1: \documentclass[showpacs,aps,prd,,nofootinbib,floats]{revtex4}
2: \bibliographystyle{u in order not to be obliged to extrapolate thensrt}
3: \usepackage{graphicx}
4: \usepackage{amssymb,latexsym}
5: \usepackage[draft=false]{hyperref}
6:
7: \def\lsim{\hbox{ \raise.35ex\rlap{$<$}\lower.6ex\hbox{$\sim$}\ }}
8: \def\gsim{\hbox{ \raise.35ex\rlap{$>$}\lower.6ex\hbox{$\sim$}\ }}
9: \def\setC{\mathbb{C}}
10: \def\setR{\mathbb{R}}
11:
12: \begin{document}
13:
14: \title{${D}$-term inflation, cosmic strings, and consistency with cosmic
15: microwave background measurements}
16:
17:
18: \author{Jonathan Rocher}
19: \email{rocher@iap.fr}
20: \affiliation{Institut d'Astrophysique
21: de Paris, {${\cal G}\setR\varepsilon\setC{\cal O}$}, FRE 2435-CNRS, 98bis
22: boulevard Arago, 75014 Paris, France.}
23:
24: \author{Mairi Sakellariadou} \email{msakel@cc.uoa.gr, mairi@mpej.unige.ch}
25: \affiliation{Division of Astrophysics, Astronomy, and Mechanics, Department of
26: Physics, University of Athens, Panepistimiopolis, GR-15784 Zografos, Hellas,
27: and \\ D\'epartement de Physique Th\'eorique, Universit\'e de Gen\`eve, 24
28: quai E. Ansermet, CH-1211 Gen\`eve 4, Switzerland.}
29:
30: \pacs{12.10.Dm, 98.80.Cq, 11.27.+d}
31:
32: \begin{abstract}
33: Standard D-term inflation is studied in the framework of
34: supergravity. D-term inflation produces cosmic strings,
35: however it can still be compatible with CMB measurements without
36: invoking any new physics. The cosmic strings contribution to the CMB
37: data is not constant, nor dominant, contrary to some previous
38: results. Using current CMB measurements, the free parameters (gauge
39: and superpotential couplings, as well as the Fayet-Iliopoulos term) of
40: D-term inflation are constrained.
41: \end{abstract}
42:
43: \maketitle
44:
45: \section{Introduction}
46:
47: The inflationary paradigm~\cite{infl} offers simple answers to the
48: shortcomings of the standard hot big bang model. In addition, simple
49: inflationary models offer successful candidates for the initial
50: density fluctuations leading to the observed structure formation. One
51: crucial question though is to answer how generic is the onset of
52: inflation~\cite{onset} and to find consistent and natural models of
53: inflation from the point of view of particle physics. One can argue
54: that the initial conditions which favor successful inflationary models
55: are the likely outcome of the quantum era before
56: inflation~\cite{onset}. It is more difficult however to find
57: natural ways to guarantee the flatness of the inflaton potential.
58:
59: The early history of the Universe at energies below the Planck scale
60: is described by an effective N=1 supergravity theory. Since inflation
61: should have taken place at an energy scale $V^{1/4}\lesssim 4\times
62: 10^{16}$ GeV, this implies that inflationary models should be
63: constructed in the framework of supergravity. Here is where the
64: problem arises: it is difficult to implement slow-roll inflation
65: within supergravity. The positive false vacuum of the inflaton field
66: breaks spontaneously global supersymmetry, which gets restored after
67: the end of inflation (when $V$ disappears). In supergravity theories,
68: the supersymmetry breaking is transmitted to all fields by gravity,
69: and thus any scalar field, which could play the r\^ole of the
70: inflaton, gets an effective mass $\sim\sqrt{8\pi V}/M_{\rm Pl}\sim H$,
71: where $H$ stands for the expansion rate during inflation, and $M_{\rm Pl}$
72: denotes the reduced Planck mass. This
73: problem, known as the problem of ``Hubble-induced mass'', originates
74: from F-term interactions and thus it is resolved if we consider the
75: vacuum energy as being dominated by non-zero D-terms of some
76: superfields~\cite{dterm}. This result led to a dramatic interest in
77: D-term inflation, since in addition this model can easily be implemented
78: in string theory.
79: However, later on D-term inflation in its turn was
80: though to be plagued with problems.
81:
82: In D-term inflation, the inflationary era ends when a U(1)
83: gauge symmetry is spontaneously broken by a vacuum expectation value
84: of some scalar field, leading to the formation of gauge cosmic strings.
85: As it was explicitly shown in Ref.~\cite{rjs}, cosmic strings are
86: generically expected to be formed at the end of a hybrid inflation phase,
87: in the context of supersymmetric grand unified theories.
88: It was claimed~\cite{rj} that the cosmic strings
89: contribution to the angular power spectrum of the Cosmic Microwave
90: Background (CMB) temperature anisotropies is constant and dominant
91: ($75\%$). From the observational point of view however, strong
92: constraints~\cite{bprs} are placed on the allowed cosmic strings
93: contribution to the CMB: it can not exceed $\sim 10\%$.
94: Thus, standard D-term inflation was thought
95: to be inconsistent with cosmology. To rescue D-term inflation there
96: have been proposed different mechanisms which either consider more
97: complicated models, or they require additional ingredients so that
98: cosmic strings are not produced at the end of hybrid inflation. For
99: example~\cite{nocs}, one can add a nonrenormalisable term in the
100: potential, or add an additional discrete symmetry, or consider GUT
101: models based on non-simple groups. More recently, a new pair of charged
102: superfields has been introduced in D-term inflation so that cosmic
103: strings formation is avoided~\cite{jaa}.
104:
105: The aim of our study is to show that standard D-term inflation
106: leading to the production of cosmic strings is still compatible with
107: cosmological data, and in particular CMB, without invoking any new
108: physical mechanisms. We find that in D-term inflation the cosmic
109: strings contribution to the CMB data depends on the free parameters, as for
110: F-term inflation~\cite{rs}. The maximum allowed cosmic
111: strings contribution to the CMB measurements places upper limits on
112: the inflationary scale (which is also the cosmic string energy scale),
113: or equivalently on the coupling of the superpotential.
114:
115: We first review the results for F-term hybrid inflation, in which case
116: the supersymmetric renormalisable superpotential reads
117: \begin{equation}\label{superpot}
118: W_{\rm infl}^{\rm F}=\kappa S(\Phi_+\Phi_- - M^2)~,
119: \end{equation}
120: where $S, \Phi_+, \Phi_-$ are three chiral superfields, and $\kappa$,
121: $M$ are two constants. The cosmic strings contribution to the CMB is
122: a function of the coupling $\kappa$, or equivalently of the mass scale
123: $M$. It can be consistent
124: with the most recent measurements, which require that it is at most
125: equal to $9\%$~\cite{bprs}, provided~\cite{rs}
126: \begin{equation}
127: M\lsim 2\times 10^{15} {\rm GeV} ~~\Leftrightarrow ~~\kappa \lsim
128: 7\times10^{-7}~.
129: \end{equation}
130: The above limit was obtained in the context of SO(10) gauge group.
131: Upper limits of the same order of magnitude are found for other gauge
132: groups~\cite{rs}.
133:
134: This result implies that F-term inflation leading to the production of
135: cosmic strings of the GUT scale can be compatible with measurements,
136: provided the coupling is sufficiently small. Thus, hybrid supersymmetric
137: inflation losses some of its appeal since it is required some amount
138: of fine tuning of its free parameter, $\kappa$ should be of the
139: order of $10^{-6}$ or smaller. This constraint on $\kappa$ is in
140: agreement with the one given in Ref.~\cite{kl}. The parameter
141: $\kappa$ is also subject to the gravitino constraint which imposes an
142: upper limit to the reheating temperature, to avoid gravitino
143: overproduction. Within supersymmetric GUTs, and assuming a see-saw mechanism to
144: give rise to massive neutrinos, the inflaton field will decay during
145: reheating into pairs of right-handed neutrinos. Using the constraints
146: on the see-saw mechanism it is possible~\cite{SenoSha,rs} to convert
147: the constraint on the reheating temperature to a constraint on the
148: coupling parameter $\kappa$, namely $\kappa \lesssim 8\times
149: 10^{-3}~$, which is clearly a weaker constraint.
150:
151: The superpotential coupling $\kappa$ is allowed to get higher values,
152: namely it can approach the upper limit permitted by the gravitino
153: constraint, if one employs the curvaton
154: mechanism~\cite{lw2002}. Such a mechanism can be easily
155: accommodated within supersymmetric theories, where one expects to have
156: a number of scalar fields. For fixed $\kappa$, the cosmic strings
157: contribution decreases rapidly as the initial value of the curvaton
158: field, ${\cal\psi}_{\rm init}$, decreases. Thus, the WMAP measurements
159: lead to an upper limit on ${\cal\psi}_{\rm init}$, namely
160: ${\cal\psi}_{\rm init}\lsim 5\times 10^{13}(\kappa/10^{-2})$ GeV~\cite{rs}.
161: This limit holds for $\kappa$ in the range $[5\times 10^{-5}, 1]$; for
162: lower values of $\kappa$, the cosmic strings contribution is always
163: suppressed and thus lower than the WMAP limit.
164:
165: The above results hold also if one includes supergravity corrections.
166: This is expected since the value of the inflaton
167: field is several orders of magnitude below the Planck scale.
168:
169: \section{D-term inflation}
170: D-term inflation is derived from the superpotential
171: \begin{equation}
172: \label{superpotD}
173: W^{\rm D}_{\rm infl}=\lambda S \Phi_+\Phi_-~,
174: \end{equation}
175: where $S, \Phi_-, \Phi_+$ are three chiral superfields and $\lambda$
176: is the superpotential coupling. D-term inflation requires the
177: existence of a nonzero Fayet-Illiopoulos term $\xi$, permitted only if
178: an extra U(1) symmetry beyond the GUT framework, is introduced. In the
179: context of supersymmetry we calculate the radiative corrections leading to the
180: effective potential,
181: \begin{equation}
182: \label{VexactD}
183: V^{{\rm D}-{\rm SUSY}}_{\rm eff}(|S|) =
184: \frac{g^2\xi^2}{2}\left\{1+\frac{g^2}{16\pi^2}
185: \left[2\ln\frac{|S|^2\lambda^2}{\Lambda^2}+
186: (z+1)^2\ln(1+z^{-1})+(z-1)^2\ln(1-z^{-1})\right]\right\}~,
187: \end{equation}
188: where $z=\lambda^2 |S|^2/(g^2\xi)$, with $g$ the gauge
189: coupling of the U(1) symmetry and $\xi$ the Fayet-Illiopoulos term,
190: chosen to be positive; $\Lambda$ stands for a renormalisation scale.
191: In the absence of the curvaton mechanism, the
192: quadrupole anisotropy is the sum of the inflaton field (scalar and tensor
193: parts) and cosmic strings contributions and we normalise it to the COBE data.
194:
195: We compute the mass scale of the symmetry breaking, given by $\sqrt\xi$,
196: and we find that it increases with $\lambda$. We then
197: calculate the cosmic strings contribution to the temperature anisotropies.
198: We find that within supersymmetry D-term inflation is consistent
199: with CMB data provided the superpotential coupling $\lambda$ is quite
200: small, namely $\lambda\lesssim 3\times 10^{-5}$.
201:
202: However, the dependence of $z_{\rm Q}$ (the index Q denotes the scale
203: corresponding to the quadrupole anisotropy) on the superpotential coupling
204: $\lambda$ results to values of the inflaton field $S_{\rm Q}$ above the Planck
205: mass. This implies that the correct analysis has to be done in the framework
206: of supergravity. For small values of the gauge coupling $g$, the study in the
207: context of supergravity becomes just the analysis within supersymmetry. Some
208: previous studies~\cite{rj,jap} found in the literature kept only the first
209: term of the radiative corrections. We find that it is necessary to perform the
210: analysis using the {\sl full} effective potential, which we calculated
211: for minimal supergravity. More precisely, using a minimal
212: K\"ahler potential ($K=|\phi_-|^2+|\phi_+|^2+|S|^2$) and a minimal gauge
213: kinetic function ($f(\Phi_i)=1$), the scalar potential
214: reads~\cite{rs}
215: \begin{equation}
216: \label{VexactDsugra}
217: V^{\rm D-SUGRA}_{\rm eff} =
218: \frac{g^2\xi^2}{2}\left\{1+\frac{g^2}{16\pi^2}
219: \left[2\ln\frac{|S|^2\lambda^2}{\Lambda^2}\exp\left({|S|^2\over M_{\rm
220: Pl}^2}\right)+
221: (z+1)^2\ln(1+z^{-1})+(z-1)^2\ln(1-z^{-1})\right]\right\}~,
222: \end{equation}
223: where $z=[\lambda^2 |S|^2/(g^2\xi)]\exp(|S|^2/M_{\rm Pl}^2)$.
224: The number of e-foldings is
225: \begin{equation}
226: N_{\rm Q}={2\pi^2\over g^2}\, \int_1^{z_{\rm Q}}
227: \frac{{\rm W}(c\,{\tilde z})}
228: {{\tilde z}^2 f({\tilde z})[1+{\rm W}(c\,{\tilde z})]^2}\, {\rm d}{\tilde z}~,
229: \end{equation}
230: where ${\rm W}(x)$ denotes the ``W-Lambert function'' defined by
231: ${\rm W}(x)\exp[{\rm W}(x)]=x$, and $c\equiv
232: (g^2 \xi)/(\lambda^2 M_{\rm Pl}^2)$. The number of e-foldings
233: $N_{\rm Q}$ is thus a function of $c$
234: and $z_{\rm Q}$, for $g$ fixed. Setting $N_{\rm Q}=60$ we obtain
235: a numerical relation between $c$ and $z_{\rm Q}$ which allows us
236: to construct a function $z_{\rm Q}(\xi)$ and express the three
237: contributions to the CMB only as a function of $\xi$. The total $(\delta T/T)$
238: is given by
239: \begin{equation}\label{eqnumDsugra}
240: \left[\left({\delta T\over T}\right)_{\rm Q-tot}\right]^2 \sim
241: \left(\frac{\xi}{M_{\rm Pl}^2}\right)^2\left\{ \frac{\pi^2}{90g^2}z^{-2}_{\rm
242: Q}f^{-2}(z_{\rm Q}) \frac{{\rm W}(c\,z_{\rm Q})} {\left[1+{\rm W}(c\,
243: z_{\rm Q})\right]^2} +\left(\frac{0.77
244: g}{8\sqrt{2}\pi}\right)^2 +
245: \left(\frac{9\pi}{4}\right)^2\right\}~,
246: \end{equation}
247: where the three contributions come from the scalar and tensor parts
248: of the inflaton field, and the cosmic strings, respectively.
249: We normalise the {\sl l.h.s.} of
250: Eq.~(\ref{eqnumDsugra}) to the COBE data, i.e.,
251: $\left(\delta T/ T\right)_{\rm Q}^{\rm COBE} \sim 6.3\times 10^{-6}$,
252: and we solve it numerically to obtain $\xi$, and thus, the three
253: contributions for given values of $g$ and $\lambda$.
254:
255: The cosmic strings contribution to the CMB data, is found to be an increasing
256: function of the mass scale $\sqrt{\xi}$, as shown in Fig.~\ref{xifig} below.
257:
258: \begin{figure}[htbp]
259: \includegraphics[scale=.5]{ContribFuncMassSugra.eps}\\
260: \caption{The cosmic strings contribution to the CMB data, as a function of
261: the mass scale $\sqrt{\xi}$ in units of $10^{15}$ GeV. }\label{xifig}
262: \end{figure}
263:
264:
265: \begin{figure}[htbp]
266: \includegraphics[scale=.45]{PRL4.eps}\\
267: \caption{Cosmic strings contribution to the CMB temperature anisotropies
268: as a function
269: of the superpotential coupling $\lambda$ for different values of the gauge
270: coupling $g$. The maximal contribution allowed by WMAP is represented by a
271: dotted line.}\label{PRL4}
272: \end{figure}
273:
274: Our results, summarized in Fig.~\ref{PRL4}, differ from the
275: results obtained in the framework of supersymmetry unless $\lambda
276: \gtrsim 10^{-3}$ or $g \lesssim 10^{-4}$. The cosmic strings
277: contribution to the CMB turns out to be dependent on the free parameters,
278: with however the robust result that the cosmic strings contribution is not
279: constant, nor is it always dominant, in contradiction to
280: Ref.~\cite{rj}. This implies that contrary to what is often assumed,
281: the simplest D-term inflation is still an open possibility and one
282: does not need to consider more complicated models. Our
283: analysis shows that $g\gtrsim 1$ necessitates multiple-stage
284: inflation, since otherwise we cannot have sufficient e-foldings to resolve
285: the horizon problem of standard cosmology, while
286: $g\gtrsim 2\times 10^{-2}$ is incompatible with the WMAP measurements.
287: For $g\lesssim 2\times
288: 10^{-2}$, we can also constrain the superpotential coupling $\lambda$
289: and get $\lambda \lesssim 3\times 10^{-5}$. This limit was already
290: found in the framework of supersymmetry~\cite{rs} and it is in
291: agreement with the constraint $\lambda\lesssim {\cal O}(10^{-4}-10^{-5})$
292: of Ref.~\cite{jap}.
293: Supergravity corrections impose in addition a lower limit to
294: the coupling $\lambda$. If for example $g=10^{-2}$, the cosmic strings
295: contribution imposes $10^{-8}\lesssim \lambda \lesssim 3\times
296: 10^{-5}$. The constraint induced by CMB measurements is
297: expressed as a single constraint on the Fayet-Iliopoulos term $\xi$,
298: namely $\sqrt\xi \lesssim 2\times 10^{15}~{\rm GeV}$.
299:
300: As a next step we examine whether there is a mechanism to
301: allow more natural values of the couplings. Assuming the existence of a scalar
302: field, that is subdominant during inflation as well as at the
303: beginning of the radiation dominated era, such a field (the curvaton)
304: gives an additional contribution to the temperature anisotropies,
305: which we calculate below for the case of D-term inflation.
306: The curvaton contribution, in terms of the metric perturbation,
307: reads~\cite{mt2002}
308: \begin{equation}
309: \left(\frac{\delta T}{T}\right)_{\rm curv} \equiv \frac{\Psi_{\rm
310: curv}}{3} ={4\over 9}{\delta{\cal \psi_{\rm init}}\over \psi_{\rm
311: init}}~.
312: \end{equation}
313: The initial quantum fluctuations of the curvaton
314: field are given by
315: $\delta{\cal \psi}_{\rm init}=H_{\rm inf}/( 2\pi)$.
316: The expansion rate during inflation, $H_{\rm infl}$,
317: is a function of the inflaton field and it is given by the Friedmann
318: equation: $H_{\rm infl}^2(\varphi)=(8\pi/3)V(\varphi)$.
319: Thus, for the D-term tree-level effective potential,
320: the additional curvaton contribution to the total
321: temperature anisotropies is given by~\cite{jap}
322: \begin{equation}
323: \left[\left(\frac{\delta T}{T}\right)_{\rm curv}\right]^2={1\over
324: 6}\left({2\over 27\pi}\right)^2\left(
325: {g\xi\over M_{\rm Pl}\psi_{\rm init}}\right)^2~.
326: \end{equation}
327: Normalising the total $(\delta T/ T)$ to COBE we then obtain the
328: contributions of the different sources (inflaton field splitted into scalar
329: and tensor parts, cosmic strings, curvaton field) to the CMB as a function
330: of one of the three parameters $\psi_{\rm init}, \lambda, g$, keeping the
331: other two fixed.
332: We show in Fig.~\ref{prl2} the three contributions as a function of
333: $\psi_{\rm init}$, for $\lambda=10^{-1}$ and $g=10^{-1}$.
334: Clearly, there are values of $\psi_{\rm init}$ which allow bigger
335: values of the superpotential coupling $\lambda$ and of the gauge coupling $g$,
336: than the upper bounds obtained in the absence of a curvaton field.
337:
338: \begin{figure}[htbp]
339: \includegraphics[scale=.5]{ContribDtermCurvg1.eps}\\
340: \caption{The cosmic strings (dark gray), curvaton (light gray) and
341: inflaton (gray) contributions to the CMB temperature anisotropies as a
342: function of the the initial value of the curvaton field
343: ${\cal\psi}_{\rm init}$, for $\lambda=10^{-1}$ and
344: $g=10^{-1}$.}\label{prl2}
345: \end{figure}
346:
347: More explicitely, the fine tuning on the couplings can be avoided provided
348: \begin{equation}
349: \psi_{\rm init}\lsim 3\times 10^{14}\left({g\over 10^{-2}}\right) ~{\rm GeV}
350: ~~~{\rm for}~~ \lambda\in [10^{-1}, 10^{-4}]~.
351: \end{equation}
352: Clearly, for smaller values of $\lambda$, the curvaton mechanism is not
353: necessary.
354:
355: We would like to bring to the attention of the reader that in the
356: above study we have neglected the quantum gravitational effects, which
357: would lead to a contribution to the effective potential, even though
358: $S_{\rm Q}\sim {\cal O}(10 M_{\rm Pl})$. Our analysis is however
359: still valid, since the effective potential given in
360: Eq.~(\ref{VexactDsugra}) satisfies the conditions~\cite{lindebook}
361: $V(|S|)\ll M_{\rm Pl}^4$ and $m^2_S={\rm d}^2 V/{\rm d}S^2 \ll M_{\rm
362: Pl}^2$, and thus the quantum gravitational corrections $[\Delta
363: V(|S|)]_{\rm QG}$ are negligible when compared to the effective
364: potential $V_{\rm eff}^{\rm D-SUGRA}$.
365:
366: \section{Conclusions}
367: D-term inflation gained a lot of interest since it was shown that it
368: avoids the problem of ``Hubble-induced mass'', but it was later thought
369: to be plagued with an inconsistency with the data.
370: Standard D-term inflation ends with the formation of cosmic
371: strings which was claimed to lead to a constant and dominant contribution
372: to the CMB data, much higher than the one allowed by measurements.
373: In this study, we show that this is not the case, and therefore,
374: standard D-term hybrid
375: inflation can still be compatible with cosmological data.
376:
377: We consider standard D-term inflation in its simplest form and without
378: any additional ingredients. We perform our analysis in the
379: framework of supergravity, since we reach scales above the Planck
380: scale, and we consider all one-loop radiative corrections. We show that
381: the cosmic strings produced at the end of D-term inflation can lead to a
382: contribution to the CMB data which is allowed by the measurements.
383: The price to be paid is that the couplings must be small.
384: However, this constraint can be less severe
385: if one invokes the curvaton mechanism.
386:
387: \section*{Acknowledgements}
388: It is a pleasure to thank
389: G. Esposito-Far\`ese, A.\ Linde and P.\ Peter
390: for discussions and comments.
391:
392: \begin{thebibliography}{50}
393:
394: \bibitem{infl} A.~Guth, Phys.~Rev.~D{\bf 23}, 347 (1981); A.~Linde,
395: Phys.~Lett.~B{\bf 108}, 389 (1982).
396:
397: \bibitem{onset}
398: E~Calzetta and M.~Sakellariadou, Phys.~Rev.~D{\bf 45}, 2802 (1992);
399: Phys.~Rev.~D{\bf 47}, 3184 (1993).
400:
401: \bibitem{dterm}
402: E.~Halyo, Phys.~Lett.~B{\bf 387}, 43 (1996); P.~Binetruy and D.~Dvali,
403: Phys.~Lett.~B{\bf 388}, 241 (1996).
404:
405: \bibitem{rjs} R.\ Jeannerot, J.\ Rocher, and M. \ Sakellariadou,
406: Phys.\ Rev. D{\bf 68}, 103514 (2003).
407:
408: \bibitem{rj}
409: R.\ Jeannerot, Phys.\ Rev. D{\bf 56}, 6205 (1997).
410:
411: \bibitem{bprs}
412: F.\ R.\ Bouchet, P.\ Peter, A.\ Riazuelo, and M.\ Sakellariadou,
413: Phys.\ Rev. D{\bf 65}, 021301 (2002);
414: L.\ Pogosian, M.\ Wyman, and I.\ Wasserman,
415: \emph{Observational constraints on cosmic strings: Bayesian analysis
416: in a three dimensional parameter space}, [arXiv:astro-ph/0403268].
417:
418: \bibitem{nocs}
419: R.\ Jeannerot, S.\ Khalil, G.\ Lazarides, and Q.\
420: Shafi, JHEP\textbf{0010},012 (2000);
421: G.\ Lazarides, and C.\ Panagiotakopoulos, Phys.\ Rev.\ D{\bf 52}, 559 (1995);
422: T.\ Watari, and T.\ Yanagida, Phys.\ Lett.\ B{\bf589}, 71 (2004).
423:
424: \bibitem{jaa}
425: J.~Urrestilla, A.~Ach\'ucarro and A.~C.~Davis, Phys.~Rev.~Lett.~{\bf
426: 92}, 251302 (2004).
427:
428: \bibitem{rs}
429: J.~Rocher and M.~Sakellariadou, \emph{Supersymmetric
430: grand unified theories and cosmology}, [arXiv:hep-ph/0406120].
431:
432: \bibitem{kl}
433: R.\ Kallosh and A.\ Linde, JCAP {\bf 0310}, 008 (2003).
434:
435: \bibitem{SenoSha}
436: V.\ N.\ Senoguz, and Q.\ Shafi, Phys.\ Lett.\ B{\bf567}, 79 (2003).
437:
438: \bibitem{lw2002}
439: D.\ H.\ Lyth and D.\ Wands, Phys.\ Lett.\ B{\bf524}, 5 (2002);
440: T.\ Moroi, and T.\ Takahashi, Phys.\ Lett.\ B{\bf522}, 215 (2001);
441: Erratum-ibid.\ B{\bf 539}, 303 (2002).
442:
443: \bibitem{jap}
444: M.\ Endo, M.\ Kawasaki, and T.\ Moroi, Phys.\ Lett.\ B.{\bf569},73 (2003).
445:
446: \bibitem{mt2002}
447: T.\ Moroi, and T.\ Takahashi, Phys.\ Rev. D{\bf 66}, 063501 (2002).
448:
449: \bibitem{lindebook}
450: A.\ Linde, {\sl Particle Physics and Inflationary Cosmology}, Contemporary
451: Concepts in Physics, {\bf 5} (1990), Hardwood Academic Publishers.
452:
453: \end{thebibliography}
454:
455: \end{document}
456: \end
457:
458: