Quasi-cyclic modules and coregular sequences (Q2231128)

Let \(A\) be a ring, \(I\) an ideal, and \(M\) a non-zero \(A\)-module with \(\mathrm{Supp} (M)\subset V(I )\). A sequence \(x_1, \dots, x_r\) of elements of \(I\) is \textit{coregular for \(M\)} if \(x_1M = M\) and \(x_{i+1} (0 :_M (x_1,\dots, x_i )) = 0 :_M (x_1,\dots , x_i )\) for \(1 \leq i \leq r-1\). Recall that an irreducible curve in \(\mathbb{P}^3\) is called \textit{set-theoretic complete intersection} if it is the intersection of two surfaces. \textit{M. Hellus} [Local cohomology and Matlis duality. Habilitationsschrift. Leipzig (2006)] gave the following criterion for a subvariety \(V\subset\mathbb{P}^n\) of codimension \(r\), with homogeneous ideal \(I\) in \(A\), the coordinate ring of \(\mathbb{P}^n\) to be a set-theoretic complete intersection: \(V\) is a set-theoretic complete intersection in \(\mathbb{P}^n\) if and only if the local cohomology modules \(H_I^i (A)\) are zero for all \(i\neq r\), and the Matlis dual \(D(H^r_I (A))\) has depth \(r\). In the paper under review, the authors develop the theory of coregular sequences and codepths and give a new version of Hellus's theorem: ``A variety \(V\) of codimension \(r\) in \(\mathbb{P}^n\) is a set-theoretic complete intersection if and only if \(H^i_I (A) = 0\) for all \(i > r\) and \(H^r_I (A)\) has codepth \(r\)'' (To avoid dealing with Matlis duals of large modules, they state their theorem in term of codepth). They also define quasi-cyclic modules as increasing unions of cyclic modules, and show that modules of codepth at least two are quasi-cyclic. Then they focus on curves in \(\mathbb{P}^3\) and give a number of necessary conditions for a curve to be a set-theoretic complete intersection. Thus an example of a curve for which any of these necessary conditions does not hold would provide a negative answer to the still open problem, whether every connected curve in \(\mathbb{P}^3\) is a set-theoretic complete intersection.