CHIN.PHYS.LETT. Vol. 25, No.

12 (2008) 4497

Phantom Energy with Variable G and Λ

Arbab I. Arbab∗∗
1
Department of Physics, Faculty of Science, University of Khartoum, P.O. Box 321, Khartoum 11115, Sudan
Department of Physics and Applied Mathematics, Faculty of Applied Sciences and Computer, Omdurman Ahlia
University, P.O. Box 786, Omdurman, Sudan

(Received 27 March 2008)

We investigate a cosmological model of a phantom energy with a variable cosmological constant (Λ) depending on
the energy density (ρ) as Λ ∝ ρ−α , α = const and a variable gravitational constant G. The model requires α < 0
and a negative gravitational constant. The cosmological constant evolves with time as Λ ∝ t−2 . For ω > −1 and
α < −1 the cosmological constant Λ < 0, G > 0 and ρ decrease with cosmic expansion. For ordinary energy (or
dark energy), i.e. ω > −1, we have −1 < α < 0 and β > 0 so that G > 0 increases with time and ρ decreases
with time. Cosmic acceleration with dust particles is granted, provided −2/3 < α < 0 and Λ > 0.

PACS: 98. 80. Cq, 95. 36. +x, 98. 80. Es

Cosmologists have wondered whether our present
universe will eventually re-collapse and end with a
Big Crunch, or expand indefinitely and eventually be-
comes cold and empty. However, recent evidence from
supernovae type I ushers into a flat universe, possi-
bly with a cosmological constant or some other sort
of negative-pressure dark energy, has suggested that
our fate is accelerating.[1,2] However, the data may
actually be pointing toward an astonishingly different
cosmic end game. Caldwell et al.[3] explored the con-
sequences that follow if the dark energy is a phantom
energy, i.e. the sum of the pressure and energy den-
sity is negative. The positive phantom-energy density
becomes infinite in a finite time, overcoming all other
forms of matter, which will rapidly brings the epoch of
cosmic structure to a halt.[4] The phantom energy rips
apart every bound matter before the Universe ends
into a Big-Rip.
However, the phantom energy scenario does violate
the the strong energy condition (SEC) ρ + 3p > 0, a
principle that keeps energies positive and imposes en-
ergy conservation on a global scale. It is the strong
energy condition that is helpful to rule out wormholes,
warp drives, and time machines. Dark energy requires
an equation of state p + 3ρ < 0. The violation of the
null energy condition (NEC) p+ρ > 0 results in energy
flows faster than the speed of light. A phantom be-
haviour is predicted by several scenarios, e.g., kineti-
cally driven models[5] and some versions of braneworld
cosmologies.[6]
Another possibility for dark energy is an energy of
a scalar field known as quintessence having an equa-
tion of state such that −1 < ω < −1/3.[7−11] Assum-
ing that the quintessence field coupled minimally to
gravity, one can write the Lagrangian as
1 2
L= φ̇ + V (φ), (1)
2
∗∗ Email:
aiarbab@uofk.edu
c 2008 Chinese Physical Society and IOP Publishing Ltd
with energy density and pressure given by
1 2 1 2
ρ= φ̇ + V (φ), p= φ̇ − V (φ). (2)
2 2
where a dot presents a derivative with respect to time.
In the so-called tracker models the scalar field density
(and its equation of state) remains close to that of the
dominant background matter during most of the cos-
mological evolution. The equation of state is given by
p φ̇2 /2 − V (φ)
= . (3)
ρ φ̇2 /2 + V (φ)
A comprehensive study of quintessence is investigated
by Ratra and Peebles.[12] However, A tracker poten-
tial of the form V (φ) ∝ φ−n , n = const is considered
recently by Sahni.[13]
Moreover, one may include the possibility of an
equation of state p = −ρ. This is attributed to exis-
tence of vacuum energy or the cosmological constant.
At the present time it is difficult to tell which form of
energy our universe consists of.
In this Letter, we investigate the evolution of dark
energy and phantom energy arising from the intro-
duction of a cosmological constant that evolves as
Λ = 3β/ρα , where α, β = const. With this assump-
tion, a phantom energy arises whenever p + ρ < 0
and α > 0. However, the gravitational constant be-
comes negative. In the present model, the dark en-
ergy models do no necessarily require the condition
p > −1/3ρ. Cosmic acceleration is generated for
α > 0, −1 < α < 0 and α < −1. Phantom en-
ergy with variable G has been recently considered by
Stefancic.[14] We have shown that during the evolution
of the domain-walls the cosmological constant flips its
sign. We have seen that the whole evolution of the
universe is characterized by the two constants α and
β.
4498 Arbab I. Arbab Vol. 25
h 3  β 1/2 i
The Friedmann–Lematre–Robertson–Walker met- where F = − α(1 + ω) . Using
ric 2 (1 + α)
Eq. (15), Eq. (11) becomes
ds2 = dt2 − a2 (t) dr2 + r2 (dθ2 + sin2 θ dφ2 ) ,


−3αβ
G= F −2(1+α)/α t−2(1+α)/α , α 6= −1, 0.
for a flat universe yields 8π (1 + α)
(16)
3(ȧ/a)2 = 8π Gρ + Λ, (4) Using Eq. (15), Eq. (10) reads
ρ = F 2/α t2/α , α 6= 0, −1, (17)

3(ä/a) = −4π G(ρ + 3p) + Λ, (5) which shows that the energy density of the phantom
increases with expansion, if α > 0. Using Eq. (17),
and the energy conservation equation reads Eq. (9) yields
3β
ρ̇ + 3(ȧ/a)(ρ + p) = 0. (6) Λ = 2 t−2 . (18)
F
This form is found to emerge from many of the vacuum
The pressure p and energy density ρ of an ideal fluid
decaying models.[16,17,19−21] Since we believe that the
are related by the equation of state
vacuum decays with time then β > 0.
p = ω ρ, ω = const. (7) For an expanding universe, we require that −α(1+
ω) > 0 as is evident from Eq. (15). If α > 0, one must
We now allow Λ and G to vary with time, i.e. Λ = have the relation 1 + ω < 0. This is the familiar condi-
Λ(t) and G = G(t), so that the energy conservation, tion for the existence of phantom energy. This relation
Eq. (6), entitles that[15−18] implies that p < −ρ. In this case, the energy density
grows with time, as it is evident from Eq. (17). This is
8π Ġρ + Λ̇ = 0. (8) the condition for phantom energy. It is very interest-
ing to see that such a cosmological constant variation
We consider here the ansatz leads to phantom energy solution. We, however, no-
tice that an increasing energy density requires α > 0.
Λ = 3β/ρα , β, α = const. (9)
We see from Eq. (13) that since H 2 ≥ 0 the signs of
Integrating Eq. (6), using Eq. (7), we obtain β and 1 + α are the same. However, Eq. (13) implies
that when α > 0 then β > 0 and hence Eq. (16) im-
ρ = Aa−3(1+ω) , A = const. (10) plies that G < 0. Since we have considered α > 0
then β > 0. Thus, phantom energy existence requires
Substituting this in Eq. (8) using Eq. (9), one obtains G < 0 and Λ > 0 so that Ġ > 0 and Λ̇ < 0 (as can be
  seen from Eqs. (8) and (9)). We, therefore, see that
−3αβ the phantom energy has negative gravity. One may
G= a3(1+ω)(1+α) , α 6= −1, 0.
8π(1 + α)A1+α attribute this to the anti-gravitating nature of phan-
(11) tom energy. Such a bizarre behaviour could be the
When α = 0, Eq. (8) implies that Λ = const and reason why phantom energy has a negative pressure,
G = const. Substituting Eqs. (10) and (11) into unlike the ordinary matter.
Eq. (4) we obtain In the scalar field theory, the phantom energy is
 β  modelled by a field with a negative kinetic energy, but
H2 = A−α a3(1+ω)α , α 6= −1, 0, (12) negative pressure and positive energy. In our present
1+α scenario, G < 0 and p < 0, but ρ > 0. Hence, the
where H = ȧ/a, or two pictures of phantom energy evolution could be
equivalent. However, in our present scenario while the
 β 
gravitational increases, the cosmological constants de-
H2 = ρ−α , α 6= −1, 0. (13)
1+α creases with time. Unlike the standard phantom en-
ergy model, we see that as t → 0 a → 0, ρ → 0; and
Using Eq. (13), Eq. (11) can be written as
as t → ∞, a → ∞, ρ → ∞. Consider now the case
 −3αβ  ω = −3/2 and α = 2/3. In this case, a ∝ t2 , and
8πG = ρ−(α+1) , α 6= −1, 0. (14) hence, ρ ∝ t3 and G ∝ t−5 . An equivalent case corre-
1+α
sponds to α = 1 and ω = −4/3. However, in the latter
Integrating Eq. (13), using Eq. (10), one obtains case ρ ∝ t2 and G ∝ t−4 . The present observational
data favour negative values for ω rather than positive
a(t) = A−1/3(1+ω) F −2/3α(1+ω) t−2/3α(1+ω) , ones. In particular, models with −1.62 < ω < −0.74
α 6= 0, −1, (15) are favoured observationally.[22]
No. 12 Arbab I. Arbab
The present observational data usher toward an
accelerated expansion of the universe.[1,2] The decel-
eration parameter is given by q = −(ä/a)/H 2 . Using
Eq. (15), this yields
1
q = − [2 + 3α(1 + ω)]. (19)
2 G=
For an accelerating universe, one must have q < 0,
−2
i.e., α > . For non-phantom energy one has
3(1 + ω)
1 + ω > 0 so that α < 0. We treat the case α = −1
separately.
For ω > −1 and α < −1, the scale factor grows
with time. For a positive gravitational constant,
G > 0, Eq. (14) implies β < 0, so that the cosmo-
logical constant becomes negative, Λ < 0. Equation
(16) and (17) imply that G and ρ decrease with cos-
mic expansion. For α = −2 and ω = −1/2 one has
a ∝ t2/3 . The evolution of dust particles in Fried-
mann cosmology mimics a dark energy. Similarly for
α = −2 and ω = −1/3, the scale factor evolves as,
a ∝ t1/2 . Hence, cosmic strings in this case evolve
like radiation in the standard Friedmann cosmology.
For the two cases above the energy density evolves as,
ρ ∝ t−1 and G ∝ t−1 .
Now consider the case −1 < α < 0, β > 0 and
ω > −1. This is the case for ordinary matter (or dark
energy). In this case, G > 0, Λ > 0. Equations (16)
and (17) imply that G increases with time and ρ de-
creases with time. For α = −1/2 and ω = 1, the scale
factor varies as a ∝ t2/3 . Thus, stiff matter mimics
ordinary matter. In this case, G ∝ t2 and ρ ∝ t−4 .
We notice that in the present scenario one has the re-
lation Gρ ∝ H 2 . Such a relation is known to satisfy
the Machian cosmology.[17]
In this model, it is possible to have a cosmic ac-
celeration in the present epoch with ordinary matter
(dust or radiation). To my knowledge this solution
has not been considered before. This arises due to the
presence of the cosmological constant of the form sug-
gested in Eq.(9). We consider here ω > 0 and β > 0.
In this case we have
dust : ω = 0 ⇒ α > −2/3.
radiation : ω = 1/3 ⇒ α > −1/2.
For the two cases, both G > 0 and Λ > 0. Moreover,
G increases while Λ decreases with cosmic time.
In this case, we consider −1 < ω < 0. Cosmic ac-
celeration for strings-like and domain-walls like fluid
respectively imply
cosmic − strings : ω = −1/3 ⇒ α > −1,
domain − walls : ω = −2/3 ⇒ α > −2.
We observe that domain-walls proceed in two differ-
ent ways: (i) when α < −1, β < 0 for G > 0 so that
4499
Λ < 0, (ii) when α > −1, β > 0 for G > 0 so that
Λ > 0. Hence, during the domain-walls evolution the
cosmological constant changes its sign.
We study in this section the case α = −1, i.e.,
Λ = 3βρ. Equations (8) and (9) imply that
3β
(1 + ω) ln(Ca), C = const. (20)
8π
This equation implies that when a → 0, G → −∞
unless β or 1 + ω becomes negative. Notice that when
a → ∞ then G → ∞. Hence, these solutions show
that the Universe can’t start from a singularity or end
at infinity as far as |G| cannot be infinite. Hence, one
may say that in the case Λ = 3βρ, the initial singu-
larity is avoided provided the gravitational constant is
allowed to vary with time. For an expanding universe
the gravitational constant must increase with cosmic
time. We notice that G increases with time for β > 0
(Λ > 0) and ω > −1 (ordinary/dark energy), or β < 0
(Λ < 0) and ω < −1 (phantom energy). Upon using
this Eq. (20), Eq. (4) becomes
ȧ2 = β[1 + (1 + ω) ln(Ca)]Aa−(1+3ω) . (21)
The solution of the above equation gives the time
dependence of the gravitational constant and energy
density. However, for ω = −1, Eq. (21)√yields the
exponential expansion of the form H = βA. This
requires that both β > 0 and A > 0.
We consider here the case ω = −1. In this case,
Eq. (10) yields ρ = const and then Eq. (9) implies
Λ = const. Now Eq. (8) implies that G = const 6= 0, as
far as α 6= −1, and hence Eq. (4) gives a ∝ exp(H0 t),
H0 = const. This is the standard de-Sitter expansion.
We notice from Eq. (20) that during inflation the grav-
itational constant vanishes when α = −1. This may
have assisted the universe to inflate with a constant
p
expansion rate, H = Λ/3. This solution is the same
as the one obtained where Λ/3 = βA.
We have considered in this paper a cosmological
model with a cosmological constant varies as Λ =
3β/ρα . We have found that cosmic acceleration is
guaranteed in radiation (−1/2 < α < 0) and mat-
ter (−2/3 < α < 0) dominated epochs with Λ > 0
and G > 0. For dark energies (cosmic-strings/domain-
walls) cosmic acceleration occurs when α < −1. Phan-
tom energy with ω < −1 is allowed provided α > 0
and G < 0. The phantom energy density varies as
ρ ∝ t2/α . For instance, for α = 1 and ω = −4/3 the
scale factor increases as a ∝ t2 . The negative grav-
itational constant is tantamount to negative kinetic
energy for phantom field. We could have obtained a
positive gravitational constant if we have not set the
integration constant to zero. The case ω = −1 gives
the familiar de-Sitter inflationary solution.
I would like to thank M. Jamil for useful commu-
nication.
4500 Arbab I. Arbab Vol. 25

References Peebles P J E and Ratra B 1988 Astrophys. J. 325 L17

[13] Sahni V 2004 Lect. Notes Phys. 653 141
[1] Perlmutter S et al 1998 Astrophys. J. 517 565 [14] Stefancic H 2004 Phys. Lett. B 586 5
[2] Reiss A et al 1998 Astrophys. J. 116 1009 [15] Beesham A 1986 Int. J. Theor. Phys. 25 1295
[3] Caldwell R, Kamionkowski M and Weinberg N 2003 Phys. [16] Abdel Rahman A.-M M 1990 Gen. Rel. Gravit. 22 655
Rev. Lett. 91 071301 [17] Arbab A I 2003 Class. Quantm. Gravit. 20 93
[4] Dabrowski M P, Stachowiak T and Szydlowski M 2003
Arbab A I 1997 Gen. Rel. Gravit. 29 61
Phys. Rev. D 68 103519
Arbab A I 2003 J. Cosmo. Astroparticle Phys. 05 008
[5] Chiba T, Okabe T and Yamaguchi M 2000 Phys. Rev. D
62 023511 [18] Shapiro Ilya L and Hrvoje Stefancic J S 2005 J. Cosmol.
[6] Sahni V and Shtanov Y 2003 J. Cosmol. Astropart. Phys. Astropart. Phys. 0501 012
11 014 [19] Berman M S, Som M M and Gomide F M 1989 Gen. Rel.
[7] Caldwell R 2002 Phys. Lett. B 545 23 Grav. 21 287
[8] Ratra B and Peebles P J E 1988 Phys. Rev. D 37 3406 [20] Berman M S and Som M M 1990 Int. J. Theor. Phys. 29
[9] Wetterich C 1995 Astron. Astrophys. 301 321 1411
[10] Turner M S and White M 1997 Phys. Rev. D 56 R4439 [21] Lima J A S and Carvalho J C 1994 Gen. Rel. Gravit. 26
[11] Caldwell R R, Dave R and Steinhardt P J 1998 Phys. Rev. 909
Lett. 80 1582 [22] Carroll M S, Hoffman M and Trodden M 2003 Phys. Rev.
[12] Peebles P J E and Ratra B 2003 Rev. Mod. Phys. 75 559 D 68 023509