Unit 3: Complex Numbers

Real Number System —

The set of real number (ℝ) is a collection of elements with two operations ‘+’ (addition) and
‘.’ (scalar multiplication) such that it satisfies the following axioms:
i) 𝑥 + 𝑦 ∈ ℝ; 𝑥, 𝑦 ∈ ℝ ∀ 𝑥, 𝑦 ∈ ℝ
ii) 𝑥 + (𝑦 + 𝑧) = (𝑥 + 𝑦) + 𝑧; 𝑥. (𝑦. 𝑧) = (𝑥. 𝑦). 𝑧 ∀ 𝑥, 𝑦, 𝑧 ∈ ℝ
iii) 𝑥 + 𝑦 = 𝑦 + 𝑥; 𝑥. 𝑦 = 𝑦. 𝑥 ∀ 𝑥, 𝑦 ∈ ℝ
iv) For each 𝑥 ∈ ℝ ∃ 0 ∈ ℝ such that 𝑥 + 0 = 𝑥 = 0 + 𝑥 (0 is known as additive identity).
v) For each 𝑥 ∈ ℝ ∃ 𝑦 ∈ ℝ such that 𝑥 + 𝑦 = 0 = 𝑦 + 𝑥 (such 𝑦 is known as additive
inverse of 𝑥 and is denoted by ‘−𝑥’)
vi) For each 𝑥 ∈ ℝ ∃ 1 ∈ ℝ such that 𝑥. 1 = 𝑥 = 1. 𝑥 (such 1 is known as multiplicative
identity)
vii) For each 𝑥 ∈ ℝ ∃ 𝑦 ∈ ℝ such that 𝑥. 𝑦 = 1 = 𝑦. 𝑥 (such 𝑦 is known as multiplicative
inverse of 𝑥 and is denoted by ‘𝑥 −1 ’)
Geometrically ℝ along with the above axioms can be represented by a straight line centred
at 0 and all 𝑥 (i.e., positive numbers) towards right and all −𝑥 (i.e., negative numbers)
towards left. Thus, if 𝑥 ∈ ℝ then 𝑥 < 0, 𝑥 = 0, 𝑥 > 0
Let us consider an equation 𝑥 2 + 1 = 0 ⇒ 𝑥 2 = −1 ⇒ 𝑥 = ±√−1
For simplicity let 𝑥 = √−1
Now, does 𝑥 ∈ ℝ?
Because 𝑥 can not fill any of the 3 categories 𝑥 < 0, 𝑥 = 0, 𝑥 > 0
Assume 𝑥 = 0, 𝑥 2 = −1 ⇒ 0 = −1 (not possible)
Assume 𝑥 > 0
Consider a right-angled triangle ABC with ̅̅̅̅ 𝐴𝐵 = 1 ̅̅̅̅
𝐵𝐶 = √−1
2
Then 𝐴𝐶̅̅̅̅ 2 = ̅̅̅̅
𝐴𝐵2 + ̅̅̅̅
𝐵𝐶 2 = 12 + √−1 = 1 − 1 = 0 (not possible)
Assume 𝑥 < 0 ⇒ −𝑥 > 0 ⇒ −√−1 > 0
Consider a right-angled triangle ABC with 𝐴𝐵 ̅̅̅̅ = 1 𝐵𝐶
̅̅̅̅ = −√−1
2
̅̅̅̅ 2 = ̅̅̅̅
Then 𝐴𝐶 𝐴𝐵2 + ̅̅̅̅
𝐵𝐶 2 = 12 + (−√−1) = 1 − 1 = 0 (not possible)
Thus, the solution of the equation 𝑥 2 + 1 = 0 is neither 𝑥 > 0 not 𝑥 = 0 and not 𝑥 < 0
Thus, the solution of 𝑥 2 = −1 is not a real number.
Then √−1 is given a new name 𝑖 after years of debate and discussion.
Thus, we need another collection of elements containing 𝑖 with some operations.
Now, consider the following equation 𝑥 2 + 4𝑥 + 13
−4±√4 2 −4.1.13 −4±√−36
Then 𝑥 = = = −2 ± 3𝑖
2 2
Therefore, we need not only number like 𝑖 but also 𝑎 + 𝑖𝑏 (𝑎, 𝑏 ∈ ℝ) such number is called
complex numbers. Thus, complex number is an expression of the form 𝑎 + 𝑖𝑏 where 𝑎, 𝑏
are real numbers. The set of complex number is denoted by ℂ.
Two complex numbers 𝑎 + 𝑖𝑏 and 𝑐 + 𝑖𝑑 are equal if and only if 𝑎 = 𝑐 and 𝑏 = 𝑑. Clearly
ℝ ⊂ ℂ. Any real number is also a complex number. For 𝑥 ∈ ℂ.
Complex Plane —
Elements in the complex number system are observed as points on the plane. This plane is
called the complex plane. Thus, a complex number 𝑎 + 𝑖𝑏 is sometimes denoted by (𝑎, 𝑏)
where 𝑎 is the real part and 𝑏 is the imaginary part.
1|DRC
The set of all points 𝑥 + 𝑖. 0 is called real axis and the set of all point 0 + 𝑖. 𝑦 is called
imaginary axis. The point 0 + 𝑖. 0 is common to both the axis is called the origin.
Operations on Complex Number —
In order to make complex number system we need two operations:
Addition: 𝑧1 = 𝑎1 + 𝑖𝑏1 , 𝑧2 = 𝑎2 + 𝑖𝑏2 then 𝑧1 + 𝑧2 = (𝑎1 + 𝑎2 ) + 𝑖(𝑏1 + 𝑏2 )
Multiplication:
Case I: Product of a real number and a complex number
Let 𝑐 ∈ ℝ and 𝑧 = 𝑎 + 𝑖𝑏 ∈ ℂ then 𝑐𝑧 = 𝑐𝑎 + 𝑖𝑐𝑏
Case II: Product of two complex numbers
Since we are developing a number system of complex numbers. So, product of two complex
numbers needs to be defined.
Let 𝑧1 = 𝑎1 + 𝑖𝑏1 , 𝑧2 = 𝑎2 + 𝑖𝑏2 then 𝑧1 𝑧2 = (𝑎1 + 𝑖𝑏1 )(𝑎2 + 𝑖𝑏2 )
= (𝑎1 𝑎2 − 𝑏1 𝑏2 ) + 𝑖(𝑎2 𝑏1 + 𝑎1 𝑏2 )
Now, subtraction and division are inverse operation of addition and multiplication
respectively i.e., 𝑧1 − 𝑧2 = (𝑎1 − 𝑎2 ) + 𝑖(𝑏1 − 𝑏2 )
𝑧1 𝑎 +𝑖𝑏
= 1 1
𝑧2 𝑎2 +𝑖𝑏2
(𝑎1 +𝑖𝑏1 )(𝑎2 −𝑖𝑏2 )
= (𝑎
2 +𝑖𝑏2 )(𝑎2 −𝑖𝑏2 )
(𝑎1 𝑎2 +𝑏1 𝑏2 )+𝑖(𝑎2 𝑏1 −𝑎1 𝑏2 )
=
𝑎2 2 +𝑏2 2
𝑎1 𝑎2 +𝑏1 𝑏2 𝑎2 𝑏1 −𝑎1 𝑏2
= +𝑖
𝑎2 2 +𝑏2 2 𝑎2 2 +𝑏2 2
Basic Properties of Complex Numbers —
If 𝑧 = 𝑎 + 𝑖𝑏 is a complex number then 𝑎 is called the real part of 𝑧 and 𝑏 is the imaginary
part of 𝑧 i.e., 𝑎 = Re(𝑧) and 𝑏 = Im(𝑧).
Also 𝑧̅ = 𝑎 − 𝑖𝑏 is called the complex conjugate of 𝑧.
Since 𝑎 − 𝑖𝑏 can be located by the point (𝑎, −𝑏) i.e., location of 𝑎 − 𝑖𝑏 in complex plane is
same as the location of (𝑎, −𝑏) in the real plane.
1. 𝑧 + 𝑧̅ = 2Re(𝑧)
2. 𝑧 − 𝑧̅ = 2𝑖(𝐼𝑚(𝑧))
̅̅̅̅) = 𝑧
3. (𝑧̅
4. 𝑧̅̅̅̅̅̅̅̅̅
1 + 𝑧2 = 𝑧̅1 + 𝑧̅2
5. 𝑧 = 𝑧̅ iff 𝑧 is real i.e., Im(𝑧) = 0
6. Re(𝑧) = Re(𝑧̅)
7. |𝑧| = √𝑎2 + 𝑏 2 , |𝑧| is the distance of the complex number 𝑧 = 𝑎 + 𝑖𝑏 from the origin of
the complex plane.
8. 𝑧. 𝑧̅ = 𝑎2 + 𝑏 2 = |𝑧|2
9. |𝑧1 + 𝑧2 | ≤ |𝑧1 | + |𝑧2 |
Polar Form of Complex Number —
Another way of representing complex number is polar form. We know that any pair of real
number (𝑥, 𝑦) can be transformed to (𝑟, 𝜃) coordinate system by setting 𝑥 = 𝑟𝐶𝑜𝑠𝜃, 𝑦 =
𝑟𝑆𝑖𝑛𝜃 then 𝑧 = 𝑥 + 𝑖𝑦
= 𝑟 cos 𝜃 + 𝑖𝑟 sin 𝜃
= 𝑟(cos 𝜃 + 𝑖 sin 𝜃)
2|DRC
 = 𝑟𝑒 𝑖𝜃 (by Euler’s theorem)
Therefore, |𝑧| = √𝑥 2 + 𝑦 2
= √(𝑟 2 𝑐𝑜𝑠 2 𝜃 + 𝑟 2 𝑠𝑖𝑛2 𝜃)
=𝑟
So, 𝑟 is the |𝑧| and θ is called the argument of 𝑧 and it is written as arg(𝑧). Thus, argument
of a complex number is just the angle it makes with positive real axis.
The argument of 𝑧 is not unique because we know that
sin 𝜃 = sin(𝜃 + 2𝑛𝜋); 𝑛 = 0, ±1, ±2, ±3, ….
cos 𝜃 = cos(𝜃 + 2𝑛𝜋); 𝑛 = 0, ±1, ±2, ±3, ….
Therefore, arg(𝑧) = {𝜃, 𝜃 − 2𝜋, 𝜃 + 2𝜋, 𝜃 − 4𝜋, 𝜃 + 4𝜋, … . } i.e., arg(𝑧) is the collection
of angles θ that satisfies 𝑧 = 𝑟𝑒 𝑖𝜃 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃)
Therefore, arg(𝑧) = {𝜃: 𝑧 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃)}
Principal Argument —
The principal argument of 𝑧 is the unique argument that lies on (0,2𝜋] or (−𝜋, 𝜋]
Now, let us consider the following cases:
Case I: If 𝑧 lies in the first quadrant.
𝑦 2
Let 𝑧 = 2 + 2𝑖 then tan 𝜃 = = = 1
𝑥 2
−1 𝜋
⇒θ= 𝑡𝑎𝑛 1 =
4
0
⇒ θ = 45 < 2π
Therefore, principal arg(𝑧) = 450
Case II: If 𝑧 lies in the second quadrant
Let 𝑧 = −2 + 2𝑖 which can be imagined as the rotation of 𝑧 = 2 + 2𝑖 by 900
Hence, θ = 450 + 900 = 1350 < 2π
Therefore, principal arg(𝑧) = 1350
Case III: If 𝑧 lies in the third quadrant
Let 𝑧 = −2 − 2𝑖
When we say angle made by 𝑧 with positive 𝑥-axis then we have 2 values θ and α. Also 𝑧
= −2 − 2𝑖 can be imagined as the rotation of 𝑧 = 2 + 2𝑖 by 1800
Hence, θ = 450 + 1800 = 2250 < 2π
Again, α = −(𝜋 − 450 ) = −1350 ∈ (−𝜋, 𝜋]
Therefore, principal arg(𝑧) is 2250 or −1350
Case IV: If 𝑧 lies in the fourth quadrant
Let 𝑧 = 2 − 2𝑖
When we say angle made by 𝑧 with positive 𝑥-axis then we have 2 values θ and α. Also 𝑧 =
2 − 2𝑖 can be imagined as the rotation of 𝑧 = 2 + 2𝑖 by 2700
Hence, θ = 450 + 2700 = 3150 < 2π
Again, α = −450 ∈ (−𝜋, 𝜋]
Therefore, principal arg(𝑧) is 3150 or −450
Question 1 – Find modulus and principal argument of
i) 1 − 𝑖
ii) 3 + 3√3𝑖
iii) −√3 − 𝑖
3|DRC
Solution – i) We have 𝑧 = 1 − 𝑖 which is in the form 𝑧 = 𝑥 + 𝑖𝑦 where 𝑥 = 1 and 𝑦 = −1
Then |𝑧| = √𝑥 2 + 𝑦 2 = √12 + (−1)2 = √2
Now, let us consider the angle made by the coordinate (1, −1) in anticlockwise direction
with positive real axis as α and the angle made by the coordinate (1, −1) in clockwise
direction with real axis as θ.
𝑦 −1 𝜋
Thus, α = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 1 =
𝑥 1 4
𝜋 7𝜋 0
Now, θ = 2𝜋 − 𝛼 = 2𝜋 − = or 315
4 4
Therefore, principal arg(𝑧) = 3150
ii) We have 𝑧 = 3 + 3√3𝑖 which is in the form 𝑧 = 𝑥 + 𝑖𝑦 where 𝑥 = 3 and 𝑦 = 3√3
2
Then |𝑧| = √𝑥 2 + 𝑦 2 = √32 + (3√3) = √36 = 6
Now, let us consider the angle made by the coordinate (3,3√3) in clockwise direction with
real axis as θ.
𝑦 3√3 𝜋
Thus, θ = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 √3 =
𝑥 3 3
𝜋 0
Therefore, principal arg(𝑧) = or 60
3
iii) We have 𝑧 = −√3 − 𝑖 which is in the form 𝑧 = 𝑥 + 𝑖𝑦 where 𝑥 = −√3 and 𝑦 = −1
2
Then |𝑧| = √𝑥 2 + 𝑦 2 = √(−√3) + (−1)2 = √4 = 2
Now, let us consider the angle made by the coordinate (−√3, −1) in clockwise direction
with negative real axis as α and the angle made by the coordinate (−√3, −1) in clockwise
direction with positive real axis as θ
𝑦 −1 1 𝜋
Thus, α = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 =
𝑥 −√3 √3 6
𝜋 7𝜋 0
Now, θ = 𝜋 + 𝛼 = 𝜋 + = or 210
6 6
Therefore, principal arg(𝑧) = 2100
Some Important Relations —
1. If 𝑧1 = 𝑟1 𝑒 𝑖𝜃1 , 𝑧2 = 𝑟2 𝑒 𝑖𝜃2 then 𝑧1 𝑧2 = 𝑟1 𝑟2 𝑒 𝑖(𝜃1+𝜃2) . Thus, |𝑧1 𝑧2 | = 𝑟1 𝑟2 = |𝑧1 ||𝑧2 | and
arg(𝑧1 𝑧2 ) = 𝜃1 + 𝜃2 = arg(𝑧1 ) + arg(𝑧2 )
2. If 𝑧1 = 𝑧2 = 𝑧, |𝑧 2 | = |𝑧|2 . Similarly, arg(𝑧 2 ) = 2 arg(𝑧)
3. In general, |𝑧1 𝑧2 … . 𝑧𝑛 | = |𝑧1 ||𝑧2 | … . |𝑧𝑛 |
arg(𝑧1 𝑧2 … . 𝑧𝑛 ) = arg(𝑧1 ) + arg(𝑧2 ) +. … + arg(𝑧𝑛 )
|𝑧 𝑛 | = |𝑧|𝑛
arg(𝑧 𝑛 ) = 𝑛 arg(𝑧)
Example 1 – Determine the modulus and principal argument of 𝑐 5 where 𝑐 = 1 + 𝑖√3
Solution – Here, 𝑐 = 1 + 𝑖 √3
2
Therefore, |𝑐| = √12 + (√3) = 2
Hence, |𝑐 5 | = |𝑐|5 = 25 = 32
Now, let us consider the angle made by the coordinate (1, √3) in clockwise direction with
real axis as θ.
4|DRC
 𝑦 √3 𝜋
Thus, θ = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 √3 =
𝑥 1 3
𝜋
Therefore, arg(𝑐) =
3
𝜋 5𝜋
Now, arg(𝑐 = 5 arg(𝑐) = 5 × = or 3000
5)
3 3
Therefore, principal arg(𝑧) = 3000
Example 2 – Determine the modulus and principal argument of 𝑧 = 3 + √3𝑖. Further find
the principal argument of 𝑧 7 .
Solution – Here, 𝑧 = 3 + √3𝑖
2
Therefore, |𝑧| = √32 + (√3) = 2√3
2
Hence, |𝑧 7 | = |𝑧|7 = (2√3) = 12
Now, let us consider the angle made by the coordinate (3, √3) in clockwise direction with
real axis as θ.
𝑦 √3 1 𝜋
Thus, θ = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 =
𝑥 3 √3 6
𝜋
Therefore, arg(𝑧) =
6
7) 7𝜋
Now, arg(𝑧 = 7 arg(𝑧) = or 2100
6
Example 3 – Determine the modulus and principal argument of 𝑧 = −1 + √3𝑖. Further find
the principal argument of 𝑧 4 .
Solution – Here, 𝑧 = −1 + √3𝑖
2
Therefore, |𝑧| = √(−1)2 + (√3) = 2
Hence, |𝑧 4 | = |𝑧|4 = 24 = 16
Now, let us consider the angle made by the coordinate (−1, √3) in anticlockwise direction
with negative real axis as α and the angle made by the coordinate (−1, √3) in clockwise
direction with positive real axis as θ.
𝑦 √3 𝜋
Thus, α = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 √3 =
𝑥 −1 3
𝜋 2𝜋 0
Now, θ = 𝜋 − 𝛼 = 𝜋 − = or 120
3 3
Therefore, arg(𝑧) = 1200
Now, arg(𝑧 4 ) = 4 arg(𝑧) = 4 × 1200 = 4800
Therefore, principal arg(𝑧 4 ) = 4800 − 2𝜋 = 1200
1 7
Example 4 – Suppose 𝑧 = ( ) . Find the principal argument of 𝑧.
1+𝑖
1 1−𝑖 1 1
Solution – Let 𝑐 = = (1+𝑖)(1−𝑖) = − 𝑖
1+𝑖 2 2
1 2 1 2 1
Therefore, |𝑐| = √( ) + (− ) = √
2 2 2
7
1 1 1
Hence, |𝑧| = |𝑐 7 | = |𝑐|7 =√ = √
2 8 2

5|DRC
 1 1
Now, let us consider the angle made by the coordinate ( , − ) in anticlockwise direction
2 2
1 1
with real axis as α and the angle made by the coordinate ( , − ) in clockwise direction with
2 2
positive real axis as θ.
1
𝑦 − 𝜋
Thus, α = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛 −1
| 1
2
| = 𝑡𝑎𝑛−1 1 = 4
𝑥
2
𝜋 7𝜋
Now, θ = 2𝜋 − 𝛼 = 2𝜋 − = or 3150
4 4
Therefore, arg(𝑐) = 3150
Now, arg(𝑧) = arg(𝑐 7 ) = 7 arg(𝑐) = 7 × 3150 = 22050
Therefore, principal arg(𝑧) = 22050 − 6𝜋 = 450
Example 5 – If 𝑧1 = −1 + 𝑖√3, 𝑧2 = −√3 + 𝑖. Find principal argument of 𝑧1 𝑧2 .
Solution – Here, 𝑧1 = −1 + 𝑖 √3, 𝑧2 = −√3 + 𝑖
We know that 𝑧1 𝑧2 = (𝑎1 𝑎2 − 𝑏1 𝑏2 ) + 𝑖(𝑎2 𝑏1 + 𝑎1 𝑏2 )
Accordingly, 𝑎1 = −1, 𝑎2 = −√3, 𝑏1 = √3, 𝑏2 = 1
Therefore, 𝑧1 𝑧2 = (√3 − √3) + 𝑖(−3 − 1) = −4𝑖 = 0 + (−4)𝑖
Let us consider 𝑐 = 𝑧1 𝑧2 = 0 + (−4)𝑖
Hence, |𝑐| = √02 + (−4)2 = 4
Now, let us consider the angle made by the coordinate (0, −4) in anticlockwise direction
with real axis as α and the angle made by the coordinate (0, −4) in clockwise direction with
positive real axis as θ.
𝑦 −4 𝜋
Thus, α = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 | | = 𝑡𝑎𝑛−1 ∞ =
𝑥 0 2
𝜋 3𝜋 0
Now, θ = 2𝜋 − 𝛼 = 2𝜋 − = or 270
2 2
Therefore, principal arg(𝑐) = 2700
Roots of a Complex Number —
A complex number W is called the 𝑛𝑡ℎ root of a complex number z if 𝑊 𝑛 = z for 𝑛 ∈ 𝑁
𝑛 𝑛
implies W = √𝑧. The 𝑛 values of √𝑧 can easily be obtained as follows:
Let 𝑧 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃)
Also, W = 𝑅(cos 𝜙 + 𝑖 sin 𝜙)
Now, 𝑊 𝑛 = 𝑧
⇒ [𝑅(cos 𝜙 + 𝑖 sin 𝜙)]𝑛 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃)
⇒ 𝑅𝑛 (cos 𝜙 + 𝑖 sin 𝜙)𝑛 = 𝑟(cos 𝜃 + 𝑖 sin 𝜃)
⇒ 𝑅𝑛 (cos 𝑛𝜙 + 𝑖 sin 𝑛𝜙) = 𝑟(cos 𝜃 + 𝑖 sin 𝜃)
𝑛
Therefore, 𝑅𝑛 = 𝑟 ⇒ 𝑅 = √𝑟
𝑛𝜙 = 𝜃 + 2𝑘𝜋, where 𝑘 is an integer
𝜃 2𝑘𝜋
⇒𝜙= +
𝑛 𝑛
For 𝑘 = 0, 1, 2, …., 𝑛 − 1 we get 𝑛 distinct values of W. Further integers of 𝑘 would give
2𝑘𝜋
values already obtained. For instance, 𝑘 = 𝑛 gives | = 2𝜋. Hence, W corresponds to
𝑛 𝑘=𝑛
𝑛
𝑘 = 0. Consequently, √𝑧 for 𝑧 ≠ 0 has 𝑛 distinct roots.
𝑛 𝑛 𝜃+2𝑘𝜋 𝜃+2𝑘𝜋
Thus, W = √𝑧 = √𝑟 (cos + 𝑖 sin
𝑛
) where 𝑘 = 0, 1, 2, …., 𝑛 − 1
𝑛
6|DRC
 𝑛
These 𝑛 values lie on the circle of radius √𝑟 with centre at the origin and constitute the
𝑛
vertices of a regular polygon of size 𝑛. The value √𝑧 obtained by taking the principal
𝑛
argument of 𝑧 and 𝑘 = 0 is called the principal value of W = √𝑧
Example 1 – Find the 𝑛𝑡ℎ root of unity.
Solution – Let 𝑊 𝑛 = 1
⇒ 𝑊 𝑛 = cos 2𝑘𝜋 + 𝑖 sin 2𝑘𝜋
1
⇒ W = (cos 2𝑘𝜋 + 𝑖 sin 2𝑘𝜋)𝑛
2𝑘𝜋 2𝑘𝜋
⇒ W = cos + 𝑖 sin ; 𝑘 = 0, 1, 2, …., 𝑛 − 1
𝑛 𝑛
2𝑘𝜋
𝑖
⇒W=𝑒 𝑛 ; 𝑘 = 0, 1, 2, …., 𝑛 − 1
2𝜋
2𝜋 2𝜋
Putting ω = cos + 𝑖 sin = 𝑒 𝑖 𝑛 the 𝑛𝑡ℎ roots of unity are 1, ω, 𝜔2 , …., 𝜔𝑛−1
𝑛 𝑛
Example 2 – Find the roots of 𝑊 𝑛 = −1
Solution – Let 𝑊 𝑛 = −1
⇒ 𝑊 𝑛 = cos(𝜋 + 2𝑘𝜋) + 𝑖 sin(𝜋 + 2𝑘𝜋)
1
⇒ W = [cos(𝜋 + 2𝑘𝜋) + 𝑖 sin(𝜋 + 2𝑘𝜋)]𝑛
𝜋+2𝑘𝜋 𝜋+2𝑘𝜋
⇒ W = cos ( ) + 𝑖 sin (
𝑛
); 𝑘 = 0, 1, 2, …., 𝑛 − 1
𝑛
𝜋+2𝑘𝜋
𝑖( )
⇒W=𝑒 ; 𝑘 = 0, 1, 2, …., 𝑛 − 1
𝑛

Example 3 – Find the roots of 𝑧 3 = 8𝑖

Solution – Let 𝑧 3 = 8𝑖
𝜋 𝜋
⇒ 𝑧 3 = 8 [cos ( + 2𝑘𝜋) + 𝑖 sin ( + 2𝑘𝜋)]; 𝑘 = 0, 1, 2
2 2
1
𝜋 𝜋 3
⇒ 𝑧 = {8 [cos ( + 2𝑘𝜋) + 𝑖 sin ( + 2𝑘𝜋)]}
2 2
𝜋 𝜋
+2𝑘𝜋 +2𝑘𝜋
2 2
⇒ 𝑧 = 2 [cos ( ) + 𝑖 sin ( )]; 𝑘 = 0, 1, 2
3 3
𝜋 𝜋 √3 1
For 𝑘 = 0, 1st root of 𝑧 = 2 (cos + 𝑖 sin ) = 2 ( + 𝑖 ) = √3 + 𝑖
6 6 2 2
5𝜋 5𝜋 √3 1
For 𝑘 = 1, 2nd root of 𝑧 = 2 (cos + 𝑖 sin ) = 2 (− + 𝑖 ) = −√3 + 𝑖
6 6 2 2
9𝜋 9𝜋
For 𝑘 = 2, 3 root of 𝑧 = 2 (cos + 𝑖 sin ) = 2(0 + (−1)𝑖) = −2𝑖
rd
6 6
Functions of Complex Variable —
Assume 𝑧 = 𝑥 + 𝑖𝑦 ∈ ℂ ∀ 𝑓(𝑧) = 𝑧 2
So that 𝑓(𝑧) can be written as 𝑓(𝑧) = 𝑧 2
= (𝑥 + 𝑖𝑦)2
= 𝑥 2 + 2𝑖𝑥𝑦 + (𝑖𝑦)2
= 𝑥 2 − 𝑦 2 + 𝑖2𝑥𝑦
= 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦)
2 2
with 𝑢(𝑥, 𝑦) = 𝑥 − 𝑦 , 𝑣(𝑥, 𝑦) = 2𝑥𝑦
In general, a complex valued functions are given by 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦), 𝑧 = 𝑥 + 𝑖𝑦
where 𝑢(𝑥, 𝑦) and 𝑣(𝑥, 𝑦) are real valued function.

7|DRC
Using 𝑧 = 𝑟𝑒 𝑖𝜃 in the expression of a complex function 𝑓 we get 𝑓(𝑧) = 𝑓(𝑟𝑒 𝑖𝜃 ) =
𝑢(𝑟, 𝜃) + 𝑖𝑣(𝑟, 𝜃) where 𝑢 and 𝑣 are real functions of the real variable 𝑟 and θ.
For example,
1. 𝑓(𝑧) = 𝑧 2
2
= (𝑟𝑒 𝑖𝜃 )
= 𝑟 2 𝑒 𝑖2𝜃
= 𝑟 2 (cos 2𝜃 + 𝑖 sin 2𝜃)
= 𝑟 2 cos 2𝜃 + 𝑖𝑟 2 sin 2𝜃
= 𝑢(𝑟, 𝜃) + 𝑖𝑣(𝑟, 𝜃)
2. 𝑓(𝑧) = 𝑧 5 + 4𝑧 2 − 6
5 2
= (𝑟𝑒 𝑖𝜃 ) + 4(𝑟𝑒 𝑖𝜃 ) − 6
= 𝑟 5 𝑒 𝑖5𝜃 + 4𝑟 2 𝑒 𝑖2𝜃 − 6
= 𝑟 5 (cos 5𝜃 + 𝑖 sin 5𝜃) + 4𝑟 2 (cos 2𝜃 + 𝑖 sin 2𝜃) − 6
= (𝑟 5 cos 5𝜃 + 4𝑟 2 cos 2𝜃 − 6) + 𝑖(𝑟 5 sin 5𝜃 + 4𝑟 2 sin 2𝜃)
= 𝑢(𝑟, 𝜃) + 𝑖𝑣(𝑟, 𝜃)
Derivation of Complex Valued Function —
All the usual result for real valued function also holds for complex valued function.
For example,
i) If 𝑓(𝑧) = 𝑐, 𝑓 ′ (𝑧) = 0
ii) If 𝑓(𝑧) = 𝑔(𝑧) ± ℎ(𝑧), 𝑓 ′ (𝑧) = 𝑔′ (𝑧) ± ℎ′ (𝑧)
𝑔(𝑧) ℎ(𝑧)𝑔′ (𝑧)−𝑔(𝑧)ℎ′ (𝑧)
iii) If 𝑓(𝑧) = , 𝑓 ′ (𝑧) = {ℎ(𝑧)}2
ℎ(𝑧)
′ (𝑧)
iv) If 𝑓(𝑧) = 𝑔(𝑧). ℎ(𝑧), 𝑓 = 𝑔(𝑧)ℎ′ (𝑧) + ℎ(𝑧)𝑔′ (𝑧)
𝑑 𝑧
v) 𝑒 = 𝑒𝑧
𝑑𝑧
𝑑
vi) cos 𝑧 = − sin 𝑧
𝑑𝑧
If a function 𝑓 has a derivative at a point 𝑧0 we say that 𝑓 is differentiable at 𝑧0 .
If 𝑓 is differentiable at every point in a ball 𝐵(𝑧0 , 𝛿 ) centred at 𝑧0 with radius δ (𝛿 > 0)
may be too small then the function 𝑓 is called analytic at 𝑧 = 𝑧0 . Here, 𝐵(𝑧0 , 𝛿) =
{𝑧 ∈ ℂ; |𝑧 − 𝑧0 | < 𝛿}
NOTE : The ball 𝐵(𝑧0 , 𝛿) is also called neighbourhood of 𝑧0 .
If a function 𝑓(𝑧) is analytic at every point on a set A, we say 𝑓 is analytic on the set A or
holomorphic in A.
If a function 𝑓(𝑧) is analytic at every point in the complex plane i.e., analytic on the set of
the complex number then 𝑓 is called entire function. sin 𝑧, cos 𝑧, 𝑒 𝑧 , etc. are all expected to
be entire functions. But how to determine whether a function is analytic at a point or not,
there is set of conditions with the help of which we can see the analyticity of a function.
Cauchy-Reiman Conditions —
Let 𝑤 = 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) = 𝑢 + 𝑖𝑣 be a complex valued function of 𝑧 = 𝑥 + 𝑖𝑦.
𝛿𝑢 𝛿𝑣
The set of equations 𝑢𝑥 = 𝑣𝑦 i.e., ( = )
𝛿𝑥 𝛿𝑦
𝛿𝑢 𝛿𝑣
𝑢𝑦 = −𝑣𝑥 i.e., ( = )
𝛿𝑦 𝛿𝑥

8|DRC
are called Cauchy-Reiman partial differential equations or C-R conditions or C-R equations
which were discovered by Cauchy and used extensively by Cauchy and Reiman.
Necessary and Sufficient Condition for 𝒇(𝒛) to be Analytic —
Necessary Condition :—
The necessary condition for 𝑤 = 𝑓(𝑧) = 𝑢 + 𝑖𝑣 to be analytic in a region R is that
i) Partially derivation of 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 exist.
ii) They satisfy C-R equation.
Proof : Given that the function 𝑓(𝑧) is analytic in R
To show that C-R equation are satisfied.
𝑑𝑤 ∆𝑤
Since 𝑤 = 𝑓(𝑧) = 𝑢 + 𝑖𝑣 is analytic in R i.e., = lim = 𝑓 ′ (𝑧) must exist and is unique
𝑑𝑧 ∆𝑧→0 ∆𝑧
also.
Now, ∆𝑧 may approach to zero along the real axis (𝑥 axis) or along the imaginary axis (𝑦
axis).
𝑑𝑤 ∆𝑤 ∆𝑤
If along 𝑥 axis then ∆𝑧 = ∆(𝑥 + 𝑖𝑦) = ∆𝑥 only and = lim = lim ……(1)
𝑑𝑧 ∆𝑧→0 ∆𝑧 ∆𝑥→0 ∆𝑥
𝑑𝑤 ∆𝑤 ∆𝑤
If its along 𝑦 axis then ∆𝑧 = ∆(𝑥 + 𝑖𝑦) = 𝑖∆𝑦 only and = lim = lim ……(2)
𝑑𝑧 ∆𝑧→0 ∆𝑧 ∆𝑦→0 𝑖∆𝑦
𝑑𝑤 ∆𝑤 ∆𝑤
Since exist so lim and lim exists.
𝑑𝑧 ∆𝑥→0 ∆𝑥 ∆𝑦→0 𝑖∆𝑦
Now, 𝑓(𝑧) cannot be analytic if the two units viz., (1) and (2) are not identical. Thus, the
𝑑𝑤 ∆𝑤 ∆𝑤
necessary condition for 𝑓(𝑧) to be analytic is that = lim = lim
𝑑𝑧 ∆𝑥→0 ∆𝑥 ∆𝑦→0 𝑖∆𝑦
𝑑𝑤 𝛿𝑤 1 𝛿𝑤
Now, = = ……(3)
𝑑𝑧 𝛿𝑥 𝑖 𝛿𝑦
𝛿𝑤 𝛿𝑢 𝛿𝑣 𝛿𝑤 𝛿𝑢 𝛿𝑣
Also = +𝑖 and = +𝑖 ……(4)
𝛿𝑥 𝛿𝑥 𝛿𝑥 𝛿𝑦 𝛿𝑦 𝛿𝑦
Using (4) and (3) we get
𝛿𝑢 𝛿𝑣 1 𝛿𝑢 𝛿𝑣
+𝑖 = ( +𝑖 )
𝛿𝑥 𝛿𝑥 𝑖 𝛿𝑦 𝛿𝑦
𝛿𝑢 𝛿𝑣 𝛿𝑢 𝛿𝑣
⇒𝑖 + 𝑖2 = +𝑖
𝛿𝑥 𝛿𝑥 𝛿𝑦 𝛿𝑦
⇒ 𝑖𝑢𝑥 − 𝑣𝑥 = 𝑢𝑦 + 𝑖𝑣𝑦
⇒ 𝑖𝑢𝑥 − 𝑢𝑦 = 𝑖𝑣𝑦 + 𝑣𝑥
⇒ −𝑢𝑦 + 𝑖𝑢𝑥 = 𝑣𝑥 + 𝑖𝑣𝑦
Equating real and imaginary part we get −𝑢𝑦 = 𝑣𝑥 and 𝑢𝑥 = 𝑣𝑦 i.e., 𝑢𝑦 = −𝑣𝑥 and 𝑢𝑥 = 𝑣𝑦 .
Sufficient Condition :—
The sufficient condition for 𝑤 = 𝑓(𝑧) = 𝑢 + 𝑖𝑣 to be analytic in a region R are that
i) 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 exist and are continuous in R
ii) They satisfy C-R equation 𝑢𝑥 = 𝑣𝑦 , 𝑢𝑦 = 𝑣𝑥
Proof : Given that 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 exist and are continuous in R and these satisfy C-R
equations
To prove that 𝑓(𝑧) is analytic.
We have 𝑤 = 𝑓(𝑧) = 𝑢 + 𝑖𝑣 such that ∆𝑤 = ∆𝑢 + 𝑖∆𝑣
and 𝑧 = 𝑥 + 𝑖𝑦 such that ∆𝑧 = ∆𝑥 + 𝑖∆𝑦
Now, ∆𝑢 = 𝑢(𝑥 + ∆𝑥, 𝑦 + ∆𝑦) − 𝑢(𝑥, 𝑦)

9|DRC
 = {𝑢(𝑥 + ∆𝑥, 𝑦 + ∆𝑦) − 𝑢(𝑥, 𝑦 + ∆𝑦)} + {𝑢(𝑥, 𝑦 + ∆𝑦) − 𝑢(𝑥, 𝑦)}
= 𝑇1 + 𝑇2 ……(1)
Now, 𝑇1 = 𝑢(𝑥 + ∆𝑥, 𝑦 + ∆𝑦) − 𝑢(𝑥, 𝑦 + ∆𝑦)
𝛿
= ∆𝑥 𝑢(𝑥 ∗ , 𝑦 + ∆𝑦) ……(2)
𝛿𝑥
{Using MVT for 𝑥 argument where 𝑥 ∗ = 𝑥 + 𝜃1 ∆𝑥; 0 < 𝜃1 < 1}
Again, due to continuity of 𝑢𝑥 we have lim 𝑢𝑥 (𝑥 ∗ , 𝑦 + ∆𝑦)
∆𝑥→0
∆𝑦→0
= lim 𝑢𝑥 (𝑥 + 𝜃1 ∆𝑥, 𝑦 + ∆𝑦)
∆𝑥→0
∆𝑦→0
= 𝑢𝑥 (𝑥, 𝑦)
(𝑥 ∗
which means that 𝑢𝑥 , 𝑦 + ∆𝑦) = 𝑢𝑥 (𝑥, 𝑦) +∈1 where ∈1 → 0 as ∆𝑥, ∆𝑦 → 0 ……(3)
Using (2) and (3) we get
𝑇1 = ∆𝑥(𝑢𝑥 (𝑥, 𝑦) +∈1 ) where ∈1 → 0 as ∆𝑥, ∆𝑦 → 0 ……(4)
Similarly, 𝑇2 = 𝑢(𝑥, 𝑦 + ∆𝑦) − 𝑢(𝑥, 𝑦)
𝛿
= ∆𝑦 𝑢(𝑥, 𝑦 ∗ ) where 𝑦 ∗ = 𝑦 + 𝜃2 ∆𝑦; 0 < 𝜃2 < 1
𝛿𝑦
This implies 𝑇2 = ∆𝑦(𝑢𝑦 (𝑥, 𝑦) +∈2 ) where ∈2 → 0 as ∆𝑥, ∆𝑦 → 0 ……(5)
Now, using (4), (5) and (1) we get
∆𝑢 = ∆𝑥(𝑢𝑥 (𝑥, 𝑦) +∈1 ) + ∆𝑦(𝑢𝑦 (𝑥, 𝑦) +∈2 ) ……(6)
Similarly, ∆𝑣 = ∆𝑥(𝑣𝑥 (𝑥, 𝑦) + 𝜂1 ) + ∆𝑦(𝑣𝑦 (𝑥, 𝑦) + 𝜂2 ) ……(7)
where 𝜂1 , 𝜂2 → 0 as ∆𝑥, ∆𝑦 → 0
Again, ∆𝑤 = ∆𝑢 + 𝑖∆𝑣
= ∆𝑥(𝑢𝑥 (𝑥, 𝑦) +∈1 ) + ∆𝑦(𝑢𝑦 (𝑥, 𝑦) +∈2 ) + 𝑖[∆𝑥(𝑣𝑥 (𝑥, 𝑦) + 𝜂1 ) +
∆𝑦(𝑣𝑦 (𝑥, 𝑦) + 𝜂2 )]
= ∆𝑥[𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦)] + ∆𝑦[𝑢𝑦 (𝑥, 𝑦) + 𝑖𝑣𝑦 (𝑥, 𝑦)] + ∆𝑥(∈1 + 𝑖𝜂1 ) +
∆𝑦(∈2 + 𝑖𝜂2 )
Using C-R equations
∆𝑤 = ∆𝑥[𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦)] + ∆𝑦[−𝑣𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑦 (𝑥, 𝑦)] + ∆𝑥(∈1 + 𝑖𝜂1 ) +
∆𝑦(∈2 + 𝑖𝜂2 )
= ∆𝑥[𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦)] + ∆𝑦[−𝑣𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑦 (𝑥, 𝑦)] + 𝑇 ……(8)
where 𝑇 = ∆𝑥(∈1 + 𝑖𝜂1 ) + ∆𝑦(∈2 + 𝑖𝜂2 )
⇒ ∆𝑤 = 𝑢𝑥 (𝑥, 𝑦){∆𝑥 + 𝑖∆𝑦} + 𝑣𝑥 (𝑥, 𝑦){𝑖∆𝑥 − ∆𝑦} + 𝑇
⇒ ∆𝑤 = 𝑢𝑥 (𝑥, 𝑦)∆𝑧 + 𝑖𝑣𝑥 (𝑥, 𝑦)∆𝑧 + 𝑇
⇒ ∆𝑤 = {𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦)}∆𝑧 + 𝑇
∆𝑤 𝑇
⇒ = 𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦) + ……(9)
∆𝑧 ∆𝑧
𝑑𝑤 ∆𝑤 𝑇
Therefore, = lim = 𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦) + lim ……(10)
𝑑𝑧 ∆𝑧→0 ∆𝑧 ∆𝑧→0 ∆𝑧
∆𝑥 |∆𝑥| |∆𝑥| |∆𝑥|
Now, | | = |∆𝑧| = ≤ |∆𝑥| = 1
∆𝑧 √(∆𝑥)2 +(∆𝑦)2
∆𝑥
Therefore, 0 < |(∈1 + 𝑖𝜂1 ) | ≤ |∈1 + 𝑖𝜂1 | → 0 as ∆𝑧 → 0
∆𝑧
∆𝑥
Therefore, |(∈1 + 𝑖𝜂1 ) | → 0 as ∆𝑧 → 0
∆𝑧

10 | D R C
 ∆𝑥
Therefore, (∈1 + 𝑖𝜂1 ) → 0 as ∆𝑧 → 0
∆𝑧
∆𝑥
Similarly, (∈2 + 𝑖𝜂2 ) → 0 as ∆𝑧 → 0
∆𝑧
𝑇 ∆𝑥 ∆𝑥
Therefore, lim = lim (∈1 + 𝑖𝜂1 ) + (∈2 + 𝑖𝜂2 ) =0
∆𝑧→0 ∆𝑧 ∆𝑧→0 ∆𝑧 ∆𝑧
𝑑𝑤
Thus, from (10) we get = 𝑢𝑥 (𝑥, 𝑦) + 𝑖𝑣𝑥 (𝑥, 𝑦)
𝑑𝑧
Therefore, 𝑤 is differentiable at any point (𝑥, 𝑦) ∈ ℝ
Remark : In the previous theorem we have discussed analyticity of a function in a region ℝ
through C-R equations. Now, we will discuss differentiability at a single point through C-R
equations.
Theorem 1 : Let 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) be a continuous function that is defined in some
neighbourhood of the point 𝑧0 = 𝑥0 + 𝑖𝑦0 . If all the partial derivatives 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 are
continuous at the point 𝑧0 = (𝑥0 , 𝑦0 ) and if the C-R equations 𝑢𝑥 = 𝑣𝑦 , 𝑢𝑦 = −𝑣𝑥 hold at
𝑧0 , the function 𝑓(𝑧) is differentiable at 𝑧0 and the derivative 𝑓 ′ (𝑧) can be computed using
any one of the following – 𝑓 ′ (𝑧0 ) = 𝑢𝑥 (𝑥0 , 𝑦0 ) + 𝑖𝑣𝑥 (𝑥0 , 𝑦0 ) or 𝑓 ′ (𝑧) = 𝑣𝑦 (𝑥0 , 𝑦0 ) −
𝑖𝑢𝑦 (𝑥0 , 𝑦0 )
Example 1 – Show that the function 𝑓(𝑧) = 𝑒 −𝑦 cos 𝑥 + 𝑖𝑒 −𝑦 sin 𝑥 is differentiable for all
𝑧. Find its derivative.
Solution – Here, 𝑢(𝑥, 𝑦) = 𝑒 −𝑦 cos 𝑥
𝑣(𝑥, 𝑦) = 𝑒 −𝑦 sin 𝑥
Now, 𝑢𝑥 = −𝑒 −𝑦 sin 𝑥
𝑢𝑦 = −𝑒 −𝑦 cos 𝑥
𝑣𝑥 = 𝑒 −𝑦 cos 𝑥
𝑣𝑦 = −𝑒 −𝑦 sin 𝑥
Therefore, 𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥
We know that 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 are continuous function and the C-R equations hold for all
values of 𝑥, 𝑦
Therefore, the function is differentiable for all 𝑧 and 𝑓 ′ (𝑧) = −𝑒 −𝑦 sin 𝑥 + 𝑖𝑒 −𝑦 cos 𝑥
Example 2 – Let 𝑓(𝑧) = 𝑥 3 − 3𝑥𝑦 2 + 𝑖(3𝑥 2 𝑦 − 𝑦 3 ) . Show that this function is
differentiable for all 𝑧 and find its derivative.
Solution – Here, 𝑢(𝑥, 𝑦) = 𝑥 3 − 3𝑥𝑦 2
𝑣(𝑥, 𝑦) = 3𝑥 2 𝑦 − 𝑦 3
Now, 𝑢𝑥 = 3𝑥 2 − 3𝑦 2
𝑢𝑦 = −6𝑥𝑦
𝑣𝑥 = 6𝑥𝑦
𝑣𝑦 = 3𝑥 2 − 3𝑦 2
Therefore, 𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥
We know that 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 are continuous function and the C-R equations hold for all
values of 𝑥, 𝑦
Therefore, the function is differentiable for all 𝑧 and 𝑓 ′ (𝑧) = 3𝑥 2 − 3𝑦 2 + 𝑖6𝑥𝑦 i.e., 𝑓 ′ (𝑧)
= 3(𝑥 2 + (𝑖𝑦)2 + 2𝑥. 𝑖𝑦) = 3𝑧 2 which is also the derivative of an entire function 𝑧 3 .
Example 3 – Show that 𝑓(𝑧) = 𝑧̅ is nowhere differentiable.
11 | D R C
Solution – Here, 𝑓(𝑧) = 𝑧̅ = ̅̅̅̅̅̅̅̅
𝑥 + 𝑖𝑦 = 𝑥 − 𝑖𝑦
𝑢(𝑥, 𝑦) = 𝑥
𝑣(𝑥, 𝑦) = 𝑦
Now, 𝑢𝑥 = 1
𝑢𝑦 = 0
𝑣𝑥 = 0
𝑣𝑦 = −1
Hence, C-R equations are not satisfied for any point 𝑧 = (𝑥, 𝑦)
Therefore, 𝑓(𝑧) is nowhere differentiable.
Example 4 – Find all points at which the function 𝑓 given by 𝑓(𝑧) = 𝑥 3 + 𝑖(1 − 𝑦)3
Solution – Here, 𝑢(𝑥, 𝑦) = 𝑥 3
𝑣(𝑥, 𝑦) = (1 − 𝑦)3
Now, 𝑢𝑥 = 3𝑥 2
𝑢𝑦 = 0
𝑣𝑥 = 0
𝑣𝑦 = −3(1 − 𝑦)2
All 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 are continuous. Thus, 𝑓(𝑧) = 𝑢 + 𝑖𝑣 is differentiable at 𝑧0 = 𝑥0 + 𝑖𝑦0 if
and only if C-R equations are satisfied. Hence,
𝑢𝑥 (𝑥0 , 𝑦0 ) = 𝑣𝑦 (𝑥0 , 𝑦0 )
⇒ 3𝑥0 2 = −3(1 − 𝑦0 )2
⇒ 𝑥0 2 + (1 − 𝑦0 )2 = 0
So, 𝑥0 = 0 and 1 − 𝑦0 = 0 ⇒ 𝑦0 = 1
Therefore, the function is differentiable only at (0, 1)
Example 5 – Show that 𝑓(𝑧) = 𝑧 2 is an entire function.
Solution – Let 𝑓(𝑤) = 𝑧 2 = (𝑥 + 𝑖𝑦)2 = 𝑥 2 − 𝑦 2 + 𝑖2𝑥𝑦
Here, 𝑢(𝑥, 𝑦) = 𝑥 2 − 𝑦 2
𝑣(𝑥, 𝑦) = 2𝑥𝑦
Now, 𝑢𝑥 = 2𝑥
𝑢𝑦 = −2𝑦
𝑣𝑥 = 2𝑦
𝑣𝑦 = 2𝑥
Therefore, 𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥
We know that 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 are continuous function and the C-R equations hold for all
values of 𝑥, 𝑦
Therefore, the function is differentiable for all 𝑧 and 𝑓 ′ (𝑤) = 2𝑥 + 𝑖2𝑦 i.e., 𝑓 ′ (𝑤) =
2(𝑥 + 𝑖𝑦) = 2𝑧
Again, 𝑓 ′ (𝑧) = 2𝑧
Thus, 𝑧 2 is an entire function.
Example 6 – Find a point 𝑧 where 𝑓(𝑧) = |𝑧|2 is differentiable.
2
Solution – We have 𝑓(𝑧) = |𝑧|2 = (√x 2 + y 2 ) = 𝑥 2 + 𝑦 2
Here, 𝑢(𝑥, 𝑦) = 𝑥 2 + 𝑦 2
𝑣(𝑥, 𝑦) = 0
12 | D R C
Now, 𝑢𝑥 = 2𝑥
𝑢𝑦 = 2𝑦
𝑣𝑥 = 0
𝑣𝑦 = 0
All 𝑢𝑥 , 𝑢𝑦 , 𝑣𝑥 , 𝑣𝑦 are continuous. Thus, 𝑓(𝑧) = 𝑢 + 𝑖𝑣 is differentiable at 𝑧0 = 𝑥0 + 𝑖𝑦0 if
and only if C-R equations are satisfied. Hence,
𝑢𝑥 (𝑥0 , 𝑦0 ) = 𝑣𝑦 (𝑥0 , 𝑦0 )
⇒ 2𝑥0 = 0
⇒ 𝑥0 = 0
Again, 𝑢𝑦 (𝑥0 , 𝑦0 ) = −𝑣𝑥 (𝑥0 , 𝑦0 )
⇒ 2𝑦0 = 0
⇒ 𝑦0 = 0
Therefore, the function is differentiable only at (0, 0)
Polar Form of Cauchy Reiman Equations —
𝑦
We have 𝑥 = 𝑟 cos 𝜃 and 𝑦 = 𝑟 sin 𝜃 then 𝑟 2 = 𝑥 2 + 𝑦 2 and 𝜃 = 𝑡𝑎𝑛−1
𝑥
𝛿𝑟 2𝑥 𝑥
Now, = = = cos 𝜃
𝛿𝑥 2𝑟 𝑟
𝛿𝑟 2𝑦 𝑦
= = = sin 𝜃
𝛿𝑦 2𝑟 𝑟
𝛿𝜃 1 −𝑦 −𝑦 sin 𝜃
Again, = 𝑦 2
( 𝑥2 ) = =−
𝛿𝑥 1+( ) 𝑟2 𝑟
𝑥
𝛿𝜃 1 1 𝑥 cos 𝜃
𝛿𝑦
= 𝑦 2
(𝑥) = 𝑟 2 = 𝑟
1+( )
𝑥
𝛿𝜃 sin 𝜃 𝛿𝜃 cos 𝜃
Therefore, =− and =
𝛿𝑥 𝑟 𝛿𝑦 𝑟
𝛿𝑢 𝛿𝑢 𝛿𝑟 𝛿𝑢 𝛿𝜃 𝛿𝑢 𝛿𝑢 sin 𝜃
Now, we have = . + . = cos 𝜃 + (− )
𝛿𝑥 𝛿𝑟 𝛿𝑥 𝛿𝜃 𝛿𝑥 𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑢 𝛿𝑢 𝛿𝑟 𝛿𝑢 𝛿𝜃 𝛿𝑢 𝛿𝑢 cos 𝜃
= . + . = sin 𝜃 +
𝛿𝑦 𝛿𝑟 𝛿𝑦 𝛿𝜃 𝛿𝑦 𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑣 𝛿𝑣 𝛿𝑟 𝛿𝑣 𝛿𝜃 𝛿𝑣 𝛿𝑣 sin 𝜃
= . + . = cos 𝜃 + (− )
𝛿𝑥 𝛿𝑟 𝛿𝑥 𝛿𝜃 𝛿𝑥 𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑣 𝛿𝑣 𝛿𝑟 𝛿𝑣 𝛿𝜃 𝛿𝑣 𝛿𝑣 cos 𝜃
= . + . = sin 𝜃 +
𝛿𝑦 𝛿𝑟 𝛿𝑦 𝛿𝜃 𝛿𝑦 𝛿𝑟 𝛿𝜃 𝑟
Cauchy Reiman equations in cartesian plane are
𝛿𝑢 𝛿𝑣 𝛿𝑢 𝛿𝑢 sin 𝜃 𝛿𝑣 𝛿𝑣 cos 𝜃
= ⇒ cos 𝜃 + (− ) = sin 𝜃 + ……(1)
𝛿𝑥 𝛿𝑦 𝛿𝑟 𝛿𝜃 𝑟 𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑢 𝛿𝑣 𝛿𝑢 𝛿𝑢 cos 𝜃 𝛿𝑣 𝛿𝑣 sin 𝜃
=− ⇒ sin 𝜃 + = −[ cos 𝜃 + (− )] ……(2)
𝛿𝑦 𝛿𝑥 𝛿𝑟 𝛿𝜃 𝑟 𝛿𝑟 𝛿𝜃 𝑟
Multiplying (1) by cos 𝜃 and (2) by sin 𝜃 and then adding we get
𝛿𝑢 𝛿𝑢 sin 𝜃 cos 𝜃 𝛿𝑢 𝛿𝑢 sin 𝜃 cos 𝜃 𝛿𝑣 𝛿𝑣 𝑐𝑜𝑠 2 𝜃
𝑐𝑜𝑠 2 𝜃 + (− ) + 𝛿𝑟 𝑠𝑖𝑛2 𝜃 + 𝛿𝜃 = cos 𝜃 sin 𝜃 + −
𝛿𝑟 𝛿𝜃 𝑟 𝑟 𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑣 𝛿𝑣 𝑠𝑖𝑛2 𝜃
sin 𝜃 cos 𝜃 − (− )
𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑢 𝛿𝑣 𝑠𝑖𝑛2 𝜃+𝑐𝑜𝑠 2 𝜃
⇒ [𝑠𝑖𝑛2 𝜃 + 𝑐𝑜𝑠 2 𝜃] = [ ]
𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑢 1 𝛿𝑣
⇒ =
𝛿𝑟 𝑟 𝛿𝜃
Again, multiplying (1) by sin 𝜃 and (2) by cos 𝜃 and subtracting we get
13 | D R C
 𝛿𝑢 𝛿𝑢 𝑠𝑖𝑛2 𝜃 𝛿𝑢 𝛿𝑢 𝑐𝑜𝑠 2 𝜃 𝛿𝑣 𝛿𝑣 sin 𝜃 cos 𝜃
sin 𝜃 cos 𝜃 + (− ) − 𝛿𝑟 sin 𝜃 cos 𝜃 − 𝛿𝜃 = 𝑠𝑖𝑛2 𝜃 + +
𝛿𝑟 𝛿𝜃 𝑟 𝑟 𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑣 2 𝛿𝑣 sin 𝜃 cos 𝜃
𝑐𝑜𝑠 𝜃 + (− )
𝛿𝑟 𝛿𝜃 𝑟
𝛿𝑢 𝑠𝑖𝑛2 𝜃+𝑐𝑜𝑠 2 𝜃 𝛿𝑣
⇒− [ ]= [𝑠𝑖𝑛2 𝜃 + 𝑐𝑜𝑠 2 𝜃]
𝛿𝜃 𝑟 𝛿𝑟
𝛿𝑣 1 𝛿𝑢
⇒ =−
𝛿𝑟 𝑟 𝛿𝜃
𝛿𝑢 1 𝛿𝑣 1 𝛿𝑣 1 𝛿𝑢
Thus, in polar form the C-R equations are given by = ⇒ 𝑢𝑟 = 𝑣𝜃 and =− ⇒
𝛿𝑟 𝑟 𝛿𝜃 𝑟 𝛿𝑟 𝑟 𝛿𝜃
1
𝑣𝑟 = − 𝑢𝜃
𝑟
Theorem 2 : Let 𝑓(𝑧) = 𝑓(𝑟𝑒 𝑖𝜃 ) = 𝑢(𝑟, 𝜃) + 𝑖𝑣(𝑟, 𝜃) be a continuous function that is
defined in some neighbourhood of the point 𝑧0 = 𝑟0 𝑒 𝑖𝜃0 . If all partial derivatives 𝑢𝑟 , 𝑢𝜃 , 𝑣𝑟 ,
𝑣𝜃 are continuous at the point 𝑧0 = (𝑟0 , 𝜃0 ) and if the polar form of the C-R equations
1 1
𝑢𝑟 (𝑟0 , 𝜃0 ) = 𝑣𝜃 (𝑟0 , 𝜃0 ); 𝑣𝑟 (𝑟0 , 𝜃0 ) = − 𝑢𝜃 (𝑟0 , 𝜃0 ) holds then 𝑓 is differentiable at 𝑧0
𝑟0 𝑟0
and we can compute by 𝑓 ′ (𝑧0 ) 𝑓 ′ (𝑧0 ) = 𝑒 −𝑖𝜃0 [𝑢𝑟 (𝑟0 , 𝜃0 ) + 𝑖𝑣𝑟 (𝑟0 , 𝜃0 )] or 𝑓 ′ (𝑧0 ) =
1 −𝑖𝜃
𝑒 0 [𝑣𝜃 (𝑟0 , 𝜃0 ) − 𝑖𝑢𝜃 (𝑟0 , 𝜃0 )]
𝑟0
1 1 1 1
𝑖𝜃 𝜃 𝜃
Example 1 – If 𝑓(𝑧) = 𝑓(𝑟𝑒 ) = 𝑧 = 2 (𝑟𝑒 𝑖𝜃 )2 = 𝑟 2 cos + 𝑖𝑟 2 sin where the domain is
2 2
1
1 1 𝜃
𝑖𝜃
restricted to be {𝑟𝑒 ; 𝑟 > 0, −𝜋 < 𝜃 < 𝜋}. Then show that 𝑓 ′ (𝑧)
= 1 = 𝑟 −2 cos −
𝑟 2
2𝑧 2
1
1 − 𝜃
𝑖 𝑟 2 sin for every point in the domain.
𝑟 2
1
𝜃
Solution – Here, 𝑢(𝑟, 𝜃) = 𝑟 2 cos
2
1
𝜃
𝑣(𝑟, 𝜃) = 𝑟 sin 2
2
𝜃 1 1 1 𝜃
Now, 𝑢𝑟 = cos . = cos
2 2 12 1
2
𝑟 2𝑟 2
1
𝜃 1 1 1 𝜃
𝑢𝜃 = 𝑟 (− sin ) = − 𝑟 2 sin
2
2 2 2 2
𝜃 1 1 1 𝜃
𝑣𝑟 = sin . = sin
2 2 12 1
2
𝑟 2𝑟 2
1
1 𝜃
𝑣𝜃 = 𝑟 cos 2
2 2
1 1 𝜃 1
Now, 𝑣𝜃 = 1 cos . Therefore, 𝑢𝑟 = 𝑣𝜃
𝑟 2 𝑟
2𝑟 2
1 1 𝜃 1
Again, − 𝑢𝜃 = 1 sin . Therefore, 𝑣𝑟 = − 𝑢𝜃
𝑟 2 𝑟
2𝑟 2
Since 𝑢𝑟 , 𝑢𝜃 , 𝑣𝑟 , 𝑣𝜃 are continuous at every point in the domain and C-R equations holds.
Therefore, the function is differentiable in the given domain. Then
1 𝜃 1 𝜃
𝑓 ′ (𝑧) = 𝑒 −𝑖𝜃 [ 1 cos + 𝑖 1 sin ]
2 2
2𝑟 2 2𝑟 2
1 𝜃 𝜃
= 1 𝑒 −𝑖𝜃 [cos + 𝑖 sin ]
2 2
2𝑟 2
𝜃
1 𝑖
= 1 𝑒 −𝑖𝜃 𝑒 2
2𝑟 2

14 | D R C
 1
1 −𝑖𝜃( −1)
= 1 𝑒 2
2𝑟 2
𝜃
1
= 1 𝑒 −𝑖 2
2𝑟 2
1
1 −
= (𝑟𝑒 𝑖𝜃 ) 2
2
1
= 1
2(𝑟𝑒 𝑖𝜃 )2
1
= 1
2𝑧 2
Harmonic Function —
Any function 𝑤(𝑥, 𝑦) of 𝑥, 𝑦 which possesses continuous 2nd order derivatives and satisfies
𝛿2𝑤
Laplace equation is called a harmonic function where Laplace equation is given by +
𝛿𝑥 2
𝛿2𝑤
= 0. If in 𝑓(𝑧) = 𝑢 + 𝑖𝑣 = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) both 𝑢 and 𝑣 are harmonic function then
𝛿𝑦 2
such function 𝑢 and 𝑣 are called conjugate harmonic function or simply conjugate function.
Theorem 1 : Let 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) be an analytic function on a domain D. Then
both 𝑢 and 𝑣 are harmonic functions on D. In other words, the real and imaginary parts of
an analytic function are harmonic.
Proof : If 𝑓(𝑧) is analytic, then all partial derivatives of 𝑢 and 𝑣 are continuous and it
satisfies C-R equations i.e., 𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥 . Taking the partial derivative with
respect to 𝑥 of each size of those equations, we get
𝑢𝑥 = 𝑣𝑦 𝑢𝑦 = −𝑣𝑥
𝛿𝑢 𝛿𝑣 𝛿𝑢 𝛿𝑣
⇒ = ⇒ =−
𝛿𝑥 𝛿𝑦 𝛿𝑦 𝛿𝑥
Taking partial derivative with respect to 𝑥
𝛿2𝑢 𝛿2𝑣 𝛿2𝑢 𝛿2𝑣
2
= ……(1) =− ……(2)
𝛿𝑥 𝛿𝑥𝛿𝑦 𝛿𝑥𝛿𝑦 𝛿𝑥 2
𝛿2𝑢 𝛿2𝑣 𝛿2𝑢 𝛿2𝑣
Similarly, = ……(3) =− ……(4)
𝛿𝑥𝛿𝑦 𝛿𝑦 2 𝛿𝑦 2 𝛿𝑥𝛿𝑦
𝛿2𝑢 𝛿2𝑢
Combining (1) and (4), we get 2
+ =0
𝛿𝑥 𝛿𝑦 2
𝛿2𝑣 𝛿2𝑣
Combining (2) and (3), we get + =0
𝛿𝑥 2 𝛿𝑦 2
Therefore, both 𝑢 and 𝑣 are harmonic functions.
Remark : If we have a function 𝑢(𝑥, 𝑦) that is harmonic on the domain D and if we can find
another harmonic function 𝑣(𝑥, 𝑦) such that the partial derivative for 𝑢 and 𝑣 satisfies C-R
equations throughout the domain D. Then, we say that 𝑣(𝑥, 𝑦) is a harmonic conjugate of
𝑢(𝑥, 𝑦). If it follow that the function 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) is analytic at on D.
For example, if 𝑢(𝑥, 𝑦) = 𝑥 2 − 𝑦 2
Then 𝑢𝑥 = 2𝑥, 𝑢𝑥𝑥 = 2
Also 𝑢𝑦 = −2𝑦, 𝑢𝑦𝑦 = −2
Therefore, 𝑢𝑥𝑥 + 𝑢𝑦𝑦 = 0
Hence, 𝑢(𝑥, 𝑦) is a harmonic function ∀ 𝑧
We find that 𝑣(𝑥, 𝑦) = 2𝑥𝑦
15 | D R C
Then 𝑣𝑥 = 2𝑦, 𝑣𝑥𝑥 = 0
Also 𝑣𝑦 = 2𝑥, 𝑣𝑦𝑦 = 0
Therefore, 𝑣𝑥𝑥 + 𝑣𝑦𝑦 = 0
Hence, 𝑣(𝑥, 𝑦) is a harmonic function ∀ 𝑧
Also, we have 𝑢𝑥 = 𝑣𝑦 and 𝑢𝑦 = −𝑣𝑥
Therefore, they satisfy C-R equations ∀ 𝑧
Hence, 𝑣(𝑥, 𝑦) is a harmonic conjugate of 𝑢(𝑥, 𝑦) and 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦)
= 𝑥 2 − 𝑦 2 + 2𝑖𝑥𝑦
= (𝑥 + 𝑖𝑦)2
= 𝑧 2 is an analytic function.
Construction of Harmonic Conjugate —
Let 𝑢(𝑥, 𝑦) be harmonic in an ϵ neighbourhood of observed point (𝑥0 , 𝑦0 ), then there exist
a conjugate harmonic function 𝑣(𝑥, 𝑦) defined in this neighbourhood such that 𝑓(𝑧) =
𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) is an analytic function. It can be determined by a two-step process:
Step 1 – First, we integrate 𝑣𝑦 (which is equal to 𝑢𝑥 , by C-R equations) with respect to 𝑦
and we get
𝑣(𝑥, 𝑦) = ∫ 𝑣𝑦 (𝑥, 𝑦) 𝑑𝑦 = ∫ 𝑢𝑥 (𝑥, 𝑦) 𝑑𝑦 + 𝐶(𝑥) ……(1)
where 𝐶(𝑥) is a function of 𝑥 alone that is yet to be determined.
Step 2 – We compute 𝐶 ′ (𝑥) by differentiating both sides of (1) with respect to 𝑥 and
replacing 𝑣𝑥 with −𝑢𝑦 on the left side which gives
𝑑
−𝑢𝑦 (𝑥, 𝑦) = ∫ 𝑢𝑥 (𝑥, 𝑦) 𝑑𝑦 + 𝐶 ′ (𝑥) ……(2)
𝑑𝑥
It can be shown that because it is harmonic all terms except those involving 𝑥 in the last
equation (2), we cancel, revealing a formula for 𝐶 ′ (𝑥) involving 𝑥 alone. Elementary
integration of the single variable function 𝐶 ′ (𝑥) can be used to discover 𝐶(𝑥). We finally
observe the function 𝑣 so created that has the properties we seek.
Example 1 – Show that 𝑢(𝑥, 𝑦) = 𝑥𝑦 3 − 𝑥 3 𝑦 is a harmonic function and find a conjugate
harmonic function 𝑣(𝑥, 𝑦)
Solution – Given 𝑢(𝑥, 𝑦) = 𝑥𝑦 3 − 𝑥 3 𝑦
𝑢𝑥 = 𝑦 3 − 3𝑥 2 𝑦 𝑢𝑦 = 3𝑥𝑦 2 − 𝑥 3
𝑢𝑥𝑥 = −6𝑥𝑦 𝑢𝑦𝑦 = 6𝑥𝑦
Therefore, 𝑢𝑥𝑥 + 𝑢𝑦𝑦 = −6𝑥𝑦 + 6𝑥𝑦 = 0
Hence, 𝑢(𝑥, 𝑦) is a harmonic function.
To construct 𝑣 we start with step 1:
𝑦4 𝑦2
𝑣(𝑥, 𝑦) = ∫ 𝑣𝑦 𝑑𝑦 = ∫ 𝑢𝑥 𝑑𝑦 = ∫(𝑦 3 − 3𝑥 2 𝑦) 𝑑𝑦 = − 3𝑥 2 + 𝐶(𝑥)
4 2
𝑦4 𝑦2
Therefore, 𝑣(𝑥, 𝑦) = − 3𝑥 2 + 𝐶(𝑥) ……(1)
4 2
Following step 2, i.e., differentiating (1) with respect to 𝑥 and using −𝑢𝑦 = 𝑣𝑥 we get
3
𝑣𝑥 = − (2𝑥)𝑦 2 + 𝐶 ′ (𝑥)
2
⇒ −(3𝑥𝑦 − 𝑥 3 ) = −3𝑥𝑦 2 + 𝐶 ′ (𝑥)
2

⇒ −3𝑥𝑦 2 + 𝑥 3 = −3𝑥𝑦 2 + 𝐶 ′ (𝑥)

⇒ 𝐶 ′ (𝑥) = 𝑥 3

16 | D R C
 𝑥4
Integrating both sides with respect to 𝑥, we have 𝐶(𝑥) = +𝐶 ……(1)
4
Using (2) in (1), we have
1 3 1
𝑣(𝑥, 𝑦) = 𝑦 4 − 𝑥 2 𝑦 2 + 𝑥 4 + 𝐶
4 2 4
1 3
= (𝑥 4
+ 𝑦 − 𝑥 2𝑦2 + 𝐶
4)
4 2
Example 2 – Does an analytic function 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) exist for which 𝑣(𝑥, 𝑦) =
𝑥 3 + 𝑦 3 ? Why or why not?
Solution – We have 𝑣𝑥𝑥 + 𝑣𝑦𝑦 = 6𝑥 + 6𝑦 = 6(𝑥 + 𝑦) which is 0 only when 𝑥 + 𝑦 = 0.
Hence, 𝑣(𝑥, 𝑦) is not harmonic on a domain D. Therefore, there do not exist any analytic
function 𝑓(𝑧).
Example 3 – Find an analytic function 𝑓(𝑧) = 𝑢(𝑥, 𝑦) + 𝑖𝑣(𝑥, 𝑦) for the following function:
a) 𝑢(𝑥, 𝑦) = 𝑦 3 − 3𝑥 2 𝑦
b) 𝑣(𝑥, 𝑦) = 𝑒 𝑦 sin 𝑥
Solution – a) Given 𝑢(𝑥, 𝑦) = 𝑦 3 − 3𝑥 2 𝑦
𝑢𝑥 = −6𝑥𝑦 𝑢𝑦 = 3𝑦 2 − 3𝑥 2
𝑢𝑥𝑥 = −6𝑦 𝑢𝑦𝑦 = 6𝑦
Therefore, 𝑢𝑥𝑥 + 𝑢𝑦𝑦 = 0
Hence, 𝑢(𝑥, 𝑦) is a harmonic function.
To construct 𝑣 we start with step 1 and integrating with respect to 𝑦, we get
𝑦2
𝑣(𝑥, 𝑦) = ∫ 𝑣𝑦 𝑑𝑦 = ∫ 𝑢𝑥 𝑑𝑦 = ∫(−6𝑥𝑦)𝑑𝑦 = −6𝑥 + 𝐶(𝑥)
2
Therefore, 𝑣(𝑥, 𝑦) = −3𝑥𝑦 2 + 𝐶(𝑥) ……(1)
Following step 2 i.e., differentiating with respect to 𝑥 and using −𝑢𝑦 = 𝑣𝑥 we get
𝑣𝑥 = −3𝑦 2 + 𝐶 ′ (𝑥)
⇒ −(3𝑦 2 − 3𝑥 2 ) = −3𝑦 2 + 𝐶 ′ (𝑥)
⇒ −3𝑦 2 + 3𝑥 2 + 3𝑦 2 = 𝐶 ′ (𝑥)
⇒ 𝐶 ′ (𝑥) = 3𝑥 2
Integrating both sides with respect to 𝑥, we have
𝑥3
𝐶(𝑥) = 3 + 𝐶
3
= 𝑥3 + 𝐶 ……(2)
2 3
Using (2) in (1) we get 𝑣(𝑥, 𝑦) = −3𝑥𝑦 + 𝑥 + 𝐶
= 𝑥 3 − 3𝑥𝑦 2 + 𝐶
b) Given 𝑣(𝑥, 𝑦) = 𝑒 𝑦 sin 𝑥
𝑣𝑥 = 𝑒 𝑦 cos 𝑥 𝑣𝑦 = 𝑒 𝑦 sin 𝑥
𝑣𝑥𝑥 = −𝑒 𝑦 sin 𝑥 𝑣𝑦𝑦 = 𝑒 𝑦 sin 𝑥
Therefore, 𝑣𝑥𝑥 + 𝑣𝑦𝑦 = 0
Hence, 𝑣(𝑥, 𝑦) is a harmonic function.
To construct 𝑢 we start with step 1 and integrating with respect to 𝑥, we get
𝑢(𝑥, 𝑦) = ∫ 𝑢𝑥 𝑑𝑥 = ∫ 𝑣𝑦 𝑑𝑥 = ∫ 𝑒 𝑦 sin 𝑥 𝑑𝑥 = −𝑒 𝑦 cos 𝑥 + 𝐶(𝑥)
Therefore, 𝑢(𝑥, 𝑦) = −𝑒 𝑦 cos 𝑥 + 𝐶(𝑥) ……(1)
Following step 2 i.e., differentiating with respect to 𝑦 and using −𝑣𝑥 = 𝑢𝑦 we get
𝑢𝑦 = −𝑒 𝑦 cos 𝑥 + 𝐶 ′ (𝑥)
17 | D R C
 ⇒ −(𝑒 𝑦 cos 𝑥) = −𝑒 𝑦 cos 𝑥 + 𝐶 ′ (𝑥)
⇒ 𝐶 ′ (𝑥) = −𝑒 𝑦 cos 𝑥 + 𝑒 𝑦 cos 𝑥 = 0
Integrating both sides with respect to 𝑦, we have
𝐶(𝑥) = 0 + 𝐶 ……(2)
𝑦
Using (2) in (1) we get 𝑢(𝑥, 𝑦) = −𝑒 cos 𝑥 + 𝐶
Complex Integration —
Complex integration is divided into two parts:
— Integration of complex valued function with real domain.
— Integration of complex valued function with complex domain.
Integration of Complex Valued Function with Real Domain :—
Let 𝑓 : D → ℂ is a function of real interval D = [𝑎, 𝑏] ⊂ ℝ.
Therefore, for each 𝑡 ∈ [𝑎, 𝑏], 𝑓(𝑡) is a complex number.
Thus, 𝑓(𝑡) = 𝑢(𝑡) + 𝑖𝑣(𝑡)
𝑏 𝑏 𝑏
Therefore, ∫𝑎 𝑓(𝑡)𝑑𝑡 = ∫𝑎 𝑢(𝑡)𝑑𝑡 + 𝑖 ∫𝑎 𝑣(𝑡)𝑑𝑡
If 𝐹1 ′ (𝑡) = 𝑢(𝑡) and 𝐹2 ′ (𝑡) = 𝑣(𝑡)
𝑏
Then ∫𝑎 𝑓(𝑡)𝑑𝑡 = 𝐹1 (𝑡)]𝑏𝑎 + 𝑖𝐹2 (𝑡)]𝑏𝑎 = [𝐹1 (𝑏) − 𝐹1 (𝑎)] + 𝑖[𝐹2 (𝑏) − 𝐹2 (𝑎)]
Remark : This situation is simple to explain and evaluation is done as in case of real valued
function.
1
Example 1 – Let 𝑓(𝑡) = (1 + 𝑡 2 ) + 𝑖𝑡 3 , 𝑡 ∈ [0, 1] then find ∫0 𝑓(𝑡)𝑑𝑡
1 1 1
Solution – We have ∫0 𝑓(𝑡)𝑑𝑡 = ∫0 (1 + 𝑡 2 )𝑑𝑡 + 𝑖 ∫0 𝑡 3 𝑑𝑡
1 1
𝑡3 𝑡4
=𝑡+ ] +𝑖 ]
3 0 4 0
4 1
= +𝑖
3 4
Integration of Complex Valued Function with Complex Domain :—
Let 𝑓 : D → ℂ is a complex valued function on a subset D of the complex plane. (i.e., D ⊂
ℂ) and suppose 𝑧1 and 𝑧2 are complex numbers in the domain of 𝑓. In one dimension there
is only one path joining one number to another but here we must specify the path from 𝑧1 to
𝑧2 . Let 𝑐1 be the curve joining 𝑧1 and 𝑧2 in D then ∫𝑐 𝑓(𝑧)𝑑𝑧 is called a line integration
1
from 𝑧1 to 𝑧2 along 𝑐1 . Since, there may be more than one curve (path) joining 𝑧1 and 𝑧2 , if
𝑐2 be another path joining 𝑧1 and 𝑧2 then ∫𝑐 𝑓(𝑧)𝑑𝑧 is called a line integration from 𝑧1 to
2
𝑧2 along 𝑐2 .
NOTE : We called those curves (or path) as contour. So, line integration are also known as
contour integration. So, to define a contour integration we need
i) 𝑓(𝑧)
ii) interval [𝑧1 , 𝑧2 ] and
iii) path joining 𝑧1 and 𝑧2
𝑧
Example 2 – Evaluate the integral I = ∫𝑧 2 𝑧 2 𝑑𝑧 from
1
i) 𝑧1 = 0 to 𝑧2 = 1 + 𝑖 along 𝑦 = 𝑥 2
ii) along 𝑦 axis from 𝑧1 = 0 to 𝑧2 = 𝑖
iii) 𝑧1 = 0 to 𝑧2 = 1 + 𝑖 along the line joining A(0, 0) to C(0, 1) and C(0, 1) to B(1, 1)

18 | D R C
 𝑧 1+𝑖
Solution – i) I = ∫𝑧 2 𝑓(𝑧)𝑑𝑧 = ∫0 𝑧 2 𝑑𝑧
1
1+𝑖
= ∫0 (𝑥 + 𝑖𝑦)2 (𝑑𝑥 + 𝑖𝑑𝑦)
1+𝑖
= ∫0 (𝑥 2 − 𝑦 2 + 2𝑖𝑥𝑦)(𝑑𝑥 + 𝑖𝑑𝑦)
1+𝑖
= ∫0 [(𝑥 2 − 𝑦 2 )𝑑𝑥 + 2𝑖𝑥𝑦𝑑𝑥 + 𝑖(𝑥 2 − 𝑦 2 )𝑑𝑦 − 2𝑥𝑦𝑑𝑦]
1+𝑖 1+𝑖
= ∫0 [(𝑥 2 − 𝑦 2 )𝑑𝑥 − 2𝑥𝑦𝑑𝑦] + 𝑖 ∫0 [2𝑥𝑦𝑑𝑥 + (𝑥 2 − 𝑦 2 )𝑑𝑦]
……(1)
Since 0 = (0, 0) and 1 + 𝑖 = (1, 1) thus 𝑥 is moving from 0 to 1 and 𝑦 is moving from 0 to 1
along the path 𝑦 = 𝑥 2
1 1
Hence, (1) gives I = ∫0 [(𝑥 2 − 𝑥 4 )𝑑𝑥 − 2𝑥 3 (2𝑥𝑑𝑥)] + 𝑖 ∫0 [2𝑥 3 𝑑𝑥 + (𝑥 2 − 𝑥 4 )(2𝑥𝑑𝑥)]
1 1 1 1
= ∫0 (𝑥 2 − 𝑥 4 )𝑑𝑥 − 4 ∫0 𝑥 4 𝑑𝑥 + 2𝑖 ∫0 𝑥 3 𝑑𝑥 + 2𝑖 ∫0 (𝑥 3 − 𝑥 5 )𝑑𝑥
1 1 1 1
𝑥3 𝑥5 𝑥5 𝑥4 𝑥4 𝑥6
= − ] − 4 ] + 2𝑖 ] + 2𝑖 − ]
3 5 0 5 0 4 0 4 6 0
2 4 1 1
= − +𝑖 +𝑖
15 5 2 6
2 2
=− +𝑖
3 3
2
= (−1 + 𝑖)
3
𝑧2 𝑖
ii) I = = ∫0 𝑧 2 𝑑𝑧
∫𝑧 𝑓(𝑧)𝑑𝑧
1
𝑖
= ∫0 (𝑥 + 𝑖𝑦)2 (𝑑𝑥 + 𝑖𝑑𝑦)
𝑖
= ∫0 (𝑥 2 − 𝑦 2 + 2𝑖𝑥𝑦)(𝑑𝑥 + 𝑖𝑑𝑦)
𝑖
= ∫0 [(𝑥 2 − 𝑦 2 )𝑑𝑥 + 2𝑖𝑥𝑦𝑑𝑥 + 𝑖(𝑥 2 − 𝑦 2 )𝑑𝑦 − 2𝑥𝑦𝑑𝑦]
𝑖 𝑖
= ∫0 [(𝑥 2 − 𝑦 2 )𝑑𝑥 − 2𝑥𝑦𝑑𝑦] + 𝑖 ∫0 [2𝑥𝑦𝑑𝑥 + (𝑥 2 − 𝑦 2 )𝑑𝑦] ……(1)
Since 𝑧1 = 0 = 0 + 0𝑖 = (0, 0) and 𝑧2 = 𝑖 = 0 + 1𝑖 = (0, 1) thus 𝑥 is moving from 0 to 0 and
𝑦 is moving from 0 to 1 along the path 𝑥 = 0
1
1 2 𝑦3 𝑖
Hence, (1) gives I = 𝑖 ∫0 −𝑦 𝑑𝑦 = −𝑖 ] = −
3 0 3
𝑧2 1+𝑖 2
iii) I = ∫𝑧 𝑓(𝑧)𝑑𝑧 = ∫0 𝑧 𝑑𝑧
1
1+𝑖
= ∫0 (𝑥 + 𝑖𝑦)2 (𝑑𝑥 + 𝑖𝑑𝑦)
1+𝑖
= ∫0 (𝑥 2 − 𝑦 2 + 2𝑖𝑥𝑦)(𝑑𝑥 + 𝑖𝑑𝑦)
1+𝑖
= ∫0 [(𝑥 2 − 𝑦 2 )𝑑𝑥 + 2𝑖𝑥𝑦𝑑𝑥 + 𝑖(𝑥 2 − 𝑦 2 )𝑑𝑦 − 2𝑥𝑦𝑑𝑦]
1+𝑖 1+𝑖
= ∫0 [(𝑥 2 − 𝑦 2 )𝑑𝑥 − 2𝑥𝑦𝑑𝑦] + 𝑖 ∫0 [2𝑥𝑦𝑑𝑥 + (𝑥 2 − 𝑦 2 )𝑑𝑦] ……(1)
Let us complete this in two parts:
1st Part : From A(0, 0) to C(0, 1) thus 𝑥 is moving from 0 to 0 and 𝑦 is moving from 0 to 1
along the path 𝑥 = 0
1
1 𝑦3 𝑖
Hence, (1) gives I = 𝑖 ∫0 −𝑦 2 𝑑𝑦 = −𝑖 ] =−
3 0 3

19 | D R C
2nd Part : From C(0, 1) to B(1, 1) thus 𝑥 is moving from 0 to 1 and 𝑦 is moving from 1 to 1
along the path 𝑦 = 1
1 1
Hence, (1) gives I = ∫0 (𝑥 2 − 1)𝑑𝑥 + 𝑖 ∫0 2𝑥𝑑𝑥
1 1
𝑥3 𝑥2
= − 𝑥] + 2𝑖 ]
3 0 2 0
2
=− +𝑖
3
𝑖 2 2
Now, adding both the parts we get I = − + (− + 𝑖) = (−1 + 𝑖)
3 3 3
Remark : Look at problem 1, we have 𝑓(𝑧) = 𝑧 2 from 0 to 1 + 𝑖 along the curve 𝑦 = 𝑥 2 and
2
the answer was (−1 + 𝑖). On the other hand, in problem 3, we have integrated 𝑧 2 from 0
3
to 1 + 𝑖 along the path 𝑐1 ∪ 𝑐2 i.e., 𝑥 = 0 and 𝑦 = 1 but we reach the same answer.
Question 1 – Does the line integration is independent of the path joining 𝑧1 and 𝑧2 .
Solution – If 𝑓(𝑧) is analytic in and on a region D and 𝑧1 and 𝑧2 are any two points in D
𝑧
then ∫𝑧 2 𝑓(𝑧)𝑑𝑧 is independent of the path in D joining 𝑧1 and 𝑧2 .
1
Cauchy’s Theorem —
Statement – Suppose 𝑓(𝑧) is analytic in a region D and on its boundary and c is any closed
curve inside D i.e., ∮𝑐 𝑓(𝑧)𝑑𝑧 = 0 around any closed curve c in D.
Consequences of Cauchy’s Theorem —
Consequence 1 :— Suppose 𝑓(𝑧) is analytic in a region D. Let 𝑧1 and 𝑧2 are two points in
D. Let 𝑐1 and 𝑐2 are paths joining 𝑧1 and 𝑧2 . Let c be a closed path joining 𝑧1 and 𝑧2 along
𝑐1 and 𝑧2 to 𝑧1 along 𝑐2 . Thus, 𝑓(𝑧) is analytic inside and on c and hence, due to Cauchy’s
theorem ∮𝑐 𝑓(𝑧)𝑑𝑧 = 0
𝑧 𝑧
⇒ ∫𝑧 2 𝑓(𝑧)𝑑𝑧 + ∫𝑧 1 𝑓(𝑧)𝑑𝑧 = 0
1 2
along 𝑐1 along 𝑐2
𝑧2 𝑧
⇒ ∫𝑧 𝑓(𝑧)𝑑𝑧 − ∫𝑧 2 𝑓(𝑧)𝑑𝑧 = 0
1 1
along 𝑐1 along 𝑐2
𝑧2 𝑧
⇒ ∫𝑧 𝑓(𝑧)𝑑𝑧 = ∫𝑧 2 𝑓(𝑧)𝑑𝑧
1 1
along 𝑐1 along 𝑐2
𝑧2
Therefore, ∫𝑧 𝑓(𝑧)𝑑𝑧 is independent of the path joining 𝑧1 and 𝑧2 in a region where 𝑓(𝑧)
1
is analytic.
Consequence 2 :— Suppose 𝑓(𝑧) is analytic in a region D and 𝐹 ′ (𝑧) = 𝑓(𝑧) in D then
𝑧2 𝑧2
∫𝑧 𝑓(𝑧)𝑑𝑧 = 𝐹(𝑧2 ) − 𝐹(𝑧1 ). For example, consider our previous example I = ∫𝑧 𝑧 2 𝑑𝑧.
1 1
Clearly, 𝑓(𝑧) = 𝑧 2 is analytic in any region D containing 𝑧1 = 0 and 𝑧2 = 1 + 𝑖.
1+𝑖
1+𝑖 𝑧3 (1+𝑖)3 1−𝑖+3𝑖−3 2
Therefore, ∫0 𝑧 2 𝑑𝑧 = ] = = = (−1 + 𝑖)
3 0 3 3 3
Consequence 3 :— Let 𝑓(𝑧) be analytic in a region bounded by 2 simple closed curves c
and 𝑐1 (where 𝑐1 lies inside c) and on these curves then ∮𝑐 𝑓(𝑧)𝑑𝑧 = ∮𝑐 𝑓(𝑧)𝑑𝑧 . Then
1

20 | D R C
connect the curve c and 𝑐1 by means of a cut GD the 𝑓(𝑧) is
analytic inside and on the region GDFEDGHIJG then <c
J
∮𝐺𝐷𝐹𝐸𝐷𝐺𝐻𝐼𝐽𝐺 𝑓(𝑧)𝑑𝑧 = 0 𝑐
F <1 I
⇒ ∫𝐺𝐷 𝑓(𝑧)𝑑𝑧 + ∮𝐷𝐹𝐸𝐷 𝑓(𝑧)𝑑𝑧 + ∫𝐷𝐺 𝑓(𝑧)𝑑𝑧 +
∮𝐺𝐻𝐼𝐽𝐺 𝑓(𝑧)𝑑𝑧 = 0 G D
⇒ ∫𝐺𝐷 𝑓(𝑧)𝑑𝑧 − ∮𝑐 𝑓(𝑧)𝑑𝑧 − ∫𝐺𝐷 𝑓(𝑧)𝑑𝑧 + E
1
∮𝑐 𝑓(𝑧)𝑑𝑧 = 0
⇒ ∮𝑐 𝑓(𝑧)𝑑𝑧 = ∮𝑐 𝑓(𝑧)𝑑𝑧 H
1

Consequence 4 :— Let 𝑓(𝑧) be analytic in a region bounded by non-overlapping simple

closed curves c, 𝑐1 , 𝑐2 , …., 𝑐𝑛 [where 𝑐1 , 𝑐2 , …., 𝑐𝑛 are inside c as in the following figure].
Then ∮𝑐 𝑓(𝑧)𝑑𝑧 = ∮𝑐 𝑓(𝑧)𝑑𝑧 + ∮𝑐 𝑓(𝑧)𝑑𝑧 +. … + ∮𝑐 𝑓(𝑧)𝑑𝑧
1 2 𝑛
Cauchy’s Integral Formula —
Statement :— Let 𝑓(𝑧) is analytic inside and on a simple closed curve c and a is any point
1 𝑓(𝑧)
inside c, then 𝑓(𝑎) = ∮ 𝑑𝑧 where c moves in counterclockwise.
2𝜋𝑖 𝑐 𝑧−𝑎
Proof : Let us draw a circle 𝜈 with radius 𝜖 and centred at 𝑧 = 𝑎 .
𝑓(𝑧)
Clearly, 𝑔(𝑧) = is analytic in the region between c and 𝜈. Thus, Cauchy’s theorem gives
𝑧−𝑎
I = ∮𝑐 𝑔(𝑧)𝑑𝑧 = ∮𝜈 𝑔(𝑧)𝑑𝑧
𝑓(𝑧) 𝑓(𝑧)
⇒ ∮𝑐 𝑑𝑧 = ∮𝜈 𝑑𝑧 ……(1)
𝑧−𝑎 𝑧−𝑎
Since, 𝜈 is a circle we have |𝑧 − 𝑎| = 𝜖
and 𝑧 − 𝑎 = 𝜖𝑒 𝑖𝜃 ; 0 ≤ 𝜃 ≤ 2π
⇒ 𝑧 = 𝑎 + 𝜖𝑒 𝑖𝜃
𝑑𝑧 = 𝜖𝑖𝑒 𝑖𝜃 𝑑𝜃
𝑓(𝑧) 2𝜋 𝑓(𝑎+𝜖𝑒 𝑖𝜃 ) 2𝜋
Therefore, from (1) ∮𝑐 𝑑𝑧 = ∫0 𝜖𝑖𝑒 𝑖𝜃 𝑑𝜃 = 𝑖 ∫0 𝑓(𝑎 + 𝜖𝑒 𝑖𝜃 ) 𝑑𝜃
𝑧−𝑎 𝜖𝑒 𝑖𝜃
Therefore, 𝜖 → 0 ⇒ 𝑎 + 𝜖𝑒 𝑖𝜃 → 𝑎
⇒ 𝑓(𝑎 + 𝜖𝑒 𝑖𝜃 ) → 𝑓(𝑎)
Since, 𝑓(𝑧) is continuous.
𝑓(𝑧) 2𝜋
Therefore, I = ∮𝑐 𝑑𝑧 = 𝑖 ∫0 𝑓(𝑎) 𝑑𝜃
𝑧−𝑎
2𝜋
= 𝑖𝑓(𝑎) ∫0 𝑑𝜃
= 𝑖𝑓(𝑎)𝜃]2𝜋0
= 𝑖𝑓(𝑎)2𝜋
1 𝑓(𝑧)
⇒ 𝑓(𝑎) = ∮ 𝑑𝑧
2𝜋𝑖 𝑐 𝑧−𝑎
Consequences of Cauchy’s Integral Formula —
Consequence 1 :— Suppose 𝑓(𝑧) is analytic inside and on a closed curve c and a is any
1 𝑓(𝑧)
point inside c. Then 𝑓 ′ (𝑎) = ∮𝑐 (𝑧−𝑎)2 𝑑𝑧
2𝜋𝑖
1 𝑓(𝑧)
Cauchy’s integral formula gives 𝑓(𝑎) = ∮ 𝑑𝑧 ……(1)
2𝜋𝑖 𝑐 𝑧−𝑎

21 | D R C
Let (𝑎 + ℎ) is a point very closed to a, then 𝑎 + ℎ is also inside c.
1 𝑓(𝑧)
𝑓(𝑎 + ℎ) = ∮𝑐 𝑑𝑧 ……(2)
2𝜋𝑖 𝑧−𝑎−ℎ
𝑓(𝑎+ℎ)−𝑓(𝑎) 1 1 𝑓(𝑧) 𝑓(𝑧)
Therefore, = ∮ [ − ] 𝑑𝑧
ℎ 2𝜋𝑖 𝑐 ℎ (𝑧−𝑎) (𝑧−𝑎−ℎ)
1 1 𝑓(𝑧) 𝑓(𝑧)
= ∮ [ − ] 𝑑𝑧
2𝜋𝑖 𝑐 ℎ (𝑧−𝑎−ℎ) (𝑧−𝑎)
1 𝑓(𝑧) 𝑧−𝑎−𝑧+𝑎+ℎ
= ∮ [ ] 𝑑𝑧
2𝜋𝑖 𝑐 ℎ (𝑧−𝑎−ℎ)(𝑧−𝑎)
𝑓(𝑎+ℎ)−𝑓(𝑎) 1 𝑓(𝑧)
⇒ = ∮ [ ] 𝑑𝑧
ℎ 2𝜋𝑖 𝑐 (𝑧−𝑎−ℎ)(𝑧−𝑎)
1 𝑓(𝑧)
As ℎ → 0 we have 𝑓 ′ (𝑎) = ∮ 𝑐 (𝑧−𝑎)2
𝑑𝑧
2𝜋𝑖
Consequence 2 :— Similarly, it can be proved that if 𝑓(𝑧) is analytic within and on a closed
𝑛! 𝑓(𝑧)
curve c and ‘a’ is any point inside c. Then 𝑓 𝑛 (𝑎) = ∮𝑐 (𝑧−𝑎)𝑛+1 𝑑𝑧 2𝜋𝑖
Consequence 3 :— (Taylor’s Series Expansion) If 𝑓(𝑧) is analytic inside and on boundary
of a region D containing a point a, then
𝑓𝑛 (𝑎) 𝑓′′ (𝑎)
𝑓(𝑧) = ∑∞ 𝑛 ′
𝑛=0 𝑛! (𝑧 − 𝑎) = 𝑓(𝑎) + 𝑓 (𝑎)(𝑧 − 𝑎) + (𝑧 − 𝑎)2 +. …
2!
for any 𝑧 inside D.
Let 𝑧 be a point inside D and keep it fixed.
Let c be a circle of radius 𝜖 centred at ‘a’ enclosing 𝑧.
Let 𝑓(𝑤) is analytic inside and on c.
Therefore, Cauchy’s integral formula gives
1 𝑓(𝑤)
𝑓(𝑧) = ∮𝑐 𝑑𝑤
2𝜋𝑖 𝑤−𝑧
1 1 1
Again, = (𝑤−𝑎)−(𝑧−𝑎) = 𝑧−𝑎
𝑤−𝑧 (𝑤−𝑎)[1− ]
𝑤−𝑎
1 𝑧−𝑎
= (𝑤−𝑎)(1−𝑡), where 𝑡 =
𝑤−𝑎
|𝑧−𝑎| 𝜖
⇒ |𝑡| = |𝑤−𝑎| <
𝜖
⇒ |𝑡| < 1
1
Again, = 1 + 𝑡 + 𝑡 2 + 𝑡 3 +. …, |𝑡| < 1
1−𝑡
1 1 𝑧−𝑎 (𝑧−𝑎)2
= (𝑤−𝑎) (1 + + (𝑤−𝑎)2 +. … )
𝑤−𝑧 𝑤−𝑎
𝑓(𝑤) 𝑓(𝑤) (𝑧−𝑎)𝑓(𝑤) (𝑧−𝑎)2 𝑓(𝑤)
⇒ = + (𝑤−𝑎)2
+ (𝑤−𝑎)3
+. …
𝑤−𝑧 𝑤−𝑎
1 𝑓(𝑤) 1 𝑓(𝑤) 1 (𝑧−𝑎) 1 (𝑧−𝑎)2
⇒ ∮ 𝑑𝑤 = ∮ 𝑑𝑤 + ∮ 𝑓(𝑤)𝑑𝑤 + ∮ 𝑓(𝑤)𝑑𝑤 +. …
2𝜋𝑖 𝑐 𝑤−𝑧 2𝜋𝑖 𝑐 𝑤−𝑎 2𝜋𝑖 𝑐 (𝑤−𝑎)2 2𝜋𝑖 𝑐 (𝑤−𝑧)3
′ 𝑓′′ (𝑎) 2 𝑓′′′ (𝑎) 3
⇒ 𝑓(𝑧) = 𝑓(𝑎) + 𝑓 (𝑎)(𝑧 − 𝑎) + (𝑧 − 𝑎) + (𝑧 − 𝑎) +. …
2! 3!
Example 1 – Evaluate the integral
𝑧 2 sin 𝜋𝑧
a) ∮𝑐 1 𝑑𝑧
(𝑧−2)
cos 𝑧
b) ∮𝑐 3 𝑑𝑧
𝑧
around |𝑧| = 1

22 | D R C
 1
Solution – a) Let 𝑓(𝑧) = 𝑧 2 sin 𝜋𝑧 which is analytic everywhere and 𝑎 = lies inside the
2
circle |𝑧| = 1.
Therefore, by Cauchy’s integral formula
𝑧 2 sin 𝜋𝑧
I = ∮𝑐 1 𝑑𝑧
(𝑧−2)
1
= 2𝜋𝑖𝑓 ( )
2
1 2 𝜋
= 2𝜋𝑖 ( ) sin
2 2
𝜋𝑖
=
2
b) Let 𝑓(𝑧) = cos 𝑧 which is analytic everywhere and 𝑎 = 0 is inside |𝑧| = 1
Therefore, by Cauchy’s integral formula
cos 𝑧
I = ∮𝑐 3 𝑑𝑧
𝑧
2𝜋𝑖
= 𝑓 ′′ (𝑎)
2!
= −𝜋𝑖
𝑧 2 −1
Example 2 – Evaluate ∮𝑐 (𝑧−2)2 𝑑𝑧 along (a) |𝑧| = 1, (b) |𝑧| = 3
𝑧 2 −1
Solution – Here, I = ∮𝑐 (𝑧−2)2 𝑑𝑧
a) Along |𝑧| = 1
Here, 𝑎 = 2 lies outside the circle. Hence, by Cauchy’s theorem
𝑧 2 −1
I = ∮𝑐 (𝑧−2)2 𝑑𝑧 = 0
b) Along |𝑧| = 3
Let 𝑓(𝑧) = 𝑧 2 − 1 which is analytic everywhere and 𝑎 = 2 lies inside the circle |𝑧| = 3.
𝑧 2 −1
By Cauchy’s Integral formula, I = ∮𝑐 (𝑧−2)2 𝑑𝑧
= 2𝜋𝑖𝑓 ′ (2)
= 2𝜋𝑖. 2(2)
= 8𝜋𝑖
1
Example 3 – Evaluate ∮𝑐 𝑑𝑧 along |𝑧| = 3
𝑧(𝑧−1)
1
Solution – Here, I = ∮𝑐 𝑑𝑧 along |𝑧| = 3
𝑧(𝑧−1)
1 1 1 1
Now, I = ∮𝑐 ( − ) 𝑑𝑧 = ∮𝑐 𝑑𝑧 − ∮𝑐 𝑑𝑧 = 𝐼1 − 𝐼2
𝑧−1 𝑧 𝑧−1 𝑧
Here, in case of both 𝐼1 and 𝐼2 we have 𝑓(𝑧) = 1 is analytic everywhere and for 𝐼1 , 𝑎 = 1 is
inside the circle |𝑧| = 3. Also, for 𝐼2 , 𝑎 = 0 is also inside the circle |𝑧| = 3
1 1
By Cauchy’s Integral formula, I = ∮𝑐 𝑑𝑧 − ∮𝑐 𝑑𝑧
𝑧−1 𝑧
= 2𝜋𝑖. 𝑓(1) − 2𝜋𝑖. 𝑓(0)
=0
𝑧2
Example 4 – Evaluate ∮𝑐 𝑑𝑧 along (a) |𝑧 − 𝑖| = 1, (b) |𝑧 − 1| = 2
𝑧 2 +1
𝑧2
Solution – Here, I = ∮𝑐 2 𝑑𝑧
𝑧 +1
a) Along |𝑧 − 𝑖| = 1
23 | D R C
 𝑧2 𝑧2 𝑧 2 ⁄(𝑧+𝑖)
Now, I = ∮𝑐 𝑑𝑧 = ∮𝑐 (𝑧+𝑖)(𝑧−𝑖) 𝑑𝑧 = ∮𝑐 𝑑𝑧
𝑧 2 +1 (𝑧−𝑖)
𝑧2
Let 𝑓(𝑧) = is analytic in the given region (except for 𝑧 = −𝑖 it is not analytic but this
𝑧+𝑖
lies outside the circle |𝑧 − 𝑖| = 1) and 𝑎 = 𝑖 lies inside the circle |𝑧 − 𝑖| = 1
𝑧 2 ⁄(𝑧+𝑖)
By Cauchy’s Integral, I = ∮𝑐 (𝑧−𝑖)
𝑑𝑧
𝑓(𝑧)
= ∮𝑐 (𝑧−𝑖) 𝑑𝑧
= 2𝜋𝑖. 𝑓(𝑖)
−1
= 2𝜋𝑖.
2𝑖
= −𝜋
b) Along |𝑧 − 1| = 2
𝑧2 𝑧2 1 1 1
Now, I = ∮𝑐
𝑧 2 +1
𝑑𝑧 = ∮𝑐 (𝑧+𝑖)(𝑧−𝑖) 𝑑𝑧 = ∮ 𝑧 2 (𝑧−𝑖 − 𝑧+𝑖) 𝑑𝑧
2𝑖 𝑐
1 𝑧2 1 𝑧2
= ∮𝑐 𝑑𝑧 − ∮𝑐 𝑑𝑧
2𝑖 𝑧−𝑖 2𝑖 𝑧+𝑖
= 𝐼1 − 𝐼2
Here, in case of both 𝐼1 and 𝐼2 we have 𝑓(𝑧) = 𝑧 2 is analytic everywhere and for 𝐼1 , 𝑎 =
𝑖 lies inside the circle |𝑧 − 1| = 2. Also, for 𝐼2 , 𝑎 = −𝑖 lies inside the circle |𝑧 − 1| = 2.
1 𝑧2 1 𝑧2
By Cauchy’s integral formula, I = ∮ 𝑑𝑧 − ∮ 𝑑𝑧
2𝑖 𝑐 𝑧−𝑖 2𝑖 𝑐 𝑧+𝑖
1 1
= . 2𝜋𝑖. 𝑓(𝑖) − . 2𝜋𝑖. 𝑓(−𝑖)
2𝑖 2𝑖
1 1
= . 2𝜋𝑖. (−1) − . 2𝜋𝑖. (−1)
2𝑖 2𝑖
=0
(𝑧−1)
Example 5 – Evaluate ∮𝑐 2 𝑑𝑧 around the square whose vertices are (1, 1), (-1, 1), (1,
2𝑧 +3𝑧−2
-1), (-1, -1).
(𝑧−1)
Solution – Here, I = ∮𝑐 2 𝑑𝑧
2𝑧 +3𝑧−2
The vertex of the square are (1, 1), (-1, 1), (1, -1), (-1, -1)
(𝑧−1)
Now, I = ∮𝑐 2 𝑑𝑧
2𝑧 +3𝑧−2
(𝑧−1)
= ∮𝑐 𝑑𝑧
2𝑧 2 +4𝑧−𝑧−2
(𝑧−1)
= ∮𝑐 𝑑𝑧
2𝑧(𝑧+2)−1(𝑧+2)
(𝑧−1)
= ∮𝑐 (2𝑧−1)(𝑧+2) 𝑑𝑧
(𝑧−1)
= ∮𝑐 1 𝑑𝑧
2(𝑧+2)(𝑧−2)
(𝑧−1)⁄2(𝑧+2)
= ∮𝑐 1 𝑑𝑧
(𝑧−2)
𝑧−1
Let 𝑓(𝑧) = is analytic in the given region (except for 𝑧 = −2 it is not analytic but this
2(𝑧+2)
1
lies outside the given square) and 𝑎 = lies inside the given square.
2
(𝑧−1)⁄2(𝑧+2)
By Cauchy’s integral formula, I = ∮𝑐 1 𝑑𝑧
(𝑧−2)

24 | D R C
 𝑓(𝑧)
= ∮𝑐 1 𝑑𝑧
(𝑧−2)
1
= 2𝜋𝑖. 𝑓 ( )
2
1
−1
2
= 2𝜋𝑖. 1
2(2+2)
1
−
2
= 2𝜋𝑖.
5
𝜋𝑖
=−
5
𝑧 2 +1
Example 6 – Evaluate ∮𝑐 𝑑𝑧 along a circle of radius 1 with centre at 1.
𝑧 2 −1
2
𝑧 +1
Solution – Here , I = ∮𝑐 𝑑𝑧 along |𝑧 − 1| = 1
𝑧 2 −1
𝑧 2 +1 𝑧 2 +1 (𝑧 2 +1)⁄(𝑧+1)
Now, I = ∮𝑐 𝑑𝑧 = ∮𝑐 (𝑧+1)(𝑧−1) 𝑑𝑧 = ∮𝑐 𝑑𝑧
𝑧 2 −1 (𝑧−1)
𝑧 2 +1
Let 𝑓(𝑧) = is analytic in the given region (for 𝑧 = −1 it is not analytic but this lies
𝑧+1
outside the circle) and 𝑎 = 1 lies inside the circle |𝑧 − 1| = 1
(𝑧 2 +1)⁄(𝑧+1)
By Cauchy’s integral formula, I = ∮𝑐 (𝑧−1)
𝑑𝑧
𝑓(𝑧)
= ∮𝑐 (𝑧−1) 𝑑𝑧
= 2𝜋𝑖. 𝑓(1)
12 +1
= 2𝜋𝑖.
1+1
= 2𝜋𝑖
Liouville’s Theorem —
Statement :— Every bounded entire function is constant.
Proof : Let 𝑓 be an entire function. Suppose ∃ M ∈ ℝ+
Consider a circle 𝐵(𝑧, 𝑅) with radius R and centred at 𝑧
Let c be a closed curve as shown in the figure.
Then by definition c = {𝑤 ∈ ℂ: |𝑤 − 𝑧| = 𝑅}. Since, 𝑓 is analytic in 𝐵(𝑧, 𝑅) and on c and 𝑧
is a point inside c.
1 𝑓(𝑤)
Therefore, by Cauchy’s integral formula 𝑓 ′ (𝑧) = ∮𝑐 (𝑤−𝑧)2 𝑑𝑤 2𝜋𝑖
𝑖𝜃
Now, |𝑤 − 𝑧| = R ⇒ 𝑤 − 𝑧 = 𝑅𝑒 ; 0 < 𝜃 ≤ 2𝜋
⇒ 𝑤 = 𝑧 + 𝑅𝑒 𝑖𝜃
𝑑𝑤 = 𝑖𝑅𝑒 𝑖𝜃 𝑑𝜃
1 2𝜋 𝑓(𝑧+ 𝑅𝑒 𝑖𝜃 )
Therefore, 𝑓 ′ (𝑧) = ∫ 𝑖𝑅𝑒 𝑖𝜃 𝑑𝜃
2𝜋𝑖 0 𝑅2
1 2𝜋 𝑓(𝑧+ 𝑅𝑒 𝑖𝜃 )𝑒 𝑖𝜃
Now, |𝑓 ′ (𝑧)| = |2𝜋 ∫0 𝑑𝜃|
𝑅
1 2𝜋 |𝑓(𝑧+ 𝑅𝑒 𝑖𝜃 )||𝑒 𝑖𝜃 | 𝑏 𝑏
≤ ∫ 𝑑𝜃 {|∫𝑎 𝑓(𝑧)𝑑𝑧| ≤ ∫𝑎 |𝑓(𝑧)|𝑑𝑧}
2𝜋 0 𝑅
12𝜋
= ∫ |𝑓(𝑧 + 𝑅𝑒 𝑖𝜃 )|𝑑𝜃
2𝜋𝑅 0
1 2𝜋
≤ ∫0
𝑀𝑑𝜃
2𝜋𝑅

25 | D R C
 𝑀 2𝜋
= ∫ 𝑑𝜃
2𝜋𝑅 0
𝑀
= . 2𝜋
2𝜋𝑅
𝑀
=
𝑅
′ 𝑀
Therefore, |𝑓 (𝑧)| ≤
𝑅
Now, since R is arbitrary.
𝑀
lim |𝑓 ′ (𝑧)| ≤ lim
𝑅→∞ 𝑅→∞ 𝑅
⇒ |𝑓 ′ (𝑧)|
≤0
′ (𝑧)
⇒𝑓 =0
⇒ 𝑓(𝑧) = c (constant)
Hence, 𝑓(𝑧) is constant.

26 | D R C