Let R be a commutative ring and S⊆R a multiplicative set
. Write S−1R for the localization
.
Theorem (prime correspondence)
There is an inclusion-preserving bijection
{prime ideals P⊂S−1R}⟷{prime ideals p⊂R with p∩S=∅}given by:
- contraction: P↦p={r∈R:r/1∈P},
- extension: p↦S−1p={r/s:r∈p,s∈S}.
In particular, if p⊂R is prime
and we localize at p (so Rp is the localization at \(\mathfrak p\)
), then prime ideals of Rp correspond exactly to prime ideals q⊆p in R.
Maximal ideals correspond similarly: maximal ideals of S−1R correspond to maximal ideals
of R disjoint from S.
Examples
Invert 2 in Z.
Let S={1,2,4,8,…}. Then S−1Z=Z[1/2].
Prime ideals of Z[1/2] correspond to primes p⊂Z with p∩S=∅, i.e. (0) and (p) for odd primes p. The prime (2) disappears because it meets S.
Localize Z at (p).
In R=Z, take S=Z∖(p), so S−1R=Z(p).
Prime ideals of Z(p) correspond to primes of Z contained in (p), namely (0) and (p). Thus Spec(Z(p))={(0),(p)}.
Localize k[x,y] at powers of x.
Let R=k[x,y] and S={1,x,x2,…}. Then S−1R=Rx.
Prime ideals in Rx correspond to prime ideals of R not containing x. For instance, (y) survives (it does not contain x), while (x,y) does not (it contains x, hence meets S).