PHYS851 Quantum Mechanics I, Fall 2009

HOMEWORK ASSIGNMENT 9: SOLUTIONS

1. The Parity ROperator: [20 pts] Determine the matrix element hx|Π|x′ i and use it to simplify the
identity Π = dx dx′ |xihx|Π|x′ ihx′ |, then use this identity to compute Π2 , Π3 , and Πn .
exp[Πu]
From these results find an expression for S(u) = cosh u in the form f (u) + g(u)Π.
What is hx|S(u)|ψi? Express your answer in terms of ψeven (x) = 21 (ψ(x) + ψ(−x)) and ψodd (x) =
1
2 (ψ(x) − ψ(−x)).
Compute hx|S(0)|ψi, lim hx|S(u)|ψi, and lim hx|S(u)|ψi.
u→∞ u→−∞
Answer:

hx|Π|x′ i = hx| − x′ i = δ(x + x′ )

Z Z
′ ′ ′
Π = dxdx |xiδ(x + x )hx | = dx|xih−x|
Z Z Z
Π2 = dxdx′ |xih−x|x′ ih−x′ | = dxdx′ |xiδ(−x − x′ )h−x′ | = dx|xihx| = 1

So Π3 = Π2 · Π = Π, which generalizes to

1; n = even
Πn =
Π; n = odd

Now we have
2 3
eΠu I + Πu + Π2 u2 + Π3 u3! + . . .
S(u) = =
cosh u cosh u
u2 u4 2 5
1 + 2 + 4! + . . . u + u3! + u5! + . . .
= I +Π
cosh u cosh u
cosh u + Π sinh u
=
cosh u
= I + Π tanh u

hx|S(u)|ψi = hx|ψi + tanh uh−x|ψi

= ψ(x) + tanh uψ(−x)
= ψeven (x) + ψodd (x) + tanh u (ψeven (−x) + ψodd (−x))
= (1 + tanh u)ψeven (x) + (1 − tanh u)ψodd (x)

So that
hx|S(0)|ψi = ψeven (x) + ψodd (x) = ψ(x)
lim hx|S(u)|ψi = 2ψeven (x)
u→∞

lim hx|S(u)|ψi = 2ψodd (x)

u→−∞

1
 2 /2 P∞ αn
2. [15 pts]The coherent state |αi is defined by |αi = e−|α| n=0
√ |ni, where the states {|ni} are
n!
the harmonic oscillator energy eigenstates.
First, show that for α = 0, the coherent state |α=0i is exactly equal to the harmonic oscillator
ground-state, |n=0i.
Then show that any other coherent state can be created by acting on the ground-state, |0i, with the
‘displacement operator’ D(α), i.e. show that |αi = D(α)|0i, where
† −α∗ A
D(α) := eαA (1)
You may need the Zassenhaus formula eB+C = eB eC e−[B,C]/2 , which is valid only when [B, [B, C]] =
[C, [B, C]] = 0.
What is D(α2 )|α1 i?
Answer:
Part i) For α = 0, we have
∞ ∞
 0n δ 
√n,0 |ni = |ni
X X
|αi = e−0 √ |ni =
 
α=0
n=0 n! n=0 n! n=0

Part ii) Let B = αA† and C = −α∗ A. Then [B, C] = −|α|2 [A† , A] = |α|2 , which commutes with
everything. We can therefore use the Zassenhaus formula, which gives
† −α∗ A 2 /2 † ∗A
eαA = e−|α| eαA e−α
Now we have
∞ ∞
∗A
X (−α∗ )n X (−α∗ )n
e−α |0i = An |0i = δn,0 |0i = |0i
n! n!
n=0 n=0
So we end up with
∞ ∞
2 /2 † 2 /2
X αn 2 /2
X αn
D(α)|0i = e−|α| eαA |0i = e−|α| (A† )n |0i = e−|α| √ |ni = |αi
n! n!
n=0 n=0

Part iii) Using the Zassenhaus formula, we have

e[α1 A −α1 A ,α2 A −α2 A]/2
† −α∗ A+α A† −α∗ A † ∗ † ∗
D(α1 )D(α2 ) = eα1 A 1 2 2

now h i
†
α1 A − α∗1 A, α2 A† − α∗2 A = −α1 α∗2 [A† , A] − α∗1 α2 [A, A† ] = α1 α∗2 − α2 α∗1
and
† −α∗ A+α A† −α∗ A
eα1 A 1 2 2 = D(α1 + α2 )
This gives
(α∗ ∗
1 α2 −α2 α1 )
D(α1 )D(α2 ) = e− 2 D(α1 + α2 )
3. [15 pts] Consider a system described by the Hamiltonian H = ~κ(A + A† ). Use your results from the
previous problem to determine |ψ(t)i for a system initially in the ground-state, |ψ(0)i = |0i.

We know that |ψ(t)i = e−iHt/~|ψ(0)i, so that

† −iκA)t
|ψ(t)i = e(−iκA |0i
with α(t) = −iκt, we have −α∗ (t) = −(iκt) = −iκt, so we have
† −α∗ A
|ψ(t)i = eαA |0i = |α(t)i

2
4. [10pts each] Cohen Tannoudji, pp341-350: problems 3.6, 3.7, 3.11

3.6 For these problems, the primary task is to set up the integral which gives the desired probability:

a. Z ∞
N2 dxdydz e−|x|/a−|y|/b−|z|/c = 1
−∞
Z ∞ Z ∞
−|x|/a
dxe =2 dxe−x/a = 2a
−∞ 0
So that from symmetry we get
N 2 8abc = 1
√
which gives N = 1/ 8abc.
b.
a ∞ ∞
e−|x|/a−|y|/b−|z|/c
Z Z Z
P = dx dy dz
0 −∞ −∞ 8abc
Z a
1
= dx e−|x|/a
2a 0
1 1
Z
= du e−u
2 0
e−1
= (2)
2e
c. Based on the previous result and symmetry, we have
(e − 1)2
P=
e2
d. The requested quantity is
P = |hpx = 0, py = 0, pz = ~/c|ψi|2 dpx dpy dpz
Z
h0, 0, ~/c|ψi = dxdydz h0, 0, ~/c|xyzihxyz|ψi

1 −|x|/2a−|y|/2b−|z|/2c
−iz/c e
Z
= dxdydz e √
(2π~)3/2 8abc
Z ∞ Z ∞ Z ∞
1 −|x|/2a −|y|/2b
= √ dx e dye dz e−|z|/2c−iz/c
(2π~)3/2 8abc −∞ −∞ −∞
√ Z ∞ Z ∞ Z ∞
8 −x/2a −y/2b
= √ dx e dye dz e−z/2c cos(z/c)
(2π~)3/2 abc 0 0 0
√
8 8abc
= √
(2π~)3/2 abc 5
√
8 8abc
= (3)
5(2π~)3/2
So we have
512abc 64abc
P= 3
dpx dpy dpz = dpx dpy dpz
25(2π~) 25(π~)3

3
3.7 a. Z x2 Z ∞ Z ∞
P= dx dy dz|ψ(x, y, z)|2
x1 −∞ −∞

b. Z p2 Z ∞ Z ∞
P= dpx dy dz|ψ(px , y, z)|2
p1 −∞ −∞
R∞
where ψ(px , y, z) = √1 dxe−ipx x/~ψ(x, y, z).
2π~ −∞
c. Z x2 Z ∞ Z ∞
P= dx dy dpz |ψ(x, y, pz )|2
x1 0 0
R∞
where ψ(x, y, pz ) = √1 dz e−ipz z/~ψ(x, y, z)
2π~ −∞
d. Z p2 Z p4 Z p6
P= dpx dpy dpz |ψ(px , py , pz )|2
p1 p3 p5
1
R∞ R∞ R∞
−i(px x+py y+pz z)/~ψ(x, y, z)
where ψ(px , py , pz ) = (2π~)3/2 −∞ dx −∞ dy −∞ dz e

If we extend the py and pz limits to infinity we get

Z p2 Z ∞ Z ∞
P = dpx dpy dpz hψ|px , py , pz ihpx , py , pz |ψi
p1 −∞ −∞
Z p2 Z ∞ Z ∞  
= dpx dpy dpz hψ| |px ihpx | ⊗ |py , pz ihpy , pz | |ψi
p1 −∞ −∞

Now the identity operator can be written as

I = Ix ⊗ Iy ⊗ Iz
Z ∞ Z ∞
= Ix ⊗ dpy dpz |py , pz ihpy , pz |
−∞ −∞
Z ∞ Z ∞
= Ix ⊗ dy dz |y, zihy, z|
−∞ −∞

This shows that

Z ∞ Z ∞ Z ∞ Z ∞
dpy dpz |py , pz ihpy , pz | = dy dz |y, zihy, z|
−∞ −∞ −∞ −∞

which clearly makes sense.

Substituting this into the expression for P gives
Z p2 Z ∞ Z ∞
P= dpx dy dz hψ|px , y, zihpx , y, z|ψi
p1 −∞ −∞

which agrees with the answer to b.

4
e. Method: Treat as probability problem.
Standard probability theory tells us that if u = f (x, y, z) then the probability density p(u)
is given by Z
ρ(u) = d3 r ρ(u|~r)ρ(~r)

where ρ(u|~r) is the probability density over u for fixed ~r, and QM tells us that ρ(~r) = |ψ(~r)|2 .
Now we clearly must have ρ(u|~r) = aδ(u − f (~r)), where the normalization constant is
determined by requiring that
Z ∞ Z ∞
1= du ρ(u|~r) = a du δ(u − f (~r)) = a
−∞ −∞

so that a = 1. This gives

Z
ρ(u) = d3 r |ψ(~r)|2 δ(u − f (~r))

and then Z u2 Z u2 Z
P= du ρ(u) = du d3 r |ψ(~r)|2 δ(u − f (~r))
u1 u1

5
3.11 a. Z β
P = dx1 dx2 |ψ(x1 , x2 )|2
α
b. Z ∞
P = dx1 dx2 |ψ(x1 , x2 )|2
−∞
c.
Z β Z ∞ Z ∞ Z β Z β Z β
2 2
P= dx1 dx2 |ψ(x1 , x2 )| + dx1 dx2 |ψ(x1 , x2 )| − dx1 dx2 |ψ(x1 , x2 )|2
α −∞ −∞ α α α

d.
Z β Z α Z β Z ∞ Z α Z β
2 2
P = dx1 dx2 |ψ(x1 , x2 )| + dx1 dx2 |ψ(x1 , x2 )| + dx1 dx2 |ψ(x1 , x2 )|2
α −∞ α β −∞ α
Z ∞ Z β
+ dx1 dx2 |ψ(x1 , x2 )|2
β α

or equivalently
Z β Z ∞ Z ∞ Z β Z β Z β
2 2
P= dx1 dx2 |ψ(x1 , x2 )| + dx1 dx2 |ψ(x1 , x2 )| −2 dx1 dx2 |ψ(x1 , x2 )|2
α −∞ −∞ α α α

e. Z p′′ Z β
P= dp1 dx2 |ψ(p1 , x2 )|2
p′ α
R∞
where ψ(p1 , x2 ) = √1 dx1 e−ip1 x1 /~ψ(x1 , x2 )
2π~ −∞
f. Z p′′ Z p′′′′
P= dp1 dp2 |ψ(p1 , p2 )|2
p′ p′′′
1
R∞ R∞ −i(p1 x1 +p2 x2 )/~ψ(x , x )
where ψ(p1 , p2 ) = 2π~ −∞ dx 1 −∞ dx2 e 1 2
g. From the results of e., we find
Z p′′ Z ∞
P= dp1 dx2 |ψ(p1 , x2 )|2
p′ −∞

from the results of f., we find

Z p′′ Z ∞
P= dp1 dp2 |ψ(p1 , p2 )|2
p′ −∞

this shows that Z ∞ Z ∞

2
dx2 |ψ(p1 , x2 )| = dp2 |ψ(p1 , p2 )|2
−∞ −∞
which follows because they are both equal to
 
hψ| |p1 ihp1 | ⊗ I2 |ψi

6
h.
Z d Z ∞ Z ∞
P = dx dx1 dx2 δ(x − x1 + x2 )|ψ(x1 , x2 )|2
−d −∞ −∞
Z d Z ∞
= dx dx1 |ψ(x1 , x1 − x)|2
−d −∞
Z ∞ Z ∞
x̄ = hX1 − X2 i = dx1 dx2 (x1 − x2 )|ψ(x1 , x2 )|2
−∞ −∞