The category of firm modules need not be Abelian. (Q2468537)

Let \(R\) be a nonunital ring. As generalizations of the usual category of unital modules can be considered the categories of closed modules (\(M\cong\Hom_R(R,M)\)), of firm modules (\(M\cong R\otimes_RM\)), of \(R\)-unitary modules (\(M=RM\)), etc. In the present paper the category of firm modules \(R\)-DMod is studied. A description of monomorphisms and kernels in this category is given. Firm modules are in particular \(R\)-unitary, but the converse is not true in general. A ring \(R\) is called left xst, if the class of \(R\)-unitary modules is hereditary. The following conditions are equivalent: 1) The ring \(R\) is left xst; 2) Every \(R\)-unitary module is firm. In this case the category of firm modules is Abelian. The main result is the following Theorem: Let \(K\) be a field and \(X=\{x,y,z_1,z_2,\dots,z_n,\dots\}\) a countable set. Let \(P\) be the set of words over \(X\) given by \(\{xz_1\}\cup\{xz_n-z_{n-1}y\mid n\in\mathbb{N}\), \(n\geq 2\}\) and \(I\) the ideal of \(k\langle X\rangle\) generated by \(P\). Finally, let \(R\) be the nonunital ring \(k\langle X\rangle/I\). Then \(R\)-DMod is not an Abelian category. It is interesting that for right \(R\)-modules the category DMod-\(R\) over the same ring \(R\) is Abelian.