3
$\begingroup$

A proof I'm reading tries to evaluate the integral (where $i$ is the regular imaginary unit)

$$\int_{-\infty}^{\infty} e^{-(x-\alpha i)^2}\mathrm{d}x$$

by doing a substitution $u=x-\alpha i$. Normally, one would also have to change the bounds of integration.

$$\int_{-\infty+\alpha i}^{\infty+\alpha i} e^{-x^2}\mathrm{d}x$$

But this proof leaves the bounds as +/- infinity.

$$\int_{-\infty}^{\infty} e^{-x^2}\mathrm{d}x$$

Why is this valid?

$\endgroup$
  • 1
    $\begingroup$ Integrate $e^{-z^2}$ around the rectangle with vertices at $\pm R$ and $-ai \pm R$, then send $R \to \infty$. Show that the integrals over the vertical sides of the rectangle vanish, then conclude that the remaining integrals over the top and bottom must be equal. $\endgroup$ – Antonio Vargas Jan 22 '14 at 21:34
  • $\begingroup$ A related problem. $\endgroup$ – Mhenni Benghorbal Jan 23 '14 at 4:26
6
$\begingroup$

Consider

$$\oint_C dz \, e^{-z^2}$$

where $C$ is a rectangle having vertices $-R$, $R$, $R+i \alpha$, $-R+i \alpha$. By Cauchy's Theorem, this integral is zero. On the other hand, it is also equal to

$$\int_{-R+i \alpha}^{R+i \alpha} dx \, e^{-x^2} + i \int_{\alpha}^0 dy \, e^{-(R+i y)^2} -\int_{-R}^R dx \, e^{-x^2} -i \int_0^{\alpha} dy \, e^{-(-R + i y)^2} $$

As $R\to\infty$, the 2nd and 4th integrals vanish because each integral is bounded by the value

$$e^{-R^2} \int_0^{\alpha} dy \, e^{y^2} \le |\alpha| e^{-(R^2-\alpha^2)}$$

Thus, we are left with the equality

$$\int_{-\infty}^{\infty} dx \, e^{-x^2} = \int_{-R+i \alpha}^{R+i \alpha} dx \, e^{-x^2}$$

as was to be shown.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.