The Łojasiewicz exponent of nondegenerate surface singularities (Q2434156): Difference between revisions

Let \(f \colon (\mathbb{C}^{n},0)\rightarrow (\mathbb{C},0)\) be an isolated singularity. The Łojasiewicz exponent \(\mathcal{L}_{0}(f)\) of \(f\) is defined by \[ \mathcal{L}_{0}(f)=\inf \{\alpha \in \mathbb{R}_{+}:\exists _{C>0}\exists_{r>0}\forall _{\left| z\right| <r}\;\left| f(z)\right| \geq C\left| z\right| ^{\alpha }\}. \] Let \(\Gamma (f)\subset \mathbb{R}^{n}\) be the Newton polytope of \(f\) and \(\Gamma ^{k}(f)\subset \Gamma (f)\) be the set of all compact \(k\)-dimensional faces of \(\Gamma (f),\) \(\;k=0,1,\ldots ,n-1.\) A face \(S\in \Gamma ^{n-1}(f)\) is an exceptional face if there exists an axis \(0X_{i}\) such that one of its vertices is at distance \(1\) to the axis \(0X_{i}\) and the remaining vertices define an \((n-2)\)-dimensional face which lies in one of the coordinate hyperplanes including \(0X_{i}.\) Denote by \(E_{f}\) the set of all exceptional faces of \(f.\) The main result concerns the case \(n=3:\) if \(f:(\mathbb{C}^{3},0)\rightarrow (\mathbb{C},0)\) is an isolated and non-degenerate singularity then \[ \mathcal{L}_{0}(f)\leq \max_{S\in \Gamma ^{2}(f)\setminus E_{f}}m(S)-1\quad \text{if }\Gamma ^{2}(f)\setminus E_{f}\neq \emptyset \] and \[ \mathcal{L}_{0}(f)=m(I)-1\quad \text{if }\Gamma ^{2}(f)\setminus E_{f}=\emptyset, \] where \(m(S)\) equals to the maximum of coordinates of the intersection of the hyperplane \(L\) supporting \(S\) (\(S\subset L\)) with the union of all axes and \(m(I)\) the coordinate of the intersection of some segment \(I\in \Gamma^{1}(f)\) uniquely determined in the case \(\Gamma ^{2}(f)\setminus E_{f}=\emptyset \) with the union of all axes. The result is stronger than the Fukui inequality [\textit{T. Fukui}, Proc. Am. Math. Soc. 112, No. 4, 1169--1183 (1991; Zbl 0737.58001)] in the case \(n=3\) and it is a generalization of the Lenarcik theorem [\textit{A. Lenarcik}, Banach Cent. Publ. 44, 149--166 (1998; Zbl 0924.32007)] for the case \(n=2\).