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

Limit of a complex exponential

  1. May 26, 2014 #1


    The two terms should vanish at infinity according to the Quantum textbook of Griffiths, but I don't see how?

    I mean a complex exponential is a periodic function so how can it vanish at infinity?

    If you split up the first term

    exp(ikx) * exp(-ax)

    Take the limit of infinity of this expression

    lim exp(ikx) is not defined
    limit exp(-ax) turns zero

    But 0 * an undefined limit is still undefined, right?

    So help me out, please.
    Last edited: May 26, 2014
  2. jcsd
  3. May 26, 2014 #2


    User Avatar

    Staff: Mentor

    Wrong :tongue2:. ##0 \times \infty## is undefined, but ##0 \times a## where ##|a| < \infty## is always 0.
  4. May 26, 2014 #3

    lim (exp(ik-a)*x) to infinity equals lim exp (ikx) times lim exp(-ax) so it is ∞ * 0 ?
  5. May 26, 2014 #4


    User Avatar
    Homework Helper

    How do you think [itex]e^{-ax} \cos(kx)[/itex] behaves as [itex]x \to \infty[/itex]?

    If [itex]\lim_{x \to \infty} f(x)[/itex] and [itex]\lim_{x \to \infty} g(x)[/itex] exist, then it follows that [tex]
    \lim_{x \to \infty} f(x)g(x) = \left(\lim_{x \to \infty} f(x) \right)\left(\lim_{x \to \infty} g(x) \right).[/tex] However, even if one or both of [itex]\lim_{x \to \infty} f(x)[/itex] and [itex]\lim_{x \to \infty} g(x)[/itex] do not exist, it might still be the case that [itex]\lim_{x \to \infty} f(x)g(x)[/itex] exists.
  6. May 26, 2014 #5


    User Avatar

    Staff: Mentor

    0 \leq \left| e^{i k x} \right| \leq 1
  7. May 26, 2014 #6
    So lim exp (ikx) does not exist , but lim exp( |ikx|) does?

    But how make it work in lim (exp(ik-a)x)

    Do you know where this rule can be found or what this rule can be called?

    But then you've already calcultated lim exp (-ax) that turns zero and lim exp (ikx) does not become a number?

    I really don't get it.

    If you calculate the limit as a whole (lim exp(ik-a)), you don't get a * 0 or you do?
  8. May 26, 2014 #7

    So because of that absolute value (the norm?) that is always 1 it converges? I think I got it.

    e^ikx converges? |e^ikx| = 1 right ? e^ikx * e^-ikx = e ^ (ikx-ikx) = e^0 = 1 right

    but why is the limit not defined?

    Wait, because it's periodic. Limit is not defined, but the max and min stays fixed. And thats why when you multiply with a function that goes to zero at infinity the product of the two functions also does?

    God, I was too tired to make sense out of anything I think.
  9. May 26, 2014 #8


    User Avatar

    Staff: Mentor

    Sorry, I should've been more clear. Indeed
    \lim_{x \rightarrow \infty} e^{i k x} = \textrm{undefined}
    but ##0 \leq \left| e^{i k x} \right| \leq 1##, which means it is finite. I use the absolute value since the result is complex, but the important thing is that ##e^{i k x}## stays bounded even as ##x \rightarrow \infty##.

    Therefore, since
    \lim_{x \rightarrow \infty} e^{-a x} = 0
    you have
    \lim_{x \rightarrow \infty} e^{i k x} e^{-a x} = 0
    The fact that the limit of the first term is undefined doesn't affect the fact that the product of something finite with zero will always be zero.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook