1: % JHEP.cls available at http://jhep.cern.ch/JOURNAL/tex.html

2: \documentclass[12pt,nohyper,letterpaper]{JHEP3}         % For preprint

3: %\documentclass[12pt,draft,nohyper]{JHEP3}    % For draft

4: \usepackage{amssymb,amsfonts}

5: %\usepackage{draftfil}

6: %\usepackage{showkeys}

7: %\usepackage{epsfig}

8: %gives names of refs and cites

9: %If you do not have the msbm fonts, delete the following 10 lines

10: \font\mybb=msbm10 at 10pt

11: %\font\mybb=msbm12 at 12pt

12: 

13: \renewcommand{\arraystretch}{1.2}

14: 

15: \usepackage{epsfig}

16: %\usepackage{amssymb}

17: 

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

19: %\renewcommand{\theequation}{\arabic{section}.\arabic{equation}}

20: %\def\appendix{{\newpage\section*{Appendix}}\let\appendix\section%

21: %        {\setcounter{section}{0}

22: %        \gdef\thesection{\Alph{section}}}\section}

23: 

24: 

25: %%%%%%%%%%%%%%%%%%%%%%%%%%    My Macros    %%%%%%%%%%%%%%%%%%%%%%%%%

26: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

27: 

28: \def\be{\begin{eqnarray}}

29: \def\ee{\end{eqnarray}}

30: \newcommand{\nn}{\nonumber}

31: \newcommand\para{\paragraph{}}

32: \newcommand{\ft}[2]{{\textstyle\frac{#1}{#2}}}

33: \newcommand{\eqn}[1]{(\ref{#1})}

34: \newcommand{\pl}[1]{\frac{\partial L}{\partial{#1}}}

35: \newcommand{\ppp}[2]{\frac{\partial {#1}}{\partial {#2}}}

36: \newcommand{\ph}[1]{\frac{\partial H}{\partial{#1}}}

37: 

38: \newcommand\balpha{\mbox{\boldmath $\alpha$}}

39: \newcommand\bbeta{\mbox{\boldmath $\beta$}}

40: \newcommand\bgamma{\mbox{\boldmath $\gamma$}}

41: \newcommand\bomega{\mbox{\boldmath $\omega$}}

42: \newcommand\blambda{\mbox{\boldmath $\lambda$}}

43: \newcommand\bmu{\mbox{\boldmath $\mu$}}

44: \newcommand\bphi{\mbox{\boldmath $\phi$}}

45: \newcommand\bzeta{\mbox{\boldmath $\zeta$}}

46: \newcommand\bsigma{\mbox{\boldmath $\sigma$}}

47: \newcommand\bepsilon{\mbox{\boldmath $\epsilon$}}

48: \newcommand\btau{\mbox{\boldmath $\tau$}}

49: \newcommand\beeta{\mbox{\boldmath $\eta$}}

50: \newcommand\btheta{\mbox{\boldmath $\theta$}}

51: \newcommand\valpha{\vec{\alpha}}

52: \newcommand\vg{\vec{g}}

53: 

54: \def\Dslash{\,\,{\raise.15ex\hbox{/}\mkern-12mu D}}

55: \def\Dbarslash{\,\,{\raise.15ex\hbox{/}\mkern-12mu {\bar D}}}

56: \def\delslash{\,\,{\raise.15ex\hbox{/}\mkern-9mu \partial}}

57: \def\delbarslash{\,\,{\raise.15ex\hbox{/}\mkern-9mu {\bar\partial}}}

58: \def\pslash{\,\,{\raise.15ex\hbox{/}\mkern-9mu p}}

59: \def\calDslash{\,\,{\raise.15ex\hbox{/}\mkern-12mu {\cal D}}}

60: \newcommand\Bprime{B${}^\prime$}

61: \newcommand{\sign}{{\rm sign}}

62: \newcommand{\D}{{\cal D}}

63: 

64: \newcommand\bx{{\bf x}}

65: \newcommand\br{{\bf r}}

66: \newcommand\bF{{\bf F}}

67: \newcommand\bp{{\bf p}}

68: \newcommand\bL{{\bf L}}

69: \newcommand\bR{{\bf R}}

70: \newcommand\bP{{\bf P}}

71: \newcommand\bE{{\bf E}}

72: \newcommand\bB{{\bf B}}

73: \newcommand\bA{{\bf A}}

74: \newcommand\bee{{\bf e}}

75: \newcommand\bte{\tilde{\bf e}}

76: \newcommand{\NN}{{\mathbb N}}

77: \newcommand{\Z}{{\bf Z}}

78: \newcommand{\ZZ}{{\mathbb Z}}

79: \newcommand{\Q}{{\bf Q}}

80: \newcommand{\QQ}{{\mathbb Q}}

81: \newcommand{\R}{{\bf R}}

82: \newcommand{\RR}{{\mathbb R}}

83: \newcommand{\C}{{\bf C}}

84: \newcommand{\CC}{{\mathbb C}}

85: \newcommand{\PP}{{\mathbb P}}

86: \newcommand{\e}{\,{\rm e}}

87: \newcommand{\CP}{{\bf CP}}

88: \newcommand{\bra}{\langle}

89: \newcommand{\ket}{\rangle}

90: \newcommand{\Tr}{{\rm Tr}}

91: \newcommand{\N}{{\cal N}}

92: \newcommand{\hW}{\hat{\cal W}}

93: 

94: 

95: \def\implies{\Rightarrow}

96: \def\To{\longrightarrow}

97: \def\longlongrightarrow{\relbar\joinrel\relbar\joinrel\rightarrow}

98: \def\ridiculousrightarrow{\relbar\joinrel\relbar\joinrel\relbar%

99: \joinrel\rightarrow}

100: 

101: \def\underarrow#1{\mathrel{\mathop{\longrightarrow}\limits_{#1}}}

102: \def\onnearrow#1{\mathrel{\mathop{\nearrow}\limits^{#1}}}

103: \def\undernearrow#1{\mathrel{\mathop{\nearrow}\limits_{#1}}}

104: \def\onarrow#1{\mathrel{\mathop{\longrightarrow}\limits^{#1}}}

105: \def\onArrow#1{\mathrel{\mathop{\longlongrightarrow}\limits^{#1}}}

106: \def\OnArrow#1{\mathrel{\mathop{\ridiculousrightarrow}\limits^{#1}}}

107: \def\lae{\mathrel{\mathop{\smash{\lower .5 ex \hbox{$\stackrel<\sim$}}}}}

108: \def\lae{\mathrel{\mathop{\smash{\lower .5 ex \hbox{$\stackrel>\sim$}}}}}

109: \def\eqq{\stackrel?=}

110: 

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

112: 

113: %%%%%%%%%%%%%%%%%%%%%   NOTES START  %%%%%%%%%%%%%%%%%%%%%%%%%%%

114: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

115: 

116: 

117: 

118: 

119: 

120: 

121: 

122: \title{Heterotic Vortex Strings}

123: \author{Mohammad Edalati\\

124: Department of Physics, \\

125: University of Cincinnati, \\

126: P. O. Box 210011,  \\

127: Cincinnati, OH 45221-0011, USA \\

128: {\tt edalati@physics.uc.edu}}

129: 

130: \author{David Tong\\

131: Department of Applied Mathematics and Theoretical Physics, \\

132: University of Cambridge, \\

133: Cambridge, CB3 0WA, UK\\{\tt d.tong@damtp.cam.ac.uk}}

134: 

135: \abstract{We determine the low-energy ${\cal N}=(0,2)$ worldsheet

136: dynamics of vortex strings in a large class of non-Abelian ${\cal

137: N}=1$ supersymmetric gauge theories.}

138: 

139: 

140: \begin{document}

141: \pagestyle{plain} \setcounter{page}{1}

142: \newcounter{bean}

143: \baselineskip16pt \setcounter{section}{0}

144: 

145: 

146: 

147: \tableofcontents

148: \section{Introduction}

149: 

150: Vortex strings provide a map between four-dimensional non-Abelian

151: gauge theories and two-dimensional sigma-models. The

152: four-dimensional theories in question have a  $U(N_c)$ gauge group

153: and a sufficient number of scalar fields to allow complete gauge

154: symmetry breaking, so that the system lies in the Higgs phase.

155: Theories with this property admit vortex strings. The embedding of

156: the vortex within the non-Abelian gauge group endows the string

157: with a number of orientation modes which parameterize the complex

158: projective space ${\bf CP}^{N_c-1}$. Further bosonic and fermionic

159: zero modes of the vortex live in line bundles over ${\bf

160: CP}^{N_c-1}$. In this manner, the low-energy dynamics of a single,

161: straight, infinite vortex string is described by some variant of

162: the ${\bf CP}^{N_c-1}$ sigma-model living on the $d=1+1$

163: dimensional worldsheet \cite{vib,auzzi}.

164: 

165: \para

166: When the four-dimensional gauge theory has ${\cal N}=2$

167: supersymmetry, a pleasing story emerges. The strings are

168: $1/2$-BPS, ensuring that the worldsheet dynamics inherits ${\cal

169: N}=(2,2)$ supersymmetry. It was shown in \cite{sy,vstring},

170: following earlier work of \cite{nick,dht}, that the quantum

171: dynamics of the worldsheet theory encodes quantitative information

172: about the quantum dynamics of the parent four-dimensional theory,

173: including the Seiberg-Witten curve and the exact BPS  mass

174: spectrum. More recently, the correspondence was extended to

175: superconformal points, with a matching between the scaling

176: dimensions of chiral primary operators in the four-dimensional

177: bulk and on the worldsheet \cite{scvs}. For a review of the

178: classical and quantum dynamics of these strings, see \cite{tasi}.

179: 

180: \para

181: The purpose of this paper is to present a detailed study of the

182: classical dynamics of vortex strings in ${\cal N}=1$

183: four-dimensional gauge theories. For certain choices of parameters

184: the strings once again preserve $1/2$ of supersymmetry, now

185: guaranteeing ${\cal N}=(0,2)$ supersymmetry on the worldsheet. For

186: this reason, we refer to vortices in ${\cal N}=1$ theories as

187: ``heterotic vortex strings". We will determine the explicit ${\cal

188: N}=(0,2)$ ${\bf CP}^{N_c-1}$ sigma-models, and their variations,

189: which describe the low-energy dynamics of vortex strings in a

190: large class of ${\cal N}=1$ gauge theories\footnote{Vortex strings

191: in various non-Abelian theories with less supersymmetry were

192: previously studied in [9-13]

193: %\cite{other,others,bolog,syss,gsy}

194: and in some cases qualitative agreement was found between the

195: dynamics of the worldsheet theory and the bulk. We will comment

196: more on the relationship of our work to some of these papers in

197: Section 4.}.

198: 

199: \para

200: The paper is organized as follows: Section 2 contains a detailed

201: discussion of the ${\cal N}=(2,2)$ worldsheet dynamics of vortex

202: strings in ${\cal N}=2$ four-dimensional theories. This section is

203: mostly a review of previous work, although explicit expressions

204: for bosonic and fermionic zero modes are provided which generalize

205: results in the literature from $U(2)$ gauge theories to $U(N_c)$

206: gauge theories. Particular attention is paid to the chirality of

207: different fermionic zero modes since this will prove important in

208: later sections. Section 3 also contains review material,

209: describing the basics of the superfield formalism for ${\cal

210: N}=(0,2)$ supersymmetry in $d=1+1$ dimensions.

211: 

212: \para

213: The meat of the paper is in Section 4. We consider two different

214: classes of  deformations, each of  which breaks the

215: four-dimensional supersymmetry from ${\cal N}=2$ to ${\cal N}=1$

216: through  the introduction of a superpotential for the adjoint

217: chiral multiplet. In each case, we show that there is a unique

218: ${\cal N}=(0,2)$ worldsheet theory which correctly captures all

219: BPS properties of the vortex string and predicts the interaction

220: of fermionic zero modes. We also include an appendix which

221: collates the notation for bulk and worldsheet fields used

222: throughout the paper.

223: 

224: 

225: 

226: 

227: 

228: 

229: \section{The ${\cal N}=(2,2)$ Dynamics of Vortex Strings}

230: 

231: In this section we review the dynamics of vortex strings in

232: four-dimensional gauge theories with ${\cal N}=2$ supersymmetry.

233: The vortices are $1/2$-BPS, ensuring that the $d=1+1$ dimensional

234: worldsheet dynamics of the string inherits ${\cal N}=(2,2)$

235: supersymmetry.

236: 

237: \subsection{The Four-Dimensional Theory}

238: 

239: Our starting point is the  $d=3+1$, ${\cal N}=2$ supersymmetric

240: $U(N_c)$ gauge theory, with $N_f$ flavors transforming in the

241: fundamental representation\footnote{Conventions: We pick Hermitian

242: generators $T^m$ with Killing form $\Tr\, T^mT^n= \ft12

243: \delta^{mn}$. We write the gauge field as $A_\mu=A_\mu^mT^m$ and

244: $F_{\mu\nu}=\partial_\mu A_\nu-\partial_\nu A_\mu -

245: i[A_\mu,A_\nu]$. Fundamental  covariant derivatives are ${\cal

246: D}_\mu Q=\partial_\mu Q-iA_\mu Q$; adjoint covariant derivatives

247: are ${\cal D}_\mu A =

248: \partial_\mu A - i [A_\mu,A]$. Our summation

249: conventions are inconsistent: a sum over repeated indices is

250: usually left implicit unless there is some ambiguity or a point

251: that requires emphasis.}. We describe the theory in the language

252: of four-dimensional ${\cal N}=1$ superfields. The ${\cal N}=2$

253: vector multiplet consists of an ${\cal N}=1$ vector multiplet $V$

254: and an ${\cal N}=1$ adjoint chiral multiplet $A$. Similarly, each

255: flavor hypermultiplet splits into two chiral multiplets, $Q_i$ and

256: $\tilde{Q}_i$ where $i=1,\ldots, N_f$ is the flavor index. Each

257: $Q_i$ transforms in the fundamental ${\bf N}_c$ of the gauge

258: group, while each $\tilde{Q}_i$ transforms in the anti-fundamental

259: $\bar{\bf N}_c$. We denote the complexified gauge coupling of the

260: theory as

261: %

262: \be \tau=\frac{2\pi i}{e^2} + \frac{\theta}{2\pi}\ .

263: \label{tau}\ee

264: %

265: The four

266: dimensional theory has the usual superpotential required for

267: ${\cal N}=2$ supersymmetry,

268: %

269: \be{\cal W}_{{\cal

270: N}=2}=\sqrt{2}\,\sum_{i=1}^{N_f}\tilde{Q}_iAQ_i\ .\label{4dn2}\ee

271: %

272: The scalar potential of the theory is dictated by the D-term and

273: the F-terms arising from this  superpotential. In components it is

274: given by,

275: %

276: \be V_{4d}&=& \frac{e^2}{2}\Tr(\,\sum_{i=1}^{N_f}\,Q_iQ_i^\dagger

277: - \tilde{Q}_i\tilde{Q}_i^\dagger - v^2\,1_{N_c})^2 +e^2\Tr|\,

278: \sum_{i=1}^{N_f}\tilde{Q}_iQ_i|^2\nn\\

279: &&+\sum_{i=1}^{N_f}\left(Q_i^\dagger \{A,A^\dagger\}Q_i  +

280: \tilde{Q}_i\{A,A^\dagger\}\tilde{Q}_i^\dagger\right) +

281: \frac{1}{2e^2}\Tr|[A,A^\dagger]|^2\label{v4dn2}\ee

282: %

283: where we have taken the liberty of denoting the component scalar

284: fields by the same Roman letter as the superfield in which they

285: reside. We have included a D-term Fayet-Iliopoulos (FI) parameter

286: $v^2$ for the central $U(1)\subset U(N_c)$. This is consistent

287: with ${\cal N}=2$ supersymmetry and forces the theory into the

288: Higgs phase, with $Q_i$ gaining a vacuum expectation value (vev).

289: For $N_f<N_c$, the rank condition ensures the D-term cannot vanish

290: and there is no supersymmetric ground state. We do not consider

291: this case. When $N_f>N_c$, the D-term and F-term conditions do not

292: fix the vevs of $Q_i$ and $\tilde{Q}_i$ completely and there is a

293: Higgs branch of vacua; we shall discuss this situation in Section

294: \ref{add}. For now we restrict attention to the case $N_f=N_c$ for

295: which there is a unique supersymmetric ground state in which the

296: gauge group is completely broken. Up to a gauge transformation the

297: ground state is given by

298: %

299: \be Q^a_{\ i}=v\delta^a_{\ i}\ \ \ ,\ \ \

300: \tilde{Q}_i=A=0\label{vac1}\ee

301: %

302: where $a=1,\ldots, N_c$ is the color index. The theory lies in the

303: color-flavor locked phase, with the vacuum expectation value

304: preserved by a simultaneous gauge and flavor rotation. The

305: symmetry breaking pattern is thus broken to the diagonal

306: combination of the two (recall that we are looking at the theory with $N_f=N_c$)

307: %

308: \be U(N_c)\times SU(N_f)\rightarrow SU(N_c)_{\rm

309: diag}\ .\label{breaking}\ee

310: %

311: 

312: 

313: 

314: \subsection{The Vortex}

315: 

316: The central $U(1)\subset U(N_c)$ does not survive the symmetry

317: breaking \eqn{breaking}, a fact which provides sufficient topology

318: to ensure the presence of vortex strings in the theory \cite{no}.

319: These vortices preserve $1/2$ of the supersymmetry, ensuring

320: ${\cal N}=(2,2)$ supersymmetric dynamics on their $d=1+1$

321: dimensional worldvolume. Infinite, straight strings oriented in

322: the $x^3$ direction satisfy the first-order equations,

323: %

324: \be F_{12}=e^2(\sum_{i=1}^{N_f}Q_iQ_i^\dagger - v^2\,1_{N_c})\nn\\

325: {\cal D}_{{z}}Q_i\equiv \frac{1}{2}({\cal D}_1Q_i-i{\cal D}_2Q_i)

326: =0 \label{vort}\ee

327: %

328: where $z=x^1+ix^2$ parameterizes the transverse plane.

329: %\footnote{Anti-vortices satisfies

330: %$F_{12}=-e^2(\sum_{i=1}^{N_f}Q_iQ_i^\dagger - v^2)$ and ${\cal

331: %D}_{\bar{z}}Q_i=0$.}.

332: These are the non-Abelian vortex equations.

333: %(For example, each

334: %side of the first equation is an $N_c\times N_c$ matrix).

335: Solutions to these equations have tension

336: %

337: \be T_k=2\pi k v^2 \ee

338: %

339: where $k =-\Tr\,\int (F_{12}/2\pi) \in {\bf Z}^+$ is the winding

340: number.

341: 

342: 

343: \para

344: \EPSFIGURE{profile.eps,height=110pt}{} Solutions to the vortex

345: equations with winding number $k$ have $2kN$ bosonic collective

346: coordinates. For a single $k=1$ vortex, they break down as

347: follows: there are $2$ collective coordinates corresponding to the

348: position of the string in the $z=x^1+ix^2$ plane. The remaining

349: $2(N-1)$ collective coordinates are Goldstone modes arising from

350: the action of the surviving symmetry \eqn{breaking} on the vortex

351: string. They parameterize $SU(N_c)/[SU(N_c-1)\times U(1)] \cong

352: {\bf CP}^{N_c-1}$ \cite{vib,auzzi}.

353: 

354: \para

355: An explicit realization of the orientational modes is most simply

356: given in {\it singular gauge} in which $Q$ does not wind

357: asymptotically, with the flux instead arising from a singular

358: gauge potential \cite{auzzi}. Suppose that the Abelian $N_c=1$

359: vortex equations are solved by two profile functions $q(\rho)$ and

360: $a(\rho)$, where $\rho=\sqrt{(x^1)^2+(x^2)^2}$ is the radial

361: distance from the string

362: %

363: \be Q_{\rm Abelian} = v q(\rho)\ \ \ \ {\rm and}\ \ \ \ (A_z)_{\rm

364: Abelian}= -i\bar{z} a(\rho)\ .  \ee

365: %

366: Here the complexified gauge connection is $A_z=\ft12 (A_1-iA_2)$.

367: Plugging this ansatz into the vortex equations gives two first

368: order ordinary differential equations,

369: %

370: \be  q'=2\rho aq\ \ \ {\rm  and} \ \ \ 4a+2\rho a'=e^2v^2

371: (q^2-1)\ee

372: %

373: with prime denoting the derivative with respect to $\rho$. These

374: equations are known to admit a unique solution satisfying the

375: appropriate boundary conditions,

376: %

377: \be q(\rho)\rightarrow \left\{\begin{array}{c} 1 \\ 0

378: \end{array}\right.\ \ \ \ ,\ \ \ \ \ a(\rho)\rightarrow

379: \left\{\begin{array}{lc} 0 \ \ \ \ \ & {\rm as}\ \rho\rightarrow

380: \infty

381: \\ 1/2\rho^2 & {\rm as}\ \rho\rightarrow 0\ .

382: \end{array}\right.\ee

383: %

384: However, the solution does not have a simple analytic form. A

385: sketch of the profile $q(\rho)$ is shown in figure 1.

386: 

387: \para

388: With the $k=1$ Abelian vortex solution in hand, one may simply

389: construct a solution to the non-Abelian equations by embedding

390: thus,

391: %

392: \be Q^a_{\ i}= \left(\frac{\phi^a\bar{\phi}_i}{r}\right)\,

393: v[q(\rho)-1]+v\delta^a_{\ i} \ \ \ \ {\rm and}\ \ \ \ (A_z)^a_{\

394: b} = -i\bar{z}a(\rho)\,\left(\frac{\phi^a\bar{\phi}_b}{r}\right) \

395: .\label{solutions}\ee

396: %

397: The $\phi^a \in {\bf C}^{N_c}$ define the orientation of the

398: vortex in the gauge and flavor groups. In order that this reduce

399: to the Abelian solution, we require

400: %

401: \be \sum_{a=1}^{N_c} |\phi^a|^2=r \label{phiphi}\ee

402: %

403: with $r$ a constant that will be fixed shortly. The solutions

404: \eqn{solutions} are invariant under the simultaneous rotation,

405: %

406: \be \phi^a\rightarrow e^{i\alpha}\phi^a\ .\label{iden}\ee

407: %

408: The $\phi^a$, subject to the constraint \eqn{phiphi} and

409: identification \eqn{iden}, provide homogeneous coordinates on

410: ${\bf CP}^{N_c-1}$. The $SU(N_c)$ symmetry of four-dimensions

411: descends to the vortex string, with the $\phi^a$ transforming in

412: the fundamental representation. This ensures that the ${\bf

413: CP}^{N_c-1}$ is endowed with the symmetric Fubini-Study metric.

414: The K\"ahler class of this space is $r$.

415: 

416: \para

417: A comment on notation: since $N_f=N_c$, both $Q^a_{\ i}$ and

418: $(A_z)^a_{\ b}$ are $N_c\times N_c$ matrices. In what

419: follows, we shall often neglect to write the indices on both. In

420: this notation, $Q$ is a matrix on which gauge rotations $U\in

421: U(N_c)$ act from the left, while flavor rotations $V\in SU(N_f)$

422: act from the right, so that $Q\rightarrow UQV^\dagger$.

423: 

424: \subsubsection{Bosonic Zero Modes}

425: 

426: For general winding number $k$, the vortex zero modes are defined

427: to be solutions to the linearized vortex equations,

428: %

429: \be {\cal D}_z\delta A_{\bar{z}}-{\cal D}_{\bar{z}}\delta A_z &=&

430: \frac{ie^2}{2}(\delta Q Q^\dagger + Q\delta Q^\dagger) \nn\\

431: {\cal D}_z \delta Q &=& i\delta A_z Q\ .\ee

432: %

433: These are to be supplemented with a suitable gauge fixing

434: condition which is derived from Gauss' law and reads

435: %

436: \be {\cal D}_z\delta A_{\bar{z}}+{\cal D}_{\bar{z}}\delta A_z =

437: -\frac{ie^2}{2}(\delta Q Q^\dagger - Q\delta Q^\dagger)\ .\ee

438: %

439: This gauge fixing condition combines with the first of the

440: linearized vortex equations to leave us with two, complex, first

441: order equations to be solved around the background of a fixed

442: vortex configuration,

443: %

444: \be 2{\cal D}_{\bar{z}}\delta A_z &=& -ie^2\delta Q Q^\dagger

445: \nn\\ {\cal D}_z \delta Q &=& i\delta A_z Q\label{bogzero}.\ee

446: %

447: We now derive the solutions to these equations that arise from the

448: symmetries of the system.

449: 

450: 

451: \subsubsection*{Translational Mode}

452: 

453: For any winding number $k$, the two translational modes are always

454: given by

455: %

456: \be \delta A_z=F_{\bar{z}z}\ \ \ {\rm and}\ \ \ \delta Q={\cal

457: D}_{\bar{z}} Q \label{transzero}\ee

458: %

459: which can be checked to satisfy \eqn{bogzero} using the fact that

460: the background fields obey the second order equations of motion.

461: 

462: \subsubsection*{Orientational Modes}

463: 

464: The zero modes corresponding to orientation are only slightly more

465: complicated. In general they can be written as

466: %

467: \be \delta A_z &=& {\cal D}_z\Omega \nn\\

468: \delta Q &=& i(\Omega Q - Q\hat{\Omega}).\label{orientzero}\ee

469: %

470: Here $\Omega(x)$ is an infinitesimal gauge rotation, while

471: $\hat{\Omega}$ is an infinitesimal flavor rotation. Since only the

472: diagonal subgroup \eqn{breaking} of these is preserved in the

473: vacuum, we require that $\Omega(x)\rightarrow \hat{\Omega}$ as

474: $x\rightarrow \infty$. In terms of our orientation coordinates

475: $\phi^i$, this diagonal rotation can be written as,

476: %

477: \be \hat{\Omega}^i_{\ j}=-i\left[\delta\phi^i\bar{\phi}_j - \phi^i

478: \delta \bar{\phi}_j - 2iu \phi^i\bar{\phi}_j\right]\ee

479: %

480: which holds for any $u$. Requiring that $\hat{\Omega} \in su(N_c)$

481: fixes $u$ to be

482: %

483: \be u = -i\bar{\phi}_i\,\delta\phi^i\ .\ee

484: %

485: Later $u$ will become a gauge field on the worldsheet whose role

486: is to implement the identification \eqn{iden}. For now, we can

487: treat $u$ as a connection and introduce the covariant variation

488: $\nabla\phi^i=\delta\phi^i-iu\phi^i$ which satisfies

489: $\nabla\phi^i\cdot\bar{\phi}_i=0$. In this notation

490: %

491: \be \hat{\Omega}^i_{\ j} =

492: -i[(\nabla\phi^i)\bar{\phi}_j-\phi^i\nabla\bar{\phi}_j]. \ee

493: %

494: The zero mode equations \eqn{bogzero} translate to the requirement

495: that $\Omega(x)$ satisfy the second order differential equation

496: %

497: \be {\cal D}^2\Omega = e^2 \left[\{\Omega,

498: QQ^\dagger\}-2Q\hat{\Omega} Q^\dagger\right].\label{hard}\ee

499: %

500: Everything above holds for arbitrary winding number $k$. For a

501: single vortex, with $k=1$, the solution to \eqn{hard} was provided

502: in \cite{others} (see equation (28) of that paper) and depends

503: only on the profile function $q(\rho)$ of the

504: vortex\footnote{Equation \eqn{osol}  solves \eqn{hard} by virtue

505: of the vortex profile obeying the second order equation

506: $4\partial_z\partial_{\bar{z}}q-4a^2\rho^2q=e^2v^2q(q^2-1)$.}

507: %

508: \be \Omega(\rho) = q(\rho)\,\hat{\Omega}.\label{osol}\ee

509: %

510: Using the solution \eqn{osol}, we can now be more explicit about

511: the orientation zero modes for a single vortex. Making use of the

512: vortex equations \eqn{vort}, we find

513: %

514: \be (\delta A_z)^a_{\ b} &=& -2i(\partial_z q)

515: \,(\nabla\phi^a)\bar{\phi}_b \nn\\ \delta Q^a_{\ i} &=&

516: v(q^2-1)\,(\nabla\phi^a)\bar{\phi}_i.\ee

517: %

518: 

519: 

520: \subsection{Fermions}

521: 

522: We now turn to a study of the fermionic zero modes \cite{jackr}. We start by

523: describing the Dirac equations in four-dimensions and their

524: solutions for a single $k=1$ vortex. We will pay particular

525: attention to the correlation between the  chirality of the

526: worldsheet and four-dimensional fermions.

527: 

528: \para

529: In the following we use four-dimensional Weyl fermions

530: $\psi_\alpha$ and $\bar{\lambda}^{\dot{\alpha}}$ with

531: $\alpha,\dot{\alpha}=1,2$. The notation is standard Wess and

532: Bagger fare \cite{wb} with, for example,

533: $\psi\lambda=\psi^\alpha\lambda_\alpha=\lambda\psi$ and

534: $\bar{\psi}\bar{\lambda}=\bar{\psi}_{\dot{\alpha}}\bar{\lambda}^{\dot{\alpha}}

535: =\bar{\lambda}\bar{\psi}$. Indices are raised and lowered with

536: $\epsilon^{\alpha\beta}=\epsilon^{\dot{\alpha}\dot{\beta}}=i\sigma_2$.

537: Our signature is mostly minus and we define

538: $(\sigma^\mu)_{\alpha\dot{\alpha}}=(-1,\sigma^i)$ and

539: $(\bar{\sigma}^\mu)^{\dot{\alpha}\alpha}=(-1,-\sigma^i)$.

540: 

541: \para

542: The ${\cal N}=2$ vector multiplet in four dimensions contains two

543: Weyl fermions, $\lambda$ and $\eta$, each transforming in the

544: adjoint representation of the $U(N_c)$ gauge group. The fermion

545: $\lambda$ lives in the ${\cal N}=1$ vector multiplet while $\eta$

546: lives in the adjoint chiral multiplet $A$. Each hypermultiplet

547: also contains two Weyl fermions, $\psi$ and $\tilde{\psi}$. These

548: live in $Q$ and $\tilde{Q}$, and transform in the ${\bf N}_c$ and

549: $\bar{\bf N}_c$ representations respectively. The Dirac equations

550: in the ${\cal N}=2$ theory are

551: %\footnote{Each of these comes with its complex

552: %conjugate equation. Just to get conventions right, let me spell

553: %out the first of these which is

554: %

555: %\be

556: %-i\Dslash\bar{\lambda}-i\sqrt{2}[{\eta},a^\dagger]-i\sqrt{2}{\psi}_iq_i^\dagger+

557: %i\sqrt{2}\tilde{q}_i^\dagger\tilde{\psi}_i=0\ee.}

558: %

559: \be

560: -\frac{i}{e^2}\Dbarslash\lambda-\frac{i\sqrt{2}}{e^2}[\bar{\eta},A]+i\sqrt{2}Q_i\bar{\psi}_i-

561: i\sqrt{2}\bar{\tilde{\psi}}_i\tilde{Q}_i&=&0 \nn\\

562: -\frac{i}{e^2}\Dbarslash\eta-\frac{i\sqrt{2}}{e^2}

563: [A,\bar{\lambda}]-\sqrt{2}\tilde{Q}^\dagger_i\bar{\psi}_i-

564: \sqrt{2}\bar{\tilde{\psi}}_i{Q}_i^{\dagger}&=&0 \label{diracs}\\

565: -i\Dbarslash\psi_i+i\sqrt{2}\bar{\lambda}Q_i-\sqrt{2}A^\dagger\bar{\tilde{\psi}}_i

566: -\sqrt{2}\bar{\eta}\tilde{Q}_i^\dagger &=&0\nn\\

567: -i\Dbarslash\tilde{\psi}_i-i\sqrt{2}\tilde{Q}_i\bar{\lambda}-\sqrt{2}\bar{\psi}_i

568: A^\dagger -\sqrt{2}Q_i^{\dagger}\bar{\eta}&=&0. \nn\ee

569: %

570: We wish to study these equations in the background of the vortex.

571: Here they simplify considerably since we have $A=\tilde{Q}_i=0$.

572: The equations decouple into two pairs: the first set of equations

573: are for $\lambda$ and $\bar{\psi}_i$

574: %

575: \be -\frac{i}{e^2}\Dbarslash\lambda+i\sqrt{2}Q_i\bar{\psi}_i=0\ \

576: \ &{\rm and}&\ \ \ \

577: -i\Dslash\bar{\psi}_i-i\sqrt{2}Q_i^{\dagger}\lambda=0.

578: \label{lambdap}\ee

579: %

580: The second set of equations are for $\eta$ and

581: $\bar{\tilde{\psi}}_i$,

582: %

583: \be

584: -\frac{i}{e^2}\Dbarslash\eta-\sqrt{2}\bar{\tilde{\psi}}_iQ_i^{\dagger}=0\

585: \ \ &{\rm and}&\ \ \ -i\Dslash\bar{\tilde{\psi}}_i-\sqrt{2}\eta

586: Q_i=0.\label{etap}\ee

587: %

588: 

589: \subsubsection{Chirality}

590: 

591: Each  pair of four-dimensional fermions gives rise to a fermi zero

592: mode on the vortex string of a specific chirality. Since this will

593: be important in later sections, we dwell on the point a little

594: here. The first step is to see which components of the spinors can

595: turn on in the background of a vortex or anti-vortex. We will need

596: the following identities,

597: %

598: \be

599: \Dslash_{\alpha\dot{\alpha}}\equiv(\sigma^\mu)_{\alpha\dot{\alpha}}

600: {\cal D}_\mu=2\left(\begin{array}{cc} -{\cal D}_- & {\cal D}_z \\

601: {\cal D}_{\bar{z}} & -{\cal D}_+\end{array}\right) \ \ {\rm and}\

602: \ \Dbarslash^{\dot{\alpha}\alpha}

603: \equiv(\bar{\sigma}^\mu)^{\dot{\alpha}\alpha}

604: {\cal D}_\mu=-2\left(\begin{array}{cc} {\cal D}_+ & {\cal D}_z \\

605: {\cal D}_{\bar{z}} & {\cal D}_-\end{array}\right)\label{ds}\ \ \ \

606: \ \ \  \ee

607: %

608: where ${\cal D}_\pm=\ft12({\cal D}_0\pm{\cal D}_3)$ and ${\cal

609: D}_z=\ft12({\cal D}_1-i{\cal D}_2)$ and ${\cal

610: D}_{\bar{z}}=\ft12({\cal D}_1+i{\cal D}_2)$. Our strings are

611: static and oriented in the $x^3$ direction, so in searching for

612: zero modes of the Dirac equation in the presence of a vortex we

613: may initially set ${\cal D}_\pm=0$. We decompose the spinors as

614: $(\lambda_1,\lambda_2)=(\lambda_-,\lambda_+)$ and

615: $(\lambda^1,\lambda^2)=(\lambda^-,\lambda^+)$ so that, with our

616: raising and lowering conventions, $\lambda^+=-\lambda_-$ and

617: $\lambda^-=\lambda_+$. To see which components turn on in the

618: background of the vortex, we act on the first and second equations

619: in \eqn{lambdap} with $\Dslash$ and $\Dbarslash$ respectively.

620: Making use of the vortex equation ${\cal D}_{{z}}Q_i=0$, we find

621: %

622: \be &&(-\frac{4}{e^2}{\cal D}_z{\cal

623: D}_{\bar{z}}+2Q_iQ_i^{\dagger})\lambda_- =0\nn\\&&

624: (-\frac{4}{e^2}{\cal D}_{\bar{z}}{\cal

625: D}_z+2Q_iQ_i^{\dagger})\lambda_+ - \sqrt{2}({\cal

626: D}_{\bar{z}}Q_i)\,\bar{\psi}_{+i}=0 \ee

627: %

628: and

629: %

630: \be &&(-{\cal D}_z{\cal D}_{\bar{z}}\delta_{ij}

631: +2Q_iQ_j^{\dagger})\bar{\psi}_{+j} - \sqrt{2}({\cal D}_z

632: Q_i^{\dagger})\, \lambda_+ =0\nn\\

633: &&(-{\cal D}_{\bar{z}}{\cal D}_z\delta_{ij}+2Q_i^{\dagger}

634: Q_j)\bar{\psi}_{-j}  =0. \ee

635: %

636: %

637: %\be (-{\cal D}_z{\cal D}_{\bar{z}}+2Q_iQ^{i\dagger})\lambda_-

638: %-\sqrt{2}{\cal D}_zQ_i\,\bar{\psi}^{i+}

639: %&=&0\nn\\

640: %(-{\cal D}_{\bar{z}}{\cal D}_z+2Q_iQ^{i\dagger})\lambda_+&=&0 \ee

641: %

642: %and

643: %%

644: %\be (-{\cal D}_z{\cal D}_{\bar{z}}\delta^j_{\

645: %i}+2Q_iQ^{j\dagger})\bar{\psi}^{j-}&=&0\nn\\

646: %(-{\cal D}_{\bar{z}}{\cal D}_z\delta_j^{\ i}+2Q^{i\dagger}

647: %Q_j)\bar{\psi}^{j+} - \sqrt{2}{\cal D}_{\bar{z}}Q^{i\dagger}

648: %\lambda_- &=&0 \ee

649: %

650: The operators appearing in the equations for $\lambda_-$ and

651: $\bar{\psi}_{-i}$ are positive definite: these components can have

652: no zero modes. All zero modes live in the components $\lambda_+$

653: and $\bar{\psi}_{+i}$.

654: 

655: \para

656: To see how this correlates with the chirality of the worldsheet

657: fermions, we now allow these zero modes to vary along the string

658: so that $\lambda_+=\lambda_+(x^0,x^3)$ and

659: $\bar{\psi}_{+i}=\bar{\psi}_{+i}(x^0,x^3)$. Plugging this ansatz

660: back into the Dirac equation, including now the derivatives ${\cal

661: D}_\pm$ in \eqn{ds}, we find the equations of motion

662: ${\partial}_-\lambda_+=\partial_-\bar{\psi}_{+i}=0$. We call these

663: fermions {\it right movers}.

664: 

665: \para

666: Repeating this analysis for Dirac equations \eqn{etap}, we find

667: that $\eta_-$ and $\bar{\tilde{\psi}}_{-i}$ both carry zero modes

668: in the background of the vortex. They are {\it left movers} on the

669: string worldsheet\footnote{In the background of an anti-vortex,

670: with  ${\cal D}_{\bar{z}}Q_i=0$, the chirality of the fermi zero

671: modes is reversed, so that $(\lambda, \bar{\psi})$ donate left

672: movers, while $(\eta,\bar{\tilde{\psi}})$ donate right movers.}.

673: 

674: \subsubsection{Fermi Zero Modes}

675: 

676: From the previous analysis, we learn that the right moving fermi

677: zero modes solve

678: %

679: \be \sqrt{2}{\cal D}_z\lambda_+ &=& -e^2\,Q_i\bar{\psi}_{+i} \nn\\

680: \sqrt{2}{\cal D}_{\bar{z}} \bar{\psi}_{+i}&=&-

681: Q_i^\dagger\lambda_+\ee

682: %

683: while the equations for the left moving fermi zero modes solve

684: %

685: \be \sqrt{2}i{\cal D}_{\bar{z}} \eta_- &=& -e^2

686: \bar{\tilde{\psi}}_{-i} Q_i^\dagger \nn\\

687: \sqrt{2}i{\cal D}_z \bar{\tilde{\psi}}_{-i} &=& \eta_-Q_i.\ee

688: %

689: Each of these pairs of equations is the same as the equations for

690: bosonic zero modes \eqn{bogzero} that are derived by linearizing

691: the vortex equations and imposing a gauge fixing constraint. The

692: relationship between the bosonic and fermionic  zero modes is

693: given by

694: %

695: \be \lambda_+\leftrightarrow \delta A_{\bar{z}}\ \ &{\rm  and}& \

696: \

697: \sqrt{2}\bar{\psi}_{+i}\leftrightarrow -i\delta Q_i^\dagger \nn\\

698: \eta_-\leftrightarrow \delta A_z \ \ &{\rm and}& \ \

699: \sqrt{2}\bar{\tilde{\psi}}_{-i}\leftrightarrow -\delta

700: Q_i.\label{relzero}\ee

701: %

702: This mapping between the zero mode profiles is a consequence of

703: the preserved supersymmetry in the background of the vortex. Using

704: this, it is trivial to derive the explicit zero modes in the case

705: of a single $k=1$ vortex.

706: 

707: \subsubsection*{Goldstino Modes}

708: 

709: The bosonic translational modes were given in \eqn{transzero}.

710: Their fermionic counterparts are

711: %

712: \be \lambda_+=F_{z\bar{z}}\bar{\chi}_+ \ \ &{\rm and}& \ \

713: \bar{\psi}_{+i}=-\frac{i}{\sqrt{2}} {\cal D}_{z}Q_i^\dagger

714: \bar{\chi}_+

715: \nn\\

716: \eta_-=F_{\bar{z}z}\chi_- \ \ &{\rm and}& \ \

717: \bar{\tilde{\psi}}_{-i}=-\frac{1}{\sqrt{2}} {\cal D}_{\bar{z}}Q_i

718: \chi_- .\label{goldy}\ee

719: %

720: Both of these are  Goldstino modes, arising from acting on the

721: bosonic vortex profile \eqn{solutions} with the two broken

722: supersymmetries parameterized by $\chi_\pm$.  The above formulae

723: hold for arbitrary $k$; if we restrict to the explicit $k=1$

724: solution, we may write these in terms of the vortex profile

725: function $q(\rho)$,

726: %

727: \be  (\lambda_+)^a_{\ b}= \frac{ie^2v^2}{2r}

728: (q^2-1)\phi^a\bar{\phi}_i\bar{\chi}_+ \ \  &{\rm and}& \ \

729: \bar{\psi}^A_{+i}=-\frac{i\sqrt{2}v}{r}(\partial_zq)\,{\phi^a\bar{\phi}_b}

730: \bar{\chi}_+

731: \nn\\

732: \eta_-= -\frac{ie^2v^2}{2r} (q^2-1)\phi^a\bar{\phi}_b\chi_-

733:  \ \ &{\rm and}& \ \ \bar{\tilde{\psi}}^a_{-i}=- \frac{\sqrt{2}v}{r}

734:  (\partial_{\bar{z}}q)\,{\phi^a\bar{\phi}_i}

735: \chi_- .\label{goldy2}\ee

736: %

737: 

738: 

739: \subsubsection*{Super-Orientation Modes}

740: 

741: The superpartners of the orientational modes are equally easy to

742: write down. Given the bosonic zero modes \eqn{orientzero}, we have

743: %

744: \be (\lambda_+)^a_{\ b}=2i (\partial_{\bar{z}}q)\,

745: \phi^a\bar{\xi}_{+b}

746:  \ \ &{\rm and}& \ \

747:  \bar{\psi}^a_{+i}=-\frac{iv}{\sqrt{2}}(q^2-1)\phi_i\bar{\xi}^a_{+}

748: \nn\\

749: (\eta_-)^a_{\ b}= -2i(\partial_z q)\, {\xi}^a_-\bar{\phi}_b \ \

750: &{\rm and}& \ \ \bar{\tilde{\psi}}^a_{-i}=

751: -\frac{v}{\sqrt{2}}(q^2-1) \xi_-^a \bar{\phi}_i .\label{somodes}\ee

752: %

753: It is clear from these expressions, that the redundancy \eqn{iden}

754: which acts among the $\phi_i$ orientational coordinates, must also

755: act on the superpartners $\xi_{\pm i}$, so that

756: %

757: \be \phi_i\rightarrow e^{i\alpha}\phi_i\ \ \ {\rm and} \ \ \

758: \xi_i\rightarrow e^{i\alpha}\xi_i .\label{iden2}\ee

759: %

760: Moreover, the fact that there do not exist orientational

761: coordinates in the $N=1$ Abelian theory means that we must impose

762: a constraint on the $\xi_{\pm i}$, namely

763: %

764: \be \sum_{i=1}^{N_c}\bar{\phi}^i\xi_{\pm i}=0 .\label{fermcons}\ee

765: %

766: 

767: 

768: \subsection{Supersymmetric Dynamics}

769: 

770: 

771: The low-energy dynamics of the vortex string arises by promoting

772: the collective coordinates $z$, $\chi_\pm$, $\phi_i$ and $\xi_{\pm

773: i}$ to dynamical fields on the string worldsheet, depending on

774: $y^0\equiv x^0$ and $y^1\equiv x^3$. The fact that the vortices

775: are BPS, preserving 1/2 of the ${\cal N}=2$ four-dimensional

776: supersymmetry, ensures that the resulting worldsheet dynamics is

777: invariant under  ${\cal N}=(2,2)$ supersymmetry. Indeed, the

778: various bosonic and fermionic collective coordinates are easily

779: packaged into ${\cal N}=(2,2)$ superfields. The translational mode

780: $z$ and the two Goldstino modes $\chi_\pm$ sit in an ${\cal

781: N}=(2,2)$ chiral multiplet $Z$. Our notation is

782: standard\footnote{The one deviation from standard notation is to

783: label the complex auxiliary fields in each chiral multiplet as

784: $G$. This distinguishes them from the auxiliary $F$ fields in

785: four-dimensions.} and follows, for example, \cite{phases}

786: %

787: \be

788: Z=z+\theta^+\chi_++\theta^-\chi_-+\theta^+\theta^-G_Z+\ldots\ .\ee

789: %

790: Similarly, the orientation modes $\phi_i$ and their superpartners

791: $\xi_{\pm i}$ also sit in $(2,2)$ chiral multiplets,

792: %

793: \be

794: \Phi_i=\phi_i+\theta^+\xi_{+i}+\theta^-\xi_{-i}+\theta^+\theta^-G_i+\ldots\ .

795: \ee

796: %

797: The two constraints $\bar{\phi}^i\phi_i=r$, and

798: $\bar{\phi}^i\xi_{\pm i}=0$, together with the identification

799: \eqn{iden2}, are imposed on the worldsheet theory by introducing

800: an auxiliary ${\cal N}=(2,2)$ vector multiplet which, in

801: Wess-Zumino gauge, has components

802: %

803: \be

804: U&=&-\theta^-\bar{\theta}^-(u_0-u_1)+\theta^+\bar{\theta}^+(u_0+u_1)

805: - \theta^-\bar{\theta}^+\sigma -

806: \theta^+\bar{\theta}^-\bar{\sigma}

807: \\ && + \sqrt{2}{i}\theta^i\theta^+(\bar{}\theta}^-\bar{\zeta}_-

808: +\bar{\theta}^+\bar{\zeta}_+)+\sqrt{2}{i\bar{\theta}^+\bar{\theta}^-

809: (\theta^-\zeta_-+\theta^+\zeta_+)+2\theta^-\theta^+

810: \bar{\theta}^+\bar{\theta}^- D .\nn\ee

811: %

812: The two dimensional field strength

813: $u_{01}=\partial_0u_1-\partial_1 u_0$ is naturally housed in a

814: twisted chiral multiplet, defined by $\Sigma =

815: \bar{D}_+D_-U/\sqrt{2}$, with component expansion

816: %

817: \be \Sigma = \sigma -

818: i\sqrt{2}\theta^+\bar{\zeta}_+-i\sqrt{2}\bar{\theta}^-\zeta_-+\sqrt{2}

819: \theta^+\bar{\theta}^-(D-iu_{01})+\ldots\ .\label{twistedc}\ee

820: %

821: The fields $\sigma$, $\zeta_\pm$ and $D$ are all auxiliary. Their

822: role will become clear shortly.

823: 

824: \para

825: With the exception of a single integration constant $t$, the

826: dynamics of a $k=1$ vortex string is fixed entirely by the

827: symmetries of the theory. In particular, the $SU(N_c)_{\rm diag}$

828: symmetry of \eqn{breaking} descends to an $SU(N_c)$ global

829: symmetry on the worldsheet, under which the $\Phi_i$ transform in

830: the fundamental ${\bf N_c}$ representation. The resulting

831: dynamics is given by

832: %

833: \be {\cal L}_{\rm vortex} = \int d^4\theta\ T\bar{Z}Z+

834: \sum_{i=1}^{N_c}\bar{\Phi}_ie^{2U}\Phi_i + \frac{it}{2\sqrt{2}}\int d\theta^+

835: d\bar{\theta}^-\ \Sigma\ee

836: %

837: where $T=2\pi v^2$ is the tension of the vortex, while

838: %

839: \be t = ir+\frac{\theta}{2\pi}\ee

840: %

841: is the integration constant that needs to be fixed, and plays the

842: role of a complexified worldsheet FI parameter. After integrating

843: out the auxiliary fields $G_Z$ and $G_i$, the purely bosonic part

844: of the worldsheet Lagrangian reads,

845: %

846: \be {\cal L}_{\rm bose}=T\,|\partial_m z|^2 +

847: \sum_{i=1}^{N_c}\left(|{\cal D}_m\phi^i|^2 -

848: 2|\sigma|^2|\phi_i|^2\right)

849: +D(\sum_{i=1}^{N_c}|\phi_i|^2-r)+\frac{\theta}{2\pi}u_{01}.\ \ \ \

850: \ \ee

851: %

852: Here the $\phi_i$ fields carry charge $+1$ under the gauge

853: symmetry, with ${\cal D}\phi_i=\partial\phi_i - iu\phi_i$.

854: Dividing out by this symmetry imposes the identification

855: \eqn{iden}: $\phi_i\rightarrow e^{i\alpha}\phi_i$. Meanwhile, the

856: $D$-field in this Lagrangian plays the role of a Lagrange

857: multiplier, imposing the condition \eqn{phiphi}:

858: $\sum_i|\phi_i|^2=r$. The value of $r$ is fixed by the requirement

859: that the kinetic terms for $\phi_i$ are canonical. One finds the

860: result,

861: %

862: \be r=\frac{2\pi}{e^2}.\ee

863: %

864: This result was first shown using a brane construction in

865: \cite{vib}, and later re-derived by explicitly computing the

866: overlap of zero modes\footnote{This follows by taking the time

867: dependent ansatz for the orientational modes: ${\cal

868: D}_tQ_i=\delta Q_i$ and $F_{0z}=\delta A_z$, with

869: $\delta\phi_i=\dot{\phi}_i$. Inserting this into the four

870: dimensional kinetic terms gives,

871: %

872: \be \int dx^1dx^2 \left(\frac{1}{2e^2}F_{0i}^2+|{\cal

873: D}_tQ_i|^2\right) = \int dx^1dx^2 \left(\frac{1}{e^2}|{\cal

874: D}_iq|^2 +v^2(q-1)^2(q+1)^2\right)\frac{|{\cal D}_t\phi_i|^2}{r} =

875: \frac{2\pi}{e^2} \frac{|{\cal D}_t\phi_i|^2}{r}\nn\ee where the

876: integral is recognized as the same one that appears in computing

877: the vortex tension.} in \cite{others}. Similarly, it can be shown

878: that the four-dimensional $\theta$-angle descends to a worldsheet

879: $\theta$-angle \cite{vstring,others}. The end result is that

880: worldsheet complexified FI parameter $t$ is identified with the

881: bulk complexified gauge coupling $\tau$ \eqn{tau}:

882: %

883: \be t = \tau\ .\ee

884: %

885: We now turn to the fermionic part of the worldsheet Lagrangian,

886: given by

887: %

888: \be L_{\rm fermi} &=& 2iT\left(\bar{\chi}_-\partial_+\chi_-

889: +\bar{\chi}_+\partial_-\chi_+\right) +

890: 2i\sum_{i=1}^{N_c}\left(\bar{\xi}_{-i}{\cal D}_+\xi_{-i} +

891: \bar{\xi}_{+i}{\cal D}_-\xi_{+i}\right)\\ &&

892: -\sqrt{2}\sum_{i=1}^{N_c}\left(

893: \bar{\sigma}\bar{\xi}_{+i}\xi_{-i}+\sigma \bar{\xi}_{-i}\xi_{+i}

894: +\bar{\phi}_i(\xi_{-i}\zeta_+-\xi_{+i}\zeta_-)+\phi_i(\bar{\zeta}_-\bar{\xi}_{+i}

895: -\bar{\zeta}_+\bar{\xi}_{-i})\right)\nn\ee

896: %

897: The fermions $\xi_{\pm i}$ both have charge $+1$ under the $U(1)$

898: gauge symmetry, which is now seen to implement the full

899: identification \eqn{iden2}: $\phi_i\rightarrow e^{i\alpha}\phi_i$

900: and $\xi_\pm\rightarrow e^{i\alpha}\xi_{\pm}$. The vector

901: multiplet fermions $\zeta_\pm$ have no kinetic term and act as

902: Grassmannian Lagrange multipliers, imposing the constraint

903: \eqn{fermcons}: $\sum_i\bar{\phi}_i\xi_{\pm i}=0$. Finally, the

904: role of $\sigma$ is to mediate a four-fermi interaction for the

905: super-orientation modes. Upon integrating out $\sigma$, we have

906: %

907: \be {\cal L}_{\rm 4-fermi} =

908: -2|\sigma|^2|\phi_i|^2-\sqrt{2}\bar{\sigma}\bar{\xi}_{+i}\xi_{-i}-\sqrt{2}

909: \sigma\bar{\xi}_{-i} \xi_{+i} =

910: -\frac{|\bar{\xi}_{-i}\xi_{+i}|^2}{r}.\label{224fermi}\ee

911: %

912: Four-fermi terms of this kind are typical for soliton dynamics in

913: supersymmetric theories. We pause here to review how they arise.

914: In deriving the Dirac equations \eqn{lambdap} and \eqn{etap} we

915: set $A=\tilde{Q}_i=0$. This is valid in the background of the

916: bosonic vortex. However, it is no longer true in the presence of

917: fermions since fermi bilinears act as a source for these fields.

918: For example, the Yukawa couplings involving $A^\dagger$ contribute

919: to the equation of motion,

920: %

921: \be {\cal D}^2A +

922: {i\sqrt{2}}[\lambda,\eta]-\sqrt{2}e^2\bar{\tilde{\psi}}_i\tilde{\psi}_i

923: +e^2\{Q_iQ_i^\dagger + \tilde{Q}_i^\dagger\tilde{Q}_i, A\} =0

924: \label{yukyuk}\ee

925: % \nn\\ {\cal D}^2\tilde{Q}_i &=& i\sqrt{2}\tilde{\psi}_i\lambda

926: %- \sqrt{2}\eta\psi_i\ee

927: %

928: The solution to this equation then feeds back into the Dirac

929: equations \eqn{diracs} and must be solved iteratively, order by

930: order in the number of Grassmannian collective coordinates. This

931: is a finite, but somewhat complicated procedure (see \cite{stefan}

932: for a simple quantum mechanical model where it may be carried

933: through to completion). Thankfully, the end result \eqn{224fermi}

934: is dictated by supersymmetry.

935: 

936: 

937: \subsubsection{Symmetries}

938: \label{symmetries}

939: 

940: The four-dimensional ${\cal N}=2$ theory has two $U(1)$

941: R-symmetries\footnote{In the absence of the FI parameter $v^2$,

942: the theory has an $SU(2)_R$ symmetry, under which $(\lambda,\eta)$

943: and $(Q,\tilde{Q}^\dagger)$ both transform as doublets. The FI

944: parameter breaks $SU(2)_R\rightarrow U(1)_V$.} that we will call

945: $U(1)_R$ and $U(1)_V$. The charges of the various fields under

946: $U(1)_R\times U(1)_V$ are listed in the table.

947: %

948: \begin{center}\begin{tabular}{c|ccccccc} & $A$ & $\lambda$ & $\eta$ &

949: $Q$ & $\tilde{Q}$ & $\psi$

950: & $\tilde{\psi}$ \\ \hline $U(1)_R$ & 2 & 1 & 1 & 0 & 0 & -1 & -1 \\

951: $U(1)_V$ & 0 & 1 & -1 & 1 & 1 & 0 & 0\ .

952: \end{tabular}\end{center}

953: %

954: Both of these symmetries descend to the vortex worldsheet, where

955: they appear as the two R-symmetries of the ${\cal N}=(2,2)$

956: superalgebra. The action on the fermionic collective coordinates

957: of the vortex can  can be read directly from  \eqn{relzero}. We

958: have

959: %

960: \begin{center}\begin{tabular}{c|ccccccccc} & $\sigma$ & $\zeta_+$

961: & $\zeta_-$ & $\phi$ & $\xi_+$ & $\xi_-$ & $z$ & $\chi_+$ &

962: $\chi_-$ \\ \hline $U(1)_R$ & 2 & -1 & 1 & 0 & -1 & 1 & 0 & -1 & 1 \\

963: $U(1)_V$ & 0 & 1 & 1 & 0 & -1 & -1 & 0 & -1 & -1

964: \\ $U(1)_Z$ & 0 & 1 & 1 & 0 & -1 & -1 & 2 & 1 & 1\ .

965: \end{tabular}\end{center}

966: %

967: The $U(1)_R$ symmetry is axial; it suffers an anomaly in the

968: quantum theory of the vortex (as, indeed, does the $U(1)_R$ in

969: four-dimensions). In contrast $U(1)_V$ is a vector R-symmetry on

970: the worldsheet.

971: 

972: \para

973: The vortex theory also includes a further global $U(1)_Z$

974: symmetry, which arises from rotating the vortex string in the

975: $z=x^1+ix^2$ plane. The charges of the worldsheet fields under

976: $U(1)_Z$ are listed in the table and follow from \eqn{goldy} and

977: \eqn{somodes}. There exists a suitable linear combination of

978: $U(1)_Z$ and $U(1)_V$ which simply rotates the phase of the chiral

979: multiplet $Z$, leaving all other fields invariant.

980: 

981: \para

982: There are other, translational, symmetries of the worldsheet

983: theory that reflect the fact that $z$ and $\chi_\pm$ are all

984: Goldstone modes, arising from broken translation and supersymmetry

985: invariance respectively. In both cases, this ensures they have

986: only derivative couplings. In particular, it is the existence of

987: these symmetries that prevents the Goldstino modes $\chi_\pm$ from

988: appearing in the four-fermi term \eqn{224fermi}.

989: 

990: 

991: 

992: \subsection{${\cal N}=2$ Preserving Deformations}

993: 

994: So far we have described vortices in only the simplest ${\cal

995: N}=2$ theory with $N_f=N_c$. There are a number of ways to deform

996: and augment our theory that preserve ${\cal N}=2$ supersymmetry.

997: Here we list them and describe their effect on the worldsheet. We

998: postpone until Section 4 a discussion of deformations that break

999: the four dimensional supersymmetry to ${\cal N}=1$.

1000: 

1001: \subsubsection{Adding Masses}

1002: 

1003: The simplest deformation of our theory that preserves ${\cal N}=2$

1004: supersymmetry is to add a complex mass parameter $m_i$ for each

1005: hypermultiplet. The superpotential \eqn{4dn2} now becomes

1006: %

1007: \be {\cal W}_{{\cal

1008: N}=2}=\sqrt{2}\sum_{i=1}^{N_f}\tilde{Q}_i(A-m_i)Q_i .\ee

1009: %

1010: The vacuum \eqn{vac1} survives only if we turn on the adjoint

1011: scalar field $A$ to cancel the  F-term contributions,

1012: %

1013: \be Q^a_{\ i}=v\delta^a_{\ i}\ \ \ ,\ \ \ \tilde{Q}_i=0\ \ \ \ , \

1014: \ \ A={\rm diag}(m_1,\ldots,m_{N_c}) .\label{vac2}\ee

1015: %

1016: The vortex moduli space does not fare well under this deformation.

1017: It can be simply shown that the masses $m_i$ lift the internal

1018: ${\bf CP}^{N_c-1}$ vortex moduli space, leaving behind $N_c$

1019: distinct, isolated vortex solutions, each of which carries

1020: magnetic flux in a different diagonal $U(1)$ subgroup of the

1021: $U(N_c)$ gauge group, supported by a different $Q_i$ winding at

1022: infinity.

1023: 

1024: \para

1025: It was shown in \cite{memono,sy,vstring} that the 4d masses $m_i$

1026: induce ``twisted masses" \cite{hh} for the fields on the vortex

1027: worldsheet. In the language of ${\cal N}=(2,2)$ superfields, this

1028: deformation replaces the standard kinetic terms for $\Phi_i$

1029: by

1030: %

1031: \be \sum_{i=1}^{N_c}\bar{\Phi}_i e^{2U}\Phi_i \longrightarrow

1032: \sum_{i=1}^{N_c}\ \bar{\Phi}_i

1033: \exp\left(2U-2\theta^-\bar{\theta}^+m_i-2\theta^+\bar{\theta}^-m_i^\dagger\right)

1034: \Phi_i.\ee

1035: %

1036: In terms of components, the vortex theory (neglecting for now the

1037: $Z$ multiplet whose dynamics remains unchanged) becomes,

1038: %

1039: \be {\cal L}_{\rm vortex} &=& \sum_{i=1}^{N_c}\left(|{\cal D}_m\phi_i|^2

1040: +|F_i|^2- 2|\sigma-m_i|^2|\phi_i|^2\right)

1041: +D(\sum_{i=1}^{N_c}|\phi_i|^2-r)+\frac{\theta}{2\pi}u_{01} \nn\\

1042: && +\sum_{i=1}^{N_c} 2i\left(\bar{\xi}_{-i}{\cal D}_+\xi_{-i} +

1043: \bar{\xi}_{+i}{\cal D}_-\xi_{+i}\right) -\sqrt{2}

1044: (\bar{\sigma}-\bar{m}_i)\bar{\xi}_{+i}\xi_{-i}

1045: -\sqrt{2}(\sigma-m_i) \bar{\xi}_{-i}\xi_{+i} \nn\\ &&

1046: -i\sqrt{2}\sum_{i=1}^{N_c}\Big(

1047: \bar{\phi}_i(\xi_{-i}\zeta_+-\xi_{+i}\zeta_-)+\phi_i(\bar{\zeta}_-\bar{\xi}_{+i}

1048: -\bar{\zeta}_+\bar{\xi}_{-i})\Big).\ee

1049: %

1050: Note that the masses $m_i$ break the $U(1)_R$ symmetry, both in

1051: four-dimensions and on the vortex worldsheet. The twisted masses

1052: on the worldsheet have the desired effect of lifting the ${\bf

1053: CP}^{N_c-1}$ moduli space of the vortex theory, leaving behind

1054: $N_c$ isolated vacua given by

1055: %

1056: \be |\phi_i|^2 = r\delta_{ij}\ \ \ ,\ \ \ \sigma = m_j\ \ \ \

1057: j=1,\ldots, N_c .\ee

1058: %

1059: These different vacua of the worldsheet theory are identified with

1060: the different vortex solutions in four-dimensions. Kinks

1061: interpolating between these vacua on the worldsheet correspond to

1062: magnetic monopoles in four-dimensions, confined to lie on the

1063: vortex string by the Meissner effect \cite{memono}.

1064: 

1065: 

1066: \subsubsection{Adding Flavors}

1067: \label{add}

1068: 

1069: We now consider the theory with $N_f>N_c$ fundamental

1070: hypermultiplets. The D-term and F-term vacuum conditions in

1071: four-dimensions  read

1072: %

1073: \be \sum_{i=1}^{N_f}Q_iQ_i^\dagger - \tilde{Q}_i^\dagger

1074: \tilde{Q}_i = v^2\ \ \ {\rm and}\ \ \ \sum_{i=1}^{N_f}

1075: Q_i\tilde{Q}_i= 0 .\label{higgsbranch}\ee

1076: %

1077: When $m_i=0$, there are no further conditions, and there is a

1078: $2N_c(N_f-N_c)$ dimensional Higgs branch of the theory. For the

1079: purposes of this section, we place ourselves in the  particular

1080: vacuum $\tilde{Q}_i=0$ and

1081: %

1082: \be Q^a_{\ i}=v\delta^a_{\ i} \ \ \ {a,i=1,\ldots N_c}

1083: \label{0vac}\ee

1084: %

1085: with $Q_i=0$ for $i=N_c+1,\ldots, N_f$. This is to be supplemented

1086: by $A=0$. If we now turn on masses for the hypermultiplets, the

1087: vacuum \eqn{0vac} survives, with $A={\rm

1088: diag}(m_1,\ldots,m_{N_c})$.

1089: 

1090: 

1091: \para

1092: Vortices in theories with $N_f>N_c$ have a rather different

1093: character than those in the $N_f=N_c$ theory. The most noticeable

1094: difference is that they gain extra bosonic collective coordinates,

1095: among them a scale size. These additional collective coordinates

1096: are non-normalizable when $m_i=0$ \cite{ward,ls} but become

1097: normalizable when finite masses $m_i$ are turned on for

1098: $i=N_c+1,\ldots, N_f$ \cite{semi}. Vortices of this kind are

1099: sometimes referred to as semi-local vortices: a review of these

1100: objects in Abelian theories can be found in \cite{semi2}, while a

1101: detailed discussion in non-Abelian theories was given in

1102: \cite{semi}.

1103: 

1104: 

1105: \para

1106: An effective dynamics for the vortex worldsheet in theories with

1107: $N_f>N_c$ was proposed in \cite{vib}, based on a D-brane

1108: construction. It is once again an ${\cal N}=(2,2)$ supersymmetric

1109: $U(1)$ gauge theory, now  with $N_c$ chiral multiplets $\Psi_i$ of

1110: charge $+1$ and a further $(N_f-N_c)$ chiral multiplets

1111: $\tilde{\Psi}_j$ of charge $-1$. The D-term for this theory reads

1112: %

1113: \be D= \sum_{i=1}^{N_c} |\phi_i|^2 -

1114: \sum_{j=1}^{N_f-N_f}|\tilde{\phi}_j|^2-r = 0 \label{tilphi}\ee

1115: %

1116: which, together with the gauge action $\phi_i\rightarrow

1117: e^{i\alpha}\phi_i$ and $\tilde{\phi}_i\rightarrow

1118: e^{-i\alpha}\tilde{\phi}_i$, defines the Higgs branch of the

1119: vortex theory. This Higgs branch  is conjectured to coincide the

1120: vortex moduli space. As in the previous section, assigning complex

1121: masses $m_i$ to the four dimensional hypermultiplets induces

1122: twisted masses $m_i$, $i=1,\ldots,N_c$ for the $\Phi_i$ fields,

1123: and twisted masses $\tilde{m}_j=m_{j+N_c}$, $j=1,\ldots, N_f-N_c$

1124: for $\tilde{\Phi}_j$.

1125: 

1126: \para

1127: The presence of the negatively charged fields $\tilde{\phi}_j$

1128: means that the moduli space \eqn{tilphi} is now non-compact,

1129: corresponding to the scaling mode of the vortex. Note however that

1130: the natural metric on the Higgs branch does not coincide with the

1131: natural metric on the vortex moduli space. In particular, the

1132: non-normalizability of the scaling modes as $m_i\rightarrow 0$ is

1133: not reproduced in this model.  Nonetheless, it has been shown that

1134: the vortex theory \eqn{tilphi} does indeed correctly capture the

1135: quantum dynamics of the vortex string \cite{vstring,scvs}.

1136: 

1137: 

1138: 

1139: 

1140: \subsubsection*{Higgs Expectation Values}

1141: 

1142: When $N_f>N_c$ and $m_i=0$, the vacuum conditions

1143: \eqn{higgsbranch} in the four-dimensional theory have a moduli

1144: space of solutions. We may ask what happens to the vortex string

1145: as we change the expectation values of $Q_i$ and $\tilde{Q}_i$

1146: such that \eqn{higgsbranch} remains satisfied. The answer to this

1147: question was given in \cite{scvs}: turning on expectation values

1148: for $\tilde{Q}_i$ induces a superpotential on the vortex string

1149: worldsheet. For completeness, we briefly describe this deformation

1150: here.

1151: 

1152: \para

1153: First some notation: define the gauge invariant meson operator

1154: %

1155: \be M_i^j\equiv \tilde{Q}^j Q_i\label{mij} .\ee

1156: %

1157: It is not hard to show that in four-dimensional vacua for which

1158: $M\neq 0$, the space of BPS vortex solutions is greatly reduced.

1159: The key point is that a vacuum expectation value for $\tilde{Q}$

1160: does not allow a BPS vortex to live in the associated part of the

1161: gauge group. This follows from the mathematical fact that there is

1162: no holomorphic line bundle of negative degree. In a more physical

1163: language, a direct analysis of the vortex equations reveals that

1164: BPS vortices do not exist in $U(1)$ theories if both negatively

1165: and positively charged fields gain an expectation value

1166: \cite{penin,davis}. The upshot of this is that the vortex moduli

1167: space is partly lifted in four dimensional vacua for which $M\neq

1168: 0$. It was shown in \cite{scvs} that this effect is captured on

1169: the vortex worldsheet by the introduction of a superpotential of

1170: the form,

1171: %

1172: \be {\cal W}_{(2,2)}\sim \sum_{i=1}^{N_f} \sum_{j=1}^{N_f-N_c}

1173: M^{j+N_c}_i \,\tilde{\Phi}_j\Phi^i .\ee

1174: %

1175: 

1176: 

1177: 

1178: 

1179: \subsection{Multiple Vortices}

1180: 

1181: %

1182: \EPSFIGURE{vib.eps,height=160pt}{}

1183: %

1184: \noindent So far we have focussed on the theory for a single

1185: vortex string for which, at least in the case $N_f=N_c$,

1186: symmetries are sufficient to dictate the dynamics. In \cite{vib},

1187: a D-brane construction was used to derive a worldsheet theory

1188: which describes the interactions of $k>1$ parallel vortex strings.

1189: The D-brane construction starts with the usual  Hanany-Witten

1190: set-up for ${\cal N}=2$ four-dimensional gauge theories

1191: \cite{hanwit,w}, consisting of D4-branes attached to parallel

1192: NS5-branes. Separating the NS5-branes in the direction out of the

1193: page induces the FI parameter $v^2$. The vortex strings arise as

1194: stretched D2-branes as shown in figure 2. The worldvolume theory

1195: of $k$ vortex strings is given by an ${\cal N}=(2,2)$  $U(k)$

1196: non-Abelian gauge theory with matter content,

1197: %

1198: \para

1199: 

1200: \be \mbox{$U(k)$ Vector Multiplet $U$} &+& \mbox{Adjoint Chiral

1201: Multiplet $Z$}\nn\\ &+& \mbox{$N_c$ Fundamental Chiral Multiplets

1202: $\Phi_i$} \nn\\ &+& \mbox{$N_f-N_c$ Anti-Fundamental Chiral

1203: Multiplets $\tilde{\Phi}_j$} . \nn\ee

1204: %

1205: The complexified worldsheet FI parameter is again equated to the

1206: 4d complexified gauge coupling, $t=\tau$, or

1207: %

1208: \be ir+\frac{\theta_{2d}}{2\pi}=\frac{2\pi

1209: i}{e^2}+\frac{\theta_{4d}}{2\pi}\label{fi} . \ee

1210: %

1211: The D-term condition for the worldsheet theory is now $u(k)$

1212: valued and is given by,

1213: %

1214: \be \sum_{i=1}^{N_c}\phi_i\phi_i^\dagger

1215: -\sum_{j=1}^{N_f-N_c}\tilde{\phi}_j^\dagger\tilde{\phi}_j+T[z,z^\dagger]=r\,1_k .

1216: \label{vortexd}\ee

1217: %

1218: This provides $k^2$ constraints on the $2k(N_f+k)$ degrees of

1219: freedom in $\phi_i$, $\tilde{\phi}_j$ and $z$. After dividing by

1220: $U(k)$ gauge transformations, we are left with a $2kN_f$

1221: dimensional manifold which defines the target space for the vortex

1222: string sigma-model. This was conjectured in \cite{vib} to coincide

1223: with $2kN_f$ dimensional vortex moduli space. This quotient

1224: construction has subsequently been derived from a direct analysis

1225: of the non-Abelian vortex equations \cite{mmatrix,mmreview}.

1226: 

1227: \para

1228: %The fermionic terms in the $U(k)$ worldsheet theory are given by,

1229: %

1230: %\be L_{\rm fermi}&=&2iT\ \Tr_k\left(\bar{\chi}_-{\cal

1231: %D}_+\chi_-+\bar{\chi}_+{\cal D}_-\chi_+\right) +

1232: %2i\sum_{i=1}^{N_c}\left(\bar{\xi}_{-i}{\cal

1233: %D}_+\xi_{-i}+\bar{\xi}_{+i}{\cal D}_-\xi_{+i}\right) \nn\\

1234: %&&

1235: %-\sqrt{2}\,\Tr_k\left([\bar{\chi}_-,[\sigma,\chi_+]]-[\bar{\chi}_+,[\bar{\zeta}_-,

1236: %z]]-[\bar{\xi}_-,[\bar{\zeta}_+,z]]\right) + {\rm h.c.} \label{22phew}\\

1237: %&& -\sqrt{2}\sum_{i=1}^{N_c}\left( \bar{\xi}_{-i}\sigma\xi_{+i}

1238: %-\bar{\xi}_{+i}\bar{\zeta}_-\phi_i +

1239: %\bar{\xi}_{-i}\bar{\zeta}_+\phi_i\right) + {\rm h.c.} \nn\ee

1240: %%

1241: The $4kN_c$ fermionic zero modes of $k$ parallel vortex strings

1242: live in the $U(k)$ adjoint valued $\chi_\pm$ and the fundamental

1243: $\xi_{\pm i}$, subject to the $2k^2$ complex constraints arising

1244: from the auxiliary fermions $\zeta_\pm$,

1245: %

1246: \be \sum_i\phi_i\bar{\xi}_{\pm i}+[z,\bar{\chi}_\pm]=0 .\ee

1247: %

1248: In the case of a single $k=1$ vortex, these reduce to the

1249: constraints \eqn{fermcons}.

1250: 

1251: \para

1252: Note that the vacuum moduli space \eqn{vortexd} inherits a metric

1253: from the canonical kinetic terms for $\phi_i$ and $z$. This metric

1254: is known not to agree with the standard Manton metric

1255: \cite{manton,samols} on the vortex moduli space (except in the

1256: special case $k=1$ and $N_f=N_c$ that we described in detail

1257: earlier). This is because the limit in which the $d=1+1$ gauge

1258: theory on the D2-branes decouples from other stringy modes is

1259: different from the limit in which the D2-branes are described as

1260: vortices in the $d=3+1$ dimensional theory on the D4-branes; the

1261: two descriptions hold in different regimes of validity as we vary

1262: the parameters of the brane set-up. Nonetheless, if one is

1263: interested in computing objects protected by supersymmetry

1264: --- such as the classical, or quantum, masses of BPS states in the

1265: vortex theory ---- it should be valid to work with the gauge

1266: linear sigma model. In practice, this claim has been confirmed

1267: only for the $k=1$ theory with $N_f>N_c$. It has also been

1268: confirmed that the intricate topology of the $k=2$ vortex string

1269: moduli space in the $N_f=N_c=2$ is correctly captured by the

1270: gauged linear sigma model \cite{asy,kt,lots}.

1271: 

1272: 

1273: \section{${\cal N}=(0,2)$ Supersymmetry}

1274: 

1275: In the previous section we have worked with both ${\cal N}=1$

1276: superfields in four dimensions, $A$, $Q_i$ and $\tilde{Q}_j$, as

1277: well as ${\cal N}=(2,2)$ superfields in two dimensions, $\Sigma$,

1278: $\Phi_i$ and $\tilde{\Phi}_j$. From now on we will deal with

1279: ${\cal N}=(0,2)$ supersymmetry in two dimensions. Since this may

1280: be less familiar to some readers we devote this section to a

1281: review of the structure of ${\cal N}=(0,2)$ superfields \cite{hw,ds} and

1282: their relationship to ${\cal N}=(2,2)$ theories. The presentation

1283: follows \cite{phases} and \cite{abs}.

1284: 

1285: \subsection{Superfields}

1286: 

1287: ${\cal N}=(0,2)$ supersymmetry is generated by two right-moving,

1288: and no left-moving, supersymmetries. The two chiral supercharges

1289: are $Q_+$ and $\bar{Q}_+$. The $(0,2)$ superspace is parameterized

1290: by the bosonic coordinates $y^\pm=(y^0\pm y^1)$ and their

1291: fermionic partners $\theta^+$ and $\bar{\theta}^+$. The action of

1292: the supersymmetry generators in superspace is given as

1293: %

1294: \be

1295: Q_+&=&\frac{\partial}{\partial\theta^+}+i\bar{\theta}^+(\partial_0+\partial_1)

1296: %\ \ \ \ ,\ \ \ \

1297: \nn\\

1298: \bar{Q}_+&=&-\frac{\partial}{\partial\bar{\theta}^+}-i\theta^+(\partial_0+\partial_1) .

1299: \ee

1300: %

1301: These commute with the superderivatives,

1302: %

1303: \be D_+&=&\frac{\partial}{\partial\theta^+}-i\bar{\theta}^+

1304: (\partial_0+\partial_1)

1305: %\ \ \ \ ,\ \ \ \

1306: \nn\\

1307: \bar{D}_+&=&-\frac{\partial}{\partial\bar{\theta}^+}+i\theta^+

1308: (\partial_0+\partial_1) \ee

1309: %

1310: which satisfy $\{D_+,D_+\}=\{\bar{D}_+\bar{D}_+\}=0$ and

1311: $\{D_+,\bar{D}_+\}=2i\partial_+$. We now describe the different

1312: superfields of interest.

1313: 

1314: \subsubsection*{Gauge Multiplets}

1315: 

1316: We start with the real, adjoint valued,  gauge multiplet $U$,

1317: which has the component expansion

1318: %

1319: \be

1320: U=(u_0-u_1)-2i\theta^+\bar{\zeta}_--2i\bar{\theta}^+\zeta_-+2\theta^+\bar{\theta}^+D .

1321: \label{02vec}\ee %%

1322: Already we see the chiral nature of the supersymmetry, since only

1323: the combination $u_-=u_0-u_1$ of the two-dimensional gauge field

1324: appears in the superfield, together with a left moving fermion

1325: $\zeta_-$. The scalar field $D$ will be seen to be auxiliary. The

1326: covariant superderivatives are given by

1327: %

1328: \be {\cal D}_+&=&\frac{\partial}{\partial

1329: \theta^+}-i\bar{\theta}^+({\cal D}_0+{\cal D}_1)

1330: %\ \ \ \ {\rm and}\

1331: \ \ \ \nn\\ \bar{{\cal D}}_+&=&-\frac{\partial}{\partial

1332: \bar{\theta}^+}+i{\theta}^+({\cal D}_0+{\cal D}_1)\nn\ee

1333: %

1334: where

1335: %

1336: \be {\cal D}_0+{\cal D}_1=\partial_0+\partial_1-i(u_0+u_1) \nn\ee

1337: %

1338: includes the $u_+$ component of the gauge field, but no fermions.

1339: Meanwhile, the gauginos are included in the remaining covariant

1340: superderivative,

1341: %

1342: \be {\cal D}_0-{\cal D}_1=\partial_0-\partial_1-iU .\ee

1343: %

1344: The field strength lives naturally in a fermi multiplet (which we

1345: shall define shortly) given by the usual commutator of

1346: derivatives:

1347: %

1348: \be \Upsilon=[\bar{\cal D}_+,{\cal D}_0-{\cal

1349: D}_1]=-2\Big(\zeta_--i\theta^+(D-iv_{01})-i\theta^+\bar{\theta}^+({\cal

1350: D}_0+{\cal D}_1)\zeta_-\Big) .\ee

1351: %

1352: Here the field strength is

1353: $u_{01}=\partial_0u_1-\partial_1u_0-i[u_0,u_1]$. The kinetic terms

1354: for the gauge multiplet are then given by integration over all of

1355: superspace $d^2\theta =d\theta^+\,d\bar{\theta}^+$,

1356: %

1357: \be S_{\rm gauge}&=&\frac{1}{8g^2}\,\Tr\,\int d^2y \,d^2\theta\ \Upsilon^\dagger \Upsilon \nn\\

1358: &=&\frac{1}{g^2}\,\Tr\,\int d^2y\ \Big(\ft12\

1359: u_{01}^2+i\bar{\zeta}_-({\cal D}_0+{\cal D}_1)\zeta_-+D^2\Big)

1360: \ee

1361: %

1362: but, in fact, will not be required in the following.

1363: 

1364: \subsubsection*{Chiral Multiplets}

1365: 

1366: The chiral multiplets of $(0,2)$ theories are bosonic superfields

1367: $\Phi$, living in any representation $R$ of the gauge group. They

1368: satisfy

1369: %

1370: \be \bar{\cal D}_+\Phi=0 .\ee

1371: %

1372: Chiral multiplets contain right-moving fermions $\xi_+$, paired

1373: with a complex boson $\phi$. Their component expansion gives

1374: %

1375: \be

1376: \Phi=\phi+\sqrt{2}\theta^+\xi_+-i\theta^+\bar{\theta}^+(D_0+D_1)\phi

1377: \label{02chi}\ee

1378: %

1379: where $(D_0+D_1)$ is now the usual bosonic covariant derivative.

1380: The kinetic terms for the chiral multiplet are given by the

1381: action,

1382: %

1383: \be S_{\rm chiral}&=&-\frac{i}{2}\int d^2y\,d^2\theta\

1384: \bar{\Phi}({\cal D}_0-{\cal D}_1)\Phi \label{chirallag}\\ &=& \int

1385: d^2y\ \left(-|D_\alpha\phi|^2+i\bar{\xi}_+(D_0-D_1)\xi_+ -

1386: i\sqrt{2}\bar{\phi}\zeta_-\xi_+ +

1387: i\sqrt{2}\bar{\xi}_+\bar{\zeta}_-\phi + \bar{\phi}D\phi\right) .

1388: \nn\ee

1389: %

1390: The scalar field $\phi$ couples to the auxiliary field $D$, to

1391: give rise to the usual D-term (Note that for Abelian theories, if

1392: $\Phi$ has charge $p$ then one should replace $\zeta_-\rightarrow

1393: p\zeta_-$ and $D\rightarrow pD$ in the above action.).

1394: 

1395: \subsubsection*{Fermi Multiplets}

1396: 

1397: One novel feature of $(0,2)$ theories that is not shared by the

1398: non-chiral $(2,2)$ theories is the existence of a fermionic

1399: multiplet $\Gamma$, containing only left moving fermions $\chi_-$

1400: and no propagating bosons. Like the chiral multiplets, they can

1401: live in any representation $R$ of the gauge group. The fermi

1402: multiplet satisfies

1403: %

1404: \be \bar{\cal D}_+\Gamma=\sqrt{2}E\label{e}\ee

1405: %

1406: where $\bar{\cal D}_+E=0$, which can be solved by taking $E$ to be

1407: a holomorphic function of chiral superfields $E=E(\Phi_i)$. The

1408: fermi multiplet has component expansion

1409: %

1410: \be \Gamma = \chi_--\sqrt{2}\theta^+ G

1411: -i\theta^+\bar{\theta}^+(D_0+D_1)\chi_--\sqrt{2}\bar{\theta}^+E .

1412: \label{02fer}\ee

1413: %

1414: Note that the superfield $\Upsilon$ containing the field strength

1415: is of this type, with $\bar{\cal D}_+\Upsilon = 0$. In general,

1416: $E$ itself will also have a $\theta$ expansion,

1417: %

1418: \be E(\Phi_i)=E(\phi_i)+\sqrt{2}\theta^+\frac{\partial

1419: E}{\partial\phi_i}\xi_{+i}

1420: -i\theta^+\bar{\theta}^+(D_0+D_1)E({\phi_i}) \ee

1421: %

1422: The kinetic terms for the fermi multiplet are

1423: %

1424: \be S_{\rm fermi}&=&-\frac{1}{2}\int d^2y\,d^2\theta\

1425: \bar{\Gamma}\Gamma \label{fermilag}\\ &=&

1426: \left(i\bar{\chi}_-(D_0+D_1)\chi_-+|G|^2-|E(\phi_i)|^2

1427: -\bar{\chi}_-\frac{\partial E}{\partial\phi^i}\xi_{+i}

1428: +\bar{\xi}_{+i}\frac{\partial\bar{E}}{\partial\bar{\phi}_i}\chi_-\right)

1429: \nn\ee

1430: %

1431: We see that the complex scalar $G$ is an auxiliary field, lacking

1432: a kinetic term. Also note that the function $E(\phi)$ appears as a

1433: potential term in the Lagrangian.

1434: 

1435: \subsection{Superpotentials}

1436: 

1437: In ${\cal N}=(0,2)$ theories the auxiliary field $G$ lives in a fermi

1438: multiplet $\Gamma$, rather than a chiral multiplet. A

1439: superpotential $J(\Phi_i)$ is a holomorphic function of chiral

1440: superfields and a suitable action may be constructed by

1441: integrating terms of the form $\Gamma J$ over half of superspace.

1442: Most generally we can introduce a superpotential $J^a$ for each

1443: fermi multiplet $\Gamma^a$,

1444: %

1445: \be S_{J} &=& -\frac{1}{\sqrt{2}}\sum_a\int d^2y\,d\theta^+ \ \

1446: \Gamma_a\left.

1447: J^a(\Phi_i)\right|_{\bar{\theta}^+=0}+\ {\rm h.c.} \nn\\

1448: &=& \sum_{a}\int d^2y\ \

1449: G_aJ^a(\phi_i)+\sum_i\chi_{-a}\frac{\partial

1450: J^a}{\partial\phi_i}\xi_{+i}+\ {\rm h.c.}\ .\label{sj}\ee

1451: %

1452: This integration over half of superspace yields an ${\cal

1453: N}=(0,2)$ supersymmetric invariant action if and only if

1454: $\bar{D}_+(\Gamma_aJ^a)=0$, which requires

1455: %

1456: \be \sum_a E_aJ^a=0 \label{ej} .\ee

1457: %

1458: Of course, the combination $\Gamma_aJ^a$ is also required to be

1459: gauge invariant. An important example of the superpotential is the

1460: Fayet-Iliopoulos and theta term which are packaged in the complex

1461: combination $t=ir+\theta/2\pi$. The interaction can be written as

1462: %

1463: \be S_{D\theta}&=&\frac{t}{4}\,\Tr\int d^2y d\theta^+\

1464: \left.\Upsilon\right|_{\bar{\theta}^+=0} +{\rm \ h.c.}\nn\\ &=&

1465: \Tr\int d^2y\ (-rD+\frac{\theta}{2\pi}u_{01}) .\label{fipot}\ee

1466: %

1467: 

1468: \subsection{${\cal N}=(2,2)$ Decomposition}

1469: 

1470: It will prove useful for orientation to recall how the more

1471: familiar ${\cal N}=(2,2)$ superfields decompose into their ${\cal

1472: N}=(0,2)$ counterparts. The conventions below are taken from

1473: \cite{phases}.

1474: 

1475: \para

1476: One can enlarge ${\cal N} =(0,2)$ superspace to ${\cal N}=(2,2)$ superspace through

1477: the addition of two further fermionic components $\theta^-$ and

1478: $\bar{\theta}^-$. The corresponding superderivatives are

1479: %

1480: \be

1481: D_-=\frac{\partial}{\partial\theta^-}-i\bar{\theta}^-(\partial_0-\partial_1)\

1482: \ \ \ ,\ \ \ \

1483: \bar{D}_-=-\frac{\partial}{\partial\bar{\theta}^-}+i\theta^-(\partial_0-\partial_1).\

1484: \ee

1485: %

1486: The ${\cal N}=(2,2)$ vector multiplet $V_{(2,2)}$ decomposes into

1487: an ${\cal N}=(0,2)$ vector multiplet $V$ described in \eqn{02vec},

1488: together with an ${\cal N}=(0,2)$ chiral multiplet $\Sigma$. This

1489: chiral multiplet inherits the right moving fermion $\zeta_+$ and

1490: the complex scalar field $\sigma$ contained in $V_{(2,2)}$. It is

1491: most simply described by reduction from the ${\cal N}=(2,2)$

1492: twisted chiral multiplet containing the field strength

1493: $\Sigma_{(2,2)}=(1/\sqrt{2})\{\bar{\cal D}_+,{\cal D}_-\}$, in

1494: terms of which the ${\cal N}=(0,2)$ chiral multiplet is given by

1495: %

1496: \be \Sigma=\left.\Sigma_{(2,2)}\right|_{\theta^-=\bar{\theta}_-=0}.

1497: \ee

1498: %

1499: An ${\cal N}=(2,2)$ chiral multiplet $\Phi_{(2,2)}$ satisfies

1500: $\bar{D}_+\Phi_{(2,2)}=\bar{D}_-\Phi_{(2,2)}=0$. This chiral

1501: multiplet decomposes into an ${\cal N}=(0,2)$ chiral multiplet

1502: $\Phi$ and a fermi multiplet $\Gamma$, defined by

1503: %

1504: \be \Phi &=& \left.\Phi_{(2,2)}\right|_{\theta^-=\bar{\theta}^-=0}

1505: \nn\\

1506: \Gamma&=&\frac{1}{\sqrt{2}}\,{\cal

1507: D}_-\left.\Phi_{(2,2)}\right|_{\theta^-=\bar{\theta}^-=0}.\ee

1508: %

1509: If $\Phi_{(2,2)}$ transforms under a representation $R$ of the

1510: gauge group, then both $\Phi$ and $\Gamma$ also transform under

1511: $R$. A quick computation yields $\bar{\cal D}_+\Gamma=2i \Sigma

1512: \Phi$, meaning that, in the notation of \eqn{e}, ${\cal N}=(2,2)$

1513: supersymmetry imposes,

1514: %

1515: \be E=i\sqrt{2}\Sigma \Phi .\ee

1516: %

1517: The final ${\cal N}=(2,2)$ multiplet of interest is a twisted

1518: chiral multiplet $\Sigma_{(2,2)}$, satisfying

1519: $\bar{D}_+\Sigma_{(2,2)}=D_-\Sigma_{(2,2)}=0$. Like the ${\cal

1520: N}=(2,2)$ chiral multiplet, this too decomposes into an ${\cal

1521: N}=(0,2)$ chiral multiplet $\Sigma$ and a fermi multiplet $F$.

1522: They are given by,

1523: %

1524: \be \Sigma&=&\left.\Sigma_{(2,2)}\right|_{\theta^-=\bar{\theta}^-=0}\nn\\

1525: F&=&

1526: -\frac{1}{\sqrt{2}}\bar{D}_-\left.\Sigma_{(2,2)}\right|_{\theta^-=\bar{\theta}^-=0} .

1527: \ee

1528: %

1529: Note, however, that from the expansion \eqn{twistedc}, the

1530: $\theta^+$ component of the ${\cal N}=(0,2)$ chiral multiplet

1531: $\Sigma$ contains the barred fermion, rather than the unbarred

1532: fermion,

1533: %

1534: \be \Sigma = \sigma - i\sqrt{2}\theta^+\bar{\zeta}_+

1535: -i2\theta^+\bar{\theta}^+\partial_+\sigma .\label{02tc}\ee

1536: %

1537: This subtlety will prove important in what follows. Since twisted

1538: chiral multiplets $\Sigma_{(2,2)}$ are always uncharged under the

1539: gauge group, the corresponding fermi multiplet satisfies

1540: $\bar{\cal D}_+F=0$.

1541: 

1542: \subsubsection{The Vortex Theory in ${\cal N}=(0,2)$ Language}

1543: 

1544: Let us finish this section by describing the ${\cal N}=(2,2)$

1545: vortex theory of Section 2 in the language of ${\cal N}=(0,2)$

1546: superfields. This will serve to fix notation for what is to come.

1547: We decompose the fields as

1548: %

1549: \be \mbox{${\cal N}=(2,2)$ $U(k)$ Vector Multiplet}

1550: &\longrightarrow& \mbox{$U(k)$ Vector Multiplet, $U$} \nn\\

1551: &&+\ \mbox{Adjoint Chiral Multiplet

1552: $\Sigma$}\nn\\

1553: \mbox{${\cal N}=(2,2)$ Adjoint Chiral Multiplet} &\longrightarrow&

1554: \mbox{Adjoint Chiral Multiplet, $Z$} \nn\\ &&+\

1555: \mbox{Adjoint Fermi Multiplet $\Xi$} \nn\\

1556: \mbox{${\cal N}=(2,2)$ Fund. Chiral Multiplets} &\longrightarrow&

1557: \mbox{Fund. Chiral Multiplets, $\Phi_i$} \nn\\

1558: &&+\ \mbox{Fund. Fermi Multiplets $\Gamma_i$}\nn\\

1559: \mbox{${\cal N}=(2,2)$ Anti-Fund. Chiral Multiplet}

1560: &\longrightarrow& \mbox{Anti-Fund. Chiral Multiplets, $\tilde{\Phi}_j$} \nn\\

1561: &&+\ \mbox{Anti-Fund. Fermi Multiplet $\tilde{\Gamma}_j$}\nn\ee

1562: %

1563: where all the objects on the right are ${\cal N}=(0,2)$

1564: superfields. As before, $i=1,\ldots, N_c$ for  $\Phi_i$ and

1565: $j=1,\ldots N_f-N_c$ for $\tilde{\Phi}_j$. Appendix A contains a

1566: list of the different component fields which appear in each of

1567: these multiplets.

1568: 

1569: \para

1570: The ${\cal N}=(2,2)$ supersymmetry imposes the relations,

1571: %

1572: \be \bar{\cal D}_+\Xi=2i[\Sigma,Z] \ \ \ ,\ \ \ \bar{\cal

1573: D}_+\Gamma_i=2i(\Sigma-m_i)\Phi_i\ \ \ ,\ \ \ \bar{\cal

1574: D}_+\tilde{\Gamma}^j=-2i(\Sigma-\tilde{m}_j)\tilde{\Phi}_j\ \ \ \

1575: \ \label{22rels}\ee

1576: %

1577: (There is no sum over $i$ and $j$ on the right-hand side of these

1578: equations). As we have seen, the right-hand side of each of these

1579: equations appears as a potential ``$|E|^2$" arising in equation

1580: \eqn{fermilag}. A further contribution to the worldsheet scalar

1581: potential arises from the D-term, which provides the constraint

1582: \eqn{vortexd}.

1583: 

1584: 

1585: 

1586: \section{The ${\cal N}=(0,2)$ Dynamics of Vortex Strings}

1587: 

1588: It is now time to present new results for the dynamics of vortex

1589: strings in theories with ${\cal N}=1$ supersymmetry. Most of this

1590: section is devoted to the  discussion of a simple deformation of

1591: the ${\cal N}=2$ theory by the addition of a superpotential. In

1592: Section \ref{more} we discuss a second class of  deformations.

1593: 

1594: 

1595: 

1596: \subsection{Adding a Superpotential}

1597: 

1598: We start by considering a ``Dijkgraaf-Vafa"-like deformation

1599: \cite{dv}, breaking ${\cal N}=2$ to ${\cal N}=1$ through the

1600: addition of a superpotential for the adjoint superfield $A$. The

1601: superpotential now reads

1602: %

1603: \be {\cal W}=\sqrt{2}

1604: \sum_{i=1}^{N_f}\tilde{Q}_i(A-m_i)Q_i+\hat{\cal

1605: W}(A)\label{def}\ee

1606: %

1607: which gives rise to the scalar potential

1608: %

1609: %

1610: \be V_{4d}&=& \frac{e^2}{2}\Tr(\,\sum_{i=1}^{N_f}\,Q_iQ_i^\dagger

1611: - \tilde{Q}_i\tilde{Q}_i^\dagger - v^2{\bf 1}_{N_c})^2 +e^2\Tr|\,

1612: \sum_{i=1}^{N_f}\tilde{Q}_iQ_i-\partial\hat{W}/\partial A|^2\label{v4dv}\\

1613: &&+\sum_{i=1}^{N_f}\left(Q_i^\dagger

1614: \{A-m_i,\bar{A}-\bar{m}_i\}Q_i +

1615: \tilde{Q}_i\{A-m_i,\bar{A}-m_i\}\tilde{Q}_i^\dagger\right) +

1616: \frac{1}{2e^2}\Tr|[A,A^\dagger]|^2 .\nn\ee

1617: %

1618: Let's look at how this superpotential affects the vacuum

1619: structure. If ${\hat{\cal W}}$ is linear in $A$ then there is

1620: merely  a constant piece in the F-term above and the Lagrangian

1621: still preserves ${\cal N}=2$ supersymmetry. We can perform an

1622: $SU(2)_R$ rotation of the scalar fields

1623: $(Q_i,\tilde{Q}_i^\dagger)$ to bring the Lagrangian back to the

1624: form \eqn{v4dn2}. We will assume that $\hW$ does not contain a

1625: linear piece. In this case, for a generic superpotential $\hW(A)$,

1626: $\tilde{Q}_i$ must turn on in the vacuum. Without loss of

1627: generality, we choose the vacuum to be of the form,

1628: %

1629: \be Q^a_{\ i} = p_i\,\delta^a_{\ i} \ \ \ , \ \ \ \tilde{Q}^a_{\

1630: i}=\tilde{p}_i\,\delta^a_{\ i}\ \ \ ,\ \ \ A={\rm

1631: diag}(m_1,\ldots,m_{N_c})\label{wbac}\ee

1632: %

1633: with

1634: %

1635: \be  |p_i|^2-|\tilde{p}_i|^2=v^2 \ \ \ {\rm and} \ \ \

1636: \tilde{p}_ip_i = \left.\frac{\partial\hW}{\partial a}\right|_{m_i}

1637: \ \ \ \ \ \mbox{for each $i=1,\ldots,N_c$} \label{whatsleft}\ee

1638: %

1639: 

1640: \subsection{What Becomes of the Vortex?}

1641: 

1642: Our goal is to understand how this deformation affects the

1643: dynamics of the vortex string\footnote{Vortices in a  similar

1644: system were studied in \cite{bolog}, but in the limit with

1645: $v^2=0$, so that the vortex is built around a linear piece of

1646: $\hW$. This gives rise to somewhat different physics from that

1647: considered here.}. Let us firstly consider the case with distinct

1648: masses $m_i$. Before adding the superpotential $\hW$ there were

1649: $N_c$ different BPS vortices, each living in a different

1650: $U(1)\subset U(N_c)$ and each with a different $Q_i$,

1651: $i=1,\ldots,N_c$ carrying the asymptotic winding. What changes in

1652: the presence of $\hW$?

1653: 

1654: \para

1655: The crucial point to note is that something rather special happens

1656: when the superpotential is tuned so that a critical point

1657: coincides with one of the masses, say $m_k$ for some $k=1,\ldots,

1658: N_c$

1659: %

1660: \be \left.\frac{\partial\hW(a)}{\partial

1661: a}\right|_{a=m_k}=0 .\label{survive}\ee

1662: %

1663: If this is case, the vacuum equation \eqn{wbac} sets

1664: $\tilde{Q}_k=0$. There is then no obstacle in constructing the

1665: $k^{\rm th}$ vortex in which $Q_k$ winds; indeed the ${\cal N}=2$

1666: vortex solution remains a solution in the deformed theory.

1667: 

1668: \para

1669: Vortices of this type in ${\cal N}=1$ theories are often called

1670: D-term vortices (the name arises because the symmetry breaking is

1671: induced by a FI parameter, or D-term). It was shown in \cite{ddt}

1672: that such objects are 1/2 BPS, preserving two of the four

1673: supercharges of the four-dimensional ${\cal N}=1$ theory. In two

1674: dimensions, there are two distinct superalgebras with two

1675: supercharges: the non-chiral $(1,1)$ algebra, and the chiral

1676: $(0,2)$ algebra. Given that the previous section was devoted to a

1677: review of ${\cal N}=(0,2)$ theories, the reader may guess this

1678: will be relevant for the vortex string. Let's now see that this is

1679: indeed the case \cite{ddt}. The ${\cal N}=1$ supersymmetry

1680: transformations for the vector multiplet fields are,

1681: %

1682: \be \delta A_\mu &=& -i\bar{\epsilon}\sigma_\mu\lambda +

1683: i\bar{\lambda}\sigma_\mu \epsilon \nn\\ \delta

1684: D&=&\bar{\epsilon}\bar{\sigma}^\mu {\cal D}_\mu \lambda + {\cal

1685: D}_\mu \lambda \bar{\sigma}^\mu\epsilon \nn\\ \delta \lambda &=&

1686: \ft12 \sigma^{\mu\nu}\epsilon F_{\mu\nu} + i\epsilon D .\ee

1687: %

1688: For each chiral multiplet $Q_i$, they take the form

1689: %

1690: \be \delta Q_i &=& \sqrt{2}\epsilon \psi_i\nn\\ \delta F_i&=&

1691: i\sqrt{2}\bar{\epsilon}\bar{\Dslash}\psi_i -

1692: 2i\bar{\epsilon}\lambda Q_i\nn\\ \delta \psi_i &=&

1693: \sqrt{2}\epsilon F_i + i\sqrt{2}(\!\Dslash Q_i) \bar{\epsilon} .\ee

1694: %

1695: Similar transformations also hold for the chiral multiplets

1696: $\tilde{Q}_i$ with the appropriate substitutions. Finally, the

1697: supersymmetry transformations for the adjoint chiral multiplet $A$

1698: take the form,

1699: %

1700: \be \delta A &=& \sqrt{2}\epsilon \eta\nn\\ \delta F &=&

1701: i\sqrt{2}\bar{\epsilon}\bar{\Dslash}\eta -

1702: 2i\bar{\epsilon}[\lambda, A]\nn\\ \delta \eta &=& \sqrt{2}\epsilon

1703: F + i\sqrt{2}(\!\Dslash A) \bar{\epsilon} .\ee

1704: %

1705: The key point here is that the vortex equations \eqn{vort},

1706: together with the requirement that $F_i=F=0$, provide solutions to

1707: $\delta\lambda = \delta \psi_i = \delta\eta = 0$. The latter

1708: condition $F=0$ is trivially satisfied when \eqn{survive} holds,

1709: for then $\tilde{Q}_i=0$, while $A$ remains constant. To see which

1710: supersymmetries are preserved in this case, it will suffice to

1711: examine the $\delta \psi_i$ transformation. Using \eqn{ds}, in the

1712: background of a stationary vortex so that ${\cal D}_+={\cal

1713: D}_-=0$, we have

1714: %

1715: \be \delta\psi_{-i} = -2\sqrt{2}i ({\cal D}_z Q_i)

1716: \bar{\epsilon}_- =0 \ \ \ {\rm and}\ \ \ \delta\psi_{+i} =

1717: 2\sqrt{2}i ({\cal D}_{\bar{z}}Q_i)\bar{\epsilon}_+ .\ee

1718: %

1719: In the background of a vortex, with the scalar field satisfying

1720: ${\cal D}_zQ_i=0$, we learn that $\bar{\epsilon}_-$ is the

1721: preserved supersymmetry; it descends to provide the supersymmetry

1722: variation parameter on the worldsheet. Meanwhile,

1723: $\bar{\epsilon}_+$ is the broken supersymmetry which generates a

1724: single  Goldstino mode on the worldsheet. In our notation

1725: \eqn{goldy}, we have $\bar{\epsilon}_+=\chi_+/4$. (The $\chi_-$

1726: collective coordinate in \eqn{goldy} arises from the second

1727: supersymmetry transformation of the ${\cal N}=2$ theory. Its fate

1728: in our ${\cal N}=1$ theory will be discussed shortly). The spinors

1729: $\epsilon_\pm$ have definite, and opposite, chirality on the

1730: worldsheet. This is the statement that the worldsheet theory

1731: preserves chiral ${\cal N}=(0,2)$ supersymmetry, rather than

1732: ${\cal N}=(1,1)$.

1733: 

1734: 

1735: 

1736: \para

1737: We have seen that, in the special case that a critical point of

1738: $\hW$ coincides with a mass \eqn{survive}, there exists at least

1739: one BPS vortex preserving ${\cal N}=(0,2)$ supersymmetry. But what

1740: happens if this is not the case? If \eqn{survive} is not

1741: satisfied, then there can be no BPS vortex solutions. To see this,

1742: note that \eqn{whatsleft} tells us that $\tilde{Q}_k$  gains

1743: an expectation value in the 4d vacuum. This means it cannot now

1744: remain constant but, must wind asymptotically to ensure that its

1745: kinetic term remains finite. A putative BPS vortex must now

1746: satisfy,

1747: %

1748: \be {\cal D}_zQ_i={\cal D}_z \tilde{Q}_i=0 .\ee

1749: %

1750: Yet $Q_i$ and $\tilde{Q}_i$ have opposite charges.  A standard

1751: theorem in mathematics

1752: --- that a line bundle of negative degree has no non-zero

1753: holomorphic section

1754: --- states that there can only be simultaneous solutions to these

1755: equations when either $\tilde{Q}_i=0$ or $Q_i=0$. (See, for

1756: example, equation (3.43) of \cite{phases}). One can reach the same

1757: conclusion by noting that $A$ is now also sourced in the vortex

1758: background and $\delta\eta\neq 0$\footnote{The lack of BPS

1759: vortices in this case is entirely analogous to the statement that

1760: $F$-term vortices are not BPS in ${\cal N}=1$ theories \cite{ddt}.

1761: In our set-up, the value of $\partial \hW/\partial a$ evaluated at

1762: $a=m_k$ plays the role of the constant in the F-term in

1763: \cite{ddt}.}. Of course, simple topological  arguments imply that

1764: vortex strings still exist. However, they must satisfy the full

1765: second order equations of motion, rather than the first order

1766: Bogomolnyi equations, and their tension is strictly greater than

1767: the BPS bound $T=2\pi v^2$.

1768: 

1769: 

1770: \subsection{Vortex Dynamics}

1771: 

1772: In section 2, we described the ${\cal N}=(2,2)$ $U(k)$ theory on

1773: the vortex worldsheet that captures the dynamics of $k$ parallel

1774: vortex strings in  ${\cal N}=2$ four dimensional gauge theories.

1775: We would like to understand how the worldsheet theory reacts to

1776: the superpotential $\hW(A)$, breaking the four dimensional

1777: supersymmetry from ${\cal N}=2$ to ${\cal N}=1$. We have seen

1778: above that the vortices in the theory with superpotential $\hW(A)$

1779: are classically BPS, preserving ${\cal N}=(0,2)$ supersymmetry,

1780: when equation \eqn{survive} holds; otherwise there are no BPS

1781: vortices. We would like to see this from the worldsheet.

1782: 

1783: \para

1784: In fact, there is a unique deformation on the vortex worldsheet

1785: that preserves ${\cal N}=(0,2)$ supersymmetry and reproduces the

1786: expected vacuum structure described above. Recall from Section 3.2

1787: that superpotentials in ${\cal N}=(0,2)$ theories are constructed

1788: from fermi multiplets. The only such multiplet with a suitable

1789: transformation under the $U(k)$ gauge symmetry is $\Xi$,

1790: containing $\chi_-$ and the complex auxiliary field $G_Z$. The

1791: worldsheet deformation is given by the ${\cal N}=(0,2)$

1792: superpotential,

1793: %

1794: \be S_{{\cal W}} &\equiv& -\frac{1}{\sqrt{2}} \Tr_k\left.\int

1795: d\theta^+\ \Xi\,J(\Sigma)\right|_{\bar{\theta}^+=0} - {\rm h.c.}

1796: \nn\\ &=& -\frac{1}{\sqrt{2}}\Tr_k\,\left.\int d\theta^+\

1797: \Xi\,\frac{\partial\hW(\Sigma)}{\partial\Sigma}\right|_{\bar{\theta}^+=0}-

1798: {\rm h.c.}\label{booty}\ee

1799: %

1800: (up to some overall, unfixed, constant of proportionality). Note

1801: that a superpotential of this form is a viable holomorphic term

1802: since $\bar{\cal D}_+\Xi=2i[\Sigma, Z]\equiv i\sqrt{2}E_\Xi$ and

1803: %

1804: \be \Tr\ E_\Xi J=\sqrt{2}\Tr\ \left([\Sigma,

1805: Z]\,\frac{\partial\hW(\Sigma)}{\partial\Sigma}\right)=0\ee

1806: %

1807: which satisfies the requirement \eqn{ej}. In principle there could

1808: also be $\sigma$-dependent deformations of the kinetic terms for

1809: $\Lambda_i$ and $\Xi$. As is common in supersymmetric field

1810: theories, we will have less control over these ``D-term"

1811: deformations, but will see that the superpotential \eqn{booty}

1812: captures much of the important physics.

1813: 

1814: \para

1815: The deformation \eqn{booty} has implications for both the bosonic

1816: and fermionic zero modes of the vortex strings. We defer a

1817: discussion of the fermions to the next subsection; we start here

1818: by studying the bosonic zero modes. The extra bosonic term on the

1819: vortex worldsheet arising from \eqn{booty} is a

1820: potential\footnote{A note on dimensions: In 4d, $[\hW(A)]=3$,

1821: which ensures that the scalar potential has the correct

1822: dimensions: $[|\partial \hW/\partial A|^2]=4$. In 2d the auxiliary

1823: field has dimension $[\sigma]=1$, so that $[\partial

1824: \hW/\partial\sigma]=2$. The presence of the vortex tension, with

1825: $[T]=2$, means that the worldsheet scalar potential \eqn{kimmoy}

1826: has the correct scaling for the two dimensional worldsheet.}, %

1827: %

1828: \be V_{2d}= \Tr_k\left(T|G_Z|^2 +

1829: G_Z\frac{\partial\hW(\sigma)}{\partial\sigma} + {\rm h.c.}\right)

1830: =\frac{1}{T}\Tr_k\left|\frac{\partial\hW(\sigma)}{\partial\sigma}\right|^2 .

1831: \label{kimmoy}\ee

1832: %

1833: We will now show that this gives the expected vacuum structure by

1834: studying the $k=1$ vortex theory in some detail; the extension to

1835: $k>1$ then follows.

1836: 

1837: 

1838: \subsubsection{An Example: $k=1$ with $N_f=N_c$}

1839: 

1840: To illustrate the role of the superpotential \eqn{booty}, let's

1841: look at the familiar $k=1$ theory of a single vortex in the case

1842: with $N_f=N_c$ flavors. As we discussed in detail in Section 2,

1843: when $\hW=0$ the internal moduli space is ${\bf CP}^{N_c-1}$ with

1844: $\phi_i$ providing homogeneous coordinates. Once we turn on the

1845: superpotential $\hW$, the bosonic part of the worldsheet theory is

1846: given by

1847: %

1848: \be {\cal L}_{\rm bose}= T|\partial_mz|^2 + \sum_{i=1}^{N_c}\left( |{\cal

1849: D}_m\phi_i|^2 - 2|\sigma-m_i|^2|\phi_i|^2\right) +

1850: D(\sum_{i=1}^{N_c}|\phi_i|^2-r) - \frac{1}{T}\left|\frac{\partial

1851: \hW}{\partial \sigma}\right|^2 + \frac{\theta}{2\pi}u_{01} .\nn\ee

1852: %

1853: %

1854: In the presence of distinct, non-zero masses $m_i$, this

1855: worldsheet theory has a supersymmetric ground state (i.e. with

1856: vanishing vacuum energy) at

1857: %

1858: \be |\phi_j|^2=r\delta_{ij}\ \ \ ,\ \ \ \sigma=m_i\ee

1859: %

1860: only if $\hW(\sigma)$ has a critical point at $\sigma=m_i$

1861: %

1862: \be \left.\frac{\partial \hW(\sigma)}{\partial

1863: \sigma}\right|_{\sigma=m_i}=0 .\label{again}\ee

1864: %

1865: This coincides with the expectations of the previous section: BPS

1866: vortices only exist when \eqn{again} holds.

1867: 

1868: \para

1869: When the masses do not coincide with the critical points, and

1870: there are no BPS vortices, the potential $|\partial\hW/\partial

1871: \sigma |^2/T$ determines the vacuum energy of the vortex string.

1872: One could try to compare this to the excess tension of the non-BPS

1873: vortex string, above the bound $T=2\pi v^2$, but this

1874: unfortunately suffers from the previously mentioned ambiguity in

1875: classical wavefunction renormalization for $\chi_-$ which also

1876: affects the coefficient in front of $|G_Z|^2$.

1877: 

1878: 

1879: \para

1880: If the hypermultiplet masses vanish, $m_i=0$, then the story is a

1881: little different. We may now set $\sigma=0$ in the vacuum (recall

1882: that we assumed $\hW$ does not contain a linear piece, so

1883: $\sigma=0$ is guaranteed to be a critical point). The full ${\bf

1884: CP}^{N_c-1}$ bosonic moduli space is now restored. This is in

1885: agreement with expectations from four dimensions, where we may

1886: happily construct any vortex string, built around the vacuum with

1887: $A=\tilde{Q}_i\equiv0$. We conclude that the superpotential

1888: $\hW(A)$ does not affect the bosonic zero modes in this case, a

1889: point made previously in \cite{syss}. However, the superpotential

1890: does still affect the fermi zero modes. We now turn to a study of

1891:  these.

1892: 

1893: \subsection{Fermions}

1894: 

1895: We will study the fermions in the case with vanishing

1896: hypermultiplet masses $m_i=0$. Of all the Dirac equations in

1897: \eqn{diracs}, only that for $\eta$ is modified by the

1898: superpotential. It now reads

1899: %

1900: \be -\frac{i}{e^2}\Dbarslash\eta-\frac{i\sqrt{2}}{e^2}

1901: [A,\bar{\lambda}]-\sqrt{2}\tilde{Q}^\dagger_i\bar{\psi}_i-

1902: \sqrt{2}\bar{\tilde{\psi}}_i{Q}_i^{\dagger} -

1903: \frac{\partial^2\hW(A)}{\partial A^2}\bar{\eta}=0 .\ee

1904: %

1905: In the background of the vortex, we may again set

1906: $A=\tilde{Q}_i=0$. This means that all right-moving fermi zero

1907: modes

1908: --- those donated by $\lambda_+$ and $\bar{\psi}_{+i}$ ---

1909: remain the same as in the ${\cal N}=(2,2)$ case, given by

1910: solutions to

1911: %

1912: \be \sqrt{2}{\cal D}_z\lambda_+ &=& -e^2\,Q_i\bar{\psi}_{+i}\ , \nn\\

1913: \sqrt{2}{\cal D}_{\bar{z}} \bar{\psi}_{+i}&=&-

1914: Q_i^\dagger\lambda_+\label{same} .\ee

1915: %

1916: If the lowest order term in the superpotential $\hW$ is cubic or

1917: higher, then the left-moving fermi zero modes are similarly

1918: unaffected. However, if the superpotential $\hW(A)$ includes a

1919: quadratic mass term

1920: %

1921: \be \hW(A)=\mu_2 A^2 + \ldots \ee

1922: %

1923: then the equations for the left moving fermi zero modes become

1924: %

1925: \be \sqrt{2}i{\cal D}_{\bar{z}} \eta_- &=& -e^2

1926: \bar{\tilde{\psi}}_{-i} Q_i^\dagger - \sqrt{2}\mu_2\bar{\eta}_-\nn\\

1927: \sqrt{2}i{\cal D}_z \bar{\tilde{\psi}}_{-i} &=&

1928: \eta_-Q_i .\label{newbo}\ee

1929: %

1930: These equations are no longer related to the bosonic zero mode

1931: equations \eqn{bogzero}: this is to be expected since, in breaking

1932: to ${\cal N}=1$ supersymmetry, we have lost the half of

1933: supersymmetry which ensured the correspondence between bosonic

1934: zero modes and left-moving fermionic zero modes. Nevertheless, as

1935: stressed in \cite{syss}, the Dirac equations \eqn{newbo} must

1936: still admit the same number of solutions as the equations with

1937: $\mu_2=0$. This follows from the fact that the zero modes are

1938: chiral on the worldsheet, and cannot gain a mass through a

1939: deformation. For a single $k=1$ vortex in the $U(2)$ gauge theory,

1940: \eqn{newbo} was analyzed in \cite{syss}, both perturbatively in

1941: $\mu_2\rho$, as well as in the large $\mu_2$ limit.

1942: 

1943: \para

1944: To summarize, we learn that the deformation leaves the fermi zero

1945: modes untouched unless $\mu_2\neq 0$, in which case it deforms the

1946: profile of the left-moving fermi zero modes only. However, the

1947: number of zero modes on the worldsheet remains the same. Let us

1948: now compare this with the predictions from the proposed worldsheet

1949: deformation \eqn{booty}.

1950: 

1951: \subsubsection*{Implications for Worldsheet Fermions}

1952: 

1953: In the presence of the superpotential $\hW$, the fermionic terms

1954: in the $U(k)$ worldsheet theory read\footnote{As we mentioned

1955: previously, the deformation from ${\cal N}=2$ to ${\cal N}=1$ may

1956: also induce a finite wavefunction renormalization of the

1957: left-moving fermion kinetic terms. We will not consider this

1958: here.}

1959: %

1960: \be {\cal L}_{\rm fermi}&=&2iT\ \Tr_k\left(\bar{\chi}_-{\cal

1961: D}_+\chi_-+\bar{\chi}_+{\cal D}_-\chi_+\right) +

1962: 2i\sum_{i=1}^{N_c}\left(\bar{\xi}_{-i}{\cal

1963: D}_+\xi_{-i}+\bar{\xi}_{+i}{\cal D}_-\xi_{+i}\right) \nn\\

1964: &&

1965: -\sqrt{2}\,\Tr_k\left([\bar{\chi}_-,[\sigma,\chi_+]]-[\bar{\chi}_+,[\bar{\zeta}_-,

1966: z]]-[\bar{\xi}_-,[\bar{\zeta}_+,z]]\right) + {\rm h.c.} \label{phew}\\

1967: && -\sqrt{2}\sum_{i=1}^{N_c}\left( \bar{\xi}_{-i}\sigma\xi_{+i}

1968: -\bar{\xi}_{+i}\bar{\zeta}_-\phi_i +

1969: \bar{\xi}_{-i}\bar{\zeta}_+\phi_i\right) +

1970: \Tr_k\left(\chi_-\frac{\partial^2 \hW(\sigma)}{\partial

1971: \sigma^2}\bar{\zeta}_+\right) + {\rm h.c.}\ . \nn\ee

1972: %

1973: The ${\cal N}=(0,2)$ superpotential is responsible for only the

1974: final term. Integrating out the auxiliary fermions $\zeta_\pm$

1975: again gives  constraints on the dynamical fermions,

1976: %

1977: \be \sum_i\phi_i\bar{\xi}_{+i}+[z,\bar{\chi}_+]=0\ \  \ {\rm and}\

1978: \ \ \ \sum_i\phi_i\bar{\xi}_{-i}+[z,\bar{\chi}_-]=

1979: \frac{\partial^2\hW(\sigma)}{\partial \sigma^2}\chi_- .\ee

1980: %

1981: We see that the right-moving fermions are unaffected by the

1982: superpotential, in agreement with the Dirac equations \eqn{same}.

1983: Similarly, if $\hW$ has no quadratic term, so $\mu_2=0$, then the

1984: left-moving constraints are also left unchanged if we set

1985: $\sigma=0$ (we shall see the role played by a non-zero $\sigma$

1986: shortly). However, when $\mu_2\neq 0$, setting $\sigma=0$ still

1987: leaves deformed constraints on the left-moving fermions. For

1988: example, in the case of a single $k=1$ vortex, the constraints

1989: read

1990: %

1991: \be

1992: \sum_{i=1}^{N_c}{\phi}_i\bar{\xi}_{-i}=\mu_2{\chi}_- .\label{02cons}\ee

1993: %

1994: It's worth making a comment on this point. In the ${\cal N}=(0,2)$

1995: theory, we have defined the left-moving worldsheet fermions such

1996: that their kinetic terms are diagonal:

1997: $\bar{\chi}_-\partial_+\chi_-+\bar{\xi}_{-i}{\cal D}_+\xi_{-i}$.

1998: The constraint \eqn{02cons} holds in this basis. It is always

1999: possible to redefine the fermions so that the constraint

2000: \eqn{02cons} reverts to the original ${\cal N}=(2,2)$ constraint

2001: \eqn{fermcons},

2002: %

2003: \be

2004: \bar{\xi}'_{-i}=\bar{\xi}_{-i}-\frac{\mu_2}{r}\bar{\phi}_i\chi_- \

2005: \ \ \Rightarrow \ \ \ \sum_{i=1}^{N_c}\phi_i\,\bar{\xi}'_{-i}=0 .\ee

2006: %

2007: This will then lead to a non-diagonal form for the fermion kinetic

2008: terms.

2009: 

2010: \para

2011: It was argued in \cite{syss} that, even in the presence of the

2012: four-dimensional superpotential $\hW(A)=\mu_2A^2$, the worldsheet

2013: theory of the vortex string still retains ${\cal N}=(2,2)$

2014: supersymmetry. This argument was based on the survival of the

2015: left-moving fermi zero modes, and the lack of a suitable ${\cal

2016: N}=(0,2)$ deformation of the ${\bf CP}^{N_c-1}$ sigma-model. We

2017: disagree with this conclusion. The vortex worldsheet theory is not

2018: described by a ${\bf CP}^{N_c-1}$ sigma-model, but rather by a

2019: ${\bf C}\times {\bf CP}^{N_c-1}$ sigma-model and, as we have seen,

2020: there is a suitable deformation of the latter in which $\chi_-$,

2021: the left-moving fermion in ${\bf C}$, mixes with $\xi_{-i}$.

2022: Moreover, this mixing is necessary to correctly capture the

2023: bosonic properties of the vortex with arbitrary superpotential and

2024: masses. As we explained above, to see this mixing between $\chi_-$

2025: and $\xi_{-i}$ from an explicit analysis of the fermions would

2026: require us to solve the fermi zero mode equations \eqn{newbo}, and

2027: take their overlap to determine both the kinetic terms and the

2028: constraint condition for the Grassmann collective coordinates of

2029: the vortex.

2030: 

2031: \subsection{Symmetries and Other Aspects}

2032: 

2033: We now discuss various further aspects of the worldsheet theory,

2034: starting with an analysis of the symmetries. We will show that the

2035: worldsheet superpotential has the correct properties under

2036: R-symmetry transformations to be induced by the superpotential

2037: $\hW(A)$. The addition of the superpotential $\hW(A)$ breaks both

2038: the $U(1)_R$ and the $U(1)_V$ symmetries in four dimensions. If

2039: the superpotential takes the form,

2040: %

2041: \be \hW(A) = \sum_{n=2} \mu_n A^n .\label{mun}\ee

2042: %

2043: Treating the parameters $\mu_n$ as spurion fields, the symmetry is

2044: restored if $\mu_n$ carries charge $(2-2n, 2)$ under $U(1)_R\times

2045: U(1)_V$. Let us check that these charges descend to the worldsheet

2046: theory. The deformation \eqn{booty} once again destroys both

2047: $U(1)_R$ and $U(1)_V$ on the worldsheet, this time through the

2048: presence of the worldsheet fermi interactions. The final term in

2049: \eqn{phew} is\footnote{The presence of $\bar{\zeta}_+$ in this

2050: expression, rather than $\zeta_+$, is crucial in this analysis. It

2051: follows from the component expansion \eqn{02tc} and ultimately

2052: from the fact $\Sigma$ arises from the decomposition of a $(2,2)$

2053: twisted chiral multiplet as opposed to a $(2,2)$ chiral

2054: multiplet.},

2055: %

2056: \be \sum_n n(n-1)\mu_n \ \Tr_k\,\left(\chi_-\sigma^{n-2}

2057: \,\bar{\zeta_+}\right) .\label{yippee}\ee

2058: %

2059: Examining the table in Section \ref{symmetries}, we see that the

2060: $U(1)_R\times U(1)_V$ worldsheet symmetry is again restored if

2061: $\mu_n$ is assigned charges $(2-2n,2)$, in agreement with the

2062: analysis in four dimensions.

2063: 

2064: 

2065: \para

2066: Note that the $U(1)_Z$ symmetry on the worldsheet, which arises

2067: from rotational invariance in the $z=x^1+ix^2$ plane, is left

2068: unbroken by the deformation \eqn{yippee} as, indeed, it must be.

2069: 

2070: \subsubsection*{Discrete Symmetries}

2071: 

2072: One can also check that the deformation on the worldsheet is consistent

2073: with the discrete symmetries of the bulk theory\footnote{We thank M. Shifman and

2074: A. Yung for stressing the importance of this.}. We start by considering the action

2075: of parity, defined by

2076: %

2077: \be P: x^i \rightarrow -x^i\ \ \ \ i=1,2,3\ee

2078: %

2079: The original ${\cal N}=2$ theory can be written in terms of Dirac

2080: spinors. For example, the adjoint Dirac spinor is

2081: $\Psi = (\lambda,\bar{\eta})^T$.  Parity maps $P: \Psi \rightarrow \gamma^0\Psi$, or

2082: %

2083: \be P: \lambda \leftrightarrow \bar{\eta} \ \ \ {\rm and}\ \ \ \

2084: P: \psi_i \leftrightarrow \bar{\tilde{\psi}}_i\label{fp}\ee

2085: %

2086: while for the complex adjoint scalar $P: A\rightarrow A^\star$. (The

2087: imaginary part is really a pseudoscalar). Because the $z=x^1+ix^2

2088: \rightarrow -z$ part of the parity transformation can be undone by

2089: the rotation $U(1)_Z$ on the worldsheet, we may restrict attention to the simpler

2090: parity transformation $P: x^3\rightarrow -x^3$, with $x^1$ and $x^2$ untouched.

2091: This is the parity action under which the vortex string remains

2092: invariant. It must therefore descend to the worldsheet. Indeed, as we

2093: reviewed in Section 2, $(\lambda,\psi)$ donate right-moving zero modes

2094: $\chi_+$ and $\xi_{+i}$,

2095: while $(\bar{\eta}, \bar{{\tilde{\psi}}})$ donate left-moving zero

2096: modes $\chi_-$ and $\xi_{- i}$. So the action of parity \eqn{fp} in the

2097: bulk also exchanges left and right-movers on the worldsheet.

2098: 

2099: \para

2100: So much for the ${\cal N}=2$ theory. What happens in the presence

2101: of the ${\cal N}=1$ deformation? This pure parity symmetry \eqn{fp} is broken in

2102: the 4d theory because the interactions of $\lambda$ and $\eta$ are

2103: different. This is also seen in our ${\cal N}=(0,2)$ worldsheet

2104: theory where the interactions of left and right movers differ.

2105: 

2106: \para

2107: The 4d ${\cal N}=2$ theory is also invariant under $CP$.

2108: Under charge conjugation, $C: B\rightarrow -B$ and the vortex is

2109: mapped onto the anti-vortex. So this cannot be a symmetry of the

2110: worldsheet. However, under the particular parity transformation

2111: %

2112: \be P': x^2 \rightarrow  -x^2 \label{p1}\ee

2113: %

2114: with $x^1$ and $x^3$ invariant, we also have $B_3\rightarrow

2115: -B_3$. Moreover, the complex coordinate $z$ transverse to the

2116: vortex string is mapped to $P': z\rightarrow z^\star$. This ensures

2117: that the bosonic vortex solution is invariant under $CP'$.

2118: For example, we have

2119: %

2120: \be {\cal D}_z Q_i\ \stackrel{C}{\longrightarrow}\ {\cal

2121: D}_{z}Q^\dagger\ \stackrel{P'}{\longrightarrow}\  {\cal D}_{\bar

2122: z}Q^\dagger \ee

2123: %

2124: so the Bogomolnyi equation ${\cal D}_z Q=0$ remains invariant

2125: under $CP'$. When acting on the fermions, $CP'$ sends Weyl spinors to their

2126: complex conjugates,

2127: %

2128: \be CP': \psi_i\rightarrow -i\sigma_2\bar{\psi}_i\ \ \ , \ \ \ \

2129: CP': \lambda\rightarrow -i\sigma_2\bar{\lambda}\ \ \ ,\ \ \ {\rm etc.} \ee

2130: %

2131: This symmetry also descends to the worldsheet, where it acts as

2132: complex conjugation, as can be checked explicitly from the zero

2133: mode expressions of Section 2. We have,

2134: %

2135: \be CP': \phi_i \rightarrow \bar{\phi}_i \ \ \ {\rm and}\ \ \

2136: CP': \xi_{\pm i} \rightarrow \bar{\xi}_{\pm i}\ \ {\rm etc} \ee

2137: %

2138: Note that, just as $CP'$ in the 4d theory didn't exchange

2139: $\lambda$ and $\bar{\eta}$, so this symmetry on the worldsheet

2140: doesn't send left-movers to right-movers. This can be

2141: traced to the fact that the action $CP'$ under which the string

2142: is invariant doesn't affect $x^3$.

2143: 

2144: \para

2145: Unlike the pure parity transformation, the $CP'$ symmetry survives

2146: the deformation to ${\cal N}=1$ supersymmetry.  More precisely, the

2147: symmetry survives if the parameters in the superpotential ${\cal W}=\mu_n A^n$

2148: are real. Alternatively we can think of these parameters as transforming

2149: under $CP': \mu_n\rightarrow \mu_n^\star$. The same behavior is seen

2150: in the worldsheet theory. Invariance of the final term in \eqn{phew} requires

2151: that $CP':\mu_n\rightarrow \mu_n^\star$, in agreement with the 4d analysis.

2152: 

2153: 

2154: \subsubsection*{The Four-Fermi Term}

2155: 

2156: So far we have neglected the role of $\sigma$ on the string

2157: worldsheet. In the ${\cal N}=(2,2)$ case, we saw that $\sigma$

2158: correctly takes into account the effect of the Yukawa couplings in

2159: four-dimensions, resulting in a four-fermi term \eqn{224fermi} on

2160: the worldsheet. It will play the same role here. The equation of

2161: motion \eqn{yukyuk} for the adjoint field $A$ is now changed by

2162: the superpotential $\hW(A)$. Even if the superpotential has

2163: $\mu_2=0$, so the profiles of both left and right-moving fermionic

2164: zero modes are the same as in the ${\cal N}=2$ theory, the

2165: solutions to the full equations of motion, including Yukawa

2166: sources for $A$, will necessarily differ. We would expect this to

2167: feed back into the worldsheet dynamics. As in the ${\cal N}=(2,2)$

2168: case, it is difficult to determine this explicitly, but thankfully

2169: the lifting of the zero modes is once again dictated by the

2170: symmetries of the problem.

2171: 

2172: \para

2173: Let's start by examining the simplest case, with $\hW(A)=\mu_2

2174: A^2$, so that the fermionic constraint equation is given by

2175: \eqn{02cons}. Integrating out $\sigma$ on the worldsheet once

2176: again gives rise to a four-fermi term

2177: %

2178: \be {\cal L}_{\rm 4-fermi}=

2179: -\frac{|\bar{\xi}_{-i}\xi_{+i}|^2}{(r+2|\mu_2|^2/T)}\label{024fermi}\ee

2180: %

2181: which, up to an overall rescaling, looks the same as the ${\cal

2182: N}=(2,2)$ four-fermi term \eqn{224fermi}. However this is

2183: deceptive, for the constraints \eqn{02cons} ensure that

2184: \eqn{024fermi} now includes a component of $\chi_-$. Previously,

2185: as we discussed in Section \ref{symmetries}, $\chi_-$ was

2186: prohibited from appearing in the four-fermi term since it was a

2187: Goldstino mode in the ${\cal N}=(2,2)$ theory. It loses this

2188: protection in the ${\cal N}=(0,2)$ theory.

2189: 

2190: \para

2191: If the superpotential contains quadratic and higher order terms,

2192: then integrating out $\sigma$ results not only in a four-fermi

2193: term on the worldsheet, but also in a slew of higher order fermion

2194: lifting terms. These terms are an interesting prediction of the

2195: deformation \eqn{booty}.

2196: 

2197: 

2198: 

2199: 

2200: 

2201: 

2202: \subsubsection*{A Comment on Anomalies}

2203: 

2204: 

2205: In Section \eqn{add}, we saw that additional fundamental ${\cal

2206: N}=2$ hypermultiplets in four dimensions contributed extra zero

2207: modes to the vortex string which were captured in the gauged

2208: linear sigma model by adding $(N_f-N_c)$ chiral multiplets in the

2209: anti-fundamental representation of the $U(k)$ worldsheet gauge

2210: group.

2211: 

2212: \para

2213: There exists a trivial generalization in the ${\cal N}=1$ theories

2214: in which we add only four-dimensional chiral multiplets, instead

2215: of full hypermultiplets. For example, the addition of  a single

2216: four dimensional chiral multiplet $Q$, transforming in the ${\bf

2217: N}_c$ of $U(N_c)$, will contribute both bosonic and fermionic zero

2218: modes to the vortex string. These live in an ${\cal N}=(0,2)$

2219: chiral multiplet $\tilde{\Phi}$ of the worldsheet theory,

2220: transforming in the $\bar{\bf k}$ of $U(k)$. In contrast, the

2221: addition of $\tilde{Q}$, transforming in the $\bar{\bf N}_c$ of

2222: $U(N_c)$, will contribute only fermi zero modes, living in an

2223: ${\cal N}=(0,2)$ fermi multiplet $\tilde{\Gamma}$ which transforms

2224: in the $\bar{\bf k}$ of $U(k)$.

2225: 

2226: \para

2227: While the above observation is trivial, there is an interesting

2228: corollary in the quantum theory. The four-dimensional theory with

2229: unequal numbers of fundamental and anti-fundamental chiral

2230: multiplets is inconsistent at the quantum level, suffering a gauge

2231: anomaly. This inconsistency descends to the vortex worldsheet,

2232: which also suffers a $U(k)$ gauge anomaly unless the number of

2233: chiral multiplets $\tilde{\Phi}$ is equal to the number of fermi

2234: multiplets $\tilde{\Gamma}$. It would be interesting to study

2235: vortices in chiral, anomaly free four-dimensional gauge theories,

2236: to see if there is a corresponding delicate anomaly cancellation

2237: on the vortex worldsheet.

2238: 

2239: \subsubsection*{The SQCD Limit}

2240: 

2241: To reach the ${\cal N}=1$ SQCD limit of the four-dimensional

2242: theory, we send $\mu_2\rightarrow \infty$ to decouple the adjoint

2243: chiral multiplet $A$. On the worldsheet, this has the effect of

2244: decoupling the $U(k)$ adjoint chiral multiplet $\Sigma$. At the

2245: same time, the constraint on the left-moving fermions \eqn{02cons}

2246: becomes simply $\chi_-=0$, which effectively removes the fermi

2247: multiplet $\Xi$. The right-moving fermions on the worldsheet are

2248: still constrained to obey $\bar{\phi}\xi_{+i}=0$ (in the case

2249: $N_f=N_c$) while the left-moving fermions $\xi_{-i}$ are

2250: unconstrained. Nonetheless, the theory appears to be free of

2251: worldsheet gauge anomalies.

2252: 

2253: \para

2254: In this limit, the four-dimensional theory develops an enhanced, chiral flavor

2255: symmetry $S[U(N_f)\times U(N_f)]$, rotating left and right movers independently.

2256: (The ``S'' here is to remind us that the overall $U(1)_B$ is part of the gauge

2257: group).

2258: In the presence of the FI parameter, this is broken spontaneously and the

2259: surviving symmetry in the vacuum is,

2260: %

2261: \be S[U(N_c)\times U(N_f-N_c)] \times U(N_f) \times U(1)_R\ee

2262: %

2263: Here the $U(1)_R$ is the anomaly-free R-symmetry. The same

2264: symmetry enhancement is also seen on the vortex worldsheet theory

2265: proposed above. There is once again a particular choice for the

2266: anomaly free R-current.

2267: 

2268: \para

2269: There is an issue with the normalizability of the fermi zero modes

2270: in this limit. As $\mu_2\rightarrow 0$, the Dirac equation for the

2271: left-moving fermi zero modes become ${\cal

2272: D}_z\bar{\tilde{\psi}}_{-i}=0$ which has only non-normalizable

2273: solutions. This could be mirrored on the worldsheet by infinite

2274: kinetic terms for $\Gamma_i$, of the type that we neglected in the

2275: discussion above. Alternatively, one could add a suitable

2276: deformation to the 4d theory, such as the meson field considered

2277: in \cite{gsy}, which once again renders these zero modes finite.

2278: 

2279: 

2280: 

2281: 

2282: \subsection{A D-Brane Construction}

2283: 

2284: %

2285: \EPSFIGURE{vib2.eps,height=160pt}{}

2286: %

2287: One can construct a D-brane configuration whose low-energy

2288: dynamics is governed by the four-dimensional theory of interest,

2289: namely ${\cal N}=2$ super QCD, broken to ${\cal N}=1$ by the

2290: addition of a superpotential $\hW(A)$ for the adjoint chiral

2291: multiplet. One starts with the usual Hanany-Witten set-up for four

2292: dimensional ${\cal N}=2$ gauge theories \cite{hanwit,w}. This

2293: consists of two parallel NS5-branes lying in the $012345$

2294: directions and separated a distance $\Delta X^6\sim l_s/e^2$ in

2295: the $X^6$ direction. The ${\cal N}=2$ $U(N_c)$ gauge theory lives

2296: on $N_c$ D4-branes, with worldvolume $01236$, which are suspended

2297: between these two NS5-branes, while $N_f$ D6-branes with

2298: worldvolume $0123789$ provide the hypermultiplets. To describe the

2299: deformation \eqn{def} to ${\cal N}=1$ supersymmetry, we introduce

2300: the complex coordinates

2301: %

2302: \be v= X^4+iX^5\ \ \ ,\ \ \ w=X^8+iX^9 .\ee

2303: %

2304: A superpotential $\hW(A)$ is induced on the D4-brane worldvolume

2305: if we bend the right-hand NS5-brane so that it no longer lies at

2306: the point $w=0$, but rather on the complex curve \cite{deboeroz}

2307: %

2308: \be w= \hW(v) .\nn\ee

2309: %

2310: Note that in the limit $\mu_n\rightarrow \infty$, with $\mu_n$

2311: defined in \eqn{mun}, the curved NS5-brane becomes multiple flat

2312: NS5-branes, lying a constant values of $v=X^4+iX^5$, given by the

2313: roots of $\hW$. This is the description of the superpotential

2314: first presented in \cite{first,second}.

2315: 

2316: \para

2317: We may now pass through the series of moves described in

2318: \cite{vib}, turning on a FI parameter by separating the two

2319: NS5-branes in the $X^7$ direction, and identifying the vortices as

2320: stretched D2-branes. The final result is shown in figure 3 in the

2321: case of $N_f=N_c$. The figure shows a slice through $w=0$. The

2322: dots depict the roots of $\hW(v)$, where the curved NS5-brane

2323: intersects the $w=0$ plane; the ghostly dotted line shows where

2324: the NS5-brane has left this plane and is living at some other

2325: value of $w$. Figure 3 corresponds to a quartic superpotential,

2326: with three critical points. One can check that the theory on the

2327: D2-brane preserves ${\cal N}=(0,2)$ supersymmetry. It is clear

2328: from the brane set-up that the D2-brane has a supersymmetric

2329: ground state only when it may safely stretch from the curved

2330: NS5-brane to a D4-brane, remaining at constant $v=m_i$ and without

2331: leaving the safety of $w=0$. This requires

2332: %

2333: \be \left.\frac{\partial \hW(v)}{\partial v}\right|_{m_i}=0 .

2334: \label{itsgood}\ee

2335: %

2336: This is the brane perspective on the statement that BPS vortices

2337: only exist when \eqn{itsgood} is satisfied. It provides further

2338: evidence that a worldsheet superpotential of the form \eqn{booty}

2339: is required.

2340: 

2341: 

2342: 

2343: 

2344: \subsection{A Different Superpotential}

2345: \label{more}

2346: 

2347: To end this section, we consider a different deformation of the

2348: ${\cal N}=2$ theory which breaks the four dimensional

2349: supersymmetry to ${\cal N}=1$. We add a superpotential of the

2350: form,

2351: %

2352: \be {\cal W}_{{\cal N}=2} = \sqrt{2}\sum_{i=1}^{N_f}\,\tilde{Q}_i

2353: {\cal V}_i(A) Q_i .\label{another}\ee

2354: %

2355: Here ${\cal V}_i(A)$ is an arbitrary holomorphic function of $A$.

2356: The four-dimensional quantum dynamics of theories of this type was

2357: previously studied in \cite{kap,seiji,rab,deboeroz}. We are here

2358: interested in the effect on the vortex worldsheet. In fact, we

2359: have already met one example of such a deformation that preserves

2360: ${\cal N}=2$ supersymmetry, because the complex mass term is of

2361: this form with ${\cal V}_i(A)=A-m_i$. In that case, we saw that

2362: the effect was not to induce a superpotential on the worldsheet,

2363: but instead to change the relationship between $(0,2)$ fermi and

2364: chiral fields,

2365: %

2366: \be \bar{\cal D}_+\Gamma_i=2i\Sigma\Phi_i\ \longrightarrow \

2367: \bar{\cal D}_+\Gamma_i=2i(\Sigma-m_i)\Phi_i . \ee

2368: %

2369: Given this, it is natural to conjecture that the general

2370: deformation \eqn{another} is captured by the worldsheet theory

2371: with the relationship,

2372: %

2373: \be \bar{\cal D}_+\Gamma_i= 2i{\cal

2374: V}_i(\Sigma)\Phi_i .\label{vsup}\ee

2375: %

2376: We will now provide evidence that this is indeed the case. We will

2377: show that the deformation \eqn{vsup} is in agreement with all

2378: symmetries of the theory, and reproduces the known behavior of the

2379: vortex. The details of the calculations are similar to those

2380: presented earlier, so we shall be brief.

2381: 

2382: \para

2383: Let us firstly study what becomes of the vortex. We take the

2384: vacuum of the four-dimensional theory to be

2385: %

2386: \be Q^a_{\ i}=v\delta^a_{\ i}\ \ \ ,\ \ \ \tilde{Q}_i=0\ \ \ \

2387: A={\rm diag}(\nu_1,\ldots,\nu_{N_c})\label{4dvac}\ee

2388: %

2389: where $\nu_i$ is one of the roots of ${\cal V}_i$. If the $\nu_i$

2390: are all distinct, the situation is the same as the one we

2391: encountered in Section 2.5.1 with distinct masses $m_i$: there are

2392: $N_c$ different vortices, each supported by the winding of a

2393: different $Q_i$. In contrast, if all $\nu_i$ coincide, the full

2394: ${\bf CP}^{N_c-1}$ internal moduli space of the vortex is

2395: restored.

2396: 

2397: \para

2398: Let us see how this is reproduced on the vortex worldsheet by the

2399: deformation \eqn{vsup}. For definiteness, we take a single $k=1$

2400: vortex string in the $N_f=N_c$ theory. The bosonic part of the

2401: worldsheet theory is given by,

2402: %

2403: \be  {\cal L}_{\rm bose}=T\,|\partial_m z|^2 +

2404: \sum_{a=1}^{N_c}\left(|{\cal D}_m\phi^a|^2 - 2|{\cal

2405: V}_i(\sigma)|^2|\phi_i|^2\right)

2406: +D(\sum_{i=1}^{N_c}|\phi_i|^2-r)+\frac{\theta}{2\pi}u_{01} .\nn\ee

2407: %

2408: If the roots of $\nu_i$ of ${\cal V}_i(\sigma)$ are distinct, this

2409: theory has isolated vacua, given by

2410: %

2411: \be |\phi_j|^2=r\delta_{ij}\ \ \ ,\ \ \ \sigma

2412: =\nu_i .\label{tricky}\ee

2413: %

2414: However, there is an ambiguity here since ${\cal V}_i$ has

2415: multiple roots $\nu_i$.  Suppose, for definiteness,  that ${\cal

2416: V}_i(\sigma)$ is a polynomial of degree $P_i$. Then it appears

2417: that, for each $i=1,\ldots, N_c$, there are $P_i$ different vacua

2418: of the worldsheet theory. How are we to interpret these? In past

2419: examples \cite{memono,sy,vstring}, different vacua of the

2420: worldsheet corresponded to different physical vortices ---  see

2421: Section 2.5.1. But we certainly don't want the same interpretation

2422: here because the four-dimensional theory doesn't have $P_i$

2423: distinct vortices, each with $Q_i$ winding asymptotically.

2424: Thankfully, the interpretation of the multiple worldsheet vacua in

2425: the present case is somewhat different. For fixed

2426: $i=1,\ldots,N_c$, the $P_i$ different vacua differ only in the

2427: value of the auxiliary field $\sigma$. The field $\sigma$ is to be

2428: integrated out, set equal to its classical, algebraic equation of

2429: motion. But there are $P_i$ different solutions to this algebraic

2430: equation. The theory is only complete if we specify which of these

2431: solutions we are to take. This means that the vacuum $\sigma =

2432: \nu_i$ chosen in \eqn{tricky} is not a dynamical variable, but

2433: rather a parameter of the worldsheet theory. We are therefore free

2434: to fix it as we please, and the only natural candidate is to

2435: equate it with the four-dimensional vacuum value $\nu_i$ in

2436: \eqn{4dvac}\footnote{The equation of motion for $\sigma$ includes

2437: a term  bilinear in the fermions, seen explicitly in \eqn{ria}.

2438: The root of the equation of motion is taken to be the

2439: four-dimensional vacuum value $\nu_i$ when the fermions vanish,

2440: and is continuously connected to $\nu_i$ when the fermions turn

2441: on.}. The end result is a situation where the same worldsheet

2442: Lagrangian describes the vortex string in different

2443: four-dimensional vacua; the specific four-dimensional vacuum of

2444: interest appears as a boundary condition on the auxiliary $\sigma$

2445: field.

2446: 

2447: \para

2448: As a check of the conjecture \eqn{vsup}, we can confirm that the

2449: $U(1)_R\times U(1)_V$ charges are consistent. If we write the

2450: superpotential as

2451: %

2452: \be {\cal V}_i(A) = \sum_{n=0}\ h^{(i)}_n A^n\ee

2453: %

2454: then we are required to assign spurion charge $(2-2n,0)$ to

2455: $h_n^{(i)}$. Let's check that this is in agreement with the

2456: worldsheet. The deformation \eqn{vsup} gives rise to the terms

2457: %

2458: \be L_{\rm vortex} = \ldots + \sqrt{2}\sum_n \ (n h_n^{(i)}

2459: \bar{\xi}_{-i}\sigma^{n-1}\phi_i\bar{\zeta_+}\ +\

2460: h_n^{(i)}\bar{\xi}_{-i}\sigma^n\xi_{+i})+\ldots \label{ria}\ee

2461: %

2462: from which we learn that $h_n^{(i)}$ must again be assigned charge

2463: $(2-2n,0)$ under the worldsheet $U(1)_R\times U(1)_V$.

2464: 

2465: 

2466: 

2467: \newpage

2468: \section*{Appendix: The Alphabet}

2469: 

2470: This appendix is included to help the reader keep track of the

2471: burgeoning conventions. The four dimensional fields are all

2472: components of ${\cal N}=1$ superfields,

2473: %

2474: \be

2475: \begin{array}{ll}

2476: A_\mu: & \mbox{4d gauge field  in the vector multiplet $V$} \\

2477: A: & \mbox{Adjoint valued 4d scalar in the chiral multiplet $A$} \\

2478: Q_i: & \mbox{Fundamental 4d scalar  in the chiral multiplet $Q_i$} \\

2479: \tilde{Q}_j: & \mbox{Fundamental 4d scalar in the  chiral

2480: multiplet $\tilde{Q}_j$}

2481: \\ \lambda: & \mbox{Adjoint valued 4d

2482: fermion  in the vector multiplet $V$} \\

2483: \eta: & \mbox{Adjoint valued 4d fermion in the chiral multiplet

2484: $A$} \\ \psi_i: & \mbox{Fundamental 4d

2485: fermion living in the chiral multiplet $Q_i$} \\

2486: \tilde{\psi}_j: & \mbox{Anti-fundamental 4d fermion in the chiral

2487: multiplet $\tilde{Q}_j$} .\end{array}\nn\ee

2488: %

2489: The worldsheet fields are all components of ${\cal N}=(0,2)$

2490: superfields as described in Section 3:

2491: %

2492: \be

2493: \begin{array}{ll}

2494: z: & \mbox{Worldsheet scalar arising from broken translational

2495: invariance,} \\ & \mbox{ in the chiral multiplet $Z$} \\

2496: \phi_i: & \mbox{Worldsheet scalar corresponding to orientation

2497: modes

2498: of the string,} \\ & \mbox{in the chiral multiplet $\Phi_i$} \\

2499: \sigma: & \mbox{Worldsheet auxiliary scalar in the  chiral multiplet $\Sigma$} \\

2500: u_m: & \mbox{Worldsheet gauge field in the vector multiplet $U$} \\

2501: {\chi}_+ & \mbox{Worldsheet Goldstino fermion  in the chiral

2502: multiplet $Z$}

2503: \\ \chi_- & \mbox{Worldsheet fermion in the fermion

2504: multiplet $\Xi$} \\

2505: \xi_{+i}: & \mbox{Worldsheet fermions living in the fermion

2506: multiplet $\Phi_i$.} \\ \xi_{-i}: & \mbox{Worldsheet fermions  in

2507: the fermion multiplet $\Gamma_i$} \\ \bar{\zeta}_+: &

2508: \mbox{Worldsheet

2509: auxiliary fermion living in the chiral multiplet $\Sigma$.} \\

2510: \zeta_-: & \mbox{Worldsheet auxiliary fermion  in the vector

2511: multiplet $U$} .

2512: 

2513: \nn\end{array}\ee

2514: 

2515: \section*{Acknowledgement}

2516: We would like to thank Philip Argyres, Adam Ritz, and especially

2517: Misha Shifman

2518: and Alyosha Yung for helpful discussions. M.E. is supported in

2519: part by DOE grant FG02-84ER-40153. D.T. is supported by the Royal

2520: Society.

2521: 

2522: \begin{thebibliography}{99}

2523: 

2524: \small

2525: \parskip=0pt plus 2pt

2526: 

2527: \bibitem{vib} A.~Hanany and D.~Tong,

2528:   ``{\it Vortices, instantons and branes},''

2529:   JHEP {\bf 0307}, 037 (2003)

2530:   [arXiv:hep-th/0306150].

2531:   %%CITATION = HEP-TH 0306150;%%

2532: 

2533: \bibitem{auzzi}

2534: R.~Auzzi, S.~Bolognesi, J.~Evslin, K.~Konishi and A.~Yung,

2535:   ``{\it Nonabelian superconductors: Vortices and confinement in N = 2 SQCD},''

2536:   Nucl.\ Phys.\ B {\bf 673}, 187 (2003)

2537:   [arXiv:hep-th/0307287].

2538:   %%CITATION = HEP-TH 0307287;%%

2539: 

2540: 

2541: \bibitem{sy}  M.~Shifman and A.~Yung,

2542:   ``{\it Non-Abelian string junctions as confined monopoles},''

2543:   Phys.\ Rev.\ D {\bf 70}, 045004 (2004)

2544:   [arXiv:hep-th/0403149].

2545:   %%CITATION = HEP-TH 0403149;%%

2546: 

2547: 

2548: \bibitem{vstring} A.~Hanany and D.~Tong,

2549:   ``{\it Vortex strings and four-dimensional gauge dynamics},''

2550:   JHEP {\bf 0404}, 066 (2004)

2551:   [arXiv:hep-th/0403158].

2552:   %%CITATION = HEP-TH 0403158;%%

2553: 

2554: \bibitem{nick}

2555:  N.~Dorey,

2556:   ``{\it The BPS spectra of two-dimensional supersymmetric gauge theories with

2557:   twisted mass terms},''

2558:   JHEP {\bf 9811}, 005 (1998)

2559:   [arXiv:hep-th/9806056].

2560:   %%CITATION = HEP-TH 9806056;%%

2561: 

2562: \bibitem{dht}

2563:  N.~Dorey, T.~J.~Hollowood and D.~Tong,

2564:   ``{\it The BPS spectra of gauge theories in two and four dimensions},''

2565:   JHEP {\bf 9905}, 006 (1999)

2566:   [arXiv:hep-th/9902134].

2567:   %%CITATION = HEP-TH 9902134;%%

2568: 

2569: \bibitem{scvs}  D.~Tong,

2570:   ``{\it Superconformal vortex strings},''

2571:   JHEP {\bf 0612}, 051 (2006)

2572:   [arXiv:hep-th/0610214].

2573:   %%CITATION = HEP-TH 0610214;%%

2574: 

2575: \bibitem{tasi}   D.~Tong,

2576:   ``{\it TASI lectures on solitons},''

2577:   arXiv:hep-th/0509216.

2578:   %%CITATION = HEP-TH 0509216;%%

2579: 

2580: 

2581: \bibitem{other}

2582: V.~Markov, A.~Marshakov and A.~Yung,

2583:   ``{\it Non-Abelian vortices in N = 1* gauge theory},''

2584:   Nucl.\ Phys.\ B {\bf 709}, 267 (2005)

2585:   [arXiv:hep-th/0408235]. \\

2586:   %%CITATION = HEP-TH 0408235;%%

2587: 

2588: \bibitem{others}

2589:  A.~Gorsky, M.~Shifman and A.~Yung,

2590:   ``{\it Non-Abelian Meissner effect in Yang-Mills theories at weak

2591: coupling},''   Phys.\ Rev.\ D {\bf 71}, 045010 (2005)

2592:   [arXiv:hep-th/0412082].

2593:   %%CITATION = HEP-TH 0412082;%%

2594: 

2595: \bibitem{bolog}   S.~Bolognesi,

2596:   ``{\it The holomorphic tension of vortices},''

2597:   JHEP {\bf 0501}, 044 (2005)

2598:   [arXiv:hep-th/0411075]; ``{\it The holomorphic tension of nonabelian vortices and the quark = dual-quark

2599:   condensate},''

2600:   Nucl.\ Phys.\ B {\bf 719}, 67 (2005)

2601:   [arXiv:hep-th/0412241].

2602:   %%CITATION = HEP-TH 0412241;%%

2603:   %%CITATION = HEP-TH 0411075;%%

2604: 

2605: \bibitem{syss} M.~Shifman and A.~Yung,

2606:   ``{\it Non-abelian flux tubes in SQCD: Supersizing world-sheet supersymmetry},''

2607:   arXiv:hep-th/0501211.

2608:   %%CITATION = HEP-TH 0501211;%%

2609: 

2610: \bibitem{gsy}  A.~Gorsky, M.~Shifman and A.~Yung,

2611:   ``{\it N = 1 supersymmetric quantum chromodynamics: How confined non-Abelian

2612:   monopoles emerge from quark condensation},''

2613:   arXiv:hep-th/0701040.

2614:   %%CITATION = HEP-TH 0701040;%%

2615: 

2616: 

2617: 

2618: \bibitem{no}  H.~B.~Nielsen and P.~Olesen,

2619:   ``{\it Vortex-Line Models For Dual Strings},''

2620:   Nucl.\ Phys.\ B {\bf 61}, 45 (1973).

2621:   %%CITATION = NUPHA,B61,45;%%

2622: 

2623: \bibitem{jackr}   R.~Jackiw and P.~Rossi,

2624:   ``{\it Zero Modes Of The Vortex-Fermion System},''

2625:   Nucl.\ Phys.\  B {\bf 190}, 681 (1981).

2626:   %%CITATION = NUPHA,B190,681;%%

2627: 

2628: \bibitem{wb}   J.~Wess and J.~Bagger,

2629:   ``{\it Supersymmetry and supergravity},'' Princeton Univ. Pr. (1992)

2630: 

2631: \bibitem{stefan}   S.~Vandoren and P.~van Nieuwenhuizen,

2632:   ``{\it New instantons in the double-well potential},''

2633:   Phys.\ Lett.\ B {\bf 499}, 280 (2001)

2634:   [arXiv:hep-th/0010130].

2635:   %%CITATION = HEP-TH 0010130;%%

2636: 

2637: \bibitem{memono}   D.~Tong,

2638:   ``{\it Monopoles in the Higgs phase},''

2639:   Phys.\ Rev.\ D {\bf 69}, 065003 (2004)

2640:   [arXiv:hep-th/0307302].

2641:   %%CITATION = HEP-TH 0307302;%%

2642: 

2643: \bibitem{hh} A.~Hanany and K.~Hori,

2644:   ``{\it Branes and N=2 theories in two dimensions},''

2645:   Nucl.\ Phys.\ B {\bf 513}, 119 (1998)

2646:   [arXiv:hep-th/9707192].

2647:   %%CITATION = HEP-TH 9707192;%%

2648: 

2649: 

2650: 

2651: 

2652:   \bibitem{ward}

2653:   R.~S.~Ward,

2654:   ``{\it Slowly Moving Lumps In The $CP^1$ Model In (2+1)-Dimensions},''

2655:   Phys.\ Lett.\ B {\bf 158}, 424 (1985).

2656:   %%CITATION = PHLTA,B158,424;%%

2657: 

2658: \bibitem{ls}   R.~A.~Leese and T.~M.~Samols,

2659:   ``{\it Interaction of semilocal vortices},''

2660:   Nucl.\ Phys.\ B {\bf 396}, 639 (1993).

2661:   %%CITATION = NUPHA,B396,639;%%

2662: 

2663: \bibitem{semi}

2664: M.~Shifman and A.~Yung,

2665:   ``{\it Non-Abelian semilocal strings in N=2 supersymmetric QCD},''

2666:   Phys.\ Rev.\ D {\bf 73}, 125012 (2006)

2667:   [arXiv:hep-th/0603134].

2668:   %%CITATION = HEP-TH 0603134;%%

2669: 

2670: \bibitem{semi2}  A.~Achucarro and T.~Vachaspati,

2671:   ``{\it Semilocal and electroweak strings},''

2672:   Phys.\ Rept.\  {\bf 327}, 347 (2000)

2673:   [Phys.\ Rept.\  {\bf 327}, 427 (2000)]

2674:   [arXiv:hep-ph/9904229].

2675:   %%CITATION = HEP-PH 9904229;%%

2676: 

2677:   \bibitem{penin}  A.~A.~Penin, V.~A.~Rubakov, P.~G.~Tinyakov and S.~V.~Troitsky,

2678:   ``{\it What becomes of vortices in theories with flat directions},''

2679:   Phys.\ Lett.\ B {\bf 389}, 13 (1996)

2680:   [arXiv:hep-ph/9609257].

2681:   %%CITATION = HEP-PH 9609257;%%

2682: 

2683: \bibitem{davis}  A.~Achucarro, A.~C.~Davis, M.~Pickles and J.~Urrestilla,

2684:   ``{\it Vortices in theories with flat directions},''

2685:   Phys.\ Rev.\ D {\bf 66}, 105013 (2002)

2686:   [arXiv:hep-th/0109097].

2687:   %%CITATION = HEP-TH 0109097;%%

2688: 

2689: \bibitem{hanwit}   A.~Hanany and E.~Witten,

2690:   ``{\it Type IIB superstrings, BPS monopoles, and three-dimensional gauge

2691:   dynamics},''

2692:   Nucl.\ Phys.\ B {\bf 492}, 152 (1997)

2693:   [arXiv:hep-th/9611230].

2694:   %%CITATION = HEP-TH 9611230;%%

2695: 

2696: 

2697: \bibitem{w} E.~Witten,

2698:   ``{\it Solutions of four-dimensional field theories via M-theory},''

2699:   Nucl.\ Phys.\ B {\bf 500}, 3 (1997)

2700:   [arXiv:hep-th/9703166].

2701:   %%CITATION = HEP-TH 9703166;%%

2702: 

2703: \bibitem{mmatrix} M.~Eto, Y.~Isozumi, M.~Nitta, K.~Ohashi and N.~Sakai,

2704:   ``{\it Moduli space of non-Abelian vortices},''

2705:   Phys.\ Rev.\ Lett.\  {\bf 96}, 161601 (2006)

2706:   [arXiv:hep-th/0511088].

2707:   %%CITATION = HEP-TH 0511088;%%

2708: 

2709: \bibitem{mmreview} M.~Eto, Y.~Isozumi, M.~Nitta, K.~Ohashi and N.~Sakai,

2710:   ``{\it Solitons in the Higgs phase: The moduli matrix approach},''

2711:   J.\ Phys.\ A {\bf 39}, R315 (2006)

2712:   [arXiv:hep-th/0602170].

2713:   %%CITATION = HEP-TH 0602170;%%

2714: 

2715: 

2716: \bibitem{manton} N.~S.~Manton, ``{\em A Remark On The Scattering Of BPS Monopoles}'',

2717: Phys.\ Lett.\ B {\bf 110}, 54 (1982).

2718: %%CITATION = PHLTA,B110,54;%%

2719: 

2720: \bibitem{samols} T.~M.~Samols,

2721: ``{\em Vortex Scattering}'' Commun.\ Math.\ Phys.\  {\bf 145}, 149

2722: (1992).

2723: %%CITATION = CMPHA,145,149;%%

2724: 

2725: 

2726: 

2727: \bibitem{asy} R.~Auzzi, M.~Shifman and A.~Yung,

2728:   ``{\it Composite non-Abelian flux tubes in N = 2 SQCD},''

2729:   Phys.\ Rev.\ D {\bf 73}, 105012 (2006)

2730:   [arXiv:hep-th/0511150].

2731:   %%CITATION = HEP-TH 0511150;%%

2732: 

2733: \bibitem{kt} K.~Hashimoto and D.~Tong,

2734:   ``{\it Reconnection of non-abelian cosmic strings},''

2735:   JCAP {\bf 0509}, 004 (2005)

2736:   [arXiv:hep-th/0506022].

2737:   %%CITATION = HEP-TH 0506022;%%

2738: 

2739: \bibitem{lots}  M.~Eto, K.~Hashimoto, G.~Marmorini, M.~Nitta, K.~Ohashi and W.~Vinci,

2740:   ``{\it Universal reconnection of non-Abelian cosmic strings},''

2741:   arXiv:hep-th/0609214.

2742:   %%CITATION = HEP-TH 0609214;%%

2743: 

2744: 

2745: \bibitem{hw}

2746: C.~M.~Hull and E.~Witten, ``{\it Supersymmetric Sigma Models And

2747: The Heterotic String}'' Phys.\ Lett.\ B {\bf 160}, 398 (1985).

2748: %%CITATION = PHLTA,B160,398;%%

2749: 

2750: \bibitem{ds}

2751:  M.~Dine and N.~Seiberg, ``{\it $(2,0)$ Superspace},''

2752:   Phys.\ Lett.\ B {\bf 180}, 364 (1986).

2753:   %%CITATION = PHLTA,B180,364;%%

2754: 

2755: \bibitem{phases}

2756: E.~Witten, ``{\it Phases of N = 2 theories in two dimensions},''

2757: Nucl.\ Phys.\ B {\bf 403} (1993) 159 [arXiv:hep-th/9301042].

2758: %%CITATION = HEP-TH 9301042;%%

2759: 

2760: \bibitem{abs}

2761:  A.~Adams, A.~Basu and S.~Sethi, ``{\it $(0,2)$ duality},''

2762:   Adv.\ Theor.\ Math.\ Phys.\  {\bf 7}, 865 (2004)

2763:   [arXiv:hep-th/0309226].

2764:   %%CITATION = HEP-TH 0309226;%%

2765: 

2766: \bibitem{dv}   R.~Dijkgraaf and C.~Vafa,

2767:   ``{\it A perturbative window into non-perturbative physics},''

2768:   arXiv:hep-th/0208048.

2769:   %%CITATION = HEP-TH 0208048;%%

2770: 

2771: 

2772: 

2773: 

2774: \bibitem{ddt}  S.~C.~Davis, A.~C.~Davis and M.~Trodden,

2775:   ``{\it N=1 supersymmetric cosmic strings},''

2776:   Phys.\ Lett.\ B {\bf 405}, 257 (1997)

2777:   [arXiv:hep-ph/9702360].

2778:   %%CITATION = HEP-PH 9702360;%%

2779: 

2780: \bibitem{deboeroz} J.~de Boer and Y.~Oz,

2781:   ``{\it Monopole condensation and confining phase of N = 1 gauge theories via

2782:   M-theory fivebrane},''

2783:   Nucl.\ Phys.\ B {\bf 511}, 155 (1998)

2784:   [arXiv:hep-th/9708044].

2785:   %%CITATION = HEP-TH 9708044;%%

2786: 

2787: \bibitem{first} S.~Elitzur, A.~Giveon and D.~Kutasov,

2788:   ``{\it Branes and N = 1 duality in string theory},''

2789:   Phys.\ Lett.\ B {\bf 400}, 269 (1997)

2790:   [arXiv:hep-th/9702014].

2791:   %%CITATION = HEP-TH 9702014;%%

2792: 

2793: \bibitem{second}   S.~Elitzur, A.~Giveon, D.~Kutasov, E.~Rabinovici and A.~Schwimmer,

2794:   ``{\it Brane dynamics and N = 1 supersymmetric gauge theory},''

2795:   Nucl.\ Phys.\ B {\bf 505}, 202 (1997)

2796:   [arXiv:hep-th/9704104].

2797:   %%CITATION = HEP-TH 9704104;%%

2798: 

2799: 

2800: 

2801: \bibitem{kap}  A.~Kapustin,

2802:   ``{\it The Coulomb branch of N = 1 supersymmetric gauge theory with adjoint  and

2803:   fundamental matter},''

2804:   Phys.\ Rev.\D {\bf 66}, 010001 (2002)

2805:   arXiv:hep-th/9611049.

2806:   %%CITATION = HEP-TH 9611049;%%

2807: 

2808: \bibitem{seiji}

2809: T.~Kitao, S.~Terashima and S.~K.~Yang,

2810:   ``{\it N = 2 curves and a Coulomb phase in N = 1 SUSY gauge theories with  adjoint

2811:   and fundamental matters},''

2812:   Phys.\ Lett.\  B {\bf 399}, 75 (1997)

2813:   [arXiv:hep-th/9701009].

2814:   %%CITATION = PHLTA,B399,75;%%

2815: 

2816: 

2817: \bibitem{rab} A.~Giveon, O.~Pelc and E.~Rabinovici,

2818:   ``{\it The Coulomb phase in N = 1 gauge theories with an LG-type

2819:   superpotential},''

2820:   Nucl.\ Phys.\ B {\bf 499}, 100 (1997)

2821:   [arXiv:hep-th/9701045].

2822:   %%CITATION = HEP-TH 9701045;%%

2823: 

2824: 

2825: %\bibitem{seiberg}   N.~Seiberg,

2826: %  ``{\it Electric-magnetic duality in supersymmetric nonAbelian gauge theories},''

2827: %  Nucl.\ Phys.\ B {\bf 435}, 129 (1995)

2828: %  [arXiv:hep-th/9411149].

2829: %  %%CITATION = HEP-TH 9411149;%%

2830: 

2831: 

2832: 

2833: 

2834: \end{thebibliography}

2835: 

2836: 

2837: \end{document}

2838: