A Course in Mechanics by Dr. J.

Tinsley Oden
Part II - Homework 3 - Solutions
Pablo Seleson

November 30, 2010

Set II.2:
7. In this exercise, we use time-independent wave functions for simplicity.

(a) We check whether eix is Hermitian:

Z ∞
ix
hψ(x), e φ(x)i = ψ ∗ (x)eix φ(x)dx,
Z−∞∞ Z ∞
−ix ∗
ix
he ψ(x), φ(x)i = e ψ (x)φ(x)dx = ψ ∗ (x)e−ix φ(x)dx 6= hψ(x), eix φ(x)i.
−∞ −∞

Then, the operator is not Hermitian.

d2
(b) We check whether dx2
is Hermitian:
∞
∞ Z ∞
d2 d2 φ dψ ∗
Z
∗ ∗ dφ  dφ
hψ(x), 2 φ(x)i = ψ (x) 2 (x)dx = ψ (x) (x) − (x) (x)dx
dx −∞ dx dx  −∞ dx dx
−∞
Z ∞
dψ ∗ dφ
=− (x) (x)dx,
−∞ dx dx

where we used integration by parts, and the fact that ψ(x) → 0 for x → ±∞ and we
assume the wave function derivatives are bounded. By another integration by parts we
obtain
∞ Z ∞ 2 ∗ Z ∞ 2 ∗
d2 dψ ∗  d ψ d ψ
hψ(x), 2 φ(x)i = − (x)φ(x) + (x)φ(x)dx = (x)φ(x)dx,

dx dx 2 2
 −∞ dx −∞ dx
−∞

where the boundary terms vanish as before. Therefore,

d2 d2
hψ(x), φ(x)i = h ψ(x), φ(x)i ,
dx2 dx2

and thus the operator is Hermitian.

1
 d
(c) We check whether x dx is Hermitian:
Z ∞
d dφ
hψ(x), x φ(x)i = ψ ∗ (x)x (x)dx
dx −∞ dx
∞ Z ∞ Z ∞
∗
 dψ ∗
= ψ (x)xφ(x) − (x)xφ(x)dx − ψ ∗ (x)φ(x)dx

 −∞ dx −∞
−∞
d d
= −hx ψ(x), φ(x)i − hψ(x), φ(x)i =
6 hx ψ(x), φ(x)i,
dx dx
where we used integration by parts and assumed wave functions decay fast enough so
√ √
that xφ(x) →= 0 and xψ ∗ (x) →= 0 as x → ±∞. This assumption implies hxi is
bounded. Thus, the operator is not Hermitian.

8. a)
Z
dhQi d d
= hΨ(x, t), Q̃Ψ(x, t)i = Ψ∗ (x, t)Q̃Ψ(x, t)d3 x
dt dt dt R3
∂Ψ∗
Z Z Z
∂ Q̃ ∂Ψ
= (x, t)Q̃Ψ(x, t)d3 x + Ψ∗ (x, t) Ψ(x, t)d3 x + Ψ∗ (x, t)Q̃ (x, t)d3 x
R 3 ∂t 3 ∂t 3 ∂t
*R + Z R
Z
∂Ψ ∗ ∂ Q̃ ∂Ψ
= (x, t)Q̃Ψ(x, t)d3 x + + Ψ∗ (x, t)Q̃ (x, t)d3 x.
R3 ∂t ∂t R3 ∂t

b) Following Schrödinger’s equation

~2
 
∂Ψ 1 1
= HΨ = − ∆ + V Ψ,
∂t i~ i~ 2m
∗ 2
 
∂Ψ 1 ~ 1
⇒ =− − ∆ + V Ψ∗ = − HΨ∗ .
∂t i~ 2m i~

c) Using the Hermitian property of H we proceed as follows

Z * + Z
dhQi 1 ∂ Q̃ 1
= − HΨ∗ (x, t)Q̃Ψ(x, t)d3 x + + Ψ∗ (x, t)Q̃ HΨ(x, t)d3 x
dt R3 i~ ∂t R3 i~
Z * + Z
1 ∂ Q̃ 1
=− Ψ∗ (x, t)H Q̃Ψ(x, t)d3 x + + Ψ∗ (x, t)Q̃HΨ(x, t)d3 x
i~ R3 ∂t i~ R3
Z * +
1   ∂ Q̃
= Ψ∗ (x, t) Q̃H − H Q̃ Ψ(x, t)d3 x +
i~ R3 ∂t
Z * + * +
1 ∗
h i
3 ∂ Q̃ 1 Dh iE ∂ Q̃
= Ψ (x, t) Q̃, H Ψ(x, t)d x + = Q̃, H + ,
i~ R3 ∂t i~ ∂t

then we have * +
dhQi 1 Dh iE ∂ Q̃
= Q̃, H + .
dt i~ ∂t

2
Set II.3:
1. a) Considering the time-independent Schrödinger’s equation

~2
− ∆ψ(x, y) = Eψ(x, y) in Ω,
2m
ψ(x, y) = 0 on ∂Ω.

We find a solution using the method of separation of variables: ψ(x, y) = X(x)Y (y).
Then, the equation in Ω is

~2 ~2
 2
∂2 ~2 d2 X d2 Y
  
∂
− ∆ψ(x, y) = − + X(x)Y (y) = − Y (y) 2 (x) + X(x) 2 (y)
2m 2m ∂x2 ∂y 2 2m dx dy
= EX(x)Y (y),

or dividing by X(x)Y (y) (assuming X(x)Y (y) 6= 0)

~2 1 d2 X ~2 1 d2 Y
   
− (x) + − (y) = E.
2m X(x) dx2 2m Y (y) dy 2

Because each term depends on a different variable, a solution exists if each term is a
constant and the sum of those constants equals E. Denote the constants eX and eY ,
then
~2 d2 X
− (x) = eX X(x), x ∈ (0, a)
2m dx2
~2 d2 Y
− (y) = eY Y (y), y ∈ (0, b),
2m dy 2

with eX + eY = E. The general solutions (assuming eX and eY are positive) are

X(x) = A sin(kx x) + B cos(kx x),

Y (y) = C sin(ky y) + D cos(ky y),

where
~2 2 ~2 2
k = eX , and k = eY .
2m x 2m y
Using the boundary conditions, we get

X(0) = 0 ⇒ B = 0,
X(a) = 0 = A sin(kx a) ⇒ kx a = nπ,
Y (0) = 0 ⇒ D = 0,
Y (b) = 0 = C sin(ky b) ⇒ ky b = n0 π,

with n, n0 , nonzero integers. We now use the normalization to find A and C.

Z a Z a   a r
x sin(2k x x) a 2
1= |X(x)|2 dx = A2 sin2 (kx x)dx = A2 −  = A2 ⇒ A = ± .

0 0 2 4kx  2 a
0

3
 Similarly,
Z b
r
2 2b 2
1= |Y (y)| dy = C ⇒C=± .
0 2 b
Then, choosing the positive normalizations, the solution is of the form
r
n0 π
 
4  nπ 
ψ(x, y) = X(x)Y (y) = sin x sin y .
ab a b

b) We now observe that

2meX  nπ 2 n2 ~2 π 2
kx2 = = ⇒ eX = ,
~2 a 2ma2
 0 2
2meY nπ n02 ~2 π 2
ky2 = 2
= ⇒ eY = .
~ b 2mb2
The energy levels are E = eX + eY , then

~2 π 2 n2 n02
 
E= + 2 .
2m a2 b

c) Assume a square box (a = b). Denote the energy levels by µnn0 , then

~2 π 2
n2 + n02 ,


µnn0 = 2
2ma
for n = 1, 2, . . . and n0 = 1, 2, . . .. The lowest energy level occurs for n = n0 = 1 and is

~2 π 2
µ11 = .
ma2
We then have
n2 + n02
 
µnn0 = µ11 .
2

The following levels corresponding to n and n0 up to 4 are:

5 18
µ12 = µ21 = µ11 , µ33 = µ11 ,
2 2
8 20
µ22 = µ11 , µ24 = µ42 = µ11 ,
2 2
10 25
µ13 = µ31 = µ11 , µ34 = µ43 = µ11 ,
2 2
13 32
µ23 = µ32 = µ11 , µ44 = µ11 ,
2 2
17
µ14 = µ41 = µ11 .
2

Note: Notice that some states are degenerate.

4
 r1 M + r2 m
2. a) Let R = and r = r1 − r2 . Then,
M +m
M r1 = (M + m)R − r2 m = (M + m)R − (r1 − r)m
m 1 Mm
r1 = R + r=R+ r.
M +m M M +m
Furthermore,

mr2 = (M + m)R − r1 M = (M + m)R − (r + r2 ) M,

M 1 mM
r2 = R − r=R− r.
M +m mM +m
Let m∗ = mM
M +m , then
m∗ m∗
r1 = R + r; r2 = R − r.
M m
b) Using the chain rule
∂ ∂ ∂Rk ∂ ∂rk M ∂ ∂ m∗ ∂ ∂
= + = δki + δki = + .
∂(r1 )i ∂Rk ∂(r1 )i ∂rk ∂(r1 )i M + m ∂Rk ∂rk m ∂Ri ∂ri
Similarly,
∂ ∂ ∂Rk ∂ ∂rk m ∂ ∂ m∗ ∂ ∂
= + = δki − δki = + .
∂(r2 )i ∂Rk ∂(r2 )i ∂rk ∂(r2 )i M + m ∂Rk ∂rk M ∂Ri ∂ri
Then, we have
m∗ m∗
∇r1 = ∇R + ∇r ; ∇r2 = ∇R − ∇ r .
m M
c) The Schrödinger’s equation for this two-particle system is

~2 ~2
 
∂Ψ
− ∆r1 − ∆r2 + V (r) Ψ(r1 , r2 , t) = i~ (r1 , r2 , t).
2M 2m ∂t

Assume a solution, using separation of variables, of the form Ψ(r1 , r2 , t) = ψ(r1 , r2 )e−iEt/~ .
Then, we get

~2 ~2
 
− ∆r1 − ∆r2 + V (r) ψ(r1 , r2 )e−iEt/~
 = Eψ(r1 , r2 ) e−iEt/~
.
2M 2m
Thus, the time independent Schrödinger’s equation is

~2 ~2
 
− ∆r − ∆r + V (r) ψ(r1 , r2 ) = Eψ(r1 , r2 ).
2M 1 2m 2
We now want to write this expression using the variables R and r. Then,
 ∗   ∗   ∗ 2
m m m m∗
∆r1 = ∇r1 · ∇r1 = ∇R + ∇r · ∇R + ∇r = ∆R + 2 ∇R · ∇ r + ∆ r ,
m m m m
 ∗   ∗   ∗ 2
m m m m∗
∆r2 = ∇r2 · ∇r2 = ∇R − ∇ r · ∇R − ∇ r = ∆R − 2 ∇R · ∇ r + ∆ r .
M M M M

5
 ~2 ~2 ~2 m∗ 2 ~2 ~2 m∗ 2 ~2
  
− ∆r1 − ∆r2 = − ∆R − ∆r − ∆R − ∆r
2M 2m 2M m 2M 2m M 2m
 !
~2 1 m∗ 2 1 m∗ 2 ~2 1
    
1
=− + ∆R − + ∆r
2 M m m M 2 m M
~2 ~2
 
M m
=− + ∆ R − ∆r
2 (M + m)2 (M + m)2 2m∗
~2 ~2
=− ∆R − ∆r .
2(M + m) 2m∗

Let ϕ(R, r) = ψ(r1 (R, r), r2 (R, r)), therefore we obtain

~2 ~2
 
− ∆R − ∆r + V (r) ϕ(R, r) = Eϕ(R, r) .
2(M + m) 2m∗

1 1
d) Assume that M  m and m  M +m , then

mM mM
m∗ = ≈ = m,
M +m M
~2 ~2 ~2 ~2
− ∆R − ∆ r ≈ − ∆ r ≈ − ∆r .
2(M + m) 2m∗ 2m∗ 2m

Then, the resulting Schrödinger’s equation involving only r is

~2
 
− ∆r + V (r) ϕ(r) = Eϕ(r) ,
2m

where we assume R is effectively constant.

e) We now take a look at the Schrödinger’s equation involving both R and r

~2 ~2
 
− ∆R − ∆ r + V (r) ϕ(R, r) = Eϕ(R, r)
2(M + m) 2m∗

and assume the solution is separable, i.e., ϕ(R, r) = ϕ(r)χ(R), then

~2 ~2
 
− ∆R − ∆r + V (r) ϕ(r)χ(R) = Eϕ(r)χ(R)
2(M + m) 2m∗
or
~2 ~2
− ϕ(r)∆R χ(R) − χ(R)∆r ϕ(r) + V (r)ϕ(r)χ(R) = Eϕ(r)χ(R).
2(M + m) 2m∗

Dividing the entire equation by ϕ(r)χ(R) (assuming ϕ(r)χ(R) 6= 0) we get

~2 ~2 1
   
1
− ∆R χ(R) + − ∗ ∆r ϕ(r) + V (r) = E.
2(M + m) χ(R) 2m ϕ(r)

6
Because each term depends on a different variable, we can only satisfy this equation if
each term is constant. We denote the constants Er and ER , which are required to satisfy
Er + ER = E, then

~2
− ∆R χ(R) + VR χ(R) = ER χ(R),
2(M + m)
~2
− ∗ ∆r ϕ(r) + V (r)ϕ(r) = Er ϕ(r),
2m
with VR = 0.