Theories of Reaction Rates

Quantitative relationship between reaction rate constant and

temperature given by Arrhenius Equation:

(1)

where k is the rate constant, A is the frequency factor, and Ea is the

activation energy.

Eq. (1) indicates that the rate constant of a reaction is essentially the
product of an “attempt rate” or collision frequency, A, and the fraction of
molecules having an energy Ea or greater, which is given by the
Boltzmann factor, Ea/RT.
This energy requirement for an “effective” collision can be viewed as a
potential energy barrier that must be surmounted if a reaction is to
occur, and can be represented by the familiar energy profile of the
reaction:
Assuming that the (usually small) difference between U and H can be
ignored, ÄrUE . ÄrHE for most reactions, and, hence, E’a = Ea !ÄrHE,
where E’a is the activation energy for the reverse reaction. Thus,
because for an exothermic forward reaction, ÄrHE < 0 and E’a > Ea, the
reverse reaction may be many orders of magnitude slower than the
forward reaction. By contrast, for an endothermic reaction, ÄrHE > 0
and E’a < Ea, and the reverse reaction is faster than the forward one. If
both k and k’ can be represented by the Arrhenius Equation, their ratio
is given by

(2)

which represents the same temperature dependence that is shown by

the equilibrium constant in the van’t Hoff Equation:

(3)

as it must, according to the principle of detailed balance:

(4)
In addition, because

(5)

then

(6)

which implies that the ratio of “attempt rates” (A/A’) depends on the
relative entropies of the reactants and products. Thus, if ÄrSE < 0, for
example, then the reactants attempt to react at a lower frequency than
the products attempt to fall apart into reactants. Additionally, the
frequency factors, A and A’, must have a slight temperature
dependence, since ÄrSE does. Typically, however, the overriding
temperature dependence is that in the exponential Boltzmann factor.
 Molecular Theories of Elementary Reactions

Collision Theory

The so-called collision theory treats Ea as a potential energy barrier and

then develops a molecular model for the frequency factor A. According
to the kinetic theory of gases, the frequency of collisions between like
molecules in a gas is given by

(7)

where Zmol is the average number of collisions made by one X molecule

per unit time, N/V = [X] is the number of X molecules per unit volume
(i.e., the concentration), and ðd2 is the mutual molecular area of collision
for molecules having a diameter of d.

For the frequency of collisions between dissimilar reactant molecules A

and B, the average velocity must be replaced by the average relative
velocity between A and B, given by

(8)

where kb is Boltzmann’s constant, and ì is the reduced mass of A and B:

(9)

The total frequency of A-B collisions per unit volume is then

(10)

where dAB is the mean A-B collision diameter, given by dAB = (dA + dB)/2,
with dA and dB being the molecular diameters of A and B. (The factor
in Eq. (7) is included to avoid counting collisions of A and B
molecules twice because in A-A and B-B collisions the two colliding
molecules are indistinguishable. But this factor is not included in Eq.
(10) because in A-B collisions the two molecules are distinguishable.)
The quantity is called the collision cross section.
Thus, the overall reaction rate is the frequency of potentially reactive
collisions, ZAB, multiplied by the fraction of colliding molecules that have
sufficient energy to react, , resulting in the following expression
for the rate of an elementary bimolecular reaction:

(11)

Therefore, because the rate of a bimolecular reaction is Rate = k[A][B],

the bimolecular rate constant k and the frequency factor A can be
identified in Eq. (11) as:

(12)
and

(13)

Thus, A is predicted to increase with temperature as T1/2, which is a

relatively mild temperature dependence that is typically masked by the
exponential term. In addition, k also should increase with decreasing ì,
in agreement with Graham’s Law of effusion.

In essence, the frequency factor A represents an upper limit to the

magnitude of k, corresponding to Ea = 0, i.e., every collision results in
reaction. Depending on the particular reaction system, this upper limit
can vary over a sizable range, but is typically ~10-10 cm3/molecule/s, or
~1011 L/mol/s.

The Table below compares experimentally observed A factors (log10A)

with those predicted by collision theory (log10Act) from Eqs. (12) and (13)
for a variety of elementary bimolecular reactions. Most of the
experimentally measured A factors are considerably smaller than those
predicted by collision theory, with the disagreement expressed as the
ratio p = A/Act, where p is the so-called steric factor.
Because collision theory assumes that every reaction must occur via a
collision, then the experimental results indicate that even if a collision
occurs with sufficient energy to overcome the activation energy barrier,
it may not result in reaction. This failure to react can be attributed to
an orientation or steric effect, in which molecules may collide with
sufficient energy, but with an “ineffective” orientation:
Because collision theory gives equal weight to all collisions, including
ones that are ineffective, it thereby overestimates the attempt rate,
resulting in p < 1 and p << 1 in many cases. Additionally, the steric p
factor is not easily calculated, which results in collision theory being a
descriptive and conceptual model of chemical kinetic theory, rather than
a predictive one.

The Lindemann Mechanism. Because collision theory is an

inherently bimolecular model involving a second-order rate law, it
seemingly omits many reactions that are known to be first-order.
Reactions such as isomerizations and decompositions, which involve
only a single reactant molecule, had been assumed to be first-order.
However, in the 1920s Frederick Lindemann (and further developed by
Cyril Hinshelwood) proposed that even these types of reactions are
mediated by collisions.

Lindemann suggested that a generic first-order reaction, such as A 6 P,

actually involves an initial collisional activation step in which two
molecules of A collide with sufficient energy to cause one of them to
become internally (i.e., vibrationally) excited:

(14)

The excited A* molecule then reacts to form P in a unimolecular process.

A rate law for the Lindemann-Hinshelwood mechanism can be
developed by using the steady-state approximation for the concentration
of the intermediate A*:

(15)

Solving for [A*], (16)

Hence, using the expression for [A*] from Eq. (16), the rate of formation
of the product P is given by

(17)

Two limiting cases for Eq. (17) can be defined. If the rate of
deactivation of [A*] back to [A] greatly exceeds its rate of decomposition
to P (i.e., if k-1[A] >> k2), then
(18)

and the reaction appears to be unimolecular and first-order, even

though bimolecular collisions are necessary for reaction to occur. This
is the situation that is typically observed for most reactions when
pressures are a few Torr or greater. (k4 represents the limiting value of
the first-order rate constant at infinite pressure.)

On the other hand, if the pressure is sufficiently low, then the initial
collisional activation is the rate-limiting step, i.e., k2 >> k-1[A], and the
reaction appears to be second-order:
(19)

These two limiting cases can be combined by defining an “apparent”

first-order rate constant:
(20)

which becomes pressure-dependent at low pressures (low [A]). Thus,

the relative rate of reaction, (dP/dt)[A], is predicted to decrease at low
[A]. This decrease in the effective first-order rate constant at low
pressures is an important test of the Lindemann mechanism. The
Figure below shows data for the “unimolecular” decomposition of
cyclopropane to propylene over a wide range of pressures and illustrates
the decrease of k at low pressures.
Transition State Theory

Originally developed by Henry Eyring in 1935 as a model for

bimolecular reactions, Transition State Theory postulates that the
transitory species which is formed as a reacting system passes over the
activation barrier is a separate thermodynamic state, called the
activated complex or the transition state. For the bimolecular
reaction A + B 6 P, for example, the transition state is denoted by AB‡.
Because this state lies atop the crest of the activation barrier, it is in
metastable equilibrium, and a “mechanism” for its involvement in a
bimolecular reaction is:
(21)
If it is assumed that AB‡ is in equilibrium with A and B, i.e., that
decomposition of AB‡ is rate-limiting, then

(22)

where K‡ is the equilibrium constant for complex formation.

The rate of product formation is the rate at which AB‡ decays into
products, which is an inherently unimolecular process:
(23)

The basis of Eyring’s theory is that the activated complex decomposes

into products when a critical bond in AB‡ breaks. The force constant of
the dissolving bond approaches zero, and its energy quantization is
therefore relaxed and can assume the classical energy å = kBT.
According to Planck’s hypothesis, the frequency corresponding to this
energy is í = å/h = kBT. In the classical limit, this frequency becomes
the classical frequency of motion along the direction of the dissolving
bond. Thus, í can be identified as the unimolecular rate constant kc,
and the bimolecular rate constant k for the overall reaction A + B 6 P is
then (24)

According to classical thermodynamics, K‡ is also related to the Gibbs

energy change for a reaction and, hence to the enthalpy and entropy
changes by
!RTln K‡ = ÄGE ‡ = ÄHE ‡ ! TÄSE ‡ (25)

where K‡ has inverse concentration units, e.g., L/mol. Thus, Eq. (24) can
be written

(26)
Comparing Eq. (26) to Eq. (1), the Arrhenius activation energy, Ea,
defined as the slope of a lnk vs. 1/T plot, can be identified as

Ea = ÄEE ‡ = ÄHE ‡ ! (Än ‡ ! 1)RT (27)

where Än ‡ is the change in number of moles of gas in going from

reactants to the transition state. For a bimolecular reaction, such as
A + B 6 AB‡, Än ‡ = !1, while for unimolecular reactions, Än ‡ = 0, as it is
for all condensed-phase reactions. At normal temperatures, the
difference between ÄEE ‡ and ÄHE ‡ is only a few kJ/mol. Thus, the
Arrhenius frequency factor can be identified from Eq. (26) as

(28)

The collision theory model results in a single unknown quantity that

must be experimentally measured, viz., the Arrhenius activation energy,
Ea. Transition state theory, by contrast, results in two unknown
quantities, Ea (or ÄHE ‡) and ÄSE ‡. However, because the transition
state model is based on thermodynamics, it is possible to use statistical
thermodynamic methods to predict at least reasonable estimated values
for ÄSE ‡. Despite the approximations involved in postulating
vibrational frequencies for the AB‡ complexes, and, hence, their
partition functions, predictions by the transition state model are far
superior to those of collision theory.
Molecular Reaction Dynamics

Real molecular collisions in the gas phase are more complex than the
simple “billiard ball” model would suggest. Molecular attractions (van
der Waals forces) and repulsions (Pauli exclusion principle) play
important roles in governing trajectories of molecular encounters:
Atom-molecule reactions of the type A + BC 6 AB + C are often
elementary steps in the mechanisms of gas-phase reactions involving
simple diatomic molecules. In this type of reaction, one bond is broken,
and one is made, and the molecular mechanics can be modeled as
occurring in three steps:
1. A@@@@@@B—C
2. A@@@B@@@C
3. A—B@@@@@@C

A approaches BC; the new A-B bond begins to form, while the original
B-C bond begins to break, and finally the liberated C atom departs from
the newly formed AB molecule. The energetics of reactions such as this
are conveniently displayed on three-dimensional potential energy
surfaces, which depict the changes in both bond lengths on the two
horizontal axes and the potential energy on the vertical axis:
Examples of such an elementary reaction are
F + H2 6 HF + H
and H + F2 6 HF + F
which are steps in the mechanism of the bimolecular reaction

H2 + F2 6 2 HF
A similar reaction to the above is H2 + Br 6H + HBr, for which the
potential energy surface is shown below. Note that the energetics
(Ea = 84 kJ and ÄE rE= +63 kJ) for this closely-related reaction are quite
different from the corresponding values for the F + H2 reaction above,
for which Ea = 7.5 kJ, and ÄE rE= !134 kJ.