On the wgsc and qsf tameness conditions for finitely presented groups (Q990733)

The present work is a significant contribution on the geometry of finitely presented groups and it continues the study, started in the PhD Thesis of the second author, of some topological properties for discrete groups, namely: the (weak)-geometric simple connectivity (wgsc), introduced in [\textit{L. Funar} and \textit{S. Gadgil}, Int. Math. Res. Not. 2004, No.~24, 1193--1248 (2004; Zbl 1088.57016)], the quasi-simple filtration (qsf), introduced in [\textit{S. G. Brick} and \textit{M. L. Mihalik}, Math. Z. 220, No.~2, 207--217 (1995; Zbl 0843.57003)] and the Tucker property studied in [\textit{M. L. Mihalik} and \textit{S. T. Tschantz}, Trans. Am. Math. Soc. 349, No.~10, 4251--4264 (1997; Zbl 0883.57003)]. Let \(G\) be a finitely presented group and \(X\) a compact polyhedron such that \(\pi_1 X=G\). The first condition means that there exists some \(X\) whose universal covering \(\widetilde X\) has an exhaustion by compact connected and simply connected sub-polyhedra. The second means that \(\widetilde X\) has an exhaustion approximable by finite simply connected complexes. The third means that \(\widetilde X\) satisfies the following condition: the fundamental group of each component of \(\widetilde X - K\) is finitely generated, for any finite sub-complex \(K \subset \widetilde X\). The first part of the paper is devoted to finding characterizations of qsf. The main theorem states that, in an appropriately defined set-up, there is an equivalence between these three conditions, in the sense that they define the same class of groups (i.e. a group is qsf if and only if it is wgsc if and only if it is Tucker). The quasi-isometry invariance of the qsf is obtained as a corollary. Notice that this theorem could help the research of new exotic examples of non qsf groups (if they exist), since, by \textit{M. L. Mihalik} and \textit{S. T. Tschantz} [loc. cit.], the Tucker property is equivalent to a combinatorial property (the existence of a tame 1-combing). Why should we be interested in studying qsf and wgsc? The answer is easy and hard at the same time: there are still no known examples of groups which are not qsf, but it is also possible that all finitely presented groups are qsf (even if there exist uncountably many open contractible manifolds which are not wgsc [\textit{L. Funar} and \textit{S. Gadgil}, loc. cit.]). Section 3 of this work shows some very complicated examples of groups, known in geometric group theory, for which the qsf is still satisfied. A relation between the qsf and the Cayley complex associated to a group presentation seems to be natural. In Section 4 it is shown that even finitely presented groups admitting a Cayley complex whose metric balls have fundamental groups generated by small loops are qsf, and have linear connectivity radius and solvable word problem in the sense of Dehn. The second part of the paper is concerned with the study of extensions of infinite finitely presented groups by some infinitely presented groups to give sufficient conditions for such extensions to be qsf, and the case of the Grigorchuk group and of its finitely presented \(HNN\) extension (constructed in [\textit{R. I. Grigorchuk}, Sb. Mat. 189, No.~1, 79--100 (1998); translation from Mat. Sb. 189, No.~1, 75--95 (1998; Zbl 0931.43003)] and in [\textit{P. de la Harpe}, Topics in geometric group theory. Chicago Lectures in Mathematics. Chicago: The University of Chicago Press (2000; Zbl 0965.20025)]) is considered in detail. Here the main results are: Theorem 1.4: Any ascending \(HNN\) extension of a finitely presented group is qsf. Theorem 1.5: The finitely presented \(HNN\) extension of the Grigorchuk group is qsf.