The extension of the \(D(-k)\)-triple \(\{1,k,k+1\}\) to a quadruple (Q2151109)

Let \(n\) be a nonzero integer. We call the set of \(m\) distinct positive integers a \(D(n)-m\)-tuple, or \(m\)-tuple with the property \(D(n)\), if the product of any two of its distinct elements increased by \(n\) is a perfect square.\par In a very recent paper \textit{N. Adžaga} et al. [Period. Math. Hung. 85, No. 1, 148--163 (2022; Zbl 07574639)] proved that, if \(\{k, k + 1, c, d\}\) is a \(D(-k)\)-quadruple with \(c < d\), then \(c = 1\), hence, a natural problem is to extend the \(D(-k)\)-triple \(\{1, k, k + 1\}\) to a \(D(-k)\)-quadruple. In the aforementioned paper it is conjectured that, if \(\{1, k, k + 1, d \}\) is a \(D(-k)\)-quadruple, then \(d = 4k + 1\). Note that, in this case, by the definition of the \(D(-k)\)-quadruple, \(3k + 1\) must be a square.\par In this paper, the authors confirm the above conjecture for the case \(k = l^2-1, l \ge 3\) (Theorem 1.3 in the paper). It is noted that the conjecture is already confirmed for the case \(k = l^2\) by \textit{Y. Fujita} and \textit{A. Togbé} [Period. Math. Hung. 65, No. 1, 75--81 (2012; Zbl 1274.11078)] and \(k = 2l^2\) by \textit{N. Adžaga} and \textit{A. Filipin} [Mosc. Math. J. 17, No. 2, 165--174 (2017; Zbl 1433.11028)]. The authors also discuss why the problem under study is out of reach for general \(k\) with known methods and give some ideas that might work in the future. The strategy of the proof is briefly described in the paper as follows: ``In Section 2, we transform the problem into a system of Diophantine equations, whose solutions can be expressed as the intersection of two recurrence sequences. The congruence method due to Dujella then gives a lower bound for the number of terms. In Section 3, the lower bound and the hypergeometric method together yield \(l \le 9460\). Finally in Section 4, using linear form in two logarithms, we completely prove Theorem 1.3''.