An improved bound for the de Bruijn-Newman constant (Q5934407)

In the study of the Riemann hypothesis, Pólya introduced the function \[ H_\lambda(x):=\int^{\infty}_{0}e^{\lambda t^2}\Phi (t)\cos (xt) dt, \] where \[ \Phi (t)=\sum^{\infty}_{n=1}(2n^4\pi^2e^{9t}-3n^2\pi e^{5t})\exp (-n^2\pi e^{4t}). \] The underlying reason being that \(8H_0(x)=\Xi(x/2)\), where \(\Xi\) is the Riemann function, with all of whose zeros being real being equivalent to the Riemann hypothesis. \textit{N.~G. de Bruijn} [Duke J. Math. 17, 197-226 (1950; Zbl 0038.23302)] and later \textit{C.~M. Newman} [Proc. Am. Math. Soc. 61, 245-251 (1977; Zbl 0342.42007)] together have shown that there exists a real number \(\Lambda\leq 1/2\), such that (i) \(H_{\lambda}(x)\) has only real zeros when \(\lambda\geq \Lambda\), and (ii) \(H_{\lambda}(x)\) has some non-real zeros when \(\lambda <\Lambda.\) This number \(\Lambda\) is now called the de Bruijn-Newman constant, and the truth of the Riemann hypothesis would imply that \(\Lambda\leq 0\), whereas Newman conjectured that \(\Lambda\geq 0.\) The lower bound \(-50<\Lambda\) was obtained by \textit{G.~Csordas}, \textit{T.~S. Norfolk} and \textit{R.~S. Varga} [Numer. Math. 52, 483-497 (1988; Zbl 0663.65017)]. The improvement on such a negative bound requires extensive numerical computations, and \textit{G.~Csordas}, \textit{A.~M. Odlyzko}, \textit{W.~Smith} and \textit{R.~S. Varga} [ETNA, Electron. Trans. Numer. Anal. 1, 104-111 (1993; Zbl 0807.11059)] established that \(-5{\cdot}895{\times}10^{-9}<\Lambda\); they also relied on finding a good `Lehmer pair', namely two zeros of the Riemann zeta function very close to each other near zero number~\(10^9\). From such a pair near zero number \(10^{20}\) found recently, the author can now establish that \(-2{\cdot}7{\times}10^{-9}<\Lambda\); he also remarks that it might seem disappointing that such further computations produced an improvement only by a factor of~2, but that the previous bound was strong because of the exceptionally good Lehmer pair.