Torus embeddings and dualizing complexes (Q1149488): Difference between revisions

Let \(X_\pi\) be the affine torus embedding defined by a cone \(\pi\) [cf. \textit{G. Kempf}, \textit{F. Knudsen}, \textit{D. Mumford} and \textit{B. Saint-Donat}, Toroidal embeddings. I. Berlin etc.: Springer Verlag (1973; Zbl 0271.14017)]. There is a natural bijective correspondence between the set of faces \(\Gamma(\pi)\) and the set of torus-orbits. A subset \(\Sigma\) of \(\Gamma(\pi)\) is said to be star closed if \(\tau\notin \Sigma\) implies no faces of \(\tau\notin \Sigma\). The above correspondence extends to a bijective correspondence \(\Sigma\rightarrow \operatorname{Spec}(S_\Sigma)\) between the set of star closed subsets of \(\Gamma(\pi)\) and the set of torus-invariant reduced closed subschemes of \(X_\pi\). Now the author constructs explicitly the dualizing complex of \(S_\Sigma\). Thanks to this construction one can give nice conditions in terms of \(\Sigma\) for \(S_\Sigma\) to be Cohen-Macaulay or Gorenstein, which generalizes the results of \textit{G. A. Reisner} [Adv. Math. 21, 30--49 (1976; Zbl 0345.13017)] and of \textit{M. Hochster} [Ring Theory II, Proc. 2nd Okla. Conf. 1975, 171--223 (1977; Zbl 0351.13009)].