Quantized GIM Lie algebras and their Lusztig symmetries (Q2488330): Difference between revisions

Motivated by the theory of isolated singularities \textit{P. Slodowy} [Lie algebras and related topics, Proc. Semin., Windsor/Ont. 1984, CMS Conf. Proc. 5, 361--371 (1986; Zbl 0582.17011)] introduced generalized intersection matrix (GIM) Lie algebras. In the paper under review the author considers quantized enveloping algebras of those GIM Lie algebras that can be obtained from \(2\)-fold affinizations of finite Cartan matrices of simply laced type in the sense of \textit{S. Berman} and \textit{R. V. Moody} [Invent. Math. 108, No. 2, 323--347 (1992; Zbl 0778.17018)]. One motivation to study quantized GIM Lie algebras comes from the quantized \(2\)-toroidal Lie algebras introduced by \textit{V. Ginzburg, M. Kapranov}, and \textit{E. Vasserot} [Math. Res. Lett. 2, No. 2, 147--160 (1995; Zbl 0914.11040)] by using a presentation with infinitely many generators and defining relations. Compared with the extensive study of quantized affine Lie algebras, the theory of quantized toroidal Lie algebras is not as well developed, partly because for \(2\)-toroidal Lie algebras no finite presentation is known nor do they have a natural triangular decomposition. Since \(2\)-toroidal Lie algebras are quotients of the corresponding GIM Lie algebras, results on quantized GIM Lie algebras might give some ideas to study quantized toroidal Lie algebras. The author defines Lusztig symmetries on quantized GIM Lie algebras following a similar construction by \textit{J. Beck} [Commun. Math. Phys. 165, No. 1, 193--199 (1994; Zbl 0828.17016)] for quantized affine Lie algebras. The main goal of the paper is to show that these Lusztig symmetries are algebra automorphisms which satisfy the braid group relations. Although quantized GIM Lie algebras have a finite presentation as usual quantized affine Lie algebras, they are neither Hopf algebras with respect to the usual definition of comultiplication nor do they have a natural triangular decomposition. As a consequence, the usual representation-theoretic tools do not carry over from quantized affine Lie algebras to quantized GIM Lie algebras. Lusztig's proof in the affine case uses Verma modules and therefore depends on the triangular decomposition. Instead the author adopts a more combinatorial approach using results of \textit{R.V. Moody} and \textit{Zhiyong Shi} [Nova J. Algebra Geom. 1, No. 4, 317--337 (1992; Zbl 0867.17015)] on toroidal Weyl groups and involving a large number of computations which spezialized to the affine case gives another proof of Lusztig's results in the simply laced case without using representation theory.