No label defined (Q5605985)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3327441
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 3327441

      Statements

      0 references
      1970
      0 references
      values of Riemann zeta-function
      0 references
      odd arguments
      0 references
      The value of \(\zeta(a)\) \((\zeta(s)= \) Riemann's zeta function, \(a \in \mathbb Z)\) for \(a\le 0\), or \(a>0\) and even are well-known, but for odd \(a>0\), the determination of the arithmetic nature of these values is still an open problem. The present paper studies this problem, but does not solve it. Specifically, the following is shown: While for even \(a>0\), \(\zeta(a)= \pi^a r_a\), with \(r_a\) a rational number, proportional to the Bernoulli number \(B_a\), for odd \(a>0\), one has \(\zeta(a)=\pi^a \{R_{a}+e(a)\}\), with \(R_a\) rational and depending quadratically on the Bernoulli numbers \(B_2, B_4, \ldots, B_{n+1}\). Here \(e(a)=O\left(\pi^{-a}\right)\) and, for \(a \equiv 3\pmod 4\), \( -\frac{1}{2} \pi^2 e(a)\) is the value at \(\mathfrak{x}=1\) of the function \[ F(\mathfrak{x})=\sum_{m=1}^{\infty}\left(\sigma_{a}(m) / m^{a}\right) e^{2 \pi i m \mathfrak{x}} \] \(\left(\sigma_{a}(m)=\sum_{d \mid m} d^a\right)\); for \(a \equiv 1\pmod 4\) the result is similar, but \(F(\mathfrak x)\) has the somewhat more complicated Fourier coefficients \((1+4\pi /m(a-1)) \sigma_a(m)/m^a\). The method is an extension of that of \textit{A. Weil} [J. Math. Soc. Japan 20, 400--402 (1968; Zbl 0174.33902)]. One sets \[ \Phi(s)=(2\pi)^{-s} \Gamma(s) \zeta'(s) \zeta(s+a) \] and computes its Mellin inverse \[ \frac{1}{2\pi i} \lim_{T \rightarrow \infty} \int_{\sigma - iT}^{\sigma+iT} \Phi(s)(\mathfrak x/i)^{-s} \,ds \] for \(\sigma (=\Re s)=\sigma_0>1\) and after shifting the line of integration to \(\sigma_1=1-a-\sigma_0\); one has to take into account the poles of \(\Phi(s)\) in the strip \(\sigma_1<\sigma<\sigma_0\) and the functional equation \[ \Phi(s)=(-1)^{(a-1)/2} \Phi(1-a-s) \] satisfied by \(\Phi(s)\). A connection with a statement of \textit{S. Ramanujan} [Notebooks. Vols. 1, 2. Bombay: Tata Institute of Fundamental Research (1957; Zbl 0138.24201). Vol.1, p. 259] is mentioned in a remark added in galleys.
      0 references

      Identifiers