Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am having an issue with the following complex analysis problem. I am suppose to find the coefficients of $z^{-1}$, $z^{-2}$ and $z^{-3}$ in the Laurent series for $\displaystyle \frac{1}{\sin z}$ around $z_0 = 0$ which is valid for $2\pi < |z| < 3\pi$.

One way I thought of doing it was to say $$ \frac{1}{\sin z} = \frac{1}{(z - 2\pi)(z - 3\pi)}\frac{(z - 2\pi)(z - 3\pi)}{\sin z} $$

Let $H(z) = \displaystyle \frac{(z - 2\pi)(z - 3\pi)}{\sin z} $. Then we have $$ \frac{1}{\sin z} = H(z)\left[ \frac{A}{z - 2\pi} - \frac{B}{z - 3\pi} \right] = H(z)\left[ \sum_{k = 0}^\infty \frac{A(2\pi)^k}{z^{k + 1}} + \sum_{k = 0}^\infty \frac{Bz^k}{(3\pi)^{k + 1}} \right]. $$ This seems to get me part of the way of where i need to go, but leaves me having to find a series that works for $H(z)$. I was going to continue in this way, but I though that perhaps it was giving me an entire series, and the problem seems to be suggesting to only solve for 3 of the coefficients.

Another way to solve for the coefficients is $\displaystyle a_k = \frac{1}{2\pi i} \int_\gamma \frac{f(w)}{(w - z_0)^{k + 1}} dw$, where $\gamma$ is a circle, of say radius $R$, that is in the annulus.

I tried to do this just for $a_{-1}$ and I got the following $$ \int_0^{2\pi} \frac{Rie^{it}}{\sin(Re^{it})} dt = \int_{u(0)}^{u(2\pi)} \frac{1}{\sin(u)} du $$ where $u = Re^{it}$, so $u(0) = R$ and $u(2\pi) = R$. Thus I'm integrating from $R$ to $R$, so $a_{-1} = 0$.

However the $u$ substitution seems like something went wrong. I am not even sure if I can actually do a $u$ substitution like that, but I can't see how to solve the integral any other way.

I am not sure what direction to go with this problem, and it seems like I am making it a lot harder than it is suppose to be. Can anyone give me some direction on it?

Should I be attempting to solve the integrals, or do something like I was at the beginning?

share|cite|improve this question
up vote 3 down vote accepted

Using integrals is the way to go here. Let $\gamma$ be a path in the specified annulus with winding number 1.

We have

$$2\pi i a_{-1} = \int_\gamma \frac{1}{\sin z}\ dz.$$

We can evaluate this with the residue theorem. We have poles at 0, $\pm \pi$, and $\pm 2\pi$. Computing the residues, we see that the coefficient is 1. Finding the rest of the coefficients is very similar.

share|cite|improve this answer
Thanks. It didn't occur to me to use residue theorem. The order things are laid out in the book are different than how we are covering them in class. Thanks! – zrbecker Jul 21 '12 at 6:00

Try not to memorize mathematics. Here is how you proceed. Check the order of the pole of your function $\frac{1}{\sin(z)}$ at the point where you want to find the expansion. In your case, the function has a pole of order one. That means, Laurent series has the form,

$$ \frac{1}{\sin(z)}=\frac{ a_{-1} }{z} + a_{0} + a_{1}z + a_{2} z^3+\dots $$

Look at the series, to get $a_{-1}$, you need to multiply both sides by $z$ and then take the limit as $z\to 0$. This gives $ a_{-1} = 1 $. To find $a_{0}$, do the following:

1) - multiply both sides by $z$, you get

$$ \frac{z}{\sin(z)} = a_{-1} + a_{0}z + a_{1} z^2 + a_{2} z^3 + \dots $$

2) - differentiate both sides w.r.t $z$ to kill the constant term $a_{-1}$ which gives

$$ \frac{d}{dz} \, \frac{z}{\sin(z)} = a_{0} + 2 a_{1}z + 3\,a_{2} z^2 + \dots\quad (1) $$

3) - take the limit as $ z\to 0 $ to both sides of the above equation, you get $a_{0}=0$.

To find $a_{1}$, just differentiate $(1)$ and take the limit as $z\to 0$, that is

$a_{1}=\frac{1}{6}$. I think you can now find other coefficients.

share|cite|improve this answer
That is neat, but not what I am looking for. I need it for $2\pi < |z| < 3\pi$. The answer provided by Potato got me where I needed. – zrbecker Jul 21 '12 at 9:16

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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