Generalized reverse derivations on semiprime rings. (Q901896): Difference between revisions

Let \(R\) be a semiprime ring, \(I\) a nonzero ideal of \(R\) and \(d\) a reverse derivation of \(R\), that is \(d\colon R\to R\) is additive and for all \(x,y\in R\), \(d(xy)=d(y)x+yd(x)\). If also \(F\colon R\to R\) is additive then call the pair \((F,d)\) an \(l\)-generalized reverse derivation of \(R\) if for all \(x,y\in R\), \(F(xy)=F(y)x+yd(x)\), and call \((F,d)\) an \(r\)-generalized reverse derivation of \(R\) if for all \(x,y\in R\), \(F(xy)=d(y)x+yF(x)\). The main result of this paper shows that \((F,d)\) is an \(l\)-generalized reverse derivation of \(R\) if and only if \(d(I)\) and \(F(I)\) centralize \(I\), \(d\) is a derivation on \(I\), and \((F,d)\) is an \(r\)-generalized derivation on \(I\). A corollary is that when \(R\) is a prime ring and \((F,d)\) is an \(l\) (or \(r\))-generalized reverse derivation on \(I\) then \(R\) is commutative.