On derivations of centralizer near-rings. (Q647245)

Let \(N\) be a (right) near-ring and \(M\) a subnear-ring of \(N\). An additive mapping \(d\) from \(M\) to \(N\) is called a derivation if \(d(xy)=xd(y)+d(x)y\) holds for all \(x,y\) in \(M\). The perhaps most important near-rings are centralizer near-rings; they form the counterpart of matrix rings in the structure theories of near-rings and rings. Centralizer near-rings arise from a group \((G,+)\) (not necessarily Abelian) and a fixed-point-free group \(C\) of automorphisms of \(G\), and are defined by \[ M_C(G):=\{f\colon G\to G\mid f(0)=0\text{ and }f(c(x))=c(f(x))\text{ for all \(x\) in \(G\) and \(c\) in }C\}. \] A special case is the near-ring \(M_0(G)\) of all maps from \(G\) into \(G\) which preserve 0 in case \(G=\{\mathrm{id}\}\). A map \(f\) in \(M_C(G)\) is said to be of finite rank if there exist \(x_1,\dots,x_n\) in \(G\) such that \(f(G)\) is contained in the union of the sets \(C(x_i)\). The main result of this paper requires several lemmata and is the following (with the notation above): If \(M\) is a subnear-ring of \(M_0(G)\) containing all transformations of finite rank and with a non-zero derivation into some \(M_C(G)\) then \(G\) is the additive group of some near-field \(F\) (hence Abelian), \(C\) is isomorphic to the multiplicative group of \(F\), and \(M\) is isomorphic (as an additive group) to \(G\) (with natural actions). It is unknown if near-fields can have non-zero derivations at all. A computer search found that at least the 7 exceptional finite near-fields cannot carry derivations.