Positive scalar curvature on manifolds with odd order abelian fundamental groups (Q2024592)

This paper studies existence of positive scalar curvature metrics in closed (without boundary) manifolds \(M\) of dimension \(\ge 5\) and with finite abelian fundamental group of odd order. The main result establishes that such metrics exist under the hypothesis that \(M\) is non-spin and \(p\)-atoral for all primes dividing the order of \(\pi_1(M)\). This is a special case of the Gromov-Lawson-Rosenberg conjecture, and the hypothesis of being non-spin and atoral might be weakened in future work, as the author of the article points out. The main tool for the proof of this result consists in using manifolds with Baas-Sullivan singularities \(A\) as domains of maps \(f:A \to X\) (where \(X\) is a topological space), and a corresponding bordism theory of equivalence classes of these maps. On the other hand, the author also introduces the concept of metric with positive scalar curvature for manifolds with Baas-Sullivan singularities. With this at hand, one can define a subgroup \(H^+_*(X;\mathbb{Z})\) of \(H_*(X;\mathbb{Z})\) consisting (roughly) of those homology classes which can be represented by maps \(f:A \to X\), with some manifold \(A\) with Baas-Sullivan singularities that admits a metric of positive sectional curvature. One of the main results is the following \textit{homology invariance principle:} \(M\) admits a metric of positive scalar curvature if and only if \(\psi_*([M]) \in H^+_*(B \, \pi_1 M;\mathbb{Z})\), being \(\psi: M \to B \, \pi_1 M\) the classifying map. It remains to prove that indeed \(\psi_*([M]) \in H^+_*(B \, \pi_1 M;\mathbb{Z})\). First, classifying spaces \(B \Gamma\) with \(\Gamma\) an abelian \(p\)-group are studied; in particular, the behaviour of \(H^+_*(B \Gamma;\mathbb{Z})\) in Künneth exact sequences. Then, in order to handle \(H^+_*(B \, \pi_1 M;\mathbb{Z})\), the author carries out a theory relating products \(A_1 \times A_2\) of manifolds with Baas-Sullivan singularities and product chain complexes. This is the most technical part of the paper, since the product \(A_1 \times A_2\) is not easy to define. Ultimately, this permits to study \(H^+_*(B \, \pi_1 M;\mathbb{Z})\) by looking at all \(H^+_*(B \Gamma;\mathbb{Z})\), with \(\Gamma\) the \(p\)-subgroups of \(\pi_1 M\).