1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Integration by Residue Calculus

  1. Dec 8, 2012 #1
    1. The problem statement, all variables and given/known data
    I'm trying to evaluate the integral

    from 0 to ∞

    2. Relevant equations

    To do this I'm going to consider the complex integral:


    Over a semi-circle of radius R in the upper half plane, then let R-->∞

    There are 2 curves, the straight curve along the real axis, and the semi-circular arc.

    3. The attempt at a solution
    Along the first curve (along the real axis) the line integral is just


    from 0 to infinity

    Now the real part of this is 2*I(a)

    The second line integral goes to zero. So 2*I(a)=Re(J).

    So I need to evaluate J by residue calculus and take the real part to get I(a).

    So there are 2 simple poles inside of the semi-circle, namely
    [itex]z=e^{i\frac{\pi}{4}}[/itex] and [itex]z=e^{i\frac{3\pi}{4}}[/itex]

    So [itex]J=2i\pi(Res(f,e^{i\frac{\pi}{4}})+Res(f,e^{i\frac{3\pi}{4}}))[/itex]

    But I'm having trouble computing the Residues.

    I know that [itex]Res(f,e^{i\frac{\pi}{4}})=\frac{e^{iaz}}{4z^{3}} z=e^{i\frac{\pi}{4}}[/itex]

    But I'm struggling with the exponential to an exponential, I can't get a "clean" answer. I'm not sure if theres a better way I should go about this. Or if theres something I'm doing wrong.

    Please Help!!
  2. jcsd
  3. Dec 8, 2012 #2


    User Avatar
    Homework Helper

    Are you sure about that result? I don't think there should be a z in the numerator. It looks like you calculated the residue using L'Hopital's rule, is that right? i.e.,

    $$\lim_{z\rightarrow z_1} (z-z_1)\frac{e^{i\alpha z}}{z^4+1} = \lim_{z\rightarrow z_1} \frac{(z-z_1)'}{(z^4+1)'}e^{i\alpha z}.$$
    When you apply the rule, you will get ##4z_1^3## in the denominator, but your numerator should just be a constant.

    You can always write the roots in cartesian form, ##z = a + ib##.
  4. Dec 8, 2012 #3
    Thanks for replying

    Thats what I got for my residue, Sorry that z is just saying plug in e^i...

    I try writing it in that form (a+bi), but its so messy!

    What I get is


    Does anyone else think I'm crazy? I really feel like somethings wrong here!
  5. Dec 8, 2012 #4


    User Avatar
    Science Advisor
    Homework Helper

    If that's your answer for the whole integral, I'll agree with it, at least for the case of a>=0. If you aren't give that, I'd write |a| instead of a. I'll agree it's not a "really clean" answer. But it's about as clean as you are going to get.
  6. Dec 8, 2012 #5
    Ya I only need to solve for a>0

    Thanks! It gives me some confidence!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook