Depth of associated graded rings via Hilbert coefficients of ideals (Q2570667)

Let \((R,m)\) be a local Cohen-Macaulay ring of dimension \(d>0\), and let \(I\) be an \(m\)-primary ideal of \(R\). Then for \(n\gg 0\), the Hilbert-Samuel function \(\lambda(R/I^{n})\) of \(I\) is a polynomial in \(n\) of degree \(d\) and we can write \(\lambda(R/I^{n})\) as \(\lambda(R/I^{n})=\sum_{i=0}^{d}(-1)^{i}e_{i}{n+d-i-1\choose d-i}\). Then \(e_{0}, e_{1}, \cdots e_{d}\) are called the Hilbert coefficients of \(I\). Numerical information on the Hilbert coefficients has been used in order to obtain information on the depth of \(\text{gr}_{I}(R)\). In this paper, the authors' goal is to characterize a sufficiently high depth of the associated graded ring of \(I\) in terms of conditions on the first Hilbert coefficients and in particular on \(e_{2}\) and \(e_{3}\). First the authors show \(e_{2}\leq \sum_{n\geq 1}n\lambda(I^{n+1}/JI^{n})\) for any minimal reduction \(J\) of \(I\). Furthermore, the upper bound is attained if and only if \(\text{depth}(\text{gr}_{I}(R))\geq d-1\). Next they characterize the depth of the associated graded ring of ideals whose second Hilbert coefficient has value `close enough' to this upper bound, and prove that \(e_{2}\geq \sum_{n\geq 1}n\lambda(I^{n+1}/JI^{n})-2\) implies \(\text{depth}(\text{gr}_{I}(R))\geq d-2\). Furthermore, Narita's positivity result on \(e_{2}\) in the case of integrally closed ideals is improved. Namely they show that for any integrally closed ideal \(I\), one has \(e_{2}\geq \lambda(I^{2}/JI)\) for any minimal reduction \(J\) of \(I\). Moreover, the authors consider Valla's conjecture. \textit{S. Itoh} [J. Algebra 176, 638--652 (1995; Zbl 0846.13008)] proved that if \(d\geq 1\) and \(I\) is integrally closed, then \(e_{2}\geq e_{1}-e_{0}+\lambda(R/I)\). Then Valla conjectured the following: If this equality holds in the case in which \(I\) is the maximal ideal \(m\) of \(R\), then the associated graded ring \(\text{gr}_{m}(R)\) is Cohen-Macaulay. But Wang gave an example which shows that this conjecture is false. In this paper, the authors consider Valla's conjecture for normal ideals. In particular they prove that if \(I\) is normal and the above equality holds, then \(e_{2}=\lambda(I^{2}/JI)\) and \(I^{3}=JI^{2}\) for some minimal reduction \(J\) of \(I\), and moreover \(\text{gr}_{I}(R)\) is Cohen-Macaulay and the Hilbert function is known. Finally, the authors improve the already known result of Itoh about non-negativity of \(e_{3}\) under a weaker assumption, and they prove the following. ``Let \((R,m)\) be a Cohen-Macaulay local ring of dimension three and with infinite residue field. Let \(I\) be an \(m\)-primary ideal of \(R\) such that \(I^{q}\) is integrally closed for some \(q\geq n(I)\), where \(n(I)\) is the smallest integer \(n\) such that \(\lambda(R/I^{n})\) is a polynomial. Then \(e_{3}\geq 0\).'' Moreover they show the following. ``Let \((R,m)\) be a Cohen-Macaulay local ring of dimension three and with infinite residue field, and let \(I\) be an \(m\)-primary ideal of \(R\) such that \(I\) is asymptotically normal. Then \(e_{3}=0\) if and only if there exists some \(n\) such that the reduction number of \(I^{n}\) is at most two. Under these circumstances then \(\text{gr}_{I^{n}}(R)\) is Cohen-Macaulay.''