Professional Documents
Culture Documents
Andrew Rechnitzer
University of British Columbia
Elyse Yeager
University of British Columbia
This text is a merger of the CLP Vector Calculus textbook and problembook.
It is, at the time that we write this, still a work in progress; some bits and
pieces around the edges still need polish. Consequently we recommend to the
student that they still consult text webpage for links to the errata — especially
if they think there might be a typo or error. We also request that you send us
an email at clp@ugrad.math.ubc.ca
Additionally, if you are not a student at UBC and using these texts please
send us an email (again using the feedback button) — we’d love to hear from
you.
Joel Feldman, Andrew Rechnitzer and Elyse Yeager
v
Feedback about the text
The CLP-4 Vector Calculus text is still undergoing testing and changes. Be-
cause of this we request that if you find a problem or error in the text then:
1. Please check the errata list that can be found at the text webpage.
vi
Contents
Preface v
1 Curves 1
1.1 Derivatives, Velocity, Etc. . . . . . . . . . . . . . . . 5
1.2 Reparametrization . . . . . . . . . . . . . . . . . 20
1.3 Curvature . . . . . . . . . . . . . . . . . . . . 23
1.4 Curves in Three Dimensions . . . . . . . . . . . . . . 32
1.5 A Compendium of Curve Formula . . . . . . . . . . . . 45
1.6 Integrating Along a Curve . . . . . . . . . . . . . . 46
1.7 Sliding on a Curve . . . . . . . . . . . . . . . . . 49
1.8 Optional — Polar Coordinates . . . . . . . . . . . . . 55
1.9 Optional — Central Forces . . . . . . . . . . . . . . 62
1.10 Optional — Planetary Motion . . . . . . . . . . . . . 65
1.11 Optional — The Astroid . . . . . . . . . . . . . . . 67
1.12 Optional — Parametrizing Circles . . . . . . . . . . . . 69
2 Vector Fields 73
2.1 Definitions and First Examples . . . . . . . . . . . . . 73
2.2 Optional — Field Lines . . . . . . . . . . . . . . . 85
2.3 Conservative Vector Fields . . . . . . . . . . . . . . 92
2.4 Line Integrals . . . . . . . . . . . . . . . . . . . 103
2.5 Optional — The Pendulum . . . . . . . . . . . . . . 120
vii
CONTENTS viii
A Appendices 312
A.1 Trigonometry . . . . . . . . . . . . . . . . . . . 312
A.2 Powers and Logarithms. . . . . . . . . . . . . . . . 315
A.3 Table of Derivatives . . . . . . . . . . . . . . . . . 317
A.4 Table of Integrals . . . . . . . . . . . . . . . . . . 318
A.5 Table of Taylor Expansions . . . . . . . . . . . . . . 319
A.6 3d Coordinate Systems . . . . . . . . . . . . . . . . 320
A.7 ISO Coordinate System Notation . . . . . . . . . . . . 324
A.8 Conic Sections and Quadric Surfaces . . . . . . . . . . . 329
A.9 Review of Linear Ordinary Differential Equations . . . . . . 332
Curves
We are now going to study vector-valued functions of one real variable. That
is, we are going to study functions that assign to each real number t (typically
in some interval) a vector1 r(t). For example
r(t) = x(t), y(t), z(t)
r(0)
r(1)
r(2)
While in some applications t will indeed be “time”, it does not have to be.
It can be simply a parameter that is used to label the different points on the
curve that r(t) sweeps out. We then say that r(t) provides a parameterization
of the curve.
Example 1.0.1 Parametrization of x2 + y 2 = a2 . While we will often
use t as the parameter in a parametrized curve r(t), there is no need to call
it t. Sometimes it is natural to use a different name for the parameter. For
example, consider the circle x2 + y 2 = a2 . It is natural
to use the angle θ in
the sketch below to label the point a cos θ , a sin θ on the circle.
y
` ˘
a cos θ , a sin θ
x2 ` y 2 “ a2
θ
x
That is,
r(θ) = a cos θ , a sin θ 0 ≤ θ < 2π
2 2 2
is a parametrization of the circle x +y = a . Just looking at the figure above,
it is clear that, as θ runs from 0 to 2π, r(θ) traces out the full circle.
1 We are going to use boldface letters, like r, to designate vectors. When writing by hand,
1
CHAPTER 1. CURVES 2
However beware that just knowing that r(t) lies on a specified curve does
not guarantee that, as t varies, r(t) covers the entire curve. For example, as t
runs over the whole real line, π2 arctan(t) runs over the interval (−1, 1). For all
t, r !
2 4
arctan(t) , 1 − 2 arctan2 (t)
r(t) = x(t), y(t) = a
π π
is well-defined and obeys x(t)2 + y(t)2 = a2 . But this r(t) does not cover the
entire circle because y(t) is always positive.
Example 1.0.2 Parametrization of (x − h)2 + (y − k)2 = a2 . We can tweak
the parametrization of Example 1.0.1 to get a parametrization of the circle of
radius a that is centred on (h, k). One way to do so is to redraw the sketch of
Example 1.0.1 with the circle translated so that its centre is at (h, k).
y
` ˘
h ` a cos θ , k ` a sin θ
a a sin θ
θ
ph, kq
px ´ hq2 ` py ´ kq2 “ a2
x
We see from the sketch that
r(θ) = h + a cos θ , k + a sin θ 0 ≤ θ < 2π
x
13 y
13
• Setting cos t = a and sin t = a turns cos2 t + sin2 t = 1 into
2/3 2/3
x y
a2/3
+ a2/3
= 1.
So
r(t) = a cos t , b sin t 0 ≤ t < 2π
3 3
r(t) = a cos t , a sin t 0 ≤ t < 2π
CHAPTER 1. CURVES 3
2 2
give parametrizations of xa2 + yb2 = 1 and x2/3 + y 2/3 = a2/3 , respectively. To
see that running t from 0 to 2π runs r(t) once around the curve, look at the
figures below.
y y
t = π/2
y2
t “ π{2
x2
a2
` b2
“1 x2/3 + y 2/3 = a2/3
t“0 t=0
t“π x t=π x
t “ 3π{2
t = 3π/2
The curve x2/3 + y 2/3 = a2/3 is called an astroid. From its equation, we
would expect its sketch to look like a deformed circle. But it is probably not so
obvious that it would have the pointy bits of the right hand figure. We will not
explain here why they arise. The astroid is studied in some detail in Example
1.1.9. In particular, the above sketch is carefully developed there.
Example 1.0.4 Parametrization of ey = 1 + x2 . A very easy method that
can often create parametrizations for a curve is to use x or y as a parameter.
Because
we can solve ey = 1 + x2 for y as a function of x, namely y = ln 1 +
2
x , we can use x as the parameter simply by setting t = x. This gives the
parametrization
r(t) = t , ln(1 + t2 )
−∞<t<∞
2 2 2
Example 1.0.5 Parametrization of x + y = a , again. It is also quite
common that one can use either x or y to parametrize part of, but not all of,
a curve. A simple example is the circle x2 + y 2 = a2 . For each −a < x < a,
there are two points on the circle with that value of x. So one cannot use x
to parametrize the whole circle. Similarly, for each −a < y < a, there are two
points on the circle with that value of y. So one cannot use y to parametrize
the whole circle. On the other hand
p
r(t) = t , a2 − t2 −a < t < a
p
r(t) = t , − a2 − t2 −a < t < a
provide parametrizations of the top half and bottom half, respectively, of the
circle using x as the parameter, and
p
r(t) = a 2 − t2 , t −a < t < a
p
2
r(t) = − a − t , t 2 −a < t < a
provide parametrizations of the right half and left half, respectively, of the
circle using y as the parameter.
Example 1.0.6 Unparametrization of r(t) = (cos t, 7−t). In this example,
we will undo the parametrization r(t) = (cos t, 7 − t) and find the Cartesian
equation of the curve in question. We may rewrite the parametrization as
x = cos t
CHAPTER 1. CURVES 4
y =7−t
Note that we can eliminate the parameter t simply by using the second equation
to solve for t as a function of y. Namely t = 7 − y. Substituting this into the
first equation gives us the Cartesian equation
x = cos(7 − y)
Curves often arise as the intersection of two surfaces. For example, the
2 2
intersection of the ellipsoid x2 + y2 + z3 = 1 with the paraboloid z = x2 + 2y 2
is the blue curve in the figure below.
z “ x2 ` 2y 2
y2 z2
x2 ` 2
` 3
“1
One way to parametrize such curves is to choose one of the three coordinates
x, y, z as the parameter, and solve the two given equations for the remaining
two coordinates, as functions of the parameter. Here are two examples.
Example 1.0.7 The set of all (x, y, z) obeying
x3 − e3y =0
2 y
x −e +z =0
x3 = e3y
x2 + z = ey
as a system of two equations for the two unknowns x and z, with y being
treated as a given constant, rather than as an unknown. We can now solve the
first equation for x, substitute the result into the second equation, and finally
solve for z.
x3 = e3y =⇒ x = ey
x2 + z = ey =⇒ e2y + z = ey =⇒ z = ey − e2y
So
r(y) = ey , y , ey − e2y
y2 z2
x2 + + =1
2 3
x2 + 2y 2 = z
3 z2
− y2 + z + =1
2 3
or, completing the square,
3 1 3 2 7
− y2 + z+ =
2 3 2 4
If, for example, we are interested in points (x, y, z) on the curve with y ≥ 0,
this can be solved to give y as a function of z.
r
2 3 2 14
y= z+ −
9 2 12
differentiation interacts quite nicely with dot and cross products too. Here are
some examples.
Example 1.1.2 Let
gives
d
γ(t)b(t) = 2te−t −t2 e−t ı̂ı + 2te−3t −3t2 e−3t ̂ + 2te−5t −5t2 e−5t k̂
dt
= 2t e−t ı̂ı + e−3t ̂ + e−5t k̂ + t2 − e−t ı̂ı − 3e−3t ̂ − 5e−5t k̂
and
gives
d
a(t) · b(t) = 2te−t −t2 e−t + 4t3 e−3t −3t4 e−3t + 6t5 e−5t −5t6 e−5t
dt
= 2te−t + 4t3 e−3t + 6t5 e−5t + −t2 e−t −3t4 e−3t −5t6 e−5t
and
ı̂ı ̂ k̂
a(t) × b(t) = det t2 t4 t6
e−t e−3t e−5t
= ı̂ı t4 e−5t − t6 e−3t ) − ̂(t2 e−5t − t6 e−t ) + k̂(t2 e−3t − t4 e−t )
gives
d
a(t) × b(t)
dt
= ı̂ı 4t3 e−5t − 6t5 e−3t ) − ̂( 2te−5t − 6t5 e−t ) + k̂( 2te−3t − 4t3 e−t )
+ ı̂ı −5t4 e−5t +3t6 e−3t ) − ̂(−5t2 e−5t +t6 e−t ) + k̂(−3t2 e−3t +t4 e−t )
= 2tı̂ı + 4t3 ̂ + 6t5 k̂ × e−t ı̂ı + e−3t ̂ + e−5t k̂
CHAPTER 1. CURVES 7
and
d
a s(t) = 2(sin t) cos tı̂ı + 4(sin t)3 cos t ̂ + 6(sin t)5 cos t k̂
=⇒
dt
= 2(sin t)ı̂ı + 4(sin t)3̂ + 6(sin t)5 k̂ cos t
= a0 s(t) s0 (t)
Of course these examples extend to general (differentiable) a(t), b(t), γ(t)
and s(t) and give us (most of) the following theorem.
Theorem 1.1.3 Arithmetic of differentiation. Let
• a(t), b(t) be vector valued differentiable functions of t ∈ R that take values
in Rn and
• α, β ∈ R be constants and
r(t)
r(t + h) − r(t) ≈ r′ (t) h
r(t + h)
When h is very small this vector
• has the essentially the same direction as the tangent vector to the curve
at r(t) and
• has length being essentially the length of the part of the curve between
r(t) and r(t + h).
Taking the limit as h → 0 yields that
CHAPTER 1. CURVES 8
• r0 (t) is a tangent vector to the curve at r(t) that points in the direction
of increasing t and
• if s(t) is the length
of the part of the curve between r(0) and r(t), then
ds dr (t).
dt (t) = dt
This is worth stating formally.
Lemma 1.1.4 Let r(t) be a parametrized curve.
a Denote by T̂(t) the unit tangent vector to the curve at r(t) pointing in
the direction of increasing t. If r0 (t) 6= 0 then
r0 (t)
T̂(t) =
|r0 (t)|
b Denote by s(t) the length of the part of the curve between r(0) and
r(t).Then
ds sptq
dr
(t) = (t)
rp0q
dt dt
Z T
dr
s(T ) − s(T0 ) = (t) dt
rptq
T0 dt T̂ptq
c In particular, if the parameter happens to be arc length, i.e. if t = s, so
that ds
ds
= 1, then
dr
(s) = 1
ds T̂(s) = r0 (s)
t, then
ds
velocity at time t = v(t) = r0 (t) = x0 (t) , y 0 (t) , z 0 (t) = (t) T̂(t)
dt
ds
speed at time t =
p
(t) = |v(t)| = |r0 (t)| = x0 (t)2 + y 0 (t)2 + z 0 (t)2
dt
acceleration at time t = a(t) = r00 (t) = v0 (t) = x00 (t) , y 00 (t) , z 00 (t)
r0 (θ)
T̂(θ) = 0
= − sin θ , cos θ
|r (θ)|
ds 0
(θ) = r (θ) = a
dθ
Z Θ
0
s(Θ) − s(0) = r (θ) dθ = aΘ
0
As1 s(Θ) is the arc length of the part of the circle with 0 ≤ θ ≤ Θ, the
circumference of the whole circle is
s(2π) = 2πa
which is reassuring, since this formula has been known2 for thousands of years.
y T̂pθq
` ˘
a cos θ , a sin θ
x2 ` y 2 “ a2
θ
x
The formula s(Θ) − s(0) = aΘ also makes sense — the part of the circle
Θ
with 0 ≤ θ ≤ Θ is the fraction 2π of the whole circle, and so should have length
Θ
2π × 2πa. Also note that
r(θ) · T̂(θ) = a cos θ , a sin θ · − sin θ , cos θ = 0
so that the tangent to the circle at any point is perpendicular to the radius
vector of the circle at that point. This is another geometric fact that has been
known3 for thousands of years.
Example 1.1.7 Arc length of a helix. Consider the curve
where the standard basis vectors ı̂ı = (1, 0, 0), ̂ = (0, 1, 0) and k̂ = (0, 0, 1).
We’ll first sketch it, by observing that
• x(t) = 6 sin(2t) and y(t) = 6 cos(2t) obey x(t)2 + y(t)2 = 36 sin2 (2t) +
36 cos2 (2t) = 36. So all points of the curve lie on the cylinder x2 +y 2 = 36
and
• as t increases, x(t), y(t) runs clockwise around the circle x2 + y 2 = 36
and at the same time z(t) = 5t just increases linearly.
t“π
π
t“ 2
y
t“0
x
We have marked three points of the curve on the above sketch. The first
has t = 0 and is 0ı̂ı + 6̂ + 0k̂. The second has t = π2 and is 0ı̂ı − 6̂ + 5π
2 k̂, and
the third has t = π and is 0ı̂ı + 6̂ + 5π k̂. We’ll now use Lemma 1.1.4 to find
a unit tangent T̂(t) to the curve at r(t) and also the arclength of the part of
curve between t = 0 and t = π.
Example 1.1.8 Velocity and acceleration. Imagine that, at time t, a
particle is at
t t
r(t) = h + a cos 2π ı̂ı + k + a sin 2π ̂
T T
As |r(t) − hı̂ı − k ̂| = a, the particle is running around the circle of radius a
t t
centred on (h, k). When t increases by T , the argument, 2π T , of cos 2π T and
t
sin 2π T increases by exactly 2π and the particle runs exactly once around
the circle. In particular, it travels a distance 2πa. So it is moving at speed
2πa
T . According to Lemma 1.1.5, it has
0 2πa t 2πa t
velocity = r (t) = − sin 2π ı
ı̂ + cos 2π ̂
T T T T
ds 2πa
speed = (t) = |r0 (t)| =
dt T
2
4π 2 a
00 4π a t t
acceleration = r (t) = − 2 cos 2π ı̂ı − sin 2π ̂
T T T2 T
4π 2
= − 2 r(t) − hı̂ı − k ̂
T
Here are some observations.
• The velocity r0 (t) has dot product zero with r(t) − hı̂ı − k ̂, which is the
radius vector from the centre of the circle to the particle. So the velocity
CHAPTER 1. CURVES 11
• The speed given by Lemma 1.1.5 is exactly the speed we found above,
just before we started applying Lemma 1.1.5.
• The acceleration r00 (t) points in the direction opposite to the radius vec-
tor.
y r1 ptq
rptq
r2 ptq
ph, kq
x
Example 1.1.9 Perimeter of the astroid. In this example, we find the
perimeter of the astroid4
x2/3 + y 2/3 = a2/3
A geometric construction of this curve, as well as a derivation of its equation
is given in the optional section 1.11 later. We’ll start by finding a convenient
parametrization.
• To do so, notice that x2/3 + y 2/3 = a2/3 looks somewhat like the equation
of the circle x2 + y 2 = a2 .
• The standard parametrization of the circle, namely x = a cos t, y = a sin t
works because of the elementary trig identity cos2 t + sin2 t = 1.
• If we can arrange that x(t)2/3 = a2/3 cos2 t and y(t)2/3 = a2/3 sin2 t, then
the same elementary trig identity will give x(t)2/3 + y(t)2/3 = a2/3 , as
desired.
• But of course its easy to arrange that: just solve x(t)2/3 = a2/3 cos2 t
for x(t), namely x(t) = a cos3 t, and solve y(t)2/3 = a2/3 sin2 t for y(t),
namely y(t) = a sin3 t.
Our parametrization is
r(t) = a cos3 tı̂ı + a sin3 t ̂
By Lemma 1.1.4
r(t) = a cos3 tı̂ı + a sin3 t ̂
r0 (t) = −3a sin t cos2 tı̂ı + 3a sin2 t cos t ̂
ds p
(t) = r0 (t) = 9a2 sin2 t cos4 t + 9a2 sin4 t cos2 t
dt q
= 3a sin2 t cos2 t(cos2 t + sin2 t)
= 3a sin t cos t
r0 (t) sin t cos t
T̂(t) = = − cos tı̂ı + sin t ̂
|r0 (t))| | sin t cos t|
CHAPTER 1. CURVES 12
= sgn sin t cos t − cos tı̂ı + sin t ̂
Here sgn sin t cos t means “the sign of sin t cos t”, i.e +1 when sin t cos t > 0
and −1 when sin t cos t < 0. So
T̂(t)
( )
1 if sin t > 0, cos t > 0 or sin t < 0, cos t < 0
= − cos tı̂ı + sin t ̂
−1 if sin t > 0, cos t < 0 or sin t < 0, cos t > 0
( )
1 if 0 < t < π2 or π < t < 3π2
= π 3π
− cos tı̂ı + sin t ̂
−1 if 2 < t < π or 2 < t < 2π
and
lim T̂ (t) =lim sgn sin t cos t lim − cos tı̂ı + sin t ̂ = ̂
t→π/2− t→π/2− t→π/2−
lim T̂ (t) = lim sgn sin t cos t lim − cos tı̂ı + sin t ̂ = −̂
t→π/2+ t→π/2+ t→π/2+
and so on. This signals cusps in the curve at t = 0, i.e. at r(0) = aı̂ı,
and at t = π2 , i.e. at r( π2 ) = â, and so on. So while the astroid looks
somewhat like a circle, it has cusps at ±aı̂ı and ±â. Here is the sketch.
CHAPTER 1. CURVES 13
y
t = π/2
x2/3 + y 2/3 = a2/3
t=0
t=π x
t = 3π/2
• The astroid is invariant under reflections in the x-axis and in the y-axis.
That is, x2/3 + y 2/3 = a2/3 is invariant under x → −x and also under
y → −y. So to find the whole perimeter, it suffices to find the arc length
of the part of the astroid in the first quadrant, and then multiply by 4.
Z π/2 Z π/2
ds
perimeter = 4 dt = 4 3a sin t cos t dt
0 dt 0
Z π/2 h cos(2t) iπ/2
= 6a sin(2t) dt = 6a −
0 2 0
= 6a
0
Example 1.1.10 r (t) = 0. In the last example, we found that the astroid had
cusps at those points r(t) where the velocity r0 (t) vanished. In this example,
we will explore a little further what can happen when r0 (t) = 0.
Suppose that you are out for a walk and that your position at time t is r(t).
If at some time you have nonzero velocity, it is very hard for you to change your
direction of motion discontinuously6 . On the other hand, when r0 (t) = 0, you
are not moving at all and it is easy for you to turn and leave in any direction
you choose. You could reverse direction completely, or make a sharp left turn,
or not change direction at all. Here are examples of all of these. They all have
r0 (t) = 0. They are sketched below.
r1 ptq r2 ptq
r3 ptq
t“0 t“0 t“0
4 Astroid should not be confused with asteroid, though both words derive from the Greek
word for star.
5 Like a cross-walk sign.
6 For your velocity to jump discontinuously, your acceleration has to be infinite, which
x
By Lemma 1.1.4
so that
ds
(t) = r0 (t)
dt q
cos2 t−2t sin t cos t+t2 sin2 t + sin2 t+2t sin t cos t+t2 cos2 t +1
=
p
= 2 + t2
To evaluate the integral, we’ll use three techniques that√ you learned in your
first integral calculus course. First, motivated by the 2 + t2 , we’ll use the
trigonometric substitution
√ √
dt = 2 sec2 u du 2 + t2 = 2 1 + tan2 u = 2 sec2 u
t = 2 tan u
√ π
When t = 0, u = 0 and when t = 2, tan u = 1 so that u = 4 and
√ Z π/4 √ √ Z π/4
2
s( 2) − s(0) = 2 sec2 u 2 sec u du = 2 sec3 u du
0 0
CHAPTER 1. CURVES 15
You may have evaluated this integral in first year. There are several ways of
doing so. Perhaps the most straight forward, but also most tedious, method is
to rewrite the integral as
√ Z π/4
cos u
s( 2) − s(0) = 2 du
0 cos4 u
Ooof!
1.1.1 Exercises
Exercises — Stage 1
Questions 1.1.1.1 through 1.1.1.5 provide practice with curve parametriza-
tion. Being comfortable with the algebra and interpretation of these descrip-
tions are essential ingredients in working effectively with parametrizations.
1. Find the specified parametrization of the first quadrant part of the
circle x2 + y 2 = a2 .
a In terms of the y coordinate.
b In terms of the angle between the tangent line and the positive
x-axis.
c In terms of the arc length from (0, a).
2. Consider the following time-parametrized curve:
π
r(t) = cos t , (t − 5)2
4
√
List the three points (−1/ 2, 0), (1, 25), and (0, 25) in chronolog-
CHAPTER 1. CURVES 16
ical order.
3. At what points in the xy-plane does the curve (sin t, t2 ) cross itself?
What is the difference in t between the first time the curve crosses
through a point, and the last?
4.
y
x
a
P
θ
r(t + h)
r(t)
r(0)
Indicate on the graph s(t + h) − s(t) and r(t + h) − r(t). Are the
quantities scalars or vectors?
8. What is the relationship between velocity and speed in a vector-valued
function of time?
9. ∗. Let r(t) be a vector valued function. Let r0 , r00 , and r000 denote
dr d2 r d3 r
dt , dt2 and dt3 , respectively. Express
d
(r × r0 ) · r00
dt
in terms of r, r0 , r00 , and r000 . Select the correct answer.
a (r0 × r00 ) · r000
b (r0 × r00 ) · r + (r × r0 ) · r000
c (r × r0 ) · r000
d 0
e None of the above.
10. Show that, if the position and velocity vectors of a moving particle are
always perpendicular, then the path of the particle lies on a sphere.
Exercises — Stage 2
11. ∗. Find the speed of a particle with the given position function
√
r(t) = 5 2 tı̂ı + e5t ̂ − e−5t k̂
√
b |v(t)| = 10 + 5et + 5e−t
p
c |v(t)| = 10 + e10t + e−10t
d |v(t)| = 5 e5t + e−5t
e |v(t)| = 5 et + e−t
12. Find the velocity, speed and acceleration at time t of the particle
whose position is
13. ∗.
a Let
r(t) = t2 , 3, 13 t3
curve at time t.
22. ∗. Find the arc length of the curve r(t) = tm , tm , t3m/2 for 0 ≤
a ≤ t ≤ b, and where m > 0. Express your result in terms of m, a,
and b.
7 The particle traces out a cycloid — see Question 1.1.1.4
8 The indefinite integral involved is one of a class of integrals called elliptic integrals
because of their connections to arc lengths of ellipses. In general, elliptic integrals cannot
be expressed in terms of elementary functions. You can easily find discussions of elliptic
integrals using your favourite search engine.
CHAPTER 1. CURVES 19
23. Let C be the part of the curve of intersection of the parabolic cylinder
x = y 2 and the hyperbolic paraboloid 3z = 2xy with y ≥ 0.
a Write a vector parametric equation for C using x as the param-
eter.
b Find the length of the part of C between the origin and the
point (9, 3, 18).
c A particle moves along C in the direction for which x is increas-
ing. If the particle moves with constant speed 9, find its velocity
vector when it is at the point (1, 1, 23 ).
d Find the acceleration vector of the particle of part (c) when it
is at the point (1, 1, 32 ).
24. If a particle has constant mass m, position r, and is moving with
velocity v, then its angular momentum is L = m(r × v).
aparticle with mass m = 1 and position function r = (sin t, cos t, t),
For
find dL
dt .
Exercises — Stage 3
25. ∗. A particle moves along the curve C of intersection of the surfaces
z2 = 12y and 18x = yz in the upward direction. When the particle is
at (1, 3, 6) its velocity v and acceleration a are given by
z
a Write a vector parametric equation for C using u = 6 as a pa-
rameter.
b Find the length of C from (0, 0, 0) to (1, 3, 6).
c If u = u(t) is the parameter value for the particle’s position at
time t, find du
dt when the particle is at (1, 3, 6).
d2 u
d Find dt2 when the particle is at (1, 3, 6).
1
26. ∗. A particle of mass m = 1 has position r0 = 2 k̂ and velocity
2
v0 = π2 ı̂ı at time 0. It moves under a force
d2 r dr
2
= −g k̂ − α
dt dt
dr
where α is a positive constant. If r = r0 and dt = v0 at time t = 0,
find r(t).
1.2 Reparametrization
There are invariably many ways to parametrize a given curve. Kind of trivially,
one can always replace t by, for example, 3u. But there are also more substan-
tial ways to reparametrize curves. It often pays to tailor the parametrization
used to the application of interest. For example, we shall see in the next couple
of sections that many curve formulae simplify a lot when arc length is used as
the parameter.
Example 1.2.1 Here are three different parametrizations of the semi-circle
x2 + y 2 = r2 , y ≥ 0.
• The first uses the polar angle θ as the parameter. We have already seen,
in Example 1.0.1, the parametrization
y
` ˘
r cos θ , r sin θ
x2 ` y 2 “ r 2
r1 (θ) = r cos θ , r sin θ
θ 0≤θ≤π
x
• The second uses x as the parameter.
√ Just solving x2 +y 2 = r2 , y ≥ 0 for y
as a function of x, gives y(x) = r2 − x2 and so gives the parametrization
CHAPTER 1. CURVES 21
y
` ? ˘
x, r 2 ´ x2
x2 ` y 2 “ r 2
p
x
r2 (x) = x , r 2 − x2
−r ≤x≤r
x
• The third uses arc length from (r, 0) as the parameter. We have seen, in
Example 1.1.6, that the arc length from (r, 0) to r1 (θ) is just s = rθ. So
the point on the semicircle that is arc length s away from (r, 0) is
y
` ˘ s
r cos sr , r sin sr r3 (s) = r1
x2 ` y 2 “ r 2 r
s s
s = r cos , r sin
r r
x 0 ≤ s ≤ πr
We shall see that, for some purposes, it is convenient to use parametrization
by arc length. Here is a messier example in which we reparametrize a curve so
as to use the arc length as the parameter.
Example 1.2.2 We saw in Example 1.1.9, that, as t runs from 0 to π2 , r(t) =
a cos3 tı̂ı + a sin3 t ̂ runs from (a, 0) to (0, a) along the astroid x2/3 + y 2/3 =
a2/3 . Suppose that we want a new parametrization R(s) chosen so that, as s
runs from 0 to some appropriate value, R(s) runs from (a, 0) to (0, a) along
x2/3 + y 2/3 = a2/3 , with s being the arc length from (a, 0) to R(s) along
x2/3 + y 2/3 = a2/3 .
y
Rpsq
rptq
s
x
π ds 3a
We saw, in Example 1.1.9, that, for 0 ≤ t ≤ 2 , dt = 2 sin(2t) so that the
arclength from (a, 0) = r(0) to r(t) is
Z t
3a 3a
sin(2t0 ) dt0 =
s(t) = 1 − cos(2t)
0 2 4
π
which runs from 0, at t = 0, to 3a2 , at t = 2 . This is relatively clean and we can
invert s(t) to find t as a function of s. The value, T (s), of t that corresponds
to any given 0 ≤ s ≤ 3a 2 is determined by
3a 1 4s
s= 1 − cos 2T (s) ⇐⇒ T (s) = arccos 1 −
4 2 3a
and
R(s) = r T (s) = a cos3 T (s) ı̂ı + a sin3 T (s) ̂
We can simplify cos3 T (s) and sin3 T (s) by just using trig identities to con-
CHAPTER 1. CURVES 22
3a
vert the cos 2T (s) in s = 4 1−cos 2T (s) into cos T (s) ’s and sin T (s) ’s.
3a 3a
1 − 2 cos2 T (s) − 1
s= 1 − cos 2T (s) =
4 4
2s
⇐⇒ cos2 T (s) = 1 −
3a
3a 3a
1 − 1 − 2 sin2 T (s)
s= 1 − cos 2T (s) =
4 4
2s
⇐⇒ sin2 T (s) =
3a
Consequently the desired parametrization is
3/2 3/2
2s 2s 3a
R(s) = a 1 − ı̂ı + a ̂ 0≤s≤
3a 3a 2
1.2.1 Exercises
Exercises — Stage 1
Z t
1. A curve r(s) is parametrized in terms of arclength. What is |r0 (s)| ds
1
when t ≥ 1?
2. The function
√
s+1 s+1 3
r(s) = sin ı̂ı + cos ̂ + (s + 1)k̂
2 2 2
Exercises — Stage 2
4. ∗.
a Let
r(t) = (2 sin3 t, 2 cos3 t, 3 sin t cos t)
Find the unit tangent vector to this parametrized curve at t =
π/3, pointing in the direction of increasing t.
1 Since we specified the derivatives are nonzero, there’s no messiness about vectors being
a Find the arc length of the piece of this spiral which is contained
in the unit circle.
b Reparametrize the logarithmic spiral with respect to arc length,
measured from t = −∞.
Exercises — Stage 3
6. Define
1 arctan t
r(t) = √ ,√ , arctan t
1 + t2 1 + t−2
for 0 ≤ t. Reparametrize the function using z = arctan t, and
describe the curve it defines. What is the geometric interpretation of
the new parameter z?
7. Reparametrize the function r(t) = ( 12 t2 , 13 t3 ) in terms of arclength
from t = −1.
1.3 Curvature
So far, when we have wanted to approximate a complicated curve by a simple
curve near some point, we drew the tangent line to the curve at the point.
That’s pretty crude. In particular tangent lines are straight — they don’t
curve. We will get a much better idea of what the complicated curve looks like
if we approximate it, locally, by a very simple “curvy curve” rather than by a
straight line. Probably the simplest “curvy curve” is a circle1 and that’s what
we’ll use.
Definition 1.3.1
a The circle which best approximates a given curve near a given point is
called the circle of curvature or the osculating circle2 at the point.
b The radius of the circle of curvature is called the radius of curvature
at the point and is normally denoted ρ.
c The curvature at the point is κ = ρ1 .
d The centre of the circle of curvature is called centre of curvature at
the point.
♦
These definitions are illustrated in the figure below. It shows (part of) the
osculating circle at the point P . The point C is the centre of curvature.
1 Circles are good for studying “curvature”, because, unlike parabolas for example, the
C
Note that when the curvature κ is large, the radius of curvature ρ is small
and we have a very curvy curve. On the other hand when the curvature κ is
small, the radius of curvature ρ is large and our curve is almost straight. In
particular, straight lines have curvature exactly zero.
We are now going to determine how to find the circle of curvature, starting
by figuring out what its radius should be. We’ll first look at curves3 that lie
in the xy-plane and then move on to curves in 3d. Consider the black curve in
the figure below.
ds
ρ dθ
That figure also contains a (portion of a) red circle that fits the curve really
well between the two radial lines that are (a very small) angle dθ apart. So the
arclength ds of the part of the black curve between the two radial lines, should
be (essentially) the same as the arc length of the circle between the two radial
lines, which is ρ |dθ|, where ρ is the radius of the circle. (We put in absolute
values to take into account the possibility that dθ could be negative.) Thus
ds = ρ |dθ|. When dθ is a macroscopic angle, this is of course an approximation.
But in the limit as dθ → 0, we should end up with
ds
ρ =
dθ
We now have a formula for the radius of curvature, but not in a very convenient
form, because to evaluate it we would need to know the arc length along the
curve as a function of the angle θ in the rightmost figure below. We’ll now
spend some time developing more convenient formulae for ρ. First consider the
three figures below. They all show the same curve as in the last figure. The
leftmost figure just shows
• the curve of interest, which is the black curve, and
• the (blue) point of interest on the black curve. We want to find the
curvature at that point.
The middle figure shows the same curve and point of interest and also shows
• the red circle of curvature (i.e. best fitting circle) for the black curve at
the blue dot.
• The red dot is the centre of curvature.
3 We’llalso assume that the curves of interest are smooth, with no cusps for example, and
not straight, so that the radius of curvature 0 < ρ < ∞.
CHAPTER 1. CURVES 25
The rightmost figure shows the same black curve, blue point of interest and
red circle of curvature (at least part of it) somewhat enlarged.
• The angle θ is the angle between ı̂ı and the radius vector from the red
dot (the centre of curvature) to the blue dot (the point of interest).
• T̂ is the tangent vector to the black curve at the blue dot.
• The angle φ is the angle between ı̂ı and T̂. The vector T̂ is also tangent
to the red circle. As the tangent and radius vectors for circles are per-
pendicular to each other4 , we have that φ = θ + π2 and hence ρ = dφ ds
too.
θ
φ ı̂ı
T̂
ı̂ı
We are now in a position to develop a bunch of formulae for the radius
of curvature ρ and the curvature κ = ρ1 , that are more convenient than κ =
ds −1
. These formulae will use the
dφ
• T̂(t) is the unit tangent vector to the curve at r(t) that points in the
direction of increasing t.
• N̂(t) is the unit normal vector to the curve at r(t) that points toward the
centre of curvature.
• κ(t) is the curvature at r(t)
• ρ(t) is the radius of curvature at r(t)
♦
Theorem 1.3.3
a Given5 s(φ), i.e. if we know the arc length along the curve as a function
of the angle6 φ = ](ı̂ı, T̂), then
−1
ds ds dφ
ρ = κ = κ =
dφ dφ ds
dx d2 y dy d2 x
2 − dt dt2
v(t) × a(t)
κ= dt dt
=
ds
3 2 3/2
dx 2
+ dy
dt dt dt
dT̂
dφ T̂
dT̂
dφ T̂
φ
φ
dφ
cos φ ą 0, ds
ą0
dφ
cos φ ą 0, ds
ă0
φ
φ
dT̂
dT̂
T̂ dφ
T̂ dφ
dφ
cos φ ă 0, ds
ă0 cos φ ă 0, dφ
ą0
ds
consider the figure on the lower left. In that figure,
dT̂
so dφ = −N̂ (the centre of curvature is the red dot above the curve) and
• as s increases (i.e. as you move in the direction of the arrow on the curve),
φ decreases (on the far right hand part of the curve φ ≈ 3π 2 , while on the
far left hand part of the curve φ ≈ π), so ds < 0 and κ = ds = − dφ
dφ dφ
ds .
dT̂ dT̂ dφ
• So by (∗), ds = dφ ds = − N̂)(−κ) = κN̂.
T̂
In each of the three other figures we also end up with dds = κ(s)N̂(s).
Note that if κ(s) = 0, then N̂(s) is not defined. This makes sense: if the
curve is (locally) a straight line, there is no “best fitting circle”.
(c) Given r(t), i.e. if we have a general parametrized curve, we can deter-
mine a unit tangent vector by using Lemma 1.1.4:
dr ds r0 (t)
v(t) = (t) = (t) T̂(t) =⇒ T̂(t) =
dt dt |r0 (t)|
Then we can determine κ and N̂ by differentiating T̂(t) and using the chain
rule:
dT̂ dT̂ ds ds |T̂0 (t)|
= = κ N̂ =⇒ κ(t) = 0
dt ds dt dt |r (t)|
ds
Also, if we differentiate v(t) = dt T̂(t), we get that the acceleration
d2 r d2 s ds dT̂ d2 s ds 2
a(t) = = T̂ + = T̂ + κ N̂
dt2 dt2 dt dt dt2 dt
CHAPTER 1. CURVES 28
(d) Given x(t) , y(t) , (for curves in the xy-plane), we can read off the
curvature from
ds d2 s ds 2
v(t) × a(t) = (t) T̂(t) × 2
T̂ + κ N̂
dt dt dt
ds 3
=κ T̂ × N̂ (since T̂ × T̂ = 0
dt
(e) Given y(x), again for curves in the xy-plane, we can parametrize the
curve using x as the parameter:
r(t) = X(t) , Y (t) with X(t) = t and Y (t) = y(t)
Then
dX d2 X dY dy d2 Y d2 y
=1 =0 = =
dt dt2 dt dx dt2 dx2
and dX d2 Y d2 y
dY d2 X
dt dt2 − dt dt2
2
dx
κ= 3/2 = 3/2
dX 2 dY 2 dy 2
dt + dt 1 + dx
a Think about why this should be the case. In particular, sketch T̂ and φ and think about
T̂
what the sketch says about ddφ .
b In each of the four figures, the arrow on the curve specifies the direction of increasing
arc length s and the red dot is the centre of curvature for the curve at the blue dot.
Take another look at Theorem 1.3.3 and note that
d2 s
• the tangential component of acceleration, i.e. dt2 , arises purely from
change in speed while
2
• the normal component of acceleration, i.e. κ dsdt , arises from curvature
and is proportional to the square of the speed ds
dt Think about what you
.
feel when you are driving. That’s why velodromes and (car) race tracks
often have banked corners.
Example 1.3.4 As a warm up example, and also a check that our formulae
make sense, we’ll find the curvature κ, radius of curvature, ρ, unit tangent
vector, T̂, unit normal vector, N̂, and centre of curvature of the parametrized
curve
r(t) = a cos tı̂ı + a sin t ̂
with the constant a > 0. This is, of course, the circle of radius a centred on
CHAPTER 1. CURVES 29
the origin. As
dr ds
v(t) = (t) = −a sin tı̂ı + a cos t ̂ =⇒ (t) = |v(t)| = a
dt dt
we have that the unit tangent vector
v(t)
T(t) = = − sin tı̂ı + cos t ̂
|v(t)|
Note, as a check, that this is indeed a vector of length one and is perpendicular
to the radius vector (as expected — the curve is a circle). As
dT̂
(t) = − cos tı̂ı − sin t ̂
dt
we have that
dT̂
(t)
N̂(t) = dt = − cos tı̂ı − sin t ̂
dT̂ (t)
dt
dT̂
(t) 1
dt
κ(t) = ds =
dt (t) a
1
ρ(t) = =a
κ(t)
N̂
To get to the centre of curvature we should start from r(t) and walk a
distance ρ(t), which after all is the radius of curvature, in the direction N̂(T ),
which is pointing towards the centre of curvature. So the centre of curvature
is
r(t) + ρ(t)N̂(t) = a cos tı̂ı + a sin t ̂ + a − cos tı̂ı − sin t ̂ = 0
This makes perfectly good sense — the radius of curvature is the radius of the
original circle and the centre of curvature is the centre of the original circle.
One alternative calculation of the curvature, using x(t) = a cos t, y(t) =
a sin t, is
dx d2 y
(t) 2 (t) − dy (t) d2 2x (t)
dt dt dt dt
κ(t) = h
dx
2 dy
2 i3/2
dt (t) + dt (t)
− a sin t − a sin t − a cos t − a cos t
= 2 2 3/2
− a sin t + a cos t
1
=
a
CHAPTER 1. CURVES 30
√
Another alternative calculation of the curvature, using y(x) = a2 − x2 (for
the part of the circle with y > 0),
x x
y 0 (x) = − √ =−
a2 − x2 y(x)
y(x) − xy 0 (x) y(x)2 + x2 a2
y 00 (x) = − 2
=− 3
=−
y(x) y(x) y(x)3
is
d2 y a2
2 (x)
dx y(x)3 a2
κ(x) = h i 3/2
=h i3/2
= 3/2
2
dy
1 + dx (x) x2
1 + y(x) 2
y(x)2 + x2
1
=
a
Example 1.3.5 As a more computationally involved example, we’ll analyze
r(t) = cos t + t sin t ı̂ı + sin t − t cos t ̂ t>0
v(t) = t cos tı̂ı + t sin t ̂
a(t) = cos t − t sin t ı̂ı + sin t + t cos t ̂
ds
(t) = |v(t)| = t
dt
d2 s
(t) = 1
dt2
v(t)
T(t) = = cos tı̂ı + sin t ̂
|v(t)|
so that t2 κ(t) is the length of −t sin tı̂ı + t cos t ̂, which is t. Thus
|v(t) × a(t)|
κ(t) =
( ds (t))3
dt
t cos tı̂ı + t sin t ̂ × cos t − t sin t ı̂ı + sin t + t cos t ̂
=
( ds
dt (t))
3
CHAPTER 1. CURVES 31
t cos t sin t + t cos t − t sin t cos t − t sin t k̂
=
( ds
dt (t))
3
|t2 k̂| 1
= 3
=
t t
It pays to think before you calculate!
1.3.1 Exercises
Exercises — Stage 1
There are a lot of constants in this chapter that might be new to you.
They can take a little getting used to. Questions 1.3.1.1-1.3.1.5 provide practice
working with and interpreting these constants and their relations to each other.
1. Sketch the curve r(t) = (3 sin t, 3 cos t). At the point (0, 3), label T̂
and N̂. Give the values of κ and ρ at this point as well.
2. Consider the circle r(t) = (3 sin t, 3 cos t). Find T̂(t) and T̂(s). Then,
use parts (b) and (c) of Theorem 1.3.3 to find N̂(t) and N̂(s).
3. The functon r(t) = (t cos t, t sin t), t ≥ 0, defines a spiral centred at
the origin. Using only geometric intuition (no calculation), predict
lim κ(t).
t→∞
ds
4. Let r(t) = (et , 3t, sin t). What is dt ?
5. In Question 1.2.1.5 of Section 1.2, we found that the spiral
Exercises — Stage 2
7. Given a curve r(t) = (et , t2 + t), compute the following quantities:
A v(t)
B a(t)
CHAPTER 1. CURVES 32
ds
C
dt
D T̂(t)
E κ(t)
8. Find the curvature κ(t) of r(t) = (cos t + sin t, sin t − cos t).
9. Find the minimum and maximum values for the curvature of the el-
lipse x(t) = a cos t, y(t) = b sin t. Here a > b > 0.
10. ∗.
a Find the curvature of y = ex at (0, 1).
b Find the equation of the circle best fitting y = ex at (0, 1).
11. ∗. Consider the motion of a thumbtack stuck in the tread of a tire
which is on a bicycle moving at constant speed. This motion is given
by the parametrized curve
r(t) = t − sin t , 1 − cos t
with t > 0.
a Sketch the curve in the xy-plane for 0 < t < 4π.
b Find and simplify the formula for the curvature κ(t).
Exercises — Stage 3
12. Find the curvature κ as a function of arclength s (measured from
(0, 0)) for the curve
Z θ Z θ
1 2 1 2
x(θ) = cos 2 πt dt y(θ) = sin 2 πt dt
0 0
3
13. ∗. Let C be the curve in R2 given by the graph of the function y = x3 .
Let κ(x) be the curvature of C at the point (x, x3 /3). Find all points
where κ(x) attains its maximal values, or else explain why such points
do not exist. What are the limits of κ(x) as x → ∞ and x → −∞?
osculating circle to the curve at r(t) to be the circle in that plane that fits the
curve best near r(t). And we still have the formulae1
dr ds
v= = T̂
dt dt
dT̂
= κN̂
ds
dT̂ ds
= κ N̂
dt dt
d2 r d2 s ds 2
a = 2 = 2 T̂ + κ N̂
dt dt dt
ds 3
v×a =κ T̂ × N̂
dt
The only 2 difference is that v, a, T̂ and N̂ are now three component vectors
rather than two component vectors.
If we are lucky and our curve happens to lie completely in a single plane,
the vectors T̂(s) and N̂(s) are mutually perpendicular unit vectors that lie in
the same plane, so that their cross product B̂(s) = T̂(s) × N̂(s) is a unit vector
that is perpendicular to the plane. By continuity, B̂(s) has to be a constant
vector, i.e. be independent of s.
If, on the other hand, B̂(s) is not constant, then our curve doesn’t lie in a
single plane, and we can use the derivative
d h i dB̂
|B̂| = 1 =⇒ 1 = B̂ · B̂ =⇒ 0 = B̂ · B̂ = 2B̂ ·
ds ds
dB̂
So ds (s) must be parallel to N̂(s).
Definition 1.4.1
a The binormal vector at r(s) is B̂(s) = T̂(s) × N̂(s). The normal
vector N̂(s) is sometimes called the unit principal normal vector to
distinguish it from the binormal vector.
b We define the torsion τ (s) by
dB̂
(s) = −τ (s)N̂(s)
ds
1 The arguments in the proof of Theorem 1.3.3 that we used to verify these formulae work
in any plane, not just the xy-plane. Just choose ı̂ı and ̂ to be any two mutually perpendicular
unit vectors in the plane.
2 However this can be a significant difference.
CHAPTER 1. CURVES 34
The negative sign is included so that τ (s) > 0 indicates “right handed
twisting”. There will be an explanation of what this means in Example
1.4.4 below.
c The osculating plane at r(s) (the plane that fits the curve best at r(s))
is the plane through r(s) with normal vector B̂(s). The equation of the
plane is
B̂(s) · (x, y, z) − r(s) = 0
♦
For each s, T̂(s), N̂(s) and B̂(s) are mutually perpendicular unit vectors.
They form an orthonormal basis for R3 , just as ı̂ı, ̂ and k̂ form an orthonormal
basis for R3 . Furthermore both (T̂(s) , N̂(s) , B̂(s)) and (ı̂ı , ̂ , k̂) are “right
handed triples”3 , meaning that B̂(s) = T̂(s) × N̂(s) and k̂ = ı̂ı × ̂.
k̂ B̂
̂
N̂
ı̂ı T̂
dT̂ dB̂
We have already computed ds and ds . It is now an easy matter to compute
dN̂ d
= B̂(s) × T̂(s)
ds ds
= −τ (s)N̂(s) × T̂(s) + B̂(s) × κ(s)N̂(s)
= τ (s)B̂(s) − κ(s)T̂(s)
To see that N̂(s) × T̂(s) = −B̂(s) and B̂(s) × N̂(s) = −T̂(s), just look at the
right hand figure above.
Now suppose that we have a curve that is parametrized by t rather than s.
How do we find the torsion τ ? The most obvious method is to
3 3
• recall that v × a = κ dsdt T̂ × N̂ = κ dsdt B̂ and that B̂(t) is a unit
vector. So
v(t) × a(t)
B̂(t) =
|v(t) × a(t)|
dB̂
• Having found B(t) we can differentiate it and use ds (s) = −τ (s)N̂(s)
and the chain rule to give
dB dB ds ds
= = −τ N̂
dt ds dt dt
ds
from which we can read off τ , provided we know dt and N̂.
There is another, often more efficient, method to find the torsion τ that
uses
da d d2 s ds 2
= T̂ + κ N̂
dt dt dt2 dt
d3 s d2 s ds d ds 2 ds 3
= 3 T̂ + 2 κN̂ + κ N̂ + κ τ B̂ − κT̂
dt dt dt dt dt dt
3 We shall stick to “right handed triples” to make it easier to get various signs right.
CHAPTER 1. CURVES 35
While this looks a little complicated, notice that, with just one exception,
3
namely κ dsdt τ (s)B̂(s), every term on the right hand side is either in the
direction T̂ or in the direction N̂ and so is perpendicular to B̂. So, dotting
3
with v × a = κ ds dt B̂ gives
da ds 6
v×a · = κ2 τ = |v × a|2 τ
dt dt
and hence
v × a · da
dt
τ=
|v × a|2
If the curvature4 κ(s) > 0 and the torsion τ (s) are known, then the system
of equations5
Equation 1.4.2 Frenet–Serret Formulae.
dT̂
(s) = κ(s) N̂(s)
ds
dN̂
(s) = τ (s) B̂(s) − κ(s) T̂(s)
ds
dB̂
(s) = −τ (s) N̂(s)
ds
d d
x(s) = M (s)x(s) ⇐⇒ e−M(s) x(s) − M (s)e−M(s) x(s) = 0
ds ds
d −M(s)
⇐⇒ e x(s) = 0
ds
by the product rule. So e−M(s) x(s) is a constant independent of s. In partic-
ular e−M(s) x(s) = e−M(0) x(0) = x0 so that x(s) = x0 eM(s) . This argument
4 As in two dimensions, if κ(s) = 0, then N̂(s) is not defined. This makes even more sense
in three dimensions than in two dimensions: if the curve is a straight line, there are infinitely
many unit vectors perpendicular to it and there is no way to distinguish between them.
5 The equations are named after the two French mathematicians who independently dis-
covered them: Jean Frédéric Frenet (1816–1900, the son of a wig maker), in his thesis of 1847
(actually he only gave two of the three equations), and Joseph Alfred Serret (1819–1885) in
1851.
CHAPTER 1. CURVES 36
can be generalized to any natural number n. But that is beyond the scope of
this book.
Since the Frenet-Serret formulae constitute a first order system of ordinary
differential equations for the vector (T̂(s) , N̂(s) , B̂(s)) and since any first
order linear initial value problem has a exactly one solution,
• the vector valued function (T̂(s) , N̂(s) , B̂(s)) is determined by the func-
tions κ(s) and τ (s) (assuming that they are continuous) together with
the initial condition (T̂(0) , N̂(0) , B̂(0)).
• Furthermore, once you know T̂(s), then r(s) is determined by r(0) and
dr
ds (s) = T̂(s).
• So any smooth curve r(s) is completely determined by r(0), (T̂(0) , N̂(0) , B̂(0)),
κ(s) and τ (s).
• That is, up to translations (you can move between any two possible
choices of r(0) by a translation) and rotations (you can move between
any two possible choices of (T̂(0) , N̂(0) , B̂(0)) by a rotation) a curve is
completely determined by the curvature κ(s) > 0 and the torsion τ (s).
This result is called “The fundamental theorem of space curves”.
Theorem 1.4.3 The Fundamental Theorem of Space Curves. Let
κ(s) > 0 and τ (s) be continuous. Then up to translations and rotations, there
is a unique curve with curvature κ(s) and torsion τ (s).
Example 1.4.4 Right circular helix. The right circular helix is the curve
t“ 5π
2
t“ 3π
2
π
t“ 2
y
t“0
x
Here is why it is called a right helix rather than a left helix. If the helix
is the thread of a bolt that you are screwing into a nut, and you turn the bolt
in the direction of the (curled) fingers of your right hand (as in the figure6 on
the right above), then it moves in the direction of your thumb (as in the long
straight arrow of the figure on the right above).
To determine the curvature and torsion of this curve we compute
ds p 2
= a + b2
dt
a a b
T̂(t) = − √ sin tı̂ı + √ cos t ̂ + √ k̂
2
a +b 2 2
a +b 2 a + b2
2
2 2
From a = ddt2s T̂ + κ ds
dt N̂ = κ(a2 + b2 )N̂, we read off that
a
κ(t) = N̂(t) = − cos tı̂ı − sin t ̂
a2 + b2
From
ı̂ı ̂ k̂
v(t) × a(t) = det −a sin t a cos t b = ab sin tı̂ı − ab cos t ̂ + a2 k̂
−a cos t −a sin t 0
|v(t) × a(t)|2 = a2 b2 + a4 = a2 (a2 + b2 )
we read off
v(t) × a(t) b b a
B̂(t) = =√ sin tı̂ı − √ cos t ̂ + √ k̂
|v(t) × a(t)| a2 + b2 a2 + b2 a2 + b2
and
v × a · da
dt a2 b b
τ (t) = 2
= 2 2 2
= 2
|v × a| a (a + b ) a + b2
Note that, for the right handed helix, τ > 0. Finally the centre of curvature is
1 a2 + b2 a2 + b2
r(t) + N̂(t) = a − cos tı̂ı + a − sin t ̂ + bt k̂
κ(t) a a
b2 b2
= − cos tı̂ı − sin t ̂ + bt k̂
a a
which is another helix. In the figure below, the red curve is the original helix
and the blue curve is the helix traced by the centre of curvature.
z
x
6 This figure is a variant of this picture.
CHAPTER 1. CURVES 38
1.4.1 Exercises
Exercises — Stage 1
1. In the sketch below of a three-dimensional curve and its osculating
circle at a point, label T̂ and N̂. Will B̂ be pointing out of the paper
towards the reader, or into the paper away from the reader?
2. In the formula
ds
(t) = |v(t)| = |r0 (t)|
dt
does s stand for speed, or for arclength?
3. Which curve (or curves) below have positive torsion, which have neg-
ative torsion, and which have zero torsion? The arrows indicate the
direction of increasing t.
z
Exercises — Stage 2
6. ∗. Let C be the space curve
r(t) = et − e−t ı̂ı + et + e−t ̂ + 2t k̂
CHAPTER 1. CURVES 40
r(t) = t2 , t, t3
Find a parametric form for the tangent line at the point corre-
sponding to t = π.
b Find the tangential component aT (t) of acceleration, as a func-
tion of t, for the parametrized space curve r(t).
13. ∗. Suppose, in terms of the time parameter t , a particle moves along
the path r(t) = (sin t − t cos t)ı̂ı + (cos t + t sin t) ̂ + t2 k̂, 1 ≤ t < ∞.
a Find the speed of the particle at time t.
N̂, the binormal vector B̂ and the curvature κ all at the point
(2, 0, 4).
15. ∗. Consider the curve C given by
1 3 1
r(t) = t ı̂ı + √ t2 ̂ + t k̂, −∞ < t < ∞.
3 2
Compute the unit tangent and unit normal vectors T̂(t) and N̂(t).
Compute the curvature κ(t). Simplify whenever possible!
18. ∗.
2 3
a Find the length of the curve r(t) = 1, t2 , t3 for 0 ≤ t ≤ 1.
t3 t2
r(t) = ı̂ı + √ ̂ + t k̂ −∞<t<∞
3 2
a Find the length of the curve r(t) from r(0) = (0, 1, 0) to r(π) =
(π, −1, π 2 ).
a Assuming that the flight ends when z = 0, find out how far the
rocket travels.
b Find the unit tangent and unit normal to the trajectory at its
highest point.
c Also, compute the curvature of the trajectory at its highest
point.
24. ∗. Consider a particle travelling in space along the path parametrized
by
x = cos3 t, y = sin3 t, z = 2 sin2 t
Exercises — Stage 3
26. ∗. Under the influence of a force field F, a particle of mass 2 kg
is moving with constant speed 3 m/s along the path given as the
intersection of the plane z = x and the parabolic cylinder z = y 2 , in
the direction of increasing y. Find F at the point (1, 1, 1). (Length is
measured in m along the three coordinate axes.)
27. ∗. Consider the curve C in 3 dimensions given by
√
r(t) = 2tı̂ı + t2̂ + 3t2 k̂
for t ∈ R.
a Compute the unit tangent vector T(t).
b Compute the unit normal vector N(t).
c Show that the binormal vector B to this curve does not depend
on t and is one of the following vectors:
1/2
√ √0 √0 0
(1) − 3/2 (2) 3/2 (3) − 3/2 (4) √−1/2
0 1/2 1/2 3/2
where a(t), b(t) and c(t) are functions you have to deter-
mine. You should find that one of these functions is zero.
(iii) Draw the curve given by a(t), b(t) in the xy-plane.
(iv) Is the drawing consistent with parts (f) and (g)? Explain.
CHAPTER 1. CURVES 44
dT̂
= κ N̂
ds
Moreover, the torsion τ and the binormal vector B̂ are defined by the
equations
dB̂
B̂ = T̂ × N̂, = −τ N̂
ds
Show that
dN̂
= −κ T̂ + τ B̂
ds
p
29. ∗. A skier descends the hill z = 4 − x2 − y 2 along a trail with
parameterization
π
x = sin(2θ), y = 1 − cos(2θ), z = 2 cos θ, 0≤θ≤
2
Let P denote the point on the trail where x = 1.
a Find the vectors T̂, N̂, B̂ and the curvature κ of the ski trail at
the point P .
√
b The skier’s acceleration at P is a = (−2, 3, −2 2). Find, at P ,
(i) the rate of change of the skier’s speed and
(ii) the skier’s velocity (a vector).
30. ∗. A particle moves so that its position vector is given by r(t) =
cos t , sin t , c sin t , where t > 0 and c is a constant.
a Find the velocity v(t) and the acceleration a(t) of the particle.
− 4, 0, 14 .
a Compute the curvature of the track at the point
b Compute the radius of the circle that best approximates the
bend at the point\\ − 4, 0, 14 (that is, the radius of the oscu-
c A car drives down the track so that its position at time t is given
by r(t2 ). (Note the relationship between t and θ is θ = t2 ).
Compute the following quantities.
(i) The speed at the point − 4, 0, 14 .
d2 r d2 s ds 2
the acceleration a(t) = 2
(t) = 2
(t) T̂(t) + κ(t) (t) N̂(t)
dt dt dt
ds dr
the speed (t) = |v(t)| = (t)
dt dt
Z T Z Tp
ds
the arc length s(T ) = (t) dt = x0 (t)2 +y 0 (t)2 +z 0 (t)2 dt
0 dt 0
|v(t) × a(t)|
the curvature κ(t) = ddtT̂ (t)/ ds
dt (t) =
( ds
dt (t))
3
dφ dT̂
κ(s) = ds (s) = ds (s)
v(t) × a(t) · da
dt (t)
the torsion τ (t) =
|v(t) × a(t)|2
v(t) × a(t)
the binormal B̂(t) = T̂(t) × N̂(t) =
|v(t) × a(t)|
dr
Under arclength parametrization (i.e. if t = s) we have T̂(s) = ds (s) and
the Frenet-Serret formulae
dT̂
(s) = κ(s) N̂(s)
ds
CHAPTER 1. CURVES 46
dN̂
(s) = τ (s) B̂(s) − κ(s) T̂(s)
ds
dB̂
(s) = −τ (s) N̂(s)
ds
When the curve lies entirely in the xy-plane the curvature is given by
When the curve lies entirely in the xy-plane and the y-coordinate is given as
a function, y(x), of the x-coordinate, the curvature is
d2 y
2 (x)
dx
κ(x) = h
dy
2 i3/2
1 + dx (x)
dx d2 x
Notice that this follows from the previous formula since dx = 1 and dx2 = 0.
rptℓ´1 q rptℓ q
In this notation the subscript C specifies the curve, and ds signifies arc
length.
b For a curve y = f (x), a ≤ x ≤ b, in R2 that we call C, and for a function
g(x, y), we define
Z Z b p
g(x, y) ds = g x, f (x) 1 + f 0 (x)2 dx
C a
♦
2
Example 1.6.2 Suppose that we have a helical wire
r(t) = x(t) , y(t) , z(t) = a cos t , a sin t , bt 0 ≤ t ≤ 2π
and that this wire has constant mass density ρ. Let’s find the centre of mass of
the wire. Recall that the centre of mass is x̄, ȳ, z̄) with, for example, x̄ being
the weighted average
R R
xρds xds
x̄ = R = R (since ρ is constant)
ρds ds
R R
yds zds
of x over the wire. Similarly ȳ = R and z̄ = R . For the given curve
ds ds
x(t) , y(t) , z(t) = a cos t , a sin t , bt
x0 (t) , y 0 (t) , z 0 (t) = − a sin t , a cos t , b
ds p
(t) = x0 (t)2 + y 0 (t)2 + z 0 (t)2
dt p
= a2 sin2 t + a2 cos2 t + b2
p
= a2 + b2
so that
R R 2π √ R 2π
xds 0
x(t) a2 + b2 dt a cos(t) dt
x̄ = R = R 2π √ = 0 =0
ds a2 + b2 dt 2π
0
R R 2π √ R 2π
yds 0
y(t) a2 + b2 dt a sin(t) dt
ȳ = R = R 2π √ = 0 =0
ds 2 2
a + b dt 2π
0
1 We could relax these conditions somewhat by instead assuming that r0 (t) and ρ(t) are
bounded and are continuous except at a finite number of points. (r0 (t) need not exist at all
at those points.)
CHAPTER 1. CURVES 48
R R 2π √ R 2π
zds 0
z(t) a2 + b2 dt bt dt b h t2 i2π
z̄ = R = R 2π √ = 0 = = bπ
ds a2 + b2 dt 2π 2π 2 0
0
So the centre of mass is right on the axis of the helix, half way up, which makes
perfect sense.
1.6.1 Exercises
Exercises — Stage 1
1. Give an equation for arclength of a curve C as a line integral.
2.
R
a Show that the integral C f (x, y) ds along the curve C given in
polar coordinates by r = r(θ), θ1 ≤ θ ≤ θ2 , is
s 2
Z θ2
2
dr
f r(θ) cos θ, r(θ) sin θ r(θ) + (θ) dθ
θ1 dθ
Exercises — Stage 2
√ 2
3. Calculate C xy 2 3
R
z ds, where C is the curve 3 t , 3t , 3t from t = 1
to t = 2.
4. A hoop of radius 1 traces out the curve x2 + y 2 = 1, where x and y are
measured in metres. At a point (x, y), its density is x2 kg per metre.
What is the mass of the hoop?
R
5. Compute C (xy + z)ds where C is the straight line from (1, 2, 3) to
(2, 4, 5).
R
6. Evaluate the path integral C f (x, y, z) ds for
a f (x, y, z) = x cos z, C : r(t) = tı̂ı + t2̂, 0 ≤ t ≤ 1.
Exercises — Stage 3
∗. Evaluate the line integral C F · n̂ ds where F(x, y) = xy 2 ı̂ı + yex ̂
R
9.
, C is the boundary of the rectangle R: 0 ≤ x ≤ 3, −1 ≤ y ≤ 1,
and n̂ is the unit vector, normal to C, pointing to the outside of the
rectangle.
10. ∗. Let C be the curve given by
ma = F
where F is the net force being applied to the bead. The bead is subject
to two forces. The gravitational force is −mĝ. By definition, absence of
friction means that the wire is does not apply any force that is in the direction
tangential to the wire. But, because it is stiff, the wire never changes shape
and instead applies just the right amount of force, in the direction normal to
1 We are mathematicians — we like idealized situations.
CHAPTER 1. CURVES 50
the wire, that is needed to keep the bead on the wire2 without bending the
wire. Call this normal force W N̂.
̂ wire
W N̂
px, yq ı̂ı
bead
T̂ ´mĝ (gravity)
So, by Newton’s law,
m a = −mg ̂ + W N̂
We’ll analyse this equation by splitting it into its tangential and normal com-
ponents.
To extract the tangential component of Newton’s law, we dot it with v =
|v|T̂. Since T̂ · N̂ = 0 this kills all normal components.
dv
mv · = −mĝ · v + W N̂ · v
dt
1 d dy
m (v · v) = −mg
2 dt dt
Here we have used
• Theorem 1.1.3.c on the left hand side and
• that ̂ · v is just the y component of v and
• that N̂ and v = |v|T̂ are perpendicular.
Moving everything to the left hand side of the equation gives
d 1 2
m|v| + mgy = 0
dt 2
and we conclude that the quantity
Equation 1.7.1 Conservation of Energy.
1
E=m|v|2 + mgy
2
is a constant, independent of time. This is, of course,
q the principle of
2E
conservation of energy. It determines the speed |v| = m − 2gy of the bead
as a function of the height y (and of the energy E, which is determined by the
initial conditions).
To extract the normal component of Newton’s law, we dot it with N̂:
ma · N̂ = −mĝ · N̂ + W
Since
d2 s ds 2 d2 s
a= T̂ + κ N̂ = T̂ + κ|v|2 N̂
dt2 dt dt2
and T̂ and N̂ are perpendicular, this gives, after a little rearrangement,
2 This force is required to keep the bead from either passing through the wire or flying off
the wire.
CHAPTER 1. CURVES 51
N̂
T̂
N̂ T̂
• If the hill is concave downward as in the figure on the left above, then
N̂ points downward and the hill is allowed to have W ≤ 0 (which corre-
sponds to the normal force W N̂ pushing upward). If ever W > 0, the hill
would have to pull on you to keep you on hill. It can’t, so you become
airborne. Since ̂ · N̂ < 0, this happens whenever
r
2 g
W > 0 ⇐⇒ mκ|v| + mĝ · N̂ > 0 ⇐⇒ |v| > |̂ · N̂|
κ
• If the hill is concave upward as in the figure on the right above, then N̂
points upward and the hill is allowed to have W ≥ 0 (which corresponds
to the normal force W N̂ pushing upward). Since ̂ · N̂ > 0 we always
have W = mκ|v|2 + mĝ · N̂ > 0. You never become airborne. On the
other hand your knees may complain.
a ´ y φ aN̂
px, yq
3 We assume that you are going downhill and that the curvature κ > 0.
CHAPTER 1. CURVES 52
(iii) y hits 2a. This is the summit of the culvert. You descend on the other
side.
Which case actually happens is determined by the relative sizes of yS , yA and
2a.
2 E 1 E 2 E
• Comparing yS = 3 mg + 3 mg and yA = 3 mg + a3 , we see that yS ≤
E
yA ⇐⇒ mg ≤ a.
2 E a
• Comparing yA = 3 mg + 3 and a = 2
3a + a3 , we see that yA ≤ a ⇐⇒
E
mg ≤ a.
2 E a
• Comparing yA = 3 mg + 3 and 2a = 35 a + a3 , we see that yA ≤ 2a ⇐⇒
E 5
mg ≤ 2 a.
• If 0 ≤ mgE
≤ a then 0 ≤ yS ≤ yA ≤ a . In this case you just oscillate
between heights 0 and yS ≤ a in the bottom half of the culvert, as in the
figure on the left below.
• If a ≤ mgE
≤ 52 a then a ≤ yA ≤ yS , 2a . In this case you make it more
than half way to the top. But you become airborne at y = yA which is
somewhere between the half way mark y = a and the top y = 2a. At
this point our model breaks down because you are no longer in contact
with the culvert. You just freely follow a parabolic arc until you crash
back into the culvert, as in the figure in the centre below.
• If 25 a < mg
E
then 2a < yA < yS . In this case you successfully go
all the way around the culvert, looping the loop, as in the figure on the
E
right below. Note that, as mg > 52 a > 2a, this requires significantly more
energy than that required to reach the top, i.e. to reach height 2a.
CHAPTER 1. CURVES 53
1.7.4 Exercises
Exercises — Stage 1
You may assume the acceleration due to gravity is g = 9.8 m/s2 . You may
also assume that the systems described function as they do in the book: so
tracks are frictionless, etc., unless otherwise mentioned.
1. The figure below represents a bead sliding down a wire. Sketch vectors
representing the normal force the wire exerts on the bead, and the
force of gravity.
N̂
−̂ T̂
Let W N̂ be the normal force exerted by the wire when the bead
is at position x. Note W > 0. Is dW
dx positive or negative?
4. A skateboarder is rolling on a frictionless, very tall parabolic ramp
with cross-section described by y = x2 . Given a boarder of mass
m with system energy E, what is the highest elevation the skater
reaches? How does this compare to a circular culvert?
Exercises — Stage 2
5. A skateboarder of mass 100 kg is freely rolling in a frictionless circular
culvert of radius 5 m. If the skateboarder oscillates between vertical
heights of 0 and 3 m, what is the energy E of the system?
6. A skateboarder is rolling on a frictionless circular culvert of radius 5
m. What should their speed be when they’re at the bottom of the
culvert (y = 0) for them to make it all the way around?
7. A ball of mass 1 kg rolls down a track with the shape r(θ) = (3 cos θ, 5 sin θ, 4+
4 cos θ) for 0 ≤ θ ≤ π2 . Coordinates are measured in metres, and the
CHAPTER 1. CURVES 54
13π
If the bead breaks off the wire at θ = 3 , how fast is the bead
moving at that point?
9. A skier is gliding down a hill. The hill can be described as r(t) =
(ln t, 1 − t), 1/e ≤ t ≤ e, with coordinates measured in kilometres.
How fast would the skier have to be moving in order to catch air?
Exercises — Stage 3
10. A wire follows the arclength-parametrized path r(s) = (x(s), y(s)). A
bead, equipped with a jet pack, slides down the wire. The jet pack
can exert a variable force in a direction tangent to the wire,
U T̂.
Assuming the bead slides with constant speed dr dt
= c dr = c, find
ds
a simplified equation for U , the signed magnitude of the force exerted
by the jet pack.
Let the acceleration due to gravity be g, and let the mass of the
bead with its jet pack be m. Give U as a function of s.
Remark: most beads this author has seen did not have jet packs.
However, in modelling a frictionful4 system, friction acts as a force
that is directly opposing the direction of motion — much like our jet
pack.
11. A snowmachine is cautiously descending a hill in low gear. Its engine
provides a force M T̂ parallel to the direction of motion. The engine
provides whatever force is necessary to keep the snowmachine moving
at a constant speed, |v|. Its treads do not slip.
a Give a formula for M in terms of the mass m of the snowma-
chine, the acceleration due to gravity g, and the tangent vector
T̂ to the hill.
3π
at the point x = 4 for a snowmachine of mass 200 kg.
12. A skateboarder rolls along a culvert with elliptical cross-section de-
scribed by
Here m is the mass of the car, g is a constant, −mg k̂ is the force due
to gravity and N r(t) is the force that the roller-coaster track applies
to the car
to keep the car on the track. Since the track is frictionless,
N r(t) is always perpendicular to v(t) = dr dt (t).
θ̂θ r̂
̂ θ
r̂(θ) = cos θ ı̂ı + sin θ ̂
r
θ̂θ (θ) = − sin θ ı̂ı + cos θ ̂
θ
ı̂ı
p0, 0q
Note that this basis has two very nice properties.
then
a the position vector of the point at time t is
r(t) = r(t) r̂ θ(t)
dr dθ
v= r̂ + r θ̂θ
dt dt
But it is important to remember that the arguments really are there.
Proof. The vector from the origin to the point whose polar coordinates are (r, θ)
is r = r r̂(θ). So if we parametrize a curve by giving the polar coordinates at
1 As usual r is the distance from the origin to the point and θ is angle between the
x-axis and the vector from the origin to the point. The symbols r, θ are the standard
mathematics symbols for the polar coordinates. Appendix A.7 gives another set of symbols
that is commonly used in the physical sciences and engineering.
CHAPTER 1. CURVES 57
time t,
r(t) = r(t) r̂ θ(t)
dr dr̂ dθ
v(t) = (t) r̂ θ(t) + r(t) θ(t) (t)
dt dθ dt
dr dθ
= (t) r̂ θ(t) + r(t) (t) θ̂θ θ(t)
dt dt
d2 r dr dr̂ dθ dr dθ d2 θ dθ 2 dθ̂θ
a(t) = 2 r̂ + + θ̂θ + r 2 θ̂θ + r
dt dt dθ dt dt dt dt dt dθ
h d2 r dθ 2 i h d2 θ dr dθ i
= −r r̂ + r 2 + 2 θ̂θ
dt2 dt dt dt dt
Example 1.8.3 As an example, consider a bead that is sliding on a frictionless
rod that has one end fixed at the origin and that is rotating about the origin
at a constant Ω rad/sec.
Ωt
Because the rod is frictionless, it is incapable of applying to the bead any
force parallel to the rod. So under Newton’s law, ma = F, the radial2 com-
ponent of the acceleration of the particle is exactly
zero. So, if the polar
coordinates of the bead at time t are r(t), θ(t) , then, by Lemma 1.8.2.c,
d2 r dθ 2
− r =0
dt2 dt
dθ
As the rod is rotating at Ω rad/sec, dt = Ω and
d2 r
− Ω2 r = 0
dt2
The general solution to this constant coefficient second order ordinary differ-
ential equation is3
r(t) = AeΩ t + Be−Ω t
where A and B are arbitrary constants that are determined by initial condi-
tions. Just as an example, if r(0) = 1 and r0 (0) = 0, then A + B = 1 and
AΩ − BΩ = 0, so that A = B = 21 and
1 Ωt
e + e−Ω t
r(t) =
2
If, again for example, θ(0) = 0, then θ(t) = Ωt and the bead follows the polar
coordinate curve
1
r(θ) = eθ + e−θ
2
Observe that r(θ) is 1 when θ = 0, increases as θ increases, and tends to ∞ as
θ → +∞. The curve is a spiral.
CHAPTER 1. CURVES 58
Example 1.8.4 Conic sections in polar coordinates. In this example,
we derive the equation of a general conic section in polar coordinates. A conic
section is the intersection of a plane with a cone. This is illustrated in the
figures below.
• whose distance from a fixed point F (called the focus of the conic)
• is a constant multiple ε ≥ 0 (called the eccentricity of the conic)
• of the distance from P to a fixed line L (called the directrix of the conic).
Choose a coordinate system with the focus F of the conic being the origin and
with the directrix L being x = p for some p > 0.
2 The θ̂
θ component of the acceleration just tells us how much normal force the rod is
applying to the bead to keep it on the rod.
3 A review of the technique used to find this solution is given in Appendix A.9. In any
2
event, it is easy to check that r(t) = AeΩ t + Be−Ω t really does obey ddt2r − Ω2 r = 0.
CHAPTER 1. CURVES 59
y L
P Q
r
θ
F x
p
If P has polar coordinates (r, θ), then P has x-coordinate r cos θ. The point
Q on the line L in the figure above has x-coordinate p. So the distance from P
to L, which is also the distance from P to Q, is p − r cos θ. The distance from
P to F is r. We require that the distance from P to F is ε times the distance
from P to L. So
εp
r = ε p − r cos θ ⇐⇒ r =
1 + ε cos θ
The numerator εp is usually renamed to ` giving the equation
`
r=
1 + ε cos θ
Example 1.8.5 Conic sections in polar coordinates, again. We’ll now
take the equation r = 1+ε`cos θ for a conic section in polar coordinates, from
the last example, and convert it to the more familiar Cartesian coordinates.
Just by the definition of polar coordinates
r 1 + ε cos θ = ` ⇐⇒ r = ` − εx
⇐⇒ x2 + y 2 = `2 − 2ε`x + ε2 x2
⇐⇒ (1 − ε2 )x2 + 2ε`x + y 2 = `2 (C)
ℓ
2 2 2
x +y =`
x
ε`
2
x+ 1−ε2 y2
`2
+ `2
=1 p´rM , 0q x
(1−ε2 )2 1−ε2
p0, ´rm q
`
and is of course an ellipse with semi-major axis rM = 1−ε2 and semi-
`
minor axis rm = √1−ε 2
.
• If ε = 1, (C) reduces to
y
pℓ{2, 0q
y 2 = `2 − 2`x x
p0, ´ℓq
ε`
2
x− ε2 −1 y2
`2
− `2
=1 x
(ε2 −1)2 ε2 −1
1.8.1 Exercises
Exercises — Stage 1
1. Consider the points
(x1 , y1 ) = (3, 0) (x2 , y2 ) = (1, 1) (x3 , y3 ) = (0, 1)
(x4 , y4 ) = (−1, 1) (x5 , y5 ) = (−2, 0)
For each 1 ≤ i ≤ 5,
• sketch, in the xy-plane, the point (xi , yi ) and
• find the polar coordinates ri and θi , with 0 ≤ θi < 2π, for the
point (xi , yi ).
CHAPTER 1. CURVES 61
2.
a Find all pairs (r, θ) such that
(−2, 0) = r cos θ , r sin θ
êr (θ) = cos θ ı̂ı + sin θ ̂ êθ (θ) = − sin θ ı̂ı + cos θ ̂
a Determine, for each angle θ, the lengths of the vectors êr (θ)
and êθ (θ) and the angle between the vectors êr (θ) and êθ (θ).
Compute êr (θ) × êθ (θ) (viewing êr (θ) and êθ (θ) as vectors in
three dimensions with zero k̂ components).
b For each 1 ≤ i ≤ 5, sketch, in the xy-plane, the point (xi , yi )
and the vectors êr (θi ) and êθ (θi ). In your sketch of the vectors,
place the tails of the vectors êr (θi ) and êθ (θi ) at (xi , yi ).
4. ∗. Match the following equations with the corresponding pictures.
Cartesian coordinates are (x, y) and polar coordinates are (r, θ).
(A) y (B) y
x x
(C) y (D) y
x x
CHAPTER 1. CURVES 62
(E) y (F) y
x x
Exercises — Stage 2
5. Recall that a point with polar coordinates r and θ has x = r cos θ and
y = r sin θ. Let r = f (θ) be the equation of a plane curve in polar
coordinates. Find the curvature of this curve at a general point θ.
6. Find the curvature of the cardioid r = a(1 − cos θ).
3. The square of the period of each planet is proportional to the cube of the
major axis of the planet’s orbit.
Newton showed that all of these behaviours follow from the assumption that
the acceleration a(t) of each planet obeys the law of motion ma = F where m
is the mass of the planet and
GM m
F=− r
r3
is the “gravitational force” applied on the planet by the sun. Here G is a
constant2 , called the “gravitational constant” or the “universal gravitational
constant”, M is the mass of the sun, r is the vector from the sun to the planet
and r = |r|.
1 The German astronomer Johannes Kepler (1571–1630) developed these laws during the
course of an attempt to relate the five extraterrestrial planets then known to the five Platonic
solids. He based the laws on a great number of careful measurements made by the Danish
Astronomer Tycho Brahe (1546–1601). Then Isaac Newton (English, 1642–1727) provided
the explanation in 1687. Kepler also wrote a paper entitled “On the Six-Cornered Snowflake”.
Tycho Brahe lost his nose in a sword duel and wore a prosthetic nose from then on. The
story is that Brahe died from a burst bladder that resulted from his refusing to leave the
dinner table before his host.
2 Its value is about 6.67408 × 10−11 m3 kg−1 sec−2 .
CHAPTER 1. CURVES 63
In this section, we’ll show that some of these properties follow from the
weaker assumption that the acceleration a(t) of each planet obeys the law of
motion ma = F with F being a central force. That is, the assumption that
F is parallel to r. The verification that the other properties follow from the
specific form of the gravitational force, proportional to r−2 , will be delayed
until the optional §1.10.
So, in this section, we assume that we have a parametrized curve r(t) and
that this curve obeys
d2 r
m 2 (t) = F r(t)
dt
where, for all r ∈ R3 , F(r) is parallel to r. We shall show that
1. The path r(t) lies in a plane through the origin and that
2. the radius vector r sweeps out equal areas in equal times.
We’ll start by trying to guess what the plane is. Pretend that we know that
r(t) lies in a fixed plane through the origin. Then v(t) = dr
dt (t) lies in the same
plane and r(t) × v(t) is perpendicular to the plane. If our path really does
lie in a fixed plane, r(t) × v(t) cannot change direction — it must always be
parallel to the normal vector to the plane. So let’s define
r(t) · Ω = 0
• If Ω 6= 0, this is exactly the statement that r(t) always lies in the plane
through the origin with normal vector Ω .
• If Ω = 0, then r(t) is always parallel to v(t) and there is some function
α(t) such that
dr
(t) = v(t) = α(t) r(t)
dt
This is a first order, linear, ordinary differential equation that we can
solve by using an integrating factor. Set
Z t
β(t) = α(t) dt
0
Then
dr dr
(t) = α(t) r(t) ⇐⇒ e−β(t) (t) − α(t)e−β(t) r(t) = 0
dt dt
3 Physicists call m Ω (t) the angular momentum at time t and refer to (A) as (an example
of) conservation of angular momentum. Conservation of angular momentum is exploited in
gyro-compasses and by ice skaters (to spin faster/slower).
CHAPTER 1. CURVES 64
d −β(t)
⇐⇒ e r(t) = 0
dt
⇐⇒ e−β(t) r(t) = r(0)
⇐⇒ r(t) = eβ(t) r(0)
so that r(t) lies on a line through the origin. This makes sense — the
particle is always moving parallel to its radius vector.
This completes the verification that r(t) lies in a plane through the origin.
Now we show that the radius vector r(t) sweeps out equal areas in equal
times. In other words, we now verify that the rate at which r(t) sweeps out
area is independent of time. To do so we rewrite the statement that |r(t)×v(t)
is constant in polar coordinates. Writing r(t) = r(t)r̂ θ(t) and then applying
Lemma 1.8.2.b gives that
dr dθ dθ
θ̂θ = r2
constant = r × v = rr̂ × r̂ + r
dt dt dt
since |r̂ × r̂| = 0, |r̂ × θ̂θ | = 1
dθ 1
πr2 = r2 dθ rpt ` dtq
2π 2
dθ
rptq
1.9.1 Exercises
Exercises — Stage 3
1. ∗. Let r(t) = x(t)ı̂ı + y(t) ̂ + z(t) k̂ be the position of a particle at
time t . Suppose the motion of the particle satisfies the differential
2
equation ddt2r = f (r)r where r = |r| .
a Suppose f (r) is an arbitrary function of r . Prove or disprove
each of the following statements.
(i) The motion of the particle is planar.
(ii) The path of the particle sweeps out equal areas in equal
times.
b Find all forms of f (r) for which the motion of the particle always
lies on a straight line.
c Give a specific form of f (r) for which the motion of the particle
could lie on an ellipse.
2. ∗. An object moves along a curve in the xy-plane having polar equa-
1
tion r = θ+α (where α is a constant) under the influence of a central
force so that the object has no transverse acceleration.
a Verify that r2 θ̇ = h remains constant as the object moves.
CHAPTER 1. CURVES 65
d2 r GM GM
2
= − 3 r = − 2 r̂
dt r r
then the curve obeys Kepler’s laws
1. r(t) runs over an ellipse having one focus at the origin and
2. r(t) sweeps out equal areas in equal times and
3. the square of the period is proportional to the cube of the major axis of
the ellipse.
We just showed, in §1.9, that the fact that − GMr 3 r is parallel to r implies that
r(t) lies in a plane through the origin and sweeps out equal area in equal times.
We now verify the remaining Kepler laws.
We start by just rewriting Newton’s laws above in polar coordinates. We
saw in Lemma 1.8.2.c, that if we write r(t) = r(t) r̂(t), then
2 !
d2 r d2 r
2
dθ d θ dr dθ
= −r r̂ + r 2 + 2 θ̂θ
dt2 dt2 dt dt dt dt
GM GM
=− r = − 2 r̂
r3 r
Then
dr 1 du dθ du dθ h
= 2 = hu2
(t) = − 2 θ(t) (t) = −h θ(t) since
dt u dθ dt dθ dt r
d2 r d2 u dθ 2
2 d2 u
2
(t) = −h 2
θ(t) (t) = −h u θ(t) 2
θ(t)
dt dθ dt dθ
and our equation becomes
Equation 1.10.2
d2 u d2 u GM
−h2 u2 2
− h2 u3 = −GM u2 or +u= 2
dθ dθ2 h
This is a second order, linear, ordinary differential equation with constant
coefficients. Recall1 that the general solution of such an equation is the sum
of a “particular solution” (i.e. any one solution, which in this case we can take
to be the constant function GM h2 ) plus the general solution of the homogeneous
equation u0 + u = 0, which one often writes as
A cos θ + B sin θ
C B
α
A
to write A = C cos α and B = C sin α so that the general solution of the
homogeneous equation u0 + u = 0 becomes
The angle α just shifts the zero point of our coordinate θ. By rotating our
coordinate system by α, we can arrange that α = 0 and then
1 ` h2 Ch2
r(t) = GM
= with ` = , ε=
h2 + C cos(θ(t)) 1 + ε cos θ GM GM
As we saw in Example 1.8.4, this is exactly the equation of a conic section with
eccentricity ε.
That leaves only the last of Kepler’s laws, relating the period to the semi-
major axis. As we are talking about planets, whose orbits remain bounded, our
conic section must be a circle or ellipse, rather than a parabola or hyperbola.
1 See Appendix A.9.
CHAPTER 1. CURVES 67
Looking back at Example 1.8.5, we see that the semi-major and semi-minor
axes of our ellipse are
` `
a= b= √
1 − ε2 1 − ε2
The period T of our orbit is just the length of time it takes the radius vector
r(t) to sweep out the area of the ellipse2 , which is πab. As the rate at which
h
the radius vector is sweeping out area is 21 r2 dθ
dt = 2 , we have
πab 2 4π 2 a2 b2 4π 2 a2 b2 4π 2 3 b2
T2 = = = = a since ` =
h/2 h2 GM ` GM a
P P
P
“oid” meaning “having the shape of”. The curve was first discussed by Johann Bernoulli in
1691–92.
CHAPTER 1. CURVES 68
̂ Q ̂
φ θ
θ ı P
ı̂ ı̂ı
O O
That is
• the vector from the centre, O, of the circle of radius a to the centre, Q,
of the ball of radius a/4 is 34 a cos θ, sin θ and
• the vector from the
1
centre, Q, of the ball of radius a/4 to the point P is
4 a cos φ, − sin φ
As θ runs from 0 to π2 , the point of contact between the two circles travels
through one quarter of the circumference of the circle of radius a, which is a
distance 14 (2πa), which, in turn, is exactly the circumference of the inner circle.
Hence if φ = 0 for θ = 0 (i.e. if P starts on the x-axis), then for θ = π2 , P
is back in contact with the big circle at the north pole of both the inner and
π
outer circles. That is, φ = 3π2 when θ = 2 . (See the figure on the right above.)
So φ = 3θ and P has coordinates
3 1 a
a cos θ, sin θ + a cos φ, − sin φ = 3 cos θ + cos 3θ, 3 sin θ − sin 3θ
4 4 4
As, recalling your double angle, or even better your triple angle, trig identities,
we have
3 cos θ + cos 3θ = cos θ[3 + cos2 θ − 3 sin2 θ] = cos θ[3 + cos2 θ − 3(1 − cos2 θ)]
= 4 cos3 θ
3 sin θ − sin 3θ = sin θ[3 − 3 cos2 θ + sin2 θ] = sin θ[3 − 3(1 − sin2 θ) + sin2 θ]
= 4 sin3 θ
Oof! As x2/3 + y 2/3 = a2/3 cos2 θ + a2/3 sin2 θ , the path traced by P obeys
the equation
x2/3 + y 2/3 = a2/3
CHAPTER 1. CURVES 69
k̂
z
c
ρ̂ r(t) = c + ρ cos tı̂ı + ρ sin t ̂
ρı̂ı 0 ≤ t < 2π
y
x
Finally, let’s consider a circle in general position. The secret to parametriz-
ing a general circle is to replace ı̂ı and ̂ by two new vectors ı̂ı0 and ̂0 which
a are unit vectors,
z k̂1
ρ̂1
c
r(t) = c + ρ cos tı̂ı0 + ρ sin t ̂0
ρı̂ı1
0 ≤ t < 2π
y
x
To check that this is correct, observe that
• r(t) − c is parallel to the plane of the desired circle because both ı̂ı0
and ̂0 are parallel to the plane of the desired circle and r(t) − c =
ρ cos tı̂ı0 + ρ sin t ̂0
• r(t) − c is of length ρ for all t because
since ı̂ı0 · ı̂ı0 = ̂0 · ̂0 = 1 (ı̂ı0 and ̂0 are both unit vectors) and ı̂ı0 · ̂0 = 0 (ı̂ı0
and ̂0 are perpendicular).
To find such a parametrization in practice, we need to find the centre c of the
circle, the radius ρ of the circle and two mutually perpendicular unit vectors,
ı̂ı0 and ̂0 , in the plane of the circle. It is often easy to find at least one point
p on the circle. Then we can take ı̂ı0 = |p−c| p−c
. It is also often easy to find a
unit vector, k̂0 , that is normal to the plane of the circle. Then we can choose
̂ 0 = k̂0 × ı̂ı0 . We’ll illustrate this now.
Example 1.12.1 Let C be the intersection of the sphere x2 + y 2 + z 2 = 4 and
the plane z = y.
• The intersection of any plane with any sphere is a circle. The plane in
question passes through the centre of the sphere, so C has the same centre
and same radius as the sphere. So C has radius 2 and centre (0, 0, 0).
• Notice that the point (2, 0, 0) satisfies both x2 +y 2 +z 2 = 4 and z = y and
so is on C. We may choose ı̂ı0 to be the unit vector in the direction from
the centre (0, 0, 0) of the circle towards (2, 0, 0). Namely ı̂ı0 = (1, 0, 0).
1
r(t) = 2 cos t (1, 0, 0) + 2 sin t √ (0, 1, 1)
2
k̂1
̂1 sin t sin t
= 2 cos t, √ , √
2 2
y 0 ≤ t < 2π
1
ı̂ı
x
√ √
To check this, note that x = 2 cos t, y = 2 sin t, z = 2 sin t satisfies both
x2 + y 2 + z 2 = 4 and z = y.
Example 1.12.2 Let C be the circle that passes through the three points
(3, 0, 0), (0, 3, 0) and (0, 0, 3).
• All three points obey x + y + z = 3. So the circle lies in the plane
x + y + z = 3. We guess, by symmetry, or by looking at the figure below,
that the centre of the circle is at the centre of mass of the three points,
which is 13 [(3, 0, 0)+(0, 3, 0)+(0, 0, 3)] = (1, 1, 1). We must check this and
can do so by checking that (1, 1, 1) is equidistant from the three points:
√
(3, 0, 0) − (1, 1, 1) = (2, −1, −1) = 6
√
(0, 3, 0) − (1, 1, 1) = (−1, 2, −1) = 6
√
(0, 0, 3) − (1, 1, 1) = (−1, −1, 2) = 6
This tells us both that (1, 1, 1) is indeed the centre (as only the centre is
equidistant
√ from any three distinct points on a circle) and that the radius
of C is 6.
• We may choose ı̂ı0 to be the unit vector in the direction from the centre
(1, 1, 1) of the circle towards (3, 0, 0). Namely ı̂ı0 = √16 (2, −1, −1).
• Then
1 1
̂0 = k̂0 × ı̂ı0 = √ (1, 1, 1) × (2, −1, −1) = √ (0, 3, −3)
18 18
1
= √ (0, 1, −1)
2
√
Substituting in c = (1, 1, 1), ρ = 6, ı̂ı0 = √1 (2, −1, −1)
6
and ̂0 = √1 (0, 1, −1)
2
gives
√ 1 √ 1
r(t) = (1, 1, 1) + 6 cos t √ (2, −1, −1) + 6 sin t √ (0, 1, −1)
6 2
√ √
= 1 + 2 cos t, 1 − cos t + 3 sin t, 1 − cos t − 3 sin t
CHAPTER 1. CURVES 72
x y
To check this, note that r(0) = (3, 0, 0), r 2π 4π
3 = (0, 3, 0) and r 3 =
√ √
(0, 0, 3) since cos 2π
3 = cos 4π
3 = − 1
2 , sin 2π
3 = 2
3
and sin 4π
3 =− 2 .
3
Chapter 2
Vector Fields
♦
Here are two typical applications that naturally involve vector fields.
• If v(x, y, z) is the velocity of a moving fluid at position (x, y, z), then v
is called a velocity field.
• If F(x, y, z) is the force at position (x, y, z), then F is called1 a force
field.
Example 2.1.2 The Point Source. Imagine
• The whole world is filled with an incompressible fluid. Call it water.
• Somehow you find a way to produce still more water at the origin. Say
you create 4πm litres per second.
• This forces the water to flow outward. Let’s suppose that it flows sym-
metrically outward from the origin.
Let’s find the resulting vector field v(x, y, z). As the flow is to be symmetric,
the velocity of the water at the point (x, y, z)
• has to be pointing radially outward from the origin. That is, the direction
of the velocity vector v(x, y, z) has to be the unit radial vector
xı̂ı + ŷ + z k̂
r̂(x, y, z) = p
x2 + y 2 + z 2
1 No, force fields are not only a sci-fi trope. Gravity is an example of a force field.
73
CHAPTER 2. VECTOR FIELDS 74
• The magnitude of the velocity, i.e. the speed |v(x, y, z)| of the water, has
to depend only on the distance from the origin. That is, the speed can
only be some function of
p
r(x, y, z) = x2 + y 2 + z 2
vprq dt thick
During a very short time interval dt seconds, 4πm dt litres of water is created
at the origin (which is the red dot). As the water is incompressible, 4πm dt
litres of water must exit through the sphere during the same time interval to
make room for the newly created water.
But, at the surface of the sphere the water is flowing radially outward with
speed v(r). So during the time interval in question the water near the surface
of the sphere moves outward a distance v(r) dt, andp in particular the water
that was in the thin spherical shell r − v(r) dt ≤ x2 + 2 2
py + z ≤ r at
the beginning of the time interval exits through the sphere x2 + y 2 + z 2 = r
during the time interval. The shell is sketched in gray in the figure above. The
volume of water in the gray shell is essentially the surface area of the shell,
which is 4πr2 , times the thickness of the shell, which is v(r) dt. So, equating
the volume of water created inside the sphere with the volume of water that
exited the sphere,
4πm m
4πm dt = (4πr2 ) v(r) dt =⇒ v(r) =
2
= 2
4πr r
Thus our vector field is
m
v(x, y, z) = r̂(x, y, z)
r(x, y, z)2
If the world were two, rather than three dimensional2 , and the source created
2πm litres per second, the same argument leads to
2πm m
2πm dt = (2πr) v(r) dt =⇒ v(r) = =
2πr r
and to the vector field
m p xı̂ı + ŷ
v(x, y) = r̂(x, y) r(x, y) = x2 + y 2 r̂(x, y) = p
r(x, y) x2 + y 2
To get a mental image of what this field looks like, imagine sketching, for each
m
point (x, y), the vector r(x,y) r̂(x, y) with its tail at (x, y). Note that the vector
m
r(x,y) r̂(x, y)
CHAPTER 2. VECTOR FIELDS 75
This naturally divides the xy-plane into nine parts according to whether each
of the components is positive, 0 or negative —
• ı̂ı · v > 0 and ̂ · v > 0 in (x, y) ∈ R2 y < 0, x > 0
• ı̂ı · v > 0 and ̂ · v = 0 in (x, y) ∈ R2 y < 0, x = 0
• ı̂ı · v > 0 and ̂ · v < 0 in (x, y) ∈ R2 y < 0, x < 0
• ı̂ı · v = 0 and ̂ · v > 0 in (x, y) ∈ R2 y = 0, x > 0
• and so on
• and so on
By the time we have looked at all nine regions we will have built up the following
sketch.
y
Figure 2.1.5
From this sketch we see that, for example, in the first quadrant,
• the fluid is moving upwards and to the left and
• the fluid crosses the x-axis vertically (so that close to the x-axis, the
arrows will be almost vertical) and
• the fluid crosses the y-axis horizontally (so that close to the y-axis, the
arrows will be almost horizontal) and
• there is one point, namely (0, 0), where the vector field is exactly zero.
It’s the black dot in the centre of the figure above. Furthermore v(x, y) =
Ω(−yı̂ı + x̂) is smaller when (x, y) is closer to (0, 0) and v(x, y) is larger
when (x, y) is farther from (0, 0),
Putting all of this accumulated wisdom together, we come up with this better
sketch of the vector field.
CHAPTER 2. VECTOR FIELDS 78
θ
ℓ
τ
mg
between the rod and vertical. The forces acting on the mass are
• gravity and
• the tension in the rod, whose magnitude, τ , automatically adjusts itself
so that the distance between the mass and the hinge is fixed at `.
In the optional4 Section 2.5, we show that the angle θ(t) obeys the second order
nonlinear5 differential equation
d2 θ g
+ sin θ = 0
dt2 `
It is often much more convenient to deal with first order, rather than second
order, differential equations. The second order pendulum equation above may
CHAPTER 2. VECTOR FIELDS 79
So x(t) is the angle at time t and y(t) is the angular velocity at time t. Then,
x0 = y
g
y 0 = − sin x
`
The right hand sides form the vector field
g
v (x, y) = y , − sin x
`
We can sketch this vector field, just as we sketched the vector field of Example
2.1.4. Noting that the horizontal component
= 0 if y = 0
ı̂ı · v(x, y) = y > 0 if y > 0
< 0 if y < 0
we have
y“0
x “ ´π x“0 x“π x “ 2π
CHAPTER 2. VECTOR FIELDS 80
Our full sketch will be less cluttered if we make all arrows the same length.
This gives
y
2.1.1 Exercises
Exercises — Stage 1
1. Below is a sketch of the vector field v(x, y).
3 While we are idealizing, let’s put everything in a vacuum.
4 In the optional Section 2.5 we also include frictional forces. In this example, we do not,
so set the β of Section 2.5 to zero here.
5 It is common, when considering only small amplitude oscillations, to approximate sin θ
by θ. This converts our nonlinear differential equation into a linear differential equation.
6 This “hack” generalizes easily and is commonly used when generating, by computer,
x
∂Q ∂Q
Determine the signs of P , Q, ∂x and ∂y at the point A.
5. Imagine that the vector field v(x, y) = xı̂ı + y ̂
is the velocity field of a moving fluid.
a At time 0 you drop a twig into the fluid at the point (1, 1). What
is the approximate position of the twig at time t = 0.01?
b At time 0 you drop a twig into the fluid at the point (0, 0). What
is the position of the twig at time t = 0.01?
c At time 0 you drop a twig into the fluid at the point (0, 0). What
is the position of the twig at time t = 10?
6. Imagine that the vector field v(x, y) = 2xı̂ı − ̂
is the velocity field of a moving fluid. At time 0 you drop a twig
into the fluid at the point (0, 0). What is the position of the twig at
time t = 10?
Exercises — Stage 2
7. A platform with many small conveyor belts is aligned on a coordinate
plane. Every conveyor belt moves an object on top of it in the direction
of the origin, and a conveyor belt at position (x, y) causes an object
on top of it to move with speed y. Assume the objects do not interfere
with one another.
Give a vector-valued formula for the velocity of an object at posi-
tion (x, y).
8. Friendly bees fly towards your face from all directions. The speed of
each bee is inversely proportional to its distance from your face. Find
a vector field for the velocity of the swarm.
9. Sketch the vector field v(x, y) = (x2 , y).
p
10. Sketch the direction field of v(x, y) =
p
x2 + y 2 , (x − 1)2 + (y − 1)2 .
c v(x, y) = √y ı̂ı−x
2
̂
2
.
x +y
Exercises — Stage 3
15.
a. A pole leans against a vertical wall. The pole has length 2, and
it touches the wall at height H = 1. The pole slides down,
still touching the wall, with its height decreasing at a rate of
dH
dt = 0.5.
dr
= v r(t)
dt
is called a
• field line or integral curve (for general vector fields) or a
• stream line or flow line (when the vector field v is being thought of as a
velocity field) or a
• line of force (when the vector field v is being thought of as a force field)
of the vector field v. ♦
Example 2.2.2 Flow Line Sketch for the Vortex of Example 2.1.4.
Consider the vortex vector field, v(x, y) = Ω −yı̂ı +x̂ of Example 2.1.4. Once
we sketched the vector field, as in Figure 2.1.6, or even made the “skeleton
sketch” of Figure 2.1.5, we can get rough idea of what the stream lines look
like just by following the arrows. For example, suppose that we start a stream
line (i.e. drop the stick into the stream) on the positive x-axis. Looking at
Figure 2.1.5, which is repeated here,
y
the stick
• starts by moving in the +y direction, i.e. straight upward.
• As it moves farther into the first quadrant it develops a larger and larger
negative x-component of velocity. So it also moves leftwards toward the
y-axis.
• Eventually it crosses the positive y-axis moving in the −x direction, i.e.
1 Think Poohsticks.
2 This is not such an unreasonable assumption. The flow often changes on a larger time
scale.
3 This is also not an unreasonable approximation.
CHAPTER 2. VECTOR FIELDS 86
to the left.
• As it moves farther into the second quadrant it develops a larger and
larger negative y-component of velocity. So it also moves downwards
toward the x-axis.
• Eventually it crosses the negative x-axis moving in the −y direction, i.e.
straight downward.
• As it moves farther into the third quadrant it develops a larger and larger
positive x-component of velocity. So it also moves rightward towards the
y-axis.
• Eventually it crosses the negative y-axis moving in the +x direction, i.e.
to the right.
In two dimensions we can still use the cross product by the simple expe-
dient of thinking of r0 (t) and v r(t) as three component vectors whose third
components are zero.
A more convenient way to implement the weaker “just parallel” condition,
involves reparametrizing our streamline. Suppose that we are in two dimen-
dy
sions with r0 (t) = dx
dt (t) , dt (t) and v(r) = v1 (r) , v2 (r) and fix some t0 . If
4 We’ll have a more careful discussion of this in the optional §2.2.1.
CHAPTER 2. VECTOR FIELDS 87
dx
is nonzero5 , we can reparametrize the curve (at least near r(t0 )) so as
dt (t0 )
to use x, rather than t as the parameter. To do so, we
• solve x = x(t) for t as a function of x. Call the solution T (x). Then
• the point on the curve which has x-coordinate
x is R(x) = X(x) , Y (x)
with X(x) = x and Y (x) = y T (x) .
Then the condition that R0 (x) = 1, Y 0 (x) is parallel to v R(x) says that
R0 (x) is a scalar multiple ofv R(x) so that there is a nonzero number c(x)
so that R0 (x) = c(x)v R(x) . That is
or equivalently
Y 0 (x) c(x) v2 x, Y (x) v2 x, Y (x)
Y 0 (x) = = =
1 c(x) v1 x, Y (x) v1 x, Y (x)
dx dy dz
⇐⇒ = =
v1 (x, y, z) v2 (x, y, z) v3 (x, y, z)
Let us apply this to two examples, in which the stream lines of the vortex
field of Example 2.1.4 are found by two different methods.
Example 2.2.6 Stream lines for the vortex field using r0 (t) k v(r(t)).
In this example we will find the stream lines for the vortex field, v(x, y) =
Ω − yı̂ı + x̂ of Example 2.1.4, by using the requirement that, on a stream
line, the velocity vector r0 (t) must be parallel to v r(t) . By 2.2.5 one way to
⇐⇒ 1 2
2 Ωx = − 12 Ωy 2 + C 0
2 2
⇐⇒ x + y = C
where C 0 and C = Ω2 C 0 are just arbitrary constants. So the stream lines of the
vortex field are exactly circles centred on the origin.
y
We can come to exactly the same conclusion by using the cross product
formulation of 2.2.4.
dx dy
(t) , (t) , 0 × v1 (r(t)) , v2 (r(t)) , 0 = 0
dt dt
dx dy
⇐⇒ (t)ı̂ı + (t) ̂ × − Ωy(t)ı̂ı + Ωx(t) ̂ = 0
dt dt
dx dy
⇐⇒ Ωx(t) (t) + Ωy(t) (t) k̂ = 0
dt dt
dx dy
⇐⇒ Ωx(t) (t) + Ωy(t) (t) = 0
dt dt
d 1 2 1 2
⇐⇒ Ωx(t) + Ωy(t) =0
dt 2 2
⇐⇒ x(t)2 + y(t)2 = C
CHAPTER 2. VECTOR FIELDS 89
Example 2.2.7 Stream lines for the vortex field using r0 (t) = v(r(t)).
This time we will find the stream lines for the vortex field, v(x, y) = Ω −yı̂ı +x̂
of Example 2.1.4, by using 2.2.3, which is
dx
= −Ωy
dt
dy
= Ωx
dt
We can convert this system of first order linear ordinary differential equations
into a single second order linear constant coefficient differential equation8 , by
2
differentiating the first equation, to get ddt2x = −Ω dy
dt , and then substituting in
the second equation to get
d2 x
+ Ω2 x = 0
dt2
This equation is a special case of the ordinary differential equation treated
in Example A.9.3 of the Appendix A.9, entitled “Review of Linear Ordinary
Differential Equations”. In fact it is exactly (A.9.6) with R = 0, L = C = Ω1 .
So the general solution is (A.9.8) with ρ = 0 and ν = Ω, which is
x(t) = A cos(Ωt − θ)
2.2.1 More about r0 (t) × v r(t) = 0
Here is a lemma that gives a more precise version of “if we only care about the
curve traced out by the stick, and not about when the stick is at each point of
the path, then it suffices to impose the weaker condition r0 (t) × v r(t) = 0”.
Lemma 2.2.8 Lat a < b and let v(r) be a vector field. Assume that, for
all a < u < b, R(u) is defined, both R0 (u) and v R(u) are continuous and
nonzero and
R0 (u) × v R(u) = 0
Proof. As R0 (u) × v R(u) = 0 and both R0 (u) and v R(u) are nonzero,
0
R (u)·v(R(u))
This a(u) = v(R(u))·v(R(u)) is necessarily nonzero and continuous. Since a(u)
is nonzero and continuous, it never changes sign. That is, either a(u) > 0 for
all u, or a(u) < 0 for all u. Let T (u) be an antiderivative of a(u). Then T (u)
is strictly monotone (and continuous) and hence is invertible. That is, there
is a continuous function U (t) that obeys U T (u) = u for all a < u < b and
8 In Example 2.1.7 we converted a second order ordinary differential equation into a system
of first order ordinary differential equations. We are now just reversing the procedure we
used there.
9 Even if you don’t know how x(t) = A cos(Ωt − θ) was arrived at, you should be able to
1
r0 (t) = R0 U (t) U 0 (t) = a U (t) v R U (t)
T 0 U (t)
1
= a U (t) v r(t)
a U (t)
= v r(t)
x0 (t) = 0 y 0 (t) = 0
which forces both x(t) and y(t) to be constants. So each field line is just a single
point. On the other hand every nonconstant R(u) obeys R0 (u) × v R(u) = 0
but is not contained in a field line. (As R(u) is not constant, it covers more
than one point, while each field line is just a single point.)
Now here is a more interesting example.
Example 2.2.10 Consider the vector field v(x, y) = xı̂ı. This vector field
takes the value 0 at each point on the y-axis, is a positive multiple of ı̂ı at every
point of the right half-plane and is a negative multiple of ı̂ı at every point of
the left half-plane. Let’s find the field lines. Any field line must obey
dx dy
(t) = x(t) (t) = 0
dt dt
So y(t) must be a constant. We can solve the linear ordinary differential equa-
tion dx
dt (t) = x(t) by moving the x(t) to the left hand side, and multiplying by
the (integrating factor) e−t . This gives
dx
e−t (t) − e−t x(t) = 0
dt
By the product rule, this is the same as
d −t
e x(t) = 0
dt
which forces e−t x(t) to be a constant. So our field lines are Cet , D , with C
in the left half-plane. It is illustrated by the blue line in the figure below
(with a different value of D than for the red line).
y
But as u runs from −∞ to +∞, R(u) runs over the full line (x, D) − ∞ <
x < ∞ . It is not contained in any single field line and, in fact, completely
covers three different field lines.
2.2.2 Exercises
Exercises — Stage 1
1. Suppose that the vector field v(x, y) sketched below represents the
velocity of moving water at the point (x, y) in the first quadrant of
the xy-plane.
y
3
x
1 2 3
Sketch the path followed by a rubber ducky dropped in at the
point
a (0, 2)
b (1, 0)
c (1, 2)
CHAPTER 2. VECTOR FIELDS 92
is a field line.
Exercises — Stage 2
3. ∗. Consider the function f (x, y) = xy.
a Explicitly determine the field lines (flow lines) of F(x, y) = ∇ f .
b Sketch the field lines of F and the level curves of f in the same
diagram.
x
4. ∗. Find the field line of the vector field F = 2y ı̂ı + y2 ̂ + ey k̂ that
passes through (1, 1, e).
5. ∗. Find and sketch the field lines of the vector field F = xı̂ı + 3y ̂.
footnote.
CHAPTER 2. VECTOR FIELDS 93
dv
=⇒ m v(t) · (t) = v(t) · ∇ ϕ x(t), y(t), z(t)
dt
∂ϕ ∂ϕ
= x0 (t) x(t), y(t), z(t) + y 0 (t)
x(t), y(t), z(t)
∂x ∂y
∂ϕ
+ z 0 (t)
x(t), y(t), z(t)
∂z
d
Next use dt v · v = 2v · dv
dt on the left hand side and the chain rule on the right
hand side.
d 1 d
=⇒ mv(t) · v(t) = ϕ x(t), y(t), z(t)
dt 2 dt
d 1
=⇒ mv(t) · v(t) − ϕ x(t), y(t), z(t) = 0
dt 2
1
m|v(t)|2 − ϕ x(t), y(t), z(t) = const
=⇒
2
So 21 m|v(t)|2 − ϕ x(t), y(t), z(t) , which is called the energy2 of the particle
1
ϕ x(t), y(t), z(t) = m|v(t)|2 − E ≥ −E
2
So even without having to find
x(t) ,y(t) , z(t) , we know
that our particle
can never escape the region (x, y, z) ϕ(x, y, z) ≥ −E .
Example 2.3.4 Gravity. The gravitational force that a body of mass M at
the origin exerts on a body of mass m at r = (x, y, z) is
GM m
F(r) = − r
r3
p
where r = |r| = x2 + y 2 + z 2 and G is the gravitational constant. This force
is conservative, with potential ϕ(r) = GMr m . To verify that this is correct,
observe that
∂ ∂ GM m 1 GM m(2x) GM m
ϕ(r) = =− =− x
r3
p 2 2 2 3/2
∂x ∂x x + y + z
2 2 2 2 [x + y + z ]
∂ ∂ GM m 1 GM m(2y) GM m
ϕ(r) = =− =− y
r3
p
∂y ∂y x2 + y 2 + z 2 2 [x2 + y 2 + z 2 ]3/2
∂ ∂ GM m 1 GM m(2z) GM m
ϕ(r) = =− =− z
r3
p
∂z ∂z x2 + y 2 + z 2 2 [x2 + y 2 + z 2 ]3/2
We have already found conservation of energy very helpful a couple of times
in Section 1.7 (Sliding on a Curve). So, at this point, there are probably several
questions gnawing away at you.
• Is every vector field conservative?
• If not, is there an easy way to tell whether or not a vector field is con-
servative?
2 1 m|v(t)|2 is the kinetic energy and −ϕ is the potential energy. See Warning 2.3.2.
2
CHAPTER 2. VECTOR FIELDS 94
x2
ϕ(x, y) = + ψ(y)
2
For this to also satisfy the second equation, we need
∂ϕ ∂ x2
−y = (x, y) = + ψ(y) = ψ 0 (y)
∂y ∂y 2
which is the case if and only if there is a constant C with
y2
ψ(y) = − +C
2
So, for any choice of the constant C,
x2 y2
− +C
2 2
is a potential. In particular, taking C = 0, one possible potential is
x2 y2
ϕ(x, y) = −
2 2
Some equipotential curves for this potential are sketched in (c) below.
They are the blue curves.
b The field lines (Optional): Recalling (2.2.5), the field lines of the vector
field F(x, y) = xı̂ı − y ̂ are determined by
dx dy
= ⇐⇒ −ydx = xdy
x −y
⇐⇒ xdy + ydx = 0
⇐⇒ d(xy) = 0 by the product rule
⇐⇒ xy = C
for some constant C. If you are not comfortable with the use of the
product rule above, here is another way to write the same computation.
dy y dy
= − ⇐⇒ x = −y
dx x dx
CHAPTER 2. VECTOR FIELDS 95
dy
⇐⇒ x +y =0
dx
d
⇐⇒ (xy) = 0 by the product rule
dx
⇐⇒ xy = C
Some field lines are sketched in (c) below. They are the red curves. Note
that they appear to cross the equipotential curves, the blue curves, at
right angles. We shall see in Lemma 2.3.6, below, that this is not a
coincidence. Also note that, while the above computation tells what the
field lines are, it does not give us the direction of motion along the field
lines. We determine the direction of motion next.
c Direction of motion (Optional): The sign data
> 0
if x > 0
ı̂ı · F(x, y) = x = 0 if x = 0
< 0 if x < 0
> 0 if y < 0
̂ · F(x, y) = −y = 0 if y = 0
< 0 if y > 0
is visually displayed in the figure on the left below. The arrows in the
figure on the left gives us the direction of motion along the field lines (in
red) in the figure on the right below. Some equipotential curves are also
sketched (in blue) in the figure on the right below.
y
We have just seen one example of a conservative vector field for which the
field lines appear to cross the equipotential curves at right angles. Here is a
result which says that that was no accident. The field lines of conservative
fields always cross the equipotential surfaces at right angles.
Lemma 2.3.6 Optional. If F is a conservative vector field, then the field
lines of F are perpendicular to the equipotential surfaces of F.
Proof. Let F = ∇ ϕ. Pick any point r0 and any nonzero vector T that is tangent
to the equipotential surface at r0 . That equipotential
surface is ϕ x, y, z =
ϕ(r0 ). Consider any curve r(t) = x(t), y(t), z(t) that
• lies in the equipotential surface of F through r0 , so that ϕ r(t) = ϕ(r0 )
for all t, and also obeys
CHAPTER 2. VECTOR FIELDS 96
• r(0) = r0 and
dr
• dt (0) = T.
Differentiating ϕ r(t) = ϕ(r0 ) with respect to t and applying the chain rule
gives
d
ϕ x(t), y(t), z(t) = 0
dt
or
∂ϕ dx ∂ϕ dy
x(t), y(t), z(t) (t) + x(t), y(t), z(t) (t)
∂x dt ∂y dt
∂ϕ dz
+ x(t), y(t), z(t) (t) = 0
∂z dt
∂ϕ ∂ϕ ∂ϕ
Notice that the left hand side is exactly the dot product of ∂x , ∂y , ∂z = ∇ϕ
dy dz
with dx
dr
dt , dt , dt = dt . So
dr
∇ ϕ r(t) · (t) = 0
dt
dr
F r(t) · (t) = 0
dt
Then set t = 0 to get
F r0 · T = 0
This says that the vector T, which is tangent to the equipotential surface at
r0 , is perpendicular to the vector field at r0 , which is a tangent vector to the
field line of F through r0 .
Here is another example in which we try to find a potential for a vector
field.
Example 2.3.7 Let’s try to find a potential for the vortex vector field v(x, y) =
Ω − yı̂ı + x̂ of Example 2.1.4. The potential would have to obey
∂ϕ
(x, y) = −Ωy
∂x
∂ϕ
(x, y) = Ωx
∂y
We proceed just as we did in Example 2.3.5. The first of these equations is
satisfied if and only if there is a ψ(y), which does not depend on x, so that
The determinant in the second row is really just a mnemonic device used to
make it easy to remember the expression after the equals sign in the first row.
One must be careful about the signs in this definition — the determinant helps
with that. ♦
Theorem 2.3.9 Screening test for conservative vector fields..
a Assume that F1 (x, y) and F2 (x, y) are continuously differentiable. If the
vector field F1 (x, y)ı̂ı + F2 (x, y)̂ is conservative, then we must have
∂F1 ∂F2
=
∂y ∂x
b Assume that F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z) are continuously dif-
ferentiable. If the vector field F1 (x, y, z)ı̂ı + F2 (x, y, z)̂ + F3 (x, y, z)k̂ is
conservative, then
∂F1 ∂F2 ∂F1 ∂F3 ∂F2 ∂F3
= = =
∂y ∂x ∂z ∂x ∂z ∂y
Equivalently,
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂ = 0
∂y ∂z ∂z ∂x ∂x ∂y
That is, F is curl free.
Proof. (a) If the vector field F1 (x, y)ı̂ı + F2 (x, y)̂ is conservative, then there is
a potential ϕ(x, y) such that
∂ϕ
(x, y) = F1 (x, y)
∂x
∂ϕ
(x, y) = F2 (x, y)
∂y
∂ ∂
Applying ∂y to the first equation and ∂x to the second equation gives
∂2ϕ ∂F1
=
∂y∂x ∂y
2
∂ ϕ ∂F2
=
∂x∂y ∂x
2 2
∂ ϕ ∂ ϕ
Recall that, for any twice continuously differentiable function, ∂y∂x = ∂x∂y .
So the two left hand sides are equal, and the two right hand sides must also be
equal.
(b) If the vector field F1 (x, y, z)ı̂ı + F2 (x, y, z)̂ + F3 (x, y, z)k̂ is conservative,
CHAPTER 2. VECTOR FIELDS 98
∂ϕ
(x, y, z) = F1 (x, y, z)
∂x
∂ϕ
(x, y, z) = F2 (x, y, z)
∂y
∂ϕ
(x, y, z) = F3 (x, y, z)
∂z
We proceed just as in (a).
∂ ∂
• Applying ∂y to the first equation and ∂x to the second equation gives
∂2ϕ ∂F1
∂y∂x = ∂y
∂F1 ∂F2
∂2ϕ ∂F2
=⇒ =
= ∂y ∂x
∂x∂y ∂x
∂ ∂
• Applying ∂z to the first equation and ∂x to the third equation gives
∂2ϕ
( )
∂F1
∂z∂x = ∂z ∂F1 ∂F3
∂2ϕ ∂F3
=⇒ =
∂x∂z = ∂x
∂z ∂x
∂ ∂
• Applying ∂z to the second equation and ∂y to the third equation gives
∂2ϕ ∂F2
∂z∂y = ∂z
∂F2 ∂F3
∂2ϕ ∂F3
=⇒ =
= ∂z ∂y
∂y∂z ∂y
2.3.9 does not guarantee that F is conservative. In fact there are fields that
are not conservative but do obey ∂F ∂F2
∂y = ∂x . We’ll see one in Example 2.3.14,
1
below. We shall later find some additional regularity conditions which, when
combined with ∂F ∂F2
∂y = ∂x , do imply conservativeness. See Theorem 2.4.8,
1
below.
Example 2.3.11 Example 2.3.7 revisited. In Example 2.3.7, we attempted
to find a potential for the vector field
v(x, y) = Ω − yı̂ı + x̂
In the end we showed that, if Ω 6= 0, no potential could exist, i.e. v(x, y) is not
conservative. Had we known the screening test of Theorem 2.3.9.a, we could
have concluded that v(x, y) is not conservative by simply observing that
∂v1 ∂
= − Ωy)= −Ω
∂y ∂y
∂v2 ∂
= Ωx) = +Ω
∂x ∂x
are not equal, unless Ω = 0. But Ω = 0 makes a rather boring vector field.
3 Use your favourite search engine to look up a list of common logical errors. One is
So F passes the screening test. Let’s look for a function ϕ(x, y, z) obeying
∂ϕ
(x, y, z) = y 2 + 2xz 2 − 1
∂x
∂ϕ
(x, y, z) = (2x + 1)y (∗)
∂y
∂ϕ
(x, y, z) = 2x2 z + z 3
∂z
∂
The partial derivative ∂x treats y and z as constants. So ϕ(x, y, z) obeys the
first equation if and only if there is a function ψ(y, z) with
ϕ(x, y, z) = xy 2 + x2 z 2 − x + ψ(y, z)
This ϕ(x, y, z) will also obey the second equation if and only if
∂
xy 2 + x2 z 2 − x + ψ(y, z) = (2x + 1)y
∂y
∂ψ
⇐⇒ 2xy + (y, z) = (2x + 1)y
∂y
∂ψ
⇐⇒ (y, z) = y
∂y
y2
⇐⇒ ψ(y, z) = + ζ(z)
2
for some function ζ(z) which depends only on z. At this stage we know that
y2
ϕ(x, y, z) = xy 2 + x2 z 2 − x + ψ(y, z) = xy 2 + x2 z 2 − x + + ζ(z)
2
CHAPTER 2. VECTOR FIELDS 100
obeys the first two equations of (∗), for any function ζ(z). Finally to have the
third equation of (∗) also satisfied, we also need to chose ζ(z) to obey
y2
∂ 2 2 2
xy + x z − x + + ζ(z) = 2x2 z + z 3
∂z 2
⇐⇒ 2x2 z + ζ 0 (z) = 2x2 z + z 3
⇐⇒ ζ 0 (z) = z 3
z4
⇐⇒ ζ(z) = +C
4
for any constant C. So one possible potential, namely that with C = 0, is
y2 z4
ϕ(x, y, z) = xy 2 + x2 z 2 − x + +
2 4
Note, as a check4 , that
as desired.
y x
Example 2.3.14 Optional: First look at + − x2 +y ı
2 ı̂
Now is a
x2 +y 2 ̂ .
good time to reread Warning 2.3.10. In this example we will show that the
vector field
y x
F(x, y) = − ı̂ı + 2 ̂
x2 + y 2 x + y2
defined for all (x, y) in R2 except (x, y) = (0, 0)
passes the screening test of Theorem 2.3.9.a. We will also begin to see why it
is not conservative on the domain R2 \ {(0, 0)}. To verify the screening test,
we compute
∂ y (x2 + y 2 ) − y(2y) y 2 − x2
− 2 2
=− 2 = 2
∂y x +y (x2 + y 2 ) (x2 + y 2 )
∂ x (x2 + y 2 ) − x(2x) y 2 − x2
= 2 = 2
∂x x2 + y 2 (x2 + y 2 ) (x2 + y 2 )
and observe that the two right hand sides are identical. So the screening test
is passed.
In order for F to be conservative on the domain R2 \ {(0, 0}, there must
exist a function ϕ(x, y), that, together with both partial derivatives ∂ϕ ∂x (x, y)
and ∂ϕ
∂y (x, y), is defined for all (x, y) in R 2
except (x, y) = (0, 0), and obeys
∂ϕ y − xy2 ∂ y
(x, y) = − 2 = 2 = arctan
∂x x + y2 1 + xy ∂x x
1
∂ϕ x x ∂ y
(x, y) = 2 2
= 2 = arctan
∂y x +y 1 + xy ∂y x
It looks like we have found a potential, namely arctan xy . But there is a problem.
Recall that, by definition, arctan xy is an angle θ(x, y) that obeys tan θ(x, y) =
arctan xy ; but for any (x, y) ∈ R2 \ {(0, 0} there are infinitely many angles
having the tangent xy . To define ϕ(x, y) we have to select exactly one such
angle. It is impossible to do so in such a way that ϕ(x, y) is continuous on all
of R2 \ {(0, 0}.
To see why, fix any r > 0, and imagine that you are walking on the circle
x2 + y 2 = r2 in the xy-plane. At time θ, you are at x = r cos θ, y = r sin θ and
then xy = tan θ and you are allowed to define ϕ(x, y) = θ + kπ, for any integer
k.
Suppose that at time θ = 0 you choose k = 0. That is, you choose ϕ(r, 0) =
0. Now start walking, choosing an allowed ϕ(x, y), i.e. choosing a k, for each
point (x, y) that you cross. Because ϕ(x, y) has to vary continuously5 with
(x, y), you have to continue choosing k = 0. But you run off a cliff as θ
approaches 2π, because then
• you are approaching (r, 0) from below, as in the figure below, and
• because you are choosing k = 0, ϕ(x, y) is just a little less than 2π, but
• you have already chosen ϕ(r, 0) = 0, not 2π. So ϕ(x, y) has a jump
discontinuity6 along the positive x-axis.
y
ϕ “ 5π{8 ϕ “ 3π{8
ϕ “ 6π{8 ϕ “ 2π{8
ϕ “ 7π{8 ϕ “ π{8
ϕ“π ϕ“0
x
ϕ “ 9π{8 ϕ “ 15π{8
ϕ “ 10π{8 ϕ “ 14π{8
ϕ “ 11π{8 ϕ “ 13π{8
If you are having trouble following this argument, don’t worry about it. We
will return with a less hand-wavy argument later.
2.3.1 Exercises
Exercises — Stage 1
1. We’ve seen two calculations of the energy E of a system. Equa-
tion 1.7.1 told us E = 12 m|v| 2
+ mgy, while Example 2.3.3 says
1 2
2 m|v(t)| − ϕ x(t), y(t), z(t) = E.
Consider a force given by F = ∇ ϕ for some differentiable function
ϕ : R3 → R. A particle of mass m is being acted on by F and no
other forces, and its position at time t is given by (x(t), y(t), 0).
True or false: mgy(t) = −ϕ(x(t), y(t), 0).
2. For each of the following fields, decide which of the following holds:
A The screening test for conservative vector fields tells us F is
conservative.
5 If ϕ(x, y) is not continuous, its gradient does not exist, and ϕ cannot be a potential.
6 Those who have taken some complex analysis may recognize this as the branch cut in
ln z.
CHAPTER 2. VECTOR FIELDS 102
C The screening test for conservative vector fields does not tell us
whether F is conservative or not.
(The screening test is Theorem 2.3.9.)
a F = xı̂ı + ẑ + y k̂
b F = y 2 zı̂ı + x2 ẑ + x2 y k̂
1
c F = (yexy + 1)ı̂ı + (xexy + z)̂ + + y k̂
z
d F = y cos(xy)ı̂ı + x sin(xy)̂
3. Suppose F is conservative and let a, b, and c be constants. Find a
potential for F + (a, b, c), OR give a conservative field F and constants
a, b, and c for which F + (a, b, c) is not conservative.
4. Prove, or find a counterexample to, each of the following statements.
a If F is a conservative field and G is a non-conservative field,
then F + G is non-conservative.
b If F and G are both non-conservative fields, then F + G is non-
conservative.
Exercises — Stage 2
5. ∗. Let D be the domain consisting of all (x, y) such that x > 1, and
let F be the vector field
y x
F=− ı̂ı + 2 ̂
x2 + y 2 x + y2
Is F conservative on D? Give reasons for your answer.
6. Find a potential ϕ for F(x, y) = (x + y)ı̂ı + (x − y)̂, or prove none
exists.
7. Find a potential ϕ for F(x, y) = x1 − y1 ı̂ı + yx2 ̂, or prove none
exists.
Find a potential ϕ for F(x, y, z) = x2 yz + xz ı̂ı + 13 x3 z + y ̂ +
8.
1 3 1 2
3 x y + 2 x + y k̂, or prove none exists.
9. Find a potential ϕ(x, y, z) for
x y z
F(x, y, z) = ı̂ı+ 2 ̂+ 2 k̂,
x2 + y 2 + z 2 x + y2 + z2 x + y2 + z2
or prove none exists.
10. Determine whether or not each of the following vector fields are con-
servative. Find the potential if it is.
a F(x, y, z) = xı̂ı − 2ŷ + 3z k̂
xı̂ı − ŷ
b F(x, y) =
x2 + y 2
CHAPTER 2. VECTOR FIELDS 103
2 2
11. Let F = e(z ) ı̂ı + 2Byz 3 ̂ + Axze(z )
+ 3By 2 z 2 k̂.
a For what values of the constants A and B is the vector field F
conservative on R3 ?
Exercises — Stage 3
12. Find the velocity field for a two dimensional incompressible fluid when
there is a point source of strength m at the origin. That is, fluid is
emitted from the origin at area rate 2πm cm2 /sec. Show that this
velocity field is conservative and find its potential.
13. A particle of mass 10 kg moves in the force field F = ∇ ϕ, where
ϕ(x, y, z) = −(x2 + y 2 + z 2 ). When its potential energy is 0, the
particle is at the origin, and it moves with a velocity 2 m/s.
Following Example 2.3.3, give a region the particle can never es-
cape.
14. A particle√with constant mass m = 1/2 moves under a force field
F = ̂ + 3 3 z k̂. At position (0, 0, 0), its speed is 1. What is its speed
at (1, 1, 1)?
(You may assume without proof that the particle does indeed reach
the point (1, 1, 1).)
15. For some differentiable, real-valued functions f, g, h : R → R, we
define
F = 2f (x)f 0 (x)ı̂ı + g 0 (y)h(z)̂ + g(y)h0 (z)k̂.
Verify that F is conservative.
16. Describe the region in R3 where the field
F = xy, xz, y 2 + z
has curl 0.
y
P1
C3
C2
C1
P0 C2 x
R
We shall calculate the work Ci
F · dr for each of the curves.
a We parametrize C1 by r(t) = tı̂ı + t ̂ with t running from 0 to 1. Then
x(t) = t and y(t) = t so that
dr
F r(t) = t2 ı̂ı + (t2 + 1) ̂
and (t) = ı̂ı + ̂
dt
so that
Z Z 1 dr
Z 1
t2 ı̂ı + (t2 + 1) ̂ · [ı̂ı + ̂] dt
F · dr = F r(t) · (t) dt =
C1 0 dt 0
Z 1
2
= 2t + 1 dt
0
5
=
3
b We split C2 into two parts, C2,x running from P0 to (1, 0) along the x-
axis and then C2,y running from (1, 0) to P1 along the line x = 1. We
parametrize C2,x by r(x) = xı̂ı with x running from 0 to 1 and C2,y by
r(y) = ı̂ı + y ̂ with y running from 0 to 1. Then4
Z Z Z
F · dr = F · dr + F · dr
C2 C2,x C2,y
ı̂ı
z }| {
Z 1 d
2
= (x)(0)ı̂ı + (0 + 1) ̂ · xı̂ı dx
0 dx
̂
z }| {
Z 1 d
2
+ (1)(y)ı̂ı + (y + 1) ̂ · ı̂ı + y ̂ dy
0 dy
Z 1 Z 1
y 2 + 1 dy
= 0 dx +
0 0
4
=
3
3 The reader should check that this vector field is not conservative.
4 You might like to think about why we can split up the integral like this.
CHAPTER 2. VECTOR FIELDS 107
You could evaluate these integrals using double angle trig identities like
you did in first year calculus. But there is a sneaky, much easier, way.
Because sin2 t and cos2 t are translates
R 2π of each other,
R 2πand both are peri-
odic of period π, the two integrals 0 sin2 t dt and 0 cos2 t dt represent
the same area and so are equal. See the figure below.
y y
y “ sin2 x 1
y “ cos2 x
1
x x
π{2 π 3π{2 2π π{2 π 3π{2 2π
Thus
Z 2π Z 2π Z 2π
2 2 1 2
sin t + cos2 t dt
sin t dt = cos t dt =
0 0 0 2
Z 2π
1
= dt = π
2 0
and
I Z 2π Z 2π
2
F · dr = −2 sin t dt + 3 cos2 t dt = π
C1 0 0
dr
R
b We parametrize C2 by r(t) = ı̂ı for all t. Then dt (t) = 0 and C2
F·dr = 0.
Again, despite theR fact that C1 Rand C2 both start at (1, 0) and end at (1, 0),
the two integrals C1 F · dr and C2 F · dr are different.
Example 2.4.5 Example 2.3.14, again. In Example 2.3.14, we saw that
the vector field
y x
F(x, y) = − 2 ı̂ı + 2 ̂
x + y2 x + y2
defined for all (x, y) in R2 except (x, y) = (0, 0)
passed the screening test of Theorem 2.3.9.a, and yet was not conservative.
In this example, we will see that this F violates the conclusion of Theorem
2.4.2, thereby providing a second proof that F(x, y) is not conservative on
5 Again, the reader should verify that this vector field is not conservative.
CHAPTER 2. VECTOR FIELDS 108
R2 with (0, 0) removed. For the curve C, of Theorem 2.4.2, we use the circle
parametrized by x = a cos θ, y = a sin θ, 0 ≤ θ ≤ 2π. Then dx = −a sin θ dθ
and dy = a cos θ dθ so that
Z 2π 2 Z 2π
x dy − y dx a cos2 θ dθ + a2 sin2 θ dθ
Z
1 1 1
= = dθ
2π C x2 + y 2 2π 0 a2 cos2 θ + a2 sin2 θ 2π 0
=1
So, if F were conservative with potential ϕ, Theorem 2.4.2 would give that
x dy − y dx
Z
1
= ϕ(P1 ) − ϕ(P0 ) = 0
2π C x2 + y 2
for any path C from P0 to P1 . The point of this section is that there is an
intimate relation between path independence and conservativeness of vector
fields, that we will get to in Theorem 2.4.7.
For simplicity, we will consider only vector fields that are defined and con-
tinuous on all of R2 (i.e. the xy-plane) or R3 (i.e. the usual three dimensional
world). Some discussion about what happens for vector fields that are defined
only on part of R2 or R3 is given in the optional §4.5.
First we show that if there is one pair of (not necessarily distinct) points
P0 , P1 such that Z Z
F · dr = F · dr
C1 C2
for all curves C1 , C2 that start at P0 and end at P1 , then it is also true that,
for any other pair of points P00 , P10
Z Z
F · dr = F · dr
C10 C20
for all curves C10 , C20 that start at P00 and end at P10 . This might seem unlikely
at first, but the idea of the proof is really intuitive.
CHAPTER 2. VECTOR FIELDS 109
Theorem 2.4.6 Let F be a vector field that is defined and continuous on all
of R2 (or R3 ). Let P0 , P1 , P00 , P10 be any four points in R2 (or R3 ). Assume
that Z Z
F · dr = F · dr
C1 C2
for all curves C10 , C20 that start at P00 and end at P10 .
Proof. Let C10 and C20 be any two curves that start at P00 and end at P10 .
P1
C11
P01 Cr
Cℓ P11
P0 C21
We start by choosing any two (auxiliary) curves
as desired.
We are now ready for our main theorem on conservative fields.
Theorem 2.4.7 Let F be a vector field that is defined and continuous on all
of R2 (or R3 ). Then the following three statements are equivalent.
a F is conservative. That is, there exists a function ϕ such that F = ∇ ϕ.
b The integral C F · dr = 0 for any closed curve C.
H
c The integral F · dr Ris path independent. That is, for any points P0 , P1
R
end at P1 .
That is, if any one of the three statements are true, then all three are true.
CHAPTER 2. VECTOR FIELDS 110
ru (t) = x + tu 0≤t≤1
where Cx + Dsu is the curve which first follows Cx from the origin to x and
then follows Dsu from x to x + su. We have
Z Z Z
F · dr = F · dr + F · dr
Cx +Dsu Cx Dsu
Z Z1
= F · dr + F(x + tsu) · (su) dt
Cx 0
In the second integral, make the change of variables τ = ts, dτ = sdt. This
gives Z Z s
ϕ(x + su) = F · dr + F(x + τ u) · u dτ
Cx 0
By the fundamental theorem of calculus, applied to the second integral,
d
ϕ(x + su) = F(x + su) · u = F(x) · u
ds s=0 s=0
CHAPTER 2. VECTOR FIELDS 111
This ϕ(x, y) will also obey the second equation if and only if
∂ϕ
F2 (x, y) = (x, y)
∂y
∂ x
Z
= F1 (X, y) dX + ψ(y)
∂y
Z x 0
∂F1
= (X, y) dX + ψ 0 (y)
0 ∂y
So we have to find a ψ(y) that obeys
Z x
∂F1
ψ 0 (y) = F2 (x, y) − (X, y) dX
0 ∂y
This looks bad — no matter what ψ(y) is, the left hand side is independent
of x, while it looks like the right hand side depends on x. Fortunately our
CHAPTER 2. VECTOR FIELDS 112
screening test hypothesis now rides in to the rescuea . (We haven’t used it yet,
and it has to come in somewhere.)
Z x Z x
∂F1 ∂F2
F2 (x, y) − (X, y) dX = F2 (x, y) − (X, y) dX
0 ∂y 0 ∂x
X=x
= F2 (x, y) − F2 (X, y)
X=0
= F2 (0, y)
In going from the first line to the second line we used the fundamental theorem
of calculus. So choosing
Z y
ψ(y) = F2 (0, Y ) dY + C
0
2.4.2 Exercises
Exercises — Stage 1
1. Evaluate C x2 y 2 dx + x3 y dy counterclockwise around the square with
R
a F = xı̂ı + ẑ + y k̂
b F = y 2 zı̂ı + x2 ẑ + x2 y k̂
P1
A B
P2
2
R
R F is a conservative vector field in R with A F · dr = 5.
Suppose
What is B F · dr?
5. ∗. Let F(x, y, z) = ex sin y ı̂ı + Raex cos y + bz ̂ + cx k̂. For which
values of the constants a, b, c is C F · dr = 0 for all closed paths C?
6. Consider the four vector fields sketched below. Exactly one of those
vector fields is conservative. Determine which three vector fields are
not conservative and explain why.
(a) y (b) y
(c) y (d) y
x x
7. ∗. Consider the vector field
x − 2y 2x + y
F(x, y, z) = ı̂ı + 2 ̂ + z k̂
x2 + y 2 x + y2
d Is F conservative?
R
8. Find the work, C F·dr, done by the force field F = (x+y)ı̂ı +(x−z)̂ +
(z − y)k̂ in moving an object from (1, 0, −1) to (0, −2, 3). Does the
work done depend on the path used to get from (1, 0, −1) to (0, −2, 3)?
Exercises — Stage 2
9. Consider the vector field
where
F = x2 ı̂ı + (x − 2y) ̂ + x2 y k̂
12. Let F = 6x2 yz 2 ı̂ı +(2x3 z 2 +2y−xz) ̂ +4x3 yz k̂ and let G = yz ı̂ı +xy k̂.
a For what value of the constant λ is the vector field H = F + λG
conservative on 3-space?
b Find a scalar potential φ(x, y, z) for the conservative field H
referred to in part (a).
R
c Find C F · dr if C is the curve of intersection of the two surfaces
z = x and y = exz from the point (0, 1, 0) to the point (1, e, 1).
13. ∗. Find the work done by the force field F(x, y, z) = (x − y 2 , y −
z 2 , z − x2 ) on a particle that moves along the line segment from
(0, 0, 1) to (2, 1, 0).
x y
14. ∗. Let F = x2 +y ı + x2 +y
2 ı̂ + x3 k̂. Let P be the path which starts
2 ̂
x2 + y 2 = 1 xez = 1
16. ∗. Let C be the upper half of the unit circle centred on (1, 0) (i.e.
that part of the circle whichR lies above the x-axis), oriented clockwise.
Compute the line integral C xy dy.
17. ∗. Show that the following line integral is independent of path and
evaluate the integral.
Z
(yex + sin y) dx + (ex + sin y + x cos y) dy
C
around the triangle with vertices (1, 0, 0), (0, 1, 0), and (0, 0, 1), ori-
ented clockwise as seen from the point (1, 1, 1).
R
19. ∗. Evaluate the line integral C F · dr, where F is the conservative
vector field
conservative?
R
b For those values of α, β and γ found in part (a), calculate C F ·
dr, where C is the curve parametrized by x = t2 , y = et , z = πt,
0 ≤ t ≤ 1.
21. ∗. Consider the vector field F(x, y, z) = (cos x, 2 + sin y, ez ).
a Compute the curl of F.
b Is there a function f such that F = ∇ f ? Justify your answer.
R
c Compute the integral C F · dr along the curve C parametrized
by r(t) = (t, cos t, sin t) with 0 ≤ t ≤ 3π.
22. ∗.
a Consider the vector field
where C is the same curve as in part (c). [Note: the “4” in the
last term is not a misprint!].
Questions 2.4.2.24 and 2.4.2.25 ask you to evaluate line integrals of vector fields
that are not conservative, but that can be expressed as a sum of a conservative
vector field and another vector field that can be written concisely.
24. ∗. Let
a Find all values of A and B for which the vector field F is con-
servative.
conservative?
(b) Let F be a conservative field from part (a). Find all functions
ϕ for which F = ∇ ϕ.
R
(c) Let F be a conservative field from part (a). Evaluate C F · dr
where C is the intersection of y = x and z = ln(1 + x) from
(0, 0, 0) to (1, 1, ln 2).
R
(d) Evaluate C G · dr where
is conservative.
Exercises — Stage 3
CHAPTER 2. VECTOR FIELDS 118
31. Let C be the curve from (0, 0, 0) to (1, 1, 1) along the intersection of
the surfaces y = x2 and z = x3 .
R
a Find C ρ ds if s is arc length along C and ρ = 8x + 36z.
R
b Find C F · dr if F = sin y ı̂ı + (x cos y + z) ̂ + (y + z) k̂.
32. ∗. The curve C is the helix that winds around the cylinder x2 +y 2 = 1
(counterclockwise, as viewed from the positive z-axis, looking down on
the xy-plane). It starts at the point (1, 0, 0), winds around the cylinder
once, and ends at the point (1, 0, 1). Compute the line integral of the
vector field
F(x, y, z) = (−y, x, z 2 )
along C.
33. ∗. Evaluate the line integrals below. (Use any method you like.)
a C (x2 +y) dx+x dy, where C is the arc of the parabola y = 9−x2
R
Questions 2.4.2.35 and 2.4.2.36 ask you to find a path that leads to a particular
value of a line integral. Many such paths are possible — you only need to find
one.
35. ∗.
a Consider the vector field F x, y = (3y, x − 1) in R2 . Compute
the line integral Z
F · dr
L
is acted on by a force
F = 4t , 6t2 , −4t
c Find κ(t) the curvature of the path traversed by the particle for
t ≥ 0.
d Find the work done by the force on the particle from t = 0 to
t = T.
43. ∗.√ The position of an airplane at time t is given by x = y =
4 2 3/2
3 t , z = t(2 − t) from take-off at t = 0 to landing at t = 2.
a What is the total distance the plane travels on this flight?
b Find the radius of curvature κ at the apex of the flight, which
occurs at t = 1.
c Two external forces are applied to the plane during the flight:
the force of gravity G = −M g k̂, where M is the mass of the
plane and g is a constant; and a friction force F = −|v|2 v, where
v is the velocity of the plane. Find the work done by each of
these forces during the flight.
d One half-hour later, a bird follows the exact same flight — path
as the plane, travelling at a constant speed v = 3. One can
show
√
that
√
at the apex of the path, i.e. when the bird is at
4 2 4 2
3 , 3 , 1 , the principal unit normal N̂ to the path points in
the −k̂ direction. Find the bird’s (vector) acceleration at that
moment.
̂
θ ı̂ı
ℓ
τ βℓ dθ
dt θ
ℓ cos θ ℓ
mg ℓ sin θ
between the rod and vertical. The forces acting on the mass are
d dθ
v(t) = ` sin θ(t), − cos θ(t) = ` cos θ(t), sin θ(t) (t)
dt dt
and the total frictional force is −β`(cos θ, sin θ) dθ
dt , for some constant β. The
acceleration vector of the mass is
d d2 θ dθ 2
a(t) = v(t) = `(cos θ, sin θ) 2 + `(− sin θ, cos θ)
dt dt dt
so that Newton’s law of motion, F = ma, now tells us
d2 θ dθ 2
ma(t) = m`(cos θ, sin θ) 2
+ m`(− sin θ, cos θ)
dt dt
dθ
= F = mg(0, −1) + τ (− sin θ, cos θ) − β`(cos θ, sin θ)
dt
To eliminate the (unknown) coefficient τ we dot this equation with (cos θ, sin θ),
which extracts the component parallel to the direction of motion of the mass.
2
Dotting with (cos θ, sin θ) gives m` ddt2θ = −mg sin θ − β` dθ
dt or
d2 θ β dθ g
2
+ + sin θ = 0
dt m dt `
which is the equation of motion of the (nonlinear) pendulum. In general, it
can be hard to analyse nonlinear differential equations. But if the amplitude
of oscillation is small enough that we may approximate sin θ by θ we get the
equation of motion of the linear pendulum2 which is
d2 θ β dθ g
2
+ + θ=0
dt m dt `
These equations may be reformulated as systems of first order ordinary differ-
ential equation, that is as equations for the flow lines of a vector field, by the
simple expedient of defining (as we did in Example 2.1.7)
The solutions of this first order system of ordinary differential equations are
flow lines for the vector field
g β
V (x, y) = y , − sin x − y
` m
When β = 0, this is exactly the vector field of Example 2.1.7.
Chapter 3
Surface Integrals
z = f (x, y) (x, y) ∈ D ⊂ R2
Here “(x, y) ∈ D ⊂ R2 ” just means that (x, y) runs over the subset D of
R2 . For example, if the surface is the top half of the sphere of radius one
centred on the origin
p
z = 1 − x2 − y 2 with x2 + y 2 ≤ 1
b Implicitly: We can also specify that the surface is the set of points (x, y, z)
that satisfy the equation G(x, y, z) = 0, or, more generally1 , satisfy the
equation G(x, y, z) = K, with K a constant. For example, the sphere of
radius one centred on the origin is the set of points that obey
x2 + y 2 + z 2 = 1
r : D ⊂ R2 → R 3
(u, v) ∈ D 7→ r(u, v) = x(u, v) , y(u, v) , z(u, v)
The upper line means that r is a function which is defined on the subset
D of R2 and which assigns to each point on D a point in R3 . The
second line means that the function r assigns to the element (u, v) of D
the element r(u, v) = x(u, v) , y(u, v) , z(u, v) in R3 . Such a surface is
called a parametrized surface — each point of the surface is labelled by
1 Of course we can always convert the equation G(x, y, z) = K into H(x, y, z) = 0 with
123
CHAPTER 3. SURFACE INTEGRALS 124
Let’s do something a bit more substantial.
Example 3.1.2 Sphere. The sphere of radius 1 centred on the origin is the
set of points (x, y, z) that obey
G(x, y, z) = x2 + y 2 + z 2 = 1
We cannot express this surface as the graph of a function because, for each
(x, y) with x2 + y 2 < 1, there are two z’s that obey x2 + y 2 + z 2 = 1, namely
p
z = ± 1 − x2 − y 2
On the other hand, locally, this surface is the graph of a function. This means
that, for any point (x0 , y0 , z0 ) on the sphere, all points of the surface that are
sufficiently near (x0 , y0 , z0 ) can be expressed in one of the forms z = f (x, y) or
x = g(y, z),√or y = h(x, z). For example, the part of the sphere that is within
a distance 2 of the point (0, 0, 1) is
√
(x, y, z) x2 + y 2 + z 2 = 1, |(x, y, z) − (0, 0, 1)| < 2
= (x, y, z) x2 + y 2 + z 2 = 1, x2 + y 2 + z 2 − 2z + 1 < 2
= (x, y, z) x2 + y 2 + z 2 = 1, z > 0
p
= (x, y, z) z = 1 − x2 − y 2 , x2 + y 2 < 1
This is illustrated in the figure below which shows the y = 0 section of the
sphere x2 + y 2 + z 2 =√1 and also the y = 0 section of the set of points that
are within a distance 2 of (0, 0, 1). (They are the points inside the dashed
circle.)
z
x2 ` y 2 ` pz ´ 1q2 “ 2
p0, 0, 1q
p1, 0, 0q x
x2 ` y 2 ` z 2 “ 1
CHAPTER 3. SURFACE INTEGRALS 125
= (x, y, z) x2 + y 2 + z 2 = 1, x2 − 2x + 1 + y 2 + z 2 < 2
= (x, y, z) x2 + y 2 + z 2 = 1, x > 0
p
= (x, y, z) x = 1 − y 2 − z 2 , y 2 + z 2 < 1
p1, 0, 0q x
x `y `z “1
2 2 2
px ´ 1q2 ` y 2 ` z 2 “ 2
z
p0, 0, zq ρ sin ϕ
px, y, zq
ρ cos ϕ
ϕ ρ
y
px, 0, 0q θ
px, y, 0q
ρ sin ϕ sin θ
x
and here are two more figures giving the side and top views of the previous
figure.
CHAPTER 3. SURFACE INTEGRALS 126
z
ρ sin ϕ px, y, zq y
p0, 0, zq
θ ρ sin ϕ
ρ cos ϕ ρ sin ϕ cos θ
ρ
ϕ px, y, 0q
px, 0, 0q
ρ sin ϕ sin θ
px, y, 0q
x
side view top view
From the figure, we see that Cartesian and spherical coordinates are related
by
x = ρ sin ϕ cos θ
y = ρ sin ϕ sin θ
z = ρ cos ϕ
The points on the sphere x2 + y 2 + z 2 = 1 are precisely the set of points with
ρ = 1. So we can use the parametrization
r(θ, ϕ) = sin ϕ cos θ , sin ϕ sin θ , cos ϕ
Here is how to see that as ϕ runs over (0, π) and θ runs over [0, 2π), r(θ, ϕ)
covers the whole sphere x2 + y 2 + z 2 = 1 except for the north pole (ϕ = 0 gives
the north pole for all values of θ) and the south pole (ϕ = π gives the south
pole for all values of θ).
• Fix θ and have ϕ run over the interval 0 < ϕ ≤ π2 . Then r(θ, ϕ) traces
out one quarter of a circle starting at the north pole r(θ, 0) = (0, 0, 1)
(but excluding the north pole itself) and ending at the point r(θ, π2 ) =
(cos θ, sin θ, 0) in the xy-plane.
z
ϕ
y
θ
x
• Keep θ fixed at the same value and extend the interval over which ϕ
runs to 0 < ϕ < π. Now r(θ, ϕ) traces out a semi-circle starting at the
north pole r(θ, 0) = (0, 0, 1), ending at the south pole r(θ, π) = (0, 0, −1)
(but excluding both the north and south poles themselves) and passing
through the point r(θ, π2 ) = (cos θ, sin θ, 0) in the xy-plane.
CHAPTER 3. SURFACE INTEGRALS 127
y
θ
• Finally have θ run over 0 ≤ θ < 2π. Then the semicircle rotates about
the z-axis, sweeping out the full sphere, except for the north and south
poles.
Recall that ϕ is the angle between the radius vector and the z-axis. If you
hold ϕ fixed and increase θ by a small amount dθ, r(θ, ϕ) sweeps out the red
circular arc in the figure on the left below. If you hold ϕ fixed and vary θ from
0 to 2π, r(θ, ϕ) sweeps out a line of latitude. The figure on the right below
gives the lines of latitude (or at least the parts of those lines in the first octant)
π 2π 3π 4π π
for ϕ = 10 , 10 , 10 , 10 and 5π
10 = 2 .
z
z
dθ
ϕ y
y
θ
x
x
On the other hand, if you hold θ fixed and increase ϕ by a small amount
dϕ, r(θ, ϕ) sweeps out the red circular arc in the figure on the left below. If
you hold θ fixed and vary ϕ from 0 to π, r(θ, ϕ) sweeps out a line of longitude.
The figure on the right below gives the lines of longitude (or at least the parts
π 2π 3π 4π π
of those lines in the first octant) for θ = 0, 10 , 10 , 10 , 10 and 5π
10 = 2 .
CHAPTER 3. SURFACE INTEGRALS 128
z
z
ϕ dϕ y
y
θ
x
x
2 2
Example 3.1.3 Cylinder. The surface x + z = 1 is an infinite cylinder.
Part of this cylinder in the first octant is sketched below.
Note that the section of this cylinder that lies in the xz-plane, and in fact
in any plane y = c, is the circle x2 + z 2 = 1. We can of course parametrize
this circle by x = cos θ, z = sin θ. So we can parametrize the whole cylinder
by using θ and y as parameters.
r(θ, y) = cos θ , y , sin θ 0 ≤ θ < 2π, −∞ < y < ∞
Example 3.1.4 Surface of Revolution. In this example, we are going to
parametrize a surface of revolution. In your first integral calculus course, you
undoubtedly encountered many surfaces created by rotating a curve y = f (x)
about the x-axis or the y-axis. In this course, we are used to having the z-axis,
rather than the y-axis, run vertically.
So in this example, we’ll parametrize the surface constructed by rotating
the curve
z = g(y) = ey 0≤y≤1
about the z-axis. Exactly the same procedure can be used to parametrize
surfaces created by rotating about the x-axis or the y-axis too.
We start by just sketching the curve, considering the yz-plane as the plane
x = 0 in R3 . The specified curve is the red curve in the figure below. Concen-
trate on any one point on that curve. It is the blue dot at (0, Y, eY )
2 The symbols ρ, ϕ, θ are the standard mathematics symbols for the spherical coordinates.
Appendix A.7 gives another set of symbols that is commonly used in the physical sciences
and engineering.
CHAPTER 3. SURFACE INTEGRALS 129
p0, Y, eY q
x
in the figure. When our curve is rotated about the z-axis, the blue dot
sweeps out a circle. The circle that the blue dot sweeps out
• lies in the horizontal plane z = eY and
• is centred on the z-axis and
• has radius Y .
We can parametrize the circle swept out in the usual way. Here is a top view
of the circle, with the parameter, named θ, indicated.
p0, Y, eY q
θ y
Y
pY sin θ, Y cos θ, eY q
x top view
The coordinates of the red dot are Y sin θ , Y cos θ , eY . This also gives a
parametrization of the surface of revolution
x(Y, θ) = Y sin θ
y(Y, θ) = Y cos θ
z(Y, θ) = eY
0 ≤ Y ≤ 1, 0 ≤ θ < 2π
runs over the entire desired curve (namely z = g(y), 0 ≤ y ≤ 1), when Y
runs over 0 ≤ Y ≤ 1 and
• for any fixed 0 ≤ Y ≤ 1, x(Y, θ) , y(Y, θ) , z(Y, θ) runs over the circle
x2 + y 2 = Y 2 , in the plane z = eY , when θ runs over 0 ≤ θ < 2π.
CHAPTER 3. SURFACE INTEGRALS 130
?
z “ e x `y
2 2
x
Example 3.1.5 Torus. In this example, we are going to parametrize a donut
(well, its surface), or an inner tube.
The formal mathematical name for the surface of a donut is a torus. Our
strategy will be to first parametrize the section of the torus in the right half of
the yz-plane, and then built up the full torus by rotating the circle about the
z-axis. The section is a circle, sketched below.
CHAPTER 3. SURFACE INTEGRALS 131
r
θ
p0,R,0q
y
We’ll assume that the centre of the circle is a distance R from the z-axis,
and that the circle has radius r. Then the red dot on the circle is at
x=0
y = R + r cos θ
z = r sin θ
In particular the red dot is a distance r sin θ above the xy-plane and is a
distance R + r cos θ from the z-axis. So when we rotate the section about the
z-axis, the red dot sweeps out a circle which is sketched below.
z
y
x
The circle that the red dot sweeps out
• lies in the plane z = r sin θ and
• is centred on the z-axis and
• has radius ρ = R + r cos θ.
We can parametrize the circle swept out in the usual way. Here is a top view
of the circle, with the parameter, named ψ, indicated.
p0, ρ, r sin θq
y
ψ
pρ cos ψ, ρ sin ψ, r sin θq
x top view
CHAPTER 3. SURFACE INTEGRALS 132
So the parametrization of the circle swept out by the red dot, and also the
parametrization of the torus, is
or
3.1.1 Exercises
Exercises — Stage 1
1. Parametrize the surface given by z = ex+1 + xy in terms of x and y.
2. ∗. Let S be the surface given by
r(u, v) = u + v , u2 + v 2 , u − v ,
−2 ≤ u ≤ 2, −2 ≤ v ≤ 2
This is a surface you are familiar with. What surface is it (it may be
just a portion of one of the following)? sphere / helicoid / ellipsoid
/ saddle / parabolic bowl / cylinder / cone / plane
Exercises — Stage 2
3. ∗. Suppose S is the part of the hyperboloid x2 + y 2 − 2z 2 = 1 that
lies inside the cylinder x2 + y 2 = 9 and above the plane z = 1 (i.e. for
which z ≥ 1).
Which of the following are parameterizations of S?
c p
r(u, θ) = u sin θ ı̂ı + u cos θ ̂ + 2 − u2 k̂
0 ≤ u ≤ 2, 0 ≤ θ ≤ 2π
d √ √ √
r(φ, θ) = 2 sin φ cos θ ı̂ı + 2 sin φ sin θ ̂ + 2 cos φ k̂
π
0 ≤ φ ≤ , 0 ≤ θ ≤ 2π
4
e p p
r(φ, z) = − 2 − z 2 sin φı̂ı + 2 − z 2 cos φ ̂ + z k̂
√
0 ≤ φ ≤ 2π, 1 ≤ z ≤ 2
5. ∗. Let S be the part of the paraboloid z + x2 + y 2 = 4 lying between
the planes z = 0 and z = 1. For each of the following, indicate whether
or not it correctly parameterizes the surface S.
a r(u, v) = uı̂ı + v ̂ + (4 − u2 − v 2 ) k̂, 0 ≤ u2 + v 2 ≤ 1
√ √
b r(u, v) = ( 4 − u cos v)ı̂ı + ( 4 − u sin v) ̂ + u k̂, 0 ≤ u ≤ 1,
0 ≤ v ≤ 2π
√
c r(u, v) = (u cos v)ı̂ı + (u sin v) ̂ + (4 − u2 ) k̂, 3 ≤ u ≤ 2, 0 ≤
v ≤ 2π
Exercises — Stage 3
6. ∗. Consider the following surfaces
• S1 is the hemisphere given by the equation x2 + y 2 + z 2 = 4 with
z ≥ 0.
• S2 is the cylinder given by the equation x2 + y 2 = 1.
• S3 is the cone given by the equation z 2 = x2 + y 2 with z ≥ 0.
Consider the following parameterizations:
A
√ √ √
r(θ, φ) = 4 cos θ sin φ , 4 sin θ sin φ , 4 cos φ
CHAPTER 3. SURFACE INTEGRALS 134
0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/6
B
√ √ √
r(θ, φ) = 4 cos θ sin φ , 4 sin θ sin φ , 4 cos φ
0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/4
C
√ √ √
r(θ, φ) = 4 cos θ sin φ , 4 sin θ sin φ , 4 cos φ
0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/3
D
p p
r(θ, z) = 4 − z 2 cos θ , 4 − z 2 sin θ , z
0 ≤ θ ≤ 2π, 1≤z≤2
E
p p
r(θ, z) = 4 − z 2 cos θ , 4 − z 2 sin θ , z
√
0 ≤ θ ≤ 2π, 2≤z≤2
F
p p
r(θ, z) = 4 − z 2 cos θ , 4 − z 2 sin θ , z
√
0 ≤ θ ≤ 2π, 3≤z≤2
G
r(θ, z) = z cos θ , z sin θ , z
0 ≤ θ ≤ 2π, 0≤z≤1
H
r(θ, z) = z cos θ , z sin θ , z
√
0 ≤ θ ≤ 2π, 0 ≤ z ≤ 2
I
r(θ, z) = z cos θ , z sin θ , z
√
0 ≤ θ ≤ 2π, 0 ≤ z ≤ 3
J
p
r(x, y) = x , y , x2 + y 2
x2 + y 2 ≤ 1
K
p
r(x, y) = x , y , x2 + y 2
√
x2 + y 2 ≤ 2
CHAPTER 3. SURFACE INTEGRALS 135
L
p
r(x, y) = x , y , x2 + y 2
x2 + y 2 ≤ 2
For each of the following, choose from above all of the valid parame-
terization of each of the given surfaces. Note that there may be one
or more valid parameterization for each surface, and not necessarily
all of the above parameterizations will be used.
a The part of S1 contained inside S2 :
b The part of S1 contained inside S3 :
c The part of S3 contained inside S2 :
r(t) = (4, 4, t)
z
(2, 2, 4) y
(x, y, z) − (x0 , y0 , z0 ) = (x − x0 , y − y0 , z − z0 )
1 Alternatively, you could find two vectors that are in the plane (and not parallel to each
other), and then construct a normal vector by taking their cross product.
CHAPTER 3. SURFACE INTEGRALS 136
lies entirely in the plane and so is perpendicular to n. This gives the following
very neat the equation for the plane.
pa, b, cq
px, y, zq
n · (x − x0 , y − y0 , z − z0 ) = 0
r : D ⊂ R 2 → R3
(u, v) ∈ D 7→ r(u, v) = x(u, v) , y(u, v) , z(u, v)
Note that none of the normal vectors n above need be of unit length.
Note that if we apply part (c) to G(x, y, z) = z − f (x, y) we get the normal
vector n = ∇ G x0 , y0 , z0 = −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂, which is the same
as the normal vector provided by part (b). Of course they had to be at least
parallel.
CHAPTER 3. SURFACE INTEGRALS 137
is a curve on the surface (the red curve in the figure on the right below) that
passes through (x0 , y0 , z0 ) (the black dot in the figure) when u = u0 .
n
Tv
Tu
∂ ∂x ∂y ∂z
Tu = r(u, v0 ) = (u0 , v0 ) , (u0 , v0 ) , (u0 , v0 )
∂u u=u0 ∂u ∂u ∂u
It is the red arrow in the figure on the right above.
Next fix u = u0 and let v vary. Then
v 7→ r(u0 , v) = x(u0 , v) , y(u0 , v) , z(u0 , v)
is a curve on the surface (the blue curve in the figure on the right above) that
passes through (x0 , y0 , z0 ) when v = v0 . The tangent vector to this curve at
(x0 , y0 , z0 ), which is also a tangent vector to the surface at (x0 , y0 , z0 ), is
∂ ∂x ∂y ∂z
Tv = r(u0 , v) = (u0 , v0 ) , (u0 , v0 ) , (u0 , v0 )
∂v v=v0 ∂v ∂v ∂v
It is the blue arrow in the figure on the right above.
We now have two vectors, namely Tu and Tv , that are tangent to the
surface at (x0 , y0 , z0 ). So their cross product
∂x ı̂ı ̂ k̂
∂y ∂z
n = Tu × Tv = det ∂u (u0 , v0 ) ∂u (u0 , v0 ) ∂u (u0 , v0 )
∂x (u , v ) ∂y (u , v ) ∂z (u , v )
∂v 0 0 ∂v 0 0 ∂v 0 0
∂r
Tx = (x0 , y0 ) = 1 , 0 , fx (x0 , y0 )
∂x
∂r
Ty = (x0 , y0 ) = 0 , 1 , fy (x0 , y0 )
∂y
CHAPTER 3. SURFACE INTEGRALS 138
and
ı̂ı ̂ k̂
n = Tx × Ty = det 1 0 fx (x0 , y0 ) = −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂
0 1 fy (x0 , y0 )
(c) Finally assume that the surface is given implicitly by the equation
G(x, y, z) = 0 or, more generally by the equation, G(x, y, z) = K, where K is
a constant. If r(t) = x(t) y(t) , z(t) is any curve with r(0) = (x0 , y0 , z0 ) that
lies on the surface, then
G r(t) = K for all t
d
=⇒ G x(t), y(t), z(t) = 0 for all t
dt
Applying the chain rule gives
∂G dx ∂G dy
x(t), y(t), z(t) (t) + x(t), y(t), z(t) (t)
∂x dt ∂y dt
∂G dz
+ x(t), y(t), z(t) (t) = 0
∂z dt
The left hand side is exactly the dot product of ∂G ∂G ∂G
∂x , ∂y , ∂z = ∇ G with
dx dy dz
dr
dt , dt , dt = dt , so that
∇ G r(t) · r0 (t) = 0
for all t
0
=⇒ ∇ G x0 , y0 , z0 · r (0) = 0
Of course the two vectors (−x0 , −y0 , z0 ) and −2(−x0 , −y0 , z0 ) are parallel to
each other. Either can be used as a normal vector and the tangent plane to
x2 + y 2 − z 2 = 0 at (x0 , y0 , z0 ) is
0 = n · (x − x0 , y − y0 , z − z0 ) = −x0 (x − x0 ) − y0 (y − y0 ) + z0 (z − z0 )
Example 3.2.3 This time we shall find the tangent planes to the surface
x2 + y 2 − z 2 = 1
and we may take (x0 , y0 , −z0 ) as a normal vector at (x0 , y0 , z0 ). So the tangent
plane to x2 + y 2 − z 2 = 1 at (x0 , y0 , z0 ) is
0 = n · (x − x0 , y − y0 , z − z0 ) = x0 (x − x0 ) + y0 (y − y0 ) − z0 (z − z0 )
This time n = (x0 , y0 , −z0 ) 6= 0, so that we have a tangent plane, at every point
of the surface. In particular, the vanishing of n = (x0 , y0 , −z0 ) at (x0 , y0 , z0 ) =
(0, 0, 0) is not a problem because (0, 0, 0) is not on the surface.
Example 3.2.4 Optional — Parametrizing the Hyperboloid of One
Sheet. The hyperboloid of one sheet, x2 + y 2 − z 2 = 1, has a symmetry. It is
invariant under rotation about the z-axis. So it is natural to parametrize the
surface using cylindrical coordinates.
z
px, y, zq
x = r cos θ
y = r sin θ z
y
z=z r
θ
px, y, 0q
x
2 2 2
√x +y −z =√
In cylindrical coordinates the surface 1 is r2 − z 2 = 1, and
we could parametrize it by r(θ, z) = 1 + z cos θ ı̂ı + 1 + z 2 sin θ ̂ + z k̂.
2
As an exercise in working with hyperbolic trig functions, we’ll use this parametriza-
tion to find n̂.
So
ı̂ı ̂ k̂
n = Tu × Tθ = det sinh u cos θ sinh u sin θ cosh u
− cosh u sin θ cosh u cos θ 0
= − cosh2 u cos θ , − cosh2 u sin θ , sinh u cosh u
3.2.1 Exercises
Exercises — Stage 1
1. Is it reasonable to say that the surfaces x2 + y 2 + (z − 1)2 = 1 and
x2 + y 2 + (z + 1)2 = 1 are tangent to each other at (0, 0, 0)?
2. Let the point r0 = (x0 , y0 , z0 ) lie on the surface G(x, y, z) = 0. Assume
that ∇ G(x0 , y0, z0 ) 6= 0. Suppose that the parametrized curve r(t) =
x(t), y(t), z(t) is contained in the surface and that r(t0 ) = r0 . Show
that the tangent line to the curve at r0 lies in the tangent plane to
G = 0 at r0 .
3. Find the parametric equations of the normal line to the surface z =
f (x, y) at the point x0 , y0 , z0 = f (x0 , y0 ) . By definition, the normal
line in question is the line through (x0 , y0 , z0 ) whose direction vector
is perpendicular to the surface at (x0 , y0 , z0 ).
4. Let F (x0 , y0 , z0 ) = G(x0 , y0 , z0 ) = 0 and let the vectors ∇ F (x0 , y0 , z0 )
and ∇ G(x0 , y0 , z0 ) be nonzero and not be parallel to each other. Find
the equation of the normal plane to the curve of intersection of the
surfaces F (x, y, z) = 0 and G(x, y, z) = 0 at (x0 , y0 , z0 ). By definition,
that normal plane is the plane through (x0 , y0 , z0 ) whose normal vector
is the tangent vector to the curve of intersection at (x0 , y0 , z0 ).
5. Let f (x0 , y0 ) = g(x0 , y0 ) and let (fx (x0 , y0 ), fy (x0 , y0 )) 6= (gx (x0 , y0 ), gy (x0 , y0 )).
Find the equation of the tangent line to the curve of intersection of
the surfaces z = f (x, y) and z = g(x, y) at (x0 , y0 , z0 = f (x0 , y0 )).
Exercises — Stage 2
x2 y
6. ∗. Let f (x, y) = . Find the tangent plane to the surface
x4
+ 2y 2
z = f (x, y) at the point −1 , 1 , 13 .
complex number that obeys i2 = −1. You can verify these formulae by just checking that
cosh u and cos(iu) have the same Taylor expansions and that sinh u and −i sin(iu) have the
same Taylor expansions.
CHAPTER 3. SURFACE INTEGRALS 142
equation of the tangent plane to this surface at the point (−1, 1, 2).
Write your answer in the form z = ax + by + c, where a, b and c are
constants.
9. ∗. A surface is given by
z = x2 − 2xy + y 2 .
x = 2u2
y = v2
z = u2 + v 3
Find an equation for the tangent plane to S at the point (8, 1, 5).
11. ∗. Let S be the surface given by
r(u, v) = u + v , u2 + v 2 , u − v ,
−2 ≤ u ≤ 2, −2 ≤ v ≤ 2
Find the tangent plane to the surface at the point (2, 2, 0).
12. ∗. Find the tangent plane and normal line to the surface z = f (x, y) =
2y
x2 +y 2 at (x, y) = (−1, 2).
Exercises — Stage 3
16. ∗.
a Find a vector perpendicular at the point (1, 1, 3) to the surface
with equation x2 + z 2 = 10.
b Find a vector tangent at the same point to the curve of inter-
section of the surface in part (a) with surface y 2 + z 2 = 10.
c Find parametric equations for the line tangent to that curve at
that point.
17. ∗. Let P be the point where the curve
18. Find all horizontal planes that are tangent to the surface with equation
2
+y 2 )/2
z = xye−(x
Concentrate on any one the small pieces. Here is a greatly magnified sketch.
u“u0 u“u0 `du
v varying v varying
P2 P3 u varying
v“v0 `dv
P1 u varying
P0 v“v0
For example, the lower red curve was constructed by holding v fixed at
the value v0 , varying u and sketching r(u, v0 ), and the upper red curve was
constructed by holding v fixed at the slightly larger value v0 + dv, varying u
and sketching r(u, v0 + dv). So the four intersection points in the figure are
Now if
R(t) = r(u0 + tdU , v0 + tdV )
(where dU and dV are any small constants) then, by Taylor expansion,
r u0 + dU , v0 + dV = R(1)
≈ R(0) + R0 (0) t − 0 t=1
∂r ∂r
= r(u0 , v0 ) + (u0 , v0 ) dU + (u0 , v0 ) dV
∂u ∂v
Applying this three times, once with dU = du, dV = 0, once with dU = 0
dV = dv, and once with dU = du, dV = dv,
P0 = r(u0 , v0 )
∂r
P1 = r(u0 + du, v0 ) ≈ r(u0 , v0 ) +
(u0 , v0 ) du
∂u
∂r
P2 = r(u0 , v0 + dv) ≈ r(u0 , v0 ) + (u0 , v0 ) dv
∂v
∂r ∂r
P3 = r(u0 + du, v0 + dv) ≈ r(u0 , v0 ) + (u0 , v0 ) du + (u0 , v0 ) dv
∂u ∂v
We have dropped all Taylor expansion terms that are of degree two or higher
in du, dv. The reason is that, in defining the integral, we take the limit
du, dv → 0. Because of that limit, all of the dropped terms contribute exactly
0 to the integral. We shall not prove this. But we shall show, in the optional
§3.3.5, why this is the case.
The small piece of our surface with corners P0 , P1 , P2 , P3 is approximately
a parallelogram with sides
P2 P3
−−−→ −−−→ ∂r
P0 P1 ≈ P2 P3 ≈
∂u
(u0 , v0 ) du ÝÝÝÑ
P0 P2
−−−→ −−−→ ∂r P1
P 0 P2 ≈ P1 P3 ≈
∂v
(u0 , v0 ) dv P0 θÝÝÝÑ
P0 P1
−−−→ −−−→
Denote by θ the angle between the vectors P0 P1 and P0 P2 . The base of the
−−−→ −−−→
parallelogram, P0 P1 , has length P0 P1 , and the height of the parallelogram is
−−−→
P0 P2 sin θ. So the area of the parallelogram is1
∂r ∂r
n̂ dS = ± (u , v) × (u , v) dudv
∂u ∂v
∂r ∂r
dS = (u , v) ×
(u , v) dudv
∂u ∂v
The ± sign in 3.3.1 is there because there are two unit normal vectors
at each point of a surface, one on each side of the surface. Typically, the
application itself tells you which of the two normal vectors should be used. We
shall see many examples shortly.
3.3.2 Graphs
The surface which is the graph z = f (x, y) can be parametrized by
As
∂r ∂f ∂r ∂f
= ı̂ı + k̂ and = ̂ + k̂
∂x ∂x ∂y ∂y
we have
ı̂ı ̂ k̂
∂r ∂r ∂f
× = det 1 0 = −fx (x, y)ı̂ı − fy (x, y) ̂ + k̂
∂x
∂x ∂y ∂f
0 1 ∂y
z
n̂ k̂
θ
dx
x dy
p
represents a little piece of our surface. It has area dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy.
The blue parallelogram in the same sketch represents the projection of the red
parallelogram onto the xy-plane. It has area dxdy. The vector n̂ in the sketch
is a unit normal for the red parallelogram. We have seen that it is parallel to
∂r ∂r
× = −fx (x, y)ı̂ı − fy (x, y) ̂ + k̂
∂x ∂y
understand why it is true, if one thinks in terms of the Taylor expansion of G about the point.
For simplicity, let’s suppose that the point is (0, 0, 0) and G happens to be exactly equal to
its first order Taylor expansion about (0, 0, 0). That is, G(x, y, z) = A + Bx + Cy + Dz, for
some constants A, B, C, D. Since (0, 0, 0) is on the surface, A = K. As ∂G ∂z
= D 6= 0 we can
1
easily solve G(x, y, z) = K for z as a function of x and y. Namely z = D (−Bx − Cy). The
general proof is based on the fact that, under reasonable hypotheses, the first order Taylor
expansion is a good approximation to G near (0, 0, 0).
CHAPTER 3. SURFACE INTEGRALS 147
function of x and y. That is, the surface also obeys z = f (x, y) for a function
f (x, y) that satisfies
G x, y, f (x, y) = K
near the point. Differentiating this with respect to x and y gives, by the chain
rule,
∂ h i
0= G x, y, f (x, y) = Gx x, y, f (x, y) + Gz x, y, f (x, y) fx (x, y)
∂x
∂ h i
0= G x, y, f (x, y) = Gy x, y, f (x, y) + Gz x, y, f (x, y) fy (x, y)
∂y
which implies
Gx x, y, f (x, y) Gy x, y, f (x, y)
fx (x, y) = − fy (x, y) = −
Gz x, y, f (x, y) Gz x, y, f (x, y)
and
Gx x, y, f (x, y) Gy x, y, f (x, y)
−fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ = ı̂ı + ̂ + k̂
Gz x, y, f (x, y) Gz x, y, f (x, y)
∇ G x, y, f (x, y)
=
Gz x, y, f (x, y)
So, by (3.3.2),
Equation 3.3.3 For the surface G(x, y, z) = K, when Gz (x, y, z) 6= 0,
∇ G x, y, z
n̂ dS = ± dxdy
∇ G x, y, z · k̂
∇ G x, y, z
dS =
dxdy
∇ G x, y, z · k̂
3.3.4 Examples of
RR
S ρ dS
We’ll start by computing, in several different ways, the surface area of the
hemisphere
x2 + y 2 + z 2 = a2 z≥0
CHAPTER 3. SURFACE INTEGRALS 148
(with a > 0). You probably know, from high school, that the answer is 12 ×
4πa2 = 2πa2 . But you have probably not seen a derivation of this answer. Note
that, since x2 +y 2 = a2 −z 2 on the hemisphere, the
set of (x, y)’s for which there
is a z with (x, y, z) on the hemisphere is exactly (x, y) ∈ R2 x2 + y 2 ≤ a2 .
CHAPTER 3. SURFACE INTEGRALS 149
0 ≤ θ < 2π.
So the hemisphere can be parametrized by
p
x(r, θ) , y(r, θ) , z(r, θ) = r cos θ , r sin θ , a2 − r2
with 0 ≤ r ≤ a, 0 ≤ θ < 2π
√
Note that we selected the positive solution z = a2 − r2 of r2 + z 2 = a2 in
order to satisfy the condition that z ≥ 0. Since
∂x ∂y ∂z r
, , = cos θ , sin θ , − √
∂r ∂r ∂r a2 − r2
∂x ∂y ∂z
, , = (−r sin θ , r cos θ , 0)
∂θ ∂θ ∂θ
3.3.1 yields
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , drdθ
∂r ∂r ∂r ∂θ ∂θ ∂θ
ı̂ı ̂ k̂
= ± det cos θ sin θ − √a2r−r2 drdθ
−r sin θ r cos θ 0
r2 cos θ 2
r sin θ
=± √ , √ , r drdθ
2
a −r 2 a2 − r2
r r
r4 a2 r2 ar
dS = 2 2
+ r2 drdθ = drdθ = √ drdθ
a −r a − r2
2
a2 − r2
So the area of the hemisphere is
Z a Z 2π Z a
ar r
dr dθ √ = 2πa dr √
2
a −r 2 a − r2
2
0 0 0
Z 0
−du/2
= 2πa √
a 2 u
with u = a2 − r2 , du = −2r dr
h √ i0
= 2πa − u
a2
2
= 2πa
as it should be.
3 The
symbols r, θ, z are the standard mathematics symbols for the cylindrical coordinates.
Appendix A.7 gives another set of symbols that is commonly used in the physical sciences
and engineering.
CHAPTER 3. SURFACE INTEGRALS 150
3.3.3 yields
∇ G x, y, z
dS = dxdy
∇ G x, y, z · k̂
2x , 2y , 2z
= dxdy
2z
p
x2 + y 2 + z 2
= dxdy
|z|
a
=p dxdy on x2 + y 2 + z 2 = a2
a2 − x2 − y 2
We already showed, in Example 3.3.4, that the value of this integral is 2πa2 .
Example 3.3.6 Area of a hemisphere — using spherical coordinates.
Of course “integrating over a sphere” cries out for spherical coordinates. So
this time we parametrize the hemisphere x2 + y 2 + z 2 = a2 , z ≥ 0, using as
parameters the angular coordinates θ, ϕ of the spherical coordinates
z
p0, 0, zq ρ sin ϕ
px, y, zq
x = ρ sin ϕ cos θ ρ cos ϕ
y = ρ sin ϕ sin θ ϕ ρ
z = ρ cos ϕ
y
px, 0, 0q θ
px, y, 0q
ρ sin ϕ sin θ
x
and then apply 3.3.1. In spherical coordinates the equation x2 +y 2 +z 2 = a2
becomes just ρ2 = a2 , and the condition z ≥ 0 is 0 ≤ ϕ ≤ π2 , 0 ≤ θ < 2π. So
the hemisphere can be parametrized4 by
x(θ, ϕ) , y(θ, ϕ) , z(θ, ϕ) = a sin ϕ cos θ , a sin ϕ sin θ , a cos ϕ
π
0 ≤ ϕ ≤ , 0 ≤ θ < 2π
2
CHAPTER 3. SURFACE INTEGRALS 151
Since
∂x ∂y ∂z
, , = − a sin ϕ sin θ , a sin ϕ cos θ , 0
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (a cos ϕ cos θ , a cos ϕ sin θ , −a sin ϕ)
∂ϕ ∂ϕ ∂ϕ
3.3.1 yields
∂x∂y ∂z ∂x ∂y ∂z
n̂ dS = ± ,, × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ
= ± − a sin ϕ sin θ, a sin ϕ cos θ, 0 ×(a cos ϕ cos θ, a cos ϕ sin θ, −a sin ϕ) dθdϕ
= ± − a2 sin2 ϕ cos θ , −a2 sin2 ϕ sin θ , −a2 sin ϕ cos ϕ dθdϕ
= ∓a2 sin ϕ sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
and
q
dS = a2 sin ϕ sin2 ϕ cos2 θ + sin2 ϕ sin2 θ + cos2 ϕ dθdϕ
= a2 sin ϕ dθdϕ
= 2πa2
There is an easier way to do this, using a little geometry.
Example 3.3.7 Area of a hemisphere — using spherical coordinates
again. We are now going to again compute the surface area of the hemisphere
using spherical coordinates. But this time instead of determining dS using the
canned formula 3.3.1, we are going to read it off of a sketch.
Sketch the part of the hemisphere that is in the first octant, x ≥ 0, y ≥ 0,
z ≥ 0. Slice it up into small pieces by drawing in curves of constant θ (the blue
lines in the figure below) and curves of constant ϕ (the red lines in the figure
below).
4 As we have noted before, the spherical coordinate system really breaks down at ϕ = 0,
because ρ = 1, ϕ = 0 gives the same point, namely the north pole (0, 0, 1), for all values
of θ. We should really treat our integral like an improper integral, first integrating over
ε < ϕ ≤ π2 and then taking the limit ε → 0+ . However the breakdown of the spherical
coordinate system at ϕ = 0, just like the breakdown of polar coordinates at r = 0, rarely
causes problem and it is routine to skip the “improper integral” step.
CHAPTER 3. SURFACE INTEGRALS 152
z
a sin ϕ dθ
a dϕ
x
Each piece is approximately a little rectangle. Concentrate on one of them,
like the piece with the thick sides in the figure above. The area, dS, of that
piece is (essentially) the product of its height and its width. Each of the two
sides of the piece is
• a segment of a circle of radius a (a fat blue line in both the figure above
and in the figure on the left below)
• that subtends an angle dϕ
dϕ
• and hence is the fraction 2π of a full circle of radius a and hence is of
length dϕ
2π 2πa = adϕ.
dθ
a sin ϕ
a a dϕ
ϕ a
ϕ dϕ y
y
x
x
So the area of our piece is
dS = adϕ a sin ϕdθ = a2 sin ϕ dθdϕ
This is exactly the same formula that we found for dS in Example 3.3.6 so
that we will, yet again, get that the area of a hemisphere of radius a is 2πa2 .
But wait! We can do it again, by yet another method!
(Phew!)
CHAPTER 3. SURFACE INTEGRALS 153
So 3.3.2 yields
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy
s
−x 2 −y 2
= 1+ p + p dxdy
a2 − x2 − y 2 a 2 − x2 − y 2
s
x2 + y 2
= 1+ 2 dxdy
a − x2 − y 2
s
a2
= dxdy
a2 − x2 − y 2
√ a
RR
So the area is x2 +y 2 ≤a2
dxdy. We already found, in Example
a2 −x2 −y 2
3.3.5, that the value of this integral in 2πa2 .
Let’s do some more substantial examples, where the integrand is not 1.
RR 2 2 2
Example 3.3.9 Evaluate S
x y z dS where S is the part of the cone
x2 + y 2 = z 2 with 0 ≤ z ≤ 1.
Solution 1. We can express S as
D2 n̂ “ k̂
V
p
z = f (x, y) = x2 + y 2 x2 +y 2 ≤ 1 S n̂
D1
n̂ “ ´k̂
Now since
x y
fx (x, y) = p fy (x, y) = p
x + y2
2 x2 + y2
3.3.2 gives6
h x2 y 2 i1/2 √
dS = 1 + + dxdy = 2 dxdy
x2 + y 2 x2 + y 2
Our integral is then
ZZ
2 2 2
√ ZZ
x y z dS = 2 x2 y 2 (x2 + y 2 ) dxdy
S x2 +y 2 ≤1
Since we are integrating over a circular domain, let’s convert to polar coordi-
nates.
ZZ
2 2 2
√ Z 2π Z 1
x y z dS = 2 dθ dr r(r cos θ)2 (r sin θ)2 r2
S 0 0
5 We promise!
CHAPTER 3. SURFACE INTEGRALS 154
√ 2π 1
Z Z
2 2 7
= 2 dθ cos θ sin θ dr r
0 0
√ Z 2π √ Z 2π
2 2
= dθ cos2 θ sin2 θ = dθ sin2 (2θ)
8 0 32 0
√ Z 2π
2
= dθ 1 − cos(4θ)
64 0
Then because
∂r ∂r
= cos θ ı̂ı + sin θ ̂ + k̂ and = −z sin θ ı̂ı + z cos θ ̂
∂z ∂θ
3.3.1 yields9
ı̂ı ̂ k̂
n̂ dS = ± det cos θ sin θ 1 dzdθ
−z sin θ z cos θ 0
= ± − z cos θ ı̂ı − z sin θ ̂ + z k̂ dzdθ
√
dS = 2z dzdθ
Let’s do something more celestial.
Example 3.3.10 Consider a spherical shell of radius a with mass density µ
per unit area. Think of it as a hollow planet10 . We are going to determine
the gravitational force that it exerts on a particle of mass m a distance b away
from its centre. This particle can be either outside the shell (b > a) or inside
6 This answer for dS is a very clean. Think about why. Hint: review the discussion
following 3.3.2.
7 If you have forgotten why, sketch the graph.
8 We did so previously, with different variable names, in Example 3.2.2.
9 Again the formula for dS is very neat. Think about why.
CHAPTER 3. SURFACE INTEGRALS 155
the shell (b < a). We can choose the coordinate system so that the centre
of the shell is at the origin and the particle is at (0, 0, b). By Newton’s law
of gravitation, the force exerted on the particle by a tiny piece of the shell of
surface area dS located at r is
G (µdS) m
(r − (0, 0, b))
|r − (0, 0, b)|3
Here G is the gravitational constant, µdS is the mass of the tiny piece of shell,
m is the mass of the particle
p0, 0, bq
r ´ p0, 0, bq r ´ p0, 0, bq dS
dS p0, 0, bq r
r
and r − (0, 0, b) is the vector from the particle to the piece of shell. If we
work in spherical coordinates, as we did in Example 3.3.6,
dS = a2 sin ϕ dϕdθ
and
Note for future reference that the square root in [a2 + b2 − 2ab cos ϕ]3/2 is the
positive square root because [b2 + a2 − 2ab cos ϕ]1/2 is the length of r − (0, 0, b),
which is positive.
This integral is a little different than other integrals that we have encoun-
tered so far in that the integrand is a vector. By definition11 ,
ZZ ZZ ZZ ZZ
G1 ı̂ı + G2 ̂ + G3 k̂ dS = ı̂ı G1 dS + ̂ G2 dS + k̂ G3 dS
S S S S
a2 + b2 − u
u = a2 + b2 − 2ab cos ϕ du = 2ab sin ϕ dϕ cos ϕ =
2ab
When ϕ = 0, u = (a − b)2 and when ϕ = π, u = (a + b)2 , so
(a+b)2 a2 +b2 −u
−b
Z
πGµma 2b
F= k̂ du
b (a−b)2 u3/2
Z (a+b)2 a2 −b2 −u
πGµma 2b
= k̂ du
b (a−b)2 u3/2
πGµma h a2 − b2 u−1/2 1 u1/2 i(a+b)
2
= k̂ −
b 2b −1/2 2b 1/2 (a−b)2
πGµma h b2 − a2 1 a + b b2 − a2 1 |a − b| i
F= k̂ − − +
b b a+b b b |a − b| b
If b > a, so that |a − b| = b − a
πGµma h b − a a + b a + b a − b i
F= k̂ − + + =0
b b b b b
The moral13 is
• if the particle is inside the shell, it feels no gravitational force at all, and
• if the particle is outside the shell, it feels the same gravitational force as
it would if the entire mass of the shell (4πa2 µ) were concentrated at the
centre of the shell.
14
Example 3.3.11 Optional — Gravity Train. The “Gravity Train”
refers to the following curious, though admittedly not very practical, thought
experiment.
10 A favourite of science fiction and fantasy writers. Plug “subterranean fiction” into your
favourite search engine. While you’re at it, also try “gravity train”. We’ll look at it in the
optional Example 3.3.11. RR RR RR
11 Under this definition we still have (A + B) dS = A dS + B dS.
12 Think about why the ı̂ı and ̂ components should both be zero. Think symmetry.
13 These two results appeared in Isaac Newton’s Principia Mathematica (1687). They are
• Pretend that the Earth is a perfect sphere of radius R and that it has a
constant mass density ρ.
• Pick any two distinct points on the surface of the Earth. Call them V
and M .
• Bore a tunnel straight through the Earth from V to M .
• Place a train in the tunnel at V . Assume that the only forces acting on
the train are gravity, G, and a normal force, N, that the tunnel imposes
on the train to keep it in the tunnel. In particular, there are no frictional
forces, like air resistance, and the train does not have an engine. Release
the train and assume that it does not melt as it passes through the centre
of the Earth.
N
M V
G
O
What happens?
We’ll simplify our analysis of the motion of the train by picking a convenient
coordinate system.
• First translate our coordinate system so that the centre of the Earth, call
it O, is at the origin, (0, 0, 0).
• Then rotate our coordinate system about the origin so that the origin, V
and M all lie in the xz-plane.
• Then rotate our coordinate system about the y-axis so that V and M
have√the same z-coordinate
Z ≥ 0. So the coordinates
√ of V and M are
± R2 − Z 2 , 0 , Z . Let’s suppose that V is at R2 − Z 2 , 0 , Z and
√
M is at − R2 − Z 2 , 0 , Z . It really doesn’t matter which is which, but
√
we can always arrange that it is V at + R2 − Z 2 , 0 , Z by rotating
around the z-axis by 180◦ if necessary.
z
? N ?
p´ R2 ´ Z 2 , 0 , Zq p R2 ´ Z 2 , 0 , Zq
G
x
p0, 0, 0q
CHAPTER 3. SURFACE INTEGRALS 158
The y- and z-coordinates of the train are always fixed at 0 and Z, respec-
tively. So let’s call the x-coordinate at time t x(t), and look at the x-component
of Newton’s law of motion.
ma = G + N
It is
mx00 (t) = G · ı̂ı
because the normal force N has no ı̂ı component. Recall that Newton’s law of
gravity says that
GM m
G=− r
|r|3
where G is the gravitational constant, r is the vector from O to the train, and
m is the mass of the train. In this case, because of our computation in Example
3.3.10, the train only feels gravity from shells of the Earth that are inside the
train, so that M is the mass of the
z
|r| G
x
part of the Earth whose distance to the centre of the Earth is no more than
|r|. So
4
M = π|r|3 ρ
3
and
Gm 4
mx00 (t) = − 3 π|r|3 ρ r · ı̂ı
|r| 3
so that
4πGρ
x00 (t) + x(t) = 0
3
This is exactly the differential equation of simple harmonic motion. We have
seen it before in Example 2.2.7. Except for the constant 4πGρ
3 , it is identical to
the equation solved in Example A.9.4 of the Appendix A.9, entitled “Review
of Linear Ordinary Differential Equations”. The general solution is
r ! r !
4πGρ 4πGρ
x(t) = C1 cos t + C2 sin t
3 3
√
q
4πGρ
The train reaches M when x(t) = − R2 − Z 2 . That is, when cos 3 t =
−1. So the transit time, T , from V to M obeys
r r r
4πGρ 3 3π
T = π =⇒ T = π =
3 4πGρ 4Gρ
Notice that this transit time depends only on the gravitational constant G and
the density of the Earth ρ. In particular it is completely independent of
• where V and M are and, in particular,
• how close together V and M are, and also of
P2 P3 u varying
v“v0 `dv
P1 u varying
P0 v“v0
assistant to Christopher Wren) Robert Hooke (1635--1703) wrote about the gravity train
idea in a letter to Isaac Newton. A gravity train was used in the 2012 movie Total Recall.
15 Remember the error in the Taylor polynomial approximations.
16 See the optional §1.1.6 of the CLP-2 text for an analogous argument concerning Riemann
sums.
CHAPTER 3. SURFACE INTEGRALS 160
Suppose that our domain of integration consists of all (u, v)’s in a rectangle
of width A and height B, as in the figure below.
v
du
dv
B
A
Subdivide the rectangle into a grid of n × n small subrectangles by drawing
lines of constant v (the red lines in the figure) and lines of constant v (the blue
lines in the figure). Each subrectangle has width du = A n and height dv = n .
B
Now suppose that in setting up the integral we make, for each subrectangle,
an error that is bounded by some constant times
A 2 B A B 2 AB(A + B)
du2 dv + du dv 2 = + =
n n n n n3
Because there are a total of n2 subrectangles, the total error that we have
introduced, for all of these subrectangles, is no larger than a constant times
AB(A + B) AB(A + B)
n2 × =
n3 n
When we define our integral by taking the limit n → 0 of the Riemann sums,
this error converges to exactly 0.
3.3.6 Exercises
Exercises — Stage 1
1. Let 0 < θ < π2 , and a, b > 0. Denote by S the part of the surface
z = y tan θ with 0 ≤ x ≤ a, 0 ≤ y ≤ b.
a Find the surface area of S without using any calculus.
b Find the surface area of S by using (3.3.2).
2. Let a, b, c > 0. Denote by S the triangle with vertices (a, 0, 0), (0, b, 0)
and (0, 0, c).
a Find the surface area of S in three different ways, each using
(3.3.2).
b Denote by Txy the projection of S onto the xy-plane. (It is the
triangle with vertices (0, 0, 0) (a, 0, 0) and (0, b, 0).) Similarly
use Txz to denote the projection of S onto the xz-plane and Tyz
to denote the projection of S onto the yz-plane. Show that
q
Area(S) = Area(Txy )2 + Area(Txz )2 + Area(Tyz )2
3. Let a, h > 0. Denote by S the part of the cylinder x2 + z 2 = a2 with
x ≥ 0, 0 ≤ y ≤ h and z ≥ 0.
CHAPTER 3. SURFACE INTEGRALS 161
z
S
y
(a, h, 0)
x
a Find the surface area of S without using any calculus.
b Parametrize S by
π
r(θ, y) = a cos θ ı̂ı + y ̂ + a sin θ k̂ 0≤θ≤ , 0≤y≤h
2
Find the surface area of S by using (3.3.1).
Exercises — Stage 2
4. Let S be the part of the surface z = xy lying inside the cylinder
x2 + y 2 = 3. Find the moment of inertia of S about the z-axis, that
is, ZZ
I= (x2 + y 2 ) dS
S
5. ∗. Find the surface area of the part of the paraboloid z = a2 − x2 − y 2
which lies above the xy--plane.
6. ∗. Find the area of the portion of the cone z 2 = x2 + y 2 lying between
the planes z = 2 and z = 3.
2
7. ∗. Determine the surface area of the surface given by z = 3 x3/2 +
y 3/2 , over the square 0 ≤ x ≤ 1, 0 ≤ y ≤ 1.
8. ∗.
RR of the surface z = f (x, y) above the
a To find the surface area
region D, we integrate D F (x, y) dA. What is F (x, y)?
b Consider a “Death Star”, a ball of radius 2 centred
√ at the origin
with another ball of radius 2 centred at (0, 0, 2 3) cut out of it.
The diagram below shows the slice where y = 0.
z
?
2 3
?
p1, 0, 3q
π
6
2
x
(i) The Rebels want to paint part of the surface of Death Star
hot pink; specifically, the concave part (indicated with a
thick line in the diagram). To help them determine how
CHAPTER 3. SURFACE INTEGRALS 162
Z Z
surface area = dr dθ
where
p S is the part of the sphere x2 + y 2 + z 2 = 2 for which x ≥
y + z2.
2
CHAPTER 3. SURFACE INTEGRALS 164
x2 + z 2 = sin2 y
x+y+z =2
that lies in the first octant oriented so that n̂ has a positive k̂ com-
ponent. Let
F = xı̂ı + y ̂ + z k̂
Evaluate the flux integral
ZZ
F · n̂ dS
S
RR
29. ∗. Find the net flux S F · n̂ dS of the vector field F(x, y, z) =
(x, y, z) upwards (with respect to the
z-axis) through the surface S
parametrized by r = uv 2 , u2 v , uv for 0 ≤ u ≤ 1, 0 ≤ v ≤ 3.
30. ∗. Let S be the surface obtained by revolving the curve z = ey ,
0 ≤ y ≤ 1, around the y-axis, with the orientation of S having n̂
pointing toward the y-axis.
a Draw a picture of S and find a parameterization of S.
b Compute the integral S ey dS.
RR
RR
c Compute the flux integral S F · n̂ dS where F = (x, 0, z).
31. ∗. Compute the net outward flux of the vector field
r xı̂ı + y ̂ + z k̂
F= =p
|r| x2 + y 2 + z 2
across the boundary of the region between the spheres of radius 1 and
radius 2 centred at the origin.
32. ∗. Evaluate the surface integral S z 2 dS where S is the part of the
RR
34. ∗. Let the thin shell S consist of the part of the surface z 2 = 2xy with
x ≥ 1, y ≥ 1 and z ≤ 2. Find the mass of S if it has surface density
given by ρ(x, y, z) = 3z kg per unit area.
35. ∗. Let S be the portion of the paraboloid x = y 2 + z 2 that satisfies
x ≤ 2y. Its unit normal vector n̂ is so chosen that n̂ · ı̂ı > 0. Find the
flux of F = 2ı̂ı + z ̂ + y k̂ out of S.
36. ∗. Let S denote the portion of the paraboloid z = 1 − 14 x2 − y 2
for which z ≥ 0. Orient S so that its unit normal has a positive k̂
component. Let
x2 + z 2 = cos2 y
1 π 1
Find an equation for the tangent plane to S at the point 2, 4 , 2 .
b Compute the integral
ZZ
sin y dS
S
where S is the part of the surface from (a) lying between the
planes y = 0 and y = 12 π.
39. ∗. Let f be a function on R3 such that all its
first order
partial deriva-
tives are continuous. Let S be the surface (x, y, z) f (x, y, z) = c
for some c ∈ R. Assume that ∇ f 6= 0 on S. Let F be the gradient
field F = ∇ f .
a Let C be a piecewise smooth curve contained
R in S (not neces-
sarily closed). Must it be true that C F · dr = 0? Explain
why.
b Prove that for any vector field G,
ZZ
(F × G) · n̂ dS = 0.
S
40. ∗.
a Give parametric descriptions of the form r(u, v) = x(u, v) , y(u, v) , z(u, v)
for the following surfaces. Be sure to state the domains of your
parametrizations.
(i) The part of the plane 2x + 4y + 3z = 16 in the first octant
(x, y, z) x ≥ 0, y ≥ 0, z ≥ 0
√
(ii) The cap of the sphere x2 + y 2 + z 2 = 16 for 4/ 2 ≤ z ≤ 4.
CHAPTER 3. SURFACE INTEGRALS 166
b Compute ZZ
xy ı̂ı + xz ̂ + zy k̂ · n̂ dS
S
42. ∗. Let S be the part of the surface (x + y + 1)2 + z 2 = 4 which lies in
the first octant. Find the flux of F downwards through S where
n̂ vdt
vdt
θ
vdt vdt
Here ρ is the density of the fluid, v is the velocity field of the fluid, and n̂(x, y, z)
is a unit normal to S at (x, y, z). If the flux integral is positive the fluid is
crossing in the direction n̂. If it is negative the fluid is crossing opposite to the
direction of n̂. The rate at which volume of fluid is crossing through a surface
S is the flux integral
ZZ
v(x, y, z, t) · n̂(x, y, z) dS
S
This is the rate at which volume of fluid is exiting the sphere. In our derivation
of the vector field we assumed that the fluid is incompressible, so it is also the
rate at which the point source is creating fluid.
Example 3.4.3 Vortex. In Figure 2.1.6, we sketched the vector field (in two
dimensions)
v(x, y) = Ω − yı̂ı + x̂
We’ll now compute the flux of this vector field across a circle C centred on the
origin. Suppose that the circle has radius R.
n̂
RBy definition, in two dimensions, the flux of a vector field across a curve C
is C
v · n̂ ds.
This is the natural analog of the flux in three dimensions — the surface S
has been replaced by the curve C, and the surface area dS of a tiny piece of S
has been replaced by the arc length ds of a tiny piece of C.
The outward pointing unit normal at a point (x, y) on our circle C is
xı̂ı + ŷ xı̂ı + ŷ
n̂(x, y) = p =
2
x +y 2 R
So
Ω
v(x, y) · n̂(x, y) = − yı̂ı + x̂ · xı̂ı + ŷ = 0
R
and the flux across C is Z
v · n̂ ds = 0
C
This should not be a surprise — no fluid is crossing C at all. This is exactly
what we would expect from looking at the arrows in Figure 2.1.6 or at the
stream lines in Example 2.2.6.
1 You
can imagine that a very small pipe pumps water to the origin.
2 It
doesn’t really matter which unit normal we pick here. We just have to be clear which
one we’re using. With the outward normal, the flux gives the rate at which fluid crosses
the sphere in the outward direction. If we were to use the inward pointing normal, the flux
would give the rate at which fluid crosses the sphere in the inward direction.
CHAPTER 3. SURFACE INTEGRALS 169
RR
Example 3.4.4 Evaluate S
F · n̂ dS where
n̂ “ k̂
St
n̂
Ss
Sb
n̂ “ ´k̂
It is natural to decompose its surface S into three parts
We’ll compute the flux through each of the three parts separately and then
add them together.
The Top: On the top, the outward pointing normal to S is n̂ = k̂ and
dS = dxdy. This is probably intuitively obvious. But if it isn’t, you can
always derive it by parametrizing the top by r(x, y) = xı̂ı + y ̂ + 5 k̂ with
x2 + y 2 ≤ 9. So the flux through the top is
ZZ ZZ ZZ
F · n̂ dS = 2 2
(x + z) dxdy = (x + 5) dxdy
x +y ≤9
St z=5 x2 +y 2 ≤9
RR
The integral x2 +y2 ≤9 x dxdy = 0 since x is odd and the domain of integration
is symmetric about x = 0. So
ZZ ZZ
F · n̂ dS = 5 dxdy = 5π(3)2 = 45π
St x2 +y 2 ≤9
n̂
∂r
= (−3 sin θ , 3 cos θ , 0)
∂θ
∂r r
= (0 , 0 , 1) θ
∂z
∂r ∂r
n̂ dS = × dθ dz
∂θ ∂z
= (3 cos θ , 3 sin θ , 0) dθ dz top view
4
Note that n̂ = (cos θ , sin θ , 0) is outward pointing , as desired. Continuing,
F x(θ, z), y(θ, z), z(θ, z) = 3(cos θ+sin θ)ı̂ı + (3 sin θ+z) ̂ + (3 cos θ+z) k̂
F · n̂ dS = 9 cos2 θ+3 sin θ cos θ+9 sin2 θ+3z sin θ dθ dz
3
= 9 + sin(2θ) + 3z sin θ dθ dz
2
So the flux through the side is
ZZ Z 2π Z 5
3
F · n̂ dS = dθ dz 9 + sin(2θ) + 3z sin θ
Ss 0 0 2
Z 2π Z 5 Z 2π Z 2π
=9 dθ dz since sin θ dθ = sin(2θ) dθ = 0
0 0 0 0
= 9 × 2π × 5 = 90π
RR 4ı 2
Example 3.4.5 Evaluate S
F · n̂ dS where F(x, y, z) = x ı̂ı + 2y ̂ + z k̂, S
is the half of the surface 41 x2 + 19 y 2 + z 2 = 1 with z ≥ 0, and n̂ is the upward
pointing unit normal.
“ignore the z-coordinate”. The top view of the can is a circle of radius 3. Then, at a generic
point, r = (cos θ, sin θ), on the can, draw the unit normal n̂ = (cos θ , sin θ) with its tail at
r. It is pointing away from the origin, just like r is. That is, n̂ is pointing outward.
CHAPTER 3. SURFACE INTEGRALS 171
implies
x2 + y 2 = cos2 θ sin2 ϕ + sin2 θ sin2 ϕ = sin2 ϕ
and then
x2 + y 2 + z 2 = sin2 ϕ + cos2 ϕ = 1
The extra minus sign in n̂ dS was put there to make the z component of n̂
positive. (The problem specified that n̂ is to be upward unit normal.) As
F x(θ, ϕ) , y(θ, ϕ) , z(θ, ϕ)
= 24 cos4 θ sin4 ϕ ı̂ı + 2 × 32 sin2 θ sin2 ϕ ̂ + cos ϕ k̂
we have
h i
F · n̂ dS = 3 × 24 cos5 θ sin6 ϕ+2 × 2 × 32 sin3 θ sin4 ϕ+6 sin ϕ cos2 ϕ dθdϕ
x(r, θ) = 2r cos θ
y(r, θ) = 3r sin θ
p
z(r, θ) = 1 − r2
= 4π
3/2
The integral was evaluated by guessing (and checking) that − 31 (1 − r2 ) is
√
an antiderivative of r 1 − r2 . It can also be done by substituting u = 1 − r2 ,
du = −2r dr.
CHAPTER 3. SURFACE INTEGRALS 173
∇G x
ı̂ı + 2y
9 ̂ + 2z k̂
n̂dS = dx dy = 2 dx dy
∇ G · k̂ 2z
x y
= ı̂ı + ̂ + k̂ dx dy
4z 9z
x5 2y 3
=⇒ F · n̂ dS = + + z dx dy
4z 9z
It is true that n̂dS, and consequently F · n̂ dS become infinite6
as z → 0. So we should really treat the integral as an improper integral,
first integrating over z ≥ ε and then taking the limit ε → 0+ . But, as we shall
see, the singularity
q is harmless. So it is standard to gloss over this point. On
2 y2 x2 y2
S, z = z(x, y) = 1 − x4 − 9 and 4 + 9 ≤ 1, so
x5 2y 3
Z ZZ
F · n̂ dS = + + z(x, y) dx dy
S
2 x
4
2
+ y9 ≤1 4z(x, y) 9z(x, y)
5 3
x 2y
Both 4z(x,y) and 9z(x,y) are odd under x → −x, y → −y and the domain of
integration is even under x → −x, y → −y, so their integrals are zero and
Z ZZ
F · n̂ dS = 2 2
z(x, y) dx dy
S x
4 + y9 ≤1
r
x2 y2
ZZ
= 1− − dx dy
x2
4
2
+ y9 ≤1 4 9
5 3
2y
x
r
+ x 2 y 2
=⇒ F · n̂ dS = q 4 9
+ 1− − dx dy
1− x2
− y2 4 9
4 9
We can view n̂+ as being associated to (or attached to) the outside of the
sphere and n̂− as being associated to (or attached to) the inside of the sphere.
Note that, as we move over the sphere, both n̂+ and n̂− change continuously.
Definition 3.5.1 An oriented surface is a surface together with a continuous
function
N̂ : S → R3
such that, for each point p of S, N̂(p) is a unit normal to S at p. ♦
2
Example 3.5.2 Sphere. One orientation of the sphere S = (x, y, z) x +
2 2
y + z = 1 is
N̂(x, y, z) = (x, y, z)
It associates to each point p of S the outward pointing unit normal to S at p.
We can think of S with this orientation as being
the outer
side of S.
The other orientation of the sphere S = (x, y, z) x2 + y 2 + z 2 = 1 is
N̂(x, y, z) = −(x, y, z)
Lay it flat and then introduce a half twist so that the arrow on the right
hand end points upwards, rather than downwards. Then glue the two ends of
the strip together, with the two arrows coinciding. That’s the Möbius strip.
Let’s parametrize it. Think of the strip of paper that we used to construct
it as consisting of a backbone (the horizontal black line in the figure below)
with a bunch of ribs (like the thick blue line in the figure) emanating from it.
v
w
θ
ℓ
When we glue the two ends of the strip together, the black line forms a
circle. If the strip has length `, the circle will have circumference ` and hence
`
radius 2π . We’ll parametrize it as the circle
`
r̂(θ) where r̂(θ) = cos(θ)ı̂ı + sin(θ) ̂
2π
This circle is in the xy-plane. It is the black circle in the figure below. (The
figure only shows the part of the circle in the first octant, i.e. with x, y, z ≥ 0.)
k̂
y
θℓ
2π
r̂pθq
r̂pθq
x
Now we’ll add in the blue ribs. We’ll put the blue rib, that is attached to
`
the backbone at 2π r̂(θ), in the plane that contains the vectors r̂(θ) and k̂.
A side view of the plane that contains the vectors r̂(θ) and k̂ is sketched in
the figure below.
CHAPTER 3. SURFACE INTEGRALS 176
k̂ upv, θ, ϕq
ϕ
ℓ r̂pθq
2π
r̂pθq
To put the half twist into the strip of paper, we want the blue rib to rotate
about the backbone by 180◦ , i.e. π radians, as θ runs from 0 to 2π. That will
be the case if we pick the angle ϕ in the figure to be θ2 . The vector that is
running along the blue rib in the figure is
where the length, v, of the vector is a parameter. If the width of our original
strip of paper is w, then as the parameter v runs from − w2 to + w2 , the tip of
the vector u(v, θ, ϕ) runs over the entire blue rib. So, choosing ϕ = θ2 , our
parametrization of the Möbius strip is
` θ
r(θ, v) = r̂(θ) + u v, θ,
2π 2
` θ θ
= r̂(θ) + v cos r̂(θ) + v sin k̂
2π 2 2
w w
0 ≤ θ < 2π, − ≤ v ≤
2 2
where r̂(θ) = cos(θ)ı̂ı + sin(θ) ̂.
Now that we have parametrized the Möbius strip, let’s return to the ques-
tion of orientability. Recall, from Definition 3.5.1, that, if the Möbius strip
were orientable, there would exist a continuous function N̂ which assigns to
each point r of the strip a unit normal vector N̂(r) at r. First, we’ll find the
normal vectors to the surface using 3.3.1. The partial derivatives
∂r ` 0 θ 0 v θ v θ
(θ, v) = r̂ (θ) + v cos r̂ (θ) − sin r̂(θ) + cos k̂
∂θ 2π 2 2 2 2 2
∂r θ θ
(θ, v) = cos r̂(θ) + sin k̂
∂v 2 2
are relatively messy, so let’s just consider the case v = 0 (i.e. find the normal
vectors on the backbone). Then
∂r ` 0
(θ, 0) = r̂ (θ)
∂θ 2π
∂r θ θ
(θ, 0) = cos r̂(θ) + sin k̂
∂v 2 2
Since
r̂0 (θ) × r̂(θ) = − sin(θ)ı̂ı + cos(θ) ̂ × cos(θ)ı̂ı + sin(θ) ̂ = −k̂
we have
∂r ∂r ` h θ θ i
(θ, 0) × (θ, 0) = − cos k̂ − sin r̂(θ)
∂θ ∂v 2π 2 2
CHAPTER 3. SURFACE INTEGRALS 177
As k̂ and r̂(θ) are mutually perpendicular unit vectors, cos θ2 k̂ − sin θ2 r̂(θ)
has length one, and the two unit normal vectors to the Möbius strip at r(θ, 0)
are
h θ θ i
± cos k̂ − sin r̂(θ)
2 2
So, for each θ, N̂ r(θ, 0) must be either
θ θ h θ θ i
cos k̂ − sin r̂(θ) or − cos k̂ − sin r̂(θ)
2 2 2 2
Imagine walking along the Möbius strip. The normal vector N̂ r(θ, v) is
our body when we are at r(θ, v) — our feet are at the tail of the vector
N̂ r(θ, v) and our head is at the arrow of N̂ r(θ, v) . We start walking at
` `
r(0, 0) = 2π ı̂ı. Our body, N̂ 2π ı̂ı = N̂ r(0, 0) has to be one of ± cos(0) k̂ −
sin(0) r̂(0) = ±k̂. Let’s suppose that N̂ r(0, 0) = +k̂. (We start up-
right.) Now we start walking along the backbone of the Möbius strip, in-
creasing θ. Because N̂ r(θ, 0) has to be continuous, N̂ r(θ, 0) has to be
+ cos θ2 k̂ − sin θ2 r̂(θ) . We keep increasing θ. By continuity, N̂ r(θ, 0)
So we have arrived back upside down. That’s a problem — N̂ r(2π, 0) =
` `
N̂ 2π ı̂ı and we have already defined N̂ 2π ı̂ı = +k̂, not −k̂. So the Möbius
strip is not orientable. The interested reader should look up M. C. Escher’s
Möbius Strip II (Red Ants).
Integral Theorems
∂f ∂f ∂f
grad f = ∇ f = ı̂ı + ̂ + k̂
∂x ∂y ∂z
Note that the input, f , for the gradient is a scalar-valued function, while
∇f , is a vector-valued function.
the output,∇
b The divergence of a vector field F(x, y, z) is the scalar-valued function
178
CHAPTER 4. INTEGRAL THEOREMS 179
∇ · E = 4πρ
∇·B=0
1 ∂B
∇×E+ =0
c ∂t
1 ∂B 4π
∇×B− = J
c ∂t c
Here E is the electric field, B is the magnetic field, ρ is the charge density, J
is the current density and c is the speed of light.
is also the Laplace of Laplace’s equation, the Laplace transform, and the Laplace-Bayes
estimator. He was Napoleon’s examiner when Napoleon attended the Ecole Militaire in
Paris.
3 To be picky, these are Maxwell’s equations in the absence of a material medium and in
Gaussian units.
4 One important consequence of Maxwell’s equations is that electromagnetic radiation,
of prominent physicists, Maxwell was voted the third greatest physicist of all time. Only
Newton and Einstein beat him.
CHAPTER 4. INTEGRAL THEOREMS 180
(c) ∇ (f g) = (∇
∇f )g + f (∇
∇g)
(e) ∇(F · G) = F × (∇
∇ × G) − (∇
∇ × F) × G + (G · ∇)F + (F · ∇)G
Here6
∂F ∂F ∂F
(G · ∇ )F = G1 + G2 + G3
∂x ∂y ∂z
Theorem 4.1.4 Divergence Identities.
(a) ∇ · (F + G) = ∇ · F + ∇ · G
(b) ∇ · (cF) = c ∇ · F, for any constant c
(c) ∇ · (f F) = (∇
∇f ) · F + f ∇ · F
(d) ∇ · (F × G) = (∇
∇ × F) · G − F · (∇
∇ × G)
Theorem 4.1.5 Curl Identities.
(a) ∇ × (F + G) = ∇ × F + ∇ × G
(b) ∇ × (cF) = c ∇ × F, for any constant c
(c) ∇ × (f F) = (∇
∇f ) × F + f ∇ × F
(d) ∇ × (F × G) = F(∇
∇ · G) − (∇
∇ · F)G + (G · ∇ )F − (F · ∇ )G
Here
∂F ∂F ∂F
(G · ∇ )F = G1 + G2 + G3
∂x ∂y ∂z
Theorem 4.1.6 Laplacian Identities.
(a) ∇ 2 (f + g) = ∇ 2 f + ∇ 2 g
(b) ∇ 2 (cf ) = c ∇ 2 f , for any constant c
(c) ∇ 2 (f g) = f ∇ 2 g + 2∇
∇f · ∇ g + g ∇ 2 f
6 This is really the only definition that makes sense. For example G · (∇
∇F) does not make
sense because you can’t take the gradient of a vector-valued function.
CHAPTER 4. INTEGRAL THEOREMS 181
(d) ∇ · (f∇ ∇f ) = f ∇ 2 g − g ∇ 2 f
∇g − g∇
(e) ∇ × (∇
∇ × F) = ∇ (∇∇ · F) − ∇ 2 F (curl of curl)
Memory Aid. Most of the vector identities (in fact all of them except
Theorem 4.1.3.e, Theorem 4.1.5.d and Theorem 4.1.7) are really easy to guess.
Just combine the conventional linearity and product rules with the facts that
• if the left hand side is a vector (scalar), then the right hand side must
also be a vector (scalar) and
• the only valid products of two vectors are the dot and cross products and
• the product of a scalar with either a scalar or a vector cannot be either
a dot or cross product and
• A × B = −B × A. (The cross product is antisymmetric.)
∇f )·F+f ∇ ·F.
For example, consider Theorem 4.1.4.c, which says ∇ ·(f F) = (∇
• The left hand side, ∇ · (f F), is a scalar, so the right hand side must also
be a scalar.
• The left hand side, ∇ · (f F), is a derivative of the product of f and
F, so, mimicking the product rule, the right hand side will be a sum
of two terms, one with F multiplying a derivative of f , and one with f
multiplying a derivative of F.
• The derivative acting on f must be ∇ f , because ∇ · f and ∇ × f are not
well-defined. To end up with a scalar, rather than a vector, we must take
the dot product of ∇ f and F. So that term is (∇ ∇f ) · F.
• The derivative acting on F must be either ∇ · F or ∇ × F. We also need
to multiply by the scalar f and end up with a scalar. So the derivative
∇ · F}.
must be a scalar, i.e. ∇ · F and that term is f {∇
∇f ) · F + f ∇ · F, which, thankfully, is
• Our final guess is ∇ · (f F) = (∇
correct.
Proof of Theorems 4.1.3, 4.1.4, 4.1.5, 4.1.6 and 4.1.7. All of the proofs (except
for those of Theorem 4.1.7.c,d, which we will return to later) consist of
• writing out the definition of the left hand side and
• writing out the definition of the right hand side and
• observing (possibly after a little manipulation) that they are the same.
For Theorem 4.1.3.a,b, Theorem 4.1.4.a,b, Theorem 4.1.5.a,b and Theorem
4.1.6.a,b, the computation is trivial — one line per identity, if one uses some
efficient notation. Rename the coordinates x, y, zP to x1 , x2 , x3 and the standard
3 ∂
unit basis vectors ı̂ı, ̂, k̂ to ı̂ı1 , ı̂ı2 , ı̂ı3 . Then ∇ = n=1 ı̂ın ∂x n
and the proof of,
for example, Theorem 4.1.4.a is
3
X ∂
∇ · (F + G) = ı̂ın · (F + G)
n=1
∂xn
CHAPTER 4. INTEGRAL THEOREMS 182
3 3
X ∂ X ∂
= ı̂ın · F + ı̂ın · G = ∇ · F + ∇ · G
n=1
∂x n n=1
∂x n
Theorem 4.1.5.d: We use the same trick. Write out the left hand side as
3
X ∂
∇ × (F × G) = ı̂ın × (F × G)
n=1
∂xn
3 ∂F 3
X X ∂G
= ı̂ın × ×G + ı̂ın × F ×
n=1
∂xn n=1
∂xn
∇ · F)G + (∇
= (G · ∇ )F − (∇ ∇ · G)F − (F · ∇ )G
because the two red terms have cancelled, the two blue terms have cancelled
and the two black terms have cancelled.
Theorem 4.1.7.b: Substituting in
∂f ∂f ∂f
∇f = ı̂ı + ̂ + k̂
∂x ∂y ∂z
gives
∂ ∂f ∂ ∂f ∂ ∂f ∂ ∂f
∇ × ∇f = − ı̂ı − − ̂
∂y ∂z ∂z ∂y ∂x ∂z ∂z ∂x
∂ ∂f ∂ ∂f
+ − k̂
∂x ∂y ∂y ∂x
=0
By Theorem 4.1.7.b, ∇ × ∇ g = ∇ × ∇ h = 0, so
∇g × ∇ f ) = ∇ f · (∇
∇ · f (∇ ∇g × ∇ h)
∇g − g∇
∇ · (f∇ ∇f ) = (∇
∇f ) · (∇
∇g) + f ∇ · (∇
∇g) − (∇
∇g) · (∇
∇f ) + g ∇ · (∇
∇f )
= f ∇2g − g ∇2f
Theorem 4.1.7.e:
3 3
X 3
X ∂ ∂ X
∇ × F) =
∇ × (∇ ı̂ı` × ı̂ım × ı̂ın Fn
∂x` m=1
∂xm n=1
`=1
3
X ∂ 2 Fn
= ı̂ı` × ı̂ım × ı̂ın
∂x` ∂xm
`,m,n=1
wherea (
1 if m = n
δm,n =
0 if m 6= n
Hence
3 3
X ∂ 2 Fn X ∂ 2 Fn
∇ × F) =
∇ × (∇ δ`,nı̂ım − δ`,mı̂ın
∂x` ∂xm ∂x` ∂xm
`,m,n=1 `,m,n=1
3 3
X ∂ ∂Fn X ∂ 2 Fn
= ı̂ım − ı̂ın
m,n=1
∂xm ∂xn m,n=1 ∂x2m
∇ · F) − ∇ 2 F
= ∇ (∇
aδ
m,n is called the Kronecker delta function. It is named after the German number
theorist and logician Leopold Kronecker (1823–1891). He is reputed to have said “God made
the integers. All else is the work of man.”
Lemma 4.1.8
a a · (b × c) = (a × b) · c
b a × (b × c) = (c · a)b − (b · a)c
Proof. (a) Here are two proofs. For the first, just write out both sides
a · (b × c) = (a1 , a2 , a3 ) · (b2 c3 − b3 c2 , b3 c1 − b1 c3 , b1 c2 − b2 c1 )
= a1 b2 c3 − a1 b3 c2 + a2 b3 c1 − a2 b1 c3 + a3 b1 c2 − a3 b2 c1
(a × b) · c = (a2 b3 − a3 b2 , a3 b1 − a1 b3 , a1 b2 − a2 b1 ) · (c1 , c2 , c3 )
= a2 b3 c1 − a3 b2 c1 + a3 b1 c2 − a1 b3 c2 + a1 b2 c3 − a2 b1 c3
and observe that they are the same.
For the second proof, we again write out both sides, but this time we express
them in terms of determinants.
ı̂ı ̂ k̂
a · b × c = (a1 , a2 , a3 ) · det b1 b2 b3
c1 c2 c3
b2 b3 b1 b3 b1 b2
= a1 det − a2 det + a3 det
c2 c3 c1 c3 c1 c2
a1 a2 a3
= det b1 b2 b3
c1 c2 c3
ı̂ı ̂ k̂
a × b · c = det a1 a2 a3 · (c1 , c2 , c3 )
b1 b2 b3
a2 a3 a1 a3 a1 a2
= c1 det − c2 det + c3 det
b2 b3 b1 b3 b1 b2
c1 c2 c3
= det a1 a2 a3
b1 b2 b3
Exchanging two rows in a determinant changes the sign of the determinant.
Moving the top row of a 3 × 3 determinant to the bottom row requires two
exchanges of rows. So the two 3 × 3 determinants are equal.
CHAPTER 4. INTEGRAL THEOREMS 185
(b) The proof is not exceptionally difficult — just write out both sides and
grind. Substituting in
The last formula that we had for the left hand side is the same as the last
formula we had for the right hand side.
Example 4.1.9 Screening tests. We have seen the vector identity Theorem
4.1.7.b before. It says that if a vector field F is of the form F = ∇ ϕ for some
some function ϕ (that is, if F is conservative), then
∇ϕ) = 0
∇ × F = ∇ × (∇
Conversely, we have also seen, in Theorem 2.4.8, that, if F is defined and has
continuous first order partial derivatives on all of R3 , and if ∇ × F = 0, then F
is conservative. The vector identity Theorem 4.1.7.b is our screening test for
conservativeness.
Because its right hand side is zero, the vector identity Theorem 4.1.7.a is
suggestive. It says that if a vector field F is of the form F = ∇ × A for some
some vector field A, then
∇ × A) = 0
∇ · F = ∇ · (∇
∇·B=0
1 ∂B
∇×E+ =0
c ∂t
CHAPTER 4. INTEGRAL THEOREMS 186
that we saw in Example 4.1.2. The first equation implies that (assuming B
is sufficiently smooth) there is a vector field A, called the magnetic potential,
with B = ∇ × A. Substituting this into the second equation gives
1∂ 1 ∂A
0=∇×E+ ∇×A=∇× E+
c ∂t c ∂t
So E+ 1c ∂A
∂t passes the screening test of Theorem 4.1.7.b and there is a function
ϕ (called the electric potential) with
1 ∂A
E+ = −∇ϕ
c ∂t
We have put in the minus sign just to provide compatibility with the usual
physics terminology.
Example 4.1.10 Let r(x, y, z) = xı̂ı +y ̂ +z k̂ and let ψ(x, y, z) be an arbitrary
function. Verify that
∇ · r × ∇ψ = 0
By the vector identity Theorem 4.1.7.b, the second term is zero. Now since
∂z ∂y ∂z ∂x ∂y ∂x
∇×r= − ı̂ı − − ̂ + − k̂ = 0
∂y ∂z ∂x ∂z ∂x ∂y
the first term is also zero. Indeed ∇ · r × ∇ψ = 0 holds for any curl free
r(x, y, z).
∇·B=0
Of course, we’ll consider the converse soon. Also note that the vector
potential, when it exists, is far from unique. Two vector fields A and à are
both vector potentials for the same vector field if and only if
∇ × A = ∇ × Ã ⇐⇒ ∇ × (A − Ã) = 0
That is, if and only if the difference A−Ã passes the conservative field screening
test of Theorems 2.3.9 and 2.4.8. In particular, if A is one vector potential for
8 Does this remind you of Theorem 2.4.8? It should.
CHAPTER 4. INTEGRAL THEOREMS 187
yz 2
A2 = − + N (x, y)
2
∂
Note that, because ∂z treats x and y as constants, the constant of integration
N is allowed to depend on x and y.
9 There is nothing special about the subscript 3 here. By precisely the same argument,
we could come up with another vector potential whose second component is zero, and with
a third vector potential whose first component is zero.
CHAPTER 4. INTEGRAL THEOREMS 188
xz 2 yz 2 x2 y
A= ı̂ı + − + ̂
2 2 2
One can, and indeed should, quickly check that ∇ × A = B.
Let’s do another.
Example 4.1.15 Let
A2 = −2xz + N (x, y)
A1 = z 2 − 2xz + M (x, y)
CHAPTER 4. INTEGRAL THEOREMS 189
Finally, the third equation is also satisfied if and only if M (x, y) and N (x, y)
obey
∂ ∂ 2
− 2xz + N (x, y) − z − 2xz + M (x, y)) = 2x − 2z
∂x ∂y
∂N ∂M
⇐⇒ −2z + (x, y) − (x, y) = 2x − 2z
∂x ∂y
∂N ∂M
⇐⇒ (x, y) − (x, y) = 2x
∂x ∂y
All of the z’s in this equation have cancelled out10 , and we can choose, for
example, M (x, y) = 0 and N (x, y) = x2 . So we have found a vector potential.
Namely
A = (z 2 − 2xz)ı̂ı + (x2 − 2xz)̂
Again it is a good idea to check that ∇ × A = B.
We can use exactly the strategy of the last examples to prove
Theorem 4.1.16 Let B be a vector field that is defined and has all of its first
order partial derivatives continuous on all of R3 . Then there exists a vector
potential for B if and only if it passes the screening test ∇ · B = 0.
Proof. We already know that the existence of a vector potential implies that
∇ · B = 0. So we just have to assume that ∇ · B = 0 and prove that this implies
the existence of a vector field A that obeys ∇ × A = B. Hence we need to
solve
∂A3 ∂A2
− = B1 (x, y, z)
∂y ∂z
∂A3 ∂A1
− + = B2 (x, y, z)
∂x ∂z
∂A2 ∂A1
− = B3 (x, y, z)
∂x ∂y
We’ll explicitly find such an A using exactly the strategy of Example 4.1.14.
In particular, we’ll look for an A that also has A3 = 0. Then the equations
simplify to
∂A2
− = B1 (x, y, z)
∂z
∂A1
= B2 (x, y, z)
∂z
∂A2 ∂A1
− = B3 (x, y, z)
∂x ∂y
The first equation is satisfied if and only if
Z z
A2 (x, y, z) = − B1 (x, y, z̃) dz̃ + N (x, y)
0
for some function N (x, y). And the second equation is satisfied if and only if
Z z
A1 (x, y, z) = B2 (x, y, z̃) dz̃ + M (x, y)
0
10 Of course, we could equally well pick A1 = 0 or A2 = 0.
11 Ifthe z’s had not cancelled out, no N (x, y) and M (x, y), which after all are independent
of z, could satisfy the equation. That would have been a sure sign of a user error.
CHAPTER 4. INTEGRAL THEOREMS 190
So all three equations are satisfied if and only only if we can find M (x, y) and
N (x, y) that obey
A2 (x,y,z)
z Z }| {
z
∂
− B1 (x, y, z̃) dz̃ + N (x, y)
∂x 0
A1 (x,y,z)
zZ }| {
∂ z
− B2 (x, y, z̃) dz̃ + M (x, y) = B3 (x, y, z)
∂y 0
Oof! At first sight, it looks like we have a very big problem here. No matter
what N and M we pick the left hand side will depend on x and y only — not
on z. But it appears like the right hand side depends on z too. Fortunately
the screening test (which we have not used to this point in the proof) rides to
the rescue and ensures that the right hand actually does does not depend on
z. By the screening test,
∂B1 ∂B2 ∂B3
∇·B= + + =0
∂x ∂y ∂z
and we have
∂B1 ∂B2 ∂B3
+ =−
∂x ∂y ∂z
so that the right hand side is
Z z iz̃=z
∂B3 h
B3 (x, y, z) + − (x, y, z̃) dz̃ = B3 (x, y, z) + − B3 (x, y, z̃)
0 ∂z z̃=0
= B3 (x, y, 0)
Recall that θ is the angle between our direction of motion and the gradient
vector ∇ f (r0 ). So to maximize the rate of change of temperature that we feel,
as we pass through r0 , we should choose our direction of motion to be the
direction of the the gradient vector ∇ f (r0 ). In conclusion
Equation 4.1.18
4 3
• 3 πεis the volume of the interior of the sphere Sε and
• by Lemma 3.4.1, Sε v(x, y, z) · n̂(x, y, z) dS is the rate11 at which fluid
RR
is exiting Sε
to conclude that
Equation 4.1.19
v(x, y, z) = A + B x + C y + D z + R(x, y, z)
where
∂v ∂v ∂v
A = v(0, 0, 0) B= (0, 0, 0) C= (0, 0, 0) D= (0, 0, 0)
∂x ∂y ∂z
So
ZZ
v(x, y, z) · n̂(x, y, z) dS
Sε
ZZ
1
= A + B x + C y + D z + R(x, y, z) · (x, y, z) dS
ε Sε
A · ı̂ı x + A · ̂ y + A · k̂ z
2
+ B · ı̂ı x + B · ̂ xy + B · k̂ xz
+ C · ı̂ı xy + C · ̂ y 2 + C · k̂ yz
+ D · ı̂ı xz + D · ̂ yz + D · k̂ z 2
+ R(x, y, z) · (x, y, z)
11 Lemma 3.4.1 is being applied with the density ρ set equal to one, so, more precisely, the
rate is the number of units of volume of fluid exiting Sε per unit time
CHAPTER 4. INTEGRAL THEOREMS 193
That’s a lot of terms. But most of them integrate to zero, simply because the
integral of an odd function over an even domain is zero. Because Sε is invariant
under x → −x and under y → −y and under z → −z we have
ZZ ZZ ZZ ZZ ZZ ZZ
x dS = y dS = z dS = xy dS = xz dS = yz dS = 0
Sε Sε Sε Sε Sε Sε
As well Sε is invariantc under the interchange of x and y and also under the
interchange of x and z. Consequently
ZZ ZZ ZZ ZZ
2 2 2 1 2
x + y 2 + z 2 dS
x dS = y dS = z dS =
Sε 3
ZSZε Sε Sε
1
= ε2 dS since x2 + y 2 + z 2 = ε2 on Sε
3 Sε
4
= πε4
3
since the surface area of the sphere Sε is 4πε2 . So far, we have
ZZ
4
v(x, y, z) · n̂(x, y, z) dS = πε3 B · ı̂ı + C · ̂ + D · k̂
Sε 3
ZZ
1
+ R(x, y, z) · (x, y, z) dS
ε Sε
ZZ
4 1
= πε3∇ · v(0) + R(x, y, z) · (x, y, z) dS
3 ε Sε
(review the definitions of B, C, D)
which implies
ZZ
1
lim v(x, y, z) · n̂(x, y, z) dS
ε→0 4 πε3 Sε
3
ZZ
3
= ∇ · v(0) + lim R(x, y, z) · (x, y, z) dS
ε→0 4πε4 Sε
Finally, it suffices to recall that |R(x, y, z)| ≤ Kε2 and, on Sε , |(x, y, z)| = ε,
so that
ZZ ZZ
3 ≤ 3
R(x, y, z) · (x, y, z) dS |R(x, y, z)| |(x, y, z)| dS
4πε4 Sε 4πε4
Sε
ZZ
3 3
≤ Kε3 dS = Kε3 4πε2 )
4πε4 Sε 4πε 4
= 3Kε
converges to zero as ε → 0. So we are left with the desired result.
a There is another, easier to understand, proof of this result given in §4.4.1. We cannot
give that proof here because it uses the divergence theorem, which we will get to later in the
chapter.
b Terms like xy, xz and yz are not needed because, for example, |xy| ≤ 1 (x2 + y 2 ). This
2 2
inequality is equivalent to |x| − |y| ≥ 0.
c Spheres have lots of symmetry!
CHAPTER 4. INTEGRAL THEOREMS 194
Example 4.1.21 Here is a sketch of the vector field v(x, y, z) = xı̂ı + y ̂ + z k̂
and a sphere centered on the origin, like Sε .
This velocity field has fluid being created and pushed out through the
sphere. We have
∇ · v(0) = 3
This velocity field just has fluid going around in circles. No fluid actually
crosses the sphere. The divergence
∇ · v(0) = 0
This velocity field just has fluid moving uniformly to the right. Fluid enters
the sphere from the left and leaves through the right at precisely the same rate,
so that the net rate at fluid crosses the sphere is zero. The divergence
∇ · v(0) = 0
ε
Cε
We shall show in Lemma 4.1.25, below, that
I
1
∇ × v(r0 ) · n̂ = lim 2 v(r) · dr
ε→0 πε Cε
t̂
ε
Cε
the paddlewheel is really expensive and has a lot more than just four pad-
dles. Pretend12 that you are one of the paddles.
• If the paddlewheel is rotating at Ω radians per unit time, then in one
unit of time you sweep out an arc of a circle of radius ε that subtends an
angle Ω. That arc has length Ωε. So you are moving at speed Ωε.
• If you are at r, the component of the fluid velocity in your direction of
dr dr
motion, i.e. tangential to Cε , is v(r) · ds , because t̂ = ds , with s denoting
arc length along the circle, is a unit vector tangential to Cε .
• All paddles have to move at the same speed. So the speed of the paddles,
dr
Ωε, should be the average value of v(r) · ds around the circle.
Thus the rate of rotation, Ω, of the paddlewheel should be determined by
dr
H H
Cε
v(r) · ds ds v(r) · dr
Ωε = H = Cε
Cε
ds 2πε
12 Method acting might help you here.
CHAPTER 4. INTEGRAL THEOREMS 196
v(x, y, z) = A + B x + C y + D z + R(x, y, z)
where
∂v ∂v ∂v
A = v(0, 0, 0) B= (0, 0, 0) C= (0, 0, 0) D= (0, 0, 0)
∂x ∂y ∂z
So
I
v(r) · dr
Cε
Z 2π
= A + B ε cos t + C ε sin t + R(r(t)) · − ε sin tı̂ı + ε cos t ̂ dt
0
Again, multiply out the dot product so that the integrand becomes
and the sin2 t and cos2 t terms are easily integrated using (see Example 2.4.4)
Z 2π Z 2π Z 2π
1
sin2 t dt = cos2 t dt = sin2 t + cos2 t dt = π
0 0 2 0
1 2π
Z Z 2π
≤ 1
R(r(t)) · − ε sin ı
tı̂ + ε cos t
̂ dt |R(r(t))| dt
πε2 0 πε
0
Z 2π
1
≤ Kε2 dt
πε 0
1
= Kε2 (2π)
πε
= 2Kε
converges to zero as ε → 0.
a There is another, easier to understand, proof of this result given in §4.4.1. We cannot
give that proof here because it uses Stokes’ theorem, which we will get to later in the chapter.
Here are some examples. We will use the same vector fields as in Examples
4.1.21, 4.1.22 and 4.1.23. In all examples, we shall orient the paddlewheel so
that n̂ = k̂ and sketch the top view, so that the paddlewheel looks like
Example 4.1.26 Here is a sketch of the vector field v(x, y, z) = xı̂ı + y ̂ + z k̂
and a circle centered on the origin, like Cε .
CHAPTER 4. INTEGRAL THEOREMS 198
This velocity field has fluid moving parallel to the paddles, so the paddle-
wheel should not rotate at all. The computation
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × v(0) = det ∂x ∂y ∂z = 0 =⇒ ∇ × v(0) · k̂ = 0
x y z
The fluid pushing on the top paddle tries to make the paddlewheel rotate
clockwise. The fluid pushing on the bottom paddle tries to make the paddle-
wheel rotate counterclockwise, at the same rate. So the paddlewheel should
not rotate at all. Our interpretation 4.1.24 predicts an angular velocity of
ı̂ı ̂ k̂
1 1 ∂ ∂ ∂
· ∇ × v(0) · k̂ = det ∂x ∂y ∂z · k̂ = 0 · k̂ = 0
2 2
1 0 0
as expected.
18 Even for small values of 2.
CHAPTER 4. INTEGRAL THEOREMS 199
4.1.6 Exercises
Exercises — Stage 1
1. ∗. Let F = P ı̂ı + Q ̂ be the two dimensional vector field shown below.
a Assuming that the vector field in the picture is a force field, the
work done by the vector field on a particle moving from point
A to B along the given path is:
(A) Positive
(B) Negative
(C) Zero
(D) Not enough information to determine.
R
b Which statement is the most true about the line integral C2
F·
dr:
Z
(A) F · dr > 0
C2
Z
(B) F · dr = 0
C2
Z
(C) F · dr < 0
C2
(D) Not enough information to determine.
y
N
C2 B
G
x
c ∇ · F at the point N (in the picture) is:
(A) Positive
(B) Negative
(C) Zero
(D) Not enough information to determine.
CHAPTER 4. INTEGRAL THEOREMS 200
c ∇ 2 (f g) = f ∇ 2 g + 2∇
∇f · ∇ g + g ∇ 2 f
Exercises — Stage 2
4. Evaluate ∇ · F and ∇ × F for each of the following vector fields.
a F = xı̂ı + y ̂ + z k̂
b F = xy 2ı̂ı − yz 2̂ + zx2 k̂
c F = √xı̂ı+ŷ
2
2
(the polar basis vector r̂ in 2d)
x +y
−yı̂ı+x̂
d F= √ 2 2
(the polar basis vector θ̂θ in 2d)
x +y
5. ∗.
a Compute and simplify ∇ · rr for r = (x, y, z) and r = |(x, y, z)|.
∇ · (rk r) = 5r2
7. ∗. Let r be the vector field r = xı̂ı + y ̂ + z k̂ and let r be the function
r = |r|. Let a be the constant vector a = a1 ı̂ı + a2 ̂ + a3 k̂. Compute
and simplify the following quantities. Answers must be expressed in
terms of a, r, and r. There should be no x’s, y’s, or z’s in your
answers.
a ∇·r
b ∇ (r2 )
c ∇ × (r × a)
d ∇ · ∇ (r)
8. ∗. Let
r = xı̂ı + y ̂ + z k̂, r = |r|
b Compute a where ∇ · r r = ar.
c Compute a where ∇ · ∇ (r3 ) = ar.
9. Find, if possible, a vector field A that has k̂ component A3 = 0 and
that is a vector potential for
a F = (1 + yz)ı̂ı + (2y + zx)̂ + (3z 2 + xy)k̂
b G = yzı̂ı + zx̂ + xy k̂
Exercises — Stage 3
10. ∗. Let
−z x
F= ı̂ı + y ̂ + 2 k̂
x2 + z 2 x + z2
The left hand side of the fundamental theorem of calculus is the integral of the
derivative of a function. The right hand side involves only values of the function
on the boundary of the domain of integration. The divergence theorem, Green’s
theorem and Stokes’ theorem also have this form, but the integrals are in more
than one dimension. So the derivatives are multidimensional, like the curl and
divergence, and the integrands can involve vector fields.
• For the divergence theorem, the integral on the left hand side is over a
(three dimensional) volume and the right hand side is an integral over
the boundary of the volume, which is a surface.
• For Green’s and Stokes’ theorems, the integral on the left hand side is
over a (two dimensional) surface and the right hand side is an integral
over the boundary of the surface, which is a curve.
The divergence theorem is going to relate a volume integral over a solid
V to a flux integral over the surface of V . First we need a couple of defini-
tions concerning the allowed surfaces. In many applications solids, for example
cubes, have corners and edges where the normal vector is not defined. On the
other hand, to be able to compute a flux integral over a surface, we certainly
need that the set of points where the normal vector is not well-defined is small
enough that the existence of the flux integral is not jeopardized. This is the
case for “piecewise smooth” surfaces, which we now define.
CHAPTER 4. INTEGRAL THEOREMS 203
Definition 4.2.1
a A surface is smooth if it has a parametrization r(u, v) with continuous
∂r ∂r ∂r ∂r
partial derivatives ∂u and ∂v and with ∂u × ∂v nonzero.
b A surface is piecewise smooth if it consists of a finite number of smooth
pieces that meet along sharp curves and at sharp corners.
♦
Here are sketches of a smooth surface (a sausage) and a piecewise smooth
surface (an ice-cream cone), followed by the divergence theorem1 .
Note that the left hand side is a sum of three terms — one involving F1 , one
involving F2 and one involving F3 — and the right hand side is a sum of three
terms — one involving F1 , one involving F2 and one involving F3 . We’ll just
show that the F3 terms on the left hand side and right hand side are equal,
1 It is also known as Gauss’s theorem. Johann Carl Friedrich Gauss (1777–1855) was a
German mathematician. Throughout the 1990’s Gauss’s portrait appeared on the German
ten-mark banknote. In addition to Gauss’s theorem, the Gaussian distribution (the bell
curve), degaussing and the CGS unit for the magnetic field, and the crater Gauss on the
Moon are named in his honour.
2 We are going to consistently use the notation ∂(thing) to denote the boundary of (thing).
CHAPTER 4. INTEGRAL THEOREMS 204
i.e. that
ZZ ZZZ
∂F3
F3 k̂ · n̂ dS = dV
∂V V ∂z
Showing that the F1 terms match and the F2 terms match is done in the same
waya .
Special Geometry
We’ll first assume that the solid has the special form
V = (x, y, z) B(x, y) ≤ z ≤ T (x, y), (x, y) ∈ Rxy
where Rx,y is some subset of the xy-plane. We can further assume that, for
each (x, y) ∈ Rxy , we have B(x, y) ≤ T (x, y). After we’re finished with this
special case, we’ll RR
handle the general case.
Let’s work on ∂V F3 k̂ · n̂ dS first. As in the figure below,
z
z “ T px, yq
S
B
z “ Bpx, yq
y
Rxy
x BRxy
the surface ∂V consists of three pieces — the top, the bottom and the side.
We’ll consider each in turn.
• The top is T = (x, y, z) z = T (x, y), (x, y) ∈ Rxy . By 3.3.2, on T
n̂ dS = + − Tx (x, y)ı̂ı − Ty (x, y) ̂ + k̂ dxdy
So all together
ZZ ZZ ZZ ZZ
F3 k̂ · n̂ dS = F3 k̂ · n̂ dS + F3 k̂ · n̂ dS + F3 k̂ · n̂ dS
∂V T B S
ZZ
= F3 (x, y, T (x, y)) − F3 (x, y, B(x, y)) dxdy + 0 (∂V )
Rxy
V1
S1
Sc
S2
V2
Call the sausage V . Cut it into two halves by running a cleaver horizontally
through its centre. This splits the solid V into two halves, V1 and V2 as in the
figure on the right above. It also splits the boundary ∂V of V into two halves
S1 and S2 , also as in the figure on the right above. Note that
• the boundary, ∂V1 , of V1 is the union of S1 and the shaded disk Sc (the
cut introduced by the cleaver). On the cut Sc , the outward pointing
normal to V1 is −k̂.
• The boundary, ∂V2 , of V2 is the union of S2 and the shaded disk Sc . On
the cut Sc , the outward pointing normal to V2 is +k̂.
Now both V1 and V2 do satisfy the assumption of the “Special Geometry”
section above. So
ZZZ ZZZ ZZZ
∂F3 ∂F3 ∂F3
dV = dV + dV
V ∂z V1 ∂z V2 ∂z
ZZ ZZ
= F3 k̂ · n̂ dS + F3 k̂ · n̂ dS
∂V1 ∂V2
CHAPTER 4. INTEGRAL THEOREMS 206
ZZ ZZ ZZ
= F3 k̂ · n̂ dS + F3 k̂ · n̂ dS + F3 k̂ · n̂ dS
S1 Sc ↓ S2
ZZ
+ F3 k̂ · n̂ dS
Sc ↑
Here Sc ↓ is the surface Sc with normal vector −k̂ and Sc ↑ is the surface Sc
with normal vector +k̂. So the second and fourth integrals are identical except
that n̂ = −k̂ in the second integral and n̂ = +k̂ in the fourth integral. So they
cancel exactly and
ZZZ ZZ ZZ ZZ
∂F3
dV = F3 k̂ · n̂ dS + F3 k̂ · n̂ dS = F3 k̂ · n̂ dS
V ∂z S1 S2 ∂V
as desired.
a Mutatis mutandis.
b We are assuming that V is “reasonable”.
CHAPTER 4. INTEGRAL THEOREMS 207
RR
Example 4.2.4 Evaluate the flux integral S F · n̂ dS where n̂ is the outward
normal to S, which is the surface of the hemispherical region
n̂
V
V =
(x, y, z) x2 +y 2 +z 2 ≤ a2 , z ≥ 0
S
n̂
and
F = xz 2 ı̂ı + (x2 y − z 3 ) ̂ + 2xy + y 2 z + ecos y k̂
Spherical coordinates are perfect for this integral. (See Appendix A.6.3, if you
need to refresh your memory.)
ZZZ Z 2π Z π2 Z a
2 2 2
dρ ρ2 sin ϕ ρ2
x + y + z dV = dθ dϕ
V 0 0 0
Z 2π Z π Z a
2
4
= dθ sin ϕ dϕ ρ dρ
0 0 0
h i π2 ρ5 a
= 2π − cos ϕ
0 5 0
2πa5
=
5
RR
Example 4.2.5 Evaluate the flux integral S F · n̂ dS where n̂ is the outward
normal to S, which is the part of the surface z 2 = x2 + y 2 with 1 ≤ z ≤ 2, and
where
F = 3xı̂ı + (5y + ecos x ) ̂ + z k̂
D2 n̂ “ k̂
S S V
n̂ n̂
D1
n̂ “ ´k̂
V = (x, y, z) x2 + y 2 ≤ z 2 , 1 ≤ z ≤ 2
and is sketched in the figure on the right above. The boundary, ∂V , is the
union of S and the two disks
D1 = (x, y, z) x2 + y 2 ≤ z 2 , z = 1
D2 = (x, y, z) x2 + y 2 ≤ z 2 , z = 2
which implies
ZZ ZZZ ZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS − F · n̂ dS
S V D1 D2
The point of this exercise is that the left hand side, which is not easy to evaluate
directly, is the integral we want, while the three integrals on the right hand side
are all easy to evaluate. We do so now. The outward normal to (the horizontal
disk) D2 is +k̂. So
ZZ ZZ ZZ
F · n̂ dS = F · k̂ dS = z dS
D2 D2 D2
Finally, as ∇ · F = 3 + 5 + 1 = 9
ZZZ
∇ · F dV = 9 Vol(V )
V
CHAPTER 4. INTEGRAL THEOREMS 209
The volume of V can be easily computed using the first year technique5 of
slicing V into thin horizontal pancakes like that sketched in the figure below.
z“2
z“1
The pancake at height z has
• thickness dz,
So
2 2
πz 3
ZZZ Z
2 7
∇ · F dV = 9 Vol(V ) = 9 πz dz = 9 =9×π = 21π
V 1 3 1 3
RR
Example 4.2.6 Evaluate the flux integral S
F · n̂ dS where n̂ is the upward
2 2 2
normal to S, which is the part of z = x + y with 0 ≤ z ≤ 1, and
2
F = x + ey
ı̂ı + (y + cos z) ̂ + k̂
Solution. This integral can be evaluated in much the same way as we eval-
uated the integral of Example 4.2.5. We first define a solid V whose boundary
∂V contains S. A good, and hopefully obvious, choice is
2
(x, y, z) x2 + y 2 ≤ z, 0 ≤ z ≤ 1
V =
k̂
D
z“1
n̂ S
which implies
ZZ ZZZ ZZ
F · n̂ dS = − ∇ · F dV + F · k̂ dS
S
Z Z ZV ZZ D
=− 2 dV + dS
V D
Again, you can see that the actual integration is quite easy. All of the work
(or at least all of the thinking) happens in the setup.
Example 4.2.7 In Warning 4.2.3 we emphasised that the conclusion of the
divergence Theorem 4.2.2 can fail if the vector field F is not defined at even a
single point of V . Here is an example. Set
r
F= where r = xı̂ı + y ̂ + z k̂
|r|3
and V = (x, y, z) x2 + y 2 + z 2 ≤ 1 . Then, if (x, y, z) 6= 0,
∂ x ∂ y
∇ · F(x, y, z) = +
∂x x2 + y 2 + z 2 3/2 ∂y x2 + y 2 + z 2 3/2
∂ z
+
∂z x2 + y 2 + z 2 3/2
x + y 2 + z 2 − x 23 (2x) x + y 2 + z 2 − y 23 (2y)
2 2
= 5/2 + 5/2
x2 + y 2 + z 2 x2 + y 2 + z 2
x + y 2 + z 2 − z 23 (2z)
2
+ 5/2
x2 + y 2 + z 2
=0
On the other hand, the boundary of V is the unit sphere ∂V = (x, y, z) x2 +
CHAPTER 4. INTEGRAL THEOREMS 211
r
y 2 + z 2 = 1 . The outward unit normal to ∂V is n̂ = |r| so that
Z Z Z Z
r r 1
F · n̂ dS = · dS = dS = dS
∂V |r|=1 |r|3 |r| |r|=1 |r|2 |r|=1
= 4π 6= 0
dS
∇T hot
the temperature gradient, which points in the direction of increasing tem-
perature, is opposite n̂. Consequently the flow rate −κn̂·∇∇T (x, y, z, t) dS
is positive, indicating flow in the direction of n̂. This is just what you
would expect — heat flows from hot regions to cold regions. Also the rate
of flow increases as the magnitude of the temperature gradient increases.
This also makes sense (and is reminiscent of Newton’s law of cooling).
6 The heat equation was formulated by the French mathematician and physicist Jean-
Baptiste Joseph Fourier in 1807. He lived from 1768 to 1830, a period which included both
the French revolution and the reign of Napoleon. Indeed Fourier served on his local Revolu-
tionary Committee, was imprisoned briefly during the Terror, and was Napoleon Bonaparte’s
scientific advisor on his Egyptian expedition of 1798. Fourier series and the Fourier transform
are named after him. Fourier is also credited with discovering the greenhouse effect.
7 Heat is now understood to arise from the internal energy of the object. In an earlier
theory, heat was viewed as measuring an invisible fluid, called the caloric. The amount of
caloric that an object could hold was called its “heat capacity” by the Scottish physician
and chemist Joseph Black (1728–1799).
CHAPTER 4. INTEGRAL THEOREMS 212
Assuming that the object is not generating or destroying8 heat itself, this must
be same as the amount of heat that entered V in the time interval dt. That is
ZZ ZZZ
∂T
κn̂ · ∇ T dS dt = Cρ dV dt
∂V V ∂t
Now we cancel the common factor of dt. We can then rewrite the left hand
side as an integral over V by applying the divergence theorem giving
ZZZ ZZZ
∂T
∇ · ∇ T dV =
κ∇ Cρ dV
V V ∂t
all volumes V in the object and for all times t. We claim that this forces
∂T
∇2 T (x, y, z, t) − Cρ
κ∇ (x, y, z, t) = 0
∂t
for all (x, y, z) in the object and all t.
Suppose that to the contrary there was a point (x0 , y0 , z0 ) in the object and
a time t0 with, for example, κ∇ ∇2 T (x0 , y0 , z0 , t0 ) − Cρ ∂T
∂t (x0 , y0 , z0 , t0 ) > 0. By
continuity, which we are assuming, κ∇ ∇2 T (x, y, z, t0 ) − Cρ ∂T ∂t (x, y, z, t0 ) must
8 The caloric theory of heat was itself destroyed by the cannon boring experiment of 1798.
which violates (H) for V = B. This completes our derivation of the heat
equation, which is
Equation 4.2.8
∂T
∇2 T (x, y, z, t)
(x, y, z, t) = α∇
∂t
where α = κ
Cρ is called the thermal diffusivity.
∂T ∂2T
(z, t) = α 2 (z, t) (HE)
∂t ∂z
We choose a coordinate system having the surface of the Earth at z = 0 and
having z increase downward. We also assume that the temperature T (0, t) at
the surface of the Earth is primarily determined by solar heating and is given
by
T (0, t) = T0 + TA cos(σt) + TD cos(δt) (BC)
Here T0 is the long term average of the temperature at the surface of the
Earth, TA cos(σt) gives seasonal temperature variations and TD cos(δt) gives
daily temperature variations.
air
z “ 0, T p0, tq “ T0 ` TA cospσtq ` TD cospδtq
earth
Tt “ αTzz
z
Now let’s see what we can learn from the solution (SLN). For any fixed
z, the time average of T (z, t) is T0 (just because the average value if cosine
is zero), the same as the average temperature at the surface z = 0. That is,
under the hypotheses that we have made, the long term average temperature
at any depth z is is the same as the long term average temperature at the
surface. √σ pσ
The term TA e− 2α z cos σt − 2α z
To get the second line, we used the vector identity Theorem 4.1.4.c. To get the
third line, we just used that a is a constant, so that all of its derivatives are
zero. Rewrite ZZZ ZZZ
∇f ) · a dV =
(∇ ∇f ) dV
a · (∇
V V
Since a is a constant, we can factor it out of both integrals, so
ZZ ZZZ
a· f n̂ dS = a · ∇ f dV
∂V V
ZZ ZZZ
=⇒ a · f n̂ dS − ∇ f dV = 0
∂V V
In particular,
RR RRRa = ı̂ı, ̂ and k̂, we see that all three components of the
choosing
vector ∂V f n̂ dS − V
∇ f dV are zero. So
ZZ ZZZ
f n̂ dS − ∇ f dV = 0
∂V V
To get the second line, we used the vector identity Theorem 4.1.4.d. To get
the third line, we again used that a is a constant, so that all of its derivatives
are zero. For all vectors (a × b) · c = a · (b × c) (in case you don’t remember
this, it was Lemma 4.1.8.a) so that
(a × F) · n̂ = a · (F × n̂)
CHAPTER 4. INTEGRAL THEOREMS 216
and
ZZ ZZZ
a· F × n dS = −a · ∇ × F dV
Z∂V
Z ZZZ V
=⇒ a · F × n dS + ∇ × F dV = 0
∂V V
In particular,
RR = ı̂ı, ̂ and k̂, we see that all three components of the
choosing aRRR
vector ∂V F × n dS + V
∇ × F dV are zero. So
ZZZ ZZ ZZ
∇ × F dV = − F × n dS = n̂ × F dS
V ∂V ∂V
• is the force per unit surface area that the fluid exerts on the object
• acts perpendicularly to the surface
• pushes on the object
Thus the force due to pressure that acts on an infinitesimal piece of the ob-
ject’s surface at r = (x, y, z) with surface area dS and outward normal n̂ is
−p(r) n̂dS. The minus sign is there because pressure is directed into the ob-
ject. If the object fills the volume V and has surface ∂V , then the total force
on the object due to fluid pressure, called the buoyant force, is
ZZ
B=− p(r) n̂ dS
∂V
WeRRR
now wish to apply a variant of the divergence theorem to rewrite B =
− V
∇ p dV . But there is a problem with this: p(r) is the fluid pressure at r
and is only defined where there is fluid. In particular, there is no fluid12 inside
the object, so p(r) is not defined for any r in the interior of V .
So we pretend that we remove the object from the fluid and we call P (r)
the fluid pressure at r when there is no object in the fluid. We also make the
assumption that at any point r outside of the object, the pressure at r does
10 The interested reader should do a net search for the story of Archimedes and the golden
crown.
11 The first design of a self-righting boat was entered by William Wouldhave in a lifeboat
not depend on whether the object is in the fluid or not. In other words, we
assume that (
P (r) if r is not in V
p(r) =
not defined if r is in the V
This assumption is only an approximation to reality, but, in practice, it is a
very good approximation. So, by Theorem 4.2.9,
ZZ ZZ ZZZ
B=− p(r) n̂ dS = − P (r) n̂ dS = − ∇ P (r) dV (4.2.1)
∂V ∂V V
´P px, y, z ` dz
2
q dxdy k̂
dy dy
P px, y ´ 2
, zq dxdz ̂ ´P px, y ` 2
, zq dxdz ̂
P px, y, z ´ dz
2
q dxdy k̂
The total force due to the pressure acting on the cube is the sum
dx dx
−P x+ , y, z dydz ı̂ı + P x − , y, z dydz ı̂ı
2 2
dy dy
− P x, y + , z dxdz ̂ + P x, y − , z dxdz ̂
2 2
dz dz
− P x, y, z + dxdy k̂ + P x, y, z − dxdy k̂
2 2
of the forces acting on the six faces. Consider the ı̂ı component and rewrite it
as
dx dx
−P x+ , y, z dydz ı̂ı + P x − , y, z dydz ı̂ı
2 2
dx dx
P (x + 2 , y, z) − P (x − 2 , y, z)
=− ı̂ı dxdydz
dx
∂P
=− (x, y, z)ı̂ı dxdydz
∂x
Doing this for the other components as well, we see that the total force due
to the pressure acting on the cube is
n ∂P ∂P ∂P o
− (x, y, z)ı̂ı + (x, y, z) ̂ + ∇P (x, y, z) dxdydz
(x, y, z) k̂ dxdydz = −∇
∂x ∂y ∂z
We shall assume that the only other force acting on the cube is gravity and
that the fluid is stationary (or at least not accelerating). Hence the total force
acting on the cube is zero. If the fluid has density ρf , then the cube has mass
ρf dxdydz so that the force of gravity is −gρf dxdydz k̂. The vanishing of the
total force now tells us that
So the torque generated at r0 by pressure over the entire surface is the same
the torque generated at r0 by a force B applied at the single point
RRR
rρf dV
CB = RRRV
ρ dV
V f
This point is called the centre of buoyancy. It is the centre of mass of the
displaced fluid.
The moral of the above discussion is that the buoyant force, B, on a rigid
body
• acts straight upward,
Because the mass distribution of the object need not be the same as the
mass distribution of the displaced fluid, buoyancy and gravity may act at two
different points. This is exploited in the design of self-righting boats.
These boats are constructed with a heavy, often lead (which is cheap and
dense), keel. As a result, the centre of gravity is lower in the boat than the
center of buoyancy, which, because the displaced fluid has constant density, is
at the geometric centre of the boat. As the figure below illustrates, a right side
up configuration of such a boat is stable, while an upside down configuration
is unstable. The boat rotates so as to keep the centre of gravity straight below
the centre of buoyancy. To see this pretend that you are holding on to the boat
with one hand holding the centre of buoyancy and the other hand holding the
CHAPTER 4. INTEGRAL THEOREMS 220
centre of gravity. Use your hands to apply forces in the directions of the arrows
and think about how the boat will respond.
B
B
G G
• at time t, the force acting on particle number i, due to the steel rod
joining particle number i to particle number j is Fi,j (t). If there is
no steel rod joining particles number i and j, just set Fi,j (t) = 0. In
particular, Fi,i (t) = 0.
14 Just 12 grams of carbon contains about 6 × 1023 atoms.
15 Mathematicians and their idealizations! Really the rods just represent the atomic/
chemical forces that hold the wood together.
CHAPTER 4. INTEGRAL THEOREMS 221
The only assumptions that we shall make about the steel rod forces are
(A1) for each i 6= j, Fi,j (t) = −Fj,i (t). In words, the steel rod joining particles
i and j applies equal and opposite forces to particles i and j.
i 6= j, there is a function Mi,j (t) such that Fi,j (t) = Mi,j (t) ri (t)−
(A2) for each
rj (t) . In words, the force due to the rod joining particles i and j
acts parallel to the line joining particles
i and j. For (A1) to be true,
that is to have Mi,j (t) ri (t) − rj (t) = −Mj,i (t) rj (t) − ri (t) , we need
Mi,j (t) = Mj,i (t).
Newton’s law of motion, applied to particle number i, now tells us that
n
X
mi r00i (t) = Fi (t) + Fi,j (t) (Ni )
j=1
particles.
CHAPTER 4. INTEGRAL THEOREMS 222
r2 (t) × F2,1 (t) = M2,1 (t) r2 (t) × r2 (t) − r1 (t)
= −M1,2 (t) r2 (t) × r1 (t) − r2 (t)
so that
r1 (t) × F1,2 (t) + r2 (t) × F2,1 (t) = M1,2 (t) r1 (t) − r2 (t) × r1 (t) − r2 (t) = 0
because the cross product of any two parallel vectors is zero.
P
The last equation says that the i = 1, j = 2 term in ri (t) × Fi,j (t)
1≤i,j≤n
exactly
P cancels the i = 2, j = 1 term. In this way all of the terms in
ri (t) × Fi,j (t) with i 6= j cancel. Each term with i = j is exactly
1≤i,j≤n P
zero because Fii = 0. So ri (t) × Fi,j (t) = 0 and (Σri × Ni ) simplifies
1≤i,j≤n
to
n
X n
X
mi ri (t) × r00i (t) = ri (t) × Fi (t) (Σri × Ni )
i=1 i=1
At this point it makes sense to define vectors
n
X
L(t) = mi ri (t) × r0i (t)
i=1
n
X
T(t) = ri (t) × Fi (t)
i=1
= mρ2 ω k̂
is proportional to ω, which is the rate of rotation about the origin and is in
the direction k̂, which is normal to the plane containing the circle.
In any event, in order for the piece of wood to remain stationary, equations
4.2.10 and 4.2.11 force F(t) = T(t) = 0.
Now suppose that the piece of wood is a seesaw17 that is supported on a
fulcrum at p. The forces consist of gravity, −mi g k̂, acting on particle number
i, for each 1 ≤ i ≤ n, and the
Φ
If the seesaw is to remain stationary, this must also be zero. This will be the
case if the fulcrum is placed at
n
1 X
p= mi ri
M i=1
n
X n
X
T=p×Φ+ ri × Fi = (ri − p) × Fi
i=1 i=1
If the piece of wood is to remain stationary, this must also be zero. That is, the
torque about point p due to all of the forces Fi , 1 ≤ i ≤ n, must be zero.
∇ 2 ϕ = 4πρ
∇ 2 ϕ(r) 3
ZZZ
r − r0
ZZ
1
ϕ(r0 ) = − d r− ϕ(r) · n̂ dS
4π V |r − r 0 | ∂V |r − r0 |3
CHAPTER 4. INTEGRAL THEOREMS 224
∇ ϕ(r)
ZZ
− · n̂ dS (V )
∂V |r − r0 |
When we take the limit as V expands to fill all of R3 then, assuming that ϕ
and ∇ ϕ go to zero sufficiently quickly18 at ∞, the two integrals over ∂V will
converge to zero and we will end up with the formula
∇ 2 ϕ(r) 3
ZZZ
1
ϕ(r0 ) = − d r
4π R3 |r − r0 |
for all r0 in R3 .
Let
Derivation of (V):
Here is the derivation of (V ). Let Vε be the part of V outside of Sε .
Sε V
r0 Vε
Note that the boundary ∂Vε of Vε consists of two parts — the boundary
∂V of V and the sphere Sε — and that the unit outward normal to ∂Vε on Sε
r−r0
is − |r−r 0|
, because it points towards r0 and hence outside of Vε .
Recall the vector identity Theorem 4.1.7.d, which says
∇f ) = f ∇ 2 g − g ∇ 2 f
∇g − g∇
∇ · (f∇
1
Applying this identity with f = |r−r0 | and g = ϕ gives
=0 by (P2)
2
}| { z
1 1 ∇ ϕ
2 1
∇· ∇
∇ ϕ − ϕ∇ = −ϕ∇
|r − r0 | |r − r0 | |r − r0 | |r − r0 |
∇2ϕ
=
|r − r0 |
which is the integrand of the first integral on the right hand side of (V). So,
by the divergence theorem
∇2 ϕ
ZZZ ZZZ 1 1
dV = ∇· ∇ ϕ − ϕ∇ ∇ dV
Vε |r − r0 | Vε |r − r0 | |r − r0 |
ZZ
1 1
= ∇
∇ ϕ − ϕ∇ · n̂ dS
∂V |r − r0 | |r − r0 |
r − r0
ZZ
1 1
+ ∇ ϕ − ϕ∇ ∇ · − dS (M)
Sε |r − r0 | |r − r0 | |r − r0 |
To see the connection between (M) and the rest of (V), note that,
• by (P1), the first term on the right hand side of (M) is
ZZ
1 1
∇
∇ ϕ − ϕ∇ · n̂ dS
∂V |r − r0 | |r − r0 |
∇ ϕ(r) r − r0
ZZ ZZ
= · n̂ dS + ϕ(r) · n̂ dS (R1)
∂V |r − r0 | ∂V |r − r0 |3
which is 4π times the second and third terms on the right hand side of
(V),
1 r−r0
• and substituting in ∇ |r−r 0|
= − |r−r 0|
3 , from (P1), and applying (P3)
with p = 2, the limit of the second term on the right hand side of (M) is
r − r0
ZZ
1 1
lim ∇ϕ − ϕ∇ ∇ · − dS
ε→0+ Sε |r − r0 | |r − r0 | |r − r0 |
ZZ
1
= − lim ∇ ϕ · (r − r0 ) + ϕ dS
ε→0+ Bε |r − r0 |2
h i
= −4π ∇ ϕ · (r − r0 ) + ϕ
r=r0
= −4πϕ(r0 ) (R2)
CHAPTER 4. INTEGRAL THEOREMS 226
∇2ϕ ∇ ϕ(r) r − r0
ZZZ ZZ ZZ
dV = · n̂ dS + ϕ(r) · n̂ dS − 4πϕ(r0 )
V |r − r0 | ∂V |r − r0 | ∂V |r − r0 |3
which is exactly equation (V).
4.2.6 Exercises
Exercises — Stage 1
1. Let V be the cube
V = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, 0 ≤ z ≤ 1
b Show that the centroid (x̄, ȳ, z̄) of a solid V with volume |V | is
given by
ZZ
1
(x̄, ȳ, z̄) = (x2 + y 2 + z 2 ) n̂ dS
2|V | ∂V
Exercises — Stage 2
3. Let S be the unit sphere centered at the origin and oriented by the
outward pointing normal. If
F(x, y, z) = x, y, z 2
9 − x2 − y 2
0≤z≤
9 + x2 + y 2
and
2 2
−y
• D be the bottom surface of V . Because 9−x 9+x2 +y 2 is positive for
x2 + y 2 < 9 and negative for x2 + y 2 > 9, the bottom surface is
z = 0, x2 + y 2 ≤ 9.
S
y
D x2 + y 2 = 9
x
Denote by |V | the volume of V and compute, in terms of |V |,
ZZ
a F · n̂ dS with n̂ pointing downward
D
ZZZ
b ∇ · F dV
V
ZZ
c F · n̂ dS with n̂ pointing outward
S
Use the divergence theorem to answer at least one of parts (a), (b)
and (c).
RR
6. Evaluate the integral S F · n̂ dS, where F = (x, y, 1), S is the surface
z = 1 − x2 − y 2 for x2 + y 2 ≤ 1, and n̂ is the upward pointing normal,
by two methods.
a First, by direct computation of the surface integral.
b Second, by using the divergence theorem.
7. ∗.
a Find the divergence of the vector field F = (z+sin y, zy, sin x cos y).
b Find the flux of the vector field F of (a) outward through the
CHAPTER 4. INTEGRAL THEOREMS 228
V(x, y, z) = (x2 yz , yz + ex z , x2 + y)
x2 + y 2 = 1 − z 4
x2 + y 2 + 2z 2 = 2
x2 y2 z2
with z ≥ 0. [Note: the ellipsoid a2 + b2 + c2 = 1 has volume 43 πabc.]
12. ∗. Let F(x, y, z) = r/r3 where r = xı̂ı + y ̂ + z k̂ and r = |r|.
a Find ∇ · F.
y2 z2
x2 + + =1
4 4
with the unit normal pointing outward.
a Parameterize S.
RR
b Compute the flux S
F · n̂ dS of the vector field
F(x, y, z) = (x, y, z)
where F(x, y, z) = cos z+xy 2 , xe−z , sin y+x2 z and S is the bound-
where
p
F = x + cos(z 2 ) ı̂ı + y + ln(x2 + z 5 ) ̂ + x2 + y 2 k̂
22. ∗. Let E be the solid region between the plane z = 4 and the
paraboloid z = x2 + y 2 . Let
1 2
1
F = − x3 + ez ı̂ı + − y 3 + x tan z ̂ + 4z k̂
3 3
xı̂ı + y ̂ + z k̂
F(x, y, z) = 3/2
[x2 + y 2 + z 2
a Compute ∇ · F.
b Let S1 be the sphere given by
x2 + (y − 2)2 + z 2 = 9
ZZ
oriented outwards. Compute F · n̂ dS.
S1
x2 + (y − 2)2 + z 2 = 1
ZZ
oriented outwards. Compute F · n̂ dS.
S2
d Are your answers to (b) and (c) the same or different? Give a
mathematical explanation of your answer.
24. ∗. Let F be the vector field defined by
2 2
F(x, y, z) = y 3 z + 2x ı̂ı + 3y − esin z ̂ + ex +y + z k̂
RR
Calculate the flux integral S F· n̂ dS where S is the boundary surface
of the solid region
E : 0 ≤ x ≤ 2, 0 ≤ y ≤ 2, 0≤z ≤2+y
x2 + y 2 + (z − 12)2 = 132 , z ≥ 0
x2 + y 2 + z 2 = 4, z≥0
2 2
32. ∗. Let D be the cylinder x + y ≤ 1, 0 ≤ z ≤ 5. Calculate the flux
of the vector field
Use the divergence theorem to answer at least one of parts (a), (b)
and (c).
Exercises — Stage 3
35. Let E(r) be the electric field due to a charge configuration that has
density ρ(r). Gauss’ law states that, if V is any solid in R3 with
surface ∂V , then the electric flux
ZZ ZZZ
E · n̂ dS = 4πQ where Q= ρ dV
∂V V
∇ · E(r) = 4πρ(r)
That is, compute the flux by evaluating the surface integral di-
rectly.
b Calculate the same flux using the divergence theorem.
38. ∗. Consider the cube of side length 1 that lies entirely in the first
octant (x ≥ 0, y ≥ 0, z ≥ 0) with one corner at the origin and
another corner at point (1, 1, 1). As such, one face lies in the plane
x = 0, one lies in the plane y = 0, and another lies in the plane z = 0.
The other three faces lie in the planes x = 1, y = 1, and z = 1. Denote
S as the open surface that consists of the union of the 5 faces of the
cube that do not lie in the plane z = 0. The surface S is oriented in
such a way that the unit normal vectors point outwards (that is, the
orientation of S is such that the unit normal vectors on the top face
point towards positive z-directions). Determine the value of
ZZ
I= F · n̂ dS
S
39. ∗.
a Find an upward pointing unit normal vector to the surface z =
xy at the point (1, 1, 1).
b Now consider the part of the surface z = xy, which lies within
the cylinder x2 + y 2 = 9 and call it S. Compute the upward flux
of F = (y, x, 3) through S.
c Find the flux of F = (y, x, 3) through the cylindrical surface
x2 + y 2 = 9 in between z = xy and z = 10. The orientation is
outward, away from the z-axis.
40. ∗.
a Find the divergence of the vector field F = (x + sin y, z + y, z 2 ).
b Find the flux of F through the upper hemisphere x2 + y 2 + z 2 =
25, z ≥ 0, oriented in the positive z-direction.
RR
c Specify an oriented closed surface S, such that the flux S F ·
n̂ dS is equal to −9.
41. ∗. Evaluate the surface integrals. (Use any method you like.)
a S z 2 dS, if S is the part of the cone x2 + y 2 = 4z 2 where
RR
0 ≤ x ≤ y and 0 ≤ z ≤ 1.
RR
b S F · n̂ dS, if F = z k̂ and S is the rectangle with vertices
(0, 2, 0), (0, 0, 4), (5, 2, 0), (5, 0, 4), oriented so that the normal
vector points upward.
c S F · n̂ dS, where F = (y − z 2 )ı̂ı + (z − x2 )̂ + z 2 k̂ and S is the
RR
early part of his life working in his father’s bakery and grain mill. He was finally admitted
as an undergraduate to Cambridge in 1832, aged nearly forty.
CHAPTER 4. INTEGRAL THEOREMS 236
Then I ZZ
∂F2 ∂F1
F1 (x, y) dx + F2 (x, y) dy = − dxdy
C R ∂x ∂y
Warning 4.3.3 Note that in Theorem 4.3.2 we are assuming that F1 and
F2 have continuous first partial derivatives at every point of R. If that is
not the case, for example because F1 or F2 is not defined on all of R, then
y
the conclusion of Green’s theorem can fail. An example is F1 = − x2 +y 2,
x
2 2
F2 = x2 +y2 , R = (x, y) x + y ≤ 1 . See Examples 4.3.7 and 4.3.8.
Here are three notational remarks before we start the proof.
• One way to remember the integrand on the right hand side is to write it
∇ × F) · k̂.
as (∇
• Many people use M instead of F1 and N instead of F . Then Green’s
RR 2∂N
theorem becomes C M (x, y) dx + N (x, y) dy = R ∂x − ∂M
H
∂y dxdy
H R
• The symbol C is just an alternate notation for C that is sometimes
used when C is a closed curve. See Notation 2.4.1.
Proof. We prove the result by reformulating it as a divergence theorem state-
ment. To that end, we define
V = (x, y, z) (x, y) ∈ R, 0 ≤ z ≤ 1
G(x, y, z) = F2 (x, y)ı̂ı − F1 (x, y) ̂
V
y
R
x C
The definition of G does not contain a typo — the x-component of G really
is F2 and the y-component of G really is −F1 . (More or less the reverse of
what you would normally write down.)
These definitions have been rigged so that the divergence theorem applied
to G and V , namely
ZZ ZZZ
G · n̂ dS = ∇ · G dV
∂V V
ZZZ ZZ Z 1
∇ · G dV = dxdydz ∇ · G
V R 0
ZZ Z 1
∂F2 ∂F1
= dxdy dz (x, y) − (x, y)
R 0 ∂x ∂y
ZZ
∂F2 ∂F1
= dxdy (x, y) − (x, y)
R ∂x ∂y
CHAPTER 4. INTEGRAL THEOREMS 237
because the integrand is independent of z. This is exactly the right hand side
of Green’s theorem.
Now for the left hand side. The boundary, ∂V , of V is the union of the (flat)
bottom, the (flat) top and the (curved) side. The outward unit normal on the
(horizontal, flat) top is +k̂ and the outward unit normal on the (horizontal,
flat) bottom is −k̂ so that
ZZ ZZ ZZ ZZ
G · n̂ dS = G · k̂ dS + G · (−k̂) dS + G · n̂ dS
∂V top bottom side
ZZ
= G · n̂ dS
side
We have used the fact that the k̂ component of G is exactly zero to discard
the integrals over the top and bottom of ∂V . To evaluate the integral over the
side, we’ll parametrize the side. Suppose that r(t) = x(t)ı̂ı + y(t) ̂, a ≤ t ≤ b, is
a parametrization of C, with the arrow in the figure above giving the direction
of increasing t. Then we can use
∂R ∂R
n̂ dS = (t, z) × (t, z) dtdz
∂t ∂z
= x0 (t)ı̂ı + y 0 (t) ̂ × k̂ dtdz
the sketch
z
z“1
k̂ V
y
R
x C
BR
n̂
Bt
“ r1
We can now compute the surface integral directly.
ZZ ZZ
G · n̂ dS = G · n̂ dS
∂V side
Z b Z 1
dz G R(t, z) · − x0 (t) ̂ + y 0 (t)ı̂ı
= dt
a 0
CHAPTER 4. INTEGRAL THEOREMS 238
Z b Z 1
dz F2 (x(t), y(t))ı̂ı − F1 (x(t), y(t)) ̂ · − x0 (t) ̂ + y 0 (t)ı̂ı
= dt
a 0
Z b
dt F2 (x(t), y(t)) y 0 (t) + F1 (x(t), y(t)) x0 (t)
=
a
since the integrand is independent of z
I
= F1 (x, y) dx + F2 (x, y) dy
C
R C
p2, 0q x
p1, 0q
Solution. Let R = (x, y) 1 ≤ x ≤ 2, 0 ≤ y ≤ 1 . By Green’s theorem
I ZZ h
∂ 3 ∂ i
(x − xy) dx + (y 3 + 1) dy =
(y + 1) − (x − xy) dxdy
C R ∂x ∂y
Z 2 Z 1 2 2
x 3
= dx dy x = =2
1 0 2 1
Here is a simple corollary of Green’s theorem that tells how to compute the
area enclosed by a curve in the xy-plane.
Corollary 4.3.5 Let
• R be a finite region in the xy-plane whose boundary
• C consists of a finite number of piecewise smooth, simple closed curves.
• Orient C (i.e. put arrows on C) so that if you walk along C in the
direction of the arrows, then R is on your left.
Then I I
I
1
Area(R) =
xdy = − xdy − ydx
ydx =
C C 2 C
Proof. This is just Green’s theorem applied first with F = x ̂, then with
F = −y ı̂ı and finally with F = 21 − y ı̂ı + x ̂ . For all three of these F’s,
∂F2 ∂F1
− =1
∂x ∂y
so that Green’s theorem gives
I ZZ ZZ
∂F2 ∂F1
F1 (x, y) dx + F2 (x, y) dy = − dxdy = dxdy
C R ∂x ∂y R
= Area(R)
CHAPTER 4. INTEGRAL THEOREMS 239
Example 4.3.6 In this example we will use Green’s theorem to compute the
area enclosed by the astroid x2/3 + y 2/3 = a2/3 .
y
1 2π
I Z
1
x(t)y 0 (t) − y(t)x0 (t) dt
Area = xdy − ydx =
2 C 2 0
3a2 2π
Z
cos3 t sin2 t cos t + sin3 t cos2 t sin t dt
=
2 0
3a2 2π
Z
cos2 t sin2 t cos2 t + sin2 t dt
=
2 0
3a2 2π
Z
= cos2 t sin2 t dt
2 0
3a2 2π 2 3a2 2π
Z Z
= sin (2t) dt = [1 − cos(4t)] dt
8 0 16 0
3
= a2 π
8
Example 4.3.7 Trick Question. Evaluate
I
B · dr
C
where
−y ı̂ı + x ̂
B=
x2 + y 2
and C is the curve
x(t) = sin(cos t)
y(t) = sin(sin t)
z(t) = 0
with 0 ≤ t ≤ 2π.
Solution. First let’s think about the curve C. If the curve were just X(t) =
cos t, Y (t) = sin t, Z(t) = 0, it would be the unit circle centred on the origin
in the xy-plane, traversed counterclockwise. For − π2 ≤ u ≤ π2 , the function
sin u increases monotonically with u and is of the same sign as u so that, since
| sin t|, | cos t| ≤ 1 < π2 ,
• x(t) = sin cos t) has the same sign as X(t) = cos t and is increasing at
precisely the same t’s as is X(t)
CHAPTER 4. INTEGRAL THEOREMS 240
• y(t) = sin sin t) has the same sign as Y (t) = sin t and is increasing at
precisely the same t’s as is Y (t)
So the extra sine in our parametrization of C just distorts the circle, straight-
ening the sides a little as depicted here.
y
1 2x2 1 2y 2
= 2 − 2 + − 2
x + y2 (x2 + y 2 ) x2 + y 2 (x2 + y 2 )
(x2 + y 2 ) − 2x2 + (x2 + y 2 ) − 2y 2
= 2
(x2 + y 2 )
=0
Since
our integral is
I I
B · dr = B1 dx + B2 dy
C C
Z2π
B1 x(t), y(t) x0 (t) + B2 x(t), y(t) y 0 (t) dt
=
0
Z 2π
sin(sin t) cos(cos t) sin t + sin(cos t) cos(sin t) cos t
= dt
0 sin2 (cos t) + sin2 (sin t)
CHAPTER 4. INTEGRAL THEOREMS 241
This is a very ugly looking integral2 . But even if we can’t evaluate Hthe integral,
we can see that the integrand is strictly positive, and that forces C B · r > 0.
Because
π
0 ≤ | sin t|, | cos t| ≤ 1 <
2
• cos(cos t) > 0, and sin(sin t) has the same sign as sin t, and sin(sin t) is
zero if and only if sin t = 0. So the first term in the numerator,
is strictly positive.
Since the integrand is strictly positive, the integral is strictly positive.
Why we got the wrong answer:
In our initial and wrong calculation above, we assumed that ∂B ∂x (x, y) −
2
∂B1
∂y (x, y) = 0 at all points (x, y) of the region R inside C. That’s not true.
While it is true for most points, it is not true for all points. The vector field
B(x, y) is not defined at (x, y) = (0, 0). So ∂B ∂B1
∂x (x, y) − ∂y (x, y) is also not
2
defined at (x, y) = (0, 0). That’s enough to invalidate Green’s theorem. Read
the statement of Theorem 4.3.2 again carefully.
Example 4.3.8 Example 4.3.7, again.. Evaluate
I
B · dr
C
where
−y ı̂ı + x ̂
B=
x2 + y 2
and C is the curve
x(t) = sin(cos t)
y(t) = sin(sin t)
z(t) = 0
with 0 ≤ t ≤ 2π.
Solution. This is the same integral that weHcomputed incorrectly in Example
4.3.7. We’ll use two ingredients to compute C B · dr correctly.
• Let a > 0 and denote by Ca the clockwise oriented circle in the xy-
plane that is of radius a and is centered on the origin. We can explicitly
2 Indeed!
CHAPTER 4. INTEGRAL THEOREMS 242
H
compute Ca B · dr. To do so just parametrize Ca by x(t) = a cos t,
y(t) = a sin t, z(t) = 0. Then x0 (t) = −a sin t, y 0 (t) = a cos t and
I Z 2π h −a sin tı̂ı + a cos t ̂ i
B · dr = 2 2 2 2 · − a sin tı̂ı + a cos t ̂ dt
Ca 0 a cos t + a sin t
Z 2π
= dt = 2π
0
• Pick an a that is small enough that Ca lies entirely inside C and apply
Green’s theorem with the region, Ra , that is between C and Ca .
y
C
Ra
´Ca
and so
I I
B · dr = B · dr = 2π
C Ca
4.3.1 Exercises
Exercises — Stage 1
1. Let R be the square
R= (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1
C
x
x dy − y dx
I
1
3. Integrate counterclockwise around
2π C x2 + y 2
a the circle x2 + y 2 = a2
b the boundary of the square with vertices (−1, −1), (−1, 1), (1, 1)
and (1, −1)
for all (x, y) 6= (0, 0). Discuss the connection between this result and
the results of Q[4.3.1.3].
Exercises — Stage 2
5. Evaluate C F · dr where F = x2 y 2 ı̂ı + 2xy ̂ and C is the boundary of
R
the square in the xy-plane having one vertex at the origin and diago-
nally opposite vertex at the point (3, 3), oriented counterclockwise.
I
6. Evaluate (x sin y 2 −y 2 ) dx+(x2 y cos y 2 +3x) dy where C is the coun-
C
terclockwise boundary of the trapezoid with vertices (0, −2), (1, −1), (1, 1)
and (0, 2).
I
1 2 3
x y − x4 y dx + xy 4 + x3 y 2 dy counterclock-
7. ∗. Evaluate I =
C 3 √
wise around the boundary of the half-disk 0 ≤ y ≤ 4 − x2 .
8. ∗. Let C be the counterclockwise boundary of the rectangle with
vertices (1, 0), (3, 0), (3, 1) and (1, 1). Evaluate
I
2 2
3y 2 + 2xey dx + 2yx2 ey dy
C
9. ∗. Consider the closed region enclosed by the curves y = x2 + 4x + 4
and y = 4 − x2 . Let C be its boundary and suppose that C is oriented
CHAPTER 4. INTEGRAL THEOREMS 244
counter-clockwise.
a Draw the oriented curve C carefully in the xy-plane.
Let C be the boundary of the triangle with vertices (0, 0), (1, 0) and
(1, 2), oriented counter-clockwise. Compute
Z
F · dr
C
11. ∗. Suppose the curve C is the boundary of the region enclosed between
the curves y = x2 − 4x + 3 and y = 3 − x2 + 2x. Determine the value
of the line integral
Z p
2xey + 2 + x2 dx + x2 (2 + ey ) dy
C
R
Find C F · dr, where C is the boundary of the triangle (0, 0), (1, −2),
(1, 2), oriented anticlockwise.
13. ∗.
a Use Green’s theorem to evaluate the line integral
−y
Z
x
2 + y2
dx + 2 dy
C x x + y2
Exercises — Stage 3
18. ∗. Evaluate the line integral
Z
(x2 + yex ) dx + (x cos y + ex ) dy
C
Use this to calculate the area inside the curve x2/3 + y 2/3 = 1.
20. ∗. Let F(x, y) = (x+3y)ı̂ı +(x+y) ̂ and G(x, y) = (x+y)ı̂ı +(2x−3y) ̂
be vector fields. Find a number A such that for each circle C in the
plane I I
F · dr = A G · dr
C C
y3 xy 2
21. ∗. Let F(x, y) = − (x, y) 6= (0, 0).
(x2 +y 2 )2
ı̂ı (x2 +y 2 )2
̂,
H
a Compute C F · dr where C is the unit circle in the xy-plane,
positively oriented.
H
b Use (a) and Green’s theorem to find C0 F · dr where C0 is the
x2 y2
ellipse 16 + 25 = 1, positively oriented.
22. ∗. Let C1 be the circle (x − 2)2 + y 2 = 1 and let C2 be the circle
y x
(x − 2)2 + y 2 = 9. Let F = − x2 +y ı + x2 +y
2 ı̂ . Find the integrals
2 ̂
H H
C1
F · dr and C2
F · dr.
23. ∗. Let R be the region in the first quadrant of the xy-plane bounded
by the coordinate axes and the curve y = 1 − x2 . Let C be the
boundary of R, oriented counterclockwise.
R
a Evaluate C x ds.
R
b Evaluate C
F · dr, where F(x, y) = sin(x2 ) − xy ı̂ı + x2 +
cos(y 2 ) ̂.
24. ∗. Let C be the curve defined by the intersection of the surfaces
z = x + y and z = x2 + y 2 .
a Show that C is a simple closed curve.
H
b Evaluate C F · dr where
addition to Stokes’ theorem, he is known for the Navier-Stokes equations of fluid dynamics
and for his work on the wave theory of light. He gave evidence to the Royal Commission on
the Use of Iron in Railway Structures after the Dee bridge disaster of 1847.
CHAPTER 4. INTEGRAL THEOREMS 247
over a finite surface in R3 with an integral over the curve bounding the surface.
Theorem 4.4.1 Stokes’ Theorem. Let
• S be a piecewise smooth oriented surface (i.e. a unit normal n̂ has been
chosen at each point of S and this choice depends continuously on the
point)
• the boundary, ∂S, of the surface S consist of a finite number of piecewise
smooth, simple curves that are oriented consistently with n̂ in the sense
that
z
n̂
S
BS
x
• F be a vector field that has continuous first partial derivatives at every
point of S.
Then I ZZ
F · dr = ∇ × F · n̂ dS
∂S S
Note that
• in Stokes’ theorem, S must be an oriented surface. In particular, S may
not be a Möbius strip. (See Example 3.5.3.)
• If S is part of the xy-plane, then Stokes’ theorem reduces to Green’s the-
orem. Our proof of Stokes’ theorem will consist of rewriting the integrals
so as to allow an application of Green’s theorem.
• If ∂S is a simple closed curve and
◦ when you look at ∂S from high on the z-axis, it is oriented coun-
terclockwise (look at the figure in Theorem 4.4.1), then
◦ n̂ is upward pointing, i.e. has positive z-component, at least near
∂S.
Proof. Write F = F1 ı̂ı + F2 ̂ + F3 k̂. Both integrals involve F1 terms and F2
terms and F3 terms. We shall show that the F1 terms in the two integrals
agree. In other words, we shall assume that F = F1ı̂ı. The proofs that the
F2 and F3 terms also agree are similar. For simplicity, we’ll assumea that the
boundary of S consists of just a single curve, and that we can
• pick a parametrization of S with
∂r
and with r(u, v) orientation preserving in the sense that n̂ dS = + ∂u ×
CHAPTER 4. INTEGRAL THEOREMS 248
∂r
∂v du dv. Also
• pick a parametrization of the curve, ∂R, bounding R as u(t), v(t) , a ≤
t ≤ b, in such a way that when you walk along ∂R in the direction of
increasing t, then R is on your left.
Then the curve ∂S bounding S can be parametrized as R(t) = r u(t), v(t) ,
a ≤ t ≤ b.
z n̂
v
rpu, vq
S R
BS pu, vq
y BR
u
x
The orientation of R(t):
We’ll now verify that the direction of increasing t for the parametrization
R(t) of ∂S is the direction of the arrow on ∂S in the figure on the left above.
By continuity, it suffices to check the orientation at a single point.
Find a point (u0 , v0 ) on ∂R where the forward pointing tangent vector is
a positive multiple of ı̂ı. The horizontal arrow on ∂R in the figure on the left
below is at such a point. Suppose that t = t0 at this point — in other words,
suppose that (u0 , v0 ) = u(t0 ), v(t0 ) . Because the forward pointing tangent
vector to ∂R at (u0 , v0 ), namely u0 (t0 ), v 0 (t0 ) , is a positive multiple of ı̂ı, we
have u0 (t0 ) > 0 and v 0 (t0 ) = 0. The tangent vector to ∂S at R(t0 ) = r u0 , v0 ,
x
Now imagine that you are walking along ∂S in the direction of increasing t.
At time t0 you are at R(t0 ). You point your right arm straight ahead of you. So
CHAPTER 4. INTEGRAL THEOREMS 249
∂r
it is pointing in the direction ∂u (u0 , v0 ). You point your left arm out sideways
∂r
into the interior of S. It is pointing in the direction ∂v (u0 , v0 ). If the direction
of increasing t is the same as the forward direction of the orientation of ∂S, then
∂r ∂r
the vector from our feet to our head, which is ∂u (u0 , v0 ) × ∂v (u0 , v0 ), should
∂r ∂r
be pointing in the same direction as n̂. And since n̂ dS = + ∂u × ∂v du dv, it
is.
Now, with our parametrization and orientation sorted out, we can examine
the integrals.
The surface integral:
Since F = F1 ı̂ı, so that
ı̂ı ̂ k̂
∂ ∂ ∂ ∂F1 ∂F1
∇ × F = det ∂x ∂y ∂z = 0, ,−
∂z ∂y
F1 0 0
and
ı̂ı ̂ k̂
∂r ∂r ∂x ∂y ∂z
n̂ dS = × du dv = det ∂u ∂u ∂u
∂u ∂v ∂x ∂y ∂z
∂v ∂v ∂v
∂y ∂z ∂z ∂y ∂z ∂x ∂x ∂z
= − ı̂ı + − ̂
∂u ∂v ∂u ∂v ∂u ∂v ∂u ∂v
∂x ∂y ∂y ∂x
+ − k̂
∂u ∂v ∂u ∂v
and
ZZ ZZ
∂F1 ∂F1 ∂r ∂r
∇ × F · n̂ dS = 0, ,− · × du dv
S R ∂z ∂y ∂u ∂v
ZZ
∂F1 ∂z ∂x ∂x ∂z ∂F1 ∂x ∂y ∂y ∂x
= − − − du dv
R ∂z ∂u ∂v ∂u ∂v ∂y ∂u ∂v ∂u ∂v
Now we examine the line integral and show that it equals this one.
The line integral:
I Z b d
F · dr = F r u(t), v(t) · r u(t), v(t) dt
∂S a dt
Z b
h ∂r du ∂r dv i
= F r u(t), v(t) · u(t), v(t) (t) + u(t), v(t) (t) dt
a ∂u dt ∂v dt
We can write this as the line integral
I
M (u, v) du + N (u, v) dv
∂R
Z b du
h dv i
= M u(t), v(t) (t) + N u(t), v(t)) (t) dt
a dt dt
around ∂R, if we choose
∂r ∂x
M (u, v) = F r(u, v) · (u, v) = F1 x(u, v), y(u, v), z(u, v) (u, v)
∂u ∂u
∂r ∂x
N (u, v) = F r(u, v) · (u, v) = F1 x(u, v), y(u, v), z(u, v) (u, v)
∂v ∂v
Finally, we show that the surface integral equals the line integral:
CHAPTER 4. INTEGRAL THEOREMS 250
C = (x, y, z) x2 + y 2 = 1, z = 0
S1 = (x, y, z) x2 + y 2 ≤ 1, z = 0
S2 = (x, y, z) z ≥ 0, x2 + y 2 + z 2 = 1
n̂
S2
n̂
S1
C
RR RR
It should not be a surprise that S1 ∇ × F · n̂ dS = S2 ∇ × F · n̂ dS,
for the following reason. Let
V = (x, y, z) x2 + y 2 + z 2 ≤ 1, z ≥ 0
CHAPTER 4. INTEGRAL THEOREMS 251
N̂
S2
V
S1
N̂
But beware that the outward pointing normal to ∂V (call it N̂) is +n̂
on S2 and −n̂ on S1 . So the divergence theorem gives
ZZ ZZ
∇ × F · n̂ dS − ∇ × F · n̂ dS
S2 S1
ZZ ZZ
= ∇ × F · N̂ dS + ∇ × F · N̂ dS
S2 S1
ZZ
= ∇ × F · N̂ dS
Z Z∂V
Z
= ∇ · ∇ × F dV
V
by the divergence theorem
=0
S = (x, y, z) x2 + y 2 ≤ 4, z = 1
z
n̂
S
C
y
x
CHAPTER 4. INTEGRAL THEOREMS 252
Since
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det
∂x ∂y ∂z
2z + sin x146 −5z −5y
" #
∂ ∂ ∂ ∂
= ı̂ı det ∂y ∂z
− ̂ det ∂x ∂z
−5z −5y 2z + sin x146 −5y
" #
∂ ∂
+ k̂ det ∂x ∂y
146
2z + sin x −5z
= 2̂
Now we’ll repeat the last example with a harder curve.
Example 4.4.3 Evaluate C F · dr where F = 2z + sin x146 ı̂ı − 5z ̂ − 5y k̂
H
n̂
C y S C y
x x
146
The x in F will probably make a direct evaluation of the integral difficult.
So we’ll use Stokes’ theorem. To do so we need a surface S with ∂S = C. The
simplest is the flat disk
S = (x, y, z) x2 + y 2 + z 2 ≤ 4, z = y
The first octant of S is shown in the figure on the right above. We saw in the
last Example 4.4.2 that
∇ × F = 2̂
So Stokes’ theorem gives
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = 2 ̂ · n̂ dS
C S S
RR
We’ll evaluate the integral 2 S ̂ · n̂ dS in two ways. The first way is more
efficient, but also requires more insight. Since ∇ (z −y) = k̂−̂, the upward unit
CHAPTER 4. INTEGRAL THEOREMS 253
To get the upward pointing normal pointing normal, we take the + sign so
that n̂ dS = (−̂ + k̂) dxdy. As (x, y, z) runs over
S = (x, y, z) x2 + y 2 + z 2 ≤ 4, z = y
= (x, y, z) x2 + 2y 2 ≤ 4, z = y
2 2
= (x, y, z) x4 + y2 ≤ 1, z = y
2 2
(x, y) runs over the elliptical disk R = (x, y) x4 + y2 ≤ 1 . The part of
this ellipse in the first octant is the shaded region in the figure below.
z
n̂
y
x
√ √
This ellipse has semiaxes a = 2 and b = 2 and hence area πab = 2 2π.
So
I ZZ ZZ ZZ
F · dr = 2 ̂ · n̂ dS = 2 ̂ · (−̂ + k̂) dxdy = −2 dxdy
C S R R
= −2Area(R)
√
= −4 2π
H 2
Example 4.4.4 Evaluate C F·dr where F = (x+y)ı̂ı +2(x−z) ̂ +(y +z) k̂ and
C is the oriented curve obtained by going from (2, 0, 0) to (0, 3, 0) to (0, 0, 6)
and back to (2, 0, 0) along straight line segments.
CHAPTER 4. INTEGRAL THEOREMS 254
C2
C3
y
C1
x
Solution 1. In this first solution, we’ll evaluate the integral directly. The
first line segment (C1 in the figure above) may be parametrized as
r(t) = (2, 0, 0) + t (0, 3, 0) − (2, 0, 0) = 2 − 2t , 3t , 0 0≤t≤1
So the integral along this segment is
Z 1 Z 1
dr
F(r(t)) · dt = (2 + t , 2(2 − 2t) , (3t)2 ) · (−2 , 3 , 0) dt
0 dt 0
Z 1
= (8 − 14t) dt
0
h i1
= 8t − 7t2 =1
0
The second line segment (C2 in the figure above) may be parametrized as
r(t) = (0, 3, 0) + t (0, 0, 6) − (0, 3, 0) = 0 , 3 − 3t , 6t 0 ≤ t ≤ 1.
So the integral along this segment is
Z 1 Z 1
dr
3(1 − t) , −12t , 9(1 − t)2 + 6t · (0, −3, 6) dt
F(r(t)) · dt =
0 dt 0
Z 1
= [36t + 54(1 − t)2 + 36t] dt
0
h i1
= 18t2 − 18(1 − t)3 + 18t2
0
= 54
The final line segment (C3 in the figure above) may be parametrized as
r(t) = (0, 0, 6) + t (2, 0, 0) − (0, 0, 6) = (2t , 0 , 6 − 6t) 0≤t≤1
So the line integral along this segment is
Z 1 Z 1
dr
F(r(t)) · dt = 2t , 4t − 12(1 − t) , 6(1 − t) · (2, 0, −6) dt
0 dt 0
Z 1 h i1
= [4t − 36(1 − t)] dt = 2t2 + 18(1 − t)2 = −16
0 0
z “ 6 ´ 3x ´ 2y
S
y
R
3x ` 2y “ 6
x z“0
Using horizontal strips as in the figure on the left below,
I ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS
= [2(y + 1)ı̂ı + k̂] · [3ı̂ı + 2̂ + k̂] dx dy
Z ZR
= [6y + 7] dx dy
R
1
Z 3 Z 3 (6−2y)
= dy dx [6y + 7]
0 0
Z 3
1
= dy [6y + 7][6 − 2y]
0 3
Z 3
1
= dy [−12y 2 + 22y + 42]
3 0
CHAPTER 4. INTEGRAL THEOREMS 256
1h i3
= − 4y 3 + 11y 2 + 42y
3 0
= − 4 × 9 + 11 × 3 + 42 = 39
y y
3 3
x “ 13 p6 ´ 2yq y “ 32 p2 ´ xq
2 x 2 x
Alternatively, using vertical strips as in the figure on the right above,
I ZZ
F · dr = [6y + 7] dx dy
C R
3
Z 2 Z 2 (2−x)
= dx dy [6y + 7]
0 0
Z 2
h 32 3 i
= dx 3 2 (2 − x)2 + 7 (2 − x)
0 2 2
h 27 1 21 1 i2
= − (2 − x)3 − (2 − x)2
4 3 2 2 0
9 21
= 8+ 4 = 39
4 4
H
Example 4.4.5 Evaluate C F·dr where F = (cos x+y+z)ı̂ı +(x+z) ̂ +(x+y) k̂
and C is the intersection of the surfaces
y2 z2
x2 + + =1 and z = x2 + 2y 2
2 3
oriented counterclockwise when viewed from above.
2 2
Solution. First, let’s sketch the curve. x2 + y2 + z3 = 1 is an ellipsoid centred
on the origin and z = x2 + 2y 2 is an upward opening paraboloid that passes
through the origin. They are sketched in the figure below. The paraboloid is
red.
z “ x2 ` 2y 2
y2 z2
x2 ` 2
` 3
“1
Their intersection, the curve C, is the blue curve in the figure. It looks like
a deformed2 circle.
One could imagine parametrizing C. For example, substituting x2 = z−2y 2
CHAPTER 4. INTEGRAL THEOREMS 257
2
into the equation of the ellipsoid gives − 32 y 2 + 13 (z+ 32 = 74 . This can be solved
to give y as a function of z and then x2 = z − 2y 2 also gives x as a function of
z. However this would clearly yield, at best, a really messy integral. So let’s
try Stokes’ theorem.
In fact, since
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det
∂x ∂y ∂z
cos x + y + z x + z x + y
= ı̂ı 1 − 1 − ̂ 1 − 1 + k̂ 1 − 1
=0
This FHis conservative! (In fact F = ∇ sin x + xy + xz + yz .) As C is a closed
curve, C F · dr = 0.
RR
Example 4.4.6 Evaluate S G·n̂ dS where G = (2x)ı̂ı +(2z−2x) ̂ +(2x−2z) k̂
and
S = (x, y, z) z = 1 − x2 − y 2 (1 − y 3 ) cos x ey , x2 + y 2 ≤ 1
x2 ` y 2 “ 1, z “ 0
only simple thing about it is that its boundary, ∂S, is the circle x2 +y 2 = 1,
z = 0. It is clear that we should not try to evaluate the integral directly3 . In
this solution we will combine the divergence theorem with the observation that
∂ ∂ ∂
∇·G= (2x) + (2z − 2x) + (2x − 2z) = 0
∂x ∂y ∂z
to avoid ever having work with the surface S. Here is an outline of what we
will do.
• We first select a simple surface S 0 whose boundary ∂S 0 is also the circle
x2 + y 2 = 1, z = 0. A nice simple choice of S 0 , and the surface that we
will use, is the disk
S 0 = (x, y, z) x2 + y 2 = 1, z = 0
• Then we define V to be the solid whose top surface is S and whose bottom
surface is S 0 . So the boundary of V is the union of S and S 0 .
3 By Salvador Dali?
CHAPTER 4. INTEGRAL THEOREMS 258
n̂ V
n̂
S1
• For S 0 , we will use the upward pointing normal n̂ = k̂, which is minus
the outward pointing normal to ∂V on S 0 . So the divergence theorem
says that
ZZZ ZZ ZZ
∇ · G dV = G · n̂ dS − G · n̂ dS
V S S0
=0
Of course this integral can be evaluated by using that one antiderivative of the
integrand cos3 t = 1 − sin2 t cos t is sin t − 31 sin3 t and that this antiderivative
is zero at t = 0 and at t = 2π. But it is easier to observe that the integral of
any odd power of sin t or cos t over any full period is zero. Look, for example,
at the graphs of sin3 x and cos3 x, below.
y y
y “ sin3 x y “ cos3 x
1 1
x x
π{2 π 3π{2 2π π{2 π 3π{2 2π
´1 ´1
Either way
ZZ
G · n̂ dS = 0
S
H
Example 4.4.7 In this example we compute, in three different ways, C
F · dr
where
F = (z − y)ı̂ı − (x + z) ̂ − (x + y) k̂
and C is the curve x2 + y 2 + z 2 = 4, z = y oriented counterclockwise when
viewed from above.
where
• c = (0, 0, 0) is the centre of C,
• ρ = 2 is the radius of C and
• ı̂ı0 and ̂0 are two vectors that
a are unit vectors,
b are parallel to the plane z = y and
4 That way lies pain.
CHAPTER 4. INTEGRAL THEOREMS 260
k̂1
̂1
y
ı̂ı1
p2, 0, 0q
x
The trickiest part is finding suitable vectors ı̂ı0 and ̂0 :
• The point (2, 0, 0) satisfies both x2 + y 2 + z 2 = 4 and z = y and so is
on C. We may choose ı̂ı0 to be the unit vector in the direction from the
centre (0, 0, 0) of the circle towards (2, 0, 0). Namely ı̂ı0 = (1, 0, 0).
1 sin t sin t
r(t) = 2 cos t (1, 0, 0) + 2 sin t √ (0, 1, 1) = 2 cos t, √ , √
2 2 2
0 ≤ t ≤ 2π
√
To check
√ that this parametrization is correct, note that x = 2 cos t, y = 2 sin t,
z = 2 sin t satisfies both x2 + y 2 + z 2 = 4 and z = √ y.
At t = 0, r(0) = (2, 0, 0). As t increases,
√ √ z(t) = 2 sin t increases and r(t)
moves upwards towards r π2 = (0, 2, 2). This is the desired counterclock-
wise direction (when viewed from above). Now that we have a parametrization,
we can set up the integral.
√ √
r(t) = 2 cos t, 2 sin t, 2 sin t
√ √
r 0 (t) = − 2 sin t, 2 cos t, 2 cos t
F r(t) = z(t) − y(t), −x(t) − z(t), −x(t) − y(t)
√ √ √ √
= 2 sin t − 2 sin t, −2 cos t − 2 sin t, −2 cos t − 2 sin t
√ √
= − 0, 2 cos t + 2 sin t, 2 cos t + 2 sin t
√
F r(t) · r 0 (t) = − 4 2 cos2 t + 4 cos t sin t
√ √
= − 2 2 cos(2t) + 2 2 + 2 sin(2t)
by the double angle formulae sin(2t) = 2 sin t cos t and cos(2t) = 2 cos2 t − 1.
So
I Z 2π
F r(t) · r 0 (t) dt
F · dr =
C 0
CHAPTER 4. INTEGRAL THEOREMS 261
2π √ √
Z
= − 2 2 cos(2t) + 2 2 + 2 sin(2t) dt
0
h√ √ i2π
= − 2 sin(2t) + 2 2t − cos(2t)
0
√
= −4 2π
C = (x, y, z) x2 + y 2 + z 2 = 4, z = y
is a closed curve, this is possible. In fact there are many possible choices of S
with ∂S = C. Three possible S’s (sketched below) are
S = (x, y, z) x2 + y 2 + z 2 ≤ 4, z = y
S 0 = (x, y, z) x2 + y 2 + z 2 = 4, z ≥ y
S 00 = (x, y, z) x2 + y 2 + z 2 = 4, z ≤ y
S1
S
S2
The first of these, which is part of a plane, is likely to lead to simpler
computations than the last two, which are parts of a sphere. So we choose
what looks like the simpler way.
In preparation for application of Stokes’ theorem, we compute ∇ × F and
n̂ dS. For the latter, we apply the formula n̂ dS = ±(−fx , −fy , 1) dxdy (of
Equation 3.3.2) to the surface z = f (x, y) = y. We use the + sign to give the
normal a positive k̂ component.
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
z − y −x − z −x − y
= ı̂ı − 1 − (−1) − ̂ − 1 − 1 + k̂ − 1 − (−1)
= 2 ̂
n̂ dS = (0, −1, 1) dxdy
∇ × F · n̂ dS = (0, 2, 0) · (0, −1, 1) dxdy = −2 dxdy
The integration variables are x and y and, by definition, the domain of inte-
gration is
R = (x, y) (x, y, z) is in S for some z
To determine precisely what this domain of integration is, we observe that since
z = y on S, x2 + y 2 + z 2 ≤ 4 is the same as x2 + 2y 2 ≤ 4 on S,
S = (x, y, z) x2 + 2y 2 ≤ 4, z = y =⇒ R = (x, y) x2 + 2y 2 ≤ 4
CHAPTER 4. INTEGRAL THEOREMS 262
So the domain
√ of integration is√an ellipse with semimajor axis a = 2, semiminor
axis b = 2 and area πab = 2 2π. The integral is then
I ZZ ZZ √
F · dr = ∇ × F · n̂ dS = (−2) dxdy = −2 Area (R) = −4 2π
C S R
y
x2 ` 2y 2 “ 4
If, instead, we slice up R using thin horizontalp slices, then,pon each such
slice, y is essentially constant and x runs from − 4 − 2y 2 to 4 − 2y 2 . The
√
bottom
√ such slice would have y = − 2 and the top such slice would have
y = 2. So the correct limits with this slicing are
ZZ Z √ 2Z √ 2 4−2y
f (x, y) dxdy = √ dy √ dx f (x, y)
R − 2 − 4−2y 2
y
x2 ` 2y 2 “ 4
x √ dy dx f (x, y)
− 2 −2
to S. So n̂ = √1 (−̂
+ k̂) and
2
I ZZ ZZ
1
F · dr = ∇ × F · n̂ dS = (2̂) · √ (−̂ + k̂) dS
C 2
Z ZS √ √
S
= − 2 dS = − 2 Area (S)
S
H √
As S is a circle of radius 2, C
F · dr = −4 2π, yet again.
Example 4.4.8 In Warning 4.1.17, we stated that if a vector field fails to
pass the screening test ∇ · B = 0 at even a single point, for example because
the vector field is not defined at that point, then B can fail to have a vector
potential. An example is the point source
r̂(x, y, z)
B(x, y, z) =
r(x, y, z)2
This vector field is defined on all of R3 , except for the origin, and its divergence
∂ x ∂ y
∇·B= +
∂x (x2 + y 2 + z 2 )3/2 ∂y (x2 + y 2 + z 2 )3/2
∂ z
+
∂z (x2 + y 2 + z 2 )3/2
3x2
1
= − 2
(x2 + y 2 + z 2 )3/2 (x + y 2 + z 2 )5/2
3y 2
1
+ − 2
(x2 + y 2 + z 2 )3/2 (x + y 2 + z 2 )5/2
3z 2
1
+ − 2
(x2 + y 2 + z 2 )3/2 (x + y 2 + z 2 )5/2
2 2 2
3 3(x + y + z )
= 2 2 2 3/2
− 2
(x + y + z ) (x + y 2 + z 2 )5/2
5 If you are uncomfortable with the surface not having a boundary, poke a very small hole
in the surface, giving it a very small boundary. Then take the limit as the hole tends to zero.
CHAPTER 4. INTEGRAL THEOREMS 264
4.4.1.1 Divergence
Let ε > 0 be a tiny positive number, and then let
Sε (x0 , y0 , z0 ) = (x, y, z) (x − x0 )2 + (y − y0 )2 + (z − z0 )2 = ε2
Think of the vector field v as the velocity of a moving fluid which has density
one. We have already seen, in §3.4, that the flux integral for a velocity field
has the interpretation
ZZ
the volume of fluid leaving Bε (x0 , y0 , z0 )
v · n̂ dS =
Sε (x0 ,y0 ,z0 ) through Sε (x0 , y0 , z0 ) per unit time
4.4.1.2 Curl
Again let ε > 0 be a tiny positive number and let Dε (x0 , y0 , z0 ) be a tiny
flat circular disk of radius ε centred on the point (x0 , y0 , z0 ) and denote by
Cε (x0 , y0 , z0 ) its boundary circle. Let n̂ be a unit normal vector to Dε . It
tells us the orientation of Dε . Give the circle Cε the corresponding orientation
using the right hand rule. That is, if the fingers of your right hand are pointing
in the corresponding direction of motion along Cε and your palm is facing Dε ,
then your thumb is pointing in the direction n̂.
n̂
ε
Cε
Because Dε (x0 , y0 , z0 ) is really small, ∇ × v is essentially constant on
Dε (x0 , y0 , z0 ) and we essentially have
ZZ
∇ × v · n̂ dS = ∇ × v(x0 , y0 , z0 ) · n̂ Area Dε (x0 , y0 , z0 )
Dε (x0 ,y0 ,z0 )
= πε2 ∇ ×v(x0 , y0 , z0 ) · n̂
Again, this is really an approximate statement which gets better and better as
ε → 0. A more precise statement is
RR
Dε (x0 ,y0 ,z0 )
∇ × v · n̂ dS
∇ ×v(x0 , y0 , z0 ) · n̂ = lim
ε→0 πε2
By Stokes’ theorem, we also have
ZZ I
∇ × v · n̂ dS = v · dr
Dε (x0 ,y0 ,z0 ) Cε (x0 ,y0 ,z0 )
Again,
H think of the vector field v as the velocity of a moving fluid. Then
Cε
v · dr is called the circulation of v around Cε .
To measure the circulation experimentally, place a small paddle wheel in
the fluid, with the axle of the paddle wheel pointing along n̂ and each of the
paddles perpendicular to Cε and centred on Cε .
n̂
t̂
ε
Cε
Each paddle moves tangentially to Cε . It would like to move with the same
speed as the tangential speed v· t̂ (where t̂ is the forward pointing unit tangent
vector to Cε at the location of the paddle) of the fluid at its location. But all
paddles are rigidly fixed to the axle of the paddle wheel and so must all move
with the same speed. That common speed will be the average value of v · t̂
around Cε . If ds represents an element of arc length of Cε , the average value
of v · t̂ around Cε is
I I
1 1
vT = v · t̂ ds = v · dr
2πε Cε 2πε Cε
CHAPTER 4. INTEGRAL THEOREMS 266
since dr has direction t̂ and length ds so that dr = t̂ds, and since 2πε is the
circumference of Cε . If the paddle wheel rotates at Ω radians per unit time,
each paddle travels a distance Ωε per unit time (remember that ε is the radius
of Cε ). That is, vT = Ωε. Combining all this information,
RR
Dε (x0 ,y0 ,z0 )
∇ × v · n̂ dS
∇ ×v(x0 , y0 , z0 ) · n̂ = lim
ε→0
H πε2
v · dr
= lim Cε 2
ε→0 πε
2πε vT
= lim
ε→0 πε2
2πε (Ωε)
= lim
ε→0 πε2
= 2Ω
so that
Ω = 21 ∇ ×v(x0 , y0 , z0 ) · n̂
The component of ∇ × v(x0 , y0 , z0 ) in any direction n̂ is twice the rate at
which the paddle wheel turns when it is put into the fluid at (x0 , y0 , z0 ) with
its axle pointing in the direction n̂. The direction of ∇ × v(x0 , y0 , z0 ) is the
axle direction which gives maximum rate of rotation and the magnitude of
∇ × v(x0 , y0 , z0 ) is twice that maximum rate of rotation. For this reason,
∇ × v is called the “vorticity”.
B
I
E · dr = voltage around C
C
ZZ
B · n̂ dS = magnetic flux through S S
S
C
Then the voltage around C is the negative of the rate of change of the
magnetic flux through S. As an equation, Faraday’s Law is
I ZZ
∂
E · dr = − B · n̂ dS
C ∂t S
an extremely influential scientist despite having only the most basic of formal educations.
CHAPTER 4. INTEGRAL THEOREMS 267
4.4.3 Exercises
Exercises — Stage 1
1. Each of the figures below contains a sketch H of a surface
RR S and its
boundary ∂S. Stokes’ theorem says that ∂S F · dr = S ∇ × F · n̂ dS
if n̂ is a correctly oriented unit normal vector to S. Add to each sketch
a typical such normal vector.
(a) (b) (c)
S ∂S
S ∂S
∂S S
2. Let
• R be a finite region in the xy-plane,
• the boundary, C, of R consist of a single piecewise smooth, sim-
ple closed curve
◦ that is oriented (i.e. an arrow is put on C) consistently with
R in the sense that if you walk along C in the direction of
the arrow, then R is on your left
y
C
x
• F1 (x, y) and F2 (x, y) have continuous first partial derivatives at
every point of R.
Use Stokes’ theorem to show that
I ZZ
∂F2 ∂F1
F1 (x, y) dx + F2 (x, y) dy = − dxdy
C R ∂x ∂y
i.e. to show Green’s theorem.
6 For the others, see Example 4.1.2
CHAPTER 4. INTEGRAL THEOREMS 268
H H
3. Verify the identity C φ∇ ∇ψ · dr = − C ψ∇∇φ · dr for any continuously
differentiable scalar fields φ and ψ and curve C that is the boundary
of a piecewise smooth surface.
Exercises — Stage 2
4. Let C be the curve of intersection of the cylinder x2 + y 2 = 1 and the
surface z = y 2 oriented in the counterclockwise direction as seen from
(0, 0, 100). Let F = (x2 − y , y 2 + x , 1). Calculate C F · dr
H
a by direct evaluation
b by using Stokes’ Theorem.
Evaluate C F · dr where F = yex ı̂ı + (x + ex ) ̂ + z 2 k̂ and C is the
H
5.
curve
(x, y, z) ∈ R3 x2 + y 2 + z 2 = 4, z ≥ 0
oriented so the surface normals point away from the centre of the
hemisphere.
2
7. ∗. Let S be the part of the surface z = 16 − (x2 + y 2 ) which lies
above the xy-plane. Let F be the vector field
Calculate ZZ
∇ × F · n̂ dS
S
where n̂ is the upward normal on S.
8. ∗. Let C be the intersection of the paraboloid z = 4 − x2 − y 2 with
the cylinder x2 + (y − 1)2 = 1, oriented counterclockwise when
H viewed
from high on the z-axis. Let F = xz ı̂ı + x ̂ + yz k̂. Find C F · dr.
9. Let F = −yez ı̂ı + x3 cos z ̂ + z sin(xy) k̂, and let S be the part of the
surface z = (1 − x2 )(1 − y 2 ) that lies above the square −1 ≤ x ≤ 1,
−1 ≤ y ≤ 1 in the xy-plane. Find the flux of ∇ × F upward through
S.
2
10. Evaluate the integral C F·dr, in which F = (ex −yz , sin y−yz , xz+
H
a Let S1 be the part of the cone that lies between the planes z = 0
and z = 4. Note that S1 does not include any part of the plane
z = 4. Use Stokes’ theorem to determine the value of
ZZ
∇ × F · n̂ dS
S1
22. ∗. Let
F = x sin y ı̂ı − y sin x ̂ + (x − y)z 2 k̂
Use Stokes’ theorem to evaluate
Z
F · dr
C
x2 + y 2 + 2(z − 1)2 = 6, z ≥ 0.
(E.g., at the high point of the surface, the unit normal is k̂.)
Define
2
+y 2 +z 2
G = ∇ × F, where F = (xz − y 3 cos z)ı̂ı + x3 ez ̂ + xyzex k̂.
RR
Find S G · n̂dS.
26. ∗. Let C be a circle of radius R lying in the plane x + y + z = 3. Use
Stokes’ Theorem to calculate the value of
I
F · dr
C
where F = z 2ı̂ı + x2̂ + y 2 k̂. (You may use either orientation of the
circle.)
27. Let S be the oriented surface consisting of the top and four sides
of the cube whose vertices are (±1, ±1, ±1), oriented outward. If
F(x, y, z) = (xyz, xy 2 , x2 yz), find the flux of ∇ × F through S.
28. Let S denote the part of the spiral ramp (that is helicoidal surface)
parametrized by
x = u cos v, y = u sin v, z = v 0 ≤ u ≤ 1, 0 ≤ v ≤ 2π
Exercises — Stage 3
29. Let C be the intersection of x + 2y − z = 7 and x2 − 2x + 4y 2 = 15.
The curve C is oriented counterclockwise when viewed from high on
the z-axis. Let
2
F = ex + yz ı̂ı + cos(y 2 ) − x2 ̂ + sin(z 2 ) + xy k̂
H
Evaluate C F · dr.
30. ∗.
a Find the curl of the vector field F = 2 + x2 + z , 0 , 3 + x2 z .
b Let C be the curve in R3 from the point (0, 0, 0) to the point
(2, 0, 0), consisting of three consecutive line segments connecting
CHAPTER 4. INTEGRAL THEOREMS 272
the points (0, 0, 0) to (0, 0, 3), (0, 0, 3) to (0, 1, 0), and (0, 1, 0) to
(2, 0, 0). Evaluate the line integral
Z
F · dr
C
a D is connected.
b D is simply connected.
c D is disconnected.
c Compute ∇ × F.
e Is F conservative?
33. ∗. A physicist studies a vector field F(x, y, z). From experiments, it
is known that F is of the form
34. ∗. Let C be the curve in the xy-plane from the point (0, 0) to the point
(5, 5) consisting of the ten line segments consecutively connecting the
points (0, 0), (0, 1), (1, 1), (1, 2), (2, 2), (2, 3), (3, 3), (3, 4), (4, 4), (4, 5),
(5, 5). Evaluate the line integral
Z
F · dr
C
where
and let
F = z ı̂ı + x ̂ + y 3 z 3 k̂
Use Stokes’ theorem to evaluate
I
F · dr
C
38. ∗. Use Stokes’ theorem to evaluate
I
z dx + x dy − y dz
C
v = xı̂ı + y ̂ − 2z k̂
downward through S.
c A vector field F has curl ∇ × F = xı̂ı + y ̂ − 2z k̂. On the xz-
plane, the vector field F is constant with F(x, 0, z) = ̂. Given
CHAPTER 4. INTEGRAL THEOREMS 274
x2 + y 2 = 1, z = 1, y ≥ 0
C1 C2
C3
different meanings in different contexts. We are here using one of the standard definitions.
Another standard definition requires that all derivatives of all orders are continuous.
CHAPTER 4. INTEGRAL THEOREMS 275
` ˘
Bε px0 , 0q
C
O
R2 because for any point (x0 , 0) ∈ X and any ε > 0, the ball Bε (x0 , 0)
contains points with nonzero y-coordinates and so is not contained in X .
d The union of open balls
B1 (0, 0) ∪ B1 (2, 0)
= (x, y) ∈ R2 x2 + y 2 < 1 or (x−2)2 + y 2 < 1
is not connected, since any continuous path from, for example, (2, 0) to
(0, 0) must leave the union. In the figure on the left below, an “empty
disk” has been sketched at (1, 0) just to emphasise that the point (1, 0)
is not in the union.
(x, y) ∈ R2 x2 + y 2 ≤ 1 or (x − 2)2 + y 2 ≤ 1
is connected. For example, the straight line segment from (2, 0) to (0, 0)
remains in the the union.
Many, but not all, of the basic facts that we developed, in §2.4.1, about
conservative fields in Rn also applies (with the same proofs) to fields on D.
CHAPTER 4. INTEGRAL THEOREMS 276
then
F is conservative on D ⇐⇒ ∇ × F = 0 on D
D1
D2 D3
This leaves the question of what happens when (H) is violated. We’ll just
look at one example, which however gives the flavour of the general theory.
The punctured disk is
be the polar angle of (x, y) with, for example, −π < θ < π. This θ is defined
on all of D, except for the negative real axis. The domain of definition, Dπ , is
sketched on the left below.
Dπ D0
If (x0 , y0 ) happens to lie on the negative real axis, just replace −π < θ < π
by a different interval of length 2π, like 0 < θ < 2π. The domain of definition
of θ would then change to the D0 , sketched on the right above.
It’s now a simple matter to check that ∇ θ(x, y) = Θ (x, y) on the domain
of definition of θ. If x 6= 0, then, from the figure below,
y px, yq
?
x2 `y 2
y
θpx, yq
x x
we have that tan θ(x, y) = xy , and cos θ(x, y) = √ x
, so that
x2 +y 2
∂ y h ∂ i y
tan θ(x, y) = − 2 =⇒ θ(x, y) sec2 θ(x, y) = − 2
∂x x ∂x x
∂ y
=⇒ θ(x, y) = − 2 cos2 θ(x, y)
∂x x
y x2 y
=− 2 2 2
=− 2
x x +y x + y2
∂ 1 h∂ i 1
tan θ(x, y) = =⇒ θ(x, y) sec2 θ(x, y) =
∂y x ∂y x
∂ 1
=⇒ θ(x, y) = cos2 θ(x, y)
∂y x
1 x2 x
= 2 2
= 2
x x +y x + y2
If x = 0, then we must have y 6= 0 (since (0, 0) is not in the domain of definition
to θ), and we can use cot θ(x, y) = xy instead and arrive at the same result.
So far we have just looked at one vector field on D. We are now ready to
consider any vector field F on D that passes the screening test ∇ × F = 0 on
D. We claim that there is a function ϕ on D such that
I
1
Θ
F = αF + ϕ ∇ where αF = F · dr (E2)
2π Cε
The significance of this claim is that it says that if a vector field on D passes the
screening test on D, then, either it is conservative (that’s the case if and only
if αF = 0) or, if it fails to be conservative, then it differs from a conservative
field (namely ∇ ϕ) only by a constant (namely αF ) times the fixed vector field
Θ . That is, there is only one nonconservative vector field on D that passes the
screening test, up to multiplication by constants and addition of conservative
fields. This is a nice simple surprise.
Observe that in the definition of αF , we did not specify the radius ε of the
circle Cε to be used for the integration curve. That’s because the answer to the
CHAPTER 4. INTEGRAL THEOREMS 279
0
choice of ε.2 To
integral does not depend on the 0<ε <ε<1
0see this, take any
and consider the surface S = (x, y) ∈ R ε < |(x, y)| < ε .
´Cε1
Cε
D
It is completely contained in D. The boundary of S consists of two parts.
The outside part is Cε , oriented counterclockwise as usual. The inside part is
Cε0 , but oriented clockwise. It is usually denoted −Cε0 . So, by Stokes’ theorem,
I I I I I
F · dr − F · dr = F · dr + F · dr = F · dr
Cε Cε0 Cε −Cε0 ∂S
ZZ
= ∇ × F · n̂ dS = 0
S
Cε
C C
D D
It is named after the French mathematician Camille Jordan (1838–1922), who first proved
it.
CHAPTER 4. INTEGRAL THEOREMS 280
Stokes’ theorem,
I I
G · dr = F − αFΘ · dr
C
Z∂S
Z ZZ
= ∇ × F · n̂ dS − αF ∇ × Θ · n̂ dS = 0 − αF 0
S S
=0
εêp1q
εêp2q
εê p3q x0
1 We’ll also use some more mathematics than before. In this section, we’ll use matrix
eigenvalues and eigenvectors and solve some simple systems of ordinary differential equations.
We’ll also need to use a lot of subscripts and superscripts. It only looks intimidating.
CHAPTER 4. INTEGRAL THEOREMS 281
Here ε > 0 is the length of each edge of the cube and is assumed to be
really small. The vectors ê(1) , ê(2) and ê(3) are three mutually perpendicular
unit vectors that give the orientation of the edges of the cube. The vectors
from the corner x0 to its three nearest neighbour corners are εê(1) , εê(2) and
εê(3) .
As time progresses, the chunk of fluid moves. In particular, the corners
move. Let us denote by εb(1) (t) the vector, at time t, joining the n1 = n2 =
n3 = 0 corner to the n1 = 1, n2 = n3 = 0 corner. Define εb(2) (t) and εb(3) (t)
similarly. For times very close to t0 we can think of our chunk of fluid as being
essentially a parallelepiped with edges εb(k) (t).
εbp1q ptq
εbp3q ptq
εbp2q ptq
By concentrating on the edges εb(k) (t) of the chunk of fluid, rather than
the corners, we are ignoring any translations that the chunk of fluid might have
undergone. We want, instead, to determine how the size and orientation of the
parallelepiped changes as t increases.
At time t0 , b(k) = ê(k) . The velocities of the corners of the chunk of fluid
at time t0 are
v x0 + n1 εê(1) + n2 εê(2) + n3 εê(3) , t0
In particular, at time t0 , the tail of εb(k) has velocity v(x0 , t0 ) and the head
of εb(k) has velocity v(x0 + εê(k) , t0 ). Consequently (using a Taylor approxi-
mation),
db(k) 3 ∂v (k)
(t0 ) = v x0 + εê(k) , t0 − v x0 , t0 = x0 , t0 êj + O(ε2 )
P
ε ε
dt j=1 ∂xj
and so
db(k) P3 ∂v (k)
(t0 ) = x0 , t0 êj + O(ε)
dt j=1 ∂xj
∂vi
Vi,j = x0 , t0 1 ≤ i, j ≤ 3 (M)
∂xj
db(k)
(t0 ) = Vb(k) (t0 ) + O(ε)
dt
Here Vb(k) (t0 ) is the product of the 3 × 3 matrix V and the 3 × 1 column vector
b(k) (t0 ). We study the behaviour of b(k) (t) for small ε and t close to t0 , by
CHAPTER 4. INTEGRAL THEOREMS 282
is
b1 (t) = e2t β1
2t
e 0
or equivalently b(t) = b(0)
b2 (t) = e3t β2 0 e3t
If one chooses ê(1) = ı̂ı and ê(2) = ̂, the edges, b(1) (t) = e2t ê(1) and b(2) (t) =
e3t ê(2) , of the chunk of fluid never change direction. But their lengths do
(k)
change. The relative rate of change of length per unit time, | dbdt (t)|/|b(k) (t)|,
is 2 for b(1) and 3 for b(2) . In the figure below, the darker rectangle is the
initial square. That is, the square with edges b(k) (t0 ) = ê(k) . The lighter
rectangle is that with edges b(k) (t) for some t a bit bigger than t0 .
As time increases the initial cube becomes a larger and larger rectangle.
Example 4.6.2 Example 4.6.1, generalized. The behaviour of Example
4.6.1 is typical of V’s that are symmetric matrices, i.e. that obey2 Vi,j = Vj,i
for all i, j. Any d × d symmetric matrix3 (with real entries)
• has d real eigenvalues
• has d mutually orthogonal real unit eigenvectors.
obeys
db(k)
(t) = λk eλk (t−t0 ) ê(k) = eλk (t−t0 ) Vê(k) = Vb(k) (t) and b(k) (t0 ) = ê(k)
dt
So b(k) (t) = eλk (t−t0 ) ê(k) satisfies (IVP) for all t and 1 ≤ k ≤ d.
If we start, at time t0 , with a cube whose edges, ê(k) , are eigenvectors of V,
then as time progresses the edges, b(k) (t), of the chunk of fluid never change
direction. But their lengths change with the relative rate of change of length
per unit time being λk for edge number k. This rate of change may be positive
(the edge grows with time) or negative (the edge shrinks in time) depending
on the sign of λk .
The volume of the chunk of fluid at time t is V (t) = eλ1 (t−t0 ) · · · eλd (t−t0 ) .
The relative rate of change of volume per unit time is V 0 (t)/V (t) = λ1 · · · + λd ,
the sum of the d eigenvalues.
Pd The sum of the eigenvalues of any d × d matrix
V is given by its trace i=1 Vi,i . For the matrix (M)
d
V 0 (t0 ) X ∂vi
= x0 , t0 = ∇ · v x0 , t0
V (t0 ) i=1
∂xi
So, at least when the matrix V defined in (M) is symmetric, the divergence
∇ · v x0 , t0 gives the relative rate of change of volume per unit time for our
tiny chunk of fluid at time t0 and position x0 . Thus when ∇ · v = 0 the volume
is fixed. In particular, this is the case when the fluid is incompressible.
In fact we can relax the symmetry condition.
Example 4.6.3 Example 4.6.1, generalized yet again. For any d × d
matrix V, the solution of
b0 (t) = Vb(t) b(t0 ) = e
is
b(t) = eV(t−t0 ) e
where the exponential of a d × d matrix B is defined by the power series
1 3 ∞ 1
eB = 1 + B + 21 B 2 + Bn
P
B + ··· =
3! n=0 n!
let E be the d × d matrix whose k th column is e(k) and E(t) be the d × d matrix
whose k th column is b(k) (t). Then the volume of the parallelepiped with edges
e(k) , 1 ≤ k ≤ d, is V (t0 ) = det E and the volume of the parallelepiped with
edges b(k) (t), 1 ≤ k ≤ d, is
V (t) = det E(t) = det eV(t−t0 ) E = det eV(t−t0 ) det E = det eV(t−t0 ) V (t0 )
(1789–1857) in 1829.
CHAPTER 4. INTEGRAL THEOREMS 284
have5 det eB = etr B , where tr B, called the trace of the matrix B, is the sum
of the diagonal matrix elements of B. So
d
V 0 (t0 ) X
V (t) = e(t−t0 ) tr V V (t0 ) ⇒ = tr V = Vi,i
V (t0 ) i=1
The solution6 to
is
b1 (t) = β1 cos t − β2 sin t cos t − sin t
or equivalently b(t) = b(0)
b2 (t) = β1 sin t + β2 cos t sin t cos t
Consequently the vector b(t) has the same length as b(0). The angle between
b(t) and b(0) is just t radians. So, in this example, no matter what direction
vectors ê(k) we pick, the chunk of fluid just rotates at one radian per unit time.
In the figure below, the outlined rectangle is the initial square. That is, the
square with edges b(k) (t0 ) = ê(k) . The shaded rectangle is that with edges
b(k) (t) for some t a bit bigger than t0 .
Example 4.6.5 Example 4.6.4, generalized. The behaviour of Example
4.6.4 is typical of V’s that are antisymmetric matrices, i.e. that obey Vi,j =
−Vj,i for all i, j. As we have already observed, for any d × d matrix V, the
solution of b0 (t) = Vb(t), b(0) = e is b(t) = eVt e. We now show that if V
is a 3 × 3 antisymmetric matrix, then eVt is a rotation.
Assuming that V is not the zero matrix (in which case eVt is the identity
matrix for all t), we can find a number Ω > 0 and a unit vector k̂ = (k1 , k2 , k3 )
4 The proof is not so hard, though we’ll only outline it. Just denote by β the magnitude
of the largest matrix element of B. Then use the definition of the matrix product to prove
that the largest matrix element of B n has magnitude at most (dβ)n .
5 Again, we won’t prove this. But for a diagonal matrix, it is easy — just compute both
(not necessarily the standard unit vector parallel to the z-axis) such that
0 −Ωk3 Ωk2
V = Ωk3 0 −Ωk1 (R)
−Ωk2 Ωk1 0
k3 = −V1,2 /Ω. Also, let ı̂ı be any unit vector orthogonal to k̂ (again, not
necessarily the standard one) and ̂ = k̂ × ı̂ı. So ı̂ı, ̂, k̂ is a right-handed
system of three mutually perpendicular unit vectors.
Observe that, for any vector e = (e1 , e2 , e3 )
0 −Ωk3 Ωk2 e1 k2 e3 − k3 e2
Ve = Ωk3 0 −Ωk1 e2 = Ω k3 e1 − k1 e3 = Ωk̂ × e
−Ωk2 Ωk1 0 e3 k1 e2 − k2 e1
In particular,
= k̂
out explicitly)
∂v1
∂v2 ∂v1
x0 , t0 − ∂v
0 ∂x x0 , t0 − ∂x x0 , t0 ∂x3 ∂x
3
x0 , t0
1 ∂v1 2 1 1
x0 , t0 + ∂v ∂v2
x0 , t0 − ∂v
− ∂x2 ∂x1 x0 , t0
2
0 ∂x3 ∂x2 x0 , t0
3
2 ∂v
x0 , t0 + ∂x13 x0 , t0 −∂v
∂v ∂v3
−∂x31 ∂x3 x0 , t0 + ∂x2 x0 , t0
2
0
Ωk̂ = 21 ∇ × v x0 , t0
• the real and imaginary parts of each complex conjugate pair of eigenvec-
tors (they act like ı̂ı, ̂ above)
Resumé so far:
We have now seen that
• when the matrix V defined in (M) is symmetric and the direction vectors
ê(k) of the cube are eigenvectors of V, then, at time t0 , the chunk of
fluid is not changing orientation
but is changing volume at instantaneous
relative rate ∇ · v x0 , t0 and
• when the matrix V defined in (M) is antisymmetric, then, at time t0 , the
chunk of fluid is not changing
shape orsize but is rotating about the axis
∇ × v x0 , t0 at rate 12 ∇ × v x0 , t0 . For this reason, ∇ × v is often
referred to as a “vorticity” meter.
These agree with our earlier interpretations of divergence and curl.
The general case:
Now consider a general matrix V. It can always be written as the sum
V =S+A
is
b(t) = eVt e = e(A+S)t e
CHAPTER 4. INTEGRAL THEOREMS 287
If S and A were ordinary numbers, we would have e(A+S)t = eAt eSt . But for
matrices this need not be the case, unless S and A happen to commute7 . For
arbitrary matrices, it is still true that
h in
e(A+S)t = lim eAt/n eSt/n
n→∞
8
This is called the Lie product formula. It shows that our tiny chunk of fluid
mixes together the behaviours of A and S, scaling a bit, then rotating a bit,
then scaling a bit and so on.
Example 4.6.8 v(x, y) = 2yı̂ı. In this example
0 2 0 1 0 1
V= =S+A with S= A=
0 0 1 0 −1 0
is
b1 (t) = β1 + 2β2 t 1 2t
or equivalently b(t) = b(0)
b2 (t) = β2 0 1
The solution to the S part of the flow
is9
b1 (t) = β1 cosh t + β2 sinh t cosh t sinh t
or equivalently b(t) = b(0)
b2 (t) = β1 sinh t + β2 cosh t sinh t cosh t
Under the S part of the flow ê(1) scales by a factor of et , which is bigger than
one for t > 0 and ê(2) scales by a factor of e−t , which is smaller than one for
t > 0.
The solution to the A part of the flow
is
b1 (t) = β1 cos t + β2 sin t cos t sin t
or equivalently b(t) = b(0)
b2 (t) = −β1 sin t + β2 cos t − sin t cos t
The A part of the flow rotates clockwise about the origin at one radian per
unit time.
Here are some figures to help us visualize this.
• The first shows a square with edges ê(1) , ê(2) and its image under the
full flow t = 0.4 later. Under this full flow the vector ê(k) → e0.4V ê(k) .
The darkly shaded parallelogram has edges e0.4V ê(k) .
• The second shows its image under 0.4 time units of the S-flow (that is,
ê(k) → e0.4S ê(k) ). The lightly shaded rectangle has edges e0.4S ê(k) .
• The third applies 0.4 time units of the A-flow to the shaded rectangle of
the second figure. So the lightly shaded rectangle of the third figure has
edges e0.4S ê(k) and the darkly shaded rectangle has edges e0.4A e0.4S ê(k) .
n“1 n“5 n “ 10
n
So we can see that, as n increases, e0.4A/n e0.4S/n ê(k) becomes a better
and better approximation to e0.4(A+S) ê(k) .
with F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z) being functions of three vari-
ables.
c A 2-form is an expression of the form
with F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z) being functions of three vari-
ables.
d A 3-form is an expression of the form f (x, y, z) dx∧dy∧dz, with f (x, y, z)
being a function of three variables.
At this stage (there’ll be more later), just think of “dx”, “dy”, “dz”, “dx ∧ dy”,
and so on, as symbols. Do not yet attempt to attach any significance to them.
♦
There are four operations involving differential forms — addition, multi-
plication (∧), differentiation (d) and integration. Here are their definitions.
First, addition is defined, and works, just the way that you would expect it to.
2 We could also define, for example, a 1-form as an ordered list
F1 (x, y, z) , F2 (x, y, z) , F3 (x, y, z) of three functions and just view F1 (x, y, z) dx +
F2 (x, y, z) dy + F3 (x, y, z) dz as another notation for F1 (x, y, z) , F2 (x, y, z) , F3 (x, y, z) .
CHAPTER 4. INTEGRAL THEOREMS 291
♦
There is one wrinkle in multiplication. It is not commutative, meaning
that α ∧ β need not be the same as β ∧ α. You have already seen some
noncommutative products. If a and b are two vectors in R3 , then a × b =
−b × a. Also, if A and B are two n × n matrices, the matrix product AB need
not be the same as BA.
Definition 4.7.3 Multiplication of differential forms. We now define a
multiplication rule for differential forms. If ω is a k-form and ω 0 is a k 0 -form
then the product will be a (k+k 0 )-form and will be denoted ω∧ω 0 (read “omega
wedge omega prime”). It is determined by the following properties.
a If f is a function (i.e. a 0-form), then
f F1 dx + F2 dy + F3 dz = (f F1 ) dx+(f F2 ) dy+(f F3 ) dz
f F1 dy ∧ dz+F2 dz ∧ dx+F3 dx ∧ dy = (f F1 ) dy ∧ dz + (f F2 ) dz ∧ dx
+ (f F3 ) dx ∧ dy
f g dx ∧ dy ∧ dz = (f g) dx ∧ dy ∧ dz
f1 ω1 + f2 ω2 ∧ ω 0 = f1 (ω1 ∧ ω 0 ) + f2 (ω2 ∧ ω 0 )
Similarly,
ω ∧ ω0 = ω0 ∧ ω
CHAPTER 4. INTEGRAL THEOREMS 292
(so that the wedge product is commutative) and if k and k 0 are both odd
then
ω ∧ ω 0 = −ω 0 ∧ ω
(so that the wedge product is anticommutative). In particular, if ω is a
d-form with d odd
ω∧ω =0
(ω ∧ ω 0 ) ∧ ω 00 = ω ∧ ω 0 ∧ ω 00
♦
So the wedge product obeys most of the usual multiplication rules, with
the one big exception that if ω is k-form and ω 0 is a k 0 -form with k and k 0 both
odd then ω ∧ ω 0 = −ω 0 ∧ ω.
The best way to get a handle on the wedge product is to work through
some examples, like these.
Example 4.7.4 Let ω = F1 dx + F2 dy + F3 dz and ω 0 = G1 dx + G2 dy + G3 dz
be any two 1-forms. Their product is
ω ∧ ω 0 = F1 dx + F2 dy + F3 dz ∧ G1 dx + G2 dy + G3 dz
= F1 dx ∧ G1 dx + F1 dx ∧ G2 dy + F1 dx ∧ G3 dz
+ F2 dy ∧ G1 dx + F2 dy ∧ G2 dy + F2 dy ∧ G3 dz
+ F3 dz ∧ G1 dx + F3 dz ∧ G2 dy + F3 dz ∧ G3 dz
(by linearity, i.e. by part (b) of the last Definition)
= F1 G1 dx ∧ dx + F1 G2 dx ∧ dy + F1 G3 dx ∧ dz
+ F2 G1 dy ∧ dx + F2 G2 dy ∧ dy + F2 G3 dy ∧ dz
+ F3 G1 dz ∧ dx + F3 G2 dz ∧ dy + F3 G3 dz ∧ dz
= F1 G2 − F2 G1 ) dx ∧ dy + F3 G1 − F1 G3 ) dz ∧ dx
+ F2 G3 − F3 G2 ) dy ∧ dz
because
dx ∧ dx = dy ∧ dy = dz ∧ dz = 0
and
Note that, looking at the last example, if we view F = (F1 , F2 , F3 ) and
G = (G1 , G2 , G3 ) as vectors, we can write the product simply as
Equation 4.7.5
F1 dx + F2 dy + F3 dz ∧ G1 dx + G2 dy + G3 dz
= (F × G)1 dy ∧ dz + (F × G)2 dz ∧ dx + (F × G)3 dx ∧ dy
where we are using (F × G)` to denote the `th component of the cross
product F × G. In the special case that F3 = G3 = 0, we have
Equation 4.7.6
F1 dx + F2 dy ∧ G1 dx + G2 dy = F1 G2 − F2 G1 ) dx ∧ dy
CHAPTER 4. INTEGRAL THEOREMS 293
F F2
= det 1 dx ∧ dy
G1 G2
We can now see why in the Definition 4.7.1.c of 2-forms
• there were no dx ∧ dx or dy ∧ dy or dz ∧ dz terms — they are all zero
and
• there were no dy ∧ dx or dz ∧ dy or dx ∧ dz terms — they can all be
rewritten using dx ∧ dy, dy ∧ dz and dz ∧ dx terms (or vice versa).
The reason that we chose to write the Definition 4.7.1.c as
F1 dy ∧ dz + F2 dz ∧ dx + F3 dx ∧ dy
f1 dx ∧ dy + f2 dx ∧ dz + f3 dy ∧ dz
was to make formulae like 4.7.5 work. The easy way to remember
F1 dy ∧ dz + F2 dz ∧ dx + F3 dx ∧ dy
ω ∧ ω 0 = F1 dx + F2 dy + F3 dz ∧ G1 dy ∧ dz + G2 dz ∧ dx + G3 dx ∧ dy
= F1 G1 dx ∧ dy ∧ dz + F2 G2 dy ∧ dz ∧ dx + F3 G3 dz ∧ dx ∧ dy
= F1 G1 + F2 G2 + F3 G3 ) dx ∧ dy ∧ dz
dy ∧ dz ∧ dx = −dy ∧ dx ∧ dz = dx ∧ dy ∧ dz
dz ∧ dx ∧ dy = −dx ∧ dz ∧ dy = dx ∧ dy ∧ dz
We have also used that any wedge product of three d{x or y or z}’s with at
least two of the coordinates being the same is zero. For example
dx ∧ dz ∧ dx = −dx ∧ dx ∧ dz = 0
So
F1 dx + F2 dy + F3 dz ∧ G1 dy ∧ dz + G2 dz ∧ dx + G3 dx ∧ dy
= F · G dx ∧ dy ∧ dz
Example 4.7.8 Combining Examples 4.7.4 and 4.7.7, we have the wedge prod-
uct of any three (general) 1-forms F1 dx + F2 dy + F3 dz and G1 dx + G2 dy +
G3 dz and H1 dx + H2 dy + H3 dz is
F1 dx+F2 dy+F3 dz ∧ G1 dx+G2 dy+G3 dz ∧ H1 dx + H2 dy + H3 dz
CHAPTER 4. INTEGRAL THEOREMS 294
= F1 dx + F2 dy + F3 dz ∧
(G × H)1 dy ∧ dz + (G × H)2 dz ∧ dx + (G × H)3 dx ∧ dy
= F1 (G × H)1 + F2 (G × H)2 + F3 (G × H)3 dx ∧ dy ∧ dz
= F1 (G2 H3 −G3 H2 )+F2 (G3 H1 −G1 H3 )+F3 (G1 H2 −G2 H1 dx ∧ dy ∧ dz
This can be expressed cleanly in terms of determinants. Recalling the rule for
expanding a determinant along its top row
F1 dx+F2 dy+F3 dz ∧ G1 dx+G2 dy+G3 dz ∧ H1 dx+H2 dy+H3 dz
F1 F2 F3
= det G1 G2 G3 dx ∧ dy ∧ dz
H1 H2 H3
Our next operation is a differential operator which unifies and generalizes
gradient, curl and divergence.
Definition 4.7.9 Differentiation of differential forms. If ω is a k-form,
then dω is a k + 1-form, with d being the unique3 such operator that obeys
a d is linear. That is, if ω1 , ω2 are k-forms and a1 , a2 ∈ R, then
d a1 ω1 + a2 ω2 = a1 dω1 + a2 dω2
∂f ∂f ∂f
df = (x, y, z) dx + (x, y, z) dy + (x, y, z) dz
∂x ∂y ∂z
= ∇ f (x, y, z) · dr where dr = dxı̂ı + dy ̂ + dz k̂
♦
Example 4.7.10
a If f (x, y, z) = x, then
∂x ∂x ∂x
df = (x, y, z) dx + (x, y, z) dy + (x, y, z) dz = dx
∂x ∂y ∂z
That is, dx really is the operator d applied to the function x. Similarly,
dy really is the operator d applied to the function y and dz really is the
operator d applied to the function z.
3 That d is unique just means that the action of d on any differential form is completely
determined by the four rules (a), (b), (c), (d). We will see in Example 4.7.10.c,d,e, that this
is indeed the case.
CHAPTER 4. INTEGRAL THEOREMS 295
d ω ∧ dx = dω ∧ dx + (−1)k ω ∧ d dx
= dω ∧ dx
Similarly
d ω ∧ dy = dω ∧ dy d ω ∧ dz = dω ∧ dz
Example 4.7.11 In Definition 4.7.9.c, we defined, for any function f (x, y, z)
of three variables
∂f ∂f ∂f
df = (x, y, z) dx + (x, y, z) dy + (x, y, z) dz
∂x ∂y ∂z
CHAPTER 4. INTEGRAL THEOREMS 296
The analogous formulae4 for functions of one or two variables also apply.
df
df (t) = (t) dt
dt
∂f ∂f
df (u, v) = (u, v) du + (u, v) dv
∂u ∂v
Let us summarize what we have seen in the Example 4.7.10.
Lemma 4.7.12
a For any 0-form
df = ∇ f (x, y, z) · dr
of two of the three variables. Similarly you can view f (u, v) as a function of three variables
that happens to be independent of one of the three variables.
CHAPTER 4. INTEGRAL THEOREMS 297
♦
Finally, after all of these definitions, we have a very compact theorem that
simultaneously covers the fundamental theorem of calculus, Green’s theorem.
Stokes’ theorem and the divergence theorem. Had we given all of our definitions
in n dimensions, rather than just three dimensions, it would cover a lot more.
This general theorem is also called Stokes’ theorem.
Theorem 4.7.14 Stokes’ Theorem. If ω is a k-form (with k = 0, 1, 2) and
D is a (k + 1)-dimensional domain of integration, then
Z Z
dω = ω
D ∂D
To see the connection between the general Stokes’ theorem 4.7.14 and the
Stokes’ and divergence theorems of the earlier part of this chapter, here are
the k = 1 and k = 2 cases of Theorem 4.7.14 again.
• Let ω = F1 dx + F2 dy + F3 dz be a 1-form and let S be a piecewise
smooth oriented surface as in (our original) Stokes’ theorem 4.4.1. Then,
by Lemma 4.7.12.b,
∇ × F)1 dy ∧ dz + (∇
dω = (∇ ∇ × F)2 dz ∧ dx + (∇
∇ × F)3 dx ∧ dy
5.2 Exercises
1. ∗. True or false?
a ∇ · (a × r) = 0, where a is a constant vector in R3 , and r is the
vector field r = (x, y, z).
b ∇ × (∇∇f ) = 0 for all scalar fields f on R3 with continuous second
partial derivatives.
∇ · F, for every vector field F in R3 with
c ∇ · (f F) = ∇ (f ) · F + f∇
continuous partial derivatives, and every scalar function f in R3
with continuous partial derivatives.
d Suppose F is a vector field with continuous partial derivatives in the
region D, where D is R3 without the origin. If ∇ · F > 0 throughout
D, then the flux of F through the sphere of radius 5 with center at
the origin is positive.
299
CHAPTER 5. TRUE/FALSE AND OTHER SHORT QUESTIONS 300
x
b Of the two functions shown below, one is a function f (x) and one is
its curvature κ(x). Which is which?
y
x
c Let C be the curve of intersection of the cylinder x2 + z 2 = 1 and
the saddle xz = y. Parametrise C. (Be sure to specify the domain
of your parametrisation.)
g Let
F(x, y) = ax + by , cx + dy
Give the full set of a, b, c and d such that F is conservative.
h If r(s) has been parametrized by arclength (i.e. s is arclength), what
is the arclength of r(s) between s = 3 and s = 5?
i Let F be a 2D vector field which is defined everywhere except at the
points marked P and Q. Suppose that ∇ × F = 0 everywhere on
the domain of F. Consider the five curves R, S, T , U , and V shown
in the picture.
R U
Q
S
T
P
j Write down a 3D vector field F such that for all closed surfaces S,
the volume enclosed by S is equal to
ZZ
F · n̂ dS
S
k Consider the vector field F in the xy-plane shown below. Is the k̂th
component of ∇ × F at P positive, negative or zero?
CHAPTER 5. TRUE/FALSE AND OTHER SHORT QUESTIONS 303
where `−S’ denotes the surface S but with the opposite orientation.
CHAPTER 5. TRUE/FALSE AND OTHER SHORT QUESTIONS 304
∇f ) × (∇
(∇ ∇f ) = ∇ × (∇
∇f )
|r0 (s)| = 1
C1
C2
d The work done by the vector field on a particle travelling from point
B to point C along the curve C1 is (a) positive (b) negative (c) zero
(d) there is not enough information to tell.
e The work done by the vector field on a particle travelling from point
B to point C along the curve C2 is (a) positive (b) negative (c) zero
(d) there is not enough information to tell.
f The vector field F is (a) the gradient of some function f (b) the curl
of some vector field G (c) not conservative (d) divergence free.
8. ∗. Which of the following statements are true (T) and which are false
(F)?
a The curve defined by
|r0 (s)| = 1
11. ∗. Say whether the following statements are true (T) or false (F). You
may assume that all functions and vector fields are defined everywhere
and have derivatives of all orders everywhere.
a The divergence of ∇ × F is zero, for every F.
R
b In a simply connected region, C F·dr depends only on the endpoints
of C.
c If ∇ f = 0, then f is a constant function.
∇ · (F) + F · ∇ f
b ∇ · (f F) = f∇
∇ · (F) + F · ∇ f
c ∇ × (f F) = f∇
d None of the above are true.
13. ∗. True or False. Consider vector fields F and scalar functions f and
g which are defined and smooth in all of three-dimensional space. Let
r = (x, y, z) represent a variable point in space, and let ω = (ω1 , ω2 , ω3 )
be a constant vector. Let Ω be a smoothly bounded domain with outer
normal n̂. Which of the following are identites, always valid under these
assumptions?
a ∇ · ∇f = 0
b F × ∇f = f ∇ · F
c ∇ 2 f = ∇ (∇
∇ · f)
d ∇ × ∇f = 0
∇ × f ) + (∇
e (∇ ∇ × g) = ∇ f × ∇ g
f ∇ ·∇ ×F = 0
r
g ∇· |r|2 = 0 for r 6= 0
h ∇ × (ω
ω × r) = 0
CHAPTER 5. TRUE/FALSE AND OTHER SHORT QUESTIONS 309
ZZZ ZZZ ZZ
i ∇ · F dV = −
f∇ ∇ f · F dV + f F · n̂ dS
Ω Ω ∂Ω
ZZ ZZZ
j f n̂ dS = − ∇ f dV
∂Ω Ω
14. ∗. Determine if the given statements are True or False. Provide a reason
or a counterexample.
a A constant vector field is conservative on R3 .
a = f (r, v)r
solid right circular cylinder of radius b with one base in the plane
z = 1 and the other base in the plane z = 3.
c Let F and G be smooth vector
H fields defined
RR in R3 . Suppose that,
for every circle C, we have C F · dr = S G · n̂ dS, where S is the
oriented disk with boundary C. Prove that G = ∇ × F.
Appendix A
Appendices
A.1 Trigonometry
A.1.1 Trigonometry — Graphs
−π −π π π 3π 2π −π −π π π 3π 2π −π −π π π 3π 2π
2 2 2 2 2 2 2 2 2
−1 −1
312
APPENDIX A. APPENDICES 313
• Reflection
sin π2 − θ = cos θ π
cos 2 − θ = sin θ
• Rotation by π
• Pythagoras
sin2 θ + cos2 θ = 1
tan2 θ + 1 = sec2 θ
1 + cot2 θ = csc2 θ
• Cosine
• Tangent
tan α + tan β
tan(α + β) =
1 − tan α tan β
tan α − tan β
tan(α − β) =
1 + tan α tan β
• Double angle
= 1 − 2 sin2 (θ)
2 tan(θ)
tan(2θ) =
1 − tan2 θ
1 + cos(2θ)
cos2 θ =
2
1 − cos(2θ)
sin2 θ =
2
1 − cos(2θ)
tan2 θ =
1 + cos(2θ)
• Products to sums
sin(α + β) + sin(α − β)
sin(α) cos(β) =
2
cos(α − β) − cos(α + β)
sin(α) sin(β) =
2
cos(α − β) + cos(α + β)
cos(α) cos(β) =
2
• Sums to products
α+β α−β
sin α + sin β = 2 sin cos
2 2
α+β α−β
sin α − sin β = 2 cos sin
2 2
α+β α−β
cos α + cos β = 2 cos cos
2 2
α+β α−β
cos α − cos β = −2 sin sin
2 2
A.1.5 Inverse Trigonometric Functions
arctan x
arcsin x arccos x
Domain: −1 ≤ x ≤ 1 Domain: −1 ≤ x ≤ 1 Domain: all real num-
bers
Range: − π2 ≤ Range: 0 ≤ arccos x ≤ π Range: − π2 <
π π
arcsin x ≤ 2 arctan x < 2
π π
2 π 2
π
−1 1 2
−π
2 −π
2
−1 1
sin(arcsin x) = x −1 ≤ x ≤ 1
APPENDIX A. APPENDICES 315
cos(arccos x) = x −1 ≤ x ≤ 1
tan(arctan x) = x any real x
π π
2 π
π
π 2
2
−1 1
−π
2
−1 1
π
arccsc x 6= 0 arcsec x 6=
2
Again
π π
arccsc(csc θ) = θ − ≤ θ ≤ , θ 6= 0
2 2
π
arcsec(sec θ) = θ 0 ≤ θ ≤ π, θ 6=
2
arccot(cot θ) = θ 0<θ<π
and
csc(arccsc x) = x |x| ≥ 1
sec(arcsec x) = x |x| ≥ 1
cot(arccot x) = x any real x
• e0 = 1, q0 = 1
ex qx
• ex+y = ex ey , ex−y = , q x+y = q x q y , q x−y =
ey qy
1 1
• e−x = , q −x =
ex qx
y y
• ex = exy , qx = q xy
d x d g(x) d x
• e = ex , e = g 0 (x)eg(x) , q = (ln q) q x
dx dx dx
• ex dx = ex + C, e dx = a1 eax + C if a 6= 0
R R ax
APPENDIX A. APPENDICES 316
∞
X xn
• ex =
n=0
n!
• lim ex = ∞, lim ex = 0
x→∞ x→−∞
lim q x = ∞, lim q x = 0 if q > 1
x→∞ x→−∞
lim q x = 0, lim q x = ∞ if 0 < q < 1
x→∞ x→−∞
• The graph of 2x is given below. The graph of q x , for any q > 1, is similar.
y y = 2x
6
2
1
x
−3 −2 −1 1 2 3
A.2.2 Logarithms
In the following, x and y are arbitrary real numbers that are strictly bigger
than 0 (except where otherwise specified), p and q are arbitrary constants that
are strictly bigger than one, and e is 2.7182818284, to ten decimal places. The
notation ln x means loge x. Some people use log x to mean log10 x, others use
it to mean loge x and still others use it to mean log2 x.
• eln x = x, q logq x = x
• ln ex = x, logq q x = x for all −∞ < x < ∞
ln x logp x logp x
• logq x = , ln x = , logq x =
ln q logp e logp q
• ln 1 = 0, ln e = 1
logq 1 = 0, logq q = 1
• ln(xy) = ln x + ln y, logq (xy) = logq x + logq y
x x
• ln = ln x − ln y, logq = logq x − logq y
y y
1 1
• ln = − ln y, logq = − logq y
y y
• ln(xy ) = y ln x, logq (xy ) = y logq x
d 1 d 1
• ln x = , logq x =
dx x dx x ln q
Z Z
x
• ln x dx = x ln x − x + C, logq x dx = x logq x − +C
ln q
APPENDIX A. APPENDICES 317
• lim ln x = ∞, lim ln x = −∞
x→∞ x→0+
lim logq x = ∞, lim logq x = −∞
x→∞ x→0+
• The graph of log10 x is given below. The graph of logq x, for any q > 1,
is similar.
y
y = log10 x
1.0
0.5
x
1 5 10 15
−0.5
−1.0
F (x) F 0 (x) = dF
dx
1
ln x x0
g (x)
ln g(x) g(x)
1
loga x x ln a
arcsin x √ 1
1−x2
g 0 (x)
arcsin g(x) √
1−g(x)2
1
arccos x − √1−x 2
1
arctan x 1+x2
g 0 (x)
arctan g(x) 1+g(x)2
arccsc x − |x|√1x2 −1
arcsec x √1
|x| x2 −1
1
arccot x − 1+x 2
f y(x) y 0 (x)
R
F y(x) where F (y) = f (y) dy
a ax + C
xa+1
xa a+1 + C if a 6= −1
1
x ln |x| + C
g(x)a+1
g(x)a g 0 (x) a+1 + C if a 6= −1
R
f (x) F (x) = f (x) dx
sin x − cos x + C
g 0 (x) sin g(x) − cos g(x) + C
cos x sin x + C
tan x ln | sec x| + C
csc x ln | csc x − cot x| + C
sec x ln | sec x + tan x| + C
cot x ln | sin x| + C
sec2 x tan x + C
csc2 x − cot x + C
sec x tan x sec x + C
csc x cot x − csc x + C
APPENDIX A. APPENDICES 319
R
f (x) F (x) = f (x) dx
ex ex + C
eg(x) g 0 (x) eg(x) + C
1 ax
eax a e +C
1
ax ln a a x
+C
ln x x ln x − x + C
√ 1 arcsin x + C
1−x2
g 0 (x)
√ arcsin g(x) + C
1−g(x)2
√ 1
a2 −x2
arcsin xa + C
1
1+x2 arctan x + C
g 0 (x)
1+g(x)2 arctan g(x) + C
1 1 x
a2 +x2 a arctan a + C
√1 arcsec x + C \quad(x > 1)
x x2 −1
for f . When x0 = 0 this is also called the Maclaurin series for f . Here are
Taylor series expansions of some important functions.
∞
X 1 n
ex = x for − ∞ < x < ∞
n=0
n!
1 1 1
= 1 + x + x2 + x3 + · · · + xn + · · ·
2 3! n!
∞
X (−1)n 2n+1
sin x = x for − ∞ < x < ∞
n=0
(2n + 1)!
1 3 1 (−1)n 2n+1
=x− x + x5 − · · · + x + ···
3! 5! (2n + 1)!
∞
X (−1)n 2n
cos x = x for − ∞ < x < ∞
n=0
(2n)!
1 2 1 (−1)n 2n
=1− x + x4 − · · · + x + ···
2! 4! (2n)!
∞
1 X
= xn for − 1 ≤ x < 1
1 − x n=0
= 1 + x + x2 + x3 + · · · + xn + · · ·
∞
1 X
= (−1)n xn for − 1 < x ≤ 1
1+x n=0
APPENDIX A. APPENDICES 320
= 1 − x + x2 − x3 + · · · + (−1)n xn + · · ·
∞
X 1 n
ln(1 − x) = − x for − 1 ≤ x < 1
n=1
n
1 1
= −x − 12 x2 − x3 − · · · − xn − · · ·
3 n
∞
X (−1)n n
ln(1 + x) = − x for − 1 < x ≤ 1
n=1
n
1 (−1)n n
= x − 12 x2 + x3 − · · · − x − ···
3 n
p(p − 1) 2 p(p − 1)(p − 2) 3
(1 + x)p = 1 + px + x + x + ···
2 3!
p(p − 1)(p − 2) · · · (p − n + 1) n
+ x + ···
n!
px, y, zq
z
y
x
y px, y, 0q
x
and here are three figures showing a surface of constant x, a surface of
constant x, and a surface of constant z.
z z z
p0, 0, zq
y y y
px, 0, 0q p0, y, 0q
x x x
surface of constant x surface of constant y surface of constant z
(a plane) (a plane) (a plane)
Finally here is a figure showing the volume element dV in cartesian coor-
dinates.
dz
dx
dy
volume element dV “ dx dy dz
APPENDIX A. APPENDICES 321
px, y, zq
z
y
θ r
x px, y, 0q
The cartesian and cylindrical coordinates are related by
r y
y y
θ
x
x x
surface of constant r surface of constant z
surface of constant θ
(a cylindrical shell) (a plane)
(a plane)
Finally here is a figure showing the volume element dV in cylindrical coor-
dinates.
r dr
dz
r dθ
volume element dV “ r dr dθ dz
z
p0, 0, zq ρ sin ϕ
px, y, zq
ρ cos ϕ
ϕ ρ
y
px, 0, 0q θ
px, y, 0q
ρ sin ϕ sin θ
x
and here are two more figures giving the side and top views of the previous
figure.
z
ρ sin ϕ px, y, zq y
p0, 0, zq
θ ρ sin ϕ
ρ cos ϕ ρ sin ϕ cos θ
ρ
ϕ px, y, 0q
px, 0, 0q
ρ sin ϕ sin θ
px, y, 0q
x
side view top view
The cartesian and spherical coordinates are related by
ϕ
ρ
y y y
θ
x x x
surface of constant ρ surface of constant θ surface of constant ϕ
(a sphere) (a plane) (a cone)
Here is a figure showing the surface element dS in spherical coordinates
APPENDIX A. APPENDICES 323
ρ sin ϕ dθ
ρ dϕ
ρ sin ϕ dθ
dθ
ρ sin ϕ
ρ dϕ
ϕ ϕ ρ
dϕ
y y
x x
Finally, here is a figure showing the volume element dV in spherical coor-
dinates
APPENDIX A. APPENDICES 324
ρ sin ϕ dθ
ρ dϕ
dρ
x
volume element dV “ ρ2 sin ϕ dρ dθ dϕ
x x
3π{2
x
π{2
ρ “ 1, φ “ ´π{2, φ “ 3π{2
It is also sometimes convenient to extend the above definitions by saying
that x = ρ cos φ and y = ρ sin φ even when ρ is negative. For example, the
following figure shows (x, y) for ρ = 1, φ = π4 and for ρ = −1, φ = π4 .
y
ρ “ 1, φ “ π{4
π{4
x
ρ “ ´1, φ “ π{4
Both points lie on the line through the origin that makes an angle of 45◦
with the x-axis and both are a distance one from the origin. But they are on
opposite sides of the the origin.
The area element in polar coordinates is
dA = ρ dρ dφ
APPENDIX A. APPENDICES 326
ρ ρ dφ
dφ dρ
px, y, zq
z
y
ρ
φ
x px, y, 0q
The cartesian and cylindrical coordinates are related by
ρ y y y
φ
x x x
surface of constant ρ surface of constant φ surface of constant z
(a cylindrical shell) (half a plane) (a plane)
Finally here is a figure showing the volume element dV in cylindrical coor-
dinates.
ρ dρ
dz
ρ dφ
volume element dV “ ρ dρ dφ dz
APPENDIX A. APPENDICES 327
z
p0, 0, zq r sin θ
px, y, zq
r cos θ
r
θ
y
px, 0, 0q φ
px, y, 0q
r sin θ sin φ
x
Here are two more figures giving the side and top views of the previous
figure.
z
r sin θ px, y, zq y
p0, 0, zq
φ r sin θ
r cos θ r sin θ cos φ
r
θ px, y, 0q
px, 0, 0q
r sin θ sin φ
px, y, 0q
x
side view top view
The cartesian and spherical coordinates are related by
θ
r
y y y
φ
x x x
surface of constant r surface of constant φ surface of constant θ
(a sphere) (half a plane) (a cone)
Here is a figure showing the surface element dS in spherical coordinates
APPENDIX A. APPENDICES 328
r sin θ dφ
r dθ
r sin θ dφ
dφ
r sin θ
r dθ
r
θ dθ θ r
y y
x x
Finally, here is a figure showing the volume element dV in spherical coor-
dinates
APPENDIX A. APPENDICES 329
r sin θ dφ
r dθ
dr
x
volume element dV “ r 2 sin θ dr dθ dφ
and two extracts of the above figure to make it easier to see how the factors
r dθ and r sin θ dφ arise.
z
z
r sin θ dφ
dφ
r sin θ
r dθ
r
θ dθ θ
dr y
y
x
x
x =constant
two lines one line two lines circle
cross-section
y =constant
two lines two lines two lines circle
cross-section
z =constant
ellipse parabola hyperbola circle
cross-section
sketch
is beyond what we can cover here, but is not too difficult for a standard linear algeba course.
APPENDIX A. APPENDICES 331
elliptic
name ellipsoid elliptic cone
paraboloid
equation in stan- x2 y2 z2 x2 y2 z x2 y2 z2
a2 + b2 + c2 =1 a2 + b2 = c a2 + b2 = c2
dard form
two lines if x = 0,
x = constant
ellipse parabola hyperbola if x 6=
cross-section
0
two lines if
y = constant
ellipse parabola y = 0,hyperbola
cross-section
if y 6= 0
z = constant
ellipse ellipse ellipse
cross-section
sketch
sketch
and F (t) are continuous functions and that a0 (t) is not zero.
a The general solution to the ODE (A.9.1) is of the form
where
• n is the order of (A.9.1)
• yp (t) is any solution to (A.9.1)
• C1 , C2 , · · ·, Cn are arbitrary constants
• y1 , y2 , · · ·, yn are n independent solutions to the homogenous equa-
tion
a0 (t)y (n) (t) + a1 (t)y (n−1) (t) + · · · + an−1 (t)y 0 (t) + an (t)y(t) = 0
b Given any constants b0 , · · ·, bn−1 there is exactly one function y(t) that
obeys the ODE (A.9.1) and the initial conditions
di q(t)
Ri(t) + L (t) + = x(t) (A.9.3)
dt C
Assuming that R, L, C and x(t) are known, this is still one differential equation
in two unknowns, i(t) and q(t). Fortunately, there is a relationship between
the two. Namely
dq
i(t) = (t) = Cy 0 (t) (A.9.4)
dt
APPENDIX A. APPENDICES 334
This just says that the capacitor cannot create or destroy charge on its own;
all charging of the capacitor must come from the current. Substituting (A.9.4)
into (A.9.3) gives
LCy 00 (t) + RCy 0 (t) + y(t) = x(t)
As a concrete example, we’ll take an ac voltage source and choose the origin
of time so that x(0) = 0, x(t) = E0 sin(ωt). Then the differential equation
becomes
LCy 00 (t) + RCy 0 (t) + y(t) = E0 sin(ωt) (A.9.5)
This is a second order, linear, constant coefficient ODE. So we know, from
Theorem A.9.2, that the general solution is of the form yp (t)+C1 y1 (t)+C2 y2 (t),
where
• yp (t), the particular solution, is any one solution to (A.9.5),
• C1 , C2 are arbitrary constants and
• y1 (t), y2 (t) are any two independent solutions of the corresponding ho-
mogeneous equation
So to find the general solution to (A.9.5), we need to find three functions: y1 (t),
y2 (t) and yp (t).
• Finding y1 (t) and y2 (t): The best way to find y1 and y2 is to guess
them. Any solution, yh (t), of (A.9.6) has to have the property that
yh (t), RCyh0 (t) and LCyh00 (t) cancel each other out for all t. We choose
our guess so that yh (t), yh0 (t) and yh00 (t) are all proportional to a single
function of t. Then it will be easy to see if yh (t), RCyh0 (t) and LCyh00 (t)
all cancel. All derivatives of the function ert are again proportional to
ert . Hence we try yh (t) = ert , with the constant r to the determined.
This guess is a solution of (A.9.6) if and only if
Substitute in
yh (t) = v(t)e−ρt
yh0 (t) = −ρv(t)e−ρt + v 0 (t)e−ρt
yh00 (t) = ρ2 v(t)e−ρt − 2ρv 0 (t)e−ρt + v 00 (t)e−ρt
= v 00 (t)e−ρt
• Finding yp (t): The best way to find yp is to guess it. We guess that the
circuit responds to an oscillating input voltage with an output voltage
APPENDIX A. APPENDICES 336
or
and hence, applying sin(A+B) = sin A cos B+cos A sin B with A = ωt−ϕ
and B = ϕ,
1 − LCω 2 A = E0 cos(ϕ)
(A.9.9)
RCωA = E0 sin(ϕ) (A.9.10)
(A.9.10) RCω
=⇒ tan(ϕ) =
(A.9.9) 1 − LCω 2
RCω
=⇒ ϕ = arctan
1 − LCω 2
p q 2
(A.9.9)2 + (A.9.10)2 =⇒ 1−LCω 2 + R2 C 2 ω 2 A = E0
E0
=⇒ A = p
(1−LCω 2 )2 + R2 C 2 ω 2
ω
Note that there is a small range of frequencies that give a large amplitude
response. This is the phenomenon of resonance. It is exploited in the design of
radio and television tuning circuitry. It has also been dramatically illustrated
in, for example, the collapse4 of the Tacoma narrows bridge.
1 Gustav Robert Kirchhoff (1824--1887) was a German physicist.
2 Don’t make the mistake of thinking that C1 and C2 have to be real numbers, forcing
D2 to be pure imaginary. In most applications, D1 and D2 will be pure real and C1 and C2
APPENDIX A. APPENDICES 337
y0 + y = 0 (A.9.11)
We shall then impose various boundary conditions and see what happens.
The function y(t) = ert is a solution to (A.9.11) if and only if
r2 ert + ert = 0 ⇐⇒ r2 + 1 = 0 ⇐⇒ r = ±i
This gives
where C1 = C10 + C20 , and C2 = i(C10 − C20 ). Note that there is nothing
stopping C10 and C20 from being complex numbers. So there is nothing stopping
C1 = C10 + C20 , and C2 = i(C10 − C20 ) from being real numbers.
The function y(t) satisfies the ODE if and only if it is of the form
physicist and music theorist Jean le Rond d’Alembert (1717-1783). The interested reader
can easily search out more about his life.
4 There are videos of the collapse on the web.
APPENDIX A. APPENDICES 338
The function y(t) satisfies the ODE if and only if it is of the form
The function y(t) satisfies the ODE if and only if it is of the form
1 · Curves
1.1 · Derivatives, Velocity, Etc.
1.1.1 · Exercises
1.1.1.1. Hint. Draw sketches. Don’t forget the range that the parameter
runs over.
1.1.1.2. Hint. Find the value of t at which the three points occur on
the curve.
1.1.1.3. Hint. The curve “crosses itself” when (sin t, t2 ) gives the same
coordinate for different values of t. When these crossings occur will depend
on which crossing you’re referring to, so your answers should all depend on
t.
1.1.1.4. Hint. For part (b), find the position of P relative to the centre
of the circle. Then combine your answer with part (a).
1.1.1.5. Hint. We aren’t concerned with x, so we can eliminate it by
solving one equation for x as a function of y and z and plugging the result
into the other equation.
1.1.1.6. Hint. To determine whether the particle is rising or falling, we
only need to consider its z-coordinate.
1.1.1.7. Hint. This is the setup from Lemma 1.1.4. The two quantities
you’re labelling are related, but different.
1.1.1.8. Hint. See the note just before Example 1.1.6.
1.1.1.9. ∗. Hint. To simplify your answer, remember: the cross product
of a and b is a vector orthogonal to both a and b; the cross product of a
vector with itself is zero; and two orthogonal vectors have dot product 0.
d 2
1.1.1.10. Hint. Evaluate dt |r(t)| .
2
1.1.1.11. ∗. Hint. Just compute |v(t)|. Note that eat + e−at =
e2at + 2 + e−2at .
1.1.1.12. Hint. To figure out what the path looks like, first concentrate
on the x- and y-coordinates.
1.1.1.13. ∗. Hint. Review §1.5. The arc length should be positive.
339
APPENDIX B. HINTS FOR EXERCISES 340
The notation looks a little confusing at first, but we can break it down
piece by piece: dr
dt (t) is a vector, whose components are functions of t. If
we take its magnitude, we’ll get one big function of t. That function is
what we integrate. Before integrating it, however, we should simplify as
much as possible.
1.1.1.16. Hint. r(t) is the position of the particle, so its acceleration is
r00 (t).
1.1.1.17. ∗. Hint. Review §1.5.
1.1.1.18. ∗. Hint. Review §1.1.
1.1.1.19. Hint. (a) First parametrize x2 + y 2 = 9.
1.1.1.20. ∗. Hint. If you got the answer 0 in part (b), you dropped
some absolute value signs.
1.1.1.22. ∗. Hint. The integral you get can be evaluated with a simple
substitution. You may want to factor the integrand first.
1
1.1.1.23. Hint. (b) 4x + 1 + x is a perfect square.
(c), (d) Let
1.1.1.24. Hint. Given the position of a particle, you can find its velocity.
−αt
1.1.1.31. Hint. Define u(t) = eαt dr dr
dt (t) and substitute dt (t) = e u(t)
into the given differential equation to find a differential equation for u.
1.2 · Reparametrization
1.2.1 · Exercises
1.2.1.1. Hint. You’re asked to find the arclength of the curve from s = 1
to s = t.
1.2.1.2. Hint. The arclength will be 0 at P .
1.2.1.3. Hint. a(t0 ) and b(s0 ) describe the same point on R.
1.3 · Curvature
1.3.1 · Exercises
1.3.1.1. Hint. The curve is a circle, so you don’t need to do any calculus.
1.3.1.2. Hint. Because r is a circle, you can parametrize it with respect
to arclength without using an integral. You found κ in Question 1.3.1.1.
1.3.1.3. Hint. When t is large, does the spiral locally look like a circle
of large radius, or small?
1.3.1.4. Hint. ds
dt = |v(t)| = |r0 (t)|
v(t) r0 (t)
1.3.1.5. Hint. T̂ = |v(t)| = |r0 (t)|
1.3.1.7. Hint. You can find the last two quantities by making use of the
first three. Looking ahead, the formula list in Section 1.5 might come in
handy.
|v(t) × a(t)|
1.3.1.8. Hint. We can calculate κ = ds 3 . We can also figure
dt
out what kind of a shape our curve is.
1.3.1.9. Hint. The maximum and minimum values of κ(t) should be
obvious from your formula for κ(t).
1.3.1.11. ∗. Hint. For part (a), determine r(0), r(π), r(2π), r(3π), and
r(4π), to help you map out the motion. Also visualize the thumbtack as
the wheel moves.
For part (d), use the fact that you only care about t = π: where is this
on your sketch? What does that mean about the direction of N̂?
APPENDIX B. HINTS FOR EXERCISES 342
1.4.1.4. Hint. (a) Show that the tangent vector T̂(s) is a constant.
(b) Guess the plane. To do so, first show that the binormal B̂(s) is a
constant. Then show that (r(s) − r(0)) · B̂ is a constant.
1
(c) Guess the circle. To do so, first show that rc (s) = r(s) + κ(s) N̂(s)
is a constant.
1.4.1.5. ∗. Hint. It is not necessary to compute anything.
1.4.1.6. ∗. Hint. Both parts of this question make use of the quantity
ds
dt .
(v(t) × a(t)) · da
dt
1.4.1.7. Hint. τ (t) =
|v(t) × a(t)|2
1.4.1.8. Hint. Review §1.5.
1.4.1.9. Hint. The vector perpendicular to the plane containing the
osculating circle is the binormal vector, B̂.
1.4.1.10. ∗. Hint. (a) The tangent vector of the curve is also a normal
vector for the specified plane.
(b) Review §1.5.
2 2
1.4.1.11. ∗. Hint. Remember a(t) = ddt2s (t) T̂(t) + κ(t) dsdt (t) N̂(t).
Remember also that B̂ is orthogonal to T̂ and N̂, which are in the plane
of C.
1.4.1.12. ∗. Hint. By Theorem 1.3.3, the tangential component of
2
acceleration is aT (t) = ddt2s
1.4.1.13. ∗. Hint. Use your answers to previous parts to calculate (d).
Tangential and normal components of acceleration are defined just before
Example 1.3.4.
1.4.1.14. ∗. Hint. (a) All points on the curve obey an equation that
contains x’s and y’s, but no z’s. There is a standard way to get a nice
parametrization of this equation, that doesn’t involve using square roots.
(b) You don’t need to compute the constants for all points: only the
given point.
1.4.1.15. ∗. Hint. For part (c), you only need to find N̂ at a point,
which is easier than finding it for all t.
APPENDIX B. HINTS FOR EXERCISES 343
1.4.1.26. ∗. Hint. You need to find the acceleration at (1, 1, 1). Think
about what strategies are available for computing the acceleration.
1.4.1.27. ∗. Hint. For part (a), T(t) will be a vector of the form
T(t) = (1,at,bt)
√
1+4t2
where a and b are nonzero constant real numbers.
For part (b), N(t) will be a vector of the form N(t) = (−4t,α,β)
√
2 1+4t2
where
α and β are nonzero constant real numbers.
For part (e), κ(t) will be a function of the form κ(t) = (1+4tγ2 )3/2 , where
γ is a positive constant real number.
xy
1.6.1.3. Hint. Following Definition 1.6.1, set f (x, y, z) = z , x(t) = 23 t3 ,
√
y(t) = 3t2 , and z(t) = 3t.
1.6.1.4. Hint. Parametrize the circle in the usual way.
1.6.1.5. Hint. C can be parametrized as (1+t, 2+2t, 3+2t) for 0 ≤ t ≤ 1.
d √1
1.6.1.7. Hint. Simplify! Also: dt {arcsec t} = |t| t2 −1
.
1.6.1.9. ∗. Hint. Sketch C and determine the normal vectors from the
sketch. You can use x or y as the integration variable in your integrals.
2 2
1.6.1.10. ∗. Hint. (c) How is x(t) + y(t) related to z(t)?
(d) First, sketch x(t) , y(t) .
R
xρ ds
1.6.1.11. Hint. Remember x̄ = RC , etc. The integrals you evalu-
C
ρ ds
ate should all be straightforward applications of the power rule.
1.7.4.5. Hint. The highest vertical height occurs just as the skate-
E
boarder’s speed reduces to 0, at yS = mg .
1.7.4.6. Hint. At the bottom of the culvert, all the skater’s energy is
kinetic, not potential. That is, in the equation E = 12 m|v|2 +mgy, we have
y = 0.
1.7.4.7. Hint. Equation 1.7.2 tells us the normal force exerted by the
track is W N̂, where W = mκ|v|2 + mg k̂ · N̂. Equation 1.3.3 part (c) says
d2 s ds 2
a(θ) = dθ 2 T̂ + κ dθ N̂.
d2 s
1.7.4.8. Hint. When θ = 13π/3, dθ 2 = 0, which is handy for a quicker
calculation.
APPENDIX B. HINTS FOR EXERCISES 345
1.7.4.10. Hint. There are now three forces acting on the bead: one
parallel to ̂ (exerted by gravity), one parallel to N̂ (exerted by the wire),
and one parallel to T̂ (exerted by the jet pack).
Follow the reasoning in the sliding bead section of the text, focusing on
the tangential forces.
1.7.4.11. Hint. If the snowmachine is moving at a constant speed, the
tangential component of its acceleration is zero. Part (a) is similar to
Question 1.7.4.10.
1.7.4.12. Hint. Follow the discussion in §1.7.3.
It’s fine to leave part (b) pretty messy. Your answer for part (c) involves
the root of a cubic function, but you don’t need a high degree of accuracy
to decide between the three options given.
1.8.1.5. Hint. The curve can be parametrized by r(θ) = f (θ) cos θ ı̂ı +
sin θ ̂
2 · Vector Fields
2.1 · Definitions and First Examples
2.1.1 · Exercises
APPENDIX B. HINTS FOR EXERCISES 346
2.1.1.7. Hint. If you know the speed and direction of an object, you can
find its velocity.
2.1.1.8. Hint. Set your face to be at the origin, (0, 0, 0).
If A is “inversely proportional” to B, then there exists a constant α such
that AB = α. That way when |B| goes up, |A| goes down, and vice-versa.
2.1.1.9. Hint. Start with the regions where v(x, y) · ı̂ı and v(x, y) · ̂ are
positive and negative. As you move up/down/left/right, do the vectors get
longer or shorter? More horizontal or more vertical?
2.1.1.10. Hint. v(x, y) · ı̂ı is the distance from (x, y) to the origin, while
v(x, y) · ̂ is the distance from (x, y) to the point (1, 1).
2.1.1.11. Hint. Factor x2 + xy = x(x + y) and y 2 − xy = y(x − y). Chop
the plane up into eight regions using the two coordinate axes and the lines
y = x, y = −x.
2.1.1.12. Hint. What is the geometric interpretation of each summand?
2.1.1.13. Hint. (a), (c) Intrepret the vector field geometrically.
2.1.1.14. Hint. The constant G is the same for all masses, but M differs.
The net force is the sum of three force vectors.
2.1.1.15. Hint. For part a., make a triangle with P as one of its vertices
that is similar to the triangle made by the pole, the wall, and the ground.
Its hypotenuse has length p; let its base be b and its height be h. Find a
way to translate between (b, h) and (x, y).
For part b., use your answer from part a. Start by describing a point
as itsdistance from the lower end of the pole, p. Then, consider
on a pole
dz dx dy
dt and dt , dt separately. If you’re having a hard time simplifying your
p √
answer, note x + y 2 = 3(1 − z) for any point (x, y, z) on a pole when
2
H = 1.
2.3.1.1. Hint. Carefully consider the context that lead to each of these
equations.
2.3.1.2. Hint. One of the three options will NEVER be true, for any F.
2.3.1.3. Hint. Modify ϕ, the potential for F.
2.3.1.4. Hint. a. If F + G is conservative, what has to be true?
b. What if F and G are quite similar?
c. Find a potential for F + G.
2.3.1.5. ∗. Hint. Note that the domain is D = (x, y) x > 1 .
Compare to Example 2.3.14.
2.3.1.6. Hint. A potential does exist.
d
2.3.1.7. Hint. Recall dx ln |x| = x1 .
2.3.1.8. Hint. Try the screening test, Theorem 2.3.9.
Z
x
2.3.1.9. Hint. dx can be evaluated by inspection, or
x2 + y 2 + z 2
with the substitution u = x2 + y 2 + z 2 .
2.3.1.11. Hint. For what values of the constants A and B does the
vector field F pass the screening test ∇ × F = 0?
R
2.4.2.24. ∗. Hint. Write the integral of part (c) as C
G · dr. What is
the difference between G and F?
2.4.2.25. ∗. Hint. (d) How are G and F related?
2.4.2.35. ∗. Hint. Note that the curve goes from (2, 2) to (1, 1) — not
the other way around.
For part (b), one possibility is to look for a path consisting of the line
segment from (2, 2) to (2, Y ), followed by the line segment from (2, Y ) to
(1, Y ), followed by the line segment from (1, Y ) to (1, 1), with Y being a
parameter to be determined.
2.4.2.36. ∗. Hint. One possibility is to look for a path consisting of the
line segment from (0, 0) to (0, Y ), followed by the line segment from (0, Y )
to (2, Y ), followed by the line segment from (2, Y ) to (2, 0), with Y being
a parameter to be determined.
2.4.2.37. ∗. Hint. Is F conservative?
2.4.2.38. ∗. Hint. On S, note z = 2 + x2 − 3y 2 . Further,R the vector
field F̃(x, y, z) = z 2 k̂ is conservative (with potential 13 z 3 ), so C1 F̃ · dr =
R
C2
F̃ · dr for any two curves C1 and C2 from P1 to P2 . Compare this to
Questions 2.4.2.24 through 2.4.2.25.
2.4.2.39. ∗. Hint. Simplify the answer in part (a) as much as possible.
3x2
For part (c), start with ∂f
∂y = xe and ∂f 2 2
∂z = x cos(x z).
For part (d), notice the difference between the given vector field and the
conservative vector field of part (c). The resulting integral can be directly
evaluated using methods from integral calculus.
dr
2.4.2.40. ∗. Hint. For (b), remember ds dt = dt \\ Is the vector field of
part (c) conservative?
2.4.2.41.
R ∗. Hint. For part (d), what is the difference between J and
C
F · dr?
For part (e), many parts of the integral are zero: find as many as you
can.
2.4.2.42. ∗. Hint. By Newton’s law of motion, mr00 (t) = F(t).
0 00
Recall κ(t) = |r (t)×r
|r0 (t)|3
(t)|
.
3 · Surface Integrals
3.1 · Parametrized Surfaces
3.1.1 · Exercises
APPENDIX B. HINTS FOR EXERCISES 350
3.1.1.1. Hint. Your answer will have the form r(x, y) = ψ1 (x, y)ı̂ı +
ψ2 (x, y)̂ + ψ3 (x, y)k̂.
3.1.1.3. ∗. Hint. First think about what properties r(u, v) has to have
in order to be a parametrization.
3.1.1.4. ∗. Hint. First think about what properties r has to have in
order to be a parametrization.
3.1.1.5. ∗. Hint. First think about what properties r(u, v) has to have
in order to be a parametrization.
3.2.1.17. ∗. Hint. The angle between the curve and the surface at P
is 90◦ minus the angle between the curve and the normal vector to the
surface at P .
3.2.1.18. Hint. At the highest and lowest points of the surface, the
tangent plane is horizontal.
3.3.6.8. ∗. Hint. The total surface area of (b) (ii) can be determined
without evaluating any integrals.
3.3.6.11. Hint. On S, (x, y) runs over the interior of x2 + y 2 = 2x, or
equivalently, the interior of (x − 1)2 + y 2 = 1.
3.3.6.12. Hint. See Example 3.1.5 for a parametrization of the torus.
3.3.6.13. Hint. Call the part of the sphere in the first octant S. By
definition, the centroid is (x̄, ȳ, z̄) with
RR RR RR
S
x dS S
y dS z dS
x̄ = RR ȳ = RR z̄ = RRS
S
dS S
dS S
dS
The integrals will be easy if you use spherical coordinates. You can reduce
the number of integrals evaluated by using symmetry.
3.3.6.14. Hint. Before parametrizing the cylinder, express x2 +y 2 = 2ay
in cylindrical coordinates.
3.3.6.16. Hint. (a) The integral can be easily evaluated by using that
the sphere has surface area 4πa2 .
(c) Use cylindrcial coordinates for the top part of the cone.
3.3.6.20. ∗. Hint. Beware of signs. Note that 0 ≤ z ≤ 1 on S.
3.3.6.21. ∗. Hint. The z-coordinate of the centre of mass is the weighted
average of the z-coordinate over the cone. Since a density has not been
specified, we assume that it is a constant. We RR
may takeRR
the density to be
1, so the z-coordinate of the centre of mass is S z dS/ S dS.
3.3.6.22. ∗. Hint. Use cylindrcal coordinates. Note that because of
the symmetry of the cone, only the z-component of the centre of mass
requires an integral to be calculated. The z-coordinate of the centre of
massRRis the weighted
RR average of the z-coordinate over the cone. That is
z̄ = S z dS/ S dS.
3.3.6.24. ∗. Hint. Review (3.3.2).
3.3.6.25. ∗. Hint. Don’t pbe afraid to tweak spherical coordinates so
as to fit the condition x ≥ y 2 + z 2 well.
√ To do so, first use a sketch to
y 2 +z 2
develop a geometric interpretation of x .
APPENDIX B. HINTS FOR EXERCISES 352
4 · Integral Theorems
4.1 · Gradient, Divergence and Curl
4.1.6 · Exercises
4.1.6.2. Hint. Compute ∇ × F for some simple vector fields.
4.1.6.3. Hint. For parts(a) and (b), write out the definitions of the left
and right hand sides and observe that they are equal. Part (c) can be done
easily by using other, simpler, vector identities.
4.1.6.6. ∗. Hint. (c) can be done efficiently by using (a) and (b).
4.1.6.12. Hint. (a) Find the magnitude and direction of the velocity
vector. Then verify that Ω × r has that magnitude and direction.
and
RRR RRR RRR
V
x dV V
y dV V
z dV
x̄ = ȳ = z̄ =
Volume(V) Volume(V) Volume(V)
into thin horizontal washers. See Section 1.6 in the CLP-2 text.
• Use the right hand rule to check that r0 (s) × k̂ is n̂ rather than −n̂.
4.3.1.3. Hint. Use direct evaluation!
APPENDIX B. HINTS FOR EXERCISES 355
x −y
4.3.1.4. Hint. The functions x2 +y 2 and x2 +y 2 are not defined, let alone
continuous or differentiable, at x = y = 0.
4.3.1.5. Hint. For practice, evaluate this integral twice — once directly
and once using Green’s theorem.
4.3.1.6. Hint. The sin y 2 and cos y 2 in the integrand look hard to inte-
grate. Try Green’s theorem.
4.3.1.7. ∗. Hint. Don’t do the integral directly.
4.3.1.8. ∗. Hint. Don’t do the integral directly. Sketch the rectangle.
4.3.1.9. ∗. Hint. Do not compute the integral directly.
4.3.1.10. ∗. Hint. Don’t do the integral directly. Sketch the triangle.
4.3.1.11. ∗. Hint. The integrand for direct evaluation looks complicated
— don’t evaluate this integral directly.
4.3.1.12. ∗. Hint. Direct evaluation is not the most efficient method
available.
4.3.1.13. ∗. Hint. Green’s theorem must be applied to a closed curve;
note that the curve C is not closed.
Consider carefully the pointR (0, 0) in your analysis.
dt
You may use the fact that 1+t 2 = arctan(t) + C.
S
∂S
APPENDIX B. HINTS FOR EXERCISES 356
Then imagine slowly deforming the sketch to the get specified S’s
4.4.3.2. Hint. Define the vector field F(x, y, z) = F1 (x, y)ı̂ı + F2 (x, y) ̂.
4.4.3.3. Hint. First verify the vector identity ∇ × [φ∇
∇ψ + ψ∇
∇φ] = 0
4.4.3.31. ∗. Hint. For practice, evaluate the flux of part (a) twice —
once by direct evaluation and once using Stokes’ theorem.
4.4.3.32. ∗. Hint. By definition, D is connected if any two points in D
can be joined by a curve that lies completely in D.
By definition, D is simply connected if any simple closed curve in D
can be shrunk to a point continuously in D.
4.4.3.33. ∗. Hint. Review §4.1.2.
4.4.3.34. ∗. Hint. Considering that there are ten line segments in C, it
is probably not very efficient to use direct evaluation.
4.4.3.35. ∗. Hint. Direct evaluation looks hard.
H
4.4.3.36. ∗. Hint. Rewrite C E · dr as a surface integral.
5.2.8. ∗. Hint. Read all of the statements very carefully. The details are
critical.
For part (d), note that the curve need not lie in a plane.
For part (g), note that the domain can have holes in it.
For parts (h) and (i), by definition, D is simply connected if any simply
closed curve in D can be shrunk to a point continuously in D.
5.2.9. ∗. Hint. Read all of the statements very carefully. The details are
critical.
For part (b), note that the curve need not lie in a plane.
For part (d), note that the domain can have holes in it.
For parts (i) and (j), by definition, D is simply connected if any simply
closed curve in D can be shrunk to a point continuously in D.
5.2.10. ∗. Hint. Read all of the statements very carefully. The details are
critical. R R
(a) The integral C f ds = 0 is not of the form C F · dr.
(d) F and G can be any R vector fields.
(e) Think about how C f ds is defined.
∂r
(f) Look at ∂u ∂r
× ∂u very closely.
(g) The integral is completely independent of x(u, v) and y(u, v).
5.2.11. ∗. Hint. Read all of the statements very carefully. The details are
critical.
(b) Read the statement very carefully. “simply connected” plays no role
here. The vector field F is not required to be conservative.
(e) Recall that S is closed when it is the boundary of a solid region V .
(g) Assume that the constant |v| is not zero.
(j) If you have not learned about Kepler’s three laws, skip this part.
5.2.13. ∗. Hint. (g) Be careful. The power in the denominator is important.
(j) Beware the sign.
5.2.20. ∗. Hint. Review Corollary 4.3.5.
RR
5.2.21. ∗. Hint. (b) S ∇ ×F· n̂ dS is a flux integral over the closed surface
S. H H H
(c) Consider C F · dr − C G · dr = C (F − G) · dr.
Appendix C
Answers to Exercises
1 · Curves
1.1 · Derivatives, Velocity, Etc.
1.1.1 · Exercises
p
1.1.1.1. Answer. (a) r(y) = a2 − y 2 ı̂ı + y ̂, 0 ≤ y ≤ a
(b) x(φ), y(φ) = a sin φ, −a cos φ , π2 ≤ φ ≤ π
(c) x(s), y(s) = a cos( π2 − as ), a sin( π2 − as ) , 0 ≤ s ≤ π2 a
√
1.1.1.2. Answer. (1, 25), (−1/ 2, 0), (0, 25).
1.1.1.3. Answer. The curve crosses itself at all points (0, (πn)2 ) where
n is an integer. It passes such a point twice, 2πn time units apart.
1.1.1.4. Answer. (a) (a + aθ, a)
(b)(a + aθ + a sin θ, a + a cos θ)
q
2
1.1.1.5. Answer. z = − 12 1 − y2 − y
4
r(t + h)
r(t)
r(0)
359
APPENDIX C. ANSWERS TO EXERCISES 360
1.1.1.12. Answer.
√ velocity = −a sin tı̂ı + a cos t ̂ + c k̂
speed = a2 + c2
acceleration = −a cos tı̂ı − a sin t ̂
The path is a helix with radius a and with each turn having height 2πc.
(2,0,1)
1.1.1.13. ∗. Answer. (a) T̂(1) = √
5
(b) 13 53/2 − 8
1.1.1.14. Answer. 2
√
1.1.1.15. Answer. length = a2 + b2 T
1.1.1.16. Answer. 1
1.1.1.17. ∗. Answer. (a) 20 3
π3
(b) x(t) = −2π − 2t, y(t) = −2πt, z(t) = 3 + π2 t
e−αt −1 −αt
1.1.1.31. Answer. r(t) = r0 − α v0 + g 1−αt−e
α2 k̂
1.2 · Reparametrization
1.2.1 · Exercises
1.2.1.1. Answer. t − 1
√
1.2.1.2. Answer. sin(1/2), cos(1/2), 3/2
1.2.1.3. Answer. A
√
1.2.1.4. ∗. Answer. (a) 43 , − 43 , − 21
√
1.2.1.5. ∗. Answer.
(a) 2
(b) √s2 cos ln √s2 , sin ln √s2 with s > 0
1.2.1.6. Answer. (cos z, z sin z, z) for 0 ≤ z < π/2. The curve is (the
first quarter-turn of) a spiral, with width in the x-direction 2, and increas-
ing width in the y-direction. The parameter z is the height, as well as a
radian measure for the spiral.
√
1.2.1.7. Answer. When s ≤ 13 (2 2 − 1),
1 h √ i 1h √ i3/2
R(s) = (2 2 − 3s)2/3 − 1) , − (2 2 − 3s)2/3 − 1
2 3
√
and when s > 13 (2 2 − 1),
1 h √ i 1h √ i3/2
R(s) = (3s + 2 − 2 2)2/3 − 1 , (3s + 2 − 2 2)2/3 − 1
2 3
1.3 · Curvature
1.3.1 · Exercises
1.3.1.1. Answer.
APPENDIX C. ANSWERS TO EXERCISES 362
T̂
N̂
x
1
1
ρ = 3, κ = 3
ds
√
1.3.1.4. Answer. dt = e2t + 9 + cos2 t
1.3.1.5. Answer.
dT̂ 1
= √ − sin t − cos t, − sin t + cos t
dt 2
dT̂ 1 √ √
=√ − sin ln s/ 2 − cos ln s/ 2 ),
ds 2s
√ √
− sin ln s/ 2 + cos ln s/ 2
1.3.1.7. Answer.
A v(t) = (et , 2t + 1)
B a(t) = (et , 2)
ds p 2t
C = e + (2t + 1)2
dt
!
et 2t + 1
D T̂(t) = p ,p
e2t + (2t + 1)2 e2t + (2t + 1)2
et |1 − 2t|
E κ(t) =
(e2t + (2t + 1)2 )3/2
1.3.1.8. Answer. κ(t) = √1
2
a b
1.3.1.9. Answer. κmax = b2 , κmin = a2 .
2π 4π x
(b) κ(t) = √1
23/2 1−cos t
(c) 4
(d) (x − π)2 + (y + 2)2 = 16
T̂
N̂
z = x2 + y 2
N̂ T̂
B̂
y
x=y
x
(c) The torsion is zero.
1.4.1.6. ∗. Answer. (a) r0 (t) = et + e−t ı̂ı + et − e−t ̂ + 2 k̂, r00 (t) =
ı̂ı+t ̂+t2 k̂
1.4.1.8. Answer. T̂(t) = √
1+t2 +t4
t2 ı̂ı−2t ̂+k̂
B̂(t) = √
1+4t2 +t4
3
) ı̂ı+(1−t4 ) ̂+(2t+t3 )k̂
N̂(t) = −(t+2t √ √
1+t2 +t4 1+4t2 +t4
√
1+4t +t2 4
κ(t) = [1+t 2 +t4 ]3/2
√
1.4.1.13. ∗. Answer. (a) 5 t
(b) aT (t) = sin tı̂ı + cos t ̂ + 2 k̂
(c) aN (t) = t cos tı̂ı − t sin t ̂
1
(d) κ(t) = 5t
1.4.1.14. ∗. Answer. (a) r(θ) = [−1+3 cos θ]ı̂ı +3 sin θ ̂ +[10−6 cos θ] k̂,
0 ≤ θ < 2π √
(b) At (2, 0, 4), T̂ = ̂, N̂ = −ı̂ı√+2
5
k̂ ı+k̂
, B̂ = 2ı̂√ 5
, κ(0) = 35
APPENDIX C. ANSWERS TO EXERCISES 365
√
t2 ı̂ı+ 2 t ̂+k̂
1.4.1.15. ∗. Answer. (a) T̂(t) = t2 +1
√
2
(b) (t2 +1)2
√
4 ı̂ı−3√ 2 ̂−4k̂
(c) 50
1
3/2
1.4.1.18. ∗. Answer. (a) 3 2 −1
(b) −ı̂√ı−̂
2
(c) 12
1.4.1.19. ∗. Answer. (a) v(t) = 1 , −1 , t
√
(b) ds
dt (t) = 2 + t2
(c) a(t) = 0 , 0 , 1
√
(d) κ(t) = [2+t22]3/2
(−t , t , 2)
(e) N̂(t) = √ 2 2(2+t )
(f) x + y = 3
(g) (2, 1, 2)
√
t2 ı̂ı+ 2t ̂+k̂
1.4.1.20. ∗. Answer. (a) T̂(t) = t2 +1
√
2
(b) κ(t) = (t2 +1)2
√
(c) κ(0) = 2
(d) N̂(0) = ̂
(e) B̂(0) = −ı̂ı
1.4.1.21. ∗. Answer. (a) x = 1 − 2t, y = −1 + t, z = −1 + 3t
(b) 3x − 3y − z = −1
√
5 π2
1.4.1.22. ∗. Answer. (a) 2
1
(b) κ(t) = 5t
1.4.1.23. ∗. Answer. (a) 8
(b) T̂(1) = √12 (1, 1, 0), N̂(1) = (0, 0, −1)
(c) κ(1) = 18
1.4.1.25. Answer. (a) r(θ) = cos θ ı̂ı + sin θ ̂ + cos(2θ) k̂ 0 ≤ θ < 2π
(b) 51
√ √
(c) z = 2 x − 2 y √ √
(d) radius 1/κ(π/4) = 5 and centre − 2 2 , −2 2 , 0
APPENDIX C. ANSWERS TO EXERCISES 366
4
1.4.1.26. ∗. Answer. ı
9 (ı̂ − 4 ̂ + k̂)
√
ı̂ı+t̂
√+ 3tk̂
1.4.1.27. ∗. Answer. (a) T̂(t) = 1+4t2
√
(b) N̂(t) = −4t2√ı̂ı+̂+ 3k̂
1+4t2
(c) \textcircled{3}
√
(d) − 3y + z = 0
−3/2
(e) κ(t) = (1 + 4t2 )
(f) The curvature κ(t) achieves its maximum value at r(0) = (0, 0, 0).
(g) The curvature√never achieves √
a minimum.
(h) ı̂ı = u2 , ̂ = v− 4 3 w , k̂ = 3 v+w
4 , r(t) = t u + t2 v
y
y = x2
v
C
u x
The curve a(t), b(t) = (t, t2 ) is the curve y = x2 . It is “curviest” at
the origin, which is consistent with part (f). It becomes flatter and flatter
as |t| increases, but never achieves “perfect flatness”, which is consistent
with (g).
1.4.1.28. ∗. Answer. See the solution.
√ √
1.4.1.29. ∗. Answer. (a) T̂ = √16 0, 2, − 2 , N̂ = − √139 6, 1, 2 ,
√ √ √
B̂ = √113 − 1, 2, 2 2 , κ = 3√133 = 939
√
dv
(b) (i) = 5√32
dt
√
(ii) v = (0, 2, −1).
1.4.1.30. ∗. Answer. (a) v(t) = − sin t , cos t , c cos t , a(t) = −
cos t , − sin t , −c sin t
√
(b) v(t) = 1 + c2 cos2 t
2
(c) −c
√ sin t cos t
1+c2 cos2 t
(d) The curve lies on the plane z = cy.
√
17
1.4.1.31. ∗. Answer. (a) 4
(b) √417
√
(c) (i) 4 π
(ii) 16π , −4 , −4π
√
(iii) 4 17 π
4√ 2
1.6.1.3. Answer. 21 3
(27 − 1) + √
5 3
(25 − 1)
1.6.1.4. Answer. π kg
1.6.1.5. Answer. 26
53/2 −1
1.6.1.6. Answer. (a) 12
3/2
8−3
(b) 3/2
1
1.6.1.7. Answer. 2 ln 2
t2
1.6.1.8. ∗. Answer. (a) r(t) = tı̂ı + 1 + 2 ̂ + sin t k̂
2
(b) r(π/2) = π2 ı̂ı + 1 + π8 ̂ + k̂
π2 1
(c) 8 − 2
x
y
412 92 4736
1.6.1.11. Answer. 55 , − 55 , 693
W N̂
−mĝ
1.7.4.10. Answer. U = mg dy
ds
E
1.7.4.12. Answer. (a) yS =
mg
yA −3
24 (E − mgyA ) 3
(b) = 4mg q (or equivalent)
3/2
yA −3 2 yA −3 2
9+7 3
9 + 7 3
(c) The skateboarder makes it up to the ceiling, but falls off rather than
making it all the way around. Ouch.
1.7.4.13. Answer. (a), (b) See the solution.
h 2 2 i1/2
(c) 2 a gb
+b
π
er ( π2 )
eθ ( π2 ) eθ (0)
(0, 1)
eθ (π)
and here is a sketch (to a different scale) of (xi , yi ), êr (θi ), êθ (θi ) for
i = 2, 4 (the points off the axes).
er ( 3π ) y eθ ( π ) er ( π4 )
4 4
(−1, 1) (1, 1)
3π
eθ ( 3π
4
) 4 π
4
x
1.8.1.4. ∗. Answer. (a) ↔ (E)
(b) ↔ (B)
(c) ↔ (F)
(d) ↔ (C)
(e) ↔ (A)
(f) ↔ (D)
f (θ)2 +2f 0 (θ)2 −f (θ)f 00 (θ)
1.8.1.5. Answer. κ(θ) = [f (θ)2 +f 0 (θ)2 ]3/2
2 · Vector Fields
2.1 · Definitions and First Examples
2.1.1 · Exercises
2.1.1.1. Answer.
> 0
when x > 0
v(x, y) · ı̂ı = 0 when x = 0
<0 when x < 0
and
> 0
when −2 < x < 2
v(x, y) · ̂ = 0 when x ∈ {−2, 2}
< 0 when x < −2 or x > 2
at least for (x, y) shown in the sketch.
2.1.1.2. Answer.
> 0
when y > −x
v(x, y) · ı̂ı = 0 when y = −x
< 0 when y < −x
and
> 0
when y < x
v(x, y) · ̂ = 0 when y = x
< 0 when y > x
at least for (x, y) shown in the sketch.
2.1.1.8. Answer. If your face is at the origin, then v(x, y, z) = − x2 +yα2 +z2 (x, y, z)
for some positive constant α.
2.1.1.9. Answer.
APPENDIX C. ANSWERS TO EXERCISES 371
2.1.1.10. Answer.
y
2.1.1.11. Answer.
APPENDIX C. ANSWERS TO EXERCISES 372
2.1.1.12. Answer.
y
(b)
APPENDIX C. ANSWERS TO EXERCISES 373
(c)
y
a. v(p) = 1 − p2 2√ p
1
2.1.1.15. Answer. 3
, − 4
b. V(x, y, z) = − x6 , − y6 , z2 or equivalent
x
1 2 3
2.2.2.2. Answer. v(x, y) = (−x − y , x − y)
APPENDIX C. ANSWERS TO EXERCISES 374
x2 y2
2.2.2.3. ∗. Answer. (a) 2 = 2 +C
(b)
2.2.2.4. ∗. Answer. x = y 2 , z = ey
2.2.2.5. ∗. Answer. The field lines are y = C 0 x3 with C 0 a nonzero
constant, as well as x = 0 and y = 0.
2.3.1.6. Answer. ϕ = 12 x2 + xy − 12 y 2
x
2.3.1.7. Answer. ϕ = ln |x| − y
2.4.2.2. Answer. a. A
b. B
c. A
d. B
2.4.2.3. Answer. 0
2.4.2.4. Answer. 5
2.4.2.5. ∗. Answer. a = 1, b = c = 0
2.4.2.6. Answer. (a) Not conservative
(b) Not conservative
(c) Not conservative
(d) Conservative
2.4.2.7. ∗. Answer. (a) The (largest possible) domain is D = (x, y, z) x2 +
y 2 6= 0 .
(b) ∇ × F = 0 on D
APPENDIX C. ANSWERS TO EXERCISES 376
R
(c) C F · dr = 4π
(d) F is not conservative.
2.4.2.8. Answer. 9 12 for all paths from (1, 0, −1) to (0, −2, 3)
π2
2.4.2.9. Answer. 2(e − 1) + 2 + 3π
2.4.2.16. ∗. Answer. − 32
2.4.2.17. ∗. Answer.
R The line integral is independent of path because
it is of the form C F · dr with F being a conservative field. The value of
the integral is 1 + π2 .
1
2.4.2.18. ∗. Answer. 2
R ∗. Answer.
2.4.2.22. (a) ∇ × F = 0. F is conservative.
(b) C F · dr = 2π 2
2.4.2.23. ∗. Answer. (a) a = −1, b = 3
(b) f (x, y, z) = xyex + yz 3 + C works for any constant C
(c) πeπ − 2
(d) πeπ − 32 15
2
(b) 2eπ − e−π − 1
2.4.2.27. ∗. Answer. (a) 0.
(b) F is conservative with potential ϕ(x, y, z) = x2 + y 2 + z 2 . So the
integral is ϕ(a1 , a2 , a3 ) − ϕ(0, 0, 0) = a · a.
2.4.2.28. ∗. Answer. (a) ∇ × F = 0
(b) πe
2 −1
2
3/2
2.4.2.31. Answer. (a) 3 14 − 1 ≈ 34.26
(b) sin 1 + 32 ≈ 2.3415
1
2.4.2.32. ∗. Answer. 2π + 3
3 · Surface Integrals
3.1 · Parametrized Surfaces
3.1.1 · Exercises
3.1.1.1. Answer. r(x, y) = xı̂ı + ŷ + (ex+1 + xy)k̂
3.1.1.2. ∗. Answer. parabolic bowl
3.2.1.6. ∗. Answer. 2x + y + 9z = 2
3.2.1.7. ∗. Answer. 2x + y + z = 6
3.2.1.8. ∗. Answer. z = − 43 x − 32 y + 11
4
πah
3.3.6.3. Answer. 2
116
3.3.6.4. Answer. 15 π
π 3/2
3.3.6.5. ∗. Answer. (1 + 4a2 ) − 1
6
√
3.3.6.6. ∗. Answer. 5 2π
4
√ √
3.3.6.7. ∗. Answer. 15 9 3−8 2+1
p
3.3.6.8. ∗. Answer. (a) F (x, y) = 1 + fx (x, y)2 + fy (x, y)2
R 2π R 1 2r
(b) (i) 0 dθ 0 dr √4−r 2
3.3.6.20. ∗. Answer. 9π
3.3.6.21. ∗. Answer. (a)
2 2
r(θ, z) = (3 − z) cos θ ı̂ı + (3 − z) sin θ ̂ + z k̂
3 3
0 ≤ θ < 2π, 0 ≤ z ≤ 3
(b) 1
0, 0, a3
3.3.6.22. ∗. Answer.
3.3.6.23. ∗. Answer. 2π
3.3.6.24. ∗. Answer. − 14
3
√
2π
3.3.6.25. ∗. Answer. 4
16
3.3.6.26. ∗. Answer. 3 π
0, 0, 32
3.3.6.27. ∗. Answer.
APPENDIX C. ANSWERS TO EXERCISES 381
3.3.6.28. ∗. Answer. 4
81
3.3.6.29. ∗. Answer. 16
x2 +z 2 = ey
x
h i
2π
(b) 3 (1 + e2 )3/2 − 23/2
(c) π 1 − e2
3.3.6.31. ∗. Answer. 12π
√
5 π
3.3.6.32. ∗. Answer. 8
3.3.6.39. ∗. Answer. (a) Yes. See the solution for the explanation.
(b) See the solution for the proof.
3.3.6.40. ∗. Answer. (a) (i) r(u, v) = u , v , 13 (16−2u−4v) k̂ u≥
0, v ≥ 0, u + 2v ≤ 8
(a) (ii) r(u, v) = 4 cos u sin v , 4 sin u sin v , 4 cos v 0 ≤ u ≤ 2π, 0 ≤
v ≤ π4
√
(a) (iii) r(u, v) = u , v , 1 + u2 + v 2 u2 + v 2 ≤ 99
√ √
or r(u, v) = u cos v , u sin v , 1 + u2 0 ≤ v ≤ 2π, 0 ≤ u ≤ 99
h i
(b) 32π 1 − √1
2
π
3.3.6.41. ∗. Answer. (a) 2
(b) π4 + 23
3.3.6.42. ∗. Answer. − 65
4 · Integral Theorems
4.1 · Gradient, Divergence and Curl
4.1.6 · Exercises
APPENDIX C. ANSWERS TO EXERCISES 382
4.1.6.10. ∗. Answer. (a) D = (x, y, z) x2 + z 2 6= 0
(b) ∇ × F = 0 on D
(c) ∇ · F = 1 on D
(d) F is not conservative on the domain D of part (a).
4.1.6.11. ∗. Answer. (a) α = β = −1
(b) Any function of the form g(x, y, z) = xyz + w(z) will work.
4.1.6.12. Answer. (a) See the solution
(b) ∇ × (Ω
Ω × r) = 2Ω
Ω ∇ · (Ω
Ω × r) = 0
(c) 1095km/hr
4.1.6.13. Answer. See the solution.
8π
4.2.6.3. Answer. (a), (b) 3
(b) 16π
(c) 16π, again
4.2.6.14. ∗. Answer. 40π
4.2.6.15. ∗. Answer. 24π
π
4.2.6.16. ∗. Answer. 2
3
4.2.6.17. ∗. Answer. 2π
32
4.2.6.18. ∗. Answer. 3 π
4.2.6.19. ∗. Answer. 3π
26
4.2.6.20. ∗. Answer. 3
4.2.6.21. ∗. Answer. 2π
64
4.2.6.22. ∗. Answer. (a) 3 π
(b) 128
3 π
4.2.6.28. ∗. Answer. 4π
4.2.6.29. ∗. Answer. π
4.2.6.30. ∗. Answer. 12π
188
4.2.6.31. ∗. Answer. 15 π ≈ 39.37
4.2.6.32. ∗. Answer. 5π
π 1 3
4.2.6.33. Answer. 6 − 3 a
√
4.2.6.34.√ Answer. (a) −8 2π
(b) 8 √2π
(c) 16 2π
4.3.1.8. ∗. Answer. −6
4.3.1.9. ∗. Answer. (a)
y y = x2 + 4x + 4
C y = 4 − x2
C
(−2, 0) x
(b) − 83
4.3.1.10. ∗. Answer. − 31
4.3.1.11. ∗. Answer. 54
10
4.3.1.12. ∗. Answer. 3
π
4.3.1.17. ∗. Answer. (a) 2
(b) π4 + 23
3π
4.3.1.18. ∗. Answer. 2
3π
4.3.1.19. ∗. Answer. 8
4.3.1.20. ∗. Answer. A = −2
4.3.1.21. ∗. Answer. −π
H H
4.3.1.22. ∗. Answer. C1 F · dr = 0 and C2 F · dr = 2π
z
(0,0,2)
C2
C3
(0,2,0)
(2,0,0) y
x C1
(b) S = (x, y, z) x2 + y 2 + z 2 = 4, x ≥ 0, y ≥ 0, z ≥ 0 with
4.4.3.19. ∗. Answer. −2
4.4.3.20. ∗. Answer. −π
3π
4.4.3.21. ∗. Answer. 4
π
4.4.3.22. ∗. Answer. 3
Sy
Sx y
(0, 3, 0)
(2, 0, 0)
x
R
(b) C
F · dr = 10
4.4.3.24. ∗. Answer. −π
4.4.3.25. ∗. Answer. 24π
√
4.4.3.26. ∗. Answer. 2 3πR2
APPENDIX C. ANSWERS TO EXERCISES 388
4
4.4.3.27. Answer. 3
(h) False
(i) False
(j) True
5.2.3. ∗. Answer. (a) False.
(b) N̂(t), B̂(t)
(c) True.
(d) False.
(e) False.
(f) True.
5.2.4. ∗. Answer. (a) decreasing
(b) f (x) is D
(c) r(θ) = cos θ ı̂ı + sin θ k̂ + sin θ cos θ ̂, 0 ≤ θ < 2π
(d) We want parametrisation (d) with domain |u| ≥ 2, 0 ≤ v ≤ 5.
(e) One possible answer is r(t) = tı̂ı, 0 ≤ t ≤ 1.
(f) C=6
(g) (a, b, c, d) a, b, c, d all real and b = c
(h) 2
(i) (1) True
(2) False
(3) False
(4) False
(5) False
(j) Any vector field whose divergence is 1 everywhere will work. One such
vector field is F = xı̂ı.
(k) negative
5.2.5. ∗. Answer. (a) false
(b) false
(c) true
(d) false
(e) true, assuming that the second derivatives of the vector field exist and
are continuous.
(f) silly, but true
(g) true
(h) false
(i) false
(j) false
5.2.6. ∗. Answer. (a) False
(b) False
(c) True
(d) True
(e) True
(f) True
(g) True
(h) False
5.2.7. ∗. Answer. (a) Py < 0
(b) Qx > 0
(c) ∇
R × F is in the direction of +k̂ at A
(d) C1 F · dr > 0
R
(e) C2 F · dr < 0
(f) F is not conservative
APPENDIX C. ANSWERS TO EXERCISES 390
Solutions to Exercises
1 · Curves
1.1 · Derivatives, Velocity, Etc.
1.1.1 · Exercises
1.1.1.1. Solution. (a) Since, on the specified part of thep
p circle, x =
a2 − y 2 and y runs from 0 to a, the parametrization is r(y) = a2 − y 2 ı̂ı+
y ̂, 0 ≤ y ≤ a.
(b) Let θ be the angle between
• the radius vector from the origin to the point (a cos θ, a sin θ) on the
circle and
• the positive x-axis.
The tangent line to the circle at (a cos θ, a sin θ) is perpendicular to the
radius vector and so makes angle φ = π2 + θ with the positive x axis. (See
the figure on the left below.) As θ = φ − π2 , the desired parametrization is
x2 + y 2 = a2
392
APPENDIX D. SOLUTIONS TO EXERCISES 393
1.1.1.2. Solution. We can find the time at which the curve hits a given
point by considering the two equations that arise from the two coordinates.
For the y-coordinate
√ to be 0, we must have (t − 5)2 = 0, i.e. t = 5. So, the
point (−1/ 2, 0) happens when t = 5.
Similarly, for the y-coordinate to be 25, we need (t − 5)2 = 25, so
(t − 5) = ±5. When t = 0, the curve hits (1, 25); when t = 10, the curve
hits (0, 25). √
So, in order, the curve passes through the points (1, 25), (−1/ 2, 0),
and (0, 25).
1.1.1.3. Solution. The curve “crosses itself” when the same coordinates
occur for different values of t, say t1 and t2 . So, we want to know when
sin t1 = sin t2 and also t21 = t22 . Since t1 and t2 should be different, the
second equation tells us t2 = −t1 . Then the first equation tells us sin t1 =
sin t2 = sin(−t1 ) = − sin t1 . That is, sin t1 = − sin t1 , so sin t1 = 0. That
happens whenever t1 = πn for an integer n.
So, the points at which the curve crosses itself are those points (0, (πn)2 )
where n is an integer. It passes such a point at times t = πn and t = −πn.
So, the curve hits this point 2πn time units apart.
1.1.1.4. Solution. (a) Pretend that the circle is a spool of thread. As
the circle rolls, it dispenses the thread along the ground. When the circle
rolls θ radians, it dispenses the arc length θa of thread and the circle
advances a distance θa. So the centre of the circle has moved θa units to
the right from its starting point, x = a. The centre of the circle always
has y-coordinate a. So, after rolling θ radians, the centre of the circle is at
position c(θ) = (a + aθ, a).
(b) Now, let’s consider the position of P on the circle, after the circle
has rolled θ radians.
a sin θ P
a cos θ
From the diagram, we see that P is a cos θ units above the centre of
the circle, and a sin θ units to the right of it. So, the position of P is
(a + aθ + a sin θ, a + a cos θ).
Remark: this type of curve is known as a cycloid.
1.1.1.5. Solution. We aren’t concerned with x, so we can eliminate it
by solving for it in one equation, and plugging that into the other. Since
APPENDIX D. SOLUTIONS TO EXERCISES 394
1.1.1.7. Solution.
r(t + h)
r(t)
r(0)
and hence
√ p
|v(t)| = |r0 (t)| = 5 2ı̂ı + e5t ̂ + e−5t k̂ = 5 2 + e10t + e−10t
2
Since 2 + e10t + e−10t = e5t + e−5t , that’s (d).
1.1.1.12. Solution. We are told that
So, by definition,
As t runs over an interval of length 2π, (x, y) traces out a circle of radius
a and z increases by 2πc. The path is a helix with radius a and with each
turn having height 2πc.
1.1.1.13. ∗. Solution. (a) Since r0 (t) = (2t, 0, t2 ), the specified unit
APPENDIX D. SOLUTIONS TO EXERCISES 396
tangent at t = 1 is
(2, 0, 1)
T̂(1) = √
5
(b) We are √ √ 3, −1/3). As
to find the arc length between (0, 3, 0) and (1,
ds
dt = |r0 (t)| = 4t2 + t4 , that arc length is the integral of 4t2 + t4 with
t running between the value of t for which r(t) = (0, 3, 0) and the value of
t for which r(t) = (1, 3, −1/3). To find those two values of t, we observe
that
• the first component of r(t), namely t2 , matches the first component
of (0, 3, 0), namely 0, only when t = 0. So we would guess that the t
corresponding to (0, 3, 0) is t = 0. As r(0) = (02 , 3, 31 03 ) = (0, 3, 0),
this is indeed the case.
we have
ds p p
(t) = x0 (t)2 + y 0 (t)2 + z 0 (t)2 = a2 + b2
dt
APPENDIX D. SOLUTIONS TO EXERCISES 397
ds ds
√
As the speed dt (t) is constant, the length is just dt T = a2 + b2 T .
1.1.1.16. Solution. Since r(t) is the position of the particle, its acceler-
ation is r00 (t).
ds
(t) = r0 (t) = 2 cos t − 2t sin t , 2 sin t + 2t cos t , t2
dt q 2 2
= 2 cos t − 2t sin t + 2 sin t + 2t cos t + t4
p
= 4 + 4t2 + t4
= 2 + t2
π3
(b) A tangent vector to the curve at r(π) = − 2π , 0 , 3 is
x(t) = −2π − 2t
y(t) = −2πt
π3
z(t) = + π2 t
3
1.1.1.18. ∗. Solution. (a) As r(t) = 3 cos t, 3 sin t, 4t , the velocity of
the particle is
r0 (t) = − 3 sin t, 3 cos t, 4
ds
(b) As dt , the rate of change of arc length per unit time, is
ds
(t) = |r0 (t)| = − 3 sin t, 3 cos t, 4 = 5
dt
the arclength of its path between t = 1 and t = 2 is
Z 2 Z 2
ds
dt (t) = dt 5 = 5
1 dt 1
(b) As
ds p 0 2
= x (θ) + y 0 (θ)2 + z 0 (θ)2
dθ p
= 9 sin2 θ + 9 cos2 θ + 36 sin2 θ + 81 cos2 θ − 108 sin θ cos θ
p
= 45 + 45 cos2 θ − 108 sin θ cos θ
the circumference is
Z 2π p
s= 45 + 45 cos2 θ − 108 sin θ cos θ dθ
0
since the integrand is invariant under t → π−t. So the arc length from
√ t = 0
2
to t = π is just twice the arc length from part (a), namely 27 10 10 − 1 .
1.1.1.21. ∗. Solution. Since
t3 t2 t
r(t) = ı̂ı + ̂ + k̂
3 2 2
0 2 1
r (t) = t ı̂ı + t ̂ + k̂
2
APPENDIX D. SOLUTIONS TO EXERCISES 399
r r
ds 1 1 2 1
(t) = |r0 (t)| = t4 + t2 + = t2 + = t2 +
dt 4 2 2
the length of the curve is
Z t Z t
ds 1 t3 t
s(t) = (u) du = u2 + du = +
0 dt 0 2 3 2
1.1.1.22. ∗. Solution. Since
(c) Denote by
• r(x) the position of the particle when its first coordinate is x,
• R(t) the position of the particle at time t,
• x(t) the x--coordinate of the particle at time t, and
APPENDIX D. SOLUTIONS TO EXERCISES 400
• s(x) the arc length of the curve from the origin to r(x).
We are told that |R0 (t)| = 9 for all t. So
dx
=⇒ R0 (t) = r0 x(t)
R(t) = r x(t) (t)
dt !
0 ds dx 1 p dx
=⇒ 9 = |R (t)| = x(t) (t) = p + x(t) (t)
dx dt 2 x(t) dt
√
1 dx dx
9= √ + 1 (0) =⇒ (0) = 6
2 1 dt dt
so that
0 dx
0 1
R (0) = r (1) (0) = ı̂ı + ̂ + k̂ 6 = 6ı̂ı + 3 ̂ + 6 k̂
dt 2
!−1
dx 1 p
(t) = 9 p + x(t)
dt 2 x(t)
dx
In particular, when t = 0, x(0) = 1 and dt (0) =6
−2
d2 x 3 1
(0) = −9 6 = −6
dt2 2 4
so
00 00
2 0
1 1 1
R (0) = r (1) 6 + r (1) − 6 = 36 − ̂ + k̂ − 6 ı̂ı + ̂ + k̂
4 2 2
= −6ı̂ı − 12 ̂ + 12 k̂
1.1.1.24. Solution. Given the position of the particle, we can find its
velocity:
v(t) = r0 (t) = (cos t, − sin t, 1)
Applying the given formula,
L(t) = r × v = (sin t, cos t, t) × (cos t, − sin t, 1).
• [Solution 1:] We can first compute the cross product, then differ-
entiate:
L(t) = (cos t + t sin t)ı̂ı + (t cos t − sin t)̂ − k̂
APPENDIX D. SOLUTIONS TO EXERCISES 401
z2 yz
1.1.1.25. ∗. Solution. (a) Since z = 6u, y = 12 = 3u2 and x = 18 = u3 ,
(b)
du du
6ı̂ı + 12 ̂ + 12 k̂ = R0 (t1 ) = r0 (1) (t1 ) = 3ı̂ı + 6 ̂ + 6 k̂ (t1 )
dt dt
which implies that du
dt (t1 ) = 2.
(d) By the product and chain rules,
du du 2 d2 u
R0 (t) = r0 u(t) =⇒ R00 (t) = r00 u(t) + r0 u(t)
dt dt dt2
In particular,
du 2 d2 u
27ı̂ı + 30 ̂ + 6 k̂ = R00 (t1 ) = r00 (1) (t1 ) + r0 1 (t1 )
dt dt2
d2 u
= 6ı̂ı + 6 ̂ 22 + 3ı̂ı + 6 ̂ + 6 k̂
(t1 )
dt2
Simplifying
d2 u d2 u
3ı̂ı + 6 ̂ + 6 k̂ = 3ı̂ı + 6 ̂ + 6 k̂ 2
(t1 ) =⇒ (t1 ) = 1
dt dt2
APPENDIX D. SOLUTIONS TO EXERCISES 402
mr00 (t) = F(t) so that r00 (t) = −3tı̂ı + sin t ̂ + 2e2t k̂
t2
r0 (t) = −3 ı̂ı − cos t ̂ + e2t k̂ + c
2
π2
for some constant vector c. We are told that r0 (0) = v0 = 2 ı̂ı. This forces
2
c = π2 ı̂ı + ̂ − k̂ so that
π2 3t2
r0 (t) = ı̂ı + (1 − cos t) ̂ + e2t − 1 k̂
−
2 2
2
r(x) = xı̂ı + x2 ̂ + x3 k̂
3
0
r (x) = ı̂ı + 2x ̂ + 2x2 k̂
r00 (x) = 2 ̂ + 4x k̂
ds p
= |r0 (x)| = 1 + 4x2 + 4x4 = 1 + 2x2
dx
and
3 3
2 i3
Z Z Z
ds h
1 + 2x2 dx = x + x3 = 21
ds = dx =
C 0 dx 0 3 0
7
(b) The particle travelled a distance of 21 units in 2 time units. This
21
corresponds to a speed of 7/2 = 6.
APPENDIX D. SOLUTIONS TO EXERCISES 403
ds ds dx dx dx 6
6= = = (1 + 2x2 ) =⇒ =
dt dx dt dt dt 1 + 2x2
dx
R 0 = r0 1
= ı̂ı + 2 ̂ + 2 k̂ 2 = 2ı̂ı + 4 ̂ + 4 k̂
dt
(d) By the product and chain rules,
dx dx 2 d2 x
R0 (t) = r0 x(t) =⇒ R00 (t) = r00 x(t) + r0 x(t)
dt dt dt2
d
to 6 = 1 + 2x(t)2 dx
Applying dt dt (t) gives
dx 2 d2 x
0 = 4x + (1 + 2x2 )
dt dt2
2 2
In particular, when x = 1 and dx
dt = 2, 0 = 4 × 1 2 + (3) ddt2x gives
d2 x 16
dt2 = − 3 and
2 16 8
R00 = 2 ̂ + 4 k̂ 2 − ı̂ı + 2 ̂ + 2 k̂
= − 2ı̂ı + ̂ − 2 k̂
3 3
1.1.1.28. Solution. The question is already set up as an xy-plane, with
the camera at the origin, so the vector in the direction the camera is point-
ing is (x(t), y(t)). Let θ be the angle the camera makes with the positive
x-axis (due east). The camera, the object, and the due-east direction (pos-
itive x-axis) make a right triangle.
y
y(t) object
θ
x
camera
x(t)
y
tan θ =
x
Differentiating implicitly with respect to t:
dθ xy 0 − yx0
sec2 θ =
dt x2
0 !2
xy − yx0 xy 0 − yx0
dθ x
= cos2 θ =
x2 x2
p
dt x2 + y 2
APPENDIX D. SOLUTIONS TO EXERCISES 404
xy 0 − yx0
=
x2 + y 2
1.1.1.29. Solution. Using the Theorem of Pappus, we can calculate the
surface area and volume of a pipe with the same length and radius as this
pipe. So, we need to find the length of the pipe, L.
dr √
= ( 2t, t, 1)
dt p
dr
= 2t + t2 + 1 = |t + 1|
dt
Z 10
L= (t + 1)dt = 60
0
A pipe with radius 3 and length 60 has surface area 60(2π · 3) = 360π
and volume 60(π · 32 ) = 540π.
1.1.1.30. Solution. In general a helix can be parametrized by
Our first task is to determine a and b. The radius of the helix is 3 cm,
so a = 3 cm. After 10 turns (i.e. θ = 20π) the height, bθ, is 1 cm. So
1 1
b(20π) = 1 and b = 20π cm/rad. Thus r(θ) = 3 cos θ ı̂ı + 3 sin θ ̂ + 20π θk̂.
With each full turn of the helix (i.e. each increase of θ by 2π) the height
1
of the helix increases by 2πb = 10 cm. So if we can determine the length of
wire in one full turn of the helix, we can easily determine how many turns
the helix goes through in total, and from that we can determine the total
height of the helix.
q
As r0 (θ) = −3 sin θ ı̂ı+3 cos θ ̂+ 1 k̂ we have ds = r0 (θ) = 9 + 1 2 .
20π dθ 400π
So the length of one full turn of the helix is
Z 2π r r
1 1
9+ 2
dθ = 2π 9 + 2
0 400π 400π
1
turns. Each turn adds 10 cm to the height, so the total height is
500 1 50
q · = q ≈ 5.3 cm
π 9+ 1 10 1
π 9 + 400π
400π 2 2
du dr d2 r
(t) = αeαt (t) + eαt 2 (t)
dt dt dt
αt dr dr
= αe (t) − ge k̂ − αeαt (t)
αt
dt dt
= −geαt k̂
eαT − 1
u(T ) − u(0) = −g k̂
α
so that
eαT − 1 dr eαT − 1 eαT − 1
u(T ) = u(0) − g k̂ = (0) − g k̂ = v0 − g k̂
α dt α α
−αT
Substituting in u(T ) = eαt dr
dt (T ) and multiplying through by e gives
dr 1 − e−αT
(T ) = e−αT v0 − g k̂
dt α
Integrating both sides of this equation from T = 0 to T = t gives
e−αt − 1 t e−αt − 1
r(t) − r(0) = v0 − g k̂ + g k̂
−α α −α2
so that
e−αt − 1 1 − αt − e−αt
r(t) = r0 − v0 + g k̂
α α2
1.2 · Reparametrization
1.2.1 · Exercises
1.2.1.1. Solution. By Lemma 1.1.4.c, |r0 (s)| = 1 under arclength parametriza-
Rt Rt
tion. So 1 |r0 (s)|ds = 1 ds = t − 1.
1.2.1.2. Solution. The arclength from P to P will √ be 0, so P is the
point where s = 0. That is, r(0), or sin(1/2), cos(1/2), 3/2 .
1.2.1.3. Solution 1. We consider the situation geometrically. If we plot
R in space (of the relevant dimension), regardless of its parametrization,
the derivative at a point will give a vector tangent to R, in the direction
the curve moves when the parameter is increasing. Since a(t0 ) and b(s0 )
describe the same spot on the curve, a0 (t0 ) and b0 (s0 ) will be parallel1
— they’re both tangent to the same piece of curve. Furthermore, as t
increases, so does s, so the direction of increasing t is the same as the
direction of increasing s. Therefore, A. holds.
tangent direction
a(t0 ) = b(s0 )
R
APPENDIX D. SOLUTIONS TO EXERCISES 406
db d da dt
= [a(t(s))] =
ds ds dt ds
So, the vectors db da dt
ds and dt differ only by the scalar function ds . So, at any
point along the curve, these vectors are parallel.
Furthermore, we know that t and s are positively correlated: as t in-
dt
creases, so does s, because we’re covering more arclength. So, ds is nonneg-
dt
ative. Furthermore, since the derivatives are nonzero, ds is nonzero. So,
b0 (s0 ) and a0 (t0 ) are positive scalar multiples of each other. That is, they
dt
are parallel, and pointing in the same direction. However, unless ds =1
(that is, t(s) = s + C for some constant C), the vectors do not have the
same magnitude, and hence are not equal.
So, A is the best solution.
ds
q
0
= |r (t)| = 3 sin2 t sin2 (2t) + cos2 t sin2 (2t) + cos2 (2t)
dt q
= 3 sin2 (2t) + cos2 (2t)
=3
−∞ < t ≤ 0. As
the speed
ds 0 p √
= r (t) = et (cos t − sin t)2 + (sin t + cos t)2 = 2et
dt
and the arclength from t = −∞ to r(t) is
Z t
ds
Z t √ √
s(t) = (t̃) dt̃ = 2et̃ dt̃ = 2et
−∞ dt −∞
In particular
√ the length of the part of the spiral contained in the unit circle
is s(0) = 2.
√
(b) The inverse function of s(t) = 2et is t(s) = ln √s2 with s > 0.
(As t → −∞, the arc length s → 0 and as t → +∞, the arc length
s → +∞.) So the reparametrization is
s
t
s s
R(s) = e (cos t, sin t) =√ cos ln √ , sin ln √
t=ln √s 2 2 2
2
with s > 0
Since 0 ≤ t, and arctan t < π/2 we have 0 ≤ z < π/2, so cos z and sin z are
both nonnegative.
= (cos z, z sin z, z)
If we didn’t have the restricted domain, this would make a spiral going up:
z is both the height of the spiral and a radian measure. The ı̂ı-component
of the spiral stays between −1 and 1, while the ̂-component increases. So,
our spiral gets increasingly “wide,” while staying the same “thickness.”
1.2.1.7. Solution.
r(t) = ( 21 t2 , 13 t3 )
r0 (t) = (t, t2 )
p p
|r0 (t)| = t2 + t4 = |t| 1 + t2
Z t p
s(t) = |x| 1 + x2 dx
−1
(R t √
−x 1 + x2 dx when t ≤ 0
= R−1 0 √ Rt √
−1
−x 1 + x2 dx + 0 x 1 + x2 dx when t > 0
Let u = 1 + x2 , 12 du = xdx
( R 1+t2 √
1
− 2 udu when t ≤ 0
= R 1 1 √2 R 1+t2 1 √
− 2 2 udu + 1 2 udu when t > 0
2
(
− 13 u3/2 |1+t
2 when t ≤ 0
= 1 3/2 1 1 3/2 1+t2
− 3 u |2 + 3 u |1 when t > 0
( 3/2
2 1 2 3/2
= 3 − 3 (1 + t ) when t ≤ 0
2 23/2 1 2 3/2
− 3 + 3 + 3 (1 + t ) when t > 0
1.3 · Curvature
APPENDIX D. SOLUTIONS TO EXERCISES 409
1.3.1 · Exercises
1.3.1.1. Solution. The curve is a circle of radius 3, centred at the
origin. So, the “circle of best fit” is just the curve itself. T̂ is the unit
vector tangent to the circle in direction of increasing t, and N̂ is the unit
vector pointing towards the origin.
y
T̂
N̂
x
1
1
The radius of the (osculating) circle is 3, so ρ = 3 and κ = ρ = 13 .
dT̂
(s) = κ(s) N̂(s)
ds
1 1 1
− sin(s/3), − cos(s/3) = N̂(s)
3 3 3
(− sin(s/3), − cos(s/3)) = N̂(s)
dT̂ ds
= κ N̂(t)
dt dt
1
(− sin t, − cos t) = (3)N̂(t)
3
(− sin t, − cos t) = N̂(t)
1.3.1.3. Solution. As t increases, the arms of the spiral “flatten out,”
looking like a circle of bigger and bigger radius. So, we would expect the
APPENDIX D. SOLUTIONS TO EXERCISES 410
√
1.3.1.4. Solution. ds
dt = |v(t)| = |r0 (t)| = |(et , 3, cos t)| = e2t + 9 + cos2 t
1.3.1.5. Solution.
v(t) r0 (t)
T̂(t) = = 0
|v(t)| |r (t)|
et (cos t − sin t), et (cos t + sin t)
=p
e2t (cos t − sin t)2 + e2t (cos t + sin t)2
1
= √ cos t − sin t, cos t + sin t
2
dT̂ 1
= √ − sin t − cos t, − sin t + cos t
dt 2
Since R(s) is parametrized with respect to arclength, |R0 (s)| = 1.
T̂(s) = R0 (s)
√
Making ample use of the chain rule, and setting U (s) = ln s/ 2 , we
have U 0 (s) = 1s :
1
T̂(s) = √ (cos U (s) − sin U (s), cos U (s) + sin U (s))
2
dT̂ 1
= √ (− sin U (s) − cos U (s), − sin U (s) + cos U (s))
ds 2s
1.3.1.6. Solution. The circle of radius r centred at (0, r) is x2 +(y−r)2 =
r2 . The bottom half of this circle is
p
y = g(x) = r − r2 − x2
So
x
g 0 (x) = √ g 0 (0) = 0
r2 − x2
1 x2 1
g 00 (x) = √ + 3/2
g 00 (0) =
r2 − x2 [r2 − x2 ] r
As f (x) and g(x) have the same second order Taylor approximation at
x = 0, f 00 (0) = g 00 (0) = 1r .
We may parametrize the curve by r(x) = xı̂ı + f (x) ̂. So
and
|r0 (0) × r00 (0)| |f 00 (0)ı̂ı × ̂|
κ(0) = = = f 00 (0)
|r0 (0)|3 |ı̂ı|3
So κ(0) = f 00 (0) = 1
r and r is indeed the radius of curvature of y = f (x)
at x = 0.
1.3.1.7. Solution.
A v(t) = r0 (t) = (et , 2t + 1)
1.3.1.8. Solution 1. Note that (cos t + sin t)2 + (sin t − cos t)2 = 2 for
2 2
all t. So, the
√ points (x, y) of our curve lie on x + y = 2, which is a circle
of radius 2. Indeed
√
x(t) = cos t + sin t = 2 cos t cos π4 + sin t sin π4
√
= 2 cos t − π4
√
y(t) = sin t − cos t = 2 sin t cos π4 − cos t sin π4
√
= 2 sin t − π4
√
So, r(t) circumnavigates a circle of radius 2 and consequently has curva-
ture κ = √12 .
|v(t) × a(t)|
Solution 2. We use the formula κ = ds 3 , remembering that
dt
v(t) = r0 (t), a(t) = r00 (t), and ds
dt = |r0 (t)|.
√
ds dv p
= = (− sin t + cos t)2 + (cos t + sin t)2 = 2
dt dt
v(t) × a(t) 2k̂ 1
κ= = √ 3 = √
ds 3 2
2
dt
APPENDIX D. SOLUTIONS TO EXERCISES 412
(a) We’re given y in terms of x, so let’s use Part (e) of Theorem 1.3.3:
d2 y
2
dx ex
κ= 3/2 = 2 3/2
dy 2
1 + dx 1 + ex
1
κ(0) = = 2−3/2
[1 + 1]3/2
(b)
• The radius of the circle we want is ρ = κ1 = 23/2 . If its centre is at
(a, b), then the circle will have equation (x − a)2 + (y − b)2 = 23 . So,
we will find its centre.
• The unit vector N̂ points from our point (0, 1) towards the centre
of the circle. Since the radius of the circle is 23/2 , the centre of the
circle will be at (0, 1) + 23/2 N̂. So, we’ll find N̂.
ı̂ı + ̂
• Since N̂ is a unit vector perpendicular to T̂ = √ , we know N̂ will
2
ı̂ı − ̂ −ı̂ı + ̂
be either √ or √ .
2 2
• Using Part (d) of the proof of Theorem 1.3.3:
3
ds
v(t) × a(t) = κ T̂ × N̂
dt
√ 3 ı̂ı + ̂
(ı̂ı + ̂) × (̂) = 2−3/2 2 √ × N̂
2
APPENDIX D. SOLUTIONS TO EXERCISES 413
1
k̂ = √ (ı̂ı + ̂) × N̂
2
−ı̂ı + ̂
N̂ = √
2
So, the centre of our circle is at point (0, 1)+ρN̂ = (0, 1)+23/2 −ı̂
ı+̂
21/2
=
2 2
(−2, 3). Then the equation of the circle is (x + 2) + (y − 3) = 8.
1.3.1.11. ∗. Solution. (a) Think of
r(t) = (t, 1) − (sin t, cos t)
The (t, 1) part gives the position of the centre of the wheel at time t. The
other part gives the position of the thumbtack with respect to the centre
of the wheel. In particular,
• at time t = 0, r(0) = (0, 0). The thumbtack is on the ground (i.e. at
y = 0).
• At time t = π, r(π) = (π, 2). The thumbtack is at its highest point
(i.e. at y = 2) and is above the centre of the wheel at x = π.
• At time t = 2π, r(2π) = (2π, 0). The thumbtack is back on the
ground (i.e. at y = 0) and is below the centre of the wheel at x = 2π.
• At time t = 3π, r(3π) = (3π, 2). The thumbtack is again at its
highest point (i.e. at y = 2) and is above the centre of the wheel at
x = 3π.
• At time t = 4π, r(4π) = (4π, 0). The thumbtack is back on the
ground (i.e. at y = 0) and is below the centre of the wheel at x = 4π.
Here is a sketch of the curve.
r(t) = t − sin t , 1 − cos t
y
2π 4π x
(b) Since
r(t) = t − sin t , 1 − cos t
v(t) = r0 (t) = 1 − cos t , sin t
ds √
(t) = |v(t)| = 2 − 2 cos t
dt
a(t) = v0 (t) = sin t , cos t
ı̂ı ̂ k̂
v(t) × a(t) = det 1 − cos t sin t 0 = cos t − 1 k̂
sin t cos t 0
the curvature
|v(t) × a(t)| | cos t − 1| 1
κ(t) = 3
= 3/2
= 3/2 √
|v(t)| (2 − 2 cos t) 2 1 − cos t
(c) The radius of curvature at time t = π is
1 1
ρ(π) = = √ =4
κ(π) 1/23/2 2
APPENDIX D. SOLUTIONS TO EXERCISES 414
(d) At time π, the tack is at r(π) = (π, 2), which is at the top of its
trajectory. Looking at the sketch in part (a), we see that, at that time
N̂(π) = −̂. So the osculating circle at time t = π has center
(x − π)2 + (y + 2)2 = 16
so that
dT̂
2 2
1
1
κ(s) = (s) = − πs sin 2 πs ı̂ı + πs cos 2 πs ̂ = πs
ds
x3
1.3.1.13. ∗. Solution. The curve is y = y(x) = 3 . Since y 0 (x) = x2
and y 00 (x) = 2x, the curvature is
d2 y
2 (x) 2x
dx
κ(x) = h =
2 3/2
i 3/2
dy
1 + dx (x) 1 + x4
We’d like to find the critical points of κ(x), but differentiating it looks
messy. Since κ(x) has only nonnegative values, its maxima correspond the
the maxima of the function κ2 (x). So, we find the critical points of κ2 (x)
instead, to save ourselves some computational toil.
d d 4x2 8x 16x5
0= κ(x)2 = 3 = 3 −3 4
dx dx (1 + x4 ) (1 + x4 ) (1 + x4 )
8x(1 + x4 ) − 3 × 16x5 8x(1 − 5x4 )
= 4 = 4
(1 + x4 ) (1 + x4 )
T̂
N̂
Using the right-hand rule and B̂ = T̂ × N̂, B̂ points out of the page
(towards the reader).
To see this, point the fingers of your right hand in the direction of T̂,
and curl them inwards until they are in the direction of N̂. To do this,
your thumb must be pointing towards you, not away from you. Your thumb
shows the direction of T̂ × N̂.
1.4.1.2. Solution. In this equation, s stands for arclength.
When we take a very small interval from t to t + h, the change in
arclength s(t + h) − s(t) is approximately |r(t + h) − r(t)|, because our
curve is approximated by a straight line. So, s(t+h)−s(t)
h ≈ |r(t+h)−r(t)|
h ,
ds dr
leading to dt = | dt | = |v(t)|.
The magnitude of velocity is speed; in this text we generally call this v.
That is, v = |v(t)|. This leads to the potentially confusing (but standard)
convention that s stands for arclength, while v stands for speed.
1.4.1.3. Solution 1. Curves a and b are the same curve, just parametrized
differently (replace t with −t to convince yourself if the picture isn’t enough).
So, they ought to have the same torsion.
As in Example 1.4.4, we imagine that the curve is the thread on a bolt.
Take a look at your right hand. If your thumb is pointing up (corresponding
to the +z direction), and you’re looking at the tip of your thumb, your
fingers curl anticlockwise. Imagine a screw has threads matching the curves
a and b, and we turn it anticlockwise. The screw would move down — not
in the same direction as our thumb. So these curves are not right-handed
helices, so they have negative torsion.
The curve c sits entirely in a plane (the plane x = 0) so its torsion is
zero everywhere.
Solution 2. Here is the conventional computation for both a(t) and b(t).
(The upper sign is for a and the lower sign is for b.)
r(t) = cos t , ∓2 sin t , ±t/2
v(t) = − sin t , ∓2 cos t , ±1/2
a(t) = − cos t , ±2 sin t , 0
v(t) × a(t) = − sin t , ∓ cos t/2 , ∓2
da
(t) = sin t , ±2 cos t , 0
dt
da
v(t) × a(t) · (t) = −1
dt
APPENDIX D. SOLUTIONS TO EXERCISES 416
v(t) × a(t) · da
dt (t) 1
τ (t) = 2
=− 2 1 <0
|v(t) × a(t)| sin t + 4 cos2 t + 4
T̂
1.4.1.4. Solution. (a) If κ(s) ≡ 0, then dds = κ(s)N̂(s) ≡ 0 so that T̂ is
dr
a constant. As a result ds (s) = T̂ and r(s) = sT̂ + r(0) so that the curve
is the straight line with direction vector T̂ that passes through r(0).
B̂
(b) If τ (s) ≡ 0, then dds = −τ (s)N̂(s) ≡ 0 so that B̂ is a constant. As
T̂(s) ⊥ B̂,
d
(r(s) − r(0)) · B̂ = T̂(s) · B̂ = 0
ds
and (r(s) − r(0)) · B̂ must be a constant. The constant must be zero (set
s = 0), so (r(s) − r(0)) · B̂ = 0 and r(s) always lies in the plane through
r(0) with normal vector B̂.
(c) Parametrize the curve by arc length. Define the “centre of curva-
ture” at s by
1
rc (s) = r(s) + N̂(s)
κ(s)
Since κ(s) = κ0 is a constant and τ (s) ≡ 0,
d 1 1
rc (s) = T̂(s) + τ (s)B̂ − κ(s)T̂ = T̂(s) + 0B̂ − κ0 T̂ = 0
ds κ0 κ0
Thus rc (s) = rc is a constant and r(s) − rc = κ10 lies on the sphere of
radius κ10 centred on rc . Since τ (s) ≡ 0, the curve also lies on a plane, so
it is a circle.
1.4.1.5. ∗. Solution. (a), (b): T̂ points in the direction of the curve;
N̂ is perpendicular to it, in the same plane, pointing towards the centre of
curvature. Using the right-hand rule in the picture, we see B̂ is pointing
to the left.
z
z = x2 + y 2
N̂ T̂
B̂
y
x=y
x
(c) The torsion is zero, since the curve lies in a plane (the plane x = y).
ds p √
(t) = |r0 (t)| = 4 + 2e2t + 2e−2t = 2 et + e−t
dt
r00 (t) = et − e−t ı̂ı + et + e−t ̂
APPENDIX D. SOLUTIONS TO EXERCISES 417
the curvature
√
|v(t) × a(t)| 2 4 + 2e2t + 2e−2t 1
κ(t) = = =
ds 3 [4 + 2e2t + 2e−2t ]3/2 2 + e2t + e−2t
dt
(v(t) × a(t)) · da
dt (t)
τ (t) =
|v(t) × a(t)|2
(24, −12, 2) · (0, 0, 6) 3
τ (2) = √ =
(2 181) 2 181
1.4.1.8. Solution. For the specified curve
t2 t3
r(t) = tı̂ı + ̂ + k̂
2 3
v(t) = r0 (t) = ı̂ı + t ̂ + t2 k̂
a(t) = r00 (t) = ̂ + 2t k̂
ı̂ı ̂ k̂
v(t) × a(t) = det 1 t t2
0 1 2t
= t2 ı̂ı − 2t ̂ + k̂
a0 (t) = 2 k̂
From this, we read off
v(t) ı̂ı + t ̂ + t2 k̂
T̂(t) = =√
|v(t)| 1 + t2 + t4
√
|v(t) × a(t)| 1 + 4t2 + t4
κ(t) = 3
=
|v(t)| [1 + t2 + t4 ]3/2
v(t) × a(t) t2 ı̂ı − 2t ̂ + k̂
B̂(t) = =√
|v(t) × a(t)| 1 + 4t2 + t4
N̂(t) = B̂(t) × T̂(t)
ı̂ı ̂ k̂
1
=√ √ det t2 −2t 1
1 + t2 + t4 1 + 4t2 + t4
1 t t2
APPENDIX D. SOLUTIONS TO EXERCISES 418
(v(5) × a(5)) · da
dt (5)
0 = τ (5) =
|v(5) × a(5)|2
da
0 = (v(5) × a(5)) · (5)
dt
= (c2 e5c , 15ce5c (2 − 5c), −30) · (6, 0, c3 e5c )
= 6c2 e5c (1 − 5c)
1
c = 0 or c =
5
If c = 0, then r(t) = (t3 , t, 1), and so the entire curve is contained inside
the plane z = 1. (Its torsion is zero everywhere — not just at t = 5.)
Consider the case c = 15 . When t = 5, our curve (and its osculating
circle) passes through the point r(5) = (53 , 5, e). The normal vector to
v(5)×a(5)
the plane of the osculating curve is the binormal vector B̂(5) = |v(5)×a(5)| .
Since we don’t need the normal vector to the plane to be a unit vector,
we can take as the normal vector to the plane simply v(5) × a(5), or
(e/25, 3e, −30). Then, an equation of the plane containing the osculating
circle is (e/25)x + (3e)y − 30z = −10e. An equivalent equation for this
plane is (1/25)x + 3y − (30/e)z = −10.
1.4.1.10. ∗. Solution. (a) Since r0 (t) = (2t, 1, 3t2 ), we have r0 (1) =
(2, 1, 3). So the normal plane must pass through r(1) = (1, 1, 1) and be
perpendicular to (2, 1, 3). The equation of the normal plane is then
2(x − 1) + (y − 1) + 3(z − 1) = 0 or 2x + y + 3z = 6
(b) As
ds p
v(t) = r0 (t) = 2t, 1, 3t2
= 1 + 4t2 + 9t4
dt
a(t) = v0 (t) = 2, 0, 6t v(t) × a(t) = 6t, −6t2 , −2
the curvature
√
|v(t) × a(t)| 2 1 + 9t2 + 9t4
κ(t) = =
ds 3 [1 + 4t2 + 9t4 ]3/2
dt
APPENDIX D. SOLUTIONS TO EXERCISES 419
ds √
= |v(t)| = 2
dt
2 2 d2 s
From a(t) = ddt2s (t) T̂(t) + κ(t) ds
dt (t) N̂(t), and the fact that dt2 = 0, we
read off that
ds −2 1 a
κ(t) = (t) |a| = N̂(t) = = − cos tı̂ı − sin t ̂
dt 2 |a|
we read off
v(t) × a(t) 1 1 1
B̂(t) = = √ sin tı̂ı − √ cos t ̂ + √ k̂
|v(t) × a(t)| 2 2 2
so that
√
π 1 3 1
B̂ = √ ı̂ − √ ̂ + √ k̂
ı
6 2 2 2 2 2
1.4.1.12. ∗. Solution. (a) The velocity vector is
0
r (t) = (− sin(t), cos(t), 2t)
So a tangent vector at t = π is T = (0, −1, 2π) and a parametric form for
the tangent line is
R(t) = r(π) + tT = (−1, 0, π 2 ) + t(0, −1, 2π)
(b) The speed is
ds p
= |r0 (t)| = 1 + 4t2
dt
By Theorem 1.3.3, the tangential component of acceleration is
d2 s d p 4t
aT (t) = 2 = 1 + 4t2 = √
dt dt 1 + 4t2
APPENDIX D. SOLUTIONS TO EXERCISES 420
r0 (t) 1
T̂(T ) = = √ sin tı̂ı + cos t ̂ + 2 k̂
|r0 (t)| 5
So the tangential component of acceleration at time t is
d2 s
aT (t) = (t) T̂(t) = sin tı̂ı + cos t ̂ + 2 k̂
dt2
(c) The (full) acceleration is
d 0
r00 (t) =
r (t) = sin t + t cos t ı̂ı + cos t − t sin t ̂ + 2 k̂
dt
So the normal component of acceleration at time t is
ds
2
(d) Another formula for the normal component of acceleration is κ(t) dt (t) N̂(t).
ds
2
So the magnitude of the normal component of acceleration is κ(t) dt (t)
and, by part (c),
2
ds
κ(t) (t) = t cos tı̂ı − t sin t ̂ = t
dt
x2 + y 2 = 8 − 2x or (x + 1)2 + y 2 = 9
(x + 1)2 + y 2 = 9 and z = 8 − 2x
x(θ) = −1 + 3 cos θ
y(θ) = 3 sin θ
z(θ) = 8 − 2x(θ) = 10 − 6 cos θ
APPENDIX D. SOLUTIONS TO EXERCISES 421
v(0)
T̂(0) = = ̂
|v(0)|
and, since v(0) × a(0) = 9k̂ + 18ı̂ı, the unit binormal vector and curvature
at (2, 0, 4) are
√ √
v(0) × a(0) 2ı̂ı + k̂ |v(0) × a(0)| 9 5 5
B̂(0) = = √ κ(0) = 3
= 3 =
|v(0) × a(0)| 5 |v(0)| 3 3
v(t) × a(t)
B̂ = and N̂ = B̂ × T̂
|v(t) × a(t)|
(b) As
we have
ds
q
= |r0 (θ)| = sin2 θ + cos2 θ + (sin θ − cos θ)2
dθ √
= 2 − 2 sin θ cos θ
p
= 2 − sin(2θ)
v(θ) × a(θ) = 1, 1, 1
π 5π 3π 7π
• which in turn is the case when θ = 4, 4 (θ = 4 , 4 ).
So
√
3 √ ı̂ı ̂ √
maximum curvature = = 3 at √ + √ + (1 − 2) k̂
[2 − 1]3/2 2 2
ı̂ı ̂ √
and − √ − √ + (1 + 2) k̂
2 2
√
3 1 ı̂ı ̂
minimum curvature = = at − √ + √ + k̂
[2 − (−1)]3/2 3 2 2
ı̂ı ̂
and √ − √ + k̂
2 2
1.4.1.17. ∗. Solution. For r(t) to be well-defined, we need t > 0
(because of the ln t.)
r
0 1 ds 1 1
v(t) = r (t) = 2tı̂ı + 2 ̂ + k̂ = 4t2 + 4 + 2 = 2t +
t dt t t
The unit tangent vector is
1
r0 (t) 2tı̂ı + 2 ̂ + t k̂ 2t2 ı̂ı + 2t ̂ + k̂
T̂(t) = = =
0
|r (t)| 2t + 1t 2t2 + 1
we have
t2 t3
r(t) = ı̂ı + ̂ + k̂
2 3
0 2
r (t) = t ̂ + t k̂
APPENDIX D. SOLUTIONS TO EXERCISES 424
ds p p
(t) = |r0 (t)| = t2 + t4 = t 1 + t2
dt
the length of the curve is
Z 1 Z 1 p
ds 1 3/2 1 1
t 1 + t2 dt = 1 + t2 = 23/2 − 1
(t) dt =
0 dt 0 3 0 3
ds
T̂0 (t) = κ(t) (t) N̂(t)
dt
we have, by part (d),
√
π |T̂0 π
4 | 1/ 2 1
κ = 0 π = √ =
4 |r 4 | 2 2
1.4.1.19. ∗. Solution. (a), (b), (c) We have
r(t) = t + 2 , 1 − t , t2 /2
v(t) = r0 (t) = 1 , −1 , t
ds p
(t) = |v(t)| = 2 + t2
dt
a(t) = v0 (t) = 0 , 0 , 1
√ 2
(e) Since ds
dt (t) = 2 + t2 , we have ddt2s (t) = √2+t
t
2
and
d2 s ds 2
0 , 0 , 1 = a(t) = 2 T̂(t) + κ(t) N̂(t)
dt dt √
t (1 , −1 , t) 2 p 2
=√ √ + 3/2
2 + t2 N̂(t)
2+t 2 2+t 2
[2 + t2 ]
or
√
2 (t , −t , t2 ) (−t , t , 2)
√ N̂(t) = 0 , 0 , 1 − 2
=
2+t2 2 + t 2 + t2
APPENDIX D. SOLUTIONS TO EXERCISES 425
which implies
(−t , t , 2)
N̂(t) = p
2(2 + t2 )
(f) At t = 0
r(0) = (2, 1, 0)
(1, −1, 0)
T̂(0) = √
2
N̂(0) = (0, 0, 1)
1 1
B̂(0) = T̂(0) × N̂(0) = √ (1, −1, 0) × (0, 0, 1) = √ (−1, −1, 0)
2 2
The osculating plane is the plane through r(0) which is perpendicular to
B̂(0), which is
1
√ (−1, −1, 0) · (x, y, z) − (2, 1, 0) = 0 or x+y =3
2
(g) The osculating circle has centre
1 1
r(0) + N̂(0) = (2, 1, 0) + (0, 0, 1) = (2, 1, 2)
κ(0) 1/2
1.4.1.20. ∗. Solution. First some preliminary computations.
3
t t2
r(t) = ı̂ı + √ ̂ + t k̂
3 2
0
√
r (t) = t2 ı̂ı + 2t ̂ + k̂
p
|r0 (t)| = t4 + 2t2 + 1 = t2 + 1
√
r00 (t) = 2tı̂ı + 2 ̂
ı̂ı √̂ k̂ √ √
r0 (t) × r00 (t) = det t2 2t 1 = − 2ı̂ı + 2t ̂ − 2t2 k̂
√
2t 2 0
(a) The unit tangent vector is
√
r0 (t) t2 ı̂ı + 2t ̂ + k̂
T̂(t) = 0 =
|r (t)| t2 + 1
(b) The curvature is (see §1.5)
√ √ √
|r0 (t) × r00 (t)| | − 2ı̂ı + 2t ̂ − 2t2 k̂| 2 + 4t2 + 2t4
κ(t) = 0 3
= 3 = 3
|r (t)| 2
(t + 1) (t2 + 1)
√
2
= 2
2
(t + 1)
(c) At t = 0 √
κ(0) = 2
For ease of computation, we’ll find B̂ first, then use it to find N̂.
(e) At t = 0, the binormal vector is (see §1.5)
√
r0 (0) × r00 (0) − 2ı̂ı
B̂(0) = 0 = √ = −ı̂ı
|r (0) × r00 (0)| 2
APPENDIX D. SOLUTIONS TO EXERCISES 426
r(t) = (t2 , t , t3 )
v(t) = r0 (t) = (2t , 1 , 3t2 )
a(t) = r00 (t) = (2 , 0 , 6t)
(a) In particular, a (non unit) tangent vector at r(−1) = (1, −1, −1) is
r0 (−1) = (−2, 1, 3). So the tangent line to the curve at (1, −1, −1) is
or
x = 1 − 2t
y = −1 + t
z = −1 + 3t
r0 (t) 1
T̂ (t) = = √ sin tı̂ı + cos t ̂ + 2 k̂
|r0 (t)| 5
so that
ds dT̂ 1
κ(t) (t) N̂(t) = (t) = √ cos tı̂ı − sin t ̂
dt dt 5
APPENDIX D. SOLUTIONS TO EXERCISES 427
So
√ √ √ 1/2 √ 1/2
2t−1/2 /2, 2t−1/2 /2, −1
0 2t , 2t , 1 − t
T̂ (t) = −
1+t (1 + t)2
√ √ √ √
2/2, 2/2, −1 2, 2, 0
T̂0 (1) = −
2 4
1
= 0, 0, −
2
So the principal unit normal vector is
T̂0 (1)
N̂(1) = = (0, 0, −1)
|T̂0 (1)|
(c) As
dT̂ 1 ds
(1) = 0, 0, − (1) = |v(1)| = 4
dt 2 dt
the curvature
|T̂0 (1)| 1
κ(1) = =
|v(1)| 8
1.4.1.24. ∗. Solution. (a) For the specified curve
(b) Since
we have
π √
v(t) 1 1 3 3 3
T̂(t) = = (−3 cos t, 3 sin t, 4) T̂ = − , ,4
|v(t)| 5 6 5 2 2
π √
1 13 3 3
T̂0 (t) = (3 sin t, 3 cos t, 0) T̂0 = , ,0
5 6 5 2 2
3 √
= 3, 1, 0
10
T̂0 π
1 √
π
6
π π π
N̂ = π
= 3, 1, 0 B̂( = T̂ × N̂
6 |T̂0 6 |
2 6 6 6
1 √
= − 4, 4 3, −6)
10
1 √
= − 2, 2 3, −3)
5
1.4.1.25. Solution. (a) The curve x2 + y 2 = 1 is a circle of radius 1.
So we can parametrize it by x(θ) = cos θ, y(θ) = sin θ, 0 ≤ θ < 2π. The
z-coordinate of any point on the intersection is determined by z = x2 − y 2 .
So we can use the parametrization
r(θ) = cos θ ı̂ı + sin θ ̂ + [cos2 θ − sin2 θ] k̂
= cos θ ı̂ı + sin θ ̂ + cos(2θ) k̂ 0 ≤ θ < 2π
√ √
(b) Note that r(θ) = 1/ 2 , 1/ 2 , 0 when θ = π4 . For general θ, the
N̂(π/4) √ √ √ √
rc (π/4) = r(π/4) + = 1/ 2 , 1/ 2 , 0 − 5 1/ 2 , 1/ 2 , 0
κ(π/4)
√ √
= − 2 2 , −2 2 , 0
1.4.1.26. ∗. Solution. We’ll solve this problem twice, using two differ-
ent strategies. (The second strategy will be much more efficient than the
first one.) Both strategies use that F = ma. Since we are told that m = 2,
we just have to find the acceleration a at (1, 1, 1).
Strategy 1: In the first strategy, we’ll find the position r(t), as a function
of time t and then differentiate twice to get the acceleration a(t).
• First we’ll find any old parametrization. We are told that, on the
path, z = x and z = y 2 . So let’s use y as the parameter. Then
x = z = y 2 . So the parametrization is R(y) = y 2 ı̂ı + y ̂ + y 2 k̂.
(We’ll save the notation “r(t)” for the parametrization with respect
to time.)
• Next we’ll reparametrize to get the time t as the parameter. Since
dR
= 2y ı̂ı + ̂ + 2y k̂
dy
ds p
=⇒ = 2y ı̂ı + ̂ + 2y k̂ = 1 + 8y 2
dy
We are told that the speed dsdt = 3 for all t. So, choosing our zero
point for time to coincide with our zero point for s, we have s = 3t,
or t = s/3 so that
dt 1p
= 1 + 8y 2
dy 3
We could now integrate to get t as a function of y. But that looks
quite messy. Fortunately we only need the acceleration at one point,
namely (1, 1, 1). We’ll now see that that saves quite a bit of work.
APPENDIX D. SOLUTIONS TO EXERCISES 430
dy 3
(t) = p
dt 1 + 8y(t)2
1 3 16y(t)y 0 (t)
d2 y
(t) = −
dt2 2 [1 + 8y(t)2 ]3/2
In particular
3 3
y 0 (t0 ) = p =√ =1
1 + 8y(t0 )2 1+8
24 y(t0 )y 0 (t0 ) 24 × 1 × 1 8
y 00 (t0 ) = − 3/2
=− 3/2
=−
[1 + 8y(t0 )2 ] (1 + 8) 9
16
=− 2ı̂ı + ̂ + 2 k̂ + 2 2ı̂ı + 2 k̂
9
4 16 4
= ı̂ı − ̂ + k̂
9 9 9
a = 9κN̂
So we just have to find the curvature, κ, and unit normal, N̂, at (1, 1, 1). We
have already found one parametrization of the path in strategy 1, namely
|T̂0 (1)|
κ(1) =
|R0 (1)|
T̂0 (1)
N̂(1) =
|T̂0 (1)|
T̂0 (1) 2ı̂ı − 8 ̂ + 2 k̂
F = ma = 2 × 9κ(1)N̂(1) = 18 = 18 √
|R0 (1)| 27 1 + 8 × 12
4
= (ı̂ı − 4 ̂ + k̂)
9
1.4.1.27. ∗. Solution. (a) As
√
r(t) = 2tı̂ı + t2̂ + 3t2 k̂
√
r0 (t) = 2ı̂ı + 2t̂ + 2 3tk̂
the unit tangent vector is
√ √
ı̂ı + t̂ + 3tk̂ ı̂ı + t̂ + 3tk̂
T̂(t) = √ = √
|ı̂ı + t̂ + 3tk̂| 1 + 4t2
(b) Since
√ √ √
dT̂ ̂ + 3k̂ ı̂ı + t̂ + 3tk̂ −4tı̂ı + ̂ + 3k̂
(t) = √ − 4t 3/2
= 3/2
dt 1 + 4t2 (1 + 4t2 ) (1 + 4t2 )
the unit normal is
√ √
−4tı̂ı + ̂ + 3k̂ −4tı̂ı + ̂ + 3k̂
N̂(t) = √ = √
| − 4tı̂ı + ̂ + 3k̂| 2 1 + 4t2
(c) The unit binormal is
B̂(t) = T̂(t) × N̂(t)
ı̂ı ̂ √k̂
1
= det 1 t √3t
2(1 + 4t2 )
−4t 1 3
√ 2 2
− 3(1 + 4t )̂ + (1 + 4t )k̂
=
2(1 + 4t2 )
√
3 1
=− ̂ + k̂
2 2
which is (3).
(d) The
√
plane contains the point r(0) = 0 and is perpendicular to the
vector − 23 ̂ + 12 k̂ and so is
√
− 3y + z = 0
APPENDIX D. SOLUTIONS TO EXERCISES 432
v
C
u x
1.4.1.28. ∗. Solution. The three unit vectors T̂, N̂ and B̂ are mutually
perpendicular and form a right handed triple.
B̂
N̂
T̂
So
N̂ = B̂ × T̂ N̂ × T̂ = −B̂ B̂ × N̂ = −T̂
and
This leads to
ı̂ı ̂ k̂
1 √ 1 √ √
N̂ = B̂ × T̂ = √ −1 2 2 √2 = − √ 6 2, 2, 2
78 78
0 2 − 2
1 √
= −√ 6, 1, 2 .
39
Finally, √ √ √ √ √
|v × a| 2 26 2 2 13 13 39
κ= = √ 3 = √ √ = √ = .
v3 ( 6) 6 2 3 3 3 9
(b) Now parametrize the curve by time, t, and write v = r0 (t), v =
|r (t)| and a = r00 (t). Note that in part (a) we used v, v and a with
0
different meanings. We use the dot product to extract the tangential and
normal components of a = dv 2
dt T̂ + v κN̂:
dv
a · T̂ = T̂ + v 2 κN̂ · T̂
dt
dv
= T̂ · T̂ + (v 2 κ)N̂ · T̂
dt
Since T̂ is a unit vector, T̂·T̂ = kT̂k2 = 1; since T̂ and N̂ are perpendicular,
T̂ · N̂ = 0.
dv
=
dt
dv
This gives us a nice way to compute dt , the rate of change of speed.
dv √ 1 √
= a · T̂ = (−2, 3, −2 2) · √ 0, 2, − 2
dt 6
1 10 5√
= √ [0 + 6 + 4] = √ = 6.
6 6 3
Similarly, a · N̂ = v 2 κ, so
1 9 −1 √ √ 9 × 13
v2 = a · N̂ = √ √ (−2, 3, −2 2) · 6, 1, 2 = = 3.
κ 39 39 39
√ √
Hence |v| = 3; since v = |v|, v = 3. Then v = |v|T̂ = v T̂ =
√ √ √
√3 0, 2, − 2 = (0, 2, −1).
6
APPENDIX D. SOLUTIONS TO EXERCISES 434
and
1
r(θ) = 4 cos θ , 2 sin θ , cos(2θ)
4
0
1
v(θ) = r (θ) = − 4 sin θ , 2 cos θ , − sin(2θ)
2
00
a(θ) = r (θ) = − 4 cos θ , −2 sin θ , − cos(2θ)
v(π) = 0 , −2 , 0
a(π) = 4 , 0 , −1
v(π) × a(π) = 2 , 0 , 8
So the curvature at θ = π is
√
|v(π) × a(π)| |(2, 0, 8)| 17
κ(π) = 3
= =
|v(π)| |(0, −2, 0)|3 4
(b) The radius is
1 4
=√
κ(π) 17
(c) Set R(t) = r(t2 ). Then
R0 (t) = 2t r0 (t2 )
R00 (t) = 2 r0 (t2 ) + 4t2 r00 (t2 )
In particular,
√ 1
R( π) = − 4 , 0 ,
√ √ 4 √
R0 ( π) = 2 π v(π) = 0 , −4 π , 0
√ √
speed = R0 ( π) = 4 π
√
acceleration = R00 ( π) = 2 v(π) + 4π a(π) = 16π , −4 , −4π
dr
v(θ) = (θ) = x0 (θ)ı̂ı + y 0 (θ)̂
dθ
= r0 (θ) cos θ − r(θ) sin θ ı̂ı + r0 (θ) sin θ + r(θ) cos θ ̂
ds
q 2 2
=⇒ (θ) = r0 (θ) cos θ − r(θ) sin θ + r0 (θ) sin θ + r(θ) cos θ
dθ p
= r0 (θ)2 + r(θ)2
Hence
Z Z θ2 ds
f (x, y) ds = f x(θ), y(θ) dθ
C θ1 dθ
s 2
Z θ2
dr
= f r(θ) cos θ, r(θ) sin θ r(θ)2 + (θ) dθ
θ1 dθ
for some constant vector C. To satisfy the initial condition that r0 (0) =
v0 = ı̂ı + k̂, we need
So
r0 (t) = ı̂ı + t ̂ + cos t k̂
Integrating a second time, and imposing the initial condition that r(0) = ̂,
gives
t2 t2
r(t) = tı̂ı + ̂ + sin t k̂ + ̂ = tı̂ı + 1 + ̂ + sin t k̂
2 2
(b) The particle has x(t) = π/2 when t = π/2. So
π π2
r(π/2) = ı̂ı + 1 + ̂ + k̂
2 8
(c) The work done is
Z π/2
Work = F(t) · r0 (t) dt
0
Z π/2
= ̂ − sin t k̂ · ı̂ı + t ̂ + cos t k̂ dt
0
Z π/2
= t − sin t cos t dt
0
h t2
1 iπ/2
= cos2 t +
2 2 0
π2 1
= −
8 2
L4 n̂
x
R L2
n̂
So
Z Z Z Z Z
F · n̂ ds = F · (−̂) ds+ F · ı̂ı ds+ F · (̂) ds+ F · (−ı̂ı) ds
C L1 L2 L3 L4
y x y ds
Z 3 z }| { Z 1 z}|{ Z 0 z}|{ z }| {
= − (−1) ex dx + (3) y 2 dy + (1) ex (−dx)
0 −1 3
x ds
z}|{ z }| { Z −1
(0) y 2 (−dy)
+
1
= e3 − 1 + 13 − (−1)3 + e3 − 1 + 0
= 2e3
so that r(0) = (3, 1) is the initial point of L3 and r(1) = (0, 1) is the final
point of L3 . Then
ds
r0 (t) = (−3, 0) (t) = |r0 (t)| = 3
dt
and
y(t)ex(t) ds
dt (t)
Z Z 1 ds
Z 1 z }| { z}|{
3−3t
F · ̂ ds = F r(t) · ̂ (t) dt = e 3 dt
L3 0 dt 0
1
= −e3−3t = e3 − 1
0
ds
q 2 2
= |r0 (t)| = cos t − t sin t + sin t + t cos t + (2t)2
dt p
= 1 + 5t2
r0 (π) = −ı̂ı − π ̂ + 2π k̂
r0 (t) 1
T̂(π) = 0
=√ − ı̂ı − π ̂ + 2π k̂
|r (t)| 1 + 5π 2
(b)
Z p Z π Z π p
2 2
p
2 2
ds
x + y ds = x (t) + y (t) dt = t 1 + 5t2 dt
C 0 dt 0
APPENDIX D. SOLUTIONS TO EXERCISES 439
π
1 1
(1 + 5t2 )3/2 (1 + 5π 2 )3/2 − 1
= =
15 0 15
(c) For every t, the coordinates x(t) = t cos t, y(t) = t sin t, z(t) = t2
obey x(t)2 + y(t)2 = t2 = z(t) and so the 2
curve lies on z = x + y .
2
x
y
R
xρ ds
1.6.1.11. Solution. We use the centre of mass formulae x̄ = RC ,
C
ρ ds
etc. To make the working clearer, we’ll break these calculations into several
steps.
1
x(t) = t + t2 x0 (t) = 1 + t
2
1
y(t) = t − t2 y 0 (t) = 1 − t
2
4 3/2 √
z(t) = t z 0 (t) = 2 t
3
p p
x0 (t)2 + y 0 (t)2 + z 0 (t)2 =
1 + 2t + t2 + 1 − 2t + t2 + 4t
p √
= 2(t2 + 2t + 1) = 2(t + 1)
x(t) + y(t) (t + t2 /2) + (t − t2 /2)
ρ(x(t), y(t), z(t)) = = =t
2 2
√
Z Z 4 √ 23 · 11 2
ρ ds = t 2(t + 1) dt =
C 0 3
2
Z 4 √ √ Z 2 t4
Z
1 3
xρ ds = t + t2 t 2(t + 1) dt = 2 + t3 + t2 dt
C 0 2 0 2 2
√
√
9 6
5
2 2 2 · 103 2
= 2 + 3(25 ) + =
5 3 15
2
Z 4 √ √ Z 2 t4 t3
Z
1
yρ ds = t − t2 t 2(t + 1) dt = 2 − + + t2 dt
C 0 2 0 2 2
√
√
9 6
5
2 2 2 · 23 2
= 2 − + 25 + =−
5 3 15
√ Z 2
4 3/2 √
Z 4
4 2 2 7/2
Z
zρ ds = t t 2(t + 1) dt = t + t5/2 dt
C 0 3 3 0
√ 10 8
10
√
4 2 2 2 2 · 37 2
= + =
3 9 7 7 · 33
APPENDIX D. SOLUTIONS TO EXERCISES 440
√
25 ·103 2
R
xρ ds 15 412
x= R = 3 √ = ≈ 7.5
ρ ds 2 ·11 2 55
3
5
√
− 2 ·23 2
R
yρ ds 15 92
y= R = 3 √ =− ≈ −1.7
ρ ds 2 ·11 2 55
3
√
210 ·37 2
R
zρ ds 3 4736
z= R = 37·3 √ = ≈ 6.8
ρ ds 2 ·11 2 693
3
W N̂
−mĝ
F = ma = W N̂ − mĝ
W N̂ · N̂ − mĝ · N̂ = ma · N̂
APPENDIX D. SOLUTIONS TO EXERCISES 441
d2 s ds 2
Using the equation a(t) = dt2 T̂ +κ dt N̂,
2
ds
W − mĝ · N̂ = mκ
dt
2
ds
W = mg n̂ · N̂ + mκ
dt
2
ds
= mg cos θ + mκ
dt
v × a = 5(−4, 0, 3)
|v × a| 25 1
κ(θ) = = 3 =
ds 3 5 5
dθ
2
d s
Since dθ 2 = 0, we use the following theorem to find N̂:
2
d2 s ds
a(θ) = 2 T̂ + κ N̂
dθ dθ
25
(−3 cos θ, −5 sin θ, −4 cos θ) = 0 + N̂
5
3 4
N̂(θ) = − cos θ, − sin θ, − cos θ
5 5
ds √
(13π/3) = 1/ 2
dθ
2 !
d2 s ds
a(θ) · ̂ = T̂ + κ N̂ · ̂
dθ2 dθ
√
3 √
− = 0 + 6(1/2)N̂ · ̂
2
1
N̂ · ̂ = − √
2
Now we can find the speed |v| of the bead when |W | = 100 and it
breaks off the track.
W = mκ|v|2 + mĝ · N̂
1 √
9.8 1
±100 = 6|v|2 + −√
9.8 9.8 2
s
9.8 1
|v| = √ 100 + √ ≈ 20m/s ≈ 72kph
6 2
(Because |v| > 0, the equation above has no solution for W = −100.)
Quite fast! 100 N is a lot of force for such a light object.
1.7.4.9. Solution. According to the equation in §1.7.2, the skier will
become airborne when: r
g
|v| > |̂ · N̂|
κ
We’ll use the equation of the curve to find κ and N̂.
Note that g is given in metres per second, while the other quanti-
ties are in kilometres and hours. Converting, 9.8 m/s2 is the same as
9.8 m
1 km 3600 s 2
1 s2 1000 m 1 hr = 98 · 64 km
h2
= 25 · 34 · 72 km
h2
.
r(t) = (ln t, 1 − t)
ds p
r0 (t) = v(t) = (t−1 , −1) = |v(t)| = 1 + t−2
dt
r00 (t) = a(t) = (−t−2 , 0)
|v × a| t−2 |t| t
κ(t) = 3 =√ 3 = 2 )3/2
=
ds (1 + t (1 + t2 )3/2
1+t −2
dt
ground.
r s r
g g · (1 + t2 )3/2 g(1 + t2 )
|v| = |N̂ · ̂| = =
κ t(1 + t2 )1/2 t
(It seems unlikely that one could reach this speed on skis. The skier is
probably earth-bound until they find a curvier hill.)
1.7.4.10. Solution. We now have three forces acting on the bead, rather
than the two in §1.7.1. The wire still exerts a normal force W N̂ on the
bead to keep it on the wire; gravity still exerts a force −mĝ straight down.
Now our jet-pack force also exerts a force parallel to the direction of the
bead’s motion, i.e. parallel to T̂. This force is U T̂.
W N̂
U T̂
−mĝ
The net force acting on the bead is the sum of these three forces:
F = ma = U T̂ + W N̂ − mĝ
0 = (U T̂ + W N̂ − mĝ) · T̂
= (U T̂ · T̂) + (W N̂ · T̂) − mĝ · T̂
dy
= U + 0 − mg
ds
dy
U = mg
ds
1.7.4.11. Solution. (a) There are three forces acting on the snowma-
chine. If it’s not accelerating, then F = ma = 0: that is, the forces all
cancel out.
APPENDIX D. SOLUTIONS TO EXERCISES 445
W N̂
M T̂
−mĝ
So, we have the equation
ma = W N̂ + M T̂ − mĝ
ma · T̂ = W N̂ · T̂ + M T̂ · T̂ − mĝ · T̂
= 0 + M − mĝ · T̂
0 = M − mĝ · T̂
M = mĝ · T̂
(b) We would expect, from looking at the situation, that the engine
would have to provide a “backwards” force to slow the acceleration due to
gravity. So, we would expect M < 0. Indeed, if T̂ points downhill, then
the y-component of T̂ is negative, so M = mĝ · T̂ is negative.
(This is the purpose of driving downhill in a low gear: the friction inside
the motor provides a force opposing the direction of motion, slowing the
vehicle.)
(c) To use the equation M = mĝ · T̂, we’ll need to find ̂ · T̂.
So,
21 1960
M = (200kg)(9.8m/s ) − √ m = − √ N ≈ −1131.6N
3 3
1.7.4.12. Solution. We begin with the usual computations.
|v(θ) × a(θ)| 12
κ(θ) = =
ds 3 (9 + 7 sin2 θ)3/2
dθ
(−4 sin θ, 3 cos θ)
T̂(θ) = p
9 + 7 sin2 θ
(36 cos θ, 48 sin θ)
T̂0 (θ) =
−(9 cos2 θ + 16 sin2 θ)3/2
12
|T̂0 (θ)| =
9 cos2 θ + 16 sin2 θ
(3 cos θ, 4 sin θ)
N̂(θ) = p
− 9 cos2 θ + 16 sin2 θ
We want to find the height yS where |v| = 0, and the height yA where
W = 0. Remember that v in these equations is the derivative of position
with respect to time, and is not the same as v(θ).
1
Equation 1.7.1 : E= m|v|2 + mgy
2
E
If |v| = 0 : E = mgyS =⇒ yS =
mg
This answers part a.
Using y = 3 + 3 sin θ:
yA −3
24 (E − mgyA ) 3
= − 4mg q
3/2
yA −3 2 yA −3 2
9+7 3
9+7 3
yA −3
24 (E − mgyA ) 3
= 4mg q
3/2
yA −3 2 yA −3 2
9+7 3
9 + 7 3
APPENDIX D. SOLUTIONS TO EXERCISES 447
E
6 mg − yA yA −3
3
⇔ =q
2 3/2 2
9 + 7 yA3−3 9 + 7 yA3−3
2
11
6 2·9.8 − yA yA −3
3
⇔ = q
3/2
yA −3 2
2
9+7 3
9 + 7 yA3−3
3/2
yA −3 2
To simplify to a more standard form, we multiply both sides by 9 + 7 3 :
2 !
112
yA − 3 yA − 3
6 − yA = 9+7
2 · 9.8 3 3
Now, we simplify to
7 3 2 7797
0= y − 7yA + 48yA −
9 A 49
Now, solving for yA involves solving a cubic function, which is no small
task. We could ask a computer, but we can also get an idea of its root(s)
by plugging in numbers and using the intermediate value theorem. In
particular, we need to know whether yA is greater than 6 (the skater makes
it!) or between 3 and 6 (they fall off the ceiling).
Let f (y) = 79 y 3 − 7y 2 + 48y − 7797 12941
49 . Note f (4) = − 441 , which is
1367
negative, and f (5) = 441 , which is positive. So, by the intermediate value
theorem, there is a root of f (y) between y = 4 and y = 5. That is, yA
is between 4 and 5, so the skater falls off the ceiling somewhere between
these heights, rather than making it all the way around.
1.7.4.13. Solution. (a) By Newton’s law of motion
d 1
E 0 (t) = m|v(t)|2 + mgr(t) · k̂ = mv(t) · v0 (t) + mgv(t) · k̂
dt 2
= v(t) · N r(t) − mg k̂ + mgv(t) · k̂
=0
since v(t) · N r(t) = 0. So E(t) is a constant, independent of t.
(b) By part (a),
1 2
E(t) = E(0) =⇒ 2 m|v(t)|
+ mgbθ(t) = mg(2πb)
2
=⇒ |v(t)| = 2gb 2π − θ(t)
dr dr dθ dθ
v= = = − a sin θ, a cos θ, b
dt dθ dt dt
2
dθ
=⇒ |v|2 = a2 + b2 ]
dt
2
1/2 1/2
dθ |v| 2gb(2π − θ)
=⇒ =− 2 =−
dt a + b2 a2 + b2
In the figure on the left below, the blue half-line is the set of all points with
polar coordinates θ = π, r > 0 and the pink half-line is the set of all points
with polar coordinates θ = π, r < 0. In the figure on the right below, the
blue half-line is the set of all points with polar coordinates θ = 0, r > 0
and the pink half-line is the set of all points with polar coordinates θ = 0,
r < 0.
y y
(−2, 0) π (−2, 0)
x x
√ √
(b)√The distance from (1, 1) to the origin is 2. So either r = 2 or
r = − 2.
√
• If r = 2, then θ must obey
√ √ 1
(1, 1) = 2 cos θ , 2 sin θ ⇐⇒ sin θ = cos θ = √
2
π
⇐⇒ θ = + 2nπ, n integer
4
√
• If r = − 2, then θ must obey
√ √ 1
(1, 1) = − 2 cos θ , − 2 sin θ ⇐⇒ sin θ = cos θ = − √
2
5π
⇐⇒ θ = + 2nπ, n integer
4
In the figure on the left below, the blue half-line is the set of all points with
polar coordinates θ = π4 , r > 0 and the pink half-line is the set of all points
with polar coordinates θ = π4 , r < 0. In the figure on the right below, the
blue half-line is the set of all points with polar coordinates θ = 5π 4 , r > 0
and the pink half-line is the set of all points with polar coordinates θ = 5π 4 ,
r < 0.
APPENDIX D. SOLUTIONS TO EXERCISES 450
y y
√ (1, 1) √ (1, 1)
2 2
5π
π 4
4
x x
√ √
√ distance from (−1, −1) to the origin is
(c) The 2. So either r = 2
or r = − 2.
√
• If r = 2, then θ must obey
√ √ 1
(−1, −1) = 2 cos θ , 2 sin θ ⇐⇒ sin θ = cos θ = − √
2
5π
⇐⇒ θ = + 2nπ, n integer
4
√
• If r = − 2, then θ must obey
√ √ 1
(−1, −1) = − 2 cos θ , − 2 sin θ ⇐⇒ sin θ = cos θ = √
2
π
⇐⇒ θ = + 2nπ, n integer
4
In the figure on the left below, the blue half-line is the set of all points with
polar coordinates θ = 5π 4 , r > 0 and the pink half-line is the set of all points
with polar coordinates θ = 5π 4 , r < 0. In the figure on the right below, the
blue half-line is the set of all points with polar coordinates θ = π4 , r > 0
and the pink half-line is the set of all points with polar coordinates θ = π4 ,
r < 0.
y y
5π
4 π
4
√ x √ x
2 2
(−1, −1) (−1, −1)
Here is a sketch of (xi , yi ), êr (θi ), êθ (θi ) for i = 1, 3, 5 (the points on the
axes)
y
er ( π2 )
eθ ( π2 ) eθ (0)
(0, 1)
eθ (π)
and here is a sketch (to a different scale) of (xi , yi ), êr (θi ), êθ (θi ) for
i = 2, 4 (the points off the axes).
er ( 3π ) y eθ ( π ) er ( π4 )
4 4
(−1, 1) (1, 1)
3π
eθ ( 3π
4
) 4 π
4
x
1.8.1.4. ∗. Solution. (a) Since −1 ≤ sin(4θ) ≤ 1, the coordinate
r = 2 + sin(4θ) oscillates between r = 1 and r = 3 as θ runs from 0 to
2π. The maximum value r = 3 is achieved when sin(4θ) = 1, i.e when
4θ = π2 + 2nπ, i.e. when θ = π8 + nπ 2 . That matches figure (E).
(b) Since −1 ≤ sin(4θ) ≤ 1, the coordinate r = 1 + 2 sin(4θ) takes its
maximum value r = 3 when sin(4θ) = 1, i.e. when θ = π8 + nπ 2 , just as the
case with (a). But now r can also take the value 0. That matches figure
(B).
(c) r = 1 is completely indepedent of θ. All points on the curve r = 1
are a distance 1 from the origin. That is, r = 1 is the circle of radius 1
centred on the origin. That’s figure (F).
(d) In this case, θ is subject to the restriction − π2 ≤ θ ≤ π2 , like figure
(C). Figure (C) looks like a circle. We can verify that r = 2 cos(θ) is indeed
a circle by converting to Cartesian coordinates. We can convert the right
hand side to exactly 2x = 2r cos(θ) by multiplying the whole equation by
r.
r = 2 cos(θ) ⇐⇒ r2 = 2r cos(θ) ⇐⇒ x2 + y 2 = 2x
APPENDIX D. SOLUTIONS TO EXERCISES 452
⇐⇒ (x − 1)2 + y 2 = 1
d θ/10 1 θ/10
+ e−θ/10 = − e−θ/10 > 0
e e for all θ > 0
dθ 10
r = eθ/10 + e−θ/10 increases as θ increases for all θ ≥ 0. Furthermore the
rate of increase gets bigger and bigger as θ gets bigger and bigger. So r
starts at r = 2 when θ = 0 and increases faster and faster as θ increases.
That matches figure (A).
(f) When θ = 0, r = θ = 0. As
d
θ=1 for all θ
dθ
r = θ increases as θ increases for all θ ≥ 0. Furthermore the rate of increase
is independent of θ. So r starts at r = 0 when θ = 0 and increases at a
constant rate as θ increases. That matches figure (D).
a(θ) = r00 (θ) = f 00 (θ)−f (θ) cos θ ı̂ı +sin θ ̂ +2f 0 (θ) − sin θ ı̂ı +cos θ ̂
The efficient way to compute |v(θ)| and the cross product v(θ) × a(θ) is to
observe that
v(θ) = f 0 (θ) êr (θ) + f (θ) êθ (θ)
a(θ) = f 00 (θ) − f (θ) êr (θ) + 2f 0 (θ) êθ (θ)
where êr (θ) and êθ (θ) are the vectors of Q[1.8.1.3]. As êr (θ) and êθ (θ)
are mutually perpendicular unit vectors obeying êr (θ) × êθ (θ) = k̂ and
êr (θ) × êr (θ) = êθ (θ) × êθ (θ) = 0,
|v(θ)|2 = v(θ)·v(θ) = f 0 (θ) êr (θ)+f (θ) êθ (θ) · f 0 (θ) êr (θ)+f (θ) êθ (θ)
= f 0 (θ)2 êr (θ) · êr (θ)+f (θ)2 êθ (θ) · êθ (θ)+2f 0 (θ) f (θ) êr (θ) · êθ (θ)
= f 0 (θ)2 + f (θ)2
p
|v(θ)| = f 0 (θ)2 + f (θ)2
v(θ)×a(θ) = f 0 (θ) êr (θ)+f (θ) êθ (θ) × f 00 (θ)−f (θ) êr (θ)+2f 0 (θ) êθ (θ)
= 2f 0 (θ)2 êr (θ) × êθ (θ) + f (θ)[f 00 (θ) − f (θ)] êθ (θ) × êr (θ)
= 2f 0 (θ)2 − f (θ)[f 00 (θ) − f (θ)] k̂
So
|v(θ) × a(θ)| f (θ)2 + 2f 0 (θ)2 − f (θ)f 00 (θ)
κ(θ) = =
|v(θ)|3 [f (θ)2 + f 0 (θ)2 ]
3/2
APPENDIX D. SOLUTIONS TO EXERCISES 453
we have
f (θ)2 + 2f 0 (θ)2 − f (θ)f 00 (θ)
κ(θ) =
[f (θ)2 + f 0 (θ)2 ]3/2
2
a − 2a2 cos θ + a2 cos2 θ + 2a2 sin2 θ − a2 cos θ + a2 cos2 θ
=
[a2 − 2a2 cos θ + a2 cos2 θ + a2 sin2 θ]3/2
3a − 3a2 cos θ
2
3 3
= 2 2 3/2
= 3/2 √ = p
[2a − 2a cos θ] 2 a 1 − cos θ 2 2ar(θ)
r(t) · Ω = 0
Then
dr dr
(t) = α(t) r(t) ⇐⇒ e−β(t) (t) − α(t)e−β(t) r(t) = 0
dt dt
d −β(t)
⇐⇒ e r(t) = 0
dt
⇐⇒ e−β(t) r(t) = r(0)
⇐⇒ r(t) = eβ(t) r(0)
so that r(t) lies on a line through the origin. This makes sense — the
particle is always moving parallel to its radius vector.
APPENDIX D. SOLUTIONS TO EXERCISES 454
dθ 1 rpt ` dtq
πr2 = r2 dθ
2π 2 dθ
rptq
(b) If f (r) is identically zero, then r00 (t) = 0, so that r0 (t) is a constant,
say v0 , and r(t) = r0 + v0 t, for some constant r0 . That’s a straight line.
We’ll now show that if the motion of the particle always lies on a straight
line, then f (r) must be identically zero. Suppose that r(t) is a straight line.
Then there are constant vectors r0 and T̂ such that r(t) = r0 + g(t)T̂ for
some scalar valued function g(t). Then
becomes
We may always choose the initial conditions so that, for example, r0 = ı̂ı
and T̂ = ̂. So
Taking the dot product of both sides with ı̂ı gives f r(t) = 0, as desired.
(c) We saw in §1.10 that the gravitational force − GM m
r 3 r can produce
elliptical orbits. So any f (r) which is a positive constant times − r13 does
the job.
1.9.1.2. ∗. Solution. (a) Our object is subject to a central force. So
the acceleration a(t) is parallel to r(t) and
d
r(t) × v(t) = v(t) × v(t) + r(t) × a(t) = 0 + 0 = 0
dt
and, in particular,
|r(t) × v(t)| = r(t)2 |θ̇(t)|
is a constant. As θ̇(t) is continuous, r(t)2 θ̇(t) is also constant.
(b) By Lemma 1.8.2, the acceleration
d2 r dθ 2
a(t) = 2
(t) − r(t) (t) r̂ θ(t)
dt dt
d2 θ dr dθ
+ r(t) 2 (t) + 2 (t) (t) θ̂θ θ(t)
dt dt dt
Because our object is subject to a central force, the acceleration a(t) is
parallel to r̂ θ(t) . So the θ̂θ θ(t) component of the acceleration is zero
and d2 r dθ 2
a(t) = 2
(t) − r(t) (t) r̂ θ(t)
dt dt
so that d2 r dθ 2
|a(t)| = 2 (t) − r(t) (t)
dt dt
1
Since r(t) = θ(t)+α
1
ṙ(t) = − θ̇(t) = −r(t)2 θ̇(t) = −h
[θ(t) + α]2
d2 r
So dt2 (t) = 0 and
r(t)4 θ̇(t)2 h2
|a(t)| = r(t)θ̇(t)2 = 3
=
r(t) r(t)3
2 · Vector Fields
2.1 · Definitions and First Examples
2.1.1 · Exercises
2.1.1.1. Solution. The vectors are pointing to the right when x > 0, to
the left when x < 0, and are vertical when x = 0. So, at least for (x, y)
shown in the sketch,
> 0
when x > 0
v(x, y) · ı̂ı = 0 when x = 0
<0 when x < 0
• up when 2 < x.
APPENDIX D. SOLUTIONS TO EXERCISES 456
2.1.1.2. Solution. To start out, we find the places where v(x, y) · ı̂ı = 0
(vertical vectors) or v(x, y) · ̂ = 0 (horizontal vectors). Remember the
vector v(x, y) has its tail at (x, y).
We see the vertical vectors (those with v(x, y) · ı̂ı = 0) occur at every
point along the line y = −x, while horizontal vectors (those with v(x, y)·̂ =
0) occur at every point along the line y = x.
Indeed, below the line y = −x, vectors point to the left, while above the
line y = −x they point to the right. Similarly, vectors point down when
they’re above the line y = x, and the point up when they’re below the line
y = x.
y
RIGHT
LEFT
y
DOWN
UP
APPENDIX D. SOLUTIONS TO EXERCISES 457
2.1.1.3. Solution. Since all conveyors point towards the origin, the
(−x,−y)
direction of motion of an object at location (x, y) is √ 2 2
. Its magnitude
x +y
is |y|, so v(x, y) = √−|y|
2 2
(x, y).
x +y
2.1.1.4. Solution. The arrows near the point A are pointing to the right,
indicating that P > 0, and upward, indicating that Q > 0. Moving from
left to right near A, the vertical component of the arrows is decreasing,
indicating that ∂Q
∂x < 0. Moving vertically upwards near A, the vertical
component of the arrows is increasing, indicating that ∂Q
∂y > 0.
2.1.1.5. Solution. (a) At time 0 the velocity of the twig is v(1, 1) = ı̂ı +̂.
So at time t = 0.01, the position of the twig is approximately
(0, 0) + 0.01(0, 0) = (0 , 0)
2.1.1.7. Solution. Since all conveyors point towards the origin, the
(−x,−y)
direction of motion of an object at location (x, y) is √ 2 2
. Its magnitude
x +y
is y, so v(x, y) = √ −y
2 2
(x, y).
x +y
α
with length √ for some positive constant α. That is,
x2 +y 2 +z 2
α
v(x, y, z) = − (x, y, z)
x2 + y2 + z2
2.1.1.9. Solution. Beginning as in Example 2.1.4, we note
(
2 > 0 x 6= 0
v(x, y) · ı̂ı = x
=0 x=0
and
> 0
y>0
v(x, y) · ̂ = y =0 y=0.
<0 y<0
That leads to the following picture:
y
This gives us a general idea to start with. Refining, we notice that when
x2 > |y|, then the vector v(x, y) will be more horizontal than vertical. As
we move away from the y-axis in a horizontal line, the difference between
x2 and |y| grows, so the vectors get more and more horizontal. However,
for a fixed value of x, vectors farther from the axis will be more vertical
than vectors closer to it.
APPENDIX D. SOLUTIONS TO EXERCISES 459
When (x, y) is far from the origin, its distance from (0, 0) is almost the
same as its distance from (1, 0). So, we expect v(x, y) to be approximately
a scalar multiple of (1, 1).
At (0, 0), v(0, 0)·ı̂ı = 0, so our vector is horizontal; similarly, v(1, 1)·̂ = 0
so this vector is horizontal. Vectors very near to (0, 0) are nearly horizontal,
while vectors near to (1, 1) are nearly vertical.
APPENDIX D. SOLUTIONS TO EXERCISES 460
For the direction field, we normalize our vectors to have unit length.
y
y = −x
Similarly, the sign of v(x, y) · ̂ = y(y − x) depends on the signs of y
and y − x.
y=x
All together:
y
y=x
↑ ↑
← →
↑
→ →
↓
x
↑
→ →
↓
↑ ↑
→ ←
y = −x
Refining, we notice that as we move straight up or down, |v(x, y) ·ı̂ı| has
its minimum along the lines y = −x and x = 0. So, the vectors become
more strongly vertical as we approach y = −x and x = 0 from above or
below.
Similarly, |v(x, y) · ̂| has its minima along the lines y = x and y = 0,
so the vectors become more strongly horizontal as we approach y = x
horizontally.
APPENDIX D. SOLUTIONS TO EXERCISES 462
2.1.1.12. Solution. The field v(x, y) is the sum, scaled by 1/3, of the
unit vector pointing away from the origin and the unit vector pointing away
from (1, 0). This tells us about a few regions:
• Along the x axis between (0, 0) and (1, 0), the vectors away from
these points are pointing in opposite directions (and have the same
length), so they cancel each other out. That is, v(x, 0) = 0 for all
x ∈ (0, 1).
• As the distance from (x, y) to the origin grows, the vector pointing
away from (0, 0) looks more and more like the vector pointing away
from (1, 0). So, our vectors far away from the origin look like vectors
of length about 2/3, pointing away from the origin.
APPENDIX D. SOLUTIONS TO EXERCISES 463
2.1.1.13. Solution. (a) The vector field v(x, y) = xı̂ı + y ̂ is the same
as the radius vector. It points radially outward and has length growing
linearly with the distance from the origin.
y
(b) The vertical component of v(x, y) = 2xı̂ı − ̂ is always −1. Its
horizontal component is 2x, so that
• v(x, y) is rightward pointing when x > 0 and leftward pointing when
x < 0, and
• the magnitude of the horizontal component grows linearly with the
distance from the y-axis.
It is sketched in the figure on the left below.
APPENDIX D. SOLUTIONS TO EXERCISES 464
y y
x x
• is of length 1 and
2.1.1.15. Solution.
a. Consider a point P on the pole that is a distance p away from the
bottom end. Use this point to make a smaller right triangle, as in
the picture below.
y
H P p
h
x
b
√
4 − H2
APPENDIX D. SOLUTIONS TO EXERCISES 465
When H = 1:
dy p dx p 1
=− = 1 − √
dt H=1
4 dt H=1 2 2 3
Therefore,
dx dy p 1 p
v(p) = , = 1− √ ,−
dt dt H=1 2 2 3 4
For our model, we set the domain of this function to be [0, 2].
b. Let’s start by seeing what we can salvage from our work on part a.
As in part a., consider a point P on one of the poles, p metres from
the bottom end.
(0, 0, H)
2
p
P = (x, y, z)
p
x2 + y 2
√
4 − H2
Let P have position (x, y, z). Noting that dH dt is now positive, not
negative, if we stick to this two-dimensional slice,
p −1 p
V(p) = 1− √ ,
2 2 3 4
(0, 0, H)
(x, y, z)
(0, 0, 0)
p
(x, y, 0)
c · (x, y, 0)
dz p dx dy
So, we know = , and we know , = (x, y)c
dt H=1 4 dt dt H=1
for some negative constant c with |(x, y)c| = 1 − p2 2√ 1
3
. Since we
have the direction and the magnitude of the vector, we can find the
vector:
1 − p2
dx dy
, = (x, y)c = − √ p (x, y)
dt dt H=1 2 3 x2 + y 2
Finally:
dx dy dz 1 1 1
V(x, y, z) = , , = − x, − y, z
dt dt dt H=1 6 6 2
Not all values of (x, y, z) are on the frame. But, for those values of
(x, y, z) that are on the frame, this equation holds.
y
3
x
1 2 3
(b) At every point of the positive x-axis, the velocity vector v(x, 0)
points straight to the right. So a rubber ducky placed in the water at (1, 0)
just floats rightward along the positive x-axis.
(c) At every point of the first quadrant away from the axes, the velocity
vector v(x, y) points downwards and towards the right. So a rubber ducky
placed in the water at (1, 2) always floats down and to the right. The closer
the ducky gets to the x--axis the more rightward its motion becomes.
2.2.2.2. Solution. The derivatives
dx
= v1 (x, y) = −x − y
dt
dy
= v2 (x, y) = x − y
dt
So the vector field is v(x, y) = v1 (x, y) , v2 (x, y) = (−x − y , x − y).
is visually displayed in the figure on the left below. The arrows in the figure
2 2
on the left gives us the direction of motion along the field lines x2 = y2 + C
(in red) in the figure on the right below. Some equipotential curves xy = C
are also sketched (in blue) in the figure on the right below.
y
In particular
dx y 2 dy 1 1
= =⇒ x dx = 2y 3 dy =⇒ x2 = y 4 + C
2y x 2 2
x = y2 z = ey
2.2.2.5. ∗. Solution. The field lines obey
dx dy
= if x, y 6= 0
x 3y
=⇒ 3 ln |x| = ln |y| + C
=⇒ |x|3 = eC |y|
=⇒ y = ±e−C x3
=⇒ y = C 0 x3
F = xı̂ı + ẑ + y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + (0 − 0)k̂ = 0
This field passes the screening test. That means the screening test doesn’t
rule out the possibility of F being conservative. So, we have option C.
b.
F = y 2 zı̂ı + x2 ẑ + x2 y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (x2 − x2 )ı̂ı + (y 2 − 2xy)̂ + (2xz − 2yz)k̂ 6= 0
So, F fails the screening test — it’s not conservative. That’s option B.
c.
1
F = (yexy + 1)ı̂ı + (xexy + z)̂ + + y k̂
z
∂F ∂F ∂F ∂F ∂F ∂F1
3 2 1 3 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + (exy (xy + 1) − exy (xy + 1))k̂ = 0
F = y cos(xy)ı̂ı + x sin(xy)̂
∂F2
= xy cos(xy) + sin(xy)
∂x
∂F1
= −xy sin(xy) + cos(xy)
∂y
∂F2 ∂F1
6=
∂x ∂y
F fails the screening test, so it is not conservative. That is Option B.
APPENDIX D. SOLUTIONS TO EXERCISES 470
∂ϕ − xy2 y
(x, y) = =− 2
y 2
∂x 1+ x x + y2
1
∂ϕ x x
(x, y) = 2 =
∂y 1 + xy x2 + y 2
so that F = ∇ ϕ.
2.3.1.6. Solution. If ϕ is a potential for F, then:
∂ϕ
• ∂x = x + y, so ϕ = 12 x2 + xy + ψ1 (y)
∂ϕ
• ∂y = x − y, so ϕ = xy − 12 y 2 + ψ2 (x)
1
So, for instance, ϕ(x, y, z) = 2 ln(x2 + y 2 + z 2 ) is a potential for F(x, y, z).
2.3.1.10. Solution. (a) We shall show that F(x, y, z) is conservative by
finding a potential for it. ϕ(x, y, z) is a potential for this F if and only if
∂ϕ
(x, y, z) = x
∂x
∂ϕ
(x, y, z) = −2y
∂y
∂ϕ
(x, y, z) = 3z
∂z
Integrating the first of these equations gives
x2
ϕ(x, y, z) = + f (y, z)
2
Substituting this into the second equation gives
∂f
(y, z) = −2y
∂y
which integrates to
f (y, z) = −y 2 + g(z)
x2
Finally, substituting ϕ(x, y, z) = 2 − y 2 + g(z) into the last equation gives
g 0 (z) = 3z
which integrates to
3 2
g(z) = z +C
2
with C being an arbitrary constant. So, F(x, y, z) is conservative and
ϕ(x, y, z) = 12 x2 − y 2 + 23 z 2 is one allowed potential.
(b) The field F = F1 ı̂ı + F2 ̂ can be conservative only if it passes the
screening test
∂F1 ∂F2
=
∂y ∂x
In this case
∂F1 ∂ x 2xy
= =− 2
∂y ∂y x2 + y 2 (x2 + y 2 )
is different from
∂F2 ∂ −y 2xy
= = 2
∂x ∂x x2 + y 2 (x + y 2 )
2
∂ (z2 ) ∂ 2 2 2
Axze(z ) + 3By 2 z 2 2ze(z )
= Aze(z )
e = ⇐⇒
∂z ∂x
∂ ∂ 2 2 2
2Byz 3 = Axze(z ) + 3By 2 z 2 6Bye(z )
= 6Bye(z )
⇐⇒
∂z ∂y
Hence only A = 2 works. We shall see in part (b) that any B works.
(b) When A = 2, and B is any real number.
2 2
F = e(z ) ı̂ı + 2Byz 3 ̂ + 2xze(z )
+ 3By 2 z 2 k̂
∂ϕ 2
(x, y, z) = e(z )
∂x
∂ϕ
(x, y, z) = 2Byz 3
∂y
∂ϕ 2
(x, y, z) = 2xze(z ) + 3By 2 z 2
∂z
Integrating the first of these equations gives
2
ϕ(x, y, z) = xe(z )
+ f (y, z)
which integrates to
g(z) = C
with C being an arbitrary constant. So, for each real number B, ϕ(x, y, z) =
2
xe(z ) + By 2 z 3 is one allowed potential.
2.3.1.12. Solution. In each second 2πm cm2 of fluid crosses each circle
of radius r (and hence circumference 2πr) centred on the origin. So the
speed of flow at radius r is m
r . As the direction of flow is radially outward
xı̂ı + ŷ
v=m
x2 + y 2
∂ϕ x
(x, y) = m 2
∂x x + y2
∂ϕ y
(x, y) = m 2
∂y x + y2
APPENDIX D. SOLUTIONS TO EXERCISES 473
ϕ = 21 m ln(x2 + y 2 )
2.3.1.13. Solution. Following Example 2.3.3, the particle can never
escape the region {(x, y, z) : ϕ(x, y, z) ≥ −E}. So, we should find E, then
figure out the region.
The kinetic energy of the particle is 12 m|v|2 , so the total energy of
the system (also the kinetic energy when the potential energy is 0) is
1 2
2 (10)(2 ) = 20 J.
Therefore, a region it can never escape is
(x, y, z) ϕ(x, y, z) ≥ −20
that is,
(x, y, z) x2 + y 2 + z 2 ≤ 20
√
So, it can never escape the sphere centred at the origin with radius 20.
1 2
2.3.1.14. Solution. Example 2.3.3 tells us 2 m|v(t)| −ϕ
x(t), y(t), z(t) =
E is a constant quantity, provided F is conservative with potential ϕ. So,
it would be nice if F were conservative.
If F = ∇ ϕ, then
∂ϕ
• ∂x = 0, so ϕ = ψ1 (y, z)
∂ϕ
• ∂y = 1, so ϕ = y + ψ2 (x, z)
∂ϕ
• ∂z = 3z 1/3 , so ϕ = 94 z 4/3 + ψ3 (x, y)
|v|2
1 1 1 2 9
= · |v| − ϕ(1, 1, 1) = − 1+
4 2 2 4 4
√
|v| = 14
√
So, at the point (1, 1, 1), the particle has speed 14.
APPENDIX D. SOLUTIONS TO EXERCISES 474
2.3.1.15. Solution. We can start with the screening test, Theorem 2.3.9.
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= g 0 (y)h0 (z) − g 0 (y)h0 (z) ı̂ı + 0 − 0 ̂ + 0 − 0 k̂ = 0
So, it’s possible that the field is conservative. Remember, this test alone
isn’t enough to tell us it’s conservative. (Had the test come out differently,
though, we’d be done.)
Suppose F = ∇ ϕ(x, y, z). Then:
∂ϕ
• ∂x = 2f (x)f 0 (x). By inspection, we 2
R see ϕ 0= f (x) + ψ1 (y, z). (We
could also find this by evaluating 2f (x)f (x)dx with the substitu-
tion u = f (x).)
∂ϕ
• ∂y = g 0 (y)h(z), so ϕ = g(y)h(z) + ψ2 (x, z).
∂ϕ
• ∂z = g(y)h0 (z), so ϕ = g(y)h(z) + ψ2 (x, y).
When the curl is 0ı̂ı +0̂ +0k̂, we have x = 2y and x = z. That is, our points
are of the form (2c, c, 2c) for any constant c. So, the region in question is
the line through the origin in the direction of the vector (2, 1, 2).
dy=0
(0, 1) y=1 (1, 1)
dx=0 dx=0
x=0 x=1
F = xı̂ı + ẑ + y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + (0 − 0)k̂ = 0
This field passes the screening test. Since F is defined and has continuous
first-order partial derivatives on all of R3 , it is conservative. So, we have
option A.
b.
F = y 2 zı̂ı + x2 ẑ + x2 y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (x2 − x2 )ı̂ı + (y 2 − 2xy)̂ + (2xz − 2yz)k̂ 6= 0
So, F fails the screening test. So, it’s not conservative. That’s option B.
c.
F passes the screening test. Since F is defined and has continuous first-
order partial derivatives on all of R3 , it is conservative. So, we have option
A.
d.
F = y cos(xy)ı̂ı + x sin(xy)̂
∂F2
= xy cos(xy) + sin(xy)
∂x
APPENDIX D. SOLUTIONS TO EXERCISES 476
∂F1
= −xy sin(xy) + cos(xy)
∂y
∂F2 ∂F1
6=
∂x ∂y
F fails the screening test, so it is not conservative. That is Option B.
R
2.4.2.3. Solution. Since F is conservative, C F · dr = 0 over any closed
curve C. The given curve is closed, so the integral is simply zero.
R and A and
2.4.2.4. Solution. Since F is conservative, R B start and end
at the same points, by path-independence B F · dr = A F · dr = 5.
3
2.4.2.5. ∗. Solution. Note F R is defined and continuous on all of R .
By Theorem 2.4.7, the integral C F · dr = 0 for all closed paths C if and
only if F is conservative. Furthermore, F has continuous first-order partial
derivatives on all of R3 . Using Theorem 2.4.8, F is conservative if and only
if it passes the scsreening test ∇ × F = 0:
ı̂ı ̂ k̂
∂ ∂ ∂
0 = ∇ × F = det ∂x ∂y ∂z
ex sin y aex cos y + bz cx
= (0 − b)ı̂ı − (c − 0)̂ + (aex cos y − ex cos y)k̂
L1
C
x
L2 L4
L3 x
(b) Consider the square in the figure (b) on the right above, oriented
counterclockwise. It consists of the fourline segments L1 , L2 , L3 and L4 .
On all of L1 , L2 , L3 we have that F r(t) ·r0 (t) = 0 because the vector field
is perpendicular to the line segment. On L4 we have F r(t) · r0 (t) > 0. So
I Z Z Z Z
F · dr = F · dr + F · dr + F · dr + F · dr
C L1 L2 L3 L4
Z
=0+0+0+ F · dr > 0
L4
H
So C is a closed curve with C F · dr > 0 and F is not conservative.
(c) Consider the square in the figure (c) on the left below, oriented
counterclockwise. It consists of the four line segments L1 , L2 , L3 and L4 .
APPENDIX D. SOLUTIONS TO EXERCISES 477
On L1 and L3 we have that the dot product F r(t) · r0 (t) = 0 because the
vector field is perpendicular to the line segment. On L2 we have F r(t) ·
r0 (t) < 0 while on L4 we have F r(t) 0
0· r (t) > 0. The vector field F is
longer on L4 than on L2 . So F r(t) · r (t) has a larger magnitude on L4
than L2 and
I Z Z Z Z
F · dr = F · dr + F · dr + F · dr + F · dr
C L1 L2 L3 L4
Z Z
=0+ F · dr + 0 + F · dr > 0
L2 L4
H
So C is a closed curve with C
F · dr > 0 and F is not conservative.
(c) y (d) y
L1
L2 L4
L3 x x
(d) We are told that one of the four vector fields is conservative. Only
the vector field in (d) is left, so it is conservative.
Remark: We can verify that vector field (d) is indeed conservative by
observing (look at the figure (d) on the right above) that the ı̂ı component
of the vector field is exactly zero and that the ̂ component depends only
on y. So the vector field is of the form
−2x2 + 2y 2 − 2xy
∂ x − 2y −2 2y(x − 2y)
2 2
= 2 2
− 2 = 2
∂y x + y x +y (x2 + y 2 ) (x2 + y 2 )
−2x2 + 2y 2 − 2xy
∂ 2x + y 2 2x(2x + y)
= − 2 = 2
∂x x2 + y 2 x2 + y 2 (x2 + y 2 ) (x2 + y 2 )
So the curl of F is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det
∂x ∂y ∂z
x−2y 2x+y
x2 +y 2 x2 +y 2 z
!
−2x2 + 2y 2 − 2xy −2x2 + 2y 2 − 2xy
= 2 − 2 k̂
(x2 + y 2 ) (x2 + y 2 )
APPENDIX D. SOLUTIONS TO EXERCISES 478
=0
on the domain of F.
(c) Parametrize the circle by
r(t) = 2 cos tı̂ı + 2 sin t ̂ + 3 k̂ r0 (t) = −2 sin tı̂ı + 2 cos t ̂
(d) As the integral of F around the simple closed curve C is not zero, F
cannot be conservative. See Theorem 2.4.7 and Examples 2.3.14 and 4.3.8.
2.4.2.8. Solution. The point here is that F is conservative, as F = ∇φ
with
x2 z2
+ yx − yz +
φ=
2 2
So, for all paths from r(t0 ) = (1, 0, −1) to r(t1 ) = (0, −2, 3),
Z
F · dr = φ r(t1 ) − φ r(t0 ) = φ(0, −2, 3) − φ(1, 0, −1)
C
9 1 1
= 0+0+6+ − +0−0+
2 2 2
1
=9
2
So
Z Z 1 Z 1 Z 1
1 1
F · dr = x dx + y dy + (−2) dz = + − 2 = −1
C 0 0 0 2 2
2.4.2.11. ∗. Solution. Parametrize the curve using y as a parameter.
8
Then y = t, x = 2y = 2t and z = xy = 2t82 so that:
4
r(t) = 2tı̂ı + t ̂ + k̂, 1≤t≤2
t2
8
r0 (t) = 2ı̂ı + ̂ −
k̂
t3
F(r(t)) = 4t2 ı̂ı + 4t3 k̂
F(r(t)) · r0 (t) = 8t2 − 32
Then
Z Z 2 Z 2 2
8 3
F(r(t)) · r0 (t) dt = 8t2 − 32 dt =
F · dr = t − 32t
C 1 1 3 1
40
=−
3
2.4.2.12. Solution. (a), (b) The curls of F and G are
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
6x2 yz 2 2x3 z 2 + 2y − xz 4x3 yz
APPENDIX D. SOLUTIONS TO EXERCISES 480
so that r(0) = (0, 0, 1) is the initial point of the line segment and r(1) =
(2, 1, 0) is the final point of the segment. Then
so that
Z Z π/4 Z π/4
F r(θ) · r 0 (θ) dθ = sin θ cos2 θ dθ
Work = F · dr =
P 0 0
1 π/4
= − cos3 θ
3 0
1 1
= 1 − 3/2 ≈ 0.2155
3 2
2.4.2.15. ∗. Solution. Hmmm. F looks suspiciously complicated. Let’s
guess that F is conservative and look for a potential for it. φ(x, y, z) is a
potential for this F if and only if
∂ϕ
(x, y, z) = yz cos x
∂x
∂ϕ
(x, y, z) = z sin x + 2yz
∂y
∂ϕ
(x, y, z) = y sin x + y 2 − sin z
∂z
Integrating the first of these equations gives
ϕ(x, y, z) = yz sin x + f (y, z)
Substituting this into the second equation gives
∂f ∂f
z sin x + (y, z) = z sin x + 2yz or (y, z) = 2yz
∂y ∂y
APPENDIX D. SOLUTIONS TO EXERCISES 482
which integrates to
f (y, z) = y 2 z + g(z)
Finally, substituting ϕ(x, y, z) = yz sin x + y 2 z + g(z) into the last equation
gives
which integrates to
g(z) = cos z + C
with C being an arbitrary constant. So φ(x, y, z) = yz sin x + y 2 z + cos z
is one allowed scalar potential and the specified integral is
π3 π2
Z r(π/2)
F · dr = ϕ(r) = ϕ(π/2 , π/2 , π/2) − ϕ(0, 0, 0) = + −1
C r(0) 8 4
(x − 1)2 + y 2 = 1
θ
x
1, 0
we parametrize C by
r(θ) = x(θ) , y(θ) = (1 − cos θ)ı̂ı + sin θ ̂ 0≤θ≤π
So the integral is
Z Z π Z π
xy dy = x(θ) y(θ) y 0 (θ) dθ = (1 − cos θ) sin θ cos θ dθ
C 0 0
sin y)ı̂ı + (ex + sin y + x cos y) ̂. We are to show that it is independent of
path. That is the case if and only if F is conservative. So let’s look for a
potential ϕ for F. That is, let’s look for a function ϕ that obeys
∂ϕ
(x, y) = yex + sin y
∂x
∂ϕ
(x, y) = ex + sin y + x cos y
∂y
Integrating the first of these equations gives
which integrates to
f (y) = − cos y + C
So F is indeed conservative with one potential being ϕ(x, y) = yex +x sin y−
cos y and the line integral is
Z Z (0,π/2)
x x
(ye + sin y) dx + (e + sin y + x cos y) dy = F · dr = ϕ(x, y)
C C (1,0)
h i(0,π/2)
= yex + x sin y − cos y
(1,0)
π
=1+
2
2.4.2.18. ∗. Solution. Here is a sketch of C.
z
(0, 0, 1)
C1 C2
(0, 1, 0)
y
(1, 0, 0) C3
x
Note that
• y = 0 on the
R line segment from (1, 0, 0) to (0, 0, 1) so that the integral
reduces to zx dz on that line segment and
• x = 0 on the line segment from (0, 0, 1) to (0, 1, 0) so that the integral
APPENDIX D. SOLUTIONS TO EXERCISES 484
R
reduces to yz dy on that line segment and
• z = 0 on the
R line segment from (0, 1, 0) to (1, 0, 0) so that the integral
reduces to xy dx on that line segment.
So it looks feasible to evaluate the integral directly. Label the sides of the
triangle C1 , C2 and C3 as in the sketch above.
• We parametrize C1 by r(t) = (1, 0, 0) + t[(0, 0, 1) − (1, 0, 0)] = (1 −
t , 0 , t), 0 ≤ t ≤ 1. So
x z z 0 (t)
Z Z Z 1 z }| { z}|{ z}|{
xy dx + yz dy + zx dz = zx dz = (1 − t) (t) (1) dt
C1 C1 0
Z 1
= (t − t2 ) dt
0
1 1 1
= − =
2 3 6
y z y 0 (t)
Z Z Z 1 z}|{ z }| { z}|{
xy dx + yz dy + zx dz = yz dy = (t) (1 − t) (1) dt
C2 C2 0
Z 1
= (t − t2 ) dt
0
1 1 1
= − =
2 3 6
x y x0 (t)
Z Z Z 1 z}|{ z }| { z}|{
xy dx + yz dy + zx dz = xy dx = (t) (1 − t) (1) dt
C3 C3 0
Z 1
= (t − t2 ) dt
0
1 1 1
= − =
2 3 6
All together
Z 3 Z
X 1 1
xy dx + yz dy + zx dz = xy dx + yz dy + zx dz = 3 × =
C C` 6 2
`=1
The first equation forces ϕ(x, y, z) = xy + zex + ψ(y, z). Substituting this
into the second equation gives x + ∂ψ y ∂ψ
∂y (y, z) = x + e sin z or ∂y (y, z) =
y y
e sin z which forces ψ(y, z) = e sin z + ζ(z). So far, we have ϕ(x, y, z) =
xy + zex + ey sin z + ζ(z). Substituting this into the third equation gives
2
ex +ey cos z+ζ 0 (z) = z+ex +ey cos z or ζ 0 (z) = z which forces ζ(z) = z2 +C,
for some constant C, which we take to be zero. So our potential is
z2
ϕ(x, y, z) = xy + zex + ey sin z +
2
So the line integral
Z
F · dr = ϕ r(π) − ϕ r(0) = ϕ π, eπ , 0 − ϕ 0, 1, 0 = πeπ
C
F = ∇ sin x + 2y − cos y + ez
= −4
2.4.2.22. ∗. Solution. (a) The curl is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x =0
∂y ∂z
z + ey xey − e sin y
z
1 + x + ez cos y
so F passes the screening test. Since its first-order partial derivatives are
continuous on all of R3 , it is conservative by Theorem 2.4.8 in the text.
By inspection, the potential is ϕ(x, y, z) = xz + xey + ez cos y + z —
APPENDIX D. SOLUTIONS TO EXERCISES 487
We’d like a single function ϕ(x, y, z) that simultaneously obeys all three of
these equations. An initial guess is simply the sum of the distinct terms
(without the ψj ’s) that appear in the equations above:
?
ϕ(x, y, z) = zx + xey + ez cos y + z
= π 2 + π 2 + e + 1 − (0 + 0 + e + 1) = 2π 2
which forces
g(x, z) = h(x)
Finally, substituting f (x, y, z) = xyex + yz 3 + h(x) into the first equation
gives
xyex + yex + h0 (x) = (x + 1)yex or h0 (x) = 0
So h(x) = C and hence f (x, y, z) = xyex + yz 3 + C works for any constant
C.
(c) Since F = ∇ f ,
Z Z
F · dr = ∇ f · dr = f r(π)) − f ( r(0) = f (π, 1, −1) − f (0, 1, 1)
C
C π
= πe − 1 − 1 = πeπ − 2
(d) Since
Z Z
F · dr = (x + 1)yex dx + (xex + z 3 ) dy + 3yz 2 dz
C C
we have
Z Z
I= F · dr + yz 2 dz
C C
y z2 dz
Z π z }| { z }| { z }| {
= πeπ − 2 + 2
(cos 2t) cos t (− sin t) dt
Z0 π
= πeπ − 2 + (2 cos2 t − 1) cos2 t(− sin t) dt
0
Z −1
= πeπ − 2 + (2u2 − 1)u2 du with u = cos t, du = − sin t dt
1
u3 i−1
h 2u5
= πeπ − 2 + −
5 3 1
h 4 2i
π
= πe − 2 + − +
5 3
π 32
= πe −
15
∂ϕ
(x, y, z) = 2xye3z + 3x2 y 2
∂y
=⇒ ϕ(x, y, z) = xy 2 e3z + x2 y 3 + ψ2 (x, z)
∂ϕ
(x, y, z) = 3xy 2 e3z
∂z
=⇒ ϕ(x, y, z) = xy 2 e3z + ψ3 (x, y)
So
Z
(y 2 e3z + xy 3 ) dx + (2xye3z + 3x2 y 2 ) dy + 3xy 2 e3z dz
C
Z Z
= F · dr − xy 3 dr
C C
3
xy ı̂ı r0 (t)
Z 1z }| { z }| {
2t −t 3
2t −t 1
= ϕ r(1) − ϕ r(0) − e e ı̂ı · 2e ı̂ı − e ̂ + k̂ dt
0 1+t
Z 1
2 1
2et dt
= ϕ e , , ln 2 − ϕ 1, 1, 0 −
e 0
1 2 3 ln 2 1 3
= e2 + e4
e − 1 + 1 − 2(e − 1)
e e
= 23 + e − 2 − 2e + 2
=8−e
2.4.2.25. ∗. Solution. (a) The field is conservative only if
∂F1 ∂F2 ∂F1 ∂F3 ∂F2 ∂F3
= = =
∂y ∂x ∂z ∂x ∂z ∂y
That is,
∂ ∂
(2x sin(πy) − ez ) = ax2 cos(πy) − 3ez
∂y ∂x
⇐⇒ 2πx cos(πy) = 2ax cos(πy)
∂ ∂
(2x sin(πy) − ez ) = − (x + by) ez
∂z ∂x
⇐⇒ −ez = −ez
∂ ∂
ax2 cos(πy) − 3ez = − (x + by) ez
∂z ∂y
⇐⇒ −3ez = −bez
Summing the distinct terms on the right hand sides of the three equations
above, we guess
= −8
To evaluate the remaining integral, parametrize the curve by r(t) = tı̂ı +t̂ +
ln(1 + t)k̂ with 0 ≤ t ≤ 1. Then r0 (t) = ı̂ı +̂ + 1+t
1
k̂ and 3yez k̂ = 3t(1 + t)k̂
z
so that 3ye k̂ · dr = 3t dt. Subbing in
Z Z 1
3 13
G · dr = −8 + 3t dt = −8 + = −
C 0 2 2
2.4.2.26. ∗. Solution. (a) The potential f must obey
∂f
(x, y, z) = −2y cos x sin x
∂x
∂f
(x, y, z) = cos2 x + (1 + yz)eyz
∂y
∂f
(x, y, z) = y 2 eyz
∂z
Integrating the last of these equations with respect to z gives
+ zeyz − zeyz k̂
=0
∂ϕ
(x, y, z) = xyeyz + ey =⇒ ϕ(x, y, z) = xeyz + zey + ψ3 (x, y)
∂z
All together, ϕ(x, y, z) = xeyz + zey + C works for any constant C. So the
specified work integral is
Z
πe
F · dr = ϕ r(π/2) − ϕ r(0) = ϕ 0, 1, π/2 − ϕ 1, 0, 0 = −1
C 2
2.4.2.29. ∗. Solution. (a), (b) The function f (x, y) is a potential for
F(x, y) if and only if it obeys
∂f
(x, y) = 2xy cos(x2 )
∂x
∂f
(x, y) = sin(x2 ) − sin(y)
∂y
Integrating the first of these equations gives
which integrates to
g(y) = cos(y) + C
with C an arbitrary constant. Hence f (x, y) = y sin(x2 ) + cos(y) + C is a
potential for any constant C. Because F has a potential, it is conservative.
(c) We may parametrize C by
π
r(t) = sin(t)ı̂ı + t ̂ ≤t≤π
2
As f (x, y) = y sin(x2 ) + cos(y) is a potential for F
Z
π π
F · dr = f r(π) − f r( ) = f 0, π − f 1,
C 2 2
π
= −1 − sin(1)
2
π
= −1 − sin(1)
2
2.4.2.30. ∗. Solution. (a) The stated integral property is characteristic
of conservative fields (Theorem 2.4.7). Since all partial derivatives of F are
defined on all of R3 , an equivalent property is
ı̂ı ̂ k̂
∂ ∂ ∂
0=∇×F=
∂x ∂y ∂z
(mxyz + z 2 − ny 2 ) (x2 z − 4xy) (x2 y + pxz + qz 3 )
14 14
√
Z Z
2 3/2
ρ ds = u du = u
C 1 3 1
APPENDIX D. SOLUTIONS TO EXERCISES 494
2 3/2
= 14 − 1 ≈ 34.26
3
(b) Since F(x, y, z) = ∇f (x, y, z) with f (x, y, z) = x sin y + yz + 12 z 2 ,
Z
3
F · dr = f (1, 1, 1) − f (0, 0, 0) = sin 1 + ≈ 2.3415
C 2
2.4.2.33. ∗. Solution. (a) Let’s evaluate the integral directly using the
parametrization
r(x) = xı̂ı + (9 − x2 ) ̂
with −3 ≤ x ≤ 3.
Since r0 (x) = ı̂ı − 2x ̂,
dy
y dx
Z Z 3 z }| { z }| {
(x2 + y) dx + x dy = x2 + 9 − x2 +x (−2x) dx
C −3
Z 3
9 − 2x2 dx
=
−3
3
33
Z
9 − 2x2 dx = 2 27 − 2
=2
0 3
= 18
(b) In this solution, we’ll evaluate the integral directly. Label the four
sides of the square L1 , L2 , L3 and L4 as in the figure
y
(0, 1) (1, 1)
L3
L4 L2
L1
x
(1, 0)
All together
Z Z Z Z Z
F · n̂ ds = F · n̂ ds + F · n̂ ds + F · n̂ ds + F · n̂ ds
C L1 L2 L3 L4
= 0 + 2 + (e − 1) + 0 = e + 1
APPENDIX D. SOLUTIONS TO EXERCISES 496
Integating gives
v(t) = t ̂ + cos t k̂ + c
for some constant vector c. Since v(0) = ı̂ı + k̂, we have c = ı̂ı so that
t2
r(t) = tı̂ı + ̂ + sin t k̂ + c
2
for some (new) constant vector c. Since r(0) = ̂, we have c = ̂ so that
t2
r(t) = tı̂ı + 1 + ̂ + sin t k̂
2
π2
π
r1 = r(π/2) = ı̂ı + 1 + ̂ + k̂
2 8
Solution 1:
APPENDIX D. SOLUTIONS TO EXERCISES 497
(2, 2)
L
(1, 1)
L1
L3
x
(1, Y ) (2, Y )
L2
Let’s try a family of polygonal paths CY that consist of
• the line segment L1 from (2, 2) to (2, Y ) followed by
• the line segment L2 from (2, Y ) to (1, Y ) followed by
• the line segment L3 from (1, Y ) to (1, 1).
This is a way of characterizing a family of alternate paths withR only one
parameter, Y . We are hoping that the value of the integral CY F · dr
depends on Y and that we R can choose a specific value of Y so as to make
the value of the integral CY F · dr exactly 4.
Note that
• On L1 , x = 2 is a constant (so that dx = 0) and y runs from 2 to Y .
• On L2 , y = Y is a constant (so that dy = 0) and x runs from 2 to 1.
• On L3 , x = 1 is a constant (so that dx = 0) and y runs from Y to 1
So,
Z Z Z
F · dr = (3y dx + (x − 1)dy + (3y dx + (x − 1)dy
CY L1 L2
Z
+ (3y dx + (x − 1)dy
L3
Z Y Z 1 Z 1
= dy + 3Y dx + 0 dy
2 2 Y
= (Y − 2) + 3Y (1 − 2) = −2Y − 2
=⇒ b = 1 − 3a
Using b = 1 − 3a,
a+b+c=1
=⇒ a + (1 − 3a) + c = 1
=⇒ c = 2a
(2, 2)
L
(1, 1)
y = ax2 + (1 − 3a)x + 2a
r0 (x) = 1, 2ax + 1 − 3a
Here Y is a parameter.
y
(0, Y ) L2 (2, Y )
L1 L3
(0, 0) (2, 0) x
R
We are hoping that the value of the integral CY F · dr depends on Y
and thatRwe can choose a specific value of Y so as to make the value of the
integral CY F · dr exactly 8. Note that
x
(2, 0)
In this case, F(r(t)) = (2A sin t + 2, 0) and r0 (t) = (− sin t, A cos t), so
= Aπ + 4
Thus
Z
F · dr = |x(Q) − x(P )| = distance between P and Q
C
2.4.2.38. ∗. Solution.
2
• First notice that the vector
1 3
R R y, z) = z k̂ is conservative
field F̃(x,
(with potential 3 z ), so C1 F̃ · dr = C2 F̃ · dr for any two curves C1
and C2 from P1 to P2 (whether R or not they R are on the surface S).
Consequently, the statement “ C1 F · dr = C2 F · dr” is true if and
R R
only if the statement “ C1 (F − F̃) · dr = C2 (F − F̃) · dr” is true. So
we may replace the vector field F with the vector field
• We are to consider only curves on the surface S. For any such curve
C, say parametrized by r(t) with a ≤ t ≤ b, the integral
Z Z b dr
G · dr = G r(t) · (t) dt
C a dt
= H · dr
C
H(x, y, z) = G(x, y, 2 + x2 − 3y 2 )
= x(2 + x2 − 3y 2 ) + axy 2 ı̂ı + y(2 + x2 − 3y 2 )̂
• The curl of H is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × H = det
∂x ∂y ∂z
2x + x3 − 3xy 2 + axy 2 2y + yx2 − 3y 3 0
= 2xy − [−6xy + 2axy] k̂ = (8 − 2a)xy k̂
∂f 2
(x, y, z) = (1 + 6x2 )ye3x + 2xz cos(x2 z)
∂x
∂f 2
(x, y, z) = xe3x
∂y
∂f
(x, y, z) = x2 cos(x2 z)
∂z
Integrating the second of these equations gives
2
f (x, y, z) = xye3x + g(x, z)
so that
Z Z
4
J= (zı̂ı + x yz ̂ + F) · dr = −2 + (zı̂ı + x4 yz ̂) · dr
C C
dr
Parametrize C by r(x) = xı̂ı −x ̂ +x2 k̂ with 0 ≤ x ≤ 1. As dx = ı̂ı −̂ +2x k̂
Z Z 1
(zı̂ı + x4 yz ̂) · dr = (x2ı̂ı − x7 ̂) · ı̂ı − ̂ + 2x k̂ dx
C 0
Z 1
= (x2 + x7 ) dx
0
1 1
= +
3 8
11
=
24
11 37
=⇒ J = −2 + =− ≈ −1.5417
24 24
R
(e) T is a closed path and F is conservative, so T F · dr = 0. Let T1
be the line segment from (1, 0, 0) to (0, 1, 0), T2 be the line segment from
(0, 1, 0) to (0, 0, 1) and T3 be the line segment from (0, 0, 1) to (1, 0, 0).
APPENDIX D. SOLUTIONS TO EXERCISES 504
z
(0,0,1)
T2
T3
(0,1,0)
y
(1,0,0)
T1
x
R
On T1 , z = 0, so T1 zı̂ı · dr = 0. On T2 , x = 0, so ı̂ı · dr = dx = 0 and
R
T2
zı̂ı · dr = 0. Parametrize T3 by r(t) = tı̂ı + (1 − t)k̂, 0 ≤ t ≤ 1. Then
dr
dt = ı̂ − k̂ and the z-coordinate of the path is parametrized by 1 − t. So,
ı
zı̂ı dr
Z Z Z 1 z }| { z }| {
(zı̂ı + F) · dr = zı̂ı · dr = (1 − t)ı̂ı · (ı̂ı − k̂) dt
T T3 0
Z 1
1
= (1 − t) dt =
0 2
2.4.2.42. ∗. Solution. (a) By Newton’s law of motion
ma = F =⇒ 2v0 (t) = 4t , 6t2 , −4t =⇒ v0 (t) = 2t , 3t2 , −2t
So
Z t Z t
0
2u , 3u2 , −2u du
v(t) = v(0) + v (u) du = (0, 0, 0) +
0 0
= t2 , t3 , −t2
Z t Z t
0
u2 , u3 , −u2 du
r(t) = r(0) + r (u) du = (1, 2, 3) +
0 0
3
t4 t3
3 4 3
t
= (1, 2, 3) + t /3 , t /4 , −t /3 = + 1, + 2, − + 3
3 4 3
(c) From parts (a) and (b)
p
|r0 (t)| = t2 (1, t, −1) = t2 2 + t2
and
ı̂ı ̂ k̂
r0 (t) × r00 (t) = det t2 t3 −t2
2t 3t2 −2t
= − 2t4 + 3t4 ı̂ı − − 2t3 + 2t3 ̂ + 3t4 − 2t4 k̂
= t4 ı̂ı + t4 k̂
√
=⇒ r (t) × r00 (t) = 2 t4
0
(b) As
√ √ √
v(t) = r0 (t) = 2 2t1/2 , 2 2t1/2 , 2 − 2t
v(1) = 2 2 1, 1, 0
√ −1/2 √ −1/2 √ √
a(t) = v0 (t) =
2t , 2t , −2 a(1) = 2 1, 1, − 2
√ √
v(1) × a(1) = 4 − 2, 2, 0 |v(1)| = 4
the curvature
|v(1) × a(1)| 8 1
κ(1) = 3
= 3 =
|v(1)| 4 8
3.1.1 · Exercises = 0, 0, −
8
APPENDIX D. SOLUTIONS TO EXERCISES 507
x(u, v) = u + v
y(u, v) = u2 + v 2
z(u, v) = u − v
1 2 1 2
y(u, v) = u2 + v 2 = x(u, v) + z(u, v) + x(u, v) − z(u, v)
4 4
1 1
= x(u, v)2 + z(u, v)2
2 2
All points of our surface lie on 2y = x2 + z 2 . This is a parabolic bowl:
• no points have y < 0 and
• the y = Y (with Y > 0) cross-section is the circle x2 + z 2 = 2Y ,
y=Y
• the x = 0 cross-section is the parabola 2y = z 2 , x = 0
• the z = 0 cross-section is the parabola 2y = x2 , z = 0
z
• x2 + y 2 − 2z 2 = 1 + 2v 2 − 2v 2 = 1, as desired.
• The condition x2 + y 2 ≤ 9 is equivalent to 1 + 2v 2 ≤ 9, which is
equivalent to v ≤ 2, since v ≥ 0.
• The condition x2 + y 2 ≥ 3 is equivalent to 1 + 2v 2 ≥ 3, which is
equivalent to v ≥ 1, since v ≥ 0.
• z=v≥0
√ √
p Under this parametrization, x =
(d) Yes. 1 + u sin v, y = 1 + u cos v
and z = u/2. So
• x2 + y 2 − 2z 2 = 1 + u − 2(u/2) = 1, as desired.
• The condition x2 + y 2 ≤ 9 is equivalent to 1 + u ≤ 9, which is
equivalent to u ≤ 8.
• The condition x2 + y 2 ≥ 3 is equivalent to 1 + u ≥ 3, which is
equivalent to u ≥ 2.
p
• z = u/2 ≥ 0
√ √
p No. Under this parametrization, x = u cos v, y = − u sin v and
(e)
z = (u + 1)/2. So
• 0 ≤ z = 4 − u2 ≤ 1
• For each fixed z = 4 − u2 between 0 and 1, (x, y) runs once around
the circle x2 + y 2 = 4 − z = u2 as v runs from 0 to 2π.
In the cases of D, E, F, for r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂ to lie
inside S3 we need (recalling that all points of S1 have z(θ, z) ≥ 0 and hence
z ≥ 0)
√
x(θ, z)2 + y(θ, z)2 ≤ z(θ, z)2 ⇐⇒ 4 − z 2 ≤ z 2 ⇐⇒ z ≥ 2
In the cases of J, K, L, for r(x, y) = x(x, y)ı̂ı + y(x, y) ̂ + z(x, y) k̂ to lie
inside S3 we need
In the cases of J, K, L, for r(x, y) = x(x, y)ı̂ı + y(x, y) ̂ + z(x, y) k̂ to lie
inside S3 we need
(x, y, z)
θ √1 (ı̂ı + ̂)
4 2
(2, 2, 4)
y
2
2
x
So, we can parametrize the circle as (x, y, z) = (2+ √12
cos θ, 2+ √12 cos θ, 4+
sin θ), with 0 ≤ θ ≤ 2π.
Remark: it’s easy to check that this equation satisfies the two prop-
erties we desire. Since the x- and y coordinates match, it’s in the plane
x = y. To check that it’s a circle centred at (2, 2, 4), we note the distance
from (x, y, z) to (2, 2, 4) is:
p
d = (x − 2)2 + (y − 2)2 + (z − 4)2
s 2 2
1 1
= √ cos θ + √ cos θ + (sin θ)2
2 2
r
1 1 p
= cos2 θ + cos2 θ + sin2 θ = cos2 θ + sin2 θ
2 2
=1
So, our points all have distance one from the same point — that is, they
lie on a circle of radius 1.
(b) Consider a point (x, y, z) = (2 + √12 cos θ, 2 + √12 cos θ, 4 + sin θ),
rotating φ radians about the line x = y = 4.
r(t) = (4, 4, t)
R
(x, y, z)
φ
(2, 2, 4)
The new position of the point has the same height, z = 4 + sin θ. Its
distance from the line x = y = 4 is also preserved:
p
R = (x − 4)2 + (y − 4)2 + (z − z)2
APPENDIX D. SOLUTIONS TO EXERCISES 512
s
1 1
= ( √ cos θ − 2)2 + ( √ cos θ − 2)2 + 0)
2 2
√
= cos θ − 2 2
The circle traced out by a point (x, y, z) = (2+ √12 cos θ, 2+ √12 cos θ, 4+
√
sin θ) on the circle√is centred
√ at (4, 4, z) with radius√ 2(4 − √ x), so it has
equation x = 4 + 2(2 − 2 cos θ) cos φ, y = 4 + 2(2 − 2 cos θ) sin φ,
z = 4 sin θ.
In particular,
k̂ · (x − 0, y − 0, z − 0) = 0 or z=0
or
∂G ∂G
x0 , y0 , z0 (x − x0 ) + x0 , y0 , z0 (y − y0 )
∂x ∂y
∂G
+ x0 , y0 , z0 (z − z0 ) = 0
∂z
(E2)
∂G ∂G
x0 , y0 , z0 {x − x0 } + x0 , y0 , z0 {y − y0 }
∂x ∂y
∂G
+ x0 , y0 , z0 {z − z0 }
∂z
∂G 0 ∂G
x0 , y0 , z0 t y 0 (t0 )
= x0 , y0 , z0 t x (t0 ) +
∂x ∂y
∂G
x0 , y0 , z0 t z 0 (t0 )
+
∂z
h ∂G ∂G
x0 , y0 , z0 x0 (t0 ) + x0 , y0 , z0 y 0 (t0 )
=t
∂x ∂y
∂G i
x0 , y0 , z0 z 0 (t0 )
+
∂z
=0
by (E3). That is, (x, y, z) obeys the equation, (E2), of the tangent plane
to S at r0 and so is on that tangent plane. So the tangent line to C at r0
is contained in the tangent plane to S at r0 .
APPENDIX D. SOLUTIONS TO EXERCISES 514
t = ∇ F (x0 , y0 , z0 ) × ∇ G(x0 , y0 , z0 )
n · (x − x0 , y − y0 , z − z0 ) = 0
where n = t = ∇ F (x0 , y0 , z0 ) × ∇ G(x0 , y0 , z0 )
3.2.1.5. Solution. Use S1 to denote the surface z = f (x, y), S2 to denote
the surface z = g(x, y) and C to denote the curve of intersection of S1 and
S2 .
• Since C is contained in S1 , the tangent line to C at (x0 , y0 , z0 ) is
contained in the tangent plane to S1 at (x0 , y0 , z0 ), by Q[3.2.1.2].
In particular, any tangent vector, t, to C at (x0 , y0 , z0 ) must be
perpendicular to −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂, the normal vector to
S1 at (x0 , y0 , z0 ). (See part (b) of Theorem 3.2.1.)
So the tangent line in question passes through (x0 , y0 , z0 ) and has direction
vector d = t. Consquently, the tangent line is
(x − x0 , y − y0 , z − z0 ) = t d
or
x = x0 + t gy (x0 , y0 ) − fy (x0 , y0 )
y = y0 + t fx (x0 , y0 ) − gx (x0 , y0 )
z = z0 + t fx (x0 , y0 )gy (x0 , y0 ) − fy (x0 , y0 )gx (x0 , y0 )
So (2/9 , 1/9 , 1) is a normal vector to the surface at (−1, 1, 1/3) and the
tangent plane is
2 1 1
(x + 1) + (y − 1) + z − =0
9 9 3
2 1 2 1 1 2
x+ y+z =− + + =
9 9 9 9 3 9
or 2x + y + 9z = 2.
3.2.1.7. ∗. Solution. The equation of the given surface is of the form
G(x, y, z) = 9 with G(x, y, z) = √ 2 272 2 . So, by part (c) of Theorem
x +y +z +3
3.2.1, a normal vector to the surface at (2, 1, 1) is
1 27
∇ G(2, 1, 1) = − 2x , 2y , 2z
2 (x2 + y 2 + z 2 + 3)3/2
(x,y,z)=(2,1,1)
= −(2 , 1 , 1)
−(2 , 1 , 1) · (x − 2 , y − 1 , z − 1) = 0 or 2x + y + z = 6
3.2.1.8. ∗. Solution. We may use G(x, y, z) = xyz 2 + y 2 z 3 − 3 − x2 = 0
as an equation for the surface. Note that (−1, 1, 2) really is on the surface
since
Gx (x, y, z) = yz 2 − 2x Gx (−1, 1, 2) = 6
2 3
Gy (x, y, z) = xz + 2yz Gy (−1, 1, 2) = 12
2 2
Gz (x, y, z) = 2xyz + 3y z Gz (−1, 1, 2) = 8
(6 , 12 , 8) · (x + 1 , y − 1 , z − 2) = 0 or 6x + 12y + 8z = 22
or
3 3 11
z =− x− y+
4 2 4
3.2.1.9. ∗. Solution. (a) The surface is G(x, y, z) = z−x2 +2xy−y 2 = 0.
When x = a and y = 2a and (x, y, z) is on the surface, we have z =
a2 − 2(a)(2a) + (2a)2 = a2 . So, by part (c) of Theorem 3.2.1, a normal
vector to this surface at (a, 2a, a2 ) is
∇ G(a, 2a, a2 ) = (−2x + 2y , 2x − 2y , 1) = (2a , −2a , 1)
(x,y,z)=(a,2a,a2 )
(b) The two planes are parallel when their two normal vectors, namely
(2a , −2a , 1) and (1 , −1 , 1), are parallel. This is the case if and only if
a = 21 .
3.2.1.10. ∗. Solution. A plane is determined by one point on the plane
and one vector perpendicular to the plane. We are told that (8, 1, 5) is
on the plane, so it suffices to find a normal vector. The given surface is
parametrized by
so the vectors
∂r
(u, v) = 4u , 0 , 2u
∂u
∂r
(u, v) = 0 , 2v , 3v 2
∂v
are tangent to S at r(u, v). Note that r(2, 1) = (8, 1, 5). So
∂r
(2, 1) = 8 , 0 , 4
∂u
∂r
(2, 1) = 0 , 2 , 3
∂v
are tangent to S at r(2, 1) = (8, 1, 5) and
∂r ∂r
(2, 1) × (2, 1) = 8 , 0 , 4 × 0 , 2 , 3
∂u ∂v
ı̂ı ̂ k̂
= det 8 0 4
0 2 3
APPENDIX D. SOLUTIONS TO EXERCISES 517
= − 8 , −24 , 16)
1
or −8 − 8 , −24 , 16) = 1 , 3 , −2) is normal to S at (8, 1, 5). So the
tangent plane is
(1, 3, −2) · (x, y, z) − (8, 1, 5) = 0 or x + 3y − 2z = 1
3.2.1.11. ∗. Solution. To find the tangent plane we have to find a
normal vector to the surface at (2, 2, 0). Since
∂r
= 1 , 2u , 1
∂u
∂r
= 1 , 2v , −1
∂v
a normal vector to the surface at r(u, v) is
ı̂ı ̂ k̂
∂r ∂r
× = det 1 2u 1
∂u ∂v
1 2v −1
= − 2u − 2v , 2 , 2v − 2u
(−4, 2, 0) or (−2, 1, 0)
−2(x − 2) + (y − 2) + 0z = 0 or y = 2x − 2
3.2.1.12. ∗. Solution. The first order partial derivatives of f are
4xy 8
fx (x, y) = − 2 fx (−1, 2) =
(x2
+ y2 ) 25
2 4y 2 2 16 6
fy (x, y) = 2 − 2 fy (−1, 2) = − =−
x + y2 (x2 + y 2 ) 5 25 25
So, by part (b) of Theorem 3.2.1, a normal vector to the surface at (x, y) =
8 6
(−1, 2) is ( 25 , − 25 , −1). As f (−1, 2) = 54 , the tangent plane is
8 6 4
, − , −1 · x + 1 , y − 2 , z − =0
25 25 5
8 6 8
or x− y−z =−
25 25 5
and the normal line is
4 8 6
(x, y, z) = − 1, 2, +t , − , −1
5 25 25
3.2.1.13. ∗. Solution. A normal vector to the surface x2 + 9y 2 + 4z 2 =
17 at the point (x, y, z) is (2x , 18y , 8z). A normal vector to the plane
x − 8z = 0 is (1 , 0 , −8). So we want (2x , 18y , 8z) to be parallel to
(1 , 0 , −8), i.e. to be a nonzero constant times (1 , 0 , −8). This is the
case whenever y = 0 and z = −2x with x 6= 0. In addition, we want
APPENDIX D. SOLUTIONS TO EXERCISES 518
17 = x2 + 9y 2 + 4z 2 = x2 + 4(−2x)2 = 17x2 =⇒ x = ±1
For this normal line to pass through the origin, there must be a t with
or
x0 + 2x0 t = 0 (E1)
y0 + (4y0 + 2)t = 0 (E2)
z0 − t = 0 (E3)
x0 (1 + 2z0 ) = 0 (E1)
y0 + (4y0 + 2)z0 = 0 (E2)
−y0 − 1 = 0 =⇒ y0 = −1
t t
⇐⇒ x0 = , y0 = z0 = − and z02 = 4x20 + y02 − 1
4 2
t2 t2 t2 t t
⇐⇒ = + − 1 and x0 = , y0 = z0 = −
4 4 4 4 2
1
⇐⇒ t = ±2 (x0 , y0 , z0 ) = ± 2 , −1, −1
(c) The specified tangent line passes through (1, 1, 3) and has direction
vector (1, 1, 3) and so has vector parametric equation
⇐⇒ 2t6 = 2 ⇐⇒ t = 1
The angle θ between the curve and the normal vector to the surface is
determined by
√ √
(3, 1, 2) (1, 1, 4) cos θ = (3, 1, 2) · (1, 1, 4) ⇐⇒ 14 18 cos θ = 12
√
⇐⇒ 7 × 36 cos θ = 12
2
⇐⇒ cos θ = √
7
◦
⇐⇒ θ = 40.89
The angle between the curve and the surface is 90 − 40.89 = 49.11◦ (to two
decimal places).
APPENDIX D. SOLUTIONS TO EXERCISES 520
The values of z0 at these points are 0, e−1 , −e−1 , −e−1 and e−1 , respec-
tively. So the horizontal tangent planes are z = 0, z = e−1 and z = −e−1 .
At the highest and lowest points of the surface, the tangent plane is horizon-
tal. So the largest and smallest values of z are e−1 and −e−1 , respectively.
Hence by (3.3.2)
q p
dS = 1 + fx (x, y)2 + fy (x, y)2 dx dy = 1 + 02 + tan2 θ dx dy
and
ZZ ZZ p
Area(S) = dS = 1 + tan2 θ dx dy
S D
Z a pZ b
= dx dy 1 + tan2 θ
0 0
p
= ab 1 + tan2 θ = ab sec θ
3.3.6.2. Solution. Note that all three vertices (a, 0, 0), (0, b, 0) and
(0, 0, c) lie on the plane xa + yb + zc = 1. So the triangle is part of that
plane.
Method 1. S is the part of the surface z = f (x, y) with f (x, y) =
APPENDIX D. SOLUTIONS TO EXERCISES 521
c 1 − xa − yb and with (x, y) running over the triangle Txy in the xy-plane
with vertices (0, 0, 0) (a, 0, 0) and (0, b, 0). Hence by the first part of (3.3.2),
ZZ q
Area(S) = 1 + fx (x, y)2 + fy (x, y)2 dx dy
Txy
r
c2 c2
ZZ
= 1+ 2
+ 2 dx dy
Txy a b
r
c2 c2
= 1+ 2
+ 2 A(Txy )
a b
where A(Txy ) is the area of Txy . Since the triangle Txy has base a and
height b (see the figure below), it has area 12 ab. So
r
1 c2 c2 1p 2 2
Area(S) = 1 + 2 + 2 ab = a b + a2 c2 + b2 c2
2 a b 2
z
p0, 0, cq Txy
p0, b, 0q
y
pa, 0, 0q
x
Method 2. S is the part of the surface x = g(y, z) with g(y, z) =
a 1 − yb − zc and with (y, z) running over the triangle Tyz in the yz-plane
with vertices (0, 0, 0) (0, b, 0) and (0, 0, c). Hence by the second part of
(3.3.2),
ZZ q
Area(S) = 1 + gy (y, z)2 + gz (y, z)2 dy dz
Tyz
r
a2 a2
ZZ
= 1+ 2
+ 2 dy dz
Tyz b c
r
a2 a2
= 1+ 2
+ 2 A(Tyz )
b c
where A(Tyz ) is the area of Tyz . Since Tyz has base b and height c, it has
area 21 bc. So
r
1 a2 a2 1p 2 2
Area(S) = 1 + 2 + 2 bc = a b + a2 c2 + b2 c2
2 b c 2
Method 3. S is the part of the surface y = h(x, z) with h(x, z) =
b 1 − xa − zc and with (x, z) running over the triangle Txz in the xz-plane
with vertices (0, 0, 0) (a, 0, 0) and (0, 0, c). Hence by the third part of
(3.3.2),
ZZ p
Area(S) = 1 + hx (x, z)2 + hz (x, z)2 dx dz
Txz
r
b2 b2
ZZ
= 1 + 2 + 2 dx dz
Txz a c
APPENDIX D. SOLUTIONS TO EXERCISES 522
r
b2 b2
= 1+ 2
+ 2 A(Txz )
a c
where A(Txz ) is the area of Txz . Since Txz has base a and height c, it has
area 12 ac. So
r
1 b2 b2 1p 2 2
Area(S) = 1+ 2
+ 2 bc = a b + a2 c2 + b2 c2
2 a c 2
(b) We have already seen in the solution to part (a) that
ab ac bc
Area(Txy ) = Area(Txz ) = Area(Tyz ) =
2 2 2
Hence
r
a2 b2 a2 c2 b2 c2
Area(S) = + +
4 4 4
q
= Area(Txy )2 + Area(Txz )2 + Area(Tyz )2
So
ZZ Z π/2 Z h π
Area(S) = dS = dθ dy a = a h
S 0 0 2
and
ZZ ZZ p
I= (x2 + y 2 ) dS = (x2 + y 2 ) 1 + x2 + y 2 dx dy
S x2 +y 2 ≤3
APPENDIX D. SOLUTIONS TO EXERCISES 523
√
Z 2π Z 3 p
= dθ dr r r2 1 + r2
0 0
0 ≤ z = a2 − x2 − y 2 ⇐⇒ x2 + y 2 ≤ a2
That is, if and only if (x, y) lies in the circular disk of radius a centred on
the origin. The equation of the paraboloid is of the form z = f (x, y) with
f (x, y) = a2 − x2 − y 2 . So, by (3.3.2),
ZZ q
Surface area = 1 + fx (x, y)2 + fy (x, y)2 dx dy
x2 +y 2 ≤a2
ZZ p
= 1 + 4x2 + 4y 2 dx dy
x2 +y 2 ≤a2
x y
fx (x, y) = p fy (x, y) = p
x2 + y2 x + y2
2
So, by (3.3.2),
ZZ q
Surface area = 1 + fx (x, y)2 + fy (x, y)2 dx dy
4≤x2 +y 2 ≤9
s
x2 y2
ZZ
= 1+ + 2 dx dy
4≤x2 +y 2 ≤9 x2 +y 2 x + y2
APPENDIX D. SOLUTIONS TO EXERCISES 524
√ ZZ
= 2 dx dy
4≤x2 +y 2 ≤9
So, by (3.3.2),
Z 1 Z 1 q
Surface area = dx dy 1 + fx (x, y)2 + fy (x, y)2
0 0
Z 1 Z 1 p
= dx dy 1+x+y
0 0
Z 1 h2 iy=1
= dx (1 + x + y)3/2
0 3 y=0
Z 1
2
dx (2 + x)3/2 − (1 + x)3/2
=
3 0
2 2h ix=1
= (2 + x)5/2 − (1 + x)5/2
3 5 x=0
4 5/2 5/2 5/2 5/2
= 3 −2 −2 +1
15
4 √ √
= 9 3−8 2+1
15
p
3.3.6.8. ∗. Solution. (a) By (3.3.2), F (x, y) = 1 + fx (x, y)2 + fy (x, y)2 .
(b) (i) The “dimple” to be painted is part of the upper sphere x2 + y 2 +
√ 2
z − 2 3 = 4. It is on the bottom half of the sphere and so has equation
√ p
z = f (x, y) = 2 3 − 4 − x2 − y 2 . Note that
x y
fx (x, y) = p fy (x, y) = p
4 − x2 − y 2 4 − x2 − y 2
√
The point on the dimple with the largest value of x is (1, 0, 3). (It is
marked by a dot in the figure above.) The dimple is invariant under rota-
tions around the z--axis and so has (x, y) running over x2 + y 2 ≤ 1. So, by
(3.3.2),
ZZ q
Surface area = 1 + fx (x, y)2 + fy (x, y)2 dx dy
x2 +y 2 ≤1
s
x2 y2
ZZ
= 1+ 2 2
+ dx dy
x2 +y 2 ≤1 4−x −y 4 − x2 − y 2
ZZ
2
= p dx dy
x2 +y 2 ≤1 4 − x2 − y 2
(b)
√ (ii) Observe that if we flip the dimple up by reflecting it in the plane
z = 3, as in the figure below, the “Death Star” becomes a perfect ball of
radius 2.
z
?
2 3
? ?
z“ 3 p1, 0, 3q
π
6
2
x
The area of the pink dimple in the figure above is identical to the area
of the blue cap in that figure. So the total surface area of the Death Star
is exactly the surface area of a sphere of radius 2 and so is 4π22 = 16π.
3.3.6.9. ∗. Solution. On the upper half of the cone
p x y
z = f (x, y) = x2 + y 2 fx (x, y) = p fy (x, y) = p
x2 + y 2 x2 + y 2
so that
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dx dy
s
x2 y2
= 1+ 2 + dx dy
x + y2 x2 + y 2
√
= 2 dx dy
and
ZZ √
Area = 2 dx dy
1≤x2 +y 2 ≤162
√ h i
2 area of (x, y) x2 + y 2 ≤ 162 − area of (x, y) x2 + y 2 ≤ 1
=
√ √
= 2 π162 − π12 = 255 2π ≈ 1132.9
y
fy (x, y) = − p
a − x2 − y 2
2
so that
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dx dy
APPENDIX D. SOLUTIONS TO EXERCISES 526
s
x2 y2
= 1+ + dx dy
a2 − x2 − y 2 a2 − x2 − y 2
s
a2
= dx dy
a2 − x2 − y 2
r “ a cos θ
x
pa{2, 0q
So the
Z π/2 Z a cos θ
a
Surface Area = 2 dθ dr r √
0 0 a2 − r2
Z π/2 h p ia cos θ
= 2a dθ − a2 − r2
0 0
Z π/2
= 2a dθ a − a sin θ
0
h iπ/2
2
= 2a θ + cos θ = a2 [π − 2]
0
x4 − y 4 + y 2 z 2 − z 2 x2 + 1 = x4 − y 4 + y 2 x2 + y 2 − x2 + y 2 )x2 + 1 = 1
D = (x, y) (x − 1)2 + y 2 ≤ 1
So
=1
ZZ z
4 4
}|
2 2 2 2
{ √ ZZ √ √
(x − y + y z − z x + 1) dS = 2 dx dy = 2 Area(D) = 2 π
S D
APPENDIX D. SOLUTIONS TO EXERCISES 527
Then
∂r
= (R + r cos θ) − sin ψı̂ı + cos ψ̂
∂ψ
∂r
= r − sin θ cos ψ ı̂ı − sin θ sin ψ ̂ + cos θ k̂
∂θ
and
∂r ∂r
× = r(R + r cos θ) − sin ψı̂ı + cos ψ̂ ×
∂ψ ∂θ
− sin θ cos ψ ı̂ı − sin θ sin ψ ̂ + cos θ k̂
ı̂ı ̂ k̂
= r(R + r cos θ) det − sin ψ cos ψ 0
− sin θ cos ψ − sin θ sin ψ
cos θ
= r(R + r cos θ) cos ψ cos θ ı̂ı + sin ψ cos θ ̂ + sin θ k̂
As cos ψ cos θ ı̂ı + sin ψ cos θ ̂ + sin θ k̂ is a unit vector, (we could have
shortened this computation by observing that − sin ψ ı̂ı+cos ψ ̂ and − sin θ cos ψ ı̂ı−
sin θ sin ψ ̂ + cos θ k̂ are mutually perpendicular unit vectors, so that their
cross product is automatically a unit vector) and
∂r ∂r
∂ψ × ∂θ = r(R + r cos θ) =⇒ dS = r(R + r cos θ) dψ dθ
Then
∂r
= −a sin ϕ sin θ ı̂ı + a sin ϕ cos θ ̂
∂θ
∂r
= a cos ϕ cos θ ı̂ı + a cos ϕ sin θ ̂ − a sin ϕ k̂
∂ϕ
so that
ı̂ı ̂ k̂
∂r ∂r
× = det −a sin ϕ sin θ
a sin ϕ cos θ 0
∂θ ∂ϕ
a cos ϕ cos θ a cos ϕ sin θ −a sin ϕ
= −a2 sin2 ϕ cos θ ı̂ı − a2 sin2 ϕ sin θ ̂ − a2 sin ϕ cos ϕ k̂
∂r ∂r
=⇒ dS =
× dθ dϕ = a2 sin ϕ dθ dϕ
∂θ ∂ϕ
APPENDIX D. SOLUTIONS TO EXERCISES 528
As the surface area of the part of the sphere in the first octant is 18 4πa2 =
πa2
2
RR Z π/2 Z π/2 z
z dS 1 2
x̄ = ȳ = z̄ = RRS
z }| {
= dθ dϕ (a sin ϕ)( a cos ϕ)
S
dS πa2 /2 0 0
π/2
2a π π/2
Z
1 a
= dϕ sin ϕ cos ϕ = a sin2 ϕ =
π 2 0 2 0 2
3.3.6.14. Solution. In cylindrical coordinates
We still have to determine the limits of integration. The figure on the left
below provides a top view of the cylinder.
y z z=r
x2 + y 2 = 2ax
(0, a) r
r
θ
x z = −r
r=2a sin θ
From it we see that 0 ≤ θ ≤ π. The cone z 2 = x2 + y 2 = r2 (i.e.
z = ±r) and the cylinder r = 2a sin θ intersect at z 2 = r2 = 4a2 sin2 θ. So,
for each fixed θ, z runs from −2a sin θ to z = +2a sin θ. (See the figure on
the right above. It shows a constant θ cross-section.) Finally,
Z Z π Z 2a sin θ Z π
Area = 2a dθ dz = 2a dθ dz = 8a2 dθ sin θ
|z|≤2a sin θ 0 −2a sin θ 0
h iπ
= 8a2 − cos θ
0
= 16a2
APPENDIX D. SOLUTIONS TO EXERCISES 529
3.3.6.15. ∗. Solution. (a) This right circular cone symmetric about the
z-axis projects down onto a disk D in the plane z = 0. Setting z = b gives
D = (x, y, z) x2 + y 2 ≤ a2 , z = 0
bp 2 b
x = t cos θ, y = t sin θ, z = x + y 2 = t, (∗)
a a
0 ≤ θ ≤ 2π, 0 ≤ t ≤ a.
∇g xı̂ı + ŷ + z k̂
n̂ = =p
∇
|∇ g| x2 + y 2 + z 2
n++1−1/2 n+1/2
=⇒ F · n̂ = x2 + y 2 + z 2 = a2 = a2n+1
ZZ ZZ
=⇒ F · n̂ dS = a2n+1 dS = a2n+1 Area(S) = 4πa2n+3
S S
Then
∂r
= cos θ ı̂ı + sin θ ̂ − k̂
∂r
∂r
= −r sin θ ı̂ı + r cos θ ̂
∂θ
ı̂ı ̂ k̂
∂r ∂r
× = det cos θ sin θ −1
∂r ∂θ
−r sin θ r cos θ 0
= −r cos θ ı̂ı − r sin θ ̂ + r k̂
∂r ∂r
=⇒ n̂ dS = × dr dθ
∂r ∂θ
= − r cos θ ı̂ı − r sin θ ̂ + rk̂ dr dθ
∇G −yı̂ı − x̂ + k̂
n̂ dS = dxdy = dxdy = (−yı̂ı − x̂ + k̂) dxdy
∇ G · k̂ 1
so that
x2 y x2 y p
p dS = p y 2 + x2 + 1 dxdy = x2 y dxdy
1+ x2 + y2 1+ x2 + y2
and
F · n̂ dS = xı̂ı + ŷ + k̂ · − yı̂ı − x̂ + k̂ dxdy = 1 − 2xy dxdy
(a)
1 1
x2 y
ZZ ZZ Z Z
2
p dS = x y dxdy = dx dy x2 y
S 1 + x2 + y 2 S 0 0
Z 1
1
= dx 21 x2 =
0 6
(b)
ZZ Z 1 Z 1 Z 1
1 1
F · n̂ dS = dx dy [1 − 2xy] = dx [1 − x] = 1 − =
S 0 0 0 2 2
3.3.6.19. ∗. Solution.For the surface z = f (x, y) = y 3/2 ,
r 3 √ 2 r
q
2 2
9
dS = 1 + fx + fy dxdy = 1 + y dxdy = 1 + y dxdy
2 4
by (3.3.2). So the area is
Z 1 Z 1
r Z 1
9 8 h 9 3/2 i1
dx dy 1+ y = 1+ y
dx
0 0 4 0 27 4 0
Z 1
8 h 13 3/2
i
= dx −1
0 27 4
" #
3/2
8 13
= −1
27 4
as the parametrization.
(b) By symmetry the centre of mass will lie on the z-axis. We are only
asked for the z-coordinate anyway. The z-coordinate of the centre of mass
is the weighted average of z over the cone. Since a density has not been
specified, we assume that it is a constant. We RR may takeRR the density to be
1, so the z-coordinate of the centre of mass is S z dS/ S dS.
Since
∂r 2 2
∂θ = − 3 (3 − z) sin θ , 3 (3 − z) cos θ , 0
∂r 2 2
∂z = − 3 cos θ , − 3 sin θ , 1
∂r ∂r 2 2 4
∂θ × ∂z = 3 (3 − z) cos θ , 3 (3 − z) sin θ , 9 (3 − z)
2 27
= =1
9 6
This is a little less than half way up the cone, which is reasonable since the
cone is “bottom heavy”.
3.3.6.22. ∗. Solution. Each constant z cross-section of the cone is a
circle. When z = 0, that circle has radius a. When z = a that circle has
radius 0. Thus the radius decreases linearly from a to 0 as z increases from
0 to a. So the radius at height z is a − z and we can parametrize the cone
by
Since
∂r
∂θ = − (a − z) sin θ , (a − z) cos θ , 0
∂r
∂z = − cos θ , − sin θ , 1
∂r ∂r
∂θ × ∂z = (a − z) cos θ , (a − z) sin θ , a − z
F x(θ, z), y(θ, z), z(θ, z) = cos θ ı̂ı + 2z sin θ cos θ ̂ + 16z sin4 θ k̂
So
ZZ Z 1 Z 2π
dθ 2 cos2 θ + 2z sin2 θ cos θ
F · n̂ dS = dz
S 0 0
Z 1 Z 2π
dθ 1 + cos(2θ) + 2z sin2 θ cos θ
= dz
0 0
Z 1 i2π
h 1 2
= dz θ + sin(2θ) + z sin3 θ = 2π
0 2 3 0
R 2π
For an efficient, sneaky, way to evaluate 0
cos2 θ dθ, see Example 2.4.4.
To get the downward pointing normal, we want the minus sign. Set
T = (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1 − x
Then
ZZ ZZ z
z }| {
x + 1 , y + 1 , 2 (4 − x2 − y 2 ) · 2x , 2y , 1 dxdy
F · n̂ dS = −
S
Z ZT
=− 8 + 2x + 2y dxdy
T
Z 1 Z 1−x
=− dx dy 8 + 2x + 2y
0 0
Z 1
dx 8(1 − x) + 2x(1 − x) + (1 − x)2
=−
0
Z 1
dx 9 − 8x − x2
=−
0
1 14
=− 9−4− =−
3 3
3.3.6.25. ∗. Solution. First we have √ to parametrize S. It is natural
to use spherical coordinates with ρ = 2. However if we use the standard
spherical coordinates
√ √ √
x = 2 sin ϕ cos θ y = 2 sin ϕ sin θ z = 2 cos ϕ
√ 2 2 √ 2
p y +z sin ϕ sin2 θ+cos2 ϕ
the condition x ≥ y 2 + z 2 , i.e. x ≤ 1, becomes sin ϕ cos θ ≤
1, which is very complicated. So let’s back up and think a bit before we
compute. From the sketch below
APPENDIX D. SOLUTIONS TO EXERCISES 536
(x, y, z)
y
p
(x, 0, 0) y2 + z2
x
√
y 2 +z 2
we see that x is the tangent of the angle between the radius vector
(x, y, z) and the x-axis. The angle between the radius vector (x, y, z) and
the z-axis (not the x-axis) is exactly spherical coordinate ϕ. So let’s modify
spherical coordinates to make the x-axis play the role of the z-axis. The
RR x = Z, y = X, z = Y . Then the
easy way to do is to just rename pintegral
we are to compute becomes S ZX 2 dS, and the condition x ≥ y 2 + z 2
√
becomes Z ≥ X 2 + Y 2 . Under the parametrization
√ √ √
X = 2 sin ϕ cos θ Y = 2 sin ϕ sin θ Z = 2 cos ϕ
√ √
2 2 sin ϕ
the condition Z ≥ X 2 + Y 2 is X Z+Y = cos ϕ ≤ 1, which is turn is
0 ≤ ϕ ≤ π4 . As dS = 2 sin ϕ dθdϕ (see Appendix A.6.3 and recall that
√
ρ = 2) the specified integral is
ZZ ZZ
xy 2 dS = ZX 2 dS
S S
Z π/4 √
Z 2π √
2
=2 2 cos ϕ
dϕ sin ϕ 2 sin ϕ cos θ
dθ
0 0
( ) Z
√ Z π/4 2π
3 2
=4 2 dϕ cos ϕ sin ϕ dθ cos θ
0 0
(Z ) Z
√ π/4 2π
cos(2θ) + 1
=4 2 dϕ cos ϕ sin3 ϕ dθ
0 0 2
π/4 2π
√ sin ϕ 4
sin(2θ) θ
=4 2 +
4 0 4 2 0
√
2π
=
4
R 2π
For an efficient, sneaky, way to evaluate 0
cos2 θ dθ, see Example 2.4.4.
3.3.6.26. ∗. Solution. Here is a sketch of the part of S that is in the
first octant.
z
x2 + z 2 = sin2 y
(x, y, z)
θ y
x
For each fixed y, x2 + z 2 = sin2 y is a circle of radius sin y. (It’s the
APPENDIX D. SOLUTIONS TO EXERCISES 537
Then, by (3.3.1),
∂r
= − sin y sin θ , 0 , sin y cos θ
∂θ
∂r
= cos y cos θ , 1 , cos y sin θ
∂y
∂r ∂r
× = − sin y cos θ , sin y cos y , − sin y sin θ
∂θ ∂y
∂r ∂r p
dS =
× dθ dy = sin y 1 + cos2 y dθ dy
∂θ ∂y
= (x, y, z) z = 1 − x2 − y 2 , x2 + y 2 ≤ 1
A point (x, y, z) on the plane lies in the first octant if and only if
(3.3.1) gives
n̂ dS = ± u2 v , uv 2 , −3u2 v 2 dudv
and
ZZ ZZ z F
}| {
(uv 2 , u2 v , uv) · − u2 v , −uv 2 , 3u2 v 2 dudv
F · n̂ dS =
S S
Z 1 Z 3 Z 1 Z 3
3 3 3 3
= du dv u v = du u dv v
0 0 0 0
1 34 81
= =
4 4 16
3.3.6.30. ∗. Solution. (a) We start by just sketching the curve z = ey ,
considering the yz-plane as the plane x = 0 in R3 . This curve is the red
curve in the figure below. Concentrate on any one point on that curve. It
is the blue dot at (0, Y, eY )
z
p0, Y, eY q
x
in the figure. When our curve is rotated about the y-axis, the blue dot
sweeps out a circle. The circle that the blue dot sweeps out
• lies in the vertical plane y = Y and
z
(0, Y, eY )
Y Y
(e sin θ, Y, e cos θ)
θ
eY
x
end view
The coordinates of the red dot are e sin θ , Y , eY cos θ . This also
Y
x(Y, θ) = eY sin θ
y(Y, θ) = Y
z(Y, θ) = eY cos θ
0 ≤ Y ≤ 1, 0 ≤ θ < 2π
Finally here is a sketch of the part of the surface in the first octant, x, y, z ≥
0.
z
x2 +z 2 = ey
x
(b) We are using the parametrization
so that
ı̂ı ̂ k̂
∂r ∂r
eY cos θ = − eY sin θ, e2Y , −eY cos θ ,
× = det eY sin θ 1
∂Y ∂θ
eY cos θ 0 −eY sin θ
and, by (3.3.1),
∂r ∂r p p
dS = × dY dθ = e2Y + e4Y dY dθ = eY 1 + e2Y dY dθ
∂Y ∂θ
So the integral is
ZZ Z 1 Z 2π p Z 1 p
ey dS = dY dθ e2Y 1 + e2Y = 2π dY e2Y 1 + e2Y
S 0 0 0
2π h i3/2 1
= 1 + e2Y
3
0
APPENDIX D. SOLUTIONS TO EXERCISES 541
2π h i
= (1 + e2 )3/2 − 23/2
3
(c) Again, we are using the parametrization
so that
∂r ∂r
= − eY sin θ, e2Y , −eY cos θ ,
×
∂Y ∂θ
and, by (3.3.1),
∂r ∂r
dY dθ = ± − eY sin θ, e2Y , −eY cos θ dY dθ
n̂ dS = ± ×
∂Y ∂θ
We choose the “+” sign so that n̂ points towards the y-axis. As an example,
when 0 ≤ θ ≤ π2 , then z = eY cos θ > 0 while the z-coordinate of n̂ is
−ey cos θ < 0. So the integral is
ZZ
F · n̂ dS
S
x z
Z 1 Z 2π z }| { z }| {
dθ eY sin θ, 0, eY cos θ · − eY sin θ, e2Y , −eY cos θ
= dY
0 0
Z 1 Z 2π Z 1
2Y
=− dY = −2π dθ e dY e2Y
0 0 0
= −π e2 − 1 = π 1 − e2
V = (x, y, z) 1 ≤ x2 + y 2 + z 2 ≤ 4
x2 + y 2 = 4Z 2
θ y
x
So we can parametrize S by
π
r(θ, Z) = 2Z sin θ ı̂ı + 2Z cos θ ̂ + Z k̂ 0≤θ≤ , 0≤Z≤1
4
So
∂r
= 2Z cos θ ı̂ı − 2Z sin θ ̂
∂θ
∂r
= 2 sin θ ı̂ı + 2 cos θ ̂ + k̂
∂Z
so that
ı̂ı ̂ k̂
∂r ∂r
× = det 2Z cos θ −2Z sin θ 0 = − 2Z sin θ, −2Z cos θ, 4Z ,
∂θ ∂Z
2 sin θ 2 cos θ 1
and, by (3.3.1),
√
∂r ∂r
dS =
× dθdZ = 20 Z dθdZ
∂θ ∂Z
and
√
ZZ √ Z 1 Z π/4 √ π 1 5π
z 2 dS = 20 dZ dθ Z 3 = 20 =
S 0 0 4 4 8
3.3.6.33. ∗. Solution. We’ll start by parametrizing S. Note that as
x2 + y 2 runs from 0 to 4, z runs from 5 to 1, and that, for each fixed
1 ≤ Z ≤ 5, the cross-section of S with z = Z is the circle x2 + y 2 = 5 − Z,
z = Z. So we may parametrize S by
√ √
r(θ, Z) = 5 − Z cos θ ı̂ı + 5 − Z sin θ ̂ + Z k̂
0 ≤ θ ≤ 2π, 1 ≤ Z ≤ 5
Since
∂r √ √
= − 5 − Z sin θ ı̂ı + 5 − Z cos θ ̂
∂θ
∂r 1 1
=− √ cos θ ı̂ı − √ sin θ ̂ + k̂
∂Z 2 5−Z 2 5−Z
so that
ı̂ı ̂ k̂
∂r ∂r √ √
× = det − 5 − Z sin θ 5 − Z cos θ 0
∂θ ∂Z 1 1
− 2√5−Z cos θ − 2√5−Z sin θ 1
APPENDIX D. SOLUTIONS TO EXERCISES 543
√ √
= 5 − Z cos θ , 5 − Z sin θ , 1/2 ,
(3.3.1) gives
∂r ∂r √ √
n̂ dS = ± × dθdZ = ± 5 − Z cos θ , 5 − Z sin θ , 1/2 dθdZ
∂θ ∂Z
Choosing the minus sign to give the downward pointing normal
ZZ
F · n̂ dS
S
x3 xy 2
Z 5 Z 2π
1
z }| { z }| {
=− dZ dθ − [5 − Z]3/2 cos3 θ − [5 − Z]3/2 cos θ sin2 θ ,
1 0 2
y3 z 2
1
z }| { z}|{
3/2 3
− [5 − Z] sin θ , Z 2
√ 2 √
· 5 − Z cos θ , 5 − Z sin θ , 1/2
Z 5 Z 2π 1
=− dZ dθ − [5 − Z]2 cos4 θ − [5 − Z]2 cos2 θ sin2 θ
1 0 2
1 1
− [5 − Z]2 sin4 θ + Z 2
2 2
Since
1 1 1 2 1
cos4 θ + cos2 θ sin2 θ + sin4 θ = cos2 θ + sin2 θ =
2 2 2 2
the flux
ZZ Z 5 Z 2π 1
1
F · n̂ dS = dZ [5 − Z]2 − Z 2
dθ
S 1 0 2 2
Z 5
=π dZ [5 − Z]2 − Z 2
1
5
Z3 43 53
1 1
= π − [5 − Z]3 − =π − +
3 3 1 3 3 3
= −20 π
√
3.3.6.34. ∗. Solution. The
q surface is z = f (x, y) with f (x, y) = 2xy.
py x
Since fx = 2x and fy = 2y , (3.3.2) gives
r
q
2 2
y x
dS = 1 + fx + fy dxdy = 1 + + dxdy
2x 2y
s
2xy + y 2 + x2
= dxdy
2xy
x+y
=√ dxdy
2xy
y=1
xy = 2
x=1
x
So the mass is
ZZ Z 2 Z 2/x
x+y
ρ(x, y, z) dS = dx dy 3f (x, y) √
S 1 1 2xy
Z 2 Z 2/x
= dx dy 3(x + y)
1 1
Z 2 2/x Z 2
1 2 1
=3 xy + y 2
dx =3 dx 2 + 2 − x −
1 2 1 1 x 2
( 2 2 )
2
3 2 x 3 1
=3 − − =3 −1+2−2+
2 x 1 2 1 2 2
=3
S = (x, y, z) x = y 2 + z 2 , x ≤ 2y
= (x, y, z) x = y 2 + z 2 , y 2 + z 2 ≤ 2y
= (x, y, z) x = y 2 + z 2 , (y − 1)2 + z 2 ≤ 1
= (x, y, z) x = y 2 + z 2 , (y, z) in D
where D = (x, y) (y − 1)2 + z 2 ≤ 1 is a disk with radius 1. Hence
ZZ ZZ ZZ
F · n̂ dS = (2, z, y) · (1, −2y, −2z) dydz = (2 − 4yz) dydz
S D D
= ̂ + (1 − 2x − 6y) k̂
∇ × F · n̂ dS = (2y + 1 − 2x − 6y) dxdy = (1 − 2x − 4y) dxdy
3.3.6.37. Solution. Due to the symmetry of the surface and the vector
field under reflection in the xy-plane, i.e. under z → −z, it is sufficient to
compute the integral over the upper half of the surface, where z ≥ 0, and
then multiply the result by 2. The upper half of the surface consists of two
pieces, S1 and S2 , where S1 is the part on the sphere and S2 is the part
on the hyperboloid. S1 and S2 intersect on a circle. The circle is obtained
by imposing the two equations x2 + y 2 + z 2 = 16 and x2 + y 2 − z 2 = 8
simultaneously. √ Thus we have x2 + y 2 = 12 and z = 2, or in cylindrical
coordinates r = 12, z = 1, on the circle. Here is a sketch of a cross-section
of the apple core.
z
S1
√
r= 12, z = 2
φ1
S2
r
So
∂r
= −4 sin θ sin ϕı̂ı + 4 cos θ sin ϕ ̂
∂θ
∂r
= 4 cos θ cos ϕı̂ı + 4 sin θ cos ϕ ̂ − 4 sin ϕ k̂
∂ϕ
ı̂ı ̂ k̂
∂r ∂r
× = det −4 sin θ sin ϕ 4 cos θ sin ϕ 0
∂θ ∂ϕ
4 cos θ cos ϕ 4 sin θ cos ϕ −4 sin ϕ
APPENDIX D. SOLUTIONS TO EXERCISES 546
= −4 (sin ϕ) r(θ, ϕ)
and, by (3.3.1),
∂r ∂r
n̂ dS = ± × dθdϕ = ∓4 (sin ϕ) r(θ, ϕ) dθdϕ
∂θ ∂ϕ
To get the outward pointing normal, i.e. the normal point in the same
direction as r(θ, ϕ), we take the plus sign. As F = r(θ, ϕ),
42
z }| {
F · n̂ dS = 4 |r(θ, ϕ)|2 sin ϕ dθ dϕ = 64 sin ϕ dθ dϕ
and
ZZ Z 2π Z π/3 h iπ/3
F · n̂ dS = 64 dθ dϕ sin ϕ = 64 · 2π − cos ϕ = 64π
S1 0 0 0
for the hyperboloid and we always have x = r cos θ and y = r sin θ. Thus
the hyperboloid has the following parametrization:
Note that ∂R ∂R
∂θ × ∂z is pointing downward (since z > 0) and hence outward.
Since F · ∂θ × ∂R
∂R
∂z = (x, y, z) · (x, y, −z) = x2 + y 2 − z 2 = 8 on S2 , we
have
ZZ ZZ Z 2 Z 2π
F · n̂ dS = F · (Rθ × Rz ) dθ dz = dz dθ 8 = 32π
S2 S2 0 0
Finally, the flux integral over the whole apple core surface is
Z Z ZZ
2 F · n̂ dS + F · n̂ dS = 2 64π + 32π = 192π
S1 S2
APPENDIX D. SOLUTIONS TO EXERCISES 547
G(x, y, z) = x2 + z 2 − cos2 y = 0
1 π 1
∇G , , = 2x , 2 sin y cos y , 2z 1 π 1 = (1, 1, 1)
2 4 2 2, 4 ,2)
(1, 1, 1) · x − 1/2 , y − π/4 , z − 1/2 = 0 or x + y + z = 1 + π/4
3.3.6.40. ∗. Solution. (a) (i) Here is a sketch of the part of the plane
in question.
z
(0, 0, 16/3)
y
(0, 4, 0)
(8, 0, 0)
x
We can use x and y as parameters. As we can rewrite the equation of
the plane as z = 13 16 − 2x − 4y , we have the parametrization
1
r(x, y) = xı̂ı + y ̂ +16 − 2x − 4y k̂
3
In terms of x and y, the condition z = 13 16−2x−4y ≥ 0 is 16−2x−4y ≥ 0
or x + 2y ≤ 8. So the domain is
(x, y) x ≥ 0, y ≥ 0, x + 2y ≤ 8
√ x2 + y 2 + z 2 = 16
z=2 2
S
x
The full sphere can be parametrized, using spherical coordinates with
ρ = 4, by
So our parametrization is
p
If we use x and y as parameters, then, since z = 1 + x2 + y 2 , we have
the parametrization
p
r(x, y) = xı̂ı + y ̂ + 1 + x2 + y 2 k̂
So the domain is
(x, y) x2 + y 2 ≤ 99
and, by (3.3.1),
∂r ∂r
dS = × dθdϕ
∂θ ∂ϕ
q
= 16 sin ϕ cos2 θ sin2 ϕ + sin2 θ sin2 ϕ + cos2 ϕ dθdϕ
= 16 sin ϕ dθdϕ
So the area is
ZZ Z π/4 Z 2π Z π/4
Area = dS = dϕ dθ 16 sin ϕ = 32π dϕ sin ϕ
S 0 0 0
h iπ/4
= 32π − cos ϕ
0
h 1 i
= 32π 1 − √
2
3.3.6.41. ∗. Solution. Solution 1 — using tweaked spherical coordi-
nates.
√ have to parametrize S. It is natural to use spherical coordinates
First we
with ρ = 2. However if we use the standard spherical coordinates
√ √ √
x = 2 sin ϕ cos θ y = 2 sin ϕ sin θ z = 2 cos ϕ
(3.3.1) gives
∂r ∂r
n̂ dS = ± × dθdϕ = ±2 sin ϕ sin ϕ cos θ , cos ϕ , sin ϕ sin θ dθdϕ
∂θ ∂ϕ
∂r ∂r
dS = × dθdϕ = 2 sin ϕ dθdϕ
∂θ ∂ϕ
y3
ZZ Z π/4 Z 2π z√ }| {
3
y 3 dS = 2 dϕ dθ sin ϕ 2 cos ϕ
S 0 0
√ Z π/4
= 8 2π dϕ sin ϕ cos3 ϕ
0
π/4
√ cos4 ϕ
= 8 2π −
4
0
√
1 3
= 2 2π 1 − =√ π
4 2
Since
∂r x
= ı̂ı − √ ̂
∂x 2 − x2 − z 2
∂r z
= −√ ̂ + k̂
∂z 2 − x2 − z 2
so that
ı̂ı ̂ k̂
∂r ∂r
×
1
= det − √2−xx2 −z2 0
∂x ∂z
0 − √2−xz2 −z2 1
x z
= −√ ı̂ı − ̂ − √ k̂
2
2−x −z 2 2 − x2 − z 2
(3.3.1) gives
∂r ∂r
n̂ dS = ± × dxdz
∂x
∂z
x z
=∓ √ ,1, √ dxdz
2 − x2 − z 2 2 − x2 − z 2
∂r ∂r
dS = × dxdz
∂x ∂z
r √
x2 + z 2 2
= 1+ 2 2
dxdz = √ dxdz
2−x −z 2 − x2 − z 2
Choose the plus sign to give the outward pointing normal.
(a) The specified integral is
y3
ZZ ZZ √ z }| {
2 3/2
y 3 dS = dxdz √ 2 − x2 − z 2
S x2 +z 2 ≤1 2 − x2 − z 2
√ ZZ
dxdz 2 − x2 − z 2
= 2
x2 +z 2 ≤1
4 · Integral Theorems
4.1 · Gradient, Divergence and Curl
4.1.6 · Exercises
4.1.6.1. ∗. Solution. (a) A. The angle between F and dr is less than
90◦ along the entire path. So F · dr > 0 along the entire path and the work
is positive. R
(b) B. F is perpendicular to dr along all of C2 . So C2 F · dr = 0.
(c) C. It looks like Px = Qy = 0 at N . So ∇ · F = 0 at N .
(d) A. At Q, the vertical component of F is increasing from left to right
(so that Qx > 0) and the horizontal component of F is decreasing from
bottom to top (so that Py < 0). So Qx − Py > 0 at N .
(e) B. At D, the horizontal component of F is increasing from left to
right, so that Px > 0.
4.1.6.2. Solution. No. The vector field F(x, y, z) = ı̂ı + y k̂ has
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x ∂y ∂z
1 0 y
= ı̂ı
has dot product 1 with F(x, y, z) (for all x, y, z) and so is not perpendicular
to it.
4.1.6.3. Solution. (a) By the product rule
∂ ∂ ∂
∇ · (f F) = (f F1 ) + (f F2 ) + (f F3 )
∂x ∂y ∂z
∂F1 ∂F2 ∂F3
= f +f +f
∂x ∂y ∂z
∂f ∂f ∂f
+ F1 + F2 + F3
∂x ∂y ∂z
= f ∇ · F + F · ∇f
(b) Again by the product rule
∇ · (F × G)
∂ ∂ ∂
= (F2 G3 − F3 G2 ) + (F3 G1 − F1 G3 ) + (F1 G2 − F2 G1 )
∂x ∂y ∂z
∂F2 ∂F3 ∂F3 ∂F1 ∂F1 ∂F2
= G3 − G2 + G1 − G3 + G2 − G1
∂x ∂x ∂y ∂y ∂z ∂z
∂G3 ∂G2 ∂G1 ∂G3 ∂G2 ∂G1
+ F2 − F3 + F3 − F1 + F1 − F2
∂x ∂x ∂y ∂y ∂z ∂z
∂F3 ∂F2 ∂F1 ∂F3 ∂F2 ∂F1
= − G1 + − G2 + − G3
∂y ∂z ∂z ∂x ∂x ∂y
∂G3 ∂G2 ∂G1 ∂G3 ∂G2 ∂G1
− F1 − − F2 − − F3 −
∂y ∂z ∂z ∂x ∂x ∂y
= G · (∇∇ × F) − F · (∇ ∇ × G)
∂ ∂ ∂
∇ (f g) = ı̂ı (f g) + ̂ (f g) + k̂ (f g)
∂x ∂y ∂z
APPENDIX D. SOLUTIONS TO EXERCISES 555
∂f ∂f ∂f
= ı̂ıg + ̂g + k̂g
∂x ∂y ∂z
∂g ∂g ∂g
+ ı̂ıf + ̂f + k̂f
∂x ∂y ∂z
∇f + f∇
= g∇ ∇g
∇ 2 (f g) = ∇ · g∇
∇f + ∇ · f∇∇g
= g ∇ · ∇f + ∇g · ∇f + f ∇ · ∇g + ∇f · ∇g
= f ∇ 2 g + 2∇
∇f · ∇ g + g ∇ 2 f
(b) By definition
∂ ∂ ∂
∇ · (xy 2ı̂ı − yz 2̂ + zx2 k̂) = xy 2 + − yz 2 + zx2
∂x ∂y ∂z
= y 2 − z 2 + x2
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (xy 2ı̂ı − yz 2̂ + zx2 k̂) = det ∂x ∂y ∂z
xy −yz zx2
2 2
(c) By definition
!
x y
∇· p ı̂ı + p ̂
x2 + y 2 x2 + y 2
! !
∂ x ∂ y
= p + p
∂x x2 + y 2 ∂y x2 + y 2
1 x2 1 y2
=p − 3/2
+p − 3/2
x2 + y 2 [x2 + y 2 ] x2 + y 2 [x2 + y 2 ]
x2 + y 2 − x2 + x2 + y 2 − y 2
= 3/2
[x2 + y 2 ]
1
=p
x2 + y2
!
x y
∇× p ı̂ı + p ̂
x2 + y 2 x2 + y 2
ı̂ı ̂ k̂
∂ ∂ ∂
= det ∂x ∂y ∂z
√ 2x 2 √ 2y 0
x +y x +y 2
APPENDIX D. SOLUTIONS TO EXERCISES 556
!
xy xy
= − 3/2
+ 3/2
k̂
[x2 + y 2 ] [x2 + y 2 ]
=0
(d) By definition
!
y x
∇· −p ı̂ı + p ̂
x2 + y 2 x2 + y 2
! !
∂ y ∂ x
= −p + p
∂x x2 + y 2 ∂y x2 + y 2
xy xy
= 3/2
− 3/2
2
[x + y ] 2 [x + y 2 ]
2
=0
!
y x
∇ × −p ı̂ı + p ̂
x2 + y 2 x2 + y 2
ı̂ı ̂ k̂
∂ ∂ ∂
= det
∂x ∂y ∂z
y x
−√ 2 2 √ 2 2 0
x +y x +y
!
1 x2 1 y2
= p − 3/2
+p − 3/2
k̂
x2 + y 2 [x2 + y 2 ] x2 + y 2 [x2 + y 2 ]
x2 + y 2 − x2 + x2 + y 2 − y 2
= 3/2
k̂
[x2 + y 2 ]
k̂
=p
x2 + y2
r
4.1.6.5. ∗. Solution. (a) We are to compute the divergence of r =
x ı̂ı+y ̂+z k̂
[x2 +y 2 +z 2 ]1/2
. Since
∂ x 1 1 x(2x)
1/2
= 1/2
−
∂x [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]3/2
y2 + z2
= 3/2
[x2 + y 2 + z 2 ]
∂ y 1 1 y(2y)
1/2
= 1/2
−
∂y [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]3/2
x2 + z 2
= 3/2
[x2 + y 2 + z 2 ]
∂ z 1 1 z(2z)
= −
∂z [x2 + y 2 + z 2 ]1/2 [x2 + y 2 + z 2 ]
1/2 2 [x2 + y 2 + z 2 ]3/2
x2 + y 2
= 3/2
[x2 + y 2 + z 2 ]
(b)
ı̂ı ̂ k̂
∇ × yz ı̂ı + 2xz ̂ + exy k̂ = det ∂x
∂ ∂ ∂
∂y ∂z
xy
yz 2xz e
= xexy − 2x ı̂ı − yexy − y ̂ + z k̂
k/2
4.1.6.6. ∗. Solution. (a) Since rk = x2 + y 2 + z 2 ,
∂ k k 2 k −1
r = 2x x + y 2 + z 2 2 = k (r · ı̂ı) rk−2
∂x 2
∂ k k 2 k −1
r = 2y x + y 2 + z 2 2 = k (r · ̂) rk−2
∂y 2
∂ k k 2 k −1
r = 2z x + y 2 + z 2 2 = k (r · k̂) rk−2
∂z 2
We want k = −3.
(b) Using the computation in part (a)
∂ ∂ ∂
∇ · (rk r) = (xrk ) + (yrk ) + (zrk )
∂x ∂y ∂z
∂ ∂ ∂
= 3rk + x rk + y rk + z rk
∂x ∂y ∂z
= 3rk + x kx rk−2 + y ky rk−2 + z kz rk−2
= 3 + k rk
We want k = 2.
(c) Recalling that ∇ 2 = ∇ · ∇ ,
∇ 2 (rk ) = ∇ · ∇ (rk )
We want k = −2.
4.1.6.7. ∗. Solution. (a)
∂x ∂y ∂z
∇·r= + + =3
∂x ∂y ∂z
(b)
2 ∂ ∂ ∂
x2 + y 2 + z 2 = 2xı̂ı + 2y ̂ + 2z k̂ = 2r
∇ (r ) = ı̂ı + ̂ + k̂
∂x ∂y ∂z
(c) Since
ı̂ı ̂ k̂
r × a = det x y z = ı̂ı a3 y−a2 z + ̂ a1 z−a3 x + k̂ a2 x−a1 y
a1 a2 a3
we have
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (r × a) = det
∂x ∂y ∂z
a3 y − a2 z a1 z − a3 x a2 x − a 1 y
APPENDIX D. SOLUTIONS TO EXERCISES 558
(d) Since
∂ ∂ ∂ 1/2
∇ (r) = ı̂ı + ̂ + k̂ x2 + y 2 + z 2
∂x ∂y ∂z
x y x
= ı̂ı 1/2 + ̂ 1/2 + k̂ 1/2
2 2
x +y +z 2 2
x +y +z2 2 x + y + z2
2 2
we have
∂ x ∂ y
∇ · ∇(r) = +
∂x x2 + y 2 + z 2 1/2 ∂y x2 + y 2 + z 2 1/2
∂ z
+
∂z x2 + y 2 + z 2 1/2
3 1 2x2 + 2y 2 + 2z 2
= 1/2 − 2 3/2
x2 + y 2 + z 2 x2 + y 2 + z 2
2
= 1/2
x2 + y 2 + z 2
2
=
r
4.1.6.8. ∗. Solution. (a) Since
1 ∂ ∂ ∂ −1/2
∇ = ı̂ ı + ̂ + k̂ x2 + y 2 + z 2
r ∂x ∂y ∂z
x y z
= −ı̂ı 3/2 − ̂ 3/2 − k̂ 3/2
2
x +y +z 2 2 2 2
x +y +z 2 x + y + z2
2 2
x y x
= −ı̂ı 3 − ̂ 3 − k̂ 3
r r r
we have a = −3.
(b) Since
∂ h 2 1/2 i ∂ h 2 1/2 i
x + y2 + z2 x + y2 + z2
∇ · rr = x + y
∂x ∂y
∂ h 2 1/2 i
+ x + y2 + z2 z
∂z
1/2 1 2x2 + 2y 2 + 2z 2
= 3 x2 + y 2 + z 2 +
2 x2 + y 2 + z 2 1/2
1/2
= 4 x2 + y 2 + z 2
= 4r
we have a = 4.
(c) Since
∂ ∂ ∂ 3/2
∇ (r3 ) = ı̂ı + ̂ + k̂ x2 + y 2 + z 2
∂x ∂y ∂z
2 1/2
2 2
1/2 1/2
+ ̂ 3y x2 + y 2 + z 2 + k̂ 3z x2 + y 2 + z 2
= ı̂ı 3x x + y + z
= 3rr
APPENDIX D. SOLUTIONS TO EXERCISES 559
we have
∇ · ∇ (r3 ) = ∇ · 3rr = 3 ∇ · rr = 3 4r
by part (b)
= 12r
so that a = 12.
∂ ∂ ∂
4.1.6.9. Solution. (a) Since ∇ · F = ∂x (1 + yz) + ∂y (2y + zx) + ∂z (3z 2 +
xy) = 2 + 6z 6= 0, F fails the screening test and cannot have a vector
potential.
(b) The vector field A = A1ı̂ı + A2̂ is a vector potential for G if and
only if G = ∇ × A, which is the case if and only if
∂A2 1
− = yz ⇐⇒ A2 = − yz 2 + B2 (x, y)
∂z 2
∂A1 1 2
= zx ⇐⇒ A1 = xz + B1 (x, y)
∂z 2
∂A2 ∂A1 ∂B2 ∂B1
− = xy ⇐⇒ − = xy
∂x ∂y ∂x ∂y
So the divergence of F is
∂ −z ∂ ∂ x
∇·F= + (y) + =1
∂x x2 + z 2 ∂y ∂z x2 + z 2
APPENDIX D. SOLUTIONS TO EXERCISES 560
(d) By part (b), the vector field passes the conservative field screening
test ∇ × F = 0. But we should still be suspicious because of the similarity
of F to the vector field of Examples 2.3.14 and 4.3.8.
So let’s compute the line integral of F around the (closed) circle y = 0,
x2 + z 2 = 1, parametrized by
2
−z
2
x r0 (t)
x2 +y 2
Z Z 2π zx }|
+z
{ z}|{ z }| {
F · dr = − sin t ı̂ı + cos t k̂ · − sin tı̂ı + cos t k̂ dt
C 0
Z 2π
= dt = 2π
0
As the integral of F around the simple closed curve C is not zero, F cannot
be conservative on D. See Theorem 2.4.7 and Examples 2.3.14 and 4.3.8.
4.1.6.11. ∗. Solution. (a) By the vector identity of Theorem 4.1.7.a,
∇·F=∇·∇×G=0
So we must have
0 = ∇ · F = ∇ · (xz + xy)ı̂ı + α(yz − xy)̂ + β(yz + xz)k̂
= (z + y) + α(z − x) + β(y + x)
gy = xz, gx = yz
The first equation, gy = xz, is satisfied if and only if g = xyz + h(x, z).
The second equation is also satisfied if and only if gx = yz + hx (x, z) = yz.
This is the case if and only if hx (x, z) = 0. That is, if and only if h is
independent of x. Equivalently, if and only if h(x, z) = w(z) for some
function w(z). So, in fact, any function of the form g(x, y, z) = xyz + w(z)
will work.
4.1.6.12. Solution. (a) Denote by θ the angle between â and r. The
point r is a distance ` = |r| sin θ from the axis of rotation. So as the body
rotates, the point sweeps out a circle of radius ` centred on the axis of
rotation.
APPENDIX D. SOLUTIONS TO EXERCISES 561
Ω
â
v
r
θ
0
In one second the point sweeps out an arc of this circle that subtends
Ω
an angle of Ω radians. This arc is the fraction 2π of a full circle and so
Ω
has length 2π 2π` = Ω` = Ω|r| sin θ. Thus the point is moving with speed
Ω|r| sin θ. The velocity vector of the point must have length Ω|r| sin θ and
direction perpendicular to both â and r. The vector Ω × r is perpendicular
to both r and Ω = Ωâ and has length |Ω Ω| |r| sin θ = Ω|r| sin θ as desired.
So the velocity vector is either Ω × r or its negative. By the right hand
rule it is Ω × r.
(b) By vector identities
∇ × F) − F · (∇
∇ · (F × G) = G · (∇ ∇ × G)
∇ × (F × G) = F(∇∇ · G) − (∇
∇ · F)G + (G · ∇ )F − (F · ∇ )G
(which are Theorems 4.1.4(d) and 4.1.5(d)) and the assumption that Ω is
constant
∇ × (Ω ∇ · r) − (∇
Ω × r) = Ω (∇ ∇ · Ω )r + (r · ∇ )Ω
Ω − (Ω
Ω · ∇ )r
∇ · r) − (Ω
= Ω (∇ Ω · ∇ )r
∇ · (Ω ∇ × Ω ) − Ω · (∇
Ω × r) = r · (∇ ∇ × r) = −Ω ∇ × r)
Ω · (∇
Substituting in
∂x ∂y ∂z
∇·r= + + =3
∂x ∂y ∂z
∂z ∂y ∂x ∂z ∂y ∂x
∇×r= − ı̂ı + − ̂ + − k̂ = 0
∂y ∂z ∂z ∂x ∂x ∂y
∂ ∂ ∂
Ω · ∇ )r = Ω1
(Ω + Ω2 + Ω3 xı̂ı + ŷ + z k̂
∂x ∂y ∂z
= Ω1ı̂ı + Ω2̂ + Ω3 k̂ = Ω
gives
∇ × (Ω
Ω × r) = 2 Ω ∇ · (Ω
Ω × r) = 0
(c) The students are a distance 6378 sin(90◦ − 49◦ ) = 6378 cos(49◦ ) =
4184 km from the axis of rotation. The rate of rotation is Ω = 2π
24 radians
per hour. In each hour the students sweep out an arc of 2π
24 radians from a
circle of radius 4184 km. Their speed is 2π
24 × 4184 = 1095km/hr.
∂G3 ∂G2
4.1.6.13. Solution. We shall show that ∂y − ∂z = F1 . The other
components are similar. First we have
dr
t F r(t) × (t) = t F tx, ty, tz × xı̂ı + y ̂ + z k̂
dt
APPENDIX D. SOLUTIONS TO EXERCISES 562
ı̂ı ̂ k̂
= t det F1 F2 F3
x y z
So
Z 1
∂G3 h ∂F1 ∂F2 i
= t F1 tx, ty, tz + tx, ty, tz ty − tx, ty, tz tx dt
∂y 0 ∂y ∂y
Z 1 h
∂G2 ∂F3 ∂F1 i
= t tx, ty, tz tx − tx, ty, tz tz − F1 tx, ty, tz dt
∂z 0 ∂z ∂z
so that
Z 1
∂G3 ∂G2 h
− = 2t F1 tx, ty, tz
∂y ∂z 0
∂F1 ∂F1
+ t2 y tx, ty, tz + t2 z
tx, ty, tz
∂y ∂z
∂F 2 ∂F 3 i
− t2 x tx, ty, tz − t2 x
tx, ty, tz dt
∂y ∂z
∂F1 ∂F2 ∂F3
Since, by hypothesis, ∇ · F = ∂x + ∂y + ∂z = 0, the last two terms
n ∂F ∂F3 o n ∂F o
2 1
−t2 x tx, ty, tz + tx, ty, tz = −t2 x − tx, ty, tz
∂y ∂z ∂x
so that
∂G3 ∂G2
−
∂y ∂z
Z 1h
∂F1 ∂F1
2tF1 tx, ty, tz + t2 x tx, ty, tz + t2 y
= tx, ty, tz
0 ∂x ∂y
∂F1 i
+ t2 z tx, ty, tz dt
∂z
Z 1 h it=1
d 2 i h
= t F1 (tx, ty, tz) dt = t2 F1 (tx, ty, tz)
0 dt t=0
= F1 (x, y, z)
(b) Define the vector field F(x, y, z) = f (x, y, z) k̂. Then the divergence
of F is ∇ ·F(x, y, z) = ∂f
∂z (x, y, z). The boundary of the cube V is the union
of six faces
S1 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, z = 1 with outward normal n̂ = k̂
S2 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, z = 0 with outward normal n̂ = −k̂
S3 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ z ≤ 1, y = 1 with outward normal n̂ = ̂
S4 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ z ≤ 1, y = 0 with outward normal n̂ = −̂
S5 = (x, y, z) 0 ≤ y ≤ 1, 0 ≤ z ≤ 1, x = 1 with outward normal n̂ = ı̂ı
S6 = (x, y, z) 0 ≤ y ≤ 1, 0 ≤ z ≤ 1, x = 0 with outward normal n̂ = −ı̂ı
z
n̂ = k̂
(0, 1, 1)
S1
n̂ = −̂ n̂ = ̂
y
S2
(1, 1, 0)
x
n̂ = −k̂
Observe that
+f
on S1
F · n̂ = f k̂ · n̂ = −f on S2
0 on S3 , S4 , S5 , S6
ZZ ZZZ
1 1
(x2 + y 2 + z 2 ) n̂ dS = (2xı̂ı + 2y ̂ + 2z k̂) dV
2|V | ∂V 2|V | V
= (x̄, ȳ, z̄)
4.2.6.3. Solution. (a) We’ll parametrize the sphere using the spherical
coordinates θ and ϕ.
x = sin ϕ cos θ
y = sin ϕ sin θ
z = cos ϕ
To get an outward pointing normal we need the + sign, since then n̂(θ, ϕ)
is a positive multiple, namely sin ϕ, times r(θ, ϕ). So, on S,
F
z }| {
F · n̂ dS = sin ϕ sin ϕ cos θ , sin ϕ sin θ , cos2 ϕ
· sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
= sin ϕ sin2 ϕ cos2 θ + sin2 ϕ sin2 θ + cos3 ϕ
and
ZZ Z π Z 2π
dθ sin ϕ sin2 ϕ cos2 θ + sin2 ϕ sin2 θ + cos3 ϕ
F · n̂ dS = dϕ
S 0 0
Z π 2π Z
dθ sin3 ϕ + sin ϕ cos3 ϕ
= dϕ
0
Z π 0 π
1
= 2π dϕ sin3 ϕ + − cos4 ϕ
0 4
Z π 0 π
1
= 2π dϕ sin ϕ(1 − cos ϕ) = 2π − cos ϕ + cos3 ϕ
2
0 3 0
APPENDIX D. SOLUTIONS TO EXERCISES 565
4
= 2π
3
8π
=
3
(b) Let V be the interior of S. Then, by the divergence theorem,
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (1 + 1 + 2z) dV
S V V
and
ZZ Z π Z 2π
F · n̂ dS = a3 dϕ dθ sin ϕ cos2 ϕ
S 0 0
Z π π
3 2 1 3 3
= 2πa dϕ sin ϕ cos ϕ = 2πa − cos ϕ
0 3 0
4
= πa3
3
(b) Call the solid x2 + y 2 + z 2 ≤ a2 , V . As
∂ ∂ ∂
∇·F= (0) + (0) + (z) = 1
∂x ∂y ∂z
the divergence theorem gives
ZZ ZZZ ZZZ
4 3
F · n̂ dS = ∇ · F dV = dV = Volume(V ) = πa
S V V 3
APPENDIX D. SOLUTIONS TO EXERCISES 566
n̂ = −k̂ dS = dx dy F · n̂ = −y 2
so that ZZ ZZ
F · n̂ dS = − y 2 dx dy
D D
Switching to polar coordinates
ZZ Z 3 Z 2π
2
F · n̂ dS = − dr r dθ r sin θ
D 0 0
Z 3 Z 2π
=− dr r3 dθ sin2 θ
0 0
4
Z 2π
1 − cos(2θ)
3
=− dθ
0 4 2
2π
81 θ sin(2θ)
=− −
4 2 4 0
81
=− π
4
R 2π
For an efficient, sneaky, way to evaluate 0 sin2 θ dθ, see Example 2.4.4.
(b) Observe that RRR∇ · F = x + 2. Since x is odd and V is invariant under
x → −x, we have V
x dV = 0 (more details below) so that
ZZZ ZZZ ZZZ
∇ · F dV = (x + 2) dV = 2 dV = 2|V |
V V V
RRR
Here are two more detailed arguments showing that V
x dV = 0.
9−x2 −y 2
Argument 1: We may rewrite the equation z = 9+x2 +y 2 of the curved
boundary of V as
9(1 − z)
z(9 + x2 + y 2 ) = 9 − x2 − y 2 ⇐⇒ x2 + y 2 =
1+z
q
9(1−z)
This is the equation of the circle of radius r(z) = 1+z centred on
x = y = 0. So z runs from 0 to 1, and for each fixed 0 ≤ z p ≤ 1, y runs
2 2
p −r(z) to r(z) and, for each fixed y and z, x runs from − r(z) − y
from
2 2
to r(z) − y . So
ZZZ Z Z 1 Z √ 2 2r(z) Z r(z) −y
Z 1 r(z)
x dV = dz dy √ dx x = dz dy 0 = 0
V 0 −r(z) − r(z)2 −y 2 0 −r(z)
Ra
since −a x dx = 0 for any a > 0.
Argument 2: As we have observed above, the curved boundary of V is
x2 + y 2 = 9(1−z)
1+z which is invariant under rotations about the z--axis. By
that symmetry, the centroid of V lies on the z-axis. Recall that, for any
solid V , the centroid of V is (x̄, ȳ, z̄) with
RRR RRR RRR
V
xdV V
ydV zdV
x̄ = RRR ȳ = RRR z̄ = RRRV
V
dV V
dV V
dV
So
ZZZ ZZZ
x dV = x̄ Volume(V ) = 0 and y dV = ȳ Volume(V ) = 0
V V
APPENDIX D. SOLUTIONS TO EXERCISES 567
so that
ZZ ZZZ ZZ
81
F · n̂ dS = ∇ · F dV − F · n̂ dS = 2|V | + π
S V D 4
4.2.6.6. Solution. (a) Let G(x, y, z) = x2 + y 2 + z. Then the surface is
G(x, y, z) = 1 and ∇ G(x, y, z) = 2xı̂ı + 2y ̂ + k̂ is upward pointing (i.e. the
coefficient of k̂ is positive) so that, by (3.3.3),
∇G 2xı̂ı + 2y ̂ + k̂
n̂ dS = dx dy = dx dy
∇ G · k̂ 1
= (2xı̂ı + 2y ̂ + k̂) dx dy
F · n̂ dS = xı̂ı + y ̂ + k̂ · 2xı̂ı + 2y ̂ + k̂ dx dy
= 2x2 + 2y 2 + 1 dx dy
z = 1 − x2 − y 2
y
D
x x2 + y 2 = 1
Let D denote the bottom surface of V . The disk D has radius 1, area
π, z = 0 and outward normal −k̂, so that
ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dx dy = − dx dy = −π
D D D
As
∂ ∂ ∂
∇·F= (x) + (y) + (1) = 2
∂x ∂y ∂z
the divergence theorem gives
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS
S V D
ZZZ
= 2 dV − (−π)
V
ZZZ
=π+2 dV
V
APPENDIX D. SOLUTIONS TO EXERCISES 568
RRR
To evaluate the volume V
dV , we slice the V into thin horizontal pan-
cakes. Here is a sketch of the pancake at height z.
z
x2 + y 2 = 1 − z
y
D
x
√
Its cross-section is a circular disk of radius 1 − z, and hence of area
π(1 − z). As the pancake has thickness dz, it has volume π(1 − z) dz. So
ZZ Z 1 ZZ Z 1
F · n̂ dS = π + 2 dz dx dy = π + 2 dz π(1 − z)
S 0 x2 +y 2 ≤1−z 0
1
1 2
= π + 2π z − z = 2π
2 0
∂ ∂ ∂
∇·F= z + sin y + (zy) + sin x cos y
∂x ∂y ∂z
=z
(b) Let
(x, y, z) x2 + y 2 + z 2 ≤ 9
V =
By the divergence theorem (note that we are to find the outward flux),
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = z dV = 0
∂V V V
since z is odd.
4.2.6.8. Solution. Call the silo V . Call the sides and top of the silo S.
Call the base of the silo (namely, x2 + y 2 ≤ 1, z = 0) B. By the divergence
theorem,
ZZ ZZ ZZZ
V · n̂ dS + V · (−k̂) dS = ∇ · V dV
S B
ZZ ZZ Z Z ZV
V · n̂ dS − (x2 + y) dx dy = (2xyz + z) dV
S x2 +y 2 ≤1 V
RR RRR
By oddness under Y → −y, x2 +y 2 ≤1
y dx dy = V
xyz dV = 0, so
ZZ ZZ ZZZ
V · n̂ dS = x2 dx dy + z dV
S x2 +y 2 ≤1 V
x
Z 1 Z 2π z }| { ZZZ
2
= dr dθ r (r cos θ) + z dV
0 0 V
We can evaluate the volume integral by decomposing V into thin horizontal
pancakes. See Section 1.6 in the CLP-2 text. For 0 ≤ z ≤ 1, the horizontal
APPENDIX D. SOLUTIONS TO EXERCISES 569
and
RRR RRR RRR
x dVV
y dV
V
z dV
V
x̄ = ȳ = z̄ =
Volume(V) Volume(V) Volume(V)
where (x̄, ȳ, z̄) is the centroid of V. Our ball has volume V and centroid
(x̄, ȳ, z̄) = (x0 , y0 , z0 ). So
ZZ
F · n̂ dS = V [3 + 3x̄ − ȳ] = [3 + 3x0 − y0 ] V
S
V = (x, y, z) x2 + y 2 ≤ 1 − z 4 , 0 ≤ z ≤ 1
∂ x ∂ y
∇ · F(x, y, z) = +
∂x x2 + y 2 + z 2 3/2 ∂y x2 + y 2 + z 2 3/2
∂ z
+
∂z x2 + y 2 + z 2 3/2
x + y 2 + z 2 − x 23 (2x) x + y 2 + z 2 − y 23 (2y)
2 2
= 5/2 + 5/2
x2 + y 2 + z 2 x2 + y 2 + z 2
x + y 2 + z 2 − z 23 (2z)
2
+ 5/2
x2 + y 2 + z 2
3 x2 + y 2 + z 2 − 3x2 − 3y 2 − 3z 2
= 5/2
x2 + y 2 + z 2
=0
r
unit normal to σa is n̂ = |r| so that
Z Z Z Z
r r 1 1
F · n̂ dS = 3
· dS = 2
dS = 2 dS
σa |r|=a |r| |r| |r|=a |r| a |r|=a
1
= 2 4πa2 = 4π 6= 0
a
(c) No, the results of (a) and (b) do not contradict the divergence
theorem. One hypothesis of the divergence theorem is that ∇ · F (in fact
all first order derivatives of F) be defined and continuous throughout the
solid that ∇ · F is to be integrated over. That hypothesis is violated in this
case.
(d) Let’s first figure out what the surface z 2 − x2 − y 2 + 1 = 0, i.e. the
surface x2 + y 2 = 1 + z 2 , looks like. For each z0 , the z = z0 cross-section
of this surface is the circle x2 + y 2 = 1 + z02 . The radius of this circle is 1
when z0 = 0 and grows as |z0 | increases. So the solid region E looks like
an hourglass drum, as sketched in the figure on the left below.
σa
σ
Ea
σ
We are going to use the divergence theorem to compute the flux of F
out through the surface σ of E. However we cannot apply the divergence
theorem using E as the solid, because F is not defined at the origin, (0, 0, 0),
which is a point in E. So we pick any 0 < a < 1, and define the auxiliary
solid
Ea = (x, y, z) x2 + y 2 + z 2 ≥ a2 , x2 + y 2 ≤ 1 + z 2 , −1 ≤ z ≤ 1
x2 + (y − 2)2 = 1 + z 2
R = (x, y, z) x2 + (y − 2)2 ≤ 1 + z 2 , −1 ≤ z ≤ 1
x = sin ϕ cos θ
y = sin ϕ sin θ
z = cos ϕ
y2 z2 (2ỹ)2 (2z̃)2
1 = x2 + + = x2 + + = x2 + ỹ 2 + z̃ 2
4 4 4 4
and we can parametrize
x = sin ϕ cos θ
ỹ = sin ϕ sin θ
z̃ = cos ϕ
and then
x = sin ϕ cos θ
y = 2ỹ = 2 sin ϕ sin θ
z = 2z̃ = 2 cos ϕ
or
(3.3.1) yields
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ
= ± − sin ϕ sin θ , 2 sin ϕ cos θ , 0
× (cos ϕ cos θ , 2 cos ϕ sin θ , −2 sin ϕ) dθdϕ
= ± − 4 sin2 ϕ cos θ , −2 sin2 ϕ sin θ , −2 sin ϕ cos ϕ dθdϕ
= ∓2 sin ϕ 2 sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
To get an outward pointing normal we need the + sign. For example, with
the + sign, the z-component is 2 sin ϕ cos ϕ = sin(2ϕ) so that the normal
is pointing upward when 0 < ϕ < π2 , i.e. in the northern hemisphere, and
is pointing downward when π2 < ϕ < π, i.e. in the southern hemisphere.
So
F · n̂dS = (sin ϕ cos θ)(4 sin2 ϕ cos θ) + (2 sin ϕ sin θ)(2 sin2 ϕ sin θ)
+ (2 cos ϕ)(2 sin ϕ cos ϕ) dθdϕ
= 4 sin3 ϕ cos2 θ + 4 sin3 ϕ sin2 θ + 4 sin ϕ cos2 ϕ dθdϕ
= 4 sin ϕ dθdϕ
(c) Set
y2 z2
(x, y, z) x2 +
V = 4 + 4 ≤1
Since ∇ · F = 3, the divergence theorem gives
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 3Volume(V )
S V
2 2 2
The volume contained in the ellipsoid, xa2 + yb2 + zc2 = 1, of semiaxes a, b
and c is 34 πabc. In our case a = 1, b = c = 2, so
ZZ
4
F · n̂ dS = 3Volume(V ) = 3 × π(1)(2)(2) = 16π
S 3
which is exactly what we found in part (b).
The volume of the ellipsoid V can also be found by observing that, in
V,
• x runs from −1 to 1 and
• for each fixed −1 ≤ x ≤ 1, (y, z) runs over the disk y 2 +z 2 ≤ 4(1−x2 ),
which has area 4π(1 − x2 ).
That is
(x, y, z) − 1 ≤ x ≤ 1, y 2 + z 2 ≤ 4(1 − x2 )
V =
so that
Z 1 ZZ
Volume(V ) = dx dy dz
−1 y 2 +z 2 ≤4(1−x2 )
APPENDIX D. SOLUTIONS TO EXERCISES 574
Z 1 Z 1
2 2 1
= dx 4π(1 − x ) = 2 × 4π dx (1 − x ) = 8π 1 −
−1 0 3
16π
=
3
4.2.6.14. ∗. Solution. Set
V = (x, y, z) x2 + y 2 ≤ 2, 0 ≤ z ≤ 2x + 3
because 12x, 6x3 and 6xy 2 are all odd under x → −x. To evaluate the
final remaining integral, let’s switch to polar coordinates.
√
ZZ Z 2 Z 2π
2 2
dθ 13 r cos θ)2 + 9 r sin θ)2
13x + 9y dxdy = dr r
x2 +y 2 ≤2 0 0
√
Z 2 Z 2π
dr r3 dθ 13 cos2 θ + 9 sin2 θ
=
0 0
Since
Z 2π Z 2π 2π
2 cos(2θ) + 1 sin(2θ) θ
cos θ dθ = dθ = + =π
0 0 2 4 2 0
2π 2π 2π
1 − cos(2θ)
Z Z
θ sin(2θ)
sin2 θ dθ = dθ = − =π
0 0 2 2 4 0
we finally have
√
( 2)4
ZZ
F · n̂ dS = 18π + π 13 + 9 = (18 + 22)π = 40π
S 4
R 2π R 2π
For an efficient, sneaky, way to evaluate 0
cos2 θ dθ and 0
sin2 θ dθ, see
Example 2.4.4.
4.2.6.15. ∗. Solution 1 (By divergence theorem). Set F = (x + y, x +
z, y + z). Then ∇ · F = 2. That’s really simple. So let’s try using the
divergence theorem.
APPENDIX D. SOLUTIONS TO EXERCISES 575
2 2
RR S = (x, y, z) x + z = 4, 0 ≤ y ≤ 3 . We are to compute
• Set
S
F · n̂ dS, with n̂ denoting the outward normal
RR to S. S is not the
boundary of a solid, so we cannot compute S F · n̂ dS by applying
the divergence theorem directly. The figure on the left below shows
the part of S that is in the first octant.
z y=3 z y=3
S S
−̂ Dl ̂
2
x +z =42 2 2 V
y x +z =4 y
Dr
x x
• On the other hand S, is “almost” the boundary of
V = (x, y, z) x2 + z 2 ≤ 4, 0 ≤ y ≤ 3
Dl = (x, y, z) x2 + z 2 ≤ 4, y = 0
Dr = (x, y, z) x2 + z 2 ≤ 4, y = 3
The outward normal to Dr is ̂ and the outward normal to Dl is −̂, to the
divergence theorem gives
ZZZ ZZ
∇ · F dV = F · n̂ dS
V
Z Z∂V ZZ ZZ
= F · n̂ dS + F · ̂ dS + F · (−̂) dS
S Dr Dl
Then
∂r
∂θ = − 2 sin θ , 0 , 2 cos θ
∂r
∂y = 0, 1, 0
∂r ∂r
n̂dS = ± ∂θ × ∂y dθdy = ± − 2 cos θ , 0 , −2 sin θ dθdy
and, since
F r(θ, y) = 2 cos θ + y , 2 cos θ + 2 sin θ , y + 2 sin θ
4.2.6.16. ∗. Solution. The question highlights that the vector field has
divergence 0. That strongly suggests that we use the divergence theorem.
Set
n̂
V S
V = (x, y, z) 0 ≤ z ≤ 1 − x2 +
2
y2
n̂ D
Then the boundary, ∂V , of V consists of two parts, namely S (with
normal pointing upwards) and the disk
D = (x, y, 0) x2 + y 2 ≤ 1
p
z= x2 + y 2
p
Note that the the paraboloid z = 2−x2 −y 2 and the cone z = x2 + y 2
intersect along the circle x2 + y 2 = 1, z = 1. Probably the simplest solid
whose boundary includes S is
V = (x, y, z) 1 ≤ z ≤ 2 − x2 − y 2 , x2 + y 2 ≤ 1
z
z=4
V =
(x, y, z) x2 + y 2 ≤ z ≤ 4
z = x2 + y 2
x
So the divergence theorem gives
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (x2 + y 2 ) dV
S V V
To compute the triple integral, we’ll use the cylindrical coordinates (r, θ, z).
The z-coordinate runs from 0 to 4. For each fixed 0 ≤ z ≤ 4 (see the blue
disk in the figure below — which shows the part of V in the first octant),
z
z=4
z = x2 + y 2 = r 2
x
2 2
(x, y) runs over 0 ≤
√ x + y ≤ z, which in cylindrical coordinates is
2
0 ≤ r ≤ z or 0 ≤ r ≤ z. So the flux and the triple integral are
ZZ ZZZ
F · n̂ dS = (x2 + y 2 ) dV
S V
√
Z 4 Z z Z 2π
= dz dr r dθ r2
0 0 0
√
Z 4 Z z
= 2π dz dr r3
0 0
4 2
43
Z
z
= 2π dz = 2π
0 4 3×4
32
= π
3
4.2.6.19. ∗. Solution. If we were to evaluate this integral directly using,
for example, spherical coordinates, our integrand would contain
sin(x) = sin 2 sin ϕ cos θ
That’s not very friendly looking. So let’s consider using the divergence
theorem instead. To start,
∂ y ∂ ∂ 2
∇·F= e + xz + zy + sin(x) + z − 1 = 4z
∂x ∂y ∂z
APPENDIX D. SOLUTIONS TO EXERCISES 579
z=1
Dt
V
S z=0 y S y
Db
x x
The surface S is not closed, and so is not the boundary of a solid, so
we cannot apply the divergence theorem directly. But we can easily come
up with a solid whose boundary contains S. Let
V = (x, y, z) x2 + y 2 + z 2 ≤ 4, 0 ≤ z ≤ 1
Db = (x, y, z) x2 + y 2 ≤ 4, z = 0
(x, y, z) x2 + y 2 ≤ 3, z = 1
Dt =
is a lot simpler than F itself. So let’s use the divergence theorem (Theorem
APPENDIX D. SOLUTIONS TO EXERCISES 580
4.2.2).
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (2xz + z) dV
S B B
n̂
V S
n̂ D
Then the boundary, ∂V , of V consists of two parts, namely S (with
normal pointing upwards) and the disk
D = (x, y, 0) x2 + y 2 ≤ 1
So
√
ZZ Z 4 Z z Z 2π
− r2 + 4
F · n̂ dS = dz dr r dθ
∂E 0 0 0
√
Z 4 Z z
dr 4r − r3
= 2π dz
0 0
4
z2
Z
= 2π dz 2z −
0 4
3 i4
h z h 16 i 64
= 2π z 2 − = 2π 16 − = π
12 0 3 3
(b) The boundary of S consists of two parts — S, but with downward
pointing normal, on the bottom and the disk
D = (x, y, z) z = 4, x2 + y 2 ≤ 4
z k̂
z=4
D
S z = x2 + y 2
n̂
y
x
So, by part (a),
ZZ ZZ ZZ
64
π= F · n̂ dS = − F · n̂ dS + F · k̂ dS
3 ∂E S D
APPENDIX D. SOLUTIONS TO EXERCISES 582
ZZ F·k̂
Z Z z}|{
=− F · n̂ dS + 4z dS
S D
−2x2 + y 2 + z 2
= 5/2
[x2 + y 2 + z 2 ]
∂ y 1 3 y(2y)
3/2
= 3/2
−
∂y [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]5/2
x2 − 2y 2 + z 2
= 5/2
[x2 + y 2 + z 2 ]
∂ z 1 3 z(2z)
3/2
= 3/2
−
∂z [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]5/2
x2 + y 2 − 2z 2
= 5/2
[x2 + y 2 + z 2 ]
the specified divergence is
(−2x2 + y 2 + z 2 ) + (x2 − 2y 2 + z 2 ) + (x2 + y 2 − 2z 2 )
∇·F= 5/2
=0
[x2 + y 2 + z 2 ]
if (x, y, z) 6= 0 and is not defined if (x, y, z) = 0.
(b), (c) Set
V1 = (x, y, z) x2 + (y − 2)2 + z 2 ≤ 9
V2 = (x, y, z) x2 + (y − 2)2 + z 2 ≤ 1
S1
S2
(0, 2, 0)
y y
V2
V1
Both V1 and V2 are spherical balls centred on (0, 2, 0). The difference
between them is that V1 has radius 3 while V2 has radius 1. In particular
(0, 0, 0) is not in V2 . So ∇ · F is well-defined and zero throughout V2 and,
by the divergence theorem (Theorem 4.2.2),
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 0
S2 V2
APPENDIX D. SOLUTIONS TO EXERCISES 583
Sρ = (x, y, z) x2 + y 2 + z 2 = ρ2
1
is completely contained inside V1 (for example, ρ = 2 is fine)
V3 = (x, y) x2 + (y − 2)2 + z 2 ≤ 9, x2 + y 2 + z 2 ≥ ρ2
S1
Sρ (0, 2, 0)
y
V3
So
ZZ ZZ r ZZ
r r
F · n̂ dS = F· dS = 3
· dS
S1 Sρ |r| Sρ |r| |r|
ZZ ZZ
1 1
= dS = dS
Sρ |r|2 Sρ ρ 2
1
= 2 4πρ2 = 4π
ρ
2
since Sρ is a sphere of radius
RR ρ and hence ofRRsurface area 4πρ .
(d) The flux integrals S1 F · n̂ dS and S2 F · n̂ dS are different, be-
cause the one point, (0, 0, 0), where ∇ · F fails to be well-defined and zero,
is contained inside S1 but is not contained inside S2 .
4.2.6.24. ∗. Solution. The vector field F looks pretty complicated. But
APPENDIX D. SOLUTIONS TO EXERCISES 584
its divergence
∇·F=2+3+1=6
is very simple. So let’s use the divergence theorem (Theorem 4.2.9). It
says
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = 6 dV = 6 Volume(E)
S E E
y=2
E
y
That cross-section has area 2 × 2+4
2 = 6. Consequently the volume of
E is 2 × 6 = 12 and
ZZ
F · n̂ dS = 6 × 12 = 72
S
V = (x, y, z) x2 + y 2 + z 2 ≤ 4, z ≥ 1
Thus
ZZ Z Z −F·k̂
z}|{
F · n̂ dS = −3 Volume(V ) + −3 dS
S D
= −3 Volume(V ) − 3 Area(D)
= −3 Volume(V ) − 9π
√
since D is a circular disk of radius 3. To compute the volume of V , we
slice V into thin horizontal pancakes each of thinkness dz. The pancake at
height z has cross-section the circular disk x2 + y 2 ≤ 4 − z 2 . As this disk
has area π(4 − z 2 ), the pancake has volume π(4 − z 2 ) dz. All together
2 2
z3
Z
2 7 5π
Volume(V ) = dz π(4 − z ) = π 4z − =π 4− =
1 3 1 3 3
and ZZ
5π
F · n̂ dS = −3 − 9π = −14π
S 3
4.2.6.26. ∗. Solution. Let’s try the divergence theorem. Set
(x, y, z) x2 + y 2 + z 2 ≤ 3
V =
∂ ∂ 2 ∂ 2
x + y2
∇·F= − 2xy + y + sin(xz) +
∂x ∂y ∂z
= −2y + 2y + 0 = 0
APPENDIX D. SOLUTIONS TO EXERCISES 586
= 0 dV + (x2 + y 2 )dxdy
V D
Z 5 Z 2π
=0+ dr r dθ r2
0 0
54 625
= 2π = π
4 2
4.2.6.28. ∗. Solution. The boundary of the solid V enclosed by S and
z = ±1 consists of three pieces: S, the top disk
S1 = (x, y, z) x2 + y 2 ≤ 2, z = 1
On S1 , n̂ = k̂ and
F · n̂ = F · k̂ = xy − z − z 2 = xy − 2
z=1
so that, denoting D = (x, y) x2 + y 2 ≤ 2 ,
ZZ ZZ
F · n̂ dS = (xy − 2) dxdy = −2 Area(D) = −4π
S1 D
RR
Here we have used that the integral D
xy dxdy = 0 because xy is odd
under x → −x. On S2 , n̂ = −k̂ and
F · n̂ = −F · k̂ = −(xy − z − z 2 ) = −xy
z=−1
so that ZZ ZZ
F · n̂ dS = (−xy) dxdy = 0
S2 D
By the divergence theorem (Theorem 4.2.2),
ZZ ZZ ZZ ZZ
F · n̂dS = ∇ · F dV − F · n̂ dS − F · n̂ dS
S V S1 S2
= 0 − (−4π) − 0 = 4π
since
∂ ∂ ∂
(x + eyz ) + (xy − z − z 2 )
∇·F= 2yz + sin(xz) +
∂x ∂y ∂z
= 1 + 2z − 1 − 2z
=0
APPENDIX D. SOLUTIONS TO EXERCISES 587
we have
ZZ
∇ × F · n̂ dS
S
ZZ
2xı̂ı + 2y ̂ + 4z(x, y) k̂
= − 3z(x, y)2 ı̂ı − 2z(x, y) ̂ + k̂ · dxdy
x2 +y 2 ≤1 4z(x, y)
ZZ 3
= − x z(x, y) − y + 1 dxdy
x2 +y 2 ≤1 2
q
Since y is an odd function of y and x z(x, y) = x 12 (1 − x2 − y 2 ) is an odd
function of x, they both integrate to zero. Hence
ZZ ZZ
∇ × F · n̂ dS = 1 dxdy = π
S x2 +y 2 ≤1
Hence
ZZ ZZ ZZ
∇ × F · n̂ dS = ∇ × F · k̂ dS = dS = π
S D D
so that
ZZ ZZ √ 2
F · n̂ dS = 4 dxdy = 4 Area(S 0 ) = 4 × π 3 = 12π
S S0
p
4.2.6.31. ∗. Solution. Call the hemisphere 0 ≤ z ≤ 4 − x2 − y 2 , H.
Call the bottom surface of the hemisphere D and the top surface S. The
disk D has radius 2, area 4π, z = 0 and the outward normal −k̂, so that
ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dxdy = − dxdy = −4π
D D D
As
∂ ∂ 2 ∂
∇·F= (xy 2 ) + (x y) + (1) = x2 + y 2
∂x ∂y ∂z
the divergence theorem (Theorem 4.2.2) gives
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS
S
Z Z ZH D
= (x2 + y 2 ) dV − (−4π)
R
The integral over y was zero because y is odd under y → −y. On the
bottom surface, z = 0 and the outward normal to D is −k̂, so that
ZZ ZZ
F · n̂ dS = − (3 × 0 − 0 × ye0 ) dxdy = 0
Sb x2 +y 2 ≤1
Again, the integral over y was zero because y is odd under y → −y. As
∂ ∂ 2 z ∂
∇·F= (x + xyez ) + 12 y ze + (3z − yzez )
∂x ∂y ∂z
APPENDIX D. SOLUTIONS TO EXERCISES 589
x2 + y 2 + z 2 = a2
x
Then ∂V consists of
• the x = 0 face (x, y, z) y 2 + z 2 ≤ a2 , x = 0, y ≥ 0, z ≥ 0 with
normal n̂ = −ı̂ı,
• the y = 0 face (x, y, z) x2 + z 2 ≤ a2 , x ≥ 0, y = 0, z ≥ 0 with
normal n̂ = −̂,
• the z = 0 face (x, y, z) x2 + y 2 ≤ a2 , x ≥ 0, y ≥ 0, z = 0 with
normal n̂ = −k̂,
• and the first octant part of the sphere. Call it S.
Then
ZZZ ZZZ ZZZ
1 4 3
∇ · F dV = z + 1 + z − 2z dV = dV = πa
V V V 8 3
1
= πa3
6
ZZ ZZ Z a Z π/2
F · (−k̂) dx dy = (2x + 02 ) dx dy = 2 dr r dθ r cos θ
z=0 z=0
face face 0 0
a
2a3
Z
=2 r2 dr =
3
ZZ Z0Z
F · (−̂) dx dz = − (0 + 0z) dx dz = 0
y=0 y=0
face face
ZZ ZZ Z a Z π/2
F · (−ı̂ı) dy dz = −(y + 0z) dy dz = − dr r dθ r sin θ
x=0 x=0
face face 0 0
a
a3
Z
=− r2 dr = −
0 3
By the divergence theorem
ZZ ZZZ ZZ
F · n̂ dx dy = ∇ · F dV − F · (−ı̂ı) dy dz
x=0
S V face
APPENDIX D. SOLUTIONS TO EXERCISES 590
ZZ ZZ
− F · (−̂) dx dz − F · (−k̂) dx dy
y=0 z=0
face face
π 1 3
= − a
6 3
4.2.6.34. Solution. (a) On the cylindrical surface √ S1 , use (surprise!)
cylindrical coordinates. Since the cylinder has radius 2, we may parametrize
it by
√ √
r(θ, z) = 2 cos θ ı̂ı + 2 sin θ ̂ + z k̂
∂r √ √
(θ, z) = − 2 sin θ ı̂ı + 2 cos θ ̂
∂θ
∂r
(θ, z) = k̂
∂z
∂r ∂r
n̂ dS = ± (θ, z) × (θ, z) dθ dz
∂θ ∂z
√ ı̂ı √ ̂ k̂
= ± det − 2 sin θ ı̂ı 2 cos θ 0 dθ dz
0 0 1
√ √
= ± 2 cos θ ı̂ı + 2 sin θ ̂ dθ dz
z 2 = 4 − x2 − y 2 = 4 − 2 = 2
√ √
so z runs from − 2 to 2 (see the figure below) and
√
ZZ Z 2 Z 2π √
F · n̂ dS = −2 √ dz dθ = −8 2π
S1 − 2 0
z
√ √
(0, 2, 2)
√
(0, 4 − z 2 , z)
x2 +y 2 =2
x x2 + y 2 + z 2 = 4
√ 2 √ 2
So the cross-section has area π 4 − z2 −π 2 = π 2 − z 2 and
√
ZZZ ZZZ Z 2
π 2 − z 2 dz
∇ · F dV = 3 dV = 3 √
V V − 2
√
Z 2 √ 23/2
2 − z 2 dz = 6π 2 2 −
= 6π
3
√ 0
= 8 2π
So by Gauss’ law
ZZZ ZZZ ZZZ
∇ · E dV = 4π ρ dV =⇒ ∇ · E − 4πρ dV = 0
V V V
This is true for all solids V for which the divergence theorem applies. If
there were some point in R3 for which ∇ · E − 4πρ were, say, strictly
bigger than zero, then, by continuity, we could find a ball B centered on
that
RRR point
with ∇ · E − 4πρ > 0 everywhere B . This would force
RRR on
B
∇ · E − 4πρ dV > 0, which violates V
∇ · E − 4πρ dV = 0 with
V set equal to B . Hence ∇ · E − 4πρ must be zero everywhere.
4.2.6.36. Solution. By the divergence theorem
ZZ ZZZ ZZZ
r · n̂ dS = ∇ · r dV = ∇ · (xı̂ı + y ̂ + z k̂) dV
∂V
Z Z ZV V
= 3 dV = 3 Volume(V )
V
Our geometric explanation starts with the observation that the volume of
the cone with vertex (0, 0, 0) and base a tiny piece of surface dS is 31 times
APPENDIX D. SOLUTIONS TO EXERCISES 592
the area of the base times the height of the cone. The height of the cone
is |n̂ · r|, where r is a point in dS. So the volume of the cone is 13 |n̂ · r| dS.
n̂
dS
(0, 0, 0)
First assume that (0, 0, 0) is in V and V is convex. Then
To get an outward pointing normal we need the + sign. For example, with
the + sign, the z-component is 9 sin ϕ cos ϕ = 29 sin(2ϕ) so that the normal
is pointing upward when 0 < ϕ < π2 , i.e. in the northern hemisphere, and
is pointing downward when π2 < ϕ < π, i.e. in the southern hemisphere.
(As a further consistency check, note that n̂(θ, ϕ) is parallel to r(θ, ϕ).) So
ZZ Z 2π Z π
F · n̂ dS = 9 dθ dϕ sin ϕ (0, 0, 3 sin ϕ cos θ + 3 cos ϕ)
S 0 0
APPENDIX D. SOLUTIONS TO EXERCISES 593
· sin ϕ cos θ , sin ϕ sin θ , cos ϕ
Z 2π Z π
dϕ sin2 ϕ cos ϕ cos θ + sin ϕ cos2 ϕ
= 27 dθ
0 0
Z π Z 2π
= 54π dϕ sin ϕ cos2 ϕ since cos θ dθ = 0
0 0
π
= −18π cos3 ϕ 0
= 36π
(b) Set
(x, y, z) ∈ R3 x2 + y 2 + z 2 ≤ 9
V =
Since
∂
x+z =1 ∇·F=
∂z
the divergence theorem (Theorem 4.2.2) gives
ZZ ZZZ ZZZ
4
F · n̂ dS = ∇ · F dV = dV = π33 = 36π
S V V 3
4.2.6.38. ∗. Solution. Denote by V the cube specified in the problem.
Then ∂V consists of S together with the face F in the plane z = 0, oriented
with the normal being −k̂.
z
n̂ = k̂
(0, 1, 1)
y
F
(1, 1, 0)
x
As
∂ ∂ z ∂ 2
y cos(y 2 ) + z − 1 + xyez
∇·F= +1 +
∂x ∂y x+1 ∂z
∂ 2
= xyez
∂z
the divergence theorem (Theorem 4.2.2) gives
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
S V F
Z 1 Z 1 Z 1 Z Z 1
1
∂ 2
2
= dx xyez + dy dx dy xyez
dz
0 0 0 ∂z 0 0 z=0
Z 1 Z 1 z=1
Z 1 Z 1
2 2
= dx dy xyez + dx dy xyez
0 0 z=0 0 0 z=0
Z 1 Z 1
2
= dx dy xyez
0 0 z=1
2 1 2 1
x y
=e
2 0 2 0
e
=
4
APPENDIX D. SOLUTIONS TO EXERCISES 594
∇ G(x, y, z)
n̂ dS = ± dxdy = ±(−y, −x, 1)dxdy
∇ G(x, y, z) · k̂
The “+” sign gives the upward normal, so the specified upward flux is
ZZ ZZ
F · n̂ dS = (y, x, 3) · (−y − x, 1) dxdy
S x2 +y 2 ≤9
ZZ
= (3 − x2 − y 2 ) dxdy
x2 +y 2 ≤9
x = 3 cos θ
y = 3 sin θ
z=z
1 2π
Z
= −9 × 18 × dθ sin2 (2θ)
4 0
81 2π 1 − cos(4θ)
Z
=− dθ
2 0 2
2π
81 θ sin(4θ)
=− −
2 2 8 0
81
=− π
2
R 2π
For an efficient, sneaky, way to evaluate 0 dθ sin2 (2θ) see Example 2.4.4.
(c) using the divergence theorem: Note that if x2 + y 2 ≤ 9, then |x| ≤ 3
and y ≤ 3 so that |xy| ≤ 9 < 10. Set
S̃ = (x, y, z) x2 + y 2 = 9, xy ≤ z ≤ 10
V = (x, y, z) x2 + y 2 ≤ 9, xy ≤ z ≤ 10
S̃
S
y
x
Note that ∇ · F = 0. So the divergence theorem yields
ZZZ
0= ∇ · F dV
ZZ V
= F · n̂ dS
˜
∂V
ZZ ZZ ZZ
= F · n̂ dS − F · n̂ dS + F · k̂ dS
S̃ S ST
This implies
ZZ ZZ ZZ
F · n̂ dS = F · n̂ dS − F · k̂ dS
S̃ S ST
ZZ
= − 27π
2 − 3 dS
x2 +y 2 ≤9
= − 27π
2 − 3π32 = − 81π 2
APPENDIX D. SOLUTIONS TO EXERCISES 596
(b) Set
p
(x, y, z) x2 + y 2 ≤ 25, 0 ≤ z ≤ 25 − x2 − y 2
V =
ST = (x, y, z) x2 + y 2 + z 2 = 25, z ≥ 0
SB = (x, y, z) x2 + y 2 ≤ 25, z = 0
One surface which obeys this condition is the unit cube (with outward
normal) centred on 0, 0, − 11
2 .
Since
∂r
∂θ = − 2z sin θ , 2z cos θ , 0
∂r
∂z = 2 cos θ , 2 sin θ , 1
∂r ∂r
∂θ × ∂z = 2z cos θ , 2z sin θ , −4z
(3.3.1) yields that the element of surface area for this parametrization is
∂r ∂r
dS = ∂θ × ∂z dθdz = 2z (cos θ , sin θ , −2 dθdz
√
= 2 5z dθdz
• The first uses Stokes’RR theorem. But the flux integral in Stokes’ theo-
rem is of the form S ∇ × A · n̂ dS. So to be able to apply Stokes’
theorem in the current problem, F has to be of the form ∇ ×A. That
is, F has to have a vector potential. We know that in order for F to
have a vector potential, it must pass the screening test ∇ ·F = 0. Our
F = z k̂ fails this screening test. So we can’t use Stokes’ theorem.
• The second uses the divergence theorem. But the flux integral in
the divergence theorem is over the boundary of a solid. That is not
the case for our S. So in order to apply the divergence theorem
in the current problem, we have to enlarge S to the boundary of a
solid. There are many ways to do this. But they all appear fairly
complicated. So it does not seem wise to use the divergence theorem.
APPENDIX D. SOLUTIONS TO EXERCISES 598
(5, 0, 4)
(0, 2, 0)
y
x
(5, 2, 0)
and that plane is invariant under translations parallel to the x axis. As
the plane does not pass through the origin, the equation of the plane has
to be of the form by + cz = 1. For (0, 0, 4) to be on the plane, we need
c = 14 . For (0, 2, 0) to be on the plane, we need b = 12 . So S is contained
in the plane G(x, y, z) = y2 + z4 = 1 and equation (3.3.3) gives that
∇ G(x, y, z) (0, 21 , 41 )
n̂ dS = ± dxdy = ± 1 dxdy = ±(0, 2, 1) dxdy
∇ G(x, y, z) · k̂ 4
This integral would be easy enough to evaluate directly, but we don’t need
to. The average value of z (i.e. the z-coordinate of the centre of mass with
constant density) is 23 , by symmetry. Since V has volume 6, that average
value of z is also
ZZZ
1 3
z̄ = z dV =
6 V 2
RRR
So V
z dV = 9
4.2.6.42. ∗. Solution.
ZZ (a) For theZ Zsurface
Z z = f (x, y) = 1 − x2 − y 2 ,
F · n̂ dS = 2 z dV = 18
S V
APPENDIX D. SOLUTIONS TO EXERCISES 599
Now ZZ
2ax(y 2 + z 2 )
z=1−x2 −y 2
dxdy = 0
x2 +y 2 ≤1
because the integrand is odd under x → −x and
ZZ
2cy(x2 + z 2 ) z=1−x2 −y2 dxdy = 0
x2 +y 2 ≤1
Now
Z 2π Z 2π 2π
cos(2θ) + 1 sin(2θ) θ
cos2 θ dθ = dθ = + =π
0 0 2 4 2 0
Z 2π Z 2π 2π
2 1 − cos(2θ) θ sin(2θ)
sin θ dθ = dθ = − =π
0 0 2 2 4 0
R 2π R 2π
For an efficient, sneaky, way to evaluate 0 cos2 θ dθ and 0 sin2 θ dθ, see
Example 2.4.4. So, we finally have
ZZ Z 1
dr 2πbr3 (1 − r2 ) + 2πdr3 (1 − r2 ) + πr3
F · n̂ dS =
σ1 0
h1 1i h1 1i 1 π π(b + d)
= 2πb − + 2πd − +π = +
4 6 4 6 4 4 6
(b), (c) Here is a side view of σ1 , σ2 and σ3 .
σ1 , z = 1 − x2 − y 2
σ3
σ2 , z = x2 + y 2 − 1
APPENDIX D. SOLUTIONS TO EXERCISES 600
Set
(x, y, z) 0 ≤ z ≤ 1 − x2 − y 2 , x2 + y 2 ≤ 1
Vb =
Vc = (x, y, z) x2 + y 2 − 1 ≤ z ≤ 1 − x2 − y 2 , x2 + y 2 ≤ 1
∇ · H(x, y, z)
∂ x−a
=
∂x [(x − a)2 + (y − b)2 + (z − c)2 ]3/2
∂ y−b
+
∂y [(x − a) + (y − b)2 + (z − c)2 ]3/2
2
∂ z−c
+
∂z [(x − a)2 + (y − b)2 + (z − c)2 ]3/2
(x − a)2 + (y − b)2 + (z − c)2 − (x − a) 23 (2(x − a))
= 5/2
(x − a)2 + (y − b)2 + (z − c)2
(x − a)2 + (y − b)2 + (z − c)2 − (y − b) 23 (2y)
+ 5/2
(x − a)2 + (y − b)2 + (z − c)2
(x − a)2 + (y − b)2 + z 2 − (z − c) 23 (2(z − c))
+ 5/2
(x − a)2 + (y − b)2 + (z − c)2
3 (x−a)2 + (y−b)2 + (z−c)2 − 3(x−a)2 − 3(y−b)2 − 3(z−c)2
= 5/2
(x − a)2 + (y − b)2 + (z − c)2
=0
(x, y, z) in
(x, y, z) x2 + 2y 2 + 3z 2 ≤ 16
V =
So, by the divergence theorem,
ZZ ZZZ
G · n̂ dS = ∇ · G dV = 0
S V
Sε S
Vε
(2,1,1)
Set
So ZZ ZZ
F · n̂ dS = − F · n̂ dS
S Sε
To get the second line we used the vector identity Theorem 4.1.4.d. To get
the third line, we used that a is a constant, so that all of its derivatives are
zero. For all vectors a · (b × c) = (a × b) · c (in case you don’t remember
this, it was Lemma 4.1.8.a) so that
(a × F) · n̂ = a · (F × n̂)
and
ZZ ZZZ
a· F × n dS = −a · ∇ × F dV
Z∂Ω
Z ZZZ Ω
=⇒ a · F × n dS + ∇ × F dV = 0
∂Ω Ω
RR choosing a = ı̂ıRRR
In particular, , ̂ and k̂, we see that all three components of
the vector ∂Ω F × n dS + Ω
∇ × F dV are zero. So
ZZZ ZZ ZZ
∇ × F dV = − F × n dS = n̂ × F dS
Ω ∂Ω ∂Ω
4 3
∇ · F(0, 0, 0)Volume(Va ) + O(a4 ) = πa ∇ · F(0, 0, 0) + O(a4 )
3
Here O(a4 ) is an error term that is bounded by a constant times a4 . This
is consistent with the above equation if and only if ∇ · F(0, 0, 0) = 43 .
APPENDIX D. SOLUTIONS TO EXERCISES 603
It is the interior of the sphere of radius 2a centred on (0, 0, a). The surface
of V (with outward normal) is the union of S (with normal pointing away
from the origin) and the disk
B = (x, y, 0) x2 + y 2 ≤ 3a2
To evaluate the integral over V , we note that z runs from 0 to 3a and that
the cross section of
(b) Similarly, “the total flux of G = u∇u out through the boundary of
APPENDIX D. SOLUTIONS TO EXERCISES 604
R” equals ZZ ZZZ
J= G · n̂ dS = ∇ · G dV
S R
Here G = uF (using the notation from part (a)), so by the vector identity
of Theorem 4.1.4.c,
∇ · G = |∇u|2 + 0
In conclusion,
ZZZ Z Z Z h 2 2 2 i
∂u ∂u ∂u
J= ∇ · G dV = + + dV
R R ∂x ∂y ∂z
4.2.6.49. ∗. Solution. (a) This is a classic case for the divergence
theorem. The flux we want equals
ZZ ZZZ ZZZ ZZZ
I= F · n̂ dS = ∇ · F dV = (2x + 2 − 2) dV = 2 x dV
S R R R
The solid R clearly has reflection symmetry across the plane x = 0. So the
x-coordinate of the centre of mass of R, i.e. the average value of x over R,
i.e. RRR RRR
R
x dV R
x dV
x̄ = RRR =
R
dV Vol(R)
is zero. Hence
I = 2x̄Vol(R) = 0
Alternatively, here is a direct evaluation of 2
RRR
R
x dV . The base region
x2 + (y − 1)2 ≤ 1 is the circular disk of radius 1 centred on (0, 1). In polar
coordinates it is
Because the disk is contained in the upper half plane, the polar angle θ
is restricted to 0 ≤ θ ≤ π. So, in cylindrical coordinates, the solid R is
described by
0 ≤ θ ≤ π, 0 ≤ r ≤ 2 sin θ, 0 ≤ z ≤ r2 sin2 θ
Hence
Z π Z 2 sin θ Z r 2 sin2 θ
I=2 (r cos θ) dz r dr dθ
θ=0 r=0 z=0
Z π Z 2 sin θ
=2 r4 sin2 θ cos θ dr dθ
θ=0 r=0
Z π h 25 sin5 θ i
=2 sin2 θ cos θ dθ
θ=0 5
Z π
64
= sin7 θ cos θ dθ
5 θ=0
64 h sin8 θ iπ
=
5 8 θ=0
=0
APPENDIX D. SOLUTIONS TO EXERCISES 605
so
ZZ
F · n̂dS
Sside
Z π Z 4 sin4 θ
2 sin3 θ cos θ − 2 sin5 θ cos θ − sin2 θ + 2 sin4 θ dt dθ
=8
θ=0 t=0
Z π
2 sin7 θ cos θ − 2 sin9 θ cos θ − sin6 θ + 2 sin8 θ dθ
= 32
θ=0
h sin8 θ π Z π Z π
sin10 θ i 6
= 32 2 −2 − 32 sin θ dθ + 64 sin8 θ dθ
8 10 0 θ=0 θ=0
531 7531
= −32 π + 64 π (by the Hint as above)
642 8642
15
= π.
2
(b) Offset polar alternative: We can also parametrize S using cylindrical
coordinates translated so that the centre of the base of the cylinder, namely
(0, 1, 0), plays the role of the origin. Then, looking at the figure
z
(x, y, z)
(0, 1, 0) z y
θ r
x (x, y, 0)
we see that
By (3.3.1),
ı̂ı ̂ k̂
∂r ∂r
× = det 0 0 1 = − cos θ, − sin θ, 0 ,
∂t ∂θ
− sin θ cos θ 0
∂r ∂r
n̂ dS = − × dt dθ = (cos θ, sin θ, 0)dt dθ
∂t ∂θ
where we have chosen the sign to give the outward pointing normal. So
ZZ Z 2π Z (1+sin θ)2
cos3 θ + 2(1 + sin θ) sin θ dt dθ
F · n̂dS =
S θ=0 t=0
Z 2π
(1 + sin θ)2 cos3 θ + 2(1 + sin θ)3 sin θ dθ
=
0
Z 2π
2 sin θ cos3 θ + 6 sin2 θ + 2 sin4 θ dθ
=
0
APPENDIX D. SOLUTIONS TO EXERCISES 607
2π Z π Z π
1
= − cos4 θ + 12 sin2 θ dθ + 4 sin4 θ dθ
2 0 0 0
π 3π 15
= 0 + 12 + 4 = π
2 42 2
To get the third line, we used that the integral over 0 ≤ θ ≤ 2π of any odd
power of sin θ or cos θ is zero.
4.2.6.50. ∗. Solution. The circle x2 + y 2 = 4y, or equivalently, x2 +
(y − 2)2 = 4, has radius 2 and centre (0, 2). On the bottom surface, z = 0
and the outward normal is −k̂, so that
ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dxdy = − (2x + 3y) dxdy
D D D
By symmetry, the centre of mass, (x̄, ȳ), of the circle is (0, 2). Here x̄ and
ȳ are the average values
RR RR
D
x dxdy y dxdy
x̄ = RR ȳ = RRD
D
dxdy D
dxdy
and
ZZ
F · n̂ dS = −4π(2x̄ + 3ȳ) = −4π(2 × 0 + 3 × 2) = −24π
D
As
∂ ∂ ∂
∇·F= (x + x2 y) + (y − xy 2 ) + (z + 2x + 3y)
∂x ∂y ∂z
= (1 + 2xy) + (1 − 2xy) + (1)
=3
So, using −C3 to denote the line segment from (0, 1) to (1, 1)
ZZ Z Z
∂f
(x, y) dx dy = − f (x, y) dx + f (x, y) dx
R ∂y
Z C1 Z C3
= f (x, y) dx − f (x, y) dx
−C3 C1
Z1 Z 1
= f (x, 1) dx − f (x, 0) dx
0 0
• Use the right hand rule to check that r0 (s) × k̂ is n̂ rather than −n̂.
APPENDIX D. SOLUTIONS TO EXERCISES 609
n̂(s)
r′ (s)
R
r′ (s)
C
n̂(s)
So, by Green’s theorem,
I I I
dy dx
F · n̂ ds = F1 − F2 ds = [−F2 dx + F1 dy]
C C ds ds C
ZZ
∂ ∂
= F1 − (−F2 ) dx dy
∂x ∂y
Z ZR
= ∇ · F dx dy
R
(b) The boundary of the square has four sides — one with y = −1, one
with x = 1, one with y = 1 and one with x = −1.
dy=0
(−1, 1) y=1 (1, 1)
dx=0 dx=0
x=−1 x=1
x2 + y 2 = 2
x2 +y 2 =1
y=0 y=0 x
dy=0 dy=0
The two flat pieces each give zero, since on them y = 0 and dy = 0. So
x dy − y dx
I
1 1 1
2 2
= +0− +0=0
2π C x + y 2 2
4.3.1.4. Solution. The two partial derivatives
∂ x (x2 + y 2 ) − x(2x) y 2 − x2
= 2 = 2
∂x x2 + y 2 (x2 + y 2 ) (x2 + y 2 )
∂ −y −(x2 + y 2 ) − (−y)(2y) y 2 − x2
2 2
= 2 = 2
∂y x + y (x2 + y 2 ) (x2 + y 2 )
are well-defined and equal everywhere except at the origin (0, 0).
Short discussion: Were it not for the singularity at (0, 0), theR vector
field of the last problem would be conservative and the integral F · dr
around any closed curve would be zero. But as we saw in parts (a) and (b)
of Q[4.3.1.3], this is not the case. On the other hand, by Green’s theorem
(Theorem 4.3.2), the integral around the boundary of any region that does
not contain (0, 0) is zero, as happened in part (c) of Q[4.3.1.3].
Long discussion: First consider part (c) of Q[4.3.1.3]. The curve C is
the boundary of the region
R = (x, y) 1 ≤ x2 + y 2 ≤ 2, y ≥ 0
∂ x ∂ −y
The partial derivatives ∂x x2 +y 2 and ∂y x2 +y 2 are well-defined and
equal everywhere in R. So by Green’s theorem
ZZ
x dy − y dx ∂ −y
I
1 1 ∂ x
= − dx dy
2π C x2 + y 2 2π R ∂x x + y
2 2 ∂y x2 + y 2
APPENDIX D. SOLUTIONS TO EXERCISES 611
=0
Then the boundary, ∂R, of R consists of two parts. One part is Cb . The
other part is Ca , but oriented clockwise rather than counterclockwise. We’ll
call it −Ca .
y
R Cb
−Ca
x
∂ x ∂ −y
Again the partial derivatives ∂x 2
x +y 2 and 2
∂y x +y 2 are well-defined
and equal everywhere in R. So by Green’s theorem
ZZ
x dy − y dx ∂ −y
I
1 1 ∂ x
= − dx dy
2π ∂R x2 + y 2 2π R ∂x x + y
2 2 ∂y x2 + y 2
=0
Consequently
x dy − y dx x dy − y dx x dy − y dx
I I I
1 1 1
0= 2 2
= 2 2
+
2π ∂R x + y 2π Cb x + y 2π −Ca x2 + y 2
x dy − y dx x dy − y dx
I I
1 1
= 2 2
−
2π Cb x + y 2π Ca x2 + y 2
and we conclude that the answers to parts (a) and (b) should be the
same.We did indeed see that in Q[4.3.1.3].
x=0 x=3
(1, −1)
(0, −2)
APPENDIX D. SOLUTIONS TO EXERCISES 613
By Green’s theorem,
I
(x sin y 2 − y 2 ) dx + (x2 y cos y 2 + 3x) dy
C
ZZ
∂ 2 ∂
= (x y cos y 2 + 3x) − (x sin y 2 − y 2 ) dx dy
∂x ∂y
Z ZT
2xy cos y 2 + 3 − 2xy cos y 2 + 2y dx dy
=
Z ZT
= 3 + 2y dx dy
T
RR
The integral T (2y) dx dy vanishes because 2y changes sign under y → −y
while
RR the domain of integration is invariant under y → −y. The integral
T
3 dx dy is 3 times the area of the trapezoid, which is its width (1) times
the average of its heights ( 21 [2 + 4]) = 3. So
I
(x sin y 2 − y 2 ) dx + (x2 y cos y 2 + 3x) dy = 3 × 1 × 3 = 9
C
R =
R
{(x, y)} 1 ≤ x ≤ 3, 0 ≤ y ≤ 1
(1, 0) C (3, 0) x
So Green’s theorem gives
I
2 2
3y 2 + 2xey dx + 2yx2 ey dy
C
ZZ h
∂ 2 ∂ 2
i
= 2yx2 ey − 3y 2 + 2xey dxdy
R ∂x ∂y
ZZ h i
2 2
= 4xyey − 6y − 4xyey dxdy
R
Z 3 Z 1 Z 3
1
= −6 dx dy y = −6 dx
1 0 1 2
= −6
4.3.1.9. ∗. Solution. (a) The curves y = x2 + 4x + 4 and y = 4 − x2
meet when
x2 + 4x + 4 = 4 − x2 ⇐⇒ 2x2 + 4x = 2x(x + 2) = 0
So the curves intersect at (0, 4) and (−2, 0). Here is a sketch.
y y = x2 + 4x + 4
C y = 4 − x2
R
C
(−2, 0) x
(b) Let
(x, y) ∈ R2 x2 + 4x + 4 ≤ y ≤ 4 − x2 , −2 ≤ x ≤ 0
R=
By Green’s theorem (Theorem 4.3.2)
I ZZ n o
y 2 ∂ y 2 ∂
xy dx + (e + x )dy = ∂x (e + x ) − ∂y (xy) dxdy
C R
APPENDIX D. SOLUTIONS TO EXERCISES 615
Z 0 Z 4−x2
= dx dy x
−2 x2 +4x+4
Z 0
= dx (−2x2 − 4x)x
−2
4 0
x 4x3
= − −
2 3 −2
8
=−
3
4.3.1.10. ∗. Solution. The integral that would be used for direct eval-
uation looks very complicated. So let’s try Green’s theorem. The curve C
is the boundary of the triangle
y
(1, 2)
y=2x
T = (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 2x
(0, 0) (1, 0) x
So
Z Z
2 2
y 2 − e−y + sin x dx + 2xye−y + x dy
F · dr =
C
ZCZ
2 2
∂
2xye−y + x − ∂
y 2 − e−y + sin x
= ∂x ∂y dxdy
Z ZT
2 2
2ye−y + 1 − 2y + 2ye−y
= dxdy
T
Z 1 Z 2x
= dx dy 1 − 2y
0 0
Z 1
dx 2x − 4x2
=
0
4 1
=1− =−
3 3
4.3.1.11. ∗. Solution. Here is a sketch of the two curves in question.
y
y=3−x2 +2x
(0, 3)
R
(3, 0)
x
y=x2 −4x+3
R = (x, y) x2 − 4x + 3 ≤ y ≤ 3 − x2 + 2x, 0 ≤ x ≤ 3
T = (x, y) 0 ≤ x ≤ 1, −2x ≤ y ≤
2x (0, 0) x
y=−2x
(1, −2)
It has boundary ∂T = C, oriented counterclockwise as desired. So, by
Green’s theorem,
Z Z
−y
+ sin x dx + 21 x2 + x − xe−y dy
3 2
F · dr = 2y + e
C
Z∂T
Z
−y −y
∂ 1 2 ∂ 3 2
= ∂x 2 x + x − xe − ∂y 2y + e + sin x dxdy
Z ZT
x + 1 − e−y − 3y − e−y dxdy
=
Z ZT
= x − 3y + 1 dxdy
T
Now
ZZ
1
dxdy = Area(T ) = (4)(1) = 2
T 2
ZZ
y dxdy = 0 since y is odd under y → −y
T
ZZ Z 1 Z 2x Z 1
4
x dxdy = dx dy x = 4x2 dx =
T 0 −2x 0 3
APPENDIX D. SOLUTIONS TO EXERCISES 617
So Z
4 10
F · dr = −3×0+2=
C 3 3
4.3.1.13. ∗. Solution. Set
−y x
F= ı̂ı + 2 ̂
x2
+y 2 x + y2
(a) Green’s theorem must be applied to a curve that is closed, so that
it is the boundary of a region in R2 . The given curve C is not closed. But
it is part of the boundary of
2
R = (x, y) − 2 ≤ x ≤ 2, x4 + 1 ≤ y ≤ 2
Here is a sketch of R.
y
L
(−2, 2) (2, 2)
R
x2
C y= 4
+1
x
The boundary of R consists of two parts — C on the bottom and the
line segment L from (2, 2) to (−2, 2) on the top. Note that F is well-defined
on all of R and that
∂ ∂ ∂ x ∂ y
F2 − F1 = 2 2
+
∂x ∂y ∂x x + y ∂y x + y 2
2
R = (x, y) − 2 ≤ x ≤ 2, x2 − 2 ≤ y ≤ 2
APPENDIX D. SOLUTIONS TO EXERCISES 618
y = x2 − 2
C
W = (x, y) − 2 ≤ x ≤ 2, x2 − 2 ≤ y ≤ 2, x2 + y 2 ≥ ρ2
W
Cρ
y = x2 − 2
C
F · dr = − π2 . So it remains
R
We have already found, in part (a), that −L
only to use
to evaluate
Z Z 2π
F r(θ) · r0 (θ) dθ
F · dr =
Cρ 0
F(r(θ))
r0 (θ)
Z 2π z 1 }|
1
{
z }| {
= − sin θ ı̂ı + cos θ ̂ · −ρ sin θ ı̂ı + ρ cos θ ̂ dθ
0 ρ ρ
Z 2π
= dθ
0
= 2π
All together
−y
Z Z Z
x π 3π
2 + y2
dx + 2 2
dy = F · dr + F · dr = − + 2π =
C x x + y −L Cρ 2 2
(c) No, F is not conservative. We found, in parts (a) and (b), two
different values for the integrals along two paths, both of which start at
(−2, 2) and end at (2, 2). So F does not have the “path independence”
property of Theorem 2.4.7.c and cannot be conservative.
R
4.3.1.14. ∗. Solution. The given integral is of the form C F · dr with
√
F = 2xey + 2 + x2 ı̂ı + x2 2 + ey ) ̂
It is sketched below.
y C
(0, 3)
R y = x2 − 4x + 3
(3, 0)
x
C
y = 3 − x2 + 2x
By Green’s theorem (Theorem 4.3.2),
Z √
2xey + 2 + x2 dx + x2 2 + ey )dy
C
APPENDIX D. SOLUTIONS TO EXERCISES 620
ZZ h
∂ 2 ∂ √ i
x 2 + ey ) − 2xey + 2 + x2 dxdy
=
∂x ∂y
Z ZR
2x(2 + ey ) − 2xey dxdy
=
R
Z 3 Z 3−x2 +2x
=4 dx dy x
0 x2 −4x+3
Z 3
dx x (3 − x2 + 2x) − (x2 − 4x + 3)
=4
0
Z 3
1
dx 6x2 − 2x3 = 4 2 × 33 − 34 = 54
=4
0 2
4.3.1.15. ∗. Solution. (a) Denote by
R2 = (x, y) (x − 2)2 + y 2 ≤ 1
y y
C3
C2 −C1
(2,0) (2, 0)
x x
R2
R3
R
(b) We cannot blindly apply Green’s theorem to I3 = C3 F · dr because
(0, 0) is in the interior of C3 , so that Qx − Py is not identically zero in the
interior of C3 — it is not even defined throughout the interior of C3 . We
can work around this obstruction by considering the interior of C3 with
the interior of C1 removed. That is, by considering
R3 = (x, y) x2 + (y − 2)2 ≤ 9, x2 + y 2 ≥ 1
Cρ = (x, y) x2 + y 2 = ρ2
It is sketched on the left below. We can now argue as in part (b). The
boundary of R4 consists of two parts
• the circle C4 , oriented counterclockwise, and
• the circle −Cρ . That is, the circle Cρ but oriented clockwise, rather
than counterclockwise.
R5 = (x, y) x2 + y 2 ≤ 1, x2 + y 2 ≥ ρ2
y y
C1
−Cρ
(2, 2) x
−Cρ C4 R5
x
R4
R
C
(1, 0) x
Note that (0, 0) is not in R. So Qx − Py is defined and zero throughout
R. So, by Green’s theorem (Theorem 4.3.2),
Z ZZ
F · dr = Qx − Py dxdy = 0
C R
APPENDIX D. SOLUTIONS TO EXERCISES 623
(e) Note that (0, 0) is in the interior of triangle C specified for this
part. So Qx − Py is not defined in that interior and we cannot apply
Green’s theorem precisely as we did in part (d). We can work around this
obstruction by
• picking a number r > 0 that is small enough that the circle Cr , of
radius r centred on (0, 0), is completely contained in the interior of
the triangle C.
• Then we work with the region R defined by removing the interior of
the circle Cr from the interior of the triangle C. It is the grey region
sketched below.
y
(0, 1)
R C
−Cr
(0, 1)
x
(1, 1)
The boundary of R consists of two parts
Here is a sketch of R.
y
(0, 1)
C
L R
x
x2 + y 2 = 1
(0, −1)
The boundary of R consists of two parts — C on the right and the line
segment L from (0, 1) to (0, −1) on the left. Note that F = F1 ı̂ı + F2 ̂ is
well-defined on all of R and that we still have, from part (a),
∂F2 ∂F1
− = 2y 2
∂x ∂y
on all of R. So, by Green’s theorem (Theorem 4.3.2),
Z p ZZ Z
∂F2 ∂F2
1 + x3 dx + 2xy 2 + y 2 dy =
− dxdy − F · dr
C ∂x ∂y
Z ZR Z L
= 2 2
2y 2 dxdy + F2 dy
x +y ≤1
x≥0
−L
APPENDIX D. SOLUTIONS TO EXERCISES 625
ZZ Z 1
2
= y dxdy + y 2 dy
x2 +y 2 ≤1 −1
by symmetry for the first integral and
since x = 0 and dx = 0 in the
second integral
π 2
= +
4 3
In the second line, we used the notation −L for the line segment from
(0, −1) to (0, 1).
x = cos y
x
C
So the integral is
Z
(x2 + yex ) dx + (x cos y + ex ) dy
C
Z π/2
n dx
= x(t)2 + y(t)ex(t) (t)
−π/2 dt
dy o
+ x(t) cos(y(t)) + ex(t)
(t) dt
dt
Z π/2 n
o
− cos2 t + tecos t sin t + cos2 t + ecos t dt
=
−π/2
Z π/2 n o
= − cos2 t sin t + cos2 t − tecos t sin t + ecos t dt
−π/2
π/2
d h cos3 t
Z n io
= cos2 t + + tecos t dt
−π/2 dt 3
Z π/2 n
cos(2t) + 1 d h cos3 t io
= + + tecos t dt
−π/2 2 dt 3
3 iπ/2
h sin(2t) t cos t
= + + + tecos t
4 2 3 −π/2
3π
=
2
APPENDIX D. SOLUTIONS TO EXERCISES 626
R π/2
For an efficient, sneaky, way to evaluate −π/2
cos2 t dt see Example
2.4.4.
(2) Green’s (or Stokes’) theorem: The curve C is not closed so we cannot
apply Green’s theorem directly. However the boundary of the region
π π
R = (x, y) 0 ≤ x ≤ cos y, − ≤ y ≤
2 2
(sketched below) consists
of two parts, one of which is C. The other
is the line L from 0, π2 to 0, − π2 .
x = cos y
L
R
x
C
3
Then G is conservative with potential g R= x3 + yex and H is pretty
simple, so that it is not hard to evaluate C H · dr directly. Using the
parametrization r(t) = cos tı̂ı + t ̂, − π2 ≤ t ≤ π2 as above,
Z Z Z
F · dr = G · dr + H · dr
C C C
Z Z
= ∇ g · dr + H · dr
C C
Z π/2
π π dy
= g r( ) − g r(− ) + x(t) cos(y(t)) (t) dt
2 2 −π/2 dt
Z π/2
π π
= g 0, − g 0, − + cos2 t dt
2 2 −π/2
Z π/2
π π cos(2t) + 1
= − − + dt
2 2 −π/2 2
h sin(2t) t iπ/2
=π+ +
4 2 −π/2
3π
=
2
4.3.1.19. ∗. Solution. Call the region enclosed by the curve R. By
Green’s theorem, Theorem 4.3.2,
I ZZ ZZ
1 1 ∂ ∂ 1
x dy − y dx = x− (−y) dxdy = 2 dxdy
2 C 2 R ∂x ∂y 2 R
=A
∂ ∂
⇐⇒ [(x + y) − A(2x − 3y)] = [(x + 3y) − A(x + y)]
∂x ∂y
⇐⇒ 1 − 2A = 3 − A
⇐⇒ A = −2
4.3.1.21. ∗. Solution. (a) Parametrize the circle r(θ) = (cos θ, sin θ).
Then
dr
(θ) = − sin θ ı̂ı + cos θ ̂
dθ
dr
F r(θ) · (θ) = − sin4 θ − cos2 θ sin2 θ = − sin2 θ
dθ
I Z 2π Z 2π
dr
F · dr = F r(θ) · (θ) dθ = − sin2 θ dθ
C 0 dθ 0
Z 2π
1 − cos(2θ)
=− dθ
0 2
h θ sin(2θ) i2π
=− − = −π
2 4 0
R 2π
For an efficient, sneaky, way to evaluate 0 sin2 θ dθ see Example 2.4.4.
(b) Denote by W the washer shaped region between the circle x2 +y 2 =
2 2
1 and the ellipse x16 + y25 = 1. It is sketched below. By Green’s theorem
I I ZZ h
∂ ∂ i
F · dr − F · dr = F2 − F1 dxdy
C0 C W ∂x ∂y
∂ ∂ ∂ xy 2 ∂ y3
F2 − F1 = − −
∂x ∂y ∂x (x2 + y 2 )2 ∂y (x2 + y 2 )2
y2 xy 2 (2x) 3y 2 y 3 (2y)
=− 2 +2 3 − 2 +2 3
(x2 + y 2 ) (x2 + y 2 ) (x2 + y 2 ) (x2 + y 2 )
−y 2 (x2 + y 2 ) + 4x2 y 2 − 3y 2 (x2 + y 2 ) + 4y 4
= 3
(x2 + y 2 )
=0
Consequently
I I I I
F · dr − F · dr = 0 =⇒ F · dr = F · dr = −π
C0 C C0 C
W
C C0
APPENDIX D. SOLUTIONS TO EXERCISES 629
except at (0,H 0), where F is not defined. Hence by Green’s theorem (Theo-
rem 4.3.2), C F · dr = 0 forH any closed curve that does not contain (0, 0) in
its interior. In particular, C1 F · dr = 0. On the other hand, (0, 0) is con-
tainedH in the interior of C2 , so we cannot use Green’s theorem to conclude
that C2 F · dr = 0.
Let C3 be the circle of radius one centred on (0, 0) and denote by W the
washer shaped region between the circle C2 and the circle C3 . It is sketched
below.
W
C3 C2
so that I I Z 2π
F · dr = F · dr = dθ 1 = 2π
C2 C3 0
x
C2
Then
Z Z
x ds = 0 ds = 0
C1 C1
APPENDIX D. SOLUTIONS TO EXERCISES 630
Z Z 1
1
x ds = x dx =
C2 0 2
dy
On C3 , y = 1 − x2 so that dx = −2x and
r dy 2
p p
ds = dx2 + dy 2 = 1+ dx = 1 + 4x2 dx
dx
and
Z Z 1 p
x ds = x 1 + 4x2 dx
C3 0
h1 i1 1 3/2
(1 + 4x2 )3/2 =
= 5 −1
12 0 12
All together Z
1 1 3/2
x ds = + 5 − 1 ≈ 1.3484
C 2 12
(b) By either Stokes’ theorem or Green’s theorem
Z ZZ h
∂ 2 ∂ i
x + cos(y 2 ) − sin(x2 ) − xy dxdy
F · dr =
C ∂x ∂y
Z ZR
= 3x dxdy
R
Z 1 Z 1−x2
=3 dx dy x
0 0
Z 1
=3 dx (1 − x2 )x
0
h1 1i 3
=3 − =
2 4 4
4.3.1.24. ∗. Solution. (a) If (x, y, z) is on the curve, it must obey both
z = x + y and z = x2 + y 2 and hence it must also obey x2 + y 2 = x + y or
(x − 21 )2 + (y − 12 )2 = 12 . That’s a circle. We can parametrize the curve by
1 1
x(θ) = + √ cos θ
2 2
1 1
y(θ) = + √ sin θ
2 2
1
z(θ) = x + y = 1 + √ cos θ + sin θ
2
with 0 ≤ θ < 2π. As θ runs from 0 to 2π, x(θ), y(θ) runs once around the
circle without crossing itself so that x(θ), y(θ), z(θ) runs once around the
curve without crossing itself. As x(2π), y(2π), z(2π) = x(0), y(0), z(0) ,
C is a simple closed curve.
(b) (i) The vector field F =H x2 ı̂ı + y 2 ̂ + 3ez k̂ is conservative (with
potential 31 x3 + 13 y 3 + 3ez ). So C F · dr = 0.
(b) (ii) Note that the question did not specify the orientation of C.
It should have. We’ll stick with the most commonly used orientation —
counterclockwise when viewed from high on the z-axis. The vector field
G = 3ez k̂ is conservative (with potential 3ez ). So C G · dr = 0 and, using
H
APPENDIX D. SOLUTIONS TO EXERCISES 631
the parametrization
h1 1 i h1 1 i h 1 1 i
r(θ) = + √ cos θ ı̂ı + + √ sin θ ̂ + 1+ √ sin θ+ √ cos θ k̂
2 2 2 2 2 2
1 1 h 1 1 i
r0 (θ) = − √ sin θ ı̂ı + √ cos θ ̂ + √ cos θ − √ sin θ k̂
2 2 2 2
of part (a), we have
I I
F · dr = (F − G) · dr
C C
Z 2π
y(θ)2 x0 (θ) + x(θ)2 y 0 (θ) dθ
=
0
Z 2π h1 i2 1 i2 1
n 1 h1 1 o
= − + √ sin θ √ sin θ + + √ cos θ √ cos θ dθ
0 2 2 2 2 2 2
Because the integral of any odd power of sin θ or cos θ over 0 ≤ θ ≤ 2π is
zero (see Example 4.4.6 in the text),
I Z 2π n 1 1 o
F · dr = − sin2 θ + cos2 θ dθ
c 0 2 2
=0
R = (x, y) 6x2 + 3y 2 ≤ 1
S
∂S
The correct normal to this surface is sketched in
APPENDIX D. SOLUTIONS TO EXERCISES 632
n̂
S
∂S
It is correct because
• if you walk along ∂S in the direction of the arrow on ∂S,
• with the vector from your feet to your head having direction n̂
∂S
To deal with part (c), we can first rotate the flat disk that we considered
above to get
n̂
∂S
S
We can push on this S to deform it to the S of part (c). This gives the
solution to part (c).
(c)
∂S
n̂
S
n̂
y
C R
x
We are going to apply Stokes’ theorem (Theorem 4.4.1) with S being the
APPENDIX D. SOLUTIONS TO EXERCISES 633
given region R in the xy-plane and with F(x, y, z) = F1 (x, y)ı̂ı + F2 (x, y) ̂.
Then
• the unit normal vector to S specified in Stokes theorem is k̂ (if you
walk along ∂S = C in the direction of the arrow on C with the vector
from your feet to your head having direction k̂ then S = R is on your
left hand side) and
• dS = dx dy and
• the curl of F is
ı̂ı ̂ k̂ ∂F
∂ ∂ ∂ 2 ∂F1
∇ × F = det ∂x ∂y ∂z = − k̂
∂x ∂y
F1 (x, y) F2 (x, y) 0
HWe will show below that ∇ × [φ∇ ∇ψ + ψ∇ ∇φ] = 0. This will imply that
C
∇
[φ∇ ψ + ∇
ψ∇ φ] · dr = 0. One way to see ∇ψ + ψ∇
that ∇ × [φ∇ ∇φ] = 0 is
∇ψ + ψ∇
∇ × [φ∇ ∇φ] = ∇ × [∇
∇(φψ)] (by part (c) of Theorem 4.1.3)
=0 (by part (b) of Theorem 4.1.7)
∇ψ + ψ∇
Another way to see that ∇ × [φ∇ ∇φ] = 0 is
∇ψ + ψ∇
∇ ×[φ∇ ∇φ] = ∇ φ×∇
∇ψ + φ∇
∇ ×(∇
∇ψ) + ∇ ψ×∇
∇φ + ψ∇
∇ ×(∇
∇φ)
= ∇φ × ∇ψ + ∇ψ × ∇φ
∇ × (∇
since φ∇ ∇ψ) = ψ∇
∇ × (∇
∇φ) = 0
=0
4.4.3.4. Solution. (a) Observe that x(t) = cos t and y(t) = sin t obey
x(t)2 + y(t)2 = 1. Then z(t) = y(t)2 = sin2 t. So we may parametrize the
curve by r(t) = (cos t, sin t, sin2 t) with 0 ≤ t ≤ 2π. Then
F r(t) · r0 (t) = − sin t cos2 t + sin2 t + sin2 t cos t + cos2 t + 2 sin t cos t
1 d
=1+ [cos3 t + sin3 t] + sin(2t)
3 dt
I Z 2π
1 d 3 3
F · dr = 1+ [cos t + sin t] + sin(2t) dt
C 0 3 dt
APPENDIX D. SOLUTIONS TO EXERCISES 634
2π
1 3 3 1
= t + [cos t + sin t] − cos(2t)
3 2 0
= 2π
Note that r(t) = x(t)ı̂ı +y(t) ̂ +z(t) k̂ obeys x(t)+y(t)+z(t) = 3, for every
t, and that x(t)ı̂ı + y(t) ̂ = (1 + cos t)ı̂ı + (1 + sin t) ̂ runs counterclockwise
around the circle of radius 1 centered on (1, 1). So we choose S to be the
part of the plane G(x, y, z) = x + y + z = 3 with (x − 1)2 + (y − 1)2 ≤ 1.
Then, by Stokes’ Theorem,
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = k̂ · n̂ dS
C S S
with
∇G
n̂ dS = ± dxdy = ± ı̂ı + ̂ + k̂ dxdy
∇ G · k̂
As (1 + cos t)ı̂ı + (1 + sin t) ̂ runs counterclockwise around the circle (x −
1)2 + (y − 1)2 ≤ 1, Stokes’ theorem specifies the plus sign and
I ZZ
F · dr = dx dy = π
C (x−1)2 +(y−1)2 ≤1
∂S = (x, y, z) z = 0, x2 + y 2 = 4
n̂
V S
∂S
So, by Stokes’ theorem (Theorem 4.4.1)
ZZ I
∇ × F · n̂ dS = F · dr
S ∂S
F(r(θ)) r (θ) 0
Z 2π z }| { z }| {
= (−2 sin θ ı̂ı + 2 cos θ ̂ − 2 cos θ k̂) · (−2 sin θ ı̂ı + 2 cos θ ̂) dθ
0
Z 2π
=4 dθ
0
= 8π
n̂ S
D
C
Note that C is the boundary curve for both S and D. So, by Stokes’
APPENDIX D. SOLUTIONS TO EXERCISES 636
theorem, twice
ZZ Z ZZ
∇ × F · n̂ dS = F · dr = ∇ × F · n̂ dS
S
ZCZ D
= ∇ × F · k̂ dS
D
by (3.3.2). As (x, y, z) runs over S, (x, y) runs over the circular disk
D = (x, y) x2 + (y − 1)2 ≤ 1
By oddness under x → −x, all terms integrate to zero except for the last.
So I ZZ
F · dr = dxdy = Area(D) = π
C D
S = (x, y, z) − 1 ≤ x ≤ 1, −1 ≤ y ≤ 1, z ≥ 0, z = (1 − x2 )(1 − y 2 )
= (x, y, z) − 1 ≤ x ≤ 1, −1 ≤ y ≤ 1, z = (1 − x2 )(1 − y 2 )
dy=0
(−1, 1) y=1 (1, 1)
dx=0 dx=0
x=−1 x=1
=8
4.4.3.10. Solution. We shall apply Stokes’ Theorem. The curl of F is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
2
ex − yz sin y − yz xz + 2y
= (2 + y)ı̂ı − (z + y) ̂ + (0 + z) k̂
x
The equation of the surface is z = f (x, y) = 1 − x − y. So, by (3.3.2),
n̂ dS = − fx ı̂ı − fy ̂ + k̂ dx dy
= (ı̂ı + ̂ + k̂) dx dy
Here n̂ is the upward pointing unit normal. The set (x, y) for
of points
which there is a corresponding (x, y, z) in S is T = (x, y) x ≥ 0, y ≥
0, x + y ≤ 1 , which is a triangle of area 12 . Since
we have
I ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS
= 2 dx dy = 2 Area(T ) = 1
T
APPENDIX D. SOLUTIONS TO EXERCISES 638
4.4.3.11.
H ∗. Solution.
RR Stokes’ theorem, which is Theorem 4.4.1, says
that C F · dr = S ∇ × F · n̂ dS for any surface S whose boundary is C.
For the given vector field
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F(x, y, z) = det ∂x ∂y ∂z
−z x y
= ı̂ı − ̂ + k̂
Choose
x2 y2 z2
S= (x, y, z) z = y, 4 + 2 + 2 ≤1
x2 2
= (x, y, z) z = y, 4 +y ≤1
(0, 1, 1)
x2 y2 z2
y=z 4
+ 2
+ 2
=1
y
(2, 0, 0)
x
As S is part of the plane z = f (x, y) = y, (3.3.2), gives that
n̂ dS = ± − fx , −fy , 1 dxdy
= ±(0 , −1 , 1)dxdy
I Z Z z ∇×F}| { z
n̂ dS
}| {
F · dr = (1, −1, 1) · (0 , −1 , 1)dxdy
C
ZD
Z
=2 dxdy = 2 Area(D)
D
2
The ellipse D, that is x4 + y 2 ≤ 1, has semi-axes a = 2 and b = 1 and
hence area πab = 2π. Finally
I
F · dr = 2 Area(D) = 4π
C
L2
L3
(0,2,0)
y
(2,0,0)
L1
x
Two of the sides of the triangle are (0, 2, 0) − (2, 0, 0) = (−2, 2, 0) and
(0, 0, 2) − (0, 2, 0) = (0, −2, 2) so the area of the triangle is
ı̂ı ̂ k̂ √
1 1 1
(−2, 2, 0) × (0, −2, 2) = det −2 2 0 = (4, 4, 4) = 2 3
2 2 2
0 −2 2
ı̂ı+̂
+k̂
The upward pointing unit normal to E is n̂ = √
3
. So, by Stokes’
theorem,
Z ZZ
I2 = F · dr = ∇ × F · n̂ dS
C2 R
ZZ
ı̂ı + ̂ + k̂
= 2y ı̂ı + 2z ̂ + 2x k̂ · √ dS
R 3
Z Z z=2 }| on R
2 { 4 √
= √ y + z + x dS = √ Area(R) = 4 3
3 R 3
(a) Denote by T the triangle enclosed by C1 . By the computation that
we have just done in part (b)
Z
4
I1 = F · dr = √ Area(T ) = 8
C1 3
4.4.3.13. ∗. Solution. (a) Observe that
• the curve C1 is one quarter of a circle in the xy-plane, centred on the
origin, of radius 2, starting at (2, 0, 0) and ending at (0, 2, 0) and
C2
C3
(0,2,0)
(2,0,0) y
x C1
(b) C lies completely on the sphere x2 + y 2 + z 2 = 4. So it is natural
to choose
S = (x, y, z) x2 + y 2 + z 2 = 4, x ≥ 0, y ≥ 0, z ≥ 0
(3.3.1) gives
∂r ∂r
n̂ dS = ± × dθdϕ
∂θ ∂ϕ
= ∓4 cos θ sin ϕ ı̂ı + sin θ sin ϕ ̂ + cos ϕ k̂ sin ϕ dθdϕ
(c) The vector field F looks too complicated for a direct evaluation of
the line integral. So, in preparation for an application of Stokes’ theorem,
we compute
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det
∂x ∂y ∂z
2
y + sin(x2 ) z − 3x + ln(1 + y 2 ) y + ez
APPENDIX D. SOLUTIONS TO EXERCISES 641
= −4 k̂
∂S1 ∂S2a
S2 ∂S2b
S1
n̂ n̂
(b) The boundary, ∂S2 , of S2 consists of two parts, a circle in the plane
z = 4 and a circle in the plane z = 1. We’ll call the first part p
∂S2a . It is the
same as ∂S1 . We’ll call the second part ∂S2b . It is the circle x2 + y 2 = 1,
z = 1 oriented counterclockwise when viewed from high on the z-axis. We
can parametrize it
r(t) = cos tı̂ı + sin t ̂ + k̂, 0 ≤ t ≤ 2π
So, on ∂S2b ,
F r(t) · dr = − sin t , cos t , sin t cos t cos(sin t) · − sin t , cos t , 0 dt
= dt
and, by Stokes’ theorem,
ZZ I I
∇ × F · n̂ dS = F r(t) · dr + F r(t) · dr
S2 ∂S2a ∂S2b
Z 2π
= −128π + dt
0
= −126π
APPENDIX D. SOLUTIONS TO EXERCISES 642
S = (x, y, z) z = x + 4, x2 + y 2 ≤ 4
C x2 + y 2 = 4
S
n̂
y
x
We may parametrize S by
So,
ı̂ı ̂ k̂
∂r ∂r
× = det 1 0 1 = − 1, 0, 1
∂x ∂y
0 1 0
and, by (3.3.1),
∂r ∂r
n̂ dS = − × dxdy = (1, 0, −1) dxdy
∂x ∂y
∂r ∂r
We have chosen to “−” sign in n̂ dS = ± ∂x × ∂y dxdy to give the downward
pointing normal. As the curl of F is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
x3 + 2y sin(y) + z x + sin(z 2 )
= −ı̂ı − ̂ − 2k̂
= dxdy = 4π
S
4.4.3.16. ∗. Solution. (a) Note that all three vertices, (2, 0, 0), (0, 2, 0)
and 0, 0, 2), lie in the plane x + y + z = 2. So the entire path lies in that
plane too.
APPENDIX D. SOLUTIONS TO EXERCISES 643
z
(0, 0, 2)
n̂
y
(0, 2, 0)
x (2, 0, 0)
In part (b) we will need to evaluate a line integral that clearly cannot
be computed directly — we will need to use Stokes’ theorem. So let’s use
Stokes’s theorem in part (a) too. First, we find
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x ∂y
ı ı
∂z = 2y ı̂ + 2z ̂ + 2xı̂
z 2 x2 y 2
1 2
∇ × F · n̂ = √ 2y ı̂ı + 2z ̂ + 2xı̂ı · (ı̂ı + ̂ + k̂) = √ (x + y + z)
3 3
and, by Stokes’ theorem,
I ZZ ZZ ZZ
2 4
F · dr = ∇ × F · n̂ dS = √ (x + y + z) dS = √ dS
C S 3 S 3 S
4
= √ Area(S)
3
The triangle S is half of the prallelogram with sides (0, 2, 0) − (2, 0, 0) =
(−2, 2, 0) and (0, 0, 2) − (2, 0, 0) = (−2, 0, 2). The area of the parallelogram
is
√
(−2, 2, 0) × (−2, 0, 2) = (4, 4, 4) = 4 3
So
4 √
I
F · dr = √ 2 3 = 8
C 3
The curl of F is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
1 x
z + 1+z xz 3xy − (z+1)2
1 1
= (3x − x)ı̂ı − 3y − − 1 + ̂ + z k̂
(z + 1)2 (1 + z)2
= 2xı̂ı + (1 − 3y) ̂ + z k̂
Write
(x, y, z) z = f (x, y) = 1 − x2 y, x2 + y 2 ≤ 1
S=
For S, with the upward pointing normal, by (3.3.2),
n̂ dS = − fx , −fy , 1 dxdy
= 2xy , x2 , 1 dxdy
so that
z
z }| {
∇ × F · n̂ dS = 4x2 y + (x2 − 3x2 y) + (1 − x2 y) dxdy
So
Z ZZ ZZ
x2 + 1 dxdy = π + x2 dxdy
F · dr =
C x2 +y 2 ≤1 x2 +y 2 ≤1
Since
Z 2π Z 2π 2π
1 + cos(2θ) θ sin(2θ)
cos2 θ dθ = dθ = + =π
0 0 2 2 4 0
R1 R 2π π
we finally have 0
dr r3 0
dθ cos2 θ = 4 and
Z
π 5π
F · dr = π + =
C 4 4
R 2π
For an efficient, sneaky, way to evaluate 0
cos2 θ dθ, see Example 2.4.4.
4.4.3.18. ∗. Solution. We are to evaluate a line integral around a curve
C. We are told that C is the boundary of a surface S that is contained in
the plane x + y + z = 1, but we are not told precisely what C is. So we are
APPENDIX D. SOLUTIONS TO EXERCISES 645
C
n̂
y
x
observed from the positive z-axis looking down at the plane, n̂ is to
point downwards, so that
1
n̂ dS = − √ ı̂ı + ̂ + k̂) dS
3
On S we have x + y + z = 1, so that Stokes’ theorem gives
I ZZ
F · dr = ∇ × F) · n̂ dS
(∇
C S
ZZ
1
= 2(y ı̂ı + z ̂ + x k̂) · − √ ı̂ı + ̂ + k̂) dS
S 3
ZZ ZZ
2 2
= −√ (y + z + x) dS = − √ dS
3 S 3 S
10
= −√
3
since S has area 5.
4.4.3.19. ∗. Solution. We are to evaluate the line integral of a com-
plicated vector field around a relatively complicated closed curve. That
certainly suggests that we should not try to evaluate the integral directly.
To see if Stokes’ theorem looks promising, let’s compute the curl
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det = k̂
∂x ∂y √ ∂z
x 4
−y + e sin x y z tan z
z
(0, 1, 2)
(0, 1, 1)
Q
(0, 2, 0)
(0, 0, 0)
y
T
(2, 2, 0)
x
We can choose the surface S to be the union of two flat parts:
• the quadralateral Q in the yz-plane with vertices (0, 0, 0), (0, 1, 1),
(0, 1, 2) and (0, 2, 0) and
• the triangle T in the xy-plane with vertices (0, 0, 0), (0, 2, 0) and
(2, 2, 0).
The normal to Q is −ı̂ı and the normal to T is −k̂. Then Stokes’ theorem
gives
Z ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS ZZ
= k̂ · (−ı̂ı) dS + k̂ · (−k̂) dS
Q T
ZZ
=− dS
T
= −Area(T )
base height
z}|{ z}|{
= − 12 2 2
= −2
4.4.3.20. ∗. Solution. The integral looks messy. Let’s compute the
curl of
F = (z + sin z)ı̂ı + (x3 − x2 y) ̂ + (x cos z − y) k̂
to help gauge if Stokes’ theorem would be easier.
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
z + sin z x3 − x2 y x cos z − y
= −ı̂ı + ̂ + (3x2 − 2xy) k̂
S = (x, y, z) z = xy 2 , x2 + y 2 ≤ 1
D = (x, y) x2 + y 2 ≤ 1
APPENDIX D. SOLUTIONS TO EXERCISES 647
x2 + y 2 + z 2 = 2
S z=1
∂S
y
x
The boundary, ∂S, of S is the circle x2 + y 2 = 1, z = 1, oriented
counterclockwise when viewed from above. It is parametrized by
cos(2θ) + 1
cos2 θ =
2
2
cos (2θ) + 2 cos(2θ) + 1
=⇒ cos4 θ =
4
1 cos(4θ) + 1 cos(2θ) 1
= + +
4 2 2 4
3 cos(4θ) cos(2θ)
= + +
8 8 2
APPENDIX D. SOLUTIONS TO EXERCISES 648
Finally
ZZ Z 2π 3 cos(4θ) cos(2θ) 3π
∇ × F · n̂ dS = + + dθ =
S 0 8 8 2 4
4.4.3.22. ∗. Solution. We are to evaluate the line integral of a com-
plicated vector field around a relatively complicated closed curve. That
certainly suggests that we should not try to evaluate the integral directly.
As we are to use Stokes’ theorem, let’s compute the curl
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x
∂y ∂z
2
x sin y −y sin x (x − y)z
= −z 2 ı̂ı − z 2 ̂ − (y cos x + x cos y)k̂
(π/2, 0, 1) Sy
Sx
(0, π/2, 0) y
(π/2, 0, 0)
x
We can choose the surface S to be the union of two flat parts:
• the rectangle Sx in the xz-plane with vertices (0, 0, 0), ( π2 , 0, 0), ( π2 , 0, 1)
and (0, 0, 1) and
• the rectangle Sy in the yz-plane with vertices (0, 0, 0), (0, 0, 1), (0, π2 , 1)
and (0, π2 , 0)
The normal to Sx is −̂ and the normal to Sy is −ı̂ı. Then Stokes’ theorem
gives
Z ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS ZZ
= ∇ × F · (−̂) dS + ∇ × F · (−ı̂ı) dS
Sx Sy
π π
Z 2
Z 1 Z 2
Z 1
= dx dz z 2 + dy dz z 2
0 0 0 0
Z π Z π
2 1 2 1
= dx + dy
0 3 0 3
π
=
3
4.4.3.23. ∗. Solution. (a) Here is a sketch.
APPENDIX D. SOLUTIONS TO EXERCISES 649
z
(0, 0, 2)
Sy
Sx y
(0, 3, 0)
(2, 0, 0)
x
(b) We are to evaluate the line integral of a complicated vector field
around a relatively complicated closed curve. That certainly suggests that
we should not try to evaluate the integral directly. Let’s try Stokes’ theo-
rem. First, we compute the curl
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det
∂x ∂y ∂z
2z 2 3z 2
1+y + sin(x ) 1+x + sin(y ) 5(x + 1)(y + 2)
3 2
= 5(x + 1) − ı̂ı − 5(y + 2) − ̂
1+x 1+y
3z 2z
+ − 2
+ k̂
(1 + x) (1 + y)2
• the triangle Sy in the yz-plane with vertices (0, 0, 0), (0, 0, 2), and
(0, 3, 0)
Note that
• The normal to Sx specified by Stokes’ theorem is −̂. On Sx we have
2
y = 0, so that ∇ × F · ̂ simplifies to − 5(0 + 2) − 1+0 = −8.
• The normal to Sy specified by Stokes’ theorem is −ı̂ı.On Sy we have
3
x = 0, so that ∇ × F · ı̂ı simplifies to 5(0 + 1) − 1+0 = 2.
So Stokes’ theorem gives
Z ZZ
F · dr = ∇ × F · n̂ dS
C S
ZZ 8 ZZ −2
z }| { z }| {
= ∇ × F · (−̂) dS + ∇ × F · (−ı̂ı) dS
Sx Sy
1 1
= 8 Area(Sx ) − 2 Area(Sy ) = 8 (2)(2) − 2 (3)(2)
2 2
= 10
= G · n̂ dS
D
1 + cos(2θ) 1 − cos(2θ)
cos2 θ = sin2 θ =
2 2
we have
2 2
4 1 − cos(2θ)
4 1 + cos(2θ)
sin θ + cos θ = +
4 4
2
1 cos (2θ) 1 1 + cos(4θ)
= + = +
2 2 2 4
So
ZZ Z 2π
3 1 3
G · n̂ dS = 16 + cos(4θ) dθ = 16 × × (2π) = 24π
S 0 4 4 4
4.4.3.26. ∗. Solution. Note that
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x = 2y ı̂ı + 2z ̂ + 2x k̂
2 ∂y2 ∂z2
z x y
Since
ı̂ı ̂ k̂
∂ ∂ ∂ 2
∇ × F = det ∂x = · · · , · · · , y − xz
∂y ∂z
2 2
xyz xy x yz
and n̂ = k̂ on S 0 and z = −1 on S 0
ZZ Z 1 Z 1
∇ × F · n̂ dS = dx dy · · · , · · · , y 2 − xz · k̂
S0 −1 −1 z=−1
APPENDIX D. SOLUTIONS TO EXERCISES 652
Z 1 Z 1
= dx dy (y 2 + x)
−1 −1
Z 1 Z 1 Z 1
= dx dy y 2 = 2 × 2 dy y 2
−1 −1 0
4
=
3
C4 C2
xy 2 3 1
Z Z 1 Z 1
z}|{ y 2
y 2 dy =
F · dr = F r(y) · ̂ dy = =
C2 −1 −1 3 −1 3
Parametrize C4 by y. That is, r(y) = −ı̂ı + y ̂ − k̂, with y running from 1
to −1. Since r0 (y) = ̂, and x = −1 on C4 ,
xy 2 3 −1
Z Z −1 Z −1 z }| { y 2
(−1)y 2 dy = −
F · dr = F r(y) · ̂ dy = =
C4 1 1 3 1 3
All together
ZZ Z Z Z Z
4
∇ × F · n̂ dS = F · dr + F · dr + F · dr + F · dr =
S C1 C2 C3 C4 3
APPENDIX D. SOLUTIONS TO EXERCISES 653
4.4.3.28. Solution. Let’s try Stokes’ Theorem. Call F = y ı̂ı −x ̂ +xy k̂.
Then
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x ∂y ∂z = xı̂ı − y ̂ − 2 k̂
y −x xy
To evaluate the integrals of x and y we use that, for any region R in the
xy--plane, RR RR
R
x dx dy y dx dy
x̄ = ȳ = R
Area(R) Area(R)
(x−1)2 y2
Our ellipse is 42 + = 1 and so has area πab = π × 4 × 2 = 8π and
22
2 2
centroid (x̄, ȳ) = (1, 0). So, using R = (x, y) (x−1) + y22 ≤ 1 ,
42
I ZZ
F · dr = [7 − 4x − 2y] dx dy
C R
= Area(R) 7 − 4x̄ − 2ȳ}
= 8π[7 − 4 × 1 − 2 × 0] = 24π
4.4.3.30. ∗. Solution. (a) The curl is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det = (1 − 2xz) ̂
∂x ∂y ∂z
2 + x2 + z 0 3 + x2 z
(b) We are going to use Stokes’ theorem. The specified curve C is not
closed and so is not the boundary of a surface. So we extend C to a closed
curve C̃ by appending to C the line segment L from (2, 0, 0) to (0, 0, 0). In
the figure below, C is the red curve and C̃ is C plus the blue line segment.
z
(0, 0, 3)
−ı̂ı
C C
T1
(0, 1, 0)
T2 y
L
(2, 0, 0) C
x −k̂
The closed curve C̃ is boundary of the surface S that is the union of
• the triangle T1 in the yz-plane with vertices (0, 0, 0), (0, 0, 3) and
(0, 1, 0) and with normal vector −ı̂ı and
• the triangle T2 in the xy-plane with vertices (0, 0, 0), (0, 1, 0) and
(2, 0, 0) and with normal vector −k̂.
APPENDIX D. SOLUTIONS TO EXERCISES 655
The part of S in the first octant is sketched in the figure on the left below.
S consists of two parts — the cylindrical surface
S1 = (x, y, z) y 2 + z 2 = 9, 0 ≤ x ≤ 5
(a) using Stokes’ theorem: Let’s use Stokes’ theorem. The boundary ∂S
of S is the cirlce y 2 + z 2 = 9, x = 5, oriented clockwise when viewed from
far down the x-axis. We’ll parametrize it by r(θ) = 5, 3 cos θ, −3 sin θ .
Then Stokes’ theorem gives
ZZ I
∇ × G · n̂ dS = G · dr
S ∂S
Z 2π
= 5 , 3 sin θ, 3 cos θ · 0 , −3 sin θ , −3 cos θ dθ
0
Z 2π
− 9 sin2 θ − 9 cos2 θ dθ
=
0
= −18π
APPENDIX D. SOLUTIONS TO EXERCISES 656
z z
S S
y y
∂S n̂ T n̂
x x ı̂ı
(b) This time we’ll use the divergence theorem. The surface S is not
closed. So we’ll use the auxilary surface formed by “topping S off” with
the cap T = (5, y, z) y 2 + z 2 ≤ 9 . If we give T the normal vector
ı̂ı, this auxiliary
surface, the union of S and T , is the boundary of V =
(x, y, z) y 2 + z 2 ≤ 9, 0 ≤ x ≤ 5 . So the divergence theorem gives
ZZ ZZ ZZ
F · n̂ dS + F · n̂ dS = F · n̂ dS
S T
Z Z∂V
Z
= ∇ · F dV
V
=0
(d) The integrand for direct evaluation looks very complicated. On the
other hand ∇ × F is quite simple. So let’s try Stokes’ thoerem. Denote
S = (x, y, z) 2 ≤ x ≤ 4, 2 ≤ y ≤ 4, z = 2
C
(4,2,2) (4,4,2)
n̂
x
Then, by Stokes’ theorem,
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = ∇ × F · (−k̂) dS
C
ZSZ S
2x − x1 dS
=−
S
Z 4 Z 4 Z 4
1 1
=− dx dy 2x − x =− dx 2 2x − x
2 2 2
h i4
= −2 x2 − ln x = −2 12 − ln 2
2
= 2 ln 2 − 24
So let’s find a vector potential G. That is, let’s try and find a vector field
G = G1 ı̂ı + G2 ̂ + G3 k̂ that obeys ∇ × G = F, or equivalently,
∂G3 ∂G2
− = F1 = xz
∂y ∂z
∂G3 ∂G1
− + = F2 = 2xey z − yz
∂x ∂z
∂G2 ∂G1
− = F3 = y 2 − xey z 2
∂x ∂y
APPENDIX D. SOLUTIONS TO EXERCISES 658
Let’s also require that G3 = 0. (If this is mysterious to you, review §4.1.2.)
Then the equations above simplify to
∂G2
− = xz
∂z
∂G1
= 2xey z − yz
∂z
∂G2 ∂G1
− = y 2 − xey z 2
∂x ∂y
Now the first equation contains only a single unknown, namely G2 and we
can find all G2 ’s that obey the first equation simply by integrating with
respect to z:
xz 2
G2 = − + N (x, y)
2
∂
Note that, because ∂z treats x and y as constants, the constant of integra-
tion N is allowed to depend on x and y.
Similarly, the second equation contains only a single unknown, G1 , and
is easily solved by integrating with respect to z. The second equation is
satisfied if and only if
1
G1 = xey z 2 − yz 2 + M (x, y)
2
for some function M .
Finally, the third equation is also satisfied if and only if M (x, y) and
N (x, y) obey
∂ xz 2 ∂ y 2 yz 2
− + N (x, y) − xe z − + M (x, y) = y 2 − xey z 2
∂x 2 ∂y 2
which simplifies to
∂N ∂M
(x, y) − (x, y) = y 2
∂x ∂y
This is one linear equation in two unknowns, M and N . Typically, we can
easily solve one linear equation in one unknown.
So we are free to eliminate one of the unknowns by setting, for example,
M = 0, and then choosing any N that obeys
∂N
(x, y) = y 2
∂x
Integrating with respect to x gives, as one possible choice, N (x, y) = xy 2 .
So we have found a vector potential. Namely
1 xz 2
G = xey z 2 − yz 2 ı̂ı + xy 2 − ̂
2 2
We can now evaluate the flux. Parametrize ∂S by
with 0 ≤ θ ≤ 2π. So
ZZ I
F · n̂ dS = G · dr
S ∂S
APPENDIX D. SOLUTIONS TO EXERCISES 659
G(r(θ)) r0 (θ)
Z 2π z }| { z }| {
= cos θ sin2 θ ̂ · − sin θ ı̂ı + cos θ ̂ dθ
0
Z 2π
= sin2 θ cos2 θ dθ
0
2π
1 − cos(2θ) 1 + cos(2θ)
Z
= dθ
0 2 2
1 2π
Z
1 − cos2 (2θ) dθ
=
4 0
1 2π n
Z
1 + cos(4θ) o
= 1− dθ
4 0 2
Z 2π
1 1
= 2π since cos(4θ) dθ = 0
4 2 0
π
=
4
4.4.3.34. ∗. Solution. Considering that there are ten line segments in
C, it is probably not very efficient to use direct evaluation. Two other pos-
sible methods come to mind. If F is conservative, we can use F’s potential.
Even if F is not conservative, it may be possible to efficiently use Stokes’
(or Green’s) theorem. So let’s compute
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x ∂y ∂z = k̂
y 2x − 10 0
As ∇ × F 6= 0, the vector field F is not conservative. As ∇ × F 6= 0 is
very simple, it looks like Stokes’ theorem could provide an efficient way to
compute the integral. The left figure below contains a sketch of C.
y (4, 5) y (4, 5)
(5, 5) (5, 5)
(3, 4) (3, 4)
(4, 4) (4, 4)
(2, 3) (2, 3)
(3, 3) (3, 3)
(1, 2) (1, 2) L
(2, 2) (2, 2)
C C
(0, 1) (1, 1) (0, 1) (1, 1)
(0, 0) x (0, 0) x
The curve C is not closed, and so is not the boundary of a surface, so
we cannot apply Stokes’ theorem directly. But we can easily come up with
a surface whose boundary contains C. Let R be the shaded region in the
figure on the right above. The boundary ∂R of R consists of two parts —
C and the line segment L. The normal of R for −k̂ (since ∂R is oriented
clockwise). So Stokes’ theorem gives
Z Z ZZ ZZ
F · dr + F · dr = ∇ × F · (−k̂) dS = (k̂) · (−k̂) dS
C L R R
= −Area(R)
R is the union of 5 triangles, each of height 1 and base 1. So
1 5
Area(R) = 5 × ×1×1=
2 2
APPENDIX D. SOLUTIONS TO EXERCISES 660
If we denote by −L the line segment from (0, 0) to (5, 5), we can parametrize
−L by r(t) = t(5, 5), 0 ≤ t ≤ 1 and
F(r(t)) r0 (t)
Z Z 1 z }| { z }| {
Z 1
F · dr = 5tı̂ı + (10t − 10) ̂ · (5ı̂ı + 5 ̂) dt = 5 15t − 10 dt
−L 0 0
3 25
= 25 −2 =−
2 2
All together
Z Z Z
5 25
F · dr = −Area(R) − F · dr = −Area(R) + F · dr = − −
C L −L 2 2
= −15
4.4.3.35. ∗. Solution. If we parametrize the curve as
then the term sin x(θ)2 x0 (θ) in the integral will be sin 4 cos2 θ (−2 sin θ).
That looks hard to integrate. So let’s try Stokes’ theorem. The curl of F
is
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x ∂y ∂z = −xı̂ + z k̂
ı
2 2
sin x xz z
S = (x, y, z) x2 + y 2 ≤ 4, z = x2
2π
24
Z
cos(2θ) + 1
=3 dθ
4 0 2
= 12π
R 2π
For an efficient, sneaky, way to evaluate 0
cos2 t dt see Example 2.4.4.
4.4.3.36. ∗. Solution. By Stokes’ Theorem,
I ZZ
E · dr = ∇ × E) · n̂ dS
(∇
C S
This is true for all surfaces S. So the integrand, assuming that it is con-
tinuous, must be zero.
To see this, let G = ∇ × E + 1c ∂H∂t . Suppose that G(x0 ) 6= 0. Pick
a unit vector n̂ in the direction of G(x0 ). Let S be a very small flat disk
centered on x0 with normal n̂ (the vector we picked). Then G(x0 ) · n̂ > 0
and, by continuity,G(x) · n̂ > 0 for
all x on S, if we have picked S small
1 ∂H
RR
enough. Then S ∇ × E + c ∂t · n̂ dS > 0, which is a contradiction. So
G = 0 everywhere and we conclude that
1 ∂H
∇×E+ =0
c ∂t
4.4.3.37. ∗. Solution. The curl of the specified vector field is
∇ × F = ∇ × z ı̂ı + x ̂ + y 3 z 3 k̂
ı̂ı ̂ k̂
∂ ∂ ∂
= det ∂x ∂y ∂z
z x y3 z3
= 3y 2 z 3 ı̂ı + ̂ + k̂
For every t, we have x(t) = z(t) and x(t)2 + y(t)2 + z(t)2 = 2. So the
specified curve is the intersection of the plane x = z and the sphere x2 +
y 2 + z 2 = 2. This curve is the boundary of the circular disk
D = (x, y, z) x = z, x2 + y 2 + z 2 ≤ 2
√
The curve is oriented so that x(t), y(t) = cos t, 2 sin t runs in the
standard (counterclockwise) direction. So the unit normal to D used in
Stokes’ theorem has positive k̂ component. Since the plane x − z = 0
has unit normal ± √12 (1, 0, −1), the unit normal used in Stokes’ theorem is
n̂ = √12 (−1, 0, 1). By Stokes’ theorem
I ZZ ZZ
1
F · dr = ∇ × F · n̂ dS = √ (3y 2 z 3 , 1, 1) · (−1, 0, 1) dS
C D 2 D
ZZ
1
=√ (1 − 3y 2 z 3 ) dS
2 D
The disk D is invariant under the reflection (x, y, z) → (−x, y, −z). Since
APPENDIX D. SOLUTIONS TO EXERCISES 662
I ZZ
1 1
F · dr = √ dS = √ Area(D)
2 D 2
∇ × F = −ı̂ı + ̂ + k̂
A reasonable guess for the centre of the disk is 13 (1, 1, 1). (This guess is
just based on symmetry.) To check this we just need to observe that it is
indeed on the plane x + y + z = 1 and that the distance from 13 (1, 1, 1) to
any point (x, y, z) obeying x + y + z = 1 and x2 + y 2 + z 2 = 1, namely
r
1 2 1 2 1 2
x− + y− + z−
3 3
r 3 r
2 3 2 1
= x2 + y 2 + z 2 − (x + y + z) + = 1 − +
3 9 3 3
r
2
=
3
q
is the same. This also tells us that D has radius 23 and hence area 23 π.
2π
So the specified line integral is √ .
3 3
(b) We compute
∂r
= cos θ ı̂ı + sin θ ̂ + k̂
∂r
∂r
= −r sin θ ı̂ı + r cos θ ̂
∂θ
∂r ∂r
n̂ dS = ± × drdθ = ± − r cos θ ı̂ı − r sin θ ̂ + r k̂ drdθ
∂r ∂θ
APPENDIX D. SOLUTIONS TO EXERCISES 663
(c) Solution 1 : Let P be the path along line segments from (1, 0, 1) to
(0, 0, 0) and from (0, 0, 0) to (−1, 0, 1). Here is a sketch. P is in blue.
(−1, 0, 1)
z
(1, 0, 1) p
z= x2 + y 2
x
Then Z Z ZZ
F · dr + F · dr = ∇ × F · n̂ dS
C P S
(1, 0, 1) p
z= x2 + y 2
x
Then Z Z ZZ
F · dr + F · dr = ∇ × F · (−k̂) dS
C L R
by Stokes’ Theorem. Along L, the vector R field F = ̂ is orthogonal to the
curve (which has direction −ı̂ı so that L F · dr = 0. Note that ∇ × F is
the vector field v from part (b). Thus
Z ZZ ZZ ZZ
F · dr = − v · k̂ dS = 2z dS = 2 dS = 2 Area(R) = π
C R R R
APPENDIX D. SOLUTIONS TO EXERCISES 664
As
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det ∂x = −2(ı̂ı + ̂ + k̂)
∂y ∂z
y−z z−x x−y
and, on S 0 , n̂ = √1 (ı̂ı
3
+ ̂ + k̂)
ZZ ZZ √ √
∇ × F) · n̂ dS =
(∇ − 2 3 dS = −2 3 × Area(S 0 )
S0 S0
every sphere centred on the origin. We also saw, in Example 4.2.7, that the
divergence ∇ · FS = 0 everywhere except at the origin (where it is not defined).
So if we choose m to be a very big negative number (say −10100 ) and add in
a very small vector field with positive divergence (say 10−100 (xı̂ı + ŷ + z k̂)),
APPENDIX D. SOLUTIONS TO EXERCISES 665
we will get the vector field F = −10100 |r|r 3 + 10−100 (xı̂ı + ŷ + z k̂) which has
divergence ∇ · F = 3 × 10−100 > 0 everywhere except at the origin. The flux
of this field through the specified sphere will be −4π × 10100 plus a very small
positive number.
(e) True. The statement that “the flux out of one hemisphere is equal to
the flux into the opposite hemisphere” is equivalent to the statement that “the
flux out of the sphere is equal to zero”. Since ∇ · F = 0 everywhere, that is
true by the divergence theorem.
(f) That depends.
T̂
If κ = 0, then dds dr
= 0, so that ds = T̂ is a constant. So r(s) = sT̂ + r(0)
is part of a straight line.
If κ > 0, then, because the curve is in a plane, the torsion τ is zero and the
Frenet-Serret formulae reduce to
dT̂ dN̂
= κN̂ = −κT̂
ds ds
dc dr 1 dN̂ 1
= + = T̂(s) + − κT̂(s) = 0
ds ds κ ds κ
c(s) is a constant and
1
|r(s) − c| =
κ
which says that the curve is part of the circle of radius κ1 centred on c.
(g) False. We saw in Examples 2.3.14 and 4.3.8 that the given vector field
is not conservative. H H
(h) False. For example, if P = −y, then C F · dr = − C y dx is the area
inside C. See Corollary 4.3.5.
(i) False.
If dv
dt = a is a constant, then v(t) = a t + v0 . Integrating a second time,
r(t) = 21 a t2 + v0 t + r0 . This is not a spiral, whether or not the speed is
constant. (In fact, for the speed |v(t)| = |a t + v0 | to be constant, a has to be
0, so that r(t) = v0 t + r0 is a straight line.)
Another way to come to the same conclusion uses
d2 s ds 2
a(t) = (t) T̂(t) + κ(t) (t) N̂(t)
dt2 dt
ds
As the speed dt is a constant, it reduces to
ds 2
a(t) = κ(t) (t) N̂(t)
dt
As a(t) is a constant, its direction, N̂(t), is also a constant. The normal vector
to a spiral is not constant.
5.2.2. ∗. Solution. (a) False. For any constant vector a = (a1 , a2 , a3 ),
ı̂ı ̂ k̂
a × r = det a1 a2 a3 = (a2 z − a3 y)ı̂ı − (a1 z − a3 x)̂ + (a1 y − a2 x)̂
x y z
APPENDIX D. SOLUTIONS TO EXERCISES 666
So
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (a × r) = det
∂x ∂y ∂z
a2 z − a3 y −a1 z + a3 x a1 y − a2 x
= 2a1ı̂ı + 2a2̂ + 2a3 k̂
∇x2 ) = ∇ · (2xı̂ı) = 2
∇f ) = ∇ · (∇
∇ · (∇
(d) False. The trap here is that F need not be defined at the origin. We
mr
saw, in Example 3.4.2, that the point source F = |r| 3 had flux 4πm through
every sphere centred on the origin. We also saw, in Example 4.2.7, that the
divergence ∇ · F = 0 everywhere except at the origin (where it is not defined).
(e) True. Any simple, smooth, closed curve in R3 that avoids the origin
is the boundary of a surface S that also avoids the origin. Then, by Stokes’
theorem, I ZZ
F · dr = ∇ × F · n̂ dS = 0
C S
True. Let S =
(f) r |r − c| = R be a sphere. Denote by V =
r |r − c| ≤ R the ball whose boundary is S. Let H be one hemisphere of
S with outward pointing normal and let H 0 be the other hemisphere of S with
inward point normal. Then the boundary of V , with outward pointing normal,
can be viewed as consisting of two parts, namely H and −H 0 , where by −H 0 we
mean H 0 but with outward pointing normal. Then, by the divergence theorem
ZZ ZZ ZZ
F · n̂ dS − F · n̂ dS = F · n̂ dS
H H0
Z Z∂V
Z
= ∇ · F dV > 0
V
RR RR
which implies that H F · n̂ dS > H 0 F · n̂ dS.
(g) False. The trap here is that the curve is in R3 , not R2 . As we saw in
Example 1.4.4, a helix has constant curvature, but does not lie in a plane and
so is not part of a circle.
(h) False. Even if we restrict F to the xy-plane (i.e. to z = 0), this vector
field is not conservative. We saw that in Examples 2.3.14 and 4.3.8.
(i) False. For example, the vector field F = x k̂ is always parallel to the
z-axis. So its flow lines are also all parallel to the z-axis. But if the closed
curve C consists of the line segments
• L1 from (0, 0, 0) to (1, 0, 0), followed by
• L2 from (1, 0, 0) to (1, 0, 1), followed by
• L3 from (1, 0, 1) to (0, 0, 1), followed by
z
L3
(1, 0, 1)
L4
L2
y
L1
x
then
R R1
• L1
F · dr = 0
(xk̂) · ı̂ı dx = 0 since k̂ ⊥ ı̂ı, dr = ı̂ı dx on L1 and
R R1
• L2
F · dr = 0
(1k̂) · k̂ dz = 1 since x = 1 and dr = k̂ dz on L2 and
R R1
• L3
F · dr = − 0
(xk̂) · ı̂ı dx = 0 since k̂ ⊥ ı̂ı and
R R1
• L4
F · dr = − 0
(0k̂) · k̂ dz = 0 since x = 0 on L4 .
All together
Z Z Z Z Z
F · dr = F · dr + F · dr + F · dr + F · dr = 1
C L1 L2 L3 L4
d 2 d
0= |v| = (v · v) = 2v · a
dt dt
5.2.3. ∗. Solution. (a) False. r00 (t) is the full acceleration. So |r00 (t)| is
the magnitude of the full acceleration, not just the tangential component of
acceleration. For example, if r(t) = cos tı̂ı + sin t ̂ (i.e. the particle is just going
around in circles), the acceleration r00 (t) = − cos tı̂ı − sin t ̂ is perpendicular to
the direction of motion. So the tangential component of acceleration is zero,
while |r00 (t)| = 1.
(b) T̂(t) is the tangent vector to the curve at r(t). N̂(t) and B̂(t) are both
perpendicular to T̂(t) (and to each other) and so span the plane normal to the
curve at r(t).
(c) True. This is (half of) Theorem 2.4.8.
(d) False. The statement ∇ × (∇ ∇ · F ) = 0 is just plain gibberish, because
∇ · F is a scalar valued function and there is no such thing as the curl of a
scalar valued function.
(e) False. For example if F = ı̂ı, then, by the divergence theorem,
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 0
S V
Here V = x, y, z x2 + y 2 + z 2 ≤ 1 is the inside of the sphere.
(f) True. If S is the boundary of the solid region E, then we can orient S
by always choosing the normal vector that points into E.
5.2.4. ∗. Solution. (a) The helix is approximately a bunch of circles stacked
one on top of each other. The radius of the circles increase as z increases. So
the curvature decreases as z increases.
(b) Here are two arguments both of which conclude that f (x) is D.
• If C were the graph y = f (x), then f 0 (x) would have two points of
discontinuity. The curvature κ(x) would not the defined at those two
APPENDIX D. SOLUTIONS TO EXERCISES 668
(d) We are told that the helical ramp starts starts with the y-axis when
z = 0.
• In the cases of parametrisations (a) and (c), z = 0 forces u = 0 and u = 0
forces x = y = 0. That is only the origin, not the y-axis. So we can rule
out (a) and(c).
• In the case of parametrisation (b), z = 0 forces v = 0 and v = 0 forces
y = 0 and x = u. As u varies that sweeps out the x-axis, not the y-axis.
So we can rule out (b).
• In the case of parametrisation (d), z = 0 forces v = 0 and v = 0 forces
x = 0 and y = u. As u varies that sweeps out the y-axis, which is what
we want.
Furthermore
• we are told that z = v runs from 0 to 5 and that
• x2 + y 2 = u2 ≥ 4
So we want parametrisation (d) with domain |u| ≥ 2, 0 ≤ v ≤ 5.
(e) Straight lines have curvature 0. So one acceptable parametrized curve
is r(t) = tı̂ı, 0 ≤ t ≤ 1.
(f) The cube S has six sides. So the outward flux through ∂S is 6 and, by
the divergence theorem,
ZZ ZZZ ZZZ
6= F · n̂ dS = ∇ · F dV = C dV = C
∂S S S
r(3) is 3 and the arclength between r(0) and r(5), which is the same as the
arclength between r(0) and r(3) plus the arclength between r(3) and r(5), is
5. So the arclength between r(3) and r(5) is 5 − 3 = 2.
(i) In this solution, we’ll use, for example −T to refer to the curve T , but
with the arrow pointing in the opposite direction to that of the arrow on T .
In parts (2), (3) and (4) we will choose F to be the vector field
y x
G(x, y) = − ı̂ı + 2 ̂
x2 + y 2 x + y2
R U
S
R1
T
R2
P
C
By Stokes’ theorem
ZZ I
∇ × F · k̂ dS = F · dr
S C
H
• The contribution to C F · dr coming from the left and right sides of C
will be zero, because F is perpendicular to dr there.
H
• The contribution to C F · dr coming from the top of C will be negative,
because there F is a positive number times ı̂ı and dr is a negative number
times ı̂ı.
H
• The contribution to C F·dr coming from the bottom of C will be positive,
because there F is a positive number times ı̂ı and dr is a positive number
times ı̂ı.
• The magnitude of the contribution from the top of C will be larger than
the magnitude of the contribution from the bottom of C, because |F| is
larger on the top than on the bottom.
APPENDIX D. SOLUTIONS TO EXERCISES 671
H
So, all together, C F · dr < 0, and consequently (taking a limit as the square
size tends to zero) ∇ × F · k̂ is negative at P .
5.2.5. ∗. Solution. (a) False. We could have, for example, ∇ · F zero
at one point and strictly positive elsewhere. One example would be F =
x3 ı̂ı + y 3 ̂ + z 3 k̂, with S1 and S2 being the upward oriented top and bottom
hemispheres, respectively, of the unit sphere x2 + y 2 + z 2 = 1.
(b) False. The conditions that (1) ∇ × F = 0 and (2) the domain of
F is simply-connected, are sufficient, but not necessary, to imply that F is
conservative. For example the vector field F = 0, with any domain at all,
is conservative with potential 0. Another example (which does not depend
on choosing a domain that is smaller than the largest possible domain) is
1
F = ∇ x2 +y 2 with domain (x, y, z) (x, y) 6= (0, 0) . That is, the domain is
R3 with the z-axis removed.
(c) That’s true. Consider any point r(t0 ) on a parametrized curve r(t).
That’s the blue point in the figure below.
r(t0 )
T̂
N̂
c
The centre of curvature for the curve at r(t0 ) is c = r(t0 ) + ρ(t0 )N̂(t0 ). It
is the red dot in the figure.
• The radius of the osculating circle is the distance from its centre, c, to
any point of the circle, like r(t0 ). That’s |r(t0 ) − c| = |ρ(t0 )N̂(t0 )| =
ρ(t0 ). The curvature of the osculating circle is one over its radius. So its
curvature is ρ(t10 ) = κ(t0 ).
• The unit normal to the osculating circle at r(t0 ) is a unit vector in the
opposite direction to the radius vector from the centre c to r(t0 ). The
radius vector is r(t0 ) − c0 = −ρ(t0 )N̂(t0 ), so the unit normal is N̂(t0 ).
• The osculating circle lies in the plane that best fits the curve near r(t0 ).
(See the beginning of §1.4.) So the unit tangents to the osculating circle
at r(t0 ) are perpendicular to both N̂(t0 ) and B̂(t0 ) and so are either
T̂(t0 ) or −T̂(t0 ), depending on how we orient the osculating circle.
(d) False. Kepler’s third law is that a planet orbiting a sun has the square
of the period proportional to the cube of the major axis of the orbit.
(e) True. That’s part (a) of Theorem 4.1.7.
(f) True. Every domain contains closed surfaces. This has nothing to do
with vector fields.
(g) True. We saw this in Example 2.3.4.
(h) False. Let F be an everywhere defined conservative vector field with
potential ϕ. Then ∇ × F = 0 everywhere. If P and Q are R two points and if
ϕ(P ) − ϕ(Q) = 3 and if C is a curve from Q to P , then C F · dr = 3. One
example would be ϕ(x, y, z) = x, F = ı̂ı, P = (3, 0, 0), Q = (0, 0, 0).
(i) False. The normal component of acceleration depends on speed, as well
as curvature.
APPENDIX D. SOLUTIONS TO EXERCISES 672
(j) False. The curve r1 contains only points in the xy-plane. Every r2 (t)
with t 6= 0 has a nonzero z-coordinate.
5.2.6. ∗. Solution. (a) False. Changing the orientation of a surface does
not change dS at all. (It changes n̂dS by a factor of (−1).) So
ZZ ZZ
f dS = + f dS
S −S
RR
which is not − −S f dS, unless the integral is zero.
(b) False. For every vector field with two continuous partial derivatives,
∇ × F) = 0 (see Theorem 4.1.7.a), so the divergence theorem gives
∇ · (∇
ZZ ZZZ
∇ × F) · n̂ dS =
(∇ ∇ × F) dV = 0
∇ · (∇
S V
(d) True. The left hand side, (∇∇f ) × (∇∇f ), is zero because (∇
∇f ) is parallel
to itself and the right hand side ∇ × (∇ ∇f ) is zero by Theorem 4.1.7.b (the
screening test for conservative fields).
(e) True. The curve r(t) = 2 , 0 , 1 + t3 4 , −1 , −2 is a straight line.
Straight lines have curvature 0.
(f) True. In general |r0 (t)| = ds
dt . Under arc length parametrization t = s
ds
so that dt = 1.
(g) True. If F is a constant vector fleld, then, by the divergence thoerem,
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = 0 dV = 0
S V V
(e) Along the curve C2 the magnitude R of the angle between F and dr is
greater than 90◦ , so that F · dr < 0 and C2 F · dr < 0.
R R
(f) If F were conservative, we would have C1 F·dr = C2 F·dr. As these two
integrals have opposite signs F is not conservative. (Since F is not conservative,
it is not the gradient of some function. At A, Px > 0 and Qy > 0. So F is not
divergence free and is not the curl of a vector potential.)
5.2.8. ∗. Solution. (a) False. The curve r1 contains only points with z ≥ 0.
Every r2 (t) with t < 0 has z < 0.
(b) True. r2 (t2 ) = r1 (t) and t2 runs from 0 to 1 as t runs from 0 to 1.
(c) True. In general |r0 (t)| = ds ds
dt . When t = s, dt = 1.
APPENDIX D. SOLUTIONS TO EXERCISES 673
(d) False. The curve need not even lie in a plane. For example, as we saw in
Example 1.4.4, the helix r(t) = a cos tı̂ı + b sin t ̂ + bt k̂ has constant curvature
a
κ = a2 +b 2 but is not a circle.
√
(e) True. If the speed |v| = v · v of a moving object is constant, then
d
0= v · v = 2v · a
dt
(x, y, z) x2 + y 2 > 1
provides a counterexample.
(e) False. The curve r1 contains only points with z ≥ 0. Every r2 (t) with
t < 0 has z < 0.
(f) True. r2 (t2 ) = r1 (t) and t2 runs from 0 to 1 as t runs from 0 to 1.
(g) True. ∇ · (∇∇ × F) = 0 by the vector identity of Theorem 4.1.7.a.
(h) False. A counterexample is f (x, y, z) = x2 . It has ∇ f = 2xı̂ı and hence
∇f ) = 2.
∇ · (∇
False.
(i) The curve x 2 + y 2 = 2 can not be shrunk to a point continuously
in (x, y) x + y 2 > 1 .
2
(j) True. Any curve in (x, y) y > x2 can be shrunk to a point
continuously in (x, y) y > x2 .
5.2.10. ∗. Solution. (a)R False. ∇ f = 0 if and only if f is constant. But if
f is the constant K, then C f ds is K times the length of C, which need not
be zero.
(b) False. Any curve which lies in a plane has constant binormal. For
example, the circle r(t) = cos tı̂ı + sin t ̂ + 0 k̂ has constant binormal B̂ = k̂,
but is not a straight line.
2
(c) True. If r(t) has constant speed, the ds dt (t) = r0 (t) · r0 (t) is constant
APPENDIX D. SOLUTIONS TO EXERCISES 674
and
d 0
r (t) · r0 (t) = 2r0 (t) · r00 (t)
0=
dt
R
(d) False. For the line integral C F × G · dr to be independent of the
path C, the vector field F × G has to be conservative and so has to obey
∇ × (F × G) = 0. But
• Not all vector fields are conservative. For example, the vector field H =
x ̂ obeys ∇ × H = k̂ and so is not conservative.
• We can make F×G be any vector field through judicious choices of F and
G. For example, if F = x k̂ and G = ı̂ı, then F × G = x k̂ × ı̂ı = x ̂ = H.
R
(e) True. The contribution to C f ds from an “infinitesmal piece of C” is
the value of f on the piece times the length of the piece. That does not depend
on the orientation of the piece.
∂r ∂r
(f) False. The two vectors in the cross product ∂u × ∂u are identical. So
the cross product is 0.
(g) False. The integral is completely independent of x(u, v) and y(u, v).
In particular if, for example, x(u, v) = 157u, y(u, v) = 157v, z(u, v) = 0 then
RR ∂z 2
1/2
∂z 2
D
1 + ∂u + ∂v dudv is always exactly the area of D, while the
area of S is 1572 times the area of D.
(h) True. If the fluid is incompressible then its flow preserves volumes and
consequently ∇ · F = 0.
(i) Not only False, but Ridiculous. The left had side is scalar valued while
the right hand side is vector valued.
5.2.11. ∗. Solution. (a) True. That ∇ · (∇ ∇ × F) = 0 is the vector identity
of Theorem 4.1.7.a. That identity is the basis of the vector potential screening
test. R
(b) False. If F is not conservative, then C F · dr will depend on the end-
points of C.
(c) True. If ∇ f = 0, then
∂f
(x, y, z) = 0 =⇒ f (x, y, z) = g(y, z)
∂x
∂f ∂g
(x, y, z) = 0 =⇒ (y, z) = 0 =⇒ g(y, z) = h(z)
∂y ∂y
∂f
(x, y, z) = 0 =⇒ h0 (z) = 0 =⇒ h(z) = C
∂z
for some functions g(y, z), h(z) and constant C.
(d) False. The curl ∇ × F is zero for every conservative vector fields F.
There are many nonconstant conservative vector fields, like F(x, y, z) = xı̂ı.
(e) True. As S is closed, it is the boundary of a solid region V . Then, by
the divergence theorem,
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 0
S V
R
(f) True. If C F · dr = 0 for every closed curve C, then F is conservative
by Theorem 2.4.7. Consequently, ∇ × F = 0 by Theorem 2.3.9.
(g) True. If the speed |v| is constant then
d 2 d
0= |v| = (v · v) = 2v · a
dt dt
APPENDIX D. SOLUTIONS TO EXERCISES 675
v
Since T̂ = |v| , T̂ · a = 0 too. Here, we have assumed that the constant |v| is
not zero. If the constant |v| is zero, then T̂ is not defined at all (and a = 0).
(h) False. The trap here is that the curve is in R3 , not R2 . As we saw in
Example 1.4.4, a helix has constant curvature, but does not lie in a plane and
so is not part of a circle.
(i) False. The trap here is that we are told nothing about ∇ · F. As an
example, let S1 be the hemisphere
S1 = (x, y, z) x2 + y 2 + z 2 = 1, z ≥ 0
S2 = (x, y, 0) x2 + y 2 ≤ 1
V n̂ S1
S2
Set p
(x, y, z) 0 ≤ z ≤ x2 + y 2 , x2 + y 2 ≤ 1
V =
Then the boundary, ∂V , of V consists of two parts, namely S1 (with nor-
mal pointing upwards) and S2 (but with normal pointing downwards). The
divergence theorem (Theorem 4.2.2) gives
ZZ ZZ ZZZ
F · n̂ dS − F · n̂ dS = ∇ · F dV
S1 S2 V
RR RR
If ∇ · F > 0 (as is the case, for example, if F = xı̂ı) then S1
F · n̂ dS − S2
F·
n̂ dS is definitely nonzero.
(j) True. This is one of Kepler’s laws. See §1.9.
5.2.12. ∗. Solution. It’s (b). (a) is gibberish — the left hand side is a scalar
while the right hand side is a vector. (c) is also gibberish — the left hand side
is a vector while the right hand side is a scalar. (b) is the vector identity of
Theorem 4.1.4.c.
5.2.13. ∗. Solution. (a) False. For example, if f (x, y, z) = x2 , then ∇ f =
2xı̂ı and ∇ · ∇ f = 2.
(b) Not only false, but ridiculous. The left hand side is a vector while the
right hand side is a scalar.
(c) Not only false, but ridiculous. The right hand side is a vector while the
left hand side is a scalar.
(d) True. That’s the screening test for conservative fields, Theorem 4.1.7.b.
(e) Not only false, but ridiculous. The curl of a scalar function is not
defined.
(f) True. That’s the screening test for vector potentials, Theorem 4.1.7.a.
(g) False.
r ∂ x ∂ y ∂ z
∇· = + +
|r|2 ∂x x2 + y 2 + z 2 ∂y x2 + y 2 + z 2 ∂z x2 + y 2 + z 2
APPENDIX D. SOLUTIONS TO EXERCISES 676
1 2x2 1 2y 2
= − 2 + − 2
x2 + y 2 + z 2 [x2 + y 2 + z 2 ] x2 + y 2 + z 2 [x2 + y 2 + z 2 ]
1 2z 2
+ 2 − 2
x + y2 + z2 [x2 + y 2 + z 2 ]
3[x2 + y 2 + z 2 ] − 2x2 − 2y 2 − 2z 2
= 2
[x2 + y 2 + z 2 ]
1
=
x2 + y 2 + z 2
So
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (ω
ω × r) = det
∂x ∂y ∂z
ω2 z − ω3 y −ω1 z + ω3 x ω1 y − ω2 x
= 2ω1ı̂ı + 2ω2̂ + 2ω3 k̂
Ω = (x, y, z) x2 + y 2 + z 2 ≤ 1
f (x) = x
Then
n̂
ZZ z ZZ }| { ZZ
f n̂ dS = x(xı̂ı + y ̂ + z k̂) dS = ı̂ı x2 dS
∂Ω ∂Ω ∂Ω
ZZZ ZZZ ZZZ
− ∇ f dV = − ı̂ı dV = −ı̂ı dV
Ω Ω Ω
The coefficient of ı̂ı is obviously strictly positive in the upper integral and
strictly negative in the lower integral.
5.2.14. ∗. Solution. (a) True. If the vector field is F = aı̂ı + b ̂ + c k̂, then
f (x, y, z) = ax + by + cz obeys F = ∇ f and so is a potential for F.
(b) False. For example the vector field F = xı̂ı − y ̂ obeys ∇ · F = 0 but is
APPENDIX D. SOLUTIONS TO EXERCISES 677
∇f ) = (∇
∇ × (f∇ ∇f ) × (∇
∇f ) + f ∇ × (∇
∇f ) = 0
The second term vanished because of the screening test vector identity of The-
orem 4.1.7.b.
(b) True. That’s the vector identity of Theorem 4.1.4.c.
(c) True. To have constant curvature 0 the curve must have unit tangent
vector T̂(s) obeying
dT̂
(s) = 0
ds
(See §1.5.) So r0 (s) = T̂(s) must be a constant vector. Call it T̂0 . Integrating
gives
r(s) = sT̂0 + r0
for some constant vector r0 . So r(s) lies on the same straight line for all s.
(d) False. The trap here is that the curve is in R3 , not R2 . As we saw in
Example 1.4.4, a helix has constant curvature, but does not lie in a plane and
so is not part of a circle.
(e) True. The vector field F = ∇ f is conservative. So, by Theorem 2.4.7.b,
the work integral Z Z
∇ f · dr = F · dr = 0
C C
for any closed curve C, and, in particular, for any circle C.
(f) True. The statement that “the flux out of one hemisphere is equal to
the flux into the opposite hemisphere” is equivalent to the statement that “the
flux out of the sphere is equal to zero”. Since ∇ · F = 0 everywhere, that is
true by the divergence theorem.
(g) True. Let S be the boundary of the solid region V . Then, by the
divergence theorem (Theorem 4.2.9),
ZZ ZZZ
∇ × F · n̂ dS = ∇ · ∇ × F dV
S V
But ∇ · ∇ × F is identically zero, by the screening test vector identity of
Theorem 4.1.7.a. So the integral is zero.
5.2.17. ∗. Solution. (a) True. Let F be the vector field. We are assuming
that ∇ × F = 0 on all of R3 . As a result, F = ∇φ for some potential function
∂2 ∂2 ∂2
φ. We are also assuming that 0 = ∇ · F = ∇ · ∇φ = ∂x 2 + ∂y 2 + ∂z 2 φ. This
2
∂ ∂2 ∂2
If, for example, φ = x2 , then ∂x 2 + ∂y 2 + ∂z 2 φ = 2 and the flux of F through
5.2.18. ∗. Solution. (a) True. The vector field ∇ f is conservative and the
work done by a conservative field around any closed curve is zero.
(b) False. By the vector identity Theorem 4.1.7.a, we have
∇ × F) = 0
∇ · (∇
d
v(t) × r(t) = v0 (t) × r(t) + v(t) × v(t)
dt
The first term vanishes because v0 (t) = a(t) = f (t)r(t) is parallel to r(t). The
second term vanishes because v(t) = v(t).
(c) Call the constant vector v × r of part (b) N. This vector is a constant
and is perpendicular to both v(t) and r(t). In particular
N · r(t) = 0
Assuming that N is nonzero, this is the equation of the plane through the
origin with normal vector N.
(d) Yes, as long as T̂, N̂, and B̂ are well-defined, since B̂ = T̂ × N̂.
2 2
(e) No. When the maximum speed occurs ddt2s = 0 so that a = κ(t) ds dt (t) N̂(t).
If the speed and (constant) curvature are nonzero, the acceleration is nonzero.
5.2.20. ∗. Solution. We apply Green’s Theorem:
Z ZZ
∂F2 ∂F1
F1 dx + F2 dy = − dxdy
C R ∂x ∂y
Z ZZ
1 1
(a) −y dx + x dy = 1 − (−1) dx dy = Area(R)
2 ZC 2 Z ZR
1 1
(b) −x dx + y dy = 0 dx dy = 0 6= Area(R)
2
Z C ZZ 2 R
(c) y dx = − 1 dx dy = −Area(R) 6= Area(R)
ZC R ZZ
(d) 3y dx + 4x dy = 4 − 3 dx dy = Area(R)
C R
dv
a= T̂ + v 2 κN̂ = 0T̂ + κN̂.
dt
|v × a| 1
κ= = = 1.
|v|3 1
APPENDIX D. SOLUTIONS TO EXERCISES 680
The solid R clearly has reflection symmetry across the plane z = 2. So the
z-coordinate of the centre of mass of R, i.e. the average value of z over R, i.e.
RRR RRR
R
z dV R
z dV
z̄ = RRR =
R
dV Vol(R)
is 2. Hence
ZZ
(xı̂ı − y ̂ + z 2 k̂) · n̂dS = 2z̄ Vol(R) = 4 Vol(R)
S
for all disks D. Because this is true for all disks D, the integrand must be
zero. To see this, let H = ∇ × F − G. Suppose that H(x0 ) 6= 0. Pick a unit
vector n̂ in the direction of H(x0 ). Let D be a very small flat disk centered
on x0 with normal n̂ (the vector we picked). Then H(x0 ) · n̂ > 0 and, by
RR H(x) · n̂ >0 for all x on D, if we have picked D small enough.
continuity,
Then D ∇ × F − G · n̂ dS > 0, which is a contradiction. So we conclude
that ∇ × F − G = 0 and hence G = ∇ × F.