10 1103@physrevd 87 104040
10 1103@physrevd 87 104040
104040-2
COSMOLOGICAL CONSTANT AS A FREE . . . PHYSICAL REVIEW D 87, 104040 (2013)
Reissner-Nordström black hole, and in Ref. [10] for the Z 1 ~ 1=2
energy of the Kerr-Newman black hole. EBY ¼ 1 ~
r3 d: (3.9)
0 6 3
Equation (3.1) may be justified by means of the
Brown-York energy [11] When the cosmological constant is considered as a free
variable, we may therefore understand the Brown-York
1 I energy as the energy downloaded inside of the two-sphere
EBY :¼ ðk k0 ÞdA: (3.2)
8 S r ¼ constant, when the cosmological constant is increased
In Eq. (3.2) we have integrated over a closed, spacelike from zero to its given value. According to Eq. (2.9) the
two-surface S embedded in a spacelike hypersurface, change d in the cosmological constant implies the change
where the time t is a constant. k is the trace of the exterior drC in the radius rC of the cosmological horizon such that
curvature tensor on the two-surface, and k0 the trace of the d
exterior curvature tensor on the two-surface, when the two- d ¼ drC ¼ 2 drC (3.10)
drC rC
surface has been embedded in flat spacetime. dA is the
area element on the two-surface. In stationary spacetimes and hence Eq. (3.8) may also be written, according to
the Brown-York energy may be understood as the gravita- Eq. (2.6), as
tional energy enclosed inside of the two-surface. For a two- r2
sphere r ¼ constant in the de Sitter spacetime the only dEBY ¼ a dr ; (3.11)
nonzero components of the exterior curvature tensor are, rC C
when r < rC , where a is the proper acceleration on the two-sphere.
1=2 1=2 Hence, the change in the Brown-York energy may be writ-
k ¼ 1 r2 r ¼ 1 r2 r; (3.3a) ten in terms of the change dr in the radius of the shrinked
3 3 horizon a ¼ constant as
1=2 1=2
k ¼ 1 r2 r ¼ 1 r2 rsin 2 ; a r drC
3 3 dEBY ¼ dA; (3.12)
8 rC dr
(3.3b)
where dA ¼ 8rdr is the change in the area of the shrinked
and therefore the trace of the exterior curvature tensor is horizon. In the limit, where r ! rC , we may write Eq. (3.12)
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi as
2
k ¼ k þ k ¼ 1 r2 : (3.4) dEBY ¼
a
dA: (3.13)
r 3 8
When the two-sphere is embedded in flat spacetime, the The minus sign on the right-hand side of Eq. (3.13)
trace of the exterior curvature tensor is indicates that increase in the area A of the shrinked horizon
2 decreases the amount of energy inside of the shrinked
k0 ¼ ; (3.5) horizon. This means that if we increase the shrinked hori-
r zon, energy flows outwards through the shrinked horizon.
and because the area of the two-sphere is The amount of energy flown during the process, where the
shrinked horizon area has been increased by dA, is
A ¼ 4r2 ; (3.6)
a
dE ¼ dA; (3.14)
the Brown-York energy takes the form 8
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi and hence it follows that the amount of energy flown
EBY ¼ r 1 1 r2 : (3.7) through the horizon, when its area is increased from zero
3 to A is
Consider now how the Brown-York energy will change in a
E¼ A: (3.15)
the infinitesimal changes of the cosmological constant . 8
Differentiating the right-hand side of Eq. (3.7) with respect An identification of this expression as the energy of the
to , one finds that if the cosmological constant takes on de Sitter spacetime from the point of view of an observer
an infinitesimal change d, the Brown-York energy on its shrinked horizon a ¼ constant implies Eq. (3.1).
undergoes a change,
1=2
1 IV. THE PARTITION FUNCTION
dEBY ¼ 1 r2 r3 d; (3.8)
6 3
In our model we construct the shrinked horizon
and we may write the Brown-York energy in an integral a ¼ constant of the de Sitter spacetime out of discrete
form: constituents, each of them contributing to the shrinked
104040-3
JARMO MÄKELÄ PHYSICAL REVIEW D 87, 104040 (2013)
horizon an area, which is an integer times a constant. As a It was found in Ref. [7] that the number of such strings is
consequence, the area of the shrinked horizon takes the
XN
form n1
gðEn Þ ¼ ; (4.8)
m1
A ¼ ‘2Pl ðn1 þ n2 þ n3 þ þ nN Þ; (4.1) m¼1
where n1 ; n2 ; . . . ; nN are non-negative integers, is a pure whenever N < n. In the special case, where n ¼ N,
number to be determined later, and we have
sffiffiffiffiffiffiffi n
ℏG X n1
‘Pl :¼ 1:6 1035 m (4.2) gðEn Þ ¼ ¼ 2n1 : (4.9)
c3 m¼1
m 1
is the Planck length. In Eq. (4.1) N, which is assumed to be When n N, Eq. (4.9) always holds, no matter what is N.
very large, is the number of the constituents of the shrinked We are now prepared to write the partition function for
horizon. N is assumed to be fixed. The non-negative in- the de Sitter spacetime from the point of view of an
tegers nj (j ¼ 1; 2; . . . ; N) are quantum numbers determin- observer on its shrinked horizon a ¼ constant. As in
ing the quantum states of the constituents. Constituent j is Ref. [7], the partition function turns out to be of the form
in vacuum, if nj ¼ 0; otherwise it is in an excited state.
Hence, our model is similar to the one constructed in ZðÞ ¼ Z1 ðÞ þ Z2 ðÞ; (4.10)
Ref. [7] for the stretched horizon of the Schwarzschild where
black hole, and in Ref. [8] for the stretched horizon of
the Reissner-Nordström black hole. 1 XN
Z1 ðÞ :¼ 2ð1TC Þn ; (4.11a)
We shall assume that the quantum states of the de Sitter 2 n¼1
spacetime are somehow encoded into the quantum states of 1 X
!
X N n 1 nT
its shrinked horizon. The partition function of the de Sitter Z2 ðÞ ¼
: 2 C ; (4.11b)
spacetime takes the form n¼Nþ1 k¼0 k
X
ZðÞ ¼ gðEn ÞeEn ; (4.3) and we have defined the characteristic temperature
n a
TC :¼ : (4.12)
8 ln 2
where the index n labels the possible energies En of our
system, and is the inverse of its temperature. gðEn Þ is the The partition function ZðÞ of Eq. (4.10) may be calculated
number of the degenerate states associated with the energy explicitly, and it takes the form [7]
En . Using Eqs. (3.1) and (4.1) we find that the possible Nþ1
1 1
energies of the de Sitter spacetime from the point of view ZðÞ ¼ TC 1 TC ; (4.13)
2 2 2 1
of an observer on its shrinked horizon are of the form
when TC Þ 1, and
a
En ¼ n ; (4.4) ZðÞ ¼ N þ 1; (4.14)
8
where when TC ¼ 1.
n :¼ n1 þ n2 þ þ nN : (4.5)
V. ENERGY AND ENTROPY
As in Ref. [7] we take the number gðEn Þ of the degenerate The expression obtained in Eqs. (4.13) and (4.14)
states associated with the same energy En to be the number for the partition function of the de Sitter spacetime is
of the different combinations of the nonvacuum quantum identical to the partition function obtained in Ref. [7] for
states of the constituents of the shrinked horizon yielding the Schwarzschild black hole. As a consequence, the ex-
the same energy En . As a consequence, gðEn Þ equals with pressions for the energy
the number of ways of expressing a given positive integer n
@
as a sum of at most N positive integers. More precisely, EðÞ ¼ ln ZðÞ (5.1)
gðEn Þ is the number of the ordered strings @
and the entropy
S :¼ ðn1 ; n2 ; . . . ; nm Þ; (4.6)
SðÞ ¼ EðÞ þ ln ZðÞ (5.2)
where 1 m N, and nj 2 f1; 2; 3; . . .g such that of the system from the point of view of an observer on the
shrinked horizon a ¼ constant are identical to those of the
n1 þ n2 þ þ nm ¼ n: (4.7) Schwarzschild black hole. The energy per a constituent
104040-4
COSMOLOGICAL CONSTANT AS A FREE . . . PHYSICAL REVIEW D 87, 104040 (2013)
EðÞ is the Unruh temperature measured by our observer [12].
EðÞ :¼ (5.3) We shall see in the next section that the Unruh temperature
N
TU really is the lowest possible temperature measured by
takes, in the leading approximation for large N, the form our observer. With the choice (5.9) for the numerical
EðÞ ¼ E 1 ðÞ þ E 2 ðÞ; (5.4) constant one may show, as in Ref. [7], that the entropy
of the de Sitter spacetime is zero, whenever T < TC .
where However, if T > TC , one finds that between the area A of
1 2TC the shrinked horizon and the entropy S there is the
E 1 ðÞ :¼ T ln 2; (5.5a) relationship
N 2TC 2 C
2TC 1 2A 2A Acrit
EðÞ :¼ T TC ln 2: (5.5b) SðAÞ ¼ A ln þ N ln ;
ð2 C 1ÞNþ2 2TC þ 1 4 ln 2 2A Acrit Acrit
(5.12)
Equations (5.4) and (5.5) imply that
where the critical area
lim EðÞ ¼ 0; (5.6)
N!1 Acrit :¼ 8N ln 2 (5.13)
whenever T < TC , which means that the constituents of is the area of the shrinked horizon in the limit, where
the shrinked horizon are effectively in vacuum, when the T ! TCþ . In this limit A ! Aþ
crit , and Eq. (5.12) implies
temperature T measured by our observer is less than the
characteristic temperature TC . However, if T > TC , 1
SðAÞ ¼ A: (5.14)
the average energy EðÞ per a constituent is nonzero, and 4
Eqs. (5.3) and (5.5b) imply that in the leading approxima- In other words, the entropy of the de Sitter spacetime is
tion for large N we have one-quarter of the area of its cosmological horizon in the
2TC limit, where T ! TCþ . The result is similar to the one
EðÞ ¼ NTC ln 2: (5.7) obtained for black holes [7,8]. Originally, the entropy of
2 1
TC
the de Sitter spacetime was shown to equal with the one-
An interesting aspect of this result is that it relates the value quarter of the area of its cosmological horizon by Gibbons
of the cosmological constant to the absolute temperature and Hawking in Ref. [13].
T ¼ 1= measured by our observer. Since the shrinked
horizon lies very close to the cosmological horizon of the VI. PHASE TRANSITION
de Sitter spacetime, we may effectively regard the areas of
these two horizons as equals, and Eqs. (2.2), (3.1), (3.6), As a we found in Sec. V, the constituents of the shrinked
(4.12), and (5.7) imply horizon are effectively in vacuum, and there is no cosmo-
logical horizon, when T < TC . However, when T > TC ,
12 2TC 1 Eq. (5.7) implies that the average energy per constituent
¼ : (5.8)
N 2TC is given in the large N limit, to a very good approximation,
So we observe that the cosmological constant is, in our by the formula
model, a temperature-dependent quantity, which depends 2TC
both on the absolute temperature T, and on the number N
EðÞ ¼ TC ln 2: (6.1)
2TC
1
of the constituents of the shrinked horizon. Equation (5.8)
is the thermodynamical equation of state for the cosmo- As one may observe, we have
logical constant . We shall consider the cosmological
lim EðÞ ¼ 2TC ln 2; (6.2)
implications of Eq. (5.8) in Sec. VII. T!TCþ
It is interesting that the characteristic temperature TC ,
which was defined in Eq. (4.12), is proportional to the and so we may conclude that the shrinked horizon under-
proper acceleration a of our observer. If we choose the goes a phase transition at the characteristic temperature
unspecified numerical constant such that TC . The latent heat per constituent associated with this
phase transition is
¼ 4 ln 2; (5.9)
L ¼ 2TC ln 2: (6.3)
we find
Using Eqs. (4.4), (4.5), and (4.12) we find that the average
T ¼ TU ; (5.10)
n1 þ n2 þ þ nN
nðÞ :¼ (6.4)
where N
a
of the quantum numbers n1 ; n2 ; . . . ; nN is related to EðÞ
T U :¼ (5.11)
2 such that
104040-5
JARMO MÄKELÄ PHYSICAL REVIEW D 87, 104040 (2013)
EðÞ With the currently accepted estimate
nðÞ ¼ ; (6.5)
TC ln 2 1035 s2 (6.11)
and Eq. (6.3) implies that for the cosmological constant, one finds
n ¼ 2 (6.6) T 1030 K; (6.12)
after the phase transition has been completed. So we find which is very low, indeed.
that during the phase transition the constituents of the
shrinked horizon jump, in average, from the vacuum to
the second excited states. Since the constituents of the VII. EQUATION OF STATE
shrinked horizon are effectively in vacuum, when T < Consider now in detail the thermodynamical equation of
TC , the characteristic temperature TC , which was found state found in Eq. (5.8) for the cosmological constant. In
to agree with the Unruh temperature TU , is the lowest the SI units, using Eq. (5.9), we may write Eq. (5.8) as
possible temperature, which the de Sitter spacetime may
have from the point of view of an observer on the shrinked 3c2 2kB TC 1
¼ ; (7.1)
horizon. If the temperature of the environment of N‘2Pl ln 2 2kB TC
the shrinked horizon drops below TC , the constituents of
the shrinked horizon begin to perform transitions from the where ‘Pl is the Planck length, which was defined in
second excited states to the lower states, and the shrinked Eq. (4.2). Equation (7.1) implies
horizon begins to radiate with the Unruh temperature TU . 3c2
One of the major advantages of our choice to consider limþ ¼ ; (7.2)
T!TC 2N‘2Pl ln 2
the thermodynamics of the de Sitter spacetime from the
point of view of an observer on the shrinked horizon, where and therefore we are able to write the cosmological con-
the proper acceleration a ¼ constant, is that the de Sitter stant as a function of the number N of the constituents of
spacetime has, from the point of view of our observer, a the shrinked horizon only in the special case, where the
well-defined, fixed phase transition temperature, which temperature measured by our observer on the shrinked
depends on the proper acceleration of the observer only. horizon agrees with his Unruh temperature. Since N is
Such temperature would not exist, if the observer had been assumed to be very large, the cosmological constant is
chosen otherwise. Moreover, we have seen that the calcu- very small. Even though N is very large, however, N
lations are relatively simple: The expression for the energy is finite, and therefore the cosmological constant must be
E of the de Sitter spacetime in Eq. (3.1) was very simple, nonzero. Hence, we may say that in our model nonzero
and the subsequent calculations leading to the explicit cosmological constant appears as a natural consequence of
expression in Eq. (4.13) for the partition function ZðÞ of the discrete structure of spacetime.
the de Sitter spacetime were easy to perform. It is somewhat uncertain what we should regard as the
Equations (2.6) and (5.11) imply that we may write the temperature of the cosmological horizon from the point of
characteristic temperature TC as view of an observer on the shrinked horizon. The majority
rC of physicists would probably hold the view that the tem-
TC ¼ B ; (6.7) perature in question is simply the Unruh temperature mea-
6
sured by the observer. In that case the constituents of the
where horizon lie, in average, on the second excited states, and
Eqs. (4.1) and (5.9) imply that the average area of an
2 1=2
B ¼ 1 r
: (6.8) individual constituent is
3
A ¼ 8‘2Pl ln 2 1:4 1069 m2 : (7.3)
is the blue shift factor. The Tolman relation [14] implies
that far from the cosmological horizon, i.e., when rrC 1, Using Eq. (7.2) we may write the constituent number N as
the temperature of the radiation emitted by the horizon is, a function of the cosmological constant :
when the backscattering effects are neglected, 3c2
pffiffiffiffiffiffiffi N¼ ; (7.4)
rc 3 2‘2Pl ln 2
T ¼ ¼ ; (6.9)
6 6 and putting 1035 s2 we find
where we have used Eq. (2.2). In the SI units Eq. (6.9) takes
N 10122 : (7.5)
the form
pffiffiffiffiffiffiffi This is a huge number, but nevertheless it is finite. The
3 ℏ
T ¼ : (6.10) reason why the observed value of the cosmological
6 kB constant is around 1035 s2 is, in our model, that the
104040-6
COSMOLOGICAL CONSTANT AS A FREE . . . PHYSICAL REVIEW D 87, 104040 (2013)
number of the constituents of the cosmological horizon is In the considerations performed so far, we have identi-
around 10122 . fied the temperature measured by an observer on the
shrinked horizon with his Unruh temperature. It is interest-
ing to consider what would happen, if we assumed the
VIII. CONSISTENCY OF THE MODEL
shrinked horizon to be in thermal equilibrium with the
As we found in Sec. VI, the cosmological horizon emits cosmic microwave background. In that case the tempera-
radiation with the characteristic temperature T given in ture measured by our observer for the shrinked horizon
Eq. (6.10). As a consequence, the cosmological constant would equal with his measurement for the temperature of
will increase in time. However, at the beginning of our the cosmic microwave background. Since the current tem-
analysis in Sec. II we assumed that the cosmological perature TR 3 K measured by an observer far from the
constant is strictly constant. We must therefore check, cosmological horizon for the cosmic microwave back-
whether the increase of the cosmological constant in time ground is very much higher than the temperature T in
as a consequence of the radiation of the cosmological Eq. (6.12), the temperature T measured by the observer on
horizon is slow enough such that we may regard the the shrinked horizon for the cosmic microwave back-
cosmological ‘‘constant,’’ in effect, as a true constant. ground must be much higher than TC . Since
According to the first law of thermodynamics, the
2x 1 þ x ln 2; (8.7)
amount of energy emitted by the cosmological horizon
is, from the point of view of an observer far from the when x 1, we may write Eq. (7.1) as
horizon,
3c2 TC
dE ¼ T dS; (8.1) ¼ ; (8.8)
N‘2Pl T
where dS is the change in the entropy S of the horizon. when T TC . The Tolman relation implies that
Using Eqs. (5.14) and (6.10) we find that, in the SI units,
pffiffiffiffiffiffiffi 3 TC ¼ BT ; (8.9a)
3 c
dE ¼ dA; (8.2) T ¼ BTR ; (8.9b)
24 G
where dA is the change in the area of the cosmological where B is the blueshift factor, defined in Eq. (6.8), for the
horizon. Writing dA ¼ 8rC drC , and using Eq. (2.2) we observer on the shrinked horizon. Hence we find
find that the energy emitted in a unit time is 3c2 T
sffiffiffiffiffiffi ¼ ; (8.10)
N‘2Pl TR
dE c5 3 d
¼ : (8.3)
dt 2G 3 dt and Eq. (6.10) implies for the cosmological constant an
equation:
On the other hand, the energy emitted by the cosmological pffiffiffiffiffiffiffi 5
horizon in a unit time is given by the Stefan-Boltzmann 3c
law: ¼ ; (8.11)
2NGkB TR
dE
¼ AT4 ; (8.4) which has the solution,
dt
3c10 1
where is the Stefan-Boltzmann constant. Comparing ¼ : (8.12)
4G2 ðNkB TR Þ2
Eqs. (8.3) and (8.4) we get
1 d Gℏ4 pffiffiffiffiffiffiffiffiffi From this equation we may solve N in terms of the cos-
¼ 3 4 3
33 : (8.5) mological constant :
dt 18 kB c
sffiffiffiffi
Again, putting 1035 s2 we find c5 3
N¼ : (8.13)
2GkB TR
1 d
10143 s1 ; (8.6)
dt Putting 1035 s2 and TR ¼ 3 K we find
which means that the time needed for the doubling of the N 1092 : (8.14)
cosmological constant would be around 10143 seconds.
This is a huge amount of time (the present age of the This number is 30 orders of magnitude less than the
Universe is about 1017 seconds), and hence we may regard estimate obtained in Eq. (7.5). As a consequence, the
the cosmological constant, from the practical point of view, average diameter of an individual constituent of the cos-
as a true constant of nature. This means that our analysis is mological horizon is about 1015 times the Planck length, or
self-consistent. 1020 m, which is still much less than the effective size of
104040-7
JARMO MÄKELÄ PHYSICAL REVIEW D 87, 104040 (2013)
any known elementary particle. Substituting Eq. (8.14) in The radiation emitted by the cosmological horizon does
Eq. (8.12) we may express Eq. (8.12) numerically as not violate the second law of thermodynamics any more
than does the Hawking radiation of a black hole. The
1 decrease of the entropy of the horizon as a result of
1035 K2 s2 : (8.15)
TR2 the decrease in its area is always less or the same as the
increase in the entropy of the radiation. As a consequence,
It is very interesting that according to Eq. (8.15) the the sum of the entropies of the horizon and the radiation
cosmological constant is inversely proportional to the may never decrease in time, and the second law of ther-
square of the absolute temperature TR of the cosmic mi- modynamics remains to be valid.
crowave background. So we find that when the Universe It should be noted that in our very simple model the
cools down because of its expansion and TR decreases, the Universe is not assumed to consist of matter at all, but just a
cosmological constant increases, and it seems that the positive cosmological constant. In the real Universe, of
cosmological constant was less in the past than it is now. course, there is matter as well, and the area of the cosmo-
Unfortunately, such a conclusion would lead to inconsis- logical horizon is not determined by the cosmological
tencies in our model, which may be seen, if we consider the constant alone, but also by the matter fields. During the
energy density of the cosmic microwave background. course of time matter escapes beyond the cosmological
Since the energy density of thermal radiation is propor- horizon and, as a consequence, the area of the cosmologi-
tional to the fourth power of its absolute temperature, cal horizon increases in time in the same way as does the
conservation of energy implies event horizon area of a black hole, when the hole swallows
matter. Such processes have been investigated, among
R3 TR4 ¼ constant; (8.16) other things, in a recent paper by Mimoso and Pavón,
where R is the scale factor of the Universe. Differentiating and it was found that the entropy of the horizon plus
both sides of Eq. (8.16) we find that of radiation and matter inside it increases and is
concave [16].
dR dT
3 ¼ 4 R ; (8.17)
R TR IX. CONCLUDING REMARKS
and Eq. (8.12) implies an equation: In this paper we have considered the thermodynamics of
the de Sitter spacetime from the point of view of an
d 3
¼ ; (8.18) observer on a spacelike two-sphere, which is just inside
dR 2R of the cosmological horizon. For the sake of brevity and
simplicity we called the two-sphere on question as the
which has the solution
shrinked horizon of the de Sitter spacetime. Our idea was
¼ CR3=2 ; (8.19) to consider the cosmological constant as a free thermo-
dynamical variable of the system, and we assumed that
where C is a positive constant. So we observe that the when the cosmological constant varies, the radius of the
cosmological constant increases surprisingly rapidly as a shrinked horizon will also change, but in such a way that
function of the scale factor R. Actually, the increase of the the proper acceleration a of an observer on the shrinked
cosmological constant is so rapid that it takes us into a horizon stays unchanged. We constructed the shrinked
contradiction with our basic assumption, which was that horizon out of discrete constituents, each of them contrib-
the cosmological constant may be regarded, at least effec- uting to the shrinked horizon an area, which is an integer
tively, as a constant. A careful reanalysis, where the pos- times a constant. In this sense our model was similar to
sible increase of the cosmological constant in time is taken those constructed previously for the Schwarzschild [7] and
into account right from the beginning, is therefore needed the Reissner-Nordström [8] black holes. Using our model
[15]. Nevertheless, our results are very suggestive, and it we managed to obtain an explicit, analytic expression for
would be very nice, if observational cosmologists managed the partition function of the de Sitter spacetime.
to find evidence for an increasing cosmological constant. Our partition function implied, among other things, that
It may be pretty surprising that as a result of the an observer on the shrinked horizon measures for the
Hawking-type radiation emitted by the cosmological hori- cosmological horizon a certain minimum temperature,
zon the cosmological constant increases in time leading which is proportional to the observer’s proper acceleration
to the shrinking of the cosmological horizon. If the entropy a, and may be identified as the observer’s Unruh tempera-
of the de Sitter spacetime equals to one-quarter of the area ture. When the temperature of the horizon equals, from the
of its horizon and the horizon shrinks, then what will point of view of our observer, with the observer’s Unruh
happen to the second law of thermodynamics, which states temperature, the entropy of the de Sitter spacetime is, in the
that the total entropy of an isolated system may never natural units, exactly one-quarter of the area of the cosmo-
decrease in time? logical horizon.
104040-8
COSMOLOGICAL CONSTANT AS A FREE . . . PHYSICAL REVIEW D 87, 104040 (2013)
The most important result obtained in this paper may be identified with the cosmological horizon, an esti-
concerned the properties of the cosmological constant. mate 10122 .
We found that if the temperature measured by an Taken as a whole, our model seems to suggest that dark
observer on the shrinked horizon equals with his Unruh energy is not necessarily needed to explain the acceleration
temperature, the cosmological constant is inversely pro- of the expansion of the Universe, but the accelerating
portional to the number of the constituents of the shrinked expansion of the Universe may be a simple consequence
horizon. However, because the number of the constituents of the fundamentally discrete nature of spacetime, and of
of the shrinked horizon was assumed to be finite—even the principles of thermodynamics. According to this view
though very large—it follows that the cosmological con- spacetime has certain fundamental constituents—sort of
stant must necessarily be nonzero and bounded from be- ‘‘atoms of spacetime’’—with certain quantum states, and if
low. At the Unruh temperature the constituents of the one considers the cosmological constant as a free thermo-
shrinked horizon are Planck-size objects, and using the dynamical variable, the principles of thermodynamics im-
currently accepted estimate 1035 s2 for the cosmologi- ply that is positive and bounded from below. In this sense
cal constant we obtained for the number of the constituents we may regard the accelerating expansion of the Universe
of the shrinked horizon which, for all practical purposes, as a quantum effect of gravitation.
[1] A. Einstein, Sitz. Preuss. Acad. Wiss. 142 (1917). [9] See, for example, R. Bousso, arXiv:hep-th/0205177.
[2] See, for example, S. Coleman, Nucl. Phys. B310, 643 [10] E. Frodden, A. Ghosh, and A. Perez, arXiv:1110.4055.
(1988). [11] J. D. Brown and J. W. York, Phys. Rev. D 47, 1407 (1993).
[3] A. G. Riess et al., Astron. J. 116, 1009 (1998); S. [12] W. G. Unruh, Phys. Rev. D 14, 870 (1976).
Perlmutter et al., Astrophys. J. 517, 565 (1999). [13] G. W. Gibbons and S. W. Hawking, Phys. Rev. D 15, 2738
[4] See, for example, S. M. Carroll, Living Rev. Relativity 4, 1 (1977).
(2001). [14] See, for example, N. D. Birrell and P. C. Davies, Quantum
[5] See, for example, P. J. E. Peebles and B. Ratra, Rev. Mod. Fields in Curved Space (Cambridge University Press,
Phys. 75, 559 (2003). Cambridge, England, 1982).
[6] The idea that the accelerating expansion of the Universe [15] Cosmological models, where the cosmological constant is
may be caused by a discrete structure of spacetime is not not a constant, have been studied, among others, by J. M.
new. It has been advocated, among others, by T. Overduin and F. I. Cooperstock, Phys. Rev. D 58, 043506
Padmanabhan, arXiv:1206.4916; Res. Astron. Astrophys. (1998). However, in their models the cosmological con-
12, 891 (2012); arXiv:1208.1375; arXiv:1210.4174. stant always decreases in time.
[7] J. Mäkelä, Entropy 13, 1324 (2011). [16] J. P. Mimoso and D. Pavón, Phys. Rev. D 87, 047302
[8] J. Mäkelä, arXiv:1212.1155. (2013).
104040-9