Coll Problems MATH2403
Coll Problems MATH2403
1 Line integrals
Problem 1
Evaluate the line integral
Z
F(x, y, z) · dr where F(x, y, z) = (5z 2 , 2x, x + 2y)
C
along the straight line joining the points O = (0, 0, 0) and A = (1, 1, 1). Is F
a conservative field?
Solution
The parameterization of the straight line joining O with A is
1
Integrating (1) we get f (x, y, z) = 5xz 2 + h(y, z). The unknown function h
can be determined by susbtituting f into (2). We get
∂h
= 2x
∂y
which integrated gives
h(y, z) = 2xy + g(z).
Finally, substituting f (x, y, z) = 5xz 2 + 2xy + g(z) into (3) we obtain the
following differential equation for g, namely
dg
= x + 2y − 10xz.
dz
Since g depends only on z, whereas the right hand side of the above equation
depends also on x and y we obtained a contradiction and therefore there does
not exist a scalar field f such that F = ∇f . Hence, the given vector field is
not conservative.
Problem 2
Find the line integral of the vector field F(x, y, z) = (y 2 , x, z) along the curve
given by z = y = ex from x = 0 to x = 1.
Solution
We can use x as a parameter to evaluate the integral. Then, the parametric
equation of the curve is r(x) = (x, ex , ex ) with x ∈ [0, 1]. We can rewrite the
line integral as
Z Z 1
dr
I= F(x, y, z) · dr = F(r(x)) · dx.
C 0 dx
Note that
dr
F(r(x)) · = (e2x , x, ex ) · (1, ex , ex ) = 2e2x + xex .
dx
Finally, we get Z 1
I= (2e2x + xex )dx = e2 .
0
2
Problem 3
Find the line integral Z
r × dr
C
where the curve C is the ellipse
x2 y 2
+ 2 =1
a2 b
taken in an anticlockwise direction. Assume a, b > 0. What do you notice
about the magnitude of the answer?
Solution
The given ellipse can be parameterized by setting
x = a cos ϑ, y = b sin ϑ, z = 0, t ∈ [0, 2π)
Then, C is described by the vector
r(ϑ) = (a cos ϑ, b sin ϑ, 0).
We can rewrite the line integral as
Z Z 2π
dr
I= r × dr = r(ϑ) × dϑ.
C 0 dϑ
Note that
dr
r(ϑ) × = (a cos ϑ, b sin ϑ, 0) × (−a sin ϑ, b cos ϑ, 0) = (0, 0, ab).
dϑ
Finally, we get Z 2π
I= (0, 0, ab)dϑ = (0, 0, 2πab).
0
We see that the integral only has a component in the z-direction and the
magnitude of the integral is 2πab which is twice the area of the ellipse.
2 Surface integrals
Problem 4
Evaluate the surface integral of the vector field F(x, y, z) = (xy, x, x+y) over
the surface S defined by z = 0 with x ∈ [0, 1], y ∈ [0, 2] and with normal n
directed in the positive z-direction.
3
Solution
Since the given surface is just a portion of the plane z = 0, the normal is
n = (0, 0, 1). Hence, F · n = x + y and
1 2 1 2 1
y2
Z Z Z Z Z Z
F · ndS = (x + y)dydx = xy + dx = (2x + 2)dx
S 0 0 0 2 0 0
= [x2 + 2x]10 = 3.
Note that the y-integral was carried out first. However, it is equally possible
to carry out the x-integral first, and this gives the same result.
Problem 5
Find the surface integral of F(x, y, z) = (x, y, z) over the surface of the unit
cube with x, y, z ∈ [0, 1] and normal vector pointing outward.
Solution
This surface integral has six parts, corresponding to the six faces of the cube.
On the surface x = 0 the normal is n = (−1, 0, 0) and there F = (0, y, z).
Since F · n = 0, there is no contribution to the integral from this surface. By
symmetry the same result holds for the surfaces y = 0 and z = 0. On the
surface x = 1, we have n = (1, 0, 0) and F · n = x = 1. Since this is constant,
the integral over this surface is just the value of the constant multiplied by
the area of the surface, which gives 1. The same result holds on the surfaces
y = 1 and z = 1, so the total value of the surface integral, adding the six
contributions is 3.
Problem 6
The surface S is defined to be that part of the plane z = 0 lying between the
curves y = x2 and x = y 2 . Find the surface integral of F(x, y, z) = (z, xy, x2 )
over S where the normal is n = (0, 0, 1).
Solution
The two curves y = x2 and x = y 2 intersect at (0, 0) and (1, 1). Doing the x
√
integral first, the limits are y 2 ≤ x ≤ y and 0 ≤ y ≤ 1. So we have
√
Z Z Z 1 Z y Z 1
2 1 3
F · ndS = x dxdy = (y 3/2 − y 6 )dy = .
S 0 y2 3 0 35
4
Problem 7
Let S be the part of the surface z = x + y 2 with z < 0 and x > −1. Find
the surface integral of F(x, y, z) = (2y + x, −1, 0) over S.
Solution
The surface can be parameterized in terms of x and y as follows
r(x, y) = (x, y, x + y 2 ).
Then,
rx × ry = (1, 0, 1) × (0, 1, 2y) = (−1, −2y, 1)
and
F · (rx × ry ) = −x.
In terms of x and y the region of integration is x + y 2 < 0 and x > −1, so
doing the x-integral first, −1 < x < −y 2 and −1 < y < 1 and the integral is
Z Z Z 1 Z −y2 Z 1 4
y 1
F · (rx × ry )dS = − xdxdy = − − dy = 1.
S −1 −1 −1 2 2
3 Volume integrals
Problem 8
Find the volume integral of the scalar field φ(x, y, z) = x2 + y 2 + z 2 over the
region V specified by x ∈ [0, 1], y ∈ [1, 2], and z ∈ [0, 3].
Solution
The required volume integral is
Z Z Z Z 1Z 2Z 3
φ(x, y, z)dV = (x2 + y 2 + z 2 )dzdydx,
V
Z0 1 Z1 2 0
= (3x2 + 3y 2 + 9)dydx,
Z0 1 1
= (3x2 + 16)dx = 17.
0
Problem 9
Find the volume of the section of the cylinder x2 + y 2 = 1 that lies between
the planes z = x + 1 and z = −x − 1.
5
Solution
Doing the integrals in the order z, y, x, the range
√ of z between
√ the two planes
is −x − 1 < z < x + 1 and y ranges from − 1 − x to 1 − x2 at a fixed
2
value of x. The limits on the outer x integral are −1 and 1. The volume is
√
Z Z Z Z 1 Z 1−x2 Z x+1
dV = √
dzdydx,
V −1 − 1−x2 −x−1
√
Z 1 Z 1−x2
= 2 √
(x + 1)dydx,
−1 − 1−x2
Z 1 √
= 4 (x + 1) 1 − x2 dx = 2π,
−1
Problem 10
A circular pond with radius 1 meter and a maximum depth of 1 meter has
the shape of a paraboloid with equation z = 1 − x2 − y 2 . What is the total
volume of the pond? How does this compare with the case where the pond
has the same radius and depth but has the shape of a hemisphere?
Solution
The edge of the pond is where z = 0 so x2 + y 2 = 1. The limits of the volume
integral are obtained as follows. Doing the z integral first, at a fixed value
of x and y, we get that z ranges from 0 to 1 − x2 − y 2 . The limits on x and
y are the same as in the previous exercise. Hence, the volume V is
√
Z Z Z Z 1 Z 1−x2 Z 1−x2 −y 2
dV = √
dzdydx,
V −1 − 1−x2 0
√
Z 1 Z 1−x2
= √
(1 − x2 − y 2 )dydx,
−1 − 1−x2
4
Z 1 √ π
= (1 − x2 ) 1 − x2 dx = ,
3 −1 2
6
4 Gradient
Problem 11
Let r = (x, y, z) be the position vector of an arbitrary point P in a Cartesian
coordinate systemp0xyz. By r we denote the distance of P from the origin,
that is r = |r| = x2 + y 2 + z 2 . Show that the gradient of the scalar field
φ(r) = r is
r
∇φ =
r
and interpret the result geometrically.
Solution
p
Since φ(r) = x2 + y 2 + z 2 we obtain
∂φ x ∂φ y ∂φ z
=p , =p , =p .
∂x x2 + y 2 + z 2 ∂y x2 + y 2 + z 2 ∂z x2 + y 2 + z 2
Hence,
!
x y z
∇φ = p ,p ,p ,
x2 + y 2 + z 2 x2 + y 2 + z 2 x2 + y 2 + z 2
1 r
= p (x, y, z) = .
x2 + y 2 + z 2 r
Problem 12
Find the angle between the surfaces of the sphere x2 + y 2 + z 2 = 2 and the
cylinder x2 + y 2 = 1 at a point where they intersect.
Solution
To find where the sphere and the cylinder intersect we subtitute x2 + y 2 = 1
into the equation of the given spehere. We obtain z = 1 so the points of
intersection are the two circles x2 + y 2 = 1, z = ±1. The angle between the
two surfaces is the angle between the normals to the surfaces. By taking the
gradients of the two surfaces, these normal vectors are n1 = (2x, 2y, 2z) and
7
n2 = (2x, 2y, 0). The angle ϑ between the normals is found using the dot
product n1 · n2 = |n1 ||n2 | cos ϑ. This gives
p p
4x2 + 4y 2 = 2 x2 + y 2 + z 2 2 x2 + y 2 cos ϑ.
2 2
√ have x + y = 1 and z = ±1
At a point of intersection we √ and the above
relation simplifies to 4 = 2 22 cos ϑ. Hence, cos ϑ = 1/ 2 and ϑ = π/4.
Note that this is the same for all points of intersection.
Problem 13
Let a = (a1 , a2 , a3 ) be a constant vector in R3 . Find the gradient of the
scalar field f (x, y, z) = a · r where r = (x, y, z) and interpret this result
geometrically.
Solution
We have
f (x, y, z) = a · r = a1 x + a2 y + a3 z.
Hence,
∇f = (a1 , a2 , a3 ) = a.
Geometrically, the level curves f = constant represent planes that are per-
pedicular to the vector a.
Solution
Note that ∇f = (2xy, x2 + 3y 2 − 1). The critical points are given by solving
the system xy = 0 and x2 + 3y 2 − 1 = 0. Clearly, xy = 0 either when x = 0
or y = 0. Hence, we obtain the following critical points
1
0, ± √ , (±1, 0),
3
8
where
1 2 1 2
f 0, − √ = √ , f 0, √ =− √ , f (±1, 0) = 0.
3 3 3 3 3 3
The Hessian matrix is
2y 2x
H= .
2x 6y
√ 2
√
At the
√ point (0, −1/ 3) we have detH = 4 > 0 and ∂ f (0, −1/ 3)/∂x2 =
−2/ √ 3 < 0 and we coclude that the given point√is a maximum. √ At the point
2 2
(0, 1/ 3) we have detH = 4 > 0 and ∂ f (0, 1/ 3)/∂x = 2/ 3 > 0 and we
conclude that the given point is a minimum. Finally, the points (±1, 0) are
saddle points since there detH = −4 < 0.
6 Conservative fields
Problem 15
Determine whether or not the vector field F(x, y, z) = (sin y, x, 0) is conser-
vative.
6.1 Solution
The given vector field is conservative if there exists a scalar function f such
that F = ∇f . We have
∂f
= sin y, (4)
∂x
∂f
= x, (5)
∂y
∂f
= 0. (6)
∂z
Integrating (4) we get f (x, y, z) = x sin y + h(y, z). The unknown function h
can be determined by susbtituting f into (2). We get
∂h ∂h
x cos y + = x =⇒ = x − x cos y.
∂y ∂y
This equation can never be satisfied since the left hand side is a function
depending only on y and z whereas the right hand side is a function of x and
y. Hence, the given vector field is not conservative.
9
Problem 16
Consider the vector field
y x
F(x, y) = 2 2
,− 2 .
x +y x + y2
Show that the line integral of F around the unit circle x2 +y 2 = 1 is non-zero.
Explain why this result does not contradict the Test for conservative fields.
Solution
If F can be written as ∇f for some scalar field f , then
∂f y
= 2 , (7)
∂x x + y2
∂f x
= − 2 . (8)
∂y x + y2
From (8) it follows that h must be taken to be zero and we conclude that
x
f (x, y) = arctan .
y
Now consider the line integral of F around the unit circle x2 + y 2 = 1, given
parametrically by x = cos ϑ, y = sin ϑ, and ϑ ∈ [0, 2π).
I Z 2π
F · dr = (sin ϑ, − cos ϑ) · (− sin ϑ, cos ϑ)dϑ,
C 0
Z 2π
= − dϑ = −2π.
0
In the Test for conservative fields the vector field must be defined on a simply
connected region. When we performed the above integral, we considered a
ragion containing the closed path C but we had to remove the origin because
there both the vector and scalar fields are not defined. Hence, we were
actually considering a region of the real plane that was not simply connected!
10
7 Gradient, curl and divergence
Problem 17
Find the gradient ∇φ and the Laplacian ∆φ for the scalar field φ(x, y, z) =
x2 + xy + yz 2 . Consider now the scalar field
√
k2 +`2 z
f (x, y, z) = sin (kx) sin (`y)e .
Solution
We have
1.
∇φ = (2x + y, x + z 2 , 2yz), ∆φ = 2 + 2y.
2.
√ √
k2 +`2 z
∇f = k cos (kx) sin (`y), ` sin (kx) cos (`y), k2 + `2 sin (kx) sin (`y) e
Problem 18
Find the unit normal to the surface xy 2 + 2yz = 4 at the point (−2, 2, 3).
Solution
The unit normal to the surface φ =constant is ∇φ/|∇φ|. We have
which has magnitude 6 and the unit normal is n = (2/3, −1/3, 2/3).
Problem 19
For f (x, y, z) = x2 + y 2 + z 2 + xy − 3x find ∇f and find the minimum value
of f .
11
Solution
We have ∇f = (2x + y − 3, 2y + x, 2z). Setting ∇f = 0 and solving the
corresponding set of equations we find the critical point (2, −1, 0). There,
f (2, −1, 0) = −3. Since f tends to +∞ as x, y, z → ±∞, the point we found
must be a minimum value.
Problem 20
Find the equation of the plane which is tangent to the surface x2 +y 2 −2z 3 = 0
at the point (1, 1, 1).
Solution
First find the normal to the surface. Let f (x, y, z) = x2 + y 2 − 2z 3 . We
have ∇f (1, 1, 1) = (2, 2, −6). Using the equation of a plane in the form
n · (r − r0 ) = 0 with r0 = (1, 1, 1) we obtain x + y − 3z = −2.
Problem 21
Find both the divergence and the curl of the vector fields u = (y, z, x) and
v = (xyz, z 2 , x − y).
Solution
1.
div(u) = 0, ∇ × u = (−1, −1, −1),
2.
div(v) = yz, ∇ × v = (−1 − 2z, xy − 1, −xz).
Problem 22
Show that both the divergence and the curl are linear operators, i.e.
∇ · (αu + βv) = α∇ · u + β∇ · v,
∇ × (αu + βv) = α∇ × u + β∇ × v,
where u(x, y, z) and v(x, y, z) are vector fields in R3 and α, β are real con-
stants.
12
Solution
These results follow directly from the linearity of the differential operator.
Problem 23
For which values, if any, of the constants α and β is the vector field
irrotational?
Solution
The vector field u is irrotational if ∇ × u = 0. We have
Problem 24
Show that
is irrotational.
3. Find the value of the line integral of u along the curve x = sin (πt)/2,
y = t2 − t, z = t4 with t ∈ [0, 1].
Solution
1. Compute ∇ × u and verify that it vanishes. Hence, u is irrotational.
13
3. Since u is irrotational, it is conservative and the line integral is just
the difference between the values of f at the endpoints. In the present
case the endpoints are (0, 0, 0) and (1, 0, 1) and the line integrals is
Z
u · dr = f (1, 0, 1) − f (0, 0, 0) = 1.
C
with z > 0 and n is the upward-pointing normal. Note that the surface S is
not closed.
Solution
0
To apply the divergence theorem a closed surface must be used. Let S be
the surface z = 0, x2 + y 2 < 1, forming the base of the hemisphere. The
0
divergence theorem can now be applied over the entire closed surface S ∪ S ,
giving Z Z Z Z Z Z Z
∇ · F dV = F · n dS + F · n dS.
V S S0
The surface integral over S can therefore be found by subtracting the surface
0
integral over S from the volume integral. For the given vector field we have
∇ · F = 1, so the volume integral is just the volume of the hemisphere, 2π/3.
0
The surface integral over S is
Z Z Z Z
F · n dS = − x2 dS
S0 S0
since n = (0, 0, −1). This integral can be evaluated using polar coordinates.
Then, dxdy = rdrdϑ and x = r cos ϑ with r ∈ [0, 1] and ϑ ∈ [0, 2π). Hence,
we get
Z Z Z 1 Z 2π
2
− x dS = − r3 cos2 ϑ dϑdr,
0
S 0
Z 10
= −π r3 dr,
0
π
=− .
4
14
The required integral is then
Z Z
2 π 11
F · n dS = π + = π.
S 3 4 12
Problem 26
Use the divergence theorem to evaluate the integral
{
u · n dS,
S
Solution
The surface integral is equal to the volume integral of ∇u, but ∇u = 0 and
therefore the value of the surface integral is zero. This means that the total
flux of the given vector field through the surface of the sphere is zero.
Problem 27
Verify the divergence theorem, by calculating both the volume integral and
the surface integral, for the vector field
F(x, y, z) = (y, x, z − x)
and the volume E given by the unit cube with x, y, z ∈ [0, 1].
Solution
Observe that ∇F = 1 and the volume integral is
y Z 1Z 1Z 1
∇F dV = dxdydz = 1.
E 0 0 0
The surface integral has six parts from the six faces of the cube. On the face
where x = 0, the normal is n = (−1, 0, 0) and hence F · n = −y. Similarly,
on the face where x = 1 and n = (1, 0, 0) we have F · n = y. So the surface
integrals from these two faces cancel. The same argument holds for the faces
y = 0 and y = 1. On z = 0 we have n = (0, 0, −1) and F · n = −z + x = x
15
since it has to be evaluated at z = 0. On z = 1 we find n = (0, 0, 1) and
there F · n = z − x = 1 − x. The integrals over these two surfaces then given
Z 1Z 1
(x + 1 − x) dxdy = 1.
0 0
Therefore, both the surface integral and the volume integral give the answer
1 so the divergence theorem is verified.
Problem 28
Two scalar fields f and g are related by the Poisson’ equation ∆f = g, where
∆ is the Laplace operator in three dimensions. Show that
y {
g dV = ∇f · n dS.
E S
Solution
This result follows directly from applying the divergence theorem to the
vector field ∇f since g = ∆f = div(∇f ).
Solution
First consider the line integral around the circle x2 + y 2 = 1, evaluated
parametrically using x = cos ϑ, y = sin ϑ and ϑ ∈ [0, 2π) with dr =
(− sin ϑ, cos ϑ, 0)dϑ. The value of the line integral is
Z Z 2π
F · dr = (2 cos ϑ − sin ϑ, − sin2 ϑ, 0) · (− sin ϑ, cos ϑ, 0) dϑ,
C
Z0 2π
= (−2 sin ϑ cos ϑ + sin2 ϑ − sin2 ϑ cos ϑ) dϑ,
0
Z 2π
1 − cos (2ϑ) 2
= − sin (2ϑ) + − sin ϑ cos ϑ dϑ,
0 2
= π,
16
since all terms except that with factor 1/2 give zero when integrated between
0 and 2π. Now to compute the surface integral we need (∇ × F) · n. The line
integral was taken in an anticlockwise sense in the xy-plane, so the right-
hand rule means that n points in the positive z direction, so (∇ × F) · n =
(∇ × F) · ez = 1. The value of the surface integral is then just the area of
the surface, which is π since the surface is a disk of radius 1. Thus, Stoke’s
theorem is verified since the surface integral and the line integral are both
equal to π.
Problem 30
Show that I
r · dr = 0, r = (x, y, z)
C
for any closed curve C.
Solution
Since ∇ × r = 0 the given vector field is conservative and therefore its line
integral along any closed path must vanish. Alternatively, one can apply
Stokes’ theorem and arrive at the same conclusion.
Problem 31
Use Stokes’ theorem to show that
Z Z
f ∇g · dr = − g∇f · dr
C C
Solution
Note that ∇ × ∇(f g) = 0 from an identity we derived in class. Then, Stokes’
theorem implies that
I {
∇(f g) · dr = (∇ × ∇(f g)) · n dS = 0.
C S
17
Problem 32
If u is irrotational, express the surface integral
{
(u × ∇f ) · n dS
S
Solution
Since u is irrotational, the given identity reduces to ∇ × (f u) = ∇f × u.
Using Stokes’ theorem we obtain
{ {
(u × ∇f ) · n dS = − (∇f × u) · n dS,
S S
{
= − (∇ × (f u)) · n dS,
IS
= − f u · dr.
C
Problem 33
The magnetic field B = B(t, x, y, z) in an eletcrically conducting fluid moving
with velocity v obeys the magnetic induction equation
∂B
= ∇ × (v × B).
∂t
Show that the total flux of magnetic field through a surface S enclosed by a
streamline of the flow (i.e. a closed curve which is everywhere parallel to v)
is independent of time. Suppose that the magnetic field and its first order
partial derivatives are continuous in some solid region containing the surface
S.
Solution
We show that the derivative of the flux of the magnetic field through the given
closed surface with respect to time is zero. Since the magnetic field and its
first order partial derivatives are continuous in some solid region containing
18
the surface S, we can use Leibniz rule to interchange the derivative with
respect to time with the integral, that is
d { { ∂B
B · n dS = · n dS.
dt ∂t
S S
Applying the magnetic induction equation and Stokes’ theorem we find that
the rate if change of the total magnetic flux through the surface S is
d { {
B · n dS = (∇ × (v × B)) · n dS,
dt
S
IS
= v × B · dr = 0
C
19