Constructing (non-)Catenarian rings (Q2075938)

Let \(R\) and \(S\) be commutative rings with identity, \(f:R\to S\) a ring homomorphism and \(J\) an ideal of \(S\). Then the subring \(R\bowtie^fJ:=\{(r,f(r)+j)\mid r\in R\) and \(j\in J\}\) of \(R\times S\) is called the \emph{amalgamation} of \(R\) with \(S\) along \(J\) with respect to \(f\). In the paper under review the author investigates the transfer of the catenary property between \(R\) and \(R\bowtie^fJ\). The set \(X\subseteq\mbox{Spec}(R)\) is called catenary if, for each pair \(P\subseteq Q\) of prime ideals of \(X\), all saturated chain of prime ideals of P to \(Q\) have the same finite length. The ring \(R\) is called catenary if, \(\mbox{Spec}(R)\) is catenary. Then the main result of this paper is as follows. Assume that \(f:R\to S\) is surjective. Then \(R\bowtie^fJ\) is catenary if and only if \(R\) and \(\mbox{Spec}(R)\setminus\mbox{V}(J)\) are catenary. By an example, it is shown that the surjectivity assumption on \(f\) is crucial. Also several other interesting examples are constructed. In particular, a catenary ring which is not Cohen-Macaulay and a catenary ring (with zero-divizors) which is not Prüfer are constructed.