Generalized torsion elements in groups (Q6189341): Difference between revisions

Let \(G\) be a group. An element \(x\in \) is called a \textit{generalized torsion element} if there is a natural number \(k\) and there are elements \(g_{1},g_{2},\dots ,g_{k}\in G\) such that \[ x^{g_{1}}x^{g_{2}}\cdots x^{g_{k}}=1, \] namely, identity is equal to the product of finitely many conjugates of \(x\). Clearly, every torsion element of a group is also generalized torsion. On the other hand, there are many classes of groups where these two notions do not coincide. For instance, every element of the infinite dihedral group is a generalized torsion. The authors report examples of finitely generated torsion free groups whose all elements are generalized torsion (see [\textit{Yu. M. Gorchakov}, Algebra Logika 6, No. 3, 5--7 (1967; Zbl 0167.28704); \textit{A. P. Goryushkin}, Sib. Math. J. 14, 146--148 (1973; Zbl 0269.20026); translation from Sib. Mat. Zh. 14, 204--207 (1973)]). On the other hand, they prove that if \(G\) is an \(FC\)-group or a nilpotent group, then all generalized torsion elements are torsion elements. For finite groups, the authors suggest a method using the character tables to compute the generalized orders of elements. Indeed, they present interesting examples of various groups having particular generalized exponent using some GAP computations. Finally, they conjecture that all finite non-abelian simple groups have maximal generalized exponent less than or equal to \(3\).