Let A be a Noetherian local ring with the maximal ideal m and d = dim A. Let Q be a parameter ideal in A. Let I = Q : m. The problem of when the equality I 2 = QI holds true is explored. When A is a Cohen-Macaulay ring, this problem was completely solved by A. Corso, C. Huneke, C. Polini, and W. Vasconcelos [CHV, CP, CPV], while nothing is known when A is not a Cohen-Macaulay ring. The present purpose is to show that within a huge class of Buchsbaum local rings A the equality I 2 = QI holds true for all parameter ideals Q. The result will supply [Y1, Y2] and [GN] with ample examples of ideals I, for which the Rees algebras R(I) = n≥0 I n , the associated graded rings G(I) = R(I)/IR(I), and the fiber cones F(I) = R(I)/mR(I) are all Buchsbaum rings with certain specific graded local cohomology modules. Two examples are explored.One is to show that I 2 = QI may hold true for all parameter ideals Q in A, even though A is not a generalized Cohen-Macaulay ring, and the other one is to show that the equality I 2 = QI may fail to hold for some parameter ideal Q in A, even though A is a Buchsbaum local ring with multiplicity at least three. Introduction.Let A be a Noetherian local ring with the maximal ideal m and d = dim A. Let Q be a parameter ideal in A and let I = Q : m. In this paper we will study the problem of when the equality I 2 = QI holds true. K. Yamagishi [Y1, Y2] and the first author and K. Nishida [GN] have recently showed the Rees algebras R(I) = n≥0 I n , the associated graded rings G(I) = R(I)/IR(I), and the fiber cones F(I) = R(I)/mR(I) are all Buchsbaum rings with very specific graded local cohomology modules, if I 2 = QI and the base rings A are Buchsbaum. Our results will supply [Y1, Y2] and [GN] with ample examples.Our research dates back to the remarkable results of A. Corso, C. Huneke, C. Polini, and W. Vasconcelos [CHV, CP, CPV], who asserted that if A is a Cohen-Macaulay local ring, then the equality I 2 = QI holds true for every parameter ideal Q in A, unless A is a regular local ring. Let a ♯ denote, for an ideal a in A, the integral closure of a. Then 1991 Mathematics Subject Classification. Primary 13B22, Secondary 13H10.

Abstract. Given a Noetherian local ring (R, m) it is shown that there exists an integer such that R is Gorenstein if and only if some system of parameters contained in m generates an irreducible ideal. We obtain as a corollary that R is Gorenstein if and only if every power of the maximal ideal contains an irreducible parameter ideal.

No abstract

Abstract. Quasi-socle ideals, that is, ideals of the form I = Q : m q (q ≥ 2), with Q parameter ideals in a Buchsbaum local ring (A, m), are explored in connection to the question of when I is integral over Q and when the associated graded ring G(I) = n≥0 I n /I n+1 of I is Buchsbaum for all n 0, where A ( * ) denotes the length of the module. With this notation, the purpose of this article is to prove the following. The associated graded ring G(I) = n≥0 I n /I n+1 of I is a Buchsbaum ring withas A-modules for all i < d anddenotes the homogeneous component with degree n in the ith graded local cohomology module H i M (G(I)) of G(I) with respect to M . Thus, the quasi-socle ideals I = Q : m q behave very well, inside Buchsbaum rings also, under the conditions stated in Theorem 1.1. Notice that, because A is a Buchsbaum ring, the Hilbert coefficients e i Q (A) of the parameter ideal Q are given by the formulaand one has the equalityfor all n ≥ 0 (see [Sch, Korollar 3.2]), so that {e i Q (A)} 1≤i≤d are independent of the choice of Q and are invariants of A. The crucial point in Theorem 1.1 is the equality I 2 = QI; assertions (1), (2), and (3) a d = ab for some a ∈ m q and b ∈ m is rather technical, but at this moment, we do not know whether this additional condition is superfluous.We now briefly explain the background of Theorem 1.1. Our research dates back to works of A. Corso, C. Polini, C. Huneke, W. V. Vasconcelos, and the first author in which the socle ideals Q : m for parameter ideals Q in Cohen-Macaulay rings A were explored with the following result. provided that depth G(m) ≥ 2.Since Buchsbaum rings are very akin to Cohen-Macaulay rings, it seems quite natural to expect that similar results of the Cohen-Macaulay case, such as Theorem 1.3, should be true also in the Buchsbaum case after suitable modifications of the corresponding conditions, which we now report in Theorem 1.1.The proof of Theorem 1.1 is given in Section 2, which we divide into two parts. The first part shows that m q I = m q Q. The second part proves that I 2 = QI. Since A is not necessarily a Cohen-Macaulay ring, the equality I 2 = QI does not readily follow from the fact that m q I = m q Q. We carefully analyze this phenomenon in Section 2. A similar but more restricted result also holds true in the case where G(m) is a Buchsbaum ring with depth G(m) = 1, which we discuss in Section 3. In Section 4, we give exam- Let F denote the set of all the productswhere the family {f i } 1≤i≤q of elements in m \ m 2 is assumed to satisfy the conditions stated above. Let us writeWe put g j = 1≤k≤q,k =j f k for each 1 ≤ j ≤ q, and we choose g ∈ m \ m 2 so that g * , f * j is a regular sequence in G(m) for all 1 ≤ j ≤ q. Letwhence, for all 1 ≤ i ≤ d and 1 ≤ j ≤ q, we get Hence,and soWe may assume that k > 1 and that our assertion holds true for k − 1. Hence,Let y ∈ A, and assume thatThanks to Claim 1, we getWe need the following to show the equality I 2 = QI. Proof. The former assertion is proved exactly in the same way as [GSa3, Lemma 2.3], wher...

No abstract