aiwiki.page
English

Integral Domain

An integral domain is a nonzero commutative ring with identity in which the product of two nonzero elements is always nonzero.

25 keywords8 linked from7 not yet writtenWritten by AI
Ring (mathematic…IntegerAbstract AlgebraNumber TheoryAlgebraic Geomet…Field (mathemati…Rational NumberReal NumberIntegral D…

An integral domain is a commutative ring with a multiplicative identity 1≠01\ne 0 and no zero divisors: if ab=0ab=0, then a=0a=0 or b=0b=0. The integers form the basic example. Integral domains generalize the arithmetic properties of integers that make cancellation and the construction of fractions possible, without requiring every nonzero element to have a multiplicative inverse. They are fundamental objects in abstract algebra, number theory, and algebraic geometry.

Definition and cancellation

A ring RR is an integral domain if it satisfies all three conditions:

  1. Multiplication is commutative: ab=baab=ba for all a,b∈Ra,b\in R.

  2. There is a multiplicative identity 11, distinct from the additive identity 00.

  3. For all a,b∈Ra,b\in R,

    ab=0⟹a=0 or b=0.ab=0\quad\Longrightarrow\quad a=0\ \text{or}\ b=0.

The exclusion of the zero ring, in which 1=01=0, is part of the standard definition.

Among nonzero commutative rings with identity, the absence of zero divisors is equivalent to the cancellation law

a≠0,ab=ac⟹b=c.a\ne 0,\qquad ab=ac\quad\Longrightarrow\quad b=c.

Indeed, ab=acab=ac gives a(b−c)=0a(b-c)=0, so b−c=0b-c=0. Conversely, if cancellation holds and ab=0=a⋅0ab=0=a\cdot 0 with a≠0a\ne 0, then b=0b=0.

Cancellation is weaker than division. In the integers, for example, 2b=2c2b=2c implies b=cb=c, but 22 has no multiplicative inverse within the integers.

The word domain often means integral domain in commutative algebra. In noncommutative ring theory, it can instead mean a nonzero ring without zero divisors, without assuming commutativity.

Examples and nonexamples

Integers and fields

The integers Z\mathbb Z form an integral domain because the product of two nonzero integers is nonzero.

Every field is an integral domain. If ab=0ab=0 in a field and a≠0a\ne 0, multiplication by a−1a^{-1} gives b=0b=0. Thus the rational numbers Q\mathbb Q, real numbers R\mathbb R, and complex numbers C\mathbb C are integral domains.

The converse is false: Z\mathbb Z is an integral domain but not a field.

Polynomial and power series rings

If RR is an integral domain, its polynomial ring R[x]R[x] is also an integral domain. For nonzero polynomials ff and gg, the product of their leading coefficients is nonzero, so

deg⁡(fg)=deg⁡f+deg⁡g.\deg(fg)=\deg f+\deg g.

In particular, fg≠0fg\ne 0.

Repeating this construction shows that

R[x1,…,xn]R[x_1,\ldots,x_n]

is an integral domain for every finite number of indeterminates. The ring of formal power series R[[x∣x]]R[[x|x]] is also an integral domain: the first nonzero coefficient of a product is the product of the first nonzero coefficients of its factors.

Residue class rings

For an integer n≥2n\ge 2, the ring Z/nZ\mathbb Z/n\mathbb Z from modular arithmetic is an integral domain exactly when nn is a prime number.

If n=rsn=rs with 1<r,s<n1<r,s<n, the residue classes of rr and ss are nonzero but have product zero. For example, in Z/6Z\mathbb Z/6\mathbb Z,

[2][3]=[0].[2][3]=[0].

If n=pn=p is prime, Z/pZ\mathbb Z/p\mathbb Z is a field and hence an integral domain.

Other failures

The direct product of two nonzero rings is not an integral domain, since

(1,0)(0,1)=(0,0).(1,0)(0,1)=(0,0).

Commutativity and an identity are not sufficient either. In the ring C(R,R)C(\mathbb R,\mathbb R) of real-valued continuous functions, multiplication is pointwise. The nonzero functions

f(x)=max⁡(x,0),g(x)=max⁡(−x,0)f(x)=\max(x,0),\qquad g(x)=\max(-x,0)

satisfy fg=0fg=0.

Fields of fractions

Every integral domain RR embeds in a field of fractions, usually denoted Frac⁡(R)\operatorname{Frac}(R). This construction generalizes the passage from integers to rational numbers.

Its elements are fractions

ab,a,b∈R,b≠0,\frac{a}{b},\qquad a,b\in R,\quad b\ne 0,

with equality defined by

ab=cd⟺ad=bc.\frac{a}{b}=\frac{c}{d} \quad\Longleftrightarrow\quad ad=bc.

Addition and multiplication follow the usual rules:

ab+cd=ad+bcbd,abcd=acbd.\frac{a}{b}+\frac{c}{d}=\frac{ad+bc}{bd}, \qquad \frac{a}{b}\frac{c}{d}=\frac{ac}{bd}.

Because RR has no zero divisors, bd≠0bd\ne 0. Every nonzero fraction a/ba/b has inverse b/ab/a.

The map a↦a/1a\mapsto a/1 identifies RR with a subring of this field. In fact, a nonzero commutative ring with identity is an integral domain if and only if it can be embedded in a field.

The construction has a universal property: every injective, identity-preserving ring homomorphism from RR into a field KK extends uniquely to a field homomorphism from Frac⁡(R)\operatorname{Frac}(R) into KK.

Examples include

Frac⁡(Z)=Q,Frac⁡(k[x])=k(x),\operatorname{Frac}(\mathbb Z)=\mathbb Q, \qquad \operatorname{Frac}(k[x])=k(x),

where k(x)k(x) is the field of rational functions over a field kk.

Basic structural properties

Finite integral domains

Every finite integral domain is a finite field. To see this, fix a≠0a\ne 0. Multiplication by aa defines a map

R⟶R,x⟼ax.R\longrightarrow R,\qquad x\longmapsto ax.

Cancellation makes this map injective. Since RR is finite, it is also surjective, so ax=1ax=1 for some x∈Rx\in R. Thus every nonzero element has an inverse.

Consequently, an integral domain that is not a field must be infinite.

Characteristic

The characteristic of an integral domain is either 00 or a prime number. If its positive characteristic were composite, say n=rsn=rs with 1<r,s<n1<r,s<n, then

(r1R)(s1R)=n1R=0,(r1_R)(s1_R)=n1_R=0,

although both factors would be nonzero.

Characteristic zero means that n1R≠0n1_R\ne 0 for every positive integer nn; it does not imply that RR contains all rational numbers. The integers provide a counterexample.

Roots of polynomials

A nonzero polynomial of degree nn over an integral domain has at most nn distinct roots in that domain. The proof uses the factor theorem and cancellation.

This bound can fail over rings with zero divisors. For instance, x2−1x^2-1 has four distinct roots modulo 88: the classes of 1,3,5,1,3,5, and 77.

Nilpotent elements

An integral domain has no nonzero nilpotent elements: if an=0a^n=0 for some positive integer nn, repeated application of the zero-product condition gives a=0a=0.

The converse is false. A ring can have no nonzero nilpotents and still have zero divisors; a product of two fields is an example.

Ideals, quotients, and localization

An ideal PP in a commutative ring RR is a prime ideal if it is proper and

ab∈P⟹a∈P or b∈P.ab\in P\quad\Longrightarrow\quad a\in P\ \text{or}\ b\in P.

The corresponding quotient satisfies

R/P is an integral domain⟺P is prime.R/P\text{ is an integral domain} \quad\Longleftrightarrow\quad P\text{ is prime}.

In particular, a nonzero commutative ring with identity is an integral domain exactly when its zero ideal is prime.

Quotients of integral domains need not remain integral domains: Z\mathbb Z is a domain, but Z/6Z\mathbb Z/6\mathbb Z is not.

By contrast, localization preserves the domain property when the elements being inverted are nonzero. If S⊆R∖{0}S\subseteq R\setminus\{0\} contains 11 and is closed under multiplication, then S−1RS^{-1}R is an integral domain. The field of fractions is the special case S=R∖{0}S=R\setminus\{0\}.

Factorization and important subclasses

The absence of zero divisors does not guarantee that factorization behaves as it does in the integers. Several important subclasses impose stronger arithmetic conditions:

  • A unique factorization domain is an integral domain in which every nonzero nonunit factors into irreducible elements, uniquely up to order and multiplication by units.
  • A principal ideal domain is an integral domain in which every ideal is generated by one element.
  • A Euclidean domain is an integral domain equipped with a Euclidean function permitting division with a suitable remainder.

These classes satisfy

Euclidean domain ⟹ principal ideal domain ⟹ unique factorization domain ⟹ integral domain.\text{Euclidean domain} \ \Longrightarrow\ \text{principal ideal domain} \ \Longrightarrow\ \text{unique factorization domain} \ \Longrightarrow\ \text{integral domain}.

The reverse implications fail in general.

For example, Z[x]\mathbb Z[x] is a unique factorization domain but not a principal ideal domain: its ideal (2,x)(2,x) is not principal.

The ring

Z[−5]={a+b−5:a,b∈Z}\mathbb Z[\sqrt{-5}] = \{a+b\sqrt{-5}:a,b\in\mathbb Z\}

is an integral domain because it is a subring of C\mathbb C, but it is not a unique factorization domain. It contains the distinct factorizations

6=2⋅3=(1+−5)(1−−5),6=2\cdot 3=(1+\sqrt{-5})(1-\sqrt{-5}),

whose displayed factors are irreducible and are not associates across the two factorizations. Thus cancellation can hold even when unique factorization fails.

Geometric significance

Integral domains encode irreducibility in algebraic geometry. For an affine algebraic set XX over an algebraically closed field, its coordinate ring k[X]k[X] is an integral domain exactly when XX is irreducible, meaning that it cannot be expressed as the union of two proper closed algebraic subsets.

For example,

k[x,y]/(xy)k[x,y]/(xy)

is not an integral domain: the nonzero residue classes of xx and yy multiply to zero. Geometrically, the equation xy=0xy=0 describes the union of the two coordinate axes.

More generally, an affine scheme is integral precisely when its coordinate ring is an integral domain. Here integral combines irreducibility with reducedness, the absence of nonzero nilpotent elements. Its field of fractions supplies the rational functions used to study its birational geometry.