1:
2: % TASI 2001 lectures on SFT by Taylor and Zwiebach
3: % v2: references added
4: % v3: reference, discussion added, minor errors corrected
5:
6: \documentclass[12pt]{article}
7:
8: \usepackage{epsfig}
9: \usepackage{amsfonts}
10: \usepackage{amssymb,amsmath}
11:
12: \textheight=9.2in
13: \textwidth=6.5in
14: \headheight=0in
15: \headsep=0in
16: \topmargin=0in
17: \oddsidemargin=0in
18:
19: \begin{document}
20:
21:
22:
23:
24: \def\identity{{\rlap{1} \hskip 1.6pt \hbox{1}}}
25: \def\half{{\textstyle{1\over2}}} %puts a small half in a displayed eqn
26:
27:
28: \newcommand{\NP}{{\em Nucl.\ Phys.\ }}
29: \newcommand{\AP}{{\em Ann.\ Phys.\ }}
30: \newcommand{\PL}{{\em Phys.\ Lett.\ }}
31: \newcommand{\PR}{{\em Phys.\ Rev.\ }}
32: \newcommand{\PRL}{{\em Phys.\ Rev.\ Lett.\ }}
33: \newcommand{\PRP}{{\em Phys.\ Rep.\ }}
34: \newcommand{\CMP}{{\em Comm.\ Math.\ Phys.\ }}
35: \newcommand{\MPL}{{\em Mod.\ Phys.\ Lett.\ }}
36: \newcommand{\IJMP}{{\em Int.\ J.\ Mod.\ Phys.\ }}
37:
38: \pagestyle{plain}
39: \setcounter{page}{1}
40:
41: \baselineskip16pt
42:
43: \begin{titlepage}
44:
45: \begin{flushright}
46: MIT-CTP-3430\\
47: hep-th/0311017
48: \end{flushright}
49: \vspace{13 mm}
50:
51: \begin{center}
52:
53: {\Large \bf D-Branes, Tachyons, and String Field Theory}\\
54: \vspace{3mm}
55: Lectures presented at TASI 2001, Boulder, Colorado
56:
57: \end{center}
58:
59: \vspace{7 mm}
60:
61: \begin{center}
62:
63: Washington Taylor and Barton Zwiebach\\
64:
65: \vspace{3mm}
66: {\small \sl Center for Theoretical Physics} \\
67: {\small \sl MIT, Bldg. 6} \\
68: {\small \sl Cambridge, MA 02139, U.S.A.} \\
69: {\small \tt wati {\rm at} mit.edu},
70: {\small \tt zwiebach {\rm at} mitlns.mit.edu}\\
71: \end{center}
72:
73: \vspace{8 mm}
74:
75: \begin{abstract}
76: In these notes we provide a pedagogical introduction to the subject of
77: tachyon condensation in Witten's cubic bosonic open string field
78: theory. We use both the low-energy Yang-Mills description and the
79: language of string field theory to explain the problem of tachyon
80: condensation on unstable D-branes. We give a self-contained
81: introduction to open string field theory using both conformal field
82: theory and overlap integrals. Our main subjects are the Sen
83: conjectures on tachyon condensation in open string field theory and
84: the evidence that supports these conjectures. We conclude with a
85: discussion of vacuum string field theory and projectors of the
86: star-algebra of open string fields. We comment on the possible role of
87: string field theory in the construction of a nonperturbative
88: formulation of string theory that captures all possible string
89: backgrounds.
90: \end{abstract}
91:
92:
93: %\vspace{2cm}
94: \vspace{1cm}
95: \begin{flushleft}
96: October 2003
97: \end{flushleft}
98: \end{titlepage}
99: \newpage
100:
101:
102: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
103: % Introduction: WT
104: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
105:
106: \section{Introduction}
107: \label{sec:introduction}
108:
109: The last seven years have been a very exciting time for string theory.
110: A~new understanding of nonperturbative objects in
111: string theory, such
112: as D-branes, has led to exciting new developments that relate string
113: theory to physical systems such as black holes and
114: supersymmetric gauge field theories. It has also led to the discovery of
115: unexpected relationships between Yang-Mills theories and
116: quantum theories of gravity such as closed superstring theories and
117: M-theory.
118: Finally, the analysis of unstable D-branes has elucidated
119: the long-standing mysteries associated with the
120: open string tachyon.
121:
122: The study of unstable D-branes and tachyons has also led to the
123: realization that string field theory contains significant
124: non-perturbative information. This has been somewhat of a surprise.
125: Certain forms of string field theory were known since the early
126: 1990's, but there was no concrete evidence that they could be used to
127: give a non-perturbative definition of string theory. The study of
128: tachyon condensation, however, has changed our perspective. These
129: lecture notes give an introduction to string field theory and review
130: recent work in which unstable D-branes and their associated tachyons
131: are described using string field theory. As we will discuss here,
132: this work suggests that open string field theory, or some successor of
133: it, may give a complete definition of string theory in which all
134: possible backgrounds can be obtained from a single set of degrees of
135: freedom. Such a formulation appears to be necessary to address
136: questions related to vacuum selection and string cosmology.
137:
138:
139: In the rest of this section we will review briefly the
140: current status of string theory as a
141: whole, and summarize the goals of this set of lectures. In
142: section~\ref{sec:D-branes}
143: we review some basic aspects of D-branes. In
144: section~\ref{sec:tachyon-D-branes},
145: we describe a particular D-brane configuration which exhibits a tachyonic
146: instability. This tachyon can be seen in the low-energy
147: Yang-Mills description of the D-brane system. We also
148: describe a set of conjectures made by Sen in 1999, which stated
149: that the tachyonic instability of the open bosonic string
150: is the instability
151: of the space-filling D25-brane.
152: Sen suggested that open string field theory could be used to give an
153: analytic description of this instability. In section~\ref{sec:SFT}
154: we give an introduction to
155: Witten's bosonic open string field theory (OSFT).
156: Section~\ref{sec:CFT} gives a more detailed analytic description of this theory
157: using the language of conformal field theory.
158: Section~\ref{sec:overlaps} describes
159: the string field
160: theory using the oscillator approach and overlap integrals.
161: The two approaches to OSFT described in these
162: two sections give complementary ways of
163: analyzing problems in string field theory. In section 7 we
164: summarize evidence from string field theory for Sen's conjectures. In
165: section 8 we describe ``vacuum string
166: field theory,'' a new version of open string field theory which arises when one
167: attempts to directly formulate the theory around
168: the classically stable vacuum
169: where the D-brane has disappeared.
170: This section also discusses
171: important structures in string field theory, such as projectors of the
172: star algebra
173: of open string fields. Section 9 contains concluding remarks.
174:
175:
176: Much new work has been done in this area since these lectures were
177: presented at TASI in 2001. Except for some references to
178: more recent developments
179: which are related to the topics covered,
180: these lecture notes primarily cover work
181: done before summer of 2001. Previous articles reviewing related work
182: include those of Ohmori~\cite{Ohmori}, de Smet~\cite{deSmet}, Aref'eva
183: {\it et al.}~\cite{abgkm}, Bonora {\it et al.}~\cite{Bonora}, and
184: Taylor~\cite{Taylor-Valdivia}.
185: There are a number of major related areas which we do not cover
186: significantly or at all in these lectures. We do not have any
187: substantial discussion on the dynamic process of tachyon decay; there
188: has been quite a bit of work on this subject~\cite{rolling-tachyon}
189: since the time of these lectures in 2001.
190: We do not discuss the Moyal approach to SFT taken recently by Bars and
191: collaborators~\cite{Bars-original,Bars-all,Bars}; this work is an interesting
192: alternative to the level-truncation method primarily used here. We also do not
193: discuss in any detail the alternative boundary string field theory
194: (BSFT) approach to
195: OSFT. The BSFT approach is well suited to derive certain concrete results
196: regarding the tachyon vacuum~\cite{BSFT}---for example, using this
197: approach the energy
198: of the tachyon vacuum can be computed exactly. On the other hand,
199: BSFT is not a
200: completely well-defined framework, as massive string fields cannot yet be
201: consistently incorporated into the theory.
202:
203: \subsection{The status of string theory: a brief review}
204: \label{sec:status}
205:
206: To understand the significance of developments over the last seven
207: years, it is useful to recall the status of string theory in early 1995.
208: At that time
209: it was clearly understood that there were five distinct ways in which a
210: supersymmetric string could be
211: quantized to give a microscopic definition of a theory of quantum
212: gravity in ten dimensions. Each of these quantum string theories
213: gives a set of rules for calculating scattering amplitudes of string
214: states; these states describe gravitational quanta and other massless
215: and massive particles moving in a ten-dimensional spacetime. The
216: five superstring theories are known as the type IIA, IIB, I, heterotic
217: $SO(32)$, and heterotic $E_8 \times E_8$ theories. While
218: these string theories give perturbative descriptions of quantum
219: gravity, there was little understanding in 1995 of nonperturbative aspects of
220: these theories.
221:
222: In the years between 1995 and 2002, several new ideas dramatically
223: transformed our understanding of string theory. We now
224: briefly summarize these ideas and mention some aspects of these
225: developments relevant to the main topic of these lectures.
226: \vspace*{0.07in}
227:
228: \noindent
229: {\bf Dualities:} The five different perturbative formulations of
230: superstring theory are all related to one another through duality
231: symmetries~\cite{Hull-Townsend,Witten-various}, whereby the degrees of
232: freedom in one theory can be described through a duality
233: transformation in terms of the degrees of freedom of another theory.
234: Some of these duality symmetries are nonperturbative, in the sense
235: that the string coupling $g$ in one theory is related to the inverse
236: string coupling $1/g$ in the dual theory. The web of dualities that relate
237: the different theories gives a picture in which, rather than
238: describing five distinct fundamental theories, each
239: superstring theory appears to be a particular
240: perturbative limit of a single, still unknown,
241: underlying theoretical structure.
242: \vspace*{0.07in}
243:
244: \noindent
245: {\bf M-theory:} In addition to the five perturbative string theories,
246: the web of dualities also seems to include a limit which describes a
247: quantum theory of gravity in eleven dimensions. This new theory has
248: been dubbed ``M-theory''. Although no covariant definition for
249: M-theory has been given, this theory can be related to type IIA and
250: heterotic $E_8 \times E_8$ string theories through compactification on
251: a circle $S^1$ and the space $S^1/Z_2$,
252: respectively~\cite{Townsend-11,Witten-various,Horava-Witten}. In
253: the relation to type IIA, for example, the compactification radius
254: $R_{11}$ of M-theory is equal to the product $ g_sl_s$ of the string coupling
255: $g_s$ and the string length $l_s$. Thus, M-theory in flat space,
256: which arises in the limit $R_{11} \rightarrow \infty$, can be thought
257: of as the strong coupling limit of type IIA string theory. The field theory
258: limit of M-theory is eleven-dimensional supergravity. It is also
259: suspected that M-theory may be formulated as a quantum theory of
260: membranes in eleven dimensions~\cite{Townsend-11}.
261: \vspace*{0.07in}
262:
263: \noindent
264: {\bf Branes:} In addition to strings, all five superstring
265: theories, as well as M-theory, contain extended objects
266: of various dimensionalities
267: known as ``branes''. M-theory has M2-branes and
268: M5-branes, which have two and five dimensions of spatial extent, respectively.
269: (A string is a one-brane, since it has one spatial dimension.)
270: The different superstring theories each have different sets
271: of (stable) D-branes,
272: special branes that are defined
273: by Dirichlet-type boundary conditions on strings.
274: In particular, the IIA/IIB superstring theories contain
275: (stable) D-branes of all even/odd dimensions.
276: Each superstring theory also has a fundamental string and a
277: Neveu-Schwarz five-brane. The branes of one theory can be
278: related to the branes of another through the duality transformations mentioned
279: above. Using an appropriate sequence of dualities, any brane can be mapped
280: to any other brane, including the string itself. This suggests that
281: none of these
282: objects are really any more fundamental than any others; this idea is known as
283: ``brane democracy''.
284: \vspace*{0.07in}
285:
286: \noindent
287: {\bf M(atrix) theory and AdS/CFT:} It is a remarkable consequence
288: of the above developments that for certain
289: asymptotic space-time backgrounds, M-theory and
290: string theory can be completely
291: described through supersymmetric quantum mechanics and field
292: theories related to the low-energy description of systems of
293: branes. The M(atrix) model of M-theory is a simple supersymmetric
294: matrix quantum mechanics, and it is believed to capture (in light-cone
295: coordinates) all of the
296: physics of M-theory in asymptotically flat spacetime.
297: In the AdS/CFT correspondence, certain maximally supersymmetric Yang-Mills
298: theories can be used to describe closed superstring theories in
299: asymptotic spacetime backgrounds
300: that are the product of anti-de Sitter space and a sphere.
301: It is believed that the
302: Yang-Mills theories and the matrix model of M-theory, each
303: give true nonperturbative descriptions of quantum gravity in the
304: corresponding spacetime geometry. For reviews of
305: M(atrix) theory and AdS/CFT, see Taylor~\cite{WT-RMP}
306: and Aharony{\em et.al}~\cite{agmoo}.
307:
308:
309:
310: \vspace*{0.07in}
311:
312: \noindent
313: {\bf Unstable D-branes and open string tachyons:} This is in large
314: part the subject of these lectures. The most recent chapter in our
315: new understanding of nonperturbative effects in string theory has been
316: the incorporation of unstable branes and open string tachyons into the
317: overall framework of the theory. It has turned out that an
318: understanding of unstable D-branes is necessary to properly describe
319: all D-branes. This is natural from the point of view of K-theory,
320: where brane configurations which are equivalent under the annihilation
321: of unstable branes are identified~\cite{k-theory}. The
322: long-mysterious tachyon instability of open string theory has finally
323: been given a physical interpretation: it is the instability of the
324: D-brane that supports the existence of open strings. The instability
325: disappears in the tachyon vacuum, in which the D-brane decays.
326: Moreover, the belief that D-branes are solitonic solutions of string
327: theory has been confirmed: starting with the appropriate tachyonic
328: field theory of unstable space-filling branes, one can describe lower
329: dimensional D-branes as solitonic solutions.
330: Lower dimensional D-branes are thereby essentially obtained as
331: solitons of the tachyon field theory, so, in some sense,
332: lower-dimensional D-branes
333: can be thought of as being made of tachyons!
334: It has also been shown that the physics of unstable D-branes is
335: captured by string field theory, thus making it a candidate for a
336: non-perturbative formulation of string theory capable of describing
337: changes of the string background.
338:
339:
340: \vspace*{0.1in}
341:
342: The set of ideas just summarized have greatly increased our
343: understanding of nonperturbative aspects of string theory. In
344: particular M(atrix) theory and the AdS/CFT correspondences
345: provide nonperturbative
346: definitions of M-theory and string theory in certain asymptotic space-time
347: backgrounds which can be used, in principle, to calculate any local result
348: in quantum gravity. Through string field theory we have a possibly
349: nonperturbative definition of the theory that appears to capture many open
350: string theory backgrounds.
351: The existing
352: formulations of string field theory are not manifestly
353: background independent because
354: a background must be selected to write the theory.
355: Nevertheless, as we discuss in
356: these lectures, the theory describes multiple distinct backgrounds in
357: terms of a common set of variables,
358: so it embodies, at least partially, physical background independence.
359: It remains to
360: be seen if the theory
361: incorporates full physical
362: background independence; this requires an ability
363: to describe all possible open string backgrounds,
364: as well as all possible closed string backgrounds.
365:
366:
367:
368: \subsection{The goal of these lectures}
369:
370: The goal of these lectures is to describe progress towards a
371: nonperturbative formulation of string theory that implements
372: the physics of background independence.
373: Open string field theory, as applied to tachyon condensation
374: and related matters, has shown itself capable of describing
375: non-perturbative objects in string theory, and it has demonstrated
376: an ability to represent various open string backgrounds.
377:
378:
379: A completely background independent formulation of string theory
380: may be needed
381: to address fundamental questions such as:
382: What is string theory/M-theory? How is the vacuum of string theory
383: selected? ({\it i.e.}, Why can the observable low-energy universe be
384: accurately described by the standard model of particle physics in four
385: space-time dimensions with an apparently small but nonzero positive
386: cosmological constant?), and other questions of a cosmological nature.
387: Obviously, aspiring to address these questions is an ambitious
388: undertaking, but we believe that attaining a better understanding of
389: string field theory is a
390: useful step in this direction.
391: More concretely, in these lectures we will describe recent progress on
392: open string field theory. It may be useful here to recall some basic
393: aspects of open and closed strings and the relationship between them.
394:
395: Closed strings, which are topologically equivalent to a circle $S^1$,
396: give rise upon quantization to a massless set of
397: states associated
398: with the graviton $g_{\mu \nu}$, the dilaton $\varphi$, and the
399: antisymmetric two-form $B_{\mu \nu}$, as well as an infinite family of
400: massive states.
401: For the supersymmetric closed string, further
402: massless fields
403: appear within % associated with
404: the graviton supermultiplet---these are the Ramond-Ramond $p$-form fields
405: $A^{(p)}_{\mu_1
406: \cdots \mu_p}$ and the gravitini $\psi_{\mu \alpha}$. Thus, the
407: quantum theory of closed strings is naturally associated with a theory
408: of gravity in space-time. On the other hand, open strings, which are
409: topologically equivalent to an interval $[0, \pi]$, give rise under
410: quantization to a massless gauge field $A_\mu$ in space-time. The
411: supersymmetric open string also has a massless gaugino field
412: $\psi_\alpha$.
413: It is now understood that
414: the endpoints of open strings
415: must lie
416: on a Dirichlet $p$-brane (D$p$-brane),
417: and that the massless open string fields describe the fluctuations of
418: the D-brane and the gauge field living on the world-volume of the D-brane.
419:
420: It may seem, therefore, that open and closed strings are quite
421: distinct, and describe disjoint aspects of the physics in a fixed
422: background space-time that contains some family of D-branes. At tree
423: level, the closed strings indeed describe gravitational physics in the
424: bulk space-time, while the open strings describe the D-brane dynamics.
425: At the quantum level, however, the physics of open and closed strings
426: are deeply connected. Indeed, historically open strings were
427: discovered first through the form of their
428: scattering amplitudes~\cite{Veneziano}. Looking at one-loop processes for open
429: strings led to the first discovery of closed strings, which appeared as {\em
430: poles} in nonplanar one-loop open string diagrams~\cite{NGS,Lovelace}. The
431: fact that open string diagrams naturally contain closed string
432: intermediate states indicates that in some sense all closed string
433: interactions are implicitly defined by the open
434: string diagrams. This connection underlies many of the important
435: recent developments in string theory. In particular, the M(atrix)
436: theory and AdS/CFT correspondences between gauge theories and quantum
437: gravity are essentially limits in which closed string physics in a
438: fixed space-time background is captured by
439: the Yang-Mills limit of an open string theory on a family
440: of branes (D0-branes for M(atrix) theory, D3-branes for the CFT that describes
441: AdS${}_5\times S^5$, etc.)
442:
443:
444: Since quantum gravity theories in certain fixed space-time
445: backgrounds can be described by field theory limits of open strings,
446: we may ask if a global change of the space-time background
447: can be described as well.
448: If M(atrix) theory or AdS/CFT allowed for this description, it would
449: indicate that
450: these models may have background-independent generalizations.
451: Unfortunately, such background changes involve the generally intractable
452: addition of an infinite number of nonrenormalizable interactions to the field
453: theories in question.
454: One tractable situation arises
455: for the addition of a constant background $B_{\mu\nu}$ field in
456: space-time (perhaps because this closed string background is gauge
457: equivalent to the open string background of a D-brane
458: with a magnetic field). In the
459: associated Yang-Mills theory,
460: this change
461: in the background field corresponds to replacing products of open string
462: fields with a noncommutative star-product. The resulting theory is a
463: noncommutative Yang-Mills theory. Such noncommutative theories are
464: the only well-understood example of a situation where adding an
465: infinite number of apparently nonrenormalizable terms to a field
466: theory action leads to a sensible modification of quantum field
467: theory (for a review of noncommutative field theory and its connection
468: to string theory, see Douglas and Nekrasov~\cite{Douglas-Nekrasov}).
469:
470: String field theory is a nonperturbative formulation of
471: string theory in which the infinite family of
472: fields associated with string excitations are described by a
473: space-time field theory action.
474: For open strings on a D-brane configuration, this field theory
475: contains Yang-Mills
476: fields and an entire hierarchy of massive string
477: fields. Integrating out all the massive fields from the string field
478: theory action
479: gives rise to a nonabelian Born-Infeld action for the
480: D-branes, which includes an
481: infinite set of higher-order terms that arise from string theory
482: corrections to the
483: simple Yang-Mills action. Like the case of noncommutative field
484: theory discussed
485: above, the new terms appearing in this action are apparently nonrenormalizable,
486: but the combination of terms must work together to form a sensible theory.
487:
488: In the 1980's, a great deal of work was done to formulate string
489: field theory for open and closed, bosonic and supersymmetric string
490: theories. All of this work was based on the BRST approach to
491: string
492: quantization~\cite{Kato:1982im,Siegel:1984xd,Siegel:1985tw,Banks:1985ff}.
493: For the open bosonic
494: string Witten~\cite{Witten-SFT} constructed an extremely elegant string
495: field theory based on the Chern-Simons action. This cubic open
496: string field theory
497: (OSFT) is the primary focus of the work described in these lectures.
498: Although this theory can be described in a simple abstract language,
499: practical computations rapidly become
500: complicated. The formulation of bosonic closed string field theory
501: was completed in the early
502: 1990s~\cite{Zwiebach:1992ie,Saadi:tb,Kugo:1989tk,Kaku:zw}. This theory is the
503: natural counterpart of Witten's open string field theory, but it is
504: more technically
505: challenging because of its nonpolynomiality.
506: A nonpolynomial string field theory is also required to describe
507: in a non-singular fashion open and closed string fields~\cite{Zwiebach:1997fe}.
508: For open superstrings, a cubic formulation~\cite{Witten:1986qs}
509: encountered some
510: difficulties~\cite{Wendt,Greensite-Klinkhamer} (for which there are some
511: proposed resolutions~\cite{Arefeva:1989cp,Preitschopf:fc}), but the
512: nonpolynomial formulation of Berkovits~\cite{Berkovits} appears to be fully
513: consistent. Despite a substantial amount of work in string field
514: theory in the early 90's, little
515: insight was gained at the time concerning non-perturbative physics. Work on
516: this subject stalled out until open string field theory was used to test the
517: tachyon conjectures beginning in 1999~\cite{Sen-Zwiebach}.
518:
519: \smallskip
520: One simple feature of the 26-dimensional bosonic string has been
521: problematic since the early days of string theory: both the open and
522: closed bosonic strings have tachyons in their spectra, indicating that
523: the usual perturbative vacua of these
524: theories are unstable. In 1999, Ashoke Sen had a remarkable insight
525: into the nature of the open bosonic string
526: tachyon~\cite{Sen-universality}. He observed that the open bosonic
527: string theory (the so-called Veneziano model) represents open strings that end
528: on a space-filling D25-brane. He pointed out that this D-brane is unstable,
529: as it does not carry any conserved charge, and he suggested that the open
530: string tachyon is in fact the unstable mode of the D25-brane. This led
531: him to conjecture that open string field theory could be used to precisely
532: determine a new vacuum for the open string, namely one in which the D25-brane
533: is annihilated through condensation of the tachyonic unstable mode. Sen made
534: several precise conjectures regarding the details of the string field theory
535: description of this new open string vacuum. As we describe in these lectures,
536: there is now overwhelming evidence that Sen's picture is correct, demonstrating
537: that string field theory accurately describes the nonperturbative
538: physics of D-branes. This new nonperturbative application of string
539: field theory has sparked a new wave of work on open
540: string field theory, revealing many remarkable new structures.
541: In particular, string field theory now provides a concrete framework
542: in which disconnected string backgrounds can emerge from the equations
543: of motion of a single underlying theory. Although so far this can
544: only be shown explicitly in the open string context, this work paves
545: the way for a deeper understanding of background-independence in
546: quantum theories of gravity.
547:
548:
549: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
550: % D-branes: WT
551: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
552:
553: \section{D-branes}
554: \label{sec:D-branes}
555:
556: In this section we briefly review some basic features of D-branes.
557: The concepts developed here will be useful to describe tachyonic
558: D-brane configurations in the following section. For more detailed
559: reviews of D-branes, see the reviews of
560: Polchinski~\cite{Polchinski-TASI}, and of
561: Taylor~\cite{WT-Trieste}.
562:
563: \subsection{D-branes and Ramond-Ramond charges}
564:
565: D-branes can be understood from many points of view. In these
566: lectures we primarily focus on the viewpoint motivated by the recent
567: work on tachyon condensation, which is that that D-branes are solitons
568: in string field theory. The original realization of the importance of
569: D-branes in string theory stemmed from Polchinski's realization that
570: D-branes could be described in two complementary fashions:
571: {\it a}) as extended extremal black brane solutions of
572: supergravity that carry conserved charges, and {\it b}) as hypersurfaces on
573: which strings have Dirichlet boundary conditions.
574: We now discuss these
575: two viewpoints briefly.
576: \vspace*{0.03in}
577:
578: \noindent
579: {\it a}) The ten-dimensional type IIA and IIB supergravity theories
580: each have a set of $(p + 1)$-form fields $A^{(p + 1)}_{\mu_1 \cdots
581: \mu_{(p+ 1)}}$ in the supergraviton multiplet, with $p$ even/odd for
582: type IIA/IIB supergravity. These are the Ramond-Ramond (RR) fields in the
583: massless superstring spectrum. For each of these $(p + 1)$-form
584: fields, there is a solution of the supergravity field equations that
585: is invariant under $(p + 1)$-dimensional Lorentz transformations, and which has
586: the form of an extremal black hole
587: solution in the
588: $9-p$ spatial directions that are not affected
589: by these Lorentz transformations~\cite{dkl}.
590: These ``black
591: $p$-brane'' solutions carry charge under the RR fields $A^{(p + 1)}$,
592: and are BPS states in the supergravity theory that preserve half the
593: supersymmetry of the theory.
594: These solutions represent the gravitational
595: and gauge backgrounds created by the branes, in a way similar to that in which
596: the Schwarzschild solution represents the gravitational background of
597: a point mass, or the Coulomb field represents the electric field of
598: a point charge.
599: \vspace*{0.03in}
600:
601: \noindent
602: {\it b})
603: In type IIA and IIB string theory, it is possible to consider open
604: strings with Dirichlet boundary conditions on some number $9-p$ of
605: the spatial coordinates $x^\mu (\sigma)$. The locus of points defined
606: by such Dirichlet boundary conditions defines a $(p + 1)$-dimensional
607: hypersurface $\Sigma_{p + 1}$ in the ten-dimensional spacetime.
608: When $p$ is even/odd in
609: type IIA/IIB string theory, the spectrum of the resulting quantum open
610: string theory contains a massless set of fields $A_\alpha, \alpha = 0,
611: 1, \ldots, p$ and $X^{a}, a = p + 1, \ldots, 9$.
612: These fields can be associated with a gauge field living on the
613: hypersurface $\Sigma_{p + 1}$, and a set of degrees of
614: freedom describing the transverse fluctuations of this hypersurface in
615: spacetime, respectively. Thus, the quantum fluctuations of the open string
616: describe a fluctuating $(p + 1)$-dimensional hypersurface in spacetime --- a
617: Dirichlet-brane, or ``D-brane''.
618:
619: The remarkable insight of Polchinski~\cite{Polchinski} in 1995
620: was the observation that the stable Dirichlet-branes
621: of superstring theory carry Ramond-Ramond
622: charges, and therefore should be described in the low-energy
623: supergravity limit of string theory by precisely the black $p$-branes
624: discussed in {\it a}). This connection between the string and
625: supergravity descriptions of these nonperturbative objects paved the
626: way to a dramatic series of new developments in string theory,
627: including connections between string theory and supersymmetric gauge
628: theories, string constructions of black holes, and new approaches to
629: string phenomenology. The
630: bosonic D-branes on which we concentrate
631: attention in these lectures do not carry conserved charges, and thus
632: are not associated with supergravity solutions as in {\it a)}; rather,
633: these D-branes can be described through open bosonic strings with some
634: Dirichlet boundary conditions as in {\it b)}.
635:
636: \subsection{Born-Infeld and super Yang-Mills D-brane actions}
637:
638: In this subsection we briefly review the low-energy super Yang-Mills
639: description of the dynamics of one or more D-branes. As discussed in
640: the previous subsection, the massless open string modes on a
641: D$p$-brane in type IIA or IIB superstring theory describe a $(p +
642: 1)$-component gauge field $A_\alpha$, $9-p$ transverse scalar fields
643: $X^a$, and a set of massless fermionic gaugino fields. The scalar
644: fields $X^a$ describe small fluctuations of the D-brane around a flat
645: hypersurface. If the D-brane geometry is sufficiently far from flat,
646: it is useful to describe the D-brane configuration by a general
647: embedding $X^\mu (\xi)$, where $\xi^\alpha$ are $p + 1$ coordinates on
648: the D$p$-brane world-volume $\Sigma_{(p + 1)}$, and $X^\mu$ are ten
649: functions giving a map from $\Sigma_{(p + 1)}$ into the space-time
650: % blackboard
651: manifold ${\bf R}^{ 9, 1}$. Just as the Einstein equations which govern
652: the geometry of spacetime arise from the condition that the one-loop
653: contribution to the closed
654: string beta function vanishes,
655: a set of equations of motion for a general D$p$-brane geometry and associated
656: world-volume gauge field can be derived from a calculation of the
657: one-loop open string beta function~\cite{Leigh}. These equations of
658: motion arise from the classical Born-Infeld action:
659: \begin{equation}
660: S = - T_p \int d^{p + 1} \xi
661: \;e^{-\varphi} \;\sqrt{-\det (G_{\alpha \beta} + B_{\alpha \beta} + 2
662: \pi \alpha'
663: F_{\alpha \beta}) } + S_{{\rm CS}}+{\rm fermions}
664: \label{eq:DBI}
665: \end{equation}
666: where $G$, $B$, and $\varphi$ are the pullbacks of the ten-dimensional metric,
667: antisymmetric tensor, and dilaton to the D-brane world-volume, while $F$ is the
668: field strength of the world-volume $U(1)$ gauge field $A_{\alpha}$.
669: $S_{\rm CS}$ represents a set of Chern-Simons terms which will be
670: discussed in the following subsection.
671: This action can be verified by a perturbative string
672: calculation~\cite{Polchinski-TASI}, which also gives a precise
673: expression for the brane tension
674: \begin{equation}
675: \tau_p =\frac{T_p}{g_s} =
676: \frac{1}{g_s\sqrt{\alpha'}} \frac{1}{ (2 \pi \sqrt{\alpha'})^{p}}
677: \end{equation}
678: where $g_s = e^{\langle \varphi \rangle}$ is the closed string
679: coupling, equal to
680: the exponential
681: of the dilaton expectation value.
682:
683: A particular limit of the Born-Infeld action
684: (\ref{eq:DBI}) is useful to describe many low-energy aspects of
685: D-brane dynamics. Take the background space-time $G_{\mu \nu}=
686: \eta_{\mu \nu}$ to be flat, and all other supergravity fields ($B_{\mu
687: \nu}, A^{(p + 1)}_{\mu_1 \cdots \mu_{p + 1}}$) to vanish. We then
688: assume that the D-brane is approximately flat, and is close to the
689: hypersurface $X^a = 0, a > p$, so that we may make the static gauge
690: choice $X^\alpha = \xi^\alpha$. We furthermore take the low-energy
691: limit
692: in which
693: $\partial_\alpha X^a$ and $2 \pi \alpha' F_{\alpha \beta}$ are small
694: and of the same order. The action (\ref{eq:DBI})
695: can then be expanded as
696: \begin{equation}
697: S =-\tau_pV_p -\frac{1}{4g_{{\rm YM}}^2}
698: \int d^{p + 1} \xi
699: \left(F_{\alpha \beta} F^{\alpha \beta} +\frac{2}{(2 \pi \alpha')^2}
700: \partial_\alpha X^a \partial^\alpha X^a\right) +\cdots
701: \label{eq:action-expansion}
702: \end{equation}
703: where $V_p$ is the $p$-brane world-volume and the coupling $g_{{\rm
704: YM}}$ is given by
705: \begin{equation}
706: g_{{\rm YM}}^2 = \frac{1}{4 \pi^2 \alpha'^2 \tau_p}
707: = \frac{g_s}{\sqrt{\alpha'}} (2 \pi \sqrt{\alpha'})^{p-2}\,.
708: \label{eq:ym-coupling}
709: \end{equation}
710: Ignoring fermionic terms,
711: the second term in the right-hand side of
712: (\ref{eq:action-expansion})
713: is simply the reduction to $(p + 1)$ dimensions of the ten-dimensional
714: ${\mathcal N} = 1 $ super Yang-Mills action:
715: \begin{equation}
716: S = \frac{1}{g_{{\rm YM}}^2} \int d^{10}\xi \; \left(
717: -\frac{1}{4} F_{\mu \nu}F^{\mu \nu}
718: + \frac{i}{2} \bar{\psi} \Gamma^\mu \partial_{\mu} \psi \right)
719: \label{eq:SYM1}
720: \end{equation}
721: where for $\alpha, \beta \leq p$, $F_{\alpha \beta}$ is the
722: world-volume $U(1)$ field strength, and for $a > p, \alpha \leq p$, $
723: F_{\alpha a} \rightarrow \partial_\alpha X^a/(2\pi \alpha')$.
724:
725:
726: When multiple D$p$-branes are present, the D-brane action is modified
727: in a fairly simple fashion~\cite{Witten-multiple}. Consider a system
728: of $N$ coincident D-branes. For every pair of branes $\{i, j\}$ there
729: is a set of massless fields
730: \begin{equation}
731: (A_\alpha)_i^{\; j}, \; \; \;(X^a)_i^{\; j}\,,
732: \label{eq:nonabelian-fields}
733: \end{equation}
734: associated with
735: strings stretching from the $i$th brane to the $j$th brane; the
736: indices $i, j$ are known as Chan-Paton indices. Treating the fields
737: (\ref{eq:nonabelian-fields}) as
738: $N$-by-$N$ matrices, and letting Tr denote
739: the trace operation for such matrices, the multiple
740: brane analogue of the Born-Infeld action (\ref{eq:DBI}) takes the
741: schematic form
742: \begin{equation}
743: S \sim \int {\rm Tr}\; \sqrt{-\det \left( G + B + 2\pi \alpha' F \right)}\,.
744: \label{eq:NBI}
745: \end{equation}
746: In order to properly define this nonabelian analog of the Born-Infeld
747: action (NBI), it is necessary to resolve the ordering ambiguities in
748: (\ref{eq:NBI}). Since the spacetime coordinates $X^a$ associated with
749: the D-brane positions in space-time become themselves matrix-valued,
750: even evaluating the pullbacks $G_{\alpha \beta}, B_{\alpha \beta}$
751: involves resolving ordering issues. Much work has been done recently
752: to resolve these ordering ambiguities~\cite{ordering-NBI}
753: but there is still no known
754: definition of the nonabelian Born-Infeld theory
755: (\ref{eq:NBI}) which is valid to all orders.
756:
757: The nonabelian Born-Infeld action (\ref{eq:NBI}) becomes much simpler
758: when, once again, the background space-time is
759: assumed to be flat and we take the low-energy
760: limit , leading to
761: the nonabelian $U(N)$ super Yang-Mills action in $p
762: + 1$ dimensions. This action is the reduction to $p + 1$
763: dimensions of the
764: ten-dimensional
765: $U(N)$ super Yang-Mills action (analogous to (\ref{eq:SYM1})).
766: In this reduction,
767: for $\alpha, \beta \leq p$, $F_{\alpha \beta}$ is the
768: world-volume $U(N)$ field strength,
769: and for $a >p,\alpha\leq p$,
770: $ F_{\alpha a} \rightarrow \partial_\alpha X^a$, where now $ A_\alpha,
771: X^a,$ and $F_{\alpha \beta}$ are $N \times N$ matrices.
772: Since the derivatives $\partial_a$ are set to zero in the dimensional
773: reduction, we
774: furthermore have, for $a, b > p$, $F_{ab} \rightarrow -i[X^a, X^b]$.
775:
776: The low-energy description of a system of $N$ coincident flat D-branes
777: is thus given by $U(N)$ super Yang-Mills theory in the appropriate
778: dimension. This connection between D-brane actions in string theory
779: and super Yang-Mills theory has led to many new developments,
780: including new insights into supersymmetric field theories, the
781: M(atrix) theory and AdS/CFT correspondences, and brane world
782: scenarios.
783:
784:
785: \subsection{Branes from branes}
786:
787: In this subsection we describe a remarkable feature of D-brane
788: systems:
789: one or more D-branes of a fixed
790: dimension can be used to construct additional D-branes of higher or
791: lower dimension.
792:
793: In our discussion of the D-brane action (\ref{eq:DBI}),
794: we mentioned a group of terms $S_{\rm CS}$ which we did not describe
795: explicitly. For a single D$p$-brane, these Chern-Simons terms can be
796: combined into a single expression of the form
797: \begin{equation}
798: S_{\rm CS}\sim \int_{\Sigma_{p + 1}}\;{\mathcal A} \;e^{F + B}\,,
799: \label{eq:Chern-Simons}
800: \end{equation}
801: where ${\mathcal A} = \sum_{k}A^{(k)} $ represents a formal sum over all
802: the Ramond-Ramond fields $A^{(k)}$ of various dimensions. In this
803: integral, for each term $A^{(k)}$, the nonvanishing contribution to
804: (\ref{eq:Chern-Simons}) is given by expanding the exponential of $F +
805: B$ to order $(p + 1-k)/2$, where the dimension of the resulting form
806: saturates the dimension of the brane. For example, on a D$p$-brane,
807: there is a coupling of the form
808: \begin{equation}
809: \int_{\Sigma_{(p + 1)}} A^{(p-1)} \wedge F\,.
810: % % \label{eq:}
811: \end{equation}
812: This coupling implies that the $U(1)$ field strength on the D$p$-brane
813: couples to the RR field associated with $(p-2)$-branes. Thus, we can
814: associate magnetic fields on a D$p$-brane with dissolved
815: $(p-2)$-branes living on the D$p$-brane. This result generalizes to a
816: system of multiple D$p$-branes, in which case a trace is included on
817: the right-hand side of (\ref{eq:Chern-Simons}).
818: For example, on $N$ compact
819: D$p$-branes wrapped on a $p$-torus,
820: the
821: flux
822: \begin{equation}
823: \frac{1}{2 \pi} \int {\rm Tr}\; F_{\alpha \beta},
824: \label{eq:p-2-charge}
825: \end{equation}
826: of the magnetic field over a
827: two-cycle on the torus is quantized
828: and measures the number of units of D$(p-2)$-brane charge on
829: the D$p$-branes that threads the cycle integrated over. Thus, these
830: branes are encoded in the field strength $F_{\alpha \beta}$.
831: The object in (\ref{eq:p-2-charge}) is the relevant component of the first
832: Chern class of the
833: $U(N)$ bundle described by the gauge field on the $N$ branes.
834: Similarly,
835: \begin{equation}
836: \frac{1}{8 \pi^2} \int {\rm Tr}\; F\wedge F
837: % % \label{eq:}
838: \end{equation}
839: encodes D$(p -4)$-brane charge on the D$p$-branes, etc..
840:
841: Just as lower-dimensional branes can be described in terms of the
842: degrees of freedom associated with a system of $N$ D$p$-branes through
843: the field strength $F_{\alpha \beta}$, higher-dimensional branes can
844: be described by a system of $N$ D$p$-branes in terms of the
845: commutators of the matrix-valued scalar fields $X^a$. Just as
846: $\frac{1}{2 \pi} F$ measures $(p-2)$-brane charge, the matrix
847: \begin{equation}
848: 2 \pi i
849: [X^a, X^b]
850: \label{eq:up-charge}
851: \end{equation}
852: measures $ (p + 2)$-brane charge~\cite{WT-Trieste,Mark-Wati-4,Myers}. The
853: charge (\ref{eq:up-charge}) should be interpreted as a form of local charge
854: density.
855: Just as the $N$ positions of the D$p$-branes are replaced by matrices
856: in the nonabelian theory, so the locations of the charges become matrix-valued.
857: The
858: trace of (\ref{eq:up-charge}) vanishes for finite sized matrices
859: because the net D$p$-brane charge of a
860: finite-size brane configuration in flat spacetime vanishes. Higher
861: multipole moments of the brane charge, however, have a natural
862: definition in terms of traces of the charge matrix multiplied by
863: powers of the scalars $X^a$, and generically are nonvanishing.
864:
865: A simple example of the mechanism by which a system of multiple
866: D$p$-branes form a higher-dimensional brane is given by the matrix
867: sphere. If we take a system of D0-branes with scalar matrices $X^a$
868: given by
869: \begin{equation}
870: X^a = \frac{2r}{ N} J^a, \;\;\;\;\; a = 1, 2, 3
871: \label{eq:matrix-sphere}
872: \end{equation}
873: where $J^a$ are the generators of $SU(2)$ in the $N$-dimensional
874: representation, then we have a configuration corresponding to the
875: ``matrix sphere''. This is a D2-brane of spherical geometry living on
876: the locus of points satisfying $x^2 + y^2 + z^2 = r^2$. The ``local''
877: D2-brane charge of this brane is given by (\ref{eq:up-charge});
878: here, for example, the D2-brane charge in the $x$-$y$ plane is
879: proportional to the matrix $X^3$ ($z$), as one would expect from the
880: geometry of a spherical brane.
881: The D2-brane configuration given by (\ref{eq:matrix-sphere}) is
882: rotationally invariant (up to a gauge transformation). The
883: restriction of the brane to the desired locus of
884: points can be seen from the relation $(X^1)^2 + (X^2)^2 + (X^3)^2 =
885: r^2\identity+{\mathcal O} (N^{-2})$.
886:
887: \subsection{T-duality}
888: \label{sec:T-duality}
889:
890: We conclude our discussion of D-branes with a brief description of
891: T-duality. T-duality is a perturbative and nonperturbative
892: symmetry which relates
893: the type IIA and type IIB string theories. This duality symmetry was in
894: fact crucial in the original discovery of D-branes~\cite{Polchinski}.
895: A more detailed discussion of T-duality can be found in the textbook
896: by Polchinski~\cite{Polchinski-string}.
897: Using T-duality, we construct an explicit example of a brane within a
898: brane encoded in super Yang-Mills theory, illustrating the ideas of
899: the previous subsection. This example will be used in the following
900: section to construct an analogous configuration with a tachyon.
901:
902: Consider type IIA string theory on a spacetime of the form $M^9 \times
903: S^1$ where $M^9$ is a generic 9-manifold of Lorentz signature, and
904: $S^1$ is a circle of radius $R$. T-duality is the statement that this
905: theory is precisely equivalent, even
906: at the perturbative level,
907: to type IIB
908: string theory on the spacetime $M^9 \times (S^1)'$, where
909: $(S^1)'$ is a circle of radius $R' = \alpha'/R$.
910:
911: T-duality symmetry is most easily understood
912: in the case of closed strings, where
913: it amounts to an exchange of winding and momentum modes of the string.
914: The string winding modes on $S^1$ have energy
915: $R |m|/\alpha'$, where
916: $m$ is the winding number. The T-dual momentum modes on $(S^1)'$ have
917: energy $|n|/R'$, where $n$ is the momentum quantum number.
918: These two sets of values coincide when $m$ and $n$ run over all possible
919: integers. It is in fact
920: straightforward to check that the full spectrum of closed string states is
921: unchanged under T-duality.
922: For the case of open strings,
923: T-duality maps an open string with Neumann boundary conditions on
924: $S^1$ to an open string with Dirichlet boundary conditions on
925: $(S^1)'$, and vice versa. Thus, a D$p$-brane which is wrapped around
926: the circle $S^1$ is mapped under T-duality to a D$(p-1)$-brane which
927: is localized to a point on the circle $(S^1)'$. Under T-duality the
928: low-energy $(p + 1)$-dimensional Yang-Mills theory on the $p$-brane is
929: replaced by a $p$-dimensional Yang-Mills theory on the dual
930: $(p-1)$-brane. Mathematically, the covariant derivative operator in
931: the direction $S^1$ is replaced under T-duality with an adjoint scalar
932: field $X^a$. Formally, this adjoint scalar field is an infinite size
933: matrix~\cite{WT-T-duality}, which contains information about the open
934: strings wrapped an arbitrary number of times around the compact
935: direction $(S^1)'$.
936:
937: We can summarize the relevant mappings under T-duality in the
938: following table
939:
940: \begin{center}
941: \begin{tabular}{rcr}
942: IIA/$S^1$ & $\leftrightarrow $ & IIB/$(S^1)'$\\
943: $R$ & $\leftrightarrow $ & $R' =\alpha'/R $\\
944: Dirichlet/Neumann b.c.'s & $\leftrightarrow $ & Neumann/Dirichlet b.c.'s\\
945: $p$-brane & $\leftrightarrow $ & $(p\pm 1)$-brane\\
946: $2 \pi \alpha' (i \partial_a + A_a)$ & $\leftrightarrow $ & $X^a $
947: \end{tabular}
948: \end{center}
949: \vspace*{0.2in}
950:
951: The phenomena by which field strengths in one brane describe lower- or
952: higher-dimensional branes can be easily understood using T-duality.
953: The following simple example may help to clarify this connection.
954: (For a more detailed discussion using this point of view, see
955: Taylor~\cite{WT-Trieste}.)
956: %\begin{figure}
957: %\epsfig{file=dual.eps,width=12cm}
958: %\caption[x]{\footnotesize T-duality takes a diagonal D1-brane on a
959: % two-torus (a) to a D2-brane on the dual torus with
960: constant magnetic
961: % flux encoding an embedded D0-brane (b).}
962: %\label{f:T-duality}
963: %\end{figure}
964:
965:
966: \begin{figure}[!ht]
967: \leavevmode
968: \begin{center}
969: \epsfxsize = 12 cm \epsfbox{dual.eps}
970: \end{center}
971: \caption[x]{\footnotesize T-duality takes a diagonal D1-brane on a
972: two-torus (a) to a D2-brane on the dual torus with
973: constant magnetic
974: flux encoding an embedded D0-brane (b).}
975: \label{f:T-duality}
976: \end{figure}
977:
978:
979:
980: Consider a D1-brane wrapped diagonally on a two-torus $T^2$ with sides
981: of length $L_1 = L$ and $L_2 = 2 \pi R$.
982: (Figure~\ref{f:T-duality}(a)). This configuration is described in
983: terms of the world-volume Yang-Mills theory on a D1-brane stretched in
984: the $L_1$ direction through a transverse scalar field
985: \begin{equation}
986: X^2 = 2 \pi R \xi_1/L\,.
987: % % \label{eq:}
988: \end{equation}
989: To be technically precise, this scalar field should be treated as an
990: $\infty \times \infty$ matrix~\cite{WT-T-duality} whose $(n, m)$ entry
991: is associated with strings that connect the $n$th and $m$th images of the
992: D1-brane on the covering space of $S^1$. The diagonal elements
993: $X^2_{n,n}$ of this infinite matrix are given by $2 \pi R (\xi_1 +
994: n L)/L$, while all off-diagonal elements vanish. While the resulting
995: matrix-valued function of $\xi_1$ is not periodic, it is periodic up
996: to a gauge transformation
997: \begin{equation}
998: X^2 (L) = V X^2 (0) V^{-1}
999: \label{eq:boundary-1}
1000: \end{equation}
1001: where $V$ is the shift matrix with nonzero elements $V_{n, n + 1} = 1$.
1002:
1003: Under T-duality
1004: in the $x^2$ direction
1005: the infinite matrix $X^2_{nm}$ becomes the Fourier mode representation of a
1006: gauge field on a dual D2-brane:
1007: \begin{equation}
1008: A_2 = \frac{1}{ R' L} \xi_1\,.
1009: \label{eq:a2}
1010: \end{equation}
1011: The magnetic flux associated with this gauge field is
1012: \begin{equation}
1013: F_{12} = \frac{1}{ R' L}\,,
1014: % % \label{eq:}
1015: \end{equation}
1016: so that
1017: \begin{equation}
1018: \frac{1}{2 \pi} \int F_{12} \; d \xi^1 \, d \xi^2 = 1\,.
1019: \label{eq:0-charge}
1020: \end{equation}
1021: Note that the boundary condition (\ref{eq:boundary-1}) on the
1022: infinite matrix $X^2$ transforms under T-duality to the boundary
1023: condition on the gauge field
1024: \begin{eqnarray}
1025: A_2 (L, x_2) & = &
1026: e^{2 \pi i \xi_2/L_2'}
1027: \left(A_2 (0, x_2) + i \partial_2 \right)
1028: e^{-2 \pi i \xi_2/L_2'}\\
1029: & = &%e^{2 \pi i \xi_2/L_2'}
1030: A_2 (0, x_2)
1031: %e^{-2 \pi i \xi_2/L_2'}
1032: + \frac{2 \pi}{ L_2'}, \nonumber
1033: \end{eqnarray}
1034: which (\ref{eq:a2}) clearly satisfies.
1035: The off-diagonal elements of the shift matrix $V$ in
1036: (\ref{eq:boundary-1}) describe winding modes which correspond after
1037: T-duality to the first Fourier mode $e^{2 \pi i \xi_2/L_2'}$. The
1038: boundary condition on the gauge fields in the $\xi_2$ direction is
1039: trivial, which simplifies the T-duality map; a similar construction
1040: can be done with a nontrivial boundary condition in both directions,
1041: although the configuration looks more complicated in the D1-brane
1042: picture.
1043:
1044: This construction gives a simple Yang-Mills description of the mapping
1045: of D-brane charges under T-duality: the initial configuration
1046: described above has charges associated with a single D1-brane wrapped
1047: around each of the directions of the 2-torus: D$1_1 +$ D$1_2$. Under
1048: T-duality, these D1-branes are mapped to a D2-brane and a D0-brane
1049: respectively: D$2_{12} +$ D$0$. The flux integral (\ref{eq:0-charge})
1050: is the representation in the D2-brane world-volume Yang-Mills theory
1051: of the charge associated with a D0-brane which has been uniformly
1052: distributed over the surface of the D2-brane, just as in
1053: (\ref{eq:p-2-charge}).
1054:
1055: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1056: % tachyons and D-branes: WT
1057: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1058:
1059:
1060: \section{Tachyons and D-branes}
1061: \label{sec:tachyon-D-branes}
1062:
1063:
1064:
1065:
1066: We now turn to the subject of tachyons. Certain D-brane
1067: configurations are unstable, both in supersymmetric and
1068: nonsupersymmetric string theories. This instability is manifested as
1069: a tachyon, that is, as
1070: a state with $M^2 < 0$ in the spectrum of open strings that end on the
1071: D-brane. We will explicitly describe the tachyonic mode in the case
1072: of the open bosonic string in Section \ref{sec:bosonic-string}; this
1073: open bosonic string tachyon will be the focal point of most of the
1074: developments described in these notes. In this section we list some
1075: elementary D-brane configurations where tachyons arise, and we
1076: describe a particular situation in which the tachyon can be seen in
1077: the low-energy Yang-Mills description of the D-branes. This
1078: Yang-Mills background with a tachyon provides a simple field-theory
1079: model of a system analogous to the more complicated string field
1080: theory tachyon we describe in the later part of these notes. This
1081: simpler model may be useful to keep in mind in the later analysis.
1082:
1083:
1084: \subsection{D-brane configurations with tachyonic instabilities}
1085: \label{sec:D-brane-tachyons}
1086:
1087: Some simple examples of unstable D-brane configurations where the open
1088: string contains a tachyon include the following:
1089: \vspace*{0.1in}
1090:
1091: {\bf Brane-antibrane:} A pair of parallel D$p$-branes with opposite
1092: orientation in type IIA or IIB string theory which are separated by a
1093: distance $d \ll l_s$ give rise to a tachyon in the spectrum of open
1094: strings stretched between the branes~\cite{Banks-Susskind}.
1095: The difference in orientation of the branes means that the two branes are
1096: really a brane and antibrane, carrying equal but opposite RR charges.
1097: Since the net RR charge is 0, the brane and antibrane can annihilate,
1098: leaving an uncharged vacuum configuration.
1099: \vspace*{0.05in}
1100:
1101: {\bf Wrong-dimension branes:} In type IIA/IIB string theory, a
1102: D$p$-brane of even/odd spatial dimension $p$ is a stable BPS state
1103: with nonzero RR charge. On the other hand, a D$p$-brane of the
1104: {\it wrong} dimension ({\it i.e.,} odd/even for IIA/IIB) carries no
1105: charges under the classical IIA/IIB supergravity fields, and has a
1106: tachyon in the open string spectrum. Such a brane can decay into
1107: the vacuum without violating charge conservation.
1108: \vspace*{0.05in}
1109:
1110: {\bf Bosonic D-branes:} Like the wrong-dimension branes of IIA/IIB
1111: string theory, a D$p$-brane of any dimension in the bosonic string
1112: theory carries no conserved charge and has a tachyon in the open
1113: string spectrum. Again, such a brane can decay into the vacuum
1114: without violating charge conservation.
1115:
1116:
1117:
1118:
1119: \subsection{Example: tachyon in low-energy field theory of two D-branes}
1120: \label{sec:example-SYM}
1121:
1122: In order to illustrate the physical
1123: behavior of tachyonic configurations, we consider in this subsection a
1124: simple example~\cite{Hashimoto-Taylor,gns} where a brane-antibrane tachyon
1125: can be seen in the context of the low-energy Yang-Mills theory.
1126:
1127: The system we want to consider is a simple generalization of the (D2 +
1128: D0)-brane configuration we described using Yang-Mills theory in
1129: section 2.4. Consider a pair of D2-branes wrapped on a two-torus, one
1130: of which has a D0-brane embedded in it as a constant positive magnetic
1131: flux, and the other of which has an anti-D0-brane within it described
1132: by a constant negative magnetic flux. We take the two dimensions of
1133: the torus to be $L_1, L_2$. Following the discussion of Section 2.4,
1134: this configuration is equivalent under T-duality in the $L_2$
1135: direction to a pair of crossed D1-branes (see Figure~\ref{f:crossed}).
1136: %\begin{figure}
1137: %\epsfig{file=crossing.eps,width=12cm}
1138: %\caption[x]{\footnotesize A pair of crossed D1-branes, T-dual to a
1139: % pair of D2-branes with uniformly embedded D0- and anti-D0-branes.}
1140: %\label{f:crossed}
1141: %\end{figure}
1142:
1143:
1144: \begin{figure}[!ht]
1145: \leavevmode
1146: \begin{center}
1147: \epsfxsize = 12 cm \epsfbox{crossing.eps}
1148: \end{center}
1149: \caption[x]{\footnotesize A pair of crossed D1-branes, T-dual to a
1150: pair of D2-branes with uniformly embedded D0- and anti-D0-branes.}
1151: \label{f:crossed}
1152: \end{figure}
1153:
1154:
1155:
1156:
1157:
1158:
1159:
1160:
1161: The Born-Infeld energy of this configuration is
1162: \begin{eqnarray}
1163: E_{{\rm BI}} & = & 2 \sqrt{(\tau_2L_1 L_2)^2 + \tau_0^2} \nonumber\\
1164: & = & \frac{1}{g} \left[
1165: \frac{2L_1 L_2}{\sqrt{2 \pi}} + \frac{(2 \pi)^{3/2}}{L_1 L_2} +
1166: \cdots \right]\,, \label{eq:energy-bi}
1167: \end{eqnarray}
1168: in units where $2 \pi \alpha' = 1$.
1169: This can be computed either directly from the Born-Infeld action on
1170: the D2-branes (the abelian theory can be used since the matrices are
1171: diagonal), or by simply using the Pythagorean theorem in the T-dual
1172: D1-brane picture.
1173: The second term in the last line
1174: corresponds to the Yang-Mills approximation. In this approximation
1175: (dropping the D2-brane energy) the energy is
1176: \begin{equation}
1177: E_{{\rm YM}} = \frac{\tau_2}{4} \int {\rm Tr}\; F_{\alpha \beta}
1178: F^{\alpha \beta} = \frac{1}{4 \sqrt{2 \pi} g} \int {\rm Tr}\;
1179: F_{\alpha \beta}
1180: F^{\alpha \beta}\,.
1181: % % \label{eq:}
1182: \end{equation}
1183:
1184: We are interested in studying this configuration in the Yang-Mills
1185: approximation, in which we have a $U(2)$ theory on $T^2$ with field
1186: strength
1187: \begin{equation}
1188: F_{12} = \left(\begin{array}{cc}
1189: \frac{2 \pi}{ L_1 L_2} & 0\\
1190: 0 &-\frac{2 \pi}{ L_1 L_2}
1191: \end{array}\right)
1192: = \frac{2 \pi}{ L_1 L_2} \tau_3 \,.
1193: % % \label{eq:}
1194: \end{equation}
1195: This field strength can be realized as the curvature of a linear gauge
1196: field
1197: \begin{equation}
1198: A_1 = 0, \;\;\;\;\;
1199: A_2 =\frac{2 \pi}{ L_1 L_2} \xi\tau_3\,,
1200: \label{eq:unstable-background}
1201: \end{equation}
1202: which satisfies the boundary conditions
1203: \begin{equation}
1204: A_j (L, \xi_2) = \Omega (i \partial_j+ A_j (0, \xi_2)) \Omega^{-1} \,,
1205: \label{eq:bc-tachyon}
1206: \end{equation}
1207: where
1208: \begin{equation}
1209: \Omega = e^{2 \pi i ( \xi_1/L_2) \tau_3} \,.
1210: % % \label{eq:}
1211: \end{equation}
1212:
1213: It is easy to check that this configuration indeed satisfies
1214: \begin{equation}
1215: E_{{\rm YM}} = \frac{1}{2g} \frac{(2 \pi)^{3/2}}{L_1 L_2} {\rm Tr}\;
1216: \tau_3^2 = \frac{1}{g} \frac{(2 \pi)^{3/2}}{L_1 L_2}\,,
1217: \label{eq:unstable-energy}
1218: \end{equation}
1219: as desired from (\ref{eq:energy-bi}).
1220: Since
1221: \begin{equation}
1222: {\rm Tr}\; F_{\alpha \beta} = 0,
1223: % % \label{eq:}
1224: \end{equation}
1225: the gauge field we are considering is in the same topological
1226: equivalence class as $F = 0$. This corresponds to the fact that the
1227: D0-brane and anti-D0-brane can annihilate. To understand the
1228: appearance of the tachyon, we can consider the spectrum of excitations
1229: $\delta A_\alpha$ around the background
1230: (\ref{eq:unstable-background})~\cite{Hashimoto-Taylor}. The
1231: eigenvectors of the quadratic mass terms in this background are
1232: described by torus theta functions which satisfy boundary
1233: conditions related to (\ref{eq:bc-tachyon}). There are precisely two
1234: elements in the spectrum with the negative eigenvalue $-4 \pi/L_1
1235: L_2$. These theta functions
1236: ~\cite{Hashimoto-Taylor} are tachyonic modes of the theory which
1237: are associated with the annihilation of the positive and negative
1238: fluxes that encode the D0- and anti-D0-brane. These tachyonic modes are
1239: perhaps easiest to understand in the dual configuration, where they
1240: provide a direction of instability in which the two crossed D1-branes
1241: reconnect as in Figure~\ref{f:instability}.
1242: %\begin{figure}
1243: %\epsfig{file=unstable.eps,width=12cm}
1244: %\caption[x]{\footnotesize The brane-antibrane instability of a
1245: % D0-D$\bar{0}$ system embedded in two D2-branes, as seen in the
1246: % T-dual D1-brane picture.}
1247: %\label{f:instability}
1248: %\end{figure}
1249:
1250: \begin{figure}[!ht]
1251: \leavevmode
1252: \begin{center}
1253: \epsfxsize = 12 cm \epsfbox{unstable.eps}
1254: \end{center}
1255: \caption[x]{\footnotesize The brane-antibrane instability of a
1256: D0-D$\bar{0}$ system embedded in two D2-branes, as seen in the
1257: T-dual D1-brane picture.}
1258: \label{f:instability}
1259: \end{figure}
1260:
1261:
1262:
1263:
1264: It is also interesting to note that in
1265: the T-dual picture the
1266: tachyonic modes of the gauge field have support which is localized near
1267: the two intersection points and take the off-diagonal form
1268: \begin{equation}
1269: \delta A_t \sim \left(\begin{array}{cc}
1270: 0 & \star\\
1271: \star & 0
1272: \end{array} \right) \,,
1273: % % \label{eq:}
1274: \end{equation}
1275: which naturally encodes our geometric
1276: understanding that the tachyonic modes reconnect the two D1-branes near
1277: each intersection point.
1278:
1279: The full Yang-Mills action around the background
1280: (\ref{eq:unstable-background}) can be written as a quartic function of
1281: the mass eigenstates around this background. Written in terms of
1282: these modes, there are nontrivial cubic and quartic terms which couple
1283: the tachyonic modes to all the massive modes in the system. If we
1284: integrate out the massive modes, we know from the topological
1285: reasoning above that an effective potential arises for the tachyonic
1286: mode $A_t$, with a maximum value of (\ref{eq:unstable-energy}) and a
1287: minimum value of 0. This system is highly analogous to the bosonic
1288: open string tachyon we will discuss in the remainder of these
1289: lectures. Our current understanding of the bosonic string through
1290: bosonic string field theory is analogous to that of someone who only
1291: knows the Yang-Mills theory around the background
1292: (\ref{eq:unstable-background}) in terms of a complicated quartic
1293: action for an infinite family of modes. Without knowledge of the
1294: topological structure of the theory, and given only a list of the
1295: coefficients in the quartic action, such an individual would have to
1296: systematically calculate the tachyon effective potential by explicitly
1297: integrating out all the massive modes one by one. This would give a
1298: numerical approximation to the minimum of the effective potential,
1299: which could be made arbitrarily good by raising the mass of the cutoff
1300: at which the effective action is computed. It may be helpful to keep
1301: this example
1302: in mind in the following sections, where an
1303: analogous tachyonic system is considered in string field theory. For
1304: further discussion of this unstable configuration in Yang-Mills
1305: theory, see the
1306: references~\cite{Hashimoto-Taylor,gns,Morosov,Hashimoto-Nagaoka}.
1307:
1308:
1309: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1310: % Sen conjectures: WT
1311: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1312:
1313: \subsection{The Sen conjectures}
1314: \label{conjectures}
1315:
1316: The existence of the tachyonic mode in the open bosonic string
1317: indicates that the standard choice of perturbative vacuum for this
1318: theory is unstable. In the early days of the subject, there were some
1319: calculations suggesting that this tachyon could condense, leading to a
1320: more stable vacuum~\cite{Bardakci-tachyon}. Kostelecky and Samuel
1321: argued early on that the stable vacuum could be identified in string
1322: field theory in a systematic way ~\cite{ks-open}, however there was no
1323: clear physical picture for the significance of this stable vacuum. In
1324: 1999, Ashoke Sen reconsidered the problem of tachyons in string field
1325: theory. Sen suggested that the open bosonic string should really be
1326: thought of as living on a D25-brane, and hence that the perturbative
1327: vacuum for this string theory should have a nonzero vacuum energy
1328: associated with the tension of this D25-brane. He suggested that the
1329: tachyon is simply the instability mode of the D25-brane, which carries
1330: no conserved charge and hence is not expected to be stable, as
1331: discussed in section~3.1. More precisely, Sen conjectured that the
1332: following three statements are true~\cite{Sen-universality}:
1333: \begin{enumerate}
1334: \item The tachyon potential has a locally stable minimum, whose energy
1335: density $\mathcal{E}$, measured with respect to that of the
1336: unstable critical
1337: point, is equal to minus the tension of the D25-brane:
1338: \begin{equation}
1339: \mathcal{E} = -T_{25}\,.
1340: \end{equation}
1341: \item Lower-dimensional D-branes are solitonic solutions
1342: of the string theory on the background of a D25-brane.
1343:
1344: \item The locally stable vacuum of the system is the closed string vacuum.
1345: In this vacuum the D25-brane is absent and no conventional
1346: open string excitations exist.
1347: \end{enumerate}
1348:
1349: It was also suggested by Sen that open string field theory
1350: was a natural setup to test the above conjectures. He sharpened the
1351: first conjecture by suggesting that Witten's OSFT should precisely
1352: reproduce the tension of the D25-brane, which he expressed
1353: in terms of the open string coupling constant $g$ which appears
1354: in the formulation of open string field theory:
1355: \begin{equation}
1356: T_{25}= \frac{1}{2 \pi^2 g^2} \,.
1357: \end{equation}
1358: We will give the instructive derivation of this result in
1359: section~\ref{sec:evidence}.
1360:
1361: Our first encounter with the tachyon conjectures will happen in
1362: section~\ref{sec:CFT}, where we calculate the first nontrivial term in
1363: the tachyon potential, find a minimum, and discover that, even with
1364: this rough approximation, the calculated $\mathcal{E}$ gives about
1365: 70\% of the expected answer. In Section 7 of these lectures we
1366: systematically explore the evidence for these conjectures in Witten's
1367: OSFT. First, however, we need to develop the technical tools to do
1368: specific calculations in string field theory.
1369:
1370: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1371: % Witten cubic OSFT: both
1372: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1373:
1374:
1375: \section{Witten's cubic open string field theory}
1376: \label{sec:SFT}
1377:
1378: In this and in the following two sections we give a detailed
1379: description
1380: of Witten's open string field theory~\cite{Witten-SFT}. This
1381: section contains a general introduction to this string field theory.
1382: Subsection \ref{sec:bosonic-string} reviews the quantization of the
1383: open bosonic string in 26 dimensions and sets notation. Subsection
1384: \ref{sec:Witten-SFT} gives a heuristic introduction to open string field
1385: theory, which follows Witten's original paper.
1386: In subsection \ref{sec:algebraic-structure} we discuss the
1387: algebraic structure of open string field theory
1388: which emerges naturally in the context of conformal field
1389: theory. This discussion also develops the properties of the twist
1390: operator $\Omega$ which reverses the orientation of open strings.
1391:
1392: The work in the present section prepares the ground for sections
1393: \ref{sec:CFT} and \ref{sec:overlaps}, in which
1394: precise definitions of the open bosonic SFT are given
1395: using conformal field theory and the mode decomposition of overlap
1396: equations.
1397: For further background on Witten's OSFT see the reviews of LeClair,
1398: Peskin and Preitschopf~\cite{lpp}, of Thorn~\cite{Thorn}, and of
1399: Gaberdiel and Zwiebach~\cite{Gaberdiel-Zwiebach}.
1400:
1401: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1402: % bosonic open string: WT
1403: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1404:
1405: \subsection{The bosonic open string}
1406: \label{sec:bosonic-string}
1407:
1408: In this subsection we review the quantization of the open bosonic
1409: string. For further details see the textbooks by Green, Schwarz, and
1410: Witten~\cite{gsw} and by Polchinski~\cite{Polchinski-string}.
1411: The bosonic open string can be quantized using the BRST
1412: approach starting from the action
1413: \begin{equation}
1414: S = -\frac{1}{4 \pi \alpha'} \int \sqrt{-\gamma} \gamma^{ab}
1415: \partial_a X^\mu \partial_b X_\mu,
1416: \label{eq:string-action}
1417: \end{equation}
1418: where $\gamma$ is an auxiliary dynamical metric on the world-sheet.
1419: This action can be gauge-fixed to conformal gauge $\gamma_{ab} \sim
1420: \delta_{ab}$, introducing at the same time
1421: ghost and antighost fields $c^{\pm} (\sigma), b_{\pm \pm} (\sigma)$. The
1422: gauge-fixed action is
1423: \begin{equation}
1424: S = -\frac{1}{4 \pi \alpha'} \int
1425: \partial_a X^\mu \partial^a X_\mu
1426: + \frac{1}{ \pi} \int \left( b_{++} \partial_-c^+ + b_{--} \partial_+
1427: c^-\right)\,.
1428: \label{eq:gauge-fixed-action}
1429: \end{equation}
1430:
1431: The matter fields $X^\mu$ can be expanded in modes using
1432: \begin{equation}
1433: X^\mu (\sigma, \tau) = x_0^\mu + 2 p^\mu \tau +
1434: i\sqrt{2} \sum_{n \neq 0}
1435: \frac{1}{n} \alpha^\mu_n \cos (n \sigma) e^{-in \tau}\,,
1436: \label{eq:mode-decomposition}
1437: \end{equation}
1438: where we have fixed $l_s = \sqrt{2 \alpha'} = \sqrt{2}$, so that
1439: $\alpha' = 1$. In the quantum theory, $x_0^\mu$ and $p^\mu$ obey the
1440: canonical commutation relations
1441: \begin{equation}
1442: [x^\mu_0, p^\nu] = i \eta^{\mu \nu}\,.
1443: % \label{eq:}
1444: \end{equation}
1445: The $\alpha^\mu_n$'s with negative/positive
1446: values of $n$ become raising/lowering operators for the oscillator
1447: modes on the string. They satisfy the Hermiticity conditions
1448: $(\alpha^\mu_n)^{\dagger} = \alpha^\mu_{-n}$
1449: and the
1450: commutation relations
1451: \begin{equation}
1452: [\alpha^\mu_m, \alpha^\nu_n] = m \eta^{\mu \nu} \delta_{m + n, 0}\,.
1453: % \label{eq:}
1454: \end{equation}
1455: We will often use the canonically normalized oscillators:
1456: \begin{equation}
1457: a^\mu_n = \frac{1}{ \sqrt{n}} \alpha^{\mu}_n\,, \quad n\geq 1 \,,
1458: % \label{eq:}
1459: \end{equation}
1460: which obey the commutation relations
1461: \begin{equation}
1462: [a^\mu_m, a^{\nu\dagger}_n] = \eta^{\mu \nu} \delta_{m,n}\,.
1463: % \label{eq:}
1464: \end{equation}
1465: %The raising and lowering operators satisfy $(\alpha^\mu_n)^{\dagger} =
1466: %\alpha^\mu_{-n}, (a^\mu_n)^{\dagger} =
1467: %a^\mu_{-n}$.
1468: We will also frequently use position modes $x_n$ for $n \neq 0$ and
1469: lowering and raising operators $a_0, a^{\dagger}_0$ for the zero
1470: modes. These are related to the modes in
1471: (\ref{eq:mode-decomposition}) through (dropping space-time indices)
1472: \begin{eqnarray}
1473: x_n & = & \frac{i}{ \sqrt{2n}} (a_n- a^{\dagger}_n) \\
1474: x_0 & = & \frac{i}{2} (a_0-a_0^{\dagger}) \nonumber
1475: \end{eqnarray}
1476:
1477: The ghost and antighost fields can be decomposed into modes through
1478: \begin{eqnarray}
1479: c^{\pm} (\sigma) & = & \sum_{n}c_n e^{\mp in \sigma} \\
1480: b_{\pm \pm} (\sigma) & = & \sum_{n}b_n e^{\mp in \sigma} \,.
1481: \nonumber
1482: \end{eqnarray}
1483: The ghost and antighost modes satisfy the anticommutation relations
1484: \begin{eqnarray}
1485: \{c_n, b_m\} & = & \delta_{n + m, 0}\\
1486: \{c_n, c_m\} & = & \{b_n, b_m\} = 0\,. \nonumber
1487: \end{eqnarray}
1488:
1489: A general state in the open string Fock space can be written in the
1490: form
1491: \begin{equation}
1492: \alpha^{\mu_1}_{-n_1} \cdots \alpha^{\mu_i}_{-n_i} \;
1493: c_{-m_1} \cdots c_{-m_j} \;
1494: b_{-p_1} \cdots b_{-p_l} \; | 0; k \rangle
1495: % \label{eq:}
1496: \end{equation}
1497: where
1498: \begin{equation}
1499: n_i \geq 1\,, \quad m_i \geq -1\,, \quad \hbox{and} \quad p_i \geq 2\,,
1500: \end{equation}since $| 0; k\rangle$ is
1501: annihilated by
1502: \begin{eqnarray}
1503: b_n | 0; k \rangle & = & 0, \; \; \; n \geq -1\, \nonumber\\
1504: c_n | 0; k \rangle & = & 0, \; \; \; n \geq 2\,, \\
1505: \alpha^\mu_{n} | 0; k \rangle & = & 0, \; \; \; n \geq 1 \,.\nonumber
1506: \end{eqnarray}
1507: The state $|0;k\rangle$ is a momentum eigenstate:
1508: \begin{equation}
1509: p^\mu | 0; k \rangle = k^\mu | 0; k \rangle\,.
1510: % \label{eq:}
1511: \end{equation}
1512: The zero-momentum state $| 0; 0 \rangle$ is the SL(2,R) invariant
1513: vacuum; we will often write it simply as $| 0 \rangle$. This vacuum is
1514: defined to have ghost number 0, and it is normalized by the equation
1515: \begin{equation}
1516: \label{conventionalnormalization}
1517: \langle 0; k | c_{-1} c_0c_1 | 0; k' \rangle =
1518: (2 \pi)^{26}
1519: \delta (k-k')
1520: % \label{eq:}
1521: \end{equation}
1522: For string field theory we will also find it
1523: convenient to work with the vacua of ghost number one and two:
1524: \begin{eqnarray}
1525: G = 1: & \hspace*{0.1in} & | 0_1 \rangle = c_1| 0 \rangle\\
1526: G = 2: & \hspace*{0.1in} & | 0_2 \rangle = c_0 c_1| 0 \rangle\,. \nonumber
1527: \end{eqnarray}
1528: In the notation of Polchinski~\cite{Polchinski-string}, these two vacua
1529: are written as
1530: \begin{eqnarray}
1531: | 0_1 \rangle & = & | 0 \rangle_m \otimes |\! \downarrow \rangle \nonumber\\
1532: | 0_2 \rangle & = & | 0 \rangle_m \otimes |\! \uparrow \rangle \,,
1533: \end{eqnarray}
1534: where $| 0 \rangle_m$ is the matter vacuum and $|\!\downarrow\rangle,
1535: |\!\uparrow\rangle$ are the ghost vacua annihilated by $b_0, c_0$.
1536:
1537:
1538: The BRST operator of this theory is given by
1539: \begin{equation}
1540: Q_B = \sum_{n = -\infty}^{\infty} c_nL_{-n}^{({\rm m})}
1541: + \sum_{n, m = -\infty}^{ \infty} \frac{(m-n)}{2}
1542: :c_mc_nb_{-m-n}:-c_0
1543: \label{eq:BRST}
1544: \end{equation}
1545: where the matter Virasoro operators are given by
1546: \begin{equation}
1547: L_q^{({\rm m})} = \left\{
1548: \begin{array}{ll}
1549: \frac{1}{2}\sum_{n}\alpha^\mu_{q-n} \alpha_{\mu \; n}, & q \neq 0\\[1.0ex]
1550: p^2 + \sum_{n = 1}^{ \infty} \alpha^\mu_{-n} \alpha_{\mu
1551: \; n}\,.
1552: \end{array}
1553: \right.
1554: %\label{eq:}
1555: \end{equation}
1556:
1557:
1558: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1559: % Witten approach: WT
1560: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1561:
1562: \subsection{Witten's cubic bosonic SFT}
1563: \label{sec:Witten-SFT}
1564:
1565: The discussion of the previous subsection leads to a systematic
1566: quantization of the open bosonic string in the conformal field theory
1567: framework. Using this approach it is possible, in principle, to
1568: calculate an arbitrary perturbative on-shell scattering amplitude for
1569: physical string states. To study tachyon condensation in string
1570: theory, however, we require a nonperturbative, off-shell formalism for
1571: the theory--- a string field theory.
1572:
1573: A very simple form for the
1574: off-shell open bosonic string field theory action was proposed by
1575: Witten in 1986:~\cite{Witten-SFT}
1576: \begin{equation}
1577: S = -\frac{1}{2}\int \Psi \star Q \Psi -\frac{g}{3} \int \Psi \star
1578: \Psi \star \Psi\,.
1579: \label{eq:SFT-action}
1580: \end{equation}
1581: This action has the general form of a Chern-Simons theory on a
1582: 3-manifold, although for string field theory there is no explicit
1583: interpretation of the integration in terms of a concrete 3-manifold.
1584: In Eq.~(\ref{eq:SFT-action}), $g$ is interpreted as the (open) string
1585: coupling constant. The field $\Psi$ is a string field, which takes
1586: values in a graded algebra ${\mathcal A}$. Associated with the algebra
1587: ${\mathcal A}$ there is a star product
1588: \begin{equation}
1589: \star:{\mathcal A} \otimes{\mathcal A} \rightarrow{\mathcal A}, \;\;\;\;\;
1590: % \label{eq:}
1591: \end{equation}
1592: under which the degree $G$ is additive ($G_{\Psi \star \Phi} = G_\Psi
1593: + G_\Phi$). There is also a BRST operator
1594: \begin{equation}
1595: Q:{\mathcal A} \rightarrow{\mathcal A}, \;\;\;\;\;
1596: % \label{eq:}
1597: \end{equation}
1598: of degree one ($G_{Q \Psi} = 1 + G_\Psi$). String fields can be
1599: integrated using
1600: \begin{equation}
1601: % blackboard
1602: \int:{\mathcal A} \rightarrow {\bf C}\,.
1603: % \label{eq:}
1604: \end{equation}
1605: This integral vanishes for all $\Psi$ with degree $G_\Psi \neq 3$.
1606: Thus, the action (\ref{eq:SFT-action}) is only nonvanishing for a
1607: string field $\Psi$ of degree 1.
1608:
1609: The elements $Q, \star, \int$ that define the string field theory are
1610: assumed to satisfy the following axioms:
1611: \vspace*{0.15in}
1612:
1613: \noindent {\bf (a)} Nilpotency of $Q$: $\;Q^2 \Psi = 0, \;\; \; \forall \Psi
1614: \in{\mathcal A}$.
1615: \vspace*{0.08in}
1616:
1617: \noindent {\bf (b)} $\int Q\Psi = 0, \; \; \; \forall \Psi \in{\mathcal A}$.
1618: \vspace*{0.08in}
1619:
1620: \noindent {\bf (c)} Derivation property of $Q$:\\
1621: \hspace*{0.4in}$\;Q (\Psi \star \Phi) = (Q \Psi) \star \Phi +
1622: (-1)^{G_\Psi} \Psi \star (Q \Phi), \; \; \forall \Psi, \Phi \in{\mathcal A}$.
1623: \vspace*{0.08in}
1624:
1625: \noindent {\bf (d)} Cyclicity: $\;\int \Psi \star \Phi = (-1)^{G_\Psi
1626: G_\Phi} \int \Phi \star \Psi, \; \; \; \forall \Psi, \Phi \in{\mathcal A}$.
1627: \vspace*{0.08in}
1628:
1629: \noindent {\bf (e)} Associativity: $(\Phi \star \Psi) \star \Xi =
1630: \Phi \star (\Psi \star \Xi), \; \; \;
1631: \forall \Phi, \Psi, \Xi \in{\mathcal A}$.
1632: \vspace*{0.15in}
1633:
1634: When these axioms are satisfied, the action (\ref{eq:SFT-action}) is
1635: invariant under the gauge transformations
1636: \begin{equation}
1637: \delta \Psi = Q \Lambda + \Psi\star \Lambda - \Lambda \star \Psi\,,
1638: \label{eq:SFT-gauge}
1639: \end{equation}
1640: for any gauge parameter $\Lambda \in{\mathcal A}$ with degree 0.
1641:
1642: When the string coupling $g$ is taken to vanish, the equation of
1643: motion for the theory defined by (\ref{eq:SFT-action}) simply becomes
1644: $Q \Psi = 0$, and the gauge transformations (\ref{eq:SFT-gauge})
1645: simply become
1646: \begin{equation}
1647: \delta \Psi = Q \Lambda\,.
1648: %\label{eq:}
1649: \end{equation}
1650: Thus, when $g = 0$ this string field theory gives precisely the structure
1651: needed to describe the free bosonic string. The motivation for
1652: introducing the extra structure in (\ref{eq:SFT-action}) was to
1653: find a simple interacting extension of the free theory, consistent
1654: with the perturbative expansion of open bosonic string theory.
1655:
1656: Witten presented this formal structure and argued
1657: that all the needed axioms are satisfied when ${\mathcal A}$ is taken to
1658: be the space of string fields
1659: \begin{equation}
1660: {\mathcal A} =\{\Psi[x (\sigma); c (\sigma), b
1661: (\sigma)]\}
1662: \label{eq:string-functionals}
1663: \end{equation}
1664: which can be described as functionals of the matter, ghost and
1665: antighost fields describing an open string in 26 dimensions with $0
1666: \leq \sigma \leq \pi$. Such a string field can be written as a formal
1667: sum over open string Fock space states with coefficients given by an
1668: infinite family of space-time fields
1669: \begin{equation}
1670: \Psi =
1671: \int d^{26}p \;
1672: \left[ \phi (p)\; | 0_1; p \rangle + A_\mu (p) \; \alpha^\mu_{-1} | 0_1; p
1673: \rangle + \cdots \right]
1674: \label{eq:field-expansion}
1675: % \label{eq:}
1676: \end{equation}
1677: Each Fock space state is associated with a given string functional,
1678: just as the states of a harmonic oscillator are associated with
1679: wavefunctions of a particle in one dimension. For example, the matter
1680: ground state $| 0 \rangle_m$ annihilated by $a_n$ for all $n \geq 1$
1681: is associated (up to a constant $C$) with the functional of matter
1682: modes
1683: \begin{equation}
1684: | 0 \rangle_m \rightarrow
1685: C \exp \left( -\frac{1}{2}\sum_{n > 0}^{ \infty}
1686: nx_n^2 \right)\,.
1687: %\label{eq:}
1688: \end{equation}
1689:
1690: For Witten's cubic string field theory, the BRST operator $Q$ in
1691: (\ref{eq:SFT-action}) is the usual open string BRST operator $Q_B$,
1692: given in (\ref{eq:BRST}), and the degree associated with a Fock space
1693: state is the
1694: ghost number of that state.
1695: The star product $\star$ acts on a pair
1696: of functionals $\Psi, \Phi$ by gluing the right half of one string to
1697: the left half of the other using a delta function interaction
1698:
1699: \begin{center}
1700: \begin{picture}(100,60)(- 50,- 30)
1701: \put(-40,20){\line(1,0){37}}
1702: \put(40,20){\line(-1,0){37}}
1703: \put(-3,20){\line( 0, -1){ 37}}
1704: \put(3,20){\line( 0, -1){ 37}}
1705: \put(-20,2){\makebox(0,0){$\Psi$}}
1706: \put(20, 2){\makebox(0,0){$\Phi$}}
1707: %\put(0,-25){\makebox(0,0){$\delta$}}
1708: \end{picture}
1709: \end{center}
1710:
1711: This star product factorizes into separate matter and ghost parts.
1712: In the matter sector, the star product is given by the formal functional
1713: integral
1714: \begin{eqnarray}
1715: \lefteqn{\left(\Psi \star \Phi\right) [z(\sigma)]} \label{eq:mult} \\ &
1716: \equiv&
1717: \int
1718: \prod_{{0} \leq \tilde{\tau} \leq {\pi\over 2}} dy(\tilde{\tau}) \; dx
1719: (\pi -\tilde{\tau})
1720: \prod_{{\pi\over 2} \leq
1721: \tau \leq \pi}
1722: \delta[x(\tau)-y(\pi-\tau)]
1723: \; \Psi [x(\tau)] \Phi [y(\tau)]\, ,\nonumber\\
1724: & &
1725: \hspace*{1.2in}x(\tau) = z(\tau) \quad {\rm for} \quad {0} \leq \tau \leq
1726: {\pi\over 2}\, ,
1727: \nonumber\\
1728: & &
1729: \hspace*{1.2in}y(\tau) = z(\tau)\quad {\rm for} \quad {\pi\over 2} \leq
1730: \tau \leq \pi\, .
1731: \nonumber
1732: \end{eqnarray}
1733: Similarly, the integral over a string field factorizes into matter and
1734: ghost parts, and in the matter sector is given by
1735: \begin{equation}
1736: \int \Psi = \int \prod_{0 \leq \sigma \leq \pi} dx (\sigma) \;
1737: \prod_{0 \leq
1738: \tau \leq \frac{\pi}{2} }
1739: \delta[x(\tau)-x(\pi-\tau)] \;\Psi[x (\tau)]\,.
1740: \label{eq:integral-p}
1741: \end{equation}
1742: This corresponds to gluing the left and right halves of the string
1743: together with a delta function interaction
1744:
1745: \begin{center}
1746: \begin{picture}(100,60)(- 50,- 30)
1747: \put(-3,20){\line(1,0){6}}
1748: \put(-3,20){\line( 0, -1){ 37}}
1749: \put(3,20){\line( 0, -1){ 37}}
1750: %\put(0,-25){\makebox(0,0){$\delta$}}
1751: %\put(0,27){\makebox(0,0){$\Psi$}}
1752: \put(10,-3){\makebox(0,0){$\Psi$}}
1753: \end{picture}
1754: \end{center}
1755:
1756:
1757: The ghost sector of the theory is defined in a similar fashion, but
1758: has an anomaly due to the curvature of the Riemann surface that describes
1759: the three-string vertex. The ghost sector can be described either in
1760: terms of fermionic ghost fields $c (\sigma), b (\sigma)$ or through
1761: bosonization in terms of a single bosonic scalar field $\phi_g
1762: (\sigma)$. From the functional point of view of Eqs.~(\ref{eq:mult},
1763: \ref{eq:integral-p}), it is easiest to describe the ghost sector in the
1764: bosonized language. In this language, the ghost fields $b (\sigma)$
1765: and $c (\sigma)$ are replaced by the scalar field $\phi_g (\sigma)$, and
1766: the star product in the ghost
1767: sector is given by (\ref{eq:mult}) with an extra insertion of $\exp
1768: (3i \phi_g (\pi/2)/2)$ inside the integral. Similarly, the integration
1769: of a string field in the ghost sector is given by (\ref{eq:integral-p})
1770: with an insertion of $\exp (-3i \phi_g (\pi/2)/2)$ inside the integral.
1771: Witten first described the cubic string field theory using
1772: bosonized ghosts. While this approach is useful for some purposes, we
1773: will use fermionic ghost fields in the remainder of these lecture notes.
1774: With the fermionic ghosts, there is no need for insertions at the
1775: string midpoint.
1776:
1777: The expressions (\ref{eq:mult}, \ref{eq:integral-p}) may seem rather
1778: formal, as they are written in terms of functional integrals. These
1779: expressions, however, can be given precise meaning when described in
1780: terms of creation and annihilation operators acting on the string Fock
1781: space. In Sections \ref{sec:CFT} and
1782: \ref{sec:overlaps} we give a more precise definition of the string
1783: field theory action using conformal field theory and the countable
1784: mode decomposition of the string.
1785:
1786:
1787: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1788: % algebraic structure: BZ
1789: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1790:
1791: \subsection{Algebraic structure of OSFT}
1792: \label{sec:algebraic-structure}
1793:
1794: Here we discuss an approach to the algebraic structure
1795: of OSFT that is inspired by conformal field theory.
1796: This approach can be used to write rather general
1797: string actions, including those whose interactions are not based
1798: on delta-function overlaps. In this language the string action
1799: takes the form
1800: \begin{equation} \label{e1}
1801: S (\Phi) = \,-\, {1\over g^2}\,\,\bigg[\, {1\over 2} \langle \,\Phi
1802: \,,\, Q\,
1803: \Phi
1804: \rangle + {1\over 3}\langle \,\Phi \,,\, \Phi *
1805: \Phi \rangle \bigg]\,.
1806: \end{equation}
1807: Here $g$ is the open string coupling constant, the string field
1808: $\Phi$ is a state in the total matter plus ghost CFT, $Q$ is the
1809: kinetic operator, $*$ denotes a multiplication or star-product,
1810: and $\langle \cdot \,, \cdot \rangle$ is a bilinear inner product
1811: on the state space of the CFT.
1812: We will discuss the relationship
1813: between the action (\ref{e1}) and the form of the action
1814: (\ref{eq:SFT-action}) used in the previous subsection shortly.
1815:
1816: The kinetic operator $Q$ satisfies the following identities
1817: \begin{eqnarray} \label{l1e1}
1818: && Q^2 A = 0, \nonumber \\
1819: && Q (A * B) = (QA) * B + (-1)^{A} A * (QB)\,, \\
1820: && \langle \, Q A , B \,\rangle = - (-)^A \langle A , Q B \rangle
1821: \,.\nonumber
1822: \end{eqnarray}
1823: The first equation is the nilpotency condition, the second states
1824: that $Q$ is a derivation of the star product, and the third states
1825: that $Q$ is self-adjoint.
1826: There are also identities involving the inner product and the star operation
1827: \begin{eqnarray}
1828: \label{l1e2}
1829: && \langle A, B \rangle = (-)^{AB} \langle B , A \rangle\,, \nonumber \\
1830: && \langle \, A \,, B * C \, \rangle = \, \langle A* B \,,\, C \,
1831: \rangle \, \\
1832: && A * (B * C) = (A*B) * C \,.\nonumber
1833: \end{eqnarray}
1834: In the sign factors, the exponents $A, B, \cdots$ denote the
1835: Grassmanality
1836: of the state, and should be read as $(-)^A \equiv (-)^{\epsilon (A)}$
1837: where $\epsilon (A) = 0 \, (\hbox{mod}~ 2)$ when $A$ is Grassmann even,
1838: and
1839: $\epsilon (A) = 1 \, (\hbox{mod}~ 2)$ when $A$ Grassmann odd.
1840: The first property above is a symmetry condition, the second indicates
1841: that the inner product has a cyclicity property analogous to the similar
1842: property of the trace operation. Finally, the last equation is the
1843: statement that the star product is associative.
1844:
1845: Finally, we also
1846: have that the star operation is an {\it even} product of degree zero
1847: (as before, we
1848: identify degree with ghost number). In plain english, this means that both the
1849: grassmanality and the ghost number of the star
1850: product of two string fields is obtained from those of the string
1851: fields without any additional offset:
1852: \begin{eqnarray}
1853: \label{l1e3}
1854: && \epsilon (A * B ) = \epsilon (A) + \epsilon (B) \,,\nonumber \\
1855: && \hbox{gh} (A* B ) = \hbox{gh} (A) + \hbox{gh} (B) \,.
1856: \end{eqnarray}
1857: In this language $Q$ is an odd operator of degree one:
1858: \begin{eqnarray}
1859: \label{l1e3q}
1860: && \epsilon (QA ) = \epsilon (A) +1\,, \nonumber \\
1861: && \hbox{gh} (QA ) = \hbox{gh} (A) + 1 \,.
1862: \end{eqnarray}
1863: In the conventions we shall work
1864: the SL(2,R) vacuum $|0\rangle$ is assigned ghost
1865: number zero. The Grassmanality $\epsilon (A)$
1866: of a string field $A$ is an integer mod~2.
1867: In open string field theory, Grassmanality and ghost number (degree)
1868: are related because Grassmann odd operators carry odd units of
1869: ghost number.
1870:
1871: The algebraic structure discussed here is very similar, but
1872: not identical to that in section~\ref{sec:Witten-SFT}.
1873: The string field $\Phi$ and the action (\ref{e1}) can be related to
1874: the string field $\Psi$ and the action (\ref{eq:SFT-action}) of the
1875: previous section by taking
1876: \begin{equation}
1877: \Phi = g \Psi
1878: % \label{eq:}
1879: \end{equation}
1880: and by relating the inner product used here to the integral used in
1881: (\ref{eq:SFT-action}) through
1882: %Here, we have
1883: %an inner product, while there we have an integral operation.
1884: %The two are related as
1885: \begin{equation}
1886: \label{identifystructures}
1887: \langle A \,, B\rangle = \int A \star B \,.
1888: \end{equation}
1889: The first two conditions in (\ref{l1e1}) are then clearly equivalent to
1890: properties (a) and (c) of section~\ref{sec:Witten-SFT}. You can also
1891: readily see that the first two properties in (\ref{l1e2}) hold given
1892: properties (d) and (e). Property (b), however, does not have a
1893: counterpart in this formalism. A counterpart exists if we assume the
1894: existence of a suitable identity string field $\mathcal{I}$, as we
1895: will discuss at the end of this subsection.
1896:
1897: Throughout these lectures we will go back and forth between the CFT
1898: notation with string field $\Phi$ and action (\ref{e1}) and the
1899: oscillator description with string field $\Psi$ and action
1900: (\ref{eq:SFT-action}) (which we rewrite more explicitly as
1901: (\ref{eq:action-Fock-1}) in section \ref{sec:overlaps}). While we
1902: could have chosen to use one notation and neglect the other, both
1903: formalisms are used extensively in the literature, and some results
1904: are more easily expressed in one notation than the other. When in
1905: doubt, the reader should return to the previous paragraph to see how
1906: the two notations are related.
1907:
1908: \medskip
1909: Let us now deduce some basic properties of the string field,
1910: in particular its ghost number and its Grassmanality. The Grassmanality
1911: of $\Phi$ can be deduced from the condition that the kinetic term
1912: of the string action must be non-vanishing. Using the above properties
1913: we have
1914: \begin{equation}
1915: \label{deducephi}
1916: \langle \Phi, Q \Phi \rangle = (-1)^{\Phi (1+ \Phi)}
1917: \langle Q \Phi, \Phi \rangle = \langle Q \Phi, \Phi \rangle
1918: = - (-1)^\Phi \langle \Phi, Q \Phi \rangle \,.
1919: \end{equation}
1920: It is clear that the string field $\Phi$ must be Grassmann odd.
1921: At this point we must use some CFT knowledge to decide on
1922: the Grassmanality of the SL(2,R) vacuum and on the ghost number
1923: of the string field. For bosonic strings we have that zero momentum
1924: tachyon states are of the form $t c_1 |0\rangle$, where $c_1$ is a
1925: ghost field oscillator. Since this oscillator is Grassmann odd, and
1926: the string field is also Grassmann odd, we must declare the SL(2,R)
1927: vacuum to be Grassmann even. Thus
1928: \begin{equation}
1929: \label{propzero}
1930: |0\rangle \,\, \hbox{is a Grassmann even state of ghost number zero}\,.
1931: \end{equation}
1932: Since the $c_1$ oscillator carries ghost number one, we also deduce
1933: that the open string field must have ghost number one.
1934: \begin{equation}
1935: \label{propsf}
1936: |\Phi\rangle \,\, \hbox{is a Grassmann odd state of ghost number one}\,.
1937: \end{equation}
1938:
1939:
1940:
1941:
1942: Equations (\ref{l1e1}), (\ref{l1e2}),
1943: (\ref{l1e3}), (\ref{l1e3q}), and (\ref{propsf}) guarantee that the
1944: string field action is
1945: invariant under the gauge transformations:
1946: \begin{equation} \label{l1e5}
1947: \delta \Phi = Q \Lambda + \Phi * \Lambda - \Lambda * \Phi\, ,
1948: \end{equation}
1949: for any Grassmann-even ghost-number zero state $\Lambda$.
1950: Moreover, variation of the action gives the field equation
1951: \begin{equation}
1952: Q \Phi + \Phi * \Phi = 0\,.
1953: \end{equation}
1954:
1955: \medskip
1956: \noindent
1957: {\it Exercise} Verify that the string action in (\ref{e1}) is gauge
1958: invariant under the transformations (\ref{l1e5}).
1959:
1960: \bigskip
1961:
1962:
1963: It is convenient to use the above structures to define a multilinear object
1964: that given three string fields
1965: yields a number:
1966: \begin{equation}
1967: \langle\, A\,,\, B\,,\, C \rangle \equiv \langle \, A \,, \, B * C \,
1968: \rangle
1969: \end{equation}The middle equation in (\ref{l1e2}) implies the {\it cyclicity}
1970: of the multilinear form. A small calculation immediately gives:
1971: \begin{equation}
1972: \langle\, A\,,\, B\,,\, C \rangle = (-)^{A(B+C)}\langle\, B\,,\, C\,,\,
1973: A
1974: \rangle
1975: \end{equation}
1976: A basic consistency check of the signs above is that the cubic
1977: term
1978: $\langle\, \Phi\,,\, \Phi\,,\, \Phi \rangle$ in the action
1979: (\ref{l1e1}) is strictly
1980: cyclic for odd $\Phi$, and therefore does not vanish.
1981:
1982: \bigskip
1983: Open string theory has additional algebraic structure that
1984: sometimes plays a crucial role. One such structure arises from
1985: the twist operation, which reverses the parametrization of a
1986: string. From the algebraic viewpoint this is summarized by the
1987: existence of an operator $\Omega$ that satisfies the following
1988: properties:
1989: \begin{eqnarray}
1990: \label{l1e4}
1991: && \Omega ( QA) = Q (\Omega A) \nonumber\\
1992: &&\langle \, \Omega A \,, \, \Omega B \, \rangle = \langle\, A\,, \, B
1993: \,\rangle \\
1994: &&\Omega \, (A* B) = (-)^{AB + 1} \, \Omega (B) * \Omega (A)\nonumber\,.
1995: \end{eqnarray}
1996: The first property means that the BRST operator has zero twist, or
1997: does not change the twist property of the states it acts on.
1998: The second property states that the bilinear form is twist
1999: invariant. The third property is crucial. Up to signs, twisting
2000: the star product of string fields amounts to multiplying the
2001: twisted states in {\it opposite order}. This change of order
2002: is a simple consequence of the basic multiplication rule where
2003: the second half of the first string must be glued to the first
2004: half of the second one. The sign factor is also important.
2005: For the string field $\Phi$, which is grassmann odd, it gives
2006: \begin{equation}
2007: \label{l1e59}
2008: \Omega ( \Phi * \Phi ) = + \, (\Omega \Phi) * (\Omega \Phi)
2009: \end{equation}
2010: with the plus sign. This result, together with the first
2011: two equations in (\ref{l1e4}) immediately implies that the
2012: string field action in (\ref{l1e1}) is twist invariant:
2013: \begin{equation}
2014: \label{l1e6}
2015: S (\Omega \Phi) = S (\Phi) \,.
2016: \end{equation}
2017: This invariance under twist transformations allows one to
2018: construct new string theories by truncating the spectrum to
2019: the subset of states that are twist even. Moreover, in solving
2020: the string field equations it will be possible to find consistent
2021: solutions by restricting oneself to the twist even subspace of
2022: the string field.
2023:
2024: \medskip
2025: \noindent
2026: {\it Exercise}. Letting $\Omega_A$ denote the $\Omega$ eigenvalue of $A$,
2027: show that
2028: \begin{equation}
2029: \langle A, B, C\rangle = \Omega_A\Omega_B\Omega_C (-1)^{AB + BC+CA +1}
2030: \langle C, B, A\rangle \,.
2031: \end{equation}
2032:
2033: \medskip
2034: \noindent
2035: {\it Exercise}. Let $\Omega A_\pm = \pm A$ and $\epsilon (A_\pm) =1$.
2036: Show that
2037: \begin{equation}
2038: \label{fortwistproperty}
2039: \langle A_+, A_+, A_-\rangle = 0\,.
2040: \end{equation}
2041:
2042: \medskip
2043: \noindent
2044: {\it Exercise}. We will see later
2045: that the star product of the vacuum with itself is the vacuum
2046: plus Virasoro descendents:
2047: \begin{equation}
2048: |0\rangle * |0\rangle = |0\rangle + \cdots
2049: \end{equation}
2050: Show that this implies that the vacuum is twist odd:
2051: \begin{equation}
2052: \Omega |0\rangle = -|0\rangle .
2053: \end{equation}
2054:
2055: The star algebra may have
2056: an identity element ${\mathcal I}$.
2057: If ${\mathcal I}$ exists, it is presumed to satisfy
2058: \begin{equation}
2059: {\mathcal I}* A = A * {\mathcal I} = A\,,
2060: \end{equation}
2061: for all states $A$.
2062: Some properties of ${\mathcal I}$
2063: are immediately deduced from the above definition:
2064: \begin{equation}
2065: {\mathcal I} \,\,\hbox{is Grassmann even, ghost number zero, twist odd
2066: string field.}
2067: \end{equation}
2068: The twist odd property follows from the twist property of products
2069: \begin{equation}
2070: \Omega ( {\mathcal I} *A) = (-1)^{0\cdot A + 1} (\Omega A) * (\Omega
2071: {\mathcal I})= -
2072: (\Omega A) * (\Omega {\mathcal I})\,.
2073: \end{equation}
2074: Since the left hand side is also just $(\Omega A)$ it must follow that
2075: \begin{equation}
2076: \Omega {\mathcal I} = - {\mathcal I}\,.
2077: \end{equation}
2078: This is consistent with the fact that the SL(2,R) vacuum is also
2079: twist odd. Indeed the identity string field is just the vacuum plus
2080: Virasoro descendents of the vacuum, as we shall see in
2081: Section~\ref{subsec:slivers}.
2082:
2083: Any derivation $D$ of the star algebra should
2084: annihilate the identity:
2085: \begin{equation}
2086: D ( {\mathcal I}* A) = (D{\mathcal I}) * A + {\mathcal I} * DA =
2087: (D{\mathcal I}) * A + DA\,.
2088: \end{equation}
2089: Since the left hand side also equals $DA$, one concludes that
2090: $ (D{\mathcal I}) * A=0$ for all $A$, and thus the expectation that
2091: $D{\mathcal I} =0$.
2092: In the star algebra of open strings
2093: an identity state has been identified~\cite{Gross-Jevicki-12,efhm,Schnabl} that
2094: satisfies the expected properties for most well-behaved states.
2095:
2096: Finally,
2097: if the identity string field is annihilated by the derivation $Q$, then
2098: \begin{equation}
2099: \langle Q \Psi \,, \mathcal{I} \rangle = - (-)^\Psi \langle \Psi, Q
2100: \mathcal{I}\rangle =0\,.
2101: \end{equation}
2102: The identification (\ref{identifystructures}) then yields
2103: \begin{equation}
2104: 0= \int Q\Psi \star \mathcal{I} = \int Q\Psi \,,
2105: \end{equation}
2106: which is property (b) in the axiomatic formulation of OSFT
2107: discussed in section~\ref{sec:Witten-SFT}.
2108:
2109:
2110:
2111: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2112: % SFT: CFT approach: BZ
2113: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2114:
2115: \section{String field theory: conformal field theory approach}
2116: \label{sec:CFT}
2117:
2118: A direct conformal field theory evaluation of the string action
2119: is perhaps the most economical way to proceed in the case of
2120: simple computations. We will explain this definition of the
2121: action, using at the same time the example of the
2122: action restricted to only the tachyon field
2123: to illustrate the definitions. The string action, written before
2124: in (\ref{e1}) is given by
2125: \begin{equation} \label{e1p}
2126: S (\Phi) = \,-\, {1\over g^2}\,\,\bigg[\, {1\over 2} \langle \,\Phi
2127: \,,\, Q\,
2128: \Phi
2129: \rangle + {1\over 3}\langle \,\Phi \,,\, \Phi *
2130: \Phi \rangle \bigg]\,.
2131: \end{equation}
2132: This OSFT action can be used to describe the spacetime field theory
2133: of any D-brane. For example, for a Dp-brane we would have an
2134: underlying conformal field theory of a field $X^0$
2135: and $p$ fields $X^i$
2136: with Neumann boundary
2137: conditions, and $(25-p)$ fields $X^a$
2138: with Dirichlet boundary conditions.
2139: In our computations, we will assume that the brane has unit volume,
2140: in which case the mass $M$ of the brane coincides with its tension.
2141: One
2142: can show that, in units where $\alpha'=1$,
2143: \begin{equation}
2144: \label{tf1}
2145: M = {1\over 2\pi^2} {1\over g^2}\,.
2146: \end{equation}
2147: We will prove this result in section~\ref{sec:tension}.
2148:
2149: We will evaluate the OSFT action by truncating the string field
2150: down to the zero momentum tachyon.
2151: The systematic approximation of
2152: the full theory by successive level truncation is described in detail
2153: in Section \ref{sec:vacuum}.
2154: In the level expansion this
2155: zero momentum tachyon is assigned level zero. The tachyon
2156: vertex operator is $e^{ipX(z)} c(z)$ and the associated state is
2157: $c_1 |0;p\rangle$. The zero momentum tachyon state is
2158: $c_1 |0;0\rangle$ or in simpler notation $c_1 |0\rangle$.
2159: Since we have
2160: \begin{equation}
2161: L_0 c_1 |0\rangle = - c_1 |0\rangle\,,
2162: \end{equation}
2163: the level $\ell$ of a state is related to the $L_0$ eigenvalue as
2164: \begin{equation}
2165: \ell = L_0 + 1\,.
2166: \end{equation}
2167: The string field truncated to the zero momentum tachyon is written
2168: as
2169: \begin{equation}
2170: |T\rangle = t \, c_1 |0\rangle \,,
2171: \end{equation}
2172: where the variable $t$ denotes the expectation value of the
2173: tachyon field, and it is a spacetime constant.
2174: The variable $t$ is related to the tachyon field $\phi$ in the
2175: expansion (\ref{eq:field-expansion}) through
2176: \begin{equation}
2177: t = g \phi (0) \,.
2178: % \label{eq:}
2179: \end{equation}
2180: As we alternate between notation $\Psi$ and $\Phi$ for the string
2181: field, we will use $\phi$ and $t$ for the zero-momentum tachyon.
2182: After truncating to just the tachyon degree of freedom $t$
2183: the tachyon potential $V(t)$ is just minus $S(|T\rangle)$ and thus
2184: \begin{equation}
2185: V(t) = - S(|T\rangle) = M (2\pi^2) \Bigl( {1\over 2} \langle T, Q T\rangle
2186: + {1\over 3} \langle T, T , T \rangle \Bigr)\,.
2187: \end{equation}
2188: In fact, it is convenient to define the ratio
2189: \begin{equation}
2190: \label{ftest}
2191: f(t) \equiv {V(t)\over M} = (2\pi^2) \Bigl( {1\over 2}
2192: \langle T, Q T\rangle + {1\over 3} \langle T, T , T \rangle \Bigr)\,.
2193: \end{equation}
2194: The function $f(t)$ is a rescaled version of the tachyon potential.
2195: By construction, $f(t)$ has a quadratic term and a cubic term, so $f(t=0)=0$.
2196: The Sen conjecture requires that $f(t)$ have a critical point at $t=t^*$ that
2197: satisfies
2198: \begin{equation}
2199: f(t^*) = -1 \,.
2200: \end{equation}
2201: This is indeed equivalent to
2202: saying that the energy difference between
2203: the D-brane vacuum and the stable vacuum equals the energy $M$
2204: of the D-brane. It suggests strongly that the stable vacuum is a
2205: vacuum without a D-brane. It is perhaps useful to remark that
2206: $V(t)$ as obtained directly from the OSFT action does not convey the
2207: true gravitational picture where absolute vacuum energies are important.
2208: The vacuum with the D-brane, namely at $t=0$ has a positive cosmological
2209: constant, or vacuum energy. This is in fact the D-brane energy. As the
2210: theory rolls to the stable vacuum, the vacuum energy goes to zero.
2211: Thus the tachyon potential $V(t)$ is missing an additive constant, which
2212: becomes important when coupling to gravity (which we will not consider
2213: in the present lectures). Such a constant term at least morally belongs
2214: in a more general OSFT action where the disk partition function
2215: would
2216: naturally appear as a field independent contribution to the string action.
2217: This disk partition function calculated with the boundary condition
2218: appropriate to the D-brane is in fact proportional to the D-brane energy.
2219:
2220: \subsection{Kinetic term computations}
2221:
2222: Let us begin the computation of the string action truncated to
2223: the tachyon by evaluating $\langle T , QT\rangle$.
2224: To this end we need to use the normalization condition
2225: \begin{equation}
2226: \label{ourashokenorm}
2227: \langle 0| c_{-1} c_0 c_1 |0\rangle = 1\,,
2228: \end{equation}
2229: which is appropriate if we compactify all coordinates (including
2230: time) into circles of unit circumference. Indeed, comparing with
2231: (\ref{conventionalnormalization}), we see that the right-hand side
2232: of (\ref{ourashokenorm}) should have a $(2\pi)^{26} \delta (0)$, which
2233: is equivalent to the full spacetime volume $V$. In our full compactification,
2234: $V=1$. The compactification of time is only a formal
2235: trick that facilitates computations but is not strictly necessary.
2236:
2237: \smallskip
2238: \noindent
2239: {\it Exercise:} Given $c(z) = \sum_n {c_n\over z^{n-1}}$ show that
2240: \begin{equation}
2241: \label{ghcorr}
2242: \langle 0| c(z_1) c(z_2) c(z_3) |0\rangle = (z_1-z_2) (z_1 - z_3)
2243: (z_2 - z_3)\,.
2244: \end{equation}
2245:
2246: \medskip
2247: Now that we must compute precisely we should make clear the CFT
2248: definition of the inner product
2249:
2250: \noindent
2251: {\it Definition: } $ \langle A , B\rangle = \langle bpz(A) | B\rangle$.
2252: Here $bpz: {\mathcal H} \to {\mathcal H}^*$ is BPZ conjugation,
2253: which we review
2254: next.
2255:
2256: Given a primary field $\phi(z)$ of dimension $d$, it has
2257: a mode expansion
2258: \begin{equation}\
2259: \label{oscill}
2260: \phi(z) = \sum_n {\phi_n\over z^{n+d}}
2261: \quad \to \quad \phi_n = \oint {dz\over 2\pi i} z^{n+d-1} \phi(z)\,.
2262: \end{equation}
2263: We define
2264: \begin{equation}
2265: bpz(\phi_n) \equiv \oint {dt\over 2\pi i} t^{n+d-1} \phi(t)\,, \qquad
2266: \hbox{with} \quad t = -{1\over z} \,.
2267: \end{equation}
2268: Note that this simply defines the BPZ conjugation of the oscillator with
2269: the same formula as the oscillator itself (\ref{oscill}) but referred to
2270: a coordinate at $z=\infty$. This integral is evaluated by using the
2271: transformation law
2272: \begin{equation}
2273: \phi(t) (dt)^d = \phi(z) (dz)^d\,.
2274: \end{equation}
2275: We therefore get
2276: \begin{equation}
2277: bpz(\phi_n) \equiv - \oint {dz\over 2\pi i}{1\over z^2}
2278: \Bigl( - {1\over z} \Bigr)^{n+d-1} \phi(z) (z^2)^d\,.
2279: \end{equation}
2280: The minus sign in front arises from a reversal of contour of integration
2281: (a contour circling $t=0$ clockwise circles $z=0$ counterclockwise).
2282: Moreover the transformation law was used to reexpress $\phi(t)$ in
2283: terms of the field $\phi(z)$ whose mode expansion is given. Simplifying
2284: the integral one finds
2285: \begin{equation}
2286: bpz(\phi_n) = (-1)^{n+d} \oint {dz\over 2\pi i}
2287: z^{-n+d-1} \phi(z) \, = (-1)^{n+d} \phi_{-n}\,.
2288: \end{equation}
2289: In summary, we have shown that
2290: \begin{equation}
2291: bpz(\phi_n) = (-1)^{n+d} \phi_{-n}\,.
2292: \end{equation}
2293: This equation defines BPZ conjugation when we supplement it with
2294: the rule
2295: \begin{equation}
2296: bpz \Bigl( \phi_{n_1} \cdots \phi_{n_p}|0\rangle \Bigr)
2297: = \langle 0 | bpz(\phi_{n_1}) \cdots bpz(\phi_{n_p})\,.
2298: \end{equation}
2299: This formula is correct as stated also when the oscillators
2300: are anticommuting. The only condition for its validity is that
2301: the various modes with mode numbers of the same sign must
2302: commute (or anticommute). Otherwise BPZ conjugation produces
2303: a sequence of oscillators in {\em reverse} order.
2304:
2305: A nontrivial example of the above rules arises when we calculate
2306: the BPZ conjugates of the modes $L_n$ of the stress tensor. Although the
2307: stress tensor $T(z)$ is not a primary field, it transforms as a primary under
2308: SL(2,C) transformations, and therefore it does transform as a dimension two
2309: primary under the inversion needed in the definition of BPZ. Thus we have
2310: \begin{equation}
2311: bpz(L_n) = (-1)^n L_{-n}\,,
2312: \end{equation}
2313: and for a string of oscillators we must write
2314: \begin{equation}
2315: bpz \Bigl( L_{n_1} \cdots L_{n_p}|0\rangle \Bigr)
2316: = \langle 0 | bpz(L_{n_p}) \cdots bpz(L_{n_1})\,.
2317: \end{equation}
2318:
2319: \medskip
2320: Since $c(z)$ has dimension minus one,
2321: $bpz(c_1) = (-1)^{1+1} c_{-1} = c_{-1}$, so
2322: $bpz (c_1 |0\rangle) = \langle 0| c_{-1}$. With this
2323: we have
2324: \begin{equation}
2325: \langle T, Q T \rangle = t^2 \langle 0| c_{-1} Q c_1 |0\rangle\,.
2326: \end{equation}
2327: Because of the form of the inner product only the term $c_0L_0$
2328: in $Q$ can contribute and we have
2329: \begin{equation}
2330: \label{tachkin}
2331: \langle T, Q T \rangle = t^2 \langle 0| c_{-1} c_0 L_0 c_1 |0\rangle
2332: = - t^2 \langle 0| c_{-1} c_0 c_1 |0\rangle = - t^2\,.
2333: \end{equation}
2334: This completes the computation of the quadratic term in the tachyon
2335: potential. The negative sign obtained is the expected one, showing
2336: the instability of the $t=0$ field configuration.
2337:
2338: \subsection{Interaction term computation}
2339: \medskip
2340: To compute the interaction of three tachyons we must explain
2341: how the three vertex is defined in CFT language. Consider three
2342: states $A,B,$ and $C$ and their associated vertex operators
2343: ${\mathcal O}_A$, ${\mathcal O}_B$, and ${\mathcal O}_C$. We define
2344: \begin{equation}
2345: \label{defvertex}
2346: \langle A, B, C \rangle \equiv \Bigl\langle f_1^D \circ {\mathcal O}_A(0),
2347: f_2^D \circ {\mathcal O}_B(0), \, f_3^D \circ {\mathcal O}_C(0)\Bigr\rangle_D
2348: \end{equation}
2349: Here the right hand side denotes the CFT correlator of the conformal
2350: transforms of the vertex operators
2351: ${\mathcal O}_A$, ${\mathcal O}_B$, and ${\mathcal O}_C$. The conformal
2352: transforms are specified by the functions $f_i$ as we explain now.
2353: Let there be three canonical coordinates $\xi_i$, with $i=1,2,3$.
2354: The three functions $f_i(\xi_i)$ define maps from the upper half
2355: disks $\Im (\xi_i) \geq 0, |\xi_i| \leq 1$ into a disk $D$, with the points
2356: $\xi_i=0$ being taken into points in the boundary of the disk $D$.
2357: The meaning of the conformal map of operators is that:
2358: $f_i\circ {\mathcal O}_A (0)$ {\it is the operator} ${\mathcal O}_A(\xi_i=0)$
2359: {\it expressed in terms of local operators at} $f_i(\xi_i=0)$.
2360: The disk $D$ may have the form of a unit disk, or can be the
2361: (conformally equivalent) upper half plane, or any other arbitrary
2362: form. Of course, the unit disk and the upper half plane are
2363: especially convenient for explicit computations.
2364:
2365:
2366: \begin{figure}[!ht]
2367: \leavevmode
2368: \begin{center}
2369: \epsfxsize = 11 cm \epsfbox{pictorialfig.eps}
2370: \end{center}
2371: \caption[]{\footnotesize Representation of the cubic vertex as the gluing of
2372: 3 half--disks.} \label{pictorialfig}
2373: \end{figure}
2374:
2375:
2376:
2377:
2378: For the SFT
2379: at hand, the picture
2380: is given in Fig.{\ref{pictorialfig}}.
2381: The worldsheets of the three strings
2382: are represented as the unit half--disks
2383: $\{ |\xi_i| \leq 1, \Im \, \xi \geq 0 \}$, $i=1,2,3$,
2384: in three copies of the complex plane.
2385: The boundaries $|\xi_i| =1$ in the respective upper half-disks
2386: are {\it the} strings. Thus the point $\xi_i=i$ is the string
2387: midpoint. The interaction defining the vertex is
2388: built by gluing the three half-disks to form a single disk. This is done
2389: by the half-string identifications:
2390: \begin{eqnarray}
2391: \label{wittengluing}
2392: \xi_1 \xi_2 = -1\,, && \quad {\rm for} \; |\xi_1| =1, \,~~ \Re ( \xi_1 )\leq
2393: 0\,, \nonumber \\
2394: \xi_2 \xi_3 = -1\,, && \quad {\rm for} \; |\xi_2| =1, \,~~ \Re (\xi_2) \leq
2395: 0\,, \\ \xi_3 \xi_1 = -1\,, && \quad {\rm for} \; |\xi_3| =1,
2396: \,~~ \Re (\xi_3)
2397: \leq 0\,. \nonumber
2398: \end{eqnarray}
2399: Note that the common interaction point $Q$,
2400: is indeed $\xi_i=i$ (for
2401: $i=1,2,3$), namely the mid--point of each open string $|\xi_i| =1, \,
2402: \Im (\xi_i)
2403: \geq 0$. The left-half of the first string is glued with the
2404: right-half of the second string, and the same is repeated cyclically.
2405: This construction defines a specific `three--punctured disk', a
2406: genus zero Riemann surface with a boundary, three marked points
2407: (punctures) on this boundary, and a choice of local coordinates
2408: $\xi_i$ around each puncture.
2409:
2410:
2411: The calculation of the functions $f^D_i(\xi)$ require a choice of
2412: disk $D$. We begin with the case when the disk $D$ is simply
2413: chosen to be the interior of the unit disk
2414: $|w| < 1$, as shown in Fig. \ref{3wedgesfig}.
2415: In this case the functions $f_i^{D_w} \equiv f_i$ must map each half-disk
2416: to a $120^\circ$ wedge of this unit disk.
2417: To construct the explicit maps that
2418: send
2419: $\xi_i$ to the
2420: $w$ plane, one notices that the SL(2,C) transformation
2421: \begin{equation}
2422: \label{hdefi}
2423: h(z)= \frac{1+i\xi}{1-i\xi}\,,
2424: \end{equation}
2425: maps the unit upper--half disk $\{ |\xi| \leq 1, \Im \xi \geq 0 \}$ to the
2426: `right' half--disk $\{ |h| \leq 1, \Re \, h\geq 0 \}$, with
2427: $z=0$ going to $h(0) =1$. Thus the functions
2428: \begin{eqnarray}
2429: \label{unitvertex}
2430: f_1(\xi_1) & =& e^{\frac{2 \pi i}{3}}\left(
2431: \frac{1+i\xi_1}{1-i\xi_1} \right)^{\frac{2}{3}}\,, \nonumber \\
2432: f_2(\xi_2) & =& \left(
2433: \frac{1+i\xi_2}{1-i\xi_2} \right)^{\frac{2}{3}}\,, \\
2434: f(\xi_3) & =& e^{-\frac{2 \pi i}{3}} \left(
2435: \frac{1+i\xi_3}{1-i\xi_3} \right)^{\frac{2}{3}}\,,\nonumber
2436: \end{eqnarray}
2437: will send the three half-disks to three wedges in the $w$
2438: plane of Fig. \ref{3wedgesfig}, with punctures at $e^{\frac{2 \pi
2439: i}{3}}$, $1$, and $e^{-\frac{2 \pi i}{3}}$ respectively.
2440: This specification of the functions $f_i(\xi_i)$
2441: gives the definition of the cubic vertex.
2442: In this representation cyclicity ({\it i.e.}, $\langle \Phi_1, \Phi_2 ,
2443: \Phi_3 \rangle =
2444: \langle \Phi_2, \Phi_3 , \Phi_1 \rangle $) is manifest by construction.
2445: By SL(2,C) invariance, there are many other possible representations
2446: that give exactly the same off--shell amplitudes.
2447:
2448: \medskip
2449:
2450:
2451:
2452: \begin{figure}[!ht]
2453: \leavevmode
2454: \begin{center}
2455: \epsfxsize = 6 cm \epsfbox{3wedgesfig.eps}
2456: \end{center}
2457: \caption[]{\footnotesize Representation of the cubic vertex as a 3--punctured
2458: unit disk.} \label{3wedgesfig}
2459: \end{figure}
2460:
2461:
2462:
2463:
2464:
2465:
2466: A useful choice is obtained by mapping the interacting $w$ disk
2467: symmetrically to the upper half $z$-plane $H$. This is the convention
2468: that we shall mostly be using.
2469: We can therefore
2470: define the functions
2471: $f_i^H$
2472: by composition of the earlier maps $f_i$ (that send the half-disks to
2473: the $w$ unit disk) with the map $h^{-1}(w) =-i
2474: \,\frac{w-1}{w+1}$ that takes this unit disk
2475: to the upper--half--plane, with the three punctures on the real
2476: axis (Fig. \ref{uhpfig}),
2477: \begin{eqnarray}
2478: \label{uhpvertex}
2479: f_1^H (\xi_1)& \equiv & h^{-1}\circ f_1 (\xi_1)=
2480: S ( f_3^H (\xi_1)) \cr\cr
2481: &=& \sqrt {3}+{\frac {8}{3}}\,\xi_1
2482: +{\frac {16}{9}}\,\sqrt {3}\,{\xi_1}^{2}+{\frac
2483: {248}{81}}\,{\xi_1}^{3}+O\left ({\xi_1}^{4}\right )\,. \cr\cr
2484: f_2^H (\xi_2) & \equiv & h^{-1}\circ f_2 (\xi_2) =S(f_1^H(\xi_2))=
2485: \tan
2486: \left(\frac{2}{3} \, \arctan(\xi_2) \right) \cr\cr
2487: &=&
2488: {\frac {2}{3}}\, \xi_2-{\frac {10}{81}}\, {\xi_2}^{3}
2489: +O\left
2490: ({z_2}^{5}\right )\,.
2491: \cr\cr
2492: f_3^H (\xi_3) & \equiv & h^{-1}\circ f_3\, (\xi_3) =
2493: S (f_2^H (\xi_3)) \cr\cr
2494: &=&
2495: -\sqrt {3}+{\frac {8}{3}}\, \xi_3-{\frac {16}{9}}\,\sqrt
2496: {3}\, {\xi_3}^{2}+{\frac {248}{81}}\, {\xi_3}^{3} +O({\xi_3}^{4}) \,.
2497: \end{eqnarray}
2498: The three punctures are at
2499: $ f_1^H(0)=+\sqrt{3}, f_2^H(0)=0, f_3^H(0)=-\sqrt{3},$
2500: and the SL(2,R) map $S(z) = \frac{z - \sqrt{3}}{1+\sqrt{3}z}$
2501: cycles them
2502: (thus $S\circ S\circ S(z) =z$).
2503:
2504:
2505:
2506: \begin{figure}[!ht]
2507: \leavevmode
2508: \begin{center}
2509: \epsfxsize = 10 cm \epsfbox{uhpfig.eps}
2510: \end{center}
2511: \caption[]{\footnotesize Representation of the cubic vertex as the upper--half
2512: plane with 3 punctures on the real axis.}\label{uhpfig}
2513: \end{figure}
2514:
2515:
2516:
2517: \medskip
2518: This completes the definition
2519: of the string field theory action. When the disk $D$ is presented as
2520: a unit disk the functions $f_i$ in (\ref{defvertex}) are the functions
2521: given in equation (\ref{unitvertex}). When the disk $D$ is presented
2522: as the upper half plane $H$ the relevant functions in (\ref{defvertex})
2523: are the functions $f_i^H$ given in (\ref{uhpvertex}) above.
2524:
2525: \medskip\noindent
2526: {\it Exercise:} Verify explicitly by a {\it by-hand} calculation
2527: that the first two terms in the expansion of $f_1^H$ and $f_3^H$,
2528: as well as the first term in $f_2^H$ are correct.
2529:
2530:
2531: \medskip
2532: Let us now return to the computation of the tachyon action.
2533: For our string field $|T\rangle = tc_1|0\rangle$ the interaction
2534: term $\langle T, T, T\rangle$ will be given by
2535: \begin{equation}
2536: \langle T, T, T\rangle = t^3 \langle c_1, c_1, c_1\rangle\,.
2537: \end{equation}
2538: Since the vertex operator associated to $c_1|0\rangle$ is
2539: $c(z)$, using
2540: (\ref{defvertex}) we write:
2541: \begin{eqnarray}
2542: \langle T, T, T\rangle = t^3\langle f_1^H\circ c (0), f_2^H\circ c
2543: (0), f_3^H\circ c (0) \rangle_{H}
2544: \end{eqnarray}
2545: Since the field $c(z)$ is a primary of dimension minus one, we
2546: have
2547: \begin{equation}
2548: {c(z)\over dz} = {c(\xi)\over d\xi} \quad \to c(\xi) = {c(z)\over {dz\over
2549: d\xi}}
2550: \end{equation}
2551: Therefore
2552: \begin{eqnarray}
2553: f\circ c(0) \equiv c(\xi=0) = {c(f(0)) \over f'(0)}.
2554: \end{eqnarray}
2555: Using equations (\ref{uhpvertex}) to read the values of
2556: $f_1^H(0)$ and ${df_1^H\over d\xi }(0)$ we therefore get, for
2557: example,
2558: \begin{equation}
2559: f_1^H\circ c(0) = {c(f_1^H(0)) \over {f_1^H}'(0)} =
2560: {c(\sqrt{3})\over {8/3}}\,.
2561: \end{equation}
2562: The other two insertions are dealt with similarly, and we find
2563: \begin{eqnarray}
2564: \langle T, T, T \rangle &=& t^3 \Bigl\langle\,\, { c (\sqrt{3})\over
2565: {8/ 3}}\,\,, { c
2566: (0)\over {2 / 3}}, { c (-\sqrt{3})\over {8/ 3}}
2567: \Bigr\rangle_H \cr & =& {3^3 \over 2^7}
2568: \langle c(\sqrt{3}) c(0) c(-\sqrt{3}) \rangle_H =
2569: {3^4 \sqrt{3} \over 2^6}\,,
2570: \end{eqnarray}
2571: where in the last step we made use of (\ref{ghcorr}). Our answer is
2572: therefore
2573: \begin{eqnarray}
2574: \label{tachver}
2575: \langle T, T, T \rangle = {81\sqrt{3}\over 64} t^3\equiv t^3 K^3\,,
2576: \quad \langle c_1, c_1, c_1 \rangle = {81\sqrt{3}\over 64} = K^3\,.
2577: \end{eqnarray}
2578: This completes the calculation of an interaction term.
2579:
2580: \subsection{A first test of the tachyon conjecture}\label{firsttestofit}
2581:
2582: Having obtained the kinetic term of the tachyon truncated
2583: action in (\ref{tachkin}) and the cubic term in (\ref{tachver}) we are now in
2584: a position to evaluate the function $f(t)$ in (\ref{ftest}):
2585: \begin{equation}
2586: f(t) = 2\pi^2 \Bigl( - {1\over 2} t^2 + {1\over 3} K^3 t^3\Bigr)\,.
2587: \label{eq:CFT-cubic-potential}
2588: \end{equation}
2589: We must now find the (locally) stable critical point $t=t^*$ of this
2590: potential and evaluate the value of $f(t^*)$. It is clear the answer
2591: will not be the precise one $f=-1$, since we have truncated
2592: the string field dramatically.
2593: Nevertheless, we hope
2594: to get an answer that is reasonably close, if level expansion is
2595: supposed to make sense.
2596:
2597: \medskip
2598: The
2599: equation of motion is
2600: \begin{equation}
2601: -t^* + t^{*2} K^3 = 0 \quad \to \quad t^* = {1\over K^3}\,,
2602: \end{equation}
2603: and substituting back we find
2604: \begin{equation}
2605: f(t^*) = -{1\over 3} {\pi^2\over K^6} = -\pi^2 {2^{12}\over 3^{10}}
2606: = - \pi^2 {4096\over 59049} \simeq -0.684.
2607: \end{equation}
2608: Thus is this simplest approximation, where we only kept the tachyon
2609: zero mode we have found that the critical point cancels about 70\%
2610: of the D-brane energy.
2611: In section (\ref{sec:vacuum}) we discuss the
2612: extension of this calculation to include massive string modes.
2613:
2614:
2615: \subsection{String vertex in the CFT approach: Neumannology}
2616: \label{sec:more-CFT}
2617:
2618: When doing explicit computations in OSFT we need to consider
2619: interactions of fields other than the tachyon. The explicit
2620: computation of the previous section becomes a lot more involved
2621: for massive fields, and it is useful to find an automated procedure
2622: to deal with such calculations. One such procedure is based on
2623: conformal field theory conservation
2624: laws. This is a very effective method, but we will not review it
2625: here since its
2626: explanation in Rastelli and Zwiebach~\cite{Rastelli-Zwiebach} is
2627: self-contained.
2628: Another approach uses the explicit Fock representations
2629: of the string vertex. This will be our subject of interest here.
2630: We will provide a self-contained derivation of the Neumann
2631: coefficients that define the three string vertex both in the matter
2632: and in the ghost sector. In fact, our construction will be
2633: general and applies to three string interactions other than the
2634: one used in OSFT. We will determine the full structure of
2635: the three string vertex, except for the matter zero modes.
2636:
2637: In the Fock space representation of the vertex, we must find
2638: a state $\langle V_3| \in {\mathcal H}^* \otimes {\mathcal
2639: H}^*\otimes {\mathcal H}^*$
2640: such that for any Fock space states $A, B$ and $C$ one finds that
2641: \begin{equation}
2642: \label{introvert}
2643: \langle A, B, C\rangle \equiv \langle V_3 | A\rangle_{(1)}
2644: | B\rangle_{(2)} | B\rangle_{(3)}\,.
2645: \end{equation}
2646: Since we provided in (\ref{defvertex}) a definition of the left hand
2647: side of the above equation, the vertex $\langle V_3|$ is implicitly
2648: defined. Our procedure will be general in that the functions
2649: $f_r (\xi)$ that map the canonical half-disks to the upper half plane
2650: will be kept arbitrary.
2651: There is a natural ansatz for the vertex:
2652: \begin{eqnarray}
2653: \label{ansatzv}
2654: \langle V_3| &=& {\mathcal N} (\langle 0| c_{-1} c_0 )^{(3)} )
2655: (\langle 0| c_{-1} c_0 )^{(2)}) (\langle 0| c_{-1} c_0 )^{(1)}) \\
2656: && \exp \Bigl( - {1\over 2} \sum_{r,s} \sum_{n,m\geq 1} \,
2657: \alpha_m^{(r)} \, N^{rs}_{mn} \, \alpha_n^{(s)} \Bigr)
2658: \exp \Bigl( \sum_{r,s} \sum_{m\geq 0\atop n\geq 1} \,
2659: b_m^{(r)} \, X^{rs}_{mn} \, c_n^{(s)} \Bigr) \,.\nonumber
2660: \end{eqnarray}
2661: Here ${\mathcal N}$ is a normalization factor, which will be determined
2662: shortly. In fact, its determination is essentially the tachyon computation
2663: of the previous section. Moreover, note that the nontrivial oscillator
2664: dependence in the matter sector is in the form of an exponential of
2665: a quadratic
2666: form. This is a
2667: general result that follows from the free field property of the matter
2668: CFT. Having just a quadratic form is possible also for the ghost sector,
2669: but it requires a careful choice of vacua. This is because there is
2670: a sum rule regarding ghost number-- if the vacua are not chosen
2671: conveniently, extra linear ghost factors are necessary in the vertex.
2672: Since the vertex state $\langle V_3|$ is a bra we use out-vacua,
2673: in particular the vacua
2674: $\langle0|c_{-1} c_0$. This is
2675: quite convenient because the ghost number conservation
2676: law is satisfied when each of the states $A,B$ and $C$ in
2677: (\ref{introvert}) is of ghost number one. Indeed in each of the three
2678: state spaces we must have a total ghost number of three--
2679: two are supplied by the out-vacuum, and one by the in-state.
2680: This clearly allows the nontrivial ghost dependence of the vertex
2681: to be just a pure exponential with zero ghost number.
2682: A final point concerns the sum restrictions over the ghost
2683: oscillators. These simply arise because only oscillators that
2684: do not kill the vacua $\langle 0| c_{-1} c_0$ should appear in
2685: the exponential. Thus for the antighost oscillators $b_m$
2686: we find $m\geq 0$
2687: and for the ghost oscillators $c_n$ we find $n\geq 1$
2688:
2689:
2690: The normalization factor ${\mathcal N}$ can be determined by
2691: finding the overlap of the vertex with three zero momentum
2692: tachyons $c_1|0\rangle$. In this case we have
2693: \begin{equation}
2694: \langle c_1, c_1, c_1\rangle = \langle V_3 | c_1\rangle_{(1)}
2695: | c_1\rangle_{(2)} | c_1\rangle_{(3)} = {\mathcal N}\,,
2696: \end{equation}
2697: since all oscillators in the exponentials kill the zero momentum
2698: tachyon. In (\ref{tachver}) we found the value of this constant
2699: for the case of the OSFT vertex
2700: \begin{equation}
2701: {\mathcal N} = K^3 = \frac{3^{9/2}}{2^6} \,.
2702: % \label{eq:}
2703: \end{equation}
2704: The calculation in the general
2705: case is not any more complicated and it is a good exercise!
2706:
2707: \medskip
2708: \noindent
2709: {\it Exercise: } Show that for arbitrary functions $f_i(\xi)$,
2710: $i=1,2,3$, that map half-disks to the UHP, the constant $\mathcal{N}$ in the
2711: vertex (\ref{ansatzv}) is given by:
2712: \begin{equation}
2713: \label{normv}
2714: {\mathcal N} = {(f_1(0) - f_2(0)) (f_1(0) - f_3(0)) (f_2(0) - f_3(0)) \over
2715: f_1'(0) f_1'(0) f_1'(0) }\,.
2716: \end{equation}
2717:
2718: \medskip
2719: \noindent
2720: Our goal now is to find explicit expressions for the Neumann
2721: coefficients $N^{rs}_{mn}$ and $X^{rs}_{mn}$ in terms of the
2722: functions $f_i$ that define the vertex.
2723:
2724: We begin with the matter sector,
2725: where the following conventions are used
2726: \begin{equation}
2727: \label{modex}
2728: i\partial X(z) = \sum{\alpha_n\over z^{n+1}} \,, \quad
2729: \alpha_n = \oint {dz\over 2\pi i} \, z^n \, i \partial X\,,
2730: \end{equation}
2731: \begin{equation}
2732: \label{opex}
2733: \langle i\partial X (z)\, \, i\partial X(w) \rangle = {1\over (z-w)^2}\,,\quad
2734: [\alpha_n ,
2735: \alpha_m]
2736: = n \delta_{m+n,0}\,.
2737: \end{equation}
2738: To find the matter Neumann coefficients we evaluate
2739: \begin{equation}
2740: \label{expmatt}
2741: M = \langle V_3 | \, \, R\left( i \partial X^{(r)} (z) \,\, i
2742: \partial X^{(s)} (w)
2743: \right)\, c_1^{(1)}|0\rangle_{(1)}\,
2744: c_1^{(2)} | 0\rangle_{(2)} \, c_1^{(3)}|0\rangle_{(3)}
2745: \end{equation}
2746: in
2747: two different ways. In here $R( \, \dots\,)$ denotes radial ordering,
2748: necessary when $r=s$. For our first
2749: computation we use the mode expansion (\ref{modex}) of the conformal
2750: fields to find that
2751: \begin{equation}M = \langle V_3 | \Bigl( \sum_{m,n} {1\over
2752: z^{-m+1}} {1\over w^{-n+1} }
2753: \, \alpha_{-m}^{(r)} \,
2754: \alpha_{-n}^{(s)}\, + {\delta^{rs}\over (z-w)^2} \Bigr)c_1^{(1)}
2755: |0\rangle_{(1)}\,
2756: c_1^{(2)} | 0\rangle_{(2)} \, c_1^{(3)}|0\rangle_{(3)}
2757: \end{equation}and the oscillator form (\ref{ansatzv}) of
2758: the vertex to obtain
2759: \begin{equation}\label{firsteval}
2760: M = -{\mathcal N} \sum_{m,n} z^{m-1}w^{n-1} \, m n N^{rs}_{mn} +
2761: {\mathcal N} {\delta^{rs}\over (z-w)^2} \,.
2762: \end{equation}In the second evaluation we first rewrite $M$ as
2763: \begin{equation}
2764: \label{rewritem}
2765: M = \langle V_3 | \, \, i \partial X^{(r)} (z) \,\, i \partial X^{(s)} (w) \,
2766: c^{(1)}(0)c^{(2)}(0)c^{(3)}(0) |0\rangle_{(1)}\,
2767: | 0\rangle_{(2)} \, |0\rangle_{(3)}\,,
2768: \end{equation}
2769: and reinterpret as a correlator,
2770: in the spirit of (\ref{defvertex}):
2771: \begin{eqnarray}
2772: M = \Bigl\langle f_r\circ i \partial X (z)
2773: \, f_s \circ i \partial X (w) \, f_1 \circ c(0) \, f_2 \circ c(0)
2774: \, f_3 \circ
2775: c(0)\Bigr\rangle\,.
2776: \end{eqnarray}
2777: The ghost part of this correlator gives the factor ${\mathcal N}$. The matter
2778: part, using $i\partial X(z) = i\partial X (f(z)) {df\over dz}$, and
2779: (\ref{opex}) finally gives
2780: \begin{equation}\label{seceval}
2781: M= {\mathcal N} \, f'_r(z) \, f'_s(w) \Bigl\langle
2782: \,i\partial X (f_r(z))\,\, i\partial X (f_s(w))\,\Bigr\rangle
2783: = {\mathcal N} {\, f'_r(z) \, f'_s(w)\over ( f_r(z) - f_s(w))^2}\,.
2784: \end{equation}Equating the results (\ref{firsteval}) and
2785: (\ref{seceval}) of the two
2786: evaluations of $M$ we obtain:
2787: \begin{equation}
2788: \sum_{m,n} z^{m-1}w^{n-1} \, m n N^{rs}_{mn} - {\delta^{rs}\over (z-w)^2}= -
2789: {\, f'_r(z) \, f'_s(w)\over ( f_r(z) - f_s(w))^2}\,.
2790: \end{equation}
2791: It is now simple to pick up the coefficients $N^{rs}_{mn}$ by
2792: contour integration over small circles surrounding $z=0$ and
2793: $w=0$. The second term on the left-hand side gives no contribution,
2794: and one finally finds
2795: \begin{equation}
2796: \label{neumannmatter}
2797: N^{rs}_{mn} = -{1\over mn} \oint_0{dz\over 2\pi i}{1\over z^m}
2798: \oint_0{dw\over 2\pi i}{1\over w^n} \, {\, f'_r(z) \, f'_s(w)\over ( f_r(z) -
2799: f_s(w))^2}\,.
2800: \end{equation}
2801: This is the desired expression for the Neumann coefficients of the
2802: matter sector. They can be used for an arbitrary vertex.
2803: The above contour integrals are straightforward to compute and
2804: they can be easily done by a computer in a series expansion. In terms
2805: of residues the expression above is equivalent to
2806: \begin{equation}
2807: \label{neuresidue}
2808: N^{rs}_{mn} = -{1\over mn} \hbox{Res}_{z=0} \hbox{Res}_{w=0}
2809: \Bigl[ {1\over z^m}
2810: {1\over w^n} \, {\, f'_r(z) \, f'_s(w)\over ( f_r(z) -
2811: f_s(w))^2}\Bigr]
2812: \end{equation}
2813:
2814: \medskip
2815: \noindent
2816: {\it Exercise:} Show that the contour integrals in
2817: (\ref{neumannmatter}) can be evaluated in any order.
2818: Do this both for the case when $r\not= s$ and for the
2819: case when $r=s$.
2820:
2821: \medskip
2822: We now turn to the calculation of the
2823: ghost Neumann coefficients $X^{rs}_{mn}$.
2824: For this we need mode expansions and two point functions for
2825: the ghost CFT:
2826: \begin{equation}
2827: c(z) = \sum_n {c_n\over z^{n-1}} \,, \quad b(z) = \sum_n {b_n\over z^{n+2}}
2828: \,, \quad \langle c(z) b(w) \rangle = {1\over z -w}\,.
2829: \end{equation}
2830: The strategy is once more based on the computation of a certain
2831: expression in two different ways. Indeed, we consider the overlap
2832: \begin{equation}
2833: G = \langle V_3 | \, R\left( b^{(s)} (z) \,\, c^{(r)} (w) \right)\, |
2834: \,c_1^{(1)}|0\rangle_{(1)}\,
2835: c_1^{(2)} | 0\rangle_{(2)} \, c_1^{(3)}|0\rangle_{(3)}
2836: \end{equation}
2837: and first evaluate it by using the mode expansion of the antighost
2838: and ghost fields, and then the explicit expression for the vertex in
2839: (\ref{ansatzv}). In this way we find
2840: \begin{eqnarray}
2841: \label{fghost}
2842: G &=& \langle V_3 |\Bigl( \sum_{m,n} {1\over z^{-n+2}} {1\over w^{-m-1} }
2843: \, b_{-n}^{(s)} \, c_{-m}^{(r)} +{w\over
2844: z(z-w)}\Bigr) \,\,c_1^{(1)}|0\rangle_{(1)}\, c_1^{(2)} | 0\rangle_{(2)} \,
2845: c_1^{(3)}|0\rangle_{(3)}
2846: \nonumber \\ &=& {\mathcal N} \sum_{m,n} z^{-n+2}w^{-m-1} \, X^{rs}_{mn}
2847: + \mathcal{N} {w\over
2848: z(z-w)}\,.
2849: \end{eqnarray}
2850: In the second computation $G$ is interpreted as a correlator and we
2851: have
2852: \begin{eqnarray}
2853: G &=& \Bigl\langle f_s\circ b(z)
2854: \,\, f_r\circ c (w)\,\, f_1\circ c(0) \,
2855: f_2\circ c(0) \, f_3\circ c(0)
2856: \Bigr\rangle \\
2857: &=& {(f'_s (z))^2\over f'_r(w) } {1\over f'_1(0)
2858: f'_2(0) f'_3(0)} \Bigl\langle b(f_s(z))
2859: \,\, c (f_r(w))\,\, c(f_1(0)) \,c(f_2(0)) \,c(f_3(0))
2860: \Bigr\rangle\,, \nonumber
2861: \end{eqnarray}
2862: where we used the standard conformal maps of the relevant
2863: operators, all of which are primary. The final correlator is in the upper half
2864: plane and all the field arguments refer to the coordinates in the
2865: upper half plane. The correlator can be calculated by using OPE's, but it is
2866: simpler to use the singularity structure and derive the normalization from a
2867: special configuration. Note, for example, that there must be zeroes when any
2868: pair of
2869: $c$ fields approach each other. In particular, this will include a
2870: factor $(f_1(0) - f_2(0)) (f_1(0) - f_3(0)) (f_2(0) - f_3(0))$
2871: as in ${\mathcal N}$ (see (\ref{normv})). We will also have poles
2872: when the antighost approaches any ghost. These considerations
2873: imply that
2874: \begin{equation}\label{sghost}
2875: G ={\mathcal N} {(f'_s (z))^2\over f'_r(w) } {1\over f_s(z) - f_r(w)}
2876: {\prod_{I=1}^3 (f_r(w) -f_I(0)) \over \prod_{J=1}^3 (f_s(z) - f_J(0))}\,.
2877: \end{equation}We can now equate the results
2878: obtained in (\ref{fghost}) and (\ref{sghost}). Picking up
2879: the coefficients via contour integration, and noting that the
2880: second term on the right-hand side of (\ref{fghost}) does not
2881: contribute for the relevant values of $m$ and $n$, we find
2882: \begin{equation}
2883: X^{rs}_{mn} = \oint{dz\over 2\pi i}{1\over z^{n-1}}
2884: \oint{dw\over 2\pi i}{1\over w^{m+2} } \, {(f'_s (z))^2\over f'_r(w) } {1\over
2885: f_s(z) - f_r(w)}{\prod_{I=1}^3 (f_r(w) -f_I(0)) \over \prod_{J=1}^3 (f_s(z) -
2886: f_J(0)}\,.
2887: \end{equation}
2888: This is the general result for the ghost Neumann coefficients.
2889: Again, for any vertex they are easily calculated by power series
2890: expansions and picking up residues. For particular vertices one
2891: can simplify somewhat the above expressions and find interesting
2892: relations. In fact, a fair amount of work can be done for the OSFT
2893: vertex in simplifying the above results. One can show that the
2894: matrices $N$ and $X$ are related, and while no closed form expressions
2895: are known for the coefficients, they can be generated quite
2896: efficiently from the simpler expressions.
2897:
2898: Since any specific Neumann coefficient can
2899: be calculated exactly with
2900: a finite number of operations, the exact computation
2901: of $\langle A, B, C\rangle$
2902: for any three Fock space states $A$, $B$, and $C$ requires a finite number
2903: of operations, as well.
2904:
2905:
2906: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2907: % SFT: overlaps: WT
2908: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2909:
2910: \section{SFT action: oscillator approach}
2911: \label{sec:overlaps}
2912:
2913: In this section, we give a more detailed discussion of Witten's open
2914: bosonic string field theory from the oscillator point of view.
2915: The main goal of this section is to explicitly formulate the OSFT
2916: action in the string Fock space, where the action
2917: (\ref{eq:SFT-action}) takes the form
2918: \begin{equation}
2919: S = -\frac{1}{2}\langle V_2 | \Psi, Q \Psi \rangle
2920: -\frac{g}{3} \langle V_3 | \Psi, \Psi, \Psi \rangle\,.
2921: \label{eq:action-Fock-1}
2922: \end{equation}
2923: In this expression, $\langle V_2 |$ and $\langle V_3 |$ are elements
2924: of the two-fold and three-fold product of the dual Fock space $({\mathcal
2925: H}^*)^2$ and $({\mathcal H}^*)^3$, respectively. These
2926: objects defined
2927: in terms of the string Fock space give a rigorous definition to the
2928: abstract action (\ref{eq:SFT-action}) through the replacement
2929: \begin{eqnarray*}
2930: \langle V_2 | A, B \rangle & \rightarrow & \int A\star B\\
2931: \langle V_3 | A, B, C \rangle & \rightarrow & \int A\star B\star C \,.
2932: \end{eqnarray*}
2933:
2934: Subsection \ref{sec:warmup} is a warmup, in which we review some basic
2935: features of the simple harmonic oscillator and discuss squeezed
2936: states. In subsection \ref{sec:split} we relate modes on the full
2937: string to modes on half strings, giving formulae needed to compute the
2938: three-string vertex.
2939: In subsection \ref{sec:v2} we
2940: derive the two-string vertex in oscillator form, and in subsection
2941: \ref{sec:v3} we give an explicit formula for the three-string vertex.
2942: In subsection \ref{sec:calculating} we put these pieces together and
2943: discuss the calculation of the full SFT action.
2944:
2945: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2946: % Warmup
2947: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2948:
2949: \subsection{Squeezed states and the simple harmonic oscillator}
2950: \label{sec:warmup}
2951:
2952: Let us consider a simple harmonic oscillator with annihilation
2953: operator
2954: \begin{equation}
2955: a = -i \left( \sqrt{\frac{\alpha}{ 2}}x + \frac{1}{ \sqrt{2 \alpha}}
2956: \partial_x \right)
2957: % % \label{eq:}
2958: \end{equation}
2959: where $\alpha$ is an arbitrary constant. The oscillator ground state
2960: is associated with the wavefunction
2961: \begin{equation}
2962: | 0 \rangle \rightarrow
2963: \left( \frac{\alpha}{ \pi} \right)^{1/4} e^{-\alpha x^2/2}\,.
2964: % % \label{eq:}
2965: \end{equation}
2966: In the harmonic oscillator basis $|n \rangle$, the Dirac position basis
2967: states $| x \rangle$ have a squeezed state form
2968: \begin{equation}
2969: | x \rangle = \left( \frac{\alpha}{ \pi} \right)^{1/4}
2970: \exp\left(-\frac{\alpha}{ 2}x^2
2971: -i \sqrt{2 \alpha} a^{\dagger} x +\frac{1}{2} (a^{\dagger})^2 \right)
2972: | 0 \rangle\,.
2973: % % \label{eq:}
2974: \end{equation}
2975:
2976: A general wavefunction is associated with a state through the
2977: correspondence
2978: \begin{equation}
2979: f (x) \rightarrow \int_{-\infty}^\infty dx \;f (x) | x \rangle \,.
2980: \label{eq:function-state}
2981: \end{equation}
2982: In particular, we have
2983: \begin{eqnarray}
2984: \delta (x) & \rightarrow &
2985: \left( \frac{\alpha}{ \pi} \right)^{1/4}
2986: \exp\left(\frac{1}{2} (a^{\dagger})^2 \right)
2987: | 0 \rangle \,,\label{eq:squeezed-d1}
2988: \\
2989: 1 & \rightarrow & \int dx \; | x \rangle =
2990: \left( \frac{4\pi}{ \alpha} \right)^{1/4}
2991: \exp\left(-\frac{1}{2} (a^{\dagger})^2 \right)
2992: | 0 \rangle\,.\nonumber
2993: \end{eqnarray}
2994: This shows that the delta and constant functions both have squeezed
2995: state representations in terms of the harmonic oscillator basis.
2996: The norm of a squeezed state
2997: \begin{equation}
2998: | s \rangle =\exp\left(\frac{1}{2} s (a^{\dagger})^2 \right)
2999: | 0 \rangle
3000: % % \label{eq:}
3001: \end{equation}
3002: is given by
3003: \begin{equation}
3004: \langle s| s \rangle =
3005: \frac{1}{ \sqrt{1-s^2}}\,.
3006: % % \label{eq:}
3007: \end{equation}
3008: The states (\ref{eq:squeezed-d1}) are non-normalizable, but since they
3009: have $s=\pm 1$, they are right on the border of normalizability.
3010: The states
3011: (\ref{eq:squeezed-d1}) can be used to calculate, just like we do with
3012: the Dirac basis states $| x\rangle$, which lie outside
3013: %although technically not well-defined states in
3014: the single-particle Hilbert space.
3015:
3016: It will be useful for us to generalize the foregoing considerations in
3017: several ways. A particularly simple generalization arises when we
3018: consider a pair of degrees of freedom $x, y$ described by a two-harmonic
3019: oscillator Fock space basis. In such a basis, repeating the preceding
3020: analysis leads us to a function-state correspondence for the delta
3021: functions relating $x, y$ of the form
3022: \begin{equation}
3023: \delta (x \pm y) \rightarrow
3024: \exp\left(\pm a^{\dagger}_{(x)}a^{\dagger}_{(y)}\right)
3025: \left( | 0 \rangle_x \otimes | 0 \rangle_y \right)\,.
3026: \label{eq:squeezed-two}
3027: \end{equation}
3028: Note that this result is independent of $\alpha$; like the $\delta$
3029: function, the resulting state is again non-normalizable.
3030: we will find these squeezed state expressions very useful in
3031: describing the two- and three-string vertices of Witten's open string
3032: field theory. It is worth pointing out here that there are several
3033: ways of deriving (\ref{eq:squeezed-two}). The most straightforward
3034: way is to carry out a two-dimensional Gaussian integral analogous to
3035: (\ref{eq:squeezed-d1}). We can also derive (\ref{eq:squeezed-two})
3036: indirectly, however (at least up to an overall constant) from the
3037: following argument. From the general result that delta functions give
3038: squeezed states, we expect that up to an overall constant
3039: \begin{eqnarray}
3040: \lefteqn{\delta (x \pm y) \rightarrow} \\
3041: & | D_{\pm} \rangle = &
3042: \exp\left(\pm\frac{1}{2}\left[
3043: A a^{\dagger}_{(x)}a^{\dagger}_{(x)} +
3044: 2 B a^{\dagger}_{(x)}a^{\dagger}_{(y)}+
3045: C a^{\dagger}_{(y)}a^{\dagger}_{(y)} \right]
3046: \right)
3047: \left( | 0 \rangle_x \otimes | 0 \rangle_y \right)\,. \nonumber
3048: \end{eqnarray}
3049: The state associated with the delta function must satisfy the
3050: constraints
3051: \begin{eqnarray}
3052: (x \pm y) | D_{\pm} \rangle & = & 0\\ \label{eq:delta-constraints}
3053: (p_x \mp p_y) | D_{\pm} \rangle& = & 0 \,. \nonumber
3054: \end{eqnarray}
3055: Rewriting $x, p_x$ in terms of $a_{(x)}, a^{\dagger}_{(x)}$ and
3056: similarly for $y, p_y$,
3057: these conditions impose the constraints
3058: \begin{eqnarray}
3059: \left[(A \pm B -1) a_{(x)}^{\dagger}+( B \pm C \pm 1)
3060: a_{( y)}^{\dagger} \right] | D_{\pm} \rangle& = & 0\\
3061: \left[( A \mp B + 1) a_{(x)}^{\dagger}+( B \mp C \mp 1)
3062: a_{(y)}^{\dagger} \right]
3063: | D_{\pm} \rangle & = & 0 \,, \nonumber
3064: \end{eqnarray}
3065: from which it follows that $A = C = 0$ and $B = \pm 1$, reproducing
3066: (\ref{eq:squeezed-two}) up to an overall constant. We will use this
3067: indirect method, following Gross and Jevicki, to derive the
3068: three-string vertex in subsection \ref{sec:v3}.
3069:
3070: \subsection{Half-string modes}
3071: \label{sec:split}
3072:
3073: For many computations it is useful to think of the string as being
3074: ``split'' into a left half and a right half. Formally, the string
3075: field can be expressed as a functional $\Psi[L, R]$, where $L, R$ describe
3076: the left and right parts of the string. This is a very appealing
3077: idea, since it leads to a simple picture of the star product in terms of
3078: matrix multiplication
3079: \begin{equation}
3080: (\Psi \star \Phi)[L, R] = \int{\mathcal D} A \; \Psi[L, A] \Phi[A, R] \,.
3081: % \label{eq:}
3082: \end{equation}
3083: While there has been quite a bit of work aimed at making this ``split
3084: string'' formalism precise~\cite{split,Gross-Taylor-I,Gross-Taylor-II},
3085: the technical details in this approach become quite complicated when one
3086: attempts to precisely deal with the string midpoint where the left and right
3087: parts of the string attach. In particular, the BRST operator $Q_B$ becomes
3088: rather awkward in this formulation.
3089:
3090: Nonetheless, some of the structure of the star product encoded in the
3091: three-string vertex is easiest to understand using the half-string
3092: formalism, and many formulae related to the 3-string vertex are most
3093: easily expressed in terms of the linear map from full-string modes
3094: to half-string modes. In this subsection we discuss this linear map,
3095: encoded in a matrix $X$,
3096: which we use in subsection \ref{sec:v3} to give an explicit formulae
3097: for the three-string vertex.
3098:
3099: Recall that the
3100: matter fields are expanded in modes through
3101: \begin{equation}
3102: x (\sigma) = x_0 + \sqrt{2} \sum_{n = 1}^{ \infty} x_n \cos n \sigma\,.
3103: \label{eq:mode-decomposition-2}
3104: \end{equation}
3105: (We suppress Lorentz indices in most of this section for clarity.)
3106: We are interested in considering an analogous expansion of the left
3107: and right halves of the string. We expand in odd modes with Neumann
3108: boundary conditions at the ends of the string, and Dirichlet boundary
3109: conditions at the string midpoint:
3110: \begin{eqnarray}
3111: l (\sigma) & = & x (\sigma) =
3112: \sqrt{2} \sum_{k = 0}^{ \infty} l_{2k + 1}
3113: \cos (2k + 1) \sigma,\quad \sigma < \pi/2 \\
3114: r (\sigma) & = & x (\pi -\sigma) =
3115: \sqrt{2} \sum_{k = 0}^{ \infty} r_{2k + 1}
3116: \cos (2k + 1) \sigma,\quad \sigma < \pi/2 \,. \nonumber
3117: \end{eqnarray}
3118: Note that there are subtleties associated with the midpoint in this
3119: expansion. For example, while we have taken $l (\pi/2)$ to formally
3120: vanish, by choosing coefficients like $l_{2k + 1} = (-1)^k2 \sqrt{2}
3121: a/(2k + 1) \pi$ we have $l (\sigma) = a, \forall \sigma < \pi/2$, so
3122: $\lim_{\sigma \rightarrow \pi/2_-}= a$. These subtleties become
3123: important when dealing with the full theory, but are not important in
3124: the calculation we carry out below of the three-string vertex.
3125:
3126: Let us define the quantities
3127: \begin{eqnarray}
3128: X_{2k+1,2n}=X_{2n,2k+1} & = & { 4(-1)^{k+n}(2k+1)\over
3129: \pi\left({(2k+1)^2-4n^2}\right)} \;\;\; \quad (n \neq 0)\, , \label{eq:X}\\
3130: \quad X_{ 2k+1,0} =X_{0,2k+1} & = &
3131: { 2 \sqrt{2}(-1)^{k}\over \pi{(2k+1)}}\, .\nonumber
3132: \end{eqnarray}
3133: The matrix
3134: \begin{equation}
3135: X =
3136: \begin{pmatrix} %\pmatrix{
3137: 0 & X_{2k+1,2n} \\
3138: X_{2n, 2k+1} & 0
3139: \end{pmatrix}
3140: \equiv \left(\begin{array}{cc}
3141: 0 & X_{oe}\\
3142: X_{eo} & 0
3143: \end{array}\right) \,
3144: \label{eq:matX}
3145: \end{equation}
3146: where $e, o$ refer to the set of even and odd indices respectively,
3147: is manifestly symmetric and turns out to be orthogonal: \ \ $X=X^T=X^{-1}$.
3148: Performing a Fourier decomposition we can relate the full-string and
3149: half-string modes through
3150: \begin{eqnarray}
3151: x_{2n+1} &= &\half\left(l_{2n+1}-r_{2n+1}\right) \, ,
3152: \label{eq:xrel}\\
3153: x_{2n}% -\delta_{n0}x
3154: & = & \half
3155: \sum_{k=0}^{\infty}X_{2n,2k+1}\left(l_{2k+1}+r_{2k+1}\right)\,,
3156: \nonumber
3157: \end{eqnarray}
3158: We can invert (\ref{eq:xrel}) to derive
3159: \begin{eqnarray}
3160: l_{2k+1} &=
3161: &x_{2k+1}+\sum_{n=0}^{\infty}X_{2k+1,2n}%(x_{2n} -\delta_{n0}x) \, ,
3162: x_{2n} \, ,
3163: \label{eq:half-full}\\
3164: r_{2k+1} & = &
3165: -x_{2k+1}+\sum_{n=0}^{\infty}X_{2k+1,2n}%(x_{2n} -\delta_{n0}x) \ .
3166: x_{2n} \, .
3167: \nonumber
3168: \end{eqnarray}
3169: One must be careful with the order of summation
3170: in sequences of coefficients which do not go to zero
3171: faster than $1/n$: different orders of summation
3172: may give different results. Fortunately, such associativity
3173: anomalies are not relevant for the calculations we do here.
3174:
3175:
3176:
3177:
3178: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3179: % V2
3180: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3181:
3182: \subsection{The two-string vertex $\langle V_2 |$}
3183: \label{sec:v2}
3184:
3185: We can immediately apply the oscillator formulae from subsection
3186: \ref{sec:warmup} to calculate the two-string vertex. Using the mode
3187: decomposition (\ref{eq:mode-decomposition-2}), we associate the string
3188: field functional $\Psi[x (\sigma)]$ with a function $\Psi (\{x_n\})$
3189: of the infinite family of string oscillator mode amplitudes. The
3190: overlap integral combining (\ref{eq:integral-p}) and (\ref{eq:mult})
3191: can then be expressed in modes as
3192: \begin{equation}
3193: \int \Psi \star \Phi = \int \prod_{n = 0}^{ \infty} dx_ndy_n
3194: \; \delta (x_n-(-1)^ny_n) \Psi (\{x_n\}) \Phi (\{y_n\})\,.
3195: % % \label{eq:}
3196: \end{equation}
3197: Geometrically this just encodes the overlap condition $x (\sigma) = y
3198: (\pi -\sigma)$ described through
3199:
3200: \begin{center}
3201: \centering
3202: \begin{picture}(100,40)(- 50,- 20)
3203: \put( 20,2){\vector( -1,0){40}}
3204: \put( -20,-2){\vector( 1,0){40}}
3205: \put(0,-10){\makebox(0,0){$\Psi$}}
3206: \put(0, 10){\makebox(0,0){$\Phi$}}
3207: \end{picture}
3208: \end{center}
3209: It follows from (\ref{eq:squeezed-two}) that we can write the
3210: two-string vertex as the squeezed state
3211: \begin{equation}
3212: \langle V_2 |_{{\rm matter}} =
3213: \left(\langle 0 | \otimes \langle 0 |\right)
3214: \exp\Bigl( \sum_{n, m = 0}^{ \infty}
3215: -a_n^{(1)} C_{nm} a_m^{ (2)} \Bigr)\,,
3216: \label{eq:v2o}
3217: \end{equation}
3218: where $C_{nm} = \delta_{nm} (-1)^n$ is an infinite-size matrix
3219: connecting the oscillator modes of the two single-string Fock spaces,
3220: and the sum is taken over all oscillator modes, including zero. In the
3221: expression (\ref{eq:v2o}), we have used the formalism in which $| 0
3222: \rangle$ is the vacuum annihilated by $a_0$. To translate this
3223: expression into a momentum basis, we use only $n, m > 0$, and replace
3224: \begin{equation}
3225: \left(\langle 0 | \otimes \langle 0 |\right)
3226: \exp\left(-a^{(1)}_0a^{(2)}_0\right) \rightarrow
3227: \int d^{26} p
3228: \left( \langle 0; p | \otimes \langle 0; -p | \right)\,.
3229: % % \label{eq:}
3230: \end{equation}
3231:
3232: The extension of this analysis to ghosts is straightforward. For the
3233: ghost and antighost respectively, the overlap conditions corresponding
3234: with
3235: $x_1 (\sigma) = x_2 (\pi -\sigma)$ are ~\cite{Gross-Jevicki-12}
3236: $c_1 (\sigma) = -c_2 (\pi -\sigma)$ and
3237: $b_1 (\sigma) = b_2 (\pi -\sigma)$. This leads to the overall formula
3238: for the two-string vertex
3239: \begin{eqnarray}
3240: \langle V_2 | & = & \int d^{26} p
3241: \left( \langle 0; p | \otimes \langle 0; -p | \right)
3242: (c^{(1)}_0 + c^{(2)}_0) \label{eq:v2}\\
3243: & & \hspace{0.3in} \times
3244: \exp\left(
3245: -\sum_{n = 1}^{ \infty} (-1)^n
3246: [a^{(1)}_n a^{(2)}_n+c^{(1)}_n b^{(2)}_n+c^{(2)}_n b^{(1)}_n] \right)\,.
3247: \nonumber
3248: \end{eqnarray}
3249: This expression for the two-string vertex can also be derived directly
3250: from the conformal field theory approach, computing the two-point
3251: function of an arbitrary pair of states on the disk.
3252:
3253: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3254: % v3
3255: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3256:
3257: \subsection{The three-string vertex $| V_3 \rangle$}
3258: \label{sec:v3}
3259:
3260: The three-string vertex, which is associated with the three-string
3261: overlap
3262: \begin{center}
3263: \centering
3264: \begin{picture}(100,60)(- 50,- 30)
3265: \put( 20,14){\line( -2,-1){20}}
3266: \put(0, 4){\vector( -2,1){20}}
3267: \put(-20,9){\line( 2, -1){18}}
3268: \put( -2, 0 ){\vector( 0, -1){ 20}}
3269: \put( 2,-20){\line( 0, 1){ 20}}
3270: \put(2,0){\vector( 2,1){18}}
3271: \put(-15,-10){\makebox(0,0){$\Psi_2$}}
3272: \put(0, 15){\makebox(0,0){$\Psi_1$}}
3273: \put(15, -10){\makebox(0,0){$\Psi_3$}}
3274: \end{picture}
3275: \end{center}
3276: can be computed in a very similar fashion to the two-string vertex
3277: above. The details of the calculation, however, are significantly
3278: more complicated. In this subsection we follow the original approach
3279: of Gross and Jevicki~\cite{Gross-Jevicki-12}; similar approaches were
3280: taken by other authors~\cite{cst,Samuel}. The method used by Gross and
3281: Jevicki is essentially the method used in (\ref{eq:squeezed-two}) to
3282: write a delta function of two variables in oscillator form by imposing
3283: the constraints (\ref{eq:delta-constraints}) on a general squeezed
3284: state. The 3-string vertex can also be computed by explicitly
3285: performing~\cite{Ohta,Shelton} the relevant Gaussian
3286: integrals.\footnote{Another approach to the cubic vertex has been
3287: explored extensively. By diagonalizing the Neumann matrices, the star
3288: product takes the form of a continuous Moyal
3289: product~\cite{Bars-original,Moyal}. This simplifies the vertex but
3290: complicates the propagator. For a recent discussion of this work,
3291: applications of this approach, and
3292: further references, see the review of Bars~\cite{Bars}.}
3293:
3294: {}From the general structure of the overlap conditions it is clear that,
3295: like the two-string vertex, the
3296: three-string vertex takes the form of a squeezed state:
3297: \begin{eqnarray}
3298: \lefteqn{ | V_3 \rangle = \kappa\int d^{26} p^{(1)}d^{26} p^{(2)}d^{26}
3299: p^{(3)}} \nonumber \\
3300: & &
3301: \times\exp\Bigl(
3302: -\frac{1}{2}\sum_{r, s = 1}^{3}
3303: [a^{(r)}_m V^{rs}_{mn} a^{(s)}_n+
3304: 2 a^{(r)}_m V^{rs}_{m0} p^{(s)}+
3305: p^{(r)} V^{rs}_{00} p^{(s)}+
3306: c^{(r)}_m X^{rs}_{mn} b^{(s)}_n] \Bigr)\,, \nonumber\\
3307: & &
3308: \times
3309: \delta (p^{(1)} + p^{(2)} + p^{(3)}) c_0^{(1)}c_0^{(2)}c_0^{(3)}
3310: \left( | 0; p^{(1)} \rangle \otimes
3311: | 0; p^{(2)} \rangle \otimes | 0; p^{(3)} \rangle \right)
3312: \label{eq:v3}
3313: \end{eqnarray}
3314: where $\kappa ={\mathcal N} = K^3 = 3^{9/2}/2^6$, and where the Neumann
3315: coefficients $V^{rs}_{mn}$ and $X^{rs}_{mn}$ are constants. Writing the
3316: momentum basis states in oscillator form
3317: \begin{equation}
3318: | p \rangle = \frac{1}{ ( \pi)^{1/4}}
3319: \exp \left[-\frac{1}{2}p^2
3320: + \sqrt{2}a_0^{\dagger} p -\frac{1}{2}(a^{\dagger}_0)^2 \right]
3321: | 0 \rangle,
3322: \label{eq:p-basis}
3323: \end{equation}
3324: we can write matter part of the 3-string vertex as
3325: \begin{equation}
3326: \label{eq:vertex-0}
3327: | V_3 \rangle =
3328: \left( \frac{2 \; \pi^{1/4}}{ \sqrt{3} (1 + V_{00})} \right)^{26}
3329: \exp \Bigl(-\frac{1}{2}\sum_{r, s \leq 3}
3330: \; \sum_{m, n \geq 0} V'^{rs}_{m n} (a^{(r)\dagger}_m \cdot
3331: a^{(s) \dagger}_n)\Bigr)
3332: \left( | 0 \rangle \otimes | 0 \rangle \otimes
3333: | 0 \rangle \right) \,,
3334: \end{equation}
3335: where $V_{00} = V^{rr}_{00}$ and
3336: \begin{eqnarray}
3337: V'^{rs}_{mn} & = & V^{rs}_{mn}
3338: -\frac{ 1}{1 + V_{00}} \sum_{t}V^{rt}_{m0}V^{ts}_{0n} \nonumber\\
3339: V'^{rs}_{m0} & = & V'^{sr}_{0m} =
3340: \frac{\sqrt{2}}{1 + V_{00}}
3341: V^{rs}_{m0} \label{eq:v-relations}\\
3342: V'^{rs}_{00} & = &
3343: \frac{2}{3 (1 + V_{00})}
3344: + \delta^{rs} \left( 1-2/(1 + V_{00}) \right)\,.
3345: \nonumber
3346: \end{eqnarray}
3347:
3348: We now want to determine the coefficients $V'^{rs}_{mn}$ by using
3349: overlap conditions analogous to (\ref{eq:delta-constraints}). It is
3350: convenient to use a ${\bf Z}_3$ Fourier decomposition of the three
3351: string modes $x^{(i)}(\sigma)$
3352: \begin{eqnarray}
3353: Q & = & \frac{1}{ \sqrt{3}} (x^{(1)} + \omega x^{(2)} + \omega^2
3354: x^{(3)})
3355: \label{eq:Fourier-q}\\
3356: Q^{(3)} & = & \frac{1}{ \sqrt{3}} (x^{(1)} + x^{(2)} +
3357: x^{(3)}) \nonumber
3358: \end{eqnarray}
3359: where $\omega = \exp\left(2 \pi i/3\right)$.
3360: The definitions (\ref{eq:Fourier-q}) can be used to define $Q
3361: (\sigma)$ in terms of $x^{(i)}(\sigma)$ as well as to define $Q_n$ in
3362: terms of $x^{(i)}_n$ (and similarly for $Q^{(3)}$); henceforth by $Q$
3363: we denote the collection of full-string modes $Q_n$ (and similarly for
3364: $Q^{(3)}$).
3365: We can relate the
3366: full-string modes $Q, Q^{(3)}$ to half-string modes $L, R, L^{(3)},
3367: R^{(3)}$ through the equations (\ref{eq:half-full}).
3368: In terms of these
3369: variables, the overlap conditions are
3370: \begin{eqnarray}
3371: L-\omega R & = & 0,\\
3372: L^{(3)} -R^{(3)}& = & 0 \,. \nonumber
3373: \end{eqnarray}
3374: In terms of the even and odd full-string modes (which, using the same
3375: notation as in section \ref{sec:split}, we denote by $Q_{e, o}$)
3376: these conditions are
3377: expressed as
3378: \begin{equation}
3379: Q_o -i \sqrt{3} X_{oe} Q_e = 0\,,
3380: \label{eq:condition1}
3381: \end{equation}
3382: and
3383: \begin{equation}
3384: Q^{(3)}_o = 0 \,.
3385: \label{eq:condition2}
3386: \end{equation}
3387: Multiplying by $X_{eo}$, (\ref{eq:condition1}) can be rewritten as
3388: \begin{equation}
3389: \frac{i}{ \sqrt{3}} X_{eo} Q_o + Q_e = 0 \,.
3390: % \label{eq:}
3391: \end{equation}
3392: This can be combined with (\ref{eq:condition1}) and written in the
3393: simpler form
3394: \begin{equation}
3395: (1-Y) Q = 0,
3396: \label{eq:o1}
3397: \end{equation}
3398: where
3399: \begin{equation}
3400: Y = -\frac{1}{2}C + \frac{\sqrt{3}}{2} X \,.
3401: % \label{eq:}
3402: \end{equation}
3403: Note that $Y^2 = 1$.
3404: Similarly, we can write (\ref{eq:condition2})
3405: as
3406: \begin{equation}
3407: (1-C) Q^{(3)} = 0 \,.
3408: \label{eq:o2}
3409: \end{equation}
3410: Equations (\ref{eq:o1}) and (\ref{eq:o2}) are the essential overlap
3411: equations satisfied by the three-string vertex. Writing the
3412: three-string vertex as a squeezed state in terms of oscillators $A,
3413: A^{(3)}$ related to the string oscillators $a^{(i)}$ through the
3414: analogue of (\ref{eq:Fourier-q}), we then have
3415: \begin{equation}
3416: | V_3 \rangle \sim\exp\left( -A^{\dagger} U \bar{A}^{\dagger}
3417: -\frac{1}{2}(A^{(3)})^{\dagger} C(A^{(3)})^{\dagger}\right)\,,
3418: \label{eq:vu}
3419: \end{equation}
3420: where $U$ satisfies the overlap constraint
3421: (\ref{eq:o1}).
3422: Recall that the string modes are proportional to
3423: \begin{equation}
3424: x \sim E (a-a^{\dagger})
3425: % \label{eq:}
3426: \end{equation}
3427: where
3428: \begin{equation}
3429: E_{mn} = \delta_{mn} \frac{1}{ \sqrt{m}}, \quad m \neq 0, \;\;\;\;\;
3430: E_{00} = 1/\sqrt{2} \,.
3431: % \label{eq:}
3432: \end{equation}
3433: Thus, from (\ref{eq:o1}) and (\ref{eq:vu}) we see that $U$ must
3434: satisfy the overlap constraint
3435: \begin{equation}
3436: (1-Y) E (1 + U) = 0 \,.
3437: \label{eq:oo1}
3438: \end{equation}
3439: As we discussed in the last part of subsection \ref{sec:warmup},
3440: associated with this constraint there is an analogous constraint on
3441: the derivatives in the perpendicular direction. Since $Y^2 = 1$, we
3442: have
3443: \begin{eqnarray}
3444: (1 + Y) (1-Y) & = & (1-Y) (1 + Y) = 0 \,.
3445: \end{eqnarray}
3446: Since derivatives with respect to the $x$ modes go as
3447: \begin{equation}
3448: \partial \sim E^{-1} (a + a^{\dagger}),
3449: % \label{eq:}
3450: \end{equation}
3451: we have the additional overlap constraint on $U$
3452: \begin{equation}
3453: (1 + Y)E^{-1} (1-U) = 0 \,.
3454: \label{eq:oo2}
3455: \end{equation}
3456: Equations (\ref{eq:oo1}) and (\ref{eq:oo2}) determine $U$ completely,
3457: giving
3458: \begin{equation}
3459: U = (2-EYE^{-1} + E^{-1} YE)[EYE^{-1} + E^{-1} YE]^{-1}.
3460: \label{eq:implicit}
3461: \end{equation}
3462: Unfortunately, the matrix combination in brackets is difficult to
3463: explicitly invert. This does, however, give a closed form expression
3464: for the three-string vertex (\ref{eq:vertex-0}), where
3465: \begin{eqnarray}
3466: V'^{rr} & = & \frac{1}{3} (C + U + \bar{U})\\
3467: V'^{r, r \pm 1} & = & \frac{1}{6} ( 2C -U-\bar{U}) \pm \frac{i
3468: \sqrt{3}}{6} (U- \bar{U}) \,.\nonumber
3469: \end{eqnarray}
3470:
3471: While (\ref{eq:implicit}) is difficult to directly compute, given a
3472: formula for $U$ one can check that the formula is correct by checking
3473: the overlap conditions (\ref{eq:oo1}) and (\ref{eq:oo2}).
3474: Expressions for $V$ and $X$ and hence for $U$
3475: and $V'$ were computed~\cite{Gross-Jevicki-12} by essentially the method used
3476: in the previous section. Their results for $V$ and $X$ are given as
3477: follows\footnote{Note that in some references, signs and various
3478: factors in $\kappa$ and the Neumann coefficients may be slightly
3479: different. In some papers, the cubic term in the action is taken to
3480: have an overall factor of $g/6$ instead of $g/3$; this choice of
3481: normalization gives a 3-tachyon amplitude of $g$ instead of $2g$, and
3482: gives a different value for $\kappa$. Often, the sign in the
3483: exponential of (\ref{eq:v3}) is taken to be positive, which changes
3484: the signs of the coefficients $V^{rs}_{nm}, X^{rs}_{nm}$. When the
3485: matter Neumann coefficients are defined with respect to the oscillator
3486: modes $\alpha_n$ rather than $a_n$, the matter Neumann coefficients
3487: $V^{rs}_{nm}, V^{rs}_{n0}$ must be divided by $\sqrt{nm}$ and
3488: $\sqrt{n}$. This is the case for the coefficients $N^{rs}_{nm}$
3489: computed in (\ref{neumannmatter}), which are related to the $V$'s
3490: through $N^{rs}_{nm}= V^{rs}_{nm}/\sqrt{nm}$. Finally, when $\alpha'$
3491: is taken to be $1/2$, an extra factor of $1/\sqrt{2}$ appears for each
3492: $0$ subscript in the matter Neumann coefficients. }. Define $A_n,
3493: B_n$ for $n \geq 0$ through
3494: \begin{eqnarray}
3495: \left( \frac{1 + ix}{1-ix} \right)^{1/3} & = &
3496: \sum_{n\, {\rm even}} A_n x^n + i
3497: \sum_{m\, {\rm odd}} A_m x^m \label{eq:ab}\\
3498: \left( \frac{1 + ix}{1-ix} \right)^{2/3} & = &
3499: \sum_{n\, {\rm even}} B_n x^n + i
3500: \sum_{m\, {\rm odd}} B_m x^m \,. \nonumber
3501: \end{eqnarray}
3502: These coefficients can be used to define 6-string Neumann coefficients
3503: $N^{r, \pm s}_{nm}$ through
3504: \begin{eqnarray}
3505: N^{r, \pm r}_{nm} & = &
3506: \left\{\begin{array}{l}
3507: \frac{1}{3 (n \pm m)} (-1)^n (A_nB_m \pm B_nA_m), \;\;\;\;\;
3508: m + n\, {\rm even}, \;m \neq n\\
3509: 0, \;\;\;\;\; m + n\, {\rm odd}
3510: \end{array} \right.\label{eq:n6}\\
3511: N^{r, \pm (r + \sigma)}_{nm} & = &
3512: \left\{\begin{array}{l}
3513: \frac{1}{6 (n \pm \sigma m)} (-1)^{n + 1} (A_nB_m \pm \sigma B_nA_m),
3514: \;\;\;\;\;
3515: m + n\, {\rm even}, \;m \neq n\\
3516: \sigma
3517: \frac{\sqrt{3}}{6 (n \pm \sigma m)} (A_nB_m \mp \sigma B_nA_m), \;\;\;\;\;
3518: m + n\, {\rm odd}
3519: \end{array}\right].\nonumber
3520: \end{eqnarray}
3521: where in $N^{r, \pm (r + \sigma)}$, $\sigma = \pm 1$, and $r +\sigma$
3522: is taken modulo 3 to be between 1 and 3. The 3-string matter Neumann
3523: coefficients $V^{rs}_{nm}$ are then given by
3524: \begin{eqnarray}
3525: V^{rs}_{nm} & = & -\sqrt{mn} (N^{r, s}_{nm} + N^{r, -s}_{nm}),
3526: \;\;\;\;\; m \neq n,\, {\rm and}\, m, n \neq 0 \nonumber\\
3527: V^{rr}_{nn} & = & -\frac{1}{3} \left[
3528: 2 \sum_{k = 0}^{n} (-1)^{n-k} A_k^2-(-1)^n-A_n^2 \right], \;\;\;\;\;
3529: n \neq 0 \nonumber\\
3530: V^{r, r + \sigma}_{nn} & = &\frac{1}{2} \left[ (-1)^n-V^{rr}_{nn}
3531: \right], \;\;\;\;\; n \neq 0 \label{eq:n3}\\
3532: V^{rs}_{0n}& = & -\sqrt{2n} \left( N^{r, s}_{0n} + N^{r, -s}_{0n}
3533: \right), \;\;\;\;\; n \neq 0\nonumber\\
3534: V^{rr}_{00} & = & \ln (27/16) \nonumber
3535: \end{eqnarray}
3536: The ghost Neumann coefficients $X^{rs}_{m n}, m\geq 0, n >0$ are
3537: given by
3538: \begin{eqnarray}
3539: X^{rr}_{mn} & = & \left( -N^{r, r}_{nm} + N^{ r, -r}_{nm} \right),
3540: \;\;\;\;\; n \neq
3541: m\nonumber\\
3542: X^{r (r \pm 1)}_{mn} & = & m \left(\pm N^{r, r \mp 1}_{nm} \mp N^{ r, - (r
3543: \mp 1)}_{nm} \right), \;\;\;\;\; n \neq
3544: m \label{eq:x3}\\
3545: X^{rr}_{nn} & = & \frac{1}{3}
3546: \left[ -(-1)^n-A_n^2 + 2 \sum_{k = 0}^{n} (-1)^{n-k} A_k^2 -2
3547: (-1)^nA_nB_n \right] \nonumber\\
3548: X^{r (r \pm 1)}_{nn} & = &
3549: -\frac{1}{2}(-1)^n -\frac{1}{2} X^{rr}_{nn}
3550: \nonumber
3551: \end{eqnarray}
3552:
3553: These expressions for the matter and ghost
3554: Neumann coefficients were computed by
3555: Gross and Jevicki~\cite{Gross-Jevicki-12}, and include minor corrections
3556: published later~\cite{RSZ-2}. It was shown that the resulting
3557: matter matrices $U$ indeed satisfy the overlap conditions
3558: (\ref{eq:oo1}) and (\ref{eq:oo2}). This shows that the conformal
3559: field theory method and the oscillator method give the same results
3560: for the matter part of the three-string vertex. The same is true for
3561: the ghost part of the vertex, although we will not go into the details
3562: of this discussion here.
3563:
3564: Before leaving the three-string vertex, it is worth noting that
3565: the Neumann coefficients have a number of simple symmetries. There is
3566: a cyclic symmetry under $r \rightarrow r + 1, s \rightarrow s + 1$,
3567: which corresponds to the obvious geometric symmetry of rotating the
3568: vertex. The coefficients are also symmetric under the exchange $r
3569: \leftrightarrow s, n \leftrightarrow m$. Finally, there is a
3570: twist symmetry which, as discussed in section
3571: \ref{sec:algebraic-structure}, is associated with reflection of the
3572: strings
3573: \begin{eqnarray}
3574: V^{rs}_{nm} & = & (-1)^{n + m}V^{sr}_{nm}\\
3575: X^{rs}_{nm} & = & (-1)^{n + m}X^{sr}_{nm}\,. \nonumber
3576: \end{eqnarray}
3577: This symmetry follows from the fact that half-strings carrying odd
3578: modes pick up a minus sign under reflection. Since each string
3579: carrying an odd mode gets two changes of sign, from the two ends of
3580: the string, it is straightforward to see that this symmetry guarantees
3581: that the three-vertex is invariant under reflection, and therefore
3582: satisfies condition (\ref{l1e59}).
3583:
3584:
3585:
3586:
3587:
3588:
3589: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3590: % Calculating action
3591: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3592:
3593: \subsection{Calculating the SFT action}
3594: \label{sec:calculating}
3595:
3596: Given the action
3597: \begin{equation}
3598: S = -\frac{1}{2}\langle V_2 | \Psi, Q \Psi \rangle
3599: -\frac{g}{3} \langle V_3 | \Psi, \Psi, \Psi \rangle\,,
3600: \label{eq:action-Fock}
3601: \end{equation}
3602: and the explicit formulae
3603: (\ref{eq:v2}, \ref{eq:v3}) for the two- and three-string vertices, we
3604: can in principle calculate the string field action term by term for
3605: each of the fields in the string field expansion
3606: \begin{eqnarray}
3607: \Psi & = & \int d^{26}p \;
3608: \left[ \phi (p)\; | 0_1; p \rangle + A_\mu (p) \; \alpha^\mu_{-1} | 0_1; p
3609: \rangle + \chi ( p) b_{-1} c_0| 0_1; p \rangle \right. \nonumber\\
3610: & & \hspace*{0.9in} \left.
3611: + B_{\mu \nu} ( p)
3612: \alpha^\mu_{-1} \alpha^\nu_{-1} | 0_1; p \rangle + \cdots
3613: \right]\,. \label{eq:field-expansion-2}
3614: \end{eqnarray}
3615:
3616:
3617: Since the resulting action has an enormous gauge invariance given by
3618: (\ref{eq:SFT-gauge}), it is often helpful to fix the gauge before
3619: computing the action. A particularly useful gauge choice is the
3620: Feynman-Siegel gauge
3621: \begin{equation}
3622: b_0 | \Psi \rangle = 0\,.
3623: \label{eq:FS-gauge}
3624: \end{equation}
3625: This is a good gauge choice locally, fixing the linear gauge
3626: transformations $\delta | \Psi \rangle = Q | \Lambda \rangle$. This
3627: gauge choice is not, however, globally valid; we will return to this
3628: point in subsection \ref{sec:gauge}. In this gauge, all fields
3629: in the string field expansion which are associated with states that have
3630: an antighost zero-mode $c_0$ are taken to vanish. For example, the
3631: field $\chi (p)$ in (\ref{eq:field-expansion-2}) vanishes. In
3632: Feynman-Siegel gauge, the BRST operator takes the simple form
3633: \begin{equation}
3634: Q = c_0L_0 =c_0 (N + p^2 -1)
3635: \label{eq:FS-BRST}
3636: \end{equation}
3637: where $N$ is the total (matter + ghost) oscillator number.
3638:
3639: Using (\ref{eq:FS-BRST}), it is straightforward to write the quadratic
3640: terms in the string field action. They are
3641: \begin{equation}
3642: \frac{1}{2}\langle V_2 | \Psi, Q \Psi \rangle =
3643: \int d^{26} p \; \left\{
3644: \phi (-p) \left[ \frac{p^2 -1}{2} \right]\phi (p)
3645: + A_\mu (-p) \left[ \frac{p^2}{2} \right]A^\mu (p) + \cdots
3646: \right\}\,.
3647: % % \label{eq:}
3648: \end{equation}
3649:
3650: The cubic part of the action can also be computed term by term,
3651: although the terms are somewhat more complicated. The leading terms
3652: in the cubic action are
3653: given by
3654: \begin{eqnarray}
3655: \lefteqn{\frac{1}{3}
3656: \langle V_3 | \Psi, \Psi, \Psi \rangle =
3657: %\\& &
3658: \int d^{26}pd^{26}q \; \frac{\kappa g}{3} \;
3659: e^{(\ln 16/27) (p^2 + q^2 + p \cdot q)}
3660: }\label{eq:expanded-action} \\
3661: & &\hspace{1.0in} \times
3662: \left\{\phi (-p) \phi (-q) \phi (p + q) + \frac{16}{9}
3663: A^{\mu} (-p) A_\mu (-q) \phi (p + q)
3664: \right.
3665: \nonumber
3666: \\
3667: & & \hspace{1.1in}
3668: \left.
3669: %e^{(\ln 4/3 \sqrt{3}) (p^2 + q^2 + p \cdot q)}
3670: - \frac{8}{9}
3671: (p^\mu +2q^\mu) (2p^{\nu} + q^{\nu})A^{\mu} (-p) A_\nu (-q) \phi (p +
3672: q) + \cdots
3673: \right\}\,. \nonumber
3674: \end{eqnarray}
3675: In computing the $\phi^3$ term we have used
3676: \begin{equation}
3677: V^{rs}_{00} = \delta^{rs}
3678: \ln (\frac{27}{16} )\,.
3679: % % \label{eq:}
3680: \end{equation}
3681: The $A^2 \phi$ term uses
3682: \begin{equation}
3683: V^{rs}_{11} = -\frac{16}{27} , \; \;r \neq s,
3684: % % \label{eq:}
3685: \end{equation}
3686: while the $ (A \cdot p)^2 \phi$ term uses
3687: \begin{equation}
3688: V^{12}_{10} = -V^{13}_{10} = -\frac{2 \sqrt{2}}{3 \sqrt{3}}\,.
3689: % % \label{eq:}
3690: \end{equation}
3691: The most striking feature of this action is that for a generic set of
3692: three fields, there is a {\it nonlocal} cubic interaction term that
3693: contains an exponential of a quadratic form in the momenta. This
3694: means that the target space formulation of string theory has a
3695: dramatically different character from a standard quantum field theory.
3696: {}From the point of view of quantum field theory, string field theory
3697: seems to contain an infinite number of nonrenormalizable interactions.
3698: Just like the simpler case of noncommutative field theories, however,
3699: the magic of string theory seems to combine this infinite set of
3700: interactions into a sensible model.
3701: It has been shown that all on-shell amplitudes computed from the
3702: string field theory action we have discussed here precisely reproduce
3703: the amplitudes given by the usual conformal field theory approach,
3704: including the correct measure on moduli
3705: space~\cite{Giddings-Martinec,gmw,Zwiebach-proof}.
3706: Note, though, that the
3707: bosonic open theory becomes problematic at the quantum
3708: level because of the closed string tachyon, whose instability
3709: is not yet understood.
3710: For the purposes of these lectures,
3711: we will restrict our attention to the classical bosonic open string action.
3712: Open superstring field theory
3713: should be better behaved since the closed string sector has
3714: no tachyon. There has been significant progress in understanding
3715: tachyon condensation in superstring field theory~\cite{Aref'eva,Berkovits},
3716: even though superstring field theory is less developed than
3717: bosonic string field theory.
3718: %although a complete understanding of superstring field theory is still lacking
3719: %despite recent progress
3720: %~\cite{Aref'eva,Berkovits}.
3721: %at zero momentum, where the action is quite well-behaved.
3722:
3723:
3724: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3725: % Evidence: both
3726: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3727:
3728: \section{Evidence for the Sen conjectures}
3729: \label{sec:evidence}
3730:
3731: In this section we review the evidence from Witten's OSFT for Sen's
3732: conjectures. Subsection \ref{sec:tension} contains a derivation of
3733: the formula for the tension of a bosonic D-brane. In subsection
3734: \ref{sec:constraints-symmetries}, a general discussion is given of
3735: symmetries in the string field theory action and resulting constraints
3736: on the set of string fields which take nonzero values in the tachyon
3737: vacuum. Subsection \ref{sec:vacuum} contains a summary of existing
3738: results for the determination of the stable vacuum in Witten's OSFT
3739: (Sen's first conjecture),
3740: including some results which appeared after these lectures were
3741: originally given in 2001. In subsection \ref{sec:gauge} we discuss
3742: the Feynman-Siegel gauge choice and its limitations. Subsection
3743: \ref{sec:solitons} summarizes results on lower-dimensional D-branes as
3744: solitons in OSFT (Sen's second conjecture). Subsection
3745: \ref{subsec:thebackgroundsosft} discusses the general problem of
3746: finding all open string backgrounds within OSFT. Sen's third
3747: conjecture is discussed in the following section 8, which is completely
3748: devoted to a discussion of the physics in the stable vacuum (vacuum
3749: string field theory).
3750:
3751: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3752: % Mass of D-branes: BZ
3753: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3754:
3755: \subsection{Tension of bosonic D$p$-branes}
3756: \label{sec:tension}
3757:
3758: In this section we learn how to relate the open string
3759: coupling constant of string field theory to the mass of
3760: the D-brane described by the open string field theory.
3761: The material presented in this subsection is an unpublished
3762: result due to
3763: Ashoke Sen~\cite{senunpub}, who cited the result
3764: in the paper~\cite{Sen-universality} on the subject
3765: of universality. Subsequently, the closely
3766: related computation for the superstring was explained
3767: in detail~\cite{superstring}. For an alternative
3768: check of the result, see Appendix~A of the paper by Okawa~\cite{Okawa}.
3769:
3770:
3771: \smallskip
3772: In general, an open string field theory is formulated
3773: using a BCFT (boundary conformal field theory) which
3774: describes some D-brane (or a configuration of D-branes).
3775: In order to describe a D-brane with finite mass, we
3776: consider a compactification of $p$ spatial coordinates
3777: and wrap a D$p$-brane around
3778: along these dimensions. The string field theory
3779: associated with this D-brane is written as before:
3780: \begin{equation} \label{ems340909}
3781: S (\Phi) = \,-\, {1\over g^2}\,\,\bigg[\, {1\over 2} \langle \,\Phi
3782: \,,\, Q\,
3783: \Phi
3784: \rangle + {1\over 3}\langle \,\Phi \,,\, \Phi *
3785: \Phi \rangle \bigg]\,.
3786: \end{equation}
3787: The D-brane in question is perceived by the effective
3788: $(25-p)$-dimensional observer as a point particle.
3789: The BCFT includes a Neumann field $X^0$, a set
3790: of Dirichlet fields $X^i$, with $i= 1, \ldots , 25-p$ and some set
3791: of Neumann fields $X^a$, with $a= 25-p+1, \ldots, 25$ that describe
3792: the internal sector of the BCFT. The string field theory
3793: effectively describes an infinite collection of
3794: fields $\phi_i(t, x^a)$. These fields do not depend on $x^1, \ldots
3795: , x^{25-p}$
3796: because the corresponding string coordinates are Dirichlet.
3797: Since the coordinates $x^a$ are compact, the fields $\phi_i(t, x^a)$ can be
3798: expanded in Fourier modes. These are a collection of
3799: degrees of freedom that are just time
3800: dependent. The string field theory action
3801: then reduces to an integral over
3802: time of a time-dependent Lagrangian density.
3803:
3804:
3805: We will set up the string field theory in such a way that
3806: all dimensions (including time) are compactified on circles
3807: of unit circumference. In this case, the mass $M$ of the D$p$-brane
3808: coincides with the tension of the D$p$-brane. The claim is that
3809: \begin{equation}
3810: \label{massbraneorig}
3811: M = {1\over 2\pi^2 \, g^2}\,.
3812: \end{equation}
3813: In this formula and in the following, we set
3814: $\alpha'=1$. In these units the string tension is $T_0 = 1/(2\pi)$.
3815: When we consider the string field theory of a D25-brane,
3816: (\ref{massbraneorig}) gives
3817: \begin{equation}
3818: \label{tnebraneorig}
3819: T_{25} = {1\over 2\pi^2 \, g^2}\,.
3820: \end{equation}
3821:
3822:
3823: We begin our study by considering some special momentum
3824: states of the BCFT:
3825: \begin{equation}
3826: \label{themomstatesorg}
3827: |k_0\rangle \equiv e^{ (ik_0 X^0(0))} |0\rangle \,.
3828: \end{equation}
3829: Moreover, we will normalize these states by declaring
3830: \begin{equation}
3831: \langle k_0 | \, c_{-1} c_0 c_1 \, | k_0'\rangle = \delta_{k_0, k_0'}\,,
3832: \end{equation}
3833: consistent with the discussion below (\ref{ourashokenorm}).
3834: Since the time direction has been made compact via
3835: $t \sim t+1$, the time component $k_0$ of the momentum
3836: is quantized: $k_0 = 2\pi n$, with $n$ integer, and we can use
3837: a Kronecker delta in the above inner product.
3838:
3839:
3840: \medskip
3841: We will consider the computation of the brane mass in three steps.
3842:
3843: \medskip
3844: \noindent
3845: {\em Step 1}: We consider time-dependent displacements of the D-brane.
3846: We will write down a string field that describes such a displacement
3847: and evaluate the kinetic term of the string action. This will make it
3848: clear how we can hope to calculate the brane mass.
3849:
3850: Let $X^i$ be one of the Dirichlet directions for the D-brane and
3851: assume that $x^i=0$ is the original position of the brane. Consider now a
3852: displacement field $\phi^i(t)$ that is expected to be proportional to a
3853: coordinate displacement $x^i(t)$.
3854: We expand the field $\phi^i(t)$ as:
3855: \begin{equation}
3856: \label{thefieldfordiskpvkakgk}
3857: \phi^i(t) = \sum_{k_0} e^{ik_0 t} \, \phi^i(k_0) \,,
3858: \end{equation}
3859: and we use the Fourier components $\phi^i(k_0)$ to assemble
3860: the corresponding string field:
3861: \begin{equation}
3862: \label{corrstringfieldkysla}
3863: |\Phi\rangle = \sum_{k_0} \, \phi^i(k_0) \, c_1 \alpha_{-1}^i \,
3864: |k_0\rangle\,.
3865: \end{equation}
3866: As you can see, the string field is built using states of the massless scalar
3867: field that represents translations of the D-brane.
3868: For this string field, the kinetic term $S_2 (\Phi)$ of the string action is
3869: given by
3870: \begin{equation}
3871: \label{fskuyuluay}
3872: S_2(\Phi) = -{1\over g^2} \sum_{k_0, k_0'} \phi^i(k_0) \phi^i(k_0')
3873: \langle -k_0'| c_{-1} \alpha_1^i \, c_0 L_0 \, c_1 \alpha_{-1}^i |k_0\rangle\,.
3874: \end{equation}
3875: Since $L_0= p^2 + \ldots$~ where the terms indicated by dots vanish
3876: in the present case, $L_0 =
3877: -k_0^2$ in (\ref{fskuyuluay}) and
3878: \begin{equation}
3879: \label{fskuyuluay89uf}
3880: S_2(\Phi) = {1\over 2g^2} \sum_{k_0} \phi^i(-k_0) \, k_0^2 \phi^i(k_0)\,.
3881: \end{equation}
3882: Let us now rewrite this string action in terms of the field $\phi^i(t)$
3883: introduced in (\ref{thefieldfordiskpvkakgk}). A short computation
3884: gives
3885: \begin{equation}
3886: \int_0^1 dt \, \partial_t \phi^i \, \partial_t \phi^i = \sum_{k_0}
3887: \phi^i(-k_0) \,
3888: k_0^2 \phi^i(k_0)\,.
3889: \end{equation}
3890: Comparing with (\ref{fskuyuluay89uf}) we find that
3891: \begin{equation}
3892: S_2(\Phi) = {1\over 2 g^2} \int_0^1 dt \,\partial_t \phi^i \,
3893: \partial_t \phi^i\,.
3894: \end{equation}
3895: As we mentioned earlier, the field $\phi^i(t)$ is expected to be
3896: proportional to the position $x^i(t)$ of the brane (at least for small,
3897: slowly varying displacements), so we can rewrite the above action as
3898: \begin{equation}
3899: S_2(\Phi) = {1\over g^2} \Bigl( {\delta \phi^i\over \delta x^i} \Bigr)^2
3900: \int_0^1 dt\,\, {1\over 2} \partial_t
3901: x^i \, \partial_t x^i
3902: \end{equation}
3903: where the derivatives are evaluated at zero displacement. Since
3904: $\partial_t x^i$ is the velocity of the D-brane, the above action
3905: represents the contribution from the (non-relativistic) kinetic energy
3906: of a D-brane that has a mass $M$ given
3907: by \footnote{Note that at this point, it is possible to take a shortcut
3908: to get the D-brane mass directly using the fact that SFT at tree level
3909: reproduces Yang-Mills theory~\cite{Coletti-Sigalov-Taylor}, with
3910: $g_{{\rm YM}} = g/\sqrt{2}$ ~\cite{Polchinski,Coletti-Sigalov-Taylor},
3911: where the Yang-Mills
3912: field appears in the string field expansion as $A_\mu (k)
3913: \alpha^{\mu}_{-1} | 0_1; k \rangle$, and where an additional factor of
3914: $2 \pi$
3915: arises from the T-duality relation from section \ref{sec:T-duality},
3916: $ X \rightarrow 2 \pi
3917: \alpha' A$. Thus,
3918: replacing $A^i \rightarrow X^i/2 \pi$ in the Yang-Mills action we have
3919: $1/2g_{{\rm YM}}^2
3920: F_{0i} F^{0i}\rightarrow 1/2 g^2
3921: (\partial_0 x^i/\sqrt{2} \pi)^2$, so $M = 1/2 \pi^2 g^2$. Although
3922: perhaps more transparent, this is essentially the same calculation as
3923: the original argument of Sen presented here, which we include in full
3924: as it sheds light on the structure of the theory.}
3925: \begin{equation}
3926: \label{theequationforpssmass}
3927: M = {1\over g^2} \Bigl( {\delta \phi^i\over \delta x^i} \Bigr)^2\,.
3928: \end{equation}
3929:
3930: \medskip
3931: \noindent
3932: {\em Step 2.~} To find out how $\delta \phi^i$ is related to a true
3933: displacement $\delta x^i$, we add a reference D-brane a
3934: distance $b$ away from our original brane, in the direction~$x^i$.
3935: We will then consider a string stretched between the branes.
3936: We will use the string field action to
3937: compute the change in the mass of such string when our D-brane is displaced
3938: by some $\delta \phi^i$. Since the string tension is known, we will
3939: be able to calculate the value of the physical displacement $\delta x^i$.
3940:
3941: Given a string of length $L$, its mass includes a contribution
3942: $T_0 L = L/(2\pi)$,
3943: and the corresponding contribution to the mass-squared is
3944: $L^2/(4\pi^2)$. If the original stretched string
3945: has length $b$ and its length is then changed to $b +
3946: \delta x^i$, the change $\delta m^2$ of the mass-squared is
3947: \begin{equation}
3948: \label{thechangeinwetnp}
3949: \delta m^2 = {1\over 4\pi^2} \Bigl( (b+ \delta x^i)^2 - b^2 \Bigr)
3950: \simeq {1\over 2\pi^2} \, b \, \delta x^i \,.
3951: \end{equation}
3952: Let us now consider a time independent displacement, that is, a configuration
3953: with $k_0=0$ (see (\ref{thefieldfordiskpvkakgk})). We thus set $\delta \phi^i
3954: \equiv \delta
3955: \phi^i (k_0 =0) $ and $\delta \phi^i(k_0\not= 0) =0$. The
3956: string field associated to this displacement is obtained using
3957: (\ref{corrstringfieldkysla}):
3958: \begin{equation}
3959: \label{thefirsttrysvnkeoii}
3960: |\delta \Phi\rangle = \delta \phi^i \, c_1 \alpha_{-1}^i \, |0\rangle \,.
3961: \end{equation}
3962: We want to learn the effect of this string field perturbation on
3963: the masses of stretched strings. To do so, we introduce a complex
3964: field $\eta$. The fields
3965: $\eta$ and $\eta^*$ represent the string that stretches from our
3966: brane (brane one) to the reference brane (brane two) and the string
3967: that stretches from the reference brane to our brane,
3968: respectively. The string field that describes these states is written as:
3969: \begin{equation}
3970: \label{stretchedstringvifield}
3971: |\psi\rangle = \eta \, c_1 |k_0, b\rangle \otimes
3972: \begin{pmatrix}
3973: 0&1\\ 0&0
3974: \end{pmatrix}
3975: + \eta^* \, c_1 |-k_0, -b\rangle \otimes
3976: \begin{pmatrix}
3977: 0&0\\ 1&0
3978: \end{pmatrix} \,.
3979: \end{equation}
3980: The matrices included here are Chan-Paton matrices $a_{\alpha\beta}$,
3981: with $\alpha, \beta = 1, 2$. A value of one for a given $a_{\alpha\beta}$,
3982: with zero for all other entries, is used to represent a string that stretches
3983: from brane $\alpha$ to brane $\beta$. When we have parallel
3984: D-branes, the string field is matrix-valued. The string action
3985: includes a trace operation Tr that applies to the matrices, and the
3986: star product
3987: includes matrix multiplication. The state $| k_0 , b\rangle$ represents
3988: the ground state of a string with momentum $k_0$ that stretches
3989: a distance $b$ in the $x^i$ direction. It is necessary for our analysis
3990: to determine the CFT vertex operator that corresponds
3991: to this stretched string. We claim that the operator is
3992: \begin{equation}
3993: \label{theopeforstretchstr}
3994: |k_0, b\rangle \quad \longleftrightarrow \quad e^{i k_0 X^0} \, e^{i
3995: {b\over 2\pi}
3996: (X_L^i - X_R^i) }\,.
3997: \end{equation}
3998: The $k_0$ dependence of the operator is already familiar from
3999: (\ref{themomstatesorg}). The field multiplying $b$ is formed from
4000: the left-moving and right-moving
4001: parts of the open string coordinate $X^i$, evaluated at the string
4002: endpoint. This coordinate $X^i$ satisfies Dirichlet boundary conditions, so
4003: at the boundary $X_L^i = - X_R^i$, and we can replace $X_L^i - X_R^i$ by
4004: $2X_L^i$. We also have the operator products and stress tensor
4005: \begin{equation}
4006: \partial X_L^i (x) \, \partial X_L^i (y) \sim -{1\over 2} {1\over (x-y)^2} \,,
4007: \qquad T_{X^i} = - \partial X_L^i \partial X_L^i \,.
4008: \end{equation}
4009: These relations allow us to compute the conformal dimension of
4010: an exponential. One readily finds that $\exp (i\alpha X_L^i)$ has
4011: conformal dimension $\alpha^2/4$. It follows that
4012: \begin{equation}
4013: \hbox{dimension}~\Bigl( e^{i {b\over \pi} X_L^i}\Bigr) = \Bigl( {b
4014: \over 2\pi}
4015: \Bigr)^2 \,.
4016: \end{equation}
4017: Since conformal dimension is the value of $L_0$, which, in turn, determines
4018: the mass-squared, this
4019: result confirms that the operator in (\ref{theopeforstretchstr}) has
4020: correctly
4021: reproduced
4022: the mass-squared of the stretched string.
4023: For future use, the vertex operator can be written as
4024: \begin{equation}
4025: \label{theopefohfltchstr}
4026: e^{i k_0 X^0} \, e^{i {b\over \pi} X_L^i }\,.
4027: \end{equation}
4028: The evaluation of the kinetic term for the field in
4029: (\ref{stretchedstringvifield})
4030: is relatively straightforward. The only terms that survive are the
4031: off-diagonal ones, coupling $\eta$ and $\eta^*$. There are two such
4032: terms, and their contributions are identical. The product of the two
4033: matrices give a matrix of trace one, and the overlap $\langle -k_0, -b|k_0,
4034: b\rangle$ is also equal to one. We then find
4035: \begin{equation}
4036: \label{readmassstrecvnjejj}
4037: g^2 \, S_2 (\eta, \eta^*) =-{1\over 2} \cdot 2 \, \eta^* \Bigl(
4038: -k_0^2 + {b^2\over (2\pi)^2} \Bigr) \eta= \eta^* \Bigl( k_0^2 -
4039: {b^2\over (2\pi)^2}
4040: \Bigr) \eta \,.
4041: \end{equation}
4042: In the setup with two branes, the fluctuation (\ref{thefirsttrysvnkeoii}) that
4043: represents the displacement of our brane is fully represented by
4044: \begin{equation}
4045: \label{thefirsttrysvnkmmmdeoii}\
4046: |\delta \Phi\rangle = \delta \phi^i \, c_1 \alpha_{-1}^i \, |0\rangle
4047: \otimes
4048: \begin{pmatrix}1&0\\ 0&0\end{pmatrix} \,.
4049: \end{equation}
4050: With the chosen normalization for $X^i$, the vertex operator
4051: associated with $\alpha^i_{-1} |0\rangle$ is $i \sqrt{2} \partial X_L$.
4052:
4053: \medskip
4054: \noindent
4055: {\em Step 3.~} We must now include the effects of the interactions
4056: to see how the fluctuation (\ref{thefirsttrysvnkmmmdeoii}) affects
4057: the mass of the stretched string. Since the mass can be read from
4058: equation (\ref{readmassstrecvnjejj}), we will find a term proportional
4059: to $\eta^* \eta$ that arises from the interaction and modifies the
4060: value of the mass.
4061:
4062: The interaction term takes the form
4063: \begin{equation}
4064: g^2 S_3 (\Phi) = -{1\over 3} \langle \Phi , \Phi, \Phi \rangle \,,
4065: \end{equation}
4066: and the string field is taken to be $|\Phi\rangle = |\psi\rangle
4067: +|\delta
4068: \Phi\rangle$, in order to see the effect of the fluctuation on the
4069: stretched string.
4070: We are looking for the terms of the form $\eta^* \eta \,\delta
4071: \phi^i$, so we have
4072: three different operators to insert at three different punctures.
4073: There are a total
4074: of six possible arrangements, that can be divided into two groups of
4075: three arrangements each. In each of these groups the cyclic ordering
4076: of the operators is the same. The Chan-Paton matrices imply that
4077: one cyclic ordering contributes while the other does not.
4078: Indeed,
4079: \begin{equation}
4080: \begin{pmatrix}0&1\\ 0&0\end{pmatrix}
4081: \begin{pmatrix}0&0\\ 1& 0\end{pmatrix}
4082: \begin{pmatrix}1&0\\ 0&0\end{pmatrix}
4083: = \begin{pmatrix}1&0\\ 0&0\end{pmatrix}
4084: \end{equation}
4085: is a matrix of unit trace, while
4086: \begin{equation}
4087: \begin{pmatrix}0&0\\ 1&0\end{pmatrix}
4088: \begin{pmatrix}0&1\\ 0& 0\end{pmatrix}
4089: \begin{pmatrix}1&0\\ 0&0\end{pmatrix}
4090: = \begin{pmatrix}0&0\\ 0&0\end{pmatrix}
4091: \end{equation}
4092: is a matrix of vanishing trace. We conclude that the Chan-Paton
4093: matrices contribute a factor of $+3$. The operators to be inserted
4094: can be chosen to be physical (dimension zero) so we need not worry
4095: about local coordinates at the punctures. Using punctures at
4096: $\infty$, $-1,$ and $0$ we then find:
4097: \begin{equation}
4098: g^2 S_3 (\Phi) = - \eta^* \eta \delta \phi^i
4099: \Bigl\langle e^{-i k_0 X^0-i {b\over \pi} X_L^i } c(\infty)
4100: \, \sqrt{2}\, i \, \partial X_L c(-1) \,
4101: e^{i k_0 X^0+i {b\over \pi} X_L^i } c(0) \Bigr\rangle \,.
4102: \end{equation}
4103: Since the vertex operators are on-shell, and the ghost insertions
4104: are placed at standard positions, the whole correlator gives a factor of
4105: one, except for the contraction between the $\partial X_L (-1)$ and
4106: the finitely located $\exp (i{b\over \pi} X_L^i (0))$:
4107: \begin{equation}
4108: g^2 S_3 = - \eta^* \eta \delta \phi^i\, \sqrt{2} \,i \, {ib\over \pi}\,\Bigl(
4109: -{1\over 2}\Bigr) {1\over (-1-0)} = {b\over \sqrt{2} \, \pi}
4110: \eta^* \eta \delta
4111: \phi^i \,.
4112: \end{equation}
4113: Combining this result with (\ref{readmassstrecvnjejj}) we find
4114: \begin{equation}
4115: \label{readmastkjrecvnjejj}
4116: g^2 \, (S_2 + S_3) = \eta^* \Bigl( k_0^2 - {b^2\over (2\pi)^2}+
4117: {b\over \sqrt{2}}
4118: {\delta \phi^i\over \pi}
4119: \Bigr) \eta \,.
4120: \end{equation}
4121: The last term in parenthesis corresponds to a change in $m^2$. So, comparing
4122: with (\ref{thechangeinwetnp}) we obtain
4123: \begin{equation}
4124: {1\over 2\pi^2} b\, \delta x^i = -{b\over \sqrt{2}}
4125: {\delta \phi^i\over \pi} \quad \to \quad {\delta \phi^i\over \delta x^i}
4126: = -{1\over \sqrt{2} \, \pi} \,.
4127: \end{equation}
4128: This is the needed relation between the field $\delta \phi^i$ that represents
4129: a displacement of the brane and the resulting displacement $\delta x^i$.
4130: The mass of the brane now follows directly from (\ref{theequationforpssmass}):
4131: \begin{equation}
4132: M = {1\over g^2} \Bigl( {1\over \sqrt{2}\, \pi} \Bigr)^2 = {1\over 2\pi^2
4133: g^2}\,.
4134: \end{equation}
4135: This is the result we wanted to establish.
4136:
4137:
4138: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4139: % Constraints and symmetries: BZ
4140: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4141:
4142: \subsection{Constraints and symmetries}
4143: \label{sec:constraints-symmetries}
4144:
4145: It may appear that {\em a priori} all scalar fields in the
4146: spectrum of open strings could acquire a vacuum expectation
4147: value in the tachyonic vacuum. Nevertheless, there are a
4148: set of considerations that imply that only a subset of these
4149: scalar fields acquire expectation values. In this section we explore
4150: these ideas. They are subdivided into the following:
4151:
4152: \begin{enumerate}
4153:
4154: \item[(1)] Universality conditions.
4155:
4156: \item[(2)] Twist conditions.
4157:
4158: \item[(3)] Gauge fixing conditions.
4159:
4160: \item[(4)] $SU(1,1)$ conditions.
4161:
4162: \end{enumerate}
4163:
4164: Among these conditions, the third one, which concerns gauge fixing, is
4165: on a somewhat different footing. The other three conditions apply because
4166: of a simple general argument which we discuss first.
4167:
4168: Consider a subdivision of all the scalar fields into two disjoint set of
4169: fields. The first set contains the fields $t_0, t_1, t_2, \ldots$
4170: and the second
4171: set contains the fields $u_0, u_1, u_2, \ldots$. Let us denote by $t_i$
4172: the elements of the first set and by $u_a$ the elements of the second set.
4173: Suppose the string field action $S(t_i, u_a)$ is such that there are no
4174: terms that are linear in $u_a$. We then claim that it is consistent
4175: to search for a solution of the equations of motion that assumes $u_a=0$
4176: for all $a$. The reason is easy to explain. If all terms with $u$
4177: fields contain at
4178: least two of them, the equations of motion for the $u$ fields are composed
4179: of terms all of which contain at least one $u$ field.
4180: As a result, $u_a =0$ satisfies these equations of motion.
4181: In our analysis we will try to construct a set $\{ t_i\}$ with the
4182: smallest possible number of fields,
4183: so that none of the remaining fields couples linearly in the action.
4184: The tachyon field, of course, is
4185: one of the elements of the set $\{t_i\}$.
4186:
4187: \medskip
4188: Let's begin by explaining how (1) works. For this we split the state
4189: space of the BCFT into three groups. In each of these groups, the
4190: ghost part of the states is the same: it includes all states of ghost
4191: number one. The nontrivial part of the argument uses the matter
4192: part of the conformal field theory. We write
4193: \begin{equation}
4194: \mathcal{H} = \mathcal{H}_1 \oplus \mathcal{H}_2\oplus \mathcal{H}_3\,,
4195: \end{equation}
4196: where $\mathcal{H}_1, \mathcal{H}_2,$ and $\mathcal{H}_3$ will
4197: be disjoint vector subspaces of $\mathcal{H}$ (their intersections give
4198: the zero vector). We also write
4199: \begin{equation}
4200: \mathcal{H}_i = \mathcal{M}_i \otimes |\mathcal{G}\rangle \,,\quad i=1,2,3\,.
4201: \end{equation}
4202: where $|\mathcal{G}\rangle$ denotes the general state of ghost number one
4203: in the ghost conformal field theory. The spaces
4204: $\mathcal{M}_1, \mathcal{M}_2,$ and $\mathcal{M}_3$ are disjoint subspaces
4205: of the matter CFT whose union gives the total matter CFT
4206: state space. The $\mathcal{M}$ subspaces are defined as:
4207: \begin{eqnarray}
4208: \label{sthesplitvematt}
4209: \mathcal{M}_1 : && \hbox{primary\,}~~
4210: |0\rangle ~\hbox{and descendents} ,\nonumber\\
4211: \mathcal{M}_2 : && \hbox{primaries}~ |k_0 \not=0\rangle ~\hbox{and
4212: descendents}, \\
4213: \mathcal{M}_3 : && \hbox{primaries}~ |k_0=0\rangle
4214: ~\hbox{different from}~|0\rangle ~\hbox{and descendents} \nonumber \,.
4215: \end{eqnarray}
4216: In the above, primary means Virasoro primary in the matter sector,
4217: and descendent means Virasoro descendent. The union of the spaces
4218: give the full CFT because for unitary matter CFT's (which we are
4219: assuming our CFT is) the state space can be broken into primaries and
4220: their descendents. Any matter primary belongs to one of the three
4221: spaces above. It should also be clear that the primaries in $\mathcal{M}_3$
4222: have positive conformal dimension.
4223:
4224: We now claim that the fields in $\mathcal{H}_2$ and $\mathcal{H}_3$
4225: need not acquire expectation values (they are
4226: $u$ fields); the tachyon condensate is all in $\mathcal{H}_1$. We
4227: are therefore
4228: defining
4229: the $t$ fields to be precisely the fields in $\mathcal{H}_1$.
4230: To prove that this is valid, we first note
4231: that a field in $\mathcal{H}_2$ cannot appear linearly in a term
4232: where all other fields are $t$ fields ({\it i.e.}, fields in
4233: ${\mathcal H}_1$). The reason is simply momentum
4234: conservation.
4235:
4236: A little more work is necessary to show that the
4237: fields in $\mathcal{H}_3$ cannot couple linearly to the fields in
4238: $\mathcal{H}_1$. Let us begin with the kinetic term. Since the
4239: BRST operator is composed of terms that include ghost oscillators
4240: and matter Virasoro operators, it maps each $\mathcal{H}_i$ space
4241: into itself. The primaries in $\mathcal{H}_1$ and $\mathcal{H}_3$
4242: are BPZ orthogonal, so any two states in the descendent towers are
4243: also orthogonal (this is proven by using the BPZ conjugation properties
4244: of Virasoro operators to move them from one state to the other until
4245: some state is annihilated or the whole expression reduces to the
4246: BPZ inner product of the primaries).
4247: For the interaction term a similar argument holds. First note
4248: that the three string vertex does not couple two matter primaries from
4249: $\mathcal{H}_1$ to a matter primary from $\mathcal{H}_3$. This is
4250: because in the CFT matter correlator the primaries from $\mathcal{H}_1$ appear
4251: as identity operators, so the whole correlator is proportional to the one-point
4252: function of the primary in $\mathcal{H}_3$, which vanishes because the state
4253: has non-zero dimension. The Virasoro conservation laws on the vertex then
4254: imply that the coupling of any two states in $\mathcal{H}_1$ to a state in
4255: $\mathcal{H}_3$ must vanish. This completes our proof.
4256:
4257: The space $\mathcal{H}_1$ is universal. It does not depend on the
4258: details of the matter conformal field theory, except for the existence
4259: of a zero-momentum SL(2,R) ground state. The space can be written
4260: as
4261: \begin{equation}
4262: \mathcal{H}_1 \equiv \hbox{Span} \Bigl\{ L_{-j_1}^m \ldots L_{-j_p}^m\,
4263: b_{-k_1} \ldots b_{-k_q} \, c_{-l_1} \ldots c_{-l_r} \, |0\rangle\Bigr\}
4264: \end{equation}
4265: where
4266: \begin{equation}
4267: j_1 \geq j_2 \geq \ldots \geq j_p \,, ~~j_i \geq 2 \,, ~~ k_i \geq 2
4268: \,, ~~ l_i \geq
4269: -1\,,
4270: \quad
4271: \hbox{and}\quad l-q=1 \,.
4272: \end{equation}
4273: The first inequality ensures that the descendents are built unambiguously,
4274: the second inequality is needed because $L_{-1} |0\rangle =0$.
4275: The third and fourth inequalities are familiar, and the last equality
4276: ensures that the ghost number of the state is one.
4277: \medskip
4278:
4279: Let us now explain how twist properties allow us to restrict
4280: $\mathcal{H}_1$ further. The claim is that we can restrict ourselves
4281: to the twist even subspace of $\mathcal{H}_1$.
4282: Heuristically, this follows from the fact that the two- and
4283: three-string vertices are invariant under reflection, so all terms
4284: linear in twist fields would pick up a change of sign and therefore vanish.
4285: The twist-even space, of course,
4286: contains the zero momentum tachyon $c_{1} |0\rangle$ (recall that
4287: $|0\rangle$ is twist odd, and $\Omega c_{-n} \Omega^{-1} = (-)^n c_{-n}$).
4288: The first two properties in (\ref{l1e4}) ensure that the kinetic
4289: term in the string action does not couple a twist odd field to
4290: a twist even field. We also studied the twist properties of the three string
4291: vertex. In fact, in an exercise, we considered a twist even field
4292: $A_+$ and a
4293: twist odd field
4294: $A_-$ ( $\Omega A_\pm =
4295: \pm A$ ) both of which were Grassmann odd (like the string field is).
4296: You then showed that $\langle A_+, A_+, A_-\rangle = 0$ (see
4297: (\ref{fortwistproperty})). Consider now a general string field
4298: $\Phi \in \mathcal{H}$ and split it into twist even
4299: and twist odd parts $\Phi = \Phi_+ + \Phi_-$. When the interaction
4300: vertex is evaluated, the terms linear in $\Phi_-$ are of the form
4301: $\langle \Phi_+ , \Phi_+, \Phi_-\rangle$ (any other similar looking
4302: term is related to this by cyclicity). So terms linear on twist
4303: odd fields vanish. This proves that we can indeed constrain
4304: $\mathcal{H}_1$ further.
4305:
4306: The twist eigenvalue of a state is
4307: given as $\Omega = (-1)^N$, where $N$ is the number eigenvalue of the state,
4308: defined with $N=0$ for the zero momentum tachyon. In terms of level,
4309: states at odd levels are twist odd, and states at even levels are twist even.
4310: So, the twist condition allows us to restrict ourselves to the states of
4311: $\mathcal{H}_1$ at even levels.
4312:
4313: \medskip
4314: We now turn to the gauge fixing condition. This gauge fixing
4315: condition, the Feynman-Siegel gauge condition
4316: $b_0 |\Phi\rangle =0$, restricts further the space $\mathcal{H}_1$.
4317: We will discuss the global
4318: validity of the Siegel gauge later, but here we discuss its
4319: clear validity at the linearized level and within the subspace
4320: $\mathcal{H}_1$ already restricted to states at even level.
4321: First, we show that the gauge condition can be reached
4322: starting from fields that do not satisfy it.
4323: Let $|\Phi\rangle $ be a field such that
4324: $b_0 |\Phi\rangle
4325: \not=0$. Since $|\Phi\rangle$ cannot be of level one, $L_0 |\Phi
4326: \rangle\not= 0$.
4327: Then consider the following gauge equivalent state
4328: \begin{equation}
4329: |\widetilde \Phi\rangle = |\Phi\rangle - Q {b_0\over L_0} |\Phi\rangle\,.
4330: \end{equation}
4331: Using $\{b_0 , Q\} = L_0$ one readily checks that $b_0|\widetilde \Phi \rangle
4332: =0$, so the gauge can be reached. Moreover, we now show that no gauge
4333: transformation remains in this gauge. If there were, there would exist
4334: a non-zero string field in the gauge slice that happens to be pure gauge.
4335: Such field $|\Phi \rangle$ would then satisfy $b_0 |\Phi\rangle =0$,
4336: $L_0|\Phi\rangle \not=0$, and $|\Phi\rangle = Q |\epsilon\rangle$.
4337: Since both $b_0$ and $Q$ annihilate
4338: the state:
4339: \begin{equation}
4340: 0= b_0 Q |\Phi\rangle + Q b_0 |\Phi\rangle = \{ b_0, Q\} |\Phi\rangle
4341: = L_0 |\Phi\rangle\,,
4342: \end{equation}
4343: in contradiction with the fact that the state does have non-zero
4344: dimension. The Siegel gauge is clearly a good gauge at the linearized
4345: level and within the twist truncated $\mathcal{H}_1$.
4346:
4347:
4348:
4349: \medskip
4350: Let's now consider briefly the additional truncation that is
4351: allowed by $SU(1,1)$ symmetry (item (4) of our list). Once we work
4352: in the Siegel gauge, this further truncation is allowed. This truncation
4353: is only possible because of the particular form of the string vertex.
4354: It would not be allowed for arbitrarily defined star products.
4355: Let us recall
4356: how this symmetry arises in the
4357: cubic open string field theory~\cite{Zwiebach:2000vc}. In the Siegel gauge, the
4358: string field action reads
4359: \begin{equation}
4360: \label{gfaction}
4361: S \sim {1\over 2} \langle \phi|L_0 |\phi\rangle + \,{1\over 3}
4362: \langle \phi | \langle \phi | \langle \phi \,| \,v_3 \rangle \,.
4363: \end{equation}
4364: The vertex coupling the three string fields is of the form
4365: \begin{equation}
4366: \label{thevertex}
4367: |v_3\rangle \sim \exp\, (E_{matt}) \,\,\exp \Bigl( - \sum_{r,s = 1}^3
4368: \sum_{n, m
4369: =1}^\infty c_{-n}^r\,X^{rs}_{nm}\, b_{-m}^s \Bigr) \,
4370: |0_1\rangle_{123}\,,
4371: \end{equation}
4372: where we have focused on the ghost sector.
4373: The Neumann coefficients
4374: are known to satisfy~\cite{Zwiebach:2000vc}
4375: \begin{equation}
4376: \label{exchange}
4377: X^{rs}_{nm} = \, {n\over m} \, X^{sr}_{mn} \,, \quad n,m \geq 1\,.
4378: \end{equation}
4379: This relation is not true for general three string
4380: vertices, but holds for the open string field theory vertex.
4381: Given equation (\ref{exchange}),
4382: the argument of the exponential in $|v_3\rangle_{123}$
4383: can be written as a sum of terms of the form
4384: ($r,s,n,m$, not summed)
4385: \begin{equation}
4386: \label{verpart}
4387: X^{rs}_{nm} \, c_{-n}^r b_{-m}^s + X^{sr}_{mn} \, c_{-m}^s b_{-n}^r
4388: = {1\over m} X^{sr}_{mn} \Bigl( \, n c_{-n}^r b_{-m}^s + m c_{-m}^s
4389: b_{-n}^r
4390: \Bigr)\,.
4391: \end{equation}
4392: The term in
4393: parenthesis is invariant under the continuous transformations
4394: \begin{eqnarray}
4395: \label{continuous}
4396: b_{-n} (\theta) &=& b_{-n} \cos \theta - n c_{-n} \sin \theta \,,\cr
4397: c_{-n} (\theta) &=& c_{-n} \cos \theta + {1\over n} b_{-n} \sin \theta
4398: \,.
4399: \end{eqnarray}
4400: These transformations, valid for all $n\not= 0$, imply $\{ c_n(\theta),
4401: b_m(\theta) \} = \delta_{n+m}$. One readily finds that they are generated
4402: by an operator $\mathcal{S}_1$:
4403: \begin{equation}
4404: \label{ghtrans}
4405: b_{-n} (\theta) = e^{\theta \mathcal{S}_1} \, b_{-n} e^{-\theta
4406: \mathcal{S}_1}\,,
4407: \quad c_{-n} (\theta) = e^{\theta \mathcal{S}_1} \, c_{-n} e^{-\theta
4408: \mathcal{S}_1}\,,
4409: \end{equation}
4410: where $\mathcal{S}_1$ is given by
4411: \begin{equation}
4412: \label{fgen}
4413: \mathcal{S}_1 = \sum_{n=1}^\infty \Bigl( \, {1\over n} \, b_{-n} b_n
4414: - n c_{-n}
4415: c_n\Bigr)\,.
4416: \end{equation}
4417: Since the vacuum $|0_1\rangle$ is annihilated by $\mathcal{S}_1$,
4418: the vertex $|v_3\rangle$ is invariant under
4419: this $U(1)$ symmetry: $\exp \Bigl( \theta
4420: (\mathcal{S}_1^{(1)}+\mathcal{S}_1^{(2)}+ \mathcal{S}_1^{(3)}) \, \Bigr) |
4421: v_3\rangle_{123} = | v_3\rangle_{123}$. Equivalently,
4422: \begin{equation}
4423: \label{finv}
4424: \Bigl(\mathcal{S}_1^{(1)}+\mathcal{S}_1^{(2)}+ \mathcal{S}_1^{(3)} \Bigr)
4425: | v_3\rangle_{123} = 0\,.
4426: \end{equation}
4427: Since the vertex $|v_3\rangle_{123}$ is built from ghost
4428: bilinears
4429: of zero ghost number, we deduce that the ghost number operator $\mathcal{G}$
4430: \begin{equation}
4431: \label{ghgen}
4432: \mathcal{G} = \sum_{n=1}^\infty \Bigl( \, c_{-n}b_n - b_{-n} c_n \, \Bigr).
4433: \end{equation}
4434: is also conserved:
4435: \begin{equation}
4436: \label{ginv}
4437: \Bigl(\mathcal{G}^{(1)}+\mathcal{G}^{(2)}+ \mathcal{G}^{(3)} \Bigr)
4438: | v_3\rangle_{123} = 0 \,.
4439: \end{equation}
4440: We can then form the commutator
4441: \begin{equation}
4442: \label{sgen}
4443: [ \mathcal{S}_1\, , \,\mathcal{G} \, ] = 2 \mathcal{S}_2\,, \quad \hbox{with}
4444: \quad
4445: \mathcal{S}_2 = \sum_{n=1}^\infty \Bigl( \, {1\over n} \, b_{-n} b_n
4446: + n c_{-n}
4447: c_n\Bigr)\,.
4448: \end{equation}
4449: The remaining commutators are readily computed:
4450: \begin{equation}
4451: \label{rcomm}
4452: [ \mathcal{S}_2\, , \,\mathcal{G} \, ] = 2 \mathcal{S}_1\,,\quad
4453: [ \mathcal{S}_1\, , \,\mathcal{S}_2 \, ] = -2 \mathcal{G}\,.
4454: \end{equation}
4455: These relations show that $\{ \mathcal{S}_1, \mathcal{S}_2, \mathcal{G} \}$
4456: generate the algebra of $SU(1,1)$. These generators
4457: are the same as those in the $SU(1,1)$ algebra in Siegel and
4458: Zwiebach~\cite{Siegel:1985tw}.\footnote{Defining $X= (\mathcal{S}_2 -
4459: \mathcal{S}_1)/2$,
4460: $Y=(\mathcal{S}_2 + \mathcal{S}_1)/2$, and $H= \mathcal{G}$ we recover the
4461: conventional definition of the isomorphic (real) Lie algebra $sl(2,R)$, with
4462: brackets $[X,Y]= H,~ [H, X] = 2X, ~ [H,Y] = -2Y$. Note that
4463: $T_+ = - 2X$, where $T_+$ is the operator that multiplies
4464: $b_0$ in the BRST operator.}
4465: Since both $\mathcal{S}_1$ and $\mathcal{G}$ are symmetries of the three string
4466: vertex, we also have
4467: \begin{equation}
4468: \label{sinv}
4469: \Bigl(\mathcal{S}_2^{(1)}+\mathcal{S}_2^{(2)}+ \mathcal{S}_2^{(3)} \Bigr)
4470: | v_3\rangle_{123} = 0 \,.
4471: \end{equation}
4472: In summary, the three string vertex is fully $SU(1,1)$ invariant.
4473:
4474:
4475: The set of Fock space states built with the action of ghost
4476: and antighost oscillators on the vacuum $|0_1\rangle$
4477: can be decomposed into finite dimensional
4478: irreducible representations of $SU(1,1)$. Note that $(nc_{-n}, b_{-n})$
4479: transforms as a doublet. As usual, from the tensor product of two
4480: doublets one can obtain a nontrivial singlet; this is just
4481: \begin{equation}
4482: \label{singlets}
4483: m b_{-n} c_{-m} + n b_{-m} c_{-n}\,.
4484: \end{equation}It is now simple to argue that the twist even subspace of
4485: $\mathcal{H}_1$ in the Siegel gauge can be further restricted
4486: to $SU(1,1)$ singlets. Since the kinetic operator
4487: $L_0$ commutes with the $SU(1,1)$ generators,
4488: the kinetic term cannot couple a non-singlet
4489: to a singlet. Indeed, consider such a term $\langle s| L_0 |a\rangle$,
4490: where $\langle s|$ is a singlet
4491: and $|a\rangle$ is not a singlet.
4492: Given the structure of the representations (completely analogous
4493: to the finite dimensional unitary representations of $SU(2)$), it
4494: follows that there is a state $|b\rangle$ and an $SU(1,1)$ generator
4495: $\mathcal{J}$ such that $|a\rangle = \mathcal{J} |b\rangle$. Therefore
4496: $\langle s| L_0 |a\rangle = \langle s| L_0 \mathcal{J}|b\rangle
4497: = \langle s|\mathcal{J} L_0 |b\rangle =0$, where the last step gives zero
4498: because
4499: $\mathcal{J}$ annihilates the singlet (this requires $bpz (\mathcal{J} )
4500: = \pm \mathcal{J}$, which is true). It remains
4501: to show that the vertex cannot couple a non-singlet to two singlets.
4502: Indeed, with analogous notation we have
4503: \begin{eqnarray}
4504: {}_1\langle s_1|{}_2\langle s_2| {}_3\langle a \, | v_3 \rangle
4505: &&= {}_1\langle s_1|{}_2\langle s_2| {}_3\langle b \, | \mathcal{J}^{(3)}| v_3
4506: \rangle \nonumber\\
4507: &&=- {}_1\langle s_1|{}_2\langle s_2| {}_3\langle b \, | (\mathcal{J}^{(1)}+
4508: \mathcal{J}^{(2)}) |\,v_3 \rangle =0 \,,
4509: \end{eqnarray}
4510: where we used the conservation of $\mathcal{J}$ on the vertex, and on the
4511: last step the $\mathcal{J}$ operators annihilate the singlets.
4512:
4513: \medskip
4514: This completes our discussion of the various symmetries and
4515: conditions that can be used to constrain the subspace of the
4516: string state space that acquires vacuum expectation values
4517: in the tachyon vacuum.
4518:
4519: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4520: % vacuum: WT
4521: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4522:
4523: \subsection{The nonperturbative vacuum}
4524: \label{sec:vacuum}
4525:
4526:
4527: Sen's first conjecture states that the string field theory action
4528: should lead to a nontrivial vacuum solution, with energy density
4529: \begin{equation}
4530: -T_{25} = -\frac{1}{2 \pi^2 g^2} \,.
4531: % % \label{eq:}
4532: \end{equation}
4533: In this subsection we discuss evidence for the validity of this
4534: conjecture in Witten's OSFT. As mentioned in the introduction, this
4535: result holds exactly in BSFT.
4536:
4537: The string field theory equation of motion is
4538: \begin{equation}
4539: Q \Psi + g \Psi \star \Psi = 0 \,.
4540: \label{eq:SFT-EOM}
4541: \end{equation}
4542: Despite much work over the last few years, there is still no analytic
4543: solution of this equation of motion\footnote{as of October, 2003}.
4544: There is, however, a systematic approximation scheme, known as level
4545: truncation, which can be used to solve this equation
4546: numerically~\cite{ks-open}.
4547: The level $(L, I) $ truncation of the full string field theory
4548: involves dropping all
4549: fields at level $N > L$, and disregarding any interaction term between
4550: three fields whose levels add up to a number that
4551: is greater than $I$. For
4552: example, the simplest truncation of the theory is the level (0, 0)
4553: truncation. This
4554: is the truncation which was used in section \ref{firsttestofit}.
4555: Including only
4556: the zero-momentum component of the tachyon field, since
4557: we are looking for a Lorentz-invariant vacuum, the truncated theory is
4558: simply described by a potential for the tachyon zero-mode
4559: \begin{equation}
4560: V (\phi) = -\frac{1}{2}\phi^2 + g \bar{\kappa} \phi^3 \,.
4561: % % \label{eq:}
4562: \end{equation}
4563: where $\bar{\kappa} = \kappa/3 = 3^{7/2}/2^6$. This cubic function,
4564: which was computed in (\ref{eq:CFT-cubic-potential}) using the CFT
4565: approach, and in (\ref{eq:expanded-action}) using the oscillator
4566: approach, is graphed in Figure~\ref{f:potential}.
4567: \begin{figure}
4568: % GNUPLOT: LaTeX picture
4569: \setlength{\unitlength}{0.240900pt}
4570: \ifx\plotpoint\undefined\newsavebox{\plotpoint}\fi
4571: \sbox{\plotpoint}{\rule[-0.200pt]{0.400pt}{0.400pt}}%
4572: \begin{picture}(1380,900)(120,0)
4573: \font\gnuplot=cmr10 at 10pt
4574: \gnuplot
4575: \sbox{\plotpoint}{\rule[-0.200pt]{0.400pt}{0.400pt}}%
4576: \put(280.0,675.0){\rule[-0.200pt]{4.818pt}{0.400pt}}
4577: \put(260,675){\makebox(0,0)[r]{0}}
4578: \put(1419.0,675.0){\rule[-0.200pt]{4.818pt}{0.400pt}}
4579: \put(280.0,332.0){\rule[-0.200pt]{4.818pt}{0.400pt}}
4580: \put(260,332){\makebox(0,0)[r]{$V (\phi_0)$}}
4581: \put(1419.0,332.0){\rule[-0.200pt]{4.818pt}{0.400pt}}
4582: \put(280.0,175.0){\rule[-0.200pt]{4.818pt}{0.400pt}}
4583: \put(260,175){\makebox(0,0)[r]{$-T_{25} $}}
4584: \put(1419.0,175.0){\rule[-0.200pt]{4.818pt}{0.400pt}}
4585: \put(598.0,82.0){\rule[-0.200pt]{0.400pt}{4.818pt}}
4586: \put(598,41){\makebox(0,0){0}}
4587: \put(598.0,840.0){\rule[-0.200pt]{0.400pt}{4.818pt}}
4588: \put(1127.0,82.0){\rule[-0.200pt]{0.400pt}{4.818pt}}
4589: \put(1127,41){\makebox(0,0){$\phi_0 = 1/3g\bar{\kappa}$}}
4590: \put(1127.0,840.0){\rule[-0.200pt]{0.400pt}{4.818pt}}
4591: \put(280.0,675.0){\rule[-0.200pt]{279.203pt}{0.400pt}}
4592: \put(1401,638){\makebox(0,0)[l]{$\phi$}}
4593: \put(480,841){\makebox(0,0)[l]{$V (\phi)$}}
4594: \put(598.0,82.0){\rule[-0.200pt]{0.400pt}{187.420pt}}
4595: \sbox{\plotpoint}{\rule[-0.400pt]{0.800pt}{0.800pt}}%
4596: \put(1213,786){\makebox(0,0)[r]{Level (0, 0) approximation}}
4597: \put(1233.0,786.0){\rule[-0.400pt]{24.090pt}{0.800pt}}
4598: \put(280,156){\usebox{\plotpoint}}
4599: \multiput(281.41,156.00)(0.511,1.892){17}{\rule{0.123pt}{3.067pt}}
4600: \multiput(278.34,156.00)(12.000,36.635){2}{\rule{0.800pt}{1.533pt}}
4601: \multiput(293.40,199.00)(0.512,1.935){15}{\rule{0.123pt}{3.109pt}}
4602: \multiput(290.34,199.00)(11.000,33.547){2}{\rule{0.800pt}{1.555pt}}
4603: \multiput(304.41,239.00)(0.511,1.666){17}{\rule{0.123pt}{2.733pt}}
4604: \multiput(301.34,239.00)(12.000,32.327){2}{\rule{0.800pt}{1.367pt}}
4605: \multiput(316.41,277.00)(0.511,1.621){17}{\rule{0.123pt}{2.667pt}}
4606: \multiput(313.34,277.00)(12.000,31.465){2}{\rule{0.800pt}{1.333pt}}
4607: \multiput(328.41,314.00)(0.511,1.485){17}{\rule{0.123pt}{2.467pt}}
4608: \multiput(325.34,314.00)(12.000,28.880){2}{\rule{0.800pt}{1.233pt}}
4609: \multiput(340.40,348.00)(0.512,1.536){15}{\rule{0.123pt}{2.527pt}}
4610: \multiput(337.34,348.00)(11.000,26.755){2}{\rule{0.800pt}{1.264pt}}
4611: \multiput(351.41,380.00)(0.511,1.304){17}{\rule{0.123pt}{2.200pt}}
4612: \multiput(348.34,380.00)(12.000,25.434){2}{\rule{0.800pt}{1.100pt}}
4613: \multiput(363.41,410.00)(0.511,1.259){17}{\rule{0.123pt}{2.133pt}}
4614: \multiput(360.34,410.00)(12.000,24.572){2}{\rule{0.800pt}{1.067pt}}
4615: \multiput(375.40,439.00)(0.512,1.237){15}{\rule{0.123pt}{2.091pt}}
4616: \multiput(372.34,439.00)(11.000,21.660){2}{\rule{0.800pt}{1.045pt}}
4617: \multiput(386.41,465.00)(0.511,1.078){17}{\rule{0.123pt}{1.867pt}}
4618: \multiput(383.34,465.00)(12.000,21.126){2}{\rule{0.800pt}{0.933pt}}
4619: \multiput(398.41,490.00)(0.511,0.988){17}{\rule{0.123pt}{1.733pt}}
4620: \multiput(395.34,490.00)(12.000,19.402){2}{\rule{0.800pt}{0.867pt}}
4621: \multiput(410.40,513.00)(0.512,0.988){15}{\rule{0.123pt}{1.727pt}}
4622: \multiput(407.34,513.00)(11.000,17.415){2}{\rule{0.800pt}{0.864pt}}
4623: \multiput(421.41,534.00)(0.511,0.807){17}{\rule{0.123pt}{1.467pt}}
4624: \multiput(418.34,534.00)(12.000,15.956){2}{\rule{0.800pt}{0.733pt}}
4625: \multiput(433.41,553.00)(0.511,0.762){17}{\rule{0.123pt}{1.400pt}}
4626: \multiput(430.34,553.00)(12.000,15.094){2}{\rule{0.800pt}{0.700pt}}
4627: \multiput(445.41,571.00)(0.511,0.717){17}{\rule{0.123pt}{1.333pt}}
4628: \multiput(442.34,571.00)(12.000,14.233){2}{\rule{0.800pt}{0.667pt}}
4629: \multiput(457.40,588.00)(0.512,0.639){15}{\rule{0.123pt}{1.218pt}}
4630: \multiput(454.34,588.00)(11.000,11.472){2}{\rule{0.800pt}{0.609pt}}
4631: \multiput(468.41,602.00)(0.511,0.536){17}{\rule{0.123pt}{1.067pt}}
4632: \multiput(465.34,602.00)(12.000,10.786){2}{\rule{0.800pt}{0.533pt}}
4633: \multiput(479.00,616.41)(0.491,0.511){17}{\rule{1.000pt}{0.123pt}}
4634: \multiput(479.00,613.34)(9.924,12.000){2}{\rule{0.500pt}{0.800pt}}
4635: \multiput(491.00,628.40)(0.489,0.512){15}{\rule{1.000pt}{0.123pt}}
4636: \multiput(491.00,625.34)(8.924,11.000){2}{\rule{0.500pt}{0.800pt}}
4637: \multiput(502.00,639.40)(0.674,0.516){11}{\rule{1.267pt}{0.124pt}}
4638: \multiput(502.00,636.34)(9.371,9.000){2}{\rule{0.633pt}{0.800pt}}
4639: \multiput(514.00,648.40)(0.913,0.526){7}{\rule{1.571pt}{0.127pt}}
4640: \multiput(514.00,645.34)(8.738,7.000){2}{\rule{0.786pt}{0.800pt}}
4641: \multiput(526.00,655.40)(0.913,0.526){7}{\rule{1.571pt}{0.127pt}}
4642: \multiput(526.00,652.34)(8.738,7.000){2}{\rule{0.786pt}{0.800pt}}
4643: \multiput(538.00,662.38)(1.432,0.560){3}{\rule{1.960pt}{0.135pt}}
4644: \multiput(538.00,659.34)(6.932,5.000){2}{\rule{0.980pt}{0.800pt}}
4645: \put(549,666.34){\rule{2.600pt}{0.800pt}}
4646: \multiput(549.00,664.34)(6.604,4.000){2}{\rule{1.300pt}{0.800pt}}
4647: \put(561,669.34){\rule{2.891pt}{0.800pt}}
4648: \multiput(561.00,668.34)(6.000,2.000){2}{\rule{1.445pt}{0.800pt}}
4649: \put(573,671.34){\rule{2.650pt}{0.800pt}}
4650: \multiput(573.00,670.34)(5.500,2.000){2}{\rule{1.325pt}{0.800pt}}
4651: \put(584,672.84){\rule{2.891pt}{0.800pt}}
4652: \multiput(584.00,672.34)(6.000,1.000){2}{\rule{1.445pt}{0.800pt}}
4653: \put(596,672.84){\rule{2.891pt}{0.800pt}}
4654: \multiput(596.00,673.34)(6.000,-1.000){2}{\rule{1.445pt}{0.800pt}}
4655: \put(608,671.84){\rule{2.891pt}{0.800pt}}
4656: \multiput(608.00,672.34)(6.000,-1.000){2}{\rule{1.445pt}{0.800pt}}
4657: \put(620,670.34){\rule{2.650pt}{0.800pt}}
4658: \multiput(620.00,671.34)(5.500,-2.000){2}{\rule{1.325pt}{0.800pt}}
4659: \put(631,667.84){\rule{2.891pt}{0.800pt}}
4660: \multiput(631.00,669.34)(6.000,-3.000){2}{\rule{1.445pt}{0.800pt}}
4661: \put(643,664.34){\rule{2.600pt}{0.800pt}}
4662: \multiput(643.00,666.34)(6.604,-4.000){2}{\rule{1.300pt}{0.800pt}}
4663: \multiput(655.00,662.06)(1.432,-0.560){3}{\rule{1.960pt}{0.135pt}}
4664: \multiput(655.00,662.34)(6.932,-5.000){2}{\rule{0.980pt}{0.800pt}}
4665: \multiput(666.00,657.07)(1.132,-0.536){5}{\rule{1.800pt}{0.129pt}}
4666: \multiput(666.00,657.34)(8.264,-6.000){2}{\rule{0.900pt}{0.800pt}}
4667: \multiput(678.00,651.07)(1.132,-0.536){5}{\rule{1.800pt}{0.129pt}}
4668: \multiput(678.00,651.34)(8.264,-6.000){2}{\rule{0.900pt}{0.800pt}}
4669: \multiput(690.00,645.08)(0.825,-0.526){7}{\rule{1.457pt}{0.127pt}}
4670: \multiput(690.00,645.34)(7.976,-7.000){2}{\rule{0.729pt}{0.800pt}}
4671: \multiput(701.00,638.08)(0.913,-0.526){7}{\rule{1.571pt}{0.127pt}}
4672: \multiput(701.00,638.34)(8.738,-7.000){2}{\rule{0.786pt}{0.800pt}}
4673: \multiput(713.00,631.08)(0.774,-0.520){9}{\rule{1.400pt}{0.125pt}}
4674: \multiput(713.00,631.34)(9.094,-8.000){2}{\rule{0.700pt}{0.800pt}}
4675: \multiput(725.00,623.08)(0.674,-0.516){11}{\rule{1.267pt}{0.124pt}}
4676: \multiput(725.00,623.34)(9.371,-9.000){2}{\rule{0.633pt}{0.800pt}}
4677: \multiput(737.00,614.08)(0.611,-0.516){11}{\rule{1.178pt}{0.124pt}}
4678: \multiput(737.00,614.34)(8.555,-9.000){2}{\rule{0.589pt}{0.800pt}}
4679: \multiput(748.00,605.08)(0.674,-0.516){11}{\rule{1.267pt}{0.124pt}}
4680: \multiput(748.00,605.34)(9.371,-9.000){2}{\rule{0.633pt}{0.800pt}}
4681: \multiput(760.00,596.08)(0.599,-0.514){13}{\rule{1.160pt}{0.124pt}}
4682: \multiput(760.00,596.34)(9.592,-10.000){2}{\rule{0.580pt}{0.800pt}}
4683: \multiput(772.00,586.08)(0.543,-0.514){13}{\rule{1.080pt}{0.124pt}}
4684: \multiput(772.00,586.34)(8.758,-10.000){2}{\rule{0.540pt}{0.800pt}}
4685: \multiput(783.00,576.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4686: \multiput(783.00,576.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4687: \multiput(795.00,565.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4688: \multiput(795.00,565.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4689: \multiput(807.00,554.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4690: \multiput(807.00,554.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4691: \multiput(819.00,543.08)(0.489,-0.512){15}{\rule{1.000pt}{0.123pt}}
4692: \multiput(819.00,543.34)(8.924,-11.000){2}{\rule{0.500pt}{0.800pt}}
4693: \multiput(830.00,532.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4694: \multiput(830.00,532.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4695: \multiput(842.00,521.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4696: \multiput(842.00,521.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4697: \multiput(855.40,507.55)(0.512,-0.539){15}{\rule{0.123pt}{1.073pt}}
4698: \multiput(852.34,509.77)(11.000,-9.774){2}{\rule{0.800pt}{0.536pt}}
4699: \multiput(865.00,498.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4700: \multiput(865.00,498.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4701: \multiput(877.00,487.08)(0.491,-0.511){17}{\rule{1.000pt}{0.123pt}}
4702: \multiput(877.00,487.34)(9.924,-12.000){2}{\rule{0.500pt}{0.800pt}}
4703: \multiput(889.00,475.08)(0.489,-0.512){15}{\rule{1.000pt}{0.123pt}}
4704: \multiput(889.00,475.34)(8.924,-11.000){2}{\rule{0.500pt}{0.800pt}}
4705: \multiput(900.00,464.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4706: \multiput(900.00,464.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4707: \multiput(912.00,453.08)(0.539,-0.512){15}{\rule{1.073pt}{0.123pt}}
4708: \multiput(912.00,453.34)(9.774,-11.000){2}{\rule{0.536pt}{0.800pt}}
4709: \multiput(924.00,442.08)(0.599,-0.514){13}{\rule{1.160pt}{0.124pt}}
4710: \multiput(924.00,442.34)(9.592,-10.000){2}{\rule{0.580pt}{0.800pt}}
4711: \multiput(936.00,432.08)(0.489,-0.512){15}{\rule{1.000pt}{0.123pt}}
4712: \multiput(936.00,432.34)(8.924,-11.000){2}{\rule{0.500pt}{0.800pt}}
4713: \multiput(947.00,421.08)(0.599,-0.514){13}{\rule{1.160pt}{0.124pt}}
4714: \multiput(947.00,421.34)(9.592,-10.000){2}{\rule{0.580pt}{0.800pt}}
4715: \multiput(959.00,411.08)(0.674,-0.516){11}{\rule{1.267pt}{0.124pt}}
4716: \multiput(959.00,411.34)(9.371,-9.000){2}{\rule{0.633pt}{0.800pt}}
4717: \multiput(971.00,402.08)(0.543,-0.514){13}{\rule{1.080pt}{0.124pt}}
4718: \multiput(971.00,402.34)(8.758,-10.000){2}{\rule{0.540pt}{0.800pt}}
4719: \multiput(982.00,392.08)(0.774,-0.520){9}{\rule{1.400pt}{0.125pt}}
4720: \multiput(982.00,392.34)(9.094,-8.000){2}{\rule{0.700pt}{0.800pt}}
4721: \multiput(994.00,384.08)(0.674,-0.516){11}{\rule{1.267pt}{0.124pt}}
4722: \multiput(994.00,384.34)(9.371,-9.000){2}{\rule{0.633pt}{0.800pt}}
4723: \multiput(1006.00,375.08)(0.913,-0.526){7}{\rule{1.571pt}{0.127pt}}
4724: \multiput(1006.00,375.34)(8.738,-7.000){2}{\rule{0.786pt}{0.800pt}}
4725: \multiput(1018.00,368.08)(0.700,-0.520){9}{\rule{1.300pt}{0.125pt}}
4726: \multiput(1018.00,368.34)(8.302,-8.000){2}{\rule{0.650pt}{0.800pt}}
4727: \multiput(1029.00,360.07)(1.132,-0.536){5}{\rule{1.800pt}{0.129pt}}
4728: \multiput(1029.00,360.34)(8.264,-6.000){2}{\rule{0.900pt}{0.800pt}}
4729: \multiput(1041.00,354.07)(1.132,-0.536){5}{\rule{1.800pt}{0.129pt}}
4730: \multiput(1041.00,354.34)(8.264,-6.000){2}{\rule{0.900pt}{0.800pt}}
4731: \multiput(1053.00,348.06)(1.432,-0.560){3}{\rule{1.960pt}{0.135pt}}
4732: \multiput(1053.00,348.34)(6.932,-5.000){2}{\rule{0.980pt}{0.800pt}}
4733: \put(1064,341.34){\rule{2.600pt}{0.800pt}}
4734: \multiput(1064.00,343.34)(6.604,-4.000){2}{\rule{1.300pt}{0.800pt}}
4735: \put(1076,337.34){\rule{2.600pt}{0.800pt}}
4736: \multiput(1076.00,339.34)(6.604,-4.000){2}{\rule{1.300pt}{0.800pt}}
4737: \put(1088,333.84){\rule{2.650pt}{0.800pt}}
4738: \multiput(1088.00,335.34)(5.500,-3.000){2}{\rule{1.325pt}{0.800pt}}
4739: \put(1099,331.84){\rule{2.891pt}{0.800pt}}
4740: \multiput(1099.00,332.34)(6.000,-1.000){2}{\rule{1.445pt}{0.800pt}}
4741: \put(1111,330.84){\rule{2.891pt}{0.800pt}}
4742: \multiput(1111.00,331.34)(6.000,-1.000){2}{\rule{1.445pt}{0.800pt}}
4743: \put(1135,330.84){\rule{2.650pt}{0.800pt}}
4744: \multiput(1135.00,330.34)(5.500,1.000){2}{\rule{1.325pt}{0.800pt}}
4745: \put(1146,332.34){\rule{2.891pt}{0.800pt}}
4746: \multiput(1146.00,331.34)(6.000,2.000){2}{\rule{1.445pt}{0.800pt}}
4747: \put(1158,335.34){\rule{2.600pt}{0.800pt}}
4748: \multiput(1158.00,333.34)(6.604,4.000){2}{\rule{1.300pt}{0.800pt}}
4749: \put(1170,339.34){\rule{2.400pt}{0.800pt}}
4750: \multiput(1170.00,337.34)(6.019,4.000){2}{\rule{1.200pt}{0.800pt}}
4751: \multiput(1181.00,344.39)(1.132,0.536){5}{\rule{1.800pt}{0.129pt}}
4752: \multiput(1181.00,341.34)(8.264,6.000){2}{\rule{0.900pt}{0.800pt}}
4753: \multiput(1193.00,350.40)(0.913,0.526){7}{\rule{1.571pt}{0.127pt}}
4754: \multiput(1193.00,347.34)(8.738,7.000){2}{\rule{0.786pt}{0.800pt}}
4755: \multiput(1205.00,357.40)(0.674,0.516){11}{\rule{1.267pt}{0.124pt}}
4756: \multiput(1205.00,354.34)(9.371,9.000){2}{\rule{0.633pt}{0.800pt}}
4757: \multiput(1217.00,366.40)(0.611,0.516){11}{\rule{1.178pt}{0.124pt}}
4758: \multiput(1217.00,363.34)(8.555,9.000){2}{\rule{0.589pt}{0.800pt}}
4759: \multiput(1228.00,375.41)(0.491,0.511){17}{\rule{1.000pt}{0.123pt}}
4760: \multiput(1228.00,372.34)(9.924,12.000){2}{\rule{0.500pt}{0.800pt}}
4761: \multiput(1240.00,387.41)(0.491,0.511){17}{\rule{1.000pt}{0.123pt}}
4762: \multiput(1240.00,384.34)(9.924,12.000){2}{\rule{0.500pt}{0.800pt}}
4763: \multiput(1253.40,398.00)(0.512,0.639){15}{\rule{0.123pt}{1.218pt}}
4764: \multiput(1250.34,398.00)(11.000,11.472){2}{\rule{0.800pt}{0.609pt}}
4765: \multiput(1264.41,412.00)(0.511,0.671){17}{\rule{0.123pt}{1.267pt}}
4766: \multiput(1261.34,412.00)(12.000,13.371){2}{\rule{0.800pt}{0.633pt}}
4767: \multiput(1276.41,428.00)(0.511,0.717){17}{\rule{0.123pt}{1.333pt}}
4768: \multiput(1273.34,428.00)(12.000,14.233){2}{\rule{0.800pt}{0.667pt}}
4769: \multiput(1288.41,445.00)(0.511,0.807){17}{\rule{0.123pt}{1.467pt}}
4770: \multiput(1285.34,445.00)(12.000,15.956){2}{\rule{0.800pt}{0.733pt}}
4771: \multiput(1300.40,464.00)(0.512,0.938){15}{\rule{0.123pt}{1.655pt}}
4772: \multiput(1297.34,464.00)(11.000,16.566){2}{\rule{0.800pt}{0.827pt}}
4773: \multiput(1311.41,484.00)(0.511,0.943){17}{\rule{0.123pt}{1.667pt}}
4774: \multiput(1308.34,484.00)(12.000,18.541){2}{\rule{0.800pt}{0.833pt}}
4775: \multiput(1323.41,506.00)(0.511,1.033){17}{\rule{0.123pt}{1.800pt}}
4776: \multiput(1320.34,506.00)(12.000,20.264){2}{\rule{0.800pt}{0.900pt}}
4777: \multiput(1335.40,530.00)(0.512,1.237){15}{\rule{0.123pt}{2.091pt}}
4778: \multiput(1332.34,530.00)(11.000,21.660){2}{\rule{0.800pt}{1.045pt}}
4779: \multiput(1346.41,556.00)(0.511,1.169){17}{\rule{0.123pt}{2.000pt}}
4780: \multiput(1343.34,556.00)(12.000,22.849){2}{\rule{0.800pt}{1.000pt}}
4781: \multiput(1358.41,583.00)(0.511,1.304){17}{\rule{0.123pt}{2.200pt}}
4782: \multiput(1355.34,583.00)(12.000,25.434){2}{\rule{0.800pt}{1.100pt}}
4783: \multiput(1370.40,613.00)(0.512,1.486){15}{\rule{0.123pt}{2.455pt}}
4784: \multiput(1367.34,613.00)(11.000,25.905){2}{\rule{0.800pt}{1.227pt}}
4785: \multiput(1381.41,644.00)(0.511,1.440){17}{\rule{0.123pt}{2.400pt}}
4786: \multiput(1378.34,644.00)(12.000,28.019){2}{\rule{0.800pt}{1.200pt}}
4787: \multiput(1393.41,677.00)(0.511,1.530){17}{\rule{0.123pt}{2.533pt}}
4788: \multiput(1390.34,677.00)(12.000,29.742){2}{\rule{0.800pt}{1.267pt}}
4789: \multiput(1405.41,712.00)(0.511,1.666){17}{\rule{0.123pt}{2.733pt}}
4790: \multiput(1402.34,712.00)(12.000,32.327){2}{\rule{0.800pt}{1.367pt}}
4791: \multiput(1417.40,750.00)(0.512,1.885){15}{\rule{0.123pt}{3.036pt}}
4792: \multiput(1414.34,750.00)(11.000,32.698){2}{\rule{0.800pt}{1.518pt}}
4793: \multiput(1428.41,789.00)(0.511,1.847){17}{\rule{0.123pt}{3.000pt}}
4794: \multiput(1425.34,789.00)(12.000,35.773){2}{\rule{0.800pt}{1.500pt}}
4795: \put(1123.0,332.0){\rule[-0.400pt]{2.891pt}{0.800pt}}
4796: \sbox{\plotpoint}{\rule[-0.500pt]{1.000pt}{1.000pt}}%
4797: \put(1213,745){\makebox(0,0)[r]{Level (2, 6) approximation}}
4798: \multiput(1233,745)(20.756,0.000){5}{\usebox{\plotpoint}}
4799: \put(1333,745){\usebox{\plotpoint}}
4800: \put(399,477){\usebox{\plotpoint}}
4801: \multiput(399,477)(7.708,19.271){2}{\usebox{\plotpoint}}
4802: \put(415.53,515.06){\usebox{\plotpoint}}
4803: \put(424.91,533.57){\usebox{\plotpoint}}
4804: \put(434.87,551.78){\usebox{\plotpoint}}
4805: \put(445.52,569.59){\usebox{\plotpoint}}
4806: \put(457.19,586.74){\usebox{\plotpoint}}
4807: \put(470.16,602.94){\usebox{\plotpoint}}
4808: \put(483.09,619.09){\usebox{\plotpoint}}
4809: \put(498.01,633.51){\usebox{\plotpoint}}
4810: \put(514.36,646.27){\usebox{\plotpoint}}
4811: \put(531.74,657.59){\usebox{\plotpoint}}
4812: \put(550.57,666.21){\usebox{\plotpoint}}
4813: \put(570.43,672.11){\usebox{\plotpoint}}
4814: \put(590.95,675.00){\usebox{\plotpoint}}
4815: \put(611.66,674.17){\usebox{\plotpoint}}
4816: \put(632.00,670.25){\usebox{\plotpoint}}
4817: \put(651.89,664.42){\usebox{\plotpoint}}
4818: \put(671.22,656.89){\usebox{\plotpoint}}
4819: \put(689.54,647.17){\usebox{\plotpoint}}
4820: \put(707.14,636.16){\usebox{\plotpoint}}
4821: \put(724.30,624.53){\usebox{\plotpoint}}
4822: \put(740.53,611.60){\usebox{\plotpoint}}
4823: \put(756.59,598.48){\usebox{\plotpoint}}
4824: \put(772.21,584.82){\usebox{\plotpoint}}
4825: \put(787.36,570.64){\usebox{\plotpoint}}
4826: \put(802.52,556.48){\usebox{\plotpoint}}
4827: \put(817.19,541.81){\usebox{\plotpoint}}
4828: \put(831.36,526.64){\usebox{\plotpoint}}
4829: \put(845.91,511.85){\usebox{\plotpoint}}
4830: \put(859.70,496.34){\usebox{\plotpoint}}
4831: \put(873.43,480.77){\usebox{\plotpoint}}
4832: \put(887.21,465.26){\usebox{\plotpoint}}
4833: \put(901.00,449.75){\usebox{\plotpoint}}
4834: \put(914.79,434.23){\usebox{\plotpoint}}
4835: \put(928.58,418.72){\usebox{\plotpoint}}
4836: \put(942.37,403.21){\usebox{\plotpoint}}
4837: \put(956.16,387.69){\usebox{\plotpoint}}
4838: \put(969.95,372.18){\usebox{\plotpoint}}
4839: \put(984.22,357.12){\usebox{\plotpoint}}
4840: \put(997.66,341.34){\usebox{\plotpoint}}
4841: \put(1012.19,326.53){\usebox{\plotpoint}}
4842: \put(1026.53,311.54){\usebox{\plotpoint}}
4843: \put(1041.66,297.34){\usebox{\plotpoint}}
4844: \put(1056.74,283.10){\usebox{\plotpoint}}
4845: \put(1072.36,269.44){\usebox{\plotpoint}}
4846: \put(1088.36,256.23){\usebox{\plotpoint}}
4847: \put(1104.96,243.78){\usebox{\plotpoint}}
4848: \put(1121.72,231.55){\usebox{\plotpoint}}
4849: \put(1139.45,220.77){\usebox{\plotpoint}}
4850: \put(1158.02,211.49){\usebox{\plotpoint}}
4851: \put(1176.94,203.02){\usebox{\plotpoint}}
4852: \put(1197.07,197.98){\usebox{\plotpoint}}
4853: \put(1217.57,194.93){\usebox{\plotpoint}}
4854: \put(1238.23,194.65){\usebox{\plotpoint}}
4855: \put(1258.75,197.69){\usebox{\plotpoint}}
4856: \put(1278.64,203.49){\usebox{\plotpoint}}
4857: \put(1297.20,212.75){\usebox{\plotpoint}}
4858: \put(1314.64,223.98){\usebox{\plotpoint}}
4859: \put(1330.58,237.26){\usebox{\plotpoint}}
4860: \put(1345.49,251.68){\usebox{\plotpoint}}
4861: \put(1359.28,267.19){\usebox{\plotpoint}}
4862: \put(1371.94,283.62){\usebox{\plotpoint}}
4863: \put(1383.55,300.82){\usebox{\plotpoint}}
4864: \multiput(1391,312)(10.878,17.677){2}{\usebox{\plotpoint}}
4865: \put(1415.61,354.15){\usebox{\plotpoint}}
4866: \put(1425.26,372.53){\usebox{\plotpoint}}
4867: \put(1434.38,391.17){\usebox{\plotpoint}}
4868: \put(1439,401){\usebox{\plotpoint}}
4869: \sbox{\plotpoint}{\rule[-0.200pt]{0.400pt}{0.400pt}}%
4870: \put(1213,704){\makebox(0,0)[r]{Exact value of $-T_{25}$}}
4871: \multiput(1233,704)(20.756,0.000){5}{\usebox{\plotpoint}}
4872: \put(1333,704){\usebox{\plotpoint}}
4873: \put(280,175){\usebox{\plotpoint}}
4874: \put(280.00,175.00){\usebox{\plotpoint}}
4875: \put(300.76,175.00){\usebox{\plotpoint}}
4876: \put(321.51,175.00){\usebox{\plotpoint}}
4877: \put(342.27,175.00){\usebox{\plotpoint}}
4878: \put(363.02,175.00){\usebox{\plotpoint}}
4879: \put(383.78,175.00){\usebox{\plotpoint}}
4880: \put(404.53,175.00){\usebox{\plotpoint}}
4881: \put(425.29,175.00){\usebox{\plotpoint}}
4882: \put(446.04,175.00){\usebox{\plotpoint}}
4883: \put(466.80,175.00){\usebox{\plotpoint}}
4884: \put(487.55,175.00){\usebox{\plotpoint}}
4885: \put(508.31,175.00){\usebox{\plotpoint}}
4886: \put(529.07,175.00){\usebox{\plotpoint}}
4887: \put(549.82,175.00){\usebox{\plotpoint}}
4888: \put(570.58,175.00){\usebox{\plotpoint}}
4889: \put(591.33,175.00){\usebox{\plotpoint}}
4890: \put(612.09,175.00){\usebox{\plotpoint}}
4891: \put(632.84,175.00){\usebox{\plotpoint}}
4892: \put(653.60,175.00){\usebox{\plotpoint}}
4893: \put(674.35,175.00){\usebox{\plotpoint}}
4894: \put(695.11,175.00){\usebox{\plotpoint}}
4895: \put(715.87,175.00){\usebox{\plotpoint}}
4896: \put(736.62,175.00){\usebox{\plotpoint}}
4897: \put(757.38,175.00){\usebox{\plotpoint}}
4898: \put(778.13,175.00){\usebox{\plotpoint}}
4899: \put(798.89,175.00){\usebox{\plotpoint}}
4900: \put(819.64,175.00){\usebox{\plotpoint}}
4901: \put(840.40,175.00){\usebox{\plotpoint}}
4902: \put(861.15,175.00){\usebox{\plotpoint}}
4903: \put(881.91,175.00){\usebox{\plotpoint}}
4904: \put(902.66,175.00){\usebox{\plotpoint}}
4905: \put(923.42,175.00){\usebox{\plotpoint}}
4906: \put(944.18,175.00){\usebox{\plotpoint}}
4907: \put(964.93,175.00){\usebox{\plotpoint}}
4908: \put(985.69,175.00){\usebox{\plotpoint}}
4909: \put(1006.44,175.00){\usebox{\plotpoint}}
4910: \put(1027.20,175.00){\usebox{\plotpoint}}
4911: \put(1047.95,175.00){\usebox{\plotpoint}}
4912: \put(1068.71,175.00){\usebox{\plotpoint}}
4913: \put(1089.46,175.00){\usebox{\plotpoint}}
4914: \put(1110.22,175.00){\usebox{\plotpoint}}
4915: \put(1130.98,175.00){\usebox{\plotpoint}}
4916: \put(1151.73,175.00){\usebox{\plotpoint}}
4917: \put(1172.49,175.00){\usebox{\plotpoint}}
4918: \put(1193.24,175.00){\usebox{\plotpoint}}
4919: \put(1214.00,175.00){\usebox{\plotpoint}}
4920: \put(1234.75,175.00){\usebox{\plotpoint}}
4921: \put(1255.51,175.00){\usebox{\plotpoint}}
4922: \put(1276.26,175.00){\usebox{\plotpoint}}
4923: \put(1297.02,175.00){\usebox{\plotpoint}}
4924: \put(1317.77,175.00){\usebox{\plotpoint}}
4925: \put(1338.53,175.00){\usebox{\plotpoint}}
4926: \put(1359.29,175.00){\usebox{\plotpoint}}
4927: \put(1380.04,175.00){\usebox{\plotpoint}}
4928: \put(1400.80,175.00){\usebox{\plotpoint}}
4929: \put(1421.55,175.00){\usebox{\plotpoint}}
4930: \put(1439,175){\usebox{\plotpoint}}
4931: \sbox{\plotpoint}{\rule[-0.500pt]{1.000pt}{1.000pt}}%
4932: \put(1127,332){\raisebox{-.8pt}{\circle{15}}}
4933: \put(1229,194){\raisebox{-.8pt}{\circle{15}}}
4934: \put(396,471){\raisebox{-.8pt}{\circle*{15}}}
4935: \end{picture}
4936: \caption[x]{\footnotesize The effective tachyon potential in level (0,
4937: 0) and (2, 6) truncations. The open circles denote minima in each
4938: level truncation. The filled circle denotes a branch point where
4939: the level (2, 6) truncation approximation reaches the limit of
4940: Feynman-Siegel gauge validity.}
4941: \label{f:potential}
4942: \end{figure}
4943: As discussed in section \ref{firsttestofit}, this potential has a
4944: local minimum~at
4945: \begin{equation}
4946: \phi_0 = \frac{1}{3g \bar{\kappa}} \,,
4947: % % \label{eq:}
4948: \end{equation}
4949: and at this point the potential is
4950: \begin{equation}
4951: V (\phi_0) = -\frac{1}{54} \frac{1}{g^2 \bar{\kappa}^2} =
4952: -\frac{2^{11}}{3^{10}} \frac{1}{g^2}
4953: \approx (0.68) \left( -\frac{1}{2 \pi^2 g^2} \right) \,.
4954: % % \label{eq:}
4955: \end{equation}
4956: Thus, simply including the tachyon zero-mode gives a
4957: nontrivial vacuum with $68\%$ of the vacuum energy density predicted
4958: by Sen. This vacuum is denoted by an open circle in Figure~\ref{f:potential}.
4959:
4960: At higher levels of truncation, there are a multitude of fields with
4961: various tensor structures. However, again assuming that we are
4962: looking for a vacuum which preserves Lorentz symmetry, we can restrict
4963: attention to the interactions between scalar fields at zero momentum. We
4964: will work in Feynman-Siegel gauge to simplify calculations; as shown
4965: in the previous subsection, this gauge is good at least in a local
4966: neighborhood of the point where all fields vanish. The situation is
4967: further simplified by the existence of the twist symmetry,
4968: which as mentioned in the previous subsection guarantees that no cubic
4969: vertex between (zero-momentum) scalar fields can connect three fields with a
4970: total level which is odd, and thus means that odd fields are not
4971: relevant to diagrams with only external tachyons at tree level.
4972: Therefore, we need only consider even-level scalar fields in looking
4973: for Lorentz-preserving solutions to the SFT equations of motion. With
4974: these simplifications, in a general level truncation the string field
4975: is simply expressed as a sum of a finite number of terms
4976: \begin{equation}
4977: \Psi_{\rm s} = \sum_{i} \phi_i | s_i \rangle
4978: \label{eq:scalar-expansion}
4979: \end{equation}
4980: where $\phi_i$ are the zero-modes of the scalar fields associated with
4981: even-level states $|s_i \rangle$. As discussed in the previous
4982: subsection, this set of scalar fields can be further restricted to be
4983: SU(1, 1) singlets in the universal subspace ${\mathcal H}_1$.
4984: For example, including fields up to
4985: level 2, we have
4986: \begin{equation}
4987: \Psi_{{\rm s}} = \phi| 0_1 \rangle+
4988: B \; (\alpha_{-1} \cdot \alpha_{-1})| 0_1 \rangle+
4989: \beta \; b_{-1} c_{-1}| 0_1 \rangle\,.
4990: % % \label{eq:}
4991: \end{equation}
4992: In terms of the matter Virasoro generators, the state associated with
4993: the field $B$ is
4994: \begin{equation}
4995: (\alpha_{-1} \cdot \alpha_{-1}) | 0_1 \rangle
4996: = 2L_{-2}| 0_1 \rangle\,,
4997: % \label{eq:}
4998: \end{equation}
4999: which lies in the universal subspace ${\mathcal H}_1$.
5000: The potential for all the scalars appearing in the level-truncated
5001: expansion (\ref{eq:scalar-expansion}) can be simply expressed as a
5002: cubic polynomial in the zero-modes of the scalar fields
5003: \begin{equation}
5004: V = \sum_{i, j}d_{ij} \phi_i \phi_j + g \bar{\kappa}
5005: \sum_{i, j, k}t_{ijk} \phi_i \phi_j \phi_k \,.
5006: \label{eq:scalar-potential}
5007: \end{equation}
5008: Using the expressions for the Neumann coefficients given in Section
5009: 5.3, the potential for all the scalar fields up to level $L$ can be
5010: computed in a level $(L, I)$ truncation. For example, the potential
5011: in the level $(2, 6)$ truncation is given~by
5012: \begin{eqnarray}
5013: V & = & -\frac{1}{2}\phi^2 + 26 B^2 -\frac{1}{2}\beta^2
5014: %+ 26B \eta + 3 \beta \eta + 2 \eta^2
5015: \label{eq:v26}\\
5016: & &+ \bar{\kappa} g\left[
5017: \phi^3 -\frac{130}{9} \phi^2 B -\frac{11}{9} \phi^2 \beta
5018: %-\frac{16}{9} \phi^2 \eta
5019: + \frac{30212}{243} \phi B^2
5020: + \frac{2860}{243} \phi B \beta
5021: %\right. \label{eq:v26}\\
5022: %& &\hspace*{0.4in}\left.
5023: %+ \frac{4160}{243} \phi B \eta
5024: + \frac{19}{81} \phi \beta^2
5025: %-\frac{32}{27} \phi \beta \eta
5026: %+ \frac{64}{81} \phi \eta^2
5027: \right. \nonumber \\
5028: & &\hspace*{0.4in}\left.
5029: -\frac{2178904}{6561} B^3
5030: -\frac{332332}{6561} B^2 \beta
5031: %-\frac{483392}{6561} B^2 \eta
5032: -\frac{2470}{2187} B \beta^2
5033: %+ \frac{4160}{729} B \beta \eta
5034: %-\frac{8320}{2187} B \eta^2
5035: % \right. \nonumber\\
5036: %& &\hspace*{0.4in}\left.
5037: -\frac{1}{81} \beta^3
5038: %+ \frac{11248}{6561} \beta^2 \eta
5039: %+ \frac{3008}{6561} \beta \eta^2
5040: \right] \nonumber
5041: \end{eqnarray}
5042: As an example of how these terms arise, consider the $\phi^2 B$ term.
5043: The coefficient in this term is given by
5044: \begin{equation}g\, \langle V_3 | \, \bigl(| 0_1 \rangle \otimes |
5045: 0_1 \rangle \otimes
5046: \alpha_{-1} \cdot \alpha_{-1} | 0_1 \rangle\bigr) =
5047: -g \bar{\kappa} \; (3 \cdot 26) \; V^{11}_{11}
5048: = -g \bar{\kappa} \frac{130}{9} \,,
5049: \end{equation}where we have used $V^{11}_{11} = 5/27$.
5050:
5051: In the level (2, 6) truncation of the theory, the nontrivial vacuum
5052: is found by simultaneously solving the three quadratic equations found by
5053: setting to zero the derivatives of the potential (\ref{eq:v26}) with respect to
5054: $\phi, B,$ and $\beta$. There are a number of different solutions to
5055: these equations, but only one is in the vicinity of $\phi = 1/3g
5056: \bar{\kappa}$. The solution of interest is
5057: \begin{eqnarray}
5058: \phi \approx 0.39766\, \frac{1}{g \bar{\kappa}} \,,\quad
5059: B \approx 0.02045\, \frac{1}{g \bar{\kappa}}\, ,\quad
5060: \beta \approx -0.13897\, \frac{1}{g \bar{\kappa}} \,.
5061: \end{eqnarray}
5062: Plugging these values into the potential gives
5063: \begin{equation}
5064: E_{(2, 6)} = -0.95938 \,T_{25} \,,
5065: % % \label{eq:}
5066: \end{equation}
5067: or 95.9\% of the result predicted by Sen.
5068: This vacuum is denoted by an open circle in Figure~\ref{f:potential}.
5069:
5070: It is a straightforward, computationally intensive project
5071: to generalize this calculation to higher levels of truncation. This
5072: calculation was carried out to level (4, 8) by Kostelecky and
5073: Samuel~\cite{ks-open} many years ago.
5074: They noted that the vacuum seemed to be converging, but they lacked any
5075: physical picture to give meaning to this vacuum.
5076: Following Sen's conjectures, the
5077: level (4, 8) calculation was done again using somewhat different methods by Sen
5078: and Zwiebach~\cite{Sen-Zwiebach}, who showed that the energy at this level is
5079: $-0.986 \;T_{25}$. The calculation was automated by Moeller and
5080: Taylor~\cite{Moeller-Taylor}, who calculated up to level (10, 20),
5081: where there are
5082: 252 scalar fields, including all even-level scalar fields up to level 10; this
5083: computation was done using oscillators, without restriction to the universal
5084: subspace. Up to this level, the vacuum energy converges monotonically, as
5085: shown in Table 1.
5086: \begin{table}[htp]
5087: \begin{center}
5088: \begin{tabular}{|| c || c | c ||}
5089: \hline
5090: \hline
5091: level & $ g \bar{\kappa}\langle \phi \rangle$ & $V/T_{25}$\\
5092: \hline
5093: \hline
5094: (0, 0) & 0.3333 & -0.68462\\
5095: \hline
5096: (2, 4) & 0.3957 & -0.94855\\
5097: (2, 6) & 0.3977 & -0.95938\\
5098: \hline
5099: (4, 8) & 0.4005 & -0.98640\\
5100: (4, 12) & 0.4007 & -0.98782\\
5101: \hline
5102: (6, 12) & 0.4004 & -0.99514\\
5103: (6, 18) & 0.4004 & -0.99518\\
5104: \hline
5105: (8, 16) & 0.3999 & -0.99777\\
5106: (8, 20) & 0.3997 & -0.99793\\
5107: \hline
5108: (10, 20) & 0.3992 & -0.99912\\
5109: \hline
5110: \hline
5111: \end{tabular}
5112: \caption[x]{\footnotesize Tachyon VEV and vacuum energy in stable
5113: vacua of level-truncated theory}
5114: \label{t:vacuum}
5115: \end{center}
5116: \end{table}
5117: These numerical calculations indicate that level truncation of string
5118: field theory leads to a good systematic approximation scheme for
5119: computing the nonperturbative tachyon vacuum. It is also worth noting
5120: that in these computations, level $(L, 2L)$ and $(L, 3L)$
5121: approximations give fairly similar values.
5122:
5123:
5124: The preceding results were the best values for the vacuum energy at
5125: the time of these original lectures. More recently, Gaiotto and
5126: Rastelli reported further numerical
5127: results~\cite{Gaiotto-Rastelli-strings,gr-analysis}. By programming in
5128: C++ instead of mathematica, and by computing using matter Virasoro
5129: operators rather than oscillators, so that only fields in the
5130: universal subspace ${\mathcal H}_1$ were included, they were able to
5131: extend the computation to level (18, 54). Their results are shown in
5132: Table~\ref{t:vacuum-gr}
5133: \begin{table}[htp]
5134: \begin{center}
5135: \begin{tabular}{|| c | c ||}
5136: \hline
5137: \hline
5138: level & $V/T_{25}$\\
5139: \hline
5140: \hline
5141: (12, 24) & 0.99979 \\
5142: (12, 36) & 0.99982 \\
5143: \hline
5144: (14, 28) & 1.00016 \\
5145: (14, 42) & 1.00017 \\
5146: \hline
5147: (16, 32) & 1.00037 \\
5148: (16, 48) & 1.00038 \\
5149: \hline
5150: (18, 36) & 1.00049 \\
5151: (18, 54) & 1.00049 \\
5152: \hline
5153: \hline
5154: \end{tabular}
5155: \caption[x]{\footnotesize Vacuum energy in stable
5156: vacua of level-truncated theory}
5157: \label{t:vacuum-gr}
5158: \end{center}
5159: \end{table}
5160: These results were rather surprising, indicating that
5161: while the energy monotonically approaches $-T_{25}$ up to level 12, at
5162: level (14, 42) the energy drops below $-T_{25}$, and that the
5163: energy continues to decrease, reaching $-1.00049\, T_{25}$ at level
5164: (18, 54). We will discuss the resolution of this unexpected overshoot
5165: shortly.
5166:
5167:
5168: First, however, it is interesting to consider the tachyon condensation
5169: problem from the point of view of the effective tachyon potential. If
5170: instead of trying to solve the quadratic equations for all $N$ of the
5171: fields appearing in (\ref{eq:scalar-potential}), we instead fix the
5172: tachyon field $\phi$ and solve the quadratic equations for the
5173: remaining $N -1$ fields, we can determine a effective potential $V
5174: (\phi)$ for the tachyon field. This has been done numerically up to
5175: level (16, 48)~\cite{Moeller-Taylor,gr-analysis}. At each level, the
5176: tachyon effective potential smoothly interpolates between the
5177: perturbative vacuum and the nonperturbative vacuum near $\phi = 0.4/g
5178: \bar{\kappa}$. For example, the tachyon effective potential at level
5179: (2, 6) is graphed in Figure~\ref{f:potential}. In all level
5180: truncations other than (0, 0) and (2, 4) (at least up to level (10,
5181: 20)), the tachyon effective potential has two branch point
5182: singularities at which the continuous solution for the other fields
5183: breaks down; for the level (2, 6) truncation, these branch points
5184: occur at $\phi \approx -0.127/g \bar{\kappa}$ and $\phi \approx
5185: 2.293/g \bar{\kappa}$; the lower branch point is denoted by a solid
5186: circle in Figure~\ref{f:potential}. As a result of these branch
5187: points, the tachyon effective potential is only valid for a finite
5188: range of $\phi$, ranging between approximately $-0.1/g \bar{\kappa}$
5189: and $ 0.6/g \bar{\kappa}$. In Section \ref{sec:gauge} we review
5190: results which indicate that these branch points arise because the
5191: trajectory in field space associated with this potential encounters
5192: the boundary of the region of Feynman-Siegel gauge validity. It seems
5193: almost to be a fortunate accident that the nonperturbative vacuum lies
5194: within the region of validity of this gauge choice. It is worth
5195: mentioning again here that in the BSFT approach, the tachyon potential
5196: can be computed exactly~\cite{BSFT}. In this formulation, there is no
5197: branch point in the effective potential, which is unbounded below for
5198: negative values of the tachyon. On the other hand, the nontrivial
5199: vacuum in the background-independent approach arises only as the
5200: tachyon field goes to infinity, so it is harder to study the physics
5201: of the stable vacuum from this point of view.
5202:
5203: Another interesting perspective on the tachyon effective potential is
5204: found by performing a perturbative (but off-shell)
5205: computation of the
5206: coefficients in the tachyon effective potential in the level-truncated theory.
5207: This gives a power series expansion of the effective potential
5208: \begin{eqnarray}
5209: V (\phi) & = & \sum_{n = 2}^{ \infty} c_n (\bar{\kappa} g)^{n-2} \phi^n
5210: \label{eq:v}\\
5211: & = & -\frac{1}{2}\phi^2 + (\bar{\kappa} g) \phi^3 + c_4 (\bar{\kappa} g)^2
5212: \phi^4 + c_5 (\bar{\kappa} g)^3 \phi^5 +\cdots\nonumber
5213: \end{eqnarray}
5214: The coefficients up to $c_{60}$ have been computed in the level
5215: truncations up to (10, 20)~\cite{Moeller-Taylor}. Because of the
5216: branch point singularity near $\phi = -0.1/g \bar{\kappa}$, this
5217: series has a radius of convergence much smaller than the value of
5218: $\phi$ at the nonperturbative vacuum. Thus, the energy at the stable
5219: vacuum lies outside the naive range of the potential defined by the
5220: perturbative expansion.
5221:
5222: Now, let us return to the problem of the overshoot in energy below
5223: $-T_{25}$ found at level 14 by Gaiotto and Rastelli\footnote{The
5224: material in the remainder of this section was developed only after the
5225: original TASI lectures in 2001, but is included because of its
5226: relevance to the main development in this section.}. The most
5227: straightforward way of determining whether or not this represents a
5228: real problem for string field theory would be to simply continue the
5229: calculation to higher levels. Unfortunately, at present this is not
5230: tractable, as the difficulty of computation grows exponentially in the
5231: level. Thus, we must resort to more indirect methods. It was found
5232: empirically by Taylor~\cite{WT-perturbative} that the level $L$
5233: approximations of string field theory give on-shell and off-shell
5234: amplitudes with error of order $1/L$. This work and further
5235: evidence~\cite{Coletti-Sigalov-Taylor,Beccaria-Rampino} indicates that
5236: amplitudes can be very accurately approximated by computing them in
5237: different level $L$ truncations, and matching to a power series in
5238: $1/L$. Such an approach can be taken to determine highly accurate
5239: values for the coefficients $c_n$ in (\ref{eq:v}). As noted above,
5240: the resulting power series has a finite radius of convergence, and the
5241: stable vacuum lies beyond this limit. There is a standard technique,
5242: however, known as the method of Pad\'e approximants, which allows one
5243: to extrapolate a function beyond its naive radius of convergence, if
5244: the function is sufficiently well-behaved in the direction in which it
5245: is extrapolated. The idea of Pad\'e approximants is to replace a
5246: power series having given coefficients for a fixed number of terms
5247: with a rational function with the same number of coefficients,
5248: choosing the coefficients of the rational function to give a power
5249: series which agrees with the fixed coefficients in the original power
5250: series. For example, consider the first three terms in the level $L
5251: =2 $
5252: approximation to the tachyon
5253: effective potential (\ref{eq:v}),
5254: \begin{equation}
5255: -\frac{1}{2}\phi^2 + \kappa g \phi^3-\frac{34}{27} (\kappa g) \phi^4 \,.
5256: \label{eq:truncated}
5257: \end{equation}
5258: This truncated expansion
5259: has no local minima, while the Pad\'e approximant
5260: \begin{equation}
5261: P^3_1 (\phi) =\frac{-\frac{1}{2}\phi^2 + \frac{10}{27} \kappa g \phi^3
5262: }{1 + \frac{34}{27} \kappa g \phi}
5263: % \label{eq:}
5264: \end{equation}
5265: does; this approximant thus represents a better description of the
5266: tachyon potential than the truncated expression (\ref{eq:truncated}).
5267: The advantage of Pad\'e approximants is that they allow one to
5268: incorporate poles into approximations of a function with a desired
5269: local power series behavior. For a wide class of functions,
5270: successive Pad\'e approximants converge exponentially quickly in the
5271: region where the function is smooth. Empirically, this seems to be
5272: the case for the tachyon effective potential. Thus, the energy
5273: minimum at any finite level of truncation can be determined to an
5274: arbitrary degree of accuracy from the leading coefficients in the
5275: potential. For example, the energy can be computed to 10 digits of
5276: accuracy by including approximately 40 coefficients $c_n$; this
5277: calculation is, however, highly sensitive to the accuracy of the
5278: coefficients~\cite{Ellis-Karliner-WT}.
5279:
5280:
5281: Combining Pad\'e approximants with approximations to the coefficients
5282: $c_n$, computed by matching level-truncated results in a $1/L$
5283: expansion, it is possible to predict not only the exact value of the
5284: energy at the stable minimum as $L \rightarrow \infty$, but also to
5285: predict the values of the approximate energy at intermediate values of
5286: $L$. Such a computation was performed using the level approximated
5287: values of $c_n$ up to level $ (10, 20)$~\cite{WT-Pade}. By first using
5288: these values to predict the level-approximated values at higher
5289: levels, and then inserting these values into Pad\'e approximants, the
5290: overshoot phenomenon found by Gaiotto and Rastelli was accurately
5291: reproduced. For example, compared to the value -1.0003678 found by
5292: these authors at level (16, 32), extrapolation from results at levels
5293: $(L, 2L)$ up to (10, 20) gives a predicted value of -1.0003773 at
5294: level (16, 32). Furthermore, the extrapolated values $E_L$ of the
5295: energy at the stable minimum were found to decrease up to
5296: approximately level 26, and then to increase, approaching an
5297: asymptotic value as $L \rightarrow \infty$ of $E_\infty \approx -1$
5298: with error $\sim 10^{-4}$. These results suggested that the energy at
5299: the minimum in the level-truncated theory takes the form shown in
5300: Figure~\ref{f:e-l}.
5301: %\begin{figure}
5302: %\epsfig{file=g1.eps,width=12cm}
5303: %\caption[x]{\footnotesize
5304: %Expected approximations to the vacuum energy at different
5305: % levels of truncation, extrapolated from data at lower levels of
5306: %truncation.}
5307: %\label{f:e-l}
5308: %\end{figure}
5309:
5310: \begin{figure}[!ht]
5311: \leavevmode
5312: \begin{center}
5313: \epsfxsize = 12 cm \epsfbox{g1.eps}
5314: \end{center}
5315: \caption[x]{\footnotesize
5316: Expected approximations to the vacuum energy at different
5317: levels of truncation, extrapolated from data at lower levels of
5318: truncation.}
5319: \label{f:e-l}
5320: \end{figure}
5321:
5322: In the calculation just described, there were two sources of error:
5323: 1) the coefficients $c_n$ had some numerical inaccuracy, and 2)
5324: there is some error introduced in extrapolating from low
5325: levels of truncation.
5326:
5327:
5328: This computation was improved by Gaiotto and
5329: Rastelli~\cite{gr-analysis}. These authors used a different approach:
5330: instead of extrapolating the finite $L$ results for the coefficients
5331: $c_n$, they extrapolated the nonperturbatively computed effective
5332: potential $V (\phi)$ at various values of $\phi$. Because Pad\'e
5333: approximants are so accurate, for exactly known values of $c_n$ and $V
5334: (\phi)$ this approach is equivalent to the combined
5335: Pad\'e-extrapolation in $c_n$ approach, but generally this approach
5336: trades inaccuracy in $c_n$ for inaccuracy in $V (\phi)$. In practice,
5337: it is much easier to compute the coefficients $c_n$ exactly than the
5338: nonperturbative effective potential $V (\phi)$, which
5339: requires numerically solving a large system of quadratic equations.
5340: Gaiotto and Rastelli were able, however, to
5341: use their results on $V (\phi)$ at higher levels, which greatly
5342: increased the accuracy of their extrapolations. They found that
5343: while level $(L, 2L)$ and $(L, 3L)$
5344: approximations tend to be very similar, extrapolations based on level
5345: $(L, 3L)$ truncations seem more robust. Using data up to
5346: level (16, 48) they found an extrapolated value of $E_\infty\approx
5347: -1.00003$, differing from $-1$ by an order of magnitude less than the
5348: value of the energy estimated at level 28, where the overshoot is
5349: predicted to be maximal. This gives compelling support to the
5350: conclusion that the level-truncated approximations to the energy
5351: indeed behave as shown in Figure~\ref{f:e-l}, and approach the value
5352: predicted by Sen as $L \rightarrow \infty$.
5353:
5354:
5355:
5356: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5357: % gauge fixing: WT
5358: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5359:
5360: \subsection{Gauge fixing}
5361: \label{sec:gauge}
5362:
5363: In this subsection we discuss some aspects of the Feynman-Siegel gauge
5364: choice used in most explicit calculations in OSFT to date. Let us
5365: restrict attention to the zero momentum action for even-level scalar
5366: fields. This action is invariant under (\ref{eq:SFT-gauge}) with a
5367: general gauge parameter of the form
5368: \begin{equation}\Lambda = \sum \mu^a | s_a \rangle
5369: = \mu_1 b_{-2} | 0_1 \rangle +
5370: \cdots \,.
5371: \end{equation}The ghost number zero states $| s_a \rangle$ are annihilated
5372: by $b_0$, so they do not contain
5373: $c_0$. The
5374: variation of a general zero-momentum scalar field takes the form
5375: \begin{equation}
5376: \delta \phi_i = D^{ia} \mu_a + g \bar{\kappa}\, T^{ija} \phi_j \mu_a \,.
5377: % \label{eq:}
5378: \end{equation}
5379: At $\phi_i = 0$, we have the linear
5380: variation $\delta \phi_i = D^{ia}
5381: \mu_a$. Let $\phi_q$ denote fields
5382: associated with ghost number one states that contain a $c_0$. For example, at
5383: level two there is a field $\eta$ associated with the state $c_0b_{-2} | 0_1
5384: \rangle$.
5385: At each level, the number of fields $\phi_q$ is clearly
5386: equal to the number of gauge parameters $\mu_a$; the
5387: corresponding states are simply related by removing
5388: or replacing the $c_0$.
5389: From
5390: the formula for $Q_B = c_0L_0 + \cdots$, it is easy to verify that
5391: $D^{qa}$ is a linear one-to-one map at each level, so
5392: \begin{equation}
5393: \det D^{qa} \neq 0
5394: % \label{eq:}
5395: \end{equation}
5396: holds at each level. This is why the Feynman-Siegel gauge, which sets
5397: $\phi_q=0$ at each level
5398: (and which limits us to gauge parameters
5399: associated with states without a $c_0$), is a good gauge choice near $\phi_i
5400: = 0$, as shown in subsection~\ref{sec:constraints-symmetries}.
5401:
5402: Let us now consider the gauge transformations at a general point in
5403: field space $\langle\phi_i\rangle$. We have
5404: \begin{equation}
5405: \delta \phi^i = M^{ia} \mu_a
5406: % \label{eq:}
5407: \end{equation}
5408: where
5409: \begin{equation}
5410: M^{ia} = D^{ia} + g \bar{\kappa} T^{ija} \langle\phi_j\rangle \,.
5411: % \label{eq:}
5412: \end{equation}
5413: Feynman-Siegel gauge breaks down whenever the determinant of this
5414: matrix vanishes
5415: \begin{equation}
5416: \det M^{qa} =0 \,.
5417: % \label{eq:}
5418: \end{equation}
5419: This condition defines a region in field space within which
5420: Feynman-Siegel gauge is valid. At the boundary of this region, some gauge
5421: transformations give field variations which are tangent to the
5422: Feynman-Siegel gauge-fixed hypersurface. Some gauge orbits which
5423: cross the Feynman-Siegel gauge surface inside this region will cross
5424: again outside the region, giving a form of Gribov ambiguity.
5425: Furthermore, some gauge orbits never encounter the region of gauge
5426: validity. Thus, Feynman-Siegel gauge is really only locally valid.
5427:
5428: We can study the region of Feynman-Siegel gauge validity in level
5429: truncation, using finite matrices $M^{qa}$. It is instructive to
5430: consider a simple example of the breakdown of this gauge choice.
5431: Consider dropping all fields other than the tachyon $\phi = \phi_1$
5432: and the field $\eta= \phi_4$. The gauge transformation rules then
5433: become
5434: \begin{eqnarray}
5435: \delta \phi & = & \mu g \bar{\kappa} \left[ -\frac{16}{9} \phi +
5436: \frac{128}{81} \eta \right] \\
5437: \delta \eta & = & -\mu + \mu g \bar{\kappa} \left[ -\frac{224}{81}
5438: \phi + \frac{1792}{729} \eta \right] \,.\nonumber
5439: \end{eqnarray}
5440: In this simple model, $M$ is a one-by-one matrix,
5441: \begin{equation}
5442: M = -\mu (1 + g \bar{\kappa} \frac{224}{ 81} \phi) \,.
5443: % \label{eq:}
5444: \end{equation}
5445: The gauge choice $\eta = 0$ breaks down when $\eta = \delta \eta = 0$
5446: which occurs when
5447: \begin{equation}
5448: \phi = -\frac{1}{g\bar{\kappa}} \frac{81}{224} \,.
5449: % \label{eq:}
5450: \end{equation}
5451: It is easy to see that smaller values of $\phi$ are gauge-equivalent
5452: to values of $\phi$ above this boundary value, while some gauge orbits
5453: never intersect the line $\eta = 0$.
5454:
5455: The complete action including all even level (zero momentum)
5456: scalar fields and gauge invariances has been computed up to level (8,
5457: 16)~\cite{Ellwood-Taylor-gauge}. One result of this computation is that the
5458: Feynman-Siegel gauge boundary condition $\det M^{qa} = 0$ seems to be very
5459: stable near the origin as the level of truncation is increased. This
5460: gives some
5461: confidence that there is a well-defined finite region in field space where
5462: Feynman-Siegel gauge is valid, and that the boundary of this region can be
5463: arbitrarily well approximated by level-truncation calculations. Another
5464: interesting result which can be seen from these calculations is that (to the
5465: precision possible in the level-truncated analysis) the branch points in the
5466: tachyon effective potential arise precisely at those points where the
5467: trajectory in field space associated with the effective potential
5468: crosses the Feynman-Siegel gauge boundary. Thus, these branch points
5469: are gauge artifacts. As mentioned previously, the tachyon effective
5470: potential computed from boundary string field theory does not suffer
5471: from such branch point problems.
5472:
5473: It would be very desirable, however,
5474: to have an approach which enables one to describe
5475: the full string field space, including configurations which do not have
5476: gauge representatives in the local region of Feynman-Siegel gauge
5477: validity. Other
5478: gauge choices can be made, but those which have been explored to date
5479: are only minor variations on the Feynman-Siegel gauge choice, and do
5480: not lead to qualitatively different results. One might have hoped to
5481: isolate the true vacuum without gauge fixing at all, given that level
5482: truncation breaks the gauge symmetry and thus allows a discrete set of
5483: solutions at any level. This approach, however, is not particularly promising:
5484: the solutions found at each level lie at very different
5485: places on the gauge orbit, and do not approach any natural limit.
5486: Nonetheless, it seems of paramount importance to find some method for
5487: exploring the full field space of the theory. Currently inaccessible
5488: regions of the field space may contain solutions that have not yet
5489: been found (see subsection~\ref{subsec:thebackgroundsosft}).
5490: %discussed in the conclusions, such a method is probably needed to find
5491: %multiple brane solutions in the theory defined around the vacuum
5492: %associated with a single brane.
5493:
5494:
5495:
5496: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5497: % solitons: BZ
5498: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5499:
5500: \subsection{Lower-dimensional D-branes as solitons}
5501: \label{sec:solitons}
5502:
5503: One aspect of the Sen conjectures (item (2) in the list of
5504: section~\ref{conjectures}),
5505: proposes that lower-dimensional D-branes can be viewed as solitons
5506: of the D25-brane string field theory.
5507: The solitons
5508: involve
5509: profiles for the tachyon field which arise because the tachyon
5510: potential is non-trivial.
5511: The tachyon solitons are lumps, as opposed to kinks, which appear
5512: in superstring field theory solitons.
5513:
5514:
5515:
5516:
5517: In this section we will discuss the basic ideas required to
5518: test this conjecture. We will follow the approach of
5519: M\"oeller, Sen, and Zwiebach~\cite{Moeller-sz} (other attempts~\cite{djmt}
5520: do not use level expansion). In
5521: order to be able to use a level expansion we curl up one spatial coordinate $x$
5522: into a circle of radius $R$ (the corresponding string coordinate is
5523: called $X$).
5524: We will work with $R>1$.
5525: Along this direction, we will wrap a D1-brane. We will then
5526: consider the possibility that a certain process of tachyon condensation
5527: results in
5528: the D1-brane becoming a D0-brane. Our use of D1- and D0-branes
5529: is just a matter of notational ease. Additional D-brane dimensions
5530: could be included.
5531:
5532: Recall that the mass of the D1-brane can be written in the form
5533: \begin{equation}
5534: \label{md1wvvwsmdo}
5535: M_{D1} = 2\pi R T_1 = {1\over 2 \pi^2 g^2}\,
5536: \end{equation}
5537: where $g$ is the coupling constant of the open string field theory
5538: that describes the D1-brane:
5539: \begin{equation}
5540: S = -{1\over g^2} \Bigl( {1\over 2} \langle \Phi, Q \Phi\rangle + {1\over 3}
5541: \langle \Phi, \Phi * \Phi \rangle \Bigr) \equiv -{1\over g^2 } \mathcal{V}
5542: (\Phi) \,.
5543: \end{equation}
5544: A few remarks are in order.
5545: In the above string action we have
5546: included into the string coupling factor $(1/g^2)$ the volume
5547: $(2\pi R)$ of the
5548: compact circle where the D1-brane is wrapped. By doing so, we can still use
5549: a CFT overlap with unit normalization, and the right-hand side
5550: in (\ref{md1wvvwsmdo}) gives the total mass of the brane. The zero
5551: string field here is supposed to describe the vacuum with a D1-brane stretched
5552: around the circle. For time-independent string fields (the kind of fields we
5553: consider here),
5554: $\mathcal{V}(\Phi)$ is a potential. More precisely the potential
5555: energy P.E. associated with a string field is
5556: \begin{equation}
5557: \hbox{P.E.} = - S(\Phi) = {1\over g^2} \mathcal{V} (\Phi)
5558: = (2\pi R T_1) \, 2\pi^2 \, \mathcal{V}(\Phi),
5559: \end{equation}
5560: where we used (\ref{md1wvvwsmdo}).
5561: This potential energy is really the potential energy of field configurations
5562: measured with respect to the D1-brane background. Therefore, the total
5563: energy $E_{tot}$ of the configuration is obtained by adding the energy
5564: of the D1-brane to the above P.E. We find
5565: \begin{equation}
5566: \label{totenergyfs}
5567: E_{tot}(\Phi) = (2\pi R T_1) \Bigl( 1 + 2\pi^2 \, \mathcal{V}(\Phi) \Bigr).
5568: \end{equation}
5569: Since we will use the level expansion to
5570: investigate if a D0-brane can be
5571: represented as a lump solution, it is reasonable to
5572: use the level expansion to calculate the mass of the D1-brane, as well.
5573: So, we re-express the energy of the D1-brane in (\ref{totenergyfs}) in terms of
5574: the string field potential at the vacuum. Let $\Phi
5575: = T_{vac}$ denote the string field of the D1-brane SFT that represents the
5576: tachyon vacuum. Then, we have
5577: $-1 = 2\pi^2 \mathcal{V}(T_{vac})$, and we can rewrite
5578: \begin{equation}
5579: \label{theenergycalcualtesjnvdkj}
5580: E_{tot}(\Phi) = (2\pi R T_1) \Bigl( 2\pi^2 \, \mathcal{V}(\Phi)-2\pi^2 \,
5581: \mathcal{V}(T_{vac}) \Bigr).
5582: \end{equation}
5583: Indeed, this formula works correctly: when $\Phi=0$ the total energy
5584: equals the mass of the D1-brane, and when $\Phi = T_{vac}$ the energy
5585: is zero (since the D1-brane has disappeared).
5586:
5587: Let $T_{lump}$ denote the lump (string field) solution, which
5588: is expected to represent the D0-brane in the field theory of the
5589: D1-brane.
5590: The energy of the lump solution is obtained from
5591: (\ref{theenergycalcualtesjnvdkj}) for $\Phi = T_{lump}$:
5592: \begin{equation}
5593: \label{theenerglumpcualtesjnvdkj}
5594: E_{lump} = E_{tot}(T_{lump} ) = (2\pi R T_1) \Bigl( 2\pi^2 \,
5595: \mathcal{V}(T_{lump})-2\pi^2
5596: \, \mathcal{V}(T_{vac}) \Bigr).
5597: \end{equation}
5598: The tensions
5599: $T_0$ and
5600: $T_1$ of the D0- and the D1-branes are related by $T_0 = 2\pi T_1$
5601: (the D0-brane tension is the D0-brane energy).
5602: We can therefore form the ratio $r(R) $ of the lump energy and the
5603: D0-brane energy
5604: \begin{equation}
5605: \label{predsecconjecture2}
5606: r (R) = {E_{lump}\over T_0} =
5607: R \Bigl( 2\pi^2 \,
5608: \mathcal{V}(T_{lump})-2\pi^2
5609: \, \mathcal{V}(T_{vac}) \Bigr).
5610: \end{equation}
5611: In the exact solution (or at infinite level), the ratio $r(R)$ should be equal
5612: to one. This is the content of the second tachyon conjecture. At
5613: any finite level
5614: $r(R)$ is some slowly varying function of $R$. Testing
5615: the conjecture for $R \to 1$ is quite difficult, and one must go
5616: to very high level in the computation. Testing the conjecture for
5617: $R$ very large is also laborious, since many terms enter into
5618: any finite-level expansion. So, in practice, one chooses some reasonable
5619: value of $R$ (the value $R=\sqrt{3}$ is convenient) and calculates
5620: to a fixed level.
5621:
5622: Before reviewing some of the results obtained, let's do
5623: the simplest computation explicitly.
5624: We consider a tachyon field $T(x)$ which is
5625: expanded as
5626: \begin{equation}
5627: T(x) = t_0 + \sum_{n=1}^\infty t_n \cos (nx/R) \,.
5628: \end{equation}
5629: The corresponding string field is written as
5630: \begin{eqnarray}
5631: |T\rangle &&= t_0 c_1 |0\rangle + \sum_{n=1}^\infty {1\over 2} t_n
5632: \Bigl( e^{inX(0)/R} + e^{-inX(0)/R} \Bigr) \, c_1|0\rangle \,, \nonumber\\
5633: &&= t_0 c_1 |0\rangle + \sum_{n=1}^\infty {1\over 2}\, t_n \Bigl(
5634: c_1|n/R\rangle + c_1|-n/R\rangle\Bigr) \,.
5635: \end{eqnarray}
5636: We now evaluate the string action, keeping $t_0$ and the first tachyon
5637: harmonic~$t_1$:
5638: \begin{equation}
5639: |T\rangle = t_0
5640: c_1 |0\rangle + {1\over 2}\, t_1 \Bigl( c_1|1/R\rangle +
5641: c_1|-1/R\rangle\Bigr)
5642: \,.
5643: \end{equation}
5644: Consider first the contribution of $t_1$ to the kinetic term
5645: \begin{eqnarray}
5646: {1\over 2} \langle T, QT\rangle \Bigl|_{t_1} &&=
5647: {1\over 2} {t_1\over 2} \cdot {t_1\over 2} \Bigl( \langle 1/R|
5648: + \langle -1/R|\Bigr) c_{-1} c_0 L_0 c_1 \Bigl( |1/R\rangle +|-1/R\rangle
5649: \Bigr) \nonumber\\
5650: &&= {1\over 4} t_1^2 \Bigl( -1 + {1\over R^2}\Bigr)\,.
5651: \end{eqnarray}
5652: Note that, as mentioned earlier, the overlaps have unit normalization.
5653: Let us now calculate the terms that arise from the interaction. Because
5654: of momentum conservation there are no
5655: $t_1^3 $ or $t_1 t_0^2$ terms. There is only a $t_1^2 t_0$ coupling, which is
5656: readily calculated as
5657: \begin{equation}
5658: {1\over 3} \cdot 3 \cdot {t_1\over 2} \cdot {t_1\over 2} t_0 \cdot
5659: 2\cdot \langle c_1 e^{iX/R} \,, ce^{-iX/R}, c \rangle = {1\over 2} t_0 t_1^2
5660: K^{3 - {2\over R^2}}\,.
5661: \end{equation}
5662: Let us explain the origin of the various factors. The first $1/3$ is the
5663: one that comes with the interaction term in the action. The factor
5664: of 3 is because there are three possible places to insert the operator
5665: associated with $t_0$. The factors of $t_1/2$ and $t_0$ come
5666: from the field expansion, and the factor of two arises because there
5667: are two ways in which the momentum can be conserved. The correlator
5668: has been evaluated in a way similar to the previous computation that led
5669: to (\ref{tachver}). Indeed, the only difference is that the conformal
5670: dimension of two of the operators has been shifted from $-1$ to
5671: $-1 + {1\over R^2}$.
5672:
5673: Collecting now our results and using the previously
5674: calculated potential for $t_0$ (\ref{eq:CFT-cubic-potential}) we find
5675: \begin{equation}
5676: \mathcal{V} (t_0, t_1) = -{1\over 2} t_0^2 - {1\over 4} \Bigl( 1- {1\over
5677: R^2}\Bigr) t_1^2 + {1\over 3} K^3 t_0^2 + {1\over 2} t_0 t_1^2
5678: K^{3 - {2\over R^2}}\,.
5679: \end{equation}
5680: The original tachyon is still there: it corresponds to the field $t_0$, which
5681: in the present expansion has no momentum. For $R>1$, the field $t_1$ is
5682: also a tachyon. This field is present because of the instability to
5683: form a D1-brane. Indeed, for $R>1$ the energy of the D1-brane is
5684: larger than the energy of the D0-brane, and the decay is possible.
5685: For $R<1$, the D0-brane has more energy than the D1-brane. In this
5686: case, it is
5687: not clear if some high level computation can exhibit the D0-brane
5688: as a solution of the D1-brane field theory.
5689: We return to this problem in the next subsection.
5690:
5691: Let's take $R=\sqrt{3}$. In this case the potential $\mathcal{V}(t_0, t_1)$
5692: has a critical point which represents a lump: $t_0 \simeq 0.18$ and $t_1 = -
5693: 0.34$. Of course there is also the conventional tachyon vacuum solution
5694: with $t_0 =1/K^3$ and $t_1 =0$. With these two solutions, one can
5695: readily compute the ratio $r(\sqrt{3})$ in (\ref{predsecconjecture2}).
5696: We find $r(\sqrt{3}) \simeq 0.774$ in this lowest order calculation.
5697: The result is certainly quite good. This computation is called a
5698: level (1/3; 2/3) computation since the highest level field $t_1$ has
5699: level $1/3 = 1/R^2$, and we kept terms in the potential up to level $2/3$.
5700: A computation at level (2,4) gives $r\simeq 1.02$, and
5701: for level (3,6) one finds $r \simeq 0.994$. The convergence to the
5702: answer is quite spectacular. This computation includes the tachyon
5703: harmonics $t_1$, $t_2$, and $t_3$, as well as fields from the second
5704: level and their first harmonics. No higher
5705: level computations have been done for this problem.
5706: The computations are not completely universal since the Virasoro
5707: structure of the state space depends on the radius of the circle.
5708: For rational values of $R$ one may find null states, so this is why
5709: we took $R$ irrational. Even for $R$ irrational, not all states
5710: can be written as Virasoro descendents of the vacuum $|0\rangle$.
5711: New primaries (and their descendents) are needed starting at
5712: level 4.
5713:
5714: Since we are equipped with the tachyon harmonics, one is able
5715: to construct explicitly the tachyon profile for the lump solution
5716: which represents the D0-brane. As the level is
5717: increased, the
5718: profile appears to settle into a well-defined limit. That same
5719: profile appears to arise for various values of the radius $R$ of
5720: the circle used for the computation. The profile is roughly of the
5721: form
5722: \begin{equation}
5723: T(x) \simeq a + b \, e^{-x^2/(2\sigma^2)} \,, \quad a \simeq 0.56, \quad
5724: b\simeq -0.83, \quad \sigma \simeq 1.52 \,.
5725: \end{equation}
5726: The $\sigma$ width of the lump is therefore about $1.5 \sqrt{\alpha'}$.
5727: The significance (or gauge independence) of this width is not clear.
5728: Nevertheless, it is interesting that D-branes, which are defined
5729: by definite positions in CFT, appear as thick objects in SFT. Physical
5730: questions regarding D-branes are expected to have identical answers in
5731: the two approaches.
5732:
5733: The above computations have been generalized to the case of
5734: lump solutions of codimension two. In this case, we can imagine a
5735: D2-brane wrapped on a torus $T^2$ which decays into a D0-brane.
5736: The results in the level expansion appear to confirm that the
5737: lump solutions do represent D0-branes. Less accurate results
5738: are obtained; the energy
5739: has only been estimated with about ten percent
5740: accuracy.
5741:
5742: \medskip
5743: The above results have simple analogs in field theory~\cite{Harvey-Kraus}.
5744: Consider a simple scalar field theory in $p+1$ spatial dimensions, where we
5745: single out a coordinate $x$ for special treatment:
5746: \begin{equation}
5747: S = \int dt d^py dx\Bigl\{ {1\over 2} \Bigl( {\partial \phi\over
5748: \partial t}\Bigr)^2 - {1\over 2} |\nabla_y \phi|^2 -
5749: {1\over 2} \Bigl( {\partial \phi\over
5750: \partial x}\Bigr)^2 - V(\phi)\Bigr\}\,.
5751: \end{equation}
5752: As you can readily verify, time-independent solitonic solutions $\phi(x)$,
5753: which depend only on the coordinate $x$, are
5754: obtained by solving the second-order ordinary differential equation
5755: \begin{equation}
5756: \label{eqnprofidlednkn}
5757: {d^2\phi\over dx^2} = V'(\phi(x))\,.
5758: \end{equation}
5759: This equation takes the form of the equation of motion of a unit
5760: mass particle in a one-dimensional potential $-V(x)$. As an example,
5761: we consider a theory with potential~\cite{Zwiebach-toy}
5762: \begin{equation}
5763: V(\phi) = {1\over 3} (\phi-1)^2 \left(\phi + {1\over 2}\right) \,.
5764: \end{equation}
5765: The potential has a maximum at $\phi=0$ and a local minimum
5766: at $\phi=1$. At $\phi=0$ the interpretation is that of a D$(p+1)$-brane
5767: with tension
5768: \begin{equation}
5769: T_{p+1} = V(\phi=0) = {1\over 6}\,.
5770: \end{equation}
5771: As a simple exercise, verify that
5772: \begin{equation}
5773: \phi (x) = 1 - {3\over 2}
5774: \, \hbox{sech}^2 (x/2) \,,
5775: \end{equation}
5776: is a lump solution for this potential.
5777:
5778: \noindent
5779: {\em Exercise}: Show that the lump solution is an object with
5780: tension $T_p = 6/5$.
5781:
5782:
5783: In string theory the ratio ${1\over 2\pi } {T_p\over T_{p+1}}$ is equal
5784: to one. In this field theory model with a cubic potential, we find
5785: \begin{equation}
5786: {1\over 2\pi } {T_p\over T_{p+1}} = {1\over 2\pi } {6\over 5} \cdot 6
5787: = {18\over
5788: 5\pi} \simeq 1.146\,.
5789: \end{equation}
5790: It is also a familiar result in soliton field theory that the spectrum
5791: of excitations that live on the world-volume of the lump solution
5792: $\bar \phi(x)$
5793: is governed by a Schr\"odinger equation with a potential $V'' (\bar \phi(x))$.
5794: The mass-squared values for the modes that live on the lump coincide
5795: with the Schr\"odinger energies.
5796:
5797:
5798: There has been some interest in finding potentials that
5799: accurately describe the behavior of the tachyon. While the kinetic terms
5800: are not standard, the potential
5801: \begin{equation}
5802: \label{pottachecfff}
5803: V(\phi) = -{1\over 4} \phi^2 \ln \phi^2 \,, \quad \phi >0\,.
5804: \end{equation}
5805: appears to be an exact effective tachyon potential. This potential
5806: was obtained~\cite{Minahan:2000ff} in an attempt to construct realistic
5807: tachyon potentials, and was later confirmed to appear in the BSFT approach to
5808: string field theory~\cite{BSFT}.
5809: The tachyon vacuum is at $\phi=0$, and surprisingly
5810: (but correctly!) the tachyon mass goes to infinity at this vacuum.
5811: This is consistent with the conjecture that perturbative open string
5812: degrees of freedom disappear at the tachyon vacuum.
5813:
5814: \noindent
5815: {\em Exercise}. Show that $\bar\phi(x) = \exp (-x^2/4)$ is the lump
5816: solution for the potential (\ref{pottachecfff}) and the Schr\"odinger
5817: potential for fluctuations on the lump solution is ${x^2\over 4} -
5818: {3\over 2}$, a
5819: simple harmonic oscillator potential. Finally, confirm that the values of
5820: $m^2$ for the particles that live on the lump are $-1, 0, 1, 2, \ldots$.
5821: This is the expected string spectrum!
5822:
5823: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5824: % backgrounds in open string theory: wati
5825: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5826:
5827: \subsection{Open string theory backgrounds}
5828: \label{subsec:thebackgroundsosft}
5829:
5830: We mentioned in the last subsection that when the radius $R$ of a
5831: circle on which a D1-brane is compactified becomes small, it is not
5832: known how to represent a D0-brane in the string field theory on the
5833: D1-brane. When $R < 1$, the energy of the resulting D0-brane is
5834: larger than the energy of the original D1-brane. Thus, such a
5835: solution would have positive energy with respect to the original
5836: system. The difficulty of constructing such a D0-brane solution is an
5837: example of a more general, and we believe crucial, question for OSFT:
5838: Does OSFT, either through level truncation or some more sophisticated
5839: analytic approach, admit classical solutions which describe open string
5840: backgrounds with higher energy than the configuration with respect to
5841: which the theory is originally defined? If OSFT is
5842: to be a truly complete
5843: formulation of string theory, such
5844: solutions must be possible, since all open string
5845: backgrounds must be accessible to the theory.
5846:
5847:
5848: Another problem of this type is to find, either analytically or
5849: numerically, a solution of the OSFT formulated with one D25-brane that
5850: describes {\it two} D25-branes.
5851: It should be just as feasible to go from a vacuum with one
5852: D-brane to a vacuum with two D-branes as it is to go from a
5853: vacuum with one D-brane to the empty vacuum.
5854: Despite some work on this problem~\cite{Ellwood-Taylor-2}, there is as
5855: yet no evidence of a solution.
5856: Several approaches which have
5857: been tried include: {\it i}) following a positive
5858: mass field upward, looking for a stable point; this method seems
5859: to fail because of gauge-fixing problems---the effective
5860: potential often develops a singularity before reaching the energy
5861: $+ T_{25}$, {\it ii}) following the intuition of the RSZ model
5862: (discussed in the following section)
5863: and constructing a gauge transform of the original D-brane
5864: solution which is $\star-$orthogonal to the original D-brane
5865: vacuum. It can be shown formally that such a state, when added
5866: to the original D-brane vacuum gives a new solution with the
5867: correct energy for a double D-brane; unfortunately, however, we
5868: have been unable to identify such a state numerically in level
5869: truncation.
5870:
5871: While so far no progress has been made towards the construction of
5872: solutions with higher energy than the initial vacuum, it is also
5873: interesting to consider the marginal case. An example of such a
5874: situation is embodied is the problem of translating a single D-brane
5875: of less than maximal dimension in a transverse direction. It was
5876: shown by Sen and Zwiebach~\cite{Sen-Zwiebach-translate} (in a T-dual
5877: picture) that after moving a D-brane a finite distance of order of the
5878: string length in a transverse direction, the level-truncated string
5879: field theory equations develop a singularity. Thus, in level
5880: truncation it does not seem possible to move a D-brane a macroscopic
5881: distance in a transverse direction\footnote{Although this can be done
5882: formally~\cite{Kluson}, it is unclear how the formal solution relates
5883: to an explicit expression in the oscillator language.}. In this case,
5884: a toy model~\cite{Zwiebach-toy} suggests that the problem is that the
5885: infinitesimal marginal parameter for the brane translation ceases to
5886: parameterize the marginal trajectory in field space after a finite
5887: distance, just as the coordinate $x$ ceases to parameterize the circle
5888: $x^2 + y^2 = 1$ near $x = 1$. Indeed, an explicit
5889: calculation~\cite{Coletti-Sigalov-Taylor} of the
5890: field redefinition needed to take the
5891: OSFT field $A$ associated with the transverse motion to the correct marginal
5892: parameter $a$ shows that this field redefinition has a subleading term
5893: \begin{equation}
5894: A = a + \alpha a^3 + \cdots,
5895: % \label{eq:}
5896: \end{equation}
5897: where $\alpha < 0$. Thus, as $a$ increases, eventually a point is
5898: reached where $A$ begins to decrease. This shows that $A$ is not a
5899: good parameter for marginal deformations of arbitrary size.
5900: It would be nice to have a clear understanding of how arbitrary
5901: marginal deformations are encoded in the theory.
5902:
5903: To show that open string field theory is sufficiently general to
5904: address arbitrary questions involving different vacua, it is clearly
5905: necessary to show that the formalism is powerful enough to describe
5906: multiple brane vacua, the D0-brane lump on an arbitrary radius circle,
5907: and translated brane vacua. It is currently unclear whether the
5908: obstacles to finding these vacua are technical or conceptual. It may
5909: be that the level-truncation approach is not well-suited to finding
5910: these vacua, and a new approach is needed.
5911:
5912:
5913:
5914: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5915: % VSFT: both
5916: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5917:
5918: \section{String field theory around the stable vacuum}
5919: \label{sec:VSFT}
5920:
5921:
5922: The tachyon conjectures state that the classically stable
5923: vacuum is the closed string vacuum. This
5924: implies that there
5925: should be no open string excitations in this vacuum,
5926: given that the D-brane represented by the original
5927: OSFT has decayed and exists no more. Without a D-brane
5928: conventional perturbative open string states are not expected
5929: to exist.
5930: If any perturbative states exist in this vacuum, they should be closed
5931: string states, which are only expected to appear in the quantum open
5932: string field theory.
5933:
5934: There are two natural questions concerning this conjecture. First, we
5935: ask: Can it be tested? For this, we can begin with the original OSFT
5936: on the background of a D25-brane, for example, and use the (numerical)
5937: solution $\Phi_0$ for the tachyon vacuum to expand the classical
5938: OSFT around the tachyon vacuum and to calculate the
5939: spectrum. The conjecture requires that no physical states be
5940: encountered. Second, we ask: Is
5941: there a more natural formulation of open string theory around the tachyon
5942: vacuum, in which, for example, the background independence of the
5943: theory might be more manifest?
5944: The theory around the tachyon vacuum,
5945: is, no doubt, rather unusual. In the tachyon vacuum there
5946: are no apparent physical states, at least none that take any familiar
5947: form. Physical perturbative states can arise only
5948: from quantum
5949: effects or classically after the theory is shifted to a nontrivial
5950: background that represents some D-brane configuration.
5951:
5952:
5953:
5954: The tachyon vacuum is a rather special
5955: vacuum: it is the end product of the decay of {\em any} D-brane
5956: configuration. Presumably, the theory at the tachyon vacuum is
5957: independent of the particular version of OSFT used to reach it
5958: upon tachyon condensation,
5959: in the sense that the string field theories associated with different
5960: D-brane configurations should be equivalent under field redefinition
5961: around the stable vacuum of each theory.
5962: If that is the case, there
5963: may exist a theory -- which we can call {\em Vacuum String Field Theory}, or
5964: VSFT -- which formulates the physics of the tachyon vacuum directly,
5965: {\em without} using a D-brane background to reach the tachyon vacuum.
5966:
5967:
5968: Presently, there is background
5969: dependence in the formulation of Witten's OSFT; some specific
5970: D-brane background must
5971: be chosen to define the theory, even though this D-brane configuration may be
5972: removed through tachyon condensation. As a result, even if the theory
5973: is in an abstract sense completely background independent, we are
5974: stuck with some particular choice of ``coordinates'' on the theory
5975: arising from the original choice of background, which may make physics in other
5976: backgrounds rather difficult to disentangle.
5977: The tachyon vacuum is
5978: also a specific background, but it is certainly a choice that is more
5979: canonical than one which picks one out of an infinite number of
5980: D-brane configurations. There are perhaps two canonical choices: an
5981: infinite number of space-filling D-branes, which has been motivated
5982: from the viewpoint of K-theory~\cite{k-theory}, and a background with
5983: no D-branes whatsoever -- the tachyon vacuum. In this section we
5984: investigate the second choice.
5985:
5986: A strikingly simple formulation
5987: of VSFT was proposed by Rastelli, Sen, and
5988: Zwiebach
5989: (RSZ)~\cite{RSZ}, in which the BRST operator is taken to be purely
5990: contained in the ghost sector.
5991: In this theory, closed-form analytic solutions that represent
5992: D-branes can be found and
5993: take the form of projectors of the star-algebra. One shortcoming of this VSFT
5994: is that certain computations are singular and require regularization.
5995: It remains to be seen if a regular VSFT exists.
5996:
5997: In subsection \ref{sec:vacuum-theory} we describe the form of the OSFT
5998: action when expanded around the classically stable tachyon vacuum.
5999: Subsection \ref{sec:decoupling} describes evidence from Witten's OSFT
6000: that the open
6001: string degrees of freedom truly disappear from the theory in this vacuum.
6002: In \ref{sec:RSZ-model} we introduce and discuss the RSZ model of
6003: VSFT. Subsection \ref{subsec:slivers} describes an important class of
6004: states in the star algebra: slivers and projectors, which play a key
6005: role in constructing D-branes in the RSZ model, and which may also be
6006: useful in understanding solutions of the Witten theory. Finally, in
6007: \ref{subsec:closedstringsinosft} we discuss closed strings in OSFT.
6008:
6009:
6010:
6011: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6012: % Vacuum SFT: BZ
6013: % discussion of CSFT around vacuum \Phi_0, with BRST operator \tilde{Q}
6014: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6015:
6016: \subsection{String field theory in the true vacuum}
6017: \label{sec:vacuum-theory}
6018:
6019:
6020: We have seen that numerical results from level-truncated string field
6021: theory strongly suggest the existence of a classically stable vacuum solution
6022: $\Phi_0$ to the string field theory equation of motion. The state
6023: $\Phi_0$, while
6024: still unknown analytically, has been determined numerically to a high degree of
6025: precision. This state seems like a very well-behaved string field
6026: configuration. While there is no positive-definite inner product on
6027: the string field Fock space, the state $\Phi_0$ certainly has finite
6028: norm under the natural inner product $\langle V_2| \Phi_0, c_0L_0
6029: \Phi_0\rangle$, and is even better behaved under the product
6030: $\langle V_2| \Phi_0, c_0
6031: \Phi_0\rangle$. Thus, it is natural to assume that
6032: $\Phi_0$ defines a
6033: classically stable vacuum for the theory, around which we can expand
6034: the action to find a string field theory around the tachyon vacuum.
6035:
6036:
6037:
6038: \medskip
6039: Let $\Phi_0$ be the string field
6040: configuration describing the tachyon vacuum.
6041: This string field satisfies the classical field equation
6042: \begin{equation}\label{feq}
6043: Q \Phi_0 + \Phi_0 * \Phi_0 = 0 \,.
6044: \end{equation}If $\widetilde\Phi=\Phi -
6045: \Phi_0$ denotes the shifted open string field, then
6046: the cubic string field theory action (\ref{e1}) expanded around the tachyon
6047: vacuum has the form:
6048: \begin{equation}\label{e2}
6049: S (\Phi_0 + \widetilde\Phi) = S(\Phi_0) \,-\, {1\over g^2}\,\,\bigg[\, {1\over
6050: 2} \langle
6051: \,\widetilde\Phi
6052: \,,\, \widetilde Q\,
6053: \widetilde\Phi
6054: \rangle + {1\over 3}\langle \,\widetilde\Phi \,,\, \widetilde\Phi *
6055: \widetilde\Phi \rangle \bigg] \,.
6056: \end{equation}Here $S(\Phi_0)$ is a constant,
6057: which according to the energetics
6058: part of the tachyon conjectures equals the tension
6059: of the D-brane times its volume
6060: (as before, we assume that the time
6061: interval has unit length so that the action can be identified with the
6062: negative of the potential energy for static configurations).
6063: The kinetic operator $\widetilde Q$ is given in terms
6064: of $Q$ and $\Phi_0$ as:
6065: \begin{equation}\label{e3}
6066: \widetilde Q \widetilde\Phi = Q \widetilde\Phi + \Phi_0 * \widetilde\Phi +
6067: \widetilde\Phi* \Phi_0\, .
6068: \end{equation}More generally, on arbitrary string fields one would define
6069: \begin{equation}\label{e3p}
6070: \widetilde Q A = Q A + \Phi_0 * A - (-1)^{A} A * \Phi_0\, .
6071: \end{equation}The consistency of the new action (\ref{e2})
6072: is guaranteed from the
6073: consistency of the action in (\ref{e1}). Since neither the inner
6074: product nor the star multiplication have changed,
6075: the identities
6076: in (\ref{l1e2}) still hold. One can also check that the identities
6077: in (\ref{l1e1}) hold when $Q$ is replaced by $\widetilde Q$.
6078: Just as the original action is invariant under the gauge transformations
6079: (\ref{l1e5}), the new action is invariant under
6080: $\delta \widetilde\Phi =\widetilde Q \Lambda + \widetilde\Phi * \Lambda -
6081: \Lambda * \widetilde\Phi$ for any Grassmann-even ghost-number zero state
6082: $\Lambda$.
6083:
6084: Since the energy density of the brane represents
6085: a positive cosmological constant,
6086: it is natural to add the constant $-M=-S(\Phi_0)$ to
6087: (\ref{e1}). This will cancel the $S(\Phi_0)$ term in
6088: (\ref{e2}), and will make manifest
6089: the
6090: expected zero energy density in the final vacuum without D-brane. For the
6091: analysis around this final vacuum it suffices therefore to
6092: study the action
6093: \begin{equation}\label{e2p}
6094: S_0 (\widetilde\Phi) \equiv \,-\, {1\over g^2}\,\,\bigg[\, {1\over 2} \langle
6095: \,\widetilde\Phi
6096: \,,\, \widetilde Q\,
6097: \widetilde\Phi
6098: \rangle + {1\over 3}\langle \,\widetilde\Phi \,,\, \widetilde\Phi *
6099: \widetilde\Phi \rangle \bigg] \,.
6100: \end{equation}This string field theory around the stable vacuum has
6101: precisely the
6102: same form as Witten's original cubic string field theory, only with a
6103: different BRST operator $\widetilde Q$, which so far is only determined
6104: numerically. While this is insufficient for a complete formulation,
6105: it suffices to test the conjecture that open string excitations disappear
6106: in the tachyon vacuum, as we will discuss in Section~\ref{sec:decoupling}
6107:
6108: \medskip
6109: The numerical solution for $\Phi_0$ provides a numerical definition
6110: of the string field theory around the tachyon vacuum. How could we
6111: do better? If we had a closed form solution $\Phi_0$ available, the problem
6112: of formulating SFT around the tachyon vacuum would be significantly
6113: simplified. It is not clear, however, that the resulting formulation
6114: would be the best possible one.
6115: Previous experience with background deformations (small and
6116: large) in SFT indicates that even if we knew $\Phi_0$ explicitly and
6117: constructed $S_0(\widetilde\Phi)$ using eq.(\ref{e2p}), this may not
6118: be the most
6119: convenient form of the action.
6120: Typically a nontrivial field
6121: redefinition is
6122: necessary to bring the shifted SFT action to the canonical form
6123: representing the new background~\cite{Sen:1993mh}.
6124: In fact, in some cases, such as
6125: in the formulation of open SFT for D-branes with various values of
6126: magnetic fields, it is possible to formulate the various SFT's
6127: directly~\cite{Kawano:1999fw,Srednicki},
6128: but the nontrivial classical solution relating
6129: theories
6130: with different magnetic fields are not known. This suggests that if a
6131: simple form exists for the SFT action around the tachyon vacuum it might
6132: be easier to guess it than to derive it.
6133:
6134:
6135: In fact, this is exactly the approach to the formulation of vacuum
6136: string field
6137: theory (VSFT) taken by Rastelli, Sen, and Zwiebach (RSZ)~\cite{RSZ}. These
6138: authors postulate that at the tachyon vacuum the
6139: action takes the form
6140: \begin{equation}\label{e2findpxx}
6141: \mathcal{S} (\Phi) \equiv \,-\, K_0\,\,\bigg[\, {1\over 2} \langle
6142: \,\Phi \,,\, \mathcal{Q}\, \Phi
6143: \rangle + {1\over 3}\langle \,\Phi \,,\, \Phi *
6144: \Phi \rangle \bigg] \,,
6145: \end{equation}where the new kinetic operator
6146: $\mathcal{Q}$ is an operator build solely out of ghosts fields.
6147: If this gives a consistent theory at the tachyon vacuum, they argue,
6148: their choice of $\mathcal{Q}$ must be field redefinition equivalent
6149: to the $\widetilde Q$ that arises directly by shifting the original OSFT action
6150: with the tachyon solution $\Phi_0$.
6151: We discuss the RSZ model in section~\ref{sec:RSZ-model}.
6152:
6153:
6154:
6155:
6156:
6157:
6158: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6159: % Decoupling of open strings: WT
6160: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6161:
6162: \subsection{Decoupling of open strings}
6163: \label{sec:decoupling}
6164:
6165: It may seem surprising to imagine that {\it all} the perturbative open
6166: string degrees of freedom will vanish at a particular point in field
6167: space, since these are all the
6168: apparent
6169: degrees of freedom available in the
6170: theory. This is not a familiar phenomenon from quantum field theory. To
6171: understand how the open strings can decouple, it may be helpful to
6172: begin by considering the simple example of the (0, 0) level-truncated
6173: theory. In this theory, the quadratic terms in the action become
6174: \begin{equation}
6175: -\int d^{26} p \;
6176: \phi (-p) \left[ \frac{p^2 -1}{2} + g \bar{\kappa} \left(
6177: \frac{16}{27} \right)^{p^2} \cdot 3 \langle \phi \rangle \right]
6178: \phi (p) \,.
6179: % % \label{eq:}
6180: \end{equation}
6181: Taking
6182: $\langle \phi \rangle = 1/3 \bar{\kappa} g$, we find that the
6183: quadratic term is a transcendental expression which does not vanish
6184: for any real value of $p^2$. Thus, this theory has no poles, and the
6185: tachyon has decoupled from the theory. Of course, this is not the
6186: full story, as there are still finite complex poles. It does, however
6187: suggest a mechanism by which the nonlocal parts of the action (encoded
6188: in the exponential of $p^2$) can remove physical poles.
6189:
6190: To get the full story, it is necessary to continue the analysis to
6191: higher level. At level 2, there are 7 scalar fields,
6192: the tachyon and the 6 fields associated with the Fock space states
6193: \begin{eqnarray}
6194: (\alpha_{-1} \cdot \alpha_{-1})| 0_1, p \rangle & &
6195: b_{-1} \cdot c_{-1}| 0_1, p \rangle \nonumber\\
6196: c_{0} \cdot b_{-1}| 0_1, p \rangle & &
6197: (p \cdot \alpha_{-2})| 0_1, p \rangle \\
6198: (p \cdot \alpha_{-1})^2| 0_1, p \rangle & &
6199: (p \cdot \alpha_{-1}) c_0b_1| 0_1, p \rangle \nonumber
6200: \end{eqnarray}
6201: Note that in this analysis we cannot fix Feynman-Siegel gauge, as we
6202: only believe that this gauge is valid for the zero-modes of the scalar
6203: fields in the vacuum $\Psi_0$. An attempt at analyzing the spectrum
6204: of the theory in Feynman-Siegel gauge using level truncation has been
6205: made~\cite{ks-open}, but gave no sensible results\footnote{Note added:
6206: Recently, Giusto and Imbimbo have carried out a more detailed analysis
6207: of the spectrum around the stable vacuum in Feynman-Siegel gauge
6208: \cite{Giusto-Imbimbo}. Their more careful analysis shows that
6209: spurious poles found in \cite{ks-open} are cancelled when the
6210: truncation level is sufficiently high. This approach gives a nice
6211: confirmation of the results of \cite{Ellwood-Taylor-spectrum}, while
6212: working in a fixed gauge, and extends these results by having
6213: sensitivity to cohomology associated with states which are closed for
6214: all momentum, but not exact at discrete values of momentum (type A
6215: states in the notation of \cite{Giusto-Imbimbo}).}. Diagonalizing the
6216: quadratic term in the action on the full set of 7 fields of level
6217: $\leq 2$, we find~\cite{Ellwood-Taylor-spectrum} that poles develop at
6218: $M^2 = 0.9$ and $M^2 = 2.0$ (in string units, where the tachyon has
6219: $M^2 = -1$). These poles correspond to states satisfying $\tilde{Q}
6220: \tilde{\Psi} = 0$. The question now is, are these states physical?
6221: If they are exact states, of the form $\tilde{\Psi} = \tilde{Q}
6222: \tilde{\Lambda}$, then they are simply gauge degrees of freedom. If
6223: not, however, then they are states in the cohomology of $\tilde{Q}$
6224: and should be associated with physical degrees of freedom.
6225: Unfortunately, we cannot precisely determine whether the poles we find
6226: in level truncation are due to exact states, as the level-truncation
6227: procedure breaks the condition $\tilde{Q}^2 = 0$. Thus, we can only
6228: measure {\it approximately} whether a state is exact. A detailed
6229: analysis of this question was carried out by Ellwood and
6230: Taylor~\cite{Ellwood-Taylor-spectrum}. In their paper, all terms in
6231: the SFT action of the form $\phi_i \; \psi_j (p) \; \psi_k (-p)$ were
6232: determined, where $\phi_i$ is a scalar zero-mode, and $\psi_{j, k}$
6233: are nonzero-momentum scalars. In addition, all gauge transformations
6234: involving at least one zero-momentum field were computed up to level
6235: (6, 12). At each level up to $L = 6$, the ghost number 1 states in
6236: the kernel ${\rm Ker} \;\tilde{Q}^{(1)}_{(L, 2L)}$ were computed. The
6237: extent to which each of these states lies in the exact subspace was
6238: measured using the formula
6239: \begin{equation}
6240: \% \;{\rm exactness} = \sum_{i}\frac{(s \cdot e_i)^2}{ (s \cdot s)}
6241: % % \label{eq:}
6242: \end{equation}
6243: where $\{e_i\}$ are an orthonormal basis for ${\rm Im} \;
6244: \tilde{Q}^{(0)}_{ (L, 2L)}$, the image of $\tilde{Q}$ acting on the
6245: space of ghost number 0 states in the appropriate level truncation.
6246: (Note that this measure involves a choice of inner product on the Fock
6247: space; several natural inner products were tried, giving roughly
6248: equivalent results). The result of this analysis was that up to the
6249: mass scale of the level truncation, $M^2 \leq L -1$, all the states in
6250: the kernel of $\tilde{Q}^{(1)}$ were $\geq 99.9\%$ within the exact
6251: subspace, for $L \geq 4$. This result seems to give very strong
6252: evidence for Sen's third conjecture that there are no perturbative
6253: open string excitations around the stable classical vacuum $\Psi_0$.
6254: This analysis was only carried out for even level scalar fields; it
6255: would be nice to check that a similar result holds for odd-level
6256: fields and for tensor fields of arbitrary rank.
6257:
6258: Another more abstract argument that there are no open string states in
6259: the stable vacuum was given by Ellwood, Feng, He and Moeller~\cite{efhm}.
6260: These authors argued that in the stable vacuum, the identity state $|
6261: I \rangle$
6262: in the SFT star algebra, which satisfies
6263: $ I \star A = A$ for a very general class of string fields
6264: $A$, seems to be an exact state,
6265: \begin{eqnarray}
6266: | I \rangle
6267: & = & \tilde{Q} | \Lambda
6268: \rangle\,. \label{eq:i-exact}
6269: \end{eqnarray}
6270: If indeed the identity is exact, then it follows
6271: immediately that the cohomology of $\tilde{Q}$ is empty, since
6272: $\tilde{Q}A = 0$ then implies that
6273: \begin{eqnarray}
6274: A = (\tilde{Q} \Lambda) \star A
6275: = \tilde{Q} (\Lambda \star A) -\Lambda \star \tilde{Q} A
6276: = \tilde{Q} (\Lambda \star A) \,.
6277: \end{eqnarray}
6278: So to prove that the cohomology of $\tilde{Q}$ is trivial, it suffices
6279: to show that $\tilde{Q} | \Lambda \rangle = | I \rangle$.
6280: While there are some subtleties involved with the identity
6281: string field, Ellwood {\it et al.} found a very elegant expression for
6282: this field,
6283: \begin{eqnarray}
6284: | I \rangle & = &
6285: \left( \cdots e^{\frac{1}{8} L_{-16}} e^{\frac{1}{4} L_{-8}}
6286: e^{\frac{1}{2} L_{-4}}\right)
6287: e^{L_{-2}} | 0 \rangle\,.
6288: \end{eqnarray}
6289: (Recall that $| 0 \rangle= b_{-1}| 0_1 \rangle$.)
6290: They then looked numerically for a state $| \Lambda \rangle$
6291: satisfying (\ref{eq:i-exact}). For example, truncating at level $L = 3$,
6292: \begin{eqnarray}
6293: | I \rangle & = & | 0 \rangle + L_{-2}| 0 \rangle
6294: + \cdots\label{eq:identity-2}\\
6295: & = & | 0 \rangle-b_{-3} c_{1}| 0 \rangle -2b_{-2}c_0| 0 \rangle
6296: +\frac{1}{2} (\alpha_{-1} \cdot \alpha_{-1})| 0 \rangle
6297: + \cdots\nonumber
6298: \end{eqnarray}
6299: while the only candidate for $|\Lambda\rangle$ is
6300: \begin{equation}
6301: | \Lambda \rangle = \alpha \; b_{-2}| 0 \rangle,
6302: % % \label{eq:}
6303: \end{equation}
6304: for some constant $\alpha$. The authors of ~\cite{efhm} showed that
6305: the state (\ref{eq:identity-2}) is best approximated as exact when
6306: $\alpha \sim 1.12$; for this value, their measure of exactness becomes
6307: \begin{equation}
6308: \frac{\left| \tilde{ Q} |\Lambda \rangle -| I \rangle \right
6309: |}{| I |} \rightarrow 0.17,
6310: % % \label{eq:}
6311: \end{equation}
6312: which the authors interpreted as a $17\%$ deviation from exactness.
6313: Generalizing this analysis to higher levels, they found at levels 5,
6314: 7, and 9, a deviation from exactness of $11\%, 4.5\%$ and $3.5\%$
6315: respectively. At level 9, for example, the identity field has 118
6316: components, and there are only 43 gauge parameters, so this is a
6317: highly nontrivial check on the exactness of the identity. Like the
6318: results of Ellwood and Taylor~\cite{Ellwood-Taylor-spectrum}, these
6319: results strongly support the conclusion that the cohomology of the
6320: theory is trivial in the stable vacuum. In this case, the result
6321: applies to fields of all spins and all ghost numbers.
6322:
6323: Given that the Witten string field theory seems to have a classical
6324: solution with no perturbative open string excitations, in accordance
6325: with Sen's conjectures, it is quite interesting to ask what the
6326: physics of the string field theory in the
6327: stable vacuum
6328: should describe. One natural assumption might be that this theory
6329: should include closed string states in its quantum spectrum.
6330: Unfortunately, addressing this question requires performing
6331: calculations in the quantum theory around the stable vacuum. Such
6332: calculations are quite difficult (although progress in this direction
6333: has been made by Minahan in the $p$-adic version of the theory~\cite{Minahan}).
6334: Even in the perturbative vacuum, it is difficult to systematically
6335: study closed strings in the quantum string from theory. We discuss
6336: this question again in the final subsection of this section.
6337:
6338: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6339: % ghost Ansatz: BZ
6340: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6341:
6342: \subsection{Pure ghost Vacuum String Field Theory}
6343: \label{sec:RSZ-model}
6344:
6345:
6346: Our discussion in Section~\ref{sec:vacuum-theory} suggests that
6347: a VSFT may be formulated as a cubic string field theory, with
6348: some new choice $\mathcal{Q}$ for the kinetic operator.
6349: The choice of $\mathcal{Q}$ will be required to satisfy the following
6350: properties:
6351: \begin{itemize}
6352:
6353: \item
6354: The operator $\mathcal{Q}$ must be of ghost number one
6355: and must satisfy the conditions (\ref{l1e1})
6356: that guarantee gauge invariance of the string action.
6357:
6358: \item The operator $\mathcal{Q}$ must have vanishing cohomology.
6359:
6360: \item The operator $\mathcal{Q}$ must be universal, namely, it must
6361: be possible to
6362: write without reference to the brane boundary conformal field theory.
6363:
6364: \end{itemize}
6365:
6366: The first condition is unavoidable; the theory must be gauge
6367: invariant if it is to be consistent. The second condition is
6368: reasonable, but perhaps stronger than needed: all we probably
6369: know is that there should be no cohomology at ghost number one,
6370: which is the ghost number at which physical states appear.
6371: The third constraint is the most stringent one. It
6372: implies that VSFT is an intrinsic theory that can be formulated
6373: without using an auxiliary D-brane.
6374:
6375: \smallskip
6376: The simplest possible choice is $\mathcal{Q}=0$, which
6377: gives the
6378: purely cubic version of open string field theory~\cite{CUBIC}. Indeed, it
6379: has
6380: long been tempting to identify the tachyon vacuum with a theory where
6381: the kinetic operator vanishes because, lacking the kinetic term,
6382: the string field gauge symmetries are not spontaneously broken.
6383: Nevertheless, there are well-known complications with this identification.
6384: The string field shift $\bar
6385: \Phi$ that relates the cubic to the purely cubic OSFT
6386: appears to satisfy $Q \bar\Phi=0$ as well as $\bar\Phi*\bar
6387: \Phi=0$. The
6388: tachyon condensate definitely does not satisfy these two identities.
6389: We therefore search
6390: for nonzero $\mathcal{Q}$.
6391:
6392:
6393:
6394:
6395:
6396: We can satisfy the three requirements by letting
6397: ${\mathcal{Q}}$ be constructed purely from ghost operators. In
6398: particular we claim
6399: that the ghost number one operators
6400: \begin{equation}\label{cn}
6401: {\mathcal{C}}_n \equiv c_n + (-)^n \, c_{-n} \,, \quad n=0,1,2,\cdots
6402: \end{equation}satisfy the properties
6403: \begin{eqnarray} \label{ecp}
6404: && {\mathcal{C}}_n {\mathcal{C}}_n = 0, \nonumber \\
6405: &&{\mathcal{C}}_n (A * B) = ({\mathcal{C}}_n A) * B + (-1)^{A} A *
6406: ({\mathcal{C}}_n B)\,, \\
6407: && \langle \, {\mathcal{C}}_n A , B \,\rangle = - (-)^A \langle A ,
6408: {\mathcal{C}}_n B \rangle
6409: \,.\nonumber
6410: \end{eqnarray}
6411: The first
6412: property is manifest.
6413: The last property
6414: follows
6415: because under BPZ conjugation $c_n \to (-)^{n+1} c_{-n}$.
6416: The second property
6417: follows from the
6418: conservation laws~\cite{Rastelli-Zwiebach}
6419: \begin{equation}\langle V_3| ({\mathcal{C}}_n^{(1)} + {\mathcal{C}}_n^{(2)} +
6420: {\mathcal{C}}_n^{(3)} ) = 0\,.
6421: \end{equation}These conservation laws arise by consideration of
6422: integrals of the form
6423: $\int dz c(z) \varphi(z)$ where $\varphi(z) (dz)^2$ is a globally
6424: defined quadratic differential.
6425:
6426: Each of the operators ${\mathcal{C}}_n$ has
6427: vanishing cohomology. To see this note that for each
6428: $n$ the operator $B_n ={1\over 2} (b_n + (-)^n \, b_{-n}) $ satisfies
6429: $\{ {\mathcal{C}}_n , B_n \} = 1$. It then
6430: follows that whenever ${\mathcal{C}}_n \psi=0$, we
6431: have $\psi = \{ {\mathcal{C}}_n , B_n \} \psi = {\mathcal{C}}_n (
6432: B_n \psi)$, showing that
6433: $\psi$ is ${\mathcal{C}}_n$ trivial.
6434: Since they are built from ghost oscillators, all ${\mathcal{C}}_n$'s
6435: are manifestly universal.
6436:
6437:
6438: It is clear from the structure of the consistency conditions that we can take
6439: $\mathcal{Q} = \sum_{n=0}^\infty a_n \, {\mathcal{C}}_n$, where the
6440: $a_n$'s are constant
6441: coefficients. As we will see below, many properties of the RSZ theory
6442: follow simply from the fact that $\mathcal{Q}$ is pure ghost. But,
6443: there are some
6444: computations that may require a choice of $\mathcal{Q}$ (more on this
6445: later). The
6446: work of Hata and Kawano~\cite{Hata-Kawano} gave the clue for the choice of
6447: $\mathcal{Q}$ taken by RSZ:
6448: \begin{eqnarray} \label{ep1}
6449: \mathcal{Q} &=&{1\over 2i} (c(i) - \bar c(i)) = {1\over 2i} (c(i)
6450: - c(-i)) =
6451: \sum_{n=0}^\infty
6452: (-1)^n {\mathcal{C}}_{2n}\, , \nonumber\\
6453: &=& c_0 - (c_2 + c_{-2}) + (c_4 + c_{-4}) - \cdots \,.
6454: \end{eqnarray}
6455: Since the canonical zero-time open string in the complex $z$-plane is the
6456: half-circle $|z|=1$ that lies on the upper half plane, the operator
6457: $c(i)$ represents
6458: a ghost insertion precisely at the open string midpoint. This is the
6459: most delicate
6460: point on the open string given that the three string interaction is
6461: a world-sheet
6462: with a curvature singularity at the point where the three string
6463: midpoints meet.
6464: The other
6465: operator
6466: $c(-i)$ is needed in order that
6467: $\mathcal{Q}$ is twist invariant (see the first equation in (\ref{l1e4})).
6468: With this choice of
6469: ${\mathcal Q}$,
6470: the string field action is written as
6471: \begin{equation}\label{eg1}
6472: S = -K_0 \Big[ {1\over 2} \langle \Phi, \mathcal{Q} \Phi\rangle +
6473: {1\over 3} \langle \Phi, \Phi * \Phi \rangle \Big]\, ,
6474: \end{equation}where the overall
6475: normalization $K_0$ turns out to be infinite.
6476: Although the constant $K_0$ can be absorbed into a rescaling of $\Psi$,
6477: this changes the normalization of
6478: ${\mathcal Q} $. We shall instead choose a
6479: convenient normalization of $\mathcal{Q}$ and keep the constant $K_0$ in the
6480: action as in eq.(\ref{eg1}).
6481:
6482:
6483: In this VSFT the ansatz was made that any D$p$-brane
6484: solution takes the factorized form~\cite{RSZ-2}
6485: \begin{equation}\label{eo3}
6486: \Phi = \Phi_m \otimes \Phi_g\, ,
6487: \end{equation}where $\Phi_g$
6488: denotes a state obtained by acting with the ghost
6489: oscillators on the SL(2,R) invariant vacuum of the ghost CFT, and
6490: $\Phi_m$ is a
6491: state obtained by acting with matter operators on the SL(2,R)
6492: invariant vacuum of the matter CFT.
6493: Let us denote by
6494: $*^g$ and $*^m$ the star product in the ghost and matter sector
6495: respectively.
6496: Eq.(\ref{e2}) then factorizes as
6497: \begin{equation}\label{eo4}
6498: \mathcal{Q} \Phi_g = - \Phi_g *^g \Phi_g \,,
6499: \end{equation}and
6500: \begin{equation}\label{eo5}
6501: \Phi_m = \Phi_m *^m \Phi_m\, .
6502: \end{equation}This last equation is particularly simple: it states
6503: that $\Phi_m$
6504: is a projector
6505: (a projector $P$ in an algebra with product $*$ is an element that
6506: satisfies $P* P = P$).
6507: The equation for $\Phi_g$ appears to be more complicated.
6508:
6509: For
6510: any string field configuration
6511: $\Phi$ that satisfies the equation of motion, the action is given by
6512: \begin{equation}
6513: S = -{K_0\over 6} \langle \Phi, \mathcal{Q} \Phi \rangle \,,
6514: \end{equation}
6515: and with the ansatz (\ref{eo3}) this becomes
6516: \begin{equation}
6517: S = -{K_0\over 6} \langle \Phi_g | \mathcal{Q} \Phi_g \rangle
6518: \langle \Phi_m |
6519: \Phi_m\rangle\,,
6520: \end{equation}
6521: Here the inner products are the BPZ ones for the separate matter and
6522: ghost conformal field theories. For any static
6523: solution, the action is equal to minus the potential energy. If we
6524: are describing a
6525: D$p$-brane, the action is equal to minus the volume of the brane
6526: times the tension of the
6527: brane.
6528:
6529: To proceed further it is assumed that the ghost part $\Phi_g$ is
6530: universal for all
6531: D$p$-brane solutions. Under this assumption the ratio of energies
6532: associated with two different D-brane solutions with matter parts
6533: $\Phi_m'$ and $\Phi_m$ respectively, is given by:
6534: \begin{equation}\label{eo7}
6535: {E'\over E} = {\langle \Phi_m' | \Phi_m'\rangle_m \over \langle \Phi_m |
6536: \Phi_m\rangle_m} \, .
6537: \end{equation}Thus the ghost part drops out of this calculation.
6538: The inner products in the above right-hand side
6539: include brane volume factors, which once removed, give
6540: us brane tensions. Equation
6541: (\ref{eo7}) has allowed
6542: some important tests of VSFT.
6543: If solutions $\Phi_m'$ and $\Phi_m$ are available, one
6544: can calculate the ratio of tensions of D-branes. Since the ratios
6545: are known, one has a test of VSFT. The solutions, as mentioned before, are
6546: projectors of the star algebra. The D25-brane solution, for example,
6547: can be represented by the sliver state $|\Xi\rangle$, which is the first
6548: example of a star-algebra projector that was discovered. The sliver state
6549: can be constructed for any conformal field theory (a brief discussion is
6550: given in the following subsection).
6551: Similarly, D$p$-brane solutions can be obtained as modified slivers,
6552: and numerical verification that the correct ratio of tensions emerges
6553: was obtained~\cite{RSZ-2}. Subsequently, and equipped with a better
6554: understanding
6555: of the star-algebra, Okuyama~\cite{Okuyama:2002tw} was able to demonstrate
6556: analytically that the correct ratio of tensions emerges.
6557:
6558: In a series of stimulating papers~\cite{Hata-Kawano,VSFT-tension},
6559: Hata, Kawano,
6560: and Moriyama, showed that the relationship $2\pi^2 g^2 T_{25} =1$ between the
6561: D25-brane tension and the string coupling can be tested in VSFT
6562: without knowledge of
6563: the explicit form of the purely ghost
6564: $\mathcal{Q}$. In other words, the normalization of the action, the infinite
6565: constant $K_0$, does not feature in the computation. This is easy to see.
6566: The D-brane tension, which is proportional to the value of the
6567: action evaluated
6568: on the sliver solution, is linearly proportional to $K_0$.
6569: In order to calculate the string coupling, Hata and Kawano proposed to look
6570: for the tachyon state on the D-brane; this state should appear as a fluctuation
6571: around the sliver solution. With this tachyon state, the string
6572: coupling $g$ can
6573: be obtained as the coupling for three on-shell tachyons. The effective
6574: action for the tachyon fluctuation $t$ would take the form
6575: \begin{equation}
6576: K_0 \Bigl( \alpha\,{1\over 2} \, t\, (\partial^2 + 1) t + {1\over
6577: 3}\, \beta\, t^3 \Bigr)\,,
6578: \end{equation}
6579: where $\alpha$ and $\beta$ are calculable finite constants. The
6580: field rescaling
6581: $t= T/ \sqrt{K_0 \alpha}$ brings this action to canonical form
6582: \begin{equation}
6583: \,{1\over 2} \, T\, (\partial^2 + 1) T + {1\over 3}\, {\beta\over
6584: \sqrt{K_0 \alpha} }\, T^3\,,
6585: \end{equation}
6586: and the string coupling can be read $g = \beta/\sqrt{K_0 \alpha}$.
6587: Since $T_{25} \sim
6588: K_0$, the relation $2\pi^2 g^2 T_{25} =1$ does not involve $K_0$.
6589: The original
6590: computations, however, did not work out, because the tachyon state had been
6591: incorrectly identified~\cite{0111153}. In a remarkable
6592: work~\cite{Okawa}, Okawa
6593: gave a correct identification of the tachyon state and demonstrated that
6594: the relation between the string coupling and the brane tension works
6595: out correctly. Still both the string coupling and the brane tension
6596: are singular.
6597:
6598: \medskip
6599: It is interesting to wonder what features
6600: of VSFT that depend on the particular
6601: choice of pure ghost operator $\mathcal{Q}$. It appears that
6602: a completely regular definition
6603: of the spectrum of strings around D-brane solutions may involve
6604: $\mathcal{Q}$. Indeed, Okawa has recently demonstrated that
6605: the knowledge of $\mathcal{Q}$ is necessary to produce VSFT
6606: solutions that give rise to a string coupling and brane tension both
6607: of which are finite~\cite{okawatbp1}.
6608: The specific form of $\mathcal{Q}$ may
6609: also be needed for the calculation of closed string
6610: amplitudes using VSFT. It is clear, however, that the choice in (\ref{ep1})
6611: is rather special. We remarked earlier that the equation (\ref{eo4}) for the
6612: ghost part of the solution is not just a projector equation. It
6613: turns out, however,
6614: that there is a twist of the ghost CFT of $(b,c)$ in which the
6615: antighost becomes
6616: a field of dimension one and ghost becomes a field of dimension zero. The
6617: new CFT has central charge $c=-2$. If $\mathcal{Q}$ is given by
6618: (\ref{ep1}), the solution
6619: of (\ref{eo4}) is simply the sliver state of the twisted conformal field
6620: theory.~\cite{RSZ-closed}
6621:
6622:
6623: We conclude this subsection with
6624: some comments on
6625: regularization and the singular aspects of VSFT. Arguments by
6626: Gross and Taylor~\cite{Gross-Taylor-II}, and by Schnabl (unpublished)
6627: indicated that the brane tension associated with VSFT solutions is
6628: zero for any finite $K_0$. Numerical experiments confirm these
6629: arguments. A regulation scheme
6630: was developed by Gaiotto {\em et.al} ~\cite{RSZ-closed} in which
6631: $K_0$ is replaced by $K_0(a)$, and the gauge-fixed kinetic operator
6632: of VSFT is made $a$-dependent in such a way that for for infinite $a$
6633: the pure ghost operator is recovered. The $K_0(a)$ divergence
6634: as $a\to \infty$ is determined from the requirement that the D-brane
6635: tension is correctly reproduced. The regulated theory appears to be
6636: well defined, but universality is lost in the regulation. On the other
6637: hand, the analysis of the regulated equations led to the discovery
6638: of another special projector of the star algebra: the butterfly
6639: state~\cite{RSZ-closed,Schnabl,RSZ-projectors}.
6640:
6641: We noted in section~\ref{sec:vacuum-theory} that
6642: after a shift to the tachyon vacuum the open string field theory on
6643: a D25-brane becomes a cubic string
6644: field theory with kinetic operator $\widetilde Q$. This operator is
6645: not made solely
6646: of ghosts. We would expect, however, that the RSZ theory, if fully
6647: correct, is
6648: field redefinition equivalent to the theory with $\widetilde Q$.
6649: If we consider the action (\ref{e2p}),
6650: a homogeneous field redefinition of the type
6651: \begin{equation}\label{jkl}
6652: \widetilde\Phi = \, e^K \, \Phi\,,
6653: \end{equation}has special properties if $K$ is a ghost number zero Grassmann
6654: even operator that satisfies the following relations
6655: \begin{eqnarray} \label{efderp}
6656: && K (A * B) = (K A) * B + A * (K B)\,, \nonumber \\
6657: && \langle \, K A , B \,\rangle = - \langle A , K B \rangle \,.
6658: \end{eqnarray}
6659: These properties guarantee that the form of the
6660: cubic term is unchanged, and
6661: that, after the field redefinition, the action takes the form
6662: \begin{equation}\label{e2findp}
6663: S (\Phi) = \,-\, {1\over g^2}\,\,\bigg[\, {1\over 2} \langle
6664: \,\Phi \,,\, {\widehat Q}\, \Phi
6665: \rangle + {1\over 3}\langle \,\Phi \,,\, \Phi *
6666: \Phi \rangle \bigg] \,,
6667: \end{equation}where
6668: \begin{equation}
6669: \label{kjh}
6670: {\widehat Q} = \, e^{-K} \, \widetilde Q \, e^K\,.
6671: \end{equation}It is a good exercise to verify that equations (\ref{efderp})
6672: guarantee that $\widetilde Q$ satisfies the properties listed
6673: in (\ref{l1e1}). Therefore the new action is consistent.
6674:
6675: The operator $\widetilde Q$ is, by construction, regular, while
6676: $\widehat Q$, which we
6677: want to be equal to the VSFT operator $\mathcal{Q}$, should be an infinite
6678: constant times a ghost insertion at the string midpoint
6679: (the infinite constant is necessary because $g$ is finite). A large
6680: class of string
6681: reparameterizations that leave the open string midpoint invariant
6682: can be constructed with operators $K$ that satisfy the relations
6683: (\ref{efderp}).
6684: A reparameterization in which a finite part of the string is squeezed
6685: into an infinitesimal neighborhood of the string midpoint will turn a regular
6686: $\widetilde Q$ that contains a term linear in the ghost field, into an operator
6687: $\widehat Q$ whose leading term is precisely
6688: a divergent ghost insertion at the
6689: string midpoint.~\cite{RSZ-closed} This happens because the term linear
6690: in the ghost field is the term with an operator of lowest possible dimension,
6691: and a squeezing transformation, will transform this
6692: negative-dimension operator
6693: with an infinite factor. It is thus plausible that a singular
6694: squeezing transformation
6695: relates the string field theory around the tachyon vacuum
6696: to the RSZ theory.
6697:
6698:
6699:
6700:
6701:
6702: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6703: % Slivers and projections: both
6704: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6705:
6706: \subsection{Slivers and projection operators}
6707: \label{subsec:slivers}
6708:
6709: % We still need to decide what to put in this section.
6710:
6711:
6712: {}From the point of view of the RSZ approach to VSFT just discussed,
6713: projection operators of the star algebra play a
6714: crucial role in the construction of solutions of the theory.
6715: Such projection operators may also be useful in understanding
6716: solutions in the original Witten theory.
6717: Quite a bit of work has been
6718: done on constructing and analyzing projectors in the star
6719: algebra since the RSZ model was originally proposed. Without
6720: going into the technical details,
6721: we now briefly review some of the important features of projectors.
6722:
6723: \begin{figure}[!ht]
6724: \leavevmode
6725: \begin{center}
6726: \epsfxsize = 11 cm \epsfbox{BZFig9.eps}
6727: \end{center}
6728: \caption[]{\footnotesize The
6729: sliver appears as a cone with infinite excess angle-- namely,
6730: an infinite helix. The segments $AM$ and $BM$ represent the
6731: left-half and the right-half of the string. The local
6732: coordinate patch,
6733: represented by the shaded half disk shown to the right, must be
6734: glued in to form the complete surface.
6735: } \label{f9}
6736: \end{figure}
6737:
6738:
6739:
6740: The first matter projector which was explicitly constructed is the
6741: ``sliver'' state. This state was identified as a conformal field theory
6742: surface state by Rastelli and Zwiebach~\cite{Rastelli-Zwiebach}. As such,
6743: there is a surface associated with the state: a disk with one puncture
6744: on the boundary and a specified local coordinate at the puncture. This
6745: conformal field theory picture gives a complete state; it includes both
6746: the matter and the ghost part of the state. Moreover, the state can be
6747: constructed for any conformal field theory:
6748: \begin{equation}\label{slivasvirdesc}
6749: |\Xi\rangle = \exp
6750: \Bigl( -{1\over 3} L_{-2} + {1\over 30} L_{-4} - {11\over 1890} L_{-6} +
6751: {34\over 467775} L_{-8}
6752: + \cdots \Bigr) | 0\rangle \,.
6753: \end{equation}The geometrical picture of the sliver state is shown in
6754: figure~\ref{f9}.
6755: The full punctured disk is the glued surface obtained
6756: by attaching the infinite helix and the coordinate patch,
6757: which carries the puncture~$P$.
6758: There are many
6759: alternative pictures of the sliver.
6760:
6761: \begin{figure}[!ht]
6762: \leavevmode
6763: \begin{center}
6764: \epsfxsize = 10 cm \epsfbox{BZFig8.eps}
6765: \end{center}
6766: \caption[]{\footnotesize The star multiplication of a sector
6767: state with angle $\alpha$ to a sector state with angle
6768: $\beta$ gives a sector state with angle $\alpha+ \beta$.
6769: Sector states are just another presentation of wedge
6770: states.
6771: } \label{f8}
6772: \end{figure}
6773:
6774:
6775: To understand why the sliver state squares to itself one must
6776: have a picture of star multiplication for surface states.
6777: A full discussion~\cite{Rastelli:2001vb} would take too long, but the rough
6778: idea is easily explained.
6779: The sliver state is essentially the limit $\lim_{n \rightarrow \infty}
6780: (| 0 \rangle)^n$, where multiplication is performed via the star product.
6781: A surface state in a BCFT can be viewed (by excising the
6782: coordinate patch) as a disk whose boundary has two parts: a part in which
6783: the boundary condition that defines the BCFT is imposed, and a
6784: part which represents an open string. To star-multiply two surface
6785: states, one glues the right-half of the string in the first surface to the
6786: left-half of the string in the second surface; the resulting surface is the
6787: surface that represents the star product. A particularly simple class
6788: of surface states are sector states or wedge states. One such state
6789: $\mathcal{R}_\alpha$
6790: is shown to the left of figure~\ref{f8}. The BCFT boundary condition
6791: applies to the curved boundary of
6792: the sector. The radial segment AM is
6793: the left-half of the open string and the radial segment $MB$ is the right-half
6794: of the open string. The sector state is defined by the angle
6795: $\alpha$ at the string
6796: midpoint $M$. In the figure we show the multiplication of $\mathcal{R}_\alpha$
6797: and $\mathcal{R}_\beta$. The result is a sector state
6798: $\mathcal{R}_{\alpha + \beta}$ with total angle
6799: $\alpha + \beta$. The sliver state $\Xi$ is the wedge state
6800: $\mathcal{R}_\infty$
6801: with infinite angle. It is then clear that the star product of two
6802: slivers is still
6803: a wedge state of infinite angle, and thus also a sliver. The state
6804: obtained in the
6805: limit when the angle is equal to zero is in fact the identity state of the star
6806: algebra. It is manifestly clear that the product of any surface state with the
6807: identity gives the surface state. The identity state can also be written
6808: as an exponential of Virasoro operators acting on the vacuum.
6809: In fact,
6810: as mentioned in section \ref{sec:decoupling},
6811: a very curious result was
6812: found~\cite{efhm}:
6813: \begin{eqnarray}
6814: |\mathcal{I} \rangle & = & \left( \prod\limits_{n=2}^{\infty}
6815: \exp\left\{- \frac{2}{2^n} L_{-2^n}\right\} \right)
6816: e^{L_{-2}} |0 \rangle \nonumber \\
6817: &=&
6818: \ldots
6819: \exp(-\frac{2}{2^3} L_{-2^3}) \exp(-\frac{2}{2^2} L_{-2^2})
6820: \exp(L_{-2}) |0 \rangle \,,
6821: \end{eqnarray}
6822: with the Virasoro operators of higher level stacking to the left.
6823: We thus confirm that the identity is also a Virasoro descendent of the
6824: vacuum.
6825:
6826:
6827:
6828: In an independent construction, Kostelecky and
6829: Potting~\cite{Kostelecky-Potting}
6830: constructed a state $\Psi_m$ of the matter sector of the D25-brane BCFT that
6831: squared to itself (up to a proportionality constant). The construction used
6832: the oscillator
6833: language. This matter state
6834: takes the form of a squeezed state
6835: \begin{equation}
6836: |\Psi_m\rangle = \mathcal{N} \,\exp \left[\frac{1}{2}\,
6837: a^{\dagger} \cdot S \cdot a^{\dagger} \right]| 0 \rangle \,.
6838: \label{eq:sliver-squeezed}
6839: \end{equation}
6840: By requiring that such a state satisfy the projection equation $\Psi
6841: \star \Psi = \Psi$, and by making some further assumptions about the
6842: nature of the state, an explicit formula for the matrix $S$ was
6843: found in terms of the matrix $X$ from
6844: (\ref{eq:matX})~\cite{Kostelecky-Potting}. Evidence quickly emerged
6845: that the state constructed by these authors is the matter sector of
6846: the sliver state, and a proof was given by Okuda~\cite{Okuda:2002fj}.
6847:
6848: There are many other projectors that also have a simple
6849: picture as surface states~\cite{Schnabl,RSZ-projectors,Fuchs:2002zz}.
6850: In these projectors, the open string midpoint approaches
6851: (or even coincides with) the boundary of the surface
6852: where the boundary condition is applied.
6853: One particularly useful projector, which arises in the numerical
6854: solution of VSFT, is the so-called {\em butterfly} state $\mathcal{B}$.
6855: This is a very interesting state, whose picture is shown in
6856: Figure~\ref{f2butter}.
6857: When one glues two butterfly surfaces in the manner required by
6858: star-multiplication, the resulting surface does not appear to be, at
6859: first sight,
6860: another butterfly. Nevertheless, the resulting surface is conformally
6861: equivalent
6862: to a butterfly, and this is, in fact, all that is needed in order to have a
6863: projector. It has been demonstrated that the butterfly is the state
6864: that can be represented as the tensor product $|0\rangle \otimes |0\rangle$,
6865: where $|0\rangle $ is the vacuum of the half-string state
6866: space~\cite{RSZ-projectors}.
6867: Generally, any state of the form $| a \rangle \otimes | a \rangle$
6868: where $| a \rangle$ is the same state in the left and right
6869: half-string Fock spaces is a projector~\cite{Gross-Taylor-I}.
6870: The butterfly has a remarkably simple expression
6871: as a Virasoro descendent of the vacuum
6872: \begin{equation}
6873: \label{butterflyasvirdes}
6874: |\mathcal{B}\rangle = \exp \Bigl( -{1\over 2} L_{-2} \Bigr) |0\rangle \,.
6875: \end{equation}
6876: \begin{figure}[!ht]
6877: \leavevmode
6878: \begin{center}
6879: \epsfysize=5cm
6880: \epsfbox{fig2new.eps}
6881: \end{center}
6882: \caption{\footnotesize The butterfly state arises in the limit where
6883: $\alpha\to 1$ and the angle indicated in the figure vanishes.} \label{f2butter}
6884: \end{figure}
6885:
6886:
6887:
6888: Projectors have many properties which are reminiscent
6889: of D-branes. This relationship between projection operators and
6890: D-branes is familiar from noncommutative field theory, where
6891: projectors also play the role of D-brane solitons~\cite{gms}.
6892: This connection becomes quite concrete in the presence of a background
6893: B field~\cite{Moore-Taylor,b-VSFT}.
6894: In the RSZ theory,
6895: states that describe an arbitrary but fixed
6896: configuration of D-branes are constructed
6897: by tensoring the matter
6898: projector for the appropriate BCFT with a fixed ghost state
6899: that satisfies the ghost
6900: equation of motion (\ref{eo4}). Particular projectors
6901: like the sliver can be constructed which are localized in any number
6902: of space-time dimensions, corresponding to the codimension of a
6903: D-brane. Under gauge transformations, a rank one projector can be
6904: rotated into an orthogonal rank one projector, so that configurations
6905: containing multiple branes can be constructed as higher rank
6906: projectors formed from the sum of orthogonal rank
6907: one projectors~\cite{RSZ-3,Gross-Taylor-I}.
6908: This gives a very suggestive picture of
6909: how arbitrary D-brane configurations can be constructed in string
6910: field theory.
6911:
6912: While this picture is quite compelling, however, there
6913: are some technical obstacles which make this still a somewhat
6914: incomplete story. In the RSZ model,
6915: singularities appear due to the separation of the matter and ghost sectors.
6916: Moreover, projectors are, in general, somewhat singular states.
6917: For example, the matrix $S$ associated to the matter part of
6918: the sliver state has eigenvalues of
6919: $\pm 1$ for any D$p$-brane~\cite{Moore-Taylor,RSZ-projectors}. Such
6920: eigenvalues cause
6921: these states to be non-normalizable elements of the matter Fock
6922: space
6923: In the Dirichlet directions, this lack of normalizability
6924: occurs because the
6925: state is essentially localized to a point and is analogous to a delta
6926: function. In the Neumann directions, the singularity manifests as a
6927: ``breaking'' of the strings composing the D-brane, so that the
6928: functional describing the projector state is a product of a function
6929: of the string configurations on the left and right halves of the
6930: string, with no connection mediated through the midpoint. These
6931: geometric singularities seem to be generic features of the matter part of
6932: any projector, not just the sliver state~\cite{Schnabl,RSZ-projectors}. The
6933: singular geometric features of projectors, which can be traced
6934: to the fact that the open string midpoint approaches the boundary,
6935: makes certain calculations in the RSZ theory somewhat complicated, as all
6936: singularities must be regulated. Singularities do not
6937: seem to appear in the Witten theory, where the BRST operator and
6938: the numerically calculated solutions seem to behave smoothly at the string
6939: midpoint. On the other hand, it may be that further study of the
6940: projectors will lead to analytic progress on
6941: the Witten theory, as discussed in a recent paper by Okawa~\cite{okawatbp2}.
6942:
6943: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6944: % Slivers and projections: both
6945: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6946:
6947: \subsection{Closed strings in open string theory}
6948: \label{subsec:closedstringsinosft}
6949:
6950: We have discussed in earlier sections the fact that
6951: open string field theory, formulated on the background
6952: of a certain BCFT appears to capture many other open
6953: string backgrounds as solutions of the theory.
6954: Apart from its singular features, the RSZ theory
6955: admits any BCFT as a solution of the theory. One
6956: important question remains: Can closed string backgrounds
6957: be incorporated in open string field theory? The question
6958: can be answered both in the context of OSFT and in the
6959: context of the RSZ model. As we will discuss,
6960: there is very little
6961: concrete evidence as yet that this can be done
6962: in any of the two approaches.
6963: We therefore ask a simpler question: Can closed string states
6964: be seen in open string field theory?
6965: The answer here is yes, both in OSFT, and in VSFT
6966: (modulo the usual singularities),
6967: although so far this has been understood only in certain limited contexts.
6968:
6969: As has been known since the earliest days of the subject, closed
6970: strings appear as poles in perturbative open string scattering
6971: amplitudes. This was demonstrated explicitly for Witten's theory by
6972: exhibiting the closed string poles arise in the one-loop 2-point
6973: function~\cite{fgst}
6974: (although in this calculation, spurious poles also
6975: appear which complicate the interpretation). More recently, in a
6976: similar calculation the closed string tadpole generated by the D-brane
6977: was identified in the one-loop open string 1-point
6978: function~\cite{Ellwood-Shelton-Taylor}. While in principle this type
6979: of argument
6980: can be used to construct all on-shell closed string amplitudes through
6981: factorization, it is much less clear how to think of asymptotic or
6982: off-shell closed
6983: string states in this context. If Witten's theory is well-defined as a quantum
6984: theory, it would follow from unitarity that the closed string states
6985: should also
6986: arise in some natural sense as asymptotic states of the quantum open
6987: string field
6988: theory. It is currently rather unclear, however, whether, and if so
6989: how, this might be realized. There are subtleties in the quantum
6990: formulation of the theory which have never completely been
6991: resolved~\cite{Thorn,Ellwood-Shelton-Taylor}, although most of the
6992: problems of the quantum theory seem to be generated by the closed
6993: string tachyon, and may be absent in a supersymmetric theory. Both
6994: older SFT literature~\cite{Strominger-closed,Shapiro-Thorn} and recent
6995: work~\cite{Gerasimov-Shatashvili,Moore-Taylor,RSZ-closed,Hashimoto-Itzhaki}
6996: have suggested ways in which closed strings might be incorporated into
6997: the open string field theory, but a definitive resolution of this
6998: question is still not available.
6999:
7000:
7001: In the RSZ model, one description of {\em on-shell} closed string
7002: states is reasonably
7003: natural~\cite{RSZ-closed,Hashimoto-Itzhaki,Ambjorn:2002im,Drukker:2002ct}
7004: and scattering amplitudes have been
7005: computed~\cite{Alishahiha:2002as,Takahashi:2003kq}.
7006: For each on-shell closed string vertex operator $V$ one can construct a
7007: gauge-invariant open string
7008: state $\mathcal{O}_V(\Phi)$, where $\Phi$ is the open string field,
7009: and the gauge invariance is the open string gauge invariance.
7010: The world-sheet picture of the state is that of an amputated semi-infinite
7011: strip whose edge represents the open strings, the two halves of which
7012: are glued and the closed string operator is inserted at the conical
7013: singularity. Given a set of gauge invariant operators associated
7014: with a set of on-shell closed string vertex operators, the RSZ
7015: correlator of the gauge invariant operators appears to give,
7016: up to proportionality factors that need regulation, the
7017: on-shell closed string amplitude on a surface {\em without}
7018: boundaries.
7019: This result uses a nontrivial and unusual decomposition of
7020: the moduli space of Riemann surfaces without boundaries~\cite{RSZ-closed}.
7021: The decomposition, is related to, but distinct from the one used
7022: in Witten's theory to cover the moduli space of Riemann surfaces
7023: that have at least one boundary. Other decompositions have been
7024: discussed by Drukker~\cite{Drukker:2002ct}.
7025:
7026:
7027: If it were possible to encode {\em off-shell} physics naturally
7028: into open string field theory it would be reasonable to hope
7029: that closed string backgrounds could be changed by suitable
7030: expectation values of open string fields
7031: although this would presumably be a subtle effect in the quantum
7032: theory, and difficult to compute explicitly. Attaining a description
7033: of the full closed string landscape~\cite{landscape} using quantum
7034: OSFT is clearly an optimistic scenario,
7035: but it need not be farfetched;
7036: it may represent an extension
7037: of the AdS/CFT correspondence, in which the CFT side is changed
7038: from SYM into the full open string field theory.
7039: If, as it may be,
7040: it turns out to be that the closed string sector of the theory
7041: is encoded in a singular fashion in OSFT,
7042: one may be better off directly working with
7043: closed string field theory~\cite{Zwiebach:1992ie}, or with
7044: open/closed string field
7045: theory~\cite{Zwiebach:1997fe}.
7046: Because of the nonpolynomiality of these theories, it is
7047: not known at present if
7048: level expansion can be used to extract nonperturbative information.
7049: At any rate, it would be useful to have a clear picture of how far one
7050: can incorporate closed string physics from the open string point of
7051: view. Even if this cannot be realistically achieved in our current models of
7052: SFT, understanding the difficulties involved may help us in our
7053: search for a better formulation of the theory.
7054:
7055:
7056:
7057:
7058: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
7059: % Conclusions
7060: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
7061:
7062: \section{Conclusions}
7063: \label{sec:conclusions}
7064:
7065:
7066: The work described in these lectures has brought the understanding of
7067: string field theory to a new level. We now have fairly conclusive
7068: evidence that open string field theory can successfully describe
7069: distinct vacua with very different geometrical properties, which are
7070: not related to one another through a marginal deformation. The
7071: resulting picture, in which a complicated set of degrees of freedom
7072: defined primarily through an algebraic structure, can produce
7073: different geometrical backgrounds as different solutions of the
7074: equations of motion, represents an important step beyond perturbative
7075: string theory. Such a framework is clearly necessary to discuss questions of a
7076: cosmological nature in string theory. For such questions,
7077: however, one must generalize from the work described here in which the theory
7078: describes distinct {\it open} string backgrounds, to a formalism
7079: where different
7080: {\it closed} string backgrounds also appear as solutions of the equations.
7081: Ideally, we would like to have a formulation of
7082: string/M-theory in which all the currently understood vacua can arise
7083: in terms of a single well-defined set of degrees of freedom.
7084:
7085:
7086:
7087: It is not yet clear, however, how far it is possible go towards this
7088: goal using the current formulations of string field theory. It may be
7089: that the correct lesson to take from the work described here is simply
7090: that there {\it are} nonperturbative formulations in which distinct
7091: vacua can be brought together as solutions of a single classical
7092: theory, and that one should search for some deeper fundamental
7093: algebraic formulation where geometry, and even the dimension of
7094: space-time emerge from the fundamental degrees of freedom in the same
7095: way that D-brane geometry emerges from the degrees of freedom of
7096: Witten's open string field theory. A more conservative scenario,
7097: however, might be that we could perhaps use the current framework of
7098: string field theory, or some limited refinement thereof, to achieve
7099: this goal of providing a universal nonperturbative definition of
7100: string theory and M-theory. Following this latter scenario, we propose
7101: here a series of questions aimed at continuing the recent developments
7102: in open string field theory as far as possible towards this ultimate
7103: goal. It is not certain that this research program can be carried to
7104: its conclusion, but it will be very interesting to see how far
7105: open string field theory can go in reproducing important
7106: nonperturbative aspects of string theory.
7107:
7108:
7109: \medskip
7110: There are, in our mind, two very important concrete problems
7111: related to Witten's string field theory that so far have resisted solution:
7112:
7113: \begin{enumerate}
7114:
7115: \item[1)] Finding
7116: an analytic description of the tachyonic vacuum. Despite several
7117: years of work on this problem, great success with numerical
7118: approximations, and some insight from the RSZ vacuum string field
7119: theory model, we still have no closed form expression for the
7120: string field $\Phi_0$ which represents the tachyon vacuum
7121: in Witten's open string field theory. It seems almost
7122: unbelievable that there is not some elegant analytic solution to
7123: this problem. An analytic
7124: solution would almost certainly greatly enhance our understanding of
7125: this theory and would lead to other significant advances.
7126:
7127:
7128: \item[2)] Finding certain open string backgrounds as solutions of open
7129: string field theory. As discussed in section~\ref{subsec:thebackgroundsosft},
7130: we do not know how to obtain a background with multiple
7131: D-branes starting with a background with one D-brane. Nor we
7132: know how to obtain the background which represents a D0-brane
7133: using the background of a D1-brane with lower energy.
7134: It is currently unclear whether the obstacles to finding these
7135: vacua are conceptual or technical.
7136:
7137: \end{enumerate}
7138:
7139: There are other questions that are probably important to the
7140: future development of string field theory. These represent,
7141: in our opinion, subjects that merit investigation:
7142:
7143: \begin{enumerate}
7144:
7145: \item[1)] Is there a regular formulation of VSFT ?
7146: Such a version of the theory may have further similarities with BSFT
7147: and could turn out to be a complete and flexible formulation
7148: of open string field theory.
7149:
7150:
7151: \item[2)] How do closed string backgrounds appear
7152: in open string field theory? While OSFT and VSFT
7153: appear to give somewhat singular/intractable descriptions
7154: of closed string physics, some better understood, or new,
7155: version of open string theory might provide
7156: a tractable description of closed string physics.
7157: Another possibility
7158: is that closed string fields are needed in addition to
7159: open string fields; this is the case
7160: in light-cone open string field theory and in
7161: covariant open/closed string field theory.
7162:
7163: \item[3)] What are the new features of superstring field theory?
7164: The status of the tachyon conjectures for the superstring
7165: has been reviewed by
7166: Ohmori~\cite{Ohmori:2003vq}.
7167: The large set of symmetries of superstring theory makes them, in many
7168: cases, more tractable than bosonic string theories. Nevertheless,
7169: as of yet, there is no clear sense in which superstring field theory
7170: is simpler than bosonic string field theory~\cite{smet}.
7171: There are also significant conceptual problems that have not
7172: allowed a formulation of vacuum superstring field theory~\cite{Marino:2001ny}.
7173:
7174:
7175: \item[4)] How do we describe time-dependent tachyon dynamics?
7176: String field theory gives clear and concrete evidence for the
7177: Sen conjectures. Although
7178: we have not studied this subject in the present review, there is
7179: much interest in the process by which the tachyon rolls from the
7180: unstable critical point down to the tachyon vacuum.~\cite{rolling-tachyon}
7181: In fact, the early attempts to describe the rolling of the tachyon
7182: in Witten's string field theory~\cite{Moeller:2002vx,Fujita:2003ex} appear to
7183: be in contradiction with the results that follow from conformal
7184: field theory.
7185:
7186:
7187: \end{enumerate}
7188:
7189:
7190:
7191: It is challenging to imagine a
7192: single set of degrees of freedom which could encode, in different
7193: phases, all the possible string backgrounds we are familiar with,
7194: including those associated with M-theory.
7195: In principle, a
7196: nonperturbative background-independent formulation of type II string
7197: theory should allow one to take the string coupling to infinity in
7198: such a way that the fundamental degrees of freedom of the theory
7199: remain at some finite point in the configuration space.
7200: This would lead to the vacuum associated with M-theory
7201: in flat space-time. It would be quite remarkable if this can be
7202: achieved in the framework of string field theory. Given the
7203: nontrivial relationship between string fields and low-energy
7204: effective degrees of freedom,
7205: such a result need not be farfetched.
7206: If this picture could be successfully implemented, it
7207: would give a very satisfying
7208: representation of the
7209: complicated network of dualities of string and M-theory in terms of
7210: a single underlying set of degrees of freedom.
7211:
7212:
7213:
7214:
7215:
7216: \section*{Acknowledgments}
7217: We would like to thank all our collaborators: Nathan Berkovits, Erasmo
7218: Coletti, Michael Douglas, Ian Ellwood, David Gaiotto, David Gross,
7219: Hong Liu, Nicolas Moeller, Greg Moore, Leonardo Rastelli, Ashoke Sen,
7220: Jessie Shelton, and Ilya Sigalov, who helped us understand many of the
7221: issues described in these lectures. Thanks to the organizers and
7222: students of TASI 2001, who provided a stimulating environment for
7223: these lectures and much relevant feedback. WT would like to thank the
7224: Kavli Institute for Theoretical Physics for hospitality and support
7225: during the final stages of writing up these lectures. BZ would like
7226: to thank the Harvard University Physics Department for hospitality.
7227: This work was
7228: supported by the DOE through contract \#DE-FC02-94ER40818.
7229:
7230:
7231:
7232:
7233: \begin{thebibliography}{99}
7234:
7235: \bibitem{Ohmori}
7236: % lookup 0102085
7237: K.~Ohmori,
7238: ``A review on tachyon condensation in open string field theories,''
7239: {\tt hep-th/0102085}.
7240:
7241:
7242: \bibitem{deSmet}
7243: % lookup 0109182
7244: P.~J.~De Smet,
7245: ``Tachyon condensation: Calculations in string field theory,''
7246: {\tt hep-th/0109182}.
7247:
7248:
7249:
7250: \bibitem{abgkm}
7251: % lookup 0111208
7252: I.~Y.~Aref'eva, D.~M.~Belov, A.~A.~Giryavets, A.~S.~Koshelev and
7253: P.~B.~Medvedev,
7254: ``Noncommutative field theories and (super)string field theories,''
7255: {\tt hep-th/0111208}.
7256:
7257: \bibitem{Bonora}
7258: % lookup 0304270
7259: %\bibitem{Bonora:2003xp}
7260: L.~Bonora, C.~Maccaferri, D.~Mamone and M.~Salizzoni,
7261: ``Topics in string field theory,''
7262: {\tt hep-th/0304270}.
7263:
7264:
7265:
7266: \bibitem{Taylor-Valdivia}
7267: W.\ Taylor, ``Lectures on D-branes, tachyon condensation and string
7268: field theory'', School on quantum gravity, Valdivia, Chile, January
7269: 2002; {\tt hep-th/0301094}
7270:
7271: \bibitem{rolling-tachyon}
7272: %~\cite{Sen:2002nu}
7273: %\bibitem{Sen:2002nu}
7274: A.~Sen,
7275: ``Rolling tachyon,''
7276: JHEP {\bf 0204}, 048 (2002)
7277: {\tt hep-th/0203211}.
7278: %%CITATION = HEP-TH 0203211;%%
7279: %~\cite{Sen:2002in}
7280: %\bibitem{Sen:2002in}
7281: A.~Sen,
7282: ``Tachyon matter,''
7283: JHEP {\bf 0207}, 065 (2002)
7284: {\tt hep-th/0203265}.
7285: %%CITATION = HEP-TH 0203265;%%
7286: [For up-to-date references, see the long citation list for the two
7287: papers above].
7288:
7289: \bibitem{Bars-original}
7290: % lookup 0106157
7291: %\bibitem{Bars:2001ag}
7292: I.~Bars,
7293: ``Map of Witten's * to Moyal's *,''
7294: Phys.\ Lett.\ B {\bf 517}, 436 (2001)
7295: {\tt hep-th/0106157}.
7296:
7297:
7298: \bibitem{Bars-all}
7299: % lookup 0202030
7300: %\bibitem{Bars:2002bt}
7301: I.~Bars and Y.~Matsuo,
7302: ``Associativity anomaly in string field theory,''
7303: Phys.\ Rev.\ D {\bf 65}, 126006 (2002)
7304: {\tt hep-th/0202030};
7305: % lookup 0204260
7306: %\bibitem{Bars:2002nu}
7307: I.~Bars and Y.~Matsuo,
7308: ``Computing in string field theory using the Moyal star product,''
7309: Phys.\ Rev.\ D {\bf 66}, 066003 (2002)
7310: {\tt hep-th/0204260};
7311: % lookup 0211131
7312: %\bibitem{Bars:2002qt}
7313: I.~Bars, I.~Kishimoto and Y.~Matsuo,
7314: ``String amplitudes from Moyal string field theory,''
7315: Phys.\ Rev.\ D {\bf 67}, 066002 (2003)
7316: {\tt hep-th/0211131};
7317: % lookup 0304005
7318: %\bibitem{Bars:2003gu}
7319: I.~Bars, I.~Kishimoto and Y.~Matsuo,
7320: ``Fermionic ghosts in Moyal string field theory,''
7321: JHEP {\bf 0307}, 027 (2003)
7322: {\tt hep-th/0304005}.
7323:
7324:
7325: \bibitem{Bars}
7326: % lookup 0211238
7327: I.~Bars,
7328: ``MSFT: Moyal star formulation of string field theory,''
7329: {\tt hep-th/0211238}.
7330:
7331: \bibitem{BSFT}
7332: %\bibitem{Gerasimov:2000zp}
7333: A.~A.~Gerasimov and S.~L.~Shatashvili,
7334: ``On exact tachyon potential in open string field theory,''
7335: JHEP {\bf 0010}, 034 (2000),
7336: {\tt hep-th/0009103};
7337: %\bibitem{Kutasov:2000qp}
7338: D.~Kutasov, M.~Marino and G.~Moore,
7339: ``Some exact results on tachyon condensation in string field theory,''
7340: JHEP {\bf 0010}, 045 (2000),
7341: {\tt hep-th/0009148};
7342: %\bibitem{Ghoshal:2000gt}
7343: D.~Ghoshal and A.~Sen,
7344: ``Normalisation of the background independent open string field
7345: %theory action,''
7346: JHEP {\bf 0011}, 021 (2000),
7347: {\tt hep-th/0009191};
7348: %\bibitem{Kutasov:2000aq}
7349: D.~Kutasov, M.~Marino and G.~Moore,
7350: % lookup 0010108
7351: D.~Kutasov, M.~Marino and G.~W.~Moore,
7352: ``Remarks on tachyon condensation in superstring field theory,'',
7353: {\tt hep-th/0010108};
7354: %\bibitem{Moriyama:2001dc}
7355: S.~Moriyama and S.~Nakamura,
7356: ``Descent relation of tachyon condensation from boundary string
7357: %field theory,''
7358: Phys.\ Lett.\ B {\bf 506}, 161 (2001),
7359: {\tt hep-th/0011002};
7360: %\bibitem{Gerasimov:2001ga}
7361: A.~A.~Gerasimov and S.~L.~Shatashvili,
7362: ``Stringy Higgs mechanism and the fate of open strings,''
7363: JHEP {\bf 0101}, 019 (2001),
7364: {\tt hep-th/0011009};
7365: %\bibitem{Aref'eva:2000mb}
7366: % lookup 0011117
7367: I.~Y.~Aref'eva, A.~S.~Koshelev, D.~M.~Belov and P.~B.~Medvedev,
7368: ``Tachyon condensation in cubic superstring field theory,''
7369: Nucl.\ Phys.\ B {\bf 638}, 3 (2002),
7370: {\tt hep-th/0011117};
7371: %\bibitem{Kraus:2001nj}
7372: P.~Kraus and F.~Larsen,
7373: ``Boundary string field theory of the D D-bar system,''
7374: Phys.\ Rev.\ D {\bf 63}, 106004 (2001),
7375: {\tt hep-th/0012198};
7376: %\bibitem{Alishahiha:2001du}
7377: M.~Alishahiha, H.~Ita and Y.~Oz,
7378: ``On superconnections and the tachyon effective action,''
7379: Phys.\ Lett.\ B {\bf 503}, 181 (2001),
7380: {\tt hep-th/0012222};
7381: %\bibitem{Chalmers:2001dj}
7382: % lookup 0103056
7383: G.~Chalmers,
7384: ``Open string decoupling and tachyon condensation,''
7385: JHEP {\bf 0106}, 012 (2001),
7386: {\tt hep-th/0103056};
7387: %\bibitem{Marino:2001qc}
7388: % lookup 0103089
7389: M.~Marino,
7390: ``On the BV formulation of boundary superstring field theory,''
7391: JHEP {\bf 0106}, 059 (2001),
7392: {\tt hep-th/0103089};
7393: %\bibitem{Niarchos:2001si}
7394: % lookup 0103102
7395: V.~Niarchos and N.~Prezas,
7396: ``Boundary superstring field theory,''
7397: Nucl.\ Phys.\ B {\bf 619}, 51 (2001),
7398: {\tt hep-th/0103102};
7399: %\bibitem{Viswanathan:2001cs}
7400: % lookup 0104099
7401: K.~S.~Viswanathan and Y.~Yang,
7402: ``Tachyon condensation and background independent superstring field theory,''
7403: Phys.\ Rev.\ D {\bf 64}, 106007 (2001),
7404: {\tt hep-th/0104099};
7405: %\bibitem{Alishahiha:2001tg}
7406: % lookup 0104164
7407: M.~Alishahiha,
7408: ``One-loop correction of the tachyon action in boundary superstring
7409: field theory,''
7410: Phys.\ Lett.\ B {\bf 510}, 285 (2001),
7411: {\tt hep-th/0104164}.
7412:
7413:
7414: \bibitem{Hull-Townsend}
7415: Hull, C.~M., and P.~K.~Townsend,
7416: ``Unity of superstring dualities,''
7417: Nucl.\ Phys.\ {\bf B 438}, 109, 1995;
7418: {\tt hep-th/9410167}.
7419:
7420: \bibitem{Witten-various}
7421: Witten, E.\,
7422: ``String Theory Dynamics in Various Dimensions,''
7423: \NP {\bf B 443};
7424: 85 1995; {\tt hep-th/9503124}.
7425:
7426: \bibitem{Townsend-11}
7427: Townsend, P.~K.,
7428: ``The eleven-dimensional supermembrane revisited,''
7429: Phys.\ Lett.\ {\bf B 350}, 184, 1995;
7430: {\tt hep-th/9501068}.
7431:
7432: \bibitem{Horava-Witten}
7433: Ho\v{r}ava, P.\, and E.~Witten,
7434: ``Heterotic and type I string dynamics from eleven dimensions,''
7435: Nucl.\ Phys.\ {\bf B 460}, 506, 1996;
7436: {\tt hep-th/9510209}.
7437:
7438: \bibitem{WT-RMP}
7439: W.\ Taylor,
7440: ``M(atrix) theory: matrix quantum mechanics as a fundamental theory,''
7441: Rev.\ Mod.\ Phys.\ {\bf 73}, 419, 2001;
7442: {\tt hep-th/0101126}
7443:
7444: \bibitem{agmoo}
7445: Aharony, O., S.~S.~Gubser, J.~Maldacena, H.~Ooguri, and Y.~Oz,
7446: ``Large N field theories, string theory and gravity,''
7447: Phys.\ Rept.\ {\bf 323}, 183, 2000;
7448: {\tt hep-th/9905111}.
7449:
7450: %~\cite{Witten:2000cn}
7451: %~\cite{Witten:1998cd}
7452: \bibitem{k-theory}
7453: E.~Witten,
7454: ``D-branes and K-theory,''
7455: JHEP {\bf 9812}, 019 (1998)
7456: {\tt hep-th/9810188}.
7457: %%CITATION = HEP-TH 9810188;%%
7458: %~\cite{Horava:1998jy}
7459: %\bibitem{Horava:1998jy}
7460: P.~Horava,
7461: ``Type IIA D-branes, K-theory, and matrix theory,''
7462: Adv.\ Theor.\ Math.\ Phys.\ {\bf 2}, 1373 (1999)
7463: {\tt hep-th/9812135}.
7464: %%CITATION = HEP-TH 9812135;%%
7465: %\bibitem{Witten:2000cn}
7466: E.~Witten,
7467: ``Overview of K-theory applied to strings,''
7468: Int.\ J.\ Mod.\ Phys.\ A {\bf 16}, 693 (2001)
7469: {\tt hep-th/0007175}.
7470: %%CITATION = HEP-TH 0007175;%%
7471:
7472:
7473: \bibitem{Veneziano}
7474: G.\ Veneziano, ``Construction of a crossing-symmetric, Regge-behaved
7475: amplitude for linearly rising trajectories,'' {\rm Nuovo Cim.} {\bf
7476: 57A} 190 (1968).
7477:
7478:
7479:
7480: \bibitem{NGS}
7481: D.\ J.\ Gross, A.\ Neveu, J.\ Scherk and J.\ H.\ Schwarz,
7482: ``Renormalization and unitarity in the dual resonance model,''
7483: \PR {\bf D2} (1970) 697.
7484:
7485: \bibitem{Lovelace}
7486: C.\ Lovelace, ``Pomeron form factors and dual Regge cuts,''
7487: \PL {\bf B34} (1971) 500.
7488:
7489: \bibitem{Douglas-Nekrasov}
7490: % lookup 0106048
7491: M.~R.~Douglas and N.~A.~Nekrasov,
7492: ``Noncommutative field theory,''
7493: Rev.\ Mod.\ Phys.\ {\bf 73}, 977 (2001),
7494: {\tt hep-th/0106048}.
7495:
7496:
7497: %~\cite{Kato:1982im}
7498: \bibitem{Kato:1982im}
7499: M.~Kato and K.~Ogawa,
7500: ``Covariant Quantization Of String Based On BRS Invariance,''
7501: Nucl.\ Phys.\ B {\bf 212}, 443 (1983).
7502: %%CITATION = NUPHA,B212,443;%%
7503:
7504: %~\cite{Siegel:1984xd}
7505: \bibitem{Siegel:1984xd}
7506: W.~Siegel,
7507: ``Covariantly Second Quantized String. 3,''
7508: Phys.\ Lett.\ B {\bf 149}, 162 (1984)
7509: [Phys.\ Lett.\ {\bf 151B}, 396 (1985)].
7510: %%CITATION = PHLTA,B149,162;%%
7511: %~\cite{Siegel:1984wx}
7512: %\bibitem{Siegel:1984wx}
7513: %W.~Siegel,
7514: ``Covariantly Second Quantized String. 2,''
7515: Phys.\ Lett.\ B {\bf 149}, 157 (1984)
7516: [Phys.\ Lett.\ B {\bf 151}, 391 (1985)].
7517: %%CITATION = PHLTA,B149,157;%%
7518:
7519: %~\cite{Siegel:1985tw}
7520: \bibitem{Siegel:1985tw}
7521: W.~Siegel and B.~Zwiebach,
7522: ``Gauge String Fields,''
7523: Nucl.\ Phys.\ B {\bf 263}, 105 (1986).
7524: %%CITATION = NUPHA,B263,105;%%
7525:
7526:
7527: %~\cite{Banks:1985ff}
7528: \bibitem{Banks:1985ff}
7529: T.~Banks and M.~E.~Peskin,
7530: ``Gauge Invariance Of String Fields,''
7531: Nucl.\ Phys.\ B {\bf 264}, 513 (1986).
7532: %%CITATION = NUPHA,B264,513;%%
7533:
7534: \bibitem{Witten-SFT}
7535: E.\ Witten, ``Non-commutative geometry and string field theory,''
7536: \NP {\bf B268} 253, (1986).
7537:
7538: %~\cite{Zwiebach:1992ie}
7539: \bibitem{Zwiebach:1992ie}
7540: B.~Zwiebach,
7541: ``Closed string field theory: Quantum action and the B-V master equation,''
7542: Nucl.\ Phys.\ B {\bf 390}, 33 (1993)
7543: {\tt hep-th/9206084}.
7544: %%CITATION = HEP-TH 9206084;%%
7545:
7546: %~\cite{Saadi:tb}
7547: \bibitem{Saadi:tb}
7548: M.~Saadi and B.~Zwiebach,
7549: ``Closed String Field Theory From Polyhedra,''
7550: Annals Phys.\ {\bf 192}, 213 (1989).
7551: %%CITATION = APNYA,192,213;%%
7552:
7553: %~\cite{Kugo:1989tk}
7554: \bibitem{Kugo:1989tk}
7555: T.~Kugo and K.~Suehiro,
7556: ``Nonpolynomial Closed String Field Theory: Action And Its Gauge Invariance,''
7557: Nucl.\ Phys.\ B {\bf 337}, 434 (1990). \hfill\break
7558: %%CITATION = NUPHA,B337,434;%%
7559: %~\cite{Kugo:1989aa}
7560: %\bibitem{Kugo:1989aa}
7561: T.~Kugo, H.~Kunitomo and K.~Suehiro,
7562: ``Nonpolynomial Closed String Field Theory,''
7563: Phys.\ Lett.\ B {\bf 226}, 48 (1989).
7564: %%CITATION = PHLTA,B226,48;%%
7565:
7566: %~\cite{Kaku:zw}
7567: \bibitem{Kaku:zw}
7568: M.~Kaku and J.~Lykken,
7569: ``Modular Invariant Closed String Field Theory,''
7570: Phys.\ Rev.\ D {\bf 38}, 3067 (1988).
7571: %%CITATION = PHRVA,D38,3067;%%
7572: %~\cite{Kaku:zv}
7573: %\bibitem{Kaku:zv}
7574: M.~Kaku,
7575: ``Geometric Derivation Of String Field Theory From First Principles:
7576: Closed Strings
7577: And Modular Invariance,'' Phys.\ Rev.\ D {\bf 38}, 3052 (1988).
7578: %%CITATION = PHRVA,D38,3052;%%
7579:
7580:
7581: %~\cite{Zwiebach:1997fe}
7582: \bibitem{Zwiebach:1997fe}
7583: B.~Zwiebach,
7584: ``Oriented open-closed string theory revisited,''
7585: Annals Phys.\ {\bf 267}, 193 (1998)
7586: {\tt hep-th/9705241}. \hfill\break
7587: %%CITATION = HEP-TH 9705241;%%
7588: %
7589: %~\cite{Zwiebach:1990qj}
7590: %\bibitem{Zwiebach:1990qj}
7591: B.~Zwiebach,
7592: ``Quantum Open String Theory With Manifest Closed String Factorization,''
7593: Phys.\ Lett.\ B {\bf 256}, 22 (1991).
7594:
7595: %~\cite{Witten:1986qs}
7596: \bibitem{Witten:1986qs}
7597: E.~Witten,
7598: ``Interacting Field Theory Of Open Superstrings,''
7599: Nucl.\ Phys.\ B {\bf 276}, 291 (1986).
7600: %%CITATION = NUPHA,B276,291;%%
7601:
7602: \bibitem{Wendt}
7603: C.\ Wendt, ``Scattering amplitudes and contact interactions in Witten's
7604: superstring field theory,'' \NP {\bf B314} (1989) 209.
7605:
7606: \bibitem{Greensite-Klinkhamer}
7607: J.\ Greensite and F.\ R.\ Klinkhamer, ``Superstring amplitudes and contact
7608: interactions,'' \PL {\bf B304} (1988) 108.
7609:
7610: %~\cite{Arefeva:1989cp}
7611: \bibitem{Arefeva:1989cp}
7612: I.~Y.~Arefeva, P.~B.~Medvedev and A.~P.~Zubarev,
7613: ``New Representation For String Field Solves The Consistence Problem For Open
7614: Superstring Field,'' Nucl.\ Phys.\ B {\bf 341}, 464 (1990).
7615: %%CITATION = NUPHA,B341,464;%%
7616:
7617: %~\cite
7618: \bibitem{Preitschopf:fc}
7619: C.~R.~Preitschopf, C.~B.~Thorn and S.~A.~Yost,
7620: ``Superstring Field Theory,''
7621: Nucl.\ Phys.\ B {\bf 337}, 363 (1990).
7622: %%CITATION = NUPHA,B337,363;%%
7623:
7624: \bibitem{Berkovits}
7625: N.~Berkovits,
7626: ``Super-Poincare invariant
7627: superstring field theory,'' \NP
7628: {\bf B450} (1995), 90,
7629: {\tt hep-th/9503099};
7630: % lookup 0105230
7631: ``Review of open superstring field theory,''
7632: {\tt hep-th/0105230};
7633: %N.~Berkovits,
7634: ``The Ramond sector of open superstring field theory,''
7635: JHEP {\bf 0111}, 047 (2001),
7636: {\tt hep-th/0109100}.
7637:
7638: \bibitem{Sen-Zwiebach}
7639: % lookup 9912249
7640: A.~Sen and B.~Zwiebach,
7641: ``Tachyon condensation in string field theory,''
7642: JHEP {\bf 0003}, 002 (2000),
7643: {\tt hep-th/9912249}.
7644:
7645:
7646: \bibitem{Sen-universality}
7647: % lookup 9911116
7648: A.~Sen,
7649: ``Universality of the tachyon potential,''
7650: JHEP {\bf 9912}, 027 (1999),
7651: {\tt hep-th/9911116}.
7652:
7653:
7654:
7655: \bibitem{Polchinski-TASI}
7656: % lookup 9611050
7657: J.~Polchinski,
7658: ``TASI Lectures on D-branes,''
7659: {\tt hep-th/9611050}.
7660:
7661:
7662:
7663: \bibitem{WT-Trieste}
7664: % lookup 9801182
7665: W.~Taylor,
7666: ``Trieste lectures on D-branes, gauge theory and M(atrices),''
7667: {\tt hep-th/9801182}.
7668:
7669:
7670:
7671: \bibitem{dkl}
7672: % lookup 9412184
7673: M.~J.~Duff, R.~R.~Khuri and J.~X.~Lu,
7674: ``String solitons,''
7675: Phys.\ Rept.\ {\bf 259}, 213 (1995),
7676: {\tt hep-th/9412184}.
7677:
7678:
7679:
7680: \bibitem{Polchinski}
7681: J.\ Polchinski, ``Dirichlet-Branes and
7682: Ramond-Ramond Charges,'' {\PRL }{\bf 75} (1995),
7683: 4724, {\tt hep-th/9510017}.
7684:
7685: \bibitem{Leigh}
7686: R.\ G.\ Leigh,
7687: ``Dirac-Born-Infeld action from Dirichlet
7688: sigma model,''
7689: \MPL {\bf A4} (1989), 2767.
7690:
7691: \bibitem{Witten-multiple}
7692: E.\ Witten, ``Bound States of Strings and
7693: p-Branes,'' \NP {\bf B460} (1996), 335, {\tt
7694: hep-th/9510135}.
7695:
7696: \bibitem{ordering-NBI}
7697: % lookup 9908105
7698: A.~A.~Tseytlin,
7699: ``Born-Infeld action, supersymmetry and string theory,''
7700: {\tt hep-th/9908105};
7701: see also the following for some recent developments and further references:
7702: % lookup 0208044
7703: P.~Koerber and A.~Sevrin,
7704: ``The non-abelian D-brane effective action through order alpha'**4,''
7705: JHEP {\bf 0210}, 046 (2002),
7706: {\tt hep-th/0208044};
7707: % lookup 0209064
7708: S.~Stieberger and T.~R.~Taylor,
7709: ``Non-Abelian Born-Infeld action and type I - heterotic duality. II:
7710: Nonrenormalization theorems,''
7711: Nucl.\ Phys.\ B {\bf 648}, 3 (2003),
7712: {\tt hep-th/0209064};
7713: % lookup 0210146
7714: D.~T.~Grasso,
7715: ``Higher order contributions to the effective action of N = 4 super
7716: Yang-Mills,''
7717: JHEP {\bf 0211}, 012 (2002),
7718: {\tt hep-th/0210146}.
7719:
7720:
7721:
7722: \bibitem{Mark-Wati-4}
7723: W.\ Taylor and M.\ Van Raamsdonk,
7724: ``Multiple D0-branes in weakly curved backgrounds,''
7725: \PL {\bf B558} (1999), 63,
7726: {\tt hep-th/9904095};
7727: % lookup 9910052
7728: ``Multiple Dp-branes in weak background fields,''
7729: \NP {\bf B573}, 703 (2000),
7730: {\tt hep-th/9910052}.
7731:
7732:
7733:
7734: \bibitem{Myers}
7735: R.\ C.\ Myers,
7736: ``Dielectric-branes,''
7737: {\it JHEP} {\bf 9912} (1999), 022,
7738: hep-th/9910053.
7739:
7740:
7741: \bibitem{Polchinski-string}
7742: Polchinski, J.\,
7743: {\em String theory} (Cambridge University Press,
7744: Cambridge, England, 1998).
7745:
7746: \bibitem{WT-T-duality}
7747: W.\ Taylor, ``D-brane Field Theory on Compact
7748: Spaces,'' \PL {\bf B394} (1997), 283;
7749: {\tt hep-th/9611042}.
7750:
7751: \bibitem{Banks-Susskind}
7752: % lookup 9511194
7753: T.~Banks and L.~Susskind,
7754: ``Brane - Antibrane Forces,''
7755: {\tt hep-th/9511194}.
7756:
7757:
7758:
7759: \bibitem{Hashimoto-Taylor}
7760: A.\ Hashimoto and W.\ Taylor, ``Fluctuation
7761: Spectra of Tilted and Intersecting D-branes from the
7762: Born-Infeld Action,''
7763: \NP {\bf B503} (1997), 193-219;
7764: {\tt hep-th/9703217}.
7765:
7766: \bibitem{gns}
7767: % lookup 9704006
7768: E.~Gava, K.~S.~Narain and M.~H.~Sarmadi,
7769: ``On the bound states of p- and (p+2),-branes,''
7770: Nucl.\ Phys.\ B {\bf 504}, 214 (1997),
7771: {\tt hep-th/9704006}.
7772:
7773: \bibitem{Morosov}
7774: A.~V.~Morosov,
7775: {\sl ``Classical Decay of a Non-supersymmetric Configuration of Two
7776: D-branes''},
7777: \PL {\bf B433} (1998), 291,
7778: {\tt hep-th/9803110}.
7779:
7780: \bibitem{Hashimoto-Nagaoka}
7781: K.~Hashimoto and S.~Nagaoka,
7782: {\sl ``Recombination of Intersecting D-branes by Local Tachyon
7783: Condensation,''}
7784: JHEP {\bf 0306}, 034 (2003), {\tt hep-th/0303204}.
7785:
7786:
7787:
7788:
7789:
7790:
7791:
7792: \bibitem{Bardakci-tachyon}
7793: K.\ Bardakci and M.\ B.\ Halpern, ``Explicit spontaneous breakdown in
7794: a dual model,'' \PR {\bf D10} (1974) 4230;
7795: K.\ Bardakci, ``Spontaneous symmetry breaking in the standard dual string
7796: model,'' \NP {\bf B133} (1978), 297.
7797:
7798: \bibitem{ks-open}
7799: V.\ A.\ Kostelecky and S.\ Samuel, ``On a nonperturbative vacuum for the open
7800: bosonic string,'' \NP {\bf B336} (1990), 263-296.
7801:
7802: % references: section 4
7803:
7804:
7805: \bibitem{lpp}
7806: A.\ Leclair, M.\ E.\ Peskin and C.\ R.\ Preitschopf,
7807: ``String field theory on the conformal plane (I),''
7808: \NP {\bf B317} (1989), 411-463.
7809:
7810: \bibitem{Thorn}
7811: C.\ Thorn, ``String field theory,''
7812: \PRP {\bf 175} (1989), 1.
7813:
7814: \bibitem{Gaberdiel-Zwiebach}
7815: M.\ R.\ Gaberdiel and B.\ Zwiebach,
7816: ``Tensor constructions of open string
7817: theories 1., 2.,''
7818: \NP {\bf B505} (1997), 569, {\tt hep-th/9705038};
7819: \PL {\bf B410} (1997), 151, {\tt hep-th/9707051}.
7820:
7821: \bibitem{gsw}
7822: Green, M.\ B., J.\ H.\ Schwarz, and E.\ Witten, {\em Superstring theory}
7823: (Cambridge University Press, Cambridge, England, 1987).
7824:
7825: \bibitem{Gross-Jevicki-12}
7826: D.\ J.\ Gross and A.\ Jevicki, ``Operator formulation of interacting string
7827: field theory (I), (II),'' \NP {\bf B283} (1987), 1; \NP {\bf B287}
7828: (1987), 225.
7829:
7830: \bibitem{efhm}
7831: % lookup 0105024
7832: I.~Ellwood, B.~Feng, Y.~H.~He and N.~Moeller,
7833: ``The identity string field and the tachyon vacuum,''
7834: JHEP {\bf 0107}, 016 (2001),
7835: {\tt hep-th/0105024}.
7836:
7837: \bibitem{Schnabl}
7838: %% lookup 0201095
7839: %% lookup 0202139
7840: M.~Schnabl,
7841: ``Wedge states in string field theory,''
7842: JHEP {\bf 0301}, 004 (2003),
7843: {\tt hep-th/0201095};
7844: ``Anomalous reparametrizations and butterfly states in string field theory,''
7845: Nucl.\ Phys.\ B {\bf 649}, 101 (2003),
7846: {\tt hep-th/0202139}.
7847:
7848:
7849:
7850:
7851: % references: section 5
7852:
7853:
7854: \bibitem{Rastelli-Zwiebach}
7855: % lookup 0006240
7856: L.~Rastelli and B.~Zwiebach,
7857: ``Tachyon potentials, star products and universality,''
7858: JHEP {\bf 0109}, 038 (2001),
7859: {\tt hep-th/0006240}.
7860:
7861:
7862:
7863: \bibitem{split}
7864: %\bibitem{bcnt}
7865: J.\ Bordes, Chan H.-M., L.\ Nellen, Tsou S.-T.,
7866: ``Half-string oscillator approach to string field theory,''
7867: \NP {\bf
7868: B351} (1991) 441;
7869: %\bibitem{aab}
7870: A.\ Abdurrahman, F.\ Anton and J.\ Bordes,
7871: ``Half string oscillator approach to string field theory (ghost sector
7872: 1, 2),''
7873: \NP {\bf
7874: B397} (1993) 260;
7875: \NP {\bf
7876: B411} (1994) 693;
7877: %\bibitem{Abdurrahman-Bordes}
7878: A.\ Abdurrahman and J.\ Bordes,
7879: ``The relationship between the comma theory and Witten's string field
7880: theory,'' \PR {\bf D 58}:086003 (1998);
7881: %\bibitem{rsz-3}
7882: L.~Rastelli, A.~Sen and B.~Zwiebach,
7883: ``Half-strings, projectors, and multiple D-branes in vacuum string field
7884: theory,''
7885: {\tt hep-th/0105058};
7886: %\bibitem{open-matrices}
7887: T.~Kawano and K.~Okuyama,
7888: ``Open string fields as matrices,''
7889: {\tt hep-th/0105129};
7890: % lookup 0105129
7891: % lookup 0107101
7892: %\bibitem{Furuuchi:2001df}
7893: K.~Furuuchi and K.~Okuyama,
7894: ``Comma vertex and string field algebra,''
7895: JHEP {\bf 0109}, 035 (2001)
7896: {\tt hep-th/0107101};
7897: % lookup 0110204
7898: %\bibitem{Moeller:2001ap}
7899: N.~Moeller,
7900: ``Some exact results on the matter star-product in the half-string
7901: formalism,''
7902: JHEP {\bf 0201}, 019 (2002)
7903: {\tt hep-th/0110204};
7904: % lookup 0304044
7905: %\bibitem{Erler:2003eq}
7906: T.~G.~Erler,
7907: ``A fresh look at midpoint singularities in the algebra of string fields,''
7908: {\tt hep-th/0304044};
7909: % lookup 0307148
7910: %\bibitem{Fuchs:2003wu}
7911: E.~Fuchs, M.~Kroyter and A.~Marcus,
7912: ``Continuous half-string representation of string field theory,''
7913: {\tt hep-th/0307148}.
7914:
7915: \bibitem{Gross-Taylor-I}
7916: % lookup 0105059
7917: D.~J.~Gross and W.~Taylor,
7918: ``Split string field theory I,''
7919: JHEP {\bf 0108}, 009 (2001),
7920: {\tt hep-th/0105059}.
7921:
7922: \bibitem{Gross-Taylor-II}
7923: % lookup 0106036
7924: D.~J.~Gross and W.~Taylor,
7925: ``Split string field theory II,''
7926: JHEP {\bf 0108}, 010 (2001),
7927: {\tt hep-th/0106036}.
7928:
7929:
7930:
7931:
7932: \bibitem{cst}
7933: E.\ Cremmer, A.\ Schwimmer and C.\ Thorn, ``The vertex function in Witten's
7934: formulation of string field theory'' \PL {\bf B179} 57 (1986).
7935:
7936: \bibitem{Samuel}
7937: S.\ Samuel, ``The physical and ghost vertices in Witten's string field
7938: theory,'' \PL {\bf B181} 255 (1986).
7939:
7940:
7941: \bibitem{Ohta}
7942: N.\ Ohta, ``Covariant interacting string field theory in the Fock
7943: space representation,''
7944: \PR {\bf D34} (1986), 3785;
7945: \PR {\bf D35} (1987), 2627~(E);
7946: S.\ P.\ de Alwis and N.\ Ohta,
7947: ``All Free String Theories are Theories of BRST Cohomology,''
7948: \PL {\bf B174} (1986) 388;
7949: S.\ P.\ de Alwis and N.\ Ohta,
7950: ``Fock Space Formulation of Gauge Invariant Interacting Field
7951: Theories of Open and Closed Strings,''
7952: \PL {\bf B188} (1987) 425.
7953:
7954:
7955: \bibitem{Shelton}
7956: J.\ Shelton, unpublished.
7957:
7958: \bibitem{Moyal}
7959: % lookup 0111281
7960: %\bibitem{Rastelli:2001hh}
7961: L.~Rastelli, A.~Sen and B.~Zwiebach,
7962: ``Star algebra spectroscopy,''
7963: JHEP {\bf 0203}, 029 (2002)
7964: {\tt hep-th/0111281};
7965: % lookup 0202087
7966: %\bibitem{Douglas:2002jm}
7967: M.~R.~Douglas, H.~Liu, G.~Moore and B.~Zwiebach,
7968: ``Open string star as a continuous Moyal product,''
7969: JHEP {\bf 0204}, 022 (2002)
7970: {\tt hep-th/0202087};
7971: % lookup 0202176
7972: % lookup 0203175
7973: % lookup 0203204
7974: % lookup 0204239
7975: % lookup 0205107
7976: % lookup 0306252
7977: %\bibitem{Feng:2002rm}
7978: B.~Feng, Y.~H.~He and N.~Moeller,
7979: ``The spectrum of the Neumann matrix with zero modes,''
7980: JHEP {\bf 0204}, 038 (2002)
7981: {\tt hep-th/0202176};
7982: %\bibitem{Feng:2002ib}
7983: B.~Feng, Y.~H.~He and N.~Moeller,
7984: ``Zeeman spectroscopy of the star algebra,''
7985: JHEP {\bf 0205}, 041 (2002)
7986: {\tt hep-th/0203175};
7987: %\bibitem{Chen:2002md}
7988: B.~Chen and F.~L.~Lin,
7989: ``Star spectroscopy in the constant B field background,''
7990: Nucl.\ Phys.\ B {\bf 637}, 199 (2002)
7991: {\tt hep-th/0203204};
7992: %\bibitem{Arefeva:2002jj}
7993: I.~Y.~Arefeva and A.~A.~Giryavets,
7994: ``Open superstring star as a continuous Moyal product,''
7995: JHEP {\bf 0212}, 074 (2002)
7996: {\tt hep-th/0204239};
7997: %\bibitem{Erler:2002nr}
7998: T.~G.~Erler,
7999: ``Moyal formulation of Witten's star product in the fermionic ghost sector,''
8000: {\tt hep-th/0205107};
8001: %\bibitem{Maccaferri:2003rz}
8002: C.~Maccaferri and D.~Mamone,
8003: ``Star democracy in open string field theory,''
8004: JHEP {\bf 0309}, 049 (2003)
8005: {\tt hep-th/0306252}.
8006:
8007:
8008: \bibitem{RSZ-2}
8009: % lookup 0102112
8010: L.~Rastelli, A.~Sen and B.~Zwiebach,
8011: ``Classical solutions in string field theory around the tachyon vacuum,''
8012: Adv.\ Theor.\ Math.\ Phys.\ {\bf 5}, 393 (2002),
8013: {\tt hep-th/0102112}.
8014:
8015:
8016:
8017:
8018:
8019:
8020: \bibitem{Giddings-Martinec}
8021: S.~B.~Giddings and E.~J.~Martinec,
8022: ``Conformal Geometry and String Field Theory,''
8023: \NP {\bf B278}, 91 (1986).
8024:
8025: \bibitem{gmw}
8026: S.~B.~Giddings, E.~J.~Martinec and E.~Witten,
8027: ``Modular Invariance In String Field Theory,''
8028: Phys.\ Lett.\ B {\bf 176}, 362 (1986).
8029: %%CITATION = PHLTA,B176,362;%%
8030:
8031: \bibitem{Zwiebach-proof}
8032: B.~Zwiebach,
8033: ``A Proof That Witten's Open String Theory Gives A Single Cover Of
8034: Moduli Space,''
8035: Commun.\ Math.\ Phys.\ {\bf 142}, 193 (1991).
8036: %%CITATION = CMPHA,142,193;%%
8037:
8038: \bibitem{Aref'eva}
8039: % lookup 0011117
8040: I.~Y.~Aref'eva, A.~S.~Koshelev, D.~M.~Belov and P.~B.~Medvedev,
8041: ``Tachyon condensation in cubic superstring field theory,''
8042: Nucl.\ Phys.\ B {\bf 638}, 3 (2002),
8043: {\tt hep-th/0011117};
8044: % lookup 0011208
8045: G.~Bandelloni and S.~Lazzarini,
8046: ``The geometry of W3 algebra: A twofold way for the rebirth of a symmetry,''
8047: Nucl.\ Phys.\ B {\bf 594}, 477 (2001),
8048: {\tt hep-th/0011208}.
8049:
8050:
8051: \bibitem{senunpub}
8052: A.~Sen, unpublished.
8053:
8054: \bibitem{superstring}
8055: N.~Berkovits, A.~Sen and B.~Zwiebach,
8056: ``Tachyon condensation in superstring field theory,''
8057: Nucl.\ Phys.\ B {\bf 587}, 147 (2000),
8058: {\tt hep-th/0002211};
8059:
8060:
8061:
8062: \bibitem{Okawa}
8063: % lookup 0204012
8064: Y.~Okawa,
8065: ``Open string states and D-brane tension from vacuum string field theory,''
8066: JHEP {\bf 0207}, 003 (2002),
8067: {\tt hep-th/0204012}.
8068:
8069: \bibitem{Coletti-Sigalov-Taylor}
8070: % lookup 0306041
8071: E.~Coletti, I.~Sigalov and W.~Taylor,
8072: ``Abelian and nonabelian vector field effective actions from string
8073: field theory,''
8074: JHEP {\bf 0309}, 050 (2003)
8075: {\tt hep-th/0306041}.
8076:
8077:
8078: %~\cite{Zwiebach:2000vc}
8079: \bibitem{Zwiebach:2000vc}
8080: B.~Zwiebach,
8081: ``Trimming the tachyon string field with SU(1,1),''
8082: {\tt hep-th/0010190}.
8083: %%CITATION = HEP-Th 0010190;%%
8084:
8085:
8086: \bibitem{Moeller-Taylor}
8087: % lookup 0002237
8088: N.~Moeller and W.~Taylor,
8089: ``Level truncation and the tachyon in open bosonic string field theory,''
8090: Nucl.\ Phys.\ B {\bf 583}, 105 (2000),
8091: {\tt hep-th/0002237}.
8092:
8093: \bibitem{Gaiotto-Rastelli-strings}
8094: D.\ Gaiotto and L.\ Rastelli,
8095: ``Progress in open string field theory,''
8096: Presentation by L.\ Rastelli at Strings 2002, Cambridge, England;
8097: {\tt http://www.damtp.cam.ac.uk/strings02/avt/rastelli/}.
8098:
8099: \bibitem{gr-analysis}
8100: % lookup 0211012
8101: D.~Gaiotto and L.~Rastelli,
8102: ``Experimental string field theory,''
8103: {\tt hep-th/0211012}.
8104:
8105:
8106:
8107: \bibitem{WT-perturbative}
8108: W.~Taylor,
8109: ``Perturbative diagrams in string field theory,''
8110: {\tt hep-th/0207132}.
8111:
8112: \bibitem{Beccaria-Rampino}
8113: % lookup 0308059
8114: M.~Beccaria and C.~Rampino,
8115: ``Level truncation and the quartic tachyon coupling,''
8116: JHEP {\bf 0310}, 047 (2003)
8117: {\tt hep-th/0308059}.
8118:
8119:
8120: \bibitem{Ellis-Karliner-WT}
8121: J.\ Ellis, M.\ Karliner and W.\ Taylor, unpublished.
8122:
8123:
8124: \bibitem{WT-Pade}
8125: W.~Taylor,
8126: ``A Perturbative Analysis of Tachyon Condensation,''
8127: {\tt hep-th/0208149}.
8128:
8129: \bibitem{Ellwood-Taylor-gauge}
8130: I.\ Ellwood and W.\ Taylor,
8131: ``Gauge invariance and tachyon condensation in open string field
8132: theory,''
8133: {\tt hep-th/0105156}.
8134:
8135: \bibitem{Moeller-sz}
8136: N.~Moeller, A.~Sen and B.~Zwiebach,
8137: ``D-branes as tachyon lumps in string field theory,''
8138: JHEP {\bf 0008}, 039 (2000),
8139: {\tt hep-th/0005036}.
8140:
8141:
8142: \bibitem{djmt}
8143: R.~de Mello Koch, A.~Jevicki, M.~Mihailescu and R.~Tatar,
8144: ``Lumps and p-branes in open string field theory,''
8145: Phys.\ Lett.\ B {\bf 482}, 249 (2000),
8146: {\tt hep-th/0003031}.
8147:
8148: \bibitem{Harvey-Kraus}
8149: J.~A.~Harvey and P.~Kraus,
8150: ``D-branes as unstable lumps in bosonic open string field theory,''
8151: JHEP {\bf 0004}, 012 (2000),
8152: {\tt hep-th/0002117};
8153:
8154:
8155:
8156: \bibitem{Zwiebach-toy}
8157: % lookup 0008227
8158: B.~Zwiebach,
8159: ``A solvable toy model for tachyon condensation in string field theory,''
8160: JHEP {\bf 0009}, 028 (2000),
8161: {\tt hep-th/0008227}.
8162:
8163: %~\cite{Minahan:2000ff}
8164: \bibitem{Minahan:2000ff}
8165: J.~A.~Minahan and B.~Zwiebach,
8166: ``Field theory models for tachyon and gauge field string dynamics,''
8167: JHEP {\bf 0009}, 029 (2000)
8168: {\tt hep-th/0008231}.
8169: %%CITATION = HEP-TH 0008231;%%
8170:
8171:
8172: \bibitem{Ellwood-Taylor-2}
8173: I.\ Ellwood and W.\ Taylor, unpublished.
8174:
8175: \bibitem{Sen-Zwiebach-translate}
8176: % lookup 0007153
8177: A.~Sen and B.~Zwiebach,
8178: ``Large marginal deformations in string field theory,''
8179: JHEP {\bf 0010}, 009 (2000),
8180: {\tt hep-th/0007153}.
8181:
8182:
8183:
8184: \bibitem{Kluson}
8185: % lookup 0203089
8186: J.~Kluson,
8187: ``Marginal deformations in the open bosonic string field theory for N
8188: D0-branes,''
8189: {\tt hep-th/0203089};
8190: % lookup 0209255
8191: ``Exact solutions in open bosonic string field theory and marginal
8192: deformation in CFT,''
8193: {\tt hep-th/0209255}.
8194:
8195:
8196:
8197:
8198:
8199:
8200:
8201:
8202:
8203: % section 6
8204:
8205:
8206:
8207:
8208:
8209:
8210:
8211:
8212:
8213:
8214:
8215:
8216: \bibitem{RSZ}
8217: % lookup 0012251
8218: L.~Rastelli, A.~Sen and B.~Zwiebach,
8219: ``String field theory around the tachyon vacuum,''
8220: Adv.\ Theor.\ Math.\ Phys.\ {\bf 5}, 353 (2002),
8221: {\tt hep-th/0012251}.
8222:
8223:
8224: %~\cite{Sen:1993mh}
8225: \bibitem{Sen:1993mh}
8226: A.~Sen and B.~Zwiebach,
8227: ``A Proof of local background independence of classical closed string field
8228: theory,'' Nucl.\ Phys.\ B {\bf 414}, 649 (1994)
8229: {\tt hep-th/9307088}.
8230: %%CITATION = HEP-TH 9307088;%%
8231:
8232: %~\cite{Kawano:1999fw}
8233: \bibitem{Kawano:1999fw}
8234: T.~Kawano and T.~Takahashi,
8235: ``Open string field theory on noncommutative space,''
8236: Prog.\ Theor.\ Phys.\ {\bf 104}, 459 (2000)
8237: {\tt hep-th/9912274}.
8238: %%CITATION = HEP-TH 9912274;%%
8239:
8240: \bibitem{Srednicki}
8241: M.\ Fisk and M.\ Srednicki,
8242: ``Magnetic string fields,''
8243: Phys.\ Lett.\ B {\bf 313}, 308 (1989).
8244:
8245:
8246: \bibitem{Giusto-Imbimbo}
8247: S.~Giusto and C.~Imbimbo,
8248: ``Physical states at the tachyonic vacuum of open string field theory,''
8249: {\tt hep-th/0309164}.
8250:
8251:
8252:
8253:
8254:
8255:
8256: \bibitem{Ellwood-Taylor-spectrum}
8257: % lookup 0103085
8258: I.~Ellwood and W.~Taylor,
8259: ``Open string field theory without open strings,''
8260: Phys.\ Lett.\ B {\bf 512}, 181 (2001),
8261: {\tt hep-th/0103085}.
8262:
8263:
8264:
8265: \bibitem{Minahan}
8266: % lookup 0105312
8267: J.~A.~Minahan,
8268: ``Quantum corrections in p-adic string theory,''
8269: {\tt hep-th/0105312}.
8270:
8271: \bibitem{CUBIC}
8272: G.~T.~Horowitz, J.~Lykken, R.~Rohm and A.~Strominger,
8273: ``A Purely Cubic Action For String Field Theory,"
8274: Phys.\ Rev.\ Lett.\ {\bf 57}, 283 (1986).
8275: %%CITATION = PRLTA,57,283;%%
8276:
8277:
8278: \bibitem{Hata-Kawano}
8279: % lookup 0108150
8280: H.~Hata and T.~Kawano,
8281: ``Open string states around a classical solution in vacuum string
8282: field theory,''
8283: JHEP {\bf 0111}, 038 (2001),
8284: {\tt hep-th/0108150}.
8285:
8286:
8287: %~\cite{Okuyama:2002tw}
8288: \bibitem{Okuyama:2002tw}
8289: K.~Okuyama,
8290: ``Ratio of tensions from vacuum string field theory,''
8291: JHEP {\bf 0203}, 050 (2002)
8292: {\tt hep-th/0201136}.
8293: %%CITATION = HEP-TH 0201136;%%
8294: %\bibitem{Okuyama}
8295: % lookup 0201015
8296: K.~Okuyama,
8297: ``Ghost kinetic operator of vacuum string field theory,''
8298: JHEP {\bf 0201}, 027 (2002),
8299: {\tt hep-th/0201015}.
8300:
8301:
8302: \bibitem{VSFT-tension}
8303: % lookup 0111034
8304: H.~Hata and S.~Moriyama,
8305: ``Observables as twist anomaly in vacuum string field theory,''
8306: JHEP {\bf 0201}, 042 (2002),
8307: {\tt hep-th/0111034}.
8308:
8309: \bibitem{0111153}
8310: % lookup 0111153
8311: L.~Rastelli, A.~Sen and B.~Zwiebach,
8312: ``A note on a proposal for the tachyon state in vacuum string field theory,''
8313: JHEP {\bf 0202}, 034 (2002),
8314: {\tt hep-th/0111153}.
8315: % lookup 0201177
8316: H.~Hata, S.~Moriyama and S.~Teraguchi,
8317: ``Exact results on twist anomaly,''
8318: JHEP {\bf 0202}, 036 (2002),
8319: {\tt hep-th/0201177}.
8320:
8321:
8322: %~\cite{Okawa1}
8323: \bibitem{okawatbp1}
8324: Y.~Okawa,
8325: ``Some exact computations on the twisted butterfly state
8326: in string field theory,''
8327: {\tt hep-th/0310264}.
8328: %%CITATION = HEP-TH 0310264;%%
8329:
8330:
8331: \bibitem{RSZ-closed}
8332: % lookup 0111129
8333: D.~Gaiotto, L.~Rastelli, A.~Sen and B.~Zwiebach,
8334: ``Ghost structure and closed strings in vacuum string field theory,''
8335: {\tt hep-th/0111129}.
8336:
8337:
8338:
8339: \bibitem{RSZ-projectors}
8340: % lookup 0202151
8341: D.~Gaiotto, L.~Rastelli, A.~Sen and B.~Zwiebach,
8342: ``Star algebra projectors,''
8343: JHEP {\bf 0204}, 060 (2002),
8344: {\tt hep-th/0202151}.
8345:
8346: %~\cite{Rastelli:2001vb}
8347: \bibitem{Rastelli:2001vb}
8348: L.~Rastelli, A.~Sen and B.~Zwiebach,
8349: ``Boundary CFT construction of D-branes in vacuum string field theory,''
8350: JHEP {\bf 0111}, 045 (2001)
8351: {\tt hep-th/0105168}.
8352: %%CITATION = HEP-TH ;%%
8353:
8354: \bibitem{Kostelecky-Potting}
8355: V.~A.~Kostelecky and R.~Potting,
8356: ``Analytical construction of a nonperturbative vacuum for the open
8357: bosonic string,''
8358: Phys.\ Rev.\ D {\bf 63}, 046007 (2001),
8359: {\tt hep-th/0008252}.
8360:
8361: %~\cite{Okuda:2002fj}
8362: \bibitem{Okuda:2002fj}
8363: T.~Okuda,
8364: ``The equality of solutions in vacuum string field theory,''
8365: Nucl.\ Phys.\ B {\bf 641}, 393 (2002)
8366: {\tt hep-th/0201149}.
8367: %%CITATION = HEP-TH 0201149;%%
8368:
8369: %~\cite{Fuchs:2002zz}
8370: \bibitem{Fuchs:2002zz}
8371: E.~Fuchs, M.~Kroyter and A.~Marcus,
8372: ``Squeezed state projectors in string field theory,''
8373: JHEP {\bf 0209}, 022 (2002)
8374: {\tt hep-th/0207001}.
8375: %%CITATION = HEP-TH 0207001;%%
8376:
8377:
8378:
8379: \bibitem{gms}
8380: R.~Gopakumar, S.~Minwalla and A.~Strominger,
8381: ``Noncommutative solitons,''
8382: JHEP {\bf 0005}, 020 (2000),
8383: {\tt hep-th/0003160}.
8384:
8385: \bibitem{Moore-Taylor}
8386: % lookup 0111069
8387: G.~Moore and W.~Taylor,
8388: ``The singular geometry of the sliver,''
8389: JHEP {\bf 0201}, 004 (2002),
8390: {\tt hep-th/0111069}.
8391:
8392:
8393: % B fields and VSFT
8394: \bibitem{b-VSFT}
8395: % lookup 0201060
8396: % lookup 0203188
8397: % lookup 0204233
8398: % lookup 0207044
8399: %~\cite{Bonora:2002pu}
8400: %\bibitem{Bonora:2002pu}
8401: L.~Bonora, D.~Mamone and M.~Salizzoni,
8402: ``B field and squeezed states in vacuum string field theory,''
8403: Nucl.\ Phys.\ B {\bf 630}, 163 (2002)
8404: {\tt hep-th/0201060};
8405: %\bibitem{Bonora:2002iq}
8406: L.~Bonora, D.~Mamone and M.~Salizzoni,
8407: ``Vacuum string field theory with B field,''
8408: JHEP {\bf 0204}, 020 (2002)
8409: {\tt hep-th/0203188};
8410: %\bibitem{Chen:2002jd}
8411: B.~Chen and F.~L.~Lin,
8412: ``D-branes as GMS solitons in vacuum string field theory,''
8413: Phys.\ Rev.\ D {\bf 66}, 126001 (2002)
8414: {\tt hep-th/0204233};
8415: %than\bibitem{Bonora:2002rn}
8416: L.~Bonora, D.~Mamone and M.~Salizzoni,
8417: ``Vacuum string field theory ancestors of the GMS solitons,''
8418: JHEP {\bf 0301}, 013 (2003)
8419: {\tt hep-th/0207044}.
8420:
8421:
8422: \bibitem{RSZ-3}
8423: % lookup 0105058
8424: L.~Rastelli, A.~Sen and B.~Zwiebach,
8425: ``Half strings, projectors, and multiple D-branes in vacuum string
8426: field theory,''
8427: JHEP {\bf 0111}, 035 (2001),
8428: {\tt hep-th/0105058}.
8429:
8430:
8431:
8432: %~\cite{Okawa2}
8433: \bibitem{okawatbp2}
8434: Y.~Okawa,
8435: ``Solving Witten's string field theory
8436: using the butterfly state,'' {\tt hep-th/0311115}.
8437: %%CITATION = NONE;%%
8438:
8439:
8440:
8441:
8442:
8443: \bibitem{fgst}
8444: D.~Z.~Freedman, S.~B.~Giddings, J.~A.~Shapiro and C.~B.~Thorn,
8445: ``The Nonplanar One Loop Amplitude In Witten's String Field Theory,''
8446: Nucl.\ Phys.\ B {\bf 298}, 253 (1988).
8447: %%CITATION = NUPHA,B298,253;%%
8448:
8449: \bibitem{Ellwood-Shelton-Taylor}
8450: % lookup 0304259
8451: %\bibitem{Ellwood:2003xc}
8452: I.~Ellwood, J.~Shelton and W.~Taylor,
8453: ``Tadpoles and closed string backgrounds in open string field theory,''
8454: JHEP {\bf 0307}, 059 (2003)
8455: {\tt hep-th/0304259}.
8456:
8457:
8458:
8459: \bibitem{Strominger-closed}
8460: A.\ Strominger,
8461: ``Closed strings in open string field theory,''
8462: {\PRL} {\bf 58} 629 (1987).
8463:
8464: \bibitem{Shapiro-Thorn}
8465: J.~A.~Shapiro and C.~B.~Thorn,
8466: ``BRST invariant transitions between open and closed strings,''
8467: \PR {\bf D36} 432 (1987);
8468: ``closed string-open string transitions in Witten's string field
8469: theory,''
8470: Phys.\ Lett.\ B {\bf 194}, 43 (1987).
8471:
8472:
8473: \bibitem{Gerasimov-Shatashvili}
8474: % lookup 0011009
8475: A.~A.~Gerasimov and S.~L.~Shatashvili,
8476: ``Stringy Higgs mechanism and the fate of open strings,''
8477: JHEP {\bf 0101}, 019 (2001),
8478: {\tt hep-th/0011009}.
8479: % lookup 0105076
8480: S.~L.~Shatashvili,
8481: ``On field theory of open strings, tachyon condensation and closed strings,''
8482: {\tt hep-th/0105076}.
8483:
8484:
8485:
8486:
8487: \bibitem{Hashimoto-Itzhaki}
8488: % lookup 0111092
8489: A.~Hashimoto and N.~Itzhaki,
8490: ``Observables of string field theory,''
8491: JHEP {\bf 0201}, 028 (2002),
8492: {\tt hep-th/0111092}.
8493:
8494: \bibitem{Ambjorn:2002im}
8495: J.~Ambjorn and R.~A.~Janik,
8496: ``Interpolating between open and closed strings - a BSFT approach,''
8497: Phys.\ Lett.\ B {\bf 538}, 189 (2002)
8498: {\tt hep-th/0203185}.
8499:
8500:
8501: %~\cite{Drukker:2002ct}
8502: \bibitem{Drukker:2002ct}
8503: N.~Drukker,
8504: ``Closed string amplitudes from gauge fixed string field theory,''
8505: Phys.\ Rev.\ D {\bf 67}, 126004 (2003)
8506: {\tt hep-th/0207266}.
8507: %%CITATION = HEP-TH 0207266;%%
8508:
8509: %~\cite{Alishahiha:2002as}
8510: \bibitem{Alishahiha:2002as}
8511: M.~Alishahiha and M.~R.~Garousi,
8512: ``Gauge invariant operators and closed string scattering in open string field
8513: theory,'' Phys.\ Lett.\ B {\bf 536}, 129 (2002)
8514: {\tt hep-th/0201249}.
8515: %%CITATION = HEP-TH 0201249;%%
8516:
8517: %~\cite{Takahashi:2003kq}
8518: \bibitem{Takahashi:2003kq}
8519: T.~Takahashi and S.~Zeze,
8520: ``Closed string amplitudes in open string field theory,''
8521: JHEP {\bf 0308}, 020 (2003)
8522: {\tt hep-th/0307173}.
8523: %%CITATION = HEP-TH 0307173;%%
8524:
8525:
8526: \bibitem{landscape}
8527: % lookup 0302219
8528: %\bibitem{Susskind:2003kw}
8529: L.~Susskind,
8530: ``The anthropic landscape of string theory,''
8531: {\tt hep-th/0302219}.
8532:
8533:
8534: %~\cite{Ohmori:2003vq}
8535: \bibitem{Ohmori:2003vq}
8536: K.~Ohmori,
8537: ``Level-expansion analysis in NS superstring field theory revisited,''
8538: {\tt hep-th/0305103}.
8539: %%CITATION = HEP-TH 0305103;%
8540:
8541:
8542:
8543: \bibitem{smet}
8544: P.~De Smet and J.~Raeymaekers,
8545: ``Level four approximation to the tachyon potential in superstring
8546: field theory,''
8547: JHEP {\bf 0005}, 051 (2000),
8548: {\tt hep-th/0003220};
8549: %\bibitem{Iqbal:2000st}
8550: A.~Iqbal and A.~Naqvi,
8551: ``Tachyon condensation on a non-BPS D-brane,'',
8552: {\tt hep-th/0004015};
8553: P.~De Smet and J.~Raeymaekers,
8554: ``The tachyon potential in Witten's superstring field theory,''
8555: JHEP {\bf 0008}, 020 (2000),
8556: {\tt hep-th/0004112};
8557:
8558:
8559:
8560: %~\cite{Marino:2001ny}
8561: \bibitem{Marino:2001ny}
8562: %\bibitem{Lechtenfeld:2000qj}
8563: O.~Lechtenfeld and A.~D.~Popov,
8564: ``On the integrability of covariant field theory for open N = 2 strings,''
8565: Phys.\ Lett.\ B {\bf 494}, 148 (2000)
8566: {\tt hep-th/0009144};
8567: %~\cite{Kluson:2001kk}
8568: %\bibitem{Kluson:2001kk}
8569: J.~Kluson,
8570: ``Proposal for background independent Berkovits' superstring field theory,''
8571: JHEP {\bf 0107}, 039 (2001)
8572: {\tt hep-th/0106107}.
8573: %%CITATION = HEP-TH 0106107;%%
8574: %~\cite{Arefeva:2001ke}
8575: %\bibitem{Arefeva:2001ke}
8576: I.~Y.~Arefeva, A.~A.~Giryavets and P.~B.~Medvedev,
8577: ``NS matter sliver,''
8578: Phys.\ Lett.\ B {\bf 532}, 291 (2002)
8579: {\tt hep-th/0112214}.
8580: %%CITATION = HEP-TH 0112214;%%
8581: M.~Marino and R.~Schiappa,
8582: ``Towards vacuum superstring field theory: The supersliver,''
8583: J.\ Math.\ Phys.\ {\bf 44}, 156 (2003)
8584: {\tt hep-th/0112231};
8585: %%CITATION = HEP-TH 0112231;%%
8586: %~\cite{Ohmori:2002ah}
8587: %\bibitem{Ohmori:2002ah}
8588: K.~Ohmori,
8589: ``Comments on solutions of vacuum superstring field theory,''
8590: JHEP {\bf 0204}, 059 (2002)
8591: {\tt hep-th/0204138}.
8592: %%CITATION = HEP-TH 0204138;%%
8593: %\bibitem{Lechtenfeld:2002cu}
8594: O.~Lechtenfeld, A.~D.~Popov and S.~Uhlmann,
8595: ``Exact solutions of Berkovits' string field theory,''
8596: Nucl.\ Phys.\ B {\bf 637}, 119 (2002)
8597: {\tt hep-th/0204155};
8598: %~\cite{Ohmori:2002kj}
8599: %\bibitem{Ohmori:2002kj}
8600: K.~Ohmori,
8601: ``On ghost structure of vacuum superstring field theory,''
8602: Nucl.\ Phys.\ B {\bf 648}, 94 (2003)
8603: {\tt hep-th/0208009};
8604: A.~Kling, O.~Lechtenfeld, A.~D.~Popov and S.~Uhlmann,
8605: ``On nonperturbative solutions of superstring field theory,''
8606: Phys.\ Lett.\ B {\bf 551}, 193 (2003)
8607: {\tt hep-th/0209186};
8608: %%CITATION = HEP-TH 0208009;%%
8609: % lookup 0212055
8610: %\bibitem{Koshelev:2002wc}
8611: A.~Koshelev,
8612: ``Solutions of vacuum superstring field theory,''
8613: {\tt hep-th/0212055};
8614: %\bibitem{Kling:2002ht}
8615: %\bibitem{Kling:2002vi}
8616: A.~Kling, O.~Lechtenfeld, A.~D.~Popov and S.~Uhlmann,
8617: ``Solving string field equations: New uses for old tools,''
8618: Fortsch.\ Phys.\ {\bf 51}, 775 (2003)
8619: {\tt hep-th/0212335}.
8620:
8621:
8622: \bibitem{Moeller:2002vx}
8623: N.~Moeller and B.~Zwiebach,
8624: ``Dynamics with infinitely many time derivatives and rolling tachyons,''
8625: JHEP {\bf 0210}, 034 (2002)
8626: {\tt hep-th/0207107}.
8627: %%CITATION = HEP-TH 0207107;%%
8628:
8629: %~\cite{Fujita:2003ex}
8630: \bibitem{Fujita:2003ex}
8631: M.~Fujita and H.~Hata,
8632: ``Time dependent solution in cubic string field theory,''
8633: JHEP {\bf 0305}, 043 (2003)
8634: {\tt hep-th/0304163}.
8635: %%CITATION = HEP-TH 0304163;%%
8636:
8637:
8638:
8639:
8640:
8641:
8642:
8643:
8644: \end{thebibliography}
8645:
8646: \end{document}
8647: