1 FIELD EXTENSIONS

In this chapter, we will study field extensions and its related concepts. We
know many examples of fields that are subfields of larger fields. In the present
section, we will examine these inclusions of fields and of their important
properties. Let us begin our discussion with some important terminologies:

2 Characteristic of field
Recall that a field F is a commutative ring with identity in which every
nonzero element has an inverse. Equivalently, the set F × = F − {0} of
nonzero elements of F is an abelian group under multiplication. One of
the first invariants associated with any field F is its characteristic, defined
as follows: If 1F denotes the identity of F , then F contains the elements
1F , 1F +1F , 1F + 1F + 1F , · · · of the additive subgroup of F generated by
1F . which may not all be distinct. For some positive integer n, we have
n.1F = 1F + · · · + 1F (n times). Then two possibilities arise: either all the
elements n.1F are distinct, or else n.1F = 0 for some positive integer n.
Definition 2.1. The characteristic of a field F , is defined to be the smallest
positive integer p such that p.1F = 0 if such a p exists and is defined to be 0
otherwise. It is denoted char(F ).
For positive integers m and n, it is easy to see that
n.1F + m.1F = (m + n).1F 0
and

(n.1F )(m.1F ) = mn.1F .

It follows that the characteristic of a field is either 0 or a prime p (hence
the choice of p in the definition above), since if n = ab is composite with
n.1F = 0, then ab.1F = (a.1F )(b.1F ) = 0 and since F is a field, one of
a.1F or b.1F is 0, so the smallest such integer is necessarily a prime. It also
follows that if n.1F = 0, then n is divisible by p.

Now, we discuss some important result related to characteristic of rings

and fields.

Theorem 2.2. A ring R 6= {0} of positive characteristic having an identity

and no zero divisors must have prime characteristic.
Proof. Since R contains nonzero elements, R has characteristic n ≥ 2. If n
were not prime, we could write n = km with k, m ∈ Z, 1 < k, m < n. Then
0 = ne = (km)e = (ke)(me), and this implies that either ke = 0 or me = 0
since R has no zero divisors. It follows that either kr = (ke)r = 0 for all
r ∈ R or mr = (me)r = 0 for all r ∈ R, in contradiction to the definition of
the characteristic n.

Corollary 2.3. A finite field has prime characteristic.

Proof. By Theorem 2.2 is suffices to show that a finite F has a positive
characteristic. Consider the multiplies e, 2e, 3e, · · · of the identity. Since F
contains only finitely many distinct elements, there exist integers k and m
with 1 ≤ k ≤ m such that ke = me, or (m − k)e = 0, and so F as a positive
characteristic.

The finite field Z/(p) (or, equivalently Fp ) obviously has characteristic

p, whereas the ring Z of integers and the field Q of rational numbers have
characteristic 0. We note that in a ring R of characteristic 2 we have 2a =
a + a = 0, hence a = −a for all a ∈ R. A useful property commutative rings
of prime characteristic is the following.
Theorem 2.4. Let R be a commutative ring of prime characteristic p. Then
n n n n n n
(a + b)p = ap + bp and (a − b)p = ap − bp

for a, b ∈ R and n ∈ N.
Proof. We use the fact that
 
p p(p − 1) · · · (p − i + 1) ∼
= = 0 mod p
i 1.2. · · · .i

for all i ∈ Z with 0 < i < p, which follows from pi being an integer and the

observation that the factor p in the numerator can not be cancelled. Then
by binomial theorem
   
p p p p−1 p
(a + b) = a + a b + ··· + abp−1 + bp = ap + bp ,
1 p−1
and induction on n completes the proof of the first identity. By what we
have shown, we get
n n n n
ap = ((a − b) + b)p = (a − b)p + bp ,

and the second identity follows.

2
Proposition 2.5. The characteristic of a field F is either 0 or a prime p.
If char(F ) = p, then for any α ∈ F, we have

p.α = α + α + · + α = 0
| {z }
p times
.

Proof. Only the second statement has not been proved, and this follows
immediately from the evident equality p.α = p.(1F α) = (p.1F )(α) in F .

Remark 2.6. This notion of a characteristic makes sense also for any inte-
gral domain and its characteristic will be the same as for its field of fractions.

Example 2.7. (i) The fields Q and R both have characteristic 0:

char(Q) = char(R) = 0. The integral domain Z also has characteristic
0.

(ii) The (finite) field Fp = Z/pZ has characteristic p for any prime p.

(iii) The integral domain Fp [x]of polynomials in the variable x with coeffi-
cients in the field Fp has characteristic p, as does its field of fractions
F[ x] (the field of rational functions in x with coefficients in Fp ).

If we define (−n).1F = −(n.1F ) for positive n and 0.1F = 0, then we have a

natural ring homomorphism,

ϕ:Z→F

n 7→ n.1F
and we can interpret the characteristic of F by noting that ker(ϕ) =
char(F )Z. Taking the quotient by the kernel gives us an injection of either
Z or Z/pZ into F (depending on whether char(F ) = 0 or char(F ) = p).
Since F is a field, we see that F contains a subfield isomorphic either to Q
(the field of fractions of Z) or to FP = Z/pZ (the field of fractions of Z/pZ
) depending on the characteristic of F , and in either case is the smallest
subfield of F containing 1.F (the field generated by 1.F in F ).

3
3 Sub Fields
Definition 3.1. A non-empty subsets S of a field F is called a subfield of
F if it is a field with respect to the operations on F . If S 6= F , then S is
called a proper subfield of F .

Remark 3.2. A non-empty subset S of a field F is a subfield of F iff

(i) S is subgroup of (F, +), and
(ii) the set of all non-zero elements of S forms a subgroup of the group of
non-zero elements of F under multiplication.
Theorem 3.3. A non-empty subset S of a field F is a subfield of F if and
only if
(i) a ∈ S, b ∈ S =⇒ a − b ∈ S
(ii) a ∈ S, b ∈ S, b 6= 0 =⇒ ab−1 ∈ S
Proposition 3.4. Let F be any field. Prove that the intersection of all its
subfields is a prime filed.
n
T
Proof. Let S1 , S2 , . . . Sn be subfield of F . Then S1 ∩ S2 ∩ . . . , ∩Sn = Sn .
i=1
n
T n
T
Clearly, 6= φ (∵ 0, 1 ∈ Si , i = 1, 2, . . . , n). Let x, y ∈ Si
i=1 i=1
⇒ x, y ∈ Si , i = 1, 2, . . . , n
x − y ∈ Si , i = 1, 2, . . . , n
n
T Tn
⇒x−y ∈ Si , ∀x, y ∈ Si
i=1 i=1
Also if y 6= 0, then xy −1 ∈ Si i = 1, 2, . . . , n
n n
⇒ xy −1 ∈
T T
Si for all x, y ∈ Si . Hence, Si is a subfield of F.
i=1 i=1

Exercise 3.5. Show that the union of two subfileds need not be subfield.

4 Prime Fields
This section deals with study of special kind of fields. Let P be the
intersection of all subfields of a field F, then P is a subfiled of F and every
subfield of F contains P . So that no subfield of F is contained properly
in P . Hence P has no proper subfield. This leads us to define concept of
prime subfield.

4
Definition 4.1. Let < F, +, . > be a field. Then, F is calleed a prime field
if it has no proper subfield (or A field F is said to be a prime field if it has
no proper subfield).
Example 4.2. For any prime p, the finite field Zp is a prime field. The field
of rational number Q is also a prime field.
Let p be any prime integer. then < p > is a maxima ideal of Z and
Z/ < p > is a field having p elements. If S is any subfield of Z/ < p >
it has at least two element, then as a group of prime order has no proper
subgroups, S = Z/ < p >. This proves Z/ < P > is a prime field.
Example 4.3. Let us consider the field of rationals Q. Let F 0 be a subfield
of Q, then 1 ∈ F 0 . Next, let m n ∈ Q for all m, n ∈ Z and n 6= 0. Now
we write m = m.1 ∈ F 0 and n = n.1 ∈ F 0 . But n 6= 0, we conclude that
n = n.1 6= 0 ∈ F 0 . This yields that m 0
n ∈ F . This prove that Q is a prime
field.
Our next result shows that Q and Zp (where p is any prime) are only
the prime fields.
Theorem 4.4. Every field contains a subfield isomorphic to Q or to Zp for
some prime number p.
Proof. Let F be a field. Define a function

f : Z → F : f (n) = n.1 = 1 + 1 + . . . + 1(n times).

It is easy to prove that f is a ring homomorphism and Kerf = pZ =

char(F )Z, where p is the characteristic of F . We know that charF =0 or
char F = p, a prime number. Let us discuss these two cases separately.

Case(i) (char F =0) : In this case f is one-one. ∴ Z = f (Z). Thus,

f (Z) is an integral domain contained in the field F . Since F is a field, it
will also containthe field of quotients of f (Z). This will be isomorphic to the
field of quotients of Z, i.e., Q. Thus, F has a subfield which is isomorphic
to Q.

Case(ii) (char F =p, for some prime p) : It is well known that for
any prime number p, Z/pZ is a field. Hence, by applying Fundamental
theorem of Homomorphism to f , we conclude Z/pZ ' f (Z) . Thus, f (Z) is
isomorphic to Zp and is conatained in F . Consequently, F has a subfield
isomorphic to Zp .

5
 Let us reword the Theorem 4.4 slightly. What it says that :

Theorem 4.5. Let F be a field. Then, the following hold:

(i) If char (F ) = 0, then F has a subfield isomorphic to Q.

(ii) If char (F ) = p, then F has a subfield isomorphic to Zp

Following are immediatly consequences of Theorem 4.4.

Corollary 4.6. Every prime field of characteristic 0 is isomorphic to the

field of rationals.

Corollary 4.7. Every prime field of characteristic p(where p is prime) is

isomorphic to the finite field Zp .

6
5 Field Extentions and its constructions
Definition 5.1. A field K is called a field extension of F if F is a
subfield of K. Consequently, we say that F is a subfield of K and that K is
an extension of E.

Remark 5.2. Let K and F be any two fields such that F ⊆ K.

(i) If K is an extension of F, then K is a vector space over F . That is,

K(F ) is a vector space. Hence, the dimension of K over F is denoted
by [K : F ]. Moreover, we write

dim K(F ) = [K : F ] = degree of K over F.

(ii) If K, as a vector space over F , is finite dimensional, the K is known

as finite extension of F.

Definition 5.3. Let K be an extension of F . Then, the dimension of K

considered as a vector space over F is called the degree of K over F . Con-
sequently, the degree of K over F is expressed as [K : F ]. If [K : F ] < ∞,
then K is called a finite extension of F, otherwise K is called an infinite
extension of F .

Example 5.4. Let F be any field. Then, F is a subfield of itself. Therefore,

F is an extension of the field F . Moreover, {1} being the basis of the vector
space F (F ). Consequently, [F : F ] = 1 < ∞. Thus, F (F ) is finite dimen-
sional and hence F is a finite extension over itself. In particular, Q is an
finite extension of Q. Similarly, we can say that R(R) and C(C) are finite
extensions.

Example 5.5. For the real field R and complex field C, we have R ⊂ C.
Therefore, we obtain [C : R] = 2, since {1, i} is a basis of C(R). Conse-
quently, C is finite extension of R. Similarly, C is a field extension of Q.
However, Q ⊆ R is not a field extension.

Example √ 5.6. Let √ Q be the field of rationals. Let us consider the √ field
F = Q( 3) = {a + b 3 : a, b ∈ Q}. √ Then, it is easy to verify that Q( 3) is
√ Also, notice that Q ⊆ Q( 3). Further,
a field. √ it is easy to check that the set
{1, √3} is a basis of the vector
√ space Q( 3) over the field Q. Thus, we have
[Q( 3), Q] = 2. Hence, Q( 3) is a finite field extension of Q. Similarly,
√
Q( p) (where p is any prime) is a field extension of Q.

7
5.1 Constructions of field Extensions
In the present section, we construct Field Extensions from given fields. Let
us look at a particular field extension of a field F . Since F [x] is an integral
domain, we can obtain its field of quotient. We denote this field by F(x).
Then F is a subfield of F (x) i.e., F ⊆ F (x). Hence, F (x) is a field extension
of a given field F . Moreover, its elements are expressions of the form fg(x)(x)
,
where f (x), g(x) ∈ F [x] and g(x) 6= 0. In√particular, let √ Q be a field of
rationals. Then it is easy to check that Q( 2) = {a + b 2 : a, b ∈ Q} is a
field extension of Q.

An important class of field extensions are those obtained by trying to

solve equations over a given field F . For example, if F = R is the field of
real numbers, then the simple equation x2 + 1 = 0 does not have a solution
in F . The question arises whether there is some larger field containing R
in which this equation does have a solution, and it was this question that
led Gauss to introduce the complex numbers C = R + Ri, where i is defined
so that i2 + 1 = 0. One then defines addition and multiplication in C by
the usual rules familiar from elementary algebra and checks that in fact C
so defined is a field, i.e., it is possible to find an inverse for every nonzero
element of C .

There is another way of construction of a field extension of a field F

from F [x], the ring of polynomials over F . We can construct at quotient
ring of F [x] by its maximal ideal. It is well known that an ideal is maximal
in F [x] iff it is generated by an irreducible polynomial over F . Precisely, we
have the following.

Theorem 5.7. Let F be a field and F [x] be a ring of polynomial over F . If

f (x) is irreducible over F , then K = F [x]/ < f (x) > is a field extension of
F.

Proof. Given any f (x) ∈ F [x] such that deg f (x) > 0. Then, we will show
that there is a field monomorphism from F into F [x]/ < f (x) >. This will
show that F [x]/ < f (x) > contains an isomorphic copy of F ; and hence, we
can say that it contains F . So, let us define φ : F → F [x]/ < f (x) > by
φ(a) = a+ < f (x) > . Then, it is easy to check that

φ(a + b) = φ(a) + φ(b)and φ(ab) = φ(a)φ(b).

Consequently, φ is a ring homomorphism.

8
To find Ker φ, we have

Ker φ = {a ∈ F | a+ < f (x) >=< f (x) >}

= {a ∈ F | a ∈< f (x) >}
= {a ∈ F | f (x) | a}
= {0}, since deg f >0 and deg a ≤ 0.

Therefore, φ is 1-1, and hence an inclusion. Consequently, F is embedded in

F [x]/ < f (x) >. Thus, if f (x) is irreducible in F [x], then F [x]/ < f (x) > is
a field extension of F .

We will discuss some applications of the above mentioned result.

Exercise 5.8. Check which of the following rings are field extensions of Q?

(i) Q[x]/ < x3 + 10 >

(ii) Q[x]/ < x2 + 2 >

(iii) Q[x]/ < x2 − 5x + 6 >

Proof. To prove Q[x]/ < x3 + 10 > is field extension of Q.

Let f (x) = x3 + 10. Take p = 3 a prime number. Then, f (x) is irreducible
over Q by Eisenstein criterion. In view of Theorem 2.7, we conclude that
Q[x]/ < x3 + 10 > is field extension of Q. Similarly, we can prove that
R[x]/ < x2 + 2 > is field extension of Q.

On the other hand, if we suppose f (x) = x2 − 5x + 6 = (x − 2)(x − 3),

then f (x) is reducible over Q. Consequently, Q[x]/ < x2 − 5x + 6 > is not a
field extension of Q.

Exercise 5.9. Let R be a real field and f (x) = x2 + 1 be polynomial over R

of degree 2. Find field extension of R.

9
6 Algebraic and transcedental numbers
Definition 6.1. (Algebraic, transcendental). Let F ⊆ E be a field extension.
An element u ∈ E is called algebraic over F if f (u) = 0 for some nonzero
polynomial f (x) ∈ F[x]. If u ∈ E is not algebraic over F, then it is called
transcendental over F. If every element of F is algebraic over E, then we say
F ⊆ E is an algebraic extension.

Example 6.2. We see that R ⊆ C is an algebraic extension since a+bi, a, b ∈

R, is the root of x2 ?2ax + a2 + b2 .
√
Example 6.3. (a) The real number 3 2 is algebraic over Q since it is a
root of x3 ?2.
p √
(b) The real number 5 + 2 is algebraic over Q since it is a root of
x4 ?10x2 + 23.
√ √
(c) The real number 2 − 3 is algebric over Q since it is a root of x4 −
10x2 + 1.
It is natural ask if Q ⊆ R is an algebric extension. The answer is no.
This follows from the next theorem.

Theorem 6.4. (Hermite 1873, Lindemann 1882). The real numbers e and
π are transcendental over Q.

Remark 6.5. Since the set of polynomials with integer (or rational) coeffi-
cients is countable and each polynomial has a finite number of roots, the set
of real numbers that are algebraic over Q is countable. Therefore, there are
uncountably many real numbers that are transcendental over Q.

7 Finite Field
We have dealt a lot with the finite fields Zp . Now we will look at field
extensions of these fields. We know that any finite field F has characteristic
p, for some prime p. And then F is an extension of Zp . Suppose F contains
q elements. Then q must be a power of p. That is what we will prove now.

Theorem 7.1. Let F be a finite field having q elements and characteristic

p. Then q = pn , for some positive integer n.

Proof. Since char F = p, F has a prime subfield which is isomorphic to Zp .

We lose nothing if we assume that the prime subfield is Zp . We first show

10
that F is a vector space over Zp with finite dimension.

Recall that a set V is a vector space over a field K if

1. we can define a binary operation + on V such that (V,+) is an abelian
group,

2. we can define a ’scalar multiplication’ : K×V → V such that ∀ a, b ∈ K

and v, w ∈ V ,

• a.(v+w)=a.v+a.w
• (a+b).v = a.v+b.v
• (ab).v =a.(b.v)
• 1.v = v

Now, we know that (F ,+) is an abelian group. We also know that the mul-
tiplication in F will satisfy all the conditions that the scalar multiplication
should satisfy. Thus, F is a vector space over Zp . Since F is a finite field,
it has a finite dimension over Zp . Let dimxp F = n. Then, we can find
a1 , . . . , an ∈ F such that

F = Zp a1 + Zp a2 + . . . + Zp an .

We will show that F has pn elements.

Now, any element of F is of the form

b1 a1 + b2 a2 + . . . + bn an , where b1 , . . . , bn ∈ Zp .

Now, since ◦(Zp ) = p, b1 can be any one of its p elements.

Similarly, each of b2 , b3 , . . . , bn has p choices. And, corresponding

toneach chioces we get a distinct element of F . Thus, the number of
elements in F is p × p × . . . × p (n times)= pn .

The utility of this result is somethimg similar to that of Lagrange’s theorem.

Using this result we know that, for instance, no field of order 26 exists. But
does a field of order 25 exist? Does Theorem 7.1 answer this question? It only
says that a field of order 25 can exist. But it does not say that it does exist.
The following exciting result was obtained by the American Mathematician
E.H.Moore in 1893.

11
Theorem 7.2. For any prime number p and n ∈ N, there exist a field with
pn elements. Moreover, any two finite fields having the same number of
elements are isomorphic.
Now, we can utilise our knowledge of finite fields to solve the following
exercises.
Exercise 7.3. Let F be a finite field with pn elements, where p a prime
n
number such that char(F ) = p. Prove that ap = a for all a ∈ F . Hence or
otherwise, show that
n
Y
xp − x = (x − ai ).
Proof. Let F be a finite field with pn elements, where p a prime number such
n
that char(F ) = p. Since (F \{0}, .) is a group of order pn − 1, ap −1 = 1
n
for all a ∈ {0}. Therefore, we have ap = a for all a ∈ F \0. Also we have
n n
0p = 0. Consequently, we conclude that ap = a for all a ∈ F .
n
Now, xp − x ∈ F [x] can have at the most pn roots in F (by FTA).
Also, each of the pn elements
Q of F is a root. Thus, these are all the roots of
n n
xp − x. Hence, xp − = (x − ai ). This proves the result.
ai ∈F

Exercise 7.4. Let F be a finite field with pn elements. Define φ : F → F

such that φ(a) = ap for all a ∈ F. Prove that φ is an automorphism of F of
n
order n i.e., φ is an isomorphism such that φp = I,(the identity map on F )
and φr 6= I for r < n. (Such φ is called the Frobenius automorphism of F ).
Proof. Let F be a finite field with pn elements. Define φ : F → F such that
φ(a) = ap for all a ∈ F. Then , we have φ(a + b)p = (a + b)p = ap + bp

= φ(a) + φ(b).
φ(ab) = (ab)p = ap bp = φ(a)φ(b).
φ is 1 − 1, by E 10(c) of Unit 12.
Hence, Im φ has the same number of elements as the domain of φ, i.e., F .
Further, Im φ ⊆ F . Therefore, we have Im φ = F , i.e., φ is onto.
Hence, φ is an automorphism.
Now, φn (a) = [φ(a)]n = (ap )n = (ap )n = a for all a ∈ F .
Therefore, φn = I the identity map on F .
r
Also, for r < n, φr (a) = ap .
r
Now, we can’t have ap = a for all a ∈ F , because this would mean that the
r
polynomial xp −x ∈ F [x] has more than pr roots. This gives a contradiction.
Thus, φr (a) 6= a for some a ∈ F. Therefore, φr 6= I if r < n.
Hence, o(φ) = n. This completes the proof.

12
Exercise 7.5. Let F be a field such that a ∈ F iff a is a root of x27 −x ∈ F [x].
Find char(F ). Also, check whether Z2 ⊆ F ?

Proof. Let F be a field. Then, a ∈ F iff a27 = a. This can be written as

3 3
a3 = a for all a ∈ F. That is, a3 − a = 0 for all a ∈ F. Hence, a is a
27
root of x − x ∈ F [x]. Application of Exercise 7.3 gives that char(F ) = 3.
Further, we notice that char(Z2 ) = 2. Consequently, Z2 * F. This proves
the result.

Definition 7.6. The prime subfield of a field F is the subfield of F gen-

erated by the multiplicative identity 1F of F . It is (isomorphic to) either
Q( if ch(F ) = 0) or p( if ch(F ) = p).

Remark 7.7. We shall usually denote the identity 1F of a field F simply

by 1. Then in a field of characteristic p, one has p.1 = 0, frequently written
simply p = 0 (for example, 2 = 0 in a field of characteristic 2). It should be
kept in mind, however, that this is a shorthand statement ? the element ”p”
is really p.1F and is not a distinct element in F . This notation is useful in
light of the second statement in 2.5.

Example 7.8. 1. The prime subfield of both Q and R is Q.

2. The prime subfield of the field Fp (x) is isomoiphic to Fp , given by the

constant polynomials

Definition 7.9. If K is a field containing the subfield F , then K is said to

be an extension field (or simply an extension) of F , denoted K/F

In particular, every field F is an extension of its prime subfield. The

field F is sometimes called the base field of the extension.

The notation K/F for a field extension is a shorthand for K over F

and is not the quotient of K by F .

If K/F is any extension of fields, then the multiplication defined in

K makes K into a vector space over F . In particular every field F can be
considered as a vector space over its prime field.

Definition 7.10. The degree (or relative degree or index) of a field extension
K of F , denoted [K : F ], is the dimension ofK as a vector space over F (i.e.,
[K : F ] = dimF K). The extension is said to be finite if [K : F ] is finite and
is said to be infinite otherwise.

13
 Given any field F and any polynomial p(x) ∈ F [x] one can ask a similar
question: does there exist an extension K of F containing a solution of the
equation p(x) = 0 (i.e., containing a root of p(x))? Note that we may assume
here that the polynomial p(x) is irreducible in F [x] since a root of any factor
of p(x) is certainly a root of p(x) itself. The answer is yes and follows almost
immediately from our work on the polynomial ring F [x]. We first recall the
following useful result on homomorphisms of fields which follows from the
fact that the only ideals of a field F are 0 and F .

Proposition 7.11. Let ϕ : F → F 0 be a homomorphism of fields. Then ϕ

is either identically 0 or is injective, so that the image of e is either 0 or
isomorphic to F .

Theorem 7.12. Let F be a field and let p(x) ∈ F [x] be an irreducible poly-
nomial. Then there exists a field K containing an isomorphic copy of F in
which p(x) has a root. Identifying F with this isomorphic copy shows that
there exists an extension of F in which p(x) has a root.

Proof. Consider the quotient

K = F [x]/(p(x))

of the polynomial ring F [x] by the ideal generated by p(x). Since by

assumption p(x) is an irreducible polynomial in the P.I.D. F [x], the ideal
(p()x)) is a maximal ideal. Hence K is actually a field. The canonical
projection π of F [x] to the quotient F [x]/(p(x)) restricted to F ⊂ F [x]
gives a homomorphism ϕ = πF : F → K which is not identically 0 since it
maps the identity 1 of F to the identity 1 of K. Hence by the proposition
above, ϕ(F ) ∼
= F is an isomorphic copy of F contained in K. We identify F
with its isomorphic image in K and view F as a subfield of K. If x̄ = π(x)
denotes the image of x in the quotient K, then

p(x̄) = p(x) (since πisahomomorphism)

= p(x)(modp(x)) in F [x]/p(x))

=0 in F [x]/p(x))
so that K does indeed contain a root of the polynomial p(x). Then K
is an extension of F in which the polynomial p(x) has a root.

We shall use this result later to construct extensions of F containing

all the roots of p(x) (this is the notion of a splitting field and one of the

14
central objects of interest in Galois theory).
To understand the field K = F [x]/(p()x)) constructed above more fully, it
is useful to have a simple representation for the elements of this field. Since
F is a subfield of K, we might in particular ask for a basis for K as a vector
space over F .

Theorem 7.13. Let p(x) ∈ F [x] be an irreducible polynomial of degree n

over the field F and let K be the field F [x]/(p(x)). Let θ = xmod(p(x)) ∈ K.
Then the elements 1, θ, θ2 · · · , θn−1 are a basis for K as a vector space over
F , so the degree of the extension is n, i.e., [K : F ] = n. Hence

K = {a0 + a1 θ + a2 θ2 + · + an−1 θn−1 | a0 , a1 , ·, an−1 ∈ F }

consists of all polynomials of degree < n in θ.

Proof. Let a(x) ∈ F [x] be any polynomial with coefficients in F . Since F [x]
is a Euclidean Domain, we may divide a(x) by p(x) :
a(x) = q(x)p(x) + r(x) q(x), r(x) ∈ F [x] with degr(x) < n.
Since q(x)p(x) lies in the ideal (p(x)), it follows that a(x) ≡ r(x)modp(x)),
which shows that every residue class in F [x]/(p(x)) is represented by a
polynomial of degree less than n. Hence the images 1, θ, θ2 · · · , θn−1 of
1, x, x2 , · · · , xn−1 in the quotient span the quotient as a vector space over
F . It remains to see that these elements are linearly independent, so form
a basis for the quotient over F . If the elements 1, θ, θ2 · · · , θn−1 were not
linearly independent in K, then there would be a linear combination

b0 + b1 θ + b2 θ2 + · + bn−1 θn−1 = 0

in K, with b0 , b1 , ·bn−1 ∈ F , not all 0. This is equivalent to

b0 + b1 x + b2 x2 + · + bn−1 xn−1 ≡ 0mod(p(x))

. i.e.,

p(x) divides by b0 + b1 x + b2 x2 + · + bn−1 xn−1

in F [x]. But this is impossible, since p(x) is of degree n and the degree of
the nonzero polynomial on the right is < n. This proves that 1, θ, θ2 · · · , θn−1
are a basis for K over F , so that [K : F ] = n by definition. The last
statement of the theorem is clear.

15
 This theorem provides an easy description of the elements of the field
F [x]/(p(x)) as polynomials of degree <n in θ where θ is an element (in
K) with pθ = 0. It remains only to see how to add and multiply elements
written in this form. The addition in the quotient F [x]/(p(x)) is just
usual addition of polynomials. The multiplication of polynomials a(x)
and b(x) in the quotient F [x]/(p(x)) is performed by finding the product
a(x)b(x) in F [x], then finding the representative of degree < n for the coset
a(x)b(x) + (p(x)) (as in the proof above) by dividing a(x)b(x) by p(x) and
finding the remainder.

This can also be done easily in terms of θ as follows: We may suppose

p(x) is monic (since its roots and the ideal it generates do not change by
multiplying by a constant), say p(x) = xn + pn−1 xn−1 + · + p1 x + p0 . Then
in K, since p(θ) = 0, we have

θn = −(pn−1 θn−1 + · · · + p1 θ + p0 )

i.e., θn is a linear combination of lower powers of θ. Multiplying both sides

by θ and replacing the θn on the right hand side by these lower powers
again, we see that also θn+1 is a polynomial of degree <n in θ. Similarly,
any positive power of θ can be written as a polynomial of degree <n in θ,
hence any polynomial in θ can be written as a polynomial of degree < n in θ.
Multiplication in K is now easily performed: one simply writes the product
of two polynomials of degree <n in θ as another polynomial of degree n in θ
. We summarize this as:

Corollary 7.14. Let K be as in Theorem 5.12, and let a(θ), b(θ) ∈ K be

two polynomials of degree < n in θ. Then addition in K is defined simply by
usual polynomial addition and multiplication in K is defined by

a(θ)b(θ) = r(θ)

where r(x) is the remainder (of degree < n) obtained after dividing the poly-
nomial a(x)b(x) by p(x) in F [x].

16
8 Some important properties and results
We begin with the following:

Definition 8.1.

• Let K be a subfield of the field F and M any subset of F . Then the field
K(M ) is defined to be the intersection of all subfields of F containing
both K and M ; i.e. it is the smallest subfield of F containing both
K and M . It is called the extension field obtained by adjoining the
elements of M .

• For finite M = {α1 , ..., αn }, we write K(M ) = K(α1 , ..., αn ).

• If M = {α}, then L = K(α) is called a simple extension of K and α

is called a defining element of L over K.

The following type of extension is very important in the theory of fields

in general.

Definition 8.2.

• Let K be a subfield of F and α ∈ F . If a satisfies a nontrivial polyno-

mial equation with coefficients in K, i.e.,

an αn + an−1 αn−1 + · · · + a1 α1 + a0 = 0

for some ai ∈ K not all zero, then a is algebraic over K.

• An extension L of K is called algebraic over K (or an algebraic exten-

sion of K) if every element in L is algebraic over K.

Example 8.3.
√
• The element 3 3 ∈ R is algebraic over Q, since it is a root of the
polynomial x3 − 3 ∈ Q[x].

• The element i ∈ C is algebraic over R, since it is a root of x2 +1 ∈ R[x].

• The element π ∈ R is not algebraic over Q. An element which is not

algebraic over a field F is said to be transcendental over F.

17
Given a ∈ F which is algebraic over some subfield K of F , it can be checked
(exercise!) that the set J = {f ∈ K[x] : f (α) = 0} is an ideal of F [x] and
J 6= (0). By Theorem 3.4, it follows that there exists a uniquely determined
monic polynomial g ∈ K[x] which generates J, i.e. J = (g).
Definition 8.4.
If a is algebraic over K, then the uniquely determined monic polynomial
g ∈ K[x] generating the ideal J = {f ∈ K[x] : f (α) = 0} is called the
minimal polynomial of a over K. We refer to the degree of g as the degree
of a over K.

The key properties of the minimal polynomial are summarised in the

next theorem. The third property is the one most useful in practice.
Theorem 8.5.
Let a ∈ F be algebraic over a subfield K of F , and let g be the minimal
polynomial of α. Then
(i) g is irreducible in K[x];

(ii) For f ∈ K[x], we have f (α) = 0 if and only if g divides f ;

(iii) g is the monic polynomial of Least degree having as a root.

Proof. (i) Since g has the root a, it has positive degree. Suppose g = h1 h2
in K[x] with 1 ≤ deg(hi ) < deg(g) (i = 1, 2). This implies 0 = g(α) =
h1 (α)h2 (α), and so one of h1 or h2 must lie in J and hence is divisible by g,
a contradiction.
(ii) Immediate from the definition of g. (iii) Any monic polynomial in K[x]
having α as a root must be a multiple of g by (ii), and so is either equal to
g or has larger degree than g. 
Example 8.6.
√
• The element 3 3 ∈ R is algebraic over Q since it is a root of x3 − 3 ∈
Q[x].
√ Since x3 − 3 is irreducible
√ over Q, it is the minimal polynomial
3
of 3 over Q, and hence 3 3 has degree 3 over Q.
√
• The element i = −1 ∈ C is algebraic over the subfield R of C, since
it is a root of the polynomial x2 + 1 ∈ R[x]. Since x2 + 1 is irreducible
over R, it is the minimal polynomial of i over R, and hence i has degree
2 over R.

18
9 Field extensions as vector spaces
Let L be an extension field of K. An important observation is that L may
be viewed as a vector space over K. The elements of L are the "vectors"
and the elements of K are the "scalars".
We briefly recall the main properties of a vector space.

Definition 9.1.
A vector space V over F is a non-empty set of objects (called vectors) upon
which two operations are defined
• addition: there is some rule which produces, from any two objects in
V , another object in V (denote this operation by +)

• scalar multiplication: there is some rule which produces, from an ele-

ment of F (a scalar) and an object in V , another object in V
and these objects and operations obey the Vector Space Axioms:
1 x + y = y + x for all x, y ∈ V

2 (x + y) + z = x + (y + z) for all x, yz ∈ V

3 there exists an object 0 ∈ V such that x + 0 = x for all x ∈ V

4 for every x ∈ V there exists an object −x such that x + (−x) = 0

5 λ(x + y) = λx + λy for all x, y ∈ V and all scalars λ ∈ F

6 (λ + µ)x = λx + µx for all x ∈ V and all scalars λ, µ ∈ F

7 (λµ)x = λ(µx) for all x ∈ V and all scalars λ, µ ∈ F

8 1x = x for all x ∈ V
Definition 9.2.

• A basis of a vector space V over F is defined as a subset {v1 , ..., vn }

of vectors in V that are linearly independent and span V . If v1 , ..., vn
is a list of vectors in V , then these vectors form a basis if and only if
every v ∈ V can be uniquely written as

v1 = a1 v1 + · · · + an vn

where a1 , ..., an are elements of the base field F .

19
 • A vector space will have many different bases, but there are always the
same number of basis vectors in each. The number of basis vectors in
any basis is called the dimension of V over F .

• Suppose V has dimension n over F . Then any sequence of more than

n vectors in V is linearly dependent.

To see that the vector space axioms hold for a field L over a subfield K,
note that the elements of L form an abelian group under addition, and that
any "vector" α ∈ L may be multiplied by an r ∈ K (a "scalar") to get
rα ∈ L (this is just multiplication in L). Finally, the laws for multiplication
by scalars hold since, for r, s ∈ L and β ∈ K we have r(α+β) = rα+rβ, (r +
s)α − rα + sα(rs)α = r(sα) and 1α = α.

Example 9.3.

Take L = C and let K be its subfield R. Then we can easily check that C is a
vector space over R. Since we know from school that C = {a + bi : a, b ∈ R},
it is clear that a basis is given by {1, 2}.

Definition 9.4.

Let L be an extension field of K. If L is finite-dimensional as a vector space

over K,then L is said to be a finite extension of K. The dimension of the
vector space L over K is called the degree of L over K and written [L : K].

Example 9.5.

From above, C is a finite extension of R. of degree 2.

Theorem 9.6.

If L is a finite extension of K and M is a finite extension of L, then M is a

finite extension of K with

[M : K] = [M : L][L : K].

Proof. Let [M : L] = m, [L : K] = n; let {α1 , ..., αn , } be a basis of M

over. L and let {β1 , ..., βn } be a basis of L over K. We shall use them to
form a basis of M over K of appropriate cardinality.

20
 Every α ∈ M can be expressed as a linear combination α = γ1 α1 + ... +
γm αm for some γ1 , ..., γm ∈ L. Writing each γi as a linear combination of
the βj is we get
m
X m X
X n m X
X n
α= γi αi = ( rij βj )αi = rij βj αi
i=1 i=1 j=1 i=1 j=1

with coefficients rij ∈ K. We claim that the mn elements βj αi form a basis

of M over K. Clearly they span M ; it suffices to show that they are linearly
independent over K.

Suppose we have
m X
X n
sij βj αi = 0
i=1 j=1

where the coefficients sij ∈ K. Then

m X
X n
( sij βj )αi = 0
i=1 j=1

and since the αi are linearly independent over L we must have

n
X
sij βj = 0
j=1

for 1 ≤ i ≤ m. Now, since the βi are linearly independent over K, it follows

that all the sij are 0, as required. 

Corollary 9.7. If L is a finite extension of K and M is any subfield field

of L containing K, K ⊆ M ⊆ L, then prove that [M : K] divides [L : K].

Theorem 9.8. Every finite extension of K is algebraic over K.

Proof. Let L be a finite extension of K and let [L : K] = m. For a ∈ L,

the m + 1 elements 1, α, ..., αm must be linearly dependent over K, i.e. must
satisfy a0 + a1 α + · · · + am αm for some ai ∈ K (not all zero). Thus a is
algebraic over K. hfill 

Remark 9.9.

21
The converse of Theorem 5.7 is not true, however. See the Exercise sheet for
an example of an algebraic extension of Q which is not a finite extension.

We now relate our new vector space viewpoint to the residue class rings
considered previously.

Theorem 9.10.

Let F be an extension field of K and a ∈ F be algebraic of degree n over K

and let g be the minimal polynomial of α over K. Then

(i) K(α) is isomorphic to K[x]/(g);

(ii) [K(α) : K] = n and {1, α, ..., αn−1 } is a basis of K(α) over K;

(iii) Every β ∈ K(α) is algebraic over K and its degree over K is a divisor
of n.

Proof. (i) Consider the "evaluation at α" mapping τ : K[x] → K(α),

defined by
τ (f ) = f (α) for f ∈ K[x]
; it is easily shown that this is a homomorphism. Then

kerτ = {f ∈ K[x] : f (α) = 0} = (g)

by the definition of the minimal polynomial. Let S be the image of τ , i.e.

the set of polynomial expressions in a with coefficients in K. By the First
Isomorphism Theorem for rings we have S ∼ = K[x]/(g). Since g is irreducible,
by Theorem 3.8, K[x]/(g) is a field and so S is a field. Since K ⊆ S ⊆ K(α)
and a ∈ S, we have S = K(α) by the definition of K(α),and(i)follows.
(ii) Spanning set: Since S = K(α), any β ∈ K(α) can be written in the
form β = f (α) for some polynomial f ∈ K[x]. By the division algorithm,
f = qg + r for some q, r ∈ K[x] and deg(r) < deg(g) = n. Then

β = f (α) = q(α)g(α) + r(α) = r(α),

and so β is a linear combination of 1, α, ..., αn−1 with coefficients in K.

L.I.: if a0 + a1 aα + · · · + an−1 αn−l = 0 for some a0 , ..., an−1 ∈ K, then the
polynomial h = a0 + ax1 + · · · + an−1 xn−1 ∈ K[x] has α as a root, and is
thus a multiple of its minimal polynomial g. Since deg(h) < n = deg(g),
this is possible only if h = 0, i.e. a0 = · · · = an−1 = 0. Thus the elements
1, α, ..., αn−1 are linearly independent over K.

22
(iii) K(α) is a finite extension of K by (ii), and so β ∈ K(α) is algebraic over
K by Theorem 5.7. Moreover, K(β) is a subfield of K(α). If d is the degree of
β over K,then n = [K(α) : K] = [K(α) : K(β)][K(β) : K = [K(α) : K(β)]d,
i.e. d divides n. 

Remark 9.11.

This theorem tells us that the elements of the simple extension K(α) of K
are polynomial expressions in α, and any β ∈ K(α) can be uniquely expressed
in the form β = ao + a1 α + · · · + an−1 αn−1 for some ai ∈ K.

Example 9.12.

Consider the simple extension R(i) of R. We saw earlier that i has minimal
polynomial x2 + 1 over R.

So R(i) ∼
= R[x]/(x2 + 1), and {1, i} is a basis for R(i) over R(i). So

R(i) = {a + bi : a, b ∈ R} = C

Example 9.13.
√ √
Consider the simple extension Q( 3 3) of Q. We saw earlier that 3 3 has
minimal polynomial
√ x2 − 3 over Q. √ √ √
∼
So Q 3 = [x]/(x3 − 3), and {1, 3 3, ( 3 3)2 } is a basis for Q( 3 3) over
3

Q. So √ √ √
Q( 3 = {a + b 3 + c( 3)2 : a, b, c ∈ Q}.
3 3 3

Note that we have been assuming that both K and α are embedded in
some larger field F . Next, we will consider constructing a simple algebraic
extension without reference to a previously given larger field, i.e. "from the
ground up".
The next result, due to Kronecker, is one of the most fundamental results
in the theory of fields: it says that, given any non-constant polynomial over
any fie1d, there exists an extension field in which the polynomial has a root.

Theorem 9.14.

Let f e ∈ K[x] be irreducible over the field K. Then there exists a simple
algebraic extension of K with a root of f as a defining element.

Proof.

23
 • Consider the residue class ring L = K[x]/(f ), which is a field since f
is irreducible. Its elements are the residue classes [h] = h + (f ), with
h ∈ K[x].
• For any α ∈ K, think of a as a constant polynomial in K[x] and form
the residue class [a]. The mapping a 7→ [α] gives an isomorphism from
K onto a subfield K 0 of L (exercise: check!), so K 0 may be identified
with K. Thus we can view L as an extension of K.
• For every h = a0 + a1 x + · · · + am xm ∈ K[x], we have
[h] = [a0 + a1 x + · · · + am xm ]
= [a1 ] + [a1 ][x] + · · · + [am ][x]m
= a0 + a1 [x] + · · · + am [x]m
(making the identification [ai ] = ai ). So, every element of L can be
written as a polynomial in [x] with coefficients in K. Since any field
containing K and [x] must contain these expressions K is a simple
extension of K’ obtained by adjoining [x].
• If f = b0 + b1 x + · · · + bn xn , then
f ([x]) = b0 + b1 [x] + · · · + bn [x]n = [f ] = [0],
i.e. [x] is a root of f and L is a simple algebraic extension of K.

Example 9.15.
Consider the prime field F3 and the polynomial x2 +x+2 ∈ F3 [x], irreducible
over F3 . Take θ to be a "root" of f , in the sense that θ is the residue class
[x] = n + (f ) ∈ L = F3 [x]/(f ). Explicitly, we have:
f (θ) = f ([x]) = f (x + (f ))
= (x + (f ))2 + (x + (f )) + (2 + (f ))
= x2 + x + 2 + f (f )
= f + (f )
= 0 + (f )
= [0].
The other root of f in ,L is 2θ + 2, since f (2θ + 2) = θ2 + θ + 2 = 0. By
Theorem 5.9, the simple algebraic extension L = F3 (θ) consists of the nine
elements 0, 1, 2, θ, θ + 1, θ + 2, 2θ, 2θ + 1, 2θ + 2.

24
Example 9.16.
Consider the polynomial f = x2 + x + 1 ∈ F2 [x], irreducible over F2 . Let θ
be the root [x] = x + (f ) of f ; then the simple algebraic extension L = F2 (θ)
consists of the four elements 0, 1, θ, θ + 1. (The other root is θ + 1). The
tables for addition and multiplication are precisely those of Example 3.9,
now appropriately relabelled. We give the addition table:
+ 0 1 θ θ+1
0 0 1 θ θ+1
1 1 0 θ+1 θ
θ θ θ+1 0 1
θ+1 θ+1 θ 1 0
Note that, in the above examples, adjoining either of two roots of f would
yield the same extension field.

Theorem 9.17.
Let F be an extension field of the field K and α, β ∈ F be two roots of a
polynomial f ∈ K[x] that is irreducible over K. Then K(α) and K(β) are
isomorphic under an isomorphism mapping a to θ and keeping the elements
of K fixed.
Proof, By Theorem 5.9 both are isomorphic to the field K[x]/(f ) since
the irreducible f is the minimal polynomial of both α and β. 

Given a polynomial, we now want an extension field which contains all

its roots.
Definition 9.18.
Let f ∈ K[x] be a polynomial of positive degree and F an extension field of
K. Then we say that f splits in F if f can be written as a product of linear
factors in F [x], i.e. if there exist elements α1 , ..., αn ∈ F such that
f = a(x − α1 ) · · · (x − αn )
where a is the leading coefficient of f . The field F is called a splitting field
of f over If if it splits in F and if F = K(α1 , ..., an ).

So, a splitting field F of a polynomial f over K is an extension field

containing all the roots of f , and is "smallest possible" in the sense that
no subfield of F contains all roots of f . The following result answers the
questions: can we always find a splitting field, and how many are there?

25
Theorem 9.19. (Existence and uniqueness of splitting field)

(i) It K is a field and f any polynomial of positive degree in K[x], then

there exists a splitting field of f over K.

(ii) Any two splitting fields of f over K are isomorphic under an isomor-
phism which keeps the elements of K fixed and maps roots of f into
each other.

So, we may therefore talk of the splitting field of f over K. It is obtained

by adjoining to K finitely many elements algebraic over K, and so we can
show (exercise!) that it is a finite extension of K.

Example 9.20.

Find the splitting field of the polynomial f = x2 + ∈ Q[x] over Q.

√ √
The polynomial f splits in C, where it factors as (x − i 2)(x + i 2).
However, C itself is not the splitting field for f . It turns out to be sufficient
√
to adjoin just one of the complex roots of f to Q. The field K = Q(i 2)
contains both of the roots of f , and no smaller subfield has this property, so
K is the splitting field for F .

10 Problems
1. Q + iQ is a field of C.
√
2. Z + 2Z is not a subfield of R.

3. Which of the following rings are field extensions of Q?

(i) Q[x]/ < x2 + 7 >

(ii) Q[x]/ < x3 + 2 >
(iii) Q

4. Show that the smallest subfield of any field is a prime subfield.

5. Let F be a field which has no proper subfields. Show that F is isomor-

phic to a prime field.

6. Obtain the prime subfields of (R, Z5 ).

26
 7. Show that given any field, if we know its characteristic then we can
obtain its prime subfields, and vice versa.
n
8. Let F be a finite field with pn elements. Show that ap = a ∀ a ∈ F .
p n Q
And hence, show that x − x = ai ∈F (x − ai ).

9. Define a field extension, finite extension and give some examples.

10. Determine the degree and the basis of the field of Complex numbers
over the filed of reals. Also, prove that [C : R] = 2.
√ √
11. Show that Q( 2, 3) is finite field extension over Q, where Q is the
field of all rational numbers.
√
12. Let Q be √ the field of rational numbers. Next, let K = Q( 2) and
L = Q( 3). Show that [L : K] = 2 and []K : Q] = 2.

27