ARMENDARIZ RINGS AND SEMICOMMUTATIVE RINGS (Q4532322)

A ring \(R\) is called Armendariz if whenever \(f(x)=a_0+a_1x+\cdots+a_mx^m\), \(g(x)=b_0+b_1x+\cdots+b_nx^n\in R[x]\) satisfy \(f(x)g(x)=0\) then \(a_ib_j=0\) for each \(i,j\). A ring \(R\) is called semicommutative if, for every \(a\in R\), \(\{b\in R:ab=0\}\) is an ideal of \(R\). For example, reduced rings are both Armendariz and semicommutative. A ring \(R\) is called 2-primal if its prime radical coincides with its subset of nilpotent elements. The paper under review provides an example of a semicommutative ring \(R\) for which \(R[x]\) is not semicommutative, but it is shown that for all rings \(R\), \(R[x]\) is semicommutative iff the Laurent polynomial ring \(R[x,x^{-1}]\) is also. The authors next show that (i) if \(I\) is an ideal of \(R\) with no nonzero nilpotent elements and such that \(R/I\) is Armendariz, then \(R\) must be Armendariz and (ii) \(R\) is Armendariz iff the classical right quotient ring \(Q\) of \(R\) (if it exists) is Armendariz. The main feature of the paper's final section is an example of an Armendariz ring which is not semicommutative.