Assignment Chapter 3: Solution

Problem 3-5:
• The total electric field at point P is given by:
" −−→ −−→ #
~P = 1 Q 1 P Q2 P
E Q1 + Q2
4πǫ0 Q1 P 3 Q2 P 3

where:
−−→ √ √
Q1 P = −~ax 2 − ~ay ⇒ Q1 P = 4+1= 5
−−→ √ √
Q2 P = −~ax 3 + ~ay ⇒ Q2 P = 9 + 1 = 10

• Consequently:
~ P = EP x~ax + EP y~ay
E
   
1 Q1 Q2 1 Q1 Q2
= −2 3/2 − 3 3/2 ~ax + − 3/2 + 3/2 ~ay
4πǫ0 5 10 4πǫ0 5 10

– No x-component: EP x = 0 implying that:

Q1 Q2 Q1 3
−2 3/2
− 3 3/2 = 0 ⇒ =− √
5 10 Q2 4 2

– No y-component: EP y = 0 implying that:

Q1 Q2 Q1 1
− + 3/2 = 0 ⇒ = √
53/2 10 Q2 2 2

1
Problem 3-7:

• The potential at the charge whose coordinates are (0, 0, z) (in cylindrical
coordinates) is given by:
Z
1 ρl ′
V = dl
4πǫ0 L′ R
where:
√
– R= z 2 + b2 .
– dl′ = bdφ′
• Consequently:
Z 2π
ρl b
V = √ dφ′
4πǫ0 0 z + b2
2
Z 2π
ρl b ρb
= √ dφ′ = √l
4πǫ0 z + b 0
2 2 2ǫ0 z 2 + b2

• Consequently, the electric field at the charge is:

~ = −∇V = −~az dV
E
dz
ρl bz
= ~az
2ǫ0 (z 2 + b2 )3/2
implying that the force at the charge is:
Qρl bz
F~ = QE
~ = ~az
2ǫ0 (z 2 + b2 )3/2

• Note that the direction of F~ depends on wether the charge is above (z > 0)
or below (z < 0) the loop.
• For a charge that is at a distance h above the loop:
Qρl bh
F~ = ~az
2ǫ0 (h2 + b2 )3/2

2
 – For h ≫ b, h2 + b2 ≈ h2 implying that:
Qρl b
F~ = ~az
2ǫ0 h2
which corresponds to the inverse square law.
– For h = 0:
F~ = 0
– The force is maximum when:
dF
=0
dh  
Qρl b d h
⇒ =0
2ǫ0 dh (h2 + b2 )3/2
(h2 + b2 )3/2 − h 32 2h(h2 + b2 )1/2
⇒ =0
(h2 + b2 )3
b
⇒ h2 + b2 − 3h2 = 0 ⇒ h = √
2
Second Method:
• Note that instead of determining E~ from V , it is possible to determine E
~
directly: Z
~ = ρl ~ ′
E 3
Rdl
L′ 4πǫ0 R
where:
– dl′ = bdφ′ .
~ = −b~ar + z~az implying that R = (b2 + z 2 )1/2 .
– R
• Consequently:
Z 2π Z 2π 
~ = ρl −b2 ′ bz ′
E ~ar dφ + ~az dφ
4πǫ0 0 (b2 + z 2 )3/2 0 (b2 + z 2 )3/2
 Z 2π Z 2π 
ρl b2 ′ bz ′
= − 2 ~ar dφ + 2 ~az dφ
4πǫ0 (b + z 2 )3/2 0 (b + z 2 )3/2 0
R 2π
• Note that while 0 ~az dφ′ = 2π~az since ~az is a constant (that does not
R 2π
depend on φ′ ), 0 ~ar dφ′ 6= 2π~ar since ~ar depends on φ′ .
• Since ~ar = cos(φ′ )~ax + sin(φ′ )~ay , then:
    
Z 2π Z 2π
~ = ρl − 
  b2   b2 
E 2 2 3/2
cos(φ′ )dφ′  ~ax −  2 2 3/2
sin(φ′ )dφ′ 
 ~ay
4πǫ0   (b + z )   (b + z )
|0 {z } |0 {z }
=0 =0
  
Z 2π
 bz  
′
+ 2
 dφ  ~az 

(b + z 2 )3/2 0
| {z }
=2π

3
 • Finally:
~ = ρl bz
E ~az
2ǫ0 (z 2
+ b2 )3/2
which is the same as the result that was obtained before.
• Note that because of the symmetry of the problem, it was possible to
~ has only a component along the z-axis before solving the
realize that E
problem. After this observation, it is sufficient to determine only the z-
component of E ~ thus simplifying the solution.

– Note that the xy component of E ~ caused by a differential length at

′
φ will be canceled out by the xy component of E~ that is caused by
′
a differential length at φ + π.

Problem 3-8:

• The electric field at the origin is given by:

Z
E~ = 1 ~ ρl dl′
R
4πǫ0 L′ R3
where:
– dl′ = bdφ′ .
~ is the unit vector from the differential charge to the field point O:
– R
~ = −b~ar = −b cos φ′~ax − b sin φ′~ay
R
⇒ R=b

• Consequently:
 
Z π Z π
~ = − ρl ~ax
 
E cos φ′ dφ′ +~ay sin φ′ dφ′ 
4πǫ0 b  
|0 {z } | 0 {z }
=0 =2
ρl
=− ~ay
2πǫ0 b

• Note that because of the symmetry of the problem, the electric field has
no component along the x-direction. In fact, along the x direction, the
effect of the differential linear charge at the the angle φ′ will be canceled
out by the effect of the charge at π − φ′ where 0 ≤ φ′ ≤ π/2.

4
Problem 3-10:
• Use Gauss’s law: I
~ s = Q/ǫ0
E.d~
S

Part (a):
~ is along the x direction, it is parallel to the faces at: y = 0.05
• Since E
(whose normal is ~ay ), y = −0.05 (whose normal is −~ay ), z = 0.05 (whose
normal is ~az ) and z = −0.05 (whose normal is −~az ).
• At the face x = 0.05, ~an = ~ax implying that:
~ = (100x~ax )(ds~ax ) = 100xds = 5ds
~ ds
E.
Z Z
⇒ E. ~ =5
~ ds ds = 5(0.1)2 = 0.05
(1) (1)

• At the face x = −0.05, ~an = −~ax implying that:

~ = (100x~ax )(−ds~ax ) = −100xds = 5ds
~ ds
E.
Z Z
⇒ E. ~ =5
~ ds ds = 5(0.1)2 = 0.05
(2) (2)

• Finally:
Q = ǫ0 (0.05 + 0.05) = 0.1ǫ0
Part (b):
• The flux from the top and bottom discs is zero (the electric field is parallel
to these faces).
~ = rdφdz~ar = 0.05dφdz~ar :
• On the cylindrical sidewall: ds
~ = (100x~ax )(0.05dφdz~ar )
~ ds
E.
= (100(0.05 cos φ)~ax )(0.05dφdz~ar )
= 100(0.05)2 cos φdφdz (~ax .~ar ) = 100(0.05)2 cos2 (φ)dφdz
| {z }
=cos φ

• Consequently:
I Z 2π Z 0.05
~ s = 100(0.05)2
E.d~ cos2 (φ)dφ dz
S 0 −0.05
= 100(0.05)2(π)(0.1) = 0.025π ⇒ Q = 0.025πǫ0

Note that it was possible to solve this problem by applying the divergence
theorem: I Z
~
E.d~s = ~
(∇.E)dv = Q/ǫ0
S V
~ = 100 implying that:
where: ∇.E
Z
Q = 100ǫ0 dv = 100ǫ0V
V

where:

5
 • In part (a), V = 0.13 implying that Q = 0.1ǫ0 .
• In part (b), V = π(0.05)2 0.1 implying that Q = 0.025πǫ0.

Problem 3-12:
• The electric field has no preferred z direction and no preferred φ direction.
Consequently, the electric field intensity can be written as:
~ = Er~ar
E

where Er depends only on r.

~ from
• Consider a Gaussian cylinder of radius r and height L. The flux of E
this surface is given by:
I I I
~ s = (Er~ar ).(~ar d~s) = Er
E.d~ ds = Er (2πrL)
S S S

– For r < a, the total charge enclosed in the Gaussian cylinder is Q = 0

implying that:
Er = 0 ; r < a
– For a < r < b, the total charge enclosed in the Gaussian cylinder is
Q = (2πaL)ρsa implying that:
1 aρsa
Er (2πrL) = (2πaL)ρsa ⇒ Er = ; a<r<b
ǫ0 ǫ0 r

– For r > b, the total charge enclosed in the Gaussian cylinder is Q =

(2πaL)ρsa + (2πbL)ρsb implying that:

aρsa + bρsb
Er = ; r>b
ǫ0 r

~ vanishes for r > b when:

• E
b ρsa
=−
a ρsb

6
Problem 3-13:
• The work is given by:
Z Z
W = −q ~ ~l = −q
E.d (ydx + xdy)

Part (a):
• Along the parabola: x = 2y 2 implying that dx = 4ydy. Consequently:
Z
W = −q (y(4ydy) + 2y 2 dy)
Z 2
= −6q y 2 dy = −14q = 28 (µJ)
1

Part (b):
• Along the straight line: x = 6y −4 implying that dx = 6dy. Consequently:
Z
W = −q (y(6dy) + (6y − 4)dy)
Z 2
= −q (12y − 4)dy = −14q = 28 (µJ)
1

Note that W is the same in both cases since the static electric field is irrotational.

Problem 3-19:
• Assume the circular tube sits on the xy-plane with its axis coinciding with
the z-axis.
• The surface charge density on the tube wall is: ρs = Q/(2πbh).
• The electric potential can be calculated from:
Z
1 ρs ′
V = ds
4πǫ0 S ′ R

where ds′ = bdφ′ dz ′ .

Part (a):
• At a point outside the tube: z > h:
p
R = b2 + (z − z ′ )2

7
 • Consequently:
Z h Z 2π
ρs b
V = p dφ′ dz ′
4πǫ0 0 0 b + (z − z ′ )2
2
Z h
ρs b 1
= (2π) p dz ′
4πǫ0 0 b + (z − z ′ )2
2
Z
ρs b z−h 1
= −√ dZ ′
2ǫ0 z 2
b + Z ′2
ρs b h  p iz−h
= − ln Z ′ + b2 + Z ′2
2ǫ0 z
ρs b h  p   p i
= − ln (z − h) + b2 + (z − h)2 + ln z + b2 + z 2
2ǫ0
√
ρs b z + b2 + z 2
= ln p
2ǫ0 (z − h) + b2 + (z − h)2

• The electric field is given by:

~ = −∇V = −~az dV
E
dz
ρs b d h  p   p i
=− ~az ln z + b2 + z 2 − ln (z − h) + b2 + (z − h)2
2ǫ0 dz
 
1 + 1
2z √ 1 1 + 12 2(z − h) √ 2 1
ρs b  2 2 2 b +(z−h) 2
=− ~az √ b +z − p 
2ǫ0 z + b2 + z 2 (z − h) + b2 + (z − h)2
" p √ #
ρs b (z − h) + b2 + (z − h)2 z + b2 + z 2
= p − √ ~az
2ǫ0 b2 + (z − h)2 + (z − h) b2 + (z − h)2 b2 + z 2 + z b2 + z 2

• Finally, E~ reduces to:

 p √ 
ρ b (z − h) + b 2 + (z − h)2 z + b 2 + z2
~ = s p
E hp i−√ √   ~az
2ǫ0 b2 + (z − h)2 b2 + (z − h)2 + (z − h) b2 + z 2 b2 + z 2 + z
" #
ρs b 1 1
= p −√ ~az
2ǫ0 b2 + (z − h)2 b2 + z 2

Part (b):
• At a point inside the tube: z < h:
 p
2 ′ 2 ′
R= pb + (z − z ) , 0 ≤ z ′ ≤ z;
2 2
b + (z − z) , z ≤ z ≤ h.
′

8
 • Consequently:
"Z Z #
z h
ρs b dz ′ dz ′
V = p + p
2ǫ0 0 b2 + (z − z ′ )2 z b2 + (z ′ − z)2
"Z Z h−z #
0
ρs b dZ ′ dZ ′
= −√ + √
2ǫ0 z b2 + Z ′2 0 b2 + Z ′2
"Z Z h−z #
z
ρs b dZ ′ dZ ′
= √ + √
2ǫ0 0 b2 + Z ′2 0 b2 + Z ′2
" √ p #
ρs b z + b2 + z 2 (h − z) + b2 + (h − z)2
= ln + ln
2ǫ0 b b
    
ρs b 1 p p
= ln 2 z + b2 + z 2 (h − z) + b2 + (h − z)2
2ǫ0 b

• The electric field is given by:

~ − ~az dV
E
dz " √ p #
ρs b d z + b2 + z 2 (h − z) + b2 + (h − z)2
=− ~az ln + ln
2ǫ0 dz b b
 
1 + 1
2z √ 1
1
−1 + 2 (−2)(h − z) √ 2 1
ρs b  2 2 2 b +(h−z) 2
=− ~az √ b +z + p 
2ǫ0 z + b2 + z 2 (h − z) + b2 + (h − z)2
" p √ #
ρs b (h − z) + b2 + (h − z)2 z + b2 + z 2
= p − √ ~az
2ǫ0 b2 + (h − z)2 + (h − z) b2 + (h − z)2 b2 + z 2 + z b2 + z 2

Problem 3-25:
• At the z = 0 plane:
~ 1 = ~ax 2y − ~ay 3x + ~az 5
E
| {z } |{z}
~ 1t
E ~ 1n
E

~ 1t is the tangential component of the E-field

where E ~ ~ 1n is the normal
and E
~
component of the E-field.
~ is always continuous at the boundary:
• The tangential component of E
~ 2t = E
E ~ 1t = ~ax 2y − ~ay 3x

• Since the two media are perfect dielectric media, then the normal compo-
~ is also continuous at the boundary:
nent of D
~ 2n = D
D ~ 1n
~ 2n = ǫr1 ǫ0 E
⇒ ǫr2 ǫ0 E ~ 1n

~ 2n = ǫr1 E
⇒ E ~ 1n = ǫr1 5~az = 10 ~az
ǫr2 ǫr2 3

9
 • Finally, at the boundary:

~ 2 = ~ax 2y − ~ay 3x + ~az 10

E
 3 
~ 2 = 3ǫ0 ~ax 2y − ~ay 3x + ~az 10
~ 2 = ǫr2 ǫ0 E
D
3

• Note that the above expressions are valid only at the boundary. We can
not determine the field at any point inside medium 2.

Problem 3-26:
• P~ can be determined from:
P~t = ǫ0 (ǫr − 1)E
~t ; P~n = ǫ0 (ǫr − 1)E
~n

~ implies that:
• The continuity of the tangential component of E

~ t1 = E
~ t2 ⇒ 1 ~ 1 ~
E Pt1 = Pt2
ǫr1 − 1 ǫr2 − 1

• In the case of perfect dielectrics, the continuity of the normal component

~ implies that:
of D
~ n1 = D
D ~ n2
~ n1 = ǫr2 E
~ n2 ⇒ ǫr1 ~ ǫr2 ~
⇒ ǫr1 E Pn1 = Pn2
ǫr1 − 1 ǫr2 − 1

Problem 3-28:
• The electric field in medium (2) can be written as:
~ 2 = ~ar E2r + ~aφ E2φ
E
~ while ~ar E2φ is the tangential
where ~ar E2r is the normal component of E
component of E. ~
~ implies that:
• The continuity of the tangential component of E
~ t2 = E
E ~ t1 ⇒ E2φ = −3

• In the case of perfect dielectrics, the continuity of the normal component

~ implies that:
of D

~ n1 ⇒ ǫr2 E2r = 5 ⇒ E2r = 5

~ n2 = D
D
ǫr2

• Consequently:

~ 2 = 5 [cos(π/4)~ax + sin(π/4)~ay ] − 3 [− sin(π/4)~ax + cos(π/4)~ay ]

E
ǫr2
   
1 5 1 5
=√ + 3 ~ax + √ − 3 ~ay
2 ǫr2 2 ǫr2

10
 ~ 3 (and hence E
• For E ~ 2 ) to be parallel to the x-axis, E
~ 2 must have no y
component implying that:
5 5
= 3 ⇒ ǫr2 = = 1.667
ǫr2 3

Problem 3-30:
• The permittivity of the dielectric medium can be written as:
ǫ2 − ǫ1
ǫ= y + ǫ1
d

• Assume a charge Q on the upper plate at y = d. Assuming a uniform

distribution of the charges then ρs = Q/S implying that:

~ = − ρs ~ay = −
E
Q
~ay
ǫ S( ǫ2 −ǫ
d y + ǫ1 )
1

• Consequently:
Z y=d Z d
V =− ~ ~l = Q
E.d
1
dy
y=0 S 0 ( ǫ2 −ǫ
d y
1
+ ǫ1 )
ǫ2 −ǫ1
Let Y = d y + ǫ1 , then:
ǫ2 −ǫ1
– dY = d dy.
– For y = 0, Y = ǫ1 .
– For y = d, Y = ǫ2 .
Therefore: Z ǫ2
Q d 1 Qd ln(ǫ2 /ǫ1 )
V = .dy =
S ǫ2 − ǫ1 ǫ1 Y S(ǫ2 − ǫ1 )

• Finally:
Q S(ǫ2 − ǫ1 )
C= =
V d ln(ǫ2 /ǫ1 )

Problem 3-32:
First method:
• For a length L of the coaxial cable, assume a charge Q on the surface of
the inner conductor. The electric flux density is given by:

~ = Q
D ~ar
2πrL
implying that: (
Q
~ =
E 2πǫ0 ǫr1 rL ~
ar , ri < r < b;
Q
2πǫ0 ǫr2 rL ar ,
~ b < r < ro .

11
 • Consequently:
Z ri Z ro
V =− ~
E.d~r = ~ r
E.d~
ro ri
"Z Z ro #
b
Q 1 1
= dr + dr
2πǫ0 L ri ǫr1 r b ǫr2 r
    r 
Q 1 b 1 o
= ln + ln
2πǫ0 L ǫr1 ri ǫr2 b

• Finally:
Q 2πǫ L
C= =   0
(F)
V 1
ln ri + ǫ1r2 ln
b ro
ǫr1 b

• Therefore, the capacitance per unit length is:

C 2πǫ
Cl = =   0
(F/m)
L 1
ln ri + ǫ1r2 ln
b ro
ǫr1 b

Second method:
• The equivalent capacitance can be obtained from the series connections of
two capacitances:
1 1 1
= +
Ceq C1 C2
where C1 (resp. C2 ) is the capacitance of a coaxial line of length L of
inner radius ri (resp. b) and outer radius b (resp. ro ). From lecture 7:

2πǫ0 ǫr1 L
C1 =  
ln rbi
2πǫ0 ǫr2 L
C2 =

ln rbo

• Therefore:  
b

1 ri ln
ln rbo
= +
Ceq 2πǫ0 ǫr1 L 2πǫ0 ǫr2 L
implying that:
2πǫ0 ǫr1 ǫr2 L 2πLǫ
Ceq =  
=   0
(F)
ro ro
ǫr2 ln rbi + ǫr1 ln b
1
ǫr1 ln rbi + ǫr2
1
ln b

• Therefore, the capacitance per unit length is:

C 2πǫ
Cl = =   0
(F/m)
L 1
ln ri + ǫ1r2 ln
b ro
ǫr1 b

12
Problem 3-33:
• Because of the cylindrical symmetry of the problem, the electric field in-
tensities in media (1) and (2) can be written as:
~ 1 = Er1~ar
E ; ~ 2 = Er2~ar
E

where Er1 and Er2 depend only on r.

~ at the boundary between
• The continuity of the tangential component of E
the two dielectrics implies that:

Er1 = Er2 ⇒ ~1 = E
E ~2 = E
~ = Er~ar

~ itself is tangential to the boundary).

(note that E
• Assume a charge Q on the inner conductor. Applying Gauss’s law:
I
~ s=Q
D.d~
S
⇒ Er (ǫ0 ǫr1 (πrL)) + Er (ǫ0 ǫr2 (πrL)) = Q
Q
⇒ Er =
πrLǫo (ǫr1 + ǫr2 )

• Consequently:
Z ri  
Q ro
V =− Er dr = ln
ro πLǫo (ǫr1 + ǫr2 ) ri

• Finally:
Q πLǫo (ǫr1 + ǫr2 )
C= =   (F)
V ln rroi
implying that the capacitance per unit length is:
C πǫo (ǫr1 + ǫr2 )
Cl = =   (F/m)
L ln rroi

Problem 3-37:
First method:
• Assume a charge Q on the inner sphere. Consider a Gaussian sphere of
radius R, then:
D~ = Q ~aR ; Ri < R < Ro
4πR2
implying that:

~1 = Q
E ~aR ; Ri < R < b
4πǫ0 ǫr R2
~2 = Q
E ~aR ; b < R < Ro
4πǫ0 (2ǫr )R2

13
 • Consequently:
Z Ri Z Ro
V =− ~ R
E.d ~ = ~ R
E.d ~
Ro Ri
"Z Z #
b
Q 1 1 Ro 1
= 2
dR + dR
4πǫ0 ǫr Ri R 2 b R2
   
Q 1 1 1 1 1
= − + −
4πǫ0 ǫr Ri b 2 b Ro
 
Q 1 1 1
= − −
4πǫ0 ǫr Ri 2b 2Ro

~ is given by:
• D

~ = ǫ0 ǫr V
D   ~aR ; Ri < R < Ro
1 1 1
R2 Ri − 2b − 2Ro

~ =0
D ; R < Ri , R > Ro

~ is given by:
• E

~1 = V
E   ~aR ; Ri < R < b
1 1 1
R2 Ri − 2b − 2Ro

~2 = V
E   ~aR ; b < R < Ro
2 1 1
R2 Ri − b − Ro

~ =0
E ; R < Ri , R > Ro

• The capacitance is given by:

Q 4πǫ0 ǫr
C= = 1 1
V Ri − 2b − 2R1 o

Second method:
• The equivalent capacitance can be obtained from the series connections of
two capacitances:
1 1 1
= +
Ceq C1 C2
where C1 (resp. C2 ) is the capacitance of a spherical capacitor of inner
radius Ri (resp. b) and outer radius b (resp. Ro ). From lecture 7:
4πǫ0 ǫr
C1 = 1 1
Ri − b
4πǫ0 (2ǫr )
C2 = 1 1
b − Ro

14
 • Therefore:
   
1 1 1 1 1 1 1
= − + −
Ceq 4πǫ0 ǫr Ri b 2 b Ro
resulting in:
4πǫ0 ǫr
Ceq = 1 1 1
Ri − 2b − 2R o

Problem 3-44:
Part (a):
• Inside the dielectric:
Z d Z d Z d
V0 = − E ~ =−
~ 1 .dl (E1~ay )(dy~ay ) = − E1 dy = −E1 d
0 0 0

implying that:

~ 1 = −~ay V0
E ⇒ ~ 1 = −~ay ǫ0 ǫr V0
D
d d
Since the electric field inside a conductor is zero, the surface charge density
on the upper plate (between 0 and x) is:

~1 −D
ρs1 = ~an .(D ~ ~ 1 = ǫ0 ǫr V0
) = (−~ay ).D (top plate)
| conductor
{z } d
=0

where ~an = −~ay is the outward normal from the conductor.

• In the air (between x and L − x):

~ 2 = −~ay V0 ⇒ D
E ~ 2 = −~ay ǫ0 V0
d d
V0
ρs2 = ǫ0 (top plate)
d

Part (b):
• The magnetic energy stored in the dielectric is:
Z  2 Z  2
1 1 V0 1 V0
We1 = ǫ0 ǫr E12 dv = ǫ0 ǫr dv = ǫ0 ǫr (xwd)
2 V′ 2 d V′ 2 d

• The magnetic energy stored in air is:

Z  2
1 2 1 V0
We2 = ǫ0 E2 dv = ǫ0 ((L − x)wd)
2 V′ 2 d

• We1 = We2 implies that:

L
ǫr x = L − x ⇒ x =
ǫr + 1

15