This course material is supported by the Higher Education Restructuring Fund

allocated to ELTE by the Hungarian Government

Chemical Kinetics
for Beginners

Ernő Keszei

Eötvös Loránd University, Institute of Chemistry

Budapest, 2016
Contents

Contents 3
1. Introduction 4
2. Formal kinetic description of elementary reactions 7
2.1. Solution of rate equations of integer order reactions 9
2.1.1. Zeroth-order reactions 13
2.1.2. First-order reactions 14
2.1.3. Second-order reactions 16
2.1.4. Third-order ractions 20
2.2. Generalisation and extension of the order of reaction; pseudo-order 23

Suggested literature 26

3
Chapter 1

1. Introduction

Reaction kinetics is the branch of physical chemistry which deals with the temporal evolution
of chemical reactions. It provides explanation as to why some reactions do not take place, although
their products would be much more stable thermodynamically than the reactants are; it accounts for
the fact that some reactions go very fast while some other rather slowly. Furthermore, it also
provides several useful tools to calculate the actual rate of reactions within a wide variety of
circumstances, including e. g. the dependence of the rate on temperature, pressure, the solvent
applied, etc.
The most important goal of kinetic research is the identification and characterisation of
elementary molecular events which make possible the transformation of reactants into products. At
the time of the beginning of reaction kinetic studies – in the second half of the 18th century when
mechanical models dominated natural sciences – the scheme of these simple molecular events that
constitute the overall reaction has been called as the mechanism of the reaction, which term
survived and is still widely used. This means that the elementary steps of the reaction – the most
simple events when typically two molecules directly encounter – are identified, and the complex
reaction mechanism is constructed from these elementary steps. Most real-life chemical reactions
comprise quite a number of such elementary steps; the number of them e. g. in gas reactions can be
as much as a few hundreds. The overall rate of these reactions can successfully be calculated over a
wide range of circumstances if we can calculate the rate of all the constituent elementary reactions
within the given conditions, and also know the way of their connection within the reaction
mechanism. This is the reason why the theory of elementary reactions has a paramount importance
in chemical kinetics.
However, to explore the precise mechanism of composite reactions is not an easy task. To be
able to construct a reliable mechanism, we have to identify all the components which take part in
the process of the reaction, even if they are short-lived and have rather low concentration. We also
have to keep track of the temporal evolution of possibly all these components, although this is not
always possible. In many reactions, there are so called intermediates (substances formed from the
reactants and readily removed in consecutive reactions leading to the products) that are rather
short-lived and present only in very low concentrations, thus they easily remain unnoticed for the
experimentalist. After having traced the temporal evolution of as many components of the reacting
mixture as possible, the task of the kineticist is to construct a suitable mechanism that can explain
all the concentration changes as a function of time. Consequently, theories of composite reactions
also constitute an important topic of reaction kinetics.

4
 Introduction

To successfully model reactions the way explained above, we need an unambiguous definition
of the rate of change of the amount of components taking part in the reaction. There exists an
IUPAC recommendation for the definition of the rate of reaction. For the sake of this definition, we
shall write stoichiometric equations in a special form, so that the equation is set equal to zero. The
advantage of this form is that reactants and products (species on the left-hand side and on the right-
hand side in the more common equation) can be treated the same way, thus simplifying the
mathematical treatment. The general stoichiometric equation of this form can be written as:
R

∑v A = 0
i =1
i i (1.1)

The symbol Ai in this equation denotes the stoichiometric formula of the i-th species, and νi (lower
case Greek “nu”) is the stoichiometric number of this species. The index i runs over all the reacting
species whose number is R. (Components that do not react – e. g. an inert solvent – should have a
zero stoichiometric number, thus it is superfluous to include them in the sum.) As an example, let us
write one of the possible equations of the formation of water:

– 1 H2 – ½ O2 + 1 H2O = 0 (1.2)

In this equation, A1 = H2 , A2 = O2 , A3 = H2O , ν1 = – 1, ν2 = – ½ and ν3 = 1. However, we are too

much used to write stoichiometric equations in the usual left side – right side form. Thus we usually
write the equation itself in the traditional form

H2 + ½ O2 = H2O , (1.3)

but we consider the stoichiometric number νi of the reactants (left side) to be negative, while those
of the products (right side) to be positive. In the rest of this text, we always interpret stoichiometric
equations this way.
The rate of a particular reaction should be defined naturally in such a way that it should be
independent from the choice of the component taking part in the reaction whose amount would be
used for the temporal evolution. To formulate such definition, let us introduce the extent of reaction
ξ also used in thermodynamics, but for use in the definition of the rate of reaction, it is sufficient to
specify its change as

 =


. (1.4)

The definition of the rate of the reaction which is written in the above explained general form (1.1)
can then be written as:

= = 

 
 
(1.5)


5
Chapter 1

According to this, the SI unit of the rate of reaction is mol (stoichiometric equation) / s. In the
chemical praxis, concentrations are much more convenient to measure than amounts of substances.
Taking this into account, let us calculate from the above definition the rate of change of molar
concentration of a species. Using the definition of the molar concentration as ni /V, we get

=  +  
 ()   
    
(1.6)

This equation reveals that the rate of change of the molar concentration
 ⁄
 also depends
on the rate of change of volume. If the volume does not change during the reaction, the rate of
reaction can be obtained by multiplying
 ⁄
 by the volume of the reacting system, and dividing
it by the stoichiometric number. Accordingly, we can state that in case of reactions at constant
volume, the change of molar concentration is identical to the stoichiometric number times the rate
of reaction divided by the volume:

=  =
    

   
(1.7)
Within chemical kinetic context, the quantity
 ⁄
 is usually called simply as the rate of
reaction. However, we should be aware that this quantity is proportional only to the rate of reaction
in case of constant volume reactions, when the proportionality constant is the volume V divided by
the stoichiometric number νi . In condensed phase reactions (e. g. in solutions) this is typically true
to a good approximation, thus the proportionality can be assumed. Further on in this text, we also
consider
 ⁄
 as the rate of the reaction. However, in gas phase reactions, if there is a change in
number of moles during the reaction, this proportionality does not hold and the change of volume
should also be taken into account. It is worth mentioning also that the latest recommendations of
IUPAC suggest to use the term “rate of conversion” for
⁄
 and the „rate of reaction” for 
 

.


However, naming of the terms does not change their relation as explained above.
As we have seen, the definition of the reaction rate (in both versions) refers to a particular
stoichiometric equation as it contains the corresponding stoichiometric numbers. It is worth
mentioning that even this definition is only valid if during the (composite) reaction, there won’t be
any accumulation of an intermediate, and no other “by-products” are formed beside those included
in the stoichiometric equation. (This is never the case for elementary reactions where this definition
can always be used.) Thus, in case of composite reactions it is more convenient to refer to the
change of the amount (or concentration) of a component, or equivalently to the rate of consumption
for reactants and the rate of formation for products.

6
 Formal kinetic description of elementary reactions

2. Formal kinetic description of elementary reactions

Theories of elementary reactions suggest that the traditional form of kinetic equations
describing the rate of reactions – dating back as early as the middle of the 19th century – is always
valid for elementary reactions, i. e. for the simplest molecular events comprising only a few
molecules and proceeding by the formation of only one transition state between the reactants and
the products. Thus, if the elementary reaction is the decomposition of an (non-thermally) activated
species, then the rate equation can be written as  =   , where R is the reaction rate, NA the
number density (concentration) of the reacting species A. Similarly, for the reaction of two
molecules A and B we can write the rate equation as  =    . The only other – much less
frequent – possibility is the reaction of three molecules (A, B and C) after their encounter, when the
rate equation has the form =     . We can generalise this result by writing one single
equation. Replacing the number density by the more common molar concentration and writing the
reaction rate also in terms of this concentration, the rate equation has the form

− =  ∏!  ,
 
 
(2.1)

where " is the stoichiometric number of the i-th component in the stoichiometric equation, and ci is
its actual molar concentration. The rate coefficient k in this equation is somewhat different from the
one in the rate equations containing number densities N (usually in molecules/cm3 unit). Let us
ignore here to use another notation for the rate coefficient when the concentration has the unit
mol/dm3, and interpret it further on as referring to the molar concentration. (The relation between
the two contains the Avogadro constant and 1000 to convert cm3 to dm3.) The negative sign of the
time derivative reflects the fact that reactants disappear in the course of the reaction. (If we consider
the stoichiometric number as being negative for the reactants, this negative sign can be dropped.)
The quantity denoted by n is of course the number of molecules participating in the reaction.
Rate equations having the form of Eq. (2.1) are called mass action kinetic equations. If we
consider the stoichiometry of the reaction, we can readily realise that the concentration of the n
different reactants do not change independently – which will help a lot to solve the kinetic equation.
This form of the kinetic equation is typically classified according to the number of molecules

7
Chapter 2

participating in the molecular event of the reaction. Let us write the three stoichiometric and rate
equations in the form of Eq. (2.1) for the actual cases of one, two and three molecules:

Reaction: A → products (2.2)

− = 


Rate equation: (2.3)

Reaction: A + B → products (2.4)

− =− =  
. /
 
Rate equation: (2.5)

Reaction: A + B + C → products (2.6)

− =− =− =   
. / 1
  
Rate equation: (2.7)

Observing these equations we can see the coupling between concentration changes due to the
stoichiometry, and that derivatives at the left side of the rate equations are expressed at the right
side by a product, where the variables to be multiplied are not independent. Equations having this
form are called first order homogeneous ordinary differential equations of degree n. It is ordinary,
as it contains the derivative of a function () of one single variable only. It is first order, as the
only derivative is a first derivative. It is homogeneous of degree n, as there is a single function in it
(apart from the derivative), which is the product of n variables; thus a polynomial of n-th degree.
However, in chemistry, we call the reaction characterised by these equations as n-th order reaction.
(This name dates back to ancient times when the degree of polynomials in mathematics was also
called order. This term is no more used for the degree of polynomials.) According to this naming
tradition, the three reactions listed above are first order, second order and third order reactions,
respectively.
The examples discussed above were related to molecular events happening via formation of
one transition state only from the reactants and its consecutive decomposition into products. Such
events are called elementary reactions. Their rate equation always follows the mass action kinetic
law (2.1), and the number of molecules taking part in the reaction is called their molecularity –
which, in the case of the elementary reactions – is identical to their order. From the point of view of
molecularity, the first reaction is a unimolecular, the second one a bimolecular, and the third one a
termolecular reaction. (The latter is sometimes also called a trimolecular reaction, but we shall not
use this term in this text.) It is worth to emphasise that the order and molecularity of the reaction are
not identical categories. We shall later deal with composite reactions whose rate equation can
sometimes have the form of an integer order, but molecularity is not meaningful for composite

8
 Formal kinetic description of elementary reactions

reactions which consist of several elementary reactions. We can also find composite reactions
which have an order that is not an integer number, while molecularity always refers to integers.
It is worth mentioning why termolecular reactions are not frequent. The probability of
simultaneous encounter (or, in gas phase, of simultaneous collision) of three molecules is much less
frequent than that of two molecules. For this reason, the termolecular process is extremely slow, and
in many cases, there is an equivalent route with a complex mechanism which is faster, and largely
masks an eventual termolecular process.

2.1. Solution of rate equations of integer order reactions

As we have seen, rate equations of elementary reactions are always of integer order; thus, the
title of this section might well have been also “solution of rate equations of elementary reactions”.
However, this would exclude the possibility for composite reactions to be of integer order, though
there are plenty of examples for this behaviour. One of the possible forms of a general integer order
reaction is the case when the initial concentrations of the reactants are identical. (This is necessarily
the case if, for example, two reacting molecules are identical.) In this case, the relevant rate
equation has the form

− =   ,


(2.8)

if the concentration c refers to a component with unit stoichiometric number. In mathematics, this
kind of differential equation is called separable as the dependent variable c and the independent
variable t can be separated to the opposite sides of the equation, then both sides can readily be
integrated to get an implicit function of the independent variable. The above differential equation
can be written in the separated form

−  2
 = 
 .

(2.9)

Integrating both sides we can get the solution as

− 3  2
 = 3 
 .

(2.10)

Evaluating the integrals, we can determine the primitive functions up to an undetermined additive
constant. (It is enough to write one single constant; the two constants arising from the two
integrations can be combined.) To find the primitive function, we can realise that the integrands at
both sides are simple power functions. On the right side – after factoring out k from the integration
– we get 1 =  5 , while on the left side, we get =  6 . Substituting their primitive functions in

 2

place of the integrals, the general solution is

− =  + 9 ,
 782
6
(2.11)

9
Chapter 2

where the left side can be re-written in a more transparent form:

=  + 9

(6)  287
(2.12)

From this general solution, we can get the particular solution by determining the undetermined
integration constant I with the help of the initial conditions. For the first-order ordinary differential
equation, one initial condition is sufficient; the most convenient in this case is to give the value of
the concentration at the very beginning of the reaction. Let us denote this concentration at t = 0 by
co. (Further on we will call this concentration as the initial concentration.) Substituting t = 0 and
c = co , we readily get the value of the integration constant as

9=

(6):287
(2.13)

Let us write this result into the general equation:

= (6) 287 + 
 
(6)  287
(2.14)
:

Multiplying both sides by (n – 1) we can get a somewhat simpler form

= + (; − 1) ,
 
 287 :287
(2.15)

from which it is straightforward to express the explicit solution c (t):

=

<
287
7
> (6)?
(2.16)
=287
:

We can also rewrite the (n – 1)-th root in a power expression form equivalent to the above:
7
287
=@ A

7
287 > (6)?
(2.17)
=
:

It is worth noting that there is also an alternative method to solve the differential equation by
evaluating definite integrals in accordance with the initial condition. To do this, we evaluate the
right side integral in Eq. (2.10) over the limits from the time of the initial condition 0 to an arbitrary
final time tf , and the left side integral with respect to c from c(0) to c(tf):

− 3(5)B
 = 35 B 

( )  
 2
(2.18)

Evaluating the integral we get the result

C − F = G
  
(6) [(B)]287 :287
(2.19)

10
 Formal kinetic description of elementary reactions

As this solution is valid for any final time tf and the corresponding concentration c(tf), we can
substitute t in place of tf and c in place of c(tf), thus getting the same form of the solution as before:

− = 
 
(6)  287 (6):287
(2.20)

Some textbooks and research papers commit a severe formal error while using the definite
integration by not making a difference between the dummy integration variable and the limit of the
integration; in the above example they would write incorrectly

− 3
 = 35 
 .
  
:  2
(2.21)

We can easily see that this should not and cannot give the correct result we want to have, as the
limit of integration always changes with the value of the integration variable. However, with the
correct notation of using a different symbol for the dummy integration variable and the limit of the
integration, the pencil work (especially the keyboard work) is not really simpler than in the case of
evaluating the indefinite integrals and then calculating the integration constant. This procedure has
the advantage that the possibility of introducing incorrect notation (and result) is avoided. Thus we
shall follow the way to get the general solution first and then evaluating the integration constant to
get the particular solution for the initial value problem.
Let us return to the solution as expressed in Eq. (2.16). It is readily seen that this solution is not
applicable for all values of the reaction order n; for n = 1, the time dependence disappears and the
fraction 1/0 cannot be interpreted either. This frequently happens in the practice of natural
sciences; the solution of a mathematical model does not always give meaningful result within all
physically possible conditions. Another related interesting property is that, though the function itself
can be interpreted and gives mathematically meaningful results (in case n ≠ 1) for times t < 0, this
does not have any physical (or chemical) meaning either, as no reaction occurs before the start of
the reaction, i. e. before t = 0; thus, the solution in this time region cannot be interpreted either. We
shall return to the case of n = 1, but let us first explore the properties of the function given in Eq.
(2.16).
The disappearance of reactants is traditionally characterized by the so called half-life. By
definition, this is the instant  = /I , when the concentration of the reactants becomes c = co/2. It
can readily be calculated by writing co/2 in place of c in the implicit solution (2.15):

= + (; − 1)/I ,
I 
:287 :287
(2.22)

By rearranging we get the half-life /I as:

/I =
I2876 
(6)? :287
(2.23)

11
Chapter 2

Observing this result it is obvious that – not surprisingly – this formula of the half-life is not
meaningful for first order reactions, i. e. n = 1. For any n ≠ 1 it can be seen that the sign of the first
and the second coefficient is always the same, and the second coefficient depends on the initial
concentration. According to this, reactions having an order greater than 1 “slow down” while
proceeding, in the sense that the second time period needed to reduce the concentration by half is
longer, than the first time period necessary to halve the initial concentration, etc. Conversely,
reactions having an order less than 1 “accelerate” while proceeding, in the sense that successive
half-lives during the reaction become less and less due to the reduction of the concentration.
It is worth discussing an aspect of the solution of the rate equations that is typically generously
treated in older textbooks. Before powerful computers we use nowadays would have been widely
accessible, it was much easier to determine the parameters of functions (which are co , n and k in
case of the solution of n-th order reaction rate equations) using graphical methods. The simple and
easy-to-use tool to fit functions was the (straightedge) ruler. However, as the ruler enabled only to
draw straight lines, for this purpose, functions had to be “linearized”. The linearized version of the
solution of n-th order reaction rate equation is the implicit solution in the form of Eq. (2.15):

= + (; − 1)
 
 287 :287
(2.24)


 287
We can see that plotting the transformed of the concentration as a function of time t we would

and its slope (; − 1). If we apply


:287
get a straight line whose intercept (its value at t = 0) were

the ruler to draw a line across the measured points in this plot, the intersection of this line with the
and its slope (; − 1). Knowing the value of n, we

:287
vertical axis (at t = 0) is theoretically

could determine k from the slope.

Moreover, the order of the reaction can also be determined by the same technique of fitting a
straight line with the ruler. Returning to the n-th order rate equation (2.8), let us write in place of the
derivative on the left side its absolute value in the equation:

J J =  


(2.25)

Let us take the logarithm of both sides:

log J  J = log + ; log


(2.26)

We can see that plotting the transformed values log J J as a function of the transformed values



log c, these points will be aligned along a straight line whose intercept (its value at log c = 0) will
be log k, while its slope n. (This way, not only n but – in principle – log k could also be determined.
This is the reason that this method is called the differential method to determine the rate
12
 Formal kinetic description of elementary reactions

coefficient.1) If we want to have a quick and simple picture concerning the order of reaction, the
differential method may be a suitable choice. However, we should realize that numerical derivation
of the discrete experimental data of the function c(t) is needed, along with the logarithmic
transformation of the function itself and its derivative. Furthermore, a difficulty also arises when we
should allocate the derivatives (reaction rates) calculated from adjacent discrete data to some
intermediate time between the two data points. These procedures thus contain some arbitrariness,
and the transformations (sometimes quite heavily) distort the experimental errors. In addition,
numerical derivation always increases errors. As a consequence, kinetic parameters determined this
way are necessarily charged with high uncertainty.
All the problems mentioned above can easily be avoided if we do not insist using the straight-
edge method but perform a (nonlinear) parameter estimation based on the untransformed measured
data and the explicit solution of the rate equation. Therefore, we shall only briefly mention the usual
linearization tricks of the concentration functions enabling (a typically inaccurate and distorted)
parameter estimation with a graphical procedure using a straightedge. The knowledge of this
outdated method can help to properly understand and interpret kinetic parameters and their
limitations reported in older publications, determined with graphical methods. Nowadays, with the
availability of powerful computers and a great choice of suitable statistical and numerical software
packages we should prefer direct nonlinear methods to get more reliable, less distorted kinetic
parameters.
Up to now, we only explored the solution of the rate equation of the reaction of order n for the
case of equal initial concentration of the reactants. Let us find solutions for different initial
conditions as well, and also for different values of the reaction order n. Further on in this text we
shall see that there exist some special reactions that are zeroth-order. (E. g. heterogeneous catalytic
reactions.) The solution of their rate equation can easily be derived from the solution of the n-th
order reaction discussed above. Let us begin the detailed analysis with this reaction type.

2.1.1. Zeroth-order reactions

The rate equation for a reaction which is zeroth-order can be written as

− =,


(2.27)

for the value of the power function c0 is always 1, independently of the actual value of c. The
solution of this rate equation can easily be written by substituting n = 0 into the general solution of
the n-th order reaction:

1
In addition to this name, it is also called the van’t Hoff method, from Jacobus Henricus van’t Hoff (1852-1911), the
Dutch chemist who first described it. He was the first Nobel Prize winner in 1901 “for his discovery of the laws of
chemical dynamics and osmotic pressure in solutions”.

13
Chapter 2

 = M −  (2.28)

It is obvious that this function cannot be interpreted without limitations either. Following the
start of the reaction at t = 0, after the elapsed time t = co /k, the reactant is completely consumed and
the reaction will halt; thus, negative c values after that do not have any physical meaning. To
emphasize this condition, the concentration function can be given as  = M − , if 0 < t < co /k,
and zero, if t > co /k. We can also see that the unit of the rate coefficient k for a zeroth-order reaction
is mol dm–3 s–1 in terms of molar concentration and seconds as time units – in accordance with the
condition that the unit of the ratio co /k should be seconds. The half-life of reaction – in addition to a
substitution of n = 0 into the general expression of the n-th order reaction – can be calculated easily
by realising that the decrease of the reactant concentration is proportional to time until it reaches
zero at the instant co /k, thus it is t1/2 = co /2k. Obviously, consecutive half-life periods decrease by
50% each time after the concentration becomes half of the previous one. The plot of the
concentration of the reactant is a straight line between co at t = 0 and zero at t = co /k.
The concentration as a function of time for the product(s) of the reaction can be obtained from
the above solution relying on the stoichiometry. If the stoichiometric number of the reactant is 1 and
that of a product is νP in the reaction reactant → products, then the concentration of the product as
a function of time can be given as:

"N  , ha 0 <  < M /

N = O
"N M , ha  > M /
(2.29)

This is easy to justify based on the relation that νP moles of the product are produced from 1 mol of
the reactant. Thus N = "N[M − (M − )] = "N.

2.1.2. First-order reactions

The rate equation for a first-order reaction is

− =  .


(2.30)

It is readily seen that the unit of the rate coefficient k is simply s–1. The solution of this equation
cannot be given by substituting n = 1 into the solution obtained for the general n-th order reaction,
as it does not provide a meaningful result. In such cases, we should solve the actual rate equation.
As this equation is also separable, after separation and insertion of the integrals we get:

−3
 = 3 


V
(2.31)

Writing the primitive functions of both sides, the general solution is the following:

−ln  =  + 9 (2.32)

14
 Formal kinetic description of elementary reactions

Substituting the plausible initial condition ( = M when t = 0), the integration constant can be
obtained as 9 = − ln M . Let us plug this into the above equation, and multiply it by – 1 to get

ln  = ln M −  . (2.33)

The explicit solution is readily obtained as

 = M Z 6? . (2.34)

The concentration function of the product(s) of the reaction can be obtained again relying on
the stoichiometry. If νP moles of the product are produced from 1 mol of the reactant, based on the
relation N = "N(M − ), we can write

N = "N M (1 − Z 6? ) (2.35)

which gives the concentration of a product with stoichiometric number νP as a function of time.
The validity of the above concentration functions are – in principle – not limited for times
greater than zero. However, it is worth to consider that if the reactant concentration multiplied by
the volume results in a value inferior to the inverse of the Avogadro constant, there should be less
than one reactant molecule in the reaction vessel. Obviously, this does not have a physical meaning;
thus, in this sense, the validity of the concentration function is limited also in case of a first-order
reaction. (However, the concentration would not become negative, only decrease monotonously;
thus the concentration function would predict an ever smaller fraction of the last molecule if the
amount of the reactant is divided by the Avogadro constant.)
The half-life of the first-order reaction has a unique property. We can calculate it by
substituting co/2 in place of c:

= M Z 6?7/[
V:
I
(2.36)

Dividing both sides by co and taking logarithms we get the result:

/I =
\] I
?
(2.37)

As can be seen, the unique property is that the half-life is independent of the initial
concentration co. Accordingly, the concentration of the reactant for a given first-order reaction is
reduced by 50 % within the same time intervals, also in the case of consecutive time periods. In
other words, the reactant is consumed always at the same pace, thus the reaction does not speed up
nor slow down during the reaction. (It could be foreseen from the behaviour of the half-life of the
n-th order reactions; below n = 1, the half-life decreases, above 1 it increases with decreasing initial
concentration. Approaching 1 either from below or from above would lead to the same result.)

15
Chapter 2

We can also notice that the implicit solution (2.33) is already appropriate for the estimation of
kinetic parameters using a graphical plot and a ruler; plotting the measured log c – t data in a
diagram, the discrete points are found along a straight line. The slope of this line is k and its
intercept ln M .

2.1.3. Second-order reactions

The rate equation for a second-order reaction if the concentrations of the reactants are
identical can be written in the form

− =  I .


(2.38)

We can see that the SI unit of the rate coefficient k is dm3 mol–1 s–1. The solution of this equation
can be given by substituting n = 2 into the solution obtained for the general n-th order reaction:

=

7
> ?
(2.39)
=:

The concentration function of the product(s) of the reaction can be obtained as usual, relying
on the stoichiometry. If νP moles of the product are produced from 1 mol of the reactant, i. e.
N = "N (M − ), we can write

N = "N M 1 −  >  

: ?
(2.40)

for the concentration function of the product.

It is worth noting here as well that, if the reactant concentration multiplied by the volume
results in a value inferior to the inverse of the Avogadro constant, there would remain less than one
reactant molecule in the reacting system; thus the validity range of the solution is limited also in this
case. However, negative concentrations would never result from the concentration function.
As we have already stated, the half-life of the reactant of a second-order reaction depends on
the initial concentration. Substituting co/2 in place of c, we get

=
: 
7
I > ?7/[
, (2.41)
=:

from which we can express the half-life:

/I =

? :
(2.42)

This reveals that the half-life of second order reactions is inversely proportional to the initial
concentration co. Accordingly, each consecutive reduction by 50 % of the concentration takes twice
as much time as the previous reduction by half. As the half-life increases during the reaction, we

16
 Formal kinetic description of elementary reactions

can say that – in this sense – the reaction “slows down” as it proceeds. A plot of the reactant
concentration as a function of time clearly reflects this tendency.
To get a formula for using the ruler to estimate kinetic parameters, we can start from Eq.
(2.39). Taking the reciprocal of both sides we can get the following equation:

= + 
 
 :
(2.43)

We can see that the inverse of the concentration as a function of time results in a plot where the
−  data points are aligned along a straight line. The slope of the line is the rate


discrete

.

:
coefficient k and the intercept gives

Those who paid close attention to the above derivation might have noticed that the obtained
solution of the rate equation applies only for the case when the concentrations of the two reactants
are identical, but not the reactants themselves. When the two reacting molecules are the same, the
rate equation (2.38) slightly changes, and so does its solution. Let us begin with the relevant
stoichiometric equation:

2A → Products (2.44)

The proper rate equation is the following:

− = 2I
.

(2.45)

For the sake of simplicity, let us drop the subscript A further on:

− = 2 I


(2.46)

It is worth noting that the factor 2 appears in the rate equation as the reaction rate always refers to
the stoichiometric equation; while the rate relative to “one mol equation” is proportional to  I , the
rate relative to one mol reactant is twice as large: 2 I . (We would get the same result by adding
two rate equations (2.38) which refer only for one mol reactant.) We should remember this property
of the rate equation; to get the rate of change of a component, the product of the rate coefficient and
the relevant concentration(s) should be multiplied by the stoichiometric number of the component.
It is readily seen that this change results only in a replacement of k by 2k. Accordingly, the
relevant solution has the form

=

7
> I?
, (2.47)
=:

and the half-life will also change accordingly:

/I =

I?:
(2.48)

17
Chapter 2

Compared to when two different molecules react in a second-order reaction, in case of two identical
molecules, the half-life is reduced by a factor of two. This is not surprising as it follows from the
fact that the rate of disappearance is twice as fast in the latter case than in the former.
The concentration of the product(s) of the reaction as a function of time can be obtained in this
case by replacing the stoichiometric number νP by νP /2, and writing 2k in place of k in Eq. (2.40).
Those who paid close attention to the above derivations might also have noticed that the
solution of the rate equation was applied only for the cases when the initial concentrations of the
two reactants were the same, and when the two reactants were chemically identical. We still not
discussed the case of two different reactants with non-identical initial concentrations. Let us recall
the relevant stoichiometric equation

A + B → Products , (2.49)

along with the corresponding rate equation

− =− =   .
. /
 
(2.50)

Seemingly, there are two variables on the right side, but the stoichiometry reduces this to one
underlying variable; the same amount of reactant A will always be consumed as that of the reactant
B. If the volume does not change during the reaction, this relation also applies for the
concentrations. Let us denote the decrease in concentration (the progress variable) by x, and the
(different) initial concentrations of the reactants by ,M and ,M , respectively. Using this notation,
the rate equation can be written as

− =− = (,M − a)(,M − a)
(.,: 6`) (/,: 6`)
 
(2.51)

We can simplify the two time-derivatives by realising that ,M and ,M do not depend on time:

− =− + =
(.,: 6`) .,: ` `
   
(2.52)

We get a similar result for B as well. The differential equation to solve thus simplifies in the
following form:

= (,M − a)(,M − a) .
`

(2.53)

This differential equation is also a separable one. After separation, we get the two sides of the
equation ready to integrate:

3 (
a = 3 


.,: 6`)(/,:6`)
(2.54)

18
 Formal kinetic description of elementary reactions

The right side provides the well-known primitive function of a zeroth-order power function, while
we get a rational algebraic fraction to integrate on the left side. We can recall that the integration of
this fraction can be done by resolving it into the sum of partial fractions:

3 (
a = 3 C ( + ( F
a
  
.,: 6`)(/,: 6`) /,: 6.,: )(.,:6`) .,: 6/,: )(/,: 6`)
(2.55)

The integration can readily be performed resulting in the following primitive function:

−ln ( − ln ( + b;cd;
(.,:6`) (/,: 6`)
/,: 6.,: ) .,: 6/,: )
(2.56)

The solution of the differential equation including the integration constant I, after some rearrange-
ment can be written as:

ln (.,: 6`) =  + 9
 ( 6`)
(.,: 6/,:)
(2.57)
/,:

Upon substitution of the initial condition (x = 0 at t = 0) the integration constant I becomes

9= ln
 .,:
(.,: 6/,:) /,:
.

If we plug this in the solution and make use of the identities of the logarithm function we get:

ln = 
 /,: (.,: 6`)
(.,: 6/,: ) .,:(/,: 6`)
(2.58)

Let us rewrite now the original symbols cA and cB in place of ,M − a és ,M − a, and we already
have the implicit solution:

ln = 
 /,: .
(.,: 6/,: ) .,: /
(2.59)

This result is already appropriate for the estimation of kinetic parameters using a graphical plot
and a ruler; plotting the left-side transform of the concentrations as a function of time, discrete
experimental points are found along a straight line across the origin, which has a slope of k.
To find the explicit solution, we should express the concentrations cA and cB from this
equation. To do so, let us first make use of the relation  = ,M − a , and add the difference
,M − ,M to  = ,M − a. This results in  = ,M − a = ,M − ,M + (,M − a), which is
identical to  = ,M − ,M + . This way we have eliminated  , and the remaining variable in
the solution (2.59) is  . After some rearrangement and applying inverse logarithm (exponentiation)
we can get the following result:

 =
.,: 6/,:
=/,: (=
 6 e .,: 8=/,: )fg
(2.60)
=.,:

19
Chapter 2

It is easy to see that – for symmetry reasons (the role of A and B can be interchanged) – the
time-dependence of the concentration of B is similar to this:

 =
/,: 6.,:
=.,: (=
6 e /,: 8=.,: )fg
(2.61)
=/,:

The above two equations provide the solution for the rate equation in general of second-order
reactions.
However, this solution cannot be used within all circumstances. If the initial concentrations of
A and B are identical, we get an expression 0/0 for the concentration functions that cannot be
interpreted. Luckily, we have the previous solutions (2.39) or (2.47) for this case.
It is worth noting that the half-life of the reaction is not unique in this case nor can it be defined
within all initial conditions. What is always meaningful is the half-life of the reactant whose initial
concentration is inferior with respect to the other reactant. This reactant can namely be completely
consumed during the reaction. The reactant with higher initial concentration can only be reduced by
half if its initial concentration is less than twice the concentration of the other reactant.

2.1.4. Third-order reactions

The rate equation of a third-order reaction for three different reactants with identical initial
concentration can be written in the following form:

− =  h


(2.62)

We can see that the SI unit of the rate coefficient k is dm6 mol–2 s–1. The solution of this rate
equation can be obtained by substituting n = 3 into the solution of the general n-th order rate
equation with the plausible initial condition c = co at t = 0:

=<7

[ > I?
(2.63)
=:

The concentration function of the product(s) of the reaction can be obtained as usual, taking
into account the stoichiometry. If νP moles of the product are produced from 1 mol of the reactant, i.
e. N = "N(M − ), we can write

N = "N M i1 − < j.

[fg
 > [
(2.64)
= :

The half-life of the reaction can be calculated by solving the equation

20
 Formal kinetic description of elementary reactions

=<
: 
7
I [ > I?7/[
. (2.65)
=:

After both sides are squared and their reciprocals taken, we get the result

/I = I?[ .
h
(2.66)
:

This shows that the half-life of second order reactions is inversely proportional to the square of
the initial concentration co. Accordingly, each consecutive reduction by 50 % of the concentration
takes four times as much time as the previous reduction by half. Thus the half-life of third-order
reactions increases quite largely during the reaction, which means that – in this sense – the reaction
“slows down” significantly as it proceeds. A plot of the reactant concentration as a function of time
reveals this tendency.
To estimate kinetic parameters using a graphical plot and a ruler, we can derive the necessary
linearized relation by starting from Eq. (2.63). After both sides are squared and their reciprocals
taken, we get the following equation:

= [ + 
 
 [
(2.67)
:

Thus, plotting the inverse square of the concentration as a function of time, discrete experimental

points are found along a straight line of slope of k and intercept [ .
:

It is clear that there are reactions with different initial conditions than the above one, and we
can also imagine that three identical molecules react and the reaction rate is proportional to the third
power of their concentration. In the latter case we can follow the procedure we have discussed with
second-order reactions. Starting with the relevant stoichiometric equation

3A → Products , (2.68)

a factor of 3 will appear in the corresponding rate equation:

− = 3h
.

(2.69)

The solution of this differs also from the previous case only that it contains 3k in place of k.
Dropping the subscript A from the symbol of the concentration, the proper solution reads as
follows:

=<7

[ > l?
(2.70)
=:

The half-life of the reaction can be obtained in a similar manner:

21
Chapter 2

/I =
h
l?:[
(2.71)

If the initial concentration of the reactants is not identical, we can still discern two cases. One
is when three different reactants take part in the reaction having different initial concentrations, and
the other is when two identical reactant molecules and a third, different one have different initial
concentrations. Let us begin the description of the second case by writing the related stoichiometric
equation:

2A + B → Products (2.72)

The corresponding rate equation has the following form:

−I =− = I 
 . /
 
(2.73)

The two variables on the right side can also be reduced to one by realising that the amount of
reactant A consumed will always be twice as much as that of the reactant B. If the volume does not
change during the reaction, this relation also applies for the concentrations. Let us denote the
decrease in concentration (the progress variable) by x, and the (different) initial concentrations of
the reactants by ,M and ,M , respectively. Using this notation, the rate equation can be written as

− =− = (,M − a)I (,M − a)

 (.,: 6`) (/,: 6`)
I  
(2.74)

We can simplify the two time-derivatives by taking advantage that ,M and ,M do not depend on
time:

= (,M − a)I (,M − a)

 `
I 
(2.75)

The other differential equation differs only by a factor of 2 from this one. The implicit solution of
this differential equation can be obtained by the method of partial fractions (similar to the second-
order case) and – by substituting the initial condition (x = 0 at t = 0) and re-substituting the original
time-dependent concentrations cA and cB – yields

C −  F + ( ln /,: . =  .
     
(.,: 6I/,: ) .,: .,:6I/,: )
[ (2.76)
. .,: /

This result is already appropriate for the estimation of kinetic parameters using a graphical plot
and a ruler; plotting the left-side transform of the concentrations as a function of time, discrete
experimental points are found along a straight line across the origin, with a slope of k.
Unfortunately, the explicit solution is not known for this case. Thus, if we want to estimate kinetic
parameters without relying on the graphical method, we should use numerical methods including
numerical inversion of the implicit solution, or numerical integration to get time-dependent
concentration values.

22
 Formal kinetic description of elementary reactions

Finally, let us discuss the general third-order reaction with three different initial concentrations
,M , ,M , and ,M , respectively. Starting from the relevant stoichiometric equation

A + B + C → Products , (2.77)

we can write the corresponding rate equation in the following form:

− =− =− =   
. / 1
  
(2.78)

If the volume does not change during the reaction, we can express time-dependent concentrations
again with the help of a single progress variable x. Using this notation, the rate equation can be re-
written as

− =− =− = (,M − a)(,M − a)(,M − a)

(.,: 6`) (/,: 6`) (1,:6`)
  
(2.79)

We can simplify the time-derivatives by making use of the time-independence of the initial
concentrations and get the differential equation to solve:

= (,M − a)(,M − a)(,M − a)

`

(2.80)

The solution of this differential equation can also be obtained by the method of partial fractions.
After integration, substitution of the initial condition (x = 0 at t = 0) and re-substitution of the
original time-dependent concentrations – we get the implicit solution
= = =
\] . \] / \] 1
+ + =  .
=.,: =/,: =1,:
(/,: 6.,:)(.,: 61,: ) (.,: 6/,: )(/,: 61,:) (.,: 61,: )(1,: 6/,: )
(2.81)

This result is also appropriate for the estimation of kinetic parameters using a graphical plot
and a ruler; plotting the left-side transform of the concentrations as a function of time, discrete
experimental points are found along a straight line across the origin, with a slope of k.
Unfortunately, the explicit solution is not known for this case either. Thus, if we want to estimate
kinetic parameters without relying on the graphical method, we should use numerical methods
including numerical inversion of the implicit solution, or numerical integration to get time-
dependent concentration values.

2.2. Generalisation and extension of the order of reaction; pseudo-order

As it is mentioned in the introductory part of this chapter, there exist composite reactions
which do have an order but this order is not an integer. (For elementary reactions, the order is
always a positive integer.) Accordingly, we can generalise the order of reaction for any empirical
rate equation which can be written in the form

= "n  ∏p!   .
m
−
o

(2.82)

23
Chapter 2

In this equation, r is the number of reactants, n and  are concentrations, νj is the stoichiometric
number, and the symbol q is called the order of the reaction with respect to the i-th component (or
component Ai in Eq. (1.1)). The sum of the orders ∑p! q = ; is called the overall order of the
reaction. In relation to this name, q is also called the partial order of Ai . In a general (composite)
reaction, the partial order of components as well as the overall order should not be a positive
integer; it may be a negative integer or a rational nonintegral number. It is worth noting that a
nonintegral order always implies a composite reaction.
As we shall see further (when discussing composite reactions), the rate equation of the majority
of composite reactions cannot be written in the form of Eq. (2.82). At this point we could ask why
such a generalisation of the reaction order should be considered. We can have the answer also at the
discussion of composite reactions; when studying unknown reactions, the notion of reaction order
can help to explore kinetic properties. Kineticists often begin this exploration by “forcing” the
reaction to behave as if its rate equation would have the form of Eq. (2.82), thus at least one of its
reactants would have a genuine – mostly integer – order. The simplest way of forcing this behaviour
is to add all but one reactant in such great excess that their concentrations during the progress of the
reaction remain constant to a good approximation. (This technique is sometimes referred to as
flooding, or as isolation.) A frequently applied example is the case of two reactants, when the
concentration of one reactant is many times that of the other.
Let us consider the case when the rate equation of the reaction of components A and AI can be
written in the second-order form

− =− =  I .
7 [
 
(2.83)

For example, in case of I = 100 it is easy to see that, even after the reaction is completed, the
concentration of AI only changes by 1 percent. The typical error of time-dependent concentration
measurements being in this order of magnitude, we do not make a big mistake by considering I as
a constant throughout the reaction. In that case, the product ′ = I can also be considered
constant, thus we can rewrite the rate equation as

− = ′ .
7

(2.84)

We can see that it is formally identical to a first-order rate equation. However, not being a genuine
first-order reaction, it is called a pseudo-first-order reaction1.
The pseudo-first-order rate coefficient has an interesting property; its usual unit is not s–1, but
dm3 mol–1 s–1, according to the rate equation (2.83) or the product I . Formerly, when parameter
estimation had been performed using graphical methods and a ruler, it was common practice to

1
The Greek prefix ψευδο- has the meaning false, or not a real one.

24
 Formal kinetic description of elementary reactions

apply the pseudo-first-order results to determine a second-order rate constant. To do so, calculated
linearized (ln  – t) transformed values have been plotted at different concentrations I. From these
diagrams, the slopes of the lines from different diagrams were determined, as discussed in section
2.1.2. Pseudo-first-order rate coefficients obtained this way were plotted as a function of the
corresponding concentrations I , and a staraight line was fitted to the data points in the diagram.
The slope of this line was the (graphically) estimated value of the second-order rate coefficient,
according to the relation ′ = I .
In case of the graphical parameter estimations described, there is no suitable method to
determine the uncertainty of the rate coefficients. Performing the described line fittings using
appropriate statistical methods, we could calculate correct uncertainties of the pseudo-first-order
rate coefficients, as well as the resulting second-order rate coefficient. However, it is pointless to
follow this tedious method for two reasons. The first one is that the distortion of the errors due to
the transformation of the original data would lead to biased results. The second one is that the
statistical properties of the results obtained after the two stages (e. g. the number of the degrees of
freedom of the probability distribution of the second-order rate coefficient) were unfavourable. A
more simple, more precise and statistically more favourable method is to fit the concentration
function (2.60) to all measured points (obtained by measuring both concentrations) and estimate the
second-order rate coefficient as a parameter of this function, along with its uncertainty.
However, the method of flooding (or isolation) is a usual way to study unknown reactions. If it
turns out for example, that – in case of two reactants – the reaction has a pseudo-first-order kinetics
for both reactants, it is a reasonable conclusion that the overall reaction is a second-order one, and
we can put up a suitable experimental design to determine the second-order rate coefficient using
proper non-linear parameter estimation methods.

25
Chapter 2

Suggested literature

M. J. Pilling, P. W. Seakins: Reaction Kinetics, Oxford University Press, 1995

J. de Paula, P. W. Atkins: Physical Chemistry, 10th edition, Oxford University Press, 2014
L. J. Silbey, R. A. Alberty, G. B. Moungi (2004) Physical Chemistry, 4th edn., Wiley, New York
J. I. Steinfeld, J. S. Francisco, W. L. Hase: Chemical Kinetics and dynamics, 2nd edition,
Prentice Hall 1998

26