On Rees algebras of linearly presented ideals (Q743911): Difference between revisions

Let \(R\) be a Noetherian commutative ring and \(I=(f_1,\dots,f_n)\) an \(R\)-ideal. The Rees ring of the ideal \(I\) is the graded \(R\)-algebra \(\mathcal{R}(I)=\bigoplus_{j\geq 0}I^jt^j\), where \(t\) is an indeterminate. A natural way to represent the Rees algebra \(\mathcal{R}(I)\) is to consider the map \(R[t_1,\dots,t_n]\twoheadrightarrow \mathcal{R}(I)\), \(t_i\mapsto f_it\). The kernel of this map called the defining ideal of the Rees algebra \(\mathcal{R}(I)\) is the set \(\mathcal{L}\) of all polynomials \(h\in R[t_1,\dots,t_n]\) such that \(h(f_1,\dots,f_n)=0\). Recall that the ideal \(I\) satisfies the condition \(G_s\), \(s\) an integer, if \(\mu(I_{\wp})\leq\text{Ht}\wp\) for every prime ideal \(\wp\in V(I)\) with \(\text{Ht}\wp\leq s-1\). The ideal \(I\) satisfies the condition \(G_{\infty}\) if it satisfies \(G_s\) for every \(s\). Both conditions \(G_{\infty}\) and \(G_s\) were introduced by \textit{M. Artin} and \textit{M. Nagata} [J. Math. Kyoto Univ. 12, 307--323 (1972; Zbl 0263.14019)]. If \(R=k[x_1,\dots,x_d]\) is a polynomial ring over a field \(k\) and that the ideal \(I\) has an \(n\times m\) presentation matrix \(\varphi\) whose entries are linear forms in \(R\), then there exists a unique \(d\times m\) matrix \(\mathcal{B}(\varphi)\) with linear entries in \(k[t_1,\dots,t_n]\) so that \(\underline{t}.\varphi=\underline{x}.\mathcal{B}(\phi),\) where \(\underline{t}=(t_1\dots t_n)\) and \(\underline{x}=(x_1\dots,x_d)\). The matrix \(\mathcal{B}(\varphi)\) called the Jacobian dual matrix of \(\varphi\). In the paper under review, the author prepares some results on the Jacobian dual matrix and on linkage theory. Also, the author show that if \(I\) is a height two perfect ideal with a linear presentation matrix in a polynomial ring \(R=k[x_1,\dots,x_d]\) and \(\mu(I)=d+1\) such that \(I\) satisfies the Artin-Nagata condition \(G_{d-1}\), then the Rees algebra \(\mathcal{R}(I)\) is Cohen-Macaulay.