Math 120 Problem set 1 solutions

Anssi Lahtinen
October 5, 2008
Problem 1 (1.1.9b). Let G = a + b

2 R : a, b Q. Prove that the

nonzero elements of G are a group under multiplication.
Solution. We need to check that multiplication is a binary operation on
G0 and that it satises the group axioms.
1. If a
1
+ b
1

2 and a
2
+ b
2

2 are in G0, then so is

(a
1
+ b
1

2)(a
2
+ b
2

2) = (a
1
a
2
+ 2b
1
b
2
) + (a
1
b
2
+ a
2
b
1
)

2.
Thus G0 is closed under multiplication.
2. Associativity of multiplication on G0 follows from the associativity
of multiplication of the reals.
3. 1 = 1 + 0

2 belongs to G0 and provides an identity element for

multiplication.
4. if a + b

2 is in G0, then
1
a + b

2
=
a
a
2
2b
2

b
a
2
2b
2

2
is also in G0, and provides an inverse for a + b

2. Observe that
a
2
2b
2
,= 0, so that the RHS of the above equation makes sense. For
suppose we had a
2
2b
2
= 0. Since by choice of a+b

2 at least one of
a and b is non-zero, it would then follow that both must be non-zero.
In particular, we would have b ,= 0, and the equation a
2
= 2b
2
would
imply that

2 = a/b Q,
which is a contradiction.
1
Problem 2 (1.1.22). If x and g are elements of the group G, prove that
[x[ = [g
1
xg[. Deduce that [ab[ = [ba[ for all a, b G.
Solution. Let us rst show that [g
1
xg[ [x[. If [x[ = , this is obvious.
Suppose [x[ = n < . Then
(g
1
xg)
n
= (g
1
xg)(g
1
xg) (g
1
xg)
. .
n
= g
1
x
n
g = g
1
eg = e
whence [g
1
xg[ [x[ in this case as well. Substituting g
1
xg for x and g
1
for g, we also get the the inequality
[x[ = [g(g
1
xg)g
1
[ [g
1
xg[.
Thus [x[ = [g
1
xg[ as claimed. To prove the latter claim it now suces to
observe that ba = a
1
(ab)a.
Problem 3 (1.1.25). Prove that if x
2
= 1 for all x G, then G is abelian.
Solution. Suppose a, b G. By assumption, we have
abab = (ab)
2
= 1.
Multiplying both sides of this equation from the left by a and from the right
by b and using the assumption that a
2
= 1 and b
2
= 1, we get the equation
ba = ab.
Problem 4 (1.1.31). Prove that any nite group G of even order contains
an element of order 2.
Solution. Denote
t(G) = g G : g ,= g
1
.
We can write t(G) as a disjoint union
t(G) = g
1
, g
1
1
g
2
, g
1
2
g
k
, g
1
k

where each member of the of the union has 2 elements. Thus [t(G)[ is even.
Since [G[ is even, we see that [G t(G)[ = [G[ [t(G)[ is even. Since
e G t(G), it follows that G t(G) contains at least two elements. Let
g G t(G) be an element distinct from the identity. Then g = g
1
, and
therefore (multiplying both sides by g) we have g
2
= e. Since g
1
= g ,= e, it
follows that the element g has order 2.
2
Problem 5 (1.2.16). Show that the group x
1
, y
1
: x
2
1
= y
2
1
= (x
1
y
1
)
2
= 1)
is the dihedral group D
4
.
Solution. The following computation shows that, given the relations x
2
1
=
y
2
1
= 1, the relation (x
1
y
1
)
2
= 1 is equivalent to the relation x
1
y
1
= y
1
x
1
1
:
(x
1
y
1
)
2
= 1
x
1
y
1
x
1
y
1
= 1 (expanding LHS)
x
1
y
1
x
1
y
2
1
= y
1
(multiplying by y
1
from the right)
x
1
y
1
x
1
= y
1
(using y
2
1
= 1)
x
1
y
1
= y
1
x
1
1
(multiplying by x
1
1
from the right).
It follows that
x
1
, y
1
: x
2
1
= y
2
1
= (x
1
y
1
)
2
= 1) = x
1
, y
1
: x
2
1
= y
2
1
= 1, x
1
y
1
= y
1
x
1
1
).
Here the RHS is D
4
by 1.1 on p. 26 (with x
1
= r and y
1
= s).
Problem 6 (1.3.19). Find all numbers n such that S
7
contains an element
of order n.
Solution. (Sketch) By Exercise 1.3.15, the order of an element of S
n
is given
by the least common multiple of the lengths of the cycles in its cycle de-
composition. Listing all possible cycle decomposition types for elements of
S
7
(there are 15 of them, given by the partitions of 7) and computing the
order of the corresponding elements, we arrive at the following list: 1, 2, 3,
4, 5, 6, 7, 10 and 12.
Problem 7 (1.3.20). Find a set of generators and relations for S
3
.
Solution. We have S
3
= 1, (12), (13), (23), (123), (132). Let a = (123) and
b = (12). Then
(12) = b, (13) = ab, (23) = a
2
b, (123) = a and (132) = a
2
,
so a and b generate S
3
. Observe that a
3
= b
2
= 1, and that ba = (23) = a
2
b.
We claim that
a, b : a
3
= b
2
= 1, ba = a
2
b) (1)
is a presentation for S
3
. Observe rst that the relations allow us to write
every element in the group given by the presentation (1) as a product a
k
b
l
,
where 0 k 2 and 0 l 1. Thus the group described by (1) has at most
6 elements. On the other hand, the relations in (1) hold in S
3
, which has
order 6. Therefore the group described by (1) must have exactly 6 elements,
and is isomorphic to S
3
.
3
Problem 8 (1.4.8). Show that GL
n
(F) is non-abelian for any n 2 and
any F.
Solution. Suppose n 2, and denote
A =
_
_
1 1
0 1
I
n2
_
_
, B =
_
_
0 1
1 0
I
n2
_
_
.
Then det(A) = 1 and det(B) = 1, so A and B are in GL
n
(F). Since
AB =
_
_
1 1
1 0
I
n2
_
_
and BA =
_
_
0 1
1 1
I
n2
_
_
,
we see that AB ,= BA. Thus GL
n
(F) is non-abelian, as claimed.
Problem 9 (1.5.1). Compute the order of each of the elements in Q
8
.
Solution. The neutral element 1 has order 1, and the relation (1) (1) = 1
implies that 1 has order 2. It follows that i
4
= (1)
2
= 1. Since i ,= 1,
i
2
= 1 ,= 1 and i
3
= (1) i = i ,= 1, we see that i has order 4. We also
have (i)
2
= (1) i (1) i = (1)
2
i
2
= 1, whence (i)
4
= (1)
2
= 1.
Since i ,= 1, (i)
2
= 1 ,= 1 and (i)
3
= (1) (i) = i ,= 1, it follows
that i has order 4. Now similar arguments show that j and k also have
order 4.
Problem 10 (1.6.4). Show that the multiplicative groups R0 and C
0 are not isomorphic.
Solution. Suppose x R 0 is an element of nite order. Then x
k
= 1
for some integer k 1, and therefore [x[
k
= [x
k
[ = 1. It follows that [x[ = 1,
whence x = 1, which has order 1, or x = 1, which has order 2. Thus any
element of nite order in R 0 has order at most 2. Since i C 0
has order 4, R0 and C0 cannot be isomorphic.
Problem 11 (1.6.9). Show that D
24
and S
4
are not isomorphic.
Solution. The element r D
24
has order 12. On the other hand, using the
same method as in the solution of problem 1.3.19 we see that each element
of S
4
has order 1, 2, 3 or 4, whence the claim follows.
Problem 12 (1.6.17). Let G be a group. Show that the map from G to itself
dened by g g
1
is a homomorphism if and only if G is abelian.
4
Solution. Let f : G G be the map g g
1
. Suppose rst that f is a
homomorphism. Then for any a, b G we have
ab = f((ab)
1
) = f(b
1
a
1
) = f(b
1
)f(a
1
) = ba
whence G is abelian. To see the converse, suppose that G is abelian. Then
for any a, b G we have
f(ab) = (ab)
1
= b
1
a
1
= a
1
b
1
= f(a)f(b)
whence f is a homomorphism.
Problem 13 (1.6.20). Let G be a group and let Aut(G) be the set of all
isomorphisms from G onto G. Prove that Aut(G) is a group under function
composition.
Solution. (Sketch) Observe that Aut(G) is closed under function composi-
tion, ie. that the composite of two isomorphisms from G onto G is again an
isomorphism from G onto G. Associativity of the multiplication in Aut(G)
follows from the associativity of function composition. The identity map
id
G
: G G provides a neutral element. Observe that for an isomorphism
f : G G, the inverse map f
1
: G G is also an isomorphism, and pro-
vides an inverse for f with respect to the multiplication in Aut(G). Thus
all the group axioms are satised, and Aut(G) is a group as claimed.
Problem 14 (1.6.24). Let G be a nite group and let x and y be distinct
elements of order 2 in G that generate G. Show that G D
2n
, where
n = [xy[.
Solution. Recall that from (1.1) on page 26 that D
2n
has a presentation
D
2n
= r, s : r
n
= s
2
= 1.rs = sr
1
).
Observe that the elements xy and x of G satisfy the relations (xy)
n
= 1 (by
the choice of n), x
2
= 1 (by assumption) and
(xy)x = xyx = xy
1
x
1
= x(xy)
1
where equality in the middle follows from the assumption that x
2
= y
2
= 1.
By the discussion on pages 38 and 39, it follows that the assignments f(r) =
xy, f(s) = x dene a homomorphism f : D
2n
G. Since f(s) = x and
f(sr) = f(s)f(r) = x
2
y = y, and x and y generate G, the homomorphism
f is surjective.
5
To complete the proof it is now enough to show that f is injective. Recall
that any element of D
2n
can be uniquely expressed as a product s
k
r
l
, where
0 k 1 and 0 l n 1. Suppose s
k
r
l
and s
k

r
l

, 0 k, k

1,
0 l, l

n 1. are elements of D
2n
such that f(s
k
r
l
) = f(s
k

r
l

). Then
(xy)
k
x
l
= (xy)
k

x
l

and therefore
(xy)
a
= x
b
, (2)
where a = k k

and b = l

l. Without loss of generality we may assume

that k k

. Then 0 a n 1 and b = 1, 0 or 1. To prove injectivity,

we must show that in fact a = 0 and b = 0.
Let us rst show that b = 0. Suppose we had b ,= 0. Then, since x
2
= 1,
we would have
(xy)
a
= x.
Multiplying by xy from the right and using the assumption x
2
= 1, this
would give rise to the equation
(xy)
a+1
= y;
multiplying this equation by xy from the left and using y
2
= 1 we would
then get the equation
(xy)
a+2
= x.
Continuing inductively we see that we would have either (xy)
n
= x or
(xy)
n
= y. Since (xy)
n
= 1, this would mean that either x = 1 or y = 1,
which contradicts the assumption that x and y have order 2. Therefore we
must have b = 0.
Having proved that b = 0, equation (2) gives us the equation
(xy)
a
= 1.
Since 0 a n 1 and xy has order n, it follows that a = 0. Thus a = 0
and b = 0, and the claim follows.
Problem 15 (1.6.26). Let i and j be the generators of Q
8
described in
Section 5. Prove that the map from Q
8
to GL
2
(C) dened on generators
by
(i) =
_
1 0
0

1
_
and (j) =
_
0 1
1 0
_
extends to a homomorphism. Prove that is injective.
6
Solution. (Sketch) Let us extend the given assignments by dening
(1) :=
_
1 0
0 1
_
, (k) := (i)(j) =
_
0

1 0
_
and
(a) := (a)
for a 1, i, j, k. Straightforward computation then shows that the matri-
ces (1), (i), (j) and (k) satisfy the relations analogous to the
ones given on p. 36, whence is a homomorphism. To show that is
injective, one can simply write down the matrices (1), (i), (j) and
(k) and observe that they are all distinct.
Problem 16 (1.7.18). Let H be a group acting on a set A. Prove that the
relation on A dened by
a b if and only if a = hb for some h H
is an equivalence relation. (The equivalence class of x A under is called
the orbit of x under the action of H. The orbits partition the set A.)
Solution. For any a A we have a = 1 a, and therefore a a. Thus
is reexive. If a b, then a = hb for some h H, and therefore
b = h
1
(hb) = h
1
a, whence b a. Thus is symmetric. Finally, if a b
and b c, then a = h
1
b and b = h
2
c for some h
1
, h
2
H. It follows that
a = h
1
b = h
1
(h
2
c) = (h
1
h
2
)c, whence a c. Thus is transitive, and the
proof is complete.
Problem 17 (1.7.19). Let H be a subgroup of the nite group G and let H
act on G by left multiplication. Let x G and let O be the orbit of x under
the action of H. Prove that the map
H O dened by h hx
is a bijection (hence all orbits have cardinality [H[). From this and previous
exercise deduce Lagranges theorem:
if G is a nite group and H is a subgroup of G then [H[ divides [G[.
Solution. Let f : H O be the map h hx. If y O, then y = hx for
some h x. But then y = f(h). Thus f is surjective. On the other hand,
if f(h
1
) = f(h
2
), then
h
1
x = h
2
x,
7
and multiplying from the right by x
1
shows that h
1
= h
2
. Thus f is
also injective, and hence a bijection. Therefore [O[ = [H[. The set G is
the disjoint union of the orbits of the H-action. Since the cardinality of
each orbit is [H[, the cardinality of G is an integer multiple of [H[, whence
Lagranges theorem follows.
Problem 18 (1.7.20). Show that the group of rigid motions of a tetrahedron
is isomorphic to a subgroup of S
4
.
Solution. Let G be the group of rigid motions of the tetrahedron. It suces
to nd an injective homomorphism from G to S
4
. Let V be the set of
vertices of the tetrahedron, and notice that each rigid motion g G of the
tetrahedron restricts to give a permutation
g
V
: V V, v g(v)
of V . The map
f : G S
V
, g g
V
is easily veried to be a homomorphism (in both G and S
V
the group oper-
ation is given by composition of functions). Moreover, f is injective, since a
rigid motion of the tetrahedron is completely determined by where it sends
the vertices. Since [V [ = 4, we have S
V
S
4
, and the claim follows.
Problem 19 (1.7.23). Explain why the action of the group of rigid motions
of a cube on the set of three pairs of opposite faces is not faithful. Find the
kernel of this action.
Solution. Let S be the set of pairs of opposite faces of the cube. The action
on S is not faithful, since for example the rotation of the cube by 180 degrees
about the axis through the centers of the top and bottom face acts trivially
S. Interpret the cube as the subset [1, 1]
3
of R
3
, and let e
1
, e
2
and e
3
be
the standard basis vectors of R
3
. Then the elements in the kernel of the
action on S are exactly those linear transformations f such that f(e
i
) = e
i
for all i and such that det(f) = 1. Thus the kernel can be described in terms
of matrices as
_
_
_
_
_

1
0 0
0
2
0
0 0
1

2
_
_

1
,
2
= 1
_
_
_
.
It is isomorphic to Z/2 Z/2
Problem 20. Show that every element of a nite group has nite order. Is
the converse true?
8
Solution. Let G be a nite group, and let g G. Consider the map
f : N G, n g
n
.
Since G is nite, f cannot be injective, whence there exists n, m N such
that n ,= m and
g
n
= g
m
.
Without loss of generality we may assume that n > m. Now multiplying
both sides by g
m
we see that g
nm
= 1, whence the order of g is at most
n m.
The converse is not true. For example, consider the set of all sequences
(Z/2)
N
= (x
0
, x
1
, . . .) : x
i
Z/2.
It is easy to see that addition given by the rule
(x
0
, x
1
, . . .) + (y
0
, y
1
, . . .) = (x
0
+ y
0
, x
1
+ y
1
, . . .).
makes (Z/2)
N
into an abelian group where each non-identity element has
order 2. However, (Z/2)
N
is clearly innite.
9