1348 Chapter Twenty /SOLUTIONS

(d) The vector field looks like the following:

Figure 20.11

(e) We need to show that grad φ = ~v .

∂ K ∂ K
grad φ = ( log(x2 + y 2 ))~i + ( log(x2 + y 2 ))~j
∂x 2 ∂y 2
Kx ~ Ky ~
= 2 i + 2 j
x + y2 x + y2
= K(x2 + y 2 )−1 (x~i + y~j )
= ~v

34. The chain rule for partial differentiation of the formulas u = F cos φ and v = F sin φ gives

ux = (cos φ)Fx − F (sin φ)φx

vy = (sin φ)Fy + F (cos φ)φy .

We have
~ = ux + vy
div F
= ((cos φ)Fx − F (sin φ)φx ) + ((sin φ)Fy + F (cos φ)φy )
1
= (−vφx + uφy ) + (uFx + vFy )
F
1 ~
= grad φ · (−v~i + u~j ) + grad F · F
F
~ is given by φ ~ = grad φ· N
The definition of directional derivative tells us that the derivative of φ in the direction of N ~ .
N
~ ~ ~ ~ ~
Similarly FT~ = grad F · T . Since T = (1/F )F and N = (1/F )(−v i + uj ), we have ~

~ = grad φ · (−v~i + u~j ) + 1 ~

div F grad F · F
F
1 1 ~
= F grad φ · (−v~i + u~j ) + grad F · F
F F
~ + grad F · T~
= F grad φ · N
= F φN~ + FT~ .

Solutions for Section 20.2

Exercises

1. No, because the surface S is not a closed surface.

 20.2 SOLUTIONS 1349

~ = dA~i .
2. First directly: On the faces x = 0, y = 0, z = 0, the flux is zero. On the face x = 2, a unit normal is ~i and dA
So Z Z
~ =
~r · dA (2~i + y~j + z~k ) · (dA~i )
Sx=2 Sx=2

(since on that face, x = 2) Z

= 2dA = 2 · (Area of face) = 2 · 4 = 8.
Sx=2

In exactly the same way, you get Z Z

~ =
~r · dA ~ = 8,
~r · dA
Sy=2 Sz=2
so Z
−
→
~r · dA = 3 · 8 = 24.
S
Now using divergence:
~ = ∂x + ∂y + ∂z = 3,
div F
∂x ∂y ∂z
so Z Z Z
2 2 2
Flux = 3 dx dy dz = 3 · (Volume of Cube) = 3 · 8 = 24
0 0 0

3. Finding flux directly:

1) On bottom face, z = 0 so F ~ = x2~i + 2y 2~j is parallel to face so flux is zero.
2) On front face, y = 0 so F = x2~i + 3z 2~k is parallel to face so flux is zero.
~
3) On back face, y = 1 so F ~ = x2~i + 2~j + 3z 2~k and A ~ = ~j so flux is 2.
4) On top face, z = 1 so F~ = x2~i + 2y 2~j + 3~k and A ~ = ~k so flux is 3.
5) On side x = 1, F ~ = ~i + 2y 2~j + 3z 2~k and A
~ = −~i so flux is −1.
~ ~ 2~ 2~ ~ = ~i so flux is 4.
6) On side x = 2, F = 4i + 2y j + 3z k and A
Total flux is thus 8.
By the Divergence Theorem:
~ = 2x + 4y + 6z
div F
So, if W is the interior of the box, we have
Z Z Z 2 Z 1 Z 1
~ · dA
F ~ = (2x + 4y + 6z)dV = 2 (x + 2y + 3z)dz dy dx
S W 1 0 0
Z Z  1 Z Z !
2 1 2 1
3z 2 3
=2 xz + 2yz + dy dx = 2 x + 2y + dy dx
1 0
2 0 1 0
2
Z 2 h i1 Z 2  
3y 3
=2 xy + y 2 + dx = 2 x+1+ dx
1
2 0 1
2
Z 2 2

= (2x + 5)dx = (x2 + 5x) =8

1 1

4. The location of the pyramid has not been completely specified. For instance, where is it centered on the xy plane? How
is base oriented with respect to the axes? Thus, we cannot compute the flux by direct integration with the information we
have. However, we can calculate it using the divergence theorem. First we calculate the divergence of F ~.

~ = ∂(−z) + ∂0 + ∂x = 0 + 0 + 0 = 0
div F
∂x ∂y ∂z
Thus for any closed surface the flux will be zero, so the flux through our pyramid, regardless of its location or orientation,
is zero.
1350 Chapter Twenty /SOLUTIONS

5. First directly, since the vector field is totally in the ~j direction, there is no flux through the ends. On the side of the
cylinder, a normal vector at (x, y, z) is x~i + y~j . This is in fact a unit normal, since x2 + y 2 = 1 (the cylinder has radius
1). Also, using x = cos θ, y = sin θ, in this case, the element of area dA equals 1dθdz. So
Z Z 2 Z 2π
Flux = ~ · dA
F ~ = (y~j ) · (x~i + y~j ) dθ dz
0 0
Z 2 Z 2π Z 2 Z 2π Z 2
= y 2 dθ dz = sin2 θ dθ dz = πdz = 2π.
0 0 0 0 0

Now we calculate the flux using the divergence theorem. The divergence of the field is given by the sum of the
∂y
respective partials of the components, so the divergence is simply = 1. Since the divergence is constant, we can
∂y
simply calculate the volume of the cylinder and multiply by the divergence

Flux = 1πr 2 h = 2π

R
6. First compute divF ~ = x2~i + (y − 2xy)~j + 10z~k . By the Divergence Theorem,
~ = 11 at origin, where F ~ ·
F
R sphere
dA~ = 11dV = 11(volume of R) = 11(4π53 /3) = 5500π/3 ≈ 5759.6
R
7. The divergence of the field is
~ = ∂(−z) + ∂(0) + ∂x = 0.
div F
∂x ∂y ∂z
Hence, Z Z Z
~ · dA
F ~ = ~ dV =
div F 0 dV = 0,
Sphere Ball Ball
and so the flux through the sphere is zero. This makes sense, because, as Figure 20.12 shows, the vector field is flowing
around the y-axis and is always tangent to the sphere.

Figure 20.12

8. Since the surface is closed, the flux of a constant vector field out of it is 0.
~ = 1, if W is the interior of the box, the Divergence Theorem gives
9. Since div G
Z
Flux = 1 dV = 1 · Volume of box = 1 · 2 · 3 · 4 = 24.
W

~ = y, if W is the interior of the box, the Divergence Theorem gives

10. Since div H
Z Z 4 Z 3 Z 2 3
y2
Flux = y dV = y dx dy dz = 4 · 2 · = 36.
W 0 0 0
2 0
 20.2 SOLUTIONS 1351

11. Since div J~ = 2xy, if W is the interior of the box, the Divergence Theorem gives
Z Z 4 Z 3 Z 2 2 3 4
y2
Flux = 2xy dV = 2xy dx dy dz = x2 z = 72.
W 0 0 0 0
2 0 0

~ = 0, the flux through the closed surface of the box is 0.

12. Since div N

Problems
R R R
13. By the Divergence Theorem, ~ · dA
F ~ = ~ dV =
divF 0dV = 0 for a closed surface S, where W is the region
S W W
enclosed by S.
14. Since
~ = 0 + 1 + 0 = 1,
div F
we have Z
~ · dA
F ~ ) · Vol = 4 π33 = 36π.
~ = (div F
S
3

~ = y + z + x, the flux is given by

15. Since div F
Z Z
~ · dA
F ~ = (x + y + z)dV.
Sphere Sphere

We calculate the first term of the integral

Z Z 1 Z √1−z2 Z √1−y2 −z2
x dV = √ √ x dx dy dz
Sphere −1 − 1−z 2 − 1−y 2 −z 2
Z 1 Z √1−z2 √
1−y 2 −z 2
x2
= √ √ dy dz
−1 − 1−z 2
2 − 1−y 2 −z 2

Z Z √ p p !
1 1−z 2
( 1 − y 2 − z 2 )2 (− 1 − y 2 − z 2 )2
= √ − dy dz
−1 − 1−z 2
2 2
Z 1 Z √1−z2
= √ 0 dydz = 0.
−1 − 1−z 2

The other terms in the integral are zero by a similar calculation.

The same result can be obtained by a symmetry argument, which is much shorter: Since x, y, and z each take equal
positive and negative values on half the sphere, the integral of each term is 0. Thus, the flux is zero:
Z Z
~ · dA
F ~ = (x + y + z)dV = 0.
Sphere Sphere

~ = 3x2 + 3y 2 + 3z 2 , the Divergence Theorem gives

16. Since div F
Z Z Z
~ · dA
F ~ = ~ dV =
div F 3(x2 + y 2 + z 2 ) dV.
S W W

Since W is the interior of a cylinder of radius 2 centered on the z-axis, we use cylindrical coordinates, giving
Z Z Z 2π Z 2 Z 5
~ · dA
F ~ = 2
3(x + y + z ) dV = 3 2 2
(r2 + z 2 )r dz dr dθ
S W 0 0 0
Z 2π Z 2   5 Z 2π Z 2
rz 3 125
=3 r3 z + dr dθ = 3 (5r 3 + r) dr dθ
0 0
3 0 0 0
3
  2
5r4 125 r2
= 3 · 2π + · = 620π.
4 3 2 0
1352 Chapter Twenty /SOLUTIONS

~ = 3x2 + 3y 2 + 3z 2 , the Divergence Theorem gives

17. Since div F
Z Z
Flux = ~ · dA
F ~ = (3x2 + 3y 2 + 3z 2 ) dV.
S W

In spherical coordinates, the region W lies between the spheres ρ = 2 and ρ = 3 and inside the cone φ = π/4. Since
3x2 + 3y 2 + 3z 2 = 3ρ2 , we have
Z Z 2π Z π/4 Z 3
Flux = ~ · dA
F ~ = 3ρ2 · ρ2 sin φ dρ dφ dθ
S 0 0 2
3 π/4 √
3 5 633(2 − 2)
= 2π · ρ (− cos φ) = π = 232.98.
5 2 0
5

18. By the Divergence Theorem, if W is the cylinder and S is its surface:

Z Z Z
~ · dA
F ~ = ~ dV =
div F 10 dV = 10 · Volume of cylinder = 10πa3 .
S W W

19. Apply the Divergence Theorem to the solid cone, whose interior we call W . The surface of W consists of S and D. Thus
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV.
div F
S D W

~ = 0 everywhere, since F
But div F ~ is constant. Thus
Z Z
~ · dA
F ~ =− ~ · dA
F ~ = −3.22.
D S

~ through W , the interior of the box:

20. Using the Divergence theorem, we estimate the integral of div F
Z Z
Flux = ~ · dA
F ~ = ~ dV.
div F
S W

The region W can be divided into four boxes, with sides ∆x = ∆y = 0.5 and ∆z = 1. Since div F ~ is monotonic, on
~
each box the lower bound for div F is at the corner closest to the origin, and the upper bound is at the corner farthest
from the origin. Thus, we use Table 20.1 for the lower bound and Table 20.2 for the upper bound. The volume of each
box is (0.5)(0.5)1 = 0.25. Thus,
Lower bound for flux = 2(0.25) + 3(0.25) + 4(0.25) + 5(0.25) = 3.5.
Upper bound for flux = 6(0.25) + 8(0.25) + 7(0.25) + 9(0.25) = 7.5.
21. We close the cylinder, S, by adding the circular disk, S1 , at the top, z = 3. The surface S + S1 is oriented outward, so S1
is oriented upward. Applying the Divergence Theorem to the closed surface S + S 1 enclosing the region W , we have
Z Z
~ · dA
F ~ = ~ dV
div F
S+S1 W
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV.
div F
S S1
Since
~ = div(z 2~i + x2~j + 5~k ) = 0,
div F
we have Z Z
~ · dA
F ~ =− ~ · dA
F ~.
S S1

Only the ~k -component of F ~ = ~k dx dy on S1 , so

~ contributes to the flux through S1 , and dA
Z Z Z
√
~ · dA
F ~ =− (z 2~i + x2~j + 5~k ) · ~k dx dy = − 5 dx dy = −5 · Area of S1 = −5π( 2)2 = −10π.
S S1 S1
 20.2 SOLUTIONS 1353

22. We close the cylinder, S, by adding the circular disk, S1 , at the top, z = 3. The surface S + S1 is oriented outward, so S1
is oriented upward. Applying the Divergence Theorem to the closed surface S + S 1 enclosing the region W , we have
Z Z
~ · dA
F ~ = ~ dV
div F
S+S1 W
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV.
div F
S S1
Since
~ = div(y 2~i + z 2~j + (x2 + y 2 )~k ) = 0,
div F
we have Z Z
~ · dA
F ~ =− ~ · dA
F ~.
S S1

Only the ~k -component of F ~ = ~k dx dy on S1 , so

~ contributes to the flux through S1 , and dA
Z Z Z
~ · dA
F ~ =− (y 2~i + z 2~j + (x2 + y 2 )~k ) · ~k dx dy = − (x2 + y 2 ) dx dy.
S S1 S1
√
Converting to polar coordinates, since the cylinder has radius 2, we have
√
Z Z 2π Z √
2 2
~ · dA
~ =− 2 r4
F r · r dr dθ = −2π = −2π.
S 0 0
4 0

23. We close the cylinder, S, by adding the circular disk, S1 , at the top, z = 3. The surface S + S1 is oriented outward, so S1
is oriented upward. Applying the Divergence Theorem to the closed surface S + S 1 enclosing the region W , we have
Z Z
~ · dA
F ~ = ~ dV
div F
S+S1 W
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV.
div F
S S1
Since
~ = div(z~i + x~j + y~k ) = 0,
div F
we have Z Z
~ · dA
F ~ =− ~ · dA
F ~.
S S1

Only the ~k -component of F ~ = ~k dx dy on S1 , so

~ contributes to the flux through S1 , and dA
Z Z Z
~ · dA
F ~ =− (z~i + x~j + y~k ) · ~k dx dy = − y dx dy.
S S1 S1

Since y is an odd function, by symmetry, its integral over S1 is zero. Thus,

Z Z
~ · dA
F ~ =− y dx dy = 0.
S S

24. We close the cylinder, S, by adding the circular disk, S1 , at the top, z = 3. The surface S + S1 is oriented outward, so S1
is oriented upward. Applying the Divergence Theorem to the closed surface S + S 1 enclosing the region W , we have
Z Z
~ · dA
F ~ = ~ dV
div F
S+S1 W
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV.
div F
S S1
Since
~ = div(y 2~i + x2~j + 7z~k ) = 7,
div F
1354 Chapter Twenty /SOLUTIONS

we have
Z Z Z Z
~ · dA
F ~ = 7 dV − ~ · dA
F ~ = 7 · Volume of cylinder − ~ · dA
F ~
S W S1 S1
Z Z
√
= 7 · π( 2)2 6 − ~ · dA
F ~ = 84π − ~ · dA
F ~.
S1 S1

Only the ~k -component of F ~ = ~k dx dy and z = 3 on S1 , so

~ contributes to the flux through S1 , and dA
Z Z Z
~ · dA
F ~ = 84π − (y 2~i + x2~j + 7 · 3~k ) · ~k dx dy = 84π − 21 dx dy
S S1 S1
√
= 84π − 21 · Area of S1 = 84π − 21 · π( 2)2 = 42π.

25. We close the cylinder, S, by adding the circular disk, S1 , at the top, z = 3. The surface S + S1 is oriented outward, so S1
is oriented upward. Applying the Divergence Theorem to the closed surface S + S 1 enclosing the region W , we have
Z Z
~ · dA
F ~ = ~ dV
div F
S+S1 W
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV.
div F
S S1
Since
~ = div(x3~i + y 3~j + ~k ) = 3x2 + 3y 2 ,
div F
we have Z Z Z
~ · dA
F ~ = (3x2 + 3y 2 ) dV − ~ · dA
F ~.
S W S1
~ = ~k dx dy,
To find the integral over W , we use cylindrical coordinates. For the integral over S 1 , we use the fact that dA
~
so only the k-component of F contributes to the flux.
Z Z 2π Z 3 Z √
2 Z
~ · dA
F ~ = 3r2 · r dr dz dθ − (x3~i + y 3~j + ~k ) · ~k dx dy
S 0 −3 0 S1
√
2π 3 2 Z
3 4
=θ z r − dx dy
4 S1
0 −3 0
3 √ 4
= 2π · 6 · ( 2) − Area of S1
4
√
= 36π − π( 2)2 = 34π.

26. (a) True. Flux that goes in one face goes out the other, since the vector field is constant and the surface is closed.
(b) True. The flux out of S1 along the face shared with S2 cancels with the flux out of S2 over the same face. (The
normals are in opposite directions.) The other five faces of S1 and the other five faces of S2 are each faces of S.
27. (a) Taking partial derivatives, we have for ~r 6= ~0 ,
 
~ = div x~i + y~j + z~k
div F
(x2 + y 2 + z 2 )3/2
3(x2 + y 2 + z 2 ) 3
=− + 2 = 0.
(x2 + y 2 + z 2 )5/2 (x + y 2 + z 2 )3/2
(b) We could compute the flux out of the box S by computing the flux out of each side separately. However, since
~ = ~0 everywhere except the origin, we instead consider a region W between the box S and a sphere S b of
div F
~ = ~0
radius b centered at the origin and which fits inside the box. If the sphere is oriented inward, since div F
throughout W , the Divergence Theorem says
Z Z Z
0= ~ dV =
div F ~ · dA
F ~ + ~ · dA
F ~.
W S Sb
 20.2 SOLUTIONS 1355

~ through Sb is easier to calculate than the flux through the box. Since Sb is oriented inward,
The flux of F
Z Z Z
~ · dA
~ =− ~ kkdA
~k=− 1
F kF kdAk
Sb Sb Sb
b2
1 1
=− · Surface area of Sb = − 2 · 4πb2 = −4π.
b2 b
Thus, Z Z
~ · dA
F ~ =− ~ · dA
F ~ = 4π.
S Sb

~ in terms of (x, y, z) as
28. We can rewrite F

~ = p −Gmx ~i + p −Gmy ~j + p −Gmz ~k

F
(x2 + y2 + z 2 )3 (x2 + y2 + z 2 )3 (x2 + y 2 + z 2 )3

~ in the usual manner

Now we find the divergence of F

~ = ∂ p −Gmx
div F +
∂
p
−Gmy
+
∂
p
−Gmz
∂x (x2 + y 2 + z 2 )3 ∂y (x2 + y 2 + z 2 )3 ∂z (x2 + y 2 + z 2 )3
x2 + y 2 + z 2 3Gm
= −3Gm + 2
(x2 + y 2 + z 2 )5/2 (x + y 2 + z 2 )3/2
= 0.

Thus, div F ~ = 0 for all points except the origin. We consider a region enclosed by two concentric spheres. Since the
divergence of the field is zero at all points except the origin, the volume enclosed contains only points with zero divergence.
Consequently, the flux through the surface of the enclosed volume must be zero. Since the field is always inward pointing,
this is equivalent to saying that the flux into the outer sphere must equal the flux out of the inner sphere, and so we see
that for any two spheres, the flux must be equal, which shows that the flux is independent of the radius of the spheres.
29. Since both surfaces, S and B, are closed, we use the Divergence Theorem with
     
~ = ∂
div F
1 3
(x + y 3 ) +
∂ 1 3
(y + z 3 ) +
∂ 1 3
(z + x3 ) − z = x2 + y 2 + z 2 − 1.
∂x 3 ∂y 3 ∂z 3
~ < 0 at every point inside S, which is the sphere x2 + y 2 + z 2 − 1 = 0, we have
(a) Since div F
Z Z
~ · dA
F ~ = ~ dV < 0.
div F
S Interior of S

~ > 0 at every point outside S, and B is entirely outside S, we have

(b) Since div F
Z Z
~ · dA
F ~ = ~ dV > 0.
div F
B Interior of B

30. (a) First we calculate

~i ~j ~k
~r × ~a = x y z = (ya3 − za2 )~i + (za1 − xa3 )~j + (xa2 − ya1 )~k .
a1 a2 a3

Then we calculate


div(~r × ~a ) = div (ya3 − za2 )~i + (za1 − xa2 )~j + (xa2 − ya1 )~k = 0.

(b) The cube is closed and oriented outward. By the Divergence Theorem,
Z Z
~ =
(~r × ~a ) · dA div(~r × ~a )dV = 0.
Box Interior of box
1356 Chapter Twenty /SOLUTIONS

31. (a) Since ~a · ~r = a1 x + a2 y + a3 z, we have

div ((~a · ~r )~r ) = div (a1 x + a2 y + a3 z)(x~i + y~j + z~k ) ,
Now
∂
((a1 x + a2 y + a3 z)x) = a1 (1)x + (a1 x + a2 y + a3 z)1
∂x
= 2a1 x + a2 y + a3 z,
so, performing similar calculations to find ∂/∂y and ∂/∂z, we have
div ((~a · ~r )~r ) = 2a1 x + a2 y + a3 z + a1 x + 2a2 y + a3 z + a1 x + a2 y + 2a3 z
= 4a1 x + 4a2 y + 4a3 z = 4~a · ~r .
(b) By the Divergence Theorem,
Z Z
Flux = ~ =
(~a · ~r )~r · dA div ((~a · ~r )~r ) dV
Sphere Interior of sphere
Z Z
= 4~a · ~r dV = 4 (a1 x + a2 y + a3 z) dV.
Interior of sphere Interior of sphere
We calculate the first term in the last integral:
Z Z Z √ 1 1−z 2 Z √1−y2 −z2
a1 x dV = a1 √ √ x dx dy dz
Sphere −1 − 1−z 2 − 1−y 2 −z 2
Z 1 Z √1−z2 √
1−y 2 −z 2
x2
= a1 √ √ dy dz
−1 −1−z 2
2 − 1−y 2 −z 2

Z √
Z 1−z2 p p !
1
( 1 − y 2 − z 2 )2 (− 1 − y 2 − z 2 )2
= a1 √ − dy dz
−1 − 1−z 2
2 2
Z 1 Z √1−z2
= a1 √ 0 dydz = 0.
−1 − 1−z 2

By a similar calculation, the other terms in the integral are 0. Thus the whole integral on the right-hand side is zero.
This result can also be obtained by a symmetry argument, which is much shorter: Since x, y, and z each take
equal positive and negative values on half the sphere, the integral of each term is 0. Thus, the flux is 0:
Z Z
~ · dA
F ~ =4 ~a · ~r dV = 0.
Sphere Sphere

~ = 1 · 2 · 12 = 2.
32. (a) At the point (1, 2, 1), we have div F
(b) Since the box is small, we use the approximation
~ = Flux density ≈ Flux out of box
div F .
Volume of box
Thus
Flux out of box ≈ (div F~ ) · (Volume of box) = 2(0.2)3 = 0.016.
(c) To calculate the flux exactly, we use the Divergence Theorem,
Z Z
Flux out of box = ~ dV =
div F xyz 2 dV.
Box Box

Since the box has side 0.2, it is given by 0.9 < x < 1.1, 1.9 < y < 2.1, 0.9 < z < 1.1, so
Z 1.1 Z 2.1 Z 1.1 Z 1.1 Z 2.1 1.1
2 z3
Flux = xyz dzdydx = xy dydx
0.9 1.9 0.9 0.9 1.9
3 0.9
Z 1.1 2.1 1.1
(1.1)3 − (0.9)3 xy 2 (1.1)3 − (0.9)3 (2.1)2 − (1.9)2 x2
= dx = · ·
3 0.9
2 1.9
3 2 2 0.9
(1.1)3 − (0.9)3 (2.1)2 − (1.9)2 (1.1)2 − (0.9)2
= · · = 0.016053 . . . .
3 2 2
 20.2 SOLUTIONS 1357

~ .
Notice that you can calculate the flux without knowing the vector field, F
33. Any closed surface, S, oriented inward, will work. Then,
Z Z
~ · dA
F ~ =− ~ · dA
F ~,
S (inward) S (outward)

so, by the Divergence Theorem, with W representing the region inside S,

Z Z Z
~ · dA
F ~ =− ~ dV = −
div F (x2 + y 2 + 3)dV.
S (inward) W W

The integral on the right is positive because the integrand is positive everywhere. Therefore the flux through S oriented
inward is negative.
34. Let S3 the sphere of radius 3 and let S5 be the sphere of radius 5. Then we know that
Z
~ · dA
G ~ = 8π.
S3

We consider the region W between S3 and S5 . For 2 ≤ k~r k ≤ 7, we have div G ~ = 3. We apply the Divergence
Theorem to the region W between S3 and S5 , with S5 oriented outward and S3 oriented inward:
Z Z Z
~ · dA
G ~ − ~ · dA
G ~ = ~ dV = 3 · Volume of W
div G
S5 S3 W
 
4 3 4 3
=3 π5 − π3 = 392π.
3 3
Thus, Z Z Z
~ · dA
G ~ = ~ · dA
G ~ + ~ dV = 8π + 392π = 400π
div G
S5 S3 W

35. Using the Divergence Theorem, where W is the interior of S,

Z Z
Flux = ~ · dA
F ~ = ~ dV,
div F
S W

we see that Z Z
2 dV ≤ Flux ≤ 5 dV
W W
2 · Volume of sphere ≤ Flux ≤ 5 · Volume of sphere
33 33
2 · 4π ≤ Flux ≤ 5 · 4π
3 3
72π ≤ Flux ≤ 180π.

36. (a) We cannot use the Divergence Theorem to calculate the flux through the sphere of radius 2 because G ~ is not defined
throughout the interior of the sphere. We calculate the flux directly. Since G ~ is parallel to the area vector at the
~ || = 4 · 22 · 2 = 32 on the surface, we have
surface of the sphere, and since ||G
Z
Flux = ~ · dA
G ~ = 32 · Area of surface = 32 · 4π22 = 512π.
S

~ get longer as we go away from the origin suggests that div G

(b) The fact that the vectors of G ~ > 0. This is confirmed
by calculating the divergence using the formula
~ = 4r 2~r = 4(x2 + y 2 + z 2 )(x~i + y~j + z~k ).
G
∂G1 ~ is positive everywhere outside the unit sphere.
Since = 4(3x2 + y 2 + z 2 ), and so on, div G
∂x
1358 Chapter Twenty /SOLUTIONS

The sphere is completely contained within the box. Apply the Divergence Theorem to the region, W , between
the sphere and the box. This region has surface area the sphere (oriented inward) and the box (oriented outward). The
Divergence Theorem gives
Z Z Z
~ · dA
F ~ + ~ · dA
F ~ = ~ dV > 0.
div G
Box (outward) Sphere (inward) W

So Z Z Z
~ · dA
F ~ − ~ · dA
F ~ = ~ dV > 0.
div G
Box (outward) Sphere (outward) W

Thus, Z Z Z
~ · dA
F ~ = ~ · dA
F ~ + ~ dV.
div G
Box (outward) Sphere (outward) W

So the flux through the box is larger than the flux through the sphere.
37. Since the divergence is zero at all points not containing the charge, the flux must be zero through any closed surface
containing no charge. We imagine a surface composed of two concentric cylinders and their end-caps, where the axis of
both cylinders is the z-axis. Then, since no charge is contained in the region enclosed, the flux through the surface must
be zero. Now, we know that the field points away from the axis, which means it is parallel to the end-caps. Consequently,
there must be no flux through the end-caps. This implies that the flux through the inner cylinder must equal the flux out
of the outer cylinder. Since the strength of the field only depends upon the distance from the z axis, the flux through each
cylinder is a constant. This implies that the following equation must hold

Flux through each cylinder = Ea 2πra L = Eb 2πrb L

where Ea and Eb are the strengths of the field at ra and rb respectively, and L is the length of the cylinders. Dividing
through, we can arrive at the following relationship:

Ea /Eb = rb /ra

If we take Eb to be a constant at a fixed value of rb , then the equation can be simplified to

Ea = k/ra

where k = Eb rb . Thus we see that the strength of the field is proportional to 1/r.
38. (a) Using the expression given for the force, we have
Z
~ · ~i = −
Force in ~i direction = F ~
δgz~i · dA
S
Z
= −δg ~.
z~i · dA
S

Now apply the Divergence Theorem to this integral. (Notice that in order to do this, you need to orient S outward,
hence the minus sign disappears.) Z
~ · ~i = δg ∂z
F dV = 0.
V
∂x
Similarly:
Z
~ · ~j = −δg
Force in ~j direction = F ~
z~j · dA
S
Z
∂z
= δg dV = 0
V
∂y

(b)
Z
Force in ~k direction = F
~ · ~k = −δg z~k · dA
~
S
Z Z
∂z
= δg dV = δg dV = δgV.
V
∂z V
 20.2 SOLUTIONS 1359

~ at that point. So div F

39. (a) The rate at which heat is generated at any point in the earth is div F ~ = 30 watts/km3 .
~ ~ ~
(b) Differentiating gives div(α(xi + y j + z k )) = α(1 + 1 + 1) = 3α so α = 30/3 = 10 watts/km3 . Thus, F ~ = α~r
has constant divergence. Note that F~ = α~r has flow lines going radially outward, and symmetric about the origin.
(c) The vector grad T gives the direction of greatest increase in temperature. Thus, −grad T gives the direction of
greatest decrease in temperature. The equation F ~ = −k grad T says that heat will flow in the direction of greatest
decrease in temperature (i.e. from hot regions to cold), and at a rate proportional to the temperature gradient.
(d) We assume that F ~ is given by the answer to part (b). Then, using part (c), we have

~ = 10(x~i + y~j + z~k ) = −30,000 grad T,

F

so
10
grad T = − (x~i + y~j + z~k ).
30,000
Integrating we get
−10
T = (x2 + y 2 + z 2 ) + C.
2(30,000)
At the surface of the earth, x2 + y 2 + z 2 = 64002 , and T = 20◦ C, so
−1
T = (64002 ) + C = 20.
6000
Thus,
64002
C = 20 + = 6847.
6000
At the center of the earth, x2 + y 2 + z 2 = 0, so

T = 6847◦ C.

~ gives
40. (a) Taking partial derivatives of E
∂E1 ∂
= [qx(x2 + y 2 + z 2 )−3/2 ] = q[(x2 + y 2 + z 2 )−3/2 + x(−3/2)(2x)(x2 + y 2 + z 2 )−5/2 ]
∂x ∂x
= q(y 2 + z 2 − 2x2 )(x2 + y 2 + z 2 )−5/2 .

Similarly,
∂E2
= q(x2 + z 2 − 2y 2 )(x2 + y 2 + z 2 )−5/2
∂x
∂E3
= q(x2 + y 2 − 2z 2 )(x2 + y 2 + z 2 )−5/2 .
∂x

Summing, we obtain div E ~ = 0.

~ and the area vector ∆A
(b) Since on the surface of the sphere, the vector field E ~ are parallel,

~ · ∆A
E ~ = kE
~ kk∆A
~ k.

Now, on the surface of a sphere of radius a,

~k= qk~r k q
kE = 2.
k~r k3 a
Thus, Z Z
~ · dA
~ = q ~ k = q · Surface area of sphere = q · 4πa2 = 4πq.
E kdA
Sa
a2 a2 a2
~ is not defined at the origin (which lies inside
(c) It is not possible to apply the Divergence Theorem in part (b) since E
the region of space bounded by Sa ), and the Divergence Theorem requires that the vector field be defined everywhere
inside S.
1360 Chapter Twenty /SOLUTIONS

(d) Let R be the solid region lying between a small sphere Sa , centered at the origin, and the surface S. Applying the
Divergence Theorem and the result of part (a), we get:
Z Z Z
0= ~ dV =
div E ~ · dA
E ~ + ~ · dA
E ~,
R Sa S

where S is oriented with the outward normal vector, and Sa with the inward normal vector (since this is “outward”
with respect to the region R). Since
Z Z
~ · dA
E ~ =− ~ · dA
E ~,
Sa , inward Sa , outward

the result of part (b) yields Z

~ · dA
E ~ = 4πq.
S
~ is defined everywhere
[Note: It is legitimate to apply the Divergence Theorem to the region R since the vector field E
in R.]
41. Check that div E ~ = 0 by taking partial derivatives. For instance,
∂E1 ∂
= [q(x − x0 )[(x − x0 )2 + (y − y0 )2 + (z − z0 )2 ]−3/2 ]
∂x ∂x
= q[(y − y0 )2 + (z − z0 )2 − 2(x − x0 )2 ][(x − x0 )2 + (y − y0 )2 + (z − z0 )2 ]−5/2
and similarly,
∂E2
= q[(x − x0 )2 + (z − z0 )2 − 2(y − y0 )2 ][(x − x0 )2 + (y − y0 )2 + (z − z0 )2 ]−5/2
∂y
∂E3
= q[(x − x0 )2 + (y − y0 )2 − 2(z − z0 )2 ][(x − x0 )2 + (y − y0 )2 + (z − z0 )2 ]−5/2 .
∂z
Therefore,
∂E1 ∂E2 ∂E3
+ + = 0.
∂x ∂y ∂z
~ is defined everywhere but at the point with position vector ~r 0 . If this point lies outside the surface S,
The vector field E
the Divergence Theorem can be applied to the region R enclosed by S, yielding:
Z Z
~ · dA
E ~ = ~ dV = 0.
div E
S R

If the charge q is located inside S, consider a small sphere Sa centered at q and contained in R. The Divergence Theorem
for the region R0 between the two spheres yields:
Z Z Z
~ · dA
E ~ + ~ · dA
E ~ = ~ dV = 0.
div E
S Sa R0

In this formula,
R the Divergence Theorem requires S to be given the outward orientation, and S a the inward orientation. To
compute S E ~ · dA ~ , we use the fact that on the surface of the sphere, E
~ and ∆A
~ are parallel and in opposite directions,
a
so
~ · ∆A
E ~ = −kE ~ kk∆A ~k
since on the surface of a sphere of radius a,

~k=q k~r − ~r 0 k q
kE = 2.
k~r − ~r 0 k3 a
Then, Z Z
~ · dA
E ~ = −
q ~ k = −q · Surface area of sphere = − q · 4πa2 = −4πq.
kdA
Sa
a2 a2 a2
Z
~ · dA
E ~ = −4πq.
Sa
Z Z
~ · dA
E ~ − ~ · dA
E ~ = 4πq.
S Sa
 20.3 SOLUTIONS 1361

~ = p1~i + p2~j + p3~k we have

42. (a) If p

~ (~r ) = 3 (xp1 + yp2 + zp3 )~r −

E
p
~
(x2 + y 2 + z 2 )5/2 (x2 + y 2 + z 2 )3/2
Thus, taking partial derivatives,

∂E1 ∂ 3x(xp1 + yp2 + zp3 ) − p1 (x2 + y 2 + z 2 )

=
∂x ∂x (x2 + y 2 + z 2 )3/2
∂ 2p1 x2 + 3(p2 y + p3 z)x − p1 (y 2 + z 2 )
=
∂x (x2 + y 2 + z 2 )3/2
(4p1 x + 3p2 y + 3p3 z)(x2 + y 2 + z 2 )3/2
=
(x2 + y 2 + z 2 )3
[2p1 x2 + 3(p2 y + p3 z)x − p1 (y 2 + z 2 )](3/2)(2x)(x2 + y 2 + z 2 )1/2
−
(x2 + y 2 + z 2 )3
p1 x(−2x2 + y 2 + z 2 ) + p2 y(−6x2 + 3y 2 + 3z 2 ) + p3 z(−6x2 + 3y 2 + 3z 2 )
=
(x2 + y 2 + z 2 )5/2
The expressions for ∂E2 /∂y and ∂E3 /∂z are obtained similarly; Summing, we obtain

~ = ∂E1 ∂E2 ∂E3

div E + + = 0.
∂x ∂y ∂z

(b) The Divergence Theorem cannot be applied directly since E ~ is not defined at the origin (which lies inside S) and
the theorem requires that the vector field be defined everywhere inside S. However, it can be applied to the region R
bounded by S and a small sphere Sa with center at the origin and contained inside S. We conclude that
Z Z Z
~ dV =
div E ~ · dA
E ~ − ~ · dA
E ~,
R S Sa

~ = 0 inside R, it follows that

where S and Sa are outward oriented. Since div E
Z Z
~ · dA
E ~ = ~ · dA
E ~.
S Sa

The integral on the right-hand side was computed in Problem 38 on page 936 of the text, and is equal to zero no
matter what the radius a is. Therefore, Z
E~ · dA
~ = 0.
S

Solutions for Section 20.3

Exercises

1. This vector field shows no rotation, and the circulation around any closed curve appears to be zero, so we suspect a zero
curl here.
2. This vector field is definitely swirling, so we suspect a nonzero curl here.
3. The circulation around the boundary of a square in quadrant one is positive, because the vectors on the top of the square
are larger than those at the bottom. Therefore there is a nonzero curl.
4. This vector field shows no rotation, and the circulation around any closed curve appears to be zero, so the vector field has
zero curl.
5. This vector field shows no rotation, so we expect a zero curl.
6. The circulation around the boundary of a square with sides parallel to the axes and enclosing the point is nonzero because
the vectors at the bottom are larger than those at the top, so we expect a non-zero curl at this point.