Actions of some pointed Hopf algebras on path algebras of quivers (Q5963090)

Throughout the review, \(k\) denotes a field containing a primitive \(n\)-th root of unity \(\zeta\) for a fixed integer \(n\geq 2\); \(H\) denotes a finite dimensional, pointed Hopf \(k\)-algebra with set of grouplike elements \(G(H)\); \(T(n)\) denotes the Taft \(k\)-algebra generated by a grouplike element \(g\) and a \((1,g)\)-skew-primitive element \(x\) satisfying \(g^n=1\), \(x^n=0\) and \(xg=\zeta gx\). Obviously, \(G(T(n))\) is isomorphic to the cyclic group \(\mathbb{Z}_n\) generated by \(g\). Throughout the review, \(Q\) denotes a finite, loopless, Schurian quiver with sets of vertices \(Q_0\) and of arrows \(Q_1\), respectively, and with the path \(k\)-algebra \(kQ\). Assume that if the Hopf algebra \(H\) acts on the path algebra \(kQ\), then this action preserves the ascending filtration by path length of \(kQ\). In the paper under review, the authors ask when the path algebra \(kQ\) admits a nontrivial action of the Hopf algebra \(H\). Specially, of the Taft algebra \(T(n)\). The authors give a complete answer to the latter question. Their strategy is to identify a set of quivers which is small enough so that it is possible to explicitly describe all \(T(n)\)-actions on their path algebras, but large enough so that every quiver admitting a \(T(n)\)-action on its path algebra is a union of quivers from this set. The authors call these quivers the \(\mathbb{Z}_n\)-minimal quivers. More precisely, the authors prove that a \(T(n)\)-action on the path algebra \(kQ\) determines a \(\mathbb{Z}_n\)-action on the quiver \(Q\) by automorphisms (Lemma 3.2). Then every \(\mathbb{Z}_n\)-orbit of vertices is \(T(n)\)-stable. If we let \(\{e_1,e_2,\ldots,e_m\}\) be the set of trivial paths corresponding to some \(\mathbb{Z}_n\)-orbit of vertices, numbered so that \(g\cdot e_i=e_{i+1}\) with subscripts taken modulo \(m\), then the action of the \((1,g)\)-skew-primitive element \(x\) on these paths is given by \(x\cdot e_i=\gamma\zeta^i(e_i-\zeta e_{i+1})\), where \(\gamma\in k\) can be any scalar (Proposition 3.5). For any arrow \(a\in Q_1\), the action of the element \(x\) on \(a\) is given by \(x\cdot a=\alpha a+\beta (g\cdot a)+\lambda\sigma (a)\) for some scalars \(\alpha,\beta,\gamma\in k\). Here, \(\sigma (a)\) is an arrow or trivial path with the same source as \(a\) and the same target as \(g\cdot a\) (Proposition 3.10). Furthermore, if \(Q\) is a \(\mathbb{Z}_n\)-minimal quiver, then these scalars are determined explicitly by some formulas (Theorems 6.1 and 6.3). For a quiver \(Q\) admitting a \(\mathbb{Z}_n\)-action, consider the set of \(\mathbb{Z}_n\)-minimal subquivers of \(Q\), partially ordered by the inclusion. By the \(\mathbb{Z}_n\)-components of \(Q\) the authors mean the \(\mathbb{Z}_n\)-minimal subquivers of \(Q\) which are maximal in this ordering. Then, the quiver \(Q\) decomposes uniquely into a union of its \(\mathbb{Z}_n\)-components (Lemma 7.2) and every \(T(n)\)-action on the path algebra \(kQ\) extending the given \(\mathbb{Z}_n\)-action restricts to a \(T(n)\)-action on the path algebra of every \(\mathbb{Z}_n\)-component of \(Q\) (Lemma 7.3). This decomposition gives a bijection between the \(T(n)\)-actions on the path algebra \(kQ\) extending the given \(\mathbb{Z}_n\)-action and the compatible sets of \(T(n)\)-actions on the path algebras of the \(\mathbb{Z}_n\)-components of \(Q\) (Theorem 7.5). In particular, if a quiver \(Q\) admits a faithful \(\mathbb{Z}_n\)-action, then the path algebra \(kQ\) admits an inner faithful \(T(n)\)-action extending the given \(\mathbb{Z}_n\)-action (Corollary 7.6). In Section 8, the authors extend the results on Taft algebra actions on the path algebra of quivers to actions of the quantum group \(u_q(\mathfrak{sl}_2)\), and to actions of the Drinfeld double of Taft algebras.