Physics 2102B - 1st Assignment

Your solution to problems 6, 7, and 8 has to be handed in, in class, on Monday,

February 3, 2013.

1. When considering the Maxwell-Boltzmann distribution it was assumed to have the

following form

f (v ) = α e
( )
−m vx2 +vy2 +vz2 2 kT
, (1.1)

where α is some normalization constant.

a) Consider the Gaussian integral

∞
I = ∫ e− β x dx,
2
(1.2)
−∞

with β another constant. Now take the square of this integral (i.e., consider I 2 ), change
from Cartesian to polar coordinates, solve the integral for I 2 , and then show that

π
I= . (1.3)
β

b) Use your result from a) to show in equation (1.1)

32
⎛ m ⎞
α = N⎜ (1.4)
⎝ 2π kT ⎟⎠
,

if we also require that

∞ ∞ ∞
N=∫ ∫ ∫ f ( v ) dvx dvy dvz , (1.5)
−∞ −∞ −∞

with N the total number of particles in a system that verifies the Maxwell-Boltzmann
distribution.

c) Show that

1 ⎛ 2π kT ⎞
32
∞
∫
2 −mvx2 2 kT
ve dvx = ⎜ ⎟ . (1.6)
2π ⎝ m ⎠
x
−∞

You may want to consider integrating by parts to solve this integral.

1
Solution.

a) From equation (1.2)we can write

I2 = (∫ −∞
∞
e− β x dx
2

)( ∫ ∞

−∞
e− β y dy
2

) (1.7)
=∫
∞ ∞ ( − β x 2 +y 2 ) dx dy.
−∞ −∞∫ e

We now make the change of variables

x = r cos (θ )
(1.8)
y = r sin (θ )

which implies that dx dy = rdrdθ . Equation (1.7) then becomes

2π ∞
I 2 = ∫ dθ ⋅ ∫ e− β r r dr
2

0 0
∞
⎛ 1 − βr2 ⎞
= 2π ⋅ ⎜ − e ⎟ (1.9)
⎝ 2β ⎠0
π
= ,
β

which implies that

π
I= . (1.10)
β

b) It becomes apparent from equations (1.1) to (1.3) that

α e ( x y z ) dvx dvy dvz

∞ ∞ ∞ −m v 2 +v 2 +v 2 2 kT
N=∫ ∫ ∫
−∞ −∞ −∞

= α ⎡⎣ I ( β = m 2kT ) ⎤⎦
3
(1.11)
⎛ 2π kT ⎞
32

=α⎜
⎝ m ⎟⎠
,

and
32
⎛ m ⎞
α = N⎜ (1.12)
⎝ 2π kT ⎟⎠
.

2
c) We define

∞
I1 = ∫ vx2 e−mvx
2
2 kT
dvx
−∞
∞
⎛ kT −mvx2 ⎞ kT ∞
∫ e−mvx
2
= vx ⋅ ⎜ − e 2 kT
⎟⎠ +
2 kT
dvx
⎝ m −∞
m −∞

(1.13)
⎛ kT ⎞
= 0 + ⎜ ⎟ I ( β = m 2kT )
⎝ m⎠
1 ⎛ 2π kT ⎞
32

= ⎜ ⎟ ,
2π ⎝ m ⎠

∞
where we integrated by parts using I1 = uv −∞ − ∫ v du with
∞
−∞

u = vx , du = dvx
kT −mvx2 (1.14)
dv = vx e−mvx
2
2 kT
dvx , v=− e 2 kT

and we used equation (1.10).

2. Consider equation (1.24) of Chapter 1 of the lecture notes for the one-dimensional
wave equation for the electric field

∂ 2 Ez 1 ∂ 2 Ez
− = 0. (2.1)
∂x 2 c 2 ∂t 2

a) Let us now study how this wave equation changes under a Galilean transformation

t = t′
x = x ′ − vt ′ (2.2)
= x ′ − vt,

where x′ and t ′ are the spatial and temporal coordinates attached to an inertial reference
frame K ′ moving at a velocity v relative to another inertial frame K to which x and t
are attached. First use the chain rule to show that

∂ ∂ ∂
= +v
∂t ∂t ′ ∂x′
(2.3)
∂ ∂
= .
∂x ∂ x ′

3
Second, derive corresponding equations for ∂ 2 ∂t 2 and ∂ 2 ∂x 2 , and then express
equation (2.1) in K ′ . Is the wave equation covariant (or invariant) under a Galilean
transformation? Are Maxwell’s equations covariant under a Galilean transformation?

b) Now let us consider a Lorentz transformation

⎛ v ⎞
t ′ = γ ⎜ t + 2 x⎟
⎝ c ⎠ (2.4)
x′ = γ ( x + vt ) ,

where γ = (1− v 2 c 2 )
−1 2
is the Lorentz factor. Derive the needed relations for ∂ 2 ∂t 2 and
∂ 2 ∂x 2 , and express equation (2.1) in K ′ . Is the wave equation covariant (or invariant)
under a Galilean transformation? Are Maxwell’s equations covariant under a Galilean
transformation?

[Note: To a good approximation the electric field is the same in the two referential
frames, i.e., Ez  Ez′ , for both types of transformation when v is small relative to c (i.e.,
when γ  1 .]

Solution.

a) The chain rule states that

∂ ∂t ′ ∂ ∂ x ′ ∂
= +
∂t ∂t ∂t ′ ∂t ∂ x ′
∂ ∂
= +v
∂t ′ ∂x′
(2.5)
∂ ∂t ′ ∂ ∂ x ′ ∂
= +
∂x ∂x ∂t ′ ∂x ∂ x ′
∂
=
∂x′

from equations (2.2). We then have

∂2 ⎛ ∂ ∂ ⎞⎛ ∂ ∂ ⎞
=⎜ +v ⎟ ⎜ +v ⎟
∂t 2
⎝ ∂t ′ ∂ x ′ ⎠ ⎝ ∂t ′ ∂x′ ⎠
(2.6)
∂2 ∂ ∂ ∂2
= 2 + 2v + v2
∂t ′ ∂t ′ ∂ x ′ ∂x′2

and

4
 ∂2 ∂ ⎛ ∂ ⎞
= ⎜ ⎟
∂x 2
∂x′ ⎝ ∂x′ ⎠
(2.7)
∂2
= .
∂x′2

If we now apply equations (2.6) and (2.7) to (2.1) we find that

⎛ ∂2 1 ∂2 ⎞ ⎛ ∂2 1 ∂2 v ∂ ∂ v2 ∂2 ⎞
⎜⎝ ∂x 2 − c 2 ∂t 2 ⎟⎠ Ez = ⎜⎝ ∂ x ′ 2 − c 2 ∂t ′ 2 − 2 c 2 ∂t ′ ∂ x ′ − c 2 ∂ x ′ 2 ⎟⎠ Ez

⎡ ∂2 ⎛ v2 ⎞ 1 ∂2 v ∂ ∂ ⎤
= ⎢ 2 ⎜ 1− 2 ⎟ − 2 −2 2 ⎥ Ez (2.8)
⎣ ∂ x ′ ⎝ c ⎠ c ∂t ′ c ∂t ′ ∂ x ′ ⎦
2

⎛ ∂2 1 ∂2 ⎞
≠⎜ − Ez .
⎝ ∂ x ′ 2 c 2 ∂t ′ 2 ⎟⎠

Since we have an inequality for the last equation, it follows that the wave equation
changes mathematical form when subjected to a Galilean transformation and is therefore
not covariant in that respect. The same conclusion applies to Maxwell’s equations since
the wave equation was derived from them. That is, Maxwell’s equations are not covariant
under a Galilean transformation.

b) For a Lorentz transformation we have

∂ ∂t ′ ∂ ∂ x ′ ∂
= +
∂t ∂t ∂t ′ ∂t ∂ x′
⎛ ∂ ∂ ⎞
=γ ⎜ +v ⎟
⎝ ∂t ′ ∂ x′ ⎠
(2.9)
∂ ∂t ′ ∂ ∂ x ′ ∂
= +
∂x ∂x ∂t ′ ∂x ∂ x′
⎛ v ∂ ∂ ⎞
=γ ⎜ 2 +
⎝ c ∂t ′ ∂ x′ ⎟⎠
.

It then follows that

∂2 ⎛ ∂ ∂ ⎞⎛ ∂ ∂ ⎞
=γ 2⎜ +v ⎟⎠ ⎜⎝ +v ⎟
∂t 2
⎝ ∂t ′ ∂ x ′ ∂t ′ ∂ x′ ⎠
(2.10)
⎛ ∂2 ∂2 2 ∂
2
⎞
= γ ⎜ 2 + 2v
2
+v 2⎟
⎝ ∂t ′ ∂ x ′ ∂t ′ ∂ x′ ⎠

and

5
 ∂2 ⎛ v ∂ ∂ ⎞⎛ v ∂ ∂ ⎞
=γ 2⎜ 2 + +
∂x 2
⎝ c ∂t ′ ∂ x′ ⎠ ⎝ c ∂t ′ ∂ x′ ⎟⎠
⎟ ⎜ 2

(2.11)
⎡⎛ v ⎞ 2 ∂2 v ∂2 ∂2 ⎤
= γ ⎢⎜ 2 ⎟
2
+ 2 + 2 ⎥
.
⎣ ⎝ c ⎠ ∂ t ′ 2
c 2
∂ x ′ ∂ t ′ ∂ x ′ ⎦

Using these last two equations, we can write

∂2 1 ∂2 ⎡
2 ⎛ v2 ⎞ ∂2 ⎛ v2 ⎞ 1 ∂2 ⎤
− = γ ⎢⎜ 1− 2⎟
− ⎜ 1− 2 ⎟ 2 ⎥
∂x 2 c 2 ∂t 2 ⎣ ⎝ c ⎠ ∂ x ′ 2
⎝ c ⎠ c ∂t ′ 2 ⎦
(2.12)
∂2 1 ∂2
= − ,
∂ x ′ 2 c 2 ∂t ′ 2

since γ 2 = (1− v 2 c 2 ) . It follows that both the wave equation and Maxwell’s equations
−1

are covariant under a Lorentz transformation (remember that Ez = Ez′ ).

3. A schematic of a diffraction grating is shown in Figure 1, where a plane wave is

incident on the left side and propagates in the x direction towards a detector located at a
very far distance (practically at infinity) on the right. A number of N infinitesimally
narrow slits (identified with n = 0, 1, 2, … , N − 1 ) are spaced a distance d apart in the y
direction, as shown, with each of them extending infinitely in the z direction. These slits
let part of the plane wave through. As can be seen in the figure, when considering
propagation to the detector in a direction at angle θ from the x -axis the contributions
emanating from the slits travel different distances depending on their position on the
grating.

a) Model the incident electric field associated with the plane wave with

Figure 1 – Schematic of a diffraction grating.

6
 E = E0 e− j( kx−ω t )e z , (3.1)

where k = ω c is the wave number and ω the angular frequency. Show that the
amplitude of the electric at the detector will, up to a global phase factor, scale as

sin ⎡⎣ Nkd sin (θ ) 2 ⎤⎦

Ed ∝ E0 . (3.2)
sin ⎡⎣ kd sin (θ ) 2 ⎤⎦

b) Show that the intensity of the detected electric field will peak when

d sin (θ ) = mλ , (3.3)

where m is an integer and λ the wavelength of the incident wave.

Solution.

a) If the wave emanating through the first slit located at n = 0 travels a distance L to the
detector, then contributions coming from the other slit will travel L + nd sin (θ ) . The
diffracted field at the detector will then be proportional to

N −1
Ed ∝ E0 e− jkL ∑ e− jnkd sin(θ ) . (3.4)
n=0

We now calculate the following

N −1
Ed ⎡⎣1− e− jkd sin(θ ) ⎤⎦ ∝ E0 ∑ ⎡⎣ e− jnkd sin(θ ) − e− j( n+1)kd sin(θ ) ⎤⎦
n=0

∝ E0 ⎡⎣1− e− jNkd sin(θ ) ⎤⎦ (3.5)

∝ E0 e− jNkd sin(θ ) 2 ⋅ 2 j ⋅sin ⎡⎣ Nkd sin (θ ) 2 ⎤⎦ ,

and therefore

sin ⎡⎣ Nkd sin (θ ) 2 ⎤⎦

Ed ∝ E0 e− j( N −1)kd sin(θ ) 2 . (3.6)
sin ⎡⎣ kd sin (θ ) 2 ⎤⎦

We thus find that, up to a global phase factor,

sin ⎡⎣ Nkd sin (θ ) 2 ⎤⎦

Ed ∝ E0 . (3.7)
sin ⎡⎣ kd sin (θ ) 2 ⎤⎦

7
b) The intensity is proportional to the square of the electric field, and at the detector it
will be proportional to the square of equation (3.7). This signal will then be at a
maximum when the denominator goes to zero, at which point the numerator is also zero.
Using L’Hôpital’s rule we have

lim Ed ∝ NE0 . (3.8)

kd sin(θ ) 2→mπ

The condition for this to happen is therefore

1 d
kd sin (θ ) = π sin (θ )
2 λ (3.9)
= mπ

or

d sin (θ ) = mλ , (3.10)

with m is an integer called the order number.

4. Start with Planck’s blackbody radiation law expressed as a function of the wavelength

2π c 2 h 1
F(λ ,T ) = hc λ kT
(4.1)
λ e
5
−1

and derive the corresponding law written as a function of the frequency ν = c λ instead.

Solution.

Since the energy in a given wavelength band d λ must be the same as that contained in
the corresponding frequency band

dν
dν = dλ
dλ
d (c λ )
= dλ (4.2)
dλ
c
= dλ,
λ2

we must have F (ν ,T ) dν = F ( λ ,T ) d λ . We then write

8
 dλ
F (ν ,T ) = F ( λ ,T )
dν
2π c 2 h 1 λ2
= ⋅ (4.3)
λ 5 ehc λ kT − 1 c
2π hν 3 1
= hν kT
.
c 2
e −1

5. (Prob. 51, Ch. 3, in Thornton and Rex.) Derive the relation for the recoil kinetic
energy of the electron and its recoil angle φ in Compton scattering. Show that

Δλ λ
Ke = hν
1+ Δλ λ
(5.1)
⎛ hν ⎞ ⎛θ ⎞
cot (φ ) = ⎜ 1+ 2 ⎟ tan ⎜ ⎟ .
⎝ mc ⎠ ⎝ 2⎠

Solution.

By conservation of energy we know the electron’s recoil energy equals the energy lost by
the photon, which with λ ′ = λ + Δλ yields

hc hc
Ke = −
λ λ′
hcΔλ
=
λ ′λ
(5.2)
hcΔλ λ
=
λ (1+ Δλ λ )
Δλ λ
= hν .
1+ Δλ λ

Using the last two of equations (2.48) from the lecture notes for the conservation of linear
momentum

h h
= cos (θ ) + pe2 cos (φ )
λ λ′
(5.3)
h
0 = sin (θ ) − pe2 sin (φ ) ,
λ′

we find when dividing the first by the second

9
 λ ′ λ − cos (θ )
cot (φ ) =
sin (θ )
(5.4)
Δλ λ + 1− cos (θ )
= .
sin (θ )

But using the result for Δλ derived in the lecture notes for the Compton effect (i.e.,
equation (2.52)

h
Δλ = ⎡1− cos (θ ) ⎤⎦ , (5.5)
mc ⎣

and inserting it in the last of equations (5.4) gives us

⎛ h ⎞ ⎡ 1− cos (θ ) ⎤
cot (φ ) = ⎜ + 1⎟ ⎢
⎝ mcλ ⎠ ⎣ sin (θ ) ⎥⎦

⎛ hv ⎞ ⎡ 2sin 2 (θ 2 ) ⎤
= ⎜ 1+ 2 ⎟ ⎢ ⎥ (5.6)
⎝ mc ⎠ ⎣ 2sin (θ 2 ) cos (θ 2 ) ⎦
⎛ hv ⎞
= ⎜ 1+ 2 ⎟ tan (θ 2 ) ,
⎝ mc ⎠

where we used λ = c ν and

1
sin ( a ) sin ( b ) = ⎡ cos ( a − b ) − cos ( a + b ) ⎤⎦
2⎣
(5.7)
1
sin ( a ) cos ( b ) = ⎡⎣sin ( a − b ) + sin ( a + b ) ⎤⎦ .
2

6. (Prob. 28, Ch. 3, in Thornton and Rex.) We have waves in a one-dimensional box,
such that the wave displacement ψ ( x,t ) = 0 for x = 0 and x = L , where L is the length
of the box, and

∂ 2ψ 1 ∂ 2ψ
− = 0. (6.1)
∂x 2 c 2 ∂t 2

a) Show that the solutions for equation (6.1) are of the form

⎛ nπ x ⎞
ψ ( x,t ) = a ( t ) sin ⎜ , n = 1, 2, 3, … (6.2)
⎝ L ⎟⎠

10
and a ( t ) satisfies the (harmonic oscillator) equation

d 2a (t )
+ ω n2 a ( t ) = 0, (6.3)
dt 2

where ω n = nπ c L is the angular frequency.

b) Consider a general solution for equation (6.3) and show that ψ ( x,t ) is composed in
families of waves travelling in opposite directions. Clearly explain why the waves travel
in a given direction.

Solution.

a) We start by taking spatial and time derivatives of equation (6.2) with

∂ 2ψ ⎛ nπ ⎞ ⎛ nπ x ⎞
2

= −a ( t ) ⎜ ⎟ sin ⎜
∂x 2
⎝ L⎠ ⎝ L ⎟⎠
(6.4)
∂ 2ψ d 2 a ( t ) ⎛ nπ x ⎞
= sin ⎜
⎝ L ⎟⎠
,
∂t 2 dt 2

which when inserted in the wave equation yield

d 2a (t )
+ ω n2 a ( t ) = 0, (6.5)
dt 2

with ω n = nπ c L .

b) Equation (6.5) is a typical linear second-order differential equation, which allows any
or all of the following functions as solutions: sin (ω nt ) , cos (ω nt ) , and e ± jω nt . Choosing
the sine function as an example, we find that

⎛ nπ x ⎞
ψ ( x,t ) = ∑ An sin (ω nt ) sin ⎜
n
⎝ L ⎟⎠
1 ⎡ ⎛ nπ x ⎞ ⎛ nπ x ⎞⎤
= ∑
2 n
An ⎢ cos ⎜
⎣ ⎝ L
− ω nt ⎟ − cos ⎜
⎠ ⎝ L
+ ω nt ⎟ ⎥
⎠⎦
(6.6)

1
= ∑ An ⎡⎣ cos ( kn x − ω nt ) − cos ( kn x + ω nt )⎤⎦,
2 n

where we have introduced the wave number kn = ω n c = nπ L for convenience. We can

now define the phases of the two families of cosine functions as

11
 φn± ( x,t ) = kn x ± ω nt. (6.7)

If we concentrate on the spatial location of a specific maximum on the ‘ + ’ wave where

φn+ ( x,t ) = 0 (i.e., when cos ⎡⎣φn+ ( x,t ) ⎤⎦ = 1 ), we then find that

ωn
x= t = ct, (6.8)
kn

which implies that this maximum on the wave is travelling at the speed c in the positive
x -direction as time passes on. A similar calculation for the ‘ − ’ wave reveals that
x = −ct , and the waves travels in the negative x -direction at speed c as time passes on.

7. Consider an ideal gas at temperature T composed of molecules of mass m and of

number density (i.e., number per volume) n .

a) Assume that the gas is contained within a box and show that the number of molecules
that strike a surface of area A on a wall in a time interval Δt is approximately given by
nA kT m dt 2 , with k the Boltzmann constant. To simplify this calculation, consider
that the wall is perpendicular to the x -axis and make the radical assumption that the x
components of the velocities of all molecules have the same magnitude. That is, assume
that all molecules have the same vx , with half the molecules moving to the right and the
other half moving to the left. Consider that the equipartition of energy applies for this
gas.

b) Abandon the “radical assumption” of part a) and use instead a Maxwell-Boltzmann

distribution for vx (i.e., see equation (1.1) above, but simplified to the one-dimension
case) to obtain a more exact solution to the problem of part a).

c) For molecular oxygen at the so-called standard temperature and pressure (STP:
T = 273 K and p = 1 atm = 1.01× 10 5 N/m 2 ), use the result of part b) to calculate the
average number density of particles that strike a 1 cm 2 surface in 1 s?

Solution.

a) Under the suggested “radical assumption,” with all molecules having the same vx and
half of them moving each way, the number density of molecules moving to the right is
n 2 . In a time interval Δt , these molecules move a distance vx Δt to the right. Therefore,
the molecules that hit the wall are precisely those right-moving molecules in the volume
to the right of the dashed line in the accompanying figure, with length vx Δt and area A .
The number of such molecules in this volume is nAvx Δt 2 .

12
Finally, since there is one degree of freedom associated with motions along the x -axis
equipartition of energy implies that

1 2 1
mvx = kT , (7.1)
2 2

which upon insertion in the previous equation gives nA kT m dt 2 for the number of
molecules that strike a surface of area A on a wall in a time interval Δt .

b) The Maxwell-Boltzmann distribution for this problem is

m −mvx2
f ( vx ) = N e 2 kT
, (7.2)
2π kT

for −∞ < vx < ∞ , and the number of molecules having a velocity ranging from vx to
vx + dvx being f ( vx ) dvx (as always, N is the total number of particles in the gas). The
number N w of these that hit the wall on the right within a time Δt is the number in the
volume to the right of the dashed line in the figure and is found to be

∞ Avx Δt
Nw = ∫ f ( vx ) dvx , (7.3)
0 V

with V the total volume of the box. Realizing that n = N V we write

m ∞
∫ vx e−mvx
2
N w = nAΔt 2 kT
dvx
2π kT 0

∞
m ⎛ kT −mvx2 ⎞
= nAΔt ⎜⎝ − e 2 kT
⎟⎠ (7.4)
2π kT m 0

kT
= nAΔt .
2π m

c) We use the ideal gas law for this calculation, which states that

P = nkT . (7.5)

Inserting this relation in equation (7.4) we have

13
 pAΔt
Nw =
2π kT
1.01× 10 5 N/m 2 ⋅10 −4 m 2 ⋅1 s
= (7.6)
2π ⋅1.38 × 10 −23 J/K ⋅ 273 K ⋅ ( 32 × 1.66 × 10 −27 kg )
= 2.8 × 10 23.

8. (Prob. 60, Ch. 3, in Thornton and Rex.) a) Show that the maximum kinetic energy of
the recoil electron in Compton scattering (refer to Figure 8 in Chap. 2 of the Lecture
Notes) is given by

⎛ 2hν mc 2 ⎞
K e,max = hν ⎜ , (8.1)
⎝ 1+ 2hν mc 2 ⎟⎠

where ν is the frequency of the incident photon. At what angles θ and φ does this
occur?

b) If we detect a scattered electron of 100 keV at φ = 0 , what was the energy of the
incident photon?

Solution.

a) We find from the equation

h
Δλ = ⎡1− cos (θ ) ⎤⎦ (8.2)
mc ⎣

that the maximum change in photon wavelength, which will coincide with a maximum
electron recoil, happens when θ = π . Then from the equation derived in Problem 5

⎛ hν ⎞ ⎛θ ⎞
cot (φ ) = ⎜ 1+ 2 ⎟ tan ⎜ ⎟ (8.3)
⎝ mc ⎠ ⎝ 2⎠

the corresponding electron recoil angle must be φ = 0 . This is because we then have
cot (φ ) = tan (θ 2 ) = ∞ , which is a necessary condition for equation (8.3) to remain valid.
At any rate, we have in this case

2h
Δλ = . (8.4)
mc

14
Using another equation derived in Problem 5 for the kinetic energy of the recoil electron
we write

Δλ λ
Ke = hν
1+ Δλ λ
(8.5)
2hν mc 2
= hν ,
1+ 2hν mc 2

as required.

b) We can rearrange equation (8.5) as follows

⎛ 2hν ⎞ 2 ( hν )
K e,max ⎜ 1+ 2 ⎟ = (8.6)
⎝ mc ⎠ mc 2

or

⎛ 2 ⎞ ⎛ 2K e,max ⎞
⎜⎝ 2 ⎟⎠ ( hν ) − ⎜⎝ ( hν ) − K e,max = 0.
2
(8.7)
mc mc 2 ⎟⎠

This last equation is quadratic in hν that can be solved with K e,max = 100 keV to yield

2
⎛ 2K e,max ⎞ ⎛ 2K e,max ⎞ ⎛ 2 ⎞
⎜⎝ mc 2 ⎟⎠ ± ⎜⎝ mc 2 ⎟⎠ + 4K e,max ⎜⎝ mc 2 ⎟⎠
hν = , (8.8)
⎛ 2 ⎞
2⎜ 2 ⎟
⎝ mc ⎠

for which the only positive solution using the ‘ + ’ sign (a necessary condition for the
energy of a photon…) is hν = 217 keV .

15