Spin networks and \(\text{SL}(2,\mathbb{C})\)-character varieties. (Q6827271)

This paper describes a diagrammatic calculus which applies to various algebraic problems in invariant theory. This theory was motivated by an effort to understand the coordinate rings of various `character varieties', that is Geometric Invariant Theory quotients of \(\Hom(\pi,G)\) by \(\mathrm{Inn}(G)\) where \(\pi\) is a finitely generated group and \(G\) is a reductive algebraic group over \(\mathbb{C}\). Here \(\Hom(\pi,G)\) has the natural structure of an affine algebraic set, preserved by the action of the group \(\mathrm{Inn}(G)\) of inner automorphisms of \(G\). This question is already deep and fundamental when \(\pi\) is a free group.\N\N If \(\pi\) is free of rank \(n\), then \(\Hom(\pi,G)\) identifies with \(G^n\) and its coordinate ring \(A\) is the tensor power \(\otimes^n(\mathbb{C}[G])\) where \(\mathbb{C}[G])\) is the coordinate ring of \(G\). This ring decomposes as a \(G\times G\)-module as a direct sum of \(\mathrm{End}(W)\), where \(W\) ranges over the irreducible representations of \(G\). The theory of spin networks deals with the structure of this ring in terms of the decomposition of tensor products of the irreducible representations into irreducible representations.\N\N This paper treats the case \(G=\mathrm{SL}(2,\mathbb{C})\), in which case the irreducible representations are symmetric powers \(\mathrm{Sym}^k(V)\) of the standard \(2\)-dimensional representation \(V\). The paper produces a basis \(\chi^{a,b,c}\) of \(A\) when \(n=2\), indexed by `admissible triples' of nonnegative integers \((a,b,c)\), that is, triples such that \((-a+b+c)/2\), \((a-b+c)/2\), and \((a+b-c)/2\in\mathbb{Z}\). This function corresponds to the inclusion \(\mathrm{Sym}^c\) as a direct summand of \(\mathrm{Sym}^a\otimes\mathrm{Sym}^b\). Morphisms between various \(G\)-modules are the edges of the graph, and various identities between these morphisms correspond to rules to manipulate the resulting spin networks. This theory was first studied at this depth by mathematical physicists and the theory developed in this paper is related to the classical \(6j\)-symbols.\N\N This theory is applied to give a highly explicit form of the classical Fricke-Klein-Vogt theorem that when \(n=2\) and \(G=\mathrm{SL}(2,\mathbb{C})\), the coordinate ring is a polynomial ring on three variables, the traces of the two free generators and the trace of the product. For higher \(n\) and more complicated \(G\), the structure of \(A\) is much more complicated, and the theory developed in this paper is applied to other cases in later papers by the authors.\N\N The paper is very well written and provides an accessible and readable account to this very intriguing connection between combinatorics and invariant theory.\N\NFor the entire collection see [Zbl 1158.30001].