On the tensor structure of modules for compact orbifold vertex operator algebras (Q2197656): Difference between revisions

Suppose \(V\) is a vertex operator algebra and \(G\) is a compact Lie group acting faithfully and continuously (and by vertex operator algebra automorphisms) on \(V\); write \(V^G\) for the fixed subalgebra. \textit{C. Dong} et al. provided in [Int. Math. Res. Not. 1996, No. 18, 913--921 (1996; Zbl 0873.17028)] a Schur-Weyl type duality statement: \(V\) is semisimple as a \((V^G \times G)\)-module, and decomposes as \(V = \bigoplus_I V_I \otimes I\), where \(I\) ranges over the irreducible finite-dimensional \(G\)-modules, and \(V_I\) are nonzero distinct irreducible \(V^G\) modules. In particular, this theorem provides an identification of linear semisimple categories between \(\mathrm{Rep}(G)\) and the subcategory \(\mathcal{C}_V \subset \mathrm{Rep}(V^G)\) consisting of direct sums of the \(V_I\)'s. The main result of the present paper improves this to an identification of braided monoidal categories. (The theorem is proved for arbitrary abelian intertwining algebras, which are a mild generalization of vertex operator algebras.) The only assumption needed is \(V^G\) indeed has a category of modules, containing \(\mathcal{C}_V\), with a braided tensor structure. This assumption is fairly deep in general: except in very special cases (unitary, strongly rational, etc.), the construction of braided tensor category structures on modules for vertex operator algebras requires the sophisticated theory of logarithmic vertex tensor categories developed by \textit{Y.-Z. Huang} and \textit{J. Lepowsky} [J. Phys. A, Math. Theor. 46, No. 49, Article ID 494009, 21 p. (2013; Zbl 1280.81125)], which in turn depends on subtle convergence properties of 4-point functions. Nevertheless, the assumption is known to hold for a vast assortment of examples. The proof uses a nice mixture of vertex-algebraic and tensor-categorical techniques.