On strongly flat and weakly cotorsion modules (Q2633090): Difference between revisions

One of the main goals of the paper under review is to provide a description of flat modules over Noetherian commutative rings with countable spectrum. The Govorov-Lazard theorem already provides a way to describe flat left modules over an associative ring, namely every flat left module is the filtered inductive limit of finitely generated free modules. The authors look for a description of flat modules more precise than the one provided by the Govorov-Lazard theorem. Their motivating example is the following: let \(R\) be the ring of integers \(\mathbb{Z}\). Then the Govorov-Lazard theorem implies that every flat module is the filtered inductive limit of finitely generated free abelian groups. This assertion is obvious though once one notices that a \(\mathbb{Z}\)-module is flat if and only if it is a torsion-free abelian group. There is a better way of describing torsion-free abelian groups (and hence flat \(\mathbb{Z}\)-modules), namely every such module is the direct summand of an abelian group \(G\) that fits in an exact sequence \[ 0\rightarrow U\rightarrow G\rightarrow V\rightarrow0, \] where \(U\) is a free abelian group and \(V\) a \(\mathbb{Q}\)-vector space. It is this result that the authors wish to extend to flat modules over Noetherian commutative rings with countable spectrum. To achieve this, the authors prove the following theorem: let \(R\) be a Noetherian commutative ring with countable spectrum. Then any flat \(R\)-module is the direct summand of a module that can be written as the inductive limit of modules of the form \(S^{-1}R\), where \(S\) is a countable multiplicative closed subset of \(R\). The other main result of the paper under review is a description of Enochs cotorsion modules.