EXAMPLE

Calculate
∫ (9x 2
− 2 y + 3) dx + (− 2 x − 4 ) dy
C

Where C is
1. The part of he circle x2 + y2 = 1that lies in the first
quadrant oriented anti-clockwise
2. The line segment between the points [1,0] and [0,1]

[0,1]

O [1,0]
 1. x2+y2=1
π
2

∫ (9 x − 2 y + 3) dx + (− 2 x − 4 ) dy =
∫ − 9 cos 2 t sin t + 2 sin 2 t − 3 sin t − 2 cos 2 t − 4 cos t dt =
2

C 0
π π
2 1 2

=
∫
0
∫ ∫
− 9 cos 2 t sin t − 2 cos 2t − 3 sin t − 4 cos t dt = − 9 t 2 dt − 2 cos 2t dt − 3 − 4 =
0 0
1
t 3 
= −9   − 7 = −10
 3 0

2. y = 1– x
1

−
∫ (9 x
C
2
− 2 y + 3) dx + (− 2 x − 4 ) dy =
∫ (9 x
0
2
− 2(1 − x ) + 3)dx + (− 2 x − 4 )(− dx ) =

= − ∫ 9 x 2 + 4 x + 5 dx = −[3 x 3 + 2 x 2 + 5 x ]0 = −10
1
1

0
The identical results of the above calculations are no coincidence.
The reason is that the vector components

(f1(x,y), f2(x,y)) = (9x2 – 2y + 3,–2x – 4)

are the partial derivatives by x and y respectively of the function

F (x,y) = 3x3 – 2xy + 3x – 4y.

Let f (x,y) = ( f1(x,y), f2(x,y)) be a vector field with f1(x,y), f2(x,y)
continuous in a planar area A containing a regular curve C given by
the parametric equations
x = ϕ(t), y = ψ(t), t∈[a,b]
oriented in correspondence with these parametric equations. Let A
denote the point [ϕ(a),ψ(a)] and B the point [ϕ(b),ψ(b)].
A function F(x,y) in A such that

F(x,y) has continuous first partial derivatives

∂ ∂

f1 ( x, y ) = F ( x, y ), f 2 ( x, y ) = F ( x, y )
∂x ∂y

is called the potential of the vector field f(x,y)

Let the potential F (x,y) of f ( x, y ) exist and calculate the line
integral of f ( x, y ) along C:
b

∫ f1 ( x, y ) dx + f 2 ( x, y ) dy = ∫ f1 (ϕ (t ),ψ (t ))ϕ ' (t ) + f 2 (ϕ (t ),ψ (t ))ψ ' (t ) =

C a

b
= ∫  F (ϕ (t ),ψ (t )) dt = [F (ϕ (t ),ψ (t ))]a =
 d b

a 
dt

= F (ϕ (b ),ψ (b )) − F (ϕ (a ),ψ (a )) = F ( B ) − F ( A).

This means that the result only depends on the value of F(x,y)
at A and B and is independent of the choice of curve C.
Clearly, a similar result can be obtained for 3D curves: If F
(x,y,z) is a potential of f ( x, y, z ) so that
∂
f1 ( x, y, z ) = F ( x, y, z )
∂x
∂
f 2 ( x, y , z ) = F ( x, y , z )
∂y
∂
f 3 ( x, y , z ) = F ( x, y , z )
∂z
then

∫ f ( x, y, z ) dx + f ( x, y, z ) dy + f ( x, y, z ) dz = F (B ) − F (B )
C
1 2 3

for every curve C going from A to B.

Let F(x,y,z) and G(x,y,z) be two potentials of f(x,y,z).
Then
∂ ∂
f1 ( x, y, z ) = F ( x, y , z ) f1 ( x, y, z ) = G ( x, y , z )
∂x ∂x
∂ ∂
f 2 ( x, y , z ) = F ( x, y , z ) f 2 ( x, y , z ) = G ( x, y , z )
∂y ∂y
∂ ∂
f 3 ( x, y , z ) = F ( x, y , z ) f 3 ( x, y , z ) = G ( x, y , z )
∂z ∂z

∂
(F ( x, y, z ) − G ( x, y, z )) = 0
∂x
∂
(F ( x, y, z ) − G( x, y, z )) = 0 G(x,y,z)=F(x,y,z) + const
∂y
∂
(F ( x, y, z ) − G ( x, y, z )) = 0
∂z
Given the functions f1(x,y) and f2(x,y), how can we know that a
function F (x,y) exists such that

∂ ∂
f1 ( x, y ) = F ( x, y ), f 2 ( x, y ) = F ( x, y )
∂x ∂y
 ∂ ∂2 ∂ ∂2
We have f1 ( x, y ) = F ( x, y ) and f 2 ( x, y ) = F ( x, y )
∂y ∂x∂y ∂x ∂y∂x

From what we know about F(x,y), it follows that

∂2 ∂2
F ( x, y ) = F ( x, y )
∂x∂y ∂y∂x

Thus we can conclude that

∂ ∂
f1 ( x, y ) = f 2 ( x, y )
∂y ∂x

is a condition necessary for the existence of the function F(x,y).

It can be proved that this condition is also sufficient.
In the event of a 3D curve and a function F(x,y,z), if
∂ ∂ ∂
f1 = F f2 = F f3 = F
∂x ∂y ∂z
then clearly

∂f1 ∂ 2 F ∂f 2 ∂ 2 F ∂f 3 ∂ 2 F
= = =
∂y ∂x∂y ∂x ∂y∂x ∂x ∂z∂x

∂f1 ∂ F2
∂f 2 ∂ 2 F ∂f 3 ∂ 2 F
= = =
∂z ∂x∂z ∂z ∂y∂z ∂y ∂z∂y
 This means that

∂f 3 ∂f 2 ∂f1 ∂f 3 ∂f 2 ∂f1
= = = (1)
∂y ∂z ∂z ∂x ∂x ∂y

or
∂f 3 ∂f 2 ∂f1 ∂f 3 ∂f 2 ∂f1
− =0 − =0 − =0 (2)
∂y ∂z ∂z ∂x ∂x ∂y

are conditions necessary for the existence of the function F(x,y).

Again, it can be proved that they are also sufficient.
Conditions (2) can also be expressed using the following formal
calculation.
Suppose that, next to the vector

f ( x, y, z ) = f1 ( x, y, z ) i + f 2 ( x, y, z ) j + f 3 ( x, y, z ) k

we also have a formal vector

∂ ∂ ∂
∇= i+ j+ k
∂x ∂y ∂z

This vector is also called the nabla operator.

Let us calculate the vector product

f ( x, y , z ) × ∇
 i j k
∂ ∂ ∂
f ( x, y , z ) × ∇ = =
∂x ∂y ∂z
f1 f2 f3

 ∂f 3 ∂f 2   ∂f1 ∂f 3   ∂f 2 ∂f1 
= − i +  −  j + − k
 ∂y ∂z   ∂z ∂x   ∂x ∂y 

Thus the conditions (2) can be written as

f ( x, y , z ) × ∇ = 0

where 0 is the zero vector.

 The vector field f ( x, y, z ) × ∇ = 0 is called the rotation or curl of
the vector field f ( x, y, z )

If the rotation of a vector field is the zero vector in an area A, the

vector field is said to be irrotational or conservative in A.

The line integral of a vector field is independent of the curve along

which the endpoint is reached from the starting point exactly if the
vector field is irrotational.
To find a potential to functions f1(x,y,z), f2(x,y,z), f3(x,y,z), first
we must verify that
∂f 3 ∂f 2 ∂f1 ∂f 3 ∂f 2 ∂f1
f ( x, y , z ) × ∇ = 0 or = = =
∂y ∂z ∂z ∂x ∂x ∂y

and then perform the following calculations:

F ( x, y , z ) =
∫ f (x, y, z )dx + G( y, z )
1

∂ ∂ ∂ 
G ( y, z ) = F ( x, y, z ) − ∫  f1 ( x, y, z ) dx = F2 ( y, z )
∂y ∂y  ∂y 
G ( y, z ) = ∫ F2 ( y, z ) dy + H ( z )

∂ ∂
H ( z ) = G ( y, z ) − ∫  F2 ( y, z )dy = F3 ( z )

d
dz ∂z  ∂z 

H ( z ) = ∫ F3 ( z ) dz + const

F ( x, y, z ) = ∫ f1 ( x, y, z ) dx + ∫ F2 ( y, z ) dy + ∫ F3 ( z ) dz + const
Example

Find a potential to the vector field

f ( x, y, z ) = (3 x 2 yz − 2 y 3 )i + (x 3 z − 6 xy 2 + 2 yz 2 ) j + (x 3 y + 2 y 2 z )k

Solution
We have
∂ 3 ∂ 3
(x z − 6 xy + 2 yz ) = x + 4 yz = (x y + 2 y 2 z )
2 2 3

∂z ∂y

∂ 3
(x y + 2 y 2 z ) = 3x 2 y = ∂ (3x 2 yz − 2 y 3 )
∂x ∂z

∂ ∂ 3
(3x yz − 2 y ) = 3x z − 6 y = (x z − 6 xy 2 + 2 yz 2 )
2 3 2 2

∂y ∂x
F ( x, y, z ) = ∫ 3 x 2 yz − 2 y 3 dx = x 3 yz − 2 xy 3 + G ( y, z )

∂
G ( y, z ) = x 3 z − 6 xy 2 + 2 yz 2 − (x 3 z − 6 xy 2 ) = 2 yz 2
∂y

G ( y, z ) = y 2 z 2 + H ( z )

F ( x, y, z ) = x 3 yz − 2 xy 3 + y 2 z 2 + H ( z )

H ( z ) = (x 3 y + 2 y 2 z ) − (x 3 y + 2 y 2 z ) = 0
d
dz

F ( x, y, z ) = x 3 yz − 2 xy 3 + y 2 z 2 + const
 GREEN’S THEOREM

Let f(x,y) = f1(x,y) i + f2(x,y) j be a vector field in a planar area M

with continuous first order partial derivatives and let L be a closed
regular curve in M oriented anticlockwise. Let A denote the planar
area bounded by L. Then
∂ ∂ 
∫ f1 ( x, y ) dx + f 2 ( x, y )dy = ∫∫  f 2 ( x, y ) − f1 ( x, y ) dxdy
L A 
∂x ∂y 

f1 ( x, y )i + f 2 ( x, y ) j

∂ ∂
f 2 ( x, y ) − f1 ( x, y ) L
A ∂x ∂y
 y

d
A R
L
c

x
O a b
NOTE From the picture it is clear that the line integrals round all
the rectangles of the grid covering the area A add up to the
line integral round the thick contour covering A since the
line integrals round the inner sides of the rectangle grid are
always calculated twice – each time with a different sign.

We will now calculate the line integral round the yellow

rectangle R in the picture. The line integrals round other
rectangles of the grid covering A would be calculated in the
same way.
y
opposite
opposite
orientation
orientation
x = t , y = d , t ∈ [a, b]
d
x = a, y = t x = b, y = t
R
t ∈ [c, d ] t ∈ [c, d ]
c
x = t , y = c, t ∈ [a, b]

a b x
I R = ∫ f1 ( x, y ) dx + f 2 ( x, y ) dy = I1 + I 2 + I 3 + I 4 where
R

b d
I1 = ∫ f1 ( x, c ) dx I 2 = ∫ f 2 (b, y ) dy
a c
b d
I 3 = − ∫ f1 ( x, d ) dx I 4 = − ∫ f 2 (a, y ) dy
a c

b d
I R = ∫ ( f1 ( x, c ) − f1 ( x, d )) dx + ∫ ( f 2 (b, y ) − f 2 (a, y )) dy
a c
 d b
I R = ∫ ( f 2 (b, y ) − f 2 (a, y )) dy − ∫ ( f1 ( x, d ) − f1 ( x, c )) dx
c a

Using the Lagrange theorem for both integrals, we can write

∂ ∂
d b

I R = (b − a )∫  f 2 ( β x , y ) dy − (d − c )∫  f1 (x,α y ) dx
 
c
∂x  a
∂y 
where c < β x < d and a < α y < b

Finally, using the mean value theorem for integrals, we get

∂ ∂
I R = (b − a )(d − c ) f 2 (β x , β y ) − (d − c )(b − a ) f1 (α x , α y )
∂x ∂y
where c < β y < d and a < α x < b
If we calculate IR for every rectangle of the grid covering A and
add them up, we get, on the left-hand side of the equation, the line
integral of the vector field f1(x,y) i + f2(x,y) j round the thick con-
tour in the picture (see the note) and, on the right-hand side, an in-
tegral sum lying between the lower and upper integral sums of the
double integral

∂ ∂ 
∫∫A  ∂x f 2 ( x, y ) − ∂y f1 ( x, y ) dxdy
Thus, when the norm of the grid tends to zero, this equation tends
to what Green’s theorem says.
EXAMPLE

Calculate the area of the astroid given by the equations

x = a cos3 t , y = a sin 3 t , t ∈ [0,2π ]

For comparison, the circumscribed red circle has a radius of a

By Green’s theorem, we have

1
∫∫1dxdy =
A
2L∫ − y dx + x dy

2π

∫ − y dx + x dy = ∫ − a sin 3
ta 3 cos 2
t ( − sin t ) + a cos 3
ta 3 sin 2
t cos t dt =
L 0

2π 2π
= 3a 2 ∫ sin 4 t cos 2 t + cos 4 t sin 2 t dt = 3a 2 ∫ sin 2 t cos 2 t dt =
0 0

2π 2π
3a 2 3a 2 1 − cos 4t 3πa 2
= ∫0 (sin 2t ) dt = 4 ∫0 2 = 4
2

4
3πa 2
Thus the area of the astroid is
8