Quantum Mechanical 


Operators and Wavefunctions

"well behaved" functions (φ), have the
following properties
• must be continuous (no "breaks")
• must have continuous derivatives (no "kinks")
• must be normalizable. Mathematically, this
means (“dτ” implies integration over all space):
∫ ϕ ϕ dτ = C, (C is a finite constant).
∗

if we multiply ϕ by 1 C,
⎛ϕ∗ ⎞⎛ϕ ⎞ ϕ
∫ ⎜⎝ C ⎟⎠ ⎝ C ⎠ dτ = 1, C is said to be "normalized"

Orthogonality of Wavefunctions
✿ Two functions, 𝜑 and χ, are orthogonal if

∫ϕ
∗
χ dτ = 0

✿ Eigenfunctions of QM operators are orthogonal

(proof - handout). If 2 or more eigenfunctions
(eg., 𝜑1 & 𝜑2) have the same eigenvalue, then
orthogonal eigenfunctions can be “made”:
χ1 = ϕ1 and χ 2 = ϕ 2 + cϕ1
where c = − ∫ ϕ1∗ϕ 2 dτ ∫ ϕ1∗ϕ1 dτ .
 Commuting Operators have
Common Eigenfunctions
✿ Thisis shown on your handout
✿ Symmetry operators can’t change the
energy of wavefunctions after they
“operate”. Symmetry operators therefore
commute with the Hamiltonian (H ).
✿ A setof commuting operators including
H and the symmetry operators A, B, . . .
have common eigenfunctions.

Example: The H-Atom (Digression)

Hψ nlm (r,θ ,ϕ ) = Enlmψ nlm (r,θ ,ϕ )
−! 2 2 Ze2
H = T + V (r) = ∇ −
2m r
2⎛ 2
−! ∂ 2 ∂⎞ 1 2 Ze2
= ⎜ + ⎟ + L −
2m ⎝ ∂r 2 r ∂r ⎠ 2mr 2 r
where L2 is the (squared) angular momentum operator:
2 2
⎛ ∂2 ∂ 1 ∂2 ⎞ 2
⎛ ∂2 ∂⎞ L2z
L = −! ⎜ 2 + cot θ + ⎟ = −! ⎜ 2 + cot θ ⎟ + 2
⎝ ∂θ ∂θ sin 2 θ ∂ϕ 2 ⎠ ⎝ ∂θ ∂θ ⎠ sin θ
L2 operates only on θ and ϕ ⇒ commutes with H
!⎛ ∂ ⎞
L z = ⎜ ⎟ operates only on ϕ ⇒ commutes with H and L2
i ⎝ ∂ϕ ⎠
Since H does not operate on spin variables ⇒
S 2 and S z also commute with H , L2 , and L z
 Commuting Operators have
Common Eigenfunctions
✿ For the hydrogen atom, operators that commute
with H are L2, Lz, S 2, and Sz.
✿ Therefore, the eigenfunctions of H are also
eigenfunctions of L2, Lz, S 2, and Sz.
L2: l = 0, 1, ..., n-1
Lz: ml = –l, –l+1, ..., l–1, l (2l+1 values)
S 2: (s = 1/2 for one electron)
Sz: (ms = – 1/2, + 1/2 for one electron, 2S+1 values
in general)

Group Theory & 


Quantum Mechanics
✿ Symmetry Operators commute with the
molecular electronic or vibrational
Hamiltonian:
[ R,H ] = RH − H R = 0 or RH = H R

Hψ i = ε iψ i where {ε i } and {ψ i }
are sets of energies and wavefunctions for
that Hamiltonian
 Group Theory & QM, cont.
✿ Nondegenerate case: only one wavefunction
has energy εi. Then,
H Rψ i = RHψ i
H (Rψ i ) = Rε iψ i = ε i (Rψ i )
Rψi is a wavefunction with the same energy.
Therefore, Rψi = Cψi, where C is a constant
with a magnitude of 1 (usually C = ±1, but it
could be complex, C = eiα, where α is const.)

Example: Valence MOs of Water

✿ H 2O has C2v symmetry.

✿ The symmetry operators of the C2v
group all commute with each other
(each is in its own class).
✿ The eigenfunctions of H must also be
eigenfunctions of the symmetry
operators E, C2, σv1, and σv2.
Building a MO diagram for H2O
O O
H H H H
a1
b2

b1
b2 b1

a1 b2
a1
a1

a1
b2
a1

a1 orbital of H2O
C2
σv2
σv1

Eφ(1a1) (+1)φ(1a1)
C2φ(1a1) (+1)φ(1a1)
σv1 φ(1a1) (+1)φ(1a1)
σv2 φ(1a1) (+1)φ(1a1)
 b1 orbital of H2O

σv2

Eφ(1b1) (+1)φ(1b1)
σv2φ(1b1) (+1)φ(1b1)

b1 orbital of H2O, cont.

C2

σv1

C2φ(1b1) (-1)φ(1b1)
σv1φ(1b1) (-1)φ(1b1)
 b2 orbital of H2O

σv1

Eφ(1b2) (+1)φ(1b2)
σv1φ(1b2) (+1)φ(1b2)

b2 orbital of H2O, cont.

C2
σv2

C2φ(1b2) (-1)φ(1b2)
σv2φ(1b2) (-1)φ(1b2)
 Group Theory & QM, cont.
✿ Degenerate case: more than one wavefunction
has energy εi. We still write
HRψ i = RHψ i
H (Rψ i ) = Rε iψ i = ε i (Rψ i )
Again, Rψi is a wavefunction with the same
energy, but it is possible that Rψi = Σj cj ψj where
the sum is over the all wavefunctions with the
same energy (εi = εj) —for molecules the
degeneracy is usually only 2- or 3-fold.

Higher Dimensional
Representations and Degeneracy
✿ k-fold degenerate wavefunctions form a basis for
a k-dimensional irreducible rep.
✿ Example: For a molecule with octahedral (Oh)
symmetry three p orbitals (px, py, pz) on the
central atom must be degenerate, εx = εy = εz.
Demonstrate that the p orbitals form a basis for
the T1u representation in Oh symmetry.
✿ The converse is true: If k wavefunctions are
degenerate, they form a basis for a k-dimensional
irreducible representation.
The Direct Product - What is it for?
✿ It is often important to know whether
integrals involving bases of IRs are zero.
i.e., if f and g are basis functions belonging
to Γ1 and Γ2, ?
∫ fg dτ = 0
all
space
If the integrand, fg ,belongs to the totally
symmetric IR, then this is nonzero.
Otherwise it is identically zero.

Direct Products:
Is there nothing else?
“The universe is an enormous direct
product of representations of symmetry
groups.” – Steven Weinberg1

1Steven Weinberg, Sheldon Glashow, and Abdus Salam

were awarded the 1979 Nobel Prize in Physics for their
incorporation of the weak and electromagnetic ‘forces’
into a single theory.
 States and Configurations
✿ Each of the electronic states of an atom or a
molecule are usually described as being built up
from one (or perhaps more than one) configuration.
✿ Similarly, the total vibrational state of a molecule
derives from a description of which individual
modes are excited.
✿ Atomic Example: What states arise from a 1s2 2s2
2p2 (C atom ground) configuration?

Read: Harris & Bertolucci, Symmetry and Spectroscopy...,

Chapter 4, pp. 225-45.

2px , 2py ,
2p±1
 Carbon: Atomic Energy Levels
Experimental atomic
energy levels are
shown, with
energies in cm–1.

http://physics.nist.gov/
PhysRefData/ASD/
index.html

Carbon: Atomic Energy Levels

2

3P
1
} gJ µBH
sp3 5S
2
0
-1
-2
3P 27.0 H
1S 1
28,906 3P
1 0
} gJ µBH
33,735 16.4
-1

3P
1D 0
spin-orbit Zeeman
coupling splitting
s2p2 10,163
3P
e–-e– Energy splittings
repulsion
differences displayed graphically;
energies in cm–1
µ B H = 5.6711 × 10−5 eV ~ 0.4574 cm -1 , at H = 1.0T
The Direct Product - What is it for?
✿ We often seek to know the symmetries of
electronic states that arise from
configuration with partially filled orbital
subshells. For example, what states arise
from (t2g)1 (eg)1 configuration for a d2
octahedral complex?
eg

t2g

The Direct Product - What is it?

✿ BasisFunctions: If two sets of

functions {Xj, j = 1, m} and {Yj, j = 1, n}
form bases for two representations,
then the representation spanned by
the m × n product functions, {XiYj} is
called a direct product representation.
 Direct Product Representations
Γ(Ri ) in 1st Γ(Ri ) in 2 nd
⎡ p q ⎤ ⎤
a × ⎡⎢ p q ⎤ b× ⎡
2-D basis 2-D basis
⎡ a b ⎤⊗⎡ p q ⎤=⎢ ⎣ r s ⎥⎦ ⎢⎣ r s ⎥⎦ ⎥
⎢⎣ c d ⎥⎦ ⎢⎣ r s ⎥⎦ ⎢ ⎥
⎢ c × ⎡⎢ p q ⎤ d×⎡ p q ⎤ ⎥
⎢⎣ ⎣ r s ⎥⎦ ⎢⎣ r s ⎥⎦ ⎥⎦
⎡ ap bp bq ⎤
aq
⎢ br bs ⎥ Γ(Ri ) in 4-D basis made up of
= ⎢ ar as
cp dp dq ⎥ product functions from 1st and 2 nd basis
cq
⎢⎣ cr dr ds ⎥⎦
cs
⎣⎡ a b ⎦⎤ ⊗ ⎡⎣ p q ⎤⎦ = ⎡⎣ ap aq bp bq ⎤⎦
⎡ ap ⎤
⎡ a ⎤ ⊗ ⎡ p ⎤ = ⎢ aq ⎥
⎢⎣ b ⎥⎦ ⎢⎣ q ⎥⎦ ⎢ bp ⎥
⎢ ⎥
⎣ bq ⎦

Illustration: S3 Operation in D3h

⎡ ⎤
−1 2 3 2
E ′ represention; x and y basis. S3 matrix : ⎢ ⎥
⎢ − 3 2 −1 2 ⎥
⎣ ⎦
A2′′ represention; z basis. S3 "matrix" : − 1

Direct Product rep.; xz and yz basis.

⎡ ⎤
12 − 3 2
S3 matrix : ⎢ ⎥
⎢ 3 2 12 ⎥
⎣ ⎦
 Illustration: S3 Operation in D3h
⎡ ⎤
⎢ −1 2 − 3 2 ⎥
E ′ rep.; px and p y basis. S3 matrix :
⎢ 3 2 −1 2 ⎥
⎣ ⎦
⎡ ⎤
1 2 3 2
E ′′ rep.; d xz and d yz basis. S3 matrix : ⎢ ⎥
⎢ − 3 2 12 ⎥
⎣ ⎦
Direct Product rep.; px d xz , px d yz , p y d xz , p y d yz , basis.
⎡ ⎤
⎢ −1 4 − 3 4 − 3 4 −3 4 ⎥
⎢ 3 4 −1 4 34 − 3 4 ⎥
S3 matrix : ⎢ ⎥
⎢ 3 4 34 −1 4 − 3 4 ⎥
⎢ ⎥
⎢ −3 4 3 4 3 4 −1 4 ⎥
⎣ ⎦

Properties and Uses of 


Direct Products
✿ The direct product representation ΓAB,
obtained from ΓA and ΓB,will only
contain the totally symmetric
representation if ΓA and ΓB belong to
the same IR.
 Proof
The number of times the ith IR is contained in ΓAB is

ai = 1
h ∑ χ AB ( R)χi (R)
R

χ 1 (R) = 1 for all R ⇒ a1 = 1h ∑ χ AB (R)

R
but we know that χ AB (R) = χ A ( R) χ B (R), so
a1 = 1
h ∑ χ A (R) χB (R) = δ AB
R

where the final equality arises from orthogonality.

SPECTRAL TRANSITION
PROBABILITIES
When are electronic transitions allowed?
For example, is the 2T2g → 2Eg transition
illustrated below “symmetry allowed”?
eg eg

t2g t2g
2T2g 2E g
 TRANSITION ENERGIES

The energetic restriction on the

allowedness of optical transitions like
the 2T2g → 2Eg transition illustrated
earlier is familiar:
E f − Ei = hν
But this doesn’t deal with symmetry.

Transition Probabilities,
Symmetry Aspects
I ∝ transition moment: I ∝ ∫ ψ i µψ j dτ

µ = ∑ qi x i + ∑ qi yi + ∑ qi zi
i i i

µ is the dipole I x ∝ ∫ ψ i x ψ j dτ
moment operator
Iy ∝ ∫ ψ i y ψ j dτ
See Harris & Bertolucci, pp. 130-5.
I z ∝ ∫ ψ i zψ j dτ
 Transition Probabilities,
Symmetry Rules
An electric dipole transition will be
allowed with x, y, or z polarization if the
direct product of the representations of
the two states concerned is or contains
the irreducible representation to which x,
y, or z, respectively, belongs.

Summary
• In the handout, “Transitions Between Stationary States
(Adapted from Harris and Bertolucci, p. 130)”, an expression
for the probability that a system in its ground state, ψ0 can be
stimulated by radiation into an excited state, ψ1, is derived:
Ψ(r,t) = c0 (t)Ψ 0 (r,t) + c1 (t)Ψ1 (r,t)
d ( c1*c1 )
dt
(
∝ E02 µ 01
x 2
+ µ 01
y 2
+ µ 01
z 2
) assuming c0 (0) = 1 and c1 (0) = 0.

E02 is proportional to the light intensity and

⎪⎧ x ⎪⎫
µ 01
x,y,z
=e ∫ ψ 1* ⎨ y ⎬ψ 0 dτ are the transition moment integrals.
⎩⎪ z ⎭⎪

• This expression applies to an integral over all wavelengths of

incident light (in the dipole approximation) and was the form used
to evaluate whether an electron transition was symmetry allowed.
 Allowedness
• In group theory, prescription for deciding whether a
transition is dipole-allowed is straightforward:
• If the ground and excited states respectively belong to the Γ0 and Γ1
representations of the point group of the molecule in question, then
a transition is dipole-allowed if and only if
⎧Γ x for x-polarized light
⎪
Γ 0 ⊗ Γ1 contains ⎨ Γ y for y-polarized light
⎪ Γ z for z-polarized light
⎩

• Spin doesn’t change: ∆S = 0. For states where large spin-orbit

mixing is possible (large ζ and/or small ∆E between mixed states),
the rule is weakened.

Examples
✿ Is the 2T2g → 2Eg transition illustrated
earlier “symmetry allowed”?
✿ What about a 2E → 2T2 transition for a
tetrahedral complex?
t2 t2

e e

2E 2 T2
 Td E 8C3 3C2 6S4 6σ d
A1 1 1 1 1 1 x2 + y2 + z2
A2 1 1 1 –1 –1
E 2 –1 2 0 0 (2z 2 – x 2 − y 2 , x 2 − y 2 )
T1 3 0 –1 1 –1 (Rx , Ry , Rz )
T2 3 0 –1 –1 1 (x, y, z) (xy, xz, yz)

3C2
Oh E 8C3 6C4 6C2 i 8S6 3σ h 6S4 6σ d
(= C42 )
A1g 1 1 1 1 1 1 1 1 1 1 x2 + y2 + z2
A2g 1 1 1 −1 −1 1 1 1 −1 −1
Eg 2 −1 2 0 0 2 −1 2 0 0 (2z 2 – x 2 − y 2 , x 2 − y 2 )
T1g 3 0 −1 1 −1 3 0 −1 1 −1 (Rx , Ry , Rz )
T2g 3 0 −1 −1 1 3 0 −1 −1 1 (xy, xz, yz)
A1u 1 1 1 1 1 −1 −1 −1 −1 −1
A2u 1 1 1 −1 −1 −1 −1 −1 1 1
Eu 2 −1 2 0 0 −2 1 −2 0 0
T1u 3 0 −1 1 −1 −3 0 1 −1 1 (x, y, z)
T2u 3 0 −1 −1 1 −3 0 1 1 −1
 Types of Transitions: d-d
a) intramolecular e– transfer within the d-orbital manifold
b) Values for ε vary: up to 1,000 M–1 cm–1 (moderately
intense) for spin-allowed transitions in non-
centrosymmetric molecules; ~10 M–1cm–1 for spin-
allowed transitions in centrosymmetric molecules (weak);
~10–1 M–1cm–1 for spin-forbidden transitions in
centrosymmetric molecules (very weak).
c) 1-e– transitions much more intense (2-e– transitions
typically 10–2 times weaker)
d) Shoulders and/or broad peaks expected where Jahn-Teller
distortions apply.

Identifying Nonzero Matrix

Elements
Molecular Orbitals, ψµ (µ = 1, 2, …), for molecules are
usually expanded in terms of basis functions: (ψµ =
c1µφ1 + c2µφ2 + …)
These basis functions, {φ1, φ2 , … }, may be atomic
orbitals, or as we shall see, they may be Symmetry
Adapted Linear Combinations (SALCs) of atomic
orbitals. We shall see that it is important to be able
to identify whether integrals (called matrix
elements) of the form below are zero by symmetry.
?
H ij ≡ ∫ φi Hφ j d τ = 0
 Identifying Nonzero Matrix
Elements
SALCs are constructed so that they form
basis functions for the IRs of the system
(molecule or crystal).
Hamiltonian Matrix Elements, Hij,
involving SALCs are nonzero only if φi
and φj belong to the same IR.