1 Complex numbers and functions

1.1 Complex numbers

1.1.1 Definition and properties of complex numbers

The numbers that we typically used are called real numbers, and the set of real numbers
is denoted by the symbol R. For example, the numbers 1, 0, −2, 13.49, 21 , − 32 , and 0.333 . . .
all belong to the set R. From elementary mathematics, we know that a square of any
real number always gives a positive result. For instance (−2)2 = 4 and 32 = 9. In other
words, we could theoretically state that every positive number is a result of a squaring
operation. And we cannot find a number that, upon squaring, gives a negative number.

The notion of complex numbers arises as an aswer to the question, ‘what is the square
root of a negative number? ’ We define our fundamental complex unit to be the square
root of −1 as the following.

√
Definition 1.1.1 (Imaginary unit). An imaginary unit is i = −1, which implies
i2 = −1.
A complex number, z, is a number of the form

z = x + iy, (1.1)

where x and y are real numbers. We say that x = Re(z) is the real part and
y = Im(z) is the imaginary part. The set of complex numbers is denoted C.

Example 1.1.1:
A complex number is given as z = 3 + 4i. Find

(a) Re(z), the real part of z, and

(b) Im(z), the imaginary part of z.

Ex. 1.1.1 solution:

If z = 3 + 4i, then

(a) Re(z) = 3.

1
2 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

(b) Im(z) = 4.

Definition 1.1.2 (Sum and products of complex numbers). Given any two complex
numbers z = a+ib and w = c+id, the following arithmetic properties and operations
are defined:

1. Equality
Two complex numbers are equal only when their real parts and imaginary
parts are equal. In other words,

a + ib = c + id when a = b and c = d. (1.2)

2. Addition and subtraction

We can add two complex numbers by adding their real and imaginary parts
separately.

(a + ib) ± (c + id) = (a + c) ± i (b + d) . (1.3)

3. Multiplication
We can multiply two complex numbers by using the distributive law.

(a + ib) (c + id) = ac + ibc + iad + i2 bd

= ac − bd + i (ad + bc) . (1.4)

Example 1.1.2:
Given two complex numbers z = 6 − 4i and w = 8 + 13i. Calculate

(a) z + w and z − w.
(b) zw
(c) z 2 + 2w.

Ex. 1.1.2 solution:

With z = 6 − 4i and w = 8 + 13i,
1.1. COMPLEX NUMBERS 3

(a)

z + w = (6 − 4i) + (8 + 13i) = (6 + 8) + i(−4 + 13) = 14 + 9i,

z − w = (6 − 4i) − (8 + 13i) = (6 − 8) + i(−4 − 13) = −2 − 17i.

(b)

zw = (6 − 4i) + (8 + 13i) = (6)(8) + i(−4)(8) + i(6)(13) + i2 (−4)(13)

= 100 + 46i.

(c)

z 2 + 2w = (6 − 4i)2 + 2(8 + 13i) = 36 − 2(48)i − 16 + 16 + 26i

= 36 − 22i.

The following properties of complex numbers also hold:

1. Commutativity of addition: z + w = w + z.
2. Commutativity of multiplication: zw = wz.
3. Associativity of addition: z + (w + u) = (z + w) + u.
4. Associativity of multiplication: z(wu) = (zw)u.
5. Distributive property: z(w + u) = zw + zu.
6. Zero element: z + 0 = 0 + z.
7. Identity element: z · 1 = 1 · z.

We define the division of complex numbers as the multiplicative inverse. If we think about
it, what does wz mean for complex numbers? We can first define the multiiplicative inverse
of z such that a number z −1 = w such that zw = 1, provided that z 6= 0 and w 6= 0.
Therefore, if z = x + iy, then its multiplicative inverse is z −1

1 x − iy
z −1 = = 2 . (1.5)
z x + y2

We can confirm that

x2 + y 2
 
1 x − iy
z · = (x + iy) = = 1. (1.6)
z x2 + y 2 x2 + y 2
4 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Example 1.1.3:
z
Given two complex numbers z = 1 + 2i and w = 2 − i, find w
.

Ex. 1.1.3 solution:

z 1 + 2i
=
w 2−i
1 + 2i 2 + i
= ×
2−i 2+i
2 + 4i + i + 2i2
=
4+1
2 − 2 + 5i 5i
= = = i. (1.7)
5 5

1.1.2 Argand diagram/complex plane

A complex number can be represented visually with the use of an Argand diagram.
(Also called a complex plane.) The real part is plotted as the horizontal axis and the
imaginary part is plotted as the vertical axis, as shown in Fig. 1.1 below.

Im

z = x + iy
y

x
Re

Figure 1.1: The Argand diagram.

An Argand diagram helps us easily visualise two important quantities we about a complex
number, which are the magnitude and argment, defined as follows:
1.1. COMPLEX NUMBERS 5

Im

z = x + iy
y
2
y
2 +
p x
|=
|z
θ
x
Re

Figure 1.2: Magnitude and argument of a complex number.

Definition 1.1.3 (Magnitude). The magnitude of a complex number z = x + iy

is given by |z|, where
p
|z| = x2 + y 2 . (1.8)

Alternatively, the magnitude is also called the modulus. It is the distance of the point
(x, y) from the origin, as shown by the blue line in Fig. 1.2.

Definition 1.1.4 (Argument). The argument of a complex number z = x + iy is

defined as the angle between the positive real axis and the line joining the origin
and the point (x, y) on the Argand diagram. It is denoted by arg(z).

Since the argument is an angle, it is sometimes denoted by the symbol θ, and can be
visually depicted as shown in Fig. 1.2.

Example 1.1.4:
What is the magnitude and argument for z = −3 + 3i?

Ex. 1.1.4 solution:

p √
Given z = −3 + 3i, the magnitude is |z| = (−3)2 + 32 = 18. The argument is

3 3π
arg(−3 + 3i) = π − tan−1 = .
3 4

#
6 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Another aspect of complex numbers is the complex conjugate, which is the reflection
about the real axis:

Definition 1.1.5 (Complex conjugate). The complex conjugate of a complex

number is is obtained by changing the sign of its imaginary part. If z = x + iy, then
its complex conjugate is

z ∗ = x − iy. (1.9)

From the definition of complex conjugates, we see a modulus can be obtained as follows:

z ∗ z = (x − iy)(x + iy) = x2 + y 2
z ∗ z = |z|2 . (1.10)

Also, the real and imaginary parts can be expressed using conjugates:

z + z ∗ = (x + iy) + (x − iy) = 2x = 2Re(z)

1
Re(z) = (z + z ∗ ). (1.11)
2

Similarly, Im(z) = 21 (z − z ∗ ).

Also, it can be easily checked that if a complex number z is purely real if z ∗ = z, and is
purely imaginary if z ∗ = −z. These basic result can be collected in the following theorem:

Theorem 1.1.1. Let z and w be complex numbers. Then

1. (z ∗ )∗ = z.
2. (z + w)∗ = z ∗ + w∗ .
3. (zw)∗ = (z ∗ )(w∗ ).
4. |z| = |z ∗ |.
5. |zw| = |z||w|.
6. Re(z) = 12 (z + z ∗ ) and Im(z) = 1
2i
(z − z ∗ ).
7. |z| ≥ 0 and |z| = 0 if and only if z = 0.
8. zz ∗ = z ∗ z = |z|2 .

Proof. Some of these have already been proven. Most of the rest can be verified by direct
calculation.
1.1. COMPLEX NUMBERS 7

Theorem 1.1.2. Let w and z be complex numbers. Then

1. |Re(z)| ≤ |z| and |Im(z)| ≤ |z|.

2. |z + w| ≤ |z| + |w|.
3. ||z| − |w|| ≤ |z − w|

Proof. If z = a + ib, then Re(z) = a, and furthermore |Re(z)| = |a|. Also since |a| ≤
√
a2 + b2 , we combine the two facts that conclude that

|Re(z)| ≤ |z|. (1.12)

The same reasoning can be used to prove |Im(z)| ≤ |z|.

Statement 2 is known as the triangle inequality, which is proven for vectors. In terms of
complex numbers, we can also prove by considering

0 ≤ |z + w|2 = (z + w)∗ (z + w) = (z ∗ + w∗ )(z + w) = z ∗ z + w∗ w + z ∗ w + w∗ z

= |z|2 + |w|2 + w∗ z + (zw∗ )∗
= |z|2 + |w|2 + 2Re(zw∗ ) (1.13)

Now we use the statement 1. where Re(zw∗ ) ≤ |zw∗ |. Therefore

0 ≤ |z + w|2 = |z|2 + |w|2 + 2Re(zw∗ ) ≤ |z|2 + |w|2 ≤ 2|zw∗ |. (1.14)

Also, |z|2 + |w|2 ≤ 2|zw∗ | = |z|2 + 2|z||w| + |w|2 = (|z| + |w|)2 . Therefore

|z + w|2 ≤ (|z| + |w|)2

|z + w| ≤ |z| + |w|. (1.15)

A proof of statement 3 will be left as an exercise.

1.1.3 Trigonometric and polar form of a complex number

p
Let a complex number z = x + iy has modulus |z| = r = x2 + y 2 and argument
arg(z) = θ. Calculating with trigonometry (maybe by referring to Fig. 1.2, we find that

x = r cos θ, y = r sin θ. (1.16)

8 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Therefore a complex number can be given in its trignomometric form

z = r (cos θ + i sin θ) . (1.17)

Another useful form is the Euler form. First, recall Euler’s number

e = 2.7182818 . . . (1.18)

The function ex has a power series expansion

∞
x
X xn
e =
n=0
n!
x2 x3 x4 x5 x6 x7 x8
=1+x+ + + + + + + + ... (1.19)
2! 3! 4! 5! 6! 7! 8!

If we substitute x = iθ into the series,

(iθ)2 (iθ)3 (iθ)4 (iθ)5 (iθ)6 (iθ)7 (iθ)8

eiθ = 1 + iθ + + + + + + + + ...
2! 3! 4! 5! 6! 7! 8!
θ2 iθ3 θ4 iθ5 θ6 iθ7 θ8
= 1 + iθ − − + + + − + + ...
2! 3! 4! 5! 6!  7! 8!
θ2 θ4 θ6 θ8 θ3 θ5 θ7
 
= 1− + − + − . . . +i θ − + − + ... .
2! 4! 6! 8! 3! 5! 7!
| {z } | {z }
=cos θ =sin θ

We obtain Euler’s formula which states

eiθ = cos θ + i sin θ. (1.20)

Therefore, a complex number z = r (cos θ + i sin θ) can also be written in the form

z = reiθ . (1.21)

Eq. (1.21) is called the polar form, or the exponential form of a complez number z.

From the properties of angles, we see that θ + 2π gives exactly the same angle again, after
one extra rotation. Therefore

reiθ and rei(θ+2π) (1.22)

gives the same point on the Argand diagram. More generally,

eiθ+2πki , k = 0, 1, 2, . . . (1.23)
1.2. COMPLEX FUNCTIONS 9

gives the same point for any integer k, after k times of extra rotation.

Definition 1.1.6 ( (Principal argument) ). The principal argument is the case

arg(z) = Θ where −π < Θ ≤ π.

Example 1.1.5:
Find the principal argument of z = −1 − i.

Ex. 1.1.5 solution:

By drawing the point z = −1 − i on the Argand diagram, we can find its argument to be
 
−1 1 5π
θ = arg(z) = π + tan = . (1.24)
1 4

The value that we got is not in the range −π < arg(z) ≤ π. But if we go in the clockwise
direction, the angle − 3π
4
is the same angle as 5π
4
. The principal argument is therefore
− 3π
4
.

1.2 Complex functions

1.2.1 Defining complex functions

Let f be a function of a complex variable z in a certain domain D of the complex plane.

We can interpret f as a machine that takes a complex number z, and calculates a new
complex number f (z) as an output. This can be depicted visually in Fig. 1.3.

f
D

z f (z)

Figure 1.3: A complex function.

10 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Example 1.2.1:
z+i
A complex function is given by f (z) = .
z2 + 4
(a) What is f (i)?
(b) What is f (1 − i)?

Ex. 1.2.1 solution:

(a) For z = i, f (z) = f (i) gives

i+i 2i 2i
f (i) = = = . (1.25)
i2
+4 −1 + 4 3

(b) For z = 1 − i, f (z) = f (1 − i) gives

(1 − i) + i 1 1 1 4 + 2i 4 + 2i
f (1 − i) = 2
= 2
= = = 2
(1 − i) + 4 1 − 2i + i + 4 4 − 2i 4 − 2i 4 + 2i 4 + 22
4 + 2i 2+i 1 1
= = = + i. (1.26)
20 10 5 10

Since the output of a function f (z) is also a complex number, it will also have real and
imaginary parts. If z = x + iy, then f (z) can be written in terms of its real and imaginary
parts as

f (z) = u(x, y) + iv(x, y). (1.27)

Example 1.2.2:
Express the function f (z) = z 3 in the form f (z) = u(x, y) + iv(x, y).

Ex. 1.2.2 solution:

1.2. COMPLEX FUNCTIONS 11

Let z = x + iy. Then

f (z) = z 3 = (x + iy)3 = (x + iy)(x + iy)2

= (x + iy)(x2 + 2ixy − y 2 )
= x x2 − y 2 − 2xy 2 + iy(x2 − y 2 ) + 2ix2 y

= x3 − 3xy 2 +i 3x2 y − y 3 .


| {z } | {z }
u(x,y) v(x,y)

This is the form f (z) = u(x, y)+iv(x, y) where u(x, y) = x3 −3xy 2 and v(x, y) = 3x2 y−y 3 .

1.2.2 Limits

We then define the limit of a complex function in a similar way we did for calculus of real
numbers. Let f be defined in some region G ⊂ C. The limit of f (z) as z approaches z0
is equal to a number w0 is written as

lim f (z) = w0 . (1.28)

z→z0

This means that the output f (z) can be made arbitrarily close to w0 if we choose z close
enough to z0 . Note that we mention close enough, and not exactly z = z0 . More precisely,
for each positive number , there is another positive number δ such that

|f (z) − w0 | ≤  whenever 0 < |z − z0 | < δ, (1.29)

visually, this situation is shown in Fig. 1.4. For some |z − z0 | < δ means some pont z that
is within a circle of radius δ centred at z0 . All points in this circle must map to within
another circle of radius  centred at w0 .

Loosely speaking, we can interpret this as: if we change z a little bit, the output f (z)
also changes a little bit only (no sudden jumps).
12 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

z0 

z w0
f (z)

f
Figure 1.4

Example 1.2.3: 
1 ∗ i
Show that lim iz = .
z→1 2 2

Ex. 1.2.3 solution:

Here, f (z) = 12 iz ∗ , w0 = 2i , and z0 = 1. So,

i iz ∗ i |z − 1|
f (z) − = − = . (1.30)
2 2 2 2

We see that |f (z) − w0 | gets smaller whenever |z − 1| get smaller. In particular, we have
found a δ = 2, such that

i
f (z) − < whenever 0 < |z − 1| < |{z}
2 . (1.31)
2
δ

Therefore, the limit exists and is equal to 2i .

From the definition of limits using circles of radii  and δ, we see that for this limit w0
to exist, the point z must approach z0 from any direction within the circle. If we get
different answers by approaching from different directions, the limit does not exist.

We can explore this using the previous example, where f (z) = 2i z ∗ . We write the function
as f (x, y) = 2i (x + iy)∗ = 2i (x − iy). Let us first approach the point z = 1 along the
1.2. COMPLEX FUNCTIONS 13

x-axis. That means, we fix y = 0, and take the single-variable limit

i i
lim f (x, 0) = lim (x) = . (1.32)
x→1 x→0 2 2

On the other hand, if we try taking the limit vertically, we fix x = 1 and take the limit

i i
lim f (1, y) = lim (1 − iy) = , (1.33)
y→0 y→0 2 2

so both limits agree even if we take along different directions.

The following example shows a limit that does not exist.

Example 1.2.4:
z
Show that the limit lim ∗ does not exist.
z→0 z

Ex. 1.2.4 solution:

We write f (z) = zz∗ as

x + iy
f (z) = f (x, y) = . (1.34)
x − iy

The limit is approaching z0 = 0, which means x0 = 0 and y0 = 0. We first take the limit
along the x-axis. We first fix y = y0 and take the single-variable limit

x + i0 x
lim f (x, 0) = lim = lim = 1. (1.35)
x→0 x→0 x − i0 x→0 x

On the other hand, if we take the limit along the vertical direction,

0 + iy −iy
lim f (0, y) = lim = lim = −1. (1.36)
y→0 y→0 0 − iy y→0 iy

The two limits do not agree. Therefore the limit does not exist.

#
14 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Theorem 1.2.1. Suppose that

f (z) = u(x, y) + iv(x, y), where z = x + iy, (1.37)

along with z0 = x0 + iy0 , and w0 = u0 + iv0 . Then the limit

lim f (z) = w0 (1.38)

z→z0

if and only if

lim u(x, y) = u0 and lim v(x, y) = v0 . (1.39)

(x,y)→(x0 ,y0 ) (x,y)→(x0 ,y0 )

Proof. Multi-variable calculus tells us that the limits in Eq. (1.39) exists if for each positive
number , there exist positive numbers δ1 and δ2 such that

 p
|u − u0 | < whenever 0< (x − x0 )2 + (y − y0 )2 < δ1 , and (1.40a)
2
 p
|v − v0 | < whenever 0 < (x − x0 )2 + (y − y0 )2 < δ2 (1.40b)
2

Let δ be any positive number smaller than δ1 and δ2 . Since

|(u + iv) − (u0 + iv0 )| = |(u − u0 ) + i(v − v0 )| ≤ |u − u0 | + |v − v0 | (1.41)

and
p
(x − x0 )2 + (y − y0 )2 = |(x − x0 ) + i(y − y0 )| = |(x + iy) − (x0 + iy0 )|, (1.42)

it follows from Eqs. (1.40a) and (1.40b) that

 
|(u + iv) − (u0 + iv0 )| < + =
2 2
|f (z) − w0 | <  (1.43)

and

0 < |(x + iy) − (x0 + iy0 )| < δ

0 < |z − z0 | < δ, (1.44)

where δ is a positive number smaller than δ1 or δ2 . Therefore the limit holds.

1.2. COMPLEX FUNCTIONS 15

One more concept we shall bring over from calculus is the continuity of functions.

Definition 1.2.1 (Continuity). A complex function f is continuous at a point z0

if

lim f (z) = f (z0 ). (1.45)

z→z0

Of course, for the above equation to hold, both limz→z0 f (z) and f (z0 ) must exist
in the first place.

1.2.3 Derivatives

A function f (z) is differentiable at the point z0 if the derivative

 
0 f (z0 + ∆z) − f (z0 )
f (z0 ) = lim (1.46)
∆z→0 ∆z

exists and is unique. In other words, the value of the limit does not depend upon the
direction where the limit is taken.

Example 1.2.5:
Show that f (z) = z 2 is differentiable for all values of z.

Ex. 1.2.5 solution:

Let z = x + iy. Therefore f (z) = x2 − y 2 + 2ixy. Taking ∆z = ∆x + i∆y, we have

f (z + ∆z) − f (z) (x + ∆x)2 − (y + ∆y)2 + 2i(x + ∆x)(y + ∆y) − x2 + y 2 − 2ixy

=
∆z ∆x + i∆y
2x∆x + (∆x) − 2y∆y + (∆y) + 2i (x∆y + y∆x + ∆x∆y)2
2
=
∆x + i∆y
(∆x) − (∆y)2 + 2i∆x∆y
2
= 2x + i2y + . (1.47)
∆x + i∆y

Whichever way ∆x and ∆y are allowed to tend to zero, the last term in the RHS will
vanish. Therefore, the derivative is unique and is equal to

f 0 (z) = 2x + 2iy = 2z. (1.48)

#
16 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Notice that we have found that if f (z) = z 2 , then f 0 (z) = 2z, just like in ordinary calculus.

Example 1.2.6:
Show that f (z) = z ∗ is not differentiable anywhere in the complex plane.

Ex. 1.2.6 solution:

Let z = x + iy. Therefore our function is written as

f (z) = (x + iy)∗ = x − iy,

f (z + ∆z) = (x + iy)∗ = x + ∆x − i(y + ∆y) = x − iy + ∆x − i∆y.

Therefore

f (z + ∆z) − f (z) ∆x − i∆y

= . (1.49)
∆z ∆x + i∆y

If we take the limit in the horizontal direction, we fix ∆y = 0, and take the single-variable
limit

f (z + ∆z) − f (z) ∆x − 0
∆y = 0 : lim = lim = 1. (1.50)
∆x→0 ∆z ∆x→0 ∆x + 0

But if we take the limit in the vertical direction, we fix ∆x = 0 and take the single-variable
limit

f (z + ∆z) − f (z) 0 − i∆y

∆x = 0 : lim = lim = −1. (1.51)
∆y→0 ∆z ∆y→0 0 + i∆y

Both limits do not agree. Note that the results for both direction are independent of any
specific z. Therefore the function is not differentiable for any z. In other words, not
differentiable anywhere in the complex plane. #

Example 1.2.7:
Find the points of the function f (z) = |z|2 where it is differentiable, if any.

Ex. 1.2.7 solution:

1.2. COMPLEX FUNCTIONS 17

f (z) = |z|2 = z ∗ z
f (z + ∆z) = (z + ∆z)(z + ∆z)∗ = (z + ∆z)(z ∗ + ∆z ∗ ) = zz ∗ + z∆z ∗ + z ∗ ∆z + ∆z∆z ∗ .

Therefore

f (z + ∆z) − f (z) z∆z ∗ + z ∗ ∆z + ∆z∆z ∗ ∆z ∗

= = z∗ + z + ∆z ∗ . (1.52)
∆z ∆z ∆z
∗
The second term contains ∆z
∆z
where we have shown that the limit ∆z → 0 does not exist
in Example 1.2.4. However, at the point z = 0, Eq. (1.52) becomes

f (z + ∆z) − f (z)
= 0, (1.53)
∆z

f (z + ∆z) − f (z)
regardless of any direction of ∆z. Therefore we conclude that lim
∆z→0 ∆z
exists only at z = 0, where the limit is equal to 0. We have found that the function
f (z) = |z|2 is differentiable only at exactly one point, z = 0, and nowhere else. #

1.2.4 Cauchy–Riemann relations

Here, we wish to find a convenient way to determine whether a function f (z) = u(x, y) +
iv(x, y) is differentiable.

Let us take the limit

f (z + ∆z) − f (z)
L = lim (1.54)
∆z→0 ∆z

in two different ways: First by moving parallel to the real axis, and secondly moving
parallel to the imaginary axis.

To take the limit, let f (z) = u(x, y) + iv(x, y) and ∆z = ∆x + i∆y. Then

u(x + ∆x, y + ∆y) + iv(x + ∆x, y + ∆y) − u(x, y) − iv(x, y)

L= lim . (1.55)
∆x,∆y→0 ∆x + i∆y
18 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

If we take the limit by moving along the real axis, ∆y = 0 always. Then
 
u(x + ∆x, y) − u(x, y) v(x + ∆x) − v(x, y)
L = lim +i
∆x→0 ∆x ∆x
∂u ∂v
= +i . (1.56)
∂x ∂x

On the other hand, if we take the limit along the imaginary axis, ∆x = 0 always. Then
 
u(x, y + ∆y) − u(x, y) v(x, y + ∆y) − v(x, y)
L = lim +i
∆y→0 i∆y i∆y
1 ∂u ∂v
= +
i ∂y ∂y
∂v ∂u
= −i . (1.57)
∂y ∂y

If the function f (z) is differentiable, then the limit must be the same regardless of which
direction we must take it. In other words, Eq. (1.56) must be equal to (1.57). Therefore

∂u ∂v ∂v ∂u
= , =− . (1.58)
∂x ∂y ∂x ∂y

These are known as the Cauchy–Riemann relations, and are necessary but not sufficient
conditions for a function to be differentiable.

The sufficient condition for a function to be differentiable is that u and v must additionally
have continuous first partial derivatives, in addition to satisfying the Cauchy–Riemann
relations.

Theorem 1.2.2. Let f be a complex function given by f (z) = u(x, y) + iv(x, y)

for some domain z ∈ D. If

1. u and v and their first partial derivatives are continouous on D, and

2. u and v satisfies the Cauchy–Riemann relations,

then f is differentiable at each point of D.

If the conditions are satisfied, we can calculate the derivative of a complex function f (z)
by

f (z + ∆z) − f (z) ∂u ∂v ∂v ∂u
f 0 (z) = lim = +i = −i . (1.59)
∆z→0 ∆z ∂x ∂x ∂y ∂y
1.2. COMPLEX FUNCTIONS 19

Definition 1.2.2 (Harmonic functions). A function ϕ(x, y) is harmonic if it

satisfies the equation

∂ 2ϕ ∂ 2ϕ
∇2 ϕ = + 2. (1.60)
∂x2 ∂y

Theorem 1.2.3. Let G be an open set in the complex plane and suppose f (z) =
u(x, y) + iv(x, y) is differentiable on G. Then u and v are harmonic functions on G.

Proof. Tutorial.

Definition 1.2.3 (Analytic function). A function that is single-valued and differ-

entiable at all points of a domain R ⊆ C is said to be analytic in R. In other
books, the term holomorphic function is sometimes used instead.

Definition 1.2.4 (Entire function). An entire function is a function that is ana-

lytic on the entire domain.

Properties of derivatives: Let f and g be differentiable functions at a point z, then

d
1. c = 0.
dz
d
2. cf (z) = cf 0 (z).
dz
d
3. (f (z) + g(z)) = f 0 (z) + g 0 (z).
dz
d
4. (f (z)g(z)) = f 0 (z)g(z) + f (z)g 0 (z).
dz
d f (z) g(z)f 0 (z) − f (z)g 0 (z)
5. = .
dz g(z) g(z)2

Some additional facts:

• Related to the above theorem, the composition of two analytic function is differen-
tiable. If f (z) and g(z) differentiable functions in a common domain, then a new
function built with g(f (z)) is also differentiable.
• All polynomials are entire functions. That is, all polynomials are differentiable
everywhere.
20 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Definition 1.2.5 (Singularity). If a function fails to be differentiable at some point

z0 , then we say that f is singular at z = z0 .

1.2.5 Some elementary complex functions

The exponential function

What is the exponential of a complex number? If z = x + iy, then

ez = ex+iy = ex eiy
= ex (cos y + i sin y) , (1.61)

where we have used the Euler’s formula to expand eiy .

This gives us the definition of a complex exponential in terms of u(x, y) and v(x, y),

ez = ex cos y +i ex sin y . (1.62)

| {z } | {z }
u(x,y) v(x,y)

We can check the Cauchy–Riemann equations:

∂u ∂v

= ex cos y, − = −ex sin y, 

∂x ∂x CR equations hold (1.63)
∂u x ∂v
= −e sin y, − = ex cos y 

∂y ∂y
| {z }
derivatives exist and continuous

We see that the derivatives exists and are continuous for any (x, y), and the CR relations
hold for any (x, y). Conditions 1 and 2 of Theorem 1.2.2 are satisfied. Therefore ez is
differentiable at any z.

The derivative for f (z) = ez can be calculated using

∂u ∂v
f 0 (z) = +i = ex cos y + iex sin y
∂x ∂v
= ez

Therefore, we have found that

d z
e = ez , (1.64)
dz

which is the same property as real exponentials.

1.2. COMPLEX FUNCTIONS 21

Theorem 1.2.4. Let z and w be complex numbers. Then

1. e0 = 1.
2. ez+w = ez ew .
3. ez 6= 0 for all z.
1
4. e−z = z if z 6= 0.
e

It is noteworthy to mention that if x is a real number, ex always gives a positive real

number. But if z is a complex number, ez might possibly be negative.

Also, if z is increased in the imaginary direction, the value of ez will repeat. In other
words, ex is periodic in the imaginary direction with period 2πi.

Theorem 1.2.5. The following are the periodic properties of ez .

1. ez = 1 if and only if z = 2nπi for some integer n.

2. If p is a number such that ez+p = ez for all complex z, then for some integer
n, p = 2nπi.
3. ez is periodic with periodic 2nπi for each non-zero integer n. Furthermore,
these are the only periods of the complex exponential function.

Proof. To prove (1), first observe that, if z = 2nπi, for some integer n, then

ez = e2nπi = cos(2nπ) + i sin(2nπ) = 1 (1.65)

because cos(2nπ) = 1 and sin(2nπ) = 0.

Conversely, if ez = ea+ib = 1, then by the definition of the complex exponential,

ea cos b + iea sin b = 1. (1.66)

The right hand side of this equation has no complex part. Therefore the left hand side
must also have no complex part. In other words

ea sin b = 0 → sin b = 0 since ea 6= 0. (1.67)

If sin b = 0, then b = kπ for some integer k. Substituting this back into (1.66),

ea cos kπ + 0 = 1. (1.68)
22 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

ea is always positive. So for the right hand side to be positive 1, cos kπ must be positive.
This is only possible if k is an even number. Therefore k = 2n for some integer n. So
cos 2nπ = 1 and therefore

ea cos 2nπ = 1
ea = 1. (1.69)

Since a is real, ea = 1 is only possible for a = 0. Therefore we have found

a = 0, b = 2nπ, n = integer. (1.70)

To prove (2), suppose ez+p = ez for all z. Then

ez+p = ez ep = ez . (1.71)

Therefore, ep = 1. By (1), p must be p = 2nπi for some integer n.

For (3), let n be a non-zero integer. By (1),,

ez+2nπi = ez e2nπi = ez . (1.72)

Therefore, the function f (z) = ez+2nπi satisfies the condition

f (z + 2nπi) = f (z). (1.73)

So ez has a period 2nπi. To show that any period of ez has to be an integer multiple of
2π, suppose ez+p = ez . By (2), p = 2nπi for some non-zero integer n.

The cosine and sine functions

To define the sine and cosine functions when the argument is a complex number, we start
again with the Euler’s formula and take its complex conjugate.

eiθ = cos θ + i sin θ (1.74a)

iθ ∗
= e−iθ = cos θ − i sin θ.


e (1.74b)

Taking the sum and difference of the two equations leads to the following:

1 iθ 1 iθ
e + e−iθ , e − e−iθ .



cos θ = sin θ = (1.75)
2 2i

Since we have already defined ez for complex numbers, we can use this to define the cosine
1.2. COMPLEX FUNCTIONS 23

and sine function as

1 iz 1 iz
e + e−iz , e − e−iz ,



cos z = sin z = (1.76)
2 2i

for complex number z. If z = x + iy, we can extract its real and imaginary parts:

1 i(x+iy)
+ e−i(x+iy)


cos z = e
2
1 ix −y
e e + e−ix ey


=
2
1  −y
e (cos x + i sin x) + ey (cos x − i sin x)

=
2
1 1
= (cos x) ey + e−y − i (sin x) ey − e−y



2 2
= cos x cosh y − i sin x sinh y. (1.77)

In the standard form f (z) = u(x, y) + iv(x, y), we have u = cos x cosh y and v =
− sin x sinh y. By a similar calculation, we find that

sin z = sin x cosh y + i cos x sinh y. (1.78)

We can calculate the derivative of cos z by using the Cauchy-Riemann relations

d ∂u ∂v
cos z = +i
dz ∂x ∂y
= − sin x cosh y − i cos x sinh y = − sin z. (1.79)

Theorem 1.2.6. The zeroes and periods of the complex cosine and sine functions
have the following properties.

(1) sin z = 0 if and only if z = nπ for some integer n.

(2) cos z = 0 if and only if z = (2n + 1)π/2 for some integer n.
(3) cos z and sin z are periodic with periods 2nπ for every positive integer n.
Further, these are only periods with these functions.

The complex logarithm

For real numbers, we know that the logarithm is the inverse function of the exponential;
namely ln y = x ↔ y = ex . Now we shall use this idea to include complex numbers.
24 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Given z = reiθ 6= 0, and w = a + ib, we start with the equation

ew = z
ea+ib = reiθ
ea eib = reiθ . (1.80)

Taking the magnitudes of both sides,

ea eib = reiθ
ea eib = r eiθ . (1.81)

By Euler’s formula, eib = eiθ = 1. Therefore we conclude that

ea = r → a = ln r. (1.82)

Next we calculate b. Since now we have eib = eiθ , we have

eib
= ei(b−θ) = 1. (1.83)
eiθ

Now, Theorem 1.2.3 tells us that

ei(θ+2nπ)
ei(θ+2nπ) = eiθ → = 1. (1.84)
eiθ

Comparing Eq. (1.83) and (1.84), we conclude that

b = θ + 2nπ. (1.85)

Therefore, we have w = a + ib = ln r + iθ + 2nπi, and so

ew = z ↔ w = ln r + iθ + 2nπi
ln z = ln r + iθ + 2nπi.

Notice that r is actually the modulus of z, |z| = r. Therefore

ln z = ln |z| + iθ + 2nπi. (1.86)

Note that the complex logarithm is not a ‘proper’ function, since for different n, it gives
(infinitely) different outputs.

Powers
Next we want to see how to calculate functions of the form z w , where both z and w could
1.2. COMPLEX FUNCTIONS 25

be complex numbers. Let us consider different possibilities of the power w.

(i) When w = n, a real-number integer, the function z n is straightforward and can

be calculated using the usual rules of algebra. (Basically, z 2 means multiply zz,
z 4 = zzzz, etc.)
(ii) When w = 1/n, when n is a real-number integer, these are called the n-th roots of
z, z 1/n . These can be calculated by expressing z in its polar form, and remembering
Theorem 1.2.3,

z = reiθ = reiθ+2kπi , k = some integer. (1.87)

At this point, when k takes its different integer values, the output z is the same
because of the periodicity of the complex exponential. But this will no longer happen
after we take the n-th root, because

z 1/n = r1/n ei(θ+2kπ)/n . (1.88)

If k takes its different values, the exonential changes by an amount less than 2kπ
(because it has been divided by n.) Therefore, they give different results when k
runs from 0 until n − 1.
(iii) For rational powers of the form z m/n , where m and n are integers, we simply calculate
z m first and find its n-th root.
(iv) Generally, for any number w, z w can be calculated by

z w = ew ln z , ln z = ln |z| + i (arg z + 2nπ) . (1.89)

Example 1.2.8:
Find the 4-th roots of 16i.

Ex. 1.2.8 solution:

First, write the complex number in polar form,

π
16i = 16ei( 2 +2πn) , n = integer. (1.90)

Taking the fourth roots,

π πn
(16i)1/4 = 161/4 ei( 8 + 2 ) , n = 0, 1, 2, 3.
h  π nπ   π nπ i
= 2 cos + + i cos + (1.91)
8 2 8 2
26 CHAPTER 1. COMPLEX NUMBERS AND FUNCTIONS

Note that it is sufficient to consider n from 0 to 3. Because for k = 4 onwards, we repeat

the same numbers from earlier. Considering all the cases n = 0, . . . 3, we find
 π π
n=0: (16i)1/4 = 2 cos + i sin , (1.92a)
 8 8 
5π 5π
n=1: (16i)1/4 = 2 cos + i sin , (1.92b)
8 8
 
1/4 9π 9π
n=2: (16i) = 2 cos + i sin , (1.92c)
8 8
 
1/4 13π 13π
n=3: (16i) = 2 cos + i sin . (1.92d)
8 8

Alternatively, we may use Eq. (1.89) to calculate the root. Since

π
ln(16i) = ln 16 + i + 2nπi. (1.93)
2

Using (1.89),

1 π π nπ π nπ
(16i)1/4 = e 4 [ln 16+i( 2 +2nπ)] = eln 2 ei( 8 + 2 ) = 2ei( 8 + 2 ) . (1.94)

Listing out n = 0, 1, 2, 3, we get the same answer as (1.92)

Example 1.2.9:
Calculate all values of (1 − i)1+i .

Ex. 1.2.9 solution:

The modulus and argument of the base z = 1 − i is
√ π
|1 − i| = 2, arg(1 − i) = − + 2πn, n = integer. (1.95)
4

Therefore, this number raised to the power 1 + i is

√
z 1+i = e(1+i) ln z = e(1+i)[ln 2+i(−π/2+2πn)]
√ √
= 2eπ/4−2πn ei(ln 2−π/4+2πn)
√ h  √ π  √ π i
= 2eπ/4−2πn cos ln 2 − + i sin ln 2 − . (1.96)
4 4