Localization with respect to endomorphisms (Q1402909): Difference between revisions

A monoid \(S\) is called `localizable' if (1) for all \(u,v\in S\), there exist \(u',v'\in S\) such that \(u'u=v'v\) and (2) for all \(u,v,w\in S\) such that \(uw=vw\) there is \(s\in S\) such that \(su=sv\). For monoids \(A\) and \(M\), the monoid \(M\) is called an \(A\)-monoid if \(A\) acts on \(M\). Let now \(S\) be a localizable submonoid of \(\text{End}(M)\). A relation \(\mathcal R\) on \(M\times S\) is defined as follows: \((b,s){\mathcal R}(c,t)\) if and only if there exist \(s',t'\in S\) such that \(s'\circ s=t'\circ t\) and \(s'(b)=t'(c)\). Then \(\mathcal R\) is an equivalence relation and \((M\times S)/{\mathcal R}\) turns to an \(A\)-monoid (denoted by \(S^{-1}M\)) such that \(S\) maps onto a set of \(A\)-automorphisms of \(S^{-1}M\). The author shows that this localization is essentially different from Cohn's type inversive localizations. For further development of the theory, `completely localizable submonoids' are defined. The properties of localizations are discussed paying special attention to commutative algebras over a commutative ring.