No label defined (Q4385641)

Consider positive integers \(D_1\), \(D_2\) and let \(h\) be the class number of the imaginary quadratic field \(\mathbb{Q} (\sqrt{-D_1D_2})\). The aim of the present note is to sharpen the upper bound on \(n\) obtained by \textit{M.-H. Le} [Monatsh. Math. 120, 121-125 (1995; Zbl 0877.11020)] and, which is more important, to prove that (under some extra hypothesis) for each pair \((D_1,D_2)\) there is at most one solution \(n\). Theorem. Consider the Diophantine equation \(D_1x^2+ D_2^m= 4y^n\). Then, in the range \(2\leq y\leq 5000\) the only solutions with \(n>5\) are \(n=7\) for \(y\in \{2,3,4,5\}\) and \(n=13\) for \(y=2\), and the solutions for \(n=5\) are obtained for \[ \begin{multlined} y\in \{2,3,4,5,7,8, 11,13,18, 21,29, 34, 47, 55, 76, 89, 123, 144, 199, 233,\\ 322, 377, 521, 610, 843, 987, 1364, 1597, 2207, 2584, 3571, 4181\}. \end{multlined} \] For \(5000< y< 43000\) there is at most one solution \(n\geq 5\). For \(y\geq 43000\) there is at most one solution \(n\geq 3\). Moreover, in all cases \(n\leq 52000\). The proof follows rather closely Le Maohua's using estimates of linear forms in logarithms.