Virtually cyclic dimension for 3-manifold groups (Q2245020)

Given a group \(\Gamma\), a collection of its subgroups \(\mathcal F\) is called a \textit{family}, if \(\mathcal F\) is closed under conjugation and under taking subgroups. We say a \(\Gamma\)-CW-complex \(X\) is a model for the classifying space \(E_{\mathcal F}\Gamma\) if every isotropy group of \(X\) belongs to \(\mathcal F\). The geometric dimension of \(\Gamma\) with respect to the family \(\mathcal F\), which is denoted by \(gd_{\mathcal F}(\Gamma)\), is the minimum dimension \(n\) such that \(\Gamma\) admits an \(n\)-dimensional model for \(E_{\mathcal F}\Gamma\). In this paper, the authors determine \(gd_{Vcyc}(\Gamma)\) (For short, we write \(\underline{\underline{gd}}(\Gamma)\)) when \(\Gamma\) is the fundamental group of an orientable, closed, connected \(3\)-manifold. In other words, they consider the case of the family being \(\mathcal F = Vcyc\), which means the virtually cyclic subgroups of \(\Gamma\). A group is said to be \textit{virtually cyclic} if the group contains a cyclic subgroup (finite or infinite) of finite index. Theorem. Let \(M\) be a connected, closed, orientable \(3\)-manifold, and let \(\Gamma = \pi_1(M)\) be the fundamental group of \(M\). Then, \(\underline{\underline{gd}}(\Gamma) \leq 4\). Moreover, we can classify \(\underline{\underline{gd}}(\Gamma)\) as follows: \((1)\) \(\underline{\underline{gd}}(\Gamma) = 0\) if and only if \(\Gamma\) is virtually cyclic; \((2)\) \(\underline{\underline{gd}}(\Gamma) = 2\) if and only if \(\Gamma\) is a non-elementary free product of virtually cyclic groups; \((3)\) \(\underline{\underline{gd}}(\Gamma) = 4\) if and only if \(\Gamma\) contains a \(\mathbb Z \times \mathbb Z \times \mathbb Z\) subgroup; \((4)\) \(\underline{\underline{gd}}(\Gamma) = 3\) in all other cases. Note that the authors are dealing with \(3\)-manifold groups, hence the description of the theorem above corresponds to the following geometric characterization of the virtually cyclic geometric dimension. Corollary. Let \(M\) be a connected, closed, oriented \(3\)-manifold, and let \(\Gamma = \pi_1(M)\) be the fundamental group of \(M\). Then, \(\underline{\underline{gd}}(\Gamma) \leq 4\). Moreover, we can classify \(\underline{\underline{gd}}(\Gamma)\) as follows: \((1)\) \(\underline{\underline{gd}}(\Gamma) = 0\) if and only if \(M\) is modeled on \(S^3\) or \(S^2 \times \mathbb R\). \((2)\) \(\underline{\underline{gd}}(\Gamma) = 2\) if and only if every \(P_i\) in the prime decomposition of \(M\) is modeled on \(S^3\) or \(S^2 \times \mathbb E\), and either: \(k >2\) or \(M =P_1 \# P_2\) with \(|\pi_1(P_1)|>2\); \((3)\) \(\underline{\underline{gd}}(\Gamma) = 4\) if and only if one of the prime decompositions \(P_i\) is modeled on \(\mathbb E^3\); \((4)\) \(\underline{\underline{gd}}(\Gamma) = 3\) in all other cases. A main tool to prove the theorem is the use of the Kneser-Milnor prime decomposition and the JSJ (Jaco-Shalen-Johannson) decomposition of a \(3\)-manifold, the push-out constructions associated to acylindrical splittings (see [\textit{J.-F. Lafont} and \textit{I. J. Ortiz}, J. Lond. Math. Soc., II. Ser. 79, No. 2, 309--322 (2009; Zbl 1176.19001)]), the theory of Seifert fibered and hyperbolic manifolds, and Bredon cohomology computations. Moreover, the authors discuss when a manifold is Seifert fibered, with results the reviewer finds particularly interesting. Let \(M\) denote a Seifert fibered \(3\)-manifold without boundary and \(B\) denote its base space. Then, \(B\) can be either a \textit{good} or a \textit{bad orbifold}. Let \(\Gamma\) \(=\) \(\pi_1(M)\). If the underlying space of \(B\) is modeled on \(S^2\), then \(\Gamma\) is always virtually cyclic regardless of whether \(B\) is \textit{good} or \textit{bad} and \(\underline{\underline{gd}}(\Gamma) = 0\). The proof of this statement is straightforward for those who are familiar with elliptic manifolds. For instance, if the underlying space of \(B\) is the \(2\)-sphere, then its fundamental group is \(SO(3)\) \(\cong\) Isom\((S^2)\). We apply a lemma in this paper to conclude that \(\Gamma\) is virtually cyclic. Now, we want to recall the classification of bad orbifolds (see [\textit{P. Scott}, Bull. Lond. Math. Soc. 15, 401--487 (1983; Zbl 0561.57001), Theorem 2.3]). If \(B\) bad, then its Euler characteristic is a positive value. By [loc. cit., Theorem 5.3 (ii)], \(M\) is modeled on one of \(S^3\) or \(S^2 \times \mathbb E\) giving us Isom\((S^3)\) \(\cong\) \(SO(4)\) or Isom\((S^2 \times \mathbb E)\) \(\cong\) \(SO(3) \times (\mathbb R \rtimes \mathbb Z_2)\) respectively, which are both virtually cyclic. Finally, we will also see the cases when \(B\) is modeled on \(\mathbb H^2\) or \(\mathbb E^2\) and when \(\partial M\) \(\neq\) \(\emptyset\).