(Paperhub Ir) 10 1142@s0218271820500042

February 1, 2020 10:18 WSPC/S0218-2718 142-IJMPD 2050004
International Journal of Modern Physics D

Vol. 29, No. 1 (2020) 2050004 (14 pages)
c World Scientific Publishing Company
DOI: 10.1142/S0218271820500042
by UNIVERSITY OF BIRMINGHAM on 04/09/20. Re-use and distribution is strictly not permitted, except for Open Access articles.
Gravitational wave polarization in exponential forms

of f (R) gravity with coupling to matter
Mojtaba Haghshenas∗ and Tahereh Azizi†

Department of Theoretical Physics, Faculty of Basic Sciences,
University of Mazandaran, Babolsar 47416-95447, Iran
∗[email protected]
Int. J. Mod. Phys. D 2020.29. Downloaded from www.worldscientific.com
†[email protected]
Received 16 July 2019

Revised 19 September 2019
Accepted 25 October 2019
Published 10 December 2019
We investigate polarizations of gravitational waves in generalized theories of gravity

whose action contains a nonminimal coupling (NMC) between an exponential function
of f (R) gravity and the matter field. Specifically, we consider two classes of such models.
In the first case, we assume an explicit NMC between the matter Lagrangian density and
curvature while in the second case, we consider the exponential f (R) gravity coupled
to the trace of the stress-energy tensor of a scalar field. Using the Newman–Penrose
quantities, we show that these models can be considered as suitable choices in order to
produce new polarization modes.
Keywords: Gravitational wave; polarization mode; modified gravity; nonminimal cou-

pling; Newman–Penrose formalism.
1. Introduction
After the introduction of general relativity (GR) by Albert Einstein, a dramatic
change has occurred in describing the universe. Even though since the emergence
of this theory there have been a lot of discussions about it being able to describe cos-
mic phenomena, it has had successful predictions like black holes and gravitational
waves (GWs) which nowadays have a lot of reliable evidence of their existence but
it was not able to explain many other phenomena. Some shortcomings of GR were
about the appearance of singularity in the Big Bang theory. Moreover, the gravity
from GR perspective is not renormalizable and thus it is not possible to quantize
it according to the conventional methods within this framework. Therefore, just
a little after the introduction of GR, the efforts to modify this theory began. In
1919, Weyl by providing the Lagrangian density tried to introduce general theory
2050004-1
M. Haghshenas and T. Azizi
to unify gravitational and electromagnetic fields.1 Although he did not succeed, he

opened a new horizon to what we call today the modified theories of GR. It has
been shown that the modified theories are able to describe some cosmological phe-
nomena like acceleration of the expansion of the universe in the late-time without
need to introduce a mysterious dark energy term.2 Up to now, many alternative
theories have been suggested and it seems like we are faced with a bunch of mod-
els and that each has its strengths and weaknesses at the same time. One of the
popular models of modified theories is f (R) gravity where the standard Einstein–
Hilbert action is replaced by a generic function of the Ricci scalar (for a review of
f (R) modified gravity and its cosmological implications, see Refs. 3 and 4.) In this
respect, a model exponential in the curvature has been studied in Ref. 5 and it has
been shown that the exponential model leads to several interesting consequences,
such as an improvement of the solar system features and structure agreements, and
amelioration of fine tuning of the characteristic curvature scale. In another study

by Bamba et al., it was found that the selection of the f (R) function as an exponen-
tial form reproduces the physics of ΛCDM model and they also proposed a way to
describe the early time acceleration of the universe by applying exponential gravity
to inflationary cosmology.6 It should be mentioned that in addition to Ref. 6, some
other models of f (R) gravity can naturally combine R + R2 model inflation and the
late-time acceleration using exponential function too.7–9 The f (R) gravity model
can be further generalized by considering a nonminimal coupling (NMC) between
the matter Lagrangian and the curvature.10 This scenario has a new and interest-
ing phenomenology especially, concerning the energy exchange between the matter
fields and the curvature. Another approach to extend the GR has been considered
in Ref. 11 where the gravitational Lagrangian is given by an arbitrary function of
the Ricci scalar R and of the stress-energy tensor T , dubbed as the f (R, T ) modi-
fied gravity. In this approach, the matter is coupled to the curvature scalar via the
trace of its stress-energy tensor which causes nongeodesic equation of motion for
massive particles.12
After the detection of GWs by the LIGO Scientific and Virgo Collaborations in
recent years,13 a new challenge lies in the modified gravitational models. In fact, GW
data and especially, polarization modes, provide a new tool to study gravitational
physics. Future space-borne detectors will offer access to an unprecedented signal
sensitivity, thus it is worthwhile to explore GWs in alternative theories of gravity.
In 1973, an idea was raised by Eardley et al., which had a great ability to study
the properties of GWs.14 Their idea was based on analyzing Riemann tensor and all
physically relevant components of that because these components cause the relative
acceleration between test particles. They used the fact that in linearized regime, the
Riemann tensor can be split into six algebraically independent components. These
components correspond to six polarization modes derived from the little group (E2)
of the Lorentz group. With the assumption that GW is propagating along z-axis, the
six components can be classified into six independent polarizations modes. Two of
them are tensor modes, two are vector modes x, y and two scalar modes b, l and the
2050004-2
last two modes are breathing and longitudinal modes, respectively. Their research
showed that there are six possible wave polarization modes for general theories.14
Using Newman–Penrose (NP) formalism15 to characterize GW in quadratic gravity
and in a particular class of Lagrangians, Alves et al. showed that while in GR there
are just two tensor modes, in the context of the modified gravity theories, more
polarization modes can be expected.16,17 The Palatini case has been analyzed by
Näf et al. which leads to different results. Considering terms up to linear order in
Ricci scalar and calculating the nonnull components of Riemann tensor, they found
that this model does not introduce any additional polarization modes but causes
tiny modifications in the wave amplitude.18 Polarization of GW has been extensively
studied in the context of other generalized theories of gravity approach too.19–27
In this paper, we study GWs in the presence of an NMC between an exponen-
tial form of f (R) gravity and the matter field. We choose two particular theories,
namely, a nonminimal curvature-matter coupling theory and f (R, T φ ) modified

gravity where the exponential f (R) term is coupled to the trace of the stress-energy
tensor of a self-interacting scalar field. Using the NP formalism, we explore extra
modes of GWs in these two modified gravity theories. We also discuss about the
interferometers response to the longitudinal mode of GW polarization that arises
from the nonminimal curvature-matter coupling scenario. The outline of this paper
is as follows. In Sec. 2, we consider the exponential f (R) term with a NMC to the
matter Lagrangian and we obtain the GW polarization content of this scenario. In
Sec. 3, we assume that the exponential gravity is coupled to the trace of the stress-
energy tensor of a scalar field. Using the NP quantities, we show that this model
admits extra polarization modes besides GR. In Sec. 4, we study the interferome-
ters response to the longitudinal modes that have appeared in the first model. A
summary of our results is given in Sec. 5. In Appendix A, we briefly overview the
NP formalism for classifying the polarizations of null GWs.
2. A Modified Gravity with NMC Between Curvature and Matter

We start with the action which contains an NMC between curvature and matter in
the following form10 :

√ 1
S = d4 x −g f1 (R) + f2 (R)Lm , (1)
2
where g is the metric determinant, Lm is the Lagrangian density of matter field and
f1 (R), f2 (R) are generic functions of the Ricci scalar. The usual f (R) theories of
gravity are recovered by setting f2 (R) = 1 and one obtains the standard Einstein–
Hilbert action by taking in addition the linear function f1 (R) = R. In order to
obtain the field equations, we need to vary the action (1) with respect to the metric
gμν that yields the following modified field equations:
1
(f1R + 2f2R Lm )Rμν − gμν f1 − (∇μ ∇ν − gμν )(f1R + 2f2R Lm ) = f2 Tμν , (2)
2
2050004-3
where = ∇μ ∇μ , ∇μ is covariant derivative for gμν and fiR ≡ dfi /dR. By taking
the trace of the field equations (2), we get
3(f1R + 2f2R Lm ) = 2f1 − (f1R + 2f2R Lm )R + f2 T. (3)
Obviously, T is the trace of the stress-energy tensor, which can be deduced from
matter Lagrangian density as follows:

√
−2 δ( −gLm )
Tμν = √ . (4)
−g δ(g μν )
Using the Bianchi identities, one can derive the nonconservation (covariant) of the
stress-energy tensor11
f2R μν
∇μ T μν = (g Lm − T μν )∇μ R. (5)
f2
The above relation implies that in the context of the modified gravity with NMC
to matter, an extra force acts on a test particle, hence the trajectory of a point
like particle deviates from the geodesical motion.11 We are interested to investigate
the GWs in the absence of the matter field and actually in the vacuum. Therefore,
for the convenience and also in the absence of matter, the Lagrangian density can
be defined as a cosmological constant, Lm = −Λ. As a result, the trace of the
stress-energy tensor is obtained as T = −4Λ.
2.1. Polarization modes for exponential form

The polarization modes of GWs in the context of nonminimal curvature-matter
coupling are investigated in a recent work by Bertolami et al. by choosing a poly-
nomial form for the f (R) function and they showed that the extra polarization
modes are available in this context.26 In this section, we consider the exponential
form for f (R) function and explore the polarization modes in this case. In partic-
ular, a suitable choice (for more details, see Ref. 12) is to consider f1 (R) = R and
f2 (R) = exp( 2kR2 Λ ) where k is constant, so the gravitational field equations (3) take
the following form:

3 R 1 R R
exp = R + R exp − 4Λ exp . (6)
k2 2k 2 Λ k2 2k 2 Λ 2k 2 Λ
Because we are working in low curvature (small field) regime R 1, we can ignore
the first two terms in the right-hand side of the above equation in comparison with
the last term, hence Eq. (6) reduces to the following form:

R 4 2 R
exp = − Λk exp . (7)
2k 2 Λ 3 2k 2 Λ
2050004-4
Now by assuming that the wave is propagating in the direction of the z-axis and
d2
applying the double derivative ≡ dz 2 and, Eq. (7) becomes
2 2
1 d2 R 1 dR 4
+ − Λk 2 = 0. (8)
2k 2 Λ dz 2 2k 2 Λ dz 3
The above differential equation is a second-order nonlinear, inhomogeneous type

and has the following general form:
1 2
y + y 2 + a = 0, (9)
a 3
whose solution is

2
R = a log cos a(z − 3ac1 ) + c2 , (10)
where c1 , c2 are constants of integration and a = 2k 2 Λ. Because our system is

invariant under the Lorentz transformations, we can obtain the time-dependent
solution from static form by applying Lorentz transformation as

2 z − vt
R(z, t) = a log cos a √ − 3ac1 + c2 . (11)
3 1 − v2
Using the field equations (2) and replacing the values of functions fi (R) and fiR ,
we get the following relationship for the components of the Ricci tensor in terms of
the Ricci scalar as

1 1 R R
gμν R + (∇μ ∇ν − gμν ) 1 − 2 exp − Λ exp gμν
2 k a a
Rμν = . (12)
1 R
1 − 2 exp
k a
Again, remember that we are working in the linearized form and low curvature
regime. Therefore, the Ricci tensor (12) can be reduced to

1 R R
(∇μ ∇ν − gμν ) 1 − 2 exp − Λ exp gμν
k a a
Rμν = . (13)
1 R
1 − 2 exp
k a
So, the nonvanishing components of the Ricci tensor are computed as follows:

R 7 − 3v 2
Λ exp
a 3(1 − v 2 )
Rtt = , (14)
1 R
1 − 2 exp
k a
2050004-5
2
R 7v − 11
Λ exp
a 3(1 − v 2 )
Rzz = , (15)
1 R
1 − 2 exp
k a

4Λ R −v
exp
3 a 1 − v2
Rtz = , (16)
1 R
1 − 2 exp
k a

−7Λ R
exp
3 a
Rxx = Ryy = . (17)
1 R
1 − 2 exp
k a
Finally, in order to obtain the NP quantities, by substituting all nonzero compo-
nents of Ricci tensor in Eqs. (A.8)–(A.11) of Appendix A, we get
1
Ψ2 = R, (18)
12
Ψ3 = 0, (19)
1 1
Φ22 = − Rμν lμ lν = − (Rtt lt lt + 2Rtz lt lz + Rzz lz lz ). (20)
2 2
Therefore, from Eqs. (A.7), (A.11) and (A.14), we obtain

R
Λ exp
a 1 − 3v 2
Φ22 = . (21)
1 R 3(1 − v 2 )
1 − 2 exp
k a
Since we have no more constrains on the components of the Riemann tensor, Ψ4 = 0
(see Ref. 17 for details). Consequently, the nonzero NP variables are Ψ2 , Ψ4 and
Φ22 , which correspond to four possible polarization modes. Two of them, cross
and plus, are related to Ψ4 tensor modes, while Ψ2 and Φ22 are corresponding
to longitudinal and breathing scalar modes, respectively. Notice that the NMC
theories are equivalent with a scalar–tensor theory with two scalar fields, so the new
polarization modes should be excited due to the scalar fields which are presented in
these theories. The detection of the breathing and longitudinal modes in addition
to the usual × and + modes means that the graviton is mediated by massless spin
2 and massive spin 0 fields.25 It is worthwhile to point out that we could have
both functions f1 (R) and f2 (R) in the exponential forms like f1 (R) = exp( R α ) and
f2 (R) = exp( R
β ). In this case, if we follow the same procedure as mentioned above
and assume special case α = β, we will obtain the following expression for the Ricci
scalar as

√ z − vt c1
R(z, t) = α ln cosh b √ + + c2 , (22)
1 − v2 α
2050004-6
where b = 13 (2α + 1) and c1 , c2 are integration constants. However, we will obtain

similar NP quantities for this case and therefore, like the previous case, the same
number of polarization modes will be deduced.
3. GWs in f (R, T φ ) Generalized Gravity

We start with a particular action in this case defined as follows22 :

√
S = d4 x −g[f (R, T φ) + L(φ, ∇μ φ) + Lm ], (23)
where T is the trace of the stress-energy tensor and L(φ, ∇μ φ) is the Lagrangian
density of the scalar field that is identified as follows:
1
L(φ, ∇μ φ) = ∇α φ∇α φ − V (φ), (24)
2
where V (φ) is the scalar field self-interacting potential. The stress-energy tensor for
the scalar field is given by
√
2 δ[ −gL(φ, ∇μ φ)] 1
φ
Tμν =√ = ∇μ φ∇ν φ − gμν ∇α φ∇α φ + gμν V (φ). (25)
−g δg μν 2
Taking the trace of this equation leads to
T φ = −∇α φ∇α φ + 4V (φ). (26)
Now, we take f (R, T φ) = f1 (R) + λf2 (T φ ), where f2 (T φ ) is a generic function of
the trace of the stress-energy tensor of the scalar field. Varying the action (23) with
respect to the metric yields the vacuum field equations in this case as
1
f1R Rμν − f1 gμν + (gμν − ∇μ ∇ν )f1R
2
1 φ
= − [Tμν − gμν f2 (T φ ) − 2f2T ∇μ φ∇ν φ], (27)
2
where f2T ≡ df2
dT . Taking the trace of the gravitational field equation (27) leads to
a dynamical equation for the Ricci scalar as follows:
1
f1R R − 2f1 + 3f1R = − [T φ − 4f2 − 2f2T ∇α φ∇ν φ]. (28)
2
In Sec. 3.1, we explore the polarization content of the model.
3.1. Polarization modes

Some particular models of f (R, T φ) theories have been studied by Alves et al. by
considering several different types of functions in Ref. 22 and they have shown that
the properties of GWs and consequently, the number of polarization modes in such
models depend explicitly on the particular choice for f (R, T φ ) functional forms. In
this respect, they have considered polynomial form for f1 (R) and a linear depen-
dence of f2 (T φ ) on the trace of the stress-energy tensor of the scalar field and four
2050004-7
polarization modes have been obtained for this case. In this section, we choose an
exponential form for f1 (R) and let f2 (T φ ) to be the linear form and study the polar-
ization states of the scenario. Assuming a general form like f1 (R) = exp( Rq ) where
q is an arbitrary constant, f2 (T φ ) = 2λT φ and using Eqs. (26) and (27), we get

3 R R 1 R
exp − 2 exp + R exp = 2V0 (8λ − 1), (29)

q q q q q
where we have considered only the linear terms in φ and V0 is the minimum value of
the potential Vφ . Since we are working in weak field regime which results in R 1,
we can ignore R exp( Rq ) compared to the term exp( Rq ) which yields

R 2q R 2q
exp − exp − V0 (8λ − 1) = 0. (30)
q 3 q 3
By acting the d’Alembert operator and assuming that the GW is propagating in

the direction of the z-axis, we find
2
d2 R R 1 dR R 2 2 R 2
2
exp + exp − q exp − q 2 α = 0, (31)
dz q q dz q 3 q 3
where for convenience we have put (8λ − 1)V0 = α. The differential equation (31)
can be recast in the following form:

2 2 y 1 2
q α exp − − y − y 2 + q 2 = 0, (32)
3 q q 3
to which the solution is
⎛ ⎞
2 4 2 2 2
⎜ (4α q − 6c1 q) exp q(z + c2 ) + exp − q(z + c2 ) − 4aq ⎟
⎜ 3 3 ⎟
R(z) = q ln⎜
⎜
⎟,
⎟
⎝ 4q 2 ⎠
(33)
where c1 √
and c2 are constants of integration. Again, for more convenience, we have
2
defined e 3 qc2 = ξ. For time-dependent solution, by considering a Lorentz trans-
formation, we obtain
⎛ ⎞

⎜ (4α2 q 4 − 6c1 q)ξ exp 2 z − vt ⎟
⎜ q √ ⎟
⎜ 3 1 − v2 ⎟
⎜ ⎟
⎜ ⎟
⎜ 2 z − vt ⎟
R(z, t) = q ln⎜ −1
+ ξ exp − q √ 2 ⎟.
− 4αq ⎟ (34)
⎜ 1 − v2
⎜ 3 ⎟
⎜ ⎟
⎜ 4q 2 ⎟
⎝ ⎠
2050004-8
Now, by putting these results in the field equations (27), the nonvanishing compo-
nents of the Ricci tensor are as follows:

R 3v 2 − 1
Rtt = q 1 + α exp − , (35)
q 6(1 − v 2 )
2
R v +3
Rzz = q 1 + α exp −
, (36)
q 6(1 − v 2 )

R 2v
Rtz = q 1 + α exp − , (37)
q 3(1 − v 2 )

q R
Rxx = Ryy = − 1 + α exp − . (38)
6 q
Thus, the corresponding NP parameters are calculated as
1
Ψ2 = R, (39)
12
Ψ3 = 0, (40)
2
q R 2v − 4v + 1
Φ22 = − 1 + α exp − , (41)
12 q 1 − v2
Ψ4 = 0. (42)
As a result, it is clear that there are four polarization modes. Like the previous case,
we expect two tensor modes corresponding to Ψ4 = 0, a longitudinal scalar mode
with respect to Ψ2 = 0 and a breathing scalar mode arising from Φ22 = 0 to be
produced. The presence of the extra modes is due to the extra degrees of freedom
presented in this theory.
4. Longitudinal Force and the Interferometer’s Response

In this section, we investigate the response function for the interferometer to the
longitudinal mode that is arisen in the first generalized gravity model studied in
this paper. Presence of the longitudinal mode in GWs polarization is corresponding
to a force component in the wave propagating direction.19 In order to obtain the
longitudinal force which affects on the test masses, we can use the equation of
geodesic deviation
d2 xi
ẍi = = −Rtitj xj , (43)
dt2
where Rtitj is the so-called electric part of the Riemann tensor and can be calculated
directly by using the NP quantities25
⎛ ⎞
1 1
− (Ψ4 + Φ22 ) Ψ4 −2Ψ3
⎜ 2 2 ⎟
⎜ ⎟
⎜ 1 1 ⎟
Rtitj = ⎜ Ψ4 (Ψ4 − Φ22 ) 2Ψ3 ⎟ , (44)
⎜ ⎟
⎝ 2 2 ⎠
−2Ψ3 2Ψ3 −6Ψ2
2050004-9
where and denote to the real and imaginary parts of the complex functions,
relatively. In the case of theories with NMC between curvature and matter, since
Ψ3 = 0 (Eq. (19)), then one can conclude that the electric part of the Riemann
tensor is given by
⎛ ⎞
1 1
− (Ψ + Φ ) Ψ 0
⎜ 2 4 22 4 ⎟
⎜ 2 ⎟
⎜ ⎟
⎜
Rtitj = ⎜ 1 1 ⎟
Ψ4 (Ψ4 − Φ22 ) 0 ⎟. (45)
⎜ 2 2 ⎟
⎝ ⎠
0 0 −6Ψ2
By using geodesic deviation equation (43), the longitudinal force is obtained as
z̈ = 6Ψ2 z. (46)
Notice that this force component is the key tool for the longitudinal polarization
mode detection. One can assume that the test mass is the same as an interferom-
eter, consisting of the beam-splitter and mirror which have been fixed at a certain
distance. Then by applying this force component and calculating the variation of
the proper distance of beam-splitter and mirror in the presence and absence of the
GW as a function of time and finally, with a Fourier transformation, the longitudi-
nal response function with respect to the frequency can be obtained as follows (for
more details, see Refs. 19 and 25):
1
Υl (ω) ≡ (1 − vg2 ) exp[iωL(1 + vg )] + (exp[2iωL](vg + 1)3
2ωL(vg2 − 1)2
× (−2i + ωL(vg − 1)) + 2 exp[iωL(1 + vg )](6ivg + 2ivg3
− ωL + ωLvg4 ) + (vg + 1)3 [−2i + ωL(vg + 1)]), (47)

where L is the unperturbed distance between beam-splitter and mirror and ω =
2πf is the angular frequency of the wave. vg is the group velocity of GW that
in the framework of the NMC between curvature and matter modified gravity is
defined as26
∂ω 1
vg ≡ = . (48)
∂k f1 − f2 Λ
1+ 2
k (f1R − f2R Λ)
Actually, the response function (47) is independent of the particular form of the
modified gravity model and just depends on the group velocity that is produced in
each model. For the case of the modified gravity theories with NMC between curva-
ture and matter, the group velocity of the GW is estimated as vg ≈ 1 − (7 × 10−23),
see for more detail in Ref. 26. In Figs. 1 and 2, the absolute values of response
function (47) with respect to the frequency have been depicted for Ligo interfer-
ometer (L = 4000 m) and Virgo interferometer (L = 3000 m), respectively. As the
2050004-10

Fig. 1. The absolute values of the longitudinal response functions |Υl | as a function of frequency
for Ligo L = 4000 m to a GW arising from the theory with NMC between curvature and matter
and propagating with a speed of vg ≈ 1 − (7 · 10 × −23).
Fig. 2. The absolute values of the longitudinal response functions |Υl | as a function of frequency
for Virgo L = 3000 m to a GW arising from the theory with NMC between curvature and matter
and propagating with a speed of vg ≈ 1 − (7 × 10−23 ).
2050004-11
figures show, it might be difficult to detect the longitudinal mode in the frequency
band of the ground based detectors like Virgo and Ligo interferometers. We should
mention that even in the frequency band of the space-born observatory like the
Laser Interferometer Space Antenna (LISA) (10−3 Hz), the values of longitudinal
response function are too small. On the other hand, in principle, one can detect
the longitudinal polarization mode using the pulsar time arrays with the frequency
band at about 10−7 Hz.
5. Conclusion
Detection of GWs in recent years has opened new windows in the study of gravita-
tional physics, especially the generalized theories of gravity. In this work, we have
investigated the polarization content of GWs in two alternative models: Modified
gravity with NMC to the matter field and f (R) gravity minimally coupled to the
trace of the stress-energy tensor of a self-interacting scalar field. For each of the two
models, we have considered an exponential form for the f (R) function. By using the
field equation of these models and performing the NP analysis, we have found out
that in each case, there are two extra polarization modes in addition to the usual
tensor modes that exist in GR. These extra modes correspond to two scalar modes,
namely the breathing and Longitudinal scalar modes duo to additional degrees of
freedom contained in each theories. Note that the detection of polarization modes
can be used to understand the nature of gravity for example, the existence of breath-
ing and longitudinal modes in addition to the usual × and + modes means that
the graviton is mediated by massless spin 2 and massive spin 0 fields.25 We have
also studied the response function for the Ligo and Virgo interferometers to the
longitudinal mode that arises in the first case (NMC between curvature and mat-
ter field) with respect to the frequency. It should be mentioned that according to
the figures, it is a difficult task to detect longitudinal polarization mode by using
interferometers. The reason is that, even in the frequency band of the space-born
observatory like LISA, the values of longitudinal response function are too small.
However, the pulsar time arrays (with the frequency band about 10−7 Hz) will be
able to detect the longitudinal polarization mode.
Acknowledgment
We would like to acknowledge Prof. Kourosh Nozari for his invaluable remarks.
Appendix A
In 1962, Newman and Penrose suggested a formalism to study the outgoing grav-
itational radiation. We consider Eardley et al. approach (presented in 1973) to
investigate GWs and their polarization modes throughout this paper (see, for more
details, Ref. 14). Newman and Penrose introduced the following relations in order
2050004-12
to relate complex null-tetrads (k, l, m, m̃) with Cartesian tetrads (t̂, x̂, ŷ, ẑ) in any
spacetime point:
1
k = √ (t̂ + ẑ), (A.1)
2
1
l = √ (t̂ − ẑ),
(A.2)
2
1
m = √ (x̂ + iŷ), (A.3)
2
1
m̃ = √ (x̂ − iŷ). (A.4)
2
It can be shown easily that the null-tetrads satisfy the following relationships:
−k · l = m · m̃ = 1, (A.5)
k · m = k · m̃ = l · m = l · m̃ = 0. (A.6)
Moreever, according to Alves et al. suggestion,16 any tensor in Cartesian basis can
be converted in the null-tetrads basis via the following equations:
Tabc... = Tμνλ... aμ bν cλ . . . , (A.7)
where the sets (a, b, c, . . .) and (μ, ν, λ, . . .) run over the sets (k, l, m, m̃) and
(t, x, y, z), respectively. Newman and Penrose also showed that the irreducible parts
of the Riemann tensor (the so-called NP parameters) can be defined by using the
ten Ψ’s called Weyl tensor components, nine Φ’s known as the traceless Ricci tensor
components and a curvature scalar term Λ, which are all algebraically independent.
Fortunately, there are some explicit relationships between these NP quantities and
the Riemann tensor components in the null-tetrad basis and each one corresponds
to one or two independent polarization modes as follows:
Ψ4 = −Rlm̃lm̃ , (A.8)
1
Ψ3 = − Rlklm̃ , (A.9)
2
1
Ψ2 = − Rlklk , (A.10)
6
Φ22 = −Rlmlm̃ , (A.11)
where Ψ3 and Ψ4 are complex, thus each one corresponds to two independent
polarization modes, vector-x, y modes and + and × tensor modes, respectively. In
addition, there are two scalar modes, Ψ2 and Φ22 , which represent the longitudinal
scalar mode, and a breathing scalar mode, respectively. Also, we have some use-
ful expressions in order to handle NP formalism for the Ricci tensor and for the
curvature scalar as follows:
Rlk = Rlklk , (A.12)
2050004-13
R = −2Rlk = −2Rlklk , (A.13)

Rll = 2Rlmlm̃ , (A.14)
Rlm = Rlklm , (A.15)
Rlm̃ = Rlklm̃ . (A.16)
References
1. H. Weyl, Space–Time–Matter, 4th edn. (Dover, New York, 1922).
2. S. M. Carroll, V. Duvvuri, M. Trodden and M. S. Turner, Phys. Rev. D 70 (2004)
043528.
3. H. A. Buchdahl, Mon. Not. R. Astron. Soc. 150 (1970) 1; R. Kerner, Gen. Relativ.
Gravit. 14 (1982) 453; J. P. Duruisseau, R. Kerner and P. Eysseric, Gen. Relativ.
Gravit. 15 (1983) 797; A. A. Starobinsky, Phys. Lett. B 91 (1980) 99.

4. T. P. Sotiriou and V. Faraoni, Rev. Mod. Phys. 82 (2010) 451; A. De Felice and S.
Tsujikawa, Living Rev. Rel. 13 (2010) 3.
5. E. V. Linder, Phys. Rev. D 80 (2009) 123528.
6. K. Bamba, C.-C. Geng and C.-C. Lee, J. Cosmol. Astropart. Phys. 08 (2010) 021.
7. B. Appleby and A. A. Starobinsky, J. Cosmol. Astropart. Phys. 1006 (2010) 005.
8. H. Motohashi and A. Nishizawa, Phys. Rev. D 86 (2012) 083514.
9. H. Motohashi and A. Nishizawa, Phys. Rev. D 89 (2014) 063541.
10. O. Bertolami, C. G. Boehmer, T. Harko and F. S. N. Lobo, Phys. Rev. D 75 (2007)
104016.
11. O. Bertolami, F. S. N. Lobo and J. Páramos, Phys. Rev. D 78 (2008) 064036.
12. T. Harko et al., Phys. Rev. D 84 (2011) 024020.
13. LIGO Scientific and Virgo Collab. (B. P. Abbott et al.), Phys. Rev. Lett. 116 (2016)
061102.
14. D. M. Eardley, D. L. Lee and A. P. Lightman, Phys. Rev. D 8 (1973) 3308.
15. E. Newman and R. Penrose, J. Math. Phys. 3 (1962) 566.
16. M. E. S. Alves et al., Phys. Lett. B 679 (2009) 401.
17. M. Alves, O. Miranda and J. de Araujo, Class. Quantum Grav. 27 (2010) 145010.
18. J. Näf, P. Jetzer and M. Sereno, Phys. Rev. D 79 (2009) 024014.
19. C. Corda, Int. J. Mod. Phys. A 23 (2008) 1521.
20. S. Capozziello, C. Corda and M. F. De Laurentis, Phys. Lett. 669 (2008) 228.
21. H. R. Kausar, L. Philippoz and P. Jetzer, Phys. Rev. D 93 (2016) 124071.
22. M. E. S. Alves, P. H. R. S. Moraes, J. C. N. de Araujo and M. Malheiro, Phys. Rev.
D 94 (2016) 024032.
23. D. Liang, Y. Gong, S. Hou and Y. Liu, Phys. Rev. D 95 (2017) 104034.
24. M. Sharif and A. Siddiqa, Astrophys. Space Sci. 362 (2017) 226.
25. S. Hou, Y. Gong and Y. Liu, Eur. Phys. J. C 78 (2018) 378.
26. O. Bertolami, C. Gomes and F. S. N. Lobo, Eur. Phys. J. C 78 (2018) 303.
27. G. Farrugia, J. Levi Said, V. Gakis and E. N. Saridakis, Phys. Rev. D 97 (2018)
124064.
2050004-14

(Paperhub Ir) 10 1142@s0218271820500042

Uploaded by

Copyright:

Available Formats

(Paperhub Ir) 10 1142@s0218271820500042

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Paperhub Ir) 10 1142@s0218271820500042

Uploaded by

Copyright:

Available Formats

February 1, 2020 10:18 WSPC/S0218-2718 142-IJMPD 2050004

International Journal of Modern Physics D

Gravitational wave polarization in exponential forms

Mojtaba Haghshenas∗ and Tahereh Azizi†

Received 16 July 2019

We investigate polarizations of gravitational waves in generalized theories of gravity

Keywords: Gravitational wave; polarization mode; modiﬁed gravity; nonminimal cou-

M. Haghshenas and T. Azizi

to unify gravitational and electromagnetic ﬁelds.1 Although he did not succeed, he

amelioration of ﬁne tuning of the characteristic curvature scale. In another study

Gravitational wave polarization in exponential forms

namely, a nonminimal curvature-matter coupling theory and f (R, T φ ) modiﬁed

2. A Modified Gravity with NMC Between Curvature and Matter

M. Haghshenas and T. Azizi

3(f1R + 2f2R Lm ) = 2f1 − (f1R + 2f2R Lm )R + f2 T. (3)

matter Lagrangian density as follows:

2.1. Polarization modes for exponential form

Gravitational wave polarization in exponential forms

The above diﬀerential equation is a second-order nonlinear, inhomogeneous type

where c1 , c2 are constants of integration and a = 2k 2 Λ. Because our system is

M. Haghshenas and T. Azizi

Gravitational wave polarization in exponential forms

where b = 13 (2α + 1) and c1 , c2 are integration constants. However, we will obtain

3. GWs in f (R, T φ ) Generalized Gravity

We start with a particular action in this case deﬁned as follows22 :

3.1. Polarization modes

M. Haghshenas and T. Azizi

exp − 2 exp + R exp = 2V0 (8λ − 1), (29)

By acting the d’Alembert operator and assuming that the GW is propagating in

Gravitational wave polarization in exponential forms

4. Longitudinal Force and the Interferometer’s Response

M. Haghshenas and T. Azizi

× (−2i + ωL(vg − 1)) + 2 exp[iωL(1 + vg )](6ivg + 2ivg3

− ωL + ωLvg4 ) + (vg + 1)3 [−2i + ωL(vg + 1)]), (47)

Gravitational wave polarization in exponential forms

M. Haghshenas and T. Azizi

Gravitational wave polarization in exponential forms

M. Haghshenas and T. Azizi

R = −2Rlk = −2Rlklk , (A.13)

Gravit. 15 (1983) 797; A. A. Starobinsky, Phys. Lett. B 91 (1980) 99.

You might also like