Infrared Spectroscopy

Introduction to Infrared Spectroscopy.

When a beam of electromagnetic radiation of intensity Io is passed through a substance,

it can be either absorbed or transmitted, depending upon its frequency, n, and the
structure of the molecule it encounters. Electromagnetic radiation is energy; thus when a
molecule absorbs radiation it gains energy as it undergoes a quantum transition from
one energy state (Einitial) to another (Efinal). The frequency of the absorbed radiation is
related to the energy of the transition by Planck's law:

Efinal - Einitial = DE = hn = hc/l.

If a transition exists that is related to the frequency of the incident radiation by Planck's
constant, the radiation can be absorbed. If the frequency does not satisfy the Planck
expression, then the radiation will be transmitted. A plot of the frequency of the incident
radiation vs. some measure of the percent radiation absorbed by the sample is the
absorption spectrum of the compound.

The type of absorption spectroscopy depends upon the frequency range of the
electromagnetic radiation absorbed. Microwave spectroscopy involves a transition from
one molecular rotational energy level to another. Rotational energy level spacings
correspond to radiation from the microwave portion of the electromagnetic spectrum.
Vibrational spectroscopy (or infrared spectroscopy) measures transitions from one
molecular vibrational energy level to another, and requires radiation from the infrared
portion of the electromagnetic spectrum. Ultraviolet-visible spectroscopy (also called
electronic absorption spectroscopy) involves transitions among electron energy levels in
the molecule, which require radiation from the UV-visible portion of the electromagnetic
spectrum. Such transitions alter the configuration of the valence electrons in the
molecule.

Molecules may undergo several types of motion. First, the entire molecule may move
through space in some direction and with a particular velocity. This is called translational
motion and with it we associate the translational kinetic energy of the molecule, 1/2mv2
(v = velocity of the center of mass of the molecule). The velocity of translation may be
resolved into components along the three axes of a Cartesian coordinate system, so that
we may write 1/2mv2 = 1/2mvx2 + 1/2mvy2 + 1/2mvz2, where vx is the x-component of
velocity, etc., and m is the mass of the molecule. This equation tells us that the total
translational KE of the molecule consists of three parts, each of which represents the
kinetic energy along one of the reference directions. Since any translation of the
molecule may be viewed as the vector sum of its motions along the three axes, the
kinetic energy may always be broken up into the sum of three contributions, one arising
from motion along each axis. We say that the molecule has 3 translational degrees of
freedom, one corresponding to each Cartesian axis.

Second, the molecule may rotate about some internal axis. Again, any such axis may be
resolved into components along the x, y, and z axes of a Cartesian coordinate system, so
any rotation of the molecule may be resolved into three perpendicular components. The
rotational kinetic energy of the molecule can then be written

KErot = 1/2Ixwx2 + 1/2Iywy2 + 1/2Izwz2

where Ix, Iy, and Iz are the moments of inertia and wx, wy, and wz are the angular
velocities about the x, y, and z axes respectively. We see that there are 3 degrees of
freedom associated with rotational motion, one corresponding to each Cartesian axis. An
exception to this statement arises for a linear molecule, for which one of the three axes
is normally taken as the molecular axis. Consider a diatomic molecule shown with its
bond axis coincident with the z axis of a coordinate system. Because the moment of
inertia about this axis is zero, the molecule has no rotational energy about the z axis. For
linear molecules there are only two rotational degrees of freedom, rather than 3.

Finally, the molecule may vibrate. A diatomic molecule, for example, vibrates by
repeated stretching and contraction of the bond joining the two atoms. Such a molecule
has, in addition to 3 translational and 2 rotational degrees of freedom, one vibrational
degree of freedom. For a polyatomic molecule, we may deduce the number of vibrational
degrees of freedom (sometimes called vibrational modes) by subtracting the number of
translational (3) and rotational (2 or 3) degrees of freedom from the total number of
degrees of freedom possessed by the molecule. The latter is 3N, where N = the number
of atoms in the molecule. (Each atom may independently move in any of three directions
and therefore has 3 degrees of freedom available to it, for a total of 3N for the molecule.)
The number of vibrational degrees of freedom is therefore given by 3N-6 for a non-linear
polyatomic molecule; and by 3N-5 for a linear polyatomic molecule. According to this
rule, the water molecule, which is non-linear with N = 3, has 3 vibrational degrees of
freedom and should undergo three independent types of vibration. The arrows show the
directions in which the atoms move during the vibration.

Each molecular vibrational motion occurs with a frequency characteristic of the molecule
and of the particular vibration. The energy of a vibration is measured by its amplitude
(the distance moved by the atoms during the vibration), so the higher the vibrational
energy, the larger the amplitude of the motion. According to quantum mechanics, only
certain vibrational energies are allowed to the molecule (this is also true of rotational
and translational energies), so only certain amplitudes are allowed. Associated with each
vibrational motion of the molecule is a series of energy levels (or vibrational energy
states). The molecule may go from one energy level to a higher one by absorption of a
quantum of electromagnetic radiation, such that Efinal-Einitial = hn. In undergoing the
transition, the molecule gains vibrational energy, and the amplitude of the vibration
increases. The frequency of light required to cause a transition for a particular vibration
is equal to the frequency of the vibration, so we may measure the vibrational
frequencies by measuring the frequencies of light absorbed by the molecule. Since most
vibrational motions in molecules occur at frequencies of about 1014sec-1, light of
wavelength l = c/n = 3 x 1010 cm/sec/1014sec-1 is required to cause transitions. Light of
this wavelength lies in the infrared region of the spectrum. IR spectroscopy, then, deals
with transitions between vibrational energy levels in molecules. An IR spectrum is
displayed as a plot of the energy of the infrared radiation (usually expressed in
wavenumbers) versus the percent of light transmitted by the compound.

The spectrum of the molecule appears as a series of broad absorption bands of variable
intensity, each providing structural information. Each absorption band in the spectrum
corresponds to a vibrational transition within the molecule, and gives a measure of the
frequency at which the vibration occurs. For water, with three vibrational degrees of
freedom, there are three sets of energy levels within which transitions may occur. The
spacing between energy levels depends upon the particular vibration considered. Each
spacing requires a photon of different energy to cause the transition, so we expect
photons of three different energies to be absorbed by H2O. These are found at 3500 cm-1,
1650 cm-1, and 600-300 cm-1, respectively.

The fact that a molecule vibrates does not in itself insure that the molecule will exhibit
an IR spectrum. For a particular vibrational mode to absorb infrared radiation, the
vibrational motion associated with that mode must produce a change in the dipole
moment of the molecule. HCl, for example, with a center of positive charge at the H
atom and a center of negative charge at the Cl atom, has a dipole moment. The
magnitude of the dipole moment changes as the HCl bond stretches, so this vibration
absorbs IR radiation. We say that the vibration is IR active. The N2 molecule, on the other
hand, has no dipole moment. Further, stretching the N-N bond does not produce a
change in dipole moment, so the vibration is infrared inactive (i.e., cannot directly
absorb IR radiation). It is important to realize that there are many molecules that,
although possessing no permanent dipole moment, still undergo vibrations that cause
changes in the value of the dipole moment from 0 to some non-zero value. An example
is CO2, which has no permanent dipole moment because the individual bond dipoles
exactly cancel. However, when CO2 undergoes a bending vibration, its dipole moment
changes from zero to some non-zero value. This vibration produces a change in dipole
moment and is IR active.

All three vibrational motions of water cause a change in dipole moment. Hence all three
vibrations are IR active and we expect each to produce an absorption band in the IR
region. We have already seen that this expectation is borne out. The requirement that a
vibration cause a change in the dipole moment of the molecule in order to absorb
radiation can be understood simply. Exchange of energy between electromagnetic
radiation and matter can occur only if the radiation and matter can interact (or couple) in
some way. Electromagnetic radiation consists of perpendicular electric and magnetic
fields that oscillate sinuosoidally at the frequency of the radiation. Oscillation of the
electric field is equivalent to an oscillating dipole moment. A molecule may interact with
the radiation by interacting with the oscillating electric field, but this is possible only if
the molecule also possesses an oscillating electric field. Furthermore, the frequencies of
the oscillations must be the same. This is called the requirement of frequency matching,
and is universal in spectroscopy. If a vibrational motion of a molecule is to absorb IR
radiation, the motion must generate an oscillating electric field. This is equivalent to
saying that the vibration must produce a change in dipole moment. The dipole moment
oscillates at the frequency of the vibration, so radiation of this frequency may be
absorbed.

In addition to the number of vibrations expected for a molecule, we can make

statements about where in the spectrum absorptions due to certain types of vibrations
are expected to occur. For a diatomic molecule, A-B, the wavenumber (i.e., the reciprocal
of the wavelength) of the infrared radiation absorbed by the molecular vibration is given
by

(1) n = 1/2p (kAB/mAB)1/2 where mAB = MAMB/(MA + MB)

kAB is the force constant for the bond between A and B, and is a measure of the bond
strength. MA and MB are the masses of the atoms. We conclude from this equation that
the heavier the atoms involved in the bond, the lower the absorption frequency, given a
constant bond strength.

Applications of IR Spectroscopy to Organic Molecules. Organic functional groups

(atom groups bonded in particular ways) differ both in the strengths of the bond(s) and
in the masses of the atoms involved. For instance, the O-H and C=O functional groups
each contain atoms of different masses connected by bonds of different strengths.
According to equation (1), we therefore expect the O-H and C=O groups to absorb IR
radiation at different positions in the spectrum. The presence of a strong, broad band
between 3200 and 3400 cm-1 indicates the presence of an O-H group in the molecule,
while the presence of a strong band around 1700 cm-1 confirms the presence of a C=O
group.

For organic molecules, the infrared spectrum can be divided into three regions.
Absorptions between 4000 and 1300 cm-1 are primarily due to specific functional groups
and bond types. Those between 1300 and 909 cm-1, the fingerprint region, are primarily
due to more complex vibrational motions; and those between 909 and 650 cm-1 are
usually associated with the presence of benzene rings in the molecule.

Applications of IR Spectroscopy to Inorganic Molecules.

Many so-called inorganic compounds are in reality largely organic, and for these we look
for the same functional group bands in the IR as we do for purely organic compounds.
However, the infrared spectra of relatively simple, purely inorganic compounds
containing only a few atoms--specifically, inorganic salts containing polyatomic
(complex) ions--are quite distinctive and can be used to rapidly identify the ions.
Consider a simple inorganic salt, such as KNO2. On the basis of the empirical formula, we
might naively expect there to be a total of 3(4)-6 = 6 normal modes of vibration
associated with this material. However, this assumes that KNO2 is covalent. In fact, KNO2
consists of an ionic lattice of K+ and NO2- ions arranged in an infinite and very regular
array. The crystal consists of essentially isolated K+ ions and NO2- ions. Thus we are able
to consider the vibrational modes of the cation and anion independently of one another.
In this case, since the potassium ions are monatomic, they have no vibrations (3(1)-3 =
0), so we need only consider the nitrite anions. The VSEPR (Valence Shell Electron Pair
Repulsion) Theory predicts a bent structure for the nitrite ion. We thus anticipate three
normal vibrational modes for NO2-, corresponding to the diagrams drawn earlier for H2O,
and they should all be infrared active. Indeed, three bands are observed in the IR
spectrum of KNO2: the symmetric stretch at 1335 cm-1, the asymmetric stretch at 1250
cm-1, and the bending vibration at 830 cm-1 (bending vibrations occur in general at lower
frequencies than stretching vibrations). The frequencies of these vibrations are about the
same regardless of counter ion, substantiating the independence of the anion and cation
in the crystal. (This independence is only an approximation, but we will not worry about
the complicating factors now.) We can, therefore, diagnose the presence of nitrite ion in
a salt from the infrared spectrum of the material. This diagnostic application can be
implemented only when the spectrum of the material is relatively uncomplicated, but
despite this restriction it is an enormously useful application.

Let us turn now to the somewhat more complex case of NaNO3. Here we anticipate 3(4)-
6 = 6 normal vibrational modes. The infrared spectrum, however, exhibits only three
fundamental bands, at 831, 1405, and 692 cm-1. Why? There is no doubt that there are 6
normal vibrational modes. The formula is always valid. In the case of NO3-, however, the
symmetric stretch is not IR active because it does not cause a change in the dipole
moment of the ion, and so cannot give rise to absorption of IR radiation. This eliminates
one of the anticipated bands from the IR spectrum. Among the remaining 5, there are
two sets of doubly degenerate vibrations--i.e., two instances in which 2 vibrations occur
with exactly the same frequency. Thus although 5 vibrations absorb IR radiation, they
are manifested in only three spectral bands. However, these absorptions can be used
diagnostically just as for nitrite. In similar fashion, other relatively simple anionic (and
cationic) species can be identified via their IR spectra.