Precoherent quantale completions of partially ordered semigroups (Q2308108)

There exists the concept of coherent locale, and it is well-known that a locale is coherent if and only if it is isomorphic to the locale of ideals of a distributive lattice~(cf., e.g., [\textit{P. T. Johnstone}, Stone spaces. Cambridge: Cambridge University Press (1982; Zbl 0499.54001)]). A non-commutative version of this result states that a quantale is coherent if and only if it is isomorphic to the quantale of all ideals of an \(m\)-semilattice (Proposition~4.2 on page~209 of [\textit{J. Paseka}, Arch. Math., Brno 22, 203--210 (1986; Zbl 0612.06012)]; see also [\textit{K. Keimel}, Acta Math. Acad. Sci. Hung. 23, 51--69 (1972; Zbl 0265.06016)]). The present paper studies those completions of partially ordered semigroups, which give precoherent quantales. Recall that a quantale \((Q,\otimes, \leqslant)\) is said to be \textit{precoherent} provided that \((Q, \leqslant)\) is an algebraic lattice, and \(K(Q)\) (the set of compact elements of \(Q\)) is closed under \(\otimes\). A precoherent quantale \(Q\) is then called \textit{coherent} provided that the top element \(\top_Q\) of \(Q\) is compact (Definition~2.9 on page~620). The paper first provides a representation theorem for coherent quantales, namely, shows that a quantale \(Q\) is coherent if and only if \(Q\) is isomorphic to the quantale of all distributive ideals of a partially ordered semigroup with a top element~(Theorem~3.9 on page~623). Notice that in case of \(m\)-semilattices, distributive ideals are precisely ideals, which gives back the above-mentioned result for \(m\)-semilattices. The next result shows that precoherent quantale completions of a partially ordered semigroup \(S\) are in one-to-one correspondence with quantic quotients of the quantale \(\mathcal{P}(S)\) (the powerset of \(S\)) with respect to algebraic consistent quantic nuclei on \(\mathcal{P}(S)\) (Corollary~3.16 on page~625). Moreover, Theorem~3.24 on page~627 gives the smallest and the largest precoherent quantale completion of \(S\) (notice that the set of all algebraic consistent quantic nuclei on \(\mathcal{P}(S)\) is a complete lattice and thus has the smallest and the largest element). Additionally, the authors provide the necessary and sufficient conditions on \(S\) for its Frink completion to be a coherent quantale (Corollary~3.33 on page~629). Lastly, motivated by the result that quantales are precisely the injective objects in a certain category of partially ordered semigroups (namely, by Theorem~4 on page~374 of [\textit{X. Zhang} and \textit{V. Laan}, Proc. Est. Acad. Sci. 63, No. 4, 372--378 (2014; Zbl 1332.06047)], which relies on the arguments of Theorem~4.1 on page~342 of [\textit{J. Lambek} et al., Theory Appl. Categ. 26, 338--348 (2012; Zbl 1259.06019)]), the authors define a category of algebraic partially ordered semigroups \textbf{APoSgr}\({}_{\leq}\) and a class of its morphism \(\mathscr{E}_{\leq}\) such that \(\mathscr{E}_{\leq}\)-injective objects in \textbf{APoSgr}\({}_{\leq}\) are precisely precoherent quantales (Theorem~4.4 on page~631). The authors, however, claim to be unable to find \(\mathscr{E}_{\leq}\)-injective hulls in \textbf{APoSgr}\({}_{\leq}\) (Remark~4.7(1) on page~632). The paper is sufficiently well written (with a number of typos, like, e.g., ``pr\textit{o}coherent'' instead of ``pr\textit{e}coherent'' on page~618), contains nearly all of its required preliminaries, and will be of interest to all the researchers, who work with the theory of ordered algebraic structures.