On Rate Constants: Simple Collision Theory,

Arrhenius Behavior, and Activated Complex

Theory

17th February 2010

0.1 Introduction
Up to now, we haven’t said much regarding the rate constant k. It should
be apparent from the discussions, however, that:

• k is constant at a specific temperature, T and pressure, P

• thus, k = k(T,P) (rate constant is temperature and pressure depen-

dent)

• bear in mind that the rate constant is independent of concentrations,

as the reaction rate, or velocity itself is treated explicity to be concen-
tration dependent

In the following, we will consider how the physical, microscopic detaails of

reactions can be reasoned to be embodied in the rate constant, k.

1 Simple Collision Theory

Let’s consider the following gas-phase elementary reaction:
A + B → Products

The reaction rate is straightforwardly:

1
 Rate = k[A]α [B]β
= k[A]1 [B]1 α=1β=1
NA NB
=k V = volume
V V
Recall previous discussions of the total collisional frequency for heteroge-
neous reactions:
s
8kT ∗
ZAB = σAB n nB ∗
πµ A

where the n∗A = NVA , nB ∗ = NVB are number of molecules/particles per unit
volume. We can see the following:
s
8kT NA NB
ZAB = σAB
πµ V V
| {z }
k
Here we see that the concepts of collisions from simple kinetic theory can be
fundamentally related to ideas of reactions, particularly when we consider
that elementary reactions (only for which we can write rate expressions
based on molecularity and order mapping) can be thought of proceeding
due to collisions (or interactions of some sort) of monomers (unimolecular),
dimers(bimolecular), trimers (trimolecular), etc.

If we are to naively say that reactions occur due only to collisions of parti-
cles (keep in mind the nature of the system – gas-phase,elementary reaction),
then we can at the zero’th order equate the maximum reaction rate to the
total collisional frequency for heterogeneous pairs:

Rmax = ZAB

s
8kT NA NB
k max [A]1 [B]1 = σAB
πµ V V
| {z }
k

s
max 8kT
k ≡ σAB
πµ

2
1.1 Simple Collision Theory: Caveats
Simple collision theory (SCT), we see from above, remarkably predicts an
expression for the microscopic rate constant that relates to the dimensions of
the reactive species, their mass, and temperature. The form we determined
above assumes two things:
• All collisions are of sufficient energy that chemical transformation can
occur

• Steric/orientational nature of collision is always correct/accommodating

Thus, we need to consider the effects of collision energy and collision
sterics and/or orientation in our discussion of simple collision theory
and its application to defining the reate constant, k.

• For energetic considerations, we can empirically add a factor to ac-

count for the probability of a collision having a sufficient energy,
vis-a-vis, Emin , for collision. When we multiply the total collisional
frequency by this probability, we can describe the fraction of collisions
that will energetically be able to progress from reactants to products.
The energy probability is take to be Boltzmann-like:

−Emin
P robability(Emin ) ∝ e RT

3
 • For steric/orientational nature of collision, we introduce a steric factor
(empirically), p
• p < 1 generally
Thus, we can write a more general expression for the collision theory based
reaction rate as:

−Emin
Rate = ZAB p e RT
s
8kT −Emin NA NB
= p σAB e RT
πµ V V

Thus we arrive at a corrected Simple collision theory expresion for the rate
constant:

s
8kT −Emin
kSCT = p σAB e RT
πµ

" s #
8kT Emin
ln(kSCT ) = ln p σAB −
πµ RT

Note the temperature dependence of the rate constant with SCT:

1
kSCT ∝ T 2 .

In general, we can write this expression for the rate constant as:

kgeneral = c T m e−Emin /RT (1)

2 Arrhenius Temperature Dependence

1
In general, experiments do not suggest a T 2 temperature dependence of the
rate constant. Moreover, experiments demonstrate for most reactions that
the temperature dependence of ln(k) is linear with T1 . Thus, Arrhenius pro-
posed the following relation between temperature and the rate constant:

−Eactivation
kArr = A e RT

(2)

4
Thus,

Eactivation 1
ln(kArr ) = ln(A) −
R T
(3)

The energy is the activation energy (as discussed above) and the A value is
a temperature-independent freqency factor, or pre-exponential factor.

Plotting ln(k) versus T −1 will yield a straight line with slope

equal to −Eactivation
R and y intercept of ln(A).

The pre-exponential factor in this case if independent of temperature, con-

trasted with the SCT result from above. Though many reactions (across the
spectrum of reaction orders, mechanisms, etc.) follow Arrhenius behavior,
there are exceptions (as always).

3 Activated Complex Theory; Transition State The-

ory
In this section, we will consider a further refinement of our formulation
of a reaction on the atomic level. Up to now, we have not entertained
the possibility of short-lived, highly unstable intermediates appear-
ing/being generated upon the initial ”collision” of reactive species. Eyring
and co-workers postulated the presence of these highly unstable, fleeting,
transition states, and furthermore suggested an equilibrium between this
transition state and the reactive species. The transition state is also consid-
ered an activated complex, hence the nomenclature Activated Complex
Theory (as well as Transition State Theory).

A further important aspect of invoking the activated complex is that for

a reaction such as AB + C → A + BC, the path the reaction follows (if
depicted on a three-dimensional potential energy surface) is the minimum
energy path. That is, the reaction will not follow along a reaction coor-
dinate that requires any higher energetic cost than is minimally necessary.
See Figure 36.19 in Engell and Reid for a represenation of a representative
3-D potential energy surface.

Assumptions of activated complex theory:

• Equilibrium exists between reactants and activated complex

5
 • Reaction coordinate can be mapped onto a single energetic degree of
freedom of the activated complex (i.e., a vibrational degree of freedom
corresponding to bond-stretching)

The kinetic mechanism incorporating the activated complex is now:

1 k
A+B −
)−
−*− AB †
k−1

k
AB † −→
2
P roducts

The differential rate expression for A becomes:

d[A]
= 0 = −k−1 [A][B] + k−1 [AB † ]
dt

k1 Kc†
[AB † ] = [A][B] = o [A][B]
k−1 c
Note the dependence of the equilibrium constant on the standard state con-
centration, co :

[AB † ]  
co AB † co
Kc† = [A] [B]
=
[A] [B]
co co

The rate of product, P, formation (equal to the reaction rate) is:

d[P ]
Rate = = k2 [AB † ]
dt
and since we know what [AB † ] is from equilibrium of the reactants with the
activated complex:

d[P ] k2 Kc†
Rate = = [A][B]
dt co
Now, we will consider how to refine our understanding of k 2 . Consider:

6
 • We have formulated our discussion with the assumption that there
is a single reaction coordinate mapped onto a vibrational degree of
freedom

• Let this vibrational degree of freedom be a weak bond

• Rate of product formation = Rate of reaction = Frequency of vibration

of weak bond (inverse seconds) (Maximum rate since we are taking
every initial motion along that vibration coordinate to lead to product
formation) (k2 = ν).

• If every κ of the vibrational cycles leads to product formation, then

k2 = κν

Thus, the reaction rate can be written as:

d[P ] κ ν Kc†
Rate = = [A][B]
dt co
Using elements of statistical mechanics (the details of which are outside
the current scope, but are covered in: Moore and Pearson, Kinetics and
Mechanism, John Wiley, 1981, pp. 159-186)

κ kB T Kc†
k2 =
h co
where the h in the denominator is Planck’s constant. Finally, if we recall
that we can relate the equilibrium constant to the free energy change from
reactant to activated complex,

∆G† = −RT ln(Kc† )

Thus, the rate constant becomes (taking κ = 1, which is a good approx-

imation for most situations (exceptions include surface reactions)):

kB T −∆G† /RT
k2 = e
hco
Since:

7
 ∆G† = ∆H † − T ∆S †

we finally obtain the Eyring Equation:

kB T −∆H † /RT ∆S † /R
k2 = e e
hco
In the most general sense:

kB T −∆H † /RT ∆S † /R
k(T ) = e e
hco
We see that activated complex theory allows us to incorporate free energy as
a metric for determining the rate constant. In this sense, entropic factors
come into play as well. In the next section, we will discuss the relation
between the parameters of Arrhenius theory and ∆S † and ∆H † .

4 Connection to Arrhenius Parameters

The quantities ∆H † and ∆S † are now related to the Arrhenius pre-exponential
factor and activation energy, Ea .

Table 1. Relation Between Arrhenius and Eyring Parameters.

Phase / Molecularity Activation Energy, E a Pre-Exponential, A

e kB T † /R
Solution/Bimolecular Ea = ∆H † + RT A= hco e∆S

e kB T † /R
Solution/Unimolecular Ea = ∆H † + RT A= h e∆S

e kB T † /R
Gas/Unimolecular Ea = ∆H † + RT A= h e∆S

e 2 kB T † /R
Gas/Bimolecular Ea = ∆H † + 2RT A= hco e∆S

e 3 kB T † /R
Gas/Trimolecular Ea = ∆H † + 3RT A= h(co )2
e∆S