
28 Foundations
the particular value of r − r
e
. Thus for v = 0 quantum mechanics predicts that the most
probable internuclear separation is r
e
. This is counterintuitive when one thinks of a spring,
since a vibrating spring will spend most of its time in the region of the two turning points
and is moving at its fastest at the equilibrium position! Note also that there is some ‘leakage’
of the wavefunction outside of the harmonic oscillator potential well, a phenomenon that
is impossible in the classical case. This ‘leakage’ grows in importance as the vibrational
energy is increased.
For levels v = 1 and higher, nodes appear in the wavefunction and indeed the number
of nodes is equal to the vibrational quantum number. As v increases, the wavefunction
progressively heads towards behaviour that is anticipated classically, i.e. the most probable
internuclear separations shift towards the turning points, with the probability of finding the
molecule at the equilibrium separation becoming rather small.
5.1.4 The anharmonic oscillator
The justification for treating a vibrating diatomic molecule like a quantized vibrating
spring is that the harmonic oscillator model works rather well. Spectroscopic measurements
demonstrate that the separation between adjacent pairs of vibrational levels is indeed approx-
imately constant, as equation (5.6) predicts. However, if we look more closely at experi-
mental data, and if we think more clearly about the implications of the potential well shown
in Figure 5.1 we conclude that the harmonic oscillator model is only an approximation and
there are circumstances where its failure can be very serious.
The most obvious deficiency is that no allowance is made for the fact that any bond, or
for that matter any spring, will eventually break when sufficiently stretched. The harmonic
oscillator potential energy curve is infinitely deep, which would lead to the nonsensical
conclusion that a chemical bond is infinitely strong.
Qualitatively, the potential energy curve of a real diatomic molecule would be expected
to have the same shape as that shown in Figure 5.2.Todetermine the pattern of vibrational
energy levels for such an oscillator, which is now referred to as anharmonic because of
the asymmetry of the potential energy curve, a mathematical form for the potential energy
is needed, which can be substituted into the Schr¨odinger equation (5.4)inplace of the
harmonic potential (5.3). A number of different mathematical functions give rise to a curve
of similar shape to that shown in Figure 5.2 but the most widely used is the Morse potential
function,which has the form
V = D
e
{1 − exp[−a(r −r
e
)]}
2
(5.12)
where D
e
is the dissociation energy of the molecule, measured from the bottom of the
potential well. The quantity a is a constant that varies from one molecule to another (and
one electronic state to another), as does D
e
.
Solution of the Schr¨odinger equation with the potential in equation (5.12)gives
E
v
= hω
e
v +
1
2
− hω
e
x
e
v +
1
2
2
(5.13)