PHY 314: Introduction to Quantum Mechanics, Varsha 2014
Lecture 14
Anil Shaji
School of Physics, IISER Thiruvananthapuram
Quantum mechanics in three dimensions
We have been dealing with the time independent one dimensional Schrodinger equation so far.
The wave equation, in this form, is a potential dependent condition on the spatial derivatives of
the wave functions of stationary states. The time dependent version of the equation connects the
time derivative of the wave function to its spatial derivatives and the potential. Generalizing the
equation to three dimensions is straightforward in that spatial derivatives along all three dimensions
come into play and the potential becomes a function in three dimensional space. The Hamiltonian
accordingly becomes
H=
p2y
p~ 2
p2
p2
+ V (~r) = x +
+ z + V (x, y, z)
2m
2m 2m 2m
Identifying,
px = i~
,
x
py = i~
,
y
pz = i~
,
z
we get
p~ 2 = ~2
2
2
2
+
+
x2 y 2 z 2
= ~2 2 .
The Schrodinger equation in three dimension can then be written as
i~
(x, y, z)
~2 2
=
(x, y, z) + V (x, y, z)(x, y, z).
t
2m
(1)
The time independent version is
~2 2
(x, y, z) + V (x, y, z)(x, y, z) = E(x, y, z).
2m
I.
(2)
THE CENTRAL POTENTIAL
A scenario of much interest where we would like to solve the three dimensional Schrodinger
equation is when the potential depends only on the radial coordinate, r. These central potential,
V (r) arise naturally when considering several two body problems where one can extract out the
center of mass motion of the two particles. The remaining problem of the relative coordinate can
then be mapped into a one body problem with a central potential.
The technique of separation variables is a way of simplifying the differential equation in several
variables to a larger number of differential equations, each in fewer variables. We have already
employed this method to extract the time independent Schrodinger equation from the full time
2
dependent version. Since quantum mechanics is a linear theory, the set of solutions that one gets
using this technique can be superposed to obtain all solutions. If n (~r) are these solutions, then a
general solution of the time dependent Schrodinger equation has the form
X
(~r, t) =
cn n (~r)eiEn t/~ .
n
A.
The Schr
odinger equation in spherical polar coordinates
For applying the separation of variables technique, we have to first rewrite Eq. (2) in spherical
polar coordinates. The Laplacian in terms of r, and is
2
1
1
1
2
2
r
+ 2
sin
+ 2 2
.
= 2
r r
r
r sin
r sin 2
The time independent Schr
odinger equation is then
2
~2 1
1
1
2
r
+ 2
sin
+ 2 2
+ V (~r) = E.
2
2m r r
r
r sin
r sin 2
(3)
As mentioned earlier we look for solution of the form,
(r, , ) = R(r)Y (, ).
Putting this into Eq. (3), we get
2
~2 Y
Y
R
Y
R
2 R
r
+ 2
sin
+ 2 2
+ V (~r)RY = ERY.
2
2m r r
r
r sin
r sin 2
Dividing by RY and multiplying by 2mr2 /~2 we get
1 d
2mr2
2 dR
r
V (r) E
R dr
dr
~2
2
Y
1
1
Y
1
+
sin
+
= 0.
2
Y sin
sin 2
(4)
(5)
(6)
Since the first term in curly brackets depends only on r and the second term only on and , each
of them have to be independently equal to a separation constant that we choose to be of the form
l(l + 1),
1 d
2mr2
2 dR
r
V (r) E = l(l + 1)
2
R dr
dr
~
2
1
1
Y
1
Y
sin
+
= l(l + 1).
2
Y sin
sin 2
II.
THE ANGULAR EQUATION
The angular equation is
1
Y (, )
1 2 Y (, )
sin
+
= l(l + 1) Y (, ).
sin
2
sin2
(7)
3
We can write this equation as
1
1 2
sin
+
Y (, ) = l(l + 1) Y (, ).
sin
sin2 2
(8)
Before we try to solve the angular equation we notice that this is again an eigenvalue equation
of the form
L, Y (, ) = l(l + 1)Y (, ),
where
L,
1
=
sin
sin
+
1 2
sin2 2
Let us take a brief diversion and investigate what this operator is. We turn to classical mechanics
and the classical central force problem for hints as to what this operator might be. The stable
configurations of a classical particle in a central potential are closed orbits around the origin. These
orbits are the analogue of the quantum stationary states that we are interested in. Each orbit is
in turn characterized by its radius (or semi-major and semi-minor axes in case of elliptical orbits)
and the orientation of the plane of the orbit. The radius, in turn, depends on the magnitude of
the angular momentum and the orientation is fixed by the direction of the angular momentum of
the particle.
Taking a cue from classical mechanics, one suspects after looking at the radial and angular
equations, that the separation constant, l(l + 1) connecting the two equations has something to do
with angular momentum. So we guess that the operator L, has something to do with angular
momentum.
A.
The angular momentum operator
In classical mechanics, the angular momentum is
~ = ~r p~,
L
In quantum mechanics the variables ~r and p~ are replaced by the operators r
r and i~ respectively.
In spherical polar coordinates,
= r
1
1
+
+
r
r sin
we get
Using r r = 0, r = and r = ,
= i~
L
.
sin
Writing the unit vectors, and in cartesian coordinates as
= (cos cos )i + (cos sin )j cos k
= (sin )i + (cos )j,
(9)
4
we get
= i~ ( sin i + cos j) (cos cos i + cos sin j cos k) 1 ,
L
sin
Collecting terms together we get
Lx = i~ sin
+ cos cot
Ly = i~ cos
+ sin cot
Lz = i~ .
We also have
2
2
sin
cos
cot
+
2
sin
cos
cot
+ cos2 cot
2
2
sin cos cot
+ cos2 cot2 2
2
2
L2y = ~2 cos2 2 + sin cos cot2
2 sin cos cot
+ sin2 cot
2
2
2
+ sin cos cot
+ sin cot 2
L2z = ~2 2 .
L2x
= ~
sin2
Adding the three equations together, we get
1 2
1
sin
+
.
L2 = ~2
sin
sin2 2
(10)
So we identify
L, =
1 2
L
~2
and we find that Y (, ) are actually eigenfunctions of the L2 operator with eigenvalues l(l + 1)~2 .
