Localization in a Morita context with applications to fixed rings (Q1178896)

This very nice paper shows how the second layer condition and links between prime ideals transfer from one ring to another. To start with, let \(e\) be an idempotent in the right Noetherian ring \(T\). Using the relationship between Spec\((eTe)\) and Spec\((T)\), the authors are able to transfer results from Spec\((eTe)\) to Spec\(((1-e)T(1-e))\) under suitable hypotheses. In particular, when these hypotheses are satisfied, the second layer condition for \(eTe\) is inherited by \((1-e)T(1-e)\). Furthermore, if every prime of \(eTe\) belongs to a finite classical set, then the same is true of \((1-e)T(1-e)\). As a consequence, under suitable hypotheses, the above mentioned conditions transfer between right Noetherian rings which are Morita equivalent, and this has applications to the study of fixed rings. For example, let \(R\) be a Noetherian ring and let \(G\) be a finite group of automorphisms of \(R\) with \(| G|^{-1}\in R\). If \(R\) satisfies the second layer condition, then it is shown that the same is true for the fixed ring \(R^ G\). Furthermore, if \(N\) is a \(G\)-invariant classical semiprime ideal of \(R\), then \(N\cap R^ G\) is a classical semiprime ideal of \(R^ G\).