The Newton's law of gravitation states that two bodies with masses m and
M located at a relative distance r attract each other with the force
F=G
mM
r2
,
(1)
where G ≈ 6.67×10−11Nm2kg−2 is the gravitational
constant measured by Cavendish.
In other words a distribution of masses with density μ creates a
gravitational potential φ([(r)\vec]) which satisfies the Poisson equation
∆φ = 4πGμ,
(2)
and a test body m in the gravitational potential φ is affected by a
force
→
F
=−m
→
∇
φ.
(3)
The last equation can be cast into a variational form with the action
S=
⌠ ⌡
dt
−mc2+
1
2
mv2−mφ
= −mc
⌠ ⌡
dt
c−
v2
2c
+
φ
c
.
(4)
Comparing with S=−mc∫ds we get (in the limit c→∞)
ds2=
1+
2φ
c2
c2dt2−d
→
r
2
.
(5)
In other words in the Newtonian limit the metric tensor can be
approximated1 by
gab=ηab+hab (where ηab is the Minkowski metric
tensor and hab is a small correction) where the g00 component is
g00=1+
2φ
c2
.
(6)
Newtonian limit of general relativity
In this limit all fields are weak and all velocities are small. Only the
00 component of the energy-momentum tensor is non-vanishing,
T00=μ where μ is the mass density of the matter.
Therefore we shall only consider the 00
component of the Einstein's equation,
R00=κ(T00−[1/2]g00T).
In the Newtonian limit g00=1 + 2φ,
Γα00=−φ,α,
R00=−φ,α,α=∆φ, where
α = 1,2,3. The Einstein equation thus turns into the
the Poisson's equation
∆φ = [1/2]κμ
which is equivalent to the Newtonian theory if we put
κ = [(8πG)/(c4)].
Gravitational waves.
In a weak gravitational field the space-time is almost flat and
the metric tensor gab is equal to the flat metric ηab plus a
small correction hab, gab = ηab+hab. The Riemann tensor
to the lowest order in hab is
Rabcd =
1
2
(had,bc + hbc,ad − hac,bd − hbd,ac).
(7)
If we choose coordinates such that
(hab − [1/2]hδab),b=0,
the Ricci tensor is simply
Rab = −
1
2
hab,c,c
(8)
and the vacuum Einstein's
equations turn into the ordinary wave equations
∂2
∂t2
− ∆
hab = 0.
(9)
The intensity of gravitational radiation by a system of slowly moving
bodies is determined by its quadrupole moment Dαβ
−
dE
dt
=
G
45c5
(D"′αβ)2
(10)
Exercises
Calculate the energy-momentum tensor Tab for a particle of mass
m with the action S=−m∫ds. Hint: calculate the variation of the
action with respect to δgab and represent it in the form
δS = −1/2∫Tabδgabds.
In the Minkowski space consider a scalar field ϕ with action
S =
⌠ ⌡
dΩ
−
1
2
ϕ,aϕ,a−
1
2
m2ϕ2
and calculate its "translation-invariance" energy-momentum tensor,
Tab=
∂L
∂ϕ,a
ϕ,b−Lδab .
Rewrite the action in a generally covariant form and calculate its
"metric-derivative" energy-momentum tensor,
1
2
√
−g
Tab =
δ(
√
−g
L)
δgab
.
Footnotes:
1 where we have neglected the terms
gαβ, αβ = 1,2,3
since their contribution to ds2 is not multiplied by c2 and
is thus neglidible compared to the contribution from g00.
File translated from
TEX
by
TTH,
version 3.77. On 6 Oct 2007, 00:05.