On the degree of polynomial subgroup growth in class 2 nilpotent groups. (Q877479)

In this short paper the author provides a lower bound for the degree of polynomial normal subgroup growth of a finitely generated torsion-free class-\(2\) nilpotent group. This bound can be expressed in terms of the Hirsch lengths of the group itself, its Abelianisation and its centre. The normal zeta function of a finitely generated group \(G\) is the Dirichlet series \(\zeta_G^\triangleleft(s)=\sum_{n=1}^\infty a_n^\triangleleft n^{-s}\), where \(s\) stands for a complex variable and where \(a_n^\triangleleft=a_n^\triangleleft(G)\) denotes the number of normal subgroups of index \(n\) in \(G\). The study of these and related Dirichlet series, associated to groups in the spirit of the Dedekind zeta function of a number field, was initiated by \textit{F. J. Grunewald, D. Segal} and \textit{G. C. Smith} [Invent. Math. 93, No. 1, 185-223 (1988; Zbl 0651.20040)]. If \(G\) is a finitely generated torsion-free nilpotent group, the Dirichlet series \(\zeta_G^\triangleleft(s)\) converges on some right half-plane. Moreover, the abscissa of convergence \(\alpha_G^\triangleleft\) of \(\zeta_G^\triangleleft(s)\) can be interpreted as the degree of polynomial growth of the summatory sequence \(s_n^\triangleleft=\sum_{k=1}^na_k^\triangleleft\). It is known that in the given circumstances \(\alpha_G^\triangleleft\) is rational, but not necessarily an integer; [see \textit{M. du Sautoy} and \textit{F. Grunewald}, Ann. Math. (2) 152, No. 3, 793-833 (2000; Zbl 1006.11051)]. In general, it is a hard problem to describe the degree \(\alpha_G^\triangleleft\) of a finitely generated torsion-free nilpotent group \(G\) in terms of suitable group-theoretic data. A prominent basic invariant of such a group \(G\) is its Hirsch length \(h(G)\). The general bounds \(h(G/[G,G])\leq\alpha_G^\triangleleft\leq h(G)\) are not difficult to prove, but especially the lower bound is rather unsatisfactory. In the paper under review the author shows that the degree of growth of a finitely generated torsion-free class-\(2\) nilpotent group \(G\) satisfies \[ \alpha_G^\triangleleft\geq\max\left(\{d\}\cup\left\{\tfrac{(m-k)(d+k)+1}{h-k}\mid 1\leq k\leq m-1\right\}\right), \] where \(h=h(G)\), \(d=h(G/[G,G])\) and \(m=h(Z(G))\). This improves previously known estimates in the class-\(2\) setting, typically when \(m\) is large. In order to construct sufficently many normal subgroups, the author takes advantage of the Euler decomposition of \(\zeta_G^\triangleleft(s)\) into local factors. The problem of finding more accurate estimates for the degree of growth \(\alpha_G^\triangleleft\) of a finitely generated nilpotent group \(G\) of class greater than \(2\) remains open. \textit{C. Voll} [Math. Ann. 332, No. 1, 67-79 (2005; Zbl 1155.20308)] has shown that the stated lower bound is actually attained for finitely generated free class-\(2\) nilpotent groups. It is a challenge to determine the precise degree of polynomial normal subgroup growth for free nilpotent groups of higher class. Similar problems surround the degree of polynomial subgroup growth of polycyclic groups and compact \(p\)-adic Lie groups; [e.g., see Section~1.4 in \textit{B. Klopsch}, Math. Z. 245, No. 2, 335-370 (2003; Zbl 1045.20017)].