1: \documentclass[12pt]{article}
2: \usepackage{graphicx, cite, epsf, amssymb,epsfig}
3: \textheight21cm
4: %
5: %
6: \newsavebox{\LSIM}
7: \sbox{\LSIM}{\raisebox{-1ex}{$\ \stackrel{\textstyle<}{\sim}\ $}}
8: \newcommand{\lsim}{\usebox{\LSIM}}
9: \newsavebox{\GSIM}
10: \sbox{\GSIM}{\raisebox{-1ex}{$\ \stackrel{\textstyle>}{\sim}\ $}}
11: \newcommand{\gsim}{\usebox{\GSIM}}
12: %
13: %
14: \begin{document}
15: %
16: % T I T L E P A G E
17: %
18: \begin{titlepage}
19: \begin{flushright}
20: BA-01-50\\
21: DESY 01-201\\
22: hep-ph/0111465
23: \end{flushright}
24: $\mbox{ }$
25: \vspace{.1cm}
26: \begin{center}
27: \vspace{.5cm}
28: {\bf\large Kaluza-Klein Excitations of W and Z at the LHC?} \\[.3cm]
29: \vspace{1cm}
30: Stephan J. Huber$^{b,}$\footnote{stephan.huber@desy.de},
31: Chin-Aik Lee$^{a,}$\footnote{jlca@udel.edu}
32: and
33: Qaisar Shafi$^{a,}$\footnote{shafi@bartol.udel.edu} \\
34:
35: \vspace{1cm} {\em
36: $^a$Bartol Research Institute, University of Delaware, Newark, USA}\\[.2cm]
37: {\em $^b$Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany}
38: \end{center}
39: \bigskip\noindent
40: \vspace{1.cm}
41: %
42: %
43: %
44: %
45: \begin{abstract}
46: Deviations from standard electroweak physics arise in the framework
47: of a
48: Randall-Sundrum model, with matter and gauge fields in the bulk and the
49: Higgs field localized on the TeV brane. We focus in particular on
50: modifications associated with the weak mixing angle. Comparison with the
51: electroweak precision data yields a rather stringent lower bound of about
52: 10 TeV on the masses of the lowest Kaluza-Klein excitation of the W and Z
53: bosons. With some optimistic assumptions the bound could be lowered to about
54: 7 TeV.
55: \end{abstract} \end{titlepage}
56:
57: \section{Introduction}
58: The standard model (SM), as formulated for a flat
59: four-dimensional space-time, can be easily generalized
60: to the RS model \cite{1,CHNOY,GP,DHR,HS2}. In the
61: RS model, the four dimensional world
62: around us arises from the compactification
63: of a curved 5D geometry. The fifth dimension is a
64: $S_1/\mathbb{Z}_2$ orbifold and contains two 4D branes at the
65: orbifold fixed points at $y=0$ and $y=\pi R$.
66: The 5D geometry is a slice of AdS with the line element
67: $ds^2=dy^2+e^{-2\sigma(y)}\eta_{\mu\nu}dx^\mu dx^\nu$,
68: where $\sigma(y)=k|y|$, $k$ measures the
69: curvature along the fifth dimension, $x$ represents the familiar
70: four dimensions and $y$ represents the fifth dimension. The brane at $y=0$
71: is called the Planck-brane
72: and the brane at $y=\pi R$ is called the TeV-brane
73: because the typical mass scale on each of the branes is of the order
74: of the effective 4D Planck mass and a TeV respectively. This exponential
75: hierarchy of energy scales is
76: generated by the warp factor $\Omega=e^{-\pi k R}$, with $kR\simeq 11$,
77: and provides a new solution to the gauge hierarchy problem.
78:
79: By compactifying the fifth dimension, a field living in the
80: bulk, i.e.~in the full 5D space-time,
81: can be decomposed into an infinite number of 4D fields with
82: different effective 4D masses using a method known as the
83: Kaluza-Klein (KK) decomposition. Depending on the choice of $k$ and $R$,
84: there is a large gap between the mass of the
85: lightest KK mode of a bulk field and its next excited state.
86: The mass of the first excited state is typically of the
87: order of 10 TeV (since experimental constraints rule out
88: excited states which are significantly lighter), while the mass
89: of the ground state is constrained by experimental data \cite{4,Wa01},
90: since such fields correspond to the ones we
91: actually observe in accelerators.
92:
93: Bulk fields allow one to address problems related to non-renormalizable
94: operators within the RS framework, fermion masses
95: and mixings
96: \cite{GP,HS2}, neutrino masses \cite{GN,HS3} and gauge coupling
97: unification \cite{Po,RS01}. With bulk gauge fields, the couplings
98: and masses of the weak gauge bosons deviate from their SM values.
99: In ref.~\cite{HS1} the modifications of couplings and masses
100: were treated independently of each other. This is
101: a good approximation if the SM fermions live close
102: to one of the two branes. In this letter we present a
103: combined analysis of the two effects in order to cover the case
104: where the fermions are weakly localized or delocalized
105: in the extra dimension as well. Weakly localized fermions are expected for
106: instance, if the warped geometry also induces
107: the fermion mass hierarchy \cite{HS2}.
108:
109: We find that the lowest KK excitations of gauge bosons and fermions
110: have to be heavier than about 10 TeV. There is no window
111: for ``light'' KK states even for delocalized fermions, as was
112: the case for the constraints derived in refs.~\cite{GP,DHR}.
113: Hence, it is questionable if these excitations can be
114: directly studied at the LHC. Indirect evidence for extra
115: dimensions, e.g.~rare processes such as $n$--$\bar n$
116: oscillations or $\mu \rightarrow e+\gamma$ \cite{K,HS3}, are
117: therefore very important. Nevertheless, one may speculate
118: if the scenario presented here could account for the small
119: deviations of the SM predictions from recent electroweak precision
120: data \cite{W01,M01}. This might allow a lowering of the bound to around 7
121: TeV.
122:
123:
124: \section{The Kaluza-Klein reduction}
125: With some slight modifications, the SM is
126: easily embedded in a warped geometry. All of the fields are
127: assumed to live in the bulk, except the Higgs field, which is
128: confined to the TeV-brane \cite{CHNOY,HS1}. (This can be
129: considered as an approximation to the case where the Higgs field lives
130: in the bulk but is concentrated exponentially around
131: the TeV-brane. The ground state of a scalar field behaves as $e^{4\sigma}$
132: and for values of $kR$ around 11, this
133: approximation is accurate up to a few permille \cite{HS3}.)
134:
135: \begin{figure}[t]
136: \begin{picture}(100,230)
137: \put(95,-40){\epsfxsize7cm \epsffile{gaugewaves.eps}}
138: \put(270,210){{TeV-brane}}
139: \put(70,210){{Planck-brane}}
140: \put(70,140){$f_n\uparrow$}
141: \put(300,100){$\longrightarrow$}
142: \put(330,109){$y$}
143: \put(322,91){[GeV]$^{-1}$}
144: \put(200,120){$n=0$}
145: \put(300,170){$n=1$}
146: \put(300,40){$n=2$}
147: \end{picture}
148: \caption{The ground state and the first two excited states of the $Z$ boson when
149: $kR=10.83$, $k=\overline{M}_{\rm Pl}=2.44\times 10^{18}$ GeV and $a_Z$=0.1849.}
150: \label{f_1}
151: \end{figure}
152:
153: The masses
154: of the weak gauge bosons are generated by spontaneous
155: symmetry breaking arising from the Higgs mechanism,
156: $M^2(y)=a^2k\delta(y-\pi R)$, where $a$ is a dimensionless
157: parameter which is determined by how strongly the gauge boson
158: couples to the Higgs field. Since there is no large hierarchy
159: between the weak and KK scales, the 5D mass term
160: $M^2$ should be included in the KK reduction of the weak
161: gauge bosons from the very beginning \cite {HS1}. The masses
162: and 5D wave functions of a gauge field are obtained from its
163: 5D equations of motion,
164: $\frac{1}{\sqrt{-g}}\partial_M(\sqrt{-g}g^{MN}g^{RS}F_{NS})-
165: M^2(y)g^{RS}A_S=0$.
166: Here, the $A_4=0$ gauge is imposed and this is only possible because the
167: Higgs field is localized on a brane. If the Higgs were to propagate in the
168: bulk instead, it would not be possible to impose both the unitarity and the
169: $A_4=0$ gauge simultaneously and we would have to contend with
170: $A_4$
171: excitations as well. Using the method of
172: separation of variables, a gauge field can be decomposed as follows:
173: \begin{equation}A_\mu({x},y)=\frac{1}{\sqrt{2\pi R}}\sum^{\infty}_{n=0}
174: A_\mu^{(n)}({x})f_n(y),\label{KK}\end{equation} where
175: $A_\mu^{(n)}$ satisfies the field equation for a gauge boson of mass
176: $m_n$. This expansion is only valid in the weak coupling limit for
177: non-Abelian gauge fields. However, if we drop the requirement that the
178: $A_\mu^{(n)}$'s satisfy the 4D equations of motion for a Yang-Mills field,
179: any field configuration can be decomposed according to eq. (\ref{KK}). The
180: wave functions $f_n$ follow
181: from $(\partial _y^2-2\sigma'\partial_y-M^2+e^{2\sigma}m_n^2)f_n=0$,
182: with $\sigma'=d\sigma/dy$. The $f_n$ are normalized by the
183: condition $\frac{1}{2\pi R}\int^{\pi R}_{-\pi
184: R}f_n(y)^2dy=1$. The solution is given by \cite{DHRP}
185: \begin{equation}f_n=
186: \frac{e^\sigma(J_1(\frac{m_n}{k}e^\sigma)+b_n
187: Y_1(\frac{m_n}{k}e^\sigma))}{N_n},\end{equation}
188: where $N_n$
189: is the normalization constant, and $b_n$ and $m_n$ are
190: obtained by solving the following system of equations \cite{HS1}:
191: \begin{eqnarray}
192: b(m,a^2)&=&\frac{-\frac{a^2}{2}J_1(\frac{m}{k})+
193: \frac{m}{k}J_0(\frac{m}{k})}{-\frac{a^2}{2}Y_1(\frac{m}{k
194: })+\frac{m}{k}Y_0(\frac{m}{k})}, \label{3}
195: \\[.2cm]
196: b_n(e^{-\pi kR}k x_n,0)&=&b(k x_n,-a^2), \nonumber
197: \end{eqnarray}
198: where $x_n=e^{\pi
199: kR} m_n/k$. In fig.~\ref{f_1} we present the wave functions
200: of the ground state and the first two excited states of the
201: $Z$ boson.
202:
203: Bulk fermions are described by the 5D equation of motion
204: $(g^{MN}\gamma_M(\partial_M+\Gamma_M)+m_\Psi)\Psi=0$, where
205: $m_\Psi=c\sigma'$ and $\Gamma_M$ is the spin connection in the
206: tetrad formulation. The Dirac mass term $m_\Psi$ takes on the functional
207: form it has because of the parity restriction $\Psi(-y)=\pm\gamma_5\Psi(y)$.
208: For the same reason, the KK
209: ground state of a fermion can only be either left-handed or right-handed.
210: These ground states are identified with the
211: fermion fields actually observed by experiments.
212: Again, the 5D field is decomposed into a KK tower \cite{GN,GP},
213: \begin{equation}\Psi({x},y)=\frac{1}{\sqrt{2\pi
214: R}}\sum^\infty _{n=0}\Psi^{(n)}({x})f^c_n(y),
215: \end{equation}
216: where $\Psi^{(n)}({x})$ obeys the equation of motion for a
217: fermion of mass $m_n$,
218: \begin{equation}f^c_0(y)=\frac{e^{(2-c)\sigma}}{N_0},\end{equation}
219: and
220: \begin{equation}f^c_n(y)=\frac{e^{\frac{5}{2}\sigma}(J_\alpha(\frac{m_n}{k}e^\sigma)+b_n
221: Y_\alpha(\frac{m}{k}e^\sigma))}{N_n},\end{equation} for $n>0$.
222: The wave functions $f^c_n$ are normalized by
223: \begin{equation}\int^{\pi
224: R}_{-\pi R}\frac{e^{-3\sigma}f^c_n(y)^2}{2\pi kR}dy=1.\end{equation}
225: Fig.~\ref{f_2} shows the wave function for the ground state fermions
226: for different values of $c$. For $c>$1/2 ($<$1/2) the zero mode is localized
227: towards the Planck-brane (TeV-brane). For $c$ close to 1/2 the zero mode is
228: only weakly localized or even delocalized ($c=1/2$).
229:
230: \begin{figure}[t]
231: \begin{picture}(100,190)
232: \put(60,-50){\epsfxsize8cm \epsffile{f0waves.eps}}
233: \put(255,180){{TeV-brane}}
234: \put(90,180){{Planck-brane}}
235: \put(65,130){$e^{-\frac{3}{2}\sigma}f_0\uparrow$}
236: \put(298,9){$\longrightarrow$}
237: \put(328,16){$y$}
238: \put(320,-2){[GeV]$^{-1}$}
239: \put(190,40){\small $c=0.5$}
240: \put(295,140){\small$c=0$}
241: \put(230,60){\small$c=0.25$}
242: \put(130,120){\small$c=1$}
243: \put(140,60){\small$c=0.75$}
244: \end{picture}
245: \caption{The wave functions for the ground state fermions for different values of $c$.}
246: \label{f_2}
247: \end{figure}
248:
249:
250: \section{Measuring the weak mixing angle}
251:
252: Experimentally, there are several ways to measure the weak mixing
253: angle, $\theta_W$. Although the equations of the SM remain unchanged
254: in 5D, the effective 4D behavior of the fields is slightly modified.
255: In particular, measurements of $\theta_W$ made on
256: the assumption of a flat 4D space-time would result in a
257: value different from the actual 5D weak mixing angle, $\theta_5$.
258: The weak mixing angle can be calculated from the ratio
259: of the masses of the $W$ and $Z$ bosons,
260: \begin{equation}\label{8}
261: \cos^2{\theta_1}=\frac{m_W^2}{m_Z^2},
262: \end{equation}
263: where the subscript 1 indicates that this is the first possible
264: definition. Alternatively, $\theta_W$
265: can also be obtained from the corresponding ratio of the gauge
266: coupling strengths. In the SM, both definitions are equivalent
267: (at the tree level).
268:
269: In the framework of warped models, deviations from standard 4D electroweak
270: physics arise.
271: The relationship between the (5D) gauge coupling strengths and the
272: physical gauge boson mass is no longer linear because of
273: eq.~\ref{3} \cite{HS1}. As a result, the $W$ mass gets shifted
274: upward compared to the SM.\footnote{
275: Curiously, the new experimental value for $m_W$ is indeed
276: slightly above the SM prediction \cite{W01}.
277: }
278: Furthermore, the couplings of the weak gauge bosons
279: to fermions are modified. In particular, they become dependent
280: on the localization of the fermion within the extra dimension \cite{HS1}.
281: The effective 4D coupling between the ground state of a
282: fermion and the $n^{th}$ excited state of a gauge boson is
283: given by \cite{GP}
284: \begin{equation} \label{9}
285: g^{(n)}=\frac{g^{(5)}}{(2\pi R)^{3/2}}\int^{\pi R}_{-\pi R}
286: e^{-3\sigma}f^c_0(y)^2f_n(y)~dy,
287: \end{equation}
288: where $g^{(5)}$ is the 5D gauge coupling.
289:
290:
291: In the approach presented above electroweak symmetry
292: breaking is treated in the 5D framework. Corrections to
293: weak gauge boson masses and gauge coupling strengths
294: arise automatically in the KK reduction. One can interpret
295: these results also as a mixing effect between a massless
296: ($y$ independent) zero mode of a gauge field and its ($y$
297: dependent) KK states.
298: The mixing between the weak gauge bosons and their KK
299: excitations is on the order of $m_W^2/M_{KK}^2$. This
300: explains why all deviations from SM physics are
301: proportional $1/M_{KK}^2$. Note that somewhat similar
302: results are obtained for the case of a flat extra dimension \cite{DPQ}.
303:
304:
305: In the following, we perform a {\em tree level} comparison
306: between the warped model with bulk fields and the ordinary
307: SM. This procedure is justified because SM-like quantum
308: corrections are essentially the same in both models. The
309: tiny modifications (of order $10^{-3}$) of the tree level
310: physics will only cause sub-leading corrections in the loops.
311: Equivalently, we can just as well work in the picture of a
312: gauge field zero mode mixing with its KK states.
313: In the treatment of the zero mode we then include the usual
314: SM quantum corrections, while the mixing with the excited
315: states is approximated at tree-level. In the literature this
316: approach has been successfully used to investigate the impact
317: of extra Z bosons on electroweak observables \cite{EL}, which is
318: very similar to what is discussed here.
319: In the warped model (like in models with extra Z bosons),
320: there are additional radiative corrections
321: coming from loops involving KK states (extra Z bosons).
322: However, their influence is small because of the large
323: masses (about 10 TeV) of the excited states \cite{EL}.
324:
325:
326: After electroweak symmetry breaking, the neutral gauge bosons
327: in the bulk mix via the 5D weak mixing angle
328: $\tan\theta_5=g_1^{(5)}/g_2^{(5)}$, where $g_2^{(5)}$
329: and $g_1^{(5)}$ denote the $SU(2)$ and $U(1)$ gauge coupling strengths in
330: five dimensions respectively. The gauge coupling strength of the $Z$
331: boson in 5D is then given by $g_Z^{(5)}=g_2^{(5)}/\cos \theta_5$.
332: Analogous relations hold for the $W$ boson and the photon.
333: The effective 4D gauge couplings to fermions, $g_W$, $g_Z$ and
334: $g_{\gamma}$ are obtained from
335: the $n=0$ case of eq.~\ref{9}, using the relevant 5D gauge
336: coupling strengths.
337: From these quantities, the weak mixing angle can be defined as
338: \begin{equation} \label{10}
339: \cos^2\theta_2=\frac{g_W^2}{g_Z^2},
340: \end{equation}
341: or
342: \begin{equation} \label{11}
343: \cos^2\theta_3=1-\frac{g_{\gamma}^2}{g_W^2}.
344: \end{equation}
345: Experimentally, these definitions can be regarded
346: as comparisons of the strengths of the charged and neutral
347: current, and of the electric and charged current respectively.
348:
349: In the 4D SM, the values of the weak mixing angle inferred from
350: eqs.~\ref{8}, \ref{10} and \ref{11} agree, of course. Since in the
351: warped model, the gauge coupling strengths and gauge bosons
352: masses are obtained from eqs.~\ref{3} and \ref{9}, this
353: is no longer the case.
354: The differences between these angles, as measured by
355: \begin{equation} \label{deltas}
356: \Delta_{ij}=\cos^2\theta_i-\cos^2\theta_j,
357: \end{equation}
358: is too small to be
359: measured, given the current range of uncertainty in the
360: experimental data. However, upper bounds can be set on the
361: $\Delta$'s based on the range of uncertainty. This in turn
362: provides important constraints on the parameters of the warped SM
363: model. Rather than listing the permissible range of values for the
364: internal parameters of the model, a more transparent way of expressing
365: these constraints
366: from an experimental perspective is to
367: provide a lower bound for the masses of the first
368: KK excitation of the weak gauge bosons.
369:
370: \begin{figure}[t]
371: \begin{picture}(100,210)
372: \put(50,-50){\epsfxsize9cm \epsffile{rizzo.eps}}
373: \put(325,10){{$c~\longrightarrow$}}
374: \put(45,120){$M_{KK}\uparrow$}
375: \put(45,105){[TeV]}
376: \put(140,60){{\small $\Delta_{12}$}}
377: \put(143,85){{\small $\Delta_{13}$}}
378: \put(173,140){{\small $\Delta_{23}$}}
379: \put(143,30){{\small \cite{DHR}}}
380: \end{picture}
381: \caption{The three lower bound constraints and the limit of
382: ref.~\cite{DHR} for $k=M_5$ as a function of $c$.}
383: \label{f_3}
384: \end{figure}
385:
386: \section{Numerical analysis}
387:
388: The graphs in this section show the constraints on the mass of the
389: first KK excitation for different values of $k/M_5$, where
390: $M_5$ (the Planck mass in 5D) has been chosen to fit the observed
391: value for the Planck mass in 4D, $M_{\rm Pl}^2\approx M_5^3/k$ \cite{1}.
392: In each of the graphs, the lower bounds for the
393: mass of the excited $Z$ boson, $M_{KK}\equiv M_Z^{(1)}$ (the
394: difference in mass between
395: the excited $Z$ and $W$ bosons is insignificant) arising from
396: each of the three constraints are plotted.
397: The mass of the first
398: KK state is adjusted in our model by varying the radius of the
399: orbifold. In all the following graphs, we use the experimental
400: constraints $\Delta_{12}\leq 1.2\times 10^{-3}$,
401: $\Delta_{13}\leq 1.2\times 10^{-3}$ and $\Delta_{23}\leq 1.6\times
402: 10^{-4}$, based on data from the Particle Data Group \cite{4}.
403: The bound on $\Delta_{12,13}$ is dominated by the experimental
404: error of the $W$ mass, $M_W=80.419\pm0.056$ GeV, which
405: results in an uncertainty in $\cos^2\theta_1$ of $1.1\times 10^{-3}$.
406: For the weak mixing angle, we take $\sin^2\theta_W=0.23117\pm0.00016$ with
407: the
408: error providing the constraint on $\Delta_{23}$.
409:
410: \begin{figure}[t]
411: \begin{picture}(100,190)
412: \put(80,-40){\epsfxsize7cm \epsffile{mKK.01.eps}}
413: \put(300,5){{$c~\longrightarrow$}}
414: \put(45,140){$M_{KK}\uparrow$}
415: \put(45,125){[TeV]}
416: \put(133,80){{$\Delta_{13}$}}
417: \put(133,35){{$\Delta_{12}$}}
418: \put(139,150){{$\Delta_{23}$}}
419: \end{picture}
420: \caption{The three lower bound constraints for $k=M_5/100$ as a function of $c$.}
421: \label{f_4}
422: \end{figure}
423:
424: The numerical analysis is done in the following manner. We fix
425: the values of $kR$, $k/M_5$ and $c$. From the equation for the
426: KK gauge boson spectrum (\ref{3}), we determine the gauge
427: boson mass parameters, $a^2(W)$ and $a^2(Z)$ by fitting
428: the ground state mass to the experimental $W$ and $Z$
429: masses. The masses of the KK excitations are then also
430: fixed, as is the value of $\cos^2\theta_1$.
431: The 5D weak mixing angle follows from $\cos^2\theta_5=a^2(W)/a^2(Z)$.
432: With this information, we can calculate the ratios of the effective
433: gauge coupling strengths (\ref{9}) which enter the definitions of
434: $\cos^2\theta_{2,3}$. The deviations from SM physics
435: are then easily obtained from eq.~(\ref{deltas}).
436: We stress again that for different
437: values of the KK mass, the deviations scale as
438: $\Delta_{ij}\sim1/M_{KK}^2$.
439:
440: \begin{figure}[t]
441: \begin{picture}(100,190)
442: \put(80,-40){\epsfxsize7cm \epsffile{mKKD12.eps}}
443: \put(300,10){{$c~\longrightarrow$}}
444: \put(45,140){$M_{KK}\uparrow$}
445: \put(45,125){[TeV]}
446: \put(290,175){\small 1}
447: \put(290,142){\small 0.001}
448: \end{picture}
449: \caption{The constraints obtained from $\Delta_{12}$ for different values of
450: $k/M_5=1$, 0.1, 0.01, 0.001 (from top to bottom).}
451: \label{f_5}
452: \end{figure}
453:
454: In fig.~\ref{f_3} we display the three different constraints on
455: $M_{KK}$ as a function of the 5D fermion mass for $k/M_5=1$.
456: For fermions located close to the Planck-brane, $c>1/2$,
457: the strongest bound, $M_{KK}\gsim 11.1$ TeV, arises from
458: $\Delta_{12}$, i.e.~from the modification of the ratio of the
459: weak gauge boson masses. The constraint from $\Delta_{23}$,
460: which only involves the effective gauge coupling strengths, gives a
461: somewhat lower bound, $M_{KK}\gsim 7.8$ TeV.
462: This is because near the Planck-brane, the gauge boson wave
463: function is almost flat and independent of the gauge boson mass
464: \cite{HS1}. Therefore, $\Delta_{12}$ and $\Delta_{13}$ lead to
465: almost identical results. For $c\lsim1/2$, the fermions
466: become sensitive to the dip in the gauge boson wave function
467: at the TeV-brane \cite{HS1} and the gauge couplings deviate
468: from their SM values. The constraint from $\Delta_{23}$ takes
469: over and pushes the value of $M_{KK}$ far above 10 TeV. For fermions
470: strictly confined to the TeV-brane ($c\rightarrow-\infty$), we
471: find $M_{KK}\gsim 62$ TeV. For comparison, we also include the
472: lower bound on $M_{KK}$ from
473: ref.~\cite{DHR} in fig.~\ref{f_3}. It arises from the contribution of the
474: excited
475: gauge bosons to electroweak observables, like for example, effective
476: 4-fermion
477: interactions. Our bounds turn out to be much stronger for
478: all values of $c$.\footnote{This
479: justifies our neglect of loops containing KK excitations in the
480: comparison with the SM.}
481: In particular, $M_{KK}\sim 1$ TeV is
482: excluded even for fermions close to the Planck-brane. This is the
483: main result of the present paper.
484:
485: In fig.~\ref{f_4}, we give the corresponding results for $k/M_5=0.01$.
486: Qualitatively, the picture is unchanged, but the constraints on $M_{KK}$
487: are slightly looser by about 7 percent. For fermions localized close
488: to the Planck-brane, the bound from $\Delta_{12}$ on $M_{KK}$
489: is relaxed to 10.3 TeV. In general, the bound on $M_{KK}$ depends
490: logarithmically on $k/M_5$, with the relative change given approximately
491: by $0.015\ln(k/M_5)$. In fig.~\ref{f_5}, this behavior is demonstrated
492: for $\Delta_{12}$.
493:
494: So far we have only included a single 5D mass parameter $c$. If all
495: SM fermions have a common location in the extra dimension, the
496: constraint $c\lsim0.3$ applies. Otherwise the overlap of the fermion
497: wave functions and the Higgs is insufficient to generate the observed top
498: quark mass. Moreover, universal values of $c\gsim 1/2$ are disfavored by
499: deviations from universality \cite{DS}.
500: Requiring $c\lsim0.3$ for the universal $c$ parameter, the KK scale
501: has to be larger than at least 25 TeV,
502: making this scenario less attractive. However, in the SM there is no
503: symmetry that requires the $c$ parameters of different fermion
504: species to be degenerate. The $c$ parameters can be chosen in such
505: a way as to generate the fermion masses and mixings without
506: introducing hierarchies in the 5D Yukawa couplings \cite{HS2}.
507: The leptons and light quarks then turn out to have $c$ parameters
508: larger than 1/2. Since these particles are the ones used in experiments,
509: the 11 TeV bound on $M_{KK}$ applies also for this case.
510:
511: \section{Discussion and summary}
512: According to ref.~\cite{DHR}, the LHC can probe $M_{KK}$ up to
513: at most 7 TeV. From our results so far, it seems unlikely
514: that KK excitations of gauge bosons can be produced directly at
515: the LHC. Still, there may be interesting experimental signatures of the
516: presented scenario, like for instance, exotic processes such as
517: lepton flavor violation or $n-\bar n$ oscillations \cite{HS3}.
518:
519: One should keep in mind however, that the crucial input
520: in deriving the 10 TeV constraint is the tolerated deviation in the
521: $W$ mass. We took $\Delta M_W=56$ MeV \cite{4}. The current
522: experimental accuracy on the $W$ mass is about 34 MeV \cite{Wa01}.
523: If we were to take this value to be $\Delta M_W$, the quantities
524: $\Delta_{12,13}$
525: would have to be smaller than $0.86\times10^{-3}$. The lower bound
526: on $M_{KK}$ would increase to 13.1 TeV (for $k=M_5$).
527: Most interestingly, however, recent data seem to favor a $W$
528: mass
529: which is somewhat
530: above the SM prediction. At the $1\sigma$ level, the SM prediction
531: and the experimental result have no overlap \cite{W01}. The central
532: values differ by about 76 MeV for a Higgs mass of 115 GeV. Taking
533: this value for $\Delta M_W$ results in $M_{KK}\gsim 9.4$ TeV
534: (for $k=M_5$). For larger Higgs masses, the SM deviates even
535: more from experiment. For a Higgs mass of 250 GeV,
536: the SM prediction is about 130 MeV lower than the value found
537: experimentally.
538: This could be accounted for by KK gauge bosons with a mass of
539: 7 TeV. Thus, being very optimistic, KK gauge boson excitations may
540: be at the verge of being discovered at the LHC.
541:
542:
543: For $M_{KK}\sim7$ TeV, the modifications to the effective gauge coupling
544: strengths push $\Delta_{23}$ close to its allowed value, as can be
545: observed in fig.~\ref{f_3}. Since $\Delta_{23}$ is rather
546: sensitive to the location of fermions, experiments with different
547: quarks and/or leptons may actually result in different values for
548: $\sin^2\theta_W$. This may be of some relevance to the
549: $3.6 \sigma$ discrepancy in measuring $\sin^2\theta_{\rm eff}^{\rm lept}$ using
550: leptonic and hadronic asymmetries \cite{M01}.
551:
552: In more elaborate models using extra branes or non-trivial bulk
553: mass profiles multi-localization of gauge bosons and fermions is possible.
554: \cite{multi}. Then exceptionally light KK states can arise
555: which may escape the bounds we found in our analysis. To avoid
556: conflict with experiment these states have to decouple from
557: SM physics.
558:
559: In conclusion, our calculations suggest that the KK excitations of W and Z
560: within the RS model are expected to be rather heavy, of the
561: order of 10 TeV or so. With some caveats, this may be brought down to
562: about 7 TeV, which would make their discovery at the LHC much easier.
563:
564: \section*{Acknowledgements}
565: S.J.H. thanks S. Heinemeyer for a useful discussion, the Bartol Research
566: Institute for its hospitality during the crucial stages of this work and
567: the Alexander von Humboldt Stiftung for a Feodor-Lynen fellowship.
568: We thank Tony Gherghetta for useful comments on the draft and for bringing
569: ref.~\cite{DPQ} to our attention.
570:
571: This work was supported in part by DOE Contract DE-FG02-91ER40626.
572:
573: \begin{thebibliography}{5}
574:
575:
576: \bibitem{1} L. Randall and R. Sundrum, {\em Phys. Rev. Lett.} {\bf 83} (1999) 3370.
577:
578: \bibitem{CHNOY} S.~Chang, J.~Hisano, H.~Nakano, N.~Okada and Yamaguchi,
579: {\em Phys. Rev.} {\bf D62} (2000) 084025 [hep-ph/9912498].
580:
581: \bibitem{GP} T.~Gherghetta and A.~Pomarol,
582: {\em Nucl. Phys.} {\bf B586} (2000) 141 [hep-ph/0003129].
583:
584: \bibitem{DHR} H.~Davoudiasl, J.L.~Hewett and T.G.~Rizzo,
585: {\em Phys. Rev.} {\bf D63} (2001) 075004 [hep-ph/0006041].
586:
587: \bibitem{HS2} S.J.~Huber and Q.~Shafi, {\em Phys. Lett.} {\bf B498} (2001) 256
588: [hep-ph/0010196].
589:
590: \bibitem{4} D.E.~Groom {\em et al}., Particle Data Group, {\em Eur. Phys. J.}
591: {\bf C15} (2000) 1.
592:
593: \bibitem{Wa01}N.K.~Watson, hep-ph/0105230.
594:
595: \bibitem{GN} Y.~Grossman and M.~Neubert,
596: {\em Phys. Lett.} {\bf B474} (2000) 361 [hep-ph/9912408].
597:
598: \bibitem{HS3} S.J.~Huber and Q.~Shafi, {\em Phys. Lett.} {\bf B512} (2001) 365
599: [hep-ph/0104293].
600:
601: \bibitem{Po} A.~Pomarol, {\em Phys. Rev. Lett.} {\bf 85} (2000) 4004 [hep-ph/0005293].
602:
603: \bibitem{RS01} L.J.~Randall and M.D. Schwartz, hep-th/0108114, hep-th/0108115.
604:
605: \bibitem{HS1} S.J.~Huber and Q.~Shafi, {\em Phys. Rev.} {\bf D63} (2001) 045010
606: [hep-ph/0005286].
607:
608: \bibitem{K} R.~Kitano, {\em Phys. Lett.} {\bf B481} (2000) 39 [hep-ph/0002279].
609:
610: \bibitem{W01} G.~Weiglein, hep-ph/0108063.
611:
612: \bibitem{M01}The LEP Collaborations ALEPH, DELPHI, L3, OPAL, the LEP
613: Electroweak Working Group and the SLD Heavy Flavour and
614: Electroweak Groups, LEPEWWG/2001-01.
615:
616: \bibitem{DPQ} A.~Delgado, A.~Pomarol and M. Quiros,
617: {\em JHEP} {\bf 0001} (2000) 030 [hep-ph/9911252].
618:
619: \bibitem{EL}see e.g, J.~Erler and P.~Langacker, {\em Phys. Lett.} {\bf B456} (1999) 68
620: [hep-ph/9903476] and references therein.
621:
622: \bibitem{DHRP} H.~Davoudiasl, J.L.~Hewett and T.G.~Rizzo,
623: {\em Phys. Lett.} {\bf B473} (2000) 43;
624: A.~Pomarol, {\em Phys. Lett.} {\bf B486} (2000) 153.
625: \bibitem{DS} F.~del Aguila and J.~Santiago, {\em Phys. Lett.} {\bf B493} (200) 175
626: [hep-ph/0008143].
627:
628: \bibitem{multi}I.I.~Kogan, S.~Mouslopoulos, A.~Papazoglou and G.G.~Ross,
629: {\em Nucl. Phys.} {\bf B615} (2001) 191 [hep-ph/0107307];
630: F.~Del Aguila and J.~Santiago, hep-ph/0111047.
631: \end{thebibliography}
632:
633:
634:
635: \end{document}
636:
637: