1: \documentclass{article}
2: %%\documentstyle{article}
3: %\usepackage{epsfig}
4: \usepackage{graphicx}
5:
6: %------------------------------------------
7: % Definition for style
8: % Don't change these definition
9: %------------------------------------------
10: \topmargin 0pt
11: \oddsidemargin 0pt \evensidemargin 0pt
12: \textheight 23cm \textwidth 16.0cm
13: \def\abstract#1{\vskip 7mm
14: \begin{center}{\large Abstract}\par \smallskip
15: \begin{minipage}[c]{12cm}
16: \small #1
17: \end{minipage}
18: \end{center}
19: }
20: \def\title#1{\begin{center}{\Large\bf #1}\end{center}}
21: \def\author#1{\vskip 5mm \begin{center}{#1}\end{center}}
22: \def\address#1{\begin{center}{\it #1}\end{center}}
23: %------------------------------------------
24: % End of definition for style
25: %-------------------------------------------
26: % Define your macro here.
27: \def\be{\begin{equation}}
28: \def\ee{\end{equation}}
29: \newcommand{\bea}{\begin{eqnarray}}
30: \newcommand{\eea}{\end{eqnarray}}
31: \newcommand{\vx}{\ensuremath{\vec{x}}}
32: \newcommand{\vk}{\ensuremath{\vec{k}}}
33: %-------------------------------------------
34: \makeatletter
35: % You can use \lesssim ( < ) and \gtrsim ( > ).
36: % ~ ~
37: \@ifundefined{lesssim}{\def\lesssim{\mathrel{\mathpalette\vereq<}}}{}
38: \@ifundefined{gtrsim}{\def\gtrsim{\mathrel{\mathpalette\vereq>}}}{}
39: \def\vereq#1#2{\lower3pt\vbox{\baselineskip1.5pt \lineskip1.5pt
40: \ialign{$\m@th#1\hfill##\hfil$\crcr#2\crcr\sim\crcr}}}
41: \makeatother
42:
43: %--------------------------------------------
44: \begin{document}
45:
46: \title{%
47: Clarifying Slow Roll Inflation and the Quantum Corrections to the Observable
48: Power Spectra
49: % \smallskip \\
50: % {\large --- Please use this file to complete your manuscript ---}
51: }
52: \author{%
53: D. Boyanovsky$^{c,a,}$\footnote{E-mail:boyan@pitt.edu},
54: \underline{H. J. de Vega}$^{b,a,}$\footnote{E-mail:devega@lpthe.jussieu.fr},
55: N. G. Sanchez $^{a,}$\footnote{E-mail:Norma.Sanchez@obspm.fr}
56: }
57: \address{%
58: $^{a}$
59: Observatoire de Paris, LERMA, Laboratoire Associ\'e au CNRS UMR 8112,
60: \\ 61, Avenue de l'Observatoire, 75014 Paris, France. %}
61: \\$^{b}$ LPTHE, Laboratoire Associ\'e au CNRS UMR 7589,
62: Universit\'e Pierre et Marie Curie (Paris VI) \\ et Denis Diderot (Paris VII),
63: Tour 24, 5 \`eme. \'etage, 4, Place Jussieu, 75252 Paris, Cedex 05,
64: France. \\
65: $^{c}$Department of Physics and
66: Astronomy, University of Pittsburgh, Pittsburgh, Pennsylvania 15260,
67: USA }
68: \abstract{Slow-roll inflation can be studied as an effective field theory.
69: The form of the inflaton potential consistent with the data is
70: $ V(\phi) = N \; M^4 \; w\left(\frac{\phi}{\sqrt{N} \; M_{Pl}}\right) $
71: where $ \phi $ is the inflaton field, $ M $ is the inflation energy scale,
72: and $ N \sim 50 $ is the number of efolds since the cosmologically
73: relevant modes exited the Hubble radius until the end of
74: inflation. The dimensionless function $ w(\chi) $ and field
75: $ \chi $ are generically $ \mathcal{O}(1) $. The WMAP value for
76: the amplitude of scalar adiabatic fluctuations yields $ M \sim
77: 0.77\times 10^{16}$GeV. This form of the potential encodes
78: the slow-roll expansion as an expansion in $1/N$. A
79: Ginzburg-Landau (polynomial) realization of $ w(\chi) $
80: reveals that the Hubble parameter, inflaton mass and non-linear
81: couplings are of the see-saw form in terms of the small ratio $
82: M/M_{Pl} $. The quartic coupling is $ \lambda \sim
83: \frac{1}{N} \left(\frac{M}{M_{Pl}}\right)^4 $. The smallness of
84: the non-linear couplings is {\bf not} a result of fine tuning but
85: a {\bf natural} consequence of the validity of the effective field
86: theory and slow roll approximation. Our observations suggest that slow-roll
87: inflation may well be described by an almost critical
88: theory, near an infrared stable gaussian fixed point.
89: Quantum corrections to slow roll inflation are computed and turn to
90: be an expansion in powers $ \left(H/M_{Pl}\right)^2 $.
91: The corrections to the inflaton effective potential and its equation of motion
92: are computed, as well as the quantum corrections to the observable
93: power spectra. The near scale invariance of the fluctuations
94: introduces a strong infrared behavior naturally regularized by the
95: slow roll parameter $ \Delta \equiv \eta_V-\epsilon_V = \frac12(n_s -1)+
96: r/8 $. We find the \emph{effective} inflaton potential during slow roll
97: inflation including the contributions from scalar curvature and tensor
98: perturbations as well as from light scalars and Dirac
99: fermions coupled to the inflaton. The scalar and tensor
100: superhorizon contributions feature infrared enhancements regulated by
101: slow roll parameters. Fermions and gravitons
102: do not exhibit infrared enhancement. The subhorizon part is
103: completely specified by the trace anomaly of the fields with
104: different spins and is solely determined by the space-time geometry.
105: This inflationary effective potential is strikingly {\bf different}
106: from the usual Minkowski space-time result.
107: Quantum corrections to the power spectra are expressed in terms of the
108: CMB observables: $ n_s, \; r $ and $ dn_s/d \ln k $.
109: Trace anomalies (especially the
110: graviton part) dominate these quantum corrections in a definite
111: direction: they {\bf enhance} the scalar curvature fluctuations and
112: {\bf reduce} the tensor fluctuations. }
113:
114: \section{Inflation as an Effective Field Theory}
115:
116: Inflation was originally proposed to solve several outstanding
117: problems of the standard Big Bang model
118: \cite{guthsato} thus becoming an important
119: paradigm in cosmology. At the same time, it provides a natural
120: mechanism for the generation of scalar density fluctuations that
121: seed large scale structure, thus explaining the origin of the
122: temperature anisotropies in the cosmic microwave background (CMB),
123: as well as that of tensor perturbations (primordial gravitational
124: waves). A distinct aspect of
125: inflationary perturbations is that these are generated by quantum
126: fluctuations of the scalar field(s) that drive inflation.
127: Their physical wavelengths grow faster than the Hubble radius and when
128: they cross the horizon they freeze out and decouple.
129: Later on, these fluctuations are amplified and grow, becoming classical and
130: decoupling from causal microphysical processes. Upon re-entering
131: the horizon, during the matter era, these scalar (curvature)
132: perturbations induce temperature anisotropies imprinted
133: on the CMB at the last scattering surface and
134: seed the inhomogeneities which generate structure upon gravitational
135: collapse\cite{pert,hu}. Generic
136: inflationary models predict that these fluctuations are adiabatic
137: with an almost scale invariant spectrum. Moreover, they are Gaussian
138: to a very good approximation.
139: These generic predictions are in spectacular agreement with the CMB
140: observations as well as with a variety of large scale structure data
141: \cite{WMAP}. The WMAP data \cite{WMAP} clearly display an anti-correlation
142: peak in the temperature-polarization (TE) angular power spectra at
143: $l\sim 150$, providing one of the most striking confirmations of superhorizon
144: adiabatic fluctuations as predicted by inflation\cite{WMAP}.
145:
146: The robust predictions of inflation (value of the entropy of the universe,
147: solution of the flatness problem, small adiabatic Gaussian density
148: fluctuations explaining the CMB anisotropies, ...) which are common to the
149: available inflationary scenarios, show the predictive power of the
150: inflationary paradigm. While there is a great diversity of inflationary models,
151: they predict fairly generic features: a gaussian,
152: nearly scale invariant spectrum of
153: (mostly) adiabatic scalar and tensor primordial fluctuations.
154: More precisely, the WMAP\cite{WMAP} data can be fit outstandingly well
155: by simple single field slow roll models.
156: These generic predictions of inflationary models make the
157: inflationary paradigm robust. Whatever the microscopic model for the early
158: universe (GUT theory) would be, it should include inflation with the generic
159: features we know today.
160:
161: Inflationary dynamics is typically studied by treating the
162: inflaton as a homogeneous classical scalar
163: field whose evolution is determined
164: by its classical equation of motion, while the inflaton
165: quantum fluctuations (around the classical value and in the
166: Gaussian approximation) provide the seeds for the scalar
167: density perturbations of the metric. The classical dynamics of the
168: inflaton (a massive scalar field)
169: coupled to a cosmological background clearly shows that inflationary
170: behaviour is an {\bf attractor} \cite{bgzk}. This is a generic and robust
171: feature of inflation.
172:
173: In quantum field theory,
174: this classical inflaton corresponds to the expectation value
175: of a quantum field operator in a translational invariant state.
176: Important aspects of the inflationary dynamics, as resonant
177: particle production and the nonlinear back-reaction that it generates,
178: require a full quantum treatment of the inflaton for their consistent
179: description. The quantum dynamics of the inflaton
180: in a non-perturbative framework and its consequences on the CMB anisotropy
181: spectrum were treated in refs.\cite{cosmo,cosmo2,staro,anom}.
182:
183: The quantum fluctuations are of two different kinds: (a) Large quantum
184: fluctuations generated at the begining of inflation through
185: spinodal or parametric resonance depending on the inflationary scenario
186: chosen. They have comoving wavenumbers in the range of $ 10^{13} \mbox{GeV}
187: \lesssim k \lesssim 10^{15}$GeV and they become superhorizon a few efolds
188: after the begining of inflation. Their physical wavenumbers
189: become subsequently very small compared with the inflaton mass.
190: Therefore, the assembly of these modes can be treated as part of the zero mode
191: after $5-10$ efolds \cite{cosmo,cosmo2}. That is, the use of an homogeneous
192: classical inflaton is thus justified by the full quantum theory treatment of
193: the inflaton.
194: (b) Small fluctuations of high comoving wavenumbers
195: $$
196: e^{N_T-60} \; 10^{16} \, GeV < k < e^{N_T-60} \; 10^{20} \, GeV
197: $$
198: where $ N_T \geq 60 $ stands for the total number of efolds (see for example
199: Ref. \cite{sd}). These are the cosmologically relevant modes that exit
200: the horizon about 50 efolds before the end of inflation and
201: reenter later on (during the matter dominated era) being the source of
202: primordial power for the CMB anisotropies as well as for the structure
203: formation. While modes (b) obey linear evolution equations with
204: great accuracy, modes (a) strongly interact with themselves calling
205: for nonperturbative quantum field theory treatments as in
206: refs.\cite{cosmo,cosmo2}. Notice that particle production is governed by
207: linear unstabilities (parametric or spinodal) only at the begining
208: of inflation. Particle production keeps strong during the nonlinear
209: regime till particles eventually dominate the energy density and inflation
210: stops\cite{cosmo}. The modes (b) correspond to physical scales that were microscopic
211: (even transplanckian) at the
212: begining of inflation, then after they become astronomical and produce the CMB
213: anisotropies as well as the large scale structure of the universe.
214:
215: The crucial fact is that the excitations can
216: cross the horizon {\bf twice}, coming back and bringing information
217: from the inflationary era. We depict in fig. \ref{infl} the physical
218: wavelengths of modes (a) and (b) vs. the logarithm of the scale factor
219: showing that modes (b) crossed twice the horizon, modes (a) are
220: out of the horizon still today.
221:
222: \begin{figure}[h]
223: \includegraphics[width=12cm,height=8cm]{inflacion}
224: %\epsfig{file=inflacion.eps,width=12cm,height=10cm}
225: \caption{ Physical lengths $ \lambda = a(t) \;
226: \lambda_{comoving} $ vs. the scale factor $ a(t) $ in a log-log plot.
227: The causal horizon $ d_H $ is shown for the inflationary
228: (de~Sitter), radiation dominated and matter dominated eras. The
229: physical wavelengths ($ \lambda $) for today's Hubble radius
230: and a typical galactic scale ($ \lambda_{gal} $) are shown. One sees the modes (b)
231: can cross the horizon {\bf twice} bringing information from
232: extremely short wavelength modes during the
233: inflationary era. Modes (a) which have large amplitudes during inflation
234: and dominated the energy of the universe, have not yet crossed back inside
235: the horizon.}
236: \label{infl}
237: \end{figure}
238:
239: Recently, particle decay in a de Sitter background as well as during slow
240: roll inflation has been studied in ref.\cite{pardec} together with
241: its implication for the decay of the density fluctuations. Quantum
242: effects during slow roll inflation including quantum corrections
243: to the effective inflaton potential and its equation of motion are
244: derived in ref.\cite{anom,qua}.
245: Recent studies of quantum corrections during
246: inflation\cite{anom,pardec,qua} revealed the robustness of classical
247: single field slow roll inflationary models as a result of the
248: validity of the {\bf effective field theory} description. The reliability
249: of an effective field theory of inflation hinges on a wide
250: separation between the energy scale of inflation, determined by $H$
251: and that of the underlying microscopic theory which is taken to be
252: the Planck scale $ M_{Pl} $.
253: Inflation as known today should be considered as an {\bf effective theory},
254: that is, it is not a fundamental theory but a theory of a
255: condensate (the inflaton field) which follows from a more fundamental one
256: (the GUT model). The inflaton field is just an {\bf effective}
257: description while the microscopic description should come from the GUT model
258: in the cosmological spacetime. Such derivation is not yet available.
259:
260: Bosonic fields do
261: not need to be fundamental fields, for example they may emerge as
262: condensates of fermion-antifermion pairs $ < {\bar \Psi} \Psi> $
263: in a grand unified theory (GUT) in the cosmological background. In
264: order to describe the cosmological evolution is enough to consider
265: the effective dynamics of such condensates. The relation between
266: the low energy effective field theory of inflation and the
267: microscopic fundamental theory is akin to the relation between the
268: effective Ginzburg-Landau theory of superconductivity and the
269: microscopic BCS theory, or like the relation of the $ O(4) $ sigma
270: model, an effective low energy theory of pions, photons and chiral
271: condensates with quantum chromodynamics (QCD)\cite{quir}. The
272: guiding principle to construct the effective theory is to include
273: the appropriate symmetries\cite{quir}. Contrary to the sigma model
274: where the chiral symmetry strongly constraints the
275: model\cite{quir}, only general covariance can be imposed on the
276: inflaton model.
277:
278: The inflationary scenarios are usually
279: characterized as small and large fields scenarii. In small fields
280: scenarios the initial classical amplitude of the inflaton is assumed
281: small compared with $ M_{Pl} $ while in large field scenarii
282: the inflaton amplitude is initially of the order $ M_{Pl} $. The
283: first type of scenarii is usually realized with broken symmetric
284: potentials ($ m^2 < 0 $) while for the second type scenarii (`chaotic
285: inflation') one can just use unbroken potentials with $ m^2 > 0 $.
286:
287: Gravity can be treated semiclassically for inflation: the geometry is
288: classical and the metric follows from the Einstein-Friedman equations
289: where the r.h.s. is the expectation value of a quantum
290: operator. Quantum gravity corrections can be neglected during
291: inflation because the energy scale of inflation $ M \sim M_{GUT} \sim 10^{-3}
292: \; M_{Planck} $. That is, quantum gravity
293: effects are at most $ (M/M_{Planck})^2 \sim 10^{-6} $ and can be neglected in this
294: context. The studies in ref.\cite{pardec,qua,clar} reveal
295: that quantum corrections in the effective field theory yields an
296: expansion in $ \left(\frac{H}{ M_{Pl}}\right)^2 $ for
297: \emph{general inflaton potentials}. This indicates
298: that the use of the inflaton potential $ V(\phi) $ from effective
299: field theory is consistent for
300: $$
301: \left(\frac{H}{M_{Pl}}\right)^2 \ll 1 \quad {\rm and~hence} \quad
302: V(\phi) \ll M_{Pl}^4 \; ,
303: $$
304: allowing amplitudes of the inflaton field $ \phi $
305: {\bf well beyond} $ M_{Pl} $\cite{clar}.
306:
307: \subsection{Slow-roll Inflation as an expansion in $1/N_{efolds}$ and
308: no fine tuning}
309:
310: In single field inflation the energy density is dominated by a
311: homogeneous scalar \emph{condensate}, the inflaton, whose dynamics
312: can be described by an \emph{effective} Lagrangian
313: \be\label{lagra}
314: {\cal L} = a^3({ t}) \left[ \frac{{\dot
315: \phi}^2}{2} - \frac{({\nabla \phi})^2}{2 \; a^2({ t})} - V(\phi) \right]
316: \; ,
317: \ee
318: together to the Einstein-Friedman equation
319: \be \label{ef}
320: \left[ \frac{1}{a(t)} \; \frac{da}{d t} \right]^2 =
321: \frac{\rho( t)}{3 \; M^2_{Pl}} \; ,
322: \ee
323: where the energy density
324: for an homogeneous inflaton is given by
325: $$
326: \rho(t) = \frac{{\dot \phi}^2}{2} + V(\phi) \; .
327: $$
328: \noindent The inflaton potential $ V(\phi) $ is a slowly varying
329: function of $ \phi $ in order to permit a slow-roll solution for
330: the inflaton field $ \phi(t) $. All this in a spatially flat FRW metric
331: \be \label{metrica}
332: ds^2= dt^2-a^2(t) \; d{\vec x}^2 = C^2(\eta) \left[ (d \eta)^2 - d{\vec x}^2
333: \right] \; ,
334: \ee
335: where $ \eta $ is the conformal time and $ C(\eta) \equiv a(t(\eta)) $.
336:
337: The inflaton evolution equation takes the form,
338: $$
339: {\ddot \phi} + 3 \, H \, {\dot \phi} + V'(\phi) = 0 \; .
340: $$
341: In the slow-roll approximation $ {\ddot \phi} \ll
342: 3 \, H \, {\dot \phi} $ and $ \frac{{\dot \phi}^2}{2} \ll V(\phi) $,
343: and the inflaton evolution equations become
344: \be \label{slow}
345: 3 \, H(t) \, {\dot \phi} + V'(\phi) = 0 \quad , \quad
346: \left[ H(t) \right]^2 = \frac{ V(\phi)}{3 \; M^2_{Pl}} \; ,
347: \ee
348: where $ H(t) \equiv \frac{1}{a(t)} \; \frac{da}{dt} $ stands for the Hubble
349: parameter. Eq.(\ref{slow}) can be integrated by quadratures
350: with the result
351: $$
352: \ln a(t) = - \frac1{M^2_{Pl}} \int^{\phi(t)}_{\phi(0)} \frac{V(\phi)}{V'(\phi)}
353: \; d\phi \; .
354: $$
355: This formula shows that the inflaton field $ \phi $ scales as $ M_{Pl} $
356: and as the square root of the number of efolds \cite{clar}.
357: This suggest to introduce the
358: dimensionless field $ \chi $ and the dimensionless potential $ w(\chi) $,
359: \be \label{defxi}
360: \chi \equiv \frac{\phi}{\sqrt{N} \; M_{Pl}}
361: \qquad , \qquad w(\chi) \equiv \frac{V(\phi)}{ N \; M^4 }
362: \ee
363: where $ M $ stands for the scale of inflation.
364: The dimensionless field $ \chi $ is \emph{slowly} varying during the stage
365: of slow roll inflation: a large change in the field amplitude $\phi$
366: results in a small change in $\chi$ amplitude ,
367: \be \label{slofield}
368: \Delta \chi = \frac{1}{\sqrt{N}} \frac{\Delta \phi}{M_{Pl}} \; ,
369: \ee
370: a change in $\phi$ with $\Delta \phi \sim
371: M_{Pl}$ results in a change $\Delta \chi \sim 1/\sqrt{N}$.
372:
373: Introducing a \emph{stretched} (slow) dimensionless time variable $\tau$ and a
374: rescaled dimensionless Hubble parameter $h$ as follows
375: \be\label{time}
376: t = \sqrt{N} \; \frac{M_{Pl}}{M^2} \; \tau \qquad
377: , \qquad H = \sqrt{N} \; \frac{M^2}{M_{Pl}}\; h \; ,
378: \ee
379: the Einstein-Friedman equation reads
380: \be \label{efa}
381: h^2(\tau) = \frac13\left[\frac1{2\;N} \left(\frac{d\chi}{d \tau}\right)^2 +
382: w(\chi) \right] \; ,
383: \ee
384: and the evolution equation for the inflaton field
385: $\chi$ is given by
386: \be \label{eqnmot}
387: \frac1{N} \; \frac{d^2
388: \chi}{d \tau^2} + 3 \; h \; \frac{d\chi}{d \tau} + w'(\chi) = 0 \; .
389: \ee
390: The slow-roll approximation follows by neglecting the
391: $\frac1{N}$ terms in eqs.(\ref{efa}) and (\ref{eqnmot}). Both
392: $w(\chi)$ and $h(\tau)$ are of order $N^0$ for large $N$. Both
393: equations make manifest the slow roll expansion as a {\bf systematic expansion in}
394: $1/N$\cite{clar}.
395:
396: Following the spirit of the Ginsburg-Landau theory of phase transitions,
397: the simplest choice is a quartic trinomial for the inflaton potential\cite{nos}:
398: \be \label{wxi}
399: w(\chi)= w_0 \pm \frac12 \; \chi^2 + \frac{G_3}3 \; \chi^3 +
400: \frac{G_4}{4} \; \chi^4 \; .
401: \ee
402: where the coefficients $ w_0, \; G_3 $ and $ G_4 $ are of order one
403: and the signs $ \pm $ correspond to large and small field inflation,
404: respectively. Inserting eq.(\ref{wxi}) in eq.(\ref{defxi}) yields,
405: \be\label{VI}
406: V(\phi)= V_0 \pm \frac{m^2}{2} \; \phi^2 + \frac{ m
407: \; g }{3} \; \phi^3 + \frac{\lambda}{4}\; \phi^4 \; .
408: \ee
409: where $ m², \; g $ and $ \lambda $ are given by the
410: following see-saw-like relations,
411: \be
412: m = \frac{M^2}{M_{Pl}} \qquad , \qquad
413: g = \frac{G_3}{\sqrt{N}} \left( \frac{M}{M_{Pl}}\right)^2
414: \qquad , \qquad
415: \lambda = \frac{G_4}{{N}} \left( \frac{M}{M_{Pl}}\right)^4
416: \label{aco} \; .
417: \ee
418: We can now input the results from WMAP\cite{WMAP} to constrain the
419: scale $M$. The amplitude of adiabatic scalar perturbations in
420: slow-roll is expressed as
421: \be \label{ampliI}
422: |{\Delta}_{k\;ad}^{(S)}|^2 = \frac{1}{12 \, \pi^2 \; M_{Pl}^6}
423: \; \frac{V^3}{V'^2}= \frac{N^2}{12 \, \pi^2} \;
424: \left(\frac{M}{M_{Pl}}\right)^4 \; \frac{w^3(\chi)}{w'^2(\chi)} \; ,
425: \ee
426: Since, $ w(\chi) $ and $ w'(\chi) $ are of order one, we find
427: \be\label{Mwmap}
428: \left(\frac{M}{M_{Pl}}\right)^2 \sim \frac{2
429: \, \sqrt{3} \, \pi}{N} \; |{\Delta}_{k\;ad}^{(S)}| \simeq 1.02
430: \times 10^{-5} \; .
431: \ee
432: where we used $ N \simeq 50 $ and the WMAP
433: value for $ |{\Delta}_{k\;ad}^{(S)}| = (4.67 \pm 0.27)\times
434: 10^{-5} $ \cite{WMAP}. This fixes the scale of inflation to be
435: $$
436: M \simeq 3.19 \times 10^{-3} \; M_{PL} \simeq 0.77
437: \times 10^{16}\,\textrm{GeV} \; .
438: $$
439: This value pinpoints the scale of
440: the potential during inflation to be at the GUT scale suggesting a
441: deep connection between inflation and the physics at the GUT
442: scale in cosmological space-time.
443:
444: That is, the WMAP data {\bf fix} the scale of inflation $ M $ for
445: single field potentials with the form given by eq.(\ref{defxi}).
446: This value for $M$ is below the WMAP upper bound on the inflation scale
447: $ 3.3 \; 10^{16}$GeV \cite{WMAP}.
448: Furthermore, the Hubble scale during (slow roll) inflation and the
449: inflaton mass near the minimum of the potential are thereby determined from
450: eqs.(\ref{time}) and (\ref{aco}) to be $
451: m = \frac{M^2}{M_{Pl}} = 2.45 \times 10^{13} \,\textrm{GeV},
452: \quad H = \sqrt{N} \; m \; h \sim 1.0 \times 10^{14}\,\textrm{GeV}
453: = 4.1 \; m \; $ since $ h = {\cal O}(1) $.
454: In addition, the order of magnitude of the couplings naturally follows
455: from eq.(\ref{aco}): $ g \sim 10^{-6}, \quad \lambda \sim 10^{-12}$,
456: since $ M/M_{Pl} \sim 3 \times 10^{-3} $.
457:
458: Since $M/M_{Pl} \sim 3 \times 10^{-3}$,
459: these relations are a {\bf natural} consequence
460: of the validity of the effective field theory and of slow roll and
461: relieve the {\bf fine tuning problem}. We emphasize that the
462: `see-saw-like' form of the couplings is a natural consequence of
463: the form of the potential eq.(\ref{defxi}).
464: While the hierarchy between the Hubble parameter,
465: the inflaton mass and the Planck scale during slow roll inflation
466: is well known, our analysis reveals that small couplings are
467: naturally explained in terms of powers of the ratio between the
468: inflationary and Planck scales \emph{and} integer powers of $ 1/\sqrt{N} $.
469:
470: This is one of the main results presented in this lecture:
471: the effective field theory and slow roll descriptions of inflation,
472: both validated by WMAP, lead us to conclude that there is \emph{no
473: fine tuning} problem\cite{clar}. The smallness of the inflaton mass and the
474: coupling constants in this trinomial realization of the
475: inflationary potential is a \emph{direct} consequence of the
476: validity of both the effective field theory and the slow roll
477: approximations through a see-saw-like mechanism.
478:
479: It must be stressed that these order of magnitude estimates follow
480: from the statement that $ w(\chi) $ and $ \chi $ are of order one.
481: They are thus independent of the details of the model.
482: Indeed, model-dependent factors of order one appearing in the
483: observables should allow to exclude or
484: accept a given model by using the observational data.
485:
486: The WMAP results rule out
487: the purely quartic potential ($ m=0, \; g=0 $). From the point of view
488: of an effective field theory this is not surprising:
489: it is rather \emph{unnatural} to set
490: $m=0$, since $m=0$ is a particular point at which the correlation
491: length is infinite and the theory is critical. Indeed the
492: systematic study in ref.\cite{nos} shows that the best fit to the
493: WMAP data requires $ m^2 \neq 0 $.
494:
495: The general quartic Lagrangian eq.(\ref{V}) describes a
496: renormalizable theory. However, one can choose in the present context
497: arbitrary high order polynomials for $ V(\phi) $.
498: These nonrenormalizable models
499: are also effective theories where $ M_{Pl} $ plays the r\^ole of UV cutoff.
500: However, already a quartic potential is rich enough to describe the full
501: physics and to reproduce accurately the WMAP data \cite{nos}.
502:
503: For a general potential $ V(\phi) $,
504: \be \label{serie}
505: V(\phi) = \sum_{n=0}^{\infty} \lambda_n \; \phi^n \quad i. e. \quad
506: w(\chi) = \sum_{n=0}^{\infty} G_n \; \chi^n \; ,
507: \ee
508: we find from eq.(\ref{defxi}):
509: \be\label{Gn}
510: \lambda_n = \frac{G_n \; m^2}{\left( N \; M_{Pl}^2 \right)^{\frac{n}2-1}}\; ,
511: %\quad {\rm where} \quad w(\chi) = \sum_{n=0}^{\infty} G_n \; \chi^n
512: \ee
513: where the dimensionless coefficients $ G_n $ are of order one.
514: We find the scaling behavior $\lambda_n \sim 1/N^{\frac{n}2-1}$.
515: Eq. (\ref{aco}) displays particular cases of eq.(\ref{Gn}) for $ n=3 $
516: and $ 4 $.
517:
518: \medskip
519:
520: The slow-roll parameters naturally result of the order $1/N, \; 1/N^2 $ etc.
521: when expressed in terms of the inflaton potential
522: \be\label{VV}
523: V(\phi) = N \; M^4 \; w(\chi) \; .
524: \ee
525: That is,
526: \bea
527: &&\epsilon_V = \frac{M^2_{Pl}}{2} \;
528: \left[\frac{V^{'}(\Phi_0)}{V(\Phi_0)} \right]^2 = \frac1{2 \; N} \;
529: \left[\frac{w'(\chi)}{w(\chi)} \right]^2
530: \quad , \quad
531: \eta_V = M^2_{Pl} \; \frac{V^{''}(\Phi_0)}{V(\Phi_0)} =
532: \frac1{N} \; \frac{w''(\chi)}{w(\chi)} \; , \label{etav}
533: \eea
534: The spectral index $ n_s $, its running and the ratio of tensor to scalar
535: fluctuations are expressed as
536: \bea \label{indi}
537: &&n_s - 1 = -\frac3{N} \; \left[\frac{w'(\chi)}{w(\chi)} \right]^2
538: + \frac2{N} \; \frac{w''(\chi)}{w(\chi)} \quad , \cr \cr
539: &&\frac{d n_s}{d \ln k}= - \frac2{N^2} \; \frac{w'(\chi) \;
540: w'''(\chi)}{w^2(\chi)} - \frac6{N^2} \; \frac{[w'(\chi)]^4}{w^4(\chi)}
541: + \frac8{N^2} \; \frac{[w'(\chi)]^2 \; w''(\chi)}{w^3(\chi)}\quad , \cr \cr
542: &&r = \frac8{N} \; \left[\frac{w'(\chi)}{w(\chi)} \right]^2 \quad .
543: \eea
544: In eqs.(\ref{ampliI}), (\ref{etav}) and (\ref{indi}), the field $ \chi $ is computed
545: at horizon exiting. We choose $ N[\chi] = N = 50 $.
546: The WMAP results favoured single inflaton models and among them new and hybrid
547: inflation emerge to be preferable than chaotic inflation\cite{nos}.
548:
549: The inflationary era ends when the particles produced during inflation
550: dominate the energy density overcoming the vacuum energy.
551: At such stage the universe slows down its expansion to a radiation dominated
552: regime.
553:
554: \subsection{Connection with Supersymmetry}
555:
556: The form of the inflaton potential
557: \be \label{V}
558: V(\phi) = N \; M^4 \; w(\chi)
559: \ee
560: resembles (besides the factor $ N $) the moduli potential arising
561: from supersymmetry breaking
562: \be\label{susy}
563: V_{susy}(\phi) = m_{susy}^4 \; v\!\left(\frac{\phi}{M_{Pl}}\right) \; ,
564: \ee
565: where $ m_{susy} $ stands for the supersymmetry breaking scale. In our
566: context, eq.(\ref{susy}) indicates that $ m_{susy} \simeq M
567: \simeq M_{GUT} $. That is, the susy breaking scale $ m_{susy} $
568: turns out to be at the GUT and inflation scales. This may be a
569: \emph{first} observational indication of the presence of
570: supersymmetry. It should be noticed that this supersymmetry scale
571: is unrelated to an eventual existence of supersymmetry at the TeV scale.
572:
573: Notice that the invariance of the basic interactions (the lagrangian)
574: and the invariance of the physical states (or density matrices) describing
575: the matter are different issues.
576: For example, no thermal state at non-zero temperature
577: can be invariant under supersymmetry since Bose-Einstein and
578: Fermi-Dirac distributions are different.
579: More generally, the inflationary stage is described by a scalar condensate
580: (the inflaton) while fermions cannot condense due to Pauli principle.
581: This makes quite hard to uncover supersymmetric properties of the lagrangian
582: from the physics of the early universe.
583:
584: \subsection{Conjecture: inflation is near
585: a trivial infrared fixed point}\label{conjecture}
586: There are several remarkable features and consistency checks of the
587: relations (\ref{aco}):
588: \begin{itemize}
589: \item{Note the relation $\lambda \sim g^2$. This is the correct
590: consistency relation in a renormalizable theory because at one
591: loop level there is a renormalization of the quartic coupling (or
592: alternatively a contribution to the four points correlation
593: function) of orders $ \lambda^2 , \; g^4 $ and $ \lambda \; g^2 $ which
594: are of the same order for $ \lambda \sim g^2 $. Similarly, at one
595: loop level there is a renormalization of the cubic coupling
596: (alternatively, a contribution to the three point function) of
597: orders $ g^3 $ and $ \lambda \; g $ which again require $ g^2 \sim
598: \lambda $ for consistency. }
599:
600: \item {In terms of the effective field theory ratio $ (H/M_{Pl})^2 $ and
601: slow roll parameters, the dimensionless couplings are
602: \be
603: \frac{m \; g}{H} \sim \frac1{N} \;
604: \frac{H}{M_{Pl}} \qquad , \qquad
605: \lambda \sim \frac{1}{N^2} \, \left(\frac{H}{M_{Pl}} \right)^2
606: \label{lam} \; .
607: \ee
608: These relations agree with those found for the
609: dimensionless couplings in ref.\cite{pardec,qua} once the slow
610: roll parameters are identified with the expressions
611: (\ref{etav}) in terms of $1/N$. The results of
612: refs. \cite{pardec,qua} revealed that the loop expansion is indeed an expansion
613: in the effective field theory ratio $ (H/M_{Pl})^2 $ and the slow roll
614: parameters. The study in ref.\cite{clar} allows us to go further in this direction
615: and state that
616: the loop expansion is a consistent double series in the effective
617: field theory ratio $ (H/M_{Pl})^2 $ \emph{and} in $ 1/N $. Loops are either
618: powers of $ g^2 $ or of $ \lambda $ which
619: implies that for each loop there is a factor $ (H/M_{Pl})^2 $. The
620: counting of powers of $ 1/N $ is more subtle: the nearly scale invariant
621: spectrum of fluctuations leads to infrared enhancements of quantum
622: corrections in which the small factor $ 1/N $ enters as an infrared
623: regulator. Therefore, large denominators that feature the infrared
624: regulator of order $ 1/N $ cancel out factors $ 1/N $ in the
625: numerator. The final power of $ 1/N $ must be computed in
626: detail in each loop contribution. }
627:
628: \item{We find the relation (\ref{aco}) truly remarkable.
629: Since the scale of inflation $ M $ is fixed, presumably by the
630: underlying microscopic (GUT) theory, the scaling of $ \lambda $
631: with the inverse of the number of e-folds strongly suggests a
632: \emph{renormalization group explanation of the effective field
633: theory} because the number of e-folds is associated with the
634: logarithm of the scale $ N=\ln a $. A renormalization group
635: improved scale dependent quartic coupling \cite{weipes} behaves as
636: $ \lambda(K) \propto 1/\ln K $ with $ K $ the scale at which the
637: theory is studied. Since in an expanding cosmology the physical
638: scale grows with the scale factor it is natural to expect that a
639: renormalization group resummation yields the
640: renormalized coupling scaling as
641: $$
642: \lambda \sim 1/\ln a \sim 1/N \; .
643: $$
644: }
645: \end{itemize}
646:
647: There are several aspects of slow roll inflation, which when
648: considered together, lead us to conjecture that \emph{the
649: effective field theory of inflation is an almost critical theory
650: near but not at a trivial fixed point of the renormalization
651: group.} These aspects are the following:
652:
653: \begin{itemize}
654: \item{The fluctuations of the inflaton are almost \emph{massless},
655: this is the statement that the slow roll parameter
656: $$
657: \eta_V = M^2_{Pl} \; \frac{V''(\phi) }{V(\phi)} \simeq 3 \;
658: \frac{V''(\phi)}{H^2} \ll 1 \; .
659: $$
660: The slow roll relation (\ref{etav}) states that the dimensionless
661: ratio of the inflaton mass and the Hubble scale is $\sim 1/N\sim 1/\ln a $.}
662:
663: \item{The higher order couplings are suppressed by further
664: powers of $ 1/N \sim 1/\ln a $ [see eq.(\ref{Gn})].
665: In the language of critical
666: phenomena, the mass is a relevant operator in the infrared,
667: the quartic coupling $ \lambda $ is marginal,
668: and higher order couplings are irrelevant. }
669:
670: \end{itemize}
671:
672: These ingredients taken together strongly suggest that for large
673: $ N\sim \ln a $, the effective field theory is reaching a
674: \emph{trivial gaussian infrared fixed point}\cite{clar}. The evidence for
675: this is manifest in that:
676:
677: i) the power spectrum of scalar
678: fluctuations is \emph{nearly scale invariant} (a consequence of
679: $\eta_V \ll 1$),
680:
681: ii) the coupling constants
682: vanish in the asymptotic limit $N \rightarrow \infty$. \emph{If}
683: the number of e-folds were infinite, the theory would be
684: sitting at the trivial fixed point: a massless free field
685: theory.
686:
687: \medskip
688:
689: The physical situation in inflationary cosmology
690: is \emph{not} to be sitting \emph{exactly} at the fixed point,
691: inflation \emph{must} end, and is to be followed
692: by a radiation dominated standard big bang cosmology. Therefore,
693: we conclude that \emph{during the stage of slow roll inflation, the
694: theory is hovering near a trivial gaussian infrared fixed point}
695: but the presence of
696: a small relevant operator, namely the inflaton mass which
697: eventually becomes large at the end of slow roll, drives the
698: theory away from criticality.
699:
700: \medskip
701:
702: Our investigations\cite{clar} reveal that it is \emph{not} the
703: ultraviolet behavior of the renormalization group that is
704: responsible for the near criticality of the effective field theory,
705: but rather the infrared, superhorizon physics. That this is the
706: case can be gleaned in eq. (\ref{aco}): the coefficient
707: $(M/M_{Pl})^4$ in front of the term $1/\ln a $ \emph{cannot} be
708: obtained from the usual Minkowski-space renormalization group solution for
709: the running coupling. Furthermore, the true trivial fixed point is obtained in
710: the infrared limit when the scale factor $a\rightarrow \infty$, namely
711: infinitely long physical wavelengths.
712:
713: \bigskip
714:
715: If this conjecture\cite{clar} proves correct, it will have a major
716: fundamental appeal as a description of inflation because field
717: theories near a fixed point feature \emph{universal behavior}
718: independent of the underlying microscopic degrees of freedom. In
719: fact, if the effective field theory of
720: inflation is indeed near (not exactly) an infrared gaussian
721: fixed point, its predictions would be nearly
722: universal, and single field slow roll inflation describes a large
723: \emph{universality class} that features the same robust
724: predictions. Such is the case in critical phenomena
725: described by field theories where widely
726: different systems feature the same behavior near a critical point.
727:
728: \section{Quantum corrections to the equation of motion for the
729: inflaton and its effective potential.}\label{eftsr}
730:
731: We consider single field inflationary models described by a
732: general self-interacting scalar field theory in a spatially flat
733: Friedmann-Robertson-Walker cosmological space time with scale
734: factor $a(t)$. In comoving coordinates the action is given by
735: \begin{equation}\label{action}
736: S= \int d^3x \; dt \; a^3(t) \Bigg[ \frac{1}{2} \;
737: {\dot{\phi}^2}-\frac{(\nabla \phi)^2}{2a^2}-V(\phi) \Bigg] \;.
738: \end{equation}
739: We consider a \emph{generic} potential $V(\phi)$, the only
740: requirement is that its \emph{derivatives} be small in order to
741: justify the slow roll expansion\cite{hu,barrow}.
742: In order to study the corrections from the quantum fluctuations we
743: separate the classical homogeneous expectation value of the scalar
744: field from the quantum fluctuations by writing
745: \be\label{tad}
746: \phi(\vec{x},t)= \Phi_0(t)+\varphi(\vec{x},t)\;,
747: \ee
748: \noindent with
749: \be\label{exp}
750: \Phi_0(t)=\langle \phi(\vx,t) \rangle~~;~~ \langle
751: \varphi(\vx,t)\rangle =0 \;,
752: \ee
753: where the expectation value is in the non-equilibrium quantum
754: state. Expanding the Lagrangian density and integrating by parts, the
755: action becomes
756: \be\label{Split}
757: S= \int d^3x \; dt \; a^3(t)\,
758: \mathcal{L}[\Phi_0(t),\varphi(\vx,t)]\;,
759: \ee
760: \noindent with
761: \bea\label{lagraQ}
762: &&\mathcal{L}[\Phi_0(t),\varphi(\vx,t)] =
763: \frac{1}{2} \; {\dot{\Phi}^2_0}-V(\Phi_0)+\frac{1}{2} \;
764: {\dot{\varphi}^2}-\frac{(\nabla \varphi)^2}{2 \, a^2} -\frac{1}{2}\;
765: V^{''}(\Phi_0)\; \varphi^2 \cr \cr &&- \varphi\;
766: \left[\ddot{\Phi}_0+3 \, H \,\dot{\Phi}_0+V^{'}(\Phi_0)\right] -
767: \frac{1}{6}\; V^{'''}(\Phi_0)\; \varphi^3 - \frac{1}{24}\;
768: V^{(IV)}(\Phi_0)\; \varphi^4+ \textmd{higher orders in}\,
769: \varphi \; .
770: \eea
771: We obtain the equation of motion for the homogeneous
772: expectation value of the inflaton field by implementing the tadpole
773: method (see \cite{pardec} and references
774: therein). This method consists in requiring the condition $\langle
775: \varphi(\vx,t)\rangle =0 $ consistently in a perturbative expansion
776: by treating the \emph{linear}, cubic, quartic (and higher order)
777: terms in the Lagrangian density eq.(\ref{lagraQ}) as
778: \emph{perturbations}\cite{pardec}.
779:
780: Our approach relies on two distinct and fundamentally different
781: expansions: i) the effective field theory (EFT) expansion and ii)
782: the slow-roll expansion.
783:
784: Quantum corrections to the equations of motion for the inflaton and
785: for the fluctuations are
786: obtained by treating the second line in eq.(\ref{lagraQ}),
787: namely, the \emph{linear} and the non-linear terms in $\varphi$, in
788: perturbation theory.
789:
790: The generating functional of non-equilibrium real time correlation
791: functions requires a path integral along a complex contour in time:
792: the forward branch corresponds to time evolution forward $(+)$ and
793: backward $(-)$ in time as befits the time evolution of a density
794: matrix. Fields along these branches are labeled $\varphi^+$ and
795: $\varphi^-$, respectively (see
796: refs.\cite{pardec} and references therein).
797: The tadpole conditions
798: \be\label{tads}
799: \langle \varphi^\pm(\vx,t)
800: \rangle =0 \; , \ee \noindent both lead to the (same) equation of
801: motion for the expectation value $\Phi_0(t)$ by considering the
802: \emph{linear, cubic} and higher order terms in the Lagrangian
803: density as interaction vertices. To one loop order we find
804: \be\label{1lupeqn}
805: \ddot{\Phi}_0(t)+3 \, H \; \dot{\Phi}_0(t)+V'(\Phi_0)+g(\Phi_0) \; \langle
806: [\varphi^+(\vx,t)]^2\rangle =0 \;.
807: \ee
808: The first three terms in eq.(\ref{1lupeqn}) are the familiar ones for the
809: classical equation of motion of the inflaton.
810:
811: The last term is the one-loop correction to the equations of motion
812: of purely quantum mechanical origin. Another derivation of this
813: quantum correction can be found in\cite{cosmo,ramsey}.
814: The fact that the tadpole method, which in this case results in a one-loop
815: correction, leads to a covariantly conserved and fully renormalized energy
816: momentum tensor has been established in the most general case in
817: refs.\cite{cosmo,erice,mottola}.
818:
819: The coupling $g$, effective `mass term' $M^2$ and the quartic coupling
820: are defined by
821: \be \label{lambda}
822: M^2 \equiv M^2(\Phi_0) = V''(\Phi_0) = 3 \; H_0^2 \;
823: \eta_V + \mathcal{O}(\frac1{N}) \; , \;
824: g\equiv g(\Phi_0) = \frac{1}{2} \; V^{'''}(\Phi_0) \; , \;
825: \lambda \equiv \lambda (\Phi_0) = \frac{1}{6} \;
826: V^{(IV)}(\Phi_0)\; .
827: \ee
828: The $\langle(\cdots)\rangle$ is computed in the free
829: field (Gaussian) theory of the fluctuations $\varphi$ with an
830: effective `mass term' $M^2$, the
831: quantum state will be specified below. Furthermore, it is
832: straightforward to see that $\langle [\varphi^+(\vx,t)]^2\rangle =
833: \langle [\varphi^-(\vx,t)]^2\rangle=\langle
834: [\varphi(\vx,t)]^2\rangle$. In terms of the spatial Fourier
835: transform of the fluctuation field $\varphi(\vx,t)$, the one-loop
836: contribution can be written as
837: \be\label{lupPS}
838: \langle [\varphi(\vx,t)]^2\rangle = \int \frac{d^3 k}{(2\pi)^3} \; \langle
839: |\varphi_{\vk}(t)|^2 \rangle = \int_0^{\infty} \frac{dk}{k} \;
840: \mathcal{P}_{\varphi}(k,t)\,,
841: \ee
842: \noindent where $\varphi_{\vk}(t)$
843: is the spatial Fourier transform of the fluctuation field
844: $\varphi(\vx,t)$ and we have introduced the power spectrum of the
845: fluctuation
846: \be\label{PS} \mathcal{P}_{\varphi}(k,t) = \frac{k^3}{2
847: \, \pi^2} \; \langle |\varphi_{\vk}(t)|^2 \rangle \,.
848: \ee
849: During slow roll inflation the scale factor is quasi de Sitter and
850: to lowest order in slow roll it is given by :
851: \be\label{quasiDS}
852: C(\eta)=-\frac{1}{H_0 \; \eta} \; \frac{1}{1-\epsilon_V}=
853: -\frac{1}{H_0 \; \eta} (1+\epsilon_V) + \mathcal{O}(\epsilon_V^2) \,.
854: \ee
855: The spatial Fourier transform of the rescaled free field Heisenberg
856: operators $\chi(\vx,\eta) \equiv C(\eta) \varphi(\vx,t) $ obey the equation
857: \be\label{heiseqn}
858: \chi^{''}_{\vk}(\eta)+ \left[k^2 + M^2 \; C^2(\eta)-
859: \frac{C^{''}(\eta)}{C(\eta)} \right]\chi_{\vk}(\eta)=0 \,.
860: \ee
861: Using the slow roll expressions eqs.(\ref{lambda}) and (\ref{quasiDS}),
862: it becomes
863: \be\label{heiseqn2}
864: \chi^{''}_{\vk}(\eta)+ \left[k^2
865: -\frac{\nu^2-\frac{1}{4}}{\eta^2} \right]\chi_{\vk}(\eta)=0
866: \ee
867: \noindent where the index $ \nu $ and the quantity $ \Delta $ are
868: given by
869: \be\label{nu} \nu = \frac{3}{2} + \epsilon_V-\eta_V
870: +\mathcal{O}(\frac1{N^2}) \quad , \quad
871: \Delta= \frac{3}{2}-\nu = \eta_V-\epsilon_V +\mathcal{O}(\frac1{N^2}) \; .
872: \ee
873: The scale invariant case $ \nu = \frac{3}{2} $ corresponds to
874: massless inflaton fluctuations in the de Sitter background.
875: $ \Delta $ measures the departure from scale invariance.
876: In terms of the spectral index
877: of the scalar adiabatic perturbations $ n_s $ and the ratio $ r $ of
878: tensor to scalar perturbations, $ \Delta $ takes the form, \be
879: \Delta=\frac12 \left( n_s - 1 \right) + \frac{r}8 \; . \ee The free
880: Heisenberg field operators $\chi_{\vk}(\eta)$ are written in terms
881: of annihilation and creation operators that act on Fock states as
882: \be\label{ope} \chi_{\vk}(\eta) = a_{\vk} \; S_{\nu}(k,\eta)+
883: a^{\dagger}_{-\vk} \; S^{*}_{\nu}(k,\eta) \ee \noindent where the
884: mode functions $S_{\nu}(k,\eta)$ are solutions of the eqs.
885: (\ref{heiseqn2}). For Bunch-Davis boundary conditions we have
886: \be\label{BDS}
887: S_{\nu}(k,\eta) = \frac{1}{2}
888: \; \sqrt{-\pi\eta} \; e^{i\frac{\pi}{2}(\nu+\frac{1}{2})} \;
889: H^{(1)}_\nu(-k\eta)\, ,
890: \ee
891: this defines the Bunch-Davis vacuum $ a_{\vk} |0>_{BD} =0 $.
892:
893: There is no unique choice of an initial state, and a recent body of work
894: has began to address this issue (see ref.\cite{mottola}
895: for a discussion and further references). A full study of the
896: \emph{quantum loop} corrections with different initial
897: states must first elucidate the behavior of the propagators for the
898: fluctuations in such states. Here we focus on the standard
899: choice in the literature\cite{hu} which allows us to include the
900: quantum corrections into the standard results in the literature. A study
901: of quantum loop corrections with different initial states is an important
902: aspect by itself which we postpone to later work.
903:
904: The index $\nu$ in the mode functions eq.(\ref{BDS}) depends on the
905: expectation value of the scalar field, via the slow roll variables,
906: hence it slowly varies in time. Therefore, it is consistent to treat
907: this time dependence of $\nu$ as an \emph{adiabatic approximation}.
908: This is well known and standard in the slow roll
909: expansion\cite{hu}.
910: Indeed, there are corrections to the mode functions which are higher order
911: in slow roll.
912: However, these mode functions enter in the propagators in loop corrections,
913: therefore they yield higher order contributions in slow roll
914: and we discard them consistently to lowest order in slow roll.
915:
916: With this choice and to lowest order in slow roll, the power
917: spectrum eq.(\ref{PS}) is given by \be\label{PSSR}
918: \mathcal{P}_{\varphi}(k,t) = \frac{H^2}{8 \, \pi} \; (-k\eta)^3 \;
919: |H^{(1)}_\nu(- k \eta)|^2 \,. \ee For large momenta $|k\eta| \gg 1$
920: the mode functions behave just like free field modes in Minkowski
921: space-time, namely \be S_{\nu}(k,\eta) \buildrel{|k\eta| \gg
922: 1}\over= \frac{1}{\sqrt{2k}} \; e^{-ik\eta} \ee \noindent Therefore,
923: the quantum correction to the equation of motion for the inflaton
924: eqs.(\ref{1lupeqn}) and (\ref{lupPS}) determined by the momentum
925: integral of $ \mathcal{P}_{\varphi}(k,t) $ features both quadratic
926: and logarithmic divergences. Since the field theory inflationary
927: dynamics is an \emph{effective field theory} valid below a comoving
928: cutoff $\Lambda$ of the order of the Planck scale, the one loop
929: correction (\ref{lupPS}) becomes \be\label{PSint} \int^{{\Lambda}}_0
930: \frac{dk}{k} \,\mathcal{P}_{\varphi}(k,t) = \frac{H^2}{8 \, \pi}
931: \int^{\Lambda_p}_0 \frac{dz}{z} \; z^3\,
932: \left|H^{(1)}_\nu(z)\right|^2 \,, \ee \noindent where $
933: \Lambda_p(\eta)$ is the ratio of the cutoff in physical coordinates
934: to the scale of inflation, namely \be\label{physcut} \Lambda_p(\eta)
935: \equiv\frac{\Lambda}{H \; C(\eta)}=-\Lambda\, \eta\;. \ee The
936: integration variable $ z=-k \, \eta $ has a simple interpretation at
937: leading order in slow roll \be\label{zSR} z \equiv -k \, \eta =
938: \frac{k}{H_0 \, C(\eta)}= \frac{k_p(\eta)}{H_0} \,, \ee \noindent
939: where $k_p(\eta)=k/C(\eta)$ is the wavevector in physical
940: coordinates. If the spectrum of scalar fluctuations were strictly
941: scale invariant, (namely for massless inflaton fluctuations in de
942: Sitter space-time), then the index would be $\nu=3/2$ and the
943: integrand in (\ref{PSint}) given by \be\label{integ} z^3 \,
944: \left|H^{(1)}_{\frac{3}{2}}(z)\right|^2 =
945: \frac{2}{\pi}\left[1+z^2\right]\,. \ee In this strictly scale
946: invariant case, the integral of the power spectrum also features an
947: \emph{infrared} logarithmic divergence. While the ultraviolet
948: divergences are absorbed by the renormalization counterterms in the
949: effective field theory, this is not possible for the
950: infrared divergence. Obviously, the origin of this infrared behavior
951: is the {\bf exact} scale invariance of superhorizon fluctuations.
952: However, during slow roll inflation there are small corrections to
953: scale invariance, in particular the index $\nu$ is slightly different
954: from $3/2$ and this slight departure introduces a natural infrared
955: regularization. In ref.\cite{pardec} we have
956: introduced an expansion in the parameter $\Delta = 3/2-\nu=
957: \eta_V-\epsilon_V+\mathcal{O}(\epsilon^2_V,\eta^2_V,\epsilon_V\eta_V)$
958: which is small during slow roll and we showed in \cite{qua}
959: that the infrared divergences featured by
960: the quantum correction manifest as \emph{simple poles} in $\Delta$.
961:
962: The quantum correction to the equation of motion for the inflaton by
963: isolating the pole in $\Delta$ as well as the leading logarithmic divergences
964: were computed in ref.\cite{qua} with the result
965: \be\label{QC}
966: \frac12 \langle[\varphi(\vx,t)]^2\rangle = \left(\frac{H_0}{4 \, \pi}\right)^2
967: \left[ {\Lambda_p}^2 + \ln \Lambda_p^2 +\frac1{\Delta}
968: + 2 \, \gamma - 4 + \mathcal{O}(\Delta) \right]\,,
969: \ee
970: \noindent
971: where $\gamma$ is the Euler-Mascheroni constant. While the
972: quadratic and logarithmic \emph{ultraviolet} divergences are
973: regularization scheme dependent, the pole in $\Delta$ arises from
974: the infrared behavior and is independent of the regularization
975: scheme. In particular this pole coincides with that found
976: in the expression for $<\phi^2(\vx,t)>$ in ref.\cite{fordbunch}. The
977: \emph{ultraviolet divergences}, in whichever renormalization scheme,
978: require that the effective field theory be
979: defined to contain \emph{renormalization counterterms} in the bare
980: effective lagrangian, so that these counterterms will systematically
981: cancel the divergences encountered in the calculation of quantum
982: corrections in the (EFT) and slow roll approximations.
983:
984: \subsection{Renormalized effective field theory: renormalization counterterms}
985:
986: The renormalized effective field theory is obtained by writing the
987: potential $V[\phi]$ in the Lagrangian density eq.(\ref{action}) in
988: the following form \be\label{count}
989: V(\phi)=V_R(\phi)+\delta\,V_R(\phi,\Lambda)\;, \ee \noindent where
990: $V_R(\phi)$ is the renormalized \emph{classical} inflaton potential
991: and $\delta\,V_R(\phi,\Lambda)$ includes the renormalization
992: counterterms which are found systematically in a slow roll expansion
993: by canceling the ultraviolet divergences. In
994: this manner, the equations of motion and correlation functions in
995: this effective field theory \emph{are cutoff independent}.
996: We find from eqs.(\ref{1lupeqn}) and (\ref{QC})
997: \be\label{equalup}
998: \ddot{\Phi}_0(t)+3\,H\,\dot{\Phi}_0(t)+V'(\Phi_0)+ V^{'''}(\Phi_0)
999: \left(\frac{H_0}{4 \, \pi}\right)^2 \left[\Lambda_p^2+
1000: \ln\Lambda_p^2 +\frac{1}{\Delta}+2 \, \gamma - 4 +
1001: \mathcal{O}(\Delta)\right]=0 \; .
1002: \ee
1003: \noindent From this equation it becomes clear that the one-loop ultraviolet
1004: divergences can be canceled by choosing appropriate counterterms\cite{qua}
1005: leading to the final form of the
1006: renormalized inflaton equation of motion to leading order in the
1007: slow roll expansion
1008: \be\label{fineq}
1009: \ddot{\Phi}_0(t)+3\,H_0\,\dot{\Phi}_0(t)+V^{'}_R(\Phi_0)+
1010: \left(\frac{H_0}{4 \, \pi}\right)^2
1011: \frac{V^{'''}_R(\Phi_0)}{\Delta}=0 \;.
1012: \ee
1013: Although the quantum correction is
1014: of order $ V^{'''}_R(\Phi_0) $, (second order in slow
1015: roll), the strong infrared divergence arising from the quasi
1016: scale invariance
1017: of inflationary fluctuations brings about a denominator
1018: which is of first order in slow roll. Hence the lowest order
1019: quantum correction in the slow roll expansion,
1020: is actually of the same order as $ V^{'}_R(\Phi_0) $.
1021: To highlight this observation, it proves convenient to write
1022: eq.(\ref{fineq}) in terms of the EFT and slow roll parameters,
1023: \be\label{fineqsr}
1024: \ddot{\Phi}_0(t)+3\,H_0\,\dot{\Phi}_0(t)+
1025: V^{'}_R(\Phi_0)\left[1+\left(\frac{H_0}{2\pi \, M_{Pl}}\right)^2
1026: \frac{\xi_V}{2\,\epsilon_V\,\Delta}\right]=0 \;.
1027: \ee
1028: Since $\xi_V \sim \epsilon^2_V$ and $\Delta \sim \epsilon_V$ the leading quantum
1029: corrections are of zeroth order in slow roll. This is a consequence
1030: of the infrared enhancement resulting from the nearly scale
1031: invariance of the power spectrum of scalar fluctuations. The quantum
1032: correction is suppressed by an EFT factor $H^2/M^2_{Pl} \ll 1$.
1033:
1034: Restoring the dependence of $\Delta$ on $\Phi_0$ through the
1035: definitions (\ref{etav}) and (\ref{nu}) we
1036: find the equation of motion for the inflaton field
1037: to leading order in slow roll and in $ (H/M_{Pl})^2 $,
1038: \be\label{eqnslor}\ddot{\Phi}_0(t)+3\,H\,\dot{\Phi}_0(t)+
1039: V^{'}_R(\Phi_0)+\frac{1}{24 \, (\pi \; M_{Pl})^2} \;
1040: \frac{V_R^3(\Phi_0)\,
1041: V^{'''}_R(\Phi_0)}{2 \, V_R(\Phi_0)V^{''}_R(\Phi_0)-V^{'\,2}_R(\Phi_0)
1042: }=0 \;.
1043: \ee
1044:
1045: \subsection{Quantum corrections to the Friedmann equation:
1046: the effective potential}\label{friedeq}
1047:
1048: The zero temperature effective potential in Minkowski space-time is often
1049: used to describe the scalar field dynamics during inflation
1050: \cite{hu,riottorev}. However, as we see below the resulting effective potential
1051: [see eq.(\ref{Veff})] is remarkably different from the Minkowski one
1052: [see Appendix A]. The focus of this Section is to derive the effective potential for
1053: slow-roll inflation.
1054:
1055: Since the fluctuations of the inflaton field are quantized, the
1056: interpretation of the `scalar condensate' $\Phi_0$ is that of the
1057: expectation value of the full quantum field $\phi$ in a homogeneous
1058: coherent quantum state. Consistently with this, the Friedmann
1059: equation must necessarily be understood in terms of the
1060: \emph{expectation} value of the field energy momentum tensor, namely
1061: \be\label{FRW2}
1062: H^2= \frac{1}{3 \, M^2_{Pl}}\left\langle \frac{1}{2}
1063: \; \dot{\phi}^2+\frac{1}{2} \; \left(\frac{\nabla
1064: \phi}{a(t)}\right)^2+V[\phi] \right\rangle \;.
1065: \ee
1066: Separating the homogeneous condensate from the fluctuations as in eq. (\ref{tad})
1067: and imposing the tadpole equation (\ref{exp}), the Friedmann equation becomes
1068: \be\label{FRexp}
1069: H^2= \frac{1}{3 \, M^2_{Pl}}\left[ \frac{1}{2} \;
1070: {\dot{\Phi_0}}^2 + V_R(\Phi_0)+\delta V_R(\Phi_0)\right]+ \frac{1}{3
1071: \, M^2_{Pl}}\left\langle \frac{1}{2} \; \dot{\varphi}^2+\frac{1}{2}
1072: \; \left(\frac{\nabla \varphi}{a(t)}\right)^2+\frac{1}{2} \;
1073: V^{''}(\Phi_0)\; \varphi^2 +\cdots\right\rangle
1074: \ee
1075: The dots inside the angular brackets correspond to terms with higher
1076: derivatives of the potential which are smaller in the slow roll
1077: expansion. The quadratic term $\langle \varphi^2 \rangle$
1078: to leading order in slow roll is given by eq.(\ref{QC}).
1079: Calculating the expectation value in eq.(\ref{FRexp}) in
1080: free field theory corresponds to obtaining the corrections to the
1081: energy momentum tensor by integrating the fluctuations \emph{up to
1082: one loop}\cite{qua}.
1083: The first two terms of the expectation value in eq.(\ref{FRexp})
1084: \emph{do not} feature infrared divergences for $\nu=3/2$ because of
1085: the two extra powers of the loop momentum in the integral. These
1086: contributions are given by
1087: \bea\label{kinterm}
1088: &&\left\langle
1089: \frac{1}{2} \; \dot{\varphi}^2 \right\rangle = \frac{H^4_0}{16
1090: \,\pi} \; \int^{\Lambda_p}_{0} \frac{dz}{z} \; z^2 \;
1091: \left|\frac{d}{dz}\left[z^{\frac{3}{2}} H^{(1)}_{\nu}(z) \right]
1092: \right|^2 = \frac{H^4_0 \; \Lambda^4_p}{32 \,\pi^2}+
1093: \mathcal{O}( H^4_0 \Delta)\;,\\
1094: \label{grad} &&\left\langle\frac{1}{2}\left(\frac{\nabla
1095: \varphi}{a(t)}\right)^2 \right \rangle = \frac{H^4_0}{16 \,\pi} \;
1096: \int^{\Lambda_p}_{0} \frac{dz}{z} \; z^{5} \; \left|
1097: H^{(1)}_{\nu}(z) \right|^2 = \frac{H^4_0 \; \Lambda^4_p}{32 \,
1098: \pi^2}+ \frac{H^4 \; \Lambda^2_p}{16 \, \pi^2} +\mathcal{O}( H^4_0
1099: \Delta)\;.
1100: \eea
1101: The counterterms cancel the ultraviolet divergences arising from the third term
1102: in the angular brackets in eq. (\ref{FRexp})\cite{qua}. Finally,
1103: the fully renormalized Friedmann equation to one loop and to lowest order
1104: in the slow roll expansion is \cite{qua}
1105: \be\label{FRren} H^2 =
1106: \frac{1}{3 \, M^2_{Pl}}\left[ \frac{1}{2} \; {\dot{\Phi_0}}^2 +
1107: V_R(\Phi_0) + \left(\frac{H_0}{4 \,
1108: \pi}\right)^2\frac{V^{''}_R(\Phi_0)}{\Delta} +\textmd{higher orders
1109: in slow roll}\right] \equiv H^2_0 + \delta H^2 \;, \ee \noindent
1110: where $H_0$ is the Hubble parameter in absence of quantum
1111: fluctuations:
1112: $$
1113: H^2_0 = \frac{V_R(\Phi_0)}{3 \, M^2_{Pl}} \left[1+\frac{\epsilon_V}{3}+
1114: \mathcal{O}(\epsilon^2_V,\epsilon_V \; \eta_V) \right] \; .
1115: $$
1116: Using the lowest order slow roll relation eq. (\ref{lambda}), the
1117: last term in eq.(\ref{FRren}) can be written as follows
1118: \be\label{delH}
1119: \frac{\delta H^2}{H^2_0} = \left(\frac{H_0}{4 \,
1120: \pi\,M_{Pl}}\right)^2 \frac{\eta_V}{\Delta}\;.
1121: \ee
1122: This equation defines the back-reaction correction to the scale factor
1123: arising from the quantum fluctuations of the inflaton.
1124:
1125: Hence, while the ratio $\eta_V/\Delta$ is of order zero in slow roll,
1126: the one loop correction to the Friedmann equation is of the order
1127: $H^2_0/M^2_{Pl} \ll 1$ consistently with the EFT expansion.
1128: The Friedmann equation suggests the identification of the effective potential
1129: \bea\label{Veff}
1130: &&V_{eff}(\Phi_0) = V_R(\Phi_0)+ \left(\frac{H_0}{4
1131: \, \pi}\right)^2\frac{V^{''}_R(\Phi_0)}{\Delta} +\textmd{higher
1132: orders in slow roll} = \\ \cr
1133: &&= V_R(\Phi_0)\left[1+
1134: \left(\frac{H_0}{4 \, \pi\,M_{Pl}}\right)^2\frac{\eta_V}{
1135: \eta_V-\epsilon_V} + \textmd{higher orders in slow roll} \right]
1136: \; . \label{Vefsr}
1137: \eea
1138: We see that the equation of motion for the inflaton eq.(\ref{fineq})
1139: takes the natural form
1140: $$
1141: \ddot{\Phi}_0(t)+3\,H_0\,\dot{\Phi}_0(t)+
1142: \frac{\partial V_{eff}}{\partial\Phi_0}(\Phi_0)
1143: = 0 \;.
1144: $$
1145: where the derivative of $ V_{eff} $ with respect to $ \Phi_0 $
1146: is taken at fixed Hubble and slow roll parameters. That is,
1147: $ H_0 $ and $ \Delta $ must be considered in the present context
1148: as gravitational degrees of freedom and not as matter (inflaton)
1149: degrees of freedom.
1150:
1151: Eqs.(\ref{delH}) and (\ref{Veff}) make manifest the nature of the
1152: effective field theory expansion in terms of the ratio
1153: $ \left(H_0/M_{Pl}\right)^2 $. The coefficients of the powers of this
1154: ratio are obtained in the slow roll expansion. To leading order,
1155: these coefficients are of $ \mathcal{O}(N^0) $ because of the
1156: infrared enhancement manifest in the poles in $\Delta$, a
1157: consequence of the nearly scale invariant power spectrum of scalar
1158: perturbations.
1159:
1160: A noteworthy result is the rather different form of the effective
1161: potential eq.(\ref{Veff}) as compared to the result in Minkowski
1162: space time at zero temperature.
1163: In the appendix we show explicitly that the same
1164: definition of the effective potential as the expectation value of
1165: $T_{00}$ in Minkowski space-time at zero temperature
1166: is strikingly different from eq.(\ref{Veff}) valid for slow roll inflation.
1167:
1168: \section{Quantum Corrections to the Scalar and Tensor Power.
1169: Scalar, Tensor, Fermion and Light Scalar Contributions.}
1170:
1171: During slow-roll inflation many quantum fields may be coupled to the inflaton
1172: (besides itself) and can contribute to the quantum corrections to the equations
1173: of motion and to the inflaton effective potential. The scalar curvature
1174: and tensor fluctuations are certainly there. We also consider light fermions and
1175: scalars coupled to the inflaton.
1176: We take the fermions to be Dirac fields with a generic Yukawa-type coupling
1177: but it is straightforward to generalize to Weyl or Majorana fermions.
1178: The Lagrangian density is taken to be
1179: \be
1180: \mathcal{L} = \sqrt{-g} \; \Bigg\{ \frac12 \; \dot{\varphi}^2 -
1181: \left(\frac{\vec{\nabla}\varphi}{2\,a}\right)^2-V(\varphi)+\frac12 \;
1182: \dot{\sigma}^2 - \left(\frac{\vec{\nabla}\sigma}{2\,a}\right)^2 -
1183: \frac12 \; m^2_\sigma \; \sigma^2 -G(\varphi) \; \sigma^2+
1184: \overline{\Psi}\Big[i\,\gamma^\mu \; \mathcal{D}_\mu \Psi -m_f -
1185: Y(\varphi)\Big]\Psi \Bigg\} \label{lagrangian}
1186: \ee
1187: \noindent where $G(\Phi)$ and $Y(\Phi)$ are generic interaction terms
1188: between the inflaton and the scalar and fermionic fields.
1189: For simplicity we
1190: consider one bosonic and one fermionic Dirac field.
1191: The $\gamma^\mu$ are the curved space-time Dirac matrices
1192: and $ \mathcal{D}_\mu $ the fermionic covariant derivative\cite{anom,BD}.
1193: We will consider that the light scalar field $\sigma$ has vanishing
1194: expectation value at all times, therefore inflationary dynamics is
1195: driven by one single scalar field, the inflaton $\phi$.
1196:
1197: We consider the contributions from the quadratic fluctuations
1198: to the energy momentum tensor. There are \emph{four} distinct
1199: contributions: i) scalar metric (density) perturbations, ii) tensor
1200: (gravitational waves) perturbations, iii) fluctuations of the light
1201: bosonic scalar field $\sigma$, iv) fluctuations of the light
1202: fermionic field $\Psi$.
1203:
1204: Fluctuations in the metric are studied as
1205: usual \cite{hu,mukhanov}. Writing the metric as
1206: $$
1207: g_{\mu\nu}= g^0_{\mu\nu}+\delta^s g_{\mu\nu}+\delta^t g_{\mu\nu}
1208: $$
1209: \noindent where $g^0_{\mu\nu}$ is the spatially flat FRW background
1210: metric eq.(\ref{metrica}), $\delta^{s,t} g_{\mu\nu}$ correspond
1211: to the scalar and tensor perturbations respectively, and we neglect
1212: vector perturbations. In longitudinal gauge
1213: \be
1214: \delta^{s} g_{00} = C^2(\eta) \; 2 \; \phi \quad , \quad
1215: \delta^{s} g_{ij} = C^2(\eta) \; 2 \; \psi \; \delta_{ij}
1216: \label{curvpot} \quad , \quad
1217: \delta^{t} g_{ij} = -C^2(\eta) \; h_{ij}
1218: \ee
1219: \noindent where $ h_{ij} $ is transverse and
1220: traceless and we neglect vector modes since they are not
1221: generated in single field inflation\cite{hu,mukhanov}.
1222:
1223: Expanding up to quadratic order in the
1224: scalar fields, fermionic fields and metric perturbations the part
1225: of the Lagrangian density that is quadratic in these fields is
1226: given by
1227: $$
1228: \mathcal{L}_Q =\mathcal{L}_s[\delta\varphi^{gi},\phi^{gi},\psi^{gi}]+
1229: \mathcal{L}_t[h]+\mathcal{L}_\sigma[\sigma]+
1230: \mathcal{L}_\Psi[\overline{\Psi},\Psi] \; ,
1231: $$
1232: \noindent where
1233: \bea
1234: &&\mathcal{L}_t[h] = \frac{M^2_{Pl}}{8} \; C^2(\eta) \;
1235: \partial_\alpha h^j_i \; \partial_\beta h^i_j \; \eta^{\alpha \beta} \; ,
1236: \cr \cr &&
1237: \mathcal{L}_\sigma[\sigma]=
1238: C^4(\eta) \; \Bigg\{\frac{1}{2} \;
1239: \left(\frac{\sigma'}{C}\right)^2-\frac{1}{2} \; \left(\frac{\nabla
1240: \sigma}{ C}\right)^2 -\frac{1}{2} \; M^2_\sigma[\Phi_0]\;
1241: \sigma^2\Bigg\} \; , \cr \cr &&
1242: \mathcal{L}_\Psi[\overline{\Psi},\Psi]= \overline{\Psi}\Big[i \;
1243: \gamma^\mu \; \mathcal{D}_\mu \Psi -M_\Psi[\Phi_0]\Big]\Psi \; ,
1244: \nonumber
1245: \eea \noindent
1246: here the prime stands for derivatives
1247: with respect to conformal time and the labels (gi) refer to gauge
1248: invariant quantities\cite{mukhanov}. The explicit expression for
1249: $\mathcal{L}[\delta\varphi^{gi},\phi^{gi},\psi^{gi}]$ is given in
1250: eq. (10.68) in ref.\cite{mukhanov}. The effective masses for the
1251: bosonic and fermionic fields are given by
1252: \be
1253: M^2_\sigma[\Phi_0] = m^2_\sigma + G(\Phi_0) \label{sigmamass}
1254: \quad , \quad
1255: M_\Psi[\Phi_0] = m_f+Y(\Phi_0) \; .
1256: \ee
1257: We focus on the study of the quantum corrections to the Friedmann
1258: equation, for the case in which both the scalar and fermionic
1259: fields are light in the sense that during slow roll inflation,
1260: \be
1261: M_\sigma[\Phi_0]\; , \; M_\Psi[\Phi_0] \ll H_0 \; ,
1262: \ee
1263: \noindent at least during the cosmologically relevant stage corresponding to
1264: the 60 or so e-folds before the end of inflation.
1265:
1266: From the quadratic Lagrangian given above the quadratic quantum
1267: fluctuations to the energy momentum tensor can be extracted.
1268:
1269: The effective potential is identified with $\langle T^0_0 \rangle$
1270: in a spatially translational invariant state in which the
1271: expectation value of the inflaton field is $\Phi_0$. During slow
1272: roll inflation the expectation value $\Phi_0$ evolves very slowly
1273: in time, the slow roll approximation is indeed an adiabatic
1274: approximation, which justifies treating $\Phi_0$ as a constant in
1275: order to obtain the effective potential. The time variation of
1276: $\Phi_0$ only contributes to higher order corrections in
1277: slow-roll. The energy momentum tensor is computed in the
1278: FRW inflationary background determined by the \emph{classical}
1279: inflationary potential $V(\Phi_0)$, and the slow roll parameters
1280: are also explicit functions of $\Phi_0$. Therefore the energy
1281: momentum tensor depends \emph{implicitly} on $\Phi_0$ through the
1282: background and \emph{explicitly} on the masses for the light
1283: bosonic and fermionic fields given above.
1284:
1285: We can write the effective potential as
1286: \be
1287: V_{eff}(\Phi_0) = V(\Phi_0)+ \delta V(\Phi_0) \; , \label{Veff2}
1288: \ee
1289: \noindent where
1290: \be
1291: \delta V(\Phi_0)= \langle T^{0}_{0}[\Phi_0] \rangle_s + \langle
1292: T^{0}_{0}[\Phi_0] \rangle_t + \langle T^{0}_{0}[\Phi_0]
1293: \rangle_\sigma +\langle T^{0}_{0}[\Phi_0]
1294: \rangle_\Psi\label{dVeff}
1295: \ee
1296: \noindent $(s,t,\sigma,\Psi)$
1297: correspond to the energy momentum tensors of the quadratic
1298: fluctuations of the scalar metric, tensor (gravitational waves),
1299: light boson field $\sigma$ and light fermion field $\Psi$
1300: fluctuations respectively. Since these are the expectation values
1301: of a quadratic energy momentum tensor, $\delta V(\Phi_0)$
1302: corresponds to the \emph{one loop correction} to the effective
1303: potential.
1304:
1305: \subsection{Light scalar fields}
1306:
1307: During slow roll inflation the effective mass of the $\sigma$
1308: field is given by eq. (\ref{sigmamass}), just as for the inflaton
1309: fluctuation in sec. 2. It is convenient to introduce a parameter
1310: $\eta_\sigma$ defined to be
1311: \be\label{etasig} \eta_\sigma =
1312: \frac{M^2_\sigma[\Phi_0]}{3 \; H^2_0 } \; .
1313: \ee
1314: Hence, the sigma field contributions to the inflaton equations of motion
1315: and inflaton effective potential can be obtained from sec. 2 just replacing
1316: the slow roll parameter $ \eta_V $ by $ \eta_\sigma $. In particular,
1317: infrared divergences are now regulated by the parameter $ \Delta_\sigma \equiv
1318: \eta_\sigma - \epsilon_V $.
1319:
1320: To leading order in the slow roll expansion and in
1321: $\eta_\sigma \ll 1$, the infrared contribution is given by\cite{anom},
1322: $$
1323: \frac{M^2_\sigma[\Phi_0]}{2} \; \big \langle \sigma^2(\vec{x},t) \big
1324: \rangle =
1325: \frac{3\,H^4_0}{(4 \; \pi)^2} \; \frac{\eta_\sigma}{ \eta_\sigma-\epsilon_V}+
1326: \mathrm{subleading ~in ~ slow ~ roll}.
1327: $$
1328: The fully renormalized contribution from from the sigma field to $ T^0_0 $
1329: to leading order in slow roll takes the form \cite{anom},
1330: \be
1331: \langle T^0_0 \rangle_\sigma =
1332: \frac{3\,H^4_0}{(4 \; \pi)^2}\frac{\eta_\sigma}{ \Delta_\sigma}+
1333: \frac{1}{2}\,\Bigg\langle\dot{\sigma}^2+ \left(\frac{\nabla
1334: \sigma}{C(\eta)}\right)^2 \Bigg\rangle_{ren} \label{T00siglead}
1335: \ee
1336: In calculating here the second term we can set to zero the slow
1337: roll parameters $\epsilon_V, \eta_V$ as well as the mass of the
1338: light scalar, namely $\eta_\sigma=0$. Hence, to leading order,
1339: the second term is identified with the $00$ component of the
1340: renormalized energy momentum tensor for a free massless minimally
1341: coupled scalar field in exact de Sitter space time. Therefore
1342: we can extract this term from refs.\cite{fordbunch,BD},
1343: \be\label{TmunudS}
1344: \langle T_{\mu \nu}\rangle_{ren} = \frac{g_{\mu
1345: \nu}}{(4 \, \pi)^2}\Bigg\{m^2_\sigma \; H^2_0
1346: \left(1-\frac{m^2_\sigma}{2 \, H^2_0}\right)\left[-\psi\left(\frac{3}{2}+
1347: \nu\right)-\psi\left(\frac{3}{2}-\nu\right)+
1348: \ln\frac{m^2_\sigma}{H^2_0} \right]+
1349: \frac{2}{3} \; m^2_\sigma \; H^2_0-\frac{29}{30} \; H^4_0 \Bigg\} \; ,
1350: \ee
1351: where $ \nu \equiv \sqrt{\frac{9}{4}-\frac{m^2_\sigma}{H^2_0}} $ and
1352: $ \psi(z) $
1353: stands for the digamma function. This expression corrects a factor of two in
1354: ref.\cite{BD}. In eq. (6.177) in \cite{BD} the
1355: D'Alambertian acting on $G^{1}(x,x')$ was neglected. However, in
1356: computing this term, the D'Alambertian must be calculated
1357: \emph{before} taking the coincidence limit. Using the equation of
1358: motion yields the extra factor 2 and the expression eq.(\ref{TmunudS}).
1359:
1360: The pole at $ \nu=3/2 $ manifest in eq.(\ref{TmunudS})
1361: coincides with the pole in eq.(\ref{T00siglead}) using that
1362: $ m^2_\sigma = 3 \; H^2 \; \eta_\sigma $ [eq.(\ref{etasig})].
1363: This pole originates in the term $m_\sigma^2 \;
1364: <\sigma^2>$, which features an infrared divergence in the
1365: limit $\nu_\sigma=3/2$. All the terms with space-time derivatives are infrared
1366: finite in this limit. Therefore, we can extract from eq.(\ref{TmunudS})
1367: the renormalized expectation value in the limit $H_0 \gg m_\sigma$,
1368: \be
1369: \langle T^0_0 \rangle_\sigma =
1370: \frac{H^4_0}{(4 \; \pi)^2}\left[\frac{3\,\eta_\sigma}{
1371: \eta_\sigma-\epsilon_V}-\frac{29}{30}+
1372: \mathcal{O}(\epsilon_V,\eta_\sigma,\eta_V)\right]\label{Tsiglea}
1373: \ee
1374: The second term is completely determined by the {\bf trace
1375: anomaly} of the minimally coupled scalar fields\cite{fordbunch,BD,ddhf}.
1376:
1377: \subsection{Quantum Corrections to the Inflaton potential from
1378: the Scalar metric perturbations}
1379:
1380: The gauge invariant energy momentum tensor for quadratic scalar
1381: metric fluctuations has been obtained in ref.\cite{abramo}.
1382: In longitudinal gauge and in cosmic time it is given by
1383: \bea
1384: \langle T^0_0 \rangle_s = && M^2_{Pl} \Bigg[12 \; H_0 \; \langle \phi
1385: \dot{\phi} \rangle - 3 \; \langle (\dot{\phi})^2 \rangle +
1386: \frac{9}{C^2(\eta)} \; \langle (\nabla \phi)^2\rangle \Bigg] \nonumber \\
1387: & & + \frac{1}{2} \; \langle (\dot{\delta \varphi})^2\rangle + \frac{\langle
1388: (\nabla \delta \varphi)^2\rangle}{2\,C^2(\eta)} + \frac{V''(\Phi_0)}{2} \;
1389: \langle(\delta \varphi)^2\rangle + 2 \; V'(\Phi_0) \;
1390: \langle \phi \,\delta \varphi
1391: \rangle \label{T00s}
1392: \eea
1393: \noindent where the dots stand for derivatives with respect to cosmic time.
1394: In longitudinal gauge, the equations of motion in cosmic time for
1395: the Fourier modes are\cite{hu,mukhanov}
1396: \bea \label{phieq}
1397: &&\ddot{\phi}_{\vec k}+ \left(H_0-2 \;
1398: \frac{\ddot{\Phi}_0}{\dot{\Phi}_0}\right)\dot{\phi}_{\vec k}+
1399: \left[2 \; \left(\dot{H}_0-2 \; H_0 \;
1400: \frac{\ddot{\Phi}_0}{\dot{\Phi}_0}\right)+
1401: \frac{k^2}{C^2(\eta)}\right]{\phi}_{\vec k}=0 \cr \cr &&
1402: \label{delfieqn} \ddot{\delta \varphi}_{\vec k}+3 \; H \;
1403: \dot{\delta \varphi}_{\vec
1404: k}+\left[V''[\Phi_0]+\frac{k^2}{C^2(\eta)} \right]\delta
1405: \varphi_{\vec k}+2 \; V'[\Phi_0] \; \phi_{\vec k}- 4 \;
1406: \dot{\Phi}_0 \; \dot{\phi}_{\vec k}=0 \; ,
1407: \eea
1408: \noindent with the constraint equation
1409: \be \label{constraint} \dot{\phi}_{\vec
1410: k}+H_0 \; \phi_{\vec k}= \frac{1}{2M_{Pl}} \; \delta \varphi_{\vec
1411: k} \; \dot{\Phi}_0 \; .
1412: \ee
1413: We split the contributions to the energy momentum tensor as those
1414: from superhorizon modes, which yield the infrared
1415: enhancement, and the subhorizon modes for which we can set all
1416: slow roll parameters to zero. Just as discussed above for the case
1417: of the $\sigma$ field, since spatio-temporal derivatives bring
1418: higher powers of the momenta, we can neglect all derivative terms
1419: for the contribution from the superhorizon modes and set\cite{abramo}
1420: \be \langle
1421: T^0_0 \rangle_{IR} \approx \frac{1}{2} \; V''[\Phi_0] \; \langle
1422: \left(\delta \varphi (\vec{x},t)\right)^2 \rangle + 2 \;
1423: V'[\Phi_0] \; \langle \phi(\vec{x},t)\,\delta \varphi(\vec{x},t)
1424: \rangle \; .
1425: \ee
1426: The analysis of the solution of eq.(\ref{phieq})
1427: for superhorizon wavelengths in ref. \cite{mukhanov} shows that in
1428: exact de Sitter space time $\phi_{\vec k} \sim \mathrm{constant}$,
1429: hence it follows that during quasi-de Sitter slow roll inflation
1430: for superhorizon modes
1431: \be \label{fipun} \dot{\phi}_{\vec k} \sim
1432: (\mathrm{slow~roll}) \times H_0 \; \phi_{\vec k}
1433: \ee Therefore,
1434: for superhorizon modes, the constraint equation
1435: (\ref{constraint}) yields \be \label{rela} \phi_{\vec k} = -\,
1436: \frac{V'(\Phi_0)}{2 \; V(\Phi_0)} \; \delta \varphi_{\vec k} +
1437: {\rm higher ~ orders ~ in ~ slow ~ roll} \; . \ee Inserting this
1438: relation in eq.(\ref{delfieqn}) and consistently neglecting the
1439: term $\dot{\phi}_{\vec k}$ according to eq.(\ref{fipun}), we find
1440: the following equation of motion for the gauge invariant scalar
1441: field fluctuation in longitudinal gauge
1442: \be \label{delfieqn2}
1443: \ddot{\delta \varphi}_{\vec k}+3 \; H_0 \; \dot{\delta
1444: \varphi}_{\vec k}+\left[\frac{k^2}{C^2(\eta)}+3 \; H^2_0 \;
1445: \,\eta_\delta \right]\delta \varphi_{\vec k}=0 \; ,
1446: \ee \noindent
1447: where we have used the definition of the slow roll parameters
1448: $\epsilon_V; \; \eta_V$ given in eq.(\ref{etav}), and introduced
1449: $ \eta_\delta \equiv \eta_V-2 \; \epsilon_V $.
1450: This is the equation of motion for a minimally coupled scalar
1451: field with mass squared $3 \; H^2_0 \; \eta_\delta$ and we can
1452: use the results obtained in the case of the scalar field $\sigma$.
1453:
1454: Repeating the analysis presented
1455: in the case of the scalar field $\sigma$ above, we finally find\cite{anom}
1456: \be
1457: \langle T^0_0 \rangle_{IR} = \frac{3 \; H^4_0}{(4 \; \pi)^2}
1458: \; \frac{\eta_V-4\,\epsilon_V}{\eta_V-3 \, \epsilon_V} +
1459: \mathrm{subleading~in~slow~roll}
1460: \ee
1461: It can be shown that the contribution from subhorizon modes to
1462: $ \langle T^0_{0s} \rangle$ is given by \cite{anom}
1463: \be \langle T^0_{0s}\rangle_{UV} \simeq
1464: \frac{1}{2}\langle (\dot{\delta \varphi})^2 \rangle +
1465: \frac{\langle (\nabla \delta \varphi)^2\rangle}{2\,a^2}
1466: \ee
1467: \noindent where we have also neglected the term with $ V''[\Phi_0]
1468: \sim 3 \; H^2_0 \; \eta_V $ since $ k^2/a^2 \gg H^2_0 $ for
1469: subhorizon modes. Therefore, to leading order in slow roll we find
1470: the renormalized expectation value of $T_{00s}$ is given by
1471: \be
1472: \label{T00sfin} \langle T^0_{0s}\rangle_{ren} \simeq \frac{3
1473: H^4_0}{(4 \; \pi)^2} \frac{\eta_V-4\,\epsilon_V}{\eta_V-3 \,
1474: \epsilon_V} + \frac{1}{2}\Bigg\langle \dot{\delta \varphi}^2 +
1475: \left(\frac{\nabla \delta \varphi}{C(\eta)}\right)^2
1476: \Bigg\rangle_{ren}
1477: \ee To obtain the renormalized expectation
1478: value in eq.(\ref{T00sfin}) one can set all slow roll parameters
1479: to zero to leading order and simply consider a massless scalar
1480: field minimally coupled in de Sitter space time
1481: and borrow the result from eq.(\ref{Tsiglea}). We find\cite{anom}
1482: \be \langle
1483: T^0_{0s} \rangle_{ren} =
1484: \frac{H^4_0}{(4 \; \pi)^2}\left[\frac{\eta_V-4\,\epsilon_V}{\eta_V-3 \,
1485: \epsilon_V}-\frac{29}{30}+
1486: \mathcal{O}(\epsilon_V,\eta_\sigma,\eta_V)\right]\label{T00sfinal2}
1487: \ee
1488: The last term in eq. (\ref{T00sfinal2}) is completely determined
1489: by the {\bf trace anomaly} of a minimally coupled scalar field in de
1490: Sitter space time\cite{fordbunch,BD,ddhf}
1491:
1492: \subsection{Quantum Corrections to the Inflaton potential from
1493: the Tensor perturbations}
1494:
1495: Tensor perturbations correspond to massless fields with two physical
1496: polarizations. The energy momentum tensor for gravitons only depends on
1497: derivatives of the field $ h^i_j $ therefore its expectation value in
1498: the Bunch Davies (BD) vacuum does not feature infrared singularities in
1499: the limit $\epsilon_V \rightarrow 0$. The absence of infrared
1500: singularities in the limit of exact de Sitter space time, entails
1501: that we can extract the leading contribution to the effective
1502: potential from tensor perturbations by evaluating the expectation
1503: value of $T_{00}$ in the BD vacuum in exact de Sitter
1504: space time, namely by setting all slow roll parameters to zero. This
1505: will yield the leading order in the slow roll expansion.
1506:
1507: Because de Sitter space time is maximally symmetric, the expectation
1508: value of the energy momentum tensor is given by\cite{BD}
1509: \be \label{trace}
1510: \langle T_{\mu \nu} \rangle_{BD} = \frac{g_{\mu
1511: \nu}}{4} \; \langle T^{ \alpha}_{\alpha } \rangle_{BD}
1512: \ee
1513: \noindent and $ T^{ \alpha}_{\alpha }$ is a space-time constant,
1514: therefore the energy momentum tensor is manifestly covariantly
1515: conserved. A large body of work has been
1516: devoted to study the trace anomaly in de Sitter space time
1517: implementing a variety of powerful covariant regularization methods that
1518: preserve the symmetry\cite{ddhf,BD,fordbunch}
1519: yielding a renormalized value of $ \langle T_{\mu \nu} \rangle_{BD} $
1520: given by eq. (\ref{trace}). Therefore, the full
1521: energy momentum tensor is completely determined by the {\bf trace
1522: anomaly}. The contribution to the trace anomaly from gravitons has been
1523: given in refs.\cite{fordbunch,BD,ddhf}, it is
1524: \be \label{traza}
1525: \langle T^{\alpha}_{\alpha } \rangle_t = -\frac{717}{80 \; \pi^2} \; H^4_0
1526: \quad \mbox{and} \quad
1527: \langle T^0_{0} \rangle_t = -\frac{717}{320 \; \pi^2}\; H^4_0\; .
1528: \ee
1529:
1530: \subsection{Summary of Quantum Corrections to the Observable Spectra}
1531:
1532: In summary, we find that the effective potential at one-loop is given by,
1533: $$
1534: \delta V(\Phi_0) = \frac{H^4_0}{(4 \; \pi)^2} \Bigg[\frac{\eta_V-4 \;
1535: \epsilon_V}{\eta_V-3 \; \epsilon_V}+
1536: \frac{3\,\eta_\sigma}{\eta_\sigma-\epsilon_V}+
1537: \mathcal{T}_{\Phi}+ \mathcal{T}_s+\mathcal{T}_t +\mathcal{T}_{\Psi}+
1538: \mathcal{O}(\epsilon_V,\eta_V,\eta_\sigma,\mathcal{M}^2) \Bigg] \; ,
1539: $$
1540: \noindent where $(s,t,\sigma,\Psi)$ stand for the contributions
1541: of the scalar metric, tensor fluctuations, light boson field $\sigma$ and light
1542: fermion field $\Psi$, respectively. We have
1543: \be
1544: \mathcal{T}_{\Phi} = \mathcal{T}_s = -\frac{29}{30} \quad , \quad
1545: \mathcal{T}_t = -\frac{717}{5} \quad , \quad
1546: \mathcal{T}_{\Psi} = \frac{11}{60} \label{Tt}
1547: \ee
1548: The terms that feature the \emph{ratios } of combinations of slow
1549: roll parameters arise from the infrared or superhorizon
1550: contribution from the scalar density perturbations and scalar
1551: fields $\sigma$ respectively. The terms $\mathcal{T}_{s,t,\Psi}$
1552: are completely determined by the trace anomalies of scalar,
1553: graviton and fermion fields respectively. Writing $H^4_0 =
1554: V(\Phi_0) \; H^2_0/[3 \; M^2_{Pl}]$ we can finally write the effective
1555: potential to leading order in slow roll
1556: \be
1557: V_{eff}(\Phi_0) = V(\Phi_0)\Bigg[1+ \frac{H^2_0}{3 \; (4 \; \pi)^2 \;
1558: M^2_{Pl}}\Bigg( \frac{\eta_V-4\,\epsilon_V}{\eta_V-3 \; \epsilon_V}+
1559: \frac{3\,\eta_\sigma}{ \eta_\sigma-\epsilon_V}-\frac{2903}{20}\Bigg)
1560: \Bigg] \label{Veffin}
1561: \ee
1562: There are several {\bf remarkable} aspects of this result:
1563:
1564: i) the infrared
1565: enhancement as a result of the near scale invariance of scalar
1566: field fluctuations, both from scalar density perturbations as
1567: well as from a light scalar field, yield corrections of \emph{zeroth
1568: order in slow roll}. This is a consequence of the fact that
1569: during slow roll the particular combination $ \Delta_\sigma = \eta_\sigma-
1570: \epsilon_V $ of slow roll parameters yield a natural infrared cutoff.
1571:
1572: ii) the final one loop contribution to the effective potential displays the
1573: effective field theory dimensionless parameter $H^2_0/M^2_{Pl}$.
1574:
1575: iii) the last term is completely
1576: determined by the trace anomaly, a purely geometric result of the
1577: short distance properties of the theory.
1578:
1579: The quantum corrections to the effective potential lead to quantum corrections
1580: to the amplitude of scalar and tensor fluctuations with the result\cite{anom}
1581: \bea &&
1582: |{\Delta}_{k,eff}^{(S)}|^2 = |{\Delta}_{k}^{(S)}|^2 \left\{ 1+
1583: \frac23 \left(\frac{H_0}{4 \; \pi \; M_{Pl}}\right)^2
1584: \left[1+\frac{\frac38 \; r \; (n_s - 1) + 2 \; \frac{dn_s}{d \ln
1585: k}}{(n_s - 1)^2} + \frac{2903}{40} \right] \right\} \cr \cr &&
1586: |{\Delta}_{k,eff}^{(T)}|^2 =|{\Delta}_{k}^{(T)}|^2 \left\{ 1
1587: -\frac13 \left(\frac{H_0}{4 \; \pi \; M_{Pl}}\right)^2
1588: \left[-1+\frac18 \; \frac{r}{n_s - 1}+ \frac{2903}{20}
1589: \right]\right\} \; , \cr \cr &&
1590: r_{eff} \equiv \frac{|{\Delta}_{k,eff}^{(T)}|^2}{|{\Delta}_{k,eff}^{(S)}|^2}
1591: = r \; \left\{ 1-\frac13 \left(\frac{H_0}{4 \; \pi \; M_{Pl}}\right)^2
1592: \left[1+\frac{\frac38 \; r \; (n_s - 1) +
1593: \frac{dn_s}{d \ln k}}{(n_s - 1)^2} + \frac{8709}{20} \right] \right\} \; .
1594: \eea
1595: The quantum corrections turn out to {\bf
1596: enhance} the scalar curvature fluctuations and to {\bf reduce} the
1597: tensor fluctuations as well as their ratio $r$.
1598:
1599: Acknowledgment: H J de V thanks the organizers of JGRG15 for their
1600: kind hospitality in Tokyo.
1601:
1602: \appendix
1603:
1604: \section{One loop effective potential and equations of motion in
1605: Minkowski space-time: a comparison}
1606:
1607: In this appendix we establish contact with familiar effective
1608: potential both at the level of the equation of motion for the
1609: expectation value of the scalar field, as well as the expectation
1610: value of $T_{00}$.
1611:
1612: In Minkowski space time the spatial Fourier transform of the field
1613: operator is given by
1614: \be \varphi_{\vk}(t) = \frac{1}{\sqrt{2\omega_k}}\left[ a_{\vk} \;
1615: e^{-i\omega_k\,t} + a^{\dagger}_{-\vk} \;
1616: e^{i\omega_k\,t}\right]\,, \ee
1617: \noindent where the vacuum state is annihilated by $a_{\vk}$ and the
1618: frequency is given by
1619: \be
1620: \omega_k = \sqrt{k^2+V^{''}(\Phi_0)} \,.
1621: \ee
1622: The one-loop contribution to the equation of motion (\ref{1lupeqn})
1623: is given by
1624: \be\label{1lupmink}
1625: \frac{V^{'''}(\Phi_0)}{2} \; \langle
1626: [\varphi(\vx,t)]^2\rangle = \frac{V^{'''}(\Phi_0)}{8 \, \pi^2}
1627: \int^{\Lambda}_0 \frac{k^2}{\omega_k} \; dk = \frac{d}{d\Phi_0}
1628: \left[ \frac{1}{4\pi^2} \int^{\Lambda}_0 k^2 \; \omega_k \; dk
1629: \right] \,.
1630: \ee
1631: The expectation value of $T_{00}=\mathcal{H}$ (Hamiltonian density)
1632: in Minkowski space time is given up to one loop by the following
1633: expression
1634: \be \langle T_{00} \rangle = \left\langle
1635: \frac{1}{2} \; \dot{\phi}^2+\frac{1}{2} \; \left(\nabla \phi
1636: \right)^2+V(\phi) \right\rangle = \frac{1}{2} \; {\dot{\Phi_0}}^2 +
1637: V(\Phi_0)+ \left\langle
1638: \frac{1}{2} \; \dot{\varphi}^2+\frac{1}{2} \; \left({\nabla
1639: \varphi}\right)^2+\frac{1}{2} \; V^{''}(\Phi_0) \; \varphi^2
1640: +\cdots\right\rangle \;. \ee
1641: The expectation value of the fluctuation contribution is given by
1642: \be \left\langle
1643: \frac{1}{2} \; \dot{\varphi}^2+\frac{1}{2} \; \left({\nabla
1644: \varphi}\right)^2+\frac{1}{2} \; V^{''}(\Phi_0) \; \varphi^2
1645: +\cdots\right\rangle = \frac{1}{4\pi^2} \int^{\Lambda}_0
1646: k^2 \; \omega_k \; dk =
1647: \frac{\Lambda^4}{16\pi^2}+\frac{V^{''}(\Phi_0) \; \Lambda^2}{16 \,\pi^2}-
1648: \frac{[V^{''}(\Phi_0)]^2}{64 \, \pi^2}
1649: \ln\frac{4 \,\Lambda^2}{V^{''}(\Phi_0)} \;.
1650: \ee
1651: Renormalization proceeds as usual by writing the bare Lagrangian in
1652: terms of the renormalized potential and counterterms. Choosing the
1653: counterterms to cancel the quartic,
1654: quadratic and logarithmic ultraviolet divergences, we obtain the
1655: familiar renormalized one loop effective potential
1656: \be \label{potefM}
1657: V_{eff}(\Phi_0) = V_R(\Phi_0)+\frac{[V^{''}_R(\Phi_0)]^2}{64 \, \pi^2}
1658: \ln\frac{V^{''}_R(\Phi_0)}{M^2} \; ,
1659: \ee
1660: \noindent where $M$ is a renormalization scale. Furthermore from
1661: eq. (\ref{1lupmink}) it is clear that the equation of motion for
1662: the expectation value is given by
1663: \be\label{eqnmink}
1664: \ddot{\Phi}_0+ V^{'}_{eff}(\Phi_0)= 0 \;.
1665: \ee
1666:
1667:
1668: \begin{thebibliography}{99}
1669: \bibitem{guthsato} D. Kazanas, ApJ 241, L59 (1980);
1670: A. Guth, Phys. Rev. \textbf{D23}, 347 (1981);
1671: K. Sato, MNRAS, {\bf 195}, 467 (1981).
1672:
1673: \bibitem{pert}
1674: M. S. Turner, Phys. Rev. D48, 3502 (1993).
1675: S. Dodelson, W. H. Kinney, E. W. Kolb
1676: Phys. Rev. D56, 3207 (1997).
1677: S. M. Leach, A. R. Liddle, J. Martin, D. J. Schwarz,
1678: Phys. Rev. D66, 023515 (2002).
1679: N. Bartolo, S. Matarrese, A. Riotto, Phys. Rev. D64, 083514 (2001).
1680: N. Bartolo, E. Komatsu, S. Matarrese, A. Riotto, astro-ph/0406398.
1681:
1682: \bibitem{hu} See for example, W. Hu and S. Dodelson,
1683: Ann. Rev. Astron. Ap. \textbf{40}, 171 (2002); J. Lidsey, A.
1684: Liddle, E. Kolb, E. Copeland, T. Barreiro and M. Abney, Rev. of
1685: Mod. Phys. {\bf 69} 373, (1997). W. Hu, astro-ph/0402060.
1686: A. Riotto, hep-ph/0210162, M. Giovannini, astro-ph/0412601 (2004).
1687:
1688: \bibitem{WMAP} C. L. Bennett \emph{et.al.} (WMAP collaboration),
1689: Ap.J.Suppl. \textbf{148}, 1 (2003).
1690: A. Kogut \emph{et.al.} (WMAP collaboration),
1691: Ap.J.Suppl. \textbf{148}, 161 (2003).
1692: D. N. Spergel \emph{et. al. }(WMAP collaboration),
1693: Ap. J. Suppl.\textbf{148}, 175 (2003).
1694: H. V. Peiris \emph{et.al.} (WMAP collaboration),
1695: Ap. J. Suppl.\textbf{148}, 213 (2003).
1696:
1697: \bibitem{bgzk} V. A. Belinsky, L. P. Grishchuk, Ya. B. Zeldovich,
1698: I. M. Khalatnikov,
1699: Phys. Lett. {\bf B 155}, 232, (1985), JETP {\bf 62}, 195 (1985).
1700:
1701: \bibitem{cosmo}
1702: D. Boyanovsky, H. J. de Vega, in the Proceedings of the VIIth.
1703: Chalonge School, `Current Topics in Astrofundamental Physics', p.
1704: 37-97, edited by N. G. Sanchez, Kluwer publishers, Series C, vol.
1705: 562, (2001), astro-ph/0006446. D. Boyanovsky, D. Cormier, H. J.
1706: de Vega, R. Holman, S. P. Kumar, Phys. Rev. {\bf D 57}, 2166
1707: (1998). D. Boyanovsky, H. J. de Vega, R. Holman, Phys. Rev. {\bf D
1708: 49}, 2769 (1994). D. Boyanovsky, F. J. Cao and H. J. de Vega,
1709: Nucl. Phys. {\bf B 632}, 121 (2002).
1710: D. Boyanovsky, D. Cormier, H. J. de Vega, R. Holman,
1711: A. Singh, and M. Srednicki, Phys. Rev. \textbf{D56}, 1939 (1997);
1712: D. Boyanovsky, D. Cormier, H. J. de Vega, R. Holman,
1713: Phys. Rev. {\bf D55}, 3373 (1997).
1714:
1715: \bibitem{cosmo2} F. J. Cao, H. J. de Vega, N.
1716: G. Sanchez, Phys. Rev. {\bf D70}, 083528 (2004).
1717:
1718: \bibitem{staro} A. A. Starobinsky and J. Yokoyama, Phys.Rev.{\bf D50},
1719: 6357 (1994).
1720:
1721: \bibitem{anom} D. Boyanovsky, H. J. de Vega, N. S\'anchez,
1722: Phys. Rev. {\bf 72}, 103006 (2005).
1723:
1724: \bibitem{sd} S. Dodelson, {\em Modern Cosmology}, Academic Press, 2003.
1725:
1726: \bibitem{pardec} D. Boyanovsky, H. J. de Vega, Phys. Rev. {\bf D70},
1727: 063508 (2004). D. Boyanovsky, H. J. de Vega, N. S\'anchez,
1728: Phys. Rev. {\bf D 71}, 023509 (2005).
1729:
1730: \bibitem{qua} D. Boyanovsky, H. J. de Vega, N. S\'anchez, astro-ph/0503669.
1731:
1732: \bibitem{quir} H. Leutwyler, Ann. Phys. 235, 165 (1994), hep-ph/9409423.
1733: S. Weinberg, hep-ph/9412326 and `The Quantum Theory of Fields', vol.
1734: 2, Cambridge University Press, Cambridge, 2000.
1735:
1736: \bibitem{clar} D. Boyanovsky, H. J. de Vega, N. S\'anchez, astro-ph/0507595.
1737:
1738: \bibitem{nos} D. Cirigliano, H. J. de Vega, N. G. Sanchez,
1739: Phys. Rev. {\bf D 71}, 103518 (2005).
1740:
1741: \bibitem{weipes} S. Weinberg, \textit{The Quantum Theory of Fields
1742: I}, Cambridge University Press,Cambridge, 1995; M. E. Peskin and
1743: D. V. Schroeder, \textit{An Introduction to Quantum Field Theory},
1744: Advanced Book Program, Addison Wesley Pub. Co, Reading,
1745: Massachusetts, 1995.
1746:
1747: \bibitem{barrow} A. R. Liddle, P. Parsons , J. D. Barrow, Phys.
1748: Rev. \textbf{D50}, 7222 (1994).
1749:
1750: \bibitem{ramsey} S. A. Ramsey and B. L. Hu, Phys. Rev. \textbf{D56}.
1751: 661, (1997); \emph{ibid} 678, (1997).
1752:
1753: \bibitem{erice} D. Boyanovsky, D. Cormier, H. J. de Vega, R. Holman and
1754: S. P. Kumar, in the Proceedings of the VIth. Erice Chalonge School on
1755: Astrofundamental Physics, Eds. N. S\'anchez and A. Zichichi,
1756: Kluwer, 1998.
1757:
1758: \bibitem{mottola} P. Anderson, C. Molina-Paris, and E. Mottola,
1759: Phys. Rev. D72, 043515(2005).
1760:
1761: \bibitem{fordbunch} T. S. Bunch and P. C. W. Davies, Proc.
1762: R. Soc. London \textbf{A360}, 117 (1978); A. Vilenkin and L. H.
1763: Ford, Phys. Rev. \textbf{D26}, 1231 (1982).
1764: M. A. Castagnino, J. P. Paz and N. Sanchez, Phys.
1765: Lett. \textbf{B193}, 13 (1987).
1766:
1767: \bibitem{riottorev} D. H. Lyth , A. Riotto, Phys. Rept.
1768: \textbf{314}, 1 (1999).
1769:
1770: \bibitem{BD} N. D. Birrell and P. C. W. Davies, \emph{Quantum fields in curved
1771: space}, Cambridge Monographs in Mathematical Physics, Cambridge
1772: University Press, Cambridge, 1982.
1773:
1774: \bibitem{mukhanov} V. F. Mukhanov, H. A. Feldman , R. H.
1775: Brandenberger, Phys. Rept. \textbf{215}, 203 (1992).
1776:
1777: \bibitem{ddhf} M. J. Duff, Nucl. Phys. \textbf{B125}, 334 (1977); S.
1778: M. Christensen and M. J. Duff, Nucl. Phys. \textbf{B170}, 480
1779: (1980); Phys. Lett. \textbf{B76}, 571 (1978).
1780: J. S. Dowker and R. Critchley, Phys. Rev.
1781: \textbf{D15}, 1484 (1977); \emph{ibid} \textbf{D16}, 3390 (1977),
1782: \textbf{D13}, 224 (1976), \textbf{D13}, 3224 (1976).
1783: M. V. Fischetti, J. B. Hartle and B. L. Hu, Phys.
1784: Rev. \textbf{D20}, 1757 (1979).
1785: K. Fujikawa, Phys. Rev. \textbf{D23}, 2262 (1981).
1786:
1787: \bibitem{abramo} L. R. Abramo, R. H. Brandenberger and V. F.
1788: Mukhanov, Phys. Rev. \textbf{D56}, 3248 (1997); L. R. Abramo, Ph.
1789: D. Thesis, gr-qc/9709049.
1790: \end{thebibliography}
1791: \end{document}
1792:
1793:
1794: