I discovered that $$\zeta(s)=\int_0^1\frac{(-\log(1-x))^{s-1}}{x(s-1)!}dx.$$ Is this an obvious result that is not worth much interest or is this new and unique?
Asked
Active
Viewed 128 times
3
-
1The former (change of variable $x=1-e^{-t}$ plus expansion of $1/(1-e^{-t})$ as a geometric series in $e^{-t}$ do the job). Sorry. – Did Jul 08 '15 at 18:56
-
Yeah I expected as much. – tyobrien Jul 08 '15 at 18:57
1 Answers
4
This is, sadly, already known. Let $x\mapsto1-e^{-t}$, then $x\mapsto x/k$:
$$\begin{align}\int_0^1\frac{(-\log(1-x))^s}x~\mathrm dx&=\int_0^\infty\frac{x^s}{e^x-1}~\mathrm dx\\&=\int_0^\infty x^se^{-x}\left(\frac1{1-e^{-x}}\right)~\mathrm dx\\&=\int_0^\infty x^se^{-x}\sum_{k=0}^\infty e^{-kx}~\mathrm dx\\&=\sum_{k=1}^\infty\int_0^\infty x^se^{-kx}~\mathrm dx\\&=\sum_{k=1}^\infty\frac1{k^{s+1}}\int_0^\infty x^se^{-x}~\mathrm dx\\&=\zeta(s+1)\Gamma(s+1)\end{align}$$
As seen here.

Simply Beautiful Art
- 74,685