Weakly clean rings. (Q397968)

Building on his ground-breaking work from ``Corner rings of a clean ring need not be clean'' [Commun. Algebra 40, No. 5, 1595-1604 (2012; Zbl 1247.16034)], the author continues his study of weakly clean rings and elements. Let \(R\) be a ring. The author calls \(a\in R\) \textit{weakly clean} exactly when the diagonal matrix \(\text{diag}(a,0)\) is clean (i.e.\ the sum of a unit and an idempotent) in the \(2\times 2\) matrix ring over \(R\). A ring is \textit{weakly clean} if every element is weakly clean. This class of rings is interesting as it is Morita invariant, unlike the class of clean rings. In the paper under review, this class of rings is shown to have some additional interesting properties. First, it contains all complex unital \(C^*\)-algebras of real rank 0, and all semiregular rings. It also contains the \(\pi\)-regular rings. Although the paper does not state it explicitly, the author in fact proves the stronger result that \(\pi\)-regular \textit{elements} are weakly clean. Second, the author revisits the theme from his earlier paper that weakly clean rings are often corner rings of clean rings. In particular, whenever a weakly clean ring \(R\) contains a \(\pi\)-regular subring in its center, then it is shown to be a corner ring of a clean ring. As a corollary, all \(\pi\)-regular rings are corner rings of clean rings. Third, the author provides two important examples showing that there are exchange rings which are not weakly clean. Finally, the author extends the definition of weakly clean rings and elements to the non-unital setting. Somewhat strangely, the definition given for weakly clean elements in non-unital rings does not correspond to the one given for unital rings. In particular if \(R\) is a ring with \(1\), then \(a\in R\) is weakly clean in the non-unital sense if and only if \(-a\in R\) is weakly clean in the unital sense. Fortunately, this discrepancy can easily be fixed by changing the definition in the non-unital case (by changing a few signs) so that it does match the definition in the unital case. After this change, many of the remarks and proofs in Section 4 of the paper become much easier, and also hold at the element level (rather than just at the level of rings). This paper is very well written, with interesting and important results and examples that will be of use to those studying clean rings and exchange rings.