EnergyMomentum Tensor
Incoherent Matter Case
One of the simplest energymomentum tensors is the dust energy momentum tensor. This type of matter field consists of noninteracting incoherent matter. The matter field depends on one scalar quantity and one vector quantity. These two quantities are as follows:
(15.1)
This density is the density that would be measured by an observer moving with the flow of the dust fluid. Such an observer is called a comoving observer.
(15.2)



where is the proper time along the path that the dust particle is taking. The simplest energymomentum tensor that can be constructed from these two dust quantities is the following:
Note that this pattern of physical terms is similar to the pattern for kinetic energy in classical mechanics.
We will now look at the behaviour of this energymomentum tensor by seeing how it works in special relativity. The relation between the proper time and normal coordinate time is found by considering the Minkowski line element (coordinates given by ).
(15.4)



Rearranging terms we get
(15.5) 


where is the usual special relativity speed factor
(15.6)

. 

The symbol is the 3D velocity magnitude of the dust fluid particle.
We now take a look at the 00 component of the dust tensor.
(15.7)
This can then be written
(15.8) 


This equation is reminiscent of the relativistic mass relation where
(15.9) 


An extra factor is picked up in the case of because we are dealing with a density with volume involved. We need to transform the volume along the direction of motion of the dust particle and this brings in the factor to make .
We note that in a comoving frame we would expect the following condition.
(15.10) 


since by definition in such a frame. Thus we have
(15.11)
If a matter field of dust, with proper density and 3D velocity , flows past a fixed observer, that observer measures the density as
(15.12) 
. 

(15.13) 


We are therefore led to a physical interpretation of the 00 component of the dust energymomentum tensor as follows.
(15.14)
The other components of the dust energy momentum tensor (15.3) can be written
The divergence of the energy momentum tensor for dust leads to two very important equations involving energy and momentum. We shall now investigate how these two equations arise. The divergence equation is written as follows:
. 

Consider the case when the free index of the divergence equation above has . Substituting the matrix equation (15.15) into the above equation then gives (exercise)
(15.17)
This equation can be rewritten more succinctly as


This is just the equation of continuity of a fluid. It links up the change in time of the density with a corresponding change in space. In special relativity conservation of mass density is essentially the same as conservation of energy density we can interpret the 0 component of the divergence equation
(15.19) 


as a relativistic conservation of energy equation for the dust fluid. We now consider the other important equation that comes out of the divergence equation (15.16) by letting . We have
(15.20)
.
When combined these three equation gives the following result (exercise).
(15.21)
Using the equation of continuity (15.18) we see that this equation can be succinctly written as


In classical fluid mechanics the equivalent of Newton's 2nd Law equation of motion is called the NavierStokes Equation. It is written as
(15.23)
where is the pressure in the fluid and is called the external body force per unit mass. We see that the dust equation of motion (15.22) is just a special case of the NavierStokes equation when there's no pressure and no external forces acting on the body. From the two derived equations (15.18) and (15.22) we can conclude that the divergence equation (15.16) in special relativity is a statement of conservation of energy and conservation of linear momentum in the matter field. This explains where this tensor gets its name: energymomentum tensor. Making the simplest generalization we see that the equation of energy momentum conservation in General Relativity should be given as
(15.24) 
. 

This equation in General Relativity actually contains the geodesic equation of motion. We will now derive this fact. With dust the divergence equation becomes
(15.25) 


Using the product rule we can get the equivalent equation


Contract this equation with to give
Since we are using proper time the 4velocity is normalized such that
(15.28) 
. 

This automatically implies that
(15.29) 
. 

Thus the equation (15.27) reduces to
(15.30) 


This is a continuity type of equation and when it is substituted back into (15.26) we arrive at the result
(15.31) 


This equation implies that the vector is tangent to a geodesic. Thus geodesic motion must apply for dust fluid particles. This is made explicit by expanding this equation in terms of the proper time derivatives to give:
(15.32) 


Perfect Fluid Case
A perfect fluid is characterized by two scalar quantities (energy density and pressure) and one vector quantity (4velocity).
(15.33) 


(15.34) 


(15.35) 


In the limit that , the perfect fluid tensor should reduce to the dust tensor. It can be shown that in order to have this reduction and also give rise to the equivalent of the equation of continuity and the NavierStokes in Special Relativity the perfect fluid tensor must have the following form:
(15.36) 


Note that in terms of units
(15.37) 


and
(15.38) 


Sticking to pure General Relativity it can be shown that the conservation equations
(15.39) 
, 

when applied to the perfect fluid tensor, also give (1) an equationofcontinuitytype equation and (2) an equationofmotion equation that looks like a geodesic equation with a forcing function. This forcing function arises from the derivative of the pressure. This pressure derivative makes the fluid deviate from pure geodesic motion.
In general when we model some matterenergy distribution in General Relativity we need another equation that tells us how the matter reacts with pressure and absolute temperature. This equation is called the equation of state of the matterenergy system. It can be generically written as
(15.40) 


In many cases the physical system is in equilibrium and we can assume that . Then the equation of state for an equilibrium situation is generically written as
(15.41) 


If you are modelling a star you will need an equation of state that depends on the type of star. If you are modelling the universe in a cosmological calculation you will need an equation of state depending on what stage of the universe after the Big Bang you are considering.