Page 85

Note that, for elements g and h in a group (G, ⋆), we sometimes write gh rather
than g ⋆ h. Furthermore, we also define
gg = g 2 , ggg = g 3 , etc, (4.10)
as well as
g −1 g −1 = g −2 , g −1 g −1 g −1 = g −3 , etc, (4.11)
and
g 0 = I. (4.12)
Definition 4.0.5. The order of an element g in a group (G, ⋆) is the least positive
integer n such that g n = I, or is infinite if g m ̸= I for any m > 0. It is denoted by
o(g).

Example 4.0.7. Some examples are as follows.

ˆ Consider again the group presented in Example 4.0.3, which consists of the
set G = {1, −1} together with the operation of multiplication. The identity
element of this group is 1. The element −1 in this group has order 2:

−1 ̸= 1,
(4.13)
(−1)2 = 1.

ˆ Consider again the group presented in Example 4.0.4, which consists of the
set G = {1, i, −1, −i} together with the operation of multiplication. Again,
the identity element of this group is 1. The element −1 has order 2, since
(−1)2 = 1, while the element i has order 4:

(i)2 = −1,
(i)3 = −i, (4.14)
(i)4 = 1.

ˆ The element  
−1 0
g= (4.15)
0 1
of the group GL(2, R) has order 2:
    
2 −1 0 −1 0 1 0
g = = . (4.16)
0 1 0 1 0 1

ˆ The element  √ 
−1/2 − 3/2
g= √ (4.17)
3/2 −1/2
of the group GL(2, R) has order 3. Exercise: check this.

ˆ Every non-zero element of the group (R, +) is of infinite order.

 Mathematical Methods in Chemistry (MATH0042) Page 86

4.1.1 Cyclic groups

Definition 4.0.6. A group (G, ⋆) is said to be cyclic if there exists an element g ∈ G
such that
G = {g n : n ∈ Z}. (4.18)
Then we say that G is generated by g.
Note that if g is the generator of a cyclic group (G, ⋆), then so is the inverse,
g −1 , of g, since {g n : n ∈ Z} = {g −n : n ∈ Z}.

Example 4.0.8. Some examples are as follows.

ˆ Consider again the group presented in Example 4.0.3, which consists of the
set G = {1, −1} together with the operation of multiplication. This is a cyclic
group with generator −1:

(−1) × (−1) = 1,
(−1) × (−1) × (−1) = −1,
(−1) × (−1) × (−1) × (−1) = 1, (4.19)
..
.

ˆ Consider again the group presented in Example 4.0.4, which consists of the
set G = {1, i, −1, −i} together with the operation of multiplication. This is a
cyclic group with generator i:

i × i = −1,
i × i × i = −i,
i × i × i × i = 1, (4.20)
i × i × i × i × i = i,
..
.

One can check that −i is also a generator of this group ((−i)2 = −1, (−i)3 = i,
(−i)3 = 1). −i is the inverse of i in this group. However, −1 is not a generator
of this group.

ˆ (Z, +) is a cyclic group of infinite order, with generator 1. Note, for example,
that 2 is not a generator of this group since (under addition) it generates only
the even integers.

We will not prove this, but one can show that the order of a cyclic group
is equal to the order of its generator (one can show that if o(g) = n, say, then
g, g 2 , g 3 , . . . , g n−1 and the identity element I are all distinct).
All cyclic groups have the same structure, namely, every cyclic group of finite
order n takes the form
Cn = {I, g, g 2 , . . . , g n−1 }, (4.21)
Page 87

while every cyclic group of infinite order takes the form

C∞ = {. . . , g −2 , g −1 , I, g, g 2 , . . . }, (4.22)

for some generator g. Cn and C∞ are commonly referred to as the cyclic groups of
order n and the infinite cyclic group, respectively.

Example 4.0.9. The group table for C3 = {I, g, g 2 } is

I g g2
I I g g2
g g g2 I
g2 g2 I g
Chapter 5

Symmetries

A symmetry operation is an action that leaves an object looking the same as to

begin with. Consider a molecule of water, as illustrated in figure 5.1. What are its
symmetry operations?

Figure 5.1: A water molecule and its symmetry operations.

Suppose that it is oriented as shown in figure 5.1, with the oxygen atom above
the two hydrogen atoms. Then the following are its symmetry operations:
(i) rotation by 180 degrees about the red vertical axis,
(ii) reflection in the (green) plane of the molecule,
(iii) reflection in the (blue) vertical plane that is perpendicular to the plane of the
molecule,

88
Page 89

(iv) leave it as it is - this is called the identity operation.

The water molecule thus has four symmetry operations. Molecular symmetry can
be used to predict or explain many of a molecule’s chemical properties.
If you perform one symmetry operation on an object and then another, obviously
the object still looks the same as to begin with. Therefore the composition of two
symmetry operations (one after the other) of an object is also a symmetry operation
of that object. (For example, one can check that if you reflect the water molecule
in one plane of symmetry and then rotate it by 180 degrees, this is the same as
reflecting it in the other plane of symmetry - one can either check this visually, or
demonstrate it in a formulaic way, as we will later.) Thus, the set of symmetry
operations of an object under the composition of these operations evidently satisfies
the closure axiom of a group. Also, evidently, this set contains an identity element
- the identity operation. In fact, symmetries of a molecule or of a physical system
often form a group.

5.1 Representation of symmetries in matrix form

In this section we will consider transformations acting on points in 3-dimensional
space, R3 . We will describe how one can represent these transformations - including
symmetry operations - in terms of matrices.

5.1.1 Reflections
Reflections in the 3-dimensional space are characterised by a plane of reflection,
which is preserved under the transformation. Some important examples are as
follows.
The reflection in the xy-plane of a point (x, y, z) is the point (x, y, −z) - see
figure 5.2.
z

(x, y, z)

x y

(x, y, −z)

Figure 5.2: Reflection in the xy-plane.

 Mathematical Methods in Chemistry (MATH0042) Page 90

One can check that the position vectors of these two points are related by
    
1 0 0 x x
0 1 0   y  =  y  . (5.1)
0 0 −1 z −z

Hence reflection in the xy-plane corresponds to the following matrix Pxy :

 
1 0 0
Pxy = 0 1 0  (5.2)
0 0 −1

Similarly, reflections in the zx-plane and the yz-plane correspond to the matrices
Pzx and Pxz , respectively, which are given by
 
1 0 0
Pzx = 0 −1 0 (5.3)
0 0 1

and  
−1 0 0
Pyz =  0 1 0 . (5.4)
0 0 1

5.1.2 Rotations
Rotations in 3-dimensional space are characterised by an axis of rotation (which is
preserved by the rotation) and an angle of rotation. Some important examples are
as follows.

ˆ Rotation about the x-axis through an angle θ in the counter-clockwise direction

in the yz-plane:  
1 0 0
Rx (θ) = 0 cos θ − sin θ (5.5)
0 sin θ cos θ

ˆ Rotations about the y-axis through an angle θ in the counter-clockwise direc-

tion in the zx-plane:
 
cos θ 0 sin θ
Ry (θ) =  0 1 0  (5.6)
− sin θ 0 cos θ

ˆ Rotations about the z-axis through an angle θ in the counter-clockwise direc-

tion in the xy-plane:
 
cos θ − sin θ 0
Rz (θ) =  sin θ cos θ 0 (5.7)
0 0 1
Page 91

It is perhaps easiest to deduce (5.7). Evidently, under any rotation about the z-axis,
the z-coordinate of a point will remain unchanged. One way to deduce the change in
the x- and y-coordinates is as follows. Consider the point (x, y) that lies in the xy-
plane. Let (x′ , y ′ ) denote the image of (x, y) after rotation about the origin through
an angle θ in the counter-clockwise direction. Suppose that the polar coordinates of
(x, y) are r and ϕ, i.e.,
x = r cos ϕ, y = r sin ϕ. (5.8)
Then
x′ = r cos(ϕ + θ) = r(cos ϕ cos θ − sin ϕ sin θ) = (cos θ)x − (sin θ)y
(5.9)
y ′ = r sin(ϕ + θ) = r(sin ϕ cos θ + cos ϕ sin θ) = (sin θ)x + (cos θ)y.
Or, in vector form,  ′   
x cos θ − sin θ x
= . (5.10)
y′ sin θ cos θ y
(Alternatively, one can represent these points by complex numbers: (x, y) by x + iy
and (x′ , y ′ ) by x′ + iy ′ . Then we must have
x′ + iy ′ = eiθ (x + iy)
= (cos θ + i sin θ)(x + iy) (5.11)
= (x cos θ − y sin θ) + i(x sin θ + y cos θ),
which - after comparing real and imaginary parts - leads one again to (5.10).)
Hence it follows that the rotation of a point in 3-dimensional space through an
angle θ in the counter-clockwise direction about the z-axis is obtained by applying
the matrix Rz (θ) - as given by (5.7) - to the position vector of the point. One can
then deduce the forms (5.5) and (5.6) for the matrices corresponding to rotations
about the x and y-axes, respectively, by swapping the entries of (5.7) accordingly.
More specifically, if we swap x, y and z by y, z and x, respectively, then (5.10)
becomes  ′   
y cos θ − sin θ y
′ = , (5.12)
z sin θ cos θ z
which leads to (5.5) (remember that rotation about the x-axis will leave the x-
coordinate of a point unchanged). While if we swap x, y and z by z, x and y,
respectively, then (5.10) becomes
 ′   
z cos θ − sin θ z
′ = , (5.13)
x sin θ cos θ x
which leads to (5.6).

5.1.3 Inversion
The inversion with respect to the origin of a point (x, y, z) is defined to be the point
(−x, −y, −z). It is straightforward to check that this operation is described by the
matrix  
−1 0 0
Inv =  0 −1 0  . (5.14)
0 0 −1
 Mathematical Methods in Chemistry (MATH0042) Page 92

It can be shown that the matrix that corresponds to a rotation (such as Rx (θ),
Ry (θ) and Rz (θ) above) is an orthogonal matrix with determinant 1 (see Problem
Sheet 9), and hence an element of the special orthogonal group SO(3). Rotations
in the common sense of the word are sometimes referred to more specifically as
proper rotations. Another important type of transformation is what is known as an
improper rotation; an improper rotation consists of a rotation followed by reflection
in a plane that is orthogonal to the axis of the preceding rotation. It can be shown
that the matrix that corresponds to an improper rotation is an orthogonal matrix
with determinant −1. Special cases of improper rotations are simply reflections
(these are improper rotations for which the corresponding initial rotation is though
an angle simply of 0). Hence the matrix that corresponds to a reflection is an
orthogonal matrix with determinant −1. (See Problem Sheet 9.) Note also that the
inversion matrix Inv is also an orthogonal matrix with determinant −1.

5.1.4 Translations
Consider now the translation of a point by a vector v,
 
v1
v = v2  ,
v3
namely,    
x x + v1
y  7→ y + v2  .
z z + v3
Describing this in matrix form is slightly trickier. We do so as follows. We replace
the three-dimensional vector  
x
y 
z
by a four-dimensional vector with last component equal to 1:
 
x
y 
 . (5.15)
z 
1
Then, the translation by v is represented by the 4 × 4 matrix
 
1 0 0 v1
0 1 0 v2 
T (v) = 
0
.
0 1 v3 
0 0 0 1
Applying T (v) to our four dimensional vector, we get
      
x 1 0 0 v1 x x + v1
y  0 1 0 v2  y  y + v2 
T (v) 
z  = 0 0 1 v3  z  = z + v3  .
     

1 0 0 0 1 1 1
Page 93

Finally, our translated three-dimensional vector is recovered by dropping the fourth

component (equal to 1) of the resulting vector.

5.1.5 Composition of transformations

The matrix that corresponds to the composition of two transformations is equal to
the product of the matrices that correspond to these two transformations (note that
these must be multiplied in the correct order).
For example, a rotation through π/4 (counter-clockwise) about the z-axis fol-
lowed by a reflection in the xy-plane corresponds to the matrix
  √ √  √ √ 
π  1 0 0 √ 2/2 −√ 2/2 0 √ 2/2 −√ 2/2 0
Pxy Rz = 0 1 0   2/2 2/2 0 =  2/2 2/2 0 .
4
0 0 −1 0 0 1 0 0 −1

The only exception to this is when the composition involves a translation. In

this case, before multiplying them together, we need to add to all of the above
transformation matrices a fourth row and a fourth column whose entries are all zero
with the exception of the entry in the fourth row and fourth column (i.e., the lower
right-hand corner), where we put a 1. Then, we end up with 4×4 matrices which can
be multiplied together and which act on four-dimensional vectors of the form (5.15)
(applying these 4 × 4 matrices to such a four-dimensional vector, gives another four-
dimensional vector, and - as described above for translations - we recover the three-
dimensional position vector of the transformed point in three-dimensional space by
dropping the fourth component of this final four-dimensional vector). Where there
is no risk of confusion, we will denote these ‘extended’ matrices by the same name.
For example, the rotation matrix Rz (θ) becomes
 
cos θ − sin θ 0 0
 sin θ cos θ 0 0
Rz (θ) = 
 0
,
0 1 0
0 0 0 1

whereas the reflection in the zx-plane becomes

 
1 0 0 0
0 −1 0 0
Pzx = 0 0 1
.
0
0 0 0 1

Example 5.0.1. Write down the matrix associated with a counter-clockwise rota-
tion through an angle θ about the x axis followed by a translation by the vector
 
1
v = 2 .

0
 Mathematical Methods in Chemistry (MATH0042) Page 94

Such matrix will be given by

    
1 0 0 1 1 0 0 0 1 0 0 1
0 1 0 2 0 cos θ − sin θ 0 0 cos θ − sin θ 2
T (v) Rx (θ) = 
0
 = .
0 1 0 0 sin θ cos θ 0 0 sin θ cos θ 0
0 0 0 1 0 0 0 1 0 0 0 1

Note that, in general, the order in which we perform the operations matters.
For instance, if in the above example we translate first and subsequently rotate, we
instead end up with
    
1 0 0 0 1 0 0 1 1 0 0 1
0 cos θ − sin θ 0 0 1 0 2 0 cos θ − sin θ 2 cos θ
Rx (θ) T (v) = 
0 sin θ cos θ 0 0 0 1 0 = 0 sin θ cos θ
   .
0 
0 0 0 1 0 0 0 1 0 0 0 1

5.1.6 Inverse operations

The above transformations admit inverses, as described next.
Reflections. Every reflection is its own inverse. One can check this by multi-
plying the matrix associated with a reflection by itself - one will obtain the identity
matrix. For example
    
1 0 0 1 0 0 1 0 0
Pxy Pxy = 0 1 0  0 1 0  = 0 1 0 . (5.16)
0 0 −1 0 0 −1 0 0 1

Rotations. The inverse of a rotation by θ in the counter-clockwise direction

about some axis, is a rotation by −θ in the counter-clockwise direction about the
same axis. For example, the inverse of Rx (θ) is Rx (−θ):
    
1 0 0 1 0 0 1 0 0
Rx (θ) Rx (−θ) = 0 cos θ − sin θ 0 cos θ sin θ  = 0 1 0 .
0 sin θ cos θ 0 − sin θ cos θ 0 0 1

Inversion. The inversion is its own inverse. Indeed, one can check that

Inv Inv = I.

Translation. The inverse of a translation by a vector v is the translation by the

vector −v. Indeed, one can check that

T (−v)T (v) = T (v)T (−v) = I, (5.17)

where here I denotes the 4 × 4 identity matrix.

Page 95

Figure 5.3: A water molecule and its symmetry operations.

5.1.7 Example
Let us return now to the symmetry operations of a water molecule. Consider again
the molecule as drawn here. Take the red vertical axis to be the z-axis, the (green)
plane of the molecule to be the yz-plane, and the (blue) vertical plane that is per-
pendicular to the plane of the molecule to be the zx-plane.
Then the symmetries of the molecule are a rotation through 180 degrees - or
π radians - about the z-axis, and reflections in the yz- and zx-planes, along with
the identity operation. These correspond to the matrices Rz (π), Pyz , Pzx and the
identity matrix, respectively. One can show that the set of these matrices under the
operation of matrix multiplication is a group, in fact an Abelian group (see Problem
Sheet 9). Equivalently, the set of symmetry operations of the water molecule under
the composition of these operations, is an Abelian group.