Solution Outlines for Chapter 15

# 11: Prove that the intersection of any collection of subfields of a field F is a

subfield of F .

We know that every field contains 1 and 0, so the intersection of subfields is non-empty.
Since a, b are in the intersection means that a, b is in each subfield, ab and a − b are in each
subfield. Hence the intersection is an additive subgroup and is closed under multiplication.
The only other thing to check that is not inherited is that every element in the intersection
has a multiplicative inverse in the intersection but this is clear because the inverse must be
in each subfield.

# 12: Let Z3 [i] = {a + bi|a, b ∈ Z3 }. Show that the field Z3 [i] is ring-isomorphic to
the field Z3 [x]/ < x2 + 1 >.

Define φ(a + bi) = a + bi+ < x2 + 1 >. Then φ((a + bi) + (c + di)) = φ((a + c) + (b + d)i) =
(a+c)+(b+d)i+ < x2 +1 >= (a+bi)+(c+di)+ < x2 +1 >= (a+bi+ < x2 +1 >)+(c+di)+ <
x2 + 1 >= φ(a + bi) + φ(c + di). Farther, φ((a + bi)(c + di)) = φ((ac − bd) + (ad + bc)i) =
(ac − bd) + (ad + bc)i+ < x2 + 1 >= (a + bi)(c + di)+ < x2 + 1 >= ((a + bi)+ < x2 + 1 >
)((c + di)+ < x2 + 1 >. Hence, φ is a ring homomorphism.  
a b
# 15: Consider the mapping from M2 (Z) into Z given by 7→ a. Prove or
c d
disprove that this is a ring homomorphism.

The map is not a ring homomorphism.

  While
 addition
 is preserved,
 multiplication is
a b f g af + bh ∗
not. To see this observe: φ( = φ( ) = af + bh 6= af =
    c d h i ∗ ∗
a b f g
φ( )φ( ).
c d h i
  
a b
# 16: Let R = a, b, c ∈ Z . Prove or disprove that the mapping
  0 c
a b
7→ a is a ring homomorphism.
0 c
   
a b d f
This map is a ring homomorphism. Addition is preserved because φ( + )=
      0 c 0 g
a+d b+f a b d f
φ( ) = a + d = φ( ) + φ( ). Farther, the multiplication is
0 c+g    0 c   0 g    
a b f g af ∗ a b f g
preserved since φ( = φ( ) = af = φ( )φ( ).
0 d 0 i ∗ ∗ c d h i
# 17: Is the mapping from Z5 to Z30 given by x 7→ 6x a ring homomorphism? Note
that the image of the unity is the unity of the image but not the unity of
Z30 .
The mapping is a ring homomorphism. Addition is straightforward to show since φ(x + y) =
6(x + y) = 6x + 6y = φ(x) + φ(y). For multiplication, φ(xy) = 6xy and φ(x)φ(y) = 6x6y =
36xy = 6xy. Thus multiplication is preserved.

# 20: Recall that a ring element a is called an idempotent if a2 = a. Prove that

a ring homomorphism carries an idempotent to an idempotent.

Let a be an idempotent of a ring R. Then φ(a) = φ(a2 ) = (φ(a))2 . Hence, φ(a) is an

idempotent.

# 21: Determine all ring homomorphisms from Z6 to Z6 . Determine all ring

homomorphisms from Z20 to Z30 .

Part 1: We know that a ring homomorphism must be a group homomorphism as well.

Hence we know that the image of 1 has order 1, 2, 3 or 6. For the image to have order
1, 1 7→ 0 and this map clearly preserves multiplication. For the order to be 2, 1 7→ 3.
Now φ(xy) = 3xy and φ(x)φ(y) = 9xy = 3xy so this is indeed a ring homomorphism as
well. In the order three case, 1 maps to either 2 or 4. We observe that in the first instance
φ(xy) = 2xy 6= 4xy = φ(x)φ(y) so this is not a homomorphism. However, in the second
instance φ(xy) = 4xy and φ(x)φ(y) = 16xy = 4xy so it is a ring homomorphism. Finally,
for the image to have order 5, the element 1 maps to either 1 or 5. If it maps to 1, then
this is the identity map and is clearly a ring homomorphism. However, if it maps to 5,
φ(xy) = 5xy but φ(x)φ(y) = 25xy = xy so it is not a ring homomorphism. Hence the ring
homomorphisms are those determined by 1 7→ 0, 1 7→ 3, 1 7→ 4 and 1 7→ 1.

Part 2: We can use a similar logic for this part. Doing this shows that the homomorphisms
are those defined by 1 maps to 0, 6, 15 or 21.

# 30: Prove that the sum of the squares of three consecutive integers can not
be a square.

Let n be an integer. Then the sum of the squares of three consecutive integers can be rep-
resented as n2 + (n + 1)2 + (n + 2)2 . Now consider this expression modulus 3. Then it is
n2 + n2 + 2n + 1 + n2 + 4n + 4 = 3n2 + 6n + 5 = 2. For this expression to be a square then
there must exist an element, x, in Z3 such that x2 = 2. But 02 = 0, 12 = 1 and 22 = 1 so
there is no such solution.

# 37: Show that no integer of the form 111, 111, 111, . . . , 111 is prime.

Consider φ3 as defined in class. Then φ3 (111, 111, 111, . . . , 111) = φ3 (1 · 10k + 1 · 10k−1 +
 k k
φ3 (1)φ3 (10)i =
P P
. . . + 1 · 10 + 1) = 1 · · · 1 = 1 + 1 + 1 · · · + 1 but the 1’s come in threes
i=0 i=0
so this expression is 0 in Z3 . Hence 3 divides the integer so it can not be prime.

# 39: Suppose n is a positive integer written in the form n = ak 3k + ak−1 3k−1 +

· · · + a1 3 + a0 , where each of the ai ’s is 0, 1, or 2. Show that n is even if and
only if ak + ak−1 + · · · + a1 + a0 is even.

Consider φ2 as defined in class. Then φ2 (n) = φ2 (ak 3k + ak−1 3k−1 + · · · + a1 3 + a0 ) =

k k
φ2 (ai )φ2 (3)i =
P P
φ2 (ai ) · 1 = φ2 (a0 + a1 + a2 + . . . + ak ). Hence, n is even iff the sum of
i=1 i=1
its coefficients is even.

# 45: Is there a ring homomorphism from the reals to some ring whose kernel
is the integers?

No. The kernel of a ring homomorphism is an ideal but the integers are not an ideal of the
real numbers. For instance, π · 1 = π 6∈ Z.

# 50: Show that if m and n are distinct positive integers, then mZ is not ring-
isomorphic to nZ.

Let m and n be distinct positive integers. Because a ring isomorphism must take gener-
ators to generators, m would have to map to ±n. Consider the case of m 7→ n. Then
φ(nn) = φ(n + n + n + · · · + n) with n copies of n. This is equal to φ(n) + φ(n) + · · · + φ(n) =
m+m+· · ·+m = nm. But φ(nn) = φ(n)φ(n) = mm. Since n and m are distinct, nm 6= mm
so there can not be any such isomorphism.
    
a b a b
# 66: Let R = a, b ∈ Z , and let φ be the mapping that takes
b a b a
to a − b.
a. Show that φ is a homomorphism.
     
a b c d a+c b+d
Addition is preserved since φ( + ) = φ( ) = (a +
b a d c 
 b + d a + c
a b c d
c) − (b + d) = (a − b) + (c − d) = φ( ) + φ( ). We can also see that
 b a
  d c  
a b c d ac + bd ad + bc
multiplication is preserved since φ( ) = φ( )) =
b a d c bc + ad bd + ac
(ac + bd) −
 (ad + bc) = a(c − d) + b(d − c) = a(c − d) − b(c − d) = (a − b)(c − d) =
a b c d
φ( )φ( ).
b a d c

b. Determine the kernel of φ.

 The kernel is the set of matrixes
 such that a − b = 0, or a = b. Hence it is the set of
a a
matrices of the form where a ∈ Z.
a a

c. Show that R/Ker φ is isomorphic to Z.

The image of φ defined as above is Z. We know this since φ(R) ⊆ Z, and the matrix
with a = a and b = 0 maps to a for any a ∈ Z.

d. Is Ker φ a prime ideal?

Since Z is an integral domain, the kernel is a prime ideal.

e. Is Ker φ a maximal ideal?

Since Z is not a field, the kernel is not maximal.