MHB POTW for University Students week 13

Click For Summary
The discussion revolves around evaluating the integral \(\int_{-\infty}^\infty \frac{e^{ax}}{1+e^x}\,dx\). A participant expresses frustration over their initial approach, feeling it was inadequate after reviewing the solution. Another contributor suggests using contour integration and outlines the process, emphasizing the importance of changing limits during substitution. They note that while the initial method had mistakes, the contour integration method provides a valid solution using the residue theorem. The conversation highlights the complexity of the problem and offers insight into alternative evaluation techniques.
veronica1999
Messages
61
Reaction score
0
I know this problem is way beyond my level, but I worked really hard. (1 full week)
After seeing the solution I see my work is nonsense.
Is it all rubbish or is there a tiny bit that can be saved?
 

Attachments

Physics news on Phys.org
veronica1999 said:
I know this problem is way beyond my level, but I worked really hard. (1 full week)
After seeing the solution I see my work is nonsense.
Is it all rubbish or is there a tiny bit that can be saved?

Hi veronica1999, :)

When you substitute, \(t=e^{ax}\) the limits of integration should be changed appropriately.

As \(x\rightarrow\infty\), \(t\rightarrow\infty\) and as \(x\rightarrow -\infty\), \(t\rightarrow 0\). Also, \(dt=ae^{ax}\,dx=at\,dx\). Therefore after substitution the integral becomes,

\[\int_{-\infty}^{\infty}\frac{e^{ax}}{1+e^x}\,dx=\frac{1}{a} \int_{0}^{ \infty}\frac{1}{1+\sqrt[a]{t}}dt\]

I cannot guarantee that this approach would work. However this is one of the mistakes that you have done at the very beginning.

Kind Regards,
Sudharaka.
 
To my mind, the best way to evaluate $\displaystyle \int_{-\infty}^\infty \frac{e^{ax}}{1+e^x}\,dx$ is to use contour integration. Admittedly, this means that you need to know the residue theorem, but on the other hand it avoids the use of beta and gamma functions (which I dislike).

The idea is to evaluate $\displaystyle \oint_C \frac{e^{az}}{1+e^z}\,dz$, where $C$ is the contour that goes anticlockwise round the rectangle with vertices at $\pm R$ on the real axis and the points $\pm R + 2\pi i.$ When $R$ is large, the integrals along the two vertical sides of the rectangle become small (because of the condition $0<a<1$). The sum of the integrals along the two horizontal sides is $\displaystyle \int_{-R}^R \frac{e^{ax}(1-e^{2\pi ai})}{1+e^x}\,dx.$ There is just one singularity of the integrand inside the contour, at the point $\pi i$, and the residue there is $-e^{\pi ai}.$

Letting $R\to\infty$ and using the residue theorem, you find that $$\int_{-\infty}^\infty \frac{e^{ax}}{1+e^x}\,dx = \frac{-2\pi ie^{\pi ai}}{1-e^{2\pi ai}} = \frac\pi{{\scriptscriptstyle\frac1{2i}}(e^{\pi ai} - e^{-\pi ai})} = \frac\pi{\sin \pi a}\,.$$
 
Last edited:
Thread 'Problem with calculating projections of curl using rotation of contour'
Hello! I tried to calculate projections of curl using rotation of coordinate system but I encountered with following problem. Given: ##rot_xA=\frac{\partial A_z}{\partial y}-\frac{\partial A_y}{\partial z}=0## ##rot_yA=\frac{\partial A_x}{\partial z}-\frac{\partial A_z}{\partial x}=1## ##rot_zA=\frac{\partial A_y}{\partial x}-\frac{\partial A_x}{\partial y}=0## I rotated ##yz##-plane of this coordinate system by an angle ##45## degrees about ##x##-axis and used rotation matrix to...

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
Replies
4
Views
2K
Replies
4
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
Replies
1
Views
1K
  • · Replies 10 ·
Replies
10
Views
482
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K