An Introduction to Quantum Mechanics:

Exercises and Solutions

SERP-CHEM: 2011-15

Camillo Imbimbo

Dipartimento di Fisica dell’Università di Genova

Via Dodecaneso, I-16136, Genova, Italia
Contents
1 The Failure of Classical Physics 2
1.1 Atom stability . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Equipartion theorem . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 The Photoelectric Effect . . . . . . . . . . . . . . . . . . . . . 4
1.4 Bohr’s quantum theory of atomic spectra. . . . . . . . . . . . 5
1.5 Quantum theory of gas specific heat capacities . . . . . . . . . 10
1.5.1 Einstein and Debye theories of specific heat capacities
of solids . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2 The Principles of Quantum Mechanics 12

3 Simple systems 19

4 Time evolution 37

5 Symmetries 43

1 The Failure of Classical Physics

1.1 Atom stability
Exercise 1. Estimate the speed of the electron rotating around the nucleus
for the hydrogen atom, assuming that it orbitates alonga circular orbit of
radius r0 = 10−8 cm. Compare it to the speed of light.
Solution.
e2 1 1 1 m2
v2 = ≈ (1.6)2 × 10−38 9.0 × 109 ≈
4 π 0 m r0 9.1 × 10−31 1010 sec2
m2
≈ (1.6)2 0.99 × 1012 (1.1)
sec2
Hence
v 1.6 × 106
≈ = 0.53 × 10−2 (1.2)
c 3.0 × 108

2
Exercise 2. Check that formula (1.7) of the lecture notes has the correct
dimensions.
Solution.
2
Recall that 4 πe0 r2 has the dimension of a force. Therefore
h e2 a2 i h M × L3 L2 T 3 i h M × L2 i h E i
= = = (1.3)
4 π 0 c3 T2 T 4 L3 T3 T
Exercise 3. Estimate the time that it takes an electron of the hydrogen
atom to fall onto the nucleus, assuming it starts from a circular orbit of radius
r0 = 10−8 cm.
Solution.

(4 π 0 )2 c3 m2 3  4 π 0 m r0 c2 2 r0 c4 r0
τ= r0 = = 4 ≈
4 e4 e2 4c v 4c
1 10−10
≈ × sec ≈ 1.1 × 10−10 sec (1.4)
0.534 × 10−8 4 × 3.0 × 108

1.2 Equipartion theorem

Exercise 3. Prove that a mass m which is constrained to move freely on a
circle on a plane by a rigid rod of radius R is described by an Hamiltonian of
the form
p2θ
H(p, θ) = (1.5)
2I
where pθ is the momentum conjugate to angle θ which measures the position
of the particle on the circle, and I is called the momentum of inertia. Express
I in terms of m and R.
Solution.
The velocity of the particle is
dθ
v=R (1.6)
dt
Therefore its energy is
1 1
H= m v 2 = m R2 θ̇2 (1.7)
2 2

3
The momentum conjugate to θ is

pθ = m R2 θ̇ (1.8)

Thus
1 p2θ
H= (1.9)
2 m R2
Thus

I = m R2 (1.10)

Exercise 4. Prove that a diatomic molecule schematized as two identical

masses m2 held at a fixed distance R by a rigid rod of negligible mass is
described by the Hamiltonian

p~2 1  2 p2φ 
H(p, q) = + pθ + (1.11)
2m 2I sin2 θ
Compute the momentum of inertia I in terms of m and R.
Problem 2. Generalize the derivation of the equipartition theorem to the
Hamiltonian of the bi-atomic molecule (1.11).

1.3 The Photoelectric Effect

Problem 5. A lamp of 100 W emits light whose wave length is 5.890 10−7 m.
How many photons are emitted in 1 second?
Solution The energy of photons of such frequency is
12400
Eν = eV = 2.10 eV = 3.37 × 10−19 J (1.12)
5890
dN
Therefore the number dt
of photons emitted per unit of time is

dN 100W
= = 2.97 × 1020 sec−1 (1.13)
dt 3.37 10−19 J

4
1.4 Bohr’s quantum theory of atomic spectra.
Problem 6. Classical electromagnetism predicts that a charge rotating on a
circular orbit with angular orbital frequency ω emits electromagnetic radiation
with the same frequency. Show that the frequency of the photon emitted by
H
an electron of the Bohr hydrogen atom which goes from the level En+1 to the
H
level En , is, for n large, approximately equal to the frequency of the classical
circular motion corresponding to that energy.
H
Solution: The difference of energy of the levels En+1 and EnH for n >> 1 is

H e4 m 1 e4 m 1
En+1 − EnH = − 2 2 2
+ =
(4 π 0 ) ~ 2 (n + 1) (4 π 0 ) ~ 2 n2
2 2

e4 m 1
≈ (1.14)
(4 π 0 )2 ~2 n3
The angular frequency of the corresponding emitted photon is
e4 m 1
ωn+1→n = (1.15)
(4 π 0 )2 ~3 n3
On the other hand, the angular frequency for a classical circular orbit of
radius r is
v
ω= (1.16)
r
Since
e2 1
v2 = (1.17)
4 π 0 m r
we have
4 π 0
ω= m v3 (1.18)
e2
The velocity vn of the electron in the circular orbit of energy En is given by
1 e4 m 1
En = − m vn2 = − (1.19)
2 (4 π 0 ) ~ 2 n2
2 2

that is
1 e2
vn = (1.20)
n (4 π 0 ) ~

5
The angular frequency of the circular orbit with this energy is therefore
4 π 0 1 e4 m
ω = vn3 m = (1.21)
e2 n3 (4 π 0 )2 ~3
which matches the frequency of the photon emitted in the transition n+1 → n
as predicted by Bohr theory.
Problem 9. Apply the Bohr-Sommerfeld quantization condition to an
unidimensional free particle moving in a potential well V (x) defined by
(
−V0 for 0 < x < L
V (x) = (1.22)
0 for x < 0 and x > L

where V0 > 0 is the depth of the well. Determine the discrete energy levels.
How many they are? What is the minimum value of V0 for which there exist
discrete levels?
Solution: The bound classical trajectories are those with E < 0. The
momentum for these orbits is quantized according to
nh
pn = (1.23)
2L
The energy is therefore
p2n n2 h2
En = − V0 = − V0 < 0 (1.24)
2m 8 m L2
Hence there are a finite number of closed trajectories, with n = 1, 2, . . . nmax
where nmax is the highest positive integer such that
8 m L2 V0
n2max < (1.25)
h2
The minimal value of V0 for which there are discrete levels is the minimal
value for which this equation admits an integer positive solution

8 m L2 V0min min h2
= 1 ⇔ V 0 = (1.26)
h2 8 m L2
Problem 10. Apply Bohr-Sommerfeld quantization condition to a satellite
of mass m = 1 kg rotating around the earth. Determine the allowed radiuses
in terms of the mass of the earth M and the universal constant of gravitation

6
G. Suppose the satellite is on on circular orbit of radius near the radius of
the earth R = 6400 km, with a given Bohr integer number n̄. How much does
the radius change if the satellite shifts to an orbit with n = n̄ + 1? Express
the answer in meters.
Solution: The classical equation of motion for a circular orbit of radius r is

GM v2 GM
2
= ⇔ v2 = (1.27)
r r r
The momentum is therefore
r
GM
p = mv = m (1.28)
r
and the Bohr-Sommerfeld quantization condition is
I √
p dr = 2 π r p = 2 π m G M r = n h (1.29)

that is
n2 ~2
rn = (1.30)
G M m2
If n changes from n → n + ∆n = n + 1 the radius changes

2 n ∆n ~2 2 ∆n rn 2 rn
∆rn = rn+1 − rn ≈ 2
= = (1.31)
GM m n n
We have
n2 ~ 2 2 G M R m2 g m2 R 3
rn = ≈ R ⇔ n ≈ = (1.32)
G M m2 ~2 ~2
GM m
where g = R2
≈ 9.8 sec2 is the gravity acceleration. Thus

9.8 × 12 × 6.43 × 1018

n2 ≈ ≈ 24 × 1088 ⇒ n ≈ 5 × 1044 (1.33)
1.052 × 10−68
Then
2
∆rn ≈ × 6.4 × 106 m ≈ 2.6 × 10−38 m (1.34)
5 × 1044

7
Problem 12. Apply the Bohr-Sommerfeld quantization condition to compute
the energy levels and the orbits of a particle of mass m moving in circular
orbits of radius r in a potential V (r) = σ r.
Solution: The classical equation of motion is
m v2
=σ (1.35)
r
The quantization condition on the angular momentum for a circular orbit of
radius r is therefore
σr √
r
3
L = mrv = mr = m σ r2 = n ~ (1.36)
m
The orbits have therefore radiuses
2
(n ~) 3
rn = 1 (1.37)
(m σ) 3
and the corresponding energies are
2
3 3 (n ~ σ) 3
En = σ rn = (1.38)
2 2 m 13
Problem 13. Apply the Bohr-Sommerfeld quantization condition to derive
energy levels and allowed orbit radiuses for a particle of mass m moving in
circular orbits of radius r attracted toward the center of the orbit by an elastic
force F = −k r.
Solution: The classical equations of motion for circular orbits are
v2
m = kr (1.39)
r
Therefore
r
k 2 k
v2 = r ⇒v= r ≡ ωr (1.40)
m m
q
k
where ω = m is the angular frequency of the harmonic oscillations. The
angular momentum of the circular orbit is therefore

L = m v r = m ω r2 (1.41)

8
and the Bohr quantization condition Ln = n ~ gives for the radiuses of the
circular orbits
n~
rn2 = (1.42)
mω
The corresponding energy levels are therefore
1 1 1 1
En = m vn2 + m ω 2 rn2 = m ω 2 rn2 + m ω 2 rn2 =
2 2 2 2
2 2
= m ω rn = n ~ ω (1.43)

Problem 14. Derive the numerical values for the Rydberg constants pre-
dicted by Bohr atomic model for the following hydrogenoids: hydrogen H,
deuteron D (whose nucleus has one proton and one neutron), He+ , Li++ and,
for the same elements, the wave lengths of the spectral lines corresponding to
the transition between the first excited level and the fundamental level.
Solution:
Let us denote by R∞ the Rydberg constant for an infinitely massive
nucleus
e2
2 1 me
R∞ = ≈ 109737 cm−1 (1.44)
4 π 0 4 π ~3 c
The Rydberg constant R(M ) for a nucleus of mass M and charge Z is related
to the energy levels by the formula

R(M ) h c
En = −Z 2 (1.45)
n2
a
1 me

R(M ) = R∞ me ≈ R∞ 1 − (1.46)
1+ M M

Recall that
mp
= 1836.15 (1.47)
me

9
Therefore
me

RH = R(mp ) ≈ R∞ 1 − = 109677 cm−1
mp
me

RD = R(2 ∗ mp ) ≈ R∞ 1 − = 109707 cm−1
2 mp
me

RHe+ = R(4 ∗ mp ) ≈ R∞ 1 − = 109722 cm−1
4 mp
me

RLi++ = R(7 ∗ mp ) ≈ R∞ 1 − = 109728 cm−1 (1.48)
7 mp
The wave lengths of the first spectral line corresponding to the transition
from n = 1 to n = 2 are
4 1
λH = = 1215.69 Å
3 RH
4 1
λD = = 1215.36 Å
3 RD
4 1
λHe+ = = 303.798 Å
3 4 RHe+
4 1
λLi++ = = 135.013 Å (1.49)
3 9 RLi++

1.5 Quantum theory of gas specific heat capacities

1.5.1 Einstein and Debye theories of specific heat capacities of
solids
Problem 16. Suppose that the density of states of phonons grows quadrati-
cally with the frequency
dn(ω)
= A ω 2 + O(ω 3 ) for ω  ωmax (1.50)
dω
only for small frequencies ω. n(ω) is an otherwise generic function of ω for
bigger ω up to a certain given ωmax , which defines a Debye temperature TD .
Derive the predictions of the Debye model for the internal energy in the low
and high temperature limit.
Solution:
Let us start from
Z ωmax
~ ω e−β ~ ω
U= dω n0 (ω) (1.51)
0 1 − e−β ~ ω

10
 ~ ωmax
For T  TD ≡ k
,
~ β ω ≤ ~ β ωmax  1 (1.52)
Therefore the high temperature limit of the internal energy coincides with
the classical result, as expected on general grounds
Z ωmax
1 ωmax
Z
0 ~ω 3 NA
U≈ dω n (ω) = dω n0 (ω) = T  TD (1.53)
0 β ~ω β 0 β
On the other hand for low temperatures, when T  TD ,
1
 ωmax (1.54)
~β
the integrand in (1.51) is significantly different than zero only if
1
ω.  ωmax (1.55)
~β
Therefore, in the region of frequencies when the integrand is significantly
different than zero we can approximate the density of state with (1.50) and
Z ωmax Z ∞
~ ω 3 e−β ~ ω A~ x3 e−x
U ≈A dω ≈ dx =
0 1 − e−β ~ ω (~ β)4 0 1 − e−x
A π4 1
= 3 (1.56)
~ 15 β 4
In the lecture notes we derived the expression of A in terms of the sound
speed in the solid and the volume
3 L3
A= (1.57)
2 π2 v3
Therefore
L3 π 2 4 4
U≈ 3 3 k T T  TD (1.58)
v ~ 10
Problem 17. Discuss the effect of quantization on the specific heat capacities
of bi-atomic gases, knowing that: a) the rotational degrees of freedom around
an axis orthogonal to the axis connecting the two atoms give rise to discrete
energy levels with separation ∆E ∼ 10−4 − 10−2 eV ; b) vibrational degrees
of freedom have ∆E ∼ 10−1 eV ; c) rotational degrees of freedom around the
axis connecting the two atoms have ∆E ∼ 10 eV .
Solution. TO BE GIVEN

11
2 The Principles of Quantum Mechanics
Problem 10. A system is described by a 2-dimensional space of states. Let
|1i and |2i be the normalized eigenstates of an√observable O with eigenvalues
λ = 1 and λ = 2 respectively. Let v = |1i + i 3|2i be a state of the system.
a) What is the probability that the measure of O on v gives the result
λ = 2?
b) Write the 2 × 2 matrix which represents O in the basis {|1i, |2i}.
c) Let O0 be another observable, such that O0 |1i = |2i and O0 |2i = |1i.
Write the 2 × 2 matrix which represents O0 in the basis {|1i, |2i}.
d) What are the eigenvalues of O0 ?
e) What is the average of O0 on the state v?
Solution:

• a) Let us compute the norm of v

√
|v|2 = 1 + |i 3|2 = 1 + 3 = 4 (2.1)

Thus the normalized vector

1 √
v0 =


|1i + i 3|2i (2.2)
2
describes the same physical state as v. From (2.2) we read the proba-
bilities

√
1 2 1 i 3 2 3
Pλ=1 (v) = = Pλ=2 (v) = = (2.3)
2 4 2 4

• b) Since {|1i, |2i} are the eigenstates of O, in this basis O is represented

by a diagonal matrix
 
1 0
O= (2.4)
0 2

• c) The matrix representing O0 in the basis {|1i, |2i} is

 
0 0 1
O = (2.5)
1 0

12
 • d) The eigenvalues of O0 are obtained from the secular equation
 0 
−λ 1
det = (λ0 )2 − 1 = 0 ⇔ λ0 = 1, −1 (2.6)
1 −λ0

• e) The average of O0 on v is
1 √
hv 0 , O0 v 0 i = hv 0 , O0


|1i + i 3|2i i =
2
1 √
= hv 0 , |2i + i 3|1i i =


2
1 √ 1 √

= h |1i + i 3|2i, |2i + i 3|1i i =
2 2
1 √ √ 
= i 3−i 3 =0 (2.7)
4

Problem 11. On a spin 1/2 system the following 3 observables are defined
     
~ 1 0 ~ 0 1 ~ 0 −i
Sz = Sx = Sy = (2.8)
2 0 −1 2 1 0 2 i 0

a) Find the eigenvalues and the normalized eigenvectors of the 3 observables

{Sx , Sy , Sz }.
b) Suppose the system is in the state described by the vector
 
1
v= (2.9)
3i

Find the probabilities to obtain the various possible values of {Sx , Sy , Sz } on

v. (You should in other words compute 6 probabilities, 2 for each of the 3
observables.)
c) Compute the 2x2 unitary matrix which connects the basis of eigenvectors
of Sz with the basis of eigenvectors of Sy .
d) Find the eigenvalues and the normalized eigenvectors of the observable

S~n ≡ nx Sx + ny Sy + nz Sz (2.10)

where

~n = (nx , ny , nz ) = (sin θ cos φ, sin θ sin φ, cos θ) (2.11)

13
θ and φ are polar coordinates: 0 ≤ φ ≤ 2 π and 0 ≤ θ ≤ π.
e) Compute the 2x2 unitary matrix which connects the basis of eigenvectors
of Sz with the basis of eigenvectors of S~n .
Solution:
a) The eigenvalues of any of {Sx , Sy , Sz } are
~
λ± = ± (2.12)
2
The eigenvectors of Sx are obtained from
    
0 1 x± x
=± ± (2.13)
1 0 y± y±

that is
 
~ 1 1
|Sx = ± i = √ (2.14)
2 2 ±1

The eigenvectors of Sy are obtained from

    
0 −i x± x
=± ± (2.15)
i 0 y± y±

that is
 
~ 1 1
|Sy = ± i = √ (2.16)
2 2 ±i

Finally, yhe eigenvectors of Sz are obtained from

    
1 0 x± x
=± ± (2.17)
0 −1 y± y±

that is
   
~ 1 1 ~ 1 0
|Sz = i = √ |Sz = − i = √ (2.18)
2 2 0 2 2 1

b) Let us consider the normalized vector

 
1 1
|vi = √ (2.19)
10 3 i

14
The transition amplitudes are given by the scalar products
~ 1
hSx = |vi = √ (1 + 3 i)
2 20
~ 1 1
hSy = |vi = √ (1 + 3) = √ 4
2 20 20
~ 1
hSz = |vi = √ (2.20)
2 10
The corresponding probabilities are therefore
~ 2 1 1
hSx = |vi = |1 + 3 i|2 =
2 20 2
~ 2 16 4
hSy = |vi = =
2 20 5
~ 2 1
hSz = |vi = (2.21)
2 10
Hence the other probabilities are therefore
~ 2 1 1
hSx = − |vi = 1 − =
2 2 2
~ 2 4 1
hSy = − |vi = 1 − =
2 5 5
~ 2 1 9
hSz = − |vi = 1 − = (2.22)
2 10 10
c)
 
~ 1 1 1 ~ i ~
|Sy = i = √ = √ |Sz = i + √ |Sz = − i
2 2 i 2 2 2 2
 
~ 1 1 1 ~ i ~
|Sy = − i = √ = √ |Sz = i − √ |Sz = − i (2.23)
2 2 −i 2 2 2 2
The unitary matrix is therefore
 
1 1 1
U=√ (2.24)
2 i −i
d) The matrix
 
~ nz nx − iny
S~n = Sx = (2.25)
2 nz + iny nz

15
has eigenvalues
~
λ± = ± (2.26)
2
The eigenvectors are obtained from
    
nz nx − i ny x± x
=± ± (2.27)
nx + i ny −nz y± y±
that is

nz x± + (nx − i ny )y± = ±x± (2.28)

Hence
sin θ e−i φ
   
~ 1 nx − i ny 1
|S~n = ± i = p =p
2 2(1 ∓ nz ) ±1 − nz 2(1 ∓ cos θ) ±1 − cos θ

Therefore
cos 2θ e−i φ
   θ −i φ 
~ ~ sin 2 e
|S~n = i = |S~n = − i = (2.29)
2 sin 2θ 2 − cos 2θ

e)

cos 2θ e−i φ sin 2θ e−i φ

 
U= (2.30)
sin 2θ − cos 2θ
Problem 13. Show that
Z
vp~0 ,∆ (~x) = d3 p~ vp~ (~x) φp~0 ,∆ (~p) (2.31)

with
i
e ~ ~x·~p 1 p−~
(~ p0 )2
vp~ (~x) = 3 φp~0 ,∆ (~p) = 3 3 e− 2 ∆2 (2.32)
(2 π ~) 2 (π) 4 ∆ 2
is a normalizable wave function of unit norm. Compute the average of ~x, p~
∆p2 and ∆x2 on the state described by the packet (2.31). Make use of the
formula for gaussian integrals
Z ∞ r
−α x2 +β x π β2
dx e = e4α (2.33)
−∞ α

16
where α > 0 and β is a generic complex number.

Solution: Let us compute the norm of this vector in the momentum represen-
tation
d3 p~
Z Z
2 (~ p0 )2
p−~
3
hv, vi = d p~ φp~0 ,∆ )(~p) = 3 e− ∆2 =
(π) 2 ∆3
d3 p~
Z
~2
p 1 3 3
= 3 e− ∆ 2 = 3 π 2 (∆2 ) 2 = 1 (2.34)
(π) 2 ∆3 (π) 2 ∆3

Let us compute the average of ~x in the momentum representation

d3 p~
Z
p−~
(~ p0 )2 p−~
(~ p0 )2
− ~ −
hv, ~x vi = 3 e 2 ∆2 i ~ ∇ p
~ e 2 ∆2 =
(π) 2 ∆3
d3 p~ (~p − p~0 )
Z
p−~
(~ p0 )2
−
= −i ~ 3 e ∆2 × =
(π) 2 ∆3 ∆2
d3 p~
Z
~2
p p~
= −i ~ 3 e− ∆2 × 2 = 0 (2.35)
(π) 2 ∆ 3 ∆

Let us compute the average of p~ in the momentum representation

d3 p~
Z
(~ p0 )2
p−~ p−~
(~ p0 )2
hv, p~ vi = 3 e− 2 ∆2 p~ e− 2 ∆2 =
(π) 2 ∆3
d3 p~
Z 2
− p~ 2
= 3 e ∆ (~ p + p~0 ) = p~0 (2.36)
(π) 2 ∆3

Let us compute the average of ~x2 in the momentum representation

d3 p~
Z
p−~
(~ p0 )2 p−~
(~ p )2
2 − 2 ~ 2 − 2 ∆02
hv, ~x vi = − 3 e 2 ∆2 ~ ∇ p
~ e =
(π) 2 ∆3
d3 p~ h (~p − p~ ) (~p−~p0 )2 i
Z
p−~
(~ p0 )2
− 0
= +~ 2
3 e 2 ∆2 × ~
∇ p
~ 2
e− 2 ∆2 =
(π) ∆
2 3 ∆
d3 p~ h (~p − p~ )2
Z
p−~
(~ p0 )2 3 i
2 − 0
= +~ 3 e ∆ 2
× − + 2 =
(π) 2 ∆3 ∆4 ∆
2 Z 3 2
3~ d p~ ~2 p
p ~
= + 2 − ~2 3 e− ∆ 2 4 (2.37)
∆ (π) 2 ∆3 ∆

17
Integrals like those which appear in the r.h.s. can be computed in the following
way
Z Z 3 3
3 x2 2
−α ~ 3 x2
−α ~ π2 3 π2
d ~x e ~x = −∂α d ~x e = −∂α 3 = (2.38)
α2 2 α 25
Hence

2 3 ~2 3 ~2 3 ~2
hv, ~x vi = + 2 − = (2.39)
∆ 2 ∆2 2 ∆2
Therefore the uncertainty on ~x2 is

2 2 3 ~2
2
∆~x = hv, ~x vi − hv, ~x vi = (2.40)
2 ∆2
Let us compute the average of p~2
d3 p~
Z
p−~
(~ p0 )2 p )2
p−~
(~
2 − 2 − 2 ∆02
hv, p~ vi = 3 e 2 ∆2 p
~ e =
(π) 2 ∆3
d3 p~
Z 2
− p~ 2
= 3 e ∆ (~ p + p~0 )2 =
(π) 2 ∆3
3
= p~20 + ∆2 (2.41)
2
Therefore the uncertainty on the p~2 is
3 2
∆~p2 = hv, p~2 vi − hv, p~ vi2 = ∆ (2.42)
2
Note that
√ p 3
∆~x2 ∆~p2 = ~ (2.43)
2
Since the state is rotation invariant

∆~x2 = ∆(x2 + y 2 + z 2 ) = 3 ∆x2

∆~p2 = ∆(p2x + p2y + p2z ) = 3 ∆p2x (2.44)

Therefore
~
∆x × ∆px = ∆y × ∆py = ∆z × ∆pz = (2.45)
2
18
The state considered is therefore a state of minimal uncertainty.
The same computation could be done in the Schrödinger representation.
Let us first compute the wave function in the Schrödinger representation
Z
1 3 i
p~·~
x 1 −
p−~
(~ p0 )2
vp~0 ,∆ (~x) = 3 d p
~ e ~
3 3 e 2 ∆2 =
(2 ~ π) 2 (π) 4 ∆ 2
i
e ~ p~0 ·~x
Z
1 i
~·~
p
2
x− p~ 2
= 3 d3 p~ 3 3 e
~ 2∆ =
(2 ~ π) 2 (π) 4 ∆ 2
i
e ~ p~0 ·~x p− i ∆2 ~x)2
Z
1 −
(~ ∆2 ~
x2
× e− 2 ~2 =
~
3
= 3 d p~ 3 3 e
2∆ 2

(2 ~ π) 2 (π) 4 ∆ 2
i
 ∆  23 e ~ p~0 ·~x ∆2 ~x2
−
= 3 e
2 ~2 (2.46)
~ (π) 4

The averages of ~x and p~ are

 ∆ 3 1 Z ∆2 ~x2
hv, ~x vi = 3 d3~x e− ~2 ~x = 0
~ π2
 ∆ 3 1 Z
3
2 2 h
− ∆ 2~x ∆2 ~x i
hv, p~ vi = d ~
x e ~ i + p
~ 0 = p ~0 (2.47)
~ π 32 ~
The averages of ~x and p~ write
2
 ∆ 3 1 Z
3
2 2
− ∆ 2~x 2 3 ~2
hv, ~x vi = d ~
x e ~ ~
x =
~ π 32 2 ∆2
∆4 ~x2 i
 ∆ 3 1 Z 2 2 h
− ∆ 2~x
hv, p~2 vi = d 3
~
x e ~ 3 ∆ 2
+ p~2
0 − =
~ π 32 ~2
3 3
= p~20 + 3 ∆2 − ∆2 = p~20 + ∆2 (2.48)
2 2
in agreements with the results obtained above in the momentum representa-
tion.

3 Simple systems
Problem 3. Show that the exponential factor which controls the probability
of tunneling through the barrier

+∞
 if x < 0,
V (x) = −V0 if 0 < x < L (3.1)
 Z e2

4 π 0 x
if x > L

19
in the limit
Z e2
E (3.2)
4 π 0 L
is
2 √2m 2
− 4ππZe ~ − 2Z e
T ∝e 0 E
=e  ~v0 (3.3)
Solution:
Let us compute the integral
Z Z e2 s Z 1
4 π 0 E Z e2 Z e2 dy p
dx −E = √ √ 1−y (3.4)
L 4 π 0 x 4 π 0 E 4 πZ0e2LE y
In the limit (3.2)
Z 1 Z 1 Z 1
dy p dy p p π
√ 1−y → √ 1−y = 2 dξ 1 − ξ 2 = (3.5)
4 π 0 LE
2
y 0 y 0 2
Ze

Hence
q
2m Z e2√ π
√2m Z e2 2
−2
~2 4 π 0 E 2
−π − 2Z e
T ∝e =e E 4 π 0 ~
=e  ~v
0 (3.6)
Problem 5. Find the reflection and transmission coefficients R and T for
the rectangular potential barrier for E < V0 , in the limit in which L → 0,
V0 → +∞ and V0 L = V constant. (This is the so-called “thin barrier” limit.)
Solution:
In this limit
√
0 2mV √ 2mV
L→0 Lk → L→0 (k 0 )2 L → (3.7)
~ ~2
Then
16 k 2 |k 0 |2
T = 2 =
(k + i |k 0 |)2 e−|k0 | L − (k − i |k 0 |)2 e|k0 | L
16 k 2 |k 0 |2
=
2
k 2 − |k 0 |2 e−|k0 | L − e|k0 | L + 2 i k |k 0 | e−|k0 | L + e|k0 | L

16 k 2 |k 0 |2 4 k2
→ 2 = 2 →
|k 0 |2 2 |k 0 | L + 2 i k |k 0 | 2 |k 0 |2 L + 2 i k

k2 ~2 k 2 p2
→ 2 = m2 V 2
= m2 V 2
(3.8)
mV + 2 k2 + p2
~2
+ ik ~2 ~ ~2

20
Problem 6. Determine the discrete levels and the energy eigenfunctions of
an unidimensional particle moving in the potential well:
(
0 if |x| > a2 ,
V (x) = (3.9)
V0 if |x| ≤ a2 .

for V0 < 0.
Solution: Discrete states exists only if

V0 < E < 0 (3.10)

Wave functions of discrete eigenstates are either even or odd under reflection
x → −x. Let us denote even (odd) energy eigenfunctions with ψE+ (x) (ψE− (x)).

0 a
A+ cos k x |x| < 2

ψE+ (x) = B+ e−k x x > a2 (3.11)
 +k x a
B+ e x < −2


and

0 a
A− sin k x |x| < 2

ψE− (x) = B− e−k x x > a2 (3.12)

−B− e+k x x < − a2


where
p p
2 m |E| 2 m (E − V0 )
k= k0 = (3.13)
~ ~
a
Continuity of wave function and of its first derivative at x = 2
implies

k0 a ka
A+ cos = B+ e− 2
2
k0 a ka
−k 0 A+ sin = −k B+ e− 2 (3.14)
2
and
k0 a ka
A− sin = B− e− 2
2
k0 a ka
k 0 A− cos = −k B− e− 2 (3.15)
2
21
We get therefore the eigenvalue equation for E, for

k0 a
k 0 tan =k for ψE+
2
k0 a
k 0 cot = −k for ψE− (3.16)
2
Introducing the adimensional variables

k0 a ka
ξ= η= (3.17)
2 2
the equations (3.16) become

η = ξ tan ξ for ψE+

η = −ξ cot ξ for ψE− (3.18)

Moreover we have the relationship between ξ and η

a2 m |V0 |
ξ 2 + η2 = (3.19)
2 ~2
and the Eq. (3.10) means that

a2 m |V0 | a2 m |V0 |
0<ξ< 0<η< (3.20)
2 ~2 2 ~2
The eigenvalues E are determined therefore by the intersections of the two
curves

η = ξ tan ξ for
2
a m |V0 |
ξ 2 + η2 = ψE+ (3.21)
2 ~2
and

η = −ξ cot ξ for
2
a m |V0 |
ξ 2 + η2 = ψE+ (3.22)
2 ~2
which lie in the first quadrant of (ξ, η) plane.

22
 By drawing the graphs of the curves (3.21), we see that the solutions
{ξ1 , ξ3 , ξ5 . . .} corresponding to the energy eigenvalues E1 , E3 , E3 , . . . of the
even wave functions satisfy
π 3π 5
0 < ξ1 < π < ξ3 < 2 π < ξ5 < π (3.23)
2 2 2
Similarly, the solutions {ξ2 , ξ4 , ξ6 . . .} corresponding to the energy eigenvalues
E2 , E4 , E6 , . . . of the even wave functions satisfy
π 3 5π
< ξ2 < π π < ξ4 < 2 π < ξ6 < 3 π (3.24)
2 2 2
The number n of bound states is therefore the highest positive integer such
that
r
π a2 m |V0 |
(n − 1) ≤ (3.25)
2 2 ~2
In particular we have always at leasts 1 bound state,
π
0 < ξ1 < (3.26)
2
for which
π 2 ~2
0 < E1 − V0 < (3.27)
2 m a2
The un-normalized wave functions of such bound states are obtained by
replacing (3.14) and (3.15) into (3.11) and (3.12)

0
cos k x
 |x| < a2
a
ψE+ (x) = cos k2a ek ( 2 −x) x > a2 (3.28)
k a k ( a2 +x)
 a
cos 2 e x < −2


and

0
sin k x
 |x| < a2
0 a
ψE− (x) = sin k2a ek ( 2 −x) x > a2 (3.29)
 0 a
− sin k2a ek ( 2 +x) x < − a2


23
Problem 7. Determine the discrete levels of an unidimensional particle
moving in the potential well

+∞ if x < 0

V (x) = −V0 if 0 < x < L, (3.30)

0 if x > L.


for V0 > 0. Find the value of V0 for which there is one single bound state.
Solution: Bound states are possibles only for −V0 < E < 0. The eigenfunc-
tions have the form
( i 0 i 0
A e ~ p x + B e− ~ p x if 0 < x < L
ψE (x) = 1 1 (3.31)
A0 e ~ p x + B 0 e− ~ p x if x > L

where
p p √
p0 = 2 m (E + V0 ) p= 2 m |E| = −2 m E (3.32)

Since the particle is bound to be in the region x > 0, we have the boundary
condition of the wave function

ψE (0) = 0 (3.33)

Moreover the wave function of a bound state must vanish for x → +∞.
Therefore
(  i 0 i 0 
A e ~ p x − e− ~ p x if 0 < x < L
ψE (x) = 1 (3.34)
B 0 e− ~ p x if x > L

The continuity conditions at x = L are

p0 L pL
2 iA sin = B 0 e− ~
~
i p0 L p pL
2 p0 A cos = −B 0 e− ~
~ ~ ~
(3.35)

Dividing the second equation by the first we obtain the necessary and sufficient
condition for this linear homogenous system to have a solution:
p0 L
p = −p0 cot (3.36)
~
24
Energy eigenvalues are therefore determined by the intersections of the two
curves in the positive quadrant of the (p0 , p) plane
p0 L
p = −p0 cot
~
02 2
p + p = 2 m V0
p>0 p0 > 0 (3.37)

Introducing the adimensional variables

p0 L pL 2 m V0 L2
ξ= η= α2 ≡ (3.38)
~ ~ ~2
the equations of the curves become

η = −ξ cot ξ
ξ 2 + η 2 = α2
η>0 η>0 (3.39)

Drawing the graphs of the two curves we see that minimal α for which an
intersection in the positive quadrant exists is for α = π2
π π
α= ξ= η=0 (3.40)
2 2
Thus the minimal value of V0 for which there is a bound state is
~2 π 2
V0min = (3.41)
8 m L2
When α increases and
π 3π
<α< (3.42)
2 2
there is one single intersection in the positive quadrant. Thus for
~2 π 2 9 ~2 π 2
< V0 < (3.43)
8 m L2 8 m L2
there is a single bound state with
~2 ξ π
E0 = − V0 <ξ<π (3.44)
2 m L2 2
25
 More generally when α reaches the values
π 3π 5π
αn = , , ,... (3.45)
2 2 2
new solutions with η = 0 appear, in correspondence with new bound states.
The new bound states starts at E = 0 for α = αn , and, as α goes from αn to
αn+1 , its energy becomes more negative and lies in the range
2
αn2 ~2 αn+1 ~2
− V0 < E < − V0 (3.46)
2 m L2 2 m L2
When V0 → ∞ the numbers of bound states increases. The ground state
level E0 tends to the limiting value

~2 ξ ~2 π 2
E0 = − V0 → − V0 (3.47)
2 m L2 2 m L2
Problem 8. Determine the discrete levels of an unidimensional particle
moving in the potential well by using the expression obtained for the transfer
matrix for the potential barrier
1
T = ×
4
(k+k0 )2 i (k−k0 ) L 0 )2 0 k2 −(k0 )2 0 0
!
k k0
e − (k−k
k k0
ei (k+k ) L k k0
e−i (k−k ) L − e−i (k+k ) L
× k2 −(k0 )2 0 0 (k+k0 )2 −i (k−k0 ) L 0 )2 0
− (k−k e−i (k+k ) L


k k0
ei (k−k ) L − ei (k+k ) L k k0
e k k0

Solution:
We must put in this case
√ p p
2mE 2 m |E| 2 m (E + V0 )
k= =i = i |k| k0 = (3.48)
~ ~ ~
Moreover we should impose

A = B 00 = 0 A00 = 1 (3.49)

in the equation
   00 
A A
=T (3.50)
B B 00

26
Hence the equation which determine the energy levels is

(i |k| + k 0 )2 i (k−k0 ) L (i |k| − k 0 )2 i (k+k0 ) L

T11 = e − e =0 (3.51)
k k0 k k0
from which
(i |k| + k 0 )2 0

0 2
= e2 i k L (3.52)
(i |k| − k )
Let us put
|k|
i |k| + k 0 = i |k| + k 0 ei θE tan θE = (3.53)
k0
Thus
(i |k| + k 0 )2
0 2
= e4 i θE (3.54)
(i |k| − k )
The equations for the eigenvalues becomes therefore

4 θE = 2k 0 L + 2 n π n = 0, ±1, ±2, . . . (3.55)

i.e.
k0 L n π |k| k0 L n π
θE = + ⇒ tan θE = = tan( + ) (3.56)
2 2 k0 2 2
For n even
|k| k0 L
= tan( ) (3.57)
k0 2
and for n odd
|k| k0 L
− = cot( ) (3.58)
k0 2
in agreement with (3.16).
Problem 9. Determine the energy eigenfunctions for the potential barrier
(
0 if |x| > a2 ,
V (x) = (3.59)
V0 if |x| ≤ a2 .

27
with V0 > 0 and 0 < E < V0 which are also eigenstates of the spatial inversion
operator I : ψ(x) → ψ(−x). Derive the transmission matrix for this barrier.
Solution:
The even and odd energy eigenfunctions with 0 < E < V0 have the form:

e
 +i k x
+ B+ e−i k x x < − a2
ψE+ (x) = A+ cosh k 0 x |x| < a2 (3.60)

B+ ei k x + e−i k x x > a2


and

e
 +i k x
+ B− e−i k x x < − a2
ψE− (x) = A− sinh k 0 x |x| < a2 (3.61)

−B− ei k x − e−i k x x > a2


where
√ p
2mE 2 m |V0 − E|
k= k0 = (3.62)
~ ~
Continuity of the wave-functions and of their first derivative at x = ± a2 gives
+i ka
−i ka k0 a
B+ e +e
2 = A+ cosh
2
2
ka ka
k0 a
i k B+ ei 2 − e−i 2 = A+ k 0 sinh
2
0
ka ka k a
B− ei 2 + e−i 2 = −A− sinh
2
ka ka
k0 a
−i k B− ei 2 − e−i 2 = A+ k 0 cos (3.63)
2
From this
B+ − e−i k a k0 k0 a
= −i tanh
B+ + e−i k a k 2
−i k a 0
B− − e k k0 a
= −i coth (3.64)
B− + e−i k a k 2
and
0 0
1 + i kk tanh k2a −i k a
B+ = k0 k0 a
e = e2 i θ+ (E) e−i k a
1 − i k tanh 2
0 0
1 + i kk coth k2a −i k a
B− = 0 0 e = e2 i θ− (E) e−i k a (3.65)
1 − i kk coth k2a

28
where
k0 k0 a
tan θ+ (E) = tanh
k 2
k0 k0 a
tan θ− (E) = coth (3.66)
k 2
Hence
ka B+ + e−i k a −i k2a 1 + e
2 i θ+ (E)
A+ = e+i 2
0 = e 0 =
cosh k2a cosh k2a
ka B− + e−i k a −i k2a 1 + e
2 i θ− (E)
A− = −e+i 2
0 = −e 0 (3.67)
sinh k2a sinh k2a
The relation that defines the transfer matrix is
    
A> T11 T12 A<
= (3.68)
B> T21 T22 B<
where {A> , B> } ({A< , B< }) are the coefficients of the right-moving and left-
moving plane waves to which the energy eigenfunction reduces for x → +∞ (
x → −∞). By comparing with (3.60) and (3.61) we deduce
    
B+ T11 T12 1
=
1 T21 T22 B+
    
−B− T11 T12 1
= (3.69)
−1 T21 T22 B−
and hence
    
B+ − B− T11 T12 2
=
0 T21 T22 B+ + B−
    
B+ + B− T11 T12 0
= (3.70)
2 T21 T22 B+ − B−
The reflection and transmission coefficients are therefore
B+ + B− 2 1 2 i θ+ (E) 2
R= = e + e2 i θ− (E) =
2 4

1 + cos 2 θ+ (E) − θ− (E)
= cos2 θ+ (E) − θ− (E)


=
2
B+ − B− 2 1 2 i θ+ (E) 2
T = = e − e2 i θ− (E) =
2 4

1 − cos 2 θ+ (E) − θ− (E)
= sin2 θ+ (E) − θ− (E) (3.71)


=
2
29
which shows that

R+T =1 (3.72)

We can also express the elements of the transfer matrix in terms of θ± (E)
2  i (k 2 − (k 0 )2 ) 
T22 = = ei k a cosh(k 0 a) + sinh(k 0
a)
B+ − B− 2 k k0
B+ + B− i (k 2 + (k 0 )2 )
T12 = −T21 = = sinh(k 0 a)
B+ − B− 2 k k0
2 B+ B−
T11 =− =
B+ − B−
 i (k 2 − (k 0 )2 ) 
= e−i k a cosh(k 0 a) − sinh(k 0
a) (3.73)
2 k k0
and the transmission coefficient
1 1
T = = (k2 −(k0 )2 )2
=
|T22 |2 cosh2 (k 0 a) + sinh2 (k 0 a)
4 k2 (k0 )2
1
= √ (1−2 )2 √ (3.74)
cosh2 (g 1 − ) + 4  (1−)
sinh2 (g 1 − )

where
r
E 2 m a2 V0
≡ g≡ (3.75)
V0 ~2
in agreement with the result obtained in the main text. In the limit of “high”
barrier
√
k0 a  1 ⇒ g 1 − x  1 (3.76)

we obtain
0 4 k 2 (k 0 )2 E  E  −2 k0 a
T ≈ 4 e−2 k a
= 16 1 − e (3.77)
(k 2 + (k 0 )2 )2 V0 V0
When the energy tends to the threshold value E → V0− , the transmission
coefficient tends to the value
1
lim− T = m V0 L2
(3.78)
E→V0 1+ 2 ~2

30
For E > V0 we derive analogously
1
T = (k2 +|k0 |2 )2
=
cos2 (|k 0 |a) + 4 k2 |k0 |2
sin2 (|k 0 |a)
1
= √ (1−2 )2 √ (3.79)
cos2 (g  − 1) + 4  (−1)
sin2 (g  − 1)

Problem 10.
A 1-dimensional particle moves on the half line x > 0 in the potential


 +∞ if x < 0,

0 if 0 < x < L.
V (x) = (3.80)
V0
 if L < x < L + a.


0 if L + a < x.

with V0 > 0. Determine the energy eigenfunctions.

Solution:
The spectrum is continuous and not-degenerate for E > 0. For 0 < E < V0
the energy eigenfunctions have the form

sin k x
 0<x<L
0 k 0x 0 −k 0x
ψE (x) = A e + B e L<x<L+a (3.81)

 ikx −i k x
Ae +Be x>L+a

where
√ p
2mE 2 m (V0 − E)
k= k0 = (3.82)
~ ~
Continuity of the wave function and of its first derivative at x = L gives
0 0
sin k L = A0 ek L + B 0 e−k L
k 0 0

0
cos k L = A0 ek L − B 0 e−k L (3.83)
k
From this we obtain
1 −k0 L k
A0 = e sin k L + 0 cos k L)
2 k
1 0 k
B 0 = ek L sin k L − 0 cos k L) (3.84)
2 k
31
Continuity of the wave function and of its first derivative at x = L + a gives
0 0
A ei k (L+a) + B e−i k (L+a) = A0 ek (L+a) + B 0 e−k (L+a) =
1 k 0 1 k 0
= sin k L + 0 cos k L) ek a + sin k L − 0 cos k L) e−k a =
2 k 2 k
k
= sin(k L) cosh(k 0 a) + 0 cos(k L) sinh(k 0 a)
k
0
k 0 0
A ei k (L+a) − B e−i k (L+a) = A0 ek (L+a) − B 0 e−k (L+a) =


ik
0
k
= −i sin(k L) sinh(k 0 a) − i cos(k L) cosh(k 0 a) (3.85)
k
From this we obtain
1 k
A = e−i k (L+a) sin(k L) cosh(k 0 a) + 0 cos(k L) sinh(k 0 a) +

2 k
0
k
−i cos(k L) cosh(k 0 a) − i sin(k L) sinh(k 0 a)

k
B = A∗ (3.86)
Problem 11. Compute the averages of x̂2 and of p̂2 on the ground state |0i
of the harmonic oscillator. (Hint: write x̂ and p̂ in terms of a and a† .)
Solution:
~
x̂ = i λ a − a† a + a†



p̂ = (3.87)
2λ
Therefore
h0|x̂|0i = 0 h0|p̂|0i = 0 (3.88)
Moreover
x̂2 = −λ2 a†2 + a2 − a† a − a a†

~2
p̂2 = a†2 + a2 + a† a + a a†


2
(3.89)
4λ
Therefore
h0|x̂2 |0i = −λ2 h0| a†2 + a2 − a† a − a a† |0i = λ2 h0| a a† |0i = λ2

~2 †2 † † ~2
2 2
h0| a a† |0i =


h0|p̂ |0i = 2
h0| a + a + a a + a a |0i = 2
4λ 4λ
~2
= (3.90)
4 λ2
32
In conclusion

2
∆x2 = h0|x̂2 |0i − h0|x̂|0i = λ2

2 ~2
∆p2 = h0|p̂2 |0i − h0|p̂|0i = (3.91)
4 λ2
and thus
~
∆x ∆p = (3.92)
2
The ground state is therefore a state of minimal uncertainty.
Problem 12. Consider the state ψ of an harmonic oscillator of frequency ω
and mass m which satisfies the equation

aψ = αψ (3.93)

where α is a complex number.

a) Find the normalized wave function of ψ in the Schrödinger representa-
tion.
b) Compute ∆x and ∆p on the state ψ and show it is a state of minimal
indeterminacy.
c) Consider the state
∞
X αn
ψ̃ = √ |ni (3.94)
n=0 n!

where |ni with n = 0, 1, 2, . . . are the normalized eigenstates of the harmonic

oscillator Hamiltonian. Show that ψ̃ satisfies (3.93).
d) Compute the norm of ψ̃.
e) Let ψ̂ be a normalized vector corresponding to ψ̃. By comparing the
wave function in the Schrödinger representation of ψ̂ with the normalized
wave-function computed in a), derive an explicit formula for the Hermite
polynomials for any n.
Solution:
a) The wave-function ψ(x) satisfies the differential equation

~α
−i ~ ψ 0 (x) − i m ω x ψ(x) = ψ(x) (3.95)
λ

33
that is
ψ 0 (x) iα x
= − (3.96)
ψ(x) λ 2 λ2

The solution is
x x2
ψ(x) = C ei α λ − 4 λ2 (3.97)

The normalized wave-function is obtained by imposing

Z ∞ 2
Z ∞
x y2
2 i (α−ᾱ) λ − x2 2
1 = hψ, ψi = |C| dx e 2λ = |C| λ dy ei (α−ᾱ) y− 2 =
−∞ −∞
2
√ −
(α−ᾱ)2
= |C| λ 2 π e 2 (3.98)

Hence
1 1 (α−ᾱ)2 x x2
ψ(x) = 1 1 e 4 ei α λ − 4 λ2 (3.99)
(2 π) 4 λ 2

b) Let us compute ∆x by making use of the Schrödinger representation:

Z ∞ 2
Z ∞
x y2
2 i (α−ᾱ) λ − x2 2 2
hψ, x̂ ψi = |C| dx x e 2λ = |C| λ dy y ei (α−ᾱ) y− 2 =
−∞ −∞
√ ∂ (α−ᾱ)2
= −i |C|2 λ2 2 π e− 2 =
∂(α − ᾱ)
√ (α−ᾱ)2
= i |C|2 λ2 2 π (α − ᾱ) e− 2 = i (α − ᾱ) λ (3.100)

and
Z ∞ 2
Z ∞
x y2
2 2 2 i (α−ᾱ) − x2 2 3
hψ, x̂ ψi = |C| dx x e λ 2λ = |C| λ dy y 2 ei (α−ᾱ) y− 2 =
−∞ −∞
√ ∂ 2
−
(α−ᾱ)2
= −|C|2 λ3 2 π e 2 =
∂(α − ᾱ)2
√
(α−ᾱ)2
= |C|2 λ3 2 π 1 − (α − ᾱ)2 e− 2 = λ2 1 − (α − ᾱ)2 (3.101)

Thus

∆x2 = λ2 1 − (α − ᾱ)2 + (α − ᾱ)2 = λ2

 
(3.102)

34
For ∆p, let us show how to perform the computation by using raising and
lowering operators together with Eq. (3.93). As a matter of fact, this method
is simpler:
~ ~
hψ, p̂ ψi = hψ, (a + a† ) ψi = hψ, (α + a† ) ψi =
2λ 2λ
~ ~
= (α + ha ψ, ψi) = (α + ᾱ) (3.103)
2λ 2λ
And
~2
hψ, p̂2 ψi = 2
h(a + a† )ψ, (a + a† ) ψi =
4λ
~2
|α| + ha† ψ, a† ψi + α2 + ᾱ2 =
 2 
=
4 λ2
~2  2
|α| + |α|2 + 1 + α2 + ᾱ2 =

=
4 λ2
~2
2 |α|2 + 1 + α2 + ᾱ2
 
= (3.104)
4 λ2
Hence
~2  ~2
∆p2 = 2 2 2 2 2 2

2 |α| + 1 + α + ᾱ − α − ᾱ − 2 |α| = (3.105)
4 λ2 4 λ2
We verified that the state is indeed a state of minimal indeterminacy
~
∆x ∆p = (3.106)
2
c)
∞ ∞
X αn √ X αn+1 √
a ψ̃ = √ n |n − 1i = p n + 1 |ni =
n=1 n! n=0 (n + 1)!
∞
X αn
=α √ |ni = α ψ̃ (3.107)
n=0 n!

d)
∞ ∞
X ᾱn αn X |α|2 n 2
hψ̃, ψ̃i = √ √ = = e|α| (3.108)
n=0 n! n! n=0 n!

35
e) From d) we deduce that the vector
∞
−
|α|2
−
|α|2 X αn
ψ̂ = e 2 ψ̃ = e 2 √ |ni (3.109)
n=0 n!

has unit norm. The wave function of ψ̂ in the Schrödinger representation is

∞ ∞
−
|α|2 X αn |α|2 X α
− 2
n Hn ( λx ) − x22
ψ̂(x) = e 2 √ ψn (x) = e √ 1 1 e
4λ (3.110)
n=0 n! n=0 n! (2 π) 4 λ2

Here the ψn (x) are the wave-functions of the energy eigenstates of the harmonic
oscillator and the polynomials Hn (y) are defined according to

Hn ( λx ) x2
ψn (x) = hx|ni = 1 1 e− 4 λ2 (3.111)
(2 π) 4 λ 2

Hn (y) are polynomial of degree n, of parity (−1)n , which are (simply related
to) the so-called Hermite polynomials. In the following we will derive a closed
expression for Hn (x) by comparing Eq. (3.110) with the result obtained in a).
Both ψ̂(x) and ψ(x) satisfy the equation (3.93) and they both have unit
norm: hence, they can differ at most by a x-independent factor of unit norm

ψ̂(x) = ei φ(α,ᾱ) ψ(x) (3.112)

where φ(α, ᾱ) is a real-valued function of α and ᾱ. It must therefore be that
∞
1 |α|2 X αn x x2 ei φ(α,ᾱ) (α−ᾱ)2 i α λx − x22
1 1 e − 2 √ Hn ( ) e− 4 λ2 = 1 1 e
4 e 4 λ (3.113)
(2 π) 4 λ 2 n=0 n! λ (2 π) 4 λ 2

or, dividing by the common factors,

∞
X αn α2 ᾱ2
√ Hn (y) = ei φ(α,ᾱ) e 4 + 4 ei α y (3.114)
n=0 n!

where y ≡ λx . The LHS of this equation depends only on α. This means that
the factor ei φ(α,ᾱ) must cancel the dependence on ᾱ of the RHS. In other
words the combination
ᾱ2
ei φ(α,ᾱ) e 4 = F (α) (3.115)

36
must depend on α but not on ᾱ. Since i φ(α, ᾱ) is pure imaginary, we conclude
that
ᾱ2 2
+ α4 +i c
ei φ(α,ᾱ) = e− 4 (3.116)

where c is a real constant independent of both α and ᾱ. Then

∞
X αn α2
√ Hn (y) = ei c e 2 ei α y (3.117)
n=0 n!

The value of the real constant c is determined by taking α → 0 in both sides

of this equation. In this limit the LHS reduces to H0 (y) = 1 and thus

c=0 (3.118)

We established therefore the identity

∞
X αn α2
√ Hn (y) = e 2 ei α y (3.119)
n=0 n!

By expanding the RHS in powers of α, and by equating terms with the same
power of α, we obtain the following formula for Hn (x)1
n
[2] √
X n! (−1)k
Hn (y) = in y n−2 k (3.120)
k=0
2k k! (n − 2 k)!

4 Time evolution
Problem 1. Let E+ and E− the two energy eigenvalues of a 2-states system
and |+i and |−i the two corresponding energy eigenstates. Suppose that at
1
The polynomials Hn (y) defined in (3.111) are related to the Hermite polynomials hn (y)
by the formula
in y

Hn (y) = √ hn √
n! 2n 2

The factors in stems from the fact that our definition for a† differs by a factor i from the
conventional one. The factor √n!1 2n is a factor which is included in the usual definition of
the harmonic oscillator energy eigenfunctions (3.111).

37
the time t = 0 the system is in the state ψ(0) = √12 |+i + |−i . What is the

probability that at the time t > 0 the system is in the state |+i?
Solution: The state ψ(t) at the time t is
i i 1
ψ(t) = e− ~ t Ĥ ψ(0) = e− ~ t Ĥ √ |+i + |−i =


2
1 i i
= √ e− ~ t E+ |+i + e− ~ t E− |−i


(4.121)
2
The probability that at the time t the system is in the state |+i is
i
2 e− ~ E+ t 2 1
P+ (t) = h+|ψ(t)i = √ = (4.122)
2 2

The probability that at the time t the system is in the state ψ(0) is:
i i
2 e− ~ E+ t e− ~ E− t 2
Pψ(0) (t) = hψ(0), ψ(t)i = + =
2 2
1 (E+ − E− ) t
1 (E+ − E− ) t

= 1 + 1 + 2 cos = 1 + cos =
4 ~ 2 ~
(E+ − E− ) t
= cos2 (4.123)
2~
Problem 2
An harmonic oscillator of frequency ω is, at time t = 0, in the state

ψ0 = |0i + |1i + |2i (4.124)

• a) Compute the average of p̂ and p̂2 at the time t.

• b) Write the normalized wave function of the state at the time t in the
Schrödinger representation.

• c) What is the probability that at the time t > 0 the system is still in
the state ψ0 ?

Solution: The state at the time t is

ωt 
ψ(t) = e−i 2 |0i + e−i ω t |1i + e−2 i ω t |2i

(4.125)

38
• a) Therefore

~
hψ(t), a + a† ψ(t)i =


hψ(t), p̂ ψ(t)i =
2λ
ωt ~ h √
= e−i 2 hψ(t), e−i ω t |0i + e−2 i ω t 2 |1i +
2λ
−i ω t
√ −2 i ω t
√ i
+|1i + e 2 |2i + e 3 |3i i =
~ h
−i ω t −i ω t
√ iωt iωt
√ i
= e +e 2+e +e 2 =
2λ
~ √
= (1 + 2) cos(ω t) (4.126)
λ
The average of p̂ is therefore

hψ(t), p̂ ψ(t)i ~ √
p̄(t) = = (1 + 2) cos(ω t) (4.127)
hψ(t), ψ(t)i 3λ

• b) Moreover

hψ(t), p̂2 ψ(t)i = hp̂ ψ(t), p̂ ψ(t)i =

~2 −i ω t √ −2 i ω t
= e |0i + (1 + 2e ) |1i +
4 λ2
√ √ 2
+e−i ω t 2 |2i + e−2 i ω t 3 |3i =
~2 √ −2 i ω t 2

= 1 + |1 + 2 e | + 2 + 3 =
4 λ2
~2 √

= 2
6 + (3 + 2 2 cos(2 ω t)) =
4λ
~2 √

= 9 + 2 2 cos(2 ω t) (4.128)
4 λ2
The average of p̂2 is therefore

2 1 2 ~2 √

p̄ (t) = hψ(t), p̂ ψ(t)i = 9 + 2 2 cos(2 ω t) (4.129)
3 12 λ2

• c) The probability that at the time t > 0 the system is still in the state

39
 ψ0 is
1 1 2
Pψ0 →ψ0 (t) = hψ0 , ψ(t)ih|0i + |1i + |2i, =
9 9
2 1 2
|0i + e−i ω t |1i + e−2 i ω t |2ii = 1 + e−i ω t + e−2 i ω t =
9
1

= 3 + 4 cos(ω t) + 2 cos(2ω t) (4.130)
9
Problem 4. Determine the time evolution of the free wave packet
i
(p−p̄)2 px
Z
1 −
2 ∆2
e ~
vp̄,∆p (x) = 1 1 dp e p √ (4.131)
(π) 4 ∆p2 2π~
Solution: The wave function at the time t is
i i p2
(p−p̄)2
e ~ p x− ~ 2 m t
Z
− ~i Ĥ t 1 −
2 ∆2
vp̄,∆p (x; t) = e vp̄,∆p (x) = 1 1 dp e p √ =
(π) 4 ∆p2 2π~
i i p̄2 t ∆2 2
e ~ p̄ x− ~ 2 m
Z p
−(1+i t ) p + ~i
m ~ 2 ∆2
p̄
p (x− m t)
= 1 √ 1 dp e p =
(π) 4 2 π ~ ∆p2
i √
p̄ x− ~i p̄2 t
− p
p̄
∆2 (x− m t)2
e ~ 2 π ∆p 2m
2 ~2 (1+i t
∆2
p
)
= 1 √ 1 q
2
e ~m =
i ∆
(π) 4 2 π ~ ∆p
2
~ m
p
t + 1
p̄
∆2p (x− m t)
2
i i p̄2 t 1
−
e ~ p̄ x− ~ 2m ∆p2 2 ~2 (1+i t
∆2
p
)
= 1 1 q e ~m (4.132)
~ 2 (π) 4 i ∆2p
m~
t+1
Problem 5. Compute the uncertainties of x̂ and p̂ on the free wave packet
(4.131) for t > 0. (Hint: Use the Heisenberg picture of time evolution.)
Solution: Since the particle is free, the time dependent operators x̂(t) and
p̂(t) in the Heisenberg picture are
p̂
x̂(t) = x̂ + t p̂(t) = p̂(0) = p̂ (4.133)
m
Therefore
p̂2 2 t
x̂2 (t) = x̂2 + 2
t + (x̂ p̂ + p̂ x̂)
m m
2 2 2
p̂ (t) = p̂ (0) = p̂ (4.134)

40
Denote by |vi the state described by the wave function (4.131):
p
∆p − ∆2p x22 i p̄ x
vp̄,∆p (x) = 1 √ e 2~ e~ (4.135)
(π) 4 ~
Then
t2 t
hv|x̂2 (t) |vi = hv|x̂2 |vi + hv|p̂ 2
|vi + hv|(x̂ p̂ + p̂ x̂) |vi
m2 m

2 t
2
hv|x̂(t) |vi = hv|x̂|vi + hv|p̂|vi =
m
2
t t
= hv|x̂|vi2 + 2 hv|p̂|vi2 + 2 hv|p̂|vi hv|x̂|vi
m m
t2 2
= 2 p̄
m
hv|p̂ (t) |vi = hv|p̂2 |vi
2

hv|p̂ |vi2 = p̄2 (4.136)

since
hv|p̂|vi = p̄ hv|x̂|vi = 0 (4.137)
The uncertainties on x and p at the time t are therefore
t2 2t i~t
2 2
∆x (t) = ∆x + 2 ∆p2 + hv|x̂ p̂|vi −
m m m
∆p2 (t) = ∆p2 (4.138)
We are left with the computation of the matrix element
∆2 2
∆2p

Z
∆p px
hv|x̂ p̂|vi = √ dx e− ~2 x p̄ + i x2 =
π~ ~
√ 3
∆p i ∆2p ∆2 x2 i ∆3p
Z
− p2 2 π~
=√ dx e ~ x = √ 2 =
π~ ~ π ~ 2 ∆3p
i~
= (4.139)
2
In conclusion
r
t2
∆x(t) = ∆x2 + 2 ∆p2
m
∆p(t) = ∆p (4.140)

41
Problem 6. Estimate the speed at which the following free wave packets, of
minimal indeterminacy at t = 0, spread for large t:
• An electron with ∆x ≈ 10−8 cm.
• An electron with ∆x ≈ 10−2 cm.
• A grain of sand of m = 10−12 g, with ∆x = 10−5 cm.
Solution: The speed at which the packets of minimal indeterminacy spread is,
according to the solution of the previous problem, is for large t
d∆x(t) ~
≈ (4.141)
dt 2 m ∆x
Recall that the electron Compton wavelength is
~
λc = = 0.387 × 10−12 m (4.142)
mc
Therefore for an electron
d∆x(t) λc
≈ c (4.143)
dt ∆x
For ∆x = 10−8 cm = 10−10 m
d∆x(t) 1
≈ × 0.387 × 10−2 × 3 × 108 m/s = 0.580 × 106 m/s (4.144)
dt 2
For ∆x = 10 cm = 10−4 m
−2

d∆x(t) 1
≈ × 0.387 × 10−8 × 3 × 108 m/s = 0.580 m/s (4.145)
dt 2
For a grain of sand of mass
m = 10−12 g = 10−15 kg = 1.10 × 1015 melectron (4.146)
and ∆x = 10−5 cm = 10−7 m, we have
d∆x(t) 1 0.387 × 10−5
≈ × × 3 × 108 m/s = 0.527 × 10−12 m/s (4.147)
dt 2 1.10 × 1015
“Large” time t means
m ∆x m ∆x2 ∆x
t = = ~ (4.148)
∆p ~ m ∆x
that is
~
t  ∆x (4.149)
m ∆x
42
5 Symmetries
Problem 4. A spin 1 system is in the state with Sz = +1. Find the
probabilities that the measurement of Ŝx gives the values Sx = −1, 0, 1.
Solution.
Let {|Sz i, Sz = −1, 0, 1} of eigenstates of Ŝz be the states of eigenstates
of Ŝz . Let us determine the matrix which represents Ŝx in such a basis. We
have
Ŝ+ + Ŝ− Ŝ+ − Ŝ−
Ŝx = Ŝy = (5.1)
2 2i
and
√ √
Ŝ+ |1i = 0 Ŝ+ |0i = 2 |1i Ŝ+ | − 1i = 2 |0i
√ √
Ŝ− |1i = 2 |0i Ŝ− |0i = 2 | − 1i Ŝ− | − 1i = 0 (5.2)
Therefore
1 1
1
Ŝx |1i = √ |0i Ŝx |0i = √ |1i + | − 1i Ŝx | − 1i = √ |0i
2 2 2
The matrix which represents Ŝx in the {|1i, |0i, | − 1i} basis is therefore
 
0 √12 0
Ŝx =  √12 0 √12  (5.3)
 
1
0 √2 0
The eigenvalues of this matrix are given by the equation
 
−mx √12 0
1 1 mx
0 = det  √12 −mx √12  = −mx (m2x − ) − √ (− √ ) =
 
√1
2 2 2
0 2
−mx
= mx (1 − m2x ) (5.4)
and are, as expected, mx = 1, 0, −1. The eigenstate (x, y, z) corresponding to
the eigenvalue mx satisfies
y
−mx x + √ = 0
2
x z
√ − mx y + √ = 0
2 2
y
√ − mx z = 0 (5.5)
2

43
For mx = 1 this gives
y
x=z= √ (5.6)
2
Therefore the normalized eigenstate with mx = 1 is
1
2
|Sx = 1i =  √12  (5.7)
1
2

When mx = 0 we have

y=0 x = −z (5.8)

from which one derives the normalized eigenstate with mx = 0

 1 
√
2
|Sx = 0i =  0  (5.9)
− √12

Finally, Eqs. (5.5) for mx = −1 become

y
x = z = −√ (5.10)
2
from which
 1 
2
|Sx = −1i = − √1  (5.11)
2
1
2

These equations rewrite in a representation independent form as follows

1 1 1
|Sx = 1i = |1i + √ |0i + | − 1i
2 2 2
1 1
|Sx = 0i = √ |1i − √ | − 1i
2 2
1 1 1
|Sx = −1i = |1i − √ |0i + | − 1i (5.12)
2 2 2

44
The inverse relations are
1 1 1
|1i = |Sx = 1i + √ |Sx = 0i + |Sx = −1i
2 2 2
1 1
|0i = √ |Sx = 1i − √ |Sx = −1i
2 2
1 1 1
| − 1i = |Sx = 1i − √ |Sx = 0i + |Sx = −1i (5.13)
2 2 2
We conclude that the probabilities to obtain Sx = −1, 0, 1 on the state |1i are
1 1 1
P (Sx = +1) = P (Sx = 0) = P (Sx = +1) = (5.14)
4 2 4
Problem 9. Find the basis Hn;m (x), of harmonic polynomials of degree

n = 3 and n = 4 which are eigenfunctions of L3 with m.

Solution.
(n = 3) We have
H3,±3 (x) = x3± H3,±2 = x2± x3 (5.15)
Moreover, the eigenfunctions of Lz with eigenvalues m = 0 and m = 1
must have the form
H3,1 = x+ x23 + α x2+ x− H3,0 = x33 + β x+ x− x3 (5.16)
The condition that H3,1 be harmonic gives
 ∂2 ∂ ∂ 
~ 2 H3,1 =
∇ + 4 H3,1 = 2 x+ + 8 α x+ (5.17)
∂x23 ∂x+ ∂x−
Hence
1
α=− (5.18)
4
and
1 2
H3,1 = x+ x23 − x x− (5.19)
4 +
Therefore
1 2
H3,−1 = x− x23 − x x+ (5.20)
4 −
45
 Analogously
~ 2 H3,0 = (6 + 4 β) x3
∇ (5.21)

Hence
3
β=− (5.22)
2
and
3
H3,0 = x33 − x+ x− x3 (5.23)
2
Problem 10.
Consider a 3-dimensional particle of mass m moving in a harmonic poten-
tial
1
V (r) = m ω 2 r2 (5.24)
2
Show that:

• a) The energy levels are given by En = ~ (n + 23 ) with n = 0, 1, 2, . . ..

• b) The ground state is non-degenerate and has l = 0. Find the normal-

ized radial wave function χE0 ,0 (r) with l = 0.

• c) The first excited level E1 has degeneracy 3 and l = 1. Find the

normalized radial wave function χE1 ,1 (r) with l = 1.
P3
(Hint: Since one can write the 3-dimensional hamiltonian Ĥ = i=1 Ĥi
p2i 1 2 2
where Ĥi = 2 m + 2 m ω xi is the hamiltonian of the 1-dimensional harmonic
oscillator, the energy eigenfunctions of the 3-dimensional oscillator are

ψn1 ,n2 ,n3 (~x) = ψn1 (x1 ) ψn2 (x2 ) ψn3 (x3 ) (5.25)

where ψni (xi ) are the eigenfunctions of the 1-dimensional harmonic oscillators.)
Solution.
a) The energy levels corresponding to the wave functions

ψn1 ,n2 ,n3 (~x) = ψn1 (x1 ) ψn2 (x2 ) ψn3 (x3 ) (5.26)

46
are
3
X 1 3
En1 ,n2 ,n3 = ~ ω (ni + ) = ~ ω(n + ) (5.27)
i=1
2 2

where

n ≡ n1 + n2 + n3 (5.28)

b) The ground state corresponds to

n = 0 = n1 = n2 = n3 (5.29)

It is non-degenerate and has wave function

3 
Y m ω  14 mω
x2i
ψ0,0,0 (~x) = ψ0 (x1 ) ψ0 (x2 ) ψ0 (x3 ) = e− 2 ~ =
i=1
π~
 m ω  43
−m ω
r2
= e 2~ (5.30)
π~
Since this level is non-degenerate, it necessarily has l = 0 and m = 0:

ψ0,0,0 (~x) = Y0,0 (θ, φ) RE0 ,0 (r) (5.31)

Therefore the radial wave function is

√  m ω  34 − m ω r2 2  m ω  34 − m ω r2
χE0 ,0 (r) = r RE0 ,0 (r) = 4 π re 2 ~ = 1 r e 2 ~ (5.32)
π~ π4 ~

c) The first excited level has n = 1 and therefore it corresponds to the

three degenerate normalized wave functions
r
2 m ω  m ω  34 − m ω r2
ψ1,0,0 (~x) = x1 e 2~
~ π~
r
2 m ω  m ω  34 − m ω r2
ψ0,1,0 (~x) = x2 e 2~
~ π~
r
2 m ω  m ω  34 − m ω r2
ψ0,0,1 (~x) = x3 e 2~ (5.33)
~ π~

47
The wave function ψ0,0,1 can re-written as
r  m ω  12  m ω  34 m ω 2
8π
ψ0,0,1 (~x) = Y1,0 (θ, φ) r e− 2 ~ r =
3 ~ π~
= Y1,0 (θ, φ) RE1 ,1 (r) (5.34)

where
r
8 r  m ω  45 − m ω r2
RE1 ,1 (r) = e 2~ (5.35)
3 π 14 ~
Hence this is a state with l = 1 and m = 0 and the associated radial wave
function is
r
1 8 2  m ω  54 − m ω r2
χE1 ,1 (r) = r RE1 ,1 (r) = 1 r e 2~ (5.36)
π4 3 π~
The following linear combinations of the ψ1,0,0 and ψ0,1,0 are the wave functions
of the states with l = 1 and m = ±1
1
(x1 ± i x2 )  m ω  54 − m ω r2
√ ψ1,0,0 (~x) ± i ψ0,1,0 (~x) = 3 e 2~ =
2 π 4 ~
r
8 r  m ω  54 − m ω r2
= ± i Y1,±1 (θ, φ) e 2~ =
3 π 14 ~
= ±i Y1,±1 (θ, φ) RE1 ,1 (r) (5.37)

Problem 11.
Show that the radial wave function of the ground state of the hydrogenoid

 Z  32 Zr
−a
χE,0 (r) = 2 re B (5.38)
aB
has norm one.
Solution.

Z ∞  Z 3 Z ∞
2 − 2aZ r
dr |χE,0 (r)| = 4 dr r2 e B =
0 aB 0
1 ∞
Z
= dρ ρ2 e−ρ = 1 (5.39)
2 0

48
Problem 12. Find the normalized radial wave function of the first excited
state with l = 0 of the hydrogenoid atom by looking for solutions of

χE,0 (ρ) = ρ b0 + b1 ρ e−c ρ

c>0 (5.40)

Solution. We have
h
i
χ0E,0 (ρ) = b0 + 2 b1 ρ − cρ b0 + b1 ρ e−c ρ =
h i
= b0 + (2 b1 − c b0 ) ρ − c b1 ρ e−c ρ
2

h
χ00E,0 (ρ) = (2 b1 − c b0 ) − 2 c b1 ρ − c b0 +
i
−c (2 b1 − c b0 ) ρ + c2 b1 ρ2 e−c ρ =
h i
= (2 b1 − 2 c b0 ) + (c2 b0 − 4 c b1 ) ρ + c2 b1 ρ2 e−c ρ (5.41)

Therefore
χ00E,0 (ρ) (2 b1 − 2 c b0 ) + (c2 b0 − 4 c b1 ) ρ + c2 b1 ρ2
=
=
χE,0 (ρ) ρ b0 + b1 ρ
E 2
=− 2 H − (5.42)
Z |E1 | ρ

Equivalently

(2 b1 − 2 c b0 ) + (c2 b0 − 4 c b1 ) ρ + c2 b1 ρ2 =
E


= − ρ 2 H + 2 b0 + b1 ρ =
Z |E1 |
E
E
= −2 b0 − 2 b1 + b0 2 H ρ − ρ2 2 H b1 (5.43)
Z |E1 | Z |E1 |

From this we obtain the equations

2 b1 − 2 c b0 = −2 b0
E
−c2 b0 + 4 c b1 = 2 b1 + b0
Z2 |E1H |
E
c 2 b1 = − b1 (5.44)
Z2 |E1H |

49
From the third equation we obtain
E
c2 = − (5.45)
Z2 |E1H |
Replacing this in the second equation we conclude
(4 c − 2) b1 = 0 (5.46)
If c 6= 12 , this would give b1 = 0. Replacing this value in the first equation we
would obtain
(c − 1) b0 = 0 (5.47)
Since b0 6= 0 (otherwise we would recover the trivial solution b0 = b1 = 0),
then c = 1. But this is the ground state solution ∝ e−ρ . The excited state
corresponds to choosing the solution with b1 6= 0, and therefore
1
c= (5.48)
2
The first equation gives then
1
b1 = − b0 (5.49)
2
In conclusion there is a normalizable solution, whose energy eigenvalue is
Z 2 E1
E2 = (5.50)
4
and whose radial wave function is
1
−ρ
χE2 ,0 (ρ) = b0 ρ 1 − ρ e 2 (5.51)
2
This is the |2, 0, 0i excited state of the hydrogenoid. The normalization factor
is determined by
aB ∞ aB |b0 |2 ∞
Z Z
2 1
2
1= dρ χE2 ,0 (ρ) = dρ ρ2 1 − ρ e−ρ =
Z 0 Z 0 2
2
2 |b0 | aB
= (5.52)
Z
Therefore
r
Z
b0 = (5.53)
2 aB

50