Modern Physics Letters A

Vol. 32 (2017) 1750086 (16 pages)


c World Scientic Publishing Company
DOI: 10.1142/S0217732317500869

Some exact solutions in f (G, T ) gravity via Noether symmetries

M. Farasat Shamir
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

National University of Computer and Emerging Sciences, Lahore Campus, Pakistan

farasat.shamir@nu.edu.pk
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

Mushtaq Ahmad
National University of Computer and Emerging Sciences,
Chiniot-Faisalabad Campus, Pakistan
mushtaq.sial@nu.edu.pk

Received 19 February 2017

Accepted 21 March 2017
Published 26 April 2017

This paper is devoted to investigate the recently proposed modied GaussBonnet

f (G, T ) gravity, with G, the GaussBonnet term, coupled with T , the trace of energy
momentum tensor. We have used the Noether symmetry methodology to discuss some
cosmologically important f (G, T ) gravity models with anisotropic background. In par-
ticular, the Noether symmetry equations for modied f (G, T ) gravity are reported for
locally rotationally symmetric Bianchi type I universe. Explicitly, two models have been
proposed to explore the exact solutions and the conserved quantities. It is concluded
that the specic models of modied GaussBonnet gravity may be used to reconstruct
CDM cosmology without involving any cosmological constant.

Keywords: f (G, T ) gravity; Noether symmetries; CDM.

PACS Nos.: 04.20.Jb, 98.80.Jk

1. Introduction
Recent observations through huge telescopes and satellites have conrmed the evi-
dence for an evolving universe which is expanding. The interpretations from the
supernova experiments, the data from Wilkinson Microwave Anisotropy Probe
(WMAP) and Sloan Digital Sky Surveys (SDSS) lead us toward the remarkable
conclusion that this expansion is in accelerating mode.13 Cosmological models are
facing serious issues which can be summed up as the problems of the dark matter
and dark energy. It is believed that so-called factor of dark matter and dark energy
which is the 70% of the total energy-mass of the universe is causing this accelera-
tion. There is another viewpoint that the modications of General Relativity (GR)
are behind this accelerating expansion of this universe. Investigating all the possible

1750086-1
 M. F. Shamir & M. Ahmad

reasons, including the dark energy problem, for the accelerating universe expansion
would be one of the most focused areas of research followed by a series of exper-
iments and several surveys, for the many years to come. No doubt, GR has been
a great success in the last century but it could not properly address the problems
of dark matter, dark energy, initial singularity, late-time cosmic acceleration, and
the atness issues. As an alternative to GR, dierent modied theories of gravity
have been presented by the researchers to unveil these unsolved problems, which are
believed to be the real cause of this accelerating expansion of the universe. Inspired
by the original theory, a variety of modied theories of gravity like f (R), f (R, T ),
f (G), f (R, G) and f (G, T ) have been structured.416
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

In particular, modied theories of GaussBonnet (GB) gravity have attracted

much attention of the researchers in recent years.1820 Late-time cosmic acceleration
may be caused due to the existence of de Sitter point of f (G) gravity, G be the GB
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

invariant dened as
G = R2 4R R + R R , (1)
where R is the Riemann tensor, R is the Ricci tensor and R is the Ricci scalar.
The notable attribute of this theory is that the participation of GB term may pre-
vent uncertain contributions and uniforms the gravitational action.21 The theory
has been modied further by introducing scalar curvature along with GB invariant
which is named as f (R, G) gravity.22 It is shown that CDM epoch can be recon-
structed using modied GB theories of gravity.23 A descent amount of work has
been published so far in these theories.2428 In a recent paper,16 Sharif and Ikram
introduced a new modied theory known as f (G, T ) gravity that includes the trace
of the energymomentum tensor in the function. They also determined that the
massive test particles follow non-geodesic lines of geometry due to the presence of
extra force and examined the energy conditions for FriedmannRobertsonWalker
(FRW) universe. The same authors17 reproduced the cosmic evolution correspond-
ing to de Sitter universe, power-law solutions and phantom/non-phantom eras in
this theory using reconstruction techniques. For some particular choices of f (G, T )
gravity models, it is anticipated that this theory may explain the late-time cosmic
acceleration.
The exact solutions of the dierential equations have been calculated with the
help of the symmetry methods of approximations. The complexity involved in a
system of nonlinear equations is minimized through these by nding the undeter-
mined variables of equations. The Noether symmetries act like gadgets that pro-
vide the solutions and in addition to this, their existence gives suitable conditions
so that one can choose physical models of the universe compatible with cosmo-
logical observations. Many authors have used Noether symmetries to investigate
the cosmology in dierent contexts.2937 Using approximate symmetries, Sharif
and Waheed38 re-scaled the energy of stringy charged black hole solutions. Us-
ing Noether symmetries, Kucukakca39 found the exact solutions of Bianchi type I
model. Jamil et al.40 discuss f (T ), where T is the torsion scalar, specically for the

1750086-2
 Some exact solutions in f (G, T ) gravity

phantom and quintessence models using the Noether symmetry approach. Sharif
and Shaque41 examined Noether symmetries in a modied scalartensor gravity.
The exact solutions in f (R) gravity were also studied using Noether symmetries
methods for FRW spacetime.42 In a recent paper,43 we investigated f (G, T ) gravity
using Noether symmetry approach. Two specic models were studied to determine
the conserved quantities and it was concluded that the well-known de Sitter solu-
tion could be reconstructed for some specic choice of f (G, T ) gravity model. Thus,
it seems interesting to explore further the modied f (G, T ) gravity.
In this paper, our main focus is to investigate f (G, T ) gravity with anisotropic
background. We have considered the locally rotationally symmetric (LRS) Bianchi
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

type I spacetime for this purpose. Moreover, we adopt Noether symmetry approach
for the present analysis due to the complicated and highly nonlinear nature of the
eld equations. The organization of this paper is as follows. In Sec. 2, we provide
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

the basic framework for f (G, T ) gravity. Section 3 presents the Noether equations
of LRS Bianchi type I universe model for f (G, T ) gravity. Reconstruction of some
important cosmological solutions and graphical analysis is given in Sec. 4. The last
section is comprised of a brief outlook of the paper.

2. Preliminary Formalism of f (G, T ) Gravity

The general action for the modied f (G, T ) is,16
 
1 4
A = 2 d x g[R + f (G, T )] + d4 x gLM , (2)
2
where the function f (G, T ) is comprised of the GB term G and the trace of the
energymomentum tensor T , is the coupling constant, g is the determinant of
the metric tensor, R is the Ricci Scalar, and LM denotes the matter part of the
Lagrangian. The energymomentum tensor denoted by T can be given as

2 ( gLM )
T = . (3)
g g
However, the metric dependent energymomentum tensor will have the form
LM
T = g LM 2 . (4)
g
The following eld equations are obtained by varying Eq. (2) with respect to the
metric tensor.
G = 2 T [2Rg 2 2R 4g R 4R 2

+ 4R + 4R + 4R ]fG
1
+ g f [T + ]fT
2
[2RR 4R R 4R R + 2R R ]fG , (5)

1750086-3
 M. F. Shamir & M. Ahmad

where  = 2 = stands for the dAlembertian operator, G = R 12 g R

T
is the Einstein tensor, = g g , f f (G, T ), fG f (G,T
G
)
, and fT f T
(G,T )
.
Einstein equations are reproduced by putting f (G, T ) = 0 whereas eld equations
for f (G) are obtained simply by replacing f (G, T ) with f (G) in Eq. (5). Given below
is the trace of Eq. (5) which gives us an important equation as it can be used to
nd the corresponding cosmological f (G, T ) models
R + 2 T (T + )fT + 2f + 2GfG 2R2 fG + 4R fG = 0 . (6)
The covariant divergence of Eq. (5) is given as
 
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

fT g
T = 2 (T + ) (ln fT ) + T . (7)
fT 2
Equation (7) indicates that the conservation equation of energymomentum tensor
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

is not veried in usual Einsteins theory. However, the standard conservation equa-
tion for energymomentum tensor can be established by putting some constraints
to Eq. (7). In this paper, we restrict ourselves to LRS Bianchi I spacetime:
ds2 = dt2 A2 (t)dx2 B 2 (t)[dy 2 + dz 2 ] , (8)
where A and B are the cosmic scale factors. Corresponding expressions for the Ricci
scalar and GB term are given respectively as
   
A B AB B 2 AB 2 AB B
R = 2 +2 +2 + 2 , G=8 +2 , (9)
A B AB B AB 2 AB 2

where the dot gives the derivative with respect to time t. The standard matter
energymomentum tensor is dened as
T = ( + p)u u pg , (10)

satisfying the equation of state (EoS) parameter

p
= , (11)

where and p denote energy density and pressure of the uid, respectively. It is
mentioned here that the EoS parameter is important as it may describe the
dierent epochs of the accelerating universe. For 1 < 1/3, the dark en-
ergy (DE) phase is divided by quintessence epoch whereas < 1 and = 1
correspond to phantom and cosmological constant eras, respectively. It is very
dicult to determine the exact solutions of extremely nonlinear partial dierential
equations (PDEs) as the modied eld equations for LRS Bianchi type I spacetime
are of fourth-order with unknowns including f (G, T ). The benet of exact solutions
in modied gravity has signicant importance, especially in the study of phase
transitions and recent phenomenon of accelerated expansion of universe. Noether
symmetries can be used to nd the viable cosmological models and as a result, some
solutions with physical importance can be established.

1750086-4
 Some exact solutions in f (G, T ) gravity

3. Noether Symmetries and f (G, T ) Gravity

Noether symmetries have an important role in cosmology as they help to nd the
solutions of system of nonlinear eld equations. The uniqueness of the vector eld
in the tangent space can be established through the existence of Noether symmetry
approach. More conserved quantities can be recovered using this approach. For the
case of perfect uid, we rewrite the action (2) as


A = dt g[R + f (G, T ) 1 (G G) 2 (T T ) + LM ] . (12)

Here T and G stand for dynamical constraints, while the Lagrange multipliers 1
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

and 2 are calculated as

1 = fG (G, T ) , 2 = fT (G, T ) .
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

Since matter Lagrangian may have dierent denitions so here, we consider LM =

p(B). Moreover, we assume a constant ratio of shear and expanding scale factors,
which gives A = B m , where m is an arbitrary real number and for the sake of
nontrivial solutions, we consider m = 0, 1.44 Hence, after integration by parts, the
point-like Lagrangian becomes
L(B, R, G, T, B, R, G, T ) = B m+2 [R + f GfG fT {T (B) 3p(B)} p(B)]

8mB m1 B 3 GfGG 8mB m1 B 3 T fGT . (13)

We consider the vector eld and its rst prolongation respectively as45

W = (t, uj ) + i (t, uj ) j , (14)
t u

W [1] = W + ( i,t + i,j uj ,t ui ,j uj ui ) , (15)
uj
where and are the coecients of the generator, ui provides the n number of
positions. The Noether gauge symmetry given by the vector eld W
W [1] L + (D)L = DG(t, ui ) (16)
is preserved and the gauge term is denoted by G(t, ui ) and operator D is dened
as

D= + ui i .
t u
The EulerLagrange equations are given by
 
L d L
= 0. (17)
ui dt ui
Contraction of Eq. (17) with some unknown function i i (uj ) yields
  
L d L
i = 0. (18)
ui dt ui

1750086-5
 M. F. Shamir & M. Ahmad

It can be easily veried that

     
d i L d i L i d L
= . (19)
dt ui dt ui dt ui
Use of Eq. (19) in Eq. (18) results
   
L i d i L d i L
LW L = + = , (20)
ui dt ui dt ui
where L represents the Lie derivative along the vector eld. The existence of the
Noether symmetries would be possible only if the Lie derivative of the Lagrangian
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

becomes zero, i.e. the condition

LW L = 0 .
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

Since the Lagrangian L remains unchanged along the vector eld W , as a result,
the denition of Noether current turns out to be46
 
t i L
j = (21)
ui
and in order to conserve the Noether current, we must have

j,tt = 0 . (22)

The EulerLagrangian equations (17) in this case are found as

(m + 2)B m+1 [R + f GfG fT {T ((B) 3p(B))} p(B)]

+ B m+2 [fT (,B (B) 3p,B (B)) p,B (B)] + 24m[fGG ((m 1)B m2 B 3 G

+ 2B m1 B B G + B m1 B 2 G) + fGT ((m 1)B m2 B 3 T + 2B m1 B B T

+ B m1 B 2 T ) + 2B m1 B 2 G T fGGT + B m1 B 2 (G 2 fGGG + T 2 fGT T )] = 0 , (23)

B m+2 [GfGG T fGT + fGT ((B) 3p(B))]

+ 8mfGG (3Bm1 B 2 B + (m 1)B m2 B 4 ) = 0 , (24)

B m+2 [GfGT T fT T + fT T ((B) 3p(B))]

+ 8mfGT (3Bm1 B 2 B + (m 1)B m2 B 4 ) = 0 . (25)

Also the corresponding vector eld using Eq. (15) takes the form

W = + + + + + + + , (26)
B R G T B R G T
where , , and are functions of B, R, G and T . Now using Lagrangian (13)
and Noether equation (16) without the gauge term, an over-determined system of
PDEs is obtained:

1750086-6
 Some exact solutions in f (G, T ) gravity

(m 1)B m2 fGG + B m1 fGGG + B m1 fGGT + 3B m1 fGG
B

+ B m1 fGG + B m1 fGT = 0 , (27)
G G

(m 1)B m2 fGT + B m1 fGGT + B m1 fGT T

+ B m1 fGT + B m1 fGG + B m1 fGT = 0 , (28)
B T T

by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

fGG + fGT = 0 , fGG + fGT = 0 , (29)

R R B B

fGT + fGG = 0 , (30)
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

G T

fGG = 0 , fGG = 0 , fGT = 0 , fGT = 0 , (31)
G R T R

(m + 2)B m+1 [R + f (G, T ) GfG fT (T {(B) 3p(B)}) p(B)]

+ B m+2 + B m+2 [fT (,B (B) 3p,B ) p(B),B )]

+ B m+2 [GfGG fGT (T {(B) 3p(B)})]

+ B m+2 [T fT T + fT T ((B) 3p(B))] = 0 . (32)

Using Eq. (22), conservation equation for Noether charge becomes


d
[8mB m1 B 3 (GfGG + T fGT )]
dt B

m1 3 m1 3
+ (8mB B GfGG ) + (8mB B T fGT ) = 0 . (33)
G T

Since the system of PDEs (27)(32) is complicated and highly nonlinear, so we

make dierent assumptions for f (G, T ) gravity models to investigate a solution.
We rst assume that fGG = 0. Using Eqs. (27)(32), a trivial solution is obtained
= 0, = 0, = 0, and = 0. Additionally, the conservation equation (33) is
also satised in this case. For a nontrivial solution, we have to consider fGG = 0.
Therefore, as a next step, we consider fGT = 0 and fGG = 0. This choice provides
us with f (G, T ) = a0 G 2 + b0 T 2 , where a0 and b0 are the arbitrary constants. Here,
using the Noether equations we get = 0 = , = c1 and = c1 T + c2 . Thus, the
symmetry generator takes the form

W = (c1 T + c2 ) + c1 . (34)
R T
1750086-7
 M. F. Shamir & M. Ahmad

In this case, the conservation equation (33) referring to the Noether current gives
B 3 fGT = c3 , (35)
where c3 is a constant of integration. It is to be mentioned that when we choose c3
equal to zero, this case satises the conservation equation (33). The corresponding
Lagrangian becomes
L(B, R, G, T, B, R, G, T ) = B m+2 [R a0 G 2 b0 T 2 + 2b0 {(B) 3p(B)}T 2 p]

16a0mB m1 B 3 G (36)
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

and the EulerLagrangian equations are calculated as

B m+2 (R a0 G 2 b0 T 2 + 2b0 ((B) 3p(B))T p)

+ 32a0m(m 1)B m2 B 3 G + 96a0 mB m1 B G B

Mod. Phys. Lett. A Downloaded from www.worldscientific.com

+ 48a0mB m1 B 2 G = 0 , (37)

2a0 G + 16a0 m(m 1)B m2 B 4 + 48a0 mB 2 BB m1 = 0 , (38)

2b0 B m+2 (T ((B) 3p(B))) = 0 . (39)

Using Eqs. (38) and (39), it follows that
B = 0, B = c4 (40)
and
3(1+m)
B = c6 [(3m2 + 3m + 2)t 3c5 (1 + m)] 2+3m+3m2 , (41)
where c4 , c5 and c6 are the constants of integration. We discard the trivial solutions
given in Eq. (40). For the nontrivial case (Eq. (41)), the Ricci scalar R takes the
form as
6(m + 1)(7m + 5)
R= , (42)
[(3m(m + 1) + 2)t 3c5 (m + 1)]2
while the GB term G is given as
1296m3(m + 1)3
G= . (43)
[(3m2 + 3m + 2)t 3c5 (m + 1)]4
The average scale factor a is calculated as
  13 (2m+1)
3(m+1)
2
a = c6 {(3m + 3m + 2)t 3c5 (m + 1)} 3m 2 +3m+2
. (44)

The 3D graphical behavior of the scale factor a has been shown in Fig. 1. It can be
seen from Figs. 1(a) and (b) that the scale factor is increasing with the passage of
time, thus indicating the expansion in particular dimensions. This is because of the
increase in the distance between cosmologically related objects as the time passes.

1750086-8
 Some exact solutions in f (G, T ) gravity
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

(a) (b)
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

Fig. 1. The plot of the scale factor a(t) for (a) positive values of m and (b) for the negative
values of m.

This property is uniquely owned by the scale factor which is the function of cosmic
time t. The limiting behavior of the scale factor will tend to innity for the future
era, provided this accelerating expansion of the universe continues. Using Eqs. (11)
and (37)(39), we get

2 (1 3)2 + l(t) = 0 , (45)

where
1679616m6(1 + m)6
l(t) =
[(2 + 3m + 3m2 )t 3(1 + m)c5 ]8

1
+ 4478976(m 1)m4 (1 + m)6 (2 + 3m + 3m2 )
c6

3(1+m)
8 2+3m+3m
[(2 + 3m + 3m2 )t 3(1 + m)c5 ] 2

1
11197440m4(1 + m)5 (2 + 3m + 3m2 )2
c6

3(1+m)
2 8 2+3m+3m
[(2 + 3m + 3m )t 3(1 + m)c5 ] 2

 
1 4 5 2 2 3(1 + m)
+ 4478976m (1 + m) (2 + 3m + 3m ) 1 +
c6 2 + 3m + 3m2

3(1+m)
8 2+3m+3m
[(2 + 3m + 3m2 )t 3(1 + m)c5 ] 2

6(1 + m)(5 + 7m)

.
[(2 + 3m(1 + m))t 3(1 + m)c5 ]2

1750086-9
 M. F. Shamir & M. Ahmad

Now, one can see that Eq. (45) is quadratic in and its solution gives the following
two roots

1 1
= 2 +
2(1 3)2 [(2 + 3m(1 + m))t 3(1 + m)c5 ]8 c6

24(1 3)2 (746496m4(1 + m)6 (2 + m 3m3 )[(2 + 3m(1 + m))t

3(1+m)
3(1 + m)c5 ] 2+3m(1+m) + 746496m4(1 + m)5 (3m2 1)(2 + 3m(1 + m))
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

3(1+m)
[(2 + 3m(1 + m))t 3(1 + m)c5 ] (2+3m(1+m)) + 1866240m4(1 + m)5
3(1+m)
(2 + 3m(1 + m))2 [(2 + 3m(1 + m))t 3(1 + m)c5 ]] (2+3m(1+m))
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

+ 279936m6(1 + m)6 c6 + (1 + m)(5 + 7m)

 12 
6
[(2 + 3m(1 + m))t 3(1 + m)c5 ] c6 ) . (46)

The 3D graphical behavior of the energy density when plotted against the cosmic
time t is shown in Fig. 2. The plot clearly indicates the accelerating expansion of
the universe when = 1 with the positive value of the square root term. The
EoS parameter describes the dierent epochs of the accelerating universe. For
1 < 1/3, DE phase is divided by quintessence epoch whereas < 1 and
= 1 correspond to phantom and cosmological constant eras, respectively. Thus
our solution corresponds to CDM model which describes the parametrizations of
the universe containing the cosmological constant associated with the DE. The
graphical behavior of the rst root of justies the description of the CDM model
which exhibits the negative pressure p = c2 , and hence strengthens the reason

Fig. 2. Behavior of the energy density (t) for the positive square root values when = 1.

1750086-10
 Some exact solutions in f (G, T ) gravity

of the accelerating expansion of the universe. Thus for this particular cosmological
model, the solution metric takes the form
 2m
3(m+1)
2 2 2
ds = dt c6 {(3m + 3m + 2)t 3c5 (m + 1)} 3m 2 +3m+2
dx2

6(m+1)
c26 [(3m(m + 1) + 2)t 3c5 (m + 1)] 3m(m+1)+2 [dy 2 + dz 2 ] . (47)

4. Reconstruction of CDM-Type Cosmology

In this section, we propose a cosmological model f (G, T ) = G k T 1k , where k is any
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

arbitrary real number. We develop some cosmological solutions for the case k = 2.
The Lagrangian takes the form
  
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

2 m1 3
{(B) 3p(B)}G 16B B m T
L = B m+2 R p(B) G . (48)
T2 T T

Here, EulerLagrangian equations take the form


m+1 {(B) 3p(B)}G 2 48mB m1 B 2 T
(m + 2)B R p(B)
T2 T2

2
({(B) 3p(B)}),B G 48m(m 1)B m2 B 3 G
B m+2 + p ,B (B) +
T2 T

96mB BB m1 G 48mB m1 B 2 G 48mB m1 B 2 G T

+ +
T T T2
96mB m1 B 2 T 2 48m(m 1)B m2 B 3 T 96mB BB m1 T
+
T3 T2 T2

16m(m 1)B ( m 2)B 3 T
G = 0, (49)
T T

2B m+2 ((B) 3p(B))G + 16m(m 1)T B m2 B 4

+ 48mT B m1B 2 B 16mB m1 T B 3 = 0 , (50)

B m+2 ((B) 3p(B))G 2 + 8mT B m1 B 3 G 8m(m 1)T B m2 B 4

24mB m1 T B 2 B = 0 . (51)

By putting the corresponding values for G and T and using Eq. (51), we get

64m2 (m 1)(m + 3)B m6 B 8 + 960B m4 B 4 B 2 + 64m2 (10m 19)B m5 B 6 B

...
+ 192m2B 5 B m4 B 8m(m 1)B m2 B 4 24mB m1 B 2 B = 0 . (52)

1750086-11
 M. F. Shamir & M. Ahmad

This equation admits an exponential solution

B = et , (53)
with the constraint equation
8m2 (m2 + 12m 4)8 (m2 + 2m)4 = 0 , (54)
where is an arbitrary constant. Equation (54) yields the real solutions
1
(2 + m) 4
= 0, = 1 . (55)
(32m + 96m2 + 8m3 ) 4
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

Now using Eqs. (49) and (50), we get

1
B 4 p,B (B) + (m + 2)B 3 p(B) = [48m(m 1)B 3 G + 96mB BB G + 48mB B 2 G
c7
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

(m + 2)B 3 G 2 16m(m 1)B 3 G] . (56)

It may be noted that Eq. (56) is a non-homogeneous linear dierential equation in
pressure p(B) and its solution is obtained as

m2 192c8m4 B 2 (136 152m + 22m2 9m3 + 3m5 )B 6
p(B) = c7 B
B8 m6

2(182 + 154m 14m2 + 3m3 + 3m4 )B B 4 B

+
m5
B 2 B 2 ((3m3 + 6m2 128m + 254)B 2 + 4(16 7m)B (3) B)
+
m4

6B 3 (6B 3 + B (4) B 2 + 8B (3) B B)
, (57)
m3

where c7 and c8 are the constants of integration. Now using the positive from
Eq. (55), the expression for pressure explicitly becomes function of time

(m 2) 4 m + 2t 1
p(t) = c7 exp
4
8m3 + 96m2 32m [(m 6)(m 5)(m 4)(m 3)]
8

4
m+2
192c8
4
m4 [3m8 30m7 + 63m6 + 132m5
8m + 96m2 32m
3

326m4 258m3 186m2 + 1094m + 228g]. (58)

The 3D plotting of the pressure p(t) against the time t and anisotropy param-
eter m shows the dierent behaviors of the pressure. Figures 3 and 4 have been

1750086-12
 Some exact solutions in f (G, T ) gravity
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

(a) (b)
Fig. 3. Behavior of the pressure p(t), (a) when 1 < m < 3 and (b) when 3 < m < 4, with c7 = 1,
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

c8 = 1.

(a) (b)

Fig. 4. Behavior of the pressure p(t), (a) when 4 < m < 5 and (b) when 5 < m < 6, with
c7 = c8 = 1.

Fig. 5. Behavior of the pressure p(t), when m > 6 with c7 = c8 = 1.

1750086-13
 M. F. Shamir & M. Ahmad

avoided from the singularities caused due to m = 3, 4, 5, 6, showing mostly the

positive pressure but when m > 6, the negative pressure is attained throughout as
we go on increasing the time t (as shown in Fig. 5). Thus the solution for anisotropy
parameter m > 6 corresponds to CDM model. It is the interesting feature of mod-
ied GB gravity that specic f (G, T ) gravity models may be used to reconstruct
CDM cosmology without involving any cosmological constant. Thus in this case
the solution metric becomes
ds2 = dt2 e2mt [dx2 ] e2t [dy 2 + dz 2 ] . (59)
The familiar de Sitter spacetime in GR is recovered when m = 1. Here we have con-
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

structed a physical cosmological solution with a particular f (G, T ) gravity model.

In the similar way, more solutions with some other cosmological models can be
reconstructed.
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

5. Concluding Comments
This paper is devoted to study recently proposed f (G, T ) theory of gravity16 with
anisotropy background. For this purpose, we consider LRS Bianchi type I cosmo-
logical model in the presence of perfect uid. Since the eld equations are highly
nonlinear and complicated, so we assume that ratio of shear and expansion scalars
is constant, which gives A = B m .44 Noether symmetries not only aid to investigate
the hidden symmetries, but also their existence provide suitable conditions so that
we can choose physical models of universe compatible with recent observations.
Therefore, we analyze Noether symmetries of the LRS Bianchi type I universe in
f (G, T ) theory of gravity. It would be worthwhile to mention here that our results
agree with Ref. 43 for a special case when m = 1. We have calculated the La-
grangian for LRS Bianchi type I universe model in f (G, T ) theory. The existence of
Noether charges is extremely signicant in the literature, and the equation for con-
servation of charges plays an important role to examine the Noether symmetries.
The conservation equation for Noether charges has been established.
The exact solutions of Noether equations have been discussed for two cases of
f (G, T ) gravity models. The rst case when fGG = 0 yields trivial symmetries while
we obtain nontrivial symmetries for the second case when fGT = 0 and fGG = 0.
Thus, the second case provides f (G, T ) = a0 G 2 + b0 T 2 gravity model, where a0
and b0 are arbitrary constants. Moreover, solutions in both cases satisfy the conser-
vation equation for Noether charges. We have also reconstructed some important
cosmological solutions by proposing f (G, T ) = G k T 1k , where k is an arbitrary
real number. This model yields the familiar de Sitter solution already available in
GR. Moreover, the solutions for anisotropy parameter m > 6 correspond to CDM
model. Thus, the interesting feature of modied GB gravity is that specic f (G, T )
gravity models may be used to reconstruct CDM cosmology without involving
any cosmological constant.

1750086-14
 Some exact solutions in f (G, T ) gravity

Acknowledgment
The authors are thankful to National University of Computer and Emerging
Sciences (NUCES) for funding support.

References
1. A. G. Riess et al., Astrophys. J. 607, 665 (2004).
2. C. L. Bennett et al., Astrophys. J. Suppl. 148, 1 (2003).
3. M. Tegmark et al., Phys. Rev. D 69, 103501 (2004).
4. A. Azadi, D. Momeni and M. Nouri-Zonoz, Phys. Lett. B 670, 210 (2008).
5. A. A. Starobinsky, J. Exp. Theor. Phys. Lett. 86, 157 (2009).
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

6. S. Appleby, R. Battye and A. A. Starobinsky, J. Cosmol. Astropart. Phys. 06, 005

(2010).
7. S. Capozziello, E. De-Filippis and V. Salzano, Mon. Not. R. Astron. Soc. 394, 947
(2009).
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

8. K. Bamba, S. Capozziella, S. Nojiri and S. D. Odintsov, Astrophys. Space Sci. 342,

155 (2012).
9. S. Nojiri and S. D. Odintsov, Problems of Modern Theoretical Physics, A Volume in
Honour of Prof. I. L. Buchbinder, in the Occasion of his 60th Birthday (TSPU, 2008),
pp. 266285, arXiv:0807.0685.
10. S. Nojiri and S. D. Odintsov, Phys. Lett. 599, 137 (2004).
11. M. Akbar and R. Cai, Phys. Lett. B 648, 243 (2007).
12. R. Cai, Y. Liu and Y. Sun, JHEP 0910, 080 (2009).
13. T. Harko, F. S. N. Lobo, S. Nojiri and S. D. Odintsov, Phys. Rev. D 84, 024020
(2011).
14. S. Capozziello, P. Martin-Moruno and C. Rubano, Phys. Lett. B 664, 12 (2008).
15. A. D. Felice and S. Tsujikaswa, Living Rev. Relativ. 13, 3 (2010).
16. M. Sharif and A. Ikram, Eur. Phys. J. C 76, 640 (2016).
17. M. Sharif and A. Ikram, arXiv:1612.02037.
18. G. Cognola, E. Elizalde, S. Nojiri, S. D. Odintsov and S. Zerbini, Phys. Rev. D 75,
086002 (2007).
19. G. Cognola, E. Elizalde, S. Nojiri, S. D. Odintsov and S. Zerbini, Phys. Rev. D 73,
084007 (2006).
20. S. Nojiri and S. D. Odintsov, Phys. Lett. B 631, 1 (2005).
21. T. Chiba, J. Cosmol. Astropart. Phys. 03, 008 (2005).
22. S. Nojiri, S. D. Odintsov and P. V. Tretyakov, Progr. Theor. Phys. Suppl. 172, 81
(2008).
23. E. Elizalde, R. Myrzakulov, V. V. Obukhov and D. Saez-Gomez, Class. Quantum
Grav. 27, 095007 (2010).
24. M. Sharif and H. I. Fatima, Mod. Phys. Lett. A 30, 1550142 (2015).
25. M. Sharif and H. I. Fatima, J. Exp. Theor. Phys. 149, 121 (2016).
26. K. Bamba, S. D. Odintsov, L. Sebastiani and S. Zerbini, Eur. Phys. J. C 67, 295
(2010).
27. A. D. Felice, T. Suyama and T. Tanaka, Phys. Rev. D 83, 104035 (2011).
28. L. Sebastiani, Springer Proc. Phys. 137, 261 (2011).
29. M. Jamil, F. M. Mahomed and D. Momeni, Phys. Lett. B 702, 315 (2011).
30. I. Hussain, M. Jamil and F. M. Mahomed, Astrophys. Space Sci. 337, 373 (2012).
31. U. Camci, Eur. Phys. J. C 74, 3201 (2014).
32. U. Camci, J. Cosmol. Astropart. Phys. 07, 002 (2014).
33. P. Pani, T. P. Sotiriou and D. Vernieri, Phys. Rev. D 88, 121502 (2013).

1750086-15
 M. F. Shamir & M. Ahmad

34. M. Sharif and I. Fatima, J. Exp. Theor. Phys. 122, 104 (2016).
35. M. Sharif and S. Waheed, Phys. Scripta 83, 015014 (2011).
36. M. Sharif and S. Waheed, J. Cosmol. Astropart. Phys. 02, 043 (2013).
37. M. Sharif and I. Fatima, Int. J. Mod. Phys. D 25, 1650011 (2016).
38. M. Sharif and S. Waheed, Can. J. Phys. 88, 833 (2010).
39. Y. Kucukakca, U. Camci and I. Semiz, Gen. Relat. Gravit. 44, 1893 (2012).
40. M. Jamil, D. Momeni and R. Myrzakulov, Eur. Phys. J. C 72, 2137 (2012).
41. M. Sharif and I. Shafique, Phys. Rev. D 90, 084033 (2014).
42. S. Capozziello, A. Stabile and A. Troisi, Class. Quantum Grav. 24, 2153 (2007).
43. M. Shamir and M. Ahmad, Eur. Phys. J. C 77, 55 (2017).
44. C. B. Collins et al., Gen. Relat. Gravit. 12, 805 (1980).
45. P. J. Olver, Applications of Lie Groups to Dierential Equations (Springer Science &
by UNIVERSITY OF COLORADO AT BOULDER on 04/27/17. For personal use only.

Business Media, 2000).

46. S. Capozziello, N. Frusciante and D. Vernieri, Gen. Relat. Gravit. 44, 1881 (2012).
Mod. Phys. Lett. A Downloaded from www.worldscientific.com

1750086-16