Participants

All participants are located at the

Project summary

Let $H$ be a Krull monoid with finite class group $G$ such that each class contains a prime divisor (this setting includes holomorphy rings in global fields). Then every nonunit $a \in H$ can be written as a finite product of atoms (irreducible elements). If $a = u_1 \cdot \ldots \cdot u_k$ with atoms $u_1, \ldots, u_k \in H$, then $k$ is called the length of the factorization. The set $\mathsf L (a)$ of all possible factorization lengths $k$ is called the set of lengths of $a$, and this is a finite set of positive integers. We consider the system $\mathcal L (H) = \{ \mathsf L (a) \mid a \in H\}$ of sets of lengths of $H$ (for convenience, we set $\mathsf L (a) = \{0\}$ if $a$ is a unit in $H$). The Krull monoid $H$ is factorial if and only if the class group $G$ is trivial. Furthermore, $H$ is half-factorial (i.e., $|L|=1$ for all $L \in \mathcal L (H)$) if and only if $|G| \le 2$. If $|G| \ge 3$, then for every $N \in \mathbb N$ there is a set $L \in \mathcal L (H)$ such that $|L| \ge N$.

The system $\mathcal L (H)$ depends only on the class group $G$. To make this precise, consider the set $\mathcal B (G)$ of zero-sum sequences over $G$. By a sequence over $G$, we mean a finite sequence of terms from $G$ where repetition is allowed and the order is disregarded, and we say that $S$ is a zero-sum sequence if its terms sum up to zero. Defining an operation as the concatenation of sequences we obtain a monoid structure on $\mathcal B (G)$. Indeed, $\mathcal B (G)$ is a Krull monoid with class group isomorphic to $G$ (provided that $|G| \ge 3$) and every class contains a prime divisor. Moreover, the systems of sets of lengths of $H$ and that of $\mathcal B (G)$ coincide. Thus $\mathcal L (H) = \mathcal L \big( \mathcal B (G) \big)$ can be studied with methods from Zero-Sum Theory, a flourishing subfield of Additive Combinatorics.

The Structure Theorem for Sets of Lengths states that every $L \in \mathcal L (G)$ is an almost arithmetical multiprogression with universal bounds for all parameters controlling these multiprogressions. It is a main goal of the present project to study the involved parameters (such as the possible differences of the multiprogressions) in terms of classical zero-sum invariants (such as the Davenport constant of $G$) or even in terms of the group invariants. All work on these parameters will be done with a view towards the Characterization Problem, a main open question in this area. Indeed, let $G$ and $G'$ be two finite abelian groups with $|G|\ge 4$ and $|G'|\ge 4$ and suppose that $\mathcal L \big( \mathcal B (G) \big) = \mathcal L \big( \mathcal B (G') \big)$. Does it follow that $G$ and $G'$ are isomorphic? The answer is affirmative, among others, if $G$ or $G'$ has rank at most two (see [A. Geroldinger and W. Schmid, A characterization of class groups via sets of lengths, J. Korean Math. Society, 56 (2019), 869 -- 915]). However, the Characterization Problem is far open in general, and it will be in the focus of the the present project.

Publications