Positive sectional curvature, symmetry and Chern's conjecture (Q1002010)

This paper deals with the study of the fundamental group of a closed manifold \(M\) of positive sectional curvature. It is well-known that \(\pi_1(M)\) is finite. The Chern conjecture, asserting that any Abelian subgroup of \(\pi_1(M)\) is cyclic, is true if \(M\) is a space form, but it is false in general cases. Under the hypothesis that \(M\) admits an isometric circle \(T^1\)-action, a positive answer to the mentioned conjecture is due to Rong. The authors assume that a torus \(T^k\) acts effectively by isometrics on a closed \(n\)-manifold of positive sectional curvature. Under this hypothesis a classification of the group \(\pi_1(M)\) has been recently obtained. This helps in proving the following results: If \(n> 23\) and \(k>{n+1\over 8}+ 1\), then \(\pi_1(M)\) contains no \(\mathbb{Z}_p\oplus\mathbb{Z}_p\) subgroup, with \(p\) prime, \(p\neq 3\) and any element of order 2 of \(\pi_1(M)\) belongs to the center of \(\pi_1(M)\). Furthermore, if there are no \(T^k\)-fixed points or if the order of \(\pi_1(M)\) is even, then, up to isomorphisms, \(\pi_1(M)\) is the fundamental group of a rational homology 5-sphere or of a 3-dimensional space-form with constant curvature one.