145,0101199,

4644: 0104090,0104176,

4645: 0104263,0105115,

4646: 0106142,0301119,

4647: 0407229}.

4648: 

4649: \sectiono{Time Dependent Solutions} \label{s10}

4650: 

4651: In this section we shall begin by outlining a general procedure for

4652: constructing time dependent solutions in string theory describing time

4653: evolution of a field (or a set of fields) from a given `initial

4654: condition', and then apply this to the construction of time dependent

4655: solutions describing the rolling of the tachyon on an unstable D-brane

4656: system away from the maximum of the potential. Our discussion will closely

4657: follow that in \cite{0207105}. Throughout this analysis we shall restrict 

4658: our attention to time independent closed string background for which the 

4659: matter part of the bulk CFT is given by the direct sum of the theory of a 

4660: free 

4661: scalar (super-)field $X^0$ representing the time coordinate and a 

4662: unitary conformal field theory of central charge $c=25$ ($\hat c=9$).

4663: 

4664: \subsection{General procedure}

4665: 

4666: Let us begin with some general unstable D-brane system with a tachyonic

4667: field $\phi$ of mass$^2=-m^2$. This could either describe the tachyons of

4668: the kind we discussed earlier, or more general tachyon, {\it e.g.} the

4669: tachyon on a D2-D0-brane system coming from open strings stretched from

4670: the D0 to the D2-brane. Since $\phi$ is tachyonic, its potential $V(\phi)$

4671: has a

4672: maximum at $\phi=0$ with $V''(0)=-m^2$.

4673: If $\phi$ had been described by the

4674: action of a standard scalar

4675: field theory with two derivative kinetic term plus a potential term, then

4676: the motion of $\phi$ away from the maximum will be characterized by two

4677: parameters, the initial value of $\phi$ and its first time

4678: derivative.\footnote{For simplicity we are considering only spatially

4679: homogeneous field configurations here.}$^,$\footnote{One of these

4680: parameters can be

4681: fixed using the time translation invariance of the system; we simply

4682: choose the origin of time where either the field or its first time

4683: derivative vanishes.}

4684: However, given that string field theory action has inifinite number of

4685: time (and space) derivatives, it is not {\it a priori} clear if such a

4686: set of solutions can be constructed in string theory as well. We shall now

4687: argue that it is indeed possible to construct a similar set of solutions

4688: in string theory; and give an algorithm for constructing these solutions.

4689: 

4690: The construction will be carried out by using the well-known

4691: correspondence between the solutions of classical equations of

4692: motion in string theory and two dimensional conformal field

4693: theories. Boundary conformal field theories associated with time dependent 

4694: open string field configurations involve boundary interaction terms which 

4695: depend on the time coordinate field $X^0$ in a non-trivial manner. 

4696: Since such conformal field theories are difficult to analyze directly,

4697: we shall first construct a solution that

4698: depends non-trivially on a space-like coordinate $x$ and then

4699: replace $x$ by $i x^0$. The new

4700: configuration will represent a time dependent solution of the

4701: equations of motion. All we need to ensure is that the solution

4702: obtained this way is real.

4703: 

4704: Whereas this gives a general procedure for constructing time

4705: dependent solutions in string theory, we are looking for a

4706: specific kind of time dependent solution, -- that which describes

4707: the rolling of $\phi$ away from the maximum of $V(\phi)$. So the

4708: next question is: which particular euclidean solution should we

4709: begin with in order to generate such time dependent solutions? The

4710: clue to this answer comes from looking at the solution of the

4711: linearized equation of motion near the maximum of the potential.

4712: Since the higher derivative terms in string field theory are all

4713: in the interaction term, they do not affect the linerarized

4714: equation of motion for $\phi$, which takes the standard form:

4715:  \be \label{eff1}

4716: (\p_0^2 + m^2)\phi \simeq 0\, .

4717:  \ee

4718: Using time translation invariance we can choose the boundary condition on

4719: $\phi$ to be either

4720:  \be \label{eff2}

4721: \phi=\lambda, \quad \p_0 \phi=0, \quad \hbox{at} \quad x^0=0,

4722:  \ee

4723: or

4724:  \be \label{eff3}

4725: \phi=0, \quad \p_0 \phi=m\lambda, \quad \hbox{at} \quad x^0=0\, .

4726:  \ee

4727: For a conventional scalar field

4728: \refb{eff2} holds when the total energy density of the system is less than

4729: $V(0)$

4730: so that the field comes to rest at a point away from 0 and \refb{eff3}

4731: holds when the total energy density of the system is larger than $V(0)$ so

4732: that

4733: the field $\phi$ passes 0 with non-zero velocity during its motion. We 

4734: shall see that the same interpretation holds for open string field theory 

4735: as well. For

4736: the boundary condition \refb{eff2} the solution  to \refb{eff1}

4737: is given by:

4738:  \be \label{eff4}

4739: \phi(x^0) \simeq \lambda \cosh(m x^0)\, ,

4740:  \ee

4741: whereas for the boundary condition \refb{eff3} the solution

4742: is:

4743:  \be \label{eff5}

4744: \phi(x^0) \simeq \lambda \sinh(m x^0)\, .

4745:  \ee

4746: Both solutions are valid for small $\lambda$ and finite $x^0$ {\it i.e.}

4747: as long as $\phi$ is small. Thus the one parameter family of solutions of

4748: the full string field theory equations of motion that we are looking for

4749: must have the property that for small $\lambda$ it reduces to the form

4750: \refb{eff4} or \refb{eff5}.

4751: 

4752: For definiteness we shall from now on concentrate on the class of

4753: solutions

4754: with total energy density less than $V(0)$, but the analysis can be easily

4755: generalized to the other case.\footnote{In fact the solution of type

4756: \refb{eff5}

4757: can be obtained from those of type \refb{eff4} by the formal replacement

4758: $x^0\to x^0 - i\pi/2m$, $\lambda\to i\lambda$.} We now note that

4759: \refb{eff4} can be obtained from the Euclidean

4760: solution:

4761:  \be \label{eff6}

4762: \phi(x) 

4763: \simeq \lambda \cos(mx)\, ,

4764:  \ee

4765: under the replacement $x\to ix^0$. Thus we need to search for a one

4766: parameter family of euclidean solutions which for small value of the

4767: parameter $\lambda$, reduces to \refb{eff6}. Given such a one parameter

4768: family of solutions, we can construct one parameter family of time

4769: dependent solutions by the replacement $x\to ix^0$.\footnote{The original

4770: idea of constructing time dependent solution in open string theory by Wick 

4771: rotating Euclidean

4772: solution is due to ref.\cite{0202210}.}

4773: 

4774: To proceed further, we shall, for definiteness, concentrate on the bosonic

4775: string theory, although the analysis can be easily generalized to the

4776: superstring theory. Since the tachyon has mass$^2=-m^2$,

4777: the zero momentum tachyon vertex

4778: operator $V_\phi$ must have dimension $(1-m^2)$ so that $V_\phi e^{ik.X}$

4779: has dimension 1 for $k^2=m^2$. For small $\lambda$,

4780: switching on

4781: the background \refb{eff6} corresponds to deforming the original boundary

4782: CFT by a boundary perturbation of the form:

4783:  \be \label{eff7}

4784: \lambda \, \int dt \, V_\phi(t) \, \cos(m X(t))\, ,

4785:  \ee

4786: where $t$ denotes a parameter labelling the boundary of the world-sheet.

4787: Since $V_\phi \cos(mX)$

4788: has conformal dimension 1, the deformed theory represents a conformal

4789: field theory to first order in $\lambda$. In order to construct a one

4790: parameter family of solutions of the string field theory equations of

4791: motion, we need to construct a family of conformal field theories labelled

4792: by $\lambda$, which to first order in $\lambda$ agrees with the deformed

4793: theory

4794: \refb{eff7}.

4795: 

4796: In special cases, $V_\phi\cos(mX)$ may represent an exactly

4797: marginal operator in string field theory, in which case \refb{eff7}

4798: represents a conformal field theory even for finite $\lambda$, and our

4799: task

4800: is over. However even in this case, for finite $\lambda$ the parameter

4801: $\lambda$ appearing

4802: in \refb{eff7} may not agree exactly with the value of $T$ at $x^0=0$

4803: (which in turn depends on the precise definition of the tachyon field, and

4804: varies between different formulations of string field theory for example.)

4805: Thus it is more appropriate to label the perturbation as

4806:  \be \label{eff7a}

4807: \tl \, \int dt \, V_\phi(t) \, \cos(m X(t))\, ,

4808:  \ee

4809: with $\tl=\lambda+\OO(\lambda^2)$.

4810: 

4811: In the

4812: generic case $V_\phi\cos(mX)$ is not an

4813: exactly marginal operator since its $\beta$-function $\beta_{\tl}$ will

4814: have

4815: contribution

4816: of order $\tl^3$ and higher.\footnote{Note that due to $X\to X+\pi/m$,

4817: $\tl\to-\tl$ symmetry of the deformation,  $\beta_{\tl}$ receives

4818: contribution only to odd orders in $\tl$.} In this case we proceed as

4819: follows. Instead of considering the deformation \refb{eff7a}, we consider

4820: deformation by the operator:

4821:  \be \label{eff8}

4822: \wt \lambda \, \int dt \, V_\phi(t) \, \cos(\omega X(t))\, ,

4823:  \ee

4824: where $\omega$ is a constant that will be fixed shortly. Since the

4825: perturbing operator has dimension $(\omega^2-m^2+1)$, $\beta_{\tl}$ now

4826: receives contribution linear in $\tl$, and the full $\beta$-function is

4827: given by:

4828:  \be \label{eff9}

4829:  \beta_{\tl} = (\omega^2 - m^2) \tl + g(\omega, \tl)\, ,

4830:  \ee

4831: where $g(\omega, \tl)$ denotes higher order contribution to the

4832: $\beta$-function.

4833: We now adjust $\omega$ so that the right hand side of \refb{eff9} 

4834: vanishes.

4835: Since $g(\omega, \tl)\sim \tl^3$, we can get a solution of the form:

4836:  \be \label{eff10}

4837: \omega=m + \OO(\tl^2)\, .

4838:  \ee

4839: 

4840: This gives a way to generate a one parameter family of boundary CFT's

4841: labelled by $\tl$, which for small $\tl$ corresponds to the solution

4842: \refb{eff6} with $\lambda\simeq \tl$. Given this conformal field theory,

4843: we can calculate the

4844: energy momentum tensor and sources of other massless fields like the

4845: dilaton, anti-symmetric tensor field etc. from the boundary state

4846: associated with this BCFT. (If the deformed boundary CFT is solvable then

4847: we can find an exact expression for the boundary state as in section

4848: \ref{s3.4}; otherwise we may need

4849: to compute it as a perturbation series in $\tl$.)

4850: 

4851: Note that in the above analysis we have not included the

4852: $\beta$-functions for any operator other than the original perturbing

4853: operator. In order to get a conformal field theory we need to ensure

4854: that the $\beta$-function for every other operator also vanishes. What we

4855: have done here is to implicitly assume that the other operators have been

4856: `integrated out' (in a space-time sense) so that we can talk in terms of

4857: the `effective' $\beta$-function of this single operator

4858: $V_\phi\cos(\omega X)$. To see explicily what this means

4859: let $\{\OO_i\}$ denote a complete set of boundary operators in the theory

4860: other than $V_\phi\cos(\omega X)$,

4861: and let

4862: $h_i$ be the conformal weight of $\OO_i$. We add to the action the

4863: boundary term

4864:  \be \label{eff11}

4865: \sum_i \, g_i\,  \int dt \, \OO_i(t)\, ,

4866:  \ee

4867: besides \refb{eff8}. Then the $\beta$-function $\beta_i$ of $g_i$ and

4868: $\beta_{\tl}$ have the form:

4869:  \be \label{eff12}

4870: \beta_i = (h_i - 1) g_i + F_i(\omega, \vec g, \tl)\, , \qquad

4871: \beta_{\tl} = (\omega^2 - m^2) \tl + F(\omega, \vec g, \tl)\, .

4872:   \ee

4873: $F_i$ and $F$ contains terms quadratic and higher order in $g_i$ and

4874: $\tl$. {}From

4875: this we see that as long as $h_i \ne 1$, we can solve for the $g_i$'s and

4876: $(\omega-m)$ in

4877: power series in $\tl$ beginning at quadratic or higher order. In 

4878: particular we can first solve the $\beta_i=0$ equations to express each 

4879: $g_i$ as 

4880: a function of $\tl$, and substitute this back into the expression for 

4881: $\beta_{\tl}$ to get an equation of the form \refb{eff9}. This procedure 

4882: of `integrating out' the other operators

4883: breaks down if there are operators of dimension $\simeq 1$ other than

4884: $V_\phi\cos(\omega X)$ since in this case we can no longer solve the 

4885: $\beta_i=0$ equations to express $g_i$ as a power series in $\tl$.

4886: This difficulty is of course a

4887: reflection of the well known difficulty in integrating out the massless

4888: fields.

4889: 

4890: In particular we can consider operators $\p X$ or $V_\phi \sin(\omega X)$

4891: both of

4892: which will have dimension $\simeq 1$. Fortunately both these operators are

4893: odd under $X\to -X$, and hence are not generated in the operator product

4894: of

4895: $V_\phi \cos(\omega X)$ with itself, since the latter operator is even

4896: under $X\to -X$. In the generic case we do not expect any other dimension

4897: $\simeq 1$ operator to appear in the operator product of $V_\phi

4898: \cos(\omega X)$ with itself, but there may be special cases where such

4899: operators do appear.\footnote{In fact the general method outlined here

4900: also holds

4901: for generating time dependent solutions in closed string theory, but

4902: the dimension (1,1) operators associated with the zero momentum

4903: graviton/dilaton vertex operators cause obstruction to this procedure. It

4904: may be possible to avoid this problem in

4905: case of localized tachyons\cite{0108075}, since in this case the rolling

4906: tachyon

4907: generates a localized source for the bulk graviton/dilaton field, and

4908: hence we can solve the equations of motion of these massless fields in the

4909: presence of these localized sources in order to get a consistent conformal

4910: field theory.}

4911: 

4912: In order to generate the time dependent solution, we now

4913: make the replacement

4914: $X\to iX^0$. This corresponds to deforming the original boundary CFT by

4915: the operator

4916:  \be \label{eff13}

4917: \tl\, \int dt \, V_\phi(t) \, \cosh(\omega X^0(t))\, .

4918:  \ee

4919: $\omega$ is determined in terms of $\tl$ by the same equation as in the

4920: euclidean case.\footnote{If we are using the picture in which the higher

4921: modes have not been integrated out,  then we

4922: also need

4923: to add perturbations generated by the Wick rotated version of

4924: \refb{eff11}. These pictures differ from each other by a choice of

4925: renormalization scheme for the world-sheet field theory.} The

4926: corresponding energy-momentum tensor and sources

4927: of other massless fields are also obtained by the making the replacement

4928: $x\to ix^0$ in the corresponding expressions in the euclidean theory.

4929: 

4930: This illustrates the general method for constructing a one parameter

4931: family of solutions labelled by $\tl$ which reduce to \refb{eff4} for

4932: small $\tl$. In order to generate another one parameter family of BCFT

4933: labelled by $\tl$ which reduce to \refb{eff5} for small $\tl$, we simply

4934: make a replacement $X^0\to X^0-i{\pi\over 2\omega}$, $\tl\to i\tl$ in the

4935: solution derived above. This corresponds to deforming

4936: the original BCFT by the operator:

4937:  \be \label{eff14}

4938: \tl\, \int dt \, V_\phi(t) \, \sinh(\omega X^0(t))\, ,

4939:  \ee

4940: and hence corresponds to a solution of the form \refb{eff5} for small 

4941: $\tl$ since $\omega\to m$ as $\tl\to 0$.

4942: 

4943: It is clear that the same method can be used to construct time

4944: dependent solutions describing rolling of tachyons on unstable D-brane

4945: systems in superstring theory as well.

4946: The method can also be generalized

4947: to describe simultaneous rolling of multiple tachyons\cite{0207105}.

4948: 

4949: \subsection{Specific applications}

4950: 

4951: We shall now apply the general method discussed in the last section to

4952: describe the rolling of spatially homogeneous tachyon field configuration

4953: on a D$p$-brane of bosonic string theory lying along $x^0,x^{26-p},\ldots 

4954: ,

4955: x^{25}$\cite{0203211,0203265}. In this case the vertex operator

4956: $V_\phi$ of the zero momentum tachyon is just the identity operator, and

4957: the tachyon has mass$^2=-1$. As a

4958: result, to lowest order in $\tl$, the analog of the perturbation

4959: \refb{eff7a}

4960: in the euclidean theory is given by:

4961:  \be \label{ehx1}

4962: \wt \lambda \, \int dt \, \cos(X(t))\, .

4963:  \ee

4964: This is identical to the perturbation \refb{e3.2} with $\tl$ identified to

4965: $-\al$ and represents an

4966: exactly marginal deformation of the original BCFT.\footnote{In the

4967: analysis of the perturbation \refb{e3.2}

4968: we took $X$ to be compact whereas here $X$, being related by Wick

4969: rotation to the time coordinate $X^0$, is non-compact. However the

4970: marginality

4971: of the operator does not depend on whether $X$ is compact or not.}

4972: As a result, the deformation \refb{ehx1} represents a BCFT for finite

4973: $\tl$ as well. Put another way, in this case eq.\refb{eff10} is replaced

4974: by

4975:  \be \label{ehx2}

4976: \omega = m=1\, ,

4977:  \ee

4978: for all $\tl$.

4979: 

4980: Making the replacement $X\to i X^0$ we see that the rolling

4981: tachyon solution is given by perturbing the original BCFT by the operator:

4982:  \be \label{ehx3}

4983: \wt \lambda \, \int dt \, \cosh(X^0(t))\, .

4984:  \ee

4985: The energy-momentum tensor $T_{\mu\nu}$ and the dilaton charge density $Q$

4986: associated with this solution can be obtained

4987: by making the replacement $X\to iX^0$, $X^0\to -i X^{25}$, $\al\to

4988: -\tl$ in

4989: \refb{efg21}, \refb{efg22}, \refb{edcharge}. This gives:

4990:  \ben \label{ehx4}

4991: && Q = \TT_{p} \, \wt f(x^0) \, \delta(x_\perp)\, ,

4992: \qquad T_{00} =  \TT_{p} \,

4993: \cos^2(\pi\tl) \, \delta(x_\perp), \quad

4994: \nonumber \\

4995: && T_{MN}

4996: = -\TT_{p}

4997: \,   \wt f(x^0) \, \delta_{MN} \, \delta(x_\perp) \quad \hbox{for}

4998: \quad (26-p)\le M,N\le 25\, ,

4999:  \een

5000: with all other components of $T_{\mu\nu}$ being zero.

5001: Here

5002:  \be \label{ehx5}

5003: \wt f(x^0) = f(i x^0)\big|_{\alpha=-\tl} = {1\over 1 + e^{x^0}

5004: \sin(\tl\pi)} + {1\over 1 + e^{-x^0} \sin(\tl\pi)} - 1\, .

5005:  \ee

5006: This reproduces \refb{e2.18}-\refb{e2.20} and \refb{ebosdil} after taking 

5007: into account the fact that the results in section \ref{s2} were quoted for 

5008: a D$p$-brane lying along $x^0, x^1, \ldots x^p$, whereas the results 

5009: obtained here are for D$p$-branes lying along $x^0$, $x^{26-p}$, $\ldots$ 

5010: $x^{25}$. 

5011: 

5012: As discussed in the previous subsection, the other one parameter family of

5013: solutions, valid when the total energy carried by the tachyon field

5014: configuration is larger than the tension of the D-brane, is obtained by

5015: making the replacement $x^0\to x^0 - i\pi/2$, $\tl\to i\tl$. Making these

5016: replacements in \refb{ehx4}, \refb{ehx5} gives the non-zero

5017: components of the energy momentum tensor and the dilaton charge

5018: density $Q(x)$

5019: to be

5020:  \ben \label{ehx4a}

5021: && Q = \TT_{p} \, \wt f(x^0) \, \delta(x_\perp)\, ,

5022: \qquad T_{00} = \TT_{p} \, 

5023: \cosh^2(\pi\tl) \, \delta(x_\perp),

5024: \nonumber \\

5025: && T_{MN} = -\TT_{p}

5026: \,   \wt f(x^0) \, \delta_{MN} \, \delta(x_\perp) \quad \hbox{for}

5027: \quad (26-p)\le M,N\le 25\, ,

5028:  \een

5029: where

5030:  \be \label{ehx6}

5031: \wt f(x^0) = {1\over 1 + e^{x^0} \sinh(\tl\pi)}

5032: + {1\over 1 - e^{-x^0} \sinh(\tl\pi)} - 1\, .

5033:  \ee

5034: This reproduces \refb{e2.18}, \refb{e2.22}, \refb{e2.23},

5035: \refb{ebosdil}.

5036: 

5037: The analysis in the case of D$p$-$\bar{\rm D}p$-brane pair in

5038: superstring theory proceeds in an identical manner. If the total

5039: energy per unit $p$-volume is less than $\EE_p$ where $\EE_p$ is

5040: the tension of the original brane system, we need to switch on a

5041: tachyon background\footnote{The $\sqrt 2$ multiplying $\tl$ is

5042: part of the normalization convention for $\tl$.}

5043:  \be \label{etacb}

5044: T=\sqrt 2 \, {\tl}\, \cosh(X^0/\sqrt 2)\, .

5045:  \ee

5046: This is related to the background \refb{efh5} for $R=1/\sqrt 2$ by

5047: the replacement

5048:  \be \label{ehx10}

5049:  x\to i x^0+ \pi/\sqrt 2\, , \qquad \alpha\to\tl\, ,

5050:  \ee

5051: and hence represents an exactly marginal deformation of the

5052: theory. Making these replacements in \refb{efg24}-\refb{esupdil},  we

5053: get the non-zero components of $T_{\mu\nu}$ and the dilaton charge density 

5054: $Q$ to be:

5055:  \ben \label{ehx11}

5056: T_{00} &=& \EE_p\, 

5057: \cos^2(\pi\tl) \, \delta(x_\perp), \nonumber \\

5058: T_{MN}

5059: &=& -\EE_p

5060: \,    \wt f(x^0) \,

5061: \delta_{MN} \, \delta(x_\perp) \quad \hbox{for} \quad (10-p)\le M,N

5062: \le 9\, , \nonumber \\

5063: Q(x) &=& \EE_p

5064: \,    \wt f(x^0) \,

5065:  \delta(x_\perp) \, ,

5066:  \een

5067:  \be \label{ehx12}

5068: \wt f(x^0) = {1\over 1 + e^{\sqrt 2 x^0} \sin^2(\tl\pi)}

5069: + {1\over 1 + e^{-\sqrt 2 x^0} \sin^2(\tl\pi)} - 1\, .

5070:  \ee

5071: This reproduces \refb{e2.11}-\refb{e2.13} and \refb{edilcharge} after 

5072: taking 

5073: into account the fact that the results of section \ref{s2} were quoted 

5074: for a brane-antibrane system along $x^0,\ldots x^p$ whereas here the 

5075: original brane-antibrane system is taken to be along $x^0$, 

5076: $x^{10-p},\ldots 

5077: x^9$. 

5078: 

5079: For total energy $>\EE_p$ we need to switch on a tachyon background

5080:  \be \label{eepge1}

5081:  T = \sqrt 2 \, {\tl}\, \sinh(X^0/\sqrt 2)\, .

5082:  \ee

5083: This is related to \refb{etacb} by the replacement $x^0\to

5084: x^0-i\pi/\sqrt 2$, $\tl\to i\tl$. Making these replacements in

5085: \refb{ehx11}, \refb{ehx12} gives

5086:  \ben \label{ehx13}

5087: T_{00} &=& \EE_p \, 

5088: \cosh^2(\pi\tl) \, \delta(x_\perp), \nonumber \\

5089: T_{MN} &=& -\EE_p

5090: \,    \wt f(x^0) \,

5091: \delta_{MN} \, \delta(x_\perp) \quad \hbox{for} \quad (10-p)\le M,N

5092: \le 9\, , \nonumber \\

5093: Q(x) &=& \EE_p

5094: \,    \wt f(x^0) \,

5095:  \delta(x_\perp) \, ,

5096:  \een

5097: with

5098:  \be \label{ehx14}

5099: \wt f(x^0) = {1\over 1 + e^{\sqrt 2 x^0} \sinh^2(\tl\pi)}

5100: + {1\over 1 + e^{-\sqrt 2 x^0} \sinh^2(\tl\pi)} - 1\, .

5101:  \ee

5102: All other components of $T_{\mu\nu}$ vanish.

5103: This reproduces \refb{e2.14} - \refb{edilcharge}.

5104: 

5105: Similar method can be used for getting the corresponding results

5106: for the non-BPS D-brane. In fact since a non-BPS D-brane is

5107: obtained by modding out a brane-antibrane system by $(-1)^{F_L}$,

5108: and since $T_{\mu\nu}$ is invariant under this transformation, the

5109: results for the sources for NS-NS sector fields produced by

5110: non-BPS D-brane are identical to those on a D-$\bd$ pair except

5111: for a change in the overall normalization. Information

5112: about the sources for the RR fields associated with the rolling

5113: tachyon background on a non-BPS D-brane, as given in

5114: eqs.\refb{errcharge}, can be obtained from eq.\refb{erreuclid} by

5115: the replacement \refb{ehx10}. \refb{errchargep} may be obtained

5116: from \refb{errcharge} by the replacement $x^0\to x^0-i\pi/\sqrt

5117: 2$, $\tl\to i\tl$.

5118: 

5119: The boundary state can also be used to determine the source terms

5120: for massive closed string states. For this we need to find the

5121: complete boundary state in the euclidean theory and replace $X$ by

5122: $iX^0$. We shall illustrate this in the context of bosonic string

5123: theory, but in principle the same procedure works for superstring

5124: theory. The $|\BB\ra_{c=25}$ and $|\BB\ra_{ghost}$ parts of the

5125: boundary state are given by eqs.\refb{efg11} and \refb{efg12}

5126: respectively. Thus the non-trivial part is $|\BB\ra_{c=1}$. The

5127: complete expression for $|\BB\ra_{c=1}$ for a general deformation

5128: parameter $\tl$ can be obtained by following the procedure

5129: outlined in section \ref{s3.4}\cite{9402113,9811237}.\footnote{The

5130: analysis of section \ref{s3.4} was carried out for compact $X$.

5131: The result for non-compact $X$ is obtained by simply dropping from

5132: the result of section \ref{s3.4} all terms in the boundary state

5133: carrying non-zero winding number.} In order to describe the

5134: results we need to review a few facts about the 

5135: spectrum

5136: of the $c=1$ conformal field theory described by a single

5137: non-compact scalar field $X$. This theory contains a set of

5138: Virasoro primary states labelled by SU(2) quantum numbers $(j,m)$

5139: and have the form\cite{DVV}\footnote{The underlying SU(2) group is 

5140: inherited from

5141: the theory with $X$-coordinate compactified on a circle of radius

5142: 1. Although SU(2) is not a symmetry of the theory when $X$ is

5143: non-compact, it is still useful to classify states with integer

5144: $X$-momentum. In fact the theory with compact $X$ actually has an

5145: SU(2)$_L\times$SU(2)$_R$ symmetry and the theory with non-compact

5146: $X$ contains a more general set of primary states carrying left

5147: and right SU(2) quantum numbers $(j,m)$ and $(j',m)$ respectively.

5148: The states with $j\ne j'$ will not be important for our

5149: discussion.}

5150:  \be \label{epriform}

5151: |j,m\ra = \wh \PP_{j,m} \, e^{2 \, i \, m \, X(0)} |0\ra\, ,

5152:  \ee

5153: where $\wh \PP_{j,m}$ is some combination of the $X$ oscillators

5154: of level $(j^2 - m^2, j^2 - m^2)$. Thus $|j,m\ra$ carries

5155: $X$-momentum $2m$ and $(L_0,\bar L_0)$ eigenvalue

5156: $(j^2,j^2)$. For any given primary state $|j,m\ra$ we have an

5157: associated Virasoro Ishibashi state\cite{ishibashi} $|j,m\ra\ra$, defined 

5158: as the 

5159: unique 

5160: linear combination of the primary $|j,m\ra$ and all Virasoro 

5161: descendants of $|j,m\ra$ such that $L_n - \bar L_{-n}$ 

5162: annihilate the

5163: state for every $n$. We shall normalize the Ishibashi states so

5164: that when we express $\exp\left(\sum_{n=1}^\infty \, {1\over n}

5165: \alpha_{-n}\bar\alpha_{-n}\right)e^{2im X(0)}|0\ra$ as a linear

5166: combination of the Ishibashi states built on various primaries,

5167: the state $|j,m\ra\ra$ 

5168: appears with coefficient 1 for every $j\ge |m|$:\footnote{In this 

5169: convention,

5170: the usual $\delta$-function normalized primary state $|j,m\ra$ appears in 

5171: $|j,m\ra\ra$ with unit coefficient\cite{0408064}.}

5172:  \be \label{edefjm}

5173: \exp\left(\sum_{n=1}^\infty \, {1\over

5174: n}

5175: \alpha_{-n}\bar\alpha_{-n}\right)e^{2im X(0)}|0\ra = \sum_{j\ge |m|}

5176: |j,m\ra\ra\, .

5177:  \ee

5178: 

5179: For the deformation \refb{ehx1}

5180: $|\BB\ra_{c=1}$

5181: may be expressed in terms of these Ishibashi states 

5182: as\cite{9402113,9811237,0402157}

5183:  \be \label{ebs1}

5184: |\BB\ra_{c=1} = \exp\left(\sum_{n=1}^\infty \, {1\over n}

5185: \alpha_{-n}\bar\alpha_{-n}\right)\, f(X(0)) \, |0\ra +

5186: |\wt\BB\ra_{c=1} \, ,

5187:  \ee

5188: where

5189:  \be \label{edeffagain}

5190: f(x) = {1\over 1 + \sin(\pi\tl) e^{ix}} + {1\over 1 +

5191: \sin(\pi\tl) e^{-ix}} - 1\, ,

5192:  \ee

5193:  \be \label{esecondform}

5194: |\wt\BB\ra_{c=1} = \sum_{j\ge 1} \, \sum_{m=-j+1}^{j-1} \,

5195: f_{j,m}(\tl) \, |j,m\ra\ra\, ,

5196:  \ee

5197:  \be \label{efjm}

5198: f_{j,m}(\tl) = D^j_{m,-m}(2\pi\tl) \, {(-1)^{2m} \over

5199: D^j_{m,-m}(\pi)} -

5200: (-1)^{2m} \, \sin^{2|m|}(\pi\tl)\, .

5201:  \ee

5202: $D^j_{m,m'}(\theta)$ are the representation matrices of the

5203: SU(2)

5204: group element $e^{i\theta\sigma_1/2}$ in the

5205: spin

5206: $j$

5207: representation.

5208: 

5209: 

5210: If $|\wh\BB\ra_{c=1}$ denotes the continuation of $|\wt\BB\ra_{c=1}$

5211: to

5212: the Minkowski space, then the complete boundary state $|\BB\ra$ in the

5213: Minkowski

5214: space may be expressed as:

5215:  \be \label{ebs2}

5216: |\BB\ra = |\BB_1\ra + |\BB_2\ra\, ,

5217:  \ee

5218: where

5219:  \ben \label{ebs3}

5220: |\BB_1\ra &=& \TT_p\,  \exp\left(-\sum_{n=1}^\infty \, {1\over n}

5221: \alpha^0_{-n}\bar\alpha^0_{-n}\right)\, \wt f(X^0(0))\, |0\ra \nonumber \\

5222: && \otimes

5223: \int {d^{25-p} k_\perp \over (2\pi)^{25-p}}\,

5224: \exp\left(\sum_{n=1}^\infty

5225: \sum_{s=1}^{25} (-1)^{d_s} \, {1\over

5226: n} \,  \alpha^s_{-n} \bar

5227: \alpha^s_{-n} \right) |\vec k_\parallel=0,

5228: \vec k_\perp\ra\, \nonumber \\

5229: && \otimes

5230: \exp\left(-\sum_{n=1}^\infty (\bar b_{-n} c_{-n} +

5231: b_{-n}

5232: \bar c_{-n})

5233: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\, ,

5234:  \een

5235: and

5236:  \ben \label{ebs4}

5237: |\BB_2\ra &=& \TT_p\, |\wh\BB\ra_{c=1} \, \otimes

5238: \int {d^{25-p} k_\perp \over (2\pi)^{25-p}} \,

5239: \exp\left(\sum_{n=1}^\infty

5240: \sum_{s=1}^{25} (-1)^{d_s} \,{1\over

5241: n} \,  \alpha^s_{-n} \bar

5242: \alpha^s_{-n} \right) |\vec k_\parallel=0,

5243: \vec k_\perp\ra\, \nonumber \\

5244: && \otimes

5245: \exp\left(-\sum_{n=1}^\infty (\bar b_{-n} c_{-n} +

5246: b_{-n}

5247: \bar c_{-n})

5248: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\, .

5249:  \een

5250: $d_s$ is an integer which is zero for Dirichlet directions

5251: and one for Neumann directions. $\tf(x^0)=f(ix^0)$ is given in 

5252: \refb{ehx5}.

5253: 

5254: $|\BB_1\ra$ produces source terms proportional to $\wt f(x^0)$ for

5255: various closed string fields, and hence these sources fall off to

5256: zero as $x^0\to\pm \infty$. On the other hand since

5257: $|\wt\BB\ra_{c=1}$ is a linear combination of Virasoro descendants

5258: over higher level primaries, and since these primaries occur at

5259: discrete values of momenta, these terms cannot be reorganized by

5260: summing over momenta as in eq.\refb{efg21}. Thus its Minkowski

5261: version $|\wh\BB\ra_{c=1}$ will involve linear combinations of

5262: Virasoro descendants of $\exp\left(\pm n X^0(0)\right)|0\ra$ for

5263: integer $n$ and $|\BB_2\ra$, which contains

5264: $|\wh\BB\ra_{c=1}$, will have source terms for various higher

5265: level closed string fields which grow exponentially

5266: \cite{0208196,0305177,0308172}. Thus $|\BB_2\ra$ has the form:

5267:  \be \label{ebb2}

5268: |\BB_2\ra = \TT_p\, \sum_{n=-\infty}^\infty \, \sum_{N=1}^\infty

5269: \, \int {d^{25-p} k_\perp \over (2\pi)^{25-p}} \, \wh\OO^{(n)}_N

5270: \, (c_0+\bar c_0) \, c_1 \, \bar c_1 \, e^{n X^0(0)} \, |k^0=0,

5271: \vec k_\parallel=0,\vec k_\perp\ra\, ,

5272:  \ee

5273: where $\wh\OO^{(n)}_N$ is some fixed combination of negative moded

5274: oscillators of total level $(N,N)$. We shall return to a

5275: discussion of these terms in sections \ref{s9} and \ref{s11}.

5276: 

5277: An interesting limit to consider is the $\tl\to {1\over 2}$ limit.

5278: As pointed out in eq.\refb{elhalf}, in this limit $\wt f(x^0)$

5279: vanishes, and hence $|\BB_1\ra$ vanishes. It is also easy to see

5280: using \refb{efjm} that $f_{j,m}(1/2)=0$. Thus in the

5281: $\tl\to{1\over 2}$ limit $|\wh\BB\ra_{c=1}$, its analytic

5282: continuation $|\wt\BB\ra_{c=1}$ and hence $|\BB_2\ra$ vanishes.

5283: This shows that at $\tl={1\over 2}$ not only the sources for 

5284: the

5285: massless closed string fields vanish, but the sources for all the

5286: massive closed string fields also vanish. This is consistent with

5287: the identification of the $\tl={1\over 2}$ point as the vacuum

5288: without any D-brane.

5289: 

5290: It is easy to verify that $|\BB_1\ra$ is BRST invariant, {\it i.e.}

5291:  \be \label{ebrinv}

5292: (Q_B+\bar Q_B) |\BB_1\ra = 0\, .

5293:  \ee

5294: Indeed we have the stronger relation

5295:  \ben \label{estrong}

5296: && (Q_B+\bar Q_B)\, \bigg[

5297: \exp\left(-\sum_{n=1}^\infty {1\over n}

5298: \alpha^0_{-n}\bar\alpha^0_{-n}\right)|k^0\ra \otimes

5299: \exp\left(\sum_{n=1}^\infty

5300: \sum_{s=1}^{25} (-1)^{d_s} \, {1\over

5301: n} \, \alpha^s_{-n} \bar

5302: \alpha^s_{-n} \right) |\vec k_\parallel=0,

5303: \vec k_\perp\ra\, \nonumber \\

5304: && \qquad \qquad \otimes

5305: \exp\left(-\sum_{n=1}^\infty (\bar b_{-n} c_{-n} +

5306: b_{-n}

5307: \bar c_{-n})

5308: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\,

5309: \bigg]\nonumber \\

5310: && = 0\, ,

5311:  \een

5312: for any $k^0$ and $\vec k_\perp$. Thus $(Q_B+\bar Q_B)|\BB_1\ra$,

5313: which may be expressed as

5314: a linear combination of the states appearing on the left hand side of

5315: \refb{estrong}, also vanishes. Since $|\BB\ra=|\BB_1\ra + |\BB_2\ra$ is

5316: BRST invariant, this shows that $|\BB_2\ra$ is also BRST invariant:

5317:  \be \label{eb2gi}

5318: (Q_B+\bar Q_B)|\BB_2\ra=0\, .

5319:  \ee

5320: 

5321: 

5322: {}From \refb{estrong} it follows that $|\BB_1\ra$  given in \refb{ebs3} is 

5323: BRST invariant for any choice of the function $\tf(x^0)$. Thus

5324: the BRST invariance of $|\BB_1\ra$

5325: does not impose any condition on the time dependence $\wt f(x^0)$

5326: of this boundary state. In contrast the time dependence of the boundary 

5327: state

5328: $|\BB_2\ra$ is fixed by the requirement of BRST

5329: invariance\cite{0402157,0408064} since in

5330: the $c=1$ conformal field theory the primary states $|j,m\ra$ for

5331: $|m|<j$ exist only for integer $X$-momentum $2m$. This suggests

5332: that the coefficients of $|j,m\ra\ra$ appearing in $|\BB_2\ra$ can

5333: be thought of as conserved charges\cite{0402157,0309074}. We shall 

5334: discuss this point in

5335: detail in the context of two dimensional string theory in section

5336: \ref{s11} where we shall also identify these charges with

5337: appropriate conserved charges in the matrix model description of

5338: the theory.

5339: 

5340: Before we conclude this section we note that in principle we should be

5341: able to study the time dependent solutions described here in string field

5342: theory by starting with the euclidean solution describing the lump or a

5343: kink on a

5344: circle of appropriate radius (1 for bosonic string theory and $1/\sqrt 2$

5345: for superstring) and then making an inverse Wick

5346: rotation\cite{0207107,0304163,0301137}. So far however

5347: this has

5348: not yielded any useful insight into the structure of these solutions.

5349: Various other approaches to studying these time dependent solutions in 

5350: string field theory have

5351: been discussed in 

5352: \cite{0208028,0209090,0302146,0306026,0308205,0406199,0409179}

5353: 

5354: Other aspects of time dependent classical solutions describing 

5355: rolling of the tachyon away from the maximum of the potential have been 

5356: discussed in refs.\cite{0205085,0303172,

5357: 0310253,

5358: 0306026,0308205,0311179}.

5359: 

5360: \sectiono{Effective Action Around the Tachyon Vacuum}

5361: \label{s5.3}

5362: 

5363: A question that arises naturally out of the studies in the previous

5364: sections is: Is it possible to describe the physics around the tachyon

5365: vacuum by a {\it low energy effective action}? Given that the tachyon

5366: field near the top of the potential has a mass$^2$ of the order of the

5367: string scale, one wouldn't naively expect any such action to exist.

5368: Furthermore, since around the tachyon vacuum we do not expect to get any

5369: physical open string states, there is no S-matrix with which we could

5370: compare the predictions of the effective action. Thus it would seem that

5371: not only is it unlikely that we have a low energy effective action

5372: describing

5373: the physics around the tachyon vacuum, but that the very question does not

5374: make sense due to the absence of physical states around such a vacuum.

5375: 

5376: We should keep in mind however that in string theory there is another way

5377: of checking the correctness of the classical effective action, -- namely

5378: by demanding that the solutions of the classical equations of

5379: motion derived from this effective

5380: action should

5381: correspond to appropriate conformal field theories. In particular for open

5382: string theory the classical solutions should be in one to one

5383: correspondence with conformally invariant boundary deformations of the

5384: world-sheet theory. As we

5385: saw in section \ref{s10}, there are families of known time dependent

5386: solutions around the tachyon vacuum labelled by the parameter $\tl$,

5387: and we could ask if it is possible to

5388: construct an effective action that reproduces these solutions.  Also the

5389: effective action must have the property that perturbative quantization of

5390: the theory based on this action should fail to give rise to any physical

5391: particle like states. We shall show that it is indeed possible to

5392: construct an effective action satisfying these criteria at a qualitative

5393: level. However

5394: we shall not be able to derive this effective action from first principles

5395: {\it e.g.} by comparison with any S-matrix elements, or make any

5396: definitive statement about the region of validity of this effective

5397: action. Some attempts to partially justify the validity of this effective 

5398: action has been made in 

5399: \cite{0304045,0401066}.

5400: 

5401: \subsection{Effective action involving the tachyon}

5402: 

5403: We begin with the purely tachyonic part of the action, ignoring

5404: the massless fields on the D-brane world-volume. In this case the

5405: proposed action around the tachyon vacuum is given

5406: by\cite{0003122,0003211,0003221,0204143,0209068,0303239,0304145,0307197,

5407: 0310138,0004106}:

5408:  \ben \label{ez1}

5409: S &=& \int d^{p+1} x \, \LL\, , \nonumber \\

5410: \LL &=& -  V(T) \, \sqrt{1 + \eta^{\mu\nu}\p_\mu T

5411: \p_\nu T} \, = \, - V(T) \, \sqrt{-\det A}

5412: \, ,

5413:  \een

5414: where

5415:  \be \label{ey2}

5416: A_{\mu\nu} = \eta_{\mu\nu} + \p_\mu T \p_\nu T\, .

5417:  \ee

5418: The potential $V(T)$ has a maximum at $T=0$ and has the asymptotic form

5419:  \be \label{ezz1}

5420: V(T) \simeq e^{-\alpha T/2}  \quad \hbox{for large $T$}\, ,

5421:  \ee

5422: with

5423:  \ben \label{e2}

5424: \alpha &=& 1 \qquad \hbox{for bosonic string theory} \nonumber \\

5425: &=& \sqrt 2 \qquad \hbox{for superstring theory}\, .

5426:  \een

5427: In this parametrization the potential has a minimum at infinity.

5428: The energy momentum tensor can be computed

5429: from the action \refb{ez1} by first minimally coupling it to a background

5430: metric, and then calculating the functional derivative of the action with

5431: respect to the background metric. The result is given

5432: by\footnote{In writing down the expression for the energy momentum

5433: tensor, it will be understood that these are localized on the plane of the

5434: brane by a position space delta function in the transverse coordinates.

5435: Also only the components of the energy-momentum tensor along the

5436: world-volume of the brane are non-zero.}

5437:  \be \label{etens}

5438: T_{\mu\nu} = {V(T) \, \p_\mu T \p_\nu T \over \sqrt{1 + \eta^{\rho\sigma}

5439: \p_\sigma T \p_\rho T}} - V(T)\,

5440: \eta_{\mu\nu}\, \sqrt{1 + \eta^{\rho\sigma}

5441: \p_\rho

5442: T \p_\sigma T}\, .

5443:  \ee

5444: 

5445: We shall first verify that the action \refb{ez1} produces the correct

5446: large

5447: $x^0$ behaviour of the pressure for

5448: spatially homogeneous, time dependent field configurations.

5449: For such configurations the conserved energy

5450: density is

5451: given

5452: by

5453:  \be \label{e2b}

5454: T_{00} = V(T) (1-(\p_0T)^2)^{-1/2}\, .

5455:  \ee

5456: Since $T_{00}$ is conserved, and $V(T)\to 0$ for large $T$, we see that

5457: for any given $T_{00}$,

5458: as $T\to\infty$, $\p_0 T\to 1$ and hence $T\to x^0~+$~constant. In 

5459: particular using \refb{ezz1} we can

5460: show that for large $x^0$ the

5461: solution has the

5462: form

5463:  \be \label{e2a}

5464: T = x^0 +  C e^{-\alpha x^0} + \OO(e^{-2\alpha x^0}) \, ,

5465:  \ee

5466: after shifting the origin of $x^0$ so as to remove the additive constant 

5467: that might otherwise appear in the expression for $T$.

5468: One way to check that \refb{e2a} gives the correct form of the solution is

5469: to note that the leading contribution to $T_{00}$

5470: computed

5471: from this configuration

5472: remains constant in time:

5473:  \be \label{e3}

5474: T_{00} \simeq {1\over

5475: \sqrt{2\alpha C}}\, .

5476:  \ee

5477: The pressure associated with this configuration is given

5478: by:

5479:  \be \label{e4}

5480: p= T_{11} = - V(T) (1-(\p_0T)^2)^{1/2} \simeq -\sqrt{2\alpha C}

5481: e^{-\alpha x^0}\, .

5482:  \ee

5483: Using the choice of $\alpha$ given in eq.\refb{e2} one can see that

5484: \refb{e4}

5485: is in precise agreement with the asymptotic forms of

5486: \refb{e2.20} and \refb{e2.23} for the bosonic string or

5487: \refb{e2.13} and \refb{e2.15} for the superstring for large $x^0$.

5488: In particular the pressure vanishes asymptotically.

5489: 

5490: 

5491: Given the success of the effective action in reproducing the

5492: asymptotic form of $T_{\mu\nu}$, it is natural to ask if it can

5493: also reproduce the sources for the dilaton and the RR fields

5494: associated with the rolling tachyon solution on an unstable

5495: D-brane. For this we need to know how the dilaton $\Phi_D$ and the

5496: RR $p$-form fields $C^{(p)}$ couple to this effective field

5497: theory. For a non-BPS D-brane of type II string theory, following

5498: coupling to $\Phi_D$, $C^{(p)}$ and the string metric $G_{\mu\nu}$

5499: seems to reproduce qualitatively the dilaton and RR source terms

5500: and $T_{\mu\nu}$ associated with a rolling tachyon solution:

5501:  \be \label{efullcoup}

5502: S = - \int d^{p+1} x \,  e^{-\Phi_D} \, V(T) \, \sqrt{-\det{\bf

5503: A}} + \int W(T)\, dT \wedge C^{(p)}\, ,

5504:  \ee

5505:  \be \label{efull1}

5506: {\bf A}_{\mu\nu} = G_{\mu\nu} + \p_\mu T \p_\nu T\, ,

5507:  \ee

5508: where $W(T)$ is some even function of $T$ which goes to zero

5509: asymptotically as $e^{-T/\sqrt 2}$\cite{9810188,9812135,9904207,0204143}. 

5510: The sources for $G_{\mu\nu}$,

5511: $\Phi_D$ and $C^{(p)}$ can be calculated by varying this action

5512: with respect to these closed string fields and then setting

5513: $G_{\mu\nu}=\eta_{\mu\nu}$, $\Phi_D=0$, $C^{(p)}=0$. The source

5514: for $G_{\mu\nu}$ gives us back the $T_{\mu\nu}$ given in

5515: \refb{etens} whereas the sources for $\Phi_D$ and $C^{(p)}$,

5516: evaluated for the asymptotic solution \refb{e2a} reproduces the

5517: asymptotic form of the exact stringy answers \refb{edilcharge} -

5518: \refb{errchargep}. For a D-$p$-brane in bosonic string theory and

5519: a D-$\bd$ system of type II string theory the coupling to the

5520: Ramond-Ramond field $C^{(p)}$ is absent from \refb{efullcoup}, but the 

5521: dilaton coupling has the same form.

5522: 

5523: \subsection{Classical solutions around the tachyon vacuum}

5524: 

5525: Next we shall demonstrate the absence of perturbative states upon

5526: quantization

5527: of the theory around the tachyon vacuum. Since {\it a priori} it is not

5528: clear how to quantize a

5529: non-linear theory of this type, we shall use a pragmatic definition of

5530: the absence of perturbative states. Since in conventional field theory

5531: perturbative states are associated with plane wave solutions, we shall

5532: assume that absence of perturbative quantum states implies absence of

5533: plane-wave solutions (which are not pure gauge) and vice versa. Thus we

5534: need to show the absence of plane-wave solutions around the tachyon vacuum

5535: in this theory.

5536: 

5537: This leads us to the analysis of classical solutions in this theory. Since

5538: around the tachyon vacuum $V(T)=0$ and hence the action \refb{ez1}

5539: vanishes, it is more convenient to work in the Hamiltonian

5540: formalism\cite{0009061,9901159,9704051,9807064,9910159,

5541: 0010240,0101213,

5542: 0204143,0209034}. 

5543: Defining

5544: the momentum conjugate to $T$ as:

5545:  \be \label{ex1}

5546: \Pi(x) = {\delta S \over \delta (\p_0 T(x))} = {V(T) \p_0 T\over \sqrt{1

5547: - (\p_0 T)^2 + \dt}}\, ,

5548:  \ee

5549: we can construct the Hamiltonian $H$:

5550:  \be \label{ex2}

5551: H = \int d^p x \, (\Pi \p_0 T - \LL) \equiv \int d^p x \, \HH, \qquad

5552: \HH = T_{00} = \sqrt{\Pi^2 + (V(T))^2 } \, \sqrt{1 + \dt}

5553: \, .

5554:  \ee

5555: The equations of motion derived from this hamiltonian take the form:

5556:  \be \label{ex4p}

5557: \p_0 \Pi(x)= -{\delta H\over \delta T(x)} = \p_j \bigg( \sqrt{\Pi^2 +

5558: V^2}

5559: \, { \p_j

5560: T

5561: \over

5562: \sqrt{1 + \dt} }\bigg) - {V(T) V'(T) \over \sqrt{\Pi^2 + V^2}} \,

5563: \sqrt{1 + \dt}\, ,

5564:  \ee

5565:  \be \label{ex5p}

5566: \p_0 T(x) = {\delta H \over \delta \Pi(x)} = {\Pi\over

5567: \sqrt{\Pi^2 + V^2}}\, \sqrt{1 + \dt} \, .

5568:  \ee

5569: In the limit of large $T$ ({\it i.e.}

5570: near the tachyon vacuum) at fixed $\Pi$,

5571: we can

5572: ignore the

5573: $V^2\simeq e^{-\alpha T}$ term, and the Hamiltonian and the equations

5574: of motion take the form:

5575:  \be \label{eham}

5576: H = \int d^p x \, |\Pi| \sqrt{1 + \dt}\, ,

5577:  \ee

5578:  \be \label{ex4}

5579: \p_0 \Pi(x)= \p_j \bigg( |\Pi|\, { \p_j

5580: T

5581: \over

5582: \sqrt{1 + \dt} }\bigg)\, ,

5583:  \ee

5584:  \be \label{ex5}

5585: \p_0 T(x) = {\Pi\over

5586: |\Pi|}\, \sqrt{1 + \dt} \, .

5587:  \ee

5588: {}From \refb{ex5}, we see that

5589: in this limit we have $(\p_0T)^2 - \dt= 1$.

5590: 

5591: These equations can be rewritten in a suggestive form by

5592: defining\cite{0204143}

5593:  \be \label{eadd1}

5594: u_\mu \equiv -\p_\mu T, \qquad \epsilon(x) \equiv |\Pi(x)|/

5595: \sqrt{1 + \dt}\, .

5596:  \ee

5597: Eqs.\refb{ex4}, \refb{ex5} then take the form:

5598:  \be \label{eadd2}

5599: \eta^{\mu\nu} u_\mu u_\nu = -1, \qquad \p_\mu (\epsilon(x) u^\mu) = 0\, .

5600:  \ee

5601: Expressed in terms of these new variables, $T_{\mu\nu}$ given in

5602: \refb{etens} take the form:

5603:  \be \label{eadd3}

5604: T_{\mu\nu} = \epsilon(x) u_\mu u_\nu\, ,

5605:  \ee

5606: where we have used eq.\refb{ex5p} and the small $V(T)$ approximation. We 

5607: now note that \refb{eadd2}, \refb{eadd3}

5608: are precisely the equations governing the motion of non-rotating,

5609: non-interacting dust,

5610: with $u_\mu$ interpreted as the local $(p+1)$-velocity

5611: vector\cite{0204143}, and

5612: $\epsilon(x)$ interpreted as the local rest mass density.

5613: Conversely, any configuration describing flow of non-rotating,

5614: non-interacting dust can

5615: be interpreted as a solution of the equations of motion \refb{ex4},

5616: \refb{ex5}.

5617: 

5618: It is now clear that there are no plane wave solutions in this classical

5619: theory. For example if we begin with an initial

5620: static configuration with an inhomogeneous distribution of energy, this

5621: disturbance does not propagate. On the other hand a plane wave solution

5622: always propagates. Thus the particular field theory described here does

5623: not have any plane wave solution, and is not expected to have any

5624: perturbative physical state upon quantization.

5625: 

5626: Given this large class of classical solutions in the effective field

5627: theory, we can now ask if there are boundary conformal field theories

5628: corresponding to these solutions. Existence of 

5629: such boundary

5630: conformal field theories is a necessary condition for the validity of

5631: the effective action description due to the correspondence between

5632: classical solutions of open string field theory and two dimensional

5633: boundary conformal field theories. As we have seen, a classical

5634: solution of the effective field theory at late time 

5635: corresponds to

5636: a configuration of non-rotating, non-interacting dust. The latter on

5637: the other hand can be thought of as a configuration of massive particles

5638: moving around freely in space. As there is no lower bound to the density

5639: of the dust in the effective field theory, there is no lower bound

5640: to the masses of these particles. Are there boundary CFT's

5641: corresponding to such

5642: configurations? It turns out that the answer is in the affirmative.

5643: To see this consider a non-BPS D0-brane (in bosonic or type

5644: IIB string theory) or a D0-$\bd$0 pair (in type IIA string theory) and set

5645: up the rolling tachyon solution on this. This allows us to construct a

5646: configuration of arbitrary energy by adjusting the parameter $\tl$ in

5647: eq.\refb{e2.12} or \refb{e2.19} for $p=0$. With the help of Lorentz

5648: transformation, we can now construct a

5649: configuration where this 0-brane system is moving with arbitrary

5650: velocity. Since at open string tree

5651: level different D-branes do not interact with each other, we can also

5652: construct a configuration by superposing an arbitrary number of such

5653: 0-brane systems with arbitrary mass and velocity distribution. Such

5654: configurations precisely describe a configuration of non-interacting dust,

5655: {\it i.e.} classical solutions of the field theory described by the

5656: action \refb{ez1}

5657: at late time.

5658: 

5659: 

5660: \subsection{Inclusion of other massless bosonic fields} \label{s8.3}

5661: 

5662: On a non-BPS D$p$-brane world volume we have, besides the tachyonic field

5663: and infinite tower of massive fields, a U(1) gauge field $A_\mu$

5664: ($0\le\mu\le p$), and a

5665: set of scalar fields $Y^I$, one for each

5666: direction $y^I$ transverse to the D-brane ($(p+1)\le I\le D$, $D$ being 9 for

5667: superstring theory and 25 for bosonic string theory).

5668: One could try to generalize \refb{ez1} by including these massless fields

5669: in the action.

5670: The proposed form of the action is\cite{0003122}

5671:  \be \label{exz1}

5672: S = \int d^{p+1} x \, \LL\, , \qquad \qquad

5673: \LL = -  V(T) \, \sqrt{-\det A} \, ,

5674:  \ee

5675: where

5676:  \be \label{exy2}

5677: A_{\mu\nu} = \eta_{\mu\nu} + \p_\mu T \p_\nu T + F_{\mu\nu} + \p_\mu Y^I

5678:  \p_\nu Y^I\, ,

5679:  \ee

5680:  \be \label{esst4}

5681: F_{\mu\nu} = \p_\mu A_\nu - \p_\nu A_\mu\, ,

5682:  \ee

5683: and $V(T)$ has the same form as in \refb{ezz1} for large $T$.

5684: The action described in \refb{exz1}, \refb{exy2} satisfies the

5685: requirement that

5686: for $T=0$ where

5687: $V(T)$ has a maximum it reduces to the usual

5688: Dirac-Born-Infeld form. Furthermore, this action obeys

5689: various restrictions involving the

5690: universality of the tachyon potential\cite{9909062,9911116} and T-duality

5691: invariance\cite{0003221}. As we shall see later, this form of the action

5692: can also

5693: be supersymmetrized easily.

5694: 

5695: Coupling of this action to background string metric $G_{\mu\nu}$,

5696: anti-symmetric tensor field $B_{\mu\nu}$ and the dilaton $\Phi_D$

5697: is given by:

5698:  \be \label{exz1c}

5699: S = -\int d^{p+1} x \, e^{-\Phi_D} \, V(T) \, \sqrt{-\det {\bf A}}

5700: \, ,

5701:  \ee

5702: where

5703:  \ben \label{exy2c}

5704: {\bf A}_{\mu\nu} &=& G_{\mu\nu} +B_{\mu\nu}  + \p_\mu T \p_\nu T +

5705: F_{\mu\nu} +

5706: \p_\mu Y^I

5707:  \p_\nu Y^I\, \nonumber \\

5708: && +(G_{IJ}+B_{IJ})\p_\mu Y^I

5709: \p_\nu Y^J + (G_{\mu I}+B_{\mu I}) \p_\nu Y^I + (G_{I\nu}+B_{I\nu})

5710: \p_\mu Y^I\, .

5711:  \een

5712: Using this action we can compute the source terms for various

5713: closed string fields produced by the brane. For the non-BPS

5714: D-brane of type II string theory there is also a coupling to the

5715: RR fields that generalizes \refb{efullcoup}, but we shall not

5716: describe it here.

5717: 

5718: The dynamics

5719: of this brane is again best described in the Hamiltonian formalism. If we

5720: denote by $p^I$ the momentum conjugate to $Y^I$, by $\Pi$ the momentum

5721: conjugate to $T$ and by $\Pi^i$ the momentum conjugate to $A_i$

5722: ($1\le i\le p$), and

5723: consider the limit of large

5724: $T$ where we can ignore the $V(T)$ term, the

5725: Hamiltonian of the system described by the action \refb{exz1} - 

5726: \refb{esst4} is given by\cite{0009061},

5727:  \be \label{esst1}

5728: H = \int d^p x \, \HH\, ,

5729:  \ee

5730:  \be \label{esst2}

5731: \HH = \sqrt{\Pi^i \Pi^i + p_I p_I + \Pi^2 + \Pi^i \p_i Y^I \Pi^j \p_j Y^I

5732: + \Pi^i \p_i T \Pi^j \p_j T + b_i b_i}\, ,

5733:  \ee

5734: where

5735:  \be \label{esst3}

5736: b_i \equiv F_{ij} \, \Pi^j + p_I \, \p_i Y^I + \Pi\, \p_i T \, .

5737:  \ee

5738: Furthermore the $\Pi^i$'s satisfy a constraint:

5739:  \be \label{esst5}

5740: \p_i \Pi^i = 0\, .

5741:  \ee

5742: 

5743: A particular classical solution of the equations of motion derived from

5744: this Hamiltonian is given by:

5745:  \be \label{esst6}

5746: \Pi^1 = f(\vec x_\perp), \qquad \Pi^i = 0 \quad \hbox{for} \quad i\ge 2,

5747: \qquad \Pi = 0, \qquad A_1 = x^0\, ,

5748:  \ee

5749: with all other fields set to zero. Here $\vec x_\perp$ denotes

5750: coordinates along the D$p$-brane world-volume transverse to $x^1$,

5751: and $f(\vec x_\perp)$ denotes an arbitrary function which is

5752: everywhere positive. In particular if we choose $f(\vec x_\perp) =

5753: \delta(\vec x_\perp)$ we get a string like object along the $x^1$

5754: direction localized at $\vec x_\perp=0$.\footnote{Since $f(\vec

5755: x_\perp)$ needs to be positive everywhere, it is more appropriate

5756: to regard this as the limit of a gaussian.} Using \refb{exz1c},

5757: \refb{exy2c} it can be shown that this string acts as a source for

5758: the anti-symmetric tensor field $B_{\mu\nu}$ like a fundamental

5759: string, and has the same mass to charge ratio as that of the

5760: fundamental string\cite{0002223,0009061}. Furthermore the

5761: classical dynamics of this string-like solution that follows from

5762: the Hamiltonian given in \refb{esst2} is described exactly by the

5763: Nambu-Goto action\cite{0005031,0009061,0010181,0010240} . In

5764: particular, even if we begin with a D$p$-brane that breaks the

5765: full Lorentz invariance of the string theory, the dynamics of this

5766: string-like solution has the full Lorentz

5767: invariance\cite{0010240}.

5768: 

5769: This tends to suggest that the effective field theory described by

5770: the action \refb{exz1}, \refb{exy2}, which was proposed for

5771: describing the tree level open string dynamics on an unstable

5772: D-brane system, contains closed strings as classical solutions.

5773: This interpretation, however, cannot be quite correct due to

5774: various reasons, one of them being that the same effective action

5775: contains solutions describing continuous distribution of electric

5776: flux as given in \refb{esst6}, while the fundamental strings must

5777: carry quantized flux. An interpretation, proposed in

5778: \cite{0305011,0306137,0402027} is that the classical solution of

5779: the open string effective action should be trusted only for energy

5780: densities of order $1/g_s$. In the $g_s\to 0$ limit this

5781: corresponds to a very high density of flux and energy, and the

5782: effective field theory represents average properties of a dense

5783: system of closed strings. This interpretation is consistent with

5784: the general open string completeness conjecture to be discussed in

5785: section \ref{sopenclosed}.

5786: 

5787: \subsection{Supersymmetrization of the effective action}

5788: 

5789: We now turn to the issue of supersymmetrization of the action

5790: \refb{exz1} in the case of type IIA or type IIB string theory in (9+1)

5791: dimensions. For this we shall first rewrite the action \refb{exz1} in a

5792: slightly different way. We introduce a set of 10 scalar fields $X^M$

5793: instead of $(9-p)$ scalar fields $Y^I$ and take the action:

5794:  \be \label{esst9}

5795: S = \int d^{p+1} x \, \LL\, , \qquad \qquad

5796: \LL = -  V(T) \, \sqrt{-\det A} \, ,

5797:  \ee

5798: where

5799:  \be \label{esst10}

5800: A_{\mu\nu} = \p_\mu T \p_\nu T + F_{\mu\nu} + \eta_{MN} \p_\mu X^M

5801:  \p_\nu X^N \, .

5802:  \ee

5803: $\eta_{MN}$ is the (9+1)-dimensional Minkowski metric.

5804: This action is invariant under an arbitrary reparametrization of the

5805: world-volume coordinates $\{x^\mu\}$. We can fix this reparametrization

5806: invariance by choosing the gauge condition:

5807:  \be \label{esst11}

5808: X^\mu = x^\mu \quad \hbox{for} \quad 0\le \mu \le p\, .

5809:  \ee

5810: If we furthermore call the coordinates $X^{I}$ to be $Y^I$ for $(p+1)\le

5811: I\le 9$, we recover the action \refb{exz1}.

5812: 

5813: We shall now write down the supersymmetric generalization of the action

5814: \refb{esst9}, \refb{esst10} \cite{9909062,9912255}. To do this we first 

5815: need

5816: to know the

5817: spectrum of massless fermions on the world-volume of a non-BPS D$p$-brane.

5818: For

5819: type IIA string theory the massless fermionic fields on the world volume

5820: theory

5821: can

5822: be thought of as a single non-chiral Majorana spinor $\theta$ of the

5823: (9+1)-dimensional Lorentz group, whereas for type IIB string theory we

5824: have a pair of chiral spinors $\theta_A$ ($A=1,2$) of the

5825: (9+1)-dimensional Lorentz group. We shall concentrate on the type IIA

5826: theory first. Let us define:

5827:  \be \label{esp1}

5828: \Pi^M_\mu = \p_\mu X^M -\bar\theta \Gamma^M \p_\mu\theta\, ,

5829:  \ee

5830:  \be \label{esp2}

5831: \GG_{\mu\nu}=\eta_{MN}\Pi_\mu^M\Pi_\nu^N\, ,

5832:  \ee

5833: and

5834:  \be \label{esp3}

5835: \FF_{\mu\nu}=F_{\mu\nu} -[\bar\theta \Gamma_{11}\Gamma_M\p_\mu\theta

5836: (\p_\nu X^M-{1\over 2}\bar\theta\Gamma^M\p_\nu\theta)

5837: -(\mu\leftrightarrow\nu)]\, ,

5838:  \ee

5839: where $\Gamma^M$ denote the ten dimensional gamma matrices and

5840: $\Gamma_{11}$

5841: is the product of all the gamma matrices.

5842: The supersymmetric world-volume action on the non-BPS D-brane is given by:

5843:  \be \label{esp4}

5844: S = \int d^{p+1} x \, \LL\, , \qquad \qquad

5845: \LL = -  V(T) \, \sqrt{-\det \AAA} \, ,

5846:  \ee

5847: where

5848:  \be \label{esp5}

5849: \AAA_{\mu\nu} = \p_\mu T \p_\nu T + \FF_{\mu\nu} + \GG_{\mu\nu}\, .

5850:  \ee

5851: In fact, both $\GG_{\mu\nu}$ and $\FF_{\mu\nu}$ and hence the

5852: action $S$ given in \refb{esp4} can be shown to be invariant under

5853: the supersymmetry transformation\cite{9612080}:

5854:  \ben \label{esp6}

5855: && \delta_\eps\theta=\eps, \qquad \delta_\eps X^M=\bar\eps \Gamma^M\theta,

5856: \qquad \delta_\eps T = 0\, , \nonumber \\

5857: && \delta_\eps A_\mu=\bar\eps\Gamma_{11}\Gamma_M\theta \, \p_\mu X^M

5858: -{1\over 6}(\bar\eps\Gamma_{11}\Gamma_M\theta\,

5859: \bar\theta\Gamma^M\p_\mu\theta

5860: +\bar\eps\Gamma_M\theta\, \bar\theta\Gamma_{11}\Gamma^M\p_\mu\theta)\,

5861: ,

5862:  \een

5863: where the supersymmetry transformation parameter $\eps$ is a Majorana

5864: spinor of SO(9,1) Lorentz group.

5865: 

5866: Since the gauge conditions \refb{esst11} are not invariant under

5867: supersymmetry transformation, once we fix this gauge, a supersymmetry

5868: transformation must be accompanied by a compensating world-volume

5869: reparametrization. For any world-volume field $\Phi$, this corresponds to

5870: a modified supersymmetery transformation law:

5871:  \be \label{esp7}

5872: \wh\delta_\eps\Phi = \delta_\eps\Phi - \bar\eps \Gamma^\mu\theta \,

5873: \p_\mu\Phi\, .

5874:  \ee

5875: 

5876: Let us now turn to non-BPS D$p$-branes in type IIB string theory.

5877: As mentioned earlier, in this case the world-volume theory

5878: contains a pair of Majorana-Weyl spinors $\theta_1$ and $\theta_2$

5879: of SO(9,1) Lorentz group. For definiteness we shall take these spinors to 

5880: be right-handed. Let us define

5881:  \be \label{esp18}

5882: \theta=\pmatrix{\theta_1\cr\theta_2}\, ,

5883:  \ee

5884: and let $\tau_3$ denote the matrix $\pmatrix{I & \cr & -I}$ acting on

5885: $\theta$, where $I$ denotes the identity matrix acting on $\theta_1$ and

5886: $\theta_2$. We also define

5887:  \be \label{edefga}

5888: \wh\Gamma^M = \pmatrix{\Gamma^M & 0\cr 0 & \Gamma^M}\, ,

5889:  \ee

5890:  \be \label{esp1a}

5891: \Pi^M_\mu = \p_\mu X^M -\bar\theta \wh\Gamma^M \p_\mu\theta\, ,

5892:  \ee

5893:  \be \label{esp2a}

5894: \GG_{\mu\nu}=\eta_{MN}\Pi_\mu^M\Pi_\nu^N\, ,

5895:  \ee

5896: and

5897:  \be \label{esp3a}

5898: \FF_{\mu\nu}=F_{\mu\nu} -[\bar\theta \tau_3\wh\Gamma_M\p_\mu\theta

5899: (\p_\nu X^M-{1\over 2}\bar\theta\wh\Gamma^M\p_\nu\theta)

5900: -(\mu\leftrightarrow\nu)]\, .

5901:  \ee

5902: The supersymmetric world-volume action is then given by:

5903:  \be \label{espsp4}

5904: S = \int d^{p+1} x \, \LL\, , \qquad \qquad

5905: \LL = -  V(T) \, \sqrt{-\det \AAA} \, ,

5906:  \ee

5907: where

5908:  \be \label{espsp5}

5909: \AAA_{\mu\nu} = \p_\mu T \p_\nu T + \FF_{\mu\nu} + \GG_{\mu\nu}\, .

5910:  \ee

5911: Both $\GG_{\mu\nu}$ and $\FF_{\mu\nu}$ and hence the action is invariant

5912: under the supersymmetry transformation:

5913:  \ben \label{esp6a}

5914: && \delta_\eps\theta=\eps, \qquad \delta_\eps X^M=\bar\eps

5915: \wh\Gamma^M\theta,

5916: \nonumber \\

5917: && \delta_\eps A_\mu=\bar\eps\tau_3\wh\Gamma_M\theta\p_\mu X^M

5918: -{1\over

5919: 6}(\bar\eps\tau_3\wh\Gamma_M\theta\, \bar\theta\wh\Gamma^M\p_\mu\theta

5920: +\bar\eps\wh\Gamma_M\theta\, \bar\theta\tau_3\wh\Gamma^M\p_\mu\theta)\, ,

5921:  \een

5922: where the supersymmetry transformation parameter $\eps$ is given by

5923: $\pmatrix{\eps_1\cr \eps_2}$, with $\eps_1$ and $\eps_2$ both right-handed

5924: Majorana

5925: spinor of SO(9,1) Lorentz group.

5926: 

5927: As before, if we choose to work with a gauge fixed action, then the

5928: supersymmetry transformations have to be modified by including a

5929: compensating gauge transformation as in \refb{esp7}.

5930: 

5931: Just as \refb{efullcoup}, \refb{efull1} describe the coupling of the

5932: massless closed string fields to the open string tachyon, we can

5933: generalize the supersymmetric action described in this section to

5934: include coupling to the complete set of massless closed string fields. 

5935: This can be done by following the general procedure developed in 

5936: \cite{9611173} (see 

5937: also \cite{9512062,9610148,9610249,9611159,9612080}), 

5938: but we shall not describe this

5939: construction here.

5940: 

5941: \subsection{Kink solutions of the effective field theory}

5942: 

5943: The effective field theory described by the action \refb{ez1} provides

5944: a good description of the rolling tachyon solution at late time when $|T|$

5945: becomes large and $|\dot T|$ approaches 1.

5946: Since for a non-BPS D-brane or a D-$\bd$ system in superstring theory a

5947: kink

5948: solution interpolates between the vacua at $T=\pm \infty$, and hence $T$

5949: must pass through 0, there is no reason to expect this effective field

5950: theory to

5951: provide a good description of the kink solution. Nevertheless, we shall

5952: now see that the effective field theory provides a good description of the

5953: kink solution as well.

5954: Our discussion

5955: will follow the analysis of

5956: \cite{0208217,0303057,0004106,0012222}.

5957: 

5958: To begin with we shall not commit ourselves to any specific form of $V(T)$

5959: except

5960: that it is symmetric under $T\to -T$ and falls off to 0 asymptotically

5961: with the behaviour given in \refb{ezz1}. We look for a solution that

5962: depends on one spatial direction $x\equiv x^p$ and is time independent.

5963: For such a system:

5964:  \ben \label{ezsol1}

5965: && T_{xx} = -V(T) / \sqrt{1+(\p_x T)^2}\, , \qquad T_{\mu x} = 0,

5966: \nonumber \\

5967: && T_{\mu\nu} = - V(T) \, \sqrt{1+(\p_x T)^2} \, \eta_{\mu\nu}\, ,

5968: \quad

5969: \hbox{for} \quad 0\le\mu,\nu\le (p-1)\, .

5970:  \een

5971: The energy-momentum conservation gives,

5972:  \be \label{ezsol2}

5973: \p_x T_{xx} = 0\, .

5974:  \ee

5975: Thus $T_{xx}$ is independent of $x$. Since for a kink solution

5976: $T\to\pm\infty$ as $x\to\pm\infty$, and $V(T)\to 0$ in this limit, we see

5977: that $T_{xx}$ must vanish for all $x$. This, in turn, shows that we must

5978: have\cite{0004106}

5979:  \be \label{ezsol3}

5980: T=\pm\infty \quad \hbox{or} \quad \p_x T= \infty \quad \hbox{(or

5981: both)} \quad \hbox{for all $x$}\, .

5982:  \ee

5983: Clearly the solution looks singular. We shall now show that despite this

5984: singularity, the solution has finite energy density which is independent

5985: of the way we regularize the singularity, and for which the energy

5986: density is localized on a

5987: codimension 1 subspace, just as is expected of a D$(p-1)$-brane. For this

5988: let us consider the following field configuration:

5989:  \be \label{ezsol4}

5990: T(x) = f(ax)\, ,

5991:  \ee

5992: where $f(x)$ is an odd, monotone increasing function of $x$ that

5993: approaches $\pm\infty$ as $x\to\pm\infty$ but is otherwise arbitrary, and

5994: $a$ is a constant that we shall take to $\infty$ at the end. Clearly in

5995: this limit we have $T=\infty$ for $x>0$ and $T=-\infty$ for $x<0$, therby

5996: producing a singular kink.

5997: 

5998: Let us compute the energy momentum tensor associated with the

5999: configuration \refb{ezsol4}. From \refb{ezsol1} we see that the non-zero

6000: components are:

6001:  \ben \label{ezsol5}

6002: T_{xx} &=& -V(f(ax))/ \sqrt{1 + a^2 (f'(ax))^2}\, , \nonumber \\

6003: T_{\mu\nu} &=& - V(f(ax))\, \sqrt{1 + a^2 (f'(ax))^2}\,

6004: \eta_{\mu\nu}\,

6005: , \quad 0\le \mu,\nu\le (p-1).

6006:  \een

6007: Clearly in the $a\to\infty$ limit, $T_{xx}$ vanishes everywhere since

6008: the numerator vanishes and the denominator blows up. Hence

6009: the conservation law \refb{ezsol2} is automatically satisfied. This in

6010: turn shows that this configuration is a solution of the equations of

6011: motion in this limit.

6012: 

6013: {}From \refb{ezsol5} we see furthermore that in the $a\to\infty$ limit, we

6014: can write $T_{\mu\nu}$ as:

6015:  \be \label{ezsol6}

6016: T_{\mu\nu} = -a\, \eta_{\mu\nu}\,  V(f(ax))\, f'(ax)\, .

6017:  \ee

6018: Thus the integrated $T_{\mu\nu}$, associated with the codimension 1

6019: soliton, is given by:

6020:  \be \label{ezsol7}

6021: T^{tot}_{\mu\nu} = -a\, \eta_{\mu\nu} \, \int_{-\infty}^\infty\, d x \,

6022: V(f(ax))\, f'(ax) = -\eta_{\mu\nu} \, \int_{-\infty}^\infty \, d y \,

6023: V(y)\, ,

6024:  \ee

6025: where $y=f(ax)$. This shows that

6026: the

6027: final answer depends only on the form of $V(y)$ and not on the shape of

6028: the function $f(x)$ used to describe the soliton. It is also clear from

6029: the exponential fall off in $V(y)$ for large $y$ that most of the

6030: contribution to $T^{tot}_{\mu\nu}$

6031: is contained within a finite range of $y$.

6032: The relation $y=f(ax)$ then implies that

6033: the contribution comes from a region of $x$ integral of

6034: width $1/a$ around $x=0$. In the $a\to\infty$ limit such a distribution

6035: approaches a $\delta$-function. Thus the energy density associated with

6036: this solution is given by:

6037:  \be \label{ezsol8}

6038: T_{\mu\nu} = - \eta_{\mu\nu} \, \delta(x) \, \int_{-\infty}^\infty \, d y

6039: \,

6040: V(y)\, .

6041:  \ee

6042: This is precisely what is expected of a D-$(p-1)$-brane, provided the

6043: integral $\int_{-\infty}^\infty \, d y \, V(y)$ equals the tension of the

6044: D-$(p-1)$-brane. For comparison, we recall that $V(0)$ denotes the tension

6045: of a D$p$-brane.

6046: 

6047: One can check that the solution constructed this way satisfies the

6048: complete set of equations of motion of the effective field theory.

6049: Furthermore one can construct the world-volume effective action on this

6050: kink solution, and this turns out to be exactly the Dirac-Born-Infeld

6051: action, as is expected if the kink has to describe a BPS

6052: D-$(p-1)$-brane\cite{0303057,0104218,0012080}. We shall not 

6053: discuss this

6054: construction in detail,

6055: and refer the reader to the original papers.

6056: 

6057: So far in our discussion we have not committed ourselves to a specific

6058: form of $V(T)$. It turns out that we can get a lot more quantitative

6059: agreement with string theory results if we

6060: choose\cite{0301076,0303035,0303139}:

6061:  \be \label{evform}

6062: V(T) = \wt \TT_p / \cosh(T/\sqrt 2)\, .

6063:  \ee

6064: Here for definiteness we have considered the case of a non-BPS

6065: D-$p$-brane. The overall normalization of $V$ has been adjusted so that

6066: $V(0)$ reproduces the tension of this brane. We now note that:

6067:  \begin{enumerate}

6068: \item If we expand the action around $T=0$ and keep terms up to quadratic

6069: order in $T$, we find that the field $T$ describes a particle of

6070: mass$^2=-{1\over 2}$. This agrees with the mass$^2$ of the tachyon on an

6071: unstable D-brane in superstring theory.

6072: 

6073: \item The tension of the kink solution in the effective field theory is

6074: now given by:

6075:  \be \label{evform2}

6076: \int_{-\infty}^\infty dy \, V(y) = \sqrt 2 \pi \wt\TT_p \, .

6077:  \ee

6078: This gives the correct tension of a BPS D-$(p-1)$-brane.

6079: 

6080: \item $V(T)$ given in \refb{evform} has the behaviour given in

6081: \refb{ezz1}, \refb{e2} for large $T$.

6082: \end{enumerate}

6083: Finally, it was shown in \cite{0303139} that this

6084: effective field theory admits a one parameters family of solutions of the

6085: form

6086:  \be \label{evform3}

6087: T = \sqrt 2 \sinh^{-1} \left( {\lambda} \sin \left({x\over

6088: \sqrt 2}\right) \right)\, ,

6089:  \ee

6090: where $x$ denotes any of the spatial coordinates on the D-$p$-brane.

6091: These are precisely the analogs of the solutions \refb{efh5}

6092: for $R=1/\sqrt 2$. Thus the effective

6093: field theory has solutions in one to one correspondence with the family of

6094: BCFT's associated with marginal

6095: deformations of the original BCFT. The moduli space of these solutions in 

6096: the effective field theory closely resembles the moduli space of the 

6097: corresponding BCFT's but they are not identical\cite{0312003}. This shows 

6098: that the effective field theory does not reproduce quantitatively all the 

6099: features of the classical solutions in the full open string theory.

6100: 

6101: Various other aspects of the tachyon effective action described in this

6102: section have been discussed in

6103: refs.\cite{9905195,

6104: 0102174,0107087,

6105: 0202079,0204203,0206212,

6106: 0207235,0208019,0208094,

6107: 0209142,0210108,

6108: 0210221,

6109: 0211090,

6110: 0301101,

6111: 0303204,0304108,0304180,0304197,0305092,0305229,0305249,

6112: 0306294,0306295,

6113: 0307184,0308069,

6114: 0310066,

6115: 0310079,0310253,

6116: 0312086,0312149,

6117: 0401163,

6118: 0401195,

6119: 0401236,0403073,

6120: 0403124,0403217,

6121: 0404163,0404253,

6122: 0405058,0406120,

6123: 0408073,0409030,

6124: 0409151,0410030}

6125: 

6126: \sectiono{Toy Models for Tachyon Condensation} \label{s5}

6127: 

6128: In section \ref{s5.3} we have discussed a specific form of effective field

6129: theory that reproduces many of the features of the tachyon dynamics on

6130: an

6131: unstable D-brane. In this section we describe two other types of field

6132: theory models which share some properties of the tachyon dynamics in the

6133: full string theory, namely absence of physical states around the tachyon

6134: vacuum, and lower dimensional branes as solitons. The first of these

6135: models will be based on

6136: singular potential but regular kinetic term, while the second type of

6137: model will be based on smooth potential but

6138: non-local kinetic term. Both these classes of models mimick many of the

6139: properties of the time independent solutions of the tachyon effective

6140: action in open string theory. However unlike the model described in

6141: section \ref{s5.3} these models do not seem to reproduce the features of

6142: the time dependent solutions involving the open string tachyon. For this

6143: reason we shall restrict our analysis to time independent solutions

6144: only.

6145: For some analysis of time dependent solutions in these models, see

6146: \cite{0207107,0209197,math-ph/0306018}.

6147: 

6148: \subsection{Singular potential model} \label{s5.1}

6149: 

6150: Let us consider a field theory of a scalar field $\phi$ in $(p+1)$

6151: dimensions, described by the

6152: action\cite{0008231}:

6153:  \be \label{esin1}

6154: S = -\int d^{p+1}x \left[ {1\over 2} \eta^{\mu\nu} \p_\mu \phi \p_\nu\phi

6155: +

6156: V(\phi)\right]

6157:  \ee

6158: where

6159:  \be \label{esin2}

6160: V(\phi) = -{1\over 4} \phi^2 \ln \phi^2\, .

6161:  \ee

6162: The potential has a local maximum at $\phi = e^{-1/2}$. At this point

6163:  \be \label{esin3}

6164: V(e^{-1/2}) = {1\over 4 e}, \qquad V''(e^{-1/2})=-1\, .

6165:  \ee

6166: This shows that the scalar field excitation around the maximum describes

6167: a

6168: tachyonic mode with mass$^2=-1$. The potential also has minimum at

6169: $\phi=0$ where it vanishes. The second derivative of the potential is

6170: infinite at $\phi=0$, showing that there is no finite mass scalar particle

6171: obtained by perturbative quantization of this theory around this minimum.

6172: This is consistent with the second conjecture. The difference between the

6173: values of the potential at the minimum and the local maximum can be

6174: thought of as the tension of the $p$-brane that the $\phi=e^{-1/2}$

6175: solution describes.

6176: 

6177: We now examine the classical soliton solutions in this field theory. As

6178: already mentioned, it has a translationally invariant vacuum solution

6179: $\phi=0$ and a translationally invariant solution $\phi=e^{-1/2}$

6180: corresponding to a local maximum of the potential. But the theory also has

6181: codimension $(p-q)$ lump solutions, given by\cite{0008231}:

6182:  \be \label{esin4}

6183: \phi = F\left(\sqrt{(x^{q+1})^2 + \cdots + (x^p)^2}\right)\, ,

6184:  \ee

6185: where

6186:  \be \label{esin5}

6187: F(\rho) = \exp\left(-{1\over 4} \rho^2 + {1\over 2} (p-q-1)\right)\, .

6188:  \ee

6189: The tension of this $q$-brane solution is given by:

6190:  \be \label{esin6}

6191: \TT_q = {1\over 4} e^{p-q-1} (2\pi)^{(p-q)/2}\, .

6192:  \ee

6193: Thus

6194:  \be \label{esin7}

6195: \TT_q / \TT_{q+1} = \sqrt{2\pi} e\, .

6196:  \ee

6197: This is independent of $q$. This reproduces the string theoretic

6198: feature

6199: that the ratio of the tension of a D-$q$-brane and D-$(q-1)$-brane is

6200: independent of $q$.

6201: 

6202: Given the lump solution \refb{esin4}, \refb{esin5}, we can anlyze the

6203: spectrum of small fluctuations around the solution. For a codimension

6204: $(p-q)$ lump the spectrum coincides with that of $(p-q)$ dimensional

6205: harmonic oscillator\cite{0008231} up to an overall additive constant. To 

6206: be more specific,

6207: the spectrum of excitations on a codimension $(p-q)$ lump is labelled by

6208: $(p-q)$

6209: integers $(n_1, \cdots n_{p-q})$, and the mass$^2$ of the states

6210: associated

6211: with these excitations are given by:

6212:  \be \label{esin8}

6213: m_{n_1, \cdots n_{p-q}}^2 = (n_1 +\cdots n_{p-q} - 1)\, .

6214:  \ee

6215: This shows that on each of these D-$q$-branes the lowest excitation

6216: mode, corresponding to $n_1=\ldots = n_{p-q}=0$, is tachyonic and has

6217: mass$^2=-1$. This is again in accordance with the results in the

6218: bosonic string theory that the mass$^2$ of the tachyonic mode on any

6219: D-$q$-brane has the same value $-1$ independent of the value of $q$.

6220: 

6221: Various generalizations of these models as well as other field theory 

6222: models can be found in 

6223: refs.\cite{0004131,0008227,

6224: 0009246,0011226,0104229,0106103,

6225: 0107070,0107075,0112088,

6226: 0203108,0211127,0405125,

6227: 0407081}.

6228: 

6229: \subsection{$p$-adic string theory} \label{s5.2}

6230: 

6231: The $p$-adic open string theory is obtained from ordinary bosonic open 

6232: string

6233: theory on a D-brane

6234: by replacing, in the Koba-Nielson amplitude, the integral over the

6235: real world-sheet coordinates by $p$-adic integral associated with a prime

6236: number

6237: $p$\cite{Freund1,Freund2,Frampton1,padic}. We

6238: shall not

6239: review this construction here. For our purpose it will be

6240: sufficient to know

6241: that in this case there is an exact expression for the tachyon effective

6242: action which reproduces correctly all the tree level amplitudes involving

6243: the tachyon. This effective action for the tachyon on a Dirichlet

6244: $(d-1)$-brane is given by\cite{padic,Frampton2}:

6245:  \ben \label{epa1}

6246: S &=& \int d^d x \, \LL \nonumber \\

6247: &=& {1\over g^2} {p^2 \over p-1} \int d^d x \left[ -{1\over 2}

6248: \phi \, p^{-\half\Box}\, \phi + {1\over p+1} \, \phi^{p+1} \right]\, ,

6249:  \een

6250: where $\Box$ denotes the $d$ dimensional Laplacian, $\phi$ is the

6251: tachyon field (after a rescaling and a shift), and $g$ is the open

6252: string coupling constant.\footnote{Although the $p$-adic string

6253: theory is defined only for $p$ prime, once the action \refb{epa1}

6254: is written down, we can analyze its properties for any integer

6255: $p$. In the $p\to 1$ limit the action \refb{epa1} reduces to the

6256: action given in \refb{esin1}, \refb{esin2}\cite{0009103}.} The

6257: potential of the model, defined as $-\LL$ evaluated for spatially

6258: homogeneous field configurations, is given by:

6259:  \be \label{epa22}

6260: V(\phi) = {1\over g^2} {p^2 \over p-1} \left[ {1\over 2} \phi^2 -

6261: {1\over p+1} \phi^{p+1} \right]\, .

6262:  \ee

6263: This has a local minimum at $\phi=0$ and maxima at $\phi^{p-1}=1$.

6264: 

6265: The classical equation of motion derived from the action \refb{epa1} is

6266:  \be \label{epa30}

6267: p^{-\half\Box} \phi = \phi^p\, .

6268:  \ee

6269: Different known solutions of this equation

6270: are as follows\cite{padic}:

6271:  \begin{itemize}

6272: \item The configuration $\phi=0$ is a local minimum of

6273: $V(\phi)$ with $V(0)=0$.

6274: We shall identify this solution

6275: with the tachyon vacuum configuration. By definition we have taken

6276: the energy density of this vacuum to be zero.

6277: 

6278: To analyze the spectrum of perturbative physical excitations around 

6279: $\phi=0$

6280: we examine the linearized equation of motion around this point:

6281:  \be \label{epa1d}

6282: p^{-\half\Box} \phi = 0\, .

6283:  \ee

6284: If we look for a plane-wave solution of this equation of the form:

6285:  \be \label{epa1f}

6286: \phi = \phi_0 e^{ik.x}\, ,

6287:  \ee

6288: then eq.\refb{epa1d} gives

6289:  \be \label{epa1e}

6290: p^{\half k^2} = 0\, .

6291:  \ee

6292: This has no solution for finite $k^2$. Thus there are no perturbative

6293: physical excitations around the configuration $\phi=0$. This is in 

6294: accordance with

6295: conjecture 2 if we identify the $\phi=0$ configuration as the vacuum 

6296: without any D-brane.

6297: 

6298: 

6299: \item The configuration $\phi=1$, being the maximum of $V(\phi)$,

6300: represents the original D-brane

6301: configuration around

6302: which we quantized the string.\footnote{For odd $p$, there is also

6303: an equivalent solution corresponding to $\phi=-1$. Since the action

6304: is symmetric under $\phi\to -\phi$, we shall restrict our analysis to

6305: solutions with positive $\phi$.} We shall call this the

6306: D-$(d-1)$-brane solution. The energy density associated with this

6307: configuration, which can be identified as the tension $\TT_{d-1}$ of

6308: the D-$(d-1)$-brane according to the first conjecture on open string 

6309: tachyon dynamics, is given by

6310:  \be \label{epa1a}

6311: \TT_{d-1} = - \LL(\phi=1) = {1\over 2 g^2} {p^2 \over p+1}\, .

6312:  \ee

6313: Unfortunately in $p$-adic string theory there is no independent method of 

6314: calculating the tension of the D-$(d-1)$-brane and compare this with 

6315: \refb{epa1a} to verify the first conjecture.

6316: 

6317: The linearized equation of motion around the $\phi=1$ solution is found 

6318: by

6319: defining $\chi\equiv (\phi-1)$ and expanding \refb{epa30} to first order

6320: in $\chi$. This gives:

6321:  \be \label{epa1b}

6322: \left( p^{-\half\Box} -p\right) \chi = 0\, .

6323:  \ee

6324: This has plane wave solutions of the form

6325:  \be \label{epa1c}

6326: \chi = \chi_0 e^{ik.x}\, ,

6327:  \ee

6328: provided $k^2 = 2$. Thus the excitation around the point $\phi=1$

6329: describes a tachyonic mode with mass$^2=-2$.

6330: 

6331: 

6332: \item The configuration:

6333:  \be \label{epa2}

6334: \phi(x) = f(x^{q+1}) f (x^{q+2}) \cdots f(x^{d-1})\equiv

6335: F^{(d-q-1)}(x^{q+1},\ldots,x^{d-1})\, ,

6336:  \ee

6337: with

6338:  \be \label{epa3}

6339: f(\eta) \equiv p^{1\over 2(p-1)} \exp\left(-{1\over 2}\,

6340: {p-1\over p \ln p} \, \eta^2\right) \, ,

6341:  \ee

6342: describes a soliton solution with energy density localized

6343: around the hyperplane $x^{q+1} = \cdots = x^{d-1}=0$. 

6344: Indeed, by using

6345: the identity

6346:  \be \label{epas1}

6347: p^{-\half \p_\eta^2} f(\eta) = \left(f(\eta)\right)^p

6348:  \ee

6349: one can show

6350: that \refb{epa2}, \refb{epa3} solves the classical equations of

6351: motion \refb{epa30}. 

6352: \refb{epas1} in turn can be proven easily by working in the Fourier

6353: transformed space.

6354: 

6355: We shall call \refb{epa2}

6356: the solitonic $q$-brane solution. Let us denote by

6357: $x_\perp=(x^{q+1},\ldots, x^{d-1})$ the coordinates transverse to the

6358: brane and by $x_\parallel=(x^0,\ldots,x^q)$ those tangential to it.

6359: The energy density per unit $q$-volume of this brane, which can

6360: be identified as its tension $\TT_q$, is given by

6361:  \be \label{epa4}

6362: \TT_q = - \int d^{d-q-1} x_\perp\; \LL(\phi = F^{(d-q-1)}(x_\perp))

6363: = {1\over 2g_q^2} \, {p^2 \over p+1}\, ,

6364:  \ee

6365: where,

6366:  \be \label{epa9}

6367: g_q = g \left[{p^2 - 1\over

6368: 2 \pi\, p^{2p/(2p-1)}\ln p}\right]^{(d-q-1)/4}\, .

6369:  \ee

6370: \end{itemize}

6371: {}From eqs.\refb{epa1a},\refb{epa4} and \refb{epa9} we see that the

6372: ratio

6373: of the tension of a $q$-brane and a $(q-1)$-brane is

6374:  \be \label{epa5}

6375: {\TT_q\over \TT_{q-1}} = \left[ { 2\pi\, p^{2p\over p-1}\ln p

6376: \over p^2 -1}\right]^{-{1\over 2}}  \, .

6377:  \ee

6378: This is independent of $q$. Since

6379: this is also a feature of the D-branes in ordinary

6380: bosonic string theory, it suggests that the solitonic $q$-branes of

6381: $p$-adic string theory should have interpretation as

6382: D-branes. Unfortunately so far we do not have an independent way of

6383: calculating the tension of a D-$q$-brane in $p$-adic string theory for

6384: arbitrary $q$ and verify \refb{epa5} explicitly. However, as we shall show 

6385: now, the spectrum of fluctuations around a solitonic $q$-brane does match 

6386: the spectrum of open strings on a $p$-adic D-$q$-brane. 

6387: 

6388: We can study fluctuations around a solitonic $p$-brane solution by taking

6389: the following ansatz for the field $\phi$:

6390:  \be \label{epa6}

6391: \phi(x) = F^{(d-q-1)}(x_\perp) \psi(x_\parallel) \, ,

6392:  \ee

6393: with $F^{(d-q-1)}(x_\perp)$ as defined in \refb{epa2}, \refb{epa3}.

6394: For $\psi=1$ this describes the solitonic $q$-brane. Fluctuations of

6395: $\psi$ around 1 denote fluctuations of $\phi$ localized on the

6396: soliton; thus $\psi(x_\parallel)$ can be regarded as one

6397: of the fields on its world-volume. We shall call this

6398: the tachyon field on the solitonic $q$-brane world-volume. Substituting

6399: \refb{epa6} into \refb{epa30} and using \refb{epas1} we get

6400:  \be \label{epa7}

6401: p^{-\half\Box_\parallel} \psi = \psi^p\, ,

6402:  \ee

6403: where $\Box_\parallel$ denotes the $(q+1)$ dimensional Laplacian

6404: involving the world-volume coordinates $x_\parallel$ of the $q$-brane. Any 

6405: solution of eq.\refb{epa7}, after being substituted into eq.\refb{epa6}, 

6406: gives a solution to the original equation of motion \refb{epa30}. The 

6407: action involving 

6408: $\psi$ can be obtained by substituting \refb{epa6} into \refb{epa1}:

6409:  \ben

6410: S_q(\psi) &=& S\left(\phi=F^{(d-q-1)}(x_\perp)\psi(x_\parallel)\right)

6411: \nonumber\\

6412: &=& {1\over g_q^2} {p^2 \over p-1} \int d^{q+1} x_\parallel

6413: \left[ -{1\over 2} \psi p^{-\half \Box_\parallel}

6414: \psi + {1\over p+1} \psi^{p+1} \right]\, ,\label{epa8}

6415:  \een

6416: where $g_q$ has been defined in eq.\refb{epa9}.

6417: \refb{epa7}, \refb{epa8} are precisely the tachyon equation

6418: of motion and tachyon effective action (up to an overall normalization) 

6419: that we would have

6420: gotten by quantizing the open $p$-adic string on a D-$q$-brane directly.

6421: This correspondence continues to hold for other massless and massive

6422: excitations on the

6423: solitonic brane as well\cite{0003278,0102071}.

6424: This is a strong indication that these solitonic $q$-brane solutions on

6425: the world-volume of the D-$(d-1)$-brane actually describe lower

6426: dimensional D-$q$-brane, in accordance with conjecture 3.

6427: 

6428: Various other aspects of tachyon condensation in $p$-adic string theory 

6429: have been studied in \cite{0105312,0209197,

6430: 0304213,0406120,

6431: 0406259,0409311}.

6432: 

6433: \sectiono{Closed String Emission from `Decaying' D-branes}

6434: \label{s9}

6435: 

6436: So far we have carried out our analysis in tree level open string theory.

6437: Although we have used the coupling of closed strings to D-branes to

6438: determine the sources for various closed string fields and 

6439: construct the boundary state associated with a D-brane, we have not

6440: treated the closed strings as dynamical objects and studied what kind of

6441: closed string background the D-brane produces. In this section we shall

6442: address this problem in the context of time dependent solutions associated

6443: with the rolling tachyon configuration on an unstable D-brane

6444: \cite{0303139,0304192,0402157}.

6445: For

6446: simplicity we shall restrict our analysis to bosonic string theory only,

6447: but the results can be generalized to superstring theories as

6448: well.

6449: 

6450: A brief review of some aspects

6451: of closed string field theory has been 

6452: given in appendix \ref{appb}.

6453: We begin with the closed string field equation in the presence of a

6454: D-brane, as given in eq.\refb{efg4app}:

6455:  \be \label{efg4}

6456: 2\, (Q_B+\bar Q_B)\, |\Psi_c\ra = K\, g_s^2 \, |\BB\ra\, .

6457:  \ee

6458: As described in appendix A, 

6459: the closed string field $|\Psi_c\ra$ is represented by a ghost number 

6460: two state in the CFT on the full complex plane (which is conformally 

6461: equivalent to a cylinder) satisfying the constraints \refb{efg1.5}.

6462: $g_s$ is the closed string coupling constant, $K$ is

6463: a numerical constant determined in eq.\refb{ezopen}, and $|\BB\ra$ is

6464: the boundary state associated with the D-brane. Noting that $|\BB\ra$

6465: is BRST invariant,

6466: and that $\{Q_B+\bar Q_B, b_0+\bar b_0\} = (L_0 + \bar L_0)$, we can write

6467: down a

6468: solution to equation \refb{efg4} as:

6469:  \be \label{es9.1}

6470: |\Psi_c\ra = K\, g_s^2 \, [2(L_0 + \bar L_0)]^{-1} \, (b_0+\bar b_0) \,

6471: |\BB\ra\, .

6472:  \ee

6473: This solution satisfies the Siegel gauge condition and Siegel

6474: gauge equations of motion:

6475:  \be \label{eseqom}

6476: (b_0+\bar

6477: b_0)|\Psi_c\ra=0\, , \qquad 2(L_0+\bar L_0)|\Psi_c\ra = K g_s^2 (b_0+\bar

6478: b_0)|\BB\ra\, .

6479:  \ee

6480: We can of course construct other solutions which

6481: are

6482: gauge equivalent to this one by adding to $|\Psi_c\ra$ terms of the

6483: form

6484: $(Q_B+\bar Q_B)|\Lambda\ra$. However even within Siegel gauge, the

6485: right

6486: hand side of \refb{es9.1} is not defined

6487: unambiguously due to the presence of the zero eigenvalues of the

6488: operator $(L_0+\bar L_0)$.\footnote{Since acting on a level $(N,N)$

6489: state the operator

6490: $(L_0+\bar L_0)$ takes the form of a differential operator $-{1\over

6491: 2}

6492: \square + 2(N-1)$,

6493: free closed string field theory in Minkowski space has infinite

6494: number of

6495: plane

6496: wave solutions of the equations

6497: $(L_0+\bar L_0)|\Psi_c\ra

6498: = 0$, $(b_0+\bar b_0)|\Psi_c\ra

6499: = 0$.}

6500: Thus we need to carefully choose a

6501: prescription for defining the right hand side of \refb{es9.1}.

6502: A natural prescription (known as the

6503: Hartle-Hawking prescription) is to begin with the

6504: solution of the associated equations of motion in the Euclidean theory

6505: where there is a unique solution to eq.\refb{es9.1} (which therefore

6506: satisfies the full equation \refb{efg4}) and then analytically

6507: continue

6508: the

6509: result to the Minkowski space along the branch passing through the

6510: origin $x^0=0$\cite{0303139,0304192,0402124}. This is the prescription

6511: we

6512: shall

6513: follow.

6514: 

6515: We shall now describe the results obtained using this formalism.

6516: This will be carried out in two steps:

6517:  \begin{enumerate}

6518: \item First we describe the

6519: closed string

6520: background produced by the $|\BB_1\ra$

6521: component of the boundary state

6522: as defined in \refb{ebs3}.

6523:  \item We then discuss the computation of the closed string background

6524: produced by the $|\BB_2\ra$ component of the boundary state  associated

6525: with the rolling tachyon solution.

6526: \end{enumerate}

6527: 

6528: 

6529: \subsection{Closed string radiation produced by 

6530: $|\BB_1\ra$} \label{eccl}

6531: 

6532: 

6533: Let us denote by $|\Psi_c^{(1)}\ra$ and $|\Psi_c^{(2)}\ra$ the

6534: closed string field configurations produced by $|\BB_1\ra$ and

6535: $|\BB_2\ra$ as given in \refb{ebs3} and \refb{ebs4} respectively.

6536: We begin with the analysis of $|\Psi_c^{(1)}\ra$. Let us define

6537: $\wh A_N$ to be an operator of level $(N,N)$ acting on closed string 

6538: states, composed of negative

6539: moded oscillators of $X^0$, $X^s$, $b$, $c$, $\bar b$ and $\bar c$

6540: such that

6541:  \be \label{eexpan}

6542: \exp\left[\sum_{n=1}^\infty \, \left( -{1\over n}

6543: \alpha^0_{-n}\bar\alpha^0_{-n} +

6544: \sum_{s=1}^{25} (-1)^{d_s} \, {1\over

6545: n} \,  \alpha^s_{-n} \bar

6546: \alpha^s_{-n} - (\bar b_{-n} c_{-n} +

6547: b_{-n}

6548: \bar c_{-n}) \right)

6549: \right] = \sum_{N=0}^\infty \wh A_N\, .

6550:  \ee

6551: Here $\wh A_0=1$. Then $|\BB_1\ra$ given in \refb{ebs3} can be

6552: expressed as

6553:  \be \label{ebb1exp}

6554: |\BB_1\ra = \TT_p \, \int {d^{25-p} k_\perp \over (2\pi)^{25-p}} \,

6555: \sum_{N=0}^\infty \, \wh A_N

6556: \,

6557: (c_0+\bar c_0) \, c_1 \, \bar c_1 \, \wt f\left(X^0(0)\right) \,

6558: |k^0=0, \vec k_\parallel=0, \vec k_\perp\ra\, .

6559:  \ee

6560: In terms of the operators $\wh A_N$, the result for

6561: $|\Psi_c^{(1)}\ra$ is given by:

6562:  \be \label{esa0}

6563: |\Psi_c^{(1)}\ra = 2\, K\, g_s^2 \, \TT_p \, \int {d^{25-p}

6564: k_\perp\over

6565: (2\pi)^{25-p}} \,

6566: \sum_{N\ge 0} \, \wh A_N \, h^{(N)}_{\vec k_\perp}(X^0(0)) \, c_1\,

6567: \bar c_1\, |k^0=0,k_\parallel=0, \vec k_\perp\ra\, ,

6568:  \ee

6569: where $h^{(N)}_{\vec k_\perp}(x^0)$ satisfies:

6570:  \be \label{eheq}

6571: \left( \p_0^2 +\vec k_\perp^2 + 4(N-1)\right) \, h^{(N)}_{\vec

6572: k_\perp}(x^0) =\wt f(x^0)\, .

6573:  \ee

6574: It is easy to see that \refb{esa0} satisfies eq.\refb{eseqom} since

6575: acting on a level $(N,N)$ state the operator

6576: $(L_0+\bar L_0)$ takes the form of a differential operator

6577: $-{1\over 2} \square + 2(N-1)$. It can also be shown explicitly

6578: that \refb{esa0} actually satisfies the full set of equations

6579: \refb{efg4}  with $|\BB\ra$ replaced by $|\BB_1\ra$\cite{0402157}.

6580: 

6581: The result for $h^{(N)}_{\vec k_\perp}(x^0)$ following the

6582: Hartle-Hawking prescription is\cite{0303139,0304192,0402157}

6583:  \be \label{esa-1}

6584: h^{(N)}_{\vec k_\perp}(x^0)

6585: = {i \over 2\onk} \, \left[\int_C \, e^{-i \onk

6586: (x^0 -

6587: x^{\prime 0})} \wt f(x^{\prime

6588: 0}) d x^{\prime 0} - \int_{C'} e^{ i \onk (x^0 - x^{\prime 0})}

6589: \wt f(x^{\prime 0}) d x^{\prime 0}\right]

6590: \, ,

6591:  \ee

6592: with

6593:  \be \label{defonk}

6594: \onk= \sqrt{\vec k_\perp^2

6595: + 4(N-1)}\, .

6596:  \ee

6597: Here the contour $C$ runs from $i\infty$ to the origin along the

6598: imaginary $x^{\prime 0}$ axis, and then to $x^0$ along the real

6599: $x^{\prime

6600: 0}$ axis, and the contour $C'$ runs from $-i\infty$ to the origin

6601: along

6602: the

6603: imaginary $x^{\prime 0}$ axis, and then to $x^0$ along the real

6604: $x^{\prime

6605: 0}$ axis. These are known as the Hartle-Hawking contours. It is easy to 

6606: see that $h^{(N)}_{\vec k_\perp}$ given in \refb{esa-1} satisfies 

6607: \refb{eheq}.

6608: 

6609: 

6610: Since we shall be

6611: interested in

6612: the asymptotic form of the closed string fields in the $x^0\to\infty$

6613: limit, we

6614: can

6615: take the contours $C$ and $C'$ to run all the way to $+\infty$ along

6616: the

6617: real $x^{\prime 0}$ axis. By closing the contours in the first and the

6618: fourth quadrangles

6619: respectively, we can easily show that\cite{0301038,0303139} as

6620: $x^0\to\infty$,

6621:  \ben \label{es9.16}

6622: \int_C \, e^{i \onk

6623: x^{\prime 0}} \wt f(x^{\prime

6624: 0}) d x^{\prime 0} \to -{i\pi\over \sinh (\pi\onk)} e^{-i\onk \ln

6625: \sin(\pi\tl)} \, , \nonumber \\

6626: \int_{C'} \, e^{-i \onk

6627: x^{\prime 0}} \wt f(x^{\prime

6628: 0}) d x^{\prime 0} \to {i\pi\over \sinh (\pi\onk)} e^{i\onk \ln

6629: \sin(\pi\tl)} \, .

6630:  \een

6631: Thus in this limit

6632:  \be \label{es9.17aa}

6633: h^{(N)}_{\vec k_\perp}(x^0) \to

6634: {\pi\over

6635: \sinh(\pi\onk)} \, {1\over 2\onk} \, \left[ e^{-i\onk (x^0 + \ln

6636: \sin(\pi\tl)) } +  e^{i\onk (x^0 + \ln

6637: \sin(\pi\tl)) } \right] \, .

6638:  \ee

6639: Substituting this into \refb{esa0} we get

6640:  \ben \label{es9.17bb}

6641: |\Psi_c^{(1)}\ra &\to& 2\, K\, g_s^2 \, \TT_p \, \int {d^{25-p}

6642: k_\perp\over (2\pi)^{25-p}} \, \sum_{N\ge 0} \, {\pi\over

6643: \sinh(\pi\onk)} \, {1\over 2\onk} \, \wh A_N\,  c_1\, \bar c_1\,

6644: \,

6645: \nonumber \\

6646: && \left[ e^{-i\onk \, \ln

6647: \sin(\pi\tl)} |k^0=\onk, \vec k_\parallel=0, \vec k_\perp\ra +

6648: e^{i\onk

6649: \, \ln

6650: \sin(\pi\tl)} |k^0=-\onk, \vec k_\parallel=0, \vec k_\perp\ra\right]

6651: \, .

6652: \nonumber \\

6653:  \een

6654: Since $\wt f(x^0)$ vanishes as $x^0\to\infty$, in this limit we

6655: should be left with pure closed string background satisfying free

6656: field

6657: equations of motion $(Q_B+\bar Q_B)|\Psi_c^{(1)}\ra=0$, {\it i.e.}

6658: on-shell

6659: closed string field

6660: configuration\cite{0304192}. This can be verified

6661: explicitly\cite{0304192,0402157}.

6662: 

6663: 

6664: One amusing point to note is that \refb{es9.17bb} does not vanish even in

6665: the $\tl\to {1\over 2}$ limit, although the boundary state

6666: $|\BB_1\ra$

6667: vanishes in this limit. This is because in the euclidean theory the

6668: boundary state $|\BB_1\ra$ for $\tl={1\over 2}$ represents an array of

6669: D-branes with Dirichlet

6670: boundary condition on $X=i X^0$, located at $x=(2n+1)\pi$. This produces

6671: a non-trivial background in the euclidean theory, which, upon

6672: the replacement $x\to i x^0$, produces a source free closed string

6673: background in the Minkowski

6674: theory\cite{0304192}.

6675: 

6676: \refb{es9.17bb} gives the on-shell closed string radiation produced by the

6677: rolling tachyon background. In appendix \ref{appa} we have computed

6678: the energy $\EE$ per unit

6679: $p$-volume carried by this radiation. The answer is

6680:  \be \label{eedensityr}

6681: \EE = \sum_N \EE_N = 4\, K \, (g_s \TT_p)^2 \, \sum_N \, s_N \,

6682: \int {d^{25-p} k_\perp\over (2\pi)^{25-p}} \, {\pi^2\over

6683: \sinh^2(\pi\onk)} \, ,

6684:  \ee

6685: where $s_N$ is defined through the generating function:

6686:  \ben \label{esn3r}

6687: \sum_N s_N \, q^{2N} &=&  {1\over 2} \, \la 0 |

6688: \exp\left[\sum_{m=1}^\infty \sum_{s=1}^{24} (-1)^{d_s} \, {1\over

6689: m} \,  \alpha^s_{m} \bar \alpha^s_{m}\right]  q^{L_0^{matter}+\bar

6690: L_0^{matter}} \nonumber \\  && \exp\left[\sum_{n=1}^\infty

6691: \sum_{r=1}^{24} (-1)^{d_r} \, {1\over n} \,  \alpha^r_{-n} \bar

6692: \alpha^r_{-n}\right] \, | 0\ra''_{matter}\, .

6693:  \een

6694: Here $\la \cdot|\cdot\ra''_{matter}$ denotes the BPZ inner product in

6695: the

6696: matter sector with the normalization convention $\la

6697: 0|0\ra''_{matter}=1$. $d_s$ is an integer which can take values 0 or 1

6698: but the final answer is independent of $d_s$. The $N$-th term $\EE_N$ in

6699: the

6700: sum in eq.\refb{eedensityr} gives the total energy carried by all the

6701: closed string modes at level $(N,N)$.

6702: 

6703: 

6704: Since $g_s \TT_p\sim 1$, $K$ is a numerical constant determined

6705: using eq.\refb{ezopen}, and $s_N$ is a dimensionless

6706: number we see that for a fixed $N$, $\EE_N$

6707: is of order 1. Thus the total energy per unit $p$-volume carried by closed

6708: string states of a

6709: given mass$^2$ level is of order unity. Since for small $g_s$ this is much

6710: smaller that

6711: $\TT_p\sim (g_s)^{-1}$ -- the tension of the D-$p$-brane --

6712: we see that the amount of energy per unit $p$-volume carried by the closed

6713: string modes at a

6714: given level is much smaller than the tension of the brane. However the

6715: question that we need to address is whether the

6716: total energy density $\EE$ carried by all the modes of the closed string

6717: is also small

6718: compared to $\TT_p$. For this we need to estimate the large $N$

6719: behaviour

6720: of $s_N$ and also

6721: need to carry out the momentum integral in \refb{eedensityr}. Let

6722: us first do the momentum integral. Since for large $N$,

6723: $m_N\equiv \sqrt{4(N-1)}$ is large, \refb{defonk} reduces to

6724:  \be \label{es9.21}

6725: \onk \simeq m_N + {\vec k_\perp^2 \over 2 m_N} \, .

6726:  \ee

6727: This gives

6728:  \be \label{es9.21a}

6729: \sinh(\pi\onk) \simeq {1\over 2} \exp\left(\pi m_N + \pi \, {\vec

6730: k_\perp^2

6731: \over 2 m_N} \right) \, .

6732:  \ee

6733: The integration over $\vec k_\perp$ then becomes gaussian integral and

6734: gives:

6735:  \be \label{es9.22}

6736: \int {d^{25-p} k_\perp \over (2\pi)^{25-p}} \, \, \left({\pi \over

6737: \sinh(\pi\onk)}\right)^2 \simeq 4\pi^2 \, {1 \over (2\pi)^{25-p}}

6738: \exp(-4\pi \sqrt{N}) \, (4N)^{(25-p)/4} \, .

6739:  \ee

6740: The dominant contribution comes from $|k_\perp| \sim (m_N)^{1/2}$.

6741: 

6742: To find the large $N$ behaviour of $s_N$, we note from \refb{esn3r}

6743: that $2s_N$ simply

6744: counts the number of closed string states at oscillator level $(N,N)$ 

6745: which

6746: are created by identical combination of $X^s$ oscillators

6747: in the left and the

6748: right sector. Alternatively this can be regarded as the number of states

6749: at oscillator level $N$ created by the 24 right moving oscillators

6750: $\alpha^s_{-n}$ acting on the Fock vacuum. For large $N$ this is given

6751: by\cite{0303139}:

6752:  \be \label{es9.19}

6753: 2 s_N \simeq {1\over \sqrt 2} \, N^{-27/4} \, \exp(4\pi\sqrt{N})

6754: \, .

6755:  \ee

6756: Substituting \refb{es9.22} and \refb{es9.19} into \refb{eedensityr} we

6757: get:

6758:  \be \label{es9.23}

6759: \EE_N \simeq  \, K \, (g_s \TT_p)^2 \, \pi^2 \, {2^{(30-p)/2}

6760: \over (2\pi)^{25-p}} \, N^{-1/2-p/4}\, ,

6761:  \ee

6762: for large $N$. Thus $\EE=\sum_N\EE_N$

6763: is divergent for $p\le 2$. This shows that the total amount of

6764: energy per unit $p$-volume carried by {\it all} the closed string

6765: modes during the rolling of

6766: the tachyon is infinite\cite{0303139,0304192}!

6767: 

6768: {}From now on we shall focus on the $p=0$ case. For this the results

6769: obtained so far may be summarized as follows:

6770:  \begin{enumerate}

6771: \item Total amount of energy in closed string modes below any given mass

6772: level is finite. More precisely \refb{es9.23} for $p=0$ shows that the

6773: total energy carried by the closed string modes of mass less than

6774: some fixed value $M$ is proportional to $\sum_{N\le M^2/4} N^{-1/2}

6775: \sim M$.

6776: 

6777: \item The total amount of energy in all the closed string modes is

6778: infinite since the sum over $N$ diverges.

6779: 

6780: \item The contribution to the energy of a closed string mode of mass

6781: $m_N$ comes predominantly from modes with momentum $|\vec k_\perp|\sim

6782: (m_N)^{1/2}$.

6783: 

6784: \end{enumerate}

6785: 

6786: Of course since the D0-brane has a finite energy, the total energy

6787: carried by the closed string fields cannot

6788: really be infinite. We should expect that once the backreaction of the

6789: closed string emission process on the rolling of the tachyon is taken into

6790: account there will be a natural upper cut-off on the sum over $N$ so that

6791: we get a finite answer.

6792: In particular since

6793: the original D0-brane has energy of order $1/g_s$, it

6794: suggests that the backreaction of the closed string emission on the

6795: rolling tachyon solution will put a natural cut-off of order $1/g_s$ on

6796: the emitted closed string modes.

6797: In

6798: that case the results of the calculation may be reinterpreted as follows:

6799: 

6800:  \begin{enumerate}

6801: 

6802: \item All the energy of the D0-brane is converted into closed string

6803: radiation.

6804: 

6805: \item Most of the energy is stored in the closed string modes of mass

6806: $\sim g_s^{-1}$. This follows from the fact that the total energy

6807: carried by all closed string modes of mass$\le M$ is of order $M$, and

6808: for $M<< g_s^{-1}$ this energy is small compared to $g_s^{-1}$.

6809: 

6810: \item Typical momentum of these closed string modes is of order

6811: $g_s^{-1/2}$.

6812: 

6813: \end{enumerate}

6814: 

6815: If we have a D-$p$-brane with all its tangential directions compactified

6816: on a torus, then it is related to the D0-brane via T-duality, and

6817: hence we expect that similar results will hold for this system as well. In

6818: particular since under a T-duality transformation momentum along a circle

6819: gets mapped to the winding charge along the dual circle, we expect the

6820: following results:

6821: 

6822: 

6823:  \begin{enumerate}

6824: 

6825: \item All the energy of the D$p$-brane wrapped on a torus is converted

6826: into closed string

6827: radiation.

6828: 

6829: \item Most of the energy is stored in the closed string modes of mass

6830: $\sim g_s^{-1}$.

6831: 

6832: \item For these closed string modes, the typical momentum along

6833: directions

6834: transverse to the brane and typical winding along directions tangential to

6835: the brane are of order

6836: $g_s^{-1/2}$.

6837: 

6838: \end{enumerate}

6839: 

6840: These results suggest that the effect of closed string emission from a 

6841: D-brane produces a large backreaction and invalidates the classical open 

6842: string analysis. However a different interpretation based on the open 

6843: string completeness conjecture 

6844: will be discussed in section \ref{sph}.

6845: 

6846: \subsection{Closed string fields produced by $|\BB_2\ra$} \label{ecc2}

6847: 

6848: We now turn to the analysis of closed string fields generated by

6849: $|\BB_2\ra$ given in \refb{ebb2}.

6850: Since the state $ \wh\OO^{(n)}_N

6851: \, (c_0+\bar c_0) \, c_1 \, \bar c_1 \, e^{n X^0(0)} \, |\vec

6852: k_\perp\ra$ appearing in \refb{ebb2} is an eigenstate of

6853: $2(L_0+\bar L_0)$ with eigenvalue $(4(N-1)+n^2 + \vec k_\perp^2)$,

6854: the natural choice of the closed string field produced by

6855: $|\BB_2\ra$, obtained by replacing $|\BB\ra$ by 

6856: $|\BB_2\ra$ in eq.\refb{es9.1}, is

6857:  \ben \label{ebb2cl}

6858: |\Psi_c^{(2)}\ra &=& 2 \,  K\, g_s^2 \, \TT_p \, \sum_{n\in Z} \,

6859: \sum_{N=1}^\infty

6860: \,  \int \,{ d^{25-p}

6861: k_\perp \over (2\pi)^{25-p}} \,

6862: \left(4(N-1)+n^2 + \vec

6863: k_\perp^2\right)^{-1} \nonumber \\

6864: && \qquad \qquad \qquad \qquad \qquad \, \wh \OO^{(n)}_N

6865: \, c_1 \, \bar c_1 \, e^{n X^0(0)} \, |k^0=0, \vec k_\parallel=0,\vec

6866: k_\perp\ra\, .

6867:  \een

6868: Clearly, this is the result that we shall get if we begin with the

6869: closed

6870: string background produced by the boundary state in the euclidean

6871: theory

6872: and then analytically continue it to the Minkowski space.

6873: 

6874: Since the source for the closed string fields produced by $|\BB_2\ra$

6875: is localized at $\vec x_\perp=0$, $|\Psi^{(2)}_c\ra$ should satisfy

6876: source

6877: free closed

6878: string

6879: field equations away from the origin. It is easy to see that this is

6880: indeed the case\cite{0402157}.

6881: The space-time

6882: interpretation of this state for a given value of $n$ is that it

6883: represents a field which grows as $e^{n x^0}$. For positive $n$ this

6884: diverges as $x^0\to\infty$ and for negative $n$ this diverges as

6885: $x^0\to-\infty$.

6886: On the other hand in the transverse spatial directions the solution

6887: falls

6888: off as

6889: $G(\vec x_\perp, \sqrt{4(N-1) +n^2})$ where $G(\vec x_\perp, m)$

6890: denotes

6891: the Euclidean Green's function of a scalar field of mass $m$ in

6892: $(25-p)$

6893: dimensions. Since $G(\vec x_\perp, m) \sim e^{-m |\vec x_\perp|}/

6894: |\vec

6895: x_\perp|^{(24-p)/2}$ for large $|\vec x_\perp|$, we

6896: see

6897: that in position space representation the closed 

6898: string field associated with the

6899: state $\wh \OO^{(n)}_N c_1\bar c_1|k\ra$ behaves as

6900:  \be \label{eexpon}

6901: \exp\left(n x^0 - \sqrt{4(N-1) +n^2} \, |\vec x_\perp|\right)/ |\vec

6902: x_\perp|^{(24-p)/2}\,

6903:  \ee

6904: for large $|\vec x_\perp|$.

6905: Thus at any given time $x^0$, the field associated with $\wh

6906: \OO^{(n)}_N

6907: c_1\bar c_1|k\ra$ is small for $|\vec x_\perp|>> n x^0/\sqrt{4(N-1)

6908: +n^2}$

6909: and large for $|\vec x_\perp| <<  n x^0/\sqrt{4(N-1) +n^2}$. We can

6910: view such a field configuration as a disturbance

6911: propagating outward in

6912: the transverse directions from

6913: $\vec x_\perp=0$ at a speed of $ n/\sqrt{4(N-1) +n^2}$. For $N>

6914: 1$

6915: this

6916: is less than the speed of light but approaches the speed of light for

6917: fields for which $N-1<< n^2$.

6918: 

6919: \refb{eexpon} shows that for any $\vec x_\perp$, the closed string

6920: field configuration eventually grows to a value much larger than

6921: 1, and hence the linearized closed string field equation which we

6922: have used for this computation is no longer valid. This also suggests that 

6923: the classical open string analysis of the rolling tachyon solution will 

6924: suffer a large backreaction due to these exponentially growing closed 

6925: string fields. However a different 

6926: interpretation of this phenomenon will be discussed in section

6927: \ref{sph} in the context of open string completeness conjecture.

6928: 

6929: We have already seen earlier that

6930: $|\wh\BB\ra_{c=1}$ and hence

6931: $|\BB_2\ra$

6932: vanishes for $\tl={1\over 2}$. As a result the operators

6933: $\wh\OO^{(n)}_N$

6934: defined

6935: through \refb{ebb2} vanish, and hence $|\Psi_c^{(2)}\ra$ given in

6936: \refb{ebb2cl} also vanishes. Thus in the $\tl\to {1\over 2}$ limit the

6937: $|\Psi_c^{(1)}\ra$ given in \refb{es9.17bb} is the only contribution

6938: to

6939: the closed string background. This of course is manifestly finite in

6940: the

6941: $x^0\to\infty$ limit (although, as we have seen, it carries infinite

6942: energy).

6943: 

6944: There is an alternative treatment\cite{0301038} of the boundary state

6945: associated with the rolling tachyon solution in which the

6946: exponentially growing contributions to $|\BB_2\ra$ are absent

6947: altogether. This follows a different analytic continuation

6948: prescription in which instead of beginning with the Euclidean $c=1$

6949: theory of a scalar field $X$, we begin with a theory with $c>1$

6950: by giving the scalar field $X$ a small amount of background charge. We

6951: then analytically continue the results to the Minkowski space and then

6952: take the $c\to 1$ limit. In the context of the 26 dimensional critical

6953: string theory that we have been discussing, we do not have any

6954: independent way of deciding which of

6955: the analytic continuation procedure is correct.

6956: We shall however

6957: see in section \ref{s11} that at least in the context of two

6958: dimensional string theory the

6959: exponentially growing terms in $|\BB_2\ra$ do carry some physical

6960: information about

6961: the system.

6962: 

6963: Other aspects of closed string field produced by 

6964: unstable D-branes and other time dependent configurations have been 

6965: discussed in 

6966: \cite{0005242,

6967: 0204071,0204144,

6968: 0204191,0205198,0207004,

6969: 0207089,0209222,

6970: 0212150,0301095,

6971: 0303035,0305055,

6972: 0305191,

6973: 0306096,0307034,

6974: 0307078,

6975: math-ph/0308034,

6976: 0309017,0403050,

6977: 0403147,0403156,0404039,

6978: 0407147,

6979: 0409019,

6980: 0409044}.

6981: 

6982: \sectiono{D0-brane 

6983: Decay in Two Dimensional String Theory} \label{s11}

6984: 

6985: In the last few sections we have discussed various aspects of the 

6986: dynamics of 

6987: unstable

6988: D-branes in critical string theory. Due to the complexity of the

6989: problem our analysis has been restricted mostly to the

6990: level of disk amplitudes. In this section we shall study the process of 

6991: D-brane decay

6992: in two dimensional string

6993: theory\cite{0304224,0305159,0305194,0307083,0307195}. One of the

6994: reasons for doing this is that in this theory we can carry out the

6995: analysis in two different ways: 1) by regarding this as an

6996: ordinary string theory\cite{KPZ,DAVID,DISKAW} and applying the

6997: techniques developed in the earlier sections for studying the

6998: dynamics of unstable D-branes, and 2) by using an exact description

6999: of the theory in terms of matrix model\cite{GROMIL,BKZ,GINZIN}. We

7000: shall see that the matrix model results agree with the open string

7001: tree level analysis in the appropriate limit, and allows us to

7002: extend the results beyond tree level. In section \ref{sopenclosed} we

7003: shall see that these all order results lend support to the open

7004: string completeness conjecture that will be formulated in that

7005: section.

7006: 

7007: \subsection{Two dimensional string theory} \label{s2d}

7008: 

7009: We begin by reviewing the bulk conformal field theory associated with

7010: the

7011: two dimensional string theory. The world-sheet action of this

7012: CFT is given by the sum of three separate components:

7013:  \be \label{e11.1}

7014: s = s_L + s_{X^0} + s_{ghost}\, ,

7015:  \ee

7016: where $s_L$ denotes the Liouville field theory with central charge 25,

7017: $s_{X^0}$ denotes the conformal field theory of a single scalar field

7018: $X^0$ describing the time coordinate

7019: and

7020: $s_{ghost}$ denotes the usual

7021: ghost action involving the fields $b$, $c$, $\bar b$ and $\bar c$. Of

7022: these $s_{X^0}$ and $s_{ghost}$ are familiar objects. The Liouville

7023: action

7024: $s_L$ on a flat world-sheet is given by:

7025:  \be \label{e11.2}

7026: s_L = \int d^2 z \left ({1\over 2\pi}\, \p_z\vp \p_{\bar z} \vp + \mu

7027: e^{2\vp}\right)

7028:  \ee

7029: where $\vp$ is a world-sheet scalar field and $\mu$ is a constant

7030: parametrizing the theory. We shall set $\mu=1$ by shifting $\vp$

7031: by ${1\over 2} \ln\mu$. The scalar field $\vp$ carries a

7032: background charge $Q=2$ which is not visible in the flat

7033: world-sheet action \refb{e11.2} but controls the coupling of $\vp$

7034: to the scalar curvature on a curved world-sheet. This is

7035: equivalent to switching on a background dilaton field

7036:  \be \label{ebdil}

7037: \Phi_D=Q\vp=2\vp\, .

7038:  \ee

7039: The resulting theory has a central 

7040: charge

7041:  \be \label{e11.3}

7042: c = 1 + 6 Q^2 = 25\, .

7043:  \ee

7044: Note that for large negative $\vp$ the potential term in \refb{e11.2}

7045: becomes small and $\vp$ behaves like a free scalar field with background

7046: charge. Also in this region the 

7047: string coupling constant $e^{\Phi_D}$ is small.

7048: 

7049: 

7050: For our analysis we shall not use the explicit world-sheet action

7051: \refb{e11.2},

7052: but

7053: only use the abstract properties of the Liouville field theory

7054: described

7055: in \cite{9206053,9403141,9506136,0101152,0104158,0207041,0303150,0311202}. 

7056: In particular

7057: the

7058: property of

7059: the bulk

7060: conformal

7061: field

7062: theory that we shall be using is that it has a one parameter ($P$)

7063: family

7064: of primary

7065: vertex operators, denoted by $V_{Q + i P}$, of conformal weight:

7066:  \be \label{e11.4}

7067: \left({1\over 4} (Q^2 + P^2), {1\over 4} (Q^2 + P^2)\right) =

7068: \left(1+{1\over 4} P^2, 1+{1\over 4} P^2\right)\, .

7069:  \ee

7070: A generic $\delta$-function normalizable state of the bulk Liouville

7071: field theory

7072: is

7073: given by a

7074: linear combination of the secondary states built over the primary

7075:  \be \label{defpl}

7076: |P\ra= V_{Q+iP}(0)|0\ra\, , \quad P\quad \hbox{real}\, ,

7077:  \ee

7078: where $|0\ra$ denotes the

7079: SL(2,C) invariant vacuum in the Liouville field theory.

7080: 

7081: For large negative $\vp$ the world-sheet scalar field $\vp$ behaves like a

7082: free field,

7083: and hence one might expect that the primary vertex operators in this

7084: region take the form $e^{iP\vp}$. There are however two subtleties.

7085: First, due to the linear dilaton background, the delta-function 

7086: normalizable

7087: vertex operators are not of the form $e^{iP\vp}$ but of the form

7088: $e^{(Q+iP)\vp}$. Also due to the presence of the exponentially growing

7089: potential for large positive $\vp$ we effectively have a wall that

7090: reflects any incoming wave into an outgoing wave of equal and opposite

7091: $\vp$-momentum. Thus a primary vertex operator should be an appropriate

7092: linear superposition of $e^{(Q+iP)\vp}$ and $e^{(Q-iP)\vp}$ for large 

7093: negative $\vp$. $V_{Q+iP}$

7094: represents precisely this  vertex operator. In particular for $Q=2$,

7095: $V_{2+iP}$ has the asymptotic form

7096:  \be \label{evasymp}

7097: V_{2+iP} \simeq e^{(2+iP)\vp} - \left({\Gamma(iP)\over

7098: \Gamma(-iP)}\right)^2  e^{(2-iP)\vp}\, .

7099:  \ee

7100: With this choice of normalization the primary states $|P\ra$ satisfy

7101:  \be \label{enormp}

7102: \la P|P'\ra_{liouville} = 2\pi\left( \delta(P+P') - \left({\Gamma(iP)\over

7103: \Gamma(-iP)}\right)^2 \delta(P-P')\right)\, ,

7104:  \ee

7105: where $\la\cdot|\cdot\ra_{liouville}$ denotes BPZ inner product in the

7106: Liouville sector.

7107: {}From this analysis we see that the $V_{2+iP}$ and $V_{2-iP}$ should not

7108: be regarded as independent vertex operators. Instead there is an

7109: identification\cite{9403141,9506136}

7110:  \be \label{evidenti}

7111: V_{2+iP} \equiv -  \left({\Gamma(iP)\over

7112: \Gamma(-iP)}\right)^2 \, V_{2-iP}\, .

7113:  \ee

7114: 

7115: The closed string field $|\Psi_c\ra$ in this two dimensional string

7116: theory is a ghost

7117: number 2 state satisfying \refb{efg1.5} in the combined state space

7118: of the

7119: ghost, Liouville and $X^0$ field theory. We can expand $|\Psi_c\ra$ as

7120:  \be \label{eexppsi}

7121: |\Psi_c\ra = \int_0^\infty {d P\over 2\pi} \, \int_{-\infty}^\infty\, {d 

7122: E\over 2\pi} \,

7123: \wt\phi(P,

7124: E) \,

7125: c_1 \bar c_1 e^{-i E

7126: X^0(0)} |P\ra + \cdots \, ,

7127:  \ee

7128: where $\cdots$ denote higher level terms. Note that due to the reflection

7129: symmetry \refb{evidenti} we have restricted the range of $P$ integration

7130: to be from 0 to $\infty$. $\wt\phi$ may be regarded as the Fourier

7131: transform of a scalar field $\phi(\vp,x^0)$:

7132:  \be \label{edefftphi}

7133: \phi(\vp, x^0) = \int_0^\infty {dP\over 2\pi} \int_{-\infty}^\infty \,

7134: {dE\over 2\pi} \, e^{-iEx^0} \, \left(e^{i P \vp} -\left({\Gamma(iP)\over

7135: \Gamma(-iP)}\right)^2 \,

7136: e^{-i P\vp}\right)\,

7137: \wt\phi(P, E)

7138:  \ee

7139: for large negative $\vp$. $\phi(\vp,x^0)$ is known as the closed

7140: string tachyon field.\footnote{Throughout this section we shall

7141: use the same symbol  to denote a field and its Fourier transform

7142: with respect to $x^0$. However for the liouville coordinate $\vp$,

7143: a field in the momentum space representation will carry a `tilde'

7144: whereas the corresponding field in the position space

7145: representation will be denoted by the same symbol without a

7146: `tilde'. \label{fn45}} Despite its name, it actually describes a massless

7147: particle in this $(1+1)$ dimensional string theory, since the

7148: condition that the state $c_1 \bar c_1 e^{-i E X^0(0)} |P\ra$ is

7149: on-shell is $E^2 - P^2=0$. This is the only physical closed string

7150: field in this theory. The condition that $\phi(\vp, x^0)$ is real

7151: translates to the following condition on $\wt\phi(P,E)$:

7152:  \be \label{ereal}

7153:  \wt\phi(P,E) = - \left( {\Gamma(-iP)\over \Gamma(iP)}\right)^2 \,

7154:  \wt\phi^*(P,-E) \, .

7155:  \ee

7156: 

7157: We shall normalize $|\Psi_c\ra$ so that its kinetic term is given by:

7158:  \be \label{ekinlio}

7159: -\la \Psi_c|c_0^- (Q_B+\bar Q_B) |\Psi_c\ra\, .

7160:  \ee

7161: Substituting \refb{eexppsi} into \refb{ekinlio} and using \refb{enormp}

7162: we see that the

7163: kinetic

7164: term for $\wt\phi$ is given by:

7165:  \be \label{ekinphi}

7166: {1\over 2} \, \int_0^\infty {dP\over 2\pi} \, 

7167: \int_{-\infty}^\infty {dE\over 2\pi} \, \wt\phi(P,

7168: -E)

7169: (P^2

7170: -

7171: E^2) \wt\phi(P,E)\, \left({\Gamma(iP)\over

7172: \Gamma(-iP)}\right)^2 \, .

7173:  \ee

7174: Using \refb{edefftphi} this may be expressed as

7175:  \be \label{eactionphi}

7176: -{1\over 2} \int dx^0\, \int d\vp \, (\p_0\phi\p_0\phi - \p_\vp\phi

7177: \p_\vp\phi) \, ,

7178:  \ee

7179: in the large negative $\vp$ region.

7180: Thus $\phi$ is a scalar field with conventional normalization.

7181: 

7182: 

7183: If we define a new field

7184:  \be \label{edefpsi}

7185: \wt\psi(P,E) = \left({\Gamma(iP)\over

7186: \Gamma(-iP)}\right) \, \wt\phi(P,E)\, ,

7187:  \ee

7188: then \refb{ekinphi} may be written as

7189:  \be \label{enewac}

7190: {1\over 2} \, \int_0^\infty {dP\over 2\pi} \, 

7191: \int_{-\infty}^\infty \, {dE\over 2\pi} \, \wt\psi(P,

7192: -E)

7193: (P^2

7194: -

7195: E^2) \wt\psi(P,E)\, .

7196:  \ee

7197: If we consider the Fourier transform $\psi(\vp,x^0)$ of $\wt\psi$, defined

7198: through

7199:  \be \label{edefftpsi}

7200: \psi(\vp, x^0) = \int_0^\infty {dP\over 2\pi} \int_{-\infty}^\infty \,

7201: {dE\over 2\pi} \, e^{-iEx^0} \, \left(e^{i P \vp} - e^{-i P\vp}\right)\,

7202: \wt\psi(P, E)

7203:  \ee

7204: for large negative $\vp$,

7205: then the action in terms of $\psi$ takes the form:

7206:  \be \label{eactionpsi}

7207: -{1\over 2} \int dx^0\, \int d\vp \, (\p_0\psi\p_0\psi - \p_\vp\psi

7208: \p_\vp\psi) \, .

7209:  \ee

7210: Thus $\psi$ is also a scalar field with conventional

7211: normalization. Also the reality condition \refb{ereal} guarantees

7212: that the field $\psi(\vp, x^0)$ defined in \refb{edefftpsi} is

7213: real. Although in the momentum space $\wt\psi$ and $\wt\phi$ are

7214: related to each other by multiplication by a phase factor, this

7215: translates to a non-local relation between the two fields in the

7216: position space.

7217: 

7218: \subsection{D0-brane and its boundary state in two dimensional string

7219: theory} \label{sdb2d}

7220: 

7221: The two dimensional string theory also has an unstable D0-brane

7222: obtained by

7223: putting an appropriate boundary condition on the world-sheet field $\vp$,

7224: and the

7225: usual Neumann boundary condition on $X^0$ and the ghost

7226: fields\cite{0305159,0305194}.

7227: Since

7228: $\vp$

7229: is an interacting field in the world-sheet theory, it is more

7230: appropriate to describe the

7231: corresponding

7232: boundary CFT associated with the Liouville field

7233: by specifying its abstract properties. The relevant properties are as

7234: follows:

7235:  \begin{enumerate}

7236: 

7237: \item The open string

7238: spectrum in this boundary CFT is described by a single Virasoro module

7239: built

7240: over the SL(2,R) invariant vacuum state.

7241: 

7242: \item In the Liouville theory the one point function on the disk of the

7243: closed string vertex

7244: operator $V_{Q +

7245: i P}$ is given

7246: by\cite{0101152,0305159}:

7247:  \be \label{e11.5}

7248: \la V_{Q +

7249: i P} \ra_D = {2\ C\over \sqrt \pi} \, i\, \sinh(\pi P) \,

7250: {\Gamma(i P)\over \Gamma(-iP)}\, ,

7251:  \ee

7252: where $C$ is a normalization constant to be determined in

7253: eq.\refb{e11.5cc}.

7254: \end{enumerate}

7255: Since $V_{Q +

7256: i P}$ for any real $P$ gives the complete set of primary states in the

7257: theory, we get the boundary state associated with the D0-brane to

7258: be:\footnote{The normalization factor of $1/2$ has been included for

7259: convenience so as to compensate for the factor of 2 in the ghost

7260: correlator

7261: $\la

7262: 0|c_{-1}\bar c_{-1} c_0^- c_0^+ c_1\bar c_1|0\ra$. In the end the overall

7263: normalization of $|\BB\ra$,

7264: encoded in

7265: the constant $C$, will be determined from eq.\refb{eformb} by requiring

7266: that the

7267: classical action reproduces correctly the one loop string partition

7268: function.}

7269:  \ben \label{eboulio}

7270: |\BB\ra &=& {1\over2} \, \exp\left(\sum_{n=1}^\infty

7271: {1\over

7272: n} \alpha^0_{-n} \bar

7273: \alpha^0_{-n} \right)|0\ra \, \otimes \, \int_0^\infty {d P\over 2\pi} \,

7274: \la

7275: V_{Q

7276: -

7277: i P} \ra_D |P\ra\ra \nonumber \\

7278: && \otimes \, \exp\left(-\sum_{n=1}^\infty

7279: (\bar b_{-n}

7280: c_{-n} +

7281: b_{-n}

7282: \bar c_{-n})

7283: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\, ,

7284:  \een

7285: where $|P\ra\ra$ denotes the

7286: Ishibashi state in the Liouville theory, built on the primary

7287: $|P\ra\equiv V_{Q+iP}(0)|0\ra$.\footnote{Note that $|P=0\ra$ is not the 

7288: SL(2,R) invariant vacuum $|0\ra$ in the Liouville sector.}

7289: The normalization constant $C$ is

7290: determined

7291: by requiring that if we eliminate $|\Psi_c\ra$ from the combined

7292: action

7293:  \be \label{ecombined}

7294: -\la \Psi_c|c_0^- (Q_B+\bar Q_B) |\Psi_c\ra + \la \Psi_c|c_0^-

7295: |\BB\ra\, ,

7296:  \ee

7297: using its equation of motion, then the resulting value of the action

7298: reproduces the one loop open string partition function $Z_{open}$ on

7299: the

7300: D0-brane. In fact, in this case the contribution from the higher

7301: closed string modes cancel (with the contribution from $b$ and $c$ 

7302: oscillators cancelling the contribution from the $X^0$ oscillator and the 

7303: Liouville Virasoro generators). Thus we can restrict $|\Psi_c\ra$ 

7304: to only the closed string tachyon mode given in eq.\refb{eexppsi}. 

7305: Substituting \refb{eexppsi} into

7306: \refb{ecombined}, and expressing the result in terms of the field 

7307: $\wt\psi(P,E)$ defined

7308: in \refb{edefpsi}, we get

7309:  \be \label{eforma}

7310: -{1\over 2} \, \int_0^\infty {d P\over 2\pi} \, 

7311: \int_{-\infty}^\infty \, {dE\over 2\pi} \,

7312: \wt\psi(P,-E) (P^2 - E^2) \, \wt\psi(P,E) + \int_0^\infty {dP\over 2\pi}

7313: \la

7314: V_{Q+iP}\ra_D \,  {\Gamma(-iP)\over \Gamma(iP)} \,

7315: \wt\psi(P,0)\, .

7316:  \ee

7317: Eliminating $\wt\psi$ using its equation of motion and requiring that the

7318: answer is equal to

7319: $Z_{open}$ gives

7320:  \ben \label{eformb}

7321: Z_{open} &=& -{1\over 2}\, T \, \int_0^\infty {dP\over 2\pi} \, \left(\la

7322: V_{Q+iP}\ra_D\right)^2 \, {1\over P^2}\, \left( {\Gamma(-iP)\over

7323: \Gamma(iP)}\right)^2 \nonumber \\

7324: &=& {2 C^2\over \pi} \, T\, \int_0^\infty {dP\over 2\pi} \, \sinh^2(\pi P)

7325: \, {1\over P^2}\, ,

7326:  \een

7327: where $T=2\pi\delta(E=0)=\int dx^0$ denotes the total length of the

7328: time interval.

7329: We can rewrite this as

7330:  \be \label{erew1}

7331: Z_{open} = {2 C^2\over \pi} \, T\, \int_0^\infty ds \int_0^\infty {dP\over

7332: 2\pi} \, e^{-s P^2} \, {1\over 4} \, (e^{2\pi P} + e^{-2\pi P} -2)\, .

7333:  \ee

7334: After doing the $P$ integral and making a change of variable $t=\pi/2s$ we

7335: get

7336:  \ben \label{erew2}

7337: Z_{open} &=& {T\over 4\sqrt 2 \pi} \, C^2 \, \int_0^\infty {dt\over

7338: t^{3/2}}

7339: \left(e^{2\pi t} - 1\right) \nonumber \\

7340: &=& T C^2 \int_0^\infty \, {dt\over 2t} \, \int_{-\infty}^\infty \,

7341: {dE\over 2\pi} \, e^{-2\pi t E^2} \, \left(e^{2\pi t} -1\right)\, .

7342:  \een

7343: This can be regarded as the one loop open string amplitude with the open

7344: string spectrum consisting of a single tachyonic mode of mass$^2=-1$ and a

7345: single gauge field,\footnote{In 0+1 dimension a gauge field produces a

7346: constraint and removes one degree of freedom. This is the origin of the

7347: $-1$ in \refb{erew2}.} provided we choose\cite{0305159}

7348:  \be

7349: \label{e11.5cc} C=1\, .

7350:   \ee

7351: Thus this choice of normalization reproduces the result that the open 

7352: string spectrum in the Liouville sector consists of a single Virasoro 

7353: module built over the SL(2,R) invariant vacuum . The open string tachyon 

7354: is associated with the state $c_1|0\ra$ and the gauge field is associated 

7355: with the state $\alpha^0_{-1} c_1 |0\ra$. 

7356: 

7357: We can now add a boundary interaction term $\tl\, \int dt \cosh(X^0(t))$ 

7358: to deform the free field theory involving the coordinate $X^0$ to

7359: the

7360: rolling tachyon boundary CFT, and leave the Liouville and the ghost parts 

7361: unchanged. This gives a rolling tachyon solution on the

7362: D0-brane in

7363: two dimensional string theory.

7364: As in the critical string theory, we divide the boundary state into

7365: two

7366: parts, $|\BB_1\ra$ and $|\BB_2\ra$, with

7367:  \ben \label{ebb1l}

7368: |\BB_1\ra &=& {1\over 2} \, \exp\left(-\sum_{n=1}^\infty

7369: {1\over

7370: n} \alpha^0_{-n} \bar

7371: \alpha^0_{-n} \right) \wt f(X^0(0)) |0\ra \, \otimes \,

7372: \int_0^\infty {dP\over

7373: 2\pi}

7374: \,

7375: \la

7376: V_{Q-iP}\ra_D \, |P\ra\ra \nonumber \\

7377: && \otimes

7378: \exp\left(-\sum_{n=1}^\infty (\bar

7379: b_{-n} c_{-n} +

7380: b_{-n}

7381: \bar c_{-n})

7382: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\nonumber \\

7383: &\equiv& {1\over 2} \, \int_0^\infty {dP\over 2\pi} \, \la

7384: V_{Q-iP}\ra_D \, \sum_N \, \wc A_N(P) \, \wt f(X^0(0))

7385: (c_0+\bar c_0) c_1 \bar c_1 |P\ra\, ,

7386:  \een

7387: and

7388:  \ben \label{ebb2l}

7389: |\BB_2\ra &=& {1\over 2} \, |\wh\BB\ra_{c=1} \, \otimes \,

7390: \int_0^\infty {dP\over

7391: 2\pi} \,

7392: \la

7393: V_{Q-iP}\ra_D \, |P\ra\ra \nonumber \\

7394: && \otimes 

7395: \exp\left(-\sum_{n=1}^\infty

7396: (\bar

7397: b_{-n} c_{-n} +

7398: b_{-n}

7399: \bar c_{-n})

7400: \right) (c_0+\bar c_0)c_1\bar c_1 |0\ra\, \nonumber \\

7401: &\equiv& {1\over 2} \, \sum_{n\in Z} \, \sum_{N=1}^\infty \, \int_0^\infty

7402: {d P

7403: \over 2\pi} \, \la V_{Q-iP}  \ra_D

7404: \, \wc \OO^{(n)}_N(P)

7405: \,

7406: (c_0+\bar c_0) \, c_1 \, \bar c_1 \, e^{n X^0(0)} \,

7407: |P\ra\, .

7408:  \een

7409: Here $|\wh\BB\ra_{c=1}$ is the inverse Wick rotated version of $|\wt

7410: B\ra_{c=1}$ as defined in eq.\refb{esecondform}, and $\wc A_N(P)$ and

7411: $\wc\OO^{(n)}_N(P)$ are operators of

7412: level $(N,N)$,

7413: consisting of non-zero mode ghost and $X^0$ oscillators, and the

7414: Virasoro

7415: generators

7416: of the Liouville theory. The $P$ dependence of $\wc A_N$ and $\wc 

7417: O^{(n)}_N$ originates from the fact that the Virasoro Ishibashi state in 

7418: the Liouville sector, when expressed as a linear combination of Liouville 

7419: Virasoro generators acting on the primary $|P\ra$, has $P$ dependent 

7420: coefficients.

7421: 

7422: As in the case of critical string theory, it is easy to show that

7423: $|\BB_1\ra$ and $\BB_2\ra$ are separately BRST invariant.

7424: 

7425: \refb{ebb1l} and \refb{ebb2l} shows that the sources for the various

7426: closed string fields in the momentum space are proportional to $ \la

7427: V_{Q-iP}  \ra_D$. It is instructive to see what they correspond to in

7428: the

7429: position space labelled by the Liouville coordinate $\vp$. We

7430: concentrate

7431: on the large negative $\vp$ region.

7432: In this region $V_{Q+iP}$ takes the form \refb{evasymp}

7433: and

7434:  \be \label{eprr}

7435: |P\ra\ra \sim \wh \OO_L(P) \, \left( e^{2\vp(0) +

7436: i P\vp(0)} |0\ra - \left( {\Gamma(iP)\over \Gamma(-iP)}\right)^2

7437: e^{2\vp(0) -

7438: i P\vp(0)} \right)|0\ra\, ,

7439:  \ee

7440: where $\wh\OO_L(P)$ is an appropriate operator in the Liouville

7441: field theory which is even under $P\to -P$.

7442: Thus the source terms are of the form

7443:  \ben \label{esource}

7444: &&\int_0^\infty {d

7445: P\over 2\pi} \, \left( e^{2 \vp + i

7446: P \,

7447: \vp}- \left( {\Gamma(iP)\over \Gamma(-iP)}\right)^2

7448: e^{2 \vp - i

7449: P \,

7450: \vp} \right) \la V_{Q-iP}\ra_D \, \wh O_L(P)|0\ra \otimes |s\ra_{X^0,g}

7451: \nonumber 

7452: \\

7453: &\propto&

7454: \int_{-\infty}^\infty {d

7455: P\over 2\pi} \, e^{2 \vp + i

7456: P \,

7457: \vp}  \, \, \sinh(\pi P) \,

7458: {\Gamma(-i P)\over \Gamma(iP)} \, \wh O_L(P)|0\ra \otimes |s\ra_{X^0,g}\, 

7459: ,

7460:  \een

7461: where $|s\ra_{X^0,g}$ denotes some state in the $X^0$ and ghost CFT.

7462: As it stands the integral is not well defined since $\sinh(\pi P)$

7463: blows

7464: up for large $|P|$. For negative $\vp$, we shall define this integral

7465: by

7466: closing the contour in the lower half plane, and picking up the

7467: contribution from all the poles enclosed by the contour. Since the

7468: poles

7469: of

7470: $\Gamma(-iP)$ at $P=-in$ are cancelled by the zeroes of $\sinh(\pi P)$

7471: we

7472: see that the integrand has no pole in the lower half plane and hence

7473: the

7474: integral

7475: vanishes.\footnote{In arriving at this conclusion we have to ignore poles 

7476: in $\wh O_L(P)$ in the complex $P$ plane. The residues at these poles 

7477: correspond to null states in the Liouville theory and are set to zero in 

7478: our analysis\cite{0408064}.} Thus the boundary state $|\BB_1\ra$ 

7479: and $|\BB_2\ra$ given in \refb{ebb1l} and \refb{ebb2l} do not produce any

7480: source term for large negative $\vp$. This in turn leads to the

7481: identification

7482: of this

7483: system as a D0-brane that is localized in the Liouville

7484: direction\cite{0305159}.

7485: 

7486: Following arguments similar to those for the critical string theory

7487: given at the end of section \ref{s10}, one can show that the time

7488: dependence of various terms in the $|\BB_2\ra$ component of the

7489: boundary state is fixed by the requirement of BRST invariance. This

7490: indicates

7491: that

7492: $|\BB_2\ra$ encodes information about conserved charges. To see

7493: explicitly

7494: what these conserved charges correspond to, we first need to

7495: express $|\BB_2\ra$ in a more suggestive form. Making the replacement 

7496: $x\to i x^0$ in eq.\refb{epriform} we get

7497:  \be \label{eneweq}

7498: |j,m\ra = \wh \PP_{j,m} \, e^{-2mX^0(0)}|0\ra\, ,

7499:  \ee

7500: where it is understood that in $\wh \PP_{j,m}$ we replace the $\alpha_n$'s 

7501: by $i\alpha^0_n$'s. Let us define  $\wh

7502: \NN_{j,m}$ to be

7503: an operator made of $\alpha^0_{-n}$,

7504: $\bar\alpha^0_{-n}$ for $n>0$

7505: such that the Ishibashi state $|j,m\ra\ra$ in the Minkowski $c=1$

7506: theory is

7507: given by:

7508:  \be \label{ediff3}

7509: |j,m\ra\ra = \wh \NN_{j,m} \, e^{-2m X^0(0)} \, |0\ra\, .

7510:  \ee

7511: We also define

7512: $\wh\RR_{j,m}(P)$ as

7513:  \be \label{ecrit1.5}

7514: \wh\RR_{j,m}(P) = \wh \NN_{j,m} \,

7515: \wh\OO_L(P)\, \exp\left(-\sum_{n=1}^\infty (\bar b_{-n} c_{-n}

7516: +

7517: b_{-n}

7518: \bar c_{-n})

7519: \right)\, ,

7520:  \ee

7521: where $\wh \OO_L(P)$ has been defined in eq.\refb{eprr}.

7522: {}From eqs.\refb{ediff3}, \refb{ecrit1.5} it is clear that $\wh

7523: \RR_{j,m}(P)$

7524: does not have

7525: any explicit $\tl$ dependence.

7526: We now use

7527: eqs.\refb{ebb2l}, \refb{eprr} and \refb{esecondform} to express

7528: $|\BB_2\ra$ as

7529:  \be \label{ecrit1}

7530: |\BB_2\ra = {1\over 2} \, \sum_{j=1}^\infty \, \sum_{m=-(j-1)}^{j-1}

7531: \,

7532: \int_0^\infty {d P\over

7533: 2\pi} \, \la V_{Q-iP}\ra_D\, f_{j,m} (\tl) \, \wh\RR_{j,m}(P) \,

7534: (c_0+\bar c_0)c_1\bar c_1 e^{-2 m X^0(0)} \, |k^0=0, P\ra\, ,

7535:  \ee

7536: where

7537: $f_{j,m}(\tl)$ has been defined in

7538: eq.\refb{efjm}.

7539: 

7540: If we now generalize

7541: the

7542: source term so that the boundary state has the

7543: same operator structure but

7544: arbitrary time dependence:

7545:  \be \label{ecrit2}

7546: |\BB_2\ra' ={1\over 2} \, \sum_{j=1}^\infty \, \sum_{m=-(j-1)}^{j-1}

7547: \,

7548: \int_0^\infty {d P\over

7549: 2\pi} \, \la V_{Q-iP}\ra_D \, \wh\RR_{j,m}(P) \,

7550: (c_0+\bar c_0)c_1\bar c_1 \, g_{j,m}(X^0(0)) \, |k^0=0, P\ra\, ,

7551:  \ee

7552: then requiring $(Q_B+\bar Q_B)|\BB_2\ra'=0$ gives:

7553:  \be \label{ecrit3}

7554: \p_0 \left(e^{2m x^0}  g_{j,m}(x^0)\right) = 0\, .

7555:  \ee

7556: Thus $e^{2m x^0}

7557: g_{j,m}(x^0)$

7558: can be thought of as a conserved charge which takes value $f_{j,m}

7559: (\tl)$

7560: for $|\BB_2\ra$ given in \refb{ecrit1}.

7561: In particular the total energy carried by the system is proportional

7562: to the conserved charge $g_{1,0}$\cite{0408064} and is given by

7563:  \be \label{eennc}

7564: E = {1\over g_s} \, \cos^2(\pi\tl)\, .

7565:  \ee

7566: This is the analog of eq.\refb{ehx4} 

7567: for the critical string theory. The

7568: $\tl =0$ configuration represents the original D0-brane which has mass

7569: $1/g_s$ in our convention for $g_s$.

7570: 

7571: 

7572: Note that this procedure for identifying the conserved charges requires

7573: {\it a priori} knowledge of the boundary state and hence does not provide

7574: us with a systematic method for computing these charges for an

7575: arbitrary boundary state. Such a systematic procedure was developed in

7576: \cite{0408064} where it was shown how given any boundary state in the two

7577: dimensional string theory one can construct infinite number of conserved

7578: charges. Applying this method to the present problem one finds that the

7579: conserved

7580: charges are proportional to the combinations $f_{j,m}(\tl) -

7581: f_{j-1,m}(\tl)$ for $|m|\le j-2$, and $f_{j,m}(\tl)$ for $|m|=j-1$. By

7582: taking

7583: appropriate linear combinations of these charges one can construct

7584: conserved charges whose values are directly given by $f_{j,m}(\tl)$.

7585: 

7586: 

7587: 

7588: \subsection{Closed string background produced by $|\BB_1\ra$}

7589: \label{s2closed}

7590: 

7591: We now calculate the closed string field produced by this time

7592: dependent

7593: boundary state\cite{0305159,0402157}.

7594: The contribution

7595: from

7596: the $|\BB_1\ra$ part of the boundary state can be easily computed as

7597: in

7598: the case of critical string theory, and in the $x^0\to\infty$ limit

7599: takes the form:

7600:  \be \label{eb1cont}

7601: |\Psi_c^{(1)}\ra \to \int_0^\infty {dP\over 2\pi} \la

7602: V_{Q-iP}\ra_D\, \sum_N \wc A_N(P) \, h^{(N+1)}_{

7603: P}(X^0(0))

7604: c_1 \bar c_1 |P\ra\, ,

7605:  \ee

7606: where $h^{(N)}_{\vec

7607: k_\perp}(x^0)$ has been defined in \refb{esa-1}. The $(N+1)$ in the

7608: superscript of $h$ in \refb{eb1cont} can be traced to the fact that in

7609: this theory a level $(N,N)$ state has mass$^2=4N$ whereas in the

7610: critical

7611: string theory a level $(N,N)$ state had mass$^2=4(N-1)$.

7612: 

7613: Since the source terms represented by $|\BB_1\ra$ vanish as 

7614: $x^0\to\infty$, in this limit

7615: $|\Psi_c^{(1)}\ra$ is on-shell, {\it i.e.} it is annihilated by the BRST

7616: charge

7617: $(Q_B+\bar Q_B)$. Since the only physical states in the theory come

7618: from

7619: the

7620: closed string `tachyon' field, it must be possible to remove all the

7621: other

7622: components of $|\Psi_c^{(1)}\ra$ by an on shell gauge transformation

7623: of

7624: the

7625: form

7626: $\delta |\Psi_c\ra = (Q_B+\bar Q_B)|\Lambda\ra$ by suitably choosing

7627: $|\Lambda\ra$. Thus the non-trivial contribution to $|\Psi_c^{(1)}\ra$

7628: is encoded completely in the field $\wt\phi(P,E)$ defined through 

7629: eq.\refb{eexppsi}.

7630: Furthermore, since the action of $Q_B$ and $\bar Q_B$

7631: does

7632: not

7633: mix states of different levels, the gauge transformation

7634: that removes the

7635: $N>0$ components of $|\Psi_c^{(1)}\ra$ does not modify the $N=0$

7636: component.

7637: Using \refb{eb1cont}, the expression for $h^{(1)}_{P}$ from

7638: \refb{es9.17aa}, \refb{defonk}, expression for $\la V_{Q-iP}\ra_D$ from 

7639: \refb{e11.5}, the

7640: definition of $\wt\phi(P,E)$ from \refb{eexppsi}, and the fact that $\wc

7641: A_{N=0}=1$,

7642: we get the following expression for the closed string tachyon field

7643: $\wt\phi(P,x^0)$ in the $x^0\to\infty$ limit\cite{0305159}

7644:  \ben \label{e11.6}

7645: \wt\phi(P, x^0\to\infty) &=& - {\pi\over

7646: \sinh(\pi\omega_P)} \, {1\over 2\omega_P} \, {2\over \sqrt \pi} \, i\,

7647: \sinh(\pi

7648: P) \,

7649: {\Gamma(-i P)\over \Gamma(iP)} \nonumber \\

7650: && \qquad \, \left[ e^{-i\omega_P (x^0 + \ln

7651: \sin(\pi\tl)) } +  e^{i\omega_P (x^0 + \ln

7652: \sin(\pi\tl)) } \right] \, ,

7653:  \een

7654: where $\omega_P = |P|$. As stated in footnote \ref{fn45}, $\wt\phi(P, 

7655: x^0)$ denotes the Fourier transform of $\wt\phi(P,E)$ in the variable $E$.

7656: \refb{e11.6} can be simplified as

7657:  \be \label{e11.7}

7658: \wt\phi(P, x^0\to \infty) = - i\, {\sqrt\pi\over

7659: P} \,

7660: {\Gamma(-i P)\over \Gamma(iP)}

7661: \, \left[ e^{-i P (x^0 + \ln

7662: \sin(\pi\tl)) } +  e^{i P (x^0 + \ln

7663: \sin(\pi\tl)) } \right] \, .

7664:  \ee

7665: The position space representation $\phi(\vp, x^0)$ of this field,

7666: as defined through \refb{edefftphi}, is somewhat complicated. 

7667: However the expression simplifies if we use the field 

7668: $\psi(\vp, x^0)$ to represent this configuration.

7669: Eq. \refb{edefpsi} and \refb{e11.7} gives

7670:  \be \label{e11.7a}

7671: \wt\psi(P, x^0\to \infty) = - i\, {\sqrt\pi\over

7672: P} \,

7673: \left[ e^{-i P (x^0 + \ln

7674: \sin(\pi\tl)) } +  e^{i P (x^0 + \ln

7675: \sin(\pi\tl)) } \right] \, .

7676:  \ee

7677: Its Fourier transform, defined in \refb{edefftpsi}, is given by

7678:  \be \label{epsiposition}

7679: \psi(\vp, x^0) = -\sqrt{\pi} [H(x^0+ \ln

7680: \sin(\pi\tl) - \vp) -

7681: H(x^0+ \ln

7682: \sin(\pi\tl)+\vp)] \, ,

7683:  \ee

7684: where $H$ denotes the step

7685: function:

7686:  \be \label{es9.7}

7687: H(u) = \cases{1 \quad \hbox{for} \quad u>0\cr

7688: 0  \quad \hbox{for} \quad u<0} \, .

7689:  \ee

7690: 

7691: Eq.\refb{epsiposition} is valid only in the $x^0\to\infty$ limit.

7692: Since $\vp<0$, in this limit the first term goes to a constant and

7693: we get

7694:  \be \label{epsinew}

7695: \psi(\vp, x^0) = \sqrt{\pi} \, H(x^0 + \vp + \ln

7696: \sin(\pi\tl)) + \hbox{constant}\, .

7697:  \ee

7698: Thus we see that the $\psi$ field background produced by the rolling

7699: tachyon configuration takes the form of a single step function of height

7700: $\sqrt{\pi}$ in the $x^0\to\infty$ limit. Since $\psi(\vp, x^0)$ is a 

7701: scalar field with conventional kinetic term for large negative

7702: $\vp$, \refb{epsinew} carries infinite energy as in the case

7703: of critical string theory due to the infinite spatial gradient at 

7704: $\vp=-x^0-\ln \sin(\pi\tl)$\cite{0305159}. Thus 

7705: one would again be tempted to conclude that all the energy of the D0-brane 

7706: is 

7707: converted to closed string radiation, thereby invalidating the tree level 

7708: open string analysis. In this case however there is 

7709: a simple interpretation of this infinity as will be discussed in sections

7710: \ref{s11.2} and \ref{sopenclosed}.

7711: 

7712: This finishes our discussion of closed string field configuration

7713: produced by the

7714: $|\BB_1\ra$ component of the boundary state.

7715: 

7716: \subsection{Closed string background produced by $|\BB_2\ra$}

7717: \label{s12.4}

7718: 

7719: We shall now discuss the

7720: closed string background produced by $|\BB_2\ra$\cite{0402157}.

7721: This can be analyzed

7722: in the same way as in the case of critical string theory. We begin

7723: with the expression \refb{ebb2l} of

7724: $|\BB_2\ra$.

7725: Since $ \wc\OO^{(n)}_N(P)

7726: \,

7727: (c_0+\bar c_0) \, c_1 \, \bar c_1 \, e^{n X^0(0)} \,

7728: |P\ra$

7729: in this expression is

7730: an eigenstate of $2(L_0+\bar L_0)$ with eigenvalue $(4N+n^2 +P^2)$, we

7731: can

7732: choose the closed string field produced by $|\BB_2\ra$ to be:

7733:  \be \label{es12.4b}

7734: |\Psi_c^{(2)}\ra = \sum_{n\in Z} \, \sum_{N=2}^\infty \, \int_0^\infty \,{

7735: dP

7736: \over 2\pi} \, \la V_{Q-iP}  \ra_D \,

7737: \left(4N+n^2 + P^2\right)^{-1}

7738: \, \wc\OO^{(n)}_N(P)

7739: \, c_1 \, \bar c_1 \, e^{n X^0(0)} \, |P\ra\, .

7740:  \ee

7741: This corresponds to closed string field configurations which grow as

7742: $e^{nx^0}$ for large $x^0$.

7743: 

7744: A special class of operators among the

7745: $\wc O^{(n)}_N$'s are those which involve only excitations involving

7746: the

7747: $\alpha^0$,

7748: $\bar\alpha^0$ oscillators and correspond to higher level {\it

7749: primaries}

7750: of the

7751: $c=1$

7752: conformal field theory. As described before, these primaries are

7753: characterized by SU(2) quantum numbers

7754: $(j,m)$ with $j\ge 1$, $-j< m < j$, and has dimension $(j^2,

7755: j^2)$.  From \refb{esecondform}, \refb{epriform} and

7756: \refb{ebb2l} we see

7757: that the

7758: contribution to $|\BB_2\ra$ from these primary states has the form:

7759:  \be \label{eb2new}

7760: {1\over 2} \, \sum_{j\ge 1} \, \sum_{m=-j+1}^{j-1} \, \int_0^\infty \,{ dP

7761: \over 2\pi} \, \la V_{Q-iP}  \ra_D \,

7762: f_{j,m}(\tl) \, \wh

7763: \PP_{j,m} \, (c_0+\bar c_0) \, c_1

7764: \, \bar c_1 \, e^{-2m X^0(0)} \, |P\ra\, .

7765:  \ee

7766: The level of the operators $\wh\PP_{j,m}$ is

7767:  \be \label{espace1}

7768: N = (j^2 - m^2) \, .

7769:  \ee

7770: Thus the $|\Psi^{(2)}_c\ra$ produced by this part of $|\BB_2\ra$ takes

7771: the

7772: form:

7773:  \be \label{espace2}

7774: |\bar\Psi_c^{(2)}\ra = \sum_{j,m} \, f_{j,m}(\tl) \, \int_0^\infty \,{ dP

7775: \over 2\pi} \, \la V_{Q-iP}  \ra_D

7776: \, (4 j^2 + P^2)^{-1}

7777: \, \wh\PP_{j,m}

7778: \, c_1 \, \bar c_1 \, e^{-2m X^0(0)} \, |P\ra\, .

7779:  \ee

7780: 

7781: 

7782: As in the case of critical string theory, it is instructive to

7783: study the behaviour of $|\Psi^{(2)}_c\ra$ in the position space

7784: characterized by the Liouville coordinate $\vp$ instead of the

7785: momentum space expression given in \refb{es12.4b}. We concentrate

7786: on the large negative $\vp$ region as usual. Let us first focus on the 

7787: $|\bar\Psi^{(2)}_c\ra$ part

7788: of $|\Psi^{(2)}_c\ra$ as given in \refb{espace2}. If $|\Psi\ra$

7789: contains a term

7790:  \be \label{epjm1}

7791:  \int_0^\infty {dP\over 2\pi}   \, \wh\PP_{j,m} \, c_1\bar c_1\, \wt

7792:  \psi_{j,m}(P,X^0(0)) \,

7793:  |P\ra \, ,

7794:  \ee

7795: then the corresponding position space representation

7796: $\psi_{j,m}(\vp, x^0)$ of this field is defined to be

7797:  \be \label{epjm2}

7798:  \psi_{j,m}(\vp, x^0) = \int_0^\infty {dP\over 2\pi} 

7799: \, \wt \psi_{j,m}(P,x^0)  \, \left[ e^{2\vp + i

7800:  P \vp} - \left( \Gamma(iP)\over \Gamma(-iP)\right)^2 \, e^{2\vp -

7801:  iP\vp}\right]\, .

7802:  \ee

7803: Comparing \refb{epjm1} with \refb{espace2}, and using \refb{epjm2}

7804: and the expression for $\la V_{Q-iP}\ra$ given in \refb{e11.5},

7805: we get

7806:  \ben \label{espace4}

7807: \psi_{j,m}(\vp,x^0) &=& f_{j,m}(\tl) \, e^{-2 m x^0} \, 

7808: \int_0^\infty {d

7809: P\over

7810: 2\pi}

7811: \, (4 j^2 + P^2)^{-1} \, \la V_{Q-iP}  \ra_D \nonumber \\

7812: && \qquad \left[ e^{2 \vp + i

7813: P \,

7814: \vp} -\left({\Gamma(iP)\over \Gamma(-iP)}\right)^2 \, e^{2\vp - iP\vp}

7815: \right]\, \nonumber \\

7816: &=& -{2\over \sqrt \pi} \, i\, f_{j,m}(\tl) \, e^{-2

7817: m

7818: x^0} \, \int_{-\infty}^\infty {d

7819: P\over 2\pi} \, e^{2 \vp + i

7820: P \,

7821: \vp} \, (4 j^2 + P^2)^{-1} \, \sinh(\pi P) \,

7822: {\Gamma(-i P)\over \Gamma(iP)}\, . \nonumber \\

7823:  \een

7824: This integral is not well defined since $\sinh(\pi P)$ blows

7825: up for large $|P|$. As in the analysis of \refb{esource}, for negative

7826: $\vp$ we shall define this integral by

7827: closing the contour in the lower half plane, and picking up the

7828: contribution from all the poles. Since the poles of

7829: $\Gamma(-iP)$ at $P=-in$ are cancelled by the zeroes of $\sinh(\pi

7830: P)$,

7831: the only pole that the integral has in the lower half plane is at

7832: $P=-2ij$. Evaluating the residue at this pole, we get

7833:  \be \label{espace6}

7834: \psi_{j,m}(\vp,x^0) = {1

7835: \over \left((2j)!\right)^2} \, 

7836: \sqrt \pi \, f_{j,m}(\tl) \,  e^{-2m x^0 +

7837: 2 (1+j) \vp} \, .

7838:  \ee

7839: In the language of string field theory, this corresponds to

7840:  \be \label{espace7}

7841: |\bar\Psi_c^{(2)}\ra = \sum_{j,m} \, {\sqrt \pi\over

7842: \left((2j)!\right)^2}

7843: \,  f_{j,m}(\tl) \, \wh \PP_{j,m} \,

7844: e^{-2 m X^0(0)}|0\ra_{X^0} \otimes e^{2 (1+j)\vp(0)}|0\ra_L \otimes

7845: c_1 \, \bar c_1 |0\ra_{ghost}\, .

7846:  \ee

7847: The states appearing in \refb{espace7} are precisely the discrete

7848: states

7849: of two dimensional string theory\cite{LIAN,9108004} (after the

7850: replacement $X^0 \to -iX$).

7851: 

7852: Contribution to $|\Psi_c^{(2)}\ra$ from the terms in

7853: $|\BB_2\ra$ involving excitations by Liouville Virasoro generators have 

7854: been analyzed in \cite{0408064}. Since we shall not need

7855: these results for later analysis we refer the reader to the original paper

7856: for details. 

7857: 

7858: \subsection{Matrix model description of the two dimensional string theory}

7859: \label{s11.2}

7860: 

7861: The two dimensional string theory described above also has an

7862: alternative

7863: description to all orders in perturbation theory as a matrix

7864: model\cite{GROMIL,BKZ,GINZIN}. This matrix description, in turn, can

7865: be

7866: shown to be equivalent to a theory of infinite number of

7867: non-interacting

7868: fermions,

7869: each moving in an

7870: inverted harmonic oscillator potential with hamiltonian

7871:  \be \label{e12.1}

7872: h(q,p) = {1\over 2} (p^2 - q^2) + {1\over g_s}\, ,

7873:  \ee

7874: where $(q,p)$ denotes a canonically conjugate pair of variables.

7875: The coordinate variable $q$ is related to the eigenvalue of an

7876: infinite dimensional matrix, but this information will not be

7877: necessary for our discussion. Clearly $h(q,p)$ has a continuous

7878: energy spectrum spanning the range $(-\infty, \infty)$. The vacuum

7879: of the theory corresponds to all states with negative $h$

7880: eigenvalue being filled and all states with positive $h$

7881: eigenvalue being empty. Thus the fermi surface is the surface of

7882: zero energy. Since we shall not go beyond perturbation theory, we

7883: shall ignore the effect of tunneling from one side of the barrier

7884: to the other side and work on only one side of the barrier. For

7885: definiteness we shall choose this to be the negative $q$ side. In

7886: the semi-classical limit, in which we represent a quantum state by

7887: an area element of size $\hbar$ in the phase space spanned by $p$

7888: and $q$, we can restrict ourselves to the negative $q$ region, and

7889: represent the vacuum by having the region $(p^2 - q^2) \le

7890: -{2\over g_s}$ filled, and rest of the region

7891: empty\cite{POLCH,9212027} (see Fig.\ref{f12.1}). In this picture

7892: the fermi surface in the phase space corresponds to the curve:

7893:  \be \label{e12.2}

7894: {1\over 2} (p^2 - q^2) + {1\over g_s} = 0\, .

7895:  \ee

7896:  \begin{figure}[!ht]

7897:  \begin{center}

7898: \leavevmode

7899: \epsfysize=5cm

7900: \epsfbox{fig12.1.eps}

7901: \end{center}

7902: \caption{Semi-classical representation of the vacuum state in the

7903: matrix

7904: model.

7905: } \label{f12.1}

7906: \end{figure}

7907: 

7908: 

7909: 

7910: If $\Psi(q,t)$ denotes the second quantized fermion field describing

7911: the

7912: above non-relativistic system, then

7913: the massless `tachyon' field in the closed string sector is identified

7914: with the scalar field obtained by the bosonization of the fermion

7915: field

7916: $\Psi$\cite{DASJEV,SENWAD,GROSSKLEB}. The precise correspondence goes

7917: as

7918: follows.

7919: The classical equation of motion satisfied by the field $\Psi(q,x^0)$

7920: has

7921: the form:

7922:  \be \label{e12.1b}

7923: i {\p\Psi\over \p x^0} + {1\over 2} \, {\p^2 \Psi\over \p q^2} +

7924: {1\over

7925: 2} \, q^2 \Psi - {1\over g_s} \, \Psi=

7926: 0\, .

7927:  \ee

7928: We now define the

7929: `time of flight' variable $\tau$ that is related to $q$ via the

7930: relation:

7931:  \be \label{e12.1a}

7932: q = -\sqrt{2\over g_s} \, \cosh\tau\, , \qquad \tau < 0\, .

7933:  \ee

7934: $|\tau|$ measures the time taken by a zero energy classical particle

7935: moving

7936: under the Hamiltonian \refb{e12.1} to travel from $-\sqrt{2\over g_s}$

7937: to $q$. We also define

7938:  \be \label{e12.1c}

7939: v(q) = -\sqrt{q^2-{2\over g_s}} = \sqrt{2\over g_s} \, \sinh \tau\, .

7940:  \ee

7941: $|v(q)|$ gives the magnitude of the

7942: classical velocity of a zero energy particle when

7943: it is at position

7944: $q$.

7945: Using these variables, it is easy to see that for large negative

7946: $\tau$

7947: the

7948: solution to eq.\refb{e12.1b} takes the form:

7949:  \be \label{e12.1d}

7950: \Psi(q,x^0) = {1\over \sqrt{-2 v(q)}} \, \left[ e^{-i\int^q v(q') dq'

7951: +

7952: i\pi/4} \, \Psi_R(\tau, x^0) + e^{i\int^q v(q') dq' -

7953: i\pi/4} \, \Psi_L(\tau, x^0)\right]\, ,

7954:  \ee

7955: where $\Psi_L$ and $\Psi_R$ satisfy the field equations:

7956:  \be \label{e12.1e}

7957: (\p_0 - \p_\tau) \, \Psi_L(\tau, x^0)=0, \qquad (\p_0 + \p_\tau) \,

7958: \Psi_R(\tau, x^0)=0 \, .

7959:  \ee

7960: Thus at large negative $\tau$ we can regard the system as a theory of

7961: a

7962: pair of

7963: chiral fermions, one left-moving and the other right-moving. Of course

7964: there is an effective boundary condition at $\tau=0$ which relates the

7965: two

7966: fermion fields, since a particle coming in from $\tau=-\infty$ will be

7967: reflected from $\tau=0$ and will go back to $\tau=-\infty$. Since

7968: $\tau$

7969: ranges from 0 to $-\infty$, we can interprete $\Psi_R$ as the incoming

7970: wave

7971: and $\Psi_L$ as the outgoing wave.

7972: 

7973: Eq.\refb{e12.1e} shows that $\Psi_L$ and $\Psi_R$ represent a pair

7974: of relativistic fermions for large negative $\tau$. 

7975: Thus we can bosonize them into a pair of

7976: chiral bosons $\chi_L$ and $\chi_R$. This pair of chiral bosons

7977: may in turn be combined into a full scalar field $\chi(\tau, x^0)$

7978: which satisfy the free field equation of motion for large negative

7979: $\tau$ and satisfies an appropriate boundary condition at

7980: $\tau=0$. If $\chi$ is defined with the standard normalization,

7981: then for large negative $\tau$ a single right moving fermion is

7982: represented by the configuration\cite{COLEMAN,MANDELSTAM}

7983:  \be \label{ebo1}

7984: \chi=\sqrt{\pi}\, H(x^0-\tau)\, ,

7985:  \ee

7986: and

7987: a

7988: single

7989: left-moving fermion is represented by the configuration

7990:  \be \label{ebo2}

7991: \chi=\sqrt{\pi}\, H(x^0+\tau)\, ,

7992:  \ee

7993: where $H(u)$ denotes the step function defined in \refb{es9.7}.

7994: 

7995: The field $\chi(\tau,x^0)$ is related to the tachyon field $\phi(\vp,

7996: x^0)$ in the continuum description of string theory by a non-local

7997: field

7998: redefinition\cite{MOORESEI}. In momentum space the relation

7999: is\cite{poly1,9108019,9109005,9402156}:

8000:  \be \label{e12.1f}

8001: \wt\chi(P, E) = {\Gamma(iP) \over \Gamma(-iP)} \,

8002: \wt\phi(P, E)\, .

8003:  \ee

8004: Using \refb{edefpsi} this gives

8005:  \be \label{echps}

8006: \wt\chi(P, E) = \psi(P,E)\, ,

8007:  \ee

8008: and hence

8009:  \be \label{echpspo}

8010: \chi(\tau, x^0) = \psi(\tau, x^0)\, .

8011:  \ee

8012: Using \refb{epsinew} and \refb{echpspo} we see that

8013: the background $\chi$ associated with a rolling tachyon

8014: solution is given by

8015:  \be \label{e12.1i}

8016: \chi(\tau, x^0) = \sqrt{\pi} \, H(x^0 + \tau + \ln

8017: \sin(\pi\tl))

8018:  \ee

8019: for large positive $x^0$ and large negative $\vp$.

8020: According to \refb{ebo2} this precisely represents a single

8021: left-moving

8022: (outgoing)

8023: fermion. This shows that

8024: the

8025: non-BPS D0-brane of the two dimensional string theory can be

8026: identified as

8027: a state of the matrix theory where a single fermion is excited from

8028: the

8029: fermi level to some energy

8030: $>0$\cite{0304224,0305194,0305159}.

8031: 

8032: The classical

8033: configuration \refb{e12.1i} 

8034: has infinite energy in the

8035: scalar field theory. In the fermionic description this infinite

8036: energy is the result of infinite quantum uncertainty in momentum

8037: for a sharply localized particle in the position space. Thus the

8038: classical limit of the fermionic theory does not have this

8039: infinite energy. This is the origin of the apparent discrepancy

8040: between the classical open string calculation of the D0-brane

8041: energy which gives a finite answer \refb{eennc} and the classical closed 

8042: string

8043: calculation which gives infinite answer\cite{0305159}. We hope that

8044: a similar interpretation can be given for the infinite energy carried

8045: by the closed string configuration produced by a `decaying' D-brane

8046: in the critical string theory as discussed in section \ref{eccl}.

8047: 

8048:  \begin{figure}[!ht]

8049: \leavevmode

8050:  \begin{center}

8051: \epsfysize=5cm

8052: \epsfbox{fig12.2.eps}

8053: \end{center}

8054: \caption{Semiclassical representation of a closed string tachyon field

8055: configuration in the matrix model.} \label{f12.2}

8056: \end{figure}

8057: 

8058: 

8059: Note that strictly speaking the above analysis, leading to the

8060: identification of the D0-brane with the single fermion excitation,

8061: holds only close to the fermi level, {\it i.e.} near $\tl=1/2$.

8062: {}From the continuum viewpoint this requirement comes from the

8063: fact that the effect of $|\BB_2\ra$ which has not been taken into

8064: account so far, can be ignored only in the $\tl\to {1\over 2}$

8065: limit. {}From the matrix model side this requirement comes due to

8066: the fact that the bosonization of the fermion system in terms of a

8067: single scalar field holds only for excitations close to the fermi

8068: level\cite{POLCH}. Nevertheless it is natural to assume that the

8069: correspondence between a D0-brane and single fermion excitations

8070: continues to hold for general $\tl$.  In this description the

8071: D0-brane with the tachyon field sitting at the maximum of the

8072: potential corresponds to the configuration $p=0$, $q=0$. The mass

8073: of the D0-brane is then given by $h(0,0)=1/g_s$. On the other hand

8074: the rolling tachyon solution in open string theory, characterized

8075: by the parameter $\tl$, corresponds to the phase space trajectory

8076:  \be \label{etrajec}

8077: q = -\sqrt{2\over g_s} \, \sin(\pi\tl) \, \cosh x^0,

8078: \qquad p = -\sqrt{2\over g_s} \, \sin(\pi\tl) \, \sinh x^0

8079:  \ee

8080: as can be seen by comparing the energies of the rolling tachyon

8081: system (eq.\refb{eennc}) and the system described by the

8082: Hamiltonian \refb{e12.1}. The $\tl\to{1\over 2}$

8083: limit corresponds to a trajectory at the fermi level.

8084: 

8085: We can now use this

8086: correspondence to find an interpretation for the exponentially

8087: growing component $|\Psi^{(2)}_c\ra$ of the string field produced

8088: by $|\BB_2\ra$. Since $|\BB_2\ra$ contains information about the

8089: conserved charges, what we need is the identification of these

8090: charges in the matrix model. An infinite set of conserved charges

8091: of this type do indeed exist in the quantum theory of a single

8092: fermion described by \refb{e12.1}. These are of the

8093: form\cite{SENWAD,MOORESEI,UTTG-16-91,9108004,9110021,9201056,9209036,9210105,

8094: 9302106,9507041}:

8095:  \be \label{econs1old}

8096: e^{(k-l) x^0} (p+q)^l (q-p)^k\, ,

8097:  \ee

8098: where $k$ and $l$ are integers.

8099: Requiring that the canonical transformations generated by these

8100: charges

8101: preserve the fermi level $h(q,p)=0$ \cite{9108004} gives us a more

8102: restricted class of

8103: charges:

8104:  \be \label{econs1}

8105: h(q,p) \, e^{(k-l) x^0} (p+q)^l (q-p)^k = \left( {1\over 2}

8106: (p^2-q^2) + {1\over g_s}\right) \, e^{(k-l) x^0} (p+q)^l

8107: (q-p)^k\, .

8108:  \ee

8109: Thus it is natural to identify these with linear combinations of

8110: the charges $e^{2m

8111: x^0} g_{j,m}(x^0)$ in the continuum theory. In order to find the

8112: precise

8113: relation between these charges we can first compare the

8114: explicit $x^0$ dependence of the two sets of charges. This gives:

8115:  \be \label{econs2}

8116: k-l = 2m\, .

8117:  \ee

8118: Thus the conserved charge $e^{2m

8119: x^0} g_{j,m}(x^0)$ should correspond to some specific linear

8120: combination

8121: of the charges given in \refb{econs1} subject to the condition

8122: \refb{econs2}:

8123:  \be \label{econs3}

8124: g_{j,m}(x^0) \leftrightarrow g_s\, \left( {1\over 2} (p^2-q^2) +

8125: {1\over g_s}\right) \, \sum_{k\in Z \atop k\ge 0, 2m}\, \left({2\over

8126: g_s}\right)^{m-k}

8127: \,  a^{(j,m)}_k \, (q-p)^k

8128: (q+p)^{k-2m}

8129: \, .

8130:  \ee

8131: Here $a^{(j,m)}_k$ are constants and the various $g_s$

8132: dependent

8133: normalization factors have been introduced for later convenience.

8134: In order to find the precise form of the coefficients $a_k^{(j,m)}$ we

8135: compare the

8136: $\tl$ dependence of the two sides for the classical trajectory

8137: \refb{etrajec}. Since for this trajectory

8138:  \be \label{econs4}

8139: q\pm p = - \sqrt{2\over g_s} \, \sin(\pi\tl) \, e^{\pm x^0}\, ,

8140:  \ee

8141: and $g_{j,m}(x^0) = e^{-2m x^0} \, f_{j,m}(\tl)$, we have:

8142:  \be \label{econs5}

8143: f_{j,m}(\tl) = (-1)^{2m} \, \left(1-\sin^2(\pi\tl)\right) \,

8144: \sum_{k\in Z \atop k\ge 0,

8145: 2m}\,

8146: a^{(j,m)}_k \,  \sin^{2k -

8147: 2m}(\pi\tl) \, .

8148:  \ee

8149: Thus by expanding $f_{j,m}(\tl)$ given in \refb{efjm} in powers of

8150: $\sin(\pi\tl)$ we can determine the coefficients $a^{(j,m)}_k$.

8151: One consistency check for this procedure is that on the right hand

8152: side the expansion in powers of $\sin(\pi\tl)$ starts at order

8153: $\sin^{2|m|}(\pi\tl)$. It can be verified that the expansion of

8154: $f_{j,m}(\tl)$ also starts at the same order. The other

8155: consistency check is that the right hand side of \refb{econs5}

8156: vanishes at $\tl={1\over 2}$, which is also the case for

8157: $f_{j,m}(\tl)$. One can also show \cite{0402157,0408064} that these

8158: relations are invertible, {\it i.e.} the charges $h(q,p)\,

8159: (q-p)^{2m+l}\, (q+p)^l$ may be expressed as linear combinations

8160: of $g_{j,m}(x^0)$ for $|m|+1\le j\le m+l+1$. This shows that the

8161: conserved charges $g_{j,m}(x^0)$ in the continuum theory contains

8162: information about the complete set of symmetry generators in the

8163: matrix model description of the D0-brane.

8164: 

8165: Given that the boundary state $|\BB_2\ra$ carries information

8166: about the conserved charges, the closed string field produced by

8167: $|\BB_2\ra$ must also carry the same information. These can be

8168: regarded as the analog of the long range electric or gravitational

8169: field produced by a particle carrying charge or mass.

8170: Ref.\cite{0408064} gives a systematic procedure for relating the

8171: conserved charges to the asymptotic closed string field

8172: configuration at large negative $\vp$ associated with the discrete

8173: states.

8174: 

8175:  \begin{figure}[!ht]

8176: \leavevmode

8177:  \begin{center}

8178: \epsfysize=5cm

8179: \epsfbox{fig12.3.eps}

8180: \end{center}

8181: \caption{Semi-classical representation of a state of the D0-brane in

8182: the

8183: matrix model.} \label{f12.3}

8184: \end{figure}

8185: 

8186: 

8187: 

8188: In the context we note that by carefully examining the bosonization

8189: rules

8190: for a fermion moving under the influence of inverted harmonic

8191: oscillator

8192: potential, ref.\cite{0401067} has argued that in order to

8193: describe the

8194: motion of a single fermion in the language of closed string theory, we

8195: need to switch on infinite number of closed string fields besides the

8196: tachyon. The essential point is that whereas a closed string tachyon

8197: mode describes a deformation of the fermi surface of the form shown in

8198: Fig.\ref{f12.2}, the D0-brane is represented by a blob in the phase

8199: space disconnected from the Fermi surface as in Fig.\ref{f12.3}.

8200: We believe that the presence of the additional closed

8201: string

8202: background \refb{espace7} associated with the discrete

8203: states is a reflection of this effect. As a consistency check we note

8204: that

8205: at $\tl=1/2$ the additional background \refb{espace7} vanish. This is

8206: expected to be true in the matrix model description as

8207: well since in this limit the blob merges with the fermi sea\cite{0401067}.

8208: 

8209: To summarize, we see that a D0-brane in the continuum two

8210: dimensional string theory is described in the matrix model as a single

8211: fermion excited from the fermi level to some positive energy state.

8212: This provides a satisfactory picture of the D0-brane. But this

8213: correspondence also raises a puzzle. In the matrix model there is

8214: another natural class of states, namely the single hole states, which

8215: correspond to a single fermion exited from some state below

8216: the fermi level to the fermi level. One might expect that just like

8217: the single fermion states, the single hole states should also have

8218: some natural description

8219: in the continuum string theory. So far however we do not have a

8220: completely satisfactory description of these states in the matrix

8221: model. Some proposal for what these states might correspond to have

8222: been made in \cite{0307195,0307221}. According to this proposal the

8223: boundary state

8224: describing a hole is obtained by analytically continuing the boundary

8225: state of a rolling tachyon configuration to $\tl={1\over 2}+i\alpha$ for a

8226: real parameter $\alpha$ and then changing the overall sign of the boundary

8227: state. Although this produces the right values for the conserved charges

8228: carried by the hole, neither the space-time nor the world-sheet

8229: interpretation of these states is very clear\cite{0408064}. An alternative

8230: suggestion, according to which the hole states correspond to ordinary

8231: D0-branes moving under the influence of the linear dilaton

8232: background has been put forward in \cite{0408064}. We hope that this

8233: question will be resolved in the near future.

8234: 

8235: \subsection{D0-brane decay in type 0B string theory} \label{s0b}

8236: 

8237: Although two dimensional bosonic string theory provides a useful

8238: arena for studying the decay of D0-branes beyond leading order in

8239: perturbation theory, it suffers from one problem; the theory,

8240: although well defined to all orders in string perturbation

8241: theory, is not non-perturbatively stable. In the continuum

8242: description of the D0-brane this is due to the fact that the

8243: tachyon potential on the D0-brane, while having a local minimum at

8244: some positive value of the tachyon field, is expected to be

8245: unbounded from below on the negative side. The problem associated 

8246: with this can be seen for

8247: example in the analysis of the time dependent solutions where the

8248: tachyon rolling on the wrong side of the potential causes a

8249: divergence in the  dilaton charge at a finite time. In the matrix

8250: model description the instability is

8251: related to the fact that the fermi

8252: level is filled on only one side of the potential leaving the

8253: other side empty. Such a vacuum, while perturbatively stable, is

8254: non-perturbatively unstable. Hence while the matrix

8255: model - two dimensional string theory correspondence illustrates

8256: many aspects of D0-brane decay in perturbation theory, it does not

8257: allow us to go beyond perturbation theory.

8258: 

8259: On the matrix model side there is an easy way to solve the

8260: problem, -- we just fill the fermi level on both sides. This will

8261: have the feature that the spectrum of closed strings, which

8262: represent excitations around the fermi level, will be doubled

8263: since now we can excite fermion hole pairs on either side of the

8264: potential. In particular bosonizing the fermion field on the two

8265: sides of the potential we should either find two different

8266: asymptotic regions with a massless scalar field living in each of

8267: these regions, or find one asymptotic region with

8268: two massless scalar fields living in this region. This is clearly 

8269: not the

8270: two dimensional bosonic string theory

8271: that we have discussed so far which has only one asymptotic region

8272: ($\vp\to -\infty$) with one massless scalar

8273: field. Is there some other two dimensional string theory that

8274: corresponds to this particular matrix model? It turns out that the

8275: answer is yes: it is two dimensional type 0B string

8276: theory\cite{0307083,0307195,0402196}.

8277: 

8278: The local world sheet dynamics of type 0 string theories is

8279: identical to that of type II string theories, {\it i.e.} it has

8280: both left and right moving world-sheet supersymmetry. Thus the

8281: critical dimension for this theory is 10, which is equivalent to

8282: saying the matter part of the theory must have $\hat c=10$. The

8283: difference with type II  string theories comes in the GSO

8284: projection rules. Whereas the type II string theories have

8285: separate GSO projection on the left and the right sector, type 0

8286: theories have one combined GSO projection\cite{polbook}. Due to

8287: modular invariance this then requires that the theory contains

8288: NS-NS and RR sector states, but no NS-R or R-NS sector states. In

8289: other words, the closed string sector does not contain fermions!

8290: As in the case of type II string theories, we can construct two

8291: types of type 0 string theories, 0A and 0B, which differ from each

8292: other in the sign of the GSO projection operator in the RR sector.

8293: 

8294: As in the case of bosonic string theory where

8295: we can construct a

8296: two dimensional string theory by replacing 25 of the scalar fields

8297: by a Liouville theory with total central charge 25,  we can also

8298: construct two dimensional type 0A and 0B string theories by

8299: replacing nine of the space-like world-sheet superfields by a 

8300: super-Liouville theory. This theory contains a single superfield with 

8301: exponential potential and background charge such that it describes a 

8302: super-conformal field theory 

8303: with $\hat 

8304: c=9$. This gives rise to the so called two dimensional type 0 string

8305: theories. In particular type 0B theory has two independent

8306: massless scalar fields in the closed string spectrum, one coming

8307: from the RR sector and the other coming from the NS-NS sector.

8308: Thus it is natural to conjecture that this two dimensional string

8309: theory is equivalent to the matrix model described earlier, with

8310: the two scalars being identified with appropriate linear

8311: combinations of the excitations on the fermi level on two sides of

8312: the potential\cite{0307083,0307195}. This is verified by computing the 

8313: S-matrix

8314: involving these fields in the type 0B theory and comparing them

8315: with the predictions of the matrix model\cite{9109005,0309148}.

8316: The only

8317: subtle point to keep in mind is that the matrix model Hamiltonian

8318: \refb{e12.1} seems to correspond to type 0B theory with

8319: $\alpha'=1/2$ rather than $\alpha'=1$. We shall work in this unit

8320: in the rest of this section.

8321: 

8322: It turns out that the type 0B string theory has an unstable

8323: D0-brane that corresponds to Dirichlet

8324: boundary condition on the Liouville coordinate, and the usual

8325: Neumann boundary condition on the time coordinate. The open string

8326: spectrum on this brane has a tachyonic mode. In the $\alpha'=1$

8327: unit the tachyon has mass$^2=-{1\over 2}$ as in the case of

8328: unstable D-branes of superstring theory, but in the

8329: $\alpha'={1\over 2}$ unit that we are using the tachyon has

8330: mass$^2=-1$. This agrees with the tachyon mass$^2$ obtained by

8331: quantizing the inverted harmonic oscillator Hamiltonian \refb{e12.1}

8332: around $q=0$.

8333: We can also construct the rolling tachyon

8334: solution by switching on tachyon background proportional to $\cosh

8335: x^0$ or $\sinh x^0$. The world-sheet analysis is identical to that

8336: in the case of unstable D-branes in superstring theory except for

8337: the scaling of $\alpha'$. In particular we can construct the

8338: boundary state describing this D0-brane following the procedure

8339: discussed earlier and use this to study closed string radiation

8340: from the D0-brane. The result gives a kink configuration of the

8341: type given in eq.\refb{e12.1i}

8342: showing that the D0-brane is naturally identified with single

8343: fermion excitations in the theory\cite{0307083,0307195}.

8344: 

8345: Thus we see that we now have an example of a completely consistent

8346: two dimensional string theory and its matrix model description.

8347: This allows us to study aspects of D0-brane dynamics in this

8348: theory not only to all orders in perturbation theory but also

8349: non-perturbatively. We hope that this can be used to derive useful

8350: insight into the dynamics of unstable D-branes in critical string

8351: theories, particularly in the context of the open string

8352: completeness conjecture to be discussed in section

8353: \ref{sopenclosed}.

8354: 

8355: Various other aspects of D0-brane decay in two dimensional string theory

8356: have been discussed in \cite{0308047,

8357: 0310106,0407270,0408049}.

8358: 

8359: \sectiono{Open String Completeness Conjecture} \label{sopenclosed}

8360: 

8361: The straightforward analysis of closed string emission from unstable

8362: D0-branes (or D$p$-branes wrapped on $T^p$) tell us that all the energy of

8363: the D0-brane is radiated away into closed strings, both in the critical

8364: string theory and in the two dimensional string theory. Naively this would

8365: suggest that the backreaction

8366: due to the closed string emission process invalidates the classical open

8367: string results. However we shall argue in this section that results

8368: obtained from the tree level

8369: open string theory actually give dual description of the closed

8370: string emission process.

8371: 

8372: \subsection{Open string completeness in the critical string theory}

8373: \label{sph}

8374: 

8375: We begin our discussion by analyzing the results for D-brane decay

8376: in critical string theory. In order to illustrate the proposed

8377: duality between open string and closed string description, let us

8378: compare the properties of the emitted closed strings from an

8379: unstable D$p$-brane wrapped on $T^p$ with those infered from the

8380: tree level open string

8381: analysis\cite{0304192,0305011,0306132,0306137}. First of all, tree

8382: level open string analysis tells us that the final system has:

8383:  \be \label{est1}

8384: Q / T_{00} = 0\, ,

8385:  \ee

8386: where $Q$ and $T_{00}$ denote the dilaton charge density and energy

8387: density of the system respectively. On the other hand by examining

8388: the closed string world-sheet action in the background  {\it

8389: string metric} $G_{\mu\nu}$, the anti-symmetric tensor field

8390: $B_{\mu\nu}$ and the dilaton $\Phi_D$ at zero momentum,

8391:  \be \label{est2}

8392: S_{world-sheet} = {1\over 2\pi} \, \int d^2 z (G_{\mu\nu}(X) +

8393: B_{\mu\nu}(X)) \p_z X^\mu \p_{\bar z} X^\nu\, ,

8394:  \ee

8395: we see that the closed string world-sheet does not

8396: couple to the zero momentum dilaton.

8397: This shows the final state closed strings carry zero total dilaton

8398: charge. Hence the dilaton charge of the final state closed

8399: strings

8400: agrees with that computed in the open string description.

8401: 

8402: Next we note that the tree level open string analysis tells us that the

8403: final system

8404: has:

8405:  \be \label{est3}

8406: p / T_{00} = 0\, ,

8407:  \ee

8408: where $p$ denotes the pressure of the system.

8409: On the other hand, closed string analysis tells us that the final

8410: closed

8411: strings have mass $m$ of order $1/g_s$, momentum $k_\perp$

8412: transverse to the

8413: D-brane of order $1/\sqrt{g_s}$ and

8414: winding $w_\parallel$ tangential to the D-brane of order $1/\sqrt{g_s}$.

8415: For such a system the ratio of transverse pressure to the energy density

8416: is of order $(k_\perp/m)^2\sim g_s$ and the ratio of tangential pressure

8417: to

8418: the energy density is of order $-(w_\parallel/m)^2\sim -g_s$.

8419: Since both these ratios

8420: vanish in the $g_s\to 0$ limit, we again see that the pressure of the

8421: final state closed strings match the

8422: result

8423: computed in the open string description.

8424: 

8425: 

8426: Such agreements between open and closed string results also hold

8427: for

8428: more general cases, {\it e.g.} in the decay of unstable branes in the

8429: presence of electric field. Consider, for example, the decay of a

8430: D$p$-brane along $x^1,\ldots x^p$ plane, with

8431: an

8432: electric field $e$ along the $x^1$ axis. In this case the final state is

8433: characterized by its energy-momentum tensor

8434: $T_{\mu\nu}$, source $S_{\mu\nu}$ for anti-symmetric tensor field

8435: $B_{\mu\nu}$ and the

8436: dilaton charge density $Q$. One can show that in the $x^0\to\infty$

8437: limit\cite{0208142,0301049,0303133}:

8438:  \be \label{est4}

8439: T^{00}=|\Pi|\, e^{-1} \, \delta(\vec x_\perp) \, , \quad T^{11} =

8440: - |\Pi| \, e\, \delta(\vec x_\perp) \,   , \quad S^{01} = \Pi \,

8441: \delta(\vec x_\perp) \, ,

8442:  \ee

8443: where $\Pi$ is a parameter labelling the solution. All other components of

8444: $T_{\mu\nu}$ and $S_{\mu\nu}$, as well as the dilaton charge vanishes in

8445: this limit. It can be shown that these tree level open string results

8446: again agree exactly with

8447: the

8448: properties of the final state closed strings into which the D-brane

8449: decays\cite{0306137,0409050}.

8450: 

8451: Since in all these cases the tree level open string results for

8452: various properties of the final state agree with the properties of

8453: the closed strings produced in the decay of the brane, we are led

8454: to conjecture that the tree level open string theory provides a

8455: description of the rolling tachyon system which is {\it dual} to

8456: the description in terms of closed  string

8457: emission\cite{0305011,0306137}.\footnote{This correspondence has

8458: been checked only for the space averaged values of various

8459: quantities, and not for example, for the local distribution of the

8460: various charges like the stress tensor, dilaton charge and

8461: anti-symmetric tensor field charge.  This is due to the fact that

8462: we can easily give a gauge invariant definition of the

8463: space-averaged quantities since they are measured by coupling to

8464: on-shell (zero momentum) closed string states, but it is more

8465: difficult to give a gauge invariant definition of local

8466: distribution of these charges\cite{0308068}.}$^,$\footnote{At present

8467: it is not clear how exactly the open string theory encodes 

8468: information about closed strings. A hint of how this might happen can

8469: be seen in the analysis of the effective field 

8470: theory\cite{9901159,9909062,0002223,0005031,0009061,0010181,0010240,

8471: 0305011,0402027}. This has been reviewed briefly in section \ref{s8.3}.}

8472: This is different

8473: from the usual open closed duality where one loop open string

8474: theory contains information about closed strings. In order to put

8475: this conjecture on a firmer footing, one must show that it arises

8476: from a more complete conjecture involving full quantum open string

8477: theory on an unstable D-brane. The full

8478: conjecture, suggested in

8479: \cite{0308068,0312153}, takes the following

8480: form:

8481: 

8482: {\it There is a quantum open string field theory

8483: (OSFT) that describes the full dynamics of an unstable Dp-brane without an

8484: explicit coupling to closed strings. Furthermore, Ehrenfest theorem holds

8485: in the weakly coupled OSFT; the classical results correctly

8486: describe the time evolution of the quantum expectation values.}

8487: 

8488: Since the conjecture in essence says that quantum open string theory is 

8489: fully capable of describing the complete dynamics of an unstable D-brane, 

8490: we call this the {\it open string completeness conjecture}. 

8491: Stated this way, this conjecture also embodies the usual perturbative 

8492: open-closed string duality for a stable D-brane where the open string loop

8493: amplitudes contain information about closed string exchange

8494: processes. In

8495: this case the quantum open string field theory

8496: is fully capable of reproducing the open string scattering

8497: amplitudes at least to all orders in perturbation

8498: theory\cite{GIDDMARW,THORN}.

8499: For example for this system quantum open 

8500: string 

8501: theory gives rise to the cylinder contribution to the string partition 

8502: function at one loop order. Coupling closed strings to this system 

8503: as in eqs.\refb{efg2}, 

8504: \refb{efg3} will ensure that by eliminating the closed string fields by 

8505: their 

8506: classical equations of motion we get the cylinder amplitude again

8507: as in eq.\refb{ezopen}.

8508: This gives a 

8509: clear indication that including closed strings in open string field theory 

8510: amounts to double counting.\footnote{It is of course possible to contruct 

8511: open-closed string field theories where part of the contribution to a 

8512: given process comes from the closed string sector and part of it comes 

8513: from the open string sector\cite{9705241}, but we are not discussing these 

8514: theories here.}

8515: 

8516: 

8517: Note that this open string completeness conjecture {\it does

8518: not} imply that the quantum open string theory on a given system

8519: of unstable (or stable) D-branes gives a complete description of

8520: the {\it full string theory}.\footnote{In this sense it is

8521: different from the cases discussed in \cite{9811131} where open

8522: string theory on a set of D-branes is completely equivalent to

8523: closed string theory on a certain background.} It only states that

8524: this open string field theory describes a quantum mechanically

8525: consistent subsector of the full string theory, and is fully

8526: capable of describing the quantum dynamics of the 

8527: D-brane.\footnote{For a similar phenomenon involving stable D-branes in 

8528: two 

8529: dimensional string theory see ref.\cite{0312196}.} 

8530: One of the

8531: consequences of this correspondence is that the notion of

8532: naturalness of a solution may differ dramatically in the open and

8533: the closed string description. A solution describing the decay of

8534: a single (or a few) unstable D-branes may look highly contrived in

8535: the closed string description, since a generic deformation of this

8536: background in closed string theory may not be describable by the

8537: dynamics of a single D-brane, and may require a large (or even

8538: infinite) number of D-branes for its description in open string

8539: theory.

8540: 

8541: 

8542: In

8543: our discussion of the properties of the closed string states

8544: emitted from a `decaying' D-brane

8545: we have so far

8546: only included closed string 

8547: radiation produced by the $|\BB_1\ra$ component of the

8548: boundary state.

8549: Hence our analysis is valid only in the

8550: $\tl\to{1\over 2}$ limit when the closed string background produced by

8551: $|\BB_2\ra$ vanishes.

8552: As we have seen in section \ref{ecc2}, for $\tl\ne {1\over 2}$

8553: the closed string fields

8554: produced by the $|\BB_2\ra$ component of the boundary state grow

8555: exponentially with time. Naively,  this exponential growth of the

8556: closed string field configurations again indicates the breakdown of

8557: classical open string description at late time. However in the

8558: spirit of the open string completeness conjecture proposed here

8559: it is more natural to  seek an alternative

8560: interpretation. What this may be indicating is an inadequacy of

8561: the weakly coupled 

8562: closed string description rather than an inadequacy of

8563: the open string

8564: description. As an analogy we can cite the example of closed

8565: string field configurations produced by static stable D-branes.

8566: Often the field configuration is singular near the core of the

8567: brane. However we do not take this as an indication of the

8568: breakdown of the open string description. Instead it is a

8569: reflection of the inadequacy of the closed string description.

8570: 

8571: 

8572: While in critical string theory the open string completeness

8573: remains a conjecture, we shall see

8574: in the next subsection that this is bourne out quite clearly

8575: in the two dimensional string theory.

8576: 

8577: \subsection{Open string completeness in two dimensional string

8578: theory} \label{sdual2d}

8579: 

8580: The analysis of section \ref{s2closed} shows that a rolling tachyon 

8581: configuration on a D0-brane in two dimensional string theory produces an 

8582: infinitely sharp kink of the closed string tachyon field. Since this 

8583: carries infinite energy, we might naively conclude that all the energy of 

8584: the D0-brane is converted to closed string radiation and hence the results 

8585: of tree level open string analysis cannot be trusted. The analysis of 

8586: section \ref{s12.4} shows that for $\tl\ne {1\over 2}$ the closed string 

8587: field configuration produced by the $|\BB_2\ra$ component of the boundary 

8588: state grows exponentially with time. This is again a potential source for 

8589: large backreaction on the tree level open string results. 

8590: 

8591: On the other 

8592: hand

8593: we have seen in section \ref{s11.2} that a D0-brane in two

8594: dimensional string theory can be identified in the matrix model

8595: description as a single fermion excitation from the fermi level to

8596: some energy level above 0. Since the fermions are non-interacting,

8597: the states with a single excited fermion do not mix with any other

8598: states in the theory (say with states where two or more fermions

8599: are excited above the fermi level or hole states where a fermion is

8600: excited from below the fermi level to the fermi level). As a

8601: result, the quantum states of a D0-brane are in one to one

8602: correspondence with the quantum states of the single particle

8603: Hamiltonian

8604:  \be \label{ehagain}

8605:  h(q,p) = {1\over 2} (p^2 - q^2) + {1\over g_s}\, ,

8606:  \ee

8607: with one additional constraint, -- the spectrum is cut off sharply

8608: for energy below zero due to Pauli exclusion principle. 

8609: Since by definition quantum open string field theory on a D-brane is a 

8610: field theory that describes the dynamics of that D-brane, we see that 

8611: in

8612: the matrix model description the `quantum open string field

8613: theory' for a single D0-brane is described by the inverted

8614: harmonic oscillator hamiltonian \refb{ehagain} with all the

8615: negative energy states removed by hand\cite{0308068}. The

8616: classical limit of this quantum Hamiltonian is described by the

8617: classical Hamiltonian \refb{ehagain}, with a sharp cut-off on the

8618: phase space variables:\footnote{This Hamiltonian is related to the

8619: D0-brane effective Hamiltonian given in \refb{ex2}, \refb{evform}

8620: by a canonical transformation\cite{0308068} that maps the curve

8621: $h(q,p)=0$ to  $\infty$ in the $\Pi-T$ plane.}

8622:  \be \label{e12.4}

8623:   {1\over 2} (p^2 - q^2) + {1\over g_s} \ge 0\, .

8624:  \ee

8625: This is the matrix model description of `classical open string

8626: field theory' describing the dynamics of a D0-brane.\footnote{It

8627: will be very interesting to understand the precise connection

8628: between this system and the cubic open string field theory that

8629: describes the dynamics of the D0-brane in the continuum

8630: description. It will be even more interesting to study similar

8631: relation between this system and the open string field theory

8632: describing the dynamics of the D0-brane in type 0B string theory,

8633: since there the system is non-perturbatively stable. Since on the

8634: matrix model side we have a free system, the approach of

8635: \cite{0308184,0402063,0409233} 

8636: might provide a useful starting point for

8637: establishing this correspondence.} 

8638: 

8639: Clearly the quantum system described above provides us with a

8640: complete description of the dynamics of a single D0-brane.  In

8641: particular there is no need to couple this system explicitly to

8642: closed strings, although closed strings could provide an

8643: alternative description of the D0-brane as a kink solution in the

8644: closed string `tachyon' field (as

8645: given in \refb{e12.1i}). This is in

8646: accordance with the open string completeness conjecture

8647: discussed in section \ref{sph}. {}From this it is clear that it is

8648: a wrong notion to think in terms of {\it backreaction of closed

8649: string fields on the open string dynamics}. Instead we should

8650: regard the closed string background produced by the D-brane as a

8651: way of characterizing the open string background (although the

8652: open string theory itself is sufficient for this purpose). For

8653: example, in the present context, we can think of the closed string

8654: tachyon field $\chi(\tau, x^0)$ at late time as the expectation

8655: value of the operator\footnote{$\p_\tau\hat\chi$ is the

8656: representation of the usual density operator of free fermions in

8657: the Hilbert space of first quantized theory of a single fermion.}

8658:  \be \label{eexpec}

8659: \hat \chi(\tau, x^0) \equiv \sqrt{\pi} \, H \left(-\hat q(x^0) -

8660: \sqrt{2\over g_s}

8661: \cosh\tau\right)  \,

8662:  \ee

8663: in the quantum open string theory on a single D0-brane, as described

8664: by

8665: \refb{ehagain}, \refb{e12.4}.

8666: In \refb{eexpec} $\hat q$ denotes the position operator in the quantum

8667: open string theory. When we calculate the expectation value of $\hat

8668: \chi$

8669: in the

8670: quantum state whose classical limit is described by the trajectory

8671: \refb{etrajec}, we can replace $\wh q$ by its classical value

8672: $q=-\sqrt{2\over g_s} \sin(\pi\tl) \cosh(x^0)$. This gives

8673:  \be \label{eexpec2}

8674: \la \hat \chi(\tau, x^0) \ra = \sqrt{\pi} \, H \left(\sqrt{2\over

8675: g_s} \, \sin(\pi\tl) \, \cosh(x^0) - \sqrt{2\over g_s}

8676: \cosh\tau\right) \simeq \sqrt{\pi} \, H(x^0 + \tau + \ln

8677: \sin(\pi\tl))\, ,

8678:  \ee

8679: for large $x^0$ and negative $\tau$. This reproduces

8680: \refb{e12.1i}.

8681: 

8682: In a similar spirit we note that while the naive analysis of the

8683: closed string field configuration produced by the $|\BB_2\ra$

8684: component of the boundary state indicates that the exponentially

8685: growing closed string fields produce large backreaction and hence

8686: invalidates the analysis based on 

8687: open string theory, the results of section \ref{s11.2} clearly

8688: point to a different direction. According to these results the

8689: exponentially growing terms in $|\BB_2\ra$ are simply consequences

8690: of the conserved charges a D0-brane carries, which in turn may be

8691: calculated completely within the framework of the open string (field)

8692: theory. Thus these exponentially growing terms do not in any way

8693: point to an inadequacy of the open string description of D0-brane

8694: dynamics.

8695: 

8696: Various other aspects of the open string completeness

8697: conjecture in the

8698: context of two dimensional string theory have been discussed in

8699: \cite{0312135,0312163,0312192}.

8700: 

8701: \subsection{Generalized holographic principle} \label{sholo}

8702: 

8703: We have seen in the previous two subsections that the analysis of

8704: unstable D-brane `decay' in critical string theory as well as in

8705: the two dimensional string theory points to the conjecture that

8706: the quantum open string field theory on a given D-brane

8707: describes complete quantum

8708: dynamics of the D-brane. However generically

8709: a given D-brane system does not have the ability to describe an

8710: arbitrary state in string theory, -- it carries only partial 

8711: information about the full theory.\footnote{The possibility of

8712: using infinite number of unstable D-instantons or D0-branes or finite 

8713: number of

8714: space-filling D-branes to give a complete description of string

8715: theory has been discussed in 

8716: refs.\cite{9812135,9904207,0009189,0108085,0212188,0305006}.

8717: Vacuum string field theory even attempts to give a complete description of 

8718: the theory in terms of open string field theory on a single 

8719: D-brane\cite{0105058,0105059,0105168}.}

8720: This conjecture makes it 

8721: clear that open string field theory on a single or

8722: a finite number of D-branes must be encoding information about the

8723: full string theory in a highly non-local manner. For example when

8724: a D-brane `decays' into closed strings we expect that at least

8725: some of the closed strings will eventually 

8726: disperse to infinity. The open string

8727: field theory, defined in terms of variables localized near the

8728: original D-brane, must be able to describe the final closed string

8729: state produced in the decay, although it may not be able to

8730: describe states of the {\it individual closed strings} into which

8731: the D-brane decays. The situation can be described by drawing

8732: analogy to a hologram. If we regard the full string theory as the

8733: complete image produced by a hologram, then a system of

8734: finite number of D-branes can be regarded as a part of the

8735: hologram. This encodes partial information about the full image.

8736: However the information contained in any given piece of the

8737: hologram does not correspond to a given part of the complete

8738: image; instead it has partial information about all parts of the

8739: image. Thus the open string completeness conjecture proposed here

8740: can be thought of as a generalization of the holographic

8741: principle\cite{9802109,9409089,9711200,9802109,9802150,9805114}. As in 

8742: these papers in our proposal

8743: the

8744: relation between the open and closed string description is

8745: non-local, but the space in which the open string degress of

8746: freedom live is not necessarily the boundary of the space in which

8747: the closed string degrees of freedom live.

8748: 

8749: 

8750: 

8751: \medskip

8752: 

8753: {\bf Acknowledgement:} I would like to thank my collaborators

8754: N.~Berkovits,

8755: D.~Gaiotto, D.~Ghoshal, J.~Majumder, N.~Moeller, P.~Mukhopadhyay,

8756: L.~Rastelli and B.~Zwiebach for collaboration on various aspects

8757: of open string tachyon dynamics.

8758: 

8759: \appendix

8760: 

8761: \sectiono{Energy-Momentum Tensor from Boundary State}

8762: \label{appb}

8763: 

8764: As discussed in section \ref{s3.4},

8765: given a boundary CFT describing a D-brane system, we define the

8766: corresponding boundary state $|\BB\ra$ such that given any closed

8767: string state

8768: $|V\ra$ and the associated vertex operator $V$,

8769:  \be \label{efg1app}

8770: \la\BB|V\ra \propto \la V(0)\ra_D\, ,

8771:  \ee

8772: where  $\la V(0)\ra_D$ is the one point function of $V$ inserted at

8773: the

8774: centre of a unit disk $D$, the boundary condition / interaction

8775: on $\p D$ being the one

8776: associated with the particular boundary CFT under consideration.

8777: {}From this definition it is clear that the boundary state

8778: contains

8779: information about what kind of source for the closed string states is

8780: produced by the D-brane system under consideration.

8781: In this appendix we shall make this more

8782: precise by working with (linearized) closed string field

8783: theory\cite{9705241}.\footnote{We shall not attempt to give a detailed

8784: description of closed string field theory. Section \ref{s4} contains a

8785: self-contained discussion of open string field theory. Closed string

8786: field theory is formulated on similar principles.}

8787: 

8788: For simplicity we

8789: shall focus on the bosonic string theory.

8790: The closed string field corresponds to a state $|\Psi_c\ra$ of ghost

8791: number 2 in the Hilbert space of matter ghost conformal field theory in

8792: the full complex plane, satisfying the constraint\cite{9705241}

8793:  \be \label{efg1.5}

8794: b_0^-|\Psi_c\ra = 0\, , \qquad L_0^-|\Psi_c\ra = 0\, ,

8795:  \ee

8796: where

8797:  \be \label{efg2.5}

8798: c_0^\pm = (c_0 \pm \bar c_0), \qquad b_0^\pm = (b_0 \pm \bar b_0)\, ,

8799: \qquad

8800: L_0^\pm=(L_0 \pm \bar L_0)\, .

8801:  \ee

8802: $c_n$, $\bar c_n$, $b_n$, $\bar b_n$ are the usual ghost

8803: oscillators and $L_n$, $\bar L_n$ are the total Virasoro generators.

8804: The quadratic part of the closed string field theory action can be

8805: taken to be:

8806:  \be \label{efg2}

8807: -{1\over K g_s^2}\la \Psi_c|c_0^- (Q_B+\bar Q_B) |\Psi_c\ra\, ,

8808:  \ee

8809: where $Q_B$ and $\bar Q_B$ are the holomorphic and

8810: anti-holomorphic components of the BRST charge, $K$ is a

8811: normalization constant to be determined in eq.\refb{ezopen}, and

8812: $g_s$ is the appropriately normalized closed string coupling

8813: constant so that \refb{e2.1} holds. In the presence of the D-brane

8814: we need to add an extra source term to the action:

8815:  \be \label{efg3}

8816: \la \Psi_c| c_0^- |\BB\ra\, .

8817:  \ee

8818: The equation of motion of

8819: $|\Psi_c\ra$ is then

8820:  \be \label{efg4app}

8821: 2\, (Q_B+\bar Q_B)\, |\Psi_c\ra = K\, g_s^2 \, |\BB\ra\, .

8822:  \ee

8823: Clearly by a rescaling of $|\Psi_c\ra$ by $\lambda$ we can change

8824: $K$ and $|\BB\ra$ to $K/\lambda^2$ and $\lambda |\BB\ra$

8825: respectively. However once the normalization of $|\BB\ra$ is fixed

8826: in a convenient manner, the normalization constant $K$ can be

8827: determined by requiring that the classical action obtained after

8828: eliminating $|\Psi_c\ra$ using its equation of motion \refb{efg4app}

8829: reproduces the one loop partition function $Z_{open}$ of the open

8830: string theory on the D-brane. Choosing the solution of \refb{efg4app}

8831: to be\footnote{Some subtleties involved in obtaining the solution

8832: in the Minkowski space have been discussed after eq.\refb{es9.1}.}

8833:  \be \label{es9.1prev}

8834: |\Psi_c\ra = {1\over 2} \, K\, g_s^2 \, b_0^+ \, (L_0^+)^{-1} \,

8835: |\BB\ra \, ,

8836:  \ee

8837: we get

8838:  \be \label{ezopen}

8839: Z_{open} = -{1\over 4} \,  K\, g_s^2 \, \la \BB| b_0^+ \, c_0^- \,

8840: (L_0^+)^{-1} \, |\BB\ra\, .

8841:  \ee

8842: We have chosen a convenient normalization of $|\BB\ra$ in

8843: eq.\refb{efg10}. This determines $K$ from

8844: eq.\refb{ezopen}. Since $Z_{open}$ is independent of $g_s$, and

8845: $|\BB\ra$ is inversely

8846: proportional to $g_s$ due to the $\TT_p$ factor in \refb{efg10},

8847: we

8848: see

8849: that $K$ is a purely numerical constant.

8850: 

8851: 

8852: Eqs.\refb{efg3}, \refb{efg4app} clearly shows that $|\BB\ra$ represents

8853: the

8854: source for the closed string fields in the presence of a D-brane. We shall

8855: now make it more explicit by expanding $|\Psi_c\ra$ and $|\BB\ra$ in the

8856: oscillator basis.

8857: The expansion of $|\Psi_c\ra$ in the oscillator basis of closed string

8858: states has, as coefficients, various closed string fields. For example the

8859: first few terms in the expansion are:

8860:  \ben \label{efg5}

8861: |\Psi_c\ra &=& \int {d^{26}k \over (2\pi)^{26}}\, \bigg[ \wt T(k) c_1

8862: \bar c_1 +

8863: \left(\wt h_{\mu\nu}(k) + \wt b_{\mu\nu}(k)\right) 

8864: \alpha^\mu_{-1}\bar\alpha^\nu_{-1}

8865: c_1

8866: \bar c_1 \nonumber \\ &&

8867: + \left(\wt \phi(k) + {1\over 2} \eta^{\mu\nu} \wt

8868: h_{\mu\nu}(k)\right) (c_1

8869: c_{-1} - \bar c_1 \bar c_{-1}) + \ldots\bigg] |k\ra

8870:  \een

8871: where $\alpha^\mu_n$, $\bar\alpha^\mu_n$ are the usual oscillators

8872: associated with the $X^\mu$ fields. In this expansion

8873: $\wt T(k)$ has the interpretation of the Fourier transform of the

8874: closed string tachyon

8875: field, $\wt h_{\mu\nu}=\wt h_{\nu\mu}$ and $\wt\phi(k)$ are

8876: Fourier transforms of the graviton (associated with the string metric) and

8877: the dilaton fields respectively, $\wt

8878: b_{\mu\nu}=-\wt b_{\nu\mu}$ is the Fourier transform of the anti-symmetric

8879: tensor field etc.\footnote{The identification of these fields can be found

8880: by substituting \refb{efg5} into the quadratic action \refb{efg2} and

8881: comparing the resulting action with the quadratic part of the known

8882: effective

8883: action involving the graviton, dilaton and anti-symmetric tensor field.}

8884: On the other hand the boundary state $|\BB\ra$ has an expansion of the

8885: form:

8886:  \ben \label{efg5.5app}

8887: |\BB\ra &=&

8888: \int {d^{26} k\over (2\pi)^{26}} \left[\wt F(k) +  

8889: \left(\wt A_{\mu\nu}(k)+ \wt

8890: C_{\mu\nu}(k)\right)

8891: \alpha^\mu_{-1}

8892: \bar\alpha^\nu_{-1}

8893: + \wt B(k) (b_{-1} \bar c_{-1}  + \bar b_{-1} c_{-1}) + 

8894: \ldots \right] \nonumber \\

8895: && \qquad \qquad  \qquad \qquad \qquad \qquad (c_0

8896: +\bar

8897: c_0)

8898: c_1

8899: \bar c_1 |k\ra\, ,

8900:  \een

8901: where $\wt F$, $\wt A_{\mu\nu}=\wt A_{\nu\mu}$, $\wt

8902: C_{\mu\nu}=-\wt C_{\nu\mu}$, $\wt B$ etc. are fixed functions

8903: which can be read out of the boundary state $|\BB\ra$.

8904: Substituting \refb{efg5} and \refb{efg5.5app} into

8905: eq.\refb{efg3} we get terms in the action proportional to

8906: \be \label{elinear}

8907: \int \, {d^{26}k\over (2\pi)^{26}} \, \left[ \wt h_{\mu\nu}(-k) 

8908: \left(\wt

8909: A^{\mu\nu}(k) + \wt B(k) \, \eta^{\mu\nu}\right) 

8910: + 2\wt\phi(-k) \, \wt B(k) +

8911: \cdots \right] \, ,

8912: \ee

8913: where $\cdots$ denote source terms for other fields. Since the

8914: Fourier transform $\wt T^{\mu\nu}$ of the energy momentum tensor couples

8915: to $\wt h_{\mu\nu}$, we see that $\wt T_{\mu\nu}$ is

8916: proportional to $\wt A_{\mu\nu}(k) + \eta_{\mu\nu}\,  \wt

8917: B(k)$.\footnote{Note however that this

8918: definition of the various source terms is not unique, since we

8919: could redefine the off-shell closed string field, and this will in

8920: general modify the source terms for various fields. A simple

8921: example of this kind is $\wt h_{\mu\nu}(k) \to f(k^2) \wt

8922: h_{\mu\nu}(k)$ with $f(0)=1$. This does not change the definition

8923: of the graviton field on-shell, but modifies it off-shell.

8924: Consequently the energy-momentum tensor $\wt T_{\mu\nu}$ to which

8925: the field couples will get modified as $\wt T_{\mu\nu}(k) \to

8926: (f(k^2))^{-1} \, \wt T_{\mu\nu}(k)$. In our analysis we shall

8927: adopt the particularly simple definition of energy momentum tensor

8928: as follows from the action \refb{efg2}, \refb{efg3}. In this convention 

8929: the quadratic terms in the closed string field theory action have at most 

8930: two 

8931: powers of the 

8932: momentum $k^\mu$. In position space this translates to these quadratic 

8933: terms having at most two derivatives. However one should keep in mind that 

8934: other choices are possible.}

8935: 

8936: We can also use an alternative but equivalent method for

8937: expressing the energy momentum tensor $T_{\mu\nu}$ in terms of

8938: $A_{\mu\nu}$ and $B$ by using the

8939: conservation law of

8940: $T_{\mu\nu}$. \refb{efg4app} together with

8941: nilpotence of $(Q_B+\bar Q_B)$ gives:

8942:  \be \label{efg6}

8943: (Q_B+\bar Q_B)|\BB\ra = 0\, .

8944:  \ee

8945: Substituting the expansion \refb{efg5.5app} 

8946: into this equation gives, besides

8947: other equations,

8948:  \be \label{efg7}

8949: k^\nu \wt

8950: A_{\mu\nu}(k)

8951: +

8952: k_\mu \wt B(k) = 0\, .

8953:  \ee

8954: If $A_{\mu\nu}(x)$ and $B(x)$ denote the Fourier transforms of $\wt

8955: A_{\mu\nu}$ and $\wt B$ respectively,

8956:  \be \label{efourab}

8957: A_{\mu\nu}(x) = \int {d^{26}k\over (2\pi)^{26}} \, e^{ik\cdot x} \wt

8958: A_{\mu\nu}(k)\, , \qquad

8959: B(x) = \int {d^{26}k\over (2\pi)^{26}} \, e^{ik\cdot x} \wt

8960: B(k)\, ,

8961:  \ee

8962: then \refb{efg7} gives us:

8963:  \be \label{efg8}

8964: \p^\nu (A_{\mu\nu}(x) + \eta_{\mu\nu} B(x) ) = 0\, .

8965:  \ee

8966: This shows that we should identify the conserved energy-momentum tensor as

8967:  \be \label{efg9aapp}

8968: T_{\mu\nu}(x) \propto (A_{\mu\nu}(x) + \eta_{\mu\nu} B(x) )\, .

8969:  \ee

8970: 

8971: {}From \refb{elinear} we also see that the function $\wt B(k)$

8972: measures the source of the dilaton field $\wt\phi(k)$. This

8973: suggests that in position space we define the source of the

8974: dilaton to be

8975:  \be \label{edchargeapp}

8976: Q(x) \propto  B(x) \, .

8977:  \ee

8978: 

8979: \sectiono{Computation of the Energy of Closed String Radiation from

8980: Unstable D-brane} \label{appa}

8981: 

8982: 

8983: \refb{es9.17bb} gives the on-shell closed string field configuration

8984: produced by the

8985: rolling tachyon background. In this appendix we shall compute the

8986: energy per unit

8987: $p$-volume carried by this background.

8988: For this we

8989: express $(Q_B+\bar Q_B)$, acting on a state carrying momentum

8990: $\{k^\mu\}$,

8991: as

8992:  \be \label{eqbbreak}

8993: Q_B+\bar Q_B = (c_0 L_0 + \bar c_0 \bar L_0) + \wh Q_1(\vec k) k^0 +

8994: \wh

8995: Q_2(\vec k) \, ,

8996:  \ee

8997: where the operators $\wh Q_1(\vec k)$ and $\wh

8998: Q_2(\vec k)$ do not have any explicit $k^0$ dependence

8999: but can depend on the spatial components of the momentum

9000: $\vec k$. The reason

9001: that the right hand side of \refb{eqbbreak} does not contain higher

9002: powers of $k^0$ is that the $k^\mu$ dependence of $(Q_B+\bar Q_B)$

9003: comes through the $k^\mu$ dependence of the matter Virasoro generators

9004: $L^{(m)}_n$ and $\bar L^{(m)}_n$, and for $n\ne 0$ $L^{(m)}_n$ and

9005: $\bar L^{(m)}_n$ are linear in $k^\mu$. Contribution from the $L_0$

9006: and $\bar L_0$ part of $Q_B+\bar Q_B$, which are quadratic in $k^\mu$,

9007: has been written separately in the right hand side of \refb{eqbbreak}.

9008: Let us denote

9009: by $\wh L$ the part of $(L_0+\bar L_0)$ that involves oscillator

9010: contribution and contribution from spatial momenta, so that

9011:  \be \label{el0bl0}

9012: L_0+\bar L_0 = -{1\over 2} (k^0)^2 + \wh L\, .

9013:  \ee

9014: We also

9015: express a general

9016: closed string field configuration $|\Psi_c\ra$ as

9017:  \be \label{eenergy}

9018: |\Psi_c\ra = \int {d k^0\over 2\pi} \, \wt\psi_A(k^0) \, |k^0, A\ra\, ,

9019:  \ee

9020: where the label $A$ runs over discrete oscillator labels as well as

9021: continuous spatial momenta. The free closed string field theory

9022: action \refb{efg2},

9023: expressed in terms of the component fields $\wt\psi_A$, takes the form

9024:  \be \label{eener1}

9025: S = - {1\over K g_s^2} \, \int \, {d k^0\over 2\pi} \, \wt \psi_A(-k^0) \,

9026: \left[ -{1\over 2}

9027: (k^0)^2 N_{AB} + M_{AB} + k^0 \,  M^{(1)}_{AB} + M^{(2)}_{AB}

9028: \right] \wt \psi_B(k^0)\, ,

9029:  \ee

9030: where $N_{AB}$, $M_{AB}$, $M^{(1)}_{AB}$ and $M^{(2)}_{AB}$ are defined

9031: through the relations:

9032:  \ben \label{eener2}

9033: {1\over 2} \la k^0, A| c_0^- c_0^+ | k^{\prime 0}, B\ra &=& N_{AB} \, 2\pi

9034: \, \delta(

9035: k^0 +  k^{\prime 0}) \, , \nonumber \\

9036: {1\over 2} \, \la k^0, A| c_0^- c_0^+ \wh L | k^{\prime 0}, B\ra &=&

9037: M_{AB}\, 2\pi

9038: \, \delta(

9039: k^0 +  k^{\prime 0}) \, , \nonumber \\

9040: \la k^0, A| c_0^- \wh Q_1 | k^{\prime 0}, B\ra &=& M^{(1)}_{AB}\, 2\pi

9041: \, \delta(

9042: k^0 +  k^{\prime 0}) \, , \nonumber \\

9043: \la k^0, A| c_0^- \wh Q_2 | k^{\prime 0}, B\ra &=& M^{(2)}_{AB}\, 2\pi

9044: \, \delta(

9045: k^0 +  k^{\prime 0}) \, . \nonumber \\

9046:  \een

9047: If we denote by

9048:  \be \label{eener3}

9049: \psi_A(x^0) = \int {d k^0\over 2\pi} \, e^{-i k^0 x^0} \, \wt

9050: \psi_A(k^0)\, ,

9051:  \ee

9052: then the action \refb{eener1} may be reexpressed as

9053:  \be \label{eener4}

9054: S = \int d x^0 L\, ,

9055:  \ee

9056: where

9057:  \be \label{eener5}

9058: L = {1\over K g_s^2} \, \left[{1\over 2} \p_0 \, \psi_A N_{AB} \,

9059: \p_0 \psi_B - i\, M^{(1)}_{AB} \, \psi_A \, \p_0 \psi_B -

9060: \left(M_{AB} + M^{(2)}_{AB}\right) \, \psi_A \, \psi_B\right]\, .

9061:  \ee

9062: {}From this we can write down an expression for the conserved energy:

9063:  \be \label{eener6}

9064: E = \p_0 \psi_A {\p L \over \p (\p_0\psi_A)} -

9065: L

9066: = {1\over K g_s^2} \, \left[{1\over 2} \p_0 \,

9067: \psi_A N_{AB} \, \p_0 \psi_B +  \left(M_{AB} + M^{(2)}_{AB}\right) \,

9068: \psi_A \,

9069: \psi_B\right]\, .

9070:  \ee

9071: 

9072: We shall now evaluate \refb{eener6} for the on-shell closed string

9073: background \refb{es9.17bb}. For this we express $|\Psi_c^{(1)}\ra$ given

9074: in

9075: \refb{es9.17bb} as

9076:  \be \label{eener7}

9077: |\Psi_c^{(1)}\ra = \int {d k^0\over 2\pi} \, \sum_N \, \int {d^{25-p}

9078: k_\perp\over

9079: (2\pi)^{25-p}} \, \wt\psi^{(1)}_N(k^0, \vec k_\perp)

9080: |N,\vec k_\perp, k^0\ra\, ,

9081:  \ee

9082: where

9083:  \be \label{eener8}

9084: |N,\vec k_\perp, k^0\ra =

9085: \wh A_N\,  c_1\, \bar c_1\, |k^0, \vec k_\parallel=0, \vec k_\perp\ra \,

9086:  \ee

9087: and

9088:  \ben \label{eener9}

9089: \wt\psi_N^{(1)}(k^0, \vec k_\perp) &=& 2\, K\, g_s^2 \, \TT_p \,

9090: {\pi\over

9091: \sinh(\pi\onk)} \, {1\over 2\onk} \nonumber \\

9092: && \left[e^{-i\onk \, \ln

9093: \sin(\pi\tl)} \delta(k^0 -

9094: \onk)

9095:  + e^{i\onk

9096: \, \ln

9097: \sin(\pi\tl)} \delta(k^0 + \onk) \right]\, .

9098:  \een

9099: This gives

9100:  \ben \label{eener10}

9101: \psi_N^{(1)} (x^0,\vec k_\perp) &\equiv& \int {d k^0\over 2\pi} \,

9102: e^{-i k^0 x^0} \,

9103: \wt\psi_N(k^0, \vec k_\perp)   \nonumber \\

9104: &=& 2\, K\, g_s^2 \, \TT_p \, {\pi\over \sinh(\pi\onk)} \, {1\over

9105: 2\onk} \, \left[e^{-i\onk \, (x^0+\ln \sin(\pi\tl))}

9106:  + e^{i\onk

9107: \, (x^0+ \ln

9108: \sin(\pi\tl))}

9109: \right]\, . \nonumber \\

9110:  \een

9111: Using the definitions \refb{eenergy} and \refb{eener3} of $\psi_A(x^0)$,

9112: definition \refb{eener2} of $N_{AB}$, $M_{AB}$, $M^{(1)}_{AB}$ and

9113: $M^{(2)}_{AB}$, and eq.\refb{eener6}, we can now write down an expression

9114: for the total energy associated with the configuration $|\Psi_c^{(1)}\ra$

9115: as:

9116:  \ben \label{eener11}

9117: E^{(1)} &=& {1\over K g_s^2}  \, \int {d^{25-p}

9118: k_\perp\over

9119: (2\pi)^{25-p}} \, \sum_N \Big[ S_N \, \p_0 \psi_N^{(1)}(x^0, \vec

9120: k_\perp) \, \p_0

9121: \psi_N^{(1)}(x^0, -\vec k_\perp)  \nonumber \\

9122: && \qquad \qquad + R_N \, \psi_N^{(1)}(x^0,

9123: \vec k_\perp) \,

9124: \psi_N^{(1)}(x^0,

9125: -\vec k_\perp)

9126: \Big] \, ,

9127:  \een

9128: where $S_N$ and $R_N$ are defined through

9129:  \be \label{eener12-}

9130: {1\over 4} \la N, \vec k_\perp, k^0| c_0^- c_0^+ | N, \vec k_\perp',

9131: k^{\prime 0}\ra

9132: = (2\pi)^{26-p} \, \delta(k^0 + k^{\prime 0}) \, \delta^{(25-p)}(\vec

9133: k_\perp +

9134: \vec k'_\perp) \,

9135: S_N\, ,

9136:  \ee

9137:  \be \label{eener12}

9138: \la N, \vec k_\perp, k^0| c_0^- ({1\over 2} \, c_0^+ \wh L + \wh Q_2) |

9139: N, \vec

9140: k_\perp',

9141: k^{\prime 0}\ra

9142: =(2\pi)^{26-p} \, \delta(k^0 + k^{\prime 0}) \, \delta^{(25-p)}(\vec

9143: k_\perp +

9144: \vec k'_\perp) \, R_N\, .

9145:  \ee

9146: {}From the definition of $|N, \vec k_\perp, k^0\ra$ and $\wh Q_2$ it

9147: follows

9148: that neither of them contains a $c_0$ or a $\bar c_0$ zero mode. As a

9149: result the matrix element of $c_0^- \wh Q_2$ appearing in \refb{eener12}

9150: vanishes. On the other hand $|N, \vec k_\perp\ra$ is an eigenstate of $\wh

9151: L$ with eigenvalue $2(N-1) + {1\over 2} \vec k_\perp^2 = {1\over 2}

9152: (\onk)^2$. Thus comparison of \refb{eener12-} and \refb{eener12}

9153: gives

9154:  \be \label{eener13}

9155: R_N = (\onk)^2 S_N\, .

9156:  \ee

9157: Substituting this into \refb{eener11}, and using \refb{eener10} we get

9158:  \be \label{eener14}

9159: E^{(1)} = 4\, K \, (g_s \TT_p)^2 \, \sum_N \, \int {d^{25-p}

9160: k_\perp\over (2\pi)^{25-p}} \, S_N \, {\pi^2\over

9161: \sinh^2(\pi\onk)} \, .

9162:  \ee

9163: 

9164: In order

9165: to compute $S_N$ we express the left hand side of

9166: \refb{eener12-} as

9167:  \ben \label{eener15}

9168: && {1\over 4} \, \la  k^0, \vec k_\parallel=0, \vec k_\perp| c_{-1} \,

9169: \bar

9170: c_{-1} \, (\wh A_N)^c \, c_0^- \, c_0^+\, \wh A_N \, c_1 \, \bar c_1 |

9171: k^{\prime 0}, \vec k_\parallel=0, \vec k_\perp'\ra \nonumber \\

9172: &=& {1\over

9173: 2} \,

9174: V_\parallel\, (2\pi)^{26-p} \, \delta(k^0 + k^{\prime 0}) \,

9175: \delta^{(25-p)}(\vec

9176: k_\perp +

9177: \vec k'_\perp) \, \la 0 |  c_{-1} \, \bar

9178: c_{-1} \, (\wh A_N)^c \, c_0 \, \bar c_0\, \wh A_N \, c_1 \, \bar c_1 |

9179: 0\ra''\, , \nonumber \\

9180:  \een

9181: where $(\wh A_N)^c$ is the BPZ conjugate of $\wh A_N$, $V_\parallel$

9182: denotes the spatial volume tangential to the D-brane,

9183: coming from the factor of $(2\pi)^p\delta^{(p)}(\vec k_\parallel)$

9184: evaluated at

9185: $\vec k_\parallel=0$, and

9186: $\la\cdot|\cdot\ra''$ denotes a renormalized

9187: BPZ inner product in the zero momentum sector such that\footnote{This

9188: is the analog of the renormalized inner product $\la \cdot|\cdot\ra'$ for 

9189: open string states

9190: as defined in eq.\refb{e4.43}.}

9191:  \be \label{ereninn}

9192: \la 0 |  c_{-1} \, \bar

9193: c_{-1} \,  c_0 \, \bar c_0\, \, c_1 \, \bar c_1 |

9194: 0\ra'' = 1\, .

9195:  \ee

9196: This gives

9197:  \be \label{esntosn}

9198: S_N = V_\parallel s_N\, ,

9199:  \ee

9200: where

9201:  \be \label{esn1}

9202: s_N = {1\over

9203: 2} \,

9204: \la 0 |  c_{-1} \, \bar

9205: c_{-1} \, (\wh A_N)^c \, c_0 \, \bar c_0\, \wh A_N \, c_1 \, \bar c_1 |

9206: 0\ra''\, .

9207:  \ee

9208: Substituting this into \refb{eener14} we get

9209:  \be \label{eener14aa}

9210: E^{(1)} = V_\parallel \EE

9211:  \ee

9212: where the energy per unit $p$-volume $\EE$ is given by:

9213:  \be \label{eedensity}

9214: \EE = \sum_N \EE_N = 4\, K \, (g_s \TT_p)^2 \, \sum_N \, s_N \,

9215: \int

9216: {d^{25-p} k_\perp\over (2\pi)^{25-p}} \, {\pi^2\over

9217: \sinh^2(\pi\onk)} \, .

9218:  \ee

9219: 

9220: In order to compute $s_N$ using \refb{esn1} we can use the

9221: generating

9222: functional

9223:  \be \label{esn1a}

9224: \sum_N s_N \, q^{2(N-1)} =  {1\over

9225: 2} \,

9226: \sum_N \sum_M \la 0 |  c_{-1} \, \bar

9227: c_{-1} \, (\wh A_N)^c \,  q^{L_0+\bar L_0} \, \wh A_M \, c_1 \, \bar c_1 |

9228: 0\ra''\, .

9229:  \ee

9230: Note that only the $M=N$ terms contribute to \refb{esn1a}

9231: due to $(L_0+\bar L_0)$ conservation. We can now

9232: replace $\sum_N \wh A_N$ and $\sum_M\wh A_M$ by the left hand side of

9233: \refb{eexpan}. The result is essentially the cylinder amplitude, with

9234: $\ln q$ denoting the ratio of the height to the circumference of

9235: the cylinder. It is however well known that in this computation

9236: the contribution from the ghost sector cancels the contribution

9237: from two of the matter sector fields, leaving behind the

9238: contribution from 24 matter sector fields. Thus in computing the

9239: right hand side of \refb{esn1a} we can replace $\sum_N \wh A_N$ by

9240:  \be \label{esn2}

9241: \exp\left[\sum_{n=1}^\infty

9242: \sum_{s=1}^{24} (-1)^{d_s} \, {1\over

9243: n} \,  \alpha^s_{-n} \bar

9244: \alpha^s_{-n}\right] \, .

9245:  \ee

9246: We can for example take the sum over $s$ to run over all the spatial

9247: directions

9248: tangential to the brane and all the transverse directions except one. The

9249: final formula is insensitive to $d_s$ (since $d_s$ can be changed by a

9250: redefinition $\alpha_s\to -\alpha_s$ without changing $\bar \alpha_s$),

9251: and so it does not matter whether we drop a Neumann or Dirichlet

9252: direction in the sum over $s$. Using the replacement \refb{esn2} in

9253: \refb{esn1a}

9254: the ghost term factorises giving a contribution of

9255: $q^{-2}$, and we get

9256:  \ben \label{esn3}

9257: \sum_N s_N \, q^{2N} &=&  {1\over 2} \, \la 0 |

9258: \exp\left[\sum_{m=1}^\infty \sum_{s=1}^{24} (-1)^{d_s} \, {1\over

9259: m} \,  \alpha^s_{m} \bar \alpha^s_{m}\right]  q^{L_0^{matter}+\bar

9260: L_0^{matter}} \nonumber \\  && \exp\left[\sum_{n=1}^\infty

9261: \sum_{r=1}^{24} (-1)^{d_r} \, {1\over n} \,  \alpha^r_{-n} \bar

9262: \alpha^r_{-n}\right] \, | 0\ra''_{matter}\, ,

9263:  \een

9264: where $\la \cdot|\cdot\ra''_{matter}$ denotes the BPZ inner product in the

9265: matter sector with the normalization convention $\la 0|0\ra''_{matter}=1$.

9266: 

9267: \begin{thebibliography}{99}

9268: 

9269: \bibitem{callan1}

9270: A.~Abouelsaood, C.~G.~.~Callan, C.~R.~Nappi and S.~A.~Yost,

9271: %``Open Strings In Background Gauge Fields,''

9272: Nucl.\ Phys.\ B {\bf 280}, 599 (1987).

9273: %%CITATION = NUPHA,B280,599;%%

9274: 

9275: \bibitem{0104263}

9276: C.~Acatrinei and C.~Sochichiu,

9277: %``A note on the decay of noncommutative solitons,''

9278: Phys.\ Rev.\ D {\bf 67}, 125017 (2003)

9279: [arXiv:hep-th/0104263].

9280: %%CITATION = HEP-TH 0104263;%%

9281: 

9282: \bibitem{0108075}

9283: A.~Adams, J.~Polchinski and E.~Silverstein,

9284: %``Don't panic! Closed string tachyons in ALE space-times,''

9285: JHEP {\bf 0110}, 029 (2001) [arXiv:hep-th/0108075].

9286: %%CITATION = HEP-TH 0108075;%%

9287: 

9288: \bibitem{9610249}

9289: M.~Aganagic, C.~Popescu and J.~H.~Schwarz,

9290: %``D-brane actions with local kappa symmetry,''

9291: Phys.\ Lett.\ B {\bf 393}, 311 (1997)

9292: [arXiv:hep-th/9610249].

9293: %%CITATION = HEP-TH 9610249;%%

9294: 

9295: \bibitem{9612080}

9296: M.~Aganagic, C.~Popescu and J.~H.~Schwarz,

9297: %``Gauge-invariant and gauge-fixed D-brane actions,''

9298: Nucl.\ Phys.\ B {\bf 495}, 99 (1997)

9299: [arXiv:hep-th/9612080].

9300: %%CITATION = HEP-TH 9612080;%%

9301: 

9302: \bibitem{0009142}

9303: M.~Aganagic, R.~Gopakumar, S.~Minwalla and A.~Strominger,

9304: %``Unstable solitons in noncommutative gauge theory,''

9305: JHEP {\bf 0104}, 001 (2001)

9306: [arXiv:hep-th/0009142].

9307: %%CITATION = HEP-TH 0009142;%%

9308: 

9309: \bibitem{0404253}

9310: G.~L.~Alberghi, R.~Casadio and A.~Tronconi,

9311: %``Particle production in tachyon condensation,''

9312: JHEP {\bf 0406}, 040 (2004)

9313: [arXiv:hep-th/0404253].

9314: %%CITATION = HEP-TH 0404253;%%

9315: 

9316: \bibitem{0012222}

9317: M.~Alishahiha, H.~Ita and Y.~Oz,

9318: %``On superconnections and the tachyon effective action,''

9319: Phys.\ Lett.\ B {\bf 503}, 181 (2001) [arXiv:hep-th/0012222].

9320: %%CITATION = HEP-TH 0012222;%%

9321: 

9322: \bibitem{0104164}

9323: M.~Alishahiha,

9324: %``One-loop correction of the tachyon action in boundary 

9325: %superstring field

9326: %theory,''

9327: Phys.\ Lett.\ B {\bf 510}, 285 (2001)

9328: [arXiv:hep-th/0104164].

9329: %%CITATION = HEP-TH 0104164;%%

9330: 

9331: \bibitem{0203185}

9332: J.~Ambjorn and R.~A.~Janik,

9333: %``Interpolating between open and closed strings - a BSFT 

9334: %approach,''

9335: Phys.\ Lett.\ B {\bf 538}, 189 (2002)

9336: [arXiv:hep-th/0203185].

9337: %%CITATION = HEP-TH 0203185;%%

9338: 

9339: \bibitem{0312163}

9340: J.~Ambjorn and R.~A.~Janik,

9341: %``The decay of quantum D-branes,''

9342: Phys.\ Lett.\ B {\bf 584}, 155 (2004) [arXiv:hep-th/0312163].

9343: %%CITATION = HEP-TH 0312163;%%

9344: 

9345: \bibitem{0407270}

9346: J.~Ambjorn and R.~A.~Janik,

9347: %``Quantum rolling tachyon,''

9348: arXiv:hep-th/0407270.

9349: %%CITATION = HEP-TH 0407270;%%

9350: 

9351: \bibitem{0010218}

9352: O.~Andreev,

9353: %``Some computations of partition functions and tachyon 

9354: %potentials in

9355: %background independent off-shell string theory,''

9356: Nucl.\ Phys.\ B {\bf 598}, 151 (2001)

9357: [arXiv:hep-th/0010218].

9358: %%CITATION = HEP-TH 0010218;%%

9359: 

9360: \bibitem{0109187}

9361: O.~Andreev and T.~Ott,

9362: %``On one-loop approximation to tachyon potentials,''

9363: Nucl.\ Phys.\ B {\bf 627}, 330 (2002)

9364: [arXiv:hep-th/0109187].

9365: %%CITATION = HEP-TH 0109187;%%

9366: 

9367: \bibitem{0112088}

9368: O.~Andreev,

9369: %``A note on nonperturbative instability in string theory,''

9370: Phys.\ Lett.\ B {\bf 534}, 163 (2002)

9371: [arXiv:hep-th/0112088].

9372: %%CITATION = HEP-TH 0112088;%%

9373: 

9374: \bibitem{0308123}

9375: O.~Andreev,

9376: %``Comments on tachyon potentials in closed and open-closed 

9377: %string theories,''

9378: Nucl.\ Phys.\ B {\bf 680}, 3 (2004)

9379: [arXiv:hep-th/0308123].

9380: %%CITATION = HEP-TH 0308123;%%

9381: 

9382: \bibitem{AREF1}

9383: I.~Y.~Arefeva, P.~B.~Medvedev and A.~P.~Zubarev,

9384: %{\em Background Formalism For Superstring Field Theory,}

9385: Phys.\ Lett.\  {\bf B240}, 356 (1990).

9386: %%CITATION = PHLTA,B240,356;%%

9387: 

9388: \bibitem{AREF2}

9389: I.~Y.~Arefeva, P.~B.~Medvedev and A.~P.~Zubarev,

9390: %{\em New Representation For String Field Solves The Consistence Problem 

9391: %For

9392: %Open Superstring Field,}

9393: Nucl.\ Phys.\  {\bf B341}, 464 (1990).

9394: %%CITATION = NUPHA,B341,464;%%

9395: 

9396: 

9397: \bibitem{0011117}

9398: I.~Y.~Aref'eva, A.~S.~Koshelev, D.~M.~Belov and P.~B.~Medvedev,

9399: %``Tachyon condensation in cubic superstring field theory,''

9400: Nucl.\ Phys.\ B {\bf 638}, 3 (2002) [arXiv:hep-th/0011117].

9401: %%CITATION = HEP-TH 0011117;%%

9402: 

9403: \bibitem{0107197}

9404: I.~Y.~Arefeva, D.~M.~Belov, A.~S.~Koshelev and P.~B.~Medvedev,

9405:  %``Gauge invariance and tachyon condensation in cubic superstring field

9406: %theory,''

9407: Nucl.\ Phys.\ B {\bf 638}, 21 (2002) [arXiv:hep-th/0107197].

9408: %%CITATION = HEP-TH 0107197;%%

9409: 

9410: \bibitem{0112214}

9411: I.~Y.~Arefeva, A.~A.~Giryavets and P.~B.~Medvedev,

9412: %``NS matter sliver,''

9413: Phys.\ Lett.\ B {\bf 532}, 291 (2002)

9414: [arXiv:hep-th/0112214].

9415: %%CITATION = HEP-TH 0112214;%%

9416: 

9417: \bibitem{0201197}

9418: I.~Y.~Arefeva, D.~M.~Belov and A.~A.~Giryavets,

9419: %``Construction of the vacuum string field theory on a non-BPS brane,''

9420: JHEP {\bf 0209}, 050 (2002)

9421: [arXiv:hep-th/0201197].

9422: %%CITATION = HEP-TH 0201197;%%

9423: 

9424: \bibitem{0203227}

9425: I.~Y.~Arefeva, A.~A.~Giryavets and A.~S.~Koshelev,

9426: %``NS ghost slivers,''

9427: Phys.\ Lett.\ B {\bf 536}, 138 (2002)

9428: [arXiv:hep-th/0203227].

9429: %%CITATION = HEP-TH 0203227;%%

9430: 

9431: \bibitem{0301137}

9432: I.~Y.~Aref'eva, L.~V.~Joukovskaya and A.~S.~Koshelev,

9433: %``Time evolution in superstring field theory on non-BPS brane. I:

9434: %Rolling  tachyon and energy-momentum conservation,''

9435: arXiv:hep-th/0301137.

9436: %%CITATION = HEP-TH 0301137;%%

9437: 

9438: \bibitem{0012080}

9439: G.~Arutyunov, S.~Frolov, S.~Theisen and A.~A.~Tseytlin,

9440: %``Tachyon condensation and universality of DBI action,''

9441: JHEP {\bf 0102}, 002 (2001) [arXiv:hep-th/0012080].

9442: %%CITATION = HEP-TH 0012080;%%

9443: 

9444: \bibitem{0105238}

9445: G.~Arutyunov, A.~Pankiewicz and B.~.~J.~Stefanski,

9446: %``Boundary superstring field theory annulus partition 

9447: %function in the  presence

9448: %of tachyons,''

9449: JHEP {\bf 0106}, 049 (2001)

9450: [arXiv:hep-th/0105238].

9451: %%CITATION = HEP-TH 0105238;%%

9452: 

9453: \bibitem{0108085}

9454: T.~Asakawa, S.~Sugimoto and S.~Terashima,

9455: %``D-branes, matrix theory and K-homology,''

9456: JHEP {\bf 0203}, 034 (2002)

9457: [arXiv:hep-th/0108085].

9458: %%CITATION = HEP-TH 0108085;%%

9459: 

9460: \bibitem{0212188}

9461: T.~Asakawa, S.~Sugimoto and S.~Terashima,

9462: %``Exact description of D-branes via tachyon condensation,''

9463: JHEP {\bf 0302}, 011 (2003)

9464: [arXiv:hep-th/0212188].

9465: %%CITATION = HEP-TH 0212188;%%

9466: 

9467: \bibitem{0305006}

9468: T.~Asakawa, S.~Sugimoto and S.~Terashima,

9469: %``Exact description of D-branes in K-matrix theory,''

9470: Prog.\ Theor.\ Phys.\ Suppl.\  {\bf 152}, 93 (2004)

9471: [arXiv:hep-th/0305006].

9472: %%CITATION = HEP-TH 0305006;%%

9473: 

9474: \bibitem{0309074}

9475: T.~Asakawa, S.~Kobayashi and S.~Matsuura,

9476: %``Closed string field theory with dynamical D-brane,''

9477: JHEP {\bf 0310}, 023 (2003) [arXiv:hep-th/0309074].

9478: %%CITATION = HEP-TH 0309074;%%

9479: 

9480: \bibitem{9209036}

9481: J.~Avan and A.~Jevicki,

9482: %``Interacting theory of collective and topological fields in

9483: %two-dimensions,''

9484: Nucl.\ Phys.\ B {\bf 397}, 672 (1993)

9485: [arXiv:hep-th/9209036].

9486: %%CITATION = HEP-TH 9209036;%%

9487: 

9488: \bibitem{0404039}

9489: V.~Balasubramanian, E.~Keski-Vakkuri, P.~Kraus and A.~Naqvi,

9490: %``String scattering from decaying branes,''

9491: arXiv:hep-th/0404039.

9492: %%CITATION = HEP-TH 0404039;%%

9493: 

9494: \bibitem{0407229}

9495: R.~Banerjee, Y.~Kim and O.~K.~Kwon,

9496: %``Noncommutative tachyon kinks as D(p-1)-branes from 

9497: %unstable D p-brane,''

9498: arXiv:hep-th/0407229.

9499: %%CITATION = HEP-TH 0407229;%%

9500: 

9501: \bibitem{9511194}

9502: T.~Banks and L.~Susskind,

9503: %``Brane - Antibrane Forces,''

9504: arXiv:hep-th/9511194.

9505: %%CITATION = HEP-TH 9511194;%%

9506: 

9507: \bibitem{bar1}

9508: K.~Bardakci,

9509: %``Dual Models And Spontaneous Symmetry Breaking,''

9510: Nucl.\ Phys.\ B {\bf 68}, 331 (1974).

9511: %%CITATION = NUPHA,B68,331;%%

9512: 

9513: \bibitem{bar2}

9514: K.~Bardakci and M.~B.~Halpern,

9515: %``Dual M - Models,''

9516: Nucl.\ Phys.\ B {\bf 73}, 295 (1974).

9517: %%CITATION = NUPHA,B73,295;%%

9518: 

9519: \bibitem{bar3}

9520: K.~Bardakci,

9521: %``Spontaneous Symmetry Breakdown In The Standard Dual String Model,''

9522: Nucl.\ Phys.\ B {\bf 133}, 297 (1978).

9523: %%CITATION = NUPHA,B133,297;%%

9524: 

9525: \bibitem{0105098}

9526: K.~Bardakci and A.~Konechny,

9527: %``Tachyon condensation in boundary string field theory 

9528: %at one loop,''

9529: arXiv:hep-th/0105098.

9530: %%CITATION = HEP-TH 0105098;%%

9531: 

9532: \bibitem{0406120}

9533: N.~Barnaby,

9534: %``Caustic formation in tachyon effective field theories,''

9535: JHEP {\bf 0407}, 025 (2004)

9536: [arXiv:hep-th/0406120].

9537: %%CITATION = HEP-TH 0406120;%%

9538: 

9539: \bibitem{0410030}

9540: N.~Barnaby and J.~M.~Cline,

9541: %``Tachyon Defect Formation and Reheating in 

9542: %Brane-Antibrane Inflation,''

9543: arXiv:hep-th/0410030.

9544: %%CITATION = HEP-TH 0410030;%%

9545: 

9546: \bibitem{0010101}

9547: I.~Bars, H.~Kajiura, Y.~Matsuo and T.~Takayanagi,

9548: %``Tachyon condensation on noncommutative torus,''

9549: Phys.\ Rev.\ D {\bf 63}, 086001 (2001)

9550: [arXiv:hep-th/0010101].

9551: %%CITATION = HEP-TH 0010101;%%

9552: 

9553: \bibitem{0302151}

9554: I.~Bars, I.~Kishimoto and Y.~Matsuo,

9555: %``Analytic study of nonperturbative solutions in open string field  

9556: %theory,''

9557: Phys.\ Rev.\ D {\bf 67}, 126007 (2003)

9558: [arXiv:hep-th/0302151].

9559: %%CITATION = HEP-TH 0302151;%%

9560: 

9561: \bibitem{0401195}

9562: D.~Bazeia, R.~Menezes and J.~G.~Ramos,

9563: %``Regular and periodic tachyon kinks,''

9564: arXiv:hep-th/0401195.

9565: %%CITATION = HEP-TH 0401195;%%

9566: 

9567: \bibitem{9806155}

9568: O.~Bergman and M.~R.~Gaberdiel,

9569: %``Stable non-BPS D-particles,''

9570: Phys.\ Lett.\ B {\bf 441}, 133 (1998) [arXiv:hep-th/9806155].

9571: %%CITATION = HEP-TH 9806155;%%

9572: 

9573: \bibitem{9901014}

9574: O.~Bergman and M.~R.~Gaberdiel,

9575: %``Non-BPS states in heterotic-type IIA duality,''

9576: JHEP {\bf 9903}, 013 (1999)

9577: [arXiv:hep-th/9901014].

9578: %%CITATION = HEP-TH 9901014;%%

9579: 

9580: \bibitem{9902160}

9581: O.~Bergman, E.~G.~Gimon and P.~Horava,

9582: %``Brane transfer operations and T-duality of non-BPS states,''

9583: JHEP {\bf 9904}, 010 (1999)

9584: [arXiv:hep-th/9902160].

9585: %%CITATION = HEP-TH 9902160;%%

9586: 

9587: \bibitem{9908126}

9588: O.~Bergman and M.~R.~Gaberdiel,

9589: %``Non-BPS Dirichlet branes,''

9590: Class.\ Quant.\ Grav.\  {\bf 17}, 961 (2000)

9591: [arXiv:hep-th/9908126].

9592: %%CITATION = HEP-TH 9908126;%%

9593: 

9594: \bibitem{0009252}

9595: O.~Bergman,

9596: %``Tachyon condensation in unstable type I D-brane systems,''

9597: JHEP {\bf 0011}, 015 (2000)

9598: [arXiv:hep-th/0009252].

9599: %%CITATION = HEP-TH 0009252;%%

9600: 

9601: \bibitem{0002223}

9602: O.~Bergman, K.~Hori and P.~Yi,

9603: %``Confinement on the brane,''

9604: Nucl.\ Phys.\ B {\bf 580}, 289 (2000) [arXiv:hep-th/0002223].

9605: %%CITATION = HEP-TH 0002223;%%

9606: 

9607: \bibitem{0402124}

9608: O.~Bergman, S.S~Razamat,

9609: arXiv:hep-th/0402124.

9610: %%CITATION = HEP-TH 0402124;%%

9611: 

9612: \bibitem{9611173}

9613: E.~Bergshoeff and P.~K.~Townsend,

9614: %``Super D-branes,''

9615: Nucl.\ Phys.\ B {\bf 490}, 145 (1997)

9616: [arXiv:hep-th/9611173].

9617: %%CITATION = HEP-TH 9611173;%%

9618: 

9619: \bibitem{0003221}

9620: E.~A.~Bergshoeff, M.~de Roo, T.~C.~de Wit, E.~Eyras and S.~Panda,

9621: %``T-duality and actions for non-BPS D-branes,''

9622: JHEP {\bf 0005}, 009 (2000) [arXiv:hep-th/0003221].

9623: %%CITATION = HEP-TH 0003221;%%

9624: 

9625: \bibitem{9108021}

9626: N.~Berkovits, M.~T.~Hatsuda and W.~Siegel,

9627: %``The Big picture,''

9628: Nucl.\ Phys.\ B {\bf 371}, 434 (1992)

9629: [arXiv:hep-th/9108021].

9630: %%CITATION = HEP-TH 9108021;%%

9631: 

9632: \bibitem{9503099}

9633: N.~Berkovits,

9634: %``SuperPoincare invariant superstring field theory,''

9635: Nucl.\ Phys.\ B {\bf 450}, 90 (1995) [Erratum-ibid.\ B {\bf 459},

9636: 439 (1996)] [arXiv:hep-th/9503099].

9637: %%CITATION = HEP-TH 9503099;%%

9638: 

9639: \bibitem{9912121}

9640: N.~Berkovits,

9641: %``A new approach to superstring field theory,''

9642: Fortsch.\ Phys.\  {\bf 48}, 31 (2000) [arXiv:hep-th/9912121].

9643: %%CITATION = HEP-TH 9912121;%%

9644: 

9645: \bibitem{0001084}

9646: N.~Berkovits,

9647: %``The tachyon potential in open Neveu-Schwarz string field theory,''

9648: JHEP {\bf 0004}, 022 (2000) [arXiv:hep-th/0001084].

9649: %%CITATION = HEP-TH 0001084;%%

9650: 

9651: \bibitem{0002211}

9652: N.~Berkovits, A.~Sen and B.~Zwiebach,

9653: %``Tachyon condensation in superstring field theory,''

9654: Nucl.\ Phys.\ B {\bf 587}, 147 (2000) [arXiv:hep-th/0002211].

9655: %%CITATION = HEP-TH 0002211;%%

9656: 

9657: \bibitem{0105230}

9658: N.~Berkovits,

9659: %``Review of open superstring field theory,''

9660: arXiv:hep-th/0105230.

9661: %%CITATION = HEP-TH 0105230;%%

9662: 

9663: \bibitem{0109100}

9664: N.~Berkovits,

9665: %``The Ramond sector of open superstring field theory,''

9666: JHEP {\bf 0111}, 047 (2001)

9667: [arXiv:hep-th/0109100].

9668: %%CITATION = HEP-TH 0109100;%%

9669: 

9670: \bibitem{0308069}

9671: S.~Bhattacharya, S.~Mukherji and S.~Roy,

9672: %``On the effective action of a space-like brane,''

9673: Phys.\ Lett.\ B {\bf 584}, 163 (2004)

9674: [arXiv:hep-th/0308069].

9675: %%CITATION = HEP-TH 0308069;%%

9676: 

9677: \bibitem{0201060}

9678: L.~Bonora, D.~Mamone and M.~Salizzoni,

9679: %``B field and squeezed states in vacuum string field theory,''

9680: Nucl.\ Phys.\ B {\bf 630}, 163 (2002)

9681: [arXiv:hep-th/0201060].

9682: %%CITATION = HEP-TH 0201060;%%

9683: 

9684: \bibitem{0203188}

9685: L.~Bonora, D.~Mamone and M.~Salizzoni,

9686: %``Vacuum string field theory with B field,''

9687: JHEP {\bf 0204}, 020 (2002)

9688: [arXiv:hep-th/0203188].

9689: %%CITATION = HEP-TH 0203188;%%

9690: 

9691: \bibitem{0207044}

9692: L.~Bonora, D.~Mamone and M.~Salizzoni,

9693: %``Vacuum string field theory ancestors of the GMS solitons,''

9694: JHEP {\bf 0301}, 013 (2003)

9695: [arXiv:hep-th/0207044].

9696: %%CITATION = HEP-TH 0207044;%%

9697: 

9698: \bibitem{0304270}

9699: L.~Bonora, C.~Maccaferri, D.~Mamone and M.~Salizzoni,

9700: %``Topics in string field theory,''

9701: arXiv:hep-th/0304270.

9702: %%CITATION = HEP-TH 0304270;%%

9703: 

9704: \bibitem{0311198}

9705: L.~Bonora, C.~Maccaferri and P.~Prester,

9706: %``Dressed sliver solutions in vacuum string field theory,''

9707: JHEP {\bf 0401}, 038 (2004)

9708: [arXiv:hep-th/0311198].

9709: %%CITATION = HEP-TH 0311198;%%

9710: 

9711: \bibitem{0404154}

9712: L.~Bonora, C.~Maccaferri and P.~Prester,

9713: %``The perturbative spectrum of the dressed sliver,''

9714: arXiv:hep-th/0404154.

9715: %%CITATION = HEP-TH 0404154;%%

9716: 

9717: \bibitem{0409063}

9718: L.~Bonora, C.~Maccaferri, R.~J.~S.~Santos and D.~D.~Tolla,

9719: %``Exact time-localized solutions in vacuum string field theory,''

9720: arXiv:hep-th/0409063.

9721: %%CITATION = HEP-TH 0409063;%%

9722: 

9723: \bibitem{0002023}

9724: P.~Bouwknegt and V.~Mathai,

9725: %``D-branes, B-fields and twisted K-theory,''

9726: JHEP {\bf 0003}, 007 (2000)

9727: [arXiv:hep-th/0002023].

9728: %%CITATION = HEP-TH 0002023;%%

9729: 

9730: \bibitem{0005242}

9731: P.~Brax, G.~Mandal and Y.~Oz,

9732: %``Supergravity description of non-BPS branes,''

9733: Phys.\ Rev.\ D {\bf 63}, 064008 (2001)

9734: [arXiv:hep-th/0005242].

9735: %%CITATION = HEP-TH 0005242;%%

9736: 

9737: \bibitem{0304197}

9738: P.~Brax, J.~Mourad and D.~A.~Steer,

9739: %``Tachyon kinks on non BPS D-branes,''

9740: arXiv:hep-th/0304197.

9741: %%CITATION = HEP-TH 0304197;%%

9742: 

9743: \bibitem{0310079}

9744: P.~Brax, J.~Mourad and D.~A.~Steer,

9745: %``On tachyon kinks from the DBI action,''

9746: arXiv:hep-th/0310079.

9747: %%CITATION = HEP-TH 0310079;%%

9748: 

9749: \bibitem{padic}

9750: L.\ Brekke, P.G.O.\ Freund, M.\ Olson and E.\ Witten,

9751: %{\em Non-archimedean string dynamics},

9752: Nucl.\ Phys.\  {\bf B302} (1988) 365.

9753: %%CITATION = NUPHA,B302,365;%%

9754: 

9755: \bibitem{BKZ}

9756: E.~Brezin, V.~A.~Kazakov and A.~B.~Zamolodchikov,

9757: %``Scaling Violation In A Field Theory Of Closed Strings In One Physical

9758: %Dimension,''

9759: Nucl.\ Phys.\ B {\bf 338}, 673 (1990).

9760: %%CITATION = NUPHA,B338,673;%%

9761: 

9762: \bibitem{0207235}

9763: A.~Buchel, P.~Langfelder and J.~Walcher,

9764: %``Does the tachyon matter?,''

9765: Annals Phys.\  {\bf 302}, 78 (2002) [arXiv:hep-th/0207235].

9766: %%CITATION = HEP-TH 0207235;%%

9767: 

9768: \bibitem{0212150}

9769: A.~Buchel and J.~Walcher,

9770: %``The tachyon does matter,''

9771: Fortsch.\ Phys.\  {\bf 51}, 885 (2003)

9772: [arXiv:hep-th/0212150].

9773: %%CITATION = HEP-TH 0212150;%%

9774: 

9775: \bibitem{0305055}

9776: A.~Buchel and J.~Walcher,

9777: %``Comments on supergravity description of S-branes,''

9778: JHEP {\bf 0305}, 069 (2003)

9779: [arXiv:hep-th/0305055].

9780: %%CITATION = HEP-TH 0305055;%%

9781: 

9782: \bibitem{0105204}

9783: C.~P.~Burgess, M.~Majumdar, D.~Nolte, F.~Quevedo, G.~Rajesh and 

9784: R.~J.~Zhang,

9785: %``The inflationary brane-antibrane universe,''

9786: JHEP {\bf 0107}, 047 (2001)

9787: [arXiv:hep-th/0105204].

9788: %%CITATION = HEP-TH 0105204;%%

9789: 

9790: \bibitem{callan2}

9791: C.~G.~.~Callan, C.~Lovelace, C.~R.~Nappi and S.~A.~Yost,

9792: %``Adding Holes And Crosscaps To The Superstring,''

9793: Nucl.\ Phys.\ B {\bf 293}, 83 (1987).

9794: %%CITATION = NUPHA,B293,83;%%

9795: 

9796: \bibitem{9402113}

9797: C.~G.~.~Callan, I.~R.~Klebanov, A.~W.~W.~Ludwig and

9798: J.~M.~Maldacena,

9799: %``Exact solution of a boundary conformal field theory,''

9800: Nucl.\ Phys.\ B {\bf 422}, 417 (1994) [arXiv:hep-th/9402113].

9801: %%CITATION = HEP-TH 9402113;%%

9802: 

9803: \bibitem{9610148}

9804: M.~Cederwall, A.~von Gussich, B.~E.~W.~Nilsson and 

9805: A.~Westerberg,

9806: %``The Dirichlet super-three-brane in ten-dimensional 

9807: %type IIB  supergravity,''

9808: Nucl.\ Phys.\ B {\bf 490}, 163 (1997)

9809: [arXiv:hep-th/9610148].

9810: %%CITATION = HEP-TH 9610148;%%

9811: 

9812: \bibitem{9611159}

9813: M.~Cederwall, A.~von Gussich, B.~E.~W.~Nilsson, 

9814: P.~Sundell and A.~Westerberg,

9815: %``The Dirichlet super-p-branes in ten-dimensional type 

9816: %IIA and IIB

9817: %supergravity,''

9818: Nucl.\ Phys.\ B {\bf 490}, 179 (1997)

9819: [arXiv:hep-th/9611159].

9820: %%CITATION = HEP-TH 9611159;%%

9821: 

9822: \bibitem{0103056}

9823: G.~Chalmers,

9824: %``Open string decoupling and tachyon condensation,''

9825: JHEP {\bf 0106}, 012 (2001)

9826: [arXiv:hep-th/0103056].

9827: %%CITATION = HEP-TH 0103056;%%

9828: 

9829: \bibitem{0204233}

9830: B.~Chen and F.~L.~Lin,

9831: %``D-branes as GMS solitons in vacuum string field 

9832: %theory,''

9833: Phys.\ Rev.\ D {\bf 66}, 126001 (2002)

9834: [arXiv:hep-th/0204233].

9835: %%CITATION = HEP-TH 0204233;%%

9836: 

9837: \bibitem{0209222}

9838: B.~Chen, M.~Li and F.~L.~Lin,

9839: %``Gravitational radiation of rolling tachyon,''

9840: JHEP {\bf 0211}, 050 (2002)

9841: [arXiv:hep-th/0209222].

9842: %%CITATION = HEP-TH 0209222;%%

9843: 

9844: \bibitem{0204071}

9845: C.~M.~Chen, D.~V.~Gal'tsov and M.~Gutperle,

9846: %``S-brane solutions in supergravity theories,''

9847: Phys.\ Rev.\ D {\bf 66}, 024043 (2002)

9848: [arXiv:hep-th/0204071].

9849: %%CITATION = HEP-TH 0204071;%%

9850: 

9851: \bibitem{0301119}

9852: T.~Chen, J.~Frohlich and J.~Walcher,

9853: %``The decay of unstable noncommutative solitons,''

9854: Commun.\ Math.\ Phys.\  {\bf 237}, 243 (2003)

9855: [arXiv:hep-th/0301119].

9856: %%CITATION = HEP-TH 0301119;%%

9857: 

9858: \bibitem{0311179}

9859: X.~g.~Chen,

9860: %``One loop evolution in rolling tachyon,''

9861: arXiv:hep-th/0311179.

9862: %%CITATION = HEP-TH 0311179;%%

9863: 

9864: \bibitem{0209142}

9865: T.~E.~Clark, M.~Nitta and T.~ter Veldhuis,

9866: %``The effective action for brane gauge fields,''

9867: Phys.\ Rev.\ D {\bf 69}, 047701 (2004)

9868: [arXiv:hep-th/0209142].

9869: %%CITATION = HEP-TH 0209142;%%

9870: 

9871: \bibitem{0401163}

9872: T.~E.~Clark, M.~Nitta and T.~ter Veldhuis,

9873: %``Non-BPS brane dynamics and duality,''

9874: arXiv:hep-th/0401163.

9875: %%CITATION = HEP-TH 0401163;%%

9876: 

9877: \bibitem{0409030}

9878: T.~E.~Clark, M.~Nitta and T.~t.~Veldhuis,

9879: %``Superconformal symmetry, the supercurrent and non-BPS 

9880: %brane dynamics,''

9881: arXiv:hep-th/0409030.

9882: %%CITATION = HEP-TH 0409030;%%

9883: 

9884: \bibitem{0409151}

9885: T.~E.~Clark, M.~Nitta and T.~ter Veldhuis,

9886: %``Non-BPS brane dynamics and dual tensor gauge theory,''

9887: arXiv:hep-th/0409151.

9888: %%CITATION = HEP-TH 0409151;%%

9889: 

9890: \bibitem{0301101}

9891: J.~M.~Cline and H.~Firouzjahi,

9892: %``Real-time D-brane condensation,''

9893: Phys.\ Lett.\ B {\bf 564}, 255 (2003)

9894: [arXiv:hep-th/0301101].

9895: %%CITATION = HEP-TH 0301101;%%

9896: 

9897: \bibitem{COLEMAN}

9898: S.~R.~Coleman,

9899: %``Quantum Sine-Gordon Equation As The Massive Thirring Model,''

9900: Phys.\ Rev.\ D {\bf 11}, 2088 (1975).

9901: %%CITATION = PHRVA,D11,2088;%%

9902: 

9903: \bibitem{0305177}

9904: N.~R.~Constable and F.~Larsen,

9905: %``The rolling tachyon as a matrix model,''

9906: arXiv:hep-th/0305177.

9907: %%CITATION = HEP-TH 0305177;%%

9908: 

9909: \bibitem{0306294}

9910: E.~J.~Copeland, P.~M.~Saffin and D.~A.~Steer,

9911: %``Singular tachyon kinks from regular profiles,''

9912: Phys.\ Rev.\ D {\bf 68}, 065013 (2003) [arXiv:hep-th/0306294].

9913: %%CITATION = HEP-TH 0306294;%%

9914: 

9915: \bibitem{0012217}

9916: S.~Corley and S.~Ramgoolam,

9917: %``Projector equivalences in K theory and families of 

9918: %non-commutative

9919: %solitons,''

9920: JHEP {\bf 0103}, 037 (2001)

9921: [arXiv:hep-th/0012217].

9922: %%CITATION = HEP-TH 0012217;%%

9923: 

9924: \bibitem{0010021}

9925: L.~Cornalba,

9926: %``Tachyon condensation in large magnetic fields with 

9927: %background  independent

9928: %string field theory,''

9929: Phys.\ Lett.\ B {\bf 504}, 55 (2001)

9930: [arXiv:hep-th/0010021].

9931: %%CITATION = HEP-TH 0010021;%%

9932: 

9933: \bibitem{0105227}

9934: B.~Craps, P.~Kraus and F.~Larsen,

9935: %``Loop corrected tachyon condensation,''

9936: JHEP {\bf 0106}, 062 (2001)

9937: [arXiv:hep-th/0105227].

9938: %%CITATION = HEP-TH 0105227;%%

9939: 

9940: \bibitem{0403217}

9941: B.~C.~da Cunha,

9942: %``Tachyon effective dynamics and de Sitter vacua,''

9943: arXiv:hep-th/0403217.

9944: %%CITATION = HEP-TH 0403217;%%

9945: 

9946: \bibitem{dai}

9947: J. Dai, R. Leigh and J. Polchinski, Mod. Phys. Lett. {\bf A4}

9948: (1989) 2073.

9949: %%CITATION = MPLAE,A4,2073;%%

9950: 

9951: \bibitem{DASJEV}

9952: S.~R.~Das and A.~Jevicki,

9953: %``String Field Theory And Physical Interpretation Of D = 1 Strings,''

9954: Mod.\ Phys.\ Lett.\ A {\bf 5}, 1639 (1990).

9955: %%CITATION = MPLAE,A5,1639;%%

9956: 

9957: \bibitem{9110021}

9958: S.~R.~Das, A.~Dhar, G.~Mandal and S.~R.~Wadia,

9959:  %``Gauge theory formulation of the C = 1 matrix model: Symmetries and

9960: %discrete

9961: %states,''

9962: Int.\ J.\ Mod.\ Phys.\ A {\bf 7}, 5165 (1992)

9963: [arXiv:hep-th/9110021].

9964: %%CITATION = HEP-TH 9110021;%%

9965: 

9966: \bibitem{0401067}

9967: S.~R.~Das,

9968: %``D branes in 2d string theory and classical limits,''

9969: arXiv:hep-th/0401067.

9970: %%CITATION = HEP-TH 0401067;%%

9971: 

9972: \bibitem{0005006}

9973: K.~Dasgupta, S.~Mukhi and G.~Rajesh,

9974: %``Noncommutative tachyons,''

9975: JHEP {\bf 0006}, 022 (2000)

9976: [arXiv:hep-th/0005006].

9977: %%CITATION = HEP-TH 0005006;%%

9978: 

9979: \bibitem{0010247}

9980: S.~Dasgupta and T.~Dasgupta,

9981: %``Renormalization group analysis of tachyon 

9982: %condensation,''

9983: JHEP {\bf 0106}, 007 (2001)

9984: [arXiv:hep-th/0010247].

9985: %%CITATION = HEP-TH 0010247;%%

9986: 

9987: \bibitem{0310106}

9988: S.~Dasgupta and T.~Dasgupta,

9989: %``Renormalization group approach to c = 1 matrix model on a circle and 

9990: %D-brane

9991: %decay,''

9992: arXiv:hep-th/0310106.

9993: %%CITATION = HEP-TH 0310106;%%

9994: 

9995: \bibitem{DAVID}

9996: F.~David,

9997: %``Conformal Field Theories Coupled To 2-D Gravity In The Conformal

9998: %Gauge,''

9999: Mod.\ Phys.\ Lett.\ A {\bf 3}, 1651 (1988).

10000: %%CITATION = MPLAE,A3,1651;%%

10001: 

10002: \bibitem{0012089}

10003: J.~R.~David,

10004: %``Tachyon condensation using the disc partition 

10005: %function,''

10006: JHEP {\bf 0107}, 009 (2001)

10007: [arXiv:hep-th/0012089].

10008: %%CITATION = HEP-TH 0012089;%%

10009: 

10010: \bibitem{0105184}

10011: J.~R.~David,

10012: %``Excitations on wedge states and on the sliver,''

10013: JHEP {\bf 0107}, 024 (2001)

10014: [arXiv:hep-th/0105184].

10015: %%CITATION = HEP-TH 0105184;%%

10016: 

10017: \bibitem{0007235}

10018: J.~R.~David,

10019: %``Tachyon condensation in the D0/D4 system,''

10020: JHEP {\bf 0010}, 004 (2000)

10021: [arXiv:hep-th/0007235].

10022: %%CITATION = HEP-TH 0007235;%%

10023: 

10024: \bibitem{0011223}

10025: S.~P.~de Alwis and A.~T.~Flournoy,

10026: %``Some issues in noncommutative solitons as D-branes,''

10027: Phys.\ Rev.\ D {\bf 63}, 106001 (2001)

10028: [arXiv:hep-th/0011223].

10029: %%CITATION = HEP-TH 0011223;%%

10030: 

10031: \bibitem{0312135}

10032: J.~de Boer, A.~Sinkovics, E.~Verlinde and J.~T.~Yee,

10033: %``String interactions in c = 1 matrix model,''

10034: JHEP {\bf 0403}, 023 (2004) [arXiv:hep-th/0312135].

10035: %%CITATION = HEP-TH 0312135;%%

10036: 

10037: \bibitem{0003031}

10038: R.~de Mello Koch, A.\ Jevicki, M.\ Mihailescu and R.\ Tatar,

10039: %{\em Lumps and p-branes in open string field theory,}

10040: Phys.\ Lett.\  {\bf B482}, 249 (2000)

10041: [hep-th/0003031].

10042: %%CITATION = HEP-TH 0003031;%%

10043: 

10044: \bibitem{0008053}

10045: R.~de Mello Koch and J.~P.~Rodrigues,

10046: %``Lumps in level truncated open string field theory,''

10047: Phys.\ Lett.\ B {\bf 495}, 237 (2000) [arXiv:hep-th/0008053].

10048: %%CITATION = HEP-TH 0008053;%%

10049: 

10050: \bibitem{0107070}

10051: P.~J.~De Smet and J.~Raeymaekers,

10052: %``Toy model for tachyon condensation in bosonic string field theory,''

10053: JHEP {\bf 0107}, 032 (2001)

10054: [arXiv:hep-th/0107070].

10055: %%CITATION = HEP-TH 0107070;%%

10056: 

10057: \bibitem{0109182}

10058: P.~J.~De Smet,

10059: %``Tachyon condensation: Calculations in string field 

10060: %theory,''

10061: arXiv:hep-th/0109182.

10062: %%CITATION = HEP-TH 0109182;%%

10063: 

10064: \bibitem{0003220}

10065: P.~J.~De Smet and J.~Raeymaekers,

10066: %``Level four approximation to the tachyon potential in superstring field

10067: %theory,''

10068: JHEP {\bf 0005}, 051 (2000) [arXiv:hep-th/0003220].

10069: %%CITATION = HEP-TH 0003220;%%

10070: 

10071: \bibitem{0004112}

10072: P.~De Smet and J.~Raeymaekers,

10073: %{\em The tachyon potential in Witten's superstring field theory,}

10074: JHEP {\bf 0008}, 020 (2000)

10075: [hep-th/0004112].

10076: %%CITATION = HEP-TH 0004112;%%

10077: 

10078: \bibitem{0305191}

10079: Y.~Demasure and R.~A.~Janik,

10080: %``Backreaction and the rolling tachyon: An effective action point of  

10081: %view,''

10082: Phys.\ Lett.\ B {\bf 578}, 195 (2004)

10083: [arXiv:hep-th/0305191].

10084: %%CITATION = HEP-TH 0305191;%%

10085: 

10086: \bibitem{0309148}

10087: O.~DeWolfe, R.~Roiban, M.~Spradlin, A.~Volovich and 

10088: J.~Walcher,

10089: %``On the S-matrix of type 0 string theory,''

10090: JHEP {\bf 0311}, 012 (2003)

10091: [arXiv:hep-th/0309148].

10092: %%CITATION = HEP-TH 0309148;%%

10093: 

10094: \bibitem{9212027}

10095: A.~Dhar, G.~Mandal and S.~R.~Wadia,

10096: %``A Time dependent classical solution of c = 1 string field theory and

10097: %nonperturbative effects,''

10098: Int.\ J.\ Mod.\ Phys.\ A {\bf 8}, 3811 (1993)

10099: [arXiv:hep-th/9212027].

10100: %%CITATION = HEP-TH 9212027;%%

10101: 

10102: \bibitem{9507041}

10103: A.~Dhar, G.~Mandal and S.~R.~Wadia,

10104: %``Discrete state moduli of string theory from the C=1 matrix model,''

10105: Nucl.\ Phys.\ B {\bf 454}, 541 (1995)

10106: [arXiv:hep-th/9507041].

10107: %%CITATION = HEP-TH 9507041;%%

10108: 

10109: \bibitem{9109005}

10110: P.~Di Francesco and D.~Kutasov,

10111: %``World sheet and space-time physics in two-dimensional (Super)string

10112: %theory,''

10113: Nucl.\ Phys.\ B {\bf 375}, 119 (1992)

10114: [arXiv:hep-th/9109005].

10115: %%CITATION = HEP-TH 9109005;%%

10116: 

10117: \bibitem{DVV}

10118: R.~Dijkgraaf, E.~Verlinde and H.~Verlinde,

10119: %``C = 1 Conformal Field Theories On Riemann Surfaces,''

10120: Commun.\ Math.\ Phys.\  {\bf 115}, 649 (1988).

10121: %%CITATION = CMPHA,115,649;%%

10122: 

10123: \bibitem{DISKAW}

10124: J.~Distler and H.~Kawai,

10125:  %``Conformal Field Theory And 2-D Quantum Gravity Or Who's Afraid Of

10126: %Joseph

10127: %Liouville?,''

10128: Nucl.\ Phys.\ B {\bf 321}, 509 (1989).

10129: %%CITATION = NUPHA,B321,509;%%

10130: 

10131: \bibitem{9707068}

10132: P.~Di Vecchia, M.~Frau, I.~Pesando, S.~Sciuto, A.~Lerda and

10133: R.~Russo,

10134: %``Classical p-branes from boundary state,''

10135: Nucl.\ Phys.\ B {\bf 507}, 259 (1997)

10136: [arXiv:hep-th/9707068].

10137: %%CITATION = HEP-TH 9707068;%%

10138: 

10139: \bibitem{9912275}

10140: P.~Di Vecchia and A.~Liccardo,

10141: %``D-branes in string theory. II,''

10142: arXiv:hep-th/9912275.

10143: %%CITATION = HEP-TH 9912275;%%

10144: 

10145: \bibitem{0011090}

10146: E.~E.~Donets, A.~P.~Isaev, C.~Sochichiu and M.~Tsulaia,

10147: %``Brane vacuum as chain of rotators,''

10148: JHEP {\bf 0012}, 022 (2000)

10149: [arXiv:hep-th/0011090].

10150: %%CITATION = HEP-TH 0011090;%%

10151: 

10152: \bibitem{9206053}

10153: H.~Dorn and H.~J.~Otto,

10154: %``On correlation functions for noncritical strings with c <= 1 d >=

10155: %1,''

10156: Phys.\ Lett.\ B {\bf 291}, 39 (1992)

10157: [arXiv:hep-th/9206053].

10158: %%CITATION = HEP-TH 9206053;%%

10159: 

10160: \bibitem{9403141}

10161: H.~Dorn and H.~J.~Otto,

10162: %``Two and three point functions in Liouville theory,''

10163: Nucl.\ Phys.\ B {\bf 429}, 375 (1994)

10164: [arXiv:hep-th/9403141].

10165: %%CITATION = HEP-TH 9403141;%%

10166: 

10167: \bibitem{9711165}

10168: M.~R.~Douglas and C.~M.~Hull,

10169: %``D-branes and the noncommutative torus,''

10170: JHEP {\bf 9802}, 008 (1998)

10171: [arXiv:hep-th/9711165].

10172: %%CITATION = HEP-TH 9711165;%%

10173: 

10174: \bibitem{0202087}

10175: M.~R.~Douglas, H.~Liu, G.~W.~Moore and B.~Zwiebach,

10176: %``Open string star as a continuous Moyal product,''

10177: JHEP {\bf 0204}, 022 (2002)

10178: [arXiv:hep-th/0202087].

10179: %%CITATION = HEP-TH 0202087;%%

10180: 

10181: \bibitem{0307195}

10182: M.~R.~Douglas, I.~R.~Klebanov, D.~Kutasov, J.~Maldacena, E.~Martinec

10183: and

10184: N.~Seiberg,

10185: %``A new hat for the c = 1 matrix model,''

10186: arXiv:hep-th/0307195.

10187: %%CITATION = HEP-TH 0307195;%%

10188: 

10189: \bibitem{0004131}

10190: N.~Drukker, D.~J.~Gross and N.~Itzhaki,

10191: %``Sphalerons, merons and unstable branes in AdS,''

10192: Phys.\ Rev.\ D {\bf 62}, 086007 (2000)

10193: [arXiv:hep-th/0004131].

10194: %%CITATION = HEP-TH 0004131;%%

10195: 

10196: \bibitem{0301079}

10197: N.~Drukker,

10198: %``On different actions for the vacuum of bosonic string 

10199: %field theory,''

10200: JHEP {\bf 0308}, 017 (2003)

10201: [arXiv:hep-th/0301079].

10202: %%CITATION = HEP-TH 0301079;%%

10203: 

10204: \bibitem{9812483}

10205: G.~R.~Dvali and S.~H.~H.~Tye,

10206: %``Brane inflation,''

10207: Phys.\ Lett.\ B {\bf 450}, 72 (1999)

10208: [arXiv:hep-ph/9812483].

10209: %%CITATION = HEP-PH 9812483;%%

10210: 

10211: \bibitem{0105203}

10212: G.~R.~Dvali, Q.~Shafi and S.~Solganik,

10213: %``D-brane inflation,''

10214: arXiv:hep-th/0105203.

10215: %%CITATION = HEP-TH 0105203;%%

10216: 

10217: \bibitem{0103085}

10218: I.~Ellwood and W.~Taylor,

10219: %``Open string field theory without open strings,''

10220: Phys.\ Lett.\ B {\bf 512}, 181 (2001) [arXiv:hep-th/0103085].

10221: %%CITATION = HEP-TH 0103085;%%

10222: 

10223: \bibitem{0105024}

10224: I.~Ellwood, B.~Feng, Y.~H.~He and N.~Moeller,

10225: %``The identity string field and the tachyon vacuum,''

10226: JHEP {\bf 0107}, 016 (2001) [arXiv:hep-th/0105024].

10227: %%CITATION = HEP-TH 0105024;%%

10228: 

10229: \bibitem{0105156}

10230: I.~Ellwood and W.~Taylor,

10231: %``Gauge invariance and tachyon condensation in open string field 

10232: %theory,''

10233: arXiv:hep-th/0105156.

10234: %%CITATION = HEP-TH 0105156;%%

10235: 

10236: \bibitem{0406199}

10237: T.~G.~Erler and D.~J.~Gross,

10238: %``Locality, causality, and an initial value formulation for open string 

10239: %field

10240: %theory,''

10241: arXiv:hep-th/0406199.

10242: %%CITATION = HEP-TH 0406199;%%

10243: 

10244: \bibitem{0409179}

10245: T.~G.~Erler,

10246: %``Level Truncation and Rolling the Tachyon in the Lightcone Basis for 

10247: %Open

10248: %String Field Theory,''

10249: arXiv:hep-th/0409179.

10250: %%CITATION = HEP-TH 0409179;%%

10251: 

10252: \bibitem{9908121}

10253: E.~Eyras,

10254: %``Stable D8-branes and tachyon condensation in type 0 

10255: %open string theory,''

10256: JHEP {\bf 9910}, 005 (1999)

10257: [arXiv:hep-th/9908121].

10258: %%CITATION = HEP-TH 9908121;%%

10259: 

10260: \bibitem{0208019}

10261: G.~N.~Felder, L.~Kofman and A.~Starobinsky,

10262: %``Caustics in tachyon matter and other Born-Infeld scalars,''

10263: JHEP {\bf 0209}, 026 (2002) [arXiv:hep-th/0208019].

10264: %%CITATION = HEP-TH 0208019;%%

10265: 

10266: \bibitem{0403073}

10267: G.~N.~Felder and L.~Kofman,

10268: %``Inhomogeneous fragmentation of the rolling tachyon,''

10269: Phys.\ Rev.\ D {\bf 70}, 046004 (2004)

10270: [arXiv:hep-th/0403073].

10271: %%CITATION = HEP-TH 0403073;%%

10272: 

10273: \bibitem{9406125}

10274: P.~Fendley, H.~Saleur and N.~P.~Warner,

10275:  %``Exact solution of a massless scalar field with a relevant boundary

10276: %interaction,''

10277: Nucl.\ Phys.\ B {\bf 430}, 577 (1994) [arXiv:hep-th/9406125].

10278: %%CITATION = HEP-TH 9406125;%%

10279: 

10280: \bibitem{0103103}

10281: B.~Feng, Y.~H.~He and N.~Moeller,

10282: %``Testing the uniqueness of the open bosonic string field theory 

10283: vacuum,''

10284: arXiv:hep-th/0103103.

10285: %%CITATION = HEP-TH 0103103;%%

10286: 

10287: \bibitem{0310253}

10288: A.~Fotopoulos and A.~A.~Tseytlin,

10289: %``On open superstring partition function 

10290: %in inhomogeneous rolling tachyon

10291: %background,''

10292: JHEP {\bf 0312}, 025 (2003)

10293: [arXiv:hep-th/0310253].

10294: %%CITATION = HEP-TH 0310253;%%

10295: 

10296: \bibitem{Frampton1}

10297: P.~H.~Frampton and Y.~Okada,

10298: %``The P-Adic String N Point Function,''

10299: Phys.\ Rev.\ Lett.\  {\bf 60}, 484 (1988).

10300: %%CITATION = PRLTA,60,484;%%

10301: 

10302: \bibitem{Frampton2}

10303: P.~H.~Frampton and Y.~Okada,

10304: %``Effective Scalar Field Theory Of P-Adic String,''

10305: Phys.\ Rev.\ D {\bf 37}, 3077 (1988).

10306: %%CITATION = PHRVA,D37,3077;%%

10307: 

10308: \bibitem{9903123}

10309: M.~Frau, L.~Gallot, A.~Lerda and P.~Strigazzi,

10310: %``Stable non-BPS D-branes in type I string theory,''

10311: Nucl.\ Phys.\ B {\bf 564}, 60 (2000)

10312: [arXiv:hep-th/9903123].

10313: %%CITATION = HEP-TH 9903123;%%

10314: 

10315: \bibitem{0012164}

10316: S.~Fredenhagen and V.~Schomerus,

10317: %``Branes on group manifolds, gluon condensates, and 

10318: %twisted K-theory,''

10319: JHEP {\bf 0104}, 007 (2001)

10320: [arXiv:hep-th/0012164].

10321: %%CITATION = HEP-TH 0012164;%%

10322: 

10323: \bibitem{0308205}

10324: S.~Fredenhagen and V.~Schomerus,

10325: %``On minisuperspace models of S-branes,''

10326: arXiv:hep-th/0308205.

10327: %%CITATION = HEP-TH 0308205;%%

10328: 

10329: \bibitem{Freund1}

10330: P.~G.~O.~Freund and M.~Olson,

10331: %``Nonarchimedean Strings,''

10332: Phys.\ Lett.\ B {\bf 199}, 186 (1987).

10333: %%CITATION = PHLTA,B199,186;%%

10334: 

10335: \bibitem{Freund2}

10336: P.~G.~O.~Freund and E.~Witten,

10337: %``Adelic String Amplitudes,''

10338: Phys.\ Lett.\ B {\bf 199}, 191 (1987).

10339: %%CITATION = PHLTA,B199,191;%%

10340: 

10341: \bibitem{THORN}

10342: D.~Z.~Freedman, S.~B.~Giddings, J.~A.~Shapiro and C.~B.~Thorn,

10343: %``The Nonplanar One Loop Amplitude In Witten's String Field Theory,''

10344: Nucl.\ Phys.\ B {\bf 298}, 253 (1988).

10345: %%CITATION = NUPHA,B298,253;%%

10346: 

10347: \bibitem{FMS1}

10348: D.~Friedan, S.~H.~Shenker and E.~J.~Martinec,

10349: %``Covariant Quantization Of Superstrings,''

10350: Phys.\ Lett.\ B {\bf 160}, 55 (1985).

10351: %%CITATION = PHLTA,B160,55;%%

10352: 

10353: \bibitem{FMS}

10354: D.~Friedan, E.~J.~Martinec and S.~H.~Shenker,

10355: %``Conformal Invariance, Supersymmetry And String Theory,''

10356: Nucl.\ Phys.\ B {\bf 271}, 93 (1986).

10357: %%CITATION = NUPHA,B271,93;%%

10358: 

10359: \bibitem{0207001}

10360: E.~Fuchs, M.~Kroyter and A.~Marcus,

10361: %``Squeezed state projectors in string field theory,''

10362: JHEP {\bf 0209}, 022 (2002)

10363: [arXiv:hep-th/0207001].

10364: %%CITATION = HEP-TH 0207001;%%

10365: 

10366: \bibitem{0304163}

10367: M.~Fujita and H.~Hata,

10368: %``Time Dependent Solution in Cubic String Field Theory,''

10369: arXiv:hep-th/0304163.

10370: %%CITATION = HEP-TH 0304163;%%

10371: 

10372: \bibitem{0403031}

10373: M.~Fujita and H.~Hata,

10374: %``Rolling tachyon solution in vacuum string field theory,''

10375: arXiv:hep-th/0403031.

10376: %%CITATION = HEP-TH 0403031;%%

10377: 

10378: \bibitem{9910109}

10379: M.~R.~Gaberdiel and B.~J.~Stefanski,

10380: %``Dirichlet branes on orbifolds,''

10381: Nucl.\ Phys.\ B {\bf 578}, 58 (2000)

10382: [arXiv:hep-th/9910109].

10383: %%CITATION = HEP-TH 9910109;%%

10384: 

10385: \bibitem{0005029}

10386: M.~R.~Gaberdiel,

10387: %``Lectures on non-BPS Dirichlet branes,''

10388: Class.\ Quant.\ Grav.\  {\bf 17}, 3483 (2000)

10389: [arXiv:hep-th/0005029].

10390: %%CITATION = HEP-TH 0005029;%%

10391: 

10392: \bibitem{0108102}

10393: M.~R.~Gaberdiel, A.~Recknagel and G.~M.~T.~Watts,

10394: %``The conformal boundary states for SU(2) at level 1,''

10395: Nucl.\ Phys.\ B {\bf 626}, 344 (2002)

10396: [arXiv:hep-th/0108102].

10397: %%CITATION = HEP-TH 0108102;%%

10398: 

10399: \bibitem{0108238}

10400: M.~R.~Gaberdiel and A.~Recknagel,

10401: %``Conformal boundary states for free bosons and fermions,''

10402: JHEP {\bf 0111}, 016 (2001) [arXiv:hep-th/0108238].

10403: %%CITATION = HEP-TH 0108238;%%

10404: 

10405: \bibitem{0111129}

10406: D.~Gaiotto, L.~Rastelli, A.~Sen and B.~Zwiebach,

10407: %``Ghost structure and closed strings in vacuum string field theory,''

10408: Adv.\ Theor.\ Math.\ Phys.\  {\bf 6}, 403 (2003)

10409: [arXiv:hep-th/0111129].

10410: %%CITATION = HEP-TH 0111129;%%

10411: 

10412: \bibitem{0201159}

10413: D.~Gaiotto, L.~Rastelli, A.~Sen and B.~Zwiebach,

10414: %``Patterns in open string field theory solutions,''

10415: JHEP {\bf 0203}, 003 (2002)

10416: [arXiv:hep-th/0201159].

10417: %%CITATION = HEP-TH 0201159;%%

10418: 

10419: \bibitem{0202151}

10420: D.~Gaiotto, L.~Rastelli, A.~Sen and B.~Zwiebach,

10421: %``Star algebra projectors,''

10422: JHEP {\bf 0204}, 060 (2002)

10423: [arXiv:hep-th/0202151].

10424: %%CITATION = HEP-TH 0202151;%%

10425: 

10426: \bibitem{0211012}

10427: D.~Gaiotto and L.~Rastelli,

10428: %``Experimental string field theory,''

10429: JHEP {\bf 0308}, 048 (2003) [arXiv:hep-th/0211012].

10430: %%CITATION = HEP-TH 0211012;%%

10431: 

10432: \bibitem{0304192}

10433: D.~Gaiotto, N.~Itzhaki and L.~Rastelli,

10434: %``Closed strings as imaginary D-branes,''

10435: arXiv:hep-th/0304192.

10436: %%CITATION = HEP-TH 0304192;%%

10437: 

10438: \bibitem{0307221}

10439: D.~Gaiotto, N.~Itzhaki and L.~Rastelli,

10440: %``On the BCFT description of holes in the c = 1 matrix model,''

10441: Phys.\ Lett.\ B {\bf 575}, 111 (2003)

10442: [arXiv:hep-th/0307221].

10443: %%CITATION = HEP-TH 0307221;%%

10444: 

10445: \bibitem{0312196}

10446: D.~Gaiotto and L.~Rastelli,

10447: %``A paradigm of open/closed duality: Liouville D-branes 

10448: %and the Kontsevich

10449: %model,''

10450: arXiv:hep-th/0312196.

10451: %%CITATION = HEP-TH 0312196;%%

10452: 

10453: \bibitem{9812226}

10454: H.~Garcia-Compean,

10455: %``D-branes in orbifold singularities and equivariant 

10456: %K-theory,''

10457: Nucl.\ Phys.\ B {\bf 557}, 480 (1999)

10458: [arXiv:hep-th/9812226].

10459: %%CITATION = HEP-TH 9812226;%%

10460: 

10461: \bibitem{0003122}

10462: M.~R.~Garousi,

10463: %``Tachyon couplings on non-BPS D-branes and Dirac-Born-Infeld action,''

10464: Nucl.\ Phys.\ B {\bf 584}, 284 (2000) [arXiv:hep-th/0003122].

10465: %%CITATION = HEP-TH 0003122;%%

10466: 

10467: \bibitem{0209068}

10468: M.~R.~Garousi,

10469: %``On-shell S-matrix and tachyonic effective actions,''

10470: Nucl.\ Phys.\ B {\bf 647}, 117 (2002)

10471: [arXiv:hep-th/0209068].

10472: %%CITATION = HEP-TH 0209068;%%

10473: 

10474: \bibitem{0303239}

10475: M.~R.~Garousi,

10476: %``Off-shell extension of S-matrix elements and tachyonic effective  

10477: %actions,''

10478: JHEP {\bf 0304}, 027 (2003)

10479: [arXiv:hep-th/0303239].

10480: %%CITATION = HEP-TH 0303239;%%

10481: 

10482: \bibitem{0304145}

10483: M.~R.~Garousi,

10484: %``Tachyon couplings on non-BPS D-branes and Dirac-Born-Infeld action,''

10485: Nucl.\ Phys.\ B {\bf 584}, 284 (2000) [arXiv:hep-th/0003122], JHEP

10486: {\bf 0305}, 058 (2003)

10487: [arXiv:hep-th/0304145].

10488: %%CITATION = HEP-TH 0304145;%%

10489: 

10490: \bibitem{0307197}

10491: M.~R.~Garousi,

10492: %``S-matrix elements and off-shell tachyon action with non-Abelian gauge

10493: %symmetry,''

10494: JHEP {\bf 0312}, 036 (2003)

10495: [arXiv:hep-th/0307197].

10496: %%CITATION = HEP-TH 0307197;%%

10497: 

10498: \bibitem{9704006}

10499: E.~Gava, K.~S.~Narain and M.~H.~Sarmadi,

10500: %``On the bound states of p- and (p+2)-branes,''

10501: Nucl.\ Phys.\ B {\bf 504}, 214 (1997)

10502: [arXiv:hep-th/9704006].

10503: %%CITATION = HEP-TH 9704006;%%

10504: 

10505: \bibitem{0009103}

10506: A.~A.~Gerasimov and S.~L.~Shatashvili,

10507: %``On exact tachyon potential in open string field theory,''

10508: JHEP {\bf 0010}, 034 (2000) [arXiv:hep-th/0009103].

10509: %%CITATION = HEP-TH 0009103;%%

10510: 

10511: \bibitem{0011009}

10512: A.~A.~Gerasimov and S.~L.~Shatashvili,

10513: %``Stringy Higgs mechanism and the fate of open strings,''

10514: JHEP {\bf 0101}, 019 (2001)

10515: [arXiv:hep-th/0011009].

10516: %%CITATION = HEP-TH 0011009;%%

10517: 

10518: \bibitem{0105245}

10519: A.~A.~Gerasimov and S.~L.~Shatashvili,

10520: %``On non-abelian structures in field theory of open strings,''

10521: JHEP {\bf 0106}, 066 (2001)

10522: [arXiv:hep-th/0105245].

10523: %%CITATION = HEP-TH 0105245;%%

10524: 

10525: \bibitem{0003278}

10526: D.~Ghoshal and A.~Sen,

10527: %``Tachyon condensation and brane descent relations in p-adic string

10528: %theory,''

10529: Nucl.\ Phys.\ B {\bf 584}, 300 (2000) [arXiv:hep-th/0003278].

10530: %%CITATION = HEP-TH 0003278;%%

10531: 

10532: \bibitem{0009191}

10533: D.~Ghoshal and A.~Sen,

10534: %``Normalisation of the background independent open string field theory

10535: %action,''

10536: JHEP {\bf 0011}, 021 (2000)

10537: [arXiv:hep-th/0009191].

10538: %%CITATION = HEP-TH 0009191;%%

10539: 

10540: \bibitem{0106231}

10541: D.~Ghoshal,

10542: %``Normalization of the boundary superstring field theory action,''

10543: arXiv:hep-th/0106231.

10544: %%CITATION = HEP-TH 0106231;%%

10545: 

10546: \bibitem{0406259}

10547: D.~Ghoshal,

10548: %``Exact noncommutative solitons in p-adic strings and BSFT,''

10549: arXiv:hep-th/0406259.

10550: %%CITATION = HEP-TH 0406259;%%

10551: 

10552: \bibitem{0409311}

10553: D.~Ghoshal and T.~Kawano,

10554: %``Towards p-Adic String in Constant B-Field,''

10555: arXiv:hep-th/0409311.

10556: %%CITATION = HEP-TH 0409311;%%

10557: 

10558: \bibitem{0009061}

10559: G.~W.~Gibbons, K.~Hori and P.~Yi,

10560: %``String fluid from unstable D-branes,''

10561: Nucl.\ Phys.\ B {\bf 596}, 136 (2001) [arXiv:hep-th/0009061].

10562: %%CITATION = HEP-TH 0009061;%%

10563: 

10564: \bibitem{0209034}

10565: G.~Gibbons, K.~Hashimoto and P.~Yi,

10566:  %``Tachyon condensates, Carrollian contraction of Lorentz group, and

10567: %fundamental strings,''

10568: JHEP {\bf 0209}, 061 (2002) [arXiv:hep-th/0209034].

10569: %%CITATION = HEP-TH 0209034;%%

10570: 

10571: \bibitem{0204008}

10572: G.~W.~Gibbons,

10573: %``Cosmological evolution of the rolling tachyon,''

10574: Phys.\ Lett.\ B {\bf 537}, 1 (2002)

10575: [arXiv:hep-th/0204008].

10576: %%CITATION = HEP-TH 0204008;%%

10577: 

10578: \bibitem{giddings}

10579: S.~B.~Giddings,

10580: %``The Veneziano Amplitude From Interacting String Field Theory,''

10581: Nucl.\ Phys.\ B {\bf 278}, 242 (1986).

10582: %%CITATION = NUPHA,B278,242;%%

10583: 

10584: \bibitem{GIDDMARW}

10585: S.~B.~Giddings, E.~J.~Martinec and E.~Witten,

10586: %``Modular Invariance In String Field Theory,''

10587: Phys.\ Lett.\ B {\bf 176}, 362 (1986).

10588: %%CITATION = PHLTA,B176,362;%%

10589: 

10590: \bibitem{GINZIN}

10591: P.~Ginsparg and J.~Zinn-Justin,

10592: %``2-D Gravity + 1-D Matter,''

10593: Phys.\ Lett.\ B {\bf 240}, 333 (1990).

10594: %%CITATION = PHLTA,B240,333;%%

10595: 

10596: \bibitem{0309164}

10597: S.~Giusto and C.~Imbimbo,

10598: %``Physical states at the tachyonic vacuum of open string field theory,''

10599: Nucl.\ Phys.\ B {\bf 677}, 52 (2004)

10600: [arXiv:hep-th/0309164].

10601: %%CITATION = HEP-TH 0309164;%%

10602: 

10603: \bibitem{0106195}

10604: C.~Gomez and P.~Resco,

10605: %``Topics in string tachyon dynamics,''

10606: arXiv:hep-th/0106195.

10607: %%CITATION = HEP-TH 0106195;%%

10608: 

10609: \bibitem{0111169}

10610: C.~Gomez and J.~J.~Manjarin,

10611: %``Dyons, K-theory and M-theory,''

10612: arXiv:hep-th/0111169.

10613: %%CITATION = HEP-TH 0111169;%%

10614: 

10615: \bibitem{9811131}

10616: R.~Gopakumar and C.~Vafa,

10617: %``On the gauge theory/geometry correspondence,''

10618: Adv.\ Theor.\ Math.\ Phys.\  {\bf 3}, 1415 (1999)

10619: [arXiv:hep-th/9811131].

10620: %%CITATION = HEP-TH 9811131;%%

10621: 

10622: \bibitem{0003160}

10623: R.~Gopakumar, S.~Minwalla and A.~Strominger,

10624: %``Noncommutative solitons,''

10625: JHEP {\bf 0005}, 020 (2000)

10626: [arXiv:hep-th/0003160].

10627: %%CITATION = HEP-TH 0003160;%%

10628: 

10629: \bibitem{0007226}

10630: R.~Gopakumar, S.~Minwalla and A.~Strominger,

10631: %{\em Symmetry restoration and tachyon condensation in open string 

10632: %theory,}

10633: hep-th/0007226.

10634: %%CITATION = HEP-TH 0007226;%%

10635: 

10636: 

10637: \bibitem{0103256}

10638: R.~Gopakumar, M.~Headrick and M.~Spradlin,

10639: %``On noncommutative multi-solitons,''

10640: Commun.\ Math.\ Phys.\  {\bf 233}, 355 (2003)

10641: [arXiv:hep-th/0103256].

10642: %%CITATION = HEP-TH 0103256;%%

10643: 

10644: \bibitem{0308184}

10645: R.~Gopakumar,

10646: %``From free fields to AdS,''

10647: Phys.\ Rev.\ D {\bf 70}, 025009 (2004)

10648: [arXiv:hep-th/0308184].

10649: %%CITATION = HEP-TH 0308184;%%

10650: 

10651: \bibitem{0402063}

10652: R.~Gopakumar,

10653: %``From free fields to AdS. II,''

10654: Phys.\ Rev.\ D {\bf 70}, 025010 (2004)

10655: [arXiv:hep-th/0402063].

10656: %%CITATION = HEP-TH 0402063;%%

10657: 

10658: \bibitem{0409233}

10659: R.~Gopakumar,

10660: %``Free Field Theory as a String Theory?,''

10661: arXiv:hep-th/0409233.

10662: %%CITATION = HEP-TH 0409233;%%

10663: 

10664: \bibitem{9403040}

10665: M.~B.~Green,

10666: %``Point - like states for type 2b superstrings,''

10667: Phys.\ Lett.\ B {\bf 329}, 435 (1994)

10668: [arXiv:hep-th/9403040].

10669: %%CITATION = HEP-TH 9403040;%%

10670: 

10671: \bibitem{9604091}

10672: M.~B.~Green and M.~Gutperle,

10673: %``Light-cone supersymmetry and D-branes,''

10674: Nucl.\ Phys.\ B {\bf 476}, 484 (1996) [arXiv:hep-th/9604091].

10675: %%CITATION = HEP-TH 9604091;%%

10676: 

10677: \bibitem{0206025}

10678: G.~Grignani, M.~Laidlaw, M.~Orselli and G.~W.~Semenoff,

10679: %``O(D) invariant tachyon condensates in the 1/D 

10680: %expansion,''

10681: Phys.\ Lett.\ B {\bf 543}, 127 (2002)

10682: [arXiv:hep-th/0206025].

10683: %%CITATION = HEP-TH 0206025;%%

10684: 

10685: \bibitem{GROMIL}

10686: D.~J.~Gross and N.~Miljkovic,

10687: %``A Nonperturbative Solution Of D = 1 String Theory,''

10688: Phys.\ Lett.\ B {\bf 238}, 217 (1990).

10689: %%CITATION = PHLTA,B238,217;%%

10690: 

10691: \bibitem{GROSSKLEB}

10692: D.~J.~Gross and I.~R.~Klebanov,

10693: %``Fermionic String Field Theory Of C = 1 Two-Dimensional Quantum

10694: %Gravity,''

10695: Nucl.\ Phys.\ B {\bf 352}, 671 (1991).

10696: %%CITATION = NUPHA,B352,671;%%

10697: 

10698: \bibitem{0105059}

10699: D.~J.~Gross and W.~Taylor,

10700: %``Split string field theory. I,''

10701: JHEP {\bf 0108}, 009 (2001)

10702: [arXiv:hep-th/0105059].

10703: %%CITATION = HEP-TH 0105059;%%

10704: 

10705: \bibitem{0106036}

10706: D.~J.~Gross and W.~Taylor,

10707: %``Split string field theory. II,''

10708: JHEP {\bf 0108}, 010 (2001)

10709: [arXiv:hep-th/0106036].

10710: %%CITATION = HEP-TH 0106036;%%

10711: 

10712: \bibitem{9802109}

10713: S.~S.~Gubser, I.~R.~Klebanov and A.~M.~Polyakov,

10714: %``Gauge theory correlators from non-critical string theory,''

10715: Phys.\ Lett.\ B {\bf 428}, 105 (1998)

10716: [arXiv:hep-th/9802109].

10717: %%CITATION = HEP-TH 9802109;%%

10718: 

10719: \bibitem{9901042}

10720: S.~Gukov,

10721: %``K-theory, reality, and orientifolds,''

10722: Commun.\ Math.\ Phys.\  {\bf 210}, 621 (2000)

10723: [arXiv:hep-th/9901042].

10724: %%CITATION = HEP-TH 9901042;%%

10725: 

10726: \bibitem{0101001}

10727: Z.~Guralnik and S.~Ramgoolam,

10728: %``On the polarization of unstable D0-branes into 

10729: %non-commutative odd

10730: %spheres,''

10731: JHEP {\bf 0102}, 032 (2001)

10732: [arXiv:hep-th/0101001].

10733: %%CITATION = HEP-TH 0101001;%%

10734: 

10735: \bibitem{9807064}

10736: H.~Gustafsson and U.~Lindstrom,

10737: %``A picture of D-branes at strong coupling. II: Spinning partons,''

10738: Phys.\ Lett.\  {\bf B440}, 43 (1998)

10739: [hep-th/9807064].

10740: %%CITATION = HEP-TH 9807064;%%

10741: 

10742: \bibitem{0202210}

10743: M.~Gutperle and A.~Strominger,

10744: %``Spacelike branes,''

10745: JHEP {\bf 0204}, 018 (2002)

10746: [arXiv:hep-th/0202210].

10747: %%CITATION = HEP-TH 0202210;%%

10748: 

10749: \bibitem{0301038}

10750: M.~Gutperle and A.~Strominger,

10751: %``Timelike boundary Liouville theory,''

10752: Phys.\ Rev.\ D {\bf 67}, 126002 (2003)

10753: [arXiv:hep-th/0301038].

10754: %%CITATION = HEP-TH 0301038;%%

10755: 

10756: \bibitem{0308047}

10757: M.~Gutperle and P.~Kraus,

10758: %``D-brane dynamics in the c = 1 matrix model,''

10759: Phys.\ Rev.\ D {\bf 69}, 066005 (2004)

10760: [arXiv:hep-th/0308047].

10761: %%CITATION = HEP-TH 0308047;%%

10762: 

10763: \bibitem{0407147}

10764: M.~Gutperle and W.~Sabra,

10765: %``S-brane solutions in gauged and ungauged supergravities,''

10766: arXiv:hep-th/0407147.

10767: %%CITATION = HEP-TH 0407147;%%

10768: 

10769: \bibitem{0409050}

10770: M.~Gutperle and P.~Yi,

10771: %``Winding strings and decay of D-branes with flux,''

10772: arXiv:hep-th/0409050.

10773: %%CITATION = HEP-TH 0409050;%%

10774: 

10775: \bibitem{0001143}

10776: J.~A.~Harvey, P.~Horava and P.~Kraus,

10777: %``D-sphalerons and the topology of string configuration 

10778: %space,''

10779: JHEP {\bf 0003}, 021 (2000)

10780: [arXiv:hep-th/0001143].

10781: %%CITATION = HEP-TH 0001143;%%

10782: 

10783: \bibitem{0002117}

10784: J.A.\ Harvey and P.\ Kraus, 

10785: %{\em

10786: %D-Branes as unstable lumps in bosonic open string field theory},

10787: JHEP {\bf 0004}, 012 (2000) [hep-th/0002117].

10788: %%CITATION = HEP-TH 0002117;%%

10789: 

10790: \bibitem{0003101}

10791: J.~A.~Harvey, D.~Kutasov and E.~J.~Martinec,

10792: %``On the relevance of tachyons,''

10793: arXiv:hep-th/0003101.

10794: %%CITATION = HEP-TH 0003101;%%

10795: 

10796: \bibitem{0005031}

10797: J.~A.~Harvey, P.~Kraus, F.~Larsen and E.~J.~Martinec,

10798: %``D-branes and strings as non-commutative solitons,''

10799: JHEP {\bf 0007}, 042 (2000) [arXiv:hep-th/0005031].

10800: %%CITATION = HEP-TH 0005031;%%

10801: 

10802: \bibitem{0008064}

10803: J.~A.~Harvey, P.~Kraus and F.~Larsen,

10804: %``Tensionless branes and discrete gauge symmetry,''

10805: Phys.\ Rev.\ D {\bf 63}, 026002 (2001)

10806: [arXiv:hep-th/0008064].

10807: %%CITATION = HEP-TH 0008064;%%

10808: 

10809: \bibitem{0009030}

10810: J.~A.~Harvey and G.~W.~Moore,

10811: %``Noncommutative tachyons and K-theory,''

10812: J.\ Math.\ Phys.\  {\bf 42}, 2765 (2001)

10813: [arXiv:hep-th/0009030].

10814: %%CITATION = HEP-TH 0009030;%%

10815: 

10816: \bibitem{0010060}

10817: J.~A.~Harvey, P.~Kraus and F.~Larsen,

10818: %{\em Exact noncommutative solitons,}

10819: hep-th/0010060.

10820: %%CITATION = HEP-TH 0010060;%%

10821: 

10822: \bibitem{0102076}

10823: J.~A.~Harvey,

10824: %``Komaba lectures on noncommutative solitons and D-branes,''

10825: arXiv:hep-th/0102076.

10826: %%CITATION = HEP-TH 0102076;%%

10827: 

10828: \bibitem{9703217}

10829: A.~Hashimoto and W.~I.~Taylor,

10830: %``Fluctuation spectra of tilted and intersecting D-branes from the  

10831: %Born-Infeld

10832: %action,''

10833: Nucl.\ Phys.\ B {\bf 503}, 193 (1997)

10834: [arXiv:hep-th/9703217].

10835: %%CITATION = HEP-TH 9703217;%%

10836: 

10837: \bibitem{0111092}

10838: A.~Hashimoto and N.~Itzhaki,

10839: %``Observables of string field theory,''

10840: JHEP {\bf 0201}, 028 (2002)

10841: [arXiv:hep-th/0111092].

10842: %%CITATION = HEP-TH 0111092;%%

10843: 

10844: \bibitem{0101145}

10845: K.~Hashimoto and K.~Krasnov,

10846: %``D-brane solutions in non-commutative gauge theory on fuzzy sphere,''

10847: Phys.\ Rev.\ D {\bf 64}, 046007 (2001)

10848: [arXiv:hep-th/0101145].

10849: %%CITATION = HEP-TH 0101145;%%

10850: 

10851: \bibitem{0102174}

10852: K.~Hashimoto and S.~Hirano,

10853: %``Metamorphosis of tachyon profile in unstable D9-branes,''

10854: Phys.\ Rev.\ D {\bf 65}, 026006 (2002) [arXiv:hep-th/0102174].

10855: %%CITATION = HEP-TH 0102174;%%

10856: 

10857: \bibitem{0202079}

10858: K.~Hashimoto and S.~Nagaoka,

10859: %``Realization of brane descent relations in effective theories,''

10860: Phys.\ Rev.\ D {\bf 66}, 026001 (2002) [arXiv:hep-th/0202079].

10861: %%CITATION = HEP-TH 0202079;%%

10862: 

10863: \bibitem{0204203}

10864: K.~Hashimoto,

10865: %``Dynamical decay of brane-antibrane and dielectric brane,''

10866: JHEP {\bf 0207}, 035 (2002)

10867: [arXiv:hep-th/0204203].

10868: %%CITATION = HEP-TH 0204203;%%

10869: 

10870: \bibitem{0209232}

10871: K.~Hashimoto and N.~Sakai,

10872: %``Brane - antibrane as a defect of tachyon condensation,''

10873: JHEP {\bf 0212}, 064 (2002)

10874: [arXiv:hep-th/0209232].

10875: %%CITATION = HEP-TH 0209232;%%

10876: 

10877: \bibitem{0211090}

10878: K.~Hashimoto, P.~M.~Ho and J.~E.~Wang,

10879: %``S-brane actions,''

10880: Phys.\ Rev.\ Lett.\  {\bf 90}, 141601 (2003)

10881: [arXiv:hep-th/0211090].

10882: %%CITATION = HEP-TH 0211090;%%

10883: 

10884: \bibitem{0303172}

10885: K.~Hashimoto, P.~M.~Ho, S.~Nagaoka and J.~E.~Wang,

10886: %``Time evolution via S-branes,''

10887: Phys.\ Rev.\ D {\bf 68}, 026007 (2003)

10888: [arXiv:hep-th/0303172].

10889: %%CITATION = HEP-TH 0303172;%%

10890: 

10891: \bibitem{0303204}

10892: K.~Hashimoto and S.~Nagaoka,

10893: %``Recombination of intersecting D-branes by local tachyon condensation,''

10894: JHEP {\bf 0306}, 034 (2003)

10895: [arXiv:hep-th/0303204].

10896: %%CITATION = HEP-TH 0303204;%%

10897: 

10898: \bibitem{0404237}

10899: K.~Hashimoto and S.~Terashima,

10900: %``Brane decay and death of open strings,''

10901: JHEP {\bf 0406}, 048 (2004)

10902: [arXiv:hep-th/0404237].

10903: %%CITATION = HEP-TH 0404237;%%

10904: 

10905: 

10906: \bibitem{0009105}

10907: H.~Hata and S.~Shinohara,

10908:  %``BRST invariance of the non-perturbative vacuum in bosonic open string

10909: %field

10910: %theory,''

10911: JHEP {\bf 0009}, 035 (2000) [arXiv:hep-th/0009105].

10912: %%CITATION = HEP-TH 0009105;%%

10913: 

10914: \bibitem{0101162}

10915: H.~Hata and S.~Teraguchi,

10916: %``Test of the absence of kinetic 

10917: %terms around the tachyon vacuum in cubic

10918: %string field theory,''

10919: JHEP {\bf 0105}, 045 (2001)

10920: [arXiv:hep-th/0101162].

10921: %%CITATION = HEP-TH 0101162;%%

10922: 

10923: \bibitem{0108150}

10924: H.~Hata and T.~Kawano,

10925: %``Open string states around a classical solution in vacuum string field

10926: %theory,''

10927: JHEP {\bf 0111}, 038 (2001) [arXiv:hep-th/0108150].

10928: %%CITATION = HEP-TH 0108150;%%

10929: 

10930: \bibitem{0111034}

10931: H.~Hata and S.~Moriyama,

10932: %``Observables as twist anomaly in vacuum string field theory,''

10933: JHEP {\bf 0201}, 042 (2002)

10934: [arXiv:hep-th/0111034].

10935: %%CITATION = HEP-TH 0111034;%%

10936: 

10937: \bibitem{0201177}

10938: H.~Hata, S.~Moriyama and S.~Teraguchi,

10939: %``Exact results on twist anomaly,''

10940: JHEP {\bf 0202}, 036 (2002)

10941: [arXiv:hep-th/0201177].

10942: %%CITATION = HEP-TH 0201177;%%

10943: 

10944: \bibitem{0206208}

10945: H.~Hata and S.~Moriyama,

10946: %``Reexamining classical solution 

10947: %and tachyon mode in vacuum string field

10948: %theory,''

10949: Nucl.\ Phys.\ B {\bf 651}, 3 (2003)

10950: [arXiv:hep-th/0206208].

10951: %%CITATION = HEP-TH 0206208;%%

10952: 

10953: \bibitem{0208067}

10954: H.~Hata and H.~Kogetsu,

10955: %``Higher level open string states from vacuum string field theory,''

10956: JHEP {\bf 0209}, 027 (2002)

10957: [arXiv:hep-th/0208067].

10958: %%CITATION = HEP-TH 0208067;%%

10959: 

10960: \bibitem{0305010}

10961: H.~Hata, H.~Kogetsu and S.~Teraguchi,

10962: %``Gauge structure of vacuum string field theory,''

10963: JHEP {\bf 0402}, 045 (2004)

10964: [arXiv:hep-th/0305010].

10965: %%CITATION = HEP-TH 0305010;%%

10966: 

10967: \bibitem{0405064}

10968: M.~Headrick, S.~Minwalla and T.~Takayanagi,

10969: %``Closed string tachyon condensation: An overview,''

10970: Class.\ Quant.\ Grav.\  {\bf 21}, S1539 (2004)

10971: [arXiv:hep-th/0405064].

10972: %%CITATION = HEP-TH 0405064;%%

10973: 

10974: \bibitem{0008023}

10975: Y.~Hikida, M.~Nozaki and T.~Takayanagi,

10976: %``Tachyon condensation on fuzzy sphere and noncommutative solitons,''

10977: Nucl.\ Phys.\ B {\bf 595}, 319 (2001)

10978: [arXiv:hep-th/0008023].

10979: %%CITATION = HEP-TH 0008023;%%

10980: 

10981: \bibitem{0101211}

10982: Y.~Hikida, M.~Nozaki and Y.~Sugawara,

10983: %``Formation of spherical D2-brane from multiple 

10984: %D0-branes,''

10985: Nucl.\ Phys.\ B {\bf 617}, 117 (2001)

10986: [arXiv:hep-th/0101211].

10987: %%CITATION = HEP-TH 0101211;%%

10988: 

10989: \bibitem{thooft}

10990: G.~'t Hooft,

10991: %``Magnetic Monopoles In Unified Gauge Theories,''

10992: Nucl.\ Phys.\ B {\bf 79}, 276 (1974).

10993: %%CITATION = NUPHA,B79,276;%%

10994: 

10995: \bibitem{9812135}

10996: P.~Horava,

10997: %``Type IIA D-branes, K-theory, and matrix theory,''

10998: Adv.\ Theor.\ Math.\ Phys.\  {\bf 2}, 1373 (1999)

10999: [arXiv:hep-th/9812135].

11000: %%CITATION = HEP-TH 9812135;%%

11001: 

11002: \bibitem{0211127}

11003: W.~H.~Huang,

11004: %``Brane-antibrane interaction under tachyon condensation,''

11005: Phys.\ Lett.\ B {\bf 561}, 153 (2003)

11006: [arXiv:hep-th/0211127].

11007: %%CITATION = HEP-TH 0211127;%%

11008: 

11009: \bibitem{0407081}

11010: W.~H.~Huang,

11011: %``Tachyon tube on non BPS D-branes,''

11012: JHEP {\bf 0408}, 060 (2004)

11013: [arXiv:hep-th/0407081].

11014: %%CITATION = HEP-TH 0407081;%%

11015: 

11016: \bibitem{0204031}

11017: Y.~Imamura,

11018: %``Gauge transformations on a D-brane in vacuum string 

11019: %field theory,''

11020: JHEP {\bf 0207}, 042 (2002)

11021: [arXiv:hep-th/0204031].

11022: %%CITATION = HEP-TH 0204031;%%

11023: 

11024: \bibitem{0004015}

11025: A.~Iqbal and A.~Naqvi,

11026: %``Tachyon condensation on a non-BPS D-brane,''

11027: arXiv:hep-th/0004015.

11028: %%CITATION = HEP-TH 0004015;%%

11029: 

11030: \bibitem{0008127}

11031: A.~Iqbal and A.~Naqvi,

11032: %``On marginal deformations in superstring field theory,''

11033: JHEP {\bf 0101}, 040 (2001)

11034: [arXiv:hep-th/0008127].

11035: %%CITATION = HEP-TH 0008127;%%

11036: 

11037: \bibitem{ishibashi}

11038: N.~Ishibashi,

11039: %``The Boundary And Crosscap States In Conformal Field Theories,''

11040: Mod.\ Phys.\ Lett.\ A {\bf 4}, 251 (1989).

11041: %%CITATION = MPLAE,A4,251;%%

11042: 

11043: \bibitem{0206102}

11044: A.~Ishida and S.~Uehara,

11045: %``Gauge fields on tachyon matter,''

11046: Phys.\ Lett.\ B {\bf 544}, 353 (2002)

11047: [arXiv:hep-th/0206102].

11048: %%CITATION = HEP-TH 0206102;%%

11049: 

11050: \bibitem{0301179}

11051: A.~Ishida and S.~Uehara,

11052: %``Rolling down to D-brane and tachyon matter,''

11053: JHEP {\bf 0302}, 050 (2003)

11054: [arXiv:hep-th/0301179].

11055: %%CITATION = HEP-TH 0301179;%%

11056: 

11057: \bibitem{0111151}

11058: J.~M.~Isidro,

11059: %``Twisted K-theory in g > 1 from D-branes,''

11060: J.\ Geom.\ Phys.\  {\bf 42}, 325 (2002)

11061: [arXiv:hep-th/0111151].

11062: %%CITATION = HEP-TH 0111151;%%

11063: 

11064: \bibitem{0007078}

11065: D.~P.~Jatkar, G.~Mandal and S.~R.~Wadia,

11066: %``Nielsen-Olesen vortices in noncommutative Abelian Higgs model,''

11067: JHEP {\bf 0009}, 018 (2000)

11068: [arXiv:hep-th/0007078].

11069: %%CITATION = HEP-TH 0007078;%%

11070: 

11071: \bibitem{0104229}

11072: D.~P.~Jatkar and R.~Vathsan,

11073: %``Stable solitons in field theory models for tachyon condensation,''

11074: JHEP {\bf 0106}, 039 (2001)

11075: [arXiv:hep-th/0104229].

11076: %%CITATION = HEP-TH 0104229;%%

11077: 

11078: \bibitem{0107075}

11079: D.~P.~Jatkar, S.~Sur and R.~Vathsan,

11080: %``A note on fermion and gauge couplings in 

11081: %field theory models for  tachyon

11082: %condensation,''

11083: Phys.\ Lett.\ B {\bf 520}, 391 (2001)

11084: [arXiv:hep-th/0107075].

11085: %%CITATION = HEP-TH 0107075;%%

11086: 

11087: \bibitem{9302106}

11088: A.~Jevicki,

11089: %``Fields and symmetries of 2-D strings,''

11090: arXiv:hep-th/9302106.

11091: %%CITATION = HEP-TH 9302106;%%

11092: 

11093: \bibitem{0007170}

11094: C.~V.~Johnson,

11095: %``D-brane primer,''

11096: arXiv:hep-th/0007170.

11097: %%CITATION = HEP-TH 0007170;%%

11098: 

11099: \bibitem{0408049}

11100: C.~V.~Johnson,

11101: %``Tachyon condensation, open-closed duality, resolvents, and minimal 

11102: %bosonic

11103: %and type 0 strings,''

11104: arXiv:hep-th/0408049.

11105: %%CITATION = HEP-TH 0408049;%%

11106: 

11107: \bibitem{0403050}

11108: G.~Jones, A.~Maloney and A.~Strominger,

11109: %``Non-singular solutions for S-branes,''

11110: Phys.\ Rev.\ D {\bf 69}, 126008 (2004)

11111: [arXiv:hep-th/0403050].

11112: %%CITATION = HEP-TH 0403050;%%

11113: 

11114: \bibitem{0211180}

11115: N.~T.~Jones and S.~H.~H.~Tye,

11116: %``An improved brane anti-brane action from 

11117: %boundary superstring field theory

11118: %and multi-vortex solutions,''

11119: JHEP {\bf 0301}, 012 (2003)

11120: [arXiv:hep-th/0211180].

11121: %%CITATION = HEP-TH 0211180;%%

11122: 

11123: \bibitem{0207041}

11124: G.~Jorjadze and G.~Weigt,

11125: %``Zero mode problem of Liouville field theory,''

11126: arXiv:hep-th/0207041.

11127: %%CITATION = HEP-TH 0207041;%%

11128: 

11129: \bibitem{0311202}

11130: G.~Jorjadze and G.~Weigt,

11131: %``Correlation functions and vertex operators of 

11132: %Liouville theory,''

11133: Phys.\ Lett.\ B {\bf 581}, 133 (2004)

11134: [arXiv:hep-th/0311202].

11135: %%CITATION = HEP-TH 0311202;%%

11136: 

11137: \bibitem{math-ph/0308034}

11138: L.~Joukovskaya and Y.~Volovich,

11139: %``Energy flow from open to closed strings in a toy model of rolling 

11140: %tachyon,''

11141: arXiv:math-ph/0308034.

11142: %%CITATION = MATH-PH 0308034;%%

11143: 

11144: \bibitem{0104143}

11145: H.~Kajiura, Y.~Matsuo and T.~Takayanagi,

11146: %``Exact tachyon condensation on noncommutative torus,''

11147: JHEP {\bf 0106}, 041 (2001)

11148: [arXiv:hep-th/0104143].

11149: %%CITATION = HEP-TH 0104143;%%

11150: 

11151: \bibitem{0003211}

11152: K.~Kamimura and J.~Simon,

11153: %``T-duality covariance of superD-branes,''

11154: Nucl.\ Phys.\ B {\bf 585}, 219 (2000)

11155: [arXiv:hep-th/0003211].

11156: %%CITATION = HEP-TH 0003211;%%

11157: 

11158: \bibitem{0210108}

11159: S.~Kar,

11160: %``A simple mechanical analog of the field theory of 

11161: %tachyon matter,''

11162: arXiv:hep-th/0210108.

11163: %%CITATION = HEP-TH 0210108;%%

11164: 

11165: \bibitem{0306132}

11166: J.~L.~Karczmarek, H.~Liu, J.~Maldacena and A.~Strominger,

11167: %``UV finite brane decay,''

11168: arXiv:hep-th/0306132.

11169: %%CITATION = HEP-TH 0306132;%%

11170: 

11171: \bibitem{0409249}

11172: F.~Katsumata, T.~Takahashi and S.~Zeze,

11173: %``Marginal deformations and closed string couplings in 

11174: %open string field

11175: %theory,''

11176: arXiv:hep-th/0409249.

11177: %%CITATION = HEP-TH 0409249;%%

11178: 

11179: \bibitem{0106103}

11180: H.~Kawai and T.~Kuroki,

11181: %``Strings as flux tube and deconfinement on branes in gauge theories,''

11182: Phys.\ Lett.\ B {\bf 518}, 294 (2001)

11183: [arXiv:hep-th/0106103].

11184: %%CITATION = HEP-TH 0106103;%%

11185: 

11186: \bibitem{9912274}

11187: T.~Kawano and T.~Takahashi,

11188: %{\em Open string field theory on noncommutative space,}

11189: hep-th/9912274.

11190: %%CITATION = HEP-TH 9912274;%%

11191: 

11192: \bibitem{9905195}

11193: C.~Kennedy and A.~Wilkins,

11194: %``Ramond-Ramond couplings on brane-antibrane systems,''

11195: Phys.\ Lett.\ B {\bf 464}, 206 (1999)

11196: [arXiv:hep-th/9905195].

11197: %%CITATION = HEP-TH 9905195;%%

11198: 

11199: \bibitem{0301076}

11200: C.~j.~Kim, H.~B.~Kim, Y.~b.~Kim and O.~K.~Kwon,

11201: %``Electromagnetic string fluid in rolling tachyon,''

11202: JHEP {\bf 0303}, 008 (2003) [arXiv:hep-th/0301076].

11203: %%CITATION = HEP-TH 0301076;%%

11204: 

11205: \bibitem{0304180}

11206: C.~j.~Kim, Y.~b.~Kim and C.~O.~Lee,

11207: %``Tachyon kinks,''

11208: JHEP {\bf 0305}, 020 (2003)

11209: [arXiv:hep-th/0304180].

11210: %%CITATION = HEP-TH 0304180;%%

11211: 

11212: \bibitem{0305092}

11213: C.~Kim, Y.~Kim, O.~K.~Kwon and C.~O.~Lee,

11214: %``Tachyon kinks on unstable Dp-branes,''

11215: JHEP {\bf 0311}, 034 (2003)

11216: [arXiv:hep-th/0305092].

11217: %%CITATION = HEP-TH 0305092;%%

11218: 

11219: \bibitem{0307184}

11220: C.~j.~Kim, Y.~b.~Kim, O.~K.~Kwon and P.~Yi,

11221: %``Tachyon tube and supertube,''

11222: JHEP {\bf 0309}, 042 (2003)

11223: [arXiv:hep-th/0307184].

11224: %%CITATION = HEP-TH 0307184;%%

11225: 

11226: \bibitem{0404163}

11227: C.~Kim, Y.~Kim and O.~K.~Kwon,

11228: %``Tubular D-branes in Salam-Sezgin model,''

11229: JHEP {\bf 0405}, 020 (2004)

11230: [arXiv:hep-th/0404163].

11231: %%CITATION = HEP-TH 0404163;%%

11232: 

11233: \bibitem{0204191}

11234: H.~Kim,

11235: %``Supergravity approach to tachyon potential in brane-antibrane 

11236: %systems,''

11237: JHEP {\bf 0301}, 080 (2003)

11238: [arXiv:hep-th/0204191].

11239: %%CITATION = HEP-TH 0204191;%%

11240: 

11241: \bibitem{0110124}

11242: I.~Kishimoto,

11243: %``Some properties of string field algebra,''

11244: JHEP {\bf 0112}, 007 (2001)

11245: [arXiv:hep-th/0110124].

11246: %%CITATION = HEP-TH 0110124;%%

11247: 

11248: \bibitem{0112169}

11249: I.~Kishimoto and K.~Ohmori,

11250: %``CFT description of identity string field: 

11251: %Toward derivation of the VSFT

11252: %action,''

11253: JHEP {\bf 0205}, 036 (2002)

11254: [arXiv:hep-th/0112169].

11255: %%CITATION = HEP-TH 0112169;%%

11256: 

11257: \bibitem{0205275}

11258: I.~Kishimoto and T.~Takahashi,

11259: %``Open string field theory around universal solutions,''

11260: Prog.\ Theor.\ Phys.\  {\bf 108}, 591 (2002)

11261: [arXiv:hep-th/0205275].

11262: %%CITATION = HEP-TH 0205275;%%

11263: 

11264: \bibitem{0012081}

11265: M.~Kleban, A.~Lawrence and S.~Shenker,

11266: %{\em Closed strings from nothing,}

11267: hep-th/0012081.

11268: 

11269: \bibitem{9108019}

11270: I.~R.~Klebanov,

11271: %``String theory in two-dimensions,''

11272: arXiv:hep-th/9108019.

11273: %%CITATION = HEP-TH 9108019;%%

11274: 

11275: \bibitem{9210105}

11276: I.~R.~Klebanov and A.~Pasquinucci,

11277: %``Infinite asymmetry and Ward identities in two-dimensional string

11278: %theory,''

11279: arXiv:hep-th/9210105.

11280: %%CITATION = HEP-TH 9210105;%%

11281: 

11282: \bibitem{0305159}

11283: I.~R.~Klebanov, J.~Maldacena and N.~Seiberg,

11284: %``D-brane decay in two-dimensional string theory,''

11285: JHEP {\bf 0307}, 045 (2003)

11286: [arXiv:hep-th/0305159].

11287: %%CITATION = HEP-TH 0305159;%%

11288: 

11289: \bibitem{0209186}

11290: A.~Kling, O.~Lechtenfeld, A.~D.~Popov and S.~Uhlmann,

11291: %``On nonperturbative solutions of superstring field 

11292: %theory,''

11293: Phys.\ Lett.\ B {\bf 551}, 193 (2003)

11294: [arXiv:hep-th/0209186].

11295: %%CITATION = HEP-TH 0209186;%%

11296: 

11297: \bibitem{0212335}

11298: A.~Kling, O.~Lechtenfeld, A.~D.~Popov and S.~Uhlmann,

11299: %``Solving string field equations: New uses for old 

11300: %tools,''

11301: Fortsch.\ Phys.\  {\bf 51}, 775 (2003)

11302: [arXiv:hep-th/0212335].

11303: %%CITATION = HEP-TH 0212335;%%

11304: 

11305: \bibitem{0004106}

11306: J.~Kluson,

11307: %``Proposal for non-BPS D-brane action,''

11308: Phys.\ Rev.\ D {\bf 62}, 126003 (2000) [arXiv:hep-th/0004106].

11309: %%CITATION = HEP-TH 0004106;%%

11310: 

11311: \bibitem{0009189}

11312: J.~Kluson,

11313: %``D-branes from N non-BPS D0-branes,''

11314: JHEP {\bf 0011}, 016 (2000)

11315: [arXiv:hep-th/0009189].

11316: %%CITATION = HEP-TH 0009189;%%

11317: 

11318: \bibitem{0102063}

11319: J.~Kluson,

11320: %``D-branes from N D(-1)-branes in bosonic and type IIA 

11321: %string theory,''

11322: JHEP {\bf 0103}, 018 (2001)

11323: [arXiv:hep-th/0102063].

11324: %%CITATION = HEP-TH 0102063;%%

11325: 

11326: \bibitem{0103079}

11327: J.~Kluson,

11328: %``N D0-branes and anti-D0-branes in the background 

11329: %independent string  field

11330: %theory,''

11331: Phys.\ Rev.\ D {\bf 64}, 126006 (2001)

11332: [arXiv:hep-th/0103079].

11333: %%CITATION = HEP-TH 0103079;%%

11334: 

11335: \bibitem{0208028}

11336: J.~Kluson,

11337: %``Time dependent solution in open bosonic string field theory,''

11338: arXiv:hep-th/0208028.

11339: %%CITATION = HEP-TH 0208028;%%

11340: 

11341: \bibitem{0310066}

11342: J.~Kluson,

11343: %``Note on D-brane effective action in the linear dilaton background,''

11344: JHEP {\bf 0311}, 068 (2003)

11345: [arXiv:hep-th/0310066].

11346: %%CITATION = HEP-TH 0310066;%%

11347: 

11348: \bibitem{0312086}

11349: J.~Kluson,

11350: %``D-brane effective actions and particle production near the beginning of 

11351: %the

11352: %tachyon condensation,''

11353: JHEP {\bf 0401}, 019 (2004)

11354: [arXiv:hep-th/0312086].

11355: %%CITATION = HEP-TH 0312086;%%

11356: 

11357: \bibitem{0401236}

11358: J.~Kluson,

11359: %``Bosonic D-brane effective action in linear dilaton background,''

11360: JHEP {\bf 0402}, 024 (2004)

11361: [arXiv:hep-th/0401236].

11362: %%CITATION = HEP-TH 0401236;%%

11363: 

11364: \bibitem{0403124}

11365: J.~Kluson,

11366: %``Proposal for the open string tachyon effective action in the linear 

11367: %dilaton

11368: %background,''

11369: JHEP {\bf 0406}, 021 (2004)

11370: [arXiv:hep-th/0403124].

11371: %%CITATION = HEP-TH 0403124;%%

11372: 

11373: \bibitem{KPZ}

11374: V.~G.~Knizhnik, A.~M.~Polyakov and A.~B.~Zamolodchikov,

11375: %``Fractal Structure Of 2d-Quantum Gravity,''

11376: Mod.\ Phys.\ Lett.\ A {\bf 3}, 819 (1988).

11377: %%CITATION = MPLAE,A3,819;%%

11378: 

11379: \bibitem{0409044}

11380: S.~Kobayashi, T.~Asakawa and S.~Matsuura,

11381: %``Open string tachyon in supergravity solution,''

11382: arXiv:hep-th/0409044.

11383: %%CITATION = HEP-TH 0409044;%%

11384: 

11385: \bibitem{0212055}

11386: A.~Koshelev,

11387: %``Solutions of vacuum superstring field theory,''

11388: arXiv:hep-th/0212055.

11389: %%CITATION = HEP-TH 0212055;%%

11390: 

11391: \bibitem{kost-sam1}

11392: V.~A.~Kostelecky and S.~Samuel,

11393: %``The Static Tachyon Potential In The Open Bosonic String Theory,''

11394: Phys.\ Lett.\ B {\bf 207}, 169 (1988).

11395: %%CITATION = PHLTA,B207,169;%%

11396: 

11397: \bibitem{kost-sam2}

11398: V.~A.~Kostelecky and S.~Samuel,

11399: %``On A Nonperturbative Vacuum For The Open Bosonic String,''

11400: Nucl.\ Phys.\ B {\bf 336}, 263 (1990).

11401: %%CITATION = NUPHA,B336,263;%%

11402: 

11403: \bibitem{kost-pot1}

11404: V.~A.~Kostelecky and R.~Potting,

11405:  %``Expectation Values, Lorentz Invariance, and CPT in the Open Bosonic

11406: %String,''

11407: Phys.\ Lett.\ B {\bf 381}, 89 (1996) [arXiv:hep-th/9605088].

11408: %%CITATION = HEP-TH 9605088;%%

11409: 

11410: \bibitem{0008252}

11411: V.~A.~Kostelecky and R.~Potting,

11412:  %``Analytical construction of a nonperturbative vacuum for the open

11413: %bosonic

11414: %string,''

11415: Phys.\ Rev.\ D {\bf 63}, 046007 (2001) [arXiv:hep-th/0008252].

11416: %%CITATION = HEP-TH 0008252;%%

11417: 

11418: \bibitem{0104090}

11419: T.~Krajewski and M.~Schnabl,

11420: %``Exact solitons on noncommutative tori,''

11421: JHEP {\bf 0108}, 002 (2001)

11422: [arXiv:hep-th/0104090].

11423: %%CITATION = HEP-TH 0104090;%%

11424: 

11425: \bibitem{0010016}

11426: P.~Kraus, A.~Rajaraman and S.~H.~Shenker,

11427: %``Tachyon condensation in noncommutative gauge theory,''

11428: Nucl.\ Phys.\ B {\bf 598}, 169 (2001)

11429: [arXiv:hep-th/0010016].

11430: %%CITATION = HEP-TH 0010016;%%

11431: 

11432: \bibitem{0012198}

11433: P.~Kraus and F.~Larsen,

11434: %``Boundary string field theory of the DD-bar system,''

11435: Phys.\ Rev.\ D {\bf 63}, 106004 (2001)

11436: [arXiv:hep-th/0012198].

11437: %%CITATION = HEP-TH 0012198;%%

11438: 

11439: \bibitem{0401003}

11440: K.~R.~Kristjansson and L.~Thorlacius,

11441: %``c = 1 boundary conformal field theory revisited,''

11442: Class.\ Quant.\ Grav.\  {\bf 21}, S1359 (2004)

11443: [arXiv:hep-th/0401003].

11444: %%CITATION = HEP-TH 0401003;%%

11445: 

11446: \bibitem{0204144}

11447: M.~Kruczenski, R.~C.~Myers and A.~W.~Peet,

11448: %``Supergravity S-branes,''

11449: JHEP {\bf 0205}, 039 (2002)

11450: [arXiv:hep-th/0204144].

11451: %%CITATION = HEP-TH 0204144;%%

11452: 

11453: \bibitem{0305229}

11454: O.~K.~Kwon and P.~Yi,

11455: %``String fluid, tachyon matter, and domain walls,''

11456: arXiv:hep-th/0305229.

11457: %%CITATION = HEP-TH 0305229;%%

11458: 

11459: \bibitem{0009148}

11460: D.~Kutasov, M.~Marino and G.~W.~Moore,

11461: %``Some exact results on tachyon condensation in string field theory,''

11462: JHEP {\bf 0010}, 045 (2000) [arXiv:hep-th/0009148].

11463: %%CITATION = HEP-TH 0009148;%%

11464: 

11465: \bibitem{0010108}

11466: D.~Kutasov, M.~Marino and G.~W.~Moore,

11467: %``Remarks on tachyon condensation in superstring field theory,''

11468: arXiv:hep-th/0010108.

11469: %%CITATION = HEP-TH 0010108;%%

11470: 

11471: \bibitem{0304045}

11472: D.~Kutasov and V.~Niarchos,

11473: %``Tachyon effective actions in open string theory,''

11474: Nucl.\ Phys.\ B {\bf 666}, 56 (2003)

11475: [arXiv:hep-th/0304045].

11476: %%CITATION = HEP-TH 0304045;%%

11477: 

11478: \bibitem{0405058}

11479: D.~Kutasov,

11480: %``D-brane dynamics near NS5-branes,''

11481: arXiv:hep-th/0405058.

11482: %%CITATION = HEP-TH 0405058;%%

11483: 

11484: \bibitem{0408073}

11485: D.~Kutasov,

11486: %``A geometric interpretation of the open string tachyon,''

11487: arXiv:hep-th/0408073.

11488: %%CITATION = HEP-TH 0408073;%%

11489: 

11490: \bibitem{0002061}

11491: N.~D.~Lambert and I.~Sachs,

11492: %``Non-abelian field theory of stable non-BPS branes,''

11493: JHEP {\bf 0003}, 028 (2000)

11494: [arXiv:hep-th/0002061].

11495: %%CITATION = HEP-TH 0002061;%%

11496: 

11497: \bibitem{0006122}

11498: N.~D.~Lambert and I.~Sachs,

11499: %``Non-Abelian Born-Infeld action and solitons for critical non-BPS  

11500: %branes,''

11501: JHEP {\bf 0008}, 024 (2000)

11502: [arXiv:hep-th/0006122].

11503: %%CITATION = HEP-TH 0006122;%%

11504: 

11505: \bibitem{0104218}

11506: N.~D.~Lambert and I.~Sachs,

11507: %``On higher derivative terms in tachyon effective actions,''

11508: JHEP {\bf 0106}, 060 (2001) [arXiv:hep-th/0104218].

11509: %%CITATION = HEP-TH 0104218;%%

11510: 

11511: \bibitem{0208217}

11512: N.~D.~Lambert and I.~Sachs,

11513: %``Tachyon dynamics and the effective action approximation,''

11514: Phys.\ Rev.\ D {\bf 67}, 026005 (2003) [arXiv:hep-th/0208217].

11515: %%CITATION = HEP-TH 0208217;%%

11516: 

11517: \bibitem{0303139}

11518: N.~Lambert, H.~Liu and J.~Maldacena,

11519: %``Closed strings from decaying D-branes,''

11520: arXiv:hep-th/0303139.

11521: %%CITATION = HEP-TH 0303139;%%

11522: 

11523: \bibitem{0010181}

11524: F.~Larsen,

11525: %``Fundamental strings as noncommutative solitons,''

11526: Int.\ J.\ Mod.\ Phys.\ A {\bf 16}, 650 (2001)

11527: [arXiv:hep-th/0010181].

11528: %%CITATION = HEP-TH 0010181;%%

11529: 

11530: \bibitem{0212248}

11531: F.~Larsen, A.~Naqvi and S.~Terashima,

11532: %``Rolling tachyons and decaying branes,''

11533: JHEP {\bf 0302}, 039 (2003)

11534: [arXiv:hep-th/0212248].

11535: %%CITATION = HEP-TH 0212248;%%

11536: 

11537: \bibitem{0303035}

11538: F.~Leblond and A.~W.~Peet,

11539: %``SD-brane gravity fields and rolling tachyons,''

11540: JHEP {\bf 0304}, 048 (2003) [arXiv:hep-th/0303035].

11541: %%CITATION = HEP-TH 0303035;%%

11542: 

11543: \bibitem{0204155}

11544: O.~Lechtenfeld, A.~D.~Popov and S.~Uhlmann,

11545: %``Exact solutions of Berkovits' string field theory,''

11546: Nucl.\ Phys.\ B {\bf 637}, 119 (2002)

11547: [arXiv:hep-th/0204155].

11548: %%CITATION = HEP-TH 0204155;%%

11549: 

11550: \bibitem{0210221}

11551: H.~w.~Lee and W.~S.~l'Yi,

11552: %``Time evolution of rolling tachyons for a brane-antibrane 

11553: %pair,''

11554: J.\ Korean Phys.\ Soc.\  {\bf 43}, 676 (2003)

11555: [arXiv:hep-th/0210221].

11556: %%CITATION = HEP-TH 0210221;%%

11557: 

11558: \bibitem{0105115}

11559: T.~Lee,

11560: %``Tachyon condensation, boundary state and 

11561: %noncommutative solitons,''

11562: Phys.\ Rev.\ D {\bf 64}, 106004 (2001)

11563: [arXiv:hep-th/0105115].

11564: %%CITATION = HEP-TH 0105115;%%

11565: 

11566: \bibitem{0105264}

11567: T.~Lee,

11568: %``Tachyon condensation and open string field theory,''

11569: Phys.\ Lett.\ B {\bf 520}, 385 (2001)

11570: [arXiv:hep-th/0105264].

11571: %%CITATION = HEP-TH 0105264;%%

11572: 

11573: \bibitem{0109032}

11574: T.~Lee, K.~S.~Viswanathan and Y.~Yang,

11575: %``Boundary string field theory at one-loop,''

11576: J.\ Korean Phys.\ Soc.\  {\bf 42}, 34 (2003)

11577: [arXiv:hep-th/0109032].

11578: %%CITATION = HEP-TH 0109032;%%

11579: 

11580: \bibitem{leigh}

11581: R. Leigh, Mod. Phys. Lett. {\bf A4} (1989) 2767.

11582: %%CITATION = MPLAE,A4,2767;%%

11583: 

11584: \bibitem{9905006}

11585: A.~Lerda and R.~Russo,

11586: %``Stable non-BPS states in string theory: A pedagogical review,''

11587: Int.\ J.\ Mod.\ Phys.\ A {\bf 15}, 771 (2000)

11588: [arXiv:hep-th/9905006].

11589: %%CITATION = HEP-TH 9905006;%%

11590: 

11591: \bibitem{9303067}

11592: K.\ Li and E.\ Witten,

11593: %{\em

11594: %Role of short distance behavior in off-shell open string field

11595: %theory},

11596: Phys.\ Rev.\  {\bf D48}, 853 (1993)

11597: [hep-th/9303067].

11598: %%CITATION = HEP-TH 9303067;%%

11599: 

11600: \bibitem{0010058}

11601: M.~Li,

11602: %``Note on noncommutative tachyon in matrix models,''

11603: Nucl.\ Phys.\ B {\bf 602}, 201 (2001)

11604: [arXiv:hep-th/0010058].

11605: %%CITATION = HEP-TH 0010058;%%

11606: 

11607: \bibitem{LIAN}

11608: B.~H.~Lian and G.~J.~Zuckerman,

11609: Phys.\ Lett.\ B {\bf 254}, 417 (1991);

11610: %%CITATION = PHLTA,B254,417;%%

11611: %``2-D Gravity With C = 1 Matter,''

11612: Phys.\ Lett.\ B {\bf 266}, 21 (1991).

11613: %%CITATION = PHLTA,B266,21;%%

11614: 

11615: \bibitem{9604156}

11616: G.~Lifschytz,

11617: %``Comparing D-branes to Black-branes,''

11618: Phys.\ Lett.\ B {\bf 388}, 720 (1996)

11619: [arXiv:hep-th/9604156].

11620: %%CITATION = HEP-TH 9604156;%%

11621: 

11622: \bibitem{9704051}

11623: U.~Lindstrom and R.~von Unge,

11624: %``A picture of D-branes at strong coupling,''

11625: Phys.\ Lett.\ B {\bf 403}, 233 (1997) [arXiv:hep-th/9704051].

11626: %%CITATION = HEP-TH 9704051;%%

11627: 

11628: \bibitem{9910159}

11629: U.~Lindstrom, M.~Zabzine and A.~Zheltukhin,

11630: %``Limits of the D-brane action,''

11631: JHEP {\bf 9912}, 016 (1999)

11632: [hep-th/9910159].

11633: %%CITATION = HEP-TH 9910159;%%

11634: 

11635: \bibitem{0101213}

11636: U.~Lindstrom and M.~Zabzine,

11637: %``Strings at the tachyonic vacuum,''

11638: JHEP {\bf 0103}, 014 (2001)

11639: [arXiv:hep-th/0101213].

11640: %%CITATION = HEP-TH 0101213;%%

11641: 

11642: \bibitem{0403147}

11643: J.~X.~Lu and S.~Roy,

11644: %``Supergravity approach to tachyon condensation on the brane-antibrane

11645: %system,''

11646: Phys.\ Lett.\ B {\bf 599}, 313 (2004)

11647: [arXiv:hep-th/0403147].

11648: %%CITATION = HEP-TH 0403147;%%

11649: 

11650: \bibitem{0409019}

11651: J.~X.~Lu and S.~Roy,

11652: %``Delocalized, non-SUSY p-branes, tachyon condensation and tachyon 

11653: %matter,''

11654: arXiv:hep-th/0409019.

11655: %%CITATION = HEP-TH 0409019;%%

11656: 

11657: \bibitem{0003124}

11658: J.~Majumder and A.~Sen,

11659: %``Vortex pair creation on brane-antibrane pair via marginal

11660: %deformation,''

11661: JHEP {\bf 0006}, 010 (2000) [arXiv:hep-th/0003124].

11662: %%CITATION = HEP-TH 0003124;%%

11663: 

11664: \bibitem{9711200}

11665: J.~M.~Maldacena,

11666: %``The large N limit of superconformal field theories and supergravity,''

11667: Adv.\ Theor.\ Math.\ Phys.\  {\bf 2}, 231 (1998)

11668: [Int.\ J.\ Theor.\ Phys.\  {\bf 38}, 1113 (1999)]

11669: [arXiv:hep-th/9711200].

11670: %%CITATION = HEP-TH 9711200;%%

11671: 

11672: \bibitem{0108100}

11673: J.~M.~Maldacena, G.~W.~Moore and N.~Seiberg,

11674: %``D-brane instantons and K-theory charges,''

11675: JHEP {\bf 0111}, 062 (2001)

11676: [arXiv:hep-th/0108100].

11677: %%CITATION = HEP-TH 0108100;%%

11678: 

11679: \bibitem{0302146}

11680: A.~Maloney, A.~Strominger and X.~Yin,

11681: %``S-brane thermodynamics,''

11682: arXiv:hep-th/0302146.

11683: %%CITATION = HEP-TH 0302146;%%

11684: 

11685: \bibitem{0008214}

11686: G.~Mandal and S.~Rey,

11687: %{\em A note on D-branes of odd codimensions from noncommutative

11688: %tachyons,}

11689: Phys.\ Lett.\ {\bf B495}, 193 (2000)

11690: [hep-th/0008214].

11691: %%CITATION = HEP-TH 0008214;%%

11692: 

11693: \bibitem{0011094}

11694: G.~Mandal and S.~R.~Wadia,

11695: %``Matrix model, noncommutative gauge theory and the 

11696: %tachyon potential,''

11697: Nucl.\ Phys.\ B {\bf 599}, 137 (2001)

11698: [arXiv:hep-th/0011094].

11699: %%CITATION = HEP-TH 0011094;%%

11700: 

11701: \bibitem{0312192}

11702: G.~Mandal and S.~R.~Wadia,

11703: %``Rolling tachyon solution of two-dimensional string theory,''

11704: arXiv:hep-th/0312192.

11705: %%CITATION = HEP-TH 0312192;%%

11706: 

11707: \bibitem{MANDELSTAM}

11708: S.~Mandelstam,

11709: %``Soliton Operators For The Quantized Sine-Gordon Equation,''

11710: Phys.\ Rev.\ D {\bf 11}, 3026 (1975).

11711: %%CITATION = PHRVA,D11,3026;%%

11712: 

11713: \bibitem{0103089}

11714: M.~Marino,

11715: %``On the BV formulation of boundary superstring field theory,''

11716: JHEP {\bf 0106}, 059 (2001)

11717: [arXiv:hep-th/0103089].

11718: %%CITATION = HEP-TH 0103089;%%

11719: 

11720: \bibitem{0112231}

11721: M.~Marino and R.~Schiappa,

11722: %``Towards vacuum superstring field theory: The supersliver,''

11723: J.\ Math.\ Phys.\  {\bf 44}, 156 (2003)

11724: [arXiv:hep-th/0112231].

11725: %%CITATION = HEP-TH 0112231;%%

11726: 

11727: \bibitem{0101199}

11728: E.~J.~Martinec and G.~W.~Moore,

11729: %``Noncommutative solitons on orbifolds,''

11730: arXiv:hep-th/0101199.

11731: %%CITATION = HEP-TH 0101199;%%

11732: 

11733: \bibitem{0212059}

11734: E.~J.~Martinec and G.~W.~Moore,

11735: %``On decay of K-theory,''

11736: arXiv:hep-th/0212059.

11737: %%CITATION = HEP-TH 0212059;%%

11738: 

11739: \bibitem{0306295}

11740: L.~Martucci and P.~J.~Silva,

11741: %``Kinky D-branes and straight strings of open string 

11742: %tachyon effective

11743: %theory,''

11744: JHEP {\bf 0308}, 026 (2003)

11745: [arXiv:hep-th/0306295].

11746: %%CITATION = HEP-TH 0306295;%%

11747: 

11748: \bibitem{0203071}

11749: P.~Matlock, R.~C.~Rashkov, K.~S.~Viswanathan and Y.~Yang,

11750: %``Dp-brane tension from tachyons and B-field in vacuum 

11751: %string field  theory,''

11752: Phys.\ Rev.\ D {\bf 66}, 026004 (2002)

11753: [arXiv:hep-th/0203071].

11754: %%CITATION = HEP-TH 0203071;%%

11755: 

11756: \bibitem{0211286}

11757: P.~Matlock,

11758: %``Butterfly tachyons in vacuum string field theory,''

11759: Phys.\ Rev.\ D {\bf 67}, 086002 (2003)

11760: [arXiv:hep-th/0211286].

11761: %%CITATION = HEP-TH 0211286;%%

11762: 

11763: \bibitem{0009002}

11764: Y.~Matsuo,

11765: %``Topological charges of noncommutative soliton,''

11766: Phys.\ Lett.\ B {\bf 499}, 223 (2001)

11767: [arXiv:hep-th/0009002].

11768: %%CITATION = HEP-TH 0009002;%%

11769: 

11770: \bibitem{0107058}

11771: A.~Mazumdar, S.~Panda and A.~Perez-Lorenzana,

11772: %``Assisted inflation via tachyon condensation,''

11773: Nucl.\ Phys.\ B {\bf 614}, 101 (2001)

11774: [arXiv:hep-ph/0107058].

11775: %%CITATION = HEP-PH 0107058;%%

11776: 

11777: \bibitem{0304224}

11778: J.~McGreevy and H.~Verlinde,

11779: %``Strings from tachyons: The c = 1 matrix reloated,''

11780: arXiv:hep-th/0304224.

11781: %%CITATION = HEP-TH 0304224;%%

11782: 

11783: \bibitem{0305194}

11784: J.~McGreevy, J.~Teschner and H.~Verlinde,

11785: %``Classical and quantum D-branes in 2D string theory,''

11786: arXiv:hep-th/0305194.

11787: %%CITATION = HEP-TH 0305194;%%

11788: 

11789: \bibitem{0206212}

11790: T.~Mehen and B.~Wecht,

11791: %``Gauge fields and scalars in rolling tachyon 

11792: %backgrounds,''

11793: JHEP {\bf 0302}, 058 (2003)

11794: [arXiv:hep-th/0206212].

11795: %%CITATION = HEP-TH 0206212;%%

11796: 

11797: \bibitem{0105246}

11798: Y.~Michishita,

11799: %``Tachyon lump solutions of bosonic D-branes on SU(2) 

11800: %group manifolds in  cubic

11801: %string field theory,''

11802: Nucl.\ Phys.\ B {\bf 614}, 26 (2001)

11803: [arXiv:hep-th/0105246].

11804: %%CITATION = HEP-TH 0105246;%%

11805: 

11806: \bibitem{0011079}

11807: M.~Mihailescu, I.~Y.~Park and T.~A.~Tran,

11808: %``D-branes as solitons of an N = 1, D = 10 

11809: %non-commutative gauge theory,''

11810: Phys.\ Rev.\ D {\bf 64}, 046006 (2001)

11811: [arXiv:hep-th/0011079].

11812: %%CITATION = HEP-TH 0011079;%%

11813: 

11814: \bibitem{0008231}

11815: J.~A.~Minahan and B.~Zwiebach,

11816: %``Field theory models for tachyon and gauge field string dynamics,''

11817: JHEP {\bf 0009}, 029 (2000) [arXiv:hep-th/0008231].

11818: %%CITATION = HEP-TH 0008231;%%

11819: 

11820: \bibitem{0009246}

11821: J.~A.~Minahan and B.~Zwiebach,

11822: %{\em Effective tachyon dynamics in superstring theory,}

11823: hep-th/0009246.

11824: %%CITATION = HEP-TH 0009246;%%

11825: 

11826: \bibitem{0011226}

11827: J.~A.~Minahan and B.~Zwiebach,

11828: %``Gauge fields and fermions in tachyon effective field theories,''

11829: JHEP {\bf 0102}, 034 (2001) [arXiv:hep-th/0011226].

11830: %%CITATION = HEP-TH 0011226;%%

11831: 

11832: \bibitem{0102071}

11833: J.~A.~Minahan,

11834: %``Mode interactions of the tachyon condensate in p-adic string theory,''

11835: JHEP {\bf 0103}, 028 (2001)

11836: [arXiv:hep-th/0102071].

11837: %%CITATION = HEP-TH 0102071;%%

11838: 

11839: \bibitem{0105312}

11840: J.~A.~Minahan,

11841: %``Quantum corrections in p-adic string theory,''

11842: arXiv:hep-th/0105312.

11843: %%CITATION = HEP-TH 0105312;%%

11844: 

11845: \bibitem{0203108}

11846: J.~A.~Minahan,

11847: %``Stretched strings in tachyon condensation models,''

11848: JHEP {\bf 0205}, 024 (2002)

11849: [arXiv:hep-th/0203108].

11850: %%CITATION = HEP-TH 0203108;%%

11851: 

11852: \bibitem{0205098}

11853: J.~A.~Minahan,

11854: %``Rolling the tachyon in super BSFT,''

11855: JHEP {\bf 0207}, 030 (2002) [arXiv:hep-th/0205098].

11856: %%CITATION = HEP-TH 0205098;%%

11857: 

11858: \bibitem{UTTG-16-91}

11859: D.~Minic, J.~Polchinski and Z.~Yang,

11860: %``Translation Invariant Backgrounds In (1+1)-Dimensional String

11861: %Theory,''

11862: Nucl.\ Phys.\ B {\bf 369}, 324 (1992).

11863: %%CITATION = NUPHA,B369,324;%%

11864: 

11865: \bibitem{0002237}

11866: N.~Moeller and W.~Taylor,

11867: %``Level truncation and the tachyon in open bosonic string field theory,''

11868: Nucl.\ Phys.\ B {\bf 583}, 105 (2000) [arXiv:hep-th/0002237].

11869: %%CITATION = HEP-TH 0002237;%%

11870: 

11871: \bibitem{0005036}

11872: N.~Moeller, A.~Sen and B.~Zwiebach,

11873: %``D-branes as tachyon lumps in string field theory,''

11874: JHEP {\bf 0008}, 039 (2000) [arXiv:hep-th/0005036].

11875: %%CITATION = HEP-TH 0005036;%%

11876: 

11877: \bibitem{0008101}

11878: N.~Moeller,

11879:  %``Codimension two lump solutions in string field theory and tachyonic

11880: %theories,''

11881: arXiv:hep-th/0008101.

11882: %%CITATION = HEP-TH 0008101;%%

11883: 

11884: \bibitem{0207107}

11885: N.~Moeller and B.~Zwiebach,

11886: %``Dynamics with infinitely many time derivatives and rolling tachyons,''

11887: arXiv:hep-th/0207107.

11888: %%CITATION = HEP-TH 0207107;%%

11889: 

11890: \bibitem{0304213}

11891: N.~Moeller and M.~Schnabl,

11892: %``Tachyon condensation in open-closed p-adic string theory,''

11893: JHEP {\bf 0401}, 011 (2004)

11894: [arXiv:hep-th/0304213].

11895: %%CITATION = HEP-TH 0304213;%%

11896: 

11897: \bibitem{MOORESEI}

11898: G.~W.~Moore and N.~Seiberg,

11899: %``From loops to fields in 2-D quantum gravity,''

11900: Int.\ J.\ Mod.\ Phys.\ A {\bf 7}, 2601 (1992).

11901: %%CITATION = IMPAE,A7,2601;%%

11902: 

11903: \bibitem{9912279}

11904: G.~W.~Moore and E.~Witten,

11905: %``Self-duality, Ramond-Ramond fields, and K-theory,''

11906: JHEP {\bf 0005}, 032 (2000)

11907: [arXiv:hep-th/9912279].

11908: %%CITATION = HEP-TH 9912279;%%

11909: 

11910: \bibitem{0111069}

11911: G.~Moore and W.~Taylor,

11912: %``The singular geometry of the sliver,''

11913: JHEP {\bf 0201}, 004 (2002)

11914: [arXiv:hep-th/0111069].

11915: %%CITATION = HEP-TH 0111069;%%

11916: 

11917: \bibitem{0304018}

11918: G.~W.~Moore,

11919: %``K-theory from a physical perspective,''

11920: arXiv:hep-th/0304018.

11921: %%CITATION = HEP-TH 0304018;%%

11922: 

11923: \bibitem{0011002}

11924: S.~Moriyama and S.~Nakamura,

11925: %``Descent relation of tachyon condensation from boundary 

11926: %string field

11927: %theory,''

11928: Phys.\ Lett.\ B {\bf 506}, 161 (2001)

11929: [arXiv:hep-th/0011002].

11930: %%CITATION = HEP-TH 0011002;%%

11931: 

11932: \bibitem{0001066}

11933: S.~Mukhi, N.~V.~Suryanarayana and D.~Tong,

11934: %``Brane-antibrane constructions,''

11935: JHEP {\bf 0003}, 015 (2000)

11936: [arXiv:hep-th/0001066].

11937: %%CITATION = HEP-TH 0001066;%%

11938: 

11939: \bibitem{0009101}

11940: S.~Mukhi and N.~V.~Suryanarayana,

11941: %``Chern-Simons terms on noncommutative branes,''

11942: JHEP {\bf 0011}, 006 (2000)

11943: [arXiv:hep-th/0009101].

11944: %%CITATION = HEP-TH 0009101;%%

11945: 

11946: \bibitem{0107087}

11947: S.~Mukhi and N.~V.~Suryanarayana,

11948: %``Ramond-Ramond couplings of noncommutative branes,''

11949: arXiv:hep-th/0107087.

11950: %%CITATION = HEP-TH 0107087;%%

11951: 

11952: \bibitem{0101014}

11953: P.~Mukhopadhyay and A.~Sen,

11954: %``Test of Siegel gauge for the lump solution,''

11955: JHEP {\bf 0102}, 017 (2001)

11956: [arXiv:hep-th/0101014].

11957: %%CITATION = HEP-TH 0101014;%%

11958: 

11959: \bibitem{0110136}

11960: P.~Mukhopadhyay,

11961: %``Oscillator representation of the BCFT construction of 

11962: %D-branes in  vacuum

11963: %string field theory,''

11964: JHEP {\bf 0112}, 025 (2001)

11965: [arXiv:hep-th/0110136].

11966: %%CITATION = HEP-TH 0110136;%%

11967: 

11968: \bibitem{0208142}

11969: P.~Mukhopadhyay and A.~Sen,

11970: %``Decay of unstable D-branes with electric field,''

11971: JHEP {\bf 0211}, 047 (2002)

11972: [arXiv:hep-th/0208142].

11973: %%CITATION = HEP-TH 0208142;%%

11974: 

11975: \bibitem{0208094}

11976: S.~Mukohyama,

11977: %``Inhomogeneous tachyon decay, light-cone structure and 

11978: %D-brane network

11979: %problem in tachyon cosmology,''

11980: Phys.\ Rev.\ D {\bf 66}, 123512 (2002)

11981: [arXiv:hep-th/0208094].

11982: %%CITATION = HEP-TH 0208094;%%

11983: 

11984: \bibitem{0309017}

11985: K.~Nagami,

11986: %``Closed string emission from unstable D-brane with background electric

11987: %field,''

11988: JHEP {\bf 0401}, 005 (2004)

11989: [arXiv:hep-th/0309017].

11990: %%CITATION = HEP-TH 0309017;%%

11991: 

11992: \bibitem{0312149}

11993: K.~Nagami,

11994: %``Rolling tachyon with electromagnetic field in linear 

11995: %dilaton background,''

11996: Phys.\ Lett.\ B {\bf 591}, 187 (2004)

11997: [arXiv:hep-th/0312149].

11998: %%CITATION = HEP-TH 0312149;%%

11999: 

12000: \bibitem{0005114}

12001: M.~Naka, T.~Takayanagi and T.~Uesugi,

12002: %``Boundary state description of tachyon condensation,''

12003: JHEP {\bf 0006}, 007 (2000)

12004: [arXiv:hep-th/0005114].

12005: %%CITATION = HEP-TH 0005114;%%

12006: 

12007: \bibitem{9402156}

12008: M.~Natsuume and J.~Polchinski,

12009: %``Gravitational Scattering In The C = 1 Matrix Model,''

12010: Nucl.\ Phys.\ B {\bf 424}, 137 (1994)

12011: [arXiv:hep-th/9402156].

12012: %%CITATION = HEP-TH 9402156;%%

12013: 

12014: \bibitem{0408157}

12015: D.~Nemeschansky and V.~Yasnov,

12016: %``Tachyon condensation at one loop,''

12017: arXiv:hep-th/0408157.

12018: %%CITATION = HEP-TH 0408157;%%

12019: 

12020: \bibitem{0401066}

12021: V.~Niarchos,

12022: %``Notes on tachyon effective actions and Veneziano amplitudes,''

12023: Phys.\ Rev.\ D {\bf 69}, 106009 (2004)

12024: [arXiv:hep-th/0401066].

12025: %%CITATION = HEP-TH 0401066;%%

12026: 

12027: \bibitem{0102085}

12028: K.~Ohmori,

12029: %``A review on tachyon condensation in open string field theories,''

12030: arXiv:hep-th/0102085.

12031: %%CITATION = HEP-TH 0102085;%%

12032: 

12033: \bibitem{0104230}

12034: K.~Ohmori,

12035: %``Tachyonic kink and lump-like solutions in superstring field theory,''

12036: JHEP {\bf 0105}, 035 (2001)

12037: [arXiv:hep-th/0104230].

12038: %%CITATION = HEP-TH 0104230;%%

12039: 

12040: \bibitem{0106068}

12041: K.~Ohmori,

12042: %``Survey of the tachyonic lump in bosonic string field theory,''

12043: JHEP {\bf 0108}, 011 (2001)

12044: [arXiv:hep-th/0106068].

12045: %%CITATION = HEP-TH 0106068;%%

12046: 

12047: \bibitem{0204138}

12048: K.~Ohmori,

12049: %``Comments on solutions of vacuum superstring field theory,''

12050: JHEP {\bf 0204}, 059 (2002)

12051: [arXiv:hep-th/0204138].

12052: %%CITATION = HEP-TH 0204138;%%

12053: 

12054: \bibitem{0208009}

12055: K.~Ohmori,

12056: %``On ghost structure of vacuum superstring field theory,''

12057: Nucl.\ Phys.\ B {\bf 648}, 94 (2003)

12058: [arXiv:hep-th/0208009].

12059: %%CITATION = HEP-TH 0208009;%%

12060: 

12061: \bibitem{0305103}

12062: K.~Ohmori,

12063: %``Level-expansion analysis in NS superstring field theory revisited,''

12064: arXiv:hep-th/0305103.

12065: %%CITATION = HEP-TH 0305103;%%

12066: 

12067: \bibitem{0306096}

12068: K.~Ohmori,

12069: %``Toward open-closed string theoretical description of rolling tachyon,''

12070: Phys.\ Rev.\ D {\bf 69}, 026008 (2004)

12071: [arXiv:hep-th/0306096].

12072: %%CITATION = HEP-TH 0306096;%%

12073: 

12074: \bibitem{0207004}

12075: K.~Ohta and T.~Yokono,

12076: %``Gravitational approach to tachyon matter,''

12077: Phys.\ Rev.\ D {\bf 66}, 125009 (2002)

12078: [arXiv:hep-th/0207004].

12079: %%CITATION = HEP-TH 0207004;%%

12080: 

12081: \bibitem{0301095}

12082: N.~Ohta,

12083: %``Intersection rules for S-branes,''

12084: Phys.\ Lett.\ B {\bf 558}, 213 (2003)

12085: [arXiv:hep-th/0301095].

12086: %%CITATION = HEP-TH 0301095;%%

12087: 

12088: \bibitem{0204012}

12089: Y.~Okawa,

12090: %``Open string states and D-brane tension from vacuum string field

12091: %theory,''

12092: JHEP {\bf 0207}, 003 (2002)

12093: [arXiv:hep-th/0204012].

12094: %%CITATION = HEP-TH 0204012;%%

12095: 

12096: \bibitem{0310264}

12097: Y.~Okawa,

12098:  %``Some exact computations on the twisted butterfly state in string field

12099: %theory,''

12100: arXiv:hep-th/0310264.

12101: %%CITATION = HEP-TH 0310264;%%

12102: 

12103: \bibitem{0311115}

12104: Y.~Okawa,

12105: %``Solving Witten's string field theory using the butterfly state,''

12106: arXiv:hep-th/0311115.

12107: %%CITATION = HEP-TH 0311115;%%

12108: 

12109: \bibitem{0201149}

12110: T.~Okuda,

12111: %``The equality of solutions in vacuum string field 

12112: %theory,''

12113: Nucl.\ Phys.\ B {\bf 641}, 393 (2002)

12114: [arXiv:hep-th/0201149].

12115: %%CITATION = HEP-TH 0201149;%%

12116: 

12117: \bibitem{0208196}

12118: T.~Okuda and S.~Sugimoto,

12119: %``Coupling of rolling tachyon to closed strings,''

12120: Nucl.\ Phys.\ B {\bf 647}, 101 (2002)

12121: [arXiv:hep-th/0208196].

12122: %%CITATION = HEP-TH 0208196;%%

12123: 

12124: \bibitem{0010028}

12125: K.~Okuyama,

12126: %``Noncommutative tachyon from background independent open string field

12127: %theory,''

12128: Phys.\ Lett.\ B {\bf 499}, 167 (2001)

12129: [arXiv:hep-th/0010028].

12130: %%CITATION = HEP-TH 0010028;%%

12131: 

12132: \bibitem{0111087}

12133: K.~Okuyama,

12134: %``Siegel gauge in vacuum string field theory,''

12135: JHEP {\bf 0201}, 043 (2002)

12136: [arXiv:hep-th/0111087].

12137: %%CITATION = HEP-TH 0111087;%%

12138: 

12139: \bibitem{0201015}

12140: K.~Okuyama,

12141: %``Ghost kinetic operator of vacuum string field theory,''

12142: JHEP {\bf 0201}, 027 (2002)

12143: [arXiv:hep-th/0201015].

12144: %%CITATION = HEP-TH 0201015;%%

12145: 

12146: \bibitem{0201136}

12147: K.~Okuyama,

12148: %``Ratio of tensions from vacuum string field theory,''

12149: JHEP {\bf 0203}, 050 (2002)

12150: [arXiv:hep-th/0201136].

12151: %%CITATION = HEP-TH 0201136;%%

12152: 

12153: \bibitem{0304108}

12154: K.~Okuyama,

12155: %``Wess-Zumino term in tachyon effective action,''

12156: JHEP {\bf 0305}, 005 (2003)

12157: [arXiv:hep-th/0304108].

12158: %%CITATION = HEP-TH 0304108;%%

12159: 

12160: \bibitem{0308172}

12161: K.~Okuyama,

12162: %``Comments on half S-branes,''

12163: JHEP {\bf 0309}, 053 (2003) [arXiv:hep-th/0308172].

12164: %%CITATION = HEP-TH 0308172;%%

12165: 

12166: \bibitem{9904153}

12167: K.~Olsen and R.~J.~Szabo,

12168: %``Brane descent relations in K-theory,''

12169: Nucl.\ Phys.\ B {\bf 566}, 562 (2000)

12170: [arXiv:hep-th/9904153].

12171: %%CITATION = HEP-TH 9904153;%%

12172: 

12173: \bibitem{9907140}

12174: K.~Olsen and R.~J.~Szabo,

12175: %``Constructing D-branes from K-theory,''

12176: Adv.\ Theor.\ Math.\ Phys.\  {\bf 3}, 889 (1999)

12177: [arXiv:hep-th/9907140].

12178: %%CITATION = HEP-TH 9907140;%%

12179: 

12180: \bibitem{0403283}

12181: A.~Parodi,

12182: %``K-theory in cutoff version of vacuum string field 

12183: %theory,''

12184: arXiv:hep-th/0403283.

12185: %%CITATION = HEP-TH 0403283;%%

12186: 

12187: \bibitem{0405125}

12188: K.~Peeters and M.~Zamaklar,

12189: %``AdS/CFT description of D-particle decay,''

12190: arXiv:hep-th/0405125.

12191: %%CITATION = HEP-TH 0405125;%%

12192: 

12193: \bibitem{9612215}

12194: V.~Periwal,

12195: %``Antibranes and crossing symmetry,''

12196: arXiv:hep-th/9612215.

12197: %%CITATION = HEP-TH 9612215;%%

12198: 

12199: \bibitem{0006223}

12200: V.~Periwal,

12201: %``D-brane charges and K-homology.  (Z)),''

12202: JHEP {\bf 0007}, 041 (2000)

12203: [arXiv:hep-th/0006223].

12204: %%CITATION = HEP-TH 0006223;%%

12205: 

12206: \bibitem{POLCH}

12207: J.~Polchinski,

12208: %``Classical Limit Of (1+1)-Dimensional String Theory,''

12209: Nucl.\ Phys.\ B {\bf 362}, 125 (1991).

12210: %%CITATION = NUPHA,B362,125;%%

12211: 

12212: \bibitem{9404008}

12213: J.~Polchinski and L.~Thorlacius,

12214: %``Free Fermion Representation Of A Boundary Conformal Field Theory,''

12215: Phys.\ Rev.\ D {\bf 50}, 622 (1994) [arXiv:hep-th/9404008].

12216: %%CITATION = HEP-TH 9404008;%%

12217: 

12218: \bibitem{9407031}

12219: J. Polchinski, Phys. Rev. {\bf D50} (1994) 6041 [hep-th/9407031].

12220: %%CITATION = HEP-TH 9407031;%%

12221: 

12222: \bibitem{9510017}

12223: J.~Polchinski,

12224: %``Dirichlet-Branes and Ramond-Ramond Charges,''

12225: Phys.\ Rev.\ Lett.\  {\bf 75}, 4724 (1995)

12226: [arXiv:hep-th/9510017].

12227: %%CITATION = HEP-TH 9510017;%%

12228: 

12229: \bibitem{9611050}

12230: J.~Polchinski,

12231: %``Lectures on D-branes,''

12232: arXiv:hep-th/9611050.

12233: %%CITATION = HEP-TH 9611050;%%

12234: 

12235: \bibitem{polbook}

12236: J.~Polchinski, String theory. Vol. 1 and 2, Cambridge Univ. Press,

12237: 1998.

12238: 

12239: \bibitem{polyakov}

12240: A.~M.~Polyakov,

12241: %``Particle Spectrum In Quantum Field Theory,''

12242: JETP Lett.\  {\bf 20}, 194 (1974)

12243: [Pisma Zh.\ Eksp.\ Teor.\ Fiz.\  {\bf 20}, 430 (1974)].

12244: %%CITATION = JTPLA,20,194;%%

12245: 

12246: \bibitem{poly1}

12247: A.~M.~Polyakov,

12248: %``Selftuning fields and resonant correlations in 2-d 

12249: %gravity,''

12250: Mod.\ Phys.\ Lett.\ A {\bf 6}, 635 (1991).

12251: %%CITATION = MPLAE,A6,635;%%

12252: 

12253: \bibitem{0204172}

12254: R.~Potting and J.~Raeymaekers,

12255: %``Ghost sector of vacuum string field theory and the projection 

12256: %equation,''

12257: JHEP {\bf 0206}, 002 (2002)

12258: [arXiv:hep-th/0204172].

12259: %%CITATION = HEP-TH 0204172;%%

12260: 

12261: \bibitem{pty}

12262: C.~R.~Preitschopf, C.~B.~Thorn and S.~A.~Yost,

12263: %{\em Superstring Field Theory,}

12264: Nucl.\ Phys.\  {\bf B337}, 363 (1990).

12265: %%CITATION = NUPHA,B337,363;%%

12266: 

12267: \bibitem{0112202}

12268: R.~Rashkov and K.~S.~Viswanathan,

12269: %``A note on the tachyon state in vacuum string field 

12270: %theory,''

12271: arXiv:hep-th/0112202.

12272: %%CITATION = HEP-TH 0112202;%%

12273: 

12274: \bibitem{0201229}

12275: R.~Rashkov and K.~S.~Viswanathan,

12276: %``A proposal for the vector state in vacuum string field 

12277: %theory,''

12278: arXiv:hep-th/0201229.

12279: %%CITATION = HEP-TH 0201229;%%

12280: 

12281: \bibitem{0006240}

12282: L.~Rastelli and B.~Zwiebach,

12283: %``Tachyon potentials, star products and universality,''

12284: JHEP {\bf 0109}, 038 (2001)

12285: [arXiv:hep-th/0006240].

12286: %%CITATION = HEP-TH 0006240;%%

12287: 

12288: \bibitem{0012251}

12289: L.~Rastelli, A.~Sen and B.~Zwiebach,

12290: %``String field theory around the tachyon vacuum,''

12291: Adv.\ Theor.\ Math.\ Phys.\  {\bf 5}, 353 (2002)

12292: [arXiv:hep-th/0012251].

12293: %%CITATION = HEP-TH 0012251;%%

12294: 

12295: \bibitem{0102112}

12296: L.~Rastelli, A.~Sen and B.~Zwiebach,

12297: %``Classical solutions in string field theory around the tachyon vacuum,''

12298: Adv.\ Theor.\ Math.\ Phys.\  {\bf 5}, 393 (2002)

12299: [arXiv:hep-th/0102112].

12300: %%CITATION = HEP-TH 0102112;%%

12301: 

12302: \bibitem{0105058}

12303: L.~Rastelli, A.~Sen and B.~Zwiebach,

12304:  %``Half strings, projectors, and multiple D-branes in vacuum string field

12305: %theory,''

12306: JHEP {\bf 0111}, 035 (2001)

12307: [arXiv:hep-th/0105058].

12308: %%CITATION = HEP-TH 0105058;%%

12309: 

12310: \bibitem{0105168}

12311: L.~Rastelli, A.~Sen and B.~Zwiebach,

12312: %``Boundary CFT construction of D-branes in vacuum string field theory,''

12313: JHEP {\bf 0111}, 045 (2001)

12314: [arXiv:hep-th/0105168].

12315: %%CITATION = HEP-TH 0105168;%%

12316: 

12317: \bibitem{0106010}

12318: L.~Rastelli, A.~Sen and B.~Zwiebach,

12319: %``Vacuum string field theory,''

12320: arXiv:hep-th/0106010.

12321: %%CITATION = HEP-TH 0106010;%%

12322: 

12323: \bibitem{0111153}

12324: L.~Rastelli, A.~Sen and B.~Zwiebach,

12325: %``A note on a proposal for the tachyon state in vacuum string field  

12326: %theory,''

12327: JHEP {\bf 0202}, 034 (2002)

12328: [arXiv:hep-th/0111153].

12329: %%CITATION = HEP-TH 0111153;%%

12330: 

12331: \bibitem{0111281}

12332: L.~Rastelli, A.~Sen and B.~Zwiebach,

12333: %``Star algebra spectroscopy,''

12334: JHEP {\bf 0203}, 029 (2002)

12335: [arXiv:hep-th/0111281].

12336: %%CITATION = HEP-TH 0111281;%%

12337: 

12338: \bibitem{9811237}

12339: A.~Recknagel and V.~Schomerus,

12340: %``Boundary deformation theory and moduli spaces of D-branes,''

12341: Nucl.\ Phys.\ B {\bf 545}, 233 (1999) [arXiv:hep-th/9811237].

12342: %%CITATION = HEP-TH 9811237;%%

12343: 

12344: \bibitem{9903139}

12345: A.~Recknagel and V.~Schomerus,

12346: %``Moduli spaces of D-branes in CFT-backgrounds,''

12347: Fortsch.\ Phys.\  {\bf 48}, 195 (2000)

12348: [arXiv:hep-th/9903139].

12349: %%CITATION = HEP-TH 9903139;%%

12350: 

12351: \bibitem{0003110}

12352: A.~Recknagel, D.~Roggenkamp and V.~Schomerus,

12353: %``On relevant boundary perturbations of unitary minimal 

12354: %models,''

12355: Nucl.\ Phys.\ B {\bf 588}, 552 (2000)

12356: [arXiv:hep-th/0003110].

12357: %%CITATION = HEP-TH 0003110;%%

12358: 

12359: \bibitem{0301049}

12360: S.~J.~Rey and S.~Sugimoto,

12361: %``Rolling tachyon with electric and magnetic fields: T-duality

12362: %approach,''

12363: Phys.\ Rev.\ D {\bf 67}, 086008 (2003)

12364: [arXiv:hep-th/0301049].

12365: 

12366: \bibitem{0303133}

12367: S.~J.~Rey and S.~Sugimoto,

12368: %``Rolling of modulated tachyon with gauge flux and emergent fundamental

12369: %string,''

12370: arXiv:hep-th/0303133.

12371: %%CITATION = HEP-TH 0303133;%%

12372: 

12373: \bibitem{0205198}

12374: S.~Roy,

12375: %``On supergravity solutions of space-like Dp-branes,''

12376: JHEP {\bf 0208}, 025 (2002)

12377: [arXiv:hep-th/0205198].

12378: %%CITATION = HEP-TH 0205198;%%

12379: 

12380: \bibitem{0010007}

12381: S.~Ryang,

12382: %``Gauge symmetry breaking and generalized monopole in 

12383: %non-BPS D-brane

12384: %action,''

12385: arXiv:hep-th/0010007.

12386: %%CITATION = HEP-TH 0010007;%%

12387: 

12388: \bibitem{0307078}

12389: %.~Sarangi and S.~H.~H.~Tye,

12390: %``Inter-brane potential and the decay of a non-BPS-D-brane to closed 

12391: %strings,''

12392: Phys.\ Lett.\ B {\bf 573}, 181 (2003)

12393: [arXiv:hep-th/0307078].

12394: %%CITATION = HEP-TH 0307078;%%

12395: 

12396: \bibitem{0010034}

12397: M.~Schnabl,

12398: %``String field theory at large B-field and 

12399: %noncommutative geometry,''

12400: JHEP {\bf 0011}, 031 (2000)

12401: [arXiv:hep-th/0010034].

12402: %%CITATION = HEP-TH 0010034;%%

12403: 

12404: \bibitem{0011238}

12405: M.~Schnabl,

12406: %{\em Constraints on the tachyon condensate from anomalous symmetries,}

12407: hep-th/0011238.

12408: %%CITATION = HEP-TH 0011238;%%

12409: 

12410: \bibitem{0201095}

12411: M.~Schnabl,

12412: %``Wedge states in string field theory,''

12413: JHEP {\bf 0301}, 004 (2003)

12414: [arXiv:hep-th/0201095].

12415: %%CITATION = HEP-TH 0201095;%%

12416: 

12417: \bibitem{0202139}

12418: M.~Schnabl,

12419: %``Anomalous reparametrizations and butterfly states in string field  

12420: %theory,''

12421: Nucl.\ Phys.\ B {\bf 649}, 101 (2003)

12422: [arXiv:hep-th/0202139].

12423: %%CITATION = HEP-TH 0202139;%%

12424: 

12425: \bibitem{9903205}

12426: V.~Schomerus,

12427: %``D-branes and deformation quantization,''

12428: JHEP {\bf 9906}, 030 (1999)

12429: [arXiv:hep-th/9903205].

12430: %%CITATION = HEP-TH 9903205;%%

12431: 

12432: \bibitem{0306026}

12433: V.~Schomerus,

12434: %``Rolling tachyons from Liouville theory,''

12435: arXiv:hep-th/0306026.

12436: %%CITATION = HEP-TH 0306026;%%

12437: 

12438: \bibitem{9908144}

12439: J.~H.~Schwarz,

12440: %``TASI lectures on non-BPS D-brane systems,''

12441: arXiv:hep-th/9908144.

12442: %%CITATION = HEP-TH 9908144;%%

12443: 

12444: \bibitem{9908091}

12445: J.~H.~Schwarz,

12446: %``Remarks on non-BPS D-branes,''

12447: Class.\ Quant.\ Grav.\  {\bf 17}, 1245 (2000)

12448: [arXiv:hep-th/9908091].

12449: %%CITATION = HEP-TH 9908091;%%

12450: 

12451: \bibitem{9908142}

12452: N.~Seiberg and E.~Witten,

12453: %``String theory and noncommutative geometry,''

12454: JHEP {\bf 9909}, 032 (1999)

12455: [arXiv:hep-th/9908142].

12456: %%CITATION = HEP-TH 9908142;%%

12457: 

12458: \bibitem{0008013}

12459: N.~Seiberg,

12460: %{\em A note on background independence

12461: %in noncommutative gauge theories,  matrix model and tachyon

12462: %condensation,}

12463: hep-th/0008013.

12464: %%CITATION = HEP-TH 0008013;%%

12465: 

12466: \bibitem{9805019}

12467: A.~Sen,

12468: %``Stable non-BPS bound states of BPS D-branes,''

12469: JHEP {\bf 9808}, 010 (1998)

12470: [arXiv:hep-th/9805019].

12471: %%CITATION = HEP-TH 9805019;%%

12472: 

12473: \bibitem{9805170}

12474: A.~Sen,

12475: %``Tachyon condensation on the brane antibrane system,''

12476: JHEP {\bf 9808}, 012 (1998) [arXiv:hep-th/9805170].

12477: %%CITATION = HEP-TH 9805170;%%

12478: 

12479: \bibitem{9808141}

12480: A.~Sen,

12481: %``SO(32) spinors of type I and other solitons on brane-antibrane pair,''

12482: JHEP {\bf 9809}, 023 (1998)

12483: [arXiv:hep-th/9808141].

12484: %%CITATION = HEP-TH 9808141;%%

12485: 

12486: \bibitem{9809111}

12487: A.~Sen,

12488: %``Type I D-particle and its interactions,''

12489: JHEP {\bf 9810}, 021 (1998)

12490: [arXiv:hep-th/9809111].

12491: %%CITATION = HEP-TH 9809111;%%

12492: 

12493: \bibitem{9812031}

12494: A.~Sen,

12495: %``BPS D-branes on non-supersymmetric cycles,''

12496: JHEP {\bf 9812}, 021 (1998) [arXiv:hep-th/9812031].

12497: %%CITATION = HEP-TH 9812031;%%

12498: 

12499: \bibitem{9902105}

12500: A.~Sen,

12501: %``Descent relations among bosonic D-branes,''

12502: Int.\ J.\ Mod.\ Phys.\ A {\bf 14}, 4061 (1999)

12503: [arXiv:hep-th/9902105].

12504: %%CITATION = HEP-TH 9902105;%%

12505: 

12506: \bibitem{9904207}

12507: A.~Sen,

12508: %``Non-BPS states and branes in string theory,''

12509: arXiv:hep-th/9904207.

12510: %%CITATION = HEP-TH 9904207;%%

12511: 

12512: \bibitem{9909062}

12513: A.~Sen,

12514: %``Supersymmetric world-volume action for non-BPS D-branes,''

12515: JHEP {\bf 9910}, 008 (1999) [arXiv:hep-th/9909062].

12516: %%CITATION = HEP-TH 9909062;%%

12517: 

12518: \bibitem{9911116}

12519: A.~Sen,

12520: %``Universality of the tachyon potential,''

12521: JHEP {\bf 9912}, 027 (1999) [arXiv:hep-th/9911116].

12522: %%CITATION = HEP-TH 9911116;%%

12523: 

12524: \bibitem{9912249}

12525: A.~Sen and B.~Zwiebach,

12526: %``Tachyon condensation in string field theory,''

12527: JHEP {\bf 0003}, 002 (2000) [arXiv:hep-th/9912249].

12528: %%CITATION = HEP-TH 9912249;%%

12529: 

12530: \bibitem{0007153}

12531: A.~Sen and B.~Zwiebach,

12532: %{\em Large marginal deformations in string field theory,}

12533: JHEP {\bf 0010}, 009 (2000)

12534: [hep-th/0007153].

12535: %%CITATION = HEP-TH 0007153;%%

12536: 

12537: \bibitem{0009038}

12538: A.~Sen,

12539: %``Some issues in non-commutative tachyon condensation,''

12540: JHEP {\bf 0011}, 035 (2000)

12541: [arXiv:hep-th/0009038].

12542: %%CITATION = HEP-TH 0009038;%%

12543: 

12544: \bibitem{0009090}

12545: A.~Sen,

12546: %``Uniqueness of tachyonic solitons,''

12547: JHEP {\bf 0012}, 001 (2000)

12548: [arXiv:hep-th/0009090].

12549: %%CITATION = HEP-TH 0009090;%%

12550: 

12551: \bibitem{0010240}

12552: A.~Sen,

12553: %``Fundamental strings in open string theory at the tachyonic vacuum,''

12554: J.\ Math.\ Phys.\  {\bf 42}, 2844 (2001) [arXiv:hep-th/0010240].

12555: %%CITATION = HEP-TH 0010240;%%

12556: 

12557: \bibitem{0203211}

12558: A.~Sen,

12559: %``Rolling tachyon,''

12560: JHEP {\bf 0204}, 048 (2002) [arXiv:hep-th/0203211].

12561: %%CITATION = HEP-TH 0203211;%%

12562: 

12563: \bibitem{0203265}

12564: A.~Sen,

12565: %``Tachyon matter,''

12566: JHEP {\bf 0207}, 065 (2002) [arXiv:hep-th/0203265].

12567: %%CITATION = HEP-TH 0203265;%%

12568: 

12569: \bibitem{0204143}

12570: A.~Sen,

12571: %``Field theory of tachyon matter,''

12572: Mod.\ Phys.\ Lett.\ A {\bf 17}, 1797 (2002)

12573: [arXiv:hep-th/0204143].

12574: %%CITATION = HEP-TH 0204143;%%

12575: 

12576: \bibitem{0207105}

12577: A.~Sen,

12578: %``Time evolution in open string theory,''

12579: JHEP {\bf 0210}, 003 (2002) [arXiv:hep-th/0207105].

12580: %%CITATION = HEP-TH 0207105;%%

12581: 

12582: \bibitem{0209122}

12583: A.~Sen,

12584: %``Time and tachyon,''

12585: Int.\ J.\ Mod.\ Phys.\ A {\bf 18}, 4869 (2003)

12586: [arXiv:hep-th/0209122].

12587: %%CITATION = HEP-TH 0209122;%%

12588: 

12589: \bibitem{0303057}

12590: A.~Sen,

12591: %``Dirac-Born-Infeld action on the tachyon kink and vortex,''

12592: Phys.\ Rev.\ D {\bf 68}, 066008 (2003) [arXiv:hep-th/0303057].

12593: %%CITATION = HEP-TH 0303057;%%

12594: 

12595: \bibitem{0305011}

12596: A.~Sen,

12597: %``Open and closed strings from unstable D-branes,''

12598: Phys.\ Rev.\ D {\bf 68}, 106003 (2003)

12599: [arXiv:hep-th/0305011].

12600: %%CITATION = HEP-TH 0305011;%%

12601: 

12602: \bibitem{0306137}

12603: A.~Sen,

12604: %``Open-closed duality at tree level,''

12605: Phys.\ Rev.\ Lett.\  {\bf 91}, 181601 (2003)

12606: [arXiv:hep-th/0306137].

12607: %%CITATION = HEP-TH 0306137;%%

12608: 

12609: \bibitem{0308068}

12610: A.~Sen,

12611: %``Open-closed duality: Lessons from matrix model,''

12612: Mod.\ Phys.\ Lett.\ A {\bf 19}, 841 (2004)

12613: [arXiv:hep-th/0308068].

12614: %%CITATION = HEP-TH 0308068;%%

12615: 

12616: \bibitem{0312003}

12617: A.~Sen,

12618: %``Moduli space of unstable D-branes on a circle of critical radius,''

12619: JHEP {\bf 0403}, 070 (2004)

12620: [arXiv:hep-th/0312003].

12621: %%CITATION = HEP-TH 0312003;%%

12622: 

12623: \bibitem{0312153}

12624: A.~Sen,

12625: %``Remarks on tachyon driven cosmology,''

12626: arXiv:hep-th/0312153.

12627: %%CITATION = HEP-TH 0312153;%%

12628: 

12629: \bibitem{0402157}

12630: A.~Sen,

12631: %``Rolling tachyon boundary state, 

12632: %conserved charges and two dimensional string

12633: %theory,''

12634: JHEP {\bf 0405}, 076 (2004)

12635: [arXiv:hep-th/0402157].

12636: %%CITATION = HEP-TH 0402157;%%

12637: 

12638: \bibitem{0403200}

12639: A.~Sen,

12640: %``Energy momentum tensor and marginal deformations in open string field

12641: %theory,''

12642: JHEP {\bf 0408}, 034 (2004)

12643: [arXiv:hep-th/0403200].

12644: %%CITATION = HEP-TH 0403200;%%

12645: 

12646: \bibitem{0408064}

12647: A.~Sen,

12648: %``Symmetries, conserved charges and (black) holes in two dimensional

12649: %string

12650: %theory,''

12651: arXiv:hep-th/0408064.

12652: %%CITATION = HEP-TH 0408064;%%

12653: 

12654: \bibitem{SENWAD}

12655: A.~M.~Sengupta and S.~R.~Wadia,

12656: %``Excitations And Interactions In D = 1 String Theory,''

12657: Int.\ J.\ Mod.\ Phys.\ A {\bf 6}, 1961 (1991).

12658: %%CITATION = IMPAE,A6,1961;%%

12659: 

12660: \bibitem{9303143}

12661: S.L.\ Shatashvili,

12662: %{\em

12663: %Comment on the background independent open string theory},

12664: Phys.\ Lett.\  {\bf B311}, 83 (1993)

12665: [hep-th/9303143].

12666: %%CITATION = HEP-TH 9303143;%%

12667: 

12668: \bibitem{9311177}

12669: S.L.\ Shatashvili,

12670: %{\em

12671: %On the problems with background independence in string theory},

12672: hep-th/9311177.

12673: %%CITATION = HEP-TH 9311177;%%

12674: 

12675: \bibitem{0105076}

12676: S.~L.~Shatashvili,

12677: %``On field theory of open strings, tachyon condensation and closed  

12678: %strings,''

12679: arXiv:hep-th/0105076.

12680: %%CITATION = HEP-TH 0105076;%%

12681: 

12682: \bibitem{siegel-zwie}

12683: W.~Siegel and B.~Zwiebach,

12684: %``Osp(1,1/2) String Field Theory With The World Sheet Metric,''

12685: Nucl.\ Phys.\ B {\bf 288}, 332 (1987).

12686: %%CITATION = NUPHA,B288,332;%%

12687: 

12688: \bibitem{0310138}

12689: M.~Smedback,

12690: %``On effective actions for the bosonic tachyon,''

12691: arXiv:hep-th/0310138.

12692: %%CITATION = HEP-TH 0310138;%%

12693: 

12694: \bibitem{0007217}

12695: C.~Sochichiu,

12696: %``Noncommutative tachyonic solitons: Interaction with 

12697: %gauge field,''

12698: JHEP {\bf 0008}, 026 (2000)

12699: [arXiv:hep-th/0007217].

12700: %%CITATION = HEP-TH 0007217;%%

12701: 

12702: \bibitem{9807138}

12703: M.~Srednicki,

12704: %``IIB or not IIB,''

12705: JHEP {\bf 9808}, 005 (1998)

12706: [arXiv:hep-th/9807138].

12707: %%CITATION = HEP-TH 9807138;%%

12708: 

12709: \bibitem{9310006}

12710: C.~R.~Stephens, G.~'t Hooft and B.~F.~Whiting,

12711: %``Black hole evaporation without information loss,''

12712: Class.\ Quant.\ Grav.\  {\bf 11}, 621 (1994)

12713: [arXiv:gr-qc/9310006].

12714: %%CITATION = GR-QC 9310006;%%

12715: 

12716: \bibitem{0209090}

12717: A.~Strominger,

12718: %``Open string creation by S-branes,''

12719: arXiv:hep-th/0209090.

12720: %%CITATION = HEP-TH 0209090;%%

12721: 

12722: \bibitem{0307034}

12723: Y.~Sugawara,

12724: %``Thermal partition functions for S-branes,''

12725: JHEP {\bf 0308}, 008 (2003)

12726: [arXiv:hep-th/0307034].

12727: %%CITATION = HEP-TH 0307034;%%

12728: 

12729: \bibitem{0205085}

12730: S.~Sugimoto and S.~Terashima,

12731: %``Tachyon matter in boundary string field theory,''

12732: JHEP {\bf 0207}, 025 (2002) [arXiv:hep-th/0205085].

12733: %%CITATION = HEP-TH 0205085;%%

12734: 

12735: \bibitem{9409089}

12736: L.~Susskind,

12737: %``The World as a hologram,''

12738: J.\ Math.\ Phys.\  {\bf 36}, 6377 (1995)

12739: [arXiv:hep-th/9409089].

12740: %%CITATION = HEP-TH 9409089;%%

12741: 

12742: \bibitem{9805114}

12743: L.~Susskind and E.~Witten,

12744: %``The holographic bound in anti-de Sitter space,''

12745: arXiv:hep-th/9805114.

12746: %%CITATION = HEP-TH 9805114;%%

12747: 

12748: \bibitem{0101002}

12749: T.~Suyama,

12750: %``BPS vortices in brane-antibrane effective theory,''

12751: arXiv:hep-th/0101002.

12752: %%CITATION = HEP-TH 0101002;%%

12753: 

12754: \bibitem{0102192}

12755: T.~Suyama,

12756: %``Tachyon condensation and spectrum of strings on 

12757: %D-branes,''

12758: Prog.\ Theor.\ Phys.\  {\bf 106}, 1017 (2001)

12759: [arXiv:hep-th/0102192].

12760: %%CITATION = HEP-TH 0102192;%%

12761: 

12762: \bibitem{0012210}

12763: T.~Takayanagi, S.~Terashima and T.~Uesugi,

12764: %``Brane-antibrane action from boundary string field theory,''

12765: JHEP {\bf 0103}, 019 (2001)

12766: [arXiv:hep-th/0012210].

12767: %%CITATION = HEP-TH 0012210;%%

12768: 

12769: \bibitem{0103021}

12770: T.~Takayanagi,

12771: %``Holomorphic tachyons and fractional D-branes,''

12772: Nucl.\ Phys.\ B {\bf 603}, 259 (2001)

12773: [arXiv:hep-th/0103021].

12774: %%CITATION = HEP-TH 0103021;%%

12775: 

12776: \bibitem{0106142}

12777: T.~Takayanagi,

12778: %``Tachyon condensation on orbifolds and McKay correspondence,''

12779: Phys.\ Lett.\ B {\bf 519}, 137 (2001)

12780: [arXiv:hep-th/0106142].

12781: %%CITATION = HEP-TH 0106142;%%

12782: 

12783: \bibitem{0307083}

12784: T.~Takayanagi and N.~Toumbas,

12785: %``A matrix model dual of type 0B string theory in two dimensions,''

12786: JHEP {\bf 0307}, 064 (2003)

12787: [arXiv:hep-th/0307083].

12788: %%CITATION = HEP-TH 0307083;%%

12789: 

12790: \bibitem{0107046}

12791: T.~Takahashi and S.~Tanimoto,

12792: %``Wilson lines and classical solutions in cubic open 

12793: %string field  theory,''

12794: Prog.\ Theor.\ Phys.\  {\bf 106}, 863 (2001)

12795: [arXiv:hep-th/0107046].

12796: %%CITATION = HEP-TH 0107046;%%

12797: 

12798: \bibitem{0202133}

12799: T.~Takahashi and S.~Tanimoto,

12800: %``Marginal and scalar solutions in cubic open string field theory,''

12801: JHEP {\bf 0203}, 033 (2002)

12802: [arXiv:hep-th/0202133].

12803: %%CITATION = HEP-TH 0202133;%%

12804: 

12805: \bibitem{0302182}

12806: T.~Takahashi,

12807: %``Tachyon condensation and universal solutions in string field theory,''

12808: Nucl.\ Phys.\ B {\bf 670}, 161 (2003)

12809: [arXiv:hep-th/0302182].

12810: %%CITATION = HEP-TH 0302182;%%

12811: 

12812: \bibitem{0304261}

12813: T.~Takahashi and S.~Zeze,

12814: %``Gauge fixing and scattering amplitudes in string field 

12815: %theory around

12816: %universal solutions,''

12817: Prog.\ Theor.\ Phys.\  {\bf 110}, 159 (2003)

12818: [arXiv:hep-th/0304261].

12819: %%CITATION = HEP-TH 0304261;%%

12820: 

12821: \bibitem{0402196}

12822: T.~Takayanagi,

12823: %``Notes on D-branes in 2D type 0 string theory,''

12824: JHEP {\bf 0405}, 063 (2004)

12825: [arXiv:hep-th/0402196].

12826: %%CITATION = HEP-TH 0402196;%%

12827: 

12828: \bibitem{0403156}

12829: G.~Tasinato, I.~Zavala, C.~P.~Burgess and F.~Quevedo,

12830: %``Regular S-brane backgrounds,''

12831: JHEP {\bf 0404}, 038 (2004)

12832: [arXiv:hep-th/0403156].

12833: %%CITATION = HEP-TH 0403156;%%

12834: 

12835: \bibitem{0001201}

12836: W.~Taylor,

12837: %``D-brane effective field theory from string field theory,''

12838: Nucl.\ Phys.\ B {\bf 585}, 171 (2000)

12839: [arXiv:hep-th/0001201].

12840: %%CITATION = HEP-TH 0001201;%%

12841: 

12842: \bibitem{0008033}

12843: W.~Taylor,

12844: %``Mass generation from tachyon condensation for vector fields on  

12845: %D-branes,''

12846: JHEP {\bf 0008}, 038 (2000)

12847: [arXiv:hep-th/0008033].

12848: %%CITATION = HEP-TH 0008033;%%

12849: 

12850: \bibitem{0208149}

12851: W.~Taylor,

12852: %``A perturbative analysis of tachyon condensation,''

12853: JHEP {\bf 0303}, 029 (2003) [arXiv:hep-th/0208149].

12854: %%CITATION = HEP-TH 0208149;%%

12855: 

12856: \bibitem{0301094}

12857: W.~Taylor,

12858: %``Lectures on D-branes, tachyon condensation, and string field theory,''

12859: arXiv:hep-th/0301094.

12860: %%CITATION = HEP-TH 0301094;%%

12861: 

12862: \bibitem{0311017}

12863: W.~Taylor and B.~Zwiebach,

12864: %``D-branes, tachyons, and string field theory,''

12865: arXiv:hep-th/0311017.

12866: %%CITATION = HEP-TH 0311017;%%

12867: 

12868: \bibitem{0101087}

12869: S.~Terashima,

12870: %``A construction of commutative D-branes from lower 

12871: %dimensional non-BPS

12872: %D-branes,''

12873: JHEP {\bf 0105}, 059 (2001)

12874: [arXiv:hep-th/0101087].

12875: %%CITATION = HEP-TH 0101087;%%

12876: 

12877: \bibitem{0104176}

12878: S.~Terashima and T.~Uesugi,

12879: %``On the supersymmetry of non-BPS D-brane,''

12880: JHEP {\bf 0105}, 054 (2001)

12881: [arXiv:hep-th/0104176].

12882: %%CITATION = HEP-TH 0104176;%%

12883: 

12884: \bibitem{0104158}

12885: J.~Teschner,

12886: %``Liouville theory revisited,''

12887: Class.\ Quant.\ Grav.\  {\bf 18}, R153 (2001)

12888: [arXiv:hep-th/0104158].

12889: %%CITATION = HEP-TH 0104158;%%

12890: 

12891: \bibitem{0303150}

12892: J.~Teschner,

12893: %``A lecture on the Liouville vertex operators,''

12894: arXiv:hep-th/0303150.

12895: %%CITATION = HEP-TH 0303150;%%

12896: 

12897: \bibitem{9512062}

12898: P.~K.~Townsend,

12899: %``D-branes from M-branes,''

12900: Phys.\ Lett.\ B {\bf 373}, 68 (1996)

12901: [arXiv:hep-th/9512062].

12902: %%CITATION = HEP-TH 9512062;%%

12903: 

12904: \bibitem{0101125}

12905: L.~S.~Tseng,

12906: %``Noncommutative solitons and intersecting D-branes,''

12907: Phys.\ Rev.\ D {\bf 64}, 126004 (2001)

12908: [arXiv:hep-th/0101125].

12909: %%CITATION = HEP-TH 0101125;%%

12910: 

12911: \bibitem{0011033}

12912: A.~A.~Tseytlin,

12913: %{\em Sigma model approach to string theory effective actions with

12914: %tachyons,}

12915: hep-th/0011033.

12916: %%CITATION = HEP-TH 0011033;%%

12917: 

12918: \bibitem{0302125}

12919: T.~Uesugi,

12920: %``Worldsheet description of tachyon condensation 

12921: %in open string theory,''

12922: arXiv:hep-th/0302125.

12923: %%CITATION = HEP-TH 0302125;%%

12924: 

12925: \bibitem{9905034}

12926: I.~V.~Vancea,

12927: %``On the algebraic K-theory of the massive D8 and M9 

12928: %branes,''

12929: Int.\ J.\ Mod.\ Phys.\ A {\bf 16}, 4429 (2001)

12930: [arXiv:hep-th/9905034].

12931: %%CITATION = HEP-TH 9905034;%%

12932: 

12933: \bibitem{0104099}

12934: K.~S.~Viswanathan and Y.~Yang,

12935: %``Tachyon condensation and background independent 

12936: %superstring field  theory,''

12937: Phys.\ Rev.\ D {\bf 64}, 106007 (2001)

12938: [arXiv:hep-th/0104099].

12939: %%CITATION = HEP-TH 0104099;%%

12940: 

12941: \bibitem{0107098}

12942: K.~S.~Viswanathan and Y.~Yang,

12943: %``Unorientable boundary superstring field theory 

12944: %with tachyon field,''

12945: Phys.\ Rev.\ D {\bf 65}, 066001 (2002)

12946: [arXiv:hep-th/0107098].

12947: %%CITATION = HEP-TH 0107098;%%

12948: 

12949: \bibitem{math-ph/0306018}

12950: V.~S.~Vladimirov and Y.~I.~Volovich,

12951: %``On the nonlinear dynamical equation in the p-adic string theory,''

12952: Theor.\ Math.\ Phys.\  {\bf 138}, 297 (2004)

12953: [Teor.\ Mat.\ Fiz.\  {\bf 138}, 355 (2004)]

12954: [arXiv:math-ph/0306018].

12955: %%CITATION = MATH-PH 0306018;%%

12956: 

12957: \bibitem{0207089}

12958: J.~E.~Wang,

12959: %``Spacelike and time dependent branes from DBI,''

12960: JHEP {\bf 0210}, 037 (2002)

12961: [arXiv:hep-th/0207089].

12962: %%CITATION = HEP-TH 0207089;%%

12963: 

12964: \bibitem{osft}

12965: E.~Witten,

12966: %``Noncommutative Geometry And String Field Theory,''

12967: Nucl.\ Phys.\ B {\bf 268}, 253 (1986).

12968: %%CITATION = NUPHA,B268,253;%%

12969: 

12970: \bibitem{inverse}

12971: E.~Witten,

12972: %``Interacting Field Theory Of Open Superstrings,''

12973: Nucl.\ Phys.\ B {\bf 276} (1986) 291.

12974: %%CITATION = NUPHA,B276,291;%%

12975: 

12976: \bibitem{9108004}

12977: E.~Witten,

12978: %``Ground ring of two-dimensional string theory,''

12979: Nucl.\ Phys.\ B {\bf 373}, 187 (1992)

12980: [arXiv:hep-th/9108004].

12981: %%CITATION = HEP-TH 9108004;%%

12982: 

12983: \bibitem{9201056}

12984: E.~Witten and B.~Zwiebach,

12985: %``Algebraic structures and differential geometry in $2-D$ string

12986: %theory,''

12987: Nucl.\ Phys.\ B {\bf 377}, 55 (1992) [arXiv:hep-th/9201056].

12988: %%CITATION = HEP-TH 9201056;%%

12989: 

12990: \bibitem{9208027}

12991: E.\ Witten,

12992: %{\em On background independent open string field theory},

12993: Phys.\ Rev.\  {\bf D46}, 5467 (1992)

12994: [hep-th/9208027].

12995: %%CITATION = HEP-TH 9208027;%%

12996: 

12997: \bibitem{9210065}

12998: E.\ Witten,

12999: %{\em

13000: %Some computations in background independent off-shell string theory},

13001: Phys.\ Rev.\  {\bf D47}, 3405 (1993)

13002: [hep-th/9210065].

13003: %%CITATION = HEP-TH 9210065;%%

13004: 

13005: \bibitem{9810188}

13006: E.~Witten,

13007: %``D-branes and K-theory,''

13008: JHEP {\bf 9812}, 019 (1998) [arXiv:hep-th/9810188].

13009: %%CITATION = HEP-TH 9810188;%%

13010: 

13011: \bibitem{9802150}

13012: E.~Witten,

13013: %``Anti-de Sitter space and holography,''

13014: Adv.\ Theor.\ Math.\ Phys.\  {\bf 2}, 253 (1998)

13015: [arXiv:hep-th/9802150].

13016: %%CITATION = HEP-TH 9802150;%%

13017: 

13018: \bibitem{0006071}

13019: E.~Witten,

13020: %``Noncommutative tachyons and string field theory,''

13021: arXiv:hep-th/0006071.

13022: %%CITATION = HEP-TH 0006071;%%

13023: 

13024: \bibitem{0007175}

13025: E.~Witten,

13026: %``Overview of K-theory applied to strings,''

13027: Int.\ J.\ Mod.\ Phys.\ A {\bf 16}, 693 (2001)

13028: [arXiv:hep-th/0007175].

13029: %%CITATION = HEP-TH 0007175;%%

13030: 

13031: \bibitem{0209197}

13032: H.~t.~Yang,

13033: %``Stress tensors in p-adic string theory and truncated OSFT,''

13034: JHEP {\bf 0211}, 007 (2002)

13035: [arXiv:hep-th/0209197].

13036: %%CITATION = HEP-TH 0209197;%%

13037: 

13038: \bibitem{0406023}

13039: H.~t.~Yang,

13040: %``Solving Witten's SFT by insertion of operators on 

13041: %projectors,''

13042: JHEP {\bf 0409}, 002 (2004)

13043: [arXiv:hep-th/0406023].

13044: %%CITATION = HEP-TH 0406023;%%

13045: 

13046: \bibitem{0402027}

13047: H.~U.~Yee and P.~Yi,

13048: %``Open / closed duality, unstable D-branes, and coarse-grained closed

13049: %strings,''

13050: arXiv:hep-th/0402027.

13051: %%CITATION = HEP-TH 0402027;%%

13052: 

13053: \bibitem{9901159}

13054: P.~Yi,

13055: %``Membranes from five-branes and fundamental strings from Dp branes,''

13056: Nucl.\ Phys.\ B {\bf 550}, 214 (1999) [arXiv:hep-th/9901159].

13057: %%CITATION = HEP-TH 9901159;%%

13058: 

13059: \bibitem{0305249}

13060: W.~S.~L'Yi,

13061: %``The energy-carrying velocity of the tachyon of 

13062: %unstable D-branes,''

13063: arXiv:hep-th/0305249.

13064: %%CITATION = HEP-TH 0305249;%%

13065: 

13066: \bibitem{9912255}

13067: T.~Yoneya,

13068: %``Spontaneously broken space-time supersymmetry in open 

13069: %string theory  without

13070: %GSO projection,''

13071: Nucl.\ Phys.\ B {\bf 576}, 219 (2000)

13072: [arXiv:hep-th/9912255].

13073: %%CITATION = HEP-TH 9912255;%%

13074: 

13075: \bibitem{9506136}

13076: A.~B.~Zamolodchikov and A.~B.~Zamolodchikov,

13077: %``Structure constants and conformal bootstrap in Liouville field

13078: %theory,''

13079: Nucl.\ Phys.\ B {\bf 477}, 577 (1996)

13080: [arXiv:hep-th/9506136].

13081: %%CITATION = HEP-TH 9506136;%%

13082: 

13083: \bibitem{0101152}

13084: A.~B.~Zamolodchikov and A.~B.~Zamolodchikov,

13085: %``Liouville field theory on a pseudosphere,''

13086: arXiv:hep-th/0101152.

13087: %%CITATION = HEP-TH 0101152;%%

13088: 

13089: \bibitem{9705241}

13090: B.~Zwiebach,

13091: %``Oriented open-closed string theory revisited,''

13092: Annals Phys.\  {\bf 267}, 193 (1998) [arXiv:hep-th/9705241], and

13093: references therein.

13094: %%CITATION = HEP-TH 9705241;%%

13095: 

13096: \bibitem{0008227}

13097: B.\ Zwiebach,

13098: %A solvable toy model for tachyon condensation in string field theory,

13099: JHEP {\bf 0009}, 028 (2000)

13100: [hep-th/0008227].

13101: %%CITATION = HEP-TH 0008227;%%

13102: 

13103: \bibitem{0010190}

13104: B.~Zwiebach,

13105: %``Trimming the tachyon string field with SU(1,1),''

13106: arXiv:hep-th/0010190.

13107: %%CITATION = HEP-TH 0010190;%%

13108: 

13109: \end{thebibliography}

13110: \end{document}

13111: 

13112: \bye

13113: