J H E P 0 8 ( 2 0 1 3 ) 0 5 3 Published for SISSA by Springer Received: March 22, 2013 Revised: July 2, 2013 Accepted: July 16, 2013 Published: August 9, 2013 Chameleonic inflation Kurt Hinterbichler,a Justin Khoury,b Horatiu Nastasec and Rogerio Rosenfeldc aPerimeter Institute for Theoretical Physics, 31 Caroline St. N, Waterloo, Ontario, N2L 2Y5 Canada bCenter for Particle Cosmology, Department of Physics and Astronomy, University of Pennsylvania, 209 South 33rd Street, Philadelphia, PA 19104, U.S.A. cICTP South American Institute for Fundamental Research, Instituto de F́ısica Teórica, UNESP-Universidade Estadual Paulista, R. Dr. Bento T. Ferraz 271, Bl. II, Sao Paulo 01140-070, SP, Brazil E-mail: khinterbichler@perimeterinstitute.ca, jkhoury@sas.upenn.edu, nastase@ift.unesp.br, rosenfel@ift.unesp.br Abstract: We attempt to incorporate inflation into a string theory realization of the chameleon mechanism. Previously, it was found that the volume modulus, stabilized by the supersymmetric potential used by Kachru, Kallosh, Linde and Trivedi (KKLT) and with the right choice of parameters, can generically work as a chameleon. In this paper, we ask whether inflation can be realized in the same model. We find that we need a large extra dimensions set-up, as well as a semi-phenomenological deformation of the Kähler potential in the quantum region. We also find that an additional KKLT term is required so that there are now two pieces to the potential, one which drives inflation in the early universe, and one which is responsible for chameleon screening at late times. These two pieces of the potential are separated by a large flat desert in field space. The scalar field must dynamically traverse this desert between the end of inflation and today, and we find that this can indeed occur under the right conditions. Keywords: Cosmology of Theories beyond the SM, Field Theories in Higher Dimensions ArXiv ePrint: 1301.6756 c© SISSA 2013 doi:10.1007/JHEP08(2013)053 mailto:khinterbichler@perimeterinstitute.ca mailto:jkhoury@sas.upenn.edu mailto:nastase@ift.unesp.br mailto:rosenfel@ift.unesp.br http://arxiv.org/abs/1301.6756 http://dx.doi.org/10.1007/JHEP08(2013)053 J H E P 0 8 ( 2 0 1 3 ) 0 5 3 Contents 1 Introduction 1 2 The model 3 2.1 Chameleon region 4 2.2 Inflationary region 9 3 Inflationary dynamics and constraints 10 3.1 Gravitational reheating 13 3.2 Extra-dimensional volume during inflation 14 4 Traversing the “desert” 15 4.1 Kinetic-dominated phase 16 4.2 Radiation-dominated phase 17 4.3 Chameleon-coupled phase 18 4.4 Time evolution of M10 21 5 Conclusions 22 1 Introduction If our universe at high energies is governed by string theory, or any other higher-dimensional theory, then some of the extra dimensions have to be compactified and/or the standard model matter must be confined to a lower dimensional brane. The size and shape of the compactification manifold, as well as the position of our brane world within the extra dimensions, appear to four dimensional observers as fundamental scalar fields. Such fundamental scalars may play a primary role in the early universe as drivers of inflation. Today, at late times, there is no sign of such a fundamental light scalar field, despite many experimental searches designed to detect the fifth forces that such a field would naively mediate [1, 2]. If light scalars are present today, there must be some screening mechanism to explain why they have not been detected experimentally (see [3] for a review). Experiments generally look for canonical linear scalars, and are performed near the Earth. Screening mechanisms work by exploiting non-linearities which become important near the Earth, making the scalars difficult to detect. Three broad classes of screening mechanisms have been developed from a phenomeno- logical, bottom-up approach. These are the chameleon mechanism [4–18], which makes the effective mass of the scalar dependent on the local density of ordinary matter; the Vain- shtein/galileon mechanism [19–21], which makes the scalar kinetic term large near matter sources; and the symmetron/dilaton mechanism [22–25] (see also [26, 27]), which exploits spontaneous symmetry breaking to weaken the coupling of the scalar near matter sources. – 1 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 A way to put these ideas on firmer theoretical footing is to realize them in string theory. Since galileons generally propagate superluminally on certain backgrounds, their UV completion cannot be a standard local quantum field theory or perturbative string theory [28]. Chameleons and symmetrons, on the other hand, are described by canonical scalar fields with self-interaction potentials, and hence are compatible with a Wilsonian UV completion. Recently, some of us have presented a scenario for embedding the chameleon screening mechanism within supergravity/string theory compactifications [29]. (See [30, 31] for related studies of supersymmetric extensions of chameleon/symmetron/dilaton screen- ing.) In this approach, the chameleon scalar field is identified with a certain function of the volume modulus of the extra dimensions. The chameleon form for the potential and its coupling to matter arises from the Kachru, Kallosh, Linde and Trivedi (KKLT) construc- tion [32] for the superpotential for the volume modulus %, albeit with a negative coefficient a in the exponential ∼ eia%. The main focus of [29] was the late universe — various con- straints were found on the KKLT parameters which ensure that the scalar is screened from experiments today. These constraints, which are briefly reviewed in section 2, lead to a large extra dimension scenario, with the most natural case consisting of 2 large extra dimensions (with the remaining 4 dimensions stabilized near the 10d fundamental scale). In this paper, we extend this scenario to the early universe and ask whether is it pos- sible that the volume modulus also drives slow-roll inflation in the early universe. We find that this is indeed possible if we include the standard KKLT superpotential term ∼ eib% with a positive b, together with a prefactor that guarantees inflation, but whose details do not influence much the subsequent evolution. We will consider a semi-phenomenological example for such a prefactor as a proof of principle. We find constraints on the parame- ters by demanding that inflation lasts sufficiently long and that the amplitude of density perturbations match observations. Thus in total, our model consists of a standard KKLT term, which is responsible for driving inflation in the early universe, and the ∼ eia% term considered in [29], which is responsible for chameleon screening at late times. Only one of the terms is relevant at each epoch, so that the term responsible for inflation plays no role at late times, and the term responsible for chameleon screening plays no role at early times. The resulting model is economical, in that it involves a single scalar field that both drives inflation and acts as a chameleon today, and allows for the possibility that the inflaton field is directly observable experimentally today. (The idea that the inflaton still is a relevant field today appeared before, for instance in the quintessence form in [33], however we are not aware of this being considered before in the chameleon context.) Because the scalar field couples weakly to matter, reheating is inefficient and pro- ceeds through gravitational particle production [34, 35]. Inflation is therefore followed by a phase of kinetic domination. The scalar must exit the inflationary regime and reach the chameleon regime by the present time. It turns out that these two regions of the potential are separated by a large region with negligible potential spanning many M−1 Pl in distance — a “desert” — which the scalar must dynamically traverse in order to accomplish this tran- sition. In the “minimal” scenario, in which kinetic domination is followed by a standard phase of radiation domination, we derive constraints on the model parameters in order for – 2 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 the scalar to successfully traverse the desert and reach the chameleon region. We also con- sider an alternative scenario, in which the universe temporarily becomes matter-dominated following kinetic domination, thanks to a massive, unstable particle X which decays well before big bang nucleosynthesis (BBN). The coupling of the volume modulus to the matter stress-energy drives the field towards the chameleon region of the potential. We find that a successful transition is achieved for a broad range of initial conditions and BBN proceeds as usual provided that the particle lifetime lies within the range 2.5×10−3 a−4/3 s < τX < 1 s. The paper is organized as follows. In section 2 we present the model, constraining both the chameleon and inflation regions. In section 3 we analyze the consequences for inflation, deriving constraints on the relevant parameters, and discuss the reheating mech- anism. In section 4 we describe the physics needed for traversing the “desert” between the inflation and chameleon region, and derive constraints on parameters required for a successful transition. We summarize the results in section 5. 2 The model The model we consider is based on that of [29], in which a scenario was proposed for embed- ding the chameleon within a supergravity/string theory compactification. In this approach, the chameleon field is a function of the volume of the extra dimensions. The scenario uses the KKLT construction [32] for the potential and the superpotential, with an important dif- ference: the sign for the coefficient a in the exponential for the superpotential, Aeia% is neg- ative (a < 0). While the original KKLT scenario has a > 0, it was argued that even in the KKLT context one can obtain extra terms with a < 0, for instance by using gluino conden- sation on an extra D9-brane with magnetic flux as in [36],1 or by imposing T-duality invari- ance on gaugino condensation superpotentials for tori compactifications as in [37].2 Another difference is that the scale relating the dimensionless % with the radius of extra dimensions was assumed in [29] to be the 4d Planck scale MPl, instead of the 10d Planck scale M10, as in KKLT. Here we will instead follow the KKLT convention and work with 10d Planck scale. Our goal in this work is to extend this framework to a more complete scenario, which allows the chameleon to also play the role of the inflaton in the early universe. In addition to the Aeia% term responsible for chameleon physics in the late universe, we also include in the superpotential the standard KKLT term Beib%, with b > 0, to drive inflation at the fundamental Planck scale. To obtain a viable inflationary model, we will see that the coefficient B must have a particular dependence on %; while the assumed form for B(%) is phenomenologically motivated and not governed by fundamental considerations, we will 1The superpotential obtained has A = Ce−m9c, where 2πS = e−φ − ic0 is the dilaton modulus, c = 8π2/N9, m9 ∈ N is a magnetic flux on D9-branes, N9 is a number of D9-branes and the coupling function on the D9-branes is 1/g2D9 = |m9ReS − w9ReT|, with T = −i%. 2In terms of the dilaton modulus S and volume modulus T , one obtains generically superpotentials of the form W (S, T ) ∼ η(iT )−6 exp(−3S/8πb), with η(x) the Dedekind eta function. At large volume Re T → ∞, corresponding to the chameleon region, we obtain W ∝ eπT/2, with a coefficient which is again exponentially small in the dilaton modulus, exp(−3S/8πb), with b a renormalization group factor of order 1. – 3 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 see that our model predictions are not sensitive to this choice.3 In this way our framework unifies economically the idea of inflation and chameleon physics, both being defined by a single scalar: the volume modulus for the extra dimensions. In KKLT, the nonperturbative exponential superpotential arises for instance from Eu- clidean worldvolume instantons, such as Euclidean D3-branes wrapping 4-cycles of volume V4 in the compact space. These contribute e−SD3 factors, where SD3 is the Euclidean ac- tion of the wrapped branes [38]. The physical 10d Newton’s constant κ10,4 the 10d Planck mass M10, the string tension α′, and the string coupling gs are related by 1 2κ2 10 ≡ M8 10 2 = 1 (2π)7α′4g2 s . (2.1) The D3-brane tension is τ3 = √ π κ10 = M4 10 √ π = 2πM4 s gs , (2.2) where Ms ≡ (2π √ α′)−1 is the 10d string scale. The dimensionless variable appearing in the exponent of the superpotential is %, whose imaginary part measures the volume V4 of the 4-cycles wrapped by the Euclidean D3-branes: σ ≡ Im % = M4 10V4 √ π . (2.3) In terms of %, our desired form of the superpotential is W (%) = W0 +Ae−i|a|% +B(%)eib% , (2.4) where W0 is a constant. The first exponential, ∼ e−i|a|%, has the opposite sign to the usual KKLT term, as mentioned earlier, and is responsible for the chameleon mechanism in the late universe. The second exponential, with b > 0, has the usual sign and will be responsible for driving inflation. Here b can range anywhere from O(1) to � 1 (in KKLT, we can have b ∼ 1/Nc with Nc a number of D7-branes). Note that the two terms will be relevant in different regions: the a term will be exponentially small in the inflation (small volume) region, whereas the b will be exponentially small in the chameleon (large volume) region. The chameleon part of the potential is reviewed in section 2.1, whereas the inflationary part, in particular the B(%) prefactor, is discussed in section 2.2. 2.1 Chameleon region We first review the constraints from the chameleon part of the potential, neglecting the B term in (2.4): Wcham(%) 'W0 +Ae−i|a|% . (2.5) 3By this we mean that the further evolution (the various stages described in the paper) of the cosmological model depends very weakly on the details of the inflationary region. Of course, the usual constraints on the inflationary potential apply, which will give constraints on the parameters of the inflationary region. 4Note that usually one defines an unphysical 1/2κ2 10 and a physical 1/2κ2 = 1/(g2s2κ2 10), but we will call κ10 the physical one, not to confuse with the 4 dimensional one. – 4 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 It is convenient to perform the analysis in terms of a new dimensionless variable R, which measures the volume of the extra dimensions in 10 dimensional Planck units:5 ds2 D = R2ds2 4 + gαβdxαdxβ ; R = 1√ V6M6 10 . (2.6) The precise relation between R and σ depends on the hierarchy of scales in the extra dimensions. For concreteness, we will consider the case where n of the extra dimensions are large and of comparable size r, while the remaining 6− n dimensions are of the order of the fundamental scale M−1 10 . In this case, R = 1√ V6M6 10 = 1 (rM10)n/2 . (2.7) Meanwhile, the 4d Planck scale upon dimensional reduction is given by6 M2 Pl ≡M8 10V6 = rnMn+2 10 . (2.8) For σ, we note that the leading instanton corrections are those with the largest action. These will be the instantons which wrap the largest cycles, i.e., those wrapping all the large dimensions. If n ≥ 4, then V4 = r4. If n ≤ 4, on the other hand, then V4 = rnMn−4 10 . In other words, in this case the definition of σ in (2.3) gives σ =  M4 10r 4√π = 2πM4 s r 4 gs = R− 8 n √ π for n ≥ 4 , Mn 10r n√π = 2n/4π(n+4)/8Mn s r n g n/4 s = R−2√π for n ≤ 4 , (2.9) where in the last step we have expressed σ in terms of R. Up to an irrelevant factor of √ π, which can be absorbed in a redefinition of parameters, we see that σ = R−k ; k =  8 n for n ≥ 4 , 2 for n ≤ 4 . (2.10) As shown in our previous paper [29], and as we will review shortly, the resulting potential in the chameleon region is dominated by a steep exponential part for values R < R? and exhibits a minimum at Rmin. (This minimum is originally an AdS minimum, but following KKLT we lift it to a suitable dS minimum via a supersymmetry breaking term.) The value of the field today is close to R∗, hence the constraint from laboratory experiments is naturally expressed in terms of R∗. In [29], we obtained the condition aR−k∗ & 1030 . (2.11) 5An important difference with [29] — see eq. (3.6) there — is that we have now defined R in terms of M10, consistent with the KKLT conventions, instead of MPl. 6Here MPl denotes the reduced Planck scale: MPl ≡ (8πGN)−1/2 ' 2.45× 1018 GeV. – 5 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 As shown in [29], this translates into a bound on the chameleon Compton wavelength at atmospheric density of m−1 lab . cm, while in the solar system the bound is m−1 solar . 105 km. Using (2.7), the above inequality translates into a bound on the present size of the large extra dimensions r∗ and the fundamental Planck scale M10. Combining with the def- inition of MPl given in (2.8), we can constrain r∗ and M10 separately. The result is clearly sensitive to the number n of large extra dimensions. If n = 6, for instance, corresponding to all extra dimensions being large, we find M10 . 25a3/4 keV ; r∗ & 100 a µm (n = 6) . (2.12) With a ∼ O(1), M10 is clearly ruled out by particle colliders, which preclude the existence of Kaluza-Klein modes with mass up to at least ∼TeV. Repeating this exercise for different number of large extra dimensions, we find that the only allowed possibility is n = 2. In this case, (2.11) implies M10 . 2.5a1/2 TeV ; r∗ & 100 a µm (n = 2) , (2.13) which are just at the experimental limit with a ∼ O(1).7 Although one might naively think that the constraints are even weaker for n = 1, this is not the case — the bound on M10 remains the same, but r∗ must be absurdly large. (Unlike the Arkani-Hamed- Dimopoulous-Dvali (ADD) scenario [39, 40], where n > 2 is generically allowed, in our case the chameleon constraints only allow n = 2, as (2.12) illustrates.) We are therefore led to consider a scenario with n = 2 large extra dimensions, with the remaining 4 extra dimensions of size at the fundamental scale M10. To avoid conflicting with collider experiments, the standard model fields must be confined to a brane, with the two large extra dimensions extending transverse to the brane. In other words, we need to live at least on a D7-brane, if not a Dp-brane with p < 7. The matter on the brane will couple to the induced metric on the brane, gbrane µν = gMN∂µX M∂νX N , (2.14) where XM are the embedding coordinates into the 10 dimensional space-time. For sim- plicity, we demand that the brane is situated at a fixed point in these transverse extra dimensions, so there are no dynamical fluctuations for the brane position. (While not necessary, this choice avoids dealing with the additional light scalars describing the brane position.) In this case, gbrane µν is just the 10d metric gMN restricted to the brane position. Hence gbrane µν is still the same gMN Jordan frame metric. 7When considering inflation later in the paper, we will find that a � 1 is necessary for the field to successfully traverse the desert and reach the chameleon region. In this case, it may be possible for n > 2 scenarios to be phenomenologically viable. For concreteness, however, we henceforth focus on n = 2, since this is the only case which allows a ∼ O(1). Also note that in the examples we have cited for a, it is a ratio of integers defining a brane construction. This means it can get to be even a� 1, though O(1) is more natural. – 6 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 We are now in a position to specify the Kähler potential, which will define the canonical scalar field in terms of σ. In the perturbative region of the chameleon, we can use the tree- level Kähler potential found in string theory for the volume modulus,8 K(%, %̄) ' −2M2 Pl ln[−i(%− %̄)] . (2.17) The corresponding kinetic term for σ is 1 2σ2 M2 Pl(∂σ)2 , (2.18) hence the canonical scalar field is identified as φ = MPl ln σ σ∗ , (2.19) where we have set φ = 0 at the present time. The supersymmetry potential that results from the superpotential (2.5) and the Kähler potential in (2.17) is VSUSY(σ)= 1 2M2 Pl [ A2a2e2|a|σ − 2A|a| σ e|a|σ ( W0 +Ae|a|σ ) − 1 2σ2 ( W0 +Ae|a|σ )2 ] . (2.20) We have the relation σ = 1/R2, which follows from (2.10) for n = 2. This potential has an AdS minimum at σmin = 1 R2 min ≈ 1 |a| ln W0 A , (2.21) which results from the combination of several exponentials in (2.20). Meanwhile, |a|(σ − σmin) & 1, the potential is well approximated by the leading exponential. This defines the value R∗ (or σ∗) introduced earlier. Specifically, for R . R∗ (σ & σ∗), the potential is a steep exponential, V (σ) ∼ e2|a|σ. As mentioned earlier, we have yet to supplement (2.20) with a symmetry-breaking term to lift the AdS minimum to a dS minimum with the right cosmological constant. Following KKLT, this is achieved by introducing an antibrane, which contributes to the potential: VSUSYbreak = D σ2 . (2.22) Since, for the parameters listed in (2.24), the depth of the AdS minimum Vtoday ∼ −10−150M4 Pl is negligible compared to the observed positive cosmological constant, the antibrane contribution is to a good approximation constrained to match the observed value D/σ2 ∗ = 10−122M4 Pl. With σ∗ ∼ 1030/a, this fixes D ∼ 10−62a−2M4 Pl. 8Our form for K can be justified as follows. If all the 6 dimensions are large, then K = −3M2 Pl ln[−i(ρ− ρ̄)] . (2.15) When reducing on n large dimensions down to d=4, we obtain in Einstein frame M2+n 10 2 ∫ d4+nx √ g(4+n)R(4+n) = M2 Pl 2 ∫ d4x √ g(4) [ R(4) − n (n 2 + 1 ) gµν∂µψ∂νψ ] , (2.16) where r = eψ, hence ρ ∝ ienψ for n ≤ 4. For n = 2, this is obtained from the stated Kähler potential. – 7 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 -20 -18 -16 -14 -12 -10 -8 Σ-Σ0 5. ´ 10-9 1. ´ 10-8 1.5 ´ 10-8 2. ´ 10-8 2.5 ´ 10-8 3. ´ 10-8 VchamHΣL Figure 1. The potential (in Planck units) in the chameleon region consists of an exponentially steep part for large σ, with minimum at σ = σmin. See the main text for details. Here A = e−|aσ0 . The total potential in the chameleon region, Vcham(σ) = VSUSY(σ) + VSUSYbreak(σ), is sketch in figure 1. For illustrative purposes, we chose the numerically-friendly values (in Planck units) |a| = 1, W0 = 10−2, σ0 = 102, A = e−|a|σ0 and D = 0.78W 2 0 . The chameleon potential is constrained by various phenomenological considerations, detailed in [29]. For concreteness, we will adopt the limiting value of R∗ ' 10−15 √ a allowed by (2.11) — recall that k = 2 for the case n = 2 of interest — corresponding to σ∗ ' 1030 a . (2.23) The parameters of the superpotential are then fixed as9 W0 ∼ 10−30a−3/2M3 Pl ; log ( M3 Pl A ) ∼ 1030 . (2.24) (While this naively implies a super-exponentially small value for A, one must keep in mind that the combination of parameters in the potential is Ae|a|σ∗ , with σ∗ ∼ 1030/a, hence Ae|a|σ∗ can be comparable to W0 ' 10−30M3 Pl at the minimum). Finally, in the Einstein frame in which we will be working, the chameleon field φ couples conformally to matter on the brane as F (φ)Tµµ, where Tµν is the matter stress tensor. For non-relativistic matter, this reduces to F (φ)ρmatter. The coupling function, F (φ) ≡ R R∗ = e−φ/2MPl , (2.25) relates the (D-dimensional) Jordan frame with the (4-dimensional) Einstein frame. (Note that the coefficient in the exponent depends on n; in particular, in the n = 6 case we had a different exponent.) 9Although some of the assumptions in [29] were different, such as having n = 6 large extra dimensions instead of 2, the expression for the potential in terms of σ is qualitatively unchanged, hence the results of the phenomenological analysis directly carry over to the present case. – 8 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 2.2 Inflationary region We have seen that at present time, corresponding to the chameleon region of the potential, 2 of the extra dimensions must be large. In the early universe, on the other hand, we envision that inflation takes place at energy near the 10d fundamental scale M10, with all extra dimensions at the fundamental scale: r ∼ M−1 10 , or σ ∼ 1. In this region, we can neglect the first exponential in (2.4) and approximate the superpotential as Winf(%) ' B(%)eib% . (2.26) Furthermore, to obtain a sufficiently flat potential to drive inflation, we need to modify the Kähler potential (2.17) as well. This is in any case to be expected, since the tree- level Kähler potential should receive α′ corrections in the stringy region r ∼ M−1 10 . The resulting modifications generally occur within the logarithm. For instance, KKLMMT [41] considered the effect of D-brane positions Φ, giving (in our case of 2 large dimensions) K = −2M2 Pl ln ( −i(%− %̄)− κ(Φ, Φ̄) ) . (2.27) In general, perturbative corrections give polynomial corrections inside the log, while non- perturbative corrections involve exponentials (see, e.g., [37]). We now specify suitable forms for B(%) and κ to achieve successful inflation. While the particular B(%) we will use is admittedly contrived, we will argue in section 4 that our results are largely insensitive to the particular forms for B(%) and κ, as long as the resulting potential displays an inflationary plateau, drops to zero as governed by the KKLT exponential e−2bσ, and is followed by a wide “desert”. The important physics for our considerations is captured by the KKLT exponential e−2bσ, and in this sense we describe physics associated with the KKLT model. Therefore we can view the form of B(%) and κ specified here as a proof of principle. Our analysis applies more generally to any nonperturbative deformation that gives this kind of inflation. A simple supergravity model that allows slow-roll inflation is [41, 42] K(%, %̄) = M2 Pl%̄% and W (%) ∼ %, corresponding to V (σ) ∼ 1 +σ4 + . . .. Following this approach, we consider nonperturbative corrections to the Kähler potential of the form10 K(%, %̄) = −2M2 Pl ln [ −i(%− %̄) + λe−α%%̄ ] . (2.28) As desired, this reduces to (2.17) at large % and tends to K(%, %̄) ' 2αM2 Pl%̄% at small %. Since we are instead interested specifically in new inflation, where the potential slowly drops from an initial value and then faster down to a flat plateau (“the desert”), we will need two additional factors in the superpotential: i) a factor of e−β1% 4 ∼ e−β1σ4 (setting % = iσ) to win over the e2ασ2 factor in the potential at large σ; and ii) a non-perturbative factor of e−β2e c/i% ∼ e−β2e −c/σ , which triggers the end of inflation. Both factors are to a good approximation equal to unity during inflation (at small σ), and hence will be irrelevant for 10Note that, since Im% ∼ V4/α ′2, the correction inside the log is ∼ (α′2/V4)e−V 2 4 /α ′2 with respect to the leading term. – 9 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 inflationary observables.11 We therefore choose B(%) of the form: B(%) = Be−β1% 4−β2ec/i%% , (2.29) where B is a constant. We should stress that it is not clear how this form would arise in string theory. The polynomial terms can conceivably arise as perturbative corrections to the instanton term eib%, but the prefactor e−β1% 4−β2ec/i% is hard to justify. We should also mention that achieving new inflation within supergravity is very difficult. Essentially, with the condition that after the drop from the inflationary plateau, there is no potential barrier higher than the plateau, and the scalar can flow unempeded towards the chameleon region, the form of B(%) seems almost uniquely determined. While we are not sure if it is truly unique, we have been unable to find another example. Here all the parameters β1, b2 and c are required for the above shape of the potential, and some possible values for them will be given later on. Therefore, for concreteness we will assume this form, in order to have a specific working inflationary model. As mentioned earlier, any nonperturbative modification that achieves inflation will work for our purposes, and we consider this form of B(%) just as a proof of concept. The complete superpotential is then W (%) = W0 +Ae−i|a|% +Beib%e−β1% 4−β2ec/i%% . (2.30) Our model is thus fully specified by the Kähler potential (2.28) and the superpoten- tial (2.30). In the next section, we will focus on the inflationary dynamics and derive constraints on the various parameters to achieve successful inflation. 3 Inflationary dynamics and constraints In this section we describe in more detail the inflationary epoch, which is assumed to occur near the 10d fundamental scale M10, with the extra dimensions of size r ∼ M−1 10 (i.e., σ ∼ 1). In this regime, the superpotential is approximately given by (2.26), with B(%) given in (2.29): Winf(%) ' Beib%e−β1%4−β2ec/i%% . (3.1) Since the real component of % is assumed to be stabilized [32], we will set % = iσ henceforth. Through trial and error, we have scanned different values of the parameters to seek regions in parameter space where inflation is possible. We found that inflation can be achieved in the region ασ2 � bσ � 1 , (3.2) provided that the model parameters satisfy αλ2 � 1 ; bλ� 1 . (3.3) 11The non-analytic factor of e−β2e −c/σ is necessary to prevent the occurrence of a maximum whose height is higher than the inflationary plateau, thus preventing the field from classically reaching the chameleon region of the potential. – 10 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 The condition bσ � 1 is intuitively clear, since the superpotential depends exponentially on this combination. The other conditions will be motivated below. We should stress that the above conditions are by no means necessary. Inflation may well be possible for other parameter values, but among various possible simplifying approximations, this is the only one we have found to be compatible with inflation. Starting with the kinetic term, the field-space metric that derives from the Kähler potential (2.28) is g%%̄ = 2M2 Pl ( 1 + λ2αe−2α%%̄ − i(%− %̄)λαe−α%%̄(1− α%%̄) [−i(%− %̄) + λe−α%%̄]2 ) ' 2M2 Plα , (3.4) where in the last step we have set % = iσ and used the conditions ασ2 � 1 and λα2 � 1 from (3.2) and (3.3), respectively. This simple form for the Kähler metric implies a linear relation between σ and the canonically-normalized scalar field φ ' 2 √ αMPlσ . (3.5) Meanwhile, ignoring contributions from the e−β1σ 4−β2e−c/σ prefactor, the scalar poten- tial in Einstein frame is approximately given by Vinf(σ) ' B2 2αλ2M2 Pl { 1− 4bσ ( 1 +O ( 1 bλ )) + 7b2σ2 ( 1 +O ( 1 bλ )) + . . . } , (3.6) where the ellipses include terms of O(α2σ4), which by (3.2) are negligible compared to the O(b2σ2) terms we have kept. The ellipses also include terms of higher-order in σ, which only become important after inflation. With these approximations, we can constrain various phenomenological quantities: • Slow-roll parameters: the standard ε and η slow-roll parameters are ε = M2 Pl 2 ( dσ dφ V,σ V )2 ' 2b2 α ; η = M2 Pl ( d2σ dφ2 V,σ V + ( dσ dφ )2 V,σσ V ) ' 7b2 2α . (3.7) The spectral tilt of the scalar power spectrum is red-tilted [43]: ns − 1 = −6ε+ 2η ' −5b2 α . (3.8) Meanwhile, the tensor-to-scalar ratio is given by [43] r = 16ε ' 32b2/α. (3.9) • Primordial amplitude: the observed primordial amplitude constrains [44] H2 inf 8π2εM2 Pl ' 2.4× 10−9 , (3.10) – 11 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 0.2 0.4 0.6 0.8 Σ 2.´10-13 4.´10-13 6.´10-13 8.´10-13 1.´10-12 VinfHΣL Figure 2. The region of the potential (in Planck units) relevant for inflation, for the parameter values (3.15). where H2 inf = Vinf/3M 2 Pl. Since Vinf ' B2/2αλ2M2 Pl, and substituting (3.7), we obtain B ' 10−3λbM3 Pl . (3.11) • Number of e-folds: as mentioned earlier, by assumption the end of inflation is triggered by the e−β2e −c/σ factor. This factor becomes relevant when the field reaches σend−inf ∼ c/ log(2β2). Numerically, we have found that this estimate is off by a factor of 2, hence a more accurate estimate is σend−inf ' c 2 log(2β2) . (3.12) The total number of e-folds, N ' 2 √ α ∫ σend−inf 0 dσ√ 2ε ' α√ 2b σend−inf , (3.13) must be at ' 60 to solve the standard problems. This imposes σend−inf > 60 √ 2b/α. Meanwhile, for consistency σend−inf must be less than 1/b, which is the value when the approximation (3.2) breaks down. Thus we have the allowed range 60 √ 2 b α < σend−inf < 1 b . (3.14) Figure 2 shows the full supergravity potential that derives from the Kähler poten- tial (2.28) and approximate superpotential (3.1), for the parameter values α = 1 2 ; b = 10−3 ; c = 1 ; β1 = 1 ; β2 = 10 ; λ = 102 , (3.15) with (3.11) fixing B ' 10−4M3 Pl. For these parameter values, the inflationary scale gives Hinf ' 9× 10−7MPl ' 2× 1012 GeV . (3.16) – 12 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 The inflationary plateau is followed by a shallow minimum and a flat region — the “desert”. In the next section, we will discuss how after inflation the scalar field manages to traverse this desert and reach the chameleon region by the present time. These parameter values satisfy (3.14), but imply a tiny deviation from scale invariance, ns − 1 ' 10−5, in tension with the measured value from the 9-year WMAP data [44]: ns = 0.972 ± 0.013. We could not easily find alternative parameter values that would satisfy all constraints while yielding a desirable spectral tilt. The fiducial model therefore at best represents a proof of concept that an inflationary potential can be connected to a chameleon region within supergravity, but is not phenomenologically viable. 3.1 Gravitational reheating As sketched in figure 2, inflation comes to an end when the field falls off the inflationary plateau. The potential energy is converted into kinetic energy, and the field rapidly reaches the “desert”, where the potential is approximately zero over a distance�MPl in field space. Since the scalar field couples weakly to matter, reheating is inefficient and proceeds through gravitational particle production [34, 35]. The reheating temperature is of order Hinf , and the energy density in radiation is12 ρrad ∼ H4 inf . (3.17) Thus most of the inflationary energy goes into scalar field kinetic energy ρkin ' 3H2 infM 2 Pl. (3.18) By the end of reheating, the modification to the Kähler potential becomes negligible, and we are back to (2.17). As usual with gravitational reheating, we must ensure that the energy density in 4d massless gravitons is consistent with BBN bounds [33]. At the end of inflation, the ratio of graviton and inflaton energy to the energy in matter fields is fend−inf ≡ ρgraviton + ρinflaton ρmatter ∣∣∣∣ a'aend−inf ∼ 3 Ns , (3.19) where Ns is the number of scalar modes with masses below Hi. This translates at BBN to the ratio fBBN ≡ ρgraviton + ρinflaton ρmatter ∣∣∣∣ a'aend−inf = 3 Ns ( NBBN Nth )1/3 , (3.20) where NBBN = 10.75 is the effective number of spin degrees of freedom in equilibrium at temperature T , while Nth is the corresponding number at thermalization. Nucleosynthesis constraints impose fBBN . 0.07, which translates to Nth ∼ 102 − 103. This is easily 12Note that, as in explained in [33], there is a factor αR3/4 in front of the expression for the reheating tem- perature, with α a gauge coupling constant assumed in [33] to be between 0.1 and 0.01 and R = 10−2Ns, with Ns the number of scalars. In MSSM, Ns = 104, and in string theory it can be larger. Also it is not clear what values the coupling α can take, it could in principle be close to 1. Therefore we will assume that αR3/4 is of order 1 and drop it, though this prefactor is model dependent, and can take larger or smaller values as well. – 13 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 satisfied in our case: the MSSM has Ns = 104 scalars, while string theory has at least as many scalars. Thus this bound is satisfied in any case, though note that the bound is only valid in case there is no second reheating. In the second model for evolution towards BBN that we will present later, there is a second reheating, where the final amount of radiation present at BBN comes from the decay of a particle X. Then the original amount of gravitational radiation produced has been diluted away by the second reheating, and a negligible amount of gravitational radiation is produced now, since X will couple more strongly to matter than gravitational strength. A more constraining effect comes from the production of bulk gravitons from the brane, an effect first considered in [39, 40]. Once the extra dimensions have stabilized, there can in principle be evaporative cooling of the brane through bulk gravitons, which can compete with the standard cooling from the FRW expansion. The condition in order for this not to happen is a bound on the “normalcy temperature”, i.e., the temperature at which we match onto the normal phase of cosmological evolution at stabilization [39]: T∗ . 10 6n−9 n+1 ( M10 TeV )n+2 n+1 MeV , (3.21) where n denotes as before the number of large extra dimensions. Setting n = 2 for the case of interest, and substituting from (2.13) M10 ' 2.5a1/2 TeV, where as before we have assumed the limiting value (2.23), we obtain T∗ . 34a2/3 MeV . (3.22) 3.2 Extra-dimensional volume during inflation We conclude this section with a few brief remarks concerning the size of the extra- dimensional volume at the end of inflation. In units of the 10d Planck scale M10, the extra-dimensional volume reaches a maximal value of V6 ∼ σend−inf during inflation, which is always small. This problem also afflicts [42], and our modified superpotential offers no improvement. Nevertheless, we would argue that this is simply an issue of initial conditions. There is certainly nothing wrong with the volume being of order unity in M10 units — it is usually assumed in standard implementations of inflation that quantum gravity effects stabilize the extra dimensions at the fundamental scale. The stabilized value of the volume should be greater than or of order unity, but the volume at some arbitrary point in the V (φ) graph is not constrained. Effective field theory is still valid, since the energy scale of inflation Vin is much less than unity in MPl units. Note that the dilaton is assumed to be fixed in the KKLT-type construction, so there are no gs corrections, only α′ corrections to the action. But α′ corrections have already been incorporated in the action for the scalar % in our semi-phenomenological approach — see, e.g., (2.28) — and these corrections are indeed important in the inflationary region. One might also wonder about α′ corrections to the gravitational action. Such corrections to 4d gravity are small, α′R4 � 1, since the energy scale of inflation is sub-Planckian. Meanwhile, corrections ∼ α′R6 to 6d gravity on the compact space are implicitly taken into account in the action for %. – 14 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 Nevertheless, let us pause and describe more explicitly what form such corrections might take. Corrections involving the 6d Ricci scalar, ∼ α′R2 6, must be small, for otherwise Einstein’s equations would relate a large background R6 to an unacceptably large back- ground R4.13 For instance, a torus (with R6 = 0) is acceptable, but a sphere is excluded. For more general compact spaces, one also expects corrections involving the 6d Riemann tensor, ∼ α′RµνρσR µνρσ. However, in our framework we assume that only two compact dimensions vary and can become large, while the remaining 4 are stabilized at the fun- damental scale — the standard geometries are S2 (excluded) and T 2 (with Rµνρσ = 0). Since the compact space is really 6 dimensional, small variations in the other 4 dimensions should induce small α′ corrections for the action for %, governing the volume of the two large dimensions, but these corrections have already been incorporated, as mentioned.14 4 Traversing the “desert” The length of the desert separating the inflationary plateau from the chameleon region is insensitive to the details of the potential. It depends solely on the assumption of two large extra dimensions with fundamental scale at current epoch of M10 and the fact that the Kähler potential for the volume modulus is given by the perturbative formula (2.17). Given the relation (2.19) between σ and the canonical scalar field φ, the distance to be traversed is ∆φdesert = MPl ln σ? σend−inf 'MPl ln ( 1030 a ) , (4.1) where in the last step we have substituted the present value σ? ' 1030/a — see (2.23) — and estimated σend−inf ∼ 1. (For the parameter values (3.15), for instance, we have σend−inf ' 0.2.) As we will see below, such a large region is impossible to traverse in the standard scenario of a chameleon coupled to radiation, unless one is willing to consider unnaturally large values of a. The initial kinetic energy (3.18) rapidly redshifts away, and once the radiation component dominates, the scalar field is rapidly brought to a halt by Hubble friction. In order for the scalar field to successfully reach the chameleon region, we will assume that the universe undergoes a temporary epoch of matter domination well before BBN, due to an unstable massive particle. The coupling of the scalar to this massive particle will give the field the extra push required to reach its destination. Before proceeding, a few comments are in order. The required trans-Planckian sep- aration is at first sight problematic in effective field theory, since perturbative higher- dimensional operators like ∼ (φ/MPl) nψ̄ψ can become relevant for ∆φ > MPl. However, following KKLT we treat the supersymmetry breaking term due to the addition of an anti-brane in the chameleon region as a small perturbation (at σ ∼ σ∗), hence to a good approximation our setting is supersymmetric. In this context, the “desert” acts as a super- symmetric valley for the minimum of the potential. General arguments suggest that such 13This is why, for instance, the AdS4 × S7 background of 11d supergravity cannot be modified to be phenomenologically acceptable. 14For % very close to zero, there will of course be other corrections from the appearance of light modes, such as string winding modes and brane wrapping modes on the shrinking cycles. But these can be neglected during the inflationary period. – 15 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 corrections will not affect the supersymmetric locus of zero potential, i.e., a desert remains a desert. (See, for instance, [45] in the KKLT context, where it is noted that a supersym- metric extremum valley of zero potential remains stable under quantum corrections.) More generally, as mentioned earlier this issue is a generic problem in scenarios with large extra dimensions. All we have used is the existence of large extra dimensions and the tree-level string theory Kähler potential for the volume modulus. It seems reasonable to assume that the potential for the volume modulus in large extra dimensions remains small and monotonic until the stabilization point. So while the issue of potential corrections due to ∆φ/MPl > 1 is an important and subtle one (see e.g., [46]), we will not address it further here, as it is outside the scope of this paper. 4.1 Kinetic-dominated phase Let us derive the distance traversed by the scalar during the kinetic-dominated phase that immediately follows inflation. Since φ̇ ∼ 1/a3, with its initial value fixed by (3.18), we have φ̇ = √ 6HinfMPl (aend−inf a )3 . (4.2) Meanwhile, during kinetic domination, H = Hinf(aend−inf/a)3. Combining these expres- sions, we obtain dφ d ln a = √ 6MPl , (4.3) which implies a total distance traversed in the kinetic-dominated phase of ∆φkin = √ 6MPl ln aend−kin aend−inf , (4.4) where aend−kin denotes the scale factor at the end of the kinetic dominated phase (i.e., at kinetic-radiation equality). The latter can be determined straightforwardly from the initial energy density in the kinetic and radiation components, given respectively by (3.17) and (3.18): aend−kin aend−inf ∼ MPl Hinf . (4.5) We therefore obtain ∆φkin ' √ 6MPl ln MPl Hinf . (4.6) Note from (4.5) that the radiation density at the end of the kinetic-dominated phase is ρend−kin ∼ H8 inf/M 4 Pl, with corresponding temperature of Tend−kin ∼ H2 inf MPl . (4.7) For the fiducial value (3.16) Hinf ' 2× 1012 GeV, this gives Tend−kin ' 1000 TeV. Demanding that the field traverses the entire length of the desert during kinetic domination — we will see shortly that the field moves by a relatively small amount during – 16 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 the subsequent radiation-dominated phase — ∆φkin & ∆φdesert, we can combine (4.1) and (4.6) to derive the following lower bound on a: a & 1030 ( Hinf MPl )√6 . (4.8) Since (4.7) in this case represents the temperature with which we reach stabilization, and hence the temperature at the onset of the standard radiation-dominated phase, BBN requires Tend−kin &MeV, i.e., Hinf & 5× 107 GeV . (4.9) On the other hand, (4.7) is bounded from above by the normalcy bound (3.22): Tend−kin . 34a2/3 MeV. This yields another lower bound on a, namely a & 0.6 · 1030(Hinf/MPl) 3, but it is easy to check that this bound is weaker than (4.8) for all values of Hinf satisfying (4.9). In other words, once (4.8) and (4.9) are satisfied, the normalcy bound (3.22) automatically follows. For instance, for the fiducial value Hinf ' 2× 1012 GeV, the bound (4.8) gives a & 1015 (with Hinf ' 2× 1012 GeV ; Tend−kin = 1000 TeV) , (4.10) which is highly unnatural. For the lowest value allowed by BBN, Hinf ' 5× 107 GeV, we instead obtain a & 6× 103 (with Hinf ' 5× 107 GeV ; Tend−kin = MeV) , (4.11) which is acceptable. To reiterate, (4.8)−(4.11) apply in the “minimal” model where the expansion history consists of inflation, followed by kinetic domination, followed by standard radiation-dominated cosmology. In section 4.3, we will consider an alternative scenario where, after kinetic domination, the universe is dominated by a massive relic X, which later decays into radiation. This alternative possibility will considerably lower the required value of a in the fiducial case from a & 1015 to the more acceptable a & 5× 104. 4.2 Radiation-dominated phase Before moving on, however, we show that the field moves by a negligible amount during the subsequent radiation-dominated phase, being quickly brought to a halt by Hubble friction. The kinetic energy continues to redshift according to φ̇ ∼ 1/a3, hence φ̇ = √ 3Hend−kinMPl (aend−kin a )3 , (4.12) where the factor of √ 2 difference compared to (4.2) follows from the fact that the kinetic energy accounts for half of the total energy density at equality. Meanwhile, during radiation domination the Hubble parameter evolves as H = Hend−kin(aend−kin/a)2. Instead of (4.3), we obtain dφ d ln a = √ 3MPl aend−kin a , (4.13) – 17 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 0 5 10 15 20 0.0 0.2 0.4 0.6 0.8 1.0 N W Figure 3. Fractional densities Ω in units of the critical density for the chameleon-coupled matter ΩX (solid line), chameleon field energy Ωφ (dotted line), and radiation Ωrad (dashed line), as a function of N for c = 1/2. with solution φ = − √ 3MPlaend−kin/a + const. In the limit a � aend−kin, the distance traversed is ∆φrad ' √ 3MPl , (4.14) which does not help much in traversing the desert. 4.3 Chameleon-coupled phase To assist the scalar field in traversing the desert, we consider the possibility that there exists an unstable massive particle, X, which comes to dominate the energy density of the universe after kinetic domination. The scalar field couples to the energy density of X, and hence is driven to larger field values during X-domination. While not necessary — we have seen in section 4.1 that it is possible for the field to traverse the entire desert during kinetic domination — this alternative scenario will broaden the phenomenologically allowed region of parameter space. In full generality, we study this problem numerically using the formalism of [47], assuming for concreteness that at the end of the kinetic-dominated era we have comparable amounts of kinetic, radiation and X energy densities. Neglecting the potential energy, and using the perturbative relation (2.19) between σ and the canonical scalar φ, the scalar equation of motion is φ̈+ 3Hφ̇ = −ρX d lnF (φ) dφ , (4.15) where ρX ∼ F (φ)/a3. The coupling function is given by (2.25), F (φ) = e−φ/2MPl , but it will be instructive to consider the more general form F (φ) = e−cφ/MPl , (4.16) – 18 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 5 10 15 20 N 5 10 15 ΦHNL Figure 4. Numerical solution for φ(N) (in Planck units) for an initial value of φi = MPl, with Ω (i) X = 0.1e−φi/2MPl , Ω (i) rad = 0.4 and Ω (i) φ = 0.5. where c is a constant. (The case of interest is c = 1/2.) The Friedmann equation is 3M2 PlH 2 = ρX + φ̇2 2 + ρrad . (4.17) It is convenient to change the time variable to the number of e-foldings, dN = d ln a, in terms of which (4.15) becomes H2φ′′ + (3H2 +HH ′)φ′ = −ρX d lnF (φ) dφ , (4.18) where ′ ≡ d/dN . The friction term ∼ (3H2 + HH ′)φ′ can be simplified by substituting the Friedmann equation (4.17) and the acceleration equation, ä a = H ′H +H2 = − 1 6M2 Pl ( ρX + 2φ̇2 + 2ρrad ) , (4.19) to obtain H2φ′′ + 1 3M2 Pl ( 3 2 ρX + ρrad ) φ′ = −ρX d lnF (φ) dφ , (4.20) where the Hubble parameter is determined by the Friedmann equation: 3M2 PlH 2 = ρX + ρrad 1− M2 Plφ ′2 6 . (4.21) Using the fact that ρrad ∼ e−4N and ρX ∼ F (φ)e−3N , and substituting (4.21) into (4.20), we obtain an equation for the scalar which only depends on φ, its derivatives and N . This can be integrated straightforwardly by specifying initial densities. Figure 3 shows the resulting fractional energy densities, Ωφ = φ′2/6M2 Pl, ΩX ≡ ρX/3H 2M2 Pl and Ωrad ≡ ρrad/3H 2M2 Pl, for the fiducial case c = 1/2. – 19 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 From the figure, we see that the densities tend to a fixed-point behavior. (This was already anticipated in [48] for the case of a constant potential and no baryons.) For c < 1/ √ 2, which includes the case of interest, the fixed point corresponds to Ωφ = 2 3 c2 ; ΩX = 1− 2 3 c2 ; Ωrad = 0 for c < 1√ 2 . (4.22) For c > 1/ √ 2, on the other hand, we found Ωφ = 1 6c2 ; ΩX = 1 3c2 ; Ωrad = 1− 1 2c2 for c > 1√ 2 . (4.23) From (4.20) and (4.21), the attractor solution for the field is φ = φi + 3c 3 2 + Ωrad ΩX N =  φi + 2cMPlN for c < 1/ √ 2 ; φi + MPlN c for c > 1/ √ 2 , (4.24) where φi is the initial field value. Note that for the case of interest, c < 1/ √ 2, the attractor solution correspond to F (φ) = e−cφ/MPl ∼ e−2c2N , and ρφ , ρX ∼ e−(3+2c2)N for c < 1√ 2 . (4.25) (Since c < 1/ √ 2, the radiation component ρrad ∼ e−4N redshifts faster than ρφ , ρX and is therefore driven to zero, consistent with (4.22).) For c > 1/ √ 2, we instead find ρφ , ρX , ρrad ∼ e−4N , that is, all components redshift like radiation, and the universe is effectively radiation-dominated. In our case, with c = 1/2, the attractor solution (4.24) gives ∆φ = MPlN . (4.26) figure 4 shows the solution for φ(N) obtained numerically, which confirms the linear be- havior (4.26) as the attractor solution. Combining (4.6) and (4.26), and neglecting the contribution (4.14) from the radiation-dominated era, the total distance traversed by the scalar field is ∆φtot = ∆φkin + ∆φX ' √ 6MPl ln MPl Hinf +MPl ln aX−decay aend−kin , (4.27) where aX−decay is the scale factor when the X particle decays. From (4.1), we need ∆φtot & MPl ln(1030/a) to successfully reach the chameleon region, which implies aend−kin aX−decay . 10−30a ( MPl Hinf )√6 . (4.28) We next compute the reheating temperature once X decays and its energy density is converted into radiation. For simplicity, let us assume that ρX is comparable to the kinetic – 20 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 energy and radiation at equality. That is, using (4.7), ρX(aeq) ∼ H8 inf/M 4 Pl. Setting c = 1/2 in (4.25), we obtain ρX(aX−decay) = H8 inf M4 Pl ( aend−kin aX−decay )7/2 . (4.29) Assuming this energy all gets converted into radiation, the reheating temperature is Tfinal reheat = H2 inf MPl ( aend−kin aX−decay )7/8 . 10 a7/8 ( Hinf MPl )2− 7 √ 6 8 eV , (4.30) where in the last step we have used (4.28). Numerically, 2− 7 √ 6/8 ' −0.14, so the upper bound is a slowly varying function of Hinf . For our fiducial value of Hinf ' 2× 1012 GeV, the inequality is Tfinal reheat . 70 a7/8 eV. On the other hand, since BBN constrains Tfinal reheat & MeV, we obtain the following lower bound on a: a & 5× 104 (with Hinf ' 2× 1012 GeV) . (4.31) Comparing with (4.10) derived in the minimal scenario, we see that including an inter- mediate X phase greatly widens the allowed range of parameters by lowering the value of the temperature at the onset of radiation domination. The normalcy bound imposes the constraint Tfinal reheat . 34a2/3 MeV, corresponding to H−1 final reheat & 2.5× 10−3 a−4/3 s. This translates to a lower bound on the lifetime of X: τX & 2.5× 10−3 a−4/3 s. Meanwhile, BBN requires Tfinal reheat &MeV, which translates to τX . 1 s. The lifetime of X must therefore lie within the window 2.5× 10−3 a−8/3 s . τX . 1 s . (4.32) For instance, in the fiducial case with the minimal value a = 5×104 allowed by (4.31), this gives 3× 10−9 s . τX . 1 s, which is a sizable window. 4.4 Time evolution of M10 In Einstein frame, the 4d Planck scale MPl is by definition constant, but the 10d funda- mental scale M10 evolves in the time: M10 ∼ σ−1/2 ∼ e−φ/2MPl . (4.33) For consistency, the typical energy scale of the dominant component must remain well below M10 at all times. We check this epoch by epoch: • During inflation, this is trivially satisfied: M10 decreases from a value�MPl at early times to 'MPl at the end of inflation (since σend−inf ∼ O(1)), while the inflationary scale is of course well below MPl. • During the kinetic domination, (4.4) implies that M10 ∼ a− √ 3/2, which therefore redshifts faster than the temperature T ∼ a−1. Fortunately, kinetic domination ends before these scales have time to cross. Indeed, from (4.6) and (4.7), we infer M10(aend−kin) ' ( Hinf MPl )√ 3 2 MPl � Tend−kin . (4.34) – 21 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 • During X-domination, (4.26) implies M10 ∼ a−1/2, which redshifts more slowly than the radiation temperate T ∼ a−1. It also redshifts more slowly than the typical energy scale EX ∼ ρ1/4 X of X particles, which can be inferred from (4.25) with c = 1/2: EX ∼ a−7/8. Hence the X-dominated phase is completely safe. • Finally, during radiation domination, M10 tends to a constant within a Hubble time as φ is brought to a halt, while the radiation temperature of course continues to redshift. 5 Conclusions In this paper we have extended the model of [29], which derived a chameleon scenario from a modified KKLT set-up, to include a complete cosmological evolution. In particular, our goal was to have the same scalar field, the volume modulus of the extra dimensions, to both drive the inflationary epoch at early times and play the role of a chameleon dark energy field at present time. Aside from economy, this scenario opens up the exciting possibility of detecting the inflaton through present-day tests of gravity. Borrowing the constraints from laboratory tests of gravity derived in [29] , we were led to a large extra dimensions scenario, with the most natural case having 2 large extra dimensions. To successfully obtain inflation, we modified in a semi-phenomenological way both the Kähler potential in the inflationary (stringy) region and the KKLT superpotential, Beib%, with b > 0 by letting B → B(%). As a proof of principle, we focused on a particular — and admittedly somewhat ad hoc — form for B(%), but we argued that the post- inflationary evolution is not sensitive to the details of the deformation. In the chameleon region of the potential, meanwhile, we have the term Aeia%, with a < 0, used in [29]. We showed that the inflationary and chameleon regions are separated by a large desert region, �MPl in field-space distance, where the potential energy is negligible. We considered two scenarios in order for the scalar field to successfully traverse the desert after inflation. The first is the “minimal” scenario which does not add any new ingredient. After inflation, the universe enters a phase of kinetic domination, during which the scalar field proceeds through the desert. Kinetic domination is generically long because reheating is inefficient and proceeds through gravitational particle production. We derived constraints on the parameters, specifically Hinf and a, in order for the field to reach the chameleon region by the onset of radiation domination. In this minimal scenario, therefore, the entire desert is traversed during kinetic domination; the subsequent radiation-dominated phase is standard. In the second scenario, the universe after kinetic domination becomes dominated by a massive, unstable relic X which couples to the scalar field. This drives the scalar toward the chameleon region and helps it to traverse the desert. As a result, we found that the allowed parameter space is broadened in this non-minimal scenario. Most of the X matter must of course decay by BBN, which constrains its lifetime. Otherwise, all experimental constraints on the cosmology can be satisfied with only the usual fine-tuning of large extra dimensions and of the cosmological constant, as already noted in [29]. One issue we have not discussed in detail is how the field settles to the minimum of the potential in the chameleon region. Let us focus on the case with X matter, for – 22 – J H E P 0 8 ( 2 0 1 3 ) 0 5 3 concreteness. The attractor solution is due to the chameleon coupling, which creates a slope in the effective potential. In its absence, the field would come to a stop due to Hubble friction before reaching the minimum. And since the attractor solution relies also on the fact that the matter energy density ρX redshifts in time, thus lowering the slope of the effective potential, the field loses velocity gradually, with the kinetic energy ρφ being always of the order of the energy density of the chameleon-coupled matter ρχ. But when reaching the minimum of the potential, the energy density of the chameleon-coupled matter, and thus the kinetic energy as well, will be of the order of the potential energy — the field will gently settle at the minimum, being stopped by Hubble friction. To improve on the scenario, it would be helpful to have a better-motivated inflationary potential. The explicit example studied here serves as a proof of principle, as mentioned earlier, but is poorly motivated. We leave this to future work. Acknowledgments We thank Amanda Weltman for reading our manuscript and for discussions, and Albion Lawrence for discussions. The work of J.K. is supported in part by NASA ATP grant NNX11AI95G, the Alfred P. Sloan Foundation and NSF CAREER Award PHY-1145525 (J.K.). Research at Perimeter Institute is supported by the Government of Canada through Industry Canada and by the Province of Ontario through the Ministry of Economic Devel- opment and Innovation. This work was made possible in part through the support of a grant from the John Templeton Foundation. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the John Templeton Foundation (K.H.). The work of H.N. is supported in part by CNPQ grant 301219/2010-9. R.R. is par- tially supported by a CNPq research grant and by a Fapesp Tematico grant 2011/11973-4. References [1] E. Fischbach and C.L. Talmadge, The search for non-Newtonian gravity, Springer, New York U.S.A. (1999). [2] C.M. Will, The confrontation between general relativity and experiment, Living Rev. Rel. 9 (2006) 3 [gr-qc/0510072] [INSPIRE]. [3] J. Khoury, Theories of dark energy with screening mechanisms, arXiv:1011.5909 [INSPIRE]. [4] J. Khoury and A. Weltman, Chameleon fields: awaiting surprises for tests of gravity in space, Phys. Rev. Lett. 93 (2004) 171104 [astro-ph/0309300] [INSPIRE]. [5] J. Khoury and A. Weltman, Chameleon cosmology, Phys. Rev. D 69 (2004) 044026 [astro-ph/0309411] [INSPIRE]. [6] S.S. Gubser and J. Khoury, Scalar self-interactions loosen constraints from fifth force searches, Phys. Rev. D 70 (2004) 104001 [hep-ph/0405231] [INSPIRE]. [7] A. Upadhye, S.S. Gubser and J. Khoury, Unveiling chameleons in tests of gravitational inverse-square law, Phys. Rev. D 74 (2006) 104024 [hep-ph/0608186] [INSPIRE]. – 23 – http://arxiv.org/abs/gr-qc/0510072 http://inspirehep.net/search?p=find+EPRINT+gr-qc/0510072 http://arxiv.org/abs/1011.5909 http://inspirehep.net/search?p=find+EPRINT+arXiv:1011.5909 http://dx.doi.org/10.1103/PhysRevLett.93.171104 http://arxiv.org/abs/astro-ph/0309300 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0309300 http://dx.doi.org/10.1103/PhysRevD.69.044026 http://arxiv.org/abs/astro-ph/0309411 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0309411 http://dx.doi.org/10.1103/PhysRevD.70.104001 http://arxiv.org/abs/hep-ph/0405231 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0405231 http://dx.doi.org/10.1103/PhysRevD.74.104024 http://arxiv.org/abs/hep-ph/0608186 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0608186 J H E P 0 8 ( 2 0 1 3 ) 0 5 3 [8] P. Brax, C. van de Bruck, A.-C. Davis, J. Khoury and A. Weltman, Detecting dark energy in orbit — the cosmological chameleon, Phys. Rev. D 70 (2004) 123518 [astro-ph/0408415] [INSPIRE]. [9] P. Brax, C. van de Bruck, A. Davis, J. Khoury and A. Weltman, Chameleon dark energy, AIP Conf. Proc. 736 (2005) 105 [astro-ph/0410103] [INSPIRE]. [10] D.F. Mota and D.J. Shaw, Strongly coupled chameleon fields: new horizons in scalar field theory, Phys. Rev. Lett. 97 (2006) 151102 [hep-ph/0606204] [INSPIRE]. [11] D.F. Mota and D.J. Shaw, Evading equivalence principle violations, cosmological and other experimental constraints in scalar field theories with a strong coupling to matter, Phys. Rev. D 75 (2007) 063501 [hep-ph/0608078] [INSPIRE]. [12] P. Brax, C. van de Bruck, A.-C. Davis, D.F. Mota and D.J. Shaw, Testing chameleon theories with light propagating through a magnetic field, Phys. Rev. D 76 (2007) 085010 [arXiv:0707.2801] [INSPIRE]. [13] P. Brax, C. van de Bruck, A.-C. Davis, D.F. Mota and D.J. Shaw, Detecting chameleons through Casimir force measurements, Phys. Rev. D 76 (2007) 124034 [arXiv:0709.2075] [INSPIRE]. [14] P. Brax, C. van de Bruck, A.-C. Davis and D. Shaw, Laboratory tests of chameleon models, arXiv:0911.1086 [INSPIRE]. [15] C. Burrage, Supernova brightening from chameleon-photon mixing, Phys. Rev. D 77 (2008) 043009 [arXiv:0711.2966] [INSPIRE]. [16] C. Burrage, A.-C. Davis and D.J. Shaw, Detecting chameleons: the astronomical polarization produced by chameleon-like scalar fields, Phys. Rev. D 79 (2009) 044028 [arXiv:0809.1763] [INSPIRE]. [17] J.H. Steffen and A. Upadhye, The GammeV suite of experimental searches for axion-like particles, Mod. Phys. Lett. A 24 (2009) 2053 [arXiv:0908.1529] [INSPIRE]. [18] A. Upadhye, J. Steffen and A. Weltman, Constraining chameleon field theories using the GammeV afterglow experiments, Phys. Rev. D 81 (2010) 015013 [arXiv:0911.3906] [INSPIRE]. [19] A. Vainshtein, To the problem of nonvanishing gravitation mass, Phys. Lett. B 39 (1972) 393 [INSPIRE]. [20] N. Arkani-Hamed, H. Georgi and M.D. Schwartz, Effective field theory for massive gravitons and gravity in theory space, Annals Phys. 305 (2003) 96 [hep-th/0210184] [INSPIRE]. [21] C. Deffayet, G. Dvali, G. Gabadadze and A.I. Vainshtein, Nonperturbative continuity in graviton mass versus perturbative discontinuity, Phys. Rev. D 65 (2002) 044026 [hep-th/0106001] [INSPIRE]. [22] K. Hinterbichler and J. Khoury, Symmetron fields: screening long-range forces through local symmetry restoration, Phys. Rev. Lett. 104 (2010) 231301 [arXiv:1001.4525] [INSPIRE]. [23] K. Hinterbichler, J. Khoury, A. Levy and A. Matas, Symmetron cosmology, Phys. Rev. D 84 (2011) 103521 [arXiv:1107.2112] [INSPIRE]. [24] P. Brax, C. van de Bruck, A.-C. Davis and D. Shaw, The dilaton and modified gravity, Phys. Rev. D 82 (2010) 063519 [arXiv:1005.3735] [INSPIRE]. – 24 – http://dx.doi.org/10.1103/PhysRevD.70.123518 http://arxiv.org/abs/astro-ph/0408415 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0408415 http://dx.doi.org/10.1063/1.1835177 http://arxiv.org/abs/astro-ph/0410103 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0410103 http://dx.doi.org/10.1103/PhysRevLett.97.151102 http://arxiv.org/abs/hep-ph/0606204 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0606204 http://dx.doi.org/10.1103/PhysRevD.75.063501 http://dx.doi.org/10.1103/PhysRevD.75.063501 http://arxiv.org/abs/hep-ph/0608078 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0608078 http://dx.doi.org/10.1103/PhysRevD.76.085010 http://arxiv.org/abs/0707.2801 http://inspirehep.net/search?p=find+EPRINT+arXiv:0707.2801 http://dx.doi.org/10.1103/PhysRevD.76.124034 http://arxiv.org/abs/0709.2075 http://inspirehep.net/search?p=find+EPRINT+arXiv:0709.2075 http://arxiv.org/abs/0911.1086 http://inspirehep.net/search?p=find+EPRINT+arXiv:0911.1086 http://dx.doi.org/10.1103/PhysRevD.77.043009 http://dx.doi.org/10.1103/PhysRevD.77.043009 http://arxiv.org/abs/0711.2966 http://inspirehep.net/search?p=find+EPRINT+arXiv:0711.2966 http://dx.doi.org/10.1103/PhysRevD.79.044028 http://arxiv.org/abs/0809.1763 http://inspirehep.net/search?p=find+EPRINT+arXiv:0809.1763 http://dx.doi.org/10.1142/S0217732309031727 http://arxiv.org/abs/0908.1529 http://inspirehep.net/search?p=find+EPRINT+arXiv:0908.1529 http://dx.doi.org/10.1103/PhysRevD.81.015013 http://arxiv.org/abs/0911.3906 http://inspirehep.net/search?p=find+EPRINT+arXiv:0911.3906 http://dx.doi.org/10.1016/0370-2693(72)90147-5 http://inspirehep.net/search?p=find+J+Phys.Lett.,B39,393 http://dx.doi.org/10.1016/S0003-4916(03)00068-X http://arxiv.org/abs/hep-th/0210184 http://inspirehep.net/search?p=find+EPRINT+hep-th/0210184 http://dx.doi.org/10.1103/PhysRevD.65.044026 http://arxiv.org/abs/hep-th/0106001 http://inspirehep.net/search?p=find+EPRINT+hep-th/0106001 http://dx.doi.org/10.1103/PhysRevLett.104.231301 http://arxiv.org/abs/1001.4525 http://inspirehep.net/search?p=find+EPRINT+arXiv:1001.4525 http://dx.doi.org/10.1103/PhysRevD.84.103521 http://dx.doi.org/10.1103/PhysRevD.84.103521 http://arxiv.org/abs/1107.2112 http://inspirehep.net/search?p=find+EPRINT+arXiv:1107.2112 http://dx.doi.org/10.1103/PhysRevD.82.063519 http://dx.doi.org/10.1103/PhysRevD.82.063519 http://arxiv.org/abs/1005.3735 http://inspirehep.net/search?p=find+EPRINT+arXiv:1005.3735 J H E P 0 8 ( 2 0 1 3 ) 0 5 3 [25] P. Brax, C. van de Bruck, A.-C. Davis, B. Li and D.J. Shaw, Nonlinear structure formation with the environmentally dependent dilaton, Phys. Rev. D 83 (2011) 104026 [arXiv:1102.3692] [INSPIRE]. [26] K.A. Olive and M. Pospelov, Environmental dependence of masses and coupling constants, Phys. Rev. D 77 (2008) 043524 [arXiv:0709.3825] [INSPIRE]. [27] M. Pietroni, Dark energy condensation, Phys. Rev. D 72 (2005) 043535 [astro-ph/0505615] [INSPIRE]. [28] A. Adams, N. Arkani-Hamed, S. Dubovsky, A. Nicolis and R. Rattazzi, Causality, analyticity and an IR obstruction to UV completion, JHEP 10 (2006) 014 [hep-th/0602178] [INSPIRE]. [29] K. Hinterbichler, J. Khoury and H. Nastase, Towards a UV completion for chameleon scalar theories, JHEP 03 (2011) 061 [Erratum ibid. 06 (2011) 072] [arXiv:1012.4462] [INSPIRE]. [30] P. Brax and A.-C. Davis, Supersymmetron, Phys. Lett. B 707 (2012) 1 [arXiv:1109.0468] [INSPIRE]. [31] P. Brax, A.-C. Davis and J. Sakstein, SUPER-screening, Phys. Lett. B 719 (2013) 210 [arXiv:1212.4392] [INSPIRE]. [32] S. Kachru, R. Kallosh, A.D. Linde and S.P. Trivedi, De Sitter vacua in string theory, Phys. Rev. D 68 (2003) 046005 [hep-th/0301240] [INSPIRE]. [33] P. Peebles and A. Vilenkin, Quintessential inflation, Phys. Rev. D 59 (1999) 063505 [astro-ph/9810509] [INSPIRE]. [34] L. Ford, Cosmological constant damping by unstable scalar fields, Phys. Rev. D 35 (1987) 2339 [INSPIRE]. [35] L. Grishchuk and Y. Sidorov, Squeezed quantum states of relic gravitons and primordial density fluctuations, Phys. Rev. D 42 (1990) 3413 [INSPIRE]. [36] H. Abe, T. Higaki and T. Kobayashi, KKLT type models with moduli-mixing superpotential, Phys. Rev. D 73 (2006) 046005 [hep-th/0511160] [INSPIRE]. [37] F. Quevedo, Lectures on superstring phenomenology, hep-th/9603074 [INSPIRE]. [38] E. Witten, Nonperturbative superpotentials in string theory, Nucl. Phys. B 474 (1996) 343 [hep-th/9604030] [INSPIRE]. [39] N. Arkani-Hamed, S. Dimopoulos and G. Dvali, Phenomenology, astrophysics and cosmology of theories with submillimeter dimensions and TeV scale quantum gravity, Phys. Rev. D 59 (1999) 086004 [hep-ph/9807344] [INSPIRE]. [40] N. Arkani-Hamed, S. Dimopoulos, N. Kaloper and J. March-Russell, Rapid asymmetric inflation and early cosmology in theories with submillimeter dimensions, Nucl. Phys. B 567 (2000) 189 [hep-ph/9903224] [INSPIRE]. [41] S. Kachru et al., Towards inflation in string theory, JCAP 10 (2003) 013 [hep-th/0308055] [INSPIRE]. [42] M. Yamaguchi, Supergravity based inflation models: a review, Class. Quant. Grav. 28 (2011) 103001 [arXiv:1101.2488] [INSPIRE]. [43] A.R. Liddle and D. Lyth, Cosmological inflation and large scale structure, Cambridge University Press, Cambridge U.K. (2000) [INSPIRE]. – 25 – http://dx.doi.org/10.1103/PhysRevD.83.104026 http://arxiv.org/abs/1102.3692 http://inspirehep.net/search?p=find+EPRINT+arXiv:1102.3692 http://dx.doi.org/10.1103/PhysRevD.77.043524 http://arxiv.org/abs/0709.3825 http://inspirehep.net/search?p=find+EPRINT+arXiv:0709.3825 http://dx.doi.org/10.1103/PhysRevD.72.043535 http://arxiv.org/abs/astro-ph/0505615 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0505615 http://dx.doi.org/10.1088/1126-6708/2006/10/014 http://arxiv.org/abs/hep-th/0602178 http://inspirehep.net/search?p=find+EPRINT+hep-th/0602178 http://dx.doi.org/10.1007/JHEP03(2011)061 http://arxiv.org/abs/1012.4462 http://inspirehep.net/search?p=find+EPRINT+arXiv:1012.4462 http://dx.doi.org/10.1016/j.physletb.2011.11.060 http://arxiv.org/abs/1109.0468 http://inspirehep.net/search?p=find+EPRINT+arXiv:1109.0468 http://dx.doi.org/10.1016/j.physletb.2013.01.044 http://arxiv.org/abs/1212.4392 http://inspirehep.net/search?p=find+EPRINT+arXiv:1212.4392 http://dx.doi.org/10.1103/PhysRevD.68.046005 http://dx.doi.org/10.1103/PhysRevD.68.046005 http://arxiv.org/abs/hep-th/0301240 http://inspirehep.net/search?p=find+EPRINT+hep-th/0301240 http://dx.doi.org/10.1103/PhysRevD.59.063505 http://arxiv.org/abs/astro-ph/9810509 http://inspirehep.net/search?p=find+EPRINT+astro-ph/9810509 http://dx.doi.org/10.1103/PhysRevD.35.2339 http://dx.doi.org/10.1103/PhysRevD.35.2339 http://inspirehep.net/search?p=find+J+Phys.Rev.,D35,2339 http://dx.doi.org/10.1103/PhysRevD.42.3413 http://inspirehep.net/search?p=find+J+Phys.Rev.,D42,3413 http://dx.doi.org/10.1103/PhysRevD.73.046005 http://arxiv.org/abs/hep-th/0511160 http://inspirehep.net/search?p=find+EPRINT+hep-th/0511160 http://arxiv.org/abs/hep-th/9603074 http://inspirehep.net/search?p=find+EPRINT+hep-th/9603074 http://dx.doi.org/10.1016/0550-3213(96)00283-0 http://arxiv.org/abs/hep-th/9604030 http://inspirehep.net/search?p=find+EPRINT+hep-th/9604030 http://dx.doi.org/10.1103/PhysRevD.59.086004 http://dx.doi.org/10.1103/PhysRevD.59.086004 http://arxiv.org/abs/hep-ph/9807344 http://inspirehep.net/search?p=find+EPRINT+hep-ph/9807344 http://dx.doi.org/10.1016/S0550-3213(99)00667-7 http://dx.doi.org/10.1016/S0550-3213(99)00667-7 http://arxiv.org/abs/hep-ph/9903224 http://inspirehep.net/search?p=find+EPRINT+hep-ph/9903224 http://dx.doi.org/10.1088/1475-7516/2003/10/013 http://arxiv.org/abs/hep-th/0308055 http://inspirehep.net/search?p=find+EPRINT+hep-th/0308055 http://dx.doi.org/10.1088/0264-9381/28/10/103001 http://dx.doi.org/10.1088/0264-9381/28/10/103001 http://arxiv.org/abs/1101.2488 http://inspirehep.net/search?p=find+EPRINT+arXiv:1101.2488 http://inspirehep.net/record/533080 J H E P 0 8 ( 2 0 1 3 ) 0 5 3 [44] WMAP collaboration, G. Hinshaw et al., Nine-year Wilkinson Microwave Anisotropy Probe (WMAP) observations: cosmological parameter results, arXiv:1212.5226 [INSPIRE]. [45] J.J. Blanco-Pillado, R. Kallosh and A.D. Linde, Supersymmetry and stability of flux vacua, JHEP 05 (2006) 053 [hep-th/0511042] [INSPIRE]. [46] N. Kaloper, A. Lawrence and L. Sorbo, An ignoble approach to large field inflation, JCAP 03 (2011) 023 [arXiv:1101.0026] [INSPIRE]. [47] J. Lopes Franca, Urbano and R. Rosenfeld, Fine tuning in quintessence models with exponential potentials, JHEP 10 (2002) 015 [astro-ph/0206194] [INSPIRE]. [48] D. Tocchini-Valentini and L. Amendola, Stationary dark energy with a baryon dominated era: solving the coincidence problem with a linear coupling, Phys. Rev. D 65 (2002) 063508 [astro-ph/0108143] [INSPIRE]. – 26 – http://arxiv.org/abs/1212.5226 http://inspirehep.net/search?p=find+EPRINT+arXiv:1212.5226 http://dx.doi.org/10.1088/1126-6708/2006/05/053 http://arxiv.org/abs/hep-th/0511042 http://inspirehep.net/search?p=find+EPRINT+hep-th/0511042 http://dx.doi.org/10.1088/1475-7516/2011/03/023 http://dx.doi.org/10.1088/1475-7516/2011/03/023 http://arxiv.org/abs/1101.0026 http://inspirehep.net/search?p=find+EPRINT+arXiv:1101.0026 http://dx.doi.org/10.1088/1126-6708/2002/10/015 http://arxiv.org/abs/astro-ph/0206194 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0206194 http://dx.doi.org/10.1103/PhysRevD.65.063508 http://arxiv.org/abs/astro-ph/0108143 http://inspirehep.net/search?p=find+EPRINT+astro-ph/0108143 Introduction The model Chameleon region Inflationary region Inflationary dynamics and constraints Gravitational reheating Extra-dimensional volume during inflation Traversing the ``desert'' Kinetic-dominated phase Radiation-dominated phase Chameleon-coupled phase Time evolution of M(10) Conclusions