A theoretical motivation for general relativity, including the motivation for the geodesic equation and the Einstein field equation, can be obtained from special relativity by examining the dynamics of particles in circular orbits about the Earth. A key advantage in examining circular orbits is that it is possible to know the solution of the Einstein Field Equation a priori. This provides a means to inform and verify the formalism.
General relativity addresses two questions:
The former question is answered with the geodesic equation. The second question is answered with the Einstein field equation. The geodesic equation and the field equation are related through a principle of least action. The motivation for the geodesic equation is provided in the section Geodesic equation for circular orbits. The motivation for the Einstein field equation is provided in the section Stress–energy tensor.
__TOC__
See main article: Geodesics in general relativity.
For definiteness consider a circular Earth orbit (helical world line) of a particle. The particle travels with speed v. An observer on Earth sees that length is contracted in the frame of the particle. A measuring stick traveling with the particle appears shorter to the Earth observer. Therefore, the circumference of the orbit, which is in the direction of motion appears longer than
\pi
In special relativity the 4-proper-velocity of the particle in the inertial (non-accelerating) frame of the earth is
u=\left(\gamma,\gamma{v\overc}\right)
where c is the speed of light,
v
\gamma
\gamma={1\over\sqrt{{1-{{v ⋅ v}\overc2}}}}
The magnitude of the 4-velocity vector is always constant
u\alphau\alpha=-1
where we are using a Minkowski metric
η\mu\nu=η\mu\nu=\begin{pmatrix} -1&0&0&0\\ 0&1&0&0\\ 0&0&1&0\\ 0&0&0&1 \end{pmatrix}
The magnitude of the 4-velocity is therefore a Lorentz scalar.
The 4-acceleration in the Earth (non-accelerating) frame is
a\equiv{{du}\over{d\tau}}={d\over{d\tau}}{\left(\gamma,\gamma{v\overc}\right)}={\left(0,\gamma2{a\overc2}\right)}={\left(0,-\gamma2{{v ⋅ v}\overc2}{{r}\overr2}\right)}
where
d\tau
cdt=\gammad\tau
Here, the 3-acceleration for a circular orbit is
a=-\omega2r=-{v ⋅ v}{{r}\overr2}
where
\omega
r
The magnitude of the 4-velocity is constant. This implies that the 4-acceleration must be perpendicular to the 4-velocity. The inner product of the 4-acceleration and the 4-velocity is therefore always zero. The inner product is a Lorentz scalar.
The equation for the acceleration can be generalized, yielding the geodesic equation
{{du\mu
{{du\mu
where
x\mu
{R\mu
{R\mu
where
{\delta\mu
u\alphau\alpha=-1
and
a\alphau\alpha=0
It is easily verified that circular orbits satisfy the geodesic equation. The geodesic equation is actually more general. Circular orbits are a particular solution of the equation. Solutions other than circular orbits are permissible and valid.
See main article: Ricci curvature and Scalar curvature. The Ricci curvature tensor is a special curvature tensor given by the contraction
R\alpha\equiv{R\nu
The trace of the Ricci tensor, called the scalar curvature, is
R\equiv{R\alpha
Consider the situation in which there are now two particles in nearby circular polar orbits of the Earth at radius
r
v
The particles execute simple harmonic motion about the Earth and with respect to each other. They are at their maximum distance from each other as they cross the equator. Their trajectories intersect at the poles.
Imagine a spacecraft co-moving with one of the particles. The ceiling of the craft, the
\acute{z
r
\acute{x}
\acute{y}
\acute{x}\mu
{{d2\acute{x}\mu
where
\acute{u}\mu={{d\acute{x}\mu
and
\acute{{R}\mu
is the curvature tensor in the local frame.
The equation of motion for a particle in flat spacetime and in the absence of forces is
{{d{u}\mu
If we require a particle to travel along a geodesic in curved spacetime, then the analogous expression in curved spacetime is
{{D\acute{u}\mu
where the derivative on the left is the covariant derivative, which is the generalization of the normal derivative to a derivative in curved spacetime. Here
{\Gamma\mu
is a Christoffel symbol.
The curvature is related to the Christoffel symbol by
\acute{{R}\mu
The interval in the local frame is
ds2=dx2+dy2+dz2-c2dt2\equivg\mud\acute{x}\mud\acute{x}\nu
=d\acute{x}2+d\acute{y}2+d\acute{z}2-c2d\acute{t}2+2\gamma\cos(\theta)\cos(\phi)vd\acute{t}d\acute{x}+2\gamma\cos(\theta)\sin(\phi)vd\acute{t}d\acute{y}-2\gamma\sin(\theta)vd\acute{t}d\acute{z}
where
\theta
z
\phi
x
This gives a metric of
g\mu\nu=\begin{pmatrix} -1&\gamma\cos(\theta)\cos(\phi)
v | |
c |
&\gamma\cos(\theta)\sin(\phi)
v | |
c |
&-\gamma\sin(\theta)
v | |
c |
\\ \gamma\cos(\theta)\cos(\phi){
v | |
c |
in the local frame.
The inverse of the metric tensor
g\mu
g\mug\alpha=
\nu | |
\delta | |
\mu |
where the term on the right is the Kronecker delta.
The transformation of the infinitesimal 4-volume
d\Omega
d\acute{\Omega}=\sqrt{-g}d{\Omega}
where g is the determinant of the metric tensor.
The differential of the determinant of the metric tensor is
dg=gg\mudg\mu=-gg\mudg\mu
The relationship between the Christoffel symbols and the metric tensor is
{{\Gamma}\alpha
{\Gamma}\beta={
1 | |
2 |
See main article: Einstein–Hilbert action.
The principle of least action states that the world line between two events in spacetime is that world line that minimizes the action between the two events. In classical mechanics the principle of least action is used to derive Newton's laws of motion and is the basis for Lagrangian dynamics. In relativity it is expressed as
S=
2 | |
\int | |
1 |
l{L}d\Omega
between events 1 and 2 is a minimum. Here S is a scalar and
l{L}
is known as the Lagrangian density. The Lagrangian density is divided into two parts, the density for the orbiting particle
l{L}p
l{L}e
l{L}=l{L}p+l{L}e
In curved spacetime, the "shortest" world line is that geodesic that minimizes the curvature along the geodesic. The action then is proportional to the curvature of the world line. Since S is a scalar, the scalar curvature is the appropriate measure of curvature. The action for the particle is therefore
Sp=C
2 | |
\int | |
1 |
\acute{R}d\acute{\Omega}=C
2 | |
\int | |
1 |
{\acute{R}}\sqrt{-g}d{\Omega}=C
2 | |
\int | |
1 |
g\alpha\acute{R}\alpha\sqrt{-g}d{\Omega}
where
C
The Lagrangian density for the particle is therefore
l{L}p=Cg\alpha\acute{R}\alpha\sqrt{-g}
The action for the particle and the Earth is
S=
2 | |
\int | |
1 |
Cg\alpha\acute{R}\alpha\sqrt{-g}d\Omega+
2 | |
\int | |
1 |
l{L}ed\Omega
Thus the world line that lies on the surface of the sphere of radius r by varying the metric tensor. Minimization and neglect of terms that disappear on the boundaries, including terms second order in the derivative of g, yields
0=\deltaS=
2 | |
\int | |
1 |
C\left(\acute{R}\alpha-{1\over2}\acute{R}g\alpha\right)\deltag\alpha\sqrt{-g}d\Omega-
2 | |
\int | |
1 |
\acute{T}\alpha\deltag\alpha\sqrt{-g}d\Omega
where[2]
\acute{T}\alpha={1\over\sqrt{-g}}\left({d\over{dx\nu}}{{\partiall{L}e}\over{\partial\left({{dg}\over{dx\nu}}\right)}}-{{\partiall{L}e}\over{\partialg}}\right)
is the Hilbert stress–energy tensor of the field generated by the Earth.
The relationship, to within an unknown constant factor, between the stress-energy and the curvature is
\acute{T}\alpha=C\left(\acute{R}\alpha-{1\over2}\acute{R}g\alpha\right)
Newton's Law of Gravitation in non-relativistic mechanics states that the acceleration on an object of mass
m
M
f={d2r\overd\tau2}=-{GM\over{c2r3}}r
where
G
r
M
m
r
\tau\equivct
The acceleration
f
m
For definiteness. consider a particle of mass
m
M
f=-{4\piG\over{3c2}}\rho(r)r
where
\rho(r)
r
Newton's law can be written
f=-{4\piG\over{3c4}}\left({Mc2\overV}\right)r
where
V
r
Mc2
r
T00=-{T0
where
\gammai\equiv{1\over{\sqrt{1-{{vi ⋅ vi}\overc2}}}}
and
vi |
mi
There are two simple relativistic entities that reduce to the 00 component of the stress–energy tensor in the nonrelativistic limit
u\alphaT\alphau\beta → T00
and the trace
T\equiv{T\alpha
where
u\alpha
The 00 component of the stress–energy tensor can be generalized to the relativistic case as a linear combination of the two terms
T00 → u\alpha\left(AT\alpha+BTη\alpha\right)u\beta
where
A+B=1
The 4-acceleration due to gravity can be written
f\mu=-8\pi{G\over{3c4}}\left({A\over2}T\alpha+{B\over2}Tη\alpha\right
\mu | |
)\delta | |
\nu |
u\alphax\nuu\beta
Unfortunately, this acceleration is nonzero for
\mu=0
f\mu → f\mu+u\muu\nuf\nu
This yields
f\mu=-8\pi{G\over{3c4}}\left({A\over2}T\alpha+{B\over2}Tη\alpha\right)\left(
\mu | |
\delta | |
\nu |
+u\muu\nu\right)u\alphax\nuu\beta
The force in the local frame is
\acute{f}\mu=-8\pi{G\over{3c4}}\left({A\over2}\acute{T}\alpha+{B\over2}\acute{T}g\alpha\right)\left(
\mu | |
\delta | |
\nu |
+\acute{u}\mu\acute{u}\nu\right)\acute{u}\alpha\acute{x}\nu\acute{u}\beta
The Einstein field equation is obtained [4] by equating the acceleration required for circular orbits with the acceleration due to gravity
a\mu=f\mu
\acute{{R}\mu
This is the relationship between curvature of spacetime and the stress–energy tensor.
The Ricci tensor becomes
\acute{R}\alpha=8\pi{G\over{c4}}\left({A\over2}\acute{T}\alpha+{B\over2}\acute{T}g\alpha\right)
The trace of the Ricci tensor is
\acute{R}=
\alpha | |
\acute{R} | |
\alpha |
=8\pi{G\over{c4}}\left({A\over
\alpha | |
2}\acute{T} | |
\alpha |
+{B\over2}\acute{T}
\alpha | |
\delta | |
\alpha |
\right)=8\pi{G\over{c4}}\left({A\over2}+2B\right)\acute{T}
Comparison of the Ricci tensor with the Ricci tensor calculated from the principle of least action, Theoretical motivation for general relativity#Principle of least action in general relativity identifying the stress–energy tensor with the Hilbert stress-energy, and remembering that A+B=1 removes the ambiguity in A, B, and C.
A=2
B=-1
and
C=\left(8\pi{G\over{c4}}\right)-1
This gives
\acute{R}=-8\pi{G\over{c4}}\acute{T}
The field equation can be written
l{G}\alpha=8\pi{G\over{c4}}\acute{T}\alpha
where
l{G}\alpha\equiv\acute{R}\alpha-{1\over2}\acute{R}g\alpha
This is the Einstein field equation that describes curvature of spacetime that results from stress-energy density. This equation, along with the geodesic equation have been motivated by the kinetics and dynamics of a particle orbiting the Earth in a circular orbit. They are true in general.
Solving the Einstein field equation requires an iterative process. The solution is represented in the metric tensor
g\mu
Typically there is an initial guess for the tensor. The guess is used to calculate Christoffel symbols, which are used to calculate the curvature. If the Einstein field equation is not satisfied, the process is repeated.
Solutions occur in two forms, vacuum solutions and non-vacuum solutions. A vacuum solution is one in which the stress–energy tensor is zero. The relevant vacuum solution for circular orbits is the Schwarzschild metric. There are also a number of exact solutions that are non-vacuum solutions, solutions in which the stress tensor is non-zero.
See main article: Solving the geodesic equations. Solving the geodesic equations requires knowledge of the metric tensor obtained through the solution of the Einstein field equation. Either the Christoffel symbols or the curvature are calculated from the metric tensor. The geodesic equation is then integrated with the appropriate boundary conditions.
See main article: Maxwell's equations in curved spacetime.
Maxwell's equations, the equations of electrodynamics, in curved spacetime are a generalization of Maxwell's equations in flat spacetime (see Formulation of Maxwell's equations in special relativity). Curvature of spacetime affects electrodynamics. Maxwell's equations in curved spacetime can be obtained by replacing the derivatives in the equations in flat spacetime with covariant derivatives. The sourced and source-free equations become (cgs units):
{4\pi\overc}Jb=\partialaFab+
a} | |
{\Gamma | |
\mua |
F\mu+
b} | |
{\Gamma | |
\mua |
Fa\equivDaFab\equiv{Fab
and
0=\partialcFab+\partialbFca+\partialaFbc=DcFab+DbFca+DaFbc
where
Ja
Fab
\epsilonabcd
{\partial\over{\partialxa}}\equiv\partiala\equiv{},a\equiv(\partial/\partialct,\nabla)
is the 4-gradient. Repeated indices are summed over according to Einstein summation convention. We have displayed the results in several common notations.
The first tensor equation is an expression of the two inhomogeneous Maxwell's equations, Gauss' law and the Ampère's law with Maxwell's correction. The second equation is an expression of the homogeneous equations, Faraday's law of induction and Gauss's law for magnetism.
The electromagnetic wave equation is modified from the equation in flat spacetime in two ways, the derivative is replaced with the covariant derivative and a new term that depends on the curvature appears.
-{A\alpha
where the 4-potential is defined such that
Fab=\partialbAa-\partialaAb
We have assumed the generalization of the Lorenz gauge in curved spacetime
{A\mu