Chapter 2
Complex numbers
2.1 Introduction
The mathematician Paul Painlevé once wrote that
[i]t came to appear that, between two truths of the real domain, the easiest and
shortest path quite often passes through the complex domain.
The theory of complex numbers, and in particular the theory of complex analysis that follows
from it, is simultaneously one of the most beautiful and applicable areas of mathematics. The
apparently purely theoretical “game” of appending a “number” i satisfying i2 = 1 to the real
numbers results in a rich theory that provides many tools used by engineers and scientists to
find real answers to real problems much quicker than thinking in terms of real numbers. In
this module you will only glimpse this. In Methods 3 (Complex Analysis) you will begin to
appreciate the full power.
Let us imagine that we can play the game described above, namely we can include a number
i, where i2 = 1, with the real numbers. Without worrying about rigour or what this actually
means, let us proceed to calculate s few things assuming that basic algebraic manipulation of
such “complex numbers” work the same as for real numbers. I could write down any polynomial
in i with real coefficients, however, given that i2 = 1, i3 = i, i4 = 1, etc., it follows that any
such expression can be written as z = x + iy, where x and y are real. We call x = Re(z) the real
part of z and y = Im(z) imaginary part of z. The complex conjugate of z is z̄ = pz iy. This is
sometimes also denoted as z ⇤ . The length of modulus of z is denoted by |z| = x2 + y 2 . Note
that z z̄ = (x + iy)(x iy) = x2 + y 2 = |z|2 .
Continuing with our game, we can write
(x1 + iy1 ) + (x2 + iy2 ) = (x1 + x2 ) + i(y1 + y2 ).
(x1 + iy1 )(x2 + iy2 ) = x1 y1 + ix1 y2 + ix2 y1 + i2 x2 y2 = (x1 y1 x2 y2 ) + i(x1 y2 + x2 y1 ).
1 1 1 x iy x y
(x + iy) = = = 2 i .
x + iy x + iy x iy x + y2 x2 + y2
But what does all this mean? What is the “number” i? Perhaps the easiest way to approach
this is to identify the “complex number” z = x + iy with the point (x, y) in the plane R2 . Then
i is just the point (0, 1). The statements above can now be interpreted as definitions of the sum,
product and ratio of two complex numbers. In this way there is a copy of the real numbers
(namely the points of the form (x, 0) living inside the complex numbers. So we can stop being
worried about whether complex numbers exist, at least so long as you’re not concerned about
the existence of pairs of real numbers.
24
CHAPTER 2. COMPLEX NUMBERS 25
2.2 Polar coordinates and the exponential
Thinking of complex numbers as points in the plane suggests that other coordinate systems
might be useful, in particular, polar coordinates x = r cos ✓, y = r sin ✓ (r 0) are useful in
many applications. Using polar coordinates, any complex number z can be written as
z = x + iy = r(cos ✓ + i sin ✓).
This polar representation of complex numbers is especially useful when multiplying complex
numbers. Let z1 = r1 (cos ✓1 + i sin ✓1 ) and z2 = r2 (cos ✓2 + i sin ✓2 ). Then
z1 z2 = r1 r2 {(cos ✓1 cos ✓2 sin ✓1 sin ✓2 ) + i(cos ✓1 sin ✓2 + sin ✓1 cos ✓2 )} .
Using the addition formulas for cos and sin we obtain
z1 z2 = r1 r2 {cos(✓1 + ✓2 ) + i sin(✓1 + ✓2 )}.
So when we multiply two complex numbers the modulus of the product is the product of the
moduli and the argument of the product is the sum of the arguments.
In particular it follows (on setting r1 = r2 = 1 and ✓1 = ✓2 = ✓ that
(cos ✓ + i sin ✓)2 = (cos ✓ + i sin ✓)(cos ✓ + i sin ✓) = cos 2✓ + i sin 2✓
and
(cos ✓+i sin ✓)3 = (cos ✓+i sin ✓)(cos ✓+i sin ✓)2 = (cos ✓+i sin ✓)(cos 2✓+i sin 2✓) = cos 3✓+i sin 3✓.
It follows by induction that for any positive integer n,
(cos ✓ + i sin ✓)n = cos n✓ + i sin n✓. (2.1)
This result is also true for n = 0. Finally, taking the reciprocal, we have
n 1 cos n✓ i sin n✓
(cos ✓+i sin ✓) = = = cos n✓ i sin n✓ = cos( n)✓+i sin( n)✓.
cos n✓ + i sin n✓ cos2 n✓ + sin2 n✓
This shows that equation (2.1) is also true for negative integer n. Equation (2.1) is known as
de Moivre’s theorem.
Example 2.1 Express sin 5✓/ sin ✓ as a polynomial in cos ✓.
Soln: From de Moivre’s theorem we have
cos 5✓ + i sin 5✓ = (cos ✓ + i sin ✓)5
= (cos ✓)5 + 5(cos ✓)4 (i sin ✓)1 + 10(cos ✓)3 (i sin ✓)2 + 10(cos ✓)2 (i sin ✓)3
+ 5(cos ✓)1 (i sin ✓)4 + (i sin ✓)5
= (cos ✓)5 + 5i(cos ✓)4 (sin ✓) + 10i2 (cos ✓)3 (sin ✓)2 + 10i3 (cos ✓)2 (sin ✓)3
+ 5i4 (cos ✓)(sin ✓)4 + i5 (sin ✓)5 .
So using the fact that i2 = 1, i3 = i2 i = i, i4 = (i2 )2 = ( 1)2 = 1 and i5 = i4 i = i, we have
cos 5✓ + i sin 5✓
= (cos ✓)5 10(cos ✓)3 (sin ✓)2 + 5(cos ✓)(sin ✓)4 + i 5(cos ✓)4 (sin ✓) 10(cos ✓)2 (sin ✓)3 + (sin ✓)5 .
CHAPTER 2. COMPLEX NUMBERS 26
Taking the imaginary part of both sides we obtain
sin 5✓ = 5(cos ✓)4 (sin ✓) 10(cos ✓)2 (sin ✓)3 + (sin ✓)5 ,
so
sin 5✓
= 5(cos ✓)4 10(cos ✓)2 (sin ✓)2 + (sin ✓)4
sin ✓
2
= 5(cos ✓)4 10(cos ✓)2 (sin ✓)2 + (sin ✓)2
2
= 5(cos ✓)4 10(cos ✓)2 1 (cos ✓)2 + 1 (cos ✓)2
= 5(cos ✓)4 10(cos ✓)2 1 (cos ✓)2 + 1 (2 cos ✓)2 + cos ✓)4
Simplifying the right side gives
sin 5✓
= 16(cos ✓)4 12(cos ✓)2 + 1.
sin ✓
2.2.1 The complex exponential and Euler’s formula
Euler’s formula:
ei✓ = cos ✓ + i sin ✓. (2.2)
Changing ✓ to ✓ in Euler’s formula (2.2) and using the fact that cos( ✓) = cos ✓ and
sin( ✓) = sin ✓, we have
e i✓ = cos ✓ i sin ✓. (2.3)
Adding equations (2.2) and (2.3) and dividing by two gives
1 ⇣ i✓ i✓
⌘
cos ✓ = e +e .
2
Similarly, (equation(2.2) equation(2.3))/(2i) gives
1 ⇣ i✓ i✓
⌘
sin ✓ = e e .
2i
1 3 15 5
Example 2.2 Show that (cos ✓)6 = cos 6✓ + cos 4✓ + cos 2✓ + . Hence evaluate
Z 32 16 32 16
(cos ✓)6 d✓.
Let z = ei✓ . Then cos ✓ = (ei✓ + e i✓ )/2 = (z + z 1 )/2. Therefore
1 6
(cos ✓)6 = 2 6
z+z
6
=2 z 6 + 6z 5 (z 1
) + 15z 4 (z 1 2
) + 20z 3 (z 1 3
) + 15z 2 (z 1 4
) + 6z(z 1 5
) + (z 1 6
)
6 6 4 2 2 4 6
=2 z + 6z + 15z + 20 + 15z + 6z + z
✓ ◆ ✓ ◆ ✓ ◆
1 z6 + z 6 3 z4 + z 4 15 z 2 + z 2 5
= 5 + 4 + 5 + 4
2 2 2 2 2 2 2
✓ 6i✓ 6i✓
◆ ✓ 4i✓ 4i✓
◆ ✓ 2i✓ ◆
1 e +e 3 e +e 15 e + e 2i✓ 5
= + + +
32 2 16 2 32 2 16
1 3 15 5
= cos 6✓ + cos 4✓ + cos 2✓ + .
32 16 32 16
Integrate this expression to answer the second part of the question.
CHAPTER 2. COMPLEX NUMBERS 27
2.3 Roots of complex numbers
Example 2.3 Find the fourth roots of 4.
Above and beyond: The square root function and “branching” in the complex domain
Many real functions can be extended to the complex plane in
a way that maintains continuity and various special algebraic
identities, e.g., polynomials, exponentials, trigonometric and
p
hyperbolic functions. However, some functions, such as z are
more problematic. At each non-zero complex number z = rei✓ ,
p p p i✓/2
we have two choices for z, namely rei✓/2 and re . For
rp = 1 and ✓ = 0, it would seem natural to make the choice
1 = +1. If we try to extend this definition in a continuous
p p
way (so that values don’t suddenlyp jump) we are left with rei✓ = rei✓/2 . However, if we
set r = 1, ✓ = 2⇡, we now have 1 = 1, contradicting our initial choice. This phenomenon
p
is called branching. It turns out that we cannot extend z to a continuous function on the
complex plane C. See Extension sheet 5 for ways to get around this.
Many important functions of mathematical physics are defined through their good properties
in the complex plane. Sofia Kowalevskaya used the absence of branching of solutions of the
di↵erential equations describing spinning tops to find a new solvable case. (See the Above and
Beyond box in this chapter.)
2.4 The geometry of complex numbers
Example 2.4 Find the set of all z 2 C such that
z 1
= a.
z+3
Clearly the set is empty unless a is real and non-negative. If a = 0 then the set is {1}. We are
left with the case a > 0. Setting z = x + iy, we have
|(x 1) + iy| = a|(x + 3) + iy|.
So
(x 1)2 + y 2 = a2 {(x + 3)2 + y 2 }.
Rearranging, we have
(1 a2 )(x2 + y 2 ) 2(1 + 3a2 )x = 9a2 1.
If a = 1 then we have the straight line x = 1. If a 6= 1 then we have
1 + 3a2 9a2 1
x2 + y 2 2 x = .
1 a2 1 a2
So ✓ ◆2 ✓ ◆2
1 + 3a2 29a2 1 1 + 3a2 16a2
x +y = + = .
1 a2 1 a2 1 a2 (1 a2 )2
CHAPTER 2. COMPLEX NUMBERS 28
So the set is a circle of radius 4a/|1 a2 | centred at (1 + 3a2 )/(1 a2 ).
Above and beyond: Sofia Kowalevskaya (1850–1891)
Sofia Kowalevskaya (many variants of spelling are in existence)
was a Russian mathematician. She was the first woman to
have a professional university career in the way that we un-
derstand it today and the first woman professor of Mathemat-
ics in northern Europe (she became a professor in Sweden).
Despite many obstacles, she did important work on partial
di↵erential equations and the applications of complex analy-
sis to mechanics. She won the prestigious Prix Bordin for her
work in which she found a new completely solvable case of the
equations of motion for a spinning top. She found the new
case by studying the behaviour of the solutions in the com-
plex plane (making sure there was no branching as described
in the Above and Beyond box on the square root function).
The functions she used to express the final solution (theta functions) also naturally live in the
complex plane and emerge from complex analysis. She published her papers using the male
version of her name: Kowalevski. She needed to marry in order to travel and faced much
opposition.
The historian of mathematics, Roger Cooke, writes
To venture, as she did, into academia, a world almost no woman had yet explored, and to be
consequently the object of curious scrutiny, while a doubting society looked on, half-expecting her
to fail, took tremendous courage and determination. To achieve, as she did, at least two major
results of lasting value to scholarship, is evidence of a considerable talent, developed through iron
discipline.
References:
Michèle Audin, Remembering Sofia Kovalevskaya, Springer-Verlag London 2011
Roger Cooke, The life of S. V. Kovalevskaya in The Kowalevski property (Leeds, 2000), 1–19,
CRM Proc. Lecture Notes, 32, Amer. Math. Soc., Providence, RI, 2002.
Ann Hibner Koblitz, A convergence of lives. Sofia Kovalevskaia: scientist, writer, revolutionary.
Birkhäuser Boston, MA, 1983