Arithmetic of seminormal weakly Krull monoids and domains (Q745157): Difference between revisions

The authors study seminormal \(v\)-noetherian weakly Krull monoids and domains. Weakly Krull domains have been introduced by Anderson, Anderson, Mott, and Zafrullah [\textit{D. D. Anderson} et al., Commun. Algebra 20, No. 5, 1447--1462 (1992; Zbl 0749.13012); Boll. Unione Mat. Ital., VII. Ser., B 6, No. 3, 613--630 (1992; Zbl 0773.13004)], and a divisor theoretic characterization was first given by \textit{F. Halter-Koch} [Boll. Unione Mat. Ital., VII. Ser., B 9, No. 2, 417--441 (1995; Zbl 0849.20041)]. A domain is a weakly Krull domain if and only if its monoid of nonzero elements is a weakly Krull monoid, and every one-dimensional noetherian domain is \(v\)-noetherian weakly Krull. Let \(H\) be a seminormal \(v\)-noetherian weakly Krull monoid, \(\widehat H\) its complete integral closure, and suppose that the conductor \(\mathfrak f = (H : \widehat H)\) is nontrivial. In this setting the authors provide a detailed study of the theory of divisorial ideals of \(H\) and in particular they provide a natural exact sequence involving the \(v\)-class groups \(\mathcal C_v (H)\) of \(H\) and \(\mathcal C_v (\widehat H)\) of \(\widehat H\) (Theorem 5.5). Based on these algebraic results the authors study the arithmetic of such weakly Krull monoids under the additional assumption that every class of \(\mathcal C_v (H)\) contains a prime \(\mathfrak p \in \mathfrak X (H)\) with \(\mathfrak p \not\supset \mathfrak f\) (this assumption holds true, among others, for orders in algebraic number fields). For \(k \in \mathbb N\), let \(\mathcal U_k (H)\) denote the set of all \(\ell \in \mathbb N\) such that there is an equation of the form \(u_1 \cdot \ldots \cdot u_k = v_1 \cdot \ldots \cdot v_{\ell}\), where all \(u_i\) and \(v_j\) are irreducible elements of \(H\). These unions of sets of lengths \(\mathcal U_k (H)\) are a frequently studied measure in factorization theory, and the authors show that they are all intervals (apart from some exceptional cases; Theorem 5.8). By definition, we have \(k \in \mathcal U_k (H)\) for every \(k \in \mathbb N\), and the monoid \(H\) is said to be half-factorial if \(\mathcal U_k (H) = \{k\}\) for all \(k \in \mathbb N\). It has been known since the 1960s that a ring of integers in an algebraic number field is half-factorial if and only if its class group has at most two elements. A characterization of half-factoriality for non-principal orders has been known so far only in the case of quadratic number fields (Theorem 3.7.14 in [\textit{A. Geroldinger} and \textit{F. Halter-Koch}, Non-unique factorizations. Algebraic, combinatorial and analytic theory. Pure and Applied Mathematics (Boca Raton) 278. Boca Raton, FL: Chapman \& Hall/CRC (2006; Zbl 1113.11002)]). In the present paper the authors provide a characterization of half-factoriality for seminormal orders in algebraic number fields (indeed, they do it in the more general setting of seminormal weakly Krull monoids; Theorem 6.2).