Static Cylindrically Symmetric Solutions of Einstein'S Equations
Static Cylindrically Symmetric Solutions of Einstein'S Equations
Static Cylindrically Symmetric Solutions of Einstein'S Equations
ABSTRACT
In recent years, a number of solution generating techniques for spherically symmetric perfect fluid solutions of Einstein’s
equations have been invented. Besides, solutions with cylinderical symmetry are much less studied because of the
complexity of calculations involving the equations for cylinderically spacetime. For our kith interest in the cylindrically
symmetric static perfect fluid solutions of Einstein’s equations, we have provided an algorithm and find a new realistic
solution.
KEYWORDS: Spherical Symmetry, Cylinderical Symmetry, Tangential Gauge, Arc-length Gauge, Metric Function, Axis
of Symmetry, Minkowski Space
Article History
Received: 04 Jun 2020 | Revised: 22 Jun 2020 | Accepted: 25 Jun 2020
1. INTRODUCTION
Spherically symmetric solutions of Einstein’s equations have attracted the interests of researchers working in this field.
This is because phenomena such as black-holes, neutron stars etc. have been found in the class of solutions [1-8] with
spherical symmetry. As a result many such solutions have been discovered. In contrast, solutions with cylindrical
symmetry are much less studied. Whereas in the case of spherical symmetry, one is to solve for two unknown metric
functions in the case of cylindrical symmetry, other is to solve for three unknown metric functions. According to
Birkhoff’s theorem, exterior solution of any spherically symmetric source is uniquely determined by the Schwarzschild
solution. On the other hand, there is no Birkoff’s theorem for space-times with cylindrical symmetry. There exist infinitely
many exterior solutions for space-times with cylindrical symmetry. Any cylindrically symmetric interior solution must be
joined with any one of those exterior solutions at the boundary of the cylindrically symmetric source. The general form of
the spherically symmetric space-time metric in spherical coordinates is unique. In contrast, general form of the
cylindrically symmetric space-time metric is not unique so that the same solution may look different for different
definitions of the radial coordinate in a cylindrical coordinate system.
The rest of this paper is organized in the following way. In Section-2, we write down the general form of the static
cylindrically symmetric space-time metric in the tangential gauge and derive all vacuum solutions. In Section-3, we write
down the form of the metric in the arc-length gauge and derive some perfect fluid solutions for specific choices of one of
the three metric functions. This is mainly a review. In Section-4, we have derived the field equations in the tangential
www.iaset.us [email protected]
2 M. A. Kauser & Q. Islam
gauge. Main result of this chapter is provided in Section-5. Here, we have shown that the field equations can be reduced to
a pair of simultaneous Riccati type differential equations whose general solution depends on the specification of a
particular solution f(r). It is also shown that, a class of physically acceptable solutions can be generated if f(r) satisfies the
condition f (0) = 0. An algorithm is provided for generating all such solutions. In Section-6, the algorithm is illustrated by
generating a new realistic solution. Finally in Section-7, some concluding remarks are given.
ds 2 = −e 2ϕ ( r ) dt 2 + e 2 Λ ( r ) dr 2 + e 2Ω ( r ) dθ 2 + e 2ψ ( r ) dz 2 (1)
Here z denotes the coordinate along the central axis of symmetry, r denotes the radial coordinate which has value
zero on the axis of symmetry and which increases as one moves away from the axis of symmetry, θ denotes the
coordinate which measures an angle around the axis of symmetry.
r ′ such that (r ′) = e
2 2Ω ( r )
Let us choose the gauge by defining the new radial coordinate . Then the metric (1)
reduces to
ds 2 = −e 2ϕ ( r ) dt 2 + e 2 Λ ( r ) dr 2 + r 2 dθ 2 + e 2ψ ( r ) dz 2 (2)
Where the prime on r has been omitted. The form in which metric (2) is written is called tangential gauge. There
are many different conventions for defining the radial coordinate. For example, the radial coordinate can be defined in such
a way that metric (1) takes the form
Rαβ = 0
We are interested to find vacuum solutions. In this case, Einstein’s equations reduce to . We find it
Rαβ
convenient to work in the tangential gauge in which the metric has the form (2). Nonzero components of for the
metric (2) are given by [9],
ϕ′
R00 = e 2(ϕ − Λ ) ϕ ′′ + (ϕ ′) 2 − ϕ ′Λ′ + + Ψ′ϕ ′
r (4)
Λ′
Rrr = −ϕ ′′ − (ϕ ′) 2 + ϕ ′Λ′ + − Ψ′′ − (Ψ′) 2 + Λ′Ψ′
r (5)
Rθθ = e −2 Λ r (Λ ′ − ϕ ′ − Ψ ′)
(6)
Ψ′
Rzz = −e 2 ( Ψ − Λ ) Ψ′′ + (Ψ′) 2 − Ψ′Λ′ + Ψ′ϕ ′ +
r (7)
Rαβ = 0
Where primes denote differentiations with respect to r. Einstein’s vacuum field equations then provide
the following system of equations,
Λ′ = ϕ ′ + Ψ ′ (8)
ϕ′
ϕ ′′ + =0
r (9)
Ψ′
Ψ ′′ + =0
r (10)
ϕ′ Ψ′
ϕ ′ Ψ′ + + =0
r r (11)
Thus we have four equations for three unknown functions. The system of equations (8) - (10) has the solution
a +b
ϕ = log(c1r a ) , Ψ = log(c2 r b ) , Λ = log(c3r )
(12a, b, c)
ab + a +b = 0 (13)
Therefore all vacuum static cylindrically symmetric solutions of Einstein’s equations are given by
ds 2 = −c1 r 2 a dt 2 + c3 r 2( a + b ) dr 2 + r 2 dθ 2 + c2 r 2b dz 2
2 2 2
(14)
2 2 2
Subject to the constraint (13). The constants
c1 and c 2 can be absorbed by rescaling t and z. The constant c3
can be absorbed by rescaling r, which affects the dθ 2 term by bringing out another constant k 2 in its coefficient. The
constant k 2 can be absorbed by rescaling θ which redefines its range from 0 to some angle α ∗ . Therefore, all vacuum
static cylindrically symmetric solutions of Einstein’s equations in the tangential gauge are given by
ds 2 = −r 2 a dt 2 + r 2( a +b ) dr 2 + r 2 dθ 2 + r 2b dz 2 (15)
Where a and b are restricted by the constraint (13) and where 0 ≤ θ ≤ α ∗ , where α ∗ may or may not be equal
2π . If a = b = 0, constraint (13) is satisfied. Then (15) reduces to
ds 2 = −dt 2 + dr 2 + r 2 dθ 2 + dz 2 (16)
Metric (16) represents the ordinary Minkowski space if it happens that α ∗ = 2π . Otherwise it represents a cone
solution.
Any interior cylindrically symmetric solution must be joined smoothly with any one of the 1-parameter family of
solutions (15) at the boundary of the cylindrically symmetric source.
www.iaset.us [email protected]
4 M. A. Kauser & Q. Islam
Previously, perfect fluid solutions with cylindrical symmetry have been discussed, among others, by Evan [10], Bronnikov
[11] and Sharif [12]. Evan and Bronnikov used various equations of state which can be written as
ρ =γ p for specific
values of
γ as well as energy conservation equation. Sharif found some solutions by specifying one of the three metric
functions. They used the form of the metric in the arc-length gauge. In Section-4.4, we will show that, using the form of
the metric in the tangential gauge, it is possible to find all static cylindrically symmetric perfect fluid solutions in closed
form. Before that, let us see how solutions are derived in [12] using the form of the metric in the arc-length gauge, which
we rewrite for convenience in the following way,
ds 2 = −eγ ( r ) dt 2 + dr 2 + eλ ( r ) dθ 2 + e µ ( r ) dz 2 (17)
If the matter content is perfect, fluid Einstein’s equations provide the system of equations [11],
″ λ µ ″
λ2 −
µ
− λ ′µ ′
8π ρ = −e e − e e 2 −
2
2
4
(18)
″ µ γ ″
µ2
−
γ
− µ ′γ ′
8π p = e e + e e 2 +
2 2
4
(19)
γ ″
γ λ λ ″
− − λ ′γ ′
8π p = e 2 e 2 + e 2 e 2 +
4
(20)
8π p =
1
(λ ′µ ′ + µ ′γ ′ + λ ′γ ′)
4 (21)
µ λ
Where
y = e 2 , z = e 2 and primes denote derivatives with respect to r. From (22b) and (22c) we get
y′ = k1 z , z′ = k2 y (23a, b)
y′′ z′′
= = k1k2
y z (24)
Equation (25) can be solved for y. Knowing y(r), z(r) can be found from (23b). Four cases arise, (i)
k1k2 < 0, (ii)
µ
e 2 = k3 cos( − k1k 2 r + k 4 )
λ
k2
e 2 = k3 − sin( − k1k 2 r + k 4 )
k1
µ
e 2 = k4 cosh( k1k2 r )
λ
k2
e 2 = k4 sinh( k1k2 r + k5 )
k1
(iii)
e 2 = c1r + c2
λ
e 2 = c3
(iv)
e 2 = c3
λ
e 2 = c1r + c2
3 1
ρ =− k1k 2 p = k1k 2
8π , 8π ,
ρ + 3p = 0
density. If
k1k 2 = 0, the solution is trivial.
www.iaset.us [email protected]
6 M. A. Kauser & Q. Islam
′
γ µ ′
2 2
e e = 0
(26)
µ µ ″ γ γ ″
− −
e 2 e 2 + e 2 e 2 = 0
(27)
′
−µ γ ′
2 2
e e = 0
(28)
In this case
ρ and
p are given by
1
ρ= k1k 2e −γ = p
8π
We are interested in finding static cylindrically symmetric internal solutions when the matter content is perfect fluid. For
this, we are to solve the equation
1
Rµν − g µν R = 8π Tµν
2 (29)
We find it convenient to use the form of the metric in the tangential gauge which we rewrite for convenience,
ds 2 = −e 2ϕ ( r ) dt 2 + e 2 Λ ( r ) dr 2 + r 2 dθ 2 + e 2ψ ( r ) dz 2 (2)
1
G µν = Rµν − g µν R
Nonzero components of the Einstein’s tensor 2 for the metric (2) are given by [9]
Λ′ Ψ′
Gtt = e 2(ϕ − Λ ) − Ψ ′′ − Ψ ′ 2 + ψ ′Λ ′ + −
r r
ϕ′ Ψ′
Grr = Ψ ′ϕ ′ + +
r r
(
Gθθ = r 2 e −2 Λ ϕ ′′ + ϕ ′ 2 − ϕ ′Λ ′ + Ψ ′′ + Ψ ′ 2 − Ψ ′Λ ′ + Ψ ′ϕ ′ )
Λ′ ϕ ′
G zz = e 2( Ψ − Λ ) ϕ ′′ + ϕ ′ 2 − ϕ ′Λ ′ − +
r r
Tµν = ( ρ + p )u µ uν − pg µν
Tµν
Nonzero components of are found to be
Λ′ Ψ ′
8π ρ = e − 2 Λ − Ψ ′′ − Ψ′2 + ψ ′Λ′ + −
r r (30)
ϕ ′ Ψ′
8π p = e − 2 Λ Ψ ′ϕ ′ + +
r r (31)
(
8π p = e −2 Λ ϕ ′′ + ϕ ′ 2 − ϕ ′Λ ′ + Ψ ′′ + Ψ ′ 2 − Ψ ′Λ ′ + Ψ ′ϕ ′ ) (32)
Λ′ ϕ ′
8π p = e − 2 Λ ϕ ′′ + ϕ ′ 2 − ϕ ′Λ ′ − +
r r (33)
We have only four equations (30) - (33) for the five unknown functions
ρ (r ) , p(r ) , ϕ (r ) , Λ(r ) and
Ψ (r ) .
In the following, we provide a formalism for obtaining all static cylindrically symmetric perfect solutions which
depends on the specification of a single solution generating function. However, not all specifications of the generating
function can generate physically acceptable solution. For obtaining physically acceptable solutions, the generating function
is required satisfy some conditions.
ϕ′ Λ′
Ψ ′′ + Ψ ′2 + Ψ ′ϕ ′ − − Ψ′Λ′ + =0
r r
ψ′ Λ′
ϕ ′′ + ϕ ′2 − Ψ′ϕ ′ − − ϕ ′Λ′ − =0
r r
These can be rearranged as
ϕ ′ − Λ′
Ψ ′′ = − (ϕ ′ − Λ ′)Ψ ′ − Ψ ′ 2
r (34)
www.iaset.us [email protected]
8 M. A. Kauser & Q. Islam
Ψ′ + Λ′
ϕ ′′ = + (Ψ ′ + Λ ′)ϕ ′ − ϕ ′ 2
r (35)
1
Ψ′ =
Equations (34) and (35) are Riccati type differential equations. Also r is a particular solution of (34) for
any choice of
ϕ ′ − Λ′ . The general solution of a Riccati type differential equation can be obtained if one particular
solution is known. Hence the general solution of (34) can be found. Equation (35) has no such particular solution.
rΨ′′ + rΨ ′2
Λ′ = ϕ ′ −
1 − rΨ ′ (36)
rϕ ′′ + rϕ ′ 2
Λ′ = − Ψ′
1 + rϕ ′ (37)
ϕ ′ − rϕ ′′ rΨ ′′ − Ψ ′ + 2Ψ ′ 2
=
1 + rϕ ′ 1 − rΨ ′ = f (r ) , say (38)
rf − 1 f
ϕ ′′ + ϕ′ = −
r r (39)
rf − 1 f
Ψ ′′ + Ψ ′ = − 2Ψ ′ 2
r r (40)
Λ ′ = ϕ ′ − Ψ ′ − f (r ) (41)
1
Ψ′ =
Given f(r), (39) can be solved for
ϕ′ as it is linear in
ϕ ′ . Since r is a particular solution of the of the
that
ϕ (0) = Ψ (0) = Λ(0) = 0 so that the metric coefficients are equal to 1 along the axis r = 0. We also require
Ψ ′(0) = ϕ ′(0) = 0 to obtain smooth solutions along r = 0. From (4.38) we see that if ϕ ′′(0) and Ψ′′(0) are finite then
Ψ ′(0) = ϕ ′(0) = 0 implies f(0) = 0. In that case (4.41) shows that Λ ′(0) = 0 . Therefore a class of physically
acceptable solutions may be generated by choosing f(r) such that f(0) = 0. Inspired by this, we now provide the following
algorithm for generating physically acceptable solutions.
THE ALGORITHM
g ′(r )
f (r ) =
Let g (r )
r g ′(r )
ϕ ′(r ) = c − ∫ r 2 dr
g (42)
dr
2r 2 g ∫
+1
r3g
Ψ ′(r ) =
dr
2r 3 g ∫ 3
r g (43)
dr
2r 2 g ∫ +1
r3g
r g ′(r ) dr g ′(r )
Λ ′(r ) = ∫ r 2 dr
c − 2r 3 g ∫ 3 −
g - r g g (r ) (44)
Using the algorithm outlined above, we have found a new solution in Section-6.
6. NEW SOLUTION
Let us choose
g (r) = k = constant.
cr cr
Ψ ′(r ) = Λ ′(r ) = ar − 2
ϕ ′(r ) = ar , cr − 1 ,
2
cr − 1
ar 2 1
ϕ (r ) = + b1 Ψ (r ) = log(1 − cr 2 ) + b2
2 , 2 ,
www.iaset.us [email protected]
10 M. A. Kauser & Q. Islam
Λ(r ) =
2
[
1 2
]
ar − log(1 − cr 2 ) + b1 − b2
The conditions
ϕ (0) = Ψ (0) = Λ(0) = 0 are satisfied if
b1 = b2 = 0 . Clearly Ψ ′(0) = ϕ ′(0) = 0 . Hence,
3c − a
ρ (r ) =
8π e ar
2
a − c − 2acr 2
p(r ) =
8π e ar
2
Both
ρ (r ) and p(r) are decreasing functions of r.
3c − a a−c
ρ ( 0) = p(0) =
8π , 8π
Clearly
ρ (0) > 0 and p(0) > 0 if c < a < 3c . Let r = R be the point where p(r) = 0
⇒ a − c − 2ac − 2acR 2 = 0 .
This gives
a−c
R= >0
2ac if a > 0, c > 0 and a > c, or if a <0, c <0 and a > c.
7. CONCLUSIONS
Choice of gauge (coordinate system) plays an important role in cylindrically symmetric solutions of Einstein’s equations.
We have found it convenient to use the tangential gauge. Using this gauge, we have found a new static cylindrically
symmetric perfect fluid solution. Generation all such solutions depends on the specific of a single input function f(r). A
g ′(r )
f (r ) =
class of realistic solutions is generated by choosing g (r ) where g(r) = constant.
REFERENCES
1. P. Boonserm, M. Visser and S. Weinfurtner, ‘Generating perfect fluid spheres in general relativity’, Phys. Rev. D
71, 124037 (2005).
2. D. Martin and M. Visser, ‘Algorithmic construction of static perfect fluid spheres’, Phys. Rev. D 69, 104028
(2004).
3. R. J. Adler, ‘A fluid sphere in general relativity’ J. Math. Phys. Vol. 15, No. 6, 727(1974).
4. H. A. Buchdahl, ‘General relativistic fluid spheres’ Phys. Rev. Vol. 116 No. 4, 1027(1959).
5. R. C. Tolman, ‘Static solutions of Einstein’s field equations for spheres of fluid’ Phys. Rev. 55, 364(1939).
6. K. Lake, ‘All static spherically symmetric perfect fluid solutions of Einstein’s equations’, Phys. Rev. D 67, 104015
(2003).
7. L. Herrera, J. Ospino and A. Di Prisco, ‘All static spherically symmetric anisotropic solutions of Einstein’s
equations’, Phys. Rev. D 77, 027502 (2008).
8. K. Lake, ‘Generating static spherically symmetric anisotropic solutions of Einstein’s equations from isotropic
Newtonian solutions’, Phys. Rev. D 80, 064039 (2009).
9. C. S. Trendafilova and S. A. Fulling, ‘Static solutions of Einstein’s equations with cylindrical symmetry’, Eur. J.
Phys., Vol.32, No.6, 1663(2011).
10. A B. Evan, ‘Static fluid cylinders in general relativity’, J. Phys. A: Math. Gen. Vol. 10 No. 8, 1303(1977).
11. K. A. Bronnikov, ‘Static fluid cylinders and plane layers in general relativity’, J. Phys. A: Math. Gen. Vol.12
No.2, 201(1979).
12. M. Sharif, ‘Cylindrically symmetric, static, perfect-fluid solutions of Einstein's field equations’, J. of the Kor.
Physical Soc., Vol. 37, No. 5, 624(2000).
www.iaset.us [email protected]