On minimum number of generators for some derivation modules (Q2153808): Difference between revisions

Let \(G\) be a finite subgroup of \(GL(2, \mathbb{C})\) acting linearly on \(\mathbb{C}[[X, Y]]\) and \(R=\mathbb{C}[[X, Y]]^{G}\) the corresponding ring of invariants. The following statements are the main results of the paper under review. I. Let \(G\) be small and non-cyclic. (A group \(G\subseteq GL(2, \mathbb{C})\) acting linearly on \(C[[X_{1},\dots, X_{n}]]\) is called small, if no element \(g\in G\) has eigenvalue \(1\) with multiplicity \(n-1\). Equivalently, no non-identity element of \(G\) fixes a hyperplane in \(\mathbb{C}^{n}\) pointwise.) Then: (1) The module of derivations \(\mathrm{Der}_{\mathbb{C}}R\) is generated by at most \(2e\) elements where \(e\) is the multiplicity of \(R\). In particular, \(\mathrm{Der}_{\mathbb{C}}R\) is generated by at most \(|G|-1\) elements. If \(G\) is not a binary dihedral group then \(\mathrm{Der}_{\mathbb{C}}R\) can be generated by at most \(|G|/2\) elements. (The actual result is much stronger for many \(G\).)\, (2) For many binary dihedral groups, \(\mathrm{Der}_{\mathbb{C}}R\) needs at least \(e\) generators, where \(e\) is the multiplicity of \(R\). (3) There is no universal upper bound on the minimum number of generators of \(\mathrm{Der}_{\mathbb{C}}R\) as \(G\) varies. II. Let \(G\) be small and non-cyclic. Then there is no universal upper bound on the minimum number of generators of divisorial ideals of \(R\) as \(G\) varies. In fact, there is no universal upper bound on the minimum number of generators of the canonical module of \(R\). The type of \(R\) (\(= \mu(\omega_{R})\)) is equal to \(e-1\), where \(e\) is the multiplicity of \(R\).