1
$\begingroup$

I'm working on some past exam papers and I wanted to see if I'm thinking of the following in the correct way;

Given a contour integral $\int_{\gamma}\frac{1}{z}dz$ I want to find all of it's possible values given that $\gamma$ is the semi-circle starting at $1$ and going to $-1$

My thinking is that as the radius of the semicircle is $1$ then $|z|=1$ and so $z=e^{i\theta}$.

We can also parametrise $\gamma$ as $e^{i\theta}, \theta \in [-\pi,\pi]$

This means we can express our integral as;

$\int_{\gamma}\frac{1}{z}dz=\int_{\pi}^{-\pi}e^{-i\theta}ie^{i\theta}d\theta=\int_{\pi}^{-\pi}id\theta=0$

So what does the question mean by find all the values of $\int_{\gamma}\frac{1}{z}$ ?

It seems to me there is only one ?

|cite|improve this question
$\endgroup$
  • 1
    $\begingroup$ There are two different arcs you can take. Also you calculated the integral wrong and your parameterization is wrong. $\endgroup$ – Cameron Williams May 10 '18 at 22:58
  • $\begingroup$ @CameronWilliams do you mean I should evaluate both $\int_{\pi}^{-\pi}id\theta$ and $\int_{-\pi}^{\pi}id\theta$ ? wouldn't they both still just give zero ? or should I evaluate $\int_{\pi}^{-\pi}e^{-i\theta}ie^{i\theta}d\theta$ and $\int_{\pi}^{-\pi}e^{-i\theta}ie^{-i\theta}d\theta$ ? $\endgroup$ – excalibirr May 10 '18 at 23:02
  • 2
    $\begingroup$ @exodius The point (amongst other points) is $\int_{\pi}^{-\pi} d\theta \ne 0.$ It is $-2\pi.$ $\endgroup$ – spaceisdarkgreen May 10 '18 at 23:03
  • $\begingroup$ @spaceisdarkgreen oh shoot sorry , I didn't see my mistake until now , So then I should have the integrals $\int_{\pi}^{-\pi}id\theta=2\pi i$ and $\int_{-\pi}^{\pi}id\theta=-2\pi i$ ? $\endgroup$ – excalibirr May 10 '18 at 23:05
  • 1
    $\begingroup$ @exodius It really depends whether you're taking the upper or lower semicircle. If you are taking the upper semicircle, from $1$ to $-1,$ then $e^{i\theta}$ for $\theta\in[0,\pi]$ is the most natural parametrization. You then get $\int_{0}^\pi e^{-i\theta}ie^{i\theta}d\theta = i\pi.$ ($-\pi$ and $\pi$ are the same angle, so going between these would be going around a whole circle, starting and ending at $-1$). $\endgroup$ – spaceisdarkgreen May 10 '18 at 23:08
3
$\begingroup$

Given a Contour integral $\int_{\gamma}\frac{1}{z}dz$ where $\gamma$ is the semicircle going from $1$ to $-1$, then there two arcs which we can integrate over.

There is the lower semi-circle, where we go from $1$ to $-1$ in a clockwise direction, and also the upper semi-circle , where we go from $1$ to $-1$ in an anti-clockwise direction .

With that in mind there are two semi-circles we can choose from;

1) moving in a clockwise direction in which case $\gamma$ can be parametrised as $e^{-i\theta}, \theta \in[0,\pi]$

So here we have $\int_{\gamma}\frac{1}{z}dz=\int_{0}^{\pi}\frac{-ie^{-i\theta }d\theta}{e^{-i\theta}}$, as z=$e^{-i\theta}$ when moving counter-clockwise

$\int_{0}^{\pi}\frac{-ie^{-i\theta }d\theta}{e^{-i\theta}}=-i\int_{0}^{\pi}d\theta=-i[\theta]^{\pi}_{0}=-i\pi$

2) moving in a counter clockwise direction , in which case we can parametrise $\gamma$ as $e^{i\theta}, \theta \in [0,\pi]$

So for this we have $\int_{\gamma}\frac{1}{z}dz=\int_{0}^{\pi}\frac{ie^{i\theta }d\theta}{e^{i\theta}}$, as z=$e^{i\theta}$

$\int_{0}^{\pi}\frac{ie^{i\theta }d\theta}{e^{i\theta}}=i\int_{0}^{\pi}d\theta=i[\theta]^{\pi}_0=\pi i$

|cite|improve this answer
$\endgroup$

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.