Conservative Fields and Potential Functions. (Sect. 16.3) : Example

Conservative fields and potential functions. (Sect. 16.
3)
Review: Line integral of a vector field.
Conservative fields.
The line integral of conservative fields.
Finding the potential of a conservative field.
Comments on exact differential forms.
The line integral of a vector field along a curve.

Definition
The line integral of a vector-valued function F : D Rn Rn ,
with n = 2, 3, along the curve associated with the function
r : [t0 , t1 ] R D R3 is given by
Z
Z t1
F dr =
F(t) r0 (t) dt
t0
Remark: An equivalent expression is:
Example
t1
F dr =
Z 0s1
Z
F dr =
F
C
r0 (t) 0
F(t) 0
|r (t)| dt,
|r (t)|
u
ds,
F
s0
r0 (t(s))
= F(t(s)).
= 0
, and F
where u
|r (t(s))|
Work done by a force on a particle.

Definition
In the case that the vector field F : D Rn Rn , with n = 2, 3,
represents a force acting on a particle with position function
r : [t0 , t1 ] R D R3 , then the line integral
Z
W = F dr,
C
is called the work done by the force on the particle.
Example
A projectile of mass m moving on the

surface of Earth.
y
r
The movement takes place on a

plane, and F = h0, mg i.
W 6 0 in the first half of the

trajectory, and W > 0 on the
second half.
Conservative fields and potential functions. (Sect. 16.3)
Definition
A vector field F : D Rn Rn , with n = 2, 3, is called
conservative iff there exists a scalar function f : D Rn R,
called potential function, such that
F = f .
Example
A projectile of mass m moving on the
surface of Earth.
y
r
The movement takes place on a

plane, and F = h0, mg i.
F = f , with f = mgy .
Example
1
hx1 , x2 , x3 i is
(x12 + x22 + x32 )3/2
conservative and find the potential function.
Show that the vector field F =
Solution: The field F = hF1 , F2 , F3 i is conservative iff there exists a

potential function f such that F = f , that is,
F1 = x1 f ,
F2 = x2 f ,
F3 = x3 f .
Since
h
i

xi
2
2
2 1/2
= xi x1 + x2 + x3
,
(x12 + x22 + x32 )3/2
i = 1, 2, 3,
then we conclude that F = f , with f = q
x12
x22
.
+
x32

Definition
A set D Rn , with n = 2, 3, is called simply connected iff every
two points in D can be connected by a smooth curve inside D and
every loop in D can be smoothly contracted to a point without
leaving D.
Remark: A set is simply connected iff it consists of one piece and

it contains no holes.
y
D
Simply connected
D
Not simply
connected
x

Notation: If the path C Rn , with n = 2, 3, has end points r0 , r1 ,
then denote the line integral of a field F along C as follows
Z
Z r1
F dr =
F dr.
r0
(This notation emphasizes the end points, not the path.)
Theorem
A smooth vector field F : D Rn Rn , with n = 2, 3, defined on
a simply connected domain D Rn is conservativeZwith F = f
iff for every two points r0 , r1 D the line integral
F dr is
C
independent of the path C joining r0 to r1 and holds

Z r1
F dr = f (r1 ) f (r0 ).
r0
Z
Remark: A field F is conservative iff F dr is path independent.
C

Z
r1
Summary: F = f equivalent to
F dr = f (r1 ) f (r0 ).
r0
Proof: Only ().

Z
r1
r1
F dr =
t1
f dr =
r0
r0
t0

(f )
r(t)
r0 (t) dt,
where r(t0 ) = r0 and r(t1 ) = r1 . Therefore,

Z t1
Z r1

d
F dr =
f (r(t) dt = f (r(t1 )) f (r(t0 )).
t0 dt
r0
Z r1
We conclude that
F dr = f (r1 ) f (r0 ).
r0
(The statement () is more complicated to prove.)

Example
Z
(1,2,3)
2x dx + 2y dy + 2z dz.
Evaluate I =
(0,0,0)
Solution: I is a line integral for a field in R3 , since

Z
(1,2,3)
h2x, 2y , 2zi hdx, dy , dzi.
I =
(0,0,0)
Introduce
Z r1 F = h2x, 2y , 2zi, r0 = (0, 0, 0) and r1 = (1, 2, 3), then
F dr. The field F is conservative, since F = f with
I =
r0
potential f (x, y , z) = x 2 + y 2 + z 2 . That is f (r) = |r|2 . Therefore,

Z r1
I =
f dr = f (r1 ) f (r0 ) = |r1 |2 |r0 |2 = (1 + 4 + 9).
r0
We conclude that I = 14.
The line integral of conservative fields. (Along a path.)

Example
Z
(1,2,3)
Evaluate I =
2x dx + 2y dy + 2z dz along a straight line.

(0,0,0)
Solution: Consider the path C given by r(t) = h1, 2, 3i t.

Then r(0) = h0, 0, 0i, and r(1) = h1, 2, 3i. We now evaluate
F = h2x, 2y , 2zi along r(t), that is, F(t) = h2t, 4t, 6ti. Therefore,
t1
F(t) r (t) dt =
I =
t0
Z
I =
h2t, 4t, 6ti h1, 2, 3i dt

0
Z
(2t + 8t + 18t) dt =
We conclude that I = 14.
t 2 1

28t dt = 28
.
2 0
C

Theorem (Characterization of potential fields)
A smooth field F = hF1 , F2 , F3 i on a simply connected domain
D R3 is a conservative field iff hold
2 F3 = 3 F2 ,
3 F1 = 1 F3 ,
1 F2 = 2 F1 .
Proof: Only ().

Since the vector field F is conservative, there exists a scalar field f
such that F = f . Then the equations above are satisfied, since
for i, j = 1, 2, 3 hold
Fi = i f
i Fj = i j f = j i f = j Fi .
(The statement () is more complicated to prove.)

Example
Show that the field F = h2xy , (x 2 z 2 ), 2yzi is conservative.
Solution: We need to show that the equations in the Theorem
above hold, that is
2 F3 = 3 F2 ,
3 F1 = 1 F3 ,
1 F2 = 2 F1 .
with x1 = x, x2 = y , and x3 = z. This is the case, since

1 F2 = 2x,
2 F3 = 2z,
3 F1 = 0,
2 F1 = 2x,
3 F2 = 2z,
1 F3 = 0.
C

Example
Find the potential function of the conservative field
F = h2xy , (x 2 z 2 ), 2yzi.
Solution: We know there exists a scalar function f solution of
F = f
x f = 2xy ,
y f = x 2 z 2 ,
Z
f =
2xy dx + g (y , z)
z f = 2yz.
f = x 2 y + g (y , z).
y f = x 2 + y g (y , z) = x 2 z 2 y g (y , z) = z 2 .
Z
g (y , z) = z 2 dy +h(z) = z 2 y +h(z) f = x 2 y z 2 y +h(z).
z f = 2zy +z h(z) = 2yz z h(z) = 0 f = (x 2 z 2 )y +c0 .
C

Notation: We call a differential form to the integrand in a line
integral for a smooth field F, that is,
F dr = hFx , Fy , Fz i hdx, dy , dzi = Fx dx + Fy dy + Fz dz.
Remark: A differential form is a quantity that can be integrated

along a path.
Definition
A differential form F dr = Fx dx + Fy dy + Fz dz is called exact iff
there exists a scalar function f such that
Fx dx + Fy dy + Fz dz = x f dx + y f dy + z f dz.
Remarks:
I
A differential form F dr is exact iff F = f .
An exact differential form is nothing else than another name

for a conservative field.

Example
Show that the differential form given below is exact, where
F dr = 2xy dx + (x 2 z 2 ) dy 2yz dz .
Solution: We need to do the same calculation we did above:
Writing F dr = F1 dx1 + F2 dx2 + F3 dx3 , show that
2 F3 = 3 F2 ,
3 F1 = 1 F3 ,
1 F2 = 2 F1 .
with x1 = x, x2 = y , and x3 = z. We showed that this is the case,

since
1 F2 = 2x,
2 F1 = 2x,
2 F3 = 2z,
3 F1 = 0,
3 F2 = 2z,
1 F3 = 0.
So, there exists f such that F dr = f dr.
Greens Theorem on a plane. (Sect. 16.4)
I
I
Review: Line integrals and flux integrals.

Greens Theorem on a plane.
I
I
Circulation-tangential form.
Flux-normal form.
Tangential and normal forms equivalence.
Review: The line integral of a vector field along a curve.

Definition
The line integral of a vector-valued function F : D Rn Rn ,
with n = 2, 3, along the curve r : [t0 , t1 ] R D R3 , with arc
length function s, is given by
Z t1
Z s1
F u ds =
F(t) r0 (t) dt,
t0
s0
r0
where u = 0 , and s0 = s(t0 ), s1 = s(t1 ).
|r |
Remark: Since F = hFx , Fy i and
Example
r(t) = hx(t), y (t)i, in components,

Z t1
F(t) r0 (t) dt
Z t1 t0

=
Fx (t)x 0 (t) + Fy (t)y 0 (t) dt.
y
{z=0}
u
x
t0
Review: The line integral of a vector field along a curve.

Example
Evaluate the line integral of F = hy , xi along the loop
r(t) = hcos(t), sin(t)i for t [0, 2].
Solution: Evaluate F along the curve: F(t) = h sin(t), cos(t)i.
Now compute the derivative vector r0 (t) = h sin(t), cos(t)i.
Then evaluate the line integral in components,
I
Z t1

F u ds =
Fx (t)x 0 (t) + Fy (t)y 0 (t) dt,
t0
Z
F u ds =
Z
F u ds =
C

( sin(t))( sin(t)) + cos(t) cos(t) dt,
2
2

sin2 (t) + cos2 (t) dt
F u ds = 2.
C
Review: The flux across a plane loop.

Definition
The flux of a vector field F : {z = 0} R3 {z = 0} R3 along
a closed plane loop r : [t0 , t1 ] R {z = 0} R3 is given by
I
F = F n ds,
C
where n is the unit outer normal vector to the curve inside the
plane {z = 0}.
Remark: Since F = hFx , Fy , 0i,

r(t) = hx(t), y (t), 0i, ds = |r0 (t)| dt, and
1
n = 0 hy 0 (t), x 0 (t), 0i, in components,
|r |
I
Z t1

F n ds =
Fx (t)y 0 (t) Fy (t)x 0 (t) dt.
Example
z
y
{z=0}
C
n
t0
Review: The flux across a plane loop.

Example
Evaluate the flux of F = hy , x, 0i along the loop
r(t) = hcos(t), sin(t), 0i for t [0, 2].
Solution: Evaluate F along the curve: F(t) = h sin(t), cos(t), 0i.
Now compute the derivative vector r0 (t) = h sin(t), cos(t), 0i.
Now compute the normal vector n(t) = hy 0 (t), x 0 (t), 0i, that is,
n(t) = hcos(t), sin(t), 0i. Evaluate the flux integral in components,
I
Z t1

F n ds =
Fx (t)y 0 (t) Fy (t)x 0 (t) dt,
t0
Z
F n ds =
2

sin(t) cos(t) cos(t)( sin(t)) dt,
Z
F u ds =
I
0 dt
F u ds = 0.
C
I
I

I
I
Flux-normal form.

Theorem (Circulation-tangential Iform)
F u ds of the field
The counterclockwise line integral

C
F = hFx , Fy i along a loop C enclosing a region R R2 and given

by the function r(t) = hx(t), y (t)i for t [t0 , t1 ] and with unit
tangent vector u, satisfies that
Z t1
ZZ

Fx (t) x 0 (t) + Fy (t) y 0 (t) dt =
x Fy y Fx dx dy .
t0
R
C
y
{z=0}
u
x
Equivalently,
I
ZZ

F u ds =
x Fy y Fx dx dy .
C

Example
Verify Greens Theorem tangential form for the field F = hy , xi
and the loop r(t) = hcos(t), sin(t)i for t [0, 2].
I
Solution: Recall: We found that
F u ds = 2.
C
ZZ

Now we compute the double integral I =
x Fy y Fx dx dy
R
and we verify that we get the same result, 2.

ZZ
ZZ
Z

I =
1 (1) dx dy = 2
dx dy = 2
R
r dr d
0
r 2 1

I = 2(2)
I = 2.

2 0
I
ZZ

We verified that
F u ds =
x Fy y Fx dx dy = 2.
C
I
I

I
I
Flux-normal form.

Theorem (Flux-normal form)
I
F n ds of the field
The counterclockwise flux integral

C
F = hFx , Fy i along a loop C enclosing a region R R2 and given

by the function r(t) = hx(t), y (t)i for t [t0 , t1 ] and with unit
normal vector n, satisfies that
Z t1
ZZ

Fx (t) y 0 (t) Fy (t) x 0 (t) dt =
x Fx + y Fy dx dy .
t0
R
y
Equivalently,
I
ZZ

F n ds =
x Fx + y Fy dx dy .
{z=0}
C
x

Example
Verify Greens Theorem normal form for the field F = hy , xi and
the loop r(t) = hcos(t), sin(t)i for t [0, 2].
I
Solution: Recall: We found that
F n ds = 0.
C
ZZ

x Fx + y Fy dx dy
R
and we verify that we get the same result, 0.

ZZ
ZZ

I =
x (y ) + y (x) dx dy =
0 dx dy = 0.
R
ZZ
F n ds =
We verified that
C

x Fx + y Fy dx dy = 0.

Example
Verify Greens Theorem normal form for the field F = h2x, 3y i
and the loop r(t) = ha cos(t), a sin(t)i for t [0, 2], a > 0.
Solution: We start with the line integral
I
Z t1

F n ds =
Fx (t)y 0 (t) Fy (t)x 0 (t) dt.
t0
It is simple to see that F(t) = h2a cos(t), 3a sin(t)i,

and also that r0 (t) = ha sin(t), a cos(t)i.
Z 2
I
2

2a cos2 (t) 3a2 sin2 (t) dt,
F n ds =
Therefore,
0
2

1
1
1 + cos(2t) 3a2 1 cos(2t) dt.
2
2
0
C
I
Z 2
F n ds = a2 .
Since
cos(2t) dt = 0, we conclude
F n ds =
2a2

Example
Verify Greens Theorem normal form for the field F = h2x, 3y i
and the loop r(t) = ha cos(t), a sin(t)i for t [0, 2], a > 0.
I
Solution: Recall:
F n ds = a2 .
C
ZZ

x Fx + y Fy dx dy .
R
ZZ
I =

x (2x) + y (3y ) dx dy =
ZZ
(2 3) dx dy .
R
r 2 a

r dr d = 2
I =
dx dy =
= a2 .
2 0
0
0
R
I
ZZ

Hence,
F n ds =
x Fx + y Fy dx dy = a2 .
C
ZZ

I
I
Flux-normal form.
I
I

Lemma
The Green Theorem in tangential form is equivalent to the Green
Theorem in normal form.
Proof: Greens Theorem in tangential form for F = hFx , Fy i says

t1

Fx (t) x (t) + Fy (t) y (t) dt =
ZZ
t0

x Fy y Fx dx dy .
= hFy , Fx i, that is, Fx = Fy and

Apply this Theorem for F
Fy = Fx . We obtain
Z t1
ZZ

0
0
Fy (t) x (t) + Fx (t) y (t) dt =
x Fx y (Fy ) dx dy ,
t0
Z
so,
t1
t0

Fx (t) y 0 (t) Fy (t) x 0 (t) dt =
ZZ

x Fx + y Fy dx dy ,
which is Greens Theorem in normal form. The converse

implication is proved in the same way.
Using Greens Theorem

Example
Use Greens Theorem to find the counterclockwise circulation of
the field F = h(y 2 x 2 ), (x 2 + y 2 )i along the curve C that is the
triangle bounded by y = 0, x = 3 and y = x.
I
ZZ

Solution: Recall:
F dr =
x Fy y Fx dx dy .
C
ZZ
F dr =
(2x 2y ) dx dy =
(2x 2y ) dy dx,
0
I
F dr =
Z 3h
0
Z 3
x x i

2x y y 2
dx =
2x 2 x 2 dx,
0
Z
F dr =
3Z x
x 3 3
2
x dx =
3 0
F dr = 9.
C
Review of Greens Theorem on a plane.
Sketch of the proof of Greens Theorem.
Divergence and curl of a function on a plane.
Area computed with a line integral.
Review: Greens Theorem on a plane.

Theorem
Given a field F = hFx , Fy i and a loop C enclosing a region R R2
described by the function r(t) = hx(t), y (t)i for t [t0 , t1 ], with
unit tangent vector u and exterior normal vector n, then holds:
I
I The counterclockwise line integral
F u ds satisfies:
C
t1
ZZ

Fx (t) x (t) + Fy (t) y (t) dt =
t0

x Fy y Fx dx dy .
I
I
F n ds satisfies:
The counterclockwise line integral

C
t1

Fx (t) y (t) Fy (t) x (t) dt =
ZZ
t0

x Fx + y Fy dx dy .
Review: Greens Theorem on a plane.

z
R
C
y
{z=0}
Circulation-tangential form:
I
ZZ

F u ds =
x Fy y Fx dx dy .
C
u
x
z
R
y
Flux-normal form:
I
ZZ

F n ds =
x Fx + y Fy dx dy .
C
{z=0}
C
x
Lemma
The Green Theorem in tangential form is equivalent to the Green
Theorem in normal form.

We want to prove that for every differentiable vector field
F = hFx , Fy i the Green Theorem in tangential form holds,
Z
ZZ

0
0
Fx (t) x (t) + Fy (t) y (t) dt =
x Fy y Fx dx dy .
C
We only consider a simple domain like the one in the pictures.

y
y1
y = g 1 (x)
x = h 0 (y)
C
x0
y0
y = g 0 (x)
x1
x = h 1(y)
Using the picture on the left we show that

Z
ZZ

Fx (t) x 0 (t) dt =
y Fx dx dy ;
C
and using the picture on the right we show that

Z
ZZ

0
Fy (t) y (t) dt =
x Fy dx dy .
C

Show that for Fx (t) = Fx (x(t), y (t)) holds
Z
ZZ

0
Fx (t) x (t) dt =
y Fx dx dy ;
y
y = g 1 (x)
y = g 0 (x)
x0
x1
The path C can be described by the curves r0 and r1 given by

r0 (t) = ht, g0 (t)i,
t [x0 , x1 ]
r1 (t) = h(x1 + x0 t), g1 (x1 + x0 t)i
t [x0 , x1 ].
Therefore,
r00 (t) = h1, g00 (t)i,
t [x0 , x1 ]
r01 (t) = h1, g10 (x1 + x0 t)i
t [x0 , x1 ].
Recall: Fx (t) = Fx (t, g0 (t)) on r0 ,

and Fx (t) = Fx ((x1 + x0 t), g1 (x1 + x0 t)) on r1 .

Z
x1
Fx (t)x (t) dt =
Z
Fx (t, g0 (t)) dt
x0
x1
Fx ((x1 + x0 t), g1 (x1 + x0 t)) dt

x0
Substitution in the second term: = x1 + x0 t, so d = dt.

Z x1
Fx ((x1 + x0 t), g1 (x1 + x0 t)) dt =
x0
x0
x1
Fx (, g1 ( )) (d ) =
Fx (, g1 ( )) d.
x0
x1
Fx (t)x 0 (t) dt =
x1

Fx (t, g0 (t)) Fx (t, g1 (t)) dt.
ZC
Zx0 x1 Z g1 (t)

We obtain:
Fx (t)x 0 (t) dt =
y Fx (t, y ) dy dt.
Therefore,
x0
g0 (t)

Z
Recall:
Fx (t)x 0 (t) dt =
x1
x0
g1 (t)

y Fx (t, y ) dy dt.
g0 (t)
This result is precisely what we wanted to prove:

Z
ZZ

Fx (t)x 0 (t) dt =
y Fx dy dx.
C
We just mention that the result

Z
ZZ

0
Fy (t) y (t) dt =
x Fy dx dy .
C
y
y1
x = h 0 (y)
is proven in a similar way using the

parametrization of the C given in the
picture.
y0
x = h 1(y)
x

Definition
The curl of a vector field F = hFx , Fy i in R2 is the scalar

curl F z = x Fy y Fx .
The divergence of a vector field F = hFx , Fy i in R2 is the scalar
div F = x Fx + y Fy .
Remark: Both forms of Greens Theorem can be written as:

ZZ
I
F u ds =
curl F

z
dx dy .
ZZ
F n ds =
div F dx dy .
R

Remark: What type of information about F is given in curl F z ?

Example: Suppose F is the velocity field of a viscous fluid and
F = hy , xi
curl F

z
= x Fy y Fx = 2.
If we place a small ball at (0, 0), the ball will

spin around
the z-axis with speed proportional
to curl F z .
If we place a small ball at everywhere in the
plane, the ball will spin around
the z-axis with
speed proportional to curl F z .
Remark: The curl of a field measures its rotation.

Remark: What type of information about F is given in div F?
Example: Suppose F is the velocity field of a gas and
F = hx, y i
div F = x Fx + y Fy = 2.
The field F represents the gas as is heated

with a heat source at (0, 0). The heated gas
expands in all directions, radially out form
(0, 0). The div F measures that expansion.
Remark: The divergence of a field measures its expansion.

Remarks:
Notice that for F = hx, y i we have curl F
Notice that for F = hy , xi we have div F = 0.
= 0.

Remark: Any of the two versions of Greens Theorem can be used
to compute areas using a line integral. For example:
ZZ
I

x Fx + y Fy dx dy =
Fx dy Fy dx)
R
If F is such that the left-hand side above has integrand 1, then

that integral is the area A(R) of the region R. Indeed:
ZZ
I
F = hx, 0i
dx dy = A(R) = x dy .
R
ZZ
F = h0, y i
I
y dx.
dx dy = A(R) =
R
ZZ
1
F = hx, y i
2
1
dx dy = A(R) =
2
R

x dy y dx .

Example
Use Greens Theorem to find the area of the region enclosed by the
ellipse r(t) = ha cos(t), b sin(t)i, with t [0, 2] and a, b positive.
I
Solution: We use: A(R) = x dy .
C
We need to compute r0 (t) = ha sin(t), b cos(t)i. Then,

2
Z
A(R) =
x(t) y (t) dt =
0
2
2
cos (t) dt = ab
A(R) = ab
0
Since
a cos(t) b cos(t) dt.

0

1
1 + cos(2t) dt.
2
cos(2t) dt = 0, we obtain A(R) =

0
A(R) = ab.
ab
2, that is,
2
C

Conservative Fields and Potential Functions. (Sect. 16.3) : Example

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Conservative Fields and Potential Functions. (Sect. 16.3) : Example

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Conservative Fields and Potential Functions. (Sect. 16.3) : Example

Uploaded by

Copyright:

Available Formats

Conservative fields and potential functions. (Sect. 16.

Review: Line integral of a vector field.

The line integral of conservative fields.

Finding the potential of a conservative field.

Comments on exact differential forms.

The line integral of a vector field along a curve.

Remark: An equivalent expression is:

Work done by a force on a particle.

is called the work done by the force on the particle.

A projectile of mass m moving on the

The movement takes place on a

W 6 0 in the first half of the