On the Gorenstein locus of the punctual Hilbert scheme of degree 11 (Q2443276): Difference between revisions

The Hilbert scheme \({\text{Hilb}}_d (\mathbb P_k^n)\) parametrizing length \(d\) closed subschemes \(X \subset \mathbb P_k^n\) is irreducible for small values of \(d\) (because it is the closure of the open locus \({\mathcal R}\) corresponding to \(d\) distinct points), but for \(d \gg 0\) it was shown to be reducible by \textit{A. Iarrobino} [Invent. Math. 15, 72--77 (1972; Zbl 0227.14006)]. The first explicit example \({\text{Hilb}}_8 (\mathbb P_k^4)\) was given by \textit{A. Iarrobino} and \textit{J. Emsalem} [Compos. Math. 36, 145--188 (1978; Zbl 0393.14002)]; more recently Cartwright, Erman, Velasco and Viray showed that \({\text{Hilb}}_8 (\mathbb P_k^n)\) has exactly two irreducible components for \(n \geq 4\) [\textit{D. A. Cartwright} et al., Algebra Number Theory 3, No. 7, 763--795 (2009; Zbl 1187.14005)]. Naturally the open Gorenstein locus \({\text{Hilb}}^G_d (\mathbb P_k^n) \subset {\text{Hilb}}_d (\mathbb P_k^n)\) also contains \(\mathcal R\) and is irreducible for small \(d\), but \textit{Iarrobino and V. Kanev} showed \({\text{Hilb}}^G_{14} (\mathbb P_k^6)\) is reducible and conjectured irreducibility for \(d < 14\) [\textit{A. Iarrobino} and \textit{V. Kanev}, Power sums, Gorenstein algebras, and determinantal loci. With an appendix `The Gotzmann theorems and the Hilbert scheme' by Anthony Iarrobino and Steven L. Kleiman. Lecture Notes in Mathematics. 1721. Berlin: Springer. (1999; Zbl 0942.14026)]. The authors of the paper under review earlier showed that \({\text{Hilb}}^G_{d} (\mathbb P_k^n)\) is irreducible for \(n \leq 3\) or \(d \leq 10\) [J. Pure Appl. Algebra 213, No. 11, 2055--2074 (2009; Zbl 1169.14003) and J. Pure Appl. Algebra 215, No. 6, 1243--1254 (2011; Zbl 1215.14009)] and here they extend the result to \(d=11\). The problem reduces to studying zero dimensional local Gorenstein \(k\)-algebras \(A\) of length \(d = 11\), which via Macaulay's theory of inverse systems can be written \(k[[x_1, \dots x_n]]/\text{Ann}(F)\) for suitable \(F \in k[y_1, \dots y_n]\) via the action of \(k[[x_1, \dots x_n]]\) on \(k[y_1, \dots y_n]\) given by \(x_i = \partial/\partial y_i\). Using a filtration of \(\mathrm{gr}(A)\) appearing in work of \textit{T. Iarrobino} [Mem. Am. Math. Soc. 514, 115 p. (1994; Zbl 0793.13010)], the authors write the equation for \(F\) in a special form with an integer invariant \(f_3\). In general they show that if \(f_3 \leq 3\), then the scheme corresponding to the algebra \(A\) is in \({\overline {\mathcal R}}\) and \(f_3 > 3\) leads to particular Hilbert functions (especially \(H_A = (1,4,4,1,1)\)) which they handle with special arguments. They also give an almost complete description of the singular locus of \({\text{Hilb}}^G_{11} (\mathbb P_k^n)\).