Topics in Algebra solution

Sung Jong Lee, lovekrand.github.io

November 6, 2020

Problems in the Section 2.7.

1. In the following, verify if the mappings defined are homomorphisms and in those cases
in which they are homomorphisms, determine the kernel.
a) G is a group of nonzero real numbers under multiplication, G = G, φ(x) = x2 all x ∈ G.

Solution. φ is a homomorphism. Note that φ(xy) = (xy)2 = x2 y 2 = φ(x)φ(y) in G. Let K

be the kernel of φ. x2 = 1 implies x = −1 or x = 1. Hence, K = {−1, 1} ' Z2 .

b) G, G as in a), φ(x) = 2x

Solution. φ is not a homomorphism as φ(2 · 1) = 22 = 4 6= 8 = 22 · 21 = φ(2)φ(1).

c) G is the group of real numbers under addition, G = G, φ(x) = x + 1 all x ∈ G.

Solution. φ is not a homomorphism as φ(2) = 3 6= 4 = 2 + 2 = φ(1) + φ(1).

d) G, G as in c), φ(x) = 13x for x ∈ G.

Solution. φ is a homomorphism. Note that φ(x + y) = 13(x + y) = 13x + 13y = φ(x) + φ(y)
in G. Let K be the kernel of φ. Since 13x = 0 ⇐⇒ x = 0, K = {0}.

e) G be any abelian group, G = G, φ(x) = x5 all x ∈ G.

Solution. φ is a homomorphism. Note that φ(xy) = (xy)5 = x5 y 5 = φ(x)φ(y) in G. Let

K be the kernel. Sicen x5 = e ⇐⇒ x is of order 5 or x = e, K is the collection of all
elements of G of order 5.

2. Let G be any group, g a fixed elements in G. Define φ : G → G by φ(x) = gxg −1 . Prove

that φ is an homomorphism of G onto G.

Proof. Note that φ(xy) = g(xy)g −1 = (gxg −1 )(gyg −1 ) = φ(x)φ(y) for all x, y ∈ G. Hence,
φ is a homomorphism. Suppose φ(x) = e. Equivalently, gxg −1 = e =⇒ x = e, so that
the kernel K of φ is (e). Thus, φ is an isomorphism(injection). Further, for all x ∈ G,
φ(g −1 xg) = x, implying φ is onto. Therefore, φ is an onto isomorphism(automorphism).

1
3. Let G be any finite abelian group of order o(G) and suppose the integer n is relatively
prime to o(G). Prove that every g ∈ G can be written as g = xn with x ∈ G.

Proof. Consider a mapping φ : G → G defined as φ(x) = xn . Note that φ(xy) = (xy)n =

xn y n = φ(x)φ(y) so that φ is a homomorphism. Suppose xn = e. Then since gcd(n, o(G)) =
1, nλ + o(G)µ = 1 for some λ, µ ∈ Z, x = xnλ+o(G)µ = e, implying the kernel K of φ is (e),
so that φ is an isomorphism(injection). Now by Pigeonhole principle, as φ is an injection
from G to G, it is also onto. Therefore, φ is a bijection. This finishes the proof.

4. a) Given any group G and a subset U , let Û be the smallest subgroup of G which
contains U . Prove there is such a subgroup Û in G.
\
Proof. Let Û = W , intersection of all subgroups of G containing U . Clearly, Û
U ⊂W ≤G
is a\subgroup of G containing U . Let W 0 be another subgroup of G containing U . Then
W ⊂ W 0 , implying Û is the smallest subgroup of G containing U .
U ⊂W ≤G

b) If gug −1 ∈ U for all g ∈ G, u ∈ U , prove that Û is a normal subgroup of G.

Proof. Note that every elements in Û can be represented as in the form of finite products
of U , raised to integer exponents. That is,

u ∈ Û ⇐⇒ u = uk11 uk22 · · · uknn , ui ∈ U, ki ∈ Z, i = 1, 2, · · · , n,

Let u ∈ Û . Adopting the representation of u introduced above,

gug −1 = guk11 uk22 · · · uknn g −1

= (gu1 g −1 )k1 (gu2 g −1 )k2 · · · (gun g −1 )kn
= (u01 )k1 (u02 )k2 · · · (u0n )kn ∈ Û (∵ gui g −1 = u0i ∈ U )

for all g ∈ G. Hence, Û is normal in G.

5. Let U = {xyx−1 y −1 : x, y ∈ G}. In this case Û is usually written as G0 and is called the
commutator subgroup of G.
a) Prove that G0 is normal in G.

Proof. Note that for all g, x ∈ G, (gxg −1 )−1 = gx−1 g −1 and hence

g(xyx−1 y −1 )g −1 = gx · (g −1 g) · y · (g −1 g) · x−1 · (g −1 g) · y −1 g −1
= (gxg −1 )(gyg −1 )(gx−1 g −1 )(gy −1 g −1 ) ∈ U.

Now apply b) of the Problem 4. We see that G0 is normal in G.

2
b) Prove that G/G0 is abelian.

Proof. For any a, b ∈ G, abG0 = baa−1 b−1 abG0 = ba(a−1 b−1 ab)G0 = baG0 . Hence, G/G0 is
abelian.

c) If G/N is abelian, prove that G0 ⊂ N .

Proof. As G/N is abelian, abN = baN =⇒ a−1 b−1 ab ∈ N for all a, b ∈ G. Since
a−1 b−1 ab ∈ G0 , and a, b arbitrary, G0 ⊂ N .

d) Prove that if H is a subgroup of G and G0 ⊂ H, then H is normal in G.

Proof. Note that ghg −1 = ghg −1 h−1 h = (ghg −1 h−1 )h ∈ H for all g ∈ G, h ∈ H. Hence, H
is normal in G.

6. If N, M are normal subgroups of G, prove that N M/M ' N/N ∩ M .

Proof. N M is subgroup of G (Problem 3, Section 2.6). Now consider a mapping φ : N →

N M/M defined by φ(n) = nM . We show that φ is a homomorphism. Note that

φ(n1 n2 ) = n1 n2 M = n1 M n2 M = φ(n1 )φ(n2 )

so that φ is a homomorphism. Moreover, for any nM ∈ N M/M , φ(n) = nM so that φ

is onto. Now, we can apply isomorphism theorem. Note that for n ∈ N , φ(n) = M ⇐⇒
n ∈ M , so that the kernel of φ is N ∩ M . Hence,
N NM
' .
N ∩M M

7. Let V be the set of real numbers, and for a, b real, a 6= 0 let τab : V → V defined
by τab (x) = ax + b. Let G = {τab : a, b ∈ R, a 6= 0} and let N = {τ1b ∈ G}. Prove
that N is a normal subgroup of G and that G/N ' group of nonzero real numbers under
multiplication.

Proof. We know that N is normal in G, by applying Problem 23 of Section 2.6. Now, we

define a mapping φ : G → R − {0} by φ(τab ) = a. Note that φ is a homomorphism since
for any non-zero a, c ∈ R, b, d ∈ R,

φ(τab · τcd ) = φ(τac,ad+b ) = a · c = φ(τab )φ(τcd ).

Moreover, φ is cleary onto. Let K be the kernel of φ. Then the elements of k ∈ K

satisfies φ(k) = 1. But by the definition of N , we see that K is exactly N . Applying the
isomorphism theorem, we have G/N ' R − {0}.

3
8. Let G be the dihedral group defined as the set of all formal symbols xi y j , i = 0, 1,
j = 0, 1, · · · , n − 1, where x2 = e, y n = e, xy = y −1 x. Prove
a) The subgroup N = {e, y, y 2 , · · · , y n−1 } is normal in G.

Proof. It it easy to see that N = (y), and hence, a cyclic subgroup of G. Moreover,
[G : N ] = 2, implying N is normal in G.

b) That G/N ' W , where W = {−1, 1} is the group under the multiplication of the real
numbers.

Proof. From o(G/N ) = 2, we have G/N ' Z2 the only possible case. Hence, G/N '
{−1, 1} = W .

9. Prove that the center of a group is always a normal subgroup.

Proof. Note that any z ∈ Z(G) satisfies gzg −1 = z ∈ Z(G) for all g ∈ G. Hence Z(G) is
clearly normal in G.

10. Prove that a group of order 9 is abelian.

Proof. Let G be the group of order 9. Suppose there is an a ∈ G such that (a) = G, we
are done. If not, for all a ∈ G, (a) ( G. It is impossible that G to have no non-trivial
subgroup, otherwise G would be a group of prime order, contradicting that o(G) = 9. So,
we can find a subgroup (a) of order 3. Now, take b ∈ G − (a) and consider (b). The only
possible order of b is 3, hence (b) is a subgroup of order 3. Note that,

o(a) · o(b)
o((a) · (b)) =
o((a) ∩ (b))

and since (a) ∩ (b) = (e), o((a)(b)) = 9 and hence (a)(b) = G. It is now possible to write G
as G = {e, a, a2 , b, b2 , ab, ab2 , a2 b, a2 b2 }. Now we investigate if which of the elements stated
initially, equals ba ∈ G. Observe that

ba 6= e, a, a2 , b, b2 trivially,
ba = a2 b =⇒ aba = b, (ba)3 = e =⇒ bababa = bab(aba) = bab2 =⇒ ba = b ⊥,
ba = ab2 =⇒ bab = a, (ba)3 = e =⇒ bababa = (bab)aba = a2 ba =⇒ ba = a ⊥,
ba = a2 b2 =⇒ (ab)2 = e ⊥,

hence the only possibility is ba = ab. This shows that G is abelian.

11. If G is a non-abelian group of order 6, prove that G ' S3 .

4
Proof. Since G is of even order, there exists an element a ∈ G such that a2 = e. Obviously,
there is no element of order 6, otherwise G would be a cyclic group. We claim that there
is an element b ∈ G of order 3. If not, every non-identity element in G would be order of
2, so that G is abelian, contradicting that G is non-abelian. Thus, there must an element
b of order 3. Since (a) ∩ (b) = (e), and from the equation
o(a) · o(b)
o((a) · (b)) = ,
o((a) ∩ (b))
we have that o((a)(b)) = 6 so that (a)(b) = G. Hence, G = {e, a, b, b2 , ab, ab2 }. We
investigate if which of in G equals b−1 a. Clearly, b−1 a 6= e, a, b, b2 . Suppose b−1 a = ab2 .
Then ab = (bb−1 ) · ab = b(ab2 )b = ba implying G is abelian, hence a contradiction. The
only remaining possibility is b−1 a = ab. But this implies that G = {e, a, b, b2 , ab, ab2 }. This
group G with an operation property ab = b−1 a is exactly isomorphic to the symmetric group
S3 . Therefore, G ' S3 .

12. If G is abelian and if N is any subgroup of G, prove that G/N is abelian.

Proof. Since G is abelian, N is normal in G. Moreover, for any a, b ∈ G, (ab)N = (ba)N .

Thus, G/N is abelian.

13. Let G be the dihedral group defined in Problem 8. Find the center of G.

Proof. We consider the case of n > 2, otherwise the center of G is G itself trivially. By
the result of the Problem 17 in Section 2.6, if the order of n is odd, Z(G) = (e), and if the
order of n is even, {e, y n/2 } ⊂ Z(G). We find if any other elements xi y j of G is in Z(G),
for the order of n is even. In general, elements of the form xy k cannot be in the center as
if

(y −1 x)xy k (xy) = y k−1 y −1 x = y k−2 x,

but k ≡ k − 2 (mod n) will not hold if n > 2. Thus, we consider only the elements of the
form y k . Also note that

(xy)y k (y −1 x) = xy k x,

so that if y k = xy k x, y k x = xy k = y −k x ⇐⇒ k = 0, n/2. Hence, the only possible

elements of the form y k are e, y n/2 . This shows that the center of G is {e, y n/2 } exactly,
whenever n > 2 and n is of even order. Summarising,

G,
 o(G) = 2, 4
Z(G) = (e), n > 2, n ≡ 1 (mod 2)

 n/2
{e, y }, n > 2, n ≡ 0 (mod 2).

5
14. Let G be as in Problem 13. Find G0 , the commutator subgroup of G.

Proof. Note that y 2k = y k xy −k x so that y 2k ∈ G0 . Equivalently, (y 2 ) ⊂ G0 . Note that

(y 2 ) = N if n is odd and (y 2 ) = {e, y 2 , · · · , y n−2 } if n is even. Moreover, (y 2 ) is normal
in G. Hence, o(G/(y 2 )) = 2 if n is odd and o(G/(y 2 )) = 4 if n is even. Thus, G/(y 2 ) is
abelian. Applying the result of Problem 5 c), G0 ⊂ (y 2 ). Therefore, G0 = (y 2 ).

15. Let G be the group of non-zero complex numbers under multiplication and let N be
the set of complex numbers of absolute value 1 (that is, a + bi ∈ N if a2 + b2 = 1). Show
that G/N is isomorphic to the group of all positive real numbers under multiplication.

Proof. Let us define a mapping φ : G → R − {0} by φ(z) = |z|. Clearly, φ is a homomor-

phism since φ(zw) = |zw| = |z| |w| = φ(z)φ(w). Now for any k > 0 ∈ R, φ(k + 0i) = k
so that φ is onto. We now investigate the kernel of φ. Obviously, it is set of all nonzero
complex numbers of absolute value 1, that is, exactly, N . Now by isomorphism theorem,
G/N ' R − {0}.

16. Let G be the group of all nonzero complex numbers under

 multiplication and let G
a −b
be the group of all real 2 × 2 matrices of the form , where not both a and b
b a
are 0, under matrix multiplication. Show that G and G are isomorphic by exhibiting an
isomorphism of G onto G.
 
a −b
Proof. Define a mapping φ : G → G by φ(a + bi) = . φ is a homomorphism since
b a

φ((a + bi) · (c + di)) = φ(ac − bd + (ad + bc)i)

     
ac − bd ad + bc a −b c −d
= = ·
ad + bc ac − bd b a d c
= φ(a + bi)φ(c + di).

Also, φ is clearly one-one and onto. Hence, φ yields an isomorphism of G onto G.

17. Let G be the group of real numbers under addition and let N be the subgroup of
G consisting of all integers. Prove that G/N is isomorphic to the group of all complex
numbers of absolute value 1 under multiplication.

Proof. For convenience of the proof, we denote S1 to be the group of all complex numbers
of absolute value 1. We define a mapping φ : G → S1 by φ(g) = e2πgi , where e is the euler
constant, i is the imaginary number. We show φ is a homomorphism. Note that

φ(g + h) = e2π(g+h)i = e2πgi · e2πhi = φ(g)φ(h)

6
so that φ is homomorphism. Moreover, from that fact that arbitrary elements of S1 is
expressible in the form e2πki , φ is onto. So, we can apply the isomorphism theorem. We
now investigate the kernel of φ. Since e2πki = 1 ⇐⇒ k ∈ Z, N is exactly the kernel of φ.
Hence, G/N ' S1 .
 
a b
18. Let G be the group of all real 2 × 2 matrices , with ad − bc 6= 0, under matrix
b a
multiplication, and let
  
a b
N= ∈ G : ad − bc = 1 .
b a
Prove that N ⊃ G0 , the commutator subgroup of G.

Proof. It is sufficient to check that the determinant of commutator is 1. Note that

det(ABA−1 B −1 ) = det(A) det(B) det(A−1 ) det(B −1 ) = 1 for all A, B ∈ G. Hence, G0 ⊂
N.

19. In Problem 18 show, in fact, that N = G0 .

Proof. Note that for any x ∈ R,

     −1  −1
1 x 2 0 1 x 2 0 1 x
= ,
0 1 0 1 0 1 0 1 0 1
  1   1 −1  −1
1 0 2 0 1 0 2 0 1 0
= ,
x 1 0 1 x 1 0 1 x 1
     −1  −1
x 0 x 0 0 1 x 0 0 1
=
0 x−1 0 1 1 0 0 1 1 0
     
1 x 1 0 x 0
so that , and are commutators. Also,
0 1 x 1 0 x−1
     −1  −1
0 1 1 2 −1 0 1 2 −1 0
=
−1 0 0 1 1 2 0 1 1 2
 
0 1
so that is also a commutator. We can also check that
−1 0

1 ab
     
a b 1 c a 0
=
c d 0 1 0 1 0 a1
for a 6= 0. If a = 0, then it is must that b 6= 0 and
 1
1 − db
    
0 b 0 1 b 0
=
c d −1 0 0 1 0 b

7
  
a b
so that in either cases, is a product of commutators. Therefore, N ⊂ G0 and hence,
c d
N = G0 .
 
a b
20. Let G be the group of all real 2 × 2 matrices of the form , where ad 6= 0, under
0 d
 
0 1 x
matrix multiplication. Show that G is precisely the set of all matrices of the form .
0 1
 
1 x
Proof. Let N be the set of all matrices of the form . Then from the Problem 21 of
0 1
0
 that N is normal subgroup of G and G/N is abelian, so that G ⊂ N .
section 2.6,We have
1 x
Now, take ∈ N . Just like we have done in Problem 19,
0 1
     −1  −1
1 x 2 0 1 x 2 0 1 x
= ,
0 1 0 1 0 1 0 1 0 1

hence a commutator of G. Hence, N ⊂ G0 . Therefore, N = G0 .

21. Let S1 and S2 be two sets. Suppose that there exists a one-to-one mapping ψ of S1
into S2 . Show that there exists an isomorphism A(S1 ) into A(S2 ), where A(S) mean the
set of all one-to-one mapping of S onto itself.

Proof. Note that ψ is one-to-one. Hence, we define g : ψ(S1 ) → S1 by g(y) = ψ −1 |ψ(S1 ) (y).
Consequently, g ◦ ψ = idS1 . Now we define a mapping f : A(S1 ) → A(S2 ) by
(
ψ ◦ φ ◦ g(y), if y ∈ ψ(S1 ),
f (φ)(y) =
idS2 (y), else

We see that f (φ) ∈ A(S2 ). Moreover, it is a homomorphism since

(
ψ ◦ φ1 φ2 ◦ g(y) = (ψ ◦ φ1 ◦ g) ◦ (ψ ◦ φ2 ◦ g)(y), if y ∈ ψ(S1 ),
f (φ1 ◦ φ2 ) =
idS2 (y), else
= f (φ1 )f (φ2 )

and also one-one clearly. Hence, we have exhibited an isomorphism(monomorphism) of

A(S1 ) to A(S2 ).