Excellent normal local domains and extensions of Krull domains (Q472995): Difference between revisions

A general inspiration in this paper comes from attempts to describe rings between a Noetherian domain \(R\) and its quotient field \(Q(R)\) or between \(R\) and its \(I\)-adic completion \(R^*\), for an ideal \(I\) of \(R\). The results here are addition to a number of results the authors already contributed in the same area. One setting is for a local Krull domain where the authors consider properties such as flatness for the extensions of such domains. If \((R, m)\) is an excellent normal local domain with the quotient field \(K\), \(y\in m\setminus 0\), denote by \(R^*\) the \(Ry\)-adic completion of \(R\). Given elements \(\tau_1,\dots,\tau_n\in yR^*\) algebraically independent over \(R\), the authors define two associated Krull domains denoted by \(A=K(\tau_1,\dots,\tau_n)\) and \(B\) as an approximation domain (defined as a direct limit of an ascending sequence of rings each dominating their predecessors) birationally dominated by \(A\). One of the interests of the authors is to find conditions that will ensure that \(A=B\). There are examples where \(B\) is a proper subobject of \(A\) with \(A\) Noetherian and \(B\) non-Noetherian. Other examples have \(A=B\) and \(A\) non-Noetherian. When additional assumptions on \(R\) are made, namely that \(R\) is countable of Krull dimension at least 2, then the authors show that \(A\) Noetherian and \(A=B\), which is equivalent to the existence of a sequence \(\tau_n, n\in\mathbb N\) with properties as above, so that the extensions \(R[\tau_1,\dots,\tau_n]\hookrightarrow R^*[1/y]\) are flat. This is then used to prove existence of a normal Noetherian local domain \(B\), such that it dominates \(R\), its \(Ry\)-adic completion is \(R^*\) it has a height-one prime ideal \(p\), such that \(R^*/pR^*\) is not reduced. This then applies that \(B\) is not a Nagata domain and consequently is not excellent. The tasks mentioned in the first sentence are harder than they appear, which is one of the reasons the paper is not straightforward to read.