4
$\begingroup$

Let $k$ be a perfect field. I am looking for a $k$-algebra $R$ with the following properties.

  • $R$ is of finite type over $k$ and is a domain;

  • for all ${\mathfrak p}\in{\rm Spec}(R)$, the local ring $R_{\mathfrak p}$ is Cohen-Macaulay and ${\rm emb.\, dim.}(R_{\mathfrak p})-{\rm dim}(R_{\mathfrak p})\leq1$;

  • for at least one ${\mathfrak p}\in{\rm Spec}(R)$, the local ring $R_{\mathfrak p}$ is not Gorenstein.

I would also be happy with the weaker condition "not a complete intersection" in place of "not Gorenstein".

A related question: is there an example of reduced local noetherian ring $T$, which is Cohen-Macaulay, has ${\rm emb.\, dim.}(T)-{\rm dim}(T)=1$, but is not Gorenstein (resp. not a complete intersection)?

Any help would be appreciated.

$\endgroup$
2
  • 2
    $\begingroup$ I think your second condition forces your ring to be a local complete intersection $\endgroup$
    – Mohan
    Commented May 9, 2023 at 16:43
  • $\begingroup$ @Mohan. Thank you for looking into this. My question is in fact just that: is there such a ring, which is not a complete intersection? and if not, why? I would be happy to see the argument. $\endgroup$ Commented May 9, 2023 at 16:46

1 Answer 1

6
$\begingroup$

There is no such example. A Cohen-Macaulay local ring with $\operatorname{embdim}(R)-\dim(R) \le 1$ is a hypersurface, which is in particular a complete intersection. Indeed, we may pass to the completion to suppose $R$ is complete. Then by Cohen's structure theorem, $R \cong S/I$ where $(S,\mathfrak{n})$ is a regular local ring and $I \subseteq \mathfrak{n}^2$. In this situation, $\dim S=\operatorname{embdim}(R)$, and as $R$ is Cohen-Macaulay, the Auslander-Buchsbaum formula implies $\operatorname{pd}_S(R) \le 1$. If it is $0$, then $R$ is regular and we're done. If it is $1$, then from the exact sequence $0 \to I \to S \to R \to 0$ we see that $I$ is a free $S$ module. As $I$ is an ideal, it must be principal, so $R \cong S/(f)$ for some $f$.

$\endgroup$
1
  • $\begingroup$ Excellent! Thank you so much! $\endgroup$ Commented May 10, 2023 at 8:38

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .