Some remarks on \(\mathfrak g\)-variant Fedosov star products and quantum momentum mappings (Q1781418)

\textit{P.~Xu} [Commun. Math. Phys. 197, No. 1, 167--197 (1998; Zbl 0939.37048)] studied the notion of quantum momentum mapping for \({\mathfrak g}\)-invariant star products and raised the question whether the existence of a quantum momentum mapping follows from the existence of a classical momentum mapping. This question is answered in the negative in the paper under review, and the method used here improves results of \textit{K.~Hamachi} [Rev. Math. Phys. 14, No. 6, 601--621 (2002; Zbl 1040.53097)] and \textit{S.~Gutt} [Star products and group actions, Bayrischzell Workshop, April 26--29 (2002)]. The first part of the present paper includes a brief review of \textit{B. V. Fedosov}'s construction of star products on symplectic manifolds. Additionally, one describes the derivations of the corresponding star products, and then one compares the Fedosov derivations obtained from different data. In Section~3 one obtains necessary and sufficient conditions in order for the Lie derivative with respect to a symplectic vector field to be a derivation of a Fedosov star product (Theorem~3.8). This result is then used to investigate the invariance of the star product with respect to a Lie algebra action. One further obtains criteria for existence of quantum Hamiltonians, eventually leading to the negative answer to P.~Xu's aforementioned question.