Chapter 6: Operator Methods in Quantum Mechanics
Ladder operators (a raising or lowering operator) are used to determine the quantized
energy levels of Hamiltonians without solving Schrodinger differential equation. They
play important role to the quantum harmonic oscillator and angular momentum.
The Hamiltonian of a particle of mass m moving in a one-dimensional harmonic potential
is
p2 1
H= + mω 2 x 2
2m 2
where the quantum mechanical operators p and x satisfy the commutation relation
p, x = −i . The problem is how to find the energy eigenvalues and eigenstates of this
Hamiltonian. This problem can be studied by two separate methods
1. Polynomial method
2. Operator method
In the polynomial way, the Schrodinger equation is taken in the simple form
d 2u (y)
2
+ ( − y 2 )u ( y ) = 0
dy
mw 2E
where y = x and = . We follow the steps to find the eigenfunctions and the
w
energy eigenvalues below:
• Look at the asymptotic behavior of u ( y )
• Define the solution u ( y ) = Ae − y /2 H ( y )
2
• Use power series solution H ( y ) = am y m to obtain recursion relation
m=0
(m + 1)(m + 2)am+2 − 2mam + ( − 1)am = 0
am+ 2 2
• H ( y)
2
For large m, ey
am m
− y /2 y
• e = lim Ae y → require that the series must terminate.
2 2 2
/2
lim Ae
y → y →
• Using the cut off condition on the series as
2n − + 1
mmax = n an + 2 = an = 0 2n − + 1 = 0 gives the energy values as
(n + 1)(n + 2)
En = ( n + 12 ) w, n = 0,1, 2,...
• The solution of the differential equation gives the eigenfunctions as
un ( x ) = e− H n ( x)
2 2
x /2
2 n!
n
1
Operator Method
This method does not deal with solving the Schrodinger equation, it deals with operator
algebra involving operators as the creation and annihilation operators (or called ladder
operators).
2
d 2u ( x ) 1
−
2m dx 2 + 2 mw x u ( x) = Eu ( x) contains three
2 2
The Schrodinger equation
constants: , w and m. With these constants we can construct an energy w and a length
/ mw . Thus, the measurement of energy in units of w becomes E = w and the
length x in units of / mw become x = q / mw . The quantities and q are
dimensionless.
Let us define dimensionless position and momentum operators
mω
q= x
1
p= p x = -i
mω q
We now introduce two non-Hermitian operators in terms of q and p as
1 1 mω 1
 = (q + ip) = x+i p x
2 mω
2
1 1 mω 1
 + = (q − ip) = x − i p
2
x
2 mω
Since x and p are Hermitian, Â+ is indeed the Hermitian conjugate of  .
Let us derive the commutator Â, Â +
1 mw px 1 mw px
A, A† = x+i , x −i
2 mw 2 mw
1 mw i i i2
= x, x − x, px + px , x − px , px
2 2 2 2mw
0 i −i 0
i
= (−2i )
2
AA† = A† A + 1
A, A = 1 AA − A A = 1 †
† † †
A A = AA − 1
†
2
The Hamiltonian in terms of these operators
1 mw px
A= x+i x= A + A†
2 mw 2mw
1 mw px 1 mw mw
A† = x −i px = A − A† = i A† − A
2 mw i 2 2
px2 1
H= + mw2 x 2
2m 2
2 2
1 1 mw 1
= A − A† + mw2 A + A†
2m i 2 2 2mw
1 mw 2 † 1 2 †
=− A + A − AA − A A + mw
†2 † 2
A + A + AA + A A
†2 †
2m 2 A† A+1 2 2mw A† A +1
w 2 w 2
=− A + A†2 − 2 A† A − 1 + A + A†2 + 2 A† A + 1
4 4
w 1
= 4 A† A + 2 = w A† A +
4 2
1
H = w A† A +
2
where N̂ = Â + Â is known as the number operator or occupation number operator which
is Hermitian. Since Nˆ = A† A is Hermitian, it has real eigenvalues, and its
eigenfunctions are orthogonal.
Consider the commutator H , Nˆ ,
H , Nˆ = w( Nˆ + ), Nˆ = w 1
1
A† A, A† A + w , A† A
2 2
A† A, A† A + A† , A† A A
0
A† A , A+ A , A† A A† A† , A + A† , A† A
0
1 −1 0
H , Nˆ = w A A − A A = 0
† †
H , Nˆ = 0 requires that they have simultaneous eigenfunctions (the eigenfuctions of H
are also eigenfunctions of N̂ ).
H n = E n n and Nˆ n = n n
where n un and n = 0,1, 2,3,...
3
The other commutation relations,
1 1
H , A = w A† A + , A = w A† A, A + w , A = − wA
2 2
A† A, A+ A† , A A
0
0 −1
1 1
H , A† = w A† A + , A† = w A† A, A† + w , A† = wA†
2 2
A† A, A† + A† , A† A
0
1 0
H , A = − wA AH = HA + wA
H , A† = wA† A† H = HA† − wA†
Consider the eigenvalue equation of Hamiltonian, H n = E n n . Let us multiply both
sides with Â
AH n = A E n n H A n = (E n − w) A n
HA + wA En A
Note that A n is an eigenfunction of H with the eigenvalue ( E n − ω) such that the
energy E n is lowered by one unit of ω . Multiply again with Â
AH A n = A(E n − w) A n H A2 n = (E n − 2 w) A2 n
HA+ w
(E n − w ) A
where A n is also an eigenfunction of H with the eigenvalue ( E n − 2ω) such that the
2
energy E n is lowered by 2ω .
We thus find that if n is an eigenstate of H with the eigenvalue E n , then A n and
A2 n are also eigenstates of H with eigenvalues (E n − w) and (E n − 2 w) , respectively.
So that we get the picture of a ladder going downward in energy. The operator Â
therefore is called a lowering operator.
Since the harmonic oscillator has only positive energy states including zero, it is clear
that there must be a lower bound on the energy of the simple harmonic oscillator. This
means that there is a state of lowest energy, the ground state. We denote it by 0 which
must satisfy
 | 0 = 0
So that energy cannot be lowered any more.
4
1
H 0 = w A† A + 0
2
1
= wA† A 0 + w 0
2
0
1 1
= w 0 E0 = w
2 2
Now, let us consider H n = E n n and multiply both sides with  +
A† H n = A† E n n H A† n = (E n + w) A† n
† †
HA − wA E n A†
Note that A† n is an eigenfunction of H with the eigenvalue (E n + w) such that the
energy E n is raised by one unit of ω . Multiply again with A†
A† H A† n = A† (E n + w) A† n H A†2 n = (E n + 2 w) A†2 n
†
HA − w
(E n + w ) A†
where A†2 n is also an eigenfunction of H with the eigenvalue (E n + 2 w) such that the
energy E n is raised by 2ω .
We thus find that if n is an eigenstate of H with the eigenvalue E n , then A† n and
A†2 n are also eigenstates of H with eigenvalues (E n + w) and (E n + 2 w) ,
respectively. So that we get the picture of a ladder going upward in energy. The operator
A† therefore is called a raising operator.
As a result, A and A† are respectively known as the lowering and raising operators, or
the annihilation and creation operators. They are also known as the ladder operators.
A† n = Cn n + 1
A n = Dn n − 1
We conclude that the energy eigenvalues of the harmonic oscillator are
1
E n = n + ω n = 0,1,2,....
2
We obtained the energy spectrum of the harmonic oscillator without solving any
differential equation.
Question: What are the values of C n and D n ?
5
Consider the following commutator,
Nˆ , A† = A† A, A† = A† A, A† + A† , A† A = A† NA
ˆ † = A† + A† Nˆ
1 0
Nˆ , A = A A, A = A A, A + A , A A = A NA
† † † ˆ = − A + ANˆ
0 1
and consider Nˆ ( A† n ) and Nˆ ( A n ) as
ˆ † n = ( A† + A† Nˆ ) n = A† n + A† Nˆ n NA
NA ˆ † n = (n + 1) A† n
A† + A† N nn
ˆ n = (− A + ANˆ ) n = − A n + A Nˆ n NA
NA ˆ n = (n − 1) A n
− A+ AN
nn
Let us consider the above relations and the eigenvalue equation of number operator,
A† n = Cn n + 1 n A = Cn* n + 1
n A A† n = Cn* n + 1 Cn n + 1
n AA† n = Cn*Cn n + 1 n + 1
1+ A† A
1
Nˆ
n n + n Nˆ n = Cn Cn = n + 1
2
1 nn
n nn
1
A n = Dn n − 1 n A† = Dn* n − 1
n A† A n = Dn* n − 1 Dn n − 1
n A† A n = Dn* Dn n + 1 n + 1
Nˆ n = n n 1
n n n = Dn Dn = n
2
We use Nˆ , A† = A† and Nˆ , A = A , we then obtain the coefficients as Cn = n + 1 and
Dn = n . Thus, when the ladder operators are applied to a state they give the following
property
A† n = n + 1 n + 1
A n = n n −1
6
We can produce an infinite set of energy eigenstates by acting on |0⟩ with the raising
operator and multiplying by suitable normalization factors. We use this property to find
the eigenstates in terms of raising operator and 0 as
1
A† n = n + 1 n + 1 n + 1 = A† n
n +1
1
n +1 = A† n
n +1
1 1 †
n = 0: 0 +1 = A† 0 1 = A 0
0 +1 1
1 1
n = 1: 1+1 = A† 1 2 = A†2 0
1+1 1
1.2
A† 0
1
1 1
n = 2: 2 +1 = A† 2 3 = A†3 0
2 +1 1 †2
1.2.3
A 0
1.2
...
1 1 †n
n = n − 1: n − 1 + 1 = A† n −1 n = A 0
n −1+1 1
n!
A†( n−1) 0
1.2.3..n
1 †n
For a given n, the eigenstate becomes n = A 0 where the state n is obtained by
n!
acting the raising operator on the ground state.
We can find the explicit forms of the eigenstates by considering the definition of ground
sate as Aˆ 0 = 0 . The lowering operator has the following form
1 mw px 1
A= x+i q + ip
2 mw 2
d
where q and p = i are dimensionless position and momentum operators.
dq
1
Aˆ 0 = 0 q + ip u0 = 0
2
1 d du0 du0
q + i (−i ) dq u0 = 0 dq + qu0 = 0 u = −qdq
2 0
du0 q2 u0 q2
u0 = − = − + = − u0 = C0e − q /2
2
qdq ln u 0 ln C0 ln
2 C0 2
q2
ln u0
2
We then obtain ground sate wave function as u0 = C0e − where 2 = mw / .
2 2
x /2
7
The ground state has a Gaussian form as u0 = C0 e − q
2
/2
. Let us return to the form of nth
1 †n 1 mw px 1 1 d
sate n = A 0 where A† = x −i q − ip = q −
n! 2 mw 2 2 dq
and 0 u0 = C0e− q
2
/2
n
1 1 d
n
1 †n − q 2 /2 1 d − q 2 /2
n = A 0 n = q − dq C0e = q − dq C0e
n! n! 2 2 n!
n
This can be normalized and can be expressed in terms of Hermite polynomials so that
u n ( q ) = N n H n ( q )e − q
2
/2
The time dependence of operators:
The equation that describes the evaluation of a system is the time-dependent Schrodinger
equation
d
i (t ) = H (t )
dt
where (t ) is a vector in an abstract space and it depends on time. The solution of time
dependent Schrodinger equation is
(t ) = U (t , 0) (0)
= e−iHt / (0)
where (0) is the vector at time t = 0 , and the operator in front is defined by
1
e− iHt / = (−iHt / ) n
n =0 n !
The solution (t ) allows us to describe the change with time of the expectation value
of any operator B̂ that does not have any explicit time dependence
B t = (t ) Bˆ (t ) = (0) e+ iHt / Be
ˆ −iHt / (0) = Bˆ (t )
0
(0) e+ iHt / e− iHt / (0) Bˆ ( t )
The expectation value of a time-independent operator B̂ in a state (t ) may be written
as the expectation value of a time-varying operator Bˆ (t ) in the time-independent state
(0) .
8
In the Heisenberg picture, operator Bˆ (t ) is time-dependent but the state (0) is time-
independent. Time evaluation of the system is determined by a time-independent wave
function. However, in the Schrodinger picture, the operator B̂ is time-independent but the
state (t ) is time-dependent. Thus, time evaluation of the system is determined by a
time-dependent wave function. The result is the same whatever picture we use.
ˆ −iHt / becomes
Time variation of the operator B(t ) = e+iHt / Be
d ˆ d iH + iHt / ˆ −iHt / dBˆ S −iHt / iH + iHt / ˆ −iHt /
BH (t ) = e + iHt / Bˆ S e −iHt / = e BS e + e + iHt / e − e BS e
dt dt ˆ
dt ˆ
BH ( t ) BH ( t )
0
i
= ( HBˆ H (t ) − Bˆ H (t ) H )
d ˆ i
BH (t ) = H , Bˆ H (t )
dt
where H and S is used for the Heisenberg and Schrodinger pictures, respectively.
For the harmonic oscillator, the time-evaluation of the ladder operators becomes
1
H = w A† A +
2
d † i
AH† (t ) = e + iHt / AS†e − iHt / AH (t ) = H , AH† (t ) = iwAH† (t )
dt
wAH† ( t )
d † dA†
AH (t ) = iwAH† (t ) †H = iw dt AH† (t ) = eiwt A† (0)
dt AH
d i
AH (t ) = e + iHt / AH e − iHt / AH (t ) = H , AH (t ) = −iwAH (t )
dt
− wAH ( t )
d dA
AH (t ) = −iwAH (t ) H = −iw dt AH (t ) = e − iwt A(0)
dt AH
d ˆ i − iwt
AH (t ) = e AS
BH (t ) = H , Bˆ H (t ) †
AH (t ) = e AS
†
iwt
dt
9
Example: position and momentum operators of a particle
mw
px
A= x+i x = A + A†
2
2mw 2mw
mw px 1 mw mw
A† = x −i p = A − A† = i A† − A
2
2mw x
i 2 2
− iwt
x(t ) = A(t ) + A† (t ) = e AS + eiwt AS†
2mw 2mw cos wt −i sin wt cos wt + i sin wt
mw p (0) mw p (0)
= (cos wt − i sin wt ) x(0) + i x + (cos wt + i sin wt ) x(0) − i x
2mw 2 2mw 2 2mw
1 1
= 2 cos wt x(0) + 2sin wt px (0)
2 2mw
px (0)
x(t ) = x(0) cos wt + sin wt
mw
1 mw 1 mw −iwt
px (t) = A(t ) − A† (t ) = e AS − eiwt AS†
i 2 i 2 cos wt −i sin wt cos wt + i sin wt
1 mw mw p (0) mw p (0)
= (cos wt − i sin wt ) x(0) + i x − (cos wt + i sin wt ) x (0) − i x
i 2 2 2mw 2 2mw
1 mw 2ipx (0) 1 mw mw
= cos wt + (−2i sin wt ) x(0)
i 2 2mw i 2 2
px (t ) = px (0) cos wt -mw x(0) sin wt
Note that we get the same results in classical case.
10
