Homework 11 – PHY 6645 – Fall 2010

These problems are worth 20 points each. Your score on the bonus problem
will be added into your homework score.
(Due Wednesday, Dec. 1, 2010 at 6 PM in the grader’s mailbox.)

Caution: There may be lots of “typos” in these problems. Email me if you have any questions.

For this assignment, the problems are a bit harder than the usual problems and are meant as a challenge to
the top students who might find the usual problems “too easy”. You are required to turn in the first
problem. I realize that many of you are busy at the end of the term so the second problem is a bonus and
should be done only if you have the time (it is worth 20 extra points). Due to Thanksgiving, this
homework will be due the Wednesday after Thanksgiving.

1. (Hydrogen atom via operator method).

a. Solve for the bound state energy eigenvalues of the hydrogen atom using the factorization
method as follows. In atomic units, the radial wave equation is

⎡ 1 d 2 l (l + 1) 1 ⎤
⎢− 2
+ − ⎥ un ,l ≡ hn ,l un ,l = En ,l un ,l
⎣ 2 dr 2r 2 r⎦

d l 1
where u (r ) = r R(r ) . Introduce the operators D±(l ) ≡ ∓ ± . Show that
dr r l

1 1
D−(l +1) D+(l +1) = − 2hl − ; D+(l ) D−(l ) = − 2hl − .
(l + 1) 2 l2

Hence, show that hl +1 D+( l +1)un ,l (r ) = En ,l D+( l +1)un ,l ( r ) , that is D+( l +1) un ,l (r ) ∝ un ,l +1 ( r ) and that
En ,l is independent of l.

b. Give an argument showing that for every En ,l < 0 there is a maximum l. (Hint: examine the
effective potential for radial motion.) Call this l value ln . Defining n = ln + 1 show that

1
En ,l = − l = 0,....., n − 1 .
2n 2

c. Continue the above game to find the eigenfunctions. Use D+( n ) un ,n −1 (r ) = 0 to show that
∞
n −r /n
un , n −1 (r ) = A r e . Work out the value of A so that un ,n −1 is normalized, i.e. ∫ dr un2, n −1 = 1.
0

d. Prove that D−(l ) un ,l ∝ un ,l −1 . By considering the diagonal matrix element of the operator
† n2 − l 2
D+(l ) D−( l ) , and using the fact that D+(l ) = − ⎡⎣ D−(l ) ⎤⎦ show that D−(l ) n, l = n, l −1
nl
where n, l , n, l − 1 are normalized kets.
2. (Bonus: Constructing Spherical Harmonics from X, Y, Z). In this problem, we will look at a
way to construct and view the spherical harmonics Yl m (θ , φ ). For simplicity, we will work in
units where = 1 . Since the Yl m are eigenfunctions of L2 and Lz , they satisfy:

L2Yl m = −(r × ∇)i(r × ∇)Yl m = l (l + 1)Yl m (2.1)

LzYl = zˆ i(r × ∇)Yl = mYl
m m m
(2.2)

We will look for solutions to (2.1) and (2.2) that are homogeneous polynomials in
X = r sin θ cos φ , Y = r sin θ sin φ , Z = r cos θ of a given degree l. That is,

Y lm (r ) = ∑
a ,b , c
Cabc X aY b Z c . (2.3)
a ,b , c > 0
a +b + c =l

Where the Cabc are the coefficients of the individual terms. Note that the homogeneity of degree l
implies that Y lm (λ X , λY , λ Z ) = λ l Y lm ( X , Y , Z ) so that in particular, Y lm (r ) = r l Y ml (θ , φ ) .
The purely angular parts are the spherical harmonics. Note that the r l piece does not affect the
fact that the Y lm (r ) are still solutions to (2.1) and (2.2) since for any function f of r (the
magnitude of r), it follows that ( r × ∇ ) f (r ) = 0.

a. Derive the vector identity: −(r × ∇)i(r × ∇)ϕ = ⎡ − r 2∇ 2 + ( r i∇ ) + ( r i∇ ) ⎤ ϕ for any

2
⎣ ⎦
scalar field ϕ (r ) .

b. Recall Euler’s Theorem for homogeneous functions. If

f (λr ) = λ n f (r ) (2.4)
then
( ri∇ ) f (r ) = n f (r ). (2.5)
Prove (2.5) by taking the derivative of equation (2.4) with respect to λ and then setting λ = 1.
Note that this says that homogeneous functions of degree n are eigenfunctions of ( r i∇ ) with
eigenvalue n.

c. Use the results of (2.5) to show that homogeneous polynomials in X, Y, Z of the form (2.3)
are eigenfunctions of L2 with eigenvalue l(l+1) provided ∇ 2 ∑
Cabc X aY b Z c = 0. Because
a ,b , c
a ,b , c > 0
a +b + c =l
of this last constraint, we note that the spherical harmonics obtained this way are exactly the same
as those obtained by solving Laplace’s equation in spherical coordinates.

d. Note that part c. above allows us to figure out whether homogeneous polynomials of degree l
in X, Y, Z (neglecting the r l piece) are linear combinations of spherical harmonics of order l.
Hence, X and Y are linear combinations of spherical harmonics of order 1,
( X ∝ Y11 + Y1−1 , Y ∝ Y11 − Y1−1 ) andXY is a linear combination of spherical harmonics of
2 2
order 2, but X Y is not a linear combination of spherical harmonics of order 4. Express XY as a
linear combination of spherical harmonics of order 2.

e. We have shown that solutions of the form (2.3) that satisfy Laplace’s equation are
eigenfunctions of L2 . We want to look at what further constraints we must have so that they are
also eigenfunctions of Lz . To do this, it is convenient to make a change of variables,
ζ = X + iY
(2.6)
η = X − iY
Note that Lz ζ = ζ , Lzη = − η . We can now write (2.3) in a slightly different form:
Y lm (r ) = ∑
a ,b , c > 0
Cabc ζ aη b Z c (2.7)
a +b + c =l

These are still homogeneous polynomials of degree l in X, Y, Z (though the coefficients Cabc in
(2.3) and (2.7) are not the same). Consider a single term in (2.7). Show that
Lz ζ aη b Z c = ( a − b ) ζ aη b Z c , (2.8)
that is, ζ aη b Z c is an eigenfunctions of Lz with eigenvalue (a-b). For (2.7) to be eigenfunctions
of Lz means that each term must be an eigenfunction with the same eigenvalue. Thus we see that
we can write the spherical harmonics (eigenfunctions of L2 and Lz ) as
r l Y ml (θ , φ ) = ∑
a ,b , c , > 0
Cabc ζ aη b Z c (2.9)
a +b + c =l
a −b = m

with the coefficients Cabc satisfying ∇ 2 ∑

a ,b , c > 0
Cabc ζ aη b Z c = 0.
a +b + c =l
a −b = m

f. For m=l there is only one possible term in the sum: Y ll (θ , φ ) ∼ ζ l = eilφ sin l θ . Show that
∂ ∂
in terms of η , ζ that one can write L− = η − 2z . Use this and the above to find all Y2m
∂z ∂ζ
in terms of η , ζ , Z and then in terms of X , Y , Z . Do not worry about the normalization.