See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/322930320

F(T, R) theory of gravity

Article in International Journal of Modern Physics D · January 2018

DOI: 10.1142/S0218271818500621

CITATIONS READS

11 1,128

5 authors, including:

Mustafa Salti Murat Korunur

Mersin University Munzur Üniversitesi
93 PUBLICATIONS 781 CITATIONS 32 PUBLICATIONS 153 CITATIONS

SEE PROFILE SEE PROFILE

Irfan Acikgoz Nurettin Pirinccioglu

Dicle University Dicle University
22 PUBLICATIONS 120 CITATIONS 14 PUBLICATIONS 30 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Mustafa Salti on 11 February 2021.

The user has requested enhancement of the downloaded file.

January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

International Journal of Modern Physics D

⃝c World Scientific Publishing Company

f (T, R) theory of gravity

Mustafa Salti∗
Department of Physics, Faculty of Art and Science, Mersin University,
Mersin-33343, Turkey.
[email protected]

Murat Korunur
Deparment of Electricity and Energy, Tunceli Vocational College, Munzur University,
Tunceli-62000, Turkey.
[email protected]

Irfan Acikgoz
No affiliation.
[email protected]

Nurettin Pirinccioglu† and Figen Binbay‡

Department of Physics, Faculty of Science, Dicle University,
Diyarbakir-21280, Turkey.
† [email protected], ‡ fi[email protected]

We mainly focus on the idea that the dynamics of the whole universe may be understood
by making use of torsion T and curvature R at the same time. The f (T, R) gravity
can be considered as a fundamental gravitational theory describing the evolution of
the universe. The model can produce the unification of the general relativity (GR),
teleparallel gravity (TPG), f (R) gravity and f (T ) gravity theories. For this purpose,
the corresponding Lagrangian density is written in terms of an arbitrary function of
the torsion and curvature scalars. Furthermore, we use the absense/existence puzzle of
relativistic neutron stars and thermodynamical laws as constraining tools for the new
proposal.

Keywords: Modified gravity; Dark energy; Neutron stars; Thermodynamics.

1. Introduction
Recent astrophysical data1,2,3,4,5,6,7,8,9 have pointed out that the observable uni-
verse expands faster than it should. It is commonly believed that this outlandish
behavior is caused by some mysterious ingredients such as dark energy and dark
matter. Understanding the phase transition from decelerating to accelerating is
one of the most challenging and interesting topics in modern cosmology and fun-
damental physics today. Although a large number of theoretical candidates have

∗ Corresponding author.

1
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

2 Salti et al.

been introduced in literature, the enigmatic speedy expansion behavior of the cur-
rent universe is still completely unknown. The cosmological constant, scalar fields
dark energy models10,11,12,13,14 , unified dark energy-matter definitions15,16 , modi-
fied gravity theories17,18,19,20 and even extra dimensions are possible dark energy
candidates. Li et al.21 and Cai et al.22 have given very useful briefs about the dark
universe including a survey of some possible proposals for the explanation of the
speedy expansion phase.
Among different dark energy descriptions mentioned above, the modified the-
ory of gravity idea is quite significant, as it incorporates some notions of the clas-
sical and quantum gravity theories. As a matter of fact, modifying a theory of
gravity is risky and we may encounter some unwanted consequences such as viola-
tions of experimental results, instabilities and ghosts. On the other hand, adopting
modified theories helps to get very significant conclusions at cosmological scales.
However, we have no acceptable physical reason to consider one of them as ca-
pable of matching the data at all cosmological scales. There are varied modifica-
tions of gravity theories: f (R) gravity, f (T ) gravity (here T implies the torsion
scalar), Gauss-Bonnet gravity, fractal gravity, multiple field modified theory and so
on23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42 . Here, we introduce a new modi-
fication of gravity to interpret dark matter-energy dominated part of the universe.
The GR is constructed by considering torsion-free Riemannian geometry. There
are many ideas which show that the torsion effects should be considered in extended
forms of the GR43,44,45,46 . The torsion is described algebraically by considering the
spin-matter fields, thus it cannot propagate and is detected under spin-matter inter-
actions only. Besides, in f (R) theory of gravity, the curvature propagates without
any matter source and this behavior is the same as the corresponding feature of
f (T ) gravitational theory35 . In our f (T, R) theory of gravity, the torsion can propa-
gate without existence of the spin-matter fields as in the f (T ) theory, which means
our model inherits the convenient features of its constituents. In other words, both
torsion and curvature do not need any matter source in order to propagate in the
f (T, R) gravity. This is the key advantage of our idea and our model also contains
four well known gravity theories as particular cases. Bahamonde et al.47 showed
that the f (T ) theory is invariant under local lorentz transformation and it is a
unique second order theory. So, the modification of gravity may contain more in-
variants and there is no reason to restrict the GR, TPG, f (R) gravity or the f (T )
theory. Since the curvature and torsion are two geometrical quantities, they can
be combined in a new modified theory without matter sources. It is important to
mention here, that there are many papers on f (R, T ) theory of gravity where T
describes the trace of stress energy tensor instead of torsion. For instance, Jamil
et al.36 derived the first law of thermodynamics in f (R, T ) gravity by using the
method proposed by Wald48 . Additionally, R. Myrzakulov37 considered a theory of
gravity including a metric-dependent torsion where T indicates the torsion in order
to study geometric the root of the theory.
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 3

We organized this studya as follows. In the next section, we introduce our pro-
posal with its basics and give some particular cases in order to show that our model
can be reduced to four main modified gravity theories. In the third section, we use
the absense/existence problem of relativistic Neutron stars as a contracting and
analysis tool for the model. For this purpose, we rewrite the corresponding field
equations of f (T, R) theory by making use of a spherically symmetric line-element.
The fourth section is devoted to the investigation of cosmology in f (T, R)-gravity
via thermodynamical laws. Finally, in the last section, we give a summary of the
work.

2. The model
We start with the fact that the tetrad field hµi is a basic entity of the TPG, while
the metric tensor constructs the basis of the GR.
It is known that the tetrad field satisfies the following relations:
hµi hjµ = δij , hµi hiν = δνµ , (1)
where hjµ denotes the inverse form of hµj . At the same time, tetrads are related to
the metric tensor as
gµν = ηab haµ hbν . (2)
Here, ηab = diag(1, −1, −1, −1) describes the Minkowski universe.
Now, we are in a position to construct the f (T, R)-gravity. For this aim, we
assume the following action as a first step:
∫
1
S=− d4 xhf (T, R) + Sm , (3)
2κ
√
where h = −g shows the determinant of the tetrad where g is the determinant
of metric tensor. Next, f (T, R) and Sm indicate a non-linear function and matter
part of the action, respectively. Note that, we take κ = 8πG for simplicity in further
calculations.
The torsion tensor is constructed as
e ανµ − Γ
T αµν = Γ e αµν , (4)

where Γe αµν = hα ∂ν hkµ denotes the curvatureless Weitzenböck connection. Hence,

k
the torsion scalar is written in the following form
T = Sρ µν T ρµν , (5)
with
S ρµν = K µνρ − g µν T σµσ + g ρµ T σνσ , (6)

a Throughout this work, we use the Greek alphabet (α, β, µ, ν...) and Latin alphabet (a, b, i, j...)

in order to represent the space-time and the tangent space, respectively. All of these indices are
running over the values (0, 1, 2, 3).
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

4 Salti et al.

1 ( νµ )
K µν ρ = T ρ + Tρ µν − T µνρ . (7)
2
Moreover, the curvature scalar is defined by R = g µν Rµν in the GR where g µν
and Rµν denote the inverse metric tensor and the Ricci tensor, respectively. We
know that the GR is built with the torsionless Levi-Civita connection of a metric
which is defined by
1 αβ
Γαµν = g (∂ν gβµ + ∂µ gβν − ∂β gµν ), (8)
2
thus, the Riemann curvature tensor is written in terms of the Levi-Civita connection
as

Rαβ = ∂η Γη βα − ∂β Γη ηα + Γη ηλ Γλ βα − Γη βλ Γλ ηα . (9)

Variation of the action can be written as given below

∫
1
δS = d4 x[f (T, R)δh + hδf (T, R)] + δSm = 0. (10)
2κ
It can be seen that δf (T, R) is a significant term which can be written as
∂f (T, R) ∂f (T, R)
δf (T, R) = δR + δT. (11)
∂R ∂T
At this step, we need to find a connection between δR and δT . So, we may use
∫ ∫
∂f (T, R) ∂f (T, R)
I = d4 xhδf (T, R) = d4 x[h δR + h δT ]. (12)
∂R ∂T
It is easy to see that

δR = δ(g µν Rµν ) = Rµν δg µν + (∇µ ∇ν − g µν ∇λ ∇λ )δgµν , (13)

and

hδT = hδ(Sρµν Tµν

ρ
) = [2∂ν (hhρk Sρµν ) − 2hhγk S ρβµ Tρβµ ]δhkµ − 2∂ν (hhρk Sρµν δhkµ ),
(14)

where ∇α shows the usual covariant derivative. Making use of integration by parts
with the above relations, we obtain
∫ [
∂f (T, R) ∂f (T, R) γ ρβµ
I = d4 x 2 ∂ν (hhρk Sρµν ) − 2h hk S Tρβµ
∂T ∂T
]
∂f (T, R) ρ µν
+2h(∂ν )hk Sρ δhkµ
∂T
∫ [ ]
∂f (T, R) ∂f (T, R) µν
+ d4 xh (∇µ ∇ν − g µν ∇λ ∇λ ) − R δgµν . (15)
∂R ∂R
Now, the relation between the metric tensor and the tetrad fields given in equa-
tion (2) can be used in order to rewrite the integral (15). After performing a few
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 5

mathematical steps, we find

∫
{ [ ]
δS = d4 xh hkν (∇µ ∇ν − g µν ∇λ ∇λ )fR − fR Rµν
1 1
+ hµk f + fT ∂ν (hhσk Sσ µν ) − fT hγk S ρβµ Tρβγ
2 h }
+hσk Sσ µν (fT T ∂ν T + fT R ∂ν R) − κhρk Θµρ δhkµ = 0. (16)
From this point of view, the corresponding field equations of f (T, R)-gravity are
given by
1 µ [ ] 1
hk f + hkν (∇µ ∇ν − g µν ∇λ ∇λ )fR − fR Rµν + fT ∂ν (hhσk Sσ µν )
2 h
−fT hγk S ρβµ Tρβγ + hσk Sσ µν (fT T ∂ν T + fT R ∂ν R) = κhρk Θµρ , (17)
∂2f 2
where fR ≡ ∂f
∂R , fT ≡ ∂f
∂T ,fT T ≡ and fT R ≡ ∂T
∂T 2
∂ f
∂R . Also, we assumed
∫
1
δSm =− d4 xhΘµk δhkµ . (18)
2
Note that Θµk = hρk Θµρ is related to the usual energy-momentum tensor Θµν by
Θµν = η kl hνl Θµk .
We can check easily whether the theory can be reduced to the other well-known
gravitation theories such as the GR, TPG, f (R)-gravity or the f (T )-gravity. For
this aim, we can assume the following specific limiting conditions.
Case i : f (T, R) = R
We get fR = 1 and fT = fT R = fT T = 0. As a result, the field equation (17)
transforms into the following form
1 µ
h R − hkν Rµν = κΘµk . (19)
2 k
After multiplying both sides of this relation with a factor hνα , we obtain
1 να µ
h h R − hνα hkν Rµν = κ hνα Θµk , (20)
2 | {z k} | {z } | {z }
g µα δνα Θµα

Hence, it is concluded that Einstein’s GR theory can be recovered exactly.

Case ii : f (T, R) = T
It can be found that fT = 1 and fR = fT R = fT T = 0. Therefore, the equation
(17) with these conditions yields
1 µ 1
h T + ∂ν (hhρk Sρµν ) − hγk S ρβµ Tρβµ = κΘµk , (21)
2 k h
which is known as the TPG’s field equation.
Case iii : f (T, R) = f (R)
Consequently, we obtain fT = fT R = fT T = 0 and find
1 µ
h f + hkν [(∇µ ∇ν − g µν ∇λ ∇λ )fR − fR Rµν ] = κΘµk . (22)
2 k
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

6 Salti et al.

Multiplying both sides of the above field equation with a factor hkα leads
δνα
z }| {
1 kα µ
h hk f + hkα hkν [(∇µ ∇ν − g µν ∇λ ∇λ )fR − fR Rµν ] = κ hkα Θµk . (23)
2 | {z } | {z }
g µα Θµα

So, one can rewrite this equation as

1
fR Rµν − g µν f = (∇µ ∇ν − g µν ∇λ ∇λ )fR + κΘµα , (24)
2
which describes the well-known f (R)-gravity.
Case iv : f (T, R) = f (T )
Here, we assume also that fT T is very small. Thus, it is found that fR = fT R = 0.
In this way, using equation (17) with the above assumptions leads to
1 µ 1
h f + fT ∂ν (hhρk Sρµν ) − fT hγk S ρβµ Tρβµ = κΘµk . (25)
2 k h
In literature, this final result is known as the field equation of f (T )-gravity.

3. Absence/existence puzzle of neutron stars

It is commonly known that a theory should be able to pass several tests before con-
sidering it as a viable model49 . In the weak gravity regime, it must be compatible
with solar system tests. At cosmological scales, it should produce the speedy expan-
sion, should not contain gravitational instabilities and must obey the constraints
of current literature. Moreover, in strong gravity regime, it should give meaningful
solutions for the neutron stars with mass-radius relation inside the current ob-
servational bounds. Here, we discuss the absense/existence problem of relativistic
neutron stars50,51,52,53 in order to analyze the f (T, R) theory in the strong field
regime.
One can describe relativistic neutron stars by starting with two assumptions 53,49
about the metric and energy-momentum tensors. Neutron stars have the following
structure,
ds2 = e2Ω(r) dt2 − e2ξ(r) dr2 + r2 (dθ2 + sin2 θdϕ2 ). (26)
Matter inside the neutron stars is a perfect fluid and it is defined by the energy-
momentum tensor Θµρ = diag(ρ, −p, −p, −p). Additionally, the surviving compo-
nents of the metric tensor and tetrad fields are obtained as
[ ]
gµν = δµ0 δν0 e2Ω(r) − δµ1 δν1 e2ξ(r) + r2 δµ2 δν2 + sin2 θδµ3 δν3 , (27)
[ ]
hiµ = δµ0 δ0i eΩ(r) + δµ1 δ1i eξ(r) + r δµ2 δ2i + sin θδµ3 δ3i . (28)
Also, it can be found that
h = r2 sin θeΩ(r)+ξ(r) , (29)
2e−2ξ
R= (1 − e2ξ + 2rΩ′ + r2 Ω′2 − 2rξ ′ − r2 Ω′ ξ ′ + r2 ξ ′′ ), (30)
r2
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 7

and
2e−2ξ
T = (1 + 2rΩ′ ), (31)
r2
where the prime indicated derivation with respect to r.
Furthermore, one can calculate the following non-zero components of the con-
travariant Ricci tensor
e−2(ξ+Ω) [ ]
R00 = (2 − rξ ′ )Ω′ + r(Ω′2 + Ω′′ ) , (32)
r
e−4ξ [ ]
R11 = (2 + rΩ′ )ξ ′ − r(Ω′2 + Ω′′ ) , (33)
r
e−2ξ ( −2ξ )
R22 = 4
e − 1 + rξ ′ − rΩ , (34)
r
e−2ξ ( −2ξ )
R33 = 2 4
e − 1 + rξ ′ − rΩ . (35)
sin θr
In addition to these results, the surviving components of torsion tensor are found
as
T010 = −T001 = e−2Ω Ω′ T221 = −T212 = r, (36)
T331 = −T313 = r sin2 θ, (37)
2
r
T332 = −T323 = sin 2θ, (38)
2
while the non-vanishing components of S µνλ are calculated as given below
2e−2(Ω+ξ)
S 010 = −S 001 = , (39)
r
e−2ξ cot θ
S 020 = −S 002 = , (40)
r2
e−2Ω cot θ
S 020 = −S 002 = S 112 = −S 121 = , (41)
r2
e−2ξ
S 221 = −S 212 = (1 + rΩ′ ), (42)
r3
e−2ξ
S 331 = −S 313 = (1 + rΩ′ ). (43)
sin2 θr3
So, making use of these calculations leads to
f 1
κρ = − (T fT + RfR ) + λ1 (fT − fR ) − 2r2 R′2 fRRR
2 2
+(λ2 − rT )R′ fRR − 4rR′ fRT − 4rT ′ fT T , (44)
f 1 1 1
−κp = − (2T fT + RfR ) + 2 (fT − fR ) + T fR − λ3 fRR , (45)
2 2 r 2
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

8 Salti et al.

f 1
−κp = − (T fT + RfR ) + λ4 (fR − fT ) + λ5 fRR
2 2
−λ6 (T ′ fT T + R′ fRT ) + e−2ξ R′2 fRRR , (46)

where
e−2ξ 2ξ
λ1 = (e + 2rξ − 1), (47)
r2
( )
1
λ2 = 2e2ξ ξ ′ + 1 − , (48)
r
R′ e−2ξ
λ3 = (2 + rΩ′ ), (49)
r
e−2ξ 2ξ
λ4 = (e + rΩ′ + rξ ′ ), (50)
r2
e−2ξ ′ ′
λ5 = [R (rξ − 3 − rΩ′ ) + rR′′ ], (51)
r
e−2ξ
λ6 = (1 + rΩ′ ). (52)
r
After these steps, we are in a position to discuss the compatibility of the above
field equations with the conservation equation. In the GR, it is commonly assumed
that the energy-momentum tensor is a covariant constant. Therefore, one can get
a definition for the hydrostatic equilibrium53 :
dp dΩ
= −(ρ + p) . (53)
dr dr
We can determine the mass-radius relation and other global aspects of the neu-
tron stars if the metric function Ω(r) is known53,49 . Although the Levi-Civita and
Weitzenböck connections are different from each other, the conservation relation
turns out to be the same in the GR and the TPG49 . In order to obtain a conserva-
tion relation in f (T, R) theory from the field equations (44), (45) and (46), we can
perform three steps: (i ) taking derivative of equation (45) with respect to r, (ii )
adding equations (44) and (46), then multiplying the result by Ω′ , (iii) adding the
results of steps i and ii. Consequently, we find
dp dΩ 1
= −(ρ + p) − F (54)
dr dr κ
with

F (r) = ω1 (fT − fR ) + ω2 fT T + ω3 fRT + ω4 fRR + ω5 fRRR + λ3 T ′ fRRT , (55)

where
2 T′
ω1 = − 3
− + (λ1 + λ4 )Ω′ , (56)
r 2
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 9

[ ]
′ 1 ′
ω2 = T − T − (4r + λ6 )Ω , (57)
r2
T′
ω3 = −R′ [T + Ω′ (4r − λ6 )] − (R − T ), (58)
2
( )
1 R−T
ω4 = λ′3 − + R′ + (λ2 − rT )R′ Ω′ − λ5 Ω′ , (59)
r2 2
ω5 = λ3 R′ − 2r2 R′2 Ω′ − e−2ξ R′2 Ω′ . (60)
We see that equation (54) is in contradiction with the conservation relation given
by equation (53) which should be satisfied in all cases. For this reason, F must be
equal to zero in order to remove contradiction:
ω1 (fT − fR ) + ω2 fT T + ω3 fRT + ω4 fRR + ω5 fRRR + λ3 T ′ fRRT = 0. (61)
Thus, we must construct a useful form of the f (T, R) function satisfying the require-
ment (61). We see that the solution of the puzzle depends on the selected f (T, R)
model. For the sake of simplicity, one can take f (T, R) = αT + βR, where α and
β are two arbitrary constants. So, it can be found that F = ω1 (α − β). Therefore,
if we assume α = β or take f (T, R) = T + R, it can be said that the relativistic
neutron stars exist in the f (T, R)-gravity. The other possibility for the existence of
the relativistic neutron stars is a suitable choice of the functions Ω(r) and ξ(r).

4. Cosmology via Thermodynamics

Cai and Kim54 showed in the GR that the Friedman equation can be rewritten in
the form of the first law of thermodynamics on the dynamical apparent horizon
reh , i.e. −dE = TH dSh where TH = (2πe rh )−1 , Sh = πe rh2 G−1 and dE are the
Hawking temperature, horizon entropy and the amount of internal energy flows
through horizon, respectively. Later, the general relativistic Friedman equations
was rewritten again in another form55 :
dE = TH dSh + W dV, (62)
where V = 4 3
3 πr̃h ,
E = ρV and W = − p) represent the volume containing the
1
2 (ρ
matter with radius of the apparent horizon, internal energy and the work density,
respectively.
In this section, we discuss the validity of the first and second laws of ther-
modynamics in the framework of f (T, R)-gravity. Bekenstein56 proved that there
is a relation between the black hole thermodynamics and the event horizon. It is
known that the event horizon is a measure of entropy in black hole physics, which
means that each horizon corresponds to an entropy. In addition to this, Wang et
al.57 also found very significant conclusions by making use of the thermodynamical
laws. The authors showed that our universe must be non-static, however the usual
descriptions of the thermodynamical quantities on the event horizon may be more
complicated than those ones defined in the static spacetime.
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

10 Salti et al.

The apparent horizon in the Friedmann-Robertson-Walker (FRW) universe is

[ ]− 1
defined by59,60,58 r̃h = H 2 + ak2 2 . Next, the corresponding Hawking temperature
on the apparent horizon is described as54
1 2H r̃h − r̃˙ h
TH = , (63)
2πr̃h 2H r̃h
where the dot means derivative with respect to t. We see that 2H r̃h > r̃˙ h provides
that the temperature is always positive. Cai et al.61 showed that the apparent
horizon of the FRW spacetime has an associated Hawking temperature.
• The first law
In the GR, the horizon (apparent) entropy is given by the Bekenstein-Hawking
A
definition, i.e. Sh = 4G where A = 4πr̃h2 is the area of the horizon62,63,56 . In
48,64
the f (R) and f (T ) theories of gravity65 , the corresponding horizon entropy
relations are written as ShR = Af T AfT
4G and Sh = 4G , respectively. Consequently, in
R

the f (T, R) theory of gravity, we define the horizon entropy as

A
Sh = (fR + fT ). (64)
2G
Taking time derivative of the above definition, it follows that
[ ]
dSh 4πr̃h ˙ r̃h ˙ ˙
= r̃h (fR + fT ) + (fR + fT ) . (65)
dt G 2
r̃h −r̃ h
˙
After multiplying both sides of this result by a factor 2H2H r̃h , we obtain
[ ]− 52
1 k k
Th dSh = H2 + 2 ( 2 + Ḣ + 2H 2 )Fe1 dt, (66)
2G a a
where
k k
Fe1 = H( 2 − Ḣ)(fR + fT ) + (H 2 + 2 )(f˙R + f˙T ). (67)
a a
On the other hand, in the GR, the energy of the Universe within the dynamical
r̃h
apparent horizon is defined by the Misner-Sharp66,67 relation, i.e. E = 2G . Inspired
by this definition, in the f (T, R)-gravity, we define the corresponding energy as
r̃h
E′ = (fR + fT ). (68)
2G
Considering the time derivative of this equation, we find
dE ′ 1 3e
= r̃ F1 . (69)
dt 2G h
Substituting this result into equation (66) yields
1 k 5 2k
Th dSh = −dE + [H 2 + 2 ]− 2 ( 2 + Ḣ + 3H 2 )Fe1 dt. (70)
2G a a
The extra term in equation (70) is the main reason for the violation of the first law
of gravitational thermodynamics in the f (T, R)-gravity. Though all matter fields
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 11

see the same horizon and temperature, it seems that some gravitational degrees
of freedom in the f (T, R)-gravity feel a different background metric, horizon and
temperature. Dubovski and Sibiryakov68 have found a similar conclusion while in-
vestigating effects of the spontaneous Lorentz invariance breaking on the black hole
thermodynamics.
In fact, we can rewrite the equation (70) in a different form:
Th dSh + Th dSe = −dE + W dV, (71)
where
( )( [ ]−1 )
V̇ 2k Fe1 k
Th dSe = + Ḣ + 3H 2 1− − Ḣ , (72)
2 a2 H a2
1 2k
W =
( + Ḣ + 3H 2 ). (73)
2 a2
The additional term seen on the left hand side of the equation (71) may be inter-
preted also as the entropy production term developed due to the non-equilibrium
framework in the f (T, R)-gravity. It is easy to see that assuming f (T, R) = R or
f (T, R) = T gives a vanishing additional term which guarantees the traditional first
law of thermodynamics to be achieved.
• The generalized second law
Entropy of the matter inside the apparent horizon is defined by Gibbs’ equation71
TH dSm = dEm + pm dV, (74)
where Em = ρm V describes the internal energy. After taking the time derivative of
the equation (74), one can get
dSm dV
TH = (ρm + pm )( − 3HV ). (75)
dt dt
Now, using equations (44) and (45), we obtain
[ ]− 52
dSm 1 k
TH =− 2
H + 2 H(Ḣ + H 2 )Fe2 , (76)
dt 2G a
where
k 3 3k
Fe2 = 2( 2 − Ḣ)fR + H(f˙R + 4f˙T ) + R̈fRR + f¨R + (2Ḣ + 2 2 − 2 )fT . (77)
a 2r a a
Evolution of the horizon entropy has already been given in equation (66). Hence,
we can write
[ ]− 52
dSh 1 k k
TH = 2
H + 2 ( 2 + Ḣ + 2H 2 )Fe1 . (78)
dt 2G a a
Therefore, adding equations (76) and (78), we find
[ ]− 52 [ ]
dSt 1 k k 2 e 2 e
TH = 2
H + 2 ( 2 + Ḣ + 2H )F1 − H(Ḣ + H )F2 , (79)
dt 2G a a
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

12 Salti et al.

where St = Sm +Sh represents the total entropy. This result shows that the validity
of the generalized second law of thermodynamics, i.e. Ṡt ≥ 0, depends on the
selected f (T, R)-gravity model. As a result, we can write the following requirement
for the second law:
( ak2 + Ḣ + 2H 2 )Fe1 − H(Ḣ + H 2 )Fe2
[ ]5 ≥ 0. (80)
H 2 + ak2 2
In the GR, i.e. f (T, R) = R, equation (79) yields
[ ]− 52 ( )2
dSt H k k
lim TH = 2
H + 2 − Ḣ , (81)
f (T,R)→R dt 2G a a2
which indicates that the second law of thermodynamics is always valid throughout
history of the Universe. Consequently, it can be written that
dSt
lim TH = 8HGπ 2 reh5 (ρm + pm )2 ≥ 0, (82)
f (T,R)→R dt
which is the same as obtained in literature72,73,74,75 . Furthermore, if we have
k
( + Ḣ + 2H 2 )Fe1 = H(Ḣ + H 2 )Fe2 , (83)
a2
then we can get Th dSt = 0 which means that the validity of the generalized sec-
ond law of thermodynamics in the f (T, R)-gravity holds for all the time. Besides,
this significant result corresponds to a reversible adiabatic expansion phase of the
Universe. In addition to this, Th dSt may tend to infinity for a special limit. This
enigmatic case may happen when the expansion rate is very high at a very large
time (t → ∞). All the usable energy contained in the Universe will be transformed
into an unusable form in this interesting case, and the nature of entropy will be
very outlandish due to the cosmological entropy reaching its maximum value. In
the language of physics, this is the heat death scenario of a system. This mysterious
scenario is one of the suggested fates of the universe. All the thermodynamics free
energy will be derogated at this stage and the motion of life cannot sustain any
more.

5. Summary
We were motivated by the recent observations in order to explain the observed ac-
celerated expansion phase of the universe by introducing a new theoretical model.
Focusing on a generalized gravitation theory with an arbitrary coupling between
the curvature and the torsion, we proceeded further. It was assumed here that the
Lagrangian density is described by an arbitrary function of the torsion and the Rie-
mann curvature scalars. Making use of this assumption, we found field equations
describing the f (T, R)-gravity which can be reduced to other gravitational theo-
ries such as the GR, TPG, f (R)-gravity and the f (T ) theory under some limiting
conditions.
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 13

To reach our purpose, we also performed some cosmological investigations. We

obtained that field equations of the f (T, R)-gravity are incompatible with conser-
vation equation of the relativistic neutron stars unless we get F = 0. On the other
hand, defining a suitable f (T, R) function makes our proposal a viable theory of
gravity. Actually, this conclusion underlines a serious problem. Although the family
of modified gravity theories might explain late-time acceleration phase of our uni-
verse, most of these theories cannot be viable due to the unsatisfactory behavior
in the strong field regime. Hence, this work also shows the importance of testing
alternative theories of gravity by relativistic neutron star physics. Moreover, we
discussed also thermodynamics in the f (T, R) framework and observed that an
auxiliary entropy term is produced in the first law of gravitational thermodynamics
as compared both in the GR and TPG theories. Next, we derived a general cir-
cumstance for the generalized second law of thermodynamics, and found that the
validity depends on the f (T, R)-gravity model.
It is obvious that the f (T, R) theory of gravity must be tested by using other
cosmological and astrophysical tests. We plan to proceed our investigation in further
studies.

Acknowledgments
We would like to thank the anonymous reviewer for giving such constructive com-
ments which substantially helped improving the quality of the paper. This academic
work was linguistically supported by the Technology Transfer Office Academic Writ-
ing Center of Mersin University.

References
1. S. Perlmutter et al., Nature 391 (1998) 51.
2. S. Perlmutter et al., Astrophys. J. 517 (1999) 565.
3. A.D. Miller et al., Astrophys. J. Lett. 524 (1999) L1.
4. N.A. Bahcall, J.P. Ostriker, S. Perlmutter and P.J. Steinhardt, Science 284 (1999)
1481.
5. C.L. Bennett et al., Astrophys. J. Suppl. 148 (2003) 1.
6. S.L. Brile, O. Lahav, J.P. Ostriker and P.J. Steinhardt, Science 299 (2003) 1532.
7. D.N. Spergel et al., The Astrophysical Journal Supplement Series 148 (2003) 175.
8. M. Tegmark et al., Phys. Rev. D 69 (2004) 103501.
9. Planck Collaboration (P.A.R. Ade et al.), A&A 594 (2016) A13.
10. A. Sen, Lect. Notes Phys 653 (2004) 141.
11. T. Chiba, T. Okabe and M. Yamaguchi, Phys. Rev. D 62 (2000) 023511.
12. B. Feng, X.L. Wang and X.M. Zhang, Phys. Lett. B 607 (2005) 35.
13. R.R. Caldwell, Phys. Lett. B 545 (2002) 204.
14. E.J. Copeland, M. Sami and S. Tsujikawa, Int. J. Mod. Phys. D 15 (2006) 1753.
15. A. Yu. Kamenshchik, U. Moschella and V. Pasquier, Phys. Lett. B 511 (2001) 265.
16. J. Christensen-Dalsgard, ”Lecture Notes on Stellar Structure and Evolution” (Aarhus
Univ. Press, Aarhus, 2004) p.13.
17. B. Boisseau, G. Esposito-Farese, D. Polarski and A. A. Starobinsky, Phys. Rev. Lett.
85 (2000) 2236.
January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

14 Salti et al.

18. S. Capozziello, Int. J. Mod. Phys. D 11 (2002) 483.

19. V. Faraoni, ”f(R)-gravity: successes and challenges”, XVIII Congresso SIGRAV ”Gen-
eral Relativity and Gravitational Physics” Cosenza, 22-25 September 2008.
20. S. Capozziello, R. Cianci, C. Stornaiolo and S. Vignolo, Phys.Scripta 78 (2008)
065010.
21. M. Li, X.D. Li, S. Wang and X. Zhang, JCAP 06 (2009) 036.
22. Y.-F. Cai, S. Capozziello, M. De Laurentis and E.N. Saridakis, Rept. Prog. Phys. 79
(2016) 106901.
23. S. Nojiri and S.D. Odintsov, Int. J. Geom. Methods Mod. Phys. 4 (2007) 115.
24. J. Frieman, M. Turner, D. Huterer, Annu. Rev. Astron. Astrophys. 46 (2008) 385.
25. A. De Felice, S. Tsujikawa, Living Rev. Relativ. 13 (2010) 3.
26. T.P. Sotiriou and V. Faraoni, Rev. Mod. Phys. 82 (2010) 451.
27. Y.F. Cai, E.N. Saridakis, M.R. Setare and J.Q. Xia, Phys. Rep. 493 (2010) 1.
28. R. Durrer and R. Maartens, ”Dark Energy: Observational and Theoretical Ap-
proaches”, ed. P Ruiz-Lapuente (Cambridge UP, 2010), pp48-91.
29. S. Capozziello, M. De Laurentis and V. Faraoni, The Open Astronomy Journal 3
(2010) 49.
30. S. Tsujikawa, Lect. Notes Phys. 800 (2010) 99.
31. S. Tsujikawa, Dark Energy: Investigation and Modeling. In: S. Matarrese, M. Colpi,
V. Gorini and U. Moschella (eds) Dark Matter and Dark Energy. Astrophysics and
Space Science Library 370 (2011) pp331-402. Springer, Dordrecht
32. M. Li, X.D. Li, S. Wang and Y. Wang, Commun. Theor. Phys. 56 (2011) 525.
33. S. Nojiri and S.D. Odintsov, Phys. Rept. 505 (2011) 59.
34. K. Bamba, S. Capozziello, S. Nojiri and S.D. Odintsov, Astrophys. Space Sci. 342
(2012) 155.
35. R. Myrzakulov, Gen. Relativ. Gravit. 44 (2012) 3059.
36. M. Jamil, D. Momeni and R. Myrzakulov, Chin. Phys. Lett. 29 (2012) 109801.
37. R. Myrzakulov, Eur. Phys. J. C 72 (2012) 2203.
38. G. Calcagni, Phys. Rev. Lett. 104 (2010) 251301.
39. M. Askin, H. Abedi and M. Salti, Rom. J. Phys. 60 (2015) 44.
40. H. Abedi and M. Salti, Gen. Rel. Gravit. 47 (2015) 93.
41. M. Salti, O. Aydogdu and I. Acikgoz, Mod. Phys. Lett. A 31 (2016) 1650185.
42. M. Salti, O. Aydogdu, H. Yanar and F. Binbay, Mod. Phys. Lett. A 32 (2017) 1750183.
43. X.-C. Ao and X.-Z. Li, J. Cosmol. Astropart. Phys. 1110 (2011) 039.
44. L. Fabbri and S. Vignolo, Int. J. Theor. Phys. 51 (2012) 3186.
45. S. Capozziello, M. De Laurentis, L. Fabbri and S. Vignolo, Eur. Phys. J. C 72 (2012)
1908.
46. N. Poplawski, Gen. Rel. Gravit. 46 (2014) 1625.
47. S. Bahamonde, C.G. Bhmer and M. Wright, Phys. Rev. D 92 (2015) 104042.
48. R.M. Wald, Phys. Rev. D 48 (1993) R3427.
49. C. Deliduman and B. Yapiskan, gr-qc:1103.2225 (2011).
50. E. Babichev and D. Langlois, Phys.Rev. D 80 (2009) 121501.
51. T. Kobayashi and K. Maeda, Phys. Rev. D 78 (2008) 064019.
52. C.G. Bohmer, A. Mussa and N. Tamanini, Class. Quant Grav. 28 (2011) 245020.
53. W.De-Hua, H. Jian-Xun and L. Liang-Gang, Commun. Theor. Phys. 45 (2006)
853857.
54. R.G. Cai and S.P. Kim, JHEP 02 (2005) 050.
55. M. Akbar and R.G. Cai, Phys. Rev. D 75 (2007) 084003.
56. J.D. Bekenstein, Phys. Rev. D 7 (1973) 2333.
57. B. Wang, Y. Gong and E. Abdalla, Phys. Rev. D 74 (2006) 083520.
 January 5, 2018 16:58 WSPC/INSTRUCTION FILE fTR

f (T, R) theory of gravity 15

58. Y.G. Gong and A. Wang, Phys. Rev. Lett. 99 (2007) 211301.
59. E. Poisson and W. Israel, Phys. Rev. D 41 (1990) 1796.
60. S.A. Hayward, Phys. Rev. D 53 (1996) 1938.
61. R.G. Cai, L.M. Cao and Y.P. Hu, Class. Quant. Grav. 26 (2009) 55018.
62. J.M. Bardeen, B. Carter and S.W. Hawking, Commun. Math. Phys. 31 (1973) 161.
63. S.W. Hawking, Commun. Math. Phys. 43 (1975) 199.
64. G. Cognola, E. Elizalde, S. Nojiri, S.D. Odintsov and S. Zerbini, JCAP 02 (2005)
010.
65. R.X. Miao, M. Li and Y.G. Miao, JCAP 11 (2011) 033.
66. C.W. Misner and D.H. Sharp, Phys. Rev. 136 (1964) B571.
67. D. Bak and S.-J. Rey, Class. Quant. Grav. 17 (2000) L83.
68. S. Dubovsky and S. Sibiryakov, Phys. Lett. B 638 (2006) 509.
69. S.A. Hawyard, Class. Quant. Gravit. 15 (1998) 3147.
70. S.A. Hawyard, S. Mukohyama and M. Ashworht, Phys. Lett. A 256 (1999) 347.
71. G. Izquierdo and D. Pavon, Phys. Lett. B 639 (2006) 1.
72. K. Karami, JCAP 01 (2010) 015.
73. K. Karami and S. Ghaffari, Phys. Lett. B 685 (2010) 115.
74. K. Karami, S. Ghaffari and M. Soltanzadeh, Class. Quant. Gravit. 27 (2010) 205021.
75. K. Karami, A. Abdolmaleki, N. Sahraei and S. Ghaffari, JHEP 08 (2011) 150.

View publication stats