I am interested in (as explicit as possible) descriptions of non-Gorenstein integral projective curves. Most of the literature on singular curves appears to be focused around the Gorenstein case, with a notable exception being the paper of Kleiman, The Canonical Model of a Singular Curve, and a handful of others. Any other references would be appreciated. Specifically I am interested in knowing what has been understood about their derived categories of coherent sheaves (or the subcategory of perfect complexes), so I would very interested in any developments along those lines.


I don't think there is a reasonably short explicit description of non-Gorenstein curves. They could be described as lacking some of those conditions that define Gorenstein curves.

In any case, here is a simple example: Let $C$ be the union of three coordinate axis in $\mathbb A^3$. Then $C$ is not Gorenstein. It is probably a good exercise for you to check that this is true.

What I find fascinating about this is that the union of three lines contained in $\mathbb A^2$ is a Gorenstein curve, so this shows that being Gorenstein or not could depend on very subtle details.

Of course, $C$ can be projected to a plane and if the projection is general, then it will be one-to-one on $C$ and an isomorphism on each component to their respective images, however, it is not an isomorphism on $C$. This follows from the fact that the image of $C$ is Gorenstein, or more directly from the fact that the Zariski tangent space of $C$ at its singular point is $3$-dimensional while that of its image is $2$-dimensional, so they can't be isomorphic.

(In case you are worried that this example is reducible, then think about how to get this singularity on an irreducible curve).

  • 1
    $\begingroup$ Thanks for your answer. I would say that an integral curve in $\mathbb{A}^3$ with a singularity at the origin analytically isomorphic to $C$ above would be non-Gorenstein, yes? If no one is able to say anything about the derived categories of such curves (or provide any other references) in a bit, I will accept your answer. $\endgroup$ – DKS Apr 30 at 15:39
  • $\begingroup$ Yes, that's right. $\endgroup$ – Sándor Kovács Apr 30 at 19:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.