Spring Semester 2020/21 TD Quantum Mechanics, Part 1

TD Quantum Mechanics, 2020-2021

Part 1: Waves and Particles
Basic mathematical notions that are useful for the exercises
Let D(R) = {f R → C|f is continuous and C ∞ with compact support} be a space of test
functions. A distribution is a map

T : D(R) −→ C ,
ϕ 7−→ (T, ϕ) (1)

which satisfies the following properties:

• linearity: ∀λ1,2 ∈ C , ∀ϕ1,2 ∈ D(R) : (T, λ1 ϕ1 + λ2 ϕ2 ) = λ1 (T, ϕ1 ) + λ2 (T, ϕ2 )

D(R)
• continuity: ∀ (ϕn )n∈N∗ a series in D(R) with limn→∞ −→ ϕ

lim (T, ϕn ) → (T, ϕ) . (2)

n→∞

The distribution δx0 with support in x0 ∈ R is the map δx0 : ϕ 7−→ (δx0 , ϕ) = ϕ(x0 ).

Basic physics notions that are useful for the exercises

• Fundamental constants in nature:

Planck constant: h = 6.6260 · 10−34 Js = 4, 1343 · 10−15 eVs ,
reduced Planck constant: ~ = h/(2π) = 1.0546 · 10−34 Js = 6.5821 · 10−16 eVs ,
speed of light in vacuum: c = 299 792 458 m/s ≈ 2.9979 · 108 m/s ,
Boltzmann constant: kB = 1.3806 · 10−23 J/K ,
electron mass: me = 9.109 · 10−31 kg ,
proton/neutron mass: mp/n = 1.675 · 10−27 kg ,
mass of an α particle: mα = 6.645 · 10−27 kg ,
electron volt: 1ev = 1.6022 · 10−19 J (1 J = 6.241 · 1018 eV) .

• group vs phase velocity: given the dispersion relation ω(k) (i.e. the functional
dependence of ω on the wave vector) we define the phase and group velocities

ω(k) dω(k)
vp = , vg = (3)
k dk
Wave packets made up from waves with this dispersion relation move as a rigid
object if vp = vg . Else, they deform (and disperse) over time.

PAGE 1 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Basic physics notions that are useful for the exercises

• Schrödinger equation: Let Ĥ be the Hamilton operator and ψ(~x, t) the (time-
dependent) wave function in position space (with ~x ∈ Rd ). ψ satisfies the following
time-dependent Schrödinger equation
∂
i~ ψ(~x, t) = Ĥ ψ(~x, t) .
∂t
iEt
If Ĥ is time-independent, we can write the wave-function as ψ(~r, t) = φ(~r) e− ~ ,
where E corresponds to the energy associated with the wave function and the spatial
wave-function satisfies

Ĥ φ(~r) = E φ(~r) . (4)

For a massive particle (mass m) moving under the influence of a potential V (~x), the
~2 ~ 2
Hamiltonian (in position space) takes the form Ĥ = − 2m ∇ + V (~x).

Exercise 1: It takes all kinds of photons!

Preparatory Questions:
In this exercise we consider first photons (questions 2) and 3)) and then other particles of different
energies.

• Recall the units of mass, energy, wavelength and frequency in the SI system.

• Write the numerical value of Planck’s constant h together with its correct unit. How is ~
defined in terms of h?

• Recall the definition of an electron volt (eV). What is its numerical value in SI units?

• Recall the mass of a proton, an electron and a photon in kg.

Solution of the Exercise:

1) Find the energy (in eV) of the photons corresponding to:

(a) a He-Ne laser, λ = 632.8 nm; (b) a CO2 laser, λ = 10.6 µm; (c) a Nd/YAG laser, main
line λ = 1.064 µm; (d) a radio station operating at frequency f = 92.2 MHz; (e) a wi-fi access
point, f = 2.4 GHz.

Hints: Express the energy of photons in terms of their wavelength or frequency. Remember
the dispersion relation for (massless) photons.

PAGE 2 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Detailed Solution: The energy E of photons of frequency f (angular frequency ω = 2πf )

and wavelength λ is given by
2π~c hc
E = ~ω = 2π~f = = , (5)
λ λ
where f and λ are related by
ω c
f= = .
2π λ
Numerically (with the fundamental constants given on p. 1), we find for the different photons

λ [m] f [s−1 ] E [J] E [ev]

a) He-Ne laser 6.328 · 10−7 4.738 · 1014 3.14 · 10−19 1.959
−5 13 −20
b) CO2 laser 1.06 · 10 2.828 · 10 1.874 · 10 0.117
c) Nd/YAG laser 1.064 · 10−6 2.8176 · 1014 1.867 · 10−19 1.165
d) radio station 3.252 9.22 · 107 6.1094 · 10−26 3.8131 · 10−7
e) wi-fi access point 0.125 2.4 · 109 1.5903 · 10−24 9.92572 · 10−6

2) With the help of the diagram below, classify electromagnetic radiation at energies E =
10x eV, where x = −9, −4, −2, 0, 4, 6.

Hints: Compute the wavelength for the photons as a function of their energy

Detailed Solution: For photons, the wavelength as a function of the energy is given by
hc
λ= ,
E
We therefore have for energies E = 10x eV

x wavelength [m] type

−9 1239.81 radio waves
−4 1.23981 · 10−2 micro waves
−2 1.23981 · 10−4 IR light
0 1.23981 · 10−6 near IR light
4 1.23981 · 10−10 X-rays
6 1.23981 · 10−12 γ-rays

PAGE 3 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

3) Find the de Broglie wavelength of the following objects:

(a) an electron with kinetic energy K = 1 eV; (b) an electron with K = 100 eV; (c) an
α particle with K = 100 eV; (d) a thermal neutron at T = 300 K; (e) a golf ball (m = 200 g,
v = 25 ms−1 ).

Hints: For an object of momentum p, recall the de Broglie wavelength as a function of the
momentum. Find the energy for the different particles .

Detailed Solution: For an object of momentum p, the de Broglie wavelength is λ = h/p.

p2
For objects of mass m, the kinetic energy in terms of the momentum p is given by K = 2m such
that the de Broglie wavelength is
h
λ= √
2mK
Numerically, we therefore find the following de Broglie wave length

energy [eV] mass [kg] de Broglie wavelength [m]

a) 1 9.109 · 10−31 1.22643 · 10−9
b) 100 9.109 · 10−31 1.22643 · 10−10
c) 100 6.645 · 10−27 1.436 · 10−12

The thermal energy of a particle of temperature T (in 3 dim.) is

3
Etherm = kB · T
2
such that

energy [eV] mass [kg] de Broglie wavelength [m]

d) 3.8776 · 10−2 1.675 · 10−27 1.45241 · 10−10

The kinetic energy of a particle with mass m and velocity ~v is

m~v 2
Ekin =
2
such that

energy [eV] mass [kg] de Broglie wavelength [m]

e) 3.9009 · 1020 0.2 1.3252 · 10−34

PAGE 4 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Exercise 2: Classical or quantum?

Preparatory Questions:
This exercise helps to develop an intuitive in which cases quantum physics needs to be applied and
in which situations classical physics is sufficient.

• Explain the main differences between classical and quantum physics.

• Recall the numerical value of Planck’s constant.

Solution of the Exercise:

1) To decide if a given system can be described by classical physics or if one has to resort
to a quantum approach, it is often useful to study the system’s properties in terms of a
characteristic quantity A that is of type “action” and which has to be compared to the

~. This action A can be (energy · time), (momentum · length) or (energy ·
universal constant
1/2
mass) · length .
Two
( cases can thus be distinguished:
A  ~ ⇒ classical theory
A ≈ ~ ⇒ quantum mechanics

1) Show that nuclear physics (mp/n = 1.67 · 10−27 kg, r0 = 10−15 m, E ≈ 1 MeV) and atomic
physics (me = 9.1 · 10−31 kg, a0 = 0.5 Å, E = 13.6 eV) belong without doubt to the quantum
regime. What can we say about a red blood cell (m ≈ 10−14 kg, v ≈ 10−3 m/s) moving in a
capillary (d ≈ 10 µm)? What about a runner during a 100-meter sprint?

Hints: Verify
that an ”action” A of the type (energy·time), (momentum·length) or (energy·

mass)1/2 · length is indeed comparable to ~. Compute such an action for the different situations.

Detailed Solution: The following actions indeed have the dimensions of (energy· time)
kg · m kg · m2
[(momentum · length)] = ·m= = [(energy · time)]
s s
1/2
kg · m2 kg · m2

1/2 1/2
 
(energy · mass) · length = (J · kg) · m = · kg · m = = [(energy · time)]
s2 s
Thus they are comparable to ~. We can therefore compute the following characteristic quantities
for the various situations (recall ~ = h/(2π) = 1.0546 · 10−34 Js)

• nuclear physics: A = E · mp r0 = 1.63575 · 10−35 Js ' ~. Therefore nuclear physics belongs

p

to the realm of quantum physics.

√
• atomic physics: A = E · me a0 = 7.04 · 10−35 Js ' ~ (with 1 Å=10−10 m). Therefore, atomic
physics belongs to the realm of quantum physics.

PAGE 5 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

q
• red blood cell: A = m· mv 2
2
d = 7.07 · 10−23 Js  ~. Therefore a red blood cell is described
by classical physics.
q
• For a runner of m = 100kg and a speed of v = 10m/s we have A = m · mv2 100m =
2

7.07 · 104 Js  ~. Therefore a runner is described by classical physics.

Note: Using the length scale of 1 − 2m (height of a human) instead of 100m does not change
the conclusion

2) Estimate the de Broglie wavelength associated with an electron of an atom and with the
red blood cell in the capillary, then comment the results.

Hints: Recall the definition of the de Broglie wavelength

Detailed Solution: The de Broglie wave length for a massive object with momentum p and
kinetic energy Ekin is
h h
λ= =√
p 2mEkin

• electron (kinetic energy Ekin = 13.6 eV)

h
λ= √ = 3.325 · 10−10 m ,
2me Ekin
which is comparable to atomic dimensions. Thus, when describing electrons as waves, we
have to account for diffraction phenomena.

• red blood cell (p = mv)

h
λ= = 6.626 · 10−17 m .
mv
Therefore, when trying to describe a red blood cell using wave mechanics there is no diffrac-
tion. It is therefore sufficient to describe the latter using classical physics.

Exercise 3: The photoelectric effect

Preparatory Questions:
• Recall the shell structure of electrons in atoms.

• Recall the photoelectric effect of photons interacting with the electrons in various shells of
an atom.

• Recall the binding energy of an electron in an orbital of an atom.

PAGE 6 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Solution of the Exercise:

1) The photoelectric effect is the main interaction mechanism between photons of energies
below 0.5 MeV and matter. Suppose that X-ray photons of wavelength 2 Å interact with
a thin sheet of iron by means of the photoelectric effect. To which atomic shells will the
emitted photoelectrons correspond and what will be their respective kinetic energies?
The following binding energies of electrons in iron are given:

electron K LI LII LIII MI MII

energy E [keV] 7.1 0.85 0.72 0.71 0.10 0.05

Hints: Establish an energy balance between the energy of the (incoming) photon Ephot , the
binding energy of the electron E and the kinetic energy of the ejected electron Ekin .

Detailed Solution: The kinetic energy of the ejected electrons Ekin is the difference between
the energy of the incoming photon Ephot and the binding energy E

Ekin = Ephot − E

Since Ekin must be non-negative, for an electron with binding energy E to be emitted, we require

Ephot ≥ E

Photons of wavelength λ = 2 Å have an energy

hc
Ephot = = 9.932 · 10−16 J = 6.2 keV ,
λ
which implies that electrons can be ejected from all shells except for the K-shell. The respective
kinetic energies of the electrons are

shell LI LII LIII MI MII

Ekin [keV] 5.35 5.48 5.49 6.1 6.15

2) The photoelectric effect in metals can arise from the electrons of the conduction
band. The minimal energy required to eject such an electron is called the work function
Ws, which is on the order of a few eV for the kinds of metals that are used in photocathodes:

cesium potassium aluminum copper tungsten platinum

Ws [eV] 1.8 2.2 3.0 4.1 4.5 5.4

a) Determine the threshold wavelength for the onset of the photoelectric effect in each of
the above cases.

PAGE 7 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Hints: Find the limiting case of the energy balance between the photon energy and the work
function.
Detailed Solution: The threshold is determined by setting the kinetic energy of the ejected
electron to zero

0 = Ekin = Ephot − Ws =⇒ Ephot = Ws

This implies for the threshold wavelength

hc
λs =
Ws
which is numerically for the various elements

Ws cesium potassium aluminum copper tungsten platinum

λs [nm] 688.78 563.55 413.27 302.39 275.51 229.6

b) Let the photocathode be irradiated with light on the low-wavelength limit of the visible
spectrum, λ = 3500 Å. Calculate the kinetic energy of the photoelectrons produced under
these conditions from each of the metals.

Hints: Use the energy balance relation to compute the kinetic energy. What does a negative
kinetic energy imply?

Detailed Solution: The kinetic energy of the electrons Ekin and the work function satisfy
the following balance relation
hc
Ekin = − Ws ,
λ
Numerically, we find the following kinetic energies for the various elements

Ws cesium potassium aluminum copper tungsten platinum

Ekin [ev] 1.74 1.34 0.54 −0.558 −0.958 −1.858

The negative kinetic energies for copper, tungsten and platinum imply that no photoelectric effect
takes place and no electrons are ejected.

Exercise 4: Gaussian wave packets

Preparatory Questions:
• Why do we consider wave packets in quantum mechanics rather than individual (plane)
waves?

PAGE 8 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

• Recall the dispersion relation of (massless) photons and massive particles.

• Recall the difference between phase and group velocity.

• Compute the Gaussian integral R dx e−x .

R 2

• Formulate Heisenberg’s uncertainty principle.

Solution of the Exercise:

We consider a wave packet in one dimension, Ψ(x, t), which is formed by superposition of
monochromatic waves. Let ω(k) denote the angular frequency of one component of the
packet and A(k) be a Gaussian distribution centered around k0 :
Z +∞
1
Ψ(x, t) = √ A(k) ei[kx−ω(k)t] dk
2π −∞

where (a is a real positive constant)

1/4
a2

2 (k−k )2 /4
A(k) = e−a 0
,
2π

In this exercise we will study the temporal evolution of such a Gaussian wave packet.
1) First of all we will consider the wave packet associated with a photon:
Write down the expression for the angular frequency ω as a function of k (the dispersion
relation) for a photon in vacuum. Determine the phase and group velocities and show that
the associated wave packet propagates like a rigid object.

For a (massless) photon, the wave number k is related to the wavelength k and the angular
frequency ω through
2π 2πf ω
k= = = =⇒ ω = ck.
λ c c
The phase velocity vp and the group velocity vg are

ω dω
vp = = c, vg = = c,
k dk
Since vp = vg the wave packet propagates as a rigid object.

2) Now we are interested in the wave packet associated with a free “massive” particle:
Demonstrate that a free, non-relativistic particle of mass m obeys the following dispersion
relation: ω(k) = ~k 2 /2m. Calculate the phase velocity and show that the associated wave
packet will deform (broaden) with time. Calculate the group velocity and verify that it is
identical to what is expected from classical mechanics.

PAGE 9 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

For a non-relativistic particle of mass m and momentum p we have for the kinetic energy Ekin and
the de Broglie wavelength

p2 h
Ekin = = ~ω, and λ= . (6)
2m p

We can associate a wave number k = 2π λ

with the de Broglie wave length, such that we obtain
p = ~ k. Together with the expression for the kinetic energy (6) we therefore find

p2 ~k 2
ω(k) = = .
2m~ 2m
The phase velocity vp and group velocity vg are

ω ~k p dω ~k p
vp = = = , and vg = = = .
k 2m 2m dk m m
Notice that p = m vg in accordance with classical mechanics. Furthermore, since vg 6= vp , the wave
packet will be deformed as it propagates in time.

3) We will now look at the magnitude of the wave-packet-broadening effect. First of all,
show that  1/4
2 2 2
Ψ(x, 0) = 2
eik0 x e−x /a
πa
2
and draw the curve for Ψ(x, 0) .

At t = 0 we have (with a > 0)

1/4
a2
Z 
1 2 (k−k )2 /4+ikx
Ψ(x, t = 0) = √ e−a 0
dk
2π R 2π

This integral is of Gaussian type and can be computed by completing the exponent into a square:
2
a2 a2 a2 k0 + 2ix x2 − ixk0 a2

2
− (k − k0 ) + ikx = − k− −
4 4 a2 a2
 
a a2 k0 +2ix
together with the change of coordinates z = 2 k − a2 (with dk = a2 dz) such that
1/4  Z ∞− ix
a2
 2
x − ixk0 a2

2 a 2
Ψ(x, 0) = √ exp − 2
e−z dz .
a 2π 2π a −∞− ix
a

ix
The integral is parallel to the imaginary axis, however, at imaginary part a
. It is of Gaussian type
and can therefore be computed directly
1/4 ∞
a2 x2 − ixk0 a2
  Z
2 2
Ψ(x, 0) = √ exp − e−z dz . (7)
a 2π 2π a2 −∞

PAGE 10 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Complex Shift in a Gaussian Integral

The result (7) is obtained using

Z ∞− ix
a
Z ∞
−z 2 2
e dz = e−z dz , (8)
−∞− ix
a
−∞

which can be shown by considering a closed integral in the complex plane (for some R ∈ R+ )
z
−R III R

IV II

I − ix
a

2
Since the integrand e−z has no poles inside the red rectangle, using Cauchy’s theorem, we
have II + III + IIII + IIV = 0 with the integrals
Z R− ix
a
Z 0 Z −R Z −x
a
−z 2 −(R+iz)2 −z 2 2
II = dz e , III = dz e , IIII = dz e , IIV = dz e−(−R+iz) .
−R− ix
a
−x
a
R 0

In the limit R → ∞ we have

Z ∞− ix Z ∞
a 2 2
lim II = dz e−z , lim III = 0 , lim IIII = − dz e−z , lim IIV = 0 ,
R→∞ −∞− ix R→∞ R→∞ −∞ R→∞
a

which implies II = −IIV , thus proving eq. (8).

|
(x,0)|^2
The remaining integral in (7) is of Gaussian
type, for which we have 1.5

Z ∞
2 √
dze−z = π ,
−∞
1.0

Therefore, we finally find

√  2 1/4  2
x − ixk0 a2

2 a
Ψ(x, 0) = exp − 0.5
a 2π a2
 1/4
2 2 2
= 2
eik0 x e−x /a .
πa x
-3 -2 -1 0 1 2 3

The absolute value of the wave function is a=2 a=1 a=1/2

 1/2
2 2 2
2
|Ψ(x, 0)| = 2
e−2x /a , Figure 1: Graph of |Ψ(x, 0)|2 for different values of a.
πa

PAGE 11 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

which is drawn in Figure 1 for different values of a. The latter determines the width (see next
question for the precise definition) as well as the maximum of the function.

2 2
4) The width of a Gaussian e−y /b is defined as the distance√between the position of the
maximum and the point at which the amplitude is divided by e . Derive an expression for
2 2
the width of a generic Gaussian. Calculate the widths ∆x and ∆k of Ψ(x, 0) and A(k) ,
respectively, and show that ∆x ∆k = 1/2.

2 /b2
We consider the function f (y) = e−y which has a maximum at y = 0:

f (y)
•
1

√1 •
e

y
∆y

Therefore, the width ∆y is determined by

∆y 2 1 √
− 2
=− , =⇒ ∆y = b/ 2 .
b 2
The width of |Ψ(x, 0)|2 and |A(k)|2 respectively are
√
a 2
∆x = , ∆k = √ = 1/a ,
2 a 2
such that ∆x · ∆k = 21 .

5) A lengthy calculation yields:

s
−2a2 (x − ~k0 t/m)2
 
2 2
Ψ(x, t) = exp
πa2 (1 + 4~2 t2 /(m2 a4 )) a4 + 4~2 t2 /m2

a) What is the velocity of the displacement of the maximum?

Starting from
s
−2a2 (x − ~k0 t/m)2
 
2 2
Ψ(x, t) = exp
πa2 (1 + 4~2 t2 /(m2 a4 )) a4 + 4~2 t2 /m2

PAGE 12 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

2 ~k0
For given time t, the maximum of Ψ(x, t) is at x(t) = m
t, which moves at a velocity of
dx(t) ~k0
= = vg ,
dt m
which is the group velocity.

5b) Confirm that there is broadening of the wave packet. Calculate the time it takes for
the wave packet to broaden to twice its initial (t = 0) width:
• for a particle with a mass of 1 g and an initial spread of 2 mm,
• for an electron starting from an initial spread of 1 Å. What is your conclusion?

2
The width of the Gaussian Ψ(x, t) is
s
2 2
r
1 a4 + 4~m2t a 4~2 t2
∆x = √ = 1 + ,
2 2a2 2 m 2 a4
which is a monotonic growing function in t: indeed for (a = m = 1, k0 = 2) the evolution of the
wave packet is plotted below at different times t
|
(x,t)|^2
0.8

0.6 t=0
1
t=
2

0.4
t=1
t= 32

0.2
t=2

x
-2 0 2 4 6 8

From this it beomces clear that the maximum of the packet moves to the right for gowing time t,
but also the width of the packet increases. The time t2 it takes to double the width is
r r √
4~2 t22 3m2 a4 3ma2
2= 1+ 2 4 =⇒ t2 = =
ma 4~2 2~
a
In terms of the initial width ∆x0 = 2 the latter becomes
√
2 3m
t2 = (∆x0 )2 .
~
Numerically we find
• particle of mass 1 g: t2 = 1.314 · 1026 s ,
• electron: t2 = 2.992 · 10−16 s ,
Thus the spread on macroscopic objects is negligible, while the spread on microscopic objects is
relevant.

PAGE 13 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Exercise 5: A square potential well of infinite depth

Preparatory Questions:
• Write the (classical) Lagrangian for a massice particle moving in one dimension under the
influence of a potential V . derive the corresponding (classical) Hamiltonian.
• Recall the eigenvalue equation of an operator O acting on a Hilbert space H.
• Explain how to pass from the time-dependenc Schrödinger equation to the time-independent
one for a massive particle propagating in a one-dimensional potential V (x) that is indepen-
dent of time.

Solution of the Exercise:

A particle of mass m is trapped in medium of thickness a. To model this phenomenon, we

suppose that the potential energy of the particle can be represented by an infinite potential
well given by (
0 if 0 ≤ x ≤ a
V (x) = .
∞ if x < 0 or x > a
1) Let φ(x) be the spatial wave function associated with this particle of total energy E.
Write down the time-independent Schrödinger equation that φ(x) has to satisfy.

Hints: Consider the problem in the regions 0 ≤ x ≤ a, x < 0 and x > a separately.

Detailed Solution: Let Ĥ be the Hamilton operator and φ(x) the (time-independent) spatial
wave function with energy eigenvalue E. φ satisfies the following Schrödinger equation
Ĥ φ(x) = E φ(x) .
For a massive particle moving in one dimension under the influence of a potential V (x) this equation
becomes
~2 d2
 
− + V (x) φ(x) = E φ(x) . (9)
2m dx2
Considering the particular potential (with a > 0)

V (x)


 ∞ ... x < 0
V (x) = 0 ... 0 ≤ x ≤ a
∞ ... x > a


x
0 a

PAGE 14 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

the equation (9) implies

~2 d2
− φ(x) = E φ(x) , for 0 ≤ x ≤ a,
2m dx2

φ(x) = 0 for x ≤ 0 or x ≥ a .

2) State the boundary conditions that are imposed by the potential well on the general
solution of the Schrödinger equation. These conditions imply that the probability is zero for
finding the particles outside of the well (or at the two limits x = 0 and x = a). Use these
conditions to derive expressions for the (non-normalized) spatial wave functions φn (x) and
for the corresponding characteristic Energies En . Show that En = n2 · E0 and provide the
formula for E0 .

Hints: Solve the Schrödinger equation in the region 0 ≤ x ≤ a, and impose the condition
that φ(x) is a continuous function at x = 0 and x = a.

Detailed Solution: For 0 ≤ x ≤ a, we are searching for the general solution of the differential
equation

~2 d2
− φ(x) = E φ(x) ∀0 ≤ x ≤ a,
2m dx2

with φ(x = 0) = 0 = φ(x = a) , (10)

for 0 ≤ x ≤ a, we need to distinguish E < 0, E = 0 and E > 0:

• E < 0: For E = −|E| < 0, the generic solution of the differential equation (10) is

α, β ∈
√
C,
φ(x) = α sinh kx + β cosh kx , with −2mE
k= ~
∈ R+ .

Imposing the condition φ(x = 0) = 0 implies β = 0. Imposing φ(x = a) = 0 we have

φ(x = a) = 0 = α sinh(ka) =⇒ α = 0,

which means that the only solution with E < 0 is the trivial solution φ(x) = 0, which we are
not interested in.

• E = 0: For E = 0, the generic solution of the differential equation (10) is

~2 d2
− φ(x) =⇒ φ(x) = α x + β ,
2m dx2
with α, β ∈ C. Imposing the boundary conditions (10) we have α = β = 0, which leaves only
the trivial solution φ(x) = 0, which we are not interested in.

PAGE 15 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

• E > 0: For E > 0, the generic solution of the differential equation (10) is

α, β ∈
√
C,
φ(x) = α sin kx + β cos kx , with 2mE
k = ~ ∈ R+ .

The boundary conditions (10) impose

0 = φ(x = 0) = β , and 0 = φ(x = a) = α sin ka ,

which has the following non-trivial solutions

ka = nπ , for n ∈ Z.

n2 π 2 ~2
We thus find the (quantised) energies En = 2ma2
for n ∈ Z

To summarise, the (only non-trivial) unnormalised, linear independent wave-functions are1

αn sin nπx


, for 0 ≤ x ≤ a ,
φn (x) a , ∀n ∈ N∗
0 for x ≤ 0 or x ≥ a ,

with the (positive) energies

n2 π 2 ~2
En = > 0.
2ma2
The energies indeed take the form

π 2 ~2
E n = E 0 n2 , with E0 = .
2ma2
which is numerically

E0 = 6.025 · 10−20 J = 0.376 eV

3) Draw the form of the wave function for the ground state and the first excited state
(which is to be defined first), and calculate the corresponding energy for the case of an
electron (me = 9.1 · 10−31 kg, a = 1 nm), expressed in eV.

Hints: Normalise the wave function first. How many zeroes does φn have in the interval [0, a]
for n = 1, 2?

Detailed Solution: We fist normalise the wave functions (for n ∈ N, i.e. n > 0) by imposing
Z ∞ Z a  nπx 
2 2
1= dx |φn (x)| = |αn | sin2 dx
−∞ 0 a
1
Notice that it is sufficient to consider n ∈ N∗ , since φn and φ−n have the same energy eigenvalues and are not
linear independent.

PAGE 16 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Using the trigonometric identity

sin2 u = 12 (1 − cos(2u)) ∀u ∈ R ,

we find
a
|αn |2 a |αn |2
Z   
2nπx
1= 1 − cos dx = .
2 0 a 2

We therefore have up to an undetermined phase θ

r
2 iθ
αn = e , with θ ∈ [0, 2π) .
a
Since physical observables do not depend on θ, we can choose θ = 0 without loss of generality,
such that
r
2
αn =
a
The normalised wave functions for the ground state n = 1 (first non-trivial wave function) and the
first excited state n = 2 take the form (for concreteness, x is plotted in units of nm)

E1 = 6.03 · 10−20 J = 0.376 eV, E2 = 2.41 · 10−19 J = 1.504 eV.

4) For the first excited state, find the probability of observing the particle between 0 et a/2
using an intuitive argument, then verify your intuition by explicit calculation.

Hints: Use the symmetries of the wave function to find an intuition for the probability.

Detailed Solution: Since the first excited state is anti-symmetric with respect to the point
a/2 (the center of the interval [0, a]) and the total probability to find the particle between 0 ≤ x ≤ a
is 1, we expect

P (0 ≤ x ≤ a/2)n=2 = 12 .

PAGE 17 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

In fact, we can verify this intuition for generic n ∈ N

Z a/2
2 a/2
Z  nπx 
2
P (0 ≤ x ≤ a/2)n = dx |φn (x)| = dx sin2
0 a 0 a
Using the trigonometric identity
sin2 u = 21 (1 − cos(2u)) ∀u ∈ R ,
we find
Z a/2   
1 2nπx 1
P (0 ≤ x ≤ a/2)n = dx 1 − cos = .
a 0 a 2

Exercise 6: Finite and infinite potential wells: application

to the deuton
Preparatory Questions:
• Recall the definition of the laboratory frame and the center of mass frame.
• Write the Hamiltonian for a particle of mass m moving in a generic potential V (x).
• Plot the curve tan(x) for x ∈ R.

Solution of the Exercise:

Of all conceivable bound states between nucleons, only the combination proton-neutron is
realized in nature: the deuton (or deuteron). The goal of this exercise is to obtain an
order-of-magnitude estimate for the energy of this state, based on an extremely simplified
model.
We start by assuming that both neutron and proton are restricted to movement along the
0x-axis. Furthermore, we only take into account the central part of the interaction potential,
which we model according to:

 ∞ for x < 0
U (x) = −U0 for 0 ≤ x ≤ R where x = x2 − x1
0 for R < x


1) Write down the total classical energy, Elab , of the proton-neutron system in the laboratory
frame as a function of the positions of the two particles, x1 and x2 , and their temporal
derivatives, ẋ1 and ẋ2 . Calculate the energy in the center-of-mass system as a function of x
and of the reduced mass µ (to be defined as is appropriate).

Hints: Write the energy in the laboratory system as the sum of the kinetic energy of the two
particles and the potential U . Re-express the energy in terms of the center of mass coordinates
and interpret the various terms.

PAGE 18 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Detailed Solution: Classically, the total energy in the lab is the sum of the kinetic energies
of the two particles and the potential energy. Calling the masses of the two particles m1 and m2
respectively2 , we have
1 1
Elab = m1 ẋ21 + m2 ẋ22 + U (x2 − x1 ) . (11)
2 2
The center of mass position is given by
m1 x1 + m2 x2
xc = .
m1 + m2
Defining furthermore

M = m1 + m2 , and x = x 2 − x1 ,

we can express x1,2 in terms of xc and x

m2  m2  m1
x1 = xc − x, x 2 = xc + 1 − x = xc + x.
M M M
Inserting into Elab in eq. (11) we obtain
1 m1 m2 2
Elab = M ẋ2c + ẋ + U (x) .
2
| {z } 2M
kin. energy CM

Here the first term corresponds to the kinetic energy of the center of mass (i.e. the energy of the
’synchronised’ motion of both particles moving together), whereas the remaining terms are the
energy of the relative motion of the two particles. The latter therefore correspond to the energy
in the center of mass frame:
µ m1 m2
ECM = ẋ2 + U (x) , with µ= ,
2 M
where µ is the reduced mass.

2) Find the expression for the Hamiltonian Ĥ of the deuton in the center-of-mass system.

Hints: Use the energy ECM in the center of mass system found in the previous question.

Detailed Solution: Changing to the (classical) Hamiltonian description, the momentum of

a massive particle (mass µ) moving in one dimension is

p = µ ẋ ,

such that the (classical) Hamiltonian in the center of mass system becomes
p2
H= + U (x) .
2µ
2
In the following we shall work with two arbitrary masses m1,2 and impose m1 = m2 = mp = mn at the end.

PAGE 19 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

The Hamiltonian of the quantised system is obtained by promoting p and x to operators

1 2
Ĥ = p̂ + U (x̂)
2µ
d
In position space, we have p̂ = −i~ dx , such that

~2 d2
Ĥ = − + U (x) , (12)
2µ dx2

3) We are interested in the bound stationary states (E < 0) given by the wave function
Ψ(x) of the Hamiltonian Ĥ. Write down the differential equation that Ψ has to obey and
establish the general form of the solution in each of the three domains. Use the following
notation: r r
2µ(U0 + E) −2µE
k= 2
and ρ=
~ ~2

Hints: Solve the problem in three different regions x < 0, 0 ≤ x ≤ R and x > R.

Detailed Solution: We recall the form of the potential U (with R > 0 and U0 ∈ R+ )

U (x)


 ∞ ... x < 0
U (x) = −U0 . . . 0 ≤ x ≤ R
0 ... x > a R

0
x

−U0

The potential is decomposed into three different regions (x < 0, 0 ≤ x ≤ R and x > R). We
introduce the following notation

 ψ1 (x) for x < 0 ,
Ψ(x) = ψ2 (x) for 0 ≤ x ≤ R , (13)

ψ3 (x) for R ≤ x .

We can adapt (12) for each of the ψ1,2,3 individually, which leads to the following overview

PAGE 20 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

region 1 region 2 region 3

range of x x<0 0≤x≤R x>R
potential U (x) ∞ −U0 < 0 0
2 2
Schrödinger eq. ψ1 = 0 − 2µ
~
ψ200 = (E + U0 ) ψ2 − 2µ
~
ψ300 = E ψ3

gen. solution ψ1 = 0 ψ2 = A2 eik x + B2 e−ik x ψ3 = A3 eρ x + B3 e−ρ x

q q
wave numbers — k = 2µ(E+U ~2
0)
∈R ρ = − 2µ~2E ∈ R

4) State the boundary conditions and show that they impose a quantization condition on
the energy, which can be expressed as
(
sin(kR) = kkR0R
p
with k0 = 2µU0 /~2
tan(kR) < 0

Hints: Impose conditions at x = 0, x = R and limx→∞ . Take into account whether the
potential U (x) has a finite or an infinite jump at this position at x = 0 or x = R and consider
which asymptotic value the wave function has to have in order to be L2 -normaliseable.

Detailed Solution: In order to impose boundary conditions, we have to consider the change
of the potential U (x) at x = 0 and x = R as well as the asymptotic value for limx→∞ :

• at x = 0 the potential U (x) has an infinite discontinuity (jump). We therefore have to impose
that the wave function Ψ is continuous:

ψ1 (x = 0) = ψ2 (x = 0) . (14)

• at x = R the potential U (x) has a finite discontinuity (jump). We therefore have to impose
that the wave function Ψ and its first derivative are continuous:

ψ2 (x = R) = ψ3 (x = R) , and ψ20 (x = R) = ψ30 (x = R) , (15)

0 dψ2,3
where ψ2,3 (x) = dx
(x) .

• The wave function Ψ(x) needs to be L2 -normaliseable, which requires

lim ψ3 (x) = 0 . (16)

x→∞

Since ρ > 0, the condition (16) implies

lim A3 eρx + B3 e−ρx = 0

=⇒ A3 = 0 ,
x→∞

PAGE 21 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

while the condition (14) implies

0 = ψ2 (x = 0) = A2 + B2 =⇒ B2 = −A2 .

Therefore we have for ψ2

eikx − e−ikx
ψ2 (x) = 2i A2 = 2i A2 sin(kx) ,
2i
ψ20 (x) = 2i k A2 cos(kx) ,

Therefore, the conditions (15) become

2i A2 sin(kR) = B3 e−ρR , and 2i k A2 cos(kR) = −ρ B3 e−ρR .

Assuming A2 6= 0 (in order to have a non-vanishing wave-function), we can insert the first equation
into the second one to find
k
sin(kR) = − cos(kR) . (17)
ρ
For cos(kR) 6= 0,3 this relation can be written as
k
tan(kR) = − < 0. (18)
ρ
The square of this relation reads
k2
sin2 (kR) = (1 − sin2 (kR))
ρ2
which implies
k2 E + U0
sin2 (kR) = 2 2
= , (19)
k +ρ U0
which is equivalent to
r r
E + U0 k 2µU0
| sin(kR)| = = , with k0 = . (20)
U0 k0 ~2

5)Experimentally, one finds that the deuton has a single bound state with an energy of
E = −2.2 MeV. Which condition does U0 R2 have to satisfy?

Hints: Plot graphically the functions | sin(kR)|, tan(kR) and k/k0 as functions of k. Find
the the possible solutions compatible with (18) and (20) and impose conditions that the former is
unique.

Detailed Solution: For fixed k0 and R we have | sin(kR)|, tan(kR) and k/k0 as functions of
k schematically as follows
3 (2n+1)π
The condition cos(kR) = 0 requires kR = 2 for which sin(kR) 6= 0, such that this case can be ignored.

PAGE 22 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

| sin(kR)|
tan(kR)
k/k0
1 P3
P2 •
P1 •
•
π π 3π 2π 5π 3π
k
2R R 2R R 2R R

Due to the condition tan(kR) < 0, only the green shaded region is admissible: within this region,
the points P1,2,3 are solutions to the conditon | sin(kR)| = kk0 .
For the deuton, only one intersection is allowed, which implies a condition on k0 :

extremal curve 1 extremal curve 2

π π 3π 2π 5π 3π
k
2R R 2R R 2R R

If the slope is steeper or shallower than the two lines shown, we either have no (admissible)
intersection, or two of them. The condition for these two slopes is
π
• curve 1: 1= ,
2Rk0
3π
• curve 2: 1= ,
2Rk0
They thus imply
π 3π π2 9π 2
< k0 R < =⇒ < k02 R2 <
2 2 4 4
q
For U0 R this implies (k0 = 2µU
2
~2
0
)

π 2 ~2 9π 2 ~2
< U0 R2 < .
8µ 8µ

PAGE 23 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

mp
Using m1 = m2 = mp , such that µ = 2
= 8.375 · 10−28 kg we have

eV2 s2 eV2 s2
1.02 · 10−42 < U0 R2 < 9.2 · 10−42
kg kg
For a typical radius of R = 10−15 m we have
100 MeV < U0 < 900 MeV .

Exercise 7: The Dirac well

Preparatory Questions:
• Recall the definition of a distribution and their derivatives.
• Recall the definition of the Heaviside theta function as well as its derivative.
• Let f : R → R a function and x, x0 ∈ R. Recall and justify the relations
Z
δ(x − x0 ) f (x) = δ(x − x0 ) f (x0 ) , and dx δ(x − x0 ) f (x) = f (x0 ) . (21)
R

Solution of the Exercise:

Suppose that a particle of mass m is subjected to the one-dimensional potential V (x) =

α δ(x), where α is a negative constant and δ(x) is the Dirac distribution defined by:
(
∞ if x = 0
δ(x) =
0 if x 6= 0

In this exercise we will consider bound states (E < 0) of such a particle.

1) Write down the Schrödinger equation for the system with the above potential V (x) and
explain which physical dimensions (units) the parameter α has to have.

Hints: Write the general form of the time-independent Schrödinger equation and specify it
for the potential at hand.

Detailed Solution: The (general) time-independent Schrödinger equation reads

Ĥ ψ(x) = E ψ(x)

where Ĥ is the Hamiltonian and ψ(x) the wave function (in position space) of energy E. For the
potential V (x) = α δ(x), it specifically takes the form
~2 d2
− ψ(x) + α δ(x) ψ(x) = E ψ(x) , (22)
2m dx2
Since δ(x) has unit [length−1 ], the parameter α has the unit of [energy× length].

PAGE 24 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

2) The solutions of the Schrödinger equation can be written as (no demonstration required):
(
Ae−κx x > 0 ~2 κ2
ψ(x) = with E = −
Beκx x<0 2m

a) Assuming that the wave function is continuous, derive the corresponding continuity con-
dition at x = 0.

Hints: Calculate the limits limx→0± ψ(x).

Detailed Solution: The solution can be motivated as follows: For x 6= 0, the potential has
no support, thus the Schrödinger equation takes the form
~2 d2
− ψ(x) = E ψ(x) , for x 6= 0
2m dx2
which has the following solution
(
A e−κx + A0 eκx for x > 0 ~2 κ2
ψ(x) = with E=− .
B 0 e−κx + B eκx for x < 0 2m

For E < 0 (i.e. κ ∈ R) we impose

lim ψ(x) = 0 =⇒ A0 = 0 = B 0 ,
x→±∞

such that
(
A e−κx for x > 0 ~2 κ2
ψ(x) = with E=− .
B eκx for x < 0 2m

Continuity at x = 0 imposes

lim ψ(x) = A = B = lim− ψ(x) . (23)

x→0+ x→0

Notice: This implies that the wave function can be written as

ψ(x) = A e−κ|x| .

2b) The first derivative of the wave function, however, is not continuous. Show that this
discontinuity can be expressed as:
2mα
ψ 0 (0+ ) − ψ 0 (0− ) = ψ(0)
~2
To obtain this relationship, integrate the Schrödinger equation between − and +, then let
 tend to 0 R
+
Reminder: − δ(x)ψ(x)dx = ψ(0).

PAGE 25 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Hints: Integrate the Schrödinger equation over the interval [−, ] and take the limit  → 0.
Use the continuity of ψ at x = 0.

Detailed Solution: We start from the Schrödinger equation

~2 d2
− ψ(x) + α δ(x) ψ(x) = E ψ(x) , (24)
2m dx2
and integrate it over the interval [−, ] (with  ∈ R and  > 0)
Z  Z  Z 
~2 00
− dx ψ (x) = E dx ψ(x) − α dx ψ(x) δ(x)
2m − − −
Z 
~2 0 0
− [ψ () − ψ (−)] = E dx ψ(x) − α ψ(0) . (25)
2m −

Since ψ(x) is a continuous function, we have

Z 
lim dx ψ(x) −→ 0 .
→0 −

The equation (25) therefore becomes in the limit  → 0

2mα
lim [ψ 0 () − ψ 0 (−)] = ψ(0)
→0 ~2
2mα
ψ 0 (0+ ) − ψ 0 (0− ) = 2 ψ(0) . (26)
~
We therefore find a (finite) discontinuity of the first derivative of the wave function at x = 0.

3) Based on what has been established so far, determine the energy E and the wave function
ψ(x) of the bound state as a function of m, α and ~; show that the normalized wave function
has the form ψ(x) = √1l exp(−|x|/l). Find the expression for l and comment on the general
shape of ψ(x).

Hints: Calculate the derivative of ψ (in the sense of a distribution) and use the relation for
the discontinuity at x = 0.

Detailed Solution: We write the wave function in the form

ψ(x) = A θ(x) e−κx + θ(−x) eκx ,

 

where θ is the Heaviside theta-function


1 if x ≥ 0 ,
θ(x) =
0 if x < 0 .

Next, we use θ0 (x) = δ(x), as well as

δ(x − x0 ) f (x) = δ(x − x0 ) f (x0 ) ,

PAGE 26 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

for any function f : R → R (along with δ(−x) = δ(x)). The derivative of ψ(x) is therefore
ψ 0 (x) = −Aκ θ(x) e−κx − θ(−x) eκx .
 

which affords the limits

lim+ ψ 0 (x) = −Aκ lim+ e−κx = −Aκ , and lim− ψ 0 (x) = Aκ lim− eκx = Aκ .
x→0 x→0 x→0 x→0
0
With these relations, the expression for the discontinuity of ψ in eq. (26) becomes explicitly
2mα 2mα
ψ(0) = ψ 0 (0+ ) − ψ 0 (0− ) =⇒ A = −2 Aκ .
~2 ~2
This implies
mα ~2 m2 α2 mα2
κ=− =⇒ E = − = − .
~2 2m ~4 2~2
Furthermore, to normalise the wave function (i.e. to find A), we impose
Z ∞
|A|2
1= dx |ψ(x)|2 = ,
−∞ κ
√
such that we can choose A = κ = √1` and ` = κ1 implying that the (normalised) wave function
can be written as
e−|x|/` mα2
ψ(x) = √ with E=− .
` 2~2
The form of ψ(x) is schematically shown in the following

Alternative Solution: Fourier Transform

This problem can be approached in a different fashion: To solve the Schrödinger equation (22),
we consider a Fourier transform of ψ(x). Indeed, using that ψ ∈ L2 (R) (such that the inversion
theorem is applicable) we have
Z Z
1 ikx b 1
ψ(x) = √ dk e ψ(k) , and δ(x) = dk eikx , (27)
2π R 2π R

PAGE 27 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

we obtain
~2 d2
Z   Z
1 ikx αψ(0)
−√ dx +E e ψ(k)
b =− dk eikx
2π 2m dx2 2π R

which is equivalent to
Z  2 2  
1 k ~ α ψ(0) ikx
−√ dx − + E ψ(k) − √
b e =0 (28)
2π 2m 2π
and therefore imposes
 2 2 
k ~ α
− + E ψ(k)
b = √ ψ(0) . (29)
2m 2π
2 2 α ψ(0) λ~2
Introducing E = − κ2m~ < 0 and √
2π
= 2m
we have

(k 2 + κ2 )~2 b λ~2 λ
ψ(k) = − , =⇒ ψ(k)
b =− . (30)
2m 2m k2 + κ2

Fourier Transformation of e−a|x| :

Consider the function

f : R −→ R
x 7−→ f (x) = e−a|x| (31)

for a ∈ R and a > 0. The Fourier transform fb = F[f ] is given by

2a
fb(k) = √ . (32)
2π(a2 + k 2 )

Comparing with (32) we have

√ √
2π λ 2κ 2π λ
ψ(k) = −
b √ , =⇒ ψ(x) = − exp(−κ|x|) . (33)
2κ 2π(k 2 + κ2 ) 2κ
Using this solution, we can write
√
2πλ 
θ(x) e−κx + θ(−x) eκx ,

ψ(x) = −
√2κ √
0 2πλ  −κx −κx 2π λ 
κx κx
θ(x) e−κx − θ(−x) eκx .
 
ψ (x) = − δ(x) e − κ θ(x) e − δ(x) e + κ θ(−x) e =
2κ 2
where we used δ(x) e−κx = δ(x) = δ(x) eκx . Since θ(x) has only support for x > 0 we have
√
0 + 0 − 2π λ √ 2mα
ψ (0 ) − ψ (0 ) = (1 + 1) = 2π λ = 2 ψ(0) . (34)
2 ~

PAGE 28 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Furthermore, we can write

√
2m α 2π~ mα
ψ(x) = − √ ψ(0) √ e−κ|x| = ψ(0) √ e−κ|x| , (35)
2π~2 2 −2mE ~ −2mE
| {z }
=λ

For x = 0 we have
mα mα2
ψ(x = 0) = √ ψ(0) , =⇒ E=− , (36)
~ −2mE 2~2
q
2πλ2 3
since E < 0. Normalising the wave function we have 1 = 4κ3
(such that λ = − 2κπ ) and therefore
√
ψ(x) = κ e−κ|x| . (37)

which is of the demanded form for κ = 1/`.

Exercise 8: Determining an unknown potential, Part II

Preparatory Questions:
• Recall the Schrödinger equation for a generic potential.

• What is the wave function for an infinite potential?

Solution of the Exercise:

We now consider a particle of mass m in another unknown potential V (x). The wave function
ϕ(x) of this second particle is given by:
(

Ax exp −ax for x ≥ 0
ϕ(x) = , where a is a positive real const.
0 for x < 0

3) Show that the second derivative of the above wave function satisfies (∀x > 0) the following
relationship:
d2 ϕ(x) 
2 a
= ϕ(x) a − 2
dx2 x

Hints: Write the wave function using the Heaviside theta-function and compute its derivatives
(in the sense of a distribution).

Detailed Solution: The wave function can be written in the form

ϕ(x) = A θ(x) x e−ax , with a > 0.

PAGE 29 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

The first derivatives ϕ0 is

ϕ0 (x) = A δ(x) x e−ax +A θ(x) (1 − ax) e−ax ,

| {z }
=0
= A θ(x) (1 − ax) e−ax .

The second derivative ϕ00 is

ϕ00 (x) = A δ(x) (1 − ax) e−ax +A θ(x) (−a − a(1 − ax)) e−ax
| {z }
=δ(x)

= A δ(x) − A a θ(x) (2 − ax) e−ax .

We can therefore write

d2 ϕ(x)
   
2 2a −ax 2 2a
= A δ(x) + a − A θ(x) x e = a − ϕ(x) + A δ(x) .
dx2 x x

Thus for x 6= 0 (where δ(x) has no support) we have

d2 ϕ(x)
 
2 2a
= a − ϕ(x) ∀x ∈ R \ {0} .
dx2 x

4) Supposing that the particle is in an eigenstate of the Hamiltonian H, determine the

potential V (x) and the total energy E of this state.

Hints: Write the Schrödinger equation for x > 0 and separate x-dependent terms multiplying
the wave function from constant ones.
Detailed Solution: The potential is infinite for x ≤ 0. Focusing on the case x > 0 we have
the Schrödinger equation

~2 d2 ϕ(x)
− + V (x) ϕ(x) = E ϕ(x)
2m dx2
~2
 
2a 2
[V (x) − E] ϕ(x) = − − a ϕ(x)
2m x

Since only the potential is x-dependent (while the energy is constant) this implies
 ~2 a 1
− m x + V0 for x > 0 ~2 a2
V (x) = and E=− + V0 .
∞ for x ≤ 0 2m

Here V0 is an arbitrary constant, which can be chosen to be zero.

5) Calculate hV i, the average of the potential energy, as a function of m, ~ and a.

PAGE 30 OF 31
Spring Semester 2020/21 TD Quantum Mechanics, Part 1

Hints: Normalise the wave function before computing the average of the operator V (x).

Detailed Solution: Before computing hV i, we first normalise the wave function

Z ∞ Z ∞
1= 2
dx |ϕ(x)| = |A| 2
dx x2 e−2ax .
−∞ 0

R∞
Integral 0
dx xn e−αx :

For n ∈ N ∪ {0} and α ∈ R with α > 0 we have

Z ∞
n!
dx xn e−αx = n+1 .
0 α
which can be shown by complete induction.

From
Z ∞ Z ∞
1= 2
dx |ϕ(x)| = |A|2
dx x2 e−2ax ,
−∞ 0

we therefore obtain the condition

2 |A|2
1 = |A|2 = =⇒ |A|2 = 4 a3 .
(2a)3 4a3

The average of the potential energy is thus given by

Z ∞
|A|2 ~2 a ∞ ~ 2 a2
Z
hV i = hx|V |xi = 2
dx V (x) |ϕ(x)| = − dx x e−2ax = − .
−∞ m 0 m

6) Based on the value of hV i, deduce the average kinetic energy of the particle in state ϕ(x).

Hints: Use the previous result for the energy E and hV i.

Detailed Solution: The average of the kinetic energy is

~2 a2 ~2 a2 ~2 a2
hT i = E − hV i = − + = .
2m m 2m
Note: We have the virial theorem for stationary states 2hT i = n hV i with n = −1.

PAGE 31 OF 31