1: \documentclass[a4paper,final]{appolb}
2: \usepackage{epsfig}
3: \usepackage{graphicx} % including PostScript
4: \usepackage{amsmath}
5: \usepackage{amssymb} % e.g. \gtrsim
6: \usepackage{bbm} % BlackBoeard letters
7: \usepackage[small]{caption2} % font of captions is different from text
8: \usepackage{fleqn} % equations are not centered
9: \usepackage[small,loose]{subfigure} % subfigures with (a), (b) etc., ... and own caption
10: \usepackage{mciteW} % multiple citations
11:
12: \newcommand{\CenterObject}[1]{\ensuremath{\vcenter{\hbox{#1}}}}
13: \newcommand{\CenterEps}[2][1]{\ensuremath{\vcenter{%
14: \hbox{\includegraphics[scale=#1]{#2.eps}}}}}
15: % Input eps files - Usage: \CenterEps[ScaleFactor]{FileName}
16: \newcommand{\D}{\mathrm{d}}
17: \newcommand{\I}{\mathrm{i}}
18: \newcommand{\BmL}{\ensuremath{B\!-\!L} }
19: \newcommand{\SimpleRoot}[1]{\alpha_{(#1)}}
20: \newcommand{\FundamentalWeight}[1]{\mu^{(#1)}}
21: \newcommand{\E}[1]{\ensuremath{\mathrm{E}_{#1}}} % e.g. \E{8}
22: \newcommand{\G}[1]{\ensuremath{\mathrm{G}_{#1}}}
23: \newcommand{\SO}[1]{\ensuremath{\mathrm{SO}(#1)}}
24: \newcommand{\SU}[1]{\ensuremath{\mathrm{SU}(#1)}}
25: \newcommand{\U}[1]{\ensuremath{\mathrm{U}(#1)}}
26: \newcommand{\Z}[1]{\ensuremath{\mathbbm{Z}_{#1}}} % Z_N ->\Z{N}
27: %------------------------------------------------------
28:
29: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30: % %
31: % BEGINNING OF TEXT %
32: % %
33: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34:
35: \begin{document}
36: \preprint{DESY 05-260}
37: % \eqsec % uncomment this line to get equations numbered by (sec.num)
38: \title{
39: \vspace*{-3cm}
40: \begin{flushright}
41: {\normalfont DESY 05-260}
42: \end{flushright}
43: \vspace*{1cm}
44: \appHuge{Local Grand Unification}%
45: \thanks{Based on talks presented at the GUSTAVOFEST, Lisbon, July 2005, and
46: at the workshop `Strings and the real world', Ohio, November 2005.}%
47: }
48: \author{Wilfried Buchm\"uller and Koichi Hamaguchi
49: \address{Deutsches Elektronen-Synchrotron DESY, 22603 Hamburg, Germany}
50: \and
51: Oleg Lebedev and Michael Ratz
52: \address{Physikalisches Institut der Universit\"at Bonn, Nussallee 12, 53115 Bonn, Germany}
53: }
54: \maketitle
55: \begin{abstract}
56: \noindent
57: In the standard model matter fields form complete representations of a grand
58: unified group whereas Higgs fields belong to incomplete `split' multiplets. This
59: remarkable fact is naturally explained by `local grand unification' in
60: higher-dimensional extensions of the standard model. Here, the generations of
61: matter fields are localized in regions of compact space which are endowed with a
62: GUT gauge symmetry whereas the Higgs doublets are bulk fields. We realize local
63: grand unification in the framework of orbifold compactifications of the
64: heterotic string, and we present an example with \SO{10} as a local GUT group,
65: which leads to the supersymmetric standard model as an effective four-dimensional
66: theory. We also discuss different orbifold GUT limits and the unification of
67: gauge and Yukawa couplings.
68: \end{abstract}
69: \preprint{DESY 05-260}
70: \PACS{11.25.Wx, 12.10.-g, 12.60.Jv, 11.25.Mj}
71:
72: \section{Grand unification in $\boldsymbol{D>4}$ and doublet--triplet splitting}
73:
74: The standard model (SM) is a remarkably successful theory of the structure
75: of matter. It is based on the gauge group
76: $G_\mathrm{SM}=\SU3_C\times\SU2_\mathrm{L}\times\U1_Y$
77: and has three generations of matter transforming as
78: \begin{equation}
79: (\boldsymbol{3},\boldsymbol{2})_{1/6}+
80: (\overline{\boldsymbol{3}},\boldsymbol{1})_{-2/3}+
81: (\overline{\boldsymbol{3}},\boldsymbol{1})_{1/3}+
82: (\boldsymbol{1},\boldsymbol{2})_{-1/2}+
83: (\boldsymbol{1},\boldsymbol{1})_{1}\;.
84: \end{equation}
85: The evidence for neutrino masses strongly supports the existence of right-handed
86: neutrinos which are singlets under the SM gauge group. A crucial ingredient of
87: the SM is further an \SU2 doublet of Higgs fields containing the Higgs boson
88: which still remains to be discovered. From a theoretical perspective, one would
89: like to amend the standard model by supersymmetry. Apart from stabilizing the
90: hierarchy between the electroweak and Planck scales and providing a convincing
91: explanation of the observed dark matter, the minimal supersymmetric extension
92: of the standard model, the MSSM, predicts unification of the gauge couplings at
93: the unification scale $M_\mathrm{GUT}\simeq 2\cdot10^{16}\,\mathrm{GeV}$.
94:
95: Even more than the unification of gauge couplings, the symmetries and the
96: particle content of the standard model point towards grand unified theories
97: (GUTs) \cite{Georgi:1974sy,*Pati:1974yy}. Remarkably, one generation of matter,
98: including the right-handed neutrino, forms a single spinor representation of
99: \SO{10} \cite{Georgi:1975qb,*Fritzsch:1974nn}. It therefore appears natural to
100: assume an underlying \SO{10} structure of the theory. The route of unification,
101: continuing via exceptional groups, terminates at $\E8$,
102: \begin{equation}
103: \SO{10} \subset \E6 \subset \E8\;,
104: \end{equation}
105: which is beautifully realized in the heterotic string
106: \cite{Gross:1984dd,*Gross:1985fr}.
107:
108: An obstacle on the path towards unification are the Higgs fields, which are
109: \SU2 doublets, while the smallest \SO{10} representation containing Higgs
110: doublets, the $\boldsymbol{10}$--plet, predicts additional $\SU3_C$ triplets.
111: The fact that Higgs fields form incomplete `split' GUT representations is
112: particularly puzzling in supersymmetric theories where both matter and Higgs
113: fields are chiral multiplets. The triplets cannot have masses below
114: $M_\mathrm{GUT}$ since otherwise proton decay would be too rapid. This then
115: raises the question why \SU2 doublets are so much lighter than $\SU3_C$
116: triplets. This is the notorious doublet-triplet splitting problem of ordinary 4D
117: GUTs.
118:
119: Higher-dimensional theories offer new possibilities for gauge symmetry breaking
120: connected with the compactification to four dimensions. A simple and elegant
121: scheme, leading to chiral fermions in four dimensions, is the compactification
122: on orbifolds, first considered in string theories
123: \cite{Dixon:1985jw,*Dixon:1986jc,Ibanez:1986tp,*Ibanez:1987xa,*Ibanez:1987sn},
124: and more recently applied to GUT field theories \cite{Kawamura:1999nj,
125: *Kawamura:2000ev,*Altarelli:2001qj,*Hall:2001pg,*Hebecker:2001wq,*Asaka:2001eh,*Hall:2001xr}.
126: In orbifold compactifications the gauge symmetry of the 4D effective theory is
127: an intersection of larger symmetries at orbifold fixed points
128: (cf.~Fig.~\ref{fig:LocalGUTs}). Zero modes located at these fixed points all
129: appear in the 4D theory and form therefore representations of the larger local
130: symmetry groups. Zero modes of bulk fields, on the contrary, are only
131: representations of the smaller 4D gauge symmetry and form in general `split
132: multiplets'. Choosing now on some orbifold fixed points \SO{10} as local
133: symmetry, we obtain the picture of `local grand unification' illustrated in
134: Fig.~\ref{fig:LocalGUTs}. The SM gauge group is obtained as an intersection of
135: different local GUT groups. Matter fields appear as $\boldsymbol{16}$--plets
136: localized at the fixed points, whereas the Higgs doublets are associated with
137: bulk fields, which provides a solution of the doublet-triplet splitting problem.
138: In this way the structure of the standard model is naturally reproduced.
139:
140: \begin{figure}[t]
141: \centerline{\CenterObject{\includegraphics{pic1.eps}}}
142: \caption{The picture of local grand unification. The gauge group $G$ is broken
143: locally to different subgroups. Each of the local subgroups contains the
144: standard model gauge group $G_\mathrm{SM}$ which emerges as an intersection of
145: the local groups. `Brane' fields which are confined to a region with certain
146: symmetry have to come in complete matter multiplets of that symmetry. Hence,
147: localized $\boldsymbol{16}$-plets of \SO{10} are an attractive explanation of
148: complete matter generations. Higgs doublets, on the other hand, are states which are
149: not confined to an \SO{10} region, and can therefore appear as `split
150: multiplets' in the low--energy spectrum.}
151: \label{fig:LocalGUTs}
152: \end{figure}
153:
154:
155: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
156: \section{Orbifold compactification of the heterotic string}
157:
158: `Local grand unification' naturally occurs in compactifications of the
159: $\E8\times\E8$ heterotic string \cite{Gross:1984dd,*Gross:1985fr}. Six of the
160: ten space-time dimensions are compactified on an orbifold
161: \cite{Dixon:1985jw,*Dixon:1986jc,Ibanez:1986tp,*Ibanez:1987sn,Katsuki:1990bf}.
162: %
163: Specifically, we consider $\Z3\times\Z2$ orbifold compactifications on the
164: lattice $\Lambda_{\G2\times\SU3\times\SO4}$, which is a product of three
165: two-tori defined by the root vectors of \G2, \SU3 and \SO4, respectively
166: \cite{Kobayashi:2004ud,*Kobayashi:2004ya}. The \Z6 action is a rotation by
167: $2\pi/6$ in the \G2 plane, by $2\pi/3$ in the \SU3 plane and by $2\pi/2$ in the
168: \SO4 plane. This \Z6 action has fixed points in the various planes as
169: illustrated in Fig.~\ref{fig:RootLattice}.
170:
171: The rotation in the compact dimensions is embedded into the gauge degrees of
172: freedom. It acts as a shift, denoted by $V_6$, on the left-moving coordinates
173: $X_\mathrm{L}^I$ ($1\le I \le 16$), i.e.\ upon a rotation in the internal
174: space-time $X_\mathrm{L}^I$ transforms as
175: $X_\mathrm{L}^I\to X_\mathrm{L}^I+V_6^I$. This shift satisfies
176: $6V_6\in\Lambda_{\E8\times\E8}$, where $\Lambda_{\E8\times\E8}$ denotes the root
177: lattice of $\E8\times\E8$, reflecting that the corresponding automorphism has
178: order 6. In addition, some tori carry Wilson lines. For instance, a torus
179: translation along one axis in the \SO4 torus is associated with a shift by $W_2$
180: which has degree two, i.e.\ $2W_2\in\Lambda_{\E8\times\E8}$. A more detailed
181: discussion can be found elsewhere
182: \cite{Forste:2004ie,Kobayashi:2004ud,*Kobayashi:2004ya,Buchmuller:2004hv,Buchmuller:2005pr}.
183:
184: \begin{figure}[t]
185: \centerline{\CenterObject{\includegraphics{pic2.eps}}}
186: \caption{Root lattice $\Lambda_{G_2\times\SU3\times\SO4}$ and fixed points of
187: the \Z6 action.}
188: \label{fig:RootLattice}
189: \end{figure}
190:
191:
192: \subsection{Orbifold construction kit}
193: \label{sec:ock}
194:
195: The orbifold action leads to local gauge symmetry breakdown at the fixed points.
196: To see this, one analyzes locally the invariance conditions for the gauge fields
197: $A_\mu^p$, corresponding to the generator $p\in\Lambda_{\E8\times\E8}$. For
198: instance, at the origin in Fig.~\ref{fig:RootLattice}, the invariance condition
199: requires that $A_\mu^p$ vanish unless the corresponding generator is
200: `orthogonal' to the local shift $V_\mathrm{local}=V_6$. This implies that any
201: gauge boson not fulfilling $p\cdot V_6\equiv0$\footnote{Here `$\equiv$' means
202: `up to integers'.} is projected out of the zero-mode spectrum. The remaining
203: gauge bosons form a sub-algebra of the original $\E8\times\E8$. One can
204: thus say that the origin carries a local gauge group.
205:
206: Repeating the analysis at other fixed points leads in general to different local
207: projection conditions. For instance, the local projection condition at the
208: origin in the \G2 and \SU3 tori and at the bottom right position in the \SO4
209: torus is the same as before except that the shift now gets amended by the Wilson
210: line $W_2$, i.e.\ $V_\mathrm{local}=V_6+W_2$. This modified projection
211: condition, $p\cdot V_\mathrm{local}\equiv0$, implies a different local gauge
212: group.
213: %
214: An analogous analysis can be carried out for the remaining fixed points. The
215: result is that each of the fixed points carries a local gauge group, and those
216: local gauge groups are in general different. If two local shifts are equal,
217: i.e.\ if the Wilson line corresponding to the translation connecting the two
218: fixed points vanishes, the local gauge groups and their embedding into
219: $\E8\times\E8$ coincide.
220:
221: The gauge boson zero-modes consist of the gauge bosons surviving all
222: local projection conditions simultaneously. In other words, the gauge group
223: after compactification is an intersection of the local gauge groups. In the
224: following, we shall focus on the possibility that this intersection is, up to
225: \U1 factors and a `hidden sector gauge group', the SM gauge group
226: $G_\mathrm{SM}$ . It is crucial to note that each of the local gauge groups
227: contains $G_\mathrm{SM}$ as a subgroup, i.e.\
228: \begin{equation}
229: G_\mathrm{SM}\,\subsetneq\, G_\mathrm{local}\;.
230: \end{equation}
231: This leads to the picture of `local grand unified theories' where
232: $G_\mathrm{SM}$ emerges as an intersection of different GUT groups residing at
233: different fixed points.
234:
235: \begin{figure}[t]
236: \centerline{
237: \subfigure[Orbifold construction kit.\label{fig:OrbifoldConstructionKit}]{%
238: \CenterObject{\includegraphics{pic3.eps}}}
239: \quad
240: \subfigure[$2+1$ family models.\label{fig:2+1}]{
241: \CenterObject{\includegraphics{pic4.eps}}}
242: }
243: \caption{(a) orbifolds can be constructed by combining `corners' carrying a
244: local gauge group emerging from the action of a local shift. (b) In $2+1$ family
245: models, two families appear as $\boldsymbol{16}$--plets residing on fixed points
246: with local \SO{10} symmetry. The third family comes from elsewhere.
247: }
248: \end{figure}
249:
250:
251: The geometry of a 2D orbifold can be visualized as follows
252: (cf.~\cite{Quevedo:1996sv,*Hebecker:2003jt}). The fundamental region of a
253: $\Z{N}$ orbifold is one $N^\mathrm{th}$ of the fundamental region of the torus.
254: One can fold it and identify the remaining edges. This yields a `pillow' or
255: `ravioli' where the orbifold fixed points correspond to the corners (cf.\
256: Fig.~\ref{fig:LocalGUTs}).
257:
258: The corners of the pillow serve as building blocks for the construction of an
259: orbifold model. One starts at one corner with a local shift leading to a local
260: gauge group. The simplest possibility is to combine identical corners which
261: leads to an orbifold model without Wilson lines.
262: %
263: In order to construct realistic orbifold models, one has to consider
264: non-vanishing Wilson lines, i.e.\ combine corners with different gauge groups.
265: Gluing these corners together leads to an orbifold model with Wilson lines (cf.\
266: Fig.~\ref{fig:OrbifoldConstructionKit}).
267: %
268: Let us emphasize that one cannot combine these corners arbitrarily. Rather,
269: there are severe constraints coming from modular invariance, which restrict the
270: allowed Wilson lines (cf.~\cite{Forste:2004ie}).
271:
272:
273:
274: \subsection{Localized $\boldsymbol{16}$--plets and 2+1 family models}
275: \label{sec:2+1}
276:
277: The `orbifold construction kit' described above is a helpful tool for model
278: building. Note that each local GUT leads to local GUT representations. Among
279: the zero modes, the representations of the first twisted sector $T_1$ play
280: a special role. They correspond to `brane' fields which are completely
281: localized at the fixed points. In particular, they only feel the local gauge
282: group and therefore are forced to furnish complete GUT representations.
283:
284: An application of this observation to localized $\boldsymbol{16}$--plets of
285: local \SO{10} GUTs leads to a simple recipe for the construction of three
286: generation models. One starts with a `corner' which carries a local \SO{10} and
287: a $\boldsymbol{16}$--plet in the $T_1$ sector. This $\boldsymbol{16}$--plet will
288: appear as a complete multiplet in the massless spectrum, even though the
289: low-energy gauge group will be an intersection of \SO{10} with other groups, and
290: therefore only some subgroup of \SO{10}. As discussed above, this provides
291: understanding of the fact that there are complete generations and split
292: multiplets at the same time.
293:
294: In the geometry introduced above there are only two shifts which produce a
295: local \SO{10} together with the $\boldsymbol{16}$--plets
296: \cite{Katsuki:1989qz,*Katsuki:1989cs},
297: \begin{eqnarray}
298: V_6
299: & = &
300: \left(\frac{1}{2},\frac{1}{2},\frac{1}{3},0,0,0,0,0\right)
301: \left(\frac{1}{3},0,0,0,0,0,0,0\right)\;,
302: \nonumber\\
303: V_6'
304: & = &
305: \left(\frac{1}{3},\frac{1}{3},\frac{1}{3},0,0,0,0,0\right)
306: \left(\frac{1}{6},\frac{1}{6},0,0,0,0,0,0\right)\;.
307: \end{eqnarray}
308:
309: In the following, we shall focus on $2+1$ family models, where two families
310: come from two \SO{10} corners while the third one comes from somewhere else
311: (cf.\ Fig.~\ref{fig:2+1}). This pattern has recently been explored in the
312: context of Pati-Salam models \cite{Kobayashi:2004ud,*Kobayashi:2004ya}. In
313: contrast to models with three families of localized $\boldsymbol{16}$--plets, in
314: $2+1$ family models the third family and the Higgs originate partially from the
315: untwisted sector. As we shall see, this leads naturally to a situation where the
316: top Yukawa coupling is related to the gauge coupling. Note that an untwisted
317: $\boldsymbol{16}$--plet, i.e. a bulk third family, occurs only in the
318: $\E8\times\E8$ heterotic string and not in the \SO{32} heterotic string.
319:
320:
321:
322: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
323:
324: \begin{figure}[t]
325: \centerline{\CenterObject{\includegraphics{pic5.eps}}}
326: \caption{Local gauge groups up to \U1 factors and subgroups of the second \E8.
327: As indicated, the fixed point come in six pairs.}
328: \label{fig:LocalGaugeGroups}
329: \end{figure}
330:
331: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
332:
333: \section{The MSSM from the heterotic string}
334: \label{sec:MSSM}
335:
336: \subsection{Model definition and gauge groups}
337:
338: Let us now discuss a specific example \cite{Buchmuller:2005jr}. We consider a
339: compactification of the $\E8\times\E8$ heterotic string on the $\Z3\times\Z2$
340: orbifold described above. In order to obtain a $2+1$ family model, we
341: require only two Wilson lines $W_2$ and $W_3$, while a possible third Wilson
342: line associated with the `vertical' translation in the \SO4 torus
343: (cf.~Fig.~\ref{fig:RootLattice}) is set to zero. In an orthonormal basis,
344: the shift and the Wilson lines are given by
345: \begin{eqnarray}
346: V_6 & = &
347: \left(\frac{1}{2},\frac{1}{2},\frac{1}{3},0,0,0,0,0\right) \,
348: \left(\frac{1}{3},0,0,0,0,0,0,0\right)
349: \;,\nonumber\\
350: W_2 & = &
351: \left(\frac{1}{2},0,\frac{1}{2},\frac{1}{2},\frac{1}{2},0,0,0\right)
352: \,\left(-\frac{3}{4},\frac{1}{4},\frac{1}{4},-\frac{1}{4},\frac{1}{4},\frac{1}{4},\frac{1}{4},-\frac{1}{4}\right)
353: \;,\nonumber\\
354: W_3 & = &
355: \left(\frac{1}{3},0,0,\frac{1}{3},\frac{1}{3},\frac{1}{3},\frac{1}{3},\frac{1}{3}\right) \,
356: \left(1,\frac{1}{3},\frac{1}{3},\frac{1}{3},0,0,0,0\right)
357: \;. %\nonumber
358: \end{eqnarray}
359: The twelve fixed points depicted in Fig.~\ref{fig:LocalGaugeGroups} come in six
360: inequivalent pairs. They carry various local gauge groups whose intersection,
361: which is the unbroken gauge group after compactification, is given by
362: \begin{equation}
363: G~=~G_\mathrm{SM}\times\left[\SO{6}\times\SU2\right]\times\U1^8\;.
364: \label{eq:G}
365: \end{equation}
366: Here, the brackets indicate a subgroup of the second \E8. One of the $\U1$
367: factors is anomalous \cite{Dine:1987xk}. As a consequence, some fields charged
368: under the anomalous \U1 attain vacuum expectation values (VEVs)
369: which break this
370: \U1 (cf.~\cite{Cleaver:1997jb}). In our model there exist no
371: fields which are charged only under the anomalous \U1.\footnote{This feature
372: occurs rather generally in orbifold models with Wilson lines.} Hence,
373: further \U1 factors are necessarily broken, which is phenomenologically
374: attractive.
375:
376: \subsection{The massless spectrum}
377:
378: The zero modes of our model are given in Tab.~\ref{tab:Spectrum}, labeled
379: by their quantum numbers w.r.t. the gauge group
380: $G_\mathrm{SM}\times[\SO{6}\times\SU2]$. Further details such as the oscillator
381: numbers will be presented elsewhere \cite{Buchmuller:2005pr}.
382: Here we only highlight some aspects of the spectrum.
383:
384: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
385: \begin{table}[t]
386: \centerline{
387: \begin{tabular}{|c|c|c|c|c|c|c|}
388: \hline
389: name & irrep & count & &
390: name & irrep & count\\
391: \hline
392: $q_i$ & $(\boldsymbol{3},\boldsymbol{2};\boldsymbol{1},\boldsymbol{1})_{1/6}$ & 3 & &
393: $\bar u_i$ & $(\overline{\boldsymbol{3}},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{-2/3}$
394: & 3\\
395: $\bar d_i$ & $(\overline{\boldsymbol{3}},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{1/3}$
396: & 7 & &
397: $d_i$ & $(\boldsymbol{3},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{-1/3}$ & 4\\
398: $\bar\ell_i$ & $(\boldsymbol{1},\boldsymbol{2};\boldsymbol{1},\boldsymbol{1})_{1/2}$ & 5 & &
399: $\ell_i$ & $(\boldsymbol{1},\boldsymbol{2};\boldsymbol{1},\boldsymbol{1})_{-1/2}$ & 8\\
400: $m_i$ & $(\boldsymbol{1},\boldsymbol{2};\boldsymbol{1},\boldsymbol{1})_{0}$ & 8 & &
401: $\bar e_i$ & $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{1}$ & 3 \\
402: $s^-_i$ & $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{-1/2}$ & 16 & &
403: $s^+_i$ & $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{1/2}$ & 16\\
404: $s^0_i$ & $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{1},\boldsymbol{1})_{0}$ &
405: 69
406: & & $h_i$ &
407: $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{1},\boldsymbol{2})_{0}$& 14\\
408: $f_i$ & $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{4},\boldsymbol{1})_{0}$ & 4 & &
409: $\bar f_i$ &
410: $(\boldsymbol{1},\boldsymbol{1};\overline{\boldsymbol{4}},\boldsymbol{1})_{0}$ & 4 \\
411: $w_i$ & $(\boldsymbol{1},\boldsymbol{1};\boldsymbol{6},\boldsymbol{1})_{0}$ &
412: 5 & & & &\\
413: \hline
414: \end{tabular}
415: }
416: \caption{The $G_\mathrm{SM}\times[\SO{6}\times\SU2]$ quantum numbers
417: of the spectrum. The hypercharge is indicated by the subscript.}
418: \label{tab:Spectrum}
419: \end{table}
420:
421: A key ingredient is the appearance of the `local GUTs'. In
422: Tab.~\ref{tab:LocalGUTs} we list the local gauge groups together with the local
423: $T_1$ representations. As discussed in Section \ref{sec:2+1} our shift $V_6$ is
424: chosen so as to produce two local $\boldsymbol{16}$--plets at
425: $n_2=n_3=0,\ n_2'=0,1$. Note that these are the only $T_1$ fields at $n_2=0$
426: which transform non-trivially under $G_\mathrm{SM}$. At $n_2=1$, there are
427: $\boldsymbol{4}$--plets and $\overline{\boldsymbol{4}}$--plets w.r.t.\ \SO6
428: subgroups of the first \E8. Although these \SO6 subgroups are
429: embedded into \E8 differently for different $n_3$, each $\boldsymbol{4}$--plet or
430: $\overline{\boldsymbol{4}}$--plet gives rise to an $\SU2_\mathrm{L}$ doublet
431: with zero hypercharge ($m_i$ of Tab.~\ref{tab:Spectrum}) and two non--Abelian
432: singlets with opposite hypercharge ($s^\pm_i$ of Tab.~\ref{tab:Spectrum}). In
433: particular, apart from the two $\boldsymbol{16}$--plets the $T_1$
434: states are vector--like w.r.t.\ $G_\mathrm{SM}$.
435:
436: \begin{table}[t]
437: \centerline{\begin{tabular}{|l|l|l|}
438: \hline
439: & $n_2=0$ & $n_2=1$ \\
440: \hline
441: $n_3=0$
442: &
443: $\SO{10}\times\SU2^2\times\left[\SO{14}\right]$
444: &
445: $\SO8\times\SO6\times\left[\SU7\right]$
446: \\
447: &
448: $\left(\boldsymbol{16},\boldsymbol{1},\boldsymbol{1};\boldsymbol{1}\right)$
449: &
450: $\left(\boldsymbol{1},\boldsymbol{4};\boldsymbol{1}\right)$
451: \\
452: &
453: $\oplus\ 2\times
454: \left(\boldsymbol{1},\boldsymbol{2},\boldsymbol{1};\boldsymbol{1}\right)
455: \oplus\left(\boldsymbol{1},\boldsymbol{1},\boldsymbol{2};\boldsymbol{1}\right)$
456: &
457: \\
458: \hline
459: $n_3=1$ & $\SO{12}\times\left[\SO8\times\SO6\right]$ &
460: $\SO8'\times\SO6'\times\left[\SU7\right]$\\
461: & $\left(\boldsymbol{1};\boldsymbol{8},\boldsymbol{1}\right)
462: \oplus \left(\boldsymbol{1};\boldsymbol{1},\boldsymbol{4}\right)$
463: & $\left(1,\boldsymbol{4};\boldsymbol{1}\right)$ \\
464: \hline
465: $n_3=2$ & $\SU7\times\left[\SO8\times\SO6\right]$
466: & $\SO8''\times\SO6''\times\left[\SO{10}\times\SU2^2\right]$\\
467: & $\left(\boldsymbol{1};\boldsymbol{1},\overline{\boldsymbol{4}}\right)$
468: & $\left(\boldsymbol{1},\overline{\boldsymbol{4}};\boldsymbol{1},
469: \boldsymbol{1},\boldsymbol{2}\right)$ \\
470: \hline
471: \end{tabular}}
472: \caption{Local GUTs and local representations. The
473: geometric interpretation of the `localization quantum numbers' $n_2$ and $n_3$
474: is given in Fig.~\ref{fig:LocalGaugeGroups}.}
475: \label{tab:LocalGUTs}
476: \end{table}
477:
478: To understand the origin of the third generation, recall that the shift $V_6$
479: breaks the first $\E8$ to $\SO{10}\times\SU2^2\times\U1$. Under this breaking,
480: some internal components of the gauge bosons transforming in the coset
481: $\E8/[\SO{10}\times\SU2^2\times\U1]$ remain massless. An important property of
482: this coset is that it contains
483: $\boldsymbol{16}$--plets. In the $U_1$ and $U_2$ sectors we obtain
484: \SO{10} $\boldsymbol{16}$--plets, while in the $U_3$ sector we obtain \SO{10}
485: $\boldsymbol{10}$--plets. Here, the $U_1$, $U_2$ and $U_3$ sectors consist of
486: the gauge bosons with spatial
487: components in the direction of the \G2, \SU3, and \SO4
488: torus, respectively. Due to the presence of Wilson lines, these
489: \SO{10} $\boldsymbol{16}$--plets and $\boldsymbol{10}$--plets are
490: subject to further projections such that we finally obtain
491: $(\overline{\boldsymbol{3}},\boldsymbol{1})_{-2/3}$ and
492: $(\boldsymbol{1},\boldsymbol{1})_{1}$ w.r.t.\ $G_\mathrm{SM}$ from $U_1$, and
493: $(\boldsymbol{3},\boldsymbol{2})_{1/6}$ from $U_2$. These states correspond to
494: a $\boldsymbol{10}$--plet of \SU5.
495: Together with localized states from higher
496: twisted sectors, one obtains an additional complete generation. This
497: seemingly miraculous completion of the third generation is related to the
498: necessary SM anomaly cancellation.
499:
500: In summary, the massless spectrum contains three generations, two of which
501: originate from the $T_1$ sector and one being a mixture of states from
502: $U$, $T_2$ and $T_4$. All other states are \emph{vector-like} w.r.t.\
503: $G_\mathrm{SM}$. Thus we have
504: \begin{equation}
505: \text{matter:}~~~~3\times\boldsymbol{16} ~~+~~ \text{vector-like} \;.
506: \end{equation}
507:
508: One of the main problems of most known string models is the presence of exotic
509: states at low energies. In our model the vector--like exotic states
510: do not appear at low energies. Their mass terms are
511: generated by the superpotential involving singlet fields and consistent with string selection rules
512: \cite{Hamidi:1986vh,*Dixon:1986qv,Font:1988tp,*Font:1988mm},
513: \begin{equation}
514: W~\supset~x_i \bar x_j \langle s_a^0 s_b^0 \dots \rangle \;.
515: \end{equation}
516: The singlet fields acquire large VEVs yielding masses for all
517: exotic states \cite{Buchmuller:2005jr,Buchmuller:2005pr}. Recently, an
518: interesting class of heterotic string compactifications has been constructed
519: where massless exotic states are absent from the beginning
520: \cite{Braun:2005ux,*Bouchard:2005ag,*Braun:2005nv}.
521: Their relation to the model described above remains to be clarified.
522:
523:
524: \begin{figure}[t]
525: \centerline{\CenterObject{\includegraphics{GaugeYukawaUnification2.eps}}}
526: \caption{Illustration of gauge--Yukawa unification. The plot shows the MSSM
527: evolution of $\alpha_i=g_i^2/(4\pi)$ and $\alpha_t=y_t^2/(4\pi)$ where $g_i$
528: denotes the gauge couplings and $y_t$ is the top Yukawa coupling.}
529: \label{fig:GaugeYukawaUnification}
530: \end{figure}
531:
532:
533: \subsection{Unification of couplings and flavour structure}
534:
535: Our model has no exotic states at low energies and admits TeV scale soft
536: masses.
537: Therefore, it is consistent with gauge coupling unification.
538: Furthermore, the top--Yukawa coupling is related to the gauge
539: coupling since the third generation originates partially from the untwisted
540: sector. As described above, the $U_3$ zero modes descend from an
541: \SO{10} $\boldsymbol{10}$--plet, and give rise to two states $\ell_1$ and
542: $\bar\ell_1$ with the quantum numbers of the MSSM Higgs doublets. The
543: up--quark and the quark doublet of the third generation come from the
544: untwisted sectors $U_1$ and $U_2$, respectively. The Yukawa coupling
545: $U_1\,U_2\,U_3$ is a gauge interaction in 10D and its strength
546: is given by the gauge coupling at the unification scale.
547: One thus obtains the
548: superpotential coupling
549: \begin{equation}
550: W~\supset~y_t\,q_3\,\bar u_3\,h_u\;,
551: \end{equation}
552: with $y_t\simeq g$ at the GUT scale, as long as the MSSM `up--type' Higgs is
553: dominated by the $U_3$ state, $h_u\simeq\bar\ell_1$
554: (cf.~Fig.~\ref{fig:GaugeYukawaUnification}). It is well known that this is
555: consistent with the measured top mass. The Yukawa couplings involving
556: the first two generation up--type quarks occur only through higher
557: dimensional operators \cite{Buchmuller:2005jr}, which leads to the
558: required suppression of these couplings.
559:
560: The light down--type quarks $\bar d_i$ and lepton doublets $\ell_i$ are linear
561: combinations of states from various twisted and untwisted sectors. A similar
562: setup has recently been proposed in the context of orbifold GUTs
563: \cite{Asaka:2003iy}. There, it has been shown that a mixing of the light
564: leptons (down--type quarks) with additional heavy leptons
565: (down--type quarks) can lead to the observed large neutrino mixings
566: together with small CKM mixings in the quark sector. In our heterotic
567: string model an analogous mixing occurs.
568: A more
569: detailed analysis of phenomenological aspects of the model will be presented
570: elsewhere.
571:
572:
573:
574: \begin{figure}[t]
575: \centerline{\subfigure[\SO4 plane large.\label{fig:OGL1}]{%
576: \CenterObject{\includegraphics{pic6.eps}}}
577: %
578: \quad
579: %
580: \subfigure[\SU3 plane large.\label{fig:OGL2}]{%
581: \CenterObject{\includegraphics{pic7.eps}}}}
582: \centerline{\subfigure[\G2 plane large.\label{fig:OGL3}]{%
583: \CenterObject{\includegraphics{pic8.eps}}}}
584: \caption{Orbifold GUT limits. Only subgroups of the first \E8 are shown, and \U1
585: factors are omitted.}
586: \label{fig:OGL}
587: \end{figure}
588:
589:
590: \subsection{Orbifold GUT limits}
591:
592: One of the main motivations for revisiting orbifold compactifications of the
593: heterotic string is the phenomenological success of orbifold GUTs. It is
594: therefore interesting to study orbifold GUT limits of the model, which
595: correspond to anisotropic compactifications where some radii are significantly
596: larger than the others. Such anisotropy may mitigate the discrepancy
597: between the GUT and the string scales \cite{Witten:1996mz,*Hebecker:2004ce}.
598: In the energy range between the compactification scale and the string scale
599: one obtains an effective higher-dimensional field theory.
600:
601: We focus on $4+2$--dimensional orbifold GUT limits, and keep only subgroups
602: of the first \E8 factor in the subsequent analysis. As an example,
603: consider the case that the compactification radii of the \SO4 plane are larger than the others.
604: For energies in the range $R_{\SO4}^{-1}<E<R_{\G2,\SU3}^{-1}$ the model is
605: described by an effective six--dimensional theory. The local gauge groups
606: at the fixed points and the gauge group in the bulk are obtained by imposing
607: the following projection conditions on the gauge bosons labeled by $p$:
608: \begin{align}
609: \text{bulk} & :&& 2V_6\cdot p~\equiv~0\;, & W_3\cdot p~\equiv~0\;,
610: \nonumber\\
611: n_2=0 & :&& V\cdot p ~\equiv~0\;,& W_3\cdot p~\equiv~0\;,
612: \nonumber\\
613: n_2=1 & : && (V+W_2)\cdot p ~\equiv~0\;, & W_3\cdot p~\equiv~0\;.
614: \end{align}
615: This leads to the bulk group \SU6 and
616: to the `local GUTs' \SU5 at $n_2=0$ and $\SO6\times\SU2_\mathrm{L}$ at $n_2=1$,
617: respectively (cf.\ Fig.~\ref{fig:OGL} (a)). The two fixed points with $n_2=0$
618: carry one generation each. The appearance of these complete generations cannot
619: be fully understood in the 6D orbifold GUT limit. To have deeper
620: understanding one has to zoom into the smaller dimensions and unravel the
621: underlying \SO{10} gauge symmetry. An important property of this limit is that
622: the bulk \SU6 is a simple group containing $G_\mathrm{SM}$. Hence
623: the running of the gauge
624: couplings in six dimensions does not discriminate between the subgroups of
625: $G_\mathrm{SM}$. This supports the picture of gauge coupling unification at the
626: compactification scale.
627:
628: An analogous analysis can be carried out when the \SU3 torus is larger than the
629: other tori. In this case, one obtains the orbifold GUT shown in Fig.~\ref{fig:OGL} (b)
630: as an effective theory. The first two generations are localized at the fixed
631: point with the Pati--Salam group
632: $G_\mathrm{PS}=\SO6\times\SU2_\mathrm{L}\times\SU2_\mathrm{R}$
633: as an unbroken local GUT. Again, the bulk
634: group is simple and contains all SM group factors.
635:
636: In the limit where the \G2 torus is large, the first two generations reside at
637: the fixed point with unbroken gauge symmetry $G_\mathrm{SM}$
638: (cf.~Fig.~\ref{fig:OGL} (c)). In this case, $\SU3_C\times\SU2_\mathrm{L}$ is a
639: subgroup of the bulk group $\SU6$, but hypercharge is not fully contained in
640: this \SU6. However, this does not lead to the running or threshold corrections
641: which discriminate between the hypercharge and the simple SM subgroups because
642: of $N=4$ supersymmetry in the bulk, implying vanishing
643: $\beta$--functions.
644:
645: In summary, we find that all orbifold limits are consistent with gauge coupling
646: unification up to possible logarithmic corrections coming from states localized
647: at the fixed points. The different geometries have, however, important
648: consequences for the values of the gauge couplings at the compactification
649: scale and also for the pattern of the Yukawa couplings. These issues will be
650: analyzed in more detail elsewhere.
651:
652: \section{Summary}
653:
654: The quest for unification is a central theme of particle physics. In a
655: bottom--up approach, starting from the standard model, one is first
656: led to conventional 4D GUTs.
657: Among the GUT groups, \SO{10} is
658: singled out since one matter generation, including
659: the right--handed neutrino, forms an irreducible
660: representation of \SO{10}, the $\boldsymbol{16}$--plet. The route of
661: unification, continuing via exceptional groups, terminates at \E8.
662:
663: An attractive starting point for a unified theory including gravity
664: is, in a top--down approach, the heterotic string with the gauge group
665: $\E8\times\E8$.
666: In its orbifold compactifications, GUT groups appear locally at orbifold
667: fixed points.
668: We have
669: presented an example with local \SO{10} symmetry and localized
670: $\boldsymbol{16}$--plets from the twisted sectors. The standard model
671: gauge group is obtained as an intersection of different local \E8
672: subgroups. The model has three quark--lepton generations, one pair
673: of Higgs doublets and no exotic matter.
674:
675:
676:
677:
678: \vspace*{0.5cm}\noindent
679: {\bf Acknowledgements.} It is a pleasure to thank T.~Kobayashi, M.~Lindner,
680: J.~Louis, H.~P.~Nilles and S.~Stieberger for valuable discussions.
681: W.~B. thanks the organizers of the GUSTAVOFEST for arranging an enjoyable
682: and stimulating Symposium in Honor of Gustavo~C.~Branco.
683: M.~R. is deeply indebted to A.~Wisskirchen for technical support. This work was
684: partially supported by the European Union 6th Framework Program
685: MRTN-CT-2004-503369 ``Quest for Unification'' and MRTN-CT-2004-005104
686: ``ForcesUniverse''.
687:
688:
689: %\clearpage
690: \bibliography{Orbifold}
691: \addcontentsline{toc}{section}{Bibliography}
692: \bibliographystyle{ArXivmciteW}
693:
694: \end{document}
695:
696: