MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $X=Spec(A)$ be an affine scheme and $U=Spec(R)$ be an affine open subset of $X$. Is it true that $R$ is an localization of $A$, i.e. $R=S^{-1}A$ for some closed multiplication subset $S\subset A$ ?

share|cite|improve this question
The answers to this question: gives a few correct results in the direction you're asking for. – Dan Petersen Jun 12 '13 at 9:30
Three years ago , such a question was asked as follows (see : : "Given a commutative ring A, is there a "ring-theoretic" characterization of the ring homomorphisms A→B that realize Spec(B) as an open affine subscheme of Spec(A) (more precisely, those morphisms such that the induced map Spec(B)→Spec(A) is an open immersion)?" That is the true meaning of the problem ! In Grothendieck's EGA, I myself found nothing. I shall soon ask one of his students ! – Al-Amrani Jun 13 '13 at 18:27
up vote 11 down vote accepted

No. This question is pretty closely related to this other question, but let me give an answer nonetheless:

Consider an elliptic curve $E$ in $\mathbb{P}^2(\mathbb{C})$, choose coordinates $[x:y:z]$ of $\mathbb{P}^2(\mathbb{C})$ in a way that the line $z = 0$ intersects $E$ at an inflection point $O$. Let $X := E \setminus \lbrace z = 0 \rbrace$. Choose a point $P \in E$ such that $P$ is not a torsion point with respect to the group structure on $E$ for which $O$ is the origin. Let $U := X \setminus \lbrace P \rbrace$. Then $U$ is an open affine subset of $X$, but there is no $f \in A:=\mathbb{C}[X]$ that vanishes only at $P$. So $\mathbb{C}[U]$ cannot contain $A_f$ for any non-constant $f \in A$.

share|cite|improve this answer
$X$ is certainly affine, but could you explain how you know that $U$ is affine? – Zhen Lin Jun 12 '13 at 16:00
By Riemann Roch: any divisor with positive degree on a non-singular curve is ample, so that $k(O + P)$ is very ample for large $k$, i.e. there exists a curve $D$ on $\mathbb{P}^2(\mathbb{C})$ such that $C \cap D = \lbrace{O,P\rbrace}$. Now use that the complement in a projective variety of a hypersurface is affine. – auniket Jun 12 '13 at 16:18

Concerning characterization of an affine open subset U in an affine scheme Spec(A) , with complementary Y, I got the following answer. One uses Serre's affinity criterion. This gives conditions on cohomology whith supports in Y (cohomological dimension). If the ring A is regular, then the characterization is : codim(Y) = 1 purely (Cartier divisor) !.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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