R

4
Embedded Into E
5
with Perfect Fluid

J. H. CALTENCO, J. LPEZ-BONILLA, R. PEA-RIVERO
Seccin de Estudios de Posgrado e Investigacin
Escuela Superior de Ingeniera Mecnica y Elctrica
Instituto Politcnico Nacional
Edif. Z, 3er Piso, Col. Lindavista, C.P. 07738 Mxico, D.F.
Email: [email protected]
ABSTRACT: We consider spacetimes with perfect fluid embedded into E
5
, and it is showed that the
Newman-Penrose formalism [1-3] gives simple proofs for results obtained by Krishna Rao [4] and
Barnes [5].
PACS : 04.20.-q ; 04.90. + e
1. - INTRODUCTION.
We here shall employ the notation and quantities of [3,5-11].
The curvature of the spacetime is generated by a perfect fluid via the Einstein equations:
(1)
where is the cosmological constant and are the 4-velocity, density and presion of the fluid,
respectively. The present work is dedicated to metrics which are solutions of (1) and class one (that is, they
accept embedding into E
5
[3,6]). Thus, it is convenient a brief resum of interesting results in the theme under
study, therefore:
a). Szekeres [12] and Greenberg [13] proved that when , then (1) not accepts geometries
with Petrov type N.
b). Wainwright [14] realized a complete analysis of algebraically special metrics satisfying (1), and
such that their degenerate Debever-Penrose vector [3] defines a null congruence geodesic without
shear and expansion. In particular, he deduced that:
"The Gdel metric [15] is the unique solution of (1) with Petrov type O, I and
p=0, whose repeated null congruence is geodesic with zero expansion and shear", (2)
and it is possible to show that the Gdel metric not admits embedding into E
5
[8,16-18].

c). Wainwright [19] and Carminati-Wainwright [20] investigated solutions of (1) with p = 0 and Petrov
type D.
d). The only known metric Petrov type III verifying (1), for , was constructed by Allnut [21] and
its corresponding degenerate null congruence is geodesic, has shear and expansion but without
rotation.
e). Bonnor-Davidson [22] obtained a solution of (1) with and Petrov type II, and its repeated
principal congruence is geodesic, non-zero expansion but without shear and rotation.
f). Stephani [23] used the embedding process to determine all solutions of (1) with Petrov type D,
class one and zero acceleration of the matter. Besides, Stephani suggested that any R
4
with perfect
fluid and Petrov type O has class one, which is correct only when [6].
g). Barnes [5] employed the Gauss equation to study the compatibility between the Petrov and
Churchill-Plebaski [24-28] types for the Weyl, Ricci and second fundamental tensors, thus he
showed that:
"All perfect fluids of class one have Petrov types D or O and must be
[11(11)] or [1(111)], respectively", (3)
which implies that the mentioned metrics of Allnut [21] and Bonnor-Davidson [22] do not accept embedding into
E
5
.
Besides, from (3) it is immediate the following theorem of Pokhariyal [29]:
"Any spacetime of class one whose elementary divisors of
are reals (but not simples), do not admit of a perfect fluid distribution". (4)
h). Krishna Rao [4] studied (1) when and he proved the results:
"All perfect fluids with Petrov type O and spherical symmetry,
have class one", (5)
and
"All perfect fluids with state equation , and spherical symmetry, have
Petrov type O and class one". (6)
The interior solution of Schwarzschild can be embedded into E
5
[30] due to (5). Tikekar [31] showed a metric
satisfying (6).
i). Szekeres [16] obtained two interesting theorems:
"If R
4
with perfect fluid has class one, then the fluid must be without rotation", (7)

and
"If R
4
embedded into E
5
has a perfect fluid with , then this spacetime

is of the Friedman (Petrov type O)", (8)
both theorems (7) and (8) imply that the Gdel metric [15] (perfect fluid with rotation and Petrov type D) has not of
class one [8,17,18,30].
Goldman-Rosen [32] have used perfect fluids embedded into E
5
to construct cosmological models in
general relativity.
In the next Section we shall employ the Newman-Penrose (NP) formalism [1-3] for the analysis of
metrics verifying (1) with class one, resulting thus one relation of Krishna Rao [4] and a part of the
theorem (3) of Barnes [5].
2.- Equations of Gauss and Codazzi.
A spacetime accepts isometric and local embedding into E
5
if and only if there is the second fundamental form
satisfying the equations [3,6]:
Gauss (9)
Codazzi (10)
where represents the curvature tensor, and ;r means the covariant derivative.
Collinson [17,33] used (9) to demonstrate the identity:
(11)
for any R
4
of class one, where is the simple dual of Riemann tensor, is the Levi-Civita tensor and
is an invariant of Lanczos [34,35] in terms of the double dual . The projection of (11) onto
a null tetrad of NP leads to a set of 14 equations (which are explicitly in [36]) for the quantities and (NP
components of the Weyl and Ricci, tensors respectively), and their complex conjugates y .
We now shall employ the NP formalism to study perfect fluids of class one. Thus, we construct an orthonormal
real tetrad with , that is, the unitary temporal vector coincides with the 4-velocity of the fluid. We define
the NP tetrad in the usual manner ( ):
(12)
If we project (1) onto (12), then:
(13)
and the another are zero; we suppose that because if then R
4
should be an
Einstein type spacetime, but it is known [16] that these space must be the DeSitter model.
When we use (13) in the 14 NP equations [36] (which are equivalent to (11)) result the restrictions:

,
(14)
,
,
which imply that:
, (15)
then the system (14) adopts the form:
(16)
and thus we have two possibilities:
A).
The equations (14) and (16) lead to with:
(17)
relation obtained by Krishna Rao [4] for the case of spherical symmetry, we here do not need some symmetry.
The condition (17) permits the options (Petrov type O) and (Petrov type D).
B).
Then (14) and (16) imply the relations:
(18)
which jointly with algorithms [3,37-39] to determine the Petrov type give us the type D.
The expressions (17) and (18) are the only alternatives for a perfect fluid, thus we have obtained a part of the
theorem:
"All perfect fluids (with any cosmological constant) of class one
are type D or conformally flat", (19)
shown by Barnes [5] via the Churchill-Plebaski algebraical classification [24-28], we here do not need such
classification.
References.
1. E.T. Newman and R. Penrose, J. Math. Phys. 3 (1962) 566.
2. S.J. Campbell and J. Wainwright, Gen. Rel. Grav. 8 (1977) 987.
3. D. Kramer, H. Stephani, E. Herlt and M. MacCallum, Exact solutions of Einsteins field equations,
Cambridge University Press (1980).
4. J. Krishna Rao, Gen. Rel. Grav. 3 (1972) 211.
5. A. Barnes, Gen. Rel. Grav. 5 (1974) 147.
6. H.F. Goenner, General relativity and gravitation, Vol. I, Ed. A. Held, Plenum NY (1980) Cap.14.
7. R. Fuentes, J. Lpez-Bonilla, T. Matos and G. Ovando, Gen. Rel. Grav. 21 (1989) 777.
8. G. Gonzlez, J. Lpez-Bonilla and M. Rosales, Pramana J. Phys. 42 (1994) 85.
9. J. Lpez-Bonilla, J. Rivera and H. Yee, Braz. J. Phys. 25 (1995) 80.
10. J. Lpez-Bonilla and H. Nez, Pramana J. Phys. 46 (1996) 219.
11. J. Lpez-Bonilla, J. Morales and G. Ovando, Indian J. Phys. B74 (2000) 397.
12. P. Szekeres, J. Math Phys. 7 (1966) 751.
13. P. Greenberg, Stud. Appl. Math. 51 (1972) 415.
14. J. Wainwright, Commun. Math. Phys. 17 (1970) 42.
15. K. Gdel, Rev. Mod. Phys. 21 (1949) 447.
16. P. Szekeres, Nuovo Cim. A43 (1966) 1062.
17. J. Lpez-Bonilla, J. Morales and, M. Rosales, Braz. J. Phys. 24 (1994) 522.
18. J. Lpez-Bonilla, G. Ovando and J. Rivera, Aligarh Bull. Math. 17 (1997-98) 63.
19. J. Wainwright, Gen. Rel. Grav. 8 (1977) 797.
20. J. Carminati and J. Wainwright, Gen. Rel. Grav. 17 (1985) 853.
21. J. A. Allnut, Gen. Rel. Grav. 13 (1981) 1017.
22. W. B. Bonnor and W. Davidson, Class. Quantum Grav. 2 (1985) 775.
23. H. Stephani, Commun. Math. Phys. 9 (1968) 53.
24. R. V. Churchill, Trans. Am. Math. Soc. 34 (1932) 784.
25. J. Plebaski, Acta Phys. Polon. 26 (1964) 963.
26. J. Plebaski and J. Stachel, J. Math. Phys. 9 (1968) 269.
27. H.F. Goenner and J. Stachel, J. Math. Phys. 11 (1970) 3358.
28. C.B.G. McIntosh, J.M. Foyster and A.W.C. Lun, J. Math. Phys. 22 (1981) 2620.
29. G.P. Pokhariyal, Gen. Rel. Grav. 3 (1972) 87.
30. C.D. Collinson, J. Math Phys. 9 (1968) 403.
31. R.S. Tikekar, Curr. Sci. 39 (1970) 460.
32. J. Goldman and N. Rosen, Gen. Rel. Grav. 2 (1971) 367.
33. C.D. Collinson, Commun. Math. Phys. 8 (1968) 1.
34. C. Lanczos, Ann. of Math 39 (1938) 842.
35. V. Gaftoi, J. Lpez-Bonilla, D. Navarrete and G. Ovando, Rev. Mex. Fs. 36 (1990) 503.
36. D. Ladino and J. Lpez-Bonilla, Rev. Mex. Fs.35 (1989) 623.
37. R.A. DInverno and R.A. Russel- Clark, J. Math. Phys.12 (1971) 1258.
38. F.W. Letniowski and R.G. McLenaghan, Gen. Rel. Grav. 20 (1988) 463.
39. G. Ares de Parga, O. Chavoya, J. Lpez-Bonilla and G. Ovando, Rev. Mex. Fs. 35 (1989) 201.