1: \documentclass[12pt]{article}

2: \usepackage{graphicx}

3: \usepackage{epsf}

4: \usepackage{cite}

5: 

6: \def\be{\begin{equation}}

7: \def\ee{\end{equation}}

8: \def\bea{\begin{eqnarray}}

9: \def\eea{\end{eqnarray}}

10: 

11: \addtolength\topmargin{-50pt}

12: \addtolength\textheight{105pt}

13: \addtolength\textwidth{60pt}

14: \addtolength\oddsidemargin{-38pt}

15: \setlength{\parindent}{20pt}

16: \setlength{\parskip}{6pt}

17: \frenchspacing

18: \sloppy

19: 

20: \begin{document}

21: 

22: \thispagestyle{empty}

23: \vspace*{.5cm}

24: \noindent

25: HD-THEP-08-6 \hspace*{\fill} 26 January, 2008\\

26: \noindent

27: OUTP-08-02P\hspace*{\fill}revised 25 June, 2009

28: 

29: \vspace*{1.0cm}

30: 

31: \begin{center}

32: {\Large\bf Inducing the $\mu$ and the $B\mu$ Term\\[.4cm] by the Radion 

33: and the 5d Chern-Simons Term}

34: \\[1.5cm]

35: {\large A.~Hebecker$\,^a$, J.~March-Russell$\,^b$ and R.~Ziegler$\,^c$}

36: \\[.5cm]

37: {\it ${}^a$ Institut f\"ur Theoretische Physik, Universit\"at Heidelberg,

38: Philosophenweg 16 und 19\\ D-69120 Heidelberg, Germany}

39: \\[.3cm]

40: {\it ${}^b$ Rudolf Peierls Centre for Theoretical Physics, University

41: of Oxford, 1 Keble Road\\ Oxford OX1 3NP, UK}

42: \\[.3cm]

43: {\it ${}^c$ International School for Advanced Studies (SISSA/ISAS), 

44: Via Beirut 2-4,\\ 34014 Trieste, Italy}

45: \\[.4cm]

46: {\small\tt (\,a.hebecker@thphys.uni-heidelberg.de\,,\,\,}

47: {\small\tt j.march-russell1@physics.ox.ac.uk}

48: \\

49: {\small and}

50: {\small\tt ziegler@sissa.it\,)}

51: \\[1.0cm]

52: 

53: {\bf Abstract}\end{center}

54: \noindent

55: In 5-dimensional models with gauge-Higgs unification, the $F$-term vacuum

56: expectation value of the radion provides, in close analogy to the 

57: Giudice-Masiero 

58: mechanism, a natural source for the $\mu$ and $B\mu$ term. Both the leading 

59: order gauge theory lagrangian and the supersymmetric Chern-Simons term 

60: contain couplings to the radion superfield which can be used for this purpose. 

61: We analyse the basic features of this mechanism for $\mu$ term generation 

62: and provide an explicit example, based on a variation of the SU(6) 

63: gauge-Higgs unification model of Burdman and Nomura. This construction 

64: contains all the relevant features used in our generic analysis. More 

65: generally, we expect our mechanism to be relevant to many of the recently 

66: discussed orbifold GUT models derived from heterotic string theory. This 

67: provides an interesting way of testing high-scale physics via Higgs mass

68: patterns accessible at the LHC. 

69: 

70: 

71: 

72: 

73: 

74: \newpage

75: \section{Introduction}

76: The generation of a $\mu$ and $B\mu$ term in the Higgs sector of the 

77: supersymmetric standard model is one of the critical issues in low-energy 

78: supersymmetry. While the $\mu$ term alone is responsible for Higgsino 

79: masses, both terms play a central role in realizing an appropriate scalar 

80: potential in the Higgs sector, ensuring the spontaneous breaking of the 

81: electroweak gauge symmetry. Since the $\mu$ term respects supersymmetry, 

82: one might also formulate the $\mu/B\mu$ term problem by asking why this 

83: term, which would naturally be either very large or exactly zero, happens 

84: to be of the same order of magnitude as the soft supersymmetry-breaking 

85: $B\mu$ term~\cite{Kim:1983dt}. 

86: 

87: The two most popular solutions to this problem are provided by the 

88: Giudice-Masiero mechanism~\cite{Giudice:1988yz} and the next-to-minimal 

89: supersymmetric standard model~\cite{Fayet:1974pd}. In the latter, 

90: the scale of the $\mu$ term is set by the vacuum 

91: expectation value of the scalar component of an extra uncharged chiral 

92: superfield. By contrast, in the former the $\mu$ term arises from a term in 

93: the K\"ahler potential, which mimics a $\mu$ term in the superpotential 

94: after the non-zero $F$ term of the spurion superfield has absorbed part of the 

95: superspace integrations. Many variants of these mechanisms as well as

96: other approaches to the problem have since been considered 

97: (see~\cite{Barbieri:2007tu} for some recent examples). 

98: 

99: In the present paper, we investigate 5-dimensional models with 

100: gauge-Higgs unification~\cite{Manton:1979kb}, where the $\mu/B\mu$ term 

101: problem is solved naturally in a way which is very similar to the 

102: Giudice-Masiero mechanism. Both these terms as well as the gaugino mass 

103: term and some of the soft scalar masses are generated at the high scale 

104: in the interplay of the $F$ term of the radion superfield and the chiral 

105: compensator of ${\cal N}=1$ supergravity with the quadratic gauge theory 

106: lagrangian~\cite{choi} (see also~\cite{str}). We point out that the resulting 

107: high-scale relations are changed significantly by the 5d Chern-Simons term 

108: which, in particular, induces a non-trivial Higgs scalar potential even in the 

109: absence of an $F$ term of the chiral compensator. 

110: 

111: At the more fundamental level, our motivation for this work is twofold:

112: On the one hand, orbifold-GUTs~\cite{orb} are arguably {\it the} modern 

113: framework for 

114: grand unification. Within this framework, gauge-Higgs unification receives 

115: a strong motivation from the requirement of a large top Yukawa coupling. 

116: Furthermore, it is natural that both the radion superfield~\cite{ky} and 

117: (after radion stabilization) also the chiral compensator develop an $F$-term 

118: vacuum expectation value. Thus, all ingredients for our mechanism are 

119: naturally present and the required terms in the supersymmetric Higgs sector 

120: arise without any further model building assumptions. 

121: 

122: On the other hand, heterotic orbifold model building has recently produced 

123: some of the most successful string-theoretic realizations of the 

124: supersymmetric standard model~\cite{Buchmuller:2005jr} (for earlier related 

125: work see~\cite{Kobayashi:2004ud}). From this perspective, 

126: the existence of an intermediate energy scale (one or two orders of 

127: magnitude below the string scale), at which the world appears to be 

128: 5-dimensional, is also well-motivated~\cite{Hebecker:2004ce}. 

129: It provides one of the few potential solutions to the 

130: string-scale/GUT-scale problem. Furthermore, gauge-Higgs unification is again 

131: a natural ingredient in all constructions where the Higgs fields come from 

132: the untwisted sector, which is indeed the case in many concrete examples. 

133: 

134: The presence of a $\mu$ term in 5d models with gauge-Higgs unification 

135: has been noticed early on~\cite{bn}.\footnote{

136: An

137: alternative proposal in closely related string-theoretic models appears in 

138: the last paper of Ref.~\cite{Buchmuller:2005jr}.

139: } 

140: The simultaneous generation of 

141: a $B\mu$ term by the $F$-term vev of the chiral compensator, leading to an 

142: interesting relations between $\mu$ term, $B\mu$ term and non-holomorphic 

143: soft Higgs masses, has been pointed out in~\cite{choi}. This relation is

144: maintained in the presence of a 5d Chern-Simons term, which however changes 

145: the relation with the gaugino masses. As we already mentioned, the 

146: Chern-Simons term is crucial in 

147: situations where the $F$ term of the chiral compensator is small. Although 

148: such a term is generically present in 5d supersymmetric gauge 

149: theories~\cite{ims} (see also~\cite{Kuzenko:2006ek}), it affects low-energy 

150: phenomenology only if some of the scalars of the 5d gauge multiplet develop 

151: large vacuum expectation values~\cite{Hebecker:2004xx}. This is, however, very 

152: well motivated in stringy realizations of our scenario, where more than 5 

153: dimensions are originally present. In most cases, some of these extra 

154: compact dimensions support non-zero Wilson lines which can, from a 

155: 5d perspective, play the role of the required scalar vacuum expectation 

156: value. In such situations, the supersymmetric Chern-Simons term is 

157: parametrically as important for low-energy phenomenology as the quadratic 

158: lagrangian. 

159: 

160: We finally note that a detailed phenomenological analysis of the proposal 

161: advocated in the present paper has subsequently appeared 

162: in~\cite{Brummer:2009ug}. In addition to demonstrating the phenomenological 

163: viability of our setting, this work was essential for bringing an earlier, 

164: partially incorrect version of this paper in its present form. We will comment 

165: on the earlier proposal, its problems and their possible resolutions in

166: more detail below.\footnote{

167: We

168: are indebted to Felix Br\"ummer pointing out the problems of the original 

169: setting.

170: }

171: 

172: Our paper is organized as follows: We begin in Sect.~\ref{tm} with the 

173: discussion of an abelian toy model which shows, in a very direct and 

174: transparent way, how the quadratic gauge theory lagrangian and the 

175: Chern-Simons term induce, in their interplay with the radion superfield, 

176: terms that are structurally similar to the $\mu$ and $B\mu$ term and 

177: soft supersymmetry breaking masses for the `Higgs field'. 

178: 

179: In Sect.~\ref{nag}, we extend our analysis to the non-abelian case, 

180: providing in particular a superfield expression for the non-abelian 

181: supersymmetric Chern-Simons term. The derivation of this term, which we

182: consider to be a very interesting by-product of our investigation, is 

183: described in more detail in the Appendix. Applying our formulae to 

184: a U(6)$\,=\,$SU(6)$\times$U(1) model, where the possibility of gauge-Higgs 

185: unification is particularly apparent from the decomposition 

186: ${\bf 35}={\bf 24}+{\bf 5}+{\bf\bar{5}}+{\bf 1}$ of the adjoint~\cite{bn}, 

187: we identify the terms involving the two Higgs superfields, the radion and 

188: the chiral compensator.

189: 

190: We use our previous results to calculate, in Sect.~\ref{cmb}, $\mu$ and $B\mu$ 

191: term, as well as soft Higgs scalar masses and gaugino masses. As an 

192: interesting observation we note that, in the absence of the Chern-Simons

193: term and of an $F$ term of the chiral compensator, $\mu$ term and soft

194: scalar masses conspire to ensure an exactly flat scalar potential in the 

195: Higgs sector. However, once the radion is stabilized, a chiral compensator 

196: $F$ term generically develops and this flatness is lifted. 

197: 

198: In Sect.~\ref{csp}, we give the complete expressions for the $\mu$ term 

199: and the soft parameters of the gauge-Higgs sector, including the effects 

200: of the Chern-Simons term and chiral compensator. We then briefly discuss 

201: the viability of this high-scale input for low-energy phenomenology after 

202: the renormalization group running down to the electroweak scale. We also 

203: comment on the influence of the squark masses and trilinear terms on this 

204: running and on the partially model-dependent high-scale origin of these 

205: terms (especially in the top quark sector) in our 5d gauge-Higgs unification 

206: scenario. 

207: 

208: Finally, we provide in Sect.~\ref{bns} an explicit phenomenologically viable 

209: construction that has all the qualitative features which we used in our 

210: previous discussion. Our model is closely related to a 5d model for gauge-Higgs

211: unification by Burdman and Nomura~\cite{bn}. We obtain our model by 

212: lifting this previous construction to 6 dimensions, where the compact space 

213: has the topology of a pillow case, and taking a different 5d limit of this 

214: geometry. In this way the non-zero 5d vev of the scalar component of the gauge 

215: multiplet is automatically enforced. The rather intricate realization of 

216: matter fields and Yukawa couplings can essentially be copied from the 

217: construction of Burdman and Nomura. 

218: 

219: Our summary and conclusions are given in Sect.~\ref{conc}. 

220: 

221: 

222: 

223: 

224: 

225: \section{The basic mechanism in an abelian toy model}\label{tm}

226: The supersymmetric 5d U(1) gauge theory has a well-known description in 

227: terms of a 4d real superfield $V$ and a chiral superfield $\Phi=\Sigma+ 

228: iA_5+\cdots$, both depending on the extra parameter $x^5$. Using this 

229: language, the quadratic 5d lagrangian reads~\cite{Marcus:1983wb,

230: Arkani-Hamed:2001tb}

231: \be

232: {\cal L}_2=\frac{1}{4g_5^2}\left[\int d^2\theta\,\,W^2+{\rm h.c.}+

233: \int d^4\theta\,\left(2\partial_5V-(\Phi+\bar{\Phi})\right)^2\right]\,,

234: \label{l2}

235: \ee

236: where $W$ is the supersymmetric field strength defined in terms of $V$. 

237: The supersymmetric Chern-Simons term which will in general be present

238: in this theory takes the form~\cite{Arkani-Hamed:2001tb}\footnote{

239: Note

240: that we find a different sign for the second term than is reported 

241: in~\cite{Arkani-Hamed:2001tb}.

242: }

243: \bea

244: {\cal L}_{cs} & = & c\,\left[\,\int d^2\theta\,\,\Phi\,W^2 + {\rm h.c.} 

245: \right. 

246: \nonumber \\ 

247: & & \hspace*{.5cm} + \left. \frac{2}{3} \int d^4\theta \; \left( \partial_5 

248: V D_\alpha V - V D_\alpha \partial_5 V  \right) W^\alpha + {\rm h.c.} \right. 

249: \nonumber \\

250: & & \hspace*{.5cm} - \left. \frac{1}{6} \int d^4\theta \; \left( 2 \partial_5 

251: V - (\Phi + \bar{\Phi}) \right)^3 \right]\,.\label{lcs}

252: \eea

253: 

254: We are interested in the 4d effective field theory obtained after $S^1$ 

255: compactification of the above model, in particular in the couplings to the 

256: radion superfield. The following discussion can be viewed as a mild 

257: generalization of~\cite{Marti:2001iw} (because of the Chern-Simons term) or 

258: as a significantly simplified version of the derivation of related formulae 

259: in~\cite{PaccettiCorreia:2004ri}. 

260: 

261: The relevant 4d lagrangian is found by 

262: simply dropping all terms involving $x^5$ derivatives, replacing $V$ and 

263: $\Phi$ by their ($x^5$-independent) zero modes, and integrating the result 

264: over $x^5$. In the rigid case, the latter amounts to a multiplication by 

265: $L=2\pi R$. By contrast, in the case where the original model is coupled to 

266: 5d supergravity, this multiplicative factor has to be replaced by the radion 

267: superfield $T$ (or $\bar{T}$) in the holomorphic (antiholomorphic) terms of 

268: Eqs.~(\ref{l2}) and (\ref{lcs}) and by $(T+\bar{T})/2$ in the $d^4\theta$ 

269: terms. Here the 4d chiral superfield $T$ is normalized such that

270: \be

271: T=L+iB_5\,\,,

272: \ee

273: where $B_M$ ($M=0\ldots 3,5$) is the graviphoton of the 5d supergravity 

274: multiplet. Its pure-derivative coupling in the component action enforces 

275: the use of the combination $T+\bar{T}$ in the $d^4\theta$ terms in 

276: Eqs.~(\ref{l2}) and (\ref{lcs}).

277: 

278: However, this is not the only way in which $T$ enters the 4d effective 

279: theory. From the fact that $\Phi$ contains the gauge field component $A_5$,

280: and $A_5$ covariantizes the derivative operator $\partial/\partial x^5$, it

281: follows that the whole superfield has to scale as the inverse size of the 

282: compact dimension. Thus, we have to perform the replacements 

283: \be

284: \Phi\,\to\,\frac{L_0}{T}\Phi\qquad\mbox{and}\qquad \Phi\,\to\,

285: \frac{2L_0}{T+\bar{T}}\Phi\label{resc}

286: \ee

287: in the $d^2\theta$ and $d^4\theta$ terms above. Here we have introduced 

288: an arbitrary constant $L_0$ with the dimension of length to insure that

289: the new superfield $\Phi$ has the dimension of mass.

290: 

291: To summarize, the 4d low-energy lagrangian follows from Eqs.~(\ref{l2}) and 

292: (\ref{lcs}) after suppressing any $x^5$ dependence, multiplying the 

293: appropriate terms by $T$, $\bar{T}$ or $(T+\bar{T})/2$, and performing the

294: redefinition of Eq.~(\ref{resc}). The results are

295: \be

296: {\cal L}_{2,\,4d}=\frac{1}{4g_5^2}\left[\int d^2\theta\,\,T\,W^2+{\rm h.c.}+

297: 2L_0^2\int d^4\theta\,\frac{(\Phi+\bar{\Phi})^2}{T+\bar{T}}\right]\label{l24}

298: \ee

299: and

300: \be

301: {\cal L}_{cs,\,4d} = c\,\left[\,L_0\,\int d^2\theta\,\,\Phi\,W^2 + {\rm h.c.} 

302: + \frac{4L_0^3}{6} \int d^4\theta\,\frac{(\Phi + \bar{\Phi})^3}

303: {(T+\bar{T})^2} \right]\,.\label{lcs4}

304: \ee

305: To check that the $T$ dependence obtained in this intuitive approach is 

306: indeed correct, one can work out the component form of the above superfield 

307: expressions and match it (with appropriate field redefinitions and 

308: keeping track of all factors $g_{55}$) to the 5d component 

309: action~\cite{ziegler}. 

310: 

311: Our main point concerning the generation of certain MSSM operators can 

312: now easily be made. Recall that we want to think of $V$ as containing the 

313: Standard model gauge multiplet and of $\Phi$ as the Higgs 

314: superfield.\footnote{

315: Of 

316: course, in this simple U(1) toy model $\Phi$ is not charged and the second

317: Higgs multiplet is missing, but that is irrelevant for now.

318: } 

319: If the radion auxiliary field $F_T$ develops a non-zero expectation value, 

320: it is immediately clear that the superspace integrals in Eq.~(\ref{l24}) 

321: induce operators

322: \be

323: \sim F_TW^2\Big|_1\quad,\quad 

324: \sim|F_T|^2\Phi^2\Big|_1 \quad\,\quad 

325: \sim\bar{F}_T\Phi^2\Big|_{\theta^2}\quad,\quad

326: \sim|F_T|^2\Phi\bar{\Phi}\Big|_1\quad\mbox{and}\quad 

327: \sim\bar{F}_T\Phi\bar{\Phi}\Big|_{\theta^2}\,.\label{ops}

328: \ee

329: The first of them provides gaugino masses, which is often referred to as 

330: radion mediation~\cite{ky}. The second, which clearly has the structure of 

331: the MSSM $\mu$ term, provides Higgsino masses.\footnote{

332: This

333: can be understood from a slightly different perspective as follows: 

334: Non-vanishing $F_T$ is the 4d manifestation of an $SU(2)_R$ symmetry twist

335: of in the 5d background. The latter induces gaugino masses and, since the

336: Higgsinos are 5d gauginos in the present setting, non-vanishing Higgsino 

337: masses are also induced~\cite{bn}.

338: }

339: Furthermore, both the second and the remaining operators in Eq.~(\ref{ops}) 

340: contribute to the scalar potential, thereby apparently inducing 

341: a $B\mu$ term and soft scalar masses in the Higgs sector. However, a more 

342: careful analysis of Eq.~(\ref{l24}) reveals that all these contributions 

343: exactly cancel and the scalar potential remains flat. (This fact, which can 

344: also be understood from a structural perspective~\cite{Barbieri:1985wq}, 

345: remains true in the non-abelian case.) 

346: 

347: To lift the flatness of the potential and to induce a non-zero $B\mu$ 

348: term and soft scalar masses in the present 

349: framework, the effect of the chiral compensator of ${\cal N}=1$ supergravity,

350: $\varphi=1+F_\varphi\theta^2$, has to be taken into account. More 

351: specifically, a factor $\varphi\bar{\varphi}$ has to be included the last 

352: term in Eq.~(\ref{l24}). If $F_\varphi$ develops a non-zero vacuum 

353: expectation value, operators analogous to those displayed in Eq.~(\ref{ops}) 

354: (but with one or both of the factors $F_T$ and $\bar{F}_T$ replaced by 

355: $F_\varphi$ and $\bar{F}_\varphi$) are induced. The total scalar potential 

356: looses its flatness, which can be described by a non-vanishing $B\mu$ and 

357: soft scalar mass terms.

358: 

359: If the lowest component of $\Phi$ develops a vacuum expectation 

360: value, then the Chern-Simons lagrangian of Eq.~(\ref{lcs4}) corrects the 

361: quadratic order lagrangian of Eq.~(\ref{l24}). Moreover, if $c={\cal O}(1)$ 

362: and $\langle \Phi\rangle\sim 1/g_5^2$ (both of which are natural values,

363: as will become clear in the following), these contributions are not 

364: parametrically suppressed relative to those of Eq.~(\ref{l24}). Thus, gaugino 

365: masses, $\mu$ and $B\mu$ term, and the Higgs sector soft scalar masses 

366: are induced on the basis of the fundamental lagrangian of Eqs.~(\ref{l2}) and 

367: (\ref{lcs}) after coupling it to supergravity and allowing for vacuum 

368: expectation values of $\Phi$, $F_T$ and $F_\varphi$. As we will explain in 

369: more detail below, in higher-dimensional unified models an interesting and 

370: realistic phenomenology can emerge on the basis of this very generic 

371: mechanism.

372: 

373: 

374: 

375: 

376: 

377: \section{Non-abelian generalization}\label{nag}

378: The ${\cal N}=1$ superfield action of the 5d non-abelian gauge theory~\cite{ 

379: Marcus:1983wb,Arkani-Hamed:2001tb} can be given in a manifestly 

380: super-gauge-invariant form using the super-gauge-covariant $x^5$ 

381: derivative~\cite{Hebecker:2001ke}

382: \be

383: \nabla_5=\partial_5+\Phi\,.

384: \ee

385: It reads

386: \be

387: {\cal L}_2=\frac{1}{2g_5^2}\mbox{tr}\left[\int d^2\theta\,\,W^2+{\rm h.c.}+

388: \int d^4\theta\,\left(e^{-2V}\nabla_5e^{2V}\right)^2\right]\,,

389: \label{l2na}

390: \ee

391: where the action of $\Phi$ on $e^{2V}$ follows from the standard gauge 

392: transformation properties of $e^{2V}$, i.e.,

393: \be

394: \nabla_5e^{2V}=\partial_5e^{2V}-\Phi^\dagger e^{2V}-e^{2V}\Phi\,.

395: \ee

396: For the non-abelian supersymmetric Chern-Simons term we have, unfortunately, 

397: not been able to derive an equally elegant superfield formula. However,

398: sacrificing manifest super gauge invariance by restricting ourselves to

399: Wess-Zumino gauge, the following expression can be derived~\cite{ziegler} 

400: (see Appendix):

401: \bea

402: {\cal L}_{cs} & = & c\,\mbox{tr}\left[\,\int d^2\theta\,\,\Phi\,W^2 + 

403: {\rm h.c.} \right. 

404: \nonumber \\ 

405: & & \hspace*{1cm} + \left. \frac{1}{3} \int d^4\theta \; \left( \{\partial_5 

406: V,D_\alpha V\} - \{V,D_\alpha\partial_5 V\}\right) W^\alpha + {\rm h.c.} 

407: \right. 

408: \nonumber \\ 

409: & & \hspace*{1cm} - \left. \frac{1}{12} \int d^4\theta \; \left( \{\partial_5 

410: V,D_\alpha V\} - \{V,D_\alpha\partial_5 V\}\right) W_{(2)}^\alpha) + 

411: {\rm h.c.} 

412: \right. 

413: \nonumber \\

414: & & \hspace*{1cm} - \left. \frac{1}{6} \int d^4\theta \; 

415: \left(e^{-2V}\nabla_5e^{2V}\right)^3 \right]\,.\label{lcsna}

416: \eea

417: Here curly brackets are used for anticommutators and $W_{(2)}^\alpha$ 

418: represents the part of $W^\alpha$ which is quadratic in $V$ (recall

419: that, in Wess-Zumino gauge, $W$ is the sum of a linear and quadratic piece 

420: in $V$).

421: 

422: Starting from Eqs.~(\ref{l2na}) and (\ref{lcsna}), which are the 

423: non-abelian generalizations of Eqs.~(\ref{l2}) and (\ref{lcs}), the coupling 

424: of the radion superfield to the zero modes of the compactified theory can 

425: be derived in complete analogy to Sect.~\ref{tm}. To recapitulate, one 

426: simply has to suppress any $x^5$ dependence, multiply the appropriate terms 

427: by $T$, $\bar{T}$ or $(T+\bar{T})/2$, and perform a redefinition analogous

428: to that of Eq.~(\ref{resc}). The results are

429: \be

430: {\cal L}_{2,\,4d}=\frac{1}{2g_5^2}\mbox{tr}\left[\int d^2\theta\,\,T\,W^2+

431: {\rm h.c.}+2L_0^2\int d^4\theta\,\frac{(\Phi+\bar{\Phi})^2}{T+\bar{T}}\right]

432: \label{l2na4}

433: \ee

434: and

435: \be

436: {\cal L}_{cs,\,4d} = c\,\mbox{tr}\left[\,L_0\,\int d^2\theta\,\,\Phi\,W^2 + 

437: {\rm h.c.} + \frac{4L_0^3}{6} \int d^4\theta\,\frac{(\Phi + \bar{\Phi})^3}

438: {(T+\bar{T})^2} \right]\,.\label{lcsna4}

439: \ee

440: Clearly, this could have also been obtained by starting from Eqs.~(\ref{l24})

441: and (\ref{lcs4}), promoting the superfields $V$ and $\Phi$ to appropriate

442: matrices and introducing the corresponding trace operations. In this 

443: sense, our above discussion of the 5d superfield expression for the 

444: non-abelian Chern-Simons term is included merely for completeness (and 

445: possible other applications). The phenomenology-oriented analysis 

446: following from now on is based entirely on Eqs.~(\ref{l2na4}) and 

447: (\ref{lcsna4}), which are straightforward generalizations of Eqs.~(\ref{l24})

448: and~(\ref{lcs4}). 

449: 

450: We can now be more specific about how we envisage the $\mu$ and $B\mu$ term

451: generation to proceed in models of this type. To be concrete, let $V$ and 

452: $\Phi$ take values in the Lie algebra of the GUT gauge group 

453: U(6)$\,=\,$SU(6)$\times$U(1). Furthermore, let the theory be compactified 

454: to 4d on an interval such that SU(6) is broken to SU(5)$\times$U(1)$'$ and 

455: the U(1) is completely broken. In the corresponding decomposition 

456: of the adjoint representation,

457: \be

458: {\bf 35}={\bf 24}+{\bf 5}+{\bf \bar{5}}+{\bf 1}\,,

459: \ee

460: we find, as parts of the superfield $\Phi$, the Higgs multiplets $H_u$ and 

461: $H_d$ in the ${\bf 5}$ and ${\bf \bar{5}}$ of SU(5). The further breaking 

462: of SU(5) to the standard model gauge group, which could for example also 

463: be realized by boundary conditions, is not important at the moment. 

464: 

465: Thus, the second term of Eq.~(\ref{l2na4}) gives rise to the following 

466: contribution to the 4d Higgs lagrangian:

467: \be

468: {\cal L}_{2,\,4d}\supset \frac{1}{g_4^2}\int d^4\theta\frac{2L_0\varphi

469: \bar{\varphi}}{T+\bar{T}}(H_u+\bar{H}_d)(H_d+\bar{H}_u)\,.\label{lh2}

470: \ee

471: Furthermore, if $\Phi$ develops a vev $\langle\Phi\rangle=v\,\mbox{\bf 1}$, 

472: consistent with the assumed boundary-breaking of the U(1)\footnote{

473: In 

474: an earlier version of this paper, a $\Phi$-vev $\sim\mbox{diag}(1,1,1,1,1,-5)$

475: inside the adjoint of SU(6) was assumed. This is inconsistent with an orbifold 

476: breaking of SU(6) to SU(5)$\times$U(1)$'$. The desired breaking by boundary 

477: conditions can nevertheless be realized, e.g. by introducing a brane localized 

478: adjoint superfield and giving it a large vev $\sim\mbox{diag}(1,1,1,1,1,-5)$. 

479: However, the bulk vev of $\Phi$ induces a bulk mass for the ${\bf 5}$ and 

480: ${\bf \bar{5}}$ Higgs fields. This is easy to see since the gauge symmetry 

481: is broken in the 5d bulk. Hence the `broken' $A_5$ components, which form 

482: some of the Higgs scalars, become massive in 5d. Equivalently, when thinking 

483: at the zero-mode level of a corresponding $S^1$ compactification, this mass

484: term must be present since the ${\bf 5}$ and ${\bf \bar{5}}$ chiral 

485: multiplets become part of the massive vector multiplet. On an interval with

486: boundary-breaking, massless 4d fields in these representations nevertheless 

487: survive since only a certain 

488: linear combination of the bulk and brane ${\bf 5}$ and ${\bf \bar{5}}$ fields 

489: is `eaten' by the vector multiplets which become massive in the breaking of

490: SU(6) to SU(5)$\times$U(1)$'$. However, these massless Higgs fields now have 

491: a non-trivial bulk profile because of their bulk mass. This profile depends 

492: on the size of the $\Phi$-vev and affects both the calculation of soft terms 

493: and of Yukawa couplings, thereby significantly complicating the subsequent 

494: analysis. This set of problems as well as its resolution by simply using U(6) 

495: instead of SU(6) was pointed out to us by Felix Br\"ummer (see 

496: also~\cite{Brummer:2009ug}). 

497: 

498: We also note that U(6) is, of course, a product gauge group allowing for 

499: independent coefficients of the SU(6)- and U(1)-kinetic terms as well as of

500: the CS terms of SU(6), U(1) and of the mixed CS terms (see e.g.~\cite{

501: Kim:2008kn}). Since, given the above $\Phi$-vev, only the mixed CS term is 

502: relevant for our analysis, we do not complicate our notation by making all 

503: those independent coeffcients explicit.

504: }, 

505: the second term of Eq.~(\ref{lcsna4}) gives rise to the following correction 

506: to this lagrangian 

507: (up to quadratic order):

508: \be

509: {\cal L}_{cs\,,4d}\supset 2cL_0v\int d^4\theta\frac{(2L_0)^2\varphi

510: \bar{\varphi}}{(T+\bar{T})^2}(H_u+\bar{H}_d)(H_d+\bar{H}_u)\,.\label{lhcs}

511: \ee

512: Here we have assumed that, with the exception of $H_u$ and $H_d$, all 

513: the zero-mode components of the chiral adjoint $\Phi$ have been eliminated

514: by orbifolding (or acquired a large mass in another way). Note also that,

515: since we are not interested in the dynamics of $T$ and $\varphi$ at the 

516: moment, we have suppressed the constant term $\sim v^3$ in Eq.~(\ref{lhcs}). 

517: A term $\sim v^2$, which would have to be linear in $H_u$ and $H_d$, does 

518: obviously not arise for group theoretic reasons. 

519: 

520: In a vacuum where $T$ and $\varphi$ develop non-zero $F$ terms, 

521: Eqs.~(\ref{lh2}) and (\ref{lhcs}) provide, in addition to the kinetic terms 

522: for the Higgs multiplets, $\mu$ term, $B\mu$ term and soft scalar masses in

523: the Higgs sector. The relevant operators are analogous to those given 

524: explicitly in the case of our abelian toy model in Eq.~(\ref{ops}) of the 

525: previous section. In addition, the first terms of both Eq.~(\ref{l2na4}) 

526: and (\ref{lcsna4}) contribute to the standard model gauge kinetic term and 

527: to the corresponding gaugino masses. We devote the following two sections 

528: to the discussion of the resulting SUSY breaking pattern.

529: 

530: 

531: 

532: 

533: 

534: \section{Calculating the $\mu$ and $B\mu$ term and the Higgs-sector soft 

535: scalar masses}\label{cmb}

536: To begin, we ignore the possible Chern-Simons term and focus on the 

537: phenomenological implications of Eq.~(\ref{lh2}) and the first term 

538: of Eq.~(\ref{l2na4}). We assume the existence of a (meta-)stable 

539: almost-Minkowski vacuum in which Re$\,T=L_0$ and both $F_T$ and $F_\varphi$ 

540: have non-zero values. Using the chiral compensator approach to supergravity, 

541: the scalar potential in the Higgs sector (with canonical 4d field 

542: normalization) is easily obtained: We simply have to integrate out the 

543: auxiliary-field vectors $F_{H_u}$ and $F_{H_d}$ on the basis of 

544: Eq.~(\ref{lh2}) while treating $T$, $F_T$ and $F_\varphi$ as fixed external 

545: sources. The result reads

546: \be

547: {\cal L}_{4,\,can.}\supset -\left(|F_\varphi|^2-\frac{F_\varphi\bar{F}_T+

548: \bar{F}_\varphi F_T}{T+\bar{T}}\right)(H_u+\bar{H}_d)(H_d+\bar{H}_u)\,.

549: \label{pot}

550: \ee

551: We emphasize that, in contrast to the last section, in this and the 

552: following equations $H_u$ and $H_d$ are the scalar components of the 

553: corresponding superfields and their normalization has been modified to 

554: make the 4d kinetic term canonical. The corresponding Higgsino mass term

555: can be directly read off from Eq.~(\ref{lh2}):

556: \be

557: {\cal L}_{4,\,can.}\supset -\left(\bar{F}_\varphi-\frac{\bar{F}_T}{T+\bar{T}}

558: \right)\lambda_u\lambda_d+\mbox{h.c.}\,\label{higgsino}

559: \ee

560: where $\lambda_u$ and $\lambda_d$ are two-component Weyl spinors. This 

561: determines the value of the $\mu$ parameter, which is conventionally 

562: defined as the coefficient of the Higgsino bilinear:

563: \be

564: \mu=\bar{F}_\varphi-\frac{\bar{F}_T}{T+\bar{T}}\,.

565: \ee

566: Similarly to the gaugino mass 

567: \be

568: m_{1/2}=\frac{\bar{F}_T}{T+\bar{T}}\,,

569: \ee

570: a non-zero $\mu$ parameter arises as a consequence of $F_T$, even if 

571: $F_\varphi$ vanishes. 

572: 

573: Furthermore, if the Higgs scalar potential is parameterized by (see 

574: e.g.~\cite{Martin:1997ns})

575: \be

576: {\cal L}_{4,\,can.}\supset -(|\mu|^2+m_{H_u}^2)|H_u|^2-(|\mu|^2+

577: m_{H_d}^2)|H_d|^2-(B\mu)H_uH_d+\mbox{h.c.}+\mbox{quart.~terms}\,,

578: \ee

579: we read off from Eq.~(\ref{pot}) that $B\mu$, $m_{H_u}$ and $m_{H_d}$ are 

580: given by (see also~\cite{choi})

581: \be

582: B\mu=|\mu|^2+m_{H_u}^2=|\mu|^2+m_{H_d}^2=|F_\varphi|^2-

583: \frac{F_\varphi\bar{F}_T+\bar{F}_\varphi F_T}{T+\bar{T}}\,.\label{spar}

584: \ee

585: In contrast to the $\mu$ parameter, these scalar mass parameters vanish 

586: if $F_\varphi=0$. This is a result of the very specific generalized no-scale 

587: structure of the superfield expression in Eq.~(\ref{l2na4}). In terms of the 

588: conventional parameterization of the component lagrangian with soft terms,

589: it implies a somewhat surprising exact cancellation between $|\mu|^2$ and

590: $m_{H_u}^2$ as well as between $|\mu|^2$ and $m_{H_d}^2$ in 

591: Eq.~(\ref{spar}). Clearly, the phenomenological implications of the above

592: formulae crucially depend on the values of $F_T$ and $F_\varphi$ (especially 

593: on their relative size), on which we now briefly comment.

594: 

595: At the tree level, the compactification of 5d supergravity on $S^1/Z_2$ or

596: $S^1/(Z_2\times Z_2')$ gives rise to a K\"ahler potential of no-scale

597: type for the radion,

598: \be

599: K_0(T,\bar{T})=-3\ln(T+\bar{T})\,.

600: \ee

601: An effective constant superpotential can be introduced if the boundary 

602: conditions at the two ends of the interval preserve different ${\cal N}=1$

603: subalgebras of the original ${\cal N}=2$ SUSY. (Alternatively, the same 

604: effect can arise as a result of some non-perturbative boundary effect, such

605: as brane gaugino condensation.) In the resulting no-scale model, 

606: supersymmetry is broken by $F_T$, but $T$ remains a flat direction. At the 

607: same time, $F_\varphi$ remains exactly zero. For our purposes, this 

608: approximation (in the case that this is a reasonable approximation to the 

609: physical vacuum) is insufficient since, as already mentioned in 

610: Sect.~\ref{tm}, the Higgs sector scalar potential remains exactly flat in 

611: this case. 

612: 

613: Thus, we have to take the stabilization of the radion $T$ seriously from 

614: the very beginning and to determine $F_T$ and $F_\varphi$ in the context 

615: of a stabilized vacuum. It is well-known that $F_\varphi$ is generically 

616: non-zero in such situations (implying, in our context, that a Higgs sector 

617: scalar potential will be generated). 

618: 

619: Starting from the no-scale situation described above, stabilization of $T$ 

620: can arise as a result of either K\"ahler corrections or $T$-dependent 

621: superpotential terms. To be as generic as possible, we assume a model where,

622: on the basis of a corrected K\"ahler potential and superpotential,

623: \be

624: K(T,\bar{T})=K_0(T,\bar{T})+\Delta K(T,\bar{T})\qquad\mbox{and}\qquad W(T)\,,

625: \ee

626: a (meta-)stable almost-Minkowski vacuum is produced (see e.g.~\cite{ 

627: Luty:1999cz,Luty:2002hj}). The equations of motion for $F_T$ and $F_\varphi$ 

628: (and thus their vacuum values) can be obtained on the basis of the flat-space 

629: superfield lagrangian

630: \be

631: \int d^4\theta \varphi\bar{\varphi}\Omega(T,\bar{T})+\int d^2\theta\varphi^3

632: W(T)+\mbox{h.c.}\,,\label{tflag}

633: \ee

634: where $\Omega=-3\exp(-K/3)$ is the so-called `superspace kinetic 

635: energy'~\cite{wb}. 

636: 

637: For the purpose of this paper, we do not want to specify a stabilization 

638: mechanism for $T$ and extremize Eq.~(\ref{tflag}) explicitly. Instead, we

639: restrict ourselves to deriving a simple relation between the $F$ terms 

640: of the radion and the chiral compensator. This can be achieved rather

641: easily: First, assume that Eq.~(\ref{tflag}) possesses a SUSY-breaking 

642: minimum with vanishing cosmological constant. In this minimum, $W$ takes 

643: some vacuum expectation value $W_0$. We now go to a different K\"ahler-Weyl 

644: frame, defined by the requirement that the superpotential $W'$ in this 

645: frame is constant, $W'=W_0$. Such a change of frames can be viewed as a 

646: redefinition of the chiral compensator. The new chiral compensator $\varphi'$ 

647: is defined in terms of $T$ and $\varphi$ by

648: \be

649: W(T)\varphi^3=W'\varphi'^3\,.

650: \ee

651: In this new frame, $F_{\varphi'}=0$, which is an immediate consequence of 

652: vanishing vacuum energy and constant superpotential (see 

653: e.g.~\cite{Hebecker:2004sb}). Thus,

654: \be

655: \varphi=\varphi'\cdot\left(\frac{W(T)}{W_0}\right)^{-1/3}=1\cdot\left(1+

656: \frac{W_TF_T\theta^2}{W_0}\right)^{-1/3}\,.

657: \ee

658: To lighten notation, we can now suppress the index `0' of $W$ and simply 

659: conclude that

660: \be

661: F_\varphi=-\frac{W_T}{3W}F_T

662: \ee

663: in the physical vacuum. This formula allows for a simple evaluation of 

664: the previously derived supersymmetric and SUSY-breaking Higgs mass terms 

665: and their relation to gaugino masses in any concrete model of radius 

666: stabilization. Note that, for a generic function $W(T)$, we expect 

667: $F_\varphi\sim F_T/T$ on dimensional grounds. This relation is also found 

668: in the specific model of~\cite{Luty:1999cz}. The SUSY-breaking effects of 

669: $F_\varphi$ and $F_T$ are then parametrically equally important. 

670: 

671: 

672: 

673: 

674: 

675: \section{Including the effect of the Chern-Simons term and some 

676: phenomenological consequences}\label{csp}

677: We now repeat the analysis of the previous section on the basis of the 

678: complete lagrangian of Eqs.~(\ref{lh2}) and (\ref{lhcs}). Integrating out 

679: $F_{H_u}$ and $F_{H_d}$, the following (canonically normalized) scalar 

680: potential arises:

681: \be

682: {\cal L}_{4,\,can.}\supset-\left[|F_\varphi|^2-\frac{(F_\varphi\bar{F}_T+

683: \mbox{h.c.})}{T+\bar{T}}\,\frac{1+2c'}{1+c'}+\frac{|F_T|^2}{(T+\bar{T})^2}

684: \,\frac{2c'^2}{(1+c')^2}\right](H_u+\bar{H}_d)(H_d+\bar{H}_u)

685: \,,\label{potcs}

686: \ee

687: where

688: \be

689: c'=2cvg_5^2\,.\label{cp}

690: \ee

691: Note that the no-scale argument ensuring the complete flatness of the 

692: scalar potential in the absence of $F_\varphi$ has broken down. The reason 

693: is as follows: While the Chern-Simons term by itself respects the generalized 

694: no-scale structure, the presence of a fixed vev $v$ breaks this structure. 

695: For this it is crucial that the vev is truly fixed in the sense that no 

696: corresponding fluctuations are allowed - a situation which indeed arises 

697: in certain orbifold models (see below). 

698: 

699: Similarly, the Higgsino mass term, Eq.~(\ref{higgsino}), is now replaced by 

700: an analogous expression following from Eqs.~(\ref{lh2}) and (\ref{lhcs}):

701: \be

702: {\cal L}_{4,\,can.}\supset \left(\bar{F}_\varphi-\frac{\bar{F}_T}{T+\bar{T}}

703: \,\frac{1+2c'}{1+c'}\right)\lambda_u\lambda_d+\mbox{h.c.}

704: \ee

705: The gaugino mass is also affected by the Chern-Simons term. Although 

706: $F_\Phi$ does not develop a vacuum expectation value, the first term in 

707: Eq.~(\ref{lcsna4}) affects the normalization of the gauge kinetic term 

708: and hence the gaugino mass. Thus, we can summarize all effects by giving 

709: the following set of SUSY-breaking parameters and the $\mu$ term:

710: \bea

711: m_{1/2}&=&\frac{\bar{F}_T}{T+\bar{T}}\,\frac{1}{1+c'}\,,\label{m12}\\ 

712: \nonumber\\

713: B\mu&=&|\mu|^2+m_{H_u}^2=|\mu|^2+m_{H_d}^2\\

714: &=&|F_\varphi|^2-\frac{(F_\varphi\bar{F}_T+\mbox{h.c.})}{T+\bar{T}}\,

715: \frac{1+2c'}{1+c'}+\frac{|F_T|^2}{(T+\bar{T})^2}\,\frac{2c'^2}{(1+c')^2}\,,

716: \nonumber\\ \nonumber\\

717: \mu&=&\bar{F}_\varphi-\frac{\bar{F}_T}{T+\bar{T}}\,\frac{1+2c'}{1+c'}\,.

718: \label{mut}

719: \eea

720: The most striking feature of this result is, as without the Chern-Simons term, 

721: the equality between the $B\mu$ term and the parameters $|\mu|^2+m_{H_u}^2$ 

722: and $|\mu|^2+m_{H_d}^2$~\cite{choi}. We now briefly discuss the 

723: phenomenological consequences of this relation:

724: 

725: It is a well-known fact (see e.g.~\cite{Martin:1997ns}) that electroweak 

726: symmetry breaking, i.e. the destabilization of the vacuum with vanishing 

727: Higgs expectation values, requires

728: \be

729: (B\mu)^2>(|\mu|^2+m_{H_u}^2)(|\mu|^2+m_{H_d}^2)\,.\label{lar}

730: \ee

731: At the same time, positivity of the quadratic part of the scalar potential 

732: along the $D$-flat directions is guaranteed if 

733: \be

734: 2(B\mu)<(|\mu|^2+m_{H_u}^2)+(|\mu|^2+m_{H_d}^2)\,.\label{sma}

735: \ee

736: For the parameters that we have found, both inequalities turn into 

737: equalities, apparently disfavouring our scenario phenomenologically. 

738: However, our previous analysis was performed at a high scale (the GUT scale 

739: or the orbifold-GUT compactification scale, which is usually only 

740: marginally lower). Thus, our findings are, in fact, very encouraging since 

741: even small running effects can easily turn the high-scale equalities into the 

742: desired inequalities of Eqs.~(\ref{lar}) and (\ref{sma}). 

743: 

744: We now discuss in more detail how this running modification of our 

745: high-scale relations may occur. The crucial renormalization group 

746: equations are

747: \bea

748: 16\pi^2\frac{d}{dt}\mu&=&\mu\left[3|y_t|^2-3g_2^2\right]\,,\\ \nonumber\\

749: 16\pi^2\frac{d}{dt}(B\mu)&=&B\mu\left[3|y_t|^2-3g_2^2\right]

750: +\mu\left[6a_t\bar{y}_t+6g_2^2M_2\right]\,,\\ \nonumber\\

751: 16\pi^2\frac{d}{dt}m_{H_u}^2&=&6|y_t|^2\left[m_{H_u}^2+m_{Q_3}^2+m_{u_3}^2

752: \right]+6|a_t|^2-6g_2^2|M_2|^2\,,\\ \nonumber\\

753: 16\pi^2\frac{d}{dt}m_{H_d}^2&=&-6g_2^2|M_2|^2\,,

754: \eea

755: where, except for writing $B\mu$ instead of $b$, we follow the conventions 

756: of~\cite{Martin:1997ns}. Since, for the purposes of this paper, we are only 

757: interested in qualitative features, we have neglected all Yukawa couplings 

758: and trilinear couplings (except those of the top) as well as the U(1) gauge 

759: coupling $g_1$. 

760: 

761: From the above equations we first immediately recognize the well-known fact 

762: that, starting with $m_{H_u}^2=m_{H_d}^2$ at a high scale, one generically 

763: finds $m_{H_u}^2<m_{H_d}^2$ at the electroweak scale, essentially because of 

764: the effects of the large top Yukawa coupling. We also see from the formulae 

765: at the beginning of this section that both $m_{H_u}^2$ and $m_{H_d}^2$ can 

766: easily be negative from the beginning in our setting.

767: 

768: Thus, $(|\mu|^2+m_{H_u}^2)<(|\mu|^2+m_{H_d}^2)$ at the low scale and the 

769: inequalities of Eqs.~(\ref{lar}) and (\ref{sma}) can, in principle, be 

770: satisfied simultaneously. Clearly, whether this actually happens depends on 

771: the running of $\mu$ and $B\mu$ and on their initial values. This depends, 

772: in turn, on the fundamental parameters $F_T$, $F_\varphi$ and $c'$ of our 

773: construction. Furthermore, the running also depends on the soft masses

774: and trilinear couplings in the top quark sector. Since, as we will discuss in 

775: more detail in Sect.~\ref{bns}, the matter fields originate in bulk 

776: hypermultiplets, the relevant terms come from the superfields 

777: expressions~\cite{Marti:2001iw}

778: \be 

779: {\cal L}_{hyp.,\,4d}\supset \int d^4\theta \varphi\bar{\varphi}\frac{1}{2}

780: (T+\bar{T})\left(H^\dagger e^{-2V}H+H^ce^{2V}H^{c\dagger}\right)+

781: \int d^2\theta\varphi^3H^c\Phi H+\mbox{h.c.}

782: \ee

783: Unfortunately, as will again be explained in Sect.~\ref{bns} referring to 

784: the model of~\cite{bn}, realistic Yukawa couplings require many such 

785: hypermultiplet terms with non-trivial bulk profiles as well mixing with 

786: brane localized charged fields. Thus, we can not simply write down the 

787: soft squark masses and trilinear couplings without entering more deeply 

788: in the matter sector of our model. 

789: 

790: Nevertheless, we see from the above that, using the freedom of choosing 

791: $F_T$, $F_\varphi$, $c'$ and of the bulk field localization and bulk-brane 

792: mixing in the matter sector, it is very plausible that realistic low-scale 

793: SUSY-breaking parameters and $\mu$ term can result from our fundamental 

794: high-scale formulae, Eqs.~(\ref{m12})--(\ref{mut}). In situations without 

795: a Chern-Simons term, a numerical analysis of the running of the relevant 

796: parameters has already been performed in Ref.~\cite{choi}, using certain 

797: plausible assumptions about soft parameters in the top-quark sector. The 

798: authors came to the conclusion that, given the strong high-scale constraints, 

799: correct electroweak symmetry breaking is difficult to achieve. They identified 

800: the prediction $m_{H_u,\,H_d}^2=-m_{1/2}^2$ as one of the main reasons for 

801: this difficulty. However, in our model with a Chern-Simons term, precisely 

802: this constraint is lifted. In fact, as one can see from 

803: Eqs.~(\ref{m12})--(\ref{mut}), the parameters $m_{1/2}^2$ and $m_{H_u,\,H_d}^2$

804: blow up for different negative values of $c'$, implying that any 

805: high-scale ratio of these quantities can, in principle, be realized. Indeed,

806: as has recently been demonstrated in~\cite{Brummer:2009ug}, the inclusion of 

807: the Chern-Simons term in this type of gauge-Higgs unification models allows 

808: for a realistic low-energy phenomenology.

809: 

810: 

811: 

812: 

813: 

814: \section{An explicit SU(6) orbifold-GUT model}\label{bns}

815: Both the U(6) model analysed above as well as the more minimal pure SU(6)

816: model briefly discussed in a footnote in Sect.~{\ref{nag}}, do not represent

817: `clean' versions of field-theoretic orbifolding. Indeed, the U(1) factor in

818: U(6) does not allow, in the presence of charged matter, for a breaking by 

819: a $Z_2$ symmetry of the original action. The pure SU(6) model, on the other

820: hand, inherently relies on the gauge symmetry breaking by (non-orbifold)

821: boundary conditions. Thus, it is interesting to see whether a 5d model can be 

822: found which realizes all the essential features of our scenario by just 

823: modding out a set of $Z_2$ symmetries. In the present section, we provide a 

824: positive answer to this question, modifying the model of~\cite{bn}

825: appropriately. However, this construction has problems of its own which 

826: are related to precision gauge coupling unification (see below). 

827: 

828: Although we are ultimately interested in 5d orbifold GUT models with 

829: gauge-Higgs unification, the simplest way to approach our model is from a 

830: 6d perspective. We start from 6d ${\cal N}=2$ super-Yang-Mills theory 

831: with gauge group SU(6) compactified on a torus $T^2$. The torus is 

832: parameterized by a complex coordinate $z$ with the fundamental domain being 

833: defined by $0\le\mbox{Re}z<2\pi R_6$ and $0\le\mbox{Im}z<2\pi R_5$. We 

834: restrict the field space of the model by requiring invariance under two 

835: orbifold projections $P$ and $P'$. With each of these operations we associate 

836: SU(6) matrices which characterize the orbifold action in gauge space and which 

837: we denote by the same symbol: $P=i\,\mbox{diag}(1,1,1,1,1,-1)$ and $P'=

838: \mbox{diag}(1,1,-1,-1,-1,-1)$. The invariance requirements for the 

839: ${\cal N}=1$ vector superfield $V$ contained in the 6d gauge multiplet are

840: \be

841: PV(z)P^{-1}=V(-z)\qquad\mbox{and}\qquad P'V(z-\pi/2)P'^{-1}=V(-(z-\pi/2))\,.

842: \ee

843: Similar relations, but with an extra minus sign, hold for the chiral 

844: superfield $\Phi$, which contains the remaining degrees of freedom of the 

845: 6d gauge multiplet. 

846: 

847: The resulting theory can be visualized as a 6d model the compactification 

848: space of which has the geometry of a pillow (cf.~Fig.~\ref{6d}). This space 

849: has four conical singularities, each with deficit angle $\pi$, two of which 

850: are due to the projection $P$ and the other two of which are due to the 

851: projection $P'$. Correspondingly, the gauge symmetry is locally restricted 

852: at these singularities to SU(5)$\times$U(1) for $P$ and to SU(4)$\times$SU(2)$

853: \times$U(1) for $P'$. 

854: 

855: \begin{figure}

856: \begin{center}

857: \includegraphics[width=10cm]{6d.eps}

858: \caption{In two different 5d limits, the 6d model described in the text 

859: goes over into the model of Burdman/Nomura or into `Our Model'.}

860: \label{6d}

861: \end{center}

862: \end{figure}

863: 

864: We now observe that by taking the limit $R_5\to 0$, we arrive precisely at 

865: the 5d orbifold GUT model with gauge-Higgs unification of Burdman and 

866: Nomura~\cite{bn}. This limit is illustrated in Fig.~\ref{6d}. Indeed, 

867: in this limit the pillow degenerates to an interval and the fixed points 

868: with gauge group SU(5)$\times$U(1) (labelled by $P$) merge into a boundary 

869: of the 5d space with the same local gauge symmetry. Analogously, the two 

870: fixed points with gauge group SU(4)$\times$SU(2)$\times$U(1) merge and 

871: play the role of the other boundary or brane. 

872: 

873: We define our model by keeping $R_5$ finite and taking the limit $R_6\to 0$. 

874: This situation, which is also visualized in the figure, corresponds 

875: again to a 5d model compactified on an interval. However, the two

876: boundaries are now equivalent and the gauge symmetry at the boundary, which 

877: is restricted by both $P$ and $P'$, is the intersection of the two groups 

878: left invariant by the two projections. It is just the gauge symmetry of the 

879: standard model plus an extra U(1) factor (the U(1) left over when SU(6) is 

880: broken to SU(5)). 

881: 

882: The model that we have thus obtained is similar but not identical to the 5d 

883: model of Sect.~\ref{nag}: The original gauge symmetry, which is SU(6) rather 

884: than U(6), is broken at each boundary of the interval to ${\cal G_{\rm SM}}

885: \times$U(1) rather than simply to SU(5)$\times$U(1). In addition, the vacuum 

886: expectation value of $\Phi$ takes a less symmetric form. To determine this 

887: vacuum expectation value, 

888: we first recall that the scalar part of the chiral superfield $\Phi$ (which 

889: we denote by the same symbol) reads $\Phi=A_6+iA_5$ in the 6d construction. 

890: Furthermore, if a charged particle encircles the stretched pillow (labelled

891: `Our model' in Fig.~\ref{6d}) in the short direction, it experiences a gauge 

892: rotation

893: \be 

894: P\cdot P'=\exp[i(\pi/4)T]=\exp\left[i\int_0^{\pi R_6}A_6dx^6\right]\,. 

895: \ee

896: Here $T=\mbox{diag}(1,1,-1,-1,-1,1)$ is the generator of the gauge twist 

897: $P\cdot P'$ which is felt in the bulk of our effective 5d space and which 

898: breaks SU(6) to SU(3)$\times$SU(3)$\times$U(1). Thus, after dimensional 

899: reduction from 6d to 5d, we find $\langle\Phi\rangle=v\,\mbox{diag}(1,1,-1,

900: -1,-1,1)$ with $v=1/(4R_6)$. 

901: 

902: This result may appear puzzling since it seems to imply that the physical 

903: effects of $v$, introduced via the Chern-Simons term, become dominant in the 

904: 5d limit $R_6\to 0$. However, this is not the case for the following reason: 

905: The smallest $R_6$ for which our 6d motivation of the 5d model makes sense is 

906: $R_6\sim g_6$. For smaller $R_6$, the 6d approach is compromised by the fact 

907: that the strong-coupling scale of the 6d gauge theory lies below the 

908: compactification scale. Through the relation 

909: $1/g_5^2\sim R_6/g_6^2$, this limiting situation gives rise to an effective 

910: 5d gauge-coupling $g_5\sim\sqrt{R_6}$. We thus conclude from Eq.~(\ref{cp})

911: that the dimensionless parameter $c'$ governing the size of the physical 

912: effects induced by $v$ is indeed ${\cal O}(1)$ if the coefficient of the 

913: Chern-Simons term in the original lagrangian is $c\sim{\cal O}(1)$. Of course, 

914: the 6d supersymmetric gauge theory does not allow for a Chern-Simons term.

915: However, the 5d theory obtained after $S^1$-compactification includes such a

916: term because of loop effects. The group-theoretic structure of these loop

917: induced Chern-Simons terms, which have been discussed in some detail in

918: Sect.~5 of~\cite{Hebecker:2004xx} (see also~\cite{ims}), is somewhat 

919: different from that of the tree-level 5d Chern-Simons term.\footnote{

920: Such 

921: structures are possible because the loop-induced prepotential does not 

922: have to be holomorphic at the origin, $\Phi=0$. This allows for gauge 

923: invariant expressions different from tr$\,\Phi^3$.

924: }

925: However, the 

926: coefficient follows entirely from group theory and matter content and 

927: is thus naturally ${\cal O}(1)$. We will not derive these terms explicitly 

928: in the present 6d-motivated model but only reiterate that, as we claimed 

929: before, the physical effects of the Chern-Simons term in the presence of 

930: $v$ do indeed arise in more fundamental constructions and are, in general, 

931: comparable to the effects derived from the quadratic lagrangian. 

932: 

933: Let us finally turn to the problem of standard model matter fields and 

934: Yukawa couplings in the presented gauge-Higgs unification model. This is, 

935: in principle, a highly non-trivial issue since charged hypermultiplets 

936: have to be introduced in the bulk in such a way that, after the orbifold 

937: projections, the correct low-energy spectrum results. Furthermore, large 

938: 4d Yukawa couplings (in particular that of the top quark) can only result

939: from bulk gauge couplings because the two Higgs doublets come from the 

940: chiral superfield $\Phi$ in the $\bf 35$, which is part of the gauge 

941: multiplet and can not have any other interactions in the 5d (or 6d) bulk. 

942: 

943: However, concerning all of these issues we can simply refer the reader to 

944: the 5d SU(6) model of~\cite{bn}. In this model, all of the above issues 

945: have been solved: For example, the down- and up-type quarks are introduced 

946: as hypermultiplets in the ${\bf 15}$ and ${\bf 20}$ of SU(6) in the bulk, 

947: which mix with extra 4d chiral superfields introduced on the branes. It has 

948: then been shown that the top- and other Yukawa couplings can be correctly 

949: reproduced from the 5d couplings with the gauge multiplet. A similar procedure 

950: works for the leptons. The hierarchies of the Yukawa couplings can be 

951: realized by allowing for 5d bulk masses for the hypermultiplets, which lead

952: to exponential profiles of the fields and hence to very different effective 

953: 4d couplings for the zero modes of the hypermultiplets. 

954: 

955: Indeed, the whole construction of~\cite{bn} can straightforwardly be lifted 

956: to 6 dimensions. The field content in 5d and 6d is exactly the same. The 

957: orbifold $S^1/(Z_2\times Z_2')$ can be replaced by $T^2/(Z_2\times Z_2')$, 

958: as is visualized in Fig.~\ref{6d}. Instead of placing extra 4d chiral 

959: superfields and 4d superpotentials on the boundaries of the 5d interval, 

960: those can equally well be placed at the conical singularities of the 6d

961: orbifold. In short, the whole construction goes through without change. 

962: A critical issue appears to be the introduction of 5d bulk masses for the 

963: hypermultiplets, which is not possible for charged hypermultiplets in 6

964: dimensions. However, the 6d hypermultiplets may be charged under extra U(1) 

965: gauge groups. Wilson lines of these gauge groups (i.e. vacuum expectation 

966: values of $A_6$) then play the same role as 5d bulk masses and lead to 

967: localization effects for the zero modes. To summarize, we could simply copy 

968: the relevant pages of~\cite{bn}, changing the language from 5d to 6d. We 

969: will not do so since, in this paper, we do not intend to go beyond the 

970: demonstration that the type of model underlying our discussion of SUSY 

971: breaking in the Higgs sector does indeed arise in phenomenologically viable 

972: GUT models. 

973: 

974: Although the 6d lift of the 5d model 

975: of~\cite{bn} and its `opposite' 5d limit appear to be a very nice motivation

976: of our 5d framework, this is not the only way to approach our construction. 

977: Instead, we could simply say that our model is defined, from the start, 

978: on a 5d interval with gauge group SU(6) in the bulk. At each boundary, the 

979: gauge group is broken to ${\cal G}_{\rm SM}\times$U(1) (which is not a

980: $Z_2$ orbifold breaking) and a non-zero vacuum expectation value for 

981: $\Sigma$ is enforced by the boundary conditions. The inclusion of matter and 

982: the generation of Yukawa couplings can be achieved in analogy to the 

983: similar 5d gauge-Higgs unification model of~\cite{bn}. From this perspective, 

984: our model remains 5-dimensional. The `pillow' of Fig.~\ref{6d} and its 5d 

985: limit merely serve to convince the reader that non-orbifold 5d boundary 

986: conditions are natural, for example as the result of two merging conical 

987: singularities with gauge breaking by $P$ and $P'$. 

988: 

989: We finally note that, since the 5d vev used in this section does not preserve 

990: the SU(5) subgroup, large threshold corrections to gauge-coupling unification 

991: will generically be present~\cite{Hebecker:2004xx}. This is not necessarily 

992: fatal since the size of these thresholds and the way in which they affect the 

993: low-energy couplings is highly model dependent. However, it would 

994: require a more detailed analysis to establish whether a fully realistic 

995: low-energy phenomenology can emerge. Such an analysis is beyond the scope

996: of the present investigation.

997: 

998: 

999: 

1000: 

1001: 

1002: \section{Conclusions}\label{conc}

1003: We have analysed supersymmetry breaking and the supersymmetric $\mu$ term 

1004: in the Higgs sector of 5-dimensional models with gauge-Higgs unification. 

1005: This setting is well-motivated both from the perspective of 5d or 6d 

1006: orbifold GUTs, which are arguably the simplest realistic grand unified 

1007: theories on the market, as well as from the perspective of the most 

1008: successful heterotic string models. 

1009: 

1010: Gaugino masses, soft Higgs masses, as well as the $\mu$ and $B\mu$ term 

1011: are generated in a natural way once the $F$ terms of the radion superfield 

1012: and the chiral compensator acquire non-zero vacuum expectation values. This 

1013: happens in many of the simplest models where the radion (the size of the 

1014: 5th dimension) is stabilized with the help of a non-trivial superpotential. 

1015: The relative size of the SUSY-breaking parameters and the $\mu$ term 

1016: depend on ratio of the two $F$ terms, $F_\varphi/F_T$. The overall scale is 

1017: set by the ratio of the radion $F$ term and the size of the extra dimension, 

1018: $F_T/T$. This means that low-scale supersymmetry is realized if the 

1019: high-scale theory exhibits weak Scherk-Schwarz breaking (known as 

1020: radion mediation). 

1021: 

1022: In addition to the effects based on the quadratic gauge theory lagrangian, 

1023: the 5d supersymmetric Chern-Simons term can play a crucial role. This is, in 

1024: fact, expected since the Chern-Simons term is an unavoidable part of generic 

1025: 5d models compactified on an interval. Its importance for the low-energy 

1026: effective theory depends on the presence of a large vacuum expectation value 

1027: of the 5d scalar in the gauge multiplet. Such a vacuum expectation value 

1028: can be viewed as a Wilson line from the perspective an underlying 6d or 

1029: string model. Its size is then naturally of the right order of magnitude to 

1030: compete with the effects of the quadratic lagrangian. 

1031: 

1032: If, as explained above, supersymmetry breaking is governed by both the 

1033: quadratic lagrangian and the Chern-Simons term, all relevant terms are 

1034: generated just on the basis of the $F$ term of the chiral compensator. 

1035: One can then consider the limit where the $F$ term of the chiral compensator 

1036: vanishes, corresponding e.g. to the stabilization of the radion purely by 

1037: K\"ahler corrections. 

1038: 

1039: The details of the resulting low-energy phenomenology are sensitive 

1040: to the various high-scale parameters, in particular $F_\varphi$, $F_T$ and 

1041: the vacuum expectation value of the 5d scalar (the real part of the chiral 

1042: adjoint). However, an interesting feature that appears to be universal 

1043: within the class of models that we have investigated is the high-scale 

1044: relation $B\mu=|\mu|^2+m_{H_u}^2=|\mu|^2+m_{H_d}^2$. This relation between 

1045: $B\mu$ term, $\mu$ term and soft Higgs masses is at the borderline of validity 

1046: of the standard inequalities which have to be imposed for successful 

1047: electroweak

1048: symmetry breaking. Thus, we rely on running effects to lift the equality 

1049: $m_{H_u}^2=m_{H_d}^2$, which is standard, and on an appropriate running of

1050: $\mu$ and $B\mu$ to satisfy the necessary low-energy constraints. As 

1051: demonstrated in~\cite{Brummer:2009ug}, the Chern-Simons term, which lifts 

1052: certain extra constraints, is crucial to avoid the negative conclusions 

1053: concerning the low-energy phenomenology of related models reached 

1054: in~\cite{choi}. Thus, the proposed version of supersymmetric gauge-Higgs 

1055: unification with a 5d Chern-Simons term defines an interesting new class of 

1056: potentially realistic GUT models. 

1057: 

1058: 

1059: \section*{Acknowledgments}

1060: We would like to thank Felix Br\"ummer for pointing out a problem in an

1061: earlier version of this paper, as well as its resolution.

1062: 

1063: 

1064: 

1065: \section*{Appendix}

1066: This appendix is devoted to the construction of a superfield expression 

1067: for the non-abelian supersymmetric Chern-Simons term. Suppressing a possible 

1068: overall prefactor, the superfield expression for the abelian 5d Chern-Simons 

1069: term is given by~\cite{Arkani-Hamed:2001tb}

1070: \bea

1071: {\cal L}_{cs} & = & \int d^2\theta\,\,\Phi\,W^2 + {\rm h.c.} 

1072: \nonumber \\ 

1073: & & \hspace*{-.7cm} + \frac{2}{3} \int d^4\theta \; \left( \partial_5 

1074: V D_\alpha V - V D_\alpha \partial_5 V  \right) W^\alpha + {\rm h.c.}  

1075: \nonumber \\

1076: & & \hspace*{-.7cm} - \frac{1}{6} \int d^4\theta \; \left( 2 \partial_5 

1077: V - (\Phi + \bar{\Phi}) \right)^3 .\label{lcsz}

1078: \eea

1079: The simple 4d procedure for the non-abelian generalization, i.e. the 

1080: replacement $V \to e^{\pm2V}$, does not work in this case. Instead, we 

1081: construct the non-abelian lagrangian by matching an appropriate 

1082: superfield expression (in Wess-Zumino gauge) to the component action. 

1083: Working within this approach is straightforward because the number of 

1084: possible superfield actions is highly restricted and the calculation can be 

1085: performed in close analogy to the abelian case.

1086: 

1087: Our starting point is the 5d Chern-Simons action of the (non-supersymmetric)

1088: non-abelian gauge theory, which can be constructed from the 5d Chern-Simons 

1089: form given in~\cite{Nakahara:1990th}:

1090: \be

1091: \label{nacscomp}

1092: {\cal L}_{cs\, gauge} = \epsilon^{MNOPQ} \, {\rm tr} \left({\frac{1}{4} 

1093: A_M F_{NO} F_{PQ} - \frac{i}{4} A_M A_N A_O F_{PQ} - \frac{1}{10} 

1094: A_M A_N A_O A_P A_Q}\right)

1095: \ee

1096: with the non-abelian field strength 

1097: \be

1098: F_{MN}=\partial_M A_N - \partial_N A_M + i [A_M,A_N]\,. 

1099: \ee

1100: This expression must be reproduced by a superfield lagrangian which contains 

1101: the fields $\Phi,V,W_{\alpha}$ with bosonic components

1102: \bea

1103: \Phi & = & \Sigma(y) + iA_5(y) + \theta^2 F_\Phi(y) \nonumber \\

1104: % Expanding y 

1105: % & = & \Sigma(x) + iA_5(x) + i\theta \sigma^\mu \bar{\theta} \partial_\mu 

1106: % (\Sigma(x) + iA_5(x)) + \theta^2 F_\chi(x) + \frac{1}{4} \theta^2 

1107: % \bar{\theta}^2 \partial^2 (\Sigma(x) + iA_5(x))

1108: V_{{\rm WZ}} & = & - \theta \, \sigma^\mu \bar{\theta} A_\mu(x) + 

1109: \frac{1}{2} \theta^2 \bar{\theta}^2 D(x) \label{cexp}\\

1110: W_\alpha & = & \theta_\alpha D(y) - i \left( \sigma^{\mu \nu} 

1111: \right)_\alpha{}^\beta \theta_\beta F_{\mu \nu}(y)\,, \nonumber 

1112: \eea

1113: where $y = x + i \theta \sigma \bar{\theta}$. Note that the field strength 

1114: superfield 

1115: \be

1116: W_\alpha=- \frac{1}{8} \bar{D}^2 \left( e^{-2V} D_\alpha e^{2V} \right)

1117: \ee

1118: gives, in Wess-Zumino gauge, only terms linear and quadratic in $V$:

1119: \be

1120: W_\alpha=W_\alpha^{(1)} + W_\alpha^{(2)}

1121: \ee

1122: with

1123: \bea

1124: W_\alpha^{(1)} & = &  -\frac{1}{4}\bar{D}^2D_\alpha V \hspace*{.8cm}=\,\, 

1125: \theta_\alpha D(y) - 2i \left( \sigma^{\mu \nu} \right)_\alpha{}^\beta 

1126: \theta_\beta \partial_\mu A_\nu(y) 

1127: \nonumber \\

1128: W_\alpha^{(2)} & = & -\frac{1}{4}\bar{D}^2[D_\alpha V,V]\,\,=\,\,2 \left( 

1129: \sigma^{\mu \nu} \right)_\alpha{}^\beta \theta_\beta A_\mu(y) A_\nu(y)\,,

1130: \eea 

1131: which reproduces the expression in Eq.~(\ref{cexp}). 

1132: 

1133: It is convenient to rewrite Eq.~(\ref{nacscomp}) as

1134: \be

1135: \label {nacscomp5}

1136: {\cal L}_{cs\,gauge} = \epsilon^{\mu \nu \rho \sigma}\,{\rm tr} \left( 

1137: \frac{3}{4} A_5 F_{\mu \nu} F_{\rho \sigma} -\frac{1}{2} \{A_\mu,

1138: \partial_5 A_\nu\} F_{\rho \sigma} + \frac{i}{4} \{A_\mu,\partial_5 A_\nu\} 

1139: A_\rho A_\sigma \right)\,,

1140: \ee

1141: where the curly brackets denote anticommutators. It can be checked that the 

1142: variation of this expression under gauge transformations is a total 

1143: derivative.

1144: 

1145: The first term in Eq.~(\ref{nacscomp5}) is obtained from a superfield 

1146: lagrangian which is of the same form as in the abelian case:

1147: \be

1148: {\rm tr} \left( \int {d^2 \theta} \, \Phi \, W^\alpha W_\alpha + {\rm h.c.} 

1149: \right)\,.\label{p1}

1150: \ee

1151: The second term is reproduced by a piece which is also similar to the abelian 

1152: case:

1153: \be

1154: {\rm tr} \left( \int {d^4 \theta} \, \left( \{\partial_5 V,D_\alpha V\} - 

1155: \{V,\partial_5 D_\alpha V\} \right) W^\alpha + {\rm h.c.} \right). \label{p2}

1156: \ee

1157: For the last term, it is necessary to use just the part of $W_\alpha$ 

1158: quadratic in $V$:

1159: \be

1160: {\rm tr} \left( \int {d^4 \theta} \; \left( \{\partial_5 V,D_\alpha V\} - 

1161: \{V,\partial_5 D_\alpha V\} \right) W^{\alpha}_{(2)} + {\rm h.c.} \right)\,.

1162: \label{p3}

1163: \ee

1164: The above three terms already reproduce the non-supersymmetric 5d CS term of 

1165: Eq.~(\ref {nacscomp5}), but 5d Lorentz invariance is violated by a term 

1166: $\sim \Sigma F_{\mu\nu}F^{\mu\nu}$ coming from Eq.~(\ref{p1}). This can be 

1167: cured by adding a further contribution, which is a simple 

1168: generalization of the last term in the abelian CS action:

1169: \be

1170: {\rm tr} \int {d^4 \theta}  \left( e^{-2V} \nabla_5 \, e^{2V} \right)^3\,.

1171: \label{p4}

1172: \ee

1173: Here we have used the super gauge covariant derivative 

1174: \be

1175: \nabla_5 \equiv \partial_5 + \Phi,

1176: \ee

1177: acting on $e^{2V}$ as 

1178: \be

1179: \nabla_5 e^{2V} = \partial_5 e^{2V} - \Phi^\dagger e^{2V} - e^{2V} \Phi.

1180: \ee

1181: 

1182: The relative prefactors of the four contributions of 

1183: Eqs.~(\ref{p1})--(\ref{p4}) are fixed by an explicit calculation and found 

1184: to be consistent with those of the abelian action. Up to an overall 

1185: constant factor, the result is that of~Eq.~(\ref{lcsna}). 

1186: Although the evaluation of this manifestly supersymmetric expression in WZ 

1187: gauge reproduces the CS component lagrangian of Eq.~(\ref{nacscomp}), we were 

1188: not able to show that it transforms into a total derivative under super 

1189: gauge transformations. Most probably this is due to missing extra terms that 

1190: vanish in WZ gauge. It would be interesting to construct these missing 

1191: contributions and achieve manifest super gauge invariance (as it is realized 

1192: for the leading order lagrangian in Eq.~(\ref{l2na})). 

1193: 

1194: It requires a certain amount of work to extract even just the bosonic part 

1195: of our full superfield Chern-Simons lagrangian. One has to integrate by 

1196: parts using the fact that $\Sigma$ vanishes at the boundaries. Furthermore, 

1197: $F_\Phi$ is set to zero by the equations of motion, while $D$ takes the value

1198: \be

1199: D = -\partial_5 \Sigma + i [\Sigma,A_5]\,.

1200: \ee

1201: The final result is

1202: \be

1203: {\cal L}_{cs} \supset c\left[\,\frac{2}{3} {\cal L}_{cs\,gauge}  - {\rm tr} 

1204: \left( \Sigma F_{MN} F^{MN} + 2 \, \Sigma (D_M \Sigma)\,(D^M \Sigma)\right)

1205: \,\right]\,,

1206: \ee

1207: where

1208: \be

1209: D_M \Sigma = \partial_M \Sigma + i [A_M,\Sigma]\,.

1210: \ee

1211: This also fixes the normalization of our superfield expression relative to 

1212: the non-supersymmetric Chern-Simons term. 

1213: 

1214: \begin{thebibliography}{99}

1215: 

1216: \bibitem{Kim:1983dt}

1217:   J.~E.~Kim and H.~P.~Nilles,

1218:   ``The Mu Problem And The Strong CP Problem,''

1219:   Phys.\ Lett.\  B {\bf 138} (1984) 150.

1220:   %%CITATION = PHLTA,B138,150;%%

1221: 

1222: \bibitem{Giudice:1988yz}

1223:   G.~F.~Giudice and A.~Masiero,

1224:   ``A Natural Solution to the mu Problem in Supergravity Theories,''

1225:   Phys.\ Lett.\  B {\bf 206} (1988) 480.

1226:   %%CITATION = PHLTA,B206,480;%%

1227: 

1228: \bibitem{Fayet:1974pd}

1229:   P.~Fayet,

1230:   ``Supergauge Invariant Extension Of The Higgs Mechanism And A Model For The

1231:   Electron And Its Neutrino,''

1232:   Nucl.\ Phys.\  B {\bf 90} (1975) 104;\\

1233:   %%CITATION = NUPHA,B90,104;%%

1234: %\bibitem{Nilles:1982dy}

1235:   H.~P.~Nilles, M.~Srednicki and D.~Wyler,

1236:   ``Weak Interaction Breakdown Induced By Supergravity,''

1237:   Phys.\ Lett.\  B {\bf 120} (1983) 346;\\

1238:   %%CITATION = PHLTA,B120,346;%%

1239: %\bibitem{Derendinger:1983bz}

1240:   J.~P.~Derendinger and C.~A.~Savoy,

1241:   ``Quantum Effects And SU(2) X U(1) Breaking In Supergravity Gauge Theories,''

1242:   Nucl.\ Phys.\  B {\bf 237} (1984) 307;\\

1243:   %%CITATION = NUPHA,B237,307;%%

1244: %\bibitem{Gunion:1984yn}

1245:   J.~F.~Gunion and H.~E.~Haber,

1246:   ``Higgs Bosons In Supersymmetric Models. 1,''

1247:   Nucl.\ Phys.\  B {\bf 272} (1986) 1

1248:   [Erratum-ibid.\  B {\bf 402} (1993) 567];\\

1249:   %%CITATION = NUPHA,B272,1;%%

1250: %\bibitem{Ellis:1988er}

1251:   J.~R.~Ellis, J.~F.~Gunion, H.~E.~Haber, L.~Roszkowski and F.~Zwirner,

1252:   ``Higgs Bosons in a Nonminimal Supersymmetric Model,''

1253:   Phys.\ Rev.\  D {\bf 39} (1989) 844.

1254:   %%CITATION = PHRVA,D39,844;%%

1255: 

1256: \bibitem{Barbieri:2007tu}

1257:   R.~Barbieri, L.~J.~Hall, A.~Y.~Papaioannou, D.~Pappadopulo and V.~S.~Rychkov,

1258:   ``An alternative NMSSM phenomenology with manifest perturbative

1259:   unification,''

1260:   arXiv:0712.2903 [hep-ph];\\

1261:   %%CITATION = ARXIV:0712.2903;%%

1262: %\bibitem{Giudice:2007ca}

1263:   G.~F.~Giudice, H.~D.~Kim and R.~Rattazzi,

1264:   ``Natural mu and Bmu in gauge mediation,''

1265:   arXiv:0711.4448 [hep-ph];\\

1266:   %%CITATION = ARXIV:0711.4448;%%

1267: %\bibitem{Murayama:2007ge}

1268:   H.~Murayama, Y.~Nomura and D.~Poland,

1269:   ``More Visible Effects of the Hidden Sector,''

1270:   Phys.\ Rev.\  D {\bf 77} (2008) 015005

1271:   [arXiv:0709.0775 [hep-ph]];\\

1272:   %%CITATION = PHRVA,D77,015005;%%

1273: %\bibitem{Lee:2007fw}

1274:   H.~S.~Lee, K.~T.~Matchev and T.~T.~Wang,

1275:   ``A U(1)' solution to the mu-problem and the proton decay problem in

1276:   supersymmetry without R-parity,''

1277:   arXiv:0709.0763 [hep-ph].

1278:   %%CITATION = ARXIV:0709.0763;%%

1279: 

1280: \bibitem{Manton:1979kb}

1281:   N.~S.~Manton,

1282:   ``A New Six-Dimensional Approach To The Weinberg-Salam Model,''

1283:   Nucl.\ Phys.\  B {\bf 158} (1979) 141.

1284:   %%CITATION = NUPHA,B158,141;%%

1285: 

1286: \bibitem{choi}

1287:   K.~w.~Choi, N.~y.~Haba, K.~S.~Jeong, K.~i.~Okumura, Y.~Shimizu and 

1288:   M.~Yamaguchi,

1289:   ``Electroweak symmetry breaking in supersymmetric gauge - Higgs  unification

1290:   models,''

1291:   JHEP {\bf 0402} (2004) 037

1292:   [arXiv:hep-ph/0312178].

1293:   %%CITATION = JHEPA,0402,037;%%

1294: 

1295: \bibitem{str}

1296: %\bibitem{LopesCardoso:1994is}

1297:   G.~Lopes Cardoso, D.~L\"ust and T.~Mohaupt,

1298:   ``Moduli spaces and target space duality symmetries in (0,2) Z(N) orbifold

1299:   theories with continuous Wilson lines,''

1300:   Nucl.\ Phys.\  B {\bf 432} (1994) 68

1301:   [arXiv:hep-th/9405002];\\

1302:   %%CITATION = NUPHA,B432,68;%

1303: %\cite{Antoniadis:1994hg}

1304: %\bibitem{Antoniadis:1994hg}

1305:   I.~Antoniadis, E.~Gava, K.~S.~Narain and T.~R.~Taylor,

1306:   ``Effective mu term in superstring theory,''

1307:   Nucl.\ Phys.\  B {\bf 432} (1994) 187

1308:   [arXiv:hep-th/9405024];\\

1309:   %%CITATION = NUPHA,B432,187;%%

1310: %\bibitem{Brignole:1995fb}

1311:   A.~Brignole, L.~E.~Ib\'a\~nez, C.~Mu\~noz and C.~Scheich,

1312:   ``Some Issues In Soft Susy Breaking Terms From Dilaton / Moduli Sectors,''

1313:   Z.\ Phys.\  C {\bf 74} (1997) 157

1314:   [arXiv:hep-ph/9508258];\\

1315:   %%CITATION = ZEPYA,C74,157;%%

1316: %\bibitem{Brignole:1996xb}

1317:   A.~Brignole, L.~E.~Ib\'a\~nez and C.~Mu\~noz,

1318:   ``Orbifold-induced mu term and electroweak symmetry breaking,''

1319:   Phys.\ Lett.\  B {\bf 387} (1996) 769

1320:   [arXiv:hep-ph/9607405];\\

1321:   %%CITATION = PHLTA,B387,769;%%

1322: %\bibitem{Brignole:1997dp}

1323:   A.~Brignole, L.~E.~Ib\'a\~nez and C.~Mu\~noz,

1324:   ``Soft supersymmetry-breaking terms from supergravity and superstring

1325:   models,''

1326:   arXiv:hep-ph/9707209.

1327:   %%CITATION = HEP-PH/9707209;%%

1328: 

1329: \bibitem{orb}

1330: Y.~Kawamura,

1331: Prog.\ Theor.\ Phys.\  {\bf 105} (2001) 999

1332: [arXiv:hep-ph/0012125];\\

1333: %%CITATION = HEP-PH 0012125;%%

1334: G.~Altarelli and F.~Feruglio, Phys.\ Lett.\ B {\bf 511} (2001) 257

1335: [arXiv:hep-ph/0102301];\\

1336: %%CITATION = HEP-PH 0102301;%%

1337: L.~J.~Hall and Y.~Nomura, Phys.\ Rev.\ D {\bf 64} (2001) 055003

1338: [arXiv:hep-ph/0103125];\\

1339: %%CITATION = HEP-PH 0103125;%%

1340: A.~Hebecker and J.~March-Russell, Nucl.\ Phys.\ B {\bf 613} (2001) 3\\{}

1341: [arXiv:hep-ph/0106166];\\

1342: %%CITATION = HEP-PH 0106166;%%

1343: %\bibitem{hns}

1344: L.~J.~Hall, Y.~Nomura and D.~R.~Smith, Nucl.\ Phys.\ B {\bf 639} (2002) 307

1345: \\{}[arXiv:hep-ph/0107331];\\

1346: %%CITATION = HEP-PH 0107331;%%

1347: %\bibitem{Asaka:2001eh}

1348: T.~Asaka, W.~Buchmuller and L.~Covi,

1349: ``Gauge unification in six dimensions,''

1350: Phys.\ Lett.\  B {\bf 523} (2001) 199

1351: [arXiv:hep-ph/0108021].

1352: %%CITATION = PHLTA,B523,199;%%

1353: 

1354: \bibitem{ky}

1355:   T.~Kobayashi and K.~Yoshioka, ``Kaluza-Klein mediated supersymmetry 

1356:   breaking,'' Phys.\ Rev.\ Lett.\  {\bf 85} (2000) 5527

1357:   [arXiv:hep-ph/0008069];\\

1358: %%CITATION = HEP-PH 0008069;%%

1359:   Z.~Chacko and M.~A.~Luty, ``Radion mediated supersymmetry breaking,'' JHEP 

1360:   {\bf 0105} (2001) 067 [arXiv:hep-ph/0008103]. 

1361: %%CITATION = HEP-PH 0008103;%%

1362: 

1363: \bibitem{Buchmuller:2005jr}

1364:   W.~Buchmuller, K.~Hamaguchi, O.~Lebedev and M.~Ratz,

1365:   ``Supersymmetric standard model from the heterotic string,''

1366:   Phys.\ Rev.\ Lett.\  {\bf 96} (2006) 121602

1367:   [arXiv:hep-ph/0511035] and 

1368:   %%CITATION = PRLTA,96,121602;%%

1369: %\bibitem{Buchmuller:2006ik}

1370:   %W.~Buchmuller, K.~Hamaguchi, O.~Lebedev and M.~Ratz,

1371:   %``Supersymmetric standard model from the heterotic string. II,''

1372:   Nucl.\ Phys.\  B {\bf 785} (2007) 149

1373:   [arXiv:hep-th/0606187];\\

1374:   %%CITATION = NUPHA,B785,149;%%

1375: %\bibitem{Lebedev:2006kn}

1376:   O.~Lebedev, H.~P.~Nilles, S.~Raby, S.~Ramos-Sanchez, M.~Ratz, 

1377:   P.~K.~S.~Vaudrevange and A.~Wingerter,

1378:   ``A mini-landscape of exact MSSM spectra in heterotic orbifolds,''

1379:   Phys.\ Lett.\  B {\bf 645} (2007) 88

1380:   [arXiv:hep-th/0611095] and 

1381:   %%CITATION = PHLTA,B645,88;%%

1382: %\bibitem{Lebedev:2007hv}

1383:   %O.~Lebedev, H.~P.~Nilles, S.~Raby, S.~Ramos-Sanchez, M.~Ratz, 

1384:   %P.~K.~S.~Vaudrevange and A.~Wingerter,

1385:   ``The Heterotic Road to the MSSM with R parity,''

1386:   arXiv:0708.2691 [hep-th].

1387:   %%CITATION = ARXIV:0708.2691;%%

1388: 

1389: \bibitem{Kobayashi:2004ud}

1390:   T.~Kobayashi, S.~Raby and R.~J.~Zhang,

1391:   ``Constructing 5d orbifold grand unified theories from heterotic strings,''

1392:   Phys.\ Lett.\  B {\bf 593} (2004) 262

1393:   [arXiv:hep-ph/0403065];\\

1394:   %%CITATION = PHLTA,B593,262;%%

1395: %\bibitem{Forste:2004ie}

1396:   S.~Forste, H.~P.~Nilles, P.~K.~S.~Vaudrevange and A.~Wingerter,

1397:   ``Heterotic brane world,''

1398:   Phys.\ Rev.\  D {\bf 70} (2004) 106008

1399:   [arXiv:hep-th/0406208];\\

1400:   %%CITATION = PHRVA,D70,106008;%%

1401: %\bibitem{Buchmuller:2004hv}

1402:   W.~Buchmuller, K.~Hamaguchi, O.~Lebedev and M.~Ratz,

1403:   ``Dual models of gauge unification in various dimensions,''

1404:   Nucl.\ Phys.\  B {\bf 712} (2005) 139

1405:   [arXiv:hep-ph/0412318].

1406:   %%CITATION = NUPHA,B712,139;%%

1407: 

1408: \bibitem{Hebecker:2004ce}

1409:   A.~Hebecker and M.~Trapletti,

1410:   ``Gauge unification in highly anisotropic string compactifications,''

1411:   Nucl.\ Phys.\  B {\bf 713} (2005) 173

1412:   [arXiv:hep-th/0411131].

1413:   %%CITATION = NUPHA,B713,173;%%

1414: 

1415: \bibitem{bn}

1416:   G.~Burdman and Y.~Nomura,

1417:   ``Unification of Higgs and gauge fields in five dimensions,''

1418:   Nucl.\ Phys.\  B {\bf 656} (2003) 3

1419:   [arXiv:hep-ph/0210257].

1420:   %%CITATION = NUPHA,B656,3;%%

1421: 

1422: \bibitem{ims}

1423:   N.~Seiberg,

1424:   ``Five dimensional SUSY field theories, non-trivial fixed points and string

1425:   dynamics,''

1426:   Phys.\ Lett.\  B {\bf 388} (1996) 753

1427:   [arXiv:hep-th/9608111];\\

1428:   %%CITATION = PHLTA,B388,753;%%

1429: %\bibitem{Intriligator:1997pq}

1430:   K.~A.~Intriligator, D.~R.~Morrison and N.~Seiberg,

1431:   ``Five-dimensional supersymmetric gauge theories and degenerations of

1432:   Calabi-Yau spaces,''

1433:   Nucl.\ Phys.\  B {\bf 497} (1997) 56

1434:   [arXiv:hep-th/9702198].

1435:   %%CITATION = NUPHA,B497,56;%%

1436: 

1437: \bibitem{Kuzenko:2006ek}

1438:   S.~M.~Kuzenko, ``Five-dimensional supersymmetric Chern-Simons action as a

1439:   hypermultiplet quantum correction,'' arXiv:hep-th/0609078.

1440:   %%CITATION = HEP-TH 0609078;%%

1441: 

1442: \bibitem{Hebecker:2004xx}

1443:   A.~Hebecker and A.~Westphal,

1444:   ``Gauge unification in extra dimensions: Power corrections vs.

1445:   higher-dimension operators,''

1446:   Nucl.\ Phys.\  B {\bf 701} (2004) 273

1447:   [arXiv:hep-th/0407014].

1448:   %%CITATION = NUPHA,B701,273;%%

1449: 

1450: \bibitem{Brummer:2009ug}

1451:   F.~Brummer, S.~Fichet, A.~Hebecker and S.~Kraml,

1452:   ``Phenomenology of Supersymmetric Gauge-Higgs Unification,''

1453:   arXiv:0906.2957 [hep-ph].

1454:   %%CITATION = ARXIV:0906.2957;%%

1455: 

1456: \bibitem{Marcus:1983wb}

1457:   N.~Marcus, A.~Sagnotti and W.~Siegel, ``Ten-Dimensional Supersymmetric

1458:   Yang-Mills Theory In Terms Of Four-Dimensional Superfields,''

1459:   Nucl.\ Phys.\ B {\bf 224}, 159 (1983).

1460:   %%CITATION = NUPHA,B224,159;%%

1461: 

1462: \bibitem{Arkani-Hamed:2001tb}

1463:   N.~Arkani-Hamed, T.~Gregoire and J.~G.~Wacker, ``Higher dimensional 

1464:   supersymmetry in 4D superspace,'' JHEP {\bf 0203} (2002) 055

1465:   [arXiv:hep-th/0101233].

1466:   %%CITATION = HEP-TH 0101233;%%

1467: 

1468: \bibitem{Marti:2001iw}

1469:   D.~Marti and A.~Pomarol, ``Supersymmetric theories with compact extra

1470:   dimensions in N = 1 superfields,'' Phys.\ Rev.\ D {\bf 64} (2001) 105025

1471:   [arXiv:hep-th/0106256].

1472:   %%CITATION = HEP-TH 0106256;%%

1473: 

1474: \bibitem{PaccettiCorreia:2004ri}

1475:   F.~Paccetti Correia, M.~G.~Schmidt and Z.~Tavartkiladze,

1476:   ``Superfield approach to 5D conformal SUGRA and the radion,''

1477:   Nucl.\ Phys.\ B {\bf 709} (2005) 141

1478:   [arXiv:hep-th/0408138] and 

1479:   %%CITATION = HEP-TH 0408138;%%

1480: %\bibitem{Correia:2006pj}

1481: %  F.~P.~Correia, M.~G.~Schmidt and Z.~Tavartkiladze,

1482:   ``4D superfield reduction of 5D orbifold SUGRA and heterotic M-theory,''

1483:   Nucl.\ Phys.\ B {\bf 751} (2006) 222 [arXiv:hep-th/0602173].

1484:   %%CITATION = HEP-TH 0602173;%%

1485: 

1486: \bibitem{ziegler}

1487:   R.~Ziegler, ``The Chern-Simons term in radion mediated SUSY breaking'',

1488:   Diploma thesis, Univ. of Heidelberg, 2006.

1489: 

1490: \bibitem{Barbieri:1985wq}

1491:   R.~Barbieri, E.~Cremmer and S.~Ferrara, ``Flat And Positive Potentials In

1492:   N=1 Supergravity,'' Phys.\ Lett.\ B {\bf 163} (1985) 143.

1493:   %%CITATION = PHLTA,B163,143;%%

1494: 

1495: \bibitem{Hebecker:2001ke}

1496:   A.~Hebecker, ``5D super Yang-Mills theory in 4-D superspace, superfield 

1497:   brane operators, and applications to orbifold GUTs,''

1498:   Nucl.\ Phys.\ B {\bf 632} (2002) 101 [arXiv:hep-ph/0112230].

1499:   %%CITATION = HEP-PH 0112230;%%

1500: 

1501: \bibitem{Kim:2008kn}

1502:   S.~Kim, K.~M.~Lee and S.~Lee,

1503:   ``Dyonic Instantons in 5-dim Yang-Mills Chern-Simons Theories,''

1504:   JHEP {\bf 0808} (2008) 064

1505:   [arXiv:0804.1207 [hep-th]].

1506:   %%CITATION = JHEPA,0808,064;%%

1507: 

1508: \bibitem{Martin:1997ns}

1509:   S.~P.~Martin, ``A supersymmetry primer,'' arXiv:hep-ph/9709356

1510:   (extended version of a contribution to ``Perspectives on supersymmetry''

1511:   by G.L.~Kane (ed.), World Scientific, Singapore 1998).

1512:   %%CITATION = HEP-PH 9709356;%%

1513: 

1514: \bibitem{Luty:1999cz}

1515:   M.~A.~Luty and R.~Sundrum,

1516:   ``Radius stabilization and anomaly-mediated supersymmetry breaking,''

1517:   Phys.\ Rev.\  D {\bf 62} (2000) 035008

1518:   [arXiv:hep-th/9910202].

1519:   %%CITATION = PHRVA,D62,035008;%%

1520: 

1521: \bibitem{Luty:2002hj}

1522:   M.~A.~Luty and N.~Okada,

1523:   ``Almost no-scale supergravity,''

1524:   JHEP {\bf 0304} (2003) 050

1525:   [arXiv:hep-th/0209178].

1526:   %%CITATION = JHEPA,0304,050;%%

1527: 

1528: \bibitem{wb}

1529: J. Wess and J. Bagger, {\it Supersymmetry and Supergravity}, Princeton Univ.

1530: Press, 1983.

1531: 

1532: \bibitem{Hebecker:2004sb}

1533:   A.~Hebecker,

1534:   ``A comment on the cosmological constant problem in spontaneously broken

1535:   supergravity,''

1536:   Phys.\ Lett.\  B {\bf 601} (2004) 197

1537:   [arXiv:hep-th/0407196].

1538:   %%CITATION = PHLTA,B601,197;%%

1539: 

1540: \bibitem{Nakahara:1990th}

1541:   M.~Nakahara,

1542:   ``Geometry, Topology And Physics,''

1543: {\it  Boca Raton, USA: Taylor \& Francis (2003) 573 p}. 

1544: 

1545: \end{thebibliography}

1546: \end{document}

1547: