Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Integral problem

  1. Feb 16, 2007 #1


    User Avatar

    1. The problem statement, all variables and given/known data

    http://www4.0zz0.com/2007/02/16/20/74624071.bmp [Broken]

    2. Relevant equations
    Can any body help me with this integral problem???Please.

    3. The attempt at a solution

    Is the only solution by the "douple integration" method ???
    Last edited by a moderator: May 2, 2017
  2. jcsd
  3. Feb 16, 2007 #2


    User Avatar
    Science Advisor
    Homework Helper

    What have you tried?
  4. Feb 16, 2007 #3
    Calc I or Calc II? Are you sure you wrote that problem down correctly?

    (edit: have you heard of the imaginary error function? Either that, or I'm completely blanking out on this one.)
    Last edited: Feb 16, 2007
  5. Feb 16, 2007 #4


    User Avatar

    I'm really sorry. I've written the problem down uncorrectly:confused:

    This is the right one:_
    http://www4.0zz0.com/2007/02/16/21/95988123.bmp [Broken]

    sorry.And thanks for your replies.
    Last edited by a moderator: May 2, 2017
  6. Feb 16, 2007 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Try rewriting the problem;

    [tex]\int e^{-x^2}\cdot e^{-\frac{1}{x^2}} dx[/tex]
  7. Feb 16, 2007 #6


    User Avatar

    I tried this step,but I couldn't continue.

    Thanks for your reply "Hootenanny"
  8. Feb 16, 2007 #7


    User Avatar

    Maybe we need "douple integration" here.
  9. Feb 16, 2007 #8
    cheat with this site:


    enter your function as follows: Exp[-x^2-1/x^2]

    You will see that it is not an easy one.

    However, I guess that for a definite integral {-Inf,+Inf} it could be easier.
    Are you sure you want the indefinite integral?

    If yes, you can still try to figure out out to get by hand the result mathematica will show you.

    The result is:

    Integrate[Exp[-x^2 - 1/x^2], x] ==
    (Sqrt[Pi]*((1 + Erf[-x^(-1) + x])/E^2 + E^2*(-1 + Erf[x^(-1) + x])))/4
  10. Feb 16, 2007 #9


    User Avatar

    Thank you very much for the wonderful site,and for replying.

    I've tried it,and I saw the result. But as you said,I want to figure out how to get by hand the result.

    I still need help.
    Last edited: Feb 16, 2007
  11. Feb 16, 2007 #10
    That's where cheating becomes really an art.
    Try calculating the derivate of the result and write the calculation in reverse order.

    In this process you may have some good idea.

    Another method is to look in the Mathematica package how it does that.
    It is very difficult to read, but it is easy to guess the ingredients.
    Probably a standard form we be used and some integral table results (not many since erf is an easy function).
    But you need Mathematica ... and that's very expensive
  12. Feb 16, 2007 #11


    User Avatar

    Great ideas,especially that one about cheating:rofl: :wink:

    I will try this one,because I can't use Mathematica package right now.
    And we will see:rolleyes:

    Thanks for your great help.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook