Solutions to Problems in Merzbacher,

Quantum Mechanics, Third Edition

1
(r) + c
2

2
(r)
where
1
(r) and
2
(r) are related to one another by time reversal,
show that the probability current density can be expressed without
an interference term involving
1
and
2
.
I found this to be a pretty cool problem! First of all, we have the probability
conservation equation:
d
dt
=

J.
To show that

J contains no cross terms, it suces to show that its divergence
has no cross terms, and to show this it suces (by probability conservation) to
show that d/dt has no cross terms. We have
=

= [c

1
+ c

2
] [c
1

1
+ c
2

2
]
= |c
1
||
1
| + |c
2
||
2
| + c
1
c

2
+ c

1
c
2

2
(1)
1
Problem 3.2
For a free particle in one dimension, calculate the variance at time
t, (x)
2
t

(x x
t
)
2
_
t
=

x
2
_
t
x
2
t
without explicit use of the
wave function by applying (3.44) repeatedly. Show that
(x)
2
t
= (x)
2
0
+
2
m
_
1
2
xp
x
+ p
x
x
0
x
0
p
x

_
t +
(p
x
)
2
m
2
t
2
and
(p
x
)
2
t
= (p
x
)
2
0
= (p
x
).
2
I nd it easiest to use a slightly dierent notation: w(t) (x)
2
t
. (The w
reminds me of width.) Then
w(t) = w(0) + t
dw
dt

t=0
+
1
2
t
2
d
2
w
dt
2

t=0
+ (2)
We have
dw
dt
=
d
dt
_

x
2
_
x
2
_
=
d
dt

x
2
_
2 x
d
dt
x (3)
d
2
w
dt
2
=
d
2
dt
2

x
2
_
2
_
d
dt
x
_
2
2 x
d
2
dt
2
x (4)
We need to compute the time derivatives of x and

x
2
_
. The relevant
equation is
d
dt
F =
1
i h
FH HF +
_
F
t
_
for any operator F. For a free particle, the Hamiltonian is H = p
2
/2m, and
the all-important commutation relation is px = xp i h. We can use this to
calculate the time derivatives:
d
dt
x =
1
i h
[x, H]
=
1
2im h

xp
2
p
2
x
_
=
1
2im h

xp
2
p(xp i h)
_
=
1
2im h

xp
2
pxp + i hp
_
2
=
1
2im h

xp
2
(xp i h)p + i hp
_
=
1
2im h
2i hp
=
p
m
(5)
d
2
dt
2
x =
1
m
d
dt
p = 0 (6)
d
dt

x
2
_
=
1
i h

[x
2
, H]
_
=
1
2im h

x
2
p
2
p
2
x
2
_
=
1
2im h

x
2
p
2
p(xp i h)x
_
=
1
2im h

x
2
p
2
pxpx + i hpx
_
=
1
2im h

x
2
p
2
(xp i h)
2
+ i h(xp i h)
_
=
1
2im h

x
2
p
2
xpxp + 2i hxp + h
2
+ h
2
+ i hxp
_
=
1
2im h

x
2
p
2
x(xp i h)p + 3i hxp + 2 h
2
_
=
1
2im h

2 h
2
+ 4i hxp
_
=
i h
m
+
2
m
xp (7)
d
2
dt
2

x
2
_
=
2
m
d
dt
xp
=
2
i hm
[xp, H]
=
1
i hm
2

xp
3
p
2
xp
_
=
1
i hm
2

xp
3
p(xp i h)p
_
=
1
i hm
2

xp
3
pxp
2
+ i hp
2
_
=
1
i hm
2

xp
3
(xp i h)p
2
+ i hp
2
_
=
1
i hm
2

2i hp
2
_
=
2
m
2

p
2
_
(8)
d
3
dt
3

x
2
_
=
2
i hm
2

[p
2
, H]
_
= 0 (9)
Now that weve computed all time derivatives of x and

x
2
_
, its time to
3
plug them into (3) and (4) to compute the time derivatives of w.
dw
dt
=
d
dt

x
2
_
2 x
d
dt
x
=
i h
m
+
2
m
xp
2
m
x p
=
2
m
_

i h
2
+ xp
_

2
m
x p
=
2
m
_
px xp
2
+ xp
_

2
m
x p
=
2
m
_
px + xp
2
_

2
m
x p (10)
d
2
w
dt
2
=
d
2
dt
2

x
2
_
2
_
d
dt
x
_
2
2 x
d
2
dt
2
x
=
2
m
2

p
2
_

2
m
2
p
2
=
2
m
2
(p)
2
(11)
Finally, we plug these into the original equation (2) to nd
w(t) = w(0) +
2
m
_
1
2
px + xp x p
_
t +
(p)
2
m
2
t.
2
The other portion of this problem, the constancy of (p)
2
, is trivial, since
(p)
2
contains expectation values of p and p
2
, which both commute with H.
4
Problem 3.3
Consider a linear harmonic oscillator with Hamiltonian
H = T + V =
p
2
2m
+
1
2
m
2
x
2
.
(a) Derive the equation of motion for the expectation value x
t
,
and show that it oscillates, similarly to the classical oscillator,
as
x
t
= x
0
cos t +
p
0
m
sint.
(b) Derive a second-order dierential equation of motion for the
expectation value T V
t
by repeated application of (3.44)
and use of the virial theorem. Integrate this equation and,
remembering conservation of energy, calculate

x
2
_
t
.
(c) Show that
(x)
2
t

x
2
_
t
x
2
t
= (x)
2
0
cos
2
t +
(p)
2
0
m
2

2
sin
2
t
+
_
1
2
xp + px
0
x
0
p
0
_
sin 2t
m
Verify that this reduces to the result of Problem 2 in the limit
0.
(d) Work out the corresponding formula for the variance (p)
2
t
.
(a) Again I like to use slightly dierent notation: e(t) = x
t
. Then
d
dt
e(t) =
1
i h
xH Hx
=
1
2i hm

xp
2
p
2
x
_
=
1
2i hm

xp
2
p(xp i h)
_
=
1
2i hm

xp
2
(xp i h)p + i hp
_
=
1
2i hm
2i hp
=
p
m
.
d
2
dt
2
e(t) =
d
dt
p
m
5
=
1
i hm
pH Hp
=

2
2i h

px
2
x
2
p
_
=

2
2i h

(xp i h)x x
2
p
_
=

2
2i h

x(xp i h) i hx x
2
p
_
=

2
2i h
2i hx
=
2
x .
So we have
d
2
dt
2
e(t) =
2
e(t)
with general solution e(t) = Acos t +Bsin t. The coecients are determined
by the boundary conditions:
e(0) = x
0
A = x
0
e

(0) =
p
0
m
B =
p
0
m
.
(b) Lets dene v(t) = T V
t
. Then
d
dt
v(t) =
1
i h
(T V )H H(T V )
=
1
i h
(T V )(T + V ) (T + V )(T V )
=
2
i h
TV V T
=

2
2i h

p
2
x
2
x
2
p
2
_
.
We already worked out this commutator in Problem 2:

p
2
x
2
x
2
p
2
_
=

4i hxp + 2 h
2
_
so
d
dt
v(t) = 2
2
xp + i h
2
.
= 2
2
xp +
2
xp px
=
2
xp + px (12)
Next,
d
2
dt
2
v(t) =
2
2
i h
xpH Hxp
=
2
2
i h
_
1
2m

xp
3
p
2
xp
_
+
m
2
2

xpx
2
x
3
p
_
_
(13)
6
The bracketed expressions are

xp
3
p
2
xp
_
=

xp
3
p(xp i h)p
_
=

xp
3
(xp i h)p
2
+ i hp
2
_
=

2i hp
2
_

xpx
2
x
3
p
_
=

x(xp i h)x x
3
p
_
=

x
2
(xp i h) i hx
2
x
3
p
_
=

2i hx
2
_
and plugging these back into (13) gives
d
2
dt
2
v(t) = = 4
2
_

p
2
_
m

m
2
2

x
2
_
_
= 4
2
v(t)
with solution
v(t) = Acos 2t + B sin 2t. (14)
Evaluating at t = 0 gives
A = T
0
V
0
.
Also, we can use (12) evaluated at t = 0 to determine B:

2
xp + px
0
+ i h
2
= 2B
so
B =
xp + px
0
2
.
The next task is to compute

x
2
_
t
:

x
2
_
t
=
2
m
2
V
t
=
1
m
2
H (T V )
t
=
1
m
2
[H
t
v(t)] .
Since H does not depend explicitly on time, H is constant in time. For v(t)
we can use (14):

x
2
_
t
=
1
m
2
_
T
0
+ V
0
[T
0
V
0
] cos 2t +
xp + px
0
2
sin 2t
_
=
1
m
2
_
2 T
0
sin
2
t + 2 V
0
cos
2
t +
xp + px
0
2
sin 2t
_
=

p
2
_
0
m
2

2
sin
2
t +

x
2
_
0
cos
2
t +
1
2
xp + px
0
sin 2t
m
. (15)
7
(c) Earlier we found that
x
t
= x
0
cos t +
p
0
m
sin t
x
2
t
= x
2
0
cos
2
t +
p
2
0
m
2

2
sin
2
t + x
0
p
0
sin 2t.
Subtracting from (15) gives
(x)
2
t
=

x
2
_
x
2
=
_

x
2
_
0
x
2
0
_
cos
2
t
+
1
m
2

2
_

p
2
_
0
p
2
0
_
+
_
1
2
xp + px
0
x
0
p
0
_
sin 2t
= (x)
2
0
cos
2
t +
(p)
2
0
m
2

2
sin
2
t +
_
1
2
xp + px
0
x
0
p
0
_
sin 2t
m
.
As 0, cos
2
t 1, (sin
2
t/
2
) 1, and (sin 2t/) 2, as needed
to ensure matchup with the result of Problem 2.
Problem 3.4
Prove that the probability density and the probability current den-
sity at position r
0
can be expressed in terms of the operators r and
p as expectation values of the operators
(r
0
) (r r
0
) j(r
0
)
1
2m
[p(r r
0
) + (r r
0
)p] .
Derive expressions for these densities in the momentum represen-
tation.
The rst one is trivial:
(r r
0
) =
_

(r)(r r
0
)(r)dr =

(r
0
)(r
0
) = (r
0
).
For the second one,
1
2m
p(r r
0
) + (r r
0
)p =
i h
2m
_
[

(r r
0
) +

(r r
0
)] dr
The gradient operator in the rst term operates on everything to its right:
=
i h
2m
_
[

(r r
0
) + 2(r r
0
)

] dr.
8
Here we can use the identity
_
f(x)

(x a)dx = f

(a) :
=
i h
2m
|(

) + 2

|
r=r0
=
i h
2m
|

|
r=r0
= j(r
0
).
Problem 3.5
For a system described by the wave function (r

), the Wigner
distribution function is dened as
W(r

, p

) =
1
(2 h)
3
exp(ip

/ h)

_
r

2
_

_
r

+
r

2
_
dr

.
(a) Show that W(r

, p

) is a real-valued function, dened over the

, p

).
(b) Prove that
_
W(r

, p

)dp

= |(r

)|
2
and that the expectation value of a function of the operator
r in a normalized state is
f(r) =
_ _
f(r

)W(r

, p

)dr

dp

.
(c) Show that the Wigner distribution function is normalized as
_
W(r

, p

)dr

dp

= 1.
(d) Show that the probability density (r
0
) at position r
0
is ob-
tained from the Wigner distribution function with
(r
0
) f(r) = (r r
0
).
(a)
9