Chapter 5

COMPLEX NUMBERS
5.1

Constructing the complex numbers

One way of introducing the field C of complex numbers is via the arithmetic
of 2 2 matrices.
DEFINITION 5.1.1 A complex number is a matrix of the form


x y
,
y
x
where x and y are real numbers.

x 0
are scalar matrices and are called
0 x
real complex numbers and are denoted by the symbol {x}.
The real complex numbers {x} and {y} are respectively
 called the real

x y
.
part and imaginary part of the complex number
y
x


0 1
The complex number
is denoted by the symbol i.
1
0
Complex numbers of the form

We have the identities



 
 
 


x y
x 0
0 y
x 0
0 1
y 0
=
+
=
+
y
x
0 x
y
0
0 x
1
0
0 y
= {x} + i{y},


 

0 1
0 1
1
0
i2 =
=
= {1}.
1
0
1
0
0 1
89

90

CHAPTER 5. COMPLEX NUMBERS

Complex numbers of the form i{y}, where y is a nonzero real number, are
called imaginary numbers.
If two complex numbers are equal, we can equate their real and imaginary
parts:
{x1 } + i{y1 } = {x2 } + i{y2 } x1 = x2 and y1 = y2 ,
if x1 , x2 , y1 , y2 are real numbers. Noting that {0} + i{0} = {0}, gives the
useful special case is
{x} + i{y} = {0} x = 0 and y = 0,
if x and y are real numbers.
The sum and product of two real complex numbers are also real complex
numbers:
{x} + {y} = {x + y}, {x}{y} = {xy}.
Also, as real complex numbers are scalar matrices, their arithmetic is very
simple. They form a field under the operations of matrix addition and
multiplication. The additive identity is {0}, the additive inverse of {x} is
{x}, the multiplicative identity is {1} and the multiplicative inverse of {x}
is {x1 }. Consequently
{x} {y} = {x} + ({y}) = {x} + {y} = {x y},
 
x
{x}
1
1
1
.
= {x}{y} = {x}{y } = {xy } =
{y}
y

It is customary to blur the distinction between the real complex number

{x} and the real number x and write {x} as x. Thus we write the complex
number {x} + i{y} simply as x + iy.
More generally, the sum of two complex numbers is a complex number:
(x1 + iy1 ) + (x2 + iy2 ) = (x1 + x2 ) + i(y1 + y2 );

(5.1)

and (using the fact that scalar matrices commute with all matrices under
matrix multiplication and {1}A = A if A is a matrix), the product of
two complex numbers is a complex number:
(x1 + iy1 )(x2 + iy2 ) = x1 (x2 + iy2 ) + (iy1 )(x2 + iy2 )
= x1 x2 + x1 (iy2 ) + (iy1 )x2 + (iy1 )(iy2 )
= x1 x2 + ix1 y2 + iy1 x2 + i2 y1 y2
= (x1 x2 + {1}y1 y2 ) + i(x1 y2 + y1 x2 )

= (x1 x2 y1 y2 ) + i(x1 y2 + y1 x2 ),

(5.2)

5.2. CALCULATING WITH COMPLEX NUMBERS

91

The set C of complex numbers forms a field under the operations of

matrix addition and multiplication. The additive identity is 0, the additive
inverse of x + iy is the complex number (x) + i(y), the multiplicative
identity is 1 and the multiplicative inverse of the nonzero complex number
x + iy is the complex number u + iv, where
x
y
u= 2
and v = 2
.
2
x +y
x + y2
(If x + iy 6= 0, then x 6= 0 or y 6= 0, so x2 + y 2 6= 0.)

From equations 5.1 and 5.2, we observe that addition and multiplication
of complex numbers is performed just as for real numbers, replacing i2 by
1, whenever it occurs.
A useful identity satisfied by complex numbers is
r2 + s2 = (r + is)(r is).
This leads to a method of expressing the ratio of two complex numbers in
the form x + iy, where x and y are real complex numbers.
x1 + iy1
x2 + iy2

=
=

(x1 + iy1 )(x2 iy2 )

(x2 + iy2 )(x2 iy2 )
(x1 x2 + y1 y2 ) + i(x1 y2 + y1 x2 )
.
x22 + y22

The process is known as rationalization of the denominator.

5.2

Calculating with complex numbers

We can now do all the standard linear algebra calculations over the field of
complex numbers find the reduced rowechelon form of an matrix whose elements are complex numbers, solve systems of linear equations, find inverses
and calculate determinants.
For example, solve the system
(1 + i)z + (2 i)w = 2 + 7i

7z + (8 2i)w = 4 9i.

The coefficient determinant is



 1+i 2i 
 = (1 + i)(8 2i) 7(2 i)

 7
8 2i 
= (8 2i) + i(8 2i) 14 + 7i
= 4 + 13i 6= 0.

92

CHAPTER 5. COMPLEX NUMBERS

Hence by Cramers rule, there is a unique solution:



 2 + 7i 2 i 


 4 9i 8 2i 
z =
4 + 13i
(2 + 7i)(8 2i) (4 9i)(2 i)
=
4 + 13i
2(8 2i) + (7i)(8 2i) {(4(2 i) 9i(2 i)}
=
4 + 13i
2
16 4i + 56i 14i {8 4i 18i + 9i2 }
=
4 + 13i
31 + 74i
=
4 + 13i
(31 + 74i)(4 13i)
=
(4)2 + 132
838 699i
=
(4)2 + 132
838 699
=

i
185 185
and similarly w =

698 229
+
i.
185
185

An important property enjoyed by complex numbers is that every complex number has a square root:
THEOREM 5.2.1
If w is a nonzero complex number, then the equation z 2 = w has a solution z C.
Proof. Let w = a + ib, a, b R.

Case1. Suppose b = 0. Then if a > 0, z = a is a solution, while if

a < 0, i a is a solution.
Case 2. Suppose b 6= 0. Let z = x + iy, x, y R. Then the equation
2
z = w becomes
(x + iy)2 = x2 y 2 + 2xyi = a + ib,
so equating real and imaginary parts gives
x2 y 2 = a

and

2xy = b.

5.2. CALCULATING WITH COMPLEX NUMBERS

93

Hence x 6= 0 and y = b/(2x). Consequently

 
b 2
2
x
= a,
2x
so 4x4 4ax2 b2 = 0 and 4(x2 )2 4a(x2 ) b2 = 0. Hence

a a2 + b2
4a 16a2 + 16b2
2
=
.
x =
8
2

However x2 > 0, so we must take the + sign, as a a2 + b2 < 0. Hence

s

2
2
a+ a +b
a + a2 + b2
2
x =
, x=
.
2
2
Then y is determined by y = b/(2x).
EXAMPLE 5.2.1 Solve the equation z 2 = 1 + i.
Solution. Put z = x + iy. Then the equation becomes
(x + iy)2 = x2 y 2 + 2xyi = 1 + i,
so equating real and imaginary parts gives
x2 y 2 = 1 and 2xy = 1.
Hence x 6= 0 and y = 1/(2x). Consequently
 
1 2
2
= 1,
x
2x
so 4x4 4x2 1 = 0. Hence
4

x =
Hence

1+ 2
x =
2
2

Then
y=
Hence the solutions are

16 + 16
1 2
=
.
8
2

and

x=

1+ 2
.
2

1
1
= p
.
2x
2 1+ 2

z =

1+ 2
i
+ p
.
2
2 1+ 2

94

CHAPTER 5. COMPLEX NUMBERS

EXAMPLE 5.2.2 Solve the equation z 2 + ( 3 + i)z + 1 = 0.

Solution. Because every complex number has a square root, the familiar
formula

b b2 4ac
z=
2a
for the solution of the general quadratic equation az 2 + bz + c = 0 can be
used, where now a(6= 0), b, c C. Hence
z =
=

( 3 + i) ( 3 + i)2 4

2
q

( 3 + i) (3 + 2 3i 1) 4

p 2

( 3 + i) 2 + 2 3i
.
=
2

Now we have to solve w2 = 2 + 2 3i. Put w = x + iy. Then w2 =

x2 y 2 + 2xyi = 2 + 2 3i and equating
real and imaginary parts gives
x2 y 2 = 2 and 2xy = 2 3. Hence y = 3/x and so x2 3/x2 = 2. So
2
x4 + 2x2 3= 0 and (x2 + 3)(x
Hence x2 1 = 0 and x = 1.
2 1) = 0.
Then y = 3. Hence (1 + 3i) = 2 + 2 3i and the formula for z now
becomes

3 i (1 + 3i)
z =
2

1 3 (1 + 3)i
1 3 + (1 + 3)i
or
.
=
2
2
EXAMPLE 5.2.3 Find the cube roots of 1.
Solution. We have to solve the equation z 3 = 1, or z 3 1 = 0. Now
z 3 1 = (z 1)(z 2 + z + 1). So z 3 1 = 0 z 1 = 0 or z 2 + z + 1 = 0.
But

1 12 4
1 3i
2
z +z+1=0z =
=
.
2
2

So there are 3 cube roots of 1, namely 1 and (1 3i)/2.

We state the next theorem without proof. It states that every non
constant polynomial with complex number coefficients has a root in the
field of complex numbers.

5.3. GEOMETRIC REPRESENTATION OF C

95

THEOREM 5.2.2 (Gauss) If f (z) = an z n + an1 z n1 + + a1 z + a0 ,

where an 6= 0 and n 1, then f (z) = 0 for some z C.
It follows that in view of the factor theorem, which states that if a F is
a root of a polynomial f (z) with coefficients from a field F , then z a is a
factor of f (z), that is f (z) = (z a)g(z), where the coefficients of g(z) also
belong to F . By repeated application of this result, we can factorize any
polynomial with complex coefficients into a product of linear factors with
complex coefficients:
f (z) = an (z z1 )(z z2 ) (z zn ).
There are available a number of computational algorithms for finding good
approximations to the roots of a polynomial with complex coefficients.

5.3

Geometric representation of C

Complex numbers can be represented as points in the plane, using the correspondence x + iy (x, y). The representation is known as the Argand
diagram or complex plane. The real complex numbers lie on the xaxis,
which is then called the real axis, while the imaginary numbers lie on the
yaxis, which is known as the imaginary axis. The complex numbers with
positive imaginary part lie in the upper half plane, while those with negative
imaginary part lie in the lower half plane.
Because of the equation
(x1 + iy1 ) + (x2 + iy2 ) = (x1 + x2 ) + i(y1 + y2 ),
complex numbers add vectorially, using the parallellogram law. Similarly,
the complex number z1 z2 can be represented by the vector from (x2 , y2 )
to (x1 , y1 ), where z1 = x1 + iy1 and z2 = x2 + iy2 . (See Figure 5.1.)
The geometrical representation of complex numbers can be very useful
when complex number methods are used to investigate properties of triangles
and circles. It is very important in the branch of calculus known as Complex
Function theory, where geometric methods play an important role.
We mention that the line through two distinct points P1 = (x1 , y1 ) and
P2 = (x2 , y2 ) has the form z = (1 t)z1 + tz2 , t R, where z = x + iy is
any point on the line and zi = xi + iyi , i = 1, 2. For the line has parametric
equations
x = (1 t)x1 + tx2 , y = (1 t)y1 + ty2
and these can be combined into a single equation z = (1 t)z1 + tz2 .

96

CHAPTER 5. COMPLEX NUMBERS

z 2 

@
@


@



 z1 + z2


 

@
R


* z1











@
@
@
R z1 z2
@
?

Figure 5.1: Complex addition and subraction.

Circles have various equation representations in terms of complex numbers, as will be seen later.

5.4

Complex conjugate

DEFINITION 5.4.1 (Complex conjugate) If z = x + iy, the complex

conjugate of z is the complex number defined by z = x iy. Geometrically,
the complex conjugate of z is obtained by reflecting z in the real axis (see
Figure 5.2).
The following properties of the complex conjugate are easy to verify:
1. z1 + z2 = z1 + z2 ;
2. z = z.
3. z1 z2 = z1 z2 ;
4. z1 z2 = z1 z2 ;
5. (1/z) = 1/z;
6. (z1 /z2 ) = z1 /z2 ;

97

5.4. COMPLEX CONJUGATE

6


Z


x
Z

>


z
y
-

~z
Z

Figure 5.2: The complex conjugate of z: z.

7. z is real if and only if z = z;

8. With the standard convention that the real and imaginary parts are
denoted by Re z and Im z, we have
Re z =

z+z
,
2

Im z =

zz
;
2i

9. If z = x + iy, then zz = x2 + y 2 .
THEOREM 5.4.1 If f (z) is a polynomial with real coefficients, then its
nonreal roots occur in complexconjugate pairs, i.e. if f (z) = 0, then
f (z) = 0.
Proof. Suppose f (z) = an z n + an1 z n1 + + a1 z + a0 = 0, where
an , . . . , a0 are real. Then
0 = 0 = f (z) = an z n + an1 z n1 + + a1 z + a0

= an z n + an1 z n1 + + a1 z + a0
= an z n + an1 z n1 + + a1 z + a0

= f (z).

EXAMPLE 5.4.1 Discuss the position of the roots of the equation

z 4 = 1
in the complex plane.
Solution. The equation z 4 = 1 has real coefficients and so its roots come
in complex conjugate pairs. Also if z is a root, so is z. Also there are

98

CHAPTER 5. COMPLEX NUMBERS

clearly no real roots and no imaginary roots. So there must be one root w
in the first quadrant, with all remaining roots being given by w, w and
w. In fact, as we shall soon see, the roots lie evenly spaced on the unit
circle.
The following theorem is useful in deciding if a polynomial f (z) has a
multiple root a; that is if (z a)m divides f (z) for some m 2. (The proof
is left as an exercise.)
THEOREM 5.4.2 If f (z) = (z a)m g(z), where m 2 and g(z) is a
polynomial, then f (a) = 0 and the polynomial and its derivative have a
common root.
From theorem 5.4.1 we obtain a result which is very useful in the explicit
integration of rational functions (i.e. ratios of polynomials) with real coefficients.
THEOREM 5.4.3 If f (z) is a nonconstant polynomial with real coefficients, then f (z) can be factorized as a product of real linear factors and
real quadratic factors.
Proof. In general f (z) will have r real roots z1 , . . . , zr and 2s nonreal
roots zr+1 , z r+1 , . . . , zr+s , z r+s , occurring in complexconjugate pairs by
theorem 5.4.1. Then if an is the coefficient of highest degree in f (z), we
have the factorization
f (z) = an (z z1 ) (z zr )

(z zr+1 )(z z r+1 ) (z zr+s )(z z r+s ).

We then use the following identity for j = r + 1, . . . , r + s which in turn

shows that paired terms give rise to real quadratic factors:
(z zj )(z z j ) = z 2 (zj + z j )z + zj z j

= z 2 2Re zj + (x2j + yj2 ),

where zj = xj + iyj .
A wellknown example of such a factorization is the following:
EXAMPLE 5.4.2 Find a factorization of z 4 +1 into real linear and quadratic
factors.

5.5. MODULUS OF A COMPLEX NUMBER

6

99

z
>
|z|
y
x

Figure 5.3: The modulus of z: |z|.

Solution. Clearly there are no real roots. Also we have the preliminary
factorization z 4 + 1 = (z2 i)(z 2 + i). Now the roots of z 2 i are easily

verified to be (1 + i)/ 2, so the roots ofz 2 + i must be (1 i)/ 2.

In other words the roots are w = (1 + i)/ 2 and w, w, w. Grouping
conjugatecomplex terms gives the factorization
z 4 + 1 = (z w)(z w)(z + w)(z + w)

= (z 2 2zRe w + ww)(z 2 + 2zRe w + ww)

= (z 2 2z + 1)(z 2 + 2z + 1).

5.5

Modulus of a complex number

DEFINITION 5.5.1 (Modulus) If z = x +

p iy, the modulus of z is the
nonnegative real number |z| defined by |z| = x2 + y 2 . Geometrically, the
modulus of z is the distance from z to 0 (see Figure 5.3).
More generally, |z1 z2 | is the distance between z1 and z2 in the complex
plane. For
|z1 z2 | = |(x1 + iy1 ) (x2 + iy2 )| = |(x1 x2 ) + i(y1 y2 )|
p
=
(x1 x2 )2 + (y1 y2 )2 .

The following properties of the modulus are easy to verify, using the identity
|z|2 = zz:
(i)

|z1 z2 | = |z1 ||z2 |;

(ii)

|z 1 | = |z|1 ;

100

CHAPTER 5. COMPLEX NUMBERS

 
 z1  |z1 |
 =
 z2  |z2 | .

(iii)

For example, to prove (i):

|z1 z2 |2 = (z1 z2 )z1 z2 = (z1 z2 )z1 z2

= (z1 z1 )(z2 z2 ) = |z1 |2 |z2 |2 = (|z1 ||z2 |)2 .

Hence |z1 z2 | = |z1 ||z2 |.

EXAMPLE 5.5.1 Find |z| when z =
Solution.
|z| =
=
=

(1 + i)4
.
(1 + 6i)(2 7i)

|1 + i|4
|1 + 6i||2 7i|

( 12 + 12 )4
p

12 + 62 22 + (7)2
4
.
37 53

THEOREM 5.5.1 (Ratio formulae) If z lies on the line through z1 and

z2 :
z = (1 t)z1 + tz2 , t R,
we have the useful ratio formulae:




 z z1 
 t 




(i)
 z z2  =  1 t 
(ii)



 z z1 


 z1 z2 

if z 6= z2 ,

|t|.

Circle equations. The equation |z z0 | = r, where z0 C and r >

0, represents the circle centre z0 and radius r. For example the equation
|z (1 + 2i)| = 3 represents the circle (x 1)2 + (y 2)2 = 9.
Another useful circle equation is the circle of Apollonius :


z a


 z b  = ,

where a and b are distinct complex numbers and is a positive real number,
6= 1. (If = 1, the above equation represents the perpendicular bisector
of the segment joining a and b.)

101

5.5. MODULUS OF A COMPLEX NUMBER

y
6

-x

Figure 5.4: Apollonius circles:

|z+2i|
|z2i|

= 14 , 38 , 21 , 58 ; 41 , 38 , 21 , 85 .

An algebraic proof that the above equation represents a circle, runs as

follows. We use the following identities:
(i)
(ii)
(iii)

|z a|2
Re (z1 z2 )
Re (tz)

=
=
=

|z|2 2Re (za) + |a|2

Re z1 Re z2
tRe z if t R.

We have


z a
2
2
2


 z b  = |z a| = |z b|

|z|2 2Re {za} + |a|2 = 2 (|z|2 2Re {zb} + |b|2 )

(1 2 )|z|2 2Re {z(a 2 b)} = 2 |b|2 |a|2

 

2 |b|2 |a|2
a 2 b
2
|z| 2Re z
=
1 2
1 2





 
 a 2 b 2 2 |b|2 |a|2  a 2 b 2
a 2 b
2
 =
 .
+ 
+ 
|z| 2Re z
1 2
1 2 
1 2
1 2 

Now it is easily verified that

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

102

CHAPTER 5. COMPLEX NUMBERS

So we obtain



2

2
2
2


z a

 = z a b  = |a b|

z b
1 2 
|1 2 |2




a 2 b  |a b|
=
.
z
1 2  |1 2 |

The last equation represents a circle centre z0 , radius r, where

z0 =

a 2 b
1 2

and

r=

|a b|
.
|1 2 |

There are two special points on the circle of Apollonius, the points z1 and
z2 defined by
z2 a
z1 a
= and
= ,
z1 b
z2 b
or
a b
a + b
z1 =
and z2 =
.
(5.3)
1
1+
It is easy to verify that z1 and z2 are distinct points on the line through a
2
and b and that z0 = z1 +z
2 . Hence the circle of Apollonius is the circle based
on the segment z1 , z2 as diameter.
EXAMPLE 5.5.2 Find the centre and radius of the circle
|z 1 i| = 2|z 5 2i|.
Solution. Method 1. Proceed algebraically and simplify the equation
|x + iy 1 i| = 2|x + iy 5 2i|
or
|x 1 + i(y 1)| = 2|x 5 + i(y 2)|.
Squaring both sides gives
(x 1)2 + (y 1)2 = 4((x 5)2 + (y 2)2 ),
which reduces to the circle equation
x2 + y 2

38
14
x y + 38 = 0.
3
3

Completing the square gives

7
19
(x )2 + (y )2 =
3
3

19
3

2

 2
7
68
+
38 = ,
3
9

5.6. ARGUMENT OF A COMPLEX NUMBER

6

r




>


103

z
y
-

Figure 5.5: The argument of z: arg z = .

7
( 19
3 , 3)

and the radius is 68

so the centre is
9 .
Method 2. Calculate the diametrical points z1 and z2 defined above by
equations 5.3:
z1 1 i = 2(z1 5 2i)

z2 1 i = 2(z2 5 2i).

We find z1 = 9 + 3i and z2 = (11 + 5i)/3. Hence the centre z0 is given by

z0 =

19 7
z1 + z2
=
+ i
2
3
3

and the radius r is given by


 


 19 7
  8 2 
68



r = |z1 z0 | = 
+ i (9 + 3i) =  i =
.
3
3
3 3
3

5.6

Argument of a complex number

p
Let z = x + iy be a nonzero complex number, r = |z| = x2 + y 2 . Then
we have x = r cos , y = r sin , where is the angle made by z with the
positive xaxis. So is unique up to addition of a multiple of 2 radians.
DEFINITION 5.6.1 (Argument) Any number satisfying the above
pair of equations is called an argument of z and is denoted by arg z. The
particular argument of z lying in the range < is called the principal
argument of z and is denoted by Arg z (see Figure 5.5).
We have z = r cos + ir sin = r(cos + i sin ) and this representation
of z is called the polar representation or modulusargument form of z.

104

CHAPTER 5. COMPLEX NUMBERS

EXAMPLE 5.6.1 Arg 1 = 0, Arg (1) = , Arg i = 2 , Arg (i) = 2 .

We note that y/x = tan if x 6= 0, so is determined by this equation up
to a multiple of . In fact
Arg z = tan1

y
+ k,
x

where k = 0 if x > 0; k = 1 if x < 0, y > 0; k = 1 if x < 0, y < 0.

To determine Arg z graphically, it is simplest to draw the triangle formed
by the points 0, x, z on the complex plane, mark in the positive acute angle
between the rays 0, x and 0, z and determine Arg z geometrically, using
the fact that = tan1 (|y|/|x|), as in the following examples:
EXAMPLE 5.6.2 Determine the principal argument of z for the followig
complex numbers:
z = 4 + 3i, 4 + 3i, 4 3i, 4 3i.
Solution. Referring to Figure 5.6, we see that Arg z has the values
, , + , ,
where = tan1 34 .
An important property of the argument of a complex number states that
the sum of the arguments of two nonzero complex numbers is an argument
of their product:
THEOREM 5.6.1 If 1 and 2 are arguments of z1 and z2 , then 1 + 2
is an argument of z1 z2 .
Proof. Let z1 and z2 have polar representations z1 = r1 (cos 1 + i sin 1 )
and z2 = r2 (cos 2 + i sin 2 ). Then
z1 z2 = r1 (cos 1 + i sin 1 )r2 (cos 2 + i sin 2 )
= r1 r2 (cos 1 cos 2 sin 1 sin 2 + i(cos 1 sin 2 + sin 1 cos 2 ))
= r1 r2 (cos (1 + 2 ) + i sin (1 + 2 )),

which is the polar representation of z1 z2 , as r1 r2 = |z1 ||z2 | = |z1 z2 |. Hence

1 + 2 is an argument of z1 z2 .
An easy induction gives the following generalization to a product of n
complex numbers:

5.6. ARGUMENT OF
y A COMPLEX NUMBER
6





>


-x

4 + 3i
}
Z

-x

-x

4 3i

=


4 + 3i

105

Z
Z
?

-x
Z

~ 4 3i
Z

Figure 5.6: Argument examples.

COROLLARY 5.6.1 If 1 , . . . , n are arguments for z1 , . . . , zn respectively,

then 1 + + n is an argument for z1 zn .
Taking 1 = = n = in the previous corollary gives
COROLLARY 5.6.2 If is an argument of z, then n is an argument for
zn.
THEOREM 5.6.2 If is an argument of the nonzero complex number
z, then is an argument of z 1 .
Proof. Let be an argument of z. Then z = r(cos +i sin ), where r = |z|.
Hence
z 1 = r1 (cos + i sin )1
= r1 (cos i sin )

= r1 (cos() + i sin()).
Now r1 = |z|1 = |z 1 |, so is an argument of z 1 .
COROLLARY 5.6.3 If 1 and 2 are arguments of z1 and z2 , then 1 2
is an argument of z1 /z2 .

106

CHAPTER 5. COMPLEX NUMBERS

In terms of principal arguments, we have the following equations:

(i)
Arg (z1 z2 ) =
(ii)
Arg (z 1 ) =
(iii)
Arg (z1 /z2 ) =
(iv) Arg (z1 zn ) =
(v)
Arg (z n ) =
where k1 , k2 , k3 , k4 , k5 are

Arg z1 +Arg z2 + 2k1 ,

Arg z + 2k2 ,
Arg z1 Arg z2 + 2k3 ,
Arg z1 + +Arg zn + 2k4 ,
n Arg z + 2k5 ,
integers.

In numerical examples, we can write (i), for example, as

Arg (z1 z2 ) Arg z1 + Arg z2 .
EXAMPLE 5.6.3 Find the modulus and principal argument of
!17

3+i
z=
1+i
and hence express z in modulusargument form.

| 3 + i|17
217

Solution. |z| =
=
= 217/2 .
|1 + i|17
( 2)17
!

3+i
Arg z 17Arg
1+i

= 17(Arg ( 3 + i) Arg (1 + i))

  17
= 17
=

.
6
4
12


+ 2k, where k is an integer. We see that k = 1 and
Hence Arg z = 17
12


7
7
hence Arg z = 12 . Consequently z = 217/2 cos 7
+
i
sin
12
12 .
DEFINITION 5.6.2 If is a real number, then we define ei by
ei = cos + i sin .
More generally, if z = x + iy, then we define ez by
ez = ex eiy .
For example,
i

e 2 = i, ei = 1, e 2 = i.
The following properties of the complex exponential function are left as
exercises:

107

5.7. DE MOIVRES THEOREM

THEOREM 5.6.3

(i)
(ii)
(iii)
(iv)
(v)
(vi)

ez1 ez2
ezn
ez
(ez )1
ez1 /ez2
ez

ez1

=
=
6=
=
=
=

ez1 +z2 ,
ez1 ++zn ,
0,
ez ,
ez1 z2 ,
ez .

THEOREM 5.6.4 The equation

ez = 1
has the complete solution z = 2ki, k Z.
Proof. First we observe that
e2ki = cos (2k) + i sin (2k) = 1.
Conversely, suppose ez = 1, z = x + iy. Then ex (cos y + i sin y) = 1. Hence
ex cos y = 1 and ex sin y = 0. Hence sin y = 0 and so y = n, n Z. Then
ex cos (n) = 1, so ex (1)n = 1, from which follows (1)n = 1 as ex > 0.
Hence n = 2k, k Z and ex = 1. Hence x = 0 and z = 2ki.

5.7

De Moivres theorem

The next theorem has many uses and is a special case of theorem 5.6.3(ii).
Alternatively it can be proved directly by induction on n.
THEOREM 5.7.1 (De Moivre) If n is a positive integer, then
(cos + i sin )n = cos n + i sin n.
As a first application, we consider the equation z n = 1.
THEOREM 5.7.2 The equation z n = 1 has n distinct solutions, namely
2ki
the complex numbers k = e n , k = 0, 1, . . . , n 1. These lie equally
spaced on the unit circle |z| = 1 and are obtained by starting at 1, moving
round the circle anticlockwise, incrementing the argument in steps of 2
n .
(See Figure 5.7)
2i
We notice that the roots are the powers of the special root = e n .

108

CHAPTER 5. COMPLEX NUMBERS

6

|z| = 1
7





* 1

2/n
 
2/n


H
0
HH2/n
HH
H
HH
j n1

Figure 5.7: The nth roots of unity.

Proof. With k defined as above,

 2ki n
2ki
kn = e n
= e n n = 1,

by De Moivres theorem. However |k | = 1 and arg k = 2k

n , so the complex numbers k , k = 0, 1, . . . , n 1, lie equally spaced on the unit circle.
Consequently these numbers must be precisely all the roots of z n 1. For
the polynomial z n 1, being of degree n over a field, can have at most n
distinct roots in that field.
The more general equation z n = a, where a , C, a 6= 0, can be reduced
to the previous case:
i
Let be argument of z, so that a = |a|ei . Then if w = |a|1/n e n , we
have


i n
wn =
|a|1/n e n
 i n
= (|a|1/n )n e n
= |a|ei = a.

So w is a particular solution. Substituting for a in the original equation,

we get z n = wn , or (z/w)n = 1. Hence the complete solution is z/w =

109

5.7. DE MOIVRES THEOREM

6



z1

|z| = (|a|)1/n




2/n





P
PP
PP
PP

* z0

PP
q zn1

Figure 5.8: The roots of z n = a.

2ki
n

, k = 0, 1, . . . , n 1, or
i

zk = |a|1/n e n e

2ki
n

= |a|1/n e

i(+2k)
n

(5.4)

k = 0, 1, . . . , n 1. So the roots are equally spaced on the circle

|z| = |a|1/n
and are generated from the special solution having argument equal to (arg a)/n,
by incrementing the argument in steps of 2/n. (See Figure 5.8.)
EXAMPLE 5.7.1 Factorize the polynomial z 5 1 as a product of real
linear and quadratic factors.
2i

2i

4i

4i

Solution. The roots are 1, e 5 , e 5 , e 5 , e 5 , using the fact that non

real roots come in conjugatecomplex pairs. Hence
z 5 1 = (z 1)(z e

2i
5

)(z e

2i
5

)(z e

4i
5

)(z e

4i
5 ).

Now
(z e

2i
5

)(z e

2i
5

) = z 2 z(e
2

2i
5

= z 2z cos

+e
2
5

2i
5

+ 1.

)+1

110

CHAPTER 5. COMPLEX NUMBERS

Similarly
(z e

4i
5

)(z e

4i
5

) = z 2 2z cos

4
5

+ 1.

This gives the desired factorization.

EXAMPLE 5.7.2 Solve z 3 = i.
Solution. |i| = 1 and Arg i =

= . So by equation 5.4, the solutions are

zk = |i|1/3 e

i(+2k)
3

, k = 0, 1, 2.

First, k = 0 gives
z0 = e

i
6

i
= cos + i sin =
+ .
6
6
2
2

Next, k = 1 gives
z1 = e

5i
6

5
5
3
i
= cos
+ i sin
=
+ .
6
6
2
2

Finally, k = 2 gives
z1 = e

9i
6

= cos

9
9
+ i sin
= i.
6
6

We finish this chapter with two more examples of De Moivres theorem.

EXAMPLE 5.7.3 If
C = 1 + cos + + cos (n 1),
S = sin + + sin (n 1),

prove that
C=
if 6= 2k, k Z.

n
2
sin 2

sin

cos

(n1)
2

and S =

n
2
sin 2

sin

sin

(n1)
,
2

111

5.8. PROBLEMS
Solution.

C + iS = 1 + (cos + i sin ) + + (cos (n 1) + i sin (n 1))

= 1 + ei + + ei(n1)

= 1 + z + + z n1 , where z = ei
1 zn
, if z 6= 1, i.e. 6= 2k,
=
1z
in

in

1 ein
e 2 (e 2 e 2 )
=
i
i
i
1 ei
e 2 (e 2 e 2 )

= ei(n1) 2

in

n
2
sin 2

sin

= (cos (n 1) 2 + i sin (n 1) 2 )

n
2
sin 2

sin

The result follows by equating real and imaginary parts.

EXAMPLE 5.7.4 Express cos n and sin n in terms of cos and sin ,
using the equation cos n + sin n = (cos + i sin )n .
Solution. The binomial theorem gives


(cos + i sin )n = cosn + n1 cosn1 (i sin ) +
+ (i sin )n .

n
2

cosn2 (i sin )2 +

Equating real and imaginary parts gives

cos n = cosn n2 cosn2 sin2 +




sin n = n1 cosn1 sin n3 cosn3 sin3 + .

5.8

PROBLEMS

1. Express the following complex numbers in the form x + iy, x, y real:

(i) (3 + i)(14 2i); (ii)

2 + 3i
(1 + 2i)2
; (iii)
.
1 4i
1i

[Answers: (i) 40 + 20i; (ii) 10

17 +
2. Solve the following equations:

11
17 i;

(iii) 72 + 2i .]

112

CHAPTER 5. COMPLEX NUMBERS

(i)

iz + (2 10i)z

3z + 2i,

(ii)

(1 + i)z + (2 i)w
(1 + 2i)z + (3 + i)w

=
=

3i
2 + 2i.

9
[Answers:(i) z = 41

i
41 ;

(ii) z = 1 + 5i, w =

19
5

8i
5 .]

3. Express 1 + (1 + i) + (1 + i)2 + . . . + (1 + i)99 in the form x + iy, x, y

real. [Answer: (1 + 250 )i.]
4. Solve the equations: (i) z 2 = 8 6i; (ii) z 2 (3 + i)z + 4 + 3i = 0.
[Answers: (i) z = (1 3i);

(ii) z = 2 i, 1 + 2i.]

5. Find the modulus and principal argument of each of the following

complex numbers:

(i) 4 + i; (ii) 23 2i ; (iii) 1 + 2i; (iv) 12 (1 + i 3).

[Answers: (i)
tan1 2.]

17, tan1 41 ; (ii)

10
2 ,

+ tan1 31 ; (iii)

5,

6. Express the following complex numbers in modulus-argument form:

(i) z = (1 + i)(1 + i 3)( 3 i).

(1 + i)5 (1 i 3)5

.
(ii) z =
( 3 + i)4
[Answers:

(i) z = 4 2(cos
7.

5
12

+ i sin

5
12 );

(ii) z = 27/2 (cos

(i) If z = 2(cos 4 +i sin 4 ) and w = 3(cos

form of
5
(a) zw; (b) wz ; (c) wz ; (d) wz 2 .

11
12

+i sin

+ i sin

6 ),

11
12 ).]

find the polar

(ii) Express the following complex numbers in the form x + iy:

 6

(a) (1 + i)12 ; (b) 1i

.
2

[Answers: (i): (a) 6(cos

5
12

(c) 32 (cos 12
+ i sin 12
);

(ii): (a) 64;

(b) i.]

+ i sin
(d)

5
12 );

(b) 32 (cos

32
11
9 (cos 12

+ i sin

12

+ i sin

11
12 );

12 );

113

5.8. PROBLEMS
8. Solve the equations:

(i) z 2 = 1 + i 3; (ii) z 4 = i; (iii) z 3 = 8i; (iv) z 4 = 2 2i.

[Answers: (i) z = ( 3+i)

; (ii) ik (cos 8 + i sin 8 ), k = 0, 1, 2, 3; (iii)
2

z = 2i, 3 i, 3 i; (iv) z = ik 2 8 (cos 16

i sin 16
), k = 0, 1, 2, 3.]
9. Find the reduced rowechelon

2+i
1+i
1 + 2i

1 i 0
[Answer: 0 0 1 .]
0 0 0

10.

form of the complex matrix

1 + 2i
2
1 + i
1 .
2 + i 1 + i

(i) Prove that the line equation lx + my = n is equivalent to

pz + pz = 2n,
where p = l + im.

(ii) Use (ii) to deduce that reflection in the straight line

pz + pz = n
is described by the equation
pw + pz = n.
[Hint: The complex number l + im is perpendicular to the given
line.]
(iii) Prove that the line |z a| = |z b| may be written as pz +pz = n,
where p = b a and n = |b|2 |a|2 . Deduce that if z lies on the
Apollonius circle |za|
|zb| = , then w, the reflection of z in the line
|z a| = |z b|, lies on the Apollonius circle

|za|
|zb|

= 1 .

11. Let a and b be distinct complex numbers and 0 < < .

(i) Prove that each of the following sets in the complex plane represents a circular arc and sketch the circular arcs on the same
diagram:

114

CHAPTER 5. COMPLEX NUMBERS

za
= , , , .
zb
za
= represents the line segment joining
Also show that Arg
zb
za
= 0 represents the remaining portion of
a and b, while Arg
zb
the line through a and b.
Arg

(ii) Use (i) to prove that four distinct points z1 , z2 , z3 , z4 are concyclic or collinear, if and only if the crossratio
z4 z1 z3 z1
/
z4 z2 z3 z2
is real.
(iii) Use (ii) to derive Ptolemys Theorem: Four distinct points A, B, C, D
are concyclic or collinear, if and only if one of the following holds:
AB CD + BC AD = AC BD

BD AC + AD BC = AB CD

BD AC + AB CD = AD BC.