Solution to

F.F. Chen’s Plasma Physics

Tao Ye
leafybillow@gmail.com

by
c Tao Ye
All rights reserved.

Please DO NOT DISTRIBUTE this document.

Instead, link to
yetao1991.wordpress.com
Last Updated: May 2, 2015
2
Contents

1 Introduction 5
Problem 1-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Problem 1-8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Problem 1-9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Problem 1-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2 Motion of Particle 9
Problem 2-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Problem 2-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Problem 2-7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Problem 2-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Problem 2-11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Problem 2-15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Problem 2-16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Problem 2-17 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Problem 2-18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Problem 2-19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Problem 2-20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Problem 2-21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3 Wave in Plasma 17
Problem 4-5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Problem 4-9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Problem 4-16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Problem 4-18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Problem 4-23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Problem 4-26 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Problem 4-27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Problem 4-37 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Problem 4-38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Problem 4-39 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Problem 4-40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Problem 4-41 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Problem 4-42 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3
4 CONTENTS

Problem 4-43 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Problem 4-45 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Problem 4-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Problem 4-50 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Chapter 1

Introduction

Problem 1-3

Problem 1-3
1e8
5
Debye length is constant at the line of
Nd is constant at the line of

4 (1)
1e6

3 (2.2)
1e4

2 (2.1)
log(kT)

100
1

(4)
0 1

(6)
-1 (5)
(3)

-2 (7)

6 1e-1 8 1e-210 1e-312 1e-414 1e-516 18 1e-720 22 24

log(n)

Figure 1.1: Label (1)(2.1)(2.2)(3)(4)(5)(6)(7) correspond to plasma in different

condition as the problem describes.

5
6 CHAPTER 1. INTRODUCTION

According to the definition of the Debye Length

0 kTe 1/2
λD = ( ) (1.1)
ne2
1 0 1 1
⇒ log(λD ) = log( 2 ) + log(kTe ) − log(n) (1.2)
2 e 2 2
λD
⇒ log(kT ) = log(n) + 2 log( ) (kT in eV) (1.3)
7430
Then we can draw the solid straight line in the Figure?? with the Debye
length as parameter ranged from 10−1 to 10−7 . Points on a certain solid line,
named with equi-Debye-length line(analog to equipotential lines in electrostatic)
, share a same Debye length. Similarily, with the given equation:

4 3
ND = πλ n (1.4)
3 D
4 kT 3
= π(7430 · )2 (kT in eV) (1.5)
3 n
3
ND (kT ) 2
⇒ = 1 (1.6)
2.7 × 106 n2
ND 3 1
⇒ log( ) = log(kT ) − log(n) (1.7)
2.7 × 106 2 2
1 2 ND
⇒ log(kT ) = log(n) + log( ) (1.8)
3 3 2.7 × 106

The dot lines in the Fig ?? can be named with equi-ND lines, which means
they share a same value of ND in a dot lines.
As for the usage of this figure, take the point (4) for example, the point(4)
falls in the region enclosed by two solid lines and two dots ones.The two solid
lines respectively have the Debye length of 10−3 m and 10−4 m. And the dots
lines have the particle numbers ND of 104 and 106 . That is tantamount to the
fact that (4) has Debye length 10−4 m < λD < 10−3 m and number of particle
104 < ND < 106 .
Recall for the criteria for plasmas, ND ≫ 1 is automatically meet since the
smallest number of particle is larger than 100.

Problem 1-8
The Debye length is

T 1
λD = 69( ) 2 (T in the unit of K)
n
5 × 107 1
= 69 × ( )2
1033
= 1.54 × 10−11 m
. PROBLEM 1-9 7

Naturally, the number of particles contained in a Debye Sphere is :

4 3
ND = πλ × n
3 D
4
= π × (1.54)3
3
≈ 15

Problem 1-9
Since protons and antiprotons have the same inertia, both of them are fixed.
Assume that protons and antiprotons follows the Maxwellian distribution.
1 2
f (u) = Ae−( 2 mu +qφ)/kTe
,
where q equals to e for protons while q equals to −e for antiprotons. Moreover,
np (φ → ∞) = n∞
np̄ (φ → ∞) = n∞ .
Then we obtain
np = n∞ exp( −eφ

kT )
eφ
np̄ = n∞ exp( kT )
The Poisson’s Equation is
∂2φ
ε0 ∇2 φ = ε0 = −e(np − np̄ )
∂x2
With eφ/kT  1,
∂2φ 2n∞ e2 φ
ε0 =
∂x2 kT
r
ε0 kT
∴ λD = = 0.4879m
2n∞ e2

Problem 1-10
Regrad it as an isotropic space, which means that Φ has no components ofθ or
φ.
e−kr
Φ = A
r
1 ∂ 2 ∂ 2 ∂Φ ∂ 2 Φ
∇2 Φ = 2
(r Φ) = +
r ∂r ∂r r ∂r ∂r2
∂Φ kr + 1 −kr
= −A e
∂r r2
∂2Φ k 2 r2 + 2kr + 2 −kr
= Ae
∂r2 r3
8 CHAPTER 1. INTRODUCTION

e−kr
∴ ∇2 Φ = Ak 2
r
ne2
∵ ∇2 Φ = Φ − q0 δ(r − a),
ε0 kB T
where q0 = 4πε0 aφ0 .
s r
ne2 ne2 1 ε0 kB T
∴ = k2 ⇒ k = ⇒ λD = =
ε0 kB T εkB T k ne2

Consider the boundary condition :

e−ka Φ0 a
A = Φ0 ⇒ A = −ka
a e
So,
Φ0 , r ∈ (0, a]
(
− r
Φ= Φ0 a e λD (1.9)
− a r , r ∈ (r, +∞)
e λD
Chapter 2

Motion of Particle

Problem 2-2
Since A=2, for deuterium ion,

m = 2mp = 3.34 × 10−27 kg

q = |e| = 1.60 × 10−19 Coulomb.

Assume that energy can be entirely converted to kinetic energy, then the mo-
mentum can be derived
p2 p
E = Ek = ⇒ p = 2mEk = 1.46 × 10−20 kg · m/s
2m
The Larmour radius
mv⊥ 1.46 × 10−20
rL = = = 0.018m  a = 0.6m
|q|B 5 × 1.60 × 10−19
So the Larmour radius satisfies the confined-ion condition.

Problem 2-3
To keep a equibrillium in the ŷ direction, the electric force
should conteract the Lorentz force

Eq = qvB

⇒ E = vB = 106 V/m

Problem 2-7
Apply the Gauss Law to obtain the magnitude of electric field at r=a

9
10 CHAPTER 2. MOTION OF PARTICLE

E · l · 2πa = ne l · πa2 /ε0

ene a
∴E= = 9.04 × 103 V /m
2ε
E
vE = = 4.52 × 103 m/s
B
(Direction:See in the figure)

Problem 2-10
The mass of a deutron is
md = 1875MeV/c2
The kinetic energy

1
Ek = md v 2 ⇒ v = 1.386 × 106 m/s
2

v⊥ = v cos 45◦ = 9.8 × 105 m/s

md v⊥
rL = 0.03m
qB

Problem 2-11
1 1
Rm = 4 ⇒ 2 = 4 ⇒ sin θm =
sin θm 2
π
∴ θm =
6
Since the velocity is isotropic distribution, the direc-
tion of velocity should distribute uniformly

dΩ = sin θdθdϕ

The total solid angle for a sphere is

Ωtotal = 4π

The solid angle for loss cone is

Z π Z 2π
√
6 3
Ωloss = 2 sin θdθ dϕ = (1 − )4π
0 0 2
√
3
∴ the fraction of the trapped is 2 .
. PROBLEM 2-15 11

Problem 2-15
Define the displacement in polarization direction is xp .
So the work done by the electric field is

~ · x~P
W = qE

The energy gain rate is

dW dxp
= qE = qEvp .
dt dt
The change of kinetic energy

dEk d 1 dvE
= ( mv 2 ) = mvE
dt dt 2 E dt
Without loss of generality, let the vE in the same direction of E

E
vE = −
B
dvE 1 dE
∴ = −
dt B dt
dEk 1 dE
∴ = mvE (− ) = qEvp
dt B dt
m 1 dE
vp =
qB B dt

Replace ± ω1c = m
qB Thus,
1 dE
vp = ±
ωc B dt

Problem 2-16
a) The Larmor frequency of electron is

eB 1.6 × 10−19 × 1
ωe = = = 1.76 × 1011 rad/sec
m 9.109 × 10−31

b) the Larmor frequency of ion is

eB 1.6 × 10−19 × 1
ωi = = = 9.58 × 107 rad/sec
mi 1.67 × 10−27

Since ω0 = 109 rad/sec, ωe  ω0  ωi . The motion of electron is adiabatic,

while that of ion not.
12 CHAPTER 2. MOTION OF PARTICLE

Problem 2-17
2
mv⊥
Since µ = 2B is conservative under this condition, it is easy to derive:
2 02
mv⊥ mv⊥
=
2B 2B 0

where 12 mv⊥
2
= 1keV, B = 0.1T, B 0 = 1T , so

1
mv 02 = 10keV.
2 ⊥
When collision happens, the direction of motion distorts, so v⊥ = vk . Then the
kinetic energy is
1
mv 002 = 5keV
2 ⊥
Implement the adiabatic characteristic of µ, we know that
002 0002
mv⊥ mv⊥ 1 0002
= ⇒ mv⊥ = 0.5keV
2B 0 2B 0 2
Finally, the energy is

1 1
E = E⊥ + Ek = mv 0002 + mvk002 = 5.5keV
2 ⊥ 2

Problem 2-18
a) At a certain moment, we calculate the motion in one periodic circular motion
to certify the invariance of µ. In this period, we assume that the Larmour radius
does not change with minor deviation of magnetic field-B. So

2 m2 v⊥
2
s = πrL =π
q B2
2

d dB m2 v 2 dB
Φ= ·S =π 2 ⊥ = ε(induced − potential)
dt dt q B 2 dt
So the change of the energy of the particle in one period is

m 2 v⊥
2
dB
δW = qε = π
qB 2 dt

Within this period of gyration, the change of magnetic field is

dB dB 2πm
∆B = τ=
dt dt qB
. PROBLEM 2-19 13

So after a period, the energy is

1 2πm dB 1
Ek0 = mv 2 + δW = (1 + ) mv 2
2 ⊥ qB 2 dt 2 ⊥
2πm dB
B0 = B + ∆B = (1 + )B
qB 2 dt

Ek0 E0
µ0 = = =µ
B0 B0
So µ is invariant for both ion and electron.
b) Assume that 12 mv⊥ 2
= KT, 12 mvk2 = KT at initial moment. Considering

1 2 1 02
2 mv⊥ 2 mv⊥
µ= = ,
B B0
and B = 13 B 0 , so 1 02
2 mv⊥ = 3KT . Furthermore, we assume that vk remains
unchanged,
1
mv 2 = KT
2 k
kT⊥ = 3Kev, kTk = 1KeV

Problem 2-19
µ0 I
The magnetic field is B = 2πr , so the gradient is :

∂B µ0 I
∇B = − =
∂r 2πr2
Apply the equipartition theorem of Maxwellian gas, the total energy is
1 2 3
ε= (p + p2y + p2z ) = kT
2m x 2
1 2
∴ ε⊥ = (p + p2y ) = kT
2m x
The perpendicular velocity is :
r
2kT
∴ v⊥ =
m
mv⊥
∴ rL =
eB
1 ∇B kT
∴ v∇B = v⊥ rL = , Upwards
2 B eBr
mvk2
r
kT kT
∵ vk = ⇒ vR = = , Upwards
m eBr eBr
14 CHAPTER 2. MOTION OF PARTICLE

2kT
∴ vR+∇B =
eBr
Area of the top surface of the cyclinder is :
a a
s = π[(R + )2 − (R − )2 ] = 2πRa
2 2
The number of charged particle, which hit the surface s with in time of dt is:

N = n · v · s · dt.

So the hit rate is

dN
= n · v · s.
dt
Since R  a, the drift velocity with [R − a2 , R + a2 ] region can be considered as
uniform. So the accumulation rate is
4πkT na
Racc = n · vR+∇B · s · e = = 20Coulomb/S
B

Problem 2-20
a) vk2 +v⊥
2
is invariant because of the conservation of energy. At z = 0, Bz = B0 ,
v⊥ = 3 v ,vk = 13 v 2 ,
2 2 2 2

2
mv⊥ mv 2
µ= = .
2B0 2B0
2
When the electron reflects, vk = 0, then v⊥ = v2

mv 2 3
µ= ⇒ Bz = B0
3B0 2
√
2
⇒ 1 + α2 z 2 ⇒ z = ±
2α
b)
2
mv⊥ 1
∵µ= ⇒ µB0 (1 + α2 z 2 ) = mv⊥
2
2Bz 2
µ0 B 2µ0 B 2 2
∴ vk2 = v 2 − v⊥
2
= − α z
m m
dz µ0 B
∴ ( )2 = (1 − 2α2 z 2 )
dt m
r
dz µ0 B0 p
= 1 − 2α2 z 2
dt m
√ r
2 2µ0 B0
⇒z= sin( αt + φ)
2α m
. PROBLEM 2-21 15

And this equation can describe the trajectory ofqthe particle.

c) It is apparent that the gyration frequency is 2µm
0 B0
α from the the equation
of motion. q √
d) Claim: θ = 2µm 0 B0
αt + φ, z = 2α2 sin θ
Z b
J= vk dz
a
√ √
where a = − 2α2 , b = 2
2α . Thus,
π π
θa = − , θb =
2 2
Z b Z r Z θb
1 m
J= vk dz = vk vk dt = v 2 dθ
a α 2µB0 θa k
where
r
1 m
dt = dθ,
α 2µB0
dz µB0
vk2 = ( )2 = sin2 θ.
dt m

r Z π2
1 m µB0
∴J = sin2 θdθ
α 2µB0 − π2 m
r
π µB0
= = constant
2α 2m

Problem 2-21
a) According to the Ampere theorem, the magnetic field can be obtained

µ0 I
2πr · B = µ0 I ⇒ B =
2πr
In cylinder coordinate:
∂B µ0 I
∇B = − =
∂r 2πr2
1 ∇B
∴ v∇B = v⊥ rL ,
2 B
mv⊥
where rL = eB , v⊥0 = v⊥ .
2
1 mv⊥0 1 πmv⊥0
∴ v∇B = v⊥0 =
2 eB r µ0 Ie
16 CHAPTER 2. MOTION OF PARTICLE

Besides,

mvk2 1
vR =
e Rc B
mvk2 1
= µ0 I
e Rc · 2πRC
2πmvk2
=
µ0 Ie
2
3πmvk0
∴ ~v = v~R + v∇B
~ = (−ẑ)
µ0 Ie
(vk0 = v⊥0 )
Chapter 3

Wave in Plasma

Problem 4-5
According to the dispersion relation

3KTe 2
ω 2 = ωp2 + k
m

Now we calculate the unknown value in the equation

√
ωp ≈ 2π n = 5.62 × 109 rad/sec

ω = 2πf = 6.908 × 109 rad/sec

3kTe 3 × 100 × 1.6E−19

= = 5.27 × 1013
m 9.109 × 10−31

ω 2 − ωp2
∴ k2 = 3kTe
= 3.05 × 105 ⇒ k = 552.377
m

2π
⇒λ= = 1.14 × 10−2 m
k

Problem 4-9
The critical density is

mε0 ω 2
n0 = = 1.12 × 1015 m−3
e2

17
18 CHAPTER 3. WAVE IN PLASMA

Problem 4-16
If the motion of ion is neglected, the dispersion relation of electron is

c2 k 2 ωp2 ω 2 − ωp2
= 1 −
ω2 ω 2 ω 2 − ωh2

1) The resonance of X-wave is found by setting k → ∞. So the dispersion

relation can be rewrite into
ω 2 − ωp2
c2 k 2 = ω 2 − ωp2
ω 2 − ωh2
Differentiate both sides of the equation
2ω(ωp2 − ωh2 2
2kc2 dk = 2ωdω − ωp dω
ω 2 − ωh2
So the group velocity is
dw kc2
vg = = ωc2 ωp2
dk ω[1 + (ω2 −ω 2 )2 ]
h

When k → ∞ ,which implies that ω = ωh .

ω2 ω2
c p c p ω2 ω2
So 1 + (ω2 −ω 2 2 → ∞, k → ∞ and 1 + (ω 2 −ω 2 )2 has a higher order than k. Thus,
h) h
vg = 0 at resonance point.
2) The cut-off X-wave is found by setting k=0, then

ωp2 ω 2 − ωp2
1− ⇒ ω = ωR or ωL
ω 2 ω 2 − ωh2
ωc2 ωp2
At this point, ω(1 + (ω 2 −ωh 2 )2 ) is a finite value. As a result

kc2
vg = ωc2 ωp2
|k=0 = 0
ω[1 + (ω 2 −ωh 2 )2 ]

∴ Q.E.D.

Problem 4-18
For L-wave,
c2 k 2 ωp2 /ω 2
= 1 −
ω2 1 + ωc /ω
And
n0 e2 eB
ωp2 = , ωc =
mε0 m
. PROBLEM 4-23 19

The cut-off is found when k = 0. With the provided condtion f = 2.8GHz, B0 =

0.3T , the critical density is

mε0 eB
n0 = 2
[(2πf )2 + 2πf ]
e m
= 3.89 × 1017 m−3

Problem 4-23
R-wave
ω2 ωp2
k2 = −
c2 c2 (1 − ωc /ωp )
L-wave
ω2 ωp2
k2 = −
c2 c2 (1 + ωc /ωp )
Thus
s
ω ωp2 /ω 2 ω 1 ωp2 /ω 2 ω 1 ωp2 ωc
kR = 1− ωc ≈ (1 − ωc ) ≈ [1 − 2
(1 + )]
c 1− ω c 21− ω c 2ω ω
s
ω ωp2 /ω 2 ω 1 ωp2 /ω 2 ω 1 ωp2 ωc
kL = 1− ωc ≈ (1 − ωc ) ≈ [1 − (1 − )]
c 1+ ω c 21+ ω c 2 ω2 ω

The difference of the phase is twice of the Faraday rotation angle

1 180◦
∆θ = (kL − kr )
2 π
180◦ 1 1 e2 e
= ( 2) B(Z)n(z)
2π c 4π mε0 m
90◦ 1 1 λ0 e3
= ( 2 )( )2 2 B(z)n(z)
π c 4π c m ε0
◦ 3
90 e
= B(z)n(z)λ20 (degree)
π 4π 2 m2 ε0 c3

And
90◦ e3
= 1.5 × 10−11 degree = 2.62 × 10−13 rad
π 4π m2 ε0 c3
2

In conclusion
L L
180◦
Z Z
θ= (kL − kR )dz = 1.5 × 10−11 λ0 B(z)n(z)dz
2π 0 0

(in the unit of degree)

20 CHAPTER 3. WAVE IN PLASMA

Problem 4-26
a)
B
va = √ = 2.18 × 108 m/s
µ0 ρ
b) The Alfvén wave represents for phase velocity. And phase velocity did not
carry information. So it does not mean that wave can travel faster than light.

Problem 4-27
ρ = n0 M = 1.67 × 10−19
B
vA = √ = 2.18 × 104 m/s
µ0 ρ

Problem 4-37
a) Consider the elastical collision from ion, the equation of eletron’s motion
∂ve
m = −eE − mve ν
∂t
Linearize the equation

n0 e2 E1
−imωve = −eE − mve ν ⇒ j1 = −n0 eve = −
im(ω + iν)
The equation of the transverse wave is

(ω 2 − c2 k 2 )E1 = −iωj1 /ε

Insert j1 into the wave equation, we get

ω
(ω 2 − c2 k 2 )E1 = ω 2 E1
ω + iν p
ωωp2
⇒ ω 2 − c2 k 2 =
ω + iν
c2 k 2 ωp2
⇒ = 1 −
ω2 ω(ω + iν)
∴ Q.E.D.

b) Apply the previous result

ωp2
c2 k 2 = ω 2 −
1 + i ων
. PROBLEM 4-38 21

ν
When ν  ω or ω  1,then
ν ν
(1 + i )−1 ≈ 1 − i
ω ω
So the dispersion relation turns to

νωp2
c2 k 2 = ω 2 − ωp2 + i
ω
Assume that ω = a + bi, then

ωp2 ν
ωp2 + c2 k 2 = (a2 − b2 ) + 2abi + i
a + bi
ωp2 vb ωp2 νa
= (a2 − b2 ) + + i(2ab + )
a2 + b2 a2 + b2
The image part should be zero, that is

−ωp2 v
ib = i = Im(ω)i
2(a2 + b2 )
So the damping rate:
ωp2 ν
γ = −Im(ω) =
2|ω|2

c) With previous conclusion of a), we get

ω 2 − ωp2 ωp2 ν
k2 = + i
c2 c2 ω
Let k = e + di, then k 2 = e2 − d2 + 2edi, we get
ω 2 −ωp2
(
e2 − d2 = c2
ωp2 ν
2ed = c2 ω

By solving the simultaneous equations, it is easy (Uh huh!)to obtain that

1 1 2c ω 2 ωp2
δ= = = ( · 2 )(1 − 2 )1/2
Im(k) d ν ωp ω

Problem 4-38
The loss part of energy will heat up the electron to oscillation. So we firstly
need to derive the vibration motion due to the microwave. The wave-length of
the microwave is λ = 0.3m So the frequency is
2πc
ω= = 6.28 × 109 rad/sec
λ
22 CHAPTER 3. WAVE IN PLASMA

The collision frequency is

¯ = 102 /sec
ν = nn σν

So the equation of motion of electron is

∂ve
m = −eE − mve ν
∂t
Linearize the equation to be
eE
iωmve = −eE − νmve ⇒ ve =
im(ω + iν)
Then the dispersion raltion is
ω 2 − ωp2 ωp2 ν
k2 = + i
c2 ωc2
k = Re(k) + Im(k)i
And microwave in plasma is

E = E0 ei(kr−ωt) = E0 ei(Re(k)r−ωt) e−Im(k)r

The term e−Im(k)r means the wave decay when penetrate the ionosphere. The
ratio of outgoing energy v.s. incident energy is
2
Eout e−Im(k)R1 2
A= 2 = ( ) = e−2Im(k)(R1 −R2 )
Eini e−Im(k)R2

And R1 − R2 = 100km = 105 m

ne e2
ωp2 = = 3.17 × 1014
ε0 m

ν ωp2 1
Im(k) = = 1.34 × 10−12
2c ω 2
q
ωp2
1− ω2
−7
∴ A = e−2.69×10 = 0.999999732
So the loss fraction is
1−A≈0

Problem 4-39
a) When k → ∞, the resonance happens, which indicates:
" #1/2
2 2
ωp ωp
2ω 2 (1 − 2 ) − ωc2 sin2 θ ± ωc ωc2 sin4 θ + 4ω 4 (1 − 2 )2 cos2 θ =0
ω ω
. PROBLEM 4-40 23

ωp2 2
⇒ 2ω 2 − 2ωp2 − ωc2 sin2 θ = ±ωc [ωc2 sin4 θ + 4ω 2 (1 − ) cos2 θ]1/2
ω2
ωp2
⇒ (2ω 2 − 2ωp2 − ωc2 sin2 θ)2 = ωc4 sin4 θ + 4ω 2 ωc2 (1 − 2 )2 cos2 θ
ω
⇒ 4ω 4 + 4ωp4 + 4ωc4 sin4 θ − 8ω 2 ωp2 − 4ω 2 ωc2 sin2 θ + 4ωp2 ωc2 sin2 θ
ωp2 ωp4
= ωc4 sin4 θ + 4ω 2 ωc2 (1 − 2 + ) cos2 θ
ω2 ω4
⇒ 4ω 6 + 4ωp4 ω 2 − 8ω 4 ωp2 − 4ω 4 ωc2 sin2 θ + 4ω 2 ωp2 ωc2 sin2 θ
= 4ωc2 ω 4 cos2 θ − 8ω 2 ωc2 ωp2 cos2 θ + 4ωc2 ωp4 cos2 θ
⇒ ω 6 + ωp4 ω 2 − 2ω 4 ωp2 − ωc2 ω 4 + ω 2 ωp2 ωc2 + ωc4 ωp2 cos2 θ − ωc2 ωp4 cos2 θ = 0
⇒ ω 6 − 2(2ωp2 + ωc )ω 4 + ω 2 ωp2 (ωp2 + ωc2 + ωc2 cos2 θ) − ωc2 ωp4 cos2 θ = 0
⇒ (ω 2 − ωp2 )(ω 4 − ωh2 ω 2 + ωp2 ωc2 cos2 θ) = 0, (ωh2 = ωp2 + ωc2 )
1 q
⇒ ω 2 = [ωh2 ± ω 4 − 4ωp2 ωc2 cos2 θ]orω 2 = ωp2
2

b) Whenk → 0 the cut-off happens, and this means that

2
ωp2 ωp2 2 4
ωp2
2ωp2 (1 − ) = 2ω 2
(1 − ) − ωc
2
sin θ ± ωc [ω 2
c sin θ + 4ω 2
(1 − cos2 θ]1/2 ,
ω2 ω2 ω2
and ωp = ω meets the requirement. So the cut off frequency is ωp .

Problem 4-40
Plasma frequenc in the slab is

n0 e2
ωp2 =
mε0

When B0 k E1 , it is an ordinary wave in plasma Dispersion relation is

ω 2 = ωp2 + c2 k 2

2π 2 n0 e2
( ) = + c2 k 2
λ mε0
so the wave vector is
k = 543.72m−1
Say, the nubmer of wavelength is

d d
N= = 2π = 8.6
λ k
24 CHAPTER 3. WAVE IN PLASMA

When B0 ⊥ E1 , is is an extraordianry wave, the dispersion relation is

c2 k 2 ωp2 ω 2 − ωp2
= 1 −
ω2 ω 2 ω 2 − ωh2

ωh2 = ωp2 + ωc2

eB
ωc = = 1.88 × 1011 rad/s
m
n0 e2
ωp2 = = 8.9 × 1021 (rad/s)2
mε0
ωh2 = 4.432 × 1022 (rad/s)2
2πc
ω = = 1.88 × 1011 rad/s
λ
∴ k = 832.2m−1
d
N = = 13.2
λ

Problem 4-41
In the equilibrium state, there is no magnetic or electric field.
B0 = 0, E0 = 0, Te = 0
For the electromagnetic wave, the relevant Maxwell Equation is
∇ × E1 = −Ḃ1
c2 ∇ × B1 = E˙1
→ (ω 2 − c2 k 2 )E1 = −iωj1 /ε0
And
j1 = n0+ Zev0+ − n0− ev0−
= Zn0+ e(v0+ − v0− )
The equation of motion is (Te = 0)
∂v0+ −ZeE1
M+ = ZeE1 ⇒ v0+ = −
∂t iM+ ω
∂v0− eE1
M− = −eE1 ⇒ v0− =−
∂t iM− ω

ZeE1 eE1 Zn0+ e2 Z 1

∴ j1 = −Zn0+ e( + )=− ( + )E
iM+ ω iM− ω iω M+ M−
THe dispersion relation is
Z 2 e2 n0+ e2 n0−
ω 2 − c2 k 2 = +
M+ ε 0 M− ε0
. PROBLEM 4-42 25

Problem 4-42
Boltzman relation is
eφ1
ne1 = n0
kTe
The plasma approximation
ZnA1 + nH1 = ne1
The equation of continuity
iωnA1 = nA ikvA1
iωnH1 = nH ikvH1
And the equation of motion
∂vA1
MA (−iω)vA1 = MA = ZeE1 = Ze(−ikφ1 )
∂t
∂vH1
MH (−iω)vH1 = MH b = eE1 = e(−ikφ1 )
∂t
Then we get
ω2 kTe Z 2 nA nH
= ( + )
k2 n 0 MA MH
The phase velocity is
s
ω kTe Z 2 nA nH
vφ = = ( + )
k n 0 MA MH

Problem 4-43
The Poisson Equation
ε0 ∇ · E1 = n1+ e − Zn1− e
And the continuity equation
k
n1+ = n0 v1+
ω
k
n1− = n0 v1−
ω
Since kT = 0, B0 = 0, there is no collision and magnetic term in the equation
of motion.
∂v+1 −eE1
M+ = eE1 ⇒ v+1 =
∂t iωM+
∂v−1 ZeE1
M− = −ZeE1 ⇒ v−1 =
∂t iωM−
26 CHAPTER 3. WAVE IN PLASMA

Then we get
k −e2 E1 k Ze2 E1
iε0 kE1 = n0 − Z n0
ω iωM+ ω iωM−
That is s
n0+ e2 Z 2 n0− e2
ω= +
M+ ε 0 M − ε0

Problem 4-45
Assume that the ionosphere is extremly cold, we can presume that
kTi = 0
So the sonic ion wave velocity is
r
kTe
vs =
M
And the Alfvenic velocity is
B
vA = √
µ0 M n0
And it indicates that super sonic wave is not super-Alfvenic. That is ,
r
kTe M v2
v> → Te < = 1.2 × 107 K.
M k
This is the upper limit of temperature.
B B
v<√ → n0 < = 4.76 × 1011 /m3 .
µ0 M n0 µ0 M v 2
This is the upper limit of density.

Problem 4-49
a)
ω2 Ω2
ωp2 Ω2p ( ωωc ω2 −ω
p Ωc p
2 − ω ω 2 −Ω2 )
2
c c
+ + 2 2 (3.1)
ω 2 − ωc2 ω 2 − Ω2c ωp
1 − ω2 −ω
Ωp
2 − ω 2 −Ω2
c p

ωc ωp2 Ωc Ω2p
( − )2
ω ω 2 − ωc2 ω ω 2 − Ω2c
ωc2 ωp4 Ω2c Ω4p 2ωc Ωc ωp2 Ω2p (ω 2 − ωc2 )(ω 2 − Ω2c )
= + −
ω 2 (ω 2 − ωc2 )2 ω 2 (ω 2 − Ω2c )2 ω2 (ω 2 − ωc2 )2 (ω 2 − Ω2c )2
ωc2 ωp4 (ω 2 − Ω2c )2 + Ω2c Ω4p (ω 2 − ωc2 )2 − 2ωc Ωc ωp2 Ω2p (ω 2 − ωc2 )(ω 2 − Ω2c )
=
ω 2 (ω 2 − ωc2 )2 (ω 2 − Ω2c )2
. PROBLEM 4-49 27

the denominator of 3.1 is:

ωp2 Ω2p
(1 − − ) · ω 2 (ω 2 − ωc2 )2 (ω 2 − Ω2c )2
ω 2 − ωc2 ω 2 − Ω2c
= ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[(ω 2 − ωc2 )(ω 2 − Ω2c ) − ωp2 (ω 2 − Ω2c ) − Ω2p (ω 2 − ωc2 )]
Now let’s calculate the numerator of Eq3.1
Let:
∆ = (ω 2 − ωc2 )(ω 2 − Ω2c ) − ωp2 (ω 2 − Ω2c ) − Ω2p (ω 2 − ωc2 )
= ω 4 − ω 2 Ω2c − ω 2 ωc2 + ωc2 Ω2c − ωp2 ω 2 + ωp2 Ω2c − Ω2p ω 2 + ωc2 Ω2p
After the reduction of fraction to a common denominator, the numerator in-
ωp2
duced by ω 2 −ωc2 is :
Term1 = ωp2 ω 2 (ω 2 − Ω2c )∆.
Ω2p
And the numerator from ω 2 −Ω2c is :

Term2 = Ω2p ω 2 (ω 2 − ωc2 )∆

ω2 Ωc Ωp
2
Numerator caused by ( ωωc ω2 −ω
p
2 − ω ω 2 −Ω2c )
2
is
c

Term3 = ωc2 ωp4 (ω 2 − Ω2c )2 + Ω2c Ω4p (ω 2 − ωc2 )2 − 2ωc Ωc ωp2 Ω2p (ω 2 − ωc2 )(ω 2 − Ω2c )
Note that
Ω2p ωp2
= ⇒ Ω2p ω = ωp2 Ωc
ωc ωc
⇒ ωc Ωc ωp2 Ω2p = ωc2 Ω4p = ωp4 Ω2c
.

Term3 = ωc2 ωp4 (ω 2 − Ω2c )2 + Ω2c Ω4p (ω 2 − ωc2 )2 − ωp4 Ω2c (ω 2 − ωc2 )(ω − Ω2c ) − ωc2 Ω4p (ω 2 − ωc2 )(ω 2 − Ω2c )
ωc2 ωp4 ω 2 − Ω2c Ω2c Ω4p ω 2 − ωc2 ωp4 Ω2c ωc2 Ω4p
= ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[ 2 2 2
+ 2 2 2
− 2
− ]
ω ω − ωc ω ω − Ωc ω ω2
ωp2
Term1 = ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[ 2 ∆],
ω − ωc2
ω 2 − Ω2c
= ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[ωp2 (ω 2 − Ω2c − Ω2p ) − ωp4 2 ],
ω − ωc2
Ω2p
Term2 = ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[ 2 ∆]
ω − Ω2c
ω 2 − ωc2 4
= ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[Ω2p (ω 2 − ωc2 − ωp2 ) − 2 Ω ],
ω − Ω2c p
Term1 + 2 + 3 = ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[Ω2p (ω 2 − ωc2 − ωp2 ) + ωp2 (ω 2 − Ω2c − Ω2p )
ωc2 ωp4 ω 2 − Ω2c Ω2c Ω4p ω 2 − ωc2 ωp4 Ω2c ωc2 Ω4p ωc2 Ω4p
+ ( 2
− ωp4 ) 2 2
+ ( 2 − Ω4p ) 2 2
− 2
− 2
− ].
ω ω − ωc ω ω − Ωc ω ω ω2
28 CHAPTER 3. WAVE IN PLASMA

Among this ,

ωc2 ωp4 2
4 ω − Ωc
2 Ω2c Ω4p 2
4 ω − ωc
2 ωp4 Ω2c ωc2 Ω4p ωc2 Ω4p
( − ωp ) + ( − Ω p ) − − −
ω2 ω 2 − ωc2 ω2 ω 2 − Ω2c ω2 ω2 ω2
2 2 2 2 2 2 2 2 2 2
ω −ω ω − Ωc 4 Ω −ω ω − ωc 4 Ωc 4 ωc
= ( c 2 )( 2 )ω + ( c 2 )( 2 )Ω − ω − Ωp 4
ω ω − ωc2 p ω ω − Ω2c p ω 2 p ω 2
Ω2 ω2 Ω2 ω2
= −ωp4 + 2c ωp4 − Ω4p + c2 Ω4p − 2c ωp4 − c2 Ω4p
ω ω ω ω
= −ωp4 − Ω4p .

∴ Term1 + 2 + 3 = ω 2 (ω 2 −ωc2 )(ω 2 −Ω2c )[Ω2p (ω 2 −ωc2 −ωp2 )+ωp2 (ω 2 −Ω2c −Ω2p )−ωp4 −Ω4p ]

This is numerator of 3.1. So,

ω2 Ω2
ωp2 Ω2p ( ωωc ω2 −ω
p Ωc p
2 − ω ω 2 −Ω2 )
2
c c
+ 2 + ωp2 Ω2p
ω 2 − ωc2 ω − Ω2c 1 − ω2 −ω 2 − ω 2 −Ω2
c p

ω (ω − ωc2 )(ω 2 − Ω2c )[Ω2p ω 2 − Ω2p ωc2 − Ω2p ωp2 + ωp2 ω 2 − ωp2 Ω2c − ωp2 Ω2p − ωp4 − Ω4p ]
2 2
=
ω 2 (ω 2 − ωc2 )(ω 2 − Ω2c )[ω 4 − ω 2 Ω2c − ω 2 ωc2 + ωc2 Ω2c − ωp2 ω 2 + ωp2 Ω2c − Ω2p ω 2 + ωc2 Ω2p ]
Ω2p ω 2 − Ω2p ωc2 − Ω2p ωp2 + ωp2 ω 2 − ωp2 Ω2c − ωp2 Ω2p − ωp4 − Ω4p
=
ω 4 − ω 2 Ω2c − ω 2 ωc2 + ωc2 Ω2c − ωp2 ω 2 + ωp2 Ω2c − Ω2p ω 2 + ωc2 Ω2p

Let
ωp2 = ωp2 + Ω2p

And

ωp2
ω 2 (ωc −Ωc )2
ω 2 − ωc Ωc +
ωp2 −ω 2 +ωc Ωc

ωp2 (ωp2 − ω 2 + ωc Ωc )
=
(ω 2 − ωc Ωc )(ωp2 − ω 2 + ωc Ωc ) + ω 2 (ωc − Ωc )2
(ωp2 + Ω2p )(ωp2 + Ω2p − ω 2 + ωc Ωc )
= . (3.2)
(ω 2 − ωc Ωc )(ωp2 + Ω2p − ω 2 + ωc Ωc ) + ω 2 (ωc − Ωc )2

Recall
Ω2p ωc = ωp2 Ωc

Then the numerator of Eq.3.2 is

ωp4 + ωp2 Ω2P − ω 2 ωp2 + ωp2 ωc Ωc + Ω2p ωp2 + Ω4p − ω 2 Ω2p + Ω2p ωc Ωc
= ωp4 + ωp2 Ω2P − ω 2 ωp2 + Ω2p ωc2 + Ω2p ωp2 + Ω4p − ω 2 Ω2p + ωp2 Ω2c ,
. PROBLEM 4-50 29

which equals to that of Eq.3.1.

The denominator of Eq.3.2:

ω 2 ωp2 + ω 2 Ω2p − ω 4 + ω 2 ωc Ωc − ωp2 ωc Ωc + Ω2p ωc Ωc

+ω 2 ωc Ωc − ωc2 Ω2c + ω 2 ωc2 − 2ω 2 ωc Ωc + ω 2 Ω2c
= ω 2 ωp2 + ω 2 Ω2p − ω 4 + 0 − Ω2p ωc2 + ωp2 Ω2c + 0 − ωc2 Ω2c + ω 2 ωc2 − 0 + ω 2 Ω2c
= −ω 4 + ω 2 Ω2c + ω 2 ωc2 − ωc2 Ω2c + ωp2 ω 2 − ωp2 Ω2c + Ω2p ω 2 − ωc2 Ω2p ,

which equals to that of Eq.3.2.

Therefore, they are identical.
Q.E.D.
b) (Incomplete solution)For cut-off, k → 0

Ω2p ω 2 − Ω2p ωc2 − Ω2p ωp2 + ωp2 ω 2 − ωp2 Ω2c − ωp2 Ω2p − ωp4 − Ω4p
= ω 4 − ω 2 Ω2c − ω 2 ωc2 + ωc2 Ω2c − ωp2 ω 2 + ωp2 Ω2c − Ω2p ω 2 + ωc2 Ω2p
⇒ ω 4 − (Ω2c + ωc2 + 2ωp2 + 2Ω2p )ω 2 + ωc2 Ω2c + 2Ω2p ωc2 + 2ωp2 Ω2c + 2Ω2p ωp2 + ωp4 + Ω4p = 0

1 q
ω2 = [(Ω2c + ωc2 + 2ωp2 + 2Ω2p ) ± (Ω2c + ωc2 )2 + 4(Ω2p − ωp2 )(Ω2c − ωc2 )]
2
The left hand cut off frequency is

2 1 q
ωL = [(Ω2c + ωc2 + 2ωp2 + 2Ω2p ) − (Ω2c + ωc2 )2 + 4(Ω2p − ωp2 )(Ω2c − ωc2 )]
2
(c)

Figure 3.1: Figure for 4-49(c)

Problem 4-50
a)
30 CHAPTER 3. WAVE IN PLASMA

b)
ωc2 Ω2p ωc2 mωp2
ωl2 = = ,
ωc2 + ωp2 M (ωc2 + ωp2 )

1 q
ωL = [−ωc + ωc2 + 4ωp2 ],
2

r ωc
ωp = (1 − ) .
a 2
And

m ωc2 1 2 m 2
ω2 = ωc2 (r = 0) = ωc = ω
2 1 2
M ωc + 4 ωc 4 5M c
√
2−1
ωL (r = 0) = ωc
2
ωL (r = a) = 0,

where ωl < ω < ωL lays in the evanescent layer.

r
m
ω = ωl (r = 0) = ωc
5M
m (1 − ar )2 )ωc2
ωl2 =
M 4 + (1 − ar )2
1 r
ωL = [−ωc + ωc (1 + (1 − )1/2 )]
2 a

when ω > ωl → r > 0

when ω < ωL → r < 0.975a
Note
m
= 1836
M
So the evanescent layer thickness is 0.975a.
. PROBLEM 4-50 31

c)
r
ωp = (1 − )2ωc ,
a
m 4m 2
ω2 = ωl2 (r = 0) = 4ωc2 = ω ,
r M 5M c
ωc r
ωL = [−1 + 1 + 16(1 − )2 ],
2 a
m 16(1 − ar )2
ωl2 = ω2 .
M 1 + 16(1 − ar )2 c

Similarly, when ω > ωl , r > 0 ; when ω < ωL , r < 0.927a .

Conclusion: the higher center density is, the thicker evanescent layer will be.