Solutions to Problems in Goldstein,

Classical Mechanics, Second Edition

Homer Reid
April 21, 2002

Chapter 7

Problem 7.2

Obtain the Lorentz transformation in which the velocity is at an infinitesimal angle

dθ counterclockwise from the z axis, by means of a similarity transformation applied
to Eq. (7-18). Show directly that the resulting matrix is orthogonal and that the
inverse matrix is obtained by substituting −v for v.

We can obtain this transformation by first applying a pure rotation to rotate

the z axis into the boost axis, then applying a pure boost along the (new) z
axis, and then applying the inverse of the original rotation to bring the z axis
back in line with where it was originally. Symbolically we have L = R−1 KR
where R is the rotation to achieve the new z axis, and K is the boost along the
z axis.
Goldstein tells us that the new z axis is to be rotated dθ counterclockise
from the original z axis, but he doesn’t tell us in which plane, i.e. we know θ
but not φ for the new z axis in the unrotated coordinates. We’ll assume the z
axis is rotated around the x axis, in a sense such that if you’re standing on the
positive x axis, looking toward the negative x axis, the rotation appears to be
counterclockwise, so that the positive z axis is rotated toward the negative y

1
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 2

axis. Then, using the real metric,

   
1 0 0 0 1 0 0 0 1 0 0 0
 0 cos dθ sin dθ 0  0 1 0 0  0 cos dθ − sin dθ 0 
L= 
 0 − sin dθ cos dθ 0   0 0
 
γ −βγ   0 sin dθ cos dθ 0 
0 0 0 1 0 0 −βγ γ 0 0 0 1
  
1 0 0 0 1 0 0 0
 0 cos dθ sin dθ 0  0 cos dθ − sin dθ 0 
= 
 0 − sin dθ cos dθ 0   0

γ sin dθ γ cos dθ −βγ 
0 0 0 1 0 −βγ sin dθ −βγ cos dθ γ
 
1 0 0 0
 0 cos2 dθ + γ sin2 dθ (γ − 1) sin dθ cos dθ −βγ sin dθ 
=  0 (γ − 1) sin dθ cos dθ sin2 dθ + γ cos2 dθ −βγ cos dθ
.

0 −βγ sin dθ −βγ cos dθ γ

Problem 7.4

A rocket of length l0 in its rest system is moving with constant speed along the z
axis of an inertial system. An observer at the origin observes the apparent length
of the rocket at any time by noting the z coordinates that can be seen for the head
and tail of the rocket. How does this apparent length vary as the rocket moves from
the extreme left of the observer to the extreme right?

Let’s imagine a coordinate system in which the rocket is at rest and centered
at the origin. Then the world lines of the rocket’s top and bottom are

xt0
µ = {0, 0, +L0 /2, τ } xb0
µ = {0, 0, −L0 /2, τ } .

where we are parameterizing the world lines by the proper time τ . Now, the rest
frame of the observer is moving in the negative z direction with speed v = βc
relative to the rest frame of the rocket. Transforming the world lines of the
rocket’s top and bottom to the rest frame of the observer, we have

xtµ = {0, 0, γ(L0 /2 + vτ ), γ(τ + βL0 /2c)} (1)

xbµ = {0, 0, γ(−L0/2 + vτ ), γ(τ − βL0 /2c)} . (2)

Now consider the observer. At any time t in his own reference frame, he is
receiving light from two events, namely, the top and bottom of the rocket moving
past imaginary distance signposts that we pretend to exist up and down the z
axis. He sees the top of the rocket lined up with one distance signpost and the
bottom of the rocket lined up with another, and from the difference between the
two signposts he computes the length of the rocket. Of course, the light that
he sees was emitted by the rocket some time in the past, and, moreover, the
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 3

light signals from the top and bottom of the rocket that the observer receives
simultaneously at time t were in fact emitted at different proper times τ in the
rocket’s rest frame.
First consider the light received by the observer at time t0 coming from
the bottom of the rocket. Suppose in the observer’s rest frame this light were
emitted at time t0 − ∆t, i.e. ∆t seconds before it reaches the observer at the
origin; then the rocket bottom was passing through z = −c∆t when it emitted
this light. But then the event identified by (z, t) = (−c∆t, t0 − ∆t ) must lie on
the world line of the rocket’s bottom, which from (2) determines both ∆t and
the proper time τ at which the light was emitted:

 1/2
−c∆t = γ(−L0 /2 + vτ ) 1+β L0
=⇒ τ= t0 − ≡ τb (t0 ).
t0 − ∆ t = γ(τ + βL0 /2c) 1−β 2c

We use the notation τb (t0 ) to indicate that this is the proper time at which the
bottom of the rocket emits the light that arrives at the observer’s origin at the
observer’s time t0 . At this proper time, from (2), the position of the bottom of
the rocket in the observer’s reference frame was

zb (τb (t0 )) = −γL0 /2 + vγτb (t0 )

( 1/2 )
1+β L0
= −γL0 /2 + vγ t0 − (3)
1−β 2c

Similarly, for the top of the rocket we have

 1/2
1+β L0
τt (t0 ) = t0 +
1−β 2c

and
( 1/2 )
1+β L0
zt (τt (t0 )) = γL0 /2 + vγ t0 + (4)
1−β 2c

Subtracting (3) from (4), we have the length for the rocket computed by the
observer from his observations at time t0 in his reference frame:

L(t0 ) = γ(1 + β)L0

 1/2
1+β
= L0 .
1−β
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 4

Problem 7.17

Two particles with rest masses m1 and m2 are observed to move along the observer’s
z axis toward each other with speeds v1 and v2 , respectively. Upon collision they
are observed to coalesce into one particle of rest mass m3 moving with speed v3
relative to the observer. Find m3 and v3 in terms of m1 , m2 , v1 , and v2 . Would it
be possible for the resultant particle to be a photon, that is m3 = 0, if neither m1
nor m2 are zero?

Equating the 3rd and 4th components of the initial and final 4-momentum
of the system yields

γ 1 m1 v 1 − γ 2 m2 v 2 = γ 3 m3 v 3
γ 1 m1 c + γ 2 m2 c = γ 3 m3 c

Solving the second for m3 yields

γ1 γ2
m3 = m1 + m2 (5)
γ3 γ3
and plugging this into the first yields v3 in terms of the properties of particles
1 and 2:
γ 1 m1 v 1 − γ 2 m2 v 2
v3 =
γ 1 m1 + γ 2 m2
Then
v3 γ 1 m 1 β1 − γ 2 m 2 β2
β3 = =
c γ 1 m1 + γ 2 m2
2 2
γ m + 2γ1 γ2 m1 m2 + γ22 m22 − [γ12 m21 β12 + γ22 m22 β22 − 2γ1 γ2 m1 m2 β1 β2 ]
1 − β32 = 1 1
(γ1 m1 + γ2 m2 )2
γ m (1 − β1 ) + γ2 m2 (1 − β22 ) + 2γ1 γ2 m1 m2 (1 − β1 β2 )
2 2 2 2 2
= 1 1
(γ1 m1 + γ2 m2 )2
2 2
m + m2 + 2γ1 γ2 m1 m2 (1 − β1 β2 )
= 1
(γ1 m1 + γ2 m2 )2

and hence

1 (γ1 m1 + γ2 m2 )2
γ32 = 2 = 2 2 . (6)
1 − β3 m1 + m2 + 2γ1 γ2 m1 m2 (1 − β1 β2 )

Now, (5) shows that, for m3 to be zero when either m1 or m2 is zero, we must
have γ3 = ∞. That this condition cannot be met for nonzero m1 , m2 is evident
from the denominator of (6), in which all terms are positive (since β1 β2 < 1 if
m1 or m2 is nonzero).
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 5

Problem 7.19

A meson of mass π comes to rest and disintegrates into a meson of mass µ and a
neutrino of zero mass. Show that the kinetic energy of motion of the µ meson (i.e.
without the rest mass energy) is

(π − µ)2 2
c .
2π

Working in the rest frame of the pion, the conservation relations are

πc2 = (µ2 c4 + p2µ c2 )1/2 + pν c (energy conservation)

0 = p µ + pν (momentum conservation).

From the second of these it follows that the muon and neutrino must have the
same momentum, whose magnitude we’ll call p. Then the energy conservation
relation becomes

πc2 = (µ2 c4 + p2 c2 )1/2 + pc

−→ (πc − p)2 = µ2 c2 + p2
π 2 − µ2
−→ p= c.
2π
Then the total energy of the muon is

Eµ = (µ2 c4 + p2 c2 )1/2
1/2
(π 2 − µ2 )2

= c 2 µ2 +
4π 2
c2
1/2
= 4π 2 µ2 + (π 2 − µ2 )2
2π
c2 2
= (π + µ2 )
2π
Then subtracting out the rest energy to get the kinetic energy, we obtain

c2 2
K = Eµ − µc2 = (π + µ2 ) − µc2
2π
c2 2
= (π + µ2 − 2πµ)
2π
c2
= (π − µ)2
2π
as advertised.
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 6

Problem 7.20

A π + meson of rest mass 139.6 MeV collides with a neutron (rest mass 939.6 MeV)
stationary in the laboratory system to produce a K + meson (rest mass 494 MeV)
and a Λ hyperon (rest mass 1115 MeV). What is the threshold energy for this
reaction in the laboratory system?

We’ll put c = 1 for this problem. The four-momenta of the pion and neutron
before the collision are

pµ,π = (pπ , γπ mπ ), pµ,n = (0, mn )

and the squared magnitude of the initial four-momentum is thus

pµ,T pµT = −|pπ |2 + (γπ mπ + mn )2

= −|pπ |2 + γπ2 m2π + m2n + 2γπ mπ mn
= m2π + m2n + 2γπ mπ mn
= (mπ + mn )2 + 2(γπ − 1)mπ mn (7)

The threshold energy is the energy needed to produce the K and Λ particles
at rest in the COM system. In this case the squared magnitude of the four-
momentum of the final system is just (mK + mΛ )2 , and, by conservation of
momentum, this must be equal to the magnitude of the four-momentum of the
initial system (7):

(mK + mΛ )2 = (mπ + mn )2 + 2(γπ − 1)mπ mn

(mK + mΛ )2 − (mπ + mn )2
=⇒ γπ = 1 + = 6.43
2mπ mn
Then the total energy of the pion is T = γπ mπ = (6.43 · 139.6 MeV) = 898
MeV, while its kinetic energy is K = T − m = 758 MeV.
The above appears to be the correct solution to this problem. On the other
hand, I first tried to do it a different way, as below. This way yields a different
and hence presumably incorrect answer, but I can’t figure out why. Can anyone
find the mistake?
The K and Λ particles must have, between them, the same total momentum
in the direction of the original pion’s momentum as the original pion had. Of
course, the K and Λ may also have momentum in directions transverse to the
original pion momentum (if so, their transverse momenta must be equal and
opposite). But any transverse momentum just increases the energy of the final
system, which increases the energy the initial system must have had to produce
the final system. Hence the minimum energy situation is that in which the K and
Λ both travel in the direction of the original pion’s motion. (This is equivalent
to Goldstein’s conclusion that, just at threshold, the produced particles are at
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 7

rest in the COM system). Then the momentum conservation relation becomes
simply

pπ = p K + p λ (8)

and the energy conservation relation is (with c = 1)

(m2π + p2π )1/2 + mn = (m2K + p2K )1/2 + (m2Λ + p2Λ )1/2 . (9)

The problem is to find the minimum value of pπ that satisfies (9) subject to the
constraint (8).
To solve this we must first resolve a subquestion: for a given pπ , what is the
relative allocation of momentum to pK and pΛ that minimizes (9) ? Minimizing

Ef = (m2K + p2K )1/2 + (m2Λ + p2Λ )1/2 .

subject to pK + pΛ = pπ , we obtain the condition

pK pΛ mK
= =⇒ pK = pΛ (10)
(m2K + p2K )1/2 (m2Λ + p2Λ )1/2 mΛ

Combining this with (8) yields

mΛ mK
pΛ = pπ pK = pπ . (11)
mK + m Λ mK + m Λ
For a given total momentum pπ , the minimum possible energy the final system
can have is realized when pπ is partitioned between pK and pΛ according to
(11). Plugging into (8), the relation defining the threshold momentum is
 2 !1/2  2 !1/2
mK mΛ
(m2π + p2π )1/2 + mn = m2K + p2π + m2Λ + p2π
mK + m Λ mK + m Λ

Solving numerically yields pπ ≈ 655 MeV/c, for a total pion energy of about
670 MeV.
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 8

Problem 7.21

A photon may be described classically as a particle of zero mass possessing never-

theless a momentum h/λ = hν/c, and therefore a kinetic energy hν. If the photon
collides with an electron of mass m at rest it will be scattered at some angle θ with
a new energy hν 0 . Show that the change in energy is related to the scattering angle
by the formula
θ
λ0 − λ = 2λc sin2 ,
2
where λc = h/mc, known as the Compton wavelength. Show also that the kinetic
energy of the recoil motion of the electron is

2 λλc sin2 2θ


T = hν .
1 + 2 λλc sin2 θ/2

Let’s assume the photon is initially travelling along the z axis. Then the sum
of the initial photon and electron four-momenta is
     
0 0 0
 0   0   0 
 h/λ  +  0  = 
pµ,i = pµ,γ + pµ,e =      . (12)
h/λ 
h/λ mc mc + h/λ

Without loss of generality we may assume that the photon and electron move
in the xz plane after the scatter. If the photon’s velocity makes an angle θ with
the z axis, while the electron’s velocity makes an angle φ, the four-momentum
after the collision is
(h/λ0 ) sin θ + pe sin φ
     
(h/λ0 ) sin θ pe sin φ
 0   0   0 
pµ,f = pµ,γ + pµ,e =   (h/λ0 ) cos θ  + 
  = 0
.
p ep cos φ   (h/λ ) cos
p + pe cos φ
θ 
0 2 2 2 0
h/λ m c + pe (h/λ ) + m2 c2 + p2e
(13)

Equating (12) and (13) yields three separate equations:

(h/λ0 ) sin θ + pe sin φ = 0 (14)

0
(h/λ ) cos θ + pe cos φ = h/λ (15)
p
h/λ0 + m2 c2 + p2e = mc + h/λ (16)

From the first of these we find

"  2 #1/2
h h 2
sin φ = − 0 sin θ =⇒ cos φ = 1 + sin θ
λ pe λ0 p e
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 9

and plugging this into (15) we find

h2 h2 h2
p2e = + − 2 cos θ. (17)
λ2 λ02 λλ0
On the other hand, we can solve (16) to obtain
 2  
1 1 1 1
p2e = h2 − + 2mch − .
λ λ0 λ λ0

Comparing these two determinations of pe yields

mc 0
cos θ = 1 − (λ − λ)
h
or
θ mc 0 1
sin2 = (λ − λ) = (λ0 − λ)
2 2h 2λc
so this is advertised result number 1.
Next, to find the kinetic energy of the electron after the collision, we can
write the conservation of energy equation in a slightly different form:
h h
= γmc + 0
mc +
λ  λ 
1 1
=⇒ (γ − 1)mc = K = h −
λ λ0
 0 
λ −λ
=h
λλ0
2λc sin2 (θ/2)
 
=h
λ[λ + 2λc sin2 (θ/2)]
2χ sin2 (θ/2)
 
h
=
λ 1 + 2χ sin2 (θ/2)

where we put χ = λc /λ.

Problem 7.22

A photon of energy E collides at angle θ with another photon of energy E. Prove

that the minimum value of E permitting formation of a pair of particles of mass m
is
2m2 c4
Eth = .
E(1 − cos θ)

We’ll suppose the photon of energy E is traveling along the positive z axis,
while that with energy E is traveling in the xz plane (i.e., its velocity has
Homer Reid’s Solutions to Goldstein Problems: Chapter 7 10

spherical polar angles θ and φ = 0). Then the 4-momenta are

 
E E
p1 = 0, 0, ,
c c
 
E E E
p2 = sin θ, 0, cos θ,
c c c
 
E E + E cos θ E + E
pt = p 1 + p 2 = sin θ, 0, ,
c c c

It’s convenient to rotate our reference frame to one in which the space portion
of the composite four-momentum of the two photons is all along the z direction.
In this frame the total four-momentum is
 
0 1p 2 2
E+E
pt = 0, 0, E + E + 2EE cos θ, . (18)
c c

At threshold energy, the two produced particles have the same four-momenta:
 
p3 = p4 = 0, 0, p, (m2c2 + p2 )1/2 (19)

and 4-momentum conservation requires that twice (19) add up to (18), which
yields two conditions:

1
√ 1
2p = c E 2 + E 2 + 2EE cos θ −→ p 2 c2 = 4 (E
2
+ E 2 + 2EE cos θ)
p E+E 1
2 m 2 c2 + p 2 = c −→ m2 c4 + p2 c2 = 4 (E
2
+ E 2 + 2EE)

Subtracting the first of these from the second, we obtain

EE
m2 c 4 = (1 − cos θ)
2
or

2m2 c4
E=
E(1 − cos θ)

as advertised.