next up previous contents index
Next: Proof of Covariance Up: Dirac Equation Previous: Free Motion of a

Covariant Form of the Dirac Equation

We now cast the Dirac equation into a more apparent covariant form. Multiplying (5.3) by $\beta/c$ and defining


\begin{displaymath}
\gamma^0\equiv\beta \quad\textrm{and}\quad
\gamma^i\equiv\beta\alpha_i,
\end{displaymath} (5.96)

where $i=1,2,3$, we have


$\displaystyle i\hbar \left(
\gamma^0\frac{\partial}{\partial x^0} +
\gamma^1\fr...
...al}{\partial x^2} +
\gamma^3\frac{\partial}{\partial x^3}
\right) \psi - mc\psi$ $\textstyle =$ $\displaystyle 0 ,$  
$\displaystyle \left( i\hbar \gamma^\mu\frac{\partial}{\partial x^\mu}
\right) \psi - mc\psi$ $\textstyle =$ $\displaystyle 0 ,$ (5.97)

where


\begin{displaymath}
\gamma^\mu\frac{\partial}{\partial x^\mu} = \frac{\gamma^0}{c}
\frac{\partial}{\partial t} + \vec{\gamma}\cdot\vec{\nabla} .
\end{displaymath} (5.98)

In terms of the momentum operator we write


\begin{displaymath}
(\gamma^\mu \hat{p}_\mu - mc)\psi = 0 .
\end{displaymath} (5.99)

Introducing the Feynman dagger, or slash notation, for 4-vector $A_\mu$, we have


\begin{displaymath}
\not{\!\!A} \equiv \gamma^\mu A_\mu = g_{\mu\nu}\gamma^\mu A^\nu =
\gamma^0A^0 - \vec{\gamma}\cdot\vec{A} .
\end{displaymath} (5.100)

Also notice that


\begin{displaymath}
\not{\!\!A}\not{\!\!A}= A^2 .
\end{displaymath} (5.101)

We write


\begin{displaymath}
\fbox{$\displaystyle
(\not{\!\hat{p}} - mc)\psi = 0
$}\ .
\end{displaymath} (5.102)

We introduce the electromagnetic interaction by the usual minimal substitution


\begin{displaymath}
\left( \not{\!p} - \frac{e}{c}\not{\!\!A} - mc \right) \psi = 0.
\end{displaymath} (5.103)

Let us study the properties of the $\gamma^\mu$ matrices.


$\displaystyle \alpha_i\alpha_k + \alpha_k\alpha_i$ $\textstyle =$ $\displaystyle 2\delta_{ik}$  
$\displaystyle \beta\alpha_i\alpha_k + \beta\alpha_k\alpha_i$ $\textstyle =$ $\displaystyle 2\beta\delta_{ik}$  
$\displaystyle -\alpha_i\beta\alpha_k - \alpha_k\beta\alpha_i$ $\textstyle =$ $\displaystyle 2\beta\delta_{ik}$  
$\displaystyle -\beta\alpha_i\beta\alpha_k - \beta\alpha_k\beta\alpha_i$ $\textstyle =$ $\displaystyle 2\delta_{ik}$  
$\displaystyle \gamma^i\gamma^k + \gamma^k\gamma^i$ $\textstyle =$ $\displaystyle -2\delta_{ik} .$ (5.104)

And


$\displaystyle \alpha_i\beta + \beta\alpha_i$ $\textstyle =$ $\displaystyle 0$  
$\displaystyle \beta\alpha_i\beta + \beta\beta\alpha_i$ $\textstyle =$ $\displaystyle 0$  
$\displaystyle \gamma^i\gamma^0 + \gamma^0\gamma^i$ $\textstyle =$ $\displaystyle 0 .$ (5.105)

Since


\begin{displaymath}
\beta^2 = 1 \Rightarrow \gamma^0\gamma^0 + \gamma^0\gamma^0 = 2 .
\end{displaymath} (5.106)

Therefore


\begin{displaymath}
\fbox{$\displaystyle
\gamma^\mu\gamma^\nu + \gamma^\nu\gamma^\mu = 2g^{\mu\nu}
$}\ ,
\end{displaymath} (5.107)

where the matrices are $4\times 4$ and $\mu, \nu=0,1,2,3$. Although the Dirac matrices $\gamma^\mu$ are written with Greek indices, they are not four vectors. Rather, they have the same value in every frame.

The $\gamma^i$ matrices are anti-hermitian and $\gamma^0$ is hermitian:


$\displaystyle (\gamma^i)^\dagger$ $\textstyle =$ $\displaystyle (\beta\alpha_i)^\dagger =
\alpha_i^\dagger\beta^\dagger = \alpha_i\beta = -\beta\alpha_i =
-\gamma^i ,$ (5.108)
$\displaystyle (\gamma^0)^\dagger$ $\textstyle =$ $\displaystyle \beta^\dagger = \beta = \gamma^0 .$ (5.109)

This can be summarized by writing


\begin{displaymath}
(\gamma^\mu)^\dagger = \gamma^0 \gamma^\mu \gamma^0 .
\end{displaymath} (5.110)

Using our previous representation (equation 5.10),


$\displaystyle \gamma^i$ $\textstyle \equiv$ $\displaystyle \beta\alpha_i = \left( \begin{array}{cc}
I & 0 \\  0 & -I
\end{ar...
...) = \left( \begin{array}{cc}
0 & \sigma_i \\  -\sigma_i & 0
\end{array}\right),$ (5.111)
$\displaystyle \gamma^0$ $\textstyle \equiv$ $\displaystyle \beta = \left(
\begin{array}{cc}
I & 0 \\  0 & -I
\end{array}\right) .$ (5.112)


next up previous contents index
Next: Proof of Covariance Up: Dirac Equation Previous: Free Motion of a
Douglas M. Gingrich (gingrich@ ualberta.ca)
2004-03-18