Professional Documents
Culture Documents
P - Brane Dynamics in (N + 1) - Dimensional FRW Universes: A Unified Framework
P - Brane Dynamics in (N + 1) - Dimensional FRW Universes: A Unified Framework
P - Brane Dynamics in (N + 1) - Dimensional FRW Universes: A Unified Framework
It is interesting to contrast this with the corresponding The frictional force originated by the interaction of the
result for domain walls, in which case the coefficient of branes with ultrarelativistic particles may be included in
the cosmological damping term is N (rather than 2) and Eq. (25), by introducing an extra term in the damping
κ = v̂· N lenghtscale, ℓ−1 −1
P −1
ĩ=1
kĩ (rather than v̂·k). Hence, the differences d = (p + 1)H + ℓf . The friction length-
between the macroscopic evolution of cosmic string and scale, ℓf , will be properly defined in in Sec. V.
domain wall networks are expected to increase with N . Energy-momentum conservation in a FRW universe
then implies that
A. Linear scaling solutions (ℓf = ∞) flat q-dimensional surfaces do in general intersect but
that is no longer true if N > 2(N − q) (or equivalently
If the friction lengthscale becomes negligible compared p < (N − 1)/2). Hence, if the p-branes are thin (p-brane
to the Hubble radius then Eqs. (25) and (29) may have thickness much smaller than L), the energy-loss param-
a linear attractor solution. This attractor solution corre- eter c̃ may be much smaller than unity if p < (N − 1)/2.
sponds to a linear scaling regime of the form Note, however, that the linear scaling solution in Eq. (30)
may be attainable for β > 1 − p/N , even if c̃ = 0 with
L = ξt∗ and v̄ = constant , (30) s s
2
k v̄ = (1 − β)D .
with ξ = (36)
β(1 − β)D(p + 1) β(p + 1)
s s
k(k + c̃) (1 − β)kD
ξ = v̄ = . As a consequence, even if c̃ ≪ 1, the p-brane network
β(1 − β)D(p + 1) β(k + c̃)(p + 1) may be able to attain a linear scaling regime in the M2+
(31) model. However, if c̃ = 0, a linear scaling solution is no
The conditions 0 < v̄ < 1 and ξ > 0 are sufficient to longer possible in a collapsing universe.
show that models M3+ and M1− do not admit linear scal-
ing solutions, which would require a negative energy loss
parameter c̃. B. Inflation and superinflation
The RMS velocity, v̄, of maximally symmetric p-branes
R
with a Sp−i ⊗ i topology oscillating periodically in a
In the case of models M3+ and M1− the expansion is fast
Minkowski spacetime is given by
enough (ä > 0) to decelerate the branes and make the
s RMS velocity arbitrarily small. As a consequence, the
p−i inflationary models M3+ and M1− undergo a stretching
vmin ≤ ≤ vmax , (32)
p−i+1 regime described by the scaling laws
2
with vmin 2
= 1/2 and vmax = p/(p + 1), which correspond, L ∝ a, v ∝ (Ha)−1 ∝ a−1−1/β → 0 . (37)
respectively, to i = p − 1 (only one of the principal cur-
vatures is non-zero) and i = 0 (for fully spherically sym- If the p-branes are the dominant energy component of
metrical p-branes). For β = 0, the curvature parameter the universe then
k must be equal to zero for a linear scaling solution with
2 2
v̄ ≤ 1 to be attained. The expansion (collapse) of the β= = . (38)
universe hinders (aids) the velocity of the branes, lead- N (1 + wb ) D + (p + 1)v̄ 2
ing to a smaller (larger) RMS velocity and a curvature
In order to accelerate the universe one needs β > 1 (or
parameter, k, larger (smaller) than zero. Therefore, one
equivalently wb < wc = (2 − N )/N ) and, consequently
expects that
2−D
0 < v̄ < vmax , for M2+ , (33) v̄ 2 < . (39)
p+1
vmin < v̄ < 1, for M1+ and M3− .
We may then conclude that only domain walls (D = N −
On the other hand the characteristic lengthscale of the p = 1) are able to drive an inflationary phase.
network is necessarily constrained by causality and, as a
consequence, L is expected to be bounded by the particle
horizon at any given time. In the case of models M1+ , M2+
C. Ultra-relativistic collapsing solution
and M3− this implies that
Z t
dt′ t∗ In the case of model M2− (which represents a collaps-
L < dH = = , (34) ing universe with k < 0 and 0 < β < 1), the comoving
ti a(t′ ) |1 − β|
Hubble radius, |H −1 | decreases with time. As a conse-
with ti = 0 or ti = −∞ (depending on whether s = + or quence, the curvature scale of the p-brane will necessarily
−, respectively), or equivalently become smaller than |H −1 | and will tend to be confor-
mally contracted in physical coordinates whilst travelling
v̄ 2 < (k + c̃)−2 . (35) at ultrarelativistic speeds with
Some remarks regarding the energy-loss term are nec- Lph ∝ a , γ̄ ∝ a−(p+1) , (40)
essary. For simplicity, we shall consider the most triv-
ial case of flat p-branes. A moving p-brane spans a q- where is defined by γ̄ = (1 − v̄ 2 )−1/2 and L ∼ γ̄ 1/D Lph ∝
dimensional surface (with q = p + 1) which has N − q a−(N +1)/D . Hence, at ultra-relativistic speeds L ceases
degrees of freedom. Hence, if N ≤ 2(N − q) then two to be an accurate measure of Lph .
5
V. FRICTION-DOMINATED REGIMES at a higher rate than H −1 (for p + 1 > 1/β) as the uni-
verse expands and eventually overcomes the character-
The interaction of p-branes with the ultrarelativistic istic length, L. This implies that the Kibble regime is
particles in a radiation fluid results in a frictional damp- necessarily transient.
ing of the form
dv 1 B. Collapsing universe
= − (1 − v 2 )v (41)
dt ℓf
In the case of a collapsing universe, the evolution of
where ℓf ∝ σp λp+1−N /ρrad ∝ ap+2 is the friction length- the network ends in a friction dominated era, with the
scale, λ ∝ a and ρrad ∝ a−(N +1) are, respectively, the p-branes coming to a standstill in comoving coordinates
typical wavelength and the energy density of relativistic and then being conformally contracted with ρ̄ ∝ a−1 .
particles. The background temperature and density will eventu-
ally approach those of the brane-forming phase transition
and, as a consequence, the branes will dissolve into the
A. Expanding universe high density background.
[1] A. H. Guth, Phys. Rev. D23, 347 (1981). [11] C. J. A. P. Martins and E. P. S. Shellard, Phys. Rev.
[2] A. D. Linde, Phys. Lett. B108, 389 (1982). D53, 575 (1996), hep-ph/9507335.
[3] G. R. Dvali and S. H. H. Tye, Phys. Lett. B450, 72 [12] C. J. A. P. Martins and E. P. S. Shellard, Phys. Rev.
(1999), hep-ph/9812483. D65, 043514 (2002), hep-ph/0003298.
[4] C. P. Burgess et al., JHEP 07, 047 (2001), hep- [13] A. Avgoustidis and E. P. S. Shellard, Phys. Rev. D71,
th/0105204. 123513 (2005).
[5] S. H. S. Alexander, Phys. Rev. D65, 023507 (2002), hep- [14] P. P. Avelino, C. J. A. P. Martins, and J. C.
th/0105032. R. E. Oliveira, Phys. Rev. D72, 083506 (2005), hep-
[6] J. Garcia-Bellido, R. Rabadan, and F. Zamora, JHEP ph/0507272.
01, 036 (2002), hep-th/0112147. [15] P. P. Avelino, R. Menezes, and J. C. R. E. Oliveira, Phys.
[7] N. T. Jones, H. Stoica, and S. H. H. Tye, JHEP 07, 051 Rev. E83, 011602 (2011), 1006.3564.
(2002), hep-th/0203163. [16] P. P. Avelino and L. Sousa (2011), 1101.3360.
[8] M. Majumdar and A. Christine-Davis, JHEP 03, 056 [17] P. P. Avelino, C. J. A. P. Martins, J. Menezes,
(2002). R. Menezes, and J. C. R. E. Oliveira, Phys. Rev. D78,
[9] S. Sarangi and S. H. H. Tye, Phys. Lett. B536, 185 103508 (2008), 0807.4442.
(2002), hep-th/0204074. [18] V. A. Topogonov, Differencial Geometry of Curves and
[10] C. J. A. P. Martins and E. P. S. Shellard, Phys.Rev. D54, Surfaces: A Concise Guide (Birkhauser, 2006).
2535 (1996). [19] D. Garfinkle and R. Gregory, Phys. Rev. D41, 1889
6