Physics 225 Problem Set #2 Solutions

Wen Han Chiu

Problem 1
The slanted face is given by the equation

x+y+z =1

So for a fixed y and fixed z, x can take values from 0 to 1 − y − z. Projecting the the surface onto
the yz plane yields a right triangle with the hypotenuse given by

y+z =1

and hence, on that surface, y can take values from 0 to 1 − z. Lastly, projecting the triangle onto
the z axis tells us that z takes value from 0 to 1. Hence, the desired integral is
Z 1 Z 1−z Z 1−z−y Z 1 Z 1−z
z 2 dxdydz = z 2 (1 − z − y)dydz
0 0 0 0 0
Z 1  
2 2 1 2
= z (1 − z) − (1 − z) dz
0 2
Z 1
1
= z 2 − 2z 3 + z 4 dz
2 0
 
1 1 2 1
= − +
2 3 4 5
1
=
60

1
Problem 2
For the line integral, we first split the curve into three sections: 1. along the y axis, 2. along
the hypotenuse, 3. along the z-axis. For the first section, the curve has the parametrization:

x(t) = z(t) = 0, y(t) = t, t ∈ [0, 2],

thus the line integral along that section is given by

Z Z 2
v · dr = [v(0, t, 0) · ŷ]dt
section 1 0
Z 2
= (2t × 0)dt
0
=0
Similarly, the curve in the third section has the parametrization:

x(t) = y(t) = 0, z(t) = t, t ∈ [0, 2],

thus the line integral along that section is given by

Z Z 0
v · dr = [v(0, 0, t) · ẑ]dt
section 3 2
Z 0
= (3t × 0)dt
2
=0
The curve in the second section has the parametrization:

x(t) = 0, y(t) = 2 − t, z(t) = t, t ∈ [0, 2],

thus, the line integral along that section is

Z Z 2
v · dr = [v(0, 2 − t, t) · (−ŷ + ẑ)]dt
section 2 0
Z 2
= [−2(2 − t)t + 3t × 0]dt
0
 
2 2 1 3
= −2 2 − 2
2 3
8
=−
3
So the total line integral is I  
8 8
v · dr = 0 + − +0=−
3 3
Computing the curl yields
X
∇×v = ijk ∂i vj êk
ijk

= (∂2 v3 − ∂3 v2 )ê1 + (∂3 v1 − ∂1 v3 )ê2 + (∂1 v2 − ∂2 v1 )ê3

= (0 − 2y)x̂ + (0 − 3z)ŷ + (0 − x)ẑ
= −2yx̂ − 3z ŷ − xẑ

2
The unit normal vector for the surface is x̂. For a fixed z, y takes values from 0 to 2-z and z takes
values from 0 to 2. Thus,
Z Z 2 Z 2−z
∇ × v) · da =
(∇ −2ydydz
0 0
Z 2
=− (2 − z)2 dz
0
Z 0
= Z 2 dZ (Z = 2 − z)
2
1
= − 23
3
8
=−
I3
= v · dr,

thus demonstrating that Stokes’ theorem holds for this case.

3
Problem 3
(a) First, we note that the dimensions of the Dirac delta function is the inverse dimension of its
argument, i.e.
[δ(·)] = [·]−1 (1)
Since we need to restrict the charge to a specific point, we need 3 delta functions. So by
dimensional analysis, the charge density should be of the form

ρ(r) = cqδ 3 (r − r0 ),

where c is some overall normalization that we need to evaluate. Demanding the volume
integral of ρ to equal q yields
Z
q = ρdτ = qc ⇒ c = 1,

and hence
ρ(r) = qδ 3 (r − r0 )

(b) Using part (a), we have

ρ(r) = −qδ 3 (r) + qδ 3 (r − a)

(c) Since we only fix the radial distance, we only have one delta function. By dimensional analysis,
we need to divide by something of dimension [L]2 to have a quantity with dimensions of charge
density. As the only quantity in this problem with length dimension is R, the answer is of
the form:
c̃
ρ(r) = 2 Qδ(r − R),
R
where c̃ is some overall normalization. Demanding the volume integral to equal Q yields
Z ∞ Z
c̃ 2
Q= Qδ(r − R)r dr dΩ = [c̃Q]4π,
0 R2
Hence,
1
ρ(r) = Qδ(r − R)
4πR2
Problem 4
The sketch of the electric field is obtained by flipping the direction of E2 in Figure 2.4(b) of the
textbook. As the magnitudes of E1 and E2 are equal, the resulting sum has no z component and
the x components conspire:
1 q
Ex = 2 sin θ,
4π0 r 2
where the angle θ is defined in the same manner as Example 2.1 in the textbook. As in the example,
s  2
d d
r = z2 + , sin θ =
2 2r
Thus,
1 qd
E= x̂
4π0 [z + (d/2)2 ]3/2
2

4
Problem 5
Since the problem is rotationally invariant about the z-axis, we know that for points along the
z-axis:
E = Ez ẑ. (2)
Now consider an infinitesimal portion of the disc of area da0 located at (ρ, φ, z) = (ρ0 , φ0 , 0). Its
contribution to the z-component of the electric field at P is given by

1 z σda0
dEz = ,
4π0 r r 2
where p
r = z 2 + ρ02 , da0 = ρ0 dρ0 dφ0 .
The total z component is obtained by summing over all such portions of the disc. Hence,
Z 2π Z R
σz 1
Ez = ρ0 dρ0 dφ0
0 0 4π0 [z 2 + ρ02 ]3/2
Z R
σz 1
= 02 3/2
ρ0 dρ0
20 0 [z 2 +ρ ]
2 2
σz R +z
Z
1
= dR (R = ρ02 + z 2 )
40 z 2 R3/2
 
σz 1 1
= −√
20 z R2 + z 2
Thus,
 
σz 1 1
E= −√ ẑ
20 z R2 + z 2
Problem 6
Consider the same problem but with 8 cubes surrounding the charge, each with the same the
overall configuration as Figure 2.17. As the problem is fully symmetric, the flux through all 24 of
the squares on the surface is identical. thus, the flux through a single square is 1/24-th of the total
flux. As the total flux is just q/0 , we have
Z
q
E · da =
shaded side 240

5
Problem 7
First note that the problem is spherically symmetric (rotationally invariant about any axis
through the center of the sphere). Rotational invariance tells us that E must be unchanged under
the transformation: φ → φ + δφ, θ → θ + δθ for any choice of δφ and δθ. This tells us that E can
only depend on r and must point in the direction of the rotation axis. As any axis can be chosen
as the rotation axis, we can conclude that

E(r) = Er (r)r̂ (3)

Choosing a spherical Gaussian surface of radius R, the flux through the surface is given by
I Z
E · da = Er (R)R2 dΩ (Using Eq. (3))

= 4πR2 Er (R)
The total charge enclosed by Gaussian surface is
Z Z Z R
Qenc = ρdτ = krr2 drdΩ = πkR4
0

Hence, by Gauss’ law,

k 2
Er (R) = R
40
Since this holds for any R less than or equal to the radius of the sphere, we can conclude that:

k 2
E(r) = r r̂
40

Problem 8
First, we note that the only r dependence is in r so

r̂r
Z  
1 0
∇r × E = ρ(r )∇∇r × dτ 0 ,
4π0 r2

where the subscript denotes the variable that the derivative is w.r.t.. Next, we note that by chain
rule,
∂ ∂(xi − x0i ) ∂ ∂
= = .
∂xi ∂x ∂(xi − x0i ) ∂(xi − x0i )
Hence,
r̂r r̂r
   
∇r × = ∇r × =0
r2 r2

⇒ ∇×E=0