Applied

Spectroscopy

Raman Spectroscopy

Recommended Reading:
Banwell and McCash Chapter 4
Skoog, Holler Nieman Chapter 18
Atkins, Chapter 16
Inelastic scattering of light is sometimes called the
Raman effect, named after its discoverer, the
Indian scientist
C V Raman
Raman won the Nobel prize in 1930 for his
discovery, accomplished using filtered sunlight as
a monochromatic source of photons, a colored
filter as a monochromator, and a human eye as
detector. The technique became widely used after
the invention of the laser.
http://www.fact-index.com/r/ra/raman_spectroscopy.html
 Raman Spectroscopy
What is it? Raman Spectroscopy determines vibrational and rotational
level spacings from the energy (wavenumber) shifts of scattered light.

How do we do it it? Molecular transition energies are observed by

measuring the shifts in frequency of light scattered when a molecule
is subjected to an intense beam of monochromatic light.

Why we do it it? Previously we noted that molecules with no

permanent dipole moment would have no pure rotational spectrum,
and that molecular vibrational motion for which there was no
oscillation dipole would have no IR absorption or emission spectra.
Raman spectroscopy allows us to determine rotational and vibrational
level spacings for such systems, and hence to determine bond
lengths and force constants for such molecules. That is we can use
Raman spectroscopy to study H2, O2, N2 ….
 C.V.Raman

Sir Chandrasekhara Venkata

RamanIndia (1888-1970)
Nobel Prize, 1930; for his work on
the scattering of light and for the
discovery of the Raman Effect
 Basic Experiment
In Raman spectroscopy the energy levels of the molecules are explored by
examining the frequencies present in the radiation scattered by the molecules.

The photons can be scattered either

1) ELASTICALLY
Scattered photon has the same
energy as the incident photon ⇒
Rayleigh Scattering
or
2) INELASTICALLY
Scattered photon has an energy
that is less than or greater than the
incident photon ⇒ Raman
Scattering.

About 1 in 107 photons are in elastically

scattered ⇒ very weak process
 Polarizability α
Electric field
α⊥ α⏐⏐

Electric field

When an electric field E is applied to a molecule, its charge distribution is

distorted.
This induces a dipole moment μ, in the molecule.
The magnitude of the induced dipole moment varies linearly with the strength of
the electric field E (for weak fields)
μ = αE
The proportionality constant α is called the polarizability of the molecule.
α is a measure of how easy it is to distort the charge distribution of the
molecule.
α depends on the direction of the electric vector relative to the axis of the
molecule. i.e. it is a Tensor quantity
 Rotational Raman Spectroscopy
Gross Selection Rule:
The molecule must be
anisotropically polarizable

Spherical molecules are

isotropically polarizable and
therefore do not have a Rotational
Raman Spectrum

All linear molecules are anisotropically polarizable, and give a Rotational

Raman Spectrum, even molecules such as O2, N2, H2… which do not
have a Pure Rotational Spectrum!
⇒ can study rotational levels of molecules that are inaccessible by
Microwave Spectroscopy.
 When a molecule is subjected to a strong external electric field E that
field will distort the molecules electron charge distribution to induce a
small dipole moment which (to first order) is proportional to the
electric field:
μ induced = αE (α is a Tensor quantity)
where α is the polarizability of the molecule, a property which indicates how
readily its electron distribution can be distorted by the external field.

If E is a static laboratory field the

induced dipole may interact with
light in the normal manner and
transitions can occur. However, if
the external filed is due to an
intense beam of light, then this is a
second order process, since the
light first creates the dipole and then
interacts with it ! Transitions created
in this manner will be very weak but
observable.
 Specific Selection Rule
Electric field is due to electromagnetic radiation of frequency ω i.
Induced dipole moment μ is then
μ = αE(t ) = αE0 cos ωit
E0 = amplitude of electric field. If molecule is rotating at an angular frequency
ω R, then its polarizability will be time dependent (if it is anisotropic), and
α(t ) = α 0 + Δα cos 2 ω R t
where αo = (2α⊥ +αVV)/3 = spherical average of parallel and perpendicular
components and Δα = αVV - α⊥ =polarizability anisotropy
α ranges from αo - Δα to α o + Δα as the molecule rotates through 2π
Factor of 2 because polarizability returns to its initial value twice during
every revolution of the molecule
Substituting the expression for the polarizability into the expression for the
induced dipole moment gives
μ = (α 0 + Δα cos 2 ωR t ) × (Ε cos ωit ) = α 0 Ε cos ωit + ΕΔα cos ωR t cos ωit
1
μ = α 0E 0 cos ωit + E 0 Δα{cos(ωi + 2 ωR ) t + cos(ωi − 2 ωR ) t }
2
 Rotational Raman Spectroscopy
1
μ = α 0 Ε cos ωit + ΕΔα{cos(ωi + 2 ωR ) t + cos(ωi − 2 ωR ) t }
2

Component oscillating at Component oscillating at Component oscillating at

the same frequency as frequency ω i + 2ω R, ⇒ frequency ω i - 2ω R, ⇒
incident radiation ⇒ Anti-Stokes Stokes Scattering
Rayleigh Scattering Scattering ⇒ Lower energy
⇒ Higher energy
Note that Stokes and Anti-Stokes components disappear if
Δα = 0 so molecule must be anisotropically polarizable.

Two photons involves in Raman process, each is a spin-1 particle,

so a maximum change in angular momentum quantum number
of ± 2 is possible This then gives a selection rule for Rotational
Raman Transitions
ΔJ = ±2
 Rotational Raman Spectrum: Stokes Lines
We can predict the form of the rotational
Raman spectrum of a linear rotor by
applying the selection rule ΔJ = ± 2 to the
rotational energy levels When the
molecule makes a transition with ΔJ = + 2
the scattered radiation leaves the ~ νi ~
~ νi − 10 B
νi
molecule in a higher rotational state, so
the wavenumber of the incident radiation, ~
νi − 6B
initially , is decreased. These ΔJ = + 2 J
transitions account for the Stokes lines in
3 12B
the spectrum.
ν (J + 2 ) − ~
~ ν (J) = ~
νi − {F(J + 2 ) − F(J)}
2 6B
=~ν − [B(J + 2 )(J + 3 ) − BJ(J + 1)]
i 2B
1
0
νi − 2B(2 J + 3 )
=~ (1) 0B

The Stokes lines appear on the low frequency side of the incident
radiation and at displacements of 6B, 10B, 14B, .... from ~
νi
for J = 0,1,2,...
 Rotational Raman Spectrum: Anti-Stokes Lines
When the molecule makes a transition with
ΔJ = -2, the scattered photon emerges with
increased energy. These transitions
account for the anti-Stokes lines of the
spectrum. ~
νi ~
νi + 10 B
ν (J − 2 ) − ~
~ ν ( J) = ~
νi + {F(J − 2 ) − F(J)} ~
νi ~
νi + 6B
=~ν + [B(J − 2 )(J − 1) − BJ(J + 1)]
i
J
νi + 2B(2 J − 1) (2)
=~ 3 12B

The anti-Stokes lines occur at 2 6B

displacements of 6B, 10B, 14B...for J =
1 2B
2,3,4..to the high frequency side of the
0 0B
incident radiation; Note that J = 2 is the
lowest state that can contribute to the
spectrum under the selection rule
ΔJ = -2.
 6 42B

J 30B
5

4 20B

3 12B

2 6B
1 2B
0 0
4B
4B

6B 6B

Stokes ~
νi anti-Stokes
The separation of adjacent lines in both
the Stokes and anti-Stokes regions is
4B, so from its measurement I can be
determined and then used to find the
bond lengths exactly as in the case of
microwave spectroscopy.
 Example: Predict the form of the rotational Raman spectrum of 14N2 for
which B = 1.99cm-1, when it is exposed to monochromatic 336.732 nm
laser radiation
The molecule is rotationally Raman active because end-over-end rotation
modulates its polarizability as viewed by a stationary observer. Because λi
=336.732 nm corresponds to 29697.2 cm-1 equations (1) and (2) give the
following line positions:
J 0 1 2 3
Stokes Lines
ν /cm-1 29685.3 29677.3 29669.3 29661.4
λ /nm 336.868 336.958 337.048 337.139
Anti-Stokes
Lines
ν /cm-1 29709.1 29717.1
λ /nm 336.597 336.507

There will be a strong central line at 336.732 nm accompanied on either

side by lines of increasing and then decreasing intensity (as a result of the
transition moment and population effects). The spread of the entire
spectrum is very small, so the incident light must be very monochromatic
(laser).
 Pure Vibrational Raman Spectroscopy
Gross Selection Rule: The polarizability of the molecule must change as
the molecule vibrates.

For all diatomic molecules the interaction between the electrons and
nuclei changes as the molecule vibrates ⇒ a change in polarizability ⇒
all diatomic molecule (homonuclear and heteronuclear) are
vibrationally Raman active.

Analysis becomes much more difficult for polyatomic molecules, need

to apply group theory to decide if a particular vibrational mode is Raman
active.

Specific Selection Rule:

1) Harmonic Approximation Δ n = ± 1.

2) Anharmonic Approximation Δ n = ± 1, ± 2, ± 3,…

I.e same selection rules as for pure vibrational (IR) spectroscopy

The classical description of vibrational Raman spectroscopy is qualitatively
similar to that presented above for rotational Raman spectroscopy.
In this case the vibration of the molecular bond is accompanied by a
periodic stretching and compression of the electron distribution which gives
an oscillation of the component of the molecular polarizability along the
direction of the electric field.

Using μ nduced = αE shows

that this gives rise to an
induced dipole which
oscillates in phase with the
vibrational motion.
α(t ) = α 0 + δα cos(2 πν vib t )
Where δα is the amplitude
of the change in the
polarizability during one
vibrational cycle and α0 is
the average value over a
cycle/
Just as in rotational Raman spectroscopy, if the external field is due to an
intense beam of monochromatic light, the substitution of
α(t ) = α 0 + δα cos(2 πν vib t ) into μ nduced = αE
gives the following expression for the time dependent induced dipole
moment
1
μ = α 0E0 cos 2 πν it + E0 δα{cos[2 π(ν 0 + ν vib ) t ] + cos[2 π(ν 0 − ν vib ) t ]}
( )
2

Just as in rotational Raman spectroscopy, this equation shows Rayleigh

scattering at the same frequency as the incident light. Stokes scattering at
the frequency (ν 0 − ν vib )
and anti-Stoles scattering at the frequency (ν 0 + ν vib )
Key difference from rotational case is the fact that here the frequency shift is
±νvib (rather than ±2νvib ) . As a result the vibrational Raman selection rule is
based on the very strong preference for Δn = ±1 which governs normal
infrared spectroscopy.
 Structure of Vibrational Raman Spectra
For every vibrational mode of the molecule, the energy of the mode is given
by 2
En ⎛ 1⎞ ~ ⎛ 1⎞
G(n) = = ⎜ n + ⎟ νo − ⎜ n + ⎟ χ e ~ νo
hc ⎝ 2⎠ ⎝ 2⎠
where ω = 1 k
o
⎡ ⎛ 1 ⎞ ⎤ ⎛ 1 ⎞ 2 πc μ
=~ν o ⎢1 − χ e ⎜ n + ⎟ ⎥ ⎜ n + ⎟
⎣ ⎝ 2 ⎠ ⎦⎝ 2⎠
exactly the same expression we had for vibrational spectroscopy.
For Raman active modes we can observe the following transitions:

Fundamental: Δ n = ± 1.
where Δ n = + 1 corresponds to the Stokes line. The molecule starts in
the n = 0 state, is excited by the incident radiation and deexcites to
the n = 1 state.
Thus the molecule has gained energy and the scattered photon has
lower energy than the incident photon. The energy of the scattered
photon is ~ νi - (~ ν0 ) = ~
ν1 - ~ νi - G(1) − G(0 ) = ~ ν0 (1 − 2 χ e )
νi - ~
where ~
νi where is the wavenumber of the incident radiation
Δ n = - 1 corresponds to the Anti-Stokes line. The molecule starts in
the n = 1 vibrational state, is excited by the incident radiation and
deexcites to the n = 0 vibrational state.
Thus the molecule has lost energy and the scattered photon has
higher energy than the incident photon. The energy of the scattered
photon is
νi + (~
~ ν1 ) = ~
ν0 - ~ νi + G(0) − G(1) = ~ ν0 (1 − 2 χ e )
νi + ~

However, recall that at room temperature very few molecules are in

the n = 1 vibrational state (Boltzmann Distribution). So the Stokes line
will be very weak compared to the Anti-Stokes line (often the Stokes
line is not observed).

It is possible to determine the temperature of the system from the

ratio of the intensity of the Stokes line to the intensity of the
AntiStokes line.

Pure Vibrational Raman spectrum is very simple, A series of intense

lines to the low wavenumber (frequency, energy) side of the exciting
(laser) line with a weaker mirror image series of lines on the high
wavenumber side.
 Vibrational Raman Process
Rayleigh Raman
Scattering Scattering
E = hν E = hν ± ΔE
E = hν
Virtual
E = hν States (Polarised molecule)

Anti-Stokes
~

Stokes
νi ~
νi
~
νi ~
νi

2
Ground
1
State
0
 Vibrational Raman Spectrum
It is also possible to observe Overtone Bands and Hot Bands in Raman
spectra.
Overtones: Δ n = 2, 3, ...
I.e overtones are produced by the following transitions
First Overtone: n = 0 → n = 2
~ νi - (~
ν=~ ν0 ) = ~
ν2 - ~ νi - G(2 ) − G(0 ) = ~ ν0 (1 − 3 χ e ) anti-Stokes
νi - 2 ~
Second Overtone: n = 0 → n = 3
~ νi - (~
ν=~ ν0 ) = ~
ν3 - ~ νi - G(3) − G(0 ) = ~ ν0 (1 − 4 χ e ) anti-Stokes
νi - 3 ~

Hot Bands: produced by the following transitions

n=1→n=2 ~ νi - (~
ν=~ ν1 ) = ~
ν2 - ~ νi - G(2 ) − G(1) = ~ ν0 (1 − 4 χ e )
νi - ~
with weaker hot bands for n = 1→ n = 3, 4, 5...

However, since the Raman Scattering Process is very weak we can ignore
all pocesses such as Overtones and Hot bands since these are weak even
in IR spectra. So we only need to consider the fundamental transitions n =
0 → n =1.
 Raman Spectrum: Diatomic Molecule
Diatomic molecule has only one vibrational frequency (i.e. only one
mode) Therefore the Raman spectrum of a diatomic molecule should
consist of a single Stokes and a single antiStokes line (ignoring
overtone and hot bands and rotational fine structure)

1 k
ν0 =
2π μ

ν0 ν0

νS νinc νAS frequency or

wavenumber
Relative intensity of antiStokes line to Stokes line gives the
temperature of the sample (Boltzman distribution)
Vibrational Raman Spectra
example: CHCl3 molecule.
Raman Symmetric top molecule with 5
atoms ⇒ 3N - 6 = 15 - 6 = 9
vibrational modes,
however, due to symmetery 3
modes are degenerate ⇒ 6
vibrational modes.

Modes are seen at 262, 366, 668,

761, 1216 and 3019 cm-1. All six
Infrared modes are both IR and Raman
active. Exciting source is Ar 488 nm
laser line

In the Raman spectrum on the left

the wavenumber of the incident
(laser) radiation is set to zero, so x-
axis showes wavenumbers of the
vibrational modes (and not absolute
values)
 Raman Activity in Polyatomics
In general for polyatomic molecules it is usually necessary to apply Group
Theory in order to decide whether a particular vibration of the molecule is
Raman active or not. But some general rules apply:

If the molecule has no symmetry (e.g. HCN) then usually all of its
vibrational modes are Raman active

In molecules that possess symmetry (e.g. CO2, H2O) then

Symmetric vibrations give rise to intense Raman lines, non symmetric
vibrations are usually weak and sometimes unobservable. In
particular bending modes usually yield a very weak Raman line.

Rule of Mutual Exclusion:

If the molecule has a centre of symmetry, then Raman active
vibrations are infrared inactive and vice versa. If there is no centr of
symmetry in the molecule then some or all of the vibrations may be
both Raman and IR active.
Raman Activity
Vib. Raman Spectra: Example CCl4
Comparison of Raman and IR Spectra
 Vibrational-Rotational Raman spectroscopy
This fine structure is rarely resolved except in the case of diatomics.
We can write the vib-rot energy levels as (see vib-rot notes)
2
⎛ 1⎞ ⎛ 1⎞
S(n, J) = ⎜ n + ⎟~ ν + BJ(J + 1)
ν - ⎜n + ⎟ χe~
⎝ 2⎠ ⎝ 2⎠
where n = 0, 1, 2, …; J = 0, 1, 2, 3 … and again we have ignored the
centrifugal distortion (D term).

For diatomic molecules ΔJ = 0, ± 2 and combining this with the

fundamental transition n = 0 → n = 1 gives

Q-Branch ΔJ = 0 : ν (1 − 2 χ e ) cm −1
Δε Q = ~ J = 0, 1, 2, 3...

S-Branch ΔJ = +2 : ν (1 − 2 χ e ) + B(4 J + 6 ) cm −1
Δε S = ~ J = 0, 1, 2, . ..

O-Branch ΔJ = −2 : ν (1 − 2 χ e ) − B(4 J + 6 ) cm −1
Δε O = ~ J = 2, 3,4, . ..
Stokes lines, lying at low frequency (wavenumber) side of exciting
radiation will occur at wavenumbers given by.

~
νQ = ~
ν ex - Δε Q = ~ ν (1 − 2 χ e ) cm −1
ν ex - ~ J = 0, 1, 2, 3...

~
νS = ~
ν ex - Δε S = ~ ν (1 − 2 χ e ) − B(4 J + 6 ) cm −1
ν ex - ~ J = 0, 1, 2, . ..

~
νO = ~
ν ex - Δε O = ~ ν (1 − 2 χ e ) + B(4 J + 6 ) cm −1 J = 2, 3,4, . ..
ν ex - ~
νex.is the frequency of the incident (exciting) radiation, e.g. laser
frequency
Pure rotation
Raman Lines
Rotation Vibration ~
νQ
Raman Lines

~
νS ~
νO

~ ν (1 − 2 χ e )
νex - ~ ν (1 − 2 χ e )
~ ~
νex
 Polarisation Effects in Raman Spectrum
If the incident light is polarised then the Raman process can change the state
of polarisation of the scattered radiation, it can become partially depolarised,
i.e. the scattered radiation can be elliptically or even circularly polarised.

I⊥
I||
 Depolarisation Ratio
we can quantify this effect as follows. The intensity of the Raman Scattered
radiation imeasured through a polariser aligned (I) parallel I⊥ and (ii)
perpendicular I|| to the plane of polarisation of the incident light.

The Depolarisation Ratio ρ of a particular Raman line is then defined as

I
ρ= ⊥
I||

If the scattered light is unpolarised then I⊥ = I|| and ρ = 1 .

If the scattered light retains its initial polarisation then I⊥ = 0 and ρ = 0 .

A Raman lines is classified as depolarised if ρ ≥ 0.75.

and is classified as polarised if ρ < 0.75.

Only totally symmetric vibrations give rise to polarised lines. So by measuring

the depolarisation ratio we can distinguish between symmetric vibration
modes and other modes (bending, asymmetric ..)
 Instrumentation for Raman Spectroscopy
Sources: Lasers are used almost exclusively for Raman Spectroscopy.
High intensity ⇒ good signal-to-noise ratio.
Narrow linewidths of laser good for resolving Raman lines.

Common laser sources are

Argon Ion 488 and 514.4 nm Uv/vis
Krypton Ion 530 and 647 nm vis
Helium Neon 623 nm vis
Diode Laser 782 ad 820 nm vis/IR
Nd/YAG 1064 nm IR

The intensity of a Raman line is proportional ν4. I ∝ ν4

⇒ Argon and Krypton Ion lasers have an advantage over other laser
sources.
Example: Argon line at 488 nm produces a line 2.7-times more intense
as those excited by He/Ne laser for the same input power.
4 4 4
⎛ ν Ar ⎞ ⎛ c λHeNe ⎞ ⎛ 623 ⎞
⎜⎜ ⎟⎟ = ⎜⎜ ⋅ ⎟⎟ = ⎜ ⎟ = 2 . 66
⎝ νHeNe ⎠ ⎝ λ Ar c ⎠ ⎝ 488 ⎠
However, the diode laser and Nd/YAG also have advantages.
They operate in the near infrared ⇒

1) They can be operated at much higher power without causing

photodissociation of the sample (up to 50 W).

2) These photons are not energetic enough to excite electronic

states in the molecule so there is little or no fluorescence radiation
to swamp the Raman signal. (see figure on next overhead)

Sample Illumination: Much easier than IR spectroscopy, since

Raman signal is usually in the visible/ near IR region this means
that special windows, mirrors, lenses are not required, ordinary
glass may be used.
Easier to focus laser to a small spot size (microRaman) ⇒
small samples may be investigated.

See Banwell and McCash Section 4.7 AND Skoog, Nieman and Holler
Section 18 B For details on Instrumentation
Ar Ion versus Nd YAG source
Raman spectrum of
anthracene.

A: conventional Raman
instrument using a Ar Ion
laser.

B: Fourier Transform
instrument using Nd/YAG
laser.

Note huge fluorescence

contribution to the
spectrum taken with Ar Ion
laser.
Conventional Raman Spectrometer
Raman Spectrum of
myoglobin
Conventional Raman Spectrometer
 Fourier Transform Raman Spectroscopy

FT Raman
spectrum of
powdered C60
Raman Microscope
 Coherent AntiStokes Raman Spectroscopy
See Atkins Physical Chemistry (6th edition) Section 16.16 (d)
The intensity of Raman transitions may be enhanced by Coherent anti-
Stokes Raman Spectroscopy (CARS). The technique relies on the fact
that, if two laser beams of frequencies ν1 and ν2 pass through a sample,
then they may mix together (non-linear optical mixing) and give rise to
coherent radiation of several different frequencies, e.g.
ν1 + ν2, ν1 + ν2 , 2ν1 + ν2, 2ν1 - ν2 , … If we consider the mixed beam at
frequency
ν 3 = 2 ν1 - ν 2
Then suppose that ν2 (dye laser) is varied until it matches any of the
Stokes lines from the sample, such as one with frequency ν1 - Δ ν, then
the coherent radiation will have frequency
ν 3 = 2 ν 1 - (ν 1 - Δν ) = ν 1 + Δν

which is the frequency of the corresponding anti-Stokes line. This

coherent radiation forms a narrow beam of high intensity, and since it is
in the anti Stokes region of the spectrum there will be no fluorescent or
background radiation to interfere with it.
Coherent AntiStokes Raman Spectroscopy

CARS Experiment
ν2 = ν1 - Δν
Laser ν2 = 2ν1 - ν2 = 2ν1 - (ν1 - Δν) = ν1 + Δν

Detector

ν1 Dye Laser ν3
ν2 Sample

ν1
 Coherent AntiStokes Raman Spectroscopy
If the two incoming beams
ν3 are focused at a small angle
2α at the sample then the
ν1 3α collimated CARS radiation
emerges at an angle 3α to
2α
the optic axis and can be
ν2 easily filtered from ν1 and ν2
using a pinhole.

Advantage of CARS: It can be used to look at Raman transitions in the

presence of competing incoherent background radiation, for example it
can look at the Raman Spectra of species in flames. The intensities of
transitions can be interpreted in terms of the temperature in different
regions of the flame.
We could also tune the dye laser ν2, until it matched an anti-Stokes
transition at ν1 + Δ ν then
ν 3 = 2 ν 1 - (ν 1 + Δν ) = ν 1 − Δν
= a frequency corresponding to a Stokes line ⇒ Coherent Stokes Raman
Scattering (CSRS) not used much
 Raman vs. IR Spectroscopy
Advantages:
1) The incident and scattered radiation is in the visible or UV ⇒
conventional optics, monochromators and sample cells can be used
2) Can focus the laser beam to very small spot size (difficult to focus
IR)⇒ can use small samples
3) Water absorbs strongly in the IR (see vib notes, double beam
spectrometer) but water is a very weak Raman scatterer ⇒ can use
aqueous solutions for Raman Spectroscopy (not possible for IR where
other solvents are required for liquid samples. ⇒ can examine biological
samples.
4) Using Pulsed laser sources we can look at time dependence of Raman
Intensities ⇒ can observe transient species in reactions.

Disadvantages:
1) Sample heating when lasers are used
2) fluorescence in sample can swamp Raman signal