Maximal theta functions universal optimality of the hexagonal lattice for Madelung-like lattice energies (Q6161894)

There is a strong connection between theta functions and the sphere packing problem. There are only few dimensions for which the sphere packing problem is completely solved. \textit{H. Cohn} and \textit{A. Kumar} [J. Am. Math. Soc. 20, No. 1, 99--148 (2007; Zbl 1198.52009)] studied the problem of universal optimality of point configurations on several manifolds. For \(\mathbb{R}^{d}\), this problem is closely related to, but much harder than the sphere packing problem and was solved by \textit{H. Cohn} and \textit{A. Kumar} [J. Am. Math. Soc. 20, No. 1, 99--148 (2007; Zbl 1198.52009)] for dimension \(d = 1\). The only dimensions up to dimension 8, where the problem is open, are dimensions 2 and 4. In dimension 2, the hexagonal lattice is the expected candidate to solve the problem of universal optimality (see [\textit{H. Cohn} and \textit{A. Kumar}, J. Am. Math. Soc. 20, No. 1, 99--148 (2007; Zbl 1198.52009)]). Universal optimality of the hexagonal lattice among lattices is known due to the work of \textit{H. L. Montgomery} [Glasg. Math. J. 30, No. 1, 75--85 (1988; Zbl 0639.10017)]. In the present work, two families of 2-dimensional lattice theta functions have been studied, which are accompanying the family of theta functions studied by \textit{H. L. Montgomery} [Glasg. Math. J. 30, No. 1, 75--85 (1988; Zbl 0639.10017)]. These families of functions are called the centered and alternating lattice theta functions which are defined for all \(\alpha>0\), and denoted respectively, by \(\theta^{c}_{\Lambda}(\alpha)\) and \(\theta^{\pm}_{\Lambda}(\alpha)\). These lattice theta functions \(\theta^{c}\) and \(\theta^{\pm}\) can also be seen as a generalization of the classical Jacobi \(\theta_{2}\) and \(\theta_{4}\) functions, respectively. This generalization can also be taken to higher dimensions. Main result in this paper on \(\theta^{c}_{\Lambda}(\alpha)\) and \(\theta^{\pm}_{\Lambda}(\alpha)\) is analogous, but contrary, to Montgomery's main result. As an immediate consequence, a new universal optimality result has been obtained for the hexagonal lattice among two-dimensional alternating charged lattices and lattices shifted by the center of their unit cell.