Quantum Physics I (8.

04) Spring 2016

Assignment 1
Massachusetts Institute of Technology
Physics Department Due R, February 11, 2016
February 4, 2016 5:00pm

Announcements
• Please put you name and section number at the top of your problem set, and place
it in the 8.05 box labeled with your section number near 8-395 by 5:00pm.

• You may find it fun to read the first few pages of Dirac’s book on Quantum Mechanics.

Problem Set 1
1. Radiative collapse of a classical atom. [10 points]
In a classical universe, we might try to build a hydrogen atom by placing an elec-
tron in a circular orbit around a proton. We know, however, that a non-relativistic,
accelerating electron radiates energy at a rate given by the Larmor formula:

dE 2 e2 a2
=− .
dt 3 c3
Here e is the electron charge and a is the magnitude of the electron acceleration. So
the classical atom may have a stability problem. We want to figure out how big is
this effect. In the units we are working the electron potential energy in the presence
of the proton is V = −e2 /r and the magnitude of the force of attraction is e2 /r 2 .

(a) Show that for a non-relativistic electron the energy ∆E lost per revolution is
small compared to the electron’s kinetic energy K. Do this by computing the
ratio ∆E/K. Hence, it is possible to regard the orbit as circular at any instant,
even though the electron eventually spirals into the proton.
(b) A good estimate for the size of the hydrogen atom is 50 pm (pico-meters), and
a good estimate for the size of the nucleus is 1fm (femto-meter). Compare the
classically calculated velocity of the electron to the velocity of light at an orbital
radius of 50 pm, 1 pm, and 1 fm.
(c) Calculate how long it would take for the electron to spiral from 50 pm to 1pm?
Are you justified in ignoring relativistic corrections? Would the answer using
the non-relativistic approximation change much for a spiral from 50pm to 1fm?
(d) As the electron approaches the proton, what happens to its energy? Is there a
minimum value of the energy the electron can have?

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

2. Quantized energies. [5 points]

Consider an electron in circular motion around a fixed (heavy) proton as a model for
the hydrogen atom. Let V = −e2 /r denote the potential energy of the electron.

(a) Assume a circular orbit and find the relations between the kinetic energy K of
the electron, its potential energy V and the total energy E.
(b) Assume that the magnitude L of the electron angular momentum is quantized
and equal to n~ where n is a positive integer. Find the quantized values En
for the total energy and the associated orbit radii rn . Express your answers in
terms of n, the rest energy Ee = me c2 of the electron, its Comption wavelength
2
λ̄ = m~e c , and the fine structure constant α = e~c .

3. DeBroglie Relations and the Scale of Quantum Effects. [10 points]

(a) Matter Particles as Waves

If a wavelength can be associated with every moving particle, then why are we not
forcibly made aware of this property in our everyday experience? In answering,
calculate the de Broglie wavelength λ = h/p (with h = 6.6 × 10−34 J.s) of each
of the following particles:
i. an automobile of mass 2000 kg traveling at a speed of 50 mph (22m/s)
ii. a marble of mass 10 g moving with a speed of 10 cm/s,
iii. a smoke particle of diameter 100nm and a mass of 1fg being jostled about
by air molecules at room temperature (T = 300K) (assume that the particle
has the same translational kinetic energy as the thermal average of the air
molecules, KE = 32 kB T , with kB = 1.38 × 10−23 J/K)
iv. an 87 Rb atom that has been laser cooled to a temperature of T = 100µK.
Again, assume KE = 32 kB T .
(b) Light Waves as Particles
The Photoelectric effect suggests that light of frequency ν can be regarded as
consisting of photons of energy E = hν, where h = 6.6 × 10−34 Js.
i. Visible light has a wavelength in the range of 400-700 nm. What are the
energy and frequency of a photon of visible light?
ii. The microwave in my kitchen operates at roughly 2.5 GHz at a max power of
300W. How many photons per second can it emit? What about a low-power
laser (10mW at 633 nm), or a cell phone (0.25W at 850MHz)?
iii. How many such microwave photons does it take to warm a 200ml glass
of water by 10 C? (The heat capacity of water is roughly 4J/gK, and the
density is 1g/ml.)
iv. At a given power of an electromagnetic wave, do you expect a classical wave
description to work better for radio frequencies, or for X-rays?
Physics 8.04, Quantum Physics 1, Spring 2016 3

4. Complex Number Practice [15 points]

A complex number can be written in either Cartesian or polar form
p
z = a + ib = reiθ , |z| ≡ a2 + b2 . (1)

The real numbers a and b are, respectively, the real and imaginary parts of z. The
real numbers r and θ are, respectively, the magnitude and phase of z. We call |z| the
norm of z. Use this definition for z in the following:

(a) Use Taylor expansions to derive the Euler formula

eiθ = cos θ + i sin θ . (2)

(b) Write a and b in terms of r and θ, and vice versa.

(c) Complex numbers are viewed a vectors in a two-dimensional “complex plane”.
Multiplication of a complex number by a phase (a complex number of unit mag-
nitude) is equivalent to a rotation in the complex plane.
i. Show that multiplication by i is equivalent to rotation by 90◦ : iz = rei(θ+π/2)
ii. Write iz in terms of a and b. What is the real part of iz?
iii. Show that multiplication by eiφ is equivalent to rotating by φ.
(d) The complex conjugate z ∗ of a complex number z = a + ib is z ∗ = a − ib. A
complex number z is actually real if z = z ∗ , meaning that its imaginary part is
zero. A complex number z is actually imaginary if z = −z ∗ , which implies that
its real part is zero.
i. Is there a number that is both real and purely imaginary?
ii. What is (z ∗ )∗ ? Show that z ∗ = re−iθ .
iii. Express the real and imaginary parts of z in terms of z and z ∗ .
iv. Show that zz ∗ is real and evaluate it to express it in terms of a and b, in
terms of r, and in terms of |z|.
(e) Using the Euler formula derive formulae for cos 2θ, sin 2θ, cos 3θ, and sin 3θ, all
in terms of sin θ and cos θ. Derive formulae for cos(A + B) and sin(A + B), both
in terms of sines and cosines of A and B.

5. Absorption? [5 points]
A photon collides with a free electron. Explain why the photon cannot be completely
absorbed.

6. Mach-Zender interferometer [10 points]

Consider
  the Mach-Zender interferometer and assume an input beam of the form
α
. Call P0 and P1 the detection probabilities at D0 and D1.
β

(a) Calculate P0 and P1 assuming we insert a phase shifter with phase δl on the
lower leg of the interferometer.
Physics 8.04, Quantum Physics 1, Spring 2016 4

(b) Calculate P0 and P1 assuming we insert a phase shifter with phase δu on the
upper leg of the interferometer.
(c) Calculate P0 and P1 assuming we insert the two phase shifters simultaneously.

7. Elitzur-Vaidman bombs! [10 points]

(a) Suppose you decide to test bombs with a Mach-Zender interferometer repeatedly
until the status of any given bomb is certain beyond reasonable doubt. What
fraction of the working bombs are certified without detonation?
(b) Suppose 80% of the bombs in your possession are defective. You choose one at
random and test it with a Mach-Zender interferometer by sending in one photon.
You detect the photon at D0. What is the probability that the bomb is defective?
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 2
Massachusetts Institute of Technology
Physics Department Due Thu. February 18, 2016
February 11, 2016 5:00pm

Problem Set 2
1. de Broglie wavelength [20 points]

(a) The de Broglie wavelength of a non-relativistic (nr) electron with kinetic energy
Ekin can be written as as
δ
λnr = √ Å .
Ekin
In this formula δ is a unit-free constant, and the value of the energy Ekin is
entered in eV as a pure number. The answer comes out in Angstroms (Å =
10−10 m) . Give the value of the unit-free constant δ.
(b) The de Broglie wavelength of a relativistic (r) electron with energy E can be
calculated in terms of the γ factor of the electron: E = γme c2 . One finds
ℓ
λr = p .
γ2 − 1

What is the value of ℓ in fm = 10−15 m? Is this a well-known length?

(c) Rewrite the expression for λnr in (a) in terms of ℓ and γ, using Ekin = (γ −
1)me c2 . Demonstrate that λr < λnr for any energy.
(d) A few numerical calculations:
i. What is the energy of an electron whose de Broglie wavelength is equal to
its Compton wavelength? Is that electron relativistic: Is it moving faster
than 0.2 c ?
ii. The de Broglie wavelength of a particle gives you the rough idea of the dis-
tance scale it can explore in a collision experiment. The International Linear
Collider, which may be built in the near future, is expected to accelerate
electrons to 1 TeV = 1000 GeV. What is the de Broglie wavelength of such
electrons? Compare to the de Broglie wavelength of 7 TeV protons at the
LHC at Geneva.
iii. What is the maximum electron kinetic energy, and the associated β = v/c,
for which the non-relativistic value of λ (in (a) or (c)) has an error less than
or equal to 10% ?

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

2. Bohr radius, electron Compton wavelength, and classical electron radius.

[10 points]
The classical electron radius r0 is the radius obtained by setting the electrostatic
energy associated to a charged ball of radius r0 equal (up to constant factors) to the
rest energy of the electron

e2 e2
= me c2 → r0 = .
r0 me c2
Here e is the charge of the electron. The bar-Compton wavelength λ̄C of the electron
is
~
λ̄C = .
me c
Finally, the fine structure constant α, which measures the strength of the electromag-
netic coupling is
e2 1
α = ≃ .
~c 137
(a) The Bohr radius a0 is the length scale that can be constructed from e2 , ~, and
me and no extra numerical constants. Find the formula for the Bohr radius by
consideration of units. Evaluate this length in terms of fm.
(b) Show that the three lengths form a geometric sequence with ratio α:

a0 : λ̄C : r0 = 1 : α : α2 .

Use this to give the values of λ̄C and r0 in fm.

3. Two-by-two matrices and linear devices. [10 points]

Consider the two-beam Mach-Zender interferometer and a beam represented by the
two-component column vector u:
 
u1
u = , with |u1 |2 + |u2 |2 = 1 .
u2

Any linear optical element in the interferometer can be represented by a two-by-two

matrix R such that with input u beam the output is a u′ beam given by

u′ = R u .

Show that conservation of probability for arbitrary u requires that R be a unitary

matrix. A (finite size) matrix R is said to be unitary if R† R = 1, where dagger
denotes the operation of transposition and complex conjugation.
Physics 8.04, Quantum Physics 1, Spring 2016 3

4. Improving on bomb detection [15 points]

We modify the Mach-Zehnder interferometer to increase the percentage of Elitzur-
Vaidman bombs that can be vouched to work without detonating them. For this
purpose we build a beam-splitter with reflectivity R and transmissivity T . A photon
incoming (from either port) has a probability R to be reflected and a probability T to
be transmitted (R + T = 1). Let r and t denote the positive square roots:
√ √
r ≡ R, t ≡ T .

(a) Build the two-by-two matrix U that represents the beam splitter. For this con-
sider
  what happens when a photon hits the beam splitterfrom
 the top side (input
1 0
) and when it hits it from the bottom side (input ). To fix conventions
0 1
U will have all entries positive (and real) except from the bottom right-most
element (the 2,2 element). Confirm that U is unitary.

The interferometer with detectors D0 and D1 (shown below) uses two identical
copies of the beam splitter. The incoming photon arrives from the top side.

(b) A defective bomb is inserted in the lower branch of the interferometer. What
are the detection probabilities P0 and P1 at D0 and D1 respectively?
A functioning bomb inserted in the lower branch of the interferometer. What
is the detonation probability Pboom and the detection probabilities P0 and P1 ?
Express your answers in terms of R and T .
(c) You test bombs until you are reasonably sure that either they malfunction or
that they are operational. What fraction f of the operational bombs can be
certified to be good without detonating them? Give your answer in terms of R.
What is the maximum possible value for f ?
Physics 8.04, Quantum Physics 1, Spring 2016 4

5. Plane waves for matter particles. [10 points] Assume we want to represent the
wave for a matter particle moving in the x direction with momentum p = ~k. A
reasonable guess for such a wave is

Ψ(x, t) = cos(kx − ωt) + γ sin(kx − ωt) ,

where γ is a constant. A physical requirement is that an arbitrary displacement of x

or an arbitrary shift of t should not alter the character of the wave. We will demand
therefore that after the shift, whose effect is to change the phase by some constant ǫ,
we have

cos(kx − ωt + ǫ) + γ sin(kx − ωt + ǫ) = a cos(kx − ωt) + γ sin(kx − ωt)]

for some constant a that may depend on ǫ.

Write the equations that follow from the above requirement. Find the two possible
solutions for γ and the associated a. Which is the solution that corresponds to our
conventional description of a matter wave?
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 3
MIT Physics Department Due Thu. February 25, 2016
February 18, 2016 5:00pm

Announcements
• Recommended Reading: Griffiths, sections 1.1, 1.2, 1.4 and 1.5.

Problem Set 3
1. Exercises with commutators [10 points] Let A, B, and C be linear operators.

(a) Show that [A, BC] = [A, B]C + B[A, C].

(b) Show that [AB, C] = A[B, C] + [A, C]B.
(c) Show that [A, [B, C]] + [B, [C, A]] + [C, [A, B]] = 0.
(d) Calculate [x̂n , p̂] and [x, p̂n ] for n an arbitrary integer greater than zero.
(e) Calculate [x̂p̂, x̂2 ] and [x̂p̂, p̂2 ].

2. Simple tests of the stationary phase approximation [10 points]

In here we consider integrals of the form
Z ∞
Ψ(x) = dk Φ(k)eikx ,
−∞

where Φ(k) is a function that is sharply localized around k = k0 . In each of the

following cases use the stationary phase argument to predict the location of the peak
of |Ψ(x)|. Then compute the integral exactly to find Ψ(x), |Ψ(x)|, and to confirm
your prediction.
2 (k−k 2
0)
(a) Φ(k) = e−L , where L is a constant with units of length.
−L2 (k−k )2
(b) Φ(k) = e 0
e−ikx0 , where x0 and L are constants with units of length.

Useful integral: Valid for complex constants a and b, with real part of a positive:
Z ∞ r
π  b2 
−ax2 +bx
e dx = exp , when Re(a) > 0 .
−∞ a 4a

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

3. Galilean invariance of the free Schrodinger equation. [15 points]

Show that the free-particle one-dimensional Schrödinger equation for the wavefunc-
tion Ψ(x, t):
∂Ψ ~2 ∂ 2 Ψ
i~ = − ,
∂t 2m ∂x2
is invariant under Galilean transformations

x′ = x − vt , t′ = t .

By this we mean that there is a Ψ′ (x′ , t′ ) of the form

Ψ′ (x′ , t′ ) = f (x, t) Ψ(x, t) ,

where the function f (x, t) involves x, t, ~, m and v, and such that Ψ′ satisfies the
corresponding Schrödinger equation in primed variables.
∂Ψ′ ~2 ∂ 2 Ψ′
i~ = − .
∂t′ 2m ∂x′2
(a) Find the function f (x, t). [Hint: Note that the function f (x, t) cannot depend
on any observable of Ψ; it is a universal function that is used to transform any
Ψ. Thus if Ψ is a (single) plane wave, f cannot depend on its momentum or its
energy.]
(b) Demonstrate that the plane wave solution

Ψ(x, t) = A ei(kx−ωt)

transforms as expected. In other words, give Ψ′ and show that it represents, in

the primed reference frame, a particle with the expected momentum and energy.

4. Re-do current conservation in 3D [10 points]

In class we derived the expression for the one-dimensional probability current J(x, t)
starting from ρ(x, t) = |Ψ(x, t)|2 and using the one-dimensional Schrödinger equation
to write
∂ρ ∂J
+ = 0.
∂t ∂x
Repeat the same steps starting from
2
ρ(x, t) = Ψ(x, t) ,

and using the three-dimensional Schrödinger equation to derive the form of the prob-
ability current J(x, t) that should appear in the conservation equation
∂ρ
+∇·J = 0.
∂t
Physics 8.04, Quantum Physics 1, Spring 2016 3

5. Time evolution of an overlap between two states. [10 points] (Merzbacher)

Consider a wavefunction that at time t = 0 is the superposition of two widely sepa-
rated narrow wave packets Ψ1 and Ψ2 :

Ψ(x, 0) = Ψ1 (x, 0) + Ψ2 (x, 0) .

Each packet is separately normalizable. We define the overlap integral γ(t) as

Z ∞
γ(t) ≡ Ψ∗1 (x, t)Ψ2 (x, t)dx .
−∞

At time equal to zero the value of |γ(0)| is very small. As the packets evolve and
spread, what will happen to the value of |γ(t)|? Will it increase as the packets
overlap?

6. Probability current in one dimension [10 points]

Calculate the probability current J(x) for the following wavefunctions, all of which
refer to t = 0:

(a) Ψ(x) = A eγx . Here A is a complex constant and γ is a real constant.

(b) Ψ(x) = N(x)eiS(x)/~ . Here N(x) and S(x) are real.
(c) Ψ(x) = Aeikx + Be−ikx . Here A, B are complex constants and k is real.
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 4
MIT Physics Department Due Fri. March 4, 2016
February 25, 2016 12:00 noon

Announcements
• Recommended Reading: Griffiths, sections 1.5, 1.6, 2.4.

Problem Set 4
1. Exercises on packets changing shape [5 points]

(a) A free proton is localized within ∆x = 10−10 m. Estimate the time ts it takes
the packet to spread appreciably. Repeat for an proton localized within 1 cm.
(b) Consider a wave packet that satisfies the relation ∆x∆p ∼ ~. Show that the
condition ∆p ≪ p guarantees that the packet does not spread appreciably in
the time it takes to pass through a fixed position.

2. Probability current in three dimensions [10 points]

In elastic scattering of particles in three-dimensional space the wavefunction takes
the form
f (θ) ikr
Ψ(x) = eikz + e , valid for large r .
r
The time dependence has been suppressed; it is just an overall time dependent phase
e−iEt/~ with E = ~2 k 2 /(2m). It will play no role here.
The first term represents the incoming particles, moving in the +z direction. The
target is located at the origin r = 0 and the second term represents the amplitude for
particles moving radially out – the scattered particles. This amplitude is θ dependent
but assumed to be φ independent; f (θ) is a complex function of θ that carries the
information about the scattering. Recall that θ is the polar angle and z = r cos θ.
When you calculate the probability current J(x) associated with Ψ there will be a
contribution J1 due to the first term (the plane wave), a contribution J2 due to the
second term (the spherical waves), and a contribution J12 due to interference between
the first and second term:

J(x) = J1 (x) + J2 (x) + J12 (x) .

(a) Calculate the probability current J1 and the total flux of this current over a
large sphere of radius R centered at the origin r = 0.

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

(b) Calculate the radial component r̂ · J2 of the probability current J2 . Here r̂ is

the unit vector in the radial direction. Calculate the flux of this current over
a radius R sphere centered at the origin in the
R limit as R R→ ∞. Your answer
should be left as an integral over solid angle dΩ, or over dθ, if you prefer.
(c) Calculate the radial component of the interference term J12 but pick up only
the leading part in 1/r (that is ignore 1/r 2 terms). Show that the answer can
be written in the form
~k 

r̂ · J12 = Im i . . .
mr
where (. . .) represent terms that your calculation should determine. These terms
depend om f (θ), cos θ, and the product kr in exponentials. Calculating the flux
of this current over the large sphere is delicate so we will leave that for later (the
end result is the so called optical theorem!).

3. Evolving the Gaussian wave packet [15 points]

Consider the normalized wave packet representing the state of a particle of mass m
at t = 0:  x2 
1
Ψa (x, 0) = √ exp − 2 .
(2π)1/4 a 4a
Here a is a length parameter that represents the width of the packet at zero time.

(a) Confirm that Ψa (x, 0) is properly normalized.

(b) Find the Fourier representation of Ψa (x, 0), namely, determine the function
Φa (k) such that Z ∞
1
Ψa (x, 0) = √ Φa (k)eikx dk .
2π −∞
(c) Assume the particle is free to find the wavefunction Ψa (x, t) for arbitrary t > 0.
The answer is a bit messy, but can be written more clearly using the time
constant τ built from the constants in the problem:

2ma2
τ ≡ .
~

(d) At time zero the probability density is

1  x2 
|Ψa (x, 0)|2 = √ exp − 2 ≡ G(x; a) ,
2π a 2a

where we defined the gaussian G(x; a) with width parameter a. What is the
probability density |Ψa (x, t)|2 for t > 0. Express your answer in terms of the
gaussian G with a time-dependent width parameter a(t). Give a(t).
Physics 8.04, Quantum Physics 1, Spring 2016 3

Useful integral: Valid for complex constants a and b, with real part of a positive:
Z ∞ r  b2 
−ax2 +bx π
e dx = exp , when Re(a) > 0 .
−∞ a 4a
4. Parseval’s identity in 1D and 3D, and application [10 points]
(a) Consider the Fourier pair (Ψ(x), Φ(p)) relevant to one dimensional (1D) wave-
functions and the Fourier pair (Ψ(x), Φ(p) ) relevant to three-dimensional (3D)
wavefunctions. Use the Fourier relations and the integral form for the delta
function to prove the 1D and 3D versions of Parseval’s identity:1
Z ∞ Z ∞
2
dx|Ψ(x)| = dp|Φ(p)|2 ,
−∞ −∞
Z Z
d3 x |Ψ(x)|2 = d3 p |Φ(p)|2 .

(b) In the hydrogen atom the ground state wavefunction takes the form Ψ(x) =
Ne−r/a0 where r = |x|, a0 is the Bohr radius, and N is a normalization constant.
Find N. The Fourier transform (which you need not derive) takes the form
N′
Φ(p) =  2 ,
a20 p2
1+ ~2

for some constant N ′ and with p ≡ |p|. Find N ′ (you may use an algebraic
manipulator to do the integral). Calculate the probability that the electron may
be found with a momentum whose magnitude exceeds ~/a0 . (Write your inte-
grals explicitly, but you may evaluate them with a computer). [The momentum
distribution was measured by ionization of atomic hydrogen by a high energy
electron beam, see, Lohan, B. and Weigold, E. (1981) “Direct measurement of
the Electron Momentum Probability Distribution in Atomic Hydrogen,” Phys.
Lett. 86A, 139-141.]
5. Ehrenfest theorem [10 points]
Consider a particle moving in one dimension with Hamiltonian H given by
p2
H= + V (x) .
2m
Show that the expectation values hxi and hpi are time-dependent functions that
satisfy the following differential equations:
d 1
hxi = h pi ,
dt m
d D ∂V E
h pi = − .
dt ∂x
1
In mathematics this is called Plancherel’s theorem. Parseval’s result was the analog for Fourier series.
Physics 8.04, Quantum Physics 1, Spring 2016 4

6. Momentum uncertainty [5 points]

Show that in a free-particle wave packet the momentum uncertainty ∆p does not
change in time.

7. Finding Meaning in the Phase of the Wavefunction [10 points]

Suppose ψo (x) is a properly-normalized wavefunction with hxiψo = xo and hpiψo = po ,
where xo and p0 are constants. Define the boost operator B ˆq to be the operator that
acts on arbitrary functions of x by multiplication by a q-dependent phase:

B̂q f (x) = eiqx/~ f (x) .

Here q is a real number with the appropriate units. Consider now a new wavefunction
obtained by boosting the initial wavefunction:
ˆq ψo (x) .
ψnew (x) = B

(a) What is the expectation value hxiψnew in the state given by ψnew (x)?
(b) What is the expectation value hpiψnew in the state given by ψnew (x)?
(c) Based on your results, what is the physical significance of adding an overall
factor eiqx/~ to a wavefunction.
(d) Compute [p̂, B ˆq ] and [x̂, B
ˆq ].
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 5
MIT Physics Department Due Fri. March 18, 2016
March 11, 2015 12:00 noon

Reading: Griffiths: 3.2, 3.3, 3.4, 2.1 and 2.2.

Problem Set 5
1. Gaussians and uncertainty product saturation [5 points]
Consider the gaussian wavefunction
 1 x2 
ψ(x) = N exp − 2 , (1)
2a
where N ∈ R a is a real positive constant with units of length. The integrals
Z ∞ r
π  β2 
−αx2 +βx
dxe = exp , Re(α) > 0 ,
−∞ α 4α
Z ∞ Z ∞
2 −αx2 1 2
dx x e = dx e−αx
−∞ 2α −∞

(a) Use the position space wavefunction (1) to calculate the uncertainties ∆x and
∆p. Confirm that your answer saturates the Heisenberg uncertainty product
~
∆x∆p ≥ 2
.
(Hints: These calculations are actually quite brief if done the right way! Using
the second of the above integrals you don’t even have to determine N. For the
evaluation of hp̂2 i in position space fold one factor of p̂ into ψ ∗ . ).
(b) Calculate the Fourier transform φ(p) of ψ(x). Use Parseval to confirm your
answer and then recalculate ∆p using momentum space.

2. Complex Gaussians and the uncertainty product [10 points]

Consider the gaussian wavefunction
 1 x2 
ψ(x) = N exp − 2
, ∆ ∈ C , Re(∆2 ) > 0 , (1)
2∆
where N is a real normalization constant and ∆ is now a complex number: ∆∗ 6= ∆.
The integrals in Problem 1 are also useful here and so is the following relation, valid
for any nonzero complex number z,
1 Re(z)
Re = (prove it!)
z |z|2

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

(a) Use the position space representation (1) of the wavefunction to calculate the
uncertainties ∆x and ∆p. Leave your answer in terms of |∆| and Re(∆2 ). (∆x
will depend on both1 , while ∆p will depend only on Re(∆2 )).
(b) Calculate the Fourier transform φ(p) of ψ(x). Use Parseval to confirm your
answer and then recalculate ∆p using momentum space.
(c) We parameterize ∆ using a phase φ∆ ∈ R as follows
∆ = |∆| eiφ∆ .
Calculate the product ∆x∆p and confirm that the answer can be put in terms of
a trigonometric function of φ∆ and that |∆| drops out. Is your answer reasonable
for φ∆ = 0 and for φ∆ = π4 ?
(d) Consider the free evolution of a gaussian wave packet in Problem 3 of Home-
work 4. What is ∆p at time equal zero? Examine the time evolution of the
gaussian (from the solution!) and read the value of the time-dependent (com-
plex) constant ∆2 . Confirm that ∆p, found in (a), gives a time-independent
result.
3. Exercises with a particle in a box [15 points]
Consider a 1D problem for a particle of mass m that is free to move in the interval
x ∈ [0, a]. The potential V (x) is zero in this interval and infinite elsewhere. For that
system consider a solution of the Schrödinger equation of the form
 nπ 
Ψn (x, t) = N sin x e−iφn (t) , x ∈ [0, a] ,
a
and Ψn (x, t) = 0 for x < 0 and x > a. Here n ≥ 1 is an integer.

(a) Find the expression for the (real) phase φn (t) so that the above wavefunction
solves the Schrödinger equation. Find the normalization constant N.
(b) Use Ψn (x, 0) to calculate hxi, hx2 i, and ∆x.
(c) Use Ψn (x, 0) to calculate hpi, hp2i, and ∆p.
(d) Is the uncertainty inequality satisfied? Is it saturated?
(e) What answers in (b) and (c) change for Ψn (x, t)? Explain.

4. A Hard Wall [5 points]

A particle of mass m is moving in one dimension, subject to the potential V (x):
(
0, for x > 0,
V (x) =
∞ for x ≤ 0 .

Find the stationary states and their energies. These states cannot be normalized.
1
Actually ∆x can be written in terms of Re(1/∆2 ) alone.
Physics 8.04, Quantum Physics 1, Spring 2016 3

5. A Step Up on the Infinite Line [10 points]

A particle of mass m is moving in one dimension, subject to the potential V (x):
(
V0 , for x > 0,
V (x) =
0 , for x ≤ 0 .
Find the stationary states that exist for energies 0 < E < V0 .
6. A Wall and Half of a Finite Well [10 points]
A particle of mass m is moving in one dimension, subject to the potential V (x):

 ∞ , for x < 0,

V (x) = −V0 , for 0 < x < a , (V0 > 0)

0 , for x > a .


Find the stationary states that correspond to bound states (E < 0, in this case). Is
there always a bound state? Find the minimum value of z0
2ma2 V0
z02 = ,
~2
for which there are three bound states. Explain the precise relation of this problem
to the problem of the finite square well of width 2a.
7. Mimicking hydrogen with a one-dimensional square well. [5 points]
The hydrogen atom the Bohr radius a0 and ground state energy E0 are given by
~2 −10 e2
a0 = ≃ 0.529 × 10 m , E0 = − = −13.6 eV.
me2 2a0
The ground state is a bound state and the potential goes to zero at infinity. We want
to design a one-dimensional finite square well
(
−V0 , for |x| < a0 , V0 > 0 ,
V (x) =
0 , for |x| > a0 ,
that simulates the hydrogen atom. Calculate the value of V0 in eV so that the ground
state of the box is at the correct depth.
8. No states with E < V (x) [5 points]
Consider a real stationary state ψ(x) with energy E:
~2 ′′
− ψ (x) + [V (x) − E]ψ(x) = 0 .
2m
(a) Prove that E must exceed the minimum value of V (x) by noting that E = hHi.
(b) Explain the claim by trying (and failing) to sketch a wavefunction consistent
with being on the classically inaccessible region for all values of x.
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 6
MIT Physics Department Due Friday April 1, 2016
March 17, 2016 12:00 noon

Reading: Griffiths section 2.6. For the following week sections 2.5 and 2.3.

Problem Set 6

1. Particle in a square well. [10 points] A particle of mass m moves in an infinite

square well of width a. Its wavefunction at time t = 0 is
r r r
1 2 2πx 2 2 3πx
Ψ(x, 0) = √ sin + sin .
3 a a 3 a a

(a) Is Ψ in an energy eigenstate? Find Ψ(x, t).

(b) What are the probabilities that a measurement of the energy at time t gives
each of the following values

~2 π 2 4~2 π 2 9~2 π 2
, , .
2ma2 2ma2 2ma2

(c) What is the expectation value of x at time t?

(d) What is the expectation value of p at time t ?

2. Non-degeneracy of bound states in one dimension [10 points]

Griffiths Problem 2.45, p. 87.

3. Infinite rectangular well in the plane [10 points]

Consider a particle of mass m moving in the x, y plane with potential that is zero
inside the rectangular box comprised of all points (x, y) for which

0 ≤ x ≤ Lx , 0 ≤ y ≤ Ly ,

and is infinite elsewhere.

(a) Use the two-dimensional Schrödinger equation to find the energy eigenstates.
Give the energies and the normalized eigenfunctions.

1
Physics 8.04, Quantum Physics 1, Fall 2015 2

(b) Consider the case Lx = Ly = L. You will see that there are degeneracies in
the energy spectrum. Some degeneracies have a simple symmetry explanation;
identify them and state why they occur. Some degeneracies are accidental; they
seem to be random. Display some examples. [Hint: 49+ 1 = 25 + 25].
(c) Show that whenever (Lx /Ly )2 is irrational there are no degeneracies.

4. An infinite square well with a step [10 points]

A particle of mass m is moving in one dimension, subject to the potential V (x):


 ∞ , for x < 0,

 0, for 0 < x < a ,
V (x) =


 V0 , for a < x < 2a , (V0 > 0)
∞ , for x > 2a .


(a) Find the equations that determine the stationary states with energies 0 < E <
V0 . For this we define

2 2mE 2 2m(V0 − E) 2ma2 V0

k ≡ , κ ≡ , z02 = , η ≡ ka , ξ ≡ κa .
~2 ~2 ~2
(We use k for classically allowed regions and κ for classically forbidden re-
gions).Your equations should be possible to write in terms of ξ, η, and z0 .
(b) As a numerical application, consider z0 = 2π. How many states do you get with
E < V0 ? Find the possible values of the energy E in terms of V0 (use at least 4
significant digits).

5. Shooting method and application [15 points]

For a particle in a quartic potential V (x) ∼ x4 , after rescaling of x into a unit-free
variable u, the Schrödinger equation takes the form

1 d2 ψ
− + (u4 − e)ψ = 0 ,
2 du 2
where e is a unit-free measure of the energy eigenvalue. The Mathematica instructions
that allow you to find the values of e for the even solutions of this potential are given
below. These instructions produce a plot for the solution ψ(u), for u ∈ [0, 3.5] with
some suitable initial conditions and for the chosen value of the energy e.
Clear[e, psi]
v[x_]:= x^4
e=0.65;
psi = psi/. NDSolve[{-(1/2)psi’’[u] + (v[u]-e)psi[u]==0, psi[0]==1,
Physics 8.04, Quantum Physics 1, Fall 2015 3

psi’[0]==0}, psi, {u, 0, 3.5}][[1]];

Plot[psi[u], {u, 0, 3.5}]
After executing these instructions, if you write psi[0.5], for example, the program
will return the value of ψ at u = 0.5.
Play with this to familiarize yourself. The initial value of e set above is 0.65 but the
ground state energy, as you can find out by trial and error, is a little bit higher.
We now revisit the previous problem: a particle of mass m on an infinite square well
with a step. Again, we take z0 = 2π. You found two bound states with E < V0 :

E1 = 0.##436 V0 , E2 = 0.##747V0 .

(a) Use x = au, with u ∈ [0, 2] unit free, and write V = V0 f (u) for a suitably defined
f (u) to obtain a differential equation for the energy eigenstates in which no units
appear and the energy eigenvalue is encoded by the pure number e = E/V0 .
Test your differential equation with the shooting method to recover the above
values of E1 and E2 . Find the next two energy levels E3 and E4 .
(b) We discussed in lecture the fact that for slowly varying potentials the amplitude
of the wavefunction is roughly proportional to the square root of the ‘local’ de
Broglie wavelength. Our potential, having a step, is not really slowly varying,
but we can still see numerically to what degree this property holds.
Construct the energy eigenstate with 8 nodes (the eighth excited state) and
determine its energy. Let AL and AR denote the amplitudes of your wavefunction
on the left and right sides of the square well. Read the ratio AL /AR from your
wavefunction and compare with the prediction for this ratio using the de Broglie
wavelength.

6. Hydrogen ion using the square well model. [10 points]

Last time you modeled the hydrogen atom size and ground state energy
~2 e2
a0 = , E0 = − = −13.6 eV,
me2 2a0
using the square well potential
(
−V0 , for |x| < a0 , V0 > 0 ,
V (x) =
0, for |x| > a0 .

You previously found that this well has

z0 = 1.3192, and V0 = z02 |E0 | = 1.7402|E0| = 23.67eV .

The square well potential mimics the potential created by the proton and the ground
state energy is the energy of the electron in this potential.
Physics 8.04, Quantum Physics 1, Fall 2015 4

To simulate the hydrogen ion H2+ (2 protons 1 electron) we will construct an even
potential with two identical square well models of hydrogen separated by a small
distance 2γ a0 where γ is a small positive unit-free constant. The potential is therefore

 0 for |x| < γa0 ,

V (x) = −V0 , for γa0 < |x| < (2 + γ)a0 , V0 > 0 ,

0 , for |x| > (2 + γ)a0 .


For definiteness work with γ = 0.2 .

(a) Use the shooting method to find the energy of the lowest energy eigenstate,
namely, the bound state energy of an electron shared by the two protons. Show
the wavefunction of the electron from the plot of your solution.
(b) The binding energy of the ion is obtained by adding the positive energy due to
the repulsion of the two protons to the above ground state energy. What binding
energy do you get? How does it compare with the experimental value?
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 7
MIT Physics Department Due Friday April 8, 2016
April 1, 2016 12:00 noon

Reading: Griffiths sections 2.5 and 2.3.

Problem Set 7

1. Two delta functions [15 points]

Consider a particle of mass m moving in a one-dimensional double well potential

V (x) = −gδ(x − a) − gδ(x + a) , g > 0.

(a) Find transcendental equations for the bound state energy eigenvalues of the
system. Plot the energy levels in units of ~2 /(ma2 ) as a function of the dimen-
sionless parameter λ ≡ mag/~2. Explain the features of the plot.
(b) In the limit of large separation 2a between the wells find a simple formula for
the splitting between the ground state and the first excited state.

2. Sketching wavefunctions. Griffiths 2.47, p. 87. [10 points]

In this problem you should try to figure out intuitively how the solutions look. It is
a good idea then to check your intuition with the shooting method and the setup of
the H2+ ion.

3. Harmonic oscillators beyond the turning points [10 points]

For the simple harmonic oscillator energy eigenstates with n = 0, 1, and 2, calculate
the probability that the coordinate x takes a value greater than the amplitude of a
classical oscillator of the same energy.

4. Harmonic oscillator computations [15 points]

(a) Calculate the expectation value of x4 on the energy eigenstate with number n.
(b) Calculate ∆x and ∆p on the energy eigenstate with number n. What is the
value of the product ∆x∆p?

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

(c) Consider the polynomials Hn (ξ) defined by the generating function

∞
2 +2sξ
X sn
e−s = Hn (ξ) .
n=0
n!

Verify that Hn (ξ) = (2ξ)n +. . . where the dots represent terms with lower powers
of ξ. Show that the polynomials Hn (ξ) so defined satisfy the Hermite differential
equation:
Hn′′ − 2ξHn′ + 2nHn = 0 .

5. Harmonic oscillator and a wall. Griffiths Problem 2.42. p. 86. [ 5 points]

6. Harmonic oscillator oscillating! [10 points]

A particle of mass m in a harmonic oscillator with frequency ω has an initial, time
zero wavefunction
1

Ψ(x, 0) = √ ϕ0 (x) + ϕ1 (x) ,
2
where ϕ0 and ϕ1 are the normalized eigenstates of the Hamiltonian with number
eigenvalue zero and one, respectively.

(a) Write down Ψ(x, t) and |Ψ(x, t)|2 . You may leave your expressions in terms of
ϕ0 and ϕ1 .
(b) Find hxi as a function of time. What is the amplitude of this oscillation and
what is its frequency?
(c) Find hpi as a function of time.
(d) Show that for any harmonic oscillator state, the probability distribution |Ψ(x, t)|2
is equal to |Ψ(x, t + T )|2 for T = 2π
ω
.
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 8
MIT Physics Department Due Friday, April 22, 2016
April 13, 2016 12:00 noon

Problem Set 8
Reading: Griffiths, pages 73-76, 81-82 (on scattering states).
Ohanian, Chapter 11: Scattering and Resonances

1. States of the harmonic oscillator [15 points]

Consider the state ψα defined by

ψα ≡ N exp(α↠) ϕ0 ,

with α ∈ C a complex number. For the first two questions below it may be helpful to
simply expand the above exponential.

(a) Find the constant N needed for the state ψα to be normalized.

(b) Show that the state ψα is an eigenstate of the annihilation operator â. What is
the eigenvalue?
(c) Find the expectation value of the Hamiltonian in the state ψα .
(d) Find the uncertainty in the energy in the state ψα .
(e) Use the eigenvalue equation, viewed as a differential equation to calculate the
explicit form of the normalized wavefunction ψα .

2. Two delta functions- again [15 points]

Consider again the problem of a particle of mass m moving in a one-dimensional
double well potential

V (x) = −gδ(x − a) − gδ(x + a) , g > 0.

You found in the previous set the value of the bound state energy E for the even state
in terms of the energy E0 = ~2 /(2ma2 ). You had ξ = κa
E ξ mag
= −ξ 2 where = λ, λ≡ ,
E0 1 + e−2ξ ~2
with λ unit free, encoding the intensity g of the delta functions, if a is constant, or
the separation of the delta functions, if g is constant. We can thus write

a ~2
λ= a0 ≡ ,
a0 mg

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

with a0 a natural length scale in the problem once g is fixed. Introduce also the
energy E∞ associated with a single delta function:
mg 2
E∞ ≡ .
2~2
Assume now that this is a model for a diatomic molecule with interatomic distance
2a. The bound state electron helps overcome the repulsive energy between the ions.
Let the repulsive potential energy Vr (x), with x the distance between the atoms, be
given by
βg
Vr (x) = , β > 0,
x
where β is a small number. The total potential energy Vtot of the configuration is the
sum of the negative energy E of the bound state and the positive repulsive energy:
Vtot = E + Vr (2a) .

(a) Write E as E = −E∞ f (ξ, λ) where f is a function you should determine. Plot
E as a function of a/a0 = λ in order to understand how the ground state energy
varies as a function of the separation between the molecules. What are the values
of E for a → 0 and for a → ∞?
(b) Write Vr in terms of E∞ , β, and λ.
(c) Now consider the total potential energy Vtot and plot it as a function of a/a0 = λ
for various values of β. You should find a critical stable point for the potential
for sufficiently small β. For β = 0.31 what is the approximate value of a/a0 at
the critical point of the potential?

3. Finite square well turning into the infinite square well [5 points]
Consider the standard square well potential
(
−V0 , for |x| ≤ a, V0 > 0 ,
V (x) = (1)
0 for |x| > a ,
and the wavefunction for an even state

 √1 cos kx , for |x| ≤ a,
a
ψ(x) = (2)
A
 √ e−κ|x| ,
a
for x > |a| ,

where we included the √1 prefactor to have consistent units for ψ.

a
We want to have a better understanding of the limit as V0 → ∞ and understand why
the discontinuity in ψ ′ in the infinite well does not give trouble. Keeping m and a
constant as we let V0 grow large is the same as letting z0 grow large.
A previous analysis has demonstrated that for the ground state, in the situation of
large z0 , the ansatz (2) is accurately normalized and
π z0
η = ka ≃ π2 (1 − 1
z0 ) , ξ = κa ≃ z0 , A ≃ 2z0 e .
Physics 8.04, Quantum Physics 1, Spring 2016 3

We want to see if the expectation value of the Hamiltonian receives a singular con-
tribution from the forbidden region. Since the potential V (x) vanishes there, we
only need to concern ourselves with the contribution from the kinetic energy operator
p̂2 ˆ from the forbidden
K̂ = 2m . Calculate the contribution to the expectation of K
region x > a Z ∞
ˆ
hKi ≡ ˆ
dx ψ ∗ (x)Kψ(x)
x>a
a
The answer should be in terms of z0 . Interpret your result.

4. Reflection of a wavepacket off a step potential [20 points]

Consider a step potential with step height V0 :
(
V0 , for x > 0
V (x) = (1)
0 , for x < 0 .

We send in from x = −∞ a wavepacket all of whose momentum components have

energies less than the energy V0 of the step. For this we need modes with k satisfying
2mV0
k ≤ kˆ , kˆ2 = . (2)
~2
We will then write the incident wavepacket as
ˆ
k
√
Z
Ψinc (x) = a dk Φ(k) eikx e−iE(k)t/~ , x < 0. (3)
0

Here a is the constant with units of length, uniquely determined by the constants
m, V0 , ~ in this problem, and Φ(k) is a real, unit-free function peaked at k0 < kˆ

~ 2 a2 (k−k 2
a ≡ √ , Φ(k) = e−β 0)
. (4)
mV0
The real constant β, to be fixed below, controls the width of the momentum distri-
√
bution. The units of Ψinc are L−1/2 and that’s why we included the a prefactor in
(3). Recall that dk has units of L−1 .

(a) Write the reflected wavefunction (valid for x < 0) as an integral similar to (3).
This integral involves the phase shift δ(E) calculated in class.

Introduce a unit free version K of the wavenumber k, a unit-free version u of the

coordinate x, and a unit-free version τ of the time t as follows
K ~
k≡ , x ≡ au , t≡ τ. (5)
a V0
Naturally, we will write k0 = K0 /a. Note that kx = Ku.
Physics 8.04, Quantum Physics 1, Spring 2016 4

(b) Show that the group velocity and the uncertainty relation for the incoming packet
take the form
du
= #K0 , ∆u ∆K ≥ # ,
dτ
where # represent numerical constants that you should fix (different constants!).
Use the approximation that we have the full gaussian |Φ(K)|2 to determine the
uncertainty ∆K in the incoming packet in terms of β. Assuming again that we
have a full gaussian, what would be (in terms of β) the minimum possible value of
the uncertainty ∆u for the associated coordinate space probability distribution?
(c) Complete the following equations by fixing the constants represented by #
p
E(k) = # V0 K 2 , e2iδ(E) = # + #K 2 + iK # + #K 2 ≡ w(K) .

(d) Show that the delay ∆t = 2~ δ′ (E) experienced by the reflected wave implies a
∆τ given by
#
∆τ = p ,
K0 # + #K02
where you must fix the constants.
(e) Prove that the complete wavefunction Ψ(x, t) valid for x < 0 and all times, which
we now view as Ψ(u, τ ) valid for u < 0 and all τ , takes the form

1
Z #  
2 (K−K 2 2τ
a 2 Ψ(u, τ ) = dKe−β 0)
e−i#K eiKu − e−iKu w(K)
0

and determine the two missing constants.

(f) Set β = 4 and K0 = 1. What are the values of ∆K and ∆u? What is the
predicted time delay ∆τ ? (Not graded: Can you make an informed guess if the
packet will change shape quickly?)
Now use Mathematica to calculate and make plots of the probability density
1
|a 2 Ψ(u, τ )|2 . Give the plot of the wavefunction for τ = −20, −5, and 0, and
using u ∈ [−30, 0]. Examine the plot for τ = 20 and determine the time delay
∆τ by looking at the position of the peak of the packet. Your answer should
come reasonably close to the analytical value you determined previously.

5. Scattering off a rectangular barrier. Based on Griffiths 2.33. p.83. [10 points]
Do only the cases E < V0 and E = V0 .
Can you get T = 1 for E < V0 ?
Find the answer for E > V0 in some book (or do it). When does one get T = 1 for
E > V0 ?
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 9
MIT Physics Department Due Friday April 29, 2016
April 21, 2016 12:00 noon

Reading:
2KDQLDQ&KDSWHU6FDWWHULQJDQG5HVRQDQFHV
Griffiths: section 4.1.

Problem Set 9
1. A numerical test of stationary phase. [10 points]
We have used stationary phase to figure out the time dependence of the position of
peaks in wavepackets constructed from integral representations. More generally, the
stationary phase approximation can help get the value of the integral itself.
Consider the integral of a Gaussian peaked at x = 2 against a phase factor:
Z ∞
2
dx e−100(x−2) eiφ(λ,x) , φ(λ, x) = 50 x − 32
1
λx4 , λ ∈ R .


f (λ) =
−∞

We want to confirm that |f (λ)| peaks at a value λ∗ selected by stationary phase and
get the value of f (λ∗ ).

(a) What is the width ∆ at half-maximum for the gaussian? In other words, what
is the largest ∆ for which for all x in |x − 2| ≤ 12 ∆ the gaussian is larger than
half-maximum? If you had to do the integral numerically, would it be safe to
integrate from 1 to 3? Explain.
(b) Use stationary phase to find the critical value λ∗ of λ for which f (λ) would have
the largest magnitude. For λ∗ write φ(λ∗ , x) as a Taylor expansion around x = 2
up to and including terms quadratic in (x − 2).
(c) What is the excursion of the phase φ(λ∗ , x) for |x − 2| < 12 ∆? Your result,
expressed in units of π, should imply that it is a decent approximation to ignore
the phase variation at the critical λ. Do so and then perform the resulting
integral analytically. The answer is a complex number. Write your answer in
terms of a phase times the magnitude.
(d) Perform the integral analytically using the quadratic approximation for the
phase. Write your answer in terms of a phase times the magnitude.
(e) Perform the integral numerically as a function of λ for the interval λ ∈ [0, 1].
Plot the absolute value |f (λ)|. What is the value of f (λ) for the critical λ?
Compare with your previous estimates. What is the value of λ that leads to the
largest |f (λ)|?

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

2. Testing Levinson’s theorem in an example [10 points].

For the potential V (x) = −V0 for 0 < x < a, V (x) = 0 for x > a and V (x) = ∞ for
x < 0 we calculated in class the phase shift δ(E) finding
′
1 − kk cot k ′ a tan ka
tan δ = ,
tan ka + kk cot k ′ a
′

with
2mE 2m(E + V0 ) 2mV0 a2
k2 = , k ′2 = , z02 =
~2 ~2 ~2
(a) As the energy E goes to zero, ka → 0. What happens to k ′ a? Show that tan δ
goes to zero, and thus we can take δ → 0 as ka → 0.
(b) What is the limit for tan δ as E → ∞? Explain in detail.
(c) Call u ≡ ka and write tan δ as a function of u and z0

tan δ = f (u; z0 ) → δ = ArcTan [f (u; z0)] .

Write the function f (u; z0 ).

In order to construct plots with Mathematica, I found it difficult to use ArcTan[...]

because it uses the range (−π/2, π/2) and the graphs give discontinuities. One option
(suggested by W. Taylor) is to differentiate the ArcTan function and then to integrate
it again! Since δ = 0 for u = 0 we can write:
Z u
d
δ(u; z0) = du′ ′ ArcTan [f (u′ ; z0 )] .
0 du
Let the computer take the derivative and integrate. If you find a simpler way to do
this let us know!

(d) Graph the phases δ(u, z0) as a function of u for z0 = 2, 5, 9. For z = 2 use
u ∈ [0, 15], for z0 = 5 use u ∈ [0, 20] and for z0 = 9 use u ∈ [0, 30]. In each
case explain how the result is consistent with Levinson’s theorem and state how
close is δ at the upper value of u to the expected value of δ(E = ∞).
Physics 8.04, Quantum Physics 1, Spring 2016 3

3. Scattering off a step and a wall [10 points]

Consider the potential

 V0 ,
 for 0 < x < a , V0 > 0 ,
V (x) = 0 , for x > a ,

∞, for x ≤ 0 .


Calculate the phase shift δ(k) as a function of k. You will have to consider two cases:

(a) E(k) > V0 . Call k ′ the wavenumber for x < a. You may want to do this
starting from the beginning (for practice). Otherwise you could try to use the
example worked in class (and Problem 2 here), where instead of a step we had
a well of depth V0 , and modify the answer suitably. Leave your answer in the
form cot δ = . . ..
(b) E(k) < V0 . You may want to do this starting from the beginning (for practice).
Otherwise you could try some analytic continuation of the result of part (a).
Leave your answer in the form cot δ = . . ..
(c) Plot δ(k) as a function of u = ka ∈ [0, ∞] for a potential with z0 = 5 (recall
2
that z02 = 2mV0a
~2
).

4. Scattering off a delta function and a wall. [15 points]

Consider our usual one-dimensional potential with V (x) = ∞ for x ≤ 0, and with
V (x) = g δ(x − a) , g > 0 x > 0.

We scatter particles with mass m and energy E > 0 off this potential. We have

22mEa2 mag
(ka) = , λ ≡ , unit free.
~2 ~2
(a) Calculate the phase shift δ(k). Write the answer in the form
sin2 (ka)
tan δ = − ,
h(ka; λ)
where h(ka; λ) is a function you must determine. Explain how, with δ known,
one readily finds the amplitude A(k) multiplying the ‘sin’ function in the wave-
function for 0 < x < a.
(b) To understand features of tan δ calculate the leading approximation to it for
ka ≪ 1. Discuss the λ dependence of the result. For arbitrary ka what does
tan δ become for λ → ∞?
(c) Plot δ, the time delay a1 dk
dδ
, and |A| as functions of ka ∈ [0, 10] for λ = 5. Do
you see resonances? If so, identify the values of ka, the time delay a1 dk
dδ
and the
magnitude of |A|. Is the plot of δ consistent with Levinson’s theorem?
Physics 8.04, Quantum Physics 1, Spring 2016 4

5. A few commutators and a few expectation values. [10 points]

(a) Calculate the commutators

[Lz , x] , [Lz , y] , and [Lz , z] .

(b) Calculate the commutators

[Lz , px ] , [Lz , py ], and [Lz , pz ] .

(c) Assume ψ0 is an Lz eigenfunction. Show that py and px have zero expectation

value in the state ψ0 .
(d) Assume ψ0 is an Lz eigenfunction. Show that y and x have zero expectation
value in the state ψ0 .

6. Angular momentum in spherical coordinates. [10 points]

(a) Calculate the nine partial derivatives of the spherical coordinates (r, θ, φ) with
respect to the Cartesian coordinates (x, y, z) expressing your answers in terms
of the spherical coordinates.
(b) Use the above results to write Lx , Ly , and Lz as differential operators in spherical
coordinates.
(c) Compute L2x , L2y , and L2z as differential operators in spherical coordinates and
use your results to derive the expected form of L2 as a differential operator in
spherical coordinates.
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.
 Quantum Physics I (8.04) Spring 2016
Assignment 10
MIT Physics Department Due Friday May 6, 2016
29 April, 2016 12:00 noon

Reading: Griffiths sections 4.1, 4.2, and 4.3.

Problem Set 10

1. Bound states from imaginary wavenumber [5 points]

Consider the scattering solution for a finite range one-dimensional potential:

ψx (x) = eiδ(k) sin(kx + δ(k)) , x > R .

Show that having a bound state means As = eiδ sin δ has a pole at k = iκ with κ > 0.

2. Simultaneous eigenfunctions [5 points]

Consider two Hermitian operators Aˆ and B ˆ that commute. Assume that at least
one of the operators, say A, ˆ has no degeneracies in its spectrum. Show that the
eigenfunctions of Aˆ are also eigenfunctions of B.
ˆ

3. Expectation values in a particular wavefunction [10 points]. (Based on Oha-

nian, Ch. 7, problem 17).
Suppose a particle has the wavefunction

√
r
1 5
sin2 θ 1 + 14 cos θ cos 2φ f (r) ,


ψ(r, θ, φ) =
4 π
with f (r) a normalized radial wavefunction.

(a) Rewrite this wavefunction in terms of spherical harmonics. What are the pos-
sible outcomes of the measurement of L2 and Lz ? What are the corresponding
probabilities?
(b) What are the expectation values of L2 and Lz ?
(c) Determine the uncertainties ∆L2 and ∆Lz .

1
Physics 8.04, Quantum Physics 1, Spring 2016 2

4. Spherical wells. [10 points]

(a) Consider ℓ = 0 states of a particle moving in the infinite spherical well

(
0 , if r < a
V (r) =
∞ , if r > a .

Solve the radial equation for the radial wavefunction u(r) and find the possible
energy levels. Recall that u must vanish at r = 0. Try matching this ℓ = 0
spectrum to that of a one-dimensional infinite well potential V (x).
(b) Consider now states of a particle moving in a finite spherical well with V0 > 0:
(
−V0 if r < a
V (r) =
0 , if r > a .

Show that there is no bound state if

π 2 ~2
V0 a2 < .
8m

5. Hydrogen atom with total momentum. [10 points] Based on Ohanian.

When the motion of the nucleus is taken into account, the state of the hydrogen atom
can be represented by a wavefunction ψ(X, x), with X the center of mass coordinate
and x = xe − xp the relative coordinate pointing from the proton to the electron.
Suppose that the atom is in such a state that the total momentum has equal probabil-
ities for the values p0 and −p0 . Moreover the internal states are φ1,0,0 (x) or φ2,1,1 (x)
with probabilities 1/4 and 3/4 respectively (we use the notation φnℓm ). These prob-
abilities are not correlated with the total momentum.

(a) Write an expression for ψ(X, x) ignoring the overall phase but including arbi-
trary constant phase factors where possible.
(b) What is the expectation value for the total energy?
Physics 8.04, Quantum Physics 1, Spring 2016 3

6. Virial Theorem and applications [15 points]

(a) Consider any time independent operator Ω and the time derivative of its expec-
tation value, which is given by
d
i~ hΩi = [Ω,H ] ,
dt
where H is the Hamiltonian. Explain carefully why the right-hand side vanishes
if the system is in a stationary state.
(b) Now take Ω = r · p and show that for any stationary state of the hydrogen atom
Hamiltonian the following relation holds

hT i = − 12 hV i .

Here T is the kinetic energy operator and V is the potential energy operator.
(c) For any hydrogen atom eigenstate write hT i = 12 mhv 2 i, where m is quite accu-
rately the mass of the electron. Express the ratio
p
hv 2 i
c
e 2 1
in terms of the fine structure constant α = ~c ≃ 137 and the principal quantum
number n. Is the electron relativistic? Give the corresponding results for the
ground state when the nucleus has Z protons.
1
(d) What is r
in a general energy eigenstate of the hydrogen atom?

7. Exercises on hydrogen atom and some generalizations [10 points]

(a) Find hri and hr 2 i in the ground state of hydrogen. What is the most probable
value of r in the ground state?
(b) Assume that the nucleus of hydrogen has radius one femtometer. Calculate the
probability that the ground state electron is found inside the nucleus. Make
approximations to simplify your work and still get a very accurate answer!
(c) Positronium is a bound state of an electron and a positron (equal mass parti-
cles!). What are the energy levels? How does the size of positronium compare
with the size of a hydrogen atom?
MIT OpenCourseWare
https://ocw.mit.edu

8.04 Quantum Physics I

Spring 2016

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.