D-branes on Singularities: New Quivers from Old
David Berenstein, Vishnu Jejjala, Robert G. Leigh
TL;DR
The paper develops a systematic orbifold-of-an-orbifold method to compute D-brane quivers at discrete (non-Abelian) singularities by using exact sequences $0 \rightarrow N \rightarrow G \rightarrow G/N \rightarrow 0$ and constructing the quiver for $M/N$ before encoding the $G/N$ action and discrete torsion. It provides explicit rules for node splitting/merging and arrow connections, and extends to product groups with a detailed treatment of discrete torsion via group cohomology $H^2$ and the interaction term $H^1\otimes H^1$, including Abelian and non-Abelian $G_2$ cases. The framework is illustrated through a variety of examples in ${\mathbb{C}}^2$ and ${\mathbb{C}}^3$ involving ${\widehat{\mathbb{D}}}_{k}$, ${\widehat{\mathbb{E}}}_{6}$, ${\widehat{\mathbb{E}}}_{7}$, and Delta groups, revealing rich torsion structures and dualities. It also highlights that distinct geometric orbifolds can yield identical quivers yet with different superpotentials, underscoring the practical utility of quivers as a robust encoding of low-energy gauge data and their AdS/CFT implications.
Abstract
In this paper we present simplifying techniques which allow one to compute the quiver diagrams for various D-branes at (non-Abelian) orbifold singularities with and without discrete torsion. The main idea behind the construction is to take the orbifold of an orbifold. Many interesting discrete groups fit into an exact sequence $N\to G\to G/N$. As such, the orbifold $M/G$ is easier to compute as $(M/N)/(G/N)$ and we present graphical rules which allow fast computation given the $M/N$ quiver.
