Let $R$ be a noetherian regular domain. Suppose that $a, b \in R$, with $b \neq 0$, and consider the ring $S:=R[\frac{a}{b}]=R[X]/(bX-a)$. Is $S$ regular? If this is not the case are there some conditions on $a$ and $b$ (or on $R$) that imply regularity? For example $a=1$ is enough.

  • 1
    $\begingroup$ For example, if $R/(a,b)$ is regular, you should be okay. $\endgroup$ – Mohan May 11 '11 at 16:31
  • $\begingroup$ If $a=b\not\in R^{\times}$, then $S$ is not regular. $\endgroup$ – Laurent Moret-Bailly May 11 '11 at 17:38

If $R = k[x,y]$ and we throw in $\frac {x^2}y$, we get $k[x,y,z]/(zy-x^2)$, not regular. This shows that it's not enough to assume $a,b$ form a regular sequence. The only sufficient condition I can come up with is that $a$ be outside the square of (every/the) maximal ideal

  • $\begingroup$ Thanks! Actually I'm interested in the case where $R= \mathbb Z_p[Y]$, with $a=Y$ and $b=p^n$ for some integer $n$. I think that in this case $S$ is always regular, do you agree with me? $\endgroup$ – Ricky May 12 '11 at 9:01
  • $\begingroup$ Yes, $\mathbb{Z}_p[Y,Z]/(Zp^n-Y)$ is regular. $\endgroup$ – Graham Leuschke May 12 '11 at 10:44

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.