Bridgeland stability conditions on wild Kronecker quivers (Q2313340)

The authors study the space of Bridgeland stability conditions \(\mathrm{Stab}(D^b(K(l)))\) over the triangulated category \(D^b(K(l))\), where \(K(l)\) is the \(l\)-Kronecker quiver (quiver with two vertices and \(l\geq 1\) parallel arrows between the two vertices). It is known by King that \(\mathrm{Stab}(D^b(K(1)))\cong \mathbb{C}^2\). It is well-known that \(D^b(K(2))\cong D^b(\mathbb{P}^1)\). \textit{S. Okada} [J. Algebr. Geom. 15, No. 3, 487--505 (2006; Zbl 1117.14021)] shows that \(\mathrm{Stab}(D^b(K(2)))\cong \mathbb{C}^2\). See also [\textit{Y. Qiu}, Exchange graphs and stability conditions for quivers. Bath: University of Bath (PhD Thesis) (2011)] for a different method. \textit{M. Kontsevich} and \textit{A. Rosenberg} [The Gelfand Mathematical Seminars, 1996--1999. Dedicated to the memory of Chih-Han Sah. Boston, MA: Birkhäuser. 85--108 (2000; Zbl 1003.14001)] introduce a notion of the non-commutative projective space \(\mathrm{N}\mathbb{P}^{l}\) and show that \(D^b(K(l))\cong D^b(\mathrm{N}\mathbb{P}^{l-1})\) for \(l\geq 3\). So the main differences for \(D^b(K(l))\) between the case \(l=2\) and the case \(l\geq 3\) are that: (i) the former is affine while the latter is wild from representation theoretic point of view, and (ii) the former is commutative while the latter is non-commutative from geometric point of view. The main result of this article is that for each integer \(l\geq 3\), there is a biholomorphism \(\mathrm{Stab}(D^b(K(l))) \cong \mathbb{C}\times\mathcal{H}\), where \(\mathcal{H}=\{z\in \mathbb{C}:\Im(z)>0\}\) is the upper half plane on the complex plane \(\mathbb{C}\). To prove the main result, the authors study a \(\mathbb{C}\times \langle A_l \rangle\cong \mathbb{C}\times \mathbb{Z}\)-action (\(l\geq 2\)) on the \(\mathrm{Stab}(D^b(K(l)))\) and the orbit of the action. In general, for a triangulated category \(\mathcal{T}\), there are two natural group actions on the stability manifold \(\mathrm{Stab}(\mathcal{T})\): a left action by the \(\mathrm{Aut}(\mathcal{T})\) and a right action by \(\widetilde{GL}^+(2,\mathbb{R})\). Now \(\mathbb{C}\) is a subgroup of the \(\widetilde{GL}^+(2,\mathbb{R})\) and \(\langle A_l \rangle\) is a subgroup of \(\mathrm{Aut}(D^b(K(l)))\). This \(\mathbb{C}\times \langle A_l \rangle\)-action generalizes the work of \textit{S. Okada} [J. Algebr. Geom. 15, No. 3, 487--505 (2006; Zbl 1117.14021)] for the case \(l=2\) by using the notion of helix of \(D^b(K(l))\) [\textit{W. Crawley-Boevey}, CMS Conf. Proc. 14, 117--124 (1993; Zbl 0828.16012)] and a result of [\textit{E. Macrì}, Math. Res. Lett. 14, No. 4, 657--672 (2007; Zbl 1151.14015)]. The main result provides an example that \(\mathrm{Stab}(\mathcal{T})\) is not biholomorphic to \(\mathrm{Stab}(\mathcal{T}_1)\times \mathrm{Stab}(\mathcal{T}_2)\) where \(\mathcal{T}\) has a semi-orthogonal decomposition \(\mathcal{T}=\langle \mathcal{T}_1, \mathcal{T}_2\rangle\). In a subsequent article [Sel. Math., New Ser. 25, No. 3, Paper No. 45, 60 p. (2019; Zbl 1434.14002)], the authors further study the stability manifold \(\mathrm{Stab}(\mathcal{T})\) under the assumption that \(\mathcal{T}\) has an orthogonal decomposition. In the last section of the article, the authors give an interesting conjecture connecting the density of phases of semistable objects on unit circle with the affineness of stability manifold.