Prime ideals

Arun Ram
Department of Mathematics and Statistics
University of Melbourne
Parkville, VIC 3010 Australia
aram@unimelb.edu.au

Last update: 12 April 2012

Prime ideals

Let R be a commutative ring.

Let R be a ring and let 𝔞 be an ideal of R.

  1. nil ( R ) = { nilpotent elements of R } = 0 = 𝔭 Spec ( R ) 𝔭 .
  2. a = r R r k 𝔞 for some k > 0 = 𝔭 Spec ( R ) 𝔭 𝔞 𝔭 .

Proof.
b. Follows from a. and the correspondence between ideals of R containing 𝔞 and ideals of R/𝔞.

a. If x 0 then x k = 0 for some k > 0 . If 𝔭 is a prime ideal then 0 𝔭 and so x 𝔭 . So x 𝔭 . If y 0 then y k k > 0 is a multiplicative subset of R that does not contain 0. By Zorn's lemma there exists an ideal 𝔭 of R maximal with respect to the condition that 𝔭 y k k > 0 = . Let x , z R such that x 𝔭 and z 𝔭 . Then 𝔭 + ( x ) 𝔭 and 𝔭 + ( z ) 𝔭 . So y m = p 1 + r 1 x and y n = p 2 + r 2 z for some p 1 , p 2 𝔭 , r 1 , r 2 R , and some m , n > 0 . So y m + n 𝔭 + ( x z ) . So x z 𝔭 .

Let R be a commutative ring.

Let X = Spec ( R ) . If E is a subset of R then V ( E ) = V ( E ) = V E , where E is the ideal generated by E. Also V ( 0 ) = X , V ( 1 ) = , V i M i = V i M i = i V ( M i ) , V ( 𝔞 𝔞 ' ) = V ( 𝔞 𝔞 ' ) = V ( 𝔞 ) V ( 𝔞 ' ) , for any family { M i } of subsets of R and any ideals 𝔞 and 𝔞' of R.

Let R be a commutative ring. Then V : { ideals of R } { closed sets in Spec ( R ) } , and I : { subsets of Spec ( R ) } { radical ideals of R } , I ( V ( 𝔞 ) ) = 𝔞 for an ideal 𝔞 of R , and V ( I ( S ) ) = S _ , for a subset S Spec ( R ) , and { closed sets in Spec ( R ) } { radical ideals of R } .

Let f : A B be a ring homomorphism. The map f * : Spec ( B ) Spec ( A ) 𝔟 f - 1 ( 𝔟 ) is a continuous map. The reason that one uses the prime spectrum instead of the maximal spectrum is that the map f * is not well defined on maximal ideals: if 𝔟 is a maximal ideal f - 1 ( 𝔟 ) may not be maximal.

Any ring homomorphism f : A B can be factored as A im f B where the first map is surjective and the second is injective.

If f : A B is surjective then f * : Spec ( B ) Spec ( A ) is injective with im f * = V ( ker f ) . When f : A B is injective, in other words, if A B is a ring extension, the situation is more subtle.

If 𝔭 is a prime ideal of A then A A 𝔭 gives Spec ( A 𝔭 ) Spec ( A ) is injective with image { prime ideals of A contained in 𝔭 } . The best results are obtained when A B is an integral extension.

Let A B be an integral extension.

  1. (Lying over.) The map Spec ( B ) Spec ( A ) 𝔟 𝔟 A is surjective.
  2. (Going up.) The Krull dimension of A and B are equal.

The names of these results comes from the fact that this theorem is often stated in the following form: Let A B be an integral extension.

  1. (Lying over.) If 𝔭 is a prime ideal of A, then there is a prime ideal 𝔮 of B such that 𝔮 A = 𝔭 , Spec ( B ) Spec ( A ) 𝔮 𝔭
  2. (Going up.) If 𝔭 1 𝔭 2 are prime ideals of A and 𝔮1 is a prime ideal of B such that Spec ( B ) Spec ( A ) 𝔭 2 𝔮 1 𝔭 1 then there exists a prime ideal 𝔮2 of B such that Spec ( B ) Spec ( A ) 𝔮 2 𝔭 2 𝔮 1 𝔭 1 If Q1 is lying over P1 and P 1 P 2 then there is an ideal Q2 of B lying over P2 such that Q 1 Q 2 .
  3. (Incomparability.) If P is an ideal of A and Q1 and Q2 are prime ideals of B lying over P then Q 1 Q 2 .
  4. (Going down.) If Q2 is lying over P2 and P 1 P 2 then there is an ideal Q1 of B lying over P1 such that Q 1 Q 2 .
WHAT IS THE PROPER STATEMENT OF THESE IN TERMS OF Spec(R)?

Prime and maximal ideals

Let A be a commutative ring.

HW: [AM, Ch.1, Ex.26] Let Y be a compact Hausdorff topological space, C(Y) = {f:Y | f is continuous }, X = Spec(C(Y)), and |X| = Maxspec(C(Y)). For yY let evy: C(Y) f f(y) so that   evy(f) = f(y). Then Y |X| y ker(evy) is a homeomorphism where the topology on |X| has basic open sets {x|X| | fx}, for   fC(Y).

Let A be a commutative ring.

HW: Show that nil(A) = {nilpotent elements of A}.

HW: Let 𝔞 be an ideal of A. Show that 𝔞 is the ideal of A corresponding to the ideal nil(A𝔞) of the ring A𝔞.

Define V: {subsets of A} {subsets of Spec(A)} S V(S) = {xSpec(A) | xS} I: {subsets of Spec(A)} {subsets of A} E I(E) = xEx

HW: Let SA. Show that V(S) = V(S) = V(S), where S is the ideal of A generated by S.

HW: Let ESpec(A). Show that I(E) = I(E_), where E_ is the closure of E in the Zariski topology on Spec(A).

HW: Let SA and ESpec(A). Show that I(V(S)) = S and V(I(E)) = E_.

HW: Let A be a commutative ring and let X=Spec(A). Show that nil(A) = I(X).

Notes and References

Where are these from?

References

References?

page history