Map to localization is injective on spectra: Difference between revisions
No edit summary |
No edit summary |
||
| Line 17: | Line 17: | ||
Further, it is also true that the topology on <math>Spec(S)</math> is such that if we give the subspace topology to its image in <math>Spec(R)</math>, the map is a homeomorphism. | Further, it is also true that the topology on <math>Spec(S)</math> is such that if we give the subspace topology to its image in <math>Spec(R)</math>, the map is a homeomorphism. | ||
==Proof== | |||
===Some preliminary observations=== | |||
* The extension of an ideal <math>I</math> of <math>R</math>, to the ring <math>S = U^{-1}R</math>, is the ideal <math>U^{-1}I</math> of <math>S</math> | |||
* If any ideal of <math>R</math> intersects the multiplicatively closed subset <math>U</math>, then its [[extension of an ideal|extension]] to <math>S = U^{-1}R</math> is the whole ring <math>U^{-1}R</math>. Hence, it does ''not'' occur as the contraction of any ideal of <math>S</math> | |||
''Proof'': If <math>I</math> is an ideal of <math>R</math> such that <math>I \cap U</math> is nonempty, then pick <math>u \in I \cap U</math>. Clearly, we have: | |||
<math>1 = \frac{u}{u} \in U^{-1}I = I^{e}</math> | |||
Thus, <math>1</math> is in the extension of <math>I</math> to <math>S</math>. | |||
* The converse is not true for arbitrary ideals; however, it ''is'' true for [[prime ideal]]s (it is true for all ideals if <math>U</math> is a [[saturated subset]]). Formally, if <math>P</math> is a prime ideal of <math>R</math> such that <math>U \cap P</math> is empty, then the extension of <math>P</math> to <math>S</math> is ''not'' the whole ring. Moreover, <math>P</math> equals the contraction of its extension to <math>S</math>. | |||
''Proof'': Consider the ideal <math>U^{-1}P</math>. We need to show that it contracts back to precisely <math>P</math> (that'll also show that it is proper). Suppose <math>a/1 \in U^{-1}P</math>. We want to show that <math>a \in P</math>. | |||
There exists <math>b \in P, c \in U</math> such that <math>a/1 = b/c</math>, which in turn means there exists <math>s \in U</math> such that <math>acs = bs</math>. The right side is in <math>P</math>, so the left side must also be in <math>P</math>. Since <math>c,s \in U</math>, we get <math>cs \in U</math>. Further, since <math>U \cap P</math> is empty, we get <math>cs \notin P</math>, so by primeness of <math>P</math>, we have <math>a \in P</math>, as desired. | |||
===Proof of set-theoretic statement=== | |||
Revision as of 22:07, 15 March 2008
This article gives the statement, and possibly proof, of a fact about how a property of a homomorphism of commutative unital rings, forces a property for the induced map on spectra
View other facts about induced maps on spectra
Statement
Set-theoretic statement
Suppose is a commutative unital ring, is a multiplicatively closed subset of and is the localization of at the multiplicatively closed subset . Then the induced map on spectra:
is injective. In fact:
- The image of this map is those primes that are disjoint from
- The inverse image of a prime ideal is precisely the prime ideal i.e. the extension of to
Topological statement
Further, it is also true that the topology on is such that if we give the subspace topology to its image in , the map is a homeomorphism.
Proof
Some preliminary observations
- The extension of an ideal of , to the ring , is the ideal of
- If any ideal of intersects the multiplicatively closed subset , then its extension to is the whole ring . Hence, it does not occur as the contraction of any ideal of
Proof: If is an ideal of such that is nonempty, then pick . Clearly, we have:
Thus, is in the extension of to .
- The converse is not true for arbitrary ideals; however, it is true for prime ideals (it is true for all ideals if is a saturated subset). Formally, if is a prime ideal of such that is empty, then the extension of to is not the whole ring. Moreover, equals the contraction of its extension to .
Proof: Consider the ideal . We need to show that it contracts back to precisely (that'll also show that it is proper). Suppose . We want to show that .
There exists such that , which in turn means there exists such that . The right side is in , so the left side must also be in . Since , we get . Further, since is empty, we get , so by primeness of , we have , as desired.