# Constant in SET Definition

1. Jan 30, 2013

### elfmotat

So I was looking through Wald when I noticed his definition of the stress-energy for an arbitrary matter field:

$$T_{ab}=-\frac{\alpha_M}{8\pi} \frac{1}{ \sqrt{-g}} \frac{\delta S_M}{\delta g^{ab}}$$

where $S_M$ is the action for the particular type of matter field being considered, and $\alpha_M$ is some constant that determines the form of the Lagrangian for the coupled Einstein-matter field equations:

$$\mathcal{L}=R\sqrt{-g}+\alpha_M \mathcal{L}_M$$

For example, for a Klein-Gordon field we take $\alpha_{KG}=16\pi$, and for an EM field we take $\alpha_{EM}=4$. Now, my question is whether or not there is some prescription for finding the value of $\alpha_M$. How could I go about finding $\alpha_M$ for an arbitrary $\mathcal{L}_M$?

I feel like I'm missing something painfully obvious.

2. Jan 31, 2013

### Bill_K

I guess everybody has his own conventions. The usual ones follow.

The Einstein Equations are Gμν = 8πG Tμν. To get this equation we use an action I = IG + IM where IG = (1/16πG) ∫√-g R d4x and Tμν = (2/√-g) δIM/δgμν.

For electromagnetism, L = (-1/4)FμνFμν. This is in Heaviside units where e2/4πħc = 1/137. In Gaussian units where e2/ħc = 1/137, the Lagrangian would instead be IM = (-1/16π)FμνFμν.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook