0% found this document useful (0 votes)
95 views4 pages

PlaneWaveExpansion PDF

1) A plane wave solution to the Schrodinger equation for a free particle can be expressed as a sum of spherical harmonics. 2) The coefficients in this expansion, clm(k), are determined by relating the plane wave to the spherical Bessel functions and Legendre polynomials. 3) It is shown that clm(k) = il4π(2l+1)Ylm(k)*, providing a partial wave expansion of the plane wave in terms of incoming and outgoing spherical waves.

Uploaded by

Francisco Torres
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
95 views4 pages

PlaneWaveExpansion PDF

1) A plane wave solution to the Schrodinger equation for a free particle can be expressed as a sum of spherical harmonics. 2) The coefficients in this expansion, clm(k), are determined by relating the plane wave to the spherical Bessel functions and Legendre polynomials. 3) It is shown that clm(k) = il4π(2l+1)Ylm(k)*, providing a partial wave expansion of the plane wave in terms of incoming and outgoing spherical waves.

Uploaded by

Francisco Torres
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Physics 215 Winter 2018

Expansion of plane waves in spherical harmonics

Consider a free particle of mass µ in three dimension. The time-independent Schrodinger


equation for the energy eigenstates in the coordinate representation is given by
~ 2 + k 2 )ψ~ (~
(∇ k r) = 0 , (1)
corresponding to an energy E = ~2 k 2 /(2µ). The solution to eq. (1) is a plane wave,
1 ~
ψ~k (~
r) = eik ·~r , (2)
(2π) 3/2

where the wave function is conventionally normalized such that


Z
d3 x ψ~k∗ (~x) ψ~k ′ (~x) = δ 3 (~
k −~

k ).

One can also solve eq. (1) in spherical coordinates. If we look for simultaneous eigenstates
of the free particle Hamiltonian, and the angular momentum operators Lz and L ~ 2 , we obtain
 1/2
ℓ 2µk
ψ~k (r, θ, φ) = hr, θ, φ| Eℓmi = i jℓ (kr)Yℓm (θ, φ) , (3)
π~2
where the normalization factor has been chosen such that hE ′ ℓ′ m′ | Eℓmi = δℓℓ′ δmm′ δ(E − E ′ ),
and the factor of iℓ is conventional. In particular, since the free particle Hamiltonian commutes
with the angular momentum operators Lz and L ~ 2 , it follows that any choice of ℓ and m in
eq. (3) yields an energy eigenstate of energy E = ~2 k 2 /(2µ).
Hence, it must be possible to express the plane wave given in eq. (2) as a sum over spherical
harmonics,1
∞ X ℓ
i~
X
e k ·~
r
= cℓm (~
k)jℓ (kr)Yℓm (r̂) . (4)
ℓ=0 m=−ℓ

The object of these notes is to determine the coefficients cℓm (~ k).


It is always possible to choose the z-axis of our problem to lie along ~ k. For ~k = k ẑ,
~
k·~r = kr cos θ (where r ≡ |~ r|) and θ is the polar angle of the vector ~ r with respect to the
z-axis. Hence, the double sum in eq. (4) must be independent of the azimuthal angle φ. This
is possible only if cℓm (k ẑ) = 0 for all m 6= 0. That is, only the m = 0 term of eq. (4) survives
and it follows that
∞  1/2
ikr cos θ
X 2ℓ + 1
e = Aℓ (k)jℓ (kr)Pℓ (cos θ) , (5)
ℓ=0

where Aℓ (k) ≡ cℓ0 (k ẑ), and we have employed the relation between Yℓ0 (θ, φ) and the Legendre
polynomial, Pℓ (cos θ).
1
since r̂ ≡ x̂ sin θ cos φ + ŷ sin θ sin φ + ẑ cos θ, it is convenient to write Yℓm (r̂) in place of Yℓm (θ, φ).

1
We can extract the coefficient Aℓ (k) by using the orthogonality relation of the Legendre
polynomials, Z 1
2
Pℓ (cos θ)Pℓ′ (cos θ) d cos θ = δℓℓ′ . (6)
−1 2ℓ + 1
Multiplying both sides of eq. (5) by Pℓ′ (cos θ) and then integrating over cos θ with the help of
eq. (6), we end up with
p Z 1
Aℓ (k)jℓ (kr) = π(2ℓ + 1) Pℓ (w)eikrw dw , (7)
−1

where w ≡ cos θ.
There are a number of different ways to obtain Aℓ (k). One technique, which involves a
direct evaluation of the integral on the right hand side of eq. (7), is given in Appendix A. The
end result is obtained in eq. (19), which we repeat here,
Z 1
Pℓ (w)eikrw dw = 2iℓ jℓ (kr) (8)
−1

Comparing eqs. (7) and (8), we conclude that


p
Aℓ (k) = iℓ 4π(2ℓ + 1) . (9)
However, one can obtain the same result by employing the following trick. Since Aℓ (k) is
independent of r, we can evaluate Aℓ (k) by examining the r → ∞ behavior of both sides of
eq. (7). The large r behavior of the left hand side is determined by the leading term of the
asymptotic expansion of jℓ (kr),
 
1 1
 1
jℓ (kr) ∼ sin kr − 2 ℓπ + O , as r → ∞. (10)
kr (kr)2
We can determine the leading asymptotic behavior of the integral on the right hand side of
eq. (7) by a repeated integration by parts,
Z 1 1 Z 1
ikrw 1 ikrw 1
eikrw Pℓ′ (w)

Pℓ (w)e dw = e Pℓ (w) −

−1 ikr −1 ikr −1
1 Z 1
1  ikr −ikr iπℓ 1 ikrw ′ 1
eikrw Pℓ′′ (w)

= e −e e + e Pℓ (w) −

ikr (kr)2 −1 (kr) 2
−1

2iℓ
 
 1
= sin kr − 21 πℓ + O , (11)
kr (kr)2
where we have used Pℓ (1) = 1 and Pℓ (−1) p = (−1)ℓ = eiπℓ . Finally, in light of eqs. (10) and
(11), it follows from eq. (7) that Aℓ (k) = iℓ 4π(2ℓ + 1), in agreement with eq. (9).
Inserting eq. (9) back into eq. (5), we end up with

X
eikr cos θ = iℓ (2ℓ + 1)jℓ (kr)Pℓ (cos θ) . (12)
ℓ=0

2
Finally, we can relax the assumption that ~
k = k ẑ by employing the addition theorem,

4π X
Pℓ (cos θ) = Yℓm (r̂)Yℓm(k̂)∗ , where cos θ = k̂· r̂.
2ℓ + 1 m=−ℓ

Inserting the addition theorem for Pℓ (cos θ) into eq. (12) yields our final result,


∞ X
~
X
eik·~r = 4π iℓ jℓ (kr)Yℓm (k̂)∗ Yℓm (r̂) . (13)
ℓ=0 m=−ℓ

That is, we have identified cℓm (~k) = 4πiℓ Yℓm (k̂)∗ in eq. (4).
Eqs. (12) and (13) are both called the partial wave expansion of the plane wave. It is
interesting to note that in light of eq. (10), which we can rewrite as
 ikr  −ikr    
1 e e iπℓ 1
jℓ (kr) ∼ ℓ+1 − e +O , as r → ∞,
2i kr kr (kr)2

it follows from eq. (13) that a plane wave moving in the direction ~
k can be decomposed into
a linear combination of incoming and outgoing spherical waves.

APPENDIX A: Evaluation of the integral appearing in eq. (7)

The spherical Bessel function is given by the following formula,


 ℓ  
ℓ 1 d sin ρ
jℓ (ρ) = (−ρ) . (14)
ρ dρ ρ

Thus, we can write


1
sin ρ 1
Z
j0 (ρ) = = eiρw dw . (15)
ρ 2 −1

Next, we note that by an integration by parts,


Z 1
i 1
Z 1
1 d i
Z
iρw iρw
e dw = w e dw = eiρw d(w 2 − 1)
ρ dρ −1 ρ −1 2ρ −1
Z 1
i 1 1 2
Z
iρw
=− 2
(w − 1)d(e ) = (w − 1)eiρw dw .
2ρ −1 2 −1

Repeating this process ℓ-times yields


 ℓ Z 1 1
1 d 1
Z
iρw
e dw = ℓ (w 2 − 1)ℓ eiρw dw . (16)
ρ dρ −1 2 ℓ! −1

3
Applying eq. (16) to eq. (15), it then follows from eq. (14) that
ℓ Z 1
(−ρ)ℓ 1 2

1 d 1
Z
ℓ iρw
jℓ (ρ) = (−ρ) e dw = ℓ+1 (w − 1)ℓ eiρw dw
ρ dρ 2 −1 2 ℓ! −1
iℓ ℓ
(−i)ℓ 1 iρw dℓ
Z 1
ℓ d
Z
2 iρw
ℓ
w 2 − 1 dw ,

= ℓ+1 (w − 1) ℓ
e dw = ℓ+1 e ℓ
(17)
2 ℓ! −1 dw 2 ℓ! −1 dw

after integration by parts ℓ times. Finally, we employ the Rodrigues formula for the Legendre
polynomials,
1 dℓ 2
ℓ
Pℓ (w) = ℓ w − 1 .
2 ℓ! dw ℓ
Then, eq. (17) yields
(−i)ℓ 1
Z
jℓ (ρ) = Pℓ (w)eiρw dw . (18)
2 −1

We have thus succeeded in evaluating the integral that appears in eq. (7),
Z 1
Pℓ (w)eikrw dw = 2iℓ jℓ (kr) (19)
−1

REFERENCE
Kevin Cahill, Physical Mathematics (Cambridge University Press, Cambridge, UK, 2013)

You might also like