]> Number Fields - Number Fields

Number Fields

An (algebraic) number field is a subfield of whose degree over is finite.

It turns out that number fields are Dedekind domains thus all their ideals factor uniquely into prime ideals. An example of a ring where this is not true is [3 ]: take the ideal I=2 ,1 +3 . Then I2 , but I 2 =2 I. Furthermore, I is the unique prime ideal containing 2 and hence 2 is not the product of prime ideals.

Example: Let ω=e 2 πi/m for an integer m. Then [ω] is the mth cyclotomic field. The mth and 2 mth cyclotomic fields are identical because we have ω=ω m+1 =(ω 2 ) (m+1 )/2 [ω 2 ].

Example: The quadratic fields are the number fields [m] for squarefree m (and these are exactly the number fields with degree 2 over ). The real quadratic fields are the [m] with m>0 , and the imaginary quadratic fields are those with m<0 . Notice the m=1 ,3 cases are also cyclotomic fields. By considering the equation m=a+bn it can be seen that the fields [m] are pairwise nonisomorphic.

An algebraic integer is a complex number that is a root of a monic polynomial in [x].

Theorem: The minimal polynomial of an algebraic integer over has coefficients in .

Proof: This follows from the following lemma.

Lemma: Let f[x] be monic. Then if f=gh for monic g,h[x] then g,h[x].

Proof: (This is a special case of Gauss' Lemma.) Let m,n be the smallest positive integers such that mg,nh[x], so the coefficients of mg,nh have no common factor. (For example, if mg did have a common factor d, we could replace m by m/d; we know dm because the leading coefficient of mg is m since g is monic.) Now if mn>1 , then let p be some prime dividing mn. Consider the equation mnf=(mg)(nh)=0 (modp). Since p[x] is an integral domain, this means (mg)=0 or (nh)=0 , implying that p divides all the coefficients of mg or nh, which is a contradiction. Thus m=n=1 .

Corollary: The only algebraic integers in are the ordinary integers.

Corollary: Let m be a squarefree integer. The set of algebraic integers in quadratic field [m] is

{a+bm:a,b} ifm=2 ,3 (mod4 ); {a+bm2 :a,b,a=b(mod2 )} ifm=1 (mod4 ).

Proof: Let α=r+sm for rationals r,s. If s0 then the minimal polynomial of α over is

x 2 2 rx+r 2 ms 2

Thus α is integral if and only if 2 r and r 2 ms 2 are integers, implying the results above.

Theorem: Let a. The following are equivalent.

  1. a is an algebraic integer.

  2. [a] is finitely generated as an additive group.

  3. a belongs to a subring of that has a finitely generated additive group.

  4. aAA for some finitely generated additive subgroup A.

Proof: (1) (2): If n is the degree of a over then [a] is generated by 1 ,a,...,a n1 .

(2) (3) (4) is clear.

(4) (1): Let a 1 ,...,a n generate A. We may write

(aa 1 aa n)=M(a 1 a n)

where M is some n×n matrix with integer entries. Thus

(aIM)(a 1 a n)=0

But since the vector (a 1 ...a n) is nonzero, we must have that det(aIM)=0 . This is a monic equation in a (of degree n), so a must be an algebraic integer.

Corollary: If a,b are integral, so are a+b,ab

Proof: Since [a],[b] are finitely generated, we have that [a,b] is finitely generated.

Thus the set of algebraic integers in form a ring, which we shall denote by 𝔸. We call a the subring of algebraic integers of K the number ring corresponding to K.

In other texts the ring of algebraic integers is denoted , and the number ring corresponding to a field K is denoted K.