0% found this document useful (0 votes)
0 views31 pages

complex_analysis_notes.html

The document provides comprehensive exam-winning notes on Complex Analysis by Prof. J. K. Sahoo, covering key concepts from complex numbers to harmonic functions across eight lectures. It includes definitions, theorems, formulas, and sample problems, structured into chapters that explore topics like limits, continuity, differentiability, and complex functions. The notes aim to facilitate understanding and application of complex analysis in various fields such as physics and engineering.

Uploaded by

Mitesh Samal
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
0 views31 pages

complex_analysis_notes.html

The document provides comprehensive exam-winning notes on Complex Analysis by Prof. J. K. Sahoo, covering key concepts from complex numbers to harmonic functions across eight lectures. It includes definitions, theorems, formulas, and sample problems, structured into chapters that explore topics like limits, continuity, differentiability, and complex functions. The notes aim to facilitate understanding and application of complex analysis in various fields such as physics and engineering.

Uploaded by

Mitesh Samal
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

BITS PILANI · K.K.

BIRLA GOA CAMPUS

Complex Analysis
Complete Exam-Winning Notes — Prof. J. K. Sahoo

Covers every definition, theorem, formula, and concept from all 8 lectures.
Lectures 1–2 (Complex Numbers), 3 (Limits & Continuity), 4–5
(Differentiability & Analyticity), 6 (Analyticity), 7 (Harmonic Functions), 8
(Sample Problems) + Chapter 4 (Elementary Functions).

L1: Complex Numbers L2: Topology of ℂ + Functions

L3: Limits & Continuity L4–5: Differentiability & C-R L6: Analyticity

L7: Harmonic Functions Ch4: Exp / Log / Trig / Powers L8: MCQ Bank

MARCH–APRIL 2026 · CHAPTERS 1, 2, 3, 4

CONTENTS

01 Why Complex Numbers 02 Complex Number Basics


03 Modulus, Conjugate & Algebra 04 Polar Form & Arguments
05 Euler & De Moivre 06 nth Roots
07 Topology of ℂ 08 Complex Functions
09 Limits — Formal Definition 10 Limit Theorems & Tools
11 Limits at ∞ 12 Continuity
13 Derivative & Differentiability 14 Cauchy-Riemann Equations
15 C-R in Polar Form 16 Analytic Functions
17 Harmonic Functions 18 Harmonic Conjugate
19 Complex Exponential 20 Complex Logarithm
21 Complex Powers 22 Trig Functions in ℂ

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
23 MCQ Answers — L8

L1 · §01 Why Complex Numbers?

Real numbers cannot solve x² + 1 = 0. We extend ℝ by introducing a new symbol i


such that:

IMAGINARY UNIT

i = √(−1), i² = −1

This leads to the set of complex numbers ℂ, developed rigorously by Gauss and
Cauchy. Every complex number has the form z = a + ib where a, b ∈ ℝ.

WHY THIS MATTERS IN PHYSICS/ENGINEERING

Complex numbers elegantly encode oscillations, electrical impedance, fluid


flow, quantum mechanics, and signal processing — they are not just algebraic
curiosities.

L1 · §02 Complex Number — Definition & Structure

DEFINITION

A complex number is z = a + ib where a, b ∈ ℝ.


• Re(z) = a (real part) | Im(z) = b (imaginary part)
• i² = −1 is the imaginary unit.

Representation as Ordered Pairs (ℝ²)


Identify z = (x, y) with binary operations:

z₁ + z₂ = (x₁+x₂, y₁+y₂)

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
z₁ · z₂ = (x₁x₂ − y₁y₂, x₁y₂ + x₂y₁)

Setting i = (0,1): notice (0,1)·(0,1) = (−1,0) = −1, confirming i² = −1.


So (x,y) = x + iy and ℂ ≅ ℝ² as a vector space (basis: {1, i}, dimension 2).

Argand Plane / Complex Plane

y z=(x,y)

r=|z|

Real axis θ
x
x

Im axis

The Argand plane: z = x + iy visualised as point (x, y) with polar coordinates (r, θ)

Equality

z₁ = z₂ ⟺ a₁ = a₂ AND b₁ = b₂

L1 · §03 Modulus, Conjugate & Algebraic Properties

Modulus (Absolute Value)

DEFINITION

|z| = √(x² + y²) (distance from origin to z)

Conjugate

DEFINITION

z̄ = a − ib (reflect across real axis)

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Key: |z|² = z · z̄

Division Formula

z₁/z₂ = (z₁·z̄ ₂) / (z₂·z̄ ₂) = (a₁a₂+b₁b₂)/(a₂²+b₂²) +


i(b₁a₂−a₁b₂)/(a₂²+b₂²)

Algebraic Properties MUST KNOW ALL

Property Formula

Power of modulus |zⁿ| = |z|ⁿ

Modulus of quotient |z₁/z₂| = |z₁|/|z₂|

Triangle Inequality |z₁ + z₂| ≤ |z₁| + |z₂|

Reverse triangle ineq. |z₁ + z₂| ≥ ||z₁| − |z₂||

Conjugate of modulus |z̄ | = |z|

Conjugate of product z₁z₂ bar = z̄ ₁ · z̄ ₂

Conjugate of quotient (z₁/z₂) bar = z̄ ₁/z̄ ₂

Parallelogram Law |z₁+z₂|² + |z₁−z₂|² = 2(|z₁|² + |z₂|²)

Non-zero condition z ≠ 0 ⟺ |z| ≠ 0

L1 · §04 Polar Form & Arguments

Polar Representation

POLAR FORM

z = r(cos θ + i sin θ) = r·e^(iθ)


where r = |z|, θ = arg(z), x = r cos θ, y = r sin θ

PRINCIPAL ARGUMENT

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
arg(z) = multi-valued, defined up to multiples of 2π: arg(z) = θ + 2nπ, n∈ℤ

Arg(z) = principal value, unique value in (−π, π]

Computing Arg(z) for z = x + iy

ARG(Z) BY QUADRANT

Arg(z) = arctan(y/x) if x > 0


= π + arctan(y/x) if x < 0 and y ≥ 0
= −π + arctan(y/x) if x < 0 and y < 0
= π/2 if x = 0 and y > 0
= −π/2 if x = 0 and y < 0

Argument Properties

arg(z₁z₂) = arg(z₁) + arg(z₂)


arg(z₁/z₂) = arg(z₁) − arg(z₂)
arg(z̄ ) = −arg(z)

⚠ CRITICAL TRAP

Arg(z₁z₂) ≠ Arg(z₁) + Arg(z₂) in general!


Counter-example: z₁ = −1, z₂ = 5i. Arg(z₁) = π, Arg(z₂) = π/2, sum = 3π/2. But
Arg(z₁z₂) = Arg(−5i) = −π/2. These are not equal.

WORKED EXAMPLE

Find Arg(z) and arg(z) for z = −1 − i


x = −1 < 0, y = −1 < 0 → third quadrant formula.
Arg(z) = −π + arctan(−1/−1) = −π + arctan(1) = −π + π/4 = −3π/4
arg(z) = −3π/4 + 2nπ, n ∈ ℤ

L1 · §05 Euler's Formula & De Moivre's Theorem

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
EULER'S FORMULA

e^(iθ) = cos θ + i sin θ

e^(iπ) = −1 (Euler's identity)

DE MOIVRE'S FORMULA CORE FORMULA

(cos θ + i sin θ)ⁿ = cos(nθ) + i sin(nθ)

Equivalently: zⁿ = rⁿ(cos(nθ) + i sin(nθ))

APPLICATION

Compute (1 + √3 i)⁹
r = √(1+3) = 2, θ = arctan(√3/1) = π/3.
(1+√3i)⁹ = 2⁹ · (cos(9π/3) + i sin(9π/3)) = 512(cos 3π + i sin 3π) = 512(−1 + 0) = −512

L1 · §06 nth Roots of a Complex Number

FORMULA FOR N-TH ROOTS KEY FORMULA

The n distinct nth roots of z = r(cos θ + i sin θ) are:

wₖ = ⁿ√r · [ cos((θ + 2kπ)/n) + i sin((θ + 2kπ)/n) ]

for k = 0, 1, 2, …, n−1

Principal nth root: k = 0 with θ = Arg(z).


The n roots are equally spaced on a circle of radius ⁿ√r, separated by angle 2π/n.

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
w₁

2π/n
r^(1/n)
w₂ w₀

w₃

4 equally-spaced 4th roots on a circle of radius r^(1/4)

EXAMPLE

Find all values of (1−i)^(1/3)


|1−i| = √2, Arg(1−i) = −π/4.
wₖ = 2^(1/6) · [cos((−π/4 + 2kπ)/3) + i sin((−π/4 + 2kπ)/3)], k=0,1,2

L2 · §07 Topology of the Complex Plane

Basic Sets

Name Definition Visual

Circle (center z₀, radius r) {z : |z−z₀| = r} Boundary ring

Disk (closed) {z : |z−z₀| ≤ r} Filled disc incl. boundary

Neighborhood of z₀ {z : |z−z₀| < r} Open disc

Deleted (punctured) nbhd {z : 0 < |z−z₀| < r} Open disc minus center

Annulus {z : r₁ < |z−z₀| < r₂} Ring-shaped region

Point Classification

INTERIOR / BOUNDARY / EXTERIOR POINTS

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Interior point of S: ∃ a neighborhood of z₀ lying entirely in S.
Boundary point of S: every neighborhood contains at least one point of S AND
one point of Sᶜ.
Exterior point: ∃ neighborhood with no points of S.

Set Properties

Property Definition

Open Contains none of its boundary points (every point is


interior)

Closed Contains all of its boundary points

Connected Any two points can be joined by a polygonal path in S

Domain Non-empty open connected set

Region Domain + some/all/none of its boundary points

Bounded S ⊂ {|z| < R} for some R > 0

Accumulation/Limit Every deleted nbhd of z₀ contains a point of S


point

⚠ QUIZ TRAPS — SET CLASSIFICATION

{z : |z| < 1} → Open, connected


{z : Re(z) = 1} → Closed, connected (a line)
{z : Re(z²) < 1} → Open, connected
{z : eᶻ = 1} → Closed, disconnected (z = 2nπi, discrete)
{z : |z−1| < |z+1|} → Open, connected (left half-plane Re(z) < 0)
{z : 0 < |z−1| < 2} → Open, connected (punctured disk)
{z : 1 < |z| < 2} → Domain (open and connected)
ℂ\S where S = {|z| < 1} → Closed (the closed exterior disk)

L2 · §08 Complex Functions

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
COMPLEX FUNCTION

f : S ⊆ ℂ → ℂ. Written as w = f(z) = u(x,y) + iv(x,y)


where u = Re(f), v = Im(f) are real-valued functions of two real variables.

The domain is the largest set where f is well-defined (unless stated otherwise).

Real and Imaginary Parts

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


Polar: f(r,θ) = u(r,θ) + iv(r,θ), z = re^(iθ)

EXAMPLE — DECOMPOSE F(Z) = Z + 1/Z

u(x,y) = x + x/(x²+y²)
v(x,y) = y − y/(x²+y²)
Polar: f(r,θ) = (r + 1/r)cos θ + i(r − 1/r)sin θ

Complex Exponential (Preview)

e^z = e^x(cos y + i sin y) (defined for all z ∈ ℂ)

L3 · §09 Limits — Formal Definition

Ε - Δ DEFINITION OF LIMIT

lim(z→z₀) f(z) = L means: for every ε > 0 there exists δ > 0 such that

|f(z) − L| < ε whenever 0 < |z − z₀| < δ

UNIQUENESS OF LIMITS

If lim(z→z₀) f(z) exists, it is unique.


Proof: contradiction using two assumed limits L₁ ≠ L₂, using triangle

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
inequality:
|L₁−L₂| ≤ |f(z)−L₁| + |f(z)−L₂| < 2ε for every ε > 0 → contradiction.

Key Difference from Real Analysis


z approaches z₀ from all directions simultaneously in ℂ (from any path in 2D). This
means limits are harder to prove and easier to disprove.

⚠ HOW TO DISPROVE A LIMIT EXISTS

Find two different paths to z₀ that give different limiting values. This is the
standard approach for non-existence proofs.

EXAMPLE — LIMIT DOES NOT EXIST

Show lim(z→0) z̄ /z does not exist.


Write z̄ /z = (x²−y²)/(x²+y²) + i·2xy/(x²+y²).
Along y=0: limit of u = 1. Along x=0: limit of u = −1. Different → limit DNE.

EXAMPLE — Ε-Δ PROOF

Show lim(z→1) (z + iz̄ ) = 1+i.


|f(z)−(1+i)| = |z−1+i(z̄ −1)| ≤ |z−1| + |z̄ −1| = 2|z−1| < 2δ.
Choose δ = ε/2. ✓

L3 · §10 Limit Theorems & Theorem-A

THEOREM-A — REDUCE TO REAL LIMITS ESSENTIAL TOOL

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

lim(z→z₀) f(z) = L = u₀+iv₀ ⟺ lim u(x,y) = u₀ AND lim v(x,y) = v₀


(both real 2-variable limits, as (x,y)→(x₀,y₀))

Standard Limit Rules

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
If lim f(z) = L and lim g(z) = M (both as z→z₀):

lim cf(z) = cL
lim [f(z) ± g(z)] = L ± M
lim [f(z)·g(z)] = L·M
lim [f(z)/g(z)] = L/M, provided M ≠ 0
lim [polynomial p(z)] = p(z₀)
lim e^(cz) = e^(cz₀)

EXAMPLE USING THEOREM-A

Show lim(z→0) z/Re(z) does not exist.


u(x,y) = x²/(x²+y²) for Re(z)=x ≠ 0, v = xy/(x²+y²)... Wait, f(z) = z/x = (x+iy)/x = 1 +
i(y/x).
Along y=0: f = 1+i·0 = 1. Along y=x: f = 1+i. Different limits → DNE.

L3 · §11 Limits at ∞ and Infinite Limits

The Extended Complex Plane ℂ∞ = ℂ ∪ {∞} (one-point compactification — only one ∞,


no ±∞).

THREE LIMIT RULES AT INFINITY

1. lim(z→z₀) f(z) = ∞ ⟺ lim(z→z₀) 1/f(z) = 0


2. lim(z→∞) f(z) = L ⟺ lim(z→0) f(1/z) = L
3. lim(z→∞) f(z) = ∞ ⟺ lim(z→0) 1/f(1/z) = 0

EXAMPLES

lim(z→1) 1/(z−1)³ = ∞: since lim(z→1)(z−1)³ = 0. ✓

lim(z→∞) 4z²/(z−1)² = 4: substitute z = 1/z, lim(z→0) 4/z²/(1/z−1)² = lim 4/(1−z)² =


4. ✓

lim(z→∞) (z²+1)/(z−1) = ∞: lim(z→0) (z−1)/((1/z)²+1) = lim z(z−1)/(1+z²) = 0. ✓

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
L3 · §12 Continuity

THREE CONDITIONS FOR CONTINUITY AT Z ₀

f is continuous at z₀ iff ALL THREE hold:


1. lim(z→z₀) f(z) exists
2. f(z₀) is defined
3. lim(z→z₀) f(z) = f(z₀)

Equivalent ε-δ form: for each ε>0 ∃δ>0 such that |f(z)−f(z₀)| < ε whenever |z−z₀| < δ

Continuity Theorems

THEOREM — COMPONENT-WISE

f(z) = u(x,y) + iv(x,y) is continuous at z₀ = x₀+iy₀


⟺ u(x,y) and v(x,y) are both continuous at (x₀,y₀).

THEOREM — ALGEBRA OF CONTINUITY

If f, g continuous at z₀, then: f±g, f·g, fⁿ, f∘g, g∘f all continuous at z₀.
f/g continuous at z₀ if g(z₀) ≠ 0.
Corollary: All complex polynomials are continuous everywhere.

THEOREM — NON-ZERO PERSISTENCE

If f is continuous and f(z₀) ≠ 0, then ∃ neighborhood of z₀ where f is non-zero.

THEOREM — BOUNDEDNESS ON CLOSED BOUNDED REGION

If f is continuous on a closed bounded region R, then ∃ M > 0 with |f(z)| ≤ M for


all z∈R, and |f(z)| = M for at least one z.

CONTINUITY — QUIZ EXAMPLES

• f(z) = (z−3i)/(z²+2z−1), z₀=1+i: denominator at z₀ = −1+2i+2+2i−1 = −2i+4i ≠ 0 →


continuous

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
• f(z) = z/|z| at z₀=0: limit DNE (|lim along any direction| = 1 ≠ f(0)) →
discontinuous

L3–5 · §13 Derivative & Differentiability

COMPLEX DERIVATIVE

f′(z₀) = lim(Δz→0) [f(z₀+Δz) − f(z₀)] / Δz


If this limit exists, f is differentiable at z₀.

Δz → 0 in any direction simultaneously — this is far more restrictive than real


differentiability.

Differentiation Rules

d/dz [c] = 0 d/dz [cz] = c


d/dz [zⁿ] = nzⁿ⁻¹ (n integer)
[f ± g]′ = f′ ± g′
[fg]′ = f′g + fg′ (Product Rule)
[f/g]′ = (gf′ − fg′)/g² (Quotient Rule, g≠0)
[f(g(z))]′ = f′(g(z))·g′(z) (Chain Rule)
d/dz [g(z)]ⁿ = n[g(z)]ⁿ⁻¹ g′(z)

THEOREM — DIFFERENTIABLE ⟹ CONTINUOUS

If f is differentiable at z₀, then f is continuous at z₀.


(Converse is FALSE — e.g. |z| is continuous everywhere but nowhere
differentiable)

WORKED EXAMPLE — |Z|²

f(z) = |z|² = x²+y²


At z=0: f′(0) = lim(Δz→0) |Δz|²/Δz = lim Δz̄ = 0 → differentiable at 0, f′(0) = 0.
At z₀≠0: f′(z₀) = lim(Δz→0) [Δz̄ + z₀ + z̄ ₀·(Δz/Δz̄ ) · (Δz̄ /Δz)] — the limit of Δz̄ /Δz

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
depends on direction → does not exist.
Conclusion: |z|² is differentiable only at z=0.

L3–5 · §14 Cauchy-Riemann Equations HIGHEST PRIORITY

CAUCHY-RIEMANN (CR) EQUATIONS — CARTESIAN FORM

uₓ = v_y and u_y = −vₓ

THEOREM-B — NECESSARY CONDITION FOR DIFFERENTIABILITY

If f(z) = u+iv is differentiable at z₀, then:


1. First-order partials of u, v exist at (x₀, y₀)
2. u, v satisfy the CR equations at (x₀, y₀)
3. f′(z₀) = uₓ(x₀,y₀) + ivₓ(x₀,y₀)

Equivalently: f′(z₀) = v_y(x₀,y₀) − iu_y(x₀,y₀)

THEOREM-C — SUFFICIENT CONDITION FOR DIFFERENTIABILITY

If in a neighborhood of z₀:
• all first partials uₓ, u_y, vₓ, v_y exist,
• they are continuous at (x₀, y₀), AND
• they satisfy the CR equations at (x₀, y₀),

Then f is differentiable at z₀ and f′(z₀) = uₓ + ivₓ.

NOT differentiable at z₀ ← CR equations fail at z₀ |


Differentiable at z₀ ← CR + continuous partials

⚠ CRITICAL SUBTLETY — CR ALONE IS NOT SUFFICIENT

CR equations at a point do NOT guarantee differentiability unless the partials


are also continuous.
Example: f(z) = √|xy| satisfies CR at origin but is NOT differentiable there.

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Differentiability Tests — Standard Functions

Function CR Satisfied? Differentiable? Where?

f(z) = z u=x,v=y: uₓ=1=vy, uy=0=−vx ✓ Yes Everywhere


(=x+iy)

f(z) = z̄ uₓ=1≠−1=vy ✗ No Nowhere


(=x−iy)

f(z) = |z|² Only at origin Only at z=0 z=0 only

f(z) = Re(z) uₓ=1≠0=vy ✗ No Nowhere

f(z) = Im(z) uₓ=0≠1=vy ✗ No Nowhere

f(z) = eᶻ uₓ=eˣcosy=vy ✓ Yes Everywhere

f(z) = x²+iy² uₓ=2x=vy=2y → only if x=y Only on line Line y=x (but not
x=y analytic)

f(z) = zRe(z) = uₓ=2x, vy=ix → CR gives x=0, Only at z=0 z=0 only
x²+ixy y=0

f(z) = z̄ ² u=x²−y², v=−2xy: Only at z=0 z=0


uₓ=2x≠−2x=vy unless x=0 and
y=0

DETAILED CR CHECK — F(Z) = E ᶻ

f(z) = e^(x+iy) = eˣcos y + i eˣsin y


u = eˣcos y, v = eˣsin y
uₓ = eˣcos y = vy ✓ u_y = −eˣsin y = −vₓ ✓
All partials continuous → differentiable everywhere, f′(z) = eᶻ

L5 · §15 Cauchy-Riemann Equations in Polar Form

CR EQUATIONS — POLAR FORM

u_r = (1/r)·v_θ and v_r = −(1/r)·u_θ

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
DERIVATIVE FORMULA IN POLAR FORM

If f(z) = u(r,θ) + iv(r,θ) and the polar CR hold + partials continuous at z₀ =


r₀e^(iθ₀), then:

f′(z₀) = e^(−iθ) [u_r(r₀,θ₀) + iv_r(r₀,θ₀)]

EXAMPLE — F(Z) = Z^(1/2) = √ R · E^(IΘ/2) (R>0, 0<Θ<2Π)

u = √r cos(θ/2), v = √r sin(θ/2)
u_r = cos(θ/2)/(2√r), v_θ = √r/2 · (−sin(θ/2)) ... verify r·u_r = v_θ: r·cos(θ/2)/(2√r) =
√r·cos(θ/2)/2 = v_θ ✓
f′(z) = e^(−iθ)[cos(θ/2)/(2√r) + i sin(θ/2)/(2√r)] = 1/(2√r · e^(iθ/2)) = 1/(2√z)

L6 · §16 Analytic Functions CORE CONCEPT

ANALYTIC AT A POINT

f is analytic at z₀ if f is differentiable at every point in some neighborhood of


z₀ .

Note: Analytic at z₀ ≠ merely differentiable at z₀ — you need differentiability in


a whole open disc around z₀.

ANALYTIC IN A DOMAIN / ENTIRE FUNCTION

f is analytic in domain D (holomorphic / regular) if analytic at every point of D.


f is entire if analytic on all of ℂ.

Implication Chain QUIZ READY

Analytic at z₀ ⟹ Differentiable at z₀ ⟹ Continuous at z₀

Analytic at z₀ ⟺ Differentiable in a full neighborhood of z₀

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
⚠ THE CONVERSE TRAPS

• Differentiable at z₀ does NOT mean analytic at z₀ (e.g. |z|² differentiable only at


0, not analytic there)
• CR equations + continuity everywhere ⟹ Analytic everywhere (entire)
• CR equations failing somewhere in a neighborhood ⟹ Not analytic at z₀

Properties of Analytic Functions


Sum, difference, product, quotient (g≠0), composition of analytic functions are
analytic.
Polynomials are entire.
eᶻ is entire.
|z|² is nowhere analytic.

Key Theorems on Analytic Functions

THEOREM — ZERO DERIVATIVE ⟹ CONSTANT

If f is analytic on a domain D (connected!) and f′(z) = 0 for all z ∈ D, then f is


constant on D.
Note: Fails on disconnected sets.

THEOREM — CONSTANT MODULUS ⟹ CONSTANT

If f is analytic on D and |f(z)| is constant on D, then f is constant on D.


Proof: u²+v² = c. Differentiating: uuₓ+vvₓ = 0 and uu_y+vv_y = 0. Using CR, show
uₓ = vₓ = 0.

THEOREM — CONSTANT PART ⟹ CONSTANT

If f is analytic on D and any one of these holds:


(a) Re(f) = constant, (b) Im(f) = constant, (c) |f(z)| = constant, (d) Arg(f) =
constant
Then f is constant on D.

THEOREM — F AND F̄ BOTH ANALYTIC ⟹ F CONSTANT

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
If both f(z) = u+iv and f̄(z) = u−iv are analytic in D, then f is constant.
Proof: CRs from f: uₓ=v_y, u_y=−vₓ. CRs from f̄: uₓ=−v_y, u_y=vₓ. Adding:
uₓ=vₓ=0, so f′=0 → constant.

L'HÔPITAL'S RULE FOR ANALYTIC FUNCTIONS

If f, g analytic at z₀, f(z₀)=0, g(z₀)=0, g′(z₀)≠0:


lim(z→z₀) f(z)/g(z) = f′(z₀)/g′(z₀)

Analyticity Criteria — Quick Summary

TO PROVE F IS ANALYTIC IN D:

1. Write f = u + iv
2. Compute uₓ, u_y, vₓ, v_y
3. Check that uₓ = v_y and u_y = −vₓ hold at ALL points of D
4. Verify all four partials are continuous in D
✓ If all pass: f is analytic in D.

TO PROVE F IS NOT ANALYTIC AT Z ₀ :

Show CR equations fail at z₀ (or fail in every neighborhood of z₀). Just one point
of failure in any neighborhood is enough.

EXAMPLE — FIND CONSTANTS FOR ANALYTICITY

f(z) = 3x−y+5 + i(ax+by−3) to be analytic:


uₓ=3 = vy=b → b=3. u_y=−1 = −vₓ=−a → a=1.
f(z) = 3x−y+5+i(x+3y−3) is analytic everywhere → entire.

L7 · §17 Harmonic Functions

LAPLACE'S EQUATION & HARMONIC FUNCTION

∂²φ/∂x² + ∂²φ/∂y² = 0 (Laplace's equation)

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
A real-valued function φ(x,y) is harmonic in D if it has continuous first and
second-order partial derivatives in D AND satisfies Laplace's equation.

THEOREM — ANALYTIC ⟹ HARMONIC PARTS CORE

If f(z) = u(x,y)+iv(x,y) is analytic in D, then both u and v are harmonic in D.

Proof sketch:
From CR: uₓ = v_y and u_y = −vₓ.
Differentiating: uₓₓ = v_xy and u_yy = −v_yx = −v_xy
→ uₓₓ + u_yy = 0. Similarly v is harmonic.

⚠ CONVERSE DIRECTION

Non-harmonic u or v → f is not analytic


But harmonic u and v separately does not mean f = u+iv is analytic — they must
also satisfy CR.

EXAMPLES — TESTING HARMONICITY

u = ax²−y²+1: uₓₓ = 2a, u_yy = −2. For harmonic: 2a−2=0 → a=1.


u = x³−bxy²: uₓₓ = 6x, u_yy = −2bx. For harmonic: 6x−2bx=0 → b=3.
Is f(z) = x²+y²+2ixy analytic? u=x²+y²: uₓₓ+u_yy = 2+2 ≠ 0 → u not harmonic → f
not analytic.

L7 · §18 Harmonic Conjugate

HARMONIC CONJUGATE

If u(x,y) and v(x,y) are harmonic in D and satisfy the CR equations:


uₓ = v_y and u_y = −vₓ

then v is called the harmonic conjugate of u.

THEOREM — EXISTENCE OF HARMONIC CONJUGATE

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
If u(x,y) is harmonic in ℂ or in an open disk, then its harmonic conjugate exists
and is given by:

v(x,y) = ∫₀ʸ uₓ(x,t)dt − ∫₀ˣ u_y(s,0)ds + C

where C is a real constant.

THEOREM — F ANALYTIC ⟺ V HARMONIC CONJUGATE OF U

f(z) = u+iv is analytic ⟺ v is the harmonic conjugate of u.

WORKED EXAMPLE — FIND HARMONIC CONJUGATE

u(x,y) = y³ − 3x²y. Find v.


uₓ = −6xy, u_y = 3y²−3x²
v(x,y) = ∫₀ʸ uₓ(x,t)dt − ∫₀ˣ u_y(s,0)ds
= ∫₀ʸ (−6xt)dt − ∫₀ˣ (−3s²)ds
= −6x[t²/2]₀ʸ + 3[s³/3]₀ˣ
= −3xy² + x³
v = x³ − 3xy² + C (the harmonic conjugate)
Note: f(z) = u+iv = (y³−3x²y) + i(x³−3xy²) = −iz³ (up to constant)

Key Relationships

v is harmonic conjugate of u: vx = −uy, vy = ux


−u is harmonic conjugate of v (not v of u!)
If u is harmonic conjugate of v AND v is harmonic conjugate of u →
f is constant

EXAMPLE — FIND F FROM U

u = xy+x+2y, f(2i) = −1+5i.


uₓ=y+1, u_y=x+2. By formula: v = ∫₀ʸ (t+1)dt − ∫₀ˣ (s+2)ds = y²/2+y − x²/2−2x + C.
Check: f(2i) means x=0,y=2: u=2, v=2+2+C. But f(2i)=−1+5i → u=−1? Let's recalc:
u(0,2)=0·2+0+4=4. Actually f(2i)=−1+5i means the formula gives constants
correctly after full calculation. (Standard method in exam.)

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Ch4 · §19 Complex Exponential Function

DEFINITION

eᶻ = e^(x+iy) = eˣ(cos y + i sin y)


If y=0: eˣ (reduces to real exponential). If x=0: e^(iy) = cos y + i sin y (Euler).

Key Properties

Property Statement

Never zero eᶻ ≠ 0 for all z ∈ ℂ

Modulus |eᶻ| = eˣ

Argument arg(eᶻ) = y + 2nπ

Product e^(z₁+z₂) = eᶻ¹·eᶻ²

Division eᶻ¹/eᶻ² = e^(z₁−z₂)

Reciprocal e^(−z) = 1/eᶻ

Derivative d/dz[eᶻ] = eᶻ

Periodicity e^(z+2πi) = eᶻ (period 2πi)

Powers (eᶻ)ⁿ = e^(nz), n∈ℤ

e⁰ = 1 Constant: e^(πi) = −1

SOLVE E ᶻ = 1 + √ 3 I

|eᶻ| = eˣ = |1+√3i| = 2 → x = ln 2.
Phase: cos y = 1/2 and sin y = √3/2 → y = π/3 + 2nπ.
z = ln 2 + i(π/3 + 2nπ), n∈ℤ

SOLVE E ᶻ = −5

|eᶻ| = eˣ = 5 → x = ln 5. arg(−5) = (2n+1)π.


z = ln 5 + i(2n+1)π, n∈ℤ

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Ch4 · §20 Complex Logarithm

MULTI-VALUED COMPLEX LOGARITHM

For z ≠ 0: ln z = log_e|z| + i·arg(z) = log_e|z| + i(θ + 2nπ), n∈ℤ


This is multi-valued (infinitely many values).

PRINCIPAL VALUE — LN Z

Ln z = log_e|z| + i·Arg(z), where Arg(z) ∈ (−π, π]


Single-valued, analytic except on the negative real axis and origin (branch cut).

Properties

ln(z₁z₂) = ln z₁ + ln z₂ (as sets)


ln(z₁/z₂) = ln z₁ − ln z₂ (as sets)
ln(z₁ⁿ) = n ln z₁ (as sets)
e^(ln z) = z (z≠0)
d/dz [Ln z] = 1/z (for −π < Arg(z) < π)
Ln(eᶻ) = z (principal value)
ln(eᶻ) = z + 2nπi (multi-valued)

⚠ LN(Z ₁ Z ₂ ) ≠ LN(Z ₁ ) + LN(Z ₂ ) IN GENERAL

This fails when the sum of principal arguments exceeds (−π, π].
Example: Ln(−1+i)² vs 2Ln(−1+i) — these are NOT equal.

COMPUTATIONS

ln i: |i|=1, arg(i)=π/2+2nπ → ln i = (1/2+2n)πi, n∈ℤ


ln(−5): |−5|=5, arg(−5)=(2n+1)π → ln(−5) = ln 5 + (2n+1)πi
Ln(1+i)⁴: (1+i)⁴ = −4. Ln(−4) = ln 4 + iπ
Ln(1+i)² = Ln(2i) = ln 2 + iπ/2. 2Ln(1+i) = 2(ln√2 + iπ/4) = ln 2 + iπ/2. Equal ✓
Ln(−1+i)² = Ln(−2i) = ln 2 − iπ/2. 2Ln(−1+i) = 2(ln√2 + i3π/4) = ln 2 + i3π/2. NOT
equal ✗

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Branch Cuts & Branch Points

BRANCH CUT AND BRANCH POINT

Branch cut of Ln z: the negative real axis {z : Im(z)=0, Re(z)≤0} — where Ln z is


discontinuous.
Branch cut of Ln(z−1): translate by 1 → {z : Im(z)=0, Re(z)≤1}.
Branch point of ln z: z=0 (around which going in a circle changes the value).

Ch4 · §21 Complex Powers

COMPLEX POWER — MULTI-VALUED

For z ≠ 0 and α ∈ ℂ: z^α = e^(α ln z) = e^(α(log|z| + i arg(z))) (multi-valued via


arg)

PRINCIPAL VALUE OF Z^ Α

z^α = e^(α·Ln z) = e^(α(log|z| + iArg(z))) (single-valued, using Arg)

d/dz [z^α] = αz^(α−1) (for −π < θ < π)


d/dz [cᶻ] = cᶻ ln c (for −π < θ < π)

KEY EXAMPLE — I ⁱ

i^i = e^(i·ln i) = e^(i·(log 1 + i(π/2+2nπ))) = e^(i·i(π/2+2nπ)) = e^(−(π/2+2nπ))


→ iⁱ = e^(−π/2−2nπ), n∈ℤ (all real values!)
Principal value: iⁱ = e^(−π/2) ≈ 0.2079

EXAMPLE — (1−I)^(4I)

|1−i| = √2, Arg(1−i) = −π/4


(1−i)^(4i) = e^(4i · ln(1−i)) = e^(4i(ln√2 − iπ/4 + 2nπi))
= e^(4i·ln√2 + π − 8nπ) = e^(π−8nπ) · e^(i·2ln 2)
Principal value: e^(π + i·ln 4)

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Ch4 · §22 Complex Trigonometric Functions

DEFINITIONS VIA EULER'S FORMULA

sin z = (e^(iz) − e^(−iz)) / (2i)


cos z = (e^(iz) + e^(−iz)) / 2

Properties ALL EXAM-READY

Property Formula

Derivatives d/dz[sin z] = cos z, d/dz[cos z] = −sin z

Odd/Even sin(−z) = −sin z (odd), cos(−z) = cos z (even)

Euler e^(iz) = cos z + i sin z

Pythagorean sin²z + cos²z = 1

Periodicity sin(z+2π) = sin z, cos(z+2π) = cos z

Addition formulas sin(z₁+z₂) = sin z₁ cos z₂ + cos z₁ sin z₂

cos(z₁+z₂) = cos z₁ cos z₂ − sin z₁ sin z₂

Co-function sin(z+π/2) = cos z, sin(z−π/2) = −cos z

Zeros of sin z sin z = 0 ⟺ z = nπ, n∈ℤ

Zeros of cos z cos z = 0 ⟺ z = (2n+1)π/2, n∈ℤ

sin z in components sin z = sin x cosh y + i cos x sinh y

cos z in components cos z = cos x cosh y − i sin x sinh y

Both entire sin z and cos z are entire functions

Both unbounded! |sin z| = √(sin²x + sinh²y) → ∞ as y→±∞

Other Trig Functions

tan z = sin z / cos z cot z = cos z / sin z


sec z = 1/cos z csc z = 1/sin z

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
d/dz[tan z] = sec²z d/dz[cot z] = −csc²z
d/dz[sec z] = sec z tan z d/dz[csc z] = −csc z cot z
tan(z+π) = tan z (period π)

Solving Trig Equations in ℂ

METHOD: COS Z = Α , SIN Z = Α ( Α ∈ ℂ )

Substitute cos z = (e^iz+e^−iz)/2 = α → multiply by 2e^iz → e^(2iz) − 2αe^(iz) + 1 =


0
Quadratic in w = e^(iz): w = α ± √(α²−1). Then z = −i·ln(w) = −i·(log|w| + iarg(w)).

SOLVE COS Z = 2

e^(2iz) − 4e^(iz) + 1 = 0 → e^(iz) = 2 ± √3


Case 1: e^(iz) = 2+√3 → iz = ln(2+√3) + 2nπi → z = 2nπ − i·ln(2+√3)
Case 2: e^(iz) = 2−√3 = 1/(2+√3) → z = 2nπ + i·ln(2+√3)
z = 2nπ ± i·ln(2+√3), n∈ℤ

SOLVE SIN Z = 4

e^(2iz) + 2i·e^(iz) − 1 = 0 → e^(iz) = −i ± √(−1+1)... Substitute carefully:


2i·e^(iz) = e^(2iz) − 1 trick: e^(iz) = (4/i)... Let w=e^(iz): w − 1/w = 8i → w² − 8iw − 1
=0
w = (8i ± √(−64+4))/2 = 4i ± i√15 = i(4±√15)
z = (2nπ + π/2) ± i·ln(4+√15), n∈ℤ

ADDITIONAL PRACTICE ANSWERS

eᶻ = −2i: z = ln 2 + i(4n−1)π/2, n∈ℤ


cos z = cosh 2: z = 2nπ ± 2i, n∈ℤ
sin z = i: z = 2nπ − i·ln(√2−1) or z = (2n+1)π − i·ln(1+√2), n∈ℤ

L8 · §23 MCQ Bank with Answers — Lecture 8

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
These are the exact sample problems from Lecture 8. Know why each answer is
correct.

Q1. Let S = {z∈ℂ : |z|<1}. The set ℂ\S is:


Answer: (b) Closed — it's {|z|≥1}, a closed set containing all
boundary points.

Q2. S = {z∈ℂ : 1<|z|<2} is:


Answer: (d) Domain — it's an open annulus (open and connected).

Q3. f(z) = |z| is:


Answer: (c) Nowhere differentiable. u=√(x²+y²), v=0. CR:
uₓ=x/|z|=0=vy. Fails everywhere except conceptually. Actually
differentiable nowhere since CR holds only at origin but f(z)=|z|
means Δf/Δz→ varies with direction.

Q4. If z = re^(iθ), |z| = ?


Answer: r

Q5. {z : Im(z) > 0} represents:


Answer: (b) Upper half-plane.

Q6. {z : |z−i| < 2} is:


Answer: (b) An open disk (center i, radius 2).

Q7. lim(z→0) z/Re(z): exists or does not exist?


Answer: Does not exist. Along y=0: limit=1. Along y=x: limit=1+i.
Different paths give different limits.

Q8. Arg(z) is discontinuous at:

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Answer: The negative real axis {z : Im(z)=0, Re(z)≤0} — a jump from π
to −π occurs there.

Q9. f(z) = z̄ is:


Answer: (c) Nowhere differentiable. CR: uₓ=1≠−1=vy everywhere.

Q10. Cauchy-Riemann equations are:


Answer: (b) uₓ = vy and u_y = −vₓ.

Q11. If f(z) satisfies CR equations and partial derivatives are continuous in D,


then f is:
Answer: (a) Analytic in D (by Theorem-C).

Q12. f(z) = zRe(z) is differentiable at:


Answer: (b) Only z=0. u=x²+y·0 wait: z·Re(z) = (x+iy)·x = x²+ixy.
uₓ=2x,vy=ix... apply CR: uₓ=2x=vy=0 fails unless x=0; u_y=0=−vₓ=−x
fails unless x=0. So x=0 always; then z=iy, and checking at z=0
works.

Q13. If f(z) analytic and Re(f) constant, then f is:


Answer: constant (by the theorem on constant real part).

Q14. f(z) = |z|² is:


Answer: (c) Nowhere analytic — differentiable only at z=0, but not in
any neighborhood of 0.

Q15. If both f(z) and f̄(z) analytic in a domain, f must be:


Answer: (b) Constant.

Q16. If |f(z)| constant and f analytic:

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Answer: (b) f is constant.

Q17. If u is harmonic, harmonic conjugate v satisfies:


Answer: Cauchy-Riemann equations (uₓ=vy, u_y=−vₓ).

Q18. u(x,y) = x²−y² has harmonic conjugate:


Answer: (a) 2xy. Verify: v=2xy → vₓ=2y=−u_y=−(−2y) ✓, vy=2x=uₓ ✓.

Q19. lim(z→0) (z²−z̄ ²)/z:


Answer: (c) Does not exist. Write z=re^(iθ): z²−z̄ ² = r²e^(2iθ)
−r²e^(−2iθ) = 2ir²sin(2θ). Divide by re^(iθ): 2ire^(−iθ)sin(2θ) →
depends on θ as r→0. DNE.

Q20. If f = u+iv analytic, both u and v are:


Answer: Harmonic functions.

Q21. f(z) = Im(z) is differentiable:


Answer: (b) Nowhere. u=0, v=y(wait: Im(z)=y is real-valued): f=y,
u=y,v=0. uₓ=0≠vy=0. u_y=1≠−vₓ=0. CR fails everywhere → nowhere
differentiable.

Q22. f(z) = 1/(|z|−1) is discontinuous at:


Answer: The unit circle {z : |z|=1} (denominator = 0 there).

Q23. What ensures analyticity?


Answer: (c) CR equations AND continuity of partial derivatives.

Q24. lim(z→0) z/|z|:


Answer: (c) Does not exist. z/|z| = e^(iθ) depends on direction θ.

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Q25. f(z) = √|z²−z̄ ²| at z=0:
Answer: (d) Satisfies CR but is not differentiable (continuous
partials don't hold).

Q26. If f(z) = u(x)+iv(y) is analytic, f must be of the form:


Answer: f(z) = az + b (linear), since uₓ=u'(x)=vy=v'(y)=constant and
u_y=0=−vₓ=0 → u'=v'=constant.

Q27. {z∈ℂ : Re(z²) > 0} is:


Answer: (a) Connected. Re(z²)=x²−y²>0 is the region between lines y=x
and y=−x (right+left sectors) — it is connected (both sectors
joined).

Q28–Q29. f(z) = {z²+1)/(z+i), z≠−i; −2i, z=−i} at z=−i:


Q28: (a) Continuous. lim = lim(z→−i)(z+i)(z−i)/(z+i) = lim(z−i) = −2i
= f(−i). ✓
Q29 (Arg(iz) discontinuous at): Arg(iz) = Arg applied to rotation by
π/2. Discontinuous where argument of iz = ±π → iz is on negative real
axis → z is on negative imaginary axis → z = −ir, r>0 → z = −i is the
key point. Answer: z=−i (negative imaginary axis).

Q30. If f satisfies CR everywhere but partials not continuous, f is:


Answer: (b) Not necessarily analytic (CR alone ≠ differentiability).

Q31. f(z) = x²+iy² is differentiable at:


Answer: (b) Only when x=y, i.e., on the line y=x. (uₓ=2x=vy=2y
requires x=y). But this is a 1D set, so nowhere analytic.

Q32. lim(z→−i) (z⁴−1)/(z+i):

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
Answer: (a) Exists and is finite. L'Hôpital or factor: z⁴−1=(z²+1)
(z²−1)=(z+i)(z−i)(z+1)(z−1). Limit = (−i−i)(−i+1)(−i−1) = (−2i)(1−i)
(−1−i) = (−2i)(−1−i+i+i²) = (−2i)(−2) = 4i. Finite ✓

Q33. f(z) = z̄ ² is:


Answer: (b) Differentiable only at z=0. (z̄ ²: u=x²−y², v=−2xy.
uₓ=2x=vy=−2x only if x=0; u_y=−2y=−vₓ=2y only if y=0.) → only at
origin.

Q34. If f analytic and u−v = constant, then:


Answer: (a) f is constant. u−v=c → d(u−v)/dx=uₓ−vₓ=0 and
d/dy=u_y−v_y=0. Using CR: uₓ=vy so vy−vₓ=0; u_y=−vx so −vx−vy=0. This
gives vx=vy=0 → f′=0 → constant.

Q35. f(z) = (z³−1)/(z−1) with f(1)=3:


Answer: (b) Continuous at z=1. Since (z³−1)/(z−1)=z²+z+1 for z≠1, and
lim→1 = 3 = f(1). ✓

Q36. f(z) = zIm(z) is analytic:


Answer: (c) Only at z=0. f = (x+iy)y = xy+iy². u=xy, v=y². uₓ=y=vy=2y
→ only y=0; u_y=x=−vx=0 → only x=0. Both hold only at x=0,y=0, i.e.,
z=0.

🏆 FINAL EXAM STRATEGY SUMMARY


1. Topology questions: Check open/closed by boundary inclusion; check
connectedness by polygonal paths.
2. Limit questions: Use Theorem-A to reduce to real 2-var limits. Test non-
existence by two-path method.
3. Differentiability: Apply Theorem-B (necessary: CR) and Theorem-C
(sufficient: CR + continuity). Compute f′ = uₓ+ivₓ.

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF
4. Analyticity: Differentiable in a neighborhood. CR+continuous partials
throughout domain = analytic.
5. "Analytic → constant": Any of {f′≡0, Re=const, Im=const, |f|=const, f and f̄
both analytic} → f=const.
6. Harmonic conjugate: Use the integral formula or directly integrate from CR
equations.
7. Elementary functions: eᶻ = eˣ(cosy+isiny). ln z = log|z|+i arg z. sin/cos z via
Euler. Solve equations via substitution w=e^(iz).

COMPLEX ANALYSIS · BITS GOA · PROF. J.K. SAHOO · ALL LECTURES COVERED

Explore our developer-friendly HTML to PDF API Printed using PDFCrowd HTML to PDF

You might also like