GhostFree Massive f (R) Theories Modelled as Eective Einstein Spaces & Cosmic Acceleration

Sergiu I. Vacaru

Theory Division, CERN, CH-1211, Geneva 23, Switzerland ; and Rector's Oce, Alexandru Ioan Cuza University, Alexandru Lapuneanu street, nr. 14, UAIC - Corpus R, oce 323; Iai, Romania, 700057

November 7, 2013

Abstract
We study how massive ghostfree gravity f (R)modied theories, MGFTs, can be encoded into generic o diagonal Einstein spaces. Using "auxiliary" connections completely dened by the metric elds and adapted to nonholonomic frames with associated to nonlinear connection structure, we decouple and integrate in certain general forms the eld equations in MGFT. Imposing additional nonholonomic constraints, we can generate LeviCivita, LC, congurations and mimic MGFT eects via odiagonal interactions of eective Einstein and/or EinsteinCartan gravity with nonholonomically induced torsion. The cosmological evolution of ghostfree odiagonal Einstein spaces is investigated. Certain compatibility of MGFT cosmology to small odiagonal deformations of CDM models is established.

Introduction:
were elaborated.

In [1, 2], two models of nonlinear massive gravitational theories including

f (R )

modications

Such theories contain the benets of the dRGT model [3] and are free of ghost modes [4].

Another advantages are those that by tuning the tion of ination and latetime acceleration etc.

f (R) functional (on such modications, see reviews [5]), we can

From general theoretical considerations, the

stabilize cosmological backgrounds, elaborate various types cosmological evolution scenarios, unied descrip-

f (R )

paradigm

has motivation from various attempts to explain the universe acceleration and dark energy/ matter problems through infrared (IR) modications of the general relativity (GR) theory and understanding possible physical implications of the massive spin2 theory. On the other hand, ultra-violet (UV) corrections expected to be of quantum origin (see Ref. [6] on possible eective actions). Cosmological implications of massive gravity were also analyzed in the framework of modied gravity theories, MGT, [7], see also cosmological models related to bimetric gravity [8]. It is the point of this paper to apply in MGFT the so-called anholonomic frame deformation method, AFDM, [9] for constructing generic odiagonal exact solutions. Such a method provides a geometric techniques which allows us to integrate systems of partial dierential equations, PDEs, with functional and parametric dependencies for the LeviCivita (zero torsion) and nontrivial torsion congurations.

a Whitney sum

The geometric setup: We shall work on a pseudo-Riemannian manifold V, dim V = n + m, (n, m 2), when N is dened for its tangent space T V, N : T V = hT V vT V. Such a decomposition denes
The local coecients

a nonholonomic (equivalently, non-integrable, or anholonomic) horizontal (h) and vertical (v) splitting, i.e. a nonlinear connection (N-connection ) structure, see details in [9].

Nia (x, y )dxi

and

v -indices a, b, ... = n + 1, n + 2, ..., n + m,1

/y a for certain local coordinates

u = (x, y ),

or u

(xi , y a ), with

{Nia (u)}, when N = h-indices i, j, ... = 1, 2, ..., n

dene naturally N-adapted frame and, respectively, dual frame

sergiu.vacaru@cern.ch; sergiu.vacaru@uaic.ro

All Rights Reserved c Nov 7, 2013 Sergiu I. Vacaru, http://www.scribd.com/people/view/1455460-sergiu associated visiting researcher

We shall use the Einstein rule on summation on up-low cross indices. Boldface letters are written in order emphasize that a

N-connection spitting is considered on a manifold

V = (V, N).
1

structures,

e = (ei , ea )

and

e = (ei , ea ),

when and

ei = /xi Nia (u)/y a , ea = a = /y a ,

In general, such local (co) bases are nonholonomic, i.e.

ei = dxi , ea = dy a + Nia (u)dxi .

(1)

[e , e ] = e e e e = W e with anholonomy b b a a a a a coecients Wia = a Ni , Wji = ij = ej (Ni ) ei (Nj ), when ij is the N-connection curvature. With respect to (1), any metric tensor g can be expressed as a distinguished metric, dmetric,

g = g (u)e e = gi (xk )dxi dxi + ga (xk , y b )ea ea .

tions,

(2)

For any prescribed Nconnection and dmetric structures, we can work equivalently with two linear connec-

(g, N)
where

: g = 0; T = 0; D : Dg = 0; hT = 0, v T = 0, hv T = 0,

is the torsionless LeviCivita, LC, connection and D = hD + v D is the socalled canonical distinguished connection, dconnection. The value D preserves the h-v splitting under parallel transports but does not have such a property. Nevertheless, there is a canonical distortion distinguished tensor, d-tensor, Z = {T },
which is an algebraic combination of the coecients of the corresponding torsion d-tensor denes a canonical distortion relation

T = {T }.

This and and For

= 0,

and curvatures,

D = + Z which is adapted to the Nsplitting. The torsions, T R = {R } and R = {R }, respectively, of D and can be dened
and

computed in standard coordinate free and/or coecient forms. The Ricci tensors of

are dened

Ric = {R

:= R } h-v

and

Ric = {R

:= R }.

instance, the Ricci d-tensor

Ric

is characterized by four subsets of

N-adapted coecients, (3)

R = {Rij := Rkijk , Ria := Rkika , Rai := Rbaib , Rab := Rcabc }.

Alternatively to the LC-scalar curvature, curvature,

R :=

g R

g ij Rij

g ab Rab .2

R := g R ,

we can introduce the scalar of canonical d

Field equations in MGFT and Nadapted variables: We follow the model elaborated in [2] and reformulate V enabled with Nconnection structure N and two dmetrics when g = {g } is the dynamical dmetric and q = {q } is the socalled nondynamical reference metric. In our approach, we work with D instead of and R is computed for g, the nonzero graviton mass is denoted by , MP is the
it on a nonholonomic manifold Planck mass.

Let us consider the dtensor

g1 q)

computed as the square root of

g q , when (

g1 q) (

g1 q) = ek (X)
can

g q ,

4
and

ek (

g1 q) = 3 tr

g1 q det

g1 q,

for some coecients

k . The values

be dened for any dtensor X and trace

k=0

X = [X ] := tr(X) =

X , when

e0 (X ) = 1, e1 (X ) = X, 2e2 (X ) = X 2 [X 2 ], 6e3 (X ) = X 3 3X [X 2 ] + 2[X 3 ], 24e4 (X ) = X 4 6X 2 [X 2 ] + 3[X 2 ]2 + 8X [X 3 ] 6[X 4 ]; ek (X ) = 0

We shall use also the massdeformed scalar curvature The action for

k > 4. g 1 q) .

:= R + 2 2 (3 tr R

g1 q det

for MGFT is postulated in the form

2 S = MP
2

) + d4 u |g|[f (R

L],

(4)

canonical distortion relations for

Any (pseudo) Riemannian geometry can be equivalently described by both geometric data (g, ) and (g, N,D), when the R = R+ Z and Ric = Ric + Z ic, with respective distortion d-tensors Z and Z ic, are computed To prove the decoupling of fundamental gravitational equations in general relativity, GR, and various MGFTs LC-congurations can be

D = + Z.

is possible for d-metrics and the canonical dconnection working with respect to N-adapted frames. in zero values for the canonical d-torsion,

extracted from certain classes of solutions of (modied) gravitational eld equations if additional conditions are imposed, resulting

T = 0.

Our system of "Nadapted notations" is similar to that considered in [10, 11].

where

m L(g, N) is the Lagrange density for the matter elds.4

The energymomentum dtensor can be

computed via Nadapted variational calculus,

m

T :=
(4), with

( |g | 2 g |g |
1 f (R )

L)

Lg + 2

( m L) . g

(5)

Applying such a calculus to

)/dR , := df (R R = ,

see details in [9], we obtain the eld equations (6)

where

+
m

+ =

, for m f

1 2 2MP

T ,

= 22 [(3 tr +

D D 1 f D2 1 f ) g + , 1f 1f 2 1f 1 g1 q det g1 q) det g1 q)]g 2 = ( f

(7)

2 {q [( g1 q)1 ] + q [( g1 q)1 ] }. 2

We note that the Bianchi identities for the data mations

D T

(g, N,D) are given by introducing nonholonomic defor1 = D Z into the standard relations (R 2 g R) = 0 and T = 0. Even, in general, = Q = 0, such a Q [g, N] is completely dened by the dmetric and chosen Nconnection structure.
A new class of eective

This is a consequence of the nonholonomic structure. A similar "problem" exists in Lagrange mechanics with nonintegrable constraints when the standard conservation laws do not hold true. and conservation laws with nonholonomic constraints written in variables variables can be introduced using Lagrange multiples. We omit cumbersome formulas for the Bianchi densities

(g, N,D).

Encoding and decoupling properties of eld equations in MGFT:

The generalized gravitational eld equations

written with respect to Nadapted frames (6) are similar to those studied in our works [9, 10, 11]. The main dierence of such MTGs is determined by a corresponding source which in this work is considered in the form (7). Applying the AFDM, we can construct very general classes of generic odiagonal solutions which encode both

f modications

and massive gravity eects with nonzero

For simplicity, we shall consider nonholonomic dynamical systems in MGTs which via frame transforms and connection deformations can be transformed into certain eective odiagonal Einstein manifolds described by dmetrics with one Killing symmetry on

4 5 variable y . This is described by ansatz (2) when

i

/y 3 ,

i.e. the gravitational and matter elds do not depend on

gi = e(x ) , ga = ha (xi , t), Ni3 = ni (xk ), Ni4 = wi (xk , t).

The eective source is chosen for a timelike coordinate

(8)

y 4 = t,

when

= diag [1 = 2 , 2 =

(xi ) +

xi ) + (

(xi ),
(9)

3 = 4 , 4 = m (xi , t) + f (xi , t) + + f + )g . = ( m e e ,

(xi , t)]

The assumption for the rst parametrization in (9) is that the matter elds and eective sources, are generated in Nadapted frames by two types functions/ distributions

The left labels refer to contributions in such sources by general, we get 4 independent Nadapted coecients

f modications and/or by mass i of = diag { (x , t)} for variations

L

= i (x ) and (xi , t). modications. In

in (5) using

(8). For cosmological applications, we can model sources of matter elds by an energymomentum tensor for

4 5

For simplicity, we consider matter actions

S=

d4 u

|g |

which only depend on the coecients of a metric eld and not

on their derivatives. It is possible to construct metrics with nonKilling symmetries depending on all spacetime coordinates. This requests a more advanced and cumbersome geometric techniques, see examples in [9, 11] and references therein.

ideal uids as in GR but with generic odiagonal metrics frames,

6 (encoding contributions from MGT). In Nadapted

(10)

T = pg + ( + p)v v
is dened for certain (eective) energy, the uid for which

= 1

and v

A tedious calculus of Nadapted

, and pressure, densities, p, respectively, v being the four-velocity of = (0, 0, 0, 1) in N-adapted comoving frames/coordinates. coecients of the Ricci dtensor for D computed for ansatz (8) and source
m m f f

(9) transform (6) into such a system of nonlinear PDEs:

+ = 2 ( h3 = 2h3 h4 (

+ +

+ +

, ) = 2 ) = 2h3 h4 ,

(11) (12) (13)

ni + ni = 0, wi i = 0,
for

i = h3 i , = h3 , = ln |h3 |3/2 /|h4 |

where and/ or

= ln |h3 / |h3 h4 ||, 1 =

and/or

:= e ,

(14)

is considered as a generating function. In above formulas, we use such notations for partial derivatives:

/x1 ,

= 2 , h3 =4 h3 .

For simplicity, we do not study in this paper dmetrics for which

= ha = 0

= 0 (such solutions in vacuum MGFT can be constructed, for instance, for f - and/or -modications

of black hole solutions, see examples in [9]). The torsionless (LeviCivita, LC) conditions are satised if there are additionally imposed the conditions

wi = (i wi 4 ) ln

|h4 |, (i wi 4 ) ln

|h3 | = 0, k wi = i wk , ni = 0, i nk = k ni .

(15)

Exact odiagonal solutions in MGFT:

The system (11)-(13) posses an important property when 1)

solution of two dimensional (2-d) Poisson equation with source on

2(...)(xk ); h3

and

h4

are related to

is the and sources

via equation (14) and the Nconnection coecients are determined correspondingly by integrating two times

the equations for

ni

and from a system of rst order algebraic equations for

wi .

For MGTs, the procedure

of nding locally anisotropic and inhomogeneous cosmological solutions is described in [11].

We can x always a sum of nontrivial constants using formulas

m + f + and redene the generating function,

2 = 2 || +
when

dt 2 ||

, 2 =

| | ||2

2 ||, dt

(16)

2 ) / . In order to solve the second equation in (15), (i wi 4 ) ln |h3 | = 0, the generating (2 ) /|| = ( function must be chosen to satisfy the conditions (i ) = i . We can parameterize the solutions for the )2 |2 2 | = = ( system (12) and (14) in the form h3 [] and h4 [] = 2 2 || . 4|| || dt
We nd in explicit form solutions of algebraic equations in (13) and the conditions second line in (15) if

k wi = i wk

from the (17)

wi = i / = i A,
with a nontrivial function two times on

A(xk , t) depending on generating function via a rst order Pfa system.

Integrating

in (13), we express

nk =
where and

1 nk

+ 2 nk

dy 4 h4 /( |h3 |)3 ,

(18)

1 nk (x

1 nk =

i ) and i 2 nk (x ) are integration functions. To generate LC-congurations we have to x 2 nk i k n(x ).

=0

such metrics can not be diagonalized by coordinate transforms because for general Nconnections the anholonomy coecients, W , are not zero

Putting together above formulas, we conclude that generic odiagonal quadratic elements

ds2 = e(x

k ,[ m + f + ])

[(dx1 )2 + (dx2 )2 ] +

2 [dy 3 + k n dxk ]2 + f + | 4| m
|

] [ m + f + ]|2 |[ 3 + f + |2 2 | m + f + | m dt
symmetry on

dxi )2 . (dt + i A[]

(19)

determine generic odiagonal solutions of the eld equations in MGFT. For welldened assumptions on Killing

and imposed at the end zero torsion conditions such metrics belong to the integral variety of

the system (11)-(15). We can generate exact solutions in "pure"

f modied

gravity if put

= 0.

If

0, we can nonholonomically induce a nontrivial . Inverse nonlinear transforms are possible if we change
with

mutually the left labels

f.

It should be noted that above classes of metrics can be extended to describe exact solutions with nonholo-

, , , ] } of D. We have to substitute in (19) k n nk (xi , t) (18) and T = {T [ take instead of (17) the value wi = i / . It is possible to re-write all coecients in terms of the generating . The LC conditions (15) are not satised for such congurations.7 function , or in terms of
nomically induced torsion

On properties of odiagonal solutions in MGFT and GR:

The metrics (19) describe locally anisotropic

and inhomogeneous spacetimes determined by certain classes of generating functions

xi , t) (

and

+ f + ]) ; sources m (xi , t), f (xi , t), (xi , t) and m ( xi ), f ( xi ), ( xi ), and integration (xk , [ m k m f , , , which can be considered as integration functions like k n(x ); and eective cosmological constants
constants. These values and one of the sines and for respective sources

should be xed in such a form that they are compatible with

observational data. We can generate inhomogeneous cosmological metrics taking certain limits

xi , t) ( t) (

(xi , t) (t).

Such solutions generalize the class of known anisotropic solutions

of Bianchi cosmology to congurations when the coecients of metrics are not subjected to typical symmetric conditions for those spacetimes and, in our approach, may encode geometric and physical data for MGFT interactions. Fixing, for instance, certain results of

= 0, h3

i.e. for the zero mass of graviton, the metrics (19) reproduce

f ( R)

gravity and cosmology theories, see [11] and references therein. So, at least for

= 0,

by introducing a conformal factor

before

and

h4

in above formulas, redening the generating functions,

and for small odiagonal coecients, we reproduce nonholonomic deformations of

CDM

universes.

The metrics (19) do not have, in general, a simple physical interpretation. Choosing the integration constants, we can extract (for instance) Kasner type solutions with dynamical chaos etc, see examples in [9] and references therein. It is necessary a rigorous study of nonperturbative and nonlinear eects of such generic odiagonal dynamical systems even for small

(this is a matter for our further research). Here we note that

the nonholonomic nonlinear coupling with redenition of generating functions by formulas (16), and by o diagonal coecients of (19), encodes of geometric and physical data for MGFT into eective Einstein spaces. This follows from the fact that such solutions are equivalent (up to frame/coordinate transforms) to the equations

g = R .

This motivates equivalent redenitions of sources

)g

as we supposed in (10). Considering solitonic congurations, we can polarize or "open" for a period of time some modes of massive gravity and then "switch o" such interactions and "pump" certain induced functions.

f modied

eects into odiagonal coecients of Einstein metrics with redened cosmological constants and generating

Scale factors and odiagonal deformations of FLRW metrics: Let us introduce a new i |h4 |t/ t, and a scale factor a(xi , t) when the dmetric (19) can be when t = t(x , t) and
form

time coordinate

t,

represented in the (20)

ds2 = a2 (xi , t)[i (xk , t)(dxi )2 + h3 (xk , t)(e3 )2 (e4 )2 ],

where

i =

a2 e , a2 h3

= h3 i

, e3

dy 3

can be modelled with a small parameter

+ k n = dt + |h4 |(i t + wi ). , with 0 < 1, when n i ( x k ) , |h4 |(i t + wi )

dxk , e4

Small odiagonal deformations

1 + i (xk , t), k n

wi (xk , t).

(21)

Such torsion elds are dierent from those in EinsteinCartan, gauge and/or string gravity where additional eld equations

and sources are considered to dene the torsion dynamics.

We can choose a subclass of generating functions and sources when

a(xi , t) a(t), h3 (xi , t) h3 (t)

etc.

Such conditions, or of type (21), have to be imposed after a locally anisotropic solution was constructed in explicit form. This results in new classes of solutions even in diagonal limits because of generic nonlinear and nonholonomic character of odiagonal systems in MGFT. For

and

a(xi , t) a(t),

we obtain scaling

factors which are very dierent from those in FriedmannLematreRoberstsonWorker, FLRW, cosmology with GR solutions. Nevertheless, they mimic such cosmological models with redened interaction parameters and possible small odiagonal deformations of cosmological evolution for modied gravity theories as we analyzed in details in [11]. In this work, we consider eective sources encoding contributions from massive gravity, when

a2 h3 =

2 2 | , where 2 4|

m + f + |+

dt 2 |

m + f + |

m + f +

The generating functions, sources and parameters in these formulas determine integral varieties (i.e. general solutions) of certain systems of nonlinear PDE. Such values have to be xed in such forms which results in certain physical values compatible with experimental data. Following the procedure from section 5 of [11], we can derive a corresponding eective eld theory, see also references therein.

Reconstructing odiagonal cosmological models in MGFT: Let us consider a model when the gravitational ) = R + M( T), where T := T+2 2 (3 tr g1 q det g1 q). We Lagrange density (4) is chosen f (R 1 denote M := dM/d T and H := a /a for a limit a(xi , t) a(t) taken for a solution (20) and consider a that an observer is in a nonholonomic basis (1) with Ni = {ni , wi (t)} for a nontrivial o-diagonal vacuum with eective polarizations (t). It should be emphasized that a(t) is dierent from a(t) for a standard FLRW
cosmology. a new shift derivative when (for instance, for a function [11]. It is described by a set of three equations The cosmological scenarios are tested in terms of the redshift

1 + z = a1 (t) for and T = T (z ), with s(t)) s = (1 + z )Hz . We can derive MGFT

odiagonal deformed FLRW equations following the procedure considered for the formulas (63) and (64) in

1 3H 2 + [f (z ) + M(z )] 2 (z ) = 0, 2 1 2 3H + (1 + z )H (z H ) {f (z ) + M(z ) + 3(1 + z )H 2 = 0, 2 (z ) z f = 0.

Re-scaling the generating function, we have xed the condition

(22)

z 1 M(z ) = 0

and can satisfy the condition The functional

z f = 0 0 (1 +

which allows nonzero densities in certain adapted frames of references. when

encodes degrees of freedom of mass gravity for the evolution of the energy-density when

= 0

a3(1+ )

M( T) =

z )a3(1+ ) . This is taken for the dust matter approximation

Using (22), it possible to elaborate reconstruction procedures for nontrivial

(1 + in

z )3 .
a form similar to that in

[11, 12, 13]. For instance, it is well known that any FLRW cosmology can be realized in a specic Let us introduce the e-folding variable

f (R)

gravity.

Here we analyze how specic MGFTs and the FLRW cosmology can be encoded into odiagonal deformations.

:= ln a/a0 = ln(1 + z ) which will be considered instead of the i m (xi , ) + f (xi , ) + (xi , ) instead of (10) cosmological time t. We take as in (4), use (x , ) = i and parameterize the geometric objects with dependencies on (x , ) (in particular, only on ), for corresponding generating functions (16), when = / with s = H s for any function s. The matter energy density is ) f (R
that from (22). With respect to N-adapted frames(1), we can repeat all computations leading to Eqs. (2)-(7) in [12] and use that method to prove that a MGFTs with

) may realize a FLRW like cosmological model. f (R

The nonholonomic

eld equation corresponding to the rst FLRW equation is

2 2 ) = (H 2 + H H ) [f (R )] 36H 2 4H + ( H )2 + H )]+2 . f (R H f (R
Introducing an eective quadratic Hubble rate, this equation transforms into

( ) := H 2 ( ),

where

) = (R

for certain parameterizations,

2 f = 18 ( )[ ( ) + 4 ( )]

d2 f 1 df 3(1+ +6 ( ) + ( ) + 20 a0 2 2 dR dR
6

) 3(1+ ) (R)

(23)

O-diagonal cosmological models are determined by metrics of type (20), when used for computing

t ,

for a functional

) f (R

is

and

Such nonlinear systems can be described eectively by the eld equations for an The value

(nonholonomic) Einstein space dependence

f ()

with

. = R = 0. Even we

df /dR

and higher derivatives vanish for any functional

work with odiagonal congurations, the recovering procedure

simplies substantially in such cases.

An example of reconstruction of MGFT and nonholonomically deformed Einstein spaces reproducing the CDM era: We consider any a( ) and H ( ) determined by an o-diagonal solution (20), with respect to correspondingly N-adapted frames. The analog of FLRW equation for CDM cosmology is
3 3 2 2 32 H 2 = 32 H0 + 0 a3 = 32 H0 + 0 a , 0 e
where (24)

H0

and

are xed to be certain constant values. Such assumptions are considered after the coecients

of o-diagonal solutions are found and when the dependencies on sources.

(xi , ) are changed into dependencies on . The

(9) which appears in a

values with "hat" are generated via a corresponding re-denition of the generating functions and the eective The rst term on the rhs is related to an eective cosmological constant re-denition (16). For this model, the second term in (24) describes, in general, an inhomogeneous distribution of cold dark mater (CDM). The similarity with the diagonalizable cosmological models in GR is kept if we chose

= 12H 2 0
tively,

to survive in the limit

wi , ni 0,

for certain approximations of type (21).

The eective quadratic Hubble rate and the modied scalar curvature,

, R

are computed using (24), respec-

3 3 2 + 2 0 a ( ) := H0 0 e
The equation (23) transforms into

and

3 3 2 = 3 R ( ) + 12 ( ) = 12H0 + 2 0 a . 0 e

X (1 X )
for certain constants, for which and

d2 f df + [3 (1 + 2 + 1)X ] 1 2 f = 0, dX 2 dX

(25)

X := 3 +

1 3 2 1 + 2 = 1 2 = 1/6 and 3 = 1/2 where 3 = ln[2 0 a0 (R 12H0 )] 2 /3H . The solutions of such equations with constant coecients and for dierent types R 0
Similarly, we denote when for some constants

of scalar curvatures were found in [12] and [11] as Gauss hypergeometric functions.

f = F (X ) := F (1 , 2 , 3 ; X ),

and

B,

F (X ) = AF (1 , 2 , 3 ; X ) + BX 13 F (1 3 + 1, 2 3 + 1, 2 3 ; X ).
This provides a proof of the statement that MGFT can indeed describe an eective cosmological constant.

CDM

scenarios without the need of

Final remarks:

One of the most interesting results of applications of the AFDM [9] to nonlinear MGFTs

systems is that via redenition of generating functions and eective sources we can mimic

f modications and

massive gravity eects. This is possible by modelling modied theories via odiagonal interactions in eective Einstein spaces. Such models are generically nonlinear, parametric and with respect to nonholonomic frames which allows to decouple and integrate the associated PDEs in general forms. This way, we have to consider in dierent forms, for instance, the problems of stability, possible mutual transforms and equivalence of one class of solutions in a theory to similar solutions in another classes of theories. It is necessary to reconsider the status of locally anisotropic and/or inhomogeneous metrics in modern cosmology etc. Stability issues for MGFTs can be analyzed and for certain eective odiagonal Einstein metrics. We can work also with a "convenient"

f model

which allows us to dene certain stable subclasses of solutions as we discuss in details in section 4.3 of [11]. It is considered that one could be certain problems with massive gravity because of a proof the absence of FLRW cosmology (see section 2.1 in [14]). That proof follows if we consider only homogeneous and isotropic ansatz for metrics in a massive theory. In this paper, we studied more general constructions both for modied gravity functionals and for odiagonal locally anisotropic and inhomogeneous metrics. For nonlinear systems, it is very important when certain assumptions and additional constraints are considered. If certain "simplications" or approximations are made at the very beginning, we can draw certain conclusions about some properties of a theory and even follow a procedure of nding of solutions. But this way we can also eliminate a number of

other types of solutions and various nonlinear characteristics. In our approach, we elaborated a more general and more realistic model with generic odiagonal eects when certain stability congurations and odiagonal modications of FLRW cosmology are generated by eective sources in nonlinear massive gravity.

Acknowledgments:

The work is partially supported by the Program IDEI, PN-II-ID-PCE-2011-3-0256

and a visiting research program at CERN. SV is grateful to N. Mavromatos, S. Odintsov, M. Sami, E. Saridakis and P. Stavrinos for important discussions.

References
[1] Yi-Fu Cai, F. Duplessis and E. N. Saridakis, arXiv: 1307.7150 [2] J. Kluso, S. Nojiri and S. D. Odintsov, arXiv: 1309.2185; S. Nojiri and S. D. Odintov, Phys. Lett. (2012) 377; S. Nojiri, S. D. Odintsov and N, Schirai, JCAP [3] C. de Rham and G. Gabadadze, Phys. Rev. Tolley, Phys. Rev. Lett.

1305 (2013) 020

B 716

106 (2011) 231101

D 82

(2010) 044020; C. de Rham, G. Gabadadze and A. J.

[4] D. G. Boulware and S. Deser, Phys. Rev. F. Hassan and R. A. Rosen, JHEP

D 6 (1972) 3368; J. Kluson, Phys. Rev. D 86 (2012) 044024; S. 1204 (2012) 123; A. Golovnev, Phys. Lett. B 707 (2012) 4004 4 (2007) 115]; Phys.

[5] S. Nojiri and S. D. Odintsov, eConf C (2006) 0602061 [Int. J. Geom. Meth. Mod. Phys. Rept.

505

(2011) 59; S. Capozziello and V. Faraoni, Beyond Einstein gravity: A survey of gravitational

theories for cosmology and astrophysics (Springer, 2010) [6] K. S. Stelle, Phys. Rev. 527 [7] E. N. Saridakis, Class. Quant. Grav. (2013) 029; M. Mohseni, JCAP

16

(1977) 953; S. Deser, J. H. Kay and K. S. Stelle, Phys. Rev. Lett.

38

(1977)

B 725 (2013) 1; M. Andrews, G. Goon, K. Hinterbichler, J. Stokes and M. Trodden, Phys. Rev. Lett. 111 (2013) 061107; R. Gannouji, M. W. Hossain, M. Sami and E. N. Saridakis, Phys. Rev. D 87 (2013) 123536; S. Capozziello and P. Martin-Morun, Phys. Lett. B 719 (2013) 14; J. Kluson, Phys. Rev. D86
(2012) 044024

30 (2013) 075003; Y. -I. Zhang, R. Saito and M. Sasaki, JCAP 1302 1211 (2012) 023; K. Hinterbichler, J. Stokes and M. Trodden, Phys. Lett.

[8] T. Damour, I. I. Kogan and A. Papazoglou, Phys. Rev. D 66 (2002) 104025; M. S. Vokov, JHEP 1201 (2012) 035; M. von Strauss, A. Schmidt-May, J. Enangder, E. Mortsell and S. F. Hassan, JCAP 1203 (2012) 042; M. Berg, I. Buchberger, J. Enander, E. Mortsell and S. Sjors, JCAP 1212 (2012) 021 [9] S. Vacaru, Eur. Phys. J. Phys.: Conf. Ser.

543

C 73

(2013) 2287; Europhys. Letters

(2013) 012021; J. Math. Phys.

and D. Singleton, Class. Quant. Grav. [10] S. Vacaru, arXiv: 1304.2916

19 (2002) 2793

46

96

(2011) 5001; IJGMMP

(2005) 042503 JHEP

04

(2011) 9; J.

(2001) 009; S. Vacaru

[11] E. Elizalde and S. Vacaru, arXiv: 1310.6868 [12] S. Nojiri, S. D. Odintsov and D. Saez-Gomez, Phys. Lett. [13] P. Stavrinos and S. Vacaru, Class. Quant. Grav. 1304.1079 [14] G. D'Amico, C. de Rham, S. Dubovsky, G. Gabadadze, D. Pirtskhalava and A. J. Tolley, Phys. Rev. (2011) 124046

B681 (2009) 74

30

(2013) 055012; S. Vacaru, arXiv: 1305.1876; arXiv:

D 84