On the period of the continued fraction expansion of \(\sqrt{2^{2n+1}+1}\) (Q555944)

In this interesting paper the authors manage to establish that the lengths of the periods of the regular continued fraction expansions of numbers of the form \(\sqrt{2^{2n+1}+1}\) tend to infinity, for \(n\) tending to infinity. In fact, the latter result is derived as a special case of the main theorem of this paper, which gives a partial (affirmative) answer to a question raised in [\textit{P. Corvaja} and \textit{U. Zannier}, Indag. Math., New Ser. 9, No. 3, 317--332 (1998; Zbl 0923.11103)]. There it was conjectured that, given that a power sum \(f\) satisfies certain technical assumptions, one then has that the periods of the regular continued fraction expansions of \(\sqrt{f(n)}\) tend to infinity, for \(n\) tending to infinity. In the paper under review the authors impose a few more technical conditions on \(f\) and show that under these additional assumptions one has that \(\sqrt{f(n)}\) is eventually irrational (that is, \(\sqrt{f(n)}\) is rational for at most finitely many \(n\)). The authors then proceed by showing that the conjecture of Corvaja and Zannier is true under these additional assumptions on \(f\).