CUST 2023-2024
Maths for Physics 2
Session 12
Analytic functions
Summary
In this session we continue our study of complex analysis, and introduce in particular
the concept of an analytic function, which is a very important concept in complex
analysis. We first define the notion of limit of a complex function at a point, which
then naturally leads us to the notion of continuity. We then discuss the notion of
differentiability of a complex function, which eventually allows to define the notion
of analyticity.
i
Contents
1 Limit of a complex function at a point 1
2 Continuity of a complex function 5
3 Differentiability and analyticity of a complex function 8
3.1 Differentiability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 Analyticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
ii
Chapter 1
Limit of a complex function at a
point
The notion of limit of a complex function at a (complex) point is constructed in a
similar way as for a real function of a real variable. Therefore, let’s first consider a
real function f of a real variable x, i.e. f (x). What does it mean to say that, for
some x0 ∈ R,
lim f (x) = L , (1.1)
x→x0
with L ∈ R? It means that, as x approaches x0 , either from the left or from the right,
the corresponding values of f (x) approach L, and this independently of whether x
approaches x0 from either the left or the right. This is illustrated in figure 1.
X K 7
Ko no
g
flat L feel L
t x
Figure 1: Schematic illustration of the meaning of the limit (1.1).
1
CHAPTER 1. LIMIT OF A COMPLEX FUNCTION. . . 2
This idea can then be extended to the case of a complex function f of a complex
variable z, i.e. f (z). That is, we want
lim f (z) = L , L ∈ C, (1.2)
z→z0
to represent the fact that, as z approaches z0 , the corresponding values of f (z)
approach L. But what does it mean now to have z approaching z0 ? This is precisely
where things get a bit more complicated in the complex case as compared to the
real case:
• in the real case, x → x0 means that x gets closer to x0 either from the left or
from the right: this is because x here is only free to move on a line;
• but in the complex case, z and z0 are points of the complex plane, which is a
2D plane, which means that z is now free to move on a plane: therefore, in
this case, z → z0 means that z gets closer to z0 from all possible directions in
the plane!
This is illustrated in figure 2.
Ease
x sx.com
ase.fz
s I
nF
Eyz
71 721 X 7 Xo
left right
716 distigs
o
Re
Figure 2: Schematic comparison between the meaning of a real limit x → x0 and of
a complex limit z → z0 .
z
CHAPTER 1. LIMIT OF A COMPLEX FUNCTION. . . 3
Therefore, in view of this, let’s now give the mathematical definition of the limit
of f as z → z0 :
Limit of a complex function
Let the function w = f (z) be defined at all points z in a
neighborhood of a point z0 . We say that f (z) has the limit w0
as z goes to z0 , i.e.
lim f (z) = w0 , (1.3)
z→z0
if for any sufficiently small ϵ > 0 we can find a number δ > 0
such that for any z
0 < |z − z0 | < δ =⇒ |f (z) − w0 | < ϵ . (1.4)
As is depicted in figure 3, this precise mathematical definition can be illustrated by
viewing the function f as a mapping between the z plane (for the variable z) and the
w plane (for the image of z under f ). First, let’s consider the w plane: the condition
|w − w0 | < ϵ simply describes a neighborhood N (w0 , ϵ) of w0 (i.e. a disk of center w0
and radius ϵ, with ϵ small) in this plane. Therefore, let’s fix an arbitrary such small
ϵ. Now, let’s consider the z plane: the condition |z − z0 | < δ here also describes a
neighborhood, namely a neighborhood N (z0 , δ) of z0 . The meaning of (1.3)-(1.4) is
thus the following: for any ϵ > 0 that we choose in the w plane, we can find a δ > 0
such that the neighborhood N (z0 , δ) of z0 of radius δ in the z plane is mapped,
under the function f , into a region R ≡ f (N (z0 , δ)) that is entirely contained inside
the neighborhood N (w0 , ϵ) in the w plane. This fact that we map regions takes into
account the fact that z must approach z0 from all possible directions in the complex
plane. If, for some small ϵ > 0 in the w plane, we can not find such a δ > 0 in the
z plane, then the limit limz→z0 f (z) does not exist.
As defined by (1.3)-(1.4), the limit satisfies the standard rules for sums, products,
etc. . .
Example: let’s show that, for f (z) = 3z − i, we have
lim f (z) = 2i . (1.5)
z→i
Here we have z0 = i and w0 = 2i. Therefore, let’s choose an arbitrary (small)
ϵ > 0 such that |f (z) − w0 | < ϵ, that is here
|f (z) − w0 | = 3|z − i| < ϵ . (1.6)
CHAPTER 1. LIMIT OF A COMPLEX FUNCTION. . . 4
R
f Wiz 8D
Plz S
IT XX we
fly
wait I
c
Re Re
wplaned zpl
Figure 3: Illustration of the definition (1.3)-(1.4) of the limit of f (z) as z → z0 .
Therefore, let’s choose for instance δ = ϵ/3, we have here
ϵ
0 < |z − z0 | < δ =⇒ 0 < |z − i| < ,
3
which is indeed precisely (1.6): therefore, we indeed found a δ > 0 (namely δ = ϵ/3,
but it can of course be any δ < ϵ/3) such that
0 < |z − i| < δ =⇒ |f (z) − 2i| < ϵ .
This indeed shows (1.5).
Chapter 2
Continuity of a complex function
We can now use the above definition of a limit to define the notion of continuity of
a function at a point, and we have by analogy with the real case:
Continuity of a complex function
A function f (z) is said to be continuous at the point z = z0 if
lim f (z) = f (z0 ) . (2.1)
z→z0
Remark: the definition (2.1) of the continuity of f at z0 hence implies (by definition
of the limit) that
• the function f must be defined at the point z0 , i.e. f (z0 ) must exist;
• the function f must in addition be defined in a neighborhood of this point z0 .
Furthermore, the definition (2.1) of the continuity of a complex function satisfies all
the standard properties for sums, products, etc. . .
Example: let’s show that is f (z) is continuous at z0 , then so is f ∗ (z). From the
continuity of f we have
lim f (z) = f (z0 ) ,
z→z0
that is, for any sufficiently small ϵ > 0, we can find δ > 0 such that
0 < |z − z0 | < δ =⇒ |f (z) − f (z0 )| < ϵ . (2.2)
Now, remember that |w| = |w∗ |, ∀w ∈ C, so that
|f (z) − f (z0 )| = |f ∗ (z) − f ∗ (z0 )| ,
5
CHAPTER 2. CONTINUITY OF A COMPLEX FUNCTION 6
so that (2.2) can be written as
0 < |z − z0 | < δ =⇒ |f ∗ (z) − f ∗ (z0 )| < ϵ ,
which is nothing but the definition of the limit
lim f ∗ (z) = f ∗ (z0 ) ,
z→z0
which in turn is just the definition of the continuity of f ∗ (z) at z0 .
Using the above example hence immediately allows us to conclude that, for any
function f (z) that is continuous at z = z0 , we have:
i) Re f (z) = [f (z) + f ∗ (z)]/2 is continuous at z0 ;
ii) Im f (z) = [f (z) − f ∗ (z)]/2i is continuous at z0 ;
iii) |f (z)|2 = f (z)f ∗ (z) is continuous at z0 .
Actually, it is also easy to show that if Re f (z) and Im f (z) are continuous at z0 ,
then f (z) itself is also continuous at z0 .
It turns out that many of these questions can also be addressed by means of real
analysis. Indeed, to do this, we write
z = x + iy , (2.3)
so that we can then write f (z) = f (x + iy) as
f (z) = u(x, y) + iv(x, y) , (2.4)
where u and v are thus real functions of the two real variables x and y. Therefore,
to say for instance that Re f (z) and Im f (z) are continuous at z0 = x0 + iy0 , i.e.
lim Re f (z) = Re f (z0 ) and lim Im f (z) = Im f (z0 ) ,
z→z0 z→z0
is thus equivalent to the continuity of the functions u and v at the point (x0 , y0 ), i.e.
lim u(x, y) = u(x0 , y0 )
x→x0
and lim v(x, y) = v(x0 , y0 ) .
x→x0
y→y0 y→y0
Therefore, we get for the limit of f (z) itself at z0
lim f (z) = x→x
lim [u(x, y) + iv(x, y)] = u(x0 , y0 ) + iv(x0 , y0 ) = f (z0 ) .
z→z0 0
y→y0
CHAPTER 2. CONTINUITY OF A COMPLEX FUNCTION 7
This interplay between a complex function f (z) and the real functions u(x, y) and
v(x, y) of two real variables, which are by construction the real and imaginary parts
of f , plays a prominent role in complex analysis.
So far, we defined the concept of continuity of a function f at a point z = z0 . It
is also straightforward to define the continuity of f in a region R: f is continuous
in R if it is continuous at every point of this region R.
Chapter 3
Differentiability and analyticity of a
complex function
3.1 Differentiability
Similarly to the continuity of a complex function, the notion of differentiability of
a complex function is defined in terms of a limit, and by analogy with the real case
we have:
Derivative of a complex function
Let f (z) be a complex function defined in some region R and
let z0 ∈ R. The complex number, which we denote by f ′ (z0 ),
defined by the limit
f (z0 + ∆z) − f (z0 )
f ′ (z0 ) ≡ lim (3.1)
∆z→0 ∆z
is called the derivative of f at z = z0 . If the limit (3.1) indeed
exists, then f is said to be differentiable at z = z0 .
An alternative, equivalent way of writing the definition (3.1) of f ′ (z0 ) is obtained
upon replacing z0 + ∆z by z, which means that ∆z = z − z0 and thus ∆z → 0
corresponds to z → z0 , so that we have
f (z) − f (z0 )
f ′ (z0 ) = lim . (3.2)
z→z0 z − z0
Remark: here again, since the definition (3.1) is based on a limit, we must thus
have that
• the function f must be defined at the point z = z0 ;
• the function f must also be defined in a neighborhood of this point z = z0 .
8
CHAPTER 3. DIFFERENTIABILITY AND. . . 9
Notation: common notations for the derivative at z0 are
df d df
f ′ (z0 ) ≡ ≡ f (z) ≡ (z0 ) . (3.3)
dz z=z0 dz z=z0 dz
Here again, to define the differentiability of f in a region R is straightforward:
f is differentiable in R if it is differentiable at every point of this region R.
Similarly to the real case, we have the following important property:
If a function f (z) is differentiable at the point z = z0 ,
then f is continuous at z0 .
Proof: we must show that, if f ′ (z0 ) exists, then we have
lim f (z) = f (z0 ) ,
z→z0
or, in other words,
lim [f (z) − f (z0 )] = 0 .
z→z0
This is indeed the case because
f (z) − f (z0 )
lim [f (z) − f (z0 )] = lim (z − z0 )
z→z0 z→z0 z − z0
f (z) − f (z0 )
= lim lim (z − z0 ) = 0 ,
z→z0 z − z0 z→z0
since the first limit in the second line precisely corresponds to f ′ (z0 ), which exists
(so that it is finite) by assumption. ■
The definition (3.1), or equivalently (3.2), of the derivative f ′ (z0 ) allows to keep
all the standard properties satisfied by derivatives in real analysis, such as
(f + g)′ = f ′ + g ′ ,
(f g)′ = f ′ g + f g ′ ,
′
f f ′g − f g′
= ,
g g2
CHAPTER 3. DIFFERENTIABILITY AND. . . 10
[f (g(z))]′ = g ′ (z)f ′ (g(z)) .
Furthermore, since the elementary functions such as z n , ez , cos z, etc. . . are defined
as natural extensions of their real counterparts xn , ex , cos x, etc. . . they admit the
same derivatives:
d n d z d
z = nz n−1 , e = ez , cos z = − sin z , etc. . .
dz dz dz
Example: let’s show explicitly that dz 2 /dz = 2z. We apply the definition (3.1) of
f ′ (z0 ) to the function f (z) = z 2 , and we have for some z0 ∈ C
f (z0 + ∆z) − f (z0 ) (z0 + ∆z)2 − z02
f ′ (z0 ) = lim = lim
∆z→0 ∆z ∆z→0 ∆z
∆z 2 + 2z0 ∆z
= lim = lim (2z0 + ∆z) = 2z0 ,
∆z→0 ∆z ∆z→0
which is indeed valid for any z0 ∈ C.
3.2 Analyticity
Rather than functions that are simply differentiable at a point, it proves to be
(very!) useful to consider functions f that are differentiable in small regions around
a point. This leads to the definition of so-called analytic functions, which are a very
important class of functions in complex analysis. Analytic functions are defined by:
Analytic function
A function f (z) is said to be analytic at a point z = z0 if f (z)
is differentiable in a neighborhood of this point z = z0 .
Careful: note the crucial difference between differentiability and analyticity of a
function f at a point z = z0 :
i) differentiability of f at z0 means that the limit (3.1) exists, which requires as a
prerequisite that f must be merely defined in a neighborhood of z0 ;
ii) analyticity of f at z0 then requires as a prerequisite that f must be differentiable
in a neighborhood of z0 , which is of course much stronger than simply asking
f to be only defined in such a neighborhood of z0 !
As usual, if we say that a function f is analytic in a region R, it means that f
is analytic at every point of R.
CHAPTER 3. DIFFERENTIABILITY AND. . . 11
Remark: in the literature, we may also see the term “holomorphic function”. The
two terms “analytic” and “holomorphic” are equivalent and can be used interchange-
ably.
Let’s conclude with some terminology:
i) if a function f is analytic at any z ∈ C, i.e. in the entire complex plane, it is
called an entire function. For instance, f (z) = z n is an entire function for any
n ∈ N;
ii) if there exists a point z = z0 at which a function f fails to be analytic, then z0
is called a singular point, or simply a singularity, of f .
Analytic functions have wonderful properties: just to give a glimpse of them,
let’s for instance mention here that if a function f is analytic in a region R, then
f actually admits derivatives of any order in this region R! In other words, by
requiring the existence of only the first derivative of a complex function in a region,
it is then guaranteed that all the higher-order derivatives of this function also exist
in this region! This sounds magical, but of course analyticity turns out to be a strong
condition, because precisely it requires differentiability in a region of the complex
plane (i.e. a 2D region), and not just at a point.
