Electronic Absorption Spectroscopy

Classical View
According to the classical theory discussed previously, the dynamics of an electron in an
atom are governed by its natural fiequency oo = .\l(klm,) and by dissipative processes-
radiation and "viscous" damping. Monochromatic electromagnetic radiation at
fiequency o will drive the electron in oscillatory motion, but the amplitude of the motion
remains small until o + oo, a resonance condition. The extent to which the atom is
polarized by an electric field, E, is determined by the polarizability, a, where

(Note that a is a tensor while m, the dipole moment, and E, the electric field, are vectors.)

As seen in the previous discussion, a is small when o >> oo and o << oo. The in-phase
component is related to the index of refraction and therefore governs the scattering of
light. The out-of-phase component of a governs the absorption of light i.e., the processes
by which, excludmg fluorescence, etc., light energy is dissipated into heat.

According to this picture the electron can have a continuous range of energies. Thus the
displacement of the electron in the harmonic (spring) potential is unrestricted. Hence the
energy emitted by an electron relaxing to its resting state depends on the maximum
amplitude of the displacement in the harmonic potential (xo).

Ouantum View
Quantum mechanics, which provides a more correct analysis o electronic motion and
ato9rnic structure, shows that these conclusions are incorrect. AccorSding to quantum
mechanics the electron in the atom must be in one of a set of defined states, each with a
definite energy. Hence, an electron in state i has an energy Ei. When the electron passes
from state i to state j, it emits (or absorbs) energy E,! - Ei, corresponding to emission or
absorption of light with frequency:

where h is Planck's constant. What causes the electron to undergo a transition from one
state to another?

First, we must remember that the position of the electron in the atom is characterized by a
probability density function, P(r) = luC(r)f,where ~ ( ris) a probability amplitude function
or wave function obtained by solving the Schroedinger equation appropriate for the
dynamic system under study. Solution of this equation also provides the Ei.
Quantum mechanics tells us how an electron can be caused to go fiom one state to
another by perturbing its energy. In particular, for an atom exposed to light, it is the
interaction of the atomic dipole moment p with the electric field vector of the light, E,
provides the perturbation energy V(t), i.e.,

Then it can be shown that the rate at which an electron passes from state b to state a is

where Bab= (2/3)(h2/4n3)-I~<blp~@fand <blp(a>(= d3r vb(r) p(r) va(r) and I(v) is the
incident intensity at fiequency v.

The transition dipole moment, <b(pla.,measures the extent to which the electron in state
a is polarized by the incident light so that its spatial distribution is similar to that of state
b. More generally,

Under ordinary conditions Nbis small, and so NbBbacan be neglected.

Incident light will promote a transition from a + b, and thereby absorption fiom the
incident radiation field if the fiequency of the light, v = (E,- Ea) /hand if <blp(a># 0.
Hence, measuring the absorption of light while varying the orientation of its electric
vector relative to the molecules axes provides information about the orientation of the
transition dipole moments within the molecule.

The Molar Extinction Coefficient.

According to the Beer-Lambert law:

where E = molar extinction coefficient

C = molar concentration of the absorbing species
I = incident intensity
1= pathlength through the solution.

Also, for a 1M solution the rate of energy absorption per cm3 of solution is
But dI(v)/dt = [dI(v)/dl]*dVdt = cdI(v)/dl.

Therefore, -dI(v) = [ h ~ N ~ B ~ d 1 0I(v)dl,

0 0 ~ ] and so Bab= 2303&~/(N&~),
or more
generally,

Bab= (2303c/Noh) 1(&(v)/v)dv

and so,

(cblp(a>f is called the "dipole strength".

Linear Dichroism

The dichroic ratio is (A, - AI)/(A,, + Al). This parameter provides information about
the orientation of the transition dipole moment(s) relative to the molecular axes.

Absorbance Properties of Proteins.

\\
Peptide chromophore

'\. a (nm)
x electrons are somewhat delocalized over the N, C, and 0 atoms. An electron in a
nonbonding, n, orbital is concentrated near the 0 atom. The Lowest energy electronic
transition fiom the peptide bond is an n + n* transition in the range 210 to 220 nm and is
very weak (because it is symmetry forbidden), E = 100 (M an)-'. The n + x* at ~ 1 9nm 0
is much more intense, c M 7000 (Man)-'and is not polarized along any specific bond.

Aromatic Chromophores

Figure 7-10
Absorption spectra of the three aromatic amino acids.
A log scale has been used in order to display all
three conveniently on one graph. [After D. B.
Wetlaufer, Adv. Protein Chem. 17: 303 (1962).]
Absorbance Properties of Nucleic Acids

The nucleotide bases dominate the near W absorption in nucleic acids. There are many
n + K* nd n + n* transitions between 200 and 300 nrn. The transition dipoles are in the
planes of the aromatic rings. On average E ~ M =
) 10,000 (M cm)-l .

Effects of Conformation

Hypochromism

Linear Dichroism of DNA

Figure 7-21
Schematic diagram showing origin of hypochromism and hyperchrornism. (a) This alignment of induce,
dipoles (unshaded)and transition dipole (black) produces hypochromism (shaded). (b) This alignmer
of induced dipoles and transition dipole produces hyperchromism.

Figure 7-22
Linear dichroism expected for the B-jonn
D N A double helix, when aligned as
shown relative to polarized incident
light. Because ( Y o ( t ( ~ , )is in the
plane of the base pairs, it is always
perpendicular to Ell.The intensity of
absorption will be periodic along the
helm because the angle between
<yo( tlY.) and E varies with each 36"
rotation of successive base pairs.
[DNA structure after A. Kornberg,
DNA Synthesis (San Francisco:
W . H. Freeman and Company, 1974).]
 Vibrational Spectroscopies

Molecular Vibrations

Infrared (IR) and Raman spectroscopies measure changes in the vibrational

energy levels of nuclei in molecules. We begin by reviewing the classical simple
harmonic oscillator. A particle of mass m is connected by a spring of stiffness constant k
to an immovable site. The spring is at equilibrium (no net force) at length x,. When the
spring is stretched or compressed, it exerts a force to restore its length to x, : F = -k(x-x,).
For simplicity let x, be the origin. Then F = -kx. The potential energy stored in the
spring is U = kx2/2. The classical equation of motion is m d2x/dt2= - kx, which yields
solutions of the form: X = A sins, where A is an amplitude of oscillation and o =
d (k/m).
The quantum mechanical analysis of the simple harmonic oscillator is discussed
at length in elementary textbooks. As expected, the system is quantized. The energies of
the system are characterized by En= (n + %)hvowhere n is a positive integer and v, =
[J(k/m)]~n.Note that the quantum mechanical and classical mechanical analyses give
the same fundamental frequency. Whereas a classical oscillator can have any energy, the
quantum oscillator is constrained to discrete energies. Even when n = 0, E, = hv,, the
"zero point" energy.

As in electronic spectroscopy, light is absorbed due to molecular vibration when

v = (E, - El)/h, where El and E2 are vibrational energy levels. A selection rule specifies
that An = f1. Hence, AEi = hv,.

The analysis of the vibrations of polyatornic molecules can be quite complex.

Equations of the form mi d2x(dt2+ Z, fijxj= 0 must be solved. This is a system of
simultaneous liner second order differential equations. There are 3N-6 equations where
N = the number of nuclei (for nonlinear molecules). Due to coupling of each nucleus, in
principle, to all the other nuclei, the solutions are in the form of normal modes. For
example, water has 3 x 3 - 6 = 3 normal modes.
For a specified normal model all the atoms move in phase at the frequency defined for
cs
that mode. Note that it is possible to find a coordinate system = CSqSxi such that the
equations of motion in the new coordinate system are uncoupled. That is,
p, d25/ dt2 + @,kI= 0. The 6, are the "normal coordinates" which define a normal mode
c,
of vibration. In the mode the system oscillates in phase at the frequency v, = d( +&).
Similarly, the Schroedinger equation can be expressed in terms of normal modes to yield
a system of uncoupled harmonic oscillator Schroedinger equations, each of which has
solutions in the form of a simple uncoupled oscillator.

Infrared and Raman Spectroscopies

Transitions between vibrational energy levels occur at much lower frequencies

than do electronic transitions: vibrational-2 pm I h I50 pm; visible (green light) h =
0.5 p.m. For a molecule to absorb an infrared photon due to a molecular vibration two
conditions must be satisfied:
(1) v = EJh and
(2) The permanent dipole moment of the molecule must change due to the molecular
vibration. Thus dWdR # 0 where p = permanent dipole moment and R = normal
coordinate

Example, CO,:

symmetric stretch +O=*O-, infrared inactive d t / d R = 0

+ t
asymmetric stretch O=C=O-, infrared active ++ap/aR
t
bending o=c=O infrared active fd~laR
1 1

Hence, one of the three vibrational modes of CO, will not show up in an IR spectrum.
Note that of the other two modes, one is polarized parallel, the other, perpendicular to the
molecular axis.
Raman spectra result from scattering rather than absorbance as in IR. In contrast
to IR the incident light, typically in the visible range, is far from the frequencies of
molecular vibrations. Rarnan complements IR; the polarizability rather than the dipole
moment must change with vibration.

When the incident light approaches an electronic absorption, there is a large

increase in the polarizability and consequently, a large increase in the amplitude of
Raman scattering. This is known as "resonance Raman" scattering.

Vibrational Spectroscopy of Proteins

In polypeptides three peptide backbone infrared bands are of greatest importance.

One is dominated by an N-H stretch at approximately 3300 cm-I, one by a C=O stretch at
 1630-1660 cm-' (amide I band), and the third by a N-H deformation at 1520-1550 cm-'
(amide 11). Characteristics of these bands are shown in the following table.

Table 8-5
Characteristics of principal infrared absorption bands of the peptide group
Hydrogen-bonded forms Non-hydrogen-bonded

a Helix fi Sheet
Frequency Frequency Frequency
Vibration $/aR (cm- ') Dichroism (cm-I) Dichroism '
(cm - )

N-H stretch +N--H+ t* 3,290-3,300 11 3,280-3,300 1 -3,400

Amide I ( e O stretch) c k 0 - t * 1,650-1,660 11 1,630 I 1,680-1,700

I
Amide I1 tC--N+
1
1,540-1,550 1 1,520-1,525 11 < 1,520?

SOURCE:Adapted from I. A. Schellman and C. Schellman, in The ProteiyL2d ed., vol. 2, ed. H. Neurath (New York: Academic Press,
1%2), p. I.

Infrared absorption dichroism is similar to electronic absorption dichroism. In the a-

helix N-H""O=C peptide hydrogen bonds are oriented parallel to the long axis of the
molecule. Then, for oriented samples, the N-H stretch and amide I bands should
preferentially absorb IR light when the polarization is parallel to the helix axis. The
amide I1 band should have orthogonal polarization.

In an actual protein or polypeptide the situation is more complex because of

interactions which couple the peptide vibrations to each other and to those of other
structural elements. In simple a-helices, however, the situation is simplified by the
symmetry of the structure. Due to the helical symmetry there are only three bands: an
intense polymer absorption at v,,polarized parallel to the helix and two degenerate bands
at vl polarized perpendicular to the helix axis. For anti-parallel P sheets there are four
vibration bands corresponding to normal vibrational modes of the fundamental
asymmetric unit (involving 4 peptides. One amide band is IR inactive, v(n, n);two other
bands are polarized perpendicular, and the fourth parallel to the helix axis as summarized
in the following table.
Table 8-6
Observed and calculated infrared spectra of polypeptides and proteins
--

Amide I

Conformation Mode Calculated Typical observed Calculated Typical observed

NCTE:Frequency "dues are given in an-'. Calculated values shown in parentheses were adjusted to equal the
correspondingobserved values by the choice of parameters.
SOURCE: Adapted from J. A. Schellman and C. Schellman, in The Proteins. 2d ed., vol. 2, ed. H. Newath (New York: 3
Academic Press, 1962), p. 1.
A schematic representation of the vibrational modes of the antiparallel P-sheet is shown
below.

I I
I
H
I
I
0
I Ilr-
N C C
c,
P I

Figunr 8-27
Schematic representation of the vibrational modes of the antipmallel j3 sheet. Arrows represent components
of transition moments in the plane of the paper: plus and minus signs represent out-of-plane
components. [AAer T. Miyazawa. J. Chem. Phys. 32: 1647 (1960).]