PC V: Spectroscopy Solution Exercise 5 FS 2021

Solution Exercise 5

Exercise 1: Molecular oxygen from a group theoretical point of view

a) The symmetry elements and operations of O2 are:

ˆ identity E with operation E

ˆ C∞ axis coinciding with an S∞ axis, which correspond to infinitely many C∞ and S∞
operations
ˆ infinitely many Cn axes with any n, which each correspond to n − 1 symmetry operations
Cnm with m = 1, . . . , n − 1
ˆ infinitely many σv -planes with ∞σv reflection operations
ˆ a σh = S1 plane with reflection operation σh
ˆ an inversion center i with inversion operation i
ˆ and infinitely many C02 axes with ∞C20 rotation operations.

These symmetry elements form the point group D∞h . Note that σh = S1 is implicitly included
in the character table Ref.[1].

Figure 1-1: Symmetry elements of O2 .

b) A vector representation of the p-orbitals of O2 can look like Figure 1-2 (one possibility).

O O

Figure 1-2: Vectors representing p-orbitals for O2 .

We now set up matrices which represent the symmetry operations of the group on our vector
basis {x1 , y1 , z1 , x2 , y2 , z2 } and determine the characters χ of the matrices.

page 1 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

ˆ The matrix representation of E is

 
x1 y1 z1 x2 y2 z2
1 0 0 0 0 0 x1 
 
0 1 0 0 0 0 y1 
 
0 0 1 0 0 0
 z1 
 → χ(E) = 6.
0 0 0 1 0 0 x2 
 
0 0 0 0 1 0 y2 
0 0 0 0 0 1 z2
ˆ C∞ : For these symmetry operations we have to take into account that a rotation of any
angle ϕ around C∞ in one of the two directions leaves the molecule unchanged. The
z vectors are not changed during rotation, resulting in the value one for the diagonal
elements. For the x and y vectors, we can look at a two dimensional rotation ofunit
ϕ − sin ϕ

x
vectors in the xy plane which is performed by the following matrix: cos
sin ϕ cos ϕ y .
Consequently, we end up with the full rotation matrix
 
cos ϕ − sin ϕ 0 0 0 0
 sin ϕ cos ϕ 0 0 0 0 
 
 0 0 1 0 0 0 
  → χ(C∞ ) = 2 + 4 cos ϕ.
 0
 0 0 cos ϕ − sin ϕ 0  
 0 0 0 sin ϕ cos ϕ 0 
0 0 0 0 0 1
ˆ i:
 
0 0 0 −1 0 0
0
 0 0 0 −1 0 
0 0 0 0 0 −1

−1 0
 → χ(i) = 0.
 0 0 0 0 

 0 −1 0 0 0 0
0 0 −1 0 0 0

ˆ σv : Here the question arises which plane to pick to derive the character of the class.
In group theory equivalent symmetry operations are grouped in classes by means of a
similarity transform: Let P̂ , Q̂ be two symmetry operations of a group. Then
R̂ = Q̂−1 P̂ Q̂ (1.1)
is also a member of the group (the equation reads: Apply Q̂, then apply P̂ , then apply
the inverse of Q̂).1 The two operations P̂ and R̂ are called “conjugate” and all conjugate
operations form a class. From (1.1) follows immediately that all conjugate operations
have the same character, as a similarity transformation preserves the trace of a matrix.
Therefore we can pick an arbitrary σv plane to determine the character of the operation.
For simplicity we pick the xz plane:
 
1 0 0 0 0 0
0 −1 0 0 0 0
 
0 0 1 0 0 0
0 0 0 1 0 0 → χ(σv ) = 2.
 
 
0 0 0 0 −1 0
0 0 0 0 0 1

1
Example: For NCl3 , apply σv for any plane, rotate by 2π/3 around the C3 axis, apply σv−1 = σv for the
previously chosen plane; you get rotation by 4π/3 around the C3 , which is thus conjugate to the 2π/3-rotation.

page 2 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

ˆ S∞ : We can look at C∞ and first rotate (same matrix as for C∞ ) and then we mirror at
the σh plane, which exchanges the non-zero and the zero blocks:

 
0 0 0 cos ϕ − sin ϕ 0
 0 0 0 sin ϕ cos ϕ 0
 
 0 0 0 0 0 −1

cos ϕ − sin ϕ 0
 → χ(Ŝ∞ ) = 0.
 0 0 0 

 sin ϕ cos ϕ 0 0 0 0
0 0 −1 0 0 0

ˆ C20 : We consider the C20 axis to be parallel to the y axis for instance, which gives the
following rotation matrix:
 
0 0 0 −1 0 0
0 0 0 0 1 0
 
0 0 0
 0 0 −1  → χ(C20 ) = 0.
−1 0 0 0 0 0 
 
0 1 0 0 0 0
0 0 −1 0 0 0

To summarize we obtain the following representation:

D∞h E 2C∞ ∞σv i 2S∞ ∞C20

Γred 6 2 + 4 cos ϕ 2 0 0 0

c) Infinite point groups cannot be treated with the reduction formula because there is an
infinite number of symmetry elements. One way to solve this problem is to reduce the
representation “by inspection” of the character table of the D∞h point group: One guesses
the right combination of irreducible representations. This is possible because there is only
one way to decompose the reducible representation. The reducible representation of the p
orbitals Γred can be reduced as follows: Γred = Πg ⊕ Πu ⊕ Σ+ +
g ⊕ Σu . Note that the Πg and Πu
irreps are of dimension two, so we obtain the required total dimension of 6. Mathematical
methods to treat infinite point groups can be found for example in Refs. [2, 3]. When we
compare these results to Fig. 1-1 of the solution to Exercise 4 we find that the irreducible
representations we have found match the labels given to the orbitals.

d) For comparison we use the orbital diagram of Exercise 4. The partially occupied molecu-
lar orbitals have πg symmetry. To find the term symbols, we multiply the corresponding
characters of this irrep under the D∞h point group:

D∞h E 2C∞ ∞σv i 2S∞ ∞C20

Γπg ⊗πg 4 4cos2 (φ) 0 4 2
4cos (φ) 0

Recognizing that cos2 (φ) = 21 [1 − cos(2φ)], the Γπg ⊗πg representation can be reduced to
−
Γπg ⊗πg = ∆g + Σ+ g + Σg . Alternatively, this could have been deducted from the direct
−
product table, which gives the following result: Πg ⊗ Πg = Σ+ g ⊕ [Σg ] ⊕ ∆g . The square
brackets indicate that this direct product has an antisymmetric spatial part and thus a
3 − 1
symmetric (triplet) spin part. This yields the following molecular terms: 1 Σ+
g , Σg , ∆g .

page 3 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

e) The irreducible representations of the dipole operator are the same as those of the linear
functions. From the character table we read that those are Πu and Σ+ u . To determine all
possible final states we must thus take the product of the ground state irreducible represen-
tation Σ− + − − + −
g with Πu and Σu , which gives Σg ⊗Πu = Πu and Σg ⊗Σu = Σu . From the product
table, we see that the totally symmetric representation is only obtained if the irreducible
representation is multiplied with itself. Consequently, only states that can be excited from
the ground state are 3 Σ− 3
u and Πu states; only those result in the totally symmetric represen-
tation (and do not change the spin). The likelihood of these transitions can be obtained from
electronic structure calculations. Group theory only allows us to determine which integrals
are guaranteed to be zero due to symmetry.

Exercise 2: Vibrations of BrF5

a) The symmetry elements of BrF5 (Figure 2-1) are E, C4 , C2 , 2σv , 2σd .

Figure 2-1: Symmetry elements of BrF5 .

b) BrF5 belongs to the C4v point group.

Additional information: There is another notation, which also specifies the symmetry of
molecules. The so called Hermann-Mauguin symbols give only the generating elements from
which all other symmetry elements can be derived. In this case it is 4mm. 4 represents the
4-fold axis and mm represents two mirror planes (σv and σd ).
c) The molecule has 3N degrees of freedom where N is the number of nuclei. Among these 18
degrees of freedom, three are rotations and three are translations leaving twelve vibrational
degrees of freedom.
d) Two equivalent approaches can be taken to derive Γtot :

ˆ To derive the matrix representation of all displacement coordinates we define three dis-
placement vectors on each nucleus, as depicted in Figure 2-2. For a given symmetry
operation, a vector →−
v only contributes to the character of the matrix if it is either un-
changed (+1) or if it is inverted −→
−
v (−1).
In the following, a detailed explanation is given for a few symmetry operations. For the
remaining ones the reasoning is similar and can be adapted from problem 1:

page 4 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

Figure 2-2: Displacement vectors for BrF5 . The z axis is along the 4-fold rotation axis. The y
and x axis were chosen to coincide with the intramolecular bonds.

– E: The identity operation leaves all vectors unchanged: χ(E) = 18.

– C4 : The vectors of the F nuclei all change their places and, therefore, do not contribute
to the character. At the central Br nucleus and at the fluorine nucleus located on the
principal axis, only the z vector is left in place and keeps its direction. This yields:
χ(C4 ) = 2 and χ(C2 ) = −2.
– σv : At each of the Br nucleus and the three fluorine nuclei in the reflection plane,
the two vectors in the plane stay the same but the vector perpendicular to the plane
changes sign. We thus have 2 − 1 = 1 for each of the four nuclei, hence χ(σv ) = 4.
– σd : F nuclei number 2-5 changes their places and do not contribute to the value of
character. At the Br atom and at the F atom (number 6), the x and y vectors are
exchanges, and, therefore the z vectors contribute with 1: χ(σd ) = 2.
In the end, one obtains for the displacement vectors of all nuclei the following reducible
representation:

C4v E 2 C4 C2 2 σv 2 σd
Γtot 18 2 −2 4 2

ˆ Alternatively, we can construct the total representation as

Γtot = Γx,y,z ⊗ Γnuclei (2.1)

where Γx,y,z is the sum of all different irreducible representations of the unit vectors.
Γnuclei is the reducible representation in the basis of atomic cores {r1 , r2 , r3 , r4 , r5 }. The
characters of the symmetry elements in this representation are given by the number of
unmoved nuclei.
We obtain the reducible representation (Γtot )for the displacement vectors of all nuclei
given in table 2.1.

C4v E 2 C4 C2 2 σv 2 σd
Γx,y,z (A1 +E) 3 1 −1 1 1
Γnuclei 6 2 2 4 2
Γtot 18 2 −2 4 2

Table 2.1

page 5 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

e) Translations transform like the vectors x, y, z, giving the irreducible representations Γtrans =
A1 ⊕ E. Rotations transform like Rx , Ry , Rz , giving the irreducible representations Γrot =
A2 ⊕ E (as can be seen from the point group table). Note that E has the dimension two,
giving three degrees of freedom for the rotation and the translation.
f) Subtraction of the translational and rotational irreducible representations gives:

C4v E 2 C4 C2 2 σv 2 σd
Γtot 18 2 −2 4 2
−Γtrans −3 −1 1 −1 −1
−Γrot −3 −1 1 1 1
Γvib 12 0 0 4 2

i.e. twelve degrees of freedom for the vibrations.

Application of the reduction formula to Γvib gives the decomposition into irreducible repre-
sentations:
1
n(A1 ) = (12 + 0 − 0 + 8 + 4) = 3
8
1
n(A2 ) = (12 + 0 − 0 − 8 − 4) = 0
8
1
n(B1 ) = (12 − 0 − 0 + 8 − 4) = 2
8
1
n(B2 ) = (12 − 0 − 0 − 8 + 4) = 1
8
1
n(E) = (12 × 2 + 0 + 0 + 0 + 0) = 3
8
Thus, the irreducible representations for the vibrational degrees of freedom are:
Γvib =3 A1 ⊕ 2 B1 ⊕ B2 ⊕ 3 E.

g) Infrared spectra result from the absorption of light by a molecule, while Raman spectroscopy
is based on inelastic scattering. Let us assume that infrared light, which excites the molecule,
is polarized along the x-axis of the coordinate system in which the position of the molecule
is defined. Since the absorptionR intensity is proportional to the matrix element of the dipole
moment component µ̂x (I ∝ ψ0 µ̂x ψ1 d~r), the selection rules establish non-zero integrals
for symmetry reasons. At room temperature, most molecules are present in their ground
vibrational state (ν0 ), which has a totally symmetric vibrational wavefunction. Hence, as ψ0
is totally symmetric, ψ1 has to belong to the same representation as µ̂x for the integral to
be non-zero and therefore Rthe same representation as x. For a mode to be Raman active,
the corresponding integral ψ0 α̂qq0 ψ1 d~r must be non-zero. Here, α̂qq0 (q, q0=x, y, z) is a com-
ponent of the polarizability tensor (qq0=x2 , y 2 , z 2 , xy, xz, yz and their linear combinations).
One can easily recognize absorption and Raman active modes by regarding the character
table:
ˆ absorption active modes have the same symmetry as translational vectors, i.e. in our case
{A1 , E}.
ˆ Raman-active modes have the same symmetry as the quadratic functions, which for the
C4v point group are {A1 , B1 , B2 , E}.
Thus, in the vibrational spectrum of the BrF5 molecule, one can expect six absorption active
modes, three of which will have a two-fold degeneracy and nine Raman-active modes, six of
which will coincide with absorption peaks.

page 6 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

Exercise 3: Hückel MO treatment of cyclobutadiene

a) The principal axis of a square-shaped cyclobutadiene is a C4 rotation axis. It has a C2 axis
along the principal axis, as well as two C20 axes, two C200 axes, one σh mirror plane, two S4
axes, two σv mirror planes, two σd mirror planes, a center of inversion i, and the identity E,
giving the point group D4h . The symmetry elements are sketched in Figure 3-1.

✝✂ ✁

✁✥✂ ✄✆ ✁✥✂ ✄☎

Figure 3-1: Some of the symmetry elements of the cyclobutadiene molecule.

b) The orbitals contributing to the π-system are all carbon pz orbitals as it is depicted in Figure
3-2 in a schematic way.

✆✟✝ ✆✠✝
✂ ✁ ✄✟ ✂ ✁ ✄✠

✆☎✝ ✆✞✝
✂ ✁ ✄☎ ✂ ✁ ✄✞

Figure 3-2: Carbon p orbitals of the cyclobutadiene viewed from above the molecular plane.

c) Operating with the symmetry elements of the group on the pz orbitals as sketched in Figure
3-1 gives the following characters for the reducible representation:

D4h E 2 C4 C2 2 C20 2 C200 i 2 S4 σh 2 σv 2 σd

Γtot 4 0 0 −2 0 0 0 −4 2 0

page 7 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

Application of the reduction formula yields:

1
nA1g = (4 − 2 · 2 − 4 + 2 · 2) = 0
16
1
nA2g = (4 + 2 · 2 − 4 − 2 · 2) = 0
16
1
nB1g = (4 − 2 · 2 − 4 + 2 · 2) = 0
16
1
nB2g = (4 + 2 · 2 − 4 − 2 · 2) = 0
16
1
nEg = (2 · 4 + 0 − 2 · (−4) + 0) = 1
16
1
nA1u = (4 − 2 · 2 + 4 − 2 · 2) = 0
16
1
nA2u = (4 + 2 · 2 + 4 + 2 · 2) = 1
16
1
nB1u = (4 − 2 · 2 + 4 − 2 · 2) = 0
16
1
nB2u = (4 + 2 · 2 + 4 + 2 · 2) = 1
16
1
nEu = (2 · 4 + 0 + 2 · (−4) + 0) = 0
16
The four pz orbitals can be decomposed into the representations Γ = Eg ⊕ A2u ⊕ B2u .
d) The projection operators for the three irreducible representations are:
2
X 2
X 2
X 2
X
P A2u
= (E + C4 + C43 + C2 − C20i − C200i − i − S4 − S43 − σh + σvi + σdi ),
i=1 i=1 i=1 i=1
Eg
P = (2 E − 2 C2 + 2 i − 2 σh ),
2
X 2
X 2
X 2
X
P B2u = (E − C4 − C43 + C2 − C20i + C200i − i + S4 + S43 − σh + σvi − σdi )
i=1 i=1 i=1 i=1

With these operators the new set of symmetry-adapted basis functions can be generated.
Prefactors from the projection operators are combined in the normalization factor 1/16 for
non-degenerate states and 2/16 for two-fold degenerate states:
1
ψA2u = P A2u φ1 = (4 φ1 + 4 φ2 + 4 φ3 + 4 φ4 )
16
2
ψEg ,1 = P Eg φ1 = (4 φ1 − 4 φ3 )
16
1
ψB2u = P B2u φ1 = (2 φ1 − 2 φ2 + 2 φ3 − 2 φ4 )
16
Since the irreducible representation Eg is two-fold degenerate, two orthogonal basis functions
exist. There are several possibilities to derive the second basis function. Here we use the
fact that each symmetry operation on a basis function for a doubly-degenerate representation
needs to give ± itself or the other member or a linear combination of both.
Therefore we apply a C4 rotation to ψEg ,1 (neglecting any normalization for the moment):
1
ψEg ,2 ≡ C4 ψEg ,1 = (φ2 − φ4 )
2

page 8 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

Thus, after normalization one obtains the following four orthonormal and symmetry adapted
basis functions:

1
ψ1 = ψA2u = (φ1 + φ2 + φ3 + φ4 )
2
1
ψ2 = ψB2u = (φ1 − φ2 + φ3 − φ4 )
2
1
ψ3 = ψEg ,1 = √ (φ1 − φ3 )
2
1
ψ4 = ψEg ,2 = √ (φ2 − φ4 )
2

e) With the new, symmetry-adapted basis functions, R ∗ the Hückel Hamiltonian matrix can be
0 eff
constructed by evaluating the integrals Hrs = ψr Ĥ ψs d~r .
R
Because all off-diagonal elements φi H eff φj d~r (i 6= j) of the original Hückel Hamiltonian
matrix H are equalR to zero unless i and j are neighboring (in which case Hijeff = β) and its
diagonal elements φi H eff φj d~r = α (for i = j), one gets for the diagonal elements of the
new Hückel Hamiltonian matrix:
Z Z
0 ∗ 1
eff
H11 = ψ1 Ĥ ψ1 d~r = (φ∗1 + φ∗2 + φ∗3 + φ∗4 )Ĥ eff (φ1 + φ2 + φ3 + φ4 )d~r
4
1
= (α + β + β + β + α + β + β + α + β + β + β + α) = α + 2 β
Z 4Z
0 1
H22 = ψ2∗ Ĥ eff ψ2 d~r = (φ∗1 − φ∗2 + φ∗3 − φ∗4 )Ĥ eff (φ1 − φ2 + φ3 − φ4 )d~r = α − 2 β
4
Z Z
0 ∗ 1
eff
H33 = ψ3 Ĥ ψ3 d~r = (φ∗1 − φ∗3 )Ĥ eff (φ1 − φ3 )d~r = α
2
Z Z
0 ∗ 1
eff
H44 = ψ4 Ĥ ψ4 d~r = (φ∗2 − φ∗4 )Ĥ eff (φ2 − φ4 )d~r = α
2

In a similar manner one obtains for the off-diagonal elements:2

0 0 0 0 0 0
H12 = H13 = H14 = H23 = H24 = H34 =0

After evaluating all matrix elements, the Hückel Hamiltonian matrix in the new basis set
can be set up:
 
α + 2β 0 0 0
0 α − 2β 0 0 
H0 = 


 0 0 α 0 
0 0 0 α

Since the basis set is still orthonormal, the overlap matrix S 0 remains the identity matrix.
As opposed to the original Hückel Hamiltonian matrix, the Hückel matrix in the new basis is
diagonal. In general, the symmetry-adapted Hückel-determinant will have a block-diagonal
form, each n × n block belonging to the n basis functions displaying the same symmetry.
0 0
In the present case, H11 and H22 belong to the non-degenerate irreducible representations
A2u and B2u , respectively. The remaining diagonal block belongs to the two-fold degenerate
representation Eg .
2
Note that this is only true for the particular functions that we have chosen.

page 9 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

f) The Hückel determinant is given by

α + 2β −  0 0 0
0 α − 2β −  0 0
det(H 0 −  S 0 ) =
0 0 α− 0
0 0 0 α−

A block diagonal structure simplifies the calculation of the energy levels. In fact, for each
subblock the problem can be solved separately, since
 
A 0
det = det(A) · det(B)
0 B

where A and B are symmetric matrices of any size.

In our case the Hückel determinant is diagonal. Due to this fact, one can straightforwardly
obtain the energy values:

1 = α + 2 β
2 = α − 2 β
3 = α (two-fold degenerate)

Note that the Eg block is diagonal, even though this is not required by symmetry. Thus,
diagonal elements of the block are the eigenvalues themselves. This is not generally true and
eigenvalues have to be calculated by diagonalization.
The parameter β describes the interaction energy due to the chemical bond and is negative.
Therefore the lowest lying orbital is a2u orbital followed by the degenerate eg orbitals. The
b2u is the antibonding orbital. Figure 3-3 shows the energy level diagram. According to
standard notation, molecular orbitals are labelled with the lower-case Mulliken symbols of
their irreducible representations.

✁✂

✄☎

✆✁✂
Figure 3-3: The energy levels for the cyclobutadiene from a HMO treatment.

g) In the ground state, the energetically lowest-lying orbital is completely filled, leading to
the totally symmetric irreducible representation A1g . The doubly degenerate state of eg
symmetry is occupied by two electrons. Therefore:

Γ1 ⊗ Γ2 ≡ eg ⊗ eg = a1g ⊕ a2g ⊕ b1g ⊕ b2g

Thus, these two electrons give rise to the states: A1g , A2g , B1g and B2g . The states could be
singlet or triplet. For the overall wavefunctions of the singlet states to be antisymmetric, the

page 10 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021

spatial part of the wavefunction needs to be symmetric (as the spin wavefunction is antisym-
metric in this case). For the triplet state, the spatial wavefunction has to be antisymmetric.
The only antisymmetric representation for the above direct product is A2g . Therefore, the
term symbol for the triplet state is 3 A2g . The possible term symbols for the singlet states
are 1 A1g , 1 B1g and 1 B2g . In accordance with Hund’s rule the ground state (state with the
lowest energy) is that with the highest multiplicity, namely 3 A2g .
h) A rectangular cyclobutadiene molecule belongs to the D2h symmetry group. Because of that,
the molecule does not have a C20 axis or a symmetry plane σv anymore.
D2h E C2 (x) C2 (y) C2 (z) i σxy σxz σyz
Γtot 4 0 0 0 0 −4 0 0

For the four pz orbitals, localized on the carbon nuclei, the representation is given by:
Γ = B2g ⊕ B3g ⊕ Au ⊕ B1u . Applying the projection operator on the φ1 orbital, one obtains
the following normalized basis functions:
1
ψ1 = ψB1u = (φ1 + φ2 + φ3 + φ4 )
2
1
ψ2 = ψAu = (φ1 − φ2 + φ3 − φ4 )
2
1
ψ3 = ψB2g = (φ1 + φ2 − φ3 − φ4 )
2
1
ψ4 = ψB3g = (φ1 − φ2 − φ3 + φ4 )
2
The functions are schematically depicted in Figure 3-4 together with their nodal planes. As
one can see, the basis functions look similar to those obtained for the square structure of
cyclobutadiene. Decreasing the symmetry leads to splitting of the doubly degenerate energy
level of eg symmetry into two non-degenerate levels of b2g and b3g symmetry.

✡ ☛ ☎ ☎ ☎ ☎
✎ ✍

✠ ☞ ☎ ☎
✌
✁✂✄ ✁✆✝ ✁✞✝ ✟✄

Figure 3-4: Molecular orbitals of cyclobutadiene in rectangular structure.

The interaction between neighbouring atoms, that we called β before, now differ depending
on whether the neighbours are along the elongated or shortened bonds. This means that
by the symmetry arguments alone, it is not possible to say what the energetic ordering of
the b2g and b3g states will be. However, chemical reasoning tells us that the b2g state will
be lower in energy because here the two pz orbitals that have a binding overlap are along
the double bond, where the distance between them is shorter than for b3g state. The most
stable configuration is (b1u )2 (b2g )2 , corresponding to two doubly occupied orbitals. The
ground state term symbol for this configuration is 1 Ag .
For the excited state configuration (b1u )2 (b2g )1 (b3g )1 , the direct product of the irreducible
representations of the two singly occupied orbitals is: B2g ⊗ B3g = B1g . The resulting
molecular term symbols are: 1 B1g , 3 B1g . Note that here both terms exist with singlet
and triplet spin multiplicity because there is no Pauli restriction due to the two different
spatial orbitals. However, one expects that the triplet state is slightly lower in energy due
to exchange interaction.

page 11 of 12
PC V: Spectroscopy Solution Exercise 5 FS 2021
✄✂ ✄✂ ✄✂

✝✆ ✝✆ ✝✆

☎✆ ☎✆ ☎✆

✁✂ ✁✂ ✁✂
Figure 3-5: Left panel: most stable configuration of the rectangular cyclobutadiene. Middle
panel: excited triplet configuration of the rectangular cyclobutadiene. Right panel: excited
singlet configuration of the rectangular cyclobutadiene.

References
[1] http://www.webqc.org

[2] J. A. Greathouse, Group Theory Calculations Involving Linear Molecules, Chem. Educator
1, 1 (1996).

[3] J. M. Alvariño, A. Chamorro, Continuous Point Groups, J. Chem. Edu. 57, 785 (1980).

page 12 of 12