Localization of a ring

The ring S^{-1}R obtained from a ring R by inverting a multiplicative set S.
Localization of a ring

Let RR be a and let SRS\subseteq R be a . The localization of RR at SS is a ring, denoted S1RS^{-1}R, together with a ring map

ι:RS1R \iota:R\longrightarrow S^{-1}R

such that every element of SS becomes a unit in S1RS^{-1}R (see ).

Construction (fractions)

As a set, S1RS^{-1}R can be constructed from pairs (r,s)R×S(r,s)\in R\times S modulo the equivalence relation

(r,s)(r,s)tS such that t(rsrs)=0 in R. (r,s)\sim (r',s') \quad \Longleftrightarrow \quad \exists\,t\in S\text{ such that } t(rs'-r's)=0 \text{ in }R.

Write the class of (r,s)(r,s) as rs\frac{r}{s}. Addition and multiplication are defined by

rs+rs=rs+rsss,rsrs=rrss. \frac{r}{s}+\frac{r'}{s'}=\frac{rs'+r's}{ss'},\qquad \frac{r}{s}\cdot\frac{r'}{s'}=\frac{rr'}{ss'}.

The canonical map is ι(r)=r1\iota(r)=\frac{r}{1}.

If 0S0\in S, then ι(0)\iota(0) is invertible, hence 1=01=0 in S1RS^{-1}R; in this case S1RS^{-1}R is the zero ring.

Universal property

The localization is characterized by the following universal mapping property:

If AA is any commutative ring and φ:RA\varphi:R\to A is a ring homomorphism such that φ(s)\varphi(s) is a unit of AA for every sSs\in S, then there exists a unique ring homomorphism φ~:S1RA\widetilde\varphi:S^{-1}R\to A with φ~ι=φ\widetilde\varphi\circ\iota=\varphi. Explicitly,

φ~ ⁣(rs)=φ(r)φ(s)1. \widetilde\varphi\!\left(\frac{r}{s}\right)=\varphi(r)\,\varphi(s)^{-1}.

This same “invert SS” idea for modules is treated in , and it can be viewed as a special case of .

A basic structural fact is that primes of S1RS^{-1}R correspond to primes of RR disjoint from SS; see .

Examples

  1. Inverting a prime number. Take R=ZR=\mathbb Z and S={1,p,p2,}S=\{1,p,p^2,\dots\}. Then

    S1ZZ ⁣[1p]={apn:aZ, n0}Q. S^{-1}\mathbb Z \cong \mathbb Z\!\left[\frac{1}{p}\right] =\left\{\frac{a}{p^n}:a\in\mathbb Z,\ n\ge 0\right\}\subseteq\mathbb Q.
  2. Laurent polynomials. If R=k[x]R=k[x] and S={1,x,x2,}S=\{1,x,x^2,\dots\}, then

    S1Rk[x,x1], S^{-1}R \cong k[x,x^{-1}],

    since xx becomes invertible.

  3. Localizing at a prime ideal. If pR\mathfrak p\subset R is prime and S=RpS=R\setminus\mathfrak p, then S1RS^{-1}R is the RpR_{\mathfrak p}, which is a .