Grothendieck rigidity and virtual retraction of higher-rank GBS groups

A rank $n$ generalized Baumslag-Solitar group ($GBS_n$ group) is a group that splits as a finite graph of groups such that all vertex and edge groups are isomorphic to $\mathbb{Z}^n$. This paper investigates Grothendieck rigidity and virtual retraction properties of $GBS_n$ groups. We show that every residually finite $GBS_n$ group is Grothendieck rigid. Further, we characterize when a $GBS_n$ group satisfies property (VRC), showing that it holds precisely when the monodromy is finite.

💡 Research Summary

The paper studies two structural properties of higher‑rank generalized Baumslag‑Solitar groups (GBS$_n$), namely Grothendieck rigidity and the virtual retraction property (VRC). A GBS$_n$ group is defined as the fundamental group of a finite graph of groups in which every vertex and edge group is isomorphic to $\mathbb Z^n$. The authors first recall Grothendieck’s question about when an embedding $u\colon H\hookrightarrow G$ between finitely generated residually finite groups induces an isomorphism of profinite completions $\widehat u\colon\widehat H\to\widehat G$, and they call a pair $(G,H)$ a Grothendieck pair if $H$ is a proper subgroup of $G$ but $\widehat u$ is an isomorphism. A group is Grothendieck rigid if it admits no such proper subgroup.

The paper’s first main result (Theorem 1.1) states that every residually finite GBS$_n$ group is Grothendieck rigid. The proof relies on the known classification of residually finite GBS$_n$ groups (Lodha‑de Graaf‑Zhang‑Sageev 2025): such a group is either a strictly ascending HNN‑extension of $\mathbb Z^n$ or it is virtually $\mathbb Z^n$‑by‑free (hence LERF). For the HNN‑extension case, the authors develop a series of lemmas. Lemma 2.1 shows that any Grothendieck pair $(G,H)$ forces $H$ to have a semidirect product structure $K\rtimes\langle t_H\rangle$, where $K=H\cap N$ and $N$ is the normal closure of the base $\mathbb Z^n$ inside $G$. Lemma 2.2 proves that any finite quotient of $G$ restricts surjectively to $H$, while Lemma 2.3 shows that $