1: %version December, 2004

2: %

3: \documentclass[12pt]{book}

4: \usepackage{amsmath,graphicx} 

5: \usepackage{multirow} 

6: \usepackage{amsfonts}

7: \usepackage{amssymb}

8: \usepackage{fancyhdr}

9: %\usepackage{cite}

10: %\usepackage{showkeys}

11: \oddsidemargin 15pt

12: \evensidemargin 0pt

13: \textwidth 6in  

14: \textheight 9in 

15: 

16: \pagestyle{fancy}

17: \renewcommand{\chaptermark}[1]{\markboth{#1}{}} 

18: \renewcommand{\sectionmark}[1]{\markright{\thesection\ #1}} 

19: \fancyhf{} % delete current setting for header and footer 

20: \fancyhead[LE,RO]{\bfseries\thepage} 

21: \fancyhead[LO]{\bfseries\rightmark} 

22: \fancyhead[RE]{\bfseries\leftmark} 

23: \renewcommand{\headrulewidth}{0.5pt} 

24: \renewcommand{\footrulewidth}{0pt} 

25: \addtolength{\headheight}{4.5pt} % make space for the rule 

26: \fancypagestyle{plain}{\fancyhead{} % get rid of headers on plain pages 

27: \renewcommand{\headrulewidth}{0pt} % and the line 

28: }

29: \parskip 3pt plus 1pt 

30: 

31: \def\hybrid{\topmargin -20pt    \oddsidemargin 0pt

32:         \headheight 0pt \headsep 15pt

33: %       \textwidth 6.5in        % US paper

34: %       \textheight 9in         % US paper

35:         \textwidth 6.25in       % A4 paper

36:         \textheight 8.6 in       % A4 paper

37:         \marginparwidth .875in

38:         \parskip 5pt plus 1pt   \jot = 1.5ex}

39: %       The default is set to be hybrid

40: \hybrid

41: 

42: \numberwithin{equation}{chapter}

43: \numberwithin{table}{section}\setlength{\multlinegap}{25pt}

44: 

45: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

46: % abbreviate  environments

47: \newcommand{\beq}{\begin{equation}}

48: \newcommand{\eeq}{\end{equation}}

49: %\newcommand{\bi}{\begin{itemize}}

50: %\newcommand{\ei}{\end{itemize}}

51: \newcommand{\bea}{\begin{eqnarray}}

52: \newcommand{\eea}{\end{eqnarray}}

53: \newcommand{\ba}{\begin{array}}

54: \newcommand{\ea}{\end{array}}

55: \newcommand{\bt}{\begin{tabular}}

56: \newcommand{\et}{\end{tabular}}

57: \newcommand{\bc}{\begin{center}}

58: \newcommand{\ec}{\end{center}}

59: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

60: % abbreviate Greek

61: \newcommand{\ax}{\alpha}

62: \newcommand{\bx}{\beta}

63: \newcommand{\cx}{\gamma}

64: \newcommand{\dx}{\delta}

65: \newcommand{\ox}{\omega}

66: \newcommand{\lx}{\lambda}

67: \newcommand{\ab}{\bar\alpha}

68: \newcommand{\bb}{\bar\beta}

69: \newcommand{\cb}{\bar\gamma}

70: \newcommand{\db}{\bar\delta}

71: \newcommand{\Sx}{\Sigma}

72: \newcommand{\Lx}{\Lambda}

73: \newcommand{\Ox}{\Omega}

74: \newcommand{\Dx}{\Delta}

75: \newcommand{\Gx}{\Gamma}

76: \newcommand{\Oxb}{\bar{\Omega}}

77: %

78: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

79: % Cal

80: \newcommand{\cC}{\mathcal{C}}

81: \newcommand{\cD}{\mathcal{D}}

82: \newcommand{\cL}{\mathcal{L}}

83: \newcommand{\cS}{\mathcal{S}}

84: \newcommand{\cK}{\mathcal{K}}

85: \newcommand{\cN}{\mathcal{N}}

86: \newcommand{\cW}{\mathcal{W}}

87: \newcommand{\cG}{\mathcal{G}}

88: \newcommand{\cA}{\mathcal{A}}

89: \newcommand{\cH}{\mathcal{H}}

90: \newcommand{\cB}{\mathcal{B}}

91: \newcommand{\cF}{\mathcal{F}}

92: \newcommand{\cI}{\mathcal{I}}

93: \newcommand{\cJ}{\mathcal{J}}

94: \newcommand{\cR}{\mathcal{R}}

95: \newcommand{\Ac}{\mathcal{A}}

96: \newcommand{\cV}{\mathcal{V}}

97: \newcommand{\Bc}{\mathcal{B}} 

98: \newcommand{\KK}{\mathcal{K}}

99: \newcommand{\MM}{\mathcal{M}}

100: \newcommand{\cM}{\mathcal M}

101: \newcommand{\cQ}{\mathcal Q}

102: \newcommand{\cO}{\mathcal{O}}

103: \newcommand{\OO}{\mathcal{O}}

104: \newcommand{\Vw}{{\mathcal K}_w\vphantom{{\mathcal V}_w}}

105: \newcommand{\Gw}{{\mathcal G}_w\vphantom{{\mathcal G}_w}}

106: %\newcommand{\Vw}{\mathcal K}

107: %\newcommand{\Gw}{\mathcal G}

108: 

109: \newcommand{\bfA}{\mathbf{A}}

110: \newcommand{\bfC}{\mathbf{C}}

111: \newcommand{\bfG}{\mathbf{G}}

112: 

113: \newcommand{\IF}{\text{Im}\, \mathcal{F}}

114: \newcommand{\IM}{\text{Im}\, \mathcal{M}}

115: \newcommand{\RF}{\text{Re}\, \mathcal{F}}

116: \newcommand{\RM}{\text{Re}\, \mathcal{M}}

117: \newcommand{\I}{\text{Im}}

118: \newcommand{\R}{\text{Re}}

119: 

120: \newcommand{\Kcs}{K^{\text{cs}}}

121: \newcommand{\Kks}{K^{\text{ks}}}

122: \newcommand{\pev}{{\varphi^{ev}}}

123: \newcommand{\pevb}{{\bar \varphi^{ev}}}

124: \newcommand{\podd}{{\varphi^{odd}}}

125: \newcommand{\poddb}{{\bar \varphi^{odd}}}

126: \newcommand{\fuh}{{\mathcal{\hat U}}}

127: \newcommand{\fu}{{\mathcal{U}}}

128: \newcommand{\fe}{{\mathcal{E}}}

129: \newcommand{\feh}{{\mathcal{\hat E}}}

130: \newcommand{\fa}{{\mathcal{ A}}}

131: \newcommand{\volume}{\text{{\small} vol}\, }

132: 

133: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

134: % indices

135: \newcommand{\bi}{{\bar \imath}}

136: \newcommand{\ib}{{\bar\imath }}

137: \newcommand{\jb}{{\bar \jmath }}

138: \newcommand{\bj}{{\bar\jmath}}

139: \newcommand{\bk}{\bar{k}}

140: \newcommand{\bl}{\bar{l}}

141: \newcommand{\Kh}{{\hat{K}}}

142: \newcommand{\Lh}{{\hat{L}}}

143: \newcommand{\Ah}{{\hat{A}}}

144: \newcommand{\Bh}{{\hat{B}}}

145: \newcommand{\Ch}{{\hat{C}}}

146: \newcommand{\Dh}{{\hat{D}}}

147: \newcommand{\Mh}{{\hat{M}}}

148: \newcommand{\Nh}{{\hat{N}}}

149: \newcommand{\kh}{{\hat{k}}}

150: \newcommand{\ah}{{\hat{a}}}

151: \newcommand{\bh}{{\hat{b}}}

152: \newcommand{\ch}{{\hat{c}}}

153: \newcommand{\lh}{{\hat{l}}}

154: \newcommand{\mh}{{\hat{m}}}

155: \newcommand{\nh}{{\hat{n}}}

156: \newcommand{\Mext}{{\mathbb{M}^{3,1}}}

157: \newcommand{\Mint}{{Y}}

158: 

159: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

160: % Dan's macros

161: \DeclareMathOperator{\SU}{\mathit{SU}}

162: \DeclareMathOperator{\SO}{\mathit{SO}}

163: \DeclareMathOperator{\Symp}{\mathit{Sp}}

164: \DeclareMathOperator{\Spin}{\mathit{Spin}}

165: \DeclareMathOperator{\so}{\mathit{so}}

166: \DeclareMathOperator{\su}{\mathit{su}}

167: \DeclareMathOperator{\symp}{\mathit{sp}}

168: \DeclareMathOperator{\spin}{\mathit{spin}}

169: \DeclareMathOperator{\GL}{\mathit{GL}}

170: \DeclareMathOperator{\SL}{\mathit{SL}}

171: 

172: \DeclareMathOperator{\vol}{vol}

173: 

174: \newcommand{\rep}[1]{\mathbf{#1}}

175: 

176: \newcommand{\dd}{d}

177: \newcommand{\ii}{\mathrm{i}}

178: 

179: \newcommand{\bbZ}{\mathbb{Z}}

180: \newcommand{\bbR}{\mathbb{R}}

181: \newcommand{\bbC}{\mathbb{C}}

182: \newcommand{\bbP}{\mathbb{P}}

183: \newcommand{\bbF}{\mathbb{F}}

184: 

185: 

186: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

187: % misc

188: \newcommand{\CY}{Calabi--Yau}

189: \newcommand{\half}{\frac12}

190: \newcommand{\quart}{\frac14}

191: \newcommand{\nn}{\nonumber}

192: %\def\Mnote#1{{\bf[MG: #1]}}

193: \def\Tnote#1{{\bf[TG: #1]}}

194: %\def\Hnote#1{{\bf[HJ: #1]}}

195: \def\Jnote#1{{\bf[JL: #1]}}

196: % \newcommand{\IM}{\textrm{Im} \,}

197: \newcommand{\RE}{\textrm{Re} \,}

198: \newcommand{\?}{{\bf [??]}}

199: %\newcommand{\xx}{{\bf [xx]}}

200: \newcommand{\addref}{{\bf [add ref]}}

201: \newcommand{\cref}{{\bf [check ref]}}

202: \newcommand{\chec}{{\bf [check]}}

203: \newcommand{\park}{{\bf [the following text/formulas are just being parked here]}}

204: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

205: % Jan's macros

206: \newcommand{\M}{M}

207: \newcommand{\N}{\Theta}

208: \newcommand{\Weff}{W^{\rm (eff)}}

209: \newcommand{\Weffb}{\bar W^{\rm (eff)}}

210: \newcommand{\Y}{Y}

211: \newcommand{\G}{\mathcal{I}}

212: \newcommand{\CHI}{\mathcal{I}}

213: \newcommand{\f}{}

214: \newcommand{\Jc}{J_{\rm c}}

215: \newcommand{\Omegac}{\Omega_{\rm c}}

216: \newcommand{\cc}{c}

217: \newcommand{\CC}{C}

218: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

219: % Mariana's macros

220: \newcommand{\hW}{\hat{W}}

221: \newcommand{\hK}{\hat{K}}

222: \newcommand{\hWb}{\hat{\bar W}}

223: \newcommand{\hphi}{{\phi}}

224: \newcommand{\Gt}{G^{(3)}\vphantom{G}} 

225: \newcommand{\Gtb}{\bar{G}^{(3)}\vphantom{\bar G}} 

226: \newcommand{\ha}{\hat{a}}

227: \newcommand{\hb}{\hat{b}}

228: \newcommand{\hab}{\hat{ a}}

229: \newcommand{\hbb}{\hat{ b}}

230: \newcommand{\cha}{\chi_{\hat{a}}}

231: \newcommand{\chab}{\bar{\chi}_{\hat{a}}}

232: \newcommand{\chb}{\chi_{\hat{b}}}

233: \newcommand{\chbb}{\bar{\chi}_{\hat{b}}}

234: \newcommand{\Dth}{\rm (D3)}

235: \newcommand{\eff}{\rm (eff)}

236: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

237: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

238: % Hans' macros - Beginning

239: \newcommand{\tr}{\mathrm{Tr}\:}

240: \newcommand{\id}{\mathbf{1}}

241: \newcommand{\com}[2]{\big[ {#1},{#2} \big]}

242: \newcommand{\lie}[2]{\left[ {#1},{#2} \right]}

243: \newcommand{\ins}[1]{\mathrm{i}_{#1}}

244: \newcommand{\D}{\mathrm{D}}        

245: \newcommand{\Kw}{\mathcal{K}_w}    % Warped CY volume

246: \newcommand{\Em}{\varphi}          % Embedding map of the world-volume

247: \newcommand{\WV}{\mathcal{W}}      % Worldvolume

248: \newcommand{\FD}{F}

249: \newcommand{\FA}{F_\mathrm{A}}%\vphantom{F_\mathrm{A}}}  % Field strength - Brane

250: \newcommand{\norm}[1]{\lVert #1\rVert}

251: \newcommand{\Riem}[4]{R_{#1\hphantom{#2}#3#4}^{\hphantom{#1}#2}}

252: % Hans' macros - End

253: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

254: 

255: 

256: \begin{document}

257: 

258: 

259: \begin{titlepage}

260: \begin{center}

261: 

262: \hfill hep-th/0507153\\

263: \hfill ZMP-HH/05-16\\

264: \vskip 1.5cm

265: 

266: {\large \bf The effective action of type II Calabi-Yau orientifolds}

267: \footnote{%

268: This article is based on the Ph.D.~thesis of the author.}\\

269: 

270: 

271: \vskip 1cm

272: 

273: 

274: 

275: {\bf Thomas W.\ Grimm }\footnote{%

276: From September 1, 2005: Department of Physics, University of Wisconsin, Madison WI 53706, USA }  \\

277: \vskip 0.5cm

278: 

279: {\em II. Institut f{\"u}r Theoretische Physik\\

280: Universit{\"a}t Hamburg, Luruper Chaussee 149\\

281:  D-22761 Hamburg, Germany}\\

282: \vskip 5pt

283: and

284: \vskip 5pt

285: {\em Zentrum f\"ur Mathematische Physik \\

286: Universit\"at Hamburg, 

287: Bundesstrasse 55\\

288: D-20146 Hamburg, Germany}\\

289: 

290: 

291: \vskip 10pt

292: 

293:  {\tt  thomas.grimm@desy.de} \\

294: 

295: \end{center}

296: 

297: \vskip .5cm

298: 

299: \begin{center} {\bf ABSTRACT } \end{center}

300: %\vspace{-2mm}

301: 

302: \noindent

303: 

304: This article first reviews the calculation of 

305: the $N = 1$ effective action for generic type IIA and type IIB Calabi-Yau orientifolds 

306: in the presence of background fluxes by using a Kaluza-Klein reduction. The K\"ahler potential, 

307: the gauge kinetic functions and the flux-induced superpotential are 

308: determined in terms of geometrical data of the Calabi-Yau orientifold and the background fluxes. 

309: As a new result, it is shown that the chiral description directly relates to Hitchin's generalized geometry 

310: encoded by special odd and even forms on a threefold, whereas a dual formulation with several 

311: linear multiplets makes contact to the underlying $N=2$ special geometry.  

312: In type IIB setups, the flux-potentials can be expressed 

313: in terms of superpotentials, D-terms and, generically, a 

314: massive linear multiplet. The type IIA superpotential depends on all geometric 

315: moduli of the theory. It is reviewed, how type IIA orientifolds arise as a special 

316: limit of M-theory compactified on specific $G_2$ manifolds by matching the effective actions. In a similar 

317: spirit type IIB orientifolds are shown to descend from F-theory 

318: on a specific class of Calabi-Yau fourfolds.

319: In addition, mirror symmetry for Calabi-Yau orientifolds

320: is briefly discussed and it is shown that the $N = 1$ chiral coordinates linearize the appropriate 

321: instanton actions.

322: 

323: 

324: \vfill

325: 

326: 

327: \end{titlepage}

328: 

329: 

330: \vspace*{10cm}

331: \begin{center}

332:    {\bf \large Acknowledgments}

333: \end{center}

334: This article is based on my Ph.D.~thesis. 

335: First of all I would like to express my deep gratitude to my supervisor Prof.~Jan Louis 

336: for his continuous support, expert advises and encouragement.  

337: The collaboration with Mariana Gra\~na, Hans Jockers, Frederic Schuller and Mattias Wohlfarth

338: was very enjoyable and fruitful. I esspecially like to thank my office

339: mates Iman Benmachiche, Olaf Hohm, Hans Jockers, Andrei Micu and Anke Knauf for providing a very 

340: delighting athmosphere and the numerous discussions about physiscs 

341: and beyond. I am also indebted to David Cerde\~no, Vincente Cort\'es, Frederik Denef, Sergei Gukov, Henning Samtleben, 

342: Sakura Sch\"afer-Nameki, Shamit Kachru, Boris~K\"ors, Paolo Merlatti, Thorsten Pr\"ustel, 

343: Waldemar Schulgin, Silvia Vaul\`a and 

344: Martin Weidner for various discussions and correspondence. I am grateful to my lovely girlfriend,

345: for supporting me through the last years. 

346: 

347: \vspace{.2cm} 

348: This work is supported by the DFG -- The German Science Foundation, 

349: the DAAD~--~the German Academic Exchange Service, and the European RTN Program MRTN-CT-2004-503369.

350: 

351: 

352: 

353: \thispagestyle{plain}

354: 

355: 

356: \tableofcontents

357: 

358: \renewcommand{\thetable}{\arabic{chapter}.\arabic{table}}

359: 

360: \chapter{Introduction}

361: 

362: The Standard Model of particle physics extended by massive neutrinos 

363: has been tested to a very high precision and is believed 

364: to correctly describe the known elementary particles and their interactions.

365: Experimentally, the only missing ingredient is the scalar Higgs particle, which gives 

366: masses to the leptons and quarks, once it acquires a vacuum expectation 

367: value. The Standard Model provides a realistic model of a renormalizable

368: gauge theory. Despite its impressive success there are also various theoretical 

369: drawbacks, such as the large number of free parameters, the hierarchy and naturalness problem as

370: well as the missing unification with gravity. These indicate that it cannot be viewed 

371: as a fundamental theory, but rather should arise as an effective description. 

372: 

373: 

374: A natural extension of the Standard Model is provided by supersymmetry, which serves as a 

375: fundamental symmetry between bosons and fermions. Supersymmetry predicts a superpartner 

376: for all known particles and thus basically doubles the particle content of the theory.

377: However, none of the superpartners was ever detected in an accelerator experiment, which 

378: implies that supersymmetry is appearing in its broken phase. The supersymmetric Standard 

379: Model solves some of the problems of the Standard Model \cite{revSusy}. 

380: Even in its (softly) broken phase it forbids large quantum corrections to scalar masses.

381: This allows the Higgs mass to remain to be of order the weak scale also in a theory 

382: with a higher mass scale. Furthermore, assuming the supersymmetric Standard Model to be valid

383: up to very high scales, the renormalization group flow predicts a unification of all

384: three gauge-couplings. This supports the idea of an underlying theory 

385: relevant beyond the Standard Model scales. However, it remains to unify 

386: these extensions with gravity.

387: 

388: On the other hand, we know that General Relativity  

389: links the geometry of spacetime with the distribution of the matter densities. Einsteins 

390: theory is very different in nature. It is a classical theory which is 

391: hard to quantize due to its ultra-violet divergences (see however \cite{LQG}). 

392: This fact constraints its range of validity to 

393: phenomena, where quantum effects are of negligible importance. However, there is no 

394: experimental evidence which contradicts large scale predictions based on General 

395: Relativity.

396: 

397: Facing these facts General Relativity and the Standard Model seems to be incompatible, 

398: in the sense that neither of them allows to naturally adapt the other. This becomes 

399: important in regimes where both theories have to be applied in order to describe the 

400: correct physics. Early time cosmology or physics of black holes are only 

401: two regimes where the interplay of quantum and gravitational effects become important. 

402: To nevertheless approach this theoretically interesting questions one might 

403: hope for a fundamental quantum theory combining the Standard Model and 

404: General Relativity. Until now one does not know what this unifying theory 

405: is, but one has at least one possible candidate. This theory is 

406: known as String Theory, which was studied intensively from various directions in

407: the last thirty years. A comprehensive introduction to the 

408: subject can be found in \cite{GSWbook,JPbook,Zwiebach}.

409: 

410: Perturbative String Theory is a quantum theory of one-dimensional extended objects which 

411: replace the ordinary point particles. These fundamental strings can appear in various vibrational 

412: modes which at low energies are identified with different particles. The characteristic

413: length of the string is $\sqrt{\alpha'}$, where $\alpha'$ is the Regge slope.

414: Hence, the extended nature of the strings only becomes apparent close to the string 

415: scale $1/\sqrt{\alpha'}$. The string spectrum naturally includes a mode corresponding to the graviton. 

416: This implies that Sting Theory indeed includes gravity and as we will further discuss below 

417: reduces to Einsteins theory at low energies. It most likely provides a renormalizable 

418: quantum theory of gravity around a given background. It avoids the ultra-violet divergences of graviton 

419: scattering amplitudes in field theory by smearing out the location of the interactions. 

420: 

421: 

422: The extended nature of the fundamental strings poses strong consistency constraints on the 

423: theory. Non-tachyonic String Theories (Superstring Theories) require space-time supersymmetry

424: and predict a ten-dimensional space-time at weak coupling. 

425: Altogether there are only five consistent String Theories, which are called 

426: type IIA, type IIB, heterotic $SO(32)$ and $E_8 \times E_8$ and type I. These theories are connected

427: by various dualities and one may eventually hope to unify all of them into one fundamental theory \cite{Dual, JPbook}.

428: 

429: As striking a proper formulation of such a fundamental theory might be, much of its uniqueness and beauty could 

430: be spoiled in attempting to extract four-dimensional results. This is equally true for the five String Theories 

431: formulated in ten dimensions. One approach to reduce String Theory from ten to four space-time dimensions 

432: is compactification on a geometric background of the form $\Mext \times \Mint$. $\Mext$ is 

433: identified with our four-dimensional world, while $\Mint$ is chosen to be small and compact, 

434: such that these six additional dimensions are not visible in experiments. This however induces a 

435: high amount of ambiguity, since String Theory allows for various consistent choices of $\Mint$.

436: Eventually one would hope to find a String Field Theory formulated in ten dimensions, which 

437: resolves this ambiguity and dynamically chooses a certain background. %also describes all sting states at once.  

438: However, such a theory is still lacking and one is forced to take a sideway 

439: to find and explore consistent string backgrounds. 

440: 

441: For a given background, the ten-dimensional theory can then be reduced to four dimensions by a 

442: Kaluza-Klein compactification \cite{KaluzaKlein} (for a review on Kaluza-Klein reduction see 

443: e.g.\ \cite{KK-review}). This amounts to expanding 

444: the fields into modes of $\Mint$ and results in a full tower of Kaluza-Klein modes for 

445: each of the string excitations. Additionally there are winding modes corresponding to 

446: strings winding around cycles in $\Mint$. 

447: Generically it is hard and phenomenologically not interesting 

448: to deal with these infinite towers of modes and an effective description is needed.

449: 

450: 

451: In order to extract an effective formulation one may first integrate out the 

452: massive string excitations with masses of order $1/\sqrt{\alpha'}$. This is possible 

453: due to the fact that the string scale $1/\sqrt{\alpha'}$ is usually set to be of order the 

454: Planck scale such that gravity couples with Newtonian strength. 

455: In the point-particle limit $\alpha' \rightarrow 0$ the effective theory 

456: describing the massless string modes is a supergravity theory (see e.g.~\cite{GSWbook,JPbook}). 

457: It can be constructed by calculating string scattering amplitudes for massless states.

458: One then infers an effective action for these fields encoding the same 

459: tree level scattering vertices. An example is the three-graviton scattering amplitude

460: in String Theory, which in an effective description can be equivalently obtained from the 

461: ten-dimensional Einstein-Hilbert term. 

462: Repeating the same reasoning for all other massless string modes 

463: yields a ten-dimensional supergravity theory for each of the five String Theories.

464: 

465: In a similar spirit one can also extract an effective Kaluza-Klein theory. For a 

466: compact internal manifold $\Mint$ the first massive Kaluza-Klein 

467: modes have a mass of order $1/R$, where $R$ is the `average radius' of $\Mint$. Hence, 

468: choosing $\Mint$ to be sufficiently small these modes become heavy and can be integrated 

469: out. On the other hand, $\Mint$ has to be large enough that winding modes of length $\sqrt{\alpha'}$ 

470: can be discarded. Together for $p$ being the characteristic momentum of the lower-dimensional fields 

471: an effective description of the massless modes is valid in the regime $1/p \gg R \gtrsim \sqrt{\alpha'}$.

472:  

473: 

474: The structure of the four-dimensional theory obtained by such a reduction highly depends

475: on the chosen internal manifold. The properties of $\Mint$ determine the amount of supersymmetry

476: and the gauge-groups of the lower-dimensional theory. Generically one insists that $\Mint$ 

477: preserves some of the ten-dimensional supersymmetries. This is due to the fact that string

478: theory on supersymmetric backgrounds is under much better control and various consistency 

479: conditions are automatically satisfied. It turns out that looking for a supersymmetric theory 

480: with a four-dimensional Minkowski background the internal manifold has to be a  

481: Calabi-Yau manifold \cite{Greene}. From a phenomenological point of view the resulting low-energy 

482: supergravity theories need to include gauged matter fields filling the spectrum of the desired 

483: gauge theory such as the supersymmetric Standard Model. However, parameters like the size and 

484: shape of the compact space appear as massless neutral scalar fields in four dimensions. 

485: They label the continuous degeneracy of consistent backgrounds 

486: $\Mint$ and are generically not driven to any particular value; 

487: they are moduli of the theory. In a Standard Model-like vacuum these moduli have to be massive, such 

488: that they are not dynamical in the low-energy effective action. Therefore one needs to identify 

489: a mechanism in String Theory which induces a potential for these scalars. As it is 

490: well-known for supergravity theories this potential can provide at the same time

491: a way to spontaneously break supersymmetry. 

492: 

493: To generate a moduli-dependent potential in a consisted String Theory setup is a non-trivial task 

494: and requires further refinements of the standard compactifications. Recently, much effort was

495: made to establish controllable mechanisms to stabilize moduli fields in type II 

496: String Theory. The three most popular approaches are the inclusion of background 

497: fluxes \cite{Strominger1}--\cite{TGL2}, instanton corrections \cite{Witten,HM,non-pert,Curio} and 

498: gaugino condensates \cite{gaugino,non-pert}.

499: This raised the hope to find examples of string vacua with all moduli being 

500: fixed \cite{KKLT,KachruK,TGL2,DSFGK}. Moreover, phenomenologically interesting scenarios 

501: for particle physics and cosmology can be constructed within these setups \cite{reviewPP,reviewcosmo}. 

502: 

503: In contrast to $E_8 \times E_8$ and $SO(32)$ heterotic String Theory and type I 

504: strings both Type II String Theory do not consist of non-Abelian gauge-groups 

505: in their original formulation. Thus most of the model building was first concentrated

506: on the heterotic String Theory as well as type I strings. 

507: This has changed after the event of the D-branes \cite{JP,JPbook,D-branes,AD}, 

508: which naturally induce non-trivial 

509: gauge theories. It turned out that compactifications with space-time filling D-branes 

510: combined with moduli potentials due to fluxes or non-perturbative

511: effects provide a rich arena for model building in particle physics as well as cosmology 

512: \cite{reviewPP,reviewcosmo}. One of the reasons is that consistent setups 

513: with D-branes and fluxes generically demand a generalization of the Kaluza-Klein Ansatz to so-called warped 

514: compactifications \cite{BB1,Verlinde,GSS,GKP}. Remarkably, these compactifications provide 

515: a String Theory realization of models with large hierarchies \cite{Verlinde,DRS,Mayr,GSS,GKP}

516: as they were first suggested in \cite{RS}.

517:  

518: One of the major motivation of this work it to analyze the low energy 

519: dynamics of the (bulk) supergravity moduli fields within a brane world setup 

520: with a non-vanishing potential. Hence, we will more carefully introduce the basic 

521: constituents in the following. 

522: 

523: 

524: \section{Compactification and moduli stabilization}

525: \label{geom_red}

526: 

527: 

528: Sting Theory is consistently formulated in a ten-dimensional space-time.

529: In order in order to make contact with our four-dimensional observed world

530: one is forced to assign six of these dimensions to an invisible sector. 

531: This can be achieve by choosing these dimensions to be 

532: small and compact and not detectable in present experiments. 

533: Even though the additional dimensions are not observed directly,

534: they influence the resulting four-dimensional physics in a crucial way.

535:  

536: The idea of geometric compactification is rather old 

537: and goes back to the work of Kaluza and Klein in 1920 considering 

538: compactification of five-dimensional gravity on a circle \cite{KaluzaKlein}. They

539: aimed at combining gravity with $U(1)$ gauge theory in a higher-dimensional 

540: theory. Through our motivations have changed, the techniques are very similar

541: and can be generalized to the reduction from ten to four dimensions.

542: 

543: In the Kaluza-Klein reduction one starts by specifying an Ansatz for the background

544: space-time \cite{KK-review}. Topologically it is assumed to be a manifold of the product structure 

545: \beq \label{product_Ansatz}

546:   \cM_{10} = \Mext \times \Mint\ ,

547: \eeq

548: where $\Mext$ represent the four observed non-compact dimensions and $\Mint$ correspond 

549: to the compact internal manifold. On this space one specifies a block-diagonal 

550: background metric

551: \beq \label{background_metric}

552:    ds^2 =  g^{(4)}_{\mu \nu}(x)\, dx^\mu dx^\nu + g^{(6)}_{mn}(y)\, dy^m dy^n  

553:    %\left(\begin{array}{cc} g^{(4)}_{\mu \nu}(x) & 0 \\ 0 &g^{(6)}_{mn}(y) \end{array} \right)\ ,

554: \eeq  

555: where $g^{(4)}_{\mu \nu}$ is a four-dimensional Minkowski metric and $g^{(6)}_{ab}$ is 

556: the metric on the compact internal subspace. More generally, one can include 

557: a nontrivial warp factor $e^{2A(y)}$ depending on the internal coordinates $y$

558: into the Ansatz \eqref{background_metric}. This amounts to replacing 

559: $g^{(4)}_{\mu \nu}(x)$ with $e^{2A(y)} g^{(4)}_{\mu \nu}(x)$ which is 

560: the most general Ansatz for a Poincar\'e invariant four-dimensional metric \cite{deWitSD,Strominger1,BB1,Verlinde,GSS,GKP}.

561: The functional form of the warp factor is then determined by demanding the 

562: background Ansatz to be a solution of the supergravity theory. It becomes 

563: a non-trivial function in the presence of localized sources such as D-branes.

564: However, for simplicity we will restrict ourselves to the Ansatz \eqref{background_metric} 

565: in the following. 

566: 

567: The lower-dimensional theory is obtained by expanding all fields into modes of 

568: the internal manifold $\Mint$. As an illustrative example we discuss

569: the Kaluza-Klein reduction of a ten-dimensional scalar $\Phi(x,y)$ fulfilling

570: the ten-dimensional Laplace equation $\Delta_{10} \Phi = 0$ \cite{KK-review}. Using the Ansatz \eqref{background_metric}

571: the Laplace operator splits as $\Delta_{10}=\Delta_4 + \Delta_6$ and we may apply the

572: fact that $\Delta_6$ on a compact space has a discrete spectrum. The coefficients arising in 

573: the expansion of $\Phi(x,y)$ into eigenfunctions of $\Delta_6$ 

574: are fields depending only on the coordinates of $\Mext$. From a four dimensional

575: point of view the term $\Delta_6 \Phi$ thus appears as a mass term. 

576: One ends up with an infinite tower of massive states with masses

577: quantized in terms of $1/R$, where $R$ is the `radius' of $\Mint$ such that 

578: $\text{Vol}(\Mint)$ is of order $R^6$. Choosing the internal manifold to be small 

579: enough the massive Kaluza-Klein states become heavy and can be integrate 

580: out. The resulting effective theory encodes the dynamics of the four-dimensional fields

581: associated with the massless Kaluza-Klein modes satisfying

582: \beq \label{zero_modes}

583:    \Delta_6 \Phi(x,y)\ =\ 0\ .

584: \eeq

585: In chapter \ref{TypeII} we review how this procedure can be generalized to all other 

586: fields present in the ten-dimensional supergravity theories. This also includes the metric itself \cite{KK-review}. 

587: Equation \eqref{background_metric} specifies the ten-dimensional background metric and a gravity theory 

588: describes variations around this Ansatz.

589: In the non-compact dimensions these correspond to the four-dimensional graviton and 

590: the effective action reduces to the standard Einstein-Hilbert term for the metric.

591: The situation changes for the internal part of the metric. Massless fluctuations of $g_{mn}(y)$

592: around its background value, such as changes of the size and shape of $\Mint$, correspond to scalar and vector

593: fields in four-dimensions. 

594: %Zero modes of the corresponding constraint equations

595: %replacing \eqref{zero_modes} encode variations around  the background metric \eqref{product_Ansatz}. 

596: %These correspond to the four-dimensional graviton as well as massless deformations of the 

597: %internal metric $g_{mn}(y)$. 

598: As a result the four-dimensional 

599: theory consists of a huge set of scalar and vector fields arising 

600: as coefficients in the expansion of the ten-dimensional fields into zero modes of $\Mint$. 

601: In order that the four dimensional theory inherits some of the supersymmetries 

602: of the underlying ten-dimensional supergravity theory one restricts to background 

603: manifolds with structure group in $SU(3)$ such as Calabi-Yau manifolds or six-tori. 

604: This implies that the Kaluza-Klein modes reside in supermultiplets with dynamics encoded 

605: by a supergravity theory.

606: 

607: 

608: As already remarked above 

609: every compactification induces a set of massless neutral 

610: scalars called moduli. In Calabi-Yau compactifications

611: it typically consists of more then 100 scalar fields parameterizing the 

612: geometry of $\Mint$, which is clearly in conflict with the known particle spectrum. 

613: It is a long-standing problem to find a  

614: mechanism within String Theory to generate a potential for these fields. 

615: Finding such a potential will fix

616: their values in a vacuum and make them sufficiently massive such 

617: that they can be discarded from the observable spectrum. 

618: Above we already listed the three most popular possibilities to generate such a

619: potential: background fluxes, instanton corrections

620: and gaugino condensation. Let us now focus our attention to the first mechanism,

621: since fluxes will play a major role in this work.

622: 

623: To include background fluxes amounts to allowing for non-trivial vacuum expectation value of certain field

624: strengths \cite{Strominger1}--\cite{TGL2}. Take as an example a tensor field $B_2$. If its field strength 

625: $H_3 = dB_2$ admits a background flux $H_3^{\rm flux} = \big<dB_2 \big>$, the kinetic term of 

626: $B_2$ yields a contribution \cite{GKP} 

627: \beq \label{flux_pot}

628:    \int_{\cM_{10}} H_3^{\rm flux} \wedge * H_3^{\rm flux}\ ,

629: \eeq

630: which via the Hodge-$*$ couples to the metric and its deformations. Insisting on four-dimensional 

631: Poincar\'e invariance of the background, non-trivial fluxes can only be induced on internal three-cycles 

632: $\gamma$. 

633: %In string theory they are quantized as 

634: %\beq

635: %  \frac{1}{(2\pi)^2 \alpha'} \int_\gamma H_3^{\rm flux}  \in \bbZ\ , \qquad \gamma \in H_3(\Mint,\bbZ) \ .

636: %\eeq

637: The terms \eqref{flux_pot} induce a non-trivial 

638: potential for the deformations of the internal metric $g_{mn}(y)$ which generically 

639: stabilizes the corresponding moduli fields at a scale $m_{\rm flux} \sim \alpha'/R^3$ \cite{GKP,DWG}. 

640: 

641: 

642: %Even though moduli stabilization with fluxes is rather controllable, t

643: There are at least 

644: two further important points to remark. Firstly, note that in general one is not completely free to 

645: choose the fluxes, but rather has to obey certain consistency conditions. 

646: Fluxes generically induce a charge which has to be canceled on a compact space. 

647: Hence, the setup needs to be enriched by objects carrying a negative charge \cite{GKP}. 

648: Secondly, it is usually the case that fluxes do not stabilize all moduli of the theory. 

649: In order to induce a potential for the remaining fields, one needs to include 

650: non-perturbative effects such as instantons and gaugino condensates. 

651: Various recent work \cite{non-pert} is intended to get some deeper insight

652: into the nature of these corrections.

653: 

654: % To find consistet setups with all moduli fixed 

655: %by a combination of these mechanisms remains a non-trivial task \cite{D}.   

656: 

657: 

658: 

659: 

660: \section{Brane World Scenarios}

661: \label{braneworlds}

662: 

663: In the middle of the 90's, the discovery of the D-brane opened 

664: a new perspective for String Theory \cite{JP}. On the one hand, D-branes where

665: required to fill the conjectured web of string dualities \cite{Dual,JPbook}. 

666: Their appearance supports the hope for a more fundamental underlying theory 

667: unifying all the known String Theories. Moreover, they led to the conjecture of various 

668: new connections between String Theories and supersymmetric gauge theories, 

669: such as the celebrated AdS/CFT correspondence \cite{reviewAdSCFT}. 

670: From a direct phenomenological point of view, they opened a whole new arena 

671: for model building \cite{reviewPP}, since they come equipped with a gauge theory.

672: 

673: More precisely, D-branes are extended objects defined as subspaces of the ten-dimensional 

674: space-time on which open strings can end \cite{JP,JPbook,D-branes,AD}. Open strings with both ends on the same D-brane 

675: correspond to an $U(1)$ gauge field in the low energy effective action. This gauge group gets enhanced 

676: to $U(\cN)$ when putting a stack of $\cN$ D-branes on top of each other. At lowest order this 

677: induces a Yang-Mills gauge theory in the low-energy effective action. 

678: This fact allows to construct phenomenologically attractive

679: models from space-time filling D-branes consistently included in a compactification

680: of type II String Theory \cite{reviewPP}. The basic idea is that the Standard Model, 

681: or rather its supersymmetric extensions, is realized on a stack of space-time filling D-branes. 

682: The matter fields arise from dynamical excitations of the brane around its background configuration. 

683: This is similar to the situation in standard compactifications discussed in the beginning of the 

684: previous section, where moduli fields parameterize fluctuations of the background 

685: metric on $\Mint$. The crucial difference is that fluctuations of the D-branes are charged under 

686: the corresponding gauge group and can yield chiral fermions in topologically non-trivial 

687: configuration \cite{reviewPP}. 

688: 

689: In addition to the applications in Particle Physics, D-branes can serve as essential

690: ingredients to construct cosmological models. Their non-perturbative nature can be 

691: used to circumvent the no-go theorem excluding the possibility of de Sitter vacua 

692: in String Theory \cite{KKLT,BKQ,reviewcosmo}. 

693: Furthermore, similar to the fundamental string, D-branes are dynamical objects, which can 

694: move through the ten-dimensional ambient space. In certain circumstances this 

695: dynamical behavior was conjectured to be linked to a cosmological evolution \cite{reviewcosmo}.  

696: 

697: There are basically three steps to extract phenomenological data from brane world 

698: scenarios. Firstly, one has to actually construct consistent examples 

699: yielding the desired gauge groups, field content and amount of supersymmetry. 

700: Secondly, to determine the dynamics of the theory one needs to evaluate the 

701: low energy effective action of the brane excitations and the gauge neutral 

702: bulk moduli. This can then be combined with the approach 

703: to generate potentials by a flux-background and non-perturbative effects. 

704: 

705: 

706: The resulting theory may exhibit various phenomenologically interesting features. 

707: As briefly discussed in section \ref{geom_red} it can yield moduli stabilization in the vacuum.

708: Moreover, if the vacuum breaks supersymmetry this generically results in a set of soft supersymmetry 

709: breaking terms for the charged matter fields on the D-branes (see ref. \cite{KL,BIM} for 

710: a generic string inspired supergravity analysis). 

711: These can be computed from the effective low energy action

712: as it has been carried out in refs.\ \cite{Soft1,GGJL,Soft2}.

713: On the other hand, anti-branes (or brane fluxes) 

714: can be used to generate a positive cosmological constant \cite{KKLT,BKQ}. 

715: 

716: Even though this general approach sounds promising, it is extremely hard to 

717: address all these issues at once. Hence, one is usually forced to 

718: either concentrate on specific models or on one or the other ingredient 

719: to develop techniques and to extract general results. 

720: 

721: As an example, one can already check if space-time filling D-branes and fluxes 

722: alone can be consistently included in a compactification. 

723: Namely, since D-branes are charged under certain fields of the 

724: bulk supergravity theory they contribute a source term in the Bianchi identities 

725: of these fields \cite{JP,JPbook,D-branes,AD}. This is similarly true for non-trivial 

726: background fluxes. One can next apply the Gau\ss~law for the compact internal space such that 

727: consistency requires internal sources to cancel. In this respect D-branes 

728: are the higher dimensional analog of say positively charged particles. 

729: Putting such a particle in a compact space, the field lines have to end somewhere 

730: and we have to require for negative sources. In String Theory these negative sources are 

731: either appropriately chosen anti-D-branes or `orientifold planes' \cite{JP,AD}. 

732: Even though it is  possible to construct consistent 

733: scenarios with D-branes and anti-D-branes only, one may further insist to keep a 

734: $D=4$ supergravity theory. This is mainly due 

735: to the fact that these models are under much better control

736: and are not plagued by instabilities.

737: This favors the inclusion of appropriate orientifold planes, since there 

738: negative tension cancels the run-away potentials for the moduli induced by D-branes.

739: In figure \ref{braneworld_fig} we schematically picture some ingredients of a

740: brane-world model.

741: 

742: \begin{figure}[h]

743: \begin{center}

744: \includegraphics[height=5cm]{braneworld.eps}

745: \caption{\textit{Brane-world scenario on $\Mext \times \Mint$ 

746:          with space-time filling D-branes, orientifold planes and background fluxes.}}

747: \label{braneworld_fig}

748: \end{center}

749: \end{figure}

750: 

751: Orientifold planes arise in String Theories constructed 

752: form type II strings by modding out world-sheet parity plus a geometric 

753: symmetry $\sigma$ of $\Mext \times \Mint$ \cite{JP,AD}. On the level of the full String Theory 

754: this implies that non-orientable string world-sheets, such as the Klein bottle 

755: or the M\"obius strip, are allowed. 

756: Focusing on the effective action orientifolds break part or all of 

757: the supersymmetry of the low-energy theory. By imposing 

758: appropriate conditions on the  orientifold projection and the included D-branes 

759: the setup can be adjusted to preserve exactly half of the original supersymmetry. 

760: 

761: {}From a phenomenological point of view spontaneously broken $\cN=1$

762: theories are of particular interest.

763: Starting from type II String Theories in ten space-time dimensions, 

764: one can compactify on Calabi-Yau

765: threefolds to obtain  $\cN=2$ theories in four dimensions.

766: This $\cN=2$ is further broken to $\cN=1$ if in addition

767: background D-branes and orientifold planes are present \cite{Ori,GKP,AAHV,BBKL,BH}.

768: The presence of background fluxes or other

769: effects generating a potential results in a 

770: spontaneously broken  $\cN=1$ theory \cite{Bachas}--\cite{TGL2}. 

771: To examine this setup on the level of the effective action is one of the

772: motivations for this work. Note that all these brane world scenarios are conjectured

773: to admit a higher dimensional origin in a more fundamental theory, which we briefly introduce next.

774: However, it is important to keep in mind that this unifying theory is much 

775: less understood then the five String Theories.   

776: 

777: 

778: \subsection{From dualities to M- and F-theory}

779: 

780: At the first glance in seems as if we have to choose one or

781: the other String Theory in which we aim to construct a specific

782: model. However, it turns out that many of these choices are 

783: actually equivalent and linked by various dualities \cite{Dual,JPbook}. The full set of 

784: dualities forms a interlocking web between all five String

785: Theories (see figure \ref{web}).

786: 

787: \begin{figure}[h]

788: \begin{center}

789: \includegraphics[height=5cm]{dual_web.eps}

790: \caption{\textit{The duality web of String Theories.}}

791: \label{web}

792: \end{center}

793: \end{figure}

794: 

795: As an example type IIA compactified on a circle of radius $R$

796: is shown to be equivalent to type IIB on a circle of radius $1/R$ \cite{JPbook,revT-dual}. 

797: This duality is termed T-duality and relates two String Theories 

798: at weak string coupling \cite{revT-dual}. There are also strong/weak dualities

799: such as S-duality, which is a symmetry of the type IIB String

800: Theory \cite{Dual}. Both of these dualities can be generalized and applied 

801: to standard Calabi-Yau compactifications as well as 

802: brane-world scenarios. 

803: 

804: A prominent example is mirror symmetry which can be interpreted as performing

805: several T-dualities \cite{SYZ}. 

806: It associates to each Calabi-Yau manifold $Y$ a corresponding mirror Calabi-Yau $\tilde Y$ \cite{Mirror}. 

807: Within the framework of String Theory it can be argued that 

808: type IIA compactified on $Y$ is fully equivalent to type IIB strings on 

809: $\tilde Y$. From a mathematical point of view mirror symmetry 

810: exchanges the odd cohomologies of $Y$ with the even cohomologies of $\tilde Y$

811: and vice versa. Even stronger it suggests that the moduli spaces of the 

812: two Calabi-Yau manifolds are identified. Remarkably, in specific 

813: examples this allows to calculate stringy corrections to the theory on $Y$

814: from geometrical data of $\tilde Y$. Mirror symmetry can be generalized

815: to setups with D-branes \cite{Aspinwall} and eventually should identify type IIA and type IIB 

816: brane world scenarios. This raises various non-trivial questions such 

817: as in which way mirror symmetry applies to flux compactifications \cite{Vafa_NCY,GLMW}.

818: 

819: Let us also introduce S-duality in slightly more detail \cite{Dual,JPbook}. Type 

820: IIB String Theory contains in addition to the fundamental string 

821: also a D-string (D1-brane). 

822: It can now be argued that the theory where the fundamental string 

823: is at low coupling $g_s$, and hence the D-string is very heavy, is 

824: dual to a theory at $1/g_s$ with the role of both strings exchanged. 

825: Carefully identifying the fields, S-duality is also shown to be a 

826: symmetry of the corresponding type IIB low-energy effective action.

827: This strong/weak duality is actually part of a larger symmetry group 

828: $Sl(2,\bbZ)$. It has been suggested in \cite{Vafa} that this duality 

829: group admits a geometric interpretation in terms of two additional 

830: toroidal dimensions. This twelve dimensional construction was named

831: F-theory. The additional two dimensions are necessarily a compact torus,

832: which however in compactifications can be non-trivially fibered over 

833: the compactification manifold. This naturally applies to type IIB 

834: brane-world scenarios, which generically admit backgrounds corresponding 

835: these non-trivial compactifications \cite{Sen,DRS,GKP}.

836: 

837: The existence of these various dualities suggests that the ten-dimensional 

838: String Theories are actually just different limits of a more fundamental 

839: theory \cite{Dual} as pictured in figure \ref{web}. 

840: This mysterious theory unifying all five String Theories was named 

841: M-theory. In general, not much is known about its actual

842: formulation and the required structures 

843: are far less understood then the one for String Theory. However,

844: there are certain regimes in which one believes to find some

845: hints of its existence. This also includes the existence 

846: D-branes, which fit into this picture as they occur from 

847: higher-dimensional objects termed M-branes. There also is  

848: a unique supergravity theory in eleven dimensions \cite{CJS}, 

849: which is interpreted to be the low-energy limit of M-theory. 

850: In the final chapter of this article it will be this 

851: low-energy theory which allows us lift the orientifold

852: compactifications to M-theory.

853: 

854: 

855: 

856: \subsection{Topics and outline of this article}

857: 

858: After this brief general introduction let us now turn to the actual topics

859: of this article. As just discussed, an essential step to extract 

860: phenomenological properties of string vacua with 

861: (spontaneously broken) $N=1$ supersymmetry in brane world 

862: scenarios is to determine the low energy effective 

863: action. In this work we focus 

864: on type IIA and IIB String Theory compactified on generic Calabi-Yau 

865: orientifolds and determine their low energy effective action in 

866: terms of geometrical data of the Calabi-Yau orientifold and 

867: the background fluxes. We include D-branes 

868: for consistency, but freeze their matter fields (and moduli) concentrating 

869: on the couplings of the bulk moduli. We also provide a detailed discussion 

870: of the resulting $N=1$ moduli space in the chiral and the dual linear 

871: multiplet description and check mirror symmetry in the large volume--large complex 

872: structure limit. Moreover, we show at the level of the effective actions 

873: that Calabi-Yau orientifolds with fluxes admit a natural embedding 

874: into F- and M-theory compactifications. 

875: 

876: In \textit{chapter \ref{TypeII}} we first briefly review standard Calabi-Yau 

877: compactifications of type IIA and type IIB supergravity and discuss the resulting 

878: $N=2$ supergravity action.  In doing so we focus 

879: on the geometry of the moduli space $\cM^{\rm SK} \times \cM^{\rm Q}$ spanned by 

880: the scalars of the $N=2$ supergravity theory. Supersymmetry constrains 

881: it to locally admit this product form, where $\cM^{\rm SK}$ is a

882: special K\"ahler manifold and $\cM^{\rm Q}$ is a quaternionic manifold.  

883: Furthermore, we introduce $N=2$ mirror symmetry on the level of 

884: the effective action and present a somewhat non-standard construction 

885: of the mirror map between the IIA and IIB quaternionic moduli spaces

886: reproducing the results of \cite{BGHL}. 

887: 

888: In \textit{chapter \ref{effective_actO}} we immediately turn to the 

889: compactification of type II theories on Calabi-Yau orientifolds. 

890: We start with a more detailed introduction to setups with 

891: D-branes and orientifold planes and comment on consistency and 

892: supersymmetry conditions. As already mentioned in section \ref{braneworlds} 

893: orientifold planes are essential ingredients to obtain supersymmetric theories in 

894: brane-world compactifications. They arise in String Theories modded

895: out by a geometrical symmetry $\sigma$ of $\Mext \times \Mint$

896: in addition to the world-sheet parity operation. We demand 

897: $\Mint$ to be a generic Calabi-Yau manifold admitting an 

898: isometric involutive symmetry $\sigma$. It turns out that in order to preserve

899: $N=1$ supersymmetry $\sigma$ has to be a holomorphic 

900: map in type IIB and an anti-holomorphic map in type IIA 

901: compactifications. Taking into account further properties of $\sigma$ 

902: one finds three supersymmetric setups \cite{AAHV,BH}: (1) IIB orientifolds with 

903: $O3/O7$ planes, (2) IIB orientifolds with $O5/O9$ planes and 

904: (3) IIA orientifolds with $O6$ planes. 

905: 

906: The spectrum of these

907: theories was first determined in \cite{BH}.

908: However, the effective action was only computed 

909: for special cases of type IIB Calabi-Yau orientifolds with $O3/O7$ 

910: planes \cite{GKP,BBHL}. In \cite{TGL1} we generalized these results and 

911: also included an analysis of $O5/O9$ setups. 

912: For type IIA brane-world scenarios the calculation of the low energy 

913: supergravity theory was mainly concerned with orbifolds of six-tori \cite{CP,reviewPP} for 

914: which conformal field theory techniques can be applied. Complementary, 

915: the dynamics of the bulk theory can extracted for general type IIA Calabi-Yau orientifolds by 

916: using a Kaluza-Klein reduction as shown in our publication \cite{TGL2}.

917: In chapter \ref{effective_actO} we review the first parts of refs. \cite{TGL1,TGL2} 

918: and determine the $N=1$ effective action of all three setups. We extract

919: the K\"ahler potential and the gauge-kinetic couplings by first assuming that 

920: no background fluxes are present. The $N=1$ moduli space is shown to 

921: be a local product $\tilde \cM^{\rm SK} \times \tilde \cM^{\rm Q}$, where 

922: $\tilde \cM^{\rm SK}$ is a special K\"ahler manifold inside $\cM^{\rm SK}$

923: and $\tilde \cM^{\rm Q}$ is a K\"ahler manifold inside the quaternionic 

924: manifold $\cM^{\rm Q}$. 

925: 

926: We end chapter \ref{effective_actO} with a discussion of

927: mirror symmetry for Calabi-Yau orientifolds and determine 

928: the necessary conditions

929: on the involutive symmetries of the mirror IIA and IIB orientifold theories. 

930: By specifying two types of 

931: special coordinates on the IIA side, we are able to identify the large complex 

932: structure limit of IIA orientifolds with the large volume limits of IIB orientifolds

933: with $O3/O7$ and $O5/O9$ planes. 

934: 

935: In \textit{chapter \ref{lin_geom_of_M}} we present a more detailed analysis of

936: the $N=1$ moduli space geometry of Calabi-Yau orientifold compactifications \cite{TGL1,TGL2}.

937: The special K\"ahler manifold $\tilde \cM^{\rm SK}$ inherits its geometrical 

938: structure directly from $N=2$, such that we focus our attention

939: to the K\"ahler manifold $\tilde \cM^Q$ inside the quaternionic space.  

940: We show that the definition of the K\"ahler coordinates as well as

941: certain no-scale type conditions can be more easily understood 

942: in terms of the `dual' formulation where some chiral multiplets of the Calabi-Yau 

943: orientifold are replaced by linear multiplets. A linear multiplet consists of 

944: a real scalar and an anti-symmetric two-tensor as bosonic fields. In the massless case 

945: this two-tensor is dual to a second real scalar and one is led back to the chiral description. 

946: In order to do set the 

947: stage for the orientifold analysis we first 

948: review $N=1$ supergravity with several linear multiplets following \cite{BGG}.

949: The transformation into linear multiplets corresponds to a Legendre transformation of 

950: the K\"ahler potentials and coordinates. In the dual picture 

951: the characteristic functions for type IIB orientifolds take a particularly 

952: simple form. Moreover, in type IIA orientifolds the Legendre transform is 

953: essential to make contact with the underlying $N=2$ special geometry.

954: As a byproduct we determine an entire new class of no-scale

955: K\"ahler potentials which in the chiral formulation

956: can only be given implicitly as the solution of some constraint equation.

957: These new insights will enable us to give an direct construction of the K\"ahler 

958: manifold $\tilde \cM^{\rm Q}$ in analogy to the moduli space of supersymmetric Lagrangian submanifolds \cite{Hitchin2}. 

959: Moreover, this sets the stage to generalize the reduction to orientifolds of certain

960: non-Calabi-Yau manifolds introduced in \cite{HitchinGCM,Gualtieri}.

961: 

962: In \textit{chapter \ref{fluxesAB}} we redo the Kaluza-Klein compactification by additionally allowing

963: for non-trivial background fluxes. For $O3/O7$ orientifolds this amounts to 

964: a generalization of the analysis presented in \cite{GKP,BBHL} and

965: confirms that the Gukov-Vafa-Witten superpotential \cite{GVW} encodes 

966: the potential due to background fluxes. However, we show that 

967: for orientifolds with $O5/O9$ planes background fluxes generically 

968: result in a non-trivial superpotential, $D$-terms as well as

969: a direct mass term for a linear multiplet.

970: Following this observation, supergravity theories with massive linear

971: multiplets coupled to vector and chiral multiplets where further analyzed 

972: in \cite{mass_tensors}. Surprisingly, in type IIA orientifolds with background fluxes 

973: the superpotential depends on all (bulk) moduli fields of the theory. 

974: In \cite{KachruK} an equivalent observation was made for the underlying $N=2$

975: theory. This suggests that all geometric moduli can be stabilized in a supersymmetric 

976: vacuum \cite{KachruK,TGL2}. In ref.~\cite{DeWGKT} this 

977: was shown to be possible at large volume and small string coupling (see also \cite{VZ}).  

978: 

979: The IIA superpotential is expected to receive non-perturbative corrections

980: from world-sheet as well as D-brane instantons. In the final section of 

981: chapter \ref{fluxesAB} we derive that for supersymmetric type IIA and type IIB instantons 

982: the respective actions are linear in the chiral coordinates and therefore can result in holomorphic

983: corrections to the superpotential.

984: 

985: In \textit{chapter \ref{M-F-embedding}} we embed type IIB and type IIA 

986: orientifolds into F- and M-theory compactifications.  

987: Orientifolds with $O3/O7$-planes can be obtained as a limit of 

988: F-theory compactified on elliptically fibered Calabi-Yau fourfolds \cite{Sen}.

989: We check this correspondence on the level of the effective action

990: by first compactifying M-theory on a specific Calabi-Yau fourfold and comparing

991: the result with the effective action of $O3/O7$ orientifolds compactified 

992: on a circle to $D=3$. The low energy effective action of 

993: M-theory compactifications on Calabi-Yau four-folds 

994: was determined in \cite{HL,BHS} and we use their results in

995: a slightly reformulated version. Moreover, it turns out 

996: that this duality is best understood in the dual pictures

997: where three-dimensional vector multiplets are kept in the 

998: spectrum and the K\"ahler potential is an explicit function of the 

999: moduli. We determine simple solutions to the 

1000: fourfold consistency conditions for which we find perfect matching between the 

1001: orientifold and M-theory compactifications. This correspondence can be lifted

1002: to $D=4$ where M-theory  on an elliptically fibered Calabi-Yau fourfold descents 

1003: to an F-theory compactification. 

1004: 

1005: We end this chapter by also discussing the embedding of type IIA

1006: orientifolds into a specific class of $G_2$ compactifications of M-theory 

1007: as suggested in \cite{KMcG}. 

1008: Restricting the general results of \cite{PT,HM,Hitchin1,GPap,BW} to a specific 

1009: $G_2$ manifold and neglecting the contributions arising from the singularities

1010: we show agreement between the low energy effective

1011: actions \cite{TGL2}. In \cite{TGL2} we discovered that only parts of the 

1012: orientifold flux superpotential decent from fluxes in an M-theory 

1013: compactifications on manifolds with $G_2$ holonomy.  

1014: However, as we will argue one of the missing terms is

1015: generated on $G_2$ structure manifolds with non-trivial

1016: fibrations. However, the higher-dimensional origin of the term 

1017: involving the mass parameter of massive type IIA supergravity 

1018: remains mysterious.

1019: 

1020: This article is mainly based on the publications 

1021: \cite{TGL1} and \cite{TGL2} of the author. 

1022: However, we also present various new results. 

1023: Namely, it turns out to be possible to reformulate the results of 

1024: \cite{TGL1,TGL2} in a very elegant and powerful way adapted to

1025: Hitchin's analysis of special even and odd forms on six-manifolds \cite{HitchinGCM,Hitchin1}.

1026: This allows for a better understanding of the N=1 moduli 

1027: space inside the quaternionic manifold and 

1028: suggests a generalization to non-Calabi-Yau orientifolds.

1029: Moreover, we included a detailed analysis of 

1030: the orientifold limit of the F-theory embedding of type IIB orientifolds. 

1031: In addition we identify the higher-dimensional origin of a second flux term of the 

1032: IIA orientifold superpotential being due to a non-trivial fibration of a 

1033: $G_2$ structure manifold.

1034: 

1035: 

1036: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1037: %

1038: %   Type II on Calabi-Yau

1039: %

1040: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1041: 

1042: 

1043: 

1044: \chapter{Calabi-Yau compactifications of Type II theories}

1045: \label{TypeII}

1046: 

1047: 

1048: In this section we review compactifications of type IIA and type IIB 

1049: supergravity on a Calabi-Yau manifold $Y$. These lead to $N=2$ supergravity

1050: theories in four dimensions expressed in terms of the characteristic data 

1051: of the Calabi-Yau space. We start our discussion with some mathematical preliminaries.

1052: In section \ref{CY-mfds} we introduce Calabi-Yau manifolds and give a short 

1053: description of their moduli spaces. In a next step we turn to compactifications

1054: of IIA and IIB supergravity on Calabi-Yau manifolds in section \ref{revIIB} and \ref{revIIA}.

1055: Finally, in section \ref{revMirror} we give a brief account of $N=2$ mirror symmetry 

1056: applied at the level of the effective action. The mirror map for the quaternionic

1057: moduli spaces will be constructed. 

1058: 

1059: \section{Calabi-Yau manifolds and their moduli space}

1060: \label{CY-mfds}

1061: 

1062: String theory is consistently formulated in a ten-dimensional target space. 

1063: In order to reduce to a four-dimensional observable world, we choose 

1064: the background to be of the form $\cM_{10}=\Mext \times \Mint$ as 

1065: already given in \eqref{product_Ansatz}.

1066: Here $\Mint$ is a compact six-dimensional manifold, which, in principle, we are 

1067: free to choose. Due to this Ansatz, the Lorentz group of $\cM_{10}$ decomposes 

1068: as $SO(9,1)\rightarrow SO(3,1)\times SO(6)$, where $SO(6)$ is the generic

1069: structure group of a sixfold. 

1070: However, demanding $\Mint$ to preserve the minimal amount of supersymmetry

1071: one has to pick a manifold with structure group $SU(3)$. 

1072: They admit one globally defined spinor $\eta$, since the $SO(6)$ spinor 

1073: representation $\bf{4}$ decomposes to $\bf{1}\oplus\bf{3}$.

1074: Further demanding this spinor $\eta$ to be covariantly constant 

1075: reduces the class of background manifolds to manifolds with 

1076: $SU(3)$ holonomy \cite{GSWbook}. These spaces are called Calabi-Yau manifolds and   

1077: are complex K\"ahler manifolds, which are in addition Ricci flat \cite{Huebsch}.

1078: 

1079: In terms of $\eta$ one can globally define a covariantly constant

1080: two-from $J$ (the K\"ahler form) and a three-form $\Omega$ (the holomorphic three-form). 

1081: For a fixed complex structure these 

1082: fulfill the algebraic conditions

1083: \beq

1084:   J \wedge J \wedge J\ \propto\ \Omega \wedge \bar \Omega\ , \qquad J \wedge \Omega = 0\ . 

1085: \eeq

1086: where the proportionality factor depends on the normalization of $\Omega$ with respect to $J$.

1087: Performing a Kaluza-Klein reduction on the background \eqref{product_Ansatz} 

1088: the massless four-dimensional fields arise as the zero modes of 

1089: the internal Laplacian \eqref{zero_modes} \cite{GSWbook,JPbook}. These zero modes are in one-to-one correspondence 

1090: with harmonic forms on $Y$ and thus their multiplicity is counted by the dimension

1091: of the non-trivial cohomologies of the Calabi-Yau manifold. The Calabi-Yau condition poses 

1092: strong constraints on the Hodge decomposition of the cohomology groups.

1093: The only non-vanishing cohomology groups are the even and odd cohomologies 

1094: \bea \label{odd_even_cohom}

1095:   H^{ev}& =& H^{(0,0)} \oplus H^{(1,1)} \oplus  H^{(2,2)} \oplus H^{(3,3)}\ , \\   

1096:   H^{odd}& =& H^{(3,0)} \oplus H^{(2,1)} \oplus  H^{(1,2)} \oplus H^{(0,3)}\ . \nn

1097: \eea  

1098: Their dimensions $h^{(p,q)}=\dim H^{(p,q)}$ can be summarized in the Hodge diamond 

1099: as follows

1100: \beq \label{hodge_diamond}

1101:   \begin{array}{c}

1102:    h^{(0,0)}\\

1103:    h^{(1,0)} \qquad h^{(0,1)}\\

1104:     h^{(2,0)}\qquad h^{(1,1)} \qquad h^{(0,2)}\\

1105:    h^{(3,0)}\qquad h^{(2,1)}\qquad h^{(1,2)} \qquad h^{(0,3)}\\

1106:    h^{(3,1)}\qquad h^{(2,2)} \qquad h^{(1,3)}\\

1107:    h^{(3,2)} \qquad h^{(2,3)}\\

1108:    h^{(3,3)}

1109:   \end{array}\ =\

1110:   \begin{array}{c}

1111:    1\\

1112:    0 \qquad 0\\

1113:    0\qquad h^{(1,1)} \qquad 0\\

1114:    1\qquad h^{(2,1)}\qquad h^{(2,1)} \qquad 1\\

1115:    0\qquad h^{(1,1)} \qquad 0\\

1116:    0 \qquad 0\\

1117:    1

1118:   \end{array}\ .

1119: \eeq

1120: Let us introduce a basis for the different cohomology 

1121: groups by always choosing the unique harmonic representative in 

1122: each cohomology class. The basis of harmonic $(1,1)$-forms

1123: we denote by $\omega_A$ with dual harmonic $(2,2)$-forms

1124: $\tilde \omega^A $ which form a basis of $H^{(2,2)}(Y)$.

1125: $(\alpha_{\hat K}, \beta^{\hat L})$ are harmonic three-forms

1126: and form a real, symplectic basis on $H^{(3)}(Y)$. Together 

1127: the non-trivial intersection numbers are summarized as

1128: \beq \label{int-numbers1}

1129:   \int_Y \omega_A \wedge \tilde \omega^B = \delta_A^B\ , \qquad 

1130:   \int_Y \alpha_{\hat K} \wedge \beta^{\hat L} = \delta^{\hat L}_{\hat K}\ ,

1131: \eeq

1132: with all other intersections vanishing. Finally, we denote by $\vol(Y)$ the 

1133: harmonic volume $(3,3)$-form of the Calabi-Yau space.  

1134: In Table \ref{CYbasis} we summarize 

1135: the non-trivial cohomology groups on $Y$ and denote their basis elements. 

1136: 

1137: \begin{table}[h]

1138: \begin{center}

1139: \begin{tabular}{| c | c | c |} \hline

1140:    \rule[-0.3cm]{0cm}{0.8cm} cohomology group&

1141:    dimension & basis

1142:    \\ \hline

1143:    \rule[-0.3cm]{0cm}{0.8cm} $H^{(1,1)}$  &

1144:    $h^{(1,1)}$ & $\omega_A$

1145:    \\ \hline

1146:    \rule[-0.3cm]{0cm}{0.8cm} $H^{(2,2)}$  & $h^{(1,1)}$ & $\tilde \omega^A$

1147:    \\ \hline

1148:    \rule[-0.3cm]{0cm}{0.8cm} $H^{(3)}$  & $2h^{(2,1)}+2$ &

1149:    $(\alpha_{\hat K},\beta^{\hat L})$ \\ \hline

1150: \rule[-0.3cm]{0cm}{0.8cm} $H^{(2,1)}$  &

1151:    $h^{(2,1)}$ &

1152:    $\chi_K$

1153:    \\ \hline

1154:    \rule[-0.3cm]{0cm}{0.8cm} $H^{(3,3)}$  &

1155:    $1$&

1156:    $\vol$

1157:    \\ \hline

1158: \end{tabular}

1159: \caption{\small \label{CYbasis}

1160: \textit{Cohomology groups on $Y$ and their basis elements.}}

1161: \end{center}

1162: \end{table}

1163: 

1164: In sections \ref{revIIA} and  \ref{revIIB} we explain how these harmonics 

1165: yield four-dimensional massless fields, when expanding the ten-dimensional 

1166: supergravity forms. Furthermore, there are additional massless modes arising as

1167: deformations of the metric $g_{i\jb}$.  

1168: Considering variations $R_{mn}(g+\delta g)$ of the Ricci-tensor which 

1169: respect the Ricci-flatness condition $R_{mn}=0$ forces $\delta g$ to satisfy a 

1170: differential equation (the Lichnerowicz equation). 

1171: Solutions to this equation can be identified in case of a Calabi-Yau manifold 

1172: with the harmonic $(1,1)$- and $(2,1)$-forms, which parameterize K\"ahler structure and 

1173: complex structure deformations of $Y$ \cite{Tian, CdO, Huebsch}.  The deformations of the 

1174: K\"ahler form $J = i {g}_{i\bj}\, dy^i \wedge d\bar y^\jb$ 

1175: give rise to $h^{(1,1)}$ real scalars $v^{A}$ and one expands \footnote{%

1176: Globally only those deformations are allowed which keep the volume 

1177: of $Y$ as well as its two- and four-cycles positive, i.e.\

1178: $\int_Y J \wedge J \wedge J \ge 0$, $\int_{S_4} J \wedge J  \ge 0$ and $\int_{S_2} J \ge 0$.

1179: These conditions are preserved under positive rescalings of the fields $v^A$, such that 

1180: they span a $h^{(1,1)}-$dimensional cone.}

1181: \beq\label{def-v}

1182:   g_{i\bj} + \delta g_{i\bj} = -i\, J_{i\bj}  = -i\, v^{A}\, (\omega_{A})_{i\bj}  \ , \qquad A = 1, \ldots, h^{(1,1)}\ .

1183: \eeq

1184: These real deformations are complexified by the $h^{(1,1)}$ real scalars $b^A(x)$ 

1185: arising in the expansion of the B-field present in both type II string theories. 

1186: More precisely we introduce the complex fields

1187: \beq \label{def-t_II}

1188: t^A = b^A + i\, v^A\ ,

1189: \eeq

1190: which parameterize the $h^{(1,1)}-$dimensional complexified K\"ahler cone \cite{CdO}.

1191: 

1192: The second set of deformations are variations of the complex structure

1193: of $Y$. They are parameterized by complex scalar fields $z^{K}$

1194: and are in one-to-one correspondence with harmonic

1195: $(1,2)$-forms 

1196: \beq\label{cs}

1197:   \delta{g}_{ij} =  \frac{i}{||\Omega||^2}\, \bar z^{K} 

1198:   (\bar \chi_{K})_{i\ib\bj}\,

1199:   \Omega^{\ib\bj}{}_j \ , \quad K=1,\ldots,h^{(1,2)}\ ,

1200: \eeq

1201: where $\Omega$ is the holomorphic (3,0)-form,

1202: $\bar\chi_{K}$ denotes a basis of $H^{(1,2)}$ and we abbreviate

1203: $||\Omega||^2\equiv \frac1{3!}\Omega_{ijk}\bar\Omega^{ijk}$.

1204: 

1205: Together the complex scalars $z^K$ and $t^A$ span the geometric moduli 

1206: space of the Calabi-Yau manifold. It is shown to be locally a product 

1207: \beq \label{geom-mod}

1208:    \cM^{\rm cs} \times \cM^{\rm ks}\ ,

1209: \eeq

1210: where both factors are special K\"ahler manifolds of complex dimension 

1211: $h^{(2,1)}$ and $h^{(1,1)}$ respectively. To make that more precise let us first 

1212: discuss $\cM^{\rm cs}$. Its metric $G_{K \bar L}$ is given by \cite{Tian,Strominger2,CdO}

1213: \beq \label{chi_barchi}

1214:   G_{K \bar L} = -\frac{\int_Y \chi_K \wedge \bar \chi_{ L}}{\int_Y \Ox \wedge \bar \Ox} \ ,

1215: \eeq

1216: where $\chi_K$ is related to the

1217: variation of the three-form

1218: $\Omega$ via Kodaira's formula

1219: \beq \label{Kod-form}

1220: \chi_K(z,\bar z) = \partial_{z^K} \Omega(z)+ \Omega(z)\, \partial_{z^K}\Kcs   \ .

1221: \eeq

1222: With the help of \eqref{Kod-form} one shows that $G_{K \bar L}$ is a K\"ahler manifold, 

1223: since we can locally find complex coordinates $z^K$ and a function

1224: $K(z,\bar z)$ such that 

1225: \beq\label{csmetric}

1226:  G_{K \bar L} = {\partial}_{z^K}\partial_{\bar z^L}\  \Kcs\ , \qquad \Kcs = -\ln\Big[ i \int_Y \Ox \wedge \bar \Ox\Big] 

1227:       = -\ln i\Big[\bar Z^\Kh\mathcal{F}_\Kh - Z^\Kh\bar{\mathcal{F}}_\Kh \Big]

1228: \ ,

1229: \eeq

1230: where the holomorphic periods  $Z^\Kh, \mathcal{F}_\Kh$ are defined as

1231: \beq \label{pre-z}

1232: %  \cF(Z) = \tfrac{1}{2} Z^\Kh \cF_{\Kh}\ ,\quad  

1233: Z^\Kh(z) = \int_Y \Omega(z) \wedge \beta^\Kh\ , \qquad 

1234: \cF_\Kh(z) = \int_Y \Omega(z) \wedge \alpha_\Kh\ , 

1235: \eeq

1236: or in other words $\Omega$ enjoys the expansion

1237: \beq\label{Omegaexp}

1238:   \Omega(z) = Z^\Kh(z)\, \alpha_\Kh - \cF_\Kh(z)\, \beta^\Kh\ .

1239: \eeq

1240: The K\"ahler manifold $\cM^{\rm cs}$ is furthermore special K\"ahler,

1241: since $\cF_\Kh$ is the first derivative with respect to $Z^\Kh$ of 

1242: a prepotential 

1243: $\cF = \frac{1}{2} Z^\Kh \cF_\Kh$. This implies that $G_{K \bar L}$ 

1244: is fully encoded in the holomorphic function $\cF$.

1245:   

1246: Note that $\Omega$ is only defined up to complex rescalings by a holomorphic function

1247: $e^{-h(z)}$ which via \eqref{csmetric} 

1248: also changes the K\"ahler potential by a K\"ahler transformation 

1249: \beq\label{crescale}

1250: \Omega\to\Omega\, e^{-h(z)}\ , \qquad \Kcs\to\Kcs + h +\bar h\ .

1251: \eeq

1252: This symmetry renders one of the periods (conventionally

1253: denoted by $Z^0$) unphysical

1254: in that one can always choose to fix a K\"ahler gauge and set $Z^0 = 1$. 

1255: The complex structure

1256: deformations can thus be identified with the remaining 

1257: $h^{(1,2)}$ periods $Z^K$ by defining the special coordinates

1258: $z^K = {Z^K}/{Z^0}$. A more  

1259: detailed discussion of special geometry can be found in appendix \ref{specialGeom}.

1260: 

1261: Let us next turn to the second factor in \eqref{geom-mod} spanned by the complexified 

1262: K\"ahler deformations $t^A$.  The metric on $\cM^{\rm ks}$ is given by \cite{Strominger,CdO}

1263: \bea \label{Kmetric} 

1264:   G_{A B} = \frac{3}{2\KK}

1265:   \int_{Y}\omega_A \wedge *\omega_B = -\frac{3}{2}\left( \frac{\KK_{AB}}{\KK}-

1266:   \frac{3}{2}\frac{\KK_A \KK_B}{\KK^2} \right) = \partial_{t^a} \partial_{\bar t^B} K^{\rm ks}\ ,

1267: \eea

1268: where $*$ is the six-dimensional Hodge-$*$ on $Y$ and the K\"ahler potential $K^{\rm ks}$

1269: is given by 

1270: \beq \label{Kpot_ks}

1271:    K^{\rm ks} = - \ln \big[\tfrac{i}{6} \cK_{ABC}(t-\bar t)^A (t-\bar t)^B (t-\bar t)^C \big] = - \ln \tfrac{4}{3} \cK\ ,

1272: \eeq  

1273: where $\frac16 \cK$ is the volume of the Calabi-Yau manifold.

1274: We abbreviated the intersection numbers as follows

1275: \bea\label{int-numbers}

1276:   \KK_{ABC} &=& \int_{Y}\omega_A \wedge \omega_B \wedge \omega_C\ , \qquad  \qquad 

1277:   \KK_{AB}  = \int_{Y}\omega_A \wedge \omega_B \wedge J 

1278: = \KK_{ABC}v^C\ ,  \\

1279:   \KK_{A}   &=& \int_{Y}\omega_A \wedge J \wedge J

1280: =\KK_{ABC}v^Bv^C \ , \qquad 

1281:   \KK = \int_{Y}J \wedge J \wedge J

1282:  =\KK_{ABC}v^Av^Bv^C \ ,\nonumber

1283: \eea

1284: with $J = v^A \omega_A $ being the K\"ahler form of $Y$. 

1285: The manifold $\cM^{\rm ks}$ is once again special K\"ahler, since 

1286: $K^{\rm ks}$ given in \eqref{Kpot_ks} can be derived from a single holomorphic 

1287: function $f(t)=-\frac{1}{6} \cK_{ABC} t^A t^B t^C$ via \eqref{Kinz}. 

1288: 

1289: 

1290: 

1291: 

1292: \section{Type IIA on Calabi-Yau manifolds}

1293: \label{revIIA}

1294: 

1295: Let us now apply these tools in Calabi-Yau compactifications of type IIA supergravity

1296: following \cite{BCF,FS}. 

1297: This theory is the maximally supersymmetric theory in ten spacetime dimensions,

1298: which posses two gravitinos of opposite chirality. It is naturally obtained as the low energy limit of 

1299: type IIA superstring theory. Thus the supergravity spectrum consists of the massless string modes. 

1300: The bosonic fields are the dilaton $\hat \phi$, the ten-dimensional metric $\hat g$ and the two-form 

1301: $\hat B_2$ in the NS-NS sector, 

1302: while the one- and three-forms $\hat C_1,\hat C_3$ arise in 

1303: the R-R sector.\footnote{We use a `hat'

1304: to denote ten-dimensional quantities and omit it for 

1305: four-dimensional fields.} Using form notation (our conventions are summarized in appendix~\ref{conventions})

1306: the corresponding ten-dimensional type IIA supergravity action in the 

1307: Einstein frame is given by \cite{JPbook}

1308: \bea \label{10dact}

1309:   S^{(10)}_{IIA} &=& \int -\tfrac{1}{2}\hat R*\mathbf{1} -\tfrac{1}{4} d\hat \phi\wedge * d\hat \phi

1310:   -\tfrac{1}{4} e^{-\hat \phi}\hat H_3 \wedge *\hat H_3 

1311:   -\tfrac{1}{2} e^{\frac{3}{2} \hat \phi}\hat F_2 \wedge *\hat F_2 \nn \\

1312:   && -\tfrac{1}{2} e^{\frac{1}{2} \hat \phi}\hat F_4 \wedge *\hat F_4 

1313:   -\tfrac{1}{2} \hat B_2 \wedge \hat F_4 \wedge \hat F_4\ ,

1314: \eea

1315: where the field strengths are defined as 

1316: \bea \label{defHFF1}

1317:   \hat H_3 = d \hat B_2\ , \quad \hat F_2 = d\hat C_1\ , \quad 

1318:   \hat F_4 = d\hat C_3 - \hat C_1 \wedge \hat H_3\ .

1319: \eea

1320: In order to dimensionally reduce type IIA to a four-dimensional  

1321: theory, we make the product Ansatz $\Mext \times \Mint$ and perform a Kaluza-Klein 

1322: reduction. Since $Y$ is a Calabi-Yau manifold it posses one covariantly constant spinor $\eta$. 

1323: Decomposing the two ten-dimensional gravitinos into $\eta$ times some four-dimensional spinor

1324: leads to two gravitinos in $D=4$. Hence, compactifying type IIA supergravity on a Calabi-Yau

1325: threefold $Y$ results in an $N=2$ theory in four space-time dimensions and the zero modes of 

1326: $Y$ have to assemble into  massless  $N=2$ multiplets.

1327: These zero modes are in one-to-one correspondence with harmonic forms

1328: on $Y$ and thus their multiplicity is counted by the dimension

1329: of the non-trivial cohomologies of the Calabi-Yau manifold.

1330: For the dimensional reduction 

1331: one chooses a block diagonal Kaluza-Klein Ansatz for the 

1332: ten-dimensional background metric

1333: \beq \label{lineel}

1334:   ds^2\ = \ \eta_{\mu \nu}(x)\, dx^\mu dx^\nu + g_{i \jb}(y)\, dy^i dy^\jb\ ,

1335: \eeq

1336: where $\eta_{\mu \nu},\mu,\nu=0,\ldots,3$ 

1337: is a four-dimensional Minkowski metric and 

1338: $g_{i \jb},i,\jb=1 \ldots 3$ is a Calabi-Yau metric. Part of the four dimensional fields 

1339: arise as variations around this background metric. They correspond to the four-dimensional 

1340: graviton and the geometric deformations $v^A(x)$ and $z^K(x)$ defined in \eqref{def-v} and \eqref{cs}.  

1341: Variations of the off-diagonal entries of this metric vanish due to the fact that $Y$ does 

1342: not admit harmonic one-forms. Accordingly we expand 

1343: the ten-dimensional gauge-potentials introduced in \eqref{defHFF1} in 

1344: terms of harmonic 

1345: forms on $Y$

1346: \bea \label{fieldexp}

1347:   \hat C_1 &=& A^0(x)\ ,\qquad 

1348: \hat B_2\, = \, B_2(x) +  b^A(x) \, \omega_A\ ,\quad  

1349:        A\ =\ 1,\ldots, h^{(1,1)}\ ,\\

1350:   \hat C_3 &=& 

1351: A^A(x) \wedge \omega_A + \,

1352:               \xi^\Kh(x)\, \alpha_\Kh - \tilde \xi_\Kh(x)\, \beta^\Kh\ , \quad \Kh\ =\ 0,\ldots, h^{(2,1)}\ . \nn 

1353: \eea  

1354: Here $b^A,\xi^\Kh,\tilde \xi_\Kh$ are four-dimensional scalars, 

1355: $A^0,A^A$ are one-forms and  $B_2$ is a two-form. 

1356: The ten-dimensional one-form $\hat C_1$

1357: only contains a four-dimensional one-form $A^0$ in the expansion

1358: \eqref{fieldexp} since a Calabi-Yau 

1359: threefold has no harmonic one-forms. 

1360: 

1361:  

1362: The geometric deformations $v^A,z^K$ together with the fields defined in 

1363: the expansions \eqref{fieldexp} assemble into a gravity multiplet $(g_{\mu\nu},A^0)$,

1364: $h^{(1,1)}$ vector multiplets $(A^A, v^A, b^A)$,

1365: $h^{(2,1)}$ 

1366: hypermultiplets $(z^K,\xi^K,\tilde \xi_K)$ and one tensor multiplet 

1367: $(B_2,\phi,\xi^0,\tilde \xi_0)$ where we only give the bosonic 

1368: components.

1369: Dualizing the two-form $B_2$ to a scalar $a$ results in one

1370: further hypermultiplet. We summarize the bosonic spectrum in 

1371: table~\ref{tab-compIIAspec}.

1372: 

1373: \begin{table}[h]

1374: \begin{center}

1375: \begin{tabular}{| c | c | c |} \hline

1376:    \rule[-0.3cm]{0cm}{0.8cm} gravity multiplet  &

1377:    $1$ & {\small $(g_{\mu \nu},A^0)$} 

1378:    \\ \hline

1379:    \rule[-0.3cm]{0cm}{0.8cm} vector multiplets &

1380:    $h^{(1,1)}$ & {\small $(A^{A}, v^A,b^A)$}\\ \hline

1381:    \rule[-0.3cm]{0cm}{0.8cm} hypermultiplets  &

1382:    $h^{(2,1)}$ & 

1383:    {\small $(z^K,\xi^K,\tilde \xi_K)$}\\ \hline

1384: \rule[-0.3cm]{0cm}{0.8cm} tensor multiplet  &

1385:    1 &

1386:    {\small $(B_2,\phi,\xi^0,\tilde \xi_0)$} \\ \hline

1387: 

1388: \end{tabular}

1389: \caption{\small \label{tab-compIIAspec}

1390: \textit{ $N=2$ multiplets for Type IIA supergravity compactified on a Calabi-Yau manifold.}}

1391: \end{center}

1392: \end{table}

1393: In order to display the low energy  effective action in the standard 

1394: $N=2$ form one needs to redefine the field variables slightly.

1395: One combines the real scalars $v^A, b^A$ into  complex fields 

1396: $t^A$ as done in \eqref{def-t_II} and defines a four-dimensional 

1397: dilaton $D$ according to

1398: \beq \label{4d-dilaton}

1399:    e^{D} = e^{\phi} (\cK/6)^{-\frac{1}{2}}\ ,

1400: \eeq

1401: where $\cK$ is defined in \eqref{int-numbers}. Note that $v^A$, and hence the volume $\cK/6 = \text{Vol}_S(Y)$, 

1402: are evaluated in string frame. In this frame the ten-dimensional 

1403: Einstein-Hilbert term takes the form $\int \frac{1}{2} e^{-2\hat \phi} R * \mathbf{1}$

1404: and $J=v^A \omega_A$ is obtained from the internal part of this string frame metric. 

1405: The kinetic term for the ten-dimensional Einstein frame metric reads 

1406: $\int \frac{1}{2} R * \mathbf{1}$ and hence $J$ is related to $J_E$ in the 

1407: Einstein frame via $J = e^{\phi/2}J_E$.

1408: Inserting the field expansions \eqref{fieldexp} into \eqref{defHFF1}, \eqref{10dact},

1409: reducing the Riemann scalar $R$ by including the complex and K\"ahler 

1410: deformations and performing a Weyl rescaling to the standard Einstein-Hilbert term,

1411: one ends up with the four-dimensional $N=2$ effective action  

1412: \cite{N=2review,BCF,FS}

1413: \bea \label{IIA-4}

1414:   S^{(4)}_{\rm IIA} & = &\int -\tfrac12 R * \mathbf{1} 

1415:                      +  \tfrac12 \I \cN_{\Ah \Bh}\, F^{\Ah} \wedge *

1416:                      F^{\Bh}

1417:                    +  \tfrac12 \R \cN_{\Ah \Bh}\, F^{\Ah} \wedge  F^{\Bh} \\

1418:                     & & - G_{A B}\, dt^A \wedge *  d\bar t^B 

1419:                         - h_{uv}\, d\tilde q^u \wedge * d\tilde q^v \ , \nn

1420: \eea

1421: where $F^{\Ah} = dA^{\Ah}$. The couplings of the vector multiplets in the action

1422: \eqref{IIA-4} are encoded by the metric $G_{A B}$ and the complex matrix 

1423: $\cN_{\Ah \Bh}$. $G_{A B}$ only depends on the moduli $t^A$ (or rather

1424: their imaginary parts) and is defined in \eqref{Kmetric} and \eqref{Kpot_ks}. 

1425: The gauge-kinetic coupling matrix $\cN_{\Ah \Bh}$ also depends on 

1426: the scalars $t^A$ and is given explicitly in \eqref{def-cN}.

1427: It can be calculated from the same holomorphic prepotential like $G_{AB}$ as 

1428: explained in appendix \ref{specialGeom}.

1429: 

1430: Next let us turn to the couplings of the hypermultiplet sector which are encoded in the

1431: quaternionic metric $h_{uv}$. From the Kaluza-Klein reduction one obtains \cite{FS}

1432: \bea \label{q-metr}

1433:   h_{uv}\, d\tilde q^u\,  d\tilde q^v &=&  (dD)^2 + G_{K \bar L}\, dz^K d\bar z^L

1434:                                +\tfrac{1}{4}e^{4D}\big(da -(\tilde \xi_\Kh d\xi^\Kh - \xi^\Kh d\tilde \xi_\Kh) \big)^2 \\

1435:                               && - \tfrac{1}{2} e^{2D} (\text{Im}\; \cM)^{-1\ \Kh \Lh}

1436:                                    \big(d\tilde \xi_\Kh - \cM_{\Kh \Nh} d\xi^\Nh \big)

1437:                                    \big(d\tilde \xi_\Lh - \bar

1438:                                    \cM_{\Lh \Mh} d\xi^\Mh \big)\ ,\nn 

1439: \eea

1440: where $G_{K \bar L}$ is the metric on the space of complex structure deformations given in 

1441: \eqref{chi_barchi} and \eqref{csmetric}. The complex coupling matrix $\cM_{\Kh \Lh}$ 

1442: appearing in \eqref{q-metr} depends on the complex structure deformations $z^K$ and is defined as

1443: \cite{CDAF}

1444: \bea \label{defM}

1445:   \int \alpha_\Kh \wedge * \alpha_\Lh&=&-(\text{Im}\; \cM +(\text{Re}\; \cM)

1446:   (\text{Im}\; \cM)^{-1}(\text{Re}\; \cM))_{\Kh \Lh}\ , \nn\\

1447:   \int \beta^\Kh \wedge * \beta^\Lh &=&-(\text{Im}\; \cM)^{-1\ \Kh \Lh}\ ,  \\

1448:   \int \alpha_\Kh\wedge * \beta^\Lh &=& 

1449:   -((\text{Re}\; \cM)(\text{Im}\; \cM)^{-1})_{\Kh}^\Lh\ .  \nn

1450: \eea

1451: It can be calculated from the periods \eqref{pre-z} by using equation \eqref{gauge-c}.

1452: Thus also in the hypermultiplet sector all couplings are determined

1453: by a holomorphic prepotential and such metrics have been called dual or special

1454: quaternionic \cite{CFGi,FS}.

1455: 

1456: As we have just reviewed the $N=2$ moduli space 

1457: has the local product structure 

1458: \beq \label{N=2modsp}

1459:   \cM^{\rm SK} \times \cM^{\rm Q}\ ,

1460: \eeq

1461: where $\cM^{\rm SK}=\cM^{\rm ks}$ is the special K\"ahler manifold spanned

1462: by the scalars in the vector multiplets or in other words

1463: the (complexified) deformations of the Calabi-Yau K\"ahler form

1464: and $\cM^{\rm Q}$ is a dual quaternionic manifold spanned by

1465: the scalars in the hypermultiplets. 

1466: $\cM^{\rm Q}$ has a special K\"ahler submanifold spanned by the 

1467: complex structure deformations $\cM^{\rm cs}$. 

1468: 

1469: This ends our short review of Calabi-Yau compactifications of type IIA

1470: supergravity. There is a second $N=2$ supersymmetric theory in 

1471: ten dimensions which is the low energy effective theory of type IIB

1472: string theory. Reviewing the Calabi-Yau reduction of this theory will be 

1473: the task of the next section.

1474: 

1475: 

1476: \section{Type IIB on Calabi-Yau manifolds}

1477: \label{revIIB}

1478: 

1479: Now we turn to the review of type IIB compactifications 

1480: on Calabi-Yau spaces \cite{BGHL}.

1481: Type IIB supergravity is maximal supersymmetric in ten dimensions

1482: and possesses two gravitinos of the same chirality. It 

1483: consists of the same NS-NS fields as type IIA: the scalar  

1484: dilaton $\hat \phi$, the metric $\hat g$ and a two-form $\hat B_2$. 

1485: In the R-R sector type IIB consists of even forms, 

1486: the axion $\hat C_0$, a two-form $\hat C_2$ and a 

1487: four-form  $\hat C_4$.

1488: The low energy effective action in the $D=10$ 

1489: Einstein frame is given by 

1490: \cite{JPbook} 

1491: \begin{eqnarray}\label{10d-lagr}

1492:   S^{(10)}_{IIB}&=&

1493:   \int -\tfrac{1}{2} \hat R * \mathbf{1} - \tfrac{1}{4} d\hat \phi\wedge *d \hat \phi

1494:   -\tfrac{1}{4} e^{-\hat \phi} \hat H_3 \wedge* \hat H_3   \\

1495:   &&- \tfrac{1}{4} e^{2\hat \phi} \hat F_1 \wedge * \hat F_1 -

1496:   \tfrac{1}{4} e^{\hat \phi} \hat F_3 \wedge * \hat F_3 -

1497:   - \tfrac{1}{8}\hat F_{5} \wedge *\hat F_{5}

1498:    -\tfrac{1}{4} \hat C_4 \wedge \hat H_3 \wedge \hat F_3\ ,

1499: \nonumber  

1500: \end{eqnarray}

1501: with the field strengths defined as

1502: \begin{eqnarray}

1503:   \hat H_3 \ =\ d \hat B_2\ , \quad \hat F_1 = d\hat C_0\ ,\quad 

1504:   \hat F_{q+1}\ =\ d \hat C_q - \hat C_{q-2} \wedge \hat H_3\ ,\quad q=2,4\ . \label{fieldstr}

1505: \end{eqnarray}

1506: The self-duality condition $\hat F_5=*\hat F_5$ is

1507: imposed at the level of the equations of motion.

1508: 

1509: As in the type IIA compactifications discussed in the previous section 

1510: we use the Ansatz \eqref{lineel} 

1511: for the ten-dimensional background metric. Fluctuations around this background 

1512: metric are parameterized by the four-dimensional graviton $g_{\mu \nu}$ and 

1513: the geometric deformations of the Calabi-Yau metric. More precisely, we find 

1514: $h^{(1,1)}$ real K\"ahler structure deformations $v^A$ 

1515: introduced in \eqref{def-v} and $h^{(2,1)}$ complex structure deformations $z^K$ 

1516: introduced in \eqref{cs}.

1517: The type IIB gauge potentials appearing in the Lagrangian 

1518: \eqref{10d-lagr} are similarly 

1519: expanded in terms of harmonic forms on $Y$ according to 

1520: \begin{eqnarray}\label{CYexpansion}

1521:   \hat B_2 &=& B_2(x) + b^A(x)\, \omega_A\ , \qquad

1522: \hat C_2\ =\ C_2(x) + c^A (x)\,\omega_A\ , \quad A=1,\ldots,h^{(1,1)}\ , \\

1523:   \hat C_4 &=& D_2^A(x) \wedge \omega_A + V^{\hat K}(x) \wedge

1524:                \alpha_{\hat K} - U_{\hat K}(x) \wedge \beta^{\hat K} +

1525:                \rho_A(x)\, \tilde \omega^A\ , 

1526: \quad \hat K=0,\ldots,h^{(1,2)}.\nonumber

1527:   \label{full-exp}

1528: \end{eqnarray}

1529: The four-dimensional fields appearing in the expansion \eqref{CYexpansion}

1530: are the scalars $b^A(x)$, $c^A(x)$ and $\rho_A(x)$, 

1531: the one-forms $V^{\hat K}(x)$ and

1532: $U_{\hat K}(x)$ and the two-forms $B_2(x),C_2(x)$ and $D_2^A(x)$.

1533: The self-duality condition of $\hat F_5$ eliminates half of the 

1534: degrees of freedom in $\hat C_4$ and in this section we choose to eliminate

1535: $D^A_2$ and $U_{\hat K}$ in favor of $\rho_A$ and $V^{\hat K}$.

1536: Finally, the two type IIB scalars $\hat \phi, \hat C_0$ also appear as

1537: scalars in $D=4$ and therefore we drop the hats henceforth

1538: and denote them by $\phi, C_0$.

1539: 

1540: In summary the massless $D=4$ spectrum consists of 

1541: the gravity multiplet with bosonic components $(g_{\mu \nu}, V^0)$,

1542: $h^{(2,1)}$ vector multiplets with bosonic components $(V^{K}, z^{K})$,

1543: $h^{(1,1)}$ hypermultiplets with bosonic components

1544: $(v^A, b^A, c^A, \rho_A)$ 

1545: and one double-tensor multiplet \cite{BVT} with bosonic components

1546: $(B_2, C_2, \phi, C_0)$ which can be dualized to an additional (universal) 

1547: hypermultiplet. The four-dimensional spectrum is

1548: summarized in Table \ref{tab-compIIBspec}.

1549: 

1550: \begin{table}[h]

1551: \begin{center}

1552: \begin{tabular}{| c | c | c |} \hline

1553:    \rule[-0.3cm]{0cm}{0.8cm} gravity multiplet  &

1554:    $1$ & {\small $(g_{\mu \nu},V^0)$} 

1555:    \\ \hline

1556:    \rule[-0.3cm]{0cm}{0.8cm} vector multiplets &

1557:    $h^{(2,1)}$ & {\small $(V^{K}, z^{K})$}\\ \hline

1558:    \rule[-0.3cm]{0cm}{0.8cm} hypermultiplets  &

1559:    $h^{(1,1)}$ &

1560:    {\small $(v^A, b^A, c^A, \rho_A)$

1561: }\\ \hline

1562: \rule[-0.3cm]{0cm}{0.8cm} double-tensor multiplet  &

1563:    1 &

1564:    {\small $(B_2, C_2,\phi,C_0)$ 

1565: }\\ \hline

1566: 

1567: \end{tabular}

1568: \caption{\small \label{tab-compIIBspec}

1569: \textit{ $N=2$ multiplets for Type IIB supergravity compactified on a Calabi-Yau manifold.}}

1570: \end{center}

1571: \end{table}

1572: 

1573: The $N=2$ low energy effective action is computed by inserting

1574: \eqref{fieldstr} and \eqref{CYexpansion} into the action \eqref{10d-lagr}

1575: and integrating over the Calabi-Yau manifold.

1576: For the details we refer the reader to the literature 

1577: \cite{BGHL,Michelson,DallAgata,LM} and only recall the results here.

1578: One finds 

1579: \begin{eqnarray}\label{N=2}

1580:   S_{IIB}^{(4)} &=& \int - \tfrac{1}{2} R *\! {\bf 1} 

1581: + \tfrac{1}{4} \RE\cM_{\hat K \hat L} {F}^{\hat K} \wedge {F}^{\hat L} + 

1582:   \tfrac{1}{4} \IM_{\hat K\hat L} {F}^{\hat K} \wedge * {F}^{\hat L}\nonumber\\

1583: &&- G_{K L} d z^K \wedge * d \bar{z}^{L} 

1584:   - G_{AB} d t^A \wedge * d \bar t^B

1585:   -  d D \wedge * d D -  \tfrac{1}{24} e^{2 D} \cK \dd l \wedge * \dd l \nonumber\\

1586: && 

1587:   - \tfrac{1}{6} e^{2D} \cK G_{AB}

1588:   \big(\dd c^A - l \dd b^A \big)\wedge * \big( \dd c^B - l \dd b^B

1589:   \big)\\

1590:   && - \tfrac{3}{8\cK} e^{2D} G^{AD} \big( \dd \rho_A - 

1591:     \cK_{ABC} c^B \dd b^C \big) \wedge\! *\big( \dd \rho_D -

1592:     \cK_{DEF} c^E \dd b^F \big) \nonumber \\

1593: && -\tfrac{1}{4}e^{-4D} \dd B_2 \wedge * \dd B_2 - \tfrac{1}{24}  

1594:    e^{-2D} \cK 

1595:   \big( \dd C_2 - l \dd B_2 \big) \wedge *\big( \dd C_2 - l \dd B_2 \big)

1596:   \nn  \\

1597:   && -  \tfrac{1}{2} dC_2 \wedge \big( \rho_A \dd b^A  - b^A d\rho_A \big)

1598:      +\tfrac{1}{2} dB_2 \wedge c^A d \rho_A - \tfrac{1}{4}\cK_{ABC} c^A c^B dB_2 \wedge \dd b^C \nonumber \ ,

1599: \end{eqnarray}

1600: where $F^{\hat K}=dV^{\hat K}$. 

1601: The gauge kinetic matrix $\cM_{\hat K\hat L}$ is related to the metric

1602: on $H^3(Y)$ and given in \eqref{defM}. The metric $G_{K L}(z,\bar z)$ which appears in \eqref{N=2}

1603: is the metric on the space of complex structure deformations given in \eqref{csmetric}.

1604: It is a special K\"ahler metric in that it is entirely determined by the holomorphic prepotential $\mathcal{F}(z)$ 

1605: \cite{Strominger2,CdO}. On the other hand, the metric $G_{AB}$ in \eqref{N=2} is the metric 

1606: on the space of K\"ahler deformations defined in \eqref{Kmetric}.

1607: 

1608: In order to entirely express \eqref{N=2}  in terms

1609: of vector- and hypermultiplets we dualize the 

1610: $D=4$ two-forms $B_2,C_2$ to scalar fields. This can be done, since $B_2$ and $C_2$ are massless 

1611: and posses the gauge symmetries $C_2  \rightarrow C_2 + d\Lambda_1$ and $B_2 \rightarrow B_2 + d\tilde \Lambda_1$. 

1612: Let us first dualize $C_2$.

1613: We replace $dC_2$ with $D_3$ and add the Lagrange multiplier $\frac12 h\, dD_3$ such 

1614: that the differentiation with respect to $h$ yields $dD_3=0$. Locally this ensures 

1615: that $D_3=dC_2$ for some two-form $C_2$.  The terms in \eqref{N=2} involving $D_3$ are simply

1616: \beq

1617:   \cL_{C_2} = - \tfrac{g}{4}   

1618:    \big( D_3 - C_0\, \dd B_2 \big) \wedge * \big( D_3 - C_0\, \dd B_2 \big)

1619:     -  \tfrac{1}{4} D_3 \wedge J_1 +\tfrac12 D_3 \wedge dh\ ,

1620: \eeq

1621: where we abbreviated $g = \frac{1}{6} e^{-2D} \cK$ and $J_1 = \rho_A \dd b^A  - b^A d\rho_A$.

1622: Now we can consistently eliminate $D_3$ in favor of $h$ by its equation of motion. The 

1623: dualized Lagrangian takes the form

1624: \beq \label{dual_h}

1625:   \cL_{h} = - \tfrac{1}{4 g} \big(dh - \tfrac12 J_1 \big) \wedge *

1626:                                       \big(dh - \tfrac12 J_1 \big)

1627:             + \tfrac12 C_0\, dB_2 \wedge \big(dh - \tfrac12 J_1 \big)\ . 

1628: \eeq  

1629: Similarly we can dualize the two-from $B_2$ into a scalar $\tilde h$. Having replaced 

1630: $C_2,B_2$ by $h,\tilde h$ in \eqref{N=2} the effective action can be written in the standard 

1631: $N=2$ form \cite{BaggerW,dWvP,N=2review}

1632: \begin{eqnarray}

1633:   S_{IIB}^{(4)} & = & \int -\tfrac{1}{2}R *{\mathbf 1} 

1634: + \tfrac{1}{4} \RE\cM_{\Kh\Lh} {F}^\Kh \wedge {F}^\Lh + \tfrac{1}{4} \I

1635:   \cM_{\Kh\Lh} {F}^\Kh \wedge * {F}^\Lh\nonumber\\

1636: &&\qquad - G_{KL} \dd z^K \wedge *\dd \bar{z}^{L} 

1637: - h_{pq}\, \dd \tilde q^{p} \wedge * \dd \tilde q^{q} \ ,

1638:   \label{action3}

1639: \end{eqnarray}

1640: where  $q^{p}$ denote the coordinates for all

1641: $h^{(1,1)}+1$ hypermultiplets. The metric $h_{pq}$ is a quaternionic

1642: metric  explicitly given by \cite{FS}

1643: \begin{align} \label{q-metrB}

1644:   h_{pq}\, d\tilde q^p\,  d\tilde q^q &=   (d D)^2 +  G_{AB} d t^A d\bar t^B

1645:                                             + \tfrac{1}{24} e^{2 D} \cK (\dd C_0)^2 \nonumber\\

1646: & + \tfrac{1}{6} e^{2D} \cK G_{AB} \big(\dd c^A - C_0\, \dd b^A \big)\big( \dd c^B - C_0\, \dd b^B \big) \\

1647:   & + \tfrac{3}{8\cK} e^{2D} G^{AD} \big( \dd \rho_A - 

1648:     \cK_{ABC} c^B \dd b^C \big) \big( \dd \rho_D -

1649:     \cK_{DEF} c^E \dd b^F \big) \nonumber \\ 

1650:    & + \tfrac{3}{2 \cK} e^{2D}\big(dh - \tfrac12 (\rho_A \dd b^A  - b^A d\rho_A) \big)^2 \nn\\

1651:   & +  \tfrac12 e^{4D} \big(d\tilde h + C_0\, dh + c^A d\rho_A  +  \tfrac12 C_0\, (\rho_A \dd b^A  - b^A d\rho_A) 

1652:                             - \tfrac{1}{4}\cK_{ABC} c^A c^B \dd b^C \big)^2  . \nn

1653: \end{align}

1654: In summary the scalar moduli space $\cM^{\rm SK}  \times  \cM^{\rm Q}$ of the $N=2$ theory is

1655: the product of a quaternionic manifold $\cM^{\rm Q}$ spanned by the scalars

1656: $q^{p}$ with metric \eqref{q-metrB} and a special

1657: K\"ahler manifold $\cM^{\rm SK} = \cM^{\rm cs}$ spanned by the scalars $z^K$.

1658: The complexified K\"ahler structure deformations span a special K\"ahler manifold $\cM^{\rm ks}$

1659: inside $\cM^{\rm Q}$. In \cite{FS} it was shown that the quaternionic space 

1660: can be constructed from the prepotential of $\cM^{\rm ks}$ such that 

1661: $\cM^{\rm Q}$ is a special quaternionic manifold.

1662:   

1663: This ends our brief summary of type IIB compactified on

1664: Calabi-Yau threefolds and its $N=2$ low energy effective action. 

1665: We have seen that the effective actions of the type IIA and type

1666: IIB indeed take the standard $N=2$ form. In both cases the metrics 

1667: on the special K\"ahler and special quaternionic manifold are encoded by 

1668: a corresponding prepotential. However, the role of the K\"ahler and complex 

1669: structure deformations is exchanged in type IIA and type IIB compactifications.

1670: As we will discuss momentarily this can be traced back to an underlying symmetry  

1671: which finally enables us to identify both effective theories in the 

1672: large volume -- large complex structure limit.

1673: 

1674: 

1675: \section{N=2 Mirror symmetry}

1676: \label{revMirror}

1677: 

1678: In this section we briefly discuss mirror symmetry for Calabi-Yau 

1679: compactifications \cite{Mirror}. From a mathematical point of view, mirror 

1680: symmetry is a duality in the moduli space of Calabi-Yau manifolds. 

1681: It states that for a given Calabi-Yau manifold $Y$, there exists

1682: a mirror Calabi-Yau $\tilde Y$ such that 

1683: \beq \label{Hod_id}

1684:   h^{(1,1)}(Y) = h^{(2,1)}(\tilde Y)\ , \qquad  h^{(2,1)}(Y) = h^{(1,1)}(\tilde Y)\ .

1685: \eeq   

1686: Applied to the Hodge diamond \eqref{hodge_diamond} this amounts to a reflection along the

1687: diagonal. In other words, mirror symmetry identifies the odd and even cohomologies \eqref{odd_even_cohom}

1688: of two topological distinct Calabi-Yau spaces

1689: \beq \label{cohom_id}

1690:   H^{ev}(Y)\  \cong \ H^{odd}(\tilde Y)\ ,\qquad H^{odd}( Y)\  \cong \ H^{ev}(\tilde Y)\ .

1691: \eeq 

1692: Moreover, it is much stronger than that, since

1693: it also implies an identification of the moduli spaces of deformations of $Y$ and $\tilde Y$. 

1694: As given in \eqref{geom-mod} the geometric moduli space of a Calabi-Yau manifold 

1695: is a local product of two special K\"ahler manifolds $\cM^{\rm ks}$ and $\cM^{\rm cs}$.

1696: Their complex dimensions are exactly given by $h^{(1,1)}$ and $h^{(2,1)}$. Motivated 

1697: by \eqref{Hod_id} mirror symmetry conjectures the identifications

1698: \beq \label{Modspace_id}

1699:   \cM^{\rm ks}(Y)\  \equiv\  \cM^{\rm cs}(\tilde Y)\ , \qquad 

1700:   \cM^{\rm cs}(Y)\ \equiv\ \cM^{\rm ks}(\tilde Y)\ ,

1701: \eeq   

1702: as special K\"ahler manifolds. Recall that the geometry of $\cM^{\rm cs}(Y)$ 

1703: and $\cM^{\rm cs}(\tilde Y)$ are encoded in the variations of the holomorphic

1704: three-forms $\Omega$ and $\tilde \Omega$ of the two Calabi-Yau manifolds 

1705: $Y$ and $\tilde Y$. These can be expanded in a real symplectic basis of $H^{3}(Y)$ and 

1706: $H^{3}(\tilde Y)$ respectively

1707: \beq

1708:   \Omega(z) = Z^\Kh \alpha_\Kh - \cF_\Kh \beta^\Kh\ , \qquad 

1709:   \tilde \Omega(\tilde z) =\tilde Z^\Ah \alpha_\Ah - \tilde \cF_\Ah \beta^\Ah\ ,

1710: \eeq

1711: Under the large volume mirror map 

1712: the coordinates on the two manifolds $\cM^{\rm ks}(Y)$ and $\cM^{\rm cs}(\tilde Y)$ as 

1713: well as $\cM^{\rm cs}(Y)$ and $\cM^{\rm ks}(\tilde Y)$ 

1714: are identified as 

1715: \beq \label{mirror-map}

1716:   t^A = {\tilde Z^A(\tilde z)}/{\tilde Z^0(\tilde z)} \ , \qquad {Z^K(z)}/{Z^0(z)} = \tilde t^K

1717: \eeq

1718: where $t^A$ and $\tilde t^K$ are the complexified K\"ahler deformations of $Y$ and $\tilde Y$. 

1719: Equation \eqref{mirror-map} implies that $t^A,\tilde t^K$ are identified with special coordinates 

1720: on $\cM^{\rm cs}$. Furthermore, recall that due to the special K\"ahler property the metric on both 

1721: moduli spaces is encoded by a prepotential.  Applying \eqref{Modspace_id} it follows that 

1722: these prepotentials $f_Y(t)$ and $f_{\tilde Y}(\tilde z)$ as well as 

1723: $f_Y(z)$ and $f_{\tilde Y}(\tilde t)$ are identified under mirror symmetry. 

1724: One immediately notices, that this can not be the full truth, since $\cM^{\rm ks}$ and $\cM^{\rm cs}$ have 

1725: a different structure. $\cM^{\rm ks}$ is a cone and admits the

1726: simple prepotential $f(t) = - \frac16 \cK_{ABC} t^A t^B t^C$, while the metric 

1727: on $\cM^{\rm cs}$ is determined in terms of the (generically complicated)

1728: periods of the holomorphic three-form. Hence, one expects corrections to

1729: $f_Y(t)$ and $f_{\tilde Y}(\tilde t)$. These corrections get a physical interpretation

1730: as soon as mirror symmetry is embedded in string theory.

1731: They are due to strings wrapping two-cycles in $Y$ called world-sheet instantons.

1732: Schematically one identifies 

1733: \beq

1734:    f_{Y}(t) = t^3 + \mathcal{O}(e^{-t})=f_{\tilde Y}(\tilde z)\ ,\qquad f_Y(z) 

1735:             = \tilde t^3 + \mathcal{O}(e^{-\tilde t})=f_{\tilde Y}(\tilde t)\ .

1736: \eeq

1737: One can also turn the argument around and use mirror symmetry as a very powerful tool to calculate the world-sheet 

1738: instanton corrections $\mathcal{O}(e^{-t})$ as done in the pioneering paper \cite{CdOGP}.

1739: In most cases this is much simpler then a direct calculation of the world-sheet instanton

1740: contributions.

1741: 

1742: The most prominent applications of mirror symmetry in string theory

1743: is the identification of type IIA string theory compactified on $Y$ with 

1744: type IIB string theory compactified on $\tilde Y$. It matches 

1745: the full string theories including their low energy limits

1746: and supersymmetric D-brane states. With the material presented in this chapter we can check

1747: it on the level of the effective action by comparing \eqref{IIA-4} with 

1748: \eqref{action3}. This amounts to matching the moduli spaces of the 

1749: corresponding four-dimensional $N=2$ theories which take the standard $N=2$ 

1750: form \eqref{N=2modsp}. Since we already discussed the special K\"ahler part in \eqref{N=2modsp},

1751: let us now concentrate on the quaternionic manifolds $\cM^{\rm Q}(Y)$

1752: and $\cM^{\rm Q}(\tilde Y)$. In accordance with \eqref{cohom_id} and \eqref{mirror-map}

1753: one identifies the basis elements $(1, \omega_K,\tilde \omega^K, \vol(Y))$ of $H^{ev}(Y)$ with 

1754: the basis $(\alpha_\Kh,\beta^\Kh)$ of $H^{odd}(\tilde Y)$ as

1755: \beq \label{basis_id}

1756:   1\leftrightarrow \alpha_0\ ,\quad \omega_K \leftrightarrow \alpha_K\ ,\quad    \vol(Y) \leftrightarrow \beta^0\ ,

1757:   \quad \tilde \omega^K \leftrightarrow \beta^K\ .

1758: \eeq

1759: We will work in this basis in the following.

1760: Let us now construct the explicit map for the quaternionic coordinates by using 

1761: a slightly non-standard argument. We intend to apply the fact, that the fields of 

1762: the quaternionic space describe the coupling to D-branes, which are extended 

1763: non-perturbative objects present in both type II string theories. We will discuss the 

1764: low energy dynamics and supersymmetry conditions of these objects more carefully in section 

1765: \ref{D-branes}. All we need for constructing the mirror map for the quaternionic spaces 

1766: is there coupling to the R-R forms in the supergravity theory and some information 

1767: about supersymmetric branes in type IIA and type IIB string theory. 

1768: It will become clear in section \ref{D-branes}, that the only supersymmetric Euclidean 

1769: D-branes wrapping a cycle in a Calabi-Yau manifold are $D2$ 

1770: branes in Type IIA and $D(-1)$, $D1$, $D3$ and $D5$ branes in type IIB.

1771: The Chern-Simons action describes the coupling of the brane world-volume to the forms

1772: \beq \label{CS_coupling}

1773:    \text{IIA:}\quad (\sum_{p\ even} \hat C_{p} \wedge e^{-\hat B_2})_3\ , \qquad 

1774:    \text{IIB:}\quad (\sum_{p\ odd} \hat C_{p} \wedge e^{-\hat B_2})_q\ ,\ 

1775:    q = 0,2,4,6\ , 

1776: \eeq 

1777: where $\hat C_p$ and $\hat B_2$ are the ten-dimensional R-R and NS-NS forms introduced in 

1778: section \ref{revIIA} and \ref{revIIB}. By $(\ldots)_q$ we indicate that we 

1779: only consider the $q-$form appearing in the sum of forms inside the parenthesis.

1780: Supersymmetry implies that the Euclidean D-branes, wrap cycles which are dual 

1781: to harmonic forms. But the only odd harmonic forms are $(\alpha_{\hat K},\beta^{\hat K})$, while the even 

1782: harmonic forms are 

1783: $(1,\omega_K, \tilde \omega^K,\vol(Y))$. 

1784: Next we match the Chern-Simons couplings \eqref{CS_coupling} for IIA and IIB Euclidean

1785: D-branes. We decompose \eqref{CS_coupling} on the respective cohomology basis elements

1786: by using the expansions \eqref{CYexpansion} of $\hat B_2$ and the R-R forms $\hat C_0,\hat C_2,\hat C_4$ 

1787: as well as the expansion \eqref{fieldexp} of $\hat C_3$.

1788: Applying the identification \eqref{basis_id} of the basis forms we find for the coefficients

1789: of $\alpha_\Kh$ and $(1,\omega_K)$ that     

1790: \beq

1791:   \xi^0 = C_0\ ,\qquad \xi^K = c^K - C_0\ b^K\ . 

1792: \eeq

1793: Identifying the coefficients of $\beta^\Kh$ and $(\tilde \omega^K, \vol(Y))$

1794: yields higher powers in $\hat B_2$ and we find \footnote{We have replaced $\int C_6$ by 

1795: $h + \tfrac12 \rho_A b^A$.

1796: This can be done since $C_6$ is dual to $C_2$, which was dualized to $h$ in \eqref{dual_h}.}

1797: \bea

1798:   \tilde \xi_K &=& \rho_K - \cK_{KLM} c^L b^M + \tfrac{1}{2} C_0\ \cK_{KLM} b^L b^M\ , \\

1799:   \tilde \xi_0 &=& h - \tfrac{1}{2} \rho_K b^K + \tfrac{1}{2} \cK_{KLM} c^K b^L b^M 

1800:                      - \tfrac{1}{6}C_0\  \cK_{KLM} b^K b^L b^M\ .\nn

1801: \eea

1802: It remains to identify the space-time two-forms from the 

1803: NS-NS sectors. Since $B_2^A$ and $B^B_2$ are the only remaining two-forms in the spectrum, we 

1804: are forced to set $B_2^A = B^B_2$. Dualized into scalars this amounts to

1805: \beq

1806:     a = 2\tilde h + C_0\, h + \rho_K(c^K - C_0\, b^K)  

1807: \eeq

1808: Thus, by using the explicit form of the Chern-Simons coupling to D-branes,

1809: one can infer the mirror map for the coordinates on the quaternionic space.

1810: Of course, that the established map indeed transforms $h^A_{uv}$ given in \eqref{q-metr} 

1811: into $h^B_{uv}$ given in \eqref{q-metrB} can be checked by direct calculation as done in \cite{BGHL}.  

1812: 

1813: This ends our review section on Calabi-Yau compactifications of type IIA 

1814: and type IIB supergravity. We now turn to their orientifold versions which 

1815: break $N=2$ to $N=1$. The aim of the next chapter is 

1816: to determine the characteristic data of the resulting supergravity theory. 

1817: 

1818: 

1819: 

1820: 

1821: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1822: %

1823: %  Chapter: Effective actions Orientifolds

1824: %

1825: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1826: 

1827: 

1828: 

1829: \chapter{Effective actions of Type II Calabi-Yau orientifolds}

1830: \label{effective_actO}

1831: 

1832: In this chapter we discuss the four-dimensional low energy effective supergravity theory 

1833: obtained by compactifying type IIA and type IIB string theory on Calabi-Yau orientifolds.

1834: Before entering the calculations we review some aspects of D-branes and orientifolds 

1835: in section \ref{D-branes}. In particular, we introduce the low energy effective action

1836: for D-branes. Later on this will allow us to comment on corrections due to wrapped 

1837: Euclidean D-branes to the bulk supergravity theory.

1838: %in sections \ref{IIB_orientifolds}, \ref{IIA_orientifolds} and \ref{non-pert_sup}. 

1839: As we already explained in section \ref{braneworlds} the inclusion of 

1840: space-time filling D-branes is essential for consistency. However, we freeze their moduli 

1841: and matter fields such that they do not appear in the low energy effective action.\footnote{%

1842: This restriction was weakened e.g.~in \cite{GGJL,JL}, where the coupling to $D3$- and $D7$-bane moduli 

1843: was determined by using the low energy effective action of the $D$ branes.}   

1844: In a next step we turn to the main issue of this chapter and perform 

1845: a Kaluza-Klein reduction by implementing the orientifold conditions and extract the 

1846: resulting $N=1$ supergravity theory (sections \ref{oprojections} -- \ref{IIA_orientifolds}). 

1847: Specifically we determine the K\"ahler potential and the gauge-kinetic 

1848: coupling functions encoding the low energy effective theory. We end our analysis by checking 

1849: mirror symmetry in the large complex structure and large volume limit in section \ref{Mirror_orientioflds}.

1850: A derivation of a flux induced superpotential and possible gaugings will be presented in 

1851: chapter \ref{fluxesAB}. 

1852: 

1853: 

1854: \section{D-branes and orientifolds}

1855: \label{D-branes}

1856: 

1857: In this section we provide more details on D-branes and orientifolds 

1858: as used in the construction of brane-world scenarios.

1859: As already mentioned in section \ref{braneworlds} 

1860: brane world scenarios are currently one of the promising approaches 

1861: to construct phenomenologically interesting models from

1862: string compactification \cite{reviewPP}. They consist of space-time filling 

1863: D-branes serving as source for Abelian or non-Abelian gauge theories.

1864: String theory implies a low energy effective action for this gauge theory

1865: as well as the couplings to the bulk fields introduced in chapter \ref{TypeII}. 

1866: More precisely, the gauge theory and the coupling to the NS-NS fields $\hat \phi$, $\hat g$ and $\hat B_2$ 

1867: is captured by the Dirac-Born-Infeld action. The most simple example is provided by a single 

1868: $Dp$-brane, which admits an $U(1)$ gauge theory on its $p+1$-dimensional world-volume. The corresponding 

1869: bosonic part of the Dirac-Born-Infeld action reads in string frame \cite{D-branes,JPbook}

1870: \begin{equation} \label{DBI}

1871:    S_{\text{DBI}}^{\text{sf}}=-T_p\int_\WV d^{p+1}\xi\:e^{-\hat \phi}

1872:          \sqrt{-\det\left(\Em^*(\hat g+\hat B_2)_{\hat \mu \hat \nu}+2\pi\alpha' \FD_{\hat \mu \hat \nu}\right)}\ ,

1873: \end{equation}

1874: where $T_p$ denotes the brane tension. The integral is taken over the $p+1$-dimensional 

1875: world-volume $\WV$ of the $Dp$-brane, which is embedded in the ten-dimensional space-time 

1876: manifold $\cM_{10}$ via the map $\Em:\WV\hookrightarrow \cM_{10}$. 

1877: The Dirac-Born-Infeld action \eqref{DBI} contains an $U(1)$ field strength 

1878: $\FD_{\hat \mu \hat \nu} = 2\partial_{[\hat \mu}A_{ \hat \nu]}$, which describes the 

1879: $U(1)$ gauge theory to all orders in $\alpha' \FD$ \cite{Leigh:jq}. 

1880: To leading order, the gauge theory reduces to an $U(1)$ gauge theory on the world-volume 

1881: $\WV$ of the brane. The dynamics of the $Dp$-brane is encoded in the embedding map $\Em$. Fluctuations

1882: around a given $\varphi$ are parameterized by charged scalar fields, which provide the matter

1883: content of the low-energy effective theory.    

1884:  

1885: Since $Dp$-branes also carry R-R charges \cite{JP}, they couple as extended objects to appropriate

1886: R-R forms of the bulk, namely the $p+1$-dimensional world-volume couples naturally to the R-R form $\hat C_{p+1}$. 

1887: Moreover, generically $D$-branes contain lower dimensional $D$-brane charges, and hence interact also with 

1888: lower degree R-R forms \cite{Douglas:1995bn}. 

1889: All these couplings to the bulk are implemented in the Chern-Simons action 

1890: \beq \label{CSaction}

1891:    S_{\text{CS}}=\mu_p\int_\WV\Em^*\Big(\sum_q \hat C_{q} \wedge e^{-\hat B_2}\Big) \wedge e^{2\pi\alpha'\FD}\  ,

1892: \eeq

1893: where $\mu_p$ is the charge of the D-branes.

1894: The lowest order terms in \eqref{CSaction} in the R-R fields are topological and represent the 

1895: R-R tadpole contributions to the low energy effective action. Additionally, 

1896: \eqref{CSaction} encodes the coupling of the gauged matter fields arising from 

1897: perturbations of $\varphi$ to the R-R forms. 

1898: The effective actions \eqref{DBI} and \eqref{CSaction} can be generalized to 

1899: stacks of D-branes \cite{Myers}. This gives rise to non-Abelian gauge theories

1900: and appropriate (intersecting) embeddings can yield Standard Model like gauge theories \cite{reviewPP}.

1901: 

1902: Generic brane world scenarios lead to non-supersymmetric low energy theories, which 

1903: are plagued by various instabilities due to runaway potentials for the bulk moduli. 

1904: In contrast, supersymmetric setups are under much better control and are therefore phenomenologically 

1905: favored. 

1906: However, the aim to preserve some supersymmetry poses strong conditions on the D-branes present 

1907: in the setup. D-branes which preserve half of the original supersymmetries 

1908: are called BPS branes and the corresponding supersymmetry conditions BPS conditions. 

1909: In brane-world setups with a ten-dimensional background space-time

1910: of the form $\Mext \times \Mint$ two types of branes will be of importance which 

1911: preserve four dimensional Poncar\'e invariance. 

1912: Firstly, one includes D-branes filling the space-time $\Mext$ and wrapping a cycle in 

1913: the manifold $\Mint$. These provide the gauge theory and matter fields just discussed. 

1914: Secondly, one might add Euclidean D-branes (called D-instantons) solely wrapping 

1915: a cycle in $\Mint$. They induce corrections to the supergravity theory and 

1916: their effects can be useful to stabilize bulk moduli. 

1917: The BPS conditions for both types of 

1918: branes demand that the brane tensions $T_p$ and charges $\mu_p$ are 

1919: equal. This ensures stability since the net force between BPS branes 

1920: vanishes \cite{JP}. Moreover, there are conditions 

1921: on the cycles $\Lambda_{Dp}$ in $\Mint$ wrapped by the branes.

1922: In \cite{BBS} it was shown that in a purely metric background with 

1923: $\Mint$ being a Calabi-Yau manifold the only allowed cycles are 

1924: special Lagrangian submanifolds of $\Mint$ in 

1925: type IIA and holomorphic submanifolds in type 

1926: IIB. More precisely special Lagrangian submanifolds are 

1927: three-cycles $\Lambda^{(3)}$ in $Y$ for which 

1928: \beq \label{spLagr-C}

1929:   \vol(\Lambda^{(3)})=\tilde \varphi^*( \R \Omega)\ ,\qquad \tilde \varphi^*( \I \Omega) = 0\ , \qquad \tilde \varphi^* J = 0\ ,

1930: \eeq

1931: where $\vol(\Lambda^{(3)})=\det^{1/2}(\tilde \varphi^* g)\, d^3\lambda$ 

1932: is the volume form on $\Lambda^{(3)}$, $J$ and $\Omega$ are the

1933: K\"ahler form and holomorphic three-form of $Y$ as in chapter \ref{TypeII}

1934: and $\tilde \varphi$ defines the embedding of the D-brane into $Y$.

1935: On the other hand, holomorphic submanifolds are even-dimensional cycles $\Lambda^{(2)},\Lambda^{(4)}$ in 

1936: $Y$ satisfying

1937: \beq \label{holom-C}

1938:   \vol(\Lambda^{(2)})=\tilde \varphi^*(J)\ ,\qquad \vol(\Lambda^{(4)})=\tfrac{1}{2}  \tilde \varphi^*(J \wedge J)\ ,

1939:   \qquad   \varphi^*(\Omega) = 0\ .

1940: \eeq

1941: It can be shown that the conditions \eqref{spLagr-C} and \eqref{holom-C} ensure that such cycles 

1942: minimizes their volume in their homology classes (see e.g. \cite{BBS}).

1943: 

1944: These conditions have to be adjusted as soon as one allows a non-trivial 

1945: background of supergravity forms \cite{MMMS,CU}. As an example, the 

1946: BPS conditions on the volume of the cycles in the presence of a non-trivial $\hat B_2$ field are given by \cite{MMMS}

1947: \bea  \label{calcond}

1948:    \text{IIA:} \qquad \vol_{DBI}(\Lambda^{(3)}_{Dp}) &=& e^{-i\theta_{Dp}}\ \tilde \varphi^* \big( \Omega \big)\ ,\\

1949:    \text{IIB:} \qquad  \vol_{DBI}(\Lambda^{(q)}_{Dp}) &=& 

1950:                        e^{-i\theta_{Dp}}\  \tilde\varphi^* \big(e^{-\hat B_2 + i J} \big)_q\ , \quad q=2,4,6\ , \nn

1951: \eea

1952: where $\vol_{DBI}(\Lambda^{(q)}_{Dp}) = \det^{1/2}(\tilde \varphi^*[g + \hat B_2])\, d^q\lambda$ is the Dirac-Born-Infeld 

1953: volume form on $\Lambda^{(q)}_{Dp}$.

1954: $e^{i\theta_{Dp}}$ denotes a constant phase which will be determined below.

1955: The BPS conditions involving the volume elements split into real and imaginary 

1956: parts, where the imaginary part has to vanish on $\Lambda^{(q)}_{Dp}$ by using reality 

1957: of $\vol_{DBI}(\Lambda^{(q)}_{Dp})$. The cycles $\Lambda^{(q)}_{Dp}$ satisfying the conditions

1958: \eqref{calcond} are called calibrated with respect to the form $e^{-i\theta_{Dp}}\,\Omega$ in type IIA

1959: and calibrated with respect to $e^{-i\theta_{Dp}}\, e^{-\hat B_2 + i J}$ in type IIB.

1960: In a setup with several D-branes some supersymmetry is preserved as 

1961: soon as all D-branes are calibrated with respect to the same form.

1962: However, as we already explained in section \ref{braneworlds} this 

1963: is not the end of the story, since consisted supersymmetric theories 

1964: have to include negative tension objects such as orientifold planes \cite{GKP}. 

1965: 

1966: Similar to D-branes, orientifold planes are hyper-planes 

1967: of the ten-dimensional background. They arise in string theories 

1968: which contain non-orientable world-sheets. Orientifold theories can 

1969: be constructed by starting from a closed string theory such as type 

1970: IIA or type IIB strings and dividing out a symmetry group \cite{AD,Ori} \footnote{%

1971: As usual, dividing out a symmetry can be understood as a gauge fixing.} 

1972: \beq \label{osym}

1973:   G \cup S\Omega_p,

1974: \eeq

1975: where $G$ is a group of target space symmetries and $\Omega_p$ is the 

1976: world-sheet parity, exchanging left and right movers. $S$ 

1977: contains operations, which render $S\Omega_p$ to be a 

1978: symmetry of the string theory. For orientifolds \eqref{osym} consists 

1979: of evidently perturbative symmetries of the string theory, which can be imposed 

1980: order by order in perturbation theory and are believed to be unbroken also 

1981: non-perturbatively. Specifically this implies that the orientifold projection 

1982: can be consistently imposed in a low energy description. 

1983: The orientifold planes are the hyper-surfaces left invariant by $S$.

1984: They naturally couple to the R-R forms and thus carry a charge. Moreover,

1985: they can have negative tension.\footnote{%

1986: Note that orientifold

1987: planes are to lowest order non-dynamical in string theory. This is not anymore true 

1988: to higher orders as can be inferred from their F-theory interpretation \cite{Sen}.} 

1989: This allows to construct consisted D-brane setups with some fraction of supersymmetry preserved.

1990: More precisely, in a background $\Mext \times \Mint$ orientifold planes 

1991: wrap cycles in $\Mint$ arising as the fix-point set of $S$. If these

1992: are calibrated with respect to the same form as the cycles wrapped 

1993: by the D-branes in the setup, the brane-orientifold setup can preserves some 

1994: supersymmetry. We will comment on these conditions later on in this chapter. 

1995: 

1996: Before we define the precise orientifold projections relevant for this work in section 

1997: \ref{oprojections}, let us first collect some possible symmetry 

1998: operations allowed in $S$. In the simplest example $S$ only consists 

1999: of a target space symmetry $\sigma:\cM_{10}\rightarrow \cM_{10}$, 

2000: such that $\Omega_p \sigma $ is a symmetry of the 

2001: underlying string theory. This will be the case for IIB orientifolds with $O5$

2002: or $O9$ planes. However, type IIB admits a second perturbative symmetry operation 

2003: denoted by $(-1)^{F_L}$, where $F_L$ is the space-time fermion number in the left-moving 

2004: sector.

2005: Under the action of $(-1)^{F_L}$ R-NS and R-R states are odd 

2006: while NS-R and NS-NS states are even. Orientifolds with $O3$ and/or $O7$ planes 

2007: arise from projections of the form $(-1)^{F_L} \Omega_p \sigma$ as we will argue 

2008: below. In summary let us display the transformation behavior of the massless bosonic 

2009: states under these two operations \cite{JPbook,AD}

2010: \beq \label{transf-AB} 

2011: \begin{array}{cllll}  

2012:   \Omega_p:&  \qquad \text{even:}\quad  &\hat \phi, \  \hat g,\ \hat C_1,\   \hat C_2 ,  

2013:               \qquad& \text{odd:}\quad &\hat C_0, \ \hat B_2 ,\ \hat C_3, \ \hat C_4 \ ,\\

2014:   (-1)^{F_L}:& \qquad \text{even:}\quad  &\hat \phi, \  \hat g,\  \hat B_2 ,  

2015:               \qquad& \text{odd:}\quad &\hat C_0,\ \hat C_1,\ \hat C_2 ,\ \hat C_3, \ \hat C_4 \ , 

2016: \end{array}

2017: \eeq

2018: where we have also displayed the transformation properties of the type IIA forms. 

2019: With these transformations at hand one  easily checks that $\Omega_p$ as well as

2020: $(-1)^{F_L}$ are symmetries of the ten-dimensional type IIB supergravity action.

2021: This is in contrast to type IIA. By using \eqref{transf-AB} one immediately notices that

2022: $\Omega_p$, \,$(-1)^{F_L}$ and $(-1)^{F_L}\Omega_p$ alone are no symmetries 

2023: of the type IIA effective action \eqref{10dact}. However, orientifolds with 

2024: $O6$ planes arise if $S$ includes $(-1)^{F_L}\Omega_p$ as well as some appropriatly 

2025: chosen target space symmetry which ensures that $S\Omega_p$ leaves \eqref{10dact} invariant.  

2026: Let us now make this more explicite by properly defining the Calabi-Yau orientifold 

2027: projections.

2028: 

2029: \section{Orientifold projections} \label{oprojections}

2030: 

2031: After this brief introduction we are now in the position to specify the orientifolds 

2032: under consideration and give an explicit definition 

2033: of the orientifold symmetry group \eqref{osym}. 

2034: We start from type II string theory and compactify 

2035: on a Calabi-Yau threefold $Y$. In addition we 

2036: mod out by orientation reversal of the string 

2037: world-sheet $\Omega_p$ together with an  `internal'

2038: symmetry $\sigma$ which acts solely on $Y$ 

2039: but leaves the $D=4$ Minkowskian space-time untouched.

2040: We will restrict ourselves to involutive symmetries ($\sigma^2 = 1$) of $Y$ 

2041: and thus set $G$ in \eqref{osym} to be empty.\footnote{Calabi-Yau manifolds have only 

2042: discrete isometries. For example in the case of the quintic, 

2043: $\sigma$ could act 

2044: by permuting the coordinates such that

2045: the defining equation is left invariant. 

2046: A classification of all possible involutions

2047: of the quintic can be found in ref.\ \cite{BH}.} 

2048: This avoids the appearance of further twisted 

2049: sectors as they appear in general orbifold models \cite{DHVW}. In 

2050: a next step we have to specify additional properties of $\sigma$ and the complete 

2051: operation $S\Omega_p$ in order that it provides a symmetry of the 

2052: string theory under consideration. To do that we discuss the type IIA and type IIB

2053: case in turn.

2054: 

2055: \subsubsection{Type IIB orientifolds}

2056: 

2057: Let us start with type IIB Calabi-Yau orientifolds and define the orientifold projections following 

2058: \cite{Sen,DP,AAHV,BH}. 

2059: %We are then in the position to check if these projections 

2060: %are a symmetry of the ten-dimensional supergravity action 

2061: %for the corresponding string theory. 

2062: Later on, in section \ref{IIB_orientifolds} we show that gauge-fixing these symmetries indeed 

2063: leads to an $N=1$ supergravity theory.  

2064: In type IIB consistency requires 

2065: $\sigma$ to be an isometric and holomorphic involution of $Y$ \cite{AAHV,BH}.

2066: A holomorphic isometry leaves both the metric

2067: and the complex structure of the Calabi-Yau manifold invariant. 

2068: As a consequence also the K\"ahler form $J$

2069: is invariant such that 

2070: \beq

2071:   \sigma^* J = J \ , 

2072: \eeq

2073: where $\sigma^*$ denotes the pull-back of the map $\sigma$.

2074: Hence in our analysis we focus on the class of Calabi-Yau threefolds which 

2075: admit such an involution but within this class we leave the

2076: threefolds arbitrary. Since the involution is holomorphic 

2077: it respects the Hodge decomposition \eqref{odd_even_cohom} and we find 

2078: in particular $\sigma^* H^{(3,0)} = H^{(3,0)}$. Picking the

2079: holomorphic three-form $\Omega$ as an representative of $H^{(3,0)}$

2080: and using that $(\sigma^*)^2 =\text{id}$ one is left with two 

2081: possible actions 

2082: \beq \label{Omegatransf}

2083:   (1)\quad O3/O7: \quad \sigma^* \Omega = - \Omega\ ,\qquad \qquad (2)\quad O5/O9: \quad  \sigma^* \Omega = + \Omega\ . 

2084: \eeq

2085: Correspondingly, depending on the transformation properties of  $\Omega$

2086: two different symmetry operations $\mathcal{O}=S\Omega_p$

2087: are possible \cite{Sen,DP,AAHV,BH} \footnote{%

2088: The factor $(-1)^{F_L}$ is included in $\mathcal{O}_{(1)}$

2089: to ensure that $\OO_{(1)}^2=1$ on all states.}

2090: \beq \label{o3-projection}

2091: \mathcal{O}_{(1)} = (-1)^{F_L} \Omega_p \, \sigma\ ,\qquad

2092: \mathcal{O}_{(2)} = \Omega_p \, \sigma

2093: \eeq 

2094: where $\Omega_p$ is the world-sheet parity, $F_L$ is the space-time fermion number

2095: in the left-moving sector introduced at the end of section \ref{D-branes}. 

2096: This specifies the operation $S \Omega_p$ in \eqref{osym} and,

2097: since $G$ is empty, the complete orientifold projection. We are now in the position to 

2098: check if the orientifold projections are indeed a symmetry of the bosonic ten-dimensional 

2099: type IIB supergravity action \eqref{10d-lagr}. We will do this check by concentrating only 

2100: on some of the terms in \eqref{10d-lagr} keeping in mind that the analysis for the 

2101: remaining terms is analoge. The background 

2102: $\cM'=\Mext \times \sigma(Y)$ denotes the image of $\cM=\Mext\times Y$ under the geometric action 

2103: $\sigma$. Also inserting the $\sigma$-transformed fields into \eqref{10d-lagr} one infers  

2104: \footnote{Here we have used \eqref{wedge*comp}

2105: in order to give the component expression of the kinetic terms in \eqref{10d-lagr}.}

2106: \beq \label{tranfact}

2107:  S^{(10)}_{IIB'}\ =\

2108:   \int_{\cM'}\big( -\tfrac{1}{2} \hat R_{g'} *{'} \mathbf{1} 

2109:                 - \tfrac{1}{4}g'^{MN} (\partial_M \hat \phi')(\partial_N \hat \phi') *{'} \mathbf{1} - \ldots

2110:      -\tfrac{1}{4} \hat C_4' \wedge \hat H_3' \wedge \hat F_3'\big)\ ,

2111: \eeq

2112: where $g'=\sigma^* g,\ \hat \phi'=\sigma^*\hat \phi$ etc.\ and the dots denote 

2113: terms transforming similar to the kinetic term of $\phi'$. The Hodge star $*'$ is evaluated on

2114: the manifold $\cM'$ with metric $g'$. Now we apply the properties of the involution. Since 

2115: $\sigma$ is an isometry we find $g=g'$ and due to the holomorphicity of $\sigma$ we can deduce 

2116: that the ten-dimensional volume element $*'\mathbf{1}$ does not change sign in going from $\cM'$ to $\cM$.\footnote{%

2117: Holomorphic maps do not change the orientation of $M$.} This ensures that the 

2118: Einstein-Hilbert term takes the from $\int_{\cM'} \sigma^* (- \frac{1}{2} R * \mathbf{1})$ and by applying 

2119: \eqref{int-form1} and \eqref{int-form2} is invariant under the isometric map $\sigma$.

2120: A similar reasoning applies to all other terms in \eqref{tranfact} and one concludes

2121: that the effective action \eqref{10d-lagr} is indeed unchanged by $\sigma$. 

2122: Combined with the invariance of \eqref{10d-lagr} under the world-sheet parity $\Omega_p$ and $(-1)^{F_L}$

2123: one infers that the orientifold operations \eqref{o3-projection} are symmetries of the effective theory.

2124: 

2125: 

2126: 

2127: 

2128: The fix-point set of the involutions $\sigma$ in \eqref{o3-projection} determines 

2129: the location of the orientifold planes.

2130: Modding out by $\mathcal{O}_{(1)}$

2131: leads to the possibility of having $O3$- and $O7$-planes

2132: while modding out by $\mathcal{O}_{(2)}$ allows 

2133: $O5$- and $O9$-planes. To see this, recall that 

2134: the four-dimensional Minkowski space is left invariant by

2135: $\sigma$ such that the orientifold planes are necessarily space-time filling.

2136: Using the fact that $\sigma$ is holomorphic they 

2137: have to be even-dimensional (including the time direction) which 

2138: selects $O3$-, $O5$-, $O7$- or $O9$-planes as the only possibilities.

2139: The actual

2140: dimensionality of the orientifold plane is then determined 

2141: by the dimensionality of the fix-point set of $\sigma$ in $Y$.

2142: In order to determine this dimensionality we need the induced

2143: action of $\sigma$ on the tangent space at any point 

2144: of the orientifold plane. 

2145: Since one can always choose $\Omega \propto dy^1 \wedge dy^2 \wedge dy^3$

2146: we see that for $\sigma^* \Omega  =  - \Omega$ 

2147: the internal part of the orientifold plane  is 

2148: either a point or a  surface of complex dimension two.

2149: Together with the space-time filling part we thus can have 

2150: $O3$- and/or $O7$-planes.

2151: The same argument can be repeated  for $\sigma^* \Omega  =  \Omega$ 

2152: which then leads to the possibility of 

2153: $O5$- or $O9$-planes. There are no models with $O5$ and $O9$ planes, since

2154: the appearance of a $O9$ plane implies that the complete background $\cM_{10}$

2155: consist of fix-points of $\sigma=\text{id}$.

2156: The case of $O9$ planes is special and coincides with type I if one 

2157: introduces the appropriate number of $D9$-branes to cancel tadpoles.

2158: 

2159: Since the involution $\sigma$ is holomorphic the fix-point set of the involution

2160: are holomorphic cycles $\Lambda_{Op}$. This implies that they are calibrated 

2161: with respect to the forms $1$ and $J \wedge J$ in orientifolds with $O3/O7$ planes 

2162: and with respect to $J$ or $J\wedge J\wedge J$ in orientifolds with $O5$ or $O9$ planes. 

2163: More precicely, one finds that the volume forms on $\Lambda_{Op}$ equals the pull-back 

2164: of $e^{iJ}$ to the cycle \footnote{%

2165: Here we abbreviate the formal sum of $(q,q)$-forms

2166: $ e^{iJ} = 1 + iJ + \frac{1}{2!}J \wedge J - \frac{i}{3!} J \wedge J\wedge J$.} 

2167: \beq \label{cal_sOp}

2168:  \vol(\Lambda_{Op}) = e^{-i\theta_{Op}}\, e^{iJ} \big|_{\Lambda_{Op}}\ , \qquad 

2169:  \theta_{O3/7} = 0\ ,\quad \theta_{O5}=\tfrac{\pi}{2}\ ,\quad \theta_{O9}=-\tfrac{\pi}{2}\ ,

2170: \eeq

2171: where the phase depends on the type of orientifold planes in the setup. Furthermore

2172: one has $\Omega|_{\Lambda_{Op}}=0$. Cycles fulfilling these conditions minimize their volume within

2173: their homology class. Note that similar to \eqref{calcond} this condition has to be modified 

2174: in the presence of a $\hat B_2$ field. In this case the form which 

2175: calibrates the supersymmetric cycles is $e^{-\hat B_2+iJ}$. 

2176: Let us check whether the fix-point sets $\Lambda_{Op}$ of $\sigma$ 

2177: remain calibrated. 

2178: In the two orientifold setups only fields are kept in

2179: the spectrum which are invariant under the respective 

2180: projection $\mathcal{O}_{(1/2)}$ given in \eqref{o3-projection}. Thus, by using \eqref{transf-AB} 

2181: one infers that $\hat B_2$ has to transform

2182: as $\sigma^* \hat B_2 = - \hat B_2$ for both orientifold projections. 

2183: This implies that $\hat B_2$ restricted to the fix-point set of 

2184: $\sigma$ vanishes. \footnote{% 

2185: Denoting $\rho^* \hat B_2 = \hat B_2|_{\Lambda_{Op}}$ the pull-back to the 

2186: fix-point set $\Lambda_{Op}$ of $\sigma$ it follows 

2187: $-\rho^*\hat B_2 = \rho^* (\sigma^*  \hat B_2) = (\sigma \circ \rho)^* \hat B_2 = \rho^* \hat B_2$

2188: such that $\rho^*\hat B_2=0$.}

2189: One concludes that the cycles $\Lambda_{Op}$ remain 

2190: calibrated with respect to the generalized calibration form, i.e.\

2191: \beq\label{cal-Op}

2192:   \vol_{DBI}(\Lambda_{Op}) = e^{-i\theta_{Op}} e^{-\hat B_2+iJ} \big|_{\Lambda_{Op}}\ ,

2193: \eeq

2194: where $\theta_{Op}$ is as given in \eqref{cal_sOp} and $\vol_{DBI}(\Lambda_{Op})$ is defined as 

2195: in \eqref{calcond}.

2196: At this point, one can compare the calibration condition \eqref{cal-Op} for the 

2197: orientifold planes with the one for the $Dp$-branes given in \eqref{calcond}. 

2198: In order to preserve some supersymmetry all orientifold planes and 

2199: D-branes, have to be calibrated with respect to the same form. This implies that 

2200: the phases $\theta_{Dp}$ in \eqref{calcond} have to coincide with $\theta_{Op}$ given 

2201: in \eqref{cal_sOp} (see also \cite{JL} for the case of $D3/D7$ branes). This is 

2202: equivalently true for $Dq$-instantons wrapping $q+1$-cycles in $Y$. In supersymmetric setups 

2203: with $O(q+3)$ planes one has to set $\theta_{Dq}=\theta_{O(q+3)}$, where $e^{i\theta_{Dq}}$

2204: is the phase in the D-instanton calibration condition.

2205: 

2206: 

2207: 

2208: 

2209: 

2210: \subsubsection{Type IIA orientifolds}

2211: 

2212: Let us now turn to the type IIA Calabi-Yau orientifolds. 

2213: In contrast to type IIB the orientifold projection has 

2214: to include an anti-holomorphic and isometric involution $\sigma$ 

2215: in order to preserve $N=1$ supersymmetry \cite{AAHV,BBKL,BH}. 

2216: Hence, the K\"ahler form on $Y$ transforms as 

2217: \beq \label{constrJ}

2218:   \sigma^* J\ =\ -J\ , 

2219: \eeq

2220: since $\sigma$ preserves the metric but yields a minus sign when applied to 

2221: the complex structure.

2222: The complete projection takes the form 

2223: \beq \label{oproj}

2224:   \mathcal{O} = (-1)^{F_L} \Omega_p \sigma\ .

2225: \eeq

2226: In addition to the condition \eqref{constrJ} 

2227: compatibility of $\sigma$ with the Calabi-Yau condition 

2228: $\Omega \wedge \bar \Omega \propto J \wedge J \wedge J$ 

2229: implies that $\sigma$ also acts non-trivially  on the three-form $\Omega$ as

2230: \beq \label{constrO}

2231:   \sigma^* \Omega\ =\ e^{2i\theta} \bar \Omega \ , 

2232: \eeq

2233: where $e^{2i\theta}$ is a constant phase and we included a factor 2 for later convenience.

2234: Similar to the type IIB case we can check that the 

2235: projection $\mathcal{O}$ is a symmetry of the type IIA supergravity 

2236: action \eqref{10dact}. Note however, that $(-1)^{F_L} \Omega_p$ alone is not a symmetry 

2237: of type IIA. Using \eqref{transf-AB} this can be already inferred from the fact that the kinetic and

2238: topological terms in \eqref{10dact} transform with a different sign. 

2239: On the other hand, under the action of the involution $\sigma$ the effective action changes as

2240: \beq

2241:  S^{(10)}_{IIA'}\ =\

2242:   \int_{\cM'}\big( -\tfrac{1}{2} \hat R_{g'} *{'} \mathbf{1} 

2243:                  - \tfrac{1}{4}g'^{MN} (\partial_M \hat \phi')(\partial_N \hat \phi') *{'} \mathbf{1} \ldots

2244:      -\tfrac{1}{2} \hat B_2' \wedge \hat F_4' \wedge \hat F_4' \big)\ ,

2245: \eeq

2246: where as in \eqref{tranfact} we have set $g'=\sigma^* g,\ \hat \phi'=\sigma^*\hat \phi$ etc.\ and the Hodge star $*'$ is on

2247: the manifold $\cM'=\Mext \times \sigma(Y)$ with metric $g'$. Using the fact that $\sigma$ is 

2248: an anti-holomorpic isometric involution it changes the sign of the volume element 

2249: $*\mathbf{1} \sim \vol(\Mext) \wedge J' \wedge J' \wedge J'$, such that $*'\mathbf{1}=-*\mathbf{1}$.

2250: From equations \eqref{int-form1} and \eqref{int-form2} one finds that the topological term transforms with a minus sign

2251: while the kinetic terms remain invariant. This extra sign cancels the minus from the action of

2252: $(-1)^{F_L} \Omega_p$ such that $\mathcal{O}$ is indeed a symmetry of \eqref{10dact}.

2253: In section \ref{IIA_orientifolds} we show that gauge-fixing this symmetry results in an $N=1$ supergravity

2254: theory.

2255: 

2256: Type IIA orientifolds with anti-holomorphic involution generically contain $O6$ planes. This 

2257: is due to the fact, that the fixed point set of $\sigma$ in $Y$ are  three-cycles 

2258: $\Lambda_{O6}$ supporting the internal part of the orientifold planes. 

2259: These cycles are special Lagrangian submanifolds of $Y$ as an 

2260: immediate consequences of \eqref{constrJ} and \eqref{constrO}

2261: which implies \cite{HitchinLec}

2262: \beq \label{OLagr}

2263:   J|_{\Lambda_{O6}} = 0\ , \qquad  \I(e^{-i\theta}\Omega)|_{\Lambda_{O6}} = 0\ .

2264: \eeq

2265: In other words, they are calibrated with respect to 

2266: %$\R(e^{-U-i\theta}\Omega)$ 

2267: $\R(e^{-i\theta}\Omega)$

2268: \beq \label{calibr-O6}

2269: \rm{vol}(\Lambda_{O6})\sim \R(e^{-i\theta}\Omega)\ ,

2270: \eeq

2271: where the overall normalization of $\Omega$ will be determined

2272: in \eqref{Omeganorm}. Once again this poses conditions on 

2273: additional D-branes in the setup, if they are demanded to preserve the same

2274: supersymmetry. More precicely, BPS branes have to be calibrated 

2275: with respect to the same form as the orientifold planes. This implies 

2276: by comparing \eqref{calcond} with \eqref{calibr-O6} that 

2277: $\theta_{D6}=\theta$ for space-time filling $D6$-branes wrapping a 

2278: three-cycle in $Y$. A similar condition $\theta_{D2}=\theta$ has 

2279: to hold for supersymmetric $D2$-instantons wrapping a three-cycle in $Y$.

2280: 

2281: 

2282: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2283: %

2284: %  IIB orientifolds 

2285: %

2286: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2287: 

2288: 

2289: \section{Type IIB Calabi-Yau orientifolds \label{IIB_orientifolds}}

2290: 

2291: 

2292: In this section we impose the projection \eqref{o3-projection}

2293: on the type IIB theory and derive the massless spectrum 

2294: (section~\ref{o3-spectrum}) and its 

2295: low energy $N=1, D=4$ effective supergravity action 

2296: (section~\ref{O37_effective_act}).

2297: This generalizes similar derivations already performed in refs.\

2298: \cite{GKP,BBHL}.

2299: We restrict our analysis  to the bosonic fields of the compactification

2300: keeping in mind that the couplings of the 

2301: fermionic partners are fixed by  supersymmetry. Furthermore, we 

2302: include space-time filling D-branes for consistency but fix their 

2303: moduli, such that they do not appear in the low energy effective 

2304: action. The compactification we perform is closely related to the compactification

2305: of type IIB string theory on Calabi-Yau threefolds reviewed

2306: in chapter \ref{TypeII}. The orientifold projection 

2307: \eqref{o3-projection}

2308: truncates the massless spectrum from $N=2$ to $N=1$

2309: multiplets and also leads to a modification of the couplings

2310: which render the low energy effective theory

2311: compatible with $N=1$ supergravity.

2312: Such truncation procedures from $N=2$ to $N=1$ supergravity has been carried

2313: out from a purely supergravity point of view  

2314: in refs.\ \cite{ADAF}.

2315: 

2316: \subsection{The $N=1$ spectrum \label{o3-spectrum}}

2317: %

2318: Before computing the effective action let us first

2319: determine the massless spectrum when 

2320: the orientifold projection is taken into account and see how 

2321: the fields assemble in $N=1$ supermultiplets \cite{BH}.

2322: In the 

2323: four-dimensional compactified theory

2324: only states invariant under the projection are kept.

2325: Using equation \eqref{transf-AB} one immediately infers that 

2326: the scalars $\hat \phi,\hat l$, the metric $\hat g$ and the

2327: four-form $\hat C_4$ are even under $(-1)^{F_L} \Omega_p$

2328: while both two forms $\hat B_2, \hat C_2$ are odd.

2329: Using \eqref{o3-projection} this implies that the invariant

2330: states have to obey

2331: \begin{equation} \label{fieldtransfB}

2332: \begin{array}{lcl}

2333: \\

2334: \sigma^*  \hat \phi &=& \  \hat \phi\ , \\

2335: \sigma^*   \hat g &=& \ \hat g\ , \\

2336: \sigma^*   \hat B_2 &=& -  \hat B_2\ ,

2337: \end{array}

2338: \hspace{1cm}

2339: \begin{array}{lcl}

2340: \multicolumn{3}{c}{ \underline{O3/O7}} \\[2ex]

2341: \sigma^*  \hat C_0 &=& \ \ \hat C_0\ , \\

2342: \sigma^*   \hat C_2 &=& - \hat C_2\ , \\

2343: \sigma^*   \hat C_4 &=& \ \ \hat C_4\ , 

2344: \end{array}

2345: \hspace{1cm}

2346: \begin{array}{lcl}

2347: \multicolumn{3}{c}{ \underline{O5/O9}} \\[2ex]

2348: \sigma^*   \hat C_0 &=& - \hat C_0\ , \\

2349: \sigma^*   \hat C_2 &=& \ \ \hat C_2\ , \\

2350: \sigma^*   \hat C_4 &=& - \hat C_4\ , 

2351: \end{array}

2352: \end{equation}

2353: where the first column is identical for both involutions $\sigma$ 

2354: in \eqref{o3-projection}. 

2355: Since $\sigma$ is a holomorphic involution the cohomology groups $H^{(p,q)}$

2356: (and thus the

2357: harmonic $(p,q)$-forms) split into two eigenspaces 

2358: under the action of $\sigma^*$ 

2359: \beq\label{H3split}

2360: H^{(p,q)} = 

2361: H^{(p,q)}_+\oplus H^{(p,q)}_-\ .

2362: \eeq

2363: $H^{(p,q)}_+$ has dimension $h_+^{(p,q)}$ and denotes

2364: the even eigenspace of $\sigma^*$ while

2365: $H^{(p,q)}_-$ has  dimension $h_-^{(p,q)}$ and denotes

2366: the odd eigenspace of $\sigma^*$. 

2367: The Hodge $*$-operator commutes with $\sigma^*$ since $\sigma$ preserves the

2368: orientation and the metric of the Calabi-Yau manifold and thus the Hodge

2369: numbers obey $h^{(1,1)}_\pm=h^{(2,2)}_\pm$. Holomorphicity of $\sigma$ 

2370: further implies $h^{(2,1)}_\pm = h^{(1,2)}_\pm$ while

2371: \eqref{Omegatransf} leads to 

2372: $h^{(3,0)}_+ = h^{(0,3)}_+=0, h^{(3,0)}_- = h^{(0,3)}_-=1$ for $O3/O7$ orientifolds 

2373: and $h^{(3,0)}_+ = h^{(0,3)}_+=1, h^{(3,0)}_- = h^{(0,3)}_-=0$ for $O5/O9$ orientifolds.

2374: Furthermore, the volume-form which is proportional

2375: to $\Omega\wedge\bar\Omega$ is invariant under $\sigma^*$ and thus one has 

2376: $h^{(0,0)}_+ = h^{(3,3)}_+=1, h^{(0,0)}_- = h^{(3,3)}_-=0$.

2377: We summarize the non-trivial cohomology groups including

2378: their basis elements in table~\ref{CYObasisB}.

2379: \begin{table}[h]

2380: \begin{center}

2381: \begin{tabular}{|c || c | c || c| c || c | c |} \cline{1-7}

2382:    setup &\multicolumn{2}{|c||}{\rule[-0.3cm]{0cm}{0.8cm} cohomology group} &

2383:    \multicolumn{2}{|c||}{dimension} & \multicolumn{2}{|c|}{basis}

2384:    \\ \cline{1-7}

2385:    \multirow{3}{1.2cm}[-.3cm]{$O3/O7$ $\text{\ \ and}$ $O5/O9$}  &\rule[-0.3cm]{0cm}{0.8cm} $H^{(1,1)}_+$ & $H^{(1,1)}_-$  &

2386:    $h^{(1,1)}_+$ & $h^{(1,1)}_- $ & $\omega_\alpha$ & $\omega_a$

2387:    \\ \cline{2-7}

2388:     &\rule[-0.3cm]{0cm}{0.8cm} $H^{(2,2)}_+$ & $H^{(2,2)}_-$  & $h^{(1,1)}_+$ & $h^{(1,1)}_-$ & 

2389:    $\tilde \omega^\alpha$ & $\tilde \omega^a$

2390:    \\ \cline{2-7}

2391:     &\rule[-0.3cm]{0cm}{0.8cm} $H^{(2,1)}_+$  & $H^{(2,1)}_-$ 

2392:    & $h^{(2,1)}_+$ & $h^{(2,1)}_-$ &

2393:    $\chi_\kappa$ & $\chi_k$

2394:    \\ \hline

2395:    O3/O7&\rule[-0.3cm]{0cm}{0.8cm} $H^{(3)}_+$ & $H^{(3)}_-$  & $2h^{(2,1)}_+$ & $2h^{(2,1)}_-+2$ &

2396:    $(\alpha_{\kappa},\beta^{\lambda})$ & $(\alpha_{\hat k},\beta^{\hat l})$ 

2397:    \\ \hline 

2398:    O5/O9&\rule[-0.3cm]{0cm}{0.8cm} $H^{(3)}_+$ & $H^{(3)}_-$  & $2h^{(2,1)}_+ +2$ & $2h^{(2,1)}_-$ &

2399:    $(\alpha_{\kappa},\beta^{\lambda})$ & $(\alpha_{\hat k},\beta^{\hat l})$ \\ \hline

2400: \end{tabular}

2401: \caption{\small \label{CYObasisB}

2402: \textit{Cohomology groups and their basis elements.}}

2403: \end{center}

2404: \end{table}

2405: 

2406: 

2407: 

2408: 

2409: 

2410: The four-dimensional invariant spectrum

2411: is found by using the Kaluza-Klein expansion 

2412: given in eqs.\ \eqref{def-v}, \eqref{cs} and \eqref{CYexpansion}

2413: keeping only the fields which in addition obey \eqref{fieldtransfB}.

2414: We see immediately that the $D=4$ scalar field arising from

2415: $\hat\phi$ remains in the spectrum for both setups and as before we denote it by 

2416: $\phi$. Since $\sigma^*$  leaves 

2417: the K\"ahler form $J$ invariant  only the

2418: $h_+^{(1,1)}$ even K\"ahler deformations $v^\alpha$ remain in the spectrum

2419: and we expand 

2420: \beq\label{transJo}

2421: J =  v^{\alpha}\, \omega_{\alpha} \ ,\qquad

2422: \alpha = 1,\ldots, h_+^{(1,1)}\ ,

2423: \eeq 

2424: where $\omega_\alpha$ denotes a basis of $H^{(1,1)}_+$.

2425: {} Similarly, from eq.\ \eqref{cs}  we infer that the

2426: invariance of the metric together with

2427: \eqref{Omegatransf} implies that the complex structure deformations 

2428: kept in the spectrum correspond to elements in $H^{(1,2)}_-$ for $O3/O7$

2429: setups and to elements of $H^{(1,2)}_+$ for $O5/O9$. Hence, \eqref{cs} is 

2430: replaced by

2431: \bea\label{cso}

2432: O3/O7:\quad \delta{g}_{ij} =  \frac{i}{||\Omega||^2}\, \bar z^{k}

2433: (\bar \chi_{ k})_{i\ib\bj}\,

2434: \Omega^{\ib\bj}{}_j \ ,\quad  k=1,\ldots,h_-^{(1,2)}\ , \\ \qquad 

2435: O5/O9:\quad \delta{g}_{ij} =  \frac{i}{||\Omega||^2}\, \bar z^{\kappa}  

2436: (\bar \chi_{\kappa})_{i\ib\bj}\,

2437: \Omega^{\ib\bj}{}_j \ , \quad  \kappa =1,\ldots,h_+^{(1,2)}\ ,\nn

2438: \eea

2439: where $\bar\chi_{k}\ (\bar \chi_{\kappa})$ denotes a basis of $H^{(1,2)}_-\ (H^{(1,2)}_+)$.\footnote{%

2440: In ref.\ \cite{BH} it is further shown that the

2441: $h_\pm^{(1,2)}$ deformations form a smooth submanifold

2442: of the Calabi-Yau  moduli space.}

2443: 

2444: {}From eqs.\ (\ref{fieldtransfB}) we learn that in the expansion of

2445: $\hat B_2$ only odd elements are kept. Thus, for both orientifold setups we

2446: have

2447: \beq \label{exp-B}

2448:   \hat B_2 = b^a\, \omega_a\ ,\qquad  a=1,\ldots, h_-^{(1,1)}\ ,

2449: \eeq 

2450: where $\omega_a$ is a basis of $H^{(1,1)}_-$. 

2451: The orientifold projections differ in the R-R sector. For $O3/O7$ orientifolds

2452: $\hat C_2$ is odd and $\hat C_4$ is even. Therefore the expansion \eqref{CYexpansion} 

2453: is replaced by 

2454: \beq\label{exp1}

2455:   \hat C_2\ =\ c^a\, \omega_a\ , \qquad 

2456:   \hat C_4\ =\  D_2^\alpha\wedge \omega_\alpha

2457: + V^{\kappa}\, \wedge \alpha_{\kappa} 

2458: + U_{\kappa}\wedge\beta^{\kappa}+

2459:  \rho_\alpha\ \tilde \omega^\alpha\ ,

2460: \eeq

2461: where $\tilde\omega^\alpha$ is a basis

2462: of $H^{(2,2)}_+$ which is dual to $\omega_\alpha$, and

2463: $(\alpha_{\kappa}, \beta^{\kappa})$ is a real, symplectic

2464: basis of $H^{(3)}_+ = H^{(1,2)}_+ \oplus H^{(2,1)}_+$ 

2465: (c.f.\ table~\ref{CYObasisB}). From \eqref{fieldtransfB} we find that 

2466: the axion $\hat C_0$ remains in the spectrum and we denote

2467: the corresponding four-dimensional field by $C_0$.

2468: Note that the two $D=4$ two-forms $B_2$ and $C_2$ present in the $N=2$

2469: compactification (see \eqref{CYexpansion})

2470: have been projected out and in the expansion of  $\hat B_2$ and $\hat C_2$

2471: only the scalar fields $c^a, b^a$ appear.

2472: The non-vanishing of $c^a,b^a$ and $V^\kappa$ is closely related to the 

2473: appearance of $O7$-planes. To understand this in more detail

2474: we recall, that $O3$-planes appear 

2475: when the fix-point set of $\sigma$ is zero-dimensional in $Y$

2476: or in other words all tangent vectors at this point are odd under

2477: the action of $\sigma$.

2478: This in turn implies that locally

2479: two-forms  are even  under $\sigma^*$, while three-forms 

2480: are odd. However, this is incompatible

2481: with the expansions given in \eqref{exp1} for 

2482: non-vanishing $b^a,c^a$ and $V^\kappa$. 

2483: For a setup also including $O7$-planes we locally

2484: get the correct transformation behavior, 

2485: so that harmonic forms in $H^{(1,1)}_-$ and 

2486: $H^{(2,1)}_+$ can be supported. 

2487: 

2488: For $O5/O9$ orientifolds the $\mathcal{O}_{(2)}$-invariant R-R forms transform exactly 

2489: with the opposite sign under $\sigma$. Thus, 

2490: the expansion \eqref{CYexpansion} reduces to 

2491: \beq\label{expO5} 

2492: \hat C_2\ =\ C_2+ c^\alpha\ \omega_\alpha\ , 

2493: \qquad \hat{C}_4 \ =\ D_{2}^a \wedge \omega_a + V^{k} \wedge 

2494: \alpha_{k} - U_{k} \wedge \beta^{k} + 

2495:   \rho_a\, \tilde \omega^a\ .

2496: \eeq

2497: In this case the axion $\hat C_0$ is projected out and replaced by 

2498: the $D=4$ antisymmetric tensor $C_2(x)$. As a consequence the $N=1$ 

2499: spectrum contains a `universal' linear multiplet $(\phi,C_2)$ which in the massless 

2500: case can be dualized to a chiral multiplet.

2501: As for Calabi-Yau compactifications

2502: imposing the self-duality on $\hat F_5$

2503: eliminates half of the degrees of freedom in the expansions \eqref{exp1} and \eqref{expO5}

2504: of $\hat C_4$. For the one-forms  $V^{\cdot},U_{\cdot}$ this corresponds to

2505: the choice of electric versus magnetic gauge potentials.

2506: On the other hand choosing the two forms $D_2^{\cdot}$ 

2507: or the scalars $\rho_{\cdot}$ determines

2508: the structure of the $N=1$ multiplets to be either a linear or a chiral 

2509: multiplet and in chapter \ref{lin_geom_of_M} we discuss both cases.

2510: 

2511: Altogether the resulting $N=1$ fields for the two setups 

2512: assembles into a gravitational

2513: multiplet, $h_\pm^{(2,1)}$ vector multiplets and 

2514: $(h_\mp^{(2,1)}+ h^{(1,1)}+1)$ chiral multiplets 

2515: and are

2516: summarized in  table~\ref{N=1spectrumtab} \cite{BH,TGL1}.

2517: 

2518: 

2519: \begin{table}[h] 

2520: \begin{center}

2521: \begin{tabular}{|c|c|c||c|c|} \hline

2522:  \rule[-0.3cm]{0cm}{0.8cm} & \multicolumn{2}{|c||}{$O3/O7$} & \multicolumn{2}{c|}{$O5/O9$}\\ \hline 

2523:  \rule[-0.3cm]{0cm}{0.85cm} 

2524:  gravity multiplet&1&$g_{\mu \nu} $ &1& $g_{\mu \nu}$ \\ \hline

2525:  \rule[-0.3cm]{0cm}{0.85cm} 

2526:  vector multiplets&   $\ h_+^{(2,1)}\ $&  $V^{\lambda} $& $\ h_-^{(2,1)}\ $ & $V^{k} $ \\ \hline

2527:  \rule[-0.3cm]{0cm}{0.85cm} 

2528:  \multirow{3}{30mm}[-3.5mm]{chiral multiplets} &   $h_-^{(2,1)}$& $z^{k} $ &   $h_+^{(2,1)}$& $z^{\lambda} $\\ \cline{2-5}

2529: \rule[-0.3cm]{0cm}{0.85cm} 

2530:  &  $ h^{(1,1)}_-$ &$( b^a, c^a)$ & $h^{(1,1)}_+$& $( v^\alpha, c^\alpha)$\\ \cline{2-5}

2531:  \rule[-0.3cm]{0cm}{0.85cm} 

2532:    & 1 & $(\phi,l)$ && \\ \hline

2533: \rule[-0.3cm]{0cm}{0.85cm} 

2534: \multirow{3}{44mm}[2mm]{chiral/linear multiplets } & $h^{(1,1)}_+$& $( v^\alpha, \rho_\alpha )$ & $h^{(1,1)}_-$& 

2535:  $( b^a, \rho_a )$\\ \cline{2-5}

2536: \rule[-0.3cm]{0cm}{0.85cm}  & && 1 & $(\phi,C_2)$ \\

2537: \hline

2538: \end{tabular}

2539: \caption{\label{N=1spectrumtab} $N =1$ spectrum of Type IIB orientifold compactifications.}

2540: \end{center}

2541: \end{table} 

2542: 

2543: Compared to the $N=2$ spectrum of the Calabi-Yau compactification

2544: given in table~\ref{tab-compIIBspec} we see that 

2545: the graviphoton `left' the gravitational multiplet

2546: while the $h^{(2,1)}$ $N=2$ vector multiplets decomposed

2547: into $h_\pm^{(2,1)}$ $N=1$ vector multiplets plus $h_\mp^{(2,1)}$ 

2548: chiral multiplets. Furthermore, the $h^{(1,1)}+1$ hypermultiplets

2549: lost half of their physical degrees of freedom and are reduced 

2550: into $h^{(1,1)}+1$ chiral multiplets. This is 

2551: consistent with the theorem of \cite{AM,ADAF} where it was shown that 

2552: any K\"ahler submanifold of a quaternionic manifold

2553: can have at most half of its (real) dimension.

2554: 

2555: 

2556: 

2557: 

2558: 

2559: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2560: %

2561: %  IIB: O3/O7 effective action

2562: %

2563: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2564: 

2565: \subsection{The effective action \label{O37_effective_act}}

2566: 

2567: In following we derive the effective actions encoding the 

2568: dynamics of the $N=1$ multiplets of the type IIB orientifold 

2569: theories. However, before entering the actual computations 

2570: a cautionary note is in order. 

2571: In the presence of localized sources such as orientifold planes and 

2572: D-branes as well as in the presence of non-trivial background fluxes 

2573: the product Ansatz \eqref{lineel} for the metric is strictly speaking not anymore

2574: suitable. This is due to the fact that the supergravity theory with source 

2575: terms and fluxes does not have the background metric \eqref{lineel} as a solution 

2576: \cite{BB1,Verlinde,GSS,GKP}. As deviation from the standard Calabi-Yau compactifications

2577: a non-trivial warp factor $e^{-2A}$ has to be included into the

2578: Ansatz for the metric \eqref{lineel} such that \cite{GKP,GP}

2579: \beq 

2580: ds^2=e^{2A(y)} {g_{\mu \nu}}(x) dx^{\mu} dx^{\nu}+ e^{-2A(y)} 

2581:      {g_{i\bj}}(y) dy^i d\bar y^{\bj}\ .

2582: \label{warpmetric}

2583: \eeq

2584: However, in this work we perform our analysis in the

2585: unwarped Calabi-Yau manifold since in the large radius limit

2586: the warp factor approaches one and the metrics 

2587: of the two manifolds coincide  \cite{GKP,FP}. 

2588: This in turn also implies that the metrics on the moduli space

2589: of deformations agree and as a consequence the kinetic terms

2590: in the low

2591: energy effective actions are the same. The difference appears 

2592: in the potential when some of the 

2593: Calabi-Yau zero modes are rendered massive. However, 

2594: the regime $e^{2A(y)} \approx 1$ should be understood as a 

2595: special limit and it would be desirable to generalize compactifications

2596: to warped backgrounds \eqref{warpmetric}.

2597: 

2598: Let us now turn to the derivation of the four-dimensional effective action 

2599: by redoing the Kaluza-Klein reduction of the ten-dimensional

2600: type IIB action given in \eqref{10d-lagr} for the truncated orientifold

2601: spectrum. 

2602: 

2603: \subsubsection{The reduction of the $N=2$ vector sector} 

2604: 

2605: We first consider the reduction of the vector sector 

2606: of the $N=2$ supergravity theory obtained by type IIB 

2607: Calabi-Yau compactification. As discussed in section \ref{revIIB} the 

2608: four-dimensional bosonic components of the vector multiplets

2609: are $(z^K,V^K)$. The complex scalars $z^K$ parameterize the

2610: complex structure deformations of $Y$. Under the 

2611: orientifold projection these $N=2$ multiplets split into

2612: chiral multiplets with bosonic components $(z^k)$ and vector

2613: multiplets $(V^\lambda)$ for $O3/O7$ orientifolds and chiral multiplets $(z^\lambda)$

2614: and vectors $(V^k)$ in $O5/O9$ orientifolds. Since the reduction of the vector 

2615: sector is very similar for both the $O3/O7$ and $O5/O9$ case

2616: we will first concentrate on the first case 

2617: and later give a rule how to translate these results to 

2618: $O5/O9$ orientifolds. 

2619: 

2620: %In order to not overload the notation 

2621: %we now choose the same indices for the $O3/O7$ and $O5/O9$ case. 

2622: %In other words, the range of the indices is 

2623: %\bea \label{indices1}

2624: %  O3/O7:\quad \kappa = 1\ldots h^{(2,1)}_+\ , \quad k = 1\ldots h^{(2,1)}_-\ , \quad \hat k = 0\ldots h^{(2,1)}_-\ ,\\

2625: %  O5/O9:\quad \kappa = 1\ldots h^{(2,1)}_-\ , \quad k = 1\ldots h^{(2,1)}_+\ , \quad \hat k = 0\ldots h^{(2,1)}_+\ ,\nn

2626: %\eea 

2627: %where the hatted indices run over one more value. This is consistent with the fact that by 

2628: %\eqref{cso} the three-form $\Omega$ is in $H^{(3)}_-$ for $O3/O7$ setups and in $H^{(3)}_+$ for 

2629: %$O5/O9$ setups.

2630: 

2631: Due to the split of the cohomology

2632: $H^{(3)}= H^{(3)}_+\oplus H^{(3)}_-$ the real symplectic basis 

2633: $(\alpha_\Kh , \beta^\Lh)$

2634: of $H^{(3)}$ can be split into $(\alpha_{\kappa} , \beta^{\lambda})$

2635: of $H^{(3)}_+$ and $(\alpha_{\hat k} , \beta^{\hat l})$

2636: of $H^{(3)}_-$.  

2637: % where the upper sign is for $O3/O7$ and the lower for $O5/O9$. 

2638: Eqs.\ \eqref{int-numbers1} continue to hold which implies

2639: that both basis are symplectic and obey

2640: \begin{equation}\label{sbasiso}

2641:  \int \alpha_{\kappa} \wedge \beta^{\lambda} 

2642: = \delta^{\lambda}_{\kappa}\ ,

2643: \qquad

2644: \int \alpha_{\hat k} \wedge \beta^{\hat l} 

2645: = \delta^{\hat l}_{\hat k}\ ,

2646: \end{equation}

2647: with all other intersections vanishing. Since $\hat C_4$ is even under $\sigma^*$

2648: the expansion \eqref{exp1} led to $h^{(3)}_+ = h^{(2,1)}_+$ vectors 

2649: $V^\kappa$. The three-form $\Omega$ is odd under $\sigma^*$ and thus has to be expanded

2650: in a basis of $H^{(3)}_-$ according to 

2651: \begin{equation}

2652: \Omega(z^k) = Z^{\hat k}\alpha_{\hat k} - \mathcal{F}_{\hat k}

2653: \beta^{\hat k}\ , 

2654: \label{cond-1}

2655: \end{equation} 

2656: while the other periods $(Z^{\kappa},\mathcal{F}_{\kappa})$

2657: vanish

2658: \beq

2659:  Z^{\kappa}|_{z^{\kappa}=0}= \int_Y \Omega\wedge\beta^\kappa = 0 \ , \qquad

2660: \mathcal{F}_{\kappa}\big|_{z^{\kappa}=0} 

2661: =  \int_Y \Omega\wedge\alpha_\kappa = 0 \ .

2662: \eeq

2663: As a consequence the metric on the space 

2664: of complex structure deformations reduces to 

2665: \beq\label{csmetrico} 

2666: G_{ kl} = 

2667: \frac{\partial}{\partial z^{k}}

2668: \frac{\partial}{\partial\bar z^{l}}

2669: \  K_{\rm cs}\ , \qquad

2670:  K_{\rm cs} = -\ln\Big[ - i \int_Y \Ox \wedge \bar \Ox\Big] 

2671: = -\ln i\Big[ Z^{\hat k} \bar \cF_{\hat k}    - \bar Z^{\hat k} {\mathcal{F}}_{\hat k} \Big]

2672: \ ,

2673: \eeq

2674: replacing \eqref{csmetric}. The reduction of the kinetic terms for 

2675: the $N=2$ vector sector thus yields \cite{TGL1}

2676: \beq \label{red-vector}

2677: S^{(4)\, vec}_{O3/O7}\ =\ \int - G_{k { l}} \; dz^{k} \wedge *d\bar z^{l}

2678:     +\tfrac{1}{4}\text{Im}\; \cM_{\kappa \lambda}\; 

2679:     F^{\kappa}\wedge *F^{\lambda}

2680:      +\tfrac{1}{4}\text{Re}\; \cM_{\kappa \lambda}\;

2681:      F^{\kappa}\wedge F^{\lambda}\ , 

2682: \eeq

2683: where $F^\lambda = dV^\lambda$. Recall that the vectors $V^k$ as well 

2684: as the graviphoton are projected out by the orientifold projection \eqref{o3-projection}

2685: and do not appear in \eqref{red-vector}. 

2686: The coupling matrix $\cM_{\kappa \lambda}(z^k)$ in front of the remaining vectors

2687: $V^\kappa$ is evaluated on the subspace where $z^\kappa=0$ and thus depends on 

2688: $z^k$ only. The analysis for $O5/O9$ orientifolds is in complete anology to the

2689: $O3/O7$ case, with the difference that the vectors $V^k$ and scalars $z^\lambda$

2690: remain in the spectrum while $V^\lambda$ and $z^k$ is projected out. 

2691: The equations \eqref{sbasiso} -- \eqref{red-vector} can be translated to this second case by replacing

2692: the indices $k,l \rightarrow \kappa, \lambda$, $\kh \rightarrow \hat \kappa$ and 

2693: $\kappa,\lambda \rightarrow k,l$. 

2694: This is consistent with the fact that by 

2695: \eqref{cso} the three-form $\Omega$ is in $H^{(3)}_-$ for $O3/O7$ setups and in $H^{(3)}_+$ for 

2696: $O5/O9$ setups.

2697:    

2698: \subsubsection{The reduction of the $N=2$ quaternionic sector} 

2699: 

2700: Similar to the vector sector, we now perform the reduction 

2701: of the hypermultiplet couplings \eqref{q-metrB}. One computes

2702: the four-dimensional effective action by redoing the Kaluza-Klein reduction 

2703: of the ten-dimensional type IIB action given in \eqref{10d-lagr} for the 

2704: truncated orientifold spectrum. Equivalently, one can impose the 

2705: orientifold constrains on the four-dimensional $N=2$ effective action 

2706: \eqref{action3}. In type IIB the metric on the quaternionic manifold depends on the 

2707: complexified K\"ahler deformations $t$ and the dilaton and is obtained from 

2708: the intersection numbers in the even cohomologies. Hence, in order to perform the reduction 

2709: to $N=1$ we first need to reconsider the structure of the metrics \eqref{Kmetric} and 

2710: the intersection numbers \eqref{int-numbers} for the orientifold. 

2711: 

2712: Note that $\sigma^*J=J$ and  $\sigma^*\hat B_2=-\hat B_2$ holds for both IIB orientifold 

2713: projections. This implies that the constraints on the space of 

2714: K\"ahler structure deformations are the same for $O3/O7$ as well as $O5/O9$ setups.

2715: Let us discuss them in the following.

2716: Corresponding to the decomposition  

2717: $H^{(1,1)}=H^{(1,1)}_+\oplus H^{(1,1)}_-$ also

2718: the harmonic (1,1)-forms $\omega_A$

2719: split into 

2720: $\omega_A = (\omega_\alpha, \omega_a)$

2721: %A= 1,\ldots,h^{(1,1)}

2722: %\alpha = 1,\ldots,h^{(1,1)}_+

2723: %a = 1,\ldots,h^{(1,1)}_-

2724: such that 

2725: $\omega_\alpha$ is a basis of 

2726: $H^{(1,1)}_+$ and 

2727: $\omega_a$ is a basis of 

2728: $H^{(1,1)}_-$. This in turn results in a decomposition of the intersection 

2729: numbers $\KK_{ABC}$ given in \eqref{int-numbers}.

2730: Under the orientifold projection

2731: only $\KK_{\alpha\beta\gamma}$ and $\KK_{\alpha bc}$ can be non-zero

2732: while $\KK_{\alpha \beta c}= \KK_{abc} =0$ has to hold. 

2733: Since the K\"ahler-form $J$ is invariant 

2734: we also conclude from \eqref{int-numbers}

2735: that $\KK_{\alpha b}=0=\KK_{a}$. To summarize,

2736: keeping only the invariant intersection numbers results in

2737: \begin{eqnarray}\label{constr}

2738:   \KK_{\alpha \beta c}= \KK_{abc} =\KK_{\alpha b}=\KK_{a}=0\ ,

2739: \end{eqnarray}

2740: while all the other intersection numbers can be non-vanishing.\footnote{From

2741: a supergravity point of view this 

2742: has been also observed in refs.\ \cite{ADAF}.}

2743: Inserting \eqref{constr} into \eqref{Kmetric} we derive

2744: \begin{eqnarray} \label{splitmetr}

2745:   G_{\alpha \beta}=

2746:   -\frac{3}{2}\left( \frac{\KK_{\alpha \beta}}{\KK}-

2747:   \frac{3}{2}\frac{\KK_\alpha \KK_\beta}{\KK^2} \right)\ , \qquad

2748:   G_{a b}=-\frac{3}{2} \frac{\KK_{a b}}{\KK}\ , \qquad

2749: G_{\alpha b}\ =\ G_{a \beta}\ =\ 0\ ,

2750: \end{eqnarray}

2751: where

2752: \begin{equation}\label{intO3}

2753:   \KK_{\alpha\beta}=\KK_{\alpha\beta\gamma}\; v^\gamma\ , 

2754: \quad \ 

2755: \KK_{ab}=\KK_{ab\gamma}\; v^\gamma\ ,\quad

2756: \KK_{\alpha}=\KK_{\alpha \beta\gamma}\; v^\beta v^\gamma\ ,

2757:   \quad \KK=\KK_{\alpha \beta \gamma} \; v^\alpha v^\beta v^\gamma

2758: \ .

2759: \end{equation}

2760: %

2761: We see that the metric $G_{AB}$ given in \eqref{Kmetric}

2762: is block-diagonal with respect to the 

2763: decomposition $H^{(1,1)}=H^{(1,1)}_+\oplus H^{(1,1)}_-$.

2764: For later use let us also record the inverse metrics

2765: \begin{eqnarray}\label{Ginvers}

2766:   G^{\alpha \beta}

2767:   =  -\frac{2}{3} \KK \KK^{\alpha \beta} + 2 v^\alpha v^\beta\ ,

2768: \qquad

2769:   G^{a b} =

2770:    - \frac{2}{3} \KK \KK^{a b}\ ,

2771: \end{eqnarray}

2772: where $\KK^{\alpha \beta}$ and $\KK^{a b}$ are the inverse 

2773: of $\KK_{\alpha \beta}$ and

2774: $\KK_{a b}$, respectively. 

2775: 

2776: The $N=2$ hypermultiplet couplings are reduced by inserting \eqref{constr}  - \eqref{Ginvers}

2777: and truncating to the orientifold spectrum as summarized in table \ref{N=1spectrumtab}. 

2778: Since this the orientifold spectrum of $O3/O7$ setups differs from the one

2779: of $O5/O9$ setups, one obtains two different effective actions. 

2780: Together with the standard Einstein-Hilbert term and the contributions 

2781: from the reduction of the $N=2$ vectors \eqref{red-vector} one finds after 

2782: Weyl rescaling \cite{TGL1}

2783: \begin{eqnarray}  \label{S_scalarO3}

2784: S^{(4)}_{O3/O7} &=&\int -\tfrac{1}{2} R *\mathbf{1} 

2785:   - G_{k \bar l} \; dz^{k} \wedge *d\bar z^{l}

2786:   -G_{\alpha \beta} \; dv^\alpha \wedge *dv^\beta - G_{ab}\; db^a \wedge * db^b \nonumber \\

2787:   && - dD \wedge * dD

2788:     -\tfrac{1}{24}e^{2 D}\cK\, dl \wedge * dl -   \tfrac{1}{6}e^{2D} \cK

2789:   G_{ab}\left(dc^a-l db^a \right) \wedge *\left(dc^b-l db^b \right)

2790:   \nonumber \\

2791:   &&-\tfrac{3}{8 \cK}e^{2D} G^{\alpha \beta}  \Big(d\rho_\alpha-

2792:   \KK_{\alpha a b} c^a db^b \Big)

2793:   \wedge

2794:   *\Big(d\rho_\beta - \KK_{\beta cd} c^c db^d 

2795: \Big) \nn \\

2796:   &&+\tfrac{1}{4}\text{Im}\; \cM_{\kappa \lambda}\; 

2797:     F^{\kappa}\wedge *F^{\lambda}

2798:      +\tfrac{1}{4}\text{Re}\; \cM_{\kappa \lambda}\;

2799:      F^{\kappa}\wedge F^{\lambda}\ ,  

2800: \end{eqnarray}

2801: and

2802: \begin{eqnarray} \label{S_scalarO5} 

2803:   S^{(4)}_{O5/O9}&=&\int -\tfrac{1}{2} R *\mathbf{1} 

2804:   - G_{\kappa \bar \lambda} \; dz^{\kappa} \wedge *d\bar z^{\lambda}

2805:   - G_{\alpha \beta} \; dv^\alpha \wedge *dv^\beta  

2806:    \nn \\ 

2807:   &&- G_{ab}\; db^a \wedge * db^b -d D \wedge * dD - \tfrac{1}{6}e^{2D} \cK G_{\alpha \beta}\; dc^\alpha \wedge * dc^\beta

2808:    \nn \\ 

2809:   &&- \tfrac{3}{2\KK}e^{2D}(dh+\tfrac{1}{2}(d\rho_a b^a -\rho_a db^a))

2810:   \wedge *(dh+\tfrac{1}{2}(d\rho_a b^a - \rho_a db^a))\nonumber \\

2811:   &&- \tfrac{3}{8 \cK}e^{2D}G^{ab}(d \rho_a - \KK_{ac\alpha} c^\alpha db^c)

2812:   \wedge *(d \rho_b - \KK_{bd\beta} c^\beta db^d)\ . \nn\\

2813:   &&+\tfrac{1}{4}\text{Im}\; \cM_{k l}\; 

2814:     F^{k}\wedge *F^{l}

2815:      +\tfrac{1}{4}\text{Re}\; \cM_{k l}\;

2816:      F^{k}\wedge F^{l}\ ,  

2817: \end{eqnarray} 

2818: where we have expressed the result in a chiral basis and used the index conventions given in table \ref{CYObasisB}. In 

2819: contrast to ref. \cite{TGL1} we have expressed the effective actions 

2820: in terms of the string frame K\"ahler structure deformations $v^\alpha$ and

2821: the four-dimensional dilaton 

2822: \beq \label{def-D_B}

2823:   e^D = e^\phi\ (\cK/6)^{-1/2}\ ,

2824: \eeq

2825: where $e^\phi$ is the ten-dimensional dilaton.  

2826: This ends our computation of the orientifold bulk action. In remains to cast 

2827: \eqref{S_scalarO3} and \eqref{S_scalarO5} into the standard $N=1$ form. 

2828: 

2829: 

2830: \subsection{The K\"ahler potentials and gauge-couplings}

2831: \label{Kpo_gaugeIIB}

2832: 

2833: Our next task is to transform the actions \eqref{S_scalarO3} and \eqref{S_scalarO5} into the standard

2834: $N=1$ supergravity form with chiral multiplets where it is 

2835: expressed in terms of a K\"ahler potential $K$, 

2836: a holomorphic superpotential $W$ and the holomorphic gauge-kinetic coupling 

2837: functions $f$ as follows \cite{WB,GGRS}

2838: \beq\label{N=1action}

2839:   S^{(4)} = -\int \tfrac{1}{2}R * \mathbf{1} +

2840:   K_{I \bar J} DM^I \wedge * D\bar M^{\bar J}  

2841:   + \tfrac{1}{2}\text{Re}f_{\kappa \lambda}\ 

2842:   F^{\kappa} \wedge * F^{\lambda}  

2843:   + \tfrac{1}{2}\text{Im} f_{\kappa \lambda}\ 

2844:   F^{\kappa} \wedge F^{\lambda} + V*\mathbf{1}\ ,

2845: \eeq

2846: where

2847: \beq\label{N=1pot}

2848: V=

2849: e^K \big( K^{I\bar J} D_I W {D_{\bar J} \bar W}-3|W|^2 \big)

2850: +\tfrac{1}{2}\, 

2851: (\text{Re}\; f)^{-1\ \kappa\lambda} D_{\kappa} D_{\lambda}

2852: \ .

2853: \eeq

2854: Here the $M^I$ collectively denote  all

2855: complex scalars in chiral multiplets present in the theory  and 

2856: $K_{I \bar J}$ is a K\"ahler metric satisfying

2857: $  K_{I\bar J} = \partial_I \bar\partial_{\bar J} K(M,\bar M)$.

2858: The scalar potential is expressed in terms of the 

2859: K\"ahler-covariant derivative $D_I W= \partial_I W + 

2860: (\partial_I K) W$. 

2861: 

2862: In the reduction we did not find any scalar potential, such that one immediately concludes

2863: $W=0$ and $D_\kappa=0$. Next we need to find a complex structure on the space of

2864: scalar fields such that the metrics computed in \eqref{S_scalarO3} and \eqref{S_scalarO5}

2865: are manifestly K\"ahler. 

2866: 

2867: \subsubsection{The K\"ahler potential: $O3/O7$ setups}

2868:  

2869: As we saw in \eqref{csmetrico} the 

2870: complex structure deformations $z^{k}$ are already good K\"ahler

2871: coordinates with $G_{k\bar l}$ being the appropriate K\"ahler metric. 

2872: For the remaining fields the definition of 

2873: the K\"ahler coordinates is not so obvious. 

2874: Guided by refs.\ \cite{HL,BBHL} we define

2875: \beq \label{def-coordsO3}

2876:  \fe - i\, \fa = i\tau + iG^a \omega_a - T_\alpha \tilde \omega^\alpha

2877: \eeq

2878: where 

2879: \beq \label{def-A}

2880:    \pev = \fe + i\, \feh =e^{-\phi} e^{-\hat B_2 + iJ}\ ,\qquad \fa = e^{-\hat B_2} \wedge \sum_{q=0,2,4,6} \hat C_q |_{scalar}\ , 

2881: \eeq

2882: are sums of even forms.

2883: In \eqref{def-A} we have defined $\hat C_q |_{scalar}$ to be the part of $\hat C_q$ yielding scalars in $D=4$, 

2884: e.g.~$\hat C_4|_{scalar} = \rho_\alpha\, \tilde \omega^\alpha$. Expanding all the forms in

2885: \eqref{def-coordsO3} by using \eqref{def-A},\eqref{exp-B} and \eqref{exp1}

2886: the coordinates take the form \cite{TGL1}

2887: \bea \label{tau}

2888:   \tau &=& C_0+ie^{-\phi} \ , \qquad  G^a =c^a-\tau b^a\ ,\nn \\

2889:   T_\alpha &=&  i( \rho_\alpha - \tfrac{1}{2} \cK_{\alpha ab}c^a b^b) + \tfrac{1}{2} e^{-\phi} \cK_\alpha

2890:                - \zeta_\alpha\ , 

2891: \eea

2892: where\footnote{The definition of $\zeta_\alpha$ is unique up to a constant

2893: which does not enter into the metric. The possibility of a non-zero constant

2894: is important for the formulation in terms of linear multiplets in 

2895: section~\ref{IIB_lin}.}  

2896: \beq\label{zetadef}

2897: \KK_{\alpha} =  \KK_{\alpha\beta\gamma} v^\beta v^\gamma \ ,\qquad

2898: \zeta_\alpha =  -\frac{i}{2(\tau-\bar \tau)}\ \KK_{\alpha b c}G^b (G- \bar G)^c\ .

2899: \eeq

2900: In ref.~\cite{TGL1} it was checked explicitly that in terms of these coordinates

2901: the metric of \eqref{S_scalarO3} is K\"ahler with the K\"ahler potential \cite{TGL1}

2902: \beq \label{kaehlerpot-O7-1}

2903:    K = K_{\rm cs}(z,\bar z)   

2904:    + K^{\rm Q}(\tau,T,G)\ , \qquad K_{\rm cs} = - \text{ln}\Big[-i\int_Y \Omega(z) \wedge \bar \Omega(\bar z) \Big]\ ,

2905: \eeq

2906: and 

2907: \beq\label{kaehlerpot-Kk}

2908:   K^{\rm Q} =  -\text{ln}\big[-i(\tau - \bar \tau)\big]

2909:   - 2 \text{ln}\big[ \text{Vol}_E(\tau,T,G)\big]=-\ln\big[2 e^{-4D}\big] \ ,

2910: \eeq

2911: where we have used \eqref{def-D_B} in order to evaluate the last equality.

2912: The Einstein frame volume 

2913: $\text{Vol}_E (Y) = \frac{1}{6} e^{-\frac{3}{2} \phi} \KK_{\alpha\beta\gamma} v^\alpha v^\beta v^\gamma$

2914: in \eqref{kaehlerpot-Kk}

2915: should be understood 

2916: as a function of the K\"ahler coordinates $(\tau,T,G)$ 

2917: which enter by solving (\ref{tau}) for $e^{-\phi/2} v^\alpha$ in terms of $(\tau,T,G)$.

2918: Unfortunately this solution cannot be given explicitly and therefore $\text{Vol}_E$ is known

2919: only implicitly via $e^{-\phi/2} v^\alpha(\tau,T,G)$.\footnote{This is in complete analogy

2920: to the situation encountered in compactifications

2921: of M-theory on Calabi-Yau fourfolds studied in \cite{HL}. 

2922: This is no coincidence and can be understood from the fact

2923: that this theory can be lifted to F-theory on Calabi-Yau fourfolds

2924: which in a specific limit  is related to  orientifold

2925: compactifications of type IIB \cite{Sen}. In section \ref{F-theory}

2926: we make this more explicit by checking this correspondence on the level of the 

2927: effective actions.} 

2928: In chapter \ref{lin_geom_of_M} we show that the definition of the

2929: K\"ahler coordinates \eqref{tau} and the K\"ahler potential

2930: \eqref{kaehlerpot-O7-1} can be understood somewhat

2931: more conceptually in a dual formalism using linear multiplets 

2932: $L^\alpha$ instead of the chiral multiplets $T_\alpha$.

2933: 

2934: Let us return to the K\"ahler potential (\ref{kaehlerpot-O7-1}).

2935: $K_{cs}$ and the first term in \eqref{kaehlerpot-Kk} are the standard

2936: K\"ahler potentials for the complex structure deformations

2937: and the dilaton, respectively. 

2938: $\text{Vol}_E(\tau,G,T)$ also depends on $\tau$ and therefore the metric mixes $\tau$

2939: with $T_\alpha$ and $G^a$. It is block diagonal in the 

2940: complex structure deformations which do not mix with the other scalars.

2941: Hence, the moduli space locally has the form

2942: \beq \label{modulispaceO3}

2943: \cM_{N=1} = \tilde \cM^{\rm SK} \times\, \tilde \cM^{\rm Q}\ ,

2944: \eeq

2945: where each factor is a K\"ahler manifold. The manifold $ \tilde \cM^{\rm SK}$ 

2946: has complex dimension $h_-^{(1,2)}$ and is a special K\"ahler manifold in 

2947: that $K_{\rm cs}$ satisfies \eqref{csmetrico}. It parameterizes the complex 

2948: structure deformations of $Y$ respecting the orientifold constraint \eqref{Omegatransf}.

2949: On the other hand, $\tilde \cM^{\rm Q}$ is a $h^{(1,1)}+1$-dimensional K\"ahler manifold inside 

2950: the quaternionic manifold $\cM^{\rm Q}$. Local coordinates are given by the 

2951: fields $\tau, G^a,T_\alpha$ arising in the expansion \eqref{def-coordsO3}. 

2952: Also the K\"ahler potential $K^{\rm Q}(\tau, G,T)$ fulfills

2953: special properties inherited from the underlying 

2954: special quaternionic manifold.   

2955: To see this, let us bring $K^{\rm Q}$ in a slightly different form. 

2956: Using the explicit expansion \eqref{def-coordsO3} of $\pev$ one checks 

2957: that up to a trivial K\"ahler transformation the K\"ahler potential 

2958: \eqref{kaehlerpot-Kk} can be rewritten as 

2959: \beq \label{def-Phi}

2960:    K^{\rm Q} = -2\ln\, \Phi_B(\fe)\ , \qquad \Phi_B(\fe)\equiv i\big<\pev, \pevb \big>\ , 

2961: \eeq

2962: where $\pev=\fe+i\feh$ is defined in \eqref{def-A} and $\feh(\fe)$ has 

2963: to be evaluated. In \eqref{def-Phi} 

2964: we abbreviated the skew-symmetric product $\big<\varphi, \psi \big>$ 

2965: for two sums of even forms $\varphi=\varphi_0+\varphi_2 + \varphi_4 + \varphi_6$

2966: and $\psi=\psi_0+\psi_2+\psi_4+\psi_6$ as \cite{HitchinGCM} 

2967: \beq \label{symp-form}

2968:   \big<\varphi, \psi \big> =  \int_Y\ \sum_{m} (-1)^{m} \varphi_{2m} \wedge \psi_{\,6-2m}\ .

2969:                           %= \int_Y \varphi_0\wedge \psi_6 - \varphi_2\wedge \psi_4 + \varphi_4\wedge \psi_2

2970:                           %   -\varphi_6\wedge \psi_0\ .

2971: \eeq

2972: The function $\Phi_B$ can be identified with Hitchins functional on 

2973: a generalized complex manifold \cite{HitchinGCM} evaluated for the simple form $\pev$ defined

2974: in \eqref{def-A} (see \cite{GLprep} for more details).  

2975: We discuss the geometry of $\tilde \cM^{\rm Q}$ in greater detail in section \ref{geom_of_modspace}. 

2976: 

2977: Although not immediately obvious from its definition

2978: $K^{\rm Q}$ obeys a no-scale type condition in that it satisfies

2979: \beq\label{NScond}

2980:     \frac{\partial K}{\partial N^I}\,  (K^{-1})^{ I\bar J}\, 

2981:     \frac{\partial K}{\partial \bar N^{\bar J}} = 4\ ,

2982: \eeq

2983: where $N^I = (\tau,G^a,T_\alpha)$.\footnote{For $G^a=0$

2984: this has already been observed in \cite{GKP,BBHL,DWG,DAFT}.} 

2985: This equality can be shown by direct computation as done in \cite{TGL1}.

2986: Alternatively, it can be deduced from the fact that $\Phi_B$ defined in 

2987: \eqref{def-Phi} is homogeneous of degree two, i.e.~$\Phi_B(a\, \fe) = a^2\, \Phi_B(\fe)$ for all 

2988: $a\in \bbR $ \cite{HitchinGCM}. Using \eqref{def-coordsO3} a

2989: simple calculation shows that $K^{\rm Q}=-2\ln \Phi_B$ satisfies \eqref{NScond}.

2990: From \eqref{N=1pot} we see that 

2991: \eqref{NScond} implies $V\ge 0$ which we also show 

2992: in the linear multiplet formalism in section \ref{IIB_lin}.

2993: For $\tau=\text{const.}$ the right hand side of \eqref{NScond} is found to

2994: be equal to $3$ as it is the case in the standard no-scale K\"ahler potentials

2995: of \cite{NS}. 

2996: 

2997: Let us relate \eqref{kaehlerpot-O7-1} to the known K\"ahler potentials 

2998: in the literature. 

2999: First of all, for $G^a=0$ and thus 

3000: $T_\alpha=i \rho_\alpha + \frac{1}{2}\KK_\alpha $

3001:  the K\"ahler potential 

3002: \eqref{kaehlerpot-O7-1} reduce to the one given in \cite{BBHL}.

3003: Secondly, for one overall K\"ahler modulus $v$ parameterizing the volume

3004: (i.e. for $h^{(1,1)}_+=1$, $T_{\alpha}\equiv T$) 

3005: the K\"ahler potential $K^{\rm Q}$ reduces to 

3006: $K=-3 \text{ln}( T +\bar T )$

3007: which coincides with the K\"ahler potential 

3008: determined in \cite{GKP}.

3009: 

3010: Before we turn to the discussion of the $O5/O9$ case let us note that 

3011: $K$ is invariant under the $SL(2,\bbR)$ transformations inherited from the

3012: ten-dimensional type IIB theory. In the orientifold theory

3013: this symmetry acts on $\tau$ by fractional linear transformations 

3014: exactly as in $D=10$ and transforms $(b^a,c^a)$ as a doublet, such that

3015: \beq \label{Sl2}

3016:   \tau \to \frac{a\tau + b}{c\tau +d}\ ,\qquad G^a \to \frac{G^a}{c\tau +d}\ , \qquad ad-bc=1\ .

3017: \eeq

3018: Under the $SL(2,{\bf R})$ only the second term of $K$ 

3019: given in \eqref{kaehlerpot-Kk}

3020: transforms but this transformation is just a K\"ahler transformation.

3021: This can be seen from \eqref{tau} and the fact that $e^{-\phi/2} v^\alpha$ and $z^k$ are invariant.

3022: This symmetry reduces to $SL(2,\bbZ)$ in the full string theory, which is nothing

3023: but the invariance group of a two-torus. This torus becomes part of the space-time 

3024: in the formulation of `F-theory' \cite{Vafa}. We discuss in section \ref{F-theory} 

3025: the embedding of $O3/O7$ orientifolds into this theory on the level of the effective action.   

3026: 

3027: 

3028: \subsubsection{The K\"ahler potential: $O5/O9$ setups}

3029: 

3030: In the action \eqref{S_scalarO5}  we immediately see that the complex

3031: structure deformations $z^\kappa$ are again already good K\"ahler coordinates.

3032: For the remaining fields we find the appropriate K\"ahler coordinates

3033: to be

3034: \beq \label{def-coordsO5} 

3035:   \feh - i \fa\ =\  t^\alpha \omega_\alpha - A_b\, \tilde \omega^b - S\, \text{vol}(Y)\ ,

3036: \eeq 

3037: where $\feh=\I\, \pev$ and $\fa$ are defined in \eqref{def-A} and we have used that in 

3038: $O5/O9$ setups the axion $C_0$ gets projected out. Furthermore, we denoted by 

3039: $\text{vol}(Y)=\cK^{-1} J\wedge J \wedge J$ the to one normalized volume form of $Y$.

3040: Using 

3041: the expansions \eqref{transJo}, \eqref{exp-B}

3042: and \eqref{expO5} we obtain the explicit expressions \cite{TGL1}

3043: \begin{eqnarray}\label{Kcoord}

3044:   t^\alpha &=& e^{-\phi} v^\alpha - i c^\alpha\ , \qquad A_a \ =\ \N_{ab} b^b + i \rho_a \ ,\\

3045:   S & = & \tfrac{1}{6}  e^{-\phi}\, \cK + i h 

3046:            - \tfrac{1}{4} (\text{Re}\N^{-1})^{ab} A_a (A+\bar A)_b\ , \nn

3047: \end{eqnarray}

3048: where we inserted 

3049: \beq\label{Ndef}

3050:    \N_{ab}(t) \equiv \KK_{ab \alpha} t^\alpha \ ,\qquad \int C_6 = h+\tfrac{1}{2} \rho_a b^a\ . 

3051: \end{equation}

3052: The matrix $\N_{ab}$ depends holomorphically on the coordinates $t^\alpha$ which

3053: ensures that $\tilde \cM^{\rm Q}$ is K\"ahler \cite{FS,HL}.

3054: In the variables given in \eqref{Kcoord} the K\"ahler potential reads \cite{TGL1}

3055: \beq \label{O5-Kaehlerpot} 

3056:   K \ =\ K_{cs}(z,\bar z) + K^{\rm Q}(S,t,A)\ , \qquad K_{cs} = -\text{ln}\Big[-i\int\Omega \wedge \bar \Omega \Big]

3057: \eeq

3058: with

3059: \bea \label{kaehlerpot-KkO5}

3060:  K^{\rm Q}& =& - \text{ln}\Big[\tfrac{1}{48}\KK_{\alpha \beta \gamma}(t+\bar t)^\alpha 

3061:         (t+\bar t)^\beta (t+\bar t)^\gamma  \Big]\nn \\

3062:      && - \text{ln}\Big[S + \bar S + \tfrac{1}{4} (A + \bar A)_a (\text{Re}\N^{-1})^{ab} 

3063:         (A + \bar A)_b \Big]\\

3064:      &=& -\ln\big[2 e^{-4D}\big]\ . \nn

3065: \eea

3066: where we used \eqref{def-D_B}.

3067: The check that $K$ indeed reproduces \eqref{S_scalarO5} is straightforward, since 

3068: \eqref{O5-Kaehlerpot} is closely related to the quaternionic

3069: `K\"ahler potential' given in \cite{FS} and we can make use

3070: of their results.\footnote{Note however,

3071: that the complex structure changed non-trivially.

3072: In \cite{FS} the standard $t \sim v + i b$ formed complex coordinates.}

3073: The same reference  already observed 

3074: that for a holomorphic matrix $\N$ the quaternionic geometry is also 

3075: K\"ahler. This situation was also found in compactifications

3076: of the heterotic string to $D=3$ on a circle \cite{HL}. 

3077: 

3078: {}From \eqref{O5-Kaehlerpot} we infer that the $N=1$ moduli space admits 

3079: the local product structure $\tilde \cM^{\rm SK}\times \tilde \cM^{\rm Q}$

3080: similar to \eqref{modulispaceO3}. However, in $O5/O9$ orientifolds $\tilde \cM^{\rm SK}$

3081: is a special K\"ahler manifold spanned by the $h^{(2,1)}_+$ complex scalars $z^\kappa$, which 

3082: are the ones projected out in $O3/O7$ orientifolds. $\tilde \cM^{\rm Q}$ is spanned by the complex scalars 

3083: $S,t^\alpha,A_a$ and thus is of complex dimension $h^{1,1} + 1$ as in $O3/O7$ setups. 

3084: Furthermore, also $K^{\rm Q}$ for orientifolds with $O5/O9$ planes can be rewritten in terms 

3085: of the functional $\Phi_B(\feh)$ as

3086: \beq \label{def-PhiO5}

3087:   K^{\rm Q}\ =\ -2 \ln \Phi_B(\feh)\ ,\qquad  \Phi_B(\feh ) \equiv i\big<\pev,\pevb \big>\ ,

3088: \eeq

3089: where $\pev=\fe(\feh)+i\feh$ are defined in \eqref{def-A}.

3090: The functional dependence of $K^{\rm Q}$ on $\pev$ is the same as in \eqref{def-Phi} for $O3/O7$ orientifolds.  

3091: This can be understood from the fact that $\pev$ only depends on the NS-NS sector 

3092: variables, which are the same in both types of orientifolds. 

3093: Nevertheless, the local structure of $\tilde \cM^{\rm Q}$ is different for both orientifold theories.

3094: This becomes appearent when one expresses $K^{\rm Q}$ in terms of

3095: proper K\"ahler coordinates. In $O5/O9$ setups this corresponds to the fact that 

3096: $\Phi_B$ is a function of $\feh$ as needed for \eqref{def-coordsO5}. Hence, in order 

3097: to express $K^{\rm Q}$ in terms of the K\"ahler coordinates $S,t,A$ as in \eqref{kaehlerpot-KkO5} 

3098: one evaluates $\fe(\feh)$. Let us end this discussion by remarking that

3099: $\Phi_B$ is also homogeneous of degree two in $\feh$, such that 

3100: by using \eqref{def-coordsO5} one extracts a no-scale type condition equivalent to \eqref{NScond}. 

3101: 

3102: 

3103: \subsubsection{The gauge-couplings: $O3/O7$ and $O5/O9$ setups}

3104: 

3105: Our next task is to determine

3106: the gauge-kinetic coupling functions $f_{\kappa \lambda}$ 

3107: and show that they are holomorphic in the moduli. We do 

3108: this only for $O3/O7$ orientifolds, since the result easily 

3109: translates to the $O5/O9$ case. As explained in section 

3110: \ref{O37_effective_act} this is achieved by an appropriate replacement of the 

3111: indices.

3112: By comparing the actions \eqref{red-vector} and \eqref{N=1action} one finds

3113: \beq \label{gauge-couplingsO3}

3114:   f_{\kappa \lambda}

3115: =-\tfrac{i}{2} \,

3116: \bar{\cM}_{\kappa \lambda}\Big|_{z^{\kappa}=0=\bar z^{\kappa}}\ ,

3117: \eeq 

3118: where 

3119: ${\cM}_{\kappa \lambda}$ is the 

3120: $N=2$ gauge kinetic matrix given in \eqref{defM}

3121: evaluated at ${z^{\kappa}=\bar z^{\kappa}}=0$.

3122: Its holomorphicity in the complex structure deformations $z^k$ is not

3123: immediately obvious but can be shown by using

3124: \eqref{defM} and  \eqref{gauge-c}.

3125: More precisely, from \eqref{defM} together with 

3126: the decomposition of $H^{(3)}$ expressed by \eqref{H3split}

3127: and \eqref{sbasiso} we infer that ${\cM}_{\hat K \hat L}$

3128: is block diagonal or in other words

3129: ${\cM}_{\kappa \hat l} = 0.$ Multiplying ${\cM}_{\kappa \hat l}$

3130: with $X^{\hat l}$ and using $X^\lambda=0$ together with

3131: \eqref{gauge-c} we further conclude 

3132: \beq\label{Fdiag}

3133: {\cF}_{\kappa \hat l}\Big|_{z^{\kappa}=0=\bar z^{\kappa}}=0\ .

3134: \eeq

3135: Finally inserting \eqref{cond-1} and \eqref{Fdiag}

3136: into \eqref{gauge-c}

3137: we arrive at \cite{TGL1}

3138: \begin{eqnarray}\label{fholo}

3139:   f_{\kappa \lambda} (z^k)

3140: =-\tfrac{i}{2} 

3141: \mathcal{F}_{\kappa \lambda}\Big|_{z^{\kappa}=0=\bar z^{\kappa}}\ ,

3142: \end{eqnarray} 

3143: which is manifestly holomorphic since $\mathcal{F}_{\kappa \lambda}(z^k)$

3144: are holomorphic functions of the complex structure

3145: deformations $z^k$.

3146: 

3147: 

3148: 

3149: 

3150: 

3151: 

3152: 

3153: 

3154: 

3155: 

3156: 

3157: 

3158: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3159: %

3160: %   IIA: The Effective action

3161: %

3162: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3163: 

3164: 

3165: 

3166: 

3167: \section{Type IIA Calabi-Yau orientifolds}

3168: \label{IIA_orientifolds}

3169: 

3170: In this section we determine the $N=1$ supergravity action 

3171: obtained by compactification of Type IIA string theory on 

3172: a Calabi-Yau orientifold. The orientifold projection 

3173: $\cO=(-1)^{F_L} \Omega_p \sigma$ was already defined in 

3174: \eqref{oproj} and includes an anti-holomorphic isometric involution 

3175: $\sigma$. In section \ref{spectrum-IIA} we extract the $N=1$ spectrum by 

3176: identifying the fields invariant under $\cO$. The corresponding 

3177: effective action is calculated in section \ref{eff_actIIA}.

3178: It is shown to be compatible with $N=1$ supersymmetry in section \ref{Kpo_gaugeIIA},

3179: where we determine the K\"ahler potential and gauge-kinetic coupling functions.  

3180: 

3181: 

3182: \subsection{The $N=1$ spectrum \label{spectrum-IIA}}

3183: 

3184: In order to determine the  $\mathcal{O}$-invariant states let us recall

3185: that the ten-dimensional RR forms $\hat C_1$ and $\hat C_3$ 

3186: are odd 

3187: under $(-1)^{F_L}$ while all other fields are even. 

3188: Under the world-sheet 

3189: parity $\Omega_p$ on the other hand $\hat B_2, \hat C_3$ are odd

3190: with all other fields being even.

3191: As a consequence the 

3192: $\mathcal{O}$-invariant states have to satisfy

3193: \cite{BH}

3194: \begin{equation} \label{fieldtransf}

3195: \begin{array}{lcl}

3196: \sigma^*  \hat \phi &=& \  \hat \phi\ , \\

3197: \sigma^*   \hat g &=& \ \hat g\ , \\

3198: \sigma^*   \hat B_2 &=& -  \hat B_2\ ,

3199: \end{array}

3200: \hspace{2cm}

3201: \begin{array}{lcl}

3202: \sigma^*   \hat C_1 &=&  -  \hat C_1\ , \\

3203: \sigma^*   \hat C_3 &=& \  \hat C_3\ ,

3204: \end{array}

3205: \end{equation}

3206: while the deformations of the Calabi-Yau metric are constrained

3207: by \eqref{constrJ} and \eqref{constrO}.\footnote{%

3208: Following the argument presented in 

3209: \cite{BH} we note that the involution does not change

3210:  under deformations of $Y$. 

3211: This is due to its involutive property 

3212: and the fact that we identify involutions which differ

3213: by diffeomorphisms.

3214: Therefore we fix an involution and restrict the deformation space by demanding 

3215: \eqref{constrJ} and \eqref{constrO}. }

3216: 

3217: 

3218: As we recalled in the previous section the massless modes are in one-to-one

3219: correspondence with the harmonic forms on $Y$. The space of harmonic forms

3220: splits under the involution $\sigma$ into even and odd eigenspaces

3221: \beq \label{cohom-split}

3222:    H^p(Y)\ =\ H^p_+ \oplus H^p_-\ \ .

3223: \eeq

3224: Depending on the transformation properties given in \eqref{fieldtransf}

3225: the $\mathcal{O}$-invariant states reside either in $H^p_+$ or in $H^p_-$

3226: and as a consequence the number of states is reduced.

3227: We summarize all non-trivial cohomology groups including their basis elements 

3228: in table \ref{CYObasis}.\\

3229: 

3230: \begin{table}[h]

3231: \begin{center}

3232: \begin{tabular}{| c || c | c| c | c | c | c |} \hline

3233:    \rule[-0.3cm]{0cm}{0.9cm} cohomology group &  $\ H^{(1,1)}_+\ $ & 

3234:    $\ H^{(1,1)}_-\ $ & $\ H^{(2,2)}_+\ $ & $\ H^{(2,2)}_-\ $ & $\ H^{(3)}_+\ $ & $\ H^{(3)}_-\ $

3235:    \\ \hline

3236:    \rule[-0.3cm]{0cm}{0.8cm} dimension &  $h^{(1,1)}_+$  & $h^{(1,1)}_- $  

3237:                                        &  $h^{(1,1)}_-$  & $h^{(1,1)}_+$ 

3238:                                        &  $h^{(2,1)}+1$  &  $h^{(2,1)}+1$ 

3239:    \\ \hline

3240:    \rule[-0.3cm]{0cm}{0.8cm} basis     & $\omega_\alpha$ & $\omega_a$

3241:                                        & $\tilde \omega^a$ & $\tilde \omega^\alpha$

3242:    & $ a_{\Kh}$ & $b^{\Kh}$ \\ \hline

3243: \end{tabular}

3244: \caption{\small \label{CYObasis}

3245: \textit{Cohomology groups and their basis elements.}}

3246: \end{center}

3247: \end{table}

3248: 

3249: $\omega_\alpha, \omega_a$ denote

3250:  even and odd $(1,1)$-forms while 

3251: $\tilde\omega^\alpha, \tilde\omega^a$ denote odd and even

3252:  $(2,2)$-forms. The number of even $(1,1)$-forms is equal to the number

3253: of odd $(2,2)$-forms and vice versa since the 

3254: volume form which is 

3255: proportional to $J \wedge J \wedge J$ is odd and thus Hodge duality

3256: demands $h^{(1,1)}_+ = h^{(2,2)}_- ,\ h^{(1,1)}_- = h^{(2,2)}_+$.

3257: This can also be seen from the fact that the non-trivial

3258: intersection numbers are 

3259: \beq \label{basis-int}

3260:   \int \omega_\alpha \wedge \tilde \omega^\beta =

3261:   \delta^{\beta}_\alpha\ ,\quad \alpha,\beta = 1, \ldots, h^{(1,1)}_+\ ,

3262:   \qquad 

3263:   \int \omega_a \wedge \tilde \omega^b = \delta^{b}_a\ , \quad

3264:   a,b=1,\ldots,h^{(1,1)}_-\ ,

3265: \eeq

3266: with all other pairings vanishing.

3267: {}From the volume-form being odd 

3268: one further infers $h^{(3,3)}_+=0,$ $h^{(3,3)}_-=1$ and 

3269: $h^{(0,0)}_+=1,\ h^{(0,0)}_-=0$.

3270: 

3271: 

3272: $H^{3}$ can be decomposed 

3273: independently of the complex structure as

3274: $H^{3}=H^3_+ \oplus H^3_-$ where 

3275: the (real) dimensions of  both $H^3_+$ and $H^3_-$

3276: is equal and given by $h^{3}_+ =h^{3}_-= h^{(2,1)}+1$.

3277: Again this is a consequence of Hodge duality

3278: together with the fact that the volume-form is odd.

3279: It implies that for each element $a_\Kh \in H^3_+$ 

3280: there is a dual element $b^\Lh \in H^3_-$

3281: with the intersections

3282: \beq \label{basis_ab}

3283:  \int a_\Kh \wedge b^\Lh = \delta^\Lh_\Kh \ , \qquad 

3284:  \Kh, \Lh = 0,\ldots, h^{(2,1)}\ .

3285: \eeq

3286: Compared to \eqref{int-numbers1} this amounts to a symplectic rotation

3287: such that all $\alpha$-elements are chosen to be even and 

3288: all $\beta$-elements are chosen to be odd but with the intersection

3289: numbers unchanged.

3290: The orientifold projection breaks this symplectic invariance 

3291: or in other words fixes a particular symplectic gauge 

3292: which groups all basis elements into even and odd. 

3293: This in turn implies that the basis $(a_\Kh,b^\Kh)$ is only one possible choice. 

3294: However, since the calculation simplifies considerably for this basis, we first restrict 

3295: to this special case and later give the general results with calculations summarized in 

3296: section \ref{IIA_lin}.  

3297: 

3298: 

3299: 

3300: In the remainder of this subsection we determine the $N=1$ spectrum 

3301: which survives the orientifold projections.

3302: Let us first discuss the K\"ahler moduli. 

3303: From the eqs. \eqref{constrJ} and \eqref{fieldtransf} we see that both

3304: $J$ and $\hat B_2$ are odd and hence have to be expanded

3305: in a basis $\omega_a$ of odd harmonic $(1,1)$-forms 

3306: \beq \label{expJB}

3307:   J\ =\ v^a(x)\, \omega_a\ ,\qquad  \hat B_2\ =\ b^a(x)\, \omega_a\ , \qquad a = 1,\ldots, h^{(1,1)}_-\ .

3308: \eeq

3309: In contrast to \eqref{fieldexp} the four-dimensional 

3310: two-form $B_2$ gets projected out due to \eqref{fieldtransf} and the fact

3311: that $\sigma$ acts trivially on the flat dimensions. 

3312: $v^a$ and $b^a$ are space-time scalars and 

3313: as in $N=2$ they can be combined into complex coordinates

3314: \beq \label{def-t}

3315:   t^a = b^a + i\,  v^a\ , \qquad \Jc = B_2 + i J\ ,

3316: \eeq

3317: where we have also introduced the complexified 

3318: K\"ahler form $\Jc$.

3319: We see that in terms of the field variables the same complex

3320: structure is chosen as in $N=2$ but the dimension of the K\"ahler moduli

3321: space is truncated from $h^{(1,1)}$ to $h^{(1,1)}_-$.

3322: 

3323: 

3324: The number of complex structure deformations is similarly reduced since

3325: \eqref{constrO} constrains the possible deformations.

3326: To see this one performs a symplectic rotation on 

3327: \eqref{Omegaexp} and expands $\Omega$ in the basis of

3328: $H^p_+ \oplus H^p_-$, i.e.\ as\footnote{Let us stress that at this

3329: point all $N=2$ relations are still intact since \eqref{Omegapm}

3330: is just a specific choice of the standard $N=2$ basis \eqref{Omegaexp}.} 

3331: \beq\label{Omegapm}

3332: \Omega(z) = Z^\Kh(z)\, a_\Kh - \cF_\Lh(z)\, b^\Lh\ .

3333: \eeq 

3334: Inserted into \eqref{constrO} one finds 

3335: \bea \label{Z=0}

3336:    \I(e^{-i\theta} Z^\Kh)\ =\  0\ , \qquad

3337:    \R(e^{-i\theta} \cF_\Kh )\ = 0\ .

3338: \eea  

3339: The first set of equations  are $h^{(2,1)}+1$ real conditions

3340: for $h^{(2,1)}$ complex scalars $z^K$.

3341: One of these equations is redundant due to the 

3342: scale invariance \eqref{crescale} of $\Omega$.

3343: More precisely, the phase of $e^{-h}$ can be used to 

3344: trivially satisfy $\I(e^{-i\theta} Z^\Kh)= 0$ for one of the $Z^\Kh$.

3345: Thus $\I(e^{-i\theta} Z^\Kh)=0$

3346: projects out $h^{(2,1)}$ real scalars, i.e.\

3347: half of the complex structure deformations. 

3348: Furthermore, in section \ref{eff_actIIA} we will see 

3349: the remaining real complex structure deformations 

3350: span a Lagrangian submanifold $\cM^{\rm cs}_\bbR$

3351: with respect to the K\"ahler form

3352: inside $\cM^{\rm cs}$. 

3353: Note that the second set of equations in \eqref{Z=0}

3354: $\R (e^{-i\theta} \cF_\Kh ) = 0$ 

3355: should not be read as equations determining

3356: the $z^K$ but is a constraint on the periods (or equivalently

3357: the Yukawa couplings) of the Calabi-Yau

3358: which has to be fulfilled in order to admit an involutive symmetry 

3359: with the property \eqref{constrO}.\footnote{This can also be seen 

3360: as conditions arising in consistent truncations of 

3361: $N=2$ to $N=1$ theories as discussed in ref.\ \cite{ADAF}.} 

3362: 

3363: 

3364: 

3365: As we have just discussed 

3366: the complex rescaling \eqref{crescale}

3367: is reduced to the freedom of a real rescaling by \eqref{constrO}. 

3368: %Thus we restrict to rescalings \eqref{crescale} with $f$ being real 

3369: %when restricted to $\cM^{\rm cs}_\bbR$, 

3370: Under these transformations $\Omega$ and the K\"ahler potential $\Kcs$ 

3371: change as

3372: \beq \label{real_K}

3373:   \Omega\to\Omega\, e^{-\R(h)}\ , \qquad \Kcs\to\Kcs + 2 \R(h)\ ,

3374: \eeq

3375: when restricted to $\cM^{\rm cs}_\bbR$. This freedom can be used to set one of 

3376: the $\R (e^{-i\theta}Z^{\Kh})$ equal to one and tells us

3377: that $\Omega$ depends only on $h^{(2,1)}$ real deformation parameters. 

3378: However, it will turn out to be more

3379: convenient to leave this gauge freedom intact and define 

3380: a complex `compensator' $C=re^{-i\theta}$ with the transformation property

3381: $C\to C e^{\R (h)}$.\footnote{This is reminiscent of the situation

3382: encountered in the computation of the entropy of $N=2$ black holes

3383: where it is also convenient to leave this scale invariance intact \cite{OSV}.}

3384: Later on we will relate $r$ to 

3385: the inverse of the four-dimensional dilaton 

3386: so that the scale invariant function $C\Omega$ depends on 

3387: $h^{(2,1)}+1$ real parameters. 

3388: Using \eqref{Omegapm} $C\Omega$ enjoys

3389: the expansion 

3390: \beq \label{decompO}

3391:   C \Omega\ =\ \R (C Z^\Kh)\, a_\Kh - i\I (C \cF_\Lh)\,  b^\Lh\ .

3392: \eeq

3393:  

3394: 

3395: 

3396: We are left with the expansion of the ten-dimensional fields

3397: $\hat C_1$ and $\hat C_3$ into  harmonic forms. 

3398: {}From \eqref{fieldtransf} we learn that $\hat C_1$ is odd

3399: and so together with the fact that

3400: $Y$ posses no harmonic one-forms 

3401: and $\sigma$ acts trivially on the flat dimensions

3402: the entire $\hat C_1$ is projected out. This 

3403: corresponds to the fact that the $N=2$ graviphoton $A^0$ is removed

3404: from the gravity multiplet, 

3405: which in $N=1$ only consists of the metric $g_{\mu \nu}$ as 

3406: bosonic component.

3407: Finally, $\hat C_3$ is even and thus can be expanded according to

3408: \beq \label{form-exp}

3409:   \hat C_3 = \cc_3(x) 

3410: + A^\alpha(x) \wedge \omega_\alpha + \CC_3\ ,\qquad

3411: \CC_3 \equiv \xi^\Kh(x)\,  a_\Kh \ ,

3412: \eeq

3413: where $\xi^\Kh$ are $h^{(2,1)}+1$ real 

3414: scalars, $A^\alpha$ are $h^{(1,1)}_+$ one-forms

3415: and $\cc_3$ is a three-form in four dimensions.

3416: $\cc_3$ contains no physical degree of freedom but as we will see 

3417: in section~\ref{O6sup} corresponds to a 

3418: constant flux parameter in the superpotential.

3419: The real scalars 

3420: $\xi^\Kh$ have to combine with the 

3421: $h^{(2,1)}$ real  complex structure deformations 

3422: and the dilaton to form chiral multiplets.

3423: In the next section we will find that the appropriate complex fields 

3424: arise from the combination

3425: \beq\label{Omegacdef}

3426:   \Omegac\ =\ \CC_3 + 2i\R(C\Omega) \ .

3427: \eeq

3428: Expanding $\Omegac$  in a basis \eqref{basis_ab} of $H^3_+(Y)$ 

3429: and using \eqref{decompO} and \eqref{form-exp} we have 

3430: \beq\label{newO}

3431: \Omegac\ =\ 2 N^\Kh a_\Kh \ ,\qquad

3432: N^\Kh= \tfrac{1}{2} \int \Omega_c\wedge \beta^\Kh =

3433: \tfrac{1}{2}\big(\xi^\Kh + 2i \R (C Z^\Kh)\big)\ . 

3434: \eeq

3435: Due to the orientifold projection the two three-forms 

3436: $\Omega$ and $C_3$

3437: each lost half of their degrees of freedom and combined

3438: into a new complex three-form $\Omegac$. 

3439: As we will show in more detail in the next section

3440: the `good' chiral coordinates in the $N=1$ orientifold

3441: are the periods of $C\Omega$ directly while in $N=2$

3442: the periods agree with the proper field variables only 

3443: in special coordinates.

3444: 

3445: Let us summarize the resulting $N=1$ spectrum.

3446: It assembles into a gravitational multiplet,

3447: $h^{(1,1)}_+$ vector multiplets and 

3448: $(h^{(1,1)}_- + h^{(2,1)}+1)$ chiral multiplets.

3449: We list the bosonic parts of the $N=1$ supermultiplets in table

3450: \ref{N=1spectrumA} \cite{BH}. We see that the $h^{(1,1)}$ $N=2$

3451: vector multiplets split into $h^{(1,1)}_+$ $N=1$ vector multiplets and 

3452: $h^{(1,1)}_-$ chiral multiplets while the 

3453: $h^{(2,1)}+1$ hypermultiplets are reduced to $h^{(2,1)}+1$ chiral multiplets.

3454: 

3455: \begin{table}[h]

3456: \begin{center}

3457: \begin{tabular}{|l|c|c|} \hline 

3458:  \rule[-0.3cm]{0cm}{0.8cm} 

3459: multiplets& multiplicity & bosonic components\\ \hline\hline

3460:  \rule[-0.3cm]{0cm}{0.8cm} 

3461:  gravity multiplet&1&$g_{\mu \nu} $ \\ \hline

3462:  \rule[-0.3cm]{0cm}{0.8cm} 

3463:  vector multiplets&   $h_+^{(1,1)}$&  $A^{\alpha} $\\ \hline

3464:  \rule[-0.3cm]{0cm}{0.8cm} 

3465:  {chiral multiplets} &   $h_-^{(1,1)}$& 

3466: $t^a$ \\ \hline

3467:  \rule[-0.3cm]{0cm}{0.8cm} 

3468: {chiral multiplets} 

3469: & $ h^{(2,1)}+1$ &$ N^\Kh$\\ 

3470: \hline

3471: \end{tabular} 

3472: \caption{\label{N=1spectrumA} \textit{$N =1$ spectrum of $O6$ orientifold compactification.}}

3473: \end{center}

3474: \end{table} 

3475: 

3476: 

3477: \subsection{The effective action}

3478: \label{eff_actIIA}

3479: 

3480: In this section we calculate the four-dimensional effective action of type 

3481: IIA orientifolds by performing a Kaluza-Klein reduction of the 

3482: ten-dimensional type IIA action \eqref{10dact} taking the 

3483: orientifold constraints into account. Equivalently this amounts to

3484: imposing the orientifold projections on the $N=2$ action of 

3485: section~\ref{revIIA}.

3486: %Additionally we discuss the reduced moduli space of K\"ahler and complex 

3487: %structure deformations preserving \eqref{constrJ} and \eqref{constrO} in 

3488: %somewhat more detail. 

3489: Inserting \eqref{expJB}, \eqref{decompO}, \eqref{form-exp} into

3490: the ten-dimensional type IIA action \eqref{10dact} and performing a Weyl 

3491: rescaling of the four-dimensional metric 

3492: %$g_{\mu \nu} \rightarrow (\cK/6) g_{\mu \nu}$ 

3493: we find \cite{TGL2}

3494: \bea \label{act1}

3495:   S^{(4)}_{O6} &=& \int -\tfrac{1}{2} R*\mathbf{1} 

3496: - G_{a b}\, dt^a \wedge * d \bar t^b 

3497: + \tfrac{1}{2} \text{Im}\, \cN_{\alpha \beta}\ F^\alpha \wedge * F^\beta 

3498:         + \tfrac{1}{2} \text{Re}\, \cN_{\alpha \beta}\ F^\alpha

3499:         \wedge F^\beta   \nn\\

3500:       && 

3501: \quad      -\, d D \wedge * dD  -\, G_{K L}(q)\, dq^K \wedge * dq^L 

3502:          +\tfrac{1}{2} e^{2D}\, \text{Im}\, \cM_{ \Kh  \Lh}\, 

3503:          d\xi^{\Kh} \wedge * d\xi^{\Lh} \ ,

3504: \eea

3505: where $F^\alpha = dA^\alpha$.

3506: Let us discuss  the different couplings appearing in \eqref{act1}

3507: in turn. 

3508: Apart from the standard Einstein-Hilbert term the first line arises

3509: from the projection of the $N=2$ vector multiplets action.

3510: As we already observed the orientifold projection reduces the number

3511: of K\"ahler moduli from $h^{(1,1)}$  to $h^{(1,1)}_-$ ($t^A\to t^a$)

3512: but leaves the complex structure on this component of the moduli space

3513: intact.  Accordingly the metric $G_{ab}(t)$ is inherited from the

3514: metric $G_{AB}$ of the $N=2$ moduli space 

3515: $\cM^{SK}$ given in \eqref{Kmetric}.

3516: Since the volume form is odd only intersection numbers with one or

3517: three odd basis elements in 

3518: \eqref{int-numbers} can be non-zero and consequently one has

3519: \beq \label{van-int}

3520:   \cK_{\alpha \beta \gamma} = \cK_{\alpha a b} = \cK_{\alpha a} = \cK_{\alpha} = 0\ ,

3521: \eeq  

3522: while all other intersection numbers can be non-vanishing.\footnote{From a supergravity 

3523: point of view this has been discussed also in \cite{ADAF}.} 

3524: This implies that the metric $G_{AB}(t^A)$ of \eqref{Kmetric} is block

3525: diagonal and obeys

3526: \begin{eqnarray} \label{splitmetrIIA}

3527:   G_{a b}=

3528:   -\frac{3}{2}\left( \frac{\KK_{a b}}{\KK}-

3529:   \frac{3}{2}\frac{\KK_a \KK_b}{\KK^2} \right)\ , \qquad

3530:   G_{\alpha \beta}=-\frac{3}{2} \frac{\KK_{\alpha \beta}}{\KK}\ , \qquad

3531: G_{\alpha b}\ =\ 0\ ,

3532: \end{eqnarray}

3533: where

3534: \begin{equation}\label{intO6}

3535:   \KK_{ab}=\KK_{abc}\; v^c\ , 

3536: \quad \ 

3537: \KK_{\alpha \beta}=\KK_{\alpha \beta a}\; v^a\ ,\quad

3538: \KK_{a}=\KK_{a b c}\; v^b v^c\ ,

3539:   \quad \KK=\KK_{abc} \; v^a v^b v^c

3540: \ .

3541: \end{equation}

3542: In comparison to type IIB orientifolds the opposite intersection 

3543: numbers vanish as can be seen by comparing \eqref{van-int} with \eqref{constr}.

3544: This is due to the fact that the K\"ahler form $J$ transforms in IIA and IIB 

3545: orientifolds with a relative minus sign under the action of $\sigma$. 

3546: 

3547: The same consideration also truncates  the $N=2$ gauge-kinetic 

3548: coupling matrix $\cN_{\Ah \Bh}$ explicitly given in \eqref{def-cN}. 

3549: Inserting \eqref{van-int} and \eqref{intO6} one arrives at 

3550: \beq \label{def-N_alph_bet} 

3551:   \text{Re} \cN_{\alpha \beta} = - \cK_{\alpha \beta a} b^a\ , \qquad 

3552:   \text{Im} \cN_{\alpha \beta} = \cK_{\alpha \beta}\ ,\qquad \cN_{a \alpha}=\cN_{0 \alpha}=0\ .

3553: \eeq

3554: (The other non-vanishing matrix elements $\cN_{\ah\bh}$ arise in the potential 

3555: \eqref{U-pot} once fluxes are turned on.) 

3556: 

3557: Let us now discuss the 

3558: terms in the second line of \eqref{act1} arising from the

3559: reduction  of the $N=2$ hypermultiplet action which is 

3560: determined by the quaternionic metric \eqref{q-metr}.

3561: $D$ is the the four-dimensional dilaton defined in \eqref{4d-dilaton}.

3562: The metric $G_{KL}$ is inherited from the $N=2$

3563: K\"ahler metric $G_{K \bar L}(z,\bar z)$ given in \eqref{csmetric}

3564: and thus is 

3565: the induced metric on the submanifold $\cM^{\rm cs}_\bbR$ 

3566: defined by the constraint \eqref{constrO}.

3567: More precisely, the complex structure deformations respecting \eqref{constrO}

3568: can be determined from \eqref{Kod-form}

3569: by considering infinitesimal variations of 

3570: $\Omega$

3571: \beq

3572:   \Omega(z + \delta z) \ =\ \Omega(z) + \delta z^K (\partial_{z^K} \Omega)_{z} \ 

3573:               =\ \Omega(z) - \delta z^K( \Kcs_{z^K} \Omega - \chi_K)_z \ .

3574: \eeq 

3575: Now we impose the condition that both 

3576: $\Omega(z+\delta z)$ and $\Omega(z)$ satisfy \eqref{constrO}. 

3577: This implies locally 

3578: \beq  \label{constr2}

3579:   \delta z^K\, \partial_{z^K} \Kcs = \delta\bar z^K\, \partial_{\bar z^K} \Kcs\ , \qquad 

3580:   \delta z^K\sigma^* \chi_K = e^{2i \theta} \delta \bar z^K\bar \chi_K \ ,

3581: \eeq 

3582: where $\partial_{z^K} \Kcs$ and $\chi_K$ are restricted to $\cM^{\rm cs}_\bbR$.

3583: Using the fact that $\Kcs$ is a K\"ahler potential and therefore $\partial_{z^K}\Kcs\neq 0$, we conclude from

3584: the first equation in \eqref{constr2} that for each $\delta z^K$ either the 

3585: real or imaginary part has to be zero. This is consistent with the observation

3586: of the previous section that coordinates of $\cM^{\rm cs}_\bbR$ can be 

3587: identified with 

3588: the real or imaginary part of the complex structure deformations $z^K$.

3589: To simplify the notation we call these deformations collectively

3590: $q^K$ and denote the embedding map by 

3591: $\rho:\cM^{\rm cs}_\bbR \hookrightarrow \cM^{\rm cs}$.

3592: Locally this corresponds to 

3593: \beq \label{embmap1}

3594:   \rho:\ q^K=(q^s,q^\sigma)\ \mapsto\ z^K=(q^s,iq^\sigma)\ ,

3595: \eeq

3596: for some splitting $z^K=(z^s,z^\sigma)$. In other words, 

3597: the local coordinates on $\cM^{\rm cs}_\bbR$ 

3598: are $\R z^s=q^s$ and $\I z^\sigma = q^\sigma$ while $\I z^s=0=\R z^\sigma$.

3599: Using the second equation in \eqref{constr2}, the embedding

3600: map \eqref{embmap1} and the expression \eqref{chi_barchi} for the $N=2$ metric $G_{K\bar L}$ we also deduce that

3601: the K\"ahler form vanishes when pulled back to $\cM^{\rm cs}_\bbR$. 

3602: In summary we 

3603: have

3604: \beq \label{def-G}

3605:   \rho^*(G_{K \bar L}\, dz^K d \bar z^L)\, \equiv\, G_{KL}(q)\, dq^K dq^L\ , \quad

3606:   \rho^*(iG_{K \bar L}\, dz^K \wedge d \bar z^L)\, =\, 0\ .

3607: \eeq

3608: The first equation defines the induced metric while the second equation

3609: implies that $\cM^{\rm cs}_\bbR$

3610: is a Lagrangian submanifold of $\cM^{\rm cs}$ with respect to the 

3611: K\"ahler-form.

3612: 

3613: 

3614: 

3615: Finally, coming back to the action \eqref{act1}

3616: the matrix $\cM_{\Kh \Lh}$ is defined in analogy with

3617: \eqref{defM} as

3618: \beq \label{defM2}

3619:   \int a_\Kh \wedge * a_\Lh = -\text{Im}\; \cM_{\Kh \Lh} \ , \qquad 

3620: %+(\text{Re}\; \cM)

3621:  % (\text{Im}\; \cM)^{-1}(\text{Re}\; \cM))_{\Kh \Lh}

3622: %\ , \qquad\qquad

3623: % \int a_\Kh\wedge * b^\Lh = 0\ , \nn\\

3624:  \int b^\Kh \wedge * b^\Lh\  = \ -(\text{Im}\; \cM)^{-1\ \Kh \Lh}\ , 

3625: %  -((\text{Re}\; \cM)(\text{Im}\; \cM)^{-1})_{\Kh}^\Lh\ .  \nn

3626: \eeq

3627: where $\text{Im} \cM_{\Kh \Lh}$ can be given explicitly 

3628: in terms of the periods by inserting \eqref{Z=0} into \eqref{gauge-c} \cite{TGL1}.

3629: %This yields

3630: %\bea \label{gauge-r}

3631: %\I \cM_{\Kh \Lh}=-\I \mathcal{F}_{\Kh \Lh}+2 \frac{(\text{Im}\; \mathcal{F})_{\Kh \Mh} \R (CZ^\Mh)

3632: %   (\text{Im}\; \mathcal{F})_{\Lh \Nh}\R (CZ^\Nh) }{\R (CZ^\Nh)(\text{Im}\; \mathcal{F})_{\Nh\Mh} 

3633: %    \R (CZ^\Mh)}\ .

3634: %\eea 

3635: Similarly one obtains $\R\cM_{\Kh \Lh}=0$ 

3636: consistent with the fact that \eqref{defM} implies that $\int a_\Kh\wedge * b^\Lh$ vanishes

3637: for the special basis $(a_\Kh,b^\Kh)$.

3638: 

3639: This ends our discussion of

3640: the effective action obtained by applying the orientifold projection. 

3641: The next step is to rewrite the action \eqref{act1} 

3642: in the standard $N=1$ supergravity form

3643: which we turn to now.

3644: 

3645: 

3646: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3647: %

3648: %   Kahler potential 

3649: %

3650: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3651: 

3652: \subsection{The K\"ahler potential and gauge-couplings}

3653: \label{Kpo_gaugeIIA}

3654: 

3655: 

3656: The standard $N=1$ supergravity the action is expressed 

3657: in terms of a K\"ahler potential $K$, a holomorphic superpotential $W$ 

3658: and the holomorphic gauge-kinetic coupling 

3659: functions $f$ as given in \eqref{N=1action}. Hence, our task is to  

3660: find $K,f$ and $W$ for the type IIA orientifolds. As an immediate 

3661: observation one finds that \eqref{act1} includes no potential, such 

3662: that $W=0$ and $D_\alpha=0$. It is also not difficult to read 

3663: off the gauge-kinetic coupling function $f_{\alpha \beta}$.

3664: Comparing \eqref{act1} with \eqref{N=1action} 

3665: using \eqref{def-N_alph_bet} and \eqref{def-t}

3666: one infers 

3667: \beq \label{gauge-A}

3668:   f_{\alpha \beta}\ =\ -i \bar \cN_{\alpha \beta}\ =\ i \cK_{\alpha \beta a}  t^a \ .

3669: \eeq

3670: As required by $N=1$ supersymmetry the $f_{\alpha \beta}$ 

3671: are indeed holomorphic. Note that they are linear in the $t^a$ moduli

3672: and do not depend on the complex structure and $\xi$-moduli.

3673: 

3674: {}From \eqref{act1} we also immediately observe that

3675: the orientifold moduli space has the product structure  

3676: \beq \label{direct-mod}

3677:   \cM_{N=1}=\tilde\cM^{\rm SK} \times \tilde\cM^{\rm Q}\ .

3678: \eeq

3679: The first factor $\tilde\cM^{\rm SK}$ is a subspace of the 

3680: $N=2$ moduli space $\cM^{\rm SK}$ with dimension $h^{(1,1)}_-$ 

3681: spanned by  the complexified K\"ahler deformations $t^a$.

3682: The second factor $\tilde\cM^{\rm Q}$ is a subspace of the quaternionic

3683: manifold $\cM^{\rm Q}$ with

3684: dimension $h^{(2,1)} +1$ 

3685: spanned by  the complex structure deformations $q^K$, the dilaton $D$

3686: and the scalars $\xi^{\hat K}$ arising from $C_3$.

3687: Let us discuss both factors in turn.

3688: 

3689: As we already stressed earlier the metric $G_{ab}$ 

3690: of \eqref{act1} defined in \eqref{splitmetr} is a trivial

3691: truncation

3692: of the $N=2$ special K\"ahler metric  \eqref{Kmetric} and therefore remains

3693: special K\"ahler. The K\"ahler potential is given by

3694: \beq \label{Kks}

3695: K^{\rm K}\ =\ - \ln \Big[\tfrac{i}{6}\cK_{a b c} (t -\bar t)^a (t -\bar t)^b (t -\bar t)^c \Big]

3696:   \ = \ - \ln \Big[\tfrac{4}{3}  \int_Y J \wedge J \wedge J\Big]\ ,

3697: \eeq

3698: where $J$ is the K\"ahler form in the string frame.

3699: Moreover, $K^{\rm K}$ can be obtained from 

3700: the prepotential $f(t)=-\tfrac{1}{6} \cK_{abc}t^a t^b t^c$ 

3701: by using equation \eqref{Kinz}. 

3702: It is well known that $K^{\rm K}$ obeys the standard no-scale condition

3703: \cite{NS}

3704: \beq \label{no-scale1}

3705:   K_{t^a} K^{t^a \bar t^b} K_{ \bar t^b}\ =\ 3\ . 

3706: \eeq

3707: 

3708: The geometry of the second component  $\tilde\cM^{\rm Q}$ in \eqref{direct-mod}

3709: is considerably more complicated. This is due to the fact that 

3710: \eqref{newO} defines a new complex structure on the field space. In the following

3711: we sketch the calculation of the K\"ahler potential for the basis 

3712: $(a_\Kh,b^\Kh)$ and only summarize the results for a generic symplectic basis.

3713: The details of this more involved calculation will be presented in section \ref{IIA_lin}. 

3714: 

3715: To begin with, let us define the compensator $C$ introduced in section \ref{spectrum-IIA} as 

3716: \beq \label{def-C}

3717:    C\ =\ e^{-D-i\theta} e^{\Kcs(q)/2}\ , \qquad C \rightarrow C e^{\R\, h(q)}\ ,

3718: \eeq 

3719: where $\Kcs$ is the K\"ahler potential defined in \eqref{csmetric} restricted

3720: to the real subspace $\cM^{\rm cs}_{\bbR}$. We also displayed the transformation 

3721: behavior of $C$ under real K\"ahler transformations \eqref{real_K}. With this at hand one 

3722: defines the scale invariant variable

3723: \beq \label{l-def}

3724:    l^\Kh \ =\  \R(C Z^\Kh(q))\ .

3725: \eeq

3726: Inserted into \eqref{act1} and using 

3727: the Jacobian matrix encoding the change of variables $(e^D,q^K) \rightarrow l^\Kh$

3728: the second line \eqref{act1} simplifies as\footnote{%

3729: The calculation of this result can be found in section \ref{IIA_lin}.}

3730: \beq \label{IIAQ}

3731:   \cL^{(4)}_{\rm Q} =  2 e^{2D}\, \text{Im}\, \cM_{\Kh \Lh}\, 

3732: (dl^\Kh \wedge * dl^\Lh  + \tfrac{1}{4} d\xi^{\Kh} \wedge * d\xi^{\Lh})\ .  

3733: \eeq

3734: We see that the scalars $l^\Kh$ and $\xi^\Kh$ nicely combine 

3735: into complex coordinates 

3736: \beq \label{Ncoords}

3737:    N^\Kh\ =\ \tfrac{1}{2}\xi^\Kh +  i l^\Kh\ 

3738: =\ \tfrac{1}{2}\xi^\Kh +  i \R(C Z^\Kh)

3739: = \tfrac{1}{2} \int \Omega_c\wedge b^\Kh 

3740: \ , 

3741: \eeq

3742: which we anticipated in equation \eqref{newO}.

3743: The important fact to note here is that $\tilde\cM^{\rm Q}$

3744: is equipped with a new complex structure and the corresponding

3745: K\"ahler coordinates 

3746: coincide with half of the periods of $\Omegac$.

3747: This is in contrast to the situation in $N=2$ where one of the periods

3748: ($Z^0$) is a gauge degree of freedom and the K\"ahler

3749: coordinates are the special coordinates $z^K = Z^K/Z^0$.

3750: 

3751: In order to show that the metric in \eqref{IIAQ} is K\"ahler we need the

3752: explicit expression for the K\"ahler potential. Using \eqref{Z=0} in \eqref{gauge-c} 

3753: one obtains straightforwardly

3754: \beq

3755: 2 e^{2D} \text{Im}\, \cM_{\Kh \Lh} = \partial_{N^\Kh} \partial_{\bar N^\Lh} K^{\rm Q}\ ,

3756: \eeq

3757: where 

3758: \beq \label{KQsimple}

3759:  K^{\rm Q} = -2 \ln\big[4i\cF(CZ)\big]\ , \qquad

3760: \cF\big(\R(CZ)\big) = \frac{i}2\,  \R(C Z^\Kh)\, \I(C\cF_\Kh)\ .

3761: \eeq

3762: Alternatively, using \eqref{decompO} and $*\Omega =- i \Omega$

3763: one derives the integral representation

3764: \beq \label{intKQ}

3765:   K^{\rm Q}\ = - 2\ln\Big[2\int_Y \R(C \Omega)\wedge *\R(C\Omega)\Big]=\ - \ln\, e^{-4D} \ , 

3766: \eeq

3767: where in the second equation we used \eqref{def-C} and \eqref{csmetric}. In 

3768: the form \eqref{intKQ} the dependence of $K^{\rm Q}$ on the coordinates $N^\Kh$

3769: is only implicit and given by means of their definition \eqref{Ncoords}.  

3770: Also $K^{\rm Q}$ obeys a no-scale type condition in that it 

3771: satisfies

3772: \bea \label{no-scale2}

3773:   K_{N^\Kh} K^{N^\Kh \bar N^\Lh} K_{\bar N^\Lh} = 4\ ,

3774: \eea

3775: which can be checked by direct calculation.

3776: 

3777: 

3778: The analysis so far started from the symplectic basis 

3779: $(a_\Kh,b^\Kh)$ introduced in \eqref{basis_ab},

3780: determined the K\"ahler coordinates in \eqref{Ncoords}

3781: and derived  the K\"ahler potential $K^{\rm Q}$

3782: in terms of the prepotential $\cF$ in \eqref{KQsimple} or as an 

3783: integral representation in \eqref{intKQ}. Now we need to ask

3784: to what extent  this result depends on the choice of 

3785: the basis \eqref{basis_ab}. Or in other words let us redo

3786: the calculation starting from an arbitrary symplectic basis

3787: and determine the K\"ahler potential and the proper field variables

3788: for the corresponding orientifold theory. 

3789: Let us first recall the situation 

3790: in the $N=2$ theory reviewed in section \ref{revIIA}. 

3791: The periods $(Z^\Kh,\cF_\Kh)$ defined in \eqref{pre-z}

3792: form a symplectic vector

3793: of $Sp(2h^{(1,2)}+2,\bf Z)$

3794: such that $\Omega$ given in \eqref{Omegaexp} and 

3795: $\Kcs$ given in \eqref{csmetric} is manifestly invariant.

3796: The prepotential  $\cF(Z) = \frac{1}{2} Z^\Kh \cF_\Kh$ on the other hand 

3797: does depend

3798: on the choice of the basis $(\alpha_\Kh,\beta^\Kh)$

3799: and is not invariant.  

3800: 

3801: For $N=1$ orientifolds this situation is different 

3802: since the orientifold projection \eqref{constrO} explicitly breaks the 

3803: symplectic invariance.\footnote{A symplectic transformation $\cS$ preserve the

3804: form $\big<\alpha,\beta\big> = \int \alpha \wedge \beta$, such that 

3805: $\big<\cS \alpha,\cS \beta \big> = \big< \alpha,\beta \big>$.

3806: On the other hand the anti-holomorphic involution satisfies

3807: $\big<\sigma^* \alpha,\sigma^* \beta \big>  

3808: = - \big< \alpha,\beta \big>$.}  

3809: This can also be seen from the form

3810: of the $N=1$ K\"ahler potential \eqref{KQsimple} which is expressed

3811: in terms of the non-invariant prepotential.

3812: One immediately concludes that the result \eqref{KQsimple} is 

3813: basis dependent and $K^Q$ takes this simple form due to the special 

3814: choice $a_\Kh \in H^{3}_+(Y)$ and $b^\Kh \in H^3_-(Y)$.\footnote{Note that this is in striking analogy to 

3815: the background dependence of the B model partition function as discussed in \cite{BCOV,Witten2}.}

3816: On the other hand, the integral representation \eqref{intKQ} only implicitly depends 

3817: on the symplectic basis through the definition of the coordinates $N^\Kh$. 

3818: This suggest, that it is possible to generalize our results by allowing for 

3819: an arbitrary choice of symplectic basis in the definition of the $N=1$ coordinates. 

3820: More precisely, let us consider the generic basis $(\alpha_\Kh,\beta^\Lh)$, 

3821: where we assume that the $h^3_+=h^{2,1}+1$ basis elements $(\alpha_k,\beta^\lambda)$

3822: span $H^3_+$ and the $h^3_-=h^{2,1}+1$ basis elements $(\alpha_\lambda,\beta^k)$ span $H^3_-$.

3823: In this basis the intersections \eqref{int-numbers1} take the form

3824: \beq \label{sp_alpha-beta}

3825:   \int_Y \alpha_k \wedge \beta^l\ =\ \delta_k^l\ , \qquad 

3826:   \int_Y \alpha_\kappa \wedge \beta^\lambda\ =\ \delta_\kappa^\lambda\ ,

3827: \eeq

3828: with all other combinations vanishing.

3829: Applying the orientifold constraint \eqref{constrO} one concludes that 

3830: the equations \eqref{Z=0} are replaced by 

3831: \beq \label{Z=0gen}

3832:   \I(C Z^k) =  \R (C \cF_k )\ =\ 0\ , \qquad 

3833:    \R (C Z^\lambda) = \I(C \cF_\lambda)\ =\ 0\ .

3834: \eeq 

3835: Correspondingly, the expansions \eqref{decompO} and \eqref{form-exp}

3836: take the form

3837: \bea \label{decompO2}

3838:   C \Omega &=& \R (C Z^k) \alpha_k + i\I (C Z^\lambda ) \alpha_\lambda -

3839:              \R (C \cF_\lambda) \beta^\lambda - i\I (C \cF_k) \beta^k\ ,\nn\\

3840: \CC_3 &=& \xi^k\, \alpha_k - \tilde \xi_\lambda\, \beta^\lambda\ ,

3841: \eea

3842: which implies that we also have to redefine the $N=1$ coordinates of 

3843: $\tilde \cM^{\rm Q}$ in an appropriate way. 

3844: In section~\ref{IIA_lin} we show that the 

3845: new K\"ahler coordinates  $(N^k,T_\lambda)$ are again determined by the periods of $\Omegac$ and given  by 

3846: \bea \label{Oexp}\label{def-NT}

3847:   N^k &=&\tfrac{1}{2} \int \Omegac \wedge \beta^k \

3848:  =\ \tfrac{1}{2}\xi^k + i \R(CZ^k)\ , \nn\\

3849:   T_\lambda &=& i \int \Omegac \wedge \alpha_\lambda\ =\

3850: i\tilde \xi_\lambda - 2 \R (C \cF_\lambda) \ ,

3851: \eea

3852: where we evaluated the integrals by using \eqref{Omegacdef} 

3853: and \eqref{decompO2}.

3854: 

3855: The K\"ahler potential takes again the form \eqref{intKQ} but now 

3856: depends on $N^k,T_\lambda$ and thus no longer simplifies to \eqref{KQsimple}.

3857: %A detailed discussion of this calculation is given in section \ref{IIA_lin}. 

3858: Let us compare the situation to the original $N=2$ theory, which

3859: was formulated in terms of the

3860: $Z^\Kh$ or equivalently the special coordinates $z^K$. Holomorphicity

3861: in these coordinates played a central role in defining the prepotential

3862: encoding the special geometry of $\cM^{\rm cs}$ in $\cM^{\rm Q}$ (cf.~section 

3863: \ref{revIIA}). In contrast, the $N=1$ orientifold constraints destroy this complex structure and force us 

3864: to combine $\R(C\Omega)$ with the RR three-form $C_3$ into $\Omegac$. The 

3865: K\"ahler coordinates are half of the periods of $\Omegac$ 

3866: but now in this more general case also the

3867: derivatives of $\cF$ can serve as coordinates as seen in \eqref{def-NT}. 

3868: However, as it is shown in section 

3869: \ref{IIA_lin}, $\R (C \cF_\lambda)$ and $e^{2D}\I (CZ^\lambda)$ are related

3870: by a Legendre transformation of the K\"ahler potential. Working with this transformed

3871: potential and the coordinates $\R(CZ^k)$ and $e^{2D}\I (CZ^\lambda)$ enables us 

3872: to make contact to the underlying $N=2$ theory in its canonical formulation. 

3873: From a supergravity point of 

3874: view, this Legendre transformation corresponds to replacing the chiral multiplets 

3875: $T_\lambda$ by linear multiplets as described in the next chapter.

3876: This is possible due to the translational isometries of $K$, 

3877: which arise as a consequence of the $C_3$ gauge invariance 

3878: and which render $K$ independent of

3879: the scalars $\xi$ and $\tilde\xi$. 

3880: We show in section \ref{geom_of_modspace}

3881: that this also enables us to construct $\tilde \cM^{\rm Q}$ from $\cM^{\rm cs}_\bbR$

3882: similar to the moduli space of supersymmetric Lagrangian submanifolds in a

3883: Calabi-Yau space as described by Hitchin \cite{Hitchin2}.  

3884: This also allows us to interpret the no-scale condition \eqref{no-scale2}

3885: geometrically.

3886: 

3887: Let us summarize the results obtained so far. We found that the moduli

3888: space of $N=1$ orientifolds is indeed the product of two K\"ahler spaces

3889: with the K\"ahler potential 

3890: \beq \label{N=1Kpot}

3891:   K\ =\ K^{\rm K} + K^{\rm Q} = - \ln \Big[\tfrac{4}{3}  \int_Y J \wedge J \wedge J\Big] 

3892:         - 2\ln\Big[2\int_Y \R(C \Omega)\wedge *\R(C\Omega)\Big]\ .

3893: \eeq 

3894: The first term depends on the K\"ahler deformations of the orientifold

3895: while the second term is a function of the real complex structure 

3896: deformations and the dilaton.

3897: The $N=1$ K\"ahler coordinates are obtained 

3898: by expanding the complex combinations\footnote{This combination 

3899: of forms has also appeared recently in ref.\ \cite{NOV}

3900: in the discussion of $D$-instanton couplings in the A-model.

3901: Here they appear as the proper chiral $N=1$ variables and as we will

3902: see in the next section they linearize the D-instanton action.}

3903: \beq \label{N=1coords}

3904:   \Omegac\ =\ \CC_3 + 2i \R(C\Omega)\ ,\qquad 

3905:   \Jc\ =\ \hat B_2 + iJ \ , 

3906: \eeq

3907: in a real harmonic basis of $H^{3}_+(Y)$ and $H^{(1,1)}_-(Y)$ respectively. 

3908: Note that $K$ does not depend on the scalars arising in the expansion of 

3909: $\hat B_2$ and $\hat C_3$, such that the K\"ahler manifold admits a set of 

3910: $h^{(1,1)}_- + h^{(2,1)}+1$ translational isometries. In other words

3911: $K$ consists of two functionals encoding the dynamics of the two-form $J$ 

3912: and the real three-form $\R(C\Omega)$.

3913: %\footnote{The functions 

3914: %$V [ \R(C\Omega) ]= \int \R(C\Omega)\wedge* \R(C\Omega)$

3915: %and $V [J ]=\int J\wedge J \wedge J$ are known as Hitchins functionals \cite{Hitchin1}. 

3916: %The orientifold constraints \eqref{constrJ} and \eqref{constrO} 

3917: %restricts their domain to $J \in H^2_-(Y)$

3918: %and $\R(C\Omega) \in H^3_+(Y)$.}   

3919: In type IIA orientifolds it is not difficult to rewrite $K^{\rm Q}$ in a form 

3920: similar to \eqref{def-Phi}. Defining the odd form 

3921: \beq \label{def-podd}

3922:  \podd=\fu +i\, \fuh = C\Omega\ ,

3923: \eeq

3924: one finds 

3925: \beq \label{symp-formodd}

3926:   K^{\rm Q} = - 2 \ln \Phi_A(\fu) \ ,\qquad   \Phi_A(\fu) \equiv i\big<\podd,\poddb \big>=i\int_Y \podd \wedge \poddb\ .

3927: \eeq

3928: The function $\Phi_A(\fu)$ is 

3929: known as Hitchins functional for the real 

3930: three-form $\fu$ \cite{Hitchin1,HitchinGCM}. The orientifold constraint \eqref{constrO} 

3931: restricts its domain to $\fu \in H^3_+(Y)$. Applying the fact that

3932: $\Phi_A(\fu)$ is a homogeneous function of degree two $K^{\rm Q}$ obeys the no-scale type conditions

3933: \eqref{no-scale2}, \eqref{no-scale4}. This is independent of the chosen basis 

3934: and can be also shown directly as done in section \ref{IIA_lin}.

3935: %Moreover, irrespective of the chosen basis the K\"ahler potential 

3936: %obeys the no-scale type conditions \eqref{no-scale1} and 

3937: 

3938: The no-scale conditions are violated when further stringy 

3939: corrections are included. $K$ receives additional contributions due

3940: to perturbative effects as well as world-sheet and $D2$ instantons.

3941: It is well-known that the combination $\Jc=\hat B_2 + i J$ 

3942: gives the proper coupling to the string world-sheet such that 

3943:  world-sheet instantons correct the holomorphic prepotential as 

3944: $f(t) = -\frac{1}{6}\cK_{abc}t^a t^b t^c + O(e^{-t})$. 

3945: Since we divided out the world-sheet parity these corrections also

3946: include non-orientable Riemann surfaces, such that the prepotential 

3947: $f(t)$ consists of two parts $f(t) = f_{or}(t) + f_{unor}(t)$. 

3948: The function $f_{or}$ counts holomorphic maps 

3949: from orientable world-sheets to $Y$, while $f_{unor}$ counts holomorphic maps 

3950: from non-orientable world-sheets to $Y$ \cite{BFM}. 

3951: In the next section we show that $D2$ instantons naturally couple to the complex three-form  

3952: $\Omegac$ and they are expected to correct 

3953: $K^{\rm Q}$.  

3954: 

3955: 

3956: 

3957: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3958: %

3959: %  Mirror symmetry

3960: %

3961: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3962: 

3963: \section{Mirror symmetry \label{Mirror_orientioflds}}

3964: 

3965: In this section we discuss mirror symmetry 

3966: for Calabi-Yau orientifolds from the point of view of the effective

3967: action derived in the large volume limit. More precisely, we compare the $N=1$

3968: data for type IIB orientifolds on $\tilde Y/\sigma_B$ (section \ref{Kpo_gaugeIIB}) 

3969: with the data for type IIA orientifolds on $Y/\sigma_A$ (section \ref{Kpo_gaugeIIA}). 

3970: Since we want to discuss mirror symmetry we choose $\tilde Y$ 

3971: to be the mirror manifold of $Y$. This implies that the

3972: non-trivial Hodge numbers $h^{(1,1)}$ and $h^{(2,1)}$ of $Y$ and $\tilde Y$ 

3973: satisfy $h^{(1,1)}(Y)=h^{(2,1)}(\tilde Y)$  and $h^{(2,1)}(Y)=h^{(1,1)}(\tilde Y)$ as already 

3974: given in section \ref{revMirror} where we briefly introduced $N=2$ mirror symmetry. 

3975: In orientifolds we also have to specify the 

3976: involutions $\sigma_A$ and $\sigma_B$ which are identified under mirror symmetry. Since the

3977: discussion

3978: in this article is quite generic and never specified any involution

3979: $\sigma$ explicitly we also keep the discussion of mirror symmetry

3980: generic. That is we assume that there exists a mirror pair

3981: of manifolds $Y$ and $\tilde Y$ with a mirror pair of involutions

3982: $\sigma_A, \sigma_B$. 

3983: Matching the number of $N=1$ multiplets summarized in table \ref{numberM} 

3984: implies an orientifold version of

3985: \eqref{Hod_id},\footnote{For the sector of $\tilde \cM^{\rm Q}$ mirror

3986:   symmetry

3987: is a constraint on the couplings rather than the Hodge

3988: numbers.}  i.e.\

3989: \bea \label{matchchohm}

3990:    O3/O7&: &\quad h^{1,1}_-(Y) = h^{2,1}_-(\tilde Y) \ , \qquad  h^{1,1}_+(Y) = h^{2,1}_+(\tilde Y) \ ,\nn\\

3991:    O5/O9&: &\quad h^{1,1}_-(Y) = h^{2,1}_+(\tilde Y)   \ , \qquad h^{1,1}_+(Y) = h^{2,1}_-(\tilde Y) \ .

3992: \eea

3993: \begin{table}[h] 

3994: \begin{center}

3995: \begin{tabular}{|l|c|c|c|} \hline 

3996:  \rule[-0.3cm]{0cm}{0.8cm} 

3997: multiplets& IIA$_Y$ \  $O6$ & IIB$_{\tilde Y}$ \  $O3/O7$ & IIB$_{\tilde Y}$ \  $O5/O9$ \\ \hline\hline

3998:  \rule[-0.3cm]{0cm}{0.9cm} 

3999:  {vector multiplets} &   $h_+^{(1,1)}$ & $h_+^{(2,1)}$ & $h_-^{(2,1)}$ \\ \hline

4000: \rule[-0.3cm]{0cm}{0.9cm} 

4001:  chiral multiplets in $\tilde \cM^{\rm SK}$& $h_-^{(1,1)}$ & $h_-^{(2,1)}$ & 

4002:                       $h_+^{(2,1)} $   \\ \hline

4003:  \rule[-0.3cm]{0cm}{0.9cm} 

4004:  chiral multiplets in $\tilde \cM^{\rm Q}$&$h^{(2,1)} + 1$&$h^{(1,1)} + 1$ & 

4005:                       $h^{(1,1)} + 1$   \\ \hline

4006: \end{tabular} 

4007: \caption{ \textit{Number of $N=1$ multiplets of orientifold compactifications.}}\label{numberM}

4008: \end{center}

4009: \end{table} 

4010: 

4011: 

4012: 

4013: 

4014: 

4015: 

4016: Our next task will be to match the couplings of the mirror theories.

4017: Since the effective actions on both sides 

4018:  are only computed in the large volume limit

4019: we can expect to find agreement only if we also take 

4020: the large complex structure limit exactly as in the $N=2$ mirror

4021: symmetry.

4022: However, if one believes in mirror symmetry one can use the 

4023: the geometrical results of the complex structure moduli space to

4024: `predict' the corrections to its mirror symmetric component.

4025: This is not quite as straightforward since the full $N=1$ moduli space is a

4026: lot more complicated than the underlying $N=2$ space \cite{BH}.

4027: Let us therefore start our analysis with the simpler situation of the 

4028: special K\"ahler sectors $\tilde \cM^{\rm SK}_A,\, \tilde \cM_B^{\rm SK}$ in \eqref{direct-mod} 

4029: and \eqref{modulispaceO3} and the vector multiplet couplings 

4030: and postpone the analysis of $\tilde M^{\rm Q}_{A,B}$ 

4031: to section \ref{O3O7mirror}.   

4032: 

4033: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4034: \subsection{Mirror symmetry  in $ \mathcal{M}^{\rm K}$}  \label{mirrorMK}

4035: %%%

4036: Recall that the manifold $\tilde \cM^{\rm SK}_A$ is spanned by the 

4037: complexified K\"ahler deformations $t^a$ preserving the constraint 

4038: \eqref{constrJ}.  Under mirror symmetry these moduli are mapped 

4039: to the complex structure deformations which respect the constraint

4040: \eqref{Omegatransf}.

4041: In both cases the K\"ahler potential is merely a truncated

4042: version of the $N=2$ K\"ahler potential and one has

4043: \beq

4044:   K^{\rm K}_A \ =\ - \ln \Big[\tfrac{4}{3}  \int_Y J \wedge J \wedge J\Big]

4045:   \quad  \leftrightarrow \quad 

4046:   K^{\rm cs}_B\ =\ -  \ln \Big[-i \int \Omega \wedge \bar \Omega \Big]\ .

4047: \eeq

4048: Both K\"ahler potentials can be expressed in terms of prepotentials

4049: $f_A(t), f_B(z)$ and in the large complex structure limit

4050: $f_B(z)$ becomes cubic and agrees with $f_A(t)$.

4051: Mirror symmetry therefore equates these prepotentials 

4052: and exchanges $J^3$ with $\Omega\wedge\bar\Omega$

4053: exactly as in $N=2$ 

4054: \beq\label{mirrorK}

4055: f_A(t) =  f_B(z) \ , \qquad J^3 \leftrightarrow

4056: \Omega\wedge\bar\Omega\ .

4057: \eeq

4058: In \cite{FMM} the $N=2$ version of this map was written into the form \footnote{The authors argued that

4059: this should be true also for mirror symmetry of certain non-Calabi-Yau backgrounds. }

4060: \beq \label{pure-spinor-map}

4061:   e^{J_c}\ \leftrightarrow\ \Omega\ ,

4062: \eeq 

4063: where $J_c$ is given in \eqref{N=1coords}.  

4064: Thus for $\tilde \cM^{\rm SK}$ mirror symmetry is a truncated

4065: version of $N=2$ mirror symmetry. As we will see momentarily this also

4066: holds for the gauge kinetic couplings 

4067: which depend holomorphically on the moduli spanning $\tilde \cM^{\rm SK}$.

4068: 

4069: In type IIA the gauge-kinetic couplings

4070: are given in \eqref{gauge-A} and read

4071: $f_{\alpha \beta}(t) = i\cK_{\alpha\beta c}  t^c$.

4072: The IIB couplings were determined in \eqref{fholo} to be 

4073: \bea \label{gauge-B}

4074:  f_{\alpha\beta}(z^a) = - {i} \bar \cM_{\alpha\beta}

4075:    = - i\cF_{\alpha\beta}\ ,

4076: \eea

4077: where in order to not overload the notation we are using the same indices

4078: for both cases.\footnote{We rescaled the type IIB gauge bosons

4079: by $\sqrt 2$ in order to properly match the normalizations.} 

4080: More precisely we are choosing

4081: \bea

4082: \alpha, \beta = 1, \ldots, h^{(2,1)}_+(\tilde Y)\ ,\qquad 

4083: a, b = 1, \ldots, h^{(2,1)}_-(\tilde Y)\ , \qquad \textrm{for} \quad O3/O7\ ,\nn\\ 

4084: \alpha, \beta = 1, \ldots, h^{(2,1)}_-(\tilde Y)\ ,\qquad 

4085: a, b = 1, \ldots, h^{(2,1)}_+(\tilde Y)\ ,\qquad \textrm{for} \quad

4086: O5/O9\ . 

4087: \eea

4088: The matrix $\cF_{\alpha\beta}(z^a)$ is  

4089: holomorphic and the second derivatives of the prepotential restricted

4090: to $\tilde \cM^{\rm  K}_B$. In the large complex structure limit 

4091: $\cF_{\alpha\beta}$ is linear

4092: in $z^a$ and therefore also agrees with the type IIA mirror

4093: couplings. 

4094: Thus mirror symmetry implies the map $\cN_{\alpha \beta}(\bar t^a) = \cM_{\alpha\beta}(\bar z^a)$

4095: in both cases.

4096: 

4097: This concludes our discussions of mirror symmetry

4098: for the chiral multiplets which span $\tilde\cM^{\rm SK}$.

4099: We have shown that 

4100: the K\"ahler potential and 

4101: the gauge-kinetic coupling functions

4102: agree in the large complex structure limit under mirror symmetry.

4103: In this sector the geometrical quantities on the type IIB side include

4104: corrections which are believed to 

4105: compute world-sheet non-perturbative effects  

4106: such as world-sheet instantons on the type

4107: IIA side. 

4108: This is analogous to the situation

4109: in $N=2$ and  may be traced back to the 

4110: fact, that it is still possible to formulate a topological 

4111: A model counting 

4112: world-sheet instantons for Calabi-Yau orientifolds \cite{AAHV,BFM}.

4113: 

4114: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4115: \subsection{Mirror symmetry in $ \cM^{\rm Q}$}

4116: %%%

4117: Let us now turn to the discussion of the K\"ahler manifolds $\tilde \cM^{\rm Q}_{A}$ and

4118: $\tilde \cM^{\rm Q}_{B}$ arising in 

4119: the reduction of the quaternionic spaces. 

4120: %As we already remarked above, the metrics on the

4121: %type IIB K\"ahler manifolds $\tilde \cM^{\rm Q}_B$ (for both projections in \eqref{constrOB}) 

4122: %are given in terms of `special coordinates'. 

4123: On the IIA side the K\"ahler potential is given in \eqref{N=1Kpot}

4124: which is expressed in terms of the $h^{(2,1)}+1$ coordinates

4125: $(N^k,T_\lambda)$ defined in \eqref{Oexp}.

4126: In this definition we did not fix the scale invariance \eqref{real_K}

4127: $\Omega\to

4128: \Omega e^{-\R (h)}$ or in other

4129: words we defined the coordinates in terms of the scale invariant

4130: combination $C\Omega$. Somewhat surprisingly there seem to be two 

4131: physically inequivalent ways to fix this scale invariance.

4132: In $N=2$ one uses the scale invariance to define special 

4133: coordinates $z^K = Z^K/Z^0, z^0 = 1$ where  $Z^0$ is the coefficient

4134: in front of the base element $\alpha_0$. The choice of $Z^0$

4135: is convention and 

4136: due to the symplectic invariance any other choice would be

4137: equally good. 

4138: However, as we already discussed in section 3.1 and 3.3 the 

4139: constraint \eqref{constrO} breaks the symplectic invariance and  

4140: $H^3$ decomposes into two eigenspaces $H^3_+\oplus H^3_-$.

4141: Thus in \eqref{decompO2} we have the choice to scale one of the $Z^k$ 

4142: equal to one or 

4143: one of the $Z^\lambda$ equal to $i$.

4144: Denoting the corresponding basis element by $\alpha_0$, 

4145: these two choices are characterized by 

4146: $\alpha_0 \in H^{3}_+$ or $\alpha_0 \in H^{3}_-$.

4147: This choice identifies the dilaton direction inside the moduli space

4148: and therefore is crucial in identifying the type IIB

4149: mirror. This is related to the fact that in type IIB

4150: the dilaton reside in a chiral multiplet for $O3/O7$ orientifolds and in a

4151: linear multiplet for $O5/O9$ orientifolds as we make more explicit in section 

4152: \ref{IIB_lin}. Let us discuss these two cases in turn.

4153: 

4154: 

4155: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4156: %

4157: %  The Mirror of O3/O7

4158: %

4159: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4160: 

4161: \subsubsection{The Mirror of IIB orientifolds with $O3/O7$ planes}

4162: \label{O3O7mirror}

4163: 

4164: We first want to show that in the large complex structure limit

4165: $K^Q_A$ given in \eqref{intKQ} coincides with

4166: $K^{\rm Q}_B$ given in \eqref{kaehlerpot-Kk} for 

4167: orientifolds with $O3/O7$ planes.

4168: It turns out that in order to do so we need to choose

4169: $\alpha_0 \in H^3_{+}$ and the dual basis element

4170: $\beta^0\in H^3_{-}$.

4171: It is convenient to keep track of this choice and therefore

4172: we mark the $\alpha$'s and $\beta$'s which contain $\alpha_0$

4173: and $\beta^0$ by putting a hat on the corresponding index. 

4174: Thus we work in the basis $(\alpha_\kh,\beta^\lambda)$ 

4175: of $H^3_+$ and $(\alpha_\lambda,\beta^\kh)$

4176: of $H^{3}_-$. Therefore, we rewrite the combination $C\Omega$ as 

4177: \beq

4178:   C\Omega = g_A^{-1}(\textbf{1}\, \alpha_0 + q^k \alpha_k + iq^\lambda \alpha_\lambda) + \ldots\ ,

4179: \eeq

4180: where we introduced $g_A$ and the real special coordinates 

4181: \beq \label{realspC1}

4182:   g_A =\frac{1}{\R(CZ^0)}\ ,\qquad q^k = \frac{\R(CZ^k)}{\R(CZ^0)}\ , \qquad q^\lambda = \frac{\I(CZ^\lambda)}{\R(CZ^0)}\ .

4183: \eeq

4184: We also need to express the prepotential $\cF(Z)$ 

4185: in the special coordinates $q^k,q^\lambda$.

4186: In analogy to \eqref{def-f} one defines a function $f(q)$ 

4187:  such that 

4188: \beq \label{def-h(q)}

4189:   \cF\big(\R[CZ^\kh],i\I[CZ^\lambda] \big)\ =\ i\big(\R[ CZ^0]\big)^2\  f(q^k,q^\lambda) \ . 

4190: \eeq

4191: We are now in the position to rewrite the $N=1$ coordinates 

4192: $N^\kh,T_\lambda$ given in 

4193: \eqref{def-NT} in terms of $g_A$ and the special coordinates $q^K$. 

4194: Inserting \eqref{realspC1} 

4195: into \eqref{def-NT} one obtains

4196: \beq \label{c-in-q37}

4197:   N^0\ =\ \tfrac{1}{2} \xi^0 + i g_A^{-1}\ , \qquad

4198:   N^k\ =\ \tfrac{1}{2} \xi^k + i g_A^{-1} q^k \ , \qquad

4199:   T_\lambda\ =\ i \tilde \xi_\lambda - 2 g_A^{-1} f_\lambda(q)\ ,

4200: \eeq

4201: where $f_\lambda$ is the first derivative of $f(q)$ with respect to $q^\lambda$. 

4202: 

4203: The final step is to specify $f(q)$ in the large complex structure

4204: limit. 

4205: In this limit the $N=2$ prepotential is known to be

4206: \beq \label{N=2pre}

4207:  \cF(Z) = \tfrac{1}{6} (Z^0)^{-1}{\kappa_{KLM} Z^K Z^L Z^M}\ .

4208: \eeq

4209: Inserted into the orientifold constraints

4210: \eqref{Z=0gen} one infers

4211: \beq \label{vankappa37}

4212:   \kappa_{klm} = \kappa_{\kappa \lambda l} = 0 \ , 

4213: \eeq 

4214: while $\kappa_{\kappa \lambda \mu}$ and $\kappa_{\kappa l m}$ can be non-zero.

4215: Using \eqref{vankappa37}, \eqref{def-h(q)} and \eqref{realspC1} 

4216: we arrive at

4217: \beq\label{fori37}

4218:   f(q)\ =\ - \tfrac{1}{6} \kappa_{\kappa \lambda \mu} q^\kappa q^\lambda q^\rho 

4219:            + \tfrac{1}{2} \kappa_{\kappa kl} q^\kappa q^k q^l\ .

4220: \eeq

4221: 

4222: In order to continue 

4223: we also have to specify the range the indices $k$ and $\lambda$ 

4224: take on the IIA side.

4225: A priori it is not fixed and can be changed by a symplectic transformation.

4226: Mirror symmetry demands 

4227: \beq \label{na-nb}

4228:  k = 1,\ldots, h^{(1,1)}_-(\tilde Y)\ , \qquad  

4229: \lambda = 1,\ldots,h^{(1,1)}_+(\tilde Y)\ ,

4230: \eeq

4231: or in other words there have to be $h^{(1,1)}_-(\tilde Y)$ 

4232: basis elements $\alpha_k$ and $h^{(1,1)}_+(\tilde Y)$ basis elements

4233: $\beta^\lambda$ in $H^3_+(Y)$. In addition the 

4234: non-vanishing couplings $\kappa_{\kappa \lambda \mu}$ and 

4235: $\kappa_{\kappa l m}$

4236: have to be identified with 

4237: $\cK_{\kappa \lambda \mu}$ and $\cK_{\kappa l m}$ appearing

4238: in the definition of the type IIB chiral coordinates \eqref{tau}.

4239: With these conditions fulfilled

4240: we can insert \eqref{fori37} into \eqref{c-in-q37} and compare with

4241: \eqref{tau}. This leads to the identification

4242: \beq

4243:  N^{\kh} = (\tau, G^k) \qquad \textrm{and}\qquad

4244:  T_{\lambda}^A = 2 T_{\lambda}^B\ ,

4245: \eeq

4246: which in terms of the Kaluza-Klein variables corresponds to 

4247: \bea\label{phi=g}

4248:  e^{\phi_B}&=& g_A \ ,\qquad  q^\lambda\ =\ v^\lambda\ ,\qquad  q^k\ =\ -b^k\ ,\nn\\

4249:   \xi_0 &=& 2 C_0\ , \quad \xi^k=2(c^k-C_0 b^k)\ , \\

4250:   \tilde \xi_\lambda &=& 2 \rho_\lambda - 2\cK_{\lambda kl}c^k b^l +

4251: C_0 \cK_{\lambda kl}b^k b^l\ .\nn

4252: \eea

4253: With these identifications one immediately shows 

4254: $e^{D_A} = e^{D_B}$, where $e^{D_A}$ and

4255: $e^{D_B}$ are the four-dimensional dilatons of the type IIA and IIB theory.

4256: This implies that the K\"ahler potentials \eqref{intKQ} and \eqref{kaehlerpot-Kk} 

4257: of the two theories coincide in the large volume -- 

4258: large complex structure limit. However, the corrections

4259: away from this limit cannot be properly understood 

4260: from a pure supergravity analysis. It is clear that 

4261: $K^{\rm Q}_A$ includes corrections of the mirror IIB

4262: theory but the precise nature of these corrections remains to be understood.

4263: 

4264: 

4265: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4266: %

4267: %  The Mirror of O5/O9

4268: %

4269: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4270: 

4271: \subsubsection{The Mirror of IIB orientifolds with $O5/O9$ planes}

4272: \label{O5O9mirror}

4273: 

4274: In this section we check mirror symmetry for type IIB orientifolds with 

4275: $O5/O9$ planes with complex coordinates and K\"ahler potential determined 

4276: in section \ref{Kpo_gaugeIIB}.

4277: In order to find the same chiral data on the IIA side, we have to examine the 

4278: case where $\alpha_0 \in H^3_{-}$. Therefore we choose a basis 

4279: $(\alpha_k,\beta^{\hat \lambda})$ of $H^3_+$ and $(\alpha_{\hat \lambda},\beta^k)$

4280: of $H^{3}_-$. We rewrite the combination $C\Omega$ in this basis as 

4281: \beq

4282:   C\Omega = g_A^{-1}(i\, \alpha_0 + i q^\lambda \alpha_\lambda  + q^k \alpha_k) + \ldots

4283: \eeq

4284: where we introduced the real special coordinates 

4285: \beq \label{realspC2}

4286:   g_A =\frac{1}{\I(CZ^0)}\ ,\qquad q^k = \frac{\R(CZ^k)}{\I(CZ^0)}\ , \qquad q^\lambda = \frac{\I(CZ^\lambda)}{\I(CZ^0)}\ .

4287: \eeq

4288: Let us also express the prepotential $\cF(Z)$ in terms of $q^k,q^\lambda$. As in $N=2$ one defines a 

4289: function $f(q)$ such that 

4290: \beq \label{def-h59}

4291:   \cF\big(\R[CZ^k],i\I[CZ^{\hat\lambda}] \big) =- i\big(\I[ CZ^0]\big)^2\,  f(q^k,q^\lambda) \ . 

4292: \eeq

4293: We can now rewrite the $N=1$ coordinates $T_{\hat\lambda}, N^k$ 

4294: given in \eqref{def-NT} in terms of 

4295: $q^k,q^\lambda$ and $g_A$ as

4296: \bea \label{c-in-q59}

4297:   N^k &=& \tfrac{1}{2} \xi^k + i g^{-1}_A q^k \ , \qquad T_\lambda = i \tilde \xi_\lambda +2 g^{-1}_A f_\lambda(q)\ , \nn\\

4298:   T_0 &=& i \tilde \xi_0 + 2 g^{-1}_A (2f(q)- f_\lambda q^\lambda - f_k q^k)\ , 

4299: \eea

4300: where $f_\lambda,f_k$ are the first derivatives of $f(q)$ with respect to $q^\lambda$ and $q^k$. 

4301: 

4302: Going to the large complex structure limit, the $N=2$ prepotential takes the form 

4303: \eqref{N=2pre}. We split the indices as $K=(k,\hat \lambda)$ and apply the constraints 

4304: \eqref{Z=0gen} to find that

4305: \beq \label{vankappa59}

4306:   \kappa_{\kappa \lambda \mu} = \kappa_{\kappa k l} = 0 \qquad \kappa_{klm} \neq 0\ ,\qquad 

4307:   \kappa_{\kappa \lambda l} \neq 0\ .

4308: \eeq 

4309: Using \eqref{vankappa59} and \eqref{def-h59} we can calculate $f(q)$ as 

4310: \beq

4311:   f(q) = \tfrac{1}{6} \kappa_{ k l m} q^k q^l q^m - \tfrac{1}{2} \kappa_{\kappa \lambda k} q^\kappa q^\lambda q^k\ .

4312: \eeq

4313: In order to match the chiral coordinates $T_0,T_\lambda,N^k$ 

4314: with the type IIB coordinates 

4315: of \eqref{Kcoord} we need again to specify the range of the indices

4316: on the type IIA side. Obviously we need

4317: \beq \label{na-nb59}

4318: k=1, \ldots, h^{(1,1)}_+(\tilde Y)\ , \qquad  \lambda= 1,\ldots,  h^{(1,1)}_-(\tilde Y)\ ,

4319: \eeq

4320: which is the equivalent of \eqref{na-nb} with the plus and minus sign interchanged. 

4321: Thus the non-vanishing intersections can be identified with 

4322: $\cK_{klm}$ and $\cK_{\kappa\lambda k}$ on the IIB side.

4323: Inserting $f(q)$ back into the equations \eqref{c-in-q59} for the chiral 

4324: coordinates $N^k,T_{\hat \lambda}$ and demanding \eqref{na-nb59} one can 

4325: compare these to the type IIB coordinates \eqref{Kcoord}. 

4326: One identifies 

4327: \beq

4328:  T_{\hat \lambda} = 2(S,A_\lambda)\ ,\qquad  N^{k} = it^k \ .

4329:  \eeq

4330: In terms of the Kaluza-Klein modes this amounts to the identification

4331: \bea

4332:  g_A &=& e^{\phi_B}\ , \qquad q^k = -v^k\ , \qquad 

4333: q^\lambda = b^\lambda\ ,\qquad 

4334:   \xi^k = -2 c^k\ , \nn \\ 

4335:   \tilde \xi_\lambda &=& 2\rho_\lambda - 2 \cK_{\lambda \kappa l} c^l b^\kappa\ , \qquad 

4336:   \tilde \xi_0 = 2h + \cK_{l\lambda \kappa} c^l b^\lambda b^\kappa - \rho_\lambda b^\lambda\ .

4337: \eea  

4338: With these identifications one shows again $e^{D_A} = e^{D_B}$ and as

4339: a consequence the  K\"ahler potentials \eqref{intKQ} and \eqref{kaehlerpot-KkO5} agree

4340: in the large volume -- large complex structure limit.

4341: 

4342: 

4343: In summary, we found that it is indeed possible to obtain both type IIB 

4344: setups as mirrors of the type IIA orientifolds. 

4345: In analogy to \eqref{pure-spinor-map}

4346: we found by comparing \eqref{N=1coords} with \eqref{def-coordsO3} and \eqref{def-coordsO5} the mirror relation

4347: \bea\label{pure-spinor-map2}

4348: O3/O7: & \quad \podd\ \leftrightarrow\ \pev\ , &\qquad C_3 \leftrightarrow \fa\ ,

4349: \nn \\

4350: O5/O9: & \quad \podd\ \leftrightarrow\ -i \pev \ ,& \qquad C_3 \leftrightarrow \fa\ ,

4351: \eea

4352: where $\podd,\pev$ and $\fa$ are defined in \eqref{def-podd} and \eqref{def-A}. 

4353: Furthermore, we found that the functionals $\Phi_A$ and $\Phi_B$ have to identified

4354: as 

4355: \beq \label{mirror-hitchin}

4356:   O3/O7: \quad \Phi_A(\fu)  \leftrightarrow \ \Phi_B(\fe)\ , \qquad  

4357:   O5/O9: \quad \Phi_B(\fu)  \leftrightarrow \ \Phi_B(\feh)\ ,

4358: \eeq 

4359: such that the K\"ahler potentials are matched. However, the crucial role of the two 

4360: definitions of special coordinates remains to be understood further. 

4361: 

4362: Let us close this chapter with a brief remark on the generalizations of this result. 

4363: Formulated in this abstract fashion equations \eqref{pure-spinor-map2} and \eqref{mirror-hitchin}

4364: are expected to hold even for orientifolds of generalized complex manifolds. This includes certain 

4365: $SU(3)$ structure manifolds, such as half-flat manifolds. This looks very promising and 

4366: deserves further investigation \cite{GLprep}.  

4367: 

4368: 

4369: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4370: %

4371: %  Linear multiplets and the geometry of the moduli space

4372: %

4373: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4374: 

4375: 

4376: \chapter{Linear multiplets and the geometry of the moduli space}

4377: \label{lin_geom_of_M}

4378: 

4379: In this chapter we explore the geometry of the $N=1$ moduli 

4380: space in more detail. Our attempt is to get some deeper understanding

4381: of the properties of the K\"ahler manifolds obtained from the 

4382: $N=2$ to $N=1$ reduction performed in the previous chapter. Recall 

4383: that the orientifold moduli space is a direct product 

4384: \beq \label{mod-spaceN=1}

4385:   \tilde \cM^{\rm SK} \times \tilde \cM^{\rm Q}\ ,

4386: \eeq

4387: where $N=1$ supersymmetry demands each factor to be a K\"ahler manifold.

4388: $\tilde \cM^{\rm SK}$ is a submanifold of the $N=2$ special K\"ahler 

4389: manifold $\cM^{\rm SK}$ parameterizing complex structure deformations 

4390: in type IIB and complexified K\"ahler structure deformations in type IIA.  

4391: As we have shown also $\tilde \cM^{\rm SK}$ is special K\"ahler, since  

4392: it inherits its complex structure from $\cM^{\rm SK}$ and admits

4393: a K\"ahler metric obtained from a prepotential. 

4394: 

4395: The reduction of the hypermultiplet

4396: sector is more `radical' since it defines a K\"ahler manifold $\tilde \cM^{\rm Q}$

4397: inside of a quaternionic manifold $\cM^{\rm Q}$, which itself is not necessarily K\"ahler. 

4398: This K\"ahler submanifold has half the dimension of the quaternionic space.

4399: In general it is a difficult 

4400: mathematical problem to characterize K\"ahler manifolds inside quaternionic

4401: ones \cite{AM}. However, the quaternionic manifolds obtained by Calabi-Yau compactifications

4402: of type IIA or type IIB supergravity posses special properties. As shown 

4403: in \cite{CFGi,FS} they can be constructed from special K\"ahler manifold $\cM^{\text{SK}}$ via the 

4404: local c-map,

4405: \beq \label{c-map}

4406:   \cM^{\rm SK}_{2n} \quad \xrightarrow{\text{c-map}}\quad  \cM^{\rm Q}_{4n+4}\ ,

4407: \eeq

4408: where $2n$ and $4n+4$ are the real dimensions of $\cM^{\rm SK}$ and $\cM^{\rm Q}$.  

4409: These quaternionic manifolds are termed special or dual quaternionic.

4410: One observes that their metric depends on only half of the bosonic fields in the 

4411: hypermultiplets, or, in other words, on half of the quaternionic coordinates.

4412: More precisely, the components of the metrics \eqref{q-metr} and \eqref{q-metrB}

4413: on $\cM^{\rm Q}$ are functions of only NS-NS scalar fields $M^I_{\text{NS}}$. 

4414: The second half are R-R scalar fields  denoted by $M_{I\, \text{RR}}$ which appear in the 

4415: quaternionic metrics only as a differential and hence posses Peccei-Quinn shift symmetries

4416: \beq

4417:    M_{I\, \text{RR}} \rightarrow M_{I\, \text{RR}} + c_I\ ,

4418: \eeq

4419: for arbitrary constants $c_I$. 

4420: 

4421: The orientifold projection truncates half of the NS-NS fields and half of

4422: the R-R fields. $N=1$ supersymmetry forces the remaining fields to span a K\"ahler 

4423: manifold $\tilde \cM^{\rm Q}$. Furthermore, it can be seen in tables \ref{N=1spectrumtab} 

4424: and \ref{N=1spectrumA} 

4425: that supersymmetry combines each NS-NS field $M^I_{\text{NS}}$ together with a R-R field 

4426: $M_{I\, \text{RR}}$ into a chiral multiplet with bosonic 

4427: components $M^I = (M^I_{\text{NS}}, M_{I\, \text{RR}})$ spanning $\tilde \cM^{\rm Q}$.

4428: The fact, that the R-R fields posses shift symmetries allows us to 

4429: chose a set $M_{\alpha \, \text{RR}}$ and dualize them into two-tensors 

4430: $D^{\alpha}_{2\, \text{RR}}$. This amounts to replacing 

4431: the chiral multiplets $M^\alpha$ by linear multiples 

4432: $L^\alpha=(M^\alpha_{\text{NS}},D^{\alpha}_{2\, \text{RR}})$, while keeping the

4433: remaining fields $M^a$ chiral. The manifold $\tilde \cM^{\rm Q}_{L^\alpha}$

4434: spanned by the real scalars $M^\alpha_{\text{NS}}$ and the complex scalars $M^a$

4435: still contains all the information about the full K\"ahler space $\tilde \cM^{\rm Q}$. In 

4436: that one can construct $\tilde \cM^{\rm Q}$ starting from $\tilde \cM^{\rm Q}_{L^\alpha}$,

4437: \beq \label{dual-map}

4438:    \tilde \cM^{\rm Q}_{L^\alpha} \quad \xrightarrow{\ \text{dualization of } D^\alpha_2\ }\quad \tilde \cM^{\rm Q}\ .

4439: \eeq

4440: This dualization procedure will be discussed in section \ref{linear_multiplets}. 

4441: As we will explain there, the kinetic terms and couplings of the chiral and linear multiplets 

4442: can be encoded by a single function $\tilde K$ being the Legendre transform of the K\"ahler potential. 

4443: As an application we determine $\tilde K$ for all three orientifold setups. Firstly, in 

4444: section \ref{IIB_lin} we apply the linear multiplet formalism to IIB orientifolds. Secondly, 

4445: in section \ref{IIA_lin} we provide the missing calculation of the K\"ahler potential for $\tilde \cM^{\rm Q}$ for

4446: general IIA orientifolds. In this derivation we apply the techniques connected with 

4447: the map \eqref{dual-map}. 

4448: 

4449: 

4450: Finally, recall that the quaternionic space can be obtained from $\cM^{\rm SK}$ 

4451: via the local c-map construction \eqref{c-map}. In section \ref{geom_of_modspace}

4452: we construct the map 

4453: \beq \label{N=1c-map}

4454:    \cM^{\rm SK} \cap \tilde \cM^{\rm Q} \quad \xrightarrow{N=1 \text{ c-map} }\quad \tilde \cM^{\rm Q}\ ,

4455: \eeq

4456: which can be interpreted as the $N=1$ analog of the local c-map \eqref{c-map}. 

4457: As we will show it is closely related to the dualization

4458: in \eqref{dual-map}, when specifying the right chiral fields $M^\alpha$ for dualization.

4459: This construction is inspired by the one presented in \cite{Hitchin2}, where 

4460: the moduli space of Lagrangian submanifolds with $U(1)$ connection is discussed.

4461: Furthermore, it provides the basis to extend the analysis to non-Calabi-Yau orientifolds.

4462: 

4463:  

4464: 

4465: 

4466: \section{Linear multiplets and Calabi-Yau orientifolds\label{linear_multiplets}}

4467: 

4468: In this section we rewrite the bulk effective action of type IIB and type IIA orientifolds  

4469: using the linear multiplet formalism of ref.\ \cite{BGG}. 

4470: In this way we will be able to understand the definition of the K\"ahler

4471: coordinates given in \eqref{tau}, \eqref{Kcoord} and \eqref{def-NT} as a superfield duality transformation

4472: and furthermore discover the no-scale properties of $K^{\rm Q}$ 

4473: somewhat more conceptually. In an analog three-dimensional situation this has 

4474: also been observed in \cite{BHS}.

4475: 

4476: Let us first briefly review $N=1$ supergravity coupled to $n$ linear multiplets 

4477: $L^\alpha, \alpha=1,\ldots, n$ and

4478: $r$ chiral multiplets $N^A, A=1,\ldots,r$ following \cite{BGG}.

4479: Linear multiplets are defined by the constraint 

4480: \beq\label{linearc}

4481: (D^2-8\bar R) L^\alpha = 0 = (\bar D^2-8R) L^\alpha\ ,

4482: \eeq

4483: where $D$ is the superspace covariant derivative and $R$ is the chiral 

4484: superfield containing the curvature scalar.

4485: As bosonic components $L$ contains a  real scalar field which we also

4486: denote by $L$ and the field strength of a

4487: two-form $D_2$.

4488: The superspace Lagrangian (omitting the gauge interactions) is given by 

4489: \beq\label{actionL}

4490: S = - 3 \int   E\, F(N^A,\bar N^A, L^\alpha) 

4491: + \frac12 \int \frac{E}{R}\, e^{K/2}\ W(N)

4492: + \frac12 \int \frac{E}{R^\dagger}\, e^{K/2}\ \bar W(\bar N)

4493: \ ,

4494: \eeq

4495: where $E$ is the super-vielbein and $W$ the superpotential.

4496: The function $F$

4497:  depends implicitly  on the K\"ahler potential 

4498: $K(N^A,\bar N^A, L^\alpha)$

4499: through the differential constraint\footnote{Strictly speaking

4500: $K(N^A,\bar N^A, L^\alpha)$ is not a K\"ahler potential 

4501: but as we will see it determines the kinetic terms in the action.}

4502: \bea\label{Fcon}

4503:  1- \frac{1}{3}L^\alpha K_{L^\alpha}  = F-L^\alpha F_{L^\alpha}\ ,

4504: \eea

4505: which ensures the correct normalization of the Einstein-Hilbert term.

4506: The subscripts on $K$ and $F$ denotes differentiation, i.e.\

4507: $K_{L^\alpha} = \frac{\partial K}{\partial L^\alpha},

4508: F_{L^\alpha} = \frac{\partial F}{\partial L^\alpha}$, etc.\ .

4509: Let us also define the kinetic potential $\tilde K$ and rewrite \eqref{Fcon} as

4510: \beq \label{kinpo-def}

4511:   \tilde K = K - 3 F\ ,\qquad F = 1-\tfrac{1}{3} \tilde K_{L^\alpha} L^\alpha\ .

4512: \eeq

4513: Expanding \eqref{actionL} into components one finds that $\tilde K$ 

4514: determines the kinetic terms of the fields. More precisely,

4515: the (bosonic) component Lagrangian derived from \eqref{actionL} 

4516: is found to be\footnote{This is a straightforward generalization

4517: of the Lagrangian for one linear multiplet given in \cite{BGG}.

4518: The potential for this case has also been given in \cite{HL}.}

4519: \bea\label{kinetic_lin}

4520: \cL &=& -\tfrac{1}{2}R*\mathbf{1} - 

4521:   \tilde K_{A\bar B}\, dN^A \wedge * d \bar N^{B}

4522:   + \tfrac{1}{4} \tilde K_{L^\alpha L^\beta}\, 

4523:   dL^\alpha \wedge * dL^\beta - V * \mathbf{1}\nn\\ 

4524:   && + \tfrac{1}{4} \tilde K_{L^\alpha L^\beta}\, dD^\alpha_2 \wedge * dD^\beta_2

4525:      - \tfrac{i}2\,  dD^\alpha_2 \wedge 

4526: \big(\tilde K_{\alpha A}\, dN^A -\tilde K_{\alpha \bar A}\,d\bar N^A\big)

4527: \ ,

4528: \eea

4529: where 

4530: \beq\label{Lsc}

4531: %\tilde K_{A\bar B} \equiv K_{A\bar B}- 3 L^\alpha \zeta^R_{\alpha,A\bar B}

4532: %\ , \qquad

4533:  V = e^K \Big(\tilde K^{A \bar B}D_AW D_{\bar B}\bar W - 

4534: (3- L^\alpha K_{L^\alpha}) |W|^2  \Big)\ .

4535: \eeq

4536: We see that the function

4537: $\tilde K(N,\bar N, L) = K - 3 F$ determines the kinetic terms of the fields 

4538: $N^A$ and $L^\alpha$ as well as the couplings of the two-forms $D^\alpha_2$ to 

4539: the chiral fields $N^I$. Note that only derivatives of  $F_{L^\alpha}$ appear leaving a 

4540: constant piece in  $F_{L^\alpha}$ undetermined. This constant 

4541: drops out from \eqref{Fcon}.

4542: 

4543: In a next step we like to recover the

4544: standard $N=1$ effective action by dualizing the linear 

4545: multiplets $L^\alpha$ into chiral multiplets $T_\alpha$. 

4546: This establishes the map \eqref{dual-map}, which 

4547: will be a useful tool in the remainder of this chapter.

4548: From here we can proceed in two ways.

4549: We can dualize the two-forms $D^\alpha_2$ 

4550: in components and show that the resulting action is 

4551: K\"ahler by determining the K\"ahler potential and 

4552: complex coordinates.

4553: This is done in appendix \ref{linm} and provides a simple,

4554: but somehow more tedious dualization procedure.

4555: However, performing the duality in superspace yields 

4556: directly the proper K\"ahler coordinates $T_\alpha$ and 

4557: K\"ahler potential $K(T,\bar T,N,\bar N)$.

4558:   

4559: The duality transformation in superfields is 

4560: performed in detail in \cite{BGG} and here we only repeat the 

4561: essential steps.

4562: One first considers  the linear multiplets $L^\alpha$ to be 

4563: unconstrained real superfields and modifies the action

4564: \eqref{actionL} to read\footnote{We omit the superpotential

4565: terms here since they only depend on $N$ and play no role

4566: in the dualization.}

4567: \beq\label{actionX}

4568: S = - 3 \int E\, \Big(F(N^A,\bar N^A, L^\alpha) + 

4569:     6 L^\alpha(T_\alpha + \bar T_\alpha) \Big) + \ldots\ ,

4570: \eeq

4571: where the $T_\alpha$ are chiral superfields and in order to be consistent

4572: with our previous conventions we have included a factor $6$

4573: in the second term. 

4574: Variation with respect to $T_\alpha$ results in the constraint that $L^\alpha$ are linear multiplets

4575: and one arrives back at the action \eqref{actionL}. 

4576: Variation with respect to the (unconstrained) $L^\alpha$ yields the 

4577: equations\footnote{Notice that there is a misprint

4578: in the equivalent equation given in \cite{BGG}.}

4579: \beq \label{bGl}

4580:   6 (T_\alpha + \bar T_\alpha)  + F_{L^\alpha}

4581: - \tfrac{1}{3} K_{L^\alpha} 

4582: \big(F+ 6 L^\beta (T_\beta + \bar T_\beta)\big)  =0 \ ,

4583: \eeq

4584: where we have used 

4585: $\delta_{L} E = -\tfrac{1}{3} E K_{L^\alpha} \delta L^\alpha$.

4586: This equation determines  

4587: $L^\alpha$ in terms of the chiral superfields $N^A,T_\alpha$ and is the looked

4588: for duality relation.

4589: However, depending on the specific form of $F$ and $K$ 

4590: one might not be able to solve \eqref{bGl} explicitly

4591: for $L^\alpha$ but instead only obtain an implicit

4592: relation  $L^\alpha(N,\bar N, T+\bar T)$.

4593: Nevertheless one should 

4594: insert  $L^\alpha(N,\bar N, T+\bar T)$ back  into \eqref{actionX} 

4595: which then expresses the Lagrangian (implicitly) in terms 

4596: of $T_\alpha$ and therefore defines a Lagrangian in the chiral superfield

4597: formalism. 

4598: The unusual feature being that the explicit functional dependence is 

4599: not known. A correctly normalized Einstein-Hilbert term is ensured by 

4600: additionally imposing

4601: \beq \label{normeq}

4602:   F(N,\bar N,L) + 6 L^{\alpha}(T_\alpha + \bar T_\alpha) = 1\ .

4603: \eeq 

4604: Contracting \eqref{bGl} with $L^\alpha$ and using equation \eqref{normeq} one obtains

4605: \eqref{Fcon}. Thus $F$

4606: has to have the same functional dependence as before

4607: and therefore eqn.~\eqref{kinpo-def} is unmodified, but 

4608: one should insert $L(N,\bar N,T+\bar T)$ implicitly 

4609: determined by \eqref{bGl}. Using \eqref{normeq} the duality 

4610: condition \eqref{bGl} can be cast into the form 

4611: \beq \label{dual_coords}

4612:     T_\alpha + \bar T_\alpha = \tfrac{1}{2}\tilde K_{L^\alpha}\ ,

4613: \eeq 

4614: where $\tilde K$ is the kinetic potential defined in \eqref{kinpo-def}.

4615: We also like to rewrite the K\"ahler potential $K\big( L(N,\bar N, T +\bar T),N,\bar N\big)$

4616: in terms of $\tilde K$. Inserting \eqref{dual_coords} into \eqref{kinpo-def} one infers

4617: \beq \label{Kpot_dual}

4618:   K(N,\bar N, T+\bar T) = \tilde K(N,\bar N,L) - 2(T_\alpha + \bar T_\alpha) L^\alpha\ ,

4619: \eeq  

4620: where we removed a constant factor by means of a K\"ahler transformation. 

4621: Equation \eqref{dual_coords} identifies $T_\alpha + \bar T_\alpha$ to be the canonical 

4622: conjugate to $L^\alpha$ with respect to $\tilde K$, while by \eqref{Kpot_dual} the

4623: K\"ahler potential $K$ is the Legendre transform of $\tilde K$. 

4624: The equations \eqref{dual_coords} and \eqref{Kpot_dual} 

4625: characterize the map \eqref{dual-map} and can 

4626: be equivalently obtained by a component field dualization as shown in appendix \ref{linm}.

4627: Before turning to the orientifold examples let us calculate the the bosonic 

4628: effective action in terms of $\tilde K$ and the coordinates

4629: \beq \label{coordinates}

4630:  N^A\ ,\qquad  T_\alpha=i\tilde \xi_\alpha 

4631: + \tfrac{1}{4} \tilde K_{L^\alpha}\ ,

4632: \eeq 

4633: where $\tilde \xi_\alpha$ is the scalar dual to $D_2^\alpha$ and we have used \eqref{dual_coords}.

4634: Using the K\"ahler potential \eqref{Kpot_dual} one obtains

4635: \bea \label{dual_lagra}

4636: \cL &=& -\tfrac{1}{2}R*\mathbf{1} - 

4637:   \tilde K_{N^k\bar N^l}\, dN^k \wedge * d \bar N^{l}

4638:   + \tfrac{1}{4} \tilde  K_{L^\kappa L^\lambda}\, 

4639:   dL^\kappa \wedge * dL^\lambda  - V * \mathbf{1} \\ 

4640:   && + 4 \tilde K^{L^\kappa L^\lambda} \Big(d\tilde \xi_\kappa - \tfrac{1}2

4641:   \I \big(\tilde K_{L^\kappa N^l}\,dN^l\big)\Big)\wedge * 

4642:   \Big(d\tilde \xi_\lambda - \tfrac{1}2

4643:   \I \big(\tilde K_{L^\lambda N^k}\,dN^k\big)\Big) \ .\nn  

4644: \eea

4645: where $\tilde K$ is the kinetic potential appearing in \eqref{Kpot_dual}.

4646: This is the dual Lagrangian to \eqref{kinetic_lin} as can be equivalently shown

4647: by component field dualization (equation \eqref{eff_act1}). 

4648: We now give some explicit examples for this dualization, by applying it

4649: to the Calabi-Yau orientifolds studied in chapter \ref{effective_actO}.

4650: 

4651: 

4652: \subsection{Two simple examples: Type IIB orientifolds \label{IIB_lin}}

4653: 

4654: \subsubsection{I.\quad $O3/O7$ orientifolds}

4655: 

4656: Let us now restrict to simple potentials $K(N,\bar N, L)$ and 

4657: $F(N,\bar N, L)$, which describe the correct kinematics 

4658: for $O3/O7$ orientifolds. Our aim is to rewrite the action \eqref{S_scalarO3}

4659: in the linear multiplet formalism. As we are going to show this enables us to

4660: circumvent the implicit definition of the K\"ahler potential \eqref{kaehlerpot-O7-1}.

4661: In other words, replacing the chiral multiplets $T_\alpha$ with linear multiplets $L^\alpha$

4662: as just described allows us to give an explicit expression for $K$ in terms of $\tau,z$ and 

4663: $L^\alpha$ \cite{TGL1}. This is achieved by the K\"ahler potential

4664: \beq\label{KL}

4665:   K = K_0(N^A, \bar N^{A}) + \alpha\ln (\KK_{\alpha \beta \gamma} L^\alpha L^\beta L^\gamma)\ ,

4666: \eeq

4667: where we

4668: leave $K_0(N^A, N^{\bar A})$ and the normalization constant

4669: $\alpha$ arbitrary for the moment.

4670: Inserting \eqref{KL} into \eqref{Fcon} shows that possible solutions $F$ have the form 

4671: \beq\label{FL}

4672:    F=1 - \alpha + \tfrac{1}{3}\, L^\alpha \zeta^R_\alpha(N^A, \bar N^{ A}) \ ,

4673: \eeq

4674: where  the real functions $\zeta^R_\alpha(N^A, \bar N^{A})$ are

4675: not further determined by \eqref{Fcon}. In that sense the 

4676: $\zeta^R_\alpha(N^A, \bar N^{A})$ are additional input functions

4677: which determine the Lagrangian since they appear in the

4678: kinetic potential \eqref{kinpo-def}.

4679: Comparing \eqref{tau} with \eqref{dual_coords} by using \eqref{KL}

4680: and \eqref{FL} we are led to identify\footnote{Strictly

4681: speaking \eqref{dual_coords} only determines the real part

4682: of $T_\alpha$. The imaginary part can be found by comparing 

4683: the explicit effective actions \eqref{S_scalarO3} and \eqref{dual_lagra}.}

4684: \beq\label{zetaid}

4685: \alpha = 1\ , \qquad  L^\alpha =  \tfrac{3}{2} e^{\phi}\, \frac{ v^\alpha}{\KK} \ , \qquad 

4686:  \zeta^R_\alpha = -\frac{i}{2(\tau-\bar \tau)} \KK_{\alpha b c} (G-\bar G)^b (G- \bar G)^c\ ,

4687: \eeq

4688: where $\zeta^R_\alpha = \zeta_\alpha+\bar \zeta_\alpha$ was already given in \eqref{zetadef}.

4689: Hence, we have shown that the definition of the K\"ahler coordinates in \eqref{tau}

4690: is nothing but the duality relation \eqref{dual_coords} obtained from the superfield

4691: dualization of  the linear multiplets $L^\alpha$ to chiral multiplets $T_\alpha$.\footnote{%

4692: The case $\alpha=1$ is a somewhat special situation 

4693: in that the function $F$ does not have a constant piece but only the term

4694: linear in $L^\alpha$.

4695: This in turn requires that the $\zeta^R_\alpha$ cannot be chosen zero but that they

4696: have at least a constant piece so that $F$ does not

4697: vanish. This constant is otherwise irrelevant since it

4698: drops out of all physical quantities.

4699: (In a slightly different context 

4700: the case $\alpha=1$ has also been discussed  in ref.\ \cite{Binetruy}.)}

4701: It remains to determine $K_0$. Comparing \eqref{KL} by using \eqref{zetaid}

4702: with \eqref{kaehlerpot-O7-1} one finds 

4703: \beq\label{K0}

4704: K_0 =  K_{\rm cs}(z,\bar z)  -\text{ln}\big[-i(\tau - \bar \tau)\big] \ .

4705: \eeq

4706: In summary, the low energy effective action for $O3/O7$ orientifolds 

4707: can be rewritten by using chiral multiplets $(z^k,\tau,G^a)$ and linear

4708: multiplets $L^\alpha$. This supergravity theory is determined (in the formalism of ref.\

4709: \cite{BGG} and apart from $W$ and $f$ which we can neglect for this discussion)

4710: by the independent functions $K$ and $F$ given in \eqref{KL} and \eqref{FL} together with

4711: \eqref{zetaid} and \eqref{K0}. 

4712: Inserted into \eqref{kinpo-def} we determine the kinetic potential

4713: \beq\label{tildeK_O3}

4714:  \tilde K(z, \tau, G,L) = K_{\rm cs}(z,\bar z) + 

4715:       \ln\Big(\frac{1}{2}\frac{\cK_{\alpha \beta \gamma} L^\alpha L^\beta L^\gamma}{l^0} \Big)  

4716:        - \frac{\cK_{\alpha a b} L^\alpha l^a l^b}{l^0}\ ,

4717: \eeq

4718: where we have defined $l^a = \I G^a$ and $l^0 = \I \tau$.

4719: In the dual formulation where the linear multiplets $L^\alpha$ are dualized

4720: to chiral multiplets $T_\alpha$ the Lagrangian is entirely determined

4721: by the K\"ahler potential given in \eqref{kaehlerpot-O7-1}  with the `unusual'

4722: feature that it is not given explicitly in terms of the chiral

4723: multiplets but only implicitly via the constraint \eqref{dual_coords}.

4724: In this sense the orientifold compactifications 

4725: (and similarly the compactifications of F-theory on elliptic Calabi-Yau

4726: fourfolds considered in \cite{HL} and section \ref{F-theory}) lead to 

4727: a more general class of K\"ahler potentials

4728: then usually considered in supergravity.

4729: In fact the same feature holds for arbitrary $K_0$ and arbitrary $\zeta^R_\alpha$,

4730: such that also $O3/O7$ orientifolds with space-time filling $D3$  and $D7$

4731: branes fall into this class as shown in \cite{GGJL,JL}.

4732: 

4733: Furthermore, these `generalized' K\"ahler potentials are all of 

4734: `no-scale type' in that they lead to a positive semi-definite potential $V$.

4735: For $\alpha=1$ (and arbitrary $K_0$ and $\zeta_\alpha$) 

4736: the K\"ahler potential \eqref{KL} obeys 

4737: \beq

4738: L^\alpha K_{L^\alpha} = 3\ ,

4739: \eeq

4740: and hence the

4741: the second term in the potential \eqref{Lsc} vanishes leaving a positive semi-definite

4742: potential with a supersymmetric Minkowskian ground state.

4743: Since in the chiral formulation $K$ cannot even be given explicitly one can

4744: consider such $K$s  as a `generalized' class of 

4745: no-scale K\"ahler potentials.

4746: The analogous property has also been observed in refs.\ \cite{HL,BBHL,DAFT}.

4747: Finally note with what ease the no-scale property follows in the 

4748: linear formulation compared to the somewhat involved computation

4749: in the chiral formulation performed in \cite{TGL1}.

4750: 

4751: \subsubsection{II.\quad $O5/O9$ orientifolds}

4752: 

4753: As second simple example let us dualize the 

4754: effective action \eqref{S_scalarO5} of orientifolds with $O5/O9$ planes.

4755: In this case our motivation is slightly different, since in contrast

4756: to $O3/O7$ orientifolds, the K\"ahler potential is already given explicitly in 

4757: terms of the K\"ahler coordinates. Recall however, that type IIB compactified

4758: on a Calabi-Yau naturally admits a double tensor multiplet $(\phi,C_0,B_2,C_2)$

4759: which is truncated to the linear multiplet $L=(\phi,C_2)$ by the $O5/O9$ orientifold

4760: projection. In section \ref{IIB_orientifolds} we  dualized $C_2$ to a scalar $h$ and extracted 

4761: the K\"ahler potential in the chiral picture. However, 

4762: with the techniques presented above, we are now in the position to 

4763: formulate this $N=1$ theory by keeping the linear multiplet $L$ \cite{TGL1}.   

4764: 

4765: Let us determine $\tilde K=K-3F$ encoding the couplings of the 

4766: chiral and linear multiplets in \eqref{kinetic_lin}. As we will show 

4767: in a moment the potential $K(N,\bar N, L)$ and the function $F(N,\bar N,L)$ are given by

4768: \beq\label{KpotL}

4769:   K =  K_0 + \text{ln}\, L\ , \qquad F = \tfrac{2}{3} + \tfrac{1}{3}\, L \zeta^R\ ,

4770: \eeq

4771: which is readily checked to be a solution of the normalization condition \eqref{Fcon}.

4772: Comparing equation \eqref{dual_coords} for $S$ by using the Ansatz \eqref{KpotL} with the 

4773: definition \eqref{Kcoord} of $S$ one determines $L$ and $\zeta^R$ as 

4774: \beq \label{Lzeta_O59}

4775:   L = \tfrac{3}{2} e^{\phi}\cK^{-1}\ , \qquad 

4776:   \zeta^R =  \tfrac{1}{4} (A+ \bar A)_a (\text{Re}\N^{-1})^{ab} (A+ \bar A)_b\ .

4777: \eeq

4778: Inserted back into \eqref{KpotL} indeed yields the K\"ahler potential \eqref{O5-Kaehlerpot}

4779: if we identify 

4780: \beq \label{KLdetail}

4781:   K_0 = K_{\rm cs}(z,\bar z)

4782:         - \text{ln}\Big[\tfrac{1}{48}\KK_{\alpha \beta \gamma}(t+\bar t)^\alpha 

4783:         (t+\bar t)^\beta (t+\bar t)^\gamma  \Big]\ , 

4784: \eeq

4785: Thus we have shown that the kinetic terms can consistently

4786: be described either in the chiral- or the linear multiplet formalism

4787: and we have determined the appropriate coordinates. 

4788: 

4789: Let us supplement our analysis with another formulation of the $O5/O9$ setups. 

4790: Namely we like to dualize the chiral multiplet $S$ 

4791: as well as the chiral multiplets $A_a$ into a linear multiplets $L^0$ and $L^a$. 

4792: As we will see, this will be a first case where $F(N,\bar N,L)$ is not linear

4793: in the linear multiplets $L^0,L^a$ in contrast to \eqref{FL} and \eqref{KpotL}.

4794: We will show  momentarily that the K\"ahler potential still has the form 

4795: \beq \label{K_dualA}

4796:   K(z,t,L) = K_0(z,t) + \ln L^0\ ,   

4797: \eeq

4798: where $K_0$ is the same as in \eqref{KLdetail}. $F$ can be deduced from 

4799: equation \eqref{dual_coords}, which translates to 

4800: \beq

4801:   \tfrac{1}{2}\tilde K_{L^0} = S + \bar S\ , \qquad \tfrac{1}{2}\tilde K_{L^a} = A_a + \bar A_a

4802: \eeq 

4803: Inserting \eqref{K_dualA} and the coordinates $S,A_a$ given in \eqref{Kcoord} one easily concludes that

4804: \beq \label{F_dualA}

4805:   L^0=\tfrac{3}{2} e^\phi \frac{1}{\cK}\ ,\qquad L^a= \tfrac{3}{2} e^\phi \frac{b^a}{\cK}\ ,\qquad 

4806:   F =  \tfrac{2}{3} - \tfrac{1}{3} (L^0)^{-1} \cK_{\alpha a b} (t^\alpha + \bar t^\alpha) L^a L^b\ .

4807: \eeq

4808: where $L^0$ is equal to $L$ in \eqref{Lzeta_O59}. Together with \eqref{K_dualA} 

4809: this is consistent with the normalization equation \eqref{Fcon}.

4810: Inserting \eqref{K_dualA} and \eqref{F_dualA} into \eqref{kinpo-def} the kinetic potential reads

4811: \beq \label{tildeK_O5}

4812:   \tilde K(z,t,A,L) = K_{\rm cs}(z,\bar z) 

4813:       - \ln\Big(\frac{1}{6} \frac{\cK_{\alpha \beta \gamma} l^\alpha l^\beta l^\gamma}{L^0} \Big) 

4814:       + 2 \frac{\cK_{\alpha a b} l^\alpha L^a L^b}{L^0}\ ,

4815: \eeq

4816: where we have defined $l^\alpha = \R\, t^\alpha$. 

4817: 

4818: Let us close this discussion by comparing this 

4819: kinetic potential with the one obtained for $O3/O7$ orientifolds in \eqref{tildeK_O3}. They are identical 

4820: under the identifications 

4821: \beq

4822:   \tilde K_{O3/O7}\ \rightarrow\ -\tilde K_{O5/O9}\ ,\qquad  L^\alpha \rightarrow\ l^\alpha \ ,\qquad 

4823:   (l^a,l^0) \rightarrow\ (L^a,L^0)\ .

4824: \eeq

4825: Note however, that this is a rather drastic step, since we identify linear multiplets of the one 

4826: theory with chiral multiplets of the other. It would be interesting to explore 

4827: this duality in more detail. It corresponds in simple cases to two T-dualities and 

4828: manifests itself by a rotation of the forms 

4829: \beq

4830:   \pev \ \rightarrow\ i\pev\ , \qquad (\fe, \feh)\ \rightarrow\ (-\feh, \fe)\ .

4831: \eeq 

4832: This ends our discussion of IIB orientifolds. As we have seen, much of the 

4833: underlying K\"ahler geometry can be directly analyzed by simply switching to 

4834: the linear multiplet picture.

4835: 

4836: \subsection{An involved example: Type IIA orientifolds \label{IIA_lin}}

4837: 

4838: Let us now turn to a more involved application of the linear multiplet 

4839: formalism or rather the Legendre transform method behind \eqref{dual_coords} 

4840: and \eqref{Kpot_dual}. Namely, we will present a more detailed analysis 

4841: of the moduli space $\tilde \cM^{\rm Q}$ for type IIA orientifolds \cite{TGL2}.

4842: Our aim is to show that the K\"ahler potential \eqref{intKQ} with coordinates 

4843: $T_\lambda,N^k$ introduced in \eqref{Oexp} indeed encodes the correct 

4844: low-energy dynamics of the theory obtained by Kaluza-Klein reduction.

4845: Furthermore, we show that $K^{\rm Q}$ always obeys a no-scale

4846: type condition equivalent to \eqref{no-scale2}.

4847: 

4848: Let us start by performing the reduction of the ten-dimensional 

4849: theory by using the general basis $(\alpha_\Kh,\beta^\Kh)$ 

4850: introduced in \eqref{decompO2}. It was chosen such that it splits on 

4851: $H^3(Y)=H^{3}_+ \oplus H^3_-$ as 

4852: \beq \label{basis1}

4853:   (\alpha_k,\beta^\lambda) \in H^{3}_+(Y)\ , \qquad  (\alpha_\lambda,\beta^k) \in H^{3}_-(Y)\ ,

4854: \eeq

4855: where both eigenspaces are spanned by $h^{2,1}+1$ basis vectors.

4856: As remarked above, we will only concentrate on the moduli space 

4857: $\tilde \cM^{\rm Q}$, such that we can set $t^a=0$ and $A^\alpha=0$.

4858: Due to \eqref{fieldtransf}, the ten-dimensional three-form $\hat C_3$ is expanded in 

4859: elements of $H^{3}_+(Y)$ as 

4860: \beq

4861:   \CC_3 = \xi^k(x)\, \alpha_k - \tilde \xi_\lambda(x)\, \beta^\lambda \ ,

4862: \eeq

4863: where $\xi^k, \tilde \xi_\lambda$ are $h^{2,1}+1$ real space-time scalars in 

4864: four-dimensions. Inserting this Ansatz into the ten-dimensional effective 

4865: action one finds

4866: \bea  \label{act2}

4867:   S^{(4)}_{\tilde \cM^{\rm Q}} &=& \int -\, d D \wedge * dD  -\, G_{K L}(q)\, dq^K \wedge * dq^L 

4868:          +\tfrac{1}{2} e^{2D}\, \text{Im}\, \cM_{ k  l}\, 

4869:          d\xi^{ k} \wedge * d\xi^{l} \\

4870:       &&  + \tfrac{1}{2} e^{2D}\, (\text{Im}\, \cM)^{-1\ \kappa \lambda}

4871:       \big(d\tilde \xi_\kappa - \text{Re}\, \cM_{\kappa  l}\, 

4872:       d\xi^{l} \big)

4873:         \wedge * \big(d\tilde \xi_\lambda-\text{Re}\, \cM_{\lambda  k}\, d\xi^{ k} \big)\ , \nn 

4874: \eea

4875: where compared to \eqref{act1} only the terms involving $\xi^{k},\tilde \xi_\lambda$ have

4876: changed. The metric $G_{K L}(q)$ was introduced in \eqref{def-G}

4877: and is the induced metric on the space of real 

4878: complex structure deformations $\cM^{\rm cs}_\bbR$ parameterized by $q^K$. 

4879: It remains to comment on the kinetic and coupling terms of the 

4880: scalars $\xi^k, \tilde \xi_\lambda$. In the quaternionic metric

4881: \eqref{q-metr} of the $N=2$ theory they couple via the 

4882: matrix $\cM_{\Kh \Lh}$ given in \eqref{defM}. Using the split of the symplectic basis 

4883: $(\alpha_\Kh, \beta^\Kh)$ as given in \eqref{basis1} and the fact that by Hodge duality 

4884: for a form $\gamma \in H^{3}_+$ one finds $ * \gamma \in H^{3}_-$ one concludes

4885: \beq

4886:   \text{Re} \cM_{\kappa \lambda}(q) = \text{Re} \cM_{k l}(q) = \text{Im} \cM_{\kappa k}(q) = 0\ , 

4887: \eeq

4888: whereas $\text{Re} \cM_{k \lambda}, \text{Im} \cM_{\kappa \lambda}, \text{Im} \cM_{k l}$ 

4889: are generally non-zero on $\cM^{\text{cs}}_{\mathbb{R}}$. The explicit form of non-vanishing

4890: components can be obtained by restricting \eqref{gauge-c} to $\cM^{\rm cs}_\bbR$ and

4891: using the constraints \eqref{Z=0gen}.

4892: 

4893: In order to combine the scalars $e^D,q^K$ with $\xi^k, \tilde \xi_\kappa$ into 

4894: complex variables, we have to redefine these fields and rewrite the first two 

4895: terms in \eqref{act2}. Thus we define the $h^{2,1}+1$ real coordinates

4896: \beq \label{lL-def}

4897:    L^\lambda\ =\ - e^{2D}\, \I \big[C Z^\lambda(q) \big]\ ,\qquad 

4898:    l^k \ =\  \R\big[C Z^k(q)\big]\ ,

4899: \eeq

4900: which is consistent with the orientifold constraint 

4901: \eqref{Z=0gen}. The additional factor of $e^{2D}$ was included in order 

4902: to match the dilaton factors later on.

4903: Using \eqref{lL-def} one calculates the Jacobian matrix 

4904: for the change of variables $(e^D,q^K)$ to $(l^k,L^\lambda)$ as 

4905: explicitly done in \cite{TGL2}.

4906: It is then straight forward to 

4907: rewrite \eqref{act2} by using the 

4908: identities \eqref{spconst} of special geometry as

4909: \begin{align} \label{IIA1}

4910:  S^{(4)}_{\tilde \cM^{\rm Q}} = & \int  2 e^{-2D} \text{Im} \cM_{\kappa \lambda}\, dL^\kappa \wedge * dL^\lambda

4911:                         +  2 e^{2D} \text{Im} \cM_{k l}\, dl^k \wedge * dl^l

4912:                         + \tfrac{e^{2D}}{2}  \text{Im} \cM_{ k  l}\, 

4913:                           d\xi^{ k} \wedge * d\xi^{l}  \nn\\   

4914:        &+ \tfrac{e^{2D}}{2} \, (\text{Im}\, \cM)^{-1\ \kappa \lambda}

4915:       \big(d\tilde \xi_\kappa - \text{Re}\, \cM_{\kappa  k}\, 

4916:       d\xi^{k} \big)

4917:         \wedge * \big(d\tilde \xi_\lambda-\text{Re}\, \cM_{\lambda  k}\, d\xi^{ k} \big)\ . 

4918: \end{align}

4919: From \eqref{IIA1} one sees that the scalars $l^k$ and $\xi^k$ nicely combine 

4920: into complex coordinates 

4921: \beq

4922:    N^k\ =\ \tfrac{1}{2}\xi^k +  i l^k\ =\ \tfrac{1}{2}\xi^k +  i \R(C Z^k)\ ,

4923: \eeq

4924: which corresponds to \eqref{def-NT}.

4925: In contrast, one observes that 

4926: the metric for the kinetic terms of the

4927: scalars $\tilde \xi_\lambda$ is exactly the inverse of the one

4928: appearing in the kinetic terms of the scalar fields $L^\lambda$. 

4929: Hence, comparing \eqref{IIA1} with \eqref{dual_lagra} on concludes that this 

4930: action is obtained by dualizing a set of linear multiplets 

4931: $(L^\lambda, D^\lambda_2)$ into chiral multiplets 

4932: $(L^\lambda,\tilde \xi_\lambda)$. To extract $\tilde K(L, N,\bar N)$ we 

4933: compare \eqref{IIA1} with \eqref{dual_lagra} and read off the metric

4934: \beq \label{lLmetric}

4935:    \tilde K_{L^\kappa  L^\lambda}\ =\ 8\, e^{-2D} \IM_{\kappa \lambda}\ , \quad 

4936:    \tilde K_{l^k l^l}\ =\ -8\, e^{2D} \IM_{k l} \ , \quad 

4937:    \tilde K_{L^\kappa  l^l} \ =\  - 8\, \RM_{\kappa l}\ ,

4938: \eeq

4939: where we have used that the metric is independent of $\xi^k,\tilde \xi_\lambda$.

4940: This metric can be obtained from a kinetic potential of the form

4941: \beq \label{kinpo1}

4942:   \tilde K(L,l)\ =\ - \ln\big[ e^{-4D} \big]+ 8e^{2D}\I \big[\rho^*\cF(CZ^k)\big]\ ,

4943: \eeq

4944: where $\cF$ is the prepotential of the special K\"ahler manifold $\cM^{\rm cs}$

4945: restricted to the real subspace $\cM^{\rm cs}_\bbR$. The map $\rho$ was given 

4946: in \eqref{embmap1} and enforces the constraints \eqref{Z=0gen}. To show that $\tilde K$

4947: indeed yields the correct metric \eqref{lLmetric} one differentiates \eqref{kinpo1}

4948: with respect to $e^{-D},q^K$ and uses the inverse of the Jacobian matrix 

4949: for the change of variables $(e^D,q^K)$ to $(l^k,L^\lambda)$. Applying equations

4950: \eqref{ML-hf} one finds its first derivatives  

4951: \beq \label{first-der}

4952:   \tilde K_{ L^\lambda} \ =\ - 8\, \R\big[C \cF_\lambda(q) \big] \qquad

4953:    \tilde K_{l^k} \ = \ 8\, e^{2D}\, \I\big[C \cF_{k}(q) \big]\ .

4954: \eeq

4955: Repeating the procedure and differentiating \eqref{first-der}

4956: with respect to $e^{-D},q^K$ and using once again the inverse Jacobian 

4957: one applies \eqref{def-M} to show \eqref{lLmetric}. Knowing \eqref{kinpo1}

4958: one can also extract the functions $F(L, N, \bar N)$ and $K(L,N,\bar N)$

4959: by applying \eqref{kinpo-def}. As we will show momentarily 

4960: $K$ and $F = \frac13 (K- \tilde K)$ are given by

4961: \beq

4962:   K(L,l) = - \ln \big[ e^{-4D} \big] \ , \qquad 

4963:   F(L,l) = - \tfrac{8}{3} e^{2D}\I \big[\rho^*\cF(CZ)\big]+\tfrac{1}{3}\ . 

4964: \eeq

4965: It suffices to determine $K$ which expressed in the correct coordinates 

4966: serves as the K\"ahler potential in the chiral description. 

4967: 

4968: As explained in the beginning of this section the actual K\"ahler potential of 

4969: $\tilde \cM^{\rm Q}$ is the Legendre transform \eqref{Kpot_dual} of $\tilde K$ with 

4970: respect to the variables $L^\lambda$. There we also found the explicit 

4971: definition of the complex coordinates $T_\lambda$ combining $(L^\lambda,\tilde \xi_\lambda)$.  

4972: Using \eqref{first-der} in \eqref{dual_coords} and fixing the normalization of the 

4973: imaginary part of $T_\lambda$ by comparing \eqref{IIA1} with \eqref{dual_lagra}

4974: one finds

4975: \beq

4976:   T_\lambda = i \tilde \xi_\lambda + \tfrac{1}{4}\tilde K_{L^\lambda} 

4977:            = i \tilde \xi_\lambda - 2\, \R\big(C F_\lambda\big) \ ,

4978: \eeq

4979: which coincides with \eqref{def-NT} already quoted in section \ref{Kpo_gaugeIIA}.

4980: To give an explicit expression for  $K^{\rm Q}$ we insert  

4981: equation \eqref{kinpo1} into \eqref{Kpot_dual}. Applying the $N=2$ identity 

4982: $\cF=\frac12 Z^\Kh \cF_\Kh$, the constraint equations \eqref{Z=0gen} 

4983: and \eqref{lL-def},\eqref{first-der} we rewrite 

4984: \beq  \label{K_lL}

4985:     K^{\rm Q}= - \ln\big[e^{-4D} \big] + \tfrac{1}{2} (l^k \tilde K_{l^k} - L^\lambda \tilde K_{ L^\lambda})\ .

4986: \eeq

4987: It is possible to evaluate the terms appearing in the parentheses. In order to do that 

4988: we combine the equations \eqref{lL-def} and \eqref{first-der} to the simple form 

4989: \bea \label{lL}

4990:   \R\big( C \Omega \big)\ =\ l^k \alpha_k + \tfrac{1}{8} \tilde K_{L^\lambda} \beta^\lambda\ ,\quad

4991:    e^{2D} \I\big( C \Omega \big)\ =\ -L^\lambda \alpha_\lambda - \tfrac{1}{8} \tilde K_{l^k} \beta^k\ . 

4992: \eea

4993: We now use equation \eqref{csmetric} and the definition \eqref{def-C} of $C$

4994: to calculate

4995: \beq \label{skconstr}

4996:  2 \int_Y \R( C\Omega) \wedge \I(C\Omega) = i \int_Y C\Omega \wedge \overline{C\Omega} = e^{-2D}\ .

4997: \eeq

4998: Inserting the equations \eqref{lL} into \eqref{skconstr} we find 

4999: \bea \label{lL=4} 

5000:   L^\lambda \tilde K_{L^\lambda} - l^k \tilde K_{l^k} = 4\ .

5001: \eea

5002: Inserted back into \eqref{K_lL} we have shown that the K\"ahler potential

5003: has indeed the form \eqref{intKQ}.\footnote{By using the equation \eqref{skconstr} and $*\Omega=-i\Omega$ 

5004: it is straight forward to show $e^{-2D}=2\int \R(C\Omega)\wedge * \R(C\Omega)$}

5005: Moreover, \eqref{lL=4} directly translates into a no-scale type condition for $K^{\rm Q}$

5006: \bea \label{no-scale4}

5007:   K_{w^\Kh} K^{w^\Kh \bar w^\Lh} K_{\bar w^\Lh} = 4\ ,

5008: \eea

5009: where $w^\Kh=(T_\kappa, N^k)$.

5010: In order to see this, one inserts the inverse K\"ahler metric \eqref{invKm1},

5011: the K\"ahler derivatives \eqref{Kder} and the derivatives of \eqref{lL=4} back into

5012: \eqref{lL=4}. In other words, we were able to translate one of 

5013: the special K\"ahler conditions present in the underlying 

5014: $N=2$ theory into a constraint on the geometry of 

5015: $\tilde \cM^{\rm Q}$. Two non-trivial examples fulfilling 

5016: \eqref{lL=4} are the $O3/O7$ and $O5/O9$ kinetic potentials 

5017: \eqref{tildeK_O3} and \eqref{tildeK_O5}. They admit this 

5018: simple form since instanton corrections are not taken into account. 

5019: 

5020: 

5021: 

5022: \section{The geometry of the moduli space}

5023: \label{geom_of_modspace}

5024: 

5025: In this section we give an alternative formulation of 

5026: the geometric structures of the moduli space $\tilde \cM^{\rm Q}$ 

5027: which is closely related the moduli space of 

5028: supersymmetric Lagrangian submanifolds in a Calabi-Yau 

5029: threefold \cite{Hitchin2}.\footnote{This 

5030: analysis can equivalently be applied to the moduli space of 

5031: $G_2$ compactifications of 

5032: M-theory.}  In this set-up also

5033: the no-scale conditions \eqref{no-scale2}, 

5034: \eqref{lL=4} are interpreted geometrically. This provides 

5035: a more elegant description of the $N=1$ moduli space and 

5036: its special properties. Moreover, we construct the $N=1$

5037: analog \eqref{N=1c-map} of the $N=2$ c-map \eqref{c-map}.

5038: Our analysis can serve as a starting point for the analysis

5039: of non-Calabi-Yau orientifolds by using the 

5040: language of generalized complex manifolds invented by 

5041: Hitchin \cite{HitchinGCM}.  

5042: 

5043: In section~\ref{IIA_orientifolds} we started from a $N=2$ quaternionic

5044: manifold $\cM^{\rm Q}$ and determined the submanifold

5045: $\tilde\cM^{\rm Q}$ by imposing the orientifold projection.

5046: $N=1$ supersymmetry ensured that this submanifold is K\"ahler.

5047: $\cM^{\rm Q}$ has a second but different K\"ahler submanifold

5048: $\cM^{\rm cs}$ which intersects with $\tilde\cM^{\rm Q}$

5049:  on the real manifold $\cM^{\rm cs}_\bbR$.

5050: The c-map is in some sense the reverse operation where 

5051: $\cM^{\rm Q}$ is constructed starting from $\cM^{\rm cs}$

5052: and shown to be quaternionic \cite{CFGi,FS}.

5053: In this section we analogously construct the K\"ahler manifold 

5054: $\tilde\cM^{\rm Q}$ starting from $\cM^{\rm cs}_\bbR$.

5055: 

5056: \begin{figure}[h]

5057: \begin{center}

5058: \includegraphics[height=4cm]{C-bundle.eps}

5059: \caption{\textit{The local moduli space $\cM_{\bbR} = \cM_{\bbR}^{\rm cs} \times \bbR$ in 

5060:        $\cM^{\rm cs} \times \bbC \simeq \cM^{\rm cs} \times H^{(3,0)}$.}}

5061: \label{loc_modspace}

5062: \end{center}

5063: \end{figure}

5064: 

5065: 

5066: 

5067: In fact the proper starting point is not $\cM^{\rm cs}_\bbR$ but rather

5068: $\cM_\bbR=\cM_\bbR^{\rm cs} \times \bbR$ which is the local product of the

5069: moduli space 

5070: of real complex structure deformations of a Calabi-Yau orientifold 

5071: times the real dilaton direction. The $N=2$ analog of 

5072: $\cM_\bbR$ is the extended moduli space 

5073: $\hat\cM^{\rm cs} = \cM^{\rm cs} \times \bbC$ where $\bbC$ 

5074: is the complex line normalizing $\Omega$. The corresponding modulus

5075: can be identified with the complex dilaton \cite{Witten2}.

5076: The orientifold projection fixes the phase of the complex dilaton

5077: (it projects out the four-dimensional $B_2$) to be $\theta$ and thus reduces

5078: $\bbC$ to   $\bbR$ (figure \ref{loc_modspace}). 

5079: The local geometry of $\cM_\bbR$ is encoded in the variations of the real and imaginary part of  the normalized holomorphic three-form $C\Omega$.

5080: This form naturally defines an embedding 

5081: \beq

5082:   E: \cM_\bbR \rightarrow V \times V^*

5083: = \ H^3_+(\mathbb{R}) \times H^3_-(\mathbb{R})\ .

5084: \eeq

5085: where $V =H^3_+(\mathbb{R})$ and we used the intersection 

5086: form $\big<\alpha,\beta \big>=\int \alpha \wedge \beta $ on $H^{3}(Y)$ 

5087: to identify $V^* \cong H^3_-(\mathbb{R})$. 

5088:  $V \times V^*$ naturally admits a

5089: symplectic form $\cW$ and an indefinite metric $\cG$ defined as

5090: \bea

5091:   \cW((\alpha_+,\alpha_-),(\beta_+,\beta_-)) = \big<\alpha_+, \beta_-\big> - \big<\beta_+ ,\alpha_-\big>\ , \nn\\

5092:   \cG((\alpha_+,\alpha_-),(\beta_+,\beta_-)) = \big<\alpha_+, \beta_-\big> + \big<\beta_+ ,\alpha_-\big>\ ,

5093: \eea

5094: where $\alpha_\pm,\beta_\pm \in H^3_\pm(\mathbb{R})$.

5095: 

5096: Now we construct $E$ in such a way that 

5097: $\cM_{\bbR}$ is a Lagrangian submanifold of $V \times V^*$ with respect to 

5098: $\cW$ and its metric is induced from $\cG$, i.e.\ 

5099: \beq\label{Lagr}

5100:  E^*(\cW)=0 \ , \qquad E^*(\cG)=g

5101: \eeq

5102: where

5103: \beq \label{metrQ}

5104:   \tfrac{1}{2} g = dD \otimes dD +  G_{K L} d q^K \otimes d q^L\ 

5105: \eeq 

5106:  is the metric on $\cM_\bbR$ as determined in \eqref{act1}.

5107: As we are going to show momentarily $E$ is given by

5108: \bea  \label{embmap}

5109:    E(q^\Kh) = 2\,\big(\fu ,  -e^{2D} \fuh \big)\ ,

5110: \eea 

5111: where $\fu+i\,\fuh = C\Omega$, $q^\Kh=(e^{-D},q^K)$ and 

5112: $\Omega$ is evaluated at  $q^K \in \cM^{\rm cs}_\bbR$.

5113: Additionally $E$ satisfies

5114: \beq \label{no-scale3}

5115:   \cG(E(q^\Kh),E(q^\Kh)) = 4\ ,

5116: \eeq

5117: for all $q^K $. This implies that the image of all points in 

5118: $\cM_\bbR$ have the same distance from the origin. Later on we will show that

5119: this translates into the no-scale condition \eqref{no-scale4}.

5120: 

5121: Let us first show that the $E$ given in \eqref{embmap} indeed satisfies

5122: \eqref{Lagr} and \eqref{no-scale3}. The explicit calculation is straightforward 

5123: and essentially included in the calculation presented in section \ref{IIA_lin}.\footnote{Formally one has to  

5124: first evaluate $E_* (\partial_{Q^\Kh})$ and expresses the result in terms of the $(3,0)$-form $\Omega$ 

5125: and the $(2,1)$-forms $\chi_K$. One then uses that by definition of the pull-back 

5126: $E^* \omega(\partial_{q^\Kh},\cdot ) = \omega(E_* (\partial_{q^\Kh}),\cdot)$ for a form $\omega$ on $V \times V^*$. 

5127: Applied to $\cG$ and $\cW$ one finds that the truncation of the special K\"ahler potential 

5128: \eqref{csmetric} and \eqref{chi_barchi} indeed imply \eqref{Lagr}. This calculation does not

5129: make use of any specific basis of $H^3_\pm$.} In order to connect with section 

5130: \ref{IIA_lin} let us first recall how we applied the map \eqref{dual-map} to

5131: extract the chiral data of the $N=1$ moduli space.

5132: We started with a special K\"ahler manifold $\cM^{\rm sk}$ with metric determined 

5133: in terms of a holomorphic prepotential $\cF(Z)$. 

5134: Next we assumed that the $N=2$ theory with quaternionic space $\cM^{\rm Q}$ constructed 

5135: via the local c-map \eqref{c-map} allows a reduction to $N=1$. Accordingly the section 

5136: $\Omega(z)=Z^\Kh \alpha_\Kh - \cF_\Kh \beta^\Kh$ fulfills equation \eqref{Z=0gen} for some basis 

5137: \beq \label{red-basis}

5138:   (\alpha_k,\beta^\lambda) \in H^{3}_+ \ ,\qquad  (\alpha_\lambda,\beta^k) \in H^{3}_-\ .

5139: \eeq

5140: Using this basis we found the kinetic potential $\tilde K(L,l)$ given in \eqref{kinpo1}, which 

5141: explicitly depends on the prepotential $\cF$. It encodes the metric on 

5142: $\tilde \cM^{\rm Q} \subset \cM^{\rm Q}$ via the K\"ahler potential \eqref{Kpot_dual}. 

5143: On the other hand, equation \eqref{dual_coords} defines the complex structure on 

5144: $\tilde \cM^{\rm Q}$. 

5145: 

5146: These steps can be translated into the language of this section. Namely,

5147: choosing the basis \eqref{red-basis} to expand the map $E$ defined in \eqref{embmap}

5148: one finds   

5149: \beq \label{Eincoords}

5150:   E(q^\Kh)\ =\ \big(2l^k \alpha_k + \tfrac{1}{4} \tilde K_{L^\lambda} \beta^\lambda, 

5151:                 2L^\lambda \alpha_\lambda + \tfrac{1}{4} \tilde K_{l^k} \beta^k\big)\ ,

5152: \eeq

5153: where $l^k,L^\lambda$ and $\tilde K_{L^\lambda}, \tilde K_{l^k}$ are functions of $q^\Kh$ as given in 

5154: \eqref{lL-def} and \eqref{first-der}.

5155: We define coordinates $u^\Kh=(2l^k, \tfrac{1}{4}\tilde K_{L^\lambda})$ on $V$ and 

5156: coordinates $v_\Kh=(\tfrac{1}{4}\tilde K_{l^k},-2L^\lambda)$ on $V^*$. 

5157: In these coordinates the first two conditions in \eqref{Lagr} simply read

5158: \beq \label{Lagrc}

5159:     E^*(du^\Kh \wedge dv_\Kh)=0\ , \qquad E^*(du^\Kh \otimes dv_\Kh) = g\ .

5160: \eeq

5161: {}From section \ref{IIA_lin} we further know that 

5162: $\tilde K_{L^\kappa}, \tilde K_{l^k}$ are derivatives 

5163:  of a kinetic potential $\tilde K$ and thus we can evaluate $du^\Kh$ 

5164: and $dv_\Kh$ in terms of $l^k,L^\kappa$.

5165: Inserting the result into \eqref{Lagrc} 

5166: the second equation can be rewritten as

5167: \beq

5168:   \tfrac{1}{2} g\ =\ \tfrac{1}{4} 

5169:         \tilde K_{l^k l^l}\, dl^k \otimes dl^l - 

5170:         \tfrac{1}{4} 

5171:         \tilde K_{L^\kappa L^\lambda}\, dL^\kappa \otimes dL^\lambda\ ,

5172: \eeq 

5173: while the first equation is trivially fulfilled due to the symmetry of $\tilde K_{l^k l^l}$

5174: and $\tilde K_{L^\kappa L^\lambda}$. This metric is exactly the one appearing in the action 

5175: \eqref{IIA1} when using \eqref{lLmetric}. Expressing $g$ in coordinates 

5176: $e^{D},q^K$ leads to \eqref{metrQ}, as we have already checked by going from 

5177: \eqref{act2} to \eqref{IIA1} above.

5178: Furthermore, inserting \eqref{Eincoords} into \eqref{no-scale3} 

5179: it exactly translates 

5180: into the no-scale condition \eqref{lL=4}, which was shown in section

5181: \ref{IIA_lin} to be equivalent to \eqref{no-scale2}. 

5182: 

5183: We have just shown that $\cM_\bbR$ is a 

5184: Lagrangian submanifold of  $V \times V^*$.

5185: Identifying $T^*V \cong V \times V^*$ we conclude that  $\cM_\bbR$ 

5186: can be obtained as the graph $(\alpha(u),u)$ 

5187: of a closed one-form $\alpha$. This implies that we can locally find 

5188: a generating function $K': V \rightarrow \mathbb{R}$ such that $\alpha = dK'$. 

5189: In local coordinates $(v_\Kh,u^\Kh)$ this amounts to

5190: \beq \label{v=K/u}

5191:   v_\Kh(u) = \frac{\partial K'}{\partial u^{\Kh}}

5192: \eeq 

5193: such that 

5194: \beq

5195:   - L^\kappa(u) = 2\, \frac{\partial K'(u)}{\partial \tilde K_{L^\kappa}}\ , \quad 

5196:   \tilde K_{l^k}(u) = 2\, \frac{\partial K'(u)}{\partial l^k}\ . 

5197: \eeq

5198: These equations are satisfied if we define $K'$ in terms of $\tilde K$ as

5199: \bea \label{Legendre}

5200:   2 K'\ =\ \tilde K(L(u),l) - \tilde K_{L^\kappa}(u)\, L^\kappa(u)\ , 

5201: \eea  

5202: which is nothing but the Legendre transform of $\tilde K$ with respect to $L^\kappa$.

5203: Later on we show that the function ${2}K'$  is identified with the K\"ahler potential 

5204: $K$ given in \eqref{intKQ}. 

5205: 

5206: In order to do that, we now extend our discussion to the 

5207: full moduli space $\tilde \cM^{\rm Q}$ including the scalars 

5208: $\zeta^\Kh=(\xi^k,\tilde \xi_\kappa)$ parameterizing the 

5209: three-form $\hat C_3$ in $H^{3}_+(\bbR)$. Locally one has

5210: \beq

5211:    \tilde \cM^{\rm Q} = \cM_\bbR \times H^{3}_+(\bbR)\ .

5212: \eeq

5213: The tangent space at a point $p$ in $ \tilde \cM^{\rm Q}$ can be identified as

5214: \beq

5215:   T_p \tilde \cM^{\rm Q} \cong H^3_+(\bbR)\oplus H^3_+(\bbR) \cong H^3_+(\bbR) \otimes \bbC\ ,

5216: \eeq

5217: where the first isomorphism is induced by the embedding $E$ given in \eqref{embmap}.

5218: This is a complex vector space and thus $\tilde \cM^{\rm Q}$ admits an 

5219: almost complex structure $I$. In components it is given by

5220: \beq \label{def-I}

5221:   I(\partial_{q^\Kh}) = (\partial u^\Lh/\partial q^\Kh)\, \partial_{\zeta^\Lh}\ ,\qquad

5222:   I((\partial u^\Lh/\partial q^\Kh)\, \partial_{\zeta^\Lh})=-\partial_{q^\Kh}\ ,

5223: \eeq

5224: where we have used that $I$ is induced by the embedding map $E$. One can show that

5225: the almost complex structure $I$ is integrable, since  

5226: \beq

5227:   dw^\Kh = du^\Kh + i d \zeta^\Kh = (\partial u^\Lh/\partial q^\Kh) dq^\Kh + i d \zeta^\Kh\ ,

5228: \eeq

5229: are a basis of $(1,0)$ forms and $w^\Kh=u^\Kh+i\zeta^\Kh$ are complex coordinates on $\tilde \cM^{\rm Q}$. 

5230: Using the definition of $u^\Kh$ one infers that as expected $w^\Kh = (N^k,T_\kappa)$. 

5231: Moreover, one naturally extends the metric $g$ on $T \cM_\bbR$ to a hermitian metric 

5232: on $T\tilde \cM^{\rm Q}$. The corresponding two-form is then given by

5233: \beq

5234:   \tilde \omega(\partial_{\zeta^\Lh}, \partial_{q^\Kh}) = g(I\partial_{\zeta^\Lh},\partial_{q^\Kh})\ ,

5235:   \qquad \tilde \omega(\partial_{\zeta^\Kh},\partial_{\zeta^\Lh})= \tilde \omega(\partial_{q^\Kh},\partial_{q^\Lh})=0\ .

5236: \eeq

5237: Using the definition \eqref{def-I} of the almost complex structure and 

5238: equation \eqref{Lagr}, one concludes that $\tilde \omega$ is given by

5239: \beq \label{tildeo}

5240:    \tilde \omega= dv_\Kh \wedge d\zeta^\Kh 

5241:                 = 2i \frac{\partial^2 K'}{\partial w^\Kh \partial \bar w^\Lh} dw^\Kh \wedge d\bar w^\Lh   \ ,

5242: \eeq

5243: where for the second equality we applied \eqref{v=K/u} and expressed the 

5244: result in coordinates $w^\Kh=u^\Kh + i \zeta^\Kh$. Note that $K'$ is a function 

5245: of $u^\Kh$ only, such that derivatives with respect to $w^\Kh$ translate to ones

5246: with respect to $u^\Kh$. Equation \eqref{tildeo} implies that $K^{\rm Q}=2K'$ is indeed

5247: the correct K\"ahler potential for the moduli space $\tilde \cM^{\rm Q}$. 

5248: 

5249: So far we restricted ourselves to type IIA orientifolds. However, by using the 

5250: mirror map \eqref{pure-spinor-map2} one easily translates the above construction to IIB 

5251: setups. In the IIB case the real manifold started with is simply the

5252: local product $\cM^B_\bbR = \cM^{\rm ks}_\bbR \times \bbR$, where $\cM^{\rm ks}_\bbR$

5253: is a real slice in the complexified K\"ahler cone $\cM^{\rm ks}$ and $\bbR$ parameterizes

5254: the four-dimensional dilaton direction. $\cM^{\rm ks}_\bbR$ is locally spanned by 

5255: the fields $v^\alpha$ and $b^a$ introduced in section \ref{IIB_orientifolds}. 

5256: Once again we aim to find the embedding map $E$ 

5257: \beq

5258:   E: \cM^B_\bbR \rightarrow V \times V^*\ .

5259: \eeq

5260: In order to be more explicit we distinguish $O3/O7$ and $O5/O9$ setups and  

5261: define 

5262: \beq

5263:   E_{O3/7}(q^\Kh)\ =\ 2\, (\fe,\ e^{2D_B} \feh)\ ,\qquad 

5264:   E_{O5/9}(q^\Kh)\ =\ 2\, (\feh,\ e^{2D_B} \fe)\ ,

5265: \eeq

5266: where $\fe+i\feh = e^{-\phi} e^{-B+iJ}$ and

5267: $q^{\Ah}=(e^{-D_B},v^\alpha,b^a)$. Correspondingly we need to set

5268: \beq

5269:   V_{O3/7}=H^{ev}_+\ ,\quad V_{O3/7}^*=H^{ev}_- \qquad

5270:   V_{O5/9}=H^{ev}_-\ ,\quad V_{O5/9}^*=H^{ev}_+\ , 

5271: \eeq

5272: where we have abbreviated \footnote{Recall that $H^{(0,0)}_- = H^{(3,3)}_-=0$ as discussed in section 

5273: \ref{IIB_orientifolds}.}

5274: \beq \label{Heven}

5275:   H^{ev}_+ = H^{(0,0)}_+ \oplus H^{(1,1)}_- \oplus H^{(2,2)}_+ \ ,\qquad 

5276:   H^{ev}_- = H^{(1,1)}_+ \oplus H^{(2,2)}_- \oplus H^{(3,3)}_+\ .

5277: \eeq

5278: Given a vector space $V$ of even forms, the identification of $V^*$ with the 

5279: respective cohomology groups is done by using the intersection form $\big<\cdot,\cdot\big>$ 

5280: defined in \eqref{symp-form}.

5281: To check that $E_{O3/7}$ and $E_{O5/9}$ are defined correctly, one 

5282: proceeds in full analogy to the type IIA case. Once again, the calculation 

5283: simplifies considerably by using the existence of the kinetic potentials \eqref{tildeK_O3} and 

5284: \eqref{tildeK_O5}.  

5285: 

5286: Let us summarize our results. We constructed the metric and complex structure of 

5287: the K\"ahler manifold $\tilde \cM^{\rm Q} \subset \cM^{\rm Q}$ by specifying a map 

5288: \beq

5289:    E: \cM_\bbR \rightarrow V \times V^*\ ,

5290: \eeq

5291: where $\cM_{\bbR}$ parameterizes the real four-dimensional dilaton direction times 

5292: certain deformations of the Calabi-Yau orientifold. $V$ is an appropriately chosen 

5293: vector space

5294: \beq

5295:   V_{IIA} = H^{odd}_+\ ,\qquad V_{IIB} = H^{ev}_\pm\ ,

5296: \eeq

5297: where $H^{odd}_+=H^3_+$ and $H^{ev}_\pm$ is given in \eqref{Heven}. 

5298: More explicitly $E$ takes the form 

5299: \beq

5300:    E(q^\Kh)=2\, \big(\rho,- \hat \rho / \Phi_{A,B}\big)\ ,

5301: \eeq

5302: where $\Phi_{A,B}(\rho)$ is given in \eqref{def-Phi}, \eqref{def-PhiO5} and 

5303: $\rho=(\fu,\fe,\feh)$ depending on the orientifold setup. In order 

5304: to evaluate $\Phi_{A,B}(\rho)=e^{-2D}$ we use the definition of the four-dimensional dilaton 

5305: \eqref{def-D_B}.

5306: Since $\cM_{\bbR}$ is embedded as a Lagrange submanifold in $V\times V^*$ it can be locally given

5307: by the graph of the one-form $dK'$. Moreover, since $E$ induces the metric on 

5308: $\cM_{\bbR}$ and a complex structure on $\cM_{\bbR}\times V$ the function $2K'$ is nothing but the K\"ahler 

5309: potential on the local moduli space $\tilde \cM^{\rm Q}=\cM_{\bbR}\times V$. Thus, the 

5310: difficulty is to find the map $E$ or, by recalling \eqref{v=K/u}, the functional dependence 

5311: $\hat \rho(\rho)$. This non-linear map 

5312: \beq

5313:   \rho\ \mapsto\ \hat \rho(\rho)\ ,   

5314: \eeq 

5315: lies at the heart of Hitchins approach to extract the geometry 

5316: of even and odd forms on six-manifolds \cite{Hitchin1,HitchinGCM}. 

5317: One may thus hope to generalize Calabi-Yau orientifolds to non-Calabi-Yau orientifolds \cite{GLprep}.

5318: 

5319: 

5320: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

5321: %

5322: %   NS-NS and R-R fluxes 

5323: %

5324: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

5325: 

5326: 

5327: \chapter{Calabi-Yau orientifolds with NS-NS and R-R background fluxes}

5328: \label{fluxesAB}

5329: 

5330: In this chapter we redo the reduction of type IIB and type IIA on Calabi-Yau 

5331: orientifolds by additionally allowing for non-trivial R-R and 

5332: NS-NS background fluxes. As we will show, these fluxes 

5333: result in non-trivial potentials for the supergravity fields and 

5334: can lead to charged scalars or massive tensors. 

5335: 

5336: 

5337: We first discuss the two type IIB setups.  In section \ref{O3_wflux} we 

5338: show that in orientifolds with $O3/O7$ planes fluxes introduce a superpotential only. 

5339: More intriguingly, we point out in section \ref{O5_wflux} that 

5340: $O5/O9$ setups with background flux in general admit a superpotential as well as 

5341: a massive linear multiplet. Thus, additionally to the kinetic terms studied in 

5342: section \ref{IIB_lin} we find $D-$terms and a direct mass 

5343: term for a linear multiplet \cite{TGL1,mass_tensors}. 

5344: In both IIB orientifold cases the induced potentials 

5345: depend only on some but not all bulk moduli fields in the theory. In order 

5346: to find potentials for the remaining moduli one has to take non-perturbative

5347: contributions into account. In \cite{Witten} it was argued that 

5348: certain D-instantons induce corrections to the superpotential. 

5349: To gain a better understanding of these corrections is subject

5350: of various recent work \cite{non-pert,DSFGK}. 

5351: In section \ref{non-pert_sup} we do only a very moderate step and check if the 

5352: resulting leading order superpotentials are holomorphic in the bulk coordinates. 

5353: Assuming a generic form of such a superpotential one might achieve that all bulk fields 

5354: are stabilized in the vacuum \cite{KKLT,non-pert,DSFGK}. 

5355: 

5356: In type IIA orientifolds the situation is slightly different. As we 

5357: show in section \ref{O6sup} generic NS-NS and R-R background fluxes 

5358: induce a superpotential which depends on all bulk moduli of the theory.

5359: Hence, appropriately chosen background fluxes could stabilize all geometric moduli 

5360: in type IIA orientifolds. Additionally, one can attempt to include corrections due to 

5361: non-perturbative effects. A brief discussion of superpotential contributions due

5362: to world-sheet or D-instantons can be found in section \ref{non-pert_sup}. 

5363: 

5364: 

5365: \section{$O3/O7$ orientifolds: GVW superpotential}

5366: \label{O3_wflux}

5367: 

5368: In this section we study $O3/O7$ orientifolds by also allowing

5369: background three-form fluxes $H_3$ and $ F_3$ on the

5370: Calabi-Yau manifold \cite{Michelson,TV,Mayr,GKP,TGL1}.

5371: The Bianchi identities together with the equations of motion imply

5372: that $H_3$ and $ F_3$ have to be harmonic three-forms. 

5373: In orientifold compactifications they are further constrained 

5374: by the orientifold

5375: projection. {}From \eqref{fieldtransfB}

5376: we see that for the projection given in \eqref{o3-projection}

5377: they both have to be odd under $\sigma^*$

5378: and hence are  parameterized by elements of $H^{(3)}_-(Y)$.\footnote{This

5379: uses the fact that the exterior derivative on $Y$ commutes with $\sigma^*$.}

5380: It is convenient to combine the two three-forms into a complex

5381: $G_3$  according to 

5382: \begin{equation}\label{fluxesB}

5383: G_3 = F_3 -\tau H_3\ , \qquad \tau= C_0 + i e^{- \phi}\ .

5384: \end{equation}

5385: $G_3$ can  be explicitly expanded into a symplectic basis of $H^{(3)}_-$

5386: as 

5387: \beq\label{G3exp}

5388: G_3 = m^{\hat k}\alpha_{\hat k} - e_{\hat k}\beta^{\hat k}\ , \qquad

5389: \hat k = 0,\ldots, h^{(1,2)}\ ,

5390: \eeq

5391: with $2(h^{(1,2)}_-+1)$ complex  flux parameters 

5392: \beq\label{mcomplex}

5393: m^{\hat k} = m^{\hat k}_F -\tau  m^{\hat k}_H\ , \qquad 

5394: e_{\hat k}  = e_{\hat k}^F -\tau  e_{\hat k}^H\ .

5395: \eeq

5396: However, in the following we do not need this explicit expansion and

5397: express our results in terms of $G_3$. 

5398: 

5399: The reduction of the IIB theory is performed by replacing 

5400: \beq

5401:   d\hat B_2 \ \rightarrow \ d\hat B_2 + H_3\ , \qquad d\hat C_2 \ \rightarrow \ d\hat C_2 + F_3\ ,

5402: \eeq

5403: in the field-strengths \eqref{fieldstr}. $H_3$ and $F_3$ are the 

5404: background value of the field strengths 

5405: $\hat F_3$ and $\hat H_3$ but do not effect 

5406: $\hat F_5$ since the only possible terms would be of

5407: the form $H_3 \wedge C_2 $ or  $B_2 \wedge F_3$ 

5408: but both $C_2$ and $B_2$ are projected 

5409: out by the orientifold projection.\footnote{%

5410: We neglect subtleties appearing when $\hat B_2,\hat C_2$ do not arise with 

5411: a derivative. These can be approached along the lines of \cite{DeWGKT}.} 

5412: The only effect of non-trivial background fluxes is the appearance of 

5413: a potential $V$. It is manifestly positive semi-definite and found to be

5414: \cite{TV,GKP,BBHL,DWG}

5415: \beq\label{potential37}

5416:  V= 

5417:   %\frac{18i\ e^{4 \phi}}{\cK^{2}\int \Omega \wedge \bar \Omega }

5418:   e^{K}

5419:   \Big( \int \Omega \wedge \bar G_3 \int \bar \Omega \wedge G_3 

5420:   + G^{k { l}} \int \chi_{ k} \wedge G_3 

5421:   \int \bar \chi_{ l} \wedge \bar G_3 \Big)\ ,

5422: \eeq

5423: where $K$ is given in \eqref{kaehlerpot-O7-1}, $\chi_{k}$ is a basis of $H^{(2,1)}_-$ defined in \eqref{cso} and 

5424: the background flux $G_3$ is defined in \eqref{fluxesB}.

5425: The details of the computation of $V$ can be found in \cite{GKP,TGL1}.

5426: 

5427: 

5428: Strictly speaking the additional term 

5429: $\cL^{(4)}_{\text{top}} \sim 

5430: %\frac{18}{\KK^2} 

5431:   \int_{Y}H_3 \wedge F_3 $

5432: arises in the Kaluza-Klein reduction. 

5433: However, consistency of the compactifications requires its cancellation 

5434: against Wess-Zumino like couplings of the orientifold planes 

5435: to the R-R flux \cite{GKP}. 

5436: 

5437: 

5438: 

5439: Finally, one checks that the potential \eqref{potential37}

5440: can be derived from a superpotential $W$ via the expression

5441: given in \eqref{N=1pot} with vanishing $D$-term $D_\kappa=0$. 

5442: For orientifolds with $c^a=b^a=0$

5443: $W$  was  shown to be \cite{GVW,TV,GKP,BBHL,DWG}

5444: \begin{equation}

5445:   W(\tau,z^k) =  \int_{Y} \Omega \wedge G_3\ .

5446:   \label{superpot}

5447: \end{equation}

5448: This continues to be the correct superpotential

5449: also if  $c^a$ and $b^a$ are in the spectrum \cite{TGL1}, which is due to the 

5450: fact that $K^{\rm Q}$ satisfies the no-scale condition \eqref{NScond}.

5451: This ends our analysis for $O3/O7$ setups. Surprisingly, for $O5/O9$ orientifolds

5452: the computation is more involved and forces us to once more apply and 

5453: extend the linear multiplet techniques developed in chapter \ref{lin_geom_of_M}.

5454: 

5455: 

5456: \section{$O5/O9$ orientifolds: Gaugings and the massive linear multiplet}

5457: \label{O5_wflux}

5458: 

5459: We now turn to the effective action of $O5/O9$ orientifolds with 

5460: background fluxes. In order to detect the 

5461: changes due to this non-trivial background, we proceed as in the $O3/O7$ case and 

5462: first evaluate the field strengths (\ref{fieldstr}) including 

5463: the possibility of background three-form fluxes 

5464: $H_3$ and $F_3$.

5465: Since $\hat B_2$ and hence  $H_3$ is odd

5466: it is again parameterized by $H^{(3)}_-$ while $\hat C_2$ and 

5467: $F_3$ are even and therefore parameterized by $H^{(3)}_+$. 

5468: As a consequence the 

5469: explicit expansions of the background fluxes $H_3$ and $F_3$

5470: are given by

5471: \bea\label{mef}

5472: H_3&=& m^k_H \alpha_k - e_k^H \beta^k\ , \qquad k = 1, \ldots, h^{(2,1)}_-\ ,

5473: \nn\\

5474: F_3 &=& m_F^{\hat \kappa} \, \alpha_{\hat \kappa} 

5475: - e^F_{\hat \kappa}\, \beta^{\hat \kappa} \ ,

5476: \qquad \hat\kappa = 0, \ldots, h^{(2,1)}_+ \ ,

5477: \eea

5478: where the $(m^k_H, e_k^H)$ are $2h^{(2,1)}_-$ constant flux 

5479: parameters determining $H_3$ and 

5480: $(m_F^{\hat \kappa}, e^F_{\hat \kappa})$ are $2h^{(2,1)}_++2$

5481: constant flux 

5482: parameters corresponding to $F_3$. 

5483: Inserting \eqref{exp-B}, \eqref{expO5} and  \eqref{mef} into \eqref{fieldstr}

5484: we obtain

5485: \bea

5486:  \hat H_3 &=& db^a \wedge \omega_a + m^k_H \alpha_{k} - 

5487:              e_{k}^H \beta^{k}\ ,\qquad

5488:  \hat F_3 \ = \ dC_2+dc^\alpha \wedge \omega_\alpha + F_3\ ,\nonumber \\

5489:  \hat F_5 &=& dD_2^a \wedge \omega_a + \tilde F^{k} \wedge \alpha_{k}

5490:                - \tilde G_{k}\wedge \beta^{k} 

5491:                + d\rho_a \wedge \tilde \omega^a  \\ 

5492:            && - db^a \wedge C_2 \wedge \omega_a - c^\alpha db^a \omega_a \wedge 

5493:                \omega_\alpha\ ,\nn

5494: \eea

5495: where we defined

5496: \beq\label{Fcech}

5497: \tilde F^{k}= dV^{k} - m^{k}_H C_2\ , \qquad

5498: \tilde G_{k}=dU_{k} - e_{k}^H C_2\ .

5499: \eeq 

5500: As in section \ref{IIB_orientifolds} the self-duality condition on $\hat F_5$

5501: is imposed by a Lagrange multiplier \cite{DallAgata} and we eliminate 

5502: $D^a_{2}$ and $U_{k}$ by inserting their equations

5503: of motion into the action.

5504: After Weyl rescaling the four-dimensional metric with a factor $\KK/6$ 

5505: the ${N}=1$ effective action reads 

5506: \bea\label{actiono5}

5507:  S^{(4)}_{O5/O9} &=& \int -\tfrac{1}{2}R*\mathbf{1}-

5508:   G_{\kappa\lambda} \; dz^{\kappa} \wedge *d\bar z^{\lambda}

5509:   -G_{\alpha \beta} \; dv^\alpha \wedge *dv^\beta - G_{ab}\; db^a \wedge * db^b \nn \\

5510:   &&   - \tfrac{e^{2D}}{6} \cK\, G_{\alpha \beta}\; dc^\alpha \wedge * dc^\beta

5511:   - \tfrac{e^{-2D}}{24}  \cK\, dC_2 \wedge * dC_2 -\tfrac{1}{4} dC_2 \wedge (\rho_a db^a - b^a d\rho_a)  \nonumber \\ 

5512:   && 

5513:   - dD \wedge * dD 

5514:   - \tfrac{3e^{2D}}{8\KK} G^{ab}(d \rho_a - \KK_{ac\alpha} c^\alpha db^c)

5515:   \wedge *(d \rho_b - \KK_{bd\beta} c^\beta db^d)- V*\mathbf{1}\nn \\[.1cm]

5516: && 

5517:  + \tfrac{1}{4} \text{Re}\; \mathcal{M}_{k l}\; \tilde 

5518:   F^{k} \wedge \tilde F^{l} + 

5519:   \tfrac{1}{4} \text{Im}\; \mathcal{M}_{k l}\; 

5520:   \tilde 

5521:   F^{k} \wedge * \tilde F^{l} +

5522:   \tfrac{1}{4} e_{k} (dV^{k} +

5523:   \tilde F^{k})\wedge C_2\ ,\nn \\

5524: \eea

5525: where 

5526: \begin{align}\label{pot5}

5527: V &=

5528:   \frac{18i\ e^{4\phi}}{\cK^2 \int \Omega \wedge \bar \Omega }

5529:   \left( \int \Omega \wedge F_3 \int \bar \Omega \wedge F_3 

5530:   + G^{\kappa \lambda} \int \chi_{\kappa} \wedge F_3 

5531:   \int \bar \chi_{\lambda} \wedge F_3 \right) \\

5532:   &- 

5533:    \tfrac{9\, e^{2\phi}}{\KK^2} \Big[

5534:   m^{k}_H\, (\text{Im} \MM)_{ k l}\,   m^{l}_H +

5535:   \big(e_{k}^H-(m_H \text{Re} \MM)_{k}\big)   \big(\text{Im} \MM\big)^{-1  k l} 

5536:   \big(e_{l}^H-(m_H \text{Re} \MM)_{l}\big) \Big]\ . \nonumber

5537: \end{align}

5538: The derivation of this potential can be found in ref. \cite{TGL1}.\footnote{Note that in this class of 

5539: orientifolds the topological term 

5540: $\int_YH_3 \wedge F_3$ vanishes since there is no intersection between 

5541: $H^{(3)}_+$ and $H^{(3)}_-$. Thus strictly speaking background

5542: D-branes have to be included in order to satisfy the tadpole cancellation

5543: condition.} 

5544: 

5545: The action \eqref{actiono5} has the standard one-form gauge invariance

5546: $V^k\to V^k+d\Lambda^k_0$ 

5547: but due to the modification in \eqref{Fcech} 

5548: also a modified (St\"uckelberg) two-form gauge invariance given by

5549: \beq\label{2gauge}

5550: C_2\to C_2 +d\Lambda_1\ ,\qquad V^k\to V^k + m^k_H\Lambda_1\ .

5551: \eeq

5552: Thus for $m^k_H\neq 0$ one vector can be set to zero by an appropriate 

5553: gauge transformation. 

5554: This is directly related to the fact that  \eqref{actiono5}

5555: includes mass terms  proportional to $m^k_H$ for $C_2$ arising from 

5556: \eqref{Fcech}. In this case gauge invariance requires

5557: the presence of Goldstone degrees of freedom which

5558: can be `eaten' by $C_2$.\footnote{Exactly the same situation

5559: occurs in Calabi-Yau compactifications of type IIB

5560: with background fluxes where both $B_2$ and $C_2$

5561: can become massive \cite{LM}.}

5562: Finally note that the last term in \eqref{actiono5} also 

5563: includes a standard $D=4$ Green-Schwarz term $F^k\wedge C_2$.

5564: 

5565: 

5566: \subsection{Vanishing magnetic fluxes $m^k_H=0$}

5567: 

5568: The next step is to show that $S^{(4)}_{O5/O9}$ is consistent 

5569: with the constraints of $N=1$ supergravity. However, due to the

5570: possibility of $C_2$ mass terms this is not completely

5571: straightforward. A massive $C_2$

5572: is no longer dual to a scalar but rather to a vector.

5573: We find it more convenient to keep the 

5574: massive tensor in the spectrum and discuss 

5575: the $N=1$ constraints in terms of a massive linear multiplet.

5576: Before doing so, let us first discuss

5577: the situation $m^k_H= 0$ where $\tilde F^{l}= F^{l}$ holds.

5578: In this case the $C_2$ remains massless 

5579: and  can be dualized to a scalar field $h$ which together with

5580: the dilaton $ \phi$ combines to form a chiral multiplet 

5581: $(\phi, h)$.

5582: Using the standard dualization procedure (see section \ref{revIIB})

5583: one obtains the effective action \eqref{S_scalarO5} plus the 

5584: potential $V$ given in \eqref{pot5} evaluated at $m^k_H=0$.

5585: Furthermore, due to electric NS-NS fluxes the scalar $h$ is gauged 

5586: and we have to replace in \eqref{S_scalarO5}

5587: \beq \label{hcov}

5588:   dh\quad  \rightarrow \quad Dh=d h - e_k^H V^k\ .

5589: \eeq

5590: Hence, $h$ couples non-trivially to the gauge fields as a direct consequence

5591: of the Green-Schwarz coupling $F^k\wedge C_2$

5592: in \eqref{actiono5}. In the dualized

5593: action the scalar $h$ then is charged 

5594: under the $U(1)$ gauge transformation $h\to h + e_k^H \Lambda_0^k$ with $V^k \to V^k + d\Lambda_0^k$. 

5595: Note that the gauge charges are set by the electric fluxes.

5596: 

5597: The K\"ahler potential \eqref{O5-Kaehlerpot} with chiral coordinates \eqref{Kcoord} 

5598: and the gauge-couplings \eqref{fholo} remain unchanged for the theory with 

5599: $m^k_H=0$. However, due to the non-trivial electric NS-NS fluxes the 

5600: covariant derivative of $h$ given in \eqref{hcov} translates into the 

5601: covariant derivative $DS = dS - i  e_{k} V^{k}$. It remains to cast 

5602: the potential $V$ given in \eqref{pot5}, evaluated at $m^k_H=0$, into 

5603: the standard $N=1$ supergravity form \eqref{N=1pot}.

5604: {}From eq.\ \eqref{hcov}

5605: we see that the axion is charged 

5606: and as a consequence we

5607: expect a non-vanishing $D$-term in the potential. Recall

5608:  the general formula

5609: for the $D$-term \cite{WB}

5610: \beq

5611: K_{I\bar J} \bar X^{\bar J}_k = i \partial_I D_k \ ,

5612: \eeq

5613: where $X^{I}$ is the Killing vector of the $U(1)$ gauge transformations 

5614: defined as $\delta M^I = \Lambda^k_0 X_k^J \partial_J M^I$.

5615: Inserting \eqref{O5-Kaehlerpot} 

5616: and \eqref{Kcoord} we obtain 

5617: \begin{equation}

5618:   D_k = - e_{k}^H\, \frac{\partial K}{\partial \bar S} = 

5619:   3\, e_{k}^H\, e^{\phi}\cK^{-1} \ .

5620: \end{equation}

5621: Using also \eqref{gauge-couplingsO3} we arrive at the $D$-term contribution 

5622: to the potential 

5623: \begin{eqnarray}

5624:   \tfrac{1}{2}   (\text{Re}\; f)^{-1\ kl} D_k D_l = -

5625:   \tfrac{9}{\cK^2} e^{2\phi} \; e_{k}^H\, (\text{Im}\; \MM)^{-1\ kl}\,

5626:   e_{l}^H\ , \label{D-term}

5627: \end{eqnarray}

5628: which indeed reproduces the last term in \eqref{pot5} for 

5629: $m^k_H=0$.

5630: 

5631: The first term in \eqref{pot5} arises from the  superpotential

5632: \begin{equation}\label{W5}

5633:   W= \int_{Y} \Omega \wedge F_3\ ,

5634: \end{equation}

5635: which follows from a calculation analog to the $O3/O7$ case \cite{TGL1}.

5636: It is interesting  that for this class of orientifolds

5637: the RR-flux $F_3$ results in a contribution to the superpotential while

5638: the NS-flux $H_3$ contributes instead to a $D$-term.

5639: 

5640: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

5641: %

5642: \subsection{Non-vanishing magnetic fluxes $m^k_H\neq 0$}

5643: 

5644: 

5645: Let us now turn to the case where both electric and magnetic fluxes are

5646: non-zero and the two-form $C_2$ is massive.

5647: In this case $C_2$ is dual to a massive vector or equivalently the massive

5648: linear multiplet is dual to massive vector multiplet.

5649: Here we do not discuss this duality but instead show how the couplings

5650: of a massive linear multiplet is consistent with the action 

5651: \eqref{actiono5} \cite{mass_tensors}.

5652: 

5653: In section \ref{IIB_lin} we already examined the kinetic terms and couplings

5654: for the $O5/O9$ theory in the presence of one tensor multiplet $L=(\phi,C_2)$. 

5655: We found that they are determined in terms of the generalized K\"ahler potential

5656: and the function $F$ both given in \eqref{KpotL}.

5657: Let us now briefly discuss the situation of a massive

5658: linear multiplet coupled to $N=1$ vector- and chiral multiplets.

5659: For simplicity we discuss the situation in flat space 

5660: and do not couple the massive linear multiplet to supergravity.

5661: However, we expect our results to generalize to the 

5662: supergravity case. More details can be found in \cite{GGRS,mass_tensors}.

5663: 

5664: As we already said, a linear multiplet $L$ contains a real scalar (also denote by $L$)

5665: and the field strength of a two-form $C_2$ as bosonic components. However, 

5666: it does not contain the two-form itself

5667: which instead is a member of the chiral `prepotential' $\Phi$ defined

5668: as\footnote{We suppress the spinorial indices and use the convention

5669: $D\Phi \equiv D^\alpha\Phi_\alpha$, 

5670: $\bar D\bar\Phi \equiv \bar D^{\dot{\alpha}}\bar\Phi_{\dot{\alpha}}$.}

5671: \beq

5672: L= D\Phi +\bar D \bar \Phi\ , \qquad \bar D\Phi = 0\ .

5673: \eeq

5674: This definition solves the constraint \eqref{linearc} (in flat space).

5675: The kinetic term for $L$ (or rather for $\Phi$) is given in 

5676: \eqref{actionL} and a mass-term can be added via the 

5677: chiral integral

5678: \beq\label{Lmasst}

5679: \cL_{m} = \tfrac14\int d^2\theta \Big[

5680: f_{kl}(N) (W^k - 2i m^k_H\Phi)(W^l - 2i m^l_H\Phi)

5681: + 2 e_k^H (W^k - i m^k_H\Phi)\Phi\Big] + {\rm h.c.}\ ,

5682: \eeq

5683: where $W^k= -\tfrac14 \bar D^2 DV^k$ are the chiral field strengths supermultiplets

5684: of the vector multiplets $V^k$ and $f_{kl}(N)$ are the gauge kinetic function

5685: which can depend holomorphically on the chiral multiplets $N$.

5686: $(m^k_H,e_k^H)$ are constant parameters which will turn out 

5687: to correspond to the flux parameters defined in \eqref{mef}.

5688: The Lagrangian \eqref{Lmasst} is invariant under the standard

5689: one-form gauge invariance $V^k\to V^k +\Lambda_0^k + \bar \Lambda_0^k$

5690: ($\Lambda_0^k$ are chiral superfields)

5691: which leaves both $W^k$ and $\Phi$ invariant.

5692: In addition \eqref{Lmasst} has a two-form gauge invariance 

5693: corresponding to \eqref{2gauge} given by

5694: \beq\label{linearg}

5695: \Phi \to \Phi +\tfrac{i}8 \bar D^2 D \Lambda_1\ ,\qquad

5696: V^k\to V^k + m^k_H \Lambda_1\ ,

5697: \eeq

5698: where $\Lambda_1 $ now is a real superfield. 

5699: {}From \eqref{linearg} we see that one entire vector multiplet

5700: can be gauged away and thus plays the role of the Goldstone degrees

5701: of freedom which are `eaten' by the massive linear multiplet.

5702: 

5703: In components one finds the bosonic action

5704: \beq

5705: \cL_{m} = -\tfrac{1}{2} \text{Re} f_{k l}\; \tilde 

5706:   F^{k} \wedge  * \tilde F^{l} -

5707:   \tfrac{1}{2} \text{Im} f_{k l}\; 

5708:   \tilde 

5709:   F^{k} \wedge \tilde F^{l} +

5710:   \tfrac{1}{4} e_{k} (dV^{k} +

5711:   \tilde F^{k})\wedge C_2 - V*\mathbf{1} \ ,

5712: \eeq

5713: where $\tilde F^{l}$ is defined exactly as in \eqref{Fcech}

5714: and the potential $V$ receives

5715: two distinct contributions

5716: \beq

5717: V=

5718: \tfrac{1}{2}\, 

5719: (\text{Re} f)^{-1 kl} D_{k} D_{l} + 2\, m^k_H\text{Re} f_{kl}\, m^l_H\, L^2\ ,

5720: \qquad

5721: D_{k} = \big(e_k^H + 2\,\text{Im}f_{kl}\, m^l_H \big)\, L \ .

5722: \eeq

5723: The first term arises from eliminating the $D$-terms in 

5724: the $U(1)$ field strength $W^k$ while the second term is a 

5725: `direct' mass term for the scalar $L$.\footnote{Note that this second term

5726: is a contribution to the potential which is neither a $D$- nor an

5727: $F$-term but instead a `direct' mass term whose presence is enforced

5728: by the massive two-form.}

5729: Inserting the $D$-term yields a second contribution to the mass term

5730: and one obtains altogether

5731: \bea

5732: V&=&

5733: \tfrac12 \big[\big(e_k^H +2\text{Im} f_{kp}\, m^p \big)

5734: (\text{Re} f)^{-1 kl} \big( e_l^H +2 \text{Im} f_{lr}\, m^r\big)

5735: +4 m^k_H\, \text{Re} f_{kl}\, m^l_H\big] L^2\ . \qquad 

5736: \eea 

5737: Using \eqref{Lzeta_O59}  and \eqref{gauge-couplingsO3} 

5738: this precisely agrees with the second term in the potential \eqref{pot5}.

5739: 

5740: As before the first term in \eqref{pot5}  can be derived from  the superpotential

5741: \eqref{W5}. This ends our discussion of 

5742: type IIB orientifolds in a general NS-NS and R-R flux background. 

5743: As we have seen, switching on fluxes yields a potential for only part of 

5744: the moduli fields. This changes in IIA orientifolds to which we will 

5745: turn now.

5746: 

5747: 

5748: \section{$O6$ orientifolds: Flux superpotentials}

5749: \label{O6sup}

5750: 

5751: In this section we derive the effective action of type IIA orientifolds

5752: in the presence of background fluxes. 

5753: For standard $N=2$ Calabi-Yau compactifications of type IIA a

5754: similar analysis is carried out in refs.\ \cite{LM,KachruK}.

5755: In order to do so

5756: we need to start from the ten-dimensional action of massive 

5757: type IIA supergravity which differs from the action \eqref{10dact} in

5758: that the two-form $\hat B_2$ is massive. In the 

5759: Einstein frame it is given by \cite{Romans}

5760: \bea \label{10dactm}

5761:   S^{(10)}_{MIIA} &=& \int -\tfrac{1}{2}\hat R*\mathbf{1} -\tfrac{1}{4} d\hat \phi\wedge * d\hat \phi

5762:   -\tfrac{1}{4} e^{-\hat \phi}\hat H_3 \wedge *\hat H_3 

5763:   -\tfrac{1}{2} e^{\frac{3}{2} \hat \phi}\hat F_2 \wedge *\hat F_2 \nn \\

5764:   && -\tfrac{1}{2} e^{\frac{1}{2} \hat \phi}\hat F_4 \wedge *\hat F_4 

5765:   -\tfrac{1}{2} e^{\frac{5}{2} \hat \phi}\, (m^0)^2 * \mathbf{1} + \cL_{\rm top}\ ,

5766: \eea

5767: where

5768: \bea

5769:   \cL_{\rm top}&=& -\tfrac{1}{2}\Big[ \hat B_2 \wedge d\hat C_3 \wedge d\hat C_3\  

5770:                                    -(\hat B_2)^2 \wedge d\hat C_3 \wedge d\hat C_1

5771:                                    + \tfrac{1}{3}(\hat B_2)^3 \wedge (d\hat C_1)^2 \nn \\

5772:                & &                 - \tfrac{m^0}{3}(\hat B_2)^3 \wedge d\hat C_3

5773:                                    + \tfrac{m^0}{4}(\hat B_2)^4 \wedge d\hat C_1

5774:                                    + \tfrac{(m^0)^2}{20}(\hat B_2)^5 \Big]\ ,

5775: \eea

5776: and the field strengths are defined as

5777: \bea \label{defHFF}

5778:   \hat H_3 = d \hat B_2\ , \quad \hat F_2 = d\hat C_1+m^0 \hat B_2\ , \quad 

5779:   \hat F_4 = d\hat C_3 - \hat C_1 \wedge \hat H_3-\tfrac{m^0}{2}(\hat B_2)^2\ .

5780: \eea

5781: Compared to the analysis of the previous section we now include

5782: non-trivial background fluxes of the field strengths

5783: $F_2$, $H_3$ and $F_4$ on the Calabi-Yau orientifold.

5784: We keep the Bianchi identity and the equation of motion intact 

5785: and therefore expand $F_2$, $H_3$ and $F_4$

5786: in terms of  harmonic forms compatible with the orientifold

5787: projection. From \eqref{fieldtransf} we infer that $F_2$ is expanded in

5788: harmonic forms of $H^{2}_-(Y)$, 

5789: $H_3$ in harmonic forms of $H^3_{-}(Y)$ and $F_4$ in harmonic forms

5790: of $H^{4}_+(Y)$.\footnote{As we observed in the previous section

5791: there is no $\hat C_1$

5792: due to the absence of one-forms on the orientifold. 

5793: Nevertheless its field strength $F_2$ 

5794: can be non-trivial on the orientifold since $Y$ generically possesses

5795: non-vanishing harmonic two-forms.}

5796:  Explicitly the expansions read 

5797: \bea \label{fluxesA}

5798:  H_3\, =\, q^\lambda \alpha_\lambda - p_k\, \beta^k\ , \quad   F_2\, =\, -m^a \omega_a\ , \quad 

5799:   F_4\, =\, e_a\, \tilde \omega^a\ ,

5800: \eea

5801: where $(q^\lambda,p_k)$ are $h^{(2,1)}+1$ real NS flux parameters 

5802: while $(e_a,m^a)$ are $2h^{1,1}_-$ real RR flux parameters.

5803: %By $m^0$ we denote the mass parameter $m$ of the massive type IIA theory which 

5804: %we combine with the $h^{1,1}_-$ magnetic RR fluxes into a

5805: %vector $m^{\hat a}=(m^0,m^a)$.

5806: The harmonic forms $(\alpha_\lambda, \beta^k)$ are the elements of the real

5807: symplectic basis of $H^3_-$ introduced in \eqref{sp_alpha-beta}. 

5808: The basis $\tilde \omega^a$ of

5809: $H^{(2,2)}_+$ is defined to be the dual basis of $\omega_a$ while the

5810: basis $\tilde \omega^\alpha$ denotes a basis of $H^{(2,2)}_-$ dual to $\omega_\alpha$. 

5811: 

5812: Inserting \eqref{expJB}, \eqref{form-exp} and \eqref{fluxesA} into

5813: \eqref{defHFF} we arrive at

5814: \bea \label{fieldst}

5815:   \hat H_3 &=& db^a\wedge \omega_a + q^\lambda \alpha_\lambda - p_k\, \beta^k\ ,  \qquad \qquad 

5816:   \hat F_2 = (m^0 b^a + m^a)\, \omega_a\ ,\\

5817:   \hat F_4 &=& dC_3 + dA^\alpha \wedge \omega_\alpha 

5818:   + d\xi^k \wedge \alpha_k - 

5819:             d\tilde \xi_\lambda \wedge \beta^\lambda +  

5820:    \big(b^a m^b  -\tfrac12 m^0 b^a b^b\big)\, 

5821: \cK_{abc}\tilde \omega^c + e_a\, \tilde \omega^a\ , \nn

5822: \eea

5823: where we have used $\omega_a \wedge \omega_b = \cK_{abc}\, \tilde \omega^c$.

5824: Now we repeat the KK-reduction of the previous section using the

5825: modified field strength 

5826: \eqref{fieldst} and the action \eqref{10dactm} instead of \eqref{10dact}.

5827: This results in%

5828: \footnote{The action $S^{(4)}_{O6}$ is given in \eqref{act1}. However, due to the fact that 

5829: we perform the Kaluza-Klein reduction in the generic basis introduced in \eqref{sp_alpha-beta} the kinetic 

5830: terms for $\tilde \cM^{\rm Q}$ are replaced by \eqref{act2}. }

5831: \beq\label{Sflux}

5832: S^{(4)} = S^{(4)}_{O6} - \int  \tfrac{g}{2}\, d\cc_3 \wedge * d\cc_3 + {h}\, d\cc_3 +

5833:  U * \mathbf{1}\ ,

5834: \eeq

5835: where $S^{(4)}_{O6}$ is given in \eqref{act1}.

5836: $\cc_3$ 

5837: is the four-dimensional part of the ten-dimensional 

5838: three-form $\hat C_3$ defined in \eqref{form-exp} and 

5839: its couplings to the scalar fields are given by

5840: \beq

5841:   g =  e^{-4 \phi} \left(\tfrac{\cK}6

5842:        \right)^3\ , \qquad  h = e_a b^a + \tilde \xi_\lambda q^\lambda 

5843:   - \xi^k p_k + \tfrac{1}{2}\R \cN_{0 \ah}\, m^\ah \ ,

5844: \eeq

5845: where we denoted $m^\ah=(m^0,m^a)$. The potential term $U$ of \eqref{Sflux}

5846: is given by

5847: \beq \label{U-pot}

5848:   U = \tfrac{9}{\cK^2} e^{2\phi} \int_Y H_3 \wedge * H_3

5849:         -  \tfrac{18}{\cK^2} e^{4\phi} \I \cN_{\ah \bh}\, m^\ah m^\bh

5850:          +\tfrac{ 27 } {\cK^3} e^{4\phi} G^{ab}(e_a - \R \cN_{a\ah}\, m^\ah)(e_b - \R \cN_{b\bh}\, m^\bh)\ . 

5851: \eeq

5852: The matrix $\cN_{\ah \bh} (t,\bar t)$ is defined to be the corresponding 

5853: part of the $N=2$ gauge-coupling matrix \eqref{def-cN} 

5854: restricted to $\tilde \cM^{\rm SK}$ by applying \eqref{van-int} and \eqref{splitmetr}. 

5855: 

5856: In four space-time dimensions 

5857: $\cc_3$ is dual to a constant which plays the role of

5858: an additional electric flux $e_0$ in complete analogy with the

5859: situation in $N=2$ discussed in \cite{LM}.

5860: Eliminating $\cc_3$ in favor of $e_0$ by following \cite{LM} or \cite{BW}

5861: the potential takes the form \cite{TGL2} 

5862: \beq \label{V-pot1}

5863:    V  = \tfrac{9}{\cK^2} e^{2\phi} \int H_3 \wedge * H_3 

5864:       - \tfrac{18}{\cK^2} e^{4\phi} (\tilde e_\ah - \cN_{\ah \ch}\, m^\ch) (\I \cN)^{-1\, \ah \bh}

5865:                                           (\tilde e_\bh -\bar \cN_{\bh

5866:         \ch}\, m^\ch)\ ,

5867: \eeq

5868: where we introduced the shorthand notation 

5869: $\tilde e_\ah=(e_0 + \xi_\lambda q^\lambda-\xi^\kh p_\kh,e_a)$ and $m^\ah=(m^0,m^a)$. 

5870: Note that in the presence of NS flux 

5871: one can absorb $e_0$ by shifting the fields 

5872: $\xi,\tilde \xi$. This corresponds to adding an integral form to 

5873: $\CC_3$ as carefully discussed in \cite{BW}. 

5874: However, for the discussion of mirror symmetry it is more convenient to

5875: keep the parameter $e_0$ explicitly in the action.

5876: 

5877: 

5878: In order to establish the consistency with $N=1$ supergravity

5879: one needs to rewrite $V$ given in \eqref{V-pot1} in terms of

5880: \eqref{N=1pot} or in other words we need express $V$ in terms 

5881: of a superpotential $W$ and appropriate $D$-terms. 

5882: From \eqref{Sflux} we infer that turning on fluxes does not 

5883: charge any of the fields and therefore all $D$-terms have to vanish. 

5884: In \cite{TGL2} it was checked that the potential \eqref{V-pot1} can be entirely expressed in

5885: terms of the superpotential 

5886: \beq \label{superpot1}

5887:   W\ =\  W^{\rm Q}(N,T) + W^{\rm K}(t)\ ,

5888: \eeq 

5889: where 

5890: \bea \label{superpot2}

5891:   W^{\rm Q}(N^k,T_\lambda)& =& \int_Y \Omegac \wedge H_3\ =\ 

5892:         - 2N^k p_k - i T_\lambda q^\lambda\ , \\

5893:   W^{\rm K}(t^a) &=& e_0 + \int_Y \Jc \wedge F_4 - \tfrac{1}{2} \int_Y \Jc \wedge \Jc \wedge F_2 

5894:        - \tfrac{1}{6} m^0 \int_Y \Jc \wedge \Jc \wedge \Jc\ ,

5895: \nn\\

5896: &=& e_0 + e_a t^a + \tfrac{1}{2}\cK _{abc} m^a t^bt^c - \tfrac{1}{6} m^0  \cK _{abc} t^a t^bt^c\nn\ ,

5897: \eea

5898: and $\Omegac$ and $\Jc$ are defined in \eqref{N=1coords}. Using the 

5899: definitions \eqref{symp-form} and \eqref{symp-formodd} of the skew-symmetric products $\big<\cdot,\cdot\big>$ 

5900: for even and odd forms $W$ is rewritten as 

5901: \beq

5902:   W = \big<e^{\Jc}, F \big> + \big<\Omegac,H_3 \big>\ , \qquad F = m^0 - F_2 - F_4 + F_6\ , 

5903: \eeq

5904: where we have defined $F_6$ via $e_0= \int_Y F_6$. 

5905: We see that the superpotential is the sum of two terms.

5906: $W^{\rm Q}$  depends on the NS fluxes $(p_k,q^\lambda)$ of $H_3$ and the 

5907: chiral fields $N^k,T_\lambda$ parameterizing the space $\tilde \cM^{\rm Q}$. 

5908: $W^{\rm K}$ depends on the RR fluxes $(e_{\hat a}, m^{\hat b})$  

5909: of $F_2$ and $F_4$ (together with $m^0$ and $e_0$) and

5910: the complexified K\"ahler deformations $t^a$ parameterizing 

5911: $\cM^{\rm SK}$. 

5912: We see that contrary to the type IIB case both types of moduli, 

5913: K\"ahler and complex structure deformations appear in the superpotential

5914: suggesting the possibility that all moduli can be fixed in this set-up.

5915: This was resently shown to be the case in refs.~\cite{VZ,DeWGKT}.

5916: 

5917: 

5918: Let us end this section by comparing the R-R superpotentials of type IIA 

5919: and type IIB orientifolds. Recall that for 

5920: both IIB orientifold setups R-R fluxes induce superpotentials \eqref{superpot} and \eqref{W5} 

5921: holomorphic in the complex structure deformations $z$. Hence, we compare 

5922: \beq

5923:   W_A(t) = \big<e^{\Jc}, F \big> \ , \qquad W_B(z) = \big<\Omega, F_3\big>\ ,

5924: \eeq  

5925: where the skew-products are defined in \eqref{symp-form} and \eqref{symp-formodd}.

5926: As just discussed $F$ depends on $2h^{(1,1)}_- + 2$ RR fluxes $(e_\ah,m^\ah)$.

5927: To count the flux parameters labeling $F_3$ recall that it transforms differently in the

5928: two IIB orientifolds. $F_3$ sits in $H^{3}_-(\tilde Y)$ and is determined in terms of

5929: $2h^{(2,1)}_-+2$ real flux parameters for the $O3/O7$ case and sits in $H^{3}_+(\tilde Y)$ 

5930: depending on $2h^{(2,1)}_+ + 2$ real flux parameters

5931: for the $O5/O9$ case. Therefore, the number of flux parameters matches when choosing mirror 

5932: involutions satisfying \eqref{matchchohm}. Exchanging \cite{FMM}

5933: \beq

5934:   e^{\Jc}(t)\ \leftrightarrow\ \Omega(z) \ , \qquad F\  \leftrightarrow\ F_3\ ,

5935: \eeq

5936: as in equation \eqref{pure-spinor-map} the two superpotentials $W_A(t)$ and $W_B(z)$ get identified. 

5937: In $N=2$ the mirror identification of the complex structure moduli space $\cM^{\rm cs}$ with the complexified K\"ahler 

5938: moduli space $\cM^{\rm ks}$ can be used to calculate world-sheet instanton corrections to $\cM^{\rm ks}$.

5939: It would be interesting to generalize this to $N=1$ orientifold theories which allow additionally for 

5940: non-oriented world-sheets as discussed at the end of section \ref{IIA_orientifolds}.

5941: In addition to world-sheet instantons also certain D-instantons induce correction terms 

5942: to the superpotential. We will end this chapter by a few comments on their 

5943: generic structure. 

5944: 

5945: \section{D-instanton corrections to the superpotentials}

5946: \label{non-pert_sup}

5947: 

5948: Let us close this chapter by briefly discussing possible D-instanton

5949: corrections to the superpotentials \eqref{superpot}, \eqref{W5} and \eqref{superpot1}. 

5950: They can arise from wrapping $D(p-1)$-branes around $p$-cycles 

5951: $\Sigma_p$ \cite{BBS}. In addition to corrections of the K\"ahler potential 

5952: D-instantons induce extra superpotential terms \cite{Witten}. These 

5953: depend on brane moduli as well as bulk fields and found recent phenomenological 

5954: application in moduli stabilization \cite{KKLT, non-pert, DSFGK}. It would be interesting to fully 

5955: incorporate these effects and to understand the additional contributions due to 

5956: non-orientable world-volumes. First steps into this direction are done in the 

5957: recent works \cite{non-pert, DSFGK}.

5958: In this section we will take only a very moderate step and 

5959: apply the calibration conditions to show that the D-brane action 

5960: becomes linear in the bulk fields. This ensures holomorphicity of the induces 

5961: superpotential terms when expressed in the proper K\"ahler variables of the 

5962: respective orientifold setup. 

5963: 

5964: To make this more precise, recall that any correlation function

5965: is weighted by the string-frame 

5966: world-volume action of the wrapped Euclidean $D(p-1)$-branes

5967: and thus includes a factor $e^{-S_{D(p-1)}}$ where

5968: \beq \label{instact}

5969:   S_{D(p-1)} = i\mu_{p}\, 

5970:   \int_{\cW_{p}}\Big(d^{p} \lambda\ e^{- \hat \phi}  \sqrt{\det\big({\varphi^*(\hat g+ \hat B_2) +  \ell F}\big)} 

5971:              - i \Em^*\Big(\sum_q \hat C_{q} \wedge e^{-\hat B_2}\Big) \wedge e^{\ell \FD}\Big)\ .

5972: \eeq 

5973: where $\ell=2\pi \alpha'$. This is the Euclidean analog of the 

5974: Dirac-Born-Infeld action  \eqref{DBI} plus the Chern-Simons action \eqref{CSaction}.  

5975: $\cW_p$ is the world-volume of the $D(p-1)$-brane and  $\varphi^*$ is the pull-back

5976: of the map $\varphi$ which embeds $\cW_p$ into Calabi-Yau orientifold $Y$, 

5977: $\varphi:\cW_p \hookrightarrow Y$.

5978: We have chosen the R-R charge $\mu_p$ equal to the tension since 

5979: the wrapped  $D(p-1)$-branes must be BPS in order to preserve $N=1$ supersymmetry.

5980: In fact, as we already discussed in section \eqref{D-branes} there are 

5981: additional condition arising from the requirement that the $Dp$-branes  preserves 

5982: the same supersymmetry that is left intact

5983: by the orientifold projections. This in turn implies 

5984: that $O3/O7$ orientifolds can admit 

5985: corrections from $D3$ instantons, $O5/O9$ setups from $D1$ and $D5$ instantons and 

5986: $O6$ setups from $D2$ instantons. Moreover, these have to be calibrated 

5987: with respect to the same forms as the internal parts of the orientifold planes.

5988: 

5989: The calibration conditions for Euclidean $D(p-1)$-branes

5990: in a Calabi-Yau manifold have been derived in refs.\ \cite{BBS,MMMS}.

5991: Let us first apply their results to type IIA orientifolds with $O6$ planes.

5992: Recall that the unbroken supercharge has to be some linear combination  

5993: $\epsilon=a^+ \epsilon_+ + a^- \epsilon_-$ of the two covariantly 

5994: constant spinors $\epsilon_+$ and $\epsilon_-$ of the 

5995: original  $N=2$ supersymmetry. Let us denote the relative phase 

5996: of $a^+$ and $a^-$ by $a^-/a^+=-ie^{i\theta_{D2}}$ while the 

5997: absolute magnitude can be fixed by the normalization of $\Omega$.

5998: {} As forms $J$ and $\Omega$ have to obey the condition 

5999: \beq

6000:    J \wedge J \wedge J = \tfrac{3i}{2}e^{-2U} \Omega \wedge \bar \Omega\ 

6001: \eeq 

6002: at every point in the moduli space. Note however, that $J$ depends on 

6003: K\"ahler structure deformations $v^a$ while $\Omega$ is a function of 

6004: the complex structure deformations $q^K$. Hence, $e^U$ is a non-trivial function 

6005: of $v^a$ and $q^K$ and from $\int J^3 =\frac{3i}{2}e^{-2U}\int \Omega\wedge\bar\Omega$

6006: one infers 

6007: \beq\label{Omeganorm}

6008:  % \gamma_{ijk}\, \epsilon_+ =e^{-U} \Omega_{ijk} \epsilon_-\ , \qquad  

6009: e^{U}=\sqrt{2}\, e^{\frac{1}{2}(K^{\rm K}-\Kcs)}\ ,

6010: \eeq

6011: where K\"ahler potential $K^{\rm K}(t)$

6012: is given in \eqref{Kks} while $\Kcs(q)$ is the restriction of the K\"ahler 

6013: potential \eqref{csmetric} to the real slice $\cM^{\rm cs}_\bbR$.

6014: The existence of $\epsilon$ imposes constraints

6015: on the map $\varphi$. These BPS conditions read  \cite{BBS,MMMS}

6016: \beq\label{sLagr-cond}

6017:   \varphi^*(\Omega)\ =\ e^{U+i\theta_{D2}} \sqrt{\det\big({\varphi^*(\hat g+ \hat B_2) +  \ell F_2}\big)} 

6018:                         d^3 \lambda\ , \qquad 

6019:   \varphi^*\Jc + i 2\pi \alpha' F_2\ =\ 0\ ,

6020: \eeq 

6021: where $\Jc$ is given in \eqref{def-t}.

6022:  The second condition in \eqref{sLagr-cond} enforces 

6023: $\varphi^*(J)=0$ as well as $\varphi^*\hat B_2 + \ell F_2 =0$, such that the first equation 

6024: simplifies to 

6025: \beq \label{cal1}

6026:   \varphi^*\R( e^{-i\theta_{D2}}\Omega)\ =\ e^U \sqrt{\det\big(\varphi^*\hat g\big)} d^3 \lambda\ , \qquad 

6027:   \varphi^*\I( e^{-i\theta_{D2}}\Omega)\ =\ 0\ ,

6028: \eeq

6029: where we have used that the volume element on $\cW_3$ is real. For vanishing $F$ these conditions 

6030: coincide with those displayed in equation \eqref{calcond}. Even in the general case 

6031: \eqref{sLagr-cond} and \eqref{cal1} imply that the Euclidean $D2$ branes have to 

6032: wrap special Lagrangian cycles in $Y$, which are calibrated with respect to 

6033: $\R(e^{-U-i\theta_{D2}}\Omega)$. 

6034: On the other hand, recall 

6035: that the orientifold planes are located 

6036: at the fixed points of the anti-holomorphic 

6037: involution $\sigma$ in $Y$ which are

6038: special Lagrangian cycles calibrated 

6039: with respect to $\R(e^{-U-i\theta}\Omega)$

6040: as was argued in  eqs.\

6041: \eqref{OLagr} and \eqref{calibr-O6}.\footnote{$e^{-U}$ is the normalization factor which was left undetermined in \eqref{calibr-O6}.}

6042: Thus, in order for the D-instantons to 

6043: preserve the same linear combination of the supercharges as the orientifold, we have to 

6044: demand  $\theta_{D2} =\theta$.

6045:  Using this constraint and inserting the calibration conditions 

6046: \eqref{cal1} back into \eqref{instact} one finds

6047: \beq \label{instact2}

6048:   S_{D2} = i\mu_3 \, 

6049:   \int_{\cW_3} \big( \varphi^*\big[2\R( C\Omega) \big] - i \varphi^*(\hat C_3) \big)\ = \ 

6050:   \, \int_{\cW_3} \varphi^*\Omegac \ ,

6051: \eeq 

6052: where $C=\frac{1}{2} e^{-\phi-i\theta} e^{-U}$ 

6053: was defined in eqs.\ \eqref{def-C}, \eqref{4d-dilaton} and 

6054: $\Omegac$ is given in \eqref{N=1coords}. The coefficients of $\Omegac$ 

6055: expanded in a basis of $H^{3}_+(Y)$

6056: are exactly the $N=1$ K\"ahler coordinates $(N^k,T_\lambda)$ introduced in \eqref{Oexp}. As a consequence the instanton action 

6057: \eqref{instact2} is linear and thus holomorphic in these coordinates

6058: which shows that $D2$-instantons 

6059: can correct the superpotential.

6060: Explicitly such corrections can be obtained by evaluating 

6061: appropriate fermionic 2-point functions which are weighted 

6062: by $e^{-S_{D2}}$ \cite{HM}. Applying \eqref{instact2}

6063: and keeping only the lowest term in the fluctuations 

6064: of the instanton one obtains corrections of the form 

6065: \beq

6066:    W_{D3} \propto  e^{-\int_{\Sigma_3}  \Omegac}\ , 

6067: \eeq

6068: where $\Sigma_3$ is the three-cycle wrapped by the $D2$ instanton.

6069:  

6070: This result can be lifted to M-theory by embedding Calabi-Yau orientifolds into 

6071: compactifications on special $G_2$ manifolds.

6072: In this case the $D2$ instantons correspond 

6073: to membranes wrapping three-cycles in the $G_2$ space 

6074: which do not extend in the 

6075: dilaton direction \cite{HM,KMcG}. The embedding of IIA 

6076: orientifolds into $G_2$ manifolds and the comparison of the

6077: respective effective actions is the subject of section \ref{G2_embedding}.

6078: 

6079: Let us next extend this observation to IIB orientifolds. For 

6080: simplicity we set $F=0$ for these cases, since brane fluxes would 

6081: correct the K\"ahler coordinates as discussed e.g.\ in \cite{JL}. 

6082: Hence, the calibration 

6083: conditions for the respective D$(p-1)$-instantons read \cite{MMMS}

6084: \beq \label{cal_IIB}

6085:    \Em^*\big( e^{-B_2 + iJ} \big)_p = e^{i\theta_{D(p-1)}}  \sqrt{\det {\varphi^*(\hat g+ \hat B_2)}}\ d^p \lambda\ ,

6086:    \qquad p=2,4,6\ ,

6087: \eeq

6088: where $\big( e^{-B_2 + iJ} \big)_p$ denotes the $p$-form in the sum over even forms.

6089: In order that these instantons preserve the same supersymmetry as the orientifold planes 

6090: we furthermore have to set $\theta_{D(p-1)} = \theta_{O(p+3)}$, where $\theta_{O(p+3)}$ is given in 

6091: \eqref{cal_sOp}. Multiplying \eqref{cal_IIB} by $e^{-\phi}$ and comparing real and imaginary parts we find 

6092: \beq \label{pull-e}

6093:   \Em^*\fe_4 = e^{-\phi} \sqrt{\det {\varphi^*(\hat g+ \hat B_2)}}\ d^4 \lambda\ , 

6094: \eeq 

6095: where $\fe_4$ is the four-form in $\fe$ defined in \eqref{def-A} and 

6096: we have only displayed the equation for $D3$ instantons. Furthermore, by comparing \eqref{instact} 

6097: and \eqref{def-A} one finds that $\int_{\cW_{p}} \Em^* \fa$  exactly reproduces the Chern-Simons action,

6098: since the vectors in the expansions of the R-R forms $C_p$ vanish when the pulled back to $\cW_p \subset Y$. 

6099: Hence, together with \eqref{pull-e} we conclude that the instanton actions take the form 

6100: \beq

6101:     S_{D3}=i\mu_4 \int_{\cW_4} \Em^*\fe_4 - i \Em^* \fa = -i\mu_4\, T_\alpha\, \int_{\cW_4} \Em^* \tilde \omega^\alpha\ , 

6102: \eeq 

6103: where the definition of $T_\alpha$ is given in \eqref{def-coordsO3}. This shows that also in type IIB

6104: orientifolds the $N=1$ K\"ahler coordinates defined in \eqref{def-coordsO3} and \eqref{def-coordsO5} linearize 

6105: the instanton actions. By a similar reasoning as in the IIA case this ensures holomorphicity

6106: of  instanton induced superpotentials in these coordinates. 

6107: 

6108: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6109: %

6110: %  Embedding into M- and F-theory

6111: %

6112: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6113: 

6114: 

6115: \chapter{Embedding into M- and F-theory}

6116: \label{M-F-embedding}

6117: 

6118: In this chapter we discuss the embedding of type IIA and 

6119: type IIB orientifolds into compactifications of M- and F-theory.

6120: Let us first review the basic idea, by briefly introducing F- and

6121: M-theory in the limit needed for our considerations. 

6122: 

6123: F-theory provides a geometrical interpretation of the non-perturbative 

6124: $Sl(2,\mathbb{Z})$ symmetry \eqref{Sl2} of type IIB string theory.

6125: Under this symmetry the complex dilaton $\tau$  

6126: transforms in a non-trivial manner and  

6127: can be interpreted as the complex structure modulus of 

6128: a two-dimensional torus. In \cite{Vafa} this idea was 

6129: put forward in arguing for a natural interpretation 

6130: in terms of a twelve-dimensional F-theory. 

6131: Compactifying this theory on a two-torus gives back type IIB in ten dimensions.

6132: However, in going to lower dimensions, this torus can be fibered over the

6133: internal manifold. Compactification of F-theory on such elliptically 

6134: fibered manifolds $Y_{n+2} \rightarrow B_{n}$ is defined to be type IIB string

6135: theory compactified on the base $B_n$, with a complex dilaton field $\tau$

6136: varying over the internal manifold. One interesting case is when 

6137: $Y_4$ is a elliptically fibered Calabi-Yau fourfold with base $B_3$. 

6138: It was shown in \cite{Sen} that in a special limit which corresponds to 

6139: a weak coupling limit of type IIB string theory the two-fold cover of $B_3$ 

6140: is a Calabi-Yau manifold. Furthermore, the compactification on $B_3$ corresponds to an 

6141: orientifold compactification with $O7$ planes and $D7$ branes, which are

6142: located at points where the torus fibers become singular. This limit is called 

6143: the orientifold limit 

6144: \beq \label{orientifold_limit}

6145:     \text{F-theory}\ /\ Y_4 \quad \xrightarrow[\text{limit}]{\text{\quad orientifold\quad }} 

6146:     \quad \text{Type IIB}\ /\ \mathcal{O}Y_6 \ . 

6147: \eeq

6148: Section \ref{F-theory} is devoted to check this correspondence for the effective bulk 

6149: actions of the two theories. However, since there is no known effective action

6150: for F-theory we will take a detour over M-theory compactified on $Y_4$.

6151: We compare the resulting three-dimensional effective action with the $D=3$ action 

6152: obtained by compactifying the $O3/O7$ orientifold action on a circle. Later on 

6153: we lift the correspondence to $D=4$ and compare it with \eqref{orientifold_limit}.   

6154: 

6155: In section \ref{G2_embedding} we discuss the embedding of Type IIA orientifolds into M-theory. 

6156: Recall that type IIA supergravity can be obtained by compactifying 

6157: 11-dimensional supergravity (the low energy limit of M-theory) on a circle. 

6158: Correspondingly the $D=4,N=2$ theories arising in Calabi-Yau compactifications 

6159: are lifted as  

6160: \beq

6161:   \text{Type IIA}\ /\ Y_6 \quad \cong \quad  \text{M-theory}\ /\ S^1 \times Y_6 \ .

6162: \eeq  

6163: Hence, the immediate question is to find some analog for 

6164: the orientifold compactifications. In order to do that, one

6165: has to identify appropriate manifolds which upon compactification

6166: of M-theory (understood as 11-dimensional supergravity) yield a 

6167: four-dimensional $N=1$ theory. Recalling that the number of supersymmetries 

6168: is related to the number of covariantly constant spinors, the only possible 

6169: candidates are seven-manifolds with structure group or holonomy $G_2$. 

6170: This implies that the reduction of the $SO(7)$ spinor 

6171: representation yields one singlet, which in the case of $G_2$ holonomy is furthermore 

6172: covariantly constant with respect to the Levi-Cevita connection.

6173: It was argued in \cite{KMcG} that for a special class of 

6174: $G_2$ manifolds $X$ the resulting four-dimensional theory coincides with 

6175: the one of IIA Calabi-Yau orientifolds. Schematically one has

6176: \beq

6177:   \text{Type IIA}\ /\ \mathcal{O}Y_6  \quad \cong \quad  \text{M-theory}\ /\ X \ .

6178: \eeq

6179: In section \ref{G2_embedding} we verify this conjecture for a certain limit of the two

6180: theories. This enables us to match the $N=1$ characteristic functions determined in 

6181: section \ref{Kpo_gaugeIIA} for IIA orientifolds with the one obtained for $G_2$ compactifications

6182: on $X$. As we will show, this includes the K\"ahler potential, the gauge-couplings as 

6183: well as the flux superpotentials. In ref. \cite{TGL2} only part of the orientifold

6184: superpotentials were found to have an origin in an M-theory compactification on a 

6185: manifold with $G_2$ holonomy. As we will show, the remaining terms are due 

6186: to a non-trivial fibration of a manifold with $G_2$ structure introduced in \cite{CS,CCDLM}. 

6187: 

6188: 

6189: 

6190: 

6191: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6192: %

6193: %    F-Theory and CY IIB orientifolds

6194: %

6195: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6196: 

6197: \section{F-theory and $O3/O7$ orientifolds}

6198: \label{F-theory}

6199: 

6200: In this section we discuss the embedding of $O3/O7$ orientifolds 

6201: into a F-theory compactification, which corresponds to the limit \eqref{orientifold_limit}.

6202: To analyze the two theories on the level of the effective bulk actions we start 

6203: by compactifying M-theory on a Calabi-Yau four-fold. 

6204: When shrinking the volume of the elliptic fiber the M-theory

6205: compactification on $Y_4$ is equivalent to an F-theory 

6206: compactification on $Y_4$. We only perform this limit at the 

6207: very end and rather compare the two theories in three 

6208: dimensions. In order to do that we first briefly review 

6209: compactifications of eleven-dimensional supergravity on 

6210: Calabi-Yau fourfolds following \cite{HL,BHS}. We determine the effective action

6211: and characteristic functions encoding the supergravity theory.

6212: Next we compactify the four-dimensional effective action 

6213: of $O3/O7$ orientifolds to three dimensions on a circle. 

6214: We are then in the position to show, that the characteristic 

6215: data of the two three-dimensional theories coincide if we 

6216: choose a Calabi-Yau fourfold of the form 

6217: \beq \label{def-Z}

6218:   Y_4 = (Y \times T^2)/\hat \sigma\ ,

6219: \eeq

6220: where $Y$ is a Calabi-Yau threefold and $\hat \sigma = (\sigma,-1,-1)$. 

6221: The involution $\hat \sigma$ acts as a holomorphic isometric involution on $Y$ and

6222: inverts both coordinates on $T^2$. Note that $Y_4$ generically admits singularities if

6223: $\sigma$ has a non-trivial fix-point set. These have to be smoothed out which 

6224: yields additional moduli in the theory. The analog on the orientifold are 

6225: moduli corresponding to D-branes and orientifold planes. However, since 

6226: we only restricted to the bulk fields we will also freeze moduli arising in 

6227: the process of smoothing out $Y_4$ defined in \eqref{def-Z}. 

6228: Having matched the three-dimensional theories we comment on the lift to 

6229: $D=4$. Finally, we also include a brief discussion on the lift 

6230: of orientifold three-form flux $G_3$ to four-form flux $G_4$. 

6231: 

6232: 

6233: 

6234: 

6235: 

6236: \subsubsection{M-theory compactified on a Calabi-Yau fourfold}

6237: 

6238: Let us start by summarizing compactification of M-theory on a

6239: Calabi-Yau fourfold by following the analysis of \cite{HL,BHS}. 

6240: The low energy effective action of 11d supergravity is given by \cite{CJS}

6241: \bea \label{11act}

6242:  S^{(11)}=\int - \tfrac{1}{2} R *\mathbf{1} - \tfrac{1}{4} F_4 \wedge * F_4 

6243:                    -\tfrac{1}{12} C_3 \wedge F_4 \wedge F_4\ ,

6244: \eea

6245: where $F_4=dC_3$ is the field strength of $C_3$. 

6246: The three-form $C_3$ together with the eleven-dimensional metric are the 

6247: only bosonic fields in the low energy description of M-theory.

6248: Recall that the action \eqref{11act} is given to lowest order 

6249: in $\kappa_{11}$. One-loop corrections associated to the sigma model anomaly of 

6250: a $M5$-brane contribute additional terms to 

6251: \eqref{11act} and induce a $C_3$ tadpole term 

6252: $-\frac{\chi(Y_4)}{24}$ \cite{DLM,SVW}. 

6253: This contribution can be canceled by considering setups 

6254: with a certain number of background $M3$-branes or 

6255: switched on background fluxes. 

6256: However, for the moment we keep our analysis simple in sticking to

6257: the action \eqref{11act} without extra source terms. 

6258: 

6259: The fields of the three-dimensional theory arise from the expansion 

6260: of the eleven-dimensional supergravity fields into harmonic forms. 

6261: For a Calabi-Yau fourfold $Y_4$, the only non-vanishing cohomologies are 

6262: given by 

6263: \bea \label{Z-cohom}

6264:  H^{0}(Y_4)& =& H^{(0,0)}\ , \qquad H^{2}(Y_4)\ =\ H^{(1,1)}\ ,\qquad  H^{3}(Y_4)\ =\ H^{(2,1)} \oplus H^{(1,2)} \ ,\nn \\[.2cm]

6265:  H^{4}(Y_4)&=&H^{(4,0)}  \oplus H^{(3,1)} \oplus H^{(2,2)} \oplus H^{(1,3)} \oplus H^{(0,4)}\ ,

6266: \eea

6267: with their Hodge duals $H^{5}$, $H^{6}$ and $H^{8}$.

6268: Let us extract the spectrum obtained by expansion into harmonic basis forms of these

6269: cohomologies. This is done in analogy to the case of type II compactifications discussed in 

6270: chapter \ref{TypeII}.

6271: The deformations of the metric of the fourfold respecting the 

6272: Calabi-Yau condition split into two sets: $h^{(1,1)}(Y_4)$ real scalar K\"ahler

6273:  structure deformations $M^\cA(x)$ and $h^{(3,1)}(Y_4)$ complex structure moduli $Z^\cK(x)$. 

6274: Similar to \eqref{def-v} and \eqref{cs} for Calabi-Yau threefolds they parameterize  

6275: the expansions

6276: \beq \label{deform_4}

6277:   J_F \ =\ M^\cA(x) e_\cA\ ,\qquad 

6278:   \delta g_{\bi \bj} = -\frac{1}{3 ||\Omega ||^2} \bar \Omega_{F\, \bi}^{\ \ \ klm} Z^\cK(x) 

6279:   \Phi_{\cK\, klm\bj}  

6280: \eeq

6281: where $J_F$ and $\Omega_F$ are the K\"ahler form and the holomorphic $(4,0)$-form on 

6282: the Calabi-Yau fourfold. The harmonic forms $e_\cA, \cA=1,\ldots, h^{(1,1)}(Y_4)$

6283: form a basis of  $H^{(1,1)}(Y_4)$, while $\Phi_\cK, \cK=1,\ldots, h^{(3,1)}(Y_4)$ 

6284: form a basis of $H^{(3,1)}(Y_4)$. 

6285: Also $C_3$ is expanded into harmonic forms via the Kaluza-Klein Ansatz

6286: \beq \label{A3exp}

6287:   C_3 = A^\cA(x)\wedge e_\cA + N^I(x)\, \Psi_I + \bar N^I(x)\, \bar \Psi_{ I}\ , 

6288: \eeq

6289: where $A^\cA(x)$ are vectors and $N^I(x)$ are complex scalars in three dimensions. 

6290: The harmonic forms $\Psi_I,\bar \Psi_I,I=1,\ldots h^{(2,1)}$ 

6291: define a basis of $H^{3}(Y_4)$, which can be chosen to obey

6292: %

6293: %

6294: \footnote{This needs some words of justification. First, recall 

6295: that for a complex manifold $Y_4$ the filtration $F^3(\MM) = H^{(3,0)}$,  

6296: $F^2(\MM) = H^{(3,0)}  \oplus H^{(2,1)}$, 

6297: etc.\ can be shown to consist of holomorphic bundles $F^i(\MM)$ over the space of complex 

6298: structure deformations. Since $H^{(3,0)}$ is empty 

6299: for Calabi-Yau fourfolds, $H^{(2,1)}$ is a holomorphic

6300: bundle and one can locally choose a basis

6301: $\psi_I(Z),\ \partial_{\bar Z^\cK}\psi_I=0 $. Hence, the holomorphic derivative is expanded as

6302: %\beq \label{psiexp}

6303: $  \partial_{Z^\cK} \psi_I = (\sigma_\cK)^{J}_{I} \psi_J  + 

6304:   (\lambda_\cK)^{\bar J}_{I} \bar \psi_{\bar J}, $

6305: %\eeq

6306: where $(\sigma_\cK)^{J}_{I},\ (\lambda_\cK)^{\bar J}_{I}$ are functions of $Z,\bar Z$.

6307: One can now show, that there exists a basis $\Psi^I = M_I^{\bar J} \bar \psi$ (for some real 

6308: $M_I^{\bar J}$) which obeys \eqref{der_Psi}.

6309: In order that this is the case one has to demand: $\partial_{Z^\cK}\ln M^{\bar I}_J=A_{\cK J}^{\ \ \ I}$,

6310: $B_{\bar \cK I}^{\ \ \ \bar J}=(M^{-1})^{\bar J}_K M^{\bar L}_I (\bar \lambda_{\bar \cK})^{K}_{\bar L}$

6311: and $A_{\cK I}^{\ \ \ \bar J}=-(\sigma_\cK)^{J}_{I}$. A possible definition of $M^{\bar I}_J$ can

6312: be found in \cite{HL}.}

6313: %

6314: %

6315: \beq \label{der_Psi}

6316:   \partial_{Z^{\cK}} \Psi_I = A_{\cK  I}^{\ \ \  J} \Psi_J \ , \qquad

6317:   \partial_{\bar Z^{\cK}} \Psi_I = B_{\bar \cK I}^{\ \ \ \bar J} \bar  \Psi_J\ ,  

6318: \eeq

6319: where $A_{\cK  I}^{\ \ \ J}$ and $B_{\bar \cK I}^{\ \ \ \bar J}$ are model dependent 

6320: functions of $Z$ and $\bar Z$. Differentiating these equations with respect 

6321: to $Z^{\cK}$ and $\bar Z^{\cL}$ and comparing 

6322: $\partial_{Z^\cK} \partial_{\bar Z^\cL} \Psi_I$ with $\partial_{\bar Z^\cL} \partial_{Z^\cK} \Psi_I$

6323: we extract the consistency conditions

6324: \beq \label{DE}

6325:   \partial_{\bar Z^{\cK}}  A_{\cL  I}^{\ \ \  J} = B_{\bar \cK I}^{\ \ \ \bar L} \, \bar B_{\cL \bar L}^{\ \ \ J}\ ,

6326:   \qquad

6327:   \partial_{\bar Z^{\cK}} \bar B_{\cL \bar I}^{\ \ \ J} = A_{\bar \cK \bar I}^{\ \ \  \bar L} \bar B_{\cL \bar L}^{\ \ \ J}\ .

6328: \eeq

6329: In summary, the bosonic part of the $D=3,N=2$ supergravity spectrum obtained by compactification 

6330: on a Calabi-Yau fourfold is displayed in table \ref{Mspectrum}.

6331:    

6332: \begin{table}[h] 

6333: \begin{center}

6334: \begin{tabular}{|l|c|c|c|} \hline 

6335:  \rule[-0.3cm]{0cm}{0.9cm}

6336:  {gravity multiplet} &   1  & $g^{(3)}_{pq}$\\ \hline

6337:  \rule[-0.3cm]{0cm}{0.9cm}

6338:  {vector multiplets} &    $h^{(1,1)}$ & $(M^\cA,A^\cA)$ \\ \hline

6339: \rule[-0.3cm]{0cm}{0.9cm} 

6340:  chiral multiplets & $h^{(3,1)} + h^{(2,1)}$ & $Z^\cK$, $N^I$ 

6341:  \\ \hline

6342: \end{tabular} 

6343: \caption{\textit{$D=3,N=2$ spectrum for M-theory 

6344:                  on a Calabi-Yau fourfold.}}\label{Mspectrum}

6345: \end{center}

6346: \end{table} 

6347: 

6348: Also the calculation of the three-dimensional low energy effective action is similar to

6349: the analysis performed in chapter \ref{TypeII}. 

6350: The field strength $F_4=dC_3$ is evaluated by using \eqref{A3exp} and \eqref{der_Psi}

6351: as

6352: \beq \label{fieldstr_4}

6353:   F_4 = dA^\cA \wedge e_\cA + D N^I \Psi_I +  D \bar N^I \bar \Psi_I\ , 

6354: \eeq

6355: with

6356: \beq

6357:  DN^I = dN^I + (N^J  A_{\cK J}^{\ \ \ I} + \bar N^J B_{\cK \bar J}^{\ \ \ I}) dZ^\cK \ , \qquad D\bar N^I = \overline{DN^I}

6358: \eeq

6359: Inserting \eqref{deform_4}, \eqref{fieldstr_4} and \eqref{A3exp} and performing the standard Weyl rescaling

6360: the effective action takes the form \cite{HL}

6361: \bea \label{F-theory_act}

6362:   S^{(3)}_{F}&= &\int -\tfrac{1}{2} R - G_{\cK \cL}\, dZ^\cK \wedge * dZ^\cL 

6363:                        - \tfrac{1}{2} d \ln \cV \wedge * d \ln \cV - \tfrac{1}{2} G_{\cA \cB}\, dM^\cA \wedge * dM^\cB\nn\\ 

6364:                 &&       - \tfrac{1}{2} \cG_{I\bar J}\ D N^I \wedge * D\bar N^J 

6365:                          - \tfrac{1}{2} \cV^2\, G_{\cA \cB}\ dA^\cA \wedge dA^\cB \nn \\

6366:                 &&       + \tfrac{i}{4} d_{\cA I \bar J}\ dA^\cA \wedge (N^I D\bar N^J - \bar N^I  D N^J)\ ,

6367: \eea

6368: where $G_{\cK \cL}$, $\cG_{I\bar J}$ and $G_{\cA \cB}$ are the metrics on $H^4$, $H^3$ and $H^2$ respectively

6369: and will be discussed in turn.

6370: Let us first comment on the complex structure and K\"ahler structure deformations. 

6371: The higher-dimensional analog 

6372: of \eqref{csmetric} is the metric $G_{\cK \cL}$ on the space of complex structure 

6373: deformations of $Y_4$. It is K\"ahler and takes the form

6374: \beq

6375:   G_{\cK \bar \cL} = \partial_{Z^\cK} \partial_{\bar Z^\cL} K^{\rm cs}_F\ , \qquad  

6376:   K^{\rm cs}_F=-\ln\big[\int_{Y_4} \Omega_F \wedge \bar \Omega_F \big]\ .

6377: \eeq

6378: In analogy to \eqref{Kmetric} and \eqref{int-numbers} 

6379: we define on the space of 

6380: $(1,1)$-forms intersection numbers $d_{\cA \cB \cC \cD}$ and 

6381: a metric $G_{\cA \cB}$  via

6382: \beq \label{d_ABCD}

6383:  d_{\cA \cB \cC \cD} = \int_{Y_4} e_\cA \wedge e_\cB \wedge e_\cC \wedge e_\cD\ , \qquad 

6384:  G_{\cA \cB} = \frac{1}{2 \cV} \int_{Y_4} e_\cA \wedge * e_\cB\ ,  

6385: \eeq

6386: where $\cV = \frac{1}{4!}\int J_F \wedge J_F \wedge J_F \wedge J_F$ is the volume of the 

6387: Calabi-Yau four-fold. 

6388: 

6389: In contrast to a Calabi-Yau threefold the four-dimensional manifold $Y_4$ admits a third non-trivial 

6390: cohomology $H^{3}(Y_4)$ with metric $G_{I \bar J}$. It has non-vanishing intersections 

6391: $d_{\cA I \bar J}$ with $H^2$ such that

6392: \beq \label{d_AIJ}

6393:   d_{\cA I \bar J} = i\int_{Y_4} e_\cA \wedge \Psi_I \wedge \bar \Psi_{J}\ ,\qquad 

6394:   \cG_{I\bar J} = \frac{1}{4 \cV} \int_{Y_4} \Psi_I \wedge * \bar \Psi_{J}= - \frac{\,M^\cA d_{\cA I \bar J}}{4 \cV}\ ,

6395: \eeq 

6396: where we have used $*\bar \Psi_I = -iJ_F \wedge \bar \Psi_I$ in order to evaluate 

6397: the last equality.

6398: However, in general $\cG_{I\bar J}$ as well as $d_{\cA I \bar J}$ depend on the complex structure

6399: deformations $Z^\cK$, since their definition involves the forms $\Psi_I(Z,\bar Z)$.

6400: Hence, by using \eqref{der_Psi} we obtain differential equations for $d_{\cA I \bar J}$ and $\cG_{I\bar J}$, 

6401: which read 

6402: \beq \label{Dd}

6403:   \partial_{Z^\cK}  d_{\cA I \bar J}  = A_{\cK I}^{\ \ \ K}\,  d_{\cA K \bar J}\ ,\qquad 

6404:   \partial_{Z^\cK} \cG_{I\bar J} =  A_{\cK I}^{\ \ \ K}\, \cG_{K\bar J}\ .

6405: \eeq

6406: 

6407: Having determined the effective action \eqref{F-theory_act} we can now proceed in two ways. Either 

6408: we dualize the vectors $A^{\cA}$ into scalars $P_\cA$ and combine them into chiral multiplets 

6409: $T_\cA=(M^\cA,P_\cA)$. The K\"ahler potential of this $D=3, N=2$ theory was determined in \cite{HL}.

6410: It takes the form

6411: \beq

6412:   K_F(Z,N,T) = - \ln\Big[ \int_{Y_4} \Omega_F \wedge \bar \Omega_F \Big] - 3 \ln \cV(T,N)\ , 

6413: \eeq 

6414: where $\cV(T,N)$ is the volume of $Y_4$, which depends implicitly on the K\"ahler coordinates.

6415: This is indeed analog to the situation in type IIB orientifolds with $O3/O7$ planes.

6416: However,  

6417: in section \ref{IIB_lin} we explored a way around this implicit definition by 

6418: changing to the dual picture. In $D=4$ this amounts to by keeping linear multiplets 

6419: $(L^\alpha,D^\alpha_2)$ in the spectrum, which allows to give $K$ as an explicit function 

6420: of $L^\alpha$. As we will review momentarily, this is equivalently 

6421: true for the $D=3$ theory \eqref{F-theory_act} and amounts to keeping the vector multiplets $(M^\cA, A^\cA)$ in 

6422: the spectrum \cite{BHS}. 

6423: 

6424: General $D=3,N=2$ supergravity theories with vector and chiral multiplets 

6425: are discussed e.g.\ in \cite{BHS}. To avoid a detailed review of their results we make contact with 

6426: section \ref{linear_multiplets} by observing that the effective action \eqref{kinetic_lin} for chiral and 

6427: linear multiplets in $D=4$ can be translated to $D=3$ chiral-vector setups 

6428: by replacing $dD_2^\cA$ with $dA^\cA$.\footnote{Furthermore, one has to replace in the potential 

6429: \eqref{Lsc} the factor $3$ by a $4$ \cite{HL}.} Using these identifications, one compares 

6430: \eqref{kinetic_lin} with \eqref{F-theory_act} to find 

6431: \beq

6432:   L^\cA = \frac{M^\cA}{\cV}\ ,\qquad  

6433:   \tilde K_{L^\cA L^\cB} = - \tfrac{1}{2}\, \cV^2\, G_{\cA \cB}\ .

6434: \eeq

6435: The kinetic potential for the vector multiplet $(L^\cA, A^\cA)$ is found to be \cite{BHS}

6436: \beq \label{kin_F}

6437:  \tilde K(L,N,Z) = - \ln\Big[ \int_{Y_4} \Omega_F \wedge \bar \Omega_F \Big] 

6438:                    + \ln\big(d_{\cA \cB \cC \cD} L^\cA L^\cB L^\cC L^\cD \big) + 

6439:                     L^\cA \zeta_A 

6440: \eeq

6441: with

6442: \beq

6443:   \zeta^R_A = \tfrac12 d_{A I \bar J} \bar N^I N^J+ \omega_{A I J} N^I N^J + \omega_{A \bar I \bar J} \bar  N^I \bar N^J \ .

6444: \eeq

6445: The functions $\omega_{A \bar I \bar J}(Z,\bar Z)$ obey

6446: \beq \label{Domega}

6447:   \partial_{\bar Z^\cK} \omega_{\cA \bar I \bar J} = B_{\bar \cK I}^{\ \ \ \bar K} d_{\cA J \bar K}\ ,

6448: \eeq

6449: but are otherwise unconstraint. It is now straight forward to check, that 

6450: the effective action determined in terms of $\tilde K(L,N,Z)$ is 

6451: indeed equivalent to \eqref{F-theory_act} up to a total derivative \cite{BHS}.\footnote{More precisely 

6452: one finds $\tfrac{i}{4} d_{\cA I \bar J}\ (\bar N^I DN^J - N^I  D \bar N^J) = \I(\tilde K_{L^\cA Q^m} d Q^m) +$ total derivative, where $Q^m=(N^I,Z^\cK)$.} This ends our review of the M-theory compactification. In order to compare 

6453: \eqref{kin_O33} with the $O3/O7$ orientifold data, we first have to compactify the orientifold theory to 

6454: three dimensions.

6455: 

6456: \subsubsection{The $O3/O7$ orientifolds in three-dimensions}

6457: 

6458: Let us now compactify the four-dimensional $O3/O7$ orientifold theory determined by 

6459: \eqref{S_scalarO3} on a circle $S^1$. In order to do that we partly follow \cite{HL}, where general 

6460: compactifications of $D=4,N=1$ theories are discussed. 

6461: Due to the fact that $D=4$ chiral multiplets reduce to $D=3$ multiplets we turn our 

6462: attention to the vectors $V^\kappa$ with kinetic terms \eqref{red-vector}. In three dimensions

6463: vectors are dual to scalars and for four supercharges the dynamics can be encoded by a K\"ahler 

6464: or kinetic potential.   

6465: The Kaluza-Klein reduction is performed by choosing the Ansatz

6466: \beq \label{3d-Ansatz}

6467:   g^{(4)}_{\mu \nu} = \left(

6468:   \begin{array}{cc}

6469:    g^{(3)}_{pq} + r^2 A_p^0 A_q^0& r^2 A_q^0\\ r^2 A_p^0 & r^2

6470:   \end{array} 

6471:   \right)\ , \qquad V^\kappa_\mu = (A^\kappa_p+A^0_p\, n^\kappa, n^\kappa)\ ,

6472: \eeq 

6473: where $A^0_p,\ A^\kappa_p,\ p=1,2,3$ are vectors and $n^\kappa$ as well as $r$ (the radius of $S^1$) are

6474: scalars in three dimensions. The resulting $D=3$ theory posses chiral multiplets $(z^k,\tau,G^a,T_\alpha)$ and 

6475: vector multiplets $(A^0,r)$ and $(A^\kappa,n^\kappa)$. Next we dualize the vectors $A^\kappa$ 

6476: into scalars $\tilde n_\kappa$

6477: by the standard Lagrange multiplier method (see section \ref{revIIB}). However, we keep the vector multiplet 

6478: $(A^0,r)$ and denote $L={r}^{-1}$.

6479: The scalars $\tilde n_\kappa$ and $n^\kappa$ combine into complex scalars $D_\kappa$ via \cite{FS,HL}

6480: \beq

6481:   D_\kappa = -f_{\kappa \lambda}(z)\, n^\lambda  + i\, \tilde n_\kappa\ ,

6482: \eeq 

6483: where $f_{kl}(z)$ are the gauge-couplings of the $O3/O7$ theory given in \eqref{fholo}. One next inserts the 

6484: Ansatz \eqref{3d-Ansatz} into the $D=4$ orientifolds action \eqref{S_scalarO3} and performs a Weyl rescaling

6485: to obtain a $D=3$ effective action with standard Einstein-Hilbert term. Using the definition of $D_\kappa$ 

6486: this action is encoded by a kinetic

6487: potential 

6488: \beq

6489:   \tilde K_3 = - \ln\Big[ \int_Y \Omega \wedge \bar \Omega \Big]  + K^{k}(\tau, G,T) + \ln (L) + L \zeta^R\ ,

6490: \eeq 

6491: where $K^{k}(\tau, G,T)$ and $\zeta^R$ are given in \eqref{kaehlerpot-Kk} and \eqref{def-zR}.

6492: Replacing the chiral multiplets $T_\alpha$ by vector multiplets $(A^\alpha,L^\alpha)$ we

6493: apply \eqref{tildeK_O3} to rewrite the kinetic potential as 

6494: \beq \label{kin_O33}

6495:   \tilde K_3 = - \ln\Big[ \int_Y \Omega \wedge \bar \Omega \Big] %K^{\rm cs}(z) 

6496:                - \ln\big(-i(\tau -\bar \tau)\big) + 

6497:                \ln( \cK_{\alpha \beta \gamma} L^\alpha L^\beta L^\gamma) +  \ln (L) + 

6498:                L^\alpha \zeta^R_\alpha + L \zeta^R\ ,

6499: \eeq

6500: where 

6501: \beq \label{def-zR}

6502:   \zeta^R_\alpha = -\frac{i}{2(\tau-\bar \tau)}\ \KK_{\alpha b c}(G-\bar G)^b (G- \bar G)^c\ , \quad 

6503:    \zeta^R = -\tfrac{1}{2} (D_k + \bar D_k) (\R f_{kl} )^{-1} (D_k + \bar D_k)\ .

6504: \eeq

6505: The function $\zeta^R_\alpha =\zeta_\alpha+\bar\zeta_\alpha$ was already given in \eqref{zetaid}.

6506: $\tilde K_3$ fully encodes the dynamics of the chiral multiplets $z^k,\tau,G^a,D_k$

6507: and the vector multiplets $(A^\alpha,L^\alpha)$ and $(A,L)$ in three-dimensions. This enables us 

6508: to compare the orientifold theory with the M-theory compactification discussed at the 

6509: beginning of this section.

6510: 

6511: \subsubsection{F-theory embedding of $O3/O7$ orientifolds}

6512: 

6513: In order to discuss the F-theory embedding of the $O3/O7$ bulk orientifold theory, we 

6514: restrict to the simple fourfolds defined in \eqref{def-Z}. Working on these manifolds 

6515: the $\hat \sigma$ invariant cohomologies split as  

6516: \bea \label{cohom_splitF}

6517:   H^2(Y_4) &=& H^{2}_+(Y) \oplus H^2_+(T^2) \ , \qquad H^3(Y_4)\ =\  H^{3}_+(Y) \oplus

6518:                                               \big(H^{2}_-(Y)\wedge H^1_-(T^2)\big)\nn \\

6519:   H^4(Y_4) &=& H^4_+ (Y) \oplus \big(H^3_-(Y)\wedge H^1_-(T^2)\big) \oplus  \big(H^{2}_+(Y) \wedge H^2_+(T^2)\big)\ ,

6520: \eea

6521: where $H^q_\pm(Y)$ are the cohomology groups of $Y$ introduced in \eqref{H3split} and we denote 

6522: by $H^1_-(T^2),\ H^2_+(T^2)$ the cohomologies of $T^2$. We denote a basis of the $T^2$-cohomologies

6523: by $\alpha^{(1,0)},\alpha^{(0,1)} \in H^1_-(T^2)$ and $\vol(T^2) \in  H^2_+(T^2)$.\footnote{ Recall, that for $T^2$ 

6524: one finds $h^{(0,0)}_+=h^{(1,1)}_+=h^{(1,0)}_-=h^{(0,1)}_-=1$.}

6525: We next analyze the spectrum and couplings of the three-dimensional 

6526: theory \eqref{F-theory_act} on the manifolds \eqref{def-Z}. Let us start with the complex structure deformations $Z^\cK$.

6527: {}From \eqref{cohom_splitF} one concludes, that the only $(3,1)$-forms in $H^4(Y_4)$ arise from 

6528: the cohomology $H^{(2,1)}_-(Y)\wedge H^{(1,0)}_-(T^2)$ and $H^{(3,0)}_-(Y)\wedge H^{(0,1)}_-(T^2)$.

6529: Hence we set 

6530: \beq

6531:   Z^\cK \equiv (\tau,\ z^k)\ ,  \qquad \cK=0,\ldots, h^{2,1}_-(Y) \ . 

6532: \eeq 

6533: This is consistent with the fact that in F-theory the complex dilaton 

6534: $\tau$ becomes the complex structure modulus

6535: of the torus fiber of the fourfold $Y_4$ given in \eqref{def-Z}. Hence, we 

6536: will set $\alpha^{(1,0)}=dq - \tau dp$ and lift $\tau$

6537: to one of the complex structure deformations of $Y_4$. 

6538: Moreover, 

6539: in the orientifold limit the complex structure deformations of the orientifold $z^k$ are

6540: the complex structure deformations of the base of $Y_4$. On \eqref{def-Z} also 

6541: the holomorphic four-form $\Omega_F$ splits as $\Omega_F=\Omega\wedge \alpha^{(1,0)}$,

6542: such that 

6543: \beq \label{split_OO}

6544:   \ln\Big[ \int_{Y_4} \Omega_F \wedge \bar \Omega_F \Big] = \ln\Big[-i\int_Y \Omega \wedge \bar \Omega \Big] 

6545:     + \ln\big[-i(\tau - \bar \tau) \big]\ ,

6546: \eeq

6547: where we have used $\int_{T^2} dq\wedge dp = 1$.

6548: 

6549: The K\"ahler structure deformations of $Y_4$ assembled into the vector multiplets 

6550: $(M^\cA/\cV, A^\cA)=(L^\cA, A^\cA)$. These split under the decomposition 

6551: \eqref{cohom_splitF} into one modulus parameterizing the torus volume and 

6552: $h^{(1,1)}_+$ K\"ahler structure deformations of $Y/\sigma$. In three dimensions

6553: this has an obvious counterpart in the orientifold theory, since an additional 

6554: K\"ahler modulus $L=r^{-1}$ arose from the compactification on $S^1$. 

6555: This leads us to identify 

6556: \beq \label{id_L}

6557:   L^\cA  \equiv (L,L^\alpha)\ ,\qquad  A^\cA \equiv (A^0,A^\alpha)\ ,\qquad \cA=0,\ldots, h^{1,1}_+(Y) \ .

6558: \eeq

6559: Note that this implies that one matches the volume modulus of $T^2$ with the inverse

6560: radius $L=r^{-1}$ of the $S^1$. Also the corresponding intersection numbers 

6561: \eqref{d_ABCD} split on the manifolds \eqref{def-Z} as

6562: \beq

6563:   d_{\cA \cB \cC \cD} \rightarrow d_{0 \alpha \beta \gamma}\ ,

6564: \eeq

6565: with all other intersections vanishing. Here we have chosen $e_0=\text{vol}(T^2)$ 

6566: to be the invariant volume form of $T^2$. This implies that in the kinetic potential

6567: \eqref{kin_F} the logarithm involving $L^\alpha$ splits as 

6568: \beq \label{split_LLLL}

6569:   \ln\big(d_{\cA \cB \cC \cD} L^\cA L^\cB L^\cC L^\cD \big) = \ln L + 

6570:   \ln\big(\cK_{\alpha \beta \gamma} L^\alpha L^\beta L^\gamma \big)\ ,

6571: \eeq 

6572: where we have identified $d_{0\alpha \beta \gamma}=\cK_{\alpha \beta \gamma}$, being the

6573: intersections of $H^2_+(Y)$.

6574: 

6575: Finally, the remaining chiral multiplets $N^I$  and the orientifold fields $G^a,D_\lambda$

6576: have to be matched

6577: \beq

6578:   N^I \equiv (G^a,D_\lambda)\ , \qquad I = 1, \ldots, h^{(1,1)}_-(Y) + h^{(2,1)}_+(Y)\ . 

6579: \eeq 

6580: Once again, this is consistent with the split \eqref{cohom_splitF}

6581: of $H^3(Y_4)$. The intersection numbers $d_{A I \bar J }$ given in \eqref{d_AIJ} 

6582: decompose as 

6583: \beq

6584:   d_{\cA I \bar J} \rightarrow d_{0\kappa \lambda}\ , d_{\alpha a b}\ , 

6585: \eeq

6586: while all other intersections vanish. Note however, that in general 

6587: $d_{\cA I \bar J}$ depends on the complex structure moduli $Z^\cK$ and

6588: a naive identification $d_{\alpha a b} \cong \cK_{\alpha a b}$ can only 

6589: be true up to a complex structure dependent part. To extract this dependence

6590: we can proceed in two ways. Either we compare the two kinetic potentials

6591: \eqref{kin_F} and \eqref{kin_O33} to determine 

6592: $d_{\cA I\bar J}$ as well as $\omega_{\cA IJ}$ and check if the equations

6593: \eqref{Dd}, \eqref{Domega} and \eqref{DE} are obeyed. However, we choose 

6594: a different route and look for simple solutions of the consistency conditions 

6595: \eqref{DE}. Having determined $A_{\cK I}^{\ \ \ J}$ and $B_{\bar \cK I}^{\ \ \ \bar J}$ 

6596: we are in the position to solve \eqref{Dd}, \eqref{Domega} to determine $d_{\cA I \bar J}$

6597: and $\omega_{\cA I \bar J}$. 

6598: 

6599: To construct a simple solution to \eqref{DE} we start with a 

6600: holomorphic functions $f_{IJ}(Z)$, which can arise e.g.~as gauge couplings of a 

6601: supersymmetric theory. In terms of $f_{IJ}$ the equations \eqref{DE} are solved by

6602: \beq

6603:   A_{\cK I}^{\ \ \ J} = - (\R f)^{-1\, JK}\, \partial_{Z^\cK} f_{KI} \ , \qquad 

6604:   B_{\bar \cK I}^{\ \ \ \bar J} = (\R f)^{-1\, JK}\, \partial_{\bar Z^\cK} \bar f_{KI}\ .

6605: \eeq

6606: Relevant for the orientifold embedding are the two special cases 

6607: \beq

6608:   f_{\kappa \lambda}(Z^k) = f_{\kappa \lambda}(z^k)\ , \qquad f_{00}(Z^0) = -i\tau\ , 

6609: \eeq

6610: where $f_{\kappa \lambda}$ are the gauge-couplings of the orientifold given in \eqref{fholo}

6611: and $-i\tau$ are the gauge-couplings of a gauge-theory on space-time filling $D3$ branes (see for example \cite{GGJL}).

6612: Not to surprisingly, these are exactly the right functions to match the kinetic 

6613: potentials \eqref{kin_F} and \eqref{kin_O33}. Namely, consistent with 

6614: \eqref{Dd} and \eqref{Domega} we identify 

6615: \beq \label{couplings}

6616:   d_{0 \kappa \lambda} = \omega_{0 \kappa \lambda} = (\R f)^{\kappa \lambda}\ , \qquad 

6617:   d_{\alpha a b} = \omega_{\alpha a b} = \frac{1}{\tau-\bar \tau} \cK_{\alpha a b}\ ,

6618: \eeq

6619: where $\cK_{\alpha a b}$  are the intersections on $Y$, which are independent of $\tau$ and $z^k$.

6620: Equations \eqref{split_OO}, \eqref{split_LLLL} and \eqref{couplings} imply 

6621: that the kinetic potential of the M-theory compactification 

6622: reduces to the one for $O3/O7$ orientifolds on the Calabi-Yau fourfold \eqref{def-Z}.

6623: 

6624: The final step is to lift this correspondence to four dimensions. On the orientifold side 

6625: this simply amounts to performing the decompactification limit $L_0=r_0^{-1} \rightarrow 0$,

6626: where $r_0$ arises in $r_0+r(x)$ as the background radius. Of course, the resulting theory 

6627: coincides with the $D=4$ orientifold theory, if identifying the correct four-vectors. 

6628: More subtle is the lift of the M-theory compactification, which is known as the F-theory limit.

6629: It amounts to shrinking the volume of the two-torus (identified in \eqref{id_L} with $L_0$)

6630: on an elliptically fibered Calabi-Yau fourfold. However, for the simple manifold \eqref{def-Z}

6631: this limes is rather straightforward and coincides with the decompactification limit for the 

6632: orientifold.  

6633: 

6634: 

6635: %Using kinetic potential \eqref{kin_O33} this amounts to 

6636: %$\tilde K_{LL} dL\wedge *dL=-(r+r^0)^{-2} dr \wedge * dr \rightarrow 0$.

6637: 

6638: 

6639: In addition to the bulk theory one can allow for non-trivial four-form background flux

6640: $G_4=\big<dC_3\big>$ on $Y_4$. The theory will be changed by a non-vanishing 

6641: potential, which is obtained from the Gukov-Vafa-Witten superpotential $\int \Omega_F \wedge G_4$. 

6642: In order to relate it to the 

6643: $O3/O7$ orientifold three-form flux $G_3$ given in \eqref{fluxesB} 

6644: one locally writes \cite{DRS,GSS,GKP} 

6645: \beq

6646:   G_{4} = - \frac{G_{3} \wedge \alpha^{(0,1)}}{\tau - \bar \tau}  + h.c. \ .

6647: \eeq 

6648: This implies that the Gukov-Vafa-Witten superpotential reduces as 

6649: \beq

6650:   \int_{Y_4} \Omega_F \wedge G_4 = \int_Y \Omega \wedge G_3\ ,

6651: \eeq

6652: which coincides with the orientifold superpotential found in \eqref{superpot}.

6653: 

6654: This ends our discussion of the F-theory embedding of $O3/O7$ orientifolds. 

6655: Their are many directions for further research. It would be

6656: desirable to include $D7$ branes into the setup, which correspond to certain 

6657: singularities on the Calabi-Yau fourfold. The naive fourfold given 

6658: \eqref{def-Z} is only valid in the regime were moduli for D-branes and orientifold

6659: branes are frozen. F-theory compactifications provide  powerful tools to

6660: approach regimes where these fields are included \cite{non-pert}. 

6661: A second issue is to discuss moduli stabilization in those setups, 

6662: resent results \cite{DSFGK} suggest that all moduli can be stabilized

6663: in F-theory compactifications by including fluxes and non-perturbative corrections. 

6664:  

6665: 

6666: 

6667: %In refs. \cite{HL2,BHS} it was argued

6668: %that after including certain higher order corrections to \eqref{11act} and 

6669: %background flux $G_4$ one obtains a gauged supergravity in three dimensions.  

6670: %Some of these results should have a four-dimensional counterpart.

6671: 

6672: 

6673: 

6674: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6675: %

6676: %    Type IIA orientifolds and special $G_2$

6677: %

6678: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6679: 

6680: 

6681: \section{Type IIA orientifolds and special $G_2$ manifolds}

6682: \label{G2_embedding}

6683: 

6684: In this section we discuss the relationship between the type IIA 

6685: Calabi-Yau orientifolds considered so far 

6686: and $G_2$ compactifications of  M-theory. 

6687: In refs.\ \cite{KMcG} it was argued that for a specific class

6688: of $G_2$ compactifications $X$, type IIA orientifolds appear at special

6689: loci in their moduli space. 

6690: More precisely,  

6691: these  $G_2$ manifolds have to be such that they admit the form 

6692: \beq \label{spG_2}

6693:   X\ =\ (Y \times S^1)/{\hat \sigma}\ ,

6694: \eeq

6695: where $Y$ is a Calabi-Yau threefold and $\hat \sigma = (\sigma,-1)$ 

6696: is an involution which inverts the coordinates of the circle $S^1$ 

6697: and acts as an anti-holomorphic isometric involution on $Y$. 

6698: $\sigma$ and $\hat \sigma$ can have a non-trivial fix-point set

6699: and as a consequence $X$ is a singular $G_2$ manifold. 

6700: In terms of the 

6701: type IIA orientifolds the fix-points of $\sigma$ are the locations 

6702: of the $O6$ planes in $Y$ and as we already discussed earlier cancellation of 

6703: the appearing  tadpoles require the presence 

6704: of appropriate $D6$-branes. In this paper we froze all

6705: excitation of the $D6$-branes and only discussed the effective action

6706: of the orientifold bulk. In terms of $G_2$ compactification this 

6707: corresponds to the limit where $X$ is smoothed out and all additional 

6708: moduli arising in this process are frozen.

6709: 

6710: 

6711: 

6712: The purpose of this section is to check the embedding of type IIA

6713: orientifolds into $G_2$ compactifications of M-theory 

6714: at the level of the $N=1$ effective action.

6715: For orientifolds the effective action was derived in sections

6716: 3 and 4 and so as a first step we need to 

6717: recall the effective action of M-theory 

6718: (or rather eleven-dimensional supergravity) on smooth $G_2$ manifolds

6719: \cite{PT, HM, Hitchin1, GPap,BW}. 

6720: 

6721: The bosonic part of the eleven-dimensional 

6722: supergravity theory was already given in equation \eqref{11act}. 

6723: It encodes the dynamic of the bosonic components $g_{11}$ and $C_3$ of the supergravity multiplet.

6724: As in the reduction on Calabi-Yau manifolds one chooses the background metric 

6725: to admit a block-diagonal form

6726: \beq \label{lin-el}

6727:   ds^2 = ds^2_4(x) + ds^2_{G_2}(y)\ ,

6728: \eeq

6729: where $ds^2_4$ and $ds^2_{G_2}$ are the line elements of a Minkowski

6730: and a $G_2$ metric, respectively.

6731: The Kaluza-Klein Ansatz 

6732: for the three-form $C_3$ reads 

6733: \beq

6734:   C_3 = c^i(x)\, \phi_i + A^\alpha(x) \wedge \omega_\alpha \ , \qquad i=1,\ldots,b^3(X)\ ,\quad \alpha = 1, \ldots, b^2(X)

6735: \eeq

6736: where $c^i$ are real scalars and $A^\alpha$ are one-forms in four space-time dimensions.

6737: The harmonic forms $\phi_i$ and $\omega_\alpha$ span a basis of

6738: $H^3(X)$ and $H^2(X)$, respectively.  

6739: The $G_2$ holonomy allows for exactly one covariantly 

6740: constant spinor which can be used to define a real, harmonic and 

6741: covariantly constant 

6742: three-form $\Phi$.\footnote{The covariantly constant

6743: three-form is the analog of the holomorphic three-form $\Omega$ 

6744: %(or the K\"ahler form $J$),

6745: on Calabi-Yau manifolds.}

6746: The deformation space of the $G_2$ metric has dimension $b^3(X)=\dim H^3(X,\bbR)$ and 

6747: can be parameterized by expanding $\Phi$ 

6748: %($d\Phi=d*\Phi=0$) 

6749: into the basis $\phi_i$ \cite{Joyce} 

6750: \beq

6751:    \Phi = s^i(x)\, \phi_i \ .

6752: \eeq

6753: One combines the real scalars

6754: $s^i$ and $c^i$ into complex coordinates according to

6755: \beq

6756:   S^i = c^i + i s^i\ ,

6757: \eeq

6758: which form the bosonic components of $b^{3}(X)$ chiral multiplets.

6759: In addition the  effective four-dimensional supergravity features

6760: $b^{2}(X)$ vector multiplets with the $A^\alpha$ as bosonic components. 

6761: Due to the $N=1$ supersymmetry, 

6762: the couplings of these multiplet are again expressed in terms of 

6763: a K\"ahler potential $K_{G_2}$,  gauge-kinetic 

6764: coupling functions $f_{G_2}$ and a (flux induced) superpotential $W_{G_2}$.

6765: Let us discuss these functions in turn.

6766: 

6767: The K\"ahler potential was found to be \cite{HM,Hitchin1,GPap,BW}

6768: \beq \label{G_2Kpo}

6769:   K_{G_2}\ =\ - 3 \ln \big(  \tfrac{1}{ \kappa^2_{11}} \tfrac{1}{7} \int_X \Phi \wedge * \Phi \big)\ ,  

6770: \eeq 

6771: where $\frac{1}{7}\int \Phi \wedge * \Phi = \text{vol}(X)$ is the volume of the $G_2$ manifold $X$. 

6772: The associated K\"ahler metric is given by 

6773: \beq \label{G_2Kmetr}

6774:   \partial_{i}\bar \partial_{\bj}  K_{G_2}\ =\ \tfrac{1}{4} \text{vol}(X)^{-1} \int_X \phi_i \wedge * \phi_j\ ,

6775:   \qquad 

6776:   \partial_{i}  K_{G_2}\ =\ \tfrac{i}{2} \text{vol}(X)^{-1} \int_X \phi_i \wedge * \Phi\ ,

6777: \eeq

6778: and obeys the no-scale type condition

6779: \beq

6780:   (\partial_{i}  K_{G_2})\,  K_{G_2}^{i \jb}\, (\partial_{\bj}  K_{G_2})  = 7\ . 

6781: \eeq 

6782: 

6783: The holomorphic gauge coupling functions 

6784: $f_{G_2}$ arise from the couplings of $C_3$ in 

6785: \eqref{11act}. At the tree level they are linear in $S^i$ 

6786: and read \cite{HM,GPap}

6787: \beq \label{gauge-kinG}

6788:   (f_{G_2})_{\alpha \beta} = \tfrac{i}{2 \kappa_{11}^2}\, 

6789: S^i\int_X \phi_i \wedge \omega_\alpha \wedge \omega_\beta\ . 

6790: \eeq

6791: 

6792: Finally, non-vanishing background flux $G_4$ of $F_4 =dC_3$ 

6793: induces a scalar potential which via \eqref{N=1pot} 

6794: can be expressed in terms of the superpotential 

6795: \cite{Gukov,AS,BW}

6796: \beq \label{G_2supo}

6797:   W_{G_2}\ =\  \tfrac{1}{4 \kappa_{11}^2}\int_X \big(\tfrac{1}{2} C_3 +i\Phi) \wedge G_4\ .

6798: \eeq 

6799: (The factor $1/2$ ensures holomorphicity of $W_{G_2}$ in the

6800: coordinates $S^i$ and compensates the quadratic dependence on $C_3$

6801: \cite{BW}.)

6802: 

6803: 

6804: In order to compare the low energy effective theory of $G_2$

6805:  compactifications

6806: % \eqref{G_2Kpo}, \eqref{gauge-kinG} and \eqref{G_2supo} 

6807: with the one of the orientifold we first have to restrict to the special $G_2$ manifolds 

6808: $X$ introduced in \eqref{spG_2}. 

6809: This can be done by analyzing how the cohomologies of $X$ are related to the 

6810: ones of $Y$. As in equation \eqref{cohom-split} we consider the splits $H^p(Y)=H^p_+ \oplus H^p_-$ 

6811: of the cohomologies into eigenspaces

6812: of the involution $\sigma$. Working on the $G_2$ manifold $X$ given in \eqref{spG_2} 

6813: we thus find the $\hat \sigma$-invariant cohomologies

6814: \beq \label{splcoho}

6815:    \begin{array}{cclcrcl}

6816:    H^2(X) &=& H^2_+(Y)\ , &\ &

6817:    H^3(X) &=& H^3_+(Y) \oplus \big[H^{2}_-(Y)\wedge H_-^1(S^1)\big] \ ,\Big. \\

6818:    H^5(X) &=& H^4_-(Y)\wedge H_-^1(S^1)\ , &&

6819:    H^4(X) &=& H^4_+(Y) \oplus \big[H^{3}_-(Y)\wedge H_-^1(S^1)\big] \ ,   

6820:   \end{array}

6821: \eeq

6822: where $H^2(X)$ and $H^5(X)$ as well as $H^3(X)$ and $H^4(X)$ are

6823: Hodge duals. $H_-^1(S^1)$ is the one-dimensional space containing

6824: the odd one-form of $S^1$. The split of $H^3(X)$ induces a split of the $G_2$-form

6825: $\Phi$ which is most easily seen by introducing locally an orthonormal basis

6826: $(e^1,\ldots,e^7) \in \Lambda^1(X)$ of one-forms. 

6827: In terms of this basis one has \cite{Joyce,Hitchin1,CS}

6828: \beq\label{Phidecomp}

6829:   \Phi

6830: %     \ =\ e^{127} + e^{347} + e^{567} + e^{135} - e^{236} - e^{146} - e^{245} 

6831:       \ =\ J_M \wedge e^7 + \text{Re} \Omega_M \ ,

6832: \qquad

6833: *\Phi= \tfrac12 J_M\wedge J_M + \I \Omega_M\wedge e^7\ , 

6834: \eeq

6835: where 

6836: \bea \label{defJO}

6837:   J_M = e^1\wedge e^2 + e^3\wedge e^4 + e^5\wedge e^6\ , \quad 

6838: \Omega_M = (e^1 + ie^2)\wedge(e^3+ie^4)\wedge(e^5+ie^6)\ .

6839: \eea

6840: Applied to the manifold \eqref{spG_2} 

6841: we may interpret $e^7=dy^7$ as being the odd one-form

6842: along $S^1$. Since  $\Phi$ is required to be invariant under 

6843: $\hat\sigma$

6844: and $\sigma$ is anti-holomorphic the decomposition 

6845: \eqref{Phidecomp} implies

6846: \beq \label{splitPhi}

6847:   \hat \sigma^* J_M = - J_M\ , \qquad  

6848: \hat \sigma^* \Omega_M = \bar \Omega_M\ .

6849: \eeq  

6850: In terms of the 

6851: basis vectors $e^1,\ldots,e^6$ this is ensured by choosing

6852: $e^4,e^5,e^6$ to be odd  and

6853: $e^1,e^2,e^3$ to be even under $\sigma$.

6854: We see that  $J_M$ and $\Omega_M$ satisfy

6855: the exact same conditions as the corresponding forms of the

6856: orientifold

6857: (c.f.\ \eqref{constrJ}, \eqref{constrO}) and thus have to be proportional to

6858: $J$ and $C\Omega$ used in section \ref{IIA_orientifolds}. In order to

6859: determine the exact relation  

6860: it is necessary to fix their relative normalization. 

6861: The relation between

6862: $J_M$ and the  K\"ahler form $J$ in the string frame 

6863: can be determined from the relation of the respective metrics.

6864: Reducing eleven-dimensional supergravity to type IIA supergravity in

6865: the string frame 

6866: requires the line element \eqref{lin-el} of the eleven-dimensional 

6867: metric to take the form 

6868: \beq \label{metransatz}

6869:   ds^2 = e^{-{2 \hat \phi}/{3}} ds_4^2(x) + 

6870:          e^{-{2 \hat \phi}/{3}} g_{(s)\, ab}\, dy^a dy^b + e^{{4 \hat \phi}/{3}} (dy^7)^2\ ,

6871: \eeq   

6872: where $a,b=1,\ldots,6$. 

6873: The factors $e^{\hat \phi}$ of  the ten-dimensional dilaton are

6874: chosen such that the type IIA

6875: supergravity action takes the standard form with 

6876: $g_{(s)}$ being the Calabi-Yau metric in string frame

6877: (see e.g.~\cite{JPbook}). 

6878: Consequently we have to identify 

6879: \beq\label{JM}

6880: J_M=e^{-{2 \hat \phi}/{3}} J\ .

6881: \eeq

6882: 

6883: Similarly, using \eqref{defJO} we find that the normalization of $\Omega_M$ is given by

6884: \bea \label{norm}

6885:   J_M \wedge J_M \wedge J_M  = \frac{3i}{4}\, \Omega_M \wedge \bar \Omega_M\ .

6886: \eea 

6887: Integrating over $Y$ and using \eqref{JM}, \eqref{Kks} and \eqref{csmetric} we obtain

6888: \bea \label{normO}

6889:   \Omega_M =  e^{-\hat \phi-i\theta} 

6890:   e^{\frac{1}{2}(\Kcs - K^{\rm K})}\, \Omega =  \sqrt{8} C\Omega\ ,

6891: \eea

6892: where $C$ is given in \eqref{def-C}.

6893: The phase $e^{i\theta}$ drops out in \eqref{norm} such 

6894: that we can choose it

6895: as in \eqref{constrO} in order to fulfill \eqref{splitPhi}.

6896: Inserting $J_M$ and $\Omega_M$ into equation \eqref{splitPhi} 

6897: one arrives at

6898: \beq \label{Phi-o}

6899:   \Phi = J \wedge d\tilde y^7 + \sqrt{8} \text{Re}(C\Omega) \ ,

6900: \eeq

6901: where we defined $d\tilde y^7 = e^{-\frac{2 \hat \phi}{3}} dy^7$. The form 

6902: $d\tilde y^7$ is normalized such that $\int_{S^1} d\tilde y^7=2\pi R$ where 

6903: the metric \eqref{metransatz} was used and 

6904: $R$ is the $\phi$-independent radius of the internal circle. 

6905: We also set $\kappa^2_{10}=\kappa_{11}^2 / 2\pi R = 1$ henceforth.

6906: Using \eqref{Phi-o}, \eqref{Phidecomp} 

6907: and \eqref{def-C} we calculate 

6908: \beq \label{voldec}

6909:           \tfrac{1}{\kappa^2_{11}}\, \tfrac{1}{7} \int \Phi \wedge * \Phi 

6910:            = e^{-\frac{4\hat \phi}{3}} \, \tfrac{1}{6} \int J \wedge J \wedge J \ , 

6911: \eeq

6912: which  equivalently  can be obtained by applying the split 

6913: $\text{vol}(X)=\text{vol}(Y)\cdot\text{vol}(S^1)$ of the $G_2$ volume when evaluated in the metric \eqref{metransatz}.

6914: Inserting \eqref{voldec} into \eqref{G_2Kpo} using \eqref{def-C}

6915: we obtain

6916: \beq\label{IIAori}

6917:   K_{G_2}\ =\

6918:        - \ln \Big[  \tfrac{1}{6} \int_Y J \wedge J \wedge J \Big]

6919:        - 2\ln\Big[2\int_Y \R(C \Omega)\wedge *_6 \R(C\Omega)\Big]\ .

6920: \eeq

6921:  Thus we find exactly the K\"ahler potential 

6922: $K$ of the type IIA orientifold as given in \eqref{N=1Kpot}.\footnote{%

6923: In terms of the Hitchin functionals \cite{Hitchin1} recently discussed in 

6924: \cite{DGNV,Nekrasov} the reduction of the 

6925: $G_2$ K\"ahler potential \eqref{G_2Kpo} corresponds to

6926: the split of the seven-dimensional Hitchin functional to the 

6927: two six-dimensional ones \ref{IIAori}.}

6928: 

6929: 

6930: In order to compare the gauge kinetic functions and the superpotential

6931: we also need to identify the 

6932: K\"ahler coordinates of the two theories. 

6933: $C_3$ splits under the decomposition \eqref{splcoho} 

6934: of the cohomologies as\footnote{We have introduced a factor of $\sqrt{2}$ for later convenience.} 

6935: \beq \label{spl-C}

6936:   C_3 = \hat B_2 \wedge d\tilde y^7 + \sqrt{2} \hat C_3 \ ,

6937: \eeq

6938: where $\hat B_2$ is an odd two-form on $Y$ 

6939: and $\hat C_3$ an even three-form on $Y$.

6940: Combining \eqref{Phi-o} and \eqref{spl-C} using \eqref{N=1coords}

6941: one finds

6942: \beq \label{spl-CPhi}

6943:   S^i \phi_i\ =\ C_3 + i\Phi\ =\ \Jc \wedge d\tilde y^7 + \sqrt{2}\, \Omegac\ .

6944: \eeq

6945: As discussed after

6946: \eqref{N=1coords} the coefficients arising in the expansions of $\Jc$ and $\Omegac$

6947: into the basis $(\alpha_k,\beta^\lambda)$ of $H^{3}_+(Y)$

6948: and $\omega_a$ of $H^2(Y)$ are exactly the orientifold coordinates and 

6949: therefore we have to identify

6950: $S^a \cong t^a$ and $S^K \cong (N^k,T_\lambda)$.

6951: With this information at hand, it is not difficult to show that the 

6952: gauge-kinetic couplings \eqref{gauge-kinG} coincide with \eqref{gauge-A}. 

6953: One splits $\phi_a = \omega_a \wedge d\tilde y^7$ and obtains 

6954: \beq

6955:   (f_{G_2})_{\alpha \beta} 

6956: = \tfrac{i}{2} S^a \int_Y \omega_a \wedge \omega_\alpha \wedge

6957: \omega_\beta\ \sim i t^a \cK_{a\alpha\beta} = (f_{OY})_{\alpha \beta}\  

6958:  ,

6959: \eeq 

6960: where the precise factor depends on the normalization of

6961: the gauge fields.

6962: 

6963: It remains to compare the flux induced superpotentials \eqref{G_2supo} 

6964: with \eqref{superpot1}. Using the 

6965: cohomology splits \eqref{splcoho} and \eqref{spl-C} 

6966: the background flux 

6967: splits accordingly as $G_4 = H_3 \wedge d\tilde y^7 + \sqrt{2} F_4$. 

6968: Inserted into \eqref{G_2supo}

6969: using \eqref{spl-CPhi} we arrive at

6970: \beq

6971:   W_{G_2} = \tfrac{1}{\sqrt{8}}\int_Y \Jc \wedge F_4 + 

6972:             \tfrac{1}{\sqrt{8}}\int_Y \Omegac \wedge H_3

6973: \eeq

6974: Compared to \eqref{superpot1} the superpotential $W_{G_2}$ only 

6975: includes terms proportional to the fluxes $H_3$ and $F_4$.\footnote{%

6976: The term proportional to $e_0$ in \eqref{superpot2} can be absorbed

6977: into a redefinition of $\R t^a$ \cite{BW}.} 

6978: An interesting question is to identify the remaining terms 

6979: in \eqref{superpot1} which are likely to arise once manifolds with

6980: $G_2$ structure (instead of $G_2$ holonomy) are considered. The term 

6981: due to $F_2$ arises in compactifications on fibered 

6982: $G_2$ manifolds $X \rightarrow Y$ \cite{CS,CCDLM}. In our case we 

6983: restrict to circle fibrations over the quotient $Y/\sigma$, where $Y$

6984: is a Calabi-Yau manifold. We introduce the projection $\pi:X \rightarrow Y$. The metric on such 

6985: a manifold takes the form 

6986: \beq

6987:    g_{G_2} = \alpha \otimes \alpha + \pi^* g\ ,

6988: \eeq 

6989: where $g$ is the metric on $Y$ and $d\alpha=\pi^* F_2$. This implies 

6990: that $X$ has not anymore $G_2$ holonomy but rather $G_2$ structure with 

6991: $d\Phi = F_2\wedge J$ being not closed. Following \cite{BJ} this 

6992: induces a superpotential term of the form 

6993: \beq \label{non-G_2}

6994:    W = \int_X (dC_3 + i d\Phi) \wedge (C_3+i\Phi) + \ldots = \int_Y F_2 \wedge \Jc \wedge \Jc + \ldots\ , 

6995: \eeq

6996: where $\Phi$ and $C_3$ are given in \eqref{Phi-o} and \eqref{spl-C} with $d\tilde y^7 = \alpha$

6997: and $dC_3 = \hat B_2 \wedge F_2 + \ldots$. This reproduces exactly the $F_2$ superpotential term \eqref{superpot2}

6998: in type IIA orientifolds. It remains to reveal the origin 

6999: the superpotential term linear in $m^0$. Unfortunately, this is less straightforward and

7000: is likely to involve more general $G_2$ manifolds \cite{Witt}.\footnote{We like to thank A. Micu for discussions on this point.}

7001: It would be nice to make this more explicit and 

7002: to point out the relation to the Scherk-Schwarz constructions of massive IIA supergravity.

7003: % One obvious possibility is to choose 

7004: % a fibred $G_2$ manifold with $d\alpha=\pi^* (m^0 J+ F_2)$. This yields the desired 

7005: % terms in \eqref{non-G_2}. It would be nice to understand this correspondence 

7006: % in more detail and to point out the relation to the Scherk-Schwarz constructions

7007: % of massive IIA supergravity.

7008: 

7009: \chapter{Conclusions}

7010: 

7011: In this work we determined the low energy effective action for type IIB and type IIA 

7012: Calabi-Yau orientifolds in the presence of background fluxes. 

7013: In our analysis we did not specify a particular Calabi-Yau manifold but

7014: merely demanded that it admits an isometric involution $\sigma$.  

7015: Furthermore, in order to preserve $N=1$ supersymmetry $\sigma$ was chosen to 

7016: be a holomorphic map in type IIB and an anti-holomorphic map in type IIA. 

7017: Depending on the explicit action of $\sigma$ on the holomorphic three-form $\Omega$, 

7018: we analyzed three distinct cases: (1) orientifolds with $O3/O7$-planes, (2) 

7019: orientifolds with $O5/O9$-planes and (3) orientifolds with $O6$-planes. 

7020: For each case we calculated the characteristic functions of the 

7021: corresponding $N=1$ supergravity theories and discussed their 

7022: generic properties. 

7023: 

7024: In chapter \ref{effective_actO} we restricted to the case where background fluxes are absent

7025: and no potential is generated. We computed the effective action by a Kaluza-Klein analysis

7026: valid in the large volume limit and determined the

7027: chiral variables, the K\"ahler potentials and the gauge kinetic functions for

7028: all three setups. We found that the moduli space of the $N=1$ theory inherits 

7029: a product structure $\tilde \cM^{\rm SK} \times \tilde \cM^Q$ from

7030: the underlying $N=2$ theory obtained by ordinary Calabi-Yau

7031: compactification of type II theories. $\tilde \cM^{\rm SK}$ is

7032: a special K\"ahler manifold parameterized by the complex structure deformations 

7033: in type IIB and by the complexified K\"ahler deformations in type IIA.

7034: For type IIB orientifolds the second component $\tilde \cM^Q$ is parameterized

7035: by the periods of the complex even form $\fe - i\, \fa$ for setups with 

7036: $O3/O7$ planes and by the periods of $\feh - i\, \fa$ for setups with $O5/O9$. The 

7037: form $\fe+ i\,\feh =e^{-\hat \phi }\, e^{-\hat B_2+iJ}$ comprises of the complexified 

7038: K\"ahler deformations while $\fa$ is a sum of the even R-R forms defined in 

7039: \eqref{def-A}. On the other hand, for type 

7040: IIA orientifolds  with $O6$ planes $\tilde \cM^{\rm Q}$

7041: is spanned by the periods of the complex three-form $\Omegac=C_3 + 2i \R C\Omega$ 

7042: containing the complex structure deformations of the Calabi-Yau orientifold. 

7043: $\tilde \cM^{\rm Q}$ is a K\"ahler submanifold inside the quaternionic manifold 

7044: with a K\"ahler potential encoding the dynamics of the even/odd 

7045: forms of the respective orientifold setup.

7046: Finally we showed that in the large volume -- large complex structure limit

7047: one finds mirror symmetric effective actions if one compares

7048: type IIA and type IIB supergravity compactified on mirror manifolds

7049: and in addition chooses a set of `mirror involutions'.

7050: For $\tilde\cM^K$ mirror symmetry amounts to a truncated versions

7051: of $N=2$ mirror symmetry in that it still relates

7052: two holomorphic prepotentials. In this case the corrections computed

7053: by mirror symmetry are likely to be analogous to the situation in $N=2$.

7054: For $\tilde\cM^Q$

7055: the situation is more involved since the geometry of the moduli

7056: space changes drastically. Nevertheless we were able to show that

7057: mirror symmetry holds in the large volume - large complex structure limit.

7058: However, understanding 

7059: the nature of the corrections computed by mirror symmetry

7060: appear to be more involved and certainly deserves further study.

7061: It is interesting to note that mirror symmetry can be understood as an exchange

7062: of the odd form $\Omegac$ with the even forms $\fe+i\, \fa$ or $\feh+i\, \fa$ in accord with    

7063: \cite{FMM}. Two choices of special coordinates in $\Omegac$ single out the corresponding orientifold 

7064: setup on the mirror side. It would be desirable to reveal 

7065: the origin of this mapping and finally to generalize it to non-Calabi-Yau compactifications. 

7066: 

7067: In chapter \ref{lin_geom_of_M} we presented a more detailed investigation of

7068: the $N=1$ moduli space of Calabi-Yau orientifold compactifications.

7069: The special K\"ahler manifold $\tilde \cM^{\rm SK}$ inherits its geometrical 

7070: structure directly from $N=2$, such that we focused on

7071: the K\"ahler manifold $\tilde \cM^Q$ inside the quaternionic space.  

7072: It turned out that the definition of the K\"ahler coordinates as well as 

7073: the no-scale type conditions on $\tilde \cM^{\rm Q}$ can be more easily understood 

7074: in terms of the `dual' formulation where some chiral multiplets of the Calabi-Yau 

7075: orientifold are replaced by linear multiplets. After a brief review of $N=1$ supergravity with 

7076: several linear multiplets we reformulated all three orientifold setups by dualizing a certain set of

7077: chiral multiplets. The transformation into 

7078: linear multiplets corresponds to a Legendre transformation of 

7079: the K\"ahler potential and coordinates. The new kinetic potential of $O3/O7$ and $O5/O9$ 

7080: orientifolds takes a particularly simple form induced from a tree-level prepotential. 

7081: In contrast for $O6$ orientifolds it is given in terms of a generic prepotential satisfying the 

7082: orientifold constrains and generically includes correction corresponding to world-sheet instantons in 

7083: type IIB. For orientifolds with $O6$ planes the Legendre transform was essential to make contact 

7084: with the underlying $N=2$ special geometry. As a byproduct we determined an entire new class of no-scale

7085: K\"ahler potentials which in the chiral formulation

7086: can only be given implicitly as the solution of some constraint equation.

7087: We closed this chapter by giving an explicit construction of the K\"ahler 

7088: manifold $\tilde \cM^{\rm Q}$ replacing the $N=2$ c-map. The space  

7089: $\tilde \cM^{\rm Q}$ was shown to admit a geometric structure similar to the one of the moduli

7090: space of supersymmetric Lagrangian submanifolds \cite{Hitchin2}. This also 

7091: provides the ground for a more general investigation of non-Calabi-Yau orientifolds. 

7092: Namely, we found that the K\"ahler potential of $\tilde \cM^{\rm Q}$ is the 

7093: logarithm of Hitchins functional for a generalized complex sixfold 

7094: evaluated for the simple even and odd forms associated to the orientifold setup.

7095: 

7096: In chapter \ref{fluxesAB} we repeated the Kaluza-Klein compactification by additionally allowing

7097: for non-trivial background fluxes. 

7098: In the $O3/O7$ case the background fluxes induce a non-trivial scalar potential

7099: which is determined in terms of a superpotential previously given in

7100: \cite{GVW,TV,GKP,BBHL}. We also included the scalar fields $(b^a,c^a)$ arising from the two

7101: type IIB two-forms $B_2$ and $C_2$. We showed that in this case the potential 

7102: is unmodified which can be traced to the no-scale property of the K\"ahler potential.

7103: For orientifolds with $O5/O9$ planes the influence of background fluxes is more

7104: involved. This is due to the fact that the space-time two-form $C_2$ arising in the 

7105: expansion of the RR field $\hat C_2$ remains in the spectrum. It combines with 

7106: the dilaton into a linear multiplet, which only if it is massless can be dualized to 

7107: a chiral multiplet. However, generic NS three-form background fluxes render this

7108: form massive. We therefore first restricted our attention to the case were the mass

7109: term vanishes which occurs if the magnetic fluxes arising from

7110: the NS three-form $H_3$ are set to zero. In the resulting

7111: chiral description the axion dual to $C_2$ is gauged with the gauge charges set by the 

7112: electric fluxes. The scalar potential now consists of two distinct contributions. 

7113: The term which depends on the RR fluxes arising from 

7114: $F_3$ is obtained from a (truncated) superpotential of the previous case

7115: whereas the second contribution depends on the electric fluxes of

7116: $H_3$ and arises from $D$-terms which are present  due to the gauged isometry.    

7117: Finally, we also analyzed non-vanishing magnetic fluxes 

7118: in the NS sector which can be described by an $N=1$ theory including a massive linear

7119: multiplet coupled to vector and chiral multiplets. 

7120: In this case the scalar potential additionally 

7121: includes a direct mass term for the scalar  

7122: in the linear multiplet which is neither a $D$- nor an $F$-term.

7123: For type IIA orientifolds all background fluxes induce a superpotential $W$ 

7124: which depends on all geometrical moduli. It splits into the sum of two terms 

7125: with one term depending on the RR fluxes and the complexified

7126: K\"ahler form $J_c$ while the second term

7127: features the NS fluxes and $\Omegac$.

7128: Both terms are expected to receive non-perturbative corrections

7129: from world-sheet- and D-brane instantons.

7130: We showed that for supersymmetric type IIA and type IIB instantons the respective actions 

7131: are linear in the chiral coordinates and thus can result in holomorphic

7132: corrections to $W$.

7133: 

7134: In the last chapter \ref{M-F-embedding} we analyzed the embedding of type IIB and type IIA 

7135: orientifolds into F- and M-theory compactifications.  

7136: Orientifolds with $O3/O7$-planes can be obtained as a limit of 

7137: F-theory compactified on elliptically fibered Calabi-Yau fourfolds \cite{Vafa,Sen}.

7138: To check this correspondence on the level of the effective action we 

7139: took a sideway by first compactifying M-theory on a Calabi-Yau fourfold.

7140: This yields a three-dimensional $N=2$ supergravity theory determined in terms 

7141: of the characteristic data of the Calabi-Yau fourfold. Restricting to a specific  

7142: fourfold this effective theory can be compared to the one obtained by  

7143: compactifying the effective action of $O3/O7$ orientifolds 

7144: on a circle to $D=3$. We determined simple solutions to the 

7145: fourfold consistency conditions for which we found perfect matching between the 

7146: orientifold and M-theory compactifications. This correspondence can be lifted

7147: to $D=4$ where the M-theory on the elliptically fibered fourfold descends to an F-theory compactification. 

7148: In our analysis we neglected contributions due to singularities of 

7149: the Calabi-Yau fourfold. Smoothed out they yield additional moduli, which 

7150: are identified with $D7$ or $O7$ moduli in the orientifold limit. 

7151: In a next step one can attempt to include these into the analysis and

7152: later deform away from the orientifold limes. Non-trivial fibrations appear if

7153: the orientifold charges are not canceled locally and the F-theory picture becomes

7154: essential.  Finally we also discussed the embedding of type IIA

7155: orientifolds into a specific class of 

7156: $G_2$ compactification of M-theory. Neglecting the 

7157: contributions arising from the singularities of the $G_2$ manifold

7158: we were able to show agreement between the low energy effective

7159: actions. Comparing the superpotentials we only discovered 

7160: the terms which are due to four-form flux from in M-theory. 

7161: However, relaxing the condition of $G_2$ holonomy we were able to 

7162: identify one of the remaining terms as corresponding to a 

7163: non-trivial fibration of a $G_2$ structure manifold. It remains to 

7164: identify the counterpart of the orientifold  

7165: superpotential term cubic in the complexified K\"ahler moduli. This 

7166: term is propotional to the mass parameter of massive IIA supergravity

7167: and plays the essential role in moduli stabilization.  

7168: 

7169: Let us end our conclusions with some directions for further research. 

7170: Firstly, it would be desirable to include D-brane matter fields into 

7171: the orientifold setups. For type IIB setups with $D3$ and $D7$ branes this was done, for example,

7172: in refs. \cite{GGJL,JL}.  An important task is to extend these results to

7173: type IIA orientifolds with space-time filling $D6$ branes. The knowledge 

7174: of the full effective action enables to perform a calculation of soft supersymmetry 

7175: breaking terms of semi-realistic D-brane scenarios. 

7176: 

7177: As already mentioned, a generalization to non-Calabi-Yau orientifolds is of 

7178: particular interest \cite{GLprep}. Orientifolds allow for consistent $D=4$ 

7179: Minkowski or Anti-de Sitter vacua for which the internal manifold possesses

7180: non-trivial torsion. As we have argued, the orientifold projections specify

7181: a K\"ahler submanifold in the quaternionic $N=2$ moduli space with geometry encoded 

7182: by special even and odd forms. The K\"ahler potential is Hitchins functional truncated

7183: by the projection. A similar analysis is likely to apply to orientifolds of 

7184: generalized complex manifolds as introduced in \cite{HitchinGCM}. 

7185:  

7186: Brane worlds in orientifolds are a prominent arena for model building in

7187: particle physics and cosmology. However, finding a particular vacuum featuring the

7188: properties of our universe is a highly non-trivial task. One major step into this 

7189: direction is to extract vacua with stabilized moduli fields. Assuming that this 

7190: can be achieved, for example by background fluxes, one encounters a huge set of possible vacua

7191: labeled by different flux quantum numbers.

7192: In the pioneering paper \cite{Douglas} it was argued that a statistical analysis of this `landscape' 

7193: could lead a deeper understanding of the vacuum structure of string theory. 

7194: These considerations were mostly applied to type IIB orientifolds and

7195: certain M-theory vacua. It is an interesting task to generalize this to type

7196: IIA orientifolds. For early time cosmology a wave-function for flux vacua could 

7197: yield an interesting attempt to approach quantum cosmological questions within

7198: the framework of string theory \cite{OVV}. It would be nice to relate these 

7199: new developments in topological string theories to the results of $N=1$

7200: flux compactifications. Surprisingly various similarities appear, which hint 

7201: to at least a formal relation.   

7202:  

7203: 

7204: 

7205: 

7206: \chapter{Appendix}

7207: 

7208: \appendix

7209: 

7210: 

7211: 

7212:  %\def\theequation{\normalsize \Alph{section}.\arabic{equation}}   

7213: \renewcommand{\thesection}{\Alph{section}}

7214: \renewcommand{\theequation}{\Alph{section}.\arabic{equation}}

7215: % \renewcommand{\thechapter}{\}

7216: %\appendix

7217: %\chapter{Appendix}

7218: \section{Conventions}\label{conventions}

7219: In this appendix we summarize our conventions.

7220: 

7221: \begin{itemize}

7222: \item

7223: The coordinates of the four-dimensional Minkowski space-time are 

7224: denoted by $x^\mu, \mu=0,\ldots,3$.

7225: The corresponding metric is chosen to have signature $(-,+,+,+)$.

7226: The coordinates of the compact Calabi-Yau manifold  $Y$

7227: are denoted by $y^i, \bar y^\bi,\ i,\bi=1,2,3$. 

7228: 

7229: \item

7230: $p$-forms are expanded into a real basis according to 

7231: \beq

7232:   A_p\ =\ \frac{1}{p!}\, 

7233: A_{\mu_1 \ldots \mu_p} dx^{\mu_1}\wedge \ldots \wedge dx^{\mu_p}\ .

7234: \eeq

7235: \item

7236: $(p,q)$-forms are expanded into a complex basis as

7237: \beq

7238:   A_{p,q} = \frac{1}{p!q!} A_{i_1 \ldots i_p \bi_1 \ldots \bi_q} dy^{i_1}\wedge \ldots \wedge dy^{i_p}

7239:             \wedge d\bar y^{\bi_1}\wedge \ldots \wedge d\bar y^{\bi_q}\ .

7240: \eeq

7241: \item

7242: The exterior derivative is defined as 

7243: \beq

7244:   dA_p=\frac{1}{p!} \partial_\mu A_{\mu_1 \ldots \mu_p} dx^\mu\wedge dx^{\mu_1}\wedge \ldots \wedge dx^{\mu_p}\ .

7245: \eeq

7246: \item

7247: The field strength of a $p$-form $F_{p+1}=dA_p$ 

7248: is given by 

7249: \beq

7250:   F_{\mu_1 \ldots \mu_{p+1}} = (p+1)\, \partial_{[\mu_1}A_{\mu_2\ldots \mu_{p+1}]}\ .

7251: \eeq

7252: \item

7253: The inner product for real forms is defined by using the Hodge-$*$ operator. In

7254: components we have 

7255: \beq \label{wedge*comp}

7256:  \int F_p \wedge * F_p = \frac{1}{p!}\int F_{\mu_1 \ldots \mu_p} F^{\mu_1 \ldots \mu_p} *\mathbf{1}\ ,

7257: \eeq

7258: where $*\mathbf{1} = d^d x\, \sqrt{-g}$ is the $d$-dimensional measure. 

7259: 

7260: \item

7261: The Hodge-$*$ satisfies 

7262: $** F_p = (-1)^{p(d-p)+\kappa} F_p$, where $\kappa=1$ for Lorentzian signature 

7263: and $\kappa=0$ for Euclidean signature. 

7264: 

7265: \item

7266: Let $\sigma_1$ and $\sigma_2$ be an orientiation preserving and an orientation reversing map

7267: $\sigma_{1,2}: M \rightarrow M$, where $M$ is an $n$-dimensional manifold. Then one finds

7268: for a $n$-form $\omega$ on $M$ that

7269: \beq \label{int-form1}

7270:   \int_{\sigma_1(M)} \omega = \int_M \sigma_1^*(\omega)\ , \qquad  \int_{\sigma_2(M)} \omega = -\int_M \sigma_2^*( \omega)\ .

7271: \eeq

7272: However, if we choose $\omega_{M}=*\mathbf{1}$ to be the canonical volume form of $M$ then

7273: $\omega_{\sigma_1(M)}= \sigma_1^* (\omega_{M})$ and $\omega_{\sigma_2(M)}= -\sigma_2^*( \omega_{M})$,

7274: such that

7275: \beq \label{int-form2}

7276:   \int_{\sigma_{1,2}(M)} \omega_{\sigma_{1,2}(M)} = \int_M \sigma_{1,2}^*(\omega_M)\ .

7277: \eeq

7278: 

7279: \end{itemize}

7280: 

7281: 

7282: 

7283: 

7284: \section{N=2 supergravity and special geometry}

7285: \label{specialGeom}

7286: 

7287: In this appendix we briefly summarize the $N=2$ special geometry

7288: of the Calabi-Yau 

7289: moduli space. A more detailed discussion can be found,

7290: for example, in refs.\ \cite{CdO,Strominger2,Freed,N=2review,CRTV}.

7291: A special K\"ahler manifold $\cM$ is a Hodge-K\"ahler manifold (with line bundle $\cL$)

7292: of real dimension $2n$ with associated 

7293: holomorphic flat $Sp(2n+2,\mathbb{R})$ vector bundle $\mathcal{H}$ over $\cM$. Furthermore 

7294: there exists a holomorphic section $\Omega(z)$ of $\cL$ such that 

7295: \beq\label{N=2KP}

7296:   K(z,\bar z) = - \ln i \big<\Omega(z) , \bar \Omega(\bar z)  \big>\ , \qquad 

7297:   \big<\Omega, \partial_{z^K} \Omega\big> = 0\ , \qquad K=1,\ldots n\ ,

7298: \eeq

7299: where $K$ is the K\"ahler potential of $\cM$ and 

7300: $\big<\cdot,\cdot \big>$ is the 

7301: symplectic product on the fibers. 

7302: This is precisely what one encounters in the moduli space

7303: of the complex structure deformations of a Calabi-Yau manifold

7304: with $\Omega$ being the holomorphic three-form.

7305: In this case 

7306: one is lead to set $n=h^{(2,1)}$ and identify 

7307: the fibers of the associated 

7308: $Sp$-bundle with $H^3(Y,\mathbb{C})$. The symplectic product is given by 

7309: the intersections on $H^3(Y,\bbC)$ as

7310: \beq \label{sympl-f}

7311:   \big<\alpha, \beta \big> = \int_Y \alpha \wedge \beta\ .

7312: \eeq

7313: The K\"ahler covariant derivatives of $\Omega$ are denoted by 

7314: $\chi_K$ as explicitly given in \eqref{Kod-form}.

7315: In terms of the symplectic basis $(\alpha_\Kh, \beta^\Kh)$

7316: introduced in \eqref{int-numbers1} both

7317:  $\Omega$ and $\chi_K$ enjoy the expansion

7318: \beq \label{def-chi}

7319:   \Omega = Z^\Kh\, \alpha_\Kh - \cF_\Kh\, \beta^\Kh\ , \qquad  

7320:   \chi_K = \chi^\Lh_{K}\, \alpha_\Lh - \chi_{\Lh|K}\, \beta^\Lh\ .

7321: \eeq 

7322: %Correspondingly $\Omega(z)$ transforms as given in 

7323: %\eqref{crescale} under K\"ahler transformations. We also introduce 

7324: %a section $\chi_K$ of $\cH\times \cL$ as 

7325: %\beq 

7326: %  \chi_K = \partial_{z^K} \Omega + K_{z^K} \Omega\ , 

7327: %\eeq

7328: %such that $\chi_K$ is the covariant derivative of $\Omega$.

7329: %On the fibers of $\cH$ one 

7330: %can choose a real symplectic basis $(\alpha_\Kh,\beta^\Kh)$ satisfying 

7331: %\beq

7332: %  \big< \alpha_\Kh, \beta^\Lh \big> =\delta^\Lh_\Kh\ , \quad 

7333: %  \big< \alpha_\Kh, \alpha_\Lh \big> = 

7334: %  \big<  \beta^\Kh , \beta^\Lh \big> =0\ , \qquad \Kh,\Lh=0,\ldots,n\ .

7335: %\eeq

7336: %In this basis the sections $\Omega(z)$ and $\chi(z,\bar z)$ admit the expansion

7337: The holomorphic functions $Z^\Kh(z)$ and $\cF_\Kh(z)$

7338: are called the periods of $\Omega$, while $\chi^\Lh_{K}(z,\bar z)$ and $\chi_{\Lh|K}(z,\bar z)$ 

7339: are the periods of $\chi_K$.  In terms of $Z^\Kh,\cF_\Kh$ the K\"ahler potential 

7340: \eqref{N=2KP} can be rewritten as in \eqref{csmetric}.

7341: 

7342: For every special K\"ahler manifold there exists

7343: a complex matrix $\cM_{\Kh \Lh}(z,\bar z)$ defined as 

7344: \beq \label{def-M}

7345:    \cM_{\Kh \Lh} = (\bar \chi_{\Kh|\bar M}\ \ \cF_\Kh)  (\bar \chi^\Lh_{\bar M}\ \ Z^\Lh)^{-1}\ ,

7346: \eeq

7347: where $\chi^{\Lh}_{K}$ and $\chi_{\Lh| K}$ are given in \eqref{def-chi}.

7348: Furthermore, one extracts from \eqref{def-M} the  

7349: identities

7350: \bea \label{ML-hf}

7351:   \cF_{\Kh} = \cM_{\Kh \Lh} Z^\Lh\ , \qquad   \chi_{\Lh| K} = \bar \cM_{\Lh \Mh} \chi^{\Mh}_{K}\ ,

7352: \eea

7353: which can be used to rewrite \eqref{N=2KP} as 

7354: \bea \label{spconst}

7355:    G_{M \bar N} & = & -2 e^{K} \chi^\Kh_M\, \I \cM_{\Kh \Lh}\, \bar \chi^\Lh_{\bar N}\ , \qquad

7356:    1 \ = \ -2 e^{K} Z^\Kh\, \text{Im}\, \cM_{\Kh \Lh}\,\bar Z^\Lh \ , \\

7357:    0 & = & -2 \bar \chi^\Kh_{\bar M}\, \text{Im}\, \cM_{\Kh \Lh}\,\bar Z^\Lh \ .\nn

7358: \eea

7359: 

7360: If one assumes that  

7361: the Jacobian matrix $\partial_{z^L}\big(Z^K/Z^0 \big)$ is invertible 

7362: $\cF_\Kh$ is the derivative of a holomorphic prepotential $\cF$ with respect to the periods $Z^\Kh$. 

7363: It is homogeneous of degree two and obeys

7364: \beq

7365:    \cF = \tfrac{1}{2}Z^\Kh \cF_{\Kh}\ , \qquad  \cF_\Kh =\partial_{Z^\Kh} \cF\ , \qquad  

7366:    \cF_{\Kh \Lh} =\partial_{Z^\Kh} \cF_{\Lh}\ ,

7367: \qquad \cF_{\Lh}= Z^\Kh \cF_{\Kh \Lh}\ ,

7368: \eeq

7369: which implies that $\cF_{\Kh \Lh}(Z)$ is invariant

7370: under rescalings of $Z^\Kh$. 

7371: Notice that $\cF$ is only invariant under a restricted 

7372: class of symplectic transformations

7373: and thus depends on the choice of symplectic basis. 

7374: 

7375: The complex matrix 

7376: $\cM_{\Kh \Lh}$ defined in \eqref{def-M} can be rewritten in terms of the 

7377: periods $Z^\Kh$ and the matrix $\cF_{\Kh\Lh}(Z)$ as

7378: \bea \label{gauge-c}

7379:    \cM_{\Kh \Lh}=\overline{ \mathcal{F}}_{\Kh \Lh}+2i \frac{(\text{Im}\; \mathcal{F})_{\Kh \Mh} Z^\Mh

7380:    (\text{Im}\; \mathcal{F})_{\Lh \Nh}Z^\Nh }{Z^\Nh(\text{Im}\; \mathcal{F})_{\Nh\Mh} 

7381:     Z^\Mh}\ .

7382: \eea

7383: 

7384: 

7385: 

7386: Whenever the 

7387: Jacobian matrix $\partial_{z^L}\big(Z^K/Z^0 \big)$ is invertible 

7388: the $Z^\Kh$ can be viewed as projective coordinates of $\mathbb{P}_{h^{(2,1)}+1}$.

7389: Going to a special gauge, i.e.~fixing the K\"ahler transformations

7390: \eqref{crescale}, one introduces

7391: special coordinates $z^K$ by setting $z^K=Z^K/Z^0$. 

7392: Due to the homogeneity of $\cF$ it is possible to define

7393: a holomorphic prepotential $f(z)$ which only depends on the special

7394: coordinates as

7395: \beq\label{def-f}

7396:   \cF(Z) = (Z^0)^2 f(z)\ .

7397: \eeq

7398: In terms of $f$  the K\"ahler potential  given in \eqref{N=2KP} 

7399: reads

7400: \beq \label{Kinz}

7401:   K\ =\ - \ln i|Z^0|^2 \big[2(f-\bar f)-(\partial_K\, f + \partial_{\bar K} \bar f)(z^K - \bar z^K) \big]\ . 

7402: \eeq

7403: 

7404: A special example of the situation just discussed is the moduli

7405: space spanned by the complexified K\"ahler deformations $t^A$ introduced 

7406: in \eqref{4d-dilaton}. These fields can be interpreted as special coordinates on 

7407: a special K\"ahler manifold $\cM^{\rm SK}(t,\bar t)$ \cite{CdO}. 

7408: The K\"ahler potential of the metric $G_{AB}$ given 

7409: in \eqref{Kmetric} is of the form \eqref{Kinz} with

7410: \beq \label{pre-K}

7411:   f(t)=-\tfrac{1}6 \cK_{ABC} t^A t^B t^C\ .

7412: \eeq

7413: Furthermore, inserting \eqref{pre-K}

7414: into \eqref{gauge-c} using \eqref{def-f} 

7415: one determine the gauge-couplings  $\cN_{\Ah \Bh}(t,\bar t)$ 

7416: to be

7417: \bea \label{def-cN}

7418:   \text{Re} \cN &=& \ \

7419:   \left(\ba{cc}-\frac13 \cK_{ABC}b^A b^B b^C &  \frac12 \cK_{ABC} b^B b^C \\

7420:               \frac12 \cK_{ABC} b^B b^C & - \cK_{ABC}b^C  \ea \right)\ , \nn \\

7421:   \text{Im} \cN &=& -\frac{\cK}{6}

7422:   \left(\ba{cc}1 + 4 G_{AB}b^A b^B & -4 G_{AB}b^B  \\

7423:              - 4 G_{AB}b^B &  4 G_{AB}  \ea \right)\ , \nn \\

7424:   (\text{Im} \cN)^{-1} &=& - \frac{6}{\cK}

7425:   \left(\ba{cc}1 & b^A  \\

7426:          b^A &  \frac14 G^{AB} + b^A b^B \ea \right)\ ,

7427: \eea

7428: where $G_{AB}$ is given in \eqref{Kmetric}.

7429: 

7430: 

7431: % In this appendix we briefly summarize $N=2$ supergravity in four space-time 

7432: % dimensions. We consentrate on effective action for the bosonic fields and the geometry of 

7433: % the corresponding $N=2$ moduli spaces. A more detailed disscussion also including possible gaugings 

7434: % can be found, for example, in ref.~\cite{N=2review}. 

7435: 

7436: % The $N=2$ theories under consideration consist of a gravity multiplet with bosonic 

7437: % components $(g_{\mu \nu},A^0)$, $n_v$ vector multiplets with bosonic components $(A^A,z^A)$

7438: % and $n_h$ hypermultiplets with bosonic components $(\tilde q^{u},\tilde q^{u+1},\tilde q^{u+2},\tilde q^{u+3})$.  

7439: % The kinetic terms and couplings of these fields are encoded by the effective

7440: % action \cite{N=2review} 

7441: % \bea \label{N=2act}

7442: %   S^{(4)}_{ N=2} & = &\int-\ \tfrac12 R * \mathbf{1}\ -\ h_{uv}\, d\tilde q^u \wedge * d\tilde q^v \\

7443: %                  &&  - G_{K \bar L}\, dz^K \wedge *  d\bar z^L + \tfrac14 \I \cM_{\Kh \Lh}\, F^{\Kh} \wedge *

7444: %                      F^{\Lh}

7445: %                    +  \tfrac14 \R \cM_{\Kh \Lh}\, F^{\Kh} \wedge  F^{\Lh}\ , \nn 

7446: % \eea

7447: % where $A^\Kh=(A^0,A^K)$ and $F^{\Ah} = dA^{\Ah}$. The first line in \eqref{N=2act}

7448: % contains the standard Einstein-Hilbert term as well as the kinetic terms for 

7449: % the hypermultiplets. $N=2$ supersymmetry demands $h_{uv}$ to be a quaternionic 

7450: % metric such that the scalars $\tilde q^u$ span a quaternionic manifold $\cM^{\rm Q}_{4 n_h}$ of real 

7451: % dimension $4n_h$ \cite{N=2review}. The second line in \eqref{N=2act} consists of

7452: % the kinetic terms for the vector multiplets. In $N=2$ supergravity the metric

7453: % $G_{K \bar L}$ has to be special K\"ahler implying that the scalars $z^K$ span 

7454: % a special K\"ahler manifold $\cM^{\rm SK}_{2n_v}$ of real dimension $2n_v$ \cite{N=2review,CRTV,Freed}. 

7455: % The complex gauge coupling matrix $\cM_{\Kh \Lh}(z,\bar z)$ is generically a non-trivial function 

7456: % of $z^K,\bar z^K$, but by the constraints of $N=2$ supersymmetry independent of the 

7457: % hypermultiplet scalars $\tilde q^u$. As we will disscuss momentarily $\cM_{\Kh \Lh}$ and

7458: % the metric $G_{K \hat L}$ are not independent, but rather linked in an intriquing 

7459: % way. In summary the $N=2$ moduli space admits the local product form 

7460: % $\cM^{\rm Q}_{4 n_h} \times \cM^{\rm SK}_{2n_v}$.

7461: % In the following we will concentrate on the special K\"ahler manifold 

7462: % $\cM^{\rm SK}_{2n_v}$ obtained in Calabi-Yau compactifications of

7463: % type II supergravity theories. We refer the reader to \cite{N=2review} for a 

7464: % detailed discussion of the quaternionic space $\cM^{\rm Q}_{4 n_h}$.

7465:    

7466: % In compactifications of type IIA or type IIB supergravity on 

7467: % Calabi-Yau manifolds the moduli space $\cM^{\rm SK}$ 

7468: % is spanned by the metric deformations of the Calabi-Yau metric \cite{CdO,Strominger2,CRTV}.

7469: % To make this more precise, let us summarize some 

7470: % facts about special geometry.

7471: % %More precely, the manifold $\cM^{\rm SK}$  consists

7472: % %of the $n_v = h^{2,1}(Y)$ complex structure deformations $z^K$ in type IIB (c.f.~section \ref{revIIB}), 

7473: % %while it consists of the $n_v = h^{1,1}(Y)$

7474: % %complexified K\"ahler deformations $t^A$ in type IIA (c.f.~section \ref{revIIA}).

7475: % %

7476: % %In this appendix we briefly summarize the $N=2$ special geometry

7477: % %of the Calabi-Yau 

7478: % %moduli space. A more detailed discussion can be found,

7479: % %for example, in refs.\ \cite{CdO,Strominger2,Freed,N=2review,CRTV}.

7480: % A special K\"ahler manifold $\cM^{\rm SK}$ is a Hodge-K\"ahler manifold (with line bundle $\cL$)

7481: % of real dimension $2n$ with associated 

7482: % holomorphic flat $Sp(2n+2,\mathbb{R})$ vector bundle $\mathcal{H}$ over $\cM^{\rm SK}$. Furthermore 

7483: % there exists a holomorphic section $\Omega(z)$ of $\cL$ such that 

7484: % \beq\label{N=2KP}

7485: %   K(z,\bar z) = - \ln i \big<\Omega(z) , \bar \Omega(\bar z)  \big>\ , \qquad 

7486: %   \big<\Omega, \partial_{z^K} \Omega\big> = 0\ , \qquad K=1,\ldots n\ ,

7487: % \eeq

7488: % where $K$ is the K\"ahler potential of $\cM^{\rm SK}$ and 

7489: % $\big<\cdot,\cdot \big>$ is the 

7490: % symplectic product on the fibers. 

7491: % This is precisely what one encounters in the moduli space

7492: % of the complex structure deformations of a Calabi-Yau manifold

7493: % with $\Omega$ being the holomorphic three-form.

7494: % In this case 

7495: % one is lead to set $n=h^{(2,1)}$ and identify 

7496: % the fibers of the associated 

7497: % $Sp$-bundle with $H^3(Y,\mathbb{C})$. The symplectic product is given by 

7498: % the intersections on $H^3(Y,\bbC)$ as

7499: % \beq \label{sympl-f}

7500: %   \big<\alpha, \beta \big> = \int_Y \alpha \wedge \beta\ .

7501: % \eeq

7502: % The K\"ahler covariant derivatives of $\Omega$ are denoted by 

7503: % $\chi_K$ as explicitly given in \eqref{Kod-form}.

7504: % In terms of the symplectic basis $(\alpha_\Kh, \beta^\Kh)$

7505: % introduced in \eqref{int-numbers1} both

7506: %  $\Omega$ and $\chi_K$ enjoy the expansion

7507: % \beq \label{def-chi}

7508: %   \Omega = Z^\Kh\, \alpha_\Kh - \cF_\Kh\, \beta^\Kh\ , \qquad  

7509: %   \chi_K = \chi^\Lh_{K}\, \alpha_\Lh - \chi_{\Lh|K}\, \beta^\Lh\ .

7510: % \eeq 

7511: % %Correspondingly $\Omega(z)$ transforms as given in 

7512: % %\eqref{crescale} under K\"ahler transformations. We also introduce 

7513: % %a section $\chi_K$ of $\cH\times \cL$ as 

7514: % %\beq 

7515: % %  \chi_K = \partial_{z^K} \Omega + K_{z^K} \Omega\ , 

7516: % %\eeq

7517: % %such that $\chi_K$ is the covariant derivative of $\Omega$.

7518: % %On the fibers of $\cH$ one 

7519: % %can choose a real symplectic basis $(\alpha_\Kh,\beta^\Kh)$ satisfying 

7520: % %\beq

7521: % %  \big< \alpha_\Kh, \beta^\Lh \big> =\delta^\Lh_\Kh\ , \quad 

7522: % %  \big< \alpha_\Kh, \alpha_\Lh \big> = 

7523: % %  \big<  \beta^\Kh , \beta^\Lh \big> =0\ , \qquad \Kh,\Lh=0,\ldots,n\ .

7524: % %\eeq

7525: % %In this basis the sections $\Omega(z)$ and $\chi(z,\bar z)$ admit the expansion

7526: % The holomorphic functions $Z^\Kh(z)$ and $\cF_\Kh(z)$

7527: % are called the periods of $\Omega$, while $\chi^\Lh_{K}(z,\bar z)$ and $\chi_{\Lh|K}(z,\bar z)$ 

7528: % are the periods of $\chi_K$.  In terms of $Z^\Kh,\cF_\Kh$ the K\"ahler potential 

7529: % \eqref{N=2KP} can be rewritten as in \eqref{csmetric}.

7530: 

7531: % For every special K\"ahler manifold there exists

7532: % a complex matrix $\cM_{\Kh \Lh}(z,\bar z)$ defined as 

7533: % \beq \label{def-M}

7534: %    \cM_{\Kh \Lh} = (\bar \chi_{\Kh|\bar M}\ \ \cF_\Kh)  (\bar \chi^\Lh_{\bar M}\ \ Z^\Lh)^{-1}\ ,

7535: % \eeq

7536: % where $\chi^{\Lh}_{K}$ and $\chi_{\Lh| K}$ are given in \eqref{def-chi}.

7537: % It is precisely this matrix, which encondes the gauge-couplings of the vector 

7538: % fields in the effective action \eqref{N=2act}. 

7539: % Furthermore, one extracts from \eqref{def-M} the  

7540: % identities

7541: % \bea \label{ML-hf}

7542: %   \cF_{\Kh} = \cM_{\Kh \Lh} Z^\Lh\ , \qquad   \chi_{\Lh| K} = \bar \cM_{\Lh \Mh} \chi^{\Mh}_{K}\ ,

7543: % \eea

7544: % which can be used to rewrite \eqref{N=2KP} as 

7545: % \bea \label{spconst}

7546: %    G_{M \bar N} & = & -2 e^{K} \chi^\Kh_M\, \I \cM_{\Kh \Lh}\, \bar \chi^\Lh_{\bar N}\ , \qquad

7547: %    1 \ = \ -2 e^{K} Z^\Kh\, \text{Im}\, \cM_{\Kh \Lh}\,\bar Z^\Lh \ , \\

7548: %    0 & = & -2 \bar \chi^\Kh_{\bar M}\, \text{Im}\, \cM_{\Kh \Lh}\,\bar Z^\Lh \ .\nn

7549: % \eea

7550: 

7551: % If one assumes that  

7552: % the Jacobian matrix $\partial_{z^L}\big(Z^K/Z^0 \big)$ is invertible 

7553: % $\cF_\Kh$ is the derivative of a holomorphic prepotential $\cF$ with respect to the periods $Z^\Kh$. 

7554: % It is homogeneous of degree two and obeys

7555: % \beq

7556: %    \cF = \tfrac{1}{2}Z^\Kh \cF_{\Kh}\ , \qquad  \cF_\Kh =\partial_{Z^\Kh} \cF\ , \qquad  

7557: %    \cF_{\Kh \Lh} =\partial_{Z^\Kh} \cF_{\Lh}\ ,

7558: % \qquad \cF_{\Lh}= Z^\Kh \cF_{\Kh \Lh}\ ,

7559: % \eeq

7560: % which implies that $\cF_{\Kh \Lh}(Z)$ is invariant

7561: % under rescalings of $Z^\Kh$. 

7562: % Notice that $\cF$ is only invariant under a restricted 

7563: % class of symplectic transformations

7564: % and thus depends on the choice of symplectic basis. 

7565: 

7566: % The complex matrix 

7567: % $\cM_{\Kh \Lh}$ defined in \eqref{def-M} can be rewritten in terms of the 

7568: % periods $Z^\Kh$ and the matrix $\cF_{\Kh\Lh}(Z)$ as

7569: % \bea \label{gauge-c}

7570: %    \cM_{\Kh \Lh}=\overline{ \mathcal{F}}_{\Kh \Lh}+2i \frac{(\text{Im}\; \mathcal{F})_{\Kh \Mh} Z^\Mh

7571: %    (\text{Im}\; \mathcal{F})_{\Lh \Nh}Z^\Nh }{Z^\Nh(\text{Im}\; \mathcal{F})_{\Nh\Mh} 

7572: %     Z^\Mh}\ .

7573: % \eea

7574: 

7575: 

7576: 

7577: % Whenever the 

7578: % Jacobian matrix $\partial_{z^L}\big(Z^K/Z^0 \big)$ is invertible 

7579: % the $Z^\Kh$ can be viewed as projective coordinates of $\mathbb{P}_{h^{(2,1)}+1}$.

7580: % Going to a special gauge, i.e.~fixing the K\"ahler transformations

7581: % \eqref{crescale}, one introduces

7582: % special coordinates $z^K$ by setting $z^K=Z^K/Z^0$. 

7583: % Due to the homogeneity of $\cF$ it is possible to define

7584: % a holomorphic prepotential $f(z)$ which only depends on the special

7585: % coordinates as

7586: % \beq\label{def-f}

7587: %   \cF(Z) = (Z^0)^2 f(z)\ .

7588: % \eeq

7589: % In terms of $f$  the K\"ahler potential  given in \eqref{N=2KP} 

7590: % reads

7591: % \beq \label{Kinz}

7592: %   K\ =\ - \ln i|Z^0|^2 \big[2(f-\bar f)-(\partial_K\, f + \partial_{\bar K} \bar f)(z^K - \bar z^K) \big]\ . 

7593: % \eeq

7594: 

7595: % A second more specific example of a situation just discussed is the moduli

7596: % space spanned by the complexified K\"ahler deformations $t^A$ introduced 

7597: % in \eqref{4d-dilaton}. These fields can be interpreted as special coordinates on 

7598: % a special K\"ahler manifold $\cM^{\rm SK}(t,\bar t)$ \cite{CdO}. 

7599: % The  K\"ahler potential of the metric $G_{AB}$ given 

7600: % in \eqref{Kmetric} is of the form \eqref{Kinz} with

7601: % \beq \label{pre-K}

7602: %   f(t)=-\tfrac{1}6 \cK_{ABC} t^A t^B t^C\ .

7603: % \eeq

7604: % Furthermore, inserting \eqref{pre-K}

7605: % into \eqref{gauge-c} using \eqref{def-f} 

7606: % one determine the gauge-couplings  $\cN_{\Ah \Bh}(t,\bar t)$ 

7607: % to be

7608: % \bea \label{def-cN}

7609: %   \text{Re} \cN &=& \ \

7610: %   \left(\ba{cc}-\frac13 \cK_{ABC}b^A b^B b^C &  \frac12 \cK_{ABC} b^B b^C \\

7611: %               \frac12 \cK_{ABC} b^B b^C & - \cK_{ABC}b^C  \ea \right)\ , \nn \\

7612: %   \text{Im} \cN &=& -\frac{\cK}{6}

7613: %   \left(\ba{cc}1 + 4 G_{AB}b^A b^B & -4 G_{AB}b^B  \\

7614: %              - 4 G_{AB}b^B &  4 G_{AB}  \ea \right)\ , \nn \\

7615: %   (\text{Im} \cN)^{-1} &=& - \frac{6}{\cK}

7616: %   \left(\ba{cc}1 & b^A  \\

7617: %          b^A &  \frac14 G^{AB} + b^A b^B \ea \right)\ ,

7618: % \eea

7619: % where $G_{AB}$ is given in \eqref{Kmetric}.

7620: 

7621: 

7622: % % In the appendix \ref{CofK} we will also need another formulation 

7623: % % of special geometry. Instead of introducing a line bundle $\cL$

7624: % % with holomorphic section $\Omega(z)$ transforming as in 

7625: % % \eqref{crescale} we now switch to the associated $U(1)$ principal bundle $\mathcal{P}$.

7626: % % A section $\varphi(z,\bar z)$ of this bundle transforms under K\"ahler transformations as 

7627: % % $\varphi \rightarrow e^{-i \I f(z)} \varphi$ and we define the connection as

7628: % % $(\partial_K + \tfrac12 \partial_K K) \varphi$ and 

7629: % % $(\partial_{\bar K} - \tfrac12 \partial_{\bar K} K) \varphi$.

7630: % % Thus the to $\Omega \in \cL$ corresponding covariantly holomorphic 

7631: % % section $V \in \mathcal{P}$ is given by 

7632: % % \bea \label{def-V}

7633: % % V = e^{K/2} \Omega\ , \qquad   V(z,\bar z) = \cL^\Kh(z,\bar z) \alpha_\Kh - \cM_\Kh(z,\bar z) \beta^\Kh\ .

7634: % % \eea

7635: % % Its covariant derivative we denote by

7636: % % \bea \label{fh-def}

7637: % %   U_K \equiv (\partial_K + \tfrac12 \partial_K K) V(z,\bar z) = f^\Lh_{K} \alpha_\Lh - h_{\Lh|K} \beta^\Lh\ .

7638: % % \eea

7639: % % Comparing \eqref{fh-def} with equation  \eqref{Kod-form}  one identifies 

7640: % % $U_{K} = e^{\frac{K}{2}}  \chi_K(z)$. In terms of $U_K$ and $V$ the 

7641: % % special K\"ahler conditions \eqref{N=2KP} and 

7642: % % \eqref{chi_barchi} take the form

7643: % % \beq \label{fcontr}

7644: % %  \big<U_K,\bar U_{L}\big> = -iG_{K \bar L}\ , \quad  

7645: % %  \big<V,\bar V\big> =- i\ , \quad \big<V,U_K\big> = 0\ ,\quad  \big<U_K,U_L\big> = 0\ .

7646: % % \eeq

7647: % % One can now define a complex matrix $\cM$ as 

7648: % % \beq \label{gauge-cL}

7649: % %    \cM_{\Kh \Lh} = (\bar h_{\Kh|\bar M}\ \ \cM_\Kh)  (\bar f^\Lh_{\bar M}\ \ \cL^\Lh)^{-1}\ ,

7650: % % \eeq

7651: % % and show that it transforms under $Sp(2n+2,\mathbb{R})$ in the appropriate way to serve as a 

7652: % % gauge coupling matrix. This translates into equation \eqref{gauge-c}, when we assume the existence 

7653: % % of a prepotential. Moreover, it follows from \eqref{gauge-cL} that 

7654: % % \bea \label{ML-hf}

7655: % %   \cM_{\Kh} = \cM_{\Kh \Lh} \cL^\Lh\ , \quad   h_{\Kh| M} = \bar \cM_{\Kh \Lh} f^{\Lh}_{M}\ .

7656: % % \eea

7657: % % These equations can be used to give the component expressions of \eqref{fcontr}

7658: % % \bea \label{spconst}

7659: % %   G_{M \bar N} & = & -2 f^\Kh_M\, \text{Im}\, \cM_{\Kh \Lh}\, \bar f^\Lh_{\bar N}\ , \qquad

7660: % %   1 \ = \ -2 \cL^\Kh\, \text{Im}\, \cM_{\Kh \Lh}\,\bar \cL^\Lh \ , \\

7661: % %   0 & = & -2 \bar f^\Kh_{\bar M}\, \text{Im}\, \cM_{\Kh \Lh}\,\bar \cL^\Lh \ .\nn

7662: % % \eea

7663: 

7664: 

7665: 

7666: \section{Supergravity with several linear multiplets} \label{linm}

7667: 

7668: In this appendix we briefly discuss the dualization of several massless linear

7669: multiplets to chiral multiplets. We only discuss the bosonic component fields and do not include possible couplings to vector 

7670: multiplets. Our aim is to extract the K\"ahler potential 

7671: for the $N=1,D=4$ supergravity theory with all linear multiplets replaced by chiral ones.

7672: Let us begin by recalling the effective action for a set of 

7673: linear multiplets $(L^\lambda, D^\lambda_2)$ couplet to  chiral multiplets

7674: $N^k$. It takes the form\footnote{This action can be obtained by a straight forward 

7675: generalization of the action for one linear multiplet given in \cite{BGG}.}

7676: \bea\label{kinetic}

7677: \cL &=& -\tfrac{1}{2}R*\mathbf{1} - 

7678:   \tilde K_{N^k\bar N^l}\, dN^k \wedge * d \bar N^{l}

7679:   + \tfrac{1}{4} \tilde K_{L^\kappa L^\lambda}\, 

7680:   dL^\kappa \wedge * dL^\lambda \nn\\ 

7681:   && + \tfrac{1}{4} \tilde K_{ L^\kappa L^\lambda}\, dD^\kappa_2 \wedge * dD^\lambda_2

7682:      -  \tfrac{i}2\,  dD^\lambda_2 \wedge 

7683: \big(\tilde K_{L^\lambda N^k}\,dN^k -\tilde K_{L^\lambda \bar N^k}\,d\bar N^k\big)

7684: \ ,

7685: \eea

7686: where  $\tilde K(L,N,\bar N)$ is a function of the scalars $L^\lambda$ and

7687: the chiral multiplets $N^k$. The kinetic potential $\tilde K$ is the 

7688: analog of 

7689: the K\"ahler potential in the sense that it encodes the dynamics of the linear

7690: and chiral multiplets. In order to dualize the linear multiplets $(L^\lambda, D^\lambda_2)$ 

7691: into chiral multiplets $(L^\lambda,\tilde \xi_\lambda)$ one replaces

7692: $dD_2^\lambda$ by the form $D_3^\lambda$ and adds the term 

7693: \beq

7694: \cL \to \cL + \delta \cL\ , \qquad

7695:   \delta \cL\ =\  

7696:   - 2\tilde \xi_\lambda\, dD^\lambda_3\ =\ - 2 D_3^\lambda \wedge d\tilde \xi_\lambda\ ,\ 

7697: \eeq

7698: where $\tilde \xi_\lambda(x)$ is a Lagrange multiplier. Eliminating $\tilde \xi_\lambda$

7699: one finds that $dD^\lambda_3=0$ such that locally $D_3^\lambda=dD_2^\lambda$ as required.

7700: Alternatively one can consistently eliminate $D^\lambda_3$ by inserting 

7701: its equations of motion

7702: \beq

7703:   *D_3^\kappa = 4 \tilde K^{L^\kappa L^\lambda}\Big(d\tilde \xi_\lambda + \tfrac{i}4 

7704:   \big(\tilde K_{L^\lambda N^k}\,dN^k -\tilde K_{L^\lambda \bar N^k}\,d\bar N^k\big)\Big) 

7705: \eeq

7706: back into the Lagrangian \eqref{kinetic}. 

7707: The resulting dual Lagrangian takes the form

7708: \bea \label{eff_act1}

7709: \cL &=& -\tfrac{1}{2}R*\mathbf{1} - 

7710:   \tilde K_{N^k\bar N^l}\, dN^k \wedge * d \bar N^{l}

7711:   + \tfrac{1}{4} \tilde  K_{L^\kappa L^\lambda}\, 

7712:   dL^\kappa \wedge * dL^\lambda  \\ 

7713:   && + 4 \tilde K^{L^\kappa L^\lambda} \Big(d\tilde \xi_\kappa - \tfrac{1}2

7714:   \I \big(\tilde K_{L^\kappa N^l}\,dN^l\big)\Big)\wedge * 

7715:   \Big(d\tilde \xi_\lambda - \tfrac{1}2

7716:   \I \big(\tilde K_{L^\lambda N^k}\,dN^k\big)\Big) \ .\nn

7717: \eea

7718: Since we intend to use these results in the effective action for Calabi-Yau

7719: orientifolds, we make a further simplification. We demand that the kinetic potential

7720: $\tilde K$ is only a function of $L^\lambda$ and the imaginary part of $N^k$, which we 

7721: denote by $l^k=\I N^k$. This implies that all chiral fields $N^k$ admit a Peccei-Quinn 

7722: shift symmetry acting on the real parts of $N^k$ as it is indeed the case for the

7723: orientifold setups. Thus the effective Lagrangian \eqref{eff_act1} simplifies to 

7724: \bea \label{linaction}

7725: \cL &=& -\tfrac{1}{2}R*\mathbf{1} - 

7726:    \tfrac{1}{4}\tilde K_{l^k l^l}\, dN^k \wedge * d \bar N^l

7727:   + \tfrac{1}{4} \tilde K_{L^\kappa L^\lambda}\, 

7728:   dL^\kappa \wedge * dL^\lambda \\ 

7729:   && + 4 \tilde K^{L^\kappa L^\lambda}

7730:   \Big(d\tilde \xi_\kappa + \tfrac{1}{4}\tilde K_{L^\kappa l^l}\, d\, \text{Re}N^l\Big)\wedge * 

7731:   \Big(d\tilde \xi_\lambda + \tfrac{1}{4}\tilde K_{L^\lambda l^k}\, d\, \text{Re}N^k\Big) \ .\nn

7732: \eea

7733: This $N=1$ Lagrangian is written completely in terms of chiral multiplets and therefore 

7734: can be derived from a K\"ahler potential when choosing appropriate complex coordinates

7735: $N^k$ and $T_\lambda=(L^\lambda, \tilde \xi_\lambda)$.

7736: As we will see in a moment, a direct calculation yields that this K\"ahler potential 

7737: is the Legendre transform of $\tilde K$ with respect to the scalars $L^\kappa$. 

7738: It takes the 

7739: form 

7740: \beq \label{LegKP}

7741:  K(T,N) = \tilde K(L, N - \bar N)  - 2 (T_\kappa +\bar T_\kappa) L^\kappa

7742: \eeq

7743: where $L^\kappa(N,T)$ is a function of the complex fields $N^k,T_\lambda$. This 

7744: dependence is implicitly given via the definition of the coordinates $T_\lambda$

7745: \beq\label{defT}

7746: T_\lambda = i\tilde \xi_\lambda + \tfrac{1}{4}\tilde K_{L^\lambda}\ . 

7747: \eeq

7748: However, in order to calculate the K\"ahler metric, one only needs to determine 

7749: the derivatives of $L^\kappa(N,T)$ with respect to 

7750: $N^k,T_\lambda$. They are obtained by differentiating \eqref{defT} and simply read 

7751: \beq \label{derL}

7752:   {\partial L^\kappa}/{\partial T_\lambda} = 2 \tilde K^{L^\kappa L^\lambda}\ , \qquad 

7753:   {\partial L^\kappa}/{\partial N^l} = - \tfrac{1}{2i} \tilde K^{L^\kappa L^\lambda} \tilde K_{L^\lambda l^l}\ .

7754: \eeq

7755: Using these identities one easily calculates the first derivatives of the K\"ahler 

7756: potential \eqref{LegKP} as 

7757: \beq \label{Kder}

7758:   K_{T_\alpha} = -2 L^\alpha\ , \qquad K_{N^A} = \tfrac{1}{2i} \tilde K_{l^A}\ .

7759: \eeq

7760: Applying the equations \eqref{derL} once more when differentiating \eqref{Kder} 

7761: one finds the K\"ahler metric

7762: \bea \label{Km1}

7763:   K_{T_\alpha \bar T_\beta} &=& -4 \tilde K^{L^\alpha L^\beta}\ , \quad 

7764:   K_{T_\alpha \bar N^A}\ =\ i \tilde K^{L^\alpha L^\beta} \tilde K_{L^\beta l^A}\ , \nn \\ 

7765:   K_{N^A \bar N^B} &=& \tfrac{1}{4} \tilde K_{l^A l^B} - \tfrac{1}{4} 

7766:                      \tilde K_{ l^A L^\alpha }\, \tilde K^{L^\alpha L^\beta}\, \tilde K_{L^\beta l^B}\ ,

7767: \eea

7768: with inverse

7769: \bea \label{invKm1}

7770:   K^{T_\alpha \bar T_\beta} &=& - \tfrac{1}{4} \tilde K_{L^\alpha L^\beta}

7771:                     + \tfrac{1}{4} \tilde K_{ l^A L^\alpha }\, \tilde K^{l^A l^B} \, \tilde K_{L^\beta l^B}\ , 

7772:                     \nn \\

7773:   K^{T_\alpha \bar N^B} & = & -i \tilde K^{l^A l^B}\, \tilde K_{ l^A L^\alpha }\ , \quad 

7774:   K^{N^A \bar N^B} \ = \ 4 \tilde K^{l^A l^B}\ .

7775: \eea

7776: Finally, one checks that $K(T,N)$ is indeed the K\"ahler potential for the chiral part of the 

7777: Lagrangian \eqref{linaction}. This is done by plugging in the definition 

7778: of $T_\kappa$ given in \eqref{defT} and the K\"ahler metric \eqref{Km1} into 

7779: \beq

7780:  \cL =  -\tfrac{1}{2}R*\mathbf{1} - K_{M^I \bar M^J}\ dM^I \wedge * d\bar M^J\ , 

7781: \eeq

7782: where $M^I=(N^k, T_\lambda)$.

7783: 

7784: 

7785: 

7786: 

7787: 

7788: 

7789: 

7790: 

7791: 

7792: 

7793: 

7794: 

7795: 

7796: 

7797: 

7798: 

7799: \newpage

7800: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

7801: \begin{thebibliography}{99}

7802: 

7803: %%%%%%%%%%%%%%%%%%%%%%%%%%%%

7804: % Introduction 

7805: \bibitem{revSusy}

7806: For a review see, for example, 

7807: H.~P.~Nilles, ``Supersymmetry, Supergravity And Particle Physics,''

7808: Phys.\ Rept.\  {\bf 110} (1984) 1;\\

7809: W.~Hollik, R.~R\"uckl and J.~Wess (eds.),

7810: ``Phenomenological aspects of supersymmetry'', Springer Lecture Notes, 1992;\\

7811: J.~A.~Bagger, ``Weak-scale supersymmetry: Theory and practice,'' arXiv:hep-ph/9604232; \\

7812: S.~P.~Martin, ``A supersymmetry primer,'' arXiv:hep-ph/9709356; \\

7813: J.~Louis, I.~Brunner and S.~J.~Huber,

7814:   ``The supersymmetric standard model,'' arXiv:hep-ph/9811341,

7815: and references therein.

7816: 

7817: \bibitem{LQG}

7818: A.~Ashtekar and J.~Lewandowski, 

7819: ``Background independent quantum gravity: A status report,''

7820: Class.\ Quant.\ Grav.\  {\bf 21}, R53 (2004) [arXiv:gr-qc/0404018].

7821: 

7822: \bibitem{GSWbook}

7823:   M.~B.~Green, J.~H.~Schwarz and E.~Witten,

7824:   ``Superstring Theory'', Vol. 1\& 2,  Cambridge University Press,

7825:    Cambridge, 1987.

7826: 

7827: \bibitem{JPbook}

7828: J.~Polchinski,

7829: ``String Theory'', Vol. 1\& 2,

7830: Cambridge University Press, Cambridge, 1998.

7831: 

7832: \bibitem{Zwiebach}

7833:   B.~Zwiebach, ``A first course in string theory'',

7834:   Cambridge University Press, Cambridge, 2004. 

7835: 

7836: \bibitem{Dual}

7837:   For a review see, for example,

7838:   P.~K.~Townsend,

7839:   ``Four lectures on M-theory,''

7840:   arXiv:hep-th/9612121;\\

7841:   J.~H.~Schwarz,

7842:   ``Lectures on superstring and M theory dualities,''

7843:   Nucl.\ Phys.\ Proc.\ Suppl.\  {\bf 55B} (1997) 1

7844:   [arXiv:hep-th/9607201]; \\

7845:    A.~Sen,

7846:   ``An introduction to non-perturbative string theory,''

7847:   arXiv:hep-th/9802051;\\

7848:   N.~A.~Obers and B.~Pioline,

7849:   ``U-duality and M-theory,''

7850:   Phys.\ Rept.\  {\bf 318} (1999) 113

7851:   [arXiv:hep-th/9809039];\\  

7852:   E.~Kiritsis,

7853:   ``Supersymmetry and duality in field theory and string theory,''

7854:   arXiv:hep-ph/9911525;\\

7855:   A.~Sen, ``An introduction to duality symmetries in string theory,''

7856:   {\it Prepared for Les Houches Summer School: Session 76: 

7857:     Euro Summer School on Unity of Fundamental Physics: 

7858:     Gravity, Gauge Theory and Strings,

7859:     Les Houches, France, 30 Jul - 31 Aug 2001},

7860:     and references therein.

7861: 

7862: 

7863: \bibitem{KaluzaKlein}

7864:   T.~Kaluza,

7865:   ``On The Problem Of Unity In Physics,''

7866:   Sitzungsber.\ Preuss.\ Akad.\ Wiss.\ Berlin (Math.\ Phys.\ ) {\bf 1921} (1921) 966;\\

7867:   O.~Klein,

7868:   ``Quantum Theory And Five-Dimensional Theory Of Relativity,''

7869:   Z.\ Phys.\  {\bf 37}, 895 (1926)

7870:   [Surveys High Energ.\ Phys.\  {\bf 5}, 241 (1986)].

7871: 

7872: 

7873: \bibitem{KK-review}

7874:   For a review see, for example, M.~J.~Duff, B.~E.~W.~Nilsson and C.~N.~Pope,

7875:   ``Kaluza-Klein Supergravity,''

7876:   Phys.\ Rept.\  {\bf 130} (1986) 1;\\ 

7877:   J.~M.~Overduin and P.~S.~Wesson,

7878:   ``Kaluza-Klein gravity,''

7879:   Phys.\ Rept.\  {\bf 283} (1997) 303

7880:   [arXiv:gr-qc/9805018], and references therein.

7881: 

7882: \bibitem{Greene}

7883:   For a review see, for example, B.~R.~Greene,

7884:   ``String theory on Calabi-Yau manifolds,''

7885:   arXiv:hep-th/9702155, and references therein.

7886: 

7887: 

7888: 

7889: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

7890: % Flux

7891: 

7892: \bibitem{Strominger1}

7893:   A.~Strominger, ``Superstrings With Torsion,''

7894:   Nucl.\ Phys.\ B {\bf 274}, 253 (1986).

7895: 

7896: \bibitem{Bachas}

7897:   C.~Bachas,

7898:   ``A Way to break supersymmetry,''

7899:   arXiv:hep-th/9503030;\\

7900:   J.~Polchinski and A.~Strominger,

7901:   ``New Vacua for Type II String Theory,''

7902:   Phys.\ Lett.\ B {\bf 388} (1996) 736

7903:   [arXiv:hep-th/9510227].

7904: 

7905: \bibitem{BB1}

7906: K.~Becker and M.~Becker,

7907: ``M-Theory on Eight-Manifolds,''

7908: Nucl.\ Phys.\ B {\bf 477}, 155 (1996)

7909: [arXiv:hep-th/9605053].

7910: 

7911: 

7912: \bibitem{Michelson}

7913: J.~Michelson,

7914: ``Compactifications of type IIB strings to four dimensions with  

7915:   non-trivial classical potential,''

7916: Nucl.\ Phys.\ B {\bf 495} (1997) 127

7917: [arXiv:hep-th/9610151].

7918: 

7919: \bibitem{Verlinde}

7920:   H.~Verlinde,

7921:   ``Holography and compactification,''

7922:   Nucl.\ Phys.\ B {\bf 580} (2000) 264

7923:   [arXiv:hep-th/9906182];\\

7924:   C.~S.~Chan, P.~L.~Paul and H.~Verlinde,

7925:   ``A note on warped string compactification,''

7926:   Nucl.\ Phys.\ B {\bf 581}, 156 (2000)

7927:   [arXiv:hep-th/0003236].

7928: 

7929: 

7930: \bibitem{GVW}S.~Gukov, C.~Vafa and E.~Witten,

7931: ``CFT's from Calabi-Yau four-folds,''

7932: Nucl.\ Phys.\ B {\bf 584}, 69 (2000)

7933: [Erratum-ibid.\ B {\bf 608}, 477 (2001)]

7934: [arXiv:hep-th/9906070].

7935: 

7936: 

7937: \bibitem{DRS}

7938: K.~Dasgupta, G.~Rajesh and S.~Sethi,

7939: ``M theory, orientifolds and G-flux,''

7940: JHEP {\bf 9908} (1999) 023

7941: [arXiv:hep-th/9908088].

7942: 

7943: 

7944: \bibitem{TV}

7945: T.~R.~Taylor and C.~Vafa,

7946: ``RR flux on Calabi-Yau and partial supersymmetry breaking,''

7947: Phys.\ Lett.\ B {\bf 474} (2000) 130

7948: [arXiv:hep-th/9912152].

7949: 

7950: 

7951: \bibitem{Mayr}

7952:   P.~Mayr,

7953:   ``On supersymmetry breaking in string theory and its realization in brane

7954:   worlds,''

7955:   Nucl.\ Phys.\ B {\bf 593} (2001) 99

7956:   [arXiv:hep-th/0003198];\\

7957:   P.~Mayr,

7958:   ``Stringy world branes and exponential hierarchies,''

7959:   JHEP {\bf 0011} (2000) 013

7960:   [arXiv:hep-th/0006204].

7961: 

7962: 

7963: 

7964: \bibitem{GSS}

7965: B.~R.~Greene, K.~Schalm and G.~Shiu,

7966: ``Warped compactifications in M and F theory,''

7967: Nucl.\ Phys.\ B {\bf 584} (2000) 480

7968: [arXiv:hep-th/0004103].

7969: 

7970: 

7971: \bibitem{GKP}

7972: S.~B.~Giddings, S.~Kachru and J.~Polchinski,

7973: ``Hierarchies from fluxes in string compactifications,''

7974: Phys.\ Rev.\ D {\bf 66} (2002) 106006

7975: [arXiv:hep-th/0105097].

7976: 

7977: 

7978: \bibitem{CKLT}

7979: G.~Curio, A.~Klemm, D.~L{\"u}st and S.~Theisen,

7980: ``On the vacuum structure of type II string compactifications on  Calabi-Yau spaces with H-fluxes,''

7981: Nucl.\ Phys.\ B {\bf 609} (2001) 3

7982: [arXiv:hep-th/0012213];\\

7983: G.~Curio and A.~Krause,

7984: ``Four-flux and warped heterotic M-theory compactifications,''

7985: Nucl.\ Phys.\ B {\bf 602}, 172 (2001)

7986: [arXiv:hep-th/0012152];\\

7987: G.~Curio, A.~Klemm, B.~K{\"o}rs and D.~L{\"u}st,

7988: ``Fluxes in heterotic and type II string compactifications,''

7989: Nucl.\ Phys.\ B {\bf 620} (2002) 237

7990: [arXiv:hep-th/0106155].

7991: 

7992: \bibitem{HL}

7993: M.~Haack and J.~Louis,

7994: ``Duality in heterotic vacua with four supercharges,''

7995: Nucl.\ Phys.\ B {\bf 575} (2000) 107

7996: [arXiv:hep-th/9912181]; \\

7997: ``M-theory compactified on Calabi-Yau fourfolds with background flux,''

7998: Phys.\ Lett.\ B {\bf 507} (2001) 296

7999: [arXiv:hep-th/0103068].

8000: 

8001: 

8002: \bibitem{BB2}

8003: K.~Becker and M.~Becker,

8004: ``Supersymmetry breaking, M-theory and fluxes,''

8005: JHEP {\bf 0107} (2001) 038

8006: [arXiv:hep-th/0107044].

8007: 

8008: \bibitem{DallAgata}

8009: G.~Dall'Agata,

8010: ``Type IIB supergravity compactified on a Calabi-Yau manifold with  H-fluxes,''

8011: JHEP {\bf 0111} (2001) 005

8012: [arXiv:hep-th/0107264].

8013: 

8014: \bibitem{KST}

8015: S.~Kachru, M.~B.~Schulz and S.~Trivedi,

8016: ``Moduli stabilization from fluxes in a simple IIB orientifold,''

8017: JHEP {\bf 0310} (2003) 007

8018: [arXiv:hep-th/0201028].

8019: 

8020: \bibitem{LM}

8021: J.~Louis and A.~Micu,

8022: ``Type II theories compactified on Calabi-Yau threefolds in the presence  of background fluxes,''

8023: Nucl.\ Phys.\ B {\bf 635} (2002) 395

8024: [arXiv:hep-th/0202168].

8025: 

8026: \bibitem{BBHL}

8027: K.~Becker, M.~Becker, M.~Haack and J.~Louis,

8028: ``Supersymmetry breaking and alpha'-corrections to flux induced  potentials,''

8029: JHEP {\bf 0206} (2002) 060

8030: [arXiv:hep-th/0204254].

8031: 

8032: \bibitem{FPo}

8033: S.~Ferrara and M.~Porrati,

8034: ``N = 1 no-scale supergravity from IIB orientifolds,''

8035: Phys.\ Lett.\ B {\bf 545} (2002) 411

8036: [arXiv:hep-th/0207135].

8037: 

8038: \bibitem{DWG} O.~DeWolfe and S.~B.~Giddings,

8039: ``Scales and hierarchies in warped compactifications and brane worlds,''

8040: Phys.\ Rev.\ D {\bf 67}, 066008 (2003)

8041: [arXiv:hep-th/0208123].

8042: 

8043: \bibitem{Soft1} 

8044: M.~Gra\~na,

8045: ``MSSM parameters from supergravity backgrounds,''

8046: Phys.\ Rev.\ D {\bf 67}, 066006 (2003)

8047: [arXiv:hep-th/0209200];\\

8048: B.~K\"ors and P.~Nath,

8049: ``Effective action and soft supersymmetry breaking for intersecting D-brane

8050: models,''

8051: Nucl.\ Phys.\ B {\bf 681}, 77 (2004)

8052: [arXiv:hep-th/0309167];\\

8053: P.~G.~C\'amara, L.~E.~Ib\'a\~nez and A.~M.~Uranga,

8054: ``Flux-induced SUSY-breaking soft terms,''

8055: Nucl.\ Phys.\ B {\bf 689} (2004) 195

8056: [arXiv:hep-th/0311241].

8057: 

8058: \bibitem{GGJL}

8059: M.~Gra\~na, T.~W.~Grimm, H.~Jockers and J.~Louis,

8060:   %``Soft supersymmetry breaking in Calabi-Yau orientifolds with D-branes  and

8061:   %fluxes,''

8062:   Nucl.\ Phys.\ B {\bf 690} (2004) 21

8063:   [arXiv:hep-th/0312232].

8064: 

8065: \bibitem{Soft2}

8066: A.~Lawrence and J.~McGreevy,

8067: ``Local string models of soft supersymmetry breaking,''

8068: arXiv:hep-th/0401034;\\

8069: ``Remarks on branes, fluxes, and soft SUSY breaking,''

8070: arXiv:hep-th/0401233;

8071: D.~L\"ust, P.~Mayr, R.~Richter and S.~Stieberger,

8072: ``Scattering of gauge, matter, and moduli fields from intersecting branes,''

8073: Nucl.\ Phys.\ B {\bf 696}, 205 (2004)

8074: [arXiv:hep-th/0404134];\\

8075: D.~L\"ust, S.~Reffert and S.~Stieberger,

8076: ``Flux-induced soft supersymmetry breaking in chiral type IIb orientifolds with

8077: D3/D7-branes,''

8078: arXiv:hep-th/0406092;\\

8079: P.~G.~C\'amara, L.~E.~Ib\'a\~nez and A.~M.~Uranga,

8080: ``Flux-induced SUSY-breaking soft terms on D7-D3 brane systems,''

8081: arXiv:hep-th/0408036;\\

8082: D.~L\"ust, S.~Reffert and S.~Stieberger,

8083: ``MSSM with soft SUSY breaking terms from D7-branes with fluxes,''

8084: arXiv:hep-th/0410074;\\

8085: A.~Font and L.~E.~Ib\'a\~nez,

8086: ``SUSY-breaking soft terms in a MSSM magnetized D7-brane model,''

8087: arXiv:hep-th/0412150;\\

8088: D.~L\"ust, P.~Mayr, S.~Reffert and S.~Stieberger,

8089:   ``F-theory flux, destabilization of orientifolds and soft terms on

8090:   D7-branes,''

8091:   arXiv:hep-th/0501139;\\

8092: K.~Choi, A.~Falkowski, H.~P.~Nilles and M.~Olechowski,

8093:   ``Soft supersymmetry breaking in KKLT flux compactification,''

8094:   arXiv:hep-th/0503216.

8095: 

8096: 

8097: \bibitem{BKL}

8098: R.~Blumenhagen, D.~L{\"u}st and T.~R.~Taylor,

8099: ``Moduli stabilization in chiral type IIB orientifold models with fluxes,''

8100: Nucl.\ Phys.\ B {\bf 663} (2003) 319

8101: [arXiv:hep-th/0303016].

8102: 

8103: \bibitem{BHS}

8104:  M.~Berg, M.~Haack and H.~Samtleben,

8105:  ``Calabi-Yau fourfolds with flux and supersymmetry breaking,''

8106:  JHEP {\bf 0304} (2003) 046

8107:  [arXiv:hep-th/0212255].

8108: 

8109: \bibitem{KKLT}

8110:   S.~Kachru, R.~Kallosh, A.~Linde and S.~P.~Trivedi,

8111:   ``De Sitter vacua in string theory,''

8112:   Phys.\ Rev.\ D {\bf 68} (2003) 046005

8113:   [arXiv:hep-th/0301240].

8114: 

8115: 

8116: \bibitem{Ferrara}

8117: R.~D'Auria, S.~Ferrara, F.~Gargiulo, M.~Trigiante and S.~Vaula,

8118: ``N = 4 supergravity Lagrangian for type IIB on T**6/Z(2) in presence of  

8119: fluxes and D3-branes,''

8120: JHEP {\bf 0306} (2003) 045

8121: [arXiv:hep-th/0303049];\\

8122: C.~Angelantonj, S.~Ferrara and M.~Trigiante,

8123: ``New D = 4 gauged supergravities from N = 4 orientifolds with fluxes,''

8124: JHEP {\bf 0310} (2003) 015

8125: [arXiv:hep-th/0306185];\\

8126: for a review see,

8127: L.~Andrianopoli, S.~Ferrara and M.~Trigiante,

8128: ``Fluxes, supersymmetry breaking and gauged supergravity,''

8129: arXiv:hep-th/0307139;\\

8130: C.~Angelantonj, S.~Ferrara and M.~Trigiante,

8131:   ``Unusual gauged supergravities from type IIA and type IIB orientifolds,''

8132:   Phys.\ Lett.\ B {\bf 582}, 263 (2004)

8133:   [arXiv:hep-th/0310136].

8134: C.~Angelantonj, R.~D'Auria, S.~Ferrara and M.~Trigiante,

8135: ``K3 x T**2/Z(2) orientifolds with fluxes, open string moduli and  critical points,''

8136: Phys.\ Lett.\ B {\bf 583}, 331 (2004)

8137: [arXiv:hep-th/0312019].

8138: 

8139: 

8140: \bibitem{BHK}

8141: M.~Berg, M.~Haack and B.~K{\"o}rs,

8142: ``An orientifold with fluxes and branes via T-duality,''

8143: Nucl.\ Phys.\ B {\bf 669} (2003) 3

8144: [arXiv:hep-th/0305183];\\

8145: M.~Berg, M.~Haack and B.~K\"ors,

8146: ``Brane/Flux Interactions in Orientifolds,''

8147: arXiv:hep-th/0312172.

8148: 

8149: \bibitem{deAlwis:2003sn}

8150: S.~P.~de Alwis,

8151: ``On potentials from fluxes,''

8152: Phys.\ Rev.\ D {\bf 68} (2003) 126001

8153: [arXiv:hep-th/0307084];\\

8154: A.~Buchel,

8155: ``On effective action of string theory flux compactifications,''

8156: Phys.\ Rev.\ D {\bf 69}, 106004 (2004)

8157: [arXiv:hep-th/0312076];\\

8158: A.~Giryavets, S.~Kachru, P.~K.~Tripathy and S.~P.~Trivedi,

8159: ``Flux compactifications on Calabi-Yau threefolds,''

8160: JHEP {\bf 0404} (2004) 003

8161: [arXiv:hep-th/0312104];\\

8162: R.~Brustein and S.~P.~de Alwis,

8163: ``Moduli potentials in string compactifications with fluxes: Mapping the

8164: discretuum,''

8165: Phys.\ Rev.\ D {\bf 69}, 126006 (2004)

8166: [arXiv:hep-th/0402088].

8167: 

8168: \bibitem{TGL1}

8169:   T.~W.~Grimm and J.~Louis,

8170:   ``The effective action of N = 1 Calabi-Yau orientifolds,''

8171:   Nucl.\ Phys.\ B {\bf 699} (2004) 387

8172:   [arXiv:hep-th/0403067].

8173: 

8174: \bibitem{KachruK}

8175:   S.~Kachru and A.~K.~Kashani-Poor,

8176:   ``Moduli potentials in type IIA compactifications with RR and NS flux,''

8177:   JHEP {\bf 0503} (2005) 066

8178:   [arXiv:hep-th/0411279].

8179:   %%CITATION = HEP-TH 0411279;%%

8180: 

8181: \bibitem{TGL2}

8182:   T.~W.~Grimm and J.~Louis,

8183:   ``The effective action of type IIA Calabi-Yau orientifolds,''

8184:   arXiv:hep-th/0412277, to appear in Nucl.\ Phys.\ B.

8185: 

8186: % end: flux

8187: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8188: 

8189: 

8190: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8191: % non-pert

8192: 

8193: 

8194: \bibitem{Witten}

8195:   E.~Witten,

8196:   ``Non-Perturbative Superpotentials In String Theory,''

8197:    Nucl.\ Phys.\ B {\bf 474} (1996) 343

8198:   [arXiv:hep-th/9604030].

8199: 

8200: \bibitem{HM}

8201: J.~A.~Harvey and G.~W.~Moore,

8202: ``Superpotentials and membrane instantons,''

8203: arXiv:hep-th/9907026.

8204: 

8205: \bibitem{non-pert}

8206:   For recent results see, for example,

8207:   L.~G\"orlich, S.~Kachru, P.~K.~Tripathy and S.~P.~Trivedi,

8208:   ``Gaugino condensation and nonperturbative superpotentials in flux

8209:   compactifications,''

8210:   arXiv:hep-th/0407130;

8211:   R.~Blumenhagen, M.~Cvetic, F.~Marchesano and G.~Shiu,

8212:   ``Chiral D-brane models with frozen open string moduli,''

8213:   JHEP {\bf 0503} (2005) 050

8214:   [arXiv:hep-th/0502095]\\

8215:   G.~Curio, A.~Krause and D.~L\"ust,

8216:   ``Moduli stabilization in the heterotic / IIB discretuum,''

8217:   arXiv:hep-th/0502168;\\

8218:   P.~K.~Tripathy and S.~P.~Trivedi,

8219:   ``D3 brane action and fermion zero modes in presence of background flux,''

8220:   arXiv:hep-th/0503072;\\

8221:   R.~Kallosh, A.~K.~Kashani-Poor and A.~Tomasiello,

8222:   ``Counting fermionic zero modes on M5 with fluxes,''

8223:   arXiv:hep-th/0503138;\\

8224:    P.~Berglund and P.~Mayr, 

8225:  ``Non-perturbative superpotentials in F-theory and string duality,''

8226:   arXiv:hep-th/0504058;\\ 

8227:   D.~L\"ust, S.~Reffert, W.~Schulgin and S.~Stieberger,

8228:   ``Moduli stabilization in type IIB orientifolds. I: Orbifold limits,''

8229:   arXiv:hep-th/0506090,

8230:   and references therein.

8231: 

8232: \bibitem{Curio}

8233:   G.~Curio and A.~Krause,

8234:   ``G-fluxes and non-perturbative stabilisation of heterotic M-theory,''

8235:   Nucl.\ Phys.\ B {\bf 643}, 131 (2002)

8236:   [arXiv:hep-th/0108220];\\

8237:   M.~Becker, G.~Curio and A.~Krause,

8238:   ``De Sitter vacua from heterotic M-theory,''

8239:   Nucl.\ Phys.\ B {\bf 693}, 223 (2004)

8240:   [arXiv:hep-th/0403027].

8241: 

8242: 

8243: \bibitem{gaugino}

8244:   J.~P.~Derendinger, L.~E.~Ib\'a\~nez and H.~P.~Nilles,

8245:  ``On The Low-Energy D = 4, N=1 Supergravity Theory Extracted From The D = 10,

8246:  N=1 Superstring,''

8247:   Phys.\ Lett.\ B {\bf 155} (1985) 65;\\

8248:   M.~Dine, R.~Rohm, N.~Seiberg and E.~Witten,

8249:   ``Gluino Condensation In Superstring Models,''

8250:   Phys.\ Lett.\ B {\bf 156} (1985) 55.

8251:   C.~P.~Burgess, J.~P.~Derendinger, F.~Quevedo and M.~Quiros,

8252:   ``On gaugino condensation with field-dependent gauge couplings,''

8253:   Annals Phys.\  {\bf 250} (1996) 193

8254:   [arXiv:hep-th/9505171].

8255: 

8256:   

8257:   %%CITATION = HEP-TH 0502095;%%

8258: 

8259: 

8260: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8261: % stab all moduli

8262: 

8263: 

8264: \bibitem{DSFGK}

8265:   F.~Denef, M.~R.~Douglas, B.~Florea, A.~Grassi and S.~Kachru,

8266:   ``Fixing All Moduli in a Simple F-Theory Compactification,''

8267:   arXiv:hep-th/0503124.

8268: 

8269: 

8270: \bibitem{reviewPP}

8271: For a review see, for example,

8272: E.~Kiritsis,

8273: ``D-branes in standard model building, gravity and cosmology,''

8274: Fortsch.\ Phys.\  {\bf 52} (2004) 200

8275: [arXiv:hep-th/0310001];\\

8276: A.~M.~Uranga,

8277: ``Chiral four-dimensional string compactifications with intersecting

8278: D-branes,''

8279: Class.\ Quant.\ Grav.\  {\bf 20}, S373 (2003)

8280: [arXiv:hep-th/0301032];\\

8281: D.~L\"ust,

8282: ``Intersecting brane worlds: A path to the standard model?,''

8283: Class.\ Quant.\ Grav.\  {\bf 21} (2004) S1399

8284: [arXiv:hep-th/0401156];\\

8285: L.~E.~Ib\'a\~nez,

8286: ``The fluxed MSSM,''

8287: Phys.\ Rev.\ D {\bf 71}, 055005 (2005)

8288:   [arXiv:hep-ph/0408064]; \\

8289: R.~Blumenhagen,

8290: ``Recent progress in intersecting D-brane models,''

8291: arXiv:hep-th/0412025;

8292: R.~Blumenhagen, M.~Cvetic, P.~Langacker and G.~Shiu,

8293: ``Toward realistic intersecting D-brane models,''

8294: arXiv:hep-th/0502005,

8295: and references therein.

8296: 

8297: 

8298: \bibitem{reviewcosmo}

8299: For a review see, for example,

8300: A.~Linde,

8301: ``Prospects of inflation,''

8302: arXiv:hep-th/0402051;\\

8303: V.~Balasubramanian,

8304: ``Accelerating universes and string theory,''

8305: Class.\ Quant.\ Grav.\  {\bf 21} (2004) S1337

8306: [arXiv:hep-th/0404075];\\

8307: C.~P.~Burgess,

8308: ``Inflationary String Theory?,''

8309: arXiv:hep-th/0408037;

8310: A.~D.~Linde, ``Particle Physics and Inflationary Cosmology,''

8311: arXiv:hep-th/0503203,

8312: and references therein.

8313: 

8314: \bibitem{JP}

8315: A.~Sagnotti,

8316: ``Open Strings And Their Symmetry Groups,''

8317: arXiv:hep-th/0208020; \\

8318: J.~Dai, R.~G.~Leigh and J.~Polchinski,

8319: ``New Connections Between String Theories,''

8320: Mod.\ Phys.\ Lett.\ A {\bf 4} (1989) 2073;\\

8321: R.~G.~Leigh,

8322: ``Dirac-Born-Infeld Action From Dirichlet Sigma Model,''

8323: Mod.\ Phys.\ Lett.\ A {\bf 4} (1989) 2767;\\

8324: M.~Bianchi and A.~Sagnotti,

8325: ``On The Systematics Of Open String Theories,''

8326: Phys.\ Lett.\ B {\bf 247} (1990) 517;

8327: ``Twist Symmetry And Open String Wilson Lines,''

8328: Nucl.\ Phys.\ B {\bf 361} (1991) 519;\\

8329: P.~Horava,

8330: ``Strings On World Sheet Orbifolds,''

8331: Nucl.\ Phys.\ B {\bf 327} (1989) 461;\\

8332: J.~Polchinski,

8333: ``Dirichlet-Branes and Ramond-Ramond Charges,''

8334: Phys.\ Rev.\ Lett.\  {\bf 75} (1995) 4724

8335: [arXiv:hep-th/9510017];\\

8336: E.~G.~Gimon and J.~Polchinski,

8337: ``Consistency Conditions for Orientifolds and D-Manifolds,''

8338: Phys.\ Rev.\ D {\bf 54} (1996) 1667

8339: [arXiv:hep-th/9601038].

8340: 

8341: \bibitem{D-branes}

8342: For a review see, for example,

8343: J.~Polchinski, ``Lectures on D-branes,'' arXiv:hep-th/9611050; \\

8344: C.~P.~Bachas, ``Lectures on D-branes,'' arXiv:hep-th/9806199;\\

8345: C.~V.~Johnson, ``D-brane primer,'' arXiv:hep-th/0007170,

8346: and references therein.

8347: 

8348: \bibitem{AD}

8349: For a review see, for example,

8350: A.~Dabholkar,

8351: ``Lectures on orientifolds and duality,''

8352: [arXiv:hep-th/9804208];\\

8353: C.~Angelantonj and A.~Sagnotti,

8354: ``Open strings,''

8355: Phys.\ Rept.\  {\bf 371} (2002) 1

8356: [Erratum-ibid.\  {\bf 376} (2003) 339]

8357: [arXiv:hep-th/0204089], and references therein.

8358: 

8359: 

8360: \bibitem{RS}

8361:   L.~Randall and R.~Sundrum,

8362:   ``An alternative to compactification,''

8363:   Phys.\ Rev.\ Lett.\  {\bf 83}, 4690 (1999)

8364:   [arXiv:hep-th/9906064];\\

8365:   L.~Randall and R.~Sundrum,

8366:   ``A large mass hierarchy from a small extra dimension,''

8367:   Phys.\ Rev.\ Lett.\  {\bf 83}, 3370 (1999)

8368:   [arXiv:hep-ph/9905221].

8369: 

8370: 

8371: \bibitem{deWitSD}

8372:   B.~de Wit, D.~J.~Smit and N.~D.~Hari Dass,

8373:   ``Residual Supersymmetry Of Compactified D = 10 Supergravity,''

8374:   Nucl.\ Phys.\ B {\bf 283} (1987) 165;\\

8375:   J.~M.~Maldacena and C.~Nunez,

8376:   ``Supergravity description of field theories on curved manifolds and 

8377:     a no go theorem,''

8378:   Int.\ J.\ Mod.\ Phys.\ A {\bf 16} (2001) 822

8379:   [arXiv:hep-th/0007018].

8380: 

8381: 

8382: 

8383: \bibitem{reviewAdSCFT}

8384: For a review see, for example, 

8385: O.~Aharony, S.~S.~Gubser, J.~M.~Maldacena, H.~Ooguri and Y.~Oz,

8386: ``Large N field theories, string theory and gravity,''

8387: Phys.\ Rept.\  {\bf 323} (2000) 183

8388: [arXiv:hep-th/9905111],

8389: and references therein.

8390: 

8391: \bibitem{KL} V.~S.~Kaplunovsky and J.~Louis,

8392: ``Model independent analysis of soft terms in effective supergravity and in string theory,''

8393: Phys.\ Lett.\ B {\bf 306}, 269 (1993)

8394: [arXiv:hep-th/9303040].

8395: 

8396: \bibitem{BIM} 

8397: L.~E.~Ib\'a\~nez and D.~L\"ust,

8398: ``Duality anomaly cancellation, minimal string unification and the effective low-energy Lagrangian of 4-D strings,''

8399: Nucl.\ Phys.\ B {\bf 382} (1992) 305

8400: [arXiv:hep-th/9202046];\\

8401: A.~Brignole, L.~E.~Ib\'a\~nez and C.~Mu\~noz,

8402: ``Towards a theory of soft terms for the supersymmetric Standard Model,''

8403: Nucl.\ Phys.\ B {\bf 422}, 125 (1994)

8404: [Erratum-ibid.\ B {\bf 436}, 747 (1995)]

8405: [arXiv:hep-ph/9308271].

8406: 

8407: 

8408: \bibitem{BKQ}

8409:   C.~P.~Burgess, R.~Kallosh and F.~Quevedo,

8410:   ``de Sitter string vacua from supersymmetric D-terms,''

8411:   JHEP {\bf 0310} (2003) 056

8412:   [arXiv:hep-th/0309187].

8413: 

8414: \bibitem{Ori}

8415: C.~Angelantonj, M.~Bianchi, G.~Pradisi, A.~Sagnotti and Y.~S.~Stanev,

8416: ``Chiral asymmetry in four-dimensional open- string vacua,''

8417: Phys.\ Lett.\ B {\bf 385} (1996) 96

8418: [arXiv:hep-th/9606169];\\

8419: M.~Berkooz and R.~G.~Leigh,

8420: ``A D = 4 N = 1 orbifold of type I strings,''

8421: Nucl.\ Phys.\ B {\bf 483} (1997) 187

8422: [arXiv:hep-th/9605049];\\

8423: G.~Aldazabal, A.~Font, L.~E.~Ib\'a\~nez and G.~Violero,

8424: ``D = 4, N = 1, type IIB orientifolds,''

8425: Nucl.\ Phys.\ B {\bf 536} (1998) 29

8426: [arXiv:hep-th/9804026];\\

8427: M.~Cvetic, G.~Shiu and A.~M.~Uranga,

8428: ``Chiral four-dimensional N = 1 supersymmetric type IIA orientifolds from

8429: intersecting D6-branes,''

8430: Nucl.\ Phys.\ B {\bf 615} (2001) 3

8431: [arXiv:hep-th/0107166];

8432: 

8433: 

8434: 

8435: 

8436: \bibitem{AAHV}

8437: B.~Acharya, M.~Aganagic, K.~Hori and C.~Vafa,

8438: ``Orientifolds, mirror symmetry and superpotentials,''

8439: [arXiv:hep-th/0202208].

8440: 

8441: 

8442: \bibitem{BBKL}

8443: R.~Blumenhagen, V.~Braun, B.~K\"ors and D.~L\"ust,

8444: ``Orientifolds of K3 and Calabi-Yau manifolds with intersecting D-branes,''

8445: JHEP {\bf 0207} (2002) 026

8446: [arXiv:hep-th/0206038];

8447: R.~Blumenhagen, V.~Braun, B.~K\"ors and D.~L\"ust,

8448: ``The standard model on the quintic,''

8449: arXiv:hep-th/0210083.

8450: 

8451: 

8452: \bibitem{BH}

8453: I.~Brunner and K.~Hori,

8454: ``Orientifolds and mirror symmetry,''

8455: JHEP {\bf 0411} (2004) 005

8456: [arXiv:hep-th/0303135];\\

8457: I.~Brunner, K.~Hori, K.~Hosomichi and J.~Walcher,

8458: ``Orientifolds of Gepner models,''

8459: arXiv:hep-th/0401137.

8460: 

8461: \bibitem{revT-dual}

8462:   For a review see, for example,

8463:   E.~Alvarez, L.~Alvarez-Gaume and Y.~Lozano,

8464:   ``An introduction to T duality in string theory,''

8465:   Nucl.\ Phys.\ Proc.\ Suppl.\  {\bf 41} (1995) 1

8466:   [arXiv:hep-th/9410237]; \\

8467:   A.~Giveon, M.~Porrati and E.~Rabinovici,

8468:   ``Target space duality in string theory,''

8469:   Phys.\ Rept.\  {\bf 244} (1994) 77

8470:   [arXiv:hep-th/9401139].

8471:   and references therein.

8472: 

8473: \bibitem{SYZ}

8474:   A.~Strominger, S.~T.~Yau and E.~Zaslow,

8475:   ``Mirror symmetry is T-duality,''

8476:   Nucl.\ Phys.\ B {\bf 479} (1996) 243

8477:   [arXiv:hep-th/9606040].

8478: 

8479: \bibitem{Mirror}

8480:  For a review see, for example,

8481:  S.~Hosono, A.~Klemm and S.~Theisen,

8482:  ``Lectures on mirror symmetry,''

8483:  arXiv:hep-th/9403096;\\

8484:  K.~Hori, S.~Katz, A.~Klemm, R.~Pandharipande, R.~Thomas, C.~Vafa, R.~Vakil and E.~Zaslow,

8485:  ``Mirror symmetry'' (Clay Mathematics Monographs, Vol. 1,2).

8486:  and references therein.

8487: 

8488: \bibitem{Aspinwall}

8489:   For a review see, for example, P.~S.~Aspinwall,

8490:   ``D-branes on Calabi-Yau manifolds,''

8491:   arXiv:hep-th/0403166.

8492:   and references therein.

8493: 

8494: \bibitem{Vafa_NCY}

8495:   C.~Vafa,

8496:   ``Superstrings and topological strings at large N,''

8497:   J.\ Math.\ Phys.\  {\bf 42} (2001) 2798

8498:   [arXiv:hep-th/0008142].

8499: 

8500: \bibitem{GLMW}

8501: S.~Gurrieri, J.~Louis, A.~Micu and D.~Waldram,

8502: ``Mirror symmetry in generalized Calabi-Yau compactifications,''

8503: Nucl.\ Phys.\ B {\bf 654}, 61 (2003)

8504: [arXiv:hep-th/0211102].

8505: 

8506: \bibitem{Vafa}

8507: C.~Vafa,

8508: ``Evidence for F-Theory,''

8509: Nucl.\ Phys.\ B {\bf 469} (1996) 403

8510: [arXiv:hep-th/9602022];\\

8511: D.~R.~Morrison and C.~Vafa,

8512:   ``Compactifications of F-Theory on Calabi--Yau Threefolds -- I,''

8513:   Nucl.\ Phys.\ B {\bf 473}, 74 (1996)

8514:   [arXiv:hep-th/9602114];\\

8515: D.~R.~Morrison and C.~Vafa,

8516:   ``Compactifications of F-Theory on Calabi--Yau Threefolds -- II,''

8517:   Nucl.\ Phys.\ B {\bf 476}, 437 (1996)

8518:   [arXiv:hep-th/9603161].

8519: 

8520: \bibitem{Sen}

8521: A.~Sen,

8522: ``F-theory and Orientifolds,''

8523: Nucl.\ Phys.\ B {\bf 475} (1996) 562

8524: [arXiv:hep-th/9605150].

8525: 

8526: \bibitem{CJS}

8527: E.~Cremmer, B.~Julia and J.~Scherk,

8528: ``Supergravity Theory In 11 Dimensions,''

8529: Phys.\ Lett.\ B {\bf 76} (1978) 409.

8530: 

8531: 

8532: %%%%%%%%%%%%% Outline

8533: 

8534: \bibitem{BGHL}

8535: M.~Bodner and A.~C.~Cadavid,

8536: ``Dimensional Reduction Of Type IIb Supergravity And Exceptional Quaternionic Manifolds,''

8537: Class.\ Quant.\ Grav.\  {\bf 7} (1990) 829;\\

8538: R.~B\"ohm, H.~G\"unther, C.~Herrmann and J.~Louis,

8539: ``Compactification of type IIB string theory on Calabi-Yau threefolds,''

8540: Nucl.\ Phys.\ B {\bf 569} (2000) 229

8541: [arXiv:hep-th/9908007].

8542: 

8543: \bibitem{CP}

8544:  D.~Cremades, L.~E.~Ib\'a\~nez and F.~Marchesano,

8545:   ``Yukawa couplings in intersecting D-brane models,''

8546:   JHEP {\bf 0307}, 038 (2003)

8547:   [arXiv:hep-th/0302105];\\

8548:   M.~Cvetic and I.~Papadimitriou,

8549:   ``Conformal field theory couplings for intersecting D-branes on

8550:   orientifolds,''

8551:   Phys.\ Rev.\ D {\bf 68} (2003) 046001

8552:   [Erratum-ibid.\ D {\bf 70} (2004) 029903]

8553:   [arXiv:hep-th/0303083];\\

8554:   D.~L\"ust, P.~Mayr, R.~Richter and S.~Stieberger,

8555:   ``Scattering of gauge, matter, and moduli fields from intersecting branes,''

8556:   Nucl.\ Phys.\ B {\bf 696}, 205 (2004)

8557:   [arXiv:hep-th/0404134].

8558: 

8559: 

8560: \bibitem{BGG}

8561: P.~Binetruy, G.~Girardi and R.~Grimm,

8562: ``Supergravity couplings: A geometric formulation,''

8563: Phys.\ Rept.\  {\bf 343} (2001) 255

8564: [arXiv:hep-th/0005225].

8565:  

8566: 

8567: \bibitem{Hitchin2}

8568: N. Hitchin, ``The moduli space of special Lagrangian submanifolds,''

8569: dg-ga/9711002.

8570: 

8571: 

8572: \bibitem{HitchinGCM}

8573:   N.~Hitchin,

8574:   ``Generalized Calabi-Yau manifolds,''

8575:   Quart.\ J.\ Math.\ Oxford Ser.\  {\bf 54} (2003) 281

8576:   [arXiv:math.dg/0209099].

8577: 

8578: 

8579: \bibitem{Gualtieri}

8580:    M.~Gualtieri, `` Generalized complex geometry,''

8581:     math.DG/0401221. 

8582: 

8583: 

8584: \bibitem{mass_tensors}

8585:   R.~D'Auria and S.~Ferrara,

8586:   ``Dyonic masses from conformal field strengths in D even dimensions,''

8587:   Phys.\ Lett.\ B {\bf 606} (2005) 211

8588:   [arXiv:hep-th/0410051];\\  

8589:   J.~Louis and W.~Schulgin,

8590:   ``Massive tensor multiplets in N = 1 supersymmetry,''

8591:   Fortsch.\ Phys.\  {\bf 53} (2005) 235

8592:   [arXiv:hep-th/0410149];\\

8593:   U.~Theis,

8594:   ``Masses and dualities in extended Freedman-Townsend models,''

8595:   Phys.\ Lett.\ B {\bf 609} (2005) 402

8596:   [arXiv:hep-th/0412177];\\

8597:   S.~M.~Kuzenko,

8598:   ``On massive tensor multiplets,''

8599:   JHEP {\bf 0501} (2005) 041

8600:   [arXiv:hep-th/0412190];\\

8601:   R.~D'Auria, S.~Ferrara, M.~Trigiante and S.~Vaula,

8602:   ``N = 1 reductions of N = 2 supergravity in the presence of tensor

8603:   multiplets,''

8604:   JHEP {\bf 0503} (2005) 052

8605:   [arXiv:hep-th/0502219].

8606: 

8607: 

8608: \bibitem{DeWGKT}

8609:   O.~DeWolfe, A.~Giryavets, S.~Kachru and W.~Taylor,

8610:   ``Type IIA Moduli Stabilization,''

8611:   arXiv:hep-th/0505160.

8612: 

8613: \bibitem{VZ}

8614:   G.~Villadoro and F.~Zwirner,

8615:   ``N = 1 effective potential from dual type-IIA D6/O6 orientifolds with

8616:   general fluxes,''

8617:   JHEP {\bf 0506} (2005) 047

8618:   [arXiv:hep-th/0503169].

8619: 

8620: 

8621: 

8622: \bibitem{KMcG}

8623: S.~Kachru and J.~McGreevy,

8624: ``M-theory on manifolds of G(2) holonomy and type IIA orientifolds,''

8625: JHEP {\bf 0106} (2001) 027

8626: [arXiv:hep-th/0103223].

8627: 

8628: \bibitem{PT}

8629: G.~Papadopoulos and P.~K.~Townsend,

8630: ``Compactification of D = 11 supergravity on spaces of exceptional holonomy,''

8631: Phys.\ Lett.\ B {\bf 357} (1995) 300

8632: [arXiv:hep-th/9506150].

8633: 

8634: \bibitem{Hitchin1}

8635: N. Hitchin, ``The geometry of three-forms in six and seven dimensions,''

8636: math.DG/0010054.

8637: 

8638: 

8639: \bibitem{GPap}

8640: J.~Gutowski and G.~Papadopoulos,

8641: ``Moduli spaces and brane solitons for M theory compactifications on  holonomy

8642: G(2) manifolds,''

8643: Nucl.\ Phys.\ B {\bf 615} (2001) 237

8644: [arXiv:hep-th/0104105].

8645: 

8646: 

8647: \bibitem{BW}

8648: C.~Beasley and E.~Witten,

8649: ``A note on fluxes and superpotentials in M-theory compactifications on

8650: manifolds of G(2) holonomy,''

8651: JHEP {\bf 0207} (2002) 046

8652: [arXiv:hep-th/0203061].

8653: 

8654: 

8655: 

8656: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8657: %% 

8658: %%    Chapter 2

8659: %%

8660: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8661: 

8662: \bibitem{Huebsch}

8663:  T. H\"ubsch,

8664:  ``Calbi-Yau manifolds -- A Bestiary for Physicists'',

8665:  World Scientific Publishing (1994). 

8666: 

8667: 

8668: \bibitem{Tian}

8669: G.~Tian, in ``Mathematical aspects of string theory'', p.629, S.-T.Yau (ed.),

8670: World Scientific, Singapore, 1987 

8671: 

8672: 

8673: \bibitem{Strominger2}

8674: A.~Strominger,

8675: ``Special Geometry,''

8676: Commun.\ Math.\ Phys.\  {\bf 133} (1990) 163.

8677: 

8678: 

8679: \bibitem{CdO}

8680: P.~Candelas and X.~de la Ossa,

8681: ``Moduli Space Of Calabi-Yau Manifolds,''

8682: Nucl.\ Phys.\ B {\bf 355}, 455 (1991).

8683: 

8684: 

8685: 

8686: \bibitem{Strominger}

8687: A.~Strominger,

8688: ``Yukawa Couplings In Superstring Compactification,''

8689: Phys.\ Rev.\ Lett.\  {\bf 55} (1985) 2547.

8690: 

8691: 

8692: 

8693: \bibitem{BCF}

8694: M.~Bodner, A.~C.~Cadavid and S.~Ferrara,

8695: ``(2,2) Vacuum Configurations For Type Iia Superstrings: N=2 Supergravity

8696: Lagrangians And Algebraic Geometry,''

8697: Class.\ Quant.\ Grav.\  {\bf 8} (1991) 789.

8698: 

8699: 

8700: \bibitem{FS}

8701: S.~Ferrara and S.~Sabharwal,

8702: ``Dimensional Reduction Of Type II Superstrings,''

8703: Class.\ Quant.\ Grav.\  {\bf 6} (1989) L77;\\

8704: ``Quaternionic Manifolds For Type II Superstring Vacua Of Calabi-Yau Spaces,''

8705: Nucl.\ Phys.\ B {\bf 332} (1990) 317.

8706: 

8707: 

8708: \bibitem{N=2review}

8709: For a review of $N=2$ supergravity see, for example,

8710: L.~Andrianopoli, M.~Bertolini, A.~Ceresole, R.~D'Auria, S.~Ferrara, P.~Fre and T.~Magri,

8711: ``N = 2 supergravity and N = 2 super Yang-Mills theory on general scalar

8712: manifolds: Symplectic covariance, gaugings and the momentum map,''

8713: J.\ Geom.\ Phys.\  {\bf 23} (1997) 111

8714: [arXiv:hep-th/9605032].

8715: 

8716: 

8717: \bibitem{CDAF}

8718: H.~Suzuki,

8719:  ``Calabi-Yau compactification of type IIB string and a mass formula of the

8720: extreme black holes,''

8721: Mod.\ Phys.\ Lett.\ A {\bf 11} (1996) 623

8722: [arXiv:hep-th/9508001];\\

8723: A.~Ceresole, R.~D'Auria and S.~Ferrara,

8724: ``The Symplectic Structure of N=2 Supergravity and its Central Extension,''

8725: Nucl.\ Phys.\ Proc.\ Suppl.\  {\bf 46} (1996) 67

8726: [arXiv:hep-th/9509160].

8727: 

8728: 

8729: \bibitem{CFGi}

8730: S.~Cecotti, S.~Ferrara and L.~Girardello,

8731: ``Geometry Of Type Ii Superstrings And The Moduli Of Superconformal Field

8732: Theories,''

8733: Int.\ J.\ Mod.\ Phys.\ A {\bf 4} (1989) 2475.

8734: 

8735: 

8736: \bibitem{BVT}

8737: F.~Brandt,

8738: ``New N = 2 supersymmetric gauge theories: The double tensor multiplet  and

8739: its interactions,''

8740: Nucl.\ Phys.\ B {\bf 587} (2000) 543

8741: [arXiv:hep-th/0005086];\\

8742: U.~Theis and S.~Vandoren,

8743: ``N = 2 supersymmetric scalar-tensor couplings,''

8744: JHEP {\bf 0304} (2003) 042

8745: [arXiv:hep-th/0303048];\\

8746: G.~Dall'Agata, R.~D'Auria, L.~Sommovigo and S.~Vaula,

8747: ``D = 4, N = 2 gauged supergravity in the presence of tensor multiplets,''

8748: arXiv:hep-th/0312210.

8749: 

8750: 

8751: \bibitem{BaggerW}

8752: J.~Bagger and E.~Witten,

8753: ``Matter Couplings In N=2 Supergravity ,''

8754: Nucl.\ Phys.\ B {\bf 222} (1983) 1.

8755: 

8756: 

8757: \bibitem{dWvP}

8758: B.~de Wit and A.~Van Proeyen,

8759: ``Potentials And Symmetries Of General Gauged N=2 Supergravity - Yang-Mills

8760: Models,''

8761: Nucl.\ Phys.\ B {\bf 245} (1984) 89;\\

8762: B.~de Wit, P.~G.~Lauwers and A.~Van Proeyen,

8763: ``Lagrangians Of N=2 Supergravity - Matter Systems,''

8764: Nucl.\ Phys.\ B {\bf 255} (1985) 569.

8765: 

8766: \bibitem{CdOGP}

8767:   P.~Candelas, X.~C.~De La Ossa, P.~S.~Green and L.~Parkes,

8768:   ``A Pair Of Calabi-Yau Manifolds As An Exactly Soluble Superconformal

8769:   Theory,''

8770:   Nucl.\ Phys.\ B {\bf 359} (1991) 21.

8771:   %%CITATION = NUPHA,B359,21;%%

8772: 

8773: 

8774: 

8775: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8776: %% 

8777: %%    Chapter 3

8778: %%

8779: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8780: 

8781: \bibitem{JL}

8782:   H.~Jockers and J.~Louis,

8783:   ``The effective action of D7-branes in N = 1 Calabi-Yau orientifolds,''

8784:   Nucl.\ Phys.\ B {\bf 705} (2005) 167

8785:   [arXiv:hep-th/0409098];\\

8786:   H.~Jockers and J.~Louis,

8787:   ``D-terms and F-terms from D7-brane fluxes,''

8788:   arXiv:hep-th/0502059.

8789: 

8790: \bibitem{Leigh:jq}

8791: R.~G.~Leigh,

8792: ``Dirac-Born-Infeld Action From Dirichlet Sigma Model,''

8793: Mod.\ Phys.\ Lett.\ A {\bf 4} (1989) 2767.

8794: 

8795: 

8796: \bibitem{Douglas:1995bn}

8797: M.~R.~Douglas,

8798: ``Branes within branes,''

8799: arXiv:hep-th/9512077.

8800: 

8801: \bibitem{Myers}

8802:   R.~C.~Myers,

8803:   ``Dielectric-branes,''

8804:   JHEP {\bf 9912} (1999) 022

8805:   [arXiv:hep-th/9910053].  

8806: 

8807: 

8808: \bibitem{BBS}

8809: K.~Becker, M.~Becker and A.~Strominger,

8810: ``Five-branes, membranes and nonperturbative string theory,''

8811: Nucl.\ Phys.\ B {\bf 456} (1995) 130

8812: [arXiv:hep-th/9507158].

8813: 

8814: 

8815: \bibitem{MMMS}

8816:   M.~Marino, R.~Minasian, G.~W.~Moore and A.~Strominger,

8817:   ``Nonlinear instantons from supersymmetric p-branes,''

8818:   JHEP {\bf 0001} (2000) 005

8819:   [arXiv:hep-th/9911206].

8820: 

8821: 

8822: \bibitem{CU}

8823:   J.~F.~G.~Cascales and A.~M.~Uranga,

8824:   ``Branes on generalized calibrated submanifolds,''

8825:   JHEP {\bf 0411} (2004) 083

8826:   [arXiv:hep-th/0407132].

8827: 

8828: \bibitem{DHVW}

8829:   L.~J.~Dixon, J.~A.~Harvey, C.~Vafa and E.~Witten,

8830:   ``Strings On Orbifolds,''

8831:   Nucl.\ Phys.\ B {\bf 261} (1985) 678;\\

8832:   L.~J.~Dixon, J.~A.~Harvey, C.~Vafa and E.~Witten,

8833:   ``Strings On Orbifolds. 2,''

8834:   Nucl.\ Phys.\ B {\bf 274} (1986) 285.

8835: 

8836: 

8837: \bibitem{DP}

8838: A.~Dabholkar and J.~Park,

8839: ``Strings on Orientifolds,''

8840: Nucl.\ Phys.\ B {\bf 477} (1996) 701

8841: [arXiv:hep-th/9604178].

8842: 

8843: 

8844: \bibitem{HitchinLec}

8845: See, for example, N.~Hitchin,

8846: ``Lectures on Special Lagrangian Submanifolds'',

8847: Lectures given at the ICTP School on Differential Geometry April 1999.

8848: 

8849: 

8850: \bibitem{OSV}

8851: A.~Strominger,

8852: ``Macroscopic Entropy of $N=2$ Extremal Black Holes,''

8853: Phys.\ Lett.\ B {\bf 383} (1996) 39

8854: [arXiv:hep-th/9602111].

8855: 

8856: 

8857: \bibitem{BFM}

8858: S.~Sinha and C.~Vafa,

8859: ``SO and Sp Chern-Simons at large N,''

8860: arXiv:hep-th/0012136;\\

8861: D.~E.~Diaconescu, B.~Florea and A.~Misra,

8862: ``Orientifolds, unoriented instantons and localization,''

8863: JHEP {\bf 0307} (2003) 041

8864: [arXiv:hep-th/0305021];\\

8865: V.~Bouchard, B.~Florea and M.~Marino,

8866: ``Counting higher genus curves with crosscaps in Calabi-Yau orientifolds,''

8867: arXiv:hep-th/0405083.

8868: 

8869: 

8870: 

8871: 

8872: 

8873: 

8874: 

8875: \bibitem{AM}

8876: D.V.\ Alekseevsky, S.\ Marchiafava,

8877: ``Hermitian and K\"ahler Submanifolds of a Quaternionic K\"ahler Manifold,'' 

8878: Osaka J.\ Math.\ 38, 4 , (2001), 869.

8879: 

8880: 

8881: \bibitem{ADAF}

8882: L.~Andrianopoli, R.~D'Auria and S.~Ferrara,

8883: ``Supersymmetry reduction of N-extended supergravities in four  dimensions,''

8884: JHEP {\bf 0203} (2002) 025

8885: [arXiv:hep-th/0110277];\\

8886: ``Consistent reduction of N = 2 $\to$ N = 1 four dimensional supergravity  coupled to matter,''

8887: Nucl.\ Phys.\ B {\bf 628} (2002) 387

8888: [arXiv:hep-th/0112192].

8889: 

8890: 

8891: \bibitem{GP}

8892:   M.~Gra\~na and J.~Polchinski,

8893:   ``Supersymmetric three-form flux perturbations on AdS(5),''

8894:   Phys.\ Rev.\ D {\bf 63} (2001) 026001

8895:   [arXiv:hep-th/0009211];\\

8896:   M.~Gra\~na and J.~Polchinski,

8897:   ``Gauge / gravity duals with holomorphic dilaton,''

8898:   Phys.\ Rev.\ D {\bf 65} (2002) 126005

8899:   [arXiv:hep-th/0106014].

8900: 

8901: 

8902: \bibitem{FP}

8903:   A.~R.~Frey and J.~Polchinski,

8904:   ``N = 3 warped compactifications,''

8905:   Phys.\ Rev.\ D {\bf 65} (2002) 126009

8906:   [arXiv:hep-th/0201029].

8907: 

8908: 

8909: \bibitem{WB}

8910:   J.~Wess and J.~Bagger,

8911:   ``Supersymmetry And Supergravity,''

8912:   Princeton University Press, Princeton, 1992.

8913: 

8914: \bibitem{GGRS}

8915:   S. J.\ Gates, M.\ T.\ Grisaru, M.\ Rocek and W.\ Siegel,

8916:   ``Superspace: or one thousand and one lessons in supersymmetry,'' 

8917:   Frontiers in Physics, 58, Benjamin/Cummings, 1983. 

8918: 

8919: \bibitem{GLprep}

8920: I.~Benmachiche, T.~W.~Grimm and J.~Louis, in preperation.

8921: 

8922: 

8923: \bibitem{DAFT}

8924: R.~D'Auria, S.~Ferrara and M.~Trigiante,

8925: ``c-map,very special quaternionic geometry and dual K\"ahler spaces,''

8926: Phys.\ Lett.\ B {\bf 587} (2004) 138

8927: [arXiv:hep-th/0401161];\\

8928: R.~D'Auria, S.~Ferrara and M.~Trigiante,

8929: ``Homogeneous special manifolds, orientifolds and solvable coordinates,''

8930: Nucl.\ Phys.\ B {\bf 693} (2004) 261

8931: [arXiv:hep-th/0403204].

8932: 

8933: 

8934: 

8935: \bibitem{NS}

8936: E.~Cremmer, S.~Ferrara, C.~Kounnas and D.~V.~Nanopoulos,

8937: ``Naturally Vanishing Cosmological Constant In N=1 Supergravity,''

8938: Phys.\ Lett.\ B {\bf 133}, 61 (1983); \\

8939: J.~R.~Ellis, A.~B.~Lahanas, D.~V.~Nanopoulos and K.~Tamvakis,

8940: ``No - Scale Supersymmetric Standard Model,''

8941: Phys.\ Lett.\ B {\bf 134}, 429 (1984);\\

8942: R.~Barbieri, E.~Cremmer and S.~Ferrara,

8943: ``Flat And Positive Potentials In N=1 Supergravity,''

8944: Phys.\ Lett.\ B {\bf 163} (1985) 143.

8945: 

8946: 

8947: \bibitem{BCOV}

8948: M.~Bershadsky, S.~Cecotti, H.~Ooguri and C.~Vafa,

8949: ``Kodaira-Spencer theory of gravity and exact results for quantum string

8950: amplitudes,''

8951: Commun.\ Math.\ Phys.\  {\bf 165} (1994) 311

8952: [arXiv:hep-th/9309140].

8953: 

8954: 

8955: \bibitem{Witten2}

8956: E.~Witten,

8957: ``Quantum background independence in string theory,''

8958: arXiv:hep-th/9306122.

8959: 

8960: 

8961: \bibitem{NOV}

8962: N.~Nekrasov, H.~Ooguri and C.~Vafa,

8963: ``S-duality and topological strings,''

8964: JHEP {\bf 0410} (2004) 009

8965: [arXiv:hep-th/0403167].

8966: 

8967: 

8968: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8969: %

8970: %   Chapter 4

8971: %

8972: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8973: 

8974: \bibitem{Binetruy}

8975: P.~Binetruy,

8976: ``Dilaton, Moduli And String / Five-Brane Duality As Seen From

8977: Four-Dimensions,''

8978: Phys.\ Lett.\ B {\bf 315} (1993) 80 [arXiv:hep-th/9305069].

8979: 

8980: 

8981: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8982: %

8983: %   Chapter 5

8984: %

8985: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

8986: 

8987: \bibitem{Romans}

8988: L.~J.~Romans,

8989: ``Massive N=2a Supergravity In Ten-Dimensions,''

8990: Phys.\ Lett.\ B {\bf 169} (1986) 374.

8991: 

8992: 

8993: \bibitem{FMM}

8994: S.~Fidanza, R.~Minasian and A.~Tomasiello,

8995: ``Mirror symmetric SU(3)-structure manifolds with NS fluxes,''

8996: arXiv:hep-th/0311122; \\

8997: M.~Gra\~na, R.~Minasian, M.~Petrini and A.~Tomasiello,

8998: ``Supersymmetric backgrounds from generalized Calabi-Yau manifolds,''

8999: JHEP {\bf 0408} (2004) 046

9000: [arXiv:hep-th/0406137].

9001: 

9002: 

9003: \bibitem{Gukov}

9004: S.~Gukov,

9005: ``Solitons, superpotentials and calibrations,''

9006: Nucl.\ Phys.\ B {\bf 574} (2000) 169

9007: [arXiv:hep-th/9911011].

9008: 

9009: 

9010: \bibitem{AS}

9011: B.~S.~Acharya and B.~Spence,

9012: ``Flux, supersymmetry and M theory on 7-manifolds,''

9013: arXiv:hep-th/0007213.

9014: 

9015: 

9016: 

9017: 

9018: 

9019: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

9020: % 

9021: %   Chapter 6

9022: %

9023: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

9024: 

9025: \bibitem{DGNV}

9026: R.~Dijkgraaf, S.~Gukov, A.~Neitzke and C.~Vafa,

9027: ``Topological M-theory as unification of form theories of gravity,''

9028: arXiv:hep-th/0411073.

9029: 

9030: 

9031: \bibitem{Nekrasov}

9032: N.~Nekrasov,

9033: ``A la recherche de la m-theorie perdue. Z theory: Chasing m/f theory,''

9034: arXiv:hep-th/0412021.

9035: 

9036: 

9037: \bibitem{CS}

9038: S.~Chiossi, S.~Salamon,

9039: ``The intrinsic torsion of SU(3) and $G_2$ structures,''

9040: Differential Geometry, Valencia 2001, World Sci. Publishing, 2002, pp 115-133

9041: [arXiv:math.DG/0202282].

9042: 

9043: 

9044: \bibitem{CCDLM}

9045:   G.~L.~Cardoso, G.~Curio, G.~Dall'Agata, D.~L\"ust, P.~Manousselis and

9046:    G.~Zoupanos,

9047:   ``Non-K\"ahler string backgrounds and their five torsion classes,''

9048:   Nucl.\ Phys.\ B {\bf 652} (2003) 5

9049:   [arXiv:hep-th/0211118]

9050: 

9051: 

9052: 

9053: \bibitem{DLM}

9054: M.~J.~Duff, J.~T.~Liu and R.~Minasian,

9055: ``Eleven-dimensional origin of string / string duality: A one-loop test,''

9056: Nucl.\ Phys.\ B {\bf 452} (1995) 261

9057: [arXiv:hep-th/9506126].

9058: %%CITATION = HEP-TH 9506126;%%

9059: 

9060: 

9061: \bibitem{SVW}

9062: S.~Sethi, C.~Vafa and E.~Witten,

9063: ``Constraints on low-dimensional string compactifications,''

9064: Nucl.\ Phys.\ B {\bf 480}, 213 (1996)

9065: [arXiv:hep-th/9606122].

9066: 

9067: 

9068: 

9069: 

9070: 

9071: 

9072: \bibitem{Joyce}

9073: D.~D.~Joyce,

9074: ``Compact Riemannian 7-manifolds with Holonomy $G_2$, I./II.''

9075: J. Diff. Geom. {\bf 43} (1996) 291-328

9076: 

9077: 

9078: 

9079: 

9080: \bibitem{BJ}

9081:   B.~S.~Acharya and B.~Spence,

9082:   ``Flux, supersymmetry and M theory on 7-manifolds,''

9083:   arXiv:hep-th/0007213;\\

9084:   K.~Behrndt and C.~Jeschek,

9085:   ``Fluxes in M-theory on 7-manifolds: G-structures and superpotential,''

9086:   Nucl.\ Phys.\ B {\bf 694} (2004) 99

9087:   [arXiv:hep-th/0311119];\\

9088:   T.~House and A.~Micu,

9089:   ``M-theory compactifications on manifolds with G(2) structure,''

9090:   arXiv:hep-th/0412006;\\

9091:   N.~Lambert,

9092:   ``Flux and Freund-Rubin superpotentials in M-theory,''

9093:   arXiv:hep-th/0502200.

9094: 

9095: \bibitem{Witt}

9096:   C.~Jeschek and F.~Witt,

9097:   ``Generalised G(2)-structures and type IIB superstrings,''

9098:   JHEP {\bf 0503} (2005) 053

9099:   [arXiv:hep-th/0412280];\\

9100:   F.~Witt,

9101:   ``Generalised $G_2$-manifolds,''

9102:   arXiv:math.dg/0411642.

9103:  

9104: \bibitem{Douglas}

9105:   M.~R.~Douglas,

9106:   ``The statistics of string / M theory vacua,''

9107:   JHEP {\bf 0305} (2003) 046

9108:   [arXiv:hep-th/0303194].

9109: 

9110: \bibitem{OVV}

9111:   H.~Ooguri, C.~Vafa and E.~Verlinde,

9112:   ``Hartle-Hawking wave-function for flux compactifications,''

9113:   arXiv:hep-th/0502211.

9114:  

9115: 

9116: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

9117: % section 6

9118: 

9119: 

9120: 

9121: 

9122: 

9123: 

9124: %%%%%%%%%%%%%%%%%%%%%%%%%%

9125: % Appendices

9126: 

9127: 

9128: \bibitem{CRTV}

9129: B.~Craps, F.~Roose, W.~Troost and A.~Van Proeyen,

9130: ``What is special K\"ahler geometry?,''

9131: Nucl.\ Phys.\ B {\bf 503} (1997) 565

9132: [arXiv:hep-th/9703082].

9133: 

9134: 

9135: \bibitem{Freed}

9136: D.~S.~Freed,

9137: ``Special K\"ahler manifolds,''

9138: Commun.\ Math.\ Phys.\  {\bf 203}, 31 (1999)

9139: [arXiv:hep-th/9712042].

9140: 

9141: 

9142: 

9143: 

9144: 

9145: %\bibitem{Acharya}

9146: %  B.~S.~Acharya,

9147: %  ``A moduli fixing mechanism in M theory,''

9148: %  arXiv:hep-th/0212294.

9149: %  %%CITATION = HEP-TH 0212294;%%

9150: 

9151: 

9152: 

9153: 

9154: \end{thebibliography}

9155: \end{document}

9156: 

9157: 

9158: 

9159: 

9160: 

9161: 

9162: