Quantum Mechanics for Nanostructures: Solutions

Manual

Vladimir V. Mitin, Dmitry I. Sementsov, Nizami Z. Vagidov

2011
Contents

1 The nanoworld and quantum physics 1

2 Wave-particle duality and its manifestation in radiation and particle

behavior 3

3 Layered nanostructures as the simplest systems to study electron be-

havior in a one-dimensional potential 11

4 Additional examples of quantized motion 25

5 Approximate methods of finding quantum states 37

6 Quantum states in atoms and molecules 49

7 Quantization in nanostructures 59

A Classical dynamics of particles and waves 75

B Electromagnetic fields and waves 91

C Crystals as atomic lattices 105

iii
Chapter 2

Wave-particle duality and its

manifestation in radiation and
particle behavior

Problem 2.1. Using Planck’s formula, find the most probable frequency, ωprob , and
mean frequency, ωmean , in the blackbody radiation spectrum at T = 2000 K. (Answer:
ωprob ≈ 7.4 × 1014 s−1 and ωmean ≈ 1015 s−1 .)

Solution. Planck’s formula for spectral energy density of blackbody radiation is de-
fined as:
ω2 h̄ω
uω = 2 3 h̄ω . (2.1)
π c e kB T − 1
The most probable value of frequency in radiation spectrum corresponds to maximum
of spectral density. Therefore:
duω
= 0, (2.2)
dω
h̄ω h̄ω kh̄ωT
3e kB T − 3 − e B = 0, (2.3)
kB T
or
h̄ω
3 − 3e−x − x = 0 , where x = . (2.4)
kB T
Numerical solution of this equation gives the root:

xprob ≈ 2.82 , (2.5)

kB T
ωprob ≈ 2.82 ≈ 7.4 × 1014 s−1 . (2.6)
h̄
The mean frequency in the emission spectrum is defined as

3
4CHAPTER 2. WAVE-PARTICLE DUALITY AND ITS MANIFESTATION IN RADIATION

 h̄ω
−1
R∞ 4
0 ω e −1 dω
R∞ kB T
ωum dω kB T
ωmean = R0 ∞ = −1 ≈ 3.83 ≈ 1015 s−1 . (2.7)
u dω h̄
 h̄ω
0 m R ∞ 3
0 ω e kB T − 1 dω

To calculate the respective integrals in Eq. (2.7) the following integrals must be used:

xn dx
(
∞ π 2 /15, for n = 3 ,
Z
= (2.8)
0 ex − 1 29.4, for n = 4 .

Problem 2.2. Monochromatic radiation of wavelength λ = 0.55 µm is incident on the

photosensitive surface of the photoelectric cell and it liberates Nph = 103 electrons from
the cell. The sensitivity of the photoelectric cell, which is defined as γ = Nph e/(Ni h̄ω),
is equal to γ = 10 mA W−1 . Estimate the number of photons, Ni , incident on the
surface of the photoelectric cell. (Answer: Ni ≈ 4.4 × 104 .)

Solution. We assume that the emission process and subsequent occurence of pho-
tocurrent to be practically noninertial and last for a time period ∆t. The energy of
incident radiation during this time is defined by the expression:

2πh̄c
ε = Ni h̄ω = Ni
, (2.9)
λ
where Ni is number of photons incident on the photosensitive surface during time ∆t.
The current, I, can be defined as:
Nph e γε γ 2πh̄c
I= = = Ni , (2.10)
∆t ∆t ∆t λ
where Nph is the number of liberated electrons. From the last equation we obtain the
expression for number of incident photons, Ni :
eλ
Ni = Nph ≈ 4.4 × 104 . (2.11)
2πh̄cγ

Problem 2.3. An isolated small iron ball of radius r = 1 mm is exposed to monochro-

matic light with wavelength λ = 0.2 µm. What is the maximum potential, ϕm , of the
ball’s charging and how many electrons, Ne , will it emit? The work function of iron is
equal to Awf = 4.36 eV. (Answer: ϕm = 1.85 V and Ne ≈ 1.3 × 106 .)

Solution. As a result of the external photoelectric effect metal sphere releases electrons
and therefore becomes positively charged. The maximum potential of the iron sphere
 5

is found from the law of conservation of energy according which the energy of photon
is spent on the work function of electron and charging of the ball. For the last emitted
photoelectron we can write:
2πh̄c
eϕm + Awf = h̄ω = , (2.12)
λ
!
1 2πh̄c
ϕm = − Awf ≈ 1.85 V . (2.13)
e λ
The potential of the iron ball can be found from the following expression:
q eNe
ϕm = = , (2.14)
4πǫ0 r 4πǫ0 r
which gives us the expression for the number of emitted electrons, Ne :
4πǫ0 rϕm
Ne = ≈ 1.3 × 106 . (2.15)
e

Problem 2.4. The cut-off wavelength for a given metal is λc = 0.62 µm. How much
larger is the highest kinetic energy, Kmax , of photoelectrons emitted by the metal
surface after its exposure to light of wavelength λ = 0.35 µm than the mean thermal
electron energy, KT . Assume room temperature conditions. (Answer: Kmax /KT ≈ 40.)

Solution. Einstein’s equation for the photoelectric effect:

2
me vmax
h̄ω = Awf + = Awf + Kmax . (2.16)
2
Taking into account that ω = 2πc/λ we can rewrite the last equation in the form:
2πc
h̄ = Awf + Kmax . (2.17)
λ
From the last equation we find the cut-off wavelength, λc , for a given metal by setting
Kmax = 0:
2πc
λc = h̄ . (2.18)
Awf
Therefore,
2πc 2πc 1 1
 
Kmax = h̄ − h̄ = 2πch̄ − . (2.19)
λ λc λ λc
The mean kinetic energy of electron thermal motion is:
3
KT = kB T . (2.20)
2
6CHAPTER 2. WAVE-PARTICLE DUALITY AND ITS MANIFESTATION IN RADIATION

Then, the ratio of corresponding energies is equal to:

Kmax 4πch̄ 1 1
 
= − ≈ 40 . (2.21)
KT 3kB T λ λc

Problem 2.5. A laser beam with wavelength λ = 0.35 µm, power N = 5 W, and
cross-sectional diameter d = 200 µm is perpendicularly incident on a cube face with
side length a = 100 µm. Find the number of incident photons per unit time, n, and
the pressure, P , and force, F , that they create. (Answer: n ≈ 2.8 × 1018 , P ≈ 1.1 Pa,
and F ≈ 1.1 × 10−8 N.)

Solution. The light pressure at its normal incidence on the surface illuminated by the
beam is defined as:
I
P = (1 + R) , (2.22)
c
where I is the intensity of the light beam and R is the reflection coefficient, which for
the mirror surface is equal to R = 1. In our case I is defined as:
4N
I= . (2.23)
πd2
By substituting last equation into Eq. (2.22), we obtain:

4N
P = (1 + R) ≈ 1.1 Pa . (2.24)
πd2 c
The intensity, I, is connected with the number of photons incident during time unit at
surface unit, n0 , as:
2πh̄c
I = n0 εph = n0 , (2.25)
λ
where εph is the energy of the photon. The total number of photons incident per unit
time can be found from:
Iλa2 4N a2 λ
 2
2N λ a
n = n0 a2 = = 2
= 2 ≈ 2.8 × 1018 s−1 . (2.26)
2πh̄c πd 2πh̄c π h̄c d
The force that the laser beam exerts on the sample is:
4N
F = P a2 = 2
(1 + R)a2 ≈ 1.1 × 10−8 N . (2.27)
πd c

Problem 2.6. In a homogeneous magnetic field with magnetic flux density B = 0.1
T an electron gyrates in a circle with a radius 100 times greater than the electron de
 7

Broglie wavelength. Calculate the radius, r, and the electron de Broglie wavelength,
λBr . (Answer: r ≈ 0.8 µm and λBr ≈ 8 nm.)

Solution. The de Broglie wavelength is defined as:

h 2πh̄
λBr = = , (2.28)
p me v
where me and v are electron mass and velocity, respectively. The electron motion along
the circle in magnetic field is caused by the Lorentz force:

FL = evB . (2.29)

This force is perpendicular to electron velocity and is directed to the center of the
circle. This force gives to electron centropetal acceleration:
v2
an = . (2.30)
r
Since,
F L = m e an , (2.31)
then,
v2
me = evB . (2.32)
r
From the last expression we find
eBr
v= . (2.33)
me
Taking into account this expression we find:
2πh̄
λBr = (2.34)
eBr
or s
2πh̄
λBr = , (2.35)
eBη
where η = r/λBr = 100. Thus, we obtain: λBr ≈ 8 nm and r ≈ 0.8 µm.

Problem 2.7. A parallel array of electrons is perpendicularly incident on a narrow

slit of width b = 5 µm. Electrons are accelerated by a voltage U = 100 V. Find the
distance, ∆X, between two maxima of the first order in the diffraction picture. The
screen is located at a distance L = 20 cm from the slit. (Answer: ∆X ≈ 14.8 µm.)

Solution. The de Broglie wavelength of electrons in the beam is defined as:

h 2πh̄
λBr = = . (2.36)
p me v
8CHAPTER 2. WAVE-PARTICLE DUALITY AND ITS MANIFESTATION IN RADIATION

Figure 2.1: Single-slit diffraction pattern.

The electron velocity in the beam can be found from

me v 2
eU = , (2.37)
2
s
2eU
v= ≈ 5.9 × 106 m/s . (2.38)
me
Thus, electrons are non-relativistic and the above-mentioned equations adequately de-
scribe the considered phenomenon.

The position of diffraction maxima can be found from the well-known condition for
light diffraction:
λ
b sin θn = ±(2n + 1) , (2.39)
2
where n = 0, 1, 2, ... is the order of the diffraction maxima. For electrons we substitute
λ by λBr . Thus, for n = 1 we have:

3λBr
sin θ1 = ± . (2.40)
2b
The distance on the screen between these two maxima (see Fig. 2.1) is given by the
expression:

3LλBr
∆X = X+1 − X−1 = 2L tan θ1 ≈ 2L sin θ1 = . (2.41)
b
In the last expression we took into account that for small θ we can use the approxima-
tion:
tan θ1 ≈ sin θ1 . (2.42)
 9

Figure 2.2: The dependence of the real part and of the square of modulus of the
wavefunction on the coordinate x.

By substituting the expression for the electron de Broglie wavelength into Eq. (2.41)
we find:
3L 2πh̄ 6πh̄L
∆X = = √ ≈ 14.8 µm . (2.43)
b me v b 2eme U

Problem 2.8. The wavefunction of a particle moving along the x−axis has the form
of Eq. (2.149) from the textbook. Considering all parameters of this function as real
parameters, find the dependence on coordinate of the real part of this function, Re
{Ψ}, and the square of its modulus, |Ψ|2 , at time t = t0 .

Solution. The wavefunction is:

i 2
Ψ(x, t) = Ce h̄ (px x−Et)−αx . (2.44)
For the real part of the wavefunction at t = t0 we can write:
i i
h 2
i 2
h i
ReΨ(x, t0 ) = Re Ce h̄ (px x−Et0 )−αx = Ce−αx Re e h̄ (px x−Et0 ) . (2.45)
Using Euler’s formula for the complex numbers we obtain:
px x − Et0
 
−αx2
ReΨ(x, t0 ) = Ce cos . (2.46)
h̄
For the wavefunction’s square of modulus we can write the following expression:
2
|Ψ(x, t0 )|2 = Ψ(x, t0 )Ψ∗ (x, t0 ) = C 2 e−2αx . (2.47)
The Eqs. (2.46) and (2.47) are plotted in Fig. 2.2. Such a form is characteristic for the
wave packet.
Chapter 3

Layered nanostructures as the

simplest systems to study electron
behavior in a one-dimensional
potential

Problem 3.1. Using the procedure of calculations of average values of physical quan-
tities, show that for a one-dimensional motion the operator of coordinate is coordinate
itself.

Solution. If the wavefunction ψ(x) describes the state of one-dimensional motion of

a particle, then the square of its modulus defines the probability of finding the particle
at the point with coordinate x. The average value of the coordinate:
Z Z Z
< x >= x|ψ(x)|2 dx = xψ ∗ (x)ψ(x)dx = ψ ∗ (x)xψ(x)dx . (3.1)

By analogy the average value of any dynamical variable, to which an operator  cor-
responds, is defined as:
Z
< A >= ψ ∗ (x)Âψ(x)dx . (3.2)

Comparing the last expression with Eq. (3.1) we see that the operator of coordinate
acts on the wavefunction as a multiplicator. Thus, x̂ = x.

Problem 3.2. A particle with mass m is moving in the region −L/2 ≤ x ≤ L/2. The
state of the particle is described by the wavefunction:

i
 
ψ(x) = A exp px , (3.3)
h̄

11
12CHAPTER 3. LAYERED NANOSTRUCTURES AS THE SIMPLEST SYSTEMS TO STUD

where p is the particle’s momentum. Find the normalization constant, A, and the
average value of the particle’s kinetic energy, < K >.

Solution. Let us find the normalization constant, A. The normalization condition for
the particle’s wavefunction has the form:
Z L/2
|ψ(x)|2 dx = 1 . (3.4)
−L/2

Substituting the expression for the wavefunction, ψ(x), we obtain the expression for
the normalization constant, A:
1
A= √ . (3.5)
L
The operator of kinetic energy of particle’s one-dimensional motion has the form:

h̄2 ∂ 2
K̂ = − . (3.6)
2m ∂x2
The average value of particle’s kinetic energy is given by:

1 h̄2 Z L/2 − i px ∂ 2
Z !
i
∗
< K >= ψ (x)K̂ψ(x)dx = − e h̄ 2
e h̄ px dx =
L 2m −L/2 ∂x
1 h̄2 Z L/2 p2
= dx . (3.7)
L 2m −L/2 h̄2
Carrying out the integration we find:

p2
< K >= . (3.8)
2m

Problem 3.3. An electron is in a one-dimensional rectangular potential well with

barriers of infinite height. Find the well width, L, if the energy difference between the
fifth and fourth quantum states, E5 − E4 , is equal to the electron’s average thermal
energy at room temperature.

Solution. Electron energy eigenvalues in a one-dimensional rectangular potential well

with infinite barriers are given by:
!2
1 πh̄n
En = . (3.9)
2me L

Using this expression we find the energy interval between two neighboring states:

π 2 h̄2
En+1 − En = (2n + 1) . (3.10)
2me L2
 13

According to the formulation of the problem we have:

En+1 − En =< ET > , (3.11)

which gives:
π 2 h̄2 3
(2n + 1) 2
= kB T . (3.12)
2me L 2
From the last equation for the width of the potential well, L, and n= 4 we get:
s
2n + 1
L = πh̄ ≈ 9.35 nm . (3.13)
3kB T me

Problem 3.4. An electron is in a one-dimensional rectangular potential well with

barriers of infinite height. The width of the well is equal to L = 5 nm. Find the
wavelengths of photons emitted during electronic transitions from the excited states
with quantum numbers n = 2, λ21 , and n = 3, λ31 , to the ground state with n = 1.
(Answer: λ21 ≈ 1.15 µm and λ31 ≈ 0.43 µm.)

Solution. From energy conservation we get:

2πh̄c
h̄ωmn = = Em − En . (3.14)
λmn
From this expression we find the photon’s wavelength:

2πh̄c
λmn = . (3.15)
Em − En
The energy of electron in the n-th quantum state is defined as:
!2
1 πh̄n
En = . (3.16)
2me L
By substituting last equation into Eq. (3.15), we can write the expression for the
wavelength of the emitted photon as:

4cme L2 12 L2
λmn = ≈ 3.45 × 10 × m. (3.17)
πh̄(m2 − n2 ) m 2 − n2

Finally, for n = 2, n = 3, and m = 1, we have:

λ21 = 1.15 µm , (3.18)

and
λ31 = 0.43 µm . (3.19)
14CHAPTER 3. LAYERED NANOSTRUCTURES AS THE SIMPLEST SYSTEMS TO STUD

Figure 3.1: Rectangular symmetric potential well with barriers of infinite height.

Problem 3.5. An electron is in a one-dimensional rectangular potential well with

barriers of infinite height. Find the normalized wavefunctions and energy levels if the
width of the potential well is equal to L and the wavefunction is symmetric with respect
to the coordinate origin placed in the center of the well (see Fig. 3.1.)

Solution. Since outside the well (|x| > L/2) the electron potential energy U (x) → ∞,
the electron cannot be found outside the well. Therefore, its wavefunction at the
boundaries of the well is equal to zero, i.e.
L L
   
ψ − =ψ = 0. (3.20)
2 2
In the limits of the quantum well (−L/2 ≤ x ≤ L/2) the potential energy is equal to
zero, i.e. U (x) = 0. Then, the Schrödinger equation takes the form:

d2 ψ 2me E
+ ψ=0 (3.21)
dx2 h̄2
or
d2 ψ
2
+ k2ψ = 0 , (3.22)
dx
2me E
where k 2 = h̄2
.
 15

The general solution of this equation can be written as:

ψ(x) = A sin(kx) + B cos(kx) . (3.23)

Since the potential energy, U (x), is symmetric (i.e. it does not change when x is
substituted by −x), all stationary states must be described either by even or by odd
wavefunctions. Therefore, the solutions of the Schrödinger equation are:

ψa (x) = A sin(ka x) , (3.24)

ψs (x) = B cos(ks x) . (3.25)

For the wavefunction ψa (x) the boundary condition

ka L
A sin =0 (3.26)
2
gives the following values of the wavenumber ka :

ka L = 2mπ , (3.27)

where m = 1, 2, ....
For the wavefunction ψs (x) the boundary condition

ks L
B cos =0 (3.28)
2
gives the following values of the wavenumber ks :

ks L = (2m − 1)π , (3.29)

where m = 1, 2, .... The expressions for the wavenumbers can written using single
expression:
kL = nπ . (3.30)
For even n, i.e. n = 2m, this expression defines the wavenumbers ka and states with
the wavefunctions ψa (x). For odd n, i.e. n = 2m − 1, this expression defines the
wavenumbers ks and states with the wavefunctions ψs (x). The expression for the
energy levels can be written as:
!2
1 πh̄n
En = , n = 1, 2, 3, ... (3.31)
2me L

For the wavefunction ψa (x) the normalization constant, A, can be found from the
following condition:

2
Z L/2
2 A2 Z L/2
A sin (ka x)dx = [1 − cos(2ka x)]dx = 1 , (3.32)
−L/2 2 −L/2
16CHAPTER 3. LAYERED NANOSTRUCTURES AS THE SIMPLEST SYSTEMS TO STUD

L
A2 = 1, (3.33)
2
s
2
, A= (3.34)
L
where we took into account that ka L = 2mπ. In a similar fashion the normalization
constant, B, for ψs (x) is defined as:
s
2
B= . (3.35)
L
Thus, the normalized wavefunctions, ψa and ψs , can be written as:
s
2
ψa (x) = sin(ka x) (3.36)
L
and s
2
ψs (x) = cos(ks x) . (3.37)
L

Problem 3.6. An electron is in a one-dimensional rectangular potential well with

barriers of infinite height and with well width equal to L shown in Fig. 3.1. Find the
average value of the coordinate, the momentum projection, and the squares of these
magnitudes for an electron that is in quantum states described by symmetric wave-
functions.

Solution. The average value of electron coordinate can be calculated using the equa-
tion: Z L/2
< x >= x|ψ(x)|2 dx . (3.38)
−L/2

Let us assume that electron is in a symmetric, with respect to the coordinate origin,
potential well. For the states with the wavefunction
s
2 πnx
ψ(x) = cos , n = 1, 3, 5, ... (3.39)
L L
we find Z L/2
∗ 2 Z L/2 πnx
< x >= ψ (x)x̂ψ(x)dx = x cos2 dx =
−L/2 L −L/2 L
1 Z L/2 2πnx
 
= x 1 + cos dx = 0 . (3.40)
L −L/2 L
The average value of electron momentum projection, px , can be found from:
!
Z L/2
∗ 2 Z L/2 πnx ∂ πnx
< px >= ψ (x)pˆx ψ(x)dx = cos −ih̄ cos dx =
−L/2 L −L/2 L ∂x L
 17

πn Z L/2 2πnx
= ih̄ sin dx = 0 . (3.41)
L2 −L/2 L
This result is plausible since for the electron in the potential well the probabilities of
motion along and opposite of x-axis are equal to each other. The average value of the
electron coordinate square can be found in a similar fashion:
Z L/2 2 Z L/2 2 πnx
< x2 >= x2 |ψ(x)|2 dx = x cos2 dx =
−L/2 L −L/2 L
1 Z L/2 2 2πnx
 
= x 1 + cos dx . (3.42)
L −L/2 L
Carrying out integration in the last expression we get:
1 1 L2
 
< x2 >= − 2 2 . (3.43)
3 π n 4
Finally, we find the average value of electron momentum projection square, < p2x >:
!2
2 Z L/2 πnx ∂ πnx
< p2x >= cos −ih̄ cos dx =
L −L/2 L ∂x L
!2 Z !2
2 πnh̄ L/2 πnx 2 πnh̄
= cos dx = . (3.44)
L L −L/2 L L

Problem 3.7. An electron is in a one-dimensional rectangular potential well with

width equal to L. One of the barriers has infinite height: U (x) → ∞ at x < 0.
The height of the second barrier at x > L is finite and is equal to U0 . In the region
0 ≤ x ≤ L the potential energy is equal zero: U (x) = 0 (see Fig. 3.2). What are the
conditions on the value U0 L2 for there to be at least one energy level or only one level
in the well? If the energy level is equal to E = U0 /2, how does this condition change?

Solution. The equation that defines corresponding energy levels in such a non-
symmetric potential well has the following form (see in the textbook Eq. (3.77), its
corresponding graphical analysis, and Fig. 3.3):
h̄
sin k1 L = ± √ k1 L = ±γk1 L , (3.45)
2me U0 L
where √
2me E
k1 = . (3.46)
h̄
The analysis of Eq. (3.45) shows that there is at least one root (therefore one energy
level in the well) when the following inequality is satisfied:
h̄ 2
γ=√ < . (3.47)
2me U0 L π
18CHAPTER 3. LAYERED NANOSTRUCTURES AS THE SIMPLEST SYSTEMS TO STUD

Figure 3.2: Rectangular non-symmetric potential well with left barrier of infinite height
and right barrier of finite height.

Figure 3.3: Graphical solution of the transcendental equation (3.45).

 19

From the last equation we obtain:

2 π 2 h̄2
U0 L > . (3.48)
8me
The root of Eq. (3.45) is unique only if the following inequalities are satisfied:
2 2
<γ≤ . (3.49)
3π π
If the energy level is the following:
U0
E= , (3.50)
2
then Eq. (3.45) takes the form:
√ √ √
2me EL h̄ 2me EL 2
sin k1 L = γ =√ = . (3.51)
h̄ 2me U0 L h̄ 2
From Eq. (3.51) we find:
3π
k1 L =
. (3.52)
4
Equations (3.51) and (3.52) give us the parameter
√
h̄ 2 2
γ=√ = , (3.53)
2me U0 L 3π
which satisfies the condition of uniqueness of the root (3.49). Thus,
9π 2 h̄2
U0 L2 = . (3.54)
16me

Problem 3.8. An electron is in a symmetric rectangular potential well with width

L = 10 nm and with barriers of finite height: U0 = 10 eV. The electron wavefunction
is symmetric with respect to the coordinate origin placed at the center of the well (see
Fig. 3.4), in contrast to the case shown in Fig. 3.5 from the textbook. For which val-
ues of quantum numbers, n, can we apply the approximation of barriers with infinite
height? (Answer: n ≪ 52.)

Solution. The condition when the barrier height can be considered as infinite is:
En << U0 . (3.55)
For the electron energy levels in the potenial well with infinite barriers, the condition
(3.55) can be rewritten as:
!2
1 πh̄n
En = << U0 . (3.56)
2me L
20CHAPTER 3. LAYERED NANOSTRUCTURES AS THE SIMPLEST SYSTEMS TO STUD

Figure 3.4: Rectangular, symmetric with respect to coordinate origin, potential well
with finite barriers.

Then, from the last equation we can write the condition when approximation of the well
with infinite barriers can be applied to a potential well with barriers of finite height:
2
L

2
n << 2me U0 (3.57)
πh̄
or
Lq
n << 2me U0 . (3.58)
πh̄
Thus, substituting the given values we obtain: n << 52.

Problem 3.9. An electron with energy E = 32 eV, which is moving in the positive
direction along the x-direction, encounters an obstacle - an infinitely long rectangular
potential step. Find the reflection, R, and transmission, D, coefficients of the electron
de Broglie waves for a given potential step in two cases: (a) the height of the potential
step is lower than the electron energy (U1 = 30 eV) and (b) the height of the potential
step is higher than the electron energy (U2 = 34 eV). (Answer: (a) R ≈ 0.36 and
D ≈ 0.64; (b) R ≈ 1.0 and D ≈ 0.)

Solution. In the first case (E > U1 ) electron moves as a free particle. In the region 1
before the potential step (see Fig. 3.5) the electron wavefunction can be written as:
ψ1 (x) = A1 eik1 x + B1 e−ik1 x , (3.59)
√
where k1 = 2me E/h̄. In the region 2 of potential step the wavefunction is:
ψ2 (x) = A2 eik2 x , (3.60)
 21

Figure 3.5: Reflection of the electron from the potential step with U = U0 .
q
where k2 = 2me (E − U1 )/h̄. The general expression for the reflection coefficient in
this case has the form:
2
B1 2 k1 − k2
R= = . (3.61)
A1 k1 + k2
Since k1 and k2 are real numbers, then
√ q 2  q 2
2me E − 2me (E − U1 ) 1 − 1 − U1 /E
R = √ q  = q  . (3.62)
2me E + 2me (E − U1 ) 1 + 1 − U1 /E

Since R + D = 1, then
q
4 1 − U1 /E
D =1−R= q . (3.63)
(1 + 1 − U1 /E)2

Thus, at U1 /E = 30/32, we get R = 0.36 and D = 0.64.

In the second case (E < U2 ) the electron wavefunction in the region 1 before the
potential step has the form:

ψ1 (x) = A1 eik1 x + B1 e−ik1 x , (3.64)

√
where k1 = 2me E/h̄. In the region of the potential step 2 the wavevector k becomes
imaginary since
q q
2me (E − U2 ) 2me (U2 − E)
k2 = =i = iβ2 . (3.65)
h̄ h̄
22CHAPTER 3. LAYERED NANOSTRUCTURES AS THE SIMPLEST SYSTEMS TO STUD

 
x2
Figure 3.6: Electron tunneling through the potential barrier U (x) = U0 1 − L2
.

Then, the wavefunction is defined as:

ψ2 (x) = A2 e−β2 x . (3.66)

In this case the reflection coefficient is always equal to unity since:

2 2
B1 k1 − iβ2
R= = = 1. (3.67)
A1 k1 + iβ2

Problem 3.10. Estimate the transmission coefficient, D, of an electron with energy

E = 18 eV for passage through a potential barrier with the following dependence of
potential energy on coordinate:

x2
!
U (x) = U0 1− 2 (3.68)
L

in the region |x| ≤ L and

U (x) = 0 (3.69)
in the region |x| > L. Here, U0 = 20 eV and L = 2 nm. (Answer: D ≈ 0.24.)

Solution. To estimate the probability of electron transmission through the consid-

ered potential barrier (see Fig. 3.6) we will use the approximate expression for the
transmission coefficient: R x2 √
− h̄2 2me [U (x)−E]
D≈e x1
dx , (3.70)
 23

where x1 and x2 are the points where electron enters the barrier and leaves the barrier,
respectively. Let us find these points solving the equation U (x) = E:

x2
!
U0 1− 2 =E, (3.71)
L
s
E
x1,2 = ±x0 = ±L 1 − . (3.72)
U0
Taking into account Eq. (3.72) we can make the following transformation:
v " ! v# !
x2 x2
u u
u u
t2m U 1− 2 − E = 2me (U0 − E) 1 − 2 .
t (3.73)
e 0
L x0

Let us carry out integration in Eq. (3.70):

√
r h   i
R x0 2 R x0 q 2
− h̄2 2me U0 1− x 2 −E dx − h̄2 2me (U0 −E) 1− x2 dx
−x0 L x
D≈e =e −x0
0 =
√ q q
2me
− h̄ 2
2me (U0 −E) πL
2
1− UE − πL
h̄ U0
(U0 −E)
=e 0
=e . (3.74)
For U0 = 20 eV, E = 18 eV, and L = 2 nm we find: D ≈ e−1.44 ≈ 0.24. The probability
of electron transmission through the barrier is equal to D since it is introduced in the
same way as the transmission coefficient.
Chapter 4

Additional examples of quantized

motion

Problem 4.1. A free electron of mass me , moves in the direction defined by the polar
angle θ and the azimuthal angle ϕ (see Fig. 4.1). Find the electron wavefunction Ψ(r, t)
if the electron energy is equal to ε.

Solution. The wavefunction of the free electron in this case can be written as:
i i
Ψ(r, t) = ψ(r)e− h̄ εt = Ae h̄ (p·r−εt) , (4.1)

where the scalar-product of vectors p and r is defined as follows:

p · r = px x + py y + pz z . (4.2)

The components of vector p can be expressed using the angles θ and ϕ (see Fig. 4.1):

px = p sin θ cos ϕ ,

py = p sin θ sin ϕ , (4.3)

pz = p cos θ ,
where the momentum magnitude, p, is related with energy, ε, as follows:
√
p = |p| = 2me ε . (4.4)

After substitution of Eqs. (4.2) - (4.4) into Eq. (4.1), the wavefunction for the free
electron can be written as:
i √
 
Ψ(x, y, z, t) = A exp 2me ε(x sin θ cos ϕ + y sin θ sin ϕ + z cos θ) − εt . (4.5)
h̄

25
26 CHAPTER 4. ADDITIONAL EXAMPLES OF QUANTIZED MOTION

Figure 4.1: Momentum p in spherical system of coordinates.

Problem 4.2. The dispersion equation Eq. (4.61) from the textbook
√ ! s
2me Ea U0
cot =− −1 (4.6)
h̄ E

defines the allowed energy states of an electron in a spherically-symmetric potential

well and in a general case it has to be solved numerically. Find the energy of stationary
states in the limit of infinitely high potential barriers.

Solution. In the limiting case when U0 → ∞ the expression under the root in the
right-hand side of Eq. (4.6) also tends to infinity. Therefore, we can assume that
√ !
2me Ea
cot → −∞ . (4.7)
h̄

From Eq. (4.7) we find that

√
2me Ea
→ ±nπ , n = 1, 2, 3, .... (4.8)
h̄
The last equation gives the following expression for energy of quantum states:

π 2 h̄2 2
En = n . (4.9)
2me a2

This expression describes reasonably well the lowest energy levels if En /U0 ≪ 1.

Problem 4.3. Find the energy levels and the wavefunctions of a positively charged
harmonic oscillator in a homogeneous electric field E. The charge of the oscillating
 27

particle is e and its mass is m.

Solution. In this problem the particle is under the influence of a constant electric
force Fe = eE. Its total potential energy is equal to:

mω 2 2
U (x) = x − e|E|x . (4.10)
2
Equation (4.10) can rewritten as:
!2
mω 2 e|E| e2 |E|2
U (x) = x− − . (4.11)
2 mω 2 2mω 2

Then, the Schrödinger equation can be reduced to the equation for a one-dimensional
harmonic oscillator:
d2 ψ(x1 ) 2m mω 2 x21
!
+ ε 1 − ψ(x1 ) = 0 , (4.12)
dx21 h̄2 2

where we have introduced a new coordinate and energy of a quantum oscillator:

e|E|
x1 = x − (4.13)
mω 2
and
e2 |E|2
ε1 = ε + . (4.14)
2mω 2
q
Changing variables to x1 = mω/h̄ξ in Eq. (4.12) we arrive to Eq. (4.75) from the
textbook:
d2 ψ(ξ) 2ε1
 
2
+ − ξ ψ(ξ) = 0 . (4.15)
dξ 2 h̄ω
The eigenfunctions of Eq. (4.15) can be found in the form (4.78) from the textbook
and the eigenvalues are defined by the expression:

1 e2 |E|2
 
εn = h̄ω n + − . (4.16)
2 2mω 2
Thus, the external electric field applied in the direction of oscillations of the charged
particle displaces the position of the point, with respect to which the oscillations take
place, by the magnitude e|E|/(mω 2 ) in the direction of the field (for positively charged
particle), and it also decreases all energy levels of the oscillator by

e2 |E|2
∆ε = . (4.17)
2mω 2
28 CHAPTER 4. ADDITIONAL EXAMPLES OF QUANTIZED MOTION

Problem 4.4. An electron is in its ground state in a cubic potential well with impen-
etrable barriers. Find the length, L, of the edge of the confining cube and the energy
of the electron if the electron’s highest probability density is equal to Pm .

Solution. The normalized electron wavefunction in a cubic potential well with impen-
etrable barriers has the form:
s
8 nx πx ny πy nz πz
     
ψ(x, y, z) = 3
sin sin sin , (4.18)
L L L L
where nα = 1, 2, 3, ... are quantum numbers and α = x, y, and z. For the electron
ground state the quantum numbers are: nx = ny = nz = 1. The probability density of
electron being in this state is given by the expression:

8 πx πy πz
     
2
P (x, y, z) = |ψ(x, y, z)| = 3 sin2 sin2 sin2 . (4.19)
L L L L
This function has maximum in the middle of the well at x = y = z = L/2:

L L L 8
 
Pm = P , , = 3. (4.20)
2 2 2 L
From the last equation we find the width of the well, L:
1/3
8

L= . (4.21)
Pm
The energy of electron in the corresponding quantum state is defined as:

π 2 h̄2
E(nx , ny , nz ) = (n2 + n2y + n2z ) . (4.22)
2me L2 x
Thus, for the ground state (1,1,1) we have:

π 2 h̄2 3π 2 h̄2 2/3

E(1, 1, 1) = (1 + 1 + 1) = P . (4.23)
2me L2 8me m

Problem 4.5. An electron is in its ground state in a cubic potential well with im-
penetrable barriers and with edge length L = 5 nm. Estimate the pressure, which the
electron puts on the walls of the potential well. (Answer: P = 0.35 × 106 Pa.)

Solution. From the solution of the problem for the potential well of the considered
form we find for the energy of electron in the ground state the following expression:

3π 2 h̄2
E(1, 1, 1) = . (4.24)
2me L2
 29

Thus, the electron energy is a function of linear dimension of the well, L, i.e. E(L).
Let us suppose that the well is homogeneously increased. The work that is necessary
to do is defined:
dA = P dV = P 3L2 dL = −dE(L) . (4.25)
Therefore,
1 dE π 2 h̄2
P =− 2 = ≈ 7.7 × 103 Pa . (4.26)
3L dL m e L5
This result also can be obtained in a different way. The corresponding momentum
component is given by the expression:
πh̄
pα = nα , α = x, y, and z , (4.27)
L
where for the ground state we have nα = 1. Let us find the force which electron applies
to the face of the cube that is perpendicular to the x-axis:

∆px 2πh̄ px π 2 h̄2

Fx = = = . (4.28)
∆t L 2Lme m e L3
In the last expression we took into account that
2L 2Lme
∆t = = (4.29)
vx px
and the fact that the change of the electron momentum, ∆px , after reflection from the
barrier is as follows:
2πh̄
∆px = 2px = . (4.30)
L
Therefore, the pressure that electron wave exerts on the face of cube is defined by the
general expression:
Fx π 2 h̄2
P = 2 = . (4.31)
L m e L5

Problem 4.6. At instant t = 0 the electron behavior is described by the following

wavefunction:
2 2
Ψ(r, 0) = Ae−r /α +ikr . (4.32)
Find the normalization constant, A, the most probable value rpr , and the radial part
of the probability current, j.

Solution. The normalization constant, A, can be calculated from the following con-
dition:
Z ∞
|Ψ(r, 0)|2 4πr2 dr = 1 . (4.33)
0
30 CHAPTER 4. ADDITIONAL EXAMPLES OF QUANTIZED MOTION

By substituting the given wavefunction into the integral of Eq. (4.33) we get:
∞ α3 π
Z r
2 /α2
4π|A|2 r2 e−2r dr = 4π|A|2 = 1. (4.34)
0 8 2
From the last equation we find the expression for the normalization constant, A:
3/2
2

|A|2 = . (4.35)
πα2
To find the most probable position of electron, let us equate to zero the following
derivative:
d   d  2 2

4πr2 |Ψ(r, 0)|2 = 4πr2 |A|2 e−2r /α = 0 . (4.36)
dr dr
From the last equation we find:
α
rpr = √ . (4.37)
2
The radial component of the current density vector in the general case is defined as:
ih̄e
jr = (Ψ∇r Ψ∗ − Ψ∗ ∇r Ψ) . (4.38)
2me
The radial component of the vector operator ∇ is a derivative ∇r = ∂/∂r. Therefore:
3/2
eh̄k 2 −2r2 /α2 eh̄k 2

2 /α2
jr = |A| e = e−2r . (4.39)
me me πα2
Equation (4.39) can be written as:

jr = evr n , (4.40)

if we take into account that h̄k/me = vr is the analog of the classical velocity.

Problem 4.7. An electron is in an infinitely deep spherically-symmetric potential

well, i.e. U (r) = 0 at r ≤ r0 and U (r) = ∞ at r > r0 , where r0 is the radius of
the well. Find the form of the radial electron
√ function, the electron energy in s-states
(l = 0), the magnitudes < r > and rsq = < r2 >, the most probable value rpr , and
the probability of the electron being in the region r ≤ rpr .

Solution. The equation for the electron wavefunction X(r) in the region r ≤ r0 for
the given potental well has the form:

d2
" #
2 d l(l + 1)
2
+ + k2 − X(r) = 0 , (4.41)
dr r dr r2
where k 2 = 2me E/h̄2 . By making substitution
 31

R(r)
X(r) = (4.42)
r
Eq. (4.41) is reduced to the folowing equation:

d2
" #
l(l + 1)
2
+ k2 − R(r) = 0 , (4.43)
dr r2
where function R(r) must satisfy the following two conditions: R(0) = 0 and R(r0 ) = 0.
In the s-state the orbital quantum number l = 0 and thus, the equation which defines
function R(r) has the form:

d2
!
+ k 2 Rs (r) = 0 . (4.44)
dr2
The solution of this equation can be written as:

Rs (r) = A sin(kr) . (4.45)

From the condition Rs (r0 ) = 0 we find
πn
k= . (4.46)
r0
Then, the normalized radial wavefunction can be written as:

1 sin(kr)
Xs (r) = √ . (4.47)
2πr0 r
The energy of corresponding states is:

π 2 h̄2 2
Esn = n . (4.48)
2me r0
Mean value of r is given by:
Z r0 1 Z r0 sin2 (kr)
2 2 r0
< r >= r|Xs (r)| 4πr dr = r 2
4πr2 dr = . (4.49)
0 2πr0 0 r 2
Before we find the mean-square distance of electron from the center of the well, rsq , let
us find < r2 >:

r0 2 Z r0 2 2 r02 3
Z  
2 2 2 2
< r >= r |Xs (r)| 4πr dr = r sin (kr)dr = 1− 2 2 . (4.50)
0 r0 0 3 2π n
Thus,
√ r0 3
 1/2
rsq = < r2 >= √ 1− 2 2 . (4.51)
3 2π n
32 CHAPTER 4. ADDITIONAL EXAMPLES OF QUANTIZED MOTION

The most probable distance of electron from the well’s center, rpr , can be determined
from the condition:

4πr2 sin2 (kr)

!
d   d
4πr2 |Xs (r)|2 = = 0. (4.52)
dr dr 2πr0 r2
From the last equation we find rpr :
π
rpr = . (4.53)
2k
Taking into account Eq. (4.46) we find:
r0
rpr = . (4.54)
2n
The probability of electron being in the region r ≤ rpr is:
Z r0 /2 1
W = |Xs (r)|2 4πr2 dr = . (4.55)
0 2

Problem 4.8. For an electron in an infinitely deep spherically-symmetric potential

well, i.e. U (r) = 0 at r ≤ r0 and U (r) = ∞ at r > r0 , where r0 is the radius of the
well, find the dispersion equation for s-states in the energy region E < U0 and interval
of values of U0 r02 when the well contains only one electron energy level.

Solution. Let us write the expression for the radial part of the Schrödinger equation
for the spherically-symmetric potential well:

d2
" #
2 d l(l + 1)
2
+ + k2 − X(r) = 0 , (4.56)
dr r dr r2

where k 2 = 2me E/h̄2 . Using the substitutions,

R(r)
X(r) = , (4.57)
r
dX 1 dR R
= − 2, (4.58)
dr r dr r
d2 X d2 R 2 dR 2R
!
1
2
= − + 2 , (4.59)
dr r dr2 r dr r
the equation for radial function (4.56) of s-states (l = 0) can be transformed to:

d2
!

2
+ k 2 R(r) = 0 . (4.60)
dr
 33

The solution of this equation in the region r ≤ r0 has the form:

√
2me E
R1 = A sin(kr + α) , k = , (4.61)
h̄
and in the region r > r0
q
2me (U0 − E)
R2 = Beκr + Ce−κr , κ = . (4.62)
h̄
From the condition that the wavefunction X(r) must be limited in the entire space,
follows that α = 0 and B = 0. Thus, we can write the solutions as:
sin(kr)
X1 (r) = A , (4.63)
r
e−κr
X2 (r) = C . (4.64)
r
From the condition of continuity of the wavefunction and its derivative we have:

sin(kr0 )
X1 (r0 ) = A (4.65)
r0
!
dX1 (r) kr0 cos(kr0 ) − sin(kr0 )
=A , (4.66)
dr r=r0
r02
e−κr0
X2 (r0 ) = C , (4.67)
r0
e−κr0
!
dX2 (r)
= −C(κr0 + 1) , (4.68)
dr r=r0
r02
at the point r = r0
X1 (r0 ) = X2 (r0 ) (4.69)
! !
dX1 (r) dX2 (r)
= . (4.70)
dr r=r0
dr r=r0

From these equations we find:

k
tan(kr0 ) = − . (4.71)
κ
The last equation can be transformed to:
h̄
sin(kr0 ) = ± q kr0 . (4.72)
2me r02 U0

Equation (4.72) defines the discrete spectrum of the energy eigenvalues. Graphical
analysis of this equation (analogous graphical analysis is carried out in the textbook
34 CHAPTER 4. ADDITIONAL EXAMPLES OF QUANTIZED MOTION

for Eq. (3.75)) shows that the interval of values U0 r02 , for which the potential well
contains only one s-level, is defined as:

π 2 h̄2 9π 2 h̄2
< U0 r02 < . (4.73)
8me 8me

Problem 4.9. The wavefunction of the ground state of a one-dimensional harmonic

oscillator (particle with mass m in a one-dimensional parabolic potential U (x) =
mω 2 x2 /2) has the following form:
2
ψ(x) = Ae−bx . (4.74)

Solve the Schrödinger equation and find the constant b and the energy of the oscillator
in the ground state.

Solution. Let us write one-dimensional Schrödinger equation taking into account the
form of particle’s potential energy U (x):

d2 mω 2 2
" !#
2m
+ 2 E− x ψ(x) = 0 . (4.75)
dx2 h̄ 2
2
Let us substitute in this equation the wavefunction of the ground state, ψ(x) = Ae−bx :

mω 2 2
" !#
2m
2 2 2
−2b + 4b x + 2 E − x Ae−bx = 0 . (4.76)
h̄ 2
It follows from the last equation that

m2 ω 2
!
2mE
 
2
− 2b + 4b − x2 = 0 . (4.77)
h̄2 h̄2
Since this equation must be satisfied for any value of x, the following condition must
be satisfied:

m2 ω 2
4b2 − = 0, (4.78)
h̄2
2mE
− 2b = 0 . (4.79)
h̄2
Thus, we have:
mω
b= , (4.80)
2h̄

bh̄2 h̄ω
E= = . (4.81)
m 2
 35

Problem 4.10. The wavefunction of the one-dimensional harmonic oscillator in a

particular state has the following form:
2
ψ(x) = Axe−bx . (4.82)
Find the constant b and the energy of the oscillator in this state.

Solution. By substituting the given wavefunction into the Schrödinger equation for
the one-dimensional harmonic oscillator we find:

mω 2 2
" !#
2 2m 2
2b(2bx − 3) + 2 E − x Axe−bx = 0 . (4.83)
h̄ 2
From the last equation it follows that
m2 ω 2
!
2mE
 
2
− 6b + 4b − x2 = 0 . (4.84)
h̄2 h̄2
Since this equation must satisfy for any value of x, the following conditions must be
satisfied:

m2 ω 2
4b2 − =0 (4.85)
h̄2
and
2mE
− 6b = 0 . (4.86)
h̄2
From the last two expressions we obtain:
mω
b= (4.87)
2h̄
and
3bh̄2 3
E= = h̄ω . (4.88)
m 2

Problem 4.11. A particle with mass m moves in a three-dimensional parabolic po-

tential U (r) = mω 2 r2 /2. Find the dependence of the oscillator’s energy on the orbital
quantum number.

Solution. Because of the spherical symmetry of the potential U (r) the states with
the specific orbital, l, and magnetic, m, quantum numbers can be found if we consider
particle’s motion in spherical coordinates. The equation for the radial part of the
wavefunction X(r) has the form:
d2 X 2 dX 2mE m2 ω 2 r2 l(l + 1)
" #
+ + − − X = 0. (4.89)
dr2 r dr h̄2 h̄2 r2
36 CHAPTER 4. ADDITIONAL EXAMPLES OF QUANTIZED MOTION

By substituting X = R/r into this equation, it can be transformed to:

d2 R
" #
2 2 l(l + 1)
+ k − br − R = 0, (4.90)
dr2 r2
where we introduced √
2mE
k= (4.91)
h̄
and
mω
.
b= (4.92)
h̄
By taking into account that the wavefunction, R, tends to zero at r → 0 and r → ∞,
we can write the solution of Eq. (4.90) in the form:
2 /2
R(r) = rl+1 e−br w(r) . (4.93)

By substituting this solution into the equation for the function R(r) we obtain the
equation for the function w(r):

d2 w
!
l+1 dw 3
  
2
+2 − br + k 2 − 2b l + w = 0. (4.94)
dr r dr 2

Let us introduce a new variable ρ = br2 . As a result the last equation transforms to
the following form:
d2 w k2
!
3 dw 3 w
 
ρ 2 + l+ −ρ − −l− = 0. (4.95)
dρ 2 dρ 2b 2 2
The solution of this equation, which increases as ρ → ∞ not faster than the limited
order of magnitude of ρ, exists if
k2
!
1 3
−l− = nr , nr = 0, 1, 2, ... (4.96)
2 2b 2
The last equation defines the energy spectrum of the three-dimensional oscillator:
3
 
Enr ,l = h̄ω l + 2nr + . (4.97)
2
Comparing this expression with the expression
3
 
EN = h̄ω N + , N = nx + ny + nz , (4.98)
2
which was derived for a three-dimensional oscillator in Cartesian coordinates, we see
that N = l + 2nr . Thus, the energy level with given quantum number N can have
different values of orbital quantum number, l, i.e. the degeneracy over the angular
momentum may exist.
Chapter 5

Approximate methods of finding

quantum states

(0) (0)
Problem 5.1. The unperturbed system has two close energy levels, E1 and E2 ,
which are separated by an energy comparable to the energy of perturbation. Find the
first-order corrections to the energy of these states.

Solution. Similar to the considered in the texbook method of determination of the

perturbed degenerate quantum states let us choose the wavefunction, ψ, in the form
(0) (0)
of a superposition of the unperturbed wavefunctions, ψ1 and ψ2 :
(0) (0)
ψ = C1 ψ1 + C2 ψ2 . (5.1)
This wavefunction describes two new states to which the initial two states evolve un-
der the influence of the perturbation. Let us substitute the wavefunction defined by
Eq. (5.1) into the Schrödinger equation
(Ĥ0 + V̂ )ψ = Eψ . (5.2)
The result of substitution is the following:
(0) (0) (0) (0)
(Ĥ0 + V̂ )(C1 ψ1 + C2 ψ2 ) = E(C1 ψ1 + C2 ψ2 ) . (5.3)
(0)∗ (0)∗
Let us multiply Eq. (5.3) by ψ1 . Then, let us multiply Eq. (5.3) by ψ2 . After this
let us integrate both of the obtained equations over the entire volume of the quantum
system. Let us take into account that the unperturbed wavefunctions are orthogonal
and normalized and that they satisfy the following equations:
(0) (0) (0)
Ĥ0 ψi = Ei ψi . (5.4)
Taking into account Eq. (5.4), we obtain a system of two linear homogeneous algebraic
equations with unknown coefficients C1 and C2 :
(0)
(E1 + V11 − E)C1 + V12 C2 = 0 ,
(5.5)
(0)
V21 C1 + (E2 + V22 − E)C2 = 0 .

37
38CHAPTER 5. APPROXIMATE METHODS OF FINDING QUANTUM STATES

By equating the determinant of this system to zero we obtain a quadratic equation:

(0) (0) (0) (0)
E 2 − (E1 + E2 + V11 + V22 )E + (E1 + V11 )(E2 + V22 ) − |V12 |2 = 0 . (5.6)

The two roots of this equation correspond to the two energy levels:
v
(0) (0) u (E (0) + V − E (0) − V )2
u
E1 + E2 + V11 + V22 t 1 11 2 22
E (±) = ± + |V12 |2 . (5.7)
2 4

If the matrix element V12 becomes equal to zero, then we come to the conventional
corrections to the energy in the first approximation: the Vnn is added to the unper-
turbed energy levels En(0) . According to the formulation of the problem the matrix
element V12 has the same order of smallness as the energy distance between the energy
(0) (0)
levels E2 − E1 . Thus, under perturbation the energy levels, E (±) , are substantially
reconfigured with respect to the unperturbed energy states and the corresponding
wavefunctions (5.1) are determined by coefficients:
  1/2
(0) (0)
V12  E1 + V11 − E2 − V22
C1± =  1± q  , (5.8)
2|V12 | (0) (0) 2
(E1 + V11 − E2 − V22 ) + 4|V12 | 2

  1/2
(0) (0)
V12  E1 + V11 − E2 − V22
C2∓ = ± 1∓ q  . (5.9)
2|V12 | (0) (0) 2
(E1 + V11 − E2 − V22 ) + 4|V12 | 2

An electron in a spherically-symmetric potential can be considered as an example of

a degenerate quantum system. As it was shown in Section 4.2 of the textbook the
quantum states with the orbital angular momentum, L 6= 0, are always degenerate
with respect to the orbital magnetic number, m, with the order of degeneracy equal to
2l + 1. Here l is the orbital quantum number. If the perturbation breaks the spherical
symmetry, then the different projections of the orbital angular momentum, L, on the
chosen axis must correspond to different electron energies. Thus, in the magnetic field,
B, the term:
∆E = µB Bm , (5.10)
is added to the unperturbed electron energy. Here m = 0, ±1, ±2, ..., ±l and µB is the
Bohr’s magneton. As a result, the unperturbed electron energy level in the potential
well under the influence of the magnetic field splits (for the electron in an atom such
a splitting is called the Zeeman effect).

Problem 5.2. A harmonic oscillator with charge e and mass m is placed in a homo-
geneous electric field E directed along the axis of oscillations. Considering the electric
field as a perturbation, find the shift in the energy levels of the oscillator.
 39

Solution. The perturbation Hamiltonian in this case is given by the following operator:

V̂ (x) = −e|E|x . (5.11)

The normalized eigenfunctions for the harmonic oscillator with unperturbed Hamilto-
nian (Eq. (4.74) from the textbook) have the following form:
 √ −1/2
−x2 /(2b2 ) x
 
ψn(0) (x) = 2 n
πbn! e Hn , (5.12)
b
q
where we have introduced the parameter b = h̄/mω. For further calculations it is
important that Hermite polynomials, Hn , satisfy the following recursive relation:

Hn+1 (x) + 2nHn−1 (x) = 2xHn (x) . (5.13)

According to Eq. (5.11) the matrix elements of the perturbation operator V̂ are ex-
pressed through the coordinate matrix elements, xnl , as:

Vnl = −e|E|xnl . (5.14)

The coordinate matrix elements taking into account the orthogonality of the unper-
turbed wavefunctions (5.12) and Eq. (5.13) are defined as:
( q
(n + 1)/2, l = n + 1,
Z
(0)
xnl = ψn(0)∗ xψl dx = b (5.15)
0, l 6= n + 1 .

According to Eq. (5.12) from the textbook, we obtain for the n-th energy level for the
second order of perturbation the following expression:
|Vln |2 1 e2 b2 |E|2
 
En = En(0) + ǫEn(1) + ǫ2 En(2) = En(0) + Vnn +
X
(0) (0)
= h̄ω n + − .
l6=n En − El 2 2h̄ω
(5.16)
We see that the first order correction is equal to ǫEn(1) = 0 and the second order
correction of perturbation theory is equal to:
e2 b2 |E|2 e2 |E|2
ǫ2 En(2) = − =− , (5.17)
2h̄ω 2mω 2
which coincides with the exact solution of this problem. Therefore, the corrections of
the third and higher order must be equal to zero.

Problem 5.3. An electron in a one-dimensional rectangular potential well with the

barriers of infinite height,
(
0, 0 ≤ x ≤ L,
U (x) =
∞, x < 0, x > L,
40CHAPTER 5. APPROXIMATE METHODS OF FINDING QUANTUM STATES

Figure 5.1: One-dimensional rectangular potential well with perturbation W (x).

is subjected to a perturbation of the following form:

(
W0 , b ≤ x ≤ L − b,
W (x) =
0, x < b, x > L − b.

Find the energy of the electron in stationary states, taking into account corrections
arising from the first-order perturbation theory.

Solution. The unperturbed wavefunctions and electron energy in the given potential
well are defined by the expressions:
s
2 πnx
ψ0n (x) = sin (5.18)
L L
π 2 h̄2 2
E0n = n , (5.19)
2me L
where n = 1, 2, 3, .... The correction of first order to the energy values, E0n , is defined
by the diagonal matrix element of the perturbation operator, Ŵ :
Z
∗
E1n = Wnn = ψ0n (x)Ŵ (x)ψ0n (x)dx , (5.20)

where ψ0n (x) are electron wavefunctions of unperturbed electron states in the potential
well. For the considered case E1n is defined as:

2W0 Z L−b 2 πnx 2W0 Z L−b 1 − cos(2πnx/L)

E1n = sin dx = dx . (5.21)
L b L L b 2
By carrying out the integration in the last equation we have:
" !#
W0 L 2πn(L − b) 2πnb
E1n = L − 2b − sin − sin =
L 2πn L L
 41
!
2b 1 4πnb
= W0 1− − sin . (5.22)
L πn L
Thus, first order perturbation theory gives the following expression for the energy of
electron stationary states:

π 2 h̄2 2
!
2b 1 4πnb
En = E0n + E1n = n + W0 1 − − sin . (5.23)
2me L L πn L

If b = 0 then the corrections to all energy levels are constant and equal to W0 , which
means just a shift of the origin of coordinate system.
Since function sin(4πnb/L) is oscillatory, the dependence on the n-th perturbation
energy is decreasing with the increase of n and if πn is substantially larger than (1 −
2b/L) the perturbation leads to the homogeneous shift by W0 (1 − 2b/L).
From Eq. (5.23) we find that these corrections are valid for the states with the following
quantum numbers:
√
2me LW0
n≫ . (5.24)
πh̄

Problem 5.4. Find the energy of the ground state of a one-dimensional harmonic
oscillator which is subjected to an anharmonic potential perturbation W (x) = γx3 .

Solution. The solution of the unperturbed harmonic oscillator problem (example of

a simple harmonic oscillator is a particle with mass m which oscillates with frequency
ω) gives us the following expressions for the normalized wavefunctions and energy:
2 /(2b2 )
ψ0n (x) = An Hn (x)e−x , (5.25)

1
 
E0n = n+ h̄ω , (5.26)
2
q
where b = h̄/(mω), Hn (x) is Hermite polynomial, and n = 0, 1, 2, .... The correction
of the first order is defined by the mean value of perturbation which is defined by the
unperturbed states. In this problem the correction is equal to zero since it is defined
by the integral with symmetric limits of the antisymmetric function:

Z ∞ Z ∞
∗ ∗ 3
E1n = Wnn =< W (x) >= ψ0n Ŵ ψ0n dx =γ ψ0n x ψ0n dx = 0 . (5.27)
−∞ −∞

The correction of the second order of perturbation theory is given by the expression:
X |Wnl |2
E2n = . (5.28)
l6=n E0n − E0l
42CHAPTER 5. APPROXIMATE METHODS OF FINDING QUANTUM STATES

Non-diagonal matrix elements of the perturbation Hamiltonian can be expressed in

terms of the matrix elements of coordinates:

Z ∞ ′
∗ 3
X
Wnn′ = γ ψ0n x ψ0n′ dx = γ < n|x|j >< j|x|l >< l|x|n > . (5.29)
−∞ j,l,n

Here, the short notation of matrix elements, used in quantum mechanics, is introduced:
Z ∞
∗
xαβ =< α|x|β >= ψ0α xψ0β dx . (5.30)
−∞

Using the unperturbed wavefunctions of the harmonic oscillator we can show that only
the following coordinate matrix elements do not vanish:

s
nh̄ n
r
xn,n−1 = xn−1,n =< n|x|n − 1 >=< n − 1|x|n >= =b . (5.31)
2mω 2
Only matrix elements, Wnn′ , of the following type are non-zero:
!3/2
nh̄
Wn,n−1 = Wn−1,n = 3γ , (5.32)
2mω
!3/2 q
h̄
Wn,n−3 = Wn−3,n =γ n(n − 1)(n − 2) . (5.33)
2mω
Using these expressions for the second order of perturbation theory we get:
X |Wnl |2 X < n|W (x)|l >< l|W (x)|n >
E2n = = =
l6=n E0n − E0l l6=n h̄ω(n − l)
!3 
15γ 2 h̄ 11

=− n2 + n + . (5.34)
4h̄ω mω 30
By taking into account the corrections of the second order, we find for the ground state
energy of harmonic oscillator the following expression:

11γ 2 h̄
!
h̄ω
E0 = 1− . (5.35)
2 4m3 ω 5

Problem 5.5. A charged one-dimensional oscillator is placed in a perturbation ho-

mogeneous electric field directed along the axis of oscillations. Find the energy of the
oscillator, taking into account the first two orders of corrections of perturbation theory.
The charge of the particle is positive and equal to e and its mass is equal to m.
 43

Solution. We start with the normalized eigenfunctions of the unperturbed problem

for the one-dimensional harmonic oscillator:
2 2
 √ −1/2
ψ0n (x) = An Hn (x)e−x /(2b ) , An = 2n n! πb , (5.36)
q
where linear dimension b = h̄/(mω). For the nonzero matrix elements of coordinate
we can write the following expressions:
s
nh̄ n
r
xn,n−1 = xn−1,n =< n|x|n − 1 >=< n − 1|x|n >= =b . (5.37)
2mω 2
A constant electric force Fe = eE acts on the charged particle in the homogeneous
electric field. The potential energy, which in this problem is the perturbation, is equal
to
W = −e|E|x . (5.38)
Therefore, all nonzero perturbation matrix elements are related with matrix elements
of coordinate as:
Wn,n−1 = −e|E|xn,n−1 . (5.39)
The correction of the first order is defined by the average value of perturbation operator
over the unperturbed states. In our case it is equal to zero:
Z ∞
∗
E1n = Wnn =< W (x) >= −e|E| ψ0n xψ0n dx = 0 . (5.40)
−∞

The second order correction is given by the expression:

X |Wnl |2 X < n|x|l >< l|x|n > e2 b2 |E|2 e2 |E|2
E2n = = =− =− . (5.41)
l6=n E0n − E0l l6=n h̄ω(n − l) 2h̄ω 2mω 2

Thus, the energy of the corresponding level of a charged harmonic oscillator in electric
field, taking into account second order corrections, is given by the expression:
1 e2 |E|2
 
En = h̄ω n + − . (5.42)
2 2mω 2

Problem 5.6. The electron is in a spherically-symmetric potential well:

(
0, r ≤ a,
U (r) =
∞, r > a .
The unperturbed magnitudes of the electron energy in the well are Enl . Find, in the
first order of perturbation theory, the electron energy and wavefunction in the magnetic
field directed along the z-axis. The perturbation operator V̂ is defined as
ih̄eB ∂
V̂ = ,
2me ∂ϕ
44CHAPTER 5. APPROXIMATE METHODS OF FINDING QUANTUM STATES

where ϕ is the azimuthal angle in plane perpendicular to z-axis. Hint: use the spherical
coordinate system.

Solution. In the absence of a magnetic field the unperturbed Schrödinger equation for
the spherically-symmetric potential well has the solution described by the wavefunction:
(0)
ψnlm = X(r)Y (θ, ϕ) = Xnl (r)Plm (cos θ)eimϕ , (5.43)

and by the energy of the unperturbed quantum states (see Eq. (4.54) from the text-
book):
(0) h̄2 gnl
2
Enl = . (5.44)
2me a2
(0)
In this case the correction to the energy Enl in the first order of approximation is
defined by Eq. (5.15) from the textbook:
Z
(1) (0)∗ (0)
Enl = ψnlm V̂ ψnlm dV . (5.45)

(0)
Taking into account the dependence of the wavefunction, ψnlm , on ϕ defined by the
equation
(0)
∂ψnlm (0)
= imψnlm , (5.46)
∂ϕ
the first order correction to the energy is equal to:

(1) eB Z
(0) eB
Enl = − h̄m |ψnlm |2 dV = − h̄m = −mµB B . (5.47)
2me 2me

Here m is the orbital magnetic quantum number and µB is the Bohr’s magneton.
In order to find the eigenfunction of this problem with perturbation and the second
order correction to the energy let us write the nondiagonal matrix elements as:

′ ih̄eB Z ∗ ∂ψnlm′ h̄eB ′ Z ∗

< nlm|V̂ |nlm >= ψnlm dV = − m ψnlm ψnlm′ dV , (5.48)
2me ∂ϕ 2me
′
which are equal to zero at m 6= m . Thus, the magnetic field does not change the
wavefunctions of the unperturbed problem. The second order correction to the energy
is also equal to zero and the unperturbed energy levels split into number of sublevels
equal to the orbital magnetic quantum number m, i.e.

(0) eB
Enlm = Enl − h̄m . (5.49)
2me
 45

Problem 5.7. An electron, which is in a one-dimensional potential U (x) = βx2 /2 (a

linear harmonic oscillator), is placed in a homogeneous electric field. The electric field
is directed along the axis of oscillations and changes with time as:
2 /τ 2
E(t) = E0 e−t .

Before the electric field is turned on (t → −∞) the electron was in one of its unper-
turbed states. Find, in the first order of perturbation theory, the probability of electron
excitation to the higher states at t → ∞.

Solution. The perturbation Hamiltonian has the form

W = −e|E(t)|x (5.50)

and its matrix elements can be written through matrix elements of harmonic oscillator
coordinate:
 √
n + 1, m = n + 1,
e|E(t)|b  √

Wmn (t) = −e|E(t)|xmn = − √ n, m = n − 1, (5.51)
2  
0, other m and n ,
q
where b = h̄/(me ω). Nonzero transition probabilities of oscillator in the first order
of corrections are equal to:
!2 (
1 Z∞ eb|I| n + 1, m = n + 1,
Pmn (t) = 2 Wmn (t)e±iωt dt = √ (5.52)
h̄ −∞ 2h̄ n, m = n − 1,

where we introduced
Z ∞ √ 2 τ 2 /4
I= |E(t)|e±iωt dt = πτ |E0 |e−ω . (5.53)
−∞

Here, the sign ′ +′ in the exponent’s ± corresponds to m = n+1 and the sign ′ −′ to m =
n − 1, but the magnitude |I|2 does not depend on this sign. Thus, nonzero probabilities
have transitions from n-state only to the nearest energy states with quantum numbers
n + 1 and n − 1. The condition, when this approximation is valid, is:
√ h̄ω
n+1≪ . (5.54)
eb|E|

Problem 5.8. The motion of a plane harmonic oscillator takes place in the xy-plane
in the following potential:
β(x2 + y 2 )
U (x, y) = .
2
46CHAPTER 5. APPROXIMATE METHODS OF FINDING QUANTUM STATES

Find, in the first order of perturbation theory, the splitting of the first excited energy
level of the oscillator under the perturbation potential W = γxy.

Solution. Let us first consider the problem of the unperturbed plane harmonic oscil-
lator. Since its operators

h̄2 ∂ 2 βx2 h̄2 ∂ 2 βy 2

Ĥx = − + and Ĥ y = − + (5.55)
2m ∂x2 2 2m ∂y 2 2
commutate with each other and with the total Hamiltonian, Ĥ = Ĥx + Ĥy , then the
eigenfunctions of the total Hamiltonian, Ĥ, can be built from the eigenfunctions of the
Hamiltonians Ĥx and Ĥy , i.e. from the wavefunctions of linear oscillators:

ψnx ny (x, y) = ψnx (x)ψny (y) . (5.56)

The eigenenergies of the plane oscillator are defined as:
(0)
EN = h̄ω(N + 1) , N = nx + ny = 0, 1, 2, ... , (5.57)
q
where ω = β/m, nx = 0, 1, 2, ... and ny = 0, 1, 2, .... Energy state EN with a given
quantum number N corresponds to linearly-independent eigenfunctions ψnx ny with the
quantum numbers nx = 0, 1, 2, ..., N and ny = N, N − 1, ..., 0. This means that the
given energy level of the plane harmonic oscillator is (N + 1) times degenerate. Thus,
(0)
the unperturbed level E1 is doubly degenerate with the set of quantum numbers
nx = 0, ny = 1 and nx = 1, ny = 0. The corresponding wavefunctions are:
(0)
ψ1 = ψ10 (x, y) = ψ1 (x)ψ0 (y) , (5.58)

(0)
ψ2 = ψ01 (x, y) = ψ0 (x)ψ1 (y) . (5.59)
Therefore, the wavefunction of the perturbed state can be written as a linear combi-
nation of the unperturbed wavefunctions:
(0) (0)
ψ(x, y) = C1 ψ1 + C2 ψ2 . (5.60)
Substituting this wavefunction into the Schrödinger equation, multiplying it by complex-
conjugated wavefunction ψ ∗ (x, y), and integrating the obtained equation we get the
system of two linear equations with unknowns C1 and C2 :
(0)
(E1 + W11 − E1 )C1 + W12 C2 = 0 ,
(0) (5.61)
W21 C1 + (E1 + W22 − E1 )C2 = 0 ,
where the matrix elements of perturbation operator are:

W11 = W22 = 0 , (5.62)

 47

b2
W12 = W21 = γ . (5.63)
2
q
Here b = h̄/(me ω). Equating the determinant of the system of equations (5.61) to
zero we obtain the following equation:
(0)
(E1 − E1 )2 − W12 W21 = 0 . (5.64)
The solution of this equation gives us for the energy of the first excited state the
following expression:

(0)γb2
E1 = E1 ± . (5.65)
2
It follows from the last equation that the degeneracy is lifted for the energy level N = 1.

Problem 5.9. Find, in the quasiclassical approximation, the probability of electron

transmission, P , through the potential barrier
(
0, x < 0,
U (x) =
U0 e−x/b , x ≥ 0 ,

for an electron whose energy is E < U0 .

Solution. The standard expression for the transmission coefficient in the quasiclassical
approximation has the form:
R x2 R x2 √
− h̄2 |p|dx − h̄2 2me (U (x)−E)dx
D≈e x1
=e 0 , (5.66)

where the upper limit of integration is equal to x2 = b ln(U0 /E). As a result of

integration we obtain the following result:
 s s 
4b q U0 U0
D ≈ exp − 2me E  − 1 − arctan − 1 . (5.67)
h̄ E E
Chapter 6

Quantum states in atoms and

molecules

Problem 6.1. Using Bohr’s model of the hydrogen atom and his postulates, find the
radii of electron stationary orbits, rn , electron velocities, vn , orbital periods, Tn , the
electron energy on the corresponding orbit, and the circular frequency of a photon,
emitted during electron transition from the nth to the mth -orbit. Find the radius and
velocity, r1 and v1 , for the first electron orbit. Find the angular frequency, ω21 , of
the electron transition from the second to the first orbit. (Answer: r1 ≈ 0.53 Å,
v1 ≈ 2.19 × 106 m s−1 , and ω21 ≈ 1.53 × 1016 s−1 .)

Solution. Let us write two equations that follow from Bohr’s model: 1) Newton’s
second law for an electron on circular stationary orbit:
me vn2 ke e2
= 2 . (6.1)
rn rn
2) The angular momentum quantization, which Bohr formulated for stationary orbits,
has the form:
Ln = me vn rn = nh̄ , (6.2)
where n = 1, 2, 3, ... is the only quantum number in Bohr’s theory of atoms, which is
called the principal quantum number. From Eqs. (6.1) and (6.2) we find:
h̄n
rn = , (6.3)
m e vn
ke e2 me vn
me vn2 = , (6.4)
h̄n
ke e2 e2
vn = = , (6.5)
h̄n 4πǫ0 h̄n
h̄n 4πǫ0 h̄2 n2
rn = = . (6.6)
m e vn me e 2

49
50 CHAPTER 6. QUANTUM STATES IN ATOMS AND MOLECULES

Thus, for the n-th orbit we have the following expressions:

rn = r1 n 2 , (6.7)
v1
.
vn = (6.8)
n
The period of electron rotation on the n-th orbit is:
2πrn
Tn = = T 1 n3 , (6.9)
vn
where
2πh̄3 4πǫ0 2
 
T1 = 2
= 1.5 × 10−16 s . (6.10)
me e
Let us find now the radius and the velocity for the first Bohr’s orbit:

4πε0 h̄2
r1 = = 5.26 × 10−11 m , (6.11)
me e 2

e2
v1 = = 2.19 × 106 m/s . (6.12)
4πε0 h̄
To find the electron total energy on the corresponding orbit let us find electron’s kinetic
and potential energy: !2
me e2
Kn = , (6.13)
2 4πǫ0 h̄n
!2
e2 e2
Un = − = −me . (6.14)
4πǫ0 rn 4πǫ0 h̄n
Equations (6.13) and (6.14) show that on the stationary orbit we have:

2Kn = |Un | . (6.15)

Thus, the total energy, En , is:

!2
E1 me e2 1
En = Kn + Un = 2 = − (6.16)
n 2 4πǫ0 h̄ n2

with !2
me e2
E1 = − = −13.6 eV . (6.17)
2 4πǫ0 h̄
When an electron makes a transition from a higher orbit to a lower orbit, it emits a
photon whose frequency is:
Em − En 1 1
 
ωmn = = 2πcR∞ − , m = n + 1, n + 2, ... (6.18)
h̄ n2 m 2
 51

Here,
me e 4
R∞ = 3 2 3
= 1.097 × 107 m−1 (6.19)
64π ǫ0 h̄ c
is the Rydberg’s constant. Thus, the frequency of photon emitted as a result of tran-
sition from the second orbit to the first is equal to:
6πcR∞
ω21 = ≈ 1.53 × 1016 s−1 . (6.20)
4

Problem 6.2. Find the minimum, λmin , and maximum, λmax , wavelength of the hy-
drogen spectral lines in the visible range of the spectrum. What is the minimum speed,
vmin , of electrons incident on a hydrogen atom to excite these spectral lines? (Answer:
λmin ≈ 365 nm, λmax ≈ 656 nm, and vmin ≈ 8.2 × 105 m s−1 .)

Solution. The wavelength of the corresponding hydrogen spectral line is the following:
1 ωmn 1 1
 
= = R∞ 2
− 2 , (6.21)
λmn 2πc n m
where
me e 4
R∞ = 3 2 3
= 1.097 × 107 m−1 , (6.22)
64π ǫ0 h̄ c
is the Rydberg’s constant. Balmer series for which n = 2 and m = 3, 4, 5, ... lies in the
visible part of electromagnetic spectrum. The minimum wavelength of spectral line in
this series corresponds to m → ∞ and is equal to
4
λmin = = 365 nm . (6.23)
R∞
The maximum wavelength of the spectral line in this series corresponds to m = 3:
36
λmax = = 656 nm . (6.24)
5R∞
To find the minimum velocity of the incident electron on hydrogen atom, let us assume
that all of its kinetic energy is transmitted to hydrogen atom. As a result of such a
collision the electron in atom must make a transition from level n = 2 to level m = 3.
Therefore,
2
me vmin 1 1 5
 
= E3 − E2 = h̄2πcR∞ 2 − 2 = h̄2πcR∞ . (6.25)
2 2 3 36
From the last expression we find vmin :
s
5 h̄2πcR∞
vmin = ≈ 8.2 × 105 m/s . (6.26)
18 me
52 CHAPTER 6. QUANTUM STATES IN ATOMS AND MOLECULES

Problem 6.3. The electron in the hydrogen atom is in the ground state described by
the following wavefunction:

ψ1s (r) = (πr13 )−1/2 e−r/r1 . (6.27)

Find the average electrostatic potential at the distance r from the nucleus.

Solution. The potential, ϕ, in an atom is the sum of potentials of the nucleus, ϕnuc ,
and electron, ϕe . The nucleus potential at the distance r is equal to:
e
ϕnuc (r) = , e > 0. (6.28)
4πǫ0 r
To find the potential created by the electron let us write the Poisson equation:
ρ
∇2 ϕe = − . (6.29)
ǫ0

Because of the spherical symmetry of the ground state wavefunction the operator ∇2
contains only the radial part. Therefore, we can write Eq. (6.29) in spherical coordi-
nates in the following form:

1 d2 e
2
(rϕ e ) = |ψ1s (r)|2 , (6.30)
r dr ǫ0
where
e|ψ1s (r)|2 = −ρ . (6.31)
Let us substitute into the last equation the wavefunction ψ1s and let us integrate
Eq. (6.30) twice. First integration gives:

d e Z −2r/r1 e r1 −2r/r1
 
−2r/r1
(rϕe ) = re dr = − re + e + A . (6.32)
dr πǫ0 r13 2πǫ0 r12 2

Second integration gives:

e 1 1 B
  
′
ϕe = + e−2r/r1 − A − , (6.33)
4πǫ0 r r1 r
where
′ 2A
A = . (6.34)
r12
′
To find the integration constants, A and B, we will use the boundary conditions. At
′
r → ∞ the potential ϕe → 0, therefore A = 0. At r → 0 the potential ϕe must stay
limited and thus B = 1.
 53

Let us write now the total potential as the sum of electron and nucleus potentials:
e 1 1 1 e e r
    
ϕ = ϕe + ϕnuc = + e−2r/r1 − + = 1+ e−2r/r1 .
4πǫ0 r r1 r 4πǫ0 r 4πǫ0 r r1
(6.35)
Near the nucleus where r ≪ r1 the atom’s potential is defined primarily by the positive
charge of nucleus. At a significant distance from the nucleus where r ≫ r1 the nucleus
charge is screened by the electron cloud. As a result the potential, ϕ, exponentially
decreases.

Problem 6.4. Find the spectrum of the vibrational levels of a diatomic molecule.
For the diatomic CO molecule estimate the distance between vibrational energy levels,
∆En+1,n , and the emission wavelength, λ, during the corresponding transitions. The
carbon atom mass m1 = 1.99 × 10−26 kg, the oxygen atom mass m2 = 2.66 × 10−26 kg,
and the force constant β = 190 kg s−2 . (Answer: ∆En+1,n ≈ 8.44 × 10−2 eV and λ ≈
24 µm. Hint: use Eq. (6.122) from the textbook.)

Solution. We place the origin of coordinates at the center of one of the atoms. Let
us introduce the magnitude x = r − r0 , where r0 is the distance between atoms in
the equilibrium state, and r is the distance between oscillating atoms. Let us assume
that the displacement of atoms from equilibrium is small. Then, we can expand the
potential energy in series:
1 ′′
U (x) = U (0) + U (0)x2 + ... . (6.36)
2
The Schrödinger equation which describes one-dimensional oscillations of atoms can
be written as:
h̄2 d2 βx2
" #
− + U (0) + ψ = Eψ , (6.37)
2µ dx2 2
where
m1 m2
µ= (6.38)
m1 + m2
is the reduced mass of the molecule, U (0) is the potential energy of the molecule in
′′
the equilibrium position, and β = U (0) is the force constant which defines chemical
q
bonding in the molecule and its fundamental frequency of oscillations ω0 = β/µ.
Equation (6.37) is the equation of a harmonic oscillator. Its energy spectrum is defined
by the well-known formula:
1
 
En = U (0) + h̄ω0 n+ , n = 0, 1, 2, ... (6.39)
2
The distance between oscillating states of the molecule is equal to:

∆En+1,n = h̄ω0 . (6.40)

54 CHAPTER 6. QUANTUM STATES IN ATOMS AND MOLECULES

This expression is approximate and is valid only for low-lying states. For the given
parameters of CO molecule we find the frequency and the distance between neighboring
states: s
β
ω0 = = 1.28 × 1014 s−1 (6.41)
µ
and
∆En+1,n = h̄ω0 = 8.44 × 10−2 eV . (6.42)

The emission wavelength of the molecule transitions between neighboring states is equal
to:
λ ≈ 24 µm . (6.43)

Problem 6.5. Find for the first-order perturbation theory the energy, E1 , of the
ground state of the helium, He, atom. Use as the perturbation the energy of interac-
tion between electrons. (Answer: E1 ≈ -18.7 eV.)

Solution. In the zeroth approximation helium atom can be considered as a system of

two non-interacting electrons in the field created by the positive charge of nucleus Ze,
where Z = 2 and e > 0. In this case the Hamiltonian of the system is:

h̄2 2 ke Ze2 h̄2 2 ke Ze2

Ĥ0 = − ∇ − − ∇ − , (6.44)
2m 1 r1 2m 2 r2

where r1 and r2 are the distances of electrons from the nucleus. The wavefunction
of the ground 1s-state of the system of electrons in the zeroth approximation can be
written as a product of the single-electron radial functions of two electrons found from
the solution of the Schrödinger equation for hydrogen atom:
3
1 Z

Ψ01 = ψ1 (r1 )ψ1 (r2 ) = e−Z(r1 +r2 )/a0 , (6.45)
π a0

where a0 is radius of the first Bohr’s orbit. The energy of the helium atom in zeroth
approximation is the doubled energy of the ground state of electron in a hydrogen-like
atom: !2
(0) me Ze2 ke Z 2 e2
E1 = 2E1 = −2 =− = −27.2 eV . (6.46)
2 4πǫ0 h̄ a0
The Coulomb interaction of the electrons we can consider as a perturbation. Therefore,
the perturbation operator can be written as:

ke e2
Ŵ (r1 , r2 ) = , r12 = kr1 − r2 k . (6.47)
r12
 55

Using first order of perturbation theory the correction to the energy is defined by the
expression:
6 Z
ke e2 ke e2 Z 1 −2Z(r1 +r2 )/a0
Z 
(1)
E1 = W11 = Ψ∗01 Ψ01 dV1 dV2 = 2 e dV1 dV2 , (6.48)
r12 π a0 r12

where dV1 = 4πr12 dr1 . Taking into account that

Z
1 −α(r1 +r2 ) 20π 2
e dV1 dV2 = 5 , (6.49)
r12 α
we arrive to the following expression:
6
ke e2 5 ke Ze2
5
Z a0
 
(1) 2
E1 = 2 20π = . (6.50)
π a0 2Z 8 a0
Thus, the energy of helium atom’s ground state, using first order of perturbation theory,
is:
ke Z 2 e2 5 ke Ze2 ke Z 2 e2 5
 
(0) (1)
E1 ≈ E1 + E1 =− + =− 1− = −18.7 eV . (6.51)
a0 8 a0 a0 8Z

Problem 6.6. In the alkaline atoms the nucleus with charge eZ and the first Z − 1
electrons form a core with positive charge equal to the elementary charge, e. A valence
electron rotates in the field of this core. Find the energy spectrum of the valence
electron if its potential energy is defined by the following expression:

ke e2 C
 
U (r) = − 1+ , (6.52)
r r
where we can consider C/r1 << 1 (r1 is the Bohr’s radius).

Solution. The radial equation for alkaline atoms is analogous to the radial equation
for hydrogen atom:

h̄2 1 d ke e2 ke e2 h̄2 l(l + 1)

! " #
2 dX
r + E+ +C 2 − X = 0. (6.53)
2me r2 dr dr r r 2me r2

Let us group the last two terms:

′ ′ me e 2
l (l + 1) = l(l + 1) − C
. (6.54)
2πǫ0 h̄2
As a result of this regrouping the equation for radial part of the wavefunction will have
the same form as for the hydrogen atom:
56 CHAPTER 6. QUANTUM STATES IN ATOMS AND MOLECULES

h̄2 1 d h̄2 l (l + 1)
′ ′
ke e2
! " #
dX
2
r2 + E+ − X = 0. (6.55)
2me r dr dr r 2me r2
By solving the last equation we find the energy spectrum of alkaline atom:
!2
me e2 1
En = − . (6.56)
2 4πǫ0 h̄ (n′ )2
Here the principal quantum number is equal to:
′ ′
n = l + nr + 1 , (6.57)
where nr is the radial quantum number.
′
The solution of the quadratic equation for l has the form:
s 2
′ 1 2C 1
l = l+ − − , (6.58)
2 a0 2
where
4πǫ0 h̄2
a0 =. (6.59)
me e 2
Here we took into account only positive root of quadratic equation. Under the square
root of Eq. (6.58) the second term is much smaller than the first term. Therefore, we
can use the following approximation:

′ 2C 1
l ≈l− = l − σ(l) . (6.60)
a0 2l + 1
Then,
′ ′
n = l + nr + 1 = l − σ(l) + nr + 1 = n − σ(l) . (6.61)
As a result the expression for alkaline atom energy spectrum has the form:
!2
me e2 1
En,l =− . (6.62)
2 4πǫ0 h̄ [n − σ(l)]2
From the last equation it follows that in alkaline atoms the energy of valence electron
depends not only on the principal quantum number, n, but also on the orbital quantum
number, l.

Problem 6.7. Find the angular momentum for electrons in an atom in 3s, 4d, and
5g states and the maximum number of electrons in an atom with the same sets of
quantum numbers n, l, and m.

Solution. The angular momentum for an electron in an atom is:

 57

q
L= l(l + 1)h̄ . (6.63)
Here the azimuthal quantum number, l, is an integer number in the interval 0 ≤ l ≤
n − 1, where n is the principal quantum number. Electrons with the same principal
quantum number form the atom’s shell. Electrons with the same principal and orbital
quantum numbers form subshells. They are denoted as s, p, d, f, g, ... . The corre- √
sponding
√ quantum numbers are l = 0, 1, 2, 3, 4, ... . Therefore, L 3s = 0, L 4d = 6h̄,
L5g = 20h̄.
According to the Pauli’s exclusion principle there cannot exist two or more electrons
with the same set of quantum numbers (n, l, m, ms ) in an atom . The magnetic (spin)
quantum number, ms , defines the projection of the spin angular momentum, S:
1
Sz = ms h̄ , ms = ± . (6.64)
2
Therefore, the maximum number of electrons with the same set of quantum numbers
(n, l, m) is equal to two.
Chapter 7

Quantization in nanostructures

Problem 7.1. Find the expressions for the wavefunction and electron energy in a
symmetric rectangular quantum well with the following potential profile:
(
U0 , |x| > Lx ,
U (r) = (7.1)
0, |x| ≤ Lx .

The electron motion along the y- and z-directions is free.

Solution. The electron wavefunction in the quantum well can be written in the fol-
lowing form:
ψ(x, y, z) = X(x)Y (y)Z(z) . (7.2)
After separation of variables in the Schrödinger equation

h̄2 2
" #
− ∇ + U (r) ψ(r) = Eψ(r) (7.3)
2me

the electron motion along x-axis in each of the regions is defined by the following
equations:
d 2 X1
+ kx2 X1 = 0 , |x| ≤ Lx , (7.4)
dx2
d 2 X2
− κ2x X2 = 0 , |x| > Lx , (7.5)
dx2
where we introduced for the electron wavenumbers inside, kx , and outside, κx , of quan-
tum well the following expressions:
s
2me Ex
kx = , (7.6)
h̄2
s
2me (U0 − Ex )
κx = . (7.7)
h̄2

59
60 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

The electron energy and wavefunctions of the stationary states are defined by the
following equations:
h̄2 2
E = Ex + (k + kz2 ) , (7.8)
2me y

ψ(x, y, z) = X(x)ei(ky y+kz z) , (7.9)

(
AF (kx x) , |x| ≤ Lx ,
X(x) = (7.10)
Be−κx , |x| > Lx .
Here the function F for even states is defined as:

F (u) = cos u (7.11)

and for odd states:

F (u) = sin u . (7.12)
The energy spectrum Ex is discrete, and the spectrum that corresponds to the free
motion in yz-plane is continuous and parabolic. Thus, electron energy spectrum in the
quantum well consists of a discrete set of parabolic subbands.
The continuity condition of the wavefunction and its derivative at the points |x| = Lx
gives the transcendental equations which define the energy spectrum Ex for even and
odd states, respectively:
s
Lx q U0
 
tan 2me Ex = − 1, (7.13)
h̄ Ex
s
Lx q U0
 
cot 2me Ex = − − 1. (7.14)
h̄ Ex
The equation for even states has Ns solutions that correspond to the same number
of discrete energy levels Ex . The number of even states, Ns , in the quantum well is
defined by the inequalities:
Lx q
(Ns − 1)π < 2me U0 ≤ Ns π . (7.15)
h̄
From these inequalities it follows that for any parameters of the quantum well there is
at least one energy level Ex that corresponds to the even state. In this case Ns = 1
and the condition that there exists at least one energy level is the inequality:

π 2 h̄2
L2x U0 ≤ . (7.16)
2me
The number of roots for the odd states is defined by the inequalities:
π Lx q π
Na π − < 2me U0 ≤ Na π + . (7.17)
2 h̄ 2
 61

The first odd state (i.e. Na = 1) appears when the following condition is satisfied:

π 2 h̄2
L2x U0 > . (7.18)
8me

Problem 7.2. Using the solution of the Problem 7.1 consider the limiting cases of
very deep and very shallow quantum wells. Show that in a shallow quantum well there
is always at least one stationary state. Derive an approximate expression for energy,
E1 , of this state and estimate it for U0 = 2.5 × 10−3 eV and L = 2 nm. (Answer: E1 ≈
2.375 meV.)

Solution. In the case of the deep quantum well its height must satify the condition

h̄2
U0 >> . (7.19)
2me L2
For the lower energy levels, Ex , which are close to the bottom of the potential well,
U0
>> 1 . (7.20)
Ex
Taking into account these inequalities and equations for Ex for the even and odd states
(Eqs. (7.13) and (7.14) from the solution of Problem 1) we arrive to the following
general expression for these states:
 v 
2 2 2 2
u
π h̄ n  u 2h̄ 
Ex ≈ 1 −
t
 , n = 1, 2, 3, ... (7.21)
8me L2 me U0 L2

We see that the energy levels in the potential well with high but finite potential barriers
are lower than the energy levels in the potential well with the barriers of infinite height.
The given U0 = 2.5 × 10−3 eV and L = 2 nm satisfy the condition of shallow well since

h̄2
U0 << ≈ 10−2 eV . (7.22)
2me L2
As it was shown in Problem 7.1 in the case of L2 U0 ≤ π 2 h̄2 /(2me ) there is only one
state and it is the ground state with the energy Ex1 . If in Eq. (7.13), from the solution
of Problem 1 for even states, at Ex1 < U0 the argument of tangent is small (tan x ≈ x
for small x), then this expression can be written in the form:
s
Lq U0
2me Ex1 ≈ −1 (7.23)
h̄ Ex1
or
L√ q
2me Ex1 ≈ U0 − Ex1 . (7.24)
h̄
62 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

The last equation can be rewritten in a quadratic form:

2 h̄2 h̄2
Ex1 + E x1 − U0 = 0 . (7.25)
2me L2 2me L2
By solving Eq. (7.25)we obtain the following expression for Ex1 :

v s 
2 4 2 2
L2
u
h̄ u h̄ h̄ h̄ 8me
Ex1 = − +t + U0 =  1+ U0 − 1 ≈
4me L2 2
16me L 4 2me L 2 4me L 2 h̄2

2me L2
!
≈ U0 1− U0 . (7.26)
h̄2
We got this result taking into account three terms of expansion of square root function
in series over the small parameter 8me L2 U0 /h̄2 . By substituting the given parameters
we get:
Ex1 ≈ U0 (1 − 0.25) = 1.875 × 10−3 eV . (7.27)
From the obtained results it follows that the energy level in shallow well is close to the
top of the well.

Problem 7.3. In an asymmetric rectangular quantum well the motion of an electron in

the yz-plane is free and motion in the x-direction is limited by the following potential:

 U1 ,
 x < 0,
U (x) =  0, 0 ≤ x ≤ Lx , (7.28)

U2 , x > Lx ,
where U2 < U1 . Find the condition for which the discrete quantum states with energy
E ≤ U2 exist. Consider the limiting cases U1 → ∞ and U1 = U2 .

Solution. The electron wavefunction in the considered quantum well can be written
as
ψ(x, y, z) = X(x)Y (y)Z(z) . (7.29)
After separation of variables in the Schrödinger equation the electron motion along
x−axis in each of the regions is defined by the function X(x):
d2 X
− κ21 X = 0 , x < 0 , (7.30)
dx2

d2 X
− κ22 X = 0 , x > Lx , (7.31)
dx2
d2 X
2
+ k 2 X = 0 , 0 ≤ x ≤ Lx , (7.32)
dx
 63

where we introduced
s
2me (U1 − Ex )
κ1 = , (7.33)
h̄2
s
2me (U2 − Ex )
κ2 = , (7.34)
h̄2
and s
2me Ex
kx = . (7.35)
h̄2
For the stationary states the expressions for the electron energy and wavefunctions,
which are limited at x → ±∞, are defined as follows

h̄2 2
E = Ex + (ky + kz2 ) , (7.36)
2me

ψ(x, y, z) = X(x)ei(ky y+kz z) , (7.37)

κ1 x

 Ae
 , x < 0,
X(x) = B cos(kx x + δ) , 0 ≤ x ≤ Lx , (7.38)
Ce−κ2 x , x > Lx .



Let us consider the solution that corresponds to the highest energy level in the quantum
well, i.e. which corresponds to Ex = U2 . In this case:
s
2me (U1 − U2 )
κ1 = , (7.39)
h̄2

κ2 = 0 , (7.40)
s
2me U2
kx = . (7.41)
h̄2
The continuity conditions for the function X(x) and its derivative at the points x = 0
and x = Lx give us:

A = B cos δ , (7.42)

κ1 A = −kx B sin δ , (7.43)

B cos(kx Lx + δ) = Ce−κ2 Lx , (7.44)

−kx B sin(kx Lx + δ) = −κ2 Ce−κ2 Lx . (7.45)

64 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

Taking into account the condition (7.40), the system of Eqs.(7.42)-(7.45) can be rewrit-
ten as:

A = B cos δ , (7.46)

κ1 A = −kx B sin δ , (7.47)

B cos(kx Lx + δ) = C , (7.48)

−kx B sin(kx Lx + δ) = 0 . (7.49)

Equation (7.49) is satisfied if:
δ = −kx Lx . (7.50)
Taking into account Eq. (7.50) we obtain from Eq. (7.48):
B =C. (7.51)
Let us divide Eq. (7.47) by Eq. (7.46):
κ1 = −kx tan δ , (7.52)
or
κ1 = kx tan(kx Lx ) . (7.53)
Thus, we can write the system of Eqs. (7.46) - (7.49) as:
A = B cos(kx Lx ) , (7.54)

κ1 = kx tan(kx Lx ) , (7.55)

B =C, (7.56)

δ = −kx Lx . (7.57)
The relationship (7.55) is the condition of the existence of the solution which does not
increase with x → ±∞. In explicit form it can be written as:
s
Lx U1
q 
tan 2me U2 = − 1. (7.58)
h̄ U2
The last equation is also a condition of ocurrence of new electron stationary states in
a quantum well as it becomes deeper. If kx Lx < π/2, then the condition of ocurrence
of the first state of discrete spectrum is the inequality
s
q Lx U1
2me U2 ≥ arctan − 1. (7.59)
h̄ U2
 65

In particular, at U1 → ∞ this condition takes the form:

π 2 h̄2
U2 ≥ . (7.60)
8me L2x

At U1 = U2 this inequality is valid for any parameters of the quantum well.

Problem 7.4. Estimate the energy of the electron ground state in a quantum well,
where electron motion in yz-plane is free but motion along the x-direction is limited
by the potential U (x): (
∞, x < 0,
U (x) = (7.61)
bx, x ≥ 0.
Solution. Let us write the electron wavefunction of the ground state as:
(
i(ky y+kz z) 0, x < 0,
ψ(x, y, z) = e −δx (7.62)
xe , x > 0.

From the condition that the average electron energy in the ground state must be
minimal we will find the parameter δ:
h 2
i
h̄
ψ ∗ (x, y, z) − 2m ∇2 + U (x) ψ(x, y, z)dV
R
e
< E >= . (7.63)
|ψ(x, y, z)|2 dV
R

Let us substitute the expressions for U (x) and ψ(x, y, z) into the integral of the last
equation:
R∞ h 2 2
i
h̄ d
0 xe−δx − 2m e dx
2 + bx xe
−δx
dx h̄2 2
< E >= R∞ + (k + kz2 ) . (7.64)
0 x2 e−2δx dx 2me y

After carrying out the integration we get:

3b h̄2 2
< E >= + (δ + ky2 + kz2 ) . (7.65)
2δ 2me
To find the minimum of this function, let us equate to zero its derivative over the
parameter δ:
d 3b h̄2 δ
< E >= − 2 + = 0. (7.66)
dδ 2δ me
Thus, the average electron energy in the ground state of a quantum well is minimal
(for the same values of ky and kz ) when the parameter δ is the following:
!1/3
3bme
δ= . (7.67)
2h̄2
66 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

In this case:
h̄2 2
2/3
3 3

< E >= bh̄me + (k + kz2 ) . (7.68)
2me 2 2me y

Problem 7.5. Find the energy and the wavefunctions of the stationary states of
an electron in a quantum wire. The electron motion in the z-direction is free and
motion in the plane perpendicular to the z-direction is limited by the potential U (ρ)
(ρ2 = x2 + y 2 ): (
0, ρ ≤ a,
U (ρ) = (7.69)
∞, ρ > a .
Solution. Let us write the electron wavefunction of the ground state as:

ψ(ρ, ϕ, z) = R(ρ)ei(mϕ+kz z) . (7.70)

Let us substitute this wavefunction into the general Schrödinger equation. After sepa-
ration of variables for the radial function R(ρ), we obtain the equation:

d2 m2
!
1 d 2
+ + kρ − R(ρ) = 0 , ρ ≤ a , (7.71)
dρ2 ρ dρ ρ2
where we took into account that
q
2me Eρ
kρ = (7.72)
h̄
and at the boundary of the quantum well the wavefunction becomes equal to zero. The
solution of Eq. (7.71), which is limited at the point ρ = 0, has the form:

R(ρ) = AJm (kρ ρ) , (7.73)

where Jm (kρ ρ) is a Bessel function. The boundary condition R(a) = 0 allows us to find
the discrete energy spectrum of the electron radial motion in the quantum wire:

h̄2
Eρ(lm) = 2
λ2lm , (7.74)
2me a
where λlm is the root of the Bessel function, i.e. Jm (λlm ) = 0. The first two roots of
the Bessel function are:
λ10 ≈ 2.4 (7.75)
and
λ11 ≈ 3.83 . (7.76)
Therefore, the energy of the electron radial motion in the quantum wire is:

h̄2
Eρ(10) ≈ 2.88 , (7.77)
m e a2
 67

h̄2
Eρ(11) ≈ 7.33 . (7.78)
m e a2
Thus, the electron energy in a quantum wire is composed of subbands. The energy in
the corresponding subband is defined by the following expression:

h̄2 λ2lm
!
Elm (kz ) = Eρ(lm) + E(kz ) = + kz2 . (7.79)
2me a2

Problem 7.6. In a quantum wire the electron motion along the z-direction is free and
in the plane perpendicular to the z-direction is limited by the potential U (x, y):

β(x2 + y 2 )
U (x, y) = . (7.80)
2
Find the energy, the wavefunctions, and the magnitudes of various projections of the
electron angular momentum onto the z-direction and occupation probabilities for the
states with quantum numbers nx = 1 and ny = 1.

Solution. The wavefunctions and the energy of electron in the given quantum wire
can be written as follows:

ψnx ny (x, y, z) = ψnosc

x
(x)ψnosc
y
(y)eikz z , (7.81)

h̄2 kz2
EN (kz ) = h̄ω(N + 1) + , N = 0, 1, 2, ... , (7.82)
2me
q
where N = nx + ny and ω = β/me . Linearly-independent wavefunctions ψnx ny
with quantum numbers nx = 0, 1, 2, ..., N and ny = N, N − 1, ..., 1, 0 correspond to
the energy subband EN (kz ) with given N (i.e. energy subband EN (kz ) is (N + 1)-
fold degenerate). Using the explicit form of the wavefunctions of harmonic oscillators,
ψnosc
x
(x) and ψnosc
y
(y), let us write the electron wavefunction in the quantum wire for the
states with nx = ny = 1:
2xy 2 2 2
ψ11 (x, y, z) = √ 3 e−(x +y )/(2a ) eikz z . (7.83)
πa
This wavefunction can be written in the cylindrical system of coordinates as:
2 cos ϕ sin ϕ 2 − ρ22 ikz z
ψ11 (ρ, ϕ, z) = √ 3 ρ e 2a e . (7.84)
πa
Taking into account that
1  2iϕ 
2 cos ϕ sin ϕ = sin 2ϕ = e − e−2iϕ , (7.85)
2i
68 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

we conclude that since m = ±2, then in the state with the wavefunction ψ11 the
projection of angular momentum on the z−axis can have only two values: Lz = ±2h̄.
The probabilities of these two values are the same and equal to 1/2.

Problem 7.7. The electron motion in a spherical quantum dot happens in the follow-
ing potential:
U (r) = U0 e−r/b , U0 < 0 . (7.86)
Find the average electron energy of a ground state in the given quantum well.

Solution. To estimate the average electron energy in the spherical quantum dot, let us
write the radial part of the electron wavefunction of the ground state in the following
form:
ψ(r) = Ae−αr/(2b) . (7.87)
The normalization condition for this function allows us to find constant A:
!3
Z
2 2
Z ∞
−αr/b 2 2 b
|ψ(r)| dV = 4πA e r dr = 8πA = 1. (7.88)
0 α

From the last equation we find: s

α3
A= . (7.89)
8πb3
Let us find now the average values of electron kinetic, < K >, and potential, < U >,
energies in the quantum dot:
2
h̄2 Z dψ 2 h̄
2
α2 Z ∞ −αr/b 2 h̄2 α2
< K >= dV = 4πA e r dr = , (7.90)
2me dr 2me 4b2 0 8me b2
3
∞ α
Z Z 
< U >= U0 ψ ∗ (r)e−r/b ψ(r)dV = 4πU0 A2 e−(α+1)r/b r2 dr = U0 .
0 1+α
(7.91)
The condition of the minimal total average energy,

h̄2 α2
3
α

< E >= 2
+ U0 , (7.92)
8me b 1+α
is:
d<E> h̄2 α 3α2
= − |U0 | = 0. (7.93)
dα 4me b2 (1 + α)4
Solving last equation numerically we find the root equal to α0 . Then, the average
electron energy in the ground state is equal to:

h̄2 α02
3
α0

< E >= 2
+ U0 . (7.94)
8me b 1 + α0
 69

Figure 7.1: Double quantum well potential.

Problem 7.8. The electron motion takes place in an infinitely deep double-quantum-
dot structure and is limited by the potential U (x, y, z):

U (x, y, z) = U (x)U (y)U (z) , (7.95)

where 
 ∞,
 x < 0 , x > Lx ,
U (x) =  0, 0 < x < b , Lx − b < x < Lx , (7.96)

U0 , b ≤ x ≤ Lx − b ,
(
∞, y < 0 , y > Ly ,
U (y) = (7.97)
0, 0 ≤ y ≤ Ly ,
(
∞, z < 0 , z > Lz ,
U (z) = (7.98)
0, 0 ≤ z ≤ Lz .
Find using first-order perturbation theory the expression for the electron energy states
in a double-quantum-dot structure.

Solution. Let us consider the ”inner part”’ of the potential U (x) (see Fig. 7.1) as a
perturbation, i.e. we will consider as unperturbed the potential:
(
∞, x < 0 , x > Lx ,
Uunper (x) = (7.99)
0, 0 < x < Lx ,

and the potential (

0, x < b, x > Lx − b ,
Uper (x) = (7.100)
U0 , b ≤ x ≤ Lx − b ,
as a perturbation.
70 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

The solution of the Schrödinger equation for the unperturbed Hamiltonian gives us the
following expressions for the unperturbed wavefunctions and energy:

ψn(0)
x ny nz
(x, y, z) = ψn(0)
x
(x)ψny (y)ψnz (z) , (7.101)

where s
2 πnx x
ψn(0) (x) = sin , (7.102)
x
Lx Lx
s
2 πnα α
ψnα (α) = sin , α = y, z , (7.103)
Lα Lα
π 2 h̄2 n2x n2y n2z
!
En(0)
x ny nz
= En(0) + E ny + E nz = + + . (7.104)
x
2me L2x L2y L2z
The correction to energy En(0)
x
for the first order of perturbation theory is defined as
follows:
Z
2U0 Z Lx −b 2 πnx x
En(1) = ψn(0)∗ Uper (x)ψn(0) dx = sin dx . (7.105)
x x x
Lx b Lx
After integration we obtain:
!
U0 Lx 2πnx b
En(1) = Lx − 2b + sin . (7.106)
x
Lx πnx Lx
Thus, the approximate expression for the electron energy in the quantum dot has the
form:

π 2 h̄2
n2x n2y n2z
! !
U0 Lx 2πnx b
En(0)
E nx ny nz = x ny nz
+ En(1) = 2
+ 2
+ 2
+ Lx − 2b + sin .
x
2me
Lx Ly Lz Lx πnx Lx
(7.107)
The condition when this expression is valid is:

π 2 h̄2 n2x
U0 ≪ . (7.108)
me L2x

This means that the approximation better describes the electron energy state in quan-
tum dot when the quantum number nx is big enough and the height of the barrier U0
in the quantum well is low.

Problem 7.9. Prove that in isotropic three-dimensional systems the density of states
per unit volume of momentum space (the number of states per unit interval of momen-
tum modulus) is proportional to the square of momentum.
 71

Solution. Let us consider a free electron in a three-dimensional potential box with

dimensions Lx , Ly , and Lz . The electron wavefunction of a free electron is a plane
wave:
ψ(r) = Aeikr (7.109)
or
ψ(x, y, z) = Aei(kx x+ky y+kz z) , (7.110)
where the wavevector, k, is related with the electron momentum, p, as:
p
k= , (7.111)
h̄
or
pα
kα = , α = x, y, z . (7.112)
h̄
For convenience we consider cyclic boundary conditions for the potential box:

ψ(x, y, z) = ψ(x + Lx , y, z) = ψ(x, y + Ly , z) = ψ(x, y, z + Lz ) . (7.113)

By substituting the wavefunction (7.110) into the last expression we find that:

eikx Lx = eiky Ly = eikz Lz = 1 . (7.114)

From the last equation we find the conditions for the wavevector components:

2π 2π 2π
kx = nx , ky = ny , kz = nz , (7.115)
Lx Ly Lz

where nα = 0, ±1, ±2, ... and α = x, y, z. Thus, because the electron motion is limited
by the boundaries of the region, its wavevector and momentum can have only discrete
values. The volume of k-space that corresponds to a single quantum state is equal to:

(2π)3
∆Vk = , where Vr = Lx × Ly × Lz . (7.116)
Vr

The number of electron states, that corresponds to a unit volume d3 k = dkx dky dkz ,
taking into account two possible spin orientations, is equal to:

d3 k 2Vr 3
dN (k) = = d k. (7.117)
∆Vk (2π)3

From the last expression follows that the number of states that correponds to the unit
volume of k-space, i.e. the density of states, does not depend on the magnitude of the
wavevector:
dN (k) 2Vr
g(k) = 3
= . (7.118)
dk (2π)3
72 CHAPTER 7. QUANTIZATION IN NANOSTRUCTURES

This means that in the entire k-space the allowed states are distributed with the same
density. Similar expressions can be obtained for the p-space:
2Vr 3
dN (p) = d p, (7.119)
(2πh̄)3

dN (p) 2Vr
g(p) = 3
= . (7.120)
dp (2πh̄)3
Let us take into account that in the case of isotropic k-space the surfaces of equal
energy are spheres whose volume is equal to:

4πk 3
Vk = . (7.121)
3
The volume of the elementary spherical shell of radius k and thickness dk is equal to:

dVk = d3 k = 4πk 2 dk . (7.122)

The following number of states corresponds to this volume:

2Vr 2 k2
dN (k) = 4πk dk = Vr 2 dk . (7.123)
(2π)3 π

From the last expression we find corresponding density of states:

dN (k) k2
g(k) = = Vr 2 . (7.124)
dk π
In a similar fashion the number and density of states in p-space can be found:

4πp2 p2
dN (p) = 2Vr dp = Vr 2 3 dp , (7.125)
(2πh̄)3 π h̄

dN (p) p2
g(p) = = Vr 2 3 . (7.126)
dp π h̄

Problem 7.10. The electron motion in a quantum wire is free along the z-direction
and that in the xy-plane is limited in the region 0 < x < Lx and 0 < y < Ly by barriers
of infinite height. Find the number and density of states in the lowest-energy subband
of the quantum wire.
Solution. For simplicity we assume that Lx = Ly = L. Then, the electron wave-
functions and energy spectrum in the given quantum wire are defined by the following
expressions: s
4 πnx x πny y ikz z
ψnx ny (x, y, z) = 2
sin sin e , (7.127)
L Lz L L
 73

h̄2
" #
πnx 2 πny 2
  
E(nx , ny , kz ) = + + kz2 , (7.128)
2me L L
where we assumed that L ≪ Lz . According to these expressions the energy spectrum of
a quantum wire consists of the separate overlapping subbands (parabolas) Enx ny (kz ),
where quantum numbers nx and ny define the subband numbers. The number of
quantum states that corresponds to the interval dkz in the unit volume of the quantum
wire is:
dkz Lz dkz
dN = = . (7.129)
Vr 2π 2πL2
For each subband we can write:
s
2me
kz (E) = (E − Enx ny ) , (7.130)
h̄2
where the magnitude
h̄2
" 2 2 #
πnx πny

E nx ny = + (7.131)
2me L L
is the bottom of the subband with the corresponding quantum numbers nx and ny .
Knowing the dependence kz (E) we can find dkz :
v
2me dE
u
u
dkz = t . (7.132)
E − Enx ny 2h̄
Taking into account the last expression we obtain for the number of states in each
subband: v
1 u 2me
u
dN = 2
t dE . (7.133)
πh̄L E − Enx ny
In Eq. (7.133) we took into account that the number of states is doubled because of
the spin degeneracy and the fact that in each subband two intervals of wavenumbers
±dkz correspond to one energy interval dE. As a result, for the density of states in
each subband of quantum wire we can write:
v
dN 1 u 2me
u
gnx ny (E) = = 2
t . (7.134)
dE πh̄L E − Enx ny
For the lowest subband, quantum numbers nx and ny are equal to unity: nx = ny = 1.
Therefore, the number and density of states in this subband, found from Eqs. (7.133)
and (7.134), are: s
1 Z∞ 2me
N11 = 2
dE (7.135)
πh̄L E11 E − E11
and s
1 2me
g11 (E) = 2
. (7.136)
πh̄L E − E11
Appendix A

Classical dynamics of particles and

waves

Problem A.4.1. A ball of mass m is thrown at point x = y = 0 at initial time t = 0

with initial momentum p0 = mv0 , at an angle α = 45◦ (see, Fig. A.1). Find the de-
pendence of the magnitude of the ball’s momentum on time during the ball’s motion,
the minimal magnitude of ball’s momentum, and the magnitude of the momentum at
the time when the ball hits the ground. Write the expression for the ball’s trajectory.
Disregard the friction of air.

Solution. Let us choose an equation for the ball’s motion in the form of Eq. (A.5)
from the textbook. After substitution of the gravitational force, Fgr = mg (Eq. (A.7)
from the texbook), into Eq. (A.5) we get:

dp(t)
= mg . (A.1)
dt
Let us rewrite this equation in the form of projections on coordinate axes x and y:

dpx
=0 (A.2)
dt
dpy
= −mg , (A.3)
dt
where g = |g|. The integration of Eq. (A.2) shows that px does not depend on time,
t. Therefore, it remains equal to its initial value of the projection of vector p0 on the
x-axis:
px (t) = p0x = p0 cos α , (A.4)
i.e. p0x , remains constant as there are no forces acting on the ball along x-axis.
Integration of Eq. (A.3): Z Z
dpy = − mgdt , (A.5)

75
76 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

Figure A.1: Trajectory of the ball with the initial momentum p0 and the initial angle
α.

gives Z
py = −mg dt = −mgt + C . (A.6)
Here the integration constant, C = p0y , is found from the condition that at t = 0 the
initial momentum is py (0) = p0y = p0 sin α. As a result we obtain:
py (t) = p0y − mgt = p0 sin α − mgt . (A.7)
Let us find the momentum magnitude, |p| = p, at any instant, t:
q q
p(t) = p2x + p2y = p20 cos2 α + (p0 sin α − mgt)2 =
q
= p20 − 2p0 mgt sin α + m2 g 2 t2 . (A.8)
Function p(t) of Eq. (A.8) is a non-monotonic function of time. The time tm that
corresponds to the minimal momentum can be found from the equation
dp
= 0. (A.9)
dt
Taking into account Eq. (A.8) and (A.9), tm is equal to:
p0
tm = sin α . (A.10)
mg
By substituting t by tm in Eq. (A.8), we find the minimal momentum, pm , that the
particle has during its motion:
pm = p(tm ) = p0 cos α . (A.11)
 77

The result of Eq. (A.11) can also be found if we take into account that at time tm the
value of py becomes equal to zero (check Eq. (A.7)).
The values of time, tm , and momentum, pm , correspond to the highest point of the
ball’s trajectory, which, as we will show now, has the form of an “inverted” parabola
(see Fig. A.1). For this purpose let us write Eq. (A.4) from the textbook in the form
of projections on the coordinate axes x and y and take into account the solutions,
Eqs. (A.4) and (A.7), and the relation between v and p (Eq. (A.3) from the textbook):

dx
m = p0x (A.12)
dt
and
dy
m
= p0y − mgt . (A.13)
dt
The integration of Eqs. (A.12) and (A.13) along with taking into account the initial
conditions, x(0) = y(0) = 0, gives us:
p0x
x(t) = t (A.14)
m
and
p0y gt2
y(t) = t− . (A.15)
m 2
By eliminating time, t, in these equations we obtain an equation for the trajectory of
ball’s flight:
p0y m2 g
y= x − 2 x2 , (A.16)
p0x 2p0x
which is an inverted parabola.
The total time of the ball’s flight is equal to:

ttot = 2tm (A.17)

because of the symmetry of the ball’s trajectory. Thus, the ball’s momentum at time
ttot is equal to:
p(ttot ) = p0 , (A.18)
i.e. the ball falls on the ground with the same magnitude of the momentum modulus
(and, respectively, with the same velocity), which it had at the initial time, t = 0. If
the angle of launch is α, then the angle of fall, α′ , is:

α′ = π − α . (A.19)

Knowing the ball’s time of flight we can find the range:

2
p0x p0 cos α p0 sin 2α

x(ttot ) = vtot ttot = ttot = 2 tm = . (A.20)
m m m g
78 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

Equation (A.20) demonstrates that the distance x(ttot ) as a function of an angle α has
its maximum at α = 45◦ .

Problem A.4.2. A particle with mass m and initial velocity v0 enters a fluid, where a
retarding force, Frt = −bv, is applied to it. Find the time during which the particle’s
velocity drops to v0 /n (n is an integer number greater than 1), and find the distance
traveled by the particle during this time. Ignore the gravitational force acting on the
particle.

Solution. The equation of a particle’s motion can be chosen in the form of Eq. (A.11)
from the textbook. Its projection on the direction of motion can be written as:
dv
m = −bv . (A.21)
dt
Let us rewrite the last equation in the form:
dv b
= − dt (A.22)
v m
and integrate the left-hand and the right-hand sides of the obtained equation:

ln v = −γt + ln C , (A.23)

where γ = mb and C is an integration constant. From the last equation taking into
account the initial condition, v(0) = v0 , we obtain:

v(t) = Ce−γt = v0 e−γt . (A.24)

This equation was obtained assuming that the particle comes to rest only at t → ∞.
As the velocity decreases exponentially with time, the parameter γ that characterizes
the rate of this decrease is referred to as a damping constant. From Eq. (A.24) we find
that the particle attains velocity v0 /n during the time
1 m
tn = ln n = ln n . (A.25)
γ b
To find the distance travelled by the particle, let us rewrite Eq. (A.24), taking into
account Eq. (A.4) from the textbook, as:
dx(t)
= v0 e−γt . (A.26)
dt
Here, x is a coordinate along the direction of motion of a particle with the initial
condition x(t = 0) = 0. By integration of Eq. (A.26) we obtain:
v0 −γt v0  
x(t) = − e + C1 = 1 − e−γt , (A.27)
γ γ
 79

where the integration constant, C1 = v0 /γ, is found from the initial condition for the
particle’s coordinate. It follows from Eq. (A.27) that until the particle stops (at t → ∞)
the particle will travel the distance
v m
xstp = = v0 ,
γ b

and during the time tn , which is defined by Eq. (A.25), this distance will be:
v0 m 1
 
xn = x(tn ) = 1− . (A.28)
b n

Problem A.4.3. A particle with mass m begins its motion from the state of rest under
the influence of an oscillating force F (t) = F0 sin(γt). Find the particle’s velocity and
the distance traveled at time t1 , when the force reaches its first maximum and at time
t2 , when the force becomes equal to zero.

Solution. Let us choose the x-axis along the direction of particle’s motion. The initial
position of the particle and its initial velocity are x(0) = 0 and v(0) = 0, respectively.
The equation for particle’s motion according to Eq. (A.12) from the textbook can be
written in the form:
d2 x
m 2 = F0 sin(γt) . (A.29)
dt
The solution of this equation can be found in two steps, namely – by its integration
repeated twice. After the first integration of Eq. (A.29), we find the particle’s velocity:

dx F0
v(t) = =− cos(γt) + C . (A.30)
dt γm

Taking into account the initial condition, v(0) = 0, this expression can be rewritten as:

F0
v(t) = [1 − cos(γt)] . (A.31)
γm

From the last equation we find that at time t1 = π/2γ, when the force is maximal
(F (t1 ) = F0 (sin γt1 ) = F0 ), the particle’s velocity is:

F0
v(t1 ) = . (A.32)
γm

In a similar fashion, at time t2 = π/γ, when the force becomes equal to zero (sin(γt2 ) =
0 and cos(γt2 ) = −1), the velocity reaches its maximum value:

F0
v(t2 ) = 2 . (A.33)
γm
80 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

By integrating Eq. (A.31) and taking into account the initial condition, x(0) = 0, we
find the dependence of the particle’s coordinate on time t:
!
Z
F0 1
x(t) = v(t)dt = t − sin(γt) . (A.34)
γm γ
The distances that the particle travels during time t1 and t2 are:
F0 π
 
x(t1 ) = −1 (A.35)
mγ 2 2
and
F0
x(t2 ) = π. (A.36)
mγ 2
Note that t2 = 2t1 and as it follows from Eqs. (A.35) and (A.36) the particle covers
almost four times shorter distance during time interval from 0 to t1 than from t1 to t2 .

Problem A.4.4. Two particles which have masses m1 and m2 and velocities before
the collision v1 and v2 , undergo a perfect inelastic collision. Find the velocity of the
particle formed after the collision and find the proportion of the mechanical energy
that has transformed into internal energy.

Solution. Let us write the equation that describes the principle of conservation of
momentum of two particles that experience perfectly inelastic collision (after perfectly
inelastic collision two particles move together as a whole):

m1 v1 + m2 v2 = (m1 + m2 )u . (A.37)

Thus, the velocity of the resulting particle, u, is:

m1 v 1 + m2 v 2
u= . (A.38)
m1 + m2
The energy which is transformed into internal energy after collision is defined by the
difference of the total energy of colliding particles and the energy of the resulting
particle:
2
1 m1 m2 µVrel
Einternal = (v1 − v2 )2 = , (A.39)
2 m1 + m2 2
where µ is the so-called reduced mass of colliding particles and Vrel = v1 − v2 is their
relative velocity. From Eq. (A.39) it follows that for m1 = m2 and v1 = −v2 , all the
energy of the colliding particles transforms into internal energy.

Problem A.4.5. Find the velocities of two particles after their perfectly elastic cen-
tral collision. The initial state of the two particles before the collision is given by the
 81

parameters m1 , v1 , m2 , and v2 . A central collision is defined as a collision that happens

along the line connecting the centers of the particles.

Solution. Let us write the equations that define the principles of conservation of
momentum and energy of colliding particles:

m1 v1 + m2 v2 = m1 u1 + m2 u2 , (A.40)

m1 |v1 |2 m2 |v1 |2 m1 |u1 |2 m2 |u2 |2

+ = + , (A.41)
2 2 2 2
where v1 and v2 are the velocities of the particles before the collision and u1 and u2
– after the collision. This system of equations, one of which, Eq. (A.40), is a vector
equation, we can solve without considering projections on the coordinate axes. Let us
regroup in each of Eqs. (A.40) and (A.41) the terms with the same masses. Taking
into account that
m1 |v1 |2 m1 |u1 |2 m1
− = (v1 + u1 )(v1 − u1 ) (A.42)
2 2 2
we obtain:
m1 (v1 − u1 ) = m2 (u2 − v2 ) , (A.43)
m1 (v1 − u1 )(v1 + u1 ) = m2 (u2 − v2 )(v2 + u2 ) . (A.44)
Let us divide Eq. (A.44) by Eq. (A.43). As a result we will obtain the linear system of
equations:
m1 (v1 − u1 ) + m2 (v2 − u2 ) = 0 ,
v1 + u1 − (v2 + u2 ) = 0 . (A.45)
By solving this system we find the unknowns, u1 and u2 :

(m1 − m2 )v1 + 2m2 v2

u1 = , (A.46)
m1 + m2
(m2 − m1 )v2 + 2m1 v1
u2 = . (A.47)
m1 + m2
From Eqs. (A.46) and (A.47) we find out that the heavier particle always transfers
momentum to the lighter particle and in the particular case of equal masses (m1 =
m2 = m), the particles, as a result of the collision, exchange their velocities:

u1 = v2 and u2 = v1 . (A.48)

Problem A.4.6. According to one of the first models of the hydrogen atom (Thom-
son’s model), an electron (a particle with mass me = 9.1 × 10−31 kg and negative
82 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

Figure A.2: Thomson’s model of hydrogen atom.

charge −e = −1.6 × 10−19 C) is located at the center of the atom and all the rest of
the atomic space is uniformly filled by a spherical cloud of positive charge, the total
charge of which is equal to +e, and, which does not hinder the motion of an electron.
Considering that the hydrogen atom’s radius is known (a ≈ 0.5 × 10−10 m) estimate
the oscillation frequency of an electron. (Answer: ωe ≈ 4.5 × 106 s−1 .)

Solution. Let us calculate the force, which acts on an electron, when it is at a distance
r(t) from the equilibrium position. At a uniform distribution of the positive charge
inside of a sphere of radius a only the part of the positive charge, qin , that is inside of
the sphere with radius r(t) affects the electron (see Fig. A.2).
The resultant force of all electric forces outside the sphere of distributed positive charge
is equal to zero. Therefore, the equation of electron motion under the action of the
“restoring” electric force has the form:

d2 r
me = −Fe , (A.49)
dt2
where
eqin
Fe = , (A.50)
4πǫ0 r2
ǫ0 = 8.85 × 10−12 F/m is the permittivity of free space, and

er3
qin = 3 (A.51)
a
is the positive charge inside of a sphere shown by the dashed line in Fig. A.2. The
 83

Figure A.3: Elliptic orbit of an electron in a field of a central force F (r)er .

force acting on an electron in such a model of atom,

e2 r
Fe = , (A.52)
4πǫ0 a3

is linearly proportional to the displacement, r. Therefore, Eq. (A.49) takes the form of
the equation of harmonic oscillations

d2 r
+ ωe2 r = 0 , (A.53)
dt2
where the oscillation frequency, ωe , is given by the equation
s
e2
ωe = . (A.54)
4πǫ0 me a3

By substituting the values of physical constants into Eq. (A.54), we obtain the mag-
nitude of oscillation frequency: ωe ≈ 4.5 × 1016 s−1 . An oscillating electron, according
to the laws of classical electrodynamics, has to emit constantly electromagnetic waves
of angular frequency ωe , i.e. in the ultraviolet part of electromagnetic spectrum.

Problem A.4.7. In Sommerfeld’s model of the hydrogen atom the electron rotates
around the proton along stationary elliptic orbits (see Fig. A.3). Find the electron’s
angular momentum in terms of the parameters of the orbit (the semimajor and semimi-
nor axes of the ellipse are a and b, respectively) and the period of the orbital rotation,
84 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

T . Find the electron’s angular momentum for a circular orbit.

Solution. Let us choose a part of the electron’s trajectory and define the position of
the electron at a time t by a vector r(t) and its elementary displacement by vector dr.
According to the definition, the angular momentum, L, is equal to:

dr
L = r × p = r × mv = mr × . (A.55)
dt
The vector-product r×dr is equal to the doubled square of the shaded area in Fig. A.3,
i.e. r × dr = 2dA. Taking into account this relation, Eq. (A.55) can be rewritten in
the following form:
dA
L = 2m (A.56)
dt
or
Ldt = 2mdA . (A.57)
By integrating the left-hand side of Eq. (A.57) with respect to time from 0 to T , and
the right-hand side from 0 to Aellipse , we obtain:

LT = 2mAellipse . (A.58)

Since Aellipse = πab, then

2πabm
L= . (A.59)
T
If ellipse degenerates into a circle, i.e. a = b = r, for a circular motion with the velocity
v defined by equation:
2πr
v= , (A.60)
T
we obtain the conventional equation for L:

L = mvr . (A.61)

Problem A.4.8. An electron with velocity v enters a uniform magnetic field B, which
is perpendicular to the electron’s velocity. Find the radius, r, of the circle along which
the electron is moving, its angular momentum, L, and its magnetic moment, µ. The
magnetic moment of an electron can be calculated according the formula

~µ = −ISeB , (A.62)

where I is the circular current induced by the electron’s motion along a circle, S is the
cross-sectional area of a current loop, and eB is a unit vector along vector B.
 85

Figure A.4: Electron in uniform magnetic field, B.

Solution. In the magnetic field, B, a magnetic force (Lorentz force), Fm , which is

perpendicular to velocity v, acts on the electron. This is why the electron moves along
a circular orbit (see Fig. A.4):
Fm = −ev × B . (A.63)
The force defined by Eq. (A.63) keeps the electron on a circular orbit. Therefore,
v2
a= . (A.64)
r
According to Eqs. (A.63) and (A.64) the Newton’s second law of motion in the projec-
tion on the direction of acceleration can be written in the form:
v2
me = evB . (A.65)
r
The radius of the circle is given by the expression:
me v v
r= = , (A.66)
eB ωc
where ωc = eB/me , which is called cyclotron frequency, is the angular velocity of the
electron’s rotation in the magnetic field.
The angular momentum of the electron, L, during its motion along a circle is constant:
L = r × p = me (r × v) . (A.67)
The magnitude of this vector is equal to:
m2e v 2
L = me vr = , (A.68)
eB
86 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

and the direction of vector L coincides with the direction of magnetic field, B .
To determine the magnetic moment from Eq. (A.62) let us calculate the circular current
and the cross-sectional area of the loop:
e ev
I=− =− , S = πr2 , (A.69)
T 2πr
where T is period or time of a single rotation of an electron along the circle. As a result
we obtain for the modulus of vector µ:
evr me v 2
µ=− =− . (A.70)
2 2B
The vectors ~µ and L are directed in opposite directions due to electron’s negative
charge. The relation between them, taking into account Eqs. (A.68), (A.70), and
(A.62), is given by the expression:
e
~µ = − L. (A.71)
2me

Problem A.4.9. A particle with positive charge q, mass m, and velocity v1 , enters
a region with an electric field. Associated with the electric force there is an electric
potential φ. While moving along the electric field lines from a point with electrostatic
potential φ1 to a point with electrostatic potential φ2 , the particle accelerates. Find
its final velocity v2 .

Solution. In accordance with the work-energy theorem and Eq. (A.46) from the
textbook, the potential energy of the electric field, U (r), is transformed into the kinetic
energy of the particle, K:
mv22 mv12
− = U (r1 ) − U (r2 ) , (A.72)
2 2
where v2 is the velocity of the particle at the point r2 with the potential U (r2 ) and v1
- at the point r1 . Introducing the electrostatic field potential, φ, related with potential
energy, U , by
U = qφ , (A.73)
we obtain the following expression for the particle’s final velocity:
s
2q(φ1 − φ2 )
v2 = v12 + . (A.74)
m
If the particle, with an initial velocity v1 equal to zero, accelerates in the potential
difference ∆φ = φ1 − φ2 , it reaches a velocity
s
2q∆φ
v2 = . (A.75)
m
 87

Figure A.5: A particle with a positive charge, q, in a homogeneous electric field, E.

Problem A.4.10. Show that the force acting on a particle in a homogeneous electro-
static field is conservative and that the field itself is a potential field.

Solution. Let us first calculate the work done to move charge q in a uniform electric
field E (see Fig. A.5). For a convenience let us choose the Cartesian coordinate system
with z-axis directed along electric field E = Ez k (k is a unit vector directed along
z-axis):
Z r2 Z r2 Z z2
W12 = Fe · dr = qE · dr = qEz dz = qE(z2 − z1 ) . (A.76)
r1 r1 z1

From the last equation we can see that the work done to move the charge from point
1 to point 2 depends only on the coordinate difference (z2 − z1 ) and does not depend
on the form of the particle’s trajectory. Therefore, the force
Fe = qE (A.77)
is conservative and the potential energy of the particle, according to Eq. (A.36) from
the textbook, is defined as
U (z) = −qzEz . (A.78)

Problem A.4.11. Show that the kinetic energy of a charged particle in a magnetic
field does not change.
88 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

Figure A.6: Linear chain of ions.

Solution. Let the particle, which is in a magnetic field, have a velocity v. The change
of a particle’s kinetic energy is defined as the work done by the magnetic force, Fm :
Z r2 Z r2
K2 − K1 = Fm · dr = q (v × B) · dr . (A.79)
r1 r1

In the scalar-vector product we can make the following identical permutations:

(v × B) · dr = dr · (v × B) = (dr × v) · B . (A.80)

Taking into account that v = dr/dt and vectors r and dr/dt are parallel, we obtain
that dr × v = 0, and the magnetic field does not perform work on the particle’s
displacement. Therefore, the particle’s kinetic energy in the magnetic field does not
change, i.e. K2 = K1 .

Problem A.4.12. Consider an infinite linear chain of ions with mass m and charge e.
The distance between the neighboring ions is a (Fig. A.6). Find the period of linear
oscillations of one of the ions, considering that all other ions are motionless. Estimate
the frequency, ω, of a chain of copper ions having e = 1.6 × 10−19 C, m = 1.06 × 10−25
kg, and a = 3 × 10−10 m. (Answer: ω ≈ 2 × 1014 s−1 .)

Solution. Let us take the origin of coordinates (x = 0) to be at the equilibrium

position of the ion under consideration, and let us direct the axis x along the chain of
ions. Then, all other ions will have coordinates xn = ±na, where n = 1, 2, ... Let us
write the expression for the potential energy of the ion under consideration at the time
 89

when it occupies the position x(t):

∞  ∞
1 1 2na

U (x) = ke e2 = ke e2
X X
+ 2 2 2
, (A.81)
n=1 na − x na + x n=1 n a − x

where ke = 1/(4πǫ0 ). Using the obtained expression, let us derive the expression for
the force acting on the displaced ion:
∞
dU 4nax
= −ke e2
X
F (x) = − 2 2
. (A.82)
dx 2 2
n=1 (n a − x )

Since the oscillations have small amplitude, then x ≪ a and we can neglect x in the
denominator of Eq. (A.82). Then, the expression for the force can be rewritten in the
form: ∞
4ke e2 x X 1 1.2e2
F (x) ≈ − ≈ − x, (A.83)
a3 n=1 n3 πǫ0 a3
where we took into account that ∞ 1
n=1 n3 ≈ 1.2. This expression coincides with
P

Eq. (A.85) from the texbook if we denote

1.2e2
k= . (A.84)
πǫ0 a3
Using Eq. (A.71) from the textbook we find the frequency of oscillations:
s s
k 1.2e2
ω= = (A.85)
m πǫ0 a3 m
and the period of oscillations
s
2π πǫ0 a3 m
T = = 2π . (A.86)
ω 1.2e2
By substituting the values for e, m, and a into the expression (A.85), we find ω =
2 × 1013 s−1 , which corresponds to the infrared region of electromagnetic spectrum.

Problem A.4.13. Consider a linear chain of ions that consists of ions with opposite
charges. The distance between ions is equal to a. Find the energy of electrostatic
interaction of a separate ion with all of the others, considering the number of ions, N ,
to be large. Find the energy of the whole chain of ions.

Solution. The energy of interaction (potential energy) of an ion in the field of the rest
of ions that are on the left and right sides can be written as:
1 1 1 1 2ke e2 1 1 1
   
Ui (a) = 2ke e2 − + − + − ... = −1 + − + − ... , (A.87)
a 2a 3a 4a a 2 3 4
90 APPENDIX A. CLASSICAL DYNAMICS OF PARTICLES AND WAVES

where ke = 1/(4πǫ0 ). By summing over the series

∞
1
(−1)n−1
X
= log 2 , (A.88)
n=1 n

we obtain
ke e2 α
Ui (a) = − , (A.89)
a
where α = 2 log 2 is the so-called Madelung’s constant, which is characteristic for a
linear geometry of an ionic crystal. Let us note, that in the case when the ion is
located at one of the ends of the chain, its energy is half of the energy of an ion located
inside of the chain. As the input of the outermost ions in the total energy of crystal is
small, then we can write the total energy of the whole chain of atoms as:

N e2 α
Utot (a) = N Ui (a) = − . (A.90)
4πǫ0 a
Appendix B

Electromagnetic fields and waves

Problem B.6.1. Maxwell’s equation (B.16) from the textbook can be rewritten in
the following form:
∇ × H = jdis + j , (B.1)
where we introduced the density, jdis , of the displacement current, which is measured
in [A m−2 ]:
∂D
jdis = . (B.2)
∂t
We see from Eq. (B.202) from the textbook that it is precisely the displacement cur-
rents, which are responsible for the appearance of vortex magnetic fields in the case
when there is no current, j, of external charges. Let a monochromatic plane wave
described by Eq. (B.56) from the textbook with an amplitude of the electric field of
E0 = 103 V m−1 and frequency ω = 3 × 1015 s−1 (red light), propagate in vacuum.
Find the amplitudes of the displacement current, jdis,0 , and magnetic field intensity,
H0 , of the wave.

Solution. Taking into account Eq. (B.56) from the textbook and Eq. (B.2) we obtain:
∂E
jdis = ǫ0 = −ǫ0 ωE0 sin(ωt − k · r) . (B.3)
∂t
From the last equation it follows that
jdis,0 = ǫ0 ωE0 . (B.4)
After substitution of the given parameters into Eq. (B.4) we obtain:
jdis,0 = 2.7 × 107 A m−2 . (B.5)
For comparison, if a conductor with 1 mm2 cross-section has a current I = 10 A, then
the current density is equal to j = 107 A m−2 . Substituting Eq. (B.3) into Eq. (B.1),
we obtain for the amplitudes of the magnetic field and displacement current (at j = 0)
the following expression:
kH0 = jdis,0 . (B.6)

91
92 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

Figure B.1: The interference of two plane waves with wavevectors k1 and k2 . Here
r1 = r cos ϕ1 and r2 = r cos ϕ2 .

Thus, the amplitude of the magnetic field intensity, H0 , is equal to:

c
H0 = jdis,0 ≃ 2.7 A m−1 . (B.7)
ω

Problem B.6.2. The point of interference, SP, of two monochromatic plane waves,
propagating in vacuum, is defined by the radius vector r, which forms angles ϕ1 and ϕ2
with the wavevectors, k1 and k2 , of the two waves (see Fig. B.1). Find the frequency of
the wave at the point SP, which corresponds to the intensity maximum of the wave field.

Solution. Let us define the phase difference, ∆, in Eq. (B.160) from the textbook.
By projecting the radius vector r on the directions of the wavevectors k1 and k2 (see
Fig. B.1), we obtain

∆ = ∆2 − ∆1 = k2 · r − k1 · r = r(k2 cos ϕ2 − k1 cos ϕ1 ) . (B.8)

As the waves propagate in the same medium (in vacuum), then k1 = k2 = ω/c.
To observe the maximum intensity at the given point, the following condition must be
satisfied:
rω
(cos ϕ2 − cos ϕ1 ) = 2πm , (B.9)
c
where m is any integer number. From the last expression it follows that at ϕ1 = ϕ2 the
zeroth order interference maximum will be observed for the entire range of frequency.
 93

Figure B.2: Young’s double-slit experiment. (a) White stripes at the screen correspond
to maxima of intensity and black stripes - to its minima. (b) Plane waves from the
distant linear source (e.g. Sun) transform into concentric wavelets, which originate
from the slits S0 , S1 , and S2 .

At ϕ1 6= ϕ2 the m-th interference maximum corresponds to

2πmc
ω= . (B.10)
r(cos ϕ2 − cos ϕ1 )

Problem B.6.3. Derive the equation for the location of maxima and minima in the in-
terference pattern. Assume that the setup of Young’s scheme (the two-slit experiment;
see Fig. B.9 from the textbook) is placed in water to perform the same experiment as
was done in the air.

Solution. Since the waves coming from the slits are not plane waves, their wave fronts
are cylindrical surfaces (see Fig. B.2) and intensities of each wave at the point SP are
defined by the expression:
G
Ij = , (B.11)
rj
where G is a constant which has the dimensionality [W m−1 ] and which depends on
the intensity of the wave in the slit plane. Then, Eq. (B.158) from the textbook can
be rewritten as
G G G
I= + + 2√ cos ∆ . (B.12)
r1 r2 r1 r2
94 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

If the distance between slits, d, is much smaller than the distance l, then we can
consider that r1 ≈ r2 ≈ r. Then, Eq. (B.12) can be rewritten as:
4G ∆
I= cos2 , (B.13)
r 2
where the phase difference of two waves at the point x is given by the expression:
2πd
∆= x. (B.14)
λl
Here we took into account that for small θ we can write:
L x x
= ≈ . (B.15)
d r l
Thus, taking into account Eq. (B.13), the dependence of intensity on coordinate x is
defined as:
4G
I(x) = cos2 (γx) , (B.16)
r
where
πd
γ= . (B.17)
λl
√
Since the wavelength of the light wave in the water, λw , becomes equal to λw = λ/ ǫ
(µ = 1 for water), then the phase difference of the waves that come from two slits to
the given point is defined as:
√
2π 2π ǫ
∆w = (r2 − r1 ) = (r2 − r1 ) , (B.18)
λw λ
where r1 and r2 are the distances from the slits S1 and S2 to the given point and
ǫ = 1.33 is the dielectric constant of water in the optical range. At the same time
the distribution of the light field intensity at the screen is given by Eq. (B.13), where
the parameter ∆ has to be replaced by ∆w , which is defined by Eq. (B.18). Since the
relations between all these characteristic lengths, d, l, and x, for the given system, are
conserved, then analogously to Eq. (B.16) we can write
√ !
4G 2 πd ǫ
I(x) = cos x . (B.19)
r lλ

From Eq. (B.19) follows that the intensity minima are located at the points:
lλ 1
 
xmin = √ m − (B.20)
d ǫ 2
and the intensity maxima - at the points
lλ
xmax = √ m . (B.21)
d ǫ
 95

Figure B.3: The right-handed system.

√
Thus, the distance between maxima and minima decreases ǫ times, i.e. the interfer-
ence pattern in water becomes denser.

Problem B.6.4. The field intensities of electric and magnetic fields in a non-absorbing
medium oscillate in phase. Show that, for a wave in an absorbing medium, the oscilla-
tions of the vectors E and H do not coincide in phase. Find the relation of the phase
shift of these two vectors to the conductivity of the medium, σ.

Solution. Let us use Maxwell’s equations in the form of Eqs. (B.95) and (B.96) from
the textbook. Let us substitute in these equations the solutions for E and H in the
form of plane monochromatic waves (Eqs. (B.99) and (B.100) from the textbook).
After cancellation of the exponential factor in the left-hand and right-hand sides of the
obtained equations, we find the following two equations that couple vector amplitudes
E0 and H0 with the wavevector k:
k × H0 = −(ǫǫ0 ω − iσ)E0 , (B.22)
k × E0 = µµ0 ωH0 . (B.23)
Let us take into account that vectors E0 , H0 , and k compose the right-handed system
of orthogonal vectors. Therefore, the directions of vectors k × E0 and H0 coincide,
and the directions of vectors k × H0 and E0 are opposite to each other (Fig. B.3).
By choosing the y-axis for the direction of the wave propagation and by projecting
Eq. (B.22) on the z-axis and Eq. (B.23) on the x-axis, we obtain:
kH0 = (ǫǫ0 ω − iσ)E0 , (B.24)
96 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

kE0 = µµ0 ωH0 . (B.25)

From this system of equations we find the relation between the amplitudes of electric
and magnetic fields: s
µµ0 ω
E0 = H0 = ZH0 . (B.26)
ǫǫ0 ω − iσ
Here Z is the complex variable called wave resistance of the medium or impedance:
s v
µµ0 ω u µµ0 ω(ǫǫ0 ω + iσ)
u
Z= ≡t 2 2
. (B.27)
ǫǫ0 ω − iσ (ǫǫ0 ω) + σ

Using the relationships from the theory of complex numbers

q
u + iv = |w|eiϕ , (B.28)

where w is a complex variable, u and v are real and imaginary parts of w, respectively,
and the modulus |w| and the angle ϕ are defined as:
√
|w| = u2 + v 2 , (B.29)
v
ϕ = arctan , (B.30)
u
we can present Z in the form:
Z = |Z|eiϕ . (B.31)
Here, we separated the modulus and the phase of the complex variable, Z:
#1/4
(µµ0 ω)2
"
|Z| = , (B.32)
(ǫǫ0 ω)2 + σ 2

1 σ
ϕ = arctan . (B.33)
2 ǫǫ0 ω
In an absorbing medium the magnetic field changes as (see Eq. (B.110) from the text-
book):
′′ ′ ′′ ′
H = Re[H0 e−k y ei(ωt−k y) ] = H0 e−k y cos(ωt − k y) . (B.34)
Then, the electric field changes as:
′′ ′ ′′ ′
E = Re[H0 |Z|e−k y ei(ωt−k y+ϕ) ] = H0 |Z|e−k y
cos(ωt − k y + ϕ) . (B.35)

Thus, in an absorbing medium the oscillations of the vectors E and H are not in-phase.
The phase difference of these two vectors, according to Eq. (B.33), increases with the
increase of conductivity, σ. At σ = 0 there is no phase difference, i.e. ϕ = 0. At
σ/ǫ0 ǫω ≫ 1 the phase difference is close to ϕ = π/4, i.e. it becomes equal to the phase
shift in the standing wave.
 97

Table B.6.1. The skin depth, δ, of sea water and copper in [m].

Frequency, Hz 10 104 107 1010

Sea water 71.4 2.26 7.1 ×10−2 2.3 × 10−3
Copper 2.1 × 10−2 6.6 × 10−4 2.1 × 10−5 6.6 × 10−7

Problem B.6.5. The conductivity of sea water is σs.w. ≈ 5 Ω−1 m−1 and of copper
is σCu ≈ 6 × 107 Ω−1 m−1 . Compare the values of skin depth for these two media at
frequencies f = 10, 104 , 107 , and 1010 Hz, considering for simplicity that the dielectric
constant, ǫ, and magnetic permeability, µ, equal unity for both media. (Answer: Table
B.6.1.)
Solution. The penetration depth of the electromagnetic field into the medium is
defined by the general expression (Eq. (B.113) from the textbook) which taking into
account Eq. (B.107) from the textbook, transforms to:
" s !#−1/2
c ǫµ σ2
δ= 1+ 2 2 2 −1 . (B.36)
ω 2 ǫ0 ǫ ω

Let us estimate the magnitude of the parameter γ = σ/(ǫǫ0 ω) for the both media. For
copper we have γ ≫ 1 in the considered entire frequency range. Therefore, Eq. (B.36),
taking into account that ω = 2πf , transforms to:
s s
2ǫ0 2
δ=c = , (B.37)
µσω µ0 µσω

which is accurate for a wide range of frequency. For sea water the parameter γ ≫ 1 for
all given frequencies, except for f = 1010 Hz, for which γ ≈ 10. Therefore, for sea water
Eq. (B.37) is valid in the entire frequency range, except f > 1010 Hz. For calculation of
the skin depth for these values of frequency it is necessary to use the general expression
(B.36). The magnitudes of the skin depth, δ, in meters for the considered media at
given frequencies are given in the Table B.6.1.

Problem B.6.6. Let us consider an optical micro-resonator that consists of two

parallel-plate mirrors applied from the opposite sides of a narrow layer of optical glass
with width L = 10 µm. Find the number, N , of possible types of electromagnetic field
oscillations in the region between the mirrors (i.e. standing waves with different wave-
lengths) that can be excited in the visible range of wavelengths. Take the dielectric
constant of glass as ǫ = 1.55 and magnetic permeability as µ = 1. (Answer: N = 27.)
98 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

Solution. The standing waves are formed

√ in the gap between the mirrors in such a
way that the optical length, Lop = L ǫ, contains only integer number of half-waves,
mλ0 /2, where λ0 is the wavelength of light in vacuum. This can be formulated also as:
the width of the gap, L, must contain integer number of half-waves, mλ/2, i.e.
λ
L=m , (B.38)
2
√
where λ = λ0 / ǫ is the wavelength in the medium. Therefore, the number m can have
all integer values in the interval
2L √ 2L √
ǫ≤m≤ ǫ, (B.39)
λ02 λ01
where λ01 and λ02 are the boundaries of the interval of wavelengths. For the visible
range we can consider λ01 = 0.35µm and λ02 = 0.65µm. After simple calculations
we find that the integer number m can have values in the interval from m1 = 28 to
m2 = 55 - the smallest and the largest values of integer numbers defined by the interval
(B.39). The sought number of the standing waves, N , is defined by the difference of
these two numbers:

N = m2 − m1 = 27 . (B.40)

Problem B.6.7. The density of charge distribution in the electron cloud of a hydrogen
atom in the ground state is defined by the following expression:
e
ρ(r) = − 3 e−2r/a , (B.41)
πa
where a is the radius of first Bohr’s orbit and e is the electron charge. Find the energy
of the electrostatic Coulomb interaction of the electron cloud with the nucleus of the
hydrogen atom (i.e. with proton).

Solution. Let us choose as the origin of coordinate system the center of atom. Then,
the problem becomes spherically symmetric. Let us consider a spherical shell with the
radius r and thickness dr. The charge of electron cloud that corresponds to this layer
is equal to:
dqe = ρdV = ρ4πr2 dr . (B.42)
The interaction energy of proton with the charge of spherical layer is equal to:
4πr2 drρe 4e2 r − 2r
dW = ke = −ke 3 e a dr . (B.43)
r a
To find the total energy, we have to integrate the last expression over the entire volume
occupied by the electron cloud, i.e.
4e2 Z ∞ − 2r
W = −ke 3 re a dr . (B.44)
a 0
 99

The integration by parts gives the following expression for the interaction energy:

e2 e2
W = −ke =− . (B.45)
a 4πǫ0 a
Negative sign indicates that the interaction has the form of attraction.

Problem B.6.8. Find the screening radius (Debye radius), rD , and potential, ϕ, of a
point defect with charge q in an n-type semiconductor at room temperature, T = 300
K. Assume that positively charged ions of univalent impurities are located at the points
of the lattice and that the free electrons are in thermal equilibrium. The energy dis-
tribution of electrons is described by Boltzmann’s classical function. The density of
impurities is n0 = 1025 m−3 and the dielectric constant of the crystal is ǫ = 10. (An-
swer: rD ≈ 1.2 nm.)

Solution. Suppose that the positively charged defect of the crystalline structure is
located at the center of coordinates, and free electrons form around it an electron
cloud. The density of negative charge outside the defect is greater than the density
of the positive charge of ions, which we consider motionless and distributed according
Boltzmann’s distribution law:

ρe (r) = −ene (r) = −en0 e−eϕ(r)/(kB T ) , (B.46)

where n0 is the number of positive ions in the volume unit. Then, the density of excess
charge is:  
∆ρ(r) = −e[ne (r) − n0 ] = −en0 e−We (r)/(kB T ) − 1 , (B.47)
where We (r) = −eϕ(r) is the potential energy of a single electron, and ϕ(r) is the
electrostatic potential at the distance r from the impurity. For high temperatures the
electrostatic energy of a single electron is small in comparison with its thermal energy.
Therefore, the exponent in the Bolzmann’s distribution law can be expanded in series:

e2 n0 ϕ(r)
!
We (r) We (r)
∆ρ(r) ≈ −en0 1− − 1 = en0 =− . (B.48)
kB T kB T kB T

Let us substitute the last expression into the Poisson’s equation:

∆ρ e 2 n0
∇2 ϕ = − = ϕ. (B.49)
ǫ0 ǫ ǫ0 ǫkB T
The left-hand side of the derived equation taking into account the form of Laplace
operator in spherical coordinates in the case of spherical symmetry takes the form:
!
1 d dϕ
2
r2 = κ2 ϕ , (B.50)
r dr dr
100 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

where s
e 2 n0
κ= . (B.51)
ǫ0 kB T
As a result of integration of this equation we obtain:
C −κr
ϕ(r) = e . (B.52)
r
Near the point defect (r → 0) the potential must not differ from Coulomb potential
and therefore,
q
C= . (B.53)
4πǫ0 ǫ
Thus,
q
ϕ(r) = e−κr . (B.54)
4πǫ0 ǫr
This potential is called screened potential and the magnitude
s
1 ǫ0 ǫkB T
rD = = (B.55)
κ e 2 n0
is called the radius of screening (Debye radius). For concentration n0 = 1025 m−3 and
dielectric permitivity ǫ = 10 at room temperature T = 300 K the Debye radius is equal
to:
rD ≈ 1.2 nm . (B.56)

Problem B.6.9. A plane electromagnetic wave from a medium with index of refrac-
tion n1 is obliquely incident on an interface with a medium with index of refraction n2 .
Find the condition of total internal reflection, the wave field in the second medium,
and the depth of penetration into this medium.

Solution. The angle of incidence, θ1 , is related with the angle of refraction, θ2 , by the
Snell’s law (see Fig. B.4):
sin θ1 n2 n1
= or sin θ2 = sin θ1 . (B.57)
sin θ2 n1 n2
At n2 < n1 and at angle θ1 ≥ θcr where angle θcr is defined by the expression
n1 n2
sin θcr = 1 or θcr = arcsin , (B.58)
n2 n1
Snell’s law can be satisfied only at complex angles θ2 . It means that the wave entering
the second medium cannot propagate deep inside of it and is forced out to the interface
between two media. This phenomenon is called the total internal reflection since there
is no refracted wave.
 101

Figure B.4: Electromagnetic plane wave E1 is incident from the medium with the index
of refraction n1 on the medium with index of refraction n2 .

Let us find wave field inside of the second medium at such angles of incidence. In the
general form it can be written as:

E2 = E20 ei(ωt−k2 r) , (B.59)

where ω is the frequency of the wave, k2 = k0 n2 is the wavenumber in the second

medium, k0 = ω/c is the wavenumber in vacuum, and c is the speed of light in vacuum.
Let us choose such a system of coordinates that z-axis is perpendicular to the interface,
and the wavevector k2 lies in the xz-plane. Then, the field of the wave in the second
medium can be written as follows:

E2 = E20 eiω[t−(x sin θ2 +z cos θ2 )/v2 ] , (B.60)

where v2 = c/n2 is the phase velocity of the wave in the medium. We can write the
expression for θ2 knowing θ1 :
v
u 2
q un
cos θ2 = ± 1 − sin θ2 = ±it 12 sin2 θ1 − 1 .
2
(B.61)
n2

In the general expression for the field E2 in the last expression we have to choose ”mi-
nus” sign in order the wave in the second medium to damp. As a result of substitution
of Eq. (B.61) into the formula for E2 (Eq. (B.60)) we obtain:

E2 = E20 eiω(t−x sin θ1 /v1 )−κz , (B.62)

102 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

Figure B.5: Propagation of electromagnetic wave with wavelength equal λ; E, H, and

k compose a right-handed system.

where we introduced the damping coefficient, κ:

v
u 2
un q
κ = k0 n2 t 12 sin2 θ1 − 1 = k0 n21 sin2 θ1 − n22 . (B.63)
n2

From the obtained expressions it follows that at angle of incidence, θ1 , greater than
the critical value, θcr , the field inside the second medium is a wave traveling along
the interface and damping inside the second medium. The penetration depth into the
second medium is:
1 1 c 1
d= = q = q . (B.64)
κ k0 n21 sin2 θ1 − n22 ω n2 sin2 θ1 − n2
1 2

Problem B.6.10. Find the energy, W , which is transmitted in vacuum by a plane

electromagnetic wave propagating in the x-direction through a plane contour with area
S = 1 m2 perpendicular to this direction during one second. Assume that the ampli-
tude of the electric field intensity is equal to E0 = 10−4 V m−1 and that the frequency
is equal to f = 106 Hz. (Answer: W ≈ 1.3 × 10−15 J.)

Solution. The energy that wave transfers during time unit across the unit cross-section
is defined by the Poynting vector:

S = [E × H] , (B.65)

where electric field is oriented along y-axis and magnetic field along z-axis (see Fig. B.5).
Electric and magnetic field intensities are defined by the following expressions:

Ey = E0 sin(ωt − kx) , Hz = H0 sin(ωt − kx) . (B.66)

 103

Here, the amplitude of magnetic field is equal to:

s
ǫ0
H0 = E0 . (B.67)
µ0

We assume that the plane surface intersects with the x-axis at x = 0. In this case the
magnitude of Poynting vector is:
s
ǫ0 2 2
S(t, 0) = E sin (2πf t) . (B.68)
µ0 0

The expression for the transfered energy is:

τ σE02 ǫ0
s s " #
ǫ0 Z τ 2 sin(4πf τ )
W = σE02 sin (2πf t)dt = 1− . (B.69)
µ0 0 2 µ0 4πf τ

Since the second term in the brackets in our case is much less than unity, then the final
answer can be written as:
τ σE02
s
ǫ0
W = ≈ 1.3 × 10−15 J . (B.70)
2 µ0

Problem B.6.11. A linearly polarized wave is normally incident on a slab of thickness

L. The dielectric constant of the slab is ǫ, its conductivity is σ = 0, and its permeabil-
ity is µ = 1. Find the amplitudes of the waves reflected from and transmitted through
the slab and of the wave inside of the slab. Find the energy coefficients of reflection
and transmission. Find the angle of minimal wave reflection from the slab.

Solution. Let us assume that the slab is perpendicular to z-axis . Let us write the
equations to find the coordinate dependence of electric field inside and outside the slab
while considering time dependence proportional to exp(iωt):

d2 E1,3
2
+ k02 E1,3 = 0 , (B.71)
dz
d2 E2
+ k 2 E2 = 0 . (B.72)
dz 2
Here, the wavenumber outside the slab is equal to k0 = ω/c, and inside the√ slab is
equal to k = k0 n, where n is refraction coefficient of the slab’s material: n = ǫ. The
solutions of Eqs. (B.71) and (B.72) can be written as follows:

E1 = E0 eik0 z + Ae−ik0 z , (B.73)

E2 = Beikz + Ce−ikz , (B.74)

104 APPENDIX B. ELECTROMAGNETIC FIELDS AND WAVES

E3 = Deik0 z . (B.75)
Taking into account the condition of continuity of the tangential components of the
electric and magnetic fields at the interfaces of media at z = 0 and z = l, we obtain
the following sytem of equations for the unknowns A, B, C, and D:
E0 + A = B + C ,
E0 − A = nB − nC ,
(B.76)
BeikL + Ce−ikL = Deik0 L ,
nBeikL − nCe−ikL = Deik0 L .
Solving this system of equations for A, B, C, and D we obtain:
√
R0 (1 − e2ikL )
A= E0 , (B.77)
1 − R0 e2ikL
2
B= E0 , (B.78)
(1 + n)(1 − R0 e2ikL )
2R0
C= E0 , (B.79)
(1 − n)(R0 − e−2ikL )
T0 eikL
D= E0 , (B.80)
1 − R0 e2ikL
where we introduced the reflection and the transmission coefficients for a semi-infinite
medium:
1−n 2
 
R0 = (B.81)
1+n
and
4n
T0 = . (B.82)
(1 + n)2
By taking into account the expressions for R and T we get:
2
A 4R0 sin2 (kL)
R= = (B.83)
E0 1 − 2R0 cos(2kL) + R02
and
D 2 T02
T = = . (B.84)
E0 1 − 2R0 cos(2kL) + R02
From the obtained expressions it follows that the reflection coefficient becomes equal
to zero when
kL = πm , (B.85)
where m is integer. From the last equation it follows that the width of the slab, L,
must be equal to
λ λ0
L=m =m , (B.86)
2 2n
where λ is the wavelength in the slab.
Appendix C

Crystals as atomic lattices

Problem C.8.1. Copper has a face-centered cubic lattice. The density of copper is
ρ = 8.96×103 kg/m3 , molar mass of copper is µ = 6.4×10−2 kg mol−1 . Find the lattice
constant, a, the atomic radius, ra , and the number of atoms, n, per unit volume for a
copper crystal. (Answer: a ≈ 3.61 × 10−10 m, ra ≈ 1.28 × 10−10 m, and n ≈ 8.49 × 1028
m−3 .)

Solution. The number of atoms that corresponds to the face-centered unit cell is equal
to Z = 4. Therefore, the volume that corresponds to one atom is equal to:

a3
V0 = , (C.1)
Z
where a is the lattice constant. Then, the volume of one mole, Vµ , is:

a3
V µ = NA V 0 = NA , (C.2)
Z
where NA is the Avogadro constant: NA = 6.02 × 1023 mol−1 . The density of a crystal
can be written through the molar mass:
µ Zµ
ρ= = . (C.3)
Vµ N A a3

From the last expression we find the lattice constant, a:

!1/3
Zµ
a= ≈ 3.61 × 10−10 m . (C.4)
NA ρ

The volume of the unit cell is equal to a3 . Therefore, concentration of atoms can be
written as:

Z
n= 3
≈ 8.49 × 1028 m−3 . (C.5)
a
105
106 APPENDIX C. CRYSTALS AS ATOMIC LATTICES

The atomic radius, ra , we find from the fact that there is only one atom in the center
of the diagonal of the unit cell’s face and there are two atoms in the vertices of the
unit cell. Thus, this diagonal is equal to four atomic radii, i.e.
√
2a
ra = = 1.28 × 10−10 m . (C.6)
4

Problem C.8.2. The energy per atom of a one-dimensional molecular crystal is given
by the expression: "   6 #
a 12 a
u(x) = A −2 , (C.7)
x x
where x is the distance between neighboring atoms; A and a are constants. Find the
equilibrium interatomic distance, the binding energy per atom, and the coefficient of
thermal expansion of such a crystal.

Solution. The energy of a molecular crystal consisting of N atoms can be written

as U (x) = N u(x). In equilibrium the potential energy of a crystal is minimal, and,
therefore, !
dU
=0 (C.8)
dx x=x0
and
a 6
 
− 1 = 0. (C.9)
x0
From the last expression we obtain that x0 = a. The binding energy of a crystal is:

W = −U (x0 ) = AN . (C.10)

Thus, the binding energy that corresponds to one atom is:

W
w= = A. (C.11)
N
The coefficient of thermal expansion is defined as:
!
1 ∂V
α= , (C.12)
V0 ∂T V =V0

where V0 is equilibrium volume. By taking into account that the crystal is one-
dimensional, we find
V =N <x> . (C.13)
Therefore, !
1 ∂<x>
α= . (C.14)
x0 ∂T x=x0
 107

The magnitude < x > can be found with the help of Boltzmann’s distribution law, i.e.:
 
x exp − UkB(x)
R∞
−∞ T
dx
< x >= R ∞   . (C.15)
−∞ exp − UkB(x)
T
dx

To carry out the integration in Eq. (C.15) we expand the function U (x) near the atom’s
equilibrium position taking into account the first anharmonic term:
d2 U d3 U
! ! !
dU 1 1 2
U (x) ≈ U (x0 )+ (x−x0 )+ (x−x0 ) + (x−x0 )3 .
dx x=x0
2! dx2 x=x0
3! dx3 x=x0
(C.16)
Using the expression for the function U (x) we find the corresponding derivatives. Thus,
we have:
U (x) ≈ −A + b(x − x0 )2 − c(x − x0 )3 , (C.17)
where we introduced the coefficients
36A
b= (C.18)
x20
and
252A
c= . (C.19)
x30
Let us approximate the exponential under the integrals in Eq. (C.15):
b(x − x0 )2 c(x − x0 )3
! !" #
U (x) A
 
exp − ≈ exp exp − 1+ . (C.20)
kB T kB T kB T kB T
This approximation corresponds to the exact accounting of harmonic input b(x − x0 )2
and approximate accounting of anharmonic input c(x−x0 )3 . Let us change the variables
in the integrals, x − x0 = y, and introduce a new variable β = 1/(kB T ):
R∞ 2 3
−∞ (x0 + y) exp(−βby )(1 + βcy )dy
< x >= R∞
2 3
. (C.21)
−∞ exp(−βby )(1 + βcy )dy

To carry out integration we use the following expressions for the integrals:
∞ π
Z r
I1 = exp(−σy 2 )dy = , (C.22)
−∞ σ
Z ∞
I2 = y n exp(−σy 2 )dy = 0 for odd n , (C.23)
−∞
∞ 3 π
Z r
4 2
I3 = y exp(−σy )dy = . (C.24)
−∞ 4 σ5
As a result of integration we obtain:
!
3c 3c 7x0
< x >= x0 1+ = x0 + kB T = x0 + kB T . (C.25)
4βb2 x0 4b 2 48A
108 APPENDIX C. CRYSTALS AS ATOMIC LATTICES

From the last expression it follows that

!
1 ∂<x> 3ckB 7kB
α= = 2
= . (C.26)
x0 ∂T x=x0
4x0 b 48A

Problem C.8.3. Find the relation between the volumes of unit cells of direct, Va , and
inverse, Vb , lattices of crystals.

Solution. Translation vector of the direct lattice is defined as

a n = n1 a 1 + n2 a 2 + n3 a 3 , (C.27)

where a1 , a2 , and a3 are the basis vectors of the direct lattice and n1 , n2 , and n3
are arbitrary integers. The volume of unit cell of the direct lattice is the volume of
parallepiped built using basis vectors:

Va = (a1 [a2 × a3 ]) . (C.28)

In the inverse (reciprocal) space (i.e. space of wavevectors) the so-called inverse lattice
corresponds to the direct lattice. The translation vector of the inverse (reciprocal)
lattice is defined by the expression

bg = g1 b1 + g2 b2 + g3 b3 , (C.29)

where g1 , g2 , and g3 are arbitrary integers and b1 , b2 , and b3 are the basis vectors of
the inverse lattice. These vectors are introduced as follows:
1
b1 = [a2 × a3 ] , (C.30)
Va
1
b2 = [a3 × a1 ] , (C.31)
Va
1
b3 = [a1 × a2 ] . (C.32)
Va
The volume of the unit cell of the periodical lattice of inverse space is equal to the
volume built on the basis vectors of this lattice:

Vb = (b1 [b2 × b3 ]) . (C.33)

Let us substitute in the last equation the expressions for bj and make series of trans-
formations. Let us first transform the vector product:
1 1
[b2 × b3 ] = 2
[[a3 × a1 ] × [a1 × a2 ]] = 2 {a1 ([a3 × a1 ]a2 ) − a2 ([a3 × a1 ]a1 )} =
Va Va
 109

a1 ([a3 × a1 ]a2 ) a1 (a2 [a3 × a1 ]) a1 (a1 [a2 × a3 ]) a1 Va a1

= = = = = . (C.34)
Va2 Va2 Va2 Va2 Va
By taking into account the last expression we get:
1 1 1
Vb = (b1 [b2 × b3 ]) = ([a 2 × a 3 ]a 1 ) = (a 1 [a 2 × a 3 ]) = . (C.35)
Va2 Va2 Va

Problem C.8.4. Crystals of salt, NaCl, have a cubic structure, their molar mass
is µ = 58.4 × 10−3 kg mol−1 and their density is ρ = 2.17 × 103 kg m−3 . Find the
lattice constant, a, and the distances between crystallographic planes d100 , d110 , and
d111 . (Answer: a ≈ 5.64 × 10−10 m, d100 ≈ 2.82 × 10−10 m, d110 ≈ 1.99 × 10−10 m, and
d111 ≈ 1.63 × 10−10 m.)

Solution. The unit cell of NaCl contains eight cubes of volume (a/2)3 whose vertices
contain ions Na+ and Cl− . Each cube contains Z = (1/8) × 8 = 1 ions or half of
molecule NaCl. The volumes which correspond to one molecule, V1 , and one mole, Vµ ,
are defined as:
2 a 3
 
V1 = (C.36)
Z 2
and
N A a3
V µ = NA V 1 = . (C.37)
4Z
Let us write the crystal density through the molar volume:

µ 4Zµ
ρ= = . (C.38)
Vµ N A a3

From the last expression we find the lattice constant, a:

!1/3
4Zµ
a= ≈ 5.64 × 10−10 m . (C.39)
NA ρ

In the cubic lattice with the edge of unit cell equal to a0 the distance between adjacent
parallel crystallographic planes, dhkl , is defined by the expression:
a0
dhkl = √ , (C.40)
h2 + k 2 + l2
where h, k, and l are Miller indices of the corresponding set of planes.
In the case of NaCl crystal the length of the edge of the cube whose vertices contain
ions Na+ and Cl− is equal to a0 = a/2. This is equal to the inter-plane distance:
a
d100 = ≈ 2.82 × 10−10 m . (C.41)
2
110 APPENDIX C. CRYSTALS AS ATOMIC LATTICES

For other two inter-plane distances we have:

a
d110 = √ ≈ 1.99 × 10−10 m (C.42)
2 2
and
a
d111 = √ ≈ 1.63 × 10−10 m . (C.43)
2 3

Problem C.8.5. A unit volume of a metal with a simple cubic lattice contains n
atoms. Each atom has Z valence electrons. In the free-electron approximation, find
the radius of the Fermi sphere in k-space and in p-space, and the Fermi energy at
T = 0 K.

Solution. Let us choose a crystal in the form of a cube whose edge is equal to L. In
the approximation of free electrons, their eigenvalues are defined by the expression:

h̄2 2
E= (kx + ky2 + kz2 ) , (C.44)
2me
where
2π
kx = nx , (C.45)
L
2π
ky = ny , (C.46)
L
and
2π
kz = nz . (C.47)
L
Here nα are integers. The allowed energy states in k-space in this case can be repre-
sented by the set of cells with the volume
3
2π

∆Vk = ∆kx ∆ky ∆kz = . (C.48)
L
The number of such cells in the unit volume of k-space is equal to:

1 L3
= . (C.49)
(2π/L)3 8π 3

In each of these cells there are two electrons with oppositely-oriented spins. Because of
the isotropy of the problem, the isoenergetic surfaces in k-space are spheres. Spherical
Fermi surface of radius kF at T = 0 contains the number of cells that includes N Z
electrons. Here N is the number of atoms in a crystal. Therefore,
3
4π 3 L

NZ = 2 k . (C.50)
3 F 2π
 111

From the last equation we find:

!1/3
3π 2  1/3
kF = NZ = 3π 2 n . (C.51)
L3

Here we took into account that the electron concentration, n, is defined as:

NZ
n= . (C.52)
L3
For the Fermi momentum we obtain the following expression:
 1/3
pF = h̄kF = h̄ 3π 2 n . (C.53)

Then, the Fermi energy can be defined as:

p2F h̄2  2 2/3

EF = = 3π n . (C.54)
2me 2me

Problem C.8.6. Work function for electrons in lithium, Li, is equal to Awf = 2.36
eV, the density is ρ = 5.34 × 102 kg m−3 , and the molar mass is µ = 6.94 × 10−3 kg
mol−1 . Find the depth of the potential well U0 , where conduction electrons of metals
are confined. (Answer: U0 ≈ 7.06 eV.)

Solution. The electron work function in lithium is equal to:

Awf = U0 − EF (C.55)

if we consider the bottom of the potential well as the bottom of the conduction band
(here we talk about electrons that are very close to the Fermi level, EF ). Therefore,

U0 = Awf + EF . (C.56)

The Fermi energy through the concentration of conduction electrons can be written as:

h̄2  2 2/3
EF ≈ EF (0) = 3π n , (C.57)
2me

where EF (0) is the Fermi energy at T = 0 K. Lithium is a univalent material (Z = 1),

therefore electron concentration coincides with the concentration of atoms. Thus, the
concentration of free electrons can be written as:
NA ρ
n= . (C.58)
µ
112 APPENDIX C. CRYSTALS AS ATOMIC LATTICES

Therefore, EF (0) ≈ 4.7 eV and U0 is equal to:

!2/3
h̄2 3π 2 NA ρ
U0 = Awf + ≈ 7.06 eV . (C.59)
2me µ

Problem C.8.7. The dispersion relation for electrons in a crystal with a simple cubic
lattice has the form:

En (k) = En (0) + 2En (a)[cos(kx a) + cos(ky a) + cos(kz a)] , (C.60)

where a is the lattice constant and En (0) is the electron energy in the nth state of an
isolated atom. Find the electron effective masses and group velocities in the center and
close to the apex of the Brillouin zone.

Solution. Near the Brillouin zone center kα a ≪ 1. Therefore,

1
cos(kα a) ≈ 1 − (kα a)2 . (C.61)
2
Here α = x, y, and z. By taking into account the expansion of cosine function (C.61),
the dispersion relation, Eq. (C.60), transforms to:

En (k) = En (0)+6En (a)−En (a)(kx2 +ky2 +kz2 )a2 = En (0)+6En (a)−En (a)k 2 a2 . (C.62)

Using this expression we find the effective mass in the Brillouin zone center (near the
bottom of energy band):
!−1
∗ 2 ∂ 2E h̄2
m (0) = h̄ =− . (C.63)
∂k 2 2a2 En (a)

The sign of the effective mass depends on the sign of the overlap integral En (a). For
the s-band (En (a) = Es (a)) Es (a) < 0 and m∗ (0) > 0.
Near the top of the Brillouin zone (the top of the energy band)
′ ′
kα a = ±(π − kα a) , kα a ≪ 1 . (C.64)

Near these points

1  ′ 2
cos(kα a) ≈ −1 + k a (C.65)
2 α
and
′
En (k) = En (0) − 6En (a) + En (a)k 2 a2 . (C.66)
From the last equation it follows that

π h̄2
 
∗
m = . (C.67)
a 2a2 En (a)
 113

Since for the s-band Es (a) < 0, then m∗ ( πa ) < 0. Therefore,

π
 
∗
m = −m∗ (0) . (C.68)
a
The projections of the group velocity
1 ∂En 2a
vα = = − En (a) sin(kα a) (C.69)
h̄ ∂kα h̄
are equal to zero in the center of the Brillouin zone (kα a = 0) as well as at the
boundaries of the Brillouin zone (kα a = ±π).

Problem C.8.8. The As atom with five valence electrons is a donor impurity in Ge,
which has four valence electrons. Estimate the ionization energy, Ei , and the radius of
the first electron orbit, r1 , of valence electrons in an impurity atom. For germanium,
Ge, the dielectric constant ǫ = 16 and effective mass of the electron is m∗ = me /2.
(Answer: Ei ≈ 0.026 eV and r1 ≈ 17 × 10−10 m.)

Solution. According to Bohr’s theory of hydrogen atom the ionization energy, EiH ,
and the radius of the electron’s first orbit, r1H , are defined as follows:
!2
me e2
EiH = = 13.6 eV (C.70)
2 4πǫ0 h̄

and
4πǫ0 h̄2
r1H = = 0.53 × 10−10 m . (C.71)
me e 2
When the Bohr’s model of hydrogen atom is used for the description of an impurity
atom in a crystal, the free electron mass, me , must be replaced by the effective mass,
m∗ , and the dielectric constant, ǫ, has to be taken into account. As a result Eqs. (C.70)
and (C.71) must be written as:
!2
m∗ e2 m∗ H
Ei = = E (C.72)
2 4πǫ0 ǫh̄ m e ǫ2 i

and
4πǫ0 ǫh̄2 me ǫ
r1 =∗ 2
= ∗ riH . (C.73)
me m
By substituting the parameters for Ge in the last two equations we find:

Ei = 0.026 eV (C.74)

and
r1 = 17 × 10−10 m . (C.75)
114 APPENDIX C. CRYSTALS AS ATOMIC LATTICES

Problem C.8.9. Find the electron concentration in the conduction band and the
hole concentration in the valence band in an intrinsic nondegenerate semiconductor at
temperature T .

Solution. The electron concentration in the conduction band is defined by the expres-
sion: Z ∞
n=2 g(E)fn (E, T )dE , (C.76)
0

where the density of states, g(E), is

√ (2me )3/2
g(E) = B E , B= , (C.77)
4π 2 h̄3
and the electron energy distribution function, fn (E, T ), is given by Fermi-Dirac func-
tion:
1
fn (E, T ) = (E−E )/(k T ) . (C.78)
e F B +1
Here, the reference point for energy E is the bottom of the conduction band. After sub-
stitution of all of these expressions into Eq. (C.76) the expression for the concentration,
n, can be written as follows:
√
Z ∞
EdE
n = 2B . (C.79)
0 e(E−EF )/(kB T ) + 1
The concentration of electrons in the conduction band (and holes - in the valence band)
is significantly smaller than in metals because of the existence of forbidden energy gap,
Eg , in semiconductor. Therefore, we can consider that fn ≪ 1. From this inequality
follows that
E − EF
≫ 1. (C.80)
kB T
In this case, the Fermi-Dirac distribution function approximately coincides with the
classical Maxwell-Boltzmann distribution function
E−EF EF
− − k ET
fn (E, T ) ≈ e kB T
= Ae B , where A = e kB T . (C.81)

Taking into account the last expression we obtain for the electron concentration, n, the
following equation: Z ∞√
EF
− E
n = 2Be kB T
Ee kB T dE . (C.82)
0

Let us calculate this frequently used in the theory of semiconductors integral:

Z ∞√ √
− k ET π
Ee B dE = (kB T )3/2 . (C.83)
0 2
 115

Taking into account the last expression, Eq. (C.82) for the electron concentration, n,
transforms to: !3/2
me kB T EF
n=2 2 e kB T . (C.84)
2πh̄
In a similar fashion, the expression for the hole concentration in the valence band can
be obtained: !3/2
mh kB T E +Eg
− F
p=2 e kB T
. (C.85)
2πh̄2

Problem C.8.10. Find the position of the Fermi level in an intrinsic non-degenerate
semiconductor, whose bandgap width, Eg , depends linearly on temperature, i.e. Eg =
Eg0 − γT , where Eg0 is the bandgap at T = 0 K.

Solution. For intrinsic semiconductors the electric neutrality condition must be sat-
isfied. Therefore, the concentration of electrons, n, in the conduction band is equal
to the concentration of holes, p, in the valence band. Let us write the corresponding
concentrations taking the top of the valence band as a reference point:
!3/2
me kB T EF −Ec
n=2 e kB T
(C.86)
2πh̄2

and !3/2
mh kB T Ev −EF
p=2 e kB T
, (C.87)
2πh̄2
where Ec is the bottom of the conduction band, Ev is the top of the valence band,
and me and mh are the effective masses of electrons and holes, respectively. From the
equality n = p we obtain:
mh 3/2
2EF −Ec −Ev
 
e kB T = . (C.88)
me
From the last equation we find:
Ev + Ec 3 mh
EF = + kB T ln . (C.89)
2 4 me
For the chosen reference point Ev = 0 and Ec = Eg . Therefore,

Eg 3 mh E 0 γT 3 mh
EF = + kB T ln = g − + kB T ln . (C.90)
2 4 me 2 2 4 me