Simple Hopf algebras and deformations of finite groups. (Q2474600): Difference between revisions

There exist the notions of \textit{extension of Hopf algebras} and of \textit{normal Hopf subalgebra}; any normal Hopf subalgebra gives rise to an extension. So, there exists the notion of simple Hopf algebra. It is a basic problem in the classification of complex semisimple Hopf algebras to determine all simple (semisimple) Hopf algebras. If \(G\) is a finite simple group, then the group algebra \(\mathbb{C} G\) and its dual are both simple. Now two Hopf algebras have the same tensor category of representations iff they are twist-equivalent, meaning that one can be obtained from the other by a Drinfeld twist. Any Hopf algebra twist-equivalent to the group algebra of a finite simple group is simple [\textit{D. Nikshych}, Commun. Algebra 26, No. 1, 321-342 (1998); erratum ibid. 26, No. 4, 1347 (1998) and ibid. 26, No. 6, 2019 (1998; Zbl 0912.16018)]. It is natural to ask whether any simple (semisimple) Hopf algebra is twist-equivalent to the group algebra of a finite simple group, or even when a semisimple Hopf algebra twist-equivalent to a simple one, is again simple. The authors show that the answers are no, exhibiting a simple Hopf algebra of dimension \(p^2q^2\) that is twist-equivalent to a group algebra (of a solvable group). This shows that there exists a simple semisimple Hopf algebra of dimension 36 (answering negatively a question of S. Montgomery); and that no analogue of the Burnside theorem exists, at least in a naive sense. The result also implies that simplicity for a Hopf algebra can not be read from the tensor category of representations. Their technique produces also examples of simple Hopf algebras twist-equivalent to the group algebra of the symmetric group, and was pushed forward in [\textit{C. Galindo, S. Natale}, Manuscr. Math. 125, No. 4, 501-514 (2008; Zbl 1178.16032)].