next up previous
Next: About this document ... Up: Dirac delta function Previous: Properties of the Dirac

Representations of $\delta (x)$.

In use, the Dirac delta function is never evaluated without multiplying by a test function and integrating over some domain. Equations involving Dirac delta functions without such integrations are a convenient half-way stage that nevertheless have enormous utility. Properly speaking, the Dirac delta function is not a function at all (it is a generalized function or a functional), however it can be represented as the limit of a sequence of ordinary functions.

Representations of the Dirac delta with familiar functions allow us to visualize the Dirac delta, and many calculations involving $\delta (x)$ can be carried out with these representations. Let $F(x,\epsilon )$ be a function that has a peak near $x=0$, and the shape of the peak is controlled by parameter $\epsilon $. If the integral of $F(x,\epsilon )$ is unity, that is,

\begin{displaymath}
\int_{-\infty }^{\infty }F(x,\epsilon )\,dx=1
\end{displaymath}

for any value of parameter $\epsilon >0$, then the Dirac delta function may be represented in the limit:

\begin{displaymath}
\delta (x)=\lim_{\epsilon \rightarrow 0}\;F(x,\epsilon )
\end{displaymath}

Example representations:

  1. Top-hat function (square step).

    \begin{displaymath}
F(x,\epsilon )=\left\{
\begin{array}{cc}
1/\epsilon & -\eps...
... /2<x<\epsilon /2 \\
0 & \mbox{otherwise}
\end{array}\right.
\end{displaymath}

  2. Diffraction peak.

    \begin{displaymath}
F(x,\epsilon )=\frac{\sin (x/\epsilon )}{\pi x}
\end{displaymath}

  3. Lorentzian.

    \begin{displaymath}
F(x,\epsilon )=\frac{\epsilon /\pi }{(x^{2}+\epsilon ^{2})}
\end{displaymath}

  4. Gaussian.

    \begin{displaymath}
F(x,\epsilon )=\frac{1}{2\epsilon }\exp (-x^{2}/\epsilon ^{2})
\end{displaymath}

  5. Fourier integral. Using a standard integral, the Lorentz representation may be written

    \begin{displaymath}
F(x,\epsilon )=\frac{\epsilon /\pi }{(x^{2}+\epsilon ^{2})}=...
...\infty }\exp (-\epsilon \left\vert k\right\vert )\cos (kx)\;dk
\end{displaymath}

    and the limit can be explicitly evaluated to obtain

    \begin{displaymath}
\lim_{\epsilon \rightarrow 0}F(x,\epsilon )=\delta (x)=\frac{1}{2\pi }%
\int_{-\infty }^{\infty }\exp (ikx)\;dk
\end{displaymath}

Figure: Representations of the Dirac delta function.
\includegraphics[height=10cm]{dirac.eps}

NOTE: Although all of the above functions $F(x,\epsilon )$ are symmetric, symmetry is not essential. Non-symmetric functions produce perfectly good representations of the Dirac delta function.


next up previous
Next: About this document ... Up: Dirac delta function Previous: Properties of the Dirac
2004-01-21