Chapter 2

Complex Numbers

Complex numbers play a significant role in mathematics, physics, engineering and many other disciplines. We
shall study both the algebraic and geometric properties of complex numbers in this chapter.

2.1 Revision on Complex Numbers

We shall review basics of complex numbers in this section. Advanced topics would be delivered in the remaining
sections.

Goals
By completing this section, students should be able to

• recall what complex numbers are, including their real parts and imaginary parts;
• recall and perform addition, subtraction, multiplication of complex numbers;
• recall and perform division of complex numbers with the help of complex conjugation;

• express real part and imaginary part of a complex number with the help of complex conju-
gation;
• recall formulas relating complex conjugation and complex arithmetic.

√
A number of the form z = a + b −1, where a and b are reals, is called a complex number. The reals a and b are
called the real part and the imaginary part of z respectively. We write also R(z) = a and I(z) = b.1
√
If b = 0, then z = a is indeed a real number. Meanwhile, if a = 0, then z = b −1 is a purely imaginary number.
The set of all complex numbers is denoted by C.

Question
Why complex numbers?

We state without proof the following.

Theorem 2.1.1 (Fundamental Theorem of Algebra). Every polynomial over C has at least one (complex) zero.
√
Let’s recall the arithmetic of complex numbers. Given complex numbers zi = ai + bi −1, i = 1, 2. Their sum,
difference and product are defined respectively by
√
z1 ± z2 = (a1 ± a2 ) + (b1 ± b2 ) −1
√
z1 z2 = (a1 a2 − b1 b2 ) + (a1 b2 + a2 b1 ) −1.
1 You may use Re(z) and Im(z) in handwriting.

11
 CHAPTER 2. COMPLEX NUMBERS

Example 2.1.2. Simplify the following

√ √ √
1. (3 + 2 −1) + (−1 + −1) Ans 2 + 3 −1
√ √ √
2. (1 − −1)(−3 + 2 −1) Ans −1 + 5 −1

It may be natural to ‘define’ the quotient of two complex numbers by the formula
√
z1 a1 + b1 −1 √
= √ provided that a2 + b2 −1 ≠ 0.
z2 a2 + b2 −1
However, what are the real part and the imaginary part of this expression?
These numbers can be obtained via a ‘rationalization’ of the denominator of this expression.
√
Definition 2.1.3. Let z = a + b −1. Its complex conjugate is the complex number
√
z ∶= a − b −1.

In other words, R(z) = R(z) while I(z) = −I(z).

√
If z2 = a2 + b2 −1 ≠ 0, the quotient of z1 and z2 is defined by
√ √ √
a1 + b1 −1 a1 + b1 −1 a2 − b2 −1
√ = √ × √
a2 + b2 −1 a2 + b2 −1 a2 − b2 −1
a1 a2 + b1 b2 b1 a2 − a1 b2 √
= + −1
a22 + b22 a22 + b22
√
1 + −1
Example 2.1.4. Find the real part and imaginary part of z = √ . Ans. R(z) = −1/13, I(z) = 5/13.
2 − 3 −1
Complex conjugation behaves well with respect to arithmetic operations of complex numbers.
Theorem 2.1.5 (Properties of Complex Conjugation). Let z, z1 , z2 ∈ C. Then

z=z
z1 ± z2 = z1 ± z2
z1 ⋅ z2 = z1 ⋅ z2
z1 /z2 = z1 /z2 provided that z2 ≠ 0.

This theorem can be proved by direct verification.

To close this section, note that the real part and the imaginary part of a complex number z can be represented in
terms of z itself and z:
1 1 √
R(z) = (z + z) and I(z) = − (z − z) −1.
2 2

2.2 Geometric Representation of Complex Numbers

Question
Recall that the set of all real numbers R can be identified with a (straight) line in the sense that
every real number is identified with a point on the line.
Indeed, addition and subtraction of any two real numbers can be interpreted geometrically; and
we can define the magnitude of any real number. Can you recall how they are defined?
Is there any similar identification available in the set of all complex numbers C?

12
MATH1013/LN/2019 Sem 1 2.2. GEOMETRIC REPRESENTATION OF COMPLEX NUMBERS

Goals
By completing this section, students should be able to

• identify every complex number with a point in a plane, and compute its magnitude;
• recall and apply basic properties of modulus;
• solve the graphs of some simple complex equations involving modulus and argument;
• convert between standard(Cartesian) form and polar form of complex numbers;

• interpret geometrically the addition/subtraction/multiplication/division of two complex

numbers;
• understand and apply de Moivre’s theorem, in particular, determine n–th roots of complex
numbers;

• recall the definition of complex exponents.

The set of all complex numbers

√ C can be represented geometrically by the Cartesian plane. To be more precise,
complex number z = a + b −1 can be identified with point P (a, b) in the Cartesian plane.
This plane is thus called the complex plane, the Argand plane or the Argand diagram.
In particular, we identify horizontal x–axis as R:

{(x, y) ∈ C ∶ y = 0} ↔ R

while the vertical y–axis is identified with the set of all purely imaginary numbers:
√
{(x, y) ∈ C ∶ x = 0} ↔ {y −1 ∶ y ∈ R} .

√ √
Example 2.2.1. Given complex numbers z1 = 2 + 3 −1, z2 = −4 + −1.
1. Mark z1 and z2 in the Argand plane.
2. Compute z1 + z2 and mark the sum in the the diagram.
3. Can you conjecture a geometric relation between these
points and 0 (the origin)?
You may explore with the aid of GeoGebra by clicking
the figure on the right. You may enter z1=2+3i into the
input bar and the corresponding point will appear in the
graph. You may drag the point and move around. The
origin may be defined by O=0+0i.

2.2.1 Modulus
√
Definition 2.2.2. The modulus of a complex number z = a + b −1 is defined by
√ √
∣z∣ ∶= zz = a2 + b2 .

Question
Modulus of a complex number measures its magnitude. Can you give a geometric interpretation
of ∣z∣? It is indeed an extension of a notion for real numbers, can you recognize it?

Theorem 2.2.1 (Properties of Modulus of Complex Number). For any z, z1 , z2 ∈ C,

1. ∣z∣ = ∣z∣;

13
 CHAPTER 2. COMPLEX NUMBERS

2. −∣z∣ ≤ R(z) ≤ ∣z∣;

3. −∣z∣ ≤ I(z) ≤ ∣z∣;

4. ∣z1 z2 ∣ = ∣z1 ∣∣z2 ∣, and ∣z1 /z2 ∣ = ∣z1 ∣/∣z2 ∣ provided that z2 ≠ 0;

5. (triangle inequality) ∣z1 + z2 ∣ ≤ ∣z1 ∣ + ∣z2 ∣;

6. ∣∣z1 ∣ − ∣z2 ∣∣ ≤ ∣z1 − z2 ∣.

Proof. The first three have quite transparent geometric reasoning. The second part of the fourth one comes from
the first part. While one can verify the first part of the fourth one by direct calculation, there is a shortcut with the
help of complex conjugation:

∣z1 z2 ∣2 = (z1 z2 )(z1 z2 ) = (z1 z1 )(z2 z2 ) = ∣z1 ∣2 ∣z2 ∣2 .

A similar trick would establish the triangle inequality and the last one is a consequence of this inequality. Their
verifications are left as exercise.

2.2.2 Complex Equations and their Graphs

Given a complex equation F (z) = 0, where z ∈ C. Its graph is simply the following subset in the Argand plane:

{z ∈ C ∶ F (z) = 0}

Example 2.2.3. Find the graph of the equation R(z) = 3.

√
Solution. Let z = x + y −1. Then R(z) = 0 if and only if x = 3. That is, the graph of this equation is the vertical
line x = 3 in the Argand plane.

Example 2.2.4. Find the graph of the equation ∣z − 1∣ = 2 in the Argand plane.

Solution. Note that

∣z − 1∣ = 2 ⇐⇒ (z − 1)(z − 1) = 22 .
√
Thus, if z = x + −1y, then

∣z∣2 − (z + z) + 1 = 4
x2 + y 2 − 2x + 1 = 22
(x − 1)2 + y 2 = 22

It is the circle centered at 1 with radius 2 in the Argand plane.

2.2.3 Polar Form of Complex Numbers

In this subsection, we are going to introduce an alternative representation of complex√
number. This representation
is very useful in some situations which would be demonstrated shortly. Let z = a + b −1 ≠ 0 and it’s represented
by the point P (a, b) in the Argand plane. It defines an angle at standard position if the ray from the origin O
passing through P is taken as the terminal side and the rotation is counterclockwise.
Magnitude of the angle (in [0, 2π)) can be given by the formula
⎧
⎪
⎪ arctan ab if a > 0, b ≥ 0
⎪
⎪
⎪
⎪
⎪ π/2 if a = 0 and b > 0
⎪
θ=⎨ π + arctan ab if a < 0
⎪
⎪
⎪
⎪
⎪ 3π/2 if a = 0 and b < 0
⎪
⎪
⎪
⎩ 2π + arctan ab if a > 0 and b < 0.

14
MATH1013/LN/2019 Sem 1 2.2. GEOMETRIC REPRESENTATION OF COMPLEX NUMBERS

θ is called the principal value or principal argument of z, and is denoted by arg z.2
√ √
If we set r = ∣z∣ = a2 + b2 , then every non–zero complex number z = a + b −1 defines a pair of real number
(r, θ) ∈ (0, +∞) × [0, 2π) satisfying
(a, b) = (r cos θ, r sin θ) .

Conversely, it’s clear that every ordered pair (r, θ) ∈ (0, +∞) × [0, 2π) defines a non–zero complex number by the
formula √
z = r cos θ + r sin θ −1.
This representation of z is called the polar form of z, and it is usually abbreviated to

z = rcis θ = ∣z∣cis arg z

√
and we refer to the representation z = a + b −1 with respect to the Cartesian coordinates (a, b) as the Cartesian
form.

Cartesian form Polar form

Also, we adopt that 0 = 0cis θ for any θ.

Example 2.2.6. Find the Cartesian form of the following complex numbers which are in polar form.

1. z = 4cis (π/6).
2. z = 2cis (−2π/3).

Solution. Consider z = 4cis (π/6). The Cartesian form is

√ √ √
z = 4 cos(π/6) + 4 sin(π/6) −1 = 2 3 + 2 −1.
√
Similarly, the Cartesian form of z = 2cis (−2π/3) is z = −1 − −3.
Example 2.2.7. Find the polar form of the following complex numbers which are in Cartesian form.
√
1. z = 2 + −1.
√
2. z = −2 + −1.
√
Solution. Consider z = 2 + −1. Then

∣z∣ = and tan arg z = .

Since z is in the first quadrant, arg z = . It follows that the polar form of z is .
√
Similarly, the polar form of z = −2 + −1 is .

Example 2.2.9. Find the graph of arg(z + 1) = π/4.

2 Some authors use the range (−π, π] instead.

15
 CHAPTER 2. COMPLEX NUMBERS

√ √
Solution. Let z = x + y −1. Then z + 1 = (x + 1) + y −1 and
y
= tan(π/4) = 1
x+1
y = x + 1.

However, note the graph of the equation is NOT the whole straight line. Can you name extra condition(s) which
are satisfied by (x, y)?
So the graph is

Remark 2.2.10. Since θ + 2nπ, where n ∈ Z, and θ define the same angle at standard position, we define extended
argument of z = rcis θ to be the set
arg z = {θ + 2nπ ∶ n ∈ Z} .
When a complex number z is written in polar form, we shall employ the principal argument of z in general.
However, one can use any member from arg z during calculation.

Geometric Interpretation of Multiplication/Division of Complex Numbers

The first application of polar form is to give a geometric interpretation of multiplication and division of complex
numbers.
Let’s consider two complex numbers in polar form
√
zi = ri (cos θi + −1 sin θi ) , i = 1, 2.

Then
√ √
z1 z2 = r1 r2 (cos θ1 + −1 sin θ1 ) (cos θ2 + −1 sin θ2 )
R(z1 z2 ) = r1 r2 (cos θ1 cos θ2 − sin θ1 sin θ2 )
= r1 r2 cos(θ1 + θ2 );
I(z1 z2 ) = r1 r2 (sin θ1 cos θ2 + cos θ1 sin θ2 )
= r1 r2 sin(θ1 + θ2 ).

As a result, when a complex number z1 is multiplied by a complex number z2 , we scale the length of z1 by a
factor ∣z2 ∣ and rotate the point representing z1 counterclockwise by an angle arg z2 in the Argand plane.

Similarly, when a complex number z1 is divided by a nonzero complex number z2 , we scale the length of z1 by a
factor 1/∣z2 ∣ and rotate the point representing z1 clockwise by an angle arg z2 in the Argand plane.

16
MATH1013/LN/2019 Sem 1 2.2. GEOMETRIC REPRESENTATION OF COMPLEX NUMBERS

Theorem 2.2.11. For any z1 , z2 ∈ C,

√
z1 z2 = ∣z1 ∣∣z2 ∣ [cos(arg z1 + arg z2 ) + −1 sin(arg z1 + arg z2 )] .

If z2 ≠ 0, then
z1 ∣z1 ∣ √
= [cos(arg z1 − arg z2 ) + −1 sin(arg z1 − arg z2 )] .
z2 ∣z2 ∣
Note that arg z1 ± arg z2 may not be in the interval [0, 2π).
√ √ √ √
Example 2.2.12. Let z1 = 1 − −1, z2 = 3 + −1 and z3 = −1 − −3. Compute their polar forms and then
z1 z2 /z3 .

Solution. The polar forms of these complex numbers are

√
z1 = 2cis (7π/4)
z2 = 2cis (π/6)
z3 = 2cis (4π/3).

Thus,
z1 z2 7π π 4π 7π
= 2 2 +1−1 cis (
1
+ − ) = 21/2 cis .
z3 4 6 3 12

2.2.4 de Moivre’s Theorem

Question √
How would you compute (1 + −1)2017 ?
One may find it by brute force, or binomial theorem. However, they are inefficient.

In fact, it can be easily obtained by the following.

Theorem 2.2.13 (De Moivre Theorem). For any n ∈ Z and θ ∈ R,

n
(cis θ) = cis (nθ).

As usual, we adopt that z 0 = 1 for any complex number z.

The proof of Theorem 2.2.13 relies on the optional tool below.

Theorem 2.2.14 (Principle of Mathematical Induction). Given a sequence of statements S(1), S(2), S(3), . . . , S(n), . . ..
Suppose that

(i) statement S(1) is true,

(ii) for any positive integer k, statement S(k + 1) is true whenever statement S(k) is true.

Then S(n) is true for all n ∈ N.

Proof of Theorem 2.2.13 (Optional). We first prove by induction on n that the formula is true for n ∈ N. Clearly,
the formula is true when n = 1. Now suppose it is valid when n = k. Then apply Theorem 2.2.11 to get

z k+1 = z k ⋅ z = cis (kθ) ⋅ cis θ = cis ((k + 1)θ).

By the principle of mathematical induction, the formula is valid for all n ∈ N. Now if n = −m for some m ∈ N,
then
z0 1 ⋅ cis 0
zn = m = = cis (0 − m)θ = cis (nθ).
z cis (mθ)

17
 CHAPTER 2. COMPLEX NUMBERS

It follows immediately that for any z = rcis θ ∈ C,

z n = rn cis (nθ).

√
Example 2.2.15. Compute (1 + −1)2017 .
√
Solution. The polar form of 1 + −1 is
π
z = 21/2 cis .
4
By De Moivre theorem,
2017π π π √
z 2017 = 22017/2 cis = 22017/2 cis (252(2π) + ) = 22017/2 cis = 21008 (1 + −1) .
4 4 4

De Moivre theorem can also be used to establish some identities.

Example 2.2.16. Show that
sin 5θ = 16 sin5 θ − 20 sin3 θ + 5 sin θ.

Solution. Let z = cis θ. On the one hand, by De Moivre theorem

√
z 5 = cos 5θ + −1 sin 5θ.

On the other hand, by the binomial theorem,

5 √
z 5 = ∑ ( −1)5−r 5 Cr cosr θ sin5−r θ.
r=0

It follows that
5 √
sin 5θ = I(z 5 ) = I [ ∑ ( −1)5−r 5 Cr cosr θ sin5−r θ]
r=0
= 5 C4 cos
4
θ sin1 θ − 5 C2 cos2 θ sin3 θ + 5 C0 cos0 θ sin5 θ.

Use the identity cos2 θ = 1 − sin2 θ to simplify the RHS.

n–th Roots of Complex Numbers

Another application of de Moivre theorem is to find n–th roots of complex numbers.

Question √ √
What do you mean by saying that a complex number z = a + b −1 is a square root of −1?

To appreciate the power of de Moivre theorem, let’s try to solve the problem below without using this theorem.
√ √
Example 2.2.17. Suppose that (a + b −1)2 = −1. Then
√
R ((a + b√ −1)2 ) = = 0
{
I ((a + b −1)2 ) = = 1

It is routine to check that either (a, b) = ( , ) or (a, b) = ( , ).

Quest √
Can you use the same method to find the cube roots, fourth roots of −1 (if any)?

18
MATH1013/LN/2019 Sem 1 2.2. GEOMETRIC REPRESENTATION OF COMPLEX NUMBERS

Definition 2.2.18. Let n ∈ N ∖ {1} and α ∈ C ∖ {0}. Any solution of the equation
zn − α = 0
is called an n–th root of the complex number α.
In particular, if α = 1, then these roots are called the n–th roots of unity.

Question √
Let α = 3 + 4 −1 and z is a square root of α.

1. What is ∣z∣?
2. Can you determine the principal argument of z?

Theorem 2.2.19 (n–th Roots of Complex Number). Let α = rcis θ where r > 0 and θ ∈ [0, 2π). Then for any
n ∈ N ∖ {1}, the equation z n − α = 0 has n distinct roots

θ + 2kπ
zk = r1/n cis , k = 0, 1, . . . , n − 1.
n
Corollary 2.2.20 (n–th Roots of Unity). For any n ∈ N ∖ {1}, the n–th roots of unity are
1 = ωn0 , ωn = ωn1 , ωn2 , . . . , ωnn−1
2π k
where ωn = cis and ωnk = (ωn ) is the k–th power of ωn .
n
Proof of Theorem 2.2.19 (Optional). It’s clear that each zk is a solution. In addition, it’s readily seen that 0 ≤
argz0 < argz1 < . . . < argzn−1 < 2π.
Conversely, suppose that r0 cis θ0 is a solution. By De Moivre theorem
n
rcis θ = α = (r0 cis θ0 ) = r0n cis (nθ0 ).
It follows that r0 = r1/n (why?) and
nθ0 ∈ argα = {θ + 2mπ ∶ m ∈ Z} .
Suppose that nθ0 = θ + 2mπ for some m ∈ Z. Then there are q ∈ Z and k ∈ {0, 1, . . . , n − 1} such that m = nq + k.
As a result,
θ + 2mπ θ + 2nqπ + 2kπ θ + 2kπ
θ0 = = = + 2qπ.
n n n
√
Example 2.2.21. Find the cube roots of α = 1 + −1.

Solution. Note that polar form of α is

√ π
α= 2cis .
4
So the cube roots of α are
π/4 + 2kπ π 2π
zk = 21/6 cis = 21/6 cis ( + k ⋅ ) , k = 0, 1, 2.
3 12 3
π 3π 1/6 17π
Or explicitly, 21/6 cis , 21/6 cis , 2 cis .
12 4 12
Example 2.2.22. Find the cube roots of unity.

Solution. Note that √

2π √ 2π 1 3√
ω3 = cos + −1 sin =− + −1.
3 3 2 2
So the cube roots of unity are √ √
1 3√ 1 3√
1, − + −1, and − − −1.
2 2 2 2

19
 CHAPTER 2. COMPLEX NUMBERS

2.3 Miscellaneous Topics

2.3.1 Complex Exponents

In this subsection, we shall define ez where z ∈ C. Our approach is heuristic. A rigorous and general discussion
of z1z2 , with z1 , z2 ∈ C, would be in MATH3404. Recall that for x ∈ R,

1 1
ex = 1 + x + x2 + ⋯ + xn + ⋯
2 n!
√
If we take x = θ −1, a purely imaginary number, then
√ √ 1 1 √
−1
eθ = 1 + θ −1 − θ2 − θ3 −1 + ⋯
2 3!
1 2 1 4 1 6
= (1 − θ + θ − θ + ⋯)
2 4! 6!
1 3 1 5 1 7 √
+ (θ − θ + θ − θ + ⋯) −1
3! 5! 7!
√
= cos θ + sin θ −1

The above derivation motivates us to define

√ √
−1
eθ = cos θ + sin θ −1 (2.1)

which is also known as Euler’s formula.

Put θ = π and rearrange the terms to get
√
−1
eπ + 1 = 0. (2.2)
You can find the five most important mathematics constants come together nicely.
√
You may ask then what ez is, where z = u + v −1 is an arbitrary complex number. A natural definition should be
√ √ √
−1 −1
eu+v = eu ev = eu (cos v + sin v −1) (2.3)

so that the rules of exponentiation are preserved.

You may observe that Rz = u determines the modulus of ez while Iz = v determines the argument of ez . In
addition, you may check that the complex exponent enjoys the same algebraic properties of real exponent.

2.3.2 Complex–valued Functions over an Interval

Suppose that u(x) and v(x) are real–valued functions (i.e., the codomain is R). Then
√
f (x) ∶= u(x) + v(x) −1

is a complex–valued function. We define

√ √ √
(u1 (x) + v1 −1) + (u2 (x) + v2 −1) = (u1 + u2 )(x) + (v1 + v2 )(x) −1,
√ √ √
(u1 (x) + v1 −1) − (u2 (x) + v2 −1) = (u1 − u2 )(x) + (v1 − v2 )(x) −1,
√ √ √
(a + b −1)(u(x) + v(x) −1) = (au(x) − bv(x)) + (bu(x) + av(x)) −1,
√ √ √
(u1 + v1 −1) ⋅ (u2 + v2 −1) = (u1 u2 − v1 v2 ) + (v1 u2 + u1 v2 ) −1.

20
MATH1013/LN/2019 Sem 1 2.4. REFERENCES

The functions u and v are respectively the real part and the imaginary part of the function f .
√
If u and v are differentiable, then define the derivative of the complex–valued function f = u + v −1 by

df d √ du(x) dv(x) √
= (u(x) + v(x) −1) = + −1.
dx dx dx dx
It is readily check that for any α ∈ C and complex–valued functions fi (x),

d df1 (x) df2 (x)

(αf1 (x) + f2 (x)) = α (x) + .
dx dx dx
√
Example 2.3.1. Let α = a + b −1 ∈ C and f (x) = eαx where x ∈ R. Then the real part and the imaginary part of
f are respectively and . In addition,

d
f ′ (x) = ( )
dx √
= (a + b −1)f (x) = αf (x).

Define complex conjugate of complex–valued function by

√ √
u(x) + v(x) −1 = u(x) − v(x) −1.
√
Example 2.3.2. Let α = a + b −1 ∈ C. Then
√ √
eαx = eax cos(bx) + eax sin(bx) −1 = eax cos(bx) − eax sin(bx) −1
√
= eax cos(−bx) + eax sin(−bx) −1 = eαx .
√
It is easy to see that for any complex–valued function f (x) = u(x) + v(x) −1,

df d √ d √ df
(x) = (u(x) + v(x) −1) = (u(x) + v(x) −1) = .
dx dx dx dx

2.4 References
• Bayin, S˛. Selc˛uk, Essentials of Mathematical Methods in Science and Engineering, John Wiley & Sons,
Inc., Hoboken, NJ, USA, 2008, pp.331–338.
• Fryres ; Jr. ; Phi. Schmidt, Schaum’s outline of theory and problems of college mathematics: algebra,
discrete mathematics, precalculus, introduction to calculus, McGraw-Hill, USA, 2003. Chapter 34.

21