1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Infinite sum with k^(1/k)

  1. Nov 11, 2012 #1
    1. The problem statement, all variables and given/known data

    Show that
    [tex]\sum_{k=0}^{\infty} \sqrt[k]k-1 [/tex]


    2. Relevant equations

    Ratio, radix theorems, comparison with other sums...

    3. The attempt at a solution

    No idea whatsoever.
    Where does one begin in this case ? With other cases I'm quite confident.
  2. jcsd
  3. Nov 11, 2012 #2
    Let the general term be denoted by:
    a_k = \sqrt[k]{k} - 1
    Do you know how to prove that [itex]\lim_{k \rightarrow \infty}{a_k} = 0[/itex]?
    Then, we have:
    k = (1 + a_k)^{k}
    Taking the same equality for [itex]k + 1[/itex], and subtracting this one, you ought to get:
    1 = (1 + a_{k + 1})^{k + 1} - (1 + a_k)^{k}
    Solve this equation for [itex]a_{k + 1}[/itex]. What do you get?

    Because as [itex]k \rightarrow \infty[/itex], [itex]a_k[/itex] is an infinitesimal quantity, you may expand your expression for [itex]a_{k + 1}[/itex] in powers of [itex]a_{k}[/itex] up to the first non-vanishing order. What do you get?

    The solution for this asymptotic recursion relation would give you a comparison general term [itex]b_n[/itex], and the series [itex]\sum_{n = 1}^{\infty}{b_n}[/itex] is pretty easy to test for convergence. Then, you may use the [STRIKE]Ratio comparison test[/STRIKE].

    Use the Asymptotic comparison test mentioned here instead.
    Last edited: Nov 11, 2012
  4. Nov 11, 2012 #3
    I don't quite follow you.

    I sort of need simpler methods, thanks anyway.
  5. Nov 11, 2012 #4
    Yep, asymptotic comparison seems a good way.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook