Radicals of skew polynomial rings and skew Laurent polynomial rings. (Q716474)

In this paper, \(R\) denotes an associative ring with identity, and \(\sigma\) stands for an automorphism of \(R\). \(W(R)\), \(L(R)\) and \(N(R)\) denote the Wedderburn radical, the Levitzki radical and the upper nil radical of \(R\), respectively. An ideal \(I\) of \(R\) is called a \(\sigma\)-ideal if \(\sigma(I)\subseteq I\). If \(\sigma(I)=I\), then \(I\) is called \(\sigma\)-invariant. An element \(a\in R\) is called \(\sigma\)-nilpotent if for any integer \(l\geqslant 1\), there exists a positive integer \(m=m(l)\), depending on \(l\), such that \(a\sigma^l(a)\sigma^{2l}(a)\cdots\sigma^{ml}(a)=0\). A subset \(S\) of \(R\) is called \(\sigma\)-nil if every element in \(S\) is \(\sigma\)-nilpotent. \(N_\sigma(R)=\sum\{I:I\) is \(\sigma\)-nil \(\sigma\)-ideal of \(R\}\) is the upper \(\sigma\)-nil radical of \(R\). A subset \(S\) of \(R\) is called \(\sigma\)-nilpotent if for any integer \(l\geqslant 1\), there exists a positive integer \(m=m(l)\) such that \(S\sigma^l(S)\sigma^{2l}(S)\cdots\sigma^{ml}(S)=0\). A subset \(S\) of \(R\) is called locally \(\sigma\)-nilpotent if every finite subset of \(S\) is \(\sigma\)-nilpotent. In this paper, the authors introduce two \(\sigma\)-radicals analogous to the Wedderburn radical and the Levitzky radical of rings, namely, the \(\sigma\)-Wedderburn radical \(W_\sigma(R)=\sum\{I:I\) is a \(\sigma\)-nilpotent \(\sigma\)-ideal of \(R\}\) of \(R\) and the \(\sigma\)-Levitzki radical \(L_\sigma(R)=\sum\{I:I\) is a locally \(\sigma\)-nilpotent \(\sigma\)-ideal of \(R\}\) of \(R\). They study the \(\sigma\)-radicals of the polynomial ring \(R[X]\), where \(X\) is a set of commuting indeterminates. They show, for example, that \(L_\sigma(R[X])=L_\sigma(R)[X]\) and \(W_\sigma(R[X])=W_\sigma(R)[X]\). Using the properties of the introduced radicals, the authors study the Wedderburn radical, the Levitzki radical and the upper nil radical of the skew polynomial ring \(R[x;\sigma]\) and the skew Laurent polynomial ring \(R[x,x^{-1};\sigma]\), and they characterize the upper nil radical of \(R[x;\sigma]\) and \(R[x,x^{-1};\sigma]\), via the upper \(\sigma\)-nil radical of \(R\). They show, for example that when \(\sigma\) is of locally finite order, that is, if for every \(r\in R\) there exists a positive integer \(n=n(r)\) such that \(\sigma^n(r)=r\), then \(W(R[x;\sigma])=W(R)[x;\sigma]\), \(L(R[x;\sigma])=L(R)[x;\sigma]\) and \(N(R)=N_\sigma(R)\).