0
$\begingroup$

I came up with the following integrations

$$\int_0^\infty \frac{1}{\sqrt{1+a^2\cosh ^2x}}dx,\quad a>0$$

I want to know the precise asymptotic behavior as $a\to 0^+$. I think it is related to the elliptic integral with complex variables. Can anyone give me help? Thx

$\endgroup$

2 Answers 2

3
$\begingroup$

Maple says the exact value is $$(1+a^2)^{-1/2}\,\mathrm{EllipticK}((1+a^2)^{-1/2})$$ $$= \ln(4/a) + \bigl(1/4+\ln(a^{1/4}/2^{1/2})\bigr)a^2 + O(a^4\ln a).$$

This checks out numerically. EllipticK is the complete elliptic integral of the first kind.

$\endgroup$
4
  • $\begingroup$ Is it possible to do it by hand to prove the integral is equal to that EllipticK function? Or can Maple show how to prove this? $\endgroup$ Commented Jan 1 at 7:48
  • $\begingroup$ Numerical shows these two are not equal. $\endgroup$ Commented Jan 1 at 8:07
  • $\begingroup$ @Slm2004 They are definitely the same numerically. Perhaps there is more than one definition of the elliptic function. Example: for a=1 both are 1.311028777146059905232419794945559706841377475715811581408410851900395293535207125115147766480714547 (100 digits). $\endgroup$ Commented Jan 1 at 9:18
  • $\begingroup$ Yes. They are the same. $\endgroup$ Commented Jan 1 at 12:48
3
$\begingroup$

A human proof of the formula from Brendan McKay's answer. In your integral, denote $\tanh x=t$, then $1-t^2=1/\cosh^2 x$ and $dx=dt/(1-t^2)$, thus your integral equals $$\int_0^1\frac{dt}{(1-t^2)\sqrt{1+\frac{a^2}{1-t^2}}}=\int_0^1\frac{dt}{\sqrt{(1-t^2)(1+a^2-t^2)}}\\= (1+a^2)^{-1/2}\int_0^1\frac{dt}{\sqrt{(1-t^2)(1-\frac{t^2}{1+a^2})}}=(1+a^2)^{-1/2}K((1+a^2)^{-1/2})$$

$\endgroup$

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.