2
$\begingroup$

I'm new here. Are there places to put specific problems or do they just go to a general list?

There are some similar problems posted, but they are all a bit different and I can't see how to use the advice.

I have two integrals I am having troubles with: $$\int_0^{ \infty } dk ~ k^{d - 1} \dfrac{1}{(k^2 + v^2)^2} = \dfrac{1}{2} (v^2)^{d/2 -2} \Gamma \left ( \dfrac{d}{2} \right ) \Gamma \left ( 2 - \dfrac{d}{2} \right )$$

and $$\int_0^{ \infty } dk ~ k^{d - 1} \dfrac{k^2}{(k^2 + v^2)^2} = \dfrac{1}{2} (v^2)^{d/2 - 1} \Gamma \left ( 1 + \dfrac{d}{2} \right ) \Gamma \left ( 1 - \dfrac{d}{2} \right ).$$

Perhaps I'm not being clever enough. I've tried substitution, series methods, taking the derivative of the integral wrt v, and contour integration. All I can seem to get is that, for most integer values of d, that the integrals do not converge. I'd appreciate a guide to the solution, but I'll settle for someone telling me what the method is.

$\endgroup$
  • 2
    $\begingroup$ Those are the same integral: $d\to d-2$ turns the second into the first. So half your problem is trivial. $\endgroup$ – G. Smith Feb 28 at 17:32
  • $\begingroup$ Ha! I never noticed that. $\endgroup$ – topsquark Feb 28 at 17:34
  • 2
    $\begingroup$ Have you used Beta function formula? $\endgroup$ – Darkseid Feb 28 at 17:39
  • $\begingroup$ See section 8.2.1 here. $\endgroup$ – G. Smith Feb 28 at 17:43
  • $\begingroup$ en.wikipedia.org/wiki/Pochhammer_contour $\endgroup$ – G. Smith Feb 28 at 20:56
4
$\begingroup$

As pointed out in the comments, the two integrals are really the same. Also, just by redefining $k \rightarrow v k$ in the integral, we can easily scale out the $v$-dependence: $$ \int_0^{ \infty } dk ~ k^{d - 1} \dfrac{1}{(k^2 + v^2)^2} = (v^2)^{d/2 -2} \int_0^{ \infty } dk ~ k^{d - 1} \dfrac{1}{(k^2 + 1)^2}, $$ so it suffices to solve the $v=1$ case.

All we will need is the definition of the gamma function: $$ \Gamma(z) = \int_0^{\infty} d \lambda \, \lambda^{z - 1} e^{-\lambda}. $$ This formula immediately implies the so-called Schwinger parametrization of your integrand, which is often useful for loop diagrams: $$ \frac{1}{A^{z}} = \frac{1}{\Gamma(z)}\int_0^{\infty} d \lambda \, \lambda^{z - 1} \, e^{- \lambda A}, $$ which follows from simply scaling $\lambda \rightarrow A \lambda$ in the definition of the gamma function. Using the $z = 2$ case of this formula, your integral becomes $$ I = \int_0^{ \infty } dk ~ k^{d - 1} \dfrac{1}{(k^2 + 1)^2} = \int_0^{ \infty } dk \int_0^{ \infty } d\lambda ~ \lambda \, k^{d - 1} e^{-\lambda(k^2 + 1)}. $$ Now we substitute $\alpha = \lambda k^2$, and after some algebra, we get $$ I = \frac{1}{2} \int_0^{ \infty } d\alpha \int_0^{ \infty } d\lambda ~ \lambda^{1 - d/2} \, \alpha^{d/2 - 1} e^{-\alpha} e^{-\lambda}. $$ Now both integrals are precisely in the form of the gamma function, so we have $$ I = \frac{1}{2} \Gamma\left( \frac{d}{2} \right) \Gamma\left(2 - \frac{d}{2} \right). $$

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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