Next: Exercises
Up: Spin Angular Momentum
Previous: Factorization of Spinor-Wavefunctions
Spin Greater Than One-Half Systems
We have seen how to deal with a spin-half particle in quantum mechanics.
But, what happens if we have a spin one or a spin three-halves particle?
It turns out that we can generalize the Pauli two-component scheme in a fairly
straightforward manner. Consider a spin-
particle: that is, a particle for which
the eigenvalue of
is
. Here,
is either an integer, or a half-integer. The eigenvalues of
are written
, where
is allowed to take the values
. In fact,
there are
distinct allowed values of
. Not surprisingly, we can represent
the state of the particle by
different wavefunctions, denoted
. Here,
specifies the probability density
for observing the particle at position
with spin angular
momentum
in the
-direction. More exactly,
 |
(5.111) |
where
denotes a state ket in the product of the position
and spin spaces. The state of the particle can be represented more
succinctly by a spinor-wavefunction,
, which is simply the
component column
vector of the
.
Thus, a spin one-half particle is represented by a two-component spinor-wavefunction,
a spin one particle by a three-component spinor-wavefunction, a spin three-halves particle
by a four-component spinor-wavefunction, and so on.
In this extended Schrödinger/Pauli
scheme, position space operators take the form of diagonal
matrix differential operators. Thus, we can represent the momentum operators
as
 |
(5.112) |
where
is the
unit matrix. [See
Equation (5.91).]
We represent the spin
operators as
 |
(5.113) |
where the
extended Pauli matrix
(which is, of course, Hermitian) has elements
 |
(5.114) |
Here,
,
are integers, or half-integers, lying in the range
to
.
But, how can we evaluate the brackets
and, thereby, construct the extended Pauli matrices? In fact, it is trivial
to construct the
matrix. By definition,
 |
(5.115) |
Hence,
 |
(5.116) |
where use has been made of the orthonormality property of the
.
Thus,
is the suitably normalized diagonal matrix of the eigenvalues
of
. The elements of
and
are most easily
obtained by considering the ladder operators,
 |
(5.117) |
We know, from Equations (4.55)-(4.56), that
It follows from Equations (5.114), and (5.117)-(5.119), that
According to Equations (5.116) and (5.120)-(5.121), the Pauli matrices for a spin one-half
(
)
particle (e.g., an electron, a proton, or a neutron) are
as we have seen previously. For a spin one (
) particle (e.g., a
-boson or a
-boson), we find that
In fact, we can now construct the Pauli matrices for a particle of arbitrary spin.
This means that we can convert the general energy eigenvalue problem for a spin-
particle, where the Hamiltonian is some function of position and spin operators,
into
coupled partial differential equations involving the
wavefunctions
. Unfortunately, such a system
of equations is generally too complicated
to solve exactly.
Next: Exercises
Up: Spin Angular Momentum
Previous: Factorization of Spinor-Wavefunctions
Richard Fitzpatrick
2016-01-22