On the \(\sigma \)-nilpotent hypercenter of finite groups (Q2093230)

Let \(\mathfrak{F}\) denote a hereditary formation. Before we introduce the topics of this article, we let \(G\) be a finite group, \(H/K\) a chief factor of \(G\) and we recall that \(H/K\) is defined to be \(\mathfrak{F}\)-central in \(G\) if and only if the semidirect product \((H/K) \rtimes (G/C_G(H/K))\) belongs to the formation \(\mathfrak{F}\). If \(N \unlhd G\), then \(N\) is said to be \(\mathfrak{F}\)-hypercentral if and only if \(N=1\) or if every chief factor of \(G\) below \(N\) is \(\mathfrak{F}\)-central. Then the \(\mathfrak{F}\)-hypercenter of \(G\) is the product of all \(\mathfrak{F}\)-hypercentral normal subgroups, and we denote it by \(Z_{\mathfrak{F}}(G)\). One of the results of this article is a characterisation of the \(\mathfrak{F}\)-hypercenter, and for this we need the concept of weak K-\(\mathfrak{F}\)-subnormalizers. A subgroup \(H\) of \(G\) is said to be K-\(\mathfrak{F}\)-subnormal in \(G\) if and only if there is a subgroup chain from \(H\) to \(G\), given by \(H = H_0 \le H_1 \le H_n = G\), \(n \in \mathbb{N}\), where for each \(i \in \{1,\dots,n\}\) we have that \(H_{i-1} \unlhd H_i\) or \(H_i/\mathrm{Core}_{H_i}(H_{i-1}) \in \mathfrak{F}\). Then \(T \le G\) is called a weak K-\(\mathfrak{F}\)-subnormalizer of \(H \le G\) if and only if \(H\) is K-\(\mathfrak{F}\)-subnormal in \(T\) as described above and if, whenever \(H\) is K-\(\mathfrak{F}\)-subnormal in \(M \le G\) and \(T \le M\), then \(T = M\). The formation of all nilpotent groups is an example where the \(\mathfrak{F}\)-hypercenter and the intersection of all weak K-\(\mathfrak{F}\)-subnormalizers of all Sylow subgroups coincide. One result of this article is a description of all formations where this occurs. This is applied to a special case of Shemetkov's problem about the intersection of \(\mathfrak{F}\)-maximal subgroups and the \(\mathfrak{F}\)-hypercenter. Moreover, in the special case of nilpotent groups, this reproduces Hall's classical result about the hypercenter of a group. At the end of the article, there are also results about the non-\(\sigma\)-nilpotent graphs of groups, where \(\sigma\) is a partition of the set of all primes, and the authors discuss how their results generalise some previously known theorems about saturated formations to hereditary formations that are not necessarily saturated.