Elliptic curves of rank zero satisfying the \(p\)-part of the Birch and Swinnerton-Dyer conjecture (Q382236)

Let \(E\) be an elliptic curve defined over \(\mathbb{Q}\), and let \(E_D\) denote a quadratic twist of \(E\), which is isomorphic to \(E\) over the quadratic extension \(\mathbb{Q}(\sqrt D)\). Let \(L(E',s)\) denote the Hasse-Weil \(L\)-function of an elliptic curve \(E'\) over \(\mathbb{Q}\). If \(L(E',1)\neq 0\), the Birch and Swinnerton-Dyer Conjecture (BSD) can be written as \[ \frac{ L(E',1) }{\Omega_{E'}} = \frac{\#\text{Ш}(E')\;\prod_q c_q(E') }{ \#E'(\mathbb{Q})_{\mathrm{tor}}^2 } \] where \(\Omega_{E'}\), \(\text{Ш}(E')\), and \( c_q(E')\) are the real period, Tate-Shafarevich group, and the local Tamagawa numbers, respectively. Let \(p\in\{3,5,7\}\). One of the main results of the paper under review is that there are \textit{infinitely} many elliptic curves \(E\) (not quadratic twists to each other) such that \[ \mathrm{ord}_p\;\;\frac{ L(E_D,1) }{\Omega_{E_D}} =\mathrm{ord}_p\;\;\frac{\#\text{Ш}(E_D)\;\prod_q c_q(E_D) }{ \#E_D(\mathbb{Q})_{\mathrm{tor}}^2 } =0 \] for infinitely many integers \(D\). In connection with Goldfeld's conjecture the density of \(D\)'s for which we can prove BSD is of interest to many number theorists, and by the works of Davenport-Heilbronn and Nakagawa-Horie this main result of the paper under review readily implies that if \(p=3\), then there are a positive proportion of such \(D\)'s. In [\textit{K. James}, J. Number Theory 76, No. 1, 16--21 (1999; Zbl 0932.11045)] just a few such elliptic curves \(E\) are introduced for \(p=3\), and this work is an improvement of [Zbl 0932.11045]. The authors use the idea of \textit{V. Vatsal} [Duke Math. J. 98, No. 2, 397--419 (1999; Zbl 0979.11027)] to find conditions on \(D\) such that \(\mathrm{ord}_p( L(E_D,1)/\Omega_{E_D})=0\) for \textit{optimal elliptic curves} \(E\), and use the idea of \textit{G. Frey} [Can. J. Math. 40, No. 3, 649--665 (1989; Zbl 0646.14024)] to find conditions on \(D\) such that \(\dim \mathrm{Sel}^{(p)}(E_D)=0\), which implies the vanishing values of the Mordell-Weil rank, and the \(p\)-parts of the torsion subgroup and the Tate-Shafarevich group. Further imposing some trivial conditions on \(D\) the authors keep the values of \(c_q(E_D)\) in \(\{ 1,2,4\}\), and obtain the equality of the main result stated above. As noted earlier, the infinitude of such optimal elliptic curves \(E\) is one of the accomplishments of their work. To obtain infinitely many optimal elliptic curves satisfying their conditions they modify results on Goldbach numbers represented by a polynomial introduced in [\textit{J. Brüdern} et al., in: Number theory. Arithmetic in Shangri-La. China, 2011. Hackensack, NJ: World Scientific, 17--76 (2013; Zbl 1325.11033)] and [\textit{A. Perelli}, Rev. Mat. Iberoam. 12, No. 2, 477--490 (1996; Zbl 0874.11069)].