[summary: An integral domain is a ring in which the only way to make 0 as a product is to multiply 0 by something. For instance, in an integral domain like [48l Z], 2×3 is not equal to 0 because neither 2 nor 3 is.]
[summary(Technical): An integral domain is a ring in which ab=0 implies a=0 or b=0. (We exclude the ring with one element: that is conventionally not considered an integral domain.)]
In keeping with ring theory as the attempt to isolate each individual property of [48l Z] and work out how the properties interplay with each other, we define the notion of integral domain to capture the fact that if a×b=0 then a=0 or b=0. That is, an integral domain is one which has no "zero divisors": 0 cannot be nontrivially expressed as a product. (For uninteresting reasons, we also exclude the ring with one element, in which 0=1, from being an integral domain.)
Examples
- Z is an integral domain.
- Any field is an integral domain. (The proof is an exercise.)
%%hidden(Show solution): Suppose ab=0, but a≠0. We wish to show that b=0.
Since we are working in a field, a has an inverse a−1; multiply both sides by a−1 to obtain a−1ab=0×a−1. Simplifying, we obtain b=0. %%
- When p is a prime integer, the ring Zp of integers mod p is an integral domain.
- When n is a [composite_number composite] integer, the ring Zn is not an integral domain. Indeed, if n=r×s with r,s positive integers, then rs=n=0 in Zn.
Properties
The reason we care about integral domains is because they are precisely the rings in which we may cancel products: if a≠0 and ab=ac then b=c. %%hidden(Proof): Indeed, if ab=ac then ab−ac=0 so a(b−c)=0, and hence (in an integral domain) a=0 or b=c.
Moreover, if we are not in an integral domain, say rs=0 but r,s≠0. Then rs=r×0, but s≠0, so we can't cancel the r from both sides. %%
Finite integral domains
If a ring R is both finite and an integral domain, then it is a field. The proof is an exercise. %%hidden(Show solution): Given r∈R, we wish to find a multiplicative inverse.
Since there are only finitely many elements of the ring, consider S={ar:a∈R}. This set is a subset of R, because the multiplication of R is closed. Moreover, every element is distinct, because if ar=br then we can cancel the r (because we are in an integral domain), so a=b.
Since there are |R|-many elements of the subset S (where |⋅| refers to the Cardinality), and since R is finite, S must in fact be R itself.
Therefore in particular 1∈S, so 1=ar for some a. %%