Good action on a finite group (Q2193211)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Good action on a finite group |
scientific article |
Statements
Good action on a finite group (English)
0 references
25 August 2020
0 references
In this article, the authors generalise the concept of coprime action. Let \(A\), \(G\) be finite groups such that \(A\) acts on \(G\) as a group of automorphisms. Then we say that this action is good if and only if for all subgroups \(B\) of \(A\) and all \(B\)-invariant subgroups \(H\) of \(G\) it is true that \(H=[H, B] \cdot C_H(B)\). We see that every coprime action is good, and some of the features of coprime action situations can be generalised, e.g. the existence of \(A\)-invariant Hall subgroups. But the authors also give examples for good actions that are not coprime. There is a large body of work by Turull and others toward bounding the Fitting height \(h(G)\) in terms of \(l(A)\) under certain restrictions on the action of \(A\) on \(G\). Here, \(l(A)\) denotes the length of the longest chain of subgroups of \(A\), and the special case of coprime action has been studied from this perspective as early as in the 1960s (e.g. by Thompson). In this line of research, the authors consider a finite nilpotent group \(A\) of odd order which is \(C_q \wr C_q\)-free for every prime \(q\), and which acts on a finite group \(G\) such that the action is good and \(C_G(A)=1\). Moreover, suppose that every subgroup of \(A\) acts on \(G\) with regular orbits and that there is a subgroup \(B\) of \(A\) such that \(\bigcap_{a\in A}[G,B]^a=1\). Then the authors prove that \(G\) is a solvable group of Fitting height at most \(l(A :B)\), where \(B\) is as above and \(l(A :B)\) is the number of prime divisors of \(|A :B|\), counted with multiplicities. We notice that the special case \(B=1\) resembles \textit{A. Turull}'s result from 1986 [J. Reine Angew. Math. 371, 67--91 (1986; Zbl 0587.20017)], and the authors mention that they followed the arguments in his original proof. It is also worth noting that they prove results of Hall-Higman type along the way that are of independent interest.
0 references
solvable group
0 references
automorphism
0 references
Fitting height
0 references
fixed point free action
0 references
regular module
0 references