Complex Analysis-I

Real and Complex Analysis

MTL122/ MTL503/ MTL506
Lecturer: A. Dasgupta [email protected]

1. Complex numbers
The complex numbers can be defined as pairs of real numbers,
C = {(x, y) : x, y ∈ R},
equipped with the addition
(x, y) + (a, b) = (x + a, y + b)
and the multiplication
(x, y) · (a, b) = (xa − yb, xb + ya).
at C is an extension of R, in the sense that the complex numbers of the form (x, 0)
behave just like real numbers; that is, (x, 0) + (y, 0) = (x + y, 0) and (x, 0) · (y, 0) =
(x · y, 0). So we can think of the real numbers being embedded in C as those complex
numbers whose second coordinate is zero.
The definition of our multiplication implies
(0, 1) · (0, 1) = (−1, 0). (1.1)
This identity together with the fact that
(a, 0) · (x, y) = (ax, ay)
allows an alternative notation for complex numbers. The latter implies that we can
write
(x, y) = (x, 0) + (0, y) = (x, 0) · (1, 0) + (y, 0) · (0, 1).)
If we think—in the spirit of our remark on the embedding of R in C—of (x, 0) and
(y, 0) as the real numbers x and y, then this means that we can write any complex
number (x, y) as a linear combination of (1, 0) and (0, 1), with the real coefficients
x and y. (1, 0), in turn, can be thought of as the real number 1. So if we give (0, 1)
a special name, say i, then the complex number that we used to call (x, y) can be
written as x · 1 + y · i, or in short,
x + iy.
The number x is called the real part and y the imaginary part of the complex number
x + iy, often denoted as Re(x + iy) = x and Im(x + iy) = y. The identity (1.1) then
reads
i2 = −1.
Although we just introduced a new way of writing complex numbers, let’s for a
moment return to the (x, y)-notation. It suggests that one can think of a complex
number as a two-dimensional real vector. When plotting these vectors in the plane
R2 , we will call the x-axis the real axis and the y-axis the imaginary axis. The plane
of two axes representing complex numbers as points is called the complex plane or

1
2

Argand Plane. The addition that we defined for complex numbers resembles vector
addition. The analogy stops at multiplication: there is no “usual” multiplication of
two vectors in R2 and certainly not one that agrees with our definition of the product
of two complex numbers.

Any vector in R2 is defined by its two coordinates. On the other hand, it is also
determined by its length and the angle it encloses with, say, the positive real axis.
• The absolute value or modulus r = |z| ∈ R of z = x + iy is
p
r = |z| = x2 + y 2 .

• The argument of z = x + iy is a number θ ∈ R such that

x = r cos θ and y = r sin θ,
that is y
argz = arctan.
x
A given complex number z = x+iy has infinitely many possible arguments.
For instance, the number 1 = 1 + 0i lies on the x-axis, and so has argument
0, but we could just as well say it has argument 2π, 4π, −2π, or 2π ∗ k for
any integer k. The number 0 = 0 + 0i has modulus 0, and every number φ is
an argument. Aside from the exceptional case of 0, for any complex number
z, the arguments of z all differ by a multiple of 2π, just as we saw for the
example z = 1.

θ is not uniquely defined: If θ = arctan( xy ) then θ + 2kπ, k = ±1 ± 2... also

fit. Note: When calculating θ you must take account of the quad-
rant in which z lies.

• The principal value of the argument is denoted by Argz, and is the

unique value of argz such that −π < argz ≤ π. Argz is obtained by adding
or subtracting integer multiples of 2π from argz.
By argz we denote the set of all possible arguments, ie.
argz = {Argz + 2πk : K ∈ Z}.
• Writing a complex numbers in terms of polar coordinates r and θ:
z = x + iy = r cos θ + ir sin θ = r (cos θ + i sin θ) = riθ .
• For any two complex numbers z1 and z2
arg(z1 z2 ) = argz1 + argz2
and for z2 6= 0,

z1
arg = argz1 − argz2 .
z2
3

The ‘metric’ d : C × C → R in C is defined by ,

d(z1 , z2 ) = |z1 − z2 | = |z2 − z1 |.
d(z2 , z1 ) = distance between the endpoints of the two vectors in R2 .
That |z1 − z2 | = |z2 − z1 | simply says that the vector from z1 to z2 has the same
length as its inverse, the vector from z2 to z1 .

Euler’s Identity
We all know what eθ is if θ is a real number. What is eiθ ?
∞
iθ
X (iθ)n
e =
n=0
n!
iθ (iθ)2 (iθ)3
= 1+ + + + ....
1! 2! 3!
(iθ)2 (iθ)4 (iθ) (iθ)3 (iθ)5

= 1+ + + ... + + + + ...
2! 4! 1! 3! 5!
(θ)2 (θ)4 (θ)3 θ5

= 1− + − ... + i θ − + − ...
2! 4! 3! 5!
= cos θ + i sin θ. (1.2)
So we have the celebrated Euler’s identity,
eiθ = cosθ + i sin θ.
Recall that the polar form of a nonzero complex number z is given by
z = r (cos θ + i sin θ) .
thus, using Euler’s identity, the polar form becomes
z = reiθ .
We have the following inequalities,
−|z| ≤ Rez ≤ |z| and − |z| ≤ Imz ≤ |z|.
p
We know |z| = |x + iy| = x2 + y 2 , that is,
|x + iy|2 = x2 + y 2 = (x + iy)(x − iy).
x − iy is called the (complex) conjugate of x + iy. We denote the conjugate by
x + iy = x − iy.
Geometrically, conjugating z means reflecting the vector corresponding to z with
respect to the real axis. The following collects some basic properties of the conjugate.
For any z, z1 , z2 ∈ C,
(a) z1 ± z2 = z1 ± z2

(b) z1 · z2 = z1 · z2

(c) zz12 = zz12
4

(d) z = z

(e) |z| = |z|

(f) |z|2 = zz

(g) Rez = 12 (z + z)
1
(h) Imz = 2i
(z − z)

(i) eiφ = e−iφ .

From part (f) we have a neat formula for the inverse of a non-zero complex number:
1 z
z −1 = = 2.
z |z|
A famous geometric inequality is the triangle inequality
|z1 + z2 | ≤ |z1 | + |z2 |.
Definition 1.1. Let z = x + iy be a complex number. Then we define ez by
ez = ex+iy = ex eiy = ex (cos y + sin y).
Let z = reiθ be a nonzero complex number. Then for every positive integer n, the
n power z n of z is given by
th

z n = rn einθ = rn (cos nθ + i sin nθ) .

Is there a formula for all the nth
√ roots of z?
To solve this problem, let ζ = z. Suppose that ζ = ρeiφ . Then
n

ζ n = z =⇒ ρn einφ = reiθ .
So,
ρn = r
and φ = n1 (θ + 2kπ), where k = 0, 1, 2, ..., n − 1. Thus, there are n nth roots of z given
by
√
ζk = n rei(θ+2kπ)/n , k = 0, 1, 2, ..., n − 1.
√
Example 1.2. Find the fourth roots of unity, i.e., 4 1.
Solution 1.3. Since 1 = 1e0 , we get r = 1 and θ = 0. Thus, the four fourth roots of
unity are
ζk = ei(2kπ)/4 = eikπ/2 , k = 01, 2, 3.
So,
ζ0 = 1, ζ1 = eiπ/2 = i, ζ2 = eiπ = −1, ζ3 = ei3π/2 = −i.
5

2. Functions of Complex Variable

A function defined on a set D is a rule that associates to each point z of D a
complex number w. Set D is called the domain and w is called the image of f at z
and is denoted by f (z) = w.
f (z) = f (x + iy) = u(x, y) + iv(x, y)

f (z) = f (reiθ ) = u(r, θ) + iv(r, θ).

Example 2.1. Write f (z) = 1/z 2 in u + iv form.
2 −y 2
u(x, y) = (xx2 +y 2xy
2 )2 and v(x, y) = (x2 +y 2 )2 .

Also u(r, θ) = r−2 cos 2θ and v(r, θ) = −r−2 sin 2θ.

Domain of f is C \ {0}.
Example 2.2. (1) Polynomials
P (z) = a0 + a1 z + a2 z 2 + ... + an z n
where the coefficients are real.
(2) Exponential Function
z2 z3
ez = 1 + z + + + ...
2! 3!
converges for all z.
(3) Trigonometric Functions

Define
eiz + e−iz
cos z =
2
eiz − e−iz
sin z =
2i
sin z
tan z =
cos z
(4) Hyperbolic Functions

ez + e−z
cos hz =
2
ez − e−z
sin hz =
2

3. Limits
Definition 3.1. The limit of the function f (z) as z → z0 is a number w0 if for any
given > 0 there exists a δ > 0 such that
|z − z0 | < δ =⇒ |f (z) − w0 | < .
6
5
Example 3.2. Prove that lim (2 + i)z = 2
(1 + 3i). According to the definition,
z→1+i
lim (2 + i)z = 52 (1 + 3i). if, for every > 0, there is a δ > 0, such that |(2 + i)z −
z→1+i
5
2
(1+3i)| < whenever 0 < |z −(1+i)| < δ. We first factor (2+i) out of the left-hand
side:
5 1 + 3i
|2 + i| · z −
< .
2 2+i
√ √
Because |2+i| = 5 and 1+3i
2+i
= (1+i), we get 5·|z−(1+i)| < or |z−(1+i)| < √ .
5
This indicate that we should take δ = √5 .
Theorem 3.3. Let f (z) = u(x, y) + iv(x, y) and w0 = u0 + iv0 . lim f (z) = w0 if and
z→z0
only if lim u = u0 and lim v = v0 .
(x,y)→(x0 ,y0 ) (x,y)→(x0 .y0 )

Example 3.4. Compute lim (z 2 + i).

z→1+i
Since f (z) = z 2 + i = x2 − y 2 + (2xy + 1)i, we set
u(x, y) = x2 − y 2 , v(x, y) = 2xy + 1
and z0 = 1 + i, i.e., x0 = 1 and y0 = 1. We next compute the two real limits:
u0 = lim (x2 − y 2 ) = 12 − 12 = 0,
(x,y)→(1,1)

v0 = lim (2xy + 1) = 2.1.1 + 1 = 3.

(x,y)→(1,1)
Therefore, L = u0 + iv0 = 0 + 3i = 3i, that is, lim = 3i.
z→1+i

• There is an important difference between the two concepts of limit:

– In a real limit, there are two directions from which x can approach x0
on the real line, from the left or from the right.
– In a complex limit, there are infinitely many directions from which z can
approach z0 in the complex plane. In order for a complex limit to exist,
each way in which z can approach z0 must yield the same limiting value.
Example 3.5. Real One-Sided Limits
There is at least one very important difference between real and complex limits.
• For real functions lim f (x) = L if and only if lim+ f (x) = L and lim− f (x) =
x→x0 x→x0 x→x0
L. . Since there are two directions from which x can approach x0 on the real
line, the real limit exists if and only if these two one-sided limits have the
same value.

• Example
Consider the real function
(
x2 , if x < 0.
f (x) == (3.1)
x − 1, if x ≥ 0.
The limit of f as x approaches 0 does not exist:
lim− f (x) = lim− x2 = 0,
x→x0 x→x0
7

but
lim f (x) = lim+ x − 1 = −1,
x→x+
0 x→x0

For limits of complex functions, z is allowed to approach z0 from any direction in the
complex plane, i.e., along any curve or path through z0 .

For lim f (z) to exist and to equal L, we require that f (z) approach the same
z→z0
complex number L along every possible curve through z0 . f (z) to exist and to equal
L, we require that f (z) approach the same complex number L along every possible
curve through z0 .
Remark 3.6. If f approaches two complex numbers L1 6= L2 for two different curves
or paths through z0 , then lim f (z) does not exist.
z→z0

Example 3.7. Show that lim z̄z does not exist.

z→0

• First, we let z approach 0 along the real axis. That is, we consider complex
numbers of the form z = x + 0i, where the real number x is approaching 0.
For these points we have:
z x + 0i
lim = lim = lim 1 = 1.
z→0 z x→0 x − 0i x→0

• On the other hand, if we let z approach 0 along the imaginary axis, then
z = 0 + iy, where the real number y is approaching 0. For this approach we
have
z 0 + iy
lim = lim = lim (−1) = −1.
z→0 z y→0 0 − iy y→0
Since the two values are not the same, we conclude that lim zz does not exist.
z→0

Properties of Complex Limits

Suppose that f and g are complex functions. If lim f (z) = L and lim g(z) = M,
z→z0 z→z0
then:
(i) lim cf (z) = cL, c a complex constant;
z→z0
(ii) lim (f (z) ± g(z)) = L ± M ;
z→z0
(iii) lim f (z) · g(z) = L · M, and
z→z0
f (z) L
(iv) lim = , provided M 6= 0.
z→z0 g(z) M

4. Continuity
A complex function f is continuous at a point z0 if lim f (z) = f (z0 ).
z→z0

A complex function f is sontinuous at a point z0 if each of the following three

conditions hold:
• lim f (z) exists;
z→z0
• f is defined at z0 :
8

• lim f (z) = f (z0 ).

z→z0
If a complex function f is not continuous at a point z0 , then we say that f is
discontinuous at z0 .
1
Example 4.1. The function f (z) = 1+z 2
is discontinuous at z = i and z = −i.
• Suppose that f (z) = u(x, y) + iv(x, y) and z0 = x0 + iy0 . Then the complex
function f is continuous at the point z0 if and only if both real functions u
and v are continuous at the point (x0 , y0 ).
Suppose that the function f (z) = u(x, y)+iv(x, y) is defined on a closed and bounded
region R in the complex plane. As with real functions, we say that the complex
functions f is bounded on R if there exists a real constant M > 0, such that |f (z)| ≤
M, for all z in R.

5. Differentiability
Let f : D → C, z0 and interior point in D. Then, f is differentiable at z0 if the
limit
f (z) − f (z0 )
lim
z→z0 z − z0
exists. The value of the limit is called the derivative of f at the point z0 and denoted
by f 0 (z0 ).

If we write h or ∆z as z − z0 , then we can write

f (z0 + ∆z) − f (z0 )
f 0 (z0 ) = lim
∆z→0 ∆z
Chain Rule:
Suppose f is differentiable at z0 and g is differentiable at f (z0 ). Then the composite
function g ◦ f is differentiable at z0 and
(g ◦ f )0 (z0 ) = g 0 (f (z0 ))f 0 (z0 ).
Question 5.1. Is there any difference between the differentiability in R2 and C?
The function f : C → C defined by f (z) = |z|2 is differentiable only at 0. We have
(using |z|2 = zz)
|z0 + h|2 − |z0 |2

z0 h + z0 h + hh h
= = z0 + z0 + h.
h h h
Note that the limit of the above exists as h → 0 iff z0 = 0.
However, looking f as R2 → R we have f (x, y) = x2 + y 2 . Observe that f is differen-
tiable everywhere on R2 .

Properties:
dc
• dz = 0, c a complex constant.

• dz
dz
= 1, d
dz
[cf (z)] = cf 0 (z).
9

• d
dz
[f (z) + g(z)] = f 0 (z) + F 0 (z)

• d
dz
= f (z)F 0 (z) + F (z)f 0 (z).
[f (z)F (z)]
0 0 (z)F (z)
• F (z) 6= 0, dz F (z) = F (z)f (z)−f
d f (z)
(F (z))2

• Chain Rule: Suppose f has derivative at z0 and g has derivative at f (z0 ).

Then F (z) = g[f (z)] has a derivative at z0 and F 0 (z0 ) = g 0 [f (z0 )]f 0 (z0 ).
A necessary condition

Suppose f is differentiable at z0 . Assume f (x + iy) = u(x, y) + iv(x, y), z0 = x0 + iy0 .

Now, lim f (z)−f
z−z0
(z0 )
exists and equals f 0 (z0 ) whichever way z approaches z0 .
z→z0
(1) Horizontally: z = x + iy0 approaches z0 as x approaches x0 . So,
u(x, y0 ) − u(x0 , y0 ) v(x, y0 ) − v(x0 , y0 )
f 0 (z0 ) = lim + i lim , (5.1)
x→x0 x − x0 x→x0 x − x0
i.e. f 0 (z0 ) = ux (x0 , y0 ) + ivx (x0 , y0 ).
(2) Vertically:z = x0 + iy approaches z0 as y approaches y0 , and z − z0 = i(y − y0 ).
So,
u(x0 , y) − u(x0 , y0 ) v(x0 , y) − v(x0 , y0 )
f 0 (z0 ) = lim + i lim , (5.2)
y→y0 i(y − y0 ) y→y0 i(y − y0 )
i.e., f 0 (z0 ) = −iuy (x0 , y0 ) + vy (x0 , y0 ).
From (5.1) and (5.2) we have at z0 = (x0 , y0 ),
ux = vy , vx = −uy .
Theorem 5.2. Suppose that f (z) = f (x + iy) = u(x, y) + iv(x, y) is differentiable at
z0 = x0 + iy0 . Then, the partial derivatives of u and v exist at the point z0 = (x0 , y0 )
and
ux = vy , uy = −vx , at z0 = x0 + iy0 (Cauchy-Riemann(CR) Equations)
Example 5.3.
• CR equations are necessary, but not sufficient conditions. A function may not
be differentiable, even if CR equations are satisfied.
Example 5.4. Consider (
z2
z
, if z 6= 0
f (z) = (5.3)
0, if z = 0.
Then
x3 −3xy 2 3 2
f (z) − f (0) z 2 /z x2 +y 2
+ i yx−3x
2 +y 2
y
= =
z−0 z x + iy
If z approaches 0 along the x-axis, then f (z)−f
z−0
(0)
= x−0x
→ 1.
f (z)−f (0)
If z approaches 0 along the line y = x, then z−0 = −x−ix x+ix
→ −1.
10

So, f is not differentiable at z = 0. Here,

x3 − 3xy 2 y 3 − 3x2 y
u= , v =
x2 + y 2 x2 + y 2
at nonzero z and 0 at 0. They satisfy CR equations at z = 0.
At (0, 0), ux = vy = 1, uy = −vx = 0.
Theorem 5.5. Let f = u + iv, defined on B(z0 , r), be such that
• ux , uy , vx , vy exist on B(z0 , r).
• ux , uy , vx , vy are continuous at z0 .
• u and v satisfies CR equations: ux = vy , vy = −ux Then f is differentiable at
z0 and f 0 (z0 ) = ux (z0 ) + ivx (z0 ).
Proof Proved in detail in class lecture.

Example 5.6. Show that the function f (z) = z1 , is differentiable everywhere except
z = 0. Let us write the function as
1
f (x + iy) = (5.4)
x + iy
x − iy
= 2 . (5.5)
x + y2
Hence the real and imaginary component functions are
x
u(x, y) = 2 ,
x + y2
y
v(x, y) = − .
x2
+ y2
Except at the point x = y = 0, these functions are differentiable, and their partial
derivatives satisfy:
∂u −x2 + y 2 ∂v
= 2 2 2
=
∂x (x + y ) ∂y
∂v 2xy ∂u
= 2 = − .
∂x x + y 2 )2 ∂y
Here ux , uy , vx , vy exist and continuous at all points except at z = 0. So f is differen-
tiable everywhere except z = 0.
Cauchy-Riemann equation in polar form:
Suppose differentiable f is given in polar coordinates:
w = f (reiθ ) = u(r, θ) + iv(r, θ).
Then the Cauchy-Riemann equation in polar form is
rur = vθ , uθ = −rvr .
Formulated in class lecture.
11

6. Analytic Functions/Regular/Holomorphic
Definition 6.1. A function f (z) is said to be analytic in an open set of the complex
plane if f (z) has a derivative at each point of that set.
Remark 6.2.

• If we speak of a function f that is analytic in a set S that is not open, we

mean that f is analytic in an open set that contains S.

• f is analytic at a point z0 if its is analytic in a neighborhood of z0 .

Example 6.3.
1
i) f (z) = z
(z 6= 0) is analytic everywhere except at z = 0.

ii) f (z) = |z|2 is not analytic at any point as f 0 (z) exists only at z = 0 but not
at any other point other in any neighborhood of 0.
• If a function f fails to be analytic at a point z0 but is analytic at some point
in every neighborhood of z0 then z0 is a singular point.

• If f (z) is analytic everywhere in the complex plane, it is called entire.

Example 6.4.

1
(1) z
is analytic except at z = 0, so the function is singular at the point.

(2) The functions z n , n a non negative integer are entire functions.

7. Harmonic Functions
The real part and the imaginary part of a complex function f (z) = u(x, y)+iv(x, y)
that is analytic in a domain D are solution of the Laplace equation
∂ 2u ∂ 2u
+ =0
∂x2 ∂y 2
and
∂ 2v ∂ 2v
+ =0
∂x2 ∂y 2
in D and have continuous partial derivatives in D. We call u and v the harmonic
functions.

If two give functions u and v are harmonic in a domain D and their first derivatives
satisfy the C-R equations, i.e., ux = vy , uy = −vx throughout D, v is said to be a
harmonic conjugate of u.
Remark 7.1. If v is a harmonic conjugate of u in some domain it is not in general
true that u is a harmonic conjugate of u.
12

8. Exponential, Trigonometric and Hyperbolic Functions

We begin with the definition of the function ez for all z ∈ C. Guided by the
properties of the exponential function ex for all x ∈ R, we expect the function ez to
be an entire function such that ez1 ez2 = ez1 +z2 , z1 , z2 ∈ C, and e0 = 1.
To do so let z = x + iy ∈ C. Then
ez = ex+iy = ex c(y) + iex s(y),
where c(y) and s(y) are to be determined. First we note that
1 = e0 = c(0) + is(0).
Hence c(0) = 1 and s(0) = 0. Let
u(x, y) = ex c(y)
and
v(x, y) = ex s(y)
for all z ∈ C. Now using C-R equations we get a system ordinary differential equations
(
s0 = c,
(8.1)
c0 = −s.
Thus we get
(
s00 + s = 0,
(8.2)
s(0) = 0.
Therefore s(y) = sin y and c(y) = cos y for all y ∈ R. This gives Euler’s identity to
the effect that
eiy = cos y + i sin y, y ∈ R,
and the very important formula
ez = ex (cos y + i sin y).
d z
Therefore dz
e = ez .
Example 8.1. Find all zeroes of ez = 1.
Solved in class.
• ez+2πi = ez , z ∈ C.
Definition 8.2. Foe every complex number z, we define cos z and sin z by
eiz + e−iz
cos z =
2
and
eiz − e−iz
sin z = .
2i
For every complex number z we define tan z, sec z, csc z and cot z as in calculus. The
derivatives and the trigonometric identities for these six functions are also the same.
13

Definition 8.3. For every complex number z, we define cosh z and sinh z by
ez + e−z
cosh z =
2
and
ez − e−z
sinh z = .
2
9. Logarithms, Branches and Cuts
Before talking about the Logarithms we mention an important notion related to
the argument of acomplex number. For every nonzero complex number z, if θ is a
value or branch of arg z, then
arg z = θ + 2kπ, k = 0, ±1, ±2, ±3, ...
Let τ be a real number. The place Cτ given by Cτ = {(r, θ) : r > 0, τ < θ ≤ τ +2π} is
known as the cut plane along the branch cut {(r, τ ) : r > 0}. The branch of arg z that
lies in (τ, τ + 2π] is denoted by argτ z. The principal branch is then arg−π z = Arg z.

A very interesting function in complex analysis is the logarithmic function

w = log z,
which we will introduce. As in calculus we want to define
w = log z ⇐⇒ z = ew .
As usual, we write w = u + iv and z = x + iy. Since |ew | = eu 6= 0, we see that z 6= 0.
So, we can write z in polar form as
z = reiθ .
Therefore
reiθ = eu eiv .
We get r = eu or u = ln r. We also get
v = θ = arg z.
So, we have the formula
log z = ln|z| + iarg z, z 6= 0.
Remark 9.1. For every nonzero complex number z,
log z = ln|z| + i(θ + 2kπ), k = 0, ±1, ±2, ±3, ...,
where θ is any value or branch of arg z.
Definition 9.2. Let τ ∈ R. Then for every nonzero complex number z, we define the
branch logτ z = ln|z| + iargτ z. If we let τ = −π then we write Logz for log−π z and
we call Logz the principal logarithm of z.
Thus it is important to note that
Logz = ln|z| + iarg−π z = ln|z| + iArgz, z 6= 0.
Example 9.3. Compute log(1 + i), Log(1 + i) and logπ (1 + i).
14

Solved in class.

Theorem 9.4. Let τ ∈ R. Then w = f (z) = logτ z is holomorphic on the domain

C◦τ given by
C◦τ = {(r, θ) : r > 0, τ < θ < τ + 2π},
and
d 1
(logτ z) = , z ∈ C◦τ .
dz z
Proof Let z0 ∈ C◦τ . Then we need to prove that f is differentiable at z0 . Let
w0 = logτ z0 . Then we need to prove that
f (z) − f (z0 ) w − w0 1
lim = lim = .
z→z0 z − z0 z→z 0 z − z0 z0
But w = logτ z ⇐⇒ z = ew and
ew − ew0 z − z0
ew0 = lim = lim .
w→w0 w − w0 w→w0 w − w0

Note that
f (z) = logτ z = ln|z| + iargτ z → ln|z0 | + iargτ z0 = logτ z0
as z → z0 . Therefore z → z0 ⇒ w → w0 . Moreover,
z 6= z0 ⇒ w 6= w0
because
w = w0 ⇒ z = ew = ew0 = z0 .
Therefore
w − w0 1 1 1
lim = lim z−z0 = w
=
z→z0 z − z0 z→z0
w−w0
e 0 z0
as required.

Example 9.5. Find the domain on which the function

w = f (z) = Log(3z − i)
is holomorphic. Compute f 0 (z).
Since f is the composition of Log and 3z − i, f is holomorphic at all points z unless
3z − i ∈ (−∞, 0]. But it is easy to check that
x i
3z − i ∈ (−∞, 0] ⇐⇒ z = + , x ≤ 0.
3 3
(9.1)
i
Thus, f is holomorphic on the domain C − {x + 3
: x ≤ 0}.
Example 9.6. Find a branch of w = f (z) = log(z 3 − 2) that is holomorphic at z = 0.
15

Let τ ∈ R and take the branch logτ . Then f is the composition of logτ and the
function z 3 − 2. So f is holomorphic at z = 0 only if −2 is not on the branch cut
{(r, τ ) : r > 0}. So we can use any branch where the branch cut {(r, τ ) : r > 0} is
not equal to (−∞, 0).
From the properties of arg one can derive the following properties of logarithm func-
tions.

For any z1 , z2 ∈ C, with z1 , z2 6= 0

i) log(z1 z2 ) = log(z1 ) + log(z2 )

ii) log( zz12 ) = log(z1 ) − log(z2 ).

Example 9.7. Let z1 = −2i, z2 = −i. Then check log(z1 z2 ) = log(z1 ) + log(z2 ) but
Log(z1 z2 ) 6= Log(z1 ) + Log(z2 ).
10. Complex-valued Function
First consider derivatives and definite integrals of complex-valued functions w of a
real variable t. We can write it as w(t) = u(t) + iv(t) where u and v are real valued
functions of t. Then w0 (t) = u0 (t) + iv 0 (t), provided u0 (t), v 0 (t) exists at t.
• Rules of differentiation of sums and products holds.
• Mean value theorem for derivatives no longer holds. For example, f (z) =
ez , z = x + iy. Here f 0 (c) 6= 0 but f (z1 ) − f (z2 ) = 0 for z2 = z1 + 2πi.
Rb Rb Rb
Definition 10.1. a w(t)dt = a u(t)dt + i a v(t)dt when the individual integral
exists.
Rb Rb
• Re a w(t)dt = a Re(w(t))dt
Rb Rb
• Im a w(t)dt = a Im(w(t))dt
Improper integrals of w(t) over unbounded intervals are defined in a similar manner.
Definition 10.2. A real valued function of a real variable t is said to be piecewise
continuous on [a, b] if it is continuous everywhere on [a, b] except possibly for a finite
number of points where only the right hand limit is required at a and only the left
hand limit is required at b.
If w(t) = u(t) + iv(t), a ≤ t ≤ b, then w is said to be piecewise continuous if both u
and v are piecewise continuous on [a, b].
Rb
If w is piecewise continuous on [a, b] then a w(t)dt exists.
Rb Rc Rb
• a w(t)dt = a w(t)dt + c w(t)dt.
Rb Rb
• z0 a w(t)dt = a z0 w(t)dt, where z0 is a complex constant.
Rb Rb Rb
• a w1 (t)dt + a w2 (t)dt = a (w1 (t) + w2 (t))dt.
Rb Ra
• a w(t)dt = − b w(t)dt.
Suppose that the functions w(t) = u(t)+iv(t) and W (t) = U (t)+iV (t) are continuous
on [a, b].
(1) If W 0 (t) = w(t) then U 0 (t) = u(t), V 0 (t) = v(t).
Rb
(2) a w(t)dt = W (b) − W (a).
16
Rb
Suppose a < b and a w(t)dt 6= 0.
Rb Rb Rb
Let a w(t)dt = r0 eiθ0 . So r0 = e−iθ0 a w(t)dt = a e−iθ0 w(t)dt. Since r0 is real,
Z b Z b Z b
−iθ0 −iθ0
r0 = e w(t)dt = Re e w(t)dt = Re(e−iθ0 w(t))dt.
a a a
−iθ0 −iθ0
Re(e w(t)) ≤ |e w| = |w|.
Rb Rb
Then a w(t)dt ≤ a |w(t)|dt, a < b.
With minor modification of the above proof we also get
Z ∞ Z ∞
w(t)dt ≤ w(t)dt
a a

11. Contour Integration

Contour integration is a powerful technique, based on complex analysis, that
allows us to calculate certain integrals that are otherwise difficult or impossible to do.
Contour integrals have important applications in many areas of physics, particularly
in the study of waves and oscillations

11.1. Contour Integrals.

You have previously studied what it means to take the integral of a real function.
To recap: if f (x) is a real function, the integral from x = a to x = b is defined by
dividing the interval into N segments, and evaluating the sum of f (x)∆x on each
segment, in the limit where N goes to infinity:
Z b N
X b−a
f (x)dx = lim ∆xf (xn ), where xn = a + n∆x, ∆x = . (11.1)
a N →0
n=0
N
Now consider the case where f is a complex function of a complex variable. The
straight-foward way to define the integral of f (z) is by an analogous expression like
this:
XN
lim ∆zf (zn ) (11.2)
N →0
n=0
However, since f takes complex inputs, the values of zn need not lie along the real
line. In general, the complex numbers zn form a set of points in the two-dimensional
complex plane. We can imagine chaining together a sequence of points z1 , z2 , ..., zN ,
which are separated by displacements ∆z1 , ∆z2 , ∆z3 , ..., ∆zN −1 such that
z2 = z1 + ∆z1
z3 = z2 + ∆z2 ,
z4 = z3 + ∆z3 ,
. = .
. = .
. = .
zN = zN −1 + ∆zN −1 . (11.3)
17

Figure shown in class. Now, the sum we are interested in is

N
X −1
∆zn f (zn ) = ∆z1 f (z1 ) + ∆z2 f (z2 ) + ... + ∆zN −1 f (zN −1 ).
n=1
Suppose we fix the end-points z1 and zN , and take the limit N → ∞, so that each
displacement ∆zn becomes infinitesimal. Then the sequence of points z1 , z2 , ..., zN
turns into a continuous trajectory in the complex plane, with a certain starting point
and end-point. Such a trajectory is called a contour, and can be denoted by an
abstract symbol, such as Γ. Now we can define a contour integral over Γ, like this:
Z N
X −1
f (z)dz = lim ∆zn f (zn ).
Γ N →∞
n=1
The symbol Γ in the subscript of the integral sign indicates that the integral is to take
place over the contour Γ. It is always necessary, when defining a contour integral, to
specify the contour to integrate over. This is roughly analogous to how we need to
specify the ends of the integration range, in order to properly define a real definite
integral. In the complex case, we must specify an entire trajectory. It is important to
note that the contour Γ specifies a direction. In fact, if we integrate along the curve
in the opposite direction, the value of the contour integral switches sign.
11.1.1. Contour Integral along parametric curve.
Simple contour integrals can be calculated by parameterizing the contour. Sup-
pose we have a contour integral Z
dzf (z),
Γ
where f is a complex function of a complex variable and Γ is a given contour. As
previously discussed, we can describe a trajectory in the complex plane by a complex
function of a real variable, z(t) :
Γ = {z(t)|t1 < t < t2 },
where t ∈ R, z(t) ∈ C. The real numbers t1 and t2 specify two complex numbers,
z(t1 ) and z(t2 ), which correspond to the endpoints of the contour. The rest of the
contour is given by the values of z(t) between these two end-points. Assuming we
are able to parameterize Γ in this way, we can express the complex displacement dz
in the contour integral by
dz
dz → dt .
dt
This allows us to express the contour integral Γ as a definite integral over t.
Z Z t2
dz
dzf (z) = dt f (z(t)) .
Γ t1 dt
The resulting definite integral can then be evaluated using standard integration tech-
niques.
Example 11.1. Calculate the following contour integral:
Z
dzz n , n ∈ Z,
Γ[θ1 ,θ2 ]
18

where the trajectory Γ[θ1 , θ2 ] is shown in the figure in class. We can parameterize the
contour by defining the function z(θ) as follows:
Γ[θ1 , θ2 ] = {z(θ) : θ1 ≤ θ ≤ θ2 },
where z(θ) = Reiθ . Then the contour integral can be converted into an integral over
the real parameter θ :
Z Z θ2
n dz
dzz = dθz n
Γ[θ1 ,θ2 ] θ1 dθ
Z θ2
= dθ(Rn einθ )(iReiθ )
θ1
Z θ2
n+1
= iR dθei(n+1)θ (11.4)
θ1

To proceed, there are two distinct cases that we need to handle separately. Firstly,
if n 6= −1, then we can evaluate the integral on the last line as follows
Z θ2 i(n+1)θ theta2
i(n+1)θ e
dθe =
θ1 i(n + 1) θ1
ei(n+1)θ2 − ei(n+1)θ1
= (11.5)
i(n + 1)
However, this doesn’t apply if n = −1, since the factor of n + 1 in the denominator
would vanish. Instead, in this case, the integrand is identically one, so we do the
integral in a different way:
Z θ2 Z θ2
i(n+1)θ
dθe = dθ
θ1 θ1
= θ2 − θ1 (11.6)
Putting these two cases together, we arrive at the result
(
i(θ2 − θ1 ), if n = −1
Z
dzz n = n+1 ei(n+1)θ2 −ei(n+1)θ1 (11.7)
Γ[θ1 ,θ2 ] iR i(n+1)
6 1.
if n =
The case where θ2 = θ1 + 2π is of particular interest. Here, Γ forms a complete loop,
and the above equation simplifies to
I (
2πi, if n = 1
z n dz = (11.8)
0, if n 6= −1.
Γ
H
Here, the special integration symbol is used to indicate that the contour integral is
taken over a loop. The value of this particular contour integral is independent ofR
(so long as R > 0). It is zero for all values of n except n = −1. For the special case
n = −1, the contour integral takes the value 2πi. This is an important result.
So we have seen that the integrals of complex-valued functions are defined on curves
in the complex plane rather than on just intervals of the real lines.
19

Definition 11.2. Arc A set of points γ : z = (x, y) in the complex plane is said to
be an arc or curve if
x = x(t), y = y(t) a ≤ t ≤ b
where x(t) and y(t) are continuous functions of the real parameter t. The set γis
described by z(t) where
z(t) = x(t) + iy(t), a ≤ t ≤ b
The point z(a) is called the initial point and the point z(b) is called the terminal point
of γ
Example 11.3. The polygonal line:
(
x + ix when 0 ≤ x ≤ 1
z= (11.9)
x+i when 1 ≤ x ≤ 2
Here x = x and y = y(x). Here x is the parameter t and z = z(t) = z(x) is the
simple arc.
Definition 11.4. An arc or curve C is called a simple arc (or curve), if it does not
cross itself, that is, C is simple if z(t1 ) 6= z(t2 ) whenever t1 6= t2 .
Example 11.5. The curve in Example 11.1 is a simple curve.
Definition 11.6. If an arc is simple except for the fact that z(b) = z(a), we say that
C is a simple closed curve or Jordan curve.
Example 11.7. i) The unit circle z = eiθ = cos θ + i sin θ, 0 ≤ θ ≤ 2π is a sim-
ple closed curve as z(0) = z(2π), oriented in the counter-clockwise direction.
ii) z = z0 + Reiθ , 0 ≤ θ ≤ 2π is similarly a simple closed curve a circle centred
at z0 with radius R.
iii) The arc z = e−iθ , 0 ≤ θ ≤ 2π is a simple closed curve , a unit circle, but not
same as the curve described in the above example. The set of points are the
same, but the circle is traversed in the clockwise direction.
Definition 11.8. Smooth Curves
A curve (or arc) is said to be smooth if it obeys the following three conditions
(1) z(t) has continuous derivative on the interval [a, b]
(2) z 0 (t) is never zero on (a, b)
(3) z(t) is a one to one function on [a, b]
If the first two conditions are met but z(a) = z(b), then it is called a smooth closed
curve.
The example of the circles are simple closed contour.
The boundary of a triangle or a rectangle taken in a specific direction is a simple
closed contour.
Definition 11.9. Contour:
A contour or a piecewise smooth arc. Its is an arc consisting of finite number of
smooth arcs joined end to end. Hence if z = z(t) = x(t) + iy(t) represents a contour,
then z(t) is continuous, whereas its derivative z 0 (t) is piecewise continuous. Also
20

z 0 (t) 6= 0 where it is defined.

If z(a) = z(b) then it is called a simple closed contour.

Example 11.10. (1) The polygonal line described earlier is a contour.
Positive Orientation
The direction of increasing values in the real parameter t corresponds to the positive
direction on a contour C. If the contour is closed the positive direction corresponds
to the counter-clockwise direction or the direction in which you would walk so that
the interior of the closed contour is always on your left.

Length of an Arc
The length of an arc is given by
Z b Z bp
0
L= |z (t)|dt = (x0 )2 + (y 0 )2 dt,
a a

where x(t) and y(t) can be thought of as parametric representations of the curve γ
which consists of a set of points in the cartesian (x, y) plane.
Theorem 11.11. Jordan Curve Theorem
A simple closed curve or simple closed contour divides the complex plane into two
sets, the interior which is bounded and the exterior which is unbounded.

12. Contour Integral

As we have discussed in the previous class that,

• z = z(t), a ≤ t ≤ b represents a contour C extending from a point z1 = z(a)

to a point z2 = z(b);
• f (z) be piecewise continuous on C, that is f (z(t)) is piecewise continuous on
the interval a ≤ t ≤ b;
The contour integral of f along C is defined as follows:
Z Z b
f (z)dz = f (z(t))z 0 (t)dt.
C a
0
Remark 12.1. Since C is a contour, z (t) is also piecewise continuous on the interval
a ≤ t ≤ b and so the above integral exists.
Associated with the contour C in the above integral is the contour −C, consisting
of the same set of points but with the order preserved so that the new contour extends
from the point z2 to the point z1 .
The contour −C has the parametric representation z = z(−t), −b ≤ t ≤ −a.

12.1. Properties of the contour integral.

R R
i) C
f (z)dz = − −C
f (z)dz
21

ii) Suppose that C consists of a contour C1 from z1 to z2 followed by a contour

C2 from z2 to z3 , the initial point of C2 being the final point of C1 .
Z Z Z
f (z)dz = f (z)dz + f (z)dz
C C1 C2
Some times the contour C is called the sum of its legs C1 and C2 and is
denoted by C1 + C2 . So
Z Z Z
f (z)dz = f (z)dz + f (z)dz
C1 +C2 C1 C2

R R
iii) z f (z)dz = z0
C 0 C
f (z)dz
R R R
iv) C
[f (z) + g(z)]dz = C
f (z)dz + C
g(z)dz
Now
Z Z b
f (z)dz = f (z(t))z 0 (t)dt
C a
Z b
≤ |f (z(t))z 0 (t)| dt. (12.1)
a
Suppose there exist an M > 0 such that |f (z)| ≤ M for all z ∈ C. So
Z Z b
f (z)dz ≤ M |z 0 (t)|dt = M L
C a
Rb
where L is the length of the contour given by a
|z 0 (t)|dt.
Remark 12.2. Suppose f is a continuous function of a real variable t defined on the
closed bounded interval [a, b]. Then it is well known that f is bounded on [a, b] and
there exists t0 ∈ [a, b] such that f (t0 ) = sup |f (t)|, that is |f (t)| ≤ |f (t0 )| for all
a≤t≤b
t ∈ [a, b]. So if f is continuous on the contour C : z = z(t), then f (z) = f (z(t)) is
continuous on [a, b] and consequently there exists an M > 0 such that |f (z(t))| ≤ M
for all t ∈ [a, b]. It now follows that such an M also exists when f is piecewise
continuous.
Example 12.3.
Find the value of the integral Z
I= zdz
C
when C is the right-handed half z = 2eiθ , − π2 ≤ θ ≤ π
2
of the circle |z| = 2 from
z = −2i to z = 2i.
Proof. z = 2eiθ so z = 2e−iθ and dθ
d
2eiθ = 2ieiθ . Then
Z Z π
2
zdz = 2e−iθ 2ieiθ dθ
C − π2
Z π
2
= 4i dθ = 4πi.
− π2
22

• When z is on the circle |z| = 2, then zz = |z|2 = 4 and so z = z4 . Hence
Z Z Z
dz 1 4 1
= dz = zdz = πi.
C z 4 Cz 4 C
• Show that C dz
R
z
= 2πi if C represents the circle z = reiθ , 0 ≤ θ ≤ 2π or
−π ≤ θ ≤ π.
Example 12.4. Let C1 denotes the contour OAB, where OA is the y-axis, 0 ≤ y ≤ 1,
AB is the line segment y = 1 with 0 ≤ x ≤ 1 and C2 is the segment OB of
the line y = x. Evaluate the integral on C1 and C2 respectively of the function
f (z) = y − x − i3x2 , where z = x + iy.

The leg OA has the parametric representation z = 0 + iy, 0 ≤ y ≤ 1 and hence on

OA, f (z) = y, 0 ≤ y ≤ 1.
So Z Z 1 Z 1
i
f (z)dz = yidy = i ydy = .
OA 0 0 2
On the leg AB, z = x + i, 0 ≤ x ≤ 1 and so
Z Z 1
1 i
f (z)dz = (1 − x − i3x2 )dx = − ..
AB 0 2 2
So
1−i
Z
f (z)dz = .
C1 2
R R
Now in C2 we have z = x + ix, 0 ≤ x ≤ 1. Hence C2 f (z)dz = OB f (z)dz =
R1
0
−i3x2 (1 + i)dx = 1 − i
Remark 12.5.

• The integrals of f (z) along the two paths C1 and C2 have different values even
though those are paths having the same initial and same final points.
• If C = C1 − C2 , then
−1 + i
Z Z Z Z
f (z)dz = f (z)dz = f (z)dz − f (z)dz = 6= 0.
C C1 −C2 C! C2 2
Let C denote an arbitrary smooth arc z = z(t), a ≤ t ≤ b from a fixed point z1 to
a fixed point z2 . Then
Z b b
[z(t)]2 [z(b)]2 − [z(a)]2 z 2 − z12
Z
0
I= zdz = z(t)z (t)dt = = = 2 .
C a 2 a 2 2
The above expression is also valid when C is a contour that is not necessarily smooth
since a contour consists of a finite number of smooth arcs Ck , k = 1, 2, ..., n−1, joined
end to end. More precisely, suppose that each Ck extends from zk to zk+1 . Then
n−1 Z n−1 2
zk+1 − zk2 zn2 − z12
Z X X
zdz = zdz = = ,
C k=1 Ck k=1
2 2
23

R z1 being the initial point of C and zn its final point. So when z1 = zn then
C
zdz = 0.
13. Antiderivative
Definition 13.1. Suppose a function f is continuous on a domain D and there exists
a function F such that F 0 (z) = f (z) for all z ∈ D. Then F is called an antiderivative
of f in the domain D.
Remark 13.2.

i) An antiderivative is necessarily analytic on the domain D.

ii) If F and G are two antiderivatives of a continuous function f on a domain

D, then F (z) − G(z) =constant for all z ∈ D, implies the analytic function
F (z) − G(z) has derivative 0 on D. Hence F (z) − G(z) is constant on D.
Theorem 13.3. Suppose that a function f is continuous on a domain D, then the
followings are equivalent:
a) f has an antiderivative F in D.

b) The integrals of f (z) along the contours lying entirely in D and extending
from any fixed point z1 to any other point z2 all have the same value.(i.e the
intergral is independent of the path in D.)

c) The integrals of f (z) around closed contours lying entirely in D all have value
0.
Example 13.4. The continuous functions f (z) = z 2 has an antiderivative F (z) =
z 3 /3 throughout the plane.
R 1+i 3
Hence 0 z 2 dz = z3 ]1+i0 = 32 (−1 + i) for every contour from z = 0 to z = 1 + i.

Example 13.5. The function 1/z 2 which is continuous every where except at the
origin has an antiderivative −1/z in the domain |z| > 0. Consequently
Z z2
dz 1 1
2
= − , z1 6= 0, z2 6= 0
z1 z z1 z2
Rfor dzany contour from z1 to z2 that doesn’t
iθ
pass through the origin. In particular
C z2
= 0 where C is the circle z = 2e , −π ≤ θ ≤ π about the origin.
Theorem 13.6. If a function
R is analytic at all points interior to and on a simple
closed contour C, then C f (z)dz = 0.
Definition 13.7. (Simply connected domain) A domain D is called simply connected
if every simple closed contour (within it) encloses points of D only.
A domain D is called multiply connected if it is not simply connected.
Example 13.8. (1) The set of points interior to a simple closed contour is a
simple connected domain.
24

(2) The sets C and D = {z : Rez > 0} are simply connected domains.

(3) The sets C∗ and the annulus A(a, b) = {z ∈ C : a < |z| < b} are not simply
connected domains.
Theorem 13.9. (An extension of Cauchy-Goursat)
If f is analytic in a simply connected domain D, then
Z
f (z)dz = 0
C
for every closed contour C lying in D.
Corollary 13.10. A function f which is analytic throughout a simple connected
domain D must have an anti derivative, F sayR in D. In this case for all the paths in
z
D joining two points z1 and z2 in D we have z12 f (z)dz = F (z1 ) − F (z2 ).
Example 13.11. If C0 is aR positively oriented circle of radius 0 centered at the
origin (for any > 0), then C0 z1 dz = 2πi.
Theorem 13.12. (Extension of Cauchy-Goursat Theorem to a multiply connected
domain)
Suppose that
(1) C is a simply closed contour in counter clockwise direction.
(2) Ck , k = 1, 2, 3, ..., n denotes a finite number of simple closed contours all
described in the clockwise direction that are interior to C and whose interiors
have no points in common.
If a function f is analytic throughout the closed region consisting of all points within
and on C except for the points interior to each Ck , then
Z n Z
X
f (z) + f (z)dz = 0.
C k=1 Ck

Proof Figure given in class.

By Cauchy Gourset Theorem,
Z Z
f (z)dz + f (z)dz = 0 + 0 = 0,
Γ1 Γ2

which gives the result.

Example 13.13. Z
dz
= 0,
Bz 2 (z 2+ 16)
where B consists of the annular region between the circle |z| = 2 in counter clock-
wise(positive) direction and the circle |z| = 1 in the negative direction. Then it can be
seen that z2 (zdz
2 +16) fails to be analytic at the points z = 0 and z = ±4i. These points

lies outside our annular region. Hence from above theorem the integral is 0.
Corollary 13.14. Let C1 be a positively oriented simple closed contour. Then, C1
breaks the complex plane up into two regions: the interior of C1 and the exterior of C1
25

(by the Jordan curve theorem). Let C2 be a positively oriented simple closed contour
entirely inside the interior of C1 . If f is analytic in between and on C1 and C2 , then
Z Z
f (z)dz = f (z)dz.
C1 C2