Introduction to Complex Analysis

George Voutsadakis1

1Mathematics and Computer Science

Lake Superior State University

LSSU Math 413

George Voutsadakis (LSSU) Complex Analysis October 2014 1 / 83

 Outline

1 Integration in the Complex Plane

Real Integrals
Complex Integrals Cauchy-
Goursat Theorem
Independence of Path
Cauchy’s Integral Formulas
Consequences of the Integral
Formulas

George Voutsadakis (LSSU) Complex Analysis October 2014 2 / 83

 Integration in the Complex Plane Real Integrals

Subsection 1

Real Integrals

George Voutsadakis (LSSU) Complex Analysis October 2014 3 / 83

 Integration in the Complex Plane Real Integrals

Definite Integrals

If F (x ) is an antiderivative of a continuous function f , i.e., F is a

function for which F ′(x ) = f (x ), then the definite integral of f on
the interval [a, b] is the number
ˆb
f (x )dx = F (xba)| = F (b) − F
a (a).
´2
2
Example: x 2 dx = 1 x3 = 8− 1− =
3. − 1 3 − 1 3 3

T´b he fundamental theorem of calculus´b is a method of

a f (x )dx ; it is not the definition of f (x )dx
evaluating
a. next define:
We
The definite (or Riemann) integral of a function f ;
Line integrals in the Cartesian plane.
Both definitions rest on the limit concept.

George Voutsadakis (LSSU) Complex Analysis October 2014 4 / 83

 Integration in the Complex Plane Real Integrals

Steps Leading to the Definition of the Definite

Integral
1. Let f be a function of a single variable x defined at all points in
a closed interval [a, b].
2. Let P be a partition:
a = x0 < x1 < x2 < · · · < xn− 1 < xn = b
of [a, b] into n subintervals [x k−1 , xk ] of length ∆x k = xk − x k− 1 .
3. Let kP k be the norm of the partition P of [a, b], i.e., the
length of the longest subinterval.
4. Choose a number xk∗ in each subinterval [x k−1 , xk ] of [a, b].

5. Form n products f (xk∗)∆xk , k = 1, 2, . . . , n, and then sum

products
these Xn
: f (xk∗ )∆ xk .
k=1
George Voutsadakis (LSSU) Complex Analysis October 2014 5 / 83
 Integration in the Complex Plane Real Integrals

The Definition of the Definite

Integral

Definition (Definite Integral)

The definit e integral of f on [a, b]
is
ˆb
Xn
f (x )dx = f
a lim
k = 1 (xk )∆ xk .
kPk→ 0 ∗

Whenever the limit exists we say that f is integrable on the

interval [a, b] or that the definite integral of f exist s.
It can be proved that if f is continuous on [a, b], then the
integral exists.

George Voutsadakis (LSSU) Complex Analysis October 2014 6 / 83

 Integration in the Complex Plane Real Integrals

Terminology About Curves

Suppose a curve C in the plane is parametrized by a set of
equations x = x (t ), y = y (t ), a ≤ t ≤ b, where x (t ) and y (t )
are continuous real functions. Let the initial and terminal points of
C (x (a), y (a)), (x (b), y (b)) be denoted by A, B . We say
that:
(i) C is a smooth curve if x ′ and y ′ are continuous on the closed interval
[a, b] and not simultaneously zero on the open interval (a, b).
(ii) C is a piecewise smooth curve if it consists of a finite number of
smooth curves C1, C2, . . . , Cn joined end to end, i.e., the terminal
point of one curve Ck coinciding with the initial point of the next
curve Ck+1.
(iii) C is a simple curve if the curve C does not cross itself except possibly
at t = a and t = b.
(iv) C is a closed curve if A = B.
(v) C is a simple closed curve if the curve C does not cross itself and
A = B, i.e., C is simple and closed.

George Voutsadakis (LSSU) Complex Analysis October 2014 7 / 83

 Integration in the Complex Plane Real Integrals

Steps Leading to the Definition of Line

Integrals
1. Let G be a function of two real variables x and y , defined at all
points on a smooth curve C that lies in some region of the xy
-plane. Let C be defined by the parametrization x = x (t), y = y
(t), a ≤ t ≤ b.
2. Let P be a partition of the parameter interval [a, b] into n
subintervals [t k − 1 , t k ] of length ∆ t k = t k − t k − 1 :
a = t 0 < t 1 < t 2 < · · · < t n− 1 < t n = b.
The partition P induces a partition of
the curve C into n subarcs of
length
∆ sk . Let the projection of each
sub- arc onto the x - and y -axes
have lengths
∆ xk and ∆ yk , respectively.
George Voutsadakis (LSSU) Complex Analysis October 2014 8 / 83
 Integration in the Complex Plane Real Integrals

Steps Leading to the Definition of Line Integrals

(Cont’d)

3. Let kP k be the norm of the partition P of [a, b], that is, the
length
of the longest subinterval.
4. Choose a point (xk∗, yk∗) on each subarc of C .
5. Form n products G (xk∗, yk∗)∆xk , G (xk∗, yk∗)∆yk , G (xk∗,
yk∗)∆sk ,
Xn Xn
k = 1, 2,G. .(x
. ,k∗n, G (x
, and then sum these )∆ yk ,
k , ykproducts
∗ ∗
G (xk∗ ,
n y ∗ )∆ x ,
k=1
X k=1 y ∗ )∆ s . k=1
k k k k
George Voutsadakis (LSSU) Complex Analysis October 2014 9 / 83
 Integration in the Complex Plane Real Integrals

The Definition of Line Integrals

Definition (Line Integrals in the Plane)
(i) The line integral of G along C with respect to x
is ˆ
G (x , y )dx = limXn G (xk∗ ,
C yk )∆ xk .
∗
kPk→ 0
k=1
(ii) The line integral of G along C with respect to y
is ˆ
G (x , y )dy = limXn G (xk∗ ,
C yk )∆ yk .
∗
kPk→ 0
k=1
(iii) The line integral of G along C with respect to arc length s
is ˆ
G (x , y )ds = limXn G (xk∗ ,
C yk )∆ sk .
∗
kPk→ 0
k=1
If G is continuous on C , then the three types of line integrals
exist. The curve C is referred to as the path of integration.
George Voutsadakis (LSSU) Complex Analysis October 2014 10 / 83
 Integration in the Complex Plane Real Integrals

Method of Evaluation: C Defined Parametrically

Convert a line integral to a definite integral in a single
variable. If C is a smooth curve parametrized by x = x (t), y
= y (t),
a ≤ t ≤ b, then replace
x and y in the integral by the functions x (t) and y
(t); the appropriate differential dx , dy , or ds by
p
x ′ (t )dt , y ′ (t )dt ,
arc length. [x ′ (t )]2 +
[y ′ (t )]2 dt .
The line integrals p become definite integrals in which the variable
of
Theintegration
term ds =is the [x parameter
′ (t )]2 + [yt: ′ (t )]2 dt is called the

´
differential of t he ´b
G (x , y ) = G (x (t ), y (t ))x ′
´ dx
C (t )dt
= ´ b G (x (t ), y (tp))y ′
, a

= )dt´,a b G (x (t ), y (t )) [x ′ (t )]2 + [y ′
G (x , y ) (t a
´C
dy (t )]2 dt .
C
George VoutsadakisG (x , y )
(LSSU) Complex Analysis October 2014 11 / 83
 Integration in the Complex Plane Real Integrals

Evaluation of a Line Integral

I ´
Evaluate C xy 2 dx , where the path of
inte- C is the quarter circle defined
gration
by= 4 cos t , y = 4 sin t , 0 ≤ t ≤
x 2 π

We
have dx = − 4 sin
tdt .
Thus
, ´ ´ π/2
C xy 2dx = 0 (4 cos t )(4 sin 2 −4 sin
t )´ (π/2 3 tdt )
= 0 −sin t cos
tdt π/2
256 1
4
4
0
= — −
256[ 64.
sin t ]
George Voutsadakis (LSSU) = Complex Analysis October 2014 12 / 83
 Integration in the Complex Plane Real Integrals

Evaluation of a Line Integral

II
´
Evaluate C xy 2 dy , where the path of integration C is the
quarter
circle defined by x = 4 cos t , y = 4 sin t , 0 ≤ tπ2
≤
We
.have dy = 4 cos
tdt .
Thus
, ´ ´ π/2
C xy 2 dy = (4 cos t )(4 sin t )2 (4 cos
tdt ) 0 ´ 2
0π/2 sin t cos tdt
2
= ´
256 π/2 1
2 sin 2tdt
= 256 0 14
´ π/2
0 (1 − cos
= 64 2 1 4t )dt π/2
32[t − 4 sin 4t0] =
= 16π.

George Voutsadakis (LSSU) Complex Analysis October 2014 13 / 83

 Integration in the Complex Plane Real Integrals

Evaluation of a Line Integral

III
´
Evaluate C xy 2ds, where the path of integration C is the
quarterdefined by x = 4 cos t , y = 4 sin t , 0 ≤ tπ
circle 2
≤
We q
.have
ds = 16(sin2 t + cos2 t )dt =
4dt .
Therefor
e, ´ ´ π/2
C xy = (4 cos t )(4 sin
2 ds t )2 (4dt0 )´ 2
0π/2 sin t cos
= 256
256[31 sintdt 3
0
= π/2
]3 .
t256
=

George Voutsadakis (LSSU) Complex Analysis October 2014 14 / 83

 Integration in the Complex Plane Real Integrals

Method of Evaluation: C Defined by a

Function
If the path of integration C is the graph of an explicit function
y = f (x ), a ≤ x ≤ b, then we can use x as a parameter:
The diffperential of y is dy = f ′(x )dx , and the differential of arc
is ds = 1 + [f ′ (x )]2 dx .
length
We, thus, obtain the definite integrals:
´ ´b
G (x , y ) = G (x , f
´C dx (x
= ))dx
´ab ,
G (x , f (xp))f ′ (x )
= , ´a b G (x , f (x )) 1 + [f ′
G (x , y ) dx a
´C
dy (x )]2 dxsmooth
.
A line integral
C along a piecewise curve C is defined as the
G (x , y )
sum of the integrals
ds ov´er the various smooth pieces.
Example: To evaluate C G (x , y )ds when C is composed of
smooth curvˆes C1 and C2 , wˆe ˆ
two
write G (x , y )ds = G (x , y )ds + G (x , y )
C ds.
George Voutsadakis (LSSU)
C
1
Complex Analysis October 2014 15 / 83
 Integration in the Complex Plane Real Integrals

Notation for Line Integrals

In many applications, line integrals appear as a
sum ˆ ˆ
P (x , y )dx + Q(x , y )
C dy .
It is common practice to write this sum
C as one integral
without
parentheses as ˆ
P (x , y )dx + Q(x ,
y )dy
or C ˆ
simply Pdx + Qdy .
C

A line integral along a

closed curve C is usually
denoted by
˛
Pdx + Qdy .
George Voutsadakis (LSSU) Complex Analysis C October 2014 16 / 83
 Integration in the Complex Plane Real Integrals

C Defined by an Explicit
Function

´
Evaluate C xydx + x 2 dy , where C is the
graph
of y = x 3 , −1 ≤ x ≤ 2.
We have dy = 3x 2 dx .
Therefore,
´ ´2
C xydx + x 2 dy = xx 3 dx + x 2 3x 2
dx ´−21 4
4
= ´− 1 (x + 3x )
dx
2
− 1 4x
= 2
44dx
x 5
= 4 (32 − (−1)) = 132 .
5
5− 1 5
=

George Voutsadakis (LSSU) Complex Analysis October 2014 17 / 83

 Integration in the Complex Plane Real Integrals

C a Closed Curve

¸
Evaluate C xdx , where C is the circle defined by x = cos t , y = sin
t,
0 ≤ t ≤ 2π.
We have dx = −¸sin tdt , whence:
2π
´C xdx = 0 cos t (− sin
= 2
tdt )12π
0
1
(1 − 1)
cos t2
2

=
= 0.

George Voutsadakis (LSSU) Complex Analysis October 2014 18 / 83

 Integration in the Complex Plane Real Integrals

C Another Closed Curve

¸
Evaluate C
y 2 dx − x 2 dy , where C is the closed curve shown
on
leftthe
. C is piecewise smooth. So, the
given integral is expressed as ¸a
of int´egrals,´ i.e., we write C
sum
´
C1 + C2 + C3 , with C1 , C2 , C3
=
as
shown on the right.
´ 2
1 y 22 dx − x dy = — x2 (0) =
C2, with y as a parameter: ´ 0dx ´
C1
On 0.
´
On yC2 dx
, with 2 as a parameter:
x 4 2 0 4
´C2 − x dy = 0 y (0) − 4dy = − 0 4dy = − 16.
On C3, we again use ´ x as a parameter.´0From y = x , we
2
2 2 2 2
get = 2xdx . Thus, y dx − x dy = 2 (x ) dx − x (2xdx )
dy 2
=´0 4 3
8
C3 1 5 1
4 0
(x − 2x )dx = ( 5 x − x2 ) 2 = 5.
Hence, ¸C y272dx − x2 dy =´ C1 + ´C2 + ´C3 = 0 + (−16) + 85 = − 5 .
2

George Voutsadakis (LSSU) Complex Analysis October 2014 19 / 83

 Integration in the Complex Plane Real Integrals

Orientation of a Curve
If C is not a closed curve, then we say the positive direction on
C , or that C has positive orientation, if we traverse C from its
initial point A to its terminal point B , i.e., if x = x (t), y = y (t),
a ≤ t ≤ b, are parametric equations for C , then the positive
direction on C corresponds to increasing values of the parameter t.
If C is traversed in the sense opposite to that of the positive
orientation, then C is said to have negative orientation.
If C has an orientation (positive or negative), then the
opposite
curve, the curve with the opposite orientation, will be denoted
ˆ
−C .
The
ˆ Pdx + Qdy = − Pdx + Qdy
n C
,
or, equivalently ˆ ˆ
− C Pdx + Qdy + Pdx + Qdy =
− C 0.
A line integral is independent of theCparametrization of C , provided
C
is Voutsadakis
George given the same orientation.
(LSSU) Complex Analysis October 2014 20 / 83
 Integration in the Complex Plane Complex Integrals

Subsection 2

Complex Integrals

George Voutsadakis (LSSU) Complex Analysis October 2014 21 / 83

 Integration in the Complex Plane Complex Integrals

Curves Revisited
Suppose the continuous real-valued functions x = x (t), y = y
(t), a ≤ t ≤ b, are parametric equations of a curve C in the
complex plane.
By considering z = x + iy , we can describe the points z on C by
means of a complex-valued function of a real variable t, called a
paramet rizat ion of C : z (t ) = x (t ) + iy (t ), a ≤ t ≤ b.
Example: The parametric equations x = cos t , y = sin t , 0 ≤ t ≤
2π, describe a unit circle centered at the origin. A parametrization
of this circle is z (t) = cos t + i sin t, or z (t) = eit , 0 ≤ t ≤ 2π.
The point z (a) = x (a) + iy (a) or A
= (x (a), y (a)) is called the init ial
point of C . and z (b) = x (b) + iy (b) or
B = (x (b), y (b)) the terminal point.
As t varies from t = a to t = b, C is being
traced out by the moving arrowhead of the
vector corresponding to z (t ).
George Voutsadakis (LSSU) Complex Analysis October 2014 22 / 83
 Integration in the Complex Plane Complex Integrals

Smooth Curves and Contours

Suppose the derivative of z (t) = x (t) + iy (t), a ≤ t ≤ b, is
z ′ (t ) = x ′ (t ) + iy ′ (t ).
We say C is smooth if z ′ (t ) is continuous and never zero
in the interval a ≤ t ≤ b.
Since the vector z ′ (t ) is not zero at any point P
on C , the vector z ′ (t ) is tangent to C at P . In
other words, a smooth curve has a continuously
turning tangent.

A piecewise smooth curve C has a continuously turning tangent,

except possibly at the points where the component smooth curves
C1 , C2 , . . . , Cn are joined together.
A curve C in the complex plane is simple if z (t 1 ) 6= z (t 2 ), for t 1
=6
A piecewise smooth curve C is also called a contour or
2 , except possibly for t = a and t = b.
tpath.
George Voutsadakis (LSSU) Complex Analysis October 2014 23 / 83
 Integration in the Complex Plane Complex Integrals

Positive and Negative Directions

We define the positive direction on a contour C to be the direction
on the curve corresponding to increasing values of the parameter t.
It is also said that the curve C has positive orientation.
In the case of a simple closed curve C , the positive direction
roughly corresponds to the counterclockwise direction or the
direction that a person must walk on C in order to keep the interior
of C to the left.

The negative direction on a contour C is the direction opposite the

positive direction.
If C has an orientation, the opposite curve, that is, a curve
with opposite orientation, is denoted by −C .
On a simple closed curve, the negative direction corresponds to
the clockwise direction.
George Voutsadakis (LSSU) Complex Analysis October 2014 24 / 83
 Integration in the Complex Plane Complex Integrals

Steps Leading to the Definition of the Complex Integral

I
1. Let f be a function of a complex variable z defined at all points on
a smooth curve C that lies in some region of the plane. Suppose C
is defined by the parametrization z (t) = x (t) + iy (t), a ≤ t ≤ b.
2. Let P be a partition of the parameter interval [a, b] into n
subintervals [t k − 1 , t k ] of length ∆ t k = t k − t k − 1 :

a = t 0 < t 1 < t 2 < · · · < t n− 1 < t n = b.

The partition P induces a partition of the curve C into n
subarcs whose initial and terminal points are the pairs of
numbers z = x (t ) + iy z = x (t ) + iy
0 0 1 1
(t 0 ), (t 1 ),
..z1 = x (t 1 ) + iy .z. 2 = x (t 2 ) + iy
z(tn−1 ),1 = x (t n− 1 ) + iy (t n− 1 ), z(tn2 ),
= x (t n ) + iy
(t n ).
Let ∆ zk = zk − zk − 1 , k = 1, 2, . . . ,
n.Voutsadakis (LSSU)
George Complex Analysis October 2014 25 / 83
 Integration in the Complex Plane Complex Integrals

Steps Leading to the Definition of the Complex Integral

II
3. Let kP k be the norm of the partition P of [a, b], i.e., the
length of the longest subinterval.
4. Choose a point zk∗ = xk∗ + iyk∗ on each subarc of C .

5. Form n proPn ducts f (zk∗ )∆ zk , k = 1, 2, . . . , n, and then sum

products k=1 f (zk
these
: ∗ )∆ z .
k
George Voutsadakis (LSSU) Complex Analysis October 2014 26 / 83
 Integration in the Complex Plane Complex Integrals

The Definition of the Complex

Integral
Definition (Complex Integral)
The complex integral of f on C
is
ˆ f (z )dz = Xn
f
C lim
k = 1 (zk )∆ zk .
kPk→ 0 ∗

If the limit exists, f is said to be integrable on C .

The limit exists whenever f is continuous at all points on C and C is
either smooth or piecewise smooth.
Thus, we always assume that these cond¸itions are fulfilled.
By convention, we will use the notation C f (z )dz to represent
a complex integra l around affpositively oriented closed curve C .
The notations C f (z )dz, C f (z )dz denote more explicitly
integration in the positive and
George Voutsadakis (LSSU) Complexnegative
Analysis directions, respectively.
October 2014 27 / 83
 Integration in the Complex Plane Complex Integrals

Complex-Valued Function of a Real

Variable
Example: If t represents a real variable, then f (t) = (2t + i )2 is a
complex number. For t = 2, f (2) = (4 + i )2 = 16 + 8i + i 2 =
15 + 8i .
If f1 and f2 are real-valued functions of a real variable t, then
f (t) = f 1 (t) + if 2 (t) is a complex-valued function of a real variable
t.
Example: On the in
We are interested interval 0 ≤ tof≤ a1,complex-valued
integration it seems reasonable
function
for
f (t) = f 1 (t) + if22 (t) of a real variable t carried out over a real
f (t ) = (2t + i ) to write ˆ1
ˆ interval.
1 (2t + i )2 dt = ˆ 1(4t 2 − 1 + 4ti )dt = ˆ 1(4t 2 − 1)dt + i 4tdt.
0 0 0
0 ´1
The integrals ´01 (4t2 − 1)dt and 4tdt are real, and could be
called ´1 0
the real and imaginary parts of 0 (2t + i 2) dt. Each can be
evaluated
using the fundamental theor em of ca lculus to
0 3 0 0 3
get:
George Voutsadakis (LSSU) Complex Analysis October 2014 28 / 83
 Integration in the Complex Plane Complex Integrals

Integral of Complex Valued Function of a Real

Variable
If f1 and f2 are real-valued functions of a real variable t continuous on
a common interval a ≤ t ≤ b, then we define the integral of the
complex-valued function f (t) = f 1 (t) + if 2 (t) on a ≤ t ≤ b by
ˆ b ˆ b ˆ b
f (t)dt = f 1 (t)dt + i
f 2 (t)dt.
a a a

The continuity of f1 and f2 on [a, b] guarantees that both integrals

on the right exist.
functions of a real variable t continuous on a ≤ t ≤ b,
If ´ b= f 1 (t) + if 2´(t) and g (t) = g1 (t) + ig2(t), are complex-
f (t)
then a kf (t )dt = k a f (t )dt , k a complex
valued´
b b ´b ´
constant;
(f (t ) + g (t ))dt = a f (t )dt + a g
´ab ´ b ´b
f(t(t)dt
)dt; = a f (t )dt + c f (t )dt , if c ∈ [a,
´a b];
a c ´ b
b f (t )dt = − a f (t )dt .

George Voutsadakis (LSSU) Complex Analysis October 2014 29 / 83

 Integration in the Complex Plane Complex Integrals

Evaluation of Contour Integrals

P
If
P we use u + ´iv for f , ∆ x + i ∆ y forP∆ z , lim kPk→ 0 and
forn lim, we get C f (z )dz = lim (u + iv )(∆ x + i ∆ y ) for
P
limk=[P=(u∆ x − v ∆ y ) + i (v ∆ x +
1
Thus,
u∆ y )].we have
ˆ ˆ ˆ
f (z )dz = udx − vdy + i vdx +
udy .
C C C
If
dxx==xx′ (t(t), = y=(t),
)dty, dy y ′ a ≤ t ≤ b, are parametric equations of C ,
´b
then
(t )dt .we obtain a [u(x (t ), y (t ))x
Now ′ (t ) − v (x (t ), y (t ))y
′
´
i(t )]dt +
b
a [v (x (t ), y (t ))x (t ) + u(x (t ), y (t ))y
′ ′
´b
(t )]dtis. the same as a f (z (t ))z (t )dt when the
This ′

integrand
f (z (t))z ′ (t) = [u(x ´b(t), y (t))′ + iv (x (t), y (t))][x (t) + iy
′ ′

(t)] is
multiplied out and a f (z (t))z (t)dt is expressed in terms of its
real imaginary parts.
and
George Voutsadakis (LSSU) Complex Analysis October 2014 30 / 83
 Integration in the Complex Plane Complex Integrals

Evaluating of a Contour
Integral
Theorem (Evaluation of a Contour Integral)
If f is continuous on a smooth curve C given by z (t) = x (t) + iy
(t),
a ≤ t ≤ b, ˆ ˆb
then f (z )dz = f (z (t ))z ′
C
a´ (t )dt .
Example: Evaluate C zdz, where C is given by x = 3t, y =
t 2 , ≤ t ≤ 4.
−1
A parametrization of the cont our C is z (t) = 3t + it 2 . Thus,
since
f (z ) = z, we have f (z (t )) = 3t + it 2 = 3t − it 2. Also,
z ′ (t ) = ´3 + 2it . Now,
= we
4 have
(3t − it 2)(3 +
´C zdz − 12it )dt ´
= ´ 3 2
− 1 (2t + 9t)dt + i − 1 3t dt
4
43 4
= 1 4 ( 92 t2 + t ) −+1 i t = 195
2
+ 65i .4 −1
George Voutsadakis (LSSU) Complex Analysis October 2014 31 / 83
 Integration in the Complex Plane Complex Integrals

Another Evaluation of a Contour

Integral

¸
Evaluate C z 1 dz , where C is the circle x = cos t , y =
sin≤ tt, ≤ 2π.
0
In this case z (t) = cos t + i sin t = eit , z ′ (t) = ieit ,
fand
(z (t )) = z 1 = e− it .
Hence, (t )
¸
´ 2π − it
c z1 dz = (e )
´0
ie dt
it
2π
= i 0 dt
= 2πi .

George Voutsadakis (LSSU) Complex Analysis October 2014 32 / 83

 Integration in the Complex Plane Complex Integrals

Using x as a
Parameter
For some curves the real variable x itself can be used as
the parameter.
´
Example: Evaluate C (8x 2 − iy )dz on the line segment y =
5x≤, x ≤ 2.
0
We write z = x + 5xi , whence dz = (1 + 5i )dx .
Therefore,
´ 2
´C (8x − iy )
2 = (1 + 5i ) 0 (8x2 − 5ix )
dz dx (1 + 5i ) 8 x 32 − (1 + 5i )i5 2 2
3 0 2 0
= x214
3 + 290 i .
= 3
If x and y are related by means of a continuous real function
y = f (x ), then the corresponding curve C can be
parametrized by
z (x ) = x + if (x ).
George Voutsadakis (LSSU) Complex Analysis October 2014 33 / 83
 Integration in the Complex Plane Complex Integrals

Properties of Contour Integrals

Theorem (Properties of Contour Integrals)

Suppose the functions f and g are continuous in a domain D, and C is
a smooth curve lying entirely in D. Then:
(i) ´ kf (z )dz = k f (z )dz, k a complex
constant.
´C C´
(ii) ´C [f (z ) + g (z )]dz´= C f (z )dz + C g
´ ´
(z )dzC. f (z )dz =´ C1 f (z )dz + C2 f (z )dz , where C consists of the
(iii)
smoothcurves C1 and C2 joined end to end.
´
(iv) ´ − C f (z )dz = − C f (z )dz, where −C denotes the curve
havingopposite
the orientation of C .

The four parts of the theorem also hold if C is a piecewise

smooth
curve in D.
George Voutsadakis (LSSU) Complex Analysis October 2014 34 / 83
 Integration in the Complex Plane Complex Integrals

C a Piecewise Smooth Curve

´
Evaluate (x 2 + iy 2)dz , where C is the contour
C
shown: ´ ´
We write C (x 2 + iy 2 )dz = C (x 2 + iy 2 )dz
+́ (x 2 + iy 2 )dz 1

C2
Since
. the curve C1 is defined by y = x , we
use x as a parameter: z (x ) = x + ix , z ′ (x
) = 1 + i , f (z ) = x 2 + iy 2 , f (z (x )) = x 2
´ + ix 2 , ´1
whence, finally, C (x 2 + iy 2 )dz = (x 2 + ix 2 )(i + 1)dx
= ´ 1
2
(1 + i )2 0 1 x 2 dx0= (1+
3
i)
= 2i
.The curve3 C2 is defined by x = 1, 1 ≤ y ≤ 2. If we use y as
´parameter, then z (y ) = 1 + iy , z ′(y ) ´= i , f (z (y´)) = 1 + iy
a
2 2
(x 2 + iy )dz = 1 (1 + iy2 )idy = − 12 y dy + i dy = − 73 + i
´,C2and
2
2 ´ 2
.
Therefore (x 2 + iy2 )dz = i + − 7+ i )1 = − 7 + 5 i
C 2 3 3 3
3 ( .
George Voutsadakis (LSSU) Complex Analysis October 2014 35 / 83
 Integration in the Complex Plane Complex Integrals

A Bounding Theorem
We find an upper bound for the modulus ´ b o f a contour
Recall the length of a plane curve L = p [x ′ (t )]2 + [y ′ (t )]2 dt .
integral.
If a
z ′ (t ) = x ′ (t ) + iy ′ (t ), then |z ′ (t )|p= [x ′ (t )]2 + [y ′ (t )]2 ,
´b ′
whence
L = a |z (t)|dt.
Theorem (A Bounding Theorem)

If f is ´continuous on a smooth curve C and if |f (z )| ≤ M, for all z on C ,

then | C f (z )dz | ≤ ML, where P L is the length of PC .
By triangle inequality, | nk=1 f (z∗k )∆ z | ≤ k=1 n
|f (zk∗ )||∆ zk |
Pn pk
≤ M k = 1 |∆ zk |. Because |∆ zk | = (∆ xk )2 + (∆ yk )2 , we can
interpret |∆ zk | as the length of the chord joining the points zk
zand
k − 1 on C . Moreover, since the sum n of the ∗lengths of the
cannot be greater than L, we get ´P k=1 f (zk )∆zk | ≤ ML. Finally,
chords
|
the continuity of f guarantees that ´C f (z )dz exists. Thus,
kP k → 0, the last inequality yields | C f (z )dz | ≤
letting
ML.
George Voutsadakis (LSSU) Complex Analysis October 2014 36 / 83
 Integration in the Complex Plane Complex Integrals

A Bound for a Contour

Integral
´
Find an upper bound for the absolute value of C z
ez dz where C is
the circle |z | = 4. +1

First, the length L (circumference) of the circle of radius 4 is

8π.
Next, for all points z on the circle, we have t z z
ez |e | = |e
hat+ 1| ≥ |z | − 1 = 4 − 1 = 3.
|z ≤ |
Thus, z x x
. Inthe
addition, |e | = |e (cos y + i sin y )| = e . For points on
circle z+1 |z | − 1 3
| z | = 4 , the maximum that x = Re(z ) can be is 4,
e
whence
z e4
have

ˆ ez dz ≤ 8π 4
.
C z+1 e3

George Voutsadakis (LSSU) Complex Analysis October 2014 37 / 83

 Integration in the Complex Plane Complex Integrals

Single Contour: Many Parametrizations

There is no unique parametrization for a contour C

. Example: All of the following:

z (t ) = eit = cos t + i sin t ,

0 ≤ t ≤ 2π,
z (t ) = e2πit = cos 2πt + i sin 2πt ,

0 ≤ t ≤ 1,
z (t ) = eπit /2 = cos πt
+ i sin πt
,
0 ≤ t ≤ 4,
2 2

are all parametrizations, oriented in the positive direction, for the

George Voutsadakis (LSSU) Complex Analysis October 2014 38 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Subsection 3

Cauchy-Goursat Theorem

George Voutsadakis (LSSU) Complex Analysis October 2014 39 / 83

 Integration in the Complex Plane Cauchy-Goursat Theorem

Simply and Multiply Connected Domains

A domain is an open connected set in the complex plane.
A domain D is simply connected if every simple closed contour
C
lying entirely in D can be shrunk to a point without leaving D.

Example: The entire complex plane is a simply connected

domain. The annulus defined by 1 < |z | < 2 is not simply
connected.
A domain that is not simply connected is called a
multiply connected domain.
A domain with one “hole” is doubly connected;
A domain with two “holes” triply connected, and so on.
Example: The open disk |z | < 2 is a simply connected domain. The
open
George circular
Voutsadakis (LSSU) annulus 1 < |z 2 is doubly connected.October 2014 40 / 83
| < Analysis
Complex
 Integration in the Complex Plane Cauchy-Goursat Theorem

Cauchy’s Theorem
Cauchy’s Theorem (1825)
Suppose that a function f is analytic in a simply connected domain D
and that f ′ is continuous in D. Th˛en, for every simple closed contour C
in D,
f (z )dz = 0.
C

We apply Green’s theorem and the Cauchy-Riemann equations.

Recall from calculus that, if C is a positively oriented, piecewise
smooth, simple closed curve forming the boundary of a region R
within D, and if the real-valued functions P (x , y ) and Q(x , y )
that contains C and R , thenC Pdx + Qdy = R (∂Q ∂x −
∂P
along with their first-order partial de¸rivatives are cont ˜inuous on a
)dA.
Since f ′ is continuous throughout
∂y D, the real and imaginary parts
domain
of f (z ) = u + iv and their first partial derivatives are continuous
throughout D.
George Voutsadakis (LSSU) Complex Analysis October 2014 41 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Proof of Cauchy’s Theorem

We have by Green’s Theorem

˛ ¨
Pdx + Qdy ( ∂Q − )dA.
∂P
R ∂x ∂y
=
C
B¸ y continuity¸of u, v and their first
f (z )dz =∂u C u(x , ˜
˜C (− ∂v
y )dx∂u− v (x , y )dy + i C v (x , y )dx + u(x ,
partial
R
derivatives,
∂x − ∂y )dA + i R ( ∂x −
¸∂v )dA. f being analytic in D, u
y )dy =
and the Cauchy-Riemann equations: ∂u ∂v ∂u
v ∂ysatisfy ∂x = ∂y ,
∂v
= − ∂x
.T herefor ∂y
e, ˜ ˜ ∂v
f (z ) = R (−∂x + ∂x )dA + i R (∂y − ∂y
∂v ∂v ∂v
¸C
dz )dA
= 0.

George Voutsadakis (LSSU) Complex Analysis October 2014 42 / 83

 Integration in the Complex Plane Cauchy-Goursat Theorem

The Cauchy-Goursat Theorem

Edouard Goursat proved in 1883 that the assumption of continuity of

f ′ is not necessary to reach the conclusion of Cauchy’s theorem:
Cauchy-Goursat Theorem
Suppose that a function f is analytic in a simply connected domain D.
Then, for every simple closed contour C in
˛
D,
f (z )dz = 0.
C

Since the interior of a simple closed contour is a simply

connected domain, the Cauchy-Goursat theorem can also be
stated
If f isas:
ana¸lytic at all points within and on a simple closed
contour
C , then C f (z )dz = 0.
George Voutsadakis (LSSU) Complex Analysis October 2014 43 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Applying the Cauchy-Goursat Theorem

I ¸
Evaluate z
C e dz, where the contour C is shown below.

f (z ) = ez is entire. Thus, it is analytic at

all points within and on the simple closed
con- tour C .
It fol¸lows from the
Cauchy-Goursat theorem that C ez dz =
¸ 0.
We have C ez dz = 0, for any simple closed contour in the
complex
plane.
Moreover, for any simple closed contour C and any entire function f ,
such as f (z ) = sin z , f (z ) = cos z , and p(z ) = anzn + a n−1 z n− 1
+a·1·z ·+ a0 , n = 0, 1, 2, . . ., we also
+
˛ ˛ ˛
have
sin zdz = 0, cos zdz = p(z )dz = 0,
C 0, C etc.
C
George Voutsadakis (LSSU) Complex Analysis October 2014 44 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Applying the Cauchy-Goursat Theorem

II ¸
Evaluate 1 dz, where C is the ellipse (x − 22) + 4
1 (y − 5) =
1. C z2 2
2
The rational function f (z ) = z 1 is analytic everywhere except
z = 0. But z = 0 is not a point interior to or on the simple
at
closed
elliptical contour C .

Thus, again by the Cauchy-Goursat Theorem, we

get
˛ 1
2
dz = 0.
C z

George Voutsadakis (LSSU) Complex Analysis October 2014 45 / 83

 Integration in the Complex Plane Cauchy-Goursat Theorem

Cauchy-Goursat Theorem for Multiply Connected

Domains
If f is analytic¸in a multiply connected domain D, then we cannot
conclude that C f (z )dz = 0, for every simple closed contour C in
D.
Suppose that D is a doubly connected domain and C and C1 are
simple closed contours placed as follows:
Suppose, also, that f is
- analytic
rior to C buton each
exterior to
contour
C1. and at each
point inte-
By introducing the crosscut AB , the ¸region bound´ed between

the cu¸rves is now sim´ply connected. So: ¸C f (z )dz + ¸AB f

(z )dz
+ − C1 f (z )dz + − AB f (z )dz = 0 or C f (z )dz = C1 f (z )
dz.
George Voutsadakis (LSSU) Complex Analysis October 2014 46 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Applying Deformation of Contours

¸
Evaluate C z
1
dz , where C is the black
contour: −i We choose the more convenient circular
contour C1 drawn in blue. By taking the
radius of the circle to be r = 1, we are
guaranteed that C1 lies within C . C1 is the
circle |z − i | = 1.
It can be parametrized by

z = i + eit , 0 ≤ t ≤ 2π.
From z − i = eit and dz = ieit dt, we
˛ ˛
1 dzget: 1 ˆ 2π it
ie
dz = dt
C z − i 0 eit
= ˆC12π
z−i = i dt =
2πi .
0
George Voutsadakis (LSSU) Complex Analysis October 2014 47 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

A Generalization
This result can be generalized: If z0 is any constant complex number

˛interior to any simp le closed contour C , and n an integer, we have

1 dz = 2πi , if n = 1
(z − z 0 )n 0, if n 6=
That
C
the
. integral is zero when n 6= 1 follows only partially from
the1 Cauchy-Goursat theorem.
1
When n = 0 or n is a polynomial and therefore entire.
¸ (z − z0 )
negative,
Then, 1
n dz = 0.
C (z − z0 )
clearly,
It is not very difficult to see that the integral is still zero when n is
a positive integer different from 1.
Analyticity of the function f at all points within¸and on a
simple closed contour C is sufficient to guarantee that C f (z )
dz = 0.
1 1
Example:
This result If emC is the circle |z | = 1, = 0, buti.e.,
dznecessary, f (zit)
¸ phasizes that analyticity isC not
z2
z2
then
is not analytic at z = 0 within =
can
C
George . happen
Voutsadakis that C f (z )dzComplex
(LSSU) = 0 Analysis
without f bei¸ng analytic within
October 2014 48 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Applying the Formula for the Integral of 1/(z −

z0)n Evaluate ¸ 5z +7 dz, where C is circle |z − 2| =
C z 2 +2z
The denominator
−3 2.
factors
as z 2 + 2z − 3 = (z − 1)(z + 3). Thus, the
integrand fails to be analytic at z = 1 and z = −3.
Of these two points, only z = 1 lies within
the contour C , which is a circle
centered at z = 2 of radius r = 2. By
partial fractions
5z + 7 3
2
¸ =
Hence, 5z +7 dz = 3 ¸ +1 dz. The first
1 + 2
1
C z 2 +2z − 3 C z − dz
the value ofC the
z
has the value 2πi , whereas . second integral is 0
+ 3¸ integral
+ 2z − Hence,
by the Cauchy-Goursatztheorem.
2 3 z−1
˛ Lecture up to here. z + 3
5z + dz = 3(2πi ) + 2(0) =
C
7 6πi .
z 2 + 2z − 3
George Voutsadakis (LSSU) Complex Analysis October 2014 49 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Cauchy-Goursat Theorem: Multiply Connnected Domains

If C , C1 , and C2 are simple closed contours as shown below
and f is analytic on each of
the three contours as well as
at each point interior to C
but exterior to both C1 and
C2,
C , we get ¸C f (z )dz +
¸ f (z )dz + − C2 f (z )dz = 0,
then by int¸roducing cros−¸Cscuts
1
between¸C1 and C and between C2
f (z
whence
C )dz =¸ C1 f (z )dz + C2 f (z )
and
dz.
Cauchy-Goursat Theorem for Multiply Connnected Domains
Suppose C , C1, . . . , Cn are simple closed curves with a positive
orientation, such that C1, C2, . . . , Cn are interior to C , but the regions
interior to each Ck , k = 1, 2, . . . , n, have no points in common. If f is
analytic
k = 1, 2,on
. .each contourf and
. , n, then at eachn p¸oint finterior
(z )dz Pto C b¸ut exterior
C (z )dz k=1 Ck
= all the Ck ,
to .
George Voutsadakis (LSSU) Complex Analysis October 2014 50 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Integrals in Multiply Connected Domains

¸
1
Evaluate C z 2 +1
dz, where C is the circle |z | =
4.
The denominator of the integrand factors as z 2 + 1 = (z − i )(z + i
1
So,
). the integrand z 2+1 is not analytic at z = i and at z = −i .
points lie within C . Using Both partial 1
= 1 1 − 1 1 .
¸ z 2 +1 2i z − i 2i z + i
fractions,
whenc ¸ 1 dz = 1 ( 1 − 1 )dz
C z 2 +1 2i C z − i z
e .
Surround+zi = i and z = −i by circular con-
tours C1 and C2 , respectively, that lie
entirelyC . The choice |z − i | = 2 for C1
within
1
and
|z + i | = ¸12 for C2 will
suffice.
We have 1 dz =
C z +1
2

¸ ¸
1 ¸ ( 1 − 1 )dz 1 ( 1 − 1 )dz = 2i1 C z 1− i dz −
1
2i ¸C1 +z − i z +i 1 2i C 1 z ¸− i 1 z 1
1 1
2i C1 z +i dz +2i C2 z − +i
i
dz −2
2i C2 z +i
dz = 2i 2πi − 0 + 0 −2i1 2πi =
1
¸ 0.
George Voutsadakis (LSSU) Complex Analysis October 2014 51 / 83
 Integration in the Complex Plane Cauchy-Goursat Theorem

Non-Simple Closed Contours

Throughout the foregoing discussion we assumed that C was a

simple closed contour, in other words, C did not intersect itself.
It can be shown that the Cauchy-Goursat theorem is valid for
any closed contour C in a simply connected domain D.
For a contour C that is closed but not
simple, if f is analytic in D, then
˛
f (z )dz = 0.
C

George Voutsadakis (LSSU) Complex Analysis October 2014 52 / 83

 Integration in the Complex Plane Independence of Path

Subsection 4

Independence of Path

George Voutsadakis (LSSU) Complex Analysis October 2014 53 / 83

 Integration in the Complex Plane Independence of Path

Path Independence
Definition (Independence of the Path)
´
Let z0 and z1 be points in a domain D. A contour integral C f (z )dz
is said to be independent of the path if its value is the same for all
contours C in D with initial point z0 and terminal point z1.
The Cauchy-Goursat theorem holds for closed contours, not
just simple closed contours, in a simply connected domain D.
Suppose that C and C1 are two contours lying
entirely in a simply connected domain D and both
with
z initial
0 and point
terminal point z1. C joined with −´C1 forms a

c´losed contour. Thus, if f is ´analytic in D´, C f (z )dz

+
− C1 f (z )dz = 0. Therefore, C f (z )dz = C1 f (z )
dz .

Theorem (Analyticity
George Voutsadakis (LSSU) Implies Path Independence)
Complex Analysis October 2014 54 / 83
 Integration in the Complex Plane Independence of Path

Choosing a Different Path

´
Evaluate C 2zdz, where C is the contour shown in blue.

The function f (z ) = 2z is entire. By

the theorem, we can replace the
piecewise smooth path C by any
convenient contour C1 joining z0 = −1 and
z1 = −1 + i . We choose the contour C1
to be the vertical line segment x = −1, 0
≤ 1..
Since z = − 1 + iy , dz≤ =y idy z=x+iy=-1+iy
Therefore, ´
´C 2zdz = dz=i dy
= ´1C2(−1
1
+ iy )
2zdz 01
= ´0 idy (−2i − 2y )dy
2 1
= (−2iy − y ) 0
= − 1 − 2i .
George Voutsadakis (LSSU) Complex Analysis October 2014 55 / 83
 Integration in the Complex Plane Independence of Path

Antiderivative
s ´
A contour integ´ralz1 C f (z )dz that is independent of the path C
usually written z0 f (z )dz, where z0 and z1 are the initial
is
terminal points of C .
and
Definition
(Antiderivative)
Suppose that a function f is continuous on a domain D. If there exists a
function F such that F ′(z ) = f (z ), for each z in D, then F is called an
antiderivative of f .
Example: The function F (z ) = − cos z is an antiderivative of
f (z ) = sin z since F ′ (z ) = sin z .
The most gener´al antiderivative, or indefinite integral, of a
function f (z ) is written f (z )dz = F (z ) + C , where F ′
(z ) = f (z ) and C is some complex constant.
Differentiability implies continuity, whence, since an antiderivative
F of a function f has a derivative at each point in a domain D, it
is necessarily analytic and hence
George Voutsadakis (LSSU)
continuous at each point
Complex Analysis
in D. 56 / 83
October 2014
 Integration in the Complex Plane Independence of Path

Fundamental Theorem for Contour

Integrals
Fundamental Theorem for Contour Integrals
Suppose that a function f is continuous on a domain D and F is an
antiderivative of f in D. Then, for any contour C in D with initial point
z0 and terminal point z1, ˆ
f (z )dz = F (z1 ) − F (z0 ).
C

We prove the FTCI in the case when C is a smooth curve

parametrized by z = z (t), a ≤ t ≤ b. The initial and terminal points
on C are
´ z (a) = z0 and´zb (b) = z1′. Since F (z )′ = f (z ),
′ for all z
in D, ´C f (z )dz = b f (z (t ))z (t )dt = a F (z (t ))z
′

= ´a (t
ab d )dt
dt F (z (t ))dt = F (z ba
= (t ))| F (z (b))
− F (z (a))
= F (z1 ) −
George Voutsadakis (LSSU) Complex Analysis October 2014 57 / 83
 Integration in the Complex Plane Independence of Path

Applying the Fundamental Theorem

I ´
The integral C 2zdz, where C is
shown

is independent of the path. Since f (z ) = 2z is an entire function, it

is continuous. Moreover, F (z ) = z 2 is an antiderivative of f since
F ′(z ) = 2z = f (z ). Hence, by the Fundamental Theorem, we have
´ −1+i − 1+ i
−1 2zdz = z2 − 1
= (−1 + i )2 − (−1)
2
= − 1 − 2i .

George Voutsadakis (LSSU) Complex Analysis October 2014 58 / 83

 Integration in the Complex Plane Independence of Path

Applying the Fundamental Theorem

II

´
Evaluate C cos zdz, where C is any contour with initial point z0 =
0
and terminal point z1 = 2 + i .
F (z ) = sin z is an antiderivative of f (z ) = cos z , since
F ′ (z ) = cos z =´f (z ). Therefore, by the Fundamental
hav
Theorem, we
e C cos zdz = ´02+ i cos
2+i
zdz
= 0
sin (2 + i ) − sin
0| (2 +
sin z sin
= i ).
=

George Voutsadakis (LSSU) Complex Analysis October 2014 59 / 83

 Integration in the Complex Plane Independence of Path

Some Conclusions

Observe that i¸f the contour C is closed, then z0 = z1

consequently, C f (z )dz = F (z 1 ) − F (z )
and, ´
=
Since0. the value of f (z )dz 0
depends only on the points z0 and
C
this
z1, value is the same for any contour C in D connecting these
points:
I´f a continuous function f has an antiderivative F in D,
then
C f (z )dzwe
M oreover, is have
independent of the path.
a sufficient
condition: ´
If f is continuous and C f (z )dz is independent of the
path C ´
Assume f is continuous and an
in a domain D, then f has antiderivative everywhere in D.
C f (z )dz is independent o f thezpath
domain D and that F is a function defined by F (z ) z=0´ f
in
wherea
(s)ds, s denotes a complex variable, z 0 is a fixed point
in D, and z
represents an y point in D. We wish to show that F ′(z ) = f (z ),
that F (z ) =´z0z f (s)ds is an antiderivative of f in
i.e.,
D.
George Voutsadakis (LSSU) Complex Analysis October 2014 60 / 83
 Integration in the Complex Plane Independence of Path

´
F (z ) = z0 z f (s)ds is an Antiderivative of f in
D We
´
F (z + ∆ z ) − F (z ) ´zz +∆ z f (s)ds − z0 f (s)ds = ´ z
have f
0 z
=
Because D is a domain,z we can choose
(s)ds . ∆z so that +∆
z +z ∆z is in
D.
Moreover, z and z + ∆z can be joi´ned by a stra´z ight segment. With
fixed, we can write f (z )∆ z = f zz +∆ z ds = z + ∆ z f (z )ds
z
(z
f (z) ) ∆1z ´z z f (z )ds. Therefore, we or
=F (z +∆ z )− F+∆ z have 1 ´
— f (z ) =∆ z zz [f (s) − f (z )]ds. Since f
(z ) ∆ z
continuous atisthe point z +∆ , for
z any ε > 0, there exists a δ > 0, so
that
|f (s) − f (z )| < ǫ whenever |s − z | < δ. Consequently, if we
choose
F (z +∆ z )− F
1
´ z +∆ z
(z ) — f (z ) = ∆ z
z [f (s) − f (z )]ds =
∆ z ∆so
´ zthat |∆z | < δ, it fo llows 1from the ML-inequal ity, that
z
∆1z [f (s) − f (z )]ds ≤ | ∆ z |ε|∆ z | = ε. Hence,
+∆ z F (z + ∆ z )− F
lim
z ∆z (z ) ∆ z
= f (z ) or F′ (z ) = f
→0 (z ).
George Voutsadakis (LSSU) Complex Analysis October 2014 61 / 83
 Integration in the Complex Plane Independence of Path

Existence of Antiderivative
If f is an analytic function in a simply connected domain D, it is
continuous throughout D. This implies, by the Path Independence
Theorem, that path independence holds for f in D. Therefore,
Theorem (Existence of Antiderivative)
Suppose that a function f is analytic in a simply connected domain D.
Then f has an antiderivative in D, i.e., there exists a function F such
that F ′(z ) = f (z ), for all z in D.
We have seen that, for |z | > 0, −π < arg(z ) < π,z1 is the
derivative
of Lnz . Thus, under some circumstances Lnz is an antiderivative
of
z , but one must be
1

If D is the entire complex plane without the origin, z1 is analytic in

careful!
this multiply connected domain. If C is any s¸im1ple closed
containing the origin, it does not follow thatC z dz = 0. In this
contour
Lnz
case,is not an antiderivative of z1 in D since Lnz
is notfails
(Lnz analytic
to beinanalytic
D on the non-positive real
axis).
George Voutsadakis (LSSU) Complex Analysis October 2014 62 / 83
 Integration in the Complex Plane Independence of Path

Using the Logarithmic Function

´
Evaluate C z
1
dz , where C is the contour
shown:
Suppose that D is the simply connected do-
main defined by x > 0, y > 0, i.e., the first
quadrant. In this case, Lnz is an
antideriva-
tive of z1 since both these functions are ana-
lytic in
D. Ln z=Log z=ln|z|+i Arg z
Therefore, ˆ
ˆ 2i 1
1 dz = Lnz 32i| = Ln(2i ) −
Cz
dz = 3 z
Ln3.
´Recall Ln(2i ) = loge 2 +2 π i and Ln32 = πloge 3.
1 dz = log 2 + π i − loge 3 = loge +
Hence,
C z e 2 3
i. 2

George Voutsadakis (LSSU) Complex Analysis October 2014 63 / 83

 Integration in the Complex Plane Independence of Path

Using an Antiderivative of z −1/2

´
Evaluate 1
dz, where C is the line segment between z = i and
C z 1/2
z1 = 9.
0
We take f1(z ) = z 1/2 to be the principal branch of the square
function.
root In the domain |z | > 0, −π < arg(z ) < π, the
function
1 ) =
z 1/2 = z
f 1 (z 1 − 1 / 2 is analytic and possesses the

(z ) = 2z .
Fantiderivative 1/2

Hence, ˆ ˆ9
1 dz 1 dz
z 1/2
C = i
z 1/2 = 9
2z 1/2 i√ √
= 2[3 − ( 22 + i 2
)] √ √
(6 − 2) 2 − i 2.

George Voutsadakis (LSSU) Complex Analysis October 2014 64 / 83

 Integration in the Complex Plane Independence of Path

Integration-By-Parts
In calculus indefinite integrals of certain kinds can be evaluated
by integration by parts:
ˆ ˆ
f (x )g (x )dx = f (x )g (x ) −
′

g (x )f ′ (x )dx .
´ ´
More compactly, udv = uv − vdu.
Suppose f and g are analytic in a simply connected domain
D. Then
ˆ ˆ
f (z )g (z )dz = f (z )g (z ) −
′

g (z )f ′ (z )dz .
f (z )g′ (z )dz = f (z )g (z
z0 )|
z1 − g (z )f ′
z0 (z )dz . z0

In addition, if z0 and z1 areComplex

George Voutsadakis (LSSU) the Analysis
initial and terminal points
Octoberof
2014 65 / 83
 Integration in the Complex Plane Independence of Path

The Mean Value Theorem for Definite

Integrals

The Mean Value Theorem for Definite Integrals: If f is a real

function continuous on the closed interval [a, b], then there
exists a number c in the open interval (a, b), such that
ˆb
f (x )dx = f (c)(b − a).
a

Let f be a complex function analytic in a simply connected domain

D. Then, f is continuous at every point on a contour C in D with
initial point z0 and terminal point z1.

Unfortunately, no´ z1analog of the Mean Value Theorem exists for

contour integral z0 f (z )
the
dz .

George Voutsadakis (LSSU) Complex Analysis October 2014 66 / 83

 Integration in the Complex Plane Cauchy’s Integral Formulas

Subsection 5

Cauchy’s Integral Formulas

George Voutsadakis (LSSU) Complex Analysis October 2014 67 / 83

 Integration in the Complex Plane Cauchy’s Integral Formulas

Cauchy’s First Formula

If f is analytic in a simply connected domain D and z0 is a point in
D, quotient zf −(z z)0 is not defined at z0 and, hence, is not analytic in
the
D.
Therefore, we cannot conclude that the integral of f (z ) z − z0
simple
aroundclosed
a contour C that contains z0 is
zero.
Indeed,the integral of zf −(zz)0 around C has the value 2πif
(z0). (Cauchy’s Integral
Theorem
Formula)
Suppose that f is analytic in a simply connected domain D and C is any
simple closed contour lying entirely within D. Then, for any point z0
within C ˛
1 f
, dz
f (z0 ) = (z ) z − z
2πi C 0.

Let D be a simply connected domain, C a simple closed contour in

D, and z0 an interior point of C . In addition, let C1 be a circle
centered
at z0 with radius small enough so that C1 lies within the in¸terior of
By the principle of deformation of contours, ¸C zf (z ) f (z )
− z0 dz = C1 z − z0
C.
George Voutsadakis (LSSU) Complex Analysis October 2014 68 / 83
 Integration in the Complex Plane Cauchy’s Integral Formulas

Proof of Cauchy’s Integral Formula

¸
From f (z ) ¸ f (z ) dz, we get by adding and
C z − z0 dz = C1 z ¸
f (z0 ) in the subtracting
−¸z0 C zf (z )
−z 0 dz = C1
f (z0 )− f (z0 )+f (z )
dz
¸ z − z0
numerator:
f (z0 ) ¸ dz + C1 f (z )− f (z0 )
dz. We know ¸
C1 z − z0 dz =
1 1
= ¸
z
C1¸ z
that
whence f (z )
dz = 2πif
− z0 (z0 ) C1f (z )− f (z0 ) 2πi ,
− z0C z − z0 z dz.
+
Since f is− continuous
z0 at z0, for any ε > 0, there exists a δ > 0,
such that |f (z ) − f (z0)| < ε, whenever |z − z0| < δ. In particular,
1
choose
if we C1 to be |z − 0z | =2 δ < δ, then by the ML-
inequality,
¸ f (z )− f 0 dz ≤ ε 2π = 2πε. Thus, the absolute value of
C (z z) −z 0 δδ/2
1
themade arbitrarily small by taking the radius of the
integral2 can be
circle C1 to be suffi¸ciently small. This implies that the integral is 0.
We conclude that C z f−(zz0) dz = 2πif
(z0 ).

George Voutsadakis (LSSU) Complex Analysis October 2014 69 / 83

 Integration in the Complex Plane Cauchy’s Integral Formulas

Using Cauchy’s Integral Formula

Cauchy’s integral formula shows that the values of an analytic

function f at points z0 inside a simple closed contour C are
determined by the values of f on the contour C .
Since we often work problems without a simply connected domain
explicitly defined, a more practical restatement is:
If f is analytic at all points within and on a simple clos¸ed
contour
C , and z0 is any point interior to C , then f (z02πi
) C zf (z )
− z0 dz.
= 1 ¸ 2
Example: Evaluate C z+4−z4z dz , where C is the circle |z |
We identify f (z ) = +iz 2 − 4z +=4 2.
and z = − i as a point within the
0
circle C . Next, we observe that f is analytic at all points within
andz 2 −o¸
4zn the contour C . Thus, by the Cauchy integral formula,
C +4z dz = 2πif (−i ) = 2πi (3 + 4i ) = π(−8 +
+i 6i ).
George Voutsadakis (LSSU) Complex Analysis October 2014 70 / 83
 Integration in the Complex Plane Cauchy’s Integral Formulas

Another Application of Cauchys Integral

Formula
¸
z
Evaluate dz, where C is the circle |z − 2i | = 4.
C z 2+ 9

By factoring the denominator as z 2 +

(z
9 =− 3i )(z + 3i ), we see that 3i is the
only point within the closed contour C at
which the integrand fails to be analytic. By
z
rewrit- z
ing the integrand as z w
z
z 2 + 9 = z − 3i e
identify f (z ) z +3i
= +3i ,
The function f is analytic at all points within and on the contour C
. H˛ ence, by Cauc˛z hy’s integral formula
z 3i
dz = z + 3i
dz = 2πif (3i ) = 2πi =
Cz +
2
C z − 3i 6i
πi .
9
George Voutsadakis (LSSU) Complex Analysis October 2014 71 / 83
 Integration in the Complex Plane Cauchy’s Integral Formulas

Cauchy’s Second Formula

We prove that the values of the derivatives f (n)(z0), n = 1, 2, 3, . . .

of an analytic function are also given by an integral formula.
Theorem (Cauchy’s Integral Formula for Derivatives)
Suppose that f is analytic in a simply connected domain D and C is any
simple closed contour lying entirely within D. Then, for any point z0
within C ˛
, f (n) (z0 ) = n! f dz.
(z ) 2πi C
Partial Proof (for n = 1): By (z −the
z0 )definition of the derivative
n+ 1

and f (z0 +∆ z )− f (z0

Cauchy’s Integral Formula, f ′ (z0 ) = ∆˛z )
=
lim 1 f → 0
dz − f ∆z
˛ z − (z(z + dz =
lim∆ z → 0 ) ∆z ) z(z− )z0
˛
2πi ∆ z C 0
1 f
C dz
lim∆ z → 0 (z )
2πi C (z − z0 − ∆ z )(z − z0 ) .
George Voutsadakis (LSSU) Complex Analysis October 2014 72 / 83
 Integration in the Complex Plane Cauchy’s Integral Formulas

Prof of Cauchy’s Second Formula for n =

1
We work out some preliminaries:
Continuity of f on the contour C guarantees that f is bounded, i.e.,
there exists real number M, such that |f (z)| ≤ M, for all points z on C
. In addition, let L be the length of C and let δ denote the shortest
distance between points on C and the point z0. Thus, for all points z
1
on C , we have |z − z0 | ≥ δ, ≤ 12
|z − z0 |2 δ
or
Furthermore, 1
if we choose |∆z| ≤ 2δ, then . |z — z0 − ∆ z| ≥
||z − z 0| − |∆ z|| ≥ δ − |∆ z| ≥ δ,12 whence |z − z012 − ∆ z |≤ δ.
˛ ˛
f f
dz − dz =
Now, (z (z
− ) 2 (z − z 0 −(z
∆ z) )(z −
C 0
˛ −∆ z f (z ) z ) 02ML|∆ z
Cz ) . The last expression
(z − z − ∆ z )(z − z )2 dz ≤ |
C 0 0
δ3
approaches zero as ∆z → 0, ˛
whence f (z0 + ∆ z ) − f (z0 ) 1 f
f (z0 ) = ∆ z
′
(z )
dz.
∆z 2πi C (z − z0 )2
lim →0
=
George Voutsadakis (LSSU) Complex Analysis October 2014 73 / 83
 Integration in the Complex Plane Cauchy’s Integral Formulas

Using Cauchy’s Integral Formula for

Derivatives ¸
z +1
Evaluate dz, where C is the circle |z | =
C z 4 +2iz 3
1.
Inspection of the integrand shows that it is not analytic at z = 0
and
z
z = −2i , but only z + = 0 lies within
+1
the closed contour. By writing
the integrand as = z + 2i
we can identify, z0 = 0, n =
1
z + 2iz
4
2, z 3
z +1 3
and f (z ) =z + 2i . The quotient rule gives f ′(z ) =(z −+2i 1+2i
)
fand(z ) = (z
′′ 2−2 4i
3 , whence f
′′ (0) = 4i2i − 1
. T herefore, we
get +2i )
˛
z+1 f ′′ (0)
=
C z 2πi
4 + 4z 3 2!
dz 2πi 2i − 1
=
2!
4i π
= −π+ i.
4 2

George Voutsadakis (LSSU) Complex Analysis October 2014 74 / 83

 Integration in the Complex Plane Cauchy’s Integral Formulas

Another Application of the Integral Formula for

Derivatives ¸
Evaluate z 3 +3 dz , where C is the figure-eight contour
shown
below:
C z (z − i )2

Although C is not a simple closed contour, we can

think of it as the union¸ of two simple c¸losed
C1 and C2. We write z 3 +3 dz = z 3 +3 dz +
contours C z (z − i )2 C1 z (z − i )
z +3
3
¸ z 3+3 2 ¸ z 3 +3
¸C2 z (z − i )2dz = − (z − i )2 z
dz =
−C 1 z dz + C2 (z − i )2
— I1 + I2 .
¸ 3
z (z +32
I1 = dz = 2πif (0) = 2πi −3) = − 6πi .
−i)
−C 1
( 3 2z3 − 3
For I2, f (z) z= z z+3 , whence f ′ (z) = z 2 , and f ′ (i ) = 3 + 2i .
¸ z 3 +3 Thus,
z
I2 = C dz = 21!π i f ′ (i ) = 2πi (3 + 2i ) = − 4π +
2 (z − i )2
¸ 3
6πi .
z +3
Finally, dz = − I1 + I = 6πi + (−4π + 6πi ) = − 4π +
C z (z − i )
12πi . 2
George Voutsadakis2 (LSSU) Complex Analysis October 2014 75 / 83
 Integration in the Complex Plane Consequences of the Integral Formulas

Next lecture will start here.

Subsection 6

Consequences of the Integral Formulas

George Voutsadakis (LSSU) Complex Analysis October 2014 76 / 83

 Integration in the Complex Plane Consequences of the Integral Formulas

The Derivatives of an Analytic Function are

Analytic
Theorem (Derivative of an Analytic Function Is Analytic)
Suppose that f is analytic in a simply connected domain D. Then f
possesses derivatives of all orders at every point z in D. The
derivatives f ′ , f ′′ , f ′′′ , . . . are analytic functions in D.

If f (z ) = u(x , y ) + iv (x , y ) is analytic in a simply connected

domain D, its derivatives of all orders exist at any point z in D.
Thus, f ′, f ′′ , f ′′′ , . . . are continuous. From
∂u
f ′ (z ) = ∂x
∂u
+ i ∂v
∂x =
∂v
− i ∂y ,
∂y2 ∂ 2u
f ′′(z ) ∂ 2 v —
∂ 2∂x
∂x u2 + i ∂∂xv = ∂y
= .. 2 ∂y ∂x i

we can also conclude that the real functions u and v have

continuous partial derivatives of all orders at a point of analyticity.
George Voutsadakis (LSSU) Complex Analysis October 2014 77 / 83
 Integration in the Complex Plane Consequences of the Integral Formulas

Cauchy’s Inequality
Theorem (Cauchy’s Inequality)
Suppose that f is analytic in a simply connected domain D and C is a
circle defined by |z − z0| = r that lies entirely in D. If |f (z )| ≤ M, for
points z on C ,
all n!M
(n)
then |f (z0)| ≤ n .
r

From the hypothesis, f (z ) = |f ≤ M . Thus, by

r n+1 r n+1
Formula for Derivatives(zand − z0 )the
n+ 1
(zML-
)| Cauchy’s
inequality, ˛ f n! M n!
|f (n) (z0 )| = n! n+1 2πr = r n
.
2π (z −(zz ) )n+1 dz ≤ M 2π r
C 0
The number M depends on the circle |z − z0| = r . But, if n = 0, then
M ≥ |f (z0)|, for any circle C centered at z0, as long as C lies within
D. Thus, an upper bound M of |f (z )| on C cannot be smaller than
|f (z0 )|.
George Voutsadakis (LSSU) Complex Analysis October 2014 78 / 83
 Integration in the Complex Plane Consequences of the Integral Formulas

Liouville’s Theorem

Although the next result is known as “Liouville’s Theorem”, it

was probably first proved by Cauchy.
The gist of the theorem is that an entire function f , one that is
analytic for all z , cannot be bounded unless f itself is a
constant:
Theorem (Liouville’s Theorem)
The only bounded entire functions are constants.
Suppose
Then, is an
forf any entire
point z0 ,bounded function,
by Cauchy’s i.e., |f |f(z
Inequality, ′ )| ≤| ≤M,M for all
r . By
(z
z . ) 0
making r arbitrarily large we can make |f (z0)| as small as we wish.
′

This means f ′(z0) = 0, for all points z0 in the complex plane. Hence,
by a preceding theorem, f must be a constant.

George Voutsadakis (LSSU) Complex Analysis October 2014 79 / 83

 Integration in the Complex Plane Consequences of the Integral Formulas

Fundamental Theorem of Algebra

Liouville’s Theorem enables us to establish the
celebrated Fundamental Theorem of Algebra
If p(z ) is a nonconstant polynomial, then the equation p(z ) = 0
has at least one root.
Suppose that the polynomial p(z ) = anzn + a n−1 z n−1 + · · · +
a1z + a0, n > 0, is not 0 for any complex number z . This
reciprocal
implies ofthe
that p, f (z ) =p(z 1 , is an entire function.
Now )
|f (z )| = |an z n +an− 1 z n −11+ ···+a1z +a0|
1
= 1 a .
|z |n|an + a z + · · · + na−11 + 0n |
n−

z z
Thus, |f (z )| → 0 as |z | → ∞ . So the function f must be
bounded for finite z . By Liouville’s Theorem, f is a constant.
Hence, p is a constant. But this contradicts p not being a
constant polynomial. Therefore, there must exist at least one z
for
George which(LSSU)
Voutsadakis p(z ) = 0. Complex Analysis October 2014 80 / 83
 Integration in the Complex Plane Consequences of the Integral Formulas

Morera’s Theorem

Morera’s theorem, which gives a sufficient condition for analyticity,

is often taken to be the converse of the Cauchy-Goursat Theorem:
Theorem (Morera’s Theorem)
¸
If f is continuous in a simply connected domain D and if C f (z )dz
= 0, for every closed contour C in D, then f is analytic in D.
¸
By the hypotheses of continuity of f and ´C f (z )dz = 0, for every
closed contour C in D, we conclude that C f (z )dz ´is independent
the path. T hen, the function F , defined by F (z )z0= z
f
of denotes
s(s)ds a complex variable, z is a fixed point in
(where 0
D, and z any
point in D) is an antiderivative of f , i.e., F ′(z ) = f (z ). Hence, F is
analytic in D. In addition, F ′(z ) is analytic in view of the analyticity
of the derivative of any analytic function. Since f (z ) = F ′(z ), we
see that f is analytic in D.
George Voutsadakis (LSSU) Complex Analysis October 2014 81 / 83
 Integration in the Complex Plane Consequences of the Integral Formulas

The Maximum Modulus Theorem

We saw that, if a function f is continuous on a closed and bounded

region R , then f is bounded, i.e., there exists some constant M,
such that |f (z )| ≤ M, for z in R .
If the boundary of R is a simple closed curve C , then the
modulus
|f (z )| assumes its maximum value at some z on the boundary C :
Theorem (Maximum Modulus Theorem)
Suppose
by thatclosed
a simple f is analytic
curve Cand nonconstant
. Then on a|fclosed
the modulus region its
(z )| attains R
bounded
maximum
on C.
If the stipulation that f (z ) 6= 0, for all z in R , is added to the
hypotheses, then the modulus |f (z )| also attains its minimum on
C.

George Voutsadakis (LSSU) Complex Analysis October 2014 82 / 83

 Integration in the Complex Plane Consequences of the Integral Formulas

Finding The Maximum Modulus

Find the maximum modulus of f (z ) = 2z + 5i on the closed

circular region defined by |z | ≤ 2.
We know that |z2| = · z. By replacing z by 2z + 5i , we
z + 5i 2| = (2z + 5i )(2z
|2z have+ 5i ) = (2z + 5i )(2z −
5i ) =
4zz − 10i (z − z) + 25. But, z − z = 2i Im(z ),
whence
|2z + 5i 2| = 4|z | + 20Im(z ) + 25. Because f is a polynomial,
2
it is
analytic on the region defined by |z | ≤ 2. Thus, max|2z + 5i |
occurs p |z |≤ 2
on the its
attains boundary = 2. Im(z
|z | when
maximum There, |2z + 5i
) attains = 41 + on |z | =
its| maximum
20Im(z ). This √
2,
namely, at the point z = 2i . Thus, max|z |≤ 2 |2z + 5i | = 81
=
Note 9. that f (z ) = 0 only at z = − 5
2 i and that this point is
outside
the region defined by |z | ≤ 2. Hence we can conclude that we have
a
minimum when Im(z ) attains √ its minimum on |z | = 2 at z = −2i
.a As
result, min|z |≤ 2 |2z + 5i | = 1 =
1.
George Voutsadakis (LSSU) Complex Analysis October 2014 83 / 83