CY 413: Spectroscopic Methods of Analysis

Class notes: ©Supratim Giri

Chapter 2: Microwave spectroscopy

Rotational motions of molecules:
I hope that after going through the introductory lectures of this course, at least we are aware of the
fact that molecular spectroscopy in the microwave region deals with the rotational motion of the
molecules.
In this chapter we will learn about the details of the energy spacing between ground and excited
states of a molecule due to rotation. Also, we will see whether all jumps from a lower lying energy
state to higher lying ones are possible. We would appreciate the relative intensity of the spectral
lines as well.

So, do you still remember Newton’s first law of motion?

Unless compelled by any external force, every object continues in its state of rest or of uniform
motion in a straight line.

It’s needless to say that the above law provides concept of “inertia", which may refer to an object's
"amount of resistance to change in velocity". Naturally, we would talk about ‘rotational inertia’ of a
three dimensional object if that object is in the state of rotational motion.

The term rotational inertia is also commonly known as moment of inertia or in other words, it is a
measure of an object's resistance to any change in its state of rotation. The moment of inertia is
usually denoted with the capital letter I:

........................................... Equation 1

The object consists of N point masses mi whose distances to the axis of rotation are denoted as ri

For molecular rotational motion, this I becomes a characteristic deciding factor that determines the
amount of energy gaps between ground and excited states and hence the peak position in
microwave spectra of molecules. The last sentence is basically the essence of microwave
spectroscopy. Is it clear?

Now we will see how I is related to the different energy states of a molecule due to rotation.

Molecules are 3D object and rotation of such object may be quiet complex. To simplify the problem
of rotation, it is convenient to resolve the component of rotational motion into 3 mutually
perpendicular axes (just like x, y and z axes found in geometry text books).

1
So, the molecule (a 3D object) has 3 components of moment of inertia: IA, IB and IC, all of them are
mutually perpendicular (and lying along x or y or z axis as shown in first page figure) and all of them
are passing through the object’s centre of gravity. These IA, IB and IC are known as ‘principal
moments of inertia’.

Now, we will see how these IA, IB and IC behave with respect to the different molecular geometry, i.e.
shape.

Case 1): Linear molecule

H---Cl O=C=O H--CΞC—H

These molecules may rotate in 3 ways, a) around its own bond axis, b) end-over-end rotation on the
plane of the paper and c) end-over-end rotation perpendicular to the plane of the paper.

In situation a), IA = 0 because, this type of rotation does not lead to any positional change of the
atoms or the bonds.

In both situation b) and c) we have IB = IC and we assume that IA = 0

Case 2): Spherical top molecule:

IA = IB = IC ≠ 0

Case 3): Symmetric top molecule:

IB = IC≠ IA and IA ≠ 0 end-over-end rotation on the plane and perpendicular to the plane of the paper
are identical. Another example is given below.

Case 4): Asymmetric top molecule:

IA ≠ IB ≠ IC All the three moment of inertia are different.

2
Rotational spectra of a diatomic molecule in rigid rotor model:

As we have a brief idea about moment of inertia of a rotating molecule, let’s start to arrive at the
expression of energy levels obtained from the rotation of a simple diatomic molecule, e.g. H2.

Does the molecule have any permanent dipole moment? Do you think that rotating H2 molecule
would interact with microwave radiation and give a molecular spectrum at microwave region?

What about other diatomic molecules O2, N2, F2, Cl2 etc?

Let’s move on to the next simple diatomic molecule, e.g. HF

Let’s assume that the molecule is a rigid rotor, i.e. it does not change its shape and hence centre of
gravity due to rotation. Let C be the centre of gravity, and the molecule rotates end-over-end
centring point C. So from the picture above following equations are true:

ro = r 1 + r2 ...............................eq. 2

m1r1 = m2r2 .............................eq. 3

The moment of inertia around point C is defined by:

I = m1r12 + m2r22 .............................eq. 5

Point to be noted: We can have knowledge about ro (bond length) m1 (mass of F atom), and m2 (mass
of H atom) but not about r1 and r2. So our target will be to obtain an expression for moment of
inertia where r1 and r2 doesn’t appear!

So, following algebraic exercise (solving two unknown r1 and r2 using eq. 2 and 3):

r1 = m2ro/(m1+m2) and r2 = m1ro/(m1+m2)

Putting the values of r1 and r2 in eq. 5, we get

I = m1m2ro2/(m1+m2) = μ ro2 (μ is called the reduced mass of the system)

Since HF is linear molecule, IA = 0 and IB = IC = μ ro2

3
According to the quantum theory in physics, due to rotation, molecules have several discrete energy
levels. That means instead of having continuous energy states, molecule will have discrete quantized
energy states as shown in the qualitative diagram below.

With the help of Schrodinger equation, the expression of any such quantized energy level εJ can be
given as:

εJ = h2J(J+1)/8π2I joules where J = 0, 1, 2, 3 ..... etc [h: Planck’s constant, 6.63 x10-34 JS molecule-1]

εJ = hJ(J+1)/8π2Ic cm-1 where J = 0, 1, 2, 3 ..... etc [c: velocity of light] ........................eq. 6

Alternatively, eq. 6 can be expressed as:

εJ = BJ(J+1) cm-1 where J = 0, 1, 2, 3 and B = h/8π2Ic cm-1 & I = IB = IC = μro2 .................eq. 7

B is known as rotational constant and contains information about I and hence bond length and
reduced mass of a molecule! Letter c stands for velocity of light in above equation.

So what are the allowed values of the individual energy levels (εJ) from eq. 7 when we put the values
of J as 0, 1, 2, 3, etc?

Value of J Value of εJ (cm-1)

0 0
1 2B
2 6B
3 12B
4 20B
5 30B
-- --
etc so on

So from the above consideration, we see that at J = 0, molecule will not have any rotational energy.
That implies, molecule is not rotating at all. The least possible energy allowed is 2B.

Now if we arrange the allowed values of εJ (cm-1) in a relative energy scale having units of B cm-1, the
energy level diagram will look like below:

4
Now, let’s consider the jumping of HF molecule in between these allowed quantized energy states.
Unfortunately, any molecule cannot jump arbitrarily from a certain state to any other state as long
as rotational energy levels are concerned. So there is some restriction, which is known as selection
rule derived from quantum mechanical calculation and the rule says any jump or transition
corresponding to change of J in ± 1 value is allowed!

Selection rule for rotational spectroscopy: Allowed ΔJ = ± 1 (any other ΔJ is forbidden)

So, from ground state ε0 HF molecule is only allowed to jump to excited state ε1 but not to state ε2.
To reach state ε2, it has to either jump upward from ε1 or jump down from ε3. Clear?

What is the expression for change in energy if the molecule jumps from any arbitrary level J = J to J =
(J+1)?

.................eq. 8

Now, putting J = 0, 1, 2, 3, 4, ... etc in the above equation, we get the possible ΔE values as 2B, 4B,
6B, 8B, 10B ... so on as evident from top figure in this page. So if you are to draw a rotational
spectrum corresponding to the above figure how the spectrum would look like?

5
The last figure in previous page shows a qualitative diagram of expected microwave spectra for any
diatomic molecule. Although the spectrum seen in reality is not quite like the one you just saw.

Above spectrum is a rotational spectrum of CO at 40K (very low temperature), recorded in real life.

So how do this figure compare with the figure in the previous page?

Similarity:
1) All peaks are separated by a gap of (almost) 2B cm-1
Difference:
1) Peaks are broader in the second figure (instead of sharp lines)
2) All peaks are of not equal height! Peaks at the beginning and at the end are shorter (smaller
in intensity) and middle peaks are taller (higher in intensity).

Being an inquisitive student, you would definitely ask about the reason for these differences. Well,
the first aspect (peak broadening aspect) will be addressed later (as mentioned before).
Here we will discuss the aspect of difference in peak heights in a rotational spectrum.

6
Intensity of rotational spectral lines:
If you look at the CO spectrum carefully, you will see that peaks originating from early transitions i.e.
jumps involving J=0 to J=1 and J=1 to J=2 are shorter in height. Same is the case for late transitions
i.e. jumps involving J=8 to J=9 and J=9 to J=10 are also shorter in height. The middle peaks (J=3 to J=4
and J=4 to J=5) the longest.
If you remember, in Chapter 1 of this course, we appreciated that ‘spectral peak’ signifies the
probability of a transition. That means longer peak corresponds to more probable transition as
opposed to least probable one in the case of shorter peaks.
Do you agree?
So, J=3 to J=4 and J=4 to J=5 transition in this case is most probable. Question is why?
Let’s address the above question in the following way:
1) It has been seen that probability of a transition from a lower energy level to a higher energy
level is proportional to the population of the lower level.
2) That means: highest peak of CO rotational spectrum involves J=4 to J=5 transition. Therefore
the rotational quantum level J = 4 is most populated. In other words, maximum number of
CO molecules are occupying J = 4 level (due to rotational motion), making that level most
populated. Naturally, when CO gets excited by microwave, maximum number of molecules
jump from J=4 to J=5 giving rise to the longest peak. Make sense?

Still there’s the question why mid levels (i.e. J=3, J=4, etc) are mostly populated rather than lower
levels such as J=0 or 1?

To find the answer we have to dig deeper into Physics. There are two opposing factors that govern
the population of rotational quantum levels. These are i) Boltzmann distribution factor and ii)
degeneracy factor.
i) We know that rotational energy in the lowest level is zero, since J = 0. So, if we have N0
molecules in this state, the number of molecules in any higher state is given by:
NJ/N0 = exp(-EJ/kT) [where k is Boltzmann constant] .................eq. 9
Now you can see higher the value of J, higher will be EJ and lower would be the value of
NJ. That means according to the Boltzmann distribution, ground level (J=0) would be the
most populated state.

ii) However, according to the quantum mechanical consideration, each J quantum level has
degeneracy (much like in case for l value 1, p orbital of atom is triply degenerate px, py
and pz with total degeneracy = 2l +1) and each quantum level is 2J+1 times degenerate.
According to the quantum physics, the more a level is degenerate the more population
will be in that level. So this effect will lead the higher J level to be most populated and
lower J levels to be least popular.

The above two effects work at the same time in a molecule during rotation. Thus we can write total
relative population is proportional to (2J+1)exp(-EJ/kT).
Therefore, a compromise happens at the mid level. This leads the mid J levels (not the lower or
higher one) to become maximum populated. For CO molecule, this scenario happens at around J = 4
level. For other molecules it can be at any number closer to 4 or 5 depending on the molecular
nature. So is it clear now why rotational spectral intensity pattern appear as ‘temple’ shape?

7
From the rotational spectrum of CO, we can identify the 2B value. We get 2B = 3.842 cm-1. What is
the bond length of CO, if we consider mass of C and O atoms are 12.00 and 15.9994 a.m.u.,
respectively?

Try it out to see whether you get the bond length as 0.113 x 10-9 m.

Review of rigid rotor model:

Below is the data of HF actually obtained from rotational spectroscopy

Transition ΔE (cm-1) B (cm-1) calculated

observed ro (nm)
J=0 to J=1 41.11 20.56 0.0929
J=1 to J=2 40.96 20.48 0.0931
J=2 to J=3 40.85 20.43 0.0932
J=3 to J=4 40.62 20.31 0.0935
J=4 to J=5 40.31 20.16 0.0938
J=5 to J=6 40.08 20.04 0.0941
J=6 to J=7 39.64 19.82 0.0946
J=7 to J=8 39.28 19.64 0.0951
J=8 to J=9 38.89 19.45 0.0955
J=9 to J=10 38.31 19.16 0.0963
J=10 to J=11 37.81 18.91 0.0969

From the above table it is evident that ΔE value or 2B value is not constant in reality. Separation
between the successive lines decreases steadily with increasing J value. As a consequence, the bond
length also increases. So we can see, during rotation HF molecule does not behave as a rigid rotor.
Rather the bold behaves like an elastic fixture between the atoms and the bond increases as the
centrifugal force due to rotation increases.

So our initial assumption of rigid rotor was just an approximation. For a more accurate approach, a
non-rigid rotor model is adopted to explain the observed rotational spectra of HF.

Expression of energy level for non-rigid rotor model:

εJ = BJ(J+1) – DJ2(J+1)2 cm-1 .................eq. 10

D is known as the centrifugal distortion constant and the above equation is valid only for a simple
harmonic force field.

We will consider the concept of rigid rotor model to apply and evaluate our knowledge in rotational
spectroscopy. Non-rigid rotor model is bit complicated with respect to the scope of this course.

8
Important point to be noted:

i) Take a closer look at the rotational spectra of CO. You will see the late spectral lines are
lying in the infrared region. Same is the case that happens for HF also. Bottom line is, the
lower rotational levels of diatomic molecules lie in the microwave region but higher
levels may enter in the IR region. So people may use an IR excitation to record the late
transition (e.g. jump from J=9 to J=10 for CO) and still can figure out the bond length of
CO assuming rigid rotor model.

ii) Also, molecular vibration energy levels lie in the IR region. That implies in IR
spectroscopy, both vibration and rotation couple together and may appear in the
spectrum! So microwave and IR spectroscopy are actually inseparable with each other.
We will learn about the details in the next chapter.

iii) The rotational spectra that you have come across so far, all originated due to the
absorption of microwave (primarily) radiation by the molecules in a particular J level and
then jumping of the molecule to the next J level. That’s why these types of spectra are
also known as ‘absorption’ rotational spectra. The molecules after jumping up to the
higher level may come down to the ground level but we are not concerned about the
form of energy released in that process. Because, the energy in that case is mostly
emitted as non-electromagnetic manner.

Polyatomic molecule:

In the light of rotational spectroscopy, polyatomic molecules are normally divided into three
categories. A) Linear B) Symmetric top and C) Asymmetric top

Existence of multiple moments of inertia as well as the application of non-rigid rotor model makes
the spectra of polyatomic molecule quite complicated. We don’t have the opportunity to study them
in this course.

However, the take home message is no matter how complicated the spectra of polyatomic
molecules are, the basic target is to find out the B value from the separation of spectral lines and
hence the bond lengths. This is how structures of many molecules were elucidated historically.

A question:

Is microwave spectroscopy anyway related with the working principle of a microwave oven?