1: \NeedsTeXFormat{LaTeX2e}
2: \documentclass[fleqn,12pt]{article}
3: \usepackage{cite}
4: \usepackage{amsmath}
5: \usepackage{amssymb}
6: \usepackage{amsfonts}
7: \usepackage{bbm}
8: \usepackage{graphicx}
9: \usepackage{subfigure}
10: \usepackage[small]{caption2}
11: \setcaptionmargin{1cm}
12:
13: \DeclareMathOperator{\tr}{tr}
14: \DeclareMathOperator{\diag}{diag}
15: \DeclareMathOperator{\ad}{ad}
16: \def\D{\mathrm{d}}
17: \def\I{\mathrm{i}}
18: \newcommand{\SimpleRoot}[1]{\alpha_{(#1)}{}}
19: \newcommand{\FundamentalWeight}[1]{\mu^{(#1)}}
20: \newcommand{\CenterObject}[1]{\ensuremath{\vcenter{\hbox{#1}}}}
21: \renewcommand{\thesection}{\arabic{section}}
22: \renewcommand{\thesubsection}{\arabic{section}.\arabic{subsection}}
23: \renewcommand{\thesubsubsection}{(\roman{subsubsection})}
24: \numberwithin{equation}{section}
25: \numberwithin{table}{section}
26: \renewcommand{\thetable}{\arabic{section}.\arabic{table}}
27: \renewcommand{\theequation}{\arabic{section}.\arabic{equation}}
28: \renewcommand{\labelenumi}{(\arabic{enumi})}
29: \unitlength=1mm
30: \allowdisplaybreaks[1]
31: \addtolength\topmargin{-50pt}
32: \addtolength\textheight{105pt}
33: \addtolength\textwidth{60pt}
34: \addtolength\oddsidemargin{-38pt}
35: \setlength{\parindent}{20pt}
36: \setlength{\parskip}{6pt}
37: \frenchspacing
38: \sloppy
39:
40: \begin{document}
41: \thispagestyle{empty}
42: \noindent
43: DESY 03-069 \hspace*{\fill} June 05, 2003\\
44: \vspace*{1.6cm}
45:
46: \begin{center}
47: {\Large\bf Group-Theoretical Aspects of\\[.3cm] Orbifold and Conifold GUTs}
48: \\[2.0cm]
49: {\large Arthur Hebecker and Michael Ratz}\\[.5cm]
50: {\it Deutsches Elektronen-Synchrotron, Notkestrasse 85, D-22603 Hamburg,
51: Germany}
52: \\[1cm]
53:
54: {\bf Abstract}\end{center}
55: \noindent
56: Motivated by the simplicity and direct phenomenological applicability
57: of field-theoretic orbifold constructions in the context of
58: grand unification, we set out to survey the immensely rich
59: group-theoretical possibilities open to this type of model
60: building. In particular, we show how every maximal-rank, regular subgroup
61: of a simple Lie group can be obtained by orbifolding and determine under
62: which conditions rank reduction is possible. We investigate how standard
63: model matter can arise from the higher-dimensional SUSY gauge multiplet.
64: New model building options arise if, giving up the global orbifold
65: construction, generic conical singularities and generic gauge twists
66: associated with these singularities are considered. Viewed from
67: the purely field-theoretic perspective, such models, which one
68: might call conifold GUTs, require only a very mild relaxation of the
69: constraints of orbifold model building. Our most interesting concrete
70: examples include the breaking of E$_7$ to SU(5) and of E$_8$ to
71: SU(4)$\times$SU(2)$\times$SU(2) (with extra factor groups), where three
72: generations of standard model matter come from the gauge sector and the
73: families are interrelated either by SU(3) R-symmetry or by an SU(3)
74: flavour subgroup of the original gauge group.
75:
76: \newpage
77:
78: \section{Introduction}
79: Arguably, the way in which fermion quantum numbers are explained by
80: \(\mathrm{SU}(5)\)-related grand unified theories (GUTs) represents one of
81: the most profound hints at fundamental physics beyond the standard model
82: (SM)~\cite{gg,gfm} (also~\cite{ps}). In this context supersymmetry (SUSY),
83: usually invoked to solve the hierarchy problem and to achieve gauge coupling
84: unification, receives a further and maybe even more fundamental motivation:
85: If the underlying gauge group contains gauge bosons with the quantum numbers
86: of SM matter, SUSY enforces the existence of the corresponding
87: fermions. This is very naturally realized in a higher-dimensional setting,
88: where the extra-dimensional gauge field components and their fermionic
89: partners can be light even though the gauge group is broken at a high scale
90: (see, e.g.,~\cite{Babu:2002ti,wy,gmn,nb}).
91:
92: Thus, we adopt the point of view that, at very high energies, we are
93: faced with a super Yang-Mills (SYM) theory in $d>4$ dimensions which is
94: compactified in such a way that the resulting 4d effective theory has
95: smaller gauge symmetry (ideally that of the SM) and contains the light
96: SM matter and Higgs fields. In the simplest models the compactification
97: space is flat except for a finite number of singularities. Although this
98: situation arises naturally in heterotic string theory~\cite{wit,dhvw} and
99: has thus been extensively studied in string model building, it has only
100: recently been widely recognized that many interesting phenomenological
101: implications do not depend on the underlying quantum gravity model and can
102: be studied directly in higher-dimensional field theory~\cite{kaw}
103: (also~\cite{af,hn,hm,abc,hnos}).
104:
105: In the purely field-theoretic context, one has an enormous freedom in
106: choosing the underlying gauge group, the number of extra dimensions and
107: their geometry, the way in which the compactification reduces the gauge
108: symmetry (e.g., the type of orbifold breaking), the possible extra field
109: content and couplings in the bulk and at the singularities. Although,
110: using all this freedom, realistic models can easily be constructed, there
111: is so far no model which, by its simplicity and direct relation to the
112: observed field content and couplings, appears to be as convincing as, say,
113: the generic SU(5) unification idea. However, we feel that the search for
114: such a model in the framework of higher-dimensional SYM theory is
115: promising and that a thorough understanding of the group-theoretical
116: possibilities of orbifold-breaking (without the restrictions of
117: string theory) will be valuable in this context. The present paper is
118: aimed at the exploration of these possibilities and their application
119: to orbifold GUT model building. In particular, we are interested in methods
120: for breaking larger gauge groups to the SM, in possibilities
121: for rank reduction, and in the derivation of matter fields from the adjoint
122: representation.
123:
124: In Sec.~\ref{gt}, we collect some of the most relevant facts
125: and methods of group theory, which serves in particular to fix our notation
126: and conventions for the rest of the paper.
127:
128: In Sec.~\ref{sg}, we begin by recalling the generic features of field
129: theoretic orbifold models. It is then shown how orbifolding can break a
130: simple Lie group to any of its maximal regular subgroups. This implies, in
131: particular, that any regular subgroup (possibly times extra
132: simple groups and U(1) factors) can be obtained by orbifold-breaking and
133: opens up an enormous variety of model building possibilities.
134:
135: We continue in Sec.~\ref{rr} by exploring rank reduction by non-Abelian
136: orbifolding. We show that simple group factors can always be broken
137: completely. In cases where a maximal subgroup contains an extra U(1) factor,
138: this factor can only be broken under certain conditions. We give a
139: criterion specifying when the extra U(1) cannot be removed. As an interesting
140: observation, we note that under special circumstances rank reduction based
141: on inner automorphisms is also possible on Abelian orbifolds.
142:
143: In Sec.~\ref{coni}, we discuss manifolds with conical singularities which
144: can not be obtained by orbifolding. In particular, such `conifolds' can
145: have conical singularities with arbitrary deficit angle. In addition, we
146: consider the possibility of having Wilson lines with unrestricted values
147: wrapped around the singularities of orbifolds or conifolds. All this
148: gives rise to many new possibilities for gauge symmetry breaking and for the
149: generation of three families of chiral matter from the field content of the
150: SYM theory.
151:
152: Finally, Sec.~\ref{mo} discusses three specific models, one with E$_7$
153: broken to SU(5) and two with E$_8$ broken to SU(4)$\times$SU(2)$\times$SU(2)
154: (with extra factor groups). In all cases, three generations of SM matter
155: come from the gauge sector. In one of the E$_8$ models, the
156: families are interrelated by an SU(3) R-symmetry, while in the two other
157: models an SU(3) flavour subgroup of the original gauge group appears.
158:
159: Sec.~\ref{co} contains our conclusions and outlines future perspectives
160: and open questions.
161:
162: \section{Basics of group theory}\label{gt}
163:
164: This section is not meant as an introduction to group theory, but merely
165: serves to remind the reader of some crucial facts and to fix our notation.
166: Relevant references include the classic papers of Dynkin~\cite{dyn1,dyn2,
167: dyn3} (partially collected in~\cite{dynb}), various textbooks
168: (e.g.,~\cite{gil,Georgi:1999jb,cahn}), and the review article~\cite{sla}.
169:
170: For each finite-dimensional, complex Lie algebra \(\mathfrak{g}\), the
171: maximal Abelian subalgebra \(\mathfrak{h}\), which is unique up to
172: automorphisms, is called Cartan subalgebra.
173: Its dimension defines the rank $r$ of the Lie algebra and
174: its generators will be denoted by \(\{\boldsymbol{H}_i\}_{i=1}^r\).
175: They are orthonormal with respect to the Killing metric, i.e.,
176: they fulfill the relation
177: \begin{equation}
178: \tr (\boldsymbol{H}_i\,\boldsymbol{H}_j)
179: \,=\,\lambda\,\delta_{ij}\;,\label{norm}
180: \end{equation}
181: where the trace is taken in the adjoint representation and
182: \(\lambda\) is some constant.
183:
184: The remaining generators can be chosen such that
185: \begin{equation}
186: [\boldsymbol{H}_i,\boldsymbol{E}_{\alpha}]
187: \,=\,
188: \alpha_i\,\boldsymbol{E}_{\alpha}\;,\label{hec}
189: \end{equation}
190: and are called roots. They are normalized as in Eq.~(\ref{norm}). Each
191: root \(\boldsymbol{E}_{\alpha}\) is determined uniquely by the root
192: vector \(\alpha\), which is an element of an $r$-dimensional Euclidean
193: space, called the root space. The set of all roots will be denoted
194: by \(\Sigma\). The \(\boldsymbol{E}_\alpha\) obey the commutation relations
195: \begin{equation}
196: \left[\boldsymbol{E}_\alpha,\boldsymbol{E}_\beta\right]
197: \,=\,
198: N_{\alpha,\beta}\,\boldsymbol{E}_{\alpha+\beta}\;,
199: \label{eq:DefOfNalphabeta}
200: \end{equation}
201: where the \(N_{\alpha,\beta}\) are normalization constants,
202: and \(N_{\alpha,\beta}=0\) means that \(\alpha+\beta\not\in\Sigma\).
203:
204: We introduce an order in the root space by
205: \begin{equation}
206: \alpha-\beta >0
207: \quad:\Leftrightarrow\quad
208: \text{first non-vanishing component of}\:\alpha-\beta>0
209: \;.
210: \end{equation}
211: Correspondingly, we will call a root `positive' if the first non-vanishing
212: component in the root basis is positive.
213: The smallest \(r\) positive roots are called simple and will be denoted
214: by \(\{\SimpleRoot{i}\}_{i=1}^r\).
215: They are linearly independent, and any root can be expressed by a linear
216: combination
217: \begin{equation}
218: \alpha = \sum\limits_{i=1}^r k^i\,\SimpleRoot{i}
219: \end{equation}
220: with integer coefficients \(k^i\). Motivated by this, a basis
221: \begin{equation}
222: e_{(i)}
223: \,=\,
224: \frac{2}{|\alpha_{(i)}|^2}\SimpleRoot{i}\;,\label{eb}
225: \end{equation}
226: is introduced. The normalization factor will be justified later.
227:
228: In this basis, the Euclidean metric of the root space is characterized by
229: \(g_{ij}=e_{(i)}\cdot e_{(j)}\). It is useful to consider also the vector
230: space dual to the root space which, given the existence of a metric in the
231: root space, can be identified with the root space by the canonical
232: isomorphism. It is spanned by the so-called fundamental weights
233: \(\FundamentalWeight{i}\) (\(1\le i\le r\)) which are defined by
234: \begin{equation}
235: \mu^{(i)}\cdot e_{(j)} \,=\,\delta^i_j\;.
236: \end{equation}
237: The components with respect to the $\mu^{(i)}$ basis are called Dynkin
238: labels. Correspondingly, the $\mu^{(i)}$ are frequently referred to as the
239: Dynkin basis, in which case the $e_{(i)}$ are called the dual basis.
240: The constant \(\lambda\) in Eq.~\eqref{norm}
241: is chosen such that \(|\SimpleRoot{i}|^2=2\) for the longest of the simple
242: roots. Then the normalization factor in Eq.~(\ref{eb}) ensures that the
243: Dynkin label of any weight (weights being the analogues of the vectors
244: $\alpha$ in an arbitrary representation) is integer valued.
245:
246: The Dynkin labels of each simple root are given by the corresponding row
247: of the Cartan-matrix
248: \begin{equation}
249: A_{ij}
250: \,=\,
251: 2\,\frac{\SimpleRoot{i}\cdot\SimpleRoot{j}}{|\SimpleRoot{j}|^2}\,=\,
252: g_{ij}\,\frac{|\SimpleRoot{i}|^2}{2}\;,
253: \end{equation}
254: which encodes the metric of the root space.
255:
256: It is well-known that there exist four infinite series of simple
257: groups \(\mathrm{A}_r\), \(\mathrm{B}_r\), \(\mathrm{C}_r\)
258: and \(\mathrm{D}_r\), corresponding to the classical groups,
259: and the exceptional groups \(\mathrm{G}_2\), \(\mathrm{F}_4\),
260: \(\mathrm{E}_6\), \(\mathrm{E}_7\) and \(\mathrm{E}_8\).
261: The scalar products of the simple roots determine the Dynkin diagrams
262: (cf.\ the captions of Tab.~\ref{tab:DynkinDiagramsOfClassicalGroups}
263: and Tab.~\ref{tab:DynkinDiagramsOfExceptionalGroups}).
264:
265: For later convenience, we introduce the most negative root \(\theta\),
266: which leads us to the extended Dynkin diagrams as listed in
267: Tab.~\ref{tab:DynkinDiagramsOfClassicalGroups} and in
268: Tab.~\ref{tab:DynkinDiagramsOfExceptionalGroups}.
269:
270: \begin{table}[ht]
271: \begin{center}
272: \begin{tabular}{|c|l|l|}
273: \hline
274: \multicolumn{1}{|c}{Name} &
275: \multicolumn{1}{|c}{Real algebra} &
276: \multicolumn{1}{|c|}{Extended Dynkin diagram} \\
277: \hline
278: & & \\[-3mm]
279: \(\mathrm{A}_n \) & \( \mathfrak{su}(n+1)\)
280: & \(\CenterObject{\vphantom{\includegraphics[scale=1.1]
281: {DynkinAnExtendedLabeled.eps}}\includegraphics{DynkinAnExtendedLabeled.eps}
282: }\) \\
283: & & \\
284: \(\mathrm{B}_n \) & \( \mathfrak{so}(2n+1)\)
285: & \(\CenterObject{\vphantom{\includegraphics[scale=1.1]
286: {DynkinBnExtendedLabeled.eps}}
287: \includegraphics{DynkinBnExtendedLabeled.eps}}\) \\
288: & & \\
289: \(\mathrm{C}_n \) & \( \mathfrak{sp}(2n)\)
290: & \(\CenterObject{\vphantom{\includegraphics[scale=1.1]
291: {DynkinCnExtendedLabeled.eps}}
292: \includegraphics{DynkinCnExtendedLabeled.eps}}\) \\
293: & & \\
294: \(\mathrm{D}_n \) & \( \mathfrak{so}(2n)\)
295: &
296: \(\CenterObject{\includegraphics{DynkinDnExtendedLabeled.eps}}\)\\
297: \hline
298: \end{tabular}
299: \caption{The classical Lie algebras and the corresponding
300: extended Dynkin diagrams. The shorter roots are hatched. If the simple
301: roots \(\SimpleRoot{i}\) and \(\SimpleRoot{j}\) enclose
302: an angle of \(90^\circ\), \(120^\circ\) or \(135^\circ\),
303: they are connected by 0, 1 or 2 lines, respectively.}
304: \label{tab:DynkinDiagramsOfClassicalGroups}
305: \end{center}
306: \end{table}
307:
308: \begin{table}[ht*]
309: \begin{center}
310: \begin{tabular}{|c|l|}
311: \hline
312: \multicolumn{1}{|c}{Name} &
313: \multicolumn{1}{|c|}{Extended Dynkin diagram}
314: \\
315: \hline
316: & \\[-0.3mm]
317: $\mathrm{G}_2$ &
318: $\CenterObject{\includegraphics{DynkinG2ExtendedLabeled.eps}}$
319: \\
320: & \\
321: $\mathrm{F}_4$ &
322: $\CenterObject{\includegraphics{DynkinF4ExtendedLabeled.eps}}$
323: \\[1mm]
324: & \\
325: $\mathrm{E}_6$ &
326: $\CenterObject{\includegraphics{DynkinE6ExtendedLabeled.eps}}$
327: \\[1mm]
328: & \\
329: $\mathrm{E}_7$ &
330: $\CenterObject{\includegraphics{DynkinE7ExtendedLabeled.eps}}$
331: \\[1mm]
332: & \\
333: $\mathrm{E}_8$ &
334: $\CenterObject{\includegraphics{DynkinE8ExtendedLabeled.eps}}$
335: \\[1mm]
336: & \\
337: \hline
338: \end{tabular}
339: \end{center}
340: \caption{The five exceptional Lie algebras. In \(\mathrm{G}_2\),
341: the two simple roots enclose \(150^\circ\), which is indicated by
342: a triple line.}
343: \label{tab:DynkinDiagramsOfExceptionalGroups}
344: \end{table}
345:
346:
347: \section{Obtaining all regular subgroups by orbifolding}\label{sg}
348:
349: Orbifold GUTs~\cite{kaw,af,hn,hm,abc,hnos} are based on a gauge theory on
350: $\mathbbm{R}^4\times M$, where $M$ is a manifold with some discrete
351: symmetry group $K$. In addition to the action of $K$ on $M$, an action in
352: internal space can be chosen using a homomorphism from $K$ to the automorphism
353: group of the Lie algebra of the gauge theory. If the classical field space
354: is restricted by the requirement of $K$ invariance, a gauge theory on a
355: manifold with singularities, $M/K$, results in general. We assume that
356: $M/K$, though not necessarily $M$, is compact. At the singularities, which
357: correspond to the fixed points of the space-time action of $K$, the gauge
358: symmetry may be restricted (orbifold breaking). An early review of the
359: structure of such models is contained in~\cite{hm1} (for more recent reviews
360: see, e.g.,~\cite{Haba:2002py,orb}).
361:
362: One of the main features of orbifold GUTs is the possibility of breaking
363: a gauge group without the use of Higgs fields. The orbifold field theory
364: possesses the full (unified) gauge symmetry everywhere except for certain
365: fixed points. Although this fixed-point breaking is `hard', in the sense
366: that the action does not possess the full gauge symmetry, gauge coupling
367: unification is not lost due to the numerical dominance of the bulk.
368: Furthermore, it is attractive for model building purposes
369: that the symmetry -- and hence the field content -- is characterized by
370: different groups at different geometric locations, such as the various
371: fixed points and the bulk.
372:
373: In this paper, we focus on inner-automorphism breaking, i.e., a
374: homomorphism from $K$ to the gauge group $G$ together with the adjoint
375: action of $G$ on itself is used to define the transformation of gauge
376: fields under $K$.
377: Only gauge fields invariant under $K$ have zero modes. The corresponding
378: generators define the symmetry of the low-energy effective theory, which is
379: a subgroup of $G$. We will assume that \(G\) is simple since it is
380: straightforward to extend our analysis to the product of simple groups and
381: \(\mathrm{U}(1)\) factors.
382:
383: To discuss the breaking in more detail, consider a group element $P$ which
384: is the image of some element of $K$. Any $P\in G$ can be written as an
385: exponential of some Lie algebra element and is therefore contained in some
386: U(1) subgroup of $G$. Constructing a maximal torus starting from this U(1)
387: and using the fact that the maximal torus in a compact Lie group is unique
388: up to isomorphism~\cite{Fegan:1991jb}, it becomes clear that one can always
389: write
390: \begin{equation}
391: P
392: \,=\,
393: \exp(-2\pi\I\, V\cdot \boldsymbol{H})\;,
394: \end{equation}
395: with some real vector $V$. Hence, the action of the gauge twist on the
396: Lie algebra is given by
397: \begin{subequations}\label{eq:OnlyOneTwist}
398: \begin{eqnarray}
399: P\,\boldsymbol{E}_\alpha\,P^{-1}
400: & = &
401: \exp(-2\pi\I\,\alpha\cdot V)\,\boldsymbol{E}_\alpha
402: \;,\label{eq:AbelianTwists4Ealpha}\\*
403: P\,\boldsymbol{H}_i\,P^{-1}
404: & = &
405: \boldsymbol{H}_i\;.
406: \end{eqnarray}
407: \end{subequations}
408:
409: We can also choose to write \(P=\exp(- 2\pi\I\,\xi\,\boldsymbol{T})\),
410: where \(\boldsymbol{T}\) is a normalized Lie algebra element,
411: \(\xi\in\mathbbm{R}\), and \(\xi\,\boldsymbol{T}=V\cdot \boldsymbol{H}\).
412: For generic $\xi$, $P$ commutes with precisely those Lie algebra elements
413: with which $\boldsymbol{T}$ commutes. Thus, the breaking is the same as
414: would follow from a Higgs VEV in the adjoint representation.
415:
416: However, it is clear from Eq.~(\ref{eq:AbelianTwists4Ealpha}) that, for
417: certain values of $\xi$, some of the \(\boldsymbol{E}_\alpha\) may pick
418: up phases which are an integer multiple of \(2\pi\) and are thus left
419: invariant. In this case, the surviving subgroup is larger than the one
420: obtained from an adjoint VEV proportional to $\boldsymbol{T}$. This
421: possibility is of particular interest since, in certain cases, such as the
422: breaking of \(\mathrm{SO}(10)\) to
423: \(\mathrm{SU}(4)\times\mathrm{SU}(2)\times\mathrm{SU}(2)\), the relevant
424: subgroup can not be realized by using Higgs VEVs in the adjoint or any smaller
425: representation.
426:
427: \subsection{Orbifold-breaking to any maximal regular subgroup}
428: \label{o2ms}
429:
430: We now show that, given a simple group \(G\) and a maximal regular\footnote{In
431: this paper, we concentrate on the breaking to regular subgroups. For a
432: discussion of non-regular embeddings (in the string theory context) see,
433: e.g.,\cite{dmr}.} subgroup \(H\), there exists a \(P\in G\) such that
434: \begin{equation}
435: H \,=\, \{g\in G;\; P\,g\,P^{-1}=g\}\,.
436: \end{equation}
437: In other words, every maximal regular subgroup can be generated by an
438: orbifold twist.
439:
440: In order to prove this statement, we first recall Dynkin's prescription
441: for generating semi-simple subgroups. It starts with the Dynkin diagram,
442: extends it by adding the most negative root, and then removes one of the
443: simple roots, the resulting Dynkin diagram being that of a semi-simple
444: subgroup. As demonstrated in~\cite{dyn3} (cf. Theorem 5.3), any
445: maximal-rank, semi-simple subgroup of a given group can be obtained by
446: successive application of this prescription. Maximal subgroups can always
447: be obtained in the first application.
448:
449: To implement Dynkin's prescription and remove the simple root $\alpha_{(i)}$,
450: one can use the fundamental weight $\mu^{(i)}$ and choose
451: \begin{equation}
452: P \,=\, \exp\left(\frac{2\pi \I}{n}\,\frac{2}{|\SimpleRoot{i}|^2}\,
453: \FundamentalWeight{i}\cdot\boldsymbol{H}
454: \right)\;.
455: \label{eq:FundamentalZnTwist}
456: \end{equation}
457: Obviously, $P$ commutes with all simple roots $\boldsymbol{E}_{\alpha_{(j)}}$
458: where $j\neq i$.
459: To discuss the roots $\alpha_{(i)}$ and $\theta$, recall first that
460: \begin{equation}
461: \theta = -\sum_{k=1}^r c_k\,\SimpleRoot{k}
462: \end{equation}
463: with the \(c_k\) being known as Coxeter labels. They can be read off from
464: Tab.~\ref{tab:HighestWeightsOfAdjoints}. Such group-theoretical methods were
465: used in~\cite{kkkot} in the context of E$_8$ breaking in string theory.
466:
467: \begin{table}[!ht!]
468: \begin{center}
469: \begin{tabular}{|c|l|l|}
470: \hline
471: Group & Dynkin labels of \(\Lambda_\mathrm{ad}\) & Coxeter labels\\
472: \hline
473: \(\mathrm{A}_n\simeq\mathrm{SU}(n+1)\) & \((1,0,\dots 0,1)\) &
474: \((1,1,\dots 1)\)\\
475: \(\mathrm{B}_n\simeq\mathrm{SO}(2n+1)\) & \((0,1,0,\dots)\) &
476: \((1,2,2,\dots 2,2)\)\\
477: \(\mathrm{C}_n\simeq\mathrm{Sp}(2n)\) & \((2,0,0,\dots)\) &
478: \((2,2,\dots 2,1)\)\\
479: \(\mathrm{D}_n\simeq\mathrm{SO}(2n)\) & \((0,1,0,\dots)\) &
480: \((1,2,2,\dots 2,1,1)\)\\
481: \(\mathrm{G}_2\) & \((1,0)\) & \((2,3)\)\\
482: \(\mathrm{F}_4\) & \((1,0,0,0)\) & \((2,3,4,2)\)\\
483: \(\mathrm{E}_6\) & \((0,0,0,0,0,1)\) & \((1,2,3,2,1,2)\)\\
484: \(\mathrm{E}_7\) & \((1,0,0,0,0,0,0)\) & \((2, 3, 4, 3, 2, 1, 2)\)\\
485: \(\mathrm{E}_8\) & \((0,0,0,0,0,0,1,0)\) & \((2, 4, 6, 5, 4, 3, 2, 3)\)\\
486: \hline
487: \end{tabular}
488: \end{center}
489: \caption{Highest weights of the adjoint representations,
490: denoted by \(\Lambda_\mathrm{ad}\),
491: of the simple groups in the Dynkin basis and their
492: Coxeter labels.}
493: \label{tab:HighestWeightsOfAdjoints}
494: \end{table}
495:
496: Thus, we can write the orbifold action on the two roots
497: \(\boldsymbol{E}_{\SimpleRoot{i}}\) and \(\boldsymbol{E}_\theta\) as
498: \begin{subequations}
499: \begin{eqnarray}
500: P\,\boldsymbol{E}_{\SimpleRoot{i}}\,P^{-1}
501: & = &
502: e^{ 2\pi\I/n}\,\boldsymbol{E}_{\SimpleRoot{i}}
503: \;,\\*
504: P\,\boldsymbol{E}_{\theta}\,P^{-1}
505: & = &
506: e^{- 2\pi\I\,c_i/n}\,\boldsymbol{E}_{\theta}
507: \;,
508: \end{eqnarray}
509: \end{subequations}
510: which shows that, for \(\boldsymbol{E}_\theta\) to be invariant and
511: \(\boldsymbol{E}_{\SimpleRoot{i}}\) to be projected out, we need \(c_i\ne 1\).
512: Using Tab.~\ref{tab:HighestWeightsOfAdjoints} and the corresponding Dynkin
513: diagrams, it is easy to convince oneself that $c_i=1$ occurs only for
514: those $i$ where the Dynkin-prescription with removal of $\alpha_{(i)}$
515: returns the original diagram. Thus, all non-trivial subgroups accessible by
516: the Dynkin-prescription can be obtained by $\mathbbm{Z}_n$ orbifolding with
517: $n=c_i$.
518:
519: An interesting and subtle observation can be made in those cases where
520: $c_i$ is not prime (only $c_i=4$ and $c_i=6$ occur). If $c_i=n=m\cdot k$,
521: a \(\mathbbm{Z}_m\) twist generated by $P^k$ is sufficient to project out
522: $\boldsymbol{E}_{\SimpleRoot{i}}$ while keeping $\boldsymbol{E}_\theta$,
523: yet the surviving subgroup is larger than for the corresponding
524: \(\mathbbm{Z}_n\) twist \(P\) and its Dynkin diagram is not the one obtained
525: by Dynkin's prescription. These are the famous five cases where Dynkin's
526: prescription produces a subgroup that is not maximal~\cite{Golubitsky:1971ex}.
527: They occur when removing the 3rd root of \(\mathrm{F}_4\), the 3rd root of
528: \(\mathrm{E}_7\), and the 2nd, 3rd or 5th root of \(\mathrm{E}_8\).
529:
530: It is easy to see that for $c_i$ prime the produced subgroup is maximal.
531: Indeed, the roots of $G$ which are not roots of the subgroup $H$ can be
532: classified according to their `level' relative to $\alpha_{(i)}$, i.e.,
533: according to the coefficient of $\alpha_{(i)}$ in their decomposition
534: in terms of simple roots. If a subgroup $H'$ with $H\subset H'\subset G$
535: exists, one of the levels below $c_i$ (which is the highest level) and
536: above 1 must be occupied (i.e., its roots belong to $H'$). Let $\ell$ be the
537: smallest of those levels. All multiples of $\ell$ are also occupied and, since
538: $c_i$ is not a multiple, the difference between $c_i$ and one of those
539: multiples must be smaller than $\ell$. However, by the way in which the
540: commutation relations
541: are realized in root space, the level corresponding to this difference must
542: also be occupied. This is in contradiction to $\ell$ being the smallest
543: occupied level in $H'$.
544:
545: Having dealt with all semi-simple maximal subgroups, we now come to maximal
546: subgroups containing $\mathrm{U}(1)$ factors. Given a maximal subgroup $H$
547: with $\mathrm{U}(1)$ factor, i.e., \(G\supset H\times\mathrm{U}(1)\), we
548: can always break to a subgroup $H'$ by an adjoint VEV along this
549: $\mathrm{U}(1)$ direction
550: or a corresponding orbifold twist. It is obvious that $H\subset H'$ since,
551: by the definition of $H$, all its elements commute with the generator of
552: the above $\mathrm{U}(1)$. Thus, $H=H'$ and our analysis of orbifold breaking
553: to all maximal-rank regular subgroups is complete. The maximal regular
554: subgroups and the corresponding twists are listed in
555: Tab.~\ref{tab:OrbifoldTwists} in App.~\ref{app:OrbifoldTwists}.
556: We would also like to mention that the maximal subgroups with
557: \(\mathrm{U}(1)\) factors can be obtained by removing one node of the
558: original Dynkin diagram which carries Coxeter label 1, and adding
559: the \(\mathrm{U}(1)\) factor.
560:
561: Now that it is clear how a given maximal regular subgroup can be generated
562: by an orbifold twist, we can take the opposite point of view and ask to
563: which subgroups an arbitrary given gauge twist $P=\exp{(-2\pi\I\,\xi\,
564: \boldsymbol{T}})$ can lead. Since an adjoint VEV proportional to
565: $\boldsymbol{T}$ breaks to a maximal rank subgroup $H\times$U(1), where the
566: \(\mathrm{U}(1)\) is generated by $\boldsymbol{T}$, we can classify all
567: $\boldsymbol{T}$'s by such maximal subgroups. These are given in various tables
568: (see, in particular,~\cite{sla}) together with the branching rules for the
569: adjoint representation
570: \begin{equation}
571: \ad G
572: \,\to\,
573: \ad H \oplus \boldsymbol{1}(0) \oplus \boldsymbol{R}_1(q_1) \oplus
574: \boldsymbol{R}_2(q_2)
575: \oplus\dots\;.
576: \end{equation}
577: Here the \(\boldsymbol{R}_i\) are representations under \(H\) and \(q_i\)
578: the corresponding \(\mathrm{U}(1)\) charges. Under the gauge twist,
579: the \(\boldsymbol{R}_i\) transform as \(\boldsymbol{R}_i\to e^{2\pi\I\,\xi
580: q_i}\,\boldsymbol{R}_i\). This allows us to determine which particular sets
581: of generators $\boldsymbol{R}_i$ survive for specific values of $\xi$, i.e.,
582: to identify those \(\boldsymbol{R}_i\) for which $\xi q_i=0\mod \mathbbm{Z}$.
583: Together with the generators
584: of $H\times$U(1), they form the Lie algebra of the new surviving subgroup
585: $H'\supset H\times$U(1). Thus, by analyzing all subgroups $H\times$U(1) and
586: all values of $\xi$, our classification is complete.
587:
588: Finally, we would like to comment on the minimal order of the twist
589: required to achieve the breaking \(G\to H\). A very useful approximate
590: rule is that under a \(\mathbbm{Z}_n\) twist
591: \begin{equation}
592: \dim H
593: \,\gtrsim\,
594: r + \frac{\dim G - r}{n}
595: \;.
596: \end{equation}
597: The reason is that the \(r\) Cartan generators survive the twist anyway,
598: and the phases of the roots are proportional to a level relative to a
599: simple root \(\SimpleRoot{i}\), or linear combination of such levels.
600: Due to the symmetries of the root lattice, the phases are therefore
601: almost evenly distributed among \(\{0,2\pi/n, 4\pi/n,\dots (n-1)2\pi/n\}\)
602: where an excess at 0 is possible if the twist acts trivially on a certain
603: part of the algebra. An inspection of Tab.~\ref{tab:OrbifoldTwists} confirms
604: our rule which becomes the more accurate the larger the group is.
605:
606:
607: \subsection{Some examples from the series $\boldsymbol{\mathrm{SO}(10)
608: \subset\mathrm{E}_6\subset\mathrm{E}_7\subset\mathrm{E}_8}$}
609:
610: At this point, some examples are in order.
611: Let us start with the \(\mathrm{SO}(10)\) GUT which contains
612: the Georgi-Glashow group
613: \(G_\mathrm{GG}=\mathrm{SU}(5)\otimes\mathrm{U}(1)\)
614: and the Pati-Salam group
615: \(G_\mathrm{PS}=\mathrm{SU}(4)\times\mathrm{SU}(2)\times\mathrm{SU}(2)\)
616: as subgroups. These properties are nicely illustrated by using
617: Dynkin's prescription:
618: \begin{figure}[!ht]
619: \begin{center}
620: \subfigure[Extended Dynkin diagram of \(\mathrm{SO}(10)\).]
621: {\begin{tabular}{c}
622: \CenterObject{\includegraphics{DynkinD5ExtendedLabeled.eps}}\\
623: {}
624: \end{tabular}}
625: \hfil
626: \subfigure[Breaking to $G_\mathrm{PS}$.]
627: {\begin{tabular}{c}
628: \CenterObject{\includegraphics{DynkinD5toPatiSalam.eps}}\\
629: {}
630: \end{tabular}}
631: \hfil
632: \subfigure[Breaking to $G_\mathrm{GG}$.]
633: {\begin{tabular}{c}
634: \CenterObject{\includegraphics{DynkinD5toGeorgiGlashow.eps}}\\
635: {}
636: \end{tabular}}
637: \end{center}
638: \caption{The breaking to the Pati-Salam and Georgi-Glashow subgroups
639: of \(\mathrm{SO}(10)\) can be illustrated by removing the
640: \(\SimpleRoot{3}\) (or \(\SimpleRoot{2}\)) node of the extended Dynkin
641: diagram (a) as shown in (b) or by removing \(\SimpleRoot{5}\)
642: (or \(\SimpleRoot{4}\)) as shown in (c).}
643: \label{fig:ExtendedDynkinDiagramOfSO10}
644: \end{figure}
645: Starting from the extended Dynkin diagram
646: (cf.~Fig.~\ref{fig:ExtendedDynkinDiagramOfSO10}), the diagram of
647: \(\mathrm{G}_\mathrm{PS}\) is obtained by deleting the third (or second)
648: node. Deleting the fourth (or fifth) node of the original diagram,
649: we arrive at \(G_\mathrm{GG}\). According to Sec.~\ref{o2ms}, twists
650: which break to \(G_\mathrm{GG}\) and \(\mathrm{G}_\mathrm{PS}\),
651: respectively, can be written as
652: \begin{subequations}\label{eq:PGGandPPS4SO10}
653: \begin{eqnarray}
654: P_\mathrm{PS}
655: & = &
656: \exp(\pi\I\,\FundamentalWeight{3}\cdot\boldsymbol{H})
657: \;,\\*
658: P_\mathrm{GG}
659: & = &
660: \exp(\pi\I\,\FundamentalWeight{4}\cdot\boldsymbol{H})
661: \;,
662: \end{eqnarray}
663: \end{subequations}
664: where we exploited the fact that \(|\SimpleRoot{i}|^2=2\) in
665: simply-laced groups.
666:
667: In \cite{abc,hnos}, it was shown that by identifying these two twists
668: as generators of \(\mathbbm{Z}_2\times\mathbbm{Z}_2'\), the gauge symmetry
669: on a \(\mathbbm{T}^2/(\mathbbm{Z}_2\times\mathbbm{Z}_2')\) orbifold is
670: reduced to \(G_\mathrm{SM}'=\mathrm{SU}(3)\times\mathrm{SU}(2)\times
671: \mathrm{U}(1)_\mathrm{Y}\times\mathrm{U}(1)_\chi\subset\mathrm{SO}
672: (10)\). The resulting geometry can be visualized as a `pillow' with the
673: corners corresponding to the fixed points.
674:
675: The relevant group theory can be understood as follows:
676: \(\FundamentalWeight{3}\cdot\boldsymbol{H}\) and
677: \(\FundamentalWeight{4}\cdot\boldsymbol{H}\) are the \(\mathrm{U}(1)\)
678: generators appearing in \(G_\mathrm{SM}'\). The corresponding decomposition
679: of the adjoint representation of \(\mathrm{SO}(10)\) reads
680: \begin{eqnarray}
681: \boldsymbol{45}
682: & \to &
683: (\boldsymbol{1},\boldsymbol{1})_{(6,4)}
684: \oplus(\overline{\boldsymbol{3}},\boldsymbol{1})_{(-4,4)}
685: \oplus(\boldsymbol{3},\boldsymbol{2})_{(1,4)}
686: \oplus(\boldsymbol{3},\boldsymbol{2})_{(-5,0)}
687: \nonumber\\*
688: & & {}
689: \oplus(\boldsymbol{1},\boldsymbol{1})_{(-6,-4)}
690: \oplus(\boldsymbol{3},\boldsymbol{1})_{(4,-4)}
691: \oplus(\overline{\boldsymbol{3}},\boldsymbol{2})_{(-1,-4)}
692: \oplus(\overline{\boldsymbol{3}},\boldsymbol{2})_{(5,0)}
693: \nonumber\\
694: & & {}
695: \oplus(\boldsymbol{8},\boldsymbol{1})_{(0,0)}
696: \oplus(\boldsymbol{1},\boldsymbol{3})_{(0,0)}
697: \oplus(\boldsymbol{1},\boldsymbol{1})_{(0,0)}
698: \oplus(\boldsymbol{1},\boldsymbol{1})_{(0,0)}
699: \;,\label{eq:DecompositionOf45}
700: \end{eqnarray}
701: where the \(\mathrm{SU}(3)\times\mathrm{SU}(2)\) representations are given
702: in boldface and the U(1) charges \((q_Y,q_\chi)\) appear as index.
703: The twist which is responsible for this breaking is generated
704: by a linear combination of the generators of the two \(\mathrm{U}(1)\)s,
705: and rotates the charged representations by a phase \(2\pi\,(y\,q_Y +
706: \chi\,q_\chi)\). For some combinations of \(\chi\) and \(y\), the orbifold
707: breaking preserves a larger symmetry than adjoint breaking. For example,
708: if \((\boldsymbol{1},\boldsymbol{1})_{(6,4)}\), \((\boldsymbol{1},
709: \boldsymbol{1})_{(-6,-4)}\), \((\overline{\boldsymbol{3}},
710: \boldsymbol{1})_{(-4,4)}\) and \((\boldsymbol{3},\boldsymbol{1})_{(4,-4)}\)
711: survive (e.g., by taking \(\chi=0\) and \(y=1/2\)), the resulting gauge group
712: is \(G_\mathrm{PS}\). If, on the other hand,
713: \((\boldsymbol{3},\boldsymbol{2})_{(-5,0)}\) and
714: \((\overline{\boldsymbol{3}},\boldsymbol{2})_{(5,0)}\) survive (e.g., by
715: taking \(\chi=1/8\) and \(y=0\)), the resulting gauge group is
716: \(G_\mathrm{GG}\). It is then clear that \(G_\mathrm{SM}'\) results as
717: an intersection of gauge fields surviving \(P_\mathrm{PS}\) and
718: \(P_\mathrm{GG}\). The breaking to \(G_\mathrm{SM}'\) can also
719: be realized on a single \(\mathbbm{Z}_4\) fixed point, e.g., by
720: using \(\chi=1/16\) and \(y=1/4\).
721:
722: As a side-remark, let us restate the above discussion in terms of matrices:
723: Consider the adjoint VEV
724: \begin{equation}\label{eq:GeorgisVev}
725: v
726: \,=\,
727: \diag(a,a,a,b,b)\otimes
728: \left(\begin{array}{cc}0 & 1 \\ -1 & 0\end{array}\right)
729: \;,
730: \end{equation}
731: which breaks \(\mathrm{SO}(10)\) to \(G_\mathrm{SM}'\) \cite{Georgi:1999jb}.
732: For the special case \(a=\pm b\), the remaining symmetry is larger and equal
733: to \(G_\mathrm{GG}\). Alternatively, these breakings can be realized by
734: a gauge twist \(P=\exp[2\pi\I\,v]\) at an orbifold fixed point. In this
735: case, taking \(a=0\) and \(b=1/2\) yields \(P_\mathrm{PS}\) and \(a=b=1/4\)
736: yields \(P_\mathrm{GG}\).
737:
738: Let us now turn to the task of extending the `pillow' of Asaka,
739: Buchm\"uller and Covi~\cite{abc1} along the chain of exceptional groups
740: \(\mathrm{SO}(10)\subset\mathrm{E}_6\subset\mathrm{E}_7\subset\mathrm{E}_8\).
741: A related discussion has already appeared in \cite{Haba:2002vc}. However,
742: as will become clear below, we disagree with some of the results of that
743: paper.
744:
745: The obvious generalizations of
746: \eqref{eq:PGGandPPS4SO10} for the exceptional groups read
747: \begin{subequations}
748: \begin{eqnarray}
749: P_\mathrm{GG}^{(r)}
750: & := &
751: \exp(\pi\I\,\FundamentalWeight{r-1}\cdot\boldsymbol{H})
752: \;,\\*
753: P_\mathrm{PS}^{(r)}
754: & := &
755: \exp(\pi\I\,\FundamentalWeight{r-2}\cdot\boldsymbol{H})
756: \;.
757: \end{eqnarray}
758: \end{subequations}
759: In other words, the generalizations of \(P_\mathrm{GG}\) and \(P_\mathrm{PS}\)
760: to higher groups along the above chain remove the nodes \(\SimpleRoot{r-1}\)
761: and \(\SimpleRoot{r-2}\) respectively. This is illustrated in
762: Tab.~\ref{tab:GGPS4Es}.
763: \begin{table}[!ht]
764: \[
765: \begin{array}{|l|l|}
766: \hline
767: \CenterObject{\includegraphics{DynkinD5toGGPS.eps}}
768: &
769: \begin{array}{l}
770: \begin{array}{rcl}
771: \mathrm{SO}(10)
772: & \xrightarrow{P_\mathrm{GG}\phantom{\big|}} &
773: \mathrm{SU}(5)\times\mathrm{U}(1)\\
774: \mathrm{SO}(10)
775: & \xrightarrow{P_\mathrm{PS}} &
776: \mathrm{SU}(4)\times\mathrm{SU}(2)\times\mathrm{SU}(2)\\
777: \mathrm{SO}(10)
778: & \xrightarrow{P_\mathrm{GG}\cdot P_\mathrm{PS}} &
779: \mathrm{SU}(5)'\times\mathrm{U}(1)'
780: \end{array}\\
781: G_\mathrm{GG}\cap G_\mathrm{PS}
782: \, = \,
783: \mathrm{SU}(3)\times\mathrm{SU}(2)\times\mathrm{U}(1)^2
784: \phantom{\Big|}
785: \end{array}
786: \\
787: \hline
788: \CenterObject{\includegraphics{DynkinE6toGGPS.eps}}
789: &
790: \begin{array}{l}
791: \begin{array}{rcl}
792: \mathrm{E}_6
793: & \xrightarrow{P_\mathrm{GG}^{(6)}\phantom{\big|}} &
794: \mathrm{SO}(10)\times\mathrm{U}(1)\\
795: \mathrm{E}_6
796: & \xrightarrow{P_\mathrm{PS}^{(6)}} &
797: \mathrm{SU}(6)\times\mathrm{SU}(2)\\
798: \mathrm{E}_6
799: & \xrightarrow{P_\mathrm{GG}^{(6)}\cdot P_\mathrm{PS}^{(6)}} &
800: \mathrm{SO}(10)'\times\mathrm{U}(1)'
801: \end{array}\\
802: G_\mathrm{GG}^{(6)}\cap G_\mathrm{PS}^{(6)}
803: \,= \,
804: \mathrm{SU}(5)\times\mathrm{U}(1)^2 \phantom{\Big|}
805: \end{array}
806: \\
807: \hline
808: \CenterObject{\includegraphics{DynkinE7toGGPS.eps}}
809: &
810: \begin{array}{l}
811: \begin{array}{rcl}
812: \mathrm{E}_7
813: & \xrightarrow{P_\mathrm{GG}^{(7)}\phantom{\big|}} &
814: \mathrm{E}_6\times\mathrm{U}(1)\\
815: \mathrm{E}_7
816: & \xrightarrow{P_\mathrm{PS}^{(7)}} &
817: \mathrm{SO}(12)\times\mathrm{SU}(2)\\
818: \mathrm{E}_7
819: & \xrightarrow{P_\mathrm{GG}^{(7)}\cdot P_\mathrm{PS}^{(7)}} &
820: \mathrm{E}_6'\times\mathrm{U}(1)'
821: \end{array}\\
822: G_\mathrm{GG}^{(7)}\cap G_\mathrm{PS}^{(7)}
823: \, = \,
824: \mathrm{SO}(10)\times\mathrm{U}(1)^2 \phantom{\Big|}
825: \end{array}
826: \\
827: \hline
828: \CenterObject{\includegraphics{DynkinE8toGGPS.eps}}
829: &
830: \begin{array}{l}
831: \begin{array}{rcl}
832: \mathrm{E}_8
833: & \xrightarrow{P_\mathrm{GG}^{(8)}\phantom{\big|}} &
834: \mathrm{E}_7\times\mathrm{SU}(2)\\
835: \mathrm{E}_8
836: & \xrightarrow{P_\mathrm{PS}^{(8)}} &
837: \mathrm{E}_7'\times\mathrm{SU}(2)'\\
838: \mathrm{E}_8
839: & \xrightarrow{P_\mathrm{GG}^{(8)}\cdot P_\mathrm{PS}^{(8)}} &
840: \mathrm{E}_7^{\prime\prime}\times\mathrm{SU}(2)^{\prime\prime}
841: \end{array}\\
842: G_\mathrm{GG}^{(8)}\cap G_\mathrm{PS}^{(8)}
843: \, = \,
844: \mathrm{E}_6\times\mathrm{U}(1)^2 \phantom{\Big|}
845: \end{array}
846: \\
847: \hline
848: \end{array}
849: \]
850: \caption{Breaking patterns to the Georgi-Glashow and Pati-Salam like
851: subgroups in \(\mathrm{SO}(10)\) and the three exceptional groups
852: \(\mathrm{E}_6\), \(\mathrm{E}_7\) and \(\mathrm{E}_8\).}
853: \label{tab:GGPS4Es}
854: \end{table}
855:
856: The breaking patterns of \(\mathrm{SO}(10)\), \(\mathrm{E}_6\) and
857: \(\mathrm{E}_7\) are easily determined by the use of Dynkin's prescription
858: because the Coxeter label corresponding to the nodes removed by
859: \(P_\mathrm{GG}^{(r)}\) and \(P_\mathrm{PS}^{(r)}\) are 1 and 2, respectively.
860: In the case of \(\mathrm{E}_8\), it is also easy to see that
861: \(P_\mathrm{GG}^{(8)}\) breaks to \(\mathrm{E}_7\times\mathrm{SU}(2)\).
862: For \(P_\mathrm{PS}^{(8)}\), the pattern is not so obvious: Since the
863: 6th Coxeter label is 3 (cf.\ Tab.~\ref{tab:HighestWeightsOfAdjoints}),
864: and we use a \(\mathbbm{Z}_2\) twist, the second level in terms of
865: \(\SimpleRoot{6}\) survives \(P_\mathrm{PS}^{(8)}\), but
866: \(\theta\) is projected out. We see that the subgroup must contain
867: \(\mathrm{E}_6\) and \(\mathrm{SU}(2)\), and it can not be
868: \(\mathrm{E}_6\times\mathrm{SU}(3)\) because this is not a symmetric
869: subgroup of \(\mathrm{E}_8\). Hence, it must be also
870: \(\mathrm{E}_7\times\mathrm{SU}(2)\).\footnote{This
871: is in contradiction to the breaking pattern given
872: in~\cite{Haba:2002vc}. Assigning negative parity
873: to \((\boldsymbol{27},\boldsymbol{3})\) and
874: \((\overline{\boldsymbol{27}},\overline{\boldsymbol{3}})\) is inconsistent
875: since, as can be seen from the commutator
876: \([(\boldsymbol{27},\boldsymbol{3}),(\boldsymbol{27},\boldsymbol{3})]
877: \subset (\overline{\boldsymbol{27}},\overline{\boldsymbol{3}})\),
878: it does not correspond to an algebra automorphism. This commutator does
879: not vanish since \((\boldsymbol{27},\boldsymbol{3})\oplus
880: (\overline{\boldsymbol{27}},\overline{\boldsymbol{3}})\) contains two
881: positive levels with respect to \(\SimpleRoot{6}\), linked
882: by the raising operator \(\boldsymbol{E}_{\SimpleRoot{6}}\).}
883: We checked this statement by using a computer algebra system.
884: In \cite{hnos}, an interesting property of the \(\mathrm{SO}(10)\)
885: twists was pointed out: \(P_\mathrm{GG}'=P_\mathrm{GG}\cdot P_\mathrm{PS}\)
886: breaks to a different \(\mathrm{SU}(5)\times\mathrm{U}(1)\) subgroup of
887: \(\mathrm{SO}(10)\), where the simple factor is often called `flipped
888: \(\mathrm{SU}(5)\)'. This property is maintained for all three exceptional
889: groups: \(P_\mathrm{GG}^{(r)}\cdot P_\mathrm{PS}^{(r)}=P_\mathrm{GG}^{(r)\,
890: \prime}\). Here \(P_\mathrm{GG}^{(r)\,\prime}\) breaks to a subgroup linked
891: by an inner automorphism to the subgroup left invariant by
892: \(P_\mathrm{GG}^{(r)}\). The reason is that \(P_\mathrm{GG}^{(r)\,\prime}=
893: \exp[\pi\I\,(\FundamentalWeight{r-1}
894: +\FundamentalWeight{r-2})\cdot\boldsymbol{H}]\) commutes with
895: \(\boldsymbol{E}_{\SimpleRoot{r-1}+\SimpleRoot{r-2}}\) which then
896: becomes a simple root of the subgroup, and projects out \(\boldsymbol{E}_{\theta}\).
897: This root encloses an angle of \(120^\circ\) with \(\SimpleRoot{r-3}\)
898: so that the resulting Dynkin diagram coincides with the one obtained
899: by employing \(P_\mathrm{GG}\). In the simple root system arising from
900: the substitution \((\SimpleRoot{r-2},\SimpleRoot{r-1})\to
901: (\SimpleRoot{r-2}+\SimpleRoot{r-1},-\SimpleRoot{r-1})\),
902: \(P_\mathrm{GG}^{(r)\,\prime}\) acts in the same way as
903: \(P_\mathrm{GG}^{(r)}\) in the original root system.
904:
905:
906: \section{Rank reduction and non-Abelian twists}\label{rr}
907:
908: It is obvious from the discussion so far that using only one
909: inner-automorphism orbifold twist can never result in rank reduction.
910: We therefore investigate the possibilities which arise when
911: two (or more) twists are applied. Rank reduction of the gauge group was
912: proposed in the context of string theory in \cite{Ibanez:1987xa}.
913: Here, we will discuss this issue in the context of field theory,
914: where one has fewer group-theoretic and geometric constraints.
915:
916: We assume that we have an additional orbifold symmetry,
917: \begin{equation}\label{eq:NonAbelianTwist}
918: P'
919: \,=\,
920: \exp \left[-2\pi\I \,\xi\,(\boldsymbol{E}_\beta+\boldsymbol{E}_{-\beta})
921: \right]
922: \quad\text{or}\quad
923: P'
924: \,=\,
925: \exp \left[2\pi\,\xi \,(\boldsymbol{E}_\beta-\boldsymbol{E}_{-\beta})\right]
926: \;,
927: \end{equation}
928: where \(\I(\boldsymbol{E}_\beta+\boldsymbol{E}_{-\beta})\) and
929: \(\boldsymbol{E}_\beta-\boldsymbol{E}_{-\beta}\) are real generators
930: outside the Cartan subalgebra. For simplicity, let us focus on the case
931: that \(\beta\) is a simple root, i.e.,
932: \begin{equation}
933: P'_j
934: \,=\,
935: \exp \left[-2\pi\I \,\xi\,
936: (\boldsymbol{E}_{\SimpleRoot{j}}+\boldsymbol{E}_{-\SimpleRoot{j}})
937: \right]
938: \;.
939: \end{equation}
940: Then the raising and lowering operators
941: \(\boldsymbol{E}_{\pm\SimpleRoot{j}}\) form an \(\mathrm{SU}(2)\)
942: group together with
943: \(\boldsymbol{h}=\SimpleRoot{j}\cdot \boldsymbol{H}\).
944: Clearly, this linear combination of Cartan generators
945: `rotates' under the action of \(P'\) like
946: \begin{equation}
947: P_j^{\prime\,-1}\,\boldsymbol{h}\,P_j'
948: \,=\,
949: \cos(4\pi\,\xi)\,\boldsymbol{h}
950: -\I\,\sin(4\pi\xi)\,(\boldsymbol{E}_{\SimpleRoot{j}}-\boldsymbol{E}_{-\SimpleRoot{j}})
951: \;,\label{eq:TransformationOfCartanGenerators}
952: \end{equation}
953: where we restricted ourselves to the case that \(\SimpleRoot{j}\) has length
954: \(\sqrt{2}\).
955: Since a linear combination of Cartan generators transforms non-trivially,
956: it is obvious that rank reduction is possible.
957: Note also that these rotations yield an extension of the well-known
958: Weyl reflections, i.e., the reflections with respect to a plane perpendicular
959: to a simple root.
960:
961: It is straightforward, but somewhat tedious to derive the
962: action on arbitrary roots \(\boldsymbol{E}_\alpha\).
963: In simply-laced gauge groups, the root chains have at most length two unless
964: they contain Cartan generators. Thus, \(N_{\alpha,\pm\SimpleRoot{j}}\ne 0\)
965: implies \(N_{\alpha,\mp\SimpleRoot{j}}=0\). For the upper sign, we obtain
966: (for \(\alpha\ne\SimpleRoot{i}\))
967: \begin{equation}
968: P^{\prime \,-1}_j\,\boldsymbol{E}_\alpha\,P'_j
969: \,= \,
970: \cos(2\pi\,N_{\alpha,\SimpleRoot{j}}\,\xi)
971: \,\boldsymbol{E}_\alpha
972: +\I\sin(2\pi\,N_{\alpha,\SimpleRoot{j}}\,\xi)\,
973: \boldsymbol{E}_{\alpha+\SimpleRoot{j}}
974: \;,
975: \end{equation}
976: where we use the normalization constants \(N_{\alpha,\beta}\)
977: as defined in equation \eqref{eq:DefOfNalphabeta} with the convention to
978: choose them positive.
979:
980: {}From the discussion so far, it is clear that we can break any simple group
981: factor completely by non-Abelian twists: The roots can always be removed
982: by suitable exponentials of the Cartan generators, and the \(\boldsymbol{H}_i\)
983: can be projected out by using Eq.~\eqref{eq:TransformationOfCartanGenerators}.
984: This observation has an obvious application: Let $H\subset G$
985: be the subgroup that we want to obtain by orbifolding. Let $H'\subset G$ be
986: the maximal subgroup that commutes with $H$ and the Cartan generators of
987: which are orthogonal to the Cartan generators of $H$. If $H'$ is semi-simple,
988: an orbifold breaking to $H$ is always possible. In this context, it is
989: interesting to observe that \(\mathrm{E}_8\) is the only simple group
990: containing a maximal regular subgroup of the form \(\mathrm{SU}(5)\times
991: H'\) with \(H'\) semi-simple, namely $\mathrm{E}_8\supset\mathrm{SU}(5)\times
992: \mathrm{SU}(5)$. Thus, one could say that \(\mathrm{E}_8\) is
993: the smallest GUT group larger than \(\mathrm{SU}(5)\) which can be
994: orbifolded to the SM without additional \(\mathrm{U}(1)\) factors.
995:
996: A further example is in order: It is clear that we can break an
997: \(\mathrm{SU}(2)\) factor completely by the methods described above. Thus,
998: since \(\mathrm{E}_6\supset\mathrm{SU}(6)\times\mathrm{SU}(2)\),
999: we can achieve \(\mathrm{E}_6\to\mathrm{SU}(6)\)
1000: by taking \(P=\exp[\pi\I\,\SimpleRoot{1}\cdot\boldsymbol{H}]\)
1001: and \(P'=\exp[\pi\I\,(\boldsymbol{E}_{\SimpleRoot{1}}+
1002: \boldsymbol{E}_{-\SimpleRoot{1}})/2]\).
1003: In addition, we can modify \(P\) in a way so that the breaking is stronger,
1004: e.g., \(\mathrm{E}_6\to\mathrm{SU}(5)\times\mathrm{U}(1)\).
1005:
1006: However, if extra \(\mathrm{U}(1)\) factors are contained in $H'$,
1007: the story becomes more complicated. One is tempted to conclude that such
1008: extra factors can not be removed, given that this is obviously not possible
1009: by adjoint VEV breaking. However, in the case of orbifold breaking
1010: this is not true. Consider, for example, \(\mathrm{SO}(5)\)
1011: which can be broken to \(\mathrm{SU}(2)\times\mathrm{U}(1)\)
1012: by using \(P=\diag(-1,-1,1,1,1)\). The extra \(\mathrm{U}(1)\) can be
1013: destroyed by invoking \(P'=\diag(1,-1,-1,-1,-1)\).
1014: This example is particularly interesting since here \(P\) and \(P'\) commute
1015: although the rank is reduced (which is possible because the corresponding
1016: generators
1017: do not commute). The above SO(5) example is special because \(P'\), which
1018: maps the U(1) generator to minus itself, acts on the other (real)
1019: representations in a way consistent with SU(2) symmetry. If we deal with
1020: complex representations, i.e., the adjoint of \(G\) branches as
1021: \begin{equation}
1022: \ad G
1023: \, \to \,
1024: \ad H \oplus\mathbbm{1}(0)\oplus\boldsymbol{R}(q)\oplus
1025: \overline{\boldsymbol{R}}(-q)
1026: \oplus \dots\;,\label{cr}
1027: \end{equation}
1028: where \(\boldsymbol{R}(q)\) and \(\overline{\boldsymbol{R}}(-q)\) are
1029: conjugate to each other, a flip of the \(\mathrm{U}(1)\) charge carries
1030: \(\boldsymbol{R}\) and \(\overline{\boldsymbol{R}}\) into each other.
1031: The flip then acts non-trivially on \(H\) so that flipping the
1032: \(\mathrm{U}(1)\) factor without affecting \(H\) is impossible.
1033:
1034: We emphasize that this excludes the possibility of
1035: orbifold breaking of the U(1) factor in a large class of cases. Namely,
1036: let $H\times$U(1)$\subset G$ such that the U(1) is the maximal group
1037: commuting with $H$. Clearly, any automorphism of $G$ leaving $H$ invariant
1038: has to map the U(1) onto itself. Since the only non-trivial automorphism
1039: of U(1) is the above sign flip, the presence of complex representations
1040: $H$ in the adjoint of $G$ (cf.~Eq.~\ref{cr}) excludes the required
1041: $H$-preserving autmorphism of $G$. The extension to $H\times H'\subset
1042: G$, where $H'$ is a product group containing U(1) factors, is
1043: straightforward.
1044:
1045: The above \(\mathrm{SO}(5)\) scenario with \(P\) and \(P'\) can, for
1046: example, be realized in \(4+2\) dimensions with compact space
1047: \(\mathbbm{T}^2/(\mathbbm{Z}_2\times\mathbbm{Z}_2')\). The \(\mathbbm{Z}_2\)
1048: generator acts on the torus as a rotation by \(180^\circ\), the
1049: \(\mathbbm{Z}_2'\) generator acts as a shift by half of one of the original
1050: torus translations (cf.~Fig.~\ref{fig:D2Torus}).
1051:
1052: It turns out that the elements of \(\mathbbm{Z}_2\times\mathbbm{Z}_2'\)
1053: comply with the multiplication law of the dihedral group $D_2$ of order
1054: 4.\footnote{Recall that the dihedral group of order \(2n\), called \(D_n\),
1055: can be envisaged as the group generated by the rotation of a regular
1056: \(n\)-polygon by
1057: \(2\pi/n\) and the flip over one of its edges \cite{Wigner}. Clearly,
1058: the dihedral group always can be embedded in an
1059: \(\mathrm{SO}(3)\simeq\mathrm{SU}(2)\).
1060: Anomalies of dihedral orbifolds are discussed in \cite{GrootNibbelink:2003gd}.}
1061: While the dihedral group of order
1062: 4 is Abelian, higher order dihedral groups are not. We illustrate a
1063: possible way of using the order 6 group $D_3$ in an orbifold construction in
1064: Fig.~\ref{fig:D3Torus}. It follows the
1065: \(\mathbbm{T}^2/(\mathbbm{Z}_2\times\mathbbm{Z}_2')\) construction up to
1066: the fact that we now divide the cell into three parts instead of two.
1067: Embedding it into a gauge group then allows for realizing non-Abelian twists.
1068:
1069: \begin{figure}[!h]
1070: \begin{center}
1071: \subfigure[$\mathbbm{T}^2/D_2$\label{fig:D2Torus}]
1072: {\CenterObject{\includegraphics{D2Torus.eps}}}
1073: \hfil
1074: \subfigure[$\mathbbm{T}^2/D_3$\label{fig:D3Torus}]
1075: {\CenterObject{\includegraphics{D3Torus.eps}}}
1076: \end{center}
1077: \caption{Examples of (a) a \(\mathbbm{T}^2/D_2\) and (b) a
1078: $\mathbbm{T}^2/D_3$
1079: orbifold where rank reduction is possible. The action of the
1080: dihedral group consists in a rotation by \(180^\circ\) around the origin,
1081: and in a translation by \(e_1'\) in case (a), and a translation by
1082: \(e_1^{\prime\prime}\) or \(2\,e_1^{\prime\prime}\) in case (b).}
1083: \label{fig:DihedralExamples}
1084: \end{figure}
1085:
1086: These examples can be generalized in the following way: The orbifold can be
1087: interpreted as \(\mathbbm{O}=\mathbbm{T}^n/R\) where \(R\) is a symmetry
1088: of the lattice, and the torus arises by modding out flat space by
1089: discrete translations, \(\mathbbm{T}^n=\mathbbm{R}^n/\Lambda\).
1090: By embedding the full symmetry group \(K\), containing the operations of
1091: \(R\) as well as the translations, into the gauge group, it is possible
1092: to achieve that the torus \(\mathbbm{T}^n\), which arises as
1093: intermediate step in this picture, carries Wilson lines \cite{Ibanez:1986tp}.
1094: Since the generators associated with the Wilson lines do not necessarily
1095: commute with the twists corresponding to embedding the operations of \(R\)
1096: into the gauge group, rank reduction is possible \cite{Ibanez:1987xa}.
1097: We believe that similar constructions will be important for model building.
1098:
1099: Let us briefly comment on non-regular embeddings. Consider the group
1100: \(\mathrm{SU}(3)\) which contains \(\mathrm{SO}(3)\) (the subgroup of real
1101: matrices) as an S-subgroup (in Dynkin's terminology). Let us pick two
1102: generators of the embedded \(\mathrm{SO}(3)\), for instance
1103: \begin{equation}
1104: T_1
1105: \,=\,
1106: \left(\begin{array}{ccc}
1107: 0 & 1 & 0\\
1108: -1 & 0 & 0 \\
1109: 0 & 0 & 0
1110: \end{array}\right)
1111: \quad\text{and}\quad
1112: T_2
1113: \,=\,
1114: \left(\begin{array}{ccc}
1115: 0 & 0 & -1\\
1116: 0 & 0 & 0 \\
1117: 1 & 0 & 0
1118: \end{array}\right)
1119: \;.
1120: \end{equation}
1121: It is then straightforward to convince oneself that imposing the twists
1122: \(P_1=\exp(2\pi\I\,T_1/4)\) and \(P_2=\exp(2\pi\I\,T_2/4)\) breaks
1123: \(\mathrm{SU}(3)\) completely. Similar constructions can be used to
1124: break larger groups with only a few twists. For instance,
1125: \(\mathrm{E}_8\) has a maximal S-subalgebra \(\mathfrak{su}(2)\) and
1126: can therefore be broken completely by only two twists, e.g.,
1127: by embedding a suitable dihedral group in the \(\mathrm{SU}(2)\).
1128:
1129:
1130: \section{Conifold GUTs}
1131: \label{coni}
1132:
1133: We now want to continue the discussion of the generic structure of orbifold
1134: GUTs given at the beginning of Sec.~\ref{sg} and show that a mild
1135: generalization of the construction principles leads to a much larger
1136: freedom in model building. Our main focus will be on 6d models.
1137:
1138: \subsection{Geometry and gauge symmetry breaking}
1139:
1140: In 5 dimensions, the geometry is very constraining. Up to isomorphism, the
1141: only smooth compact manifold is $\mathbbm{S}^1$, where one has the familiar
1142: problems of obtaining chiral matter and of fixing the Wilson line, the value
1143: of which represents a modulus which, in the SUSY setting, can not be stabilized
1144: by perturbative effects. The only compact orbifold is the interval, which can
1145: always be viewed as $\mathbbm{S}^1/(\mathbbm{Z}_2\times \mathbbm{Z}_2')$ (with
1146: $\mathbbm{S}^1/\mathbbm{Z}_2$ being a special case). The gauge breaking at
1147: each boundary is determined by a $\mathbbm{Z}_2$ automorphism and can be
1148: interpreted as explicit breaking by boundary conditions. One may try to
1149: generalize the setting by considering breaking by a boundary localized
1150: Higgs (in the limit where the VEV becomes large)~\cite{nsw} or ascribing
1151: Dirichlet and Neumann boundary conditions to different gauge fields (without
1152: the $\mathbbm{Z}_2$ automorphism restriction)~\cite{hm1}. Furthermore, it
1153: is possible to ascribe the breaking to a singular Wilson line crossing the
1154: boundary~\cite{hm1}. However, it appears to be unavoidable that geometry is
1155: used only in a fairly trivial way and that the breaking is confined entirely
1156: to the small-scale physics near the brane, outside the validity range of
1157: effective field theory.
1158:
1159: Group-theoretically, the 5d setting is also fairly constrained since the
1160: relative orientation of the gauge twists at the two boundaries is a
1161: modulus. To be more specific, let $P_1=\exp({\boldsymbol{T}_1})$ and
1162: $P_2=\exp({\boldsymbol{T}_2})$ be the two relevant twists. Even though this
1163: makes rank reduction possible in principle, we are faced with the problem
1164: that, if the Wilson line connecting the boundaries develops an appropriate
1165: VEV, the situation becomes equivalent to both $\boldsymbol{T}_1$ and
1166: $\boldsymbol{T}_2$ being in the Cartan subalgebra, in which case the
1167: symmetry is enhanced to a maximal-rank subgroup. SUSY prevents
1168: the modulus from being fixed by loop corrections.\footnote{For more
1169: details and a discussion of the non-supersymmetric case see~\cite{heb}
1170: and~\cite{Haba:2002py} respectively.}
1171:
1172: In 6 dimensions, the situation is much more complicated and interesting.
1173: Clearly, the smooth torus has the
1174: same problems as the $\mathbbm{S}^1$ discussed above. However, there is a
1175: large number of
1176: compact manifolds with conical singularities. A simple way to envisage such
1177: singular manifolds or, more precisely, conifolds is given in Fig.~\ref{sm}.
1178: The fundamental space consists of two identical triangles. The geometry
1179: is determined by gluing together the edges of the depicted triangles,
1180: thus leading to a triangle with a front and a back, a triangular
1181: `pillow'. It is flat everywhere except for the three conical singularities
1182: corresponding to the three corners of the basic triangle. Each deficit
1183: angle is $2(\pi-\varphi)$, where $\varphi$ is the corresponding angle of the
1184: triangle. Obviously, in this construction the basic triangle can be replaced
1185: by any polygon. If the polygon is non-convex, negative deficit angles
1186: appear.
1187:
1188: \begin{figure}[!ht]
1189: \begin{center}
1190: \CenterObject{\includegraphics{ConicalTriangle.eps}}
1191: \end{center}
1192: \caption{Construction of a compact manifold with singularities from two
1193: triangles.}\label{sm}
1194: \end{figure}
1195:
1196: Four specific polygons deserve a separate discussion. These are the
1197: rectangle, the equilateral triangle, the isosceles triangle with a
1198: $90^{\circ}$ angle, and the triangle with angles $30^\circ$, $60^\circ$, and
1199: $90^\circ$. The conifolds constructed in the above manner from these
1200: polygons can alternatively be derived from the torus as a $\mathbbm{Z}_2$,
1201: $\mathbbm{Z}_3$, $\mathbbm{Z}_4$ and $\mathbbm{Z}_6$ orbifold, respectively.
1202: Given that $\mathbbm{Z}_n$ can not be a symmetry
1203: of a 2-dimensional lattice for $n>6$, it is clear that this last method of
1204: constructing conifolds is highly constrained when compared to the generic
1205: conifold of Fig.~\ref{sm} with an arbitrary polygon. However, from the
1206: perspective of effective field theory model building, there appears to
1207: be no fundamental reason to discard the multitude of possibilities arising
1208: in the more general framework.
1209:
1210: Clearly, even more possibilities open up if, in addition to conical
1211: singularities one allows for 1-dimensional boundaries. These arise in
1212: orbifolding if a $\mathbbm{Z}_2$ reflection symmetry (in contrast to the
1213: $\mathbbm{Z}_n$ rotation symmetries above) of the torus is modded
1214: out~\cite{hnos} (see also~\cite{li}). However, in what follows we will
1215: concentrate on construction with conical singularities only.
1216:
1217: We now turn to the possibilities of geometric gauge symmetry breaking
1218: on conifolds. Recall first that, if a given conifold can be constructed
1219: from a smooth manifold by modding out a discrete symmetry group, i.e.,
1220: as an orbifold, then
1221: an appropriate embedding of this discrete group into the automorphism
1222: group of the gauge Lie algebra will lead to a gauge
1223: symmetry reduction. Working directly on the fundamental space (as opposed to
1224: the covering space) this gauge breaking can be ascribed to non-trivial
1225: values of Wilson lines encircling each of the conical singularities.
1226:
1227: It is now fairly obvious how to introduce this type of breaking in the
1228: generic construction of Fig.~\ref{sm} (possibly with the triangle replaced
1229: by an arbitrary polygon). First, we identify one edge of the front polygon
1230: with the corresponding edge of the back polygon.
1231: Next, when identifying along the two adjacent edges,
1232: one uses the freedom of introducing a relative gauge twist $P\in G$. In more
1233: detail, if $(x,y)$ and $(x',y')$ parametrize front and back polygon near
1234: the relevant edge (such that the edge is at $y=0$ or $y'=0$), one demands
1235: $A'(x'=x,y'=0)=P\,A(x,y=0)\,P^{-1}$ for the gauge potentials $A$ and $A'$ on
1236: the two polygons. Continuing with the identifications, one finds that there
1237: is a freedom of choosing $n-1$ gauge twists $P_i$ in the presence of $n$
1238: conical singularities. Technically, this is due to the fact that the
1239: identification along one of the edges can always be made trivial using
1240: global gauge rotations of one of the polygons. A geometric understanding
1241: follows from the fact that the global topology is that of a sphere, in which
1242: case the Wilson lines around $n-1$ singularities fix the last Wilson line
1243: (we always assume the vacuum configuration, i.e., $A$ is locally pure gauge).
1244:
1245: Clearly, we want to obtain a smooth manifold (except for the singularities)
1246: in the end so that, to be more precise, the $P_i$ have to be introduced in
1247: the appropriate transition functions of the defining atlas. However, we
1248: believe that it is not necessary to spell out this familiar construction
1249: in detail.
1250:
1251: Instead of using only inner automorphisms described by $P_i$, we could have
1252: allowed for outer automorphisms in the transition functions. In this case,
1253: which we will not pursue in this paper, the corresponding vacua are clearly
1254: disconnected from those defined only by inner automorphisms. The theory can
1255: then be thought of as defined on a generalization of a principal bundle (in the
1256: commonly used definition of principal bundles the transition functions involve
1257: only inner automorphisms).
1258:
1259: We now want to analyze the gauge fields in a small open subset including one
1260: conical singularity. A convenient parametrization is given by polar
1261: coordinates $(r,\varphi)$ with $0<r<\epsilon$ and $0\leq\varphi<\beta$, where the
1262: singularity is at $r=0$ and the deficit angle is $2\pi-\beta$. As familiar
1263: from the Hosotani mechanism on smooth manifolds~\cite{hos}, we can trade the
1264: gauge twist in the matching from $\varphi=\beta$ to $\varphi=0$ for a background
1265: gauge field which, for a twist $P=\exp(\boldsymbol{T})$, can be chosen as
1266: $A=\boldsymbol{e}_\varphi\boldsymbol{T}/(\beta r)$. Here $\boldsymbol{e}_\varphi$
1267: is the unit-vector in $\partial_\varphi$ direction so that $A$ is a
1268: Lie-algebra-valued vector. This simple exercise demonstrates explicitly
1269: that, at least locally, the breaking can be attributed to a non-vanishing
1270: gauge field VEV in a flat direction. However, in contrast to the Hosotani
1271: mechanism, the corresponding modulus can be fixed without violating the
1272: locality assumption (which we consider as fairly fundamental in effective
1273: field theory). Namely, the value of the Wilson line described by the
1274: above $A$ can be determined by some unspecified small-distance physics
1275: directly at the singularity. This is similar to the boundary breaking
1276: in 5 dimensions. In contrast to the 5d case, however, the breaking at the
1277: conical singularity is visible to the bulk observer, who can encircle the
1278: singularity and measure the Wilson line without coming close to the
1279: singularity. Thus, one might be tempted to conclude that this type of
1280: breaking has a better definition in terms of low-energy effective field
1281: theory.
1282:
1283: To conclude this subsection, we want to collect the generalizations of
1284: 6d field theoretic orbifold models discussed above. First, one can work
1285: on conifolds, i.e., use deficit angles that can not result from modding out
1286: on the basis of a smooth manifold. The gauge twist at each singularity may,
1287: however, be still required to be consistent with the geometric twist.
1288: Second, one can insist on conventional orbifolds as far as the geometry
1289: is concerned but use arbitrary gauge twists at each conical singularity,
1290: i.e., give up the connection between the rotation angles in tangent and
1291: gauge space. Third, one may drop both constraints and work on conifolds
1292: with arbitrary deficit angles and gauge twists. Obviously, such
1293: constructions can also be carried out in more than 6 dimensions.
1294: The detailed discussion of those is beyond the scope of the present paper.
1295:
1296:
1297: \subsection{Generating chiral matter}\label{chir}
1298:
1299: In general, compactification on a non-flat manifold can provide
1300: chiral matter if the holonomy group of the compact manifold fulfills
1301: certain criteria. For example, it is well-known that compactification
1302: of a 10d SYM theory on Calabi-Yau manifolds with \(\mathrm{SU}(3)\)
1303: holonomy~\cite{Candelas:en} or on orbifolds~\cite{dhvw} leads to
1304: \(N=1\) SUSY in 4d.
1305: Both constructions are not unrelated as many orbifolds can be regarded
1306: as singular limits of manifolds in which the curvature is concentrated at the
1307: fixed points. Since the reduction of SUSY is a matter of geometry,
1308: compactification of a higher-dimensional field theory on a conifold
1309: can also lead to \(N=1\) supersymmetric models in 4d.
1310:
1311: Interesting models have been constructed using the fact that
1312: the vector multiplet of \(N=(1,1)\) SUSY in 6d corresponds to
1313: one vector and three chiral multiplets in 4d language,
1314: \(A=(V,\phi_1,\phi_2,\phi_3)\). The fact that three copies of chiral
1315: multiplets appear automatically may be an explanation of the observed
1316: number of generations \cite{wy}. The above 6d theory can be interpreted as
1317: arising from a 10d SYM, in which case the scalars of the chiral
1318: multiplets are the extra components of gauge fields~\cite{mss}, for example,
1319: \(\phi_1\ni A_5+\I A_6\), \(\phi_2\ni A_7+\I A_8\) and \(\phi_3\ni
1320: A_9+\I A_{10}\) (\(A_M\) denote the components of the 10d vector).
1321: When defining our 6d models, we require the field transformations
1322: associated with going around a conical singularity to be an element
1323: of an SU(3) subgroup of the full $\mathrm{SO}(6)\simeq\mathrm{SU}(4)$
1324: symmetry of the
1325: underlying 10d SYM theory. Under this subgroup, which we call \(\mathrm{SU}(3)\)
1326: R-symmetry, the chiral superfields $\phi_i$ transform as a
1327: $\boldsymbol{3}$.\footnote{For more details see, e.g.,~\cite{mss}
1328: as well as~\cite{wy,Imamura:2001es}.}
1329: The appealing feature of such a construction is that matter
1330: multiplets are not put in `by hand' but arise in a natural way from a
1331: higher-dimensional SYM theory \cite{Babu:2002ti,wy,gmn,nb}.
1332:
1333: The action of the \(\mathrm{SU}(3)\) R-symmetry transformation on the chiral
1334: superfields is not completely arbitrary. For example, if \(\phi_1\ni A_5+\I A_6\),
1335: the transformation of \(\phi_1\) is fixed by geometry, e.g.,
1336: when modding out a rotation symmetry, a corresponding rotation has to be
1337: applied to the \(\phi_1\) superfield. Thus, when going around a conical
1338: singularity, \(\phi_1\) receives a phase which is given by \(2\varphi\),
1339: where \(\varphi\) is the corresponding angle of the polygon.
1340: Since multiplying by the phase \(e^{\I 2\varphi}\) corresponds to a rotation
1341: in the complex plane, we will call \(2\varphi\) `rotation angle' in what
1342: follows. Clearly, the rotation angle sums up with the deficit angle to
1343: \(2\pi\).
1344:
1345:
1346: \section{Specific models}\label{mo}
1347:
1348: Let us now discuss three models in which some of the main features of the
1349: last sections are exemplified. All these models are based on a SYM
1350: theory in \(4+2\) dimensions endowed with \((N_1,N_2)=(1,1)\) SUSY.
1351: In 4d we then deal with three chiral superfields
1352: \(\phi_1\), \(\phi_2\) and \(\phi_3\) where we assume that
1353: \(\phi_1=A_5+\I A_6\) so that the action of the R-symmetry on \(\phi_1\) is
1354: fixed (cf.\ Sec.~\ref{chir}).
1355:
1356: \subsection{$\boldsymbol{\mathrm{E}_7\to\mathrm{SU}(5)\times\mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)}$}
1357:
1358: Consider a SYM theory based on an \(\mathrm{E}_7\) gauge group.
1359: \(\mathrm{E}_7\) contains
1360: \(\mathrm{SU}(5)\times\mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)\),
1361: and the adjoint representation decomposes as
1362: \begin{eqnarray}
1363: \boldsymbol{133}
1364: & \to &
1365: (\boldsymbol{24},\boldsymbol{1})_0
1366: \oplus
1367: (\boldsymbol{1},\boldsymbol{1})_{0}
1368: \oplus
1369: (\boldsymbol{1},\boldsymbol{8})_{0}
1370: \oplus
1371: (\boldsymbol{5},\boldsymbol{1})_6
1372: \oplus
1373: (\overline{\boldsymbol{5}},\boldsymbol{1})_{-6}
1374: \nonumber\\*
1375: & & {}
1376: \oplus
1377: (\boldsymbol{10},\overline{\boldsymbol{3}})_{-2}
1378: \oplus
1379: (\overline{\boldsymbol{5}},\boldsymbol{3})_{-4}
1380: \oplus
1381: (\overline{\boldsymbol{10}},\boldsymbol{3})_{2}
1382: \oplus
1383: (\boldsymbol{5},\overline{\boldsymbol{3}})_{4}
1384: \;,\label{eq:DecompositionOfE7}
1385: \end{eqnarray}
1386: where we use a notation analogous to Eq.~\eqref{eq:DecompositionOf45}.
1387:
1388: As explained in Sec.~\ref{sg}, the twist \(P\) which causes the desired breaking
1389: can be understood as exponential of the \(\mathrm{U}(1)\) generator.
1390: Under this twist, the multiplets appearing in Eq.~\eqref{eq:DecompositionOfE7}
1391: acquire phases which are proportional to the \(\mathrm{U}(1)\) charge.
1392: By taking the proportionality constant to be \(-1/12\), we arrive at
1393: the phases listed in Tab.~\ref{tab:PhasesE7toSU5} where here and below phases
1394: are given in units of \(2\pi\).\footnote{The twist can be thought of as
1395: \(P=\diag(\omega,\omega,\omega,\omega,\omega,-\omega)\),
1396: with \(\omega\) defined as the 12th root of 1,
1397: acting on the \(\mathrm{SU}(6)\) embedded in \(\mathrm{E}_7\).
1398: Although the action of \(P\) on a fundamental representation
1399: of \(\mathrm{SU}(6)\) would be the one of a \(\mathbbm{Z}_{12}\) twist,
1400: its action on \(\mathrm{E}_7\) is \(\mathbbm{Z}_6\) since the adjoint
1401: of \(\mathrm{E}_7\) only contains antisymmetric and adjoint representations
1402: of \(\mathrm{SU}(6)\).}
1403:
1404: \begin{table}[!ht]
1405: \begin{center}
1406: $\begin{array}{|rcl|rcl|rcl|}
1407: \hline
1408: (\boldsymbol{24},\boldsymbol{1})_{0} & : & \boldsymbol{0}
1409: &
1410: (\boldsymbol{5},\boldsymbol{1})_{6} & : & \boldsymbol{1/2}
1411: &
1412: (\overline{\boldsymbol{5}},\boldsymbol{3})_{-4} & : & \boldsymbol{1/3}
1413: \vphantom{\overline{\overline{\boldsymbol{5}}}}
1414: \\
1415: (\boldsymbol{1},\boldsymbol{1})_{0} & : & \boldsymbol{0}
1416: &
1417: (\overline{\boldsymbol{5}},\boldsymbol{1})_{-6} & : & \boldsymbol{1/2}
1418: &
1419: (\overline{\boldsymbol{10}},\boldsymbol{3})_{2} & : & 5/6
1420: \\
1421: (\boldsymbol{1},\boldsymbol{8})_{0} & : & \boldsymbol{0}
1422: &
1423: (\boldsymbol{10},\overline{\boldsymbol{3}})_{-2} & : & \boldsymbol{1/6}
1424: &
1425: (\boldsymbol{5},\overline{\boldsymbol{3}})_{4} & : & 2/3\\
1426: \hline
1427: \end{array}$
1428: \end{center}
1429: \caption{Phases (in units of \(2\pi\)) for the different multiplets of
1430: \(\mathrm{SU}(5)\times\mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)\subset\mathrm{E}_7\).
1431: Zeros correspond to the surviving gauge bosons, other phases which are
1432: compensated by the R-symmetry transformations are written in boldface.}
1433: \label{tab:PhasesE7toSU5}
1434: \end{table}
1435:
1436: The smallest phase present is \(1/6\) so that \(P\) is a \(\mathbbm{Z}_6\)
1437: twist. Therefore, the R-symmetry acts on \(\phi_1\) as a \(-60^\circ\)
1438: rotation in the 5-6 plane, and thus the
1439: \((\boldsymbol{10},\overline{\boldsymbol{3}})_{-2}\) possesses a zero-mode.
1440: We choose the transformation of \(\phi_2\) such that the
1441: \((\overline{\boldsymbol{5}},\boldsymbol{3})_{-4}\) survives as well,
1442: and the phase of \(\phi_3\) is then fixed by the determinant condition.
1443: More explicitly, by taking
1444: \begin{equation}
1445: R
1446: \,=\,
1447: \exp[2\pi\I\,\diag(-1/6,2/3,-1/2)]
1448: \;\in\mathrm{SU}(3)
1449: \;,
1450: \end{equation}
1451: we can achieve that 3 generations of \(\boldsymbol{10}\) and
1452: \(\overline{\boldsymbol{5}}\) survive without any mirrors, indicated
1453: by boldface phases in Tab.~\ref{tab:PhasesE7toSU5},
1454: and an \(N=1\) SUSY in 4d is preserved.
1455: It is also interesting to observe that the only additional
1456: surviving superfields, namely \((\boldsymbol{5},\boldsymbol{1})_6\) and
1457: \((\overline{\boldsymbol{5}},\boldsymbol{1})_{-6}\) which acquire phases
1458: \(1/2\) and therefore have zero-modes due to the third diagonal entry
1459: of \(R\), carry the quantum numbers of the light Higgs fields
1460: in the supersymmetric \(\mathrm{SU}(5)\) theory. Thus, the \(\mathrm{SU}(5)\)
1461: part of this model looks relevant for reality, and is in particular
1462: anomaly-free.
1463:
1464: The geometry of this model, which can be constructed as a standard orbifold
1465: \(\mathbbm{T}^2/\mathbbm{Z}_6\), is given by two triangles with
1466: angles \(30^\circ\), \(60^\circ\) and \(90^\circ\)
1467: (cf.\ Fig.~\ref{fig:DihedralTriangle}). The \(\mathbbm{Z}_6\) twist
1468: \(P\) (\(P_6=P\) in Fig.~\ref{fig:DihedralTriangle}) is associated with
1469: the first of these fixed points; the twists \(P^2\) and \(P^3\) are
1470: associated with the remaining two fixed points. By construction, the order
1471: of rotation in the two extra dimensions coincides with the order of the
1472: twist in the gauge group.
1473: \begin{figure}[!ht]
1474: \begin{center}
1475: \CenterObject{\includegraphics{DihedralOrbifold.eps}}
1476: \end{center}
1477: \caption{Example of a \(4+2\) dimensional orbifold allowing for
1478: \(\mathbbm{Z}_6\) twists. The fundamental space consists of two triangles.
1479: The geometry can be illustrated by gluing together the edges of the
1480: depicted triangles, thus leading to a triangle with a front- and a backside.}
1481: \label{fig:DihedralTriangle}
1482: \end{figure}
1483: It is then straightforward to determine the gauge groups which survive
1484: at these fixed points. In the actual example, they turn out to be
1485: \(\mathrm{SU}(6)\times\mathrm{SU}(3)_\mathrm{F}\) and \(\mathrm{SU}(8)\),
1486: respectively. The content of non-vanishing fields at these fixed points
1487: is also found to be anomaly-free under the relevant surviving gauge group
1488: in both cases, which implies the absence of localized anomalies~\cite{abca}.
1489:
1490: The complete model is, however, not free of anomalies. This is due to
1491: localized anomalies at the $P_6$ fixed points, where the gauge group is
1492: SU(5)$\times$U(1)$\times\mathrm{SU}(3)_\mathrm{F}$. However, the
1493: SU(5) part by itself is free of localized anomalies even at this fixed
1494: point. Thus, if \(\mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)\) is
1495: broken, as it has to be in order to describe reality, there are no
1496: anomalies. The desired breaking of the unwanted symmetries may be due
1497: to fields which live on the fixed points, however, discussing such
1498: possibilities is beyond the scope of this study.
1499: Note also that if we were to break the additional symmetry by rank-reducing
1500: twists, fewer matter fields would survive. That is also the reason why
1501: we do not use rank-reducing twists in the next two models.
1502:
1503: \subsection{$\boldsymbol{\mathrm{E}_8\to G_\mathrm{PS}\times \mathrm{U}(1)^3}$}
1504:
1505: Under \(\mathrm{E}_8\to\mathrm{SO}(10)\times\mathrm{SU}(4)\), the
1506: adjoint representation of \(\mathrm{E}_8\) decomposes like
1507: \begin{equation}
1508: \boldsymbol{248}
1509: \,\to\,
1510: (\boldsymbol{45},\boldsymbol{1})\oplus(\boldsymbol{1},\boldsymbol{15})
1511: \oplus(\boldsymbol{16},\boldsymbol{4})
1512: \oplus(\overline{\boldsymbol{16}},\overline{\boldsymbol{4}})
1513: \oplus(\boldsymbol{10},\boldsymbol{6})
1514: \;.
1515: \label{eq:248toGPSxSU4}
1516: \end{equation}
1517: \(\mathrm{SO}(10)\) contains the Pati-Salam group \cite{ps}
1518: \(G_\mathrm{PS}=\mathrm{SU}(4)\times\mathrm{SU}(2)\times\mathrm{SU}(2)\)
1519: whereby
1520: \begin{subequations}
1521: \begin{eqnarray}
1522: \boldsymbol{45}
1523: & \to &
1524: (\boldsymbol{15},\boldsymbol{1},\boldsymbol{1})\oplus
1525: (\boldsymbol{1},\boldsymbol{3},\boldsymbol{1})\oplus
1526: (\boldsymbol{1},\boldsymbol{1},\boldsymbol{3})\oplus
1527: (\boldsymbol{6},\boldsymbol{2},\boldsymbol{2})
1528: \;,\\*
1529: \boldsymbol{16}
1530: & \to &
1531: (\boldsymbol{4},\boldsymbol{2},\boldsymbol{1})\oplus
1532: (\overline{\boldsymbol{4}},\boldsymbol{1},\boldsymbol{2})
1533: \;,\\
1534: \boldsymbol{10}
1535: & \to &
1536: (\boldsymbol{6},\boldsymbol{1},\boldsymbol{1})\oplus
1537: (\boldsymbol{1},\boldsymbol{2},\boldsymbol{2})
1538: \;.
1539: \end{eqnarray}
1540: \end{subequations}
1541: This breaking can be achieved by using the rotation
1542: \(-\mathbbm{1}\in\mathrm{SU}(2)\) for the first (or the second)
1543: \(\mathrm{SU}(2)\). In addition, we can now impose the twist
1544: \begin{equation}
1545: F
1546: \,=\,
1547: \exp \left[2\pi\,\I\,
1548: \diag\left(1/3,-1/6,x,-1/6-x\right)
1549: \right]
1550: \in\mathrm{SU}(4)
1551: \;,
1552: \end{equation}
1553: where, e.g., \(x=\frac{1}{4}\),
1554: in order to break \(\mathrm{SU}(4)\to[\mathrm{U}(1)]^3\).
1555: The charges of the \(\boldsymbol{4}\) are then given by
1556: \(q_i\in\{1/3,-1/6,x,-1/6-x\}\). The charges of the \(\boldsymbol{6}\)
1557: are \(q_i+q_j\) with \(i\ne j\) since the \(\boldsymbol{6}\) is the
1558: antisymmetric part of the \(\boldsymbol{4}\times\boldsymbol{4}\) of
1559: \(\mathrm{SU}(4)\), and finally the charges of the \(\boldsymbol{15}\)
1560: are \(q_i-q_j\).
1561:
1562: Together with the R-symmetry transformation
1563: \begin{equation}
1564: R
1565: \,=\,
1566: \exp\left[2\pi\I\,\diag\left(-1/3,-1/3,-1/3\right)\right]
1567: \;,
1568: \end{equation}
1569: three chiral generations of matter and three Higgs, i.e.,
1570: \((\boldsymbol{1},\boldsymbol{2},\boldsymbol{2})\), survive.
1571: Only for certain \(x\), additional fields will possess zero-modes, and
1572: we will choose \(x\) to equal none of these values.
1573: Here, the number of generations is due to dimensional reduction
1574: of \((N_1,N_2)=(1,1)\) SUSY in 6d to 4d.
1575: The surviving gauge group is
1576: \(G_\mathrm{PS}\times[\mathrm{U}(1)]^3\).
1577: The geometry is given by an equilateral triangle with
1578: the three corners corresponding to three identical fixed points.
1579:
1580: Obviously, for such a construction, the geometric twist, i.e., the
1581: rotation in the two extra dimensions, is of a lower order than the
1582: group theoretical twist. This requires going beyond the usual
1583: field-theoretic orbifold constructions (although the geometry is still
1584: an orbifold).
1585: As proposed in Sec.~\ref{coni}, we define a field theory on a manifold
1586: with three conical singularities, each of them possessing a deficit
1587: angle of \(2\pi/3\). This construction is then an equilateral
1588: triangle. We then add Wilson lines such that the group-theoretical
1589: twist \(P\) at two of the fixed points equals the one described above.
1590: The twist at the third fixed point is then constrained to be $P^{-2}$ by
1591: the global geometry.
1592:
1593: At each singularity of the conifold, the Pati-Salam part of the gauge
1594: group is anomaly-free. This is obvious for first two fixed points since
1595: the non-vanishing fields are those of the standard model with three Higgs
1596: doublets. At the third fixed point, the gauge symmetry is enhanced to the
1597: group SO(10), which has no 4d anomalies.\footnote{
1598: Quite
1599: generally, the anomaly at a given conical singularity can be
1600: calculated from the zero-mode anomaly by considering a conifold where this
1601: specific singularity appears several times (possibly together with other
1602: conical singularities, the anomalies of which are already
1603: known)~\cite{ggnow}. However, we do not investigate this further in the
1604: present paper. For recent work on the explicit calculation of anomalies in
1605: 6d models see~\cite{abca,gq,GrootNibbelink:2003gd}.
1606: }
1607: Again, investigating mechanisms to break the extra \(\mathrm{U}(1)\)s as well
1608: as \(G_\mathrm{PS}\) to \(G_\mathrm{SM}\) is beyond the scope of this paper.
1609:
1610: Note finally that this particular model can be viewed as an extension
1611: of~\cite{wy}, where three generations arise from the three chiral
1612: superfields present in the 4d description of a 10d SYM theory, i.e., they
1613: follow from the presence of three complex extra dimensions.\footnote{
1614: It
1615: has been claimed that this is related to the mechanism for obtaining three
1616: generations used in the string theory models reviewed in~\cite{far}.
1617: }
1618: The new points in
1619: our construction are the doublet-triplet splitting solution arising from the
1620: breaking to the Pati-Salam group (see~\cite{dm} and the recent related
1621: stringy models of~\cite{kim1}) and the realization of all rather than just
1622: part of the matter fields in terms of the SYM multiplet.
1623:
1624:
1625: \subsection{$\boldsymbol{\mathrm{E}_8\to G_\mathrm{PS}\times
1626: \mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)}$}
1627:
1628: Alternatively, we can obtain \(G_\mathrm{PS}\) from \(\mathrm{E}_8\)
1629: and maintain an \(\mathrm{SU}(3)_\mathrm{F}\) flavour symmetry by
1630: breaking the extra \(\mathrm{SU}(4)\) of the
1631: decomposition \eqref{eq:248toGPSxSU4} to
1632: \(\mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)\).
1633: In order to achieve this breaking, we take a central element
1634: of \(\mathrm{SU}(3)\),
1635: \begin{equation}
1636: P
1637: \,=\,
1638: \exp\left[2\pi\I\,\left(1/3,1/3,1/3,0\right)\right]
1639: \;.
1640: \end{equation}
1641: The phases which arise by combining this twist with
1642: \(\exp(- 2\pi\I/4\,\mathbbm{1})\in\mathrm{SU}(4)\) are listed
1643: in Tab.~\ref{tab:PhasesE8toGPSxSU3F}.
1644: \begin{table}[!ht]
1645: \[
1646: \begin{array}{|rcl|rcl|rcl|}
1647: \hline
1648: & & & & & & & & \\[-0.3cm]
1649: (\boldsymbol{15},\boldsymbol{1},\boldsymbol{1};\boldsymbol{1}_0)
1650: & : & \boldsymbol{0}
1651: &
1652: (\boldsymbol{1},\boldsymbol{1},\boldsymbol{1};\overline{\boldsymbol{3}}_{-4/3})
1653: & : & 2/3
1654: &
1655: (\overline{\boldsymbol{4}},\boldsymbol{1},\boldsymbol{2}; \overline{\boldsymbol{3}}_{-1})
1656: & : & 11/12
1657: \\
1658: (\boldsymbol{1},\boldsymbol{3},\boldsymbol{1};\boldsymbol{1}_0)
1659: & : & \boldsymbol{0}
1660: &
1661: (\boldsymbol{4},\boldsymbol{2},\boldsymbol{1};\boldsymbol{3}_{1})
1662: & : & \boldsymbol{1/12}
1663: &
1664: (\overline{\boldsymbol{4}},\boldsymbol{1},\boldsymbol{2};\boldsymbol{1}_{-3})
1665: & : & 1/4
1666: \\
1667: (\boldsymbol{1},\boldsymbol{1},\boldsymbol{3};\boldsymbol{1}_0)
1668: & : & \boldsymbol{0}
1669: &
1670: (\boldsymbol{4},\boldsymbol{2},\boldsymbol{1};\boldsymbol{1}_{-3})
1671: & : & 3/4
1672: &
1673: (\boldsymbol{6},\boldsymbol{1},\boldsymbol{1};\boldsymbol{3}_{1})
1674: & : & 5/6
1675: \\
1676: (\boldsymbol{1},\boldsymbol{1},\boldsymbol{1};\boldsymbol{8}_0)
1677: & : & \boldsymbol{0}
1678: &
1679: (\overline{\boldsymbol{4}},\boldsymbol{1},\boldsymbol{2};\boldsymbol{3}_{1})
1680: & : & \boldsymbol{7/12}
1681: &
1682: (\boldsymbol{6},\boldsymbol{1},\boldsymbol{1};\overline{\boldsymbol{3}}_{-1})
1683: & : & 1/6
1684: \\
1685: (\boldsymbol{1},\boldsymbol{1},\boldsymbol{1};\boldsymbol{1}_0)
1686: & : & \boldsymbol{0}
1687: &
1688: (\overline{\boldsymbol{4}},\boldsymbol{1},\boldsymbol{2};\boldsymbol{1}_{3})
1689: & : & 1/4
1690: &
1691: (\boldsymbol{1},\boldsymbol{2},\boldsymbol{2};\boldsymbol{3}_{1})
1692: & : & \boldsymbol{1/3}
1693: \\
1694: (\boldsymbol{6},\boldsymbol{2},\boldsymbol{2};\boldsymbol{1}_0)
1695: & : & 1/2
1696: &
1697: (\boldsymbol{4},\boldsymbol{2},\boldsymbol{1};\overline{\boldsymbol{3}}_{-1})
1698: & : & 11/12
1699: &
1700: (\boldsymbol{1},\boldsymbol{2},\boldsymbol{2};\overline{\boldsymbol{3}}_{-1})
1701: & : & 2/3
1702: \\
1703: (\boldsymbol{1},\boldsymbol{1},\boldsymbol{1};\boldsymbol{3}_{4/3})
1704: & : & \boldsymbol{1/3}
1705: &
1706: (\boldsymbol{4},\boldsymbol{2},\boldsymbol{1};\boldsymbol{1}_{3})
1707: & : & 3/4
1708: & & &\\
1709: \hline
1710: \end{array}
1711: \]
1712: \caption{Table of the phase factors for the different multiplets
1713: of \(G_\mathrm{PS}\times\mathrm{SU}(3)_\mathrm{F}\times\mathrm{U}(1)\subset
1714: \mathrm{E}_8\). Zeros correspond to surviving gauge bosons; other phases
1715: which are compensated by the R-symmetry transformation are written
1716: in boldface.
1717: }
1718: \label{tab:PhasesE8toGPSxSU3F}
1719: \end{table}
1720: Now let us simultaneously impose an R-symmetry twist
1721: \begin{equation}
1722: R\,=\,
1723: \exp\left[2\pi\I\,\diag(-1/12,-7/12,-1/3)\right]
1724: \;.
1725: \end{equation}
1726: It is then easy to see from Tab.~\ref{tab:PhasesE8toGPSxSU3F} that the
1727: zero modes which emerge in the matter sector are three generations
1728: of SM matter, three Higgs and three additional neutrinos.
1729:
1730: In order to realize such a model, we have again to relax the constraints
1731: of usual orbifold models, and therefore consider a manifold with a conical
1732: singularity with deficit angle \(2\pi\cdot 5/12\) instead (cf.\ Sec.~\ref{coni}).
1733: To be more specific, we envisage the geometry of the model as an isosceles
1734: triangle with an angle of \(2\pi\cdot 5/12\).
1735: Each corner corresponds to a fixed point,
1736: and we are free to choose both \(\pi/12\) fixed points identically.
1737: By construction, the group-theoretical twist \(P\) at the \(\pi/12\) fixed
1738: points generate a \(\mathbbm{Z}_{12}\), i.e., \(P^{12}=\mathbbm{1}\).
1739: At the remaining \(2\pi\cdot 5/12\) `corner', we choose the twist
1740: \(P^{10}=P^{-2}\) for consistency. Interestingly, a quick inspection of
1741: Tab.~\ref{tab:PhasesE8toGPSxSU3F} reveals that the there surviving
1742: gauge symmetry is \(\mathrm{SO}(10)\). Obviously, the \(\mathrm{SO}(10)\)
1743: part of the gauge theory at this fixed point is anomaly-free automatically.
1744:
1745: Once more, discussing the breaking of the extra gauge symmetry is beyond
1746: the scope of this study.
1747:
1748:
1749: \section{Conclusions}\label{co}
1750:
1751: We have explored some of the group-theoretical possibilities in orbifold GUTs.
1752: In particular, we showed that, given a simple gauge group \(G\),
1753: the breaking to any maximal-rank regular subgroup can be achieved by
1754: orbifolding.
1755:
1756: We further studied rank reduction and found that
1757: simple group factors can always be broken completely.
1758: This is possible when using non-Abelian twists, and also if twists
1759: commute but the corresponding generators do not.
1760: Using such constructions in orbifolding is made possible by embedding
1761: a non-Abelian (or even Abelian) space group into the gauge group.
1762:
1763: We then extended the familiar concept of orbifold GUTs by replacing the
1764: orbifolds by manifolds with conical singularities. The possibilities we
1765: discussed include orbifold geometries endowed with unrestricted Wilson
1766: lines wrapping the conical singularities, manifolds with conical
1767: singularities with arbitrary deficit angles, and combinations thereof.
1768:
1769: Finally, we presented three specific models where three generations of fields
1770: carrying the SM quantum numbers come from a SYM theory in 6d. While the
1771: first one is a conventional orbifold model illustrating the usefulness of
1772: our group theoretical methods, the two others are based on the two new
1773: concepts mentioned above.
1774:
1775: To summarize, we explored several new and interesting methods and
1776: possibilities which can be used in orbifold GUTs and their generalizations.
1777:
1778: As none of our models is yet completely realistic, more effort is required
1779: in order to discuss phenomenological consequences. However, it is very
1780: appealing how easily three generations can be obtained and the
1781: doublet-triplet splitting problem can be solved. Thus, promoting our
1782: models to realistic ones in future studies appears to be worthwhile.
1783:
1784: \noindent {\bf Note added:} While this paper was being finalized,
1785: Ref.~\cite{kim}
1786: appeared where Dynkin diagram techniques were used as well. Aspects of our
1787: analysis not addressed by~\cite{kim} include, in particular, the breaking
1788: of any simple group to all maximal-rank regular subgroups, rank-reduction,
1789: as well as several new field-theoretic concepts and models.
1790:
1791: \vspace*{-.3cm}
1792: \section*{Acknowledgments}
1793:
1794: \vspace*{-.3cm}
1795: We would like to thank Fabian Bachmaier, Wilfried Buchm\"{u}ller, John
1796: March-Russell, Hans-Peter Nilles, Mathias de Riese and Marco Serone for
1797: useful discussions.
1798:
1799:
1800: \renewcommand{\thesection}{\Alph{section}}
1801: \renewcommand{\thesubsection}{\Alph{section}.\arabic{subsection}}
1802:
1803: \def\theequation{\Alph{section}.\arabic{equation}}
1804:
1805: \renewcommand{\thetable}{\Alph{section}.\arabic{table}}
1806: \setcounter{section}{0}
1807: %\appendixtrue
1808:
1809:
1810: \vspace*{-.3cm}
1811: \section{Table of orbifold twists}\label{app:OrbifoldTwists}
1812:
1813: \vspace*{-.3cm}
1814: \begin{table}[!ht]
1815: \begin{center}
1816: \begin{tabular}{|l|c|l|l|}
1817: \hline
1818: Group & Twist & Symmetric subgroup & Comment\\
1819: \hline
1820: \hline
1821: \(\mathrm{SU}(N+M)\) & \(\mathbbm{Z}_2\) &
1822: \(\mathrm{SU}(N)\times\mathrm{SU}(M)\times\mathrm{U}(1)\) & \\
1823: \hline
1824: \(\mathrm{SO}(N+M)\) & \(\mathbbm{Z}_2\) &
1825: \(\mathrm{SO}(N)\times\mathrm{SO}(M)\) & \(N\) or \(M\) even\\
1826: \(\mathrm{SO}(2N)\) & \(\mathbbm{Z}_2\) &
1827: \(\mathrm{SU}(N)\times\mathrm{U}(1)\) & \\
1828: \hline
1829: \(\mathrm{Sp}(2N+2M)\) & \(\mathbbm{Z}_2\) &
1830: \(\mathrm{Sp}(2N)\times\mathrm{Sp}(2M)\) & \\
1831: \(\mathrm{Sp}(2N)\) & \(\mathbbm{Z}_2\) &
1832: \(\mathrm{SU}(N)\times\mathrm{U}(1)\) & \\
1833: \hline
1834: \hline
1835: \(\mathrm{G}_2\) & \(\mathbbm{Z}_2\) &
1836: \(\mathrm{SU}(2)\times\mathrm{SU}(2)\) & \\
1837: \(\mathrm{G}_2\) & \(\mathbbm{Z}_3\) &
1838: \(\mathrm{SU}(3)\) & \\
1839: \hline
1840: \(\mathrm{F}_4\) & \(\mathbbm{Z}_2\) &
1841: \(\mathrm{Sp}(6)\times\mathrm{SU}(2)\) & \\
1842: \(\mathrm{F}_4\) & \(\mathbbm{Z}_3\) &
1843: \(\mathrm{SU}(3)\times\mathrm{SU}(3)\) & \\
1844: \(\mathrm{F}_4\) & \(\mathbbm{Z}_4\) &
1845: \(\mathrm{SU}(4)\times\mathrm{SU}(2)\) & not maximal\\
1846: \(\mathrm{F}_4\) & \(\mathbbm{Z}_2\) &
1847: \(\mathrm{SO}(9)\) & \\
1848: \hline
1849: \(\mathrm{E}_6\) & \(\mathbbm{Z}_2\) &
1850: \(\mathrm{SO}(10)\times\mathrm{U}(1)\) & \\
1851: \(\mathrm{E}_6\) & \(\mathbbm{Z}_2\) &
1852: \(\mathrm{SU}(6)\times\mathrm{SU}(2)\)& \\
1853: \(\mathrm{E}_6\) & \(\mathbbm{Z}_3\) &
1854: \(\mathrm{SU}(3)\times\mathrm{SU}(3)\times\mathrm{SU}(3)\) & \\
1855: \hline
1856: \(\mathrm{E}_7\) & \(\mathbbm{Z}_2\) &
1857: \(\mathrm{SO}(12)\times\mathrm{SU}(2)\) & \\
1858: \(\mathrm{E}_7\) & \(\mathbbm{Z}_3\) &
1859: \(\mathrm{SU}(6)\times\mathrm{SU}(3)\) & \\
1860: \(\mathrm{E}_7\) & \(\mathbbm{Z}_4\) &
1861: \(\mathrm{SU}(4)\times\mathrm{SU}(4)\times\mathrm{SU}(2)\)
1862: & not maximal\\
1863: \(\mathrm{E}_7\) & \(\mathbbm{Z}_2\) &
1864: \(\mathrm{E}_6\times\mathrm{U}(1)\) & \\
1865: \(\mathrm{E}_7\) & \(\mathbbm{Z}_2\) &
1866: \(\mathrm{SU}(8)\) & \\
1867: \hline
1868: \(\mathrm{E}_8\) & \(\mathbbm{Z}_2\) &
1869: \(\mathrm{SO}(16)\) & \\
1870: \(\mathrm{E}_8\) & \(\mathbbm{Z}_4\) &
1871: \(\mathrm{SU}(8)\times\mathrm{SU}(2)\) & not maximal\\
1872: \(\mathrm{E}_8\) & \(\mathbbm{Z}_6\) &
1873: \(\mathrm{SU}(6)\times\mathrm{SU}(3)\times\mathrm{SU}(2)\)
1874: & not maximal\\
1875: \(\mathrm{E}_8\) & \(\mathbbm{Z}_5\) &
1876: \(\mathrm{SU}(5)\times\mathrm{SU}(5)\) &\\
1877: \(\mathrm{E}_8\) & \(\mathbbm{Z}_4\) &
1878: \(\mathrm{SO}(10)\times\mathrm{SU}(4)\) & not maximal\\
1879: \(\mathrm{E}_8\) & \(\mathbbm{Z}_3\) &
1880: \(\mathrm{E}_6\times\mathrm{SU}(3)\) & \\
1881: \(\mathrm{E}_8\) & \(\mathbbm{Z}_2\) &
1882: \(\mathrm{E}_7\times\mathrm{SU}(2)\) &\\
1883: \(\mathrm{E}_8\) & \(\mathbbm{Z}_3\) &
1884: \(\mathrm{SU}(9)\) & \\
1885: \hline
1886: \end{tabular}
1887: \end{center}
1888: \caption{Maximal subgroups of the simple groups and the corresponding
1889: \(\mathbbm{Z}_n\) orbifold twists. The five non-maximal subgroups which
1890: can be obtained by removing one node of the extended Dynkin diagram
1891: are listed for the sake of completeness.}
1892: \label{tab:OrbifoldTwists}
1893: \end{table}
1894:
1895: \begin{thebibliography}{99}
1896:
1897: \bibitem{gg}
1898: H. Georgi and S. Glashow, Phys. Rev. Lett. {\bf 32} (1974) 438.
1899: %%CITATION = PRLTA,32,438;%%
1900:
1901: \bibitem{gfm}
1902: H.~Georgi, AIP Conf.\ Proc.\ {\bf 23} (1975) 575;\\
1903: %%CITATION = APCPC,23,575;%%
1904: H.~Fritzsch and P.~Minkowski, Annals Phys.\ {\bf 93} (1975) 193.
1905: %%CITATION = APNYA,93,193;%%
1906:
1907: \bibitem{ps}
1908: J.~Pati and A.~Salam, Phys. Rev. {\bf D8} (1973) 1240;\\
1909: %%CITATION = PHRVA,D8,1240;%%
1910: J.~Pati and A.~Salam, Phys. Rev. {\bf D10} (1974) 275.
1911: %%CITATION = PHRVA,D10,275;%%
1912:
1913: \bibitem{wy}
1914: T.~Watari and T.~Yanagida, Phys.\ Lett.\ B {\bf 532} (2002) 252
1915: [arXiv:hep-ph/0201086].
1916: %%CITATION = HEP-PH 0201086;%%
1917:
1918: %\cite{Babu:2002ti}
1919: \bibitem{Babu:2002ti}
1920: K.~S.~Babu, S.~M.~Barr and B.~s.~Kyae,
1921: Phys.\ Rev.\ D {\bf 65} (2002) 115008\\{}
1922: [arXiv:hep-ph/0202178].
1923: %%CITATION = HEP-PH 0202178;%%
1924:
1925: \bibitem{nb}
1926: G.~Burdman and Y.~Nomura,
1927: Nucl.\ Phys.\ B {\bf 656} (2003) 3
1928: [arXiv:hep-ph/0210257].
1929: %%CITATION = HEP-PH 0210257;%%
1930:
1931: \bibitem{gmn}
1932: I.~Gogoladze, Y.~Mimura and S.~Nandi,
1933: arXiv:hep-ph/0304118.
1934: %%CITATION = HEP-PH 0304118;%%
1935:
1936: \bibitem{wit}
1937: E.~Witten, Nucl.\ Phys.\ B {\bf 258} (1985) 75.
1938: %%CITATION = NUPHA,B258,75;%%
1939:
1940: \bibitem{dhvw}
1941: L.~J.~Dixon, J.~A.~Harvey, C.~Vafa and E.~Witten, Nucl.\ Phys.\ B {\bf 261}
1942: (1985) 678 and {\bf 274} (1986) 285.
1943: %%CITATION = NUPHA,B274,285;%%
1944: %%CITATION = NUPHA,B261,678;%%
1945:
1946: \bibitem{kaw}
1947: Y.~Kawamura,
1948: Prog.\ Theor.\ Phys.\ {\bf 105} (2001) 999
1949: [arXiv:hep-ph/0012125].
1950: %%CITATION = HEP-PH 0012125;%%
1951:
1952: \bibitem{af}
1953: G.~Altarelli and F.~Feruglio, Phys.\ Lett.\ B {\bf 511} (2001) 257
1954: [arXiv:hep-ph/0102301].
1955: %%CITATION = HEP-PH 0102301;%%
1956:
1957: \bibitem{hn}
1958: L.~J.~Hall and Y.~Nomura, Phys.\ Rev.\ D {\bf 64} (2001) 055003
1959: [arXiv:hep-ph/0103125].
1960: %%CITATION = HEP-PH 0103125;%%
1961:
1962: \bibitem{hm}
1963: A.~Hebecker and J.~March-Russell, Nucl.\ Phys.\ B {\bf 613} (2001) 3\\{}
1964: [arXiv:hep-ph/0106166].
1965: %%CITATION = HEP-PH 0106166;%%
1966:
1967: \bibitem{abc}
1968: T.~Asaka, W.~Buchm\"uller and L.~Covi, Phys.\ Lett.\ B {\bf 523} (2001) 199
1969: \\{} [arXiv:hep-ph/0108021].
1970: %%CITATION = HEP-PH 0108021;%%
1971:
1972: \bibitem{hnos}
1973: L.~J.~Hall, Y.~Nomura, T.~Okui and D.~R.~Smith, Phys.\ Rev.\ D {\bf 65}
1974: (2002) 035008 [arXiv:hep-ph/0108071].
1975: %%CITATION = HEP-PH 0108071;%%
1976:
1977: \bibitem{dyn1}
1978: E. B. Dynkin, \emph{The structure of semi-simple algebras},
1979: Amer. Math. Soc. Transl. No. 1 (1950) pp. 1-143,\\ also
1980: Amer. Math. Soc. Transl. (1) {\bf 9} (1962) pp. 328-469.
1981:
1982: \bibitem{dyn2}
1983: E. B. Dynkin, \emph{Semi-simple subalgebras of semi-simple Lie algebras},
1984: Amer. Math. Soc. Transl., Ser. 2, {\bf 6} (1957) pp. 111-244.
1985:
1986: \bibitem{dyn3}
1987: E. B. Dynkin, \emph{Maximal subgroups of the classical groups},
1988: Amer. Math. Soc. Transl., Ser. 2, {\bf 6} (1957) pp. 245-378.
1989:
1990: \bibitem{dynb}
1991: E. B. Dynkin, \emph{Selected papers}, Amer. Math. Soc., 1999.
1992:
1993: \bibitem{gil}
1994: R.~Gilmore, \emph{Lie groups, Lie algebras, and some of their applications},
1995: \\{} Malabar Krieger, 1994.
1996:
1997: \bibitem{cahn}
1998: R.~N.~Cahn, \emph{Semisimple Lie algebras and their representations},\\{}
1999: Benjamin/Cummings, 1984.
2000:
2001: \bibitem{Georgi:1999jb}
2002: H.~Georgi, \emph{Lie algebras in particle physics. {F}rom isospin to unified
2003: theories, 2nd edition}, vol.~54, Perseus Books, 1999.
2004:
2005: \bibitem{sla}
2006: R.~Slansky, Phys.\ Rept.\ {\bf 79} (1981) 1.
2007: %%CITATION = PRPLC,79,1;%%
2008:
2009: \bibitem{hm1}
2010: A.~Hebecker and J.~March-Russell, Nucl. Phys. \textbf{B625} (2002) 128
2011: \\{}[arXiv:hep-ph/0107039].
2012: %%CITATION = HEP-PH 0107039;%%
2013:
2014: \bibitem{Haba:2002py}
2015: N.~Haba, M.~Harada, Y.~Hosotani and Y.~Kawamura,
2016: Nucl.\ Phys.\ B {\bf 657} (2003) 169
2017: [arXiv:hep-ph/0212035].
2018: %%CITATION = HEP-PH 0212035;%%
2019:
2020: \bibitem{orb}
2021: L.~J.~Hall and Y.~Nomura, arXiv:hep-ph/0212134;\\
2022: %%CITATION = HEP-PH 0212134;%%
2023: M.~Quiros, arXiv:hep-ph/0302189.
2024: %%CITATION = HEP-PH 0302189;%%
2025:
2026: \bibitem{Fegan:1991jb}
2027: H.~D.~Fegan, \emph{An introduction to compact {L}ie groups}, World Scientific
2028: Publishing, 1991, 131 P.
2029:
2030: \bibitem{dmr}
2031: K.~R.~Dienes and J.~March-Russell,
2032: %``Realizing Higher-Level Gauge Symmetries in String Theory: New Embeddings for String GUTs,''
2033: Nucl.\ Phys.\ B {\bf 479} (1996) 113
2034: \\{}[arXiv:hep-th/9604112].
2035: %%CITATION = HEP-TH 9604112;%%
2036:
2037: \bibitem{kkkot}
2038: Y.~Katsuki, Y.~Kawamura, T.~Kobayashi, N.~Ohtsubo and K.~Tanioka, Prog.\
2039: Theor.\ Phys.\ {\bf 82} (1989) 171.\\
2040: %%CITATION = PTPKA,82,171;%%
2041: %
2042: % \bibitem{Katsuki:1989bf}
2043: Y.~Katsuki, Y.~Kawamura, T.~Kobayashi, N.~Ohtsubo, Y.~Ono and K.~Tanioka,
2044: Nucl.\ Phys.\ B {\bf 341} (1990) 611.
2045: %%CITATION = NUPHA,B341,611;%%
2046:
2047: \bibitem{Golubitsky:1971ex}
2048: M.~Golubitsky and B.~Rothschild, \emph{Primitive subalgebras of exceptional
2049: {L}ie algebras}, Pac. J. Math. \textbf{39 No. 2} (1971), 371--393.
2050:
2051: \bibitem{abc1}
2052: T.~Asaka, W.~Buchm\"uller and L.~Covi, Phys.\ Lett.\ B {\bf 540} (2002) 295
2053: \\{}[arXiv:hep-ph/0204358].
2054: %%CITATION = HEP-PH 0204358;%%
2055:
2056: %\cite{Haba:2002vc}
2057: \bibitem{Haba:2002vc}
2058: N.~Haba and Y.~Shimizu,
2059: %``Gauge-Higgs unification in the 5 dimensional E(6), E(7), and E(8) GUTs on orbifold,''
2060: arXiv:hep-ph/0212166.
2061: %%CITATION = HEP-PH 0212166;%%
2062:
2063: %\cite{Ibanez:1987xa}
2064: \bibitem{Ibanez:1987xa}
2065: L.~E.~Ibanez, H.~P.~Nilles and F.~Quevedo,
2066: %``Reducing The Rank Of The Gauge Group In Orbifold Compactifications Of The Heterotic String,''
2067: Phys.\ Lett.\ B {\bf 192} (1987) 332.
2068: %%CITATION = PHLTA,B192,332;%%
2069:
2070: %\cite{Ibanez:1986tp}
2071: \bibitem{Ibanez:1986tp}
2072: L.~E.~Ibanez, H.~P.~Nilles and F.~Quevedo,
2073: %``Orbifolds And Wilson Lines,''
2074: Phys.\ Lett.\ B {\bf 187} (1987) 25.
2075: %%CITATION = PHLTA,B187,25;%%
2076:
2077: \bibitem{Wigner}
2078: E.~P.~Wigner, \emph{Group theory}, Academic press, 1949, 372 P.
2079:
2080: \bibitem{GrootNibbelink:2003gd}
2081: S.~Groot Nibbelink,
2082: %``Traces on orbifolds: Anomalies and one-loop amplitudes,''
2083: arXiv:hep-th/0305139.
2084: %%CITATION = HEP-TH 0305139;%%
2085:
2086: \bibitem{nsw}
2087: Y.~Nomura, D.~R.~Smith and N.~Weiner, Nucl.\ Phys.\ B {\bf 613} (2001) 147
2088: \\{}[arXiv:hep-ph/0104041].
2089: %%CITATION = HEP-PH 0104041;%%
2090:
2091: \bibitem{heb}
2092: A.~Hebecker, Nucl.\ Phys.\ B {\bf 632} (2002) 101 [arXiv:hep-ph/0112230].
2093: %%CITATION = HEP-PH 0112230;%%
2094:
2095: \bibitem{li}
2096: T.~j.~Li, Nucl.\ Phys.\ B {\bf 633} (2002) 83 [arXiv:hep-th/0112255].
2097: %%CITATION = HEP-TH 0112255;%%
2098:
2099: \bibitem{hos}
2100: Y.~Hosotani, Phys.\ Lett.\ B {\bf 126} (1983) 309, and Annals Phys.\
2101: {\bf 190} (1989) 233.
2102:
2103: %\cite{Candelas:en}
2104: \bibitem{Candelas:en}
2105: P.~Candelas, G.~T.~Horowitz, A.~Strominger and E.~Witten,
2106: %``Vacuum Configurations For Superstrings,''
2107: Nucl.\ Phys.\ B {\bf 258} (1985) 46.
2108: %%CITATION = NUPHA,B258,46;%%
2109:
2110: \bibitem{mss}
2111: N.~Marcus, A.~Sagnotti and W.~Siegel, Nucl.\ Phys.\ B {\bf 224} (1983) 159.
2112: %%CITATION = NUPHA,B224,159;%%
2113:
2114: \bibitem{Imamura:2001es}
2115: Y.~Imamura, T.~Watari and T.~Yanagida,
2116: %``Semi-simple group unification in the supersymmetric brane world,''
2117: Phys.\ Rev.\ D {\bf 64} (2001) 065023
2118: \\{}[arXiv:hep-ph/0103251].
2119: %%CITATION = HEP-PH 0103251;%%
2120:
2121: \bibitem{abca}
2122: T.~Asaka, W.~Buchm\"uller and L.~Covi, Nucl.\ Phys.\ B {\bf 648} (2003) 231
2123: \\{}[arXiv:hep-ph/0209144].
2124: %%CITATION = HEP-PH 0209144;%%
2125:
2126: \bibitem{ggnow}
2127: F.~Gmeiner, S.~Groot Nibbelink, H.~P.~Nilles, M.~Olechowski and
2128: M.~G.~Walter, Nucl.\ Phys.\ B {\bf 648} (2003) 35 [arXiv:hep-th/0208146].
2129: %%CITATION = HEP-TH 0208146;%%
2130:
2131: \bibitem{gq}
2132: G.~von Gersdorff and M.~Quiros, arXiv:hep-th/0305024.
2133: %%CITATION = HEP-TH 0305024;%%
2134:
2135: \bibitem{far}
2136: A.~E.~Faraggi, talk at {\it 4th Int. Conf. on Phys. Beyond the Standard
2137: Model}, Lake Tahoe, 1994 [arXiv:hep-ph/9501288].
2138: %%CITATION = HEP-PH 9501288;%%
2139:
2140: \bibitem{dm}
2141: R.~Dermisek and A.~Mafi, Phys.\ Rev.\ D {\bf 65} (2002) 055002
2142: [arXiv:hep-ph/0108139];\\
2143: %%CITATION = HEP-PH 0108139;%%
2144: H.~D.~Kim and S.~Raby, JHEP {\bf 0301} (2003) 056 [arXiv:hep-ph/0212348].
2145: %%CITATION = HEP-PH 0212348;%%
2146:
2147: \bibitem{kim1}
2148: J.~E.~Kim, arXiv:hep-th/0301177;\\
2149: %%CITATION = HEP-TH 0301177;%%
2150: K.~S.~Choi and J.~E.~Kim, arXiv:hep-ph/0305002.
2151: %%CITATION = HEP-PH 0305002;%%
2152:
2153: \bibitem{kim}
2154: K.~S.~Choi, K.~Hwang and J.~E.~Kim,
2155: %``Dynkin diagram strategy for orbifolding with Wilson lines,''
2156: arXiv:hep-th/0304243.
2157: %%CITATION = HEP-TH 0304243;%%
2158:
2159: \end{thebibliography}
2160: \end{document}
2161:
2162: