1: \documentclass[letterpaper]{JHEP3}
2: \usepackage{amssymb,amsfonts}
3: \usepackage{cite}
4: \usepackage{epsfig}
5: \newcommand{\trace}{\mbox{Tr}}
6: \newcommand{\bra}[1]{\langle #1|}
7: \newcommand{\ket}[1]{|#1\rangle}
8: \newcommand{\R}{{\bf R}}
9: \newcommand{\Z}{{\bf Z}}
10: \newcommand{\BU}{{\bf U}}
11: \newcommand{\CD}{{\cal D}}
12: \newcommand{\CF}{{\cal F}}
13: \newcommand{\CG}{{\cal G}}
14: \newcommand{\CI}{{\cal I}}
15: \newcommand{\CJ}{{\cal J}}
16: \newcommand{\CO}{{\cal O}}
17: \newcommand{\CT}{{\cal T}}
18: \newcommand{\CV}{{\cal V}}
19: \newcommand{\CZ}{{\cal Z}}
20: \newcommand{\CDX}{{\cal D}_{\!\!_{X^+}}}
21:
22: \newcommand{\p}{\partial}
23: \newcommand{\QB}{Q_{\rm BRST}}
24: \renewcommand{\bar}[1]{\overline{#1}}
25: \renewcommand{\tilde}[1]{\widetilde{#1}}
26: \newcommand{\be}{\begin{equation}}
27: \newcommand{\ee}{\end{equation}}
28: \newcommand{\bea}{\begin{eqnarray}}
29: \newcommand{\eea}{\end{eqnarray}}
30: \newcommand{\comment}[1]{{\bf (}{\tt #1}{\bf )}}
31: \usepackage{graphicx}
32: \usepackage{latexsym}
33: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34: \title{M-Theory Through the Looking Glass:\\
35: Tachyon Condensation in the $E_8$ Heterotic String}
36: %
37: \author{Petr Ho\v{r}ava and Cynthia A. Keeler\\
38: Berkeley Center for Theoretical Physics and Department of Physics\\
39: University of California, Berkeley, CA, 94720-7300\\
40: and\\
41: Theoretical Physics Group, Lawrence Berkeley National Laboratory\\
42: Berkeley, CA 94720-8162, USA}
43: %
44: \abstract{We study the spacetime decay to nothing in string theory and
45: M-theory. First we recall a nonsupersymmetric version of heterotic M-theory,
46: in which bubbles of nothing -- connecting the two $E_8$ boundaries by a throat
47: -- are expected to be nucleated. We argue that the fate of this system should
48: be addressed at weak string coupling, where the nonperturbative instanton
49: instability is expected to turn into a perturbative tachyonic one.
50: We identify the unique string theory that could describe this process: The
51: heterotic model with one $E_8$ gauge group and a singlet tachyon. We then use
52: worldsheet methods to study the tachyon condensation in the NSR formulation
53: of this model, and show that it induces a worldsheet super-Higgs effect.
54: The main theme of our analysis is the possibility of making meaningful
55: alternative gauge choices for worldsheet supersymmetry, in place of
56: the conventional superconformal gauge. We show in a version of unitary gauge
57: how the worldsheet gravitino assimilates the goldstino and becomes dynamical.
58: This picture clarifies recent results of Hellerman and Swanson. We also
59: present analogs of $R_\xi$ gauges, and note the importance of logarithmic CFT
60: in the context of tachyon condensation.}
61: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
62: \begin{document}
63:
64: \section{Introduction}
65:
66: The motivation for this paper is to further the studies of time-dependent
67: backgrounds in string theory. In particular, we concentrate on the problem
68: of closed-string tachyon condensation, and its hypothetical relation to the
69: ``spacetime decay to nothing.''
70:
71: Open-string tachyon condensation is now relatively well-understood
72: (see, {\it e.g.}, \cite{sen,headrick} for reviews), as
73: a description of D-brane decay into the vacuum (or to lower-dimensional stable
74: defects). On the other hand, the problem of the bulk closed-string tachyon
75: condensation appears related to a much more dramatic instability in which the
76: spacetime itself decays, or at least undergoes some other extensive change
77: indicating that the system is far from equilibrium. In the spacetime
78: supergravity approximation, this phenomenon has been linked to nonperturbative
79: instabilities due to the nucleation of ``bubbles of nothing'' \cite{witten}.
80: One of the first examples studied in the string and M-theory literature was
81: the nonsupersymmetric version of heterotic M-theory \cite{fh}, in which the
82: two $E_8$ boundaries of eleven-dimensional spacetime carry opposite relative
83: orientation and consequently break complementary sets of sixteen
84: supercharges. At large separation between the boundaries, this system has an
85: instanton solution that nucleates ``bubbles of nothing.'' In eleven
86: dimensions, the nucleated bubbles are smooth throats connecting the two
87: boundaries; the ``nothing'' phase is thus the phase ``on the other side''
88: of the spacetime boundary.
89:
90: In addition to this effect, the boundaries are attracted to each other by a
91: Casimir force which drives the system to weak string coupling, suggesting some
92: weakly coupled heterotic string description in ten dimensions. In the regime
93: of weak string coupling, we expect the originally nonperturbative instability
94: of the heterotic M-theory background to turn into a perturbative tachyonic
95: one.
96:
97: We claim that there is a unique viable candidate for describing this
98: system at weak string coupling: The tachyonic heterotic string with one
99: copy of $E_8$ gauge symmetry, and a singlet tachyon. In this paper, we
100: study in detail the worldsheet theory of this model -- in the NSR formalism
101: with local worldsheet $(0,1)$ supersymmetry -- when the tachyon develops a
102: condensate that grows exponentially along a lightcone direction $X^+$. There
103: is a close similarity between this background and the class of backgrounds
104: studied recently by Hellerman and Swanson \cite{hs1,hs2,hs3,hs4}.%
105: %
106: \footnote{Similar spacetime decay has also been seen in solutions of
107: noncritical string theory in $1+1$ dimensions \cite{Karczmarek}, and
108: noncritical M-theory in $2+1$ dimensions \cite{hk1}.}
109: %
110: The main novelty of our approach is
111: the use of alternative gauge choices for worldsheet supersymmetry, replacing
112: the traditional superconformal gauge. We show that the worldsheet dynamics
113: of spacetime tachyon condensation involves a super-Higgs mechanism, and
114: its picture simplifies considerably in our alternative gauge.
115:
116: Our main results were briefly reported in \cite{pixie}; in the present paper,
117: we elaborate on the conjectured connection to spacetime decay in heterotic
118: M-theory, and provide more details of the worldsheet theory of tachyon
119: condensation, including the analysis of the super-Higgs mechanism and its
120: compatibility with conformal invariance.
121:
122: Section~2 reviews the nonsupersymmetric version of heterotic M-theory, as a
123: simple configuration that exhibits the ``spacetime decay to nothing.'' We
124: argue that the dynamics of this instability should be studied at weak
125: string coupling, and advocate the role of the tachyonic $E_8$ heterotic
126: model as a unique candidate for this weakly coupled description of the decay.
127: In Section~3, we review some of the worldsheet structure of the tachyonic
128: $E_8$ heterotic string. In particular, we point out that the $E_8$ current
129: algebra of the nonsupersymmetric (left-moving) worldsheet sector is realized
130: at level two and central charge $c_L=31/2$; this is further supplemented by
131: a single real fermion $\lambda$ of $c_L=1/2$.
132:
133: Sections~4 and 5 represent the core of the paper, and are in principle
134: independent of the motivation presented in Section~2. In Section~4, we
135: specify the worldsheet theory in the NSR formulation, before and after the
136: tachyon condensate is turned on. The condensate is exponentially growing
137: along a spacetime null direction $X^+$. Conformal invariance then also
138: requires a linear dilaton along $X^-$ if we are in ten spacetime dimensions.
139: We point out that when the tachyon condensate develops, $\lambda$ transforms
140: as a candidate goldstino, suggesting a super-Higgs mechanism in worldsheet
141: supergravity.
142:
143: Section~5 presents a detailed analysis of the worldsheet super-Higgs
144: mechanism. Traditionally, worldsheet supersymmetry is fixed by working in
145: superconformal gauge, in which the worldsheet gravitino is set to zero. We
146: discuss the model briefly in superconformal gauge in Section~5.1, mainly to
147: point out that tachyon condensation leads to logarithmic CFT.
148:
149: Since the gravitino is expected to take on a more important role as a result
150: of the super-Higgs effect, in Section~5.2 and 5.3 we present a gauge choice
151: alternative to superconformal. This alternative gauge choice is inspired by
152: the ``unitary gauge'' known from the conventional Higgs mechanism in
153: Yang-Mills theories. We show in this gauge how the worldsheet gravitino
154: becomes a dynamical propagating field,
155: contributing $c_L=-11$ units of central charge. Additionally, we analyze
156: the Faddeev-Popov determinant of this gauge choice, and show that instead
157: of the conventional right-moving superghosts $\beta$, $\gamma$ of
158: superconformal gauge, we get {\it left-moving\/} superghosts $\tilde\beta$,
159: $\tilde\gamma$ of spin 1/2. In addition, we show how the proper treatment
160: of the path-integral measure in this gauge induces a shift in the linear
161: dilaton. This shift is precisely what is needed for the vanishing of the
162: central charge when the ghosts are included. Thus, this string background
163: is described in our gauge by a worldsheet conformal (but not superconformal)
164: field theory. Section~6 points out some interesting features of the
165: worldsheet theory in the late $X^+$ region, deeply in the condensed phase.
166:
167: In Appendix~A we list all of our needed worldsheet supergravity conventions.
168: Appendix~B presents a detailed evaluation of the determinants relevant for
169: the body of the paper.
170:
171: \section{Spacetime Decay to Nothing in Heterotic M-Theory}
172:
173: The anomaly cancelation mechanism that permits the existence of
174: spacetime boundaries in M-theory works locally near each boundary
175: component. The conventional realization, describing the strongly
176: coupled limit of the $E_8\times E_8$ heterotic string \cite{hw1,hw2},
177: assumes two boundary components, separated by fixed distance $R_{11}$ along
178: the eleventh dimension $y$, each breaking the same sixteen supercharges
179: and leaving the sixteen supersymmetries of the heterotic string.
180:
181: In \cite{fh}, a nonsupersymmetric variant of heterotic M-theory was
182: constructed, simply by flipping the orientation of one of the
183: boundaries. This flipped boundary breaks the complementary set of
184: sixteen supercharges, leaving no unbroken supersymmetry. The
185: motivation behind this construction was to find in M-theory a
186: natural analog of D-brane anti-D-brane systems whose study turned
187: out to be so illuminating in superstring theories. D$p$-branes
188: differ from $\bar{{\rm D}p}$-branes only in their orientation. In
189: analogy with ${\rm D}p$-$\bar{{\rm D}p}$ systems, we refer to the
190: nonsupersymmetric version of heterotic M-theory as $E_8\times\bar
191: E_8$ to reflect this similarity.%
192: %
193: \footnote{Actually, this heterotic M-theory configuration is an even closer
194: analog of a more complicated unstable string theory system: A stack of
195: D-branes together with an orientifold plane, plus anti-D-branes with an
196: anti-orientifold plane, such that each of the two collections is separately
197: neutral. These collections are only attracted to each other quantum
198: mechanically, due to the one-loop Casimir effect.}
199:
200: \subsection{The $E_8\times\bar E_8$ Heterotic M-Theory}
201:
202: This model, proposed as an M-theory analog of brane-antibrane
203: systems in \cite{fh}, exhibits two basic instabilities. First, the
204: Casimir effect produces an attractive force between the two
205: boundaries, driving the theory towards weak coupling. The strength
206: of this force per unit boundary area is given by (see \cite{fh} for
207: details):
208: %
209: \be\CF=-\frac{1}{(R_{11})^{11}}\frac{5}{2^{14}}\int_0^\infty dt\
210: t^{9/2}\theta_2(0|it),\ee
211: %
212: where $R_{11}$ is the distance between the two branes along the
213: eleventh dimension $y$.
214:
215: Secondly, as was first pointed out in \cite{fh}, at large
216: separations the theory has a nonperturbative instability. This
217: instanton is given by the Euclidean Schwarzschild solution
218: %
219: \be ds^2=\left(1-\left(\frac{4R_{11}}{\pi
220: r}\right)^8\right)dy^2+\frac{dr^2}{1-\left(\frac{4R_{11}}{\pi
221: r}\right)^8}+r^2d^2\Omega_9 \ee
222: %
223: under the ${\bf Z}_2$ orbifold action $y\rightarrow -y$. Here $r$ and the
224: coordinates in the $S_9$ are the other ten dimensions.
225: This instanton is schematically depicted in Figure~\ref{figure}(b).
226:
227: \EPSFIGURE{one.eps}{\label{figure}
228: (a) A schematic picture of the $E_8\times\bar
229: E_8$ heterotic M-theory. The two boundaries are separated by
230: distance $R_{11}$, carry opposite orientations, and support one copy
231: of $E_8$ gauge symmetry each. (b) A schematic picture of the
232: instanton responsible for the decay of spacetime to ``nothing.'' The
233: instanton is a smooth throat connecting the two boundaries. Thus,
234: the ``bubble of nothing'' is in fact a bubble of the hypothetical
235: phase on the other side of the $E_8$ boundary.}
236:
237: The probability to nucleate a single ``bubble of nothing'' of this
238: form is, per unit boundary area per unit time, of order
239: %
240: \be \exp\left(-\frac{4(2R_{11})^8}{3\pi^4G_{10}}\right), \ee
241: %
242: where $G_{10}$ is the ten-dimensional effective Newton constant. As
243: the boundaries are forced closer together by the Casimir force, the
244: instanton becomes less and less suppressed. Eventually, there should
245: be a crossover into a regime where the instability is visible in
246: perturbation theory, as a string-theory tachyon.
247:
248: \subsection{The Other Side of the $E_8$ Wall}
249:
250: The strong-coupling picture of the instanton catalyzing the decay of spacetime
251: to nothing suggests an interesting interpretation of this process. The
252: instanton has only one boundary, interpolating smoothly between the two $E_8$
253: walls. Thus, the bubble of ``nothing'' that is being nucleated
254: represents the bubble of a hypothetical phase on the other side
255: of the boundary of eleven-dimensional spacetime in heterotic M-theory.
256: In the supergravity approximation, this phase truly represents ``nothing,''
257: with no apparent spacetime interpretation. The boundary conditions at the
258: $E_8$ boundary in the supergravity approximation to heterotic M-theory are
259: reflective, and the boundary thus represents a perfect mirror.
260: However, it is possible that more refined methods, beyond supergravity, may
261: reveal a subtle world on the other side of the mirror.
262: This world could correspond to a topological phase of the theory, with very
263: few degrees of freedom (all of which are invisible in the supergravity
264: approximation).
265:
266: At first glance, it may seem that our limited understanding of M-theory would
267: restrict our ability to improve on the semiclassical picture of spacetime
268: decay at strong coupling. However, attempting to solve this problem at strong
269: coupling could be asking the wrong question, and a change of perspective
270: might be in order. Indeed, the theory itself suggests a less gloomy
271: resolution: the problem should be properly addressed at weak string coupling,
272: to which the system is driven by the attractive Casimir force. Thus, in the
273: rest of the paper, our intention is to develop worldsheet methods that
274: lead to new insight into the hypothetical phase ``behind the mirror,'' in the
275: regime of the weak string coupling.
276:
277: \subsection{Heterotic String Description at Weak Coupling}
278:
279: We conjecture that when the Casimir force has driven the $E_8$ boundaries into
280: the weak coupling regime, the perturbative string description of this system
281: is given by the little-studied tachyonic heterotic string model with
282: one copy of $E_8$ gauge symmetry \cite{klt}.%
283: %
284: \footnote{Another candidate perturbative description was suggested in
285: \cite{shanta}.}
286: %
287: The existence of a unique tachyonic $E_8$ heterotic string theory in
288: ten spacetime dimensions has always been rather puzzling. We
289: suspect that its role in describing the weakly coupled stages of the
290: spacetime decay in heterotic M-theory is the {\it raison d'\^etre\/}
291: of this previously mysterious model.
292:
293: We intend to review the structure of this nonsupersymmetric heterotic string
294: model in sufficient detail in Section~3. Anticipating its properties, we list
295: some preliminary evidence for this conjecture here:
296:
297: \begin{itemize}
298: %
299: \item The $E_8$ current algebra is realized at level two. This is
300: consistent with the anticipated Higgs mechanism $E_8\times E_8\to E_8$,
301: analogous to that observed in brane-antibrane systems where $U(N)\times U(N)$
302: is first higgsed to the diagonal $U(N)$ subgroup. (This analogy is discussed
303: in more detail in \cite{fh}.)
304: %
305: \item The nonperturbative ``decay to nothing'' instanton instability is
306: expected to become -- at weak string coupling -- a perturbative instability,
307: described by a tachyon which is a singlet under the gauge symmetry. The
308: tachyon of the $E_8$ heterotic string is just such a singlet.
309: %
310: \item The spectrum of massless fermions is nonchiral, with each
311: chirality of adjoint fermions present. This is again qualitatively the same
312: behavior as in brane-antibrane systems.
313: %
314: \item The nonsupersymmetric $E_8\times\bar E_8$ version of heterotic
315: M-theory can be constructed as a $\Z_2$ orbifold of the standard
316: supersymmetric $E_8\times E_8$ heterotic M-theory vacuum. Similarly,
317: the $E_8$ heterotic string is related to the supersymmetric
318: $E_8\times E_8$ heterotic string by a simple $\Z_2$ orbifold
319: procedure.
320: %
321: \end{itemize}
322:
323: The problem of tachyon condensation in the $E_8$ heterotic string
324: theory is interesting in its own right and can be studied
325: independently of any possible relation to instabilities in heterotic
326: M-theory. Thus, our analysis in the remainder of the paper is
327: independent of this conjectured relation to spacetime decay in
328: M-theory. As we shall see, our detailed investigation of the
329: tachyon condensation in the heterotic string at weak coupling
330: provides further corroborating evidence in support of this
331: conjecture.
332:
333: \section{The Forgotten $E_8$ Heterotic String}
334: \label{e8}
335:
336: Classical Poincar\'e symmetry in ten dimenions restricts the number of
337: consistent heterotic string theories to nine, of which six are
338: tachyonic. These tachyonic models form a natural hierarchy,
339: terminating with the $E_8$ model. We devote this section to a review
340: of some of the salient aspects of the nearly forgotten heterotic
341: $E_8$ theory. Most of these features have been known
342: for quite some time but are scattered in the literature \cite{klt,dh,forgacs,%
343: lewellen,elitzur,difrancesco}.
344:
345: \subsection{The Free Fermion Language}
346:
347: The tachyonic $E_8$ string was first discovered in the free-fermion
348: description of the nonsupersymmetric left-movers \cite{klt}. The
349: starting point of this construction is the same for all heterotic
350: models in ten dimensions (including the better-known supersymmetric
351: models): 32 real left-moving fermions $\lambda^A$, $A=1,\ldots 32$,
352: and ten right-moving superpartners $\psi_-^\mu$ of $X^\mu$,
353: described (in conformal gauge; see Appendix~A for our conventions)
354: by the free-field action
355: %
356: \be S_{\rm fermi}=\frac{i}{2\pi\alpha'}\int d^2\sigma^\pm\,
357: \left(\lambda_+^A\p_-\lambda_+^A
358: +\eta_{\mu\nu}\psi_-^\mu\p_+\psi_-^\nu \right). \ee
359: %
360: The only difference between the various models is in the assignment
361: of spin structures to various groups of fermions, and the consequent
362: GSO projection. It is convenient to label various periodicity
363: sectors by a 33-component vector
364: whose entries take values in $\Z_2=\{\pm\}$,%
365: %
366: \footnote{Here ``$+$'' and ``$-$'' correspond to the NS sector and
367: the R sector, respectively. This choice is consistent with the
368: grading on the operator product algebra of the corresponding
369: operators. Hence, the sector labeled by $+$ (or $-$) corresponds to
370: an antiperiodic (or periodic) fermion on the cylinder.}
371: %
372: \be \BU=(\underbrace{\pm,\ldots\pm}_{32}|\pm). \ee
373: %
374: The first 32 entries indicate the (anti)periodicity of the $A$-th
375: fermion $\lambda^A$, and the 33rd entry describes the
376: (anti)periodicity of the right-moving superpartners $\psi^\mu$ of
377: $X^\mu$.
378:
379: A specific model is selected by listing all the periodicities that contribute
380: to the sum over spin structures. Modular invariance requires that the
381: allowed periodicities $\BU$ are given as linear combinations of $n$
382: linearly independent basis vectors $\BU_i$
383: %
384: \be \BU=\sum_{i=1}^n\alpha_i\BU_i \ee
385: %
386: with $\Z_2$-valued coefficients $\alpha_i$. Modular invariance also
387: requires that in any given periodicity sector, the number of
388: periodic fermions is an integer multiple of eight. All six tachyonic
389: heterotic theories can be described using the following set of basis
390: vectors:
391: %
392: \bea \BU_1&=&(--------------------------------|-),\cr
393: \BU_2&=&(++++++++++++++++----------------|-),\cr
394: \BU_3&=&(++++++++--------++++++++--------|-),\cr
395: \BU_4&=&(++++----++++----++++----++++----|-),\cr
396: \BU_5&=&(++--++--++--++--++--++--++--++--|-),\cr
397: \BU_6&=&(+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-|-).\nonumber \eea
398: %
399: The theory which has only $U_1$ as a basis vector has 32 tachyons.
400: Adding $U_2$ reduces the number of tachyons to sixteen. This process can
401: be continued until the allowed periodicities are spanned by all six
402: vectors $\BU_i$; here only one bosonic tachyon is present. This
403: most intricate of the tachyonic theories is the
404: single-tachyon $E_8$ model we wish to discuss.%
405: %
406: \footnote{Adding other generators, for example $\BU_0=(-\ldots
407: -|+)$ which would relax the lock between the spin structures of the
408: left and right moving fermions, will produce other heterotic models.
409: In the case of $\BU_0$ added to any portion of the basis
410: $\BU_1,\ldots\BU_6$, the familiar $SO(16)\times SO(16)$ model is
411: produced. Adding generators which have only eight periodic fermions
412: will produce the remaining supersymmetric theories, as well as extra
413: copies of the tachyonic ones \cite{klt}.}%
414:
415: Thus, in our case there are $2^6=64$ different periodicity sectors.
416: Note that there is a perfect permutation symmetry among the
417: left-moving fermions, but this symmetry is lifted by the coupling to
418: the spin structure of the right-moving fermions $\psi^\mu$. There is
419: precisely one left-moving fermion whose spin structure is always
420: locked with the spin structure of the supersymmetric sector
421: $\psi^\mu$. Since this fermion $\lambda_+^{32}$ plays a special
422: role, we shall denote it by $\lambda_+$ and refer to it as the
423: ``lone fermion'' for brevity.
424:
425: The tachyon in this theory is a singlet, and comes from the
426: $(+\ldots+|+)$ sector in which all the fermions are Neveu-Schwarz.
427: The left-moving vacuum is excited by the lowest oscillation mode
428: $b_{-1/2}$ of the lone fermion $\lambda_+$; the right-moving vacuum is in
429: the ground state. The vertex operator for the tachyon (in picture 0) is
430: thus
431: %
432: \be
433: \CV=(F+\lambda p_\mu \psi^\mu)\exp (i p_\mu X^\mu).
434: \ee
435: %
436:
437: The spectrum also contains 248 massless vector bosons. Their spacetime
438: Yang-Mills group structure is rather obscure in the free fermion language, but
439: they do form the adjoint of $E_8$. There is one family of adjoint massless
440: fermions for each chirality; string loop effects are likely to combine these
441: into one massive field. More information about the spectrum at higher levels
442: can be extracted from the one-loop partition function calculated below, in
443: (\ref{olpf}).
444:
445: The 31 free fermions $\lambda^A$, $A=1\ldots 31$ realize a level-two
446: $E_8$ current algebra. The $E_8$ current algebra at level $k$ has
447: central charge
448: %
449: \be
450: c_{E_8,k}=\frac{k\,{\rm dim}\,E_8}{k+h}=\frac{248k}{k+30}.
451: \ee
452: %
453: Here $h=30$ is the dual Coxeter number of $E_8$. At level $k=2$,
454: this corresponds to the central charge of 31/2, which agrees with
455: the central charge of the 31 free fermions $\lambda^A$ which
456: comprise it. It is now convenient to switch to a more compact, mixed
457: representation of the left-moving sector of the worldsheet CFT.\/\
458: In this representation, the left-movers are succinctly described by
459: the lone fermion $\lambda_+(\sigma^+)$ together with the algebra of
460: $E_8$ currents $J^I(\sigma^+)$; here $I$ is the adjoint index of
461: $E_8$. The spin structure of $\lambda_+$ is locked with the spin
462: structure of the right-moving superpartners $\psi^\mu_-$ of $X^\mu$.
463:
464: At level two, the $E_8$ current algebra
465: has three integrable representations $\ket{\bf 1}$, $\ket{\bf 248}$
466: and $\ket{\bf 3875}$, where ${\bf n}$ denotes the representation
467: whose highest weight is in ${\bf n}$ of $E_8$. The conformal
468: weights of the highest weight states in $\ket{\bf 1}$, $\ket{\bf
469: 248}$ and $\ket{\bf 3875}$ are 0, 15/16 and 3/2,
470: respectively. In the spectrum of physical states, the NS and R sectors
471: $\ket{\pm}$ of $\lambda_+$ and $\psi_-^\mu$ are interwined with the
472: representations of $(E_8)_2$, which leads to the following sectors of
473: the physical Hilbert space,
474: %
475: \bea
476: \ket{+}&\otimes&\left(\ket{\bf 1}\oplus\ket{\bf 3875}\right),\nonumber\\
477: \ket{-}&\otimes&\ket{\bf 248}.\nonumber
478: \eea
479: %
480: One of the advantages of this representation is that the states
481: charged under the $E_8$ symmetry are now generated by the modes of
482: the currents $J^I$, making the $E_8$ symmetry manifest and the need
483: for its realization via 31 free fermions with a complicated GSO
484: projection obsolete.
485:
486: \subsection{The Language of Free Bosons}
487:
488: The bosonization of this model is nontrivial. In order to rewrite a
489: free fermion model in terms of bosonic fields, one typically
490: associates a bosonic field with a pair of fermions. This is,
491: however, impossible in the $E_8$ heterotic model: No two fermions
492: carry the same spin structure in all sectors, due to the intricate
493: interlacing of the spin structures reviewed above, and a twist of
494: the conventional bosonization is needed.
495:
496: As a result, the model cannot be constructed as a straight lattice
497: compactification; however, it can be constructed as a $\Z_2$
498: orbifold of one \cite{dh}. In fact, the starting point can be the
499: supersymmetric $E_8\times E_8$ heterotic string in the bosonic
500: language.
501:
502: The $E_8\times E_8$ lattice of left-moving scalars in the supersymmetric
503: $E_8\times E_8$ string has a nontrivial outer automorphism $\CI$ that simply
504: exchanges the two $E_8$ factors. One can use it to define a $\Z_2$ orbifold
505: action on the CFT via
506: %
507: \be
508: \CJ=\CI\cdot\exp\left(\pi iF_s\right),
509: \ee
510: %
511: where $F_s$ is the spacetime fermion number. Note that $\exp(\pi iF_s)$ can be
512: conveniently realized as a $2\pi$ rotation in spacetime, say in the $X^1,X^2$
513: plane. The orbifold breaks supersymmetry completely, and yields the
514: tachyonic $E_8$ heterotic string.
515:
516: This surprisingly simple orbifold relation between the
517: supersymmetric $E_8\times E_8$ theory and the tachyonic $E_8$ model
518: is possible because of some unique properties of the fusion algebra
519: and the characters of $E_8$. The fusion rules for the three integrable
520: highest-weight representations of $(E_8)_2$ are isomorphic to those of a
521: free CFT of a single real fermion
522: $\lambda_+$, with $\ket{\bf 248}$ playing the role of the spin
523: field, and $\ket{\bf 3875}$ that of the fermion. This is related to
524: the fact that the $c=1/2$ CFT of $\lambda_+$ can be represented as a
525: coset
526: %
527: \be
528: \label{ising}
529: \frac{(E_8)_1\times(E_8)_1}{(E_8)_2}.
530: \ee
531: %
532: This explains why there is such a simple relation between the supersymmetric
533: $E_8\times E_8$ heterotic string and the tachyonic $E_8$ model: They can be
534: viewed as two different ways of combining a sigle free-fermion theory
535: (\ref{ising}) with the level-two $E_8$ current algebra.
536:
537: In the bosonic form, the construction of the tachyonic $E_8$
538: heterotic string model is quite reminiscent of the CHL string
539: backgrounds \cite{chl}. In those models, a single copy of $E_8$
540: symmetry at level two is also obtained by a similar orbifold, but
541: the vacua are spacetime supersymmetric \cite{cp}. It is conceivable
542: that such supersymmetric CHL vacua in lower dimensions could
543: represent endpoints for decay of the $E_8$ model when the tachyon
544: profile is allowed an extra dependence on spatial dimensions, as in
545: \cite{hs1,hs2}.
546:
547: The one-loop partition function of the heterotic $E_8$ string theory can
548: be most conveniently calculated in lightcone gauge, by combining the bosonic
549: picture for the left-movers with the Green-Schwarz representation of the
550: right-movers. The one-loop amplitude is given by
551: %
552: \bea
553: \label{olpf}
554: {\cal Z}&=&\frac{1}{2}\int\frac{d^2\tau}{({\rm Im}\,\tau)^2}
555: \frac{1}{({\rm Im}\,\tau)^4|\eta(\tau)|^{24}}\left\{16\Theta_{E_8}(2\tau)
556: \frac{\theta_{10}^4(0,\bar\tau)}{\theta_{10}^4(0,\tau)}\right.\nonumber\\
557: & &\qquad\qquad\qquad\left.{}+\Theta_{E_8}(\tau/2)
558: \frac{\theta_{01}^4(0,\bar\tau)}{\theta_{01}^4(0,\tau)}
559: +\Theta_{E_8}\left((\tau+1)/2)\right)
560: \frac{\theta_{00}^4(0,\bar\tau)}{\theta_{00}^4(0,\tau)}\right\}.
561: \eea
562: %
563: For the remainder of the paper, we use the representation of the worldsheet
564: CFT in terms of the lone fermion $\lambda$ and the level-two $E_8$ current
565: algebra, represented by 31 free fermions.
566:
567: \section{Tachyon Condensation in the $E_8$ Heterotic String}
568:
569: \subsection{The General Philosophy}
570:
571: We wish to understand closed-string tachyon condensation as a dynamical
572: spacetime process. Hence, we are looking for a time-dependent classical
573: solution of string theory, which would describe the condensation as it
574: interpolates between the perturbatively unstable configuration at early times
575: and the endpoint of the condensation at late times. Classical solutions of
576: string theory correspond to worldsheet conformal field theories; thus, in
577: order to describe the condensation as an on-shell process, we intend to
578: maintain exact quantum conformal invariance on the worldsheet. In particular,
579: in this paper we are not interested in describing tachyon condensation in
580: terms of an abstract RG flow between two different CFTs. In addition, we limit
581: our attention to classical solutions, and leave the question of string loop
582: corrections for future work.
583:
584: \subsection{The Action}
585:
586: Before any gauge is selected, the $E_8$ heterotic string theory -- with the
587: tachyon condensate tuned to zero -- is described in the NSR formalism by
588: the covariant worldsheet action
589: %
590: \bea
591: \label{covaction}
592: S_0&=&-\frac{1}{4\pi\alpha'}\int d^2\sigma\,e\left(\vphantom{\frac{1}{2}}
593: \eta_{\mu\nu}\left(h^{mn}\p_mX^\mu\p_nX^\nu+i\psi^\mu\gamma^m\p_m\psi^\nu
594: -i\kappa\chi_m\gamma^n\gamma^m\psi^\mu\p_nX^\nu\right)\right.
595: \nonumber\\
596: &&\qquad\qquad\qquad\qquad\left.{}+i\lambda^A\gamma^m\p_m\lambda^A
597: -F^AF^A\vphantom{\frac{1}{2}}
598: \right).
599: \eea
600: %
601: where, as usual, $h_{mn}=\eta_{ab}e_m{}^ae_n{}^b$, $e=\det(e_m{}^a)$.
602: We choose not to integrate out the auxiliary fields $F^A$ from the action at
603: this stage, thus maintaining its off-shell invariance under local
604: supersymmetry, whose transformation rules on fields are given by
605: (\ref{susyright}-\ref{susygrav}). We have collected other useful formulae
606: and our choices of conventions in Appendix~\ref{appendix}.
607:
608: \subsubsection{Linear dilaton}
609:
610: In order to obtain a description of tachyon condensation in terms of
611: an exactly solvable CFT, we will consider the tachyon condensate that evolves
612: along a null direction. Thus, our tachyon condensate will depend on
613: a field, say $X^+$, which has trivial OPE with itself. In order for
614: such a condensate to maintain conformal invariance, we also need to turn
615: on a linear dilaton background,
616: %
617: \be
618: \Phi(X)=V_\mu X^\mu,
619: \ee
620: %
621: for some constant vector $V_\mu$. If we wish to maintain the
622: critical dimension equal to ten, the linear dilaton must be null,
623: $V\cdot V=0$. Hence, we can adjust our choice of spacetime
624: coordinates $X^{\pm},X^i$ such that $V$ is only nonzero in the $X^-$
625: direction. Later on, when we turn on the tachyon profile, the linear
626: dilaton will depend on the light-cone direction $X^+$
627: instead.%
628: %
629: \footnote{The need for a nonzero dilaton gradient at weak string
630: coupling is somewhat reminiscent of a similar phenomenon at strong
631: coupling: the perturbative Casimir force in the $E_8\times\bar E_8$
632: heterotic M-theory. In both cases, the ``decay-to-nothing''
633: instability plays out on top of a nontrivial spacetime dependence of
634: the string coupling. Given the absence of a reliable interpolation
635: between the weakly and strongly coupled regimes, it is not possible
636: to determine whether these two phenomena are directly related.}
637:
638: In the presence of the linear dilaton, the covariant action of the heterotic
639: model is $S=S_0+S_V$, with $S_0$ given in (\ref{covaction}) and
640: %
641: \be
642: \label{dilaction}
643: S_V=-\frac{1}{4\pi}\int d^2\sigma\,e\,V_\mu\left(\vphantom{M^M}X^\mu R(h)+
644: i\kappa\chi_{m}\gamma^n\gamma^mD_n\psi^\mu\right).
645: \ee
646: %
647: Recall that we are in Minkowski signature both on the worldsheet and in
648: spacetime; this accounts for the negative sign in front of $S_V$. In
649: the case of the null dilaton, $V\cdot V=0$, both $S_0$ and $S_V$ are
650: separately Weyl invariant; the proof of this fact for $S_V$ requires
651: the use of the equations of motion that follow from varying $X^+$ in
652: the full action. In addition, off-shell supersymmetry of $S_0+S_V$
653: also requires a modification of the supersymmetry transformation
654: rules in the supersymmetric matter sector, which now become
655: %
656: \be
657: \label{impsusy}
658: \delta X^\mu=i\epsilon\psi^\mu,\qquad\delta\psi^\mu=\gamma^m\p_m
659: X^\mu\epsilon+\alpha'V^\mu \gamma^m D_m\epsilon. \ee
660: %
661: The remaining supersymmetry transformations (\ref{susyleft}) and
662: (\ref{susygrav}) remain unmodified.
663:
664: The first term in (\ref{dilaction}) produces the standard
665: $V$-dependent term in the energy-momentum tensor, while the second
666: term yields the well-known improvement term in the supercurrent of
667: the linear dilaton theory. The second term also contributes a
668: gravitino dependent term to the energy-momentum tensor, as we will
669: show below.
670:
671: \subsubsection{Superpotential and the tachyon profile}
672:
673: At the classical level, the tachyon couples to the string as a
674: worldsheet superpotential. Classically, its coupling constant would
675: be dimensionful. Additionally, the superpotential would be neither
676: Weyl nor super-Weyl invariant: It would depend on the Liouville mode
677: $\phi$ as well as its superpartner $\chi_{-+}$.
678:
679: We are only interested in adding superpotentials that are, in
680: conformal gauge, exactly marginal deformations of the original
681: theory. The leading-order condition for marginality requires the
682: tachyon condensate $\CT(X)$ to be a dimension $(1/2,1/2)$ operator,
683: and the quantum superpotential takes the following form,
684: %
685: \be
686: \label{qsuppot}
687: S_W=-\frac{\mu}{\pi\alpha'}\int d^2\sigma\left(\vphantom{\frac{1}{2}}F\,\CT(X)
688: -i\lambda\psi^\mu\,\p_\mu\CT(X)\right);
689: \ee
690: %
691: $\mu$ is a dimensionless coupling.
692:
693: With $\CT(X)\sim\exp(k_\mu X^\mu)$ for some constant $k_\mu$, the condition
694: for $\CT(X)$ to be of dimension $(1/2,1/2)$ gives
695: %
696: \be
697: \label{tachonsh}
698: -k^2+2V\cdot k=\frac{2}{\alpha'}.
699: \ee
700: %
701: If we wish to maintain quantum conformal invariance at higher orders
702: in conformal perturbation theory in $\mu$, the profile of the
703: tachyon must be null, so that $S_W$ stays marginal. Together with
704: (\ref{tachonsh}), this leads to
705: %
706: \bea
707: \label{exptach}
708: \CT(X)&=&\exp(k_+X^+),\\
709: \label{onshelltach}
710: V_-k_+&=&-\frac{1}{2\alpha'}.
711: \eea
712: %
713: Since our $k_+$ is positive, so that the tachyon condensate grows with
714: growing $X^+$, this means that $V_-$ is negative, and the theory is weakly
715: coupled at late $X^-$.
716:
717: From now on, we will only be interested in the specific form of the
718: superpotential that follows from (\ref{exptach}) and (\ref{onshelltach}),
719: %
720: \be
721: \label{qsuppotsp}
722: S_W=-\frac{\mu}{\pi\alpha'}\int d^2\sigma\left(F-ik_+\lambda_+\psi_-^+\right)
723: \exp(k_+X^+).
724: \ee
725: %
726: Interestingly, the check of supersymmetry invariance of (\ref{qsuppotsp})
727: requires the use of (\ref{onshelltach}) together with the $V$-dependent
728: supersymmetry transformations (\ref{impsusy}).
729:
730: \subsection{The Lone Fermion as a Goldstino}
731:
732: Under supersymmetry, the lone fermion $\lambda_+$ transforms in an interesting
733: way,
734: %
735: \be
736: \delta\lambda_+=F\epsilon_+.
737: \ee
738: %
739: $F$ is an auxiliary field that can be eliminated from the theory by solving
740: its algebraic equation of motion. In the absence of the tachyon condensate,
741: $F$ is zero, leading to the standard (yet slightly imprecise) statement that
742: $\lambda_+$ is a singlet under supersymmetry. In our case, when the tachyon
743: condensate is turned on, $F$ develops a nonzero vacuum expectation value,
744: and $\lambda_+$ no longer transforms trivially under supersymmetry. In fact,
745: the nonlinear behavior of $\lambda_+$ under supersymmetry in the presence of
746: a nonzero condensate of $F$ is typical of the goldstino.
747:
748: Traditionally, the goldstino field $\eta_+$ is normalized such that
749: its leading order transformation under supersymmetry is just
750: $\delta\eta_+=\epsilon+\ldots$, where ``$\ldots$'' indicates field-dependent
751: corrections. In our case, choosing
752: %
753: \be \eta_+=\frac{\lambda_+}{F} \ee
754: %
755: gives the proper normalization for a goldstino under supersymmetry.
756: Classically, $\eta_+$ transforms as
757: %
758: \be
759: \delta\eta_+=\epsilon_+-i\eta_+(\epsilon\gamma^mD_m\eta).
760: \ee
761: %
762: This is the standard nonlinear realization of supersymmetry on the goldstino
763: in the Volkov-Akulov sense. This realization of supersymmetry has also played
764: a central role in the Berkovits-Vafa construction
765: \cite{bv,bastianelli,kunitomo,mca1,mca2}. This construction has been directly
766: linked by Hellerman and Swanson to the outcome of tachyon condensation, at
767: least in the case of Type 0 theory.
768:
769: \section{Tachyon Condensation and the Worldsheet Super-Higgs Mechanism}
770:
771: Now that we have precisely defined the worldsheet action in covariant form,
772: we will show how alternative gauge choices for worldsheet supersymmetry
773: can elucidate the dynamics of the system, and in particular, make the
774: worldsheet super Higgs mechanism manifest.
775:
776: Our alternative gauge choices will have one thing in common with
777: superconformal gauge: For fixing the bosonic part of the worldsheet gauge
778: symmetry, we always pick the conventional conformal gauge, by
779: setting (locally) $e_m{}^a=\delta_m{}^a$. This is logically consistent with
780: the fact that we turn on the tachyon condensate as an exactly marginal
781: deformation, maintaining worldsheet conformal invariance throughout.
782: In conformal gauge (and in worldsheet lightcone coordinates
783: $(\sigma^-,\sigma^+)$), the full worldsheet action becomes
784: %
785: \bea
786: \label{fulllcaction}
787: S&=&\frac{1}{\pi\alpha'}\int d^2\sigma^\pm\,\left(\p_+X^i\p_-X^i+\frac{i}{2}
788: \psi_-^i\p_+\psi_-^i-\p_+X^+\p_-X^--\frac{i}{2}\psi_-^+\p_+\psi_-^-
789: +\frac{i}{2}\lambda_+^A\p_-\lambda_+^A\right.\nonumber\\
790: &&\qquad\qquad{}-\mu^2\exp(2k_+X^+)
791: -i\kappa\chi_{++}\psi_-^i\p_-X^i+\frac{i}{2}\kappa\chi_{++}\psi_-^+\p_-X^-
792: +\frac{i}{2}\kappa\chi_{++}\psi_-^-\p_-X^+\nonumber\\
793: &&\qquad\qquad\qquad\qquad\qquad
794: \left.{}+i\kappa\alpha'V_-\chi_{++}\p_-\psi_-^-
795: +i\mu k_+\lambda_+\psi_-^+\exp(k_+X^+)\vphantom{\frac{1}{2}}\right).
796: \eea
797: %
798: In this action, we have integrated out the auxiliaries $F^A$ using their
799: algebraic equations of motion: All $F^A$s are zero with the exception of the
800: superpartner $F$ of the lone fermion, which develops a nonzero condensate:
801: %
802: \be
803: \label{condensate}
804: F=2\mu\,\exp(k_+X^+).
805: \ee
806: %
807: The supersymmetry algebra now closes only on-shell, with the use of
808: the $\lambda_+$ equation of motion. In the rest of the paper, $F$
809: always refers to the composite operator in terms of $X^+$ as given
810: by (\ref{condensate}). We will use products of powers of $F$ with
811: other fields several times below. Because the OPE of $F$ with any
812: field other than $X^-$ is trivial, these objects are quantum
813: mechanically well-defined so long as $X^-$ does not appear in them.
814:
815: We also present the energy-momentum tensor and supercurrent, again
816: in conformal gauge and worldsheet lightcone coordinates:
817: %
818: \bea
819: T_{++}&=&-\frac{1}{\alpha'}\left\{2i\kappa\chi_{++}\psi^\mu\p_+X_\mu
820: -2i\kappa\alpha'V_-\left(\frac{3}{2}\chi_{++}\p_+\psi^-_-+\frac{1}{2}(\p_+
821: \chi_{++})\psi^-_-\right)\right.\nonumber\\
822: \label{energymomentum}
823: &&\quad\left.{}+\p_+X^\mu\p_+X_\mu+\frac{i}{2}\lambda^A\p_+
824: \lambda^A-\alpha'V_-\p_+\p_+X^-\right\},\\
825: T_{--}&=&-\frac{1}{\alpha'}\left\{\p_-X^\mu\p_-X_\mu+\alpha'V_-\p_-\p_-X^-
826: +\frac{i}{2}\psi^\mu\p_-\psi_\mu\right\},\\
827: G_{--}&=&-\frac{2}{\alpha'}\left\{\psi^i_-\p_-X^i-\frac{1}{2}\psi^+_-
828: \p_-X^--\frac{1}{2}\psi^-_-\p_-X^+-\alpha'V_-\p_-\psi^-_-\right\}.
829: \label{superc}
830: \eea
831: %
832: In the classical action (\ref{fulllcaction}), the condensate
833: (\ref{condensate}) induces a bosonic potential term
834: ${}\sim\mu^2\exp(2k_+X^+)$. As shown in (\ref{energymomentum}), this
835: potential term does not contribute to the worldsheet vacuum energy,
836: since it is an operator of anomalous dimension $(1,1)$ and hence its
837: integration over the worldsheet does not require any dependence on
838: the worldsheet metric. Since this potential term does not contribute
839: to the energy-momentum tensor, it will not contribute to the BRST
840: charge either.
841:
842: On the other hand, this bosonic potential does contribute to the equation of
843: motion for $X^\pm$, which can be written locally as
844: %
845: \bea
846: \p_+\p_-X^+&=&0,\nonumber\\
847: \p_+\p_-X^-&=&2\mu^2k_+\exp(2k_+X^+)+\frac{i}{2}\kappa\p_-(\chi_{++}\psi_-^-)
848: -i\mu k_+^2\lambda_+\psi_-^+\exp(k_+X^+).
849: \eea
850: %
851: These equations imply that the generic incoming physical excitations of the
852: string are effectively shielded by the tachyon condensate from traveling too
853: deeply into the bubble of nothing. Thus, fundamental string excitations are
854: pushed away to infinity in the $X^-$ direction by the walls of the ``bubble of
855: nothing.'' A similar phenomenon has been observed numerous times in the
856: previous studies of closed-string tachyon condensation, see {\it e.g.}
857: \cite{mcgreevy,horowitz,aharony,hs1,hs2}.
858:
859: \subsection{Superconformal Gauge}
860: \label{superconformal}
861:
862: In the conventional treatment of strings with worldsheet supersymmetry,
863: superconformal gauge is virtually always selected. In this gauge, the
864: worldsheet gravitino is simply set to zero:
865: %
866: \be
867: \label{scgc}
868: \chi_{++}=0.
869: \ee
870: %
871: In our background, however, we expect the gravitino to take on a
872: prominent role as a result of the super-Higgs mechanism. For that reason,
873: we will explore alternative gauge choices, friendlier to this more
874: important role expected of the gravitino.
875:
876: Before we introduce alternative gauge choices, however, we address some
877: aspects of the theory in the conventional superconformal gauge. This exercise
878: will reveal at least one intriguing feature of the model: The emergence
879: of logarithmic CFT in the context of tachyon condensation.
880:
881: Superconformal gauge leaves residual superconformal symmetry which
882: should be realized on all fields by the action of the supercurrent
883: $G_{--}$. Consider in particular the lone fermion $\lambda_+$. Before the
884: tachyon condensate is turned on, the operator product of $G_{--}$ with
885: $\lambda_+$ is nonsingular, in accord with the fact that $\lambda_+$
886: transforms trivially under on-shell supersymmetry. As we have seen in
887: Section~4, when the auxiliary field $F$ develops a nonzero vacuum
888: expectation value in the process of tachyon condensation,
889: $\lambda_+$ transforms under supersymmetry nontrivially, as a
890: candidate goldstino. This raises an interesting question: How can
891: this nontrivial transformation be reproduced by the action of
892: $G_{--}$ on $\lambda_+$, if, as we have seen in (\ref{superc}), the
893: supercurrent $G_{--}$ is unmodified by $\mu$?
894:
895: The resolution of this puzzle must come from nontrivial OPEs that develop
896: at finite $\mu$ between the originally leftmoving field $\lambda_+$ and the
897: originally rightmoving fields $\psi_-^\pm$. Here and in the following, it
898: will be useful to introduce a rescaled version of the fields $\psi_-^\pm$,
899: %
900: \be
901: \label{firstresc}
902: \tilde\psi_-^-=\psi_-^-/F,\qquad\tilde\psi_-^+=F\psi_-^+.
903: \ee
904: %
905: We will encounter this particular rescaled version of $\psi_-^-$
906: again below, in another gauge. In terms of $\tilde\psi_-^-$, the
907: supercurrent (\ref{superc}) simplifies to
908: %
909: \be
910: \label{supcre}
911: G_{--}=-\frac{2}{\alpha'}\left\{\psi^i_-\p_-X^i-\frac{1}{2}\psi^+_-\p_-X^-
912: -\alpha'V_-F\p_-\tilde\psi^-_-\right\}.
913: \ee
914: The supersymmetry variations of fields are reproduced as follows. Consider
915: for example the supermultiplet $\psi_-^i$ and $X^i$. In superconformal
916: gauge, these are free fields, satisfying standard OPEs such as
917: %
918: \be
919: \psi_-^i(\sigma^\pm)\psi^j(\tau^\pm)
920: \sim\frac{\alpha'\delta^{ij}}{\sigma^--\tau^-},
921: \ee
922: %
923: which imply
924: %
925: \be
926: G_{--}(\sigma^\pm)\psi_-^i(\tau^\pm)\sim\frac{-2\p_-X^i(\tau^\pm)}{
927: \sigma^--\tau^-},
928: \ee
929: %
930: correctly reproducing the supersymmetry variation
931: $\delta\psi_-^i=-2\p_-X^i\epsilon$.
932:
933: Similarly, the last term in (\ref{supcre}) will reproduce the supersymmetry
934: transformation $\delta\lambda_+=F\epsilon$ if the following OPE holds,
935: %
936: \be
937: \label{expectope}
938: 2\alpha'V_-\p_-\psi^-_-(\sigma^\pm)\,\lambda_+(\tau^\pm)\sim
939: \frac{F(\tau^\pm)/F(\sigma^\pm)}{\sigma^--\tau^-}.
940: \ee
941: %
942: This required OPE can be checked by an explicit calculation: Starting with
943: the free-field OPEs
944: %
945: \be
946: \psi_-^+(\sigma^\pm)\,\psi_-^-(\tau^\pm)\sim
947: \frac{-2\alpha'}{\sigma^--\tau^-},
948: \ee
949: %
950: we get for the rescaled fields (\ref{firstresc})
951: %
952: \bea
953: \tilde\psi_-^+(\sigma^\pm)\,\tilde\psi_-^-(\tau^\pm)&\sim&
954: \frac{-2\alpha'F(\sigma^\pm)/F(\tau^\pm)}{\sigma^--\tau^-}\nonumber\\
955: &&\qquad{}\sim
956: \frac{-2\alpha'}{\sigma^--\tau^-}
957: \sum_{n=0}^\infty\frac{1}{n!}(\sigma^+-\tau^+)^n\left(\p_+^n
958: F(\tau^\pm)\right)/F(\tau^\pm).
959: \eea
960: %
961: Note that since $X^+$ satisfies locally the free equation of motion
962: $\p_+\p_- X^+=0$, the coefficient of the $1/(\sigma^--\tau^-)$ term in this
963: OPE is only a function of $\sigma^+$ and $\tau^+$.
964: The OPE between $\p_-\tilde\psi_-^-$ and $\lambda_+$ can then be determined
965: from the the equation of motion for $\lambda_+$:
966: %
967: \be
968: \p_-\lambda_+(\sigma)=-\mu k_+\psi_-^+\exp(k_+X^+)=-\frac{k_+}{2}
969: \tilde\psi_-^+.
970: \ee
971: %
972: Combining these last two equations, we get
973: %
974: \be
975: \p_-\lambda_+(\sigma^\pm)\tilde\psi_-^-(\tau^\pm)=-\frac{k_+}{2}
976: \tilde\psi_-^+(\sigma^\pm)\tilde\psi_-^-(\tau^\pm)=\alpha'k_+
977: \frac{F(\sigma^\pm)/F(\tau^\pm)}{\sigma^--\tau^-}.
978: \ee
979: %
980: Integrating the result with respect to $\sigma^-$, we finally obtain
981: %
982: \be \label{logope}\lambda_+(\sigma^\pm)\tilde\psi_-^-(\tau^\pm)=
983: \alpha'k_+\left\{\sum_{n=0}^\infty\frac{1}{n!}(\sigma^+-\tau^+)^n\left(\p_+^n
984: F(\tau^\pm)\right)/F(\tau^\pm)\right\} \log(\sigma^--\tau^-). \ee
985: %
986: One can then easily check that this OPE implies (\ref{expectope}) when
987: (\ref{onshelltach}) is invoked. This in turn leads to
988: %
989: \be
990: G_{--}(\sigma^\pm)\lambda_+(\tau^\pm)\sim\frac{F(\tau^\pm)}{
991: \sigma^--\tau^-},
992: \ee
993: %
994: and the supersymmetry transformation of $\lambda_+$ is correctly reproduced
995: quantum mechanically, even in the presence of the tachyon condensate.
996:
997: As is apparent from the form of (\ref{logope}), our theory exhibits --
998: in superconformal gauge -- OPEs with a logarithmic behavior. This
999: establishes an unexpected connection between models of tachyon
1000: condensation and the branch of 2D CFT known as ``logarithmic CFT'' (or LCFT;
1001: see, {\it e.g.}, \cite{flohr,gaberdiel} for reviews).
1002: The subject of LCFT has been vigorously studied in recent years,
1003: with applications to a wide range of physical problems, in particular to
1004: systems with disorder. The logarithmic behavior of OPEs is
1005: compatible with conformal symmetry, but not with unitarity. Hence, it
1006: can only emerge in string backgrounds in Minkowski spacetime signature, in
1007: which the time dependence plays an important role (such as our problem of
1008: tachyon condensation). We expect that the concepts and techniques
1009: developed in LCFTs could be fruitful for understanding time-dependent
1010: backgrounds in string theory. In the present work, we will not explore this
1011: connection further.
1012:
1013: \subsection{Alternatives to Superconformal Gauge}
1014: \label{nunit}
1015:
1016: In the presence of a super-Higgs mechanism, another natural choice of
1017: gauge suggests itself. In this gauge, one anticipates the assimilation of the
1018: Goldstone mode by the gauge field, by simply gauging away the Goldstone mode
1019: altogether. In the case of the bosonic Higgs mechanism, this gauge is often
1020: referred to as ``unitary gauge'' as it makes unitarity of the theory
1021: manifest.
1022:
1023: Following this strategy, we will first try eliminating the goldstino
1024: as a dynamical field, for example by simply choosing
1025: %
1026: \be
1027: \lambda_+=0
1028: \ee
1029: %
1030: as our gauge fixing condition for local supersymmetry. Of course,
1031: this supplements the conformal gauge choice that we have made for
1032: worldsheet diffeomorphisms. This gauge choice gives rise to
1033: non-propagating bosonic superghosts, together with the usual
1034: propagating fermionic $b,c$ system of central charge $c=-26$. We
1035: refer to this gauge as ``unitary gauge'' only due to its ancestry in the
1036: similar gauges that proved useful in the Higgs mechanism of Yang-Mills
1037: gauge theories. In fact, as we shall see, the proper implementation of this
1038: ``unitary'' gauge will lead to propagating superghosts, and therefore no
1039: manifest unitarity of the theory.
1040:
1041: In addition to the conventional fermionic $b$, $c$ ghosts from
1042: conformal gauge, this gauge choice would lead to non-propagating bosonic
1043: ghosts, which might be integrated out algebraically. More importantly,
1044: this gauge choice still leaves complicated interaction terms, such as
1045: %
1046: \be \label{intterm}
1047: \chi_{++}\psi^+_-\p_-X^-,\ee
1048: %
1049: in the gauge-fixed version of (\ref{fulllcaction}). Moreover, if the
1050: algebraic superghosts are integrated out, the equation of motion arising
1051: from variation of the action under $\lambda_+$,
1052: %
1053: \be \mu k_+\psi_-^+\exp(k_+X^+)=0, \ee
1054: %
1055: needs to be imposed on physical states by hand. This could be accomplished
1056: by simply setting $\psi_-^+=0$. This leads to another constraint, imposing
1057: the equation of motion obtained from the variation of $\psi_-^+$ in the
1058: original action as a constraint on physical states.
1059:
1060: Instead of resolving such difficulties, we will restart our analysis
1061: with $\psi^+_-=0$ as the gauge condition. As we will see below,
1062: this condition makes the gauge-fixing procedure transparent.
1063:
1064: \subsection{Liberating the Worldsheet Gravitino in an Alternative Gauge}
1065: \label{liber}
1066:
1067: We will now explicitly consider the gauge that begins simply by
1068: setting
1069: %
1070: \be
1071: \label{pixiegauge}
1072: \psi_-^+=0.
1073: \ee
1074: %
1075:
1076: If $\psi_-^+$ is eliminated from the action, the remaining $\chi_{++}$,
1077: $\psi_-^-$ system can be rewritten as a
1078: purely left-moving first-order system of conformal weights $(3/2,0)$ and
1079: $(-1/2,0)$.%
1080: %
1081: \footnote{This system is still coupled to the transverse fields $\psi^i$,
1082: $X^i$ but the strength of the interaction goes to zero with growing $F$.}
1083: %
1084: This can be seen as follows. Consider first the terms in the
1085: action that are bilinear in these two fields,
1086: %
1087: \be
1088: \label{bilin}
1089: \frac{1}{2}\chi_{++}\psi_-^-\p_-X^++\alpha'V_-\chi_{++}\p_-\psi_-^-.
1090: \ee
1091: %
1092: The presence of $\p_-$ here suggests $\chi_{++}$ and $\psi_-^-$
1093: ought to become purely left-moving. Additionally, these fields show
1094: up only in the energy momentum tensor $T_{++}$ in this gauge. However,
1095: their conformal weights do not reflect their left-moving nature. In order
1096: to obtain fields whose conformal weights are nonzero only in the left-moving
1097: sector, we can rescale
1098: %
1099: \be
1100: \label{rescale}
1101: \tilde\chi_{++}=F\chi_{++},\qquad\tilde\psi_-^-=\frac{\psi_-^-}{F}.
1102: \ee
1103: %
1104: This rescaling leads to an additional benefit: The bilinear terms
1105: (\ref{bilin}) in the classical action now assemble into the canonical kinetic
1106: term of a first-order system of spin 3/2 and $-1/2$,
1107: %
1108: \be
1109: \label{agra}
1110: \frac{i\kappa V_-}{\pi}\int d^2\sigma^\pm\,\tilde\chi_{++}\p_-
1111: \tilde\psi_-^-,
1112: \ee
1113: %
1114: with central charge $c_L=-11$.
1115:
1116: The first-order system (\ref{agra}) describes the worldsheet gravitino sector
1117: of the theory. In superconformal gauge in the absence of the tachyon
1118: condensate, the gravitino was non-dynamical and led to a constraint. Here,
1119: instead, the gravitino has been liberated as a result of the super-Higgs
1120: mechanism: together with its conjugate $\tilde\psi_-^-$, at late
1121: times it appears to have formed a left-moving free-field massless
1122: system, of central charge $c=-11$. In this modified unitary gauge,
1123: the gravitino has been literally set free: It has become a free
1124: propagating massless field!
1125:
1126: Our gauge choice (\ref{pixiegauge}) reduces the classical action
1127: significantly, to
1128: %
1129: \bea
1130: \label{pixieaction}
1131: S&=&\frac{1}{\pi\alpha'}\int
1132: d^2\sigma^\pm\,\left(\p_+X^i\p_-X^i+\frac{i}{2}
1133: \psi_-^i\p_+\psi_-^i-\p_+X^+\p_-X^-
1134: +\frac{i}{2}\lambda_+^A\p_-\lambda_+^A-\mu^2\exp(2k_+X^+)\right.\nonumber\\
1135: &&\qquad\qquad{} \left.{} -i\kappa\chi_{++}\psi_-^i\p_-X^i
1136: +\frac{i}{2}\kappa\chi_{++}\psi_-^-\p_-X^+
1137: +i\kappa\alpha'V_-\chi_{++}\p_-\psi_-^-\vphantom{\frac{1}{2}}
1138: \right). \eea
1139: %
1140: Note that unlike in the case of superconformal gauge, the gauge fixing
1141: condition (\ref{pixiegauge}) leaves no residual unfixed supersymmetries,
1142: at least at finite $\mu$. Thus, in this gauge, the theory will be conformal,
1143: but not superconformal. In Section~\ref{latetimes}, we shall return to the
1144: issue of residual supersymmetry in this gauge, in the late-time limit of
1145: $\mu\to\infty$.
1146:
1147: The action (\ref{pixieaction}) will be corrected by one-loop effects.
1148: The first such correction is due to the Faddeev-Popov (super)determinant
1149: $\Delta_{\rm FP}$.
1150: As we will now see, the inherent $X^+$ dependence in our gauge fixing
1151: condition renders $\Delta_{\rm FP}$ dependent on $X^+$ as well.
1152:
1153: Our full gauge fixing condition consists of the bosonic conformal
1154: gauge, $e_m{}^a=\delta_m{}^a$, as well as the fermionic condition
1155: (\ref{pixiegauge}). Note first that the corresponding Faddeev-Popov
1156: superdeterminant factorizes into the ratio of two bosonic determinants,%
1157: %
1158: \footnote{This is true because the operator whose superdeterminant
1159: is being calculated has a block-triangular form, as the action of
1160: diffeomorphisms on the fermionic gauge-fixing condition $\psi_-^+$
1161: vanishes when (\ref{pixiegauge}) is satisfied.}
1162: %
1163: \be \label{FadeevPopov}
1164: \Delta_{\rm FP}=J_{bc}/J_{\psi_-^+\epsilon}.
1165: \ee
1166: %
1167: Here $J_{bc}$ arises from the conformal gauge condition, and
1168: produces the standard set of fermionic $b,c$ ghosts with central
1169: charge $c=-26$. On the other hand, $J_{\psi_-^+\epsilon}$, which
1170: comes from the change of variables between the gauge-fixing
1171: condition $\psi_-^+$ and the infinitesimal supersymmetry parameter
1172: $\epsilon$, turns out to be more complicated. It will be useful to
1173: rewrite the variation of $\psi^+_-$ as
1174: %
1175: \bea
1176: \label{op}
1177: \delta\psi^+_-&=& -2\p_-X^+\epsilon+4\alpha'V_-D_-\epsilon\nonumber\\
1178: &=& \frac{4\alpha'V_-}{F}D_-(F\epsilon).
1179: \eea
1180: %
1181: Here, we are only fixing worldsheet diffeomorphisms and not Weyl
1182: transformations, in order to allow a systematic check of the vanishing of
1183: the Weyl anomaly. We are thus in a gauge where
1184: $h_{mn}=e^{2\phi}\hat{h}_{mn}$. The derivative $D_-$ preserves
1185: information about the spin of the field it acts on. We can now see
1186: the denominator of (\ref{FadeevPopov}) should be defined as
1187: %
1188: \be J_{\psi_-^+\epsilon}=\det\left(\frac{1}{F}D_-F\right). \ee
1189: %
1190: As in the case of $J_{bc}$, we wish to represent this determinant as
1191: a path integral over a conjugate pair of bosonic fields $\beta$,
1192: $\gamma$: the bosonic ghosts associated with the fixing of local
1193: supersymmetry. Note, however, that the variation of our fixing
1194: condition is dependent on $X^+$. This fact leads to interesting
1195: additional subtleties, which we discuss in detail in
1196: Appendix~\ref{appendixb}.
1197:
1198: As a result, the Jacobian turns out to be given by a set of ghosts
1199: $\tilde\beta$, $\tilde\gamma$ of central charge $c=-1$ whose path
1200: integral measure is independent of $X^+$, plus an extra term that depends
1201: explicitly on $X^+$. As shown in Eqn.~(\ref{Sgauge}) of
1202: Appendix~\ref{appendixb}, we find
1203: %
1204: \be \label{oneoverJpsi}\frac{1}{J_{\psi_-^+\epsilon}}=
1205: \exp\left\{-\frac{i}{\pi\alpha'}\int
1206: d^2\sigma^\pm\,\alpha'k_+^2\p_+X^+\p_-X^+\right\}
1207: \int\CD\tilde\beta\CD\tilde\gamma\exp\left(iS_{\tilde\beta\tilde
1208: \gamma}\right), \ee
1209: %
1210: where $S_{\tilde\beta\tilde\gamma}$ is the ghost action for a free
1211: left-moving bosonic ghost system of spin 1/2.
1212:
1213: The Faddeev-Popov determinant thus renormalizes the spacetime metric. If this
1214: were the whole story, we could re-diagonalize the spacetime metric to the
1215: canonical light-cone form, by redefining the spacetime light-cone coordinates
1216: %
1217: \bea
1218: Y^+ &=& X^+,\nonumber\\
1219: Y^- &=& X^- +\alpha' k_+^2X^+.
1220: \eea
1221: %
1222: In these new coordinates, the linear dilaton would acquire a shift:
1223: %
1224: \bea
1225: \Phi&=&V_-X^-,\nonumber\\
1226: &=& V_-Y^--V_-\alpha'k_+^2X^+,\nonumber\\
1227: &=& V_-Y^- +\frac{k_+}{2}Y^+.
1228: \eea
1229: %
1230: The effective change in the central charge from this shift in the linear
1231: dilaton would be
1232: %
1233: \be
1234: c_{dil}=6\alpha'V^2=-24\alpha'V_-V_+=6,
1235: \ee
1236: %
1237: where we have again used (\ref{tachonsh}).
1238:
1239: However, understanding the Faddeev-Popov determinant is not the
1240: whole story. In addition, the $X^+$-dependent rescaling of
1241: $\chi_{++}$ and $\psi_-^-$ as in (\ref{rescale}) also produces a
1242: subtle Jacobian $\tilde J$. As shown in Appendix~\ref{appendixb},
1243: this Jacobian can be expressed in Minkowski signature as
1244: %
1245: \be \label{minkjpsi}\tilde J=\exp\left\{-\frac{i}{4\pi\alpha'}\int
1246: d^2\sigma \hat e\left(\alpha'k_+^2\hat
1247: h^{mn}\p_mX^+\p_nX^++\alpha'k_+X^+R\right)\right\}. \ee
1248: %
1249: In this sense, the original $\chi_{++}$, $\psi_-^-$ system is
1250: equivalent to the canonical $\tilde\chi_{++}$, $\tilde\psi_-^-$
1251: system when this additional renormalization of the linear dilaton
1252: term and the the spacetime metric are taken into account.
1253: In lightcone coordinates, the first factor in (\ref{minkjpsi})
1254: becomes
1255: %
1256: \be
1257: \exp\left\{\frac{i}{\pi\alpha'}\int d^2\sigma^\pm\,\alpha'k_+^2\p_+X^+
1258: \p_-X^+\right\}.
1259: \ee
1260: %
1261: This contribution to the renormalization of the metric precisely cancels the
1262: contribution obtained from the Faddeev-Popov determinant (\ref{oneoverJpsi}).
1263: Thus, the combined effect of $\tilde J$ and $J_{\psi_-^+\epsilon}$ on the
1264: $X^\pm$ fields is just a simple shift of the linear dilaton in the original
1265: $X^\pm$ variables, as implied by the second term in (\ref{minkjpsi}).
1266: The contribution to the central charge due to this dilaton shift is $c=12$, as
1267: $c_{dil}=6\alpha'V^2=12$.
1268:
1269: The full action, still in conformal gauge, is thus given by:
1270: %
1271: \bea
1272: \label{cactionn}
1273: S&=&\frac{1}{\pi\alpha'}\int
1274: d^2\sigma^\pm\,\left(\p_+X^i\p_-X^i+\frac{i}{2}
1275: \psi_-^i\p_+\psi_-^i-\p_+X^+\p_-X^-
1276: +\frac{i}{2}\lambda_+^A\p_-\lambda_+^A-\mu^2\exp(2k_+X^+)\right.\nonumber\\
1277: &&\qquad\qquad{} \left.{}
1278: -i\kappa\frac{\tilde\chi_{++}}{F}\psi_-^i\p_-X^i
1279: +i\kappa\alpha'V_-\tilde\chi_{++}\p_-\tilde\psi_-^-\vphantom{\frac{1}{2}}
1280: \right)+S_{bc}+S_{\tilde\beta\tilde\gamma}.
1281: \eea
1282: %
1283: The information about the linear dilaton profile is absent in this action, but
1284: the energy-momentum tensor of the theory reveals that the dilaton is given by
1285: %
1286: \be
1287: V_\mu=(k_+,V_-,\vec{0}).
1288: \ee
1289: %
1290: In this form, the action has only canonical kinetic terms, plus the
1291: potential term and an interaction term which will vanish at late
1292: times (since it goes as $1/F$). Specifically, we note that the fields
1293: $\tilde\chi_{++}$, $\tilde\psi_-^-$ have now become purely left-moving.
1294: This matches their conformal weights, which are $(3/2,0)$ and $(1/2,0)$
1295: respectively. In addition, they contribute
1296: only to the left moving energy momentum tensor; their contribution
1297: results in $-11$ units of central charge for $c_L$, in the late $X^+$ limit
1298: when they decouple from the transverse degrees of freedom $X^i,\psi^i$.
1299:
1300: Now, let us summarize the central charge contributions of each field
1301: present in this gauge, at late times. For comparison, we also
1302: present the central charge breakdown for the free theory in
1303: superconformal gauge.
1304: \smallskip
1305: %
1306: \be
1307: \begin{tabular}{cc}
1308: Superconformal gauge & The alternative gauge\\
1309: {
1310: \begin{tabular}{|l|c|c|}
1311: \hline
1312: Field & $c_L$ & $c_R$ \\
1313: \hline
1314: $X^+$, $X^-$, $X^i$ & $10$ & $10$\\
1315: linear dilaton & $0$ & $0$\\
1316: $bc$ ghosts &$-26$ & $-26$\\
1317: $\psi^i$ & $0$ & $4$\\
1318: $\psi^+$, $\psi^-$ & $0$ & $1$\\
1319: $\beta$, $\gamma$ ghosts &$0$ & $11$ \\
1320: $\lambda_+$ & $1/2$ & $0$\\
1321: $(E_8)_2$ & $31/2$ & $0$ \\
1322: \hline
1323: \end{tabular}
1324: } & {
1325: \begin{tabular}{|l|c|c|}
1326: \hline
1327: Field & $c_L$ & $c_R$ \\
1328: \hline
1329: $X^+$, $X^-$, $X^i$ & $10$ & $10$\\
1330: linear dilaton & $12$ & $12$\\
1331: $bc$ ghosts &$-26$ & $-26$\\
1332: $\psi^i$ & $0$ & $4$\\
1333: $\tilde\chi$, $\tilde \psi^-$ & $-11$ & $0$ \\
1334: $\tilde\beta$, $\tilde\gamma$ ghosts &$-1$ & $0$ \\
1335: $\lambda_+$ & $1/2$ & $0$\\
1336: $(E_8)_2$ & $31/2$ & $0$ \\
1337: \hline
1338: \end{tabular}
1339: } \end{tabular}
1340: \ee
1341: %
1342: Our new gauge choice has resulted in $12$ units of central charge in
1343: the right-moving sector, from $\psi^\pm$ and the $\beta\gamma$
1344: system, effectively moving to become $-12$ units of central charge
1345: on the left. These left-moving central charge units come from
1346: $\tilde\chi$, $\tilde\psi^-$, and the new $\tilde\beta\tilde\gamma$
1347: ghosts. We also see that the shifted linear dilaton precisely
1348: compensates for this relocation of central charge, resulting in the
1349: theory in the alternative gauge still being exactly conformal
1350: at the quantum level.
1351:
1352: Interestingly, the equation of motion that follows from varying
1353: $\psi_-^+$ in the original action allows us to make contact with the
1354: original unitary gauge. Classically, this equation of motion is
1355: %
1356: \be
1357: \label{tildesc}
1358: \p_+\psi_-^-+\kappa\chi_{++}\p_-X^-+2\mu k_+\lambda_+\exp(k_+X^+)=0.
1359: \ee
1360: %
1361: This constraint can be interpreted and solved in a particularly
1362: natural way: Imagine solving the $X^\pm$ and $\chi_{++}, \psi_-^-$
1363: sectors first. Then one can simply use the constraint to express
1364: $\lambda_+$ in terms of those other fields. Thus, the alternative
1365: gauge still allows the gravitino to assimilate the goldstino
1366: in the process of becoming a propagating field. This is how the
1367: worldsheet super-Higgs mechanism is implemented, in a way compatible with
1368: conformal invariance.
1369:
1370: We should note that this classical constraint (\ref{tildesc}) could
1371: undergo a one-loop correction analogous to the one-loop shift in the
1372: dilaton. We might expect a term $\sim\p_-\chi_{++}$, from varying a
1373: one-loop supercurrent term $\sim\chi_{++}\p_-\psi_-^+$ in the full
1374: quantum action. Such a correction would not change the fact that
1375: $\lambda_+$ is determined in terms of the oscillators of other fields,
1376: it would simply change the precise details of such a rewriting.
1377:
1378: \subsection{$R_\xi$ Gauges}
1379:
1380: The history of understanding the Higgs mechanism in Yang-Mills theories
1381: was closely linked with the existence of a very useful family of gauge
1382: choices, known as $R_\xi$ gauges. $R_\xi$ gauges interpolate -- as one varies
1383: a control parameter $\xi$ -- between unitary gauge and one of the more
1384: traditional gauges (such as Lorentz or Coulomb gauge).
1385:
1386: In string theory, one could similarly consider families of gauge fixing
1387: conditions for worldsheet supersymmetry which interpolate between
1388: the traditional superconformal gauge and our alternative gauge.
1389: We wish to maintain conformal invariance of the theory in the new gauge, and
1390: will use conformal gauge to fix the bosonic part of the worldsheet gauge
1391: symmetries. Because they carry disparate conformal weights $(1,-1/2)$ and
1392: $(0,1/2)$ respectively, we cannot simply add the two gauge fixing conditions
1393: $\chi_{++}$ and $\psi_-^+$ with just a relative constant.
1394: In order to find a mixed gauge-fixing condition compatible with conformal
1395: invariance, we need a conversion factor that makes up for this difference in
1396: conformal weights. One could for example set
1397: %
1398: \be
1399: \label{mixed}
1400: (\p_-X^+)^2\chi_{++}+\xi F^2\psi_-^+=0,
1401: \ee
1402: %
1403: with $\xi$ a real constant (of conformal dimension 0). The added
1404: advantage of such a mixed gauge is that it interpolates between
1405: superconformal and alternative gauge not only as one changes $\xi$,
1406: but also at any fixed $\xi$ as $X^+$ changes: At early lightcone
1407: time $X^+$, the superconformal gauge fixing condition dominates,
1408: while at late $X^+$, the alternative gauge takes over, due to the
1409: relative factors of $F$ between the two terms in (\ref{mixed}).
1410:
1411: Even though (\ref{mixed}) is compatible with conformal invariance, it is
1412: highly nonlinear, and therefore impractical as a useful gauge fixing
1413: condition. Another, perhaps more practical, condition that incorporates
1414: both $\chi_{++}$ and $\psi_-^+$ is
1415: %
1416: \be
1417: \label{interesting}
1418: \p_-\chi_{++}+\xi\p_+\psi_-^+=0.
1419: \ee
1420: %
1421: Here the mismatch in conformal weights has been made up by inserting
1422: worldsheet derivatives, rather than other composite operators. This
1423: condition can be written in the following covariant form,
1424: %
1425: \be
1426: \gamma^m\gamma^nD_n\chi_m-2\xi\gamma^mD_m\psi^+=0,
1427: \ee
1428: %
1429: demonstrating its compatibility with diffeomorphism invariance.
1430:
1431: Note that
1432: (\ref{interesting}) correctly anticipates the dynamics of the gravitino:
1433: If we solve for $\p_+\psi_-^+$ using (\ref{interesting}) in the kinetic term
1434: $\psi_-^-\p_+\psi_-^+$, we get $\psi_-^-\p_-\chi_{++}$. Thus, $\chi_{++}$
1435: replaces $\psi_-^+$ as the conjugate partner of $\psi_-^-$, and turns it from
1436: a right-moving field into a left-moving one. Note also that both terms in
1437: (\ref{interesting}) transform under supersymmetry as
1438: $\sim\p_+\p_-\epsilon+\ldots $, which implies that the bosonic superghosts
1439: $\hat\beta$, $\hat\gamma$ associated with this gauge fixing will have a
1440: second-order kinetic term,
1441: %
1442: \be
1443: \sim\int d^2\sigma^\pm\,\left(\hat\beta\p_+\p_-\hat\gamma+\ldots\right).
1444: \ee
1445: %
1446: In the $\mu\to\infty$ limit, this gauge can be expected to leave a residual
1447: fermionic gauge symmetry.
1448:
1449: Various other classes of $R_\xi$-type gauges can be considered.
1450: For example, one could study combinations of superconformal gauge with
1451: the naive unitary gauge in which $\lambda_+$ is set to zero, leading to
1452: gauge fixing conditions such as
1453: %
1454: \be
1455: F\chi_{++}+\xi k_+\lambda_+\p_+X^+=0,
1456: \ee
1457: %
1458: or
1459: \be
1460: F\lambda_++\xi\chi_{++}k_+\p_-X^+=0.
1461: \ee
1462: %
1463: Another interesting possibility is
1464: %
1465: \be
1466: \chi_{++}+\xi\,\p_+\!\left(\frac{\lambda_+}{F}\right)=0,
1467: \ee
1468: %
1469: which would interpolate between superconformal gauge and a weaker form of
1470: unitary gauge, in which the left-moving part of the goldstino is set to zero.
1471:
1472: We have not investigated the worldsheet theory in these mixed gauges, but it
1473: would be interesting to see if any of them shed some new light on the
1474: dynamics of tachyon condensation.
1475:
1476: \section{The Condensed Phase: Exploring the Worldsheet Theory at $\mu=\infty$}
1477: \label{latetimes}
1478:
1479: The focus of the present paper has been on developing worldsheet
1480: techniques that can elucidate the super-Higgs mechanism and the
1481: dynamics of the worldsheet gravitino in the process of spacetime
1482: tachyon condensation. Here we comment briefly on the structure of
1483: the worldsheet theory in the regime where the tachyon has already
1484: condensed.
1485:
1486: This condensed phase corresponds to the system at late $X^+$. In the
1487: worldsheet theory, a constant translation of $X^+$ rescales the value of
1488: the superpotential coupling $\mu$, with the late $X^+$ limit mapping to
1489: $\mu\to\infty$. In that limit, the worldsheet theory simplifies in an
1490: interesting way. First, we rescale the parameter of local supersymmetry
1491: transformation,
1492: %
1493: \be
1494: \tilde\epsilon_+=F\epsilon_+.
1495: \ee
1496: %
1497: The supersymmetry variations then reduce in the $\mu=\infty$ limit and
1498: in conformal gauge to
1499: %
1500: \bea
1501: \label{latesusy}
1502: \delta X^+&=&0,\qquad\qquad\qquad\delta\tilde\psi_-^+=4\alpha'V_-\p_-
1503: \tilde\epsilon_+,\nonumber\\
1504: \delta X^-&=&-i\tilde\psi_-^-\tilde\epsilon_+,\quad\qquad\delta
1505: \tilde\psi_-^-=0,\\
1506: \delta X^i&=&0,\qquad\qquad\qquad\delta\tilde\psi_-^i=0,\nonumber\\
1507: \delta\lambda_+&=&\tilde\epsilon_+,\qquad\qquad\quad
1508: \delta\tilde\chi_{++}=\frac{2}{\kappa}\left(\p_+\tilde\epsilon_+-k_+\p_+X^+
1509: \tilde\epsilon_+\right).\nonumber
1510: \eea
1511: %
1512: Note that $X^+$ is now invariant under supersymmetry. Consequently, the
1513: terms in the action that originate from the superpotential,
1514: %
1515: \be
1516: -2\mu^2\exp(2k_+X^+)-ik_+\lambda_+\tilde\psi_-^+,
1517: \ee
1518: %
1519: are now separately invariant under (\ref{latesusy}), in the strict
1520: $\mu=\infty$ limit. This in turn implies that we are free to drop the
1521: potential term $\sim\mu^2\exp(2k_+X^+)$ without violating supersymmetry.
1522: The resulting model is then described, in the alternative gauge of
1523: Section~\ref{liber}, by a free field action:
1524: %
1525: \bea
1526: S_{\mu=\infty}&=&\frac{1}{\pi\alpha'}\int
1527: d^2\sigma^\pm\,\left(\p_+X^i\p_-X^i+\frac{i}{2}
1528: \psi_-^i\p_+\psi_-^i-\p_+X^+\p_-X^-
1529: +\frac{i}{2}\lambda_+^A\p_-\lambda_+^A\right.\nonumber\\
1530: &&\qquad\qquad\qquad{} \left.{}
1531: +i\kappa\alpha'V_-\tilde\chi_{++}\p_-\tilde\psi_-^-\vphantom{\frac{1}{2}}
1532: \right)+S_{bc}+S_{\tilde\beta\tilde\gamma}.
1533: \eea
1534: %
1535:
1536: We argued in Section~\ref{liber} that at finite $\mu$, our
1537: alternative gauge (\ref{pixiegauge}) does not leave any residual
1538: unfixed supersymmetry, making the theory conformal but not
1539: superconformal. This is to be contrasted with superconformal gauge,
1540: which leaves residual right-moving superconformal symmetry. At
1541: $\mu=\infty$, however, it turns out that the alternative gauge
1542: (\ref{pixiegauge}) does leave an exotic form of residual supersymmetry.
1543: Note first that at any $\mu$, the fermionic gauge fixing condition is
1544: respected by $\tilde\epsilon$ that satisfy
1545: %
1546: \be
1547: \label{ress}
1548: \p_-\epsilon+k_+\p_-X^+\epsilon\equiv\frac{1}{F}\p_-\tilde\epsilon_+=0.
1549: \ee
1550: %
1551: At finite $\mu$, these apparent residual transformations do not preserve the
1552: bosonic part of our gauge fixing condition, $e_m{}^a=\delta_m^a$.
1553: In the $\mu=\infty$ limit, however, the supersymmetry transformation of
1554: $e_m^a$ is trivial, and all solutions of (\ref{ress}) survive as
1555: residual supersymmetry transformations.
1556:
1557: Unlike in superconformal gauge, this residual supersymmetry is
1558: {\it left-moving\/}.
1559: Moreover, the generator $\tilde\epsilon_+$ of local supersymmetry is of
1560: conformal dimension $(1/2,0)$; hence, local supersymmetry cannot be expected
1561: to square to a worldsheet conformal transformation. As is clear from
1562: (\ref{latesusy}), this local supersymmetry is in fact nilpotent.
1563:
1564: This residual supersymmetry $\tilde\epsilon(\sigma^+)$ can be fixed
1565: by a supplemental gauge choice, setting a purely left-moving fermion
1566: to zero. The one that suggests itself is $\lambda_+$, which in this
1567: gauge satisfies the free equation of motion; moreover, $\lambda_+$
1568: transforms very simply under $\tilde\epsilon$ supersymmetry, and
1569: setting it to zero fixes that symmetry completely. Once we add the
1570: condition $\lambda_+=0$ to the gauge choice $\psi_-^+=0$, this gauge
1571: now becomes very similar to the naive unitary gauge of
1572: Section~\ref{nunit}. This is another way of seeing that the
1573: goldstino is not an independent dynamical field, since it has been
1574: absorbed into the dynamics of the other fields in the process of
1575: setting the gravitino free.
1576:
1577: Alternatively, one can leave the residual supersymmetry unfixed, and instead
1578: impose the constraint (\ref{tildesc}) which reduces in the $\mu=\infty$ limit
1579: to
1580: %
1581: \be
1582: \lambda_+=2\alpha'V_-\p_+\tilde\psi_-^-.
1583: \ee
1584:
1585: Depending on whether the coupling of the quantum bosonic potential
1586: $\exp(2k_+X^+)$ is tuned to zero or not, we have two different
1587: late-time theories: One whose equations of motion expel
1588: all degrees of freedom to future infinity along $X^-$ for some
1589: constant $X^+$, the other allowing perturbations to reach large
1590: $X^+$. It is the latter theory which may represent a good set of variables
1591: suitable for understanding the physics at late $X^+$. If the potential is
1592: retained, the physical string modes are pushed away to infinity along $X^-$
1593: before reaching too deeply into the condensed phase, confirming that very few
1594: of the original stringy degrees of freedom are supported there.
1595:
1596: \section{Conclusions}
1597:
1598: \subsection{Overview}
1599:
1600: The main focus of this paper has been on examining the worldsheet theory
1601: of tachyon condensation in the $E_8$ heterotic string model. We studied
1602: the theory with a linear dilaton $V=V_-X^-$, and a tachyon profile
1603: $\CT(X^\mu)=2\mu\exp(k_+X^+)$. At first glance, this tachyon profile
1604: produces a worldsheet potential term which expels all degrees of
1605: freedom from the large $X^+$ region, indicating a possible
1606: topological phase, conjecturally related to the ``nothing'' phase in
1607: heterotic M-theory.
1608:
1609: We found that the worldsheet dynamics of tachyon condensation involves a
1610: super-Higgs mechanism, and that its analysis simplifies when local
1611: worldsheet supersymmetry is fixed in a new gauge, specifically conformal gauge
1612: augmented by $\psi^+_-=0$. Following a detailed analysis of
1613: one-loop measure effects, we found that exact quantum conformal invariance is
1614: maintained throughout in this gauge. At late times, the worldsheet theory
1615: contains a free left-moving propagating gravitino sector
1616: (obscured in superconformal gauge, as there the gravitino is set to zero).
1617: The gravitino sector contributes $-11$ units of central charge to the
1618: left-movers. In addition, the gauge fixing leads to a set of left-moving
1619: ghosts with $c=-1$, and a spacelike shifted linear dilaton $V=V_-X^-+k_+X^+$.
1620:
1621: In the process of making the gravitino dynamical, the worldsheet goldstino
1622: $\lambda_+$ has been effectively absorbed into the rest of the system;
1623: more precisely, the constraint generated by the alternative gauge can be
1624: solved by expressing $\lambda_+$ in terms of the remaining dynamical degrees
1625: of freedom.
1626:
1627: \subsection{Further Analysis of the $E_8$ System}
1628:
1629: In this paper, we have laid the groundwork for an in-depth analysis
1630: of the late time physics of the $E_8$ heterotic string under tachyon
1631: condensation. The emphasis here has been on developing the worldsheet
1632: techniques, aimed in particular at clarifying the super-Higgs mechanism.
1633: The next step, which we leave for future work, would be to examine the
1634: spacetime physics in the regime of late $X^+$ where the tachyon has
1635: condensed. There are signs indicating that this phase contains very few
1636: conventional degrees of freedom; more work is needed to provide further
1637: evidence for the conjectured relation between tachyon condensation in the
1638: $E_8$ string and the spacetime decay to nothing in $E_8\times\bar E_8$
1639: heterotic M-theory.
1640:
1641: It would be interesting to use the standard tools of string theory, combined
1642: with the new worldsheet gauge, to study the spectrum and scattering amplitudes
1643: of BRST invariant states in this background, in particular at late times. The
1644: use of mixed $R_\xi$ gauges could possibly extend the range of such an
1645: analysis, by interpolating between the superconformal gauge and its
1646: alternative.
1647:
1648: \subsection{Towards Non-Equilibrium String Theory}
1649:
1650: In the process of the worldsheet analysis presented in this paper, we
1651: found two features which we believe may be of interest to a broader
1652: class of time-dependent systems in string theory:
1653: (1) in superconformal gauge, the spacetime tachyon condensate turns the
1654: worldsheet theory into a logarithmic CFT; and
1655: (2) the worldsheet dynamics of some backgrounds may simplify
1656: in alternative gauge choices for worldsheet supersymmetry.
1657:
1658: We have only explored the first hints of the LCFT story and its utility in the
1659: description of string solutions with substantial time dependence.
1660: The new gauge choices, however, are clearly applicable to other systems.
1661: As an example, consider the Type 0 model studied
1662: in \cite{hs3}. We can pick a gauge similar to our alternative gauge
1663: (\ref{pixiegauge}) by setting $\psi^+_-=\psi^+_+=0$, again in addition to
1664: conformal gauge. We expect to simply double the gauge fixing
1665: procedure in Section \ref{liber}, producing one copy of $c=-1$ superghosts
1666: and one copy of the propagating gravitino sector in both the left and right
1667: movers. When the one-loop determinant effects are included, the linear
1668: dilaton shifts by $2k_+$, resulting in additional 24 units of central charge.
1669: Together with the two $c=-11$ sectors and the two $c=-1$ ghost sectors, this
1670: shift again leads $c_{tot}=0$, similarly to the heterotic model studied in
1671: the present paper. Exact conformal invariance is again
1672: maintained at the quantum level. This is in accord with the results of
1673: \cite{hs3}; however, our results are not manifestly equivalent to those of
1674: \cite{hs3}, as a result of a different gauge choice. In addition, our results
1675: also suggest an interpretation of the somewhat surprising appearance of
1676: fermionic first-order systems with $c=-11$ in \cite{hs3}: It is likely that
1677: they represent the re-emergence of the dynamical worldsheet gravitino, in
1678: superconformal gauge.
1679:
1680: This picture suggest a new kind of worldsheet duality in string theory.
1681: Instead of viewing one CFT in two dual ways while in superconformal gauge,
1682: the new duality is between two different CFTs representing the same solution
1683: of string theory, but in two different worldsheet gauges. The physics of
1684: BRST observables should of course be gauge independent, but this does not
1685: require the CFTs to be isomorphic. In fact, the alternative gauge studied
1686: in the body of this paper represents an example: It leads to a conformal,
1687: but not superconformal theory, yet it should contain the same physical
1688: information as the SCFT realization of the same string background in
1689: superconformal gauge.
1690:
1691: We hope that this technique of using gauge choices other than superconformal
1692: gauge will be applicable to a wider class of time dependent string
1693: backgrounds, beyond the case of models with tachyon condensation studied here,
1694: and that it will increase our understanding of non-equilibrium string
1695: theory.
1696:
1697: \acknowledgments
1698:
1699: One of us (PH) wishes to thank Allan Adams, Michal Fabinger, Uday Varadarajan
1700: and Bruno Zumino for useful discussions in the early stages of this project,
1701: in 2001-02. This work has been supported by NSF Grants PHY-0244900 and
1702: PHY-0555662, DOE Grant DE-AC03-76SF00098, an NSF Graduate Research Fellowship,
1703: and the Berkeley Center for Theoretical Physics.
1704: % And, most of all, by the graciousness of the Universe.
1705:
1706: \appendix
1707: \section{Supersymmetry Conventions}
1708: \label{appendix}
1709:
1710: In the NSR formalism, the tachyonic $E_8$ heterotic string is described by
1711: worldsheet supergravity with $(0,1)$ supersymmetry.%
1712: %
1713: \footnote{A useful source of information on two-dimensional supergravities is,
1714: {\it e.g.}, \cite{orange}.}
1715: %
1716: Here we list our conventions for worldsheet supergravity.
1717:
1718: \subsection{Flat worldsheet}
1719:
1720: In the worldsheet coordinates $\sigma^a=(\sigma^0\equiv\tau,\sigma^1
1721: \equiv\sigma)$, the flat Lorentz metric is
1722: %
1723: \begin{equation}
1724: \eta_{ab}=\pmatrix{-1&0\cr 0&1},
1725: \end{equation}
1726: %
1727: with gamma matrices given by
1728: %
1729: \begin{equation}
1730: \gamma^0=\pmatrix{0&1\cr -1&0},\qquad
1731: \gamma^1=\pmatrix{0&1\cr 1&0},
1732: \end{equation}
1733: %
1734: and satisfying $\{\gamma^a,\gamma^b\}=2\eta^{ab}$. We define the chirality
1735: matrix
1736: %
1737: \be
1738: \Gamma\equiv\gamma^0\gamma^1=\pmatrix{1&0\cr 0&-1}.
1739: \ee
1740: %
1741:
1742: The two-component spinor indices $\alpha$ are raised and lowered using the
1743: natural symplectic structure,
1744: %
1745: \be
1746: \xi^\alpha\equiv\varepsilon^{\alpha\beta}\xi_\beta,\qquad
1747: \varepsilon^{\alpha\beta}=\pmatrix{0&-1\cr 1&0},
1748: \ee
1749: %
1750: with $\varepsilon^{\alpha\beta}\varepsilon_{\beta\gamma}=
1751: \varepsilon_{\gamma\beta}\varepsilon^{\beta\alpha}=\delta^\alpha_\gamma$.
1752: This implies, for example, that $\xi^\alpha\zeta_\alpha=\zeta^\alpha\xi_\alpha=
1753: -\xi_\alpha\zeta^\alpha$ and $\xi\gamma^a\zeta=-\zeta\gamma^a\xi$, for any
1754: two real spinors $\xi,\zeta$.
1755:
1756: Any two-component spinor $\xi_\alpha$ can be decomposed into its chiral
1757: components, defined via
1758: %
1759: \be
1760: \xi_\alpha\equiv\pmatrix{\xi_+\cr\xi_-},\qquad
1761: \xi_\pm=\frac{1}{2}(1\pm\Gamma)\xi.
1762: \ee
1763:
1764: \subsection{Local worldsheet supersymmetry}
1765:
1766: On curved worldsheets, we will distinguish the worldsheet index $m$
1767: from the internal Lorentz index $a$. The spacetime index $\mu$ runs
1768: over $0,\ldots D-1$, typically with $D=10$. The heterotic string
1769: action with $(0,1)$ worldsheet supersymmetry is given by
1770: %
1771: \bea
1772: \label{hetaction}
1773: S&=&-\frac{1}{4\pi\alpha'}\int d^2\sigma\,e\left(\vphantom{\frac{1}{2}}
1774: \eta_{\mu\nu}\left(h^{mn}\p_mX^\mu\p_nX^\nu+i\psi^\mu\gamma^m\p_m\psi^\nu
1775: -i\kappa\chi_m\gamma^n\gamma^m\psi^\mu\p_nX^\nu\right)\right.
1776: \nonumber\\
1777: &&\qquad\qquad\qquad\qquad\left.{}+i\lambda^A\gamma^m\p_m\lambda^A
1778: -F^AF^A\vphantom{\frac{1}{2}} \right). \eea
1779: %
1780: The fermions and gravitino satisfy the following chirality
1781: conditions:
1782: %
1783: \be \Gamma \psi^\mu_-=-\psi^\mu_-, \qquad \Gamma
1784: \lambda_+=\lambda_+, \qquad \Gamma \chi_+=\chi_+ \ee
1785: %
1786: The action (\ref{hetaction}) is invariant under the supersymmetry
1787: transformations given by
1788: %
1789: \be
1790: \label{susyright}
1791: \delta X^\mu=i\epsilon\psi^\mu,\qquad\delta\psi^\mu=\gamma^m\p_m
1792: X^\mu\epsilon
1793: \ee
1794: %
1795: for the right-moving sector,
1796: %
1797: \be
1798: \label{susyleft}
1799: \delta\lambda^A=F^A\epsilon,\qquad\ \delta F^A=i\epsilon\gamma^mD_m\lambda^A
1800: \ee
1801: %
1802: for the left-movers, and
1803: %
1804: \be
1805: \label{susygrav}
1806: \delta e_m{}^a=i\kappa\epsilon\gamma^a\chi_m,\qquad
1807: \delta\chi_m=\frac{2}{\kappa}D_m\epsilon
1808: \ee
1809: %
1810: in the supergravity sector. Note that we set $\kappa=2$ in \cite{pixie}.
1811: Of course, $\gamma^m=e^m{}_a\gamma^a$, and the covariant derivative on
1812: spinors is
1813: %
1814: \be
1815: D_m\zeta=\left(\p_m+\frac{1}{4}\omega_m{}^{ab}\gamma_{ab}\right)\zeta,
1816: \ee
1817: %
1818: with $\gamma^{ab}=[\gamma^a,\gamma^b]/2$. In general, the spin connection
1819: $\omega_m{}^{ab}$ in supergravity contains the piece that depends solely on
1820: the vielbein, $\omega_m{}^{ab}(e)$, plus a fermion bilinear improvement term.
1821: In conformal $(0,1)$ supergravity in two dimensions as described by
1822: (\ref{hetaction}), however, the improvement term vanishes identically, and
1823: we have $\omega_m{}^{ab}=\omega_m{}^{ab}(e)$, with
1824: %
1825: \be
1826: \omega_m{}^{ab}=\frac{1}{2}e^{na}(\p_m e_n{}^b-\p_n e_m{}^b)
1827: -\frac{1}{2}e^{nb}(\p_m e_n{}^a-\p_n e_m{}^a)
1828: -\frac{1}{2}e^{na}e^{pb}(\p_n e_{pc}-\p_p e_{nc})e_m{}^c.
1829: \ee
1830:
1831: Note also that the susy variation of $F^A$ is sometimes written in the
1832: literature as $\delta F^A=i\epsilon\gamma^m\hat D_m\lambda^A$, using the
1833: supercovariant derivative
1834: %
1835: \be
1836: \label{gross}
1837: \hat D_m\lambda^A_+\equiv\left(\p_m+\frac{1}{4}\omega_m{}^{ab}\gamma_{ab}
1838: \right)\lambda_+^A-\chi_{m+}F^A.
1839: \ee
1840: %
1841: This simplifies, however, in several ways. First of all, the gravitino drops
1842: out from (\ref{gross}) if the $(0,1)$ theory is independent of the
1843: superpartner of the Liouville field, as is the case for our heterotic
1844: worldsheet supergravity. Secondly, for terms relevant for the action, we
1845: get
1846: %
1847: \be
1848: \label{lovely}
1849: \lambda^A\gamma^m\hat D_m\lambda^A_+\equiv\lambda^A\gamma^m\p_m\lambda^A_+.
1850: \ee
1851: %
1852:
1853: \subsection{Lightcone coordinates}
1854:
1855: The worldsheet lightcone coordinates are
1856: %
1857: \be
1858: \sigma^\pm=\tau\pm\sigma,
1859: \ee
1860: %
1861: in which the Minkowski metric becomes
1862: %
1863: \be
1864: \eta_{ab}=\pmatrix{0&-\frac{1}{2}\cr -\frac{1}{2}&0},
1865: \ee
1866: %
1867: resulting in the lightcone gamma matrices
1868: %
1869: \be
1870: \gamma^+=\pmatrix{0&2\cr 0&0},\qquad
1871: \gamma^-=\pmatrix{0&0\cr -2&0}.
1872: \ee
1873: %
1874: On spin-vectors, such as the gravitino $\chi_{m\alpha}$, we put the
1875: worldsheet index first, and the spinor index second when required.
1876: We will use $\pm$ labels for spinor indices as well as lightcone
1877: worldsheet and spacetime indices, as appropriate. The nature of a
1878: given index should be clear from context. In lightcone coordinates,
1879: and in conformal gauge, the supersymmetry transformations of the
1880: matter multiplets are given by
1881: %
1882: \bea
1883: \delta X^\mu&=&i\epsilon_+\psi^\mu_-,\qquad\delta\psi_-=-2\p_-X^\mu\epsilon_+,
1884: \nonumber\\
1885: \delta\lambda_+^A&=&F^A\epsilon_+,\qquad\ \delta F^A=-2i\epsilon_+
1886: \p_-\lambda_+^A,
1887: \eea
1888: %
1889: and the linearized supersymmetry transformations of the supergravity
1890: multiplet are
1891: %
1892: \be
1893: \delta e_+{}_-=-2i\kappa\epsilon_+\chi_{++},\qquad\delta\chi_{++}=
1894: \frac{2}{\kappa}\p_+\epsilon_+.
1895: \ee
1896:
1897: Once we have picked conformal gauge, we can meaningfully assign a
1898: conformal weight to each field. The chart below lists these
1899: conformal dimensions for all relevant objects:
1900: \smallskip
1901: %
1902: \be
1903: \begin{tabular}{|l|c|l|}
1904: \hline
1905: Field & Symbol & Conformal Weight \\
1906: \hline
1907: gravitino & $\chi_{++}$ & $(1,-\frac{1}{2})$ \\
1908: goldstino & $\lambda_+$ & $(\frac{1}{2},0)$ \\
1909: fermion & $\psi^\mu_-$ & $(0,\frac{1}{2})$ \\
1910: boson & $X^\mu$ & $(0,0)$ \\
1911: aux.\ field & $F$ & $(\frac{1}{2},\frac{1}{2})$ \\
1912: SUSY\ parameter & $\epsilon$ & $(0,-\frac{1}{2})$ \\
1913: w.s.\ derivative& $\p_-$ & $(0,1)$ \\
1914: w.s.\ derivative& $\p_+$ & $(1,0)$ \\
1915: \hline
1916: \end{tabular}
1917: \ee
1918:
1919: \noindent In spacetime, similarly, we define the lightcone
1920: coordinates thus:
1921: %
1922: \be
1923: X^\pm=X^0\pm X^1,
1924: \ee
1925: %
1926: and denote the remaining transverse dimensions by $X^i$, so that the spacetime
1927: index decomposes as $\mu\equiv(+,-,i)$.
1928:
1929: When we combine the spacetime lightcone parametrization with the lightcone
1930: coordinate choice on the worldsheet, the heterotic action becomes
1931: %
1932: \bea
1933: \label{lcaction}
1934: S&=&\frac{1}{\pi\alpha'}\int d^2\sigma^\pm\,\left(\p_+X^i\p_-X^i+\frac{i}{2}
1935: \psi_-^i\p_+\psi_-^i-\frac{1}{2}\p_+X^+\p_-X^--\frac{1}{2}\p_+X^-\p_-X^+\right.
1936: \nonumber\\
1937: &&\qquad\qquad\qquad\qquad-\frac{i}{4}\psi_-^+\p_+\psi_-^--\frac{i}{4}
1938: \psi_-^-\p_+\psi_-^++\frac{i}{2}\lambda_+^A\p_-\lambda_+^A+\frac{1}{4}F^AF^A\\
1939: &&\qquad\qquad\qquad\qquad\qquad\left.{}-\frac{i}{2}\kappa\chi_{++}
1940: (2\psi_-^i\p_-X^i-\psi_-^+\p_-X^--\psi_-^-\p_-X^+)\right),\nonumber
1941: \eea
1942: %
1943: where we have defined $d^2\sigma^\pm\equiv d\sigma^-\wedge d\sigma^+
1944: =2d\tau\wedge d\sigma$.
1945:
1946: \section{Evaluation of the Determinants}
1947: \label{appendixb}
1948:
1949: We wish to calculate the determinant for the operator
1950: %
1951: \be D_F\equiv \frac{1}{F}D_-F \ee
1952: %
1953: as it acts on fields of arbitrary half-integer spin $j$. $F$ here is
1954: a shorthand for the tachyon condensate $F=2\mu\exp(k_+X^+)$, as
1955: determined in (\ref{condensate}); we will continue with this
1956: notation through this appendix. As $D_F$ is a chiral operator, we
1957: will take the usual approach of finding its adjoint and calculating
1958: the determinant of the corresponding Laplacian. The actual
1959: contribution we are interested in will be the square root of the
1960: determinant of the Laplacian. Throughout this appendix, we work in
1961: Euclidean signature; we will Wick rotate our results back to
1962: Minkowski signature before adding the results to the body of the
1963: paper. Also, we gauge fix only worldsheet diffeomorphisms by
1964: setting
1965: %
1966: \be h_{mn}=e^{2\phi}\hat{h}_{mn},\ee
1967: %
1968: where $\phi$ is the Liouville field. For most calculations, we set
1969: the fiducial metric $\hat{h}_{mn}$ to be the flat metric. In this
1970: gauge, we find
1971: %
1972: \be D_-=e^{-2\phi}\bar\p, \ee
1973: %
1974: independently of $j$.
1975:
1976: We define the adjoint $D_F^\dagger$ of the Faddeev-Popov operator
1977: $D_F$ via
1978: %
1979: \be
1980: \label{conjdef}
1981: \bra{T_1}\frac{1}{F}D_-(FT_2)\rangle=\bra{D_F^\dagger T_1}T_2\rangle,
1982: \ee
1983: %
1984: where $T_1$ is a worldsheet tensor of spin $j-1$, $T_2$ is a tensor
1985: of spin $j$, and the inner product on the corresponding tensors is
1986: the standard
1987: one, independent of $F$.%
1988: %
1989: \footnote{See Section II.E of \cite{dhoker} for more details on the
1990: corresponding inner products and the definition and properties of
1991: differential operators on Riemann surfaces.}
1992: %
1993:
1994: We find the left hand side of (\ref{conjdef}) becomes
1995: %
1996: \bea
1997: \bra{T_1}\frac{1}{F}D_-FT_2\rangle=\int d^2z
1998: e^{2\phi(2-j)}T_1^*\left(\frac{1}{F}D_-F\right)T_2&=&\int d^2z
1999: e^{2\phi(2-j)}T_1^*\frac{1}{F}e^{-2\phi}\bar\p\left(FT_2\right)\nonumber\\
2000: {}=-\int d^2z \bar\p\left(e^{-2\phi
2001: (j-1)}\frac{1}{F}T_1^*\right)FT_2 = &-&\!\int d^2z e^{2\phi
2002: (1-j)}\left(F D_+ \frac{1}{F}T_1\right)^{\!*}\!T_2.
2003: \eea
2004: %
2005: Thus, the adjoint operator is
2006: %
2007: \be D_F^\dagger=-F D_+ \frac{1}{F}, \ee
2008: %
2009: where
2010: %
2011: \be
2012: D_+=e^{2\phi(j-1)}\p e^{-2\phi(j-1)} \ee
2013: %
2014: when acting on a field of spin $j-1$. As above, our conventions are
2015: such that $D_+$ and $\p$ act on everything to their right.
2016:
2017: We are now interested in the determinant of the Laplace operator
2018: $D_F^\dagger D_F$,
2019: %
2020: \be \det
2021: \left(-\frac{1}{F}D_+F^2D_-\frac{1}{F}\right)=\det\left(-\frac{1}{F^2}D_+F^2D_-
2022: \right). \ee
2023: %
2024: Determinants of such operators were carefully evaluated
2025: in \cite{Kallosh};%
2026: %
2027: \footnote{Similar determinants have played a central role in other areas of
2028: CFT, perhaps most notably in the free-field Wakimoto realizations of WZW
2029: models; see, {\it e.g.}, \cite{gerasimov}.}
2030: %
2031: Eqn.~(3.2) of that paper gives a general formula for the
2032: determinant of $f\p g\bar{\p}$. For our case,
2033: %
2034: \be f=e^{(2j-2)\phi+2k_+X^+},\qquad g=e^{-2j\phi-2k_+X^+},
2035: \ee
2036: %
2037: and we get
2038: %
2039: \bea \label{trivialdet} \log\det\left(D_F^\dagger D_F\right)
2040: &=&-\frac{1}{24\pi}\int d^2\sigma\, \hat
2041: e\left\{[3(2j-1)^2-1]\hat{h}^{mn}\p_m\phi\p_n\phi
2042: \right.\nonumber\\
2043: &+&12(2j-1)k_+\hat{h}^{mn}\p_m\phi\p_n X^+\left.+
2044: 12k_+^2\hat{h}^{mn}\p_m X^+\p_n X^+\right\}. \eea
2045: %
2046:
2047: \subsection{$J_{\psi_-^+\epsilon}$}
2048:
2049: The case of the Faddeev-Popov operator corresponds to spin $j=1/2$.
2050: Since we are in fact interested in the inverse square root of the
2051: determinant of the Laplacian, we find that it contributes to the
2052: effective Euclidean worldsheet action
2053: %
2054: \be \label{SgaugeN} -\frac{1}{48\pi}\int
2055: d^2\sigma\,\hat{e}\left(-\hat{h}^{mn}\p_m\phi\p_n\phi
2056: +12k_+^2\hat{h}^{mn}\p_mX^+\p_nX^+\right). \ee
2057: %
2058: We can rewrite this contribution as a path integral over a bosonic
2059: ghost-antighost system,
2060: %
2061: \be \frac{1}{\det
2062: J_{\psi_-^+\epsilon}}=\int\CD\beta\,\CD\gamma\,\exp \left\{-\int
2063: d^2\sigma\beta\frac{1}{F}D_-(F\gamma)\right\}. \ee
2064: %
2065: The antighost $\beta$ has the same conformal dimension as
2066: $\psi_-^+$, or $(0,1/2)$, while the ghost field $\gamma$ has the
2067: dimension of $\epsilon$, namely $(0,-1/2)$. The path-integral
2068: measure of the ghost and antighost fields, $\CD\beta\,\CD\gamma$, is
2069: defined in the standard way, independently of $X^+$. More
2070: precisely, the standard measure on the fluctuations $\delta f$ of a
2071: spin $j$ field $f$ is induced from the covariant norm
2072: %
2073: \be \label{snorm} \|\delta f\|^2=\int d^2\sigma e^{(2-2j)\phi}\delta
2074: f^\ast\delta f, \ee
2075: %
2076: written here in conformal gauge $h_{mn}=e^{2\phi}\delta_{mn}$.
2077:
2078: The ghost fields $\beta$ and $\gamma$ have a kinetic term suggesting that they
2079: should be purely left-moving fields, but their conformal dimensions do not
2080: conform to this observation. It is natural to introduce the rescaled fields
2081: %
2082: \be
2083: \label{rescgh}
2084: \tilde\beta=\frac{\beta}{F},\qquad\tilde\gamma=F\gamma.
2085: \ee
2086: %
2087: Because the conformal weight of $F$ is $(1/2,1/2)$, both $\tilde\beta$ and
2088: $\tilde\gamma$ are now fields of conformal dimension $(1/2,0)$. Moreover, in
2089: terms of these rescaled fields, the classical ghost action takes the canonical
2090: form of a purely left-moving first-order system of central charge $c=-1$,
2091: %
2092: \be
2093: S_{\tilde\beta\tilde\gamma}=\int d^2\sigma\,\tilde\beta D_-\tilde\gamma.
2094: \ee
2095: %
2096: However, the rescaling (\ref{rescgh}) has an
2097: effect on the measure in the path integral. In terms of the rescaled
2098: variables, the originally $X^+$ independent measure acquires a non-canonical,
2099: explicit $X^+$ dependence. The new measure on $\tilde\beta$ and
2100: $\tilde\gamma$ is induced from
2101: %
2102: \be
2103: \label{xnorm}
2104: \|\delta\tilde\beta\|^2=\int d^2\sigma e^\phi F^2\delta\tilde\beta^\ast
2105: \delta\tilde\beta,\qquad
2106: \|\delta\tilde\gamma\|^2=\int d^2\sigma e^\phi F^{-2}\delta\tilde\gamma^\ast
2107: \delta\tilde\gamma.
2108: \ee
2109: %
2110: In order to distinguish it from the standard $X^+$ independent
2111: measure, we denote the measure induced from (\ref{xnorm}) by
2112: $\CDX\tilde\beta\,\CDX\tilde\gamma$. It is convenient to
2113: replace
2114: this $X^+$ dependent measure by hand with the standard measure
2115: $\CD\tilde\beta \,\CD\tilde\gamma$ for left-moving fields
2116: $\tilde\beta$, $\tilde\gamma$ of spin 1/2, defined with the use of
2117: the standard norm (\ref{snorm}) that is independent of $X^+$. In
2118: order to do so, we must correct for the error by including the
2119: corresponding Jacobian, which is precisely the $X^+$ dependent part
2120: of $J_{\psi_-^+\epsilon}$. Thus, we can now write
2121: %
2122: \bea \label{Sgauge} &&\frac{1}{\det
2123: J_{\psi_-^+\epsilon}}=\int\CDX\tilde\beta\,\CDX\tilde\gamma
2124: \,
2125: \exp\left\{-\int d^2\sigma\,\tilde\beta D_-\tilde\gamma\right\}\\
2126: &&\quad{}=\exp\left\{\frac{1}{48\pi}\int d^2\sigma\,\hat
2127: e\left(12k_+^2\hat h^{mn}
2128: \p_mX^+\p_nX^+\right)\right\}\int\CD\tilde\beta\,\CD\tilde\gamma\,
2129: \exp\left\{-\int d^2\sigma\,\tilde\beta
2130: D_-\tilde\gamma\right\}.\nonumber \eea
2131: %
2132: The canonical $\tilde\beta$, $\tilde\gamma$ ghosts correctly
2133: reproduce the Liouville dependence of $J_{\psi_-^+\epsilon}$. Of
2134: course, we now have a contribution to the effective action. Written
2135: in Minkowski signature, this correction becomes
2136: %
2137: \be \label{SgaugeMink} \frac{1}{4\pi}\int d^2\sigma\,\hat e\left(k_+^2
2138: \hat h^{mn} \p_mX^+\p_nX^+\right). \ee
2139: %
2140:
2141: \subsection{$\tilde J$}
2142:
2143: The evaluation of $J_{\psi^+_-\epsilon}$ is not the only calculation
2144: that can contribute to the shift of the linear dilaton. We also
2145: need to analyze the determinant involved in the change of variables
2146: that turns the gravitino and its conjugate field into manifestly
2147: left-moving fields. As we shall now show, this transformation also
2148: contributes to the linear dilaton shift.
2149:
2150: Consider the kinetic term between the conjugate pair of $\chi_{++}$
2151: and $\psi_-^-$. The relevant part of the path integral, written
2152: here still in the Minkowski worldsheet signature, is
2153: %
2154: \be
2155: \int\CD\chi_{++}\CD\psi_-^-\,\exp\left\{\frac{i\kappa V_-}{\pi}\int
2156: d^2\sigma^\pm\left(\chi_{++}\p_-\psi_-^--k_+\chi_{++}\p_-X^+\psi_-^-\right)
2157: \right\}.
2158: \ee
2159: %
2160: This path integral would give the determinant of the operator $D_F$
2161: acting on fields of spin $j=3/2$. Wick rotating to Euclidean
2162: signature and using (\ref{trivialdet}), we obtain for this
2163: determinant
2164: %
2165: \be \label{gravdet} \exp\left\{-\frac{1}{48\pi}\int
2166: d^2\sigma\,e\left(11\hat{h}^{mn}\p_m\phi\p_n\phi
2167: +24k_+\hat{h}^{mn}\p_m\phi\p_nX^++
2168: 12k_+^2\hat{h}^{mn}\p_mX^+\p_nX^+\right)\right\}. \ee
2169: %
2170: As in the calculation of the Faddeev-Popov determinant above, it is
2171: again useful to rescale the fields by an $X^+$ dependent factor. This
2172: rescaling was in fact introduced in Eqn.~(\ref{rescale}), repeated here for
2173: convenience:
2174: %
2175: \be
2176: \tilde\chi_{++}=F\chi_{++},\qquad\tilde\psi_-^-=\frac{\psi_-^-}{F}.
2177: \ee
2178: %
2179: The first term in the conformal anomaly can be interpreted as due to
2180: the purely left-moving conjugate pair of fields $\tilde\chi_{++}$
2181: and $\tilde\psi_-^-$ with the standard $X^+$ independent measure
2182: induced from (\ref{snorm}). In order to reproduce correctly the
2183: full determinant (\ref{gravdet}), we again have to compensate for
2184: the $X^+$ dependence of the measure by including the rest of the
2185: gravitino determinant (\ref{gravdet}) as an explicit conversion
2186: factor. Thus, the consistent transformation of fields includes the
2187: measure change
2188: %
2189: \be \CDX\tilde\chi_{++}\,\CDX\tilde\psi_-^-= \tilde
2190: J\,\CD\tilde\chi_{++}\,\CD\tilde\psi_-^-, \ee
2191: %
2192: with
2193: %
2194: \be \tilde J=\exp\left\{-\frac{1}{4\pi}\int
2195: d^2\sigma\,e\left(2k_+\hat{h}^{mn}\p_m\phi\p_nX^++
2196: k_+^2\hat{h}^{mn}\p_mX^+\p_nX^+\right)\right\}. \ee
2197:
2198: Together with the contribution from the Faddeev-Popov determinant
2199: (\ref{Sgauge}), we see that the contributions to the $\p_+X^+\p_-X^+$ term
2200: in fact cancel, and we are left with the following one-loop correction to the
2201: Euclidean action due to the measure factors,
2202: %
2203: \be \label{effact} \Delta S_E=\frac{1}{2\pi}\int
2204: d^2\sigma\,e\,k_+h^{mn}\p_m\phi\p_nX^+, \ee
2205: %
2206: together with the bosonic superghosts of spin $1/2$ (and with the canonical
2207: path integral measure) and central charge $c=-1$, plus the gravitino sector
2208: consisting of the canonical conjugate pair $\tilde\chi_{++}, \tilde\psi_-^-$
2209: also with the canonical measure and central charge $c=-11$.
2210:
2211: Integrating (\ref{effact}) by parts and using the following
2212: expression for the worldsheet scalar curvature in conformal gauge,
2213: %
2214: \be
2215: eR=-2\delta^{mn}\p_m\p_n\phi,
2216: \ee
2217: %
2218: we end up with
2219: %
2220: \be \label{dshiftE}
2221: \Delta S_E=\frac{1}{4\pi}\int d^2\sigma\,e\,k_+X^+R.%
2222: \ee
2223: %
2224: Rotating back to Minkowski signature, we find
2225: %
2226: \be
2227: \label{dshift}
2228: \Delta S=-\frac{1}{4\pi} \int d^2\sigma\,e\,k_+X^+R.
2229: \ee
2230: %
2231: This term represents an effective shift in the dilaton by $V_+=k_+$.
2232: The total central charge of the combined matter and ghost system,
2233: with the one-loop correction (\ref{dshift}) to the linear dilaton
2234: included, is zero.
2235:
2236: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2237: \bibliographystyle{JHEP}
2238: \bibliography{e8}
2239: \end{document}
2240: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2241: