Generalized radical semigroup rings (Q1369335): Difference between revisions

Let \(R\) be a ring. If \(x\circ y=x+y-xy\) for \(x,y\in R\), then \((R,\circ)\) is called the adjacent semigroup of \(R\). Then \(R\) is called quasi-regular iff its adjacent semigroup is a group; \(R\) is called generalized radical iff its adjacent semigroup is a union of groups. Let \(I(R)\) be the ideal of \(R\) generated by the set of all idempotents of \(R\), and let \(N(R)\) be the largest ideal of \(R\) such that \(N(R)/I(R)\) is nil. A semigroup \(S\) and the ring \(R\) are called torsion-disjoint if \(s^n=t^n\) and \(mr=0\) imply \(s=t\) or \(r=0\), whenever \(s,t\in S\), \(r\in R\), \(m,n\) positive integers, and \(m\) divides \(n\). If \(S\) is a commutative semigroup, \(G\) is the union of all subgroups of \(S\), and \(R\) is commutative, then \(R[S]\) is a generalized radical ring if and only if (i) \(R\) is generalized radical; (ii) if \(R\neq N(R)\), then \(S\) is periodic, (iii) \(I(R)\neq 0\), then \(S\) is periodic and \(I(R)\) is torsion-disjoint with \(G\). If \(B\) is a band, then \(R[S]\) is generalized radical if and only if (i) \(R\) is generalized radical and (ii) if \(I(R)\neq 0\), then \(B\) is normal.