Torsions of 3-dimensional small covers (Q1688143)

In this paper the authors study the integral homology of 3-manifolds constructed by gluing right-angled polyhedra. If \(P\) is a combinatorial 3-dimensional polytope whose 1-skeleton is a trivalent graph then any colouring of the faces by \((\mathbb Z_2)^3\) (we keep the authors' notation \(\mathbb Z_2\) for the ring \(\mathbb Z/2\mathbb Z\)) prescribes in a natural way a gluing pattern for eight copies of \(P\). If the colouring satisfies the condition that the colours of any three adjacent faces generate \((\mathbb Z_2)^3\) then the resulting cell complex is a 3-manifold. Manifolds obtained by this construction are called small covers, and the authors prove that the integral homology group of such a manifold has a torsion subgroup of exponent 2. The proof is by explicit computation of the cell complex associated to the polyhedral decomposition. They also prove that it is possible to construct a sequence of covers where the Betti numbers mod 2 grow linearly in the degree. The construction is by an appropriate series of doublings of the original right-angled polytope. These results should be viewed in the context of recent work on the exponential growth of torsion homology in finite covers following [\textit{N. Bergeron} and \textit{A. Venkatesh}, J. Inst. Math. Jussieu 12, No. 2, 391--447 (2013; Zbl 1266.22013)]. We note that the authors' reference [\textit{T. Koberda}, ``Homological eigenvalues of mapping classes and torsion homology growth for fibered 3-manifolds'', Preprint, \url{arXiv:1205.0215}] is obsolescent, a more recent paper on the topic is [\textit{Y. Liu}, ``Virtual homological spectral radii for automorphisms of surfaces'', Preprint, \url{arXiv:1710.05039}].