Problem 1

A cylindrical block of wood of density ρw , radius R, and height h is partially

immersed in a liquid of density ρl and then released, as shown in the figure.
a) What is the condition for the block to keep afloat?

b) What is the equilibrium height of the block above the water level zeq ?

c) If the block was initially slightly raised, so that z0 = z(t = 0) > zeq , and then
released, calculate z(t) assuming no friction.

d) Now assume that the liquid is viscous, and that the viscous force is propor-
tional to the velocity, as given by Fv = −bv. How is the motion of the block
modified? Write down the equation of motion.

e) What is the condition for the oscillatory motion to be over-damped?

Solution
A cylindrical block of wood of density ρw , radius R, and height h is partially immersed
in a liquid of density ρl and then released, as shown in the figure.

a) The block will float as long as its density does not exceed that of the liquid, ρw ≤ ρl .

b) At equilibrium the block is not moving so this is a statics problem, where the
condition that the vector sum of the external forces on the block is zero gives us what
we want. Let’s draw a free-body diagram for the block:

Here we have not included any pressure forces on the sides of the block, which cancel
by symmetry, and we have included the force due to the atmospheric pressure Pa on
the top and the force due to the liquid’s pressure at the bottom of the block, P (h − z).
Here m is the mass of the block. The pressure a distance h − z below the surface of
the liquid is
P (h − z) = Pa + ρl (h − z),
as can be seen by examining the additional force on an object of cross-sectional area
A due to the column of water of volume A(h − z) above it, and then dividing by the
area A. This means that the buoyant force due to the liquid, which is the vector sum
of the forces due to the air and the water, is

Fl = [P (h − z) − Pa ]A = ρl (h − z)Ag.

Therefore mg = ρl (h − zeq )A, which using m = ρw Ah gives

ρw Ahg = ρl (h − zeq )Ag

ρw
h − zeq = h
ρl
ρw
zeq = h(1 − ).
ρl
Note as ρw approaches ρl , we have that zeq approaches zero, as it should. The same
result follows from Archimedes principle: the magnitude of the buoyancy force is equal
to the weight of the displaced fluid.

c) To find k we need to find the restoring force F = −k∆z on the block when it is
raised a distance ∆z. This force is due to the fact that the buoyant force is lowered
when the block is raised, to

Fl = P (h − [zeq + ∆z])A = ρl (h − [zeq + ∆z])Ag,

so that the buoyant force is reduced by

∆Fl = ρl ∆zAg

so that
∆Fl
k= = ρl Ag,
∆z
and
k ρl Ag ρl g
ω02 = = = .
m ρw Ah ρw h
Therefore the motion of the block is simple harmonic motion with this frequency

∆z(t) = ∆z(0) cos(ω0 t).

z(t) = (z0 − zeq ) cos(ω0 t) + zeq .

d) Without knowing the value of b we cannot say whether the motion is underdamped,
critically damped, or overdamped. The equation of motion of the oscillator now be-
comes

m∆z̈ = −b∆ż − mω02 ∆z

 
b
∆z̈ + 2 ∆ż + ω02 ∆z = 0
2m
∆z̈ + 2β∆ż + ω02 ∆z = 0,
b b
where we have used the usual definition β ≡ 2m
= 2πρR2 h
.

If β <p ω0 the motion is underdamped and the system oscillates with frequency
ω1 = ω02 − β 2 and with exponentially decreasing amplitude with the form ∆z =
e−βt (A cos ω1 t + B sin ω1 t), where A and B are fixed by the initial condition.

If β = ω0 the motion is critically damped and the system does not oscillate but has
the form (A + Bt)e−βt , where A and B are fixed by the initial conditions.

If β > ω0 , the motion is overdamped andp the system does not oscillate but has the
form e−βt [Ae−ω2 t + Beω2 t ] where ω2 = β 2 − ω02 and again A and B are fixed by the
initial conditions.

e) From the above we have

β > ω0
r
b ρl g
>
2πρω R2 h ρω h
p
b > 2πR2 ρl ρω gh.
Problem 2
A uniform spherical ball of mass M and radius R is set rotating about a hori-
zontal axis with angular speed ω0 and is placed gently on the floor. The initial
center-of-mass velocity of the ball is zero. If the coefficient of sliding friction
between the ball and the floor is µ, find the speed of the center of mass of the
ball when it begins to roll without slipping. (The moment of inertia of the ball
about an axis passing through its center of mass is I = 25 M R2 .)

Solution
Is energy conserved? The ball will continue to rotate and so slip when it contacts the
floor, and the frictional force will cause it to accelerate. This means that energy is not
conserved and so we have to solve the problem using forces. First we must find the
value of the frictional force. Since the normal force is N = M g, we have

f = µN = µM g,

and as the ball is slipping this force points forward. This is the only horizontal force
and is the net force, since the normal force balances the gravitational force (or else the
ball would fall through the floor).
Newton’s law then tells us that
f
ẍ = = µg = constant,
M
where x is the center-of-mass position, and we see that as long as the ball is slipping
on the floor there is a constant acceleration of the center of mass, and the speed of
the center of mass is
v = ẋ = µgt.
The frictional force also causes a torque since it does not act through the center
of the ball,
τ = −Rf = −µM gR,
where the minus sign tells us that the torque acts to slow the rotation of the ball. The
angular equivalent of Newton’s law is
τ −µM gR 5 µg
θ̈ = ω̇ = = =− .
I 2M R2 /5 2 R
This means that the angular speed of the ball decreases linearly from its initial value
of ω0 according to
5 µgt
ω = ω0 − .
2 R
How do we know when the ball starts to roll without slipping? The condition that
it is not slipping is that the rate of linear motion of the center of mass is equal to ωR,
the speed at which the edge of the ball moves because of its rotation, so that
5
ωR = ω0 R − µgt = v = µgt
2
7
ω0 R = µgt
2
2 ω0 R
t = .
7 µg
and at this time
2
v = µgt = ω0 R
7
Problem 3
A large slab of dielectric material with a plane surface is inserted into an electric
field E, such that E is at an angle θ with the normal to the surface of the
dielectric. The dielectric constant of the dielectric is  = κ0 .

a) Find the magnitude of E and the magnitude of the displacement vector D

inside the dielectric.

b) Find a relationship between the dielectric constant and the angle E and D
make with the normal inside the dielectric. Sketch E and D inside the dielectric
assuming  = 30 and θ = 30◦ .

Solution
(a)

Referring to the above diagram, we have the boundary condition that the component
D⊥ of the electric displacement perpendicular to the surface between the air and the
dielectric is the same inside and outside the boundary (this can be found using Gauss’s
law). We also have, using the voltage drop around a small loop through the surface,
that the tangential component of the electric field is the same on both sides of the
boundary. So we have that

D1 cos(θ) = D2 cos(θi ) (1)

E1 sin(θ) = E2 sin(θi ) (2)
and also that D1 = 0 E1 , while D2 = E2 . These give that
E1 0
cos(θi ) = cos(θ) (3)
E2 
E1
sin(θi ) = sin(θ) , (4)
E2
so that by dividing these two equations we get

tan θi = tan(θ) = κ tan(θ).
0
This answers part (b). Now we can find E2 in terms of E1 = E using

sin(θ) sin(θ)
E2 = E1 =E
sin(θi ) sin(θi )

and we can find D2 in terms of D1 = 0 E by using

cos(θ)
D2 = 0 E1 .
cos(θi )

→ →
E1 D1
30° 30°

1 1

2 2
60° 60°
→ →
E2 D2

(b) See above. √

tan θi = 3 tan 30◦ = 3
Therefore
θi = 60◦

sin 30◦ E1
E2 = E1 = √
sin 60◦ 3
cos 30◦ √
D2 = D1 = 3D1
cos 60◦
Problem 4
A dielectric sphere of radius R with uniform dielectric constant  has a az-
imuthally symmetric density charge σ = σ0 cos θ placed on its surface. Outside
the sphere is vacuum.

(a) Obtain the electrostatic potential inside the sphere, φin .

(b) Obtain the electrostatic potential outside the sphere, φout .
(c) What is the electric field inside the sphere?

Solution
The general form of the electrostatic potential is
∞
X
φin = Al rl Pl (cos θ) ⇒ φin = A1 r cos θ (5)
l=0
∞
X 1
φout = (Bl rl + Cl r−l−1 )Pl (cos θ) ⇒ φout = C1 cos θ. (6)
r2
l=0

Ein = −∇φin , Eout = −∇φout . (7)

Boundary conditions:
continuity of transversal component of E:
1 ∂φin 1 ∂φout C1
(Ein − Eout ) × r̂ = 0 ⇒ − =− ⇒ A1 = 3 , (8)
R ∂θ R ∂θ R
discontinuity of radial component of D:
∂φin ∂φout
(Eout − Ein ) · r̂ = 4πσ ⇒  |r=R − = 4πσ
∂r ∂r
2C1 C1 4πσ0
A1 + = 4πσ0 ⇒ A1 = 3 = (9)
R3 R +2
(a)
4πσ0
φin = r cos θ. (10)
+2

(b)

4πσ0 R3
φout = cos θ (11)
 + 2 r2

(c)
4πσ0
Ein = − ẑ (12)
+2
Problem 5
A positive charge is uniformly distributed throughout a very long cylindrical
volume of radius R. The charge per unit volume is ρ.
~ everywhere as a function of the distance r from
(a) Find the electric field E
the axis of the cylinder.
(b) Find the electric potential V everywhere as a function of r. Define V = 0
at the surface of the cylinder.
(c) Sketch E and V as functions of r, from r = 0 to r = 3R, showing the values
of each at r = 0, R, and 3R.

Solution
(a) Let’s start by finding the electric field outside of the cylinder. Build a Gaussian
surface out of a cylinder of length L and radius r > R. Then by symmetry the
field must point radially outward on the curved part of the Gaussian surface
and must be parallel to the surface on the end caps. This means that the flux
of the electric field through this surface is
I
E · dA = E(2πrL).

Now Gauss’s law tells us that this is also Qencl /0 , where Qencl is the enclosed
charge. The volume of the charged cylinder cut out by our surface is V = πR2 L,
so the enclosed charge is
Qencl = πR2 Lρ
and Gauss’s law gives
1 ρR2
E(2πrL) = πR2 Lρ, E = .
0 20 r
Since the electric field points radially outward, we have
ρR2
E= r̂.
20 r
Inside the cylinder we use exactly the same technique, and everything goes as
before except now the charge enclosed by the Gaussian surface is reduced to
Qencl = (πr2 L)ρ
which means that the electric field is (take R → r in the above)
ρr
E= r̂.
20
(b) Starting again with the potential outside the cylinder, we know that the potential
will be the highest at the surface of the cylinder as the electric field always point
outward (and so a charged particle will lose energy as it moves away from the
cylinder). This means that
Z r0 Z r0  0
ρR2 ρR2 r
V (r0 ) = − E(r) · dr = − dr = − ln .
R R 2 0 r 20 R
 Inside the cylinder the potential increases as we move away from r = R, so we
have Z R Z R
ρr ρ
V (r0 ) = R2 − r02 .


E(r) · dr = dr =
r 0 r 0 2 0 4 0

(c) Note that the potential falls without bound as we move away from the wire,

essentially because the wire has an infinite charge.

Problem 6
Consider a two-dimensional infinite potential well of width L defined by the
potential

0 if 0 ≤ x ≤ L and 0 ≤ y ≤ L
V (x, y) = (13)
∞ otherwise

Assume that the well contains one particle of mass m, i.e. H = (p2x + p2y )/2m +
V (x, y).
(a) Obtain the ground state wave function(s) and energy. Is the ground state
degenerate?
(b) Obtain the wave function(s) and energy of the first excited state. Is the
first excited state degenerate?

Assume now that a perturbation of the form

H 0 = g sin (2xπ/L) sin (2yπ/L) (14)

is added to H.

(c) Calculate the first order perturbation to the ground state energy eigenvalue.
(d) Calculate the first order perturbation to the first excited state energy eigen-
value.

Solution
(a) Ground state wave function:

2 πx πy 1 h̄π
ψ00 = sin sin , E00 = ( )2 . (15)
L L L m L
Nondegenerate, no nodes.
(b) First excited state:

2 2πx πy 2 πx 2πy 5 πh̄

ψ10 = sin sin , ψ01 = sin sin , E10 = E01 = ( )2 . (16)
L L L L L L m L
Two folds degenerate, one line of nodes.
(c) First order perturbation for ground state:
Z L
2πx L
Z
2 πx 2 πy 2πy
hψ00 |H 0 |ψ00 i = g dx sin2 sin dy sin2 sin = (17)
0
0 L L L 0 L L L
(1) 1 h̄π 2
E00 = ( ) (18)
m L
(d) First order perturbation for excited states:
Z L Z L
2 2πx πx 2 2πy πy
hψ10 |H 0 |ψ01 i = hψ01 |H 0 |ψ10 i = g dx sin2 sin × dy sin2 sin
0 L L L 0 L L L
 Z π 2  2
8 8 4 16 32 2
= g dz(sin3 z − sin5 z) =g ( − ) = g( ) (19)
π 0 π 3 15 15π
hψ10 |H 0 |ψ10 i = hψ01 |H 0 |ψ01 i (20)
 2
(1) 5 πh̄ 32 2
E± = ± g( ) (21)
m L 15π

The degeneracy is lifted.

Problem 7
Three pointlike particles of mass m are confined to move on a circle of radius
r. Their mutual distances are fixed and equal(i.e. they lie on the corners of
an equilateral triangle). Determine the rotational energy levels of the system,
assuming that the particles have spin = 0, i.e. they obey Bose statistics.

Solution
The only possible motions of the mass points are along the circle, i.e. rotations of the
system around an axis through the center of the circle and perpendicular to the plane.
The Hamiltonian of this system is

L2z
H= , (22)
2I
where I is the moment of inertia, I = 3mr2 , and we have assumed that the rotation axis
as z-axis. In polar coordinates Lz = h̄i ∂φ
∂
, where φ is the azimuthal angle (coordinate
along the circle). The eigenfunctions of this Hamiltonian are ei`φ , where ` is an integer.
2 2
The energy levels are E = h̄2I` . Since the particles are bosons, the wavefunction must
be symmetric with respect to the exchange of any of the two particles, i.e. invariant
under rotation by ±120 = ± 2π 3
:

2π
ψ(φ ± ) = ψ(φ). (23)
3
Hence ` · 2π3
must be an integer multiple of 2π, = n · 2π, whence ` = 3n. The energy
levels of the system are therefore

3n2 h̄2
En = (24)
2mr2
Problem 8
Neutron scattering is often done by cooling fast neutrons, which are among
the fission products from a nuclear reactor, by thermalizing them in a moder-
ator such as solid deuterium oxide (D2 O ice) held at some constant cryogenic
temperature T . The average kinetic energy of neutrons from such a source is,
1 3
E= mhv 2 i = kT ,
2 2
where m is the neutron mass, hv 2 i is the mean-square velocity, and k is Boltz-
mann’s constant.
(a) What is the de Broglie wave length λn , for neutrons, in terms of their
energy, mass, and Planck’s constant?
(b) What is the de Broglie wave length for neutrons with the average kinetic
energy in terms of the temperature of the source?
(c) Hydrogen freezes at 20.25 K and can be used to keep the D2 O moderator
at a fixed temperature. What is the de Broglie wave length λn (in Å)
for neutrons with the average energy corresponding to that temperature?
What is the average neutron energy E (in eV)?
(d) The neutrons are thermalized at the 20.25 K and diffracted off a crystalline
target with lattice spacing d = 4.21 Å. Considering neutrons being nearly
monochromatic at that temperature, what is the scattering angle θ that
would give Bragg diffraction of the neutrons from this crystal? Use the
convention for diffraction shown in the figure.
The neutron mass is mn = 1.68 × 10−27 kg.

Figure 1: Bragg diffraction of the neutrons from the crystal.

Solution
(a) The de Broglie wave length is λ = h/p = h/(mv), and since E = mv 2 /2 (note we
are assuming
p v << c so we can use a nonrelativistic approximation here) then
v = 2E/m and so
h h h
λ= = p = √ .
mv m 2E/m 2mE
 !

! d

d sin !

Figure 2: Scattering path difference for neutrons

(b) Since E = 3kT /2 we have

h h
λ= p = √ .
m 2E/m 3mkT

(c) Substitute into the answer from (b),

6.6 × 10−34 Js
λ= p = 5.57×10−10 m = 5.57 Å.
3 · 1.68 × 10−27 kg · 1.38 × 10−23 J/K · 20.25 K

Also,
3 1.38 × 10−23 J/K · 20.25 K
E= = 0.0026 eV.
2 1.6 × 10−19 J/eV
(d) Referring to the diagram, we see that the path difference for neutrons which scat-
ter off adjacent planes in the crystal is just 2d sin(θ), so that we get constructive
interference when this is an integral number of de Broglie wavelengths, i.e.

nλ = 2d sin(θ),

and so we have
nλ 5.57Å
sin(θ) = =n = 0.661 n,
2d 8.42Å
and we see that n = 1 is the only possibility and there θ = sin−1 (0.661) = 41.4o .
Problem 9
This problem uses the variational principle to calculate an upper bound for
the ground state energy of a one dimensional quantum particle confined to a
potential given by
∞ f or x < 0
V (x) = (25)
λx f or x > 0
This potential is sketched in the figure below.

Use a function
R∞ proportional to xe−αx as the variational wave function. The
n −βx n!
integral 0 dxx e = β n+1 may be of use in solving this problem.
(a) What is the normalization factor for this wave function?
(b) Find an upper bound for the ground state energy.
(c) Copy the shape of the potential function in your answer book and draw a
line to indicate the ground state energy. On the same graph sketch the
shape of the wave function.
(d) Now extend this problem to 3 dimensions in the following manner: In
the y and z directions the particle is free to move anywhere between ±∞ ,
without forces. What are the quantum numbers of the energy eigenstates?
Which are continuous and which are discrete? Ignore spin.

Solution
(a) To find the normalization constant we do
hψ|ψi = 1
Where ψ is the trial function
ψ = Axe−αx
then Z ∞
A2 x2 e−2αx dx = 1
0
Using the integral given above we have
A2
=1
4α3
Solving for A we get that the normalization constant is
A = 2α3/2
(b) To Calculate the upper bound we use the variational principle, which tell us

hψ|H|ψi ≥ E0

The Hamiltonian is given by

−h̄2 ∂ 2
Ĥ = + λx
2m ∂x2

∞
−h̄2 ∂ 2
Z  
4α3 xe−αx 2
+ λx xe−αx dx ≥ E0
0 2m ∂x
Applying the Hamiltonian to the function we have
Z ∞  2
h̄2 2 −αx

h̄
4α3 xe−αx αe−αx − α xe + λx2 e−αx dx ≥ E0
0 m 2m
Z ∞ 2 2

3 h̄ −2αx h̄ 2 2 −2αx 3 −2αx
4α αxe − α x e + λx e dx ≥ E0
0 m 2m
Using again the given integral we have
 2
h̄2 1

h̄ 1 3λ
4α3 − + ≥ E0
2m 2α 8m α 8α4
Simplifying we get
h̄2 2 3λ
α + ≥ E0
2m 2α
Now to get the minimum we take the derivative with respect to α and make
that equal to zero
h̄2 3λ
α− =0
m 2α2
solving for alpha we get
 1/3
3mλ
α=
2h̄2
Plugging this back we have
2/3 1/3
h̄2 3mλ 3λ 2h̄2
 
+ ≥ E0
2m 2h̄2 2 3mλ
simplifying we have
1/3  2/3
h̄2

3 3λ
≥ E0
2 m 2
(c)
(d) In the y and z directions we have a free particle so the quantum numbers are the
wave numbers ky and kz , and they are continuous. For x we have the quantum
number n and it is discrete.
Problem 10
Consider an ideal gas of N electrons that move freely in 2D, and are confined
to an area A.
(a) Estimate the temperature T0 below which quantum mechanical considera-
tions are important.
(b) Calculate the Fermi energy assuming the system is at the temperature
T = 0K. Compare the Fermi energy to kB T0 , where T0 is calculated above.
(c) (c) What is the physical interpretation of the Fermi energy for a given
system?

Solution
(a) The area of a 2D system of N electrons is: A ∼ N π(∆x)2. Therefore ∆x ∼
(1/π)(A/N )1/2 . Using the uncertainty relation: ∆x∆p ∼ h̄, we can then write:
∆p ∼ h̄/∆x.
To estimate the temperature below which quantum effects become important,
we will use:

E ∼ kB T0 ∼ (∆p)2 /2m ∼ h̄2 /(2m(∆x)2 ) ∼ h̄2 N π 2 /(2mA). (26)

Therefore, as a crude approximation: T0 ∼ h̄2 N π 2 /(2mAkB ).

(b) Calculate the 2D Fermi energy to compare to the energy scale in part (a):

ZkF ZkF ZkF

d2 k 2πkdk 4πA A
N = 2A =2A = kdk = kF 2 (27)
(2π)2 (2π)2 (2π)2 2π
0 0 0

The factor of 2 is to account for the spin of the electron. The Fermi energy is
therefore:
h̄2 kF 2 h̄2 2πN πh̄2 N
 
EF = = =
2m 2m A mA
The Fermi temperature T0 is then:

h̄2 N
 
π
T0 =
kB mA

Note that this is close to the crude result estimated in part (a).
(c) The Fermi energy is a measure of the energy scale where quantum effects become
important for describing the properties of a system of Fermions. For most metals
at room temperature, for example, the Fermi energy is very large, and therefore
Fermi-Dirac statistics are needed to describe the properties of the system as
opposed to a more classical Maxwell-Boltzmann approach.
Problem 11
One mole of an ideal gas undergoes a reversible thermal expansion from volume
V to volume 2V .
(a) What is the change in entropy of the gas?

(b) What is the change in entropy of the universe?

Now suppose the same expansion takes place as a free expansion.
(c) What is the change in entropy of the gas?

(d) What is the change in entropy of the universe?

Solution
(a) Since the process is isothermal, the internal energy does not change. Thus the
work done during the expansion must be equal to the heat absorbed by the gas.
Z 2V Z 2V
RT
∆Q = W = P dV = dV = RT ln 2. (28)
V V V
The entropy change of the gas is then
∆Q
S= = R ln 2. (29)
T

(b) The entropy change of the heat bath the gas is coupled to when expanding is
∆Sbath = −∆Q/T = −R ln 2, and so the total entropy change of the universe is
zero,

∆Su = ∆Sgas + ∆Sbath = 0, (30)

as it must be for a reversible process.

(c) We recall that free expansion is an irreversible process in which a gas expands
into an insulated evacuated chamber. While for real gases the temperature may
change in the process, for ideal gases free expansion takes place at a constant
temperature. This is easy to see: in an ideal gas the molecules do not interact
and their average kinetic energy remains 3/2 kT (for monoatomic gas). For a
free expansion there is again no change in the internal energy of the gas, so the
initial and final states of the gas are the same as in part (a). The change in
entropy of the gas is thus again

∆Sgas = R ln 2. (31)

(d) For a free expansion the state of the bath does not change and so ∆Sbath = 0.
Thus for free expansion the total change in entropy of the universe is

∆Su = ∆Sgas + ∆Sbath = R ln 2 (32)

reflecting that free expansion is not reversible.

Problem 12
Give approximate numerical values for the following (be sure to give units):
(a) The binding energy of a nucleon in a typical nucleus.

(b) The binding energy of an electron in a hydrogen atom.

(c) The lifetime of a free neutron.
(d) The specific heat of Cu at room temperature (assumed greater than the
Debye temp.)

(e) The energy of a photon of light within the visible part of the spectrum.

Solution
(a) The binding energy of a typical nucleon is on the order of ∼ MeV.
(b) The binding energy of an electron within a hydrogen atom is ∼ 13.6 eV (Bohr
model of the atom).
(c) The lifetime of a free neutron is ∼ 15 minutes. This is roughly the time it takes
for the conversion of a down quark to an up quark through the weak interaction.
(d) The specific heat of Cu at room temperature is ∼ 3R (Dulong-Petit law)
(e) The energy of a photon of light within the visible part of the EM spectrum is ∼
eV (E = hf = hc/λ, with λ ∼ 600 nm).