1: \documentclass[11pt]{article}

2: %\usepackage{fancyheadings}

3: \usepackage[dvips]{graphicx}

4: \usepackage{graphics}

5: \usepackage{ifthen}

6: \usepackage{amsfonts}

7: \usepackage{amsmath}

8: \usepackage{epsfig}

9: \usepackage{youngtab}

10: \usepackage{multirow}

11: \usepackage[errorshow]{tracefnt}

12: \parindent 0pt

13:  \parskip.2cm

14:  \topmargin -1.0cm \textheight 23.0cm

15:  \textwidth 16.0cm

16:  \oddsidemargin=0.25cm\evensidemargin=0.25cm

17: \def\bsh{\backslash}

18:  \def\bdt{\dot \beta}

19:  \def\adt{\dot \alpha}

20:  \def\pl{ \: + }

21:  \def\mi{ \: - }

22:  \def\tbar{\bar \theta}

23:  \newfont{\bbbold}{msbm10 scaled \magstep1}

24:  \def\com{\mbox{\bbbold C}}

25:  \def\kvec{\mbox{\boldmath$k$}}

26:  \def\Kvec{\mbox{\boldmath$K$}}

27:  \def\bbA{\mbox{\bbbold A}}

28:  \def\bbB{\mbox{\bbbold B}}

29:  \def\bbC{\mbox{\bbbold C}}

30:  \def\bbD{\mbox{\bbbold D}}

31:  \def\bbE{\mbox{\bbbold E}}

32:  \def\bbF{\mbox{\bbbold F}}

33:  \def\bbG{\mbox{\bbbold G}}

34:  \def\bbH{\mbox{\bbbold H}}

35:  \def\bbI{\mbox{\bbbold I}}

36:  \def\bbJ{\mbox{\bbbold J}}

37:  \def\bbK{\mbox{\bbbold K}}

38:  \def\bbL{\mbox{\bbbold L}}

39:  \def\bbM{\mbox{\bbbold M}}

40:  \def\bbN{\mbox{\bbbold N}}

41:  \def\bbO{\mbox{\bbbold O}}

42:  \def\bbP{\mbox{\bbbold P}}

43:  \def\bbQ{\mbox{\bbbold Q}}

44:  \def\bbR{\mbox{\bbbold R}}

45:  \def\bbS{\mbox{\bbbold S}}

46:  \def\bbT{\mbox{\bbbold T}}

47:  \def\bbU{\mbox{\bbbold U}}

48:  \def\bbV{\mbox{\bbbold V}}

49:  \def\bbW{\mbox{\bbbold W}}

50:  \def\bbX{\mbox{\bbbold X}}

51:  \def\bbY{\mbox{\bbbold Y}}

52:  \def\bbX{\mbox{\bbbold X}}

53:  \def\cA{{\cal A}}

54:  \def\cB{{\cal B}}

55:  \def\cC{{\cal C}}

56:  \def\cD{{\cal D}}

57:  \def\cE{{\cal E}}

58:  \def\cF{{\cal F}}

59:  \def\cG{{\cal G}}

60:  \def\cH{{\cal H}}

61:  \def\cI{{\cal I}}

62:  \def\cJ{{\cal J}}

63:  \def\cK{{\cal K}}

64:  \def\cL{{\cal L}}

65:  \def\cM{{\cal M}}

66:  \def\cN{{\cal N}}

67:  \def\cO{{\cal O}}

68:  \def\cP{{\cal P}}

69:  \def\cQ{{\cal Q}}

70:  \def\cR{{\cal R}}

71:  \def\cS{{\cal S}}

72:  \def\cT{{\cal T}}

73:  \def\cU{{\cal U}}

74:  \def\cV{{\cal V}}

75:  \def\cW{{\cal W}}

76:  \def\cX{{\cal X}}

77:  \def\cY{{\cal Y}}

78:  \def\cZ{{\cal Z}}

79:  \def\cp{\mbox{\bbbold C}\mbox{\bbbold P}}

80:  \newtheorem{definition}{Definition}

81:  \newtheorem{theorem}{Theorem}

82:  \newtheorem{proposition}{Proposition}

83:  \newfont{\goth}{eufm10 scaled \magstep1}

84:  \def\gA{\mbox{\goth A}}\def\ga{\mbox{\goth a}}

85:  \def\gB{\mbox{\goth B}}\def\gb{\mbox{\goth b}}

86:  \def\gC{\mbox{\goth C}}\def\gc{\mbox{\goth c}}

87:  \def\gD{\mbox{\goth D}}\def\gd{\mbox{\goth d}}

88:  \def\gE{\mbox{\goth E}}\def\ge{\mbox{\goth e}}

89:  \def\gF{\mbox{\goth F}}\def\gf{\mbox{\goth f}}

90:  \def\gG{\mbox{\goth G}}\def\gg{\mbox{\goth g}}

91:  \def\gH{\mbox{\goth H}}\def\gh{\mbox{\goth h}}

92:  \def\gI{\mbox{\goth I}}\def\gi{\mbox{\goth i}}

93:  \def\gJ{\mbox{\goth J}}\def\gj{\mbox{\goth j}}

94:  \def\gK{\mbox{\goth K}}\def\gk{\mbox{\goth k}}

95:  \def\gL{\mbox{\goth L}}\def\gl{\mbox{\goth l}}

96:  \def\gM{\mbox{\goth M}}\def\gm{\mbox{\goth m}}

97:  \def\gN{\mbox{\goth N}}\def\gn{\mbox{\goth n}}

98:  \def\gO{\mbox{\goth O}}\def\go{\mbox{\goth o}}

99:  \def\gP{\mbox{\goth P}}\def\gp{\mbox{\goth p}}

100:  \def\gQ{\mbox{\goth Q}}\def\gq{\mbox{\goth q}}

101:  \def\gR{\mbox{\goth R}}\def\gr{\mbox{\goth r}}

102:  \def\gS{\mbox{\goth S}}\def\gs{\mbox{\goth s}}

103:  \def\gT{\mbox{\goth T}}\def\gt{\mbox{\goth t}}

104:  \def\gU{\mbox{\goth U}}\def\gu{\mbox{\goth u}}

105:  \def\gV{\mbox{\goth V}}\def\gv{\mbox{\goth v}}

106:  \def\gW{\mbox{\goth W}}\def\gw{\mbox{\goth w}}

107:  \def\gX{\mbox{\goth X}}\def\gx{\mbox{\goth x}}

108:  \def\gY{\mbox{\goth Y}}\def\gy{\mbox{\goth y}}

109:  \def\gX{\mbox{\goth X}}\def\gX{\mbox{\goth X}}

110:  \def\xX{\times}

111:  \def\a{\alpha}

112:  \def\b{\beta}

113:  \def\c{\gamma}\def\C{\Gamma}\def\cdt{\dot\gamma}

114:  \def\d{\delta}\def\D{\Delta}\def\ddt{\dot\delta}

115:  \def\e{\epsilon}\def\vare{\varepsilon}

116:  \def\f{\phi}\def\F{\Phi}\def\vf{\varphi}

117:  \def\h{\eta}\def\z{\zeta}

118:  %\def\i{\iota}

119:  \def\k{\kappa}

120:  \def\l{\lambda}\def\L{\Lambda}

121:  \def\m{\mu}

122:  \def\n{\nu}

123:  \def\p{\pi}

124:  \def\P{\Pi}

125:  \def\r{\rho}

126:  \def\s{\sigma}\def\S{\Sigma}

127:  \def\t{\tau}

128:  \def\th{\theta}\def\Th{\Theta}\def\vth{\vartheta}

129:  \def\X{\Xeta}

130:  \def\x{\xi}

131:  \def\o{\omega}\def\O{\Omega}

132:  \def\del{\partial}\def\delb{\bar\partial}

133:  \def\ua{\underline{\alpha}}

134:  \def\ub{\underline{\phantom{\alpha}}\!\!\!\beta}

135:  \def\uc{\underline{\phantom{\alpha}}\!\!\!\gamma}

136:  \def\um{\underline{\mu}}

137:  \def\ud{\underline\delta}

138:  \def\ue{\underline\epsilon}

139:  \def\una{\underline a}\def\unA{\underline A}

140:  \def\unb{\underline b}\def\unB{\underline B}

141:  \def\unc{\underline c}\def\unC{\underline C}

142:  \def\und{\underline d}\def\unD{\underline D}

143:  \def\une{\underline e}\def\unE{\underline E}

144:  \def\unf{\underline{\phantom{e}}\!\!\!\! f}\def\unF{\underline F}

145: \def\unX{\underline X}

146:  \def\ung{\underline g}

147:  \def\unm{\underline m}\def\unM{\underline M}

148:  \def\unn{\underline n}\def\unN{\underline N}

149:  \def\unp{\underline{\phantom{a}}\!\!\! p}\def\unP{\underline P}

150:  \def\unq{\underline{\phantom{a}}\!\!\! q}

151:  \def\unQ{\underline{\phantom{A}}\!\!\!\! Q}

152:  \def\unH{\underline{H}}\def\unG{\underline{G}}

153:  \def\unF{\underline{F}}\def\unT{\underline{T}}\def\unR{\underline{R}}

154:  \def\wt{\widetilde}

155:  \def\sd{\rm sdet}\def\str{\rm str}

156:  \def\xz{\times}

157:  \def\oa{\overline{\alpha}}

158:  \def\ob{\overline{\phantom{\alpha}}\!\!\!\beta}

159:  \def\oc{\overline{\gamma}}

160:  \def\ova{\overline{a}}\def\ovb{\overline{b}}\def\ovc{\overline{c}}

161:  \def\ovA{\overline{A}}\def\ovB{\overline{B}}

162:  \def\nab{\nabla}

163:  \def\tT{\tilde{T}}\def\tR{\tilde{R}}\def\tO{\tilde{\Omega}}\def\tE{\tilde{E}}

164:  \def\tpsi{\tilde{\psi}}

165:  \def\tL{\tilde{L}}

166:  \def\tJ{\tilde{J}}

167:  \def\tQ{\tilde{Q}}

168:  \def\tq{\tilde{q}}

169:  \def\hE{\hat{E}}\def\hT{\hat{T}}\def\hR{\hat{R}}\def\hO{\hat{\Omega}}

170:  \def\iE{(E^{-1})}\def\nno{\nonumber}

171:  \def\del{\partial}

172:  \def\bfk{{\bf k}}\def\bfK{{\bf K}}

173:  \def\bt{\bullet}

174:  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

175: %%

176: %%%%%%%%%

177:  \def\be{\begin{equation}}\def\ee{\end{equation}}

178:  \def\bea{\begin{eqnarray}}\def\eea{\end{eqnarray}}

179:  \def\ba{\begin{array}}\def\ea{\end{array}}

180: \def\bc{\begin{cases}}\def\ec{\end{cases}}

181:  \let\la=\label

182:  %\let\fr=\frac

183:   \def\nn{\nonumber}

184:  \def\ft#1#2{{\textstyle{{\scriptstyle #1}\over {\scriptstyle #2}}}}

185:  \def\fft#1#2{{#1 \over #2}}

186:  %\def\sst#1{{\scriptscriptstyle #1}}

187:  %\def\oneone{\rlap 1\mkern4mu{\rm l}}

188:  \newcommand{\eq}[1]{(\ref{#1})}

189:  \def\eqs#1#2{(\ref{#1}-\ref{#2})}

190:  \def\det{{\rm det\,}}

191:  \def\tr{{\rm tr}}\def\Tr{{\rm Tr}}

192:  \def\Str{{\rm Str}}

193:  \def\One{1 \!\! 1}

194:  \def\hF{\hat{F}}

195: %%%%%%%%%%%

196: \newcommand{\hoch}[1]{$\, ^{#1}$}

197: \newcommand{\kings}{\it\small Department of Mathematics, King's College,

198: London, UK}

199: \newcommand{\upp}{\it\small Department of Theoretical Physics

200: Uppsala University, Sweden}

201: 

202: 

203: 

204: \begin{document}

205: 

206:  \thispagestyle{empty}

207: 

208:  \hfill{KCL-MTH-03-13}

209: 

210:  \hfill{ITP-UU-03-11}

211: 

212:   \hfill{\today}

213: 

214:  \vspace{20pt}

215: 

216:  \begin{center}

217:  {\Large{\bf The deformed M2-brane}}

218:  \vspace{30pt}

219: 

220:  {P.S. Howe\hoch{1}, S.F. Kerstan\hoch{1}, U. Lindstr\"om\hoch{2}

221:  and D. Tsimpis\hoch{1}}

222: 

223:  \vspace{60pt}

224: 

225:  \end{center}

226: 

227: \begin{itemize}

228: \item [$^1$] \kings \item [$^2$] \upp

229: \end{itemize}

230: 

231:  {\bf Abstract}

232: 

233: The superembedding formalism is used to study correction terms to

234: the dynamics of the M2 brane in a flat background.  This is done by deforming the standard 

235: embedding constraint. It is shown

236: rigorously that the first such correction occurs at dimension

237: four. Cohomological techniques are used to determine this

238: correction explicitly. The action is derived to quadratic order in

239: fermions, and the modified $\k$-symmetry transformations are

240: given.

241: 

242: 

243: 

244: 

245: 

246:  {\vfill\leftline{}\vfill \vskip  10pt

247: 

248:  \baselineskip=15pt \pagebreak \setcounter{page}{1}

249: 

250: 

251: 

252: 

253: 

254: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

255: %%%%%%%%%%%%%%%%%%%%

256: %

257: %   Document

258: %

259: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

260: %%%%%%%%%%%%%%%%%%%%%

261: 

262: 

263: \section{Introduction}

264: 

265: An interesting problem in string/M theory is to consider

266: higher-order  corrections to the effective dynamics of various

267: branes. A number of results have been obtained in the purely

268: bosonic sector, particularly for D-branes,

269: \cite{Andreev:1988cb,Bachas:1999um,Fotopoulos:2001pt,

270: Wyllard:2000qe,Bilal:2001hb,Wyllard:2001ye,Koerber:2002zb}.\footnote{For

271: some earlier work on higher derivative corrections see

272: \cite{Lindstrom:1987ps,Lindstrom:1988fr}.} Some of these 

273: have been supersymmetrised, for example the $\del^4 F^4$ and other

274: terms in D-brane actions \cite{Collinucci:2002gd,Drummond:2003ex}, but

275: it has so far proven difficult to obtain Lagrangians or equations of motion

276: with fully-fledged kappa-symmetry. Some kappa-symmetric results have been obtained

277: for higher-derivative terms for superparticles

278: \cite{Ivanov:1990cb,Gauntlett:1991dw,Ivanov:1991ub} and

279: superstrings \cite{Curtright:1987mr}, for coincident D0-branes in

280: \cite{Sorokin:2001av}, and for branes in lower-dimensional

281: spacetimes in \cite{Howe:2001wc,Drummond:2002kg}. The last two

282: papers made use of the superembedding formalism \cite{stvz,s}

283: which offers a systematic way of incorporating kappa-symmetry

284: whilst maintaining manifest target space supersymmetry. In this

285: paper we shall examine the leading correction terms for the

286: M2-brane using this formalism. We shall derive the action up to

287: terms quadratic in the fermions as well as the modified

288: kappa-symmetry transformations.

289: 

290: 

291: The two examples studied in \cite{Howe:2001wc,Drummond:2002kg},

292: namely a membrane in four dimensions and a set of coincident

293: membranes in three dimensions, were relatively easy to analyse

294: because the worldvolume multiplets in both cases are off-shell.

295: This means that the standard superembedding constraints do not

296: need to be altered - one only needs to find higher-order

297: Lagrangians which can be constructed using standard methods.

298: However, for the 1/2 BPS branes in string theory and M-theory the

299: worldvolume multiplets are maximally supersymmetric matter or

300: gauge multiplets and it is not known how to find auxiliary fields

301: for these or even if they exist. Indeed, the standard

302: superembedding constraint leads directly to the equations of

303: motion of the brane in most cases \cite{Howe:1996mx,Howe:1996yn}.

304: This means that we need to find new techniques to discuss

305: higher-derivative interactions for these branes. The purpose of

306: the present paper is to show how this can be carried out in a

307: perturbative fashion for the case of the M2-brane. The basic idea

308: is to deform the superembedding constraint in such a way that the

309: field content of the brane theory is unchanged. The torsion

310: identities, which we describe in detail below, then impose

311: consistency requirements on the deformation which can be analysed

312: order by order in a parameter $\ell$ which has the dimensions of

313: length and which could be the Planck length or the string length

314: $\sqrt{\a'}$. In fact, what we end up with is a cohomology problem

315: which is similar in spirit to the cohomological problems

316: encountered in the analysis of higher-order corrections to

317: maximally supersymmetric field theories including supergravity

318: \cite{cnta,cntb,cntc,cntd,Howe:2003cy}. However, the cohomology

319: problem considered here is manifestly covariant with respect to

320: target space supersymmetry and local worldvolume supersymmetry

321: ($\k$-symmetry); if we were to work in the static gauge the

322: formalism would automatically have ``$N=2$'' worldvolume

323: supersymmetry (actually $d=3, N=16$) non-linearly realised.

324: 

325: We shall limit ourselves to the case of  a flat target superspace

326: in this paper for simplicity, although there will also be

327: corrections involving the target space curvature.  We show

328: rigorously that supersymmetry does not allow any deformations with

329: three or fewer powers of $\ell$, a result which is to be expected

330: in view of the results that have been obtained for D-branes. We

331: then analyse in some detail the lowest-order non-trivial

332: deformation at $\ell^4$. This turns out to be rather complicated

333: and involves up to seven powers in fields. We show that there is a

334: unique deformation with at most three fields. We believe that it

335: is unlikely that there are other independent deformations, because

336: of the number of constraints that have to be satisfied,  but we

337: have not completed the proof of this assertion. Knowledge of the

338: cubic terms in the deformation is sufficient to determine the

339: modified Lagrangian to quadratic order in fermions, and thus to

340: determine the full purely bosonic part of the action at order

341: $\ell^4$. The result is, as expected, a Lagrangian which is

342: quartic in the extrinsic curvature.

343: 

344: The organisation of the paper is as follows.  In the next section

345: we briefly review the superembedding formalism and the relation of

346: worldvolume supersymmetry  to $\k$-symmetry. In section three we

347: solve the torsion identities for the membrane to first order in

348: the deformation field $\psi$ and summarise the lowest order

349: results which give the standard dynamics. In section four we

350: introduce the relevant spinorial cohomology groups for deformed

351: superembeddings. In section five we discuss the constraints on the

352: deformation and solve the relevant cohomology problem up to order

353: $\ell^4$ in section six. In section seven we construct the action,

354: up to terms quadratic in the fermions, and in section eight we

355: discuss the modified $\k$-symmetry transformations.

356: 

357: \section{Superembeddings}

358: 

359: The superembedding formalism was pioneered in the context of

360: superparticles in three and four dimensions by Sorokin, Tkach,

361: Volkov and Zheltukhin \cite{stvz}, and has been applied by these

362: and other authors to various other branes; for a review see

363: \cite{s}.\footnote{For an earlier attempt to use source and target

364: superspaces, see, e.g., \cite{Gates:1985vk}.} In

365: \cite{Howe:1996mx} it was shown that the formalism can be applied

366: to arbitrary branes including those with various types of

367: worldvolume gauge fields, and it was then used to construct the

368: full non-linear equations of motion of the M-theory 5-brane in an

369: arbitrary supergravity background \cite{Howe:1996yn}.

370: 

371:  We consider a superembedding $f:M\rightarrow \unM$, where $M$ is

372:  the worldvolume of the brane and $\unM$ is the target space.

373:  Our index conventions are as follows: coordinate

374:  indices are taken from the middle of the alphabet with capitals

375:  for all, Latin for bosonic and Greek for fermionic, $M=(m,\m)$,

376:  tangent space indices are taken in a similar fashion

377:  from the beginning of the

378:  alphabet so that $A=(a,\a)$. We denote the coordinates of $M$

379:  by $z^M=(x^m,\th^{\m})$.

380:  The distinguished tangent space

381:  bases are related to coordinate bases by means of the

382:  supervielbein, $E_M{}^A$, and its inverse $E_A{}^M$.

383:  We use exactly the same notation

384:  for the target space but with all of the indices underlined.

385:  Indices for the normal bundle are denoted by primes, so that

386:  $A'=(a',\a')$. We shall occasionally group together tangent and

387:  normal indices and denote them by a bar, $\bar a=(a,a')$ etc.

388: 

389: 

390:  The embedding matrix is the derivative of $f$ referred to the

391:  preferred tangent frames, thus

392: 

393:   \be

394:   E_A{}^{\unA}:= E_A{}^M\del_M z^{\unM} E_{\unM}{}^{\unA}

395:   \la{2.1}

396:   \ee

397: 

398: In addition, we can specify a basis $\{E_{A'}\}$ for the normal

399: bundle in terms of the target space basis by means of the normal

400: matrix $E_{A'}{}^{\unA}$.

401: 

402:  The basic embedding condition is

403: 

404:   \be

405:   E_{\a}{}^{\una}=0

406:   \la{2.1.1}

407:   \ee

408: 

409: Geometrically this states that the odd tangent space of the brane

410: is a subspace of the odd tangent space of the target superspace at

411: any point on the brane. To see the content of this constraint we

412: can consider a linearised embedding in a flat target space in the

413: static gauge. This gauge is specified by identifying the

414: coordinates of the brane with a subset of the coordinates of the

415: worldvolume, so that

416: 

417:  \bea

418:  x^{\una}&=&\bc x^a  \\ x^{a'}(x,\th)\ec  \\

419:  \th^{\ua}&=&\bc  \th^{\a}\\ \th^{\a'}(x,\th) \ec

420:  \la{2.2}

421:  \eea

422: 

423: Since

424: 

425:  \be

426:  E^{\una}=d x^{\una} -{i\over2}d\th^{\ua} (\C^{\una})_{\ua\ub}

427:  \th^{\ub}

428:  \la{2.3}

429:  \ee

430: 

431: in flat space, it is easy to see, to first order in the transverse

432: fields, that the embedding condition implies that

433: 

434:  \be

435:  D_{\a} X^{a'}= i(\C^{a'})_{\a\b'} \th^{\b'}

436:  \la{2.4}

437:  \ee

438: 

439: where

440: 

441:  \be

442:  X^{a'}=x^{a'} + {i\over2} \th^{\a} (\C^{a'})_{\a\b'} \th^{\b'}

443:  \la{2.5}

444:  \ee

445: 

446: 

447: {}From this equation the nature of the worldvolume multiplet

448: specified by the embedding condition can be determined. Depending

449: on the dimensions involved, this multiplet can be one of three

450: types: (i) on-shell, i.e. the multiplet contains only physical

451: fields and these fields satisfy equations of motion, (ii)

452: Lagrangian off-shell, meaning that the multiplet also contains

453: auxiliary fields (in most cases), and that the equations of motion

454: of the physical fields are not satisfied, although they can be

455: derived from a superfield action, and (iii) underconstrained,

456: which means that further constraints are required to obtain

457: multiplets of one of the first two types. For thirty-two target

458: space supersymmetries, the multiplets are always of type (i) or of

459: type (iii), whereas for sixteen or fewer supersymmetries all three

460: types of multiplet can occur, with type (iii) arising typically

461: for cases of low even codimension.

462: 

463: In this paper we shall be concerned with the membrane in $D=11$

464: superspace, i.e. the M2-brane. This has been discussed as a

465: superembedding in \cite{Bandos:1995zw}. The worldvolume is

466: $d=3,N=8$ superspace, and the worldvolume multiplet is an on-shell

467: scalar multiplet (type (i)). This can be seen from \eq{2.5}. The

468: leading component of the superfield $X^{a'}$ describes eight

469: scalar fields on the worldvolume, while the leading component of

470: $\th^{\a'}$ describes eight spin one-half fields. Equation

471: \eq{2.5} implies that there are no further independent component

472: fields and that the scalars and spinors satisfy the usual

473: equations of motion for free massless fields.

474: 

475: 

476: In order to determine the consequences of the superembedding

477: condition in the non-linear case, and to find the induced

478: supergeometry on the brane, one uses the torsion equation which is

479: the pull-back  of the equation defining the target space torsion

480: two-form. This reads

481: 

482:  \be

483:  2\nab_{[A} E_{B]}{}^{\unC} + T_{AB}{}^C E_C{}^{\unC} =

484:  (-1)^{A(B+\unB)}E_B{}^{\unB}

485:  E_A{}^{\unA} T_{\unA\unB}{}^{\unC}

486:  \la{2.6}

487:  \ee

488: 

489: In this equation the covariant derivative acts on both worldvolume

490: and target space tensor indices. The latter are taken care of by

491: the pull-back of the target space connection, while the

492: worldvolume connection can be chosen in a variety of different

493: ways. One also has to parametrise the embedding matrix. Having

494: done all this, one can work through the torsion equation starting

495: at dimension zero. In this way the consequences of the embedding

496: condition \eq{2.1.1} can be worked out in a systematic and

497: covariant fashion.

498: 

499: The equations of the component or Green-Schwarz formalism can be

500: obtained by taking the leading ($\th=0$) components of the various

501: equations that describe the brane multiplet. A key feature of the

502: formalism is that these component equations are guaranteed to be

503: $\k$-symmetric because $\k$-symmetry can be identified with the

504: leading term in a  worldvolume local supersymmetry transformation.

505: We recall briefly how this works \cite{stvz}. Let $v^M$ be a

506: worldvolume vector field generating an infinitesimal

507: diffeomorphism. If we write the superembedding in local

508: coordinates as $f^{\unM}=z^{\unM}(X)$, then the effect of such a

509: transformation on $\underline{z}(z)$ is

510: 

511:  \be

512:  \d z^{\unM}= v^M\del_M z^{\unM}

513:  \la{2.7}

514:  \ee

515: 

516: If we express this in the preferred bases we find

517: 

518:  \be

519:  \d z^{\unA}:=\d z^{\unM}E_{\unM}{}^{\unA}=v^A E_A{}^{\unA}

520:  \la{2.8}

521:  \ee

522: 

523: Now if we take an odd worldvolume diffeomorphism, $v^a=0$, and use

524: the embedding condition \eq{2.1.1} we get

525: 

526:  \bea

527:  \d z^{\una} &=& 0\\

528:  \d z^{\ua} &=& v^{\a} E_{\a}{}^{\ua}

529:  \la{2.9}

530:  \eea

531: 

532: This can be brought to the more usual $\k$-symmetric form if we

533: define

534: 

535:  \be

536:  \k^{\ua}:= v^{\a} E_{\a}{}^{\ua}

537:  \la{2.10}

538:  \ee

539: 

540: and note that it satisfies

541: 

542:  \be

543:  \k^{\ua}=\k^{\ub}

544:  P_{\ub}{}^{\ua}:={1\over2}\k^{\ub}(1+\C)_{\ub}{}^{\ua}

545:  \la{2.11}

546:  \ee

547: 

548: where $P$ is the projection operator onto the worldvolume subspace

549: of the odd tangent space of the target superspace. We can always

550: write $P=1/2(1+\C)$, and so $\C$ is computable in terms of the

551: embedding matrix. Substituting this into \eq{2.9} we recover the

552: normal form of $\k$-symmetry transformations. (Strictly, we should

553: evaluate this equation at $\th=0$ to get the correct component

554: form.) The explicit form of $P$ is

555: 

556:  \be

557:  P_{\ua}{}^{\ub}=(E^{-1})_{\ua}{}^{\c} E_{\c}{}^{\ub}

558:  \la{2.12}

559:  \ee

560: 

561: where the inverse is taken in the fermionic tangent space (of

562: $\unM$).

563: 

564: For the case of the supermembrane in $D=11$, this procedure yields

565: the equations of motion found in \cite{Bergshoeff:1987cm}. To find

566: higher derivative corrections to these equations it follows that

567: we shall need to amend the basic embedding condition \eq{2.1.1}.

568: We note that a simple consequence of this is that the

569: $\k$-symmetry transformations will no longer have the standard

570: characteristic form for which $\d z^{\una}=0$.

571: 

572: 

573: 

574: \section{The torsion identities}

575: 

576: 

577: In this section we shall study the torsion identities in the case

578: that the embedding condition is relaxed. We shall take the target

579: space to be flat so that the only non-vanishing component of the

580: target space torsion is

581: 

582: 

583: \be T_{\ua\ub}{}^{\una}=-i\left(\C^{\una}\right)_{\ua\ub},

584: \la{3.1} \ee

585: 

586: We can parametrise the embedding matrix as follows:

587: 

588: 

589: 

590: \bea E_{\a}{}^{\una}=& \psi_\a{}^{a'} u_{a'}{}^{\una}\qquad

591: &E_{\a}{}^{\ua}=u_{\a}{}^{\ua}

592:  + h_{\a}{}^{\a'} u_{a'}{}^{\ua}\la{3.2}\\

593: E_a{}^{\una}=&u_a{}^{\una},\qquad &E_a{}^{\ua} =\Lambda_a{}^{\a '}

594: u_{\a'}{}^{\ua} \la{3.3} \eea

595: 

596: while the normal matrix can be chosen to have the form

597: 

598: \bea

599: E_{\a'}{}^{\una}=& 0,\qquad &E_{\a'}{}^{\ua}=u_{\a'}{}^{\ua}\la{3.4}\\

600: E_{a'}{}^{\una}=&u_{a'}{}^{\una}, \qquad &E_{a'}{}^{\ua}=0

601: \la{3.5} \eea

602: 

603: Here, the $32\xz 32$ matrix $u_{\overline\a}{}^{\

604: \ua}:=(u_{\a}{}^{\ua},\,u_{\a'}{}^{\ua})$

605:  is an element of $Spin(1,10)$

606: while the matrix $u_{\overline a}{}^{\

607: \una}:=(u_a{}^{\una},\,u_{a'}{}^{\una})$ is the corresponding

608: element of $SO(1,10)$.  The dimensions, in units of mass, of the

609: various components of the embedding matrix are given by

610: 

611: \be [E_{\a}{}^{\ua}]=[E_{a}{}^{\una}]=0,\qquad

612: [E_{\a}{}^{\una}]=-[E_{a}{}^{\ua}]=-\ft{1}{2} \ee

613: 

614: and similarly for the normal matrix. We can always bring these

615: matrices into the above forms by a suitable choice of the bases of

616: the even and odd tangent spaces on the worldvolume. We can now

617: plug this form of the embedding matrix into the torsion identity

618: 

619: 

620: \be \nabla_A E_B{}^{\unC}-(-)^{AB}\nabla_B

621: E_A{}^{\unC}+T_{AB}{}^{C} E_C{}^{\unC}=(-)^{A(B+\unB

622: )}E_B{}^{\unB} E_A{}^{\unA} T_{\unA\unB}{}^{~~\unC} \ee

623: 

624: and work out the consequences. As we noted earlier, the covariant

625: derivative here acts on both target space and worldvolume indices,

626: but since the target space is flat we only need a worldvolume

627: connection. This can be specified either by imposing some

628: constraints on the worldvolume torsion tensor or by choosing a

629: connection which is natural in the superembedding context. We

630: shall use the latter. Given the matrix $u$ we can define the set

631: of one-forms

632: 

633: \be X_{A}:= (\nab_A u) u^{-1} \la{} \ee

634: 

635: If we set

636: 

637: \be X_{A,b}{}^c = X_{A,b'}{}^{c'}=0 \la{} \ee

638: 

639: then we fix connections for the tangent and normal bundles in a

640: standard fashion. Note that

641: 

642: \be X_{A,\bar\b}{}^{\bar\c}={1\over4} (\C^{\bar b\bar

643: c})_{\bar\b}{}^{\bar\c} X_{A,\bar b\bar c} \la{} \ee

644: 

645: The torsion identities yield the following results for a flat

646: target superspace:

647: 

648: dim 0

649: 

650:  \bea

651:  T_{\a\b}{}^c&=& -i(\C^c)_{\a\b}+2\psi_{(\a}{}^{d'}X_{\b),d'}{}^c\\

652:  \nab_{(\a}\psi_{\b)}{}^{c'}&=&-ih_{(\a}{}^{\c'}(\C^{c'})_{\b)\c'}

653:  +\left(-T_{\a\b}{}^{\c}\psi_{\c}{}^{c'}\right)\la{27}

654:  \eea

655: 

656: dim 1/2

657: 

658:  \bea

659:  T_{\a b}{}^c&=&\psi_{\a}{}^{d'}

660:  X_{b,d'}{}^c+i\L_b{}^{\b'}h_{\a}{}^{\c'}(\C^c)_{\b'\c'}\\

661:  X_{\a

662:  b}{}^{c'}&=&\nab_b\psi_{\a}{}^{c'}+i\L_b{}^{\b'}(\C^{c'})_{\a\b'}\\

663:  T_{\a\b}{}^{\c}&=&-2h_{(\a}{}^{\d'}X_{\b),\d'}{}^{\c}\\

664:  2\nab_{(\a}h_{\b)}{}^{\c'}&=&-2X_{(\a,\b)}{}^{\c'}-T_{\a\b}{}^c\L_c{}^{\c'}

665:  -\left(T_{\a\b}{}^\d h_\d{}^{\c'}\right)

666:  \eea

667: 

668: dim 1

669: 

670:  \bea

671:  T_{ab}{}^c&=&i\L_a{}^{\a'}\L_b{}^{\b'}(\C^c)_{\a'\b'}\\

672:  2X_{[ab]}{}^{c'}&=&-T_{ab}{}^{\c}\psi_{\c}{}^{\c'}\\

673:  T_{a\b}{}^\c&=&-\L_a{}^{\d'}X_{\b,\d'}{}^\c

674:  -h_\b{}^{\d'}X_{a,\d'}{}^\c\\

675:  X_{a,\b}{}^{\c'}&=&\nab_\b \L_a{}^{\c'}-\nab_a h_\b{}^{\c'}

676:  \eea

677: 

678: dim 3/2

679: 

680:  \bea

681:  T_{ab}{}^{\c}&=& 2\L_{[a}{}^{\d'}X_{b],\d'}{}^\c  \\

682:  2\nab_{[a}\L_{b]}{}^{\c'}&=&-T_{ab}{}^c \L_c{}^{\c'}-T_{ab}{}^\d

683:  h_\d{}^{\c'}

684:  \eea

685: 

686: The terms in brackets will be irrelevant in what follows because

687: they will turn out to be zero at first order in the deformation

688: parameter as we shall shortly see.

689: 

690: \subsection{Zeroth order}

691: 

692: In the zeroth order theory the standard embedding condition

693: $E_\a{}^{\una}=0$ holds, so that $\psi_{\a}{}^{c'}=0$. If we

694: substitute this into the above equations we find that

695: $h_{\a}{}^{\b'}=0$ while the induced torsion is

696: 

697:  \be

698:  T_{\a\b}{}^c=-i(\C^c)_{\a\b}

699:  \ee

700: 

701: at dimension zero and vanishes at dimension one-half. The

702: dimension one and three-halves components are given by the

703: corresponding  equations above.

704: 

705: For $X$ we find, at dimension one-half,

706: 

707:  \bea

708:  X_{\a,b}{}^{c'}&=&i\L_b{}^{\b'}(\C^{c'})_{\a\b'}\\

709:  2X_{(\a\b)}{}^{\c'}&=&i(\C^c)_{\a\b}\L_c{}^{\c'}

710:  \eea

711: 

712: It is easy to show that these equations imply that

713: 

714:  \be

715:  (\C^a)_{\a'\b'}\L_a{}^{\b'}=0

716:  \la{deq}

717:  \ee

718: 

719: In the linearised theory $\L_a{}^{\b'}\sim \del_a \th^{\b'}$, so

720: that we can identify \eq{deq} as the equation of motion of the

721: fermion field in the worldvolume multiplet. At dimension one we

722: have

723: 

724:  \bea

725:  X_{[ab]}{}^{c'}&=&0\\

726:  X_{a,\b}{}^{\c'}&=&\nab_\b \L_a{}^{\c'}

727:  \eea

728: 

729: Using the fermion equation of motion we find the scalar equation

730: of motion

731: 

732:  \be

733:  \h^{ab}X_{a,b}{}^{c'}=0

734:  \ee

735: 

736: Indeed, in the linearised theory, $X_{ab}{}^{c'}\sim \del_a\del_b

737: X^{c'}$, so we get the massless Klein-Gordon equation. Finally, a

738: short calculation gives the supersymmetry variation of

739: $X_{ab}{}^{c'}$:

740: 

741:  \be

742:  \nabla_{\a i}X_{b,c}{}^{d'}=-i(\s^{d'}\nabla_{(b}\L_{c)})_{\a i}

743: +{1\over 2}(\L_b\c^e\otimes

744: \tilde{\s}^{f'd'}\L_c)(\s_{f'}\L_e)_{\a i}

745:  \ee

746: 

747: \subsection{First order}

748: 

749: A first order deformation of the theory will involve the presence

750: of a non-vanishing $\psi$ field which we can take to be of the

751: form of a dimensionful parameter $\b$, say, multiplied by a

752: function of the physical fields $\L_a{}^{\b'}$ and $X_{ab}{}^{c'}$

753: as well as derivatives. We shall write this schematically as

754: 

755:  \be

756:  \psi=\b f(\L,X,\del)

757:  \ee

758: 

759: Furthermore, $\psi$ is subject to the constraint

760: 

761:  \be

762:  \nab_{(\a}\psi_{\b)}{}^{c'}=-ih_{(\a}{}^{\c'}(\C^{c'})_{\b)\c'}

763:  \la{47}

764:  \ee

765: 

766: Note that, since $\psi,h$ and $T_{\a\b}{}^\c$ are all of order

767: $\b$ we are allowed to drop the $T_{\a\b}{}^\c \psi_{\c}{}^{c'}$

768: term from \eq{27}. The problem is then to analyse equation

769: \eq{47}. This can be done systematically in powers of the length

770: parameter $\ell$. Note that, given an explicit expression for

771: $\psi$ in terms of the physical fields, equation \eq{47} allows us

772: to solve for $h$.

773: 

774: \section{Spinorial cohomology for branes}

775: 

776: The notion of spinorial cohomology \cite{cnta,cntc} is useful for studying both the

777: space of physical fields in certain supersymmetric theories and

778: also for studying deformations of the equations of motion. In the

779: simplest case one studies spinorial $p$-forms, i.e. totally

780: symmetric $p$-spinors which are $\c$-traceless, together with a

781: differential operator which is obtained by acting with $D_\a$

782: followed by symmetrisation and removal of the $\c$-trace. This

783: defines a cohomology which is isomorphic to pure spinor cohomology

784: \cite{Berkovits:2000fe,Berkovits:2001rb}. The relevance of the latter to theories

785: in ten and eleven dimensions is explained by the fact that the

786: equations of motion can be interpreted in terms of pure spinor

787: integrability \cite{Howe:mf,Howe:1991bx}. The formalism has been

788: applied to $D=10$ Yang-Mills theory and supergravities in ten and

789: eleven dimensions \cite{cnta,cntb,cntc,cntd,Howe:2003cy}.

790: Moreover, one can also consider vector-valued spinorial cohomology

791: \cite{cntc}. For branes, the appropriate cohomology is a variant

792: on the latter.

793: 

794: Since we are interested in first-order deformations, it is

795: sufficient to work on a worldvolume whose induced supergeometry is

796: given by the zeroth order theory. We then consider the spaces

797: $\O^p_{B'}$ and $\O^p_{F'}$. The objects in these spaces are

798: spinorial $p$-forms on $M$ which take their values either in the

799: even normal bundle $B'$ or the odd normal bundle $F'$. There is a

800: natural map $\C:\O^{p-1}_{F'}\rightarrow \O^p_{B'}$ given by

801: 

802:  \be

803:  h_{\a_1\ldots \a_{p-1}}{}^{\c'}\mapsto -i

804:  h_{(\a_1\ldots \a_{p-1}}{}^{\c'}(\C^{c'})_{\a_b)\c'}

805:  \la{equiv}

806:  \ee

807: 

808: We can therefore form the quotient space

809: $\hat\O^p_{B'}:=\O^p_{B'}/\C(\O^{p-1}_{F'})$ and construct a

810: derivative $d_s:\hat\O^p_{B'}\rightarrow\hat\O^{p+1}_{B'}$ which

811: squares to zero. This derivative is defined by acting with the

812: spinorial covariant derivative $\nab_\a$ and symmetrising modulo

813: equivalences of the form of equation \eq{equiv}. In other words,

814: if $h\in\O^p_{B'}$ represents an equivalence class

815: $[h]\in\hat\O^p_{B'}$, then $d_s[h]=[\nab h]\in\hat\O^{p+1}_{B'}$

816: where

817: 

818:  \be

819:  (\nab h)_{\a_1\ldots \a_{p+1}}{}^{c'}:=\nab_{(\a_1}h_{\a_2\ldots

820:  \a_{p+1})}{}^{c'}

821:  \ee

822: 

823: It is easy to see that this definition is independent of the

824: choice of representative $h$. To see that $d_s^2=0$ we observe

825: that

826: 

827:  \be

828:  \nab_{(\a_1}\nab_{\a_2}h_{\a_3\ldots\a_{p+2})}{}^{c'}=

829:  -{1\over2} R_{(\a_1\a_2,\a_3}{}^{\c} h_{|\c|\a_4\ldots

830:  \a_{p+2})}{}^{c'}+{1\over2}

831:  R_{(\a_1\a_2,d'}{}^{c'}h_{\a_3\ldots\a_{p+2})}{}^{d'}

832:  \la{50}

833:  \ee

834: 

835: Now, from the first Bianchi identity, we have

836: 

837:  \bea

838:  R_{(\a_1\a_2\a_3)}{}^\c&=&\nab_{(\a_1}T_{\a_2\a_3)}{}^\c +

839:  T_{(\a_1\a_2}{}^B T_{B \a_3)}{}^{\c}\nn\\

840:  &=&-i(\C^b)_{(\a_1\a_2}T_{b\a_3)}{}^{\c}

841:  \eea

842: 

843: so that the first term on the RHS of \eq{50} is of the form

844: 

845:  \be

846:  -i(\C^b)_{(\a_1\a_2} k_{b\a_3\ldots \a_{p+2})}{}^{c'}

847:  \ee

848: 

849: for some $k$. However, this can be written as

850: 

851:  \be

852:  -ih_{(\a_1\ldots \a_{p+1}}{}^{\c'} (\C^{c'})_{\a_{p+2})\c'}

853:  \ee

854: 

855: where

856: 

857:  \be

858:  h_{\a_1\ldots

859:  \a_{p+1}}{}^{\c'}=(\C^{bc'})_{(\a_1}{}^{\c'}k_{b\a_2\ldots\a_{p+2})c'}

860:  \ee

861: 

862: 

863: and so this term maps to zero in the quotient space. The second

864: term on the RHS of \eq{50} can be evaluated with the aid of the

865: Gauss-Codazzi equation which can, in turn, be derived from the

866: definition of $X_A$. One finds that

867: 

868:  \be

869:  R_{\a\b c'}{}^{d'}=2X_{(\a c'}{}^e X_{\b) e}{}^{d'}

870:  \ee

871: 

872: Since $X_{\a b}{}^{c'}=i\L_b{}^{\b'}(\C^{c'})_{\b'\a}$ we have

873: 

874:  \be

875:  R_{\a\b c'}{}^{d'}=2\L^{e\c'}\L_e{}^{\d'}

876:  (\C^{c'})_{\c'(\a}(\C^{d'})_{\b)\d'}

877:  \ee

878: 

879: from which it is easy to see that the second term also gives zero

880: in the quotient space. This shows that $d_s^2=0$ so that we can

881: define the cohomology groups $H^p_{B'}:=\ker

882: d_s\cap\hat\O^p_{B'}/{\rm im} d_s\hat\O^{p-1}_{B'}$. We claim that

883: a first-order deformation of the dynamics of the membrane is given

884: by an element of $H^1_{B'}(phys)$, where the notation indicates

885: that the coefficients should be given by fields constructed from

886: the physical fields.

887: 

888: To see that this is the appropriate group we need to consider

889: redefinitions. We shall find the effect of a field redefinition of

890: the embedding coordinates

891: %

892: \begin{equation}

893: z^{\unM}\rightarrow z^{\unM}+(\d z)^{\unM} \label{trnsf}

894: \end{equation}

895: %

896: on the field $\psi_{\a}{}^{c'}$. The transformation (\ref{trnsf})

897: of the embedding coordinates with the background geometry fixed

898: may be viewed equivalently as a diffeomorphism of the background

899: geometry with the embedding coordinates fixed. Taking the latter

900: point of view, we find that the target-space vielbein transforms

901: as

902: $$

903: \d E_{\unM}{}^{\unA}=(\d z)^{\unN} T_{\unN \unM}{}^{\unA}

904: +\nab_{\unM}(\d z)^{\unA},

905: $$

906: up to a Lorentz transformation on the flat index. The

907: transformation of the embedding matrix reads

908: %

909: \begin{align}

910: \d E_A{}^{\unA}&=E_A{}^M\partial_M Z^{\unM} \d E_{\unM}{}^{\unA}\nn\\

911: &=\nab_{A}(\d z)^{\unA}+E_A{}^{\unC}(\d z)^{\unB}T_{\unB

912: \unC}{}^{\unA}\nn

913: \end{align}

914: %

915: Setting $A=\a,~ \unA=\una$ in the above and expressing

916: $E_{\a}{}^{\una}$ in terms of  $\psi_\a{}^{c'}$, we obtain

917: 

918:  \be

919:  \d \psi_\a{}^{c'}=\nab_{\a}(\d z)^{c'}+(\d z)^b X_{\a b}{}^{c'} -i

920:  (\d z)^{\c'}(\C^{c'})_{\c' \a}

921:  \ee

922: 

923: where $(\d z)^{\overline{c}}:=(\d

924: z)^{\unc}u_{\unc}{}^{\,\overline{c}}$, etc. Since $\d z$ is of

925: order $\b$ we can use the zeroth order expression for $X$ and so

926: we obtain

927: 

928:  \be

929:  \d \psi_\a{}^{c'}=\nab_{\a}(\d z)^{c'} -i

930:  h^{\c'}(\C^{c'})_{\c' \a}

931:  \la{redef}

932:  \ee

933: 

934: where

935: 

936:  \be

937:  h^{\c'}:=(\d z)^{\c'}- (\d z)^b \L_b{}^{\c'}

938:  \ee

939: 

940: Together with the fact that $\psi$ satisfies the constraint

941: \eq{47}, this proves the contention that the first-order

942: deformation is indeed given by an element of $H^1_{B'}(phys)$.

943: 

944: We observe that the zeroth order group $H^0_{B'}$  can be

945: interpreted as the space of physical fields, since a deformation

946: of the form of \eq{redef} such that the left-hand side vanishes

947: will leave the basic embedding constraint unchanged. Indeed, the

948: linearised embedding equation \eq{2.1.1} can also be viewed as

949: defining an element of this group in flat space.

950: 

951: The discussion given here is pertinent to the M2-brane but can be

952: generalised to other branes. This may require some technical

953: modifications since the field $h_\a{}^{\c'}$ does not vanish at

954: zeroth order in the presence of worldvolume gauge fields; in

955: particular, this is the case for D-branes and the M5-brane.

956: 

957: 

958: 

959: 

960: \section{Constraints on $\psi$}

961: 

962: We now derive the constraints on $\psi_\a{}^{a'}$ implied by the

963: dimension--0 torsion identity at linear order in $\b$,

964: %

965: \begin{equation}

966: \nabla_{(\alpha}\psi_{\beta

967: )}^{~~c'}=-ih_{(\a}{}^{\c'}(\C^{c'})_{\b)\c'} \label{constraint}

968: \end{equation}

969: %

970: Note that in the above we have taken into account that

971: $T_{\alpha\beta}^{~~\c}$ is of order $\b$ as implied by the

972: dimension--1/2 torsion identity.

973: 

974: The field $\psi_{\a i}{}^{a'}$ (switching to two-step notation)

975: transforms under the $(1)\times((0001)\otimes(1000))$

976: representation of $Spin(1,2)\times Spin(8)$. (See the appendix for

977: representation-theoretic conventions.) Decomposing $\psi$ into

978: irreducible representations, we find

979: 

980: $$

981: \psi_{\a i}{}^{a'}\sim (1)\times (1001)\oplus (1)\times (0010)

982: $$

983: Explicitly:

984: %

985: \begin{alignat}{3}

986: \psi_{\a i}{}^{a'}&=\S_{\a i}{}^{a'}&\qquad (1)\times (1001)\nn\\

987: &+(\s^{a'})_{ij'}\S_\a^{j'} &(1)\times (0010) \label{psxp}

988: \end{alignat}

989: %

990: {}From \eq{redef} it follows that the trace part in the

991: decomposition of $\psi_{\a i}^{a'}$ above can be eliminated using

992: field redefinitions of the embedding coordinates.  We shall

993: therefore set $\S_\a^{i'}=0$.

994: 

995: Similarly, the field $h_{\a}{}^{\b'} \to h_{\a i}{}^{\b j'}$ (in

996: two-step notation) transforms under the

997: $((1)\otimes(1))\times((0001)\otimes(0010))$ representation of

998: $Spin(1,2)\times Spin(8)$. Decomposing in irreducible

999: representations, we have

1000: 

1001: $$

1002: h_{\a i}{}^{\b j'} \sim (0)\times (1000)\oplus (0)\times

1003: (0011)\oplus (2)\times (1000)\oplus (2)\times (0011)

1004: $$

1005: Explicitly

1006: %

1007: \begin{alignat}{3}

1008: h_{\a i}{}^{\b j'}&=\d_\a{}^\b(\s^{a'})_i{}^{j'}h_{a'} &\qquad (0)\times (1000)\nn\\

1009: &+{1\over 6}\d_\a{}^\b(\s^{a'b'c'})_i{}^{j'}h_{a'b'c'} &(0)\times (0011)\nn\\

1010: &+(\c^a)_\a{}^\b(\s^{a'})_i{}^{j'}h_{aa'} &(2)\times (1000)\nn\\

1011: &+{1\over 6}(\c^a)_\a{}^\b(\s^{a'b'c'})_i{}^{j'}h_{aa'b'c'}

1012: &(2)\times (0011) \label{hexp}

1013: \end{alignat}

1014: %

1015: 

1016: In order to analyse (\ref{constraint}) we need to decompose

1017: $\nabla_{\a i} \S_{\b j}{}^{a'}$ into irreducible representations

1018: of $Spin(1,2)\times Spin(8)$,  (remember we have set $\S_\a^{i'}$

1019: to zero). This field transforms under the

1020: $((1)\otimes(1))\times((0001)\otimes(1001))$ representation.

1021: Decomposing into irreducible representations, we have

1022: %

1023: \begin{alignat}{2}

1024: \nabla_{\a i} \S_{\b j}{}^{a'} &\sim (0)\times (1000)\oplus

1025: (0)\times (0011)\oplus

1026: (0)\times (1100)\nn\\

1027: &\oplus (2)\times (1000)\oplus (2)\times (0011)\oplus (2)\times

1028: (1002)\nn

1029: \end{alignat}

1030: %

1031: Explicitly,

1032: %

1033: \begin{alignat}{3}

1034: \nabla_{\a i} \S_{\b j}^{a'} &=\vare_{\a\b} \d_{ij} Y^{a'}

1035: +{1\over 7} \vare_{\a\b} (\s^{a'b'})_{ij}

1036: Y_{b'}&\qquad (0)\times (1000)\nn\\

1037: &+{1\over 2}\vare_{\a\b}(\s_{b'c'})_{ij}Y^{a'b'c'} +{1\over

1038: 10}\vare_{\a\b}(\s^{a'b'c'd'})_{ij}Y_{b'c'd'}

1039: &(0)\times (0011)\nn\\

1040: &+{1\over 2}\vare_{\a\b}(\s_{b'c'})_{ij}Y^{b'c';a'} & (0)\times (1100)\nn\\

1041: &+{1\over 120}\vare_{\a\b}(\s_{b'c'd'e'})_{ij}Y^{b'c'd'e';a'}

1042:  &(0)\times (1002)\nn\\

1043: &+(\c^a)_{\a\b} \d_{ij} Y_a{}^{a'} +{1\over 7} (\c^a)_{\a\b}

1044: (\s^{a'b'})_{ij}

1045: Y_{ab'}&\qquad (2)\times (1000)\nn\\

1046: &+{1\over 2}(\c^a)_{\a\b}(\s_{b'c'})_{ij}Y_a{}^{a'b'c'} +{1\over

1047: 10}(\c^a)_{\a\b}(\s^{a'b'c'd'})_{ij}Y_{ab'c'd'}

1048: &(2)\times (0011)\nn\\

1049: &+{1\over 2}(\c^a)_{\a\b}(\s_{b'c'})_{ij}Y_a{}^{b'c';a'} & (2)\times (1100)\nn\\

1050: &+{1\over 120}(\c^a)_{\a\b}(\s_{b'c'd'e'})_{ij}Y_a{}^{b'c'd'e';a'}

1051: &(2)\times (1002) \label{yexp}

1052: \end{alignat}

1053: %

1054: The semi-colon notation here denotes ``hook'' representations, e.g

1055: $a'b';c'$ denotes a traceless tensor which is antisymmetric on

1056: $a'b'$, but not on all three indices. Using

1057: (\ref{hexp},\ref{yexp}) in (\ref{constraint}) we can solve for

1058: $h_{\a i}{}^{\b j'}$ in terms of the $Y$s:

1059: %

1060: %

1061: \begin{alignat}{2}

1062: h_{a'}&= -{i\over 7}Y_{a'} \nn\\

1063: h_{a'b'c'}&= -iY_{a'b'c'} \nn\\

1064: h_{aa'}&= iY_{aa'} \nn\\

1065: h_{aa'b'c'}&= -{3i\over 5}Y_{aa'b'c'}. \nn

1066: \end{alignat}

1067: %

1068: %

1069: In addition we find two constraints

1070: %

1071: \begin{alignat}{2}

1072: Y_{a'b';c'}&=0\nn\\

1073: Y_{aa'b'c'd';e'}&=0. \label{cnstr}

1074: \end{alignat}

1075: %

1076: Finally, the fields $Y_{aa'b';c'}, ~ Y_{a'b'c'd';e'}$ drop out of

1077: (\ref{constraint}) and are therefore left undetermined.

1078: 

1079: 

1080: 

1081: 

1082: \section{Solving the spinorial cohomology problem}

1083: 

1084: Let us recapitulate: the deformations of the supersymmetric M2

1085: theory are parametrised by the object $\S^{~~~a'}_{\a i}$ in the

1086: $(1)\times(1001)$ of $Spin(1,2)\times Spin(8)$.  $\S_{\a

1087: i}{}^{a'}$ is not arbitrary, but has to satisfy the two

1088: constraints (\ref{cnstr}). In the language of section 4, such a

1089: $\S^{~~~a'}_{\a i}$ determines an element of the spinorial

1090: cohomology group $H^1_{B'}$. These constraints tell us that the

1091: projection  of the spinor derivative of $\S^{~~~a'}_{\a i}$ onto

1092: the $(0)\times(1100) \oplus (2)\times (1002)$ part should vanish.

1093: This gives us the explicit form of the spinorial derivative $d_s$

1094: for this case:

1095: %

1096: %

1097: \begin{equation}

1098: d_s(\S_{\a i}{}^{a'}) \to (\s^{a_1'a_2'})^{ij}\e^{\a\b} \nabla_{\a

1099: i}\S^{b'}_{\b j}\vert_{(1100)} \oplus (\s^{a_1'\dots

1100: a_5'})^{ij}\c^{\a\b}_a \nabla_{\a i}\S^{b'}_{\b j}\vert_{(1002)},

1101: \label{dione}

1102: \end{equation}

1103: %

1104: where the bars denote the projections onto the indicated

1105: irreducible $Spin(8)$ representations. In addition one has to take

1106: into account the field redefinitions. As explained in the previous

1107: section, these are given by the projection of the spinor

1108: derivative of $(\d z)^{a'}$ onto the $(1)\times(1001)$ part.

1109: Explicitly,

1110: %

1111: \begin{equation}

1112: d_s(\d z)^{a'} \to \nabla_{\a i}(\d z)^{a'}\vert_{(1001)},

1113: \label{ditwo}

1114: \end{equation}

1115: %

1116: Our strategy is to view $\S^{~~~a'}_{\a i} ,~ (\d Z)^{a'}$ as

1117: composite operators, given in terms of the world-volume fields

1118: $X_{ab}{}^{c'}, ~\L_a{}^{\a i'}$. At any given order of $\ell$,

1119: one writes down the most general expressions for $\S^{~~~a'}_{\a

1120: i} ,~ (\d z)^{a'}$ allowed by  dimensional analysis and

1121: representation theory. In  determining the most general

1122: expressions  at a given order in $\ell$ one can assume that the

1123: worldvolume fields obey the lowest-order equations. This is

1124: because  $\S^{a'}_{\a i} ,~(\d z)^{a'}$ are already of order $\b$.

1125: In terms of irreducible representations of $Spin(1,2)\times

1126: Spin(8)$,

1127: %

1128: \begin{align}

1129: \L_{a}{}^{\a i'}&\sim (3)\times (0010)\nn\\

1130: X_{ab}^{c'}&\sim (4)\times(1000)\nn\\

1131: \nabla_{a}\L_{b}^{~\a i'}&\sim (5)\times (0010), ~~~{\rm up ~to

1132: ~terms ~of ~the ~form}

1133: ~\L^3\nn\\

1134: \nabla_a X_{bc}{}^{d'}&\sim (6)\times(1000),~~~{\rm up ~to ~terms

1135: ~of ~the ~form} ~X\L^2\nn

1136: \end{align}

1137: %

1138: 

1139: \subsection{Spinorial cohomology at $\b=\ell^2$}

1140: 

1141: By dimensional analysis $\S^{~~~a'}_{\a i}$ is schematically of

1142: the form\footnote{In this equation and other similar schematic

1143: equations, the symbol $\nab$ indicates the even covariant

1144: derivative $\nab_a$.}

1145: $$

1146: \S\sim \nabla \L\oplus X\L\oplus \L^3.

1147: $$

1148: In terms of irreducible representations of $Spin(1,2)\times

1149: Spin(8)$,

1150: %

1151: \begin{align}

1152: \L^3&\sim\yng(1,1,1)_{(3)}\times

1153: \yng(3)_{(0010)}\oplus\yng(2,1)_{(3)}\times\yng(2,1)_{(0010)}

1154: \oplus\yng(3)_{(3)}\times\yng(1,1,1)_{(0010)}\nn\\

1155: \nabla \L&\sim (5)\times(0010)\nn\\

1156: X\L&\sim (4)\times(1000)\otimes(3)\times(0010),\nn

1157: \end{align}

1158: %

1159: where we have taken into account the fact that $\L$ is

1160: anticommuting. One can verify that $(1)\times(1001)$ is {\it not}

1161: contained in the decomposition of the right-hand sides above and

1162: therefore there is no possible composite field $\S^{~~~a'}_{\a i}$

1163: at this order in $\b$. Consequently, the spinorial cohomology is

1164: trivial and there are no possible supersymmetric deformations of

1165: the theory at this order.

1166: 

1167: 

1168: 

1169: 

1170: 

1171: \subsection{Spinorial cohomology at $\b=\ell^3$ }

1172: 

1173: In this section we shall show that $H^1_{B'}(phys)=0$ at order

1174: $\b=\ell^3$. By dimensional analysis $\S^{~~~a'}_{\a i}$ has the

1175: form

1176: %

1177: \begin{equation}

1178: \S\sim \fbox{$X^2\L$}\oplus\nabla^2 \L\oplus \nabla X\L\oplus

1179: X\nabla\L \oplus X\L^3\oplus  \fbox{$\L^2\nabla\L$}

1180: \oplus\fbox{$\L^5$}. \label{cohthree}

1181: \end{equation}

1182: %

1183: Analysing this in terms of representations of $Spin(1,2)\times

1184: Spin(8)$ one finds that only the boxed terms contain $(1) \times

1185: (1001)$. The three contributing terms are

1186: %

1187: \begin{align*}

1188: \L^5 &\sim \tiny \yng(1,1,1,1,1)_{(3)} \times \yng(5)_{(0010)}

1189: \oplus \yng(5)_{(3)} \times \yng(1,1,1,1,1)_{(0010)} \oplus

1190: \yng(2,1,1,1)_{(3)}\times \yng(4,1)_{(0010)}\\

1191: &\tiny \oplus \fbox{$\yng(4,1)_{(3)}\times

1192: \yng(2,1,1,1)_{(0010)}$}\oplus

1193: \fbox{$\yng(3,1,1)_{(3)}\times\yng(3,1,1)_{(0010)}$} \oplus

1194: \yng(2,2,1)_{(3)}\times\yng(3,2)_{(0010)}\\

1195: &\tiny \oplus\yng(3,2)_{(3)}\times \yng(2,2,1)_{(0010)}\oplus\\

1196: X^2\L&\sim \tiny \fbox{$\yng(2)_{(4)}\times\yng(2)_{(1000)}

1197: \otimes (3)\times(0010)$}\oplus

1198: \fbox{$\yng(1,1)_{(4)} \times\yng(1,1)_{(1000)} \otimes (3)\times(0010)$}\\

1199: \L^2\nabla\L &\sim \tiny

1200: \fbox{$\yng(2)_{(3)}\times\yng(1,1)_{(0010)} \otimes

1201: (5)\times(0010)$} \oplus\yng(1,1)_{(3)}

1202: \times\yng(2)_{(0010)} \otimes \tiny{(5)\times(0010)}\\

1203: \end{align*}

1204: %

1205: Analysing the contents of these products in terms of irreducible

1206: representations, we find that only the boxed expressions contain

1207: the representation of $\S$, i.e. $(1) \times (1001)$. So at order

1208: $\ell^3$ a nontrivial $\S$ must be built from these five

1209: terms and must satisfy the constraints \eq{cnstr}.\\

1210: 

1211: We now identify the field redefinitions which can be used to

1212: remove some of these terms. Dimensional analysis restricts the

1213: field redefinitions $(\d z)^{a'}$ at order $\ell^3$ to be of the

1214: types $X^2$, $X\L\L$ and $\L^4$. Decomposing these into

1215: irreducible representations one finds that that only $X\L\L$

1216: contains a $Spin(8)$ vector which is a $Spin(1,2)$ scalar as

1217: required, and so the field redefinitions are given as:

1218: 

1219: $$\L^2X \sim \left(\yng(1,1)_{(3)} \times\yng(2)_{(0010)}\right)

1220: \otimes (4)\times(1000)$$.

1221: 

1222: This means that we can remove one of the five terms in $\S$ by a

1223: field redefinition.

1224: 

1225: Next, we need to find the terms which contribute to the

1226: constraints \eq{cnstr}. This can be done either by explicitly

1227: calculating the action of $d_s$ on all five terms, or by

1228: dimensional analysis and representation theory. The latter method

1229: reveals that, schematically

1230: 

1231: $$Y \sim X \L \nab \L, \L^2 \nab X \hspace{1mm} \text{ and }

1232: \hspace{1mm} \L^4 X~.$$

1233: 

1234: Figure \ref{order3} illustrates how the field redefinitions

1235: generate contributions to $\S$ and how $\S$ generates

1236: contributions to the constraints.

1237: \leavevmode

1238: \begin{figure}[h] \label {order3}

1239: \begin{center}

1240: \input{figure1.pstex_t}

1241: \end{center}

1242: \caption{Spinorial cohomology at  $\ell^3$. The three rows

1243: contain, schematically, all possible terms in $(\d z)^{a'}$ (field

1244: redefinitions),  $\S^{~~~a'}_{\a i}$ (traceless part of $\psi$),

1245: $Y_{a'b';c'}$ and $Y_{aa'b'c'd';e'}$ (constraints). Multiplicities

1246: are denoted by the numbers in parentheses. In the third row  the

1247: first number in each parenthesis denotes the multiplicity of

1248: $Y_{a'b';c'}$ and the second the multiplicity of

1249: $Y_{aa'b'c'd';e'}$. The arrows from the first to second, and

1250: second to third, rows indicate the action of $d_s$.}

1251: \end{figure}

1252: We can choose to remove the $\L^3 \nab\L$-type term, and are then

1253: left with four terms that must not violate the constraints, i.e.

1254: contribute to the $(0) \times (1100)$ or $(2) \times (1002)$. To

1255: show that any linear combination of these terms, that is any

1256: content of $\psi$ at this order in $\ell$, would violate the

1257: constraints, we need to go into more detail. We break up the

1258: possible terms for $\S$ and those for the constraints $Y$ into

1259: products of irreducible representations (which one can derive by

1260: dimensional analysis and representation theory as before). This is

1261: illustrated in figure \ref{order3.2}.

1262: \begin{figure} \label{order3.2}

1263: \begin{center}

1264: \input{figurel3.pstex_t}

1265: \end{center}

1266: \caption{The first row represents possible terms for $\S$ and the

1267: second row those for $Y$. The arrows indicate the action of $d_s$

1268: (the spinorial derivative). Some arrows are suppressed. The

1269: products of Young-Tableaux represent plethysms of $\L$, so that

1270: the left factor is a tableau of the (3) of Spin(1,2) and the right

1271: factor is a tableau of the (0010) of Spin(8).}

1272: \end{figure}

1273: As indicated in the figure, after using the field redefinition, we

1274: are left with two different $X^2\L$ and two different $\L^5$ terms

1275: in $\S$. We can choose a basis for the $X^2\L$ such that only one

1276: of them contributes to the forbidden $\L \nab\L X$ (arrows

1277: indicating its contributions to other representation in $Y$ are

1278: suppressed). The latter cannot be generated from either of the

1279: $\L^5$ terms. Therefore, the first $X^2\L$-term would violate the

1280: constraints and cannot be present. An explicit calculation then

1281: shows that the remaining $X^2\L$ term  contributes non-trivially

1282: to the $X\L^4$ with a (2,2) plethysm of the $\L$. Again, neither

1283: of the $\L^5$ terms can cancel this, and so the second $X^2\L$

1284: term must also be trivial (again, arrows indicating its

1285: contributions to other representations in $Y$ are suppressed).

1286: Finally, as the arrows in the figure indicate, the $\L^5$ terms

1287: cannot remove each other's contributions towards $Y$. This shows

1288: that there is no linear combination of the potential terms which

1289: obey the constraints.

1290: 

1291: 

1292: \subsection{Spinorial cohomology at $\b=\ell^4$ }

1293: 

1294: By dimensional analysis $\S^{~~~a'}_{\a i}$ is, schematically,

1295: %

1296: \begin{align}

1297: \S&\sim X^3\L\oplus X\nabla X\L \oplus X^2\nabla \L \oplus

1298: X\nabla^2\L \oplus \nabla^2X\L

1299: \oplus\nabla X\nabla \L \oplus\nabla^3\L \nn\\

1300: &\oplus X^2\L^3 \oplus\nabla X\L^3 \oplus X\nabla\L\L^2

1301: \oplus\L^2\nabla^2\L \oplus (\nabla\L)^2\L \oplus X\L^5

1302: \oplus\nabla\L \L^4 \oplus \L^7 \label{cohfour}

1303: \end{align}

1304: %

1305: In terms of irreducible representations of $Spin(1,2)\times

1306: Spin(8)$, we can decompose the above contributions as

1307: %

1308: \begin{align}

1309: \L^7

1310: &\sim  7~(1)\times (1001)\oplus\dots\nn\\

1311: %

1312: \L^4\nabla\L

1313: %&\sim

1314: %\big( \yng(1,1,1,1)_{(3)}\times\yng(4)_{(0010)}\oplus

1315: %\yng(4)_{(3)}\times\yng(1,1,1,1)_{(0010)}\oplus

1316: %\yng(2,1,1)_{(3)}\times\yng(3,1)_{(0010)}\nn\\

1317: %&\oplus \yng(3,1)_{(3)}\times\yng(2,1,1)_{(0010)}\oplus

1318: %\yng(2,2)_{(3)}\times\yng(2,2)_{(0010)}\big)\otimes(5)\times(0010)\nn\\

1319: &\sim 12~(1)\times(1001)\oplus \dots\nn\\

1320: %

1321: %

1322: X^2\L^3

1323: %&\sim

1324: %\big(\yng(2)_{(4)}\times\yng(2)_{(1000)}\oplus\yng(1,1)_{(4)}

1325: %\times\yng(1,1)_{(1000)}\big)\nn\\

1326: %&\otimes\big(\yng(3)_{(3)}\times\yng(1,1,1)_{(0010)}

1327: %\oplus\yng(2,1)_{(3)}

1328: %\times\yng(2,1)_{(0010)}\big)

1329: %\oplus\yng(1,1,1)_{(3)}

1330: %\times\yng(3)_{(0010)}\big) \nn\\

1331: &\sim 28~(1)\times (1001)\oplus\dots\nn\\

1332: %

1333: X^2\nabla\L

1334: %&\sim \big(\yng(2)_{(4)}\times\yng(2)_{(1000)}\oplus\yng(1,1)_{(4)}

1335: %\times\yng(1,1)_{(1000)}\big)

1336: %\otimes

1337: %(5)\times(0010)  \nn\\

1338: &\sim 2~(1)\times (1001)\oplus\dots\nn\\

1339: %

1340: %

1341:  \L^2\nabla^2\L

1342: %&\sim \big(\yng(2)_{(3)}\times\yng(1,1)_{(0010)}\oplus\yng(1,1)_{(3)}

1343: %\times\yng(2)_{(0010)}\big)

1344: %\otimes

1345: %(7)\times(0010)\nn\\

1346: &\sim 1~(1)\times (1001)\oplus\dots\nn\\

1347: %

1348:  (\nabla\L)^2\L

1349: %&\sim \big(\yng(2)_{(5)}\times\yng(1,1)_{(0010)}\oplus\yng(1,1)_{(5)}

1350: %\times\yng(2)_{(0010)}\big)

1351: %\otimes

1352: %(3)\times(0010)\nn\\

1353: &\sim 1~(1)\times (1001)\oplus\dots\nn\\

1354: X\nabla X\L

1355: %&\sim (4)\times(1000)\otimes

1356: % (6)\times(1000)\otimes  (3)\times(0010)\nn\\

1357: &\sim  4~(1)\times (1001)\oplus\dots\nn

1358: %

1359: \end{align}

1360: %

1361: In addition one can verify that there are no more contributions

1362: coming from the  rest of the terms on the right-hand side of

1363: (\ref{cohfour}). There are therefore fifty-five possible terms.

1364: 

1365: We shall now make the assumption that there is a non-trivial

1366: cohomology element cubic in the fields. This is what we expect

1367: from string theory calculations in ten dimensions. In order to

1368: determine the cubic part $(\S^{(cub)a'}_{\a i})$ of the

1369: cohomology, we shall only require the explicit form of the cubic

1370: terms. These are eight in total,

1371: %

1372: \begin{align}

1373: \S^{(1)a'}_{\a i}&=X^{b;ca'}\nabla_{b}X_{c}{}^{ab'}

1374: (\s_{b'}\L_{a})_{\a i}\vert_{(1001)}  \nn\\

1375: \S^{(2)a'}_{\a i}&=X^{bcb'}\nabla_{b}X_{c}{}^{aa'}

1376: (\s_{b'}\L_{a})_{\a i}\vert_{(1001)} \nn\\

1377: \S^{(3)a'}_{\a i}&=X^{bca'}\nabla_{b}X^{adb'}

1378: (\c_{cd}\otimes \s_{b'}\L_{a})_{\a i}\vert_{(1001)}  \nn\\

1379: \S^{(4)a'}_{\a i}&=X^{bcb'}\nabla_{b}X^{ada'}

1380: (\c_{cd}\otimes \s_{b'}\L_{a})_{\a i}\vert_{(1001)} \nn\\

1381: \S^{(5)a'}_{\a i}&=(\s_{b'}\nabla_b\L_c)_{\a i}

1382: X^{bda'}X_{d}{}^{cb'}\vert_{(1001)}  \nn\\

1383: \S^{(6)a'}_{\a i}&= (\c^{bc}\otimes\s_{b'}\nabla^d\L^e)_{\a i}

1384: X_{bd}{}^{a'}X_{ce}{}^{b'}

1385: \vert_{(1001)} \nn\\

1386: \S^{(7)a'}_{\a i}&=(\L^b\c^d\otimes \s^{a'b'}\L^c)

1387: (\s_{b'}\nabla_b\nabla_c \L_d)_{\a i}\vert_{(1001)} \nn\\

1388: \S^{(8)a'}_{\a i}&=(\L^b\c^d\otimes\s^{a'b'}\nabla_b\L^c)

1389: (\s_{b'}\nabla_c\L_d)_{\a i}\vert_{(1001)}  \nn, \label{upslns}

1390: \end{align}

1391: %

1392: so that

1393: %

1394: \begin{equation}

1395: \S^{(cub)a'}_{\a i}=\sum_{I=1}^{8}c^I\S^{(I)a'}_{\a i}.

1396: \label{cubel}

1397: \end{equation}

1398: %

1399: The coefficients $c^I$ will be determined, up to an overall

1400: factor, in the following.

1401: 

1402: Carrying out a similar analysis for the redefinitions $(\d

1403: z)^{a'}$ at this order in $\b$ we conclude that the only possible

1404: contributions are of the form,

1405: %

1406: \begin{align}

1407: \L^4X

1408: %&\sim

1409: %\big( \yng(1,1,1,1)_{(3)}\times\yng(4)_{(0010)}\oplus

1410: %\yng(4)_{(3)}\times\yng(1,1,1,1)_{(0010)}\oplus

1411: %\yng(2,1,1)_{(3)}\times\yng(3,1)_{(0010)}\nn\\

1412: %&\oplus \yng(3,1)_{(3)}\times\yng(2,1,1)_{(0010)}\oplus

1413: %\yng(2,2)_{(3)}\times\yng(2,2)_{(0010)}\big)\otimes(4)\times(1000)\nn\\

1414: &\sim 5~(0)\times(1000)\oplus \dots\nn\\

1415: \L^2\nabla X

1416: %&\sim  \big(\yng(2)_{(3)}\times\yng(1,1)_{(0010)}\oplus\yng(1,1)_{(3)}

1417: %\times\yng(2)_{(0010)}\big)\otimes

1418: %(6)\times(1000)   \oplus \dots\nn\\

1419: &\sim 1~(0)\times(1000) \oplus\dots\nn\\

1420: X\L\nabla\L

1421: %&\sim (4)\times(1000)\otimes (3)\times(0010) \otimes (5)\times (0010)\nn\\

1422: &\sim 2~(0)\times(1100)\oplus \dots\nn\\

1423: X^3&\sim 1~(0)\times(1100)\oplus \dots\nn

1424: \end{align}

1425: %

1426: There are nine terms in total, but we shall only need the explicit

1427: form of the following cubic terms,

1428: %

1429: \begin{align}

1430: (\d z)^{(1)a'}&=X^{bca'}X_{bdb'}X_{c}{}^{db'}\nn\\

1431: (\d z)^{(2)a'}&=(\L^b\c^d\otimes\s^{a'b'}\L^c)\nabla_bX_{cdb'}\nn\\

1432: (\d z)^{(3)a'}&=(\L^b\c^d\otimes\s^{a'b'}\nabla_b\L^c)X_{cdb'}\nn\\

1433: (\d z)^{(4)a'}&=(\L^d\c^b\nabla^c\L_d)X_{bc}{}^{a'}.

1434: \label{redfns}

1435: \end{align}

1436: %

1437: %

1438: Finally, we need to repeat the analysis for the constraints

1439: $Y_{a'b';c'}$ and $Y_{aa'b'c'd';e'}$. We find the following

1440: contributions,

1441: %

1442: \begin{align}

1443: X^2\nabla X&\sim 1~(0)\times(1100)\oplus \dots\nn\\

1444: \nabla X\nabla \L\L&\sim 1~(0)\times(1100)\oplus \dots\nn\\

1445: X\nabla^2\L\L&\sim 1~(0)\times(1100)\oplus\dots\nn\\

1446: X^3\L^2&\sim 12~(0)\times(1100)\oplus 13~(2)\times(1002)\oplus\dots\nn\\

1447: X\nabla\L\L^3&\sim 26~(0)\times(1100)\oplus 38~(2)\times(1002)\oplus\dots\nn\\

1448: \nabla X\L^4&\sim 6~(0)\times(1100)\oplus 7~(2)\times(1002)\oplus\dots\nn\\

1449: X\L^6&\sim 22~(0)\times(1100)\oplus

1450: 35~(2)\times(1002)\oplus\dots\nn

1451: \end{align}

1452: %

1453: We shall only need the explicit form of the following terms,

1454: %

1455: \begin{align}

1456: Y^{(1)a'b';c'}&=X^{bca'}X_a{}^{d[b'|}\nabla_bX^{ae|c']}

1457: \vare_{cde}\vert_{(1100)}\nn\\

1458: Y^{(1)a'b';c'}&=(\nabla^b\L^c\s^{b'c'}\L^a)\nabla_b

1459: X_{ca}{}^{a'}\vert_{(1100)} \nn\\

1460: Y^{(1)a'b';c'}&=(\nabla_b\nabla_c\L_a\s^{b'c'}\L^a)

1461: X^{bca'}\vert_{(1100)}. \nn \label{cnstrts}

1462: \end{align}

1463: %

1464: All the above is summarised in the following diagram.

1465: %%

1466: %%THE FIGURE

1467: %%

1468: %

1469: \leavevmode

1470: \begin{figure}[h]

1471: \begin{center}

1472: \input{figure2.pstex_t}

1473: \end{center}

1474: \caption{Spinorial cohomology at $\ell^4$.}

1475: \end{figure}

1476: %

1477: %

1478: The coefficients $c^I$ in (\ref{cubel}) are determined, up to an

1479: overall factor, as follows. First note that the redefinitions

1480: (\ref{redfns}) can be used to eliminate $\S^{(I)a'}_{\a i};

1481: ~I=5\dots 8$. Indeed an explicit

1482:  calculation gives

1483: %%

1484: \begin{align}

1485: d_s(\d z)^{(2)a'}&\to \S^{(7)a'}_{\a i}+\dots \nn\\

1486: d_s(\d z)^{(3)a'}&\to i\S^{(8)a'}_{\a i}+\dots\nn,

1487: \end{align}

1488: %

1489: so that $(\d z)^{(2,3)a'}$ can be used to eliminate

1490: $\S^{(7,8)a'}_{\a i}$. Moreover,

1491: %%

1492: %%

1493: \begin{align}

1494: d_s(\d Z)^{(1)a'}&\to 2i\S^{(5)a'}_{\a i}+\dots \nn\\

1495: d_s(\d Z)^{(4)a'}&\to-{1\over 2}\S^{(5)a'}_{\a i} +{1\over

1496: 2}\S^{(6)a'}_{\a i} +\dots \nn,

1497: \end{align}

1498: %

1499: and so $(\d z)^{(1,4)a'}$ can be used to eliminate

1500: $\S^{(5,6)a'}_{\a i}$. In the following we set

1501: $$

1502: c^I=0; ~~I=5\dots 8.

1503: $$

1504: Next we evaluate the action of $d_s$ on the remaining cubic terms,

1505: %%

1506: \begin{align}

1507: d_s\S^{(1)a'}_{\a i}&\to-4iY^{(3)a'b';c'}  +\dots \nn\\

1508: d_s\S^{(2)a'}_{\a i}&\to-4iY^{(2)a'b';c'}  +\dots \nn\\

1509: d_s\S^{(3)a'}_{\a i}&\to-16iY^{(1)a'b';c'}-4iY^{(3)a'b';c'}  +\dots  \nn\\

1510: d_s\S^{(4)a'}_{\a i}&\to-32iY^{(1)a'b';c'}+4iY^{(2)a'b';c'}

1511: +\dots \nn,

1512: \end{align}

1513: %

1514: where the ellipses stand for terms with more than three fields.

1515: Demanding the vanishing of $d_s$ fixes the remaining coefficients

1516: to be

1517: %

1518: \begin{align}

1519: c^2&={1\over 2}c^1\nn\\

1520: c^3&=-c^1\nn\\

1521: c^4&={1\over 2}c^1\nn.

1522: \end{align}

1523: %

1524: We conclude that the part of the spinorial cohomology cubic in the

1525: fields at order $\ell^4$ is determined up to an overall factor to

1526: be represented by

1527: 

1528:  \bea

1529:  \S^{~~~a'}_{\a i}&=&X^{bca'}\nabla_{b}X_{c}{}^{ab'}

1530: (\s_{b'}\L_{a})_{\a i}\vert_{(1001)}+ \frac{1}{2}

1531: X^{bcb'}\nabla_{b}X_{c}{}^{aa'}

1532: (\s_{b'}\L_{a})_{\a i}\vert_{(1001)} \nn\\

1533: &&-X^{bca'}\nabla_{b}X^{adb'} (\c_{cd}\otimes \s_{b'}\L_{a})_{\a

1534: i}\vert_{(1001)} \nn\\

1535: &&+\frac{1}{2} X^{bcb'}\nabla_{b}X^{ada'} (\c_{cd}\otimes

1536: \s_{b'}\L_{a})_{\a i}\vert_{(1001)}

1537:  \eea

1538: 

1539: 

1540: 

1541: \section{Action}

1542: 

1543: There are various ways of deriving the action corresponding to the

1544: deformed embedding constraint. One is to compute the modified

1545: equations of motion by solving the torsion constraints at

1546: dimensions one-half and one, and then to work back to the

1547: Lagrangian. This is a systematic approach, but is fairly tedious

1548: in practice. Another possibility is to use the action principle

1549: for branes. We shall discuss this below, but it turns out that one

1550: is required to solve another spinorial cohomology problem. In

1551: order to avoid this we shall find the key part of the Lagrangian

1552: up to quadratic order in fermions by brutal force. Nevertheless,

1553: the action principle does shed light on the structure of the

1554: component action.

1555: 

1556: \subsection{Action principle}

1557: 

1558: We recall that  supersymmetric Lagrangians in $D$-dimensional

1559: spacetime can be constructed from closed superspace $D$-forms

1560: \cite{D'Auria:1982pm,ggks}. Given such a form, $L$, the spacetime

1561: action is

1562: 

1563:  \be

1564:  S=\int\,\e^{m_1\ldots m_D} L_{m_1\ldots m_D}(x,0)

1565:  \la{act}

1566:  \ee

1567: 

1568: Under an infinitesimal superspace diffeomorphism generated by a

1569: vector field $v$, $\d L=\cL_v L=i_v dL+ d(i_v L)=d(i_v L)$ because

1570: $L$ is closed. By evaluating this variation at $\th=0$ on the

1571: spacetime part of the $D$-form one can see that the action

1572: \eq{act} is invariant under spacetime diffeomorphisms and local

1573: supersymmetry transformations, these being identified as the

1574: leading components of the even and odd superspace diffeomorphisms.

1575: 

1576: In the context of branes one can construct a Lagrangian form

1577: starting from the (closed) Wess-Zumino $(d+1)$-form $W$ on the

1578: super worldvolume, where $d$ is the dimension of the bosonic

1579: worldvolume. This can be written in terms of an explicit potential

1580: $Z$, $W=d Z$, but it can also be written in terms of a

1581: globally-defined $d$-form, $W=dK$. The $d$-form $L=K-Z$ is closed

1582: by construction. It is uniquely defined up to the exterior

1583: derivative of a $(d-1)$-form which is irrelevant in the action.

1584: When the standard embedding constraint $E_{\a}{}^{\una}=0$ is

1585: imposed (in the on-shell or Lagrangian off-shell cases) this

1586: procedure gives the Green-Schwarz action, including the

1587: Born-Infeld contribution in the case of $D$-branes \cite{hrs}. At

1588: higher orders, it must still be possible to write $W$ as $dK$, but

1589: we shall also require Lagrangian forms satisfying $dL=0$.

1590: 

1591: For the M2-brane $W$ is the pull-back of the target space

1592: four-form $H$ on to the brane, $W=f^*\unH:=H$. The components of

1593: the pullback are (up to signs)

1594: $$

1595: H_{ABCD}=E_D{}^{\unD}E_C{}^{\unC}

1596: E_B{}^{\unB}E_A{}^{\unA}H_{\unA\unB\unC\unD}.

1597: $$

1598: The only non-vanishing component of $\unH$ in flat superspace is

1599: 

1600:  \be

1601:  H_{\una\unb,\uc\ud}=-i(\C_{\una\unb})_{\uc\ud}

1602:  \ee

1603: 

1604: 

1605: To order $\b$ we therefore have

1606: 

1607: \noindent dim. -1

1608: %

1609: \begin{equation}

1610: H_{\a\b\c\d}=0 \label{ha}

1611: \end{equation}

1612: %

1613: \noindent dim. -1/2

1614: %

1615: \begin{equation}

1616: H_{a\b\c\d}=0

1617: \end{equation}

1618: %

1619: \noindent dim. 0

1620: %

1621: \begin{equation}

1622: H_{ab\c\d}=-i(\C_{ab})_{\c\d}-4i\L_{[a}{}^{\a'}(\C_{b]c'})_{\a'(\c}

1623: \psi_{\d)}{}^{c'}

1624: \end{equation}

1625: %

1626: \noindent dim. 1/2

1627: %

1628: \begin{equation}

1629: H_{abc\d}=-3i(\C_{[ab|})_{\c'\d'}\L_{|c]}{}^{\c'}h_{\d}{}^{\d'}

1630: \end{equation}

1631: %

1632: \noindent dim. 1

1633: \begin{equation}

1634: H_{abcd}=0 \label{hd}

1635: \end{equation}

1636: %

1637: We can now solve $H=dK$. In components,

1638: $$

1639: H_{ABCD}=4\nab_{[A}K_{BCD]}+ 6 T_{[AB|}^FK_{F|CD]}

1640: $$

1641: Note that for $\b=0$, all the components of $K$ are zero, except

1642: for the highest-dimension component $K_{abc}$. To order $\b$ we

1643: have, taking (\ref{ha}--\ref{hd}) into account ,

1644: 

1645: \noindent dim. -1

1646: $$

1647: 0=-{3i\over 2}(\C^f)_{(\a\b|}K_{f|\c\d)}+\nab_{(\a}K_{\b\c\d)}

1648: $$

1649: This is solved by

1650: $$

1651: K_{\a\b\c}=0; ~~~~~K_{a\b\c}=0.

1652: $$

1653: 

1654: \noindent dim. -1/2

1655: $$

1656: 0=(\C^f)_{(\b\c|}K_{fa|\d)}

1657: $$

1658: This implies

1659: $$

1660: K_{ab\c}=0.

1661: $$

1662: \noindent dim. 0

1663: $$

1664: -i(\C_{ab})_{\c\d}-4i\L_{[a}{}^{\a'}(\C_{b]c'})_{\a'(\c}

1665: \psi_{\d)}{}^{c'}=

1666: \big(-i(\C^e)_{\c\d}-2i\L^{e\a'}(\C_{c'})_{\a'(\c}

1667: \psi_{\d)}{}^{c'} \big)K_{eab}

1668: $$

1669: where we have taken into account the expression for $T_{\a\b}{}^e$

1670: to order $\b$, given in section three.  The above equation is

1671: satisfied when

1672: $$

1673: K_{abc}=\vare_{abc}.

1674: $$

1675: This is identical to the solution for $K_{ABC}$ in the $\b=0$

1676: case. This does not necessarily mean that there is no correction

1677: to the Green-Schwarz action because the induced Green-Schwarz

1678: metric may change. However, one can check that there is no such

1679: change to first order in $\b$ although corrections of this type

1680: will arise at higher orders.

1681: 

1682: This means that the Lagrangian form we are looking for must

1683: satisfy $dL=0$, and, furthermore, that $L_{\a\b\c}$ cannot be

1684: zero. Since the Lagrangian form can be changed by an exact form

1685: without affecting the action \eq{act}, it follows that this gives

1686: rise to a new spinorial cohomology problem. This is more

1687: complicated than the one treated in section six in that the

1688: representations involved are larger. On dimensional grounds, the

1689: order $\b$ contribution to $L_{\a\b\c}$ which is quartic in fields

1690: has the schematic form $\L^3 X$ but we have not attempted to

1691: compute it explicitly. However, it is useful to observe that this

1692: method of constructing actions tells us something about the

1693: structure of the component Lagrangian. From \eq{act} we have

1694: 

1695:  \be

1696:  \e^{mnp} L_{mnp}=\e^{mnp}\left(E_p{}^c E_n{}^b E_m{}^a L_{abc} +

1697:  3E_p{}^c E_n{}^b E_m{}^{\a} L_{\a bc}+\ldots\right)

1698:  \la{lag}

1699:  \ee

1700: 

1701: The non-leading terms on the right-hand side all involve the

1702: induced worldvolume ``gravitino'' field $E_m{}^{\a}$. This is

1703: related to  $\del_m\th^{\ua}$ but it is not expressible in terms

1704: of $\L$. The components $L_{ABC}$ will all be constructed from

1705: manifest tensors, namely $\L$ and $X$; the gravitino terms reflect

1706: the fact that the Lagrangian is a density.

1707: 

1708: \subsection{Component approach}

1709: 

1710: In the component approach we write the action as $S=S_0+S_1$,

1711: where $S_0$ is the usual Green-Schwarz action for the

1712: supermembrane \cite{Bergshoeff:1987cm}, and similarly for the

1713: worldvolume supersymmetry variations, so that at first order we

1714: have

1715: 

1716: \be \d_1 S_0 + \d_0 S_1 = 0 \ee

1717: 

1718: We can ignore the first term if we calculate $S_1$ using the

1719: lowest-order field equations. The neglected terms correspond to

1720: field redefinitions. We are therefore looking for an on-shell

1721: invariant at order $\b$.

1722: 

1723: {}From the discussion above we have seen that the first-order

1724: Lagrangian density $\cL$ will have the form

1725: 

1726:  \be

1727:  \cL=\sqrt{-g}\hat L + \e^{mnp}\left(

1728:  3E_p{}^c E_n{}^b E_m{}^{\a} L_{\a bc}+\ldots\right)

1729:  \la{82}

1730:  \ee

1731: 

1732: where $g$ is the determinant of the Green-Schwarz metric, $\hat

1733:  L:= \e^{abc} L_{abc}$ and where the right-hand side is understood

1734: to be evaluated at $\th=0$. Under a worldvolume supersymmetry

1735: variation we find

1736: 

1737:  \bea

1738:  \d\cL &=& \del_m (v^{\a} \cL^m_{\a})\nn\\

1739:  &=& \d(\sqrt{-g})\hat L + \sqrt{-g}\, \d \hat L \nn\\

1740:  &\phantom{=}& +\e^{mnp}\left(3E_p{}^c E_n{}^b E_m{}^{\a}\d L_{\a bc}

1741:  +6E_p{}^c (\d E_n{}^b) E_m{}^{\a}L_{\a

1742:  bc}+3E_p{}^c  E_n{}^b (\d E_m{}^{\a})L_{\a

1743:  bc} \ldots\right)

1744:  \eea

1745: 

1746: for some $\cL^m_{\a}$.  To quadratic order in fermions the terms

1747: represented by the dots in \eq{82}, which have two or more powers

1748: of $E_m{}^{\a}$, can be disregarded. The schematic form of the

1749: Lagrangian terms are

1750: 

1751:  \bea

1752:  \hat L&\sim& X^4 + \L^2 \nab X^2 + \cO(\L^4) \nn\\

1753:  L_{\a bc}&\sim& \L X^3 + \cO(\L^3)

1754:  \eea

1755: 

1756: while

1757: 

1758:  \bea

1759:  \d E_m{}^a &=&iv^{\b}(\C^a)_{\b\c} E_m{}^{\c}\nn\\

1760:  \d E_m{}^{\a}&=& \nab_m v^{\a} + v^{\b}E_m{}^c T_{\b c}{}^{\a}~.

1761:  \eea

1762: 

1763: We shall focus on the terms in the variation of $\cL$ which are

1764: linear in fermions. In the variation of $E_m{}^{\a}$ we can use

1765: the lowest-order expression for the induced torsion $T_{\b

1766: c}{}^{\a}$. As this is quadratic in $\L$ the corresponding term in

1767: $\d E_m{}^\a$ can be ignored. The term involving the variation of

1768: the world-volume vielbein $E_m{}^a$ can also be ignored as it

1769: leads to terms which are cubic in  fermions. The terms which are

1770: linear in fermions involve either $\L$ or $E_m{}^{\a}$, and these

1771: must cancel separately. Thus we see that the term arising from the

1772: variation of the determinant of the metric multiplied by the $X^4$

1773: term in $\hat L$ must be cancelled by the leading term coming from

1774: the variation of $L_{\a bc}$. The terms linear in $\L$ arise from

1775: varying both the $\L^0$ and $\L^2$ terms in $\hat L$ and from the

1776: $\nab v$ variation term in $E_m{}^{\a}$ multiplied by $L_{\a bc}$.

1777: It therefore follows that the terms linear in $\L$ in the

1778: variation of $\hat L$ will cancel up to a covariant divergence if

1779: the parameter $v^{\a}$ is taken to be covariantly constant. The

1780: terms involving the derivative of $v$ in the variation of $\hat L$

1781: must cancel against similar terms arising from the variation of

1782: $E_m{}^{\a}$. In fact, demanding that these terms cancel

1783: determines $L_{\a bc}$ at first order in $\L$. We shall therefore

1784: concentrate on computing $\hat L$ up to quadratic order in the

1785: fermions. In what follows we shall refer to $\hat L$ as the

1786: Lagrangian.

1787: 

1788: 

1789: It follows from the analysis in section six that, up to and

1790: including terms quadratic in the fermions, there is a unique

1791: supersymmetric invariant at order $l^4$. Indeed, we have shown

1792: that at order cubic in the fields there is a unique nontrivial

1793: element in the relevant spinorial cohomology group and this

1794: implies that there is a unique deformation of the equations of

1795: motion. We conclude that in the action there is a unique

1796: supersymmetric invariant at order quartic in the fields. In

1797: particular,  there can be no terms in the action of order cubic,

1798: or less, in the fields. There are also no terms of the form

1799: $\Lambda^2X^3$ as can be seen by the fact that there is no scalar

1800: in the decomposition of the tensor product

1801: %

1802: \be

1803: [(3)\times(0010)]^{2\otimes_a}\otimes[(4)\times(1000)]^{3\otimes_s},

1804: \ee

1805: %

1806: The bosonic part of the Lagrangian at $l^4$, is a linear

1807: combination of the following three $X^4$ terms

1808: %

1809: \bea

1810: t_1&:=&X_{ab}{}^{a'}X_{cda'}X^{abb'}X^{cd}{}_{b'} \nn\\

1811: t_2&:=&X_{ab}{}^{a'}X_{cd}{}^{b'}X^{ab}{}_{a'}X^{cd}{}_{b'} \nn\\

1812: t_3&:=&X_{ab}{}^{a'}X_{cd}{}^{b'}X^{ac}{}_{a'}X^{bd}{}_{b'}. \eea

1813: %

1814: A fourth structure

1815: %

1816: \be t_4:=X_{ab}{}^{a'}X_{cda'}X^{acb'}X^{bd}{}_{b'} \ee

1817: %

1818: can be shown to be linearly dependent

1819: %

1820: \be t_4=t_1+{1\over 2}t_2-2t_3, \label{wittgenstein} \ee

1821: %

1822: by using the identity

1823: %

1824: %

1825: \be 4X_{[a}{}^{[b|a'}   X_{c]}{}^{|d]b'}= \left(\delta_f^e

1826: X^{gha'}X_{gh}{}^{b'}

1827: -X_f{}^{ga'}X_g{}^{eb'}-X_f{}^{gb'}X_g{}^{ea'}  \right)

1828: \varepsilon_{eac} \varepsilon^{fbd}. \label{rc} \ee

1829: %

1830: %

1831: At quadratic order in the fermions, the Lagrangian at $l^4$

1832: contains terms of the form $\Lambda^2X\nabla X$ and $\Lambda

1833: \nabla\Lambda X^2$. A lengthy calculation, the details of which

1834: are given in appendix B, yields the following result for $\hat L$:

1835: 

1836: \bea \hat L&=&{1\over 4}X_{ab}{}^{a'}X_{cda'}X^{abb'}X^{cd}{}_{b'}

1837: -{1\over 4}X_{ab}{}^{a'}X_{cd}{}^{b'}X^{ac}{}_{a'}X^{bd}{}_{b'}\nn\\

1838: &&+i(\Lambda_a\gamma_b\nabla_c\Lambda_d) X^{aba'}X^{cd}{}_{a'}

1839: -{2i\over

1840: 3}(\Lambda^f\gamma_a\otimes\sigma^{a'b'}\nabla_f\Lambda_b)

1841: X^{ae}{}_{a'}X^{b}{}_{eb'}\nn\\

1842: &&-{i\over

1843: 3}(\Lambda_a\gamma_b\otimes\sigma^{a'b'}\nabla_c\Lambda_d)

1844: X^{ab}{}_{a'}X^{cd}{}_{b'}+{\cal O}(\Lambda^4). \label{alle} \eea

1845: %

1846: The Lagrangian above is determined up to a total derivative and up

1847: to terms vanishing by virtue of the lowest-order equations of

1848: motion. The latter can always by removed by appropriate field

1849: redefinitions.

1850: 

1851: We shall now compare our result to that of \cite{Bachas:1999um}.

1852: In this paper the bosonic CP-even part of the ${\cal O}(l^4)$

1853: Lagrangian of a D-$p$ brane was determined to be

1854: %

1855: \be {1\over 16}\left( (R_T)_{abcd}(R_T)^{abcd}

1856: -2(R_T)_{ab}(R_T)^{ab}-(R_N)_{abc'd'}(R_N)^{abc'd'}

1857: +2\bar{R}_{a'b'}\bar{R}^{a'b'} \right), \label{bachas} \ee

1858: %

1859: see equation (2.14) of that reference. In the above we have

1860: omitted an overall multiplicative constant involving $\alpha'$ and

1861: the string coupling.

1862: 

1863: 

1864: Let us view (\ref{bachas}) as the world-volume Lagrangian of a

1865: D2-brane and consider its lift to eleven dimensions. The various

1866: curvatures in (\ref{bachas}) are given, in the notation of the

1867: present paper and for flat target-space, by

1868: %

1869: \bea

1870: (R_T)_{abcd}&=&X_{aca'}X_{bd}{}^{a'}-X_{ada'}X_{bc}{}^{a'}   \nn\\

1871: (R_N)_{abc'd'}&=&X_{aec'}X^e{}_{bd'}-X_{bec'}X^e{}_{ad'} \nn\\

1872: (R_T)_{ab}&=&-X_{ae}{}^{a'}X^e{}_{ba'}\nn\\

1873: \bar{R}_{a'b'}&=&X_{aba'}X^{ab}{}_{b'}. \label{plop} \eea

1874: %

1875: Substituting (\ref{plop}) into (\ref{bachas}) on gets precisely

1876: the bosonic part of (\ref{alle}).

1877: 

1878: The Lagrangian (\ref{bachas}) was determined, as is the case with

1879: (\ref{alle}), modulo total derivatives \footnote{the coefficient

1880: of the Gauss-Bonnet term was fixed to zero in \cite{Bachas:1999um}

1881: by an indirect argument.} and terms which vanish by virtue of the

1882: lowest-order equations of motion. In \cite{Fotopoulos:2001pt} the

1883: investigation of these ambiguities was considered, see equation

1884: (5.4) therein. It was shown that there is a total of five

1885: ambiguous terms, two of which vanish by virtue of the lowest-order

1886: equations of motion and therefore cannot be determined

1887: perturbatively. Two of the other three terms vanish for flat

1888: target-space, whereas the remaining third term vanishes by virtue

1889: of equation (\ref{wittgenstein}) of the present paper.

1890: 

1891: \section{Kappa symmetry}

1892: 

1893: Although we have not used the modified worldvolume supersymmetry

1894: transformations in determining the Lagrangian, it is perhaps of

1895: interest to see how the usual $\k$-symmetry variations are

1896: altered. To lowest order the $\k$-symmetry transformations were

1897: derived, starting from worldvolume supersymmetry transformations,

1898: in equations (9-14). These will be amended when we deform the

1899: embedding condition. In general, a worldvolume supersymmetry

1900: transformation gives the following transformations

1901: 

1902:  \bea

1903:  \d z^{\una}&=& v^\a E_{\a}{}^{\una}  \\

1904:  \d z^{\ua}&=& v^{\a} E_{\a}{}^{\ua}~.

1905:  \eea

1906: 

1907: We can convert the second of these into $\k$-symmetry form as

1908: before:

1909: 

1910:  \be

1911:  \d z^{\ua}=\k^{\ub}

1912:  P_{\ub}{}^{\ua}=\frac{1}{2}\k^{\ub}(1+\C)_{\ub}{}^{\ua}~.

1913:  \ee

1914: 

1915: where $\k^{\ua}=v^{\a} E_{\a}{}^{\ua}$ and

1916: 

1917:  \be

1918:  P_{\ua}{}^{\ub}=(E^{-1})_{\ua}{}^{\c} E_{\c}{}^{\ub}

1919:  \ee

1920: 

1921: where the inverse refers to inverting the square matrix formed

1922: from $(E_{\a}{}^{\ua},E_{\a'}{}^{\ua})$. Explicitly

1923: 

1924:  \be

1925:  P_{\ua}{}^{\ub}=(u^{-1})_{\ua}{}^{\c}u_{\c}{}^{\ub} +

1926:  (u^{-1})_{\ua}{}^{\c}h_{\c}{}^{\d'}u_{\d'}{}^{\ub}

1927:  \ee

1928: 

1929: The first term here has the same structure as in the undeformed

1930: theory. The bosonic projection of $\d z^{\unM}$ can be written

1931: 

1932:  \be

1933:  \d z^{\una}=\k^{\ua} \Psi_{\ua}{}^{\una}=\d z^{\ua}

1934:  \Psi_{\ua}{}^{\una}~,

1935:  \ee

1936: 

1937: where

1938: 

1939:  \bea

1940:  \Psi_{\ua}{}^{\una}&=&(E^{-1})_{\ua}{}^\c E_{\c}{}^{\una}\nn\\

1941:  &=&(E^{-1})_{\ua}{}^\c \psi_{\c}{}^{a'} u_{a'}{}^{\una}

1942:  \eea

1943: 

1944: satisfies $\Psi=P\Psi$.

1945: 

1946: In the undeformed theory the matrix $\C$ is given by

1947: 

1948:  \be

1949:  \C=-\frac{1}{6}\e^{abc} u_a{}^{\una} u_b{}^{\unb} u_c{}^{\unc}

1950:  \C_{\una\unb\unc}~.

1951:  \la{plato}

1952:  \ee

1953: 

1954: This can easily be seen to be the same as

1955: 

1956:  \be

1957:  \C_{\ua}{}^{\ub}=(u^{-1})_{\ua}{}^{\c}u_{\c}{}^{\ub}-

1958:  (u^{-1})_{\ua}{}^{\c'}u_{\c'}{}^{\ub}

1959:  \la{socrates}

1960:  \ee

1961: 

1962: In the deformed theory we can write

1963: 

1964:  \be

1965:  \C=\C_0 +\frac{1}{2}u^{-1}(1+\C_0)h(1-\C_0)u

1966:  \ee

1967: 

1968: where $\C_0$ has the same form  as in \eq{plato} and \eq{socrates}

1969: and where $h$ is regarded as a $32\xz 32$ matrix satisfying $h=P_0

1970: h Q_0$, with $P_0=1/2(1+\C_0),\ Q=1/2(1-\C_0)$.

1971: 

1972: The formulae above are valid for an arbitrary deformation $\psi$.

1973: To obtain the order $\b$ term we merely have to substitute into

1974: the expressions the explicit forms for $\psi$ and $h$.

1975: 

1976: 

1977: \section{Conclusions}

1978: 

1979: In this paper we have discussed the first correction to the

1980: dynamics of the M2-brane in the superembedding formalism which we

1981: have shown to occur at order $\ell^4$. We have seen that the

1982: allowed modifications of the constraints are specified by an

1983: element of a certain spinorial cohomology group. There is a unique

1984: solution cubic in the fields and we believe that there are no

1985: other independent terms at quartic or higher orders. In principle

1986: one could continue this analysis for higher powers of $\ell$. We

1987: would expect to find both corrections induced by iterating the

1988: first correction and new cohomological terms. However, the

1989: computations are difficult, even at $\ell^4$.

1990: 

1991: The formalism developed here could be applied straightforwardly to

1992: other branes such as the M5-brane and D-branes. Indeed, we could

1993: obtain the equivalent terms in the D2-brane action by dimensional

1994: reduction. In particular, this would shed some light on the

1995: $\k$-symmetric effective action including $\del^4 F^4$ terms.

1996: Another extension of the formalism would be to include non-trivial

1997: supergravity backgrounds. In $D=11$ it is  expected that the first

1998: correction to supergravity occurs at $\ell^6$, so that this

1999: complication could be ignored. It should therefore be relatively

2000: straightforward to compute the corrections at the same order,

2001: which involve the background supergravity fields. In the case of

2002: the M5-brane these results could again be dimensionally reduced to

2003: ten dimensions, either to give results for the NS5-brane, or the

2004: D4-brane. In the latter case one would expect to see some sign of

2005: the $\hat A$-genus terms discussed in

2006: \cite{Green:1996dd,Cheung:1997az,minmo}. In principle, we should

2007: therefore be able to obtain some information about the

2008: supersymmetrisation of these terms.

2009: 

2010: 

2011: 

2012: \section*{Acknowledgements}

2013: 

2014: This work was supported in part by EU contracts HPRN-2000-00122 

2015: (which includes Queen 

2016: Mary, London as a subcontractor)

2017: and HPRN-CT-2000-00148 and PPARC grants PPA/G/S/ 1998/00613 and 

2018: PPA/G/O/2000/00451. 

2019: SFK thanks the German National Merit Foundation for

2020: financial support. UL acknowledges support in part by VR grant

2021: 650-1998368. 

2022: 

2023: 

2024: %%%%%%%%%%%%   APPENDIX    %%%%%%%%%%%%%%%

2025: 

2026: 

2027: \section*{Appendix A}

2028: 

2029: \subsection*{Notation}

2030: 

2031: The following index conventions are used: plain (underlined)

2032: indices refer to worldsurface and target space quantities

2033: respectively and primed indices refer to normal spaces; indices

2034: from the beginning of the alphabet refer to preferred bases,

2035: indices from the middle of the alphabet to coordinate bases; Latin

2036: (Greek) indices are used for even (odd) indices, while capital

2037: indices run over the whole space. Thus a worldsurface preferred

2038: basis index could be $A=(a,\a)$ while a target space coordinate

2039: index could be $\unM=(\unm,\um)$. Tensor quantities with indices

2040: are not underlined, but if the indices are omitted or if a target

2041: space tensor is projected on some of its indices we underline the

2042: tensor.

2043: 

2044: For the spinor indices a two step notation is used. Initially the

2045: target space index $\ua$, running from 1 to 32 is split in two,

2046: $\ua\rightarrow(\a,\a')$, where both the worldsurface index $\a$

2047: and the normal index $\a'$ run from 1 to 16. For some purposes

2048: this is adequate but sometimes one wishes to recognise explicitly

2049: that these indices transform under $Spin(1,2)\xz Spin(8)$. We set

2050: 

2051: \bea

2052: \psi^{\a}&\rightarrow&  \psi^{\a i}\nonumber\\

2053: \psi^{\a'}&\rightarrow& \psi^{\a i'} \eea where the spinor index

2054: on the right takes on two values while $i$ and $i'$ both  run from

2055: 1 to 8. We use space-favoured metrics throughout,

2056: $\h_{ab}=(-1,+1,\ldots,+1)$, and the spacetime $\e$-tensors have

2057: $\e^{0123\ldots}=+1$.

2058: 

2059: We use the following representation of the $D=11$ $\C$-matrices:

2060: \bea

2061: \C^a&=&\c^a\otimes \c_9\nonumber \\

2062: \C^{a'}&=& 1\otimes \c^{a'} \eea where the $d=3$ $\c$-matrices are

2063: $2\xz 2$ and the $d'=8$ $\c$-matrices are $16\xz 16$. The charge

2064: conjugation matrix is \be C=\e\otimes\c_9 \ee In indices \bea

2065: (\C^a)_{\ua}{}^{\ub}&=&(\c^a)_{\a}{}^{\b}\left(\ba{cc} \d_i{}^j & 0\\

2066: 0 & -\d_{i'}{}^{j'}\ea\right) \nonumber\\

2067: (\C^{a'})_{\ua}{}^{\ub}&=& \d_{\a}{}^{\b}\left(\ba{cc} 0&

2068: (\s^{a'})_{i}{}^{j'}\\

2069: (\tilde\s^{a'})_{i'}{}^{j}&0\ea\right) \eea and \be

2070: C_{\ua\ub}=\e_{\a\b}\left(\ba{cc}\d_{ij} &0\\

2071: 0&-\d_{i'j'}\ea\right) \ee where the $8\xz 8$ $\s$-matrices are

2072: related to the eight-dimensional $\c$-matrices by \be

2073: \c^{a'}=\left(\ba{cc}0&(\s^{a'})_{ij'}\\

2074: (\tilde\s^{\a'})_{i'j} &0\ea\right) \ee Eleven-dimensional spinor

2075: indices are raised or lowered according to the rule \be

2076: \psi^{\ua}=C^{\ua\ub} \psi_{\ub}\ \leftrightarrow\

2077: \psi_{\ua}=\psi^{\ub}C_{\ub\ua} \ee where $C$ with upper indices

2078: is the same matrix as $C$ with lower indices. Three-dimensional

2079: spinor indices are similarly raised or lowered using $\e$.

2080: Eight-dimensional indices, whether spinor or vector, are raised

2081: and lowered with the standard Euclidean metric. The $\C$-matrices

2082: with lowered indices are \bea

2083: (\C^a)_{\ua\ub}&=&(\c^a)_{\a\b}\left(\ba{cc} \d_{ij} & 0\\

2084: 0 & \d_{i'j'}\ea\right) \nonumber\\

2085: (\C^{a'})_{\ua\ub}&=& \e_{\a\b}\left(\ba{cc} 0&-(\s^{a'})_{ij'}\\

2086: (\tilde\s^{a'})_{i'j}&0\ea\right) \eea It is straightforward to

2087: decompose any of the eleven-dimensional $\C$-matrices with

2088: multi-vector indices in this way. We give the two-index

2089: $\C$-matrices as they are used most in the text: \bea

2090: (\C^{ab})_{\ua\ub}&=&(\c^{ab})_{\a\b}\left(\ba{cc} \d_{ij} & 0\\

2091: 0 & \d_{i'j'}\ea\right) \nonumber\\

2092: (\C^{ab'})_{\ua\ub}&=&-(\c^a)_{\a\b}\left(\ba{cc} 0&-(\s^{b'})_{ij'}\\

2093: (\tilde\s^{b'})_{i'j}&0\ea\right)\nonumber\\

2094: (\C^{a'b'})_{\ua\ub}&=& \e_{\a\b}\left(\ba{cc} (\s^{a'b'})_{ij'}&0\\

2095: 0&-(\tilde\s^{a'b'})_{i'j}\ea\right) \eea

2096: 

2097: The three-dimensional $\c$-matrices are real, symmetric with

2098: lowered indices, and satisfy \be \c^a\c^b=\h^{ab} + \c^{ab} \ee

2099: where \be \c^{ab}=\e^{abc}\c_c\ \leftrightarrow

2100: \c_a=-{1\over2}\e_{abc}\c^{bc} \ee

2101: 

2102: For the eight-dimensional $\s$-matrices one may take \bea

2103: \s^{a'}&=&(1,i\t_r)\nonumber\\

2104: \tilde\s^{a'}&=&(1,-i\t_r) \eea where $r=1,\ldots 7$, and where

2105: the $\t_r$ are seven-dimensional Dirac matrices which are purely

2106: imaginary and antisymmetric. The matrices $\s^{a'b'c'd'}$ and

2107: $\tilde\s^{a'b'c'd'}$ are symmetric, the former being self-dual,

2108: the latter anti-self-dual, $\s^{a'b'}$ and $\tilde\s^{a'b'}$ are

2109: antisymmetric.

2110: 

2111: \subsection*{Group theoretic conventions}

2112: 

2113: 

2114: Consider a Lie group $G$ and a $G$-module $V$. The $n$-th tensor

2115: product $V^{\otimes n}$ admits a decomposition

2116: $$

2117: \sum_{R} V_R \times R

2118: $$

2119: under $G\times S_n$, where $R$ runs over all irreducible

2120: representations of the symmetric group $S_n$ and $V_R$ is a

2121: $G$-module. As is well known, the irreducible representations of

2122: $S_n$ can be parametrised by partitions of $n$ or, equivalently,

2123: by the associated Young diagrams. If $R$ is associated to the

2124: partition $\l$ of $n$, $V_R$ is the {\sl plethysm} of $V$ with

2125: respect to $\l$.

2126: 

2127: In this paper, we are using a notation referring to the product

2128: group $Spin(1,2)\times Spin(8)$.

2129: 

2130: The highest weight of the standard module (i.e. of the vector

2131: representation) of $Spin(8)$ is given by $(1000)$ on the basis of

2132: fundamental weights. A two-form is $(0100)$ and a three-form is

2133: $(0011)$. The self-dual and anti-self-dual four-forms are

2134: represented by $(0002)$, $(0020)$ respectively. The chiral and

2135: anti-chiral spinors are $(0001)$, $(0010)$. Moreover, the highest

2136: weight of the $k$-form-spinor is given by the sum of the highest

2137: weight of the $k$-form and the highest weight of the spinor.

2138: Recall that by a $k$-form-spinor we refer to the projection onto

2139: the irreducible (gamma-traceless) part. For example $(1001)$ is a

2140: chiral vector-spinor, $(0110)$ is an anti-chiral two-form-spinor,

2141: etc. We can also consider the projection onto the highest-weight

2142: representation of the plethysm associated to the partition

2143: $[2,\dots  ,2,1,\dots ,1]$ of $n+k$, with $k$ entries equal to 2

2144: and $n-k$ entries equal to 1. This is one way to view the

2145: irreducible $(n,k)$-tensors of section 5. For example, a

2146: (2,1)-tensor is a two-form and a one-form and is represented by

2147: $(1100)$. A (3,1)-tensor is a three-form and a one-form and is

2148: represented by $(1011)$, etc. The highest-weights of irreducible

2149: (gamma-traceless) $(n,k)$-tensor-spinors are represented in the

2150: obvious way. For example $(1101)$ is the highest weight of a

2151: $(2,1)$-tensor-(chiral) spinor. This discussion generalizes

2152: straightforwardly to representations corresponding to more general

2153: partitions.

2154: 

2155: In the case of $Spin(1,2)$ the situation is simpler, in that the

2156: representation denoted by $(n)$ is of dimension $n+1$. Hence, the

2157: spinor representation is denoted by $(1)$, the vector by $(2)$,

2158: the vector spinor by $(3)$ and so on.

2159: 

2160: 

2161: 

2162: %%%%%%%%%%%APPENDIX   B%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2163: 

2164: 

2165: 

2166: \section*{Appendix B}

2167: 

2168: In this appendix we give some details of the derivation of the

2169: action in section seven. At quadratic order in the fermions the

2170: Lagrangian at $\ell^4$ contains terms of the form $\L^2 X\nab X$

2171: and $\L\nab\L X^2$. We have the following structures:

2172: 

2173: \noindent $\Lambda^2X\nabla X$:

2174: %

2175: \bea u_1&:=&(\Lambda^f\gamma_a\otimes\sigma^{a'b'}\Lambda_f)

2176: X_{cda'}\nabla^aX^{cd}{}_{b'}\nn\\

2177: u_2&:=&(\Lambda_a\gamma_b\otimes\sigma^{a'b'}\Lambda_c)

2178: X^{a}{}_{fa'}\nabla^fX^{bc}{}_{b'}\nn\\

2179: u_3&:=&(\Lambda_a\gamma_b\Lambda_c)

2180: X^{a}{}_{fa'}\nabla^fX^{bca'}\nn\\

2181: u_4&:=&(\Lambda_a\gamma_b\otimes\sigma^{a'b'}\Lambda_c)

2182: X^{b}{}_{fa'}\nabla^fX^{ac}{}_{b'} \eea

2183: %

2184: \noindent $\Lambda \nabla\Lambda X^2$:

2185: %

2186: \bea s_1&:=&(\Lambda^f\gamma_a\nabla_f\Lambda_b)

2187: X^{aea'}X^{b}{}_{ea'}\nn\\

2188: s_2&:=&(\Lambda_a\gamma_b\nabla_c\Lambda_d)

2189: X^{aba'}X^{cd}{}_{a'}\nn\\

2190: s_3&:=&(\Lambda^f\gamma_a\otimes\sigma^{a'b'}\nabla_f\Lambda_b)

2191: X^{ae}{}_{a'}X^{b}{}_{eb'}\nn\\

2192: s_4&:=&(\Lambda_a\gamma_b\otimes\sigma^{a'b'}\nabla_c\Lambda_d)

2193: X^{ab}{}_{a'}X^{cd}{}_{b'}\nn\\

2194: s_5&:=&(\Lambda_a\gamma_b\otimes\sigma^{a'b'}\nabla_c\Lambda_d)

2195: X^{ac}{}_{a'}X^{bd}{}_{b'}\nn\\

2196: s_6&:=&(\Lambda_a\gamma_b\nabla_c\Lambda_d) X^{aca'}X^{bd}{}_{a'}.

2197: \eea

2198: %

2199: Not all of the above are independent. Only three of the $u$ and

2200: four of the $s$-terms are, as can be seen by counting the number

2201: of scalars in the decomposition of the tensor products

2202: %

2203: \be [(3)\times(0010)]^{2\otimes_a}\otimes[(4)\times(1000)]

2204: \otimes[(6)\times(1000)] \ee

2205: %

2206: and

2207: %

2208: \be [(3)\times(0010)]\otimes [(5)\times(0010)]

2209: \otimes[(4)\times(1000)]^{2\otimes_s}. \ee

2210: %

2211: Indeed on has

2212: %

2213: \bea

2214: 0&=&u_2-u_4-{1\over 2}u_1\nn\\

2215: 0&=&s_6-s_2+s_1\nn\\

2216: 0&=&3s_5-s_4+s_3. \label{ra} \eea

2217: %

2218: The first line of (\ref{ra}) is derived using the identity

2219: %

2220: \be

2221: \varepsilon_{egh}(\Lambda^g\gamma^h\otimes\sigma^{a'b'}\Lambda_c)

2222: ={1\over 2} \varepsilon_{dec}

2223: (\Lambda^f\gamma^d\otimes\sigma^{a'b'}\Lambda_f). \label{rb} \ee

2224: %

2225: The second line of (\ref{ra}) can be shown by taking (\ref{rc})

2226: into account.

2227: %

2228: %

2229: Finally, to derive the third line of  (\ref{ra}) one notes that

2230: %

2231: \be

2232: 0=(\Lambda_e\gamma^{[ae}\gamma_b\otimes\sigma_{a'b'}\nabla_c\Lambda_d)

2233: X_{a}{}^{c|a'}X^{|b]db'}. \label{rd} \ee

2234: %

2235: In addition one has to take into account the fact that some terms

2236: are related to each other by integration by parts. Indeed, up to

2237: total derivatives, one has the following relations

2238: %

2239: \bea

2240: u_1&=&2s_3\nn\\

2241: u_2&=&s_4-s_5\nn\\

2242: u_2&=&u_4+s_3\nn\\

2243: u_3&=&s_2-s_6\nn\\

2244: u_3&=&s_1\nn\\

2245: u_4&=&2s_5\nn\\

2246: u_4&=&u_2-s_3. \label{re} \eea

2247: %

2248: Only six of the equations (\ref{ra},\ref{re}) above are

2249: independent. It is perhaps interesting to note that the

2250: group-theoretic relations (\ref{ra}) are all implied by the

2251: partial integration equations (\ref{re}). The solution may be

2252: given in terms of four independent variables which we take to be

2253: $s_{1-4}$,

2254: %

2255: \bea

2256: u_1&=&2s_3\nn\\

2257: u_2&=&{1\over 3}s_3+{2\over 3}s_4\nn\\

2258: u_3&=&s_1\nn\\

2259: u_4&=&-{2\over 3}s_3+{2\over 3}s_4\nn\\

2260: s_5&=&-{1\over 3}s_3+{1\over 3}s_4\nn\\

2261: s_6&=&-s_1+s_2. \eea

2262: %

2263: Therefore, the lagrangian at $l^4$ can be written as

2264: %

2265: \be \hat{L}=\sum_{i=1}^3 A_{i} t_i+\sum_{i=1}^4 B_{i} s_i +{\cal

2266: O}(\Lambda^4), \label{lagrangian} \ee

2267: %

2268: where the constant coefficients $A,B$ are determined by the

2269: requirement that the supersymmetric variation of $\hat L$ be a total

2270: derivative, up to $\nab v$ terms.

2271: 

2272: 

2273: At linear order in the fermions, the supersymmetric variation of

2274: the correction to the Lagrangian, igoniring the $\nab v$ terms, consists of terms of the form

2275: $\nabla \Lambda X^3$, $\Lambda\nabla X X^2$. Explicitly, one can

2276: write down the following structures (suppressing spinor indices and the supersymmetry 

2277: parameter $v$)

2278: %

2279: 

2280: \noindent $\nabla \Lambda X^3$:

2281: %

2282: \bea w_1&:=&(\gamma_{ab}\otimes\sigma^{a'b'c'}\nabla_c\Lambda_d)

2283: X^a{}_{ea'}X^{eb}{}_{b'}X^{cd}{}_{c'}\nn\\

2284: w_2&:=&(\gamma_{ab}\otimes\sigma^{a'}\nabla_c\Lambda_d)

2285: X^a{}_{ea'}X^{eb}{}_{b'}X^{cdb'}\nn\\

2286: w_3&:=&(\gamma_{ab}\otimes\sigma^{a'}\nabla_c\Lambda_d)

2287: X^{ac}{}_{a'}X^{eb}{}_{b'}X_e{}^{db'}\nn\\

2288: w_4&:=&(\sigma^{a'}\nabla_a\Lambda_b)

2289: X^a{}_{ea'}X^{eg}{}_{b'}X_g{}^{bb'}\nn\\

2290: w_5&:=&(\sigma^{a'}\nabla_a\Lambda_b)

2291: X^{abb'}X^{eg}{}_{a'}X_{egb'}\nn\\

2292: w_6&:=&(\sigma^{a'}\nabla_a\Lambda_b)

2293: X^{ab}{}_{a'}X^{eg}{}_{b'}X_{eg}{}^{b'}\nn\\

2294: w_7&:=&(\gamma_{ab}\otimes\sigma^{a'}\nabla_c\Lambda_d)

2295: X^{ac}{}_{b'}X^{eb}{}_{a'}X_e{}^{db'}\nn\\

2296: w_8&:=&(\gamma_{ab}\otimes\sigma^{a'}\nabla_c\Lambda_d)

2297: X^{ac}{}_{b'}X^{ed}{}_{a'}X_e{}^{bb'}\nn\\

2298: w_9&:=&(\sigma^{a'}\nabla_a\Lambda_b)

2299: X^a{}_{eb'}X^{eg}{}_{a'}X_g{}^{bb'}\nn\\

2300: w_{10}&:=&(\gamma_{ab}\otimes\sigma^{a'b'c'}\nabla_c\Lambda_d)

2301: X^{cb}{}_{a'}X^{a}{}_{eb'}X^{ed}{}_{c'} \eea

2302: %

2303: 

2304: \noindent $\Lambda\nabla X X^2$:

2305: %

2306: \bea z_1&:=&(\sigma^{a'}\Lambda^a)

2307: \nabla_a X_{bca'}X^{bdb'}X^{c}{}_{db'}\nn\\

2308: z_2&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2309: \nabla_d X_{bca'}X_{aeb'}X^{deb'}\nn\\

2310: z_3&:=&(\sigma^{a'}\Lambda_a)

2311: \nabla_b X_{cda'}X^{abb'}X^{cd}{}_{b'}\nn\\

2312: z_4&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2313: \nabla_b X_{dea'}X_c{}^{db'}X^{e}{}_{ab'}\nn\\

2314: z_5&:=&(\sigma^{a'}\Lambda^a)

2315: \nabla_a X_{bc}{}^{b'}X^{bd}{}_{a'}X^{c}{}_{db'}\nn\\

2316: z_6&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2317: \nabla_d X_{bcb'}X_{aea'}X^{deb'}\nn\\

2318: z_7&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2319: \nabla_b X_{deb'}X_{aca'}X^{deb'}\nn\\

2320: z_8&:=&(\sigma^{a'}\Lambda_a)

2321: \nabla_b X_{cdb'}X^{aba'}X^{cdb'}\nn\\

2322: z_9&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2323: \nabla_d X_{bcb'}X_{a}{}^{eb'}X^{d}{}_{ea'}\nn\\

2324: z_{10}&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2325: \nabla_c X_{de}{}^{b'}X^{d}{}_{aa'}X^{e}{}_{bb'}\nn\\

2326: z_{11}&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2327: \nabla_b X_{de}{}^{b'}X_{acb'}X^{de}{}_{a'}\nn\\

2328: z_{12}&:=&(\gamma^{ab}\otimes\sigma^{a'b'c'}\Lambda^c)

2329: \nabla_d X_{bca'}X_{aeb'}X^{de}{}_{c'}\nn\\

2330: z_{13}&:=&(\gamma^{ab}\otimes\sigma^{a'b'c'}\Lambda^c)

2331: \nabla_c X_{dea'}X^{d}{}_{ab'}X^{e}{}_{bc'}\nn\\

2332: z_{14}&:=&(\gamma^{ab}\otimes\sigma^{a'b'c'}\Lambda^c)

2333: \nabla_b X_{dea'}X_{acb'}X^{de}{}_{c'}\nn\\

2334: z_{15}&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2335: \nabla_b X_{deb'}X_{aea'}X_{c}{}^{db'}\nn\\

2336: z_{16}&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2337: \nabla_b X^{deb'}X_{aeb'}X_{cda'}\nn\\

2338: z_{17}&:=&(\gamma^{ab}\otimes\sigma^{a'b'c'}\Lambda^c)

2339: \nabla_b X^{de}{}_{a'}X_{aeb'}X_{cdc'}\nn\\

2340: z_{18}&:=&(\sigma^{a'}\Lambda_a)

2341: \nabla_b X_{cdb'}X^{abb'}X^{cda'}\nn\\

2342: z_{19}&:=&(\gamma^{ab}\otimes\sigma^{a'}\Lambda^c)

2343: \nabla_b X_{dea'}X^{deb'}X_{acb'}\nn\\

2344: z_{20}&:=&(\sigma^{a'b'c'}\Lambda_a) \nabla_b

2345: X_{cda'}X^{ab}{}_{b'}X^{cd}{}_{c'}. \eea

2346: %

2347: As before, not all of these structures are linearly independent.

2348: Only six of the $w$ and fourteen of the $z$-terms are, as one can

2349: see by counting the number of spinors $[(1)\times(0001)]$ in the

2350: decomposition of the tensor products

2351: %%

2352: \be [(5)\times(0010)]\otimes[(4)\times(1000)]^{3\otimes_s} \ee

2353: %%

2354: and

2355: %%

2356: \be [(3)\times(0010)]\otimes[(6)\times(1000)]

2357: \otimes[(4)\times(1000)]^{2\otimes_s}. \ee

2358: %%

2359: Indeed, taking into account (\ref{rc}) and the identity

2360: %

2361: \be \gamma_{[ab}\Lambda_{c]}=0, \ee

2362: %

2363: one can derive the following relations

2364: %

2365: \bea

2366: w_7&=&-w_2+w_3-w_4+{1\over 2}w_6\nn\\

2367: w_8&=&w_2+w_7\nn\\

2368: w_9&=&-2w_4+w_5+{1\over 2}w_6\nn\\

2369: w_{10}&=&{1\over 2}w_1 \label{plirpa} \eea

2370: %

2371: and

2372: %

2373: \bea

2374: z_{15}+z_{16}&=&2z_5-z_6+z_7-z_9+z_{11}\nn\\

2375: z_{15}-z_{16}&=&z_{10}\nn\\

2376: z_{17}&=&{1\over 2}z_{13}\nn\\

2377: z_{19}&=&-z_1+z_2+z_4\nn\\

2378: z_{18}&=&-z_5+z_8+z_9+z_{10}-z_{11}+z_{15}\nn\\

2379: 2z_{20}&=&z_{12}-z_{13}-z_{14}-z_{17}. \label{plirpb} \eea

2380: %

2381: The last two lines in equation (\ref{plirpb}) above are derived by

2382: noting that

2383: %

2384: \be (\gamma^{[ea}\sigma_{a'}\Lambda_e) \nabla_b

2385: X_{c}{}^{d|b'}X_{a}{}^{ba'}X^{|c]}{}_{db'}=0 \ee

2386: %

2387: and

2388: %

2389: \be (\gamma^{[ea|}\sigma^{a'b'c'}\Lambda_e) \nabla_b

2390: X_{c}{}^{|d|}{}_{a'}X_{a}{}^{b}{}_{b'}X^{|c]}{}_{dc'}=0. \ee

2391: %

2392: In addition one has to allow for integrations by parts. Up to

2393: total derivatives, one has the following relations

2394: %

2395: \bea

2396: w_1&=&2z_{17}\nn\\

2397: w_2&=&z_4-z_{15}\nn\\

2398: w_3&=&z_2-z_7\nn\\

2399: w_4&=&-z_1-z_8\nn\\

2400: w_5&=&-z_3-z_{18}\nn\\

2401: w_6&=&-2z_8\nn\\

2402: w_7&=&z_6-z_{19}\nn\\

2403: w_8&=&z_9-z_{11}\nn\\

2404: w_9&=&-z_3-z_5\nn\\

2405: w_{10}&=&-z_{12}-z_{14} \label{plirpc} \eea

2406: %

2407: and

2408: %

2409: \bea

2410: z_1&=&-2z_5\nn\\

2411: z_2&=&z_6+z_{10}\nn\\

2412: z_2&=&w_4-z_{16}\nn\\

2413: z_4&=&w_9-z_9\nn\\

2414: z_4&=&-w_7+z_{10}\nn\\

2415: z_5&=&-w_4-z_{18}\nn\\

2416: z_6&=&w_4-z_{15}\nn\\

2417: z_7&=&{1\over 2}w_6\nn\\

2418: z_{10}&=&-w_3-z_{15}\nn\\

2419: z_{11}&=&w_5-z_{19}\nn\\

2420: z_{12}&=&z_{17}\nn\\

2421: z_{13}&=&2z_{12}\nn\\

2422: z_{13}&=&w_{10}+z_{17}\nn\\

2423: z_{16}&=&-w_8\nn\\

2424: z_{20}&=&0. \label{plirpd} \eea

2425: %

2426: Equations (\ref{plirpa}, \ref{plirpb}, \ref{plirpc}, \ref{plirpd})

2427: above define an overdetermined system of thiry-five equations for

2428: thirty unknowns. It turns out however, that

2429:  only twenty-four of the equations are linearly independent.

2430: Therefore all $w$ and $z$-terms are expressible in terms of six

2431: independent structures, which we can take to be $z_{1-4},

2432: z_7,z_{12}$. Explicitly,

2433: %

2434: \bea

2435: w_1&=&2z_{12}\nn\\

2436: w_2&=&{1\over 2}z_1+z_2+z_4-z_7\nn\\

2437: w_3&=&z_2-z_7\nn\\

2438: w_4&=&-z_1+z_7\nn\\

2439: w_5&=&-{3\over 2}z_1-z_3+z_7\nn\\

2440: w_6&=&2z_7\nn\\

2441: w_7&=&{1\over 2}z_1-z_4\nn\\

2442: w_8&=&z_1+z_2-z_{7}\nn\\

2443: w_9&=&{1\over 2}z_1-z_3\nn\\

2444: w_{10}&=&z_{12} \label{fa} \eea

2445: %

2446: and

2447: %

2448: \bea

2449: z_5&=&-{1\over 2}z_1\nn\\

2450: z_6&=&-{1\over 2}z_1+z_2\nn\\

2451: z_8&=&-z_7\nn\\

2452: z_9&=&{1\over 2}z_1-z_3-z_{4}\nn\\

2453: z_{10}&=&{1\over 2}z_1\nn\\

2454: z_{11}&=&-{1\over 2}z_1-z_2-z_3-z_4+z_7\nn\\

2455: z_{13}&=&2z_{12}\nn\\

2456: z_{14}&=&-2z_{12}\nn\\

2457: z_{15}&=&-{1\over 2}z_1-z_2+z_7\nn\\

2458: z_{16}&=&-z_1-z_2+z_7\nn\\

2459: z_{17}&=&z_{12}\nn\\

2460: z_{18}&=&{3\over 2}z_1-z_7\nn\\

2461: z_{19}&=&-z_1+z_2+z_4\nn\\

2462: z_{20}&=&0. \label{fb} \eea

2463: %

2464: Again we note that the group-theoretic relations (\ref{plirpa},

2465: \ref{plirpb}) are all implied by the partial integration equations

2466: (\ref{plirpc}, \ref{plirpd}).

2467: 

2468: Next we compute the supersymmetric variation of the lagrangian

2469: (\ref{lagrangian}). One finds, up to total derivatives and terms containing $\nab v$,

2470: %

2471: \bea

2472: \delta t_1&=&-4i w_5\nn\\

2473: \delta t_2&=&-4i w_6\nn\\

2474: \delta t_3&=&-4i w_4\nn\\

2475: \delta s_1&=&{1\over 2}(w_3+w_4-z_1+z_2)\nn\\

2476: \delta s_2&=&{1\over 2}(w_2+w_5-z_3+z_{19})\nn\\

2477: \delta s_3&=&{1\over 2}(w_4-w_7+w_8-w_9-w_{10}+z_6-z_9-z_{12})\nn\\

2478: \delta s_4&=&{1\over

2479: 2}(w_1+w_2-w_5+w_6+z_7-z_8-z_{11}-z_{14}+z_{18}). \eea

2480: %

2481: Using (\ref{fa},\ref{fb}) it is straightforward to see that the

2482: vanishing of the variation of the lagrangian implies the following

2483: relations for the coefficients in (\ref{lagrangian}),

2484: %

2485: \bea

2486: A_1&=&{3i\over 4}B_4\nn\\

2487: A_2&=&0\nn\\

2488: A_3&=&-{3i\over 4}B_4\nn\\

2489: B_1&=&0\nn\\

2490: B_2&=&-3B_4\nn\\

2491: B_3&=&2B_4. \eea

2492: %

2493: These results lead to the Lagrangian \eq{alle}.

2494: %

2495: 

2496: 

2497: 

2498: 

2499: 

2500: 

2501: %

2502: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2503: %

2504: % Bibliography

2505: 

2506: 

2507: 

2508: %

2509: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2510: 

2511: \begin{thebibliography}{99}

2512: 

2513: \bibitem{Andreev:1988cb}

2514: O.~D.~Andreev and A.~A.~Tseytlin, ``Partition Function

2515: Representation For The Open Superstring Effective Action:

2516: Cancellation Of Mobius Infinities And Derivative Corrections To

2517: Born-Infeld Lagrangian,'' Nucl.\ Phys.\ B {\bf 311} (1988) 205.

2518: %%CITATION = NUPHA,B311,205;%%

2519: 

2520: %\cite{Bachas:1999um}

2521: \bibitem{Bachas:1999um}

2522: C.~P.~Bachas, P.~Bain and M.~B.~Green, ``Curvature terms in

2523: D-brane actions and their M-theory origin,'' JHEP {\bf 9905}

2524: (1999) 011 [arXiv:hep-th/9903210].

2525: %%CITATION = HEP-TH 9903210;%%

2526: 

2527: %\cite{Fotopoulos:2001pt}

2528: \bibitem{Fotopoulos:2001pt}

2529: A.~Fotopoulos, ``On $\a'^2$ corrections to the D-brane action for

2530: non-geodesic world-volume embeddings,'' JHEP {\bf 0109} (2001) 005

2531: [arXiv:hep-th/0104146].

2532: %%CITATION = HEP-TH 0104146;%%

2533: 

2534: 

2535: 

2536: 

2537: %\cite{Wyllard:2000qe}

2538: \bibitem{Wyllard:2000qe}

2539: N.~Wyllard, ``Derivative corrections to D-brane actions with

2540: constant background  fields,'' Nucl.\ Phys.\ B {\bf 598} (2001)

2541: 247 [arXiv:hep-th/0008125].

2542: %%CITATION = HEP-TH 0008125;%%

2543: 

2544: %\cite{Bilal:2001hb}

2545: \bibitem{Bilal:2001hb}

2546: A.~Bilal, ``Higher-derivative corrections to the non-abelian

2547: Born-Infeld action,'' Nucl.\ Phys.\ B {\bf 618} (2001) 21

2548: [arXiv:hep-th/0106062].

2549: %%CITATION = HEP-TH 0106062;%%

2550: 

2551: 

2552: 

2553: %\cite{Wyllard:2001ye}

2554: \bibitem{Wyllard:2001ye}

2555: N.~Wyllard, ``Derivative corrections to the D-brane Born-Infeld

2556: action: Non-geodesic  embeddings and the Seiberg-Witten map,''

2557: JHEP {\bf 0108} (2001) 027 [arXiv:hep-th/0107185].

2558: 

2559: \bibitem{Koerber:2002zb}

2560: P.~Koerber and A.~Sevrin, ``The non-abelian D-brane effective

2561: action through order $\a'^4$,'' JHEP {\bf 0210} (2002) 046

2562: [arXiv:hep-th/0208044].

2563: 

2564: \bibitem{Lindstrom:1987ps}

2565: U.~Lindstrom, M.~Rocek and P.~van Nieuwenhuizen, ``A Weyl

2566: Invariant Rigid String,'' Phys.\ Lett.\ B {\bf 199} (1987) 219.

2567: %%CITATION = PHLTA,B199,219;%%

2568: 

2569: \bibitem{Lindstrom:1988fr}

2570: U.~Lindstrom and M.~Rocek, ``Bosonic And Spinning Weyl Invariant

2571: Rigid Strings,'' Phys.\ Lett.\ B {\bf 201} (1988) 63.

2572: %%CITATION = PHLTA,B201,63;%%

2573: 

2574: \bibitem{Collinucci:2002gd}

2575: A.~Collinucci, M.~de Roo and M.~G.~Eenink, ``Derivative

2576: corrections in 10-dimensional super-Maxwell theory,'' JHEP {\bf

2577: 0301} (2003) 039 [arXiv:hep-th/0212012].

2578: 

2579: %\cite{Drummond:2003ex}

2580: \bibitem{Drummond:2003ex}

2581: J.~M.~Drummond, P.~J.~Heslop, P.~S.~Howe and S.~F.~Kerstan,

2582: ``Integral invariants in N = 4 SYM and the effective action for

2583: coincident D-branes,'' [arXiv:hep-th/0305202].

2584: %%CITATION = HEP-TH 0305202;%%

2585: 

2586: 

2587: 

2588: %\cite{Ivanov:1990cb}

2589: \bibitem{Ivanov:1990cb}

2590: E.~A.~Ivanov and A.~A.~Kapustnikov, ``Gauge Covariant Wess-Zumino

2591: Actions For Super P-Branes In Superspace,'' Int.\ J.\ Mod.\ Phys.\

2592: A {\bf 7} (1992) 2153.

2593: %%CITATION = IMPAE,A7,2153;%%

2594: 

2595: \bibitem{Gauntlett:1991dw}

2596: J.~P.~Gauntlett, ``A kappa symmetry calculus for

2597: superparticles,'' Phys.\ Lett.\ B {\bf 272} (1991) 25

2598: [arXiv:hep-th/9109039].

2599: %%CITATION = HEP-TH 9109039;%%

2600: 

2601: %\cite{Ivanov:1991ub}

2602: \bibitem{Ivanov:1991ub}

2603: E.~A.~Ivanov and A.~A.~Kapustnikov, ``Towards A Tensor Calculus

2604: For Kappa Supersymmetry,'' Phys.\ Lett.\ B {\bf 267} (1991) 175.

2605: %%CITATION = PHLTA,B267,175;%%

2606: 

2607: 

2608: 

2609: 

2610: \bibitem{Curtright:1987mr}

2611: T.~Curtright and P.~van Nieuwenhuizen, ``Supersprings,'' Nucl.\

2612: Phys.\ B {\bf 294} (1987) 125.

2613: %%CITATION = NUPHA,B294,125;%%

2614: 

2615: \bibitem{Sorokin:2001av}

2616: D.~P.~Sorokin, ``Coincident (super)-Dp-branes of codimension

2617: one,'' JHEP {\bf 0108} (2001) 022 [arXiv:hep-th/0106212].

2618: 

2619: %\cite{Panda:2003dj}

2620: \bibitem{Panda:2003dj}

2621: S.~Panda and D.~Sorokin, ``Supersymmetric and kappa-invariant

2622: coincident D0-branes,'' JHEP {\bf 0302} (2003) 055

2623: [arXiv:hep-th/0301065].

2624: %%CITATION = HEP-TH 0301065;%%

2625: 

2626: 

2627: 

2628: \bibitem{Howe:2001wc}

2629: P.~S.~Howe and U.~Lindstrom, ``Kappa-symmetric higher derivative

2630: terms in brane actions,'' Class.\ Quant.\ Grav.\  {\bf 19} (2002)

2631: 2813 [arXiv:hep-th/0111036].

2632: 

2633: \bibitem{Drummond:2002kg}

2634: J.~M.~Drummond, P.~S.~Howe and U.~Lindstrom, ``Kappa-symmetric

2635: non-Abelian Born-Infeld actions in three dimensions,'' Class.\

2636: Quant.\ Grav.\  {\bf 19} (2002) 6477 [arXiv:hep-th/0206148].

2637: 

2638: 

2639: 

2640: 

2641: 

2642: \bibitem{stvz}

2643: D. Sorokin, V. Tkach and D.V. Volkov, ``Superparticles, twistors

2644: and Siegel symmetry,'' Mod. Phys. Lett. {\bf A4} (1989) 901; D.

2645: Sorokin, V. Tkach, D.V. Volkov and A. Zheltukhin, ``From the

2646: superparticle Siegel symmetry to the spinning particle proper time

2647: supersymmetry,'' Phys. Lett. {\bf B216} (1989) 302; D.V. Volkov

2648: and A. Zheltukhin, ``Extension of the Penrose representation and

2649: its use to describe supersymmetric models,'' Sov. Phys. JETP Lett.

2650: {\bf 48} (1988) 63-66.

2651: 

2652: \bibitem{s}

2653: D. Sorokin, ``Superbranes and superembeddings,'' Phys. Report {\bf

2654: 239} (2000) 1.

2655: 

2656: \bibitem{Howe:1996mx}

2657: P.~S.~Howe and E.~Sezgin, ``Superbranes,'' Phys.\ Lett.\ B {\bf

2658: 390} (1997) 133 [arXiv:hep-th/9607227].

2659: 

2660: 

2661: 

2662: 

2663: \bibitem{Howe:1996yn}

2664: P.~S.~Howe and E.~Sezgin, ``D = 11, p = 5,'' Phys.\ Lett.\ B {\bf

2665: 394} (1997) 62 [arXiv:hep-th/9611008].

2666: 

2667: \bibitem{cnta}

2668: M.~Cederwall, B.~E.~Nilsson and D.~Tsimpis, ``The structure of

2669: maximally supersymmetric Yang-Mills theory: Constraining

2670: higher-order corrections,'' JHEP {\bf 0106} (2001) 034

2671: [arXiv:hep-th/0102009].

2672: %%CITATION = HEP-TH 0102009;%%

2673: 

2674: \bibitem{cntb}

2675: M.~Cederwall, B.~E.~Nilsson and D.~Tsimpis, ``D = 10

2676: super-Yang-Mills at $O(\alpha'^2)$,'' JHEP {\bf 0107} (2001) 042

2677: [arXiv:hep-th/0104236].

2678: %%CITATION = HEP-TH 0104236;%%

2679: 

2680: 

2681: \bibitem{cntc}

2682: M.~Cederwall, B.~E.~Nilsson and D.~Tsimpis, ``Spinorial cohomology

2683: and maximally supersymmetric theories,'' JHEP {\bf 0202} (2002)

2684: 009 [arXiv:hep-th/0110069].

2685: 

2686: 

2687: \bibitem{cntd}

2688: M.~Cederwall, B.~E.~Nilsson and D.~Tsimpis, ``Spinorial cohomology

2689: of abelian d = 10 super-Yang-Mills at  $O(\alpha'^3)$,'' JHEP {\bf

2690: 0211} (2002) 023 [arXiv:hep-th/0205165].

2691: %%CITATION = HEP-TH 0205165;%%

2692: 

2693: 

2694: %\bibitem{cgnn}

2695: %M.~Cederwall, U.~Gran, M.~Nielsen and B.~E.~Nilsson, ``Manifestly

2696: %supersymmetric M-theory,'' JHEP {\bf 0010} (2000) 041

2697: %[arXiv:hep-th/0007035];

2698: %%CITATION = HEP-TH 0007035;%%

2699: %``Generalised 11-dimensional supergravity,''

2700: %[arXiv:hep-th/0010042].

2701: 

2702: %\cite{Howe:2003cy}

2703: \bibitem{Howe:2003cy}

2704: P.~S.~Howe and D.~Tsimpis, ``On higher-order corrections in M

2705: theory,'' [arXiv:hep-th/0305129].

2706: %%CITATION = HEP-TH 0305129;%%

2707: 

2708: 

2709: %\cite{Gates:1985vk}

2710: \bibitem{Gates:1985vk}

2711: S.~J.~Gates and H.~Nishino, ``D = 2 Superfield Supergravity, Local

2712: (Supersymmetry)$^2$ And Nonlinear Sigma Models,'' Class.\ Quant.\

2713: Grav.\  {\bf 3} (1986) 391.

2714: %%CITATION = CQGRD,3,391;%%

2715: 

2716: %\cite{Bandos:1995zw}

2717: \bibitem{Bandos:1995zw}

2718: I.~A.~Bandos, D.~P.~Sorokin, M.~Tonin, P.~Pasti and D.~V.~Volkov,

2719: ``Superstrings and supermembranes in the doubly supersymmetric

2720: geometrical approach,'' Nucl.\ Phys.\ B {\bf 446} (1995) 79

2721: [arXiv:hep-th/9501113].

2722: %%CITATION = HEP-TH 9501113;%%

2723: 

2724: 

2725: 

2726: %\cite{Bergshoeff:1987cm}

2727: \bibitem{Bergshoeff:1987cm}

2728: E.~Bergshoeff, E.~Sezgin and P.~K.~Townsend, ``Supermembranes And

2729: Eleven-Dimensional Supergravity,'' Phys.\ Lett.\ B {\bf 189}

2730: (1987) 75.

2731: %%CITATION = PHLTA,B189,75;%%

2732: 

2733: \bibitem{Berkovits:2000fe}

2734: N.~Berkovits,

2735: ``Super-Poincare covariant quantization of the superstring,''

2736: JHEP {\bf 0004} (2000) 018

2737: [arXiv:hep-th/0001035].

2738: %%CITATION = HEP-TH 0001035;%%

2739: 

2740: 

2741: 

2742: 

2743: 

2744: \bibitem{Berkovits:2001rb}

2745: N.~Berkovits, ``Covariant quantization of the superparticle using

2746: pure spinors,'' JHEP {\bf 0109} (2001) 016 [arXiv:hep-th/0105050].

2747: %%CITATION = HEP-TH 0105050;%%

2748: 

2749: \bibitem{Howe:mf}

2750: P.~S.~Howe, ``Pure Spinors Lines In Superspace And Ten-Dimensional

2751: Supersymmetric Theories,'' Phys.\ Lett.\ B {\bf 258} (1991) 141

2752: [Addendum-ibid.\ B {\bf 259} (1991) 511].

2753: %%CITATION = PHLTA,B258,141;%%

2754: 

2755: %\cite{Howe:1991bx}

2756: \bibitem{Howe:1991bx}

2757: P.~S.~Howe, ``Pure spinors, function superspaces and supergravity

2758: theories in ten-dimensions and eleven-dimensions,'' Phys.\ Lett.\

2759: B {\bf 273} (1991) 90.

2760: %%CITATION = PHLTA,B273,90;%%

2761: 

2762: 

2763: 

2764: 

2765: \bibitem{D'Auria:1982pm}

2766: R.~D'Auria, P.~Fre, P.~K.~Townsend and P.~van Nieuwenhuizen,

2767: ``Invariance Of Actions, Rheonomy And The New Minimal N=1

2768: Supergravity In The Group Manifold Approach,'' Annals Phys.\  {\bf

2769: 155} (1984) 423.

2770: %%CITATION = APNYA,155,423;%%

2771: 

2772: \bibitem{ggks}

2773: S.J. Gates, M. Grisaru, M. Knutt-Wehlau and W. Siegel, ``Component

2774: actions from curved superspace: normal coordinates and

2775: ectoplasm,'' Phys.Lett. {\bf B421} (1998) 203-210, [arXiv:hep-th/9711151].

2776: 

2777: 

2778: \bibitem{hrs} P.S. Howe, O. Raetzel and E. Sezgin,

2779:  ``On brane actions and superembeddings,''

2780:  JHEP {\bf 9808} (1998) 011,

2781:  [arXiv:hep-th/9804051].

2782: 

2783: 

2784: 

2785: %\cite{Green:1996dd}

2786: \bibitem{Green:1996dd}

2787: M.~B.~Green, J.~A.~Harvey and G.~W.~Moore, ``I-brane inflow and

2788: anomalous couplings on D-branes,'' Class.\ Quant.\ Grav.\  {\bf

2789: 14} (1997) 47 [arXiv:hep-th/9605033].

2790: %%CITATION = HEP-TH 9605033;%%

2791: 

2792: %\cite{Cheung:1997az}

2793: \bibitem{Cheung:1997az}

2794: Y.~K.~Cheung and Z.~Yin, ``Anomalies, branes, and currents,''

2795: Nucl.\ Phys.\ B {\bf 517} (1998) 69 [arXiv:hep-th/9710206].

2796: %%CITATION = HEP-TH 9710206;%%

2797: 

2798: 

2799: \bibitem{minmo}

2800: R. Minasian and G.W. Moore, ``K theory and Ramond-Ramond charge,''

2801: JHEP {\bf 9711} (1997) 002 [arXiv:hep-th/9710230].

2802: 

2803: 

2804: 

2805: 

2806: 

2807: 

2808: 

2809: \end{thebibliography}

2810: \end{document}

2811: 

2812: 

2813: 

2814: 

2815: 

2816: