Modified \(6j\)-symbols and 3-manifold invariants (Q639548): Difference between revisions

In the 1940s, Wigner and Racah introduced the original \(6j\)-symbols associated with the Lie algebra \(sl_{2}(\mathbb{C})\) as a tool for studying the quantum theory of angular momentum. These were generalised within mathematics to certain tensors associated to semisimple monoidal categories and a major application of these is in the construction of knot and 3-manifold invariants via state sums; we mention particularly the work of \textit{A. N. Kirillov} and \textit{N. Y. Reshetikhin} [``Representation algebra \(U_{q}(SL(2))\), \(q\)-orthogonal polynomials and invariants of links'', LOMI preprint (1989)], \textit{V. G. Turaev} and \textit{O. Y. Viro} [Topology 31, No. 4, 865--902 (1992; Zbl 0779.57009)] and \textit{J. W. Barrett} and \textit{B. W. Westbury} [Trans. Am. Math. Soc. 348, No. 10, 3997--4022 (1996; Zbl 0865.57013)] in this context. One should also note that these invariants are quantum in nature, in the sense of arising from the representations of quantum groups such as \(U_{q}(sl_{2}(\mathbb{C}))\). The present work introduces modified \(6j\)-symbols, building on previous work of the authors [Compos. Math. 145, No. 1, 196--212 (2009; Zbl 1160.81022)]. The earlier work introduced modified quantum dimensions, where the standard quantum dimension may be zero but the modified dimension non-zero. Thus the \(6j\)-symbols arising from the modified version will certainly differ from the usual ones, although it is shown here that many of the properties one would like from a \(6j\)-symbol still hold and thus new 3-manifold invariants may be constructed. A general framework is considered in the first part of the paper, using the notion of a pivotal category and ambidextrous pairs. Some examples related to \(U_{q}(sl_{2}(\mathbb{C}))\) at a root of unity are given, where the standard \(6j\)-symbols are generically zero but the modified ones are not. This is closely related to work of \textit{R. M. Kashaev} [Mod. Phys. Lett. A 9, No. 40, 3757--3768 (1994; Zbl 1015.17500)] and his volume conjecture.