A novel realization of the Starobinsky inflationary
model within a moderate extension of the Minimal Supersymmetric
Standard Model (MSSM) is presented. The proposed superpotential is
uniquely determined by applying a continuous and a
discrete symmetry, whereas the Kähler potential is associated
with a no-scale-type Kähler manifold. The inflaton is identified with a
Higgs-like modulus whose the vacuum expectation value controls the
gravitational strength. Thanks to a strong enough coupling (with a
parameter involved) between the inflaton and the Ricci
scalar curvature, inflation can be attained even for subplanckian values
of the inflaton with and the corresponding effective
theory being valid up to the Planck scale. The inflationary
observables turn out to be in agreement with the current data and
the inflaton mass is predicted to be . At the
cost of a relatively small superpotential coupling constant, the model
offers also a resolution of the problem of MSSM.
Supplementing MSSM by three right-handed neutrinos we show that
spontaneously arising couplings between the inflaton and the
particle content of MSSM not only ensure a sufficiently low
reheating temperature but also support a scenario of non-thermal
leptogenesis consistently with the neutrino oscillation parameters
for gravitino heavier than about .
Keywords: Cosmology, Supersymmetric models, Supergravity, Modified Gravity;
PACS codes: 98.80.Cq, 11.30.Qc, 12.60.Jv, 04.65.+e, 04.50.Kd
Published in J. Cosmol. Astropart. Phys. 04, 024 (2014); J. Cosmol. Astropart. Phys. 07, 01E (2017)
]Linking Starobinsky-Type Inflation
in no-Scale Supergravity to MSSM
\@hangfromDepartament de Física Teòrica and IFIC,
Universitat de València-CSIC,
E-46100 Burjassot, SPAIN
Department of Physics, University of Cyprus,
P.O. Box 20537, Nicosia 1678, CYPRUS
After the announcement of the recent PLANCK results [1, 2], inflation based on the potential of the Starobinsky model  has gained a lot of momentum [4, 5, 6, 8, 9, 7, 10, 11] since it predicts [3, 12] a (scalar) spectral index very close to the one favored by the fitting of the observations by the standard power-law cosmological model with cold dark matter (CDM) and a cosmological constant (CDM). In particular, it has been shown that Starobinsky-type inflation can be realized within extensions of the Standard Model (SM)  or Minimal SUSY SM (MSSM) . However, the realization of this type of inflation within Supergravity (SUGRA) is not unique. Different super- and Kähler potentials are proposed [6, 7, 5] which result to the same scalar potential. Prominent, however, is the idea [4, 5] of implementing this type of inflation using a Kähler potential, , corresponding to a Kähler manifold inspired by the no-scale models [15, 16]. Such a symmetry fixes beautifully the form of up to an holomorphic function which exclusively depends on a modulus-like field and plays the role of a varying gravitational coupling. The stabilization of the non-inflaton accompanying field can to be conveniently arranged by higher order terms in . In this context, a variety of models are proposed in which inflaton can be identified with either a matter-like [4, 5, 14] or a modulus-like [6, 5] inflaton. The former option seems to offer a more suitable framework  for connecting the inflationary physics with a low-energy theory, such as the MSSM endowed with right handed neutrinos, , since the non-inflaton modulus is involved in the no-scale mechanism of soft SUSY breaking (SSB). On the other hand, the inflationary superpotential, , is arbitrarily chosen and not protected by any symmetry. Given that, the inflaton takes transplanckian values during inflation, higher order corrections – e.g., by non-renormalizable terms in – with not carefully tuned coefficients may easily invalidate or strongly affect [8, 17] the predictions of an otherwise successful inflationary scenario.
It would be interesting, therefore, to investigate if the shortcoming above can be avoided in the presence of a strong enough coupling of the inflaton to gravity [18, 19], as done [20, 21, 22, 24, 23, 25] in the models of non-minimal Inflation (nMI). This idea can be implemented keeping the no-scale structure of , since the involved can be an analytic function, selected conveniently. In view of the fact that depends only on a modulus-like field, we here focus on this kind of inflaton – contrary to Ref. . As a consequence, the direct connection of the inflationary model with the mechanism of the SSB is lost. Note, in passing, that despite their attractive features, the no-scale models  of SSB enface difficulties – e.g., viable SUSY spectra are obtained only when the boundary conditions for the SSB terms are imposed beyond the Grand Unified Theory (GUT) scale and so the low energy observables depend on the specific GUT.
Focusing on a modulus-like inflaton, the link to MSSM can be established through the adopted . Its form in our work is fixed by imposing a continuous symmetry, which reduces to the well-known -parity of MSSM, and a discrete symmetry. As a consequence, resembles the one used in the widely employed models [26, 27] of standard F-term Hybrid Inflation (FHI) – with singlet waterfall field though. As a bonus, a dynamical generation of the reduced Planck scale arises in Jordan Frame (JF) through the vacuum expectation value (v.e.v) of the inflaton. Therefore the inflaton acquires a higgs-character as in the theories of induced gravity [28, 29]. To produce an inflationary plateau with the selected , is to be taken quadratic, in accordance with the adopted symmetries. This is to be contrasted with the so-called modified Cecotti model [30, 6, 8, 7, 5] where the inflaton appears linearly in the super- and Kähler potentials. The inclusion of two extra parameters compared to the original model – cf. [6, 8, 5] – allows us to attain inflationary solutions for subplanckian values of the inflaton with the successful inflationary predictions of the model being remained intact. As a bonus, the ultaviolet (UV) cut-off scale [31, 32, 9] of the theory can be identified with the Planck scale and so, concerns regarding the naturalness of the model can be safely eluded.
Our inflationary model – let name it for short no-scale modular inflation (nSMI) – has ramifications to other fundamental open problems of the MSSM and post-inflationary cosmological evolution. As a consequence of the adopted symmetry, the generation [27, 33] of the mixing term between the two electroweak Higgses is explained via the v.e.v of the non-inflaton accompanying field, provided that a coupling constant in is rather suppressed. Finally, the observed  baryon asymmetry of the universe (BAU) can be explained via spontaneous [35, 36] non-thermal leptogenesis (nTL)  consistently with the constraint [38, 39, 40], the data [41, 42] on the neutrino oscillation parameters as long as the masses of the gravitino () lie in the multi-TeV region – as dictated in many versions [43, 44, 45] of MSSM after the recent LHC [46, 47] results on the Higgs boson mass.
The basic ingredients – particle content and structure of the super- and Kähler potentials – of our model are presented in Sec. [. In Sec. [ we describe the inflationary potential, derive the inflationary observables and confront them with observations. Sec. [ is devoted to the resolution of the problem of MSSM. In Sec. [ we outline the scenario of nTL, exhibit the relevant imposed constraints and describe our predictions for neutrino masses. Our conclusions are summarized in Sec. [. Throughout the text, the subscript of type denotes derivation with respect to (w.r.t) the field (e.g., ) and charge conjugation is denoted by a star.
We focus on a moderated extension of MSSM with three ’s augmented by two superfields, a matter-like and a modulus-like , which are singlets under . Besides the local symmetry of MSSM, , the model possesses also the baryon number symmetry , a nonanomalous symmetry and a discrete . Note that global continuous symmetries can effectively arise  in many compactified string theories. The charge assignments under the global symmetries of the various matter and Higgs superfields are listed in Table [. We below present the structure of the superpotential (Sec. [) and the Kähler potential (Sec. [) of our model.
The superpotential of our model naturally splits into two parts:
where is the part of which contains the usual terms – except for the term – of MSSM, supplemented by Yukawa interactions among the left-handed leptons and
Here the th generation doublet left-handed quark and lepton superfields are denoted by and respectively, whereas the singlet antiquark [antilepton] superfields by and [ and ] respectively. The electroweak Higgs superfields which couple to the up [down] quark superfields are denoted by .
On the other hand, is the part of which is relevant for nSMI, the generation of the term of MSSM and the Majorana masses for ’s. It takes the form
where is the reduced Planck mass. The imposed symmetry ensures the linearity of w.r.t . This fact allows us to isolate easily via its derivative the contribution of the inflaton into the F-term SUGRA scalar potential, placing at the origin. The imposed prohibits the existence of the term which, although does not drastically modifies our proposal, it complicates the determination of SUSY vacuum and the inflationary dynamics. On the other hand, the imposed symmetries do not forbid non-renormalizable terms of the form where is an integer. For this reason we are obliged to restrict ourselves to subplanckian values of .
The second term in the right-hand side (r.h.s) of Eq. (2.1c) provides the term of MSSM along the lines of Ref. [27, 33] – see Sec. [. The third term is the Majorana mass term for the ’s and we assume that it overshadows (for sufficiently low ’s) the last non-renormalizable term which is neglected henceforth. Here we work in the so-called right-handed neutrino basis, where is diagonal, real and positive. These masses together with the Dirac neutrino masses in Eq. (2.1b) lead to the light neutrino masses via the seesaw mechanism. The same term is important for the decay [35, 36] of the inflaton after the end of nSMI to , whose subsequent decay can activate nTL. As a result of the imposed , a term of the form is prohibited and so the decay of into is processed by suppressed SUGRA-induced interactions , guaranteing thereby a sufficiently low reheat temperature compatible with the constraint and successful nTL – see Sec. [.
Table 1: The global charges of the superfields of our model.
Since the no-scale SUGRA adopted here leads to the non-renormalizable F-term (scalar) potential in Eq. (3.1), we expect that it yields an effective SUSY theory which depends not only on the superpotential, in Eq. (2.1c), but also on the Kähler potential, in Eq. (2.6a) – see Sec. [. To trace out this behavior we apply the generic formula for the SUSY F-term potential :
which is obtained from the SUGRA potential in Eq. (3.1) – see Sec. [ below – if we perform an expansion in powers and take the limit . The Kähler potential, , employed here can not be expanded in powers of , since unity is not included in the argument of the logarithm – in contrast to the ’s used in Ref. [50, 51]. In Eq. (2.2) is the inverse of the Kähler metric with and where the complex scalar components of the superfields and are denoted by the same symbol whereas this of by . We find that reads
where is given in Eq. (2.7c) and we neglect the fourth order terms since we expect that these are not relevant for the low energy effective theory. The inverse of the matrix above is
Substituting this in Eq. (2.2), we end up with the following expression
The three first terms in the r.h.s of the expression above come from the terms of Eq. (2.2) for . The fourth one comes from the terms
for and ; the residual terms arise from terms of the form ; for and the fifth one and the last one.
From the potential in Eq. (2.5a), we find that the SUSY vacuum lies at
Contrary to the Cecotti model [30, 6, 5] our modulus can take values at the SUSY vacuum. Also, breaks spontaneously the imposed and so, it can comfortably decay via SUGRA-inspired decay channels – see Sec. [ – reheating the universe and rendering  spontaneous nTL possible. No domain walls are produced due to the spontaneous breaking of at the SUSY vacuum, since this is broken already during nSMI.
With the addition of SSB terms, as required in a realistic model, the position of the vacuum shifts [27, 33] to non-zero and an effective term is generated from the second term in the r.h.s of Eq. (2.1c) – see Sec. [. Let us emphasize that SSB effects explicitly break to the matter parity, under which all the matter (quark and lepton) superfields change sign. Combining with the fermion parity, under which all fermions change sign, yields the well-known -parity. Recall that this residual symmetry prevents the rapid proton decay, guarantees the stability of the lightest SUSY particle (LSP) and therefore it provides a well-motivated CDM candidate. Needless to say, finally, that such a connection of the Starobinsky-type inflation with this vital for MSSM -symmetry can not be established within the modified Cecotti model [30, 6, 7], since no symmetry can prohibit a quadratic term for the modulus-like field in conjunction with the tadpole term in .
According to the general discussion of Ref. , the Kähler manifold which corresponds to a Kähler potential of the form
with being an holomorphic function of , exhibits a global symmetry. Here is the number of scalar components of and the MSSM superfields which are collectively denoted as
Note that summation over the repeated (small or capital) Greek indices is implied. The third term in the r.h.s of Eq. (2.6a) – with coefficients being taken, for simplicity, real – is included since it has an impact on the scalar mass spectrum along the inflationary track – see Sec. [. In particular, the term with coefficient assists us to avoid the tachyonic instabilities encountered in similar models [6, 5, 7, 18, 19] – see Sec. [. The ellipsis represents higher order terms which are irrelevant for the inflationary dynamics since they do not mix the inflaton with the matter fields. This is, in practice, a great simplification compared to the models of nMI – cf. Ref. . Contrary to other realizations of the Starobinsky model – cf. Ref. [6, 5, 7] –, we choose to be quadratic and not linear with respect to , i.e.,
in accordance with the imposed symmetry which forbids a linear term – the coefficient is taken real too. As in the case of Eq. (2.1c), non-renormalizable terms of the form , with integer , are allowed but we can safely ignore them restricting ourselves to .
The interpretation of the adopted in Eq. (2.6a) can be given in the “physical” frame by writing the JF action for the scalar fields . To extract it, we start with the corresponding EF action within SUGRA [18, 21, 25] which can be written as
where with , is the determinant of the EF metric , is the EF Ricci scalar curvature, is defined in Sec. [, the dot denotes derivation w.r.t the JF cosmic time and the ellipsis represents terms irrelevant for our analysis. Performing then a suitable conformal transformation, along the lines of Ref. [21, 25] we end up with the following action in the JF
where is the JF metric with determinant , is the JF Ricci scalar curvature, and we use the shorthand notation and . The corresponding frame function can be found from the relation
The last result reveals that has no kinetic term, since . This is a crucial difference between the Starobinsky-type models and those  of nMI, with interest consequences  to the derivation of the ultraviolet cutoff scale of the theory – see Sec. [. Furthermore, given that , recovering the conventional Einstein gravity at the SUSY vacuum, Eq. (2.5b), dictates
In this section we outline the salient features of our inflationary scenario (Sec. [) and then, we present its predictions in Sec. [, calculating a number of observable quantities introduced in Sec. [. We also provide a detailed analysis of the UV behavior of the model in Sec. [.
where . Setting the fields and at the origin the only surviving term of is
It is obvious from the result above that a form of as the one proposed in Eq. (2.6c) can flatten sufficiently so that it can drive nSMI. Employing the dimensionless variables
and setting , and the corresponding Hubble parameter read
where we put – by virtue of Eq. (2.8) – in the final expressions.
Expanding and in real and imaginary parts as follows
we can check the stability of the inflationary direction
w.r.t the fluctuations of the various fields. In particular, we examine the validity of the extremum and minimum conditions, i.e.,
Here are the eigenvalues of the mass matrix with elements
– here we take into account that and are doublets –, the kinetic terms of the various scalars in Eq. (2.7a) can be brought into the following form
where the hatted fields are defined as follows
Table 2: The mass spectrum of our model along the inflationary trajectory of Eq. (3.6).
Upon diagonalization of the relevant sub-matrices of , Eq. (3.7b), we construct the scalar mass spectrum of the theory along the direction in Eq. (3.6). Our results are summarized in Table [, assuming in order to avoid very lengthy formulas for the masses of and . The various unspecified there eigenvalues are defined as follows:
where the spinors and associated with the superfields and are related to the normalized ones in Table [ as follows:
We also use the shorthand notation:
Note that, due to the large effective masses that the ’s in Eq. (3.7b) acquire during nSMI, they enter a phase of oscillations about with reducing amplitude. As a consequence – see Eq. (3.9b) –, since the quantity , involved in relating to , turns out to be negligibly small compared with – cf. Ref. . Moreover, we have numerically verified that the various masses remain greater than during the last e-foldings of nSMI, and so any inflationary perturbations of the fields other than the inflaton are safely eliminated – see also Sec. [.
as decreases from to . Here we have made use of Eqs. (3.16a) and (3.20b) – see Sec. [. We do not consider such a condition on as unnatural, given that in Eq. (2.1b) is of the same order of magnitude too – cf. Ref. . In Table [ we also present the masses squared of chiral fermions along the trajectory of Eq. (3.6), which can be served for the calculation of the one-loop radiative corrections. Employing the well-known Coleman-Weinberg formula , we find that the one-loop corrected inflationary potential is
where is a renormalization group (RG) mass scale. As we numerically verify the one-loop corrections have no impact on our results. The absence of gauge interactions and of a direct renormalizable coupling between and assists to that direction – cf. Ref. [25, 55]. Based on , we can proceed to the analysis of nSMI in the EF, employing the standard slow-roll approximation [57, 56]. It can be shown  that the results calculated this way are the same as if we had calculated them using the non-minimally coupled scalar field in the JF.
A successful inflationary scenario has to be compatible with a number of observational requirements which are outlined in the following.
The number of e-foldings, , that the scale suffers during nSMI has to be adequate to resolve the horizon and flatness problems of the standard Big Bag cosmology. Assuming that nSMI is followed in turn by a decaying-particle, radiation and matter domination and employing standard methods , we can easily derive the required for our model, with the result:
where is the value of when crosses the inflationary horizon. Also is the value of at the end of nSMI determined, in the slow-roll approximation, by the condition:
where the slow-roll parameters read
The termination of nSMI is triggered by the violation of the criterion at a value of equal to , which is calculated to be
since the violation of the criterion occurs at such that
On the other hand, can be calculated via the relation
Given that , we can find a relation between and as follows
Obviously, nSMI with subplanckian ’s can be achieved if
for . Therefore we need relatively large ’s.
Since the scalars listed in Table [ are massive enough during nSMI, the curvature perturbations generated by are solely responsible for . Substituting Eqs. (3.15b) and (3.18a) into the relation above, we obtain
Combining the last equality with Eq. (3.19), we find that is to be proportional to , for almost constant . Indeed, we obtain