Coupled Oscillations

To obtain a mathematical description of coupled oscillations we start with a pair of simple

pendulums with the coupling is provided by a light horizontal spring that connects them, as
shown in figure.

The spring is at its unstretched length when the

two pendulums are at their equilibrium positions.
The mass and length of each pendulum are m and
l, respectively, and the spring constant is k.
Displacements of the two masses from their
equilibrium positions are xa and xb, respectively,
and now, we consider oscillations in the plane of
the page.

Two simple pendulums coupled together

by a light horizontal spring of spring
constant k. The displacements of the two
pendulum masses from their equilibrium
positions are xa and xb, respectively, and
these lie in the plane of the page.
Case (i). We first displace each mass in the same direction by an equal amount as shown
in Figure and then release them. Since the pendulums have the same period the spring
retains its unstretched length and so plays no role in the motion.

• The two pendulums might just as well be

unconnected as they both oscillate at the
frequency of a simple pendulum
𝑔/𝑙
• We can then write the displacements of the
two masses, respectively, as

𝑥𝑎 = 𝐴 𝑐𝑜𝑠𝜔1 𝑡

𝑥𝑏 = 𝐴 𝑐𝑜𝑠𝜔1 𝑡

where A is the initial displacement and

𝜔1 = 𝑔/𝑙 Two simple pendulums coupled together

by a light horizontal spring of spring
constant k. The displacements of the two
The phase angles are zero because the masses
pendulum masses from their equilibrium
start from rest. positions are xa and xb, respectively, and
these lie in the plane of the page.
The variations xa and xb with time are shown
in Figure. The masses oscillate in phase with
the same frequency and amplitude. This is
the first normal mode of oscillation.

Oscillations of the two masses in the first

normal mode. These oscillations have the same
frequency and amplitude and are in phase with
each other.
Case (ii). We now displace each mass by an equal amount but in opposite directions, as
shown in Figure, and then release them. As the two pendulums swing back and forth the
spring is alternately stretched and compressed and this exerts an additional restoring force
on the masses.
The symmetry of the arrangement tells us that
the motions of the masses will be mirror
images of each other, i.e. xa = −xb. The resultant
equation of motion of mass a is then

The first term on the right-hand side of this

equation is the usual restoring force term for a The second normal mode of oscillation of
the coupled system in which xa = −xb
simple pendulum with small amplitude
oscillations
The second term is the restoring force due to the spring extension of 2xa. Hence

𝑔 𝑘
Where 𝜔22 = ( 𝑙 + 2 𝑚). The action of the spring is to increase the restoring force
acting on each mass and this increases the frequency of oscillation, i.e. ω2 > ω1.

The solution of above Equation is

𝑥𝑎 = 𝐵 𝑐𝑜𝑠𝜔2 𝑡

where B is the initial displacement. Again the phase angle is zero because the mass
started from rest. Since xa = −xb,

𝑥𝑏 = −𝐵 𝑐𝑜𝑠𝜔2 𝑡
 The variations of 𝑥𝑎 and 𝑥𝑏 with time are shown in Figure. The masses oscillate with the
same frequency and amplitude but now they are 180° out of phase. We could write xb as
𝑥𝑏 = 𝐵 𝑐𝑜𝑠(𝜔2 𝑡 + 𝜋)

to emphasise this phase relationship. This is the second normal mode of oscillation. We
see that in each normal mode:

• Both the masses oscillate at the same

frequency.
• Each of the masses performs SHM with
constant amplitude.
• There is a well defined phase difference
between the two masses; either zero or π.
• Once started in a particular normal mode,
the system stays in that mode and does not
evolve into the other one.
• The importance of normal modes, as we
shall see, is that they are entirely
Oscillations of the two masses in the second
independent of each other. normal mode. These oscillations have the
same frequency and amplitude but are in anti-
phase, i.e. are 180° out of phase with each
other.
 SUPERPOSITION OF NORMAL MODES

• In general the motion of a coupled oscillator will be much more complicated than in
cases (i) and (ii) above. Those cases were special in that the motion was confined to a
single normal mode, i.e. either xa = xb or xa = − xb at all times.
• The general case is illustrated in Figure below which shows the displacements of the
two masses at some instant in time and xa ≠ ± xb.

• This gives a spring extension (xa − xb ) and

produces a tension T = k(xa − xb ) in the
spring. The directions of the spring force
acting on the masses are as indicated by the
arrow heads on the springs. The restoring
force on mass a is

and the restoring force on mass b is

General case for the superposition of

normal modes where xa ≠ ± xb
The resultant equations of motion are

And

These Equations contain both xa and xb. Thus they cannot be solved separately but
must be solved simultaneously. We can do this as follows. Adding them gives

It is striking that this is the equation of SHM where the variable is (xa+ xb). Moreover
the frequency of oscillation 𝑔/𝑙 is the frequency ω1 of the first normal mode.

Subtraction of the above Equations leads to

 This again is the equation of SHM but now in the variable (xa- xb). Moreover, the
oscillation frequency

𝑔
𝜔= + 2𝑘/𝑚
𝑙
is the same as the frequency ω2 of the second normal mode. We introduce the new
variables q1 and q2 where

Then

and

We have two independent oscillations in which each normal mode is represented by the
oscillation of a single variable: each of above Equations involves just one coordinate, q1 or
q2, and describes SHM, with frequencies ω1 and ω2, respectively. These equations do not
involve, for example, a product q1q2: there is no coupling between the two normal modes.
The general solutions of the above Equations can be written, respectively,

𝑞1 = 𝐶1 cos(𝜔1 𝑡 + 𝜑1 )
𝑞2 = 𝐶2 cos(𝜔2 𝑡 + 𝜑2 )
Here, C1 and C2 are amplitudes and φ1 and φ2 are phase angles.

• The variables q1 and q2 are called normal coordinates and ω1 and ω2 are called normal
frequencies.
• If q1 = 0 then xa = −xb at all times, and if q2 = 0 then xa = xb at all times. It is useful to
describe coupled motion in terms of the normal coordinates because the resulting
equations of motion depend on only one variable, either q1 or q2 , so that they can be
considered separately; changes in q1 do not affect q2 and vice versa.

We can express the displacements of the two masses in terms of the normal coordinates.

and

The motion of a coupled oscillator can be broken down into a combination of two
independent harmonic oscillations (normal modes).
• The variables of these harmonic motions are the normal coordinates. The above
equations demonstrate that any motion of the two masses, can be written as a
superposition of the two normal modes.
• It follows that there are just two normal modes for our system. The four constants
C1,C2, 𝜑1 and 𝜑1 are determined by the initial positions and velocities of the two
masses, i.e. at time t = 0. If the two masses are released from rest at t = 0, the
appropriate solutions for q1 and q2, obtained by taking 𝜑1 = 𝜑1 = 0 are

and

𝑞1 = 𝐶1 𝑐𝑜𝑠𝜔1 𝑡

𝑞2 = 𝐶2 𝑐𝑜𝑠𝜔2 𝑡
The independence of the two normal modes is clearly demonstrated if we write down the
energy of the system.
E = K.E + P.E

In terms of the position coordinates xa and xb the energy is given by

The first two terms in this expression are the kinetic energies of the two masses, the third
term is their potential energies due to gravity and the last term is the energy stored in the
spring. Expressed in terms of the normal coordinates q1 and q2 , the energy E becomes

This equation represents the energy of two independent simple harmonic oscillators with
𝑔
frequencies 𝜔1 = 𝑔/𝑙 and 𝜔2 = + 2𝑘/𝑚.
𝑙
• Each of the expressions in square brackets in this equation contains only one of the
normal coordinates and represents the energy of a single isolated harmonic oscillator.
There are no ‘cross terms’ involving both q1 and q2, which would indicate coupling
between them.
• This is in contrast to the energy expressed in terms of the position coordinates xa and
xb where the last term, involving (xa − xb), represents a coupling between the two
masses.
Normal Coordinates, Degrees of Freedom and Normal Modes of Vibration

The significance of choosing q1 and q2 to describe the motion is that these

parameters give a very simple illustration of normal coordinates.

• Normal coordinates are coordinates in which the equations of motion take the
form of a set of linear differential equations with constant coefficients in which
each equation contains only one dependent variable (our simple harmonic
equations in q1 and q2 ).

• A vibration involving only one dependent variable q1 (or q2) is called a normal
mode of vibration and has its own normal frequency. In such a normal mode all
components of the system oscillate with the same normal frequency.

• The energy of a coupled system when the q1 and q2 modes are both vibrating
would then be expressed in terms of the squares of the velocities and
displacements of q1 and q2.
• The importance of the normal modes of vibration is that they are entirely
independent of each other. The energy associated with a normal mode is never
exchanged with another mode; this is why we can add the energies of the
separate modes to give the total energy. If only one mode vibrates the second
mode of our system will always be at rest, acquiring no energy from the
vibrating mode.
• Each independent way by which a system may acquire energy is called a degree
of freedom to which is assigned its own particular normal coordinate. The
number of such different ways in which the system can take up energy defines
its number of degrees of freedom and its number of normal coordinates. Each
harmonic oscillator has two degrees of freedom, it may take up both potential
energy (normal coordinate q1) and kinetic energy (normal coordinate 𝑞1).
Example
Consider the system of two identical simple pendulums connected by a light horizontal
spring. Deduce expressions for the displacement of the two masses in terms of the normal
modes of the system for the following sets of initial conditions, (at t = 0). In all cases the
masses are released from rest. (i) xa = A, xb = A, (ii) xa = A, xb = −A and (iii) xa = A, xb = 0.
Again we have a high frequency oscillation modulated by a low frequency term. Both cosine
terms in the expression for xb are exactly π/2 out of phase with respect to the corresponding
terms for xa. The variations of xa and xb with time are plotted in Figure. These results explain
the behaviour of the two coupled pendulums, where one pendulum was given an initial
displacement and the other was initially at its equilibrium position. The important point in
all of these examples, with different initial conditions, is that the subsequent motion is
always a superposition of the normal modes.

Figure. Oscillations of the coupled pendulums which, occur when one mass was
initially (t = 0) at xa = A and the other at xb = 0.
 Oscillating Masses Coupled by Springs

• Lets consider the case of oscillating

masses coupled together by springs.
Figure (a) shows two identical but
independent mass-spring oscillators
with mass m and spring constant k Figure a: Two uncoupled mass-spring oscillators.
attached to two rigid walls.
• The two oscillators are coupled
together by a third spring also of
spring constant k as shown in Figure
(b).
• This third spring provides the
coupling so that the motion of one
mass influences the motion of the Figure: Two mass-spring oscillators coupled together by
other. This coupled system has two a third spring. The arrows indicate the directions of the
displacements of the two masses expected in the two
normal modes of oscillation. normal modes.
• We wish to determine the two
frequencies at which the system will
oscillate, i.e. the normal frequencies
and the relative displacements of the Figure a: Two uncoupled mass-spring oscillators.
masses in the two normal modes.
• We could exploit the symmetry of the
system to spot the two normal modes
similar to that for the coupled
pendulums.
• Our physical intuition would suggest
that the normal modes would be (i)
where both masses move in the same Figure: Two mass-spring oscillators coupled together by
direction and (ii) where they move in a third spring. The arrows indicate the directions of the
displacements of the two masses expected in the two
opposite directions. normal modes.
• These two modes are indicated by the
arrows in Figure. We might also expect
that mode (ii) would have the higher
frequency of oscillation since all three
springs are having an effect rather than
just two as in mode (i).
• Instead of spotting the normal modes we
adopt a more general approach where we
make use of the characteristics of normal
Figure: Two mass-spring oscillators coupled together by
modes, namely that in a normal mode all a third spring. The arrows indicate the directions of the
of the masses oscillate at the same displacements of the two masses expected in the two
frequency and each mass performs SHM normal modes.
with constant amplitude.
• For the sake of simplicity we will assume
that the two masses are initially at rest,
i.e. they have zero velocity at t = 0.
• Figure shows the two masses a and b displaced by arbitrary values xa and xb,
respectively, from their equilibrium positions at some instant in time.
• In order to see more easily the
directions of the forces acting on
the masses we let xb > xa. The
left-hand spring is extended by
xa, the middle spring is stretched
by (xb − xa) and the right-hand
spring is compressed by xb.
• The directions of the resultant
forces on the masses are shown
by the directions of the arrow
heads. Figure: Two mass-spring oscillators coupled together by
• To obtain the equation of motion a third spring. The arrows indicate the directions of the
for each mass we need to displacements of the two masses expected in the two
normal modes.
consider only the forces exerted
by the springs on either side of
the mass. The resultant equations
of motion are
 (1)

(2)

We are looking for normal mode solutions of these equations, where both masses oscillate
at the same frequency ω, i.e. solutions of the form xa = A cosωt and xb = B cosωt.
Substituting for xa in equation (1) yields

(3)

(2)
 (4)

(4)
(5)

(5)

(6)

This is a quadratic equation in ω2 which is seen at once to have the solutions

(2k − mω2) = ±k, i.e. ω2 = k/m or 3k/m. These are the two normal frequencies
of the coupled system. Putting ω2 = k/m in Equation (3) gives A = B. This is
the first normal mode in which the two masses move in the same direction as
each other and with the same amplitude.
 𝑥𝑎 = 𝐴 𝑐𝑜𝑠𝜔1 𝑡 , 𝑥𝑏 = 𝐴 𝑐𝑜𝑠𝜔1 𝑡 (7)
Where 𝜔12 = 𝑘/𝑚

Putting ω2 = 3k/m in Equation (3) gives A = −B. This is the second normal mode
where the minus sign tells us that the masses move in opposite directions. Thus

𝑥𝑎 = 𝐴 𝑐𝑜𝑠𝜔2 𝑡 , 𝑥𝑏 = −𝐴 𝑐𝑜𝑠𝜔2 𝑡 (8)

Where 𝜔22 = 3𝑘/𝑚

All of these results are in agreement with our physical intuition.

Since most coupled oscillators do not have a symmetry that allows us to spot the
normal modes, the approach described here is normally essential. As usual the general
motion will be a superposition of the two normal modes, i.e.

𝑥𝑎 = 𝐶1 𝑐𝑜𝑠𝜔1 𝑡 + 𝐶2 𝑐𝑜𝑠𝜔2 𝑡
and

𝑥𝑏 = 𝐶1 𝑐𝑜𝑠𝜔1 𝑡 − 𝐶2 𝑐𝑜𝑠𝜔2 𝑡

If the masses did not have zero velocity at t = 0, we would also need to include phase
angles.
Example
Figure shows two equal masses of mass m suspended from two identical springs of spring
constant k. Determine the normal frequencies of this system for vertical oscillations and the
ratios of the amplitudes of oscillation of the masses at these frequencies.

Two equal masses m suspended from two identical springs of spring constant k. The displacements of
the two masses from their equilibrium positions are xa and xb respectively, measured in the downward
direction.
Solution

Let xa and xb be arbitrary displacements of the masses from their respective equilibrium
positions and let xb be greater than xa. Then the extensions of the upper and lower springs
are xa and (xb − xa), respectively, and the directions of the forces acting upon the two
masses are as indicated by the arrow heads. The resultant equations of motion are

𝑑 2 𝑥𝑎
𝑚 = −𝑘𝑥𝑎 + 𝑘 𝑥𝑏 − 𝑥𝑎 = 𝑘 𝑥𝑏 − 2𝑥𝑎
𝑑𝑡 2
And
𝑑 2 𝑥𝑏
𝑚 = −𝑘 𝑥𝑏 − 𝑥𝑎
𝑑𝑡 2

This time we try complex solutions of the form, 𝑥𝑎 = 𝐴 𝑒 𝑖𝜔𝑡 and 𝑥𝑎 = 𝐵 𝑒 𝑖𝜔𝑡 .
Substituting for xa and xb into the equations of motion and dividing through by 𝑒 𝑖𝜔𝑡 leads
to

A (2k –mω2) = Bk (i)

Ak = B (k –mω2) (ii)
The above equations leads to the quadratic equation (mω2)2 − 3kmω2 + k2 = 0, which
has the solutions ω2 = (k/2m)(3 ± √5), giving the two normal frequencies. Substituting
for ω2 = (k/2m)(3 − √5) in Equation (1) gives A/B = 1/2(√ 5 − 1), while substituting
for ω2 = (k/2m)(3 + √5) gives A/B = −1/2( √ 5 + 1) where the minus sign indicates that
the masses move in opposite directions, i.e. in anti-phase
A powerful way to handle the simultaneous equations that arise for coupled oscillators is
to use a matrix representation. Equation (i,ii) can be written, respectively, as

2𝑘 𝑘
𝐴− 𝐵 = 𝜔2 𝐴
𝑚 𝑚

and

𝑘 𝑘
− 𝐴+ 𝐵 = 𝜔2 𝐵
𝑚 𝑚

2𝑘 𝑘
, −𝑚 𝐴
𝑚
= 𝜔 2 𝐴 (iii)
𝑘 𝑘 𝐵 𝐵
−𝑚, 𝑚

This is an eigenvalue equation. The solutions of this equation for ω2 are called the
eigenvalues. The column vector with components A and B is an eigenvector of the
matrix. We can rewrite Equation (iii) in the following form
We can rewrite Equation (iii) in the following form

2𝑘 𝑘
− 𝜔2 −
𝑚 𝑚 𝐴
=0
𝑘 𝑘 𝐵
− − 𝜔2
𝑚 𝑚

This equation has non-zero solutions if and only if the determinant vanishes, i.e. if
2
2𝑘 𝑘 𝑘
− 𝜔2 − 𝜔2 − =0
𝑚 𝑚 𝑚

giving m2ω4 − 3kmω2 + k2 = 0 and the solutions ω2 = (k/2m)(3 ± √ 5) as before.

Substituting for these solutions in Equation (iii) yields the two values of A/B. The power
of this approach is not obvious for the case of two coupled oscillators but it quickly
becomes apparent when more than two are involved.
 We have discussed the example of two masses connected by springs where
the masses oscillate in one dimension, i.e. along the x-axis.
 We found that this system has two normal modes of oscillation and that each
mode has an associated normal coordinate q and normal angular frequency ω.
 These results can be generalised to N masses interconnected by springs and
moving in three dimensions. As for the case of two masses the N masses do
not move independently. When one mass is set oscillating the other masses
will feel the disturbance and will start to oscillate. For N coupled masses
there are 3N normal modes of oscillation where the factor of 3 corresponds to
the three perpendicular directions along which each mass can move. Again
each normal mode has a normal coordinate and normal frequency, so that we
have normal coordinates q1, q2, ... ,q3N with corresponding normal
frequencies ω1, ω2, ... ,ω3N . For each normal mode we have independent
SHM in the coordinate q with frequency ω.
 A good example of this is provided by a crystal lattice.

 The Einstein’s model assumes that the atoms vibrate totally independently of
each other about fixed lattice sites. In reality, they do not because the atoms
are coupled together.
 A macroscopic mechanical analogue of
a crystal lattice would consist of billiard
balls connected together with identical
springs.
 If one ball is set vibrating, say the one
labelled A in Figure, a disturbance will
propagate throughout the whole system until
all the balls are vibrating.
 Einstein’s theory can be improved by
describing the N atoms in a crystal in terms
of the 3N normal modes of vibration of the
whole crystal, each with its own Two-dimensional analogue of a crystal
characteristic angular frequency ω1, ω2, ... lattice, consisting of billiard balls connected
,ω3N . by springs.

 In terms of these normal modes, the lattice vibrations are equivalent to 3N

independent harmonic oscillators with these angular frequencies.
 Forced oscillations of coupled oscillators
 The amplitude of oscillation of a harmonic oscillator becomes very large when a
periodic driving force is applied at its natural frequency of oscillation.
 For the case of two oscillators coupled together we may expect similar behaviour.
 Now, however, there are two
natural frequencies corresponding
to the two normal frequencies.
Thus we may expect that the
system will exhibit large
amplitude oscillations when the
driving frequency is close to either
of these two normal frequencies.
 We can explore forced oscillations
by considering the arrangement of
two masses connected by springs
as shown in Figure.
 The end s of one of the outer springs is moved harmonically as ξ = a cosωt. The
displacements ξ, xa and xb of the masses from equilibrium are shown in Figure at
some instant of time.
 The resulting equation of motion for mass a is
𝑑2 𝑥𝑎
m 𝑑𝑡 2 = -k(𝑥𝑎 − 𝜉) + 𝑘 (𝑥𝑏 − 𝑥𝑎 ) 4.30
giving
𝑑 2 𝑥𝑎 2𝑘 𝑘 𝐹𝑜
+ 𝑥 − 𝑥 = ω𝑡, 4.31
𝑑𝑡 2 𝑚 𝑎 𝑚 𝑏 𝑚
Where 𝐹𝑜 = 𝑘𝑎. Similarly, the equation of motion for mass b is
𝑑 2 𝑥𝑏 𝑘 2𝑘
− 𝑥𝑎 + 𝑥 =0 4.32
𝑑𝑡 2 𝑚 𝑚 𝑏
We can solve these two simultaneous equations by, respectively, adding and subtracting
them. Thus
𝑑 2 (𝑥𝑎 + 𝑥𝑏 ) 𝑘 𝐹𝑜 4.33
2 + 𝑥𝑎 + 𝑥𝑏 = 𝑐𝑜𝑠𝜔𝑡
𝑑𝑡 𝑚 𝑚

and
𝑑 2 (𝑥𝑎 − 𝑥𝑏 ) 3𝑘 𝐹𝑜
+ 𝑥 − 𝑥 = 𝑐𝑜𝑠𝜔𝑡. 4.34
𝑑𝑡 2 𝑚 𝑎 𝑏
𝑚
We now change variables of the normal coordinates

𝑞1 = 𝑥𝑎 + 𝑥𝑏 𝑎𝑛𝑑 𝑞2 = (𝑥𝑎 − 𝑥𝑏 ) 4.35

 Giving
𝑑 2 𝑞1 𝑘 𝐹0
+ 𝑞 = cos 𝜔𝑡 4.36
𝑑𝑡 2 𝑚 1 𝑚

and
𝑑 2 𝑞2 3𝑘 𝐹0
+ 𝑞 = cos 𝜔𝑡
𝑑𝑡 2 𝑚 2 𝑚 4.37

This is a striking result and illustrates the power and simplicity of describing the coupled
motion in terms of the normal coordinates. For each of the independent coordinates q1
and q2 we have the equation for forced oscillations of a simple harmonic oscillator. We
can describe the steady-state solutions by the equations q1 = C1 cos ωt and q2 = C2 cos
ωt, where
𝐹𝑜 /𝑚
𝐶1 = 4.38
and (𝜔12 − 𝜔 2 )

𝐹𝑜 /𝑚
𝐶2 = 4.39
(𝜔22 − 𝜔 2 )

where 𝜔12 = 𝑘/𝑚 and 𝜔22 = 3𝑘/𝑚

 The maximum values of C1 and C2 given by these equations are infinitely
large when ω = ω1 and ω = ω2, respectively, so that the amplitudes of
oscillation would become infinite if the system were driven at one of its
normal frequencies.
 This is, of course, because we have neglected damping that would limit their
values in real situations.
 Nevertheless we can conclude that a coupled oscillator will oscillate with
large amplitude when it is driven at either of its normal frequencies. At other
driving frequencies the masses will oscillate at the driving frequency but with
much smaller amplitude.
From Equation (4.35) we have

1 1
𝑥𝑎 = 𝑞1 + 𝑞2 = 𝐶 + 𝐶2 cos 𝜔𝑡
2 2 1
and
1 1
𝑥𝑏 = 𝑞1 − 𝑞2 = 𝐶 − 𝐶2 cos 𝜔𝑡
2 2 1

It follows from Equations (4.38) and (4.39) that when the driving frequency ω is
near the first normal frequency ω1 = √k/m, we have |C1| >>|C2|, and xa ≈ xb, i.e.
the two masses oscillate in phase. When the driving frequency ω is near the
second normal frequency ω2 = √3k/m, one similarly obtains xa ≈ −xb, i.e. the two
masses oscillate in anti-phase.
Since a coupled system oscillates with large amplitude when driven at one of its
normal frequencies this provides a way of determining these frequencies
experimentally. A good example of this is provided by the vibrations of molecules
that contain more than two atoms.
For example, the molecule carbon dioxide
(CO2) can be modelled by three masses
connected by two springs in a linear
configuration.
The central mass represents the carbon atom
and the other two masses represent the oxygen
atoms while the springs represent the molecular
bonds.
This system has two normal modes of vibration
for displacements along the line connecting the
masses. These are called the symmetric stretch
mode and the asymmetric stretch mode as
illustrated in Figure (a) and (b), respectively.

 In the symmetric stretch mode, the central Figure: A model of the normal modes of
mass remains fixed in position while the two vibration of the CO2 molecule: (a) the
outer masses vibrate against it. symmetric stretch mode; (b) the
asymmetric stretch mode; and (c) the
 In the asymmetric stretch mode the two bending mode.
outer masses move in the same direction and
maintain the same distance apart.
When the frequency of the radiation matches a normal frequency, the radiation is
strongly absorbed by the molecules. The frequencies at which this absorption
occurs give directly the normal mode frequencies of the molecule. The measured
values of the frequency ν for the symmetric stretch and the asymmetric stretch
modes of the CO2 molecule are 4.0 × 1013 s−1 and 7.0 × 1013 s−1, respectively. The
CO2 molecule also has a bending mode of vibration as illustrated in Figure 4.16(c).
The frequency of this mode is 2.0 × 1013 s−1.

Figure A model of the normal modes of vibration of the CO2 molecule: (a) the symmetric stretch mode;
(b) the asymmetric stretch mode; and (c) the bending mode.
This bending motion can occur in two orthogonal planes and since these have the
same frequency of vibration they are said to be degenerate in frequency. These
frequencies lie in the far infrared region of the electromagnetic spectrum, with
corresponding wavelengths of ∼10 µm

Vibrations of CO2 molecules and some other molecules in the Earth’s atmosphere
play a key role in global warming because they strongly absorb radiation in the far
infrared. The surface temperature of the Sun is 5800 K and the radiation emitted
by the Sun peaks at about 500 nm. However, the surface of the Earth is at a much
lower temperature, ∼300 K, and its radiation peaks at ∼10 µm. The Earth’s
atmosphere is largely transparent at visible and near infrared wavelengths and the
Sun’s radiation passes through. However, the global-warming molecules absorb
the Earth’s far infrared radiation and act to trap its energy. This effect leads to an
increase in the surface temperature of the Earth