On the irrationality measure of \(\ln 3\) (Q957835)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On the irrationality measure of \(\ln 3\)
scientific article

    Statements

    On the irrationality measure of \(\ln 3\) (English)
    0 references
    1 December 2008
    0 references
    The best bound for the irrationality measure of \(\ln 3\) known so far was obtained by \textit{G.~Rhin} [Théorie des Nombres, Sémin. Paris 1985/1986, Prog. Math. 71, 155--164 (1987; Zbl 0632.10034)], who proved that \(\mu(\ln 3)\leq 8.616.\) In the paper under review, the author significantly improves this estimate and establishes the new record bound \(\mu(\ln 3)\leq 5.125.\) The main result of the paper is as follows. {Theorem.} Suppose that \(q, p_1, p_2\in \mathbb Z\) and \(Q=\max(|q|, |p_1|, |p_2|),\) \(Q\geq Q_0,\) where \(Q_0\) is a sufficiently large number. Then \[ |q+p_1\ln 2+p_2\ln 3|>\frac{1}{Q^{4.125}}. \] As a consequence, one gets that \(|\ln 3 -\frac{p}{q}|>\frac{1}{q^{5.125}}\) for any \(p, q\in {\mathbb Z}\) with \(q\geq q_0.\) The proof essentially uses \textit{M. Hata}'s construction [Acta Arith. 63, No. 4, 335--349 (1993; Zbl 0776.11033)], but with a different integral. The author considers the integrals \(I(\alpha)=\int_{35}^{\alpha}R(x)\,dx,\) where \[ R(x)=\frac{(x-28)^n(x-30)^n(x-35)^{2n}(x-40)^n(x-42)^n}{x^{2n+1}(70-x)^{2n+1}}, \] where \(\alpha\in\{40, 42\}\) and \(n\) is an even positive integer. The following property of the integrand \(R(x)\) is crucial: \(R(70-x)=R(x).\) This allows the author to show that \(35I(40)\cdot 2^{n+1}q_{2n}=A\ln\frac{4}{3}+B_1\) and \(35I(42)\cdot 2^{n+1}q_{2n}=A\ln\frac{3}{2}+B_2,\) where \(A, B_1, B_2\in {\mathbb Z},\) \(q_{2n}= \text{lcm} (1,2,\dots, 2n).\) The exact asymptotics of \(A, I(40), I(42)\) as \(n\to\infty\) is calculated by the saddle point method.
    0 references
    irrationality measure
    0 references
    linear independence measure
    0 references
    linear form in logarithms
    0 references
    Diophantine approximations
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references