\~bs.

U ni t Processes in
Organic Synthesis

EDITOR-1N-CHIEF
P. H. GROGGINS
Chemical Ellgin,"er
Technical Advisor, Food Machinery and Chemical CorPoration
Author of "Aniline and Its Derivatives"
and "Chemicals alld Food Production"

[ ~APAU CENTRAL LfBR

HYDERAHAD-500 0:"

FIFTH EDITI - ..... ,\I'AU -------~~.- CCl~' . i\ L L Ah

~ •. ..._........... -
ANGRAU Central Library
Hyderabad
-)- 6'3
1265
I tllllr 11I1I 1111111111 1111111'

A(;C: 1'0; 1265 Date: fbf);

INTERlfATIONAL S~ENT EDITION

 McGRAW. HILL KOGAKUSHA, LTD.
Tokyo Auckland Beirut Bogota DUsseldorf Johannesburg Lisbon London
Lucerne Madrid Mexico New Delhi Panama Paris San Juan Sao Paulo
Singapore Sydney
·uNtt PROCESSES IN ORGANIC SYNTHESIS
. ,.~

INTERNATIONAL STUDENT EDITION

Exclusive rights by McGraw-Hill Kogakwsha, Ltd., for manufactre and export.

This book connot be re-exported from the country to which it is consigned by
McGraw-Hill.

Xl[

Copyright © 1958 by the McGraw-Hill Book Company, Inc. Copy right, 1935,
1938, 1947, 1952, by the McGraw-Hili Book Company, Inc. All rights reserved. No
part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying,
recording, or otherwise, without the prior written permission of the publisher.
PREFACE

A quarter of a century has elapsed since the first edition of "Unit Proc esses in
Organic Synthesis" went to press. During that span of years the domestic
synthetic organic chemical industry has grown enormously. The production of
petrochemicals, plastics, protective coatings, surfactants, chlorinated solvents,
and elastomers-all involving organic synthesis-has multiplied many times. New
technics have been developed, and chemical engineers have contributed much to
make production processes more efficient.
If adoption as a teaching tool and use as a plant reference work may serve as
criteria, "Unit Processes in Organic Synthesis" has apparently served a useful
purpose. Acceptance of the work has been world-wide, and it is with
gratification that the authors note that the book has been or is cur
rently being translated into five foreign languages. Apparently this re flects the
need for such a textbook in a growing technological economy. In the preface to
the first edition, the editor distinguished between the now well-established terms
unit processes and unit operations and showed how they are intimately related in
the chemical engineering aspects of organic synthesis. The prime objective or
the present volume continues to be the presentation of the industrial technic
used in converting organic raw materials into usable products by various
processes. Accordingly, the underlying principles and factors for each unit
process are organized in a systematic manner with most of the emphasis on
fundamental scientific and technological principles. Detailed discussion of
manufacturing processes is included only to the extent required for the
elucidation of these basic principles. Thus it is expected that the present volume
will continue to serve as a "semiworks course" to facilitate the student's
transition from the university classroom to an industrial environment.
In keeping with the modern trend toward the so-called chemical engi neering
sciences, the authors have stressed fundamental advances in theory as well as in
industrial practice. Thus, thermodynamics and kinetics have received increased
empha~is. These subjects are inextricably associated with all chemical reactions

and, therefore, are particularly important in the study and practice of unit
processes. In this fifth edition, three new in troductory chapters are devoted to
these subjects.
vii
viii PREFACE
To ensure that the present text reflects the most recent advances in the
technology of unit processes, many new coauthors have been invited to
contribute to its pages. These new collaborators represent an accurate cross
section of the American chemical industry. Their contributions to the technical
literature in their specialized fields are evidence of their competence. Their
cooperation in the preparation of the present edition should assure a
technologically sound text.
The editor is deeply obligated to managements of many chemical organi
zations for making possible employee participation in the preparation of this
volume. It is accordingly hoped that their combined contributions will make
possible a better training of chemical engineers and chemists for the plants of
our expanding chemical industry.

P. H. Groggins
OONTENTS

vii
Preface
1. Applications of Thermodynamics in Unit Processes
Sources of thermodynamic data, 1.
2. Chemical Kinetics 22

3. Chemical-process Kinetics 40 Factors that affect a chemical process, 40; Reactor shape and
dIed of back mixing, 43.
4. Nitration. 60 Introduction, 60; Nitrating agents, 61; Aromatic nitration, 63; Kinetics and
mechanism of aromatic nitration, 68; Nitration of paraffinic hydrocarbonR, 73; Nitrate
 esters, 80; N-nitro compound~, 81; Thermodynamics of nitrations, 83; Process equipment for
technical nitration, 96; Mi.'wd acid for nitrations 101; Typical ind ustrialni tra tion
processes, 107.
5. Amination by Reduction 129 Introduction and definitions, 129; Methods of reduction, 133;
Iron and ucid (Bechamp) reduction, 135; Other metal and acid reductions, 165; Catalytic
hydrogenation, 168; Sulfide reductions, 186; Electrolytic reductions, 190; Metal and alkali
reductions, 192; Sulfite reductions, 198; Miscellaneous reductions, 201.

6. Halogenation 204 Introduction, 204; Thermodynamics and kinetics of hulogenation reactions,

211; Survey of halogenations, 222; Chlorination in the presence of a cat.alyst, 265;
Photohalogenation, 267; Design and construct.ion of equipment for halo genation, 268;
Technical halogenations, 270.
7. Sulfonation and Sulfation 303 Introduction, 303; Sulfonating and sulfating agents and their
principal appli cations, 305; Chemical and physical factors in sulfonation and sulfation, 337;
Kinetics, mechanism, and thermodynamics, 351; The desulfonation reaction, 358;
Working-up procedures, 362; Industrial equipment and techniques, 364; Transition from
batch to continuous processing, 368; Technical preparation of sulfonates and sulfates, 375.

8. Amination by Ammonolysis 388 General discussion, 388; Aminating agents, 389; Survey of
ami nation reac tions, 397; Physical and chemical factors affecting ammonolysis, 426; Cata lysts
used in amination reactions, 432; Corrosion and the pH of the autoclave charge, 437; Kinetics of
ammonolysis, 439; Thermodynamics of ammonolysis, i.x
x CONTENTS
444; Design oj reactors and auxiliaries, 447; Technical manufacture of amino
compounds, 450; Control of the ammonia-recovery system, 482.
9. Oxidation
486
Types of oxidative reactions, 486; Oxidizing agents, 488; Liquid-phase oxida tion
with oxidizing compounds, 503; Liquid-phase oxidation with oxygen, 507;
Vapor-phase oxidation of aliphatic compounds, 517; Vapor-phase oxidation of
aromatic hydrocarbons, 534; Kinetics and thermochemistry, 542; Apparatus for
oxidations, 549.
10. Hydrogenation . 555 Introduction, 555; Hydrogen: production and properties, 560;
Catalytic hydrogenation and hydrogenolysis: type reactions, 574; Kinetics and ther
modynamics of hydrogenation reactions, 590; General principles concerning
hydrogenation eatalysts, 600; Apparatus and materi~ls of construction, 608; Industrial
processes, 612.
11. Hydrocarbon Synthesis and Hydroformylation . 651 Introduction, 651; Technology of
Fischer-Tropsch operation, 654; Catalysts, 658; Thermodynamics and kinetics of the
Fischer-Tropsch reaction, 661; Reactor development, 6134; Commercial operation, 671;
Economics of Fischer Tropsch operations, 675; Methanation, 675; Processes related to the
Fischer Tropsch synthesis, 678.
12. Esterification
694
Esterification by organic acids, 605; Esterification of carboxylic acid deriva tives,
710; Esters by addition to unsaturated systems, 720; Esters of inor ganic acids, 723;
Esterification practice, 726.
13. Hydrolysis 750 Definition and scope, 750; Hydrolyzing agents, 752; Materials susceptible
to hydrolysis, 756; Kinetics, thermodynamics, and mechanism of hydrolysis, 761;
Equipment for hydrolysis, 772; Technical operations involving hydroly sis, 773.
14. Alkylation 804 Introduction, 804; Types of alkylation, 806; Alkylating agents, 815; 'Factors
 controlling alkylation, 819; Equipment for alkylations, 825; Effect of alkyla tion, 828;
Technical alkylations, 829.

15. Part 1. Principles of Polymer Chemistry 856 Introduction, 856; Chemistry of

polymerization reactions, 858; Methods of polymerization, 892; Polymerization kinetics,
904; Solution properties of polymers, 913; Polyelectrolytes, 923; Influence oi molecular
characteristics on polymer properties, 939; Influence of intermolecular arrangement on
polymer properties, 941.
polymers, 943.
Part 2. Polymerization" Practice
943
Industrially important polymerizations and

Index . 1037
CHAPTER 1

APPLICATIONS OF THERMODYNAMICS IN
UNIT PROCESSES

By THOMAS E. CORRIGAN AND KENNETH A. KOBE

"Unit Processes in Organic SyntheRis" deals with the major chemical

:ansformations which are of importance to the chemical industry. One f the
most important tools that can be used to predict the behavior of lese chemical
reactions is chemical thermodynamics. Whether certain hemical reactions can
take place, to what extent chemical conversions Itn occur, the effect of
temperature and pressure on the behavior of a hemical reaction, the
composition of reactor effluents if equilibrium is ~ached, the driving force of each
of several competing reactions taking lace, and the amount of heat released are
some of the aspects of chemical rocesses whieh can be determined from
thermodynamic calculations alone. 'hermodynamics cannot, however, tell
anything about the speed of the mction, the effect of the shape of the reactor
used, or the relative speeds f competing reactions. If these questions are to be
answered, it will be y the use of experimental kinetic data.
Not only is a knowledge of thermodynamics necessary to an understand Ig of
unit processes, but also it is a prerequisite to the application of ki etics. A
knowledge of both thermodynamics and kinetics is fundamental ) a complete
understanding of the unit organic processes. The process Igineer finds that
before the kinetics of a chemical reaction can be utilized s thermodynamics
must be known. Before becoming concerned with how LSt a reaction will go,
one must know if it can go at all and, if so, how far. or a more thorough
discussion of the principles reviewed here, the reader
c referred to some of the standard texts on chemical engineering thermo

lynamics.1
Energy Relations-First Law. The subject of thermodynamics deals Llmost
exclusively with energy relationships. Many problems such as the

1 DODGE, "Chemical Engineering Thermodynamies," McGraw-Hill Book Company, nc.,

New York, 1944; HOUGEN, WATSON, and RAGATZ, "Chemical Process Principles," )arts I
 and II, John Wiley & Sons, Inc., New York, 1954 and 1947; SMITH, "Introduc ion to Chemical
Engineering Thermodynamics," McGraw-Hill Book Company, Inc., 'lew York, 1949.
2 UNIT PROCESSES IN ORGANIC SYNTHESIS
equilibrium conversion in an adiabatic reactor or the heat requirements and
quantity of catalyst needed in a catalytic reactor depend upon energy balances
for their solution. Energy balances are based upon the first law of
thermodynamics, which states that energy cannot be created or destroyed
during' a process, although it may change from one form to another. The total
energy of a system entering a process plus any addition during the process must
equal the total energy of the system lea"ing the process.
Classification of Energy: The forms of energy can be classified into two
groups: (1) the first is related to the system and (2) the second is associated with
the process. The former includes energy poss('ssed by material of .the system;
the latter includes energy produced or transferred by the process
ing.
Group I Surface energy
Magnetic energy
Properties of a system:
Internal energy Group II
Flow or pressure energy Properties of a process:
Kinetic energy Heat
Potential energy Work

Internal energ:y (U) is that which a substance possesses because of the

motion and configuration of its molecules, atoms, and subatomic particles. The
flow or pressure energy PV is the product of pressure and volume. It may be
regarded as the energy a substance possesses by virtue of the space it occupies.
Kinetic energy is the energy a substance possesses be
cause of its motion and is mu2 j2, where m and u are the mass and velocity of the
substance, respectively. Specific kinetic energy is u2j2gc, where gc is the
gravitational constant, with Ijgc being the mass of a unit weight. The energy
which a material possesses bl cause of its position in relation to some datum
plane is called the potential energy. Two forms of energy, that due to surface
tension and that due to magnetic effects, are usually negligible in magnitude
compared with the other forms and are not used in the energy balance. The
forms of energy which depend upon the process are heat q and work w. Heat is a
term applied to that form of energy which flows as the result of a temperature
gradient. Work applies to the expendi ture of energy by mechanical processes.
Energy Balance over a Flow System. The first law of thermodynamics when
applied to two points in a flow system may be expressed by the fol lowing
equation:

u,' u,' U1 +p,V,+Z'+-2 +q-w= U.+P.V'+Z'+- ~ 2 ~

The term q refers to the heat absorbed by the system; w refers to the work done
by the system on the surroundings. In the over-all energy balance
THERMODYNAMICS IN UNIT PROCESSES
3
for chemical processes, the changes in kinetic and potential energy are
frequently negligible compared with internal energy and may be disre garded.
The equations for a flow system then become the same as those for a nonflow
(or batch) system. For those Cases in which the system per forms no work upon
its surroundings, the equation reduces to

UI + PI VI + q = U, + [>, V,
or or = H, - HI = 1:J.H
q = (U, + P,V,) - (UI + PIVI) q

where H (ea~py) is defined by the equation

H = U +PV

Standard States. To fix exactly the properties of a component in a system, a

standard state must be defined which specifies the temperature, pressure, and
physical state of the component. The standard temperature

is 25°C (298. 15°K) ; standard pressure is 1 atm (or unit fugacity = 1.0 atm). r
The physical state must be specified, for the energy of water de pends on
whether it exists as a liquid or a gas; or the energy of carbon depends on
whether it exists as ,a-graphite or diamond.
Heat of Reaction. The change in enthalpy of a system when a reaction
occurs at constant pressure is usually called the heat of reaction, though more
properly it is the enthalpy change on reaction. The system may have to give off
or absorb heat (q) in order to maintain a constant temperature in the system.
Exothermic
System loses heat 1:J.H =- 1:J.H = +
System gains heat
Endothermic

When reactants and products are in their standard states, the enthalpy change
is the standard heat of reaction. If the reaction is a combustion reaction, the
enthalpy change is the etandard heat of combustion flHc. If the reactants are the
elements in their standard state, the product is a
compound in its standard state, the enthalpy change is the standard heat of
formation flH!.
For the general reaction aA + bB ~ cC + dD, the standard heat of reaction
flHr~ can be calculated from either standard heat-of-formation or
heat-of-combustion data.
From heat-of-formation data:
1:J.H .. = ~(1:J.H,)produo .. - ~(1:J.H')r ... tant.

= c(1:J.H,)c + d(1:J.H,)D - a(AH,)A - b(AH')B

 Values for the heat of formations per mole can be found in the references for
thermodynamic data listed at the end of this chapter.
4 UNIT PROCESSES IN ORGANIC SYNTHESIS From
heat-of-combustion data:
AH,. = ~(-AH,)p,oducto - 2:( -AH.) .. ""t.nto
= c( -AH,)c + d( -AH,)D - a( -AH,)A - b( -AH,)B
It is important that the physical state of all components be the same in all
values for heats of formation or combustion. In a combustion reaction, the
water formed may be either liquid (the usual standard state) or gas. C,H.(g) +
3.50,(g) -> 2CO,(g) + 3H,O(I)
C,H.(g) + 3.50,(g) -> 2CO,(g) + 3H20(g)
- AH, = 372.82 kcal/g mole

- AH, = 341.26
Latent heat of vaporization of 3 moles H20 = 31.56
H 20(I) -> H20(g) AH" = 10.52

The heat of combustion when the water formed is in the liquid state is known
as the gross heat of combustion; when the water formed remains as a gas, it is
the net heat of combustion.
The heat of reaction may also be calculated by combining equations for
which thc heats of reaction are already known. To calculate the heat of reaction
for any reaction requires merely adding algebraically the equations for
reactions having known values of I:.H so that their sum gives the re
action desired. This enables the heat of reaction to be calculated for a reaction
that will not go quantitatively in a calorimeter. In the following case if the
heats of reaction of the first two chemical equations are known, the heat of
reaction of the third is the difference between the first two:
C({j) + 0, -> CO,(g)
CO(g) + Yz O,(g) -> CO,(g)
C({j) + Yz O,(g) -> CO(g)

The usual form in which reaction-heat data are available in the literature is a
tabulation of either the heat of formation or the heat of combustion of the
individual compounds in their standard states. For inorganic com pounds the
heat of formation is usually tabulated, while for organic com pounds the data
given are heats of combustion. Sources of data are listed at the end of this
chapter.
Although both temperature and pressure are specified in designating the
standard state for the reaction, the change of heat of reaction with small
changes in pressure is generally small and often may be neglected. For ideal
gases, the heat of reaction is not affected by pressure. For liquids and solids,
there is some slight effect, but it is so small that it may be ignored in most
cases. The cases in which the effect of pressure upon heat of re
action may be significant are those involving reactions of nonideal gases at
high pressures.
Effect of Temperature upon Heat of ReactiOl1. The effect of tempera ture
upon heat of reaction may be quite extensive and cannot be ignored
THERMODYNAMICS IN UNIT PROCESSES 5

in most cases. This is especially true in the case of reactions occurring at high
temperatures. The heat of reaction at any temperature may be cal culated from
the following expression:
= tJ.Hs
+ 'i:,Hp - 'i:,H
tJ.HT tJ.HT R

'i:,HRCJ'i:,HP
tJ.Hs

where !lH T = heat of reaction at temperature T

I1H s = standard heat of reaction at Ts
J:.H p = sum of enthalpies of all products between Ts and T J:.H R = sum
of enthalpies of all reactants between Ts and T J:.Hp and J:.HR may be evaluated
by the most convenient method. This may be done by looking up values for
individual enthalpies in tables, cal culating them from mes,n heat capacities, or
by integrating heat-capacity equations.

!lCpm = difference between mean heat capacities of all products and all
reactants
!lCp = difference between heat capacities of all products and all reactants tJ.Cp
= c(Cp)c + d(Cp)D - a(Cp)A - b(Cp)B
Entropy Change of a Rear.tion. Let us now consider what makes the
reaction proceed. The subject of thermodynamics contributes information on
why it is possible for a process to take place and how far it may proceed as well
as on the total energy relations involved when the reaction occurs.
Thermodynamics is concerned not alone with the total energy relations in a
given system but also with the degree to which this energy can be utilized to
cause a given process to occur.
Thermodynamic energy terms such as enthalpy and internal energy are a
measure of the total energy in a system but make no reference to the de gree to
which that energy is available. There are also terms which are a measure of the
degree to which the energy present in a system will make the process go. These
terms are the free energy G and the entropy S.
The latter term, entropy, is defined mathematically by the differential
equation relating it (the entropy) to the heat transferred in a reversible process:

dS "'" dq.
T
6 UNIT PROCESSES IN ORGANIC SYNTHESIS
A reversible process is regarded as one in which the driving force causing the
process is at all times only infinitesimally greater than that resisting the
process. It is at all times, therefore, at equilibrium. The reversible process is
also an ideal or imaginary on~ from the point of view that all sources of energy
dissipation are eliminated.
Free-energy Term. Entropy, defined in this manner and when multi plied by
the absolute temperature, gives a quantity TS which is a measure of that portion
of the energy which is unavailable. The term TS subtracted from the enthalpy
leaves a quantity which is that portion of the enthalpy in a flow process, or
constant-pressure batch process, which is available for useful work. The latter
term is called the free energy and is defined as: G = H - TS.
The free energy may be regarded as a property of a system which is a direct
measure of that portion of the total energy which will cause the re action to
proceed. The total energy that a system possesses may be rep resented as shown
in Fig. 1-1. The total energy H, can be divided into

FIG. 1-1. Relations of energy tI~rms H, G, A, and U.

internal energy U, and flow or external energy PV, or it can be divided into
energy available for useful work G, and Imavailable energy TS. A similar
function is the work function A, defined as: A = U - TS. Basic Energy
Relations. The following equations summarize the basic relationships of
thermodynamics:
7
THERMODYNAMICS IN UNIT PROCESSES
 t.U = q - w
dU = dq - dw dS = dq,/T H
=U +PV A = U - TS G = H
- TS t.H = t.U + t.(PV) t.A = t.U
- t.(TS) t.G = t.H - t.(TS)

t.S = ! dq,/T

Usually an enthalpy change is a constant-pressure process, changes in work

function are for a constant-temperature process, and changes in free energy are
for a process occurring at constant pressure and temperature; so thetle
equations are usually written
t.H = t.U + Pt.V
t.A = t.U - Tt.S
t.G = t.H - Tt.S
Chemical EqUilibrium. Two important things that a chemical engineer wis~es
to know about a chemical reaction are:
Can it go?
How long will it take?
The first question is one of thermodynamics; the second is one of kinetics. The
prediction of chemical reaction equilibria is one of the most useful aspects of
thermodynamics. It is possible to calculate the equilibrium conversion of a
given reaction from data taken on other reactions or from thermal data on the
individual substances involved.
Free Energy As a Criterion of Equilibrium. The decrease in free energy of a
system during any isothermal, isobaric process is a measure of the net
reversible work (Wj) that the system can do upon its surroundings. WI = -t.G
Equilibrium conditions are those in which the forces resisting the process
are just balanced by those which cause it. Therefore, any small change that
would take place would have to be a reversible one if no dissipative actions are
involved. Since a reversible process is one which takes place always at
equilibrium conditions, the criteria of reversibility must include the criteria for
equilibrium conditions at constant pressure and temperature.
Since for a reversible process at constant pressure and constant tempera ture
-dG = WI
and since
WI = 0

when only the work of expansion is involved, then if the process is at equi
librium for a differential change which it might undergo,
dG = 0
UNIT PROCESSES IN ORGANIC SYNTHESIS
8
As most chemical reactions of industrial importance are restricted to con stant
pressure and are not harnessed to produce useful work, the free energy is
useful in calculating the composition of the equilibrium mixture.
Equilibrium Constant for Ideal Gases. When a chemical reaction takes place
at constant temperature and pressure, the free energy of the system continues
to decrease as long as the reaction proceeds spontaneously. When the free
energy has reached the lowest value possible for the system at that temperature
and pressure, the reaction has come to equilibrium. Any further reaction in the
same direction would have to be accompanied by an increase in free energy.
Therefore, it could not occur unless some extra force (such as, for instance,
electromotive) were applied to the system. In most reactions there is no extra
force applied, and the reaction stops at the point where the free energy reaches
the minimum value. The composi tion of the mixture of reactants and products
is the equilibrium composition. It is this composition that chemical engineers
are interested in calculating for the purpose of estimating the maximum
possible yield of a process.
Development of the Equilibrium Equation. The actual value of the free
energy at this point has little practical signitlcance. However, we shall see that
through use of the free-energy relationships it is possible to calculate the
equilibrium composition.
Let us consider the application of these principles to a specific chemical
reaction. The water gas shift reaction
CO(g) + H,O(g) ;:::! Co.(g) + H_(g)
is one of universal interest to engineers. A chemical reaction may be
generalized as follows:
aA + bB ;:::! cC + dD

The standard free-energy change for this system may be calculated from the
standard free energies of formation (I:J.Gj) of the individual components:

All chemical thermodynamic texts! derive the relationships between the

standard free-energy change and the equilibrium constant K, which is -aGO =
RT In K
The equilibrium constant for the generalized reaction may be written in
terms of partial pressure p if the components are behaving as ideal gases; or
thermodynamic functions as activity a or fugacity f may be used for nonideal
systems.
,d

•b
K '" pc PD for ideal gases PA ps
I See footnote 1, p. 1.
THERMODYNAMICS IN UNIT PlWCESSES
9
Equilibrium Composition of N onideaI Gases. In order to provide equa tions
which would be useful for all substances, Lewis and RandalP defined a
 term,fugacily. This is an arbitrarily defined term that may be substituted for
pressure in the preceding equation and which will make it applicable for all
substances. The mathematical definition of fugacity .f is dG = RTdlnf
Since the numerical value of the fugacity of a substance is often difficult to
calculate, the ratio of the fugacity at one tltate to that at some other arbitrarily
defined state is often used. The second such condition is called the standard
state. The ratio is called the activity and is defined thus:

a=L
r
where r refers to the fugacity of the substance at the standard state. Although
any arbitrary standard state may be chosen, custom has limited this choice. In
the study of chemical equilibria involving gas reactions, the standard state is
chosen as that ill which the fugacity of the pure gas is equal to unity. This for
most gases means th{~ pure gas at 1 atm pressure . . The standard-state
temperature must be the flame as that of the system for which we are
calculating the equilibrium composition. (This must not be confused with the
standard thermochemical temperature of 25°C =
298. 15°K.)
Calculation of the Fugacity of Gases. In the application of the chemical
equilibrium equation to nonideal gases, it is the fugacities and not the partial
pressures that are important. Therefore, we must have a means of relating the
fugacity to tlle partial pressure in order to use the equation. Fugacity is really
another term which is a measure of the deviation of a gas from ideality. Since
this is the case, there must be a relationship between fugacity and the
compressibility factor

PV z = RT

which is another measure of this deviation.

The Law of Corresponding States. When 8 gas is nonideal, PV does not
equal RT. However, the quantity RT can be multiplied by a correc tion factor
so that for 1 mole of gas
PV = zRT

This correction factor (the compressibility factor) varies with both tem perature
and pressure for each gas. If there were no relation between the factors for
different gases, it would be necessary to have charts or tables showing z as a
function of temperature and pressure for each individual
1 LEWIS and RANDALL, "Thermodynamics," McGraw-Hill Book Company, Inc., New York,
1923.
10 UNIT PROCESSES IN ORGANIC SYNTHESIS

gas. This would make the use of the compressibility factor impractical.
However, it has been found by experiment that when gases are at the same
reduced temperature (Tr = T ITe) and reduced pressure (Pr = PIPe) their
 compressibility factors are approximately the same. The degree to which a gas
deviates from ideal behavior, therefore, depends upon its
reduced temperature and pressure. At equal reduced conditions the devia tion is
the same for all gases. This rule is an empirical one which has been found to
hold with a fair degree of accuracy for most gases. It is called the law of
corresponding states.
The principal value ~f the law of corresponding states is that it makes
possible the construction of charts or tables which give the value of the
compressibility factor for any gas as a function of reduced temperature and
pressure. The data are usually presented in chart form, with z plotted against
Pr and with Tr as"the parameter.l Such a chart is shown in Fig. 1-2.
The law of corresponding states may also be used in the construction of
charts or tables which give the deviation of enthalpy, entropy, or heat capacity
from that of the ideal gas as a function of reduced temperature and pressure.
One of the most important uses of this law is the development of a
mathematical relationship between the fugacity f of a nonideal gas and its
compressibility factor z. From this relationship, charts have been made2 which
permit the calculation of the fugacity of those nonidcal gases which obey the
law of corresponding states. A chart of this type is shown in Fig. 1-3. The
ordinate is the ratio of fugacity to pressure (f I P) and is called the fugacity
coefficient (II). This is plotted against reduced pressure, with reduced
temperature as a parameter. The fugacity-coefficient chart may also be used to
calculate the fugacity of individual components of gaseous mixtures and also to
calculate the fugacity of a pure liquid at its vapor pressure.
Equilibrium Constant for N on ideal Gases. When the equilibrium con stant is
applied to nonideal gas mixtures, the fugacities are used instead of partial
pressures. Thus, the equilibrium equation for the reaction aA + bB ~ cC + dD
becomes

but since
where

1 NELSON and OBERT, Chem. Eng., 61 (7), 203-208 (1954).

I HOUGEN and WATSON, "Chemical Process Principles," part II, p. 622, John Wiley & Sons,
Inc., New York, 1947.
-]

.,. . ,'
~11O .
~
J.,. loL IOCX)-
, -
-
·N
 < ~
-
~1 ~
~ t \
\ , \
>

'

0
"

.~

~
0
·
-

~
.,
,
~

" ~ . -,
t u
' f \1
";-
\
E
:-
f-~ , ,\
~
,\

I-~ 0 ~ ,,'I ~ .~-

~
-
,\
-
\
I-~ . N
0-
f~-
f-~

"
~~

+- L 0
~_
r-~t "
5

I-~ .::-

t-~ ~
a:

\
~
 L
' ~ I-
~
~ '" L

+-
"

1 ~~
,\'

{
~ ~
)~~
0L

<t: ~
E
1(1)
I()"':
IT CI>
+-
..
Ij ,

... : ..,

~'\
I I/ / ~ '"
I()
"
1 ..,
\

/11 / V \ '" a::

3' '"
r<>

II
I j V °i~ a
t .~

J.-..-----'
L +-

I V
I
V
,

I I <
V -:
::
:-

rt f f
I
(
)
U Z '
l
"
 (

i/ 0
1
I
W 1
)

_
t o
~ '
N
g
;
fl'
J

f
j
V
O
,~
~/
~

I
~

f
2

N
.:
~u l.~ It) N
of 00
.... Q)
.p 0.., N Q)

:Q S
';;;
on
.. L ... Iii o '" .... '" p.
Q)
<.)
 oj
;::s
.... .., ~.
"0 Q)
Q
'" .... Q)
..,
p.
<XI
"'" 0
S (I) 0
Q)
0.
.., ....
"0 Q)
10 '-' ;::s <D
0
.., 0 '-' .;:
;>.
co ,~
"0 .... :;:::
v
~ P
v '~ o m U1
... Q) Q)

.... ;::s ....

c. .; U1 p.
E0
U1
L
Q)
" .... .. p. .. '" ... "0 U
NO. Q)
N
'-'
;::s
.., "0 ci ~
c:i
'" Q) ~
u ....

.., ..,
" '" UJ
<:: N
a:: ·OJ
bil d

11
~

I'
~~\~ \
"~
.
~~ .
\

\ II - I-- - -

\
i' "'~ "'-
-
~ t"\ 1'\

i'
-
i'\.
I'. I I

~~~ l\l\~ \~'\

'
\
f\

. / II / /
~"2:\ ""'..c
\
,~ ~
 /I
/I
In!.! I / ,/1 I /
/I/
I}
/ /.1
III i//'III
'/,W 1/ I I / / 0
.....

I'
'V
0

~ '"" ""I

V N

i/ A
'!!

9
<:11

/ CD

,._

1/
'\ 0~/ /
lD

I{')

\)/}'/I l,
v

//I-Y!r<i

I IliX7;\)- ~ / / / (\J

\)1) It: I'll " I{')

I
9
C7l
/
I I /I 0
CD
Ci
,._
Ci
/) lD
/1 Ci
I{')
,I 0

/ i / I/' /1Q;.l' v 0

..... 0
 N
.;
::!
'" OJ
'" f: 0-
:::s
"0
'" " "0
<>.,..
'"
...
" .; ,";'
> "'>
.., <= e?-
·S 5l
IS
VI
'" e?
0
a. '0 " >, .. '-' ~ '0 " " oj N I{')
omcol"-U)L(') ~ rtl ":0000000 o
N

.. a: "" ::! d/I:II 'IU3!~!jjaO~ ~1!~D6n.:l

....
"0

.!:2 '" '"§

'" 12
<= '" " c? .....
_0 o
ci
~
13
THERMODYNAMICS IN UNIT PROCESSES

this may be written

eedd
K = "CPCI'DPD
dab b
VAPAI'BPB

cdc d c d
VAVBPAPB PAPB
or
K = VCVDPCPD = K" PCPD a b a b a b , d
where Kv = PCI'D • b by definition
PAPB

Here Kp is a direct measure of the effect of nonideality of the gases upon the
equilibrium composition.
Calculation of the Equilibrium Constant. The equilibrium constant, K, for a
chemical reaction can be calculated from the standard free-energy functions or
from changes in enthalpy and entropy for the process. Data will be found in the
references listed at the end of this chapter. Since,
-flGT = RTlnK
and
flG flHT - TflS;
T=
for a process occurring at constant temperature and pressure, data must be
available for the temperature at which the reaction takes place. If
values of AG
r
are not available, calculate them from enthalpy and entropy
data. t:..Hr was calculated on p. 5. ASr. can be calculated III a similar
manner:
!lST - flSs + f dT
flC: -T
Ts
tlSs = ~(SO)p,oduc" - ~(SO) , ... ta ...

T
The standard entropy change of the reaction is calculated from the values of
the absolute entropy of the products and reactants at the standard temperature
T s = 25°0.
The calculation of absolute entropy of many substances is possible be cause
the substances follow the third law of thermodynamics. This states that the
absolute entropy of a pure substance in the crystalline state is zero at the
temperature of absolute zero. Thus, if the latent heats X of all the phase
changes are known and if the heat capacity is known, the absolute entropy of
the pure substance may be calculated by the following equation:
so = l:JQe dT + 2: ~
TT
Entropies calculated in this manner were compared with absolute entropies
computed backward from measurable equilibrium constants and were found to
check quite closely.
If the standard free energy of the substances involved in a chemical I'C-
14 UNIT PROCESSES IN ORGANIC SYNTHESIS
action were tabulated, the standard free-enel~gy change and, thus, K for a given
reaction could be calculated directly.
Calculation of Equilibrium Conversion. In practically all engineering
calculations, the evaluation of the equilibrium constant is merely a means to an
end. The ultimate goal is the calculation of how far a reaction can proceed-the
equilibrium conversion.
Single Isothermal Reaction between Ideal Gases. One of the simplest cases
involving a calculation of equilibrium conversion is for an isothermal reaction
between ideal gases. Consider the reaction '.'j' A~R+S ,(
\vhere A, R, a.nd S are ideal gases. i, The first step is to calculate K. The method
involving the standard heat of reaction and the standard entropy change is
usually used because these data are available for most substances. In the tables
giving standard heats of formation (01' standard heats of combustion) and
absolute en tropies, the standard state is usually specified as unit fugacity for all
gases and is usually stated as the pure gas at 1 atm pressure. The ealculation
must be based upon the standard state specified by these tables and is not
arbitrary, as many people are led to believe. If the standard state for each
component is that of unit fugacity, the equilibrium equation becomes
K = EE'B!.
PA
It is necessary to express each partial pressure in terms of the same
variable-degree of conversion. If the initial gas consists of 1 mole of A, the
 conversion x at equilibrium may be expressed as follows: A->R+S
1-xxx

BASIS: I MOLE OF A IN FEBD

I
Moles I Mole fraction
I-~-X/l~ I - x x X/I + x
x X/I + x

1+ x
1.0

Constituent Partial pressure

----------
A 1f(J - xl/I + x
R 1fx/l + x
S 1fx/I + x

- where 'If' is defined as the total pressure on the system. Thus, the equilib
Total. " ,

rium equation becomes

K = 1f(x!_!_±2l~~/l + xl
1f(1 - x/I + x)
or or
x = ~1f !K
K = _7rX2_ I - x'
THERMODYNAMICS IN UNIT PROCESSES 15

Correction for N onideality. If the pressure were so high that the gases did
not behave ideally, the equation would have to be corrected for non ideal
behavior. Since the standard states are still unit fugacity, the equilibrium
equation would be

- j; - ---;;PA
K _ iRis _ VRPR vsps

The fugacity coefficients (v) would have to be evaluated. Since in this case
Kv = VR VS
VA

the equation becomes

K PRPS
R-" = P:;-
Solving for conversion a$,in the previous example,
x = - ,_ K/!S!'__
VII" + K/Kv
The only difference between the nonideal case and the previous case is that the
ratio K/Kv replaces K in the equation for x.
Reactions InvollJing Gases and Solids. We have demonstrated how to
calculate the equilibrium composition for a reaction involving all gases. Let us
consider the case where a gas reacts with a solid such as in the reaction
C + CO,;::::!2CO
Here

K = -- = -----
Pco v'coP'eo f co, veo,Peo,

The term fe does not appear for carbon because the carbon remains in the solid
state right up until the time it reacts and, therefore, is present only in its
standard state, that is, as the pure solid at unit fugacity. It so happens that, in
the tabulating of enthalpy and entropy data for solids, the standard state used is
the pure substance at the standard temperature. The calcula
tion which corrects t::.HO and t::.So for temperature automatically provides for
the correction of the fugacity of the solid from the standard temperature to the
temperatute of the system. Thus, the standard state for solids is the temperature
of the system at one atmosphere pressure. However, the fugacity of solids is
almost independent of pressure, and so the fugacity of a solid at the pressure of
the system is for all practical purposes the same as that at 1 atm. Thus, the
activity of any solid (carbon, in this reaction) is practically unity, and the term
does not appear in the equilib rium equation. It should be noted, however, that
the thermal properties of carbon must be used in calculating the value of -t::.Go
for the reaction and, hence, the value of K.
16 UNIT PROCESSES IN ORGANIC SYNTHESIS
Simultaneous Reactions. When two isothermal reactions take place
simultaneously, the calculation is slightly more involved, but may be solved
readily. The solution of this type of problem involves two un knowns. The
procedurc is to set up the equations involving the equilibrium constant for each
reaction and to solve the simultaneous equations graph ically.
Assume that the following two idcal gas reactions occur simultaneously and
are isothcrmal:
A+B~R A+R~S K,=~ PAPS K2 (1) (2)

=..E!_ PAPR

Compound A is the limiting reactant, B

the excess reactant. r = molc ratio of B BASIS: 1 MOLE OF A IN FEED
to A in thc feed Component Moles at equilibrium Mole fraction
x = drgrce of convcrsion of A by reaction 1-x-y
(1) A l-x-y
y = dcgrce of conversion of A by
reaction (2)
 Total. .......... l+r-x-y

l+r-x-y
r-x x-y x-y
B r-x 1+r-x-y R l+r-x-y
Y
Sy
, .i

l+r-x-y

In the case where the total pressure is 1 8,tm, the partial pressurelf'sre cqual to
mole fractions and the two equations become
K, = (x - y! (1'+ r - x - y)
(r - x) (1 - x - y)

K. = .J!_(1 + r - x - y)
(x - y) (1 - x - y)

Rillcc the K's are constant and may be calculated from thermal data and r i"
usually spccified, the two cquations can be solved simultaneously. The easiest
method is, generally, to plot y versus x in both equations, and th(' point of
intersection gives the desired x and y values.
Adiabatic Reactions. In the cases thus far considered, we have shown lto\\, to
calculate the equilibrium conversion for reactions that take place
THERMODYNAMICS IN UNIT PROCESSES
17
at a constant temperature. In many industrial practices, no heat is added or
removed during the process and the heat of reaction must originate from or be
dispersed in the reacting system itself. Thus, the temperature of the reacting
mass will change continuously during the reaction, and a calcula
tion of the equilibrium composition will also involve the calculation of the final
temperature. This type of problem is solved most easily by a graphic method.
The method of soJutjon for adiabatic reactions presented here will be for the
simplified case in which it may be assumed that ilH~ and ilS~ are relatively
constant over the temperature range involved. If this is the case, a plot of log K
versus liT wiII result in a straight line. The calcula
tion involves two relationships between T and x, the degree of conversion. One
is a simplified energy balance; the other is the relation involving the
equilibrium constant. The procedure is as follows:
1. By methods previously discussed, calculate K for two temperatures and
plot log K versus liT. This will be a straight line if ilHO and ilSo are assumed
to remain constant.
2. Assume even values of conversi(m, x, from 0 to 1, and calculate the
corresponding partial pressure of each gas present for each value of x. 3. Using
these ratios of partial pressures, calculate the values that K would have for each
value of x.
4. Using the log K versus liT
plot equt//bflum
plot, evaluate T for each value of x
conversion vs. T --!J'---I-'-t---l--l plo/
and plot x versus T (line ab in Fig. odlaba/ic
1-4). reac/ian T vs. x

5. Use a simple energy balance to

obtain the other relationship be tween
T and x.
xl1H = Cp (1' - To)
where Cp = molal heat capacity
LiH = heat of reaction

From this equation it is seen tIilt.t.t

Temperature, T
x=O T = To
FIG. 1-4. Adiabatic reactions: equilibrium
and where x = 1 conversions vs. temperature, adiabatic T vs.
Ad/aba/lc reac/ions x.

The line connecting Tl and To is drawn in.

6. The intersection of the two lines gives the desired values of x and T. This
procedure is illustrated in Fig. 1-5.
18 ..:
~ '.600

~
~ 1,400 ~

UNIT PROCESSES IN ORGANIC SYNTHESIS

~
 some specific methods of application.
These practical applications gen-
Thermodynamics and Unit Proc
°E~uilibr~: con~~~ion, ~8 1.0 erally involve three
esses. The major purpose of pre
senting a review of thermodynamics phases of ther modynamics. They are
principles in this volume is to em thermal
FIG. 1-5. Equilibrium conversion.
phasize the importance of thermo
effects, chemical equilibrium, and
dynamics to a fundamental under
physical equilibrium. Thermal cal
standing of unit processes. It is,
therefore, appropriate to discuss
culations involve only the first law of thermodynamics and are used for
calculating:
1. Heat of reaction and the effect of temperature on heat of reaction. 2.
Sensible heat transferred in preheating and cooling.
3. Heat of solution and adsorption.
4. Heat effects in phase transformations (vaporizing, melting, and
crystallizing) .
Chemical-equilibrium calculations, the subject of this chapter, are probably
the most important application. They involve: 1. Calculating the equilibrium
constants of both the major and secondary reactions of a process.
2. Calculating the effect of temperature on these constants. 3. Calculating
how far a reaction can go at a given set of conditions. 4. Calculating the
concentration of the desired product in the reactor
effluent for various process conditions. (This applies to fast reactions.) 5.
Determining the effects of solvents, inert diluents, or recycle streams on
product yields.
Physical equilibrium is also a very important aspect of thermodynamics
which applies to the chemical-process industries. It is more often used,
however, in those phases of a process which involve the separation and puri
fication of the product once it is formed. The physical-equilibrium portion is
applied almost entirely to the unit operations. It is used to calculate phase
compositions, vapor-liquid equilibria, absorption, adsorption, ex traction, and
distillation and to estimate solubilities. This phase of chemi cal engineering is
covered adequately in other texts.
Chemical Equilibrium Applied. The major use of the principles of this
chapter is to calculate the equilibrium composition of a system. This ap
plication in vapor-phase homogeneous reactions has been discussed above.
THERMODYNAMICS IN UNIT PROCESSES 19

But what about other systems? Do the methods equally apply? The answer is
"yes," provided that certain factors are considered. Vapor-phase Catalytic
Reactions. Since a catalyst does not change the equilibrium of a reaction, the
calculated composition should be the same with a catalyst present as without.
This is generally the case, but there are exceptions to bear in mind. One is the
case of a possible competing reaction which is too slow to have any effect in
the homogeneous system but which is speeded up by the catalyst to a greater
 extent than the main reaction. Consider, for example, the reactions
A;=!R
with the potential side reaction

Suppose that the equilibrium conversion of the first reaction is 98 per cent and
that the second is nonreversible but infinitely slow in the vapor phase. The
yield will be 98 per cent. If the catalyst speeds up both reactions so that the
second reaction is about the same speed as the first reaction, the yield will
drop. The chemical-equilibrium calculation would not have predicted this since
the second reaction was not considered.
Liquid-phase Reactions. There are times when the calculated equilib rium
composition of a liquid-phase reaction does not compare at all with the actual
measured value. This does not mean that the theory is wrong -just that there is
not enough of it. Here is the reason. The rigorous equilibrium equation for the
reaction
A+B;=!C+D
in the liquid phase is

where the a's represent the activities of the components. But, since we do not
know the actual activities, we generally use the expression

where the x's are the mole fractions. This means that we are assuming the
mixture to be an ideal solution. If the solution really were ideal, the cal culated
composition would match the measured one. If the solution should deviate
much from ideality, the calculated composition will not be correct.
Theoretically this could be taken care of by using the activity coefficient of
each component. Unfortunately, our knowledge of physical equilibrium has not
progressed far enough to allow us to calculate activity coefficients of complex
liquid mixtures (more than two or three com ponents).
20 UNIT PROCESSES IN ORGANIC SYNTHESIS
Thermodynamics Is Not Infallible. Chemical-equilibrium calculations are
useful, but they do not always give the whole answer. If a chemical reaction is
fast, thermodynamics calculations may tell all we need to know about the
reaction. In these cases the rate of reaction is not important, just product
distribution and temperature control. If the reaction is not possible at all or not
economically feasible because of too Iowa product concentration in the reactor
effluent, thermodynamics can tell us so.
If, on the other hand, the reaction is not extremely fast, if the time spent in
the reactor is important, or if rapid competing reactions are involved, the
chemical engineer can turn to another tool-kinetics.
 SOURCES OF THERMODYNAMIC DATA

Thermodynamic data must be sought from many sources. The American

Petroleum Institute and Manufacturing Chemists Association have large
projects to determine and collect thermodynamic data. Several govern mental
agencies, such as the National Bureau of Standards and the Bureau of Mines,
determine, collect, and publish thermodynamic data. Some of the best and
most readily available sources of thermodynamic data are listed below.

Bichowsky, F. R., and F. D. Rossini: "Thermochemistry of Chemical Substances," Reinhold

Publishing Corporation, New York, 1936.
Din, F. (ed.): "Thermodynamic Functions of Gases," vol. 1, Ammonia, Carbon Dioxide,
Carbon Monoxide; vol. 2, Acetylene, Ethylene, Propane, Argon, Butterworth's Scientific
Publications, London, 1956.
General Electric Company: "Properties of Combustion Gases," vol. 1, Thermodynamic
Properties; vol. 2, Chemical Composition of Equilibrium Mixtures; McGraw-Hill Book
Company, Inc., New York, 1956.
Hilsenrath, J., and others: "Tables of Thermal Properties of Gases," Natl. Bur. Stand ards Cire.
564, 1955.
(Additional single sheets issued as NBS-NACA sponsored tables, also as NACA
Technical Notes.)
Hottel, H. C., G. C. Williams, and C. N. Satterfield: "Thermodynamic Charts for Com bustion
Processes," 1. Text, II. Charts, John Wiley & Sons, Inc., New York, 1949. Keenan, J. H., and F.
G, Keyes: "Thermodynamic Properties of Steam," John Wiley & Sons, Inc., New York, 1936.
Keenan, J. H., and J. Kaye: "Gas Tables," John Wiley & Sons, Inc., New York, 1948. Kelley,
K. K., and others: Contribution to the Data on Theoretical Metallurgy, U.S. BUT. Mines Bull.
350, 371, 383,384, 394, 406, 407, 434, 476, 477, 542, 1932-1954. Kobe, K. A., and others:
"Thermochemistry for the Petrochemical Industry," Gulf Publishing Co., Houston, Tex., 1952
(A collection of a series of 19 articles that appeared in the Petroleum Refiner from
January, 1949, to December, 1951. Additional papers have appeared at intervals since
1951. Tables of heat capacity, mean heat capacity, and relative enthalpy are given in DC,
OK, of, and OR. Other thermodynamic data are given.)
Rossini, F. D., and others: "Selected Values of Physical and Thermodynamic Prop-
THERMODYNAMICS IN UNIT PROCESSES 21

erties of Hydrocarbons and Related Compounds," Carnegie Press, Pittsburgh, Pa., 1953.
(Lithographed sheets are being issued at intervals by API Project 44 to complete and
extend this material.)
Rossini, F. K., and others: "Tabl!'s of Selected Values of Chemical Thermodynamic Properties,"
Natl. Bur. Standards Circ. 500, 1952.
(Lithographed sheets are being issued periodically.)
Sage, B. H., and W. N. Lacey: "Some Properties of the Lighter Hydrocarbons, Hydrogen Sulfide,
and Carbon Dioxide," American Petroleum Institute, New York, 1955. Sage, B. H., and W. N.
Lacey: "Thermodynamic Properties of the Lighter Paraffin Hydrocarbons and Nitrogen,"
American Petroleum Institute, New York, 1950. Stull, D. R., and G. C. Sinke: "Thermodynamic
Properties of the Elements," American Chemical Society, Advances in Chemistry, 1957.
Zeise, H.: "Thermodynamik," Band III/1 Tabellen, S. Hirzel Verlag, Leipzig, 1954. (Tables of
heat capacities, enthaipies, entropies, free energies, and equilibrium constants.)

The following special references are used throughout this volume. The Office
of Technical Services (O.T.S.), U.S. Department of Com merce, Washington 25,
D.C., has a complete record of all declassiQed re ports of the British Intelligence
Objectives Sub-committee (B.LO.S.), the Combined Intelligence Objectives
Sub-committee, SHAEF (C.I.O.S.), and the Field Information Agency Technical
(F.LA.T.). The O.T.S. can arrange for delivery of copies, photostats, or
microfilms of these reports for a fee. The Office of the Publication Board
(O.P.B.) reports are also avail able from the O.T.S.
CHAPTER 2

CHEMICAL KINETICS

By T. E. CORRIGAN AND JOHN J, McKETTA

What Is Chemical Kinetics? Chemical kinetics is a study of the rates of

chemical reaction and the effect that process conditions have on these rates.
These process conditions are temperature, pressure, and reactant
concentration. The application of kinetics in the selection and design of
commercial reactors is of specific interest in the study of unit processes.'
Assume in the following consecutive reaction that the desired product is B
and that C is valueless:
A->B-+C
A thermodynamic calculation may show that the reaction is not practicable
because at equilibrium there is an excess of C. Yet a kinetic study may show
that the rates are such that a high yield of B may be obtained by quenching the
reaction at the proper time. This is shown clearly in Fig. 2-1.
Review of Basic Principles.2 The mass-action law, in equation form, states
that for the reaction
A+B-+R
the rate of reaction may be expressed in the following manner: r =
kCACB
This law is an empirical one based on observation and, unlike the thermo
dynamic expression for the equilibrium constant, is not a derived equation
founded on theory.
Order of Reaction.3 One method of describing homogeneous reactions is by
the order of reaction, which is the sum of exponents of the concen tration
terms. In the equation
r = kC1C~
1 CORRIGAN, Chem. Eng., 61, July, 1954.
2 CORRIGAN, Chem. Eng., 61, August, 1954; LAIDLER, "Chemical Kinetics," McGraw Hill
Book Company, Inc., New York, 1950; DANIELS, "Chemical Kinetics," Cornell University
Press, Ithaca, N.Y., 1938; FROST and PEARSON, "Kinetics and Mechanism," p. 153, John
Wiley & Sons, Inc., New York, 1953.
3 Nomenclature for Chaps. 2 and 3 will be found at the end of Chap. 2. 22

CHEMICAL KINETICS 23

the order of reaction is a + b. The molecularity of a reaction is the num ber of

reactant molecules in the stoichiometric equation. If the reaction is A+B->R
and the rate is
r = kCACB
the molccularity is 2 and til(' ol'llrr of reaction is lYz. The order of a re action is
not determined by tlw molecularity. Chemical reactions in which the order of
reaction is the same
as the molecularity arc called sim
ple-oreler reactions.
Mechanism. A reaction usually
does not· occur in the singk step
which may be representeel by the
over-all stoichiometric equation but
rather in a series of steps which
add up to the over-all equation.
These steps are the A + AB~A2B A2B~AB +R
"mechanism" of the AB~8
reaction. For example,
2A+B~R+8
the reaction (Step 1)
2A +B~R +8 (Step 2)
might take place as (Step 3)
follows: (Step 4)

c:
o .;::
o
E c: ~ c: o u

A +B~AB
The slowest step controls the rate of FIG. 2-1. Kinetic data may show that by
the reaction and may also de termine quenching a reaction economic recoveries are
the form of the over-all rate equation. feasible.
The rate equation,

rather than the mechanism, is important to engineers.

First-order Reaction. If the following is a first-order reaction A->B
a plot of t:..CA/l::..t against CA will give a straight line since the rate equation for
a first-order reaction is
(1) (2)
_ .!!:._CA = kC
A dt
Equation (1) integrates to the form In CA
= -kt +I
24 UNIT PROCESSES IN ORGANIC SYNTHESIS
where I is integration constant. Thus, if the reaction is first order, a plot of the
term In CA versus time will give a straight line of negative slope. Second-order
Reaction. If a chemical reaction has a rate equation of the following form, it is
of second order:
 _ dCA = kcl
dt (3)

A plot of ilCA/ilt against C! is a straight line. It is better to use the integral

form of Eq. (3):

or (4)

Thus, a plot of I/CA against t will give a straight line with a slope equal to k.
For a reaction of the form

with the rate equation

, - dftA = kCACS (5)

the integral form may be expressed in terms of conversion (~) and initial
concentrations; thusl
_I_In b(a - x) = kt
a - b a(b - x) (6)

A plot of the term I/(a - b)ln b(a - x)/a(b - x) against t will give a straight line
with a slope equal to - k. If a and b are equal, the equation reverts to the
integrated form
x = kt
a(a - x) (7)

Thus, a plot of x/a(a - x) against t would give a straight line with a slope of k.
Third-order Reaction. If the reaction
A-->B

were a third-order reaction, the rate equation would be

_ dCA = kC~
dt

In terms of conversion, the rate equation is now

(8)
dx a dt = k(a - x)
1 FROST and PEARSON, op. cit.
25 (9)

and the integrated form is

CHEMICAL KINETICS ",:2=;;a'i-x_-__::.x'''' = kt
2a'(a - x)'

A plot of the quantity (2ax - x2)/2a2(a - x)2 against t would be a straight line.
Zero-order Reaction. In some chemical reactions the rate of reaction is
independent of the reactant concentration. The rate equation would be
_ dGA = k
(10)
dt

and a plot of reactant concentration against time would be a straight line.

General Case. The general rate equation for a reaction involving a single
reactant is

dGA = kG;'
dt

or dx = k(a - x)" dt

where n may be any positive number. In mOfSt cases, n is a fraction. The

integral form of this equation1 is

1 (1 1 ) (n - 1) (a - x)" 1 -;z;;=I = kt (11)

Limitations of Methods. All the plots discussed above for determining the
order of a reaction are valid either for nonreversible reactions or at low values
of conversion.
The methods which can be used to determine the order of reaction may be
summarized as follows:
1. Plotting AC AI At versus concentration.
2. Calculating k from the data and takin~ the order which gives the most
nearly constant value of k.
3. Plotting the proper integral function for each order and selecting the one
which gives a straight line.
Table 2-1 is a summary of the differential and integral equations for each
order of reaction.
Effect of Temperature on Chemical Reactions. The basic reaction-rate
equations discussed above are defined only by concentration and time. The
effect of temperature appears only in the variation of k.
The relationship between the reaction-rate constant (k) and temperature
was first observed by Arrhenius to be

1 LAIDLER, op. cit.

 H

. H
I
~

~~ ~
I
III
~~ ~
..:c..:c..:c :;;;-
II II II II

·'. I

26
CHEMICAlj KINETICS 27

When In k is plotted against liT, the values of EIR and of 8 may be determined
 from the slope and intercept of the line.
Types of Chemical Reactions.! Some of the classifications of reactions deal
only with the chemistry of the reactions, and some arc concerned with the
environment in which the reaction takes place.
Some of the criteria which arc used in classifying reactions are: 1.
Heterogeneous or homogeneous reactions. This classification refers to the
number of phase'S in the reacting system. A homogeneous reaction is one which
takes place in only one phase. A heterogeneous reaction in volves more than one
phase.
2. Me~hanism of reaction, such as chain reactions, regular molecular reactions, or
photochemical reactions.
3. Catalytic or non catalytic reactions.
4. The method of operating the reactor, such as adiabatic, isothermal or
nonadiabatic-nonisothermal.
5. The stoichiometric chemical equation. Reactions may be divided in this
mannel' into -five major types: simple reactions, parallel reactions, series
reactions, complex series reactions, and reversible reactions.
Simple Reactions. A simple reaction is where only one reaction is con sidered
to take place. The more common simple reactions are represented by the
following types:
A -~R
A+B-->R
A-~R+S
A +B-,R+S
2A -; R
A -; 2R

A typical concentration-time curve is shown in Fig. 2-2 on page 30. Some

industrially important examples of simple reactions are: 1. A ---> R.
Isomerization of butane which is discussed in Chap. 14. 2. A + B ---> R +
S. Acylation of amines.
RNH2 + RCOOH --> RNIICOR + H20
RNH, + RCOCl -, RNHCOR + lICl

3. A ---> R + S. Production of vinyl chloride from ethylene dichloride.

ClCH2CH,Cl --> CH,=,"CIICl + HCl

When only one reaction is taking place, the only variables to be con-- , side red
are time and conversion. The reaction A ---> R is the simplest of this type. It
represents a molecular rearrangement. The basic rate equa tion for this reaction is
1 CORRIGAN, Chem. Eng., 61, September, 1954.
_ dCA = kC:;'
dt (12)

28 UNIT PROCESSES IN ORGANIC SYNTHESIS

where n is the order of reaction. The integrated forms of the equation for
various reaction orders are given in the accompanying table.
 Order Ditferential form Integral form
In CAo/CA = kt
o -dCA/dt =
(l/CAo) - (I/CA) =
1 -dCA/dt =
2 -dCA/dt = 2(Cj. - cj) = kt
kt
-dCA/dt =
~ n
-dCA/dt =
_l_{_1
___ l_}
CAO - CA = k
'" kt n-l C::.-'
CA.;'
If the rate for a reaction involving two reactant· molecules, such as the
reaction A + B ---7 R, is given by the equation
_ dCA = kCACB
dt
the equation can be integrated into one in terms of conversion and time only,
as shown in Eq. (6).
Parallel Reactions. Parallel reactions are those in which more than one product
is formed by separate reactions and where the products, once formed, do not
react again. Some typical parallel reactions are {A -+ R
A -+S

{A+B-+R
A+B-+S
{A -+ R
A+B-+S
An example of a set of parallel reactions of the type
A+B-+R
A+B-+S
A+B-+T
is in the nitration of toluene, which is discussed fully in Chap. 4.

O+HN~~&~

O CH. + HNO.-+ OC~'

~ NO,

CH, CH,
 O+HN~-O
NO,
CHEMICAL KINETICS A-.S
The basic rate equations for the 29

reaction {A -.R

dCs = k.C:; dt

- d~A = k,CA + ksC:

are
dCIl _ kC" (13)
Cit - , A
If the reactions are both the same
(14)
order, dCA = -(k, + ks) C;. dt
If both reactions are first order, (15) (16)
b
CA = CA.e-
(17)

where b = (k, + k,)t

The concentrations of Rand S are related by the simple ratio of the rate
constants
CR - CR. k,
Cs - C
s• = k;
Figure 2-3 shows a concentration-time curve for a parallel reaction. Series
Reactions. Series reactions are those in which the product of the reaction goes
on to react further. The simplest example is A ..... R ..... S
An example of commercial interest is the liquid-phase chlorination of
benzene.
From a typical plot of concentration versus holding time, it can be seen that
the concentration of R passes through a maximum value and also that there is
an inflection point in the curve for the concentration of S (Fig. 2-4).
The rate equations for reactions of this type are
_ dCA = k,C
A dt
 d~1l = k,C dCs _ k
A - k.C'il
C ..
(if - • Il
, (18)
30 UNIT PROCESSES IN ORGANIC SYNTHESIS
Integral reaction-rate equations have been worked out for the case of
first-order reactions. These are discussed in the standard texts on physical
chemistry.
The first-order rate equations are
Integral
Differentia!

_ ~CA = klCA dt

Time_
Fi g. 2-2 Cs = CAo - CA - CR

Complex Series Reactions. Complex

series reactions are those in which there
is further reaction between one of the
reactants and one of the products so
that both series and parallel reactions
take place simultaneously. An example
Time_ of this is
Fig. 2-3

Time_
fig.2-4
FIGS. 2-2 to '2-4. Time-concen tration curves
for simple, paral lel, and series reactions.
de CAo
dt = klCA - k,Cn CR = ~ (e-k1 ' - e-k,,)
A+B->R
where K = kdkl for the case where neither A+R->S
R nor S is recycled. The concentration A +S->T
of S is determined by material balance
This is a series reaction with respect to
B, R, and S, but it is a parallel reaction
with respect to A. A concentration-time in the conversion of alkylene oxides to
curve is shown in Fig. 2-5. the al kanolamines, which is discussed
An industrial operation exemplifying a in Chap. 8.
complex series liquid-phase reaction is
H,C---CH, + NH. ---> H,C--CH, +
'" / CH,OHCH,NH, -> HN
HOCII,CH,NH,
(Ethanolamine)
o
CH.CH,OH

"'/ o '" CH,CH,OH (:9iethanolamine)

CH,CH,OH CH,CH,OH
//

H.C~--CH. + HN -> N-CH,CH,OH "'if "'CH,CH,OH '"

CH,CH,OH (Triethanolamine)
CHEMICAL KINETICS 31

Among many similar liquid-phase reactions of the complex series type, the
following are shown later in this book:
1. Glycol ethers,
2. Dimethylol urea,
3. Glycines.
An example of a vapor-phase complex series reaction is the chlorination of
propane, which is discussed in Chap. 6.
A complex series reaction, such as the one above, is more complicated than
the corresponding series reaction. The reactant ratio CBo/CAo is an added
variable which is not present in the straight series reaction. The fact that A
reacts with each of the products as well as with B results in a more complicated
relationship between CA and time. The presence of two reactants makes the
control of the reaction easier than for a series reaction because in both cases, if
R is the desired product and S is valueless, the best way to obtain the highest
yield of R would be to run the reactor so as to obtain a fairly low conversion in
each pass and to recycle the unre acted A. If this l~'ere the case, the lower limit to
the conversion of A per pass in either case would be regulated only by the
economics of the reactor and recovery equipment cost. However, a high yield
could be obtained for either type of reaction.
If reactant A happens to
beamaterial which is so easily can be done to ob
decomposed that it cannot be tain a high yield of the desired product
recovered in a recycle system, nothing in the case of the straight series
reaction. In the case of the complex
series reaction, the yield of R can be
increased by using a high ratio of B to
__ R_

A in the feed.
Thus, higher yields of R can be
obtained for the complex series
reaction if B can be recovered and
recycled. If B cannot be recycled, the
same problem holds for the complex
series reaction as for the series
reaction.
Fortunately, in the com-
o
'" Complex Series

J~-----,~::;__
Time_
Fig.2-6
and 2-6. Time-concentration curves
FIGS. 2-5
for complex series and reversible reactions.
Time_
Fig. 2-5

l~~e=-==,,_
mercial reactions listed above, the compound B either can be recovered and
recycled (as in the reaction of alcohols with ethylene oxide) 01' is of low value
(such as in the reaction of water and ethylene oxide).
32 UNIT PROCESSES IN ORGANIC SYNTHESIS
Reversible Reactions. Reversible reactions are those in which the for ward
and reverse reactions take place simultaneously. Any of the above types of
reactions can also be reversible (Fig. 2-6). If the reaction is re versible, this is
taken into account in the rate equation. For a simple order reversible reaction of
the type
A+B:;:=R+S

the net rate equation can be written

(19)
where K is the equilibrium constant. The numerical value of K corre sponding
to concentration must be used in this equation. Examples of the reversible
reactions are:
1. Nitration of cellulose.
2. Water-gas shift reaction.
3. Suifonation of naphthalene.
4. Esterification of ethyl alcohol.
5. Alkylation of benzene.
Calculation of Conversion and Reactor Size.l The most important use of
kinetics for the engineer is in the calculation of reactor size. In the
batch-reactor calculations, the typical kinetic equations involving con centration
terms can be used without alteration. In this type of calcula tion it is usually
desirable to calculate either the length of time per batch for a reactor of a given
size or the amount of initial charge needed if the production rate is specified for
a given degree of conversion.
Heterogeneous Reactions. Heterogeneous reactions are those in which more
than one phase is involved. Some important types of heterogeneous reactions
of industrial interest are:
1. Gas-phase reactions promoted by solid catalysts.
2. Noncatalytic reactions involving gases and solids.
3. Reactions between gases and liquids.
4. Two- (or more) phase liquid reactions.
5. Reactions between liquids and solids.
The first of these is particularly important to the chemical-process and
petroleum-refining industries. An example is in vapor-phase catalytic cracking.
The second type is particularly important in fuel technology since the
combustion of any solid fuel falls in this category. The third type is very
important in the chemical-process industries. An example would be the
chlorination of a liquid hydrocarbon. Since the degree of contact of the two
phases is of major importance, reactors are designed by the same methods used
for absorption processes.
1 CORRIGAN, Chern. Eng., 61, October, 1954.
CHEMICAL KINETICS
33
Vapor-phase Catalytic Reactions.! When a gaseous reaction is promoted by
a catalyst, the reactants are first adsorbed upon the catalyst surface. The actual
transformation from reactants to products takes place in the adsorbed phase.
Therefore, the adsorption characteristics of the catalyst toward each of the
reactants and products are important factors in rate determination.
Adsorption may be either physical or chemical. The theory of physical
adsorption assumes that the adsorbed phase is a condensed liquid-phase layer
of molecules of the vapor on the solid surface. Chemical adsorption is the
chemical combination of a vapor molecule with a portion of catalyst surface.
This portion of surface is called an active center. In the applica
tion of adsorption to catalysis, chemical adsorption is of major importance.
Adsorption Equations.2 There are three important equations relating the
concentration of adsorbate on the solid with its partial pressure in the gas
phase. One of these is entirely empirical, while the other two have a
theoretical basis.
Freundlich's isotherm is the empirical relationship and is

where PA = partial pressure of A in vapor phase

CA = concentration of adsorbed material on solid

a and n are empirical constants

An equation of more interest in catalytic kinetics is Langmuir's isotherm CA

= k,PA

1 + k,pA
which was derived for the simplest possible case of chemical adsorption.
Other isotherms of a similar nature, but more complicated, can also be derived
for similar cases.
A general equation for a single adsorbate has been derived by R. A. Koble3
and can be written in general terms:

CA =
L'K'PA 1 + K'PA
where L' and K' are constants related to K and L (see below) and n is an
exponent which may be fractional, 1, or greater than 1. The special case where
n equals 1.0 is the Langmuir isotherm.
Catalytic Rate Equations.4 Catalytic rate equations based upon the
order-oi-reaction concept are not recommended because they do not ac
I HOUGEN and WATSON, "Chemical Process Principles," John Wiley & Sons, Inc., New York,

1947.
2 CORRIGAN, Chern. Eng., 61, November, 1954.
• CORRIGAN, Chern. Eng., 61, December, 1954; KOBLE and CORRIGAN, Ind. Eng. Chern.,
44, February, 1952,
• CORRIGAN, Chern. Eng., 62, January 1955; 62, February, 1955.
34 UNIT PROCESSES IN ORGANIC SYNTHESIS
count for adsorption on the catalyst. A rate equation for catalytic re actions
which has more theoretical basis is one which includes the effect of adsorption
of the reactants and products on the catalyst. Equations of this type were
derived by treating the adsorption and desorption as chemi cal steps.
Taking as an example the reaction
A~R+S

the following steps are encountered:

+1 Sl ~s +1
A +1~Al Al+l ~Rl+SI Rl ~R
K,=~ PAC! Kt ==
CRCS CAC!
 K. == PRC! CR (20) (21) (22) (23)
K, == PsC! Cs
In the above equations K I , K 2, and K. represent the equilibrium constants for
the individual steps.!
Assuming that the total number of active centers l, L (occupied or un occupied)
is independent of conversion, a relation for Cl can be obtained. L = Cj + CA + CR
+ Cs
L == Cz + _E!!PSCl + PRCl + PSCl
or
K2K.K, K3 K,

Therefore,
L
c, = 7I-+"--P-RP-S-/;-;K',~K;--3K~, --:-+-p-R-/'"""K-'-a-+"'--p-s/-:CK;7",

With this value of Cl, the rate equation for the first step of the reaction becomes
r = k,L(pA - PRPs/K)
1 + pRps/K,KaK, + PR/KC- --;--+-p-s/-;-;K~,
a
r = k(pA - pRps/K) (24) (25)
or
1 + KRSPRPS + KRPR + Ksps
where k = k,L KR = I/K.
K RS = I/(K,KaK,)
K. = I/K,

The rate equation for the second step

would be I HOUGEN and W 4TSON, op. cit.

CHEMICAL KINETICS 35

Elimination of the terms which cannot be measured experimentally leads to the

following equation:
T= k(PA - PRPsIK)

(1 + KAPA + KRPR + KsPs)t (26)

The squared term in the denominator appears because two active centers are
involved in the rate-controlling step.
If the third step were controlling, the starting equation for it would be T =
kaCR - k~CRCI (27)

and the finished equation

T= k(pAlps - PRIK)
 1 + KApA + Ksps + K RS (PAlps)
(28)

The assumptions made in the above derivations are:

l. Adsorption of all materials is in accord with the Koble equation. 2.
The resistance to diffusion is negligible.
3. Only one step is slow enough to be rate determining.
4. The specific rate constants and equilibrium rate constants are in
dependent of total pressure.
Selection of Equations and Evaluation of Constants.! Several techniques
making use of initial rates have been employed in the interpretation of
experimental data to select the proper equation for a given reaction. To use this
method, initial rate data have to be obtained at several different pressures. If
Eq. (24) is rewritten for the initial rate (zero conversion),

or TO = k'lf'
where 7C' is the total pressure and the feed is pure A, it can be seen that a plot of
ro versus 7C' is a straight ~line passing through the origin. It can also be seen that
the ratio 7C'lro is independent of pressure.
For the case where the surface reaction is the controlling step [Eq. (26)J, the
initial rate is
k-rr
(29)
T. = 0+b'lf')'

~ = (a + b'lf')'
or To

or
Therefore, a plot of (7C'lro)! versus 7C' will give a straight line of positive slope.
It may also be seen from Eq. (29) that a plot of ro versus 7C' will pass through a
maximum value.
1 CORRIGAN, Chem. Eng., 62, April, 1955; 62, May, 1955.
36 UNIT PROCESSES IN ORGANIC SYNTHESIS
If . the controlling step involved only one active center, its chemical
equation would be
Al;=Rl + S

The rate equation, then, would be

T = k(PA - PRPs/K)
1 + KAPA + KRPR + KSPB
(30)

and TO = 1 + KA1f (31)

 A plot of ro versus 11" becomes asymptotic to a positive value, but does not
pass through a maximum. Also, since

(32)

a plot of 1I"Iro against 11" would be a straight line of positive slope. If the
controlling step is that of desorption
Rl;=R + 1

the rate equation would be

r = k(pA/PS - pR/K) 1 + KApA= kIt KRS
~(31)
and + Ksps + KRS(PA/PS) TO =

Therefore, ro is independent of 11" and a plot of ro versus 11' is a straight line of

positive slope passing through the origin.
Plots of ro versus 11" and of 1I"Iro versus 11" may be quite helpful in deter
mining which step is controlling. Correlations involving the same variable in
both ordinate and abscissa must be regarded with caution, and the 11" Iro plots
should not be relied upon if the variation of ro is small compared with that of
11".

The values of initial rate ro may be determined by any of four methods: 1.

By measuring the slope of the tangent to the total conversion curve at zero
conversion.
2. By plotting rate versus time factor! (W IF) and extrapolating to zero. 3.
By plotting xl(W IF) versus x and extrapolating to W IF = O. 4. By fitting an
empirical equation to the conversion curve and differ entiating the equation.
After the initial rates have been used to give an indication of the mech anism,
the constants can be evaluated by using a plot of the appropriate
I The time factor W /F is a term which is often used in catalytic flow reactions in place of
contact time. W is the weight of catalyst in pounds and F is the feed rate in pound moles per
hour.
37
CHEMICAL KINETIC'S

group of termsl versus PA. For instance, if the controlling step were AI;=! Rl
+S
a plot of (PA - pRps/K)/r versus PA would be a straight line. If the con
trolling step were

Al + I ";=! Rl + Sl

a plot of [CPA - pRPs/K)/rjl versus PA would be a straight line. From the shapes
of the above plots the experimenter can usually elimi nate all but one or two of
the proposed mechanisms. The final test of the rate equation is whether or not it
fits the experimental data. The equation should be tested not only against the
data that were used to obtain the con stants, but also against new data which had
not been used previously in obtaining the equation. 2
Use of Rate Equations in Reactor Design.3 The method of using the rate
equations for catalytic reactions to calculate the reactor size and amount of
catalyst needed for a specified conversion and feed rate is very similar to the
method used for non catalytic reactions. The calculations may be divided into
three types, namely, those for isothermal reactors,4 adiabatic reactors,6 and
nonisothermal nonadiabatic reactors.6 In all three cases where the feed rate F
and the desired conversion x are specified, the weight of catalyst needed can be
calculated from the expression

W=
l"'dx (34)
FOr
The size of reactor is calculated from the weight and bulk density of the catalyst.
The methods of calculation for the three cases differ only in the evaluation of the
term on the right of Eq. (34).
Where the temperature may be assumed constant throughout the re actor, the
values of r may be calculated directly for various values of x. l/r is then plotted
against x, and the area under the curve up to the desired

to {ax. The amount of

conversion is measured. This area is equal Jo r
catalyst needed W may then be calculated.
In the case of the adiabatic reactor, the heat of reaction causes a change 1

CORRIGAN, Chern. Eng., 62, May, 1955; BARKLEY, CORRIGAN, WAINWRIGHT, and SANDS,
Ind. Eng. Chern., 44, 1,066 (1952); COCHRANE, Master of Science Thesis, Chemi cal Engineering
Department, West Virginia University, Morgantown, W.Va., 1951; CORRIGAN, GARVER, RASE,
and KIRK, Kinetics of Catalytic Cracking of Cumene, Chern. Eng. Progr., 49, 603 (1953).
• CORRIGAN, Chern. Eng., 62, June, 1955; 62, July, 1955.
a CORRIGAN and MILLS, Chem. Eng., 63, April, May, and June, 1956. 4
CORRIGAN and MILLS, Chern. Eng., 63, July, 1956.
• CORRIGAN and MILLS, Chern. Eng., 63, August, 1956.
I CORRIGAN and MILLS, Chem. Eng., 63, September, 1956.

38 UNIT PEOCESSES IN OEGANIC SYNTHESIS

in temperature. This, in turn, causes the constants of the rate equation to change.
To solve a problem of this type, it is necessary to use the rate equa tion and a
heat balance. The heat balance is used to establish a relationship between
conversion and temperature. A working curve of T versus x is plotted. Then as
each value of x is assumed, its corresponding temperature is obtained from the
working curve. The constants of the rate equation can then be evaluated from
plots of the log of each constant versus l/T (usually straight lines), and l/r can
be calculated. Then l/r is plotted
against x, and the area is determined as in the isothermal case. The nonadiabatic
nonisothermal case cannot be solved by plotting l/r versus x. Since heat may be
transferred in or out along the length of the reactor, there can be no direct
relat',)n between x and T which does not in volve reactor length. Since the
relationship between x and reactor length is the object of the problem, it is not
known. A relation between the heat transferred q and the reactor length must be
known. The problem may then be solved by ~ trial-and-error method.

NOMENCLATURE FOR CHAPTERS 2 AND 3

(2) Used in relative rate equations to
A, B, C, R, S, etc: represent moles of A, B, etc. Refers
to chemical species A, B, C, etc.,
viz., PA
Moles of species A, B, C, etc., in re
Sub A, B, C, etc. a, b, c, etc.
actor charge
Power of concentration function,
Exponent a, b, c, etc. viz., C~

b (k, + k,)t in Eqs. (16) and (17)

CA, CB Concentration of A, B, etc.
CAD, CBO Initial concentrations
CA, CB, etc. Concentration of adsorbate on Bur·
face of adsorbent
III Hypothetical term expressing an "ef
fective concentration" of unoccu pied
active centers
E Activation energy in Arrhenius ex
pression
e Base of natural logarithm
F Rate of feed to the reactor
I Integration constant
K Reaction equilibrium constant
K" K" Ka, etc. Equilibrium constant of individual
reaction steps
KA, KB, KR, etc. K' Adsorption equilibrium constants
k Constant in Koble equation Over-all
specific rate constant
(1) Symbol of chemical species in
volved in chemical reaction
r' Unoccupied active c~nter
s Natural logarithm
kl' k" ka, etc. k' Mole ratio of B to A in the
l' feed CBo/CAo
L t Exponent in general order of
V reaction expression
L' W Partial pressure
I W/F Universal gas consta.nt
In CHEMICAL KINETICS Component R, S; al~o moles
M of com ponent R, S
Reaction rate
m,n Specific rate constant of
Initial reaction rate
individual reaction step
Rate of reverse reaction
P Specific rate constant for
Frequency factor ill
R reverse step of a reaction
R,S Arrhenius ex pres8ion
Total concentration d active
Absolute temperature
T
centers involved in a
Time
catalytic reaction
To Reactor volume
Constant in Koble equation
 Weight of catalyst actor actant
Pounds of catalyst p~r pound Ratio of rate constants, viz., Total pressure
mole of feed to a k, : kl Degree of 39
vapor-phase catalytic re conversioJ). of limiting re
CHAPTER 3

CHEMICAL-PROCESS KINETICS

By THOMAS E. CORRIGAN AND JOHN J. McKETTA

Chemical-process kinetics is a study of the influence of the physical fac

tors that affect chemical reactions. Examples of these factors are: 1. The type
and shape of reactor used.
2. The method of operation.
3. Temperature control.
4. Batch or flow process.
5. Batchmixing.
6. Fixed or fluidized bed (in the case of catalytic reactions). These factors
are characteristics of the reacting system. This classifica tion may appear
arbitrary, but it is helpful in discussing the subject of re actor design.

Factors That Affect a Chemical Process

Problems in Choice of a Reactor. When a process engineer is faced with the
problem of designing a commercial or semi commercial unit, he must first
choose the reactor to be used. The type of reactor (tube, tower, or tank), the
type of operation (batch, continuous, recycle, or once-through), and the means
of temperature control (isothermal or adiabatic) may depend on the type of
reaction involved. In order to choose the best reactor and method of operation,
the specific type of reacting system must be considered.
Backmixing. Backmixing is an extremely important phenomenon which
must be considered carefully in all process-reactor design. Backmixing can
occur only in flow reactions in a chemical process. Backmixing is the mix ing
of the reactants and products of a chemical reaction by upstream diffu sion and
reaction while the main flow is in the downstream direction. Later, in Fig. 3-2,
it will be seen that in some cases of reactor design the failure to consider
backmixing may result in an error of one hundredfold in the cal culated reactor
size.
Type and Shape of Reactor. The most common types of chemical re actors
can be classified according to the scheme shown in Table 3-I. 40
41
CHEMICAIrPROCESS KINETICS

TABLE 3-1. CLASSIFICATION OF CHEMICAL REACTORS

1. Simple batch homogeneous reactors

Closed tank
Rocking autoclave (laboratory tool, not used commercially)
 Stirred kettle
Kettle with outside recirculation but with no material added or
removed
Coil with outside recirculation but with no material added or
removed
2. Semibatch reactors
Batch with continuous addition of one reactant:
Gas-phase addition
Liquid addition
Solid addition
Batch with continuous removal of one product:
Gas formation
Solid precipitation
Formation of immiscible liquid
Batch with combined addition of reactant and removal of one
product
3. Continuous homogeneous reactors
Longitudinal tubular reactor (no backmixing)
Stirred-tank reactor (complete back mixing)
Tubular reactor with some backmixing
Tower reactors:
Packed tower
Empty tower
Baffled reactor
Baffled-tank reactor
Longitudinal reactOl" with multiple injection of ODe, f!II'(ltant 4.
Continuous heterogeneous reactors " . " .
Packed-tower countercurrent reactors
Fixed-bed catalytic reactors
Longitudinal
Backmixing
Moving- and fluidized-bed catalytic reacton
Longitudinal
Backmixing
Distillation column

The simple batch reactor for homogeneous reactions is the most common
type. From this kind of reactor, many kinetic data have been obtained. All the
reactants are charged in at the beginning of the reaction, with no mass transfer
occurring until the reaction is complete. Examples of batch reactions are the
ammonolysis of nitrochlorobenzenes, hydrolysis of esters, and polymerization
of butadiene and styrene in aqueous suspension. These will be treated in more
detail in later chapters.
The semibatch reactor with continuous addition of one reactant is one in
which <.:me reactant is first charged and another is continuously added as
42 UNIT PROCESSES IN ORGANIC SYNTHESIS
the reaction proceeds. Examples of semi batch reactions are the production of
high-molecular-weight polyglycols or of detergents by the reaction of ethylene
oxide with ethylene glycol in the first case and alkylated phenols in the second.
In these cases a large amount of ethylene oxide is added to a smaller amount of
a less active reactant. By using the semi batch method wherein ethylene oxide
is added gradually to the second reactant, large concentrations of hazardous
ethylene oxide are avoided.
Another classification of a semibatch reactor is one in which a gas forms or a
solid precipitates during the reaction. Here, also, a volatile product may be
fractionated off continuously. An example is batch esterification with
continuous distillation. An example of this type of reaction, which will be dealt
with later in this text, is the esterification of ethyl alcohol with acetic acid to
form ethyl acetate.
Of the longitudinal tubular reactors the most common type is the single pass
long tube such as those used in thermal cracking furnaces. Another type
consists of a number of single tubes in parallel. In both cases, the ratio of
length to diameter of the tube is large. In this reactor the reacting mass is
carried through the reactor in plug flow with no diffusion in the longitudinal
direction.
In the stirred-tank continuous reactor (the case of complete backmixing),
reactants are fed and products withdrawn continuously. Continuous stirred-tank
reactors are currently used in liquid-phase nitration of hydro carbons or poly
hydric alcohols. A diagram of such a reactor is shown in Fig. 4-10.
In a tubular reactor, if the ratio of length to diameter is not large enough,
there will be some backmixing and the reactor will deviate from the hypo
thetical plug flow. This is usually true in most commercial tubular reactors. The
most important processes using a tubular reactor are the thermal cracking of
ethane and of propane to form ethylene. An example, dealt with in Chap. 6, is
the dehydrochlorination of ethylene dichloride.
Tower reactors may consist of an empty tower, a packed tower, or a baffled
tower. Where a large volume is needed, this type of reactor is less expensive
than a tubular reactor. However, the amount of backmixing is much greater. An
example of the use of a packed-tower reactor, which will be discussed in Chap.
14, is in the German process for the production of the ethers of ethylene glycol
from ethylene oxide and alcohol. In this process, ethylene oxide and alcohol
are fed to a pressure tower packed with iron Raschig rings.
The baffled-tank reactor is usually a horizontal tank with a low ratio of length
to diameter. It has about the same advantages and disadvantages as the tower
reactor.
The longitudinal reactor, with multiple injection of one reactant, is a special
case and corresponds to the semi batch reactor.
CHEMICAL-PROCESS KINETICS 43

The most common heterogeneous reactors are the fixed-, moving, and
fluidized-bed catalytic reactors. One example of vapor-phase catalytic re actions
is in the catalytic vapor-phase reduction of nitroxylene to xylidene. This process
is discussed in Chap. 5 (Figs. 5-16 and 5-17). Reaction System. Reaction
systems can be divided into the following groupings:
1. Homogeneous liquid phase, only mutually soluble liquids are involved.
This is represented by the hydrolysis of ethyl acetate, which will be dis cussed
in Chap. 13.
2. Heterogeneous liquid phase, involving two or more immiscible liquids.
Examples of this are in the nitration of toluene, hydrolysis of chlorobenzene to
phenol, and ammonolysis of ethylene dichloride.
3. Reaction of a liquid with a gas, as in the liquid-phase chlorination of
benzene or acetaldehydp •
4. Reaction of a liquid with a solid, as the sulfonation of anthraquinone. 5.
Liquid-phase reaction forming a solid, as in saponification of glyc erides.
6. Liquid-phase reaction forming a gas, as in the Friedel-Crafts reaction
liberating HCI.
7. Homogeneous gas-phase reaction, as in the vapor-phase nitration of
paraffins or in the dehydrochlorination of ethylene dichloride. 8.
Heterogeneous gas-phase catalytic reaction, such as the hydrogenation of
carbon monoxide to form methanol.
9. Heterogeneous gas-phase non catalytic reaction, as represented by the
gasification of coal or coke to form synthesis gas.
There are three general methods of operation for chemical reactors, namely,
isothermal, adiabatic, and nonisothermal nonadiabatic. In the isothermal case,
just enough heat is added or removed to keep the temperature constant
throughout. In the adiabatic reactor, no heat is added or removed during the
course of the reaction. In the non isothermal nonadiabatic reactor, some heat is
either added or removed during the reaction but the temperature does not
remain constant. Almost all com mercial reactors are operated as nonisothermal
nonadiabatic reactions.

Reactor Shape and Effect of Backmixing

The type and shape of a reactor, because of their effect on backmixing, have
a profound effect on the process kinetics. For the purpose of kinetic
calculations, reactors may be classified into three basic types. These are: 1.
Batch reactor.
2. Longitudinal flow reactor.
3. Stirred-tank flow reactor.
The longitudinal reactor with zero backmixing, and the stirred-tank reactor
with 100 per cent backmixing, are the two extremes between which
44 UNIT PROCESSES IN ORGANIC SYNTHESIS
most actual flow reactors will lie. The process engineer who designs a flow
reactor will first use kinetics to calculate the extreme cases and then must
estimate from empirical data how the actual reactor will behave. The batch
reactor and the longitudinal reactor are treated the same mathe
matically.
The kinetic behavior of these basic reactor types is indicated in the con
centration-time curves of Fig. 3-1, which shows concentration histories
Botch reactor Longitudinal reactor Stirred tonk reactor
Q

-
)
Q.
cb
~

a
~~ ? b

co f--
c

0
·~I R

0b; 1~
n
0

c
~
~
Leng 8 ~E 1/

·i volume)
>0
e
(;
1

_
;;
:f

JAR
Time
d

Ibs jbs b
E
.

. Final time Final volume

. q h Final volu!"e
~
Q)
>

..
;

c
0
u

FIG. 3-1. Kinetic behavior in bailie reactor types.

within the reactors (second row) and concentration-time curves for reactor
effluents (third row). The kinetic equations for the three basic reactor types
are shown in Table 3-2 for some of the more common reaction types. Figure
3-2 illustrates the effect of backmixing on first- and second-order reactions.
Two important items which must be known about backmixing in order to
design a reactor are:
1. If backmixing is present in the process, what additional holding time or
reactor volume is needed to compensate for it?
2. If the compensation is not made, what would be the decrease in pro
duction or yield due to the backmixing?
~
11

.., .-'"
r-,
~
~\ .... I

bI)
J ..r .$
.....
'" "ii '" I .....
.~ ~I- CJ 1/
l._.J
e 11 -'Co ..>:
... ~ I CJ

!XI ~.~ "",

'1l_
flS II
II

I
CJ >.t:> ~
~
fll! '" ci "ii ... Ql~ C'I
+ ~
'"
+ ... ... CJ _ OJ OJ
0

..c: ..c:
8 ,.., "t! CJ a: a:
f
1/ II 1/

000
~] >.t:> ~ CJ
o .~ "~ ci

II I +
11
..c:::""
~ :~ '" ,---,
f 0 "!~o

!XI]
"ii\fl I I
J ..... 10 'S2.
II II II

0 ..... 10 0 ~I~
.....
""'b
-
l._.J 1/ ~ ...
f:
'11. CJ ~ ..5 ~ 1/

..... I~ OJ
fll~ ...
OJ
..c:::
'" ..c::: a: a:
 :f ... ... " ...
] CJ CJ "ii
~ § CJ -'C oo'C -'C I

S ... "ii 0
OJ'~ 1/ 1/ II
!!::~
• ~ ::l
01~ "'I~ "'I~ 0
I
"" 0- .:l~ ~
Ul "<:l"<:lS;;"<:lS;;"<: oo'C
oj
!XI lI II .....
II ~
II
01~ "","<:l
I -tll~

------------
-
b t:: ~
.9 "'g
\ ~\ ~ ..... C'I C'I
~.,o

~---~
t"/;)

+
= . .3 8. ~ ~ '" >, t t ~., ~ J:Ci J:Ci i+ +
-.: ~ ~
45
46 UNIT PROCESSES IN ORGANIC SYNTHESIS
The graphs in Fig. 3-21 show how important the effect of backmixing can be,
especially at high conversions. It can be seen that for the extreme case of 99
per cent conversion at a mole ratio of 1: 1 the reactor volume re quired for a
stirred-tank reactor is 100 times as great as for either a longi tudinal or a batch
reactor. From the above discussion it is clear that backmixing has a
considerable effect on reactor size and conversion.
Effect of Backmixing on Product Distribution. 2 The type and shape of a
reacior, because of backmixing, also have a great effect on the yield or

Ratio of bockmix reaction

time to botch reaction time for
first and second order reactions
os a function of molar rolio
u 7
'" E '';: .c "0 .0
10 8 6 2
and froction
'" E '';:
5 converted

"E -'" u 0 (I) 4

3
 converted. x

Fraction
FIG. 3-2. Relation of backmixing to design and size of reactor.

product distribution. In some processes, especially those which require only a

small reactor, the effect of backmixing on yield is much more im portant than
the effect on reactor size.
For cases in which product distribution is important and reactor size is
secondary, equations have been worked out for the batch, longitudinal, and
backmixing reactors. These equations will be presented for the three major
types of complex reactions, namely: parallel, series, and complex series.
1 LF;SSELS, Olin Mathieson Chemical Corporation, Brandenburg, Ky., personal com-

munication; also Chem. Eng., 64 (8),251-2.56 (1957). 2 MAcMULLIN and WEBER, Trans. Am.
Inst. Che~. Eng., 31 (2),409--458 (1935).
CHEMICAL-PROCESS KINETICS 47

Parallel Reactions. The simplest case of paralkl reactions it ---> Rand A ---> S is
used here to simplify the explanation. A more common type of parallel reaction is
A+B->R
A+B-.S
A+B->T
An example of this type is in the nitration of toluene, discussed in Chap. 4.
Here the parallel reactions are

OOCH. CHs
-NO'
 CHs CH.

OO·
-I- HNO, -> ~ I _
~

-I- HNO,-> I .'

~ -NO,

The basic differential equations and the concentration-time relations for parallel
reactions for longitudinal and backmixing reactions are shown in Table 3-3, and
the corresponding product-distribution equations are shown in Table 3-4. It can
be seen from these equations that backmixing does not effect the product
distribution for parallel reactions of the same order.
Series (Consecutive) Reactions. The simplest example of series reac tions
which will be used for illustration is the reaction type
A-+R-+8
The basic differential equations, as well as the integral equations for both
longitudinal and backmixing reactors, are shown in Table 3-3. The cor responding
product-distribution equations are shown in Table 3-4. From these equations it
can be seen that hackmixing, and thus reactor shape, has a considerable effect on
product distribution as well as on required holding time.
A specific example of the kinetics of the seri:os reaction
A->R-+S-+T
is in the liquid-phase chlorination of benzene where the gaseous materials Cb and
Hel can be eliminated from consideration in the calculation of
j
.,+
~c
 JI!~!
..... ~

II II

o~
..
~
-....- "
I
J
~
I ----\ .....

J .....
\ .. MI
II
~
-....-""l ""l II
n
..

rJ ~
~ I

+ rJ o 0~ ~~
~
II

sJl~
I
- II II II

sJl~ ~I~ sJl~ I

48
~
"
110
~
I .9
.. .~ I~ ] ~I£ = !Xl + .~ =' " !Xl '" .. :€: ~
+' II II II .!!l
~I~ ~ ~
"0 +'
 I': II
" .. ~I" I I ;
!XlI"
='
= II
.... ~ ....
"0 ~I~
h
II
~
8 ~
"';
Po.

..C

I3

8'

!ii

'"

=
~..9

"00
1

., " ~
+ ~
+ ~
= '" ~ ~
~ +~
~
110 II II ~

.t:l .2 = ~
I
~
I ~
II
II

01 !Xl
--;,~
~ ~ II
.~
='
$~ ~ cS

'" ~
 ] r;j r;j
~§
~+
"<:
b ~
.,.~
II::+'
~
Q
~ II II
I ~

b
"0 I
.~ '" ~I~ sJl~ ., \.)
~ ~~
0
g. .,
II II

.~
" bj_ b "<:
\.)

r;jj_
~ ~~
II II
II
-j
'1;;!'1;;1'1;;I~

"'1-
\.)_\.)'1;
;1
~'1;;1'1;;I

I"'-l ~ ~~
= ~ .lit ~t
~~
~t .lit ~i
~8. ~ i:!:l ~
J~ JOt ++
~
~~

., .. ill
= .,; .~ i '': "'- ~" ><
Ql ., :::I .,; "S.
.~ 8
~
Jj
, 8 49
50 UNIT PROCESSES IN ORGANIC SYNTHESIS
product distribution. This application was developed by Macl\Iullin,' and the
results of his kinetic work are shown in Fig. 6~7.
Complex-series Reactions. In mosL industrial reactor pesigns, the prob lem is
usually that of obtaining the highest yield possible while using the smallest
reactor that will give the conversion and yield desired. There are some cases
where product distribution is so much more important than re actor size that
holding time is an insignificant problem. Examples are:
1. When the reaction is extremely rapid, taking place in a matter of seconds
or a fraction of a second, but where the yield is important. 2. When the reactants
and products are so expensive that even for moderate reactions the depreciation
of the reactor has an extremely small effect on total product cost.
3. When the cost of the equipment for separation of the reactor effluent,
recycle, and purification of the products is of a higher order of magnitude than
for the reactor itself.
For these cases the integral equations, which present a complete picture of
 product distribution for any given feed ratio or degree of conversion, need not
involve holding time and are, therefore, easier to use.
There are many important industrial applications of complex-series re
actions. Some of the more important examples are:
1. The chlorination of benzene or toluene.
2. The chlorination of either saturated or unsaturated hydrocarbons. 3. The
manufacture of ethylene glycol from ethylene oxide and water. 4. The
manufacture of ethanolamines.
5. The manufacture of alkylamines from either the corresponding al cohols
or halides and ammonia.
6. The manumcture of glycol ethers, polyglycols, and nonionic deter gents.
7. The vapor-phase nitration of paraffins.
8. The liquid-phase nitration of higher paraffins.
9. The reaction of carbon tetrachloride and hydrogen fluoride to form Freons.
10. The manufacture of chloral.
11. The catalytic reaction of acetylene and Hel.
12. The reaction of ethylene with hydrogen chloride and oxygen. 13.
The Raschig synthesis of chlorinated benzenes.
14. The manufacture of chloral by the chlorination of ethyl alcohol. 15. The
amination of chloroacetic acid to form glycines. The above examples of
complex-series reactions are described in sub
sequent chapters, but the kinetics of product distribution are not discussed. For
this reason the following section in this chapter is presented to explain
ANGRAU Central Library
1 MACMuLLIN, Chem. Eng. Prog., 44 (3), p. 183 Hyderabad

(194 I
I 1 ~~?_. _ ... , .. ,
CHEMICAL-PROCESS KINETICS 51

the application of kinetic theory to the study of product distribution. The effect of
backmixing on product distribution is very important to the study of unit
processes.
The manufacture of ethylene oxide derivatives is an especially good ex ample
of the industrial application of complex series.
High-molecular-weight polyglycols arc formed by the reaction of ethyl ene
oxide with ethylene glycol or diethylene glycol. Some nonionic de tergents are
formed by the reaction of ethylme oxide ",ith fatty alcohols or alkyl phenols. In
both cases, a large amount of ethylene oxide (which is highly reactive and may
polymerize with explosive violence) is added to a small amount of another
reactant. In this type of reaction, the semibatch reactor is superior. The relatively
inert reactant is charged to the reactor first, and then ethylene oxide is added
gradually as it reacts. Thus, there is no large concentration of oxide at reaction
conditions.
Glycol ethers arc manufactured by adding ethylene oxide to alcohols. In these
reactions, both reaction rate and product distribution arc impor tant. Reaction rate
is important because the reaction is slow enough to require a large, costly reactor.
Product distribution is important because the yields based on oxide and alcohols
are lowered if an excess of the higher-molecular-weight by-products is formed.
In this type of reaction, a tubular reactor would be desirable because it would
give a better control of product distribution than a tank or tower reactor.
The reaction of ethylene oxide with aqueous ammonia to form ethanol amines
is an example of a pJ'ocess where the kinetics of product distribution is most
important. In this process, the reaction is fast and the reactor required is very
small compared to the separation and finishing equipment nerded in the process.
For this reason the absolute rate of reaction is of minor importance. The relative
rates of the reactions to form mono-, di-, or triethanolamine are extremely
important for economic yield and product distribution.
The derivation of the product-distribution equations for the complex series
reaction of the type
A+B-->R
A+R---.S
is shown in Table 3-5 for both longitudinal and backmixing reactions. A
simplified flow sheet for a typical continuous process for this type of reac tion is
shown in Fig. 3-3.
Equations shown in Table 3-5 can be used to determine the yield of desired
product R, for any given feed ratio (CBO:CAo) and for any de sired per cent
conversion for a batch, longitudinal-flow, or stirred-tank re actor.
An example would be in the evaluation of laboratory batch data for a
.....
I II
Q:l1"
I iii' ~ ~
~\Q:I
.. 11
~I ++ I

~I~ I I ~Q:I s- ~ ~ Q:I

+ 11 11

Q:I ~Q:I
II

~~
++
-II ~~
"'~
I
..... i:I:ll" I I
11 11

~ -II
"'~
I ~ +
~ ....
11

~~
 '"
1
o
~ ....
11

~
~i:I:l

II
~~

Q o
1

52
CHEMICAL-PROCESS KINETICS 53
Recycle 8

Produc!
A storage
8
FIG. 3-3. Flow diagrtlm for a typical continuous process involving complex-series re
actions.

reaction which goeS to completion with respect to reactant A. In this case, the
procedure would be:
1. Obtain the e"perimental product distribution in a laboratory or pre
pilot-plant batch reactor.
2. Use an appropriate equation from Table 3-5 and a material balance to
calculate the rate-ratio constant" from the batch data.
3. Use the value of " and the same equation used in step 2 to calculate the
yield of R at other mole ratios, and plot R values against feed ratio (CBO:CAo)'
This gives the curve for a longitudinal flow reactor.
4. Calculate the yield of R, and plot R values against feed ratio. This gives
the corve for a backmixing reactor. Figure 3-4 shows a typical "plot for the
reaction
A,+B-+R
A+R-+S
All actual reactors would operate in the shaded area between the two lines.
Tubular reactors would be close to the upper line, stirred-tank reactors would
be close to the lower line, and packed-tower reactors would be ap proximately
halfway between.
Selection and Sizing of Homogeneous Reactors. One of the first deci sions
to be made when designing a reactor is which reactor type is best for the
specific proceSS. If a large-scale process is considered, then a continu ous
reactor will haNe a larger capacity and is preferable. If the process in volves
small quantities, intermittent products, a complicated sequence
of operations, or extremely long holding times, a batch reactor may be
preferable. If a flow process is used, the flow reactor could be a long coil, a
bundle of parallel tubes, an empty, packed, or baffled tower, or a stirred tank
or vessel.
54 UNIT PROCESSES IN ORGANIC SYNTHESIS
 Points on upper curve' expeflmento! values.
Performance of any real reactor wi!! fa!!
between these two c/Jrves in shaded area.

789
Mole ratio, BolAo

FIG. 3-4. Product distribution as a function of type of reactor.

For a single-phase reaction with a low holding time, a single coil is prefer
able. This gives high velocities and very little backmixing. A bundle of parallel
tubes can also be used in place of a single tube. This type will give a lower
velocity for the same holding time.
For tubular reactors with high holding time, the cost of tube or pipe is quite
high per unit volume. A tmver will provide a large reactor volume at a much
lower cost. The disadvantage is that there is considerable back mixing in a
tower or tank reactor unless packing is used.
The stirred-tank continuous reactor may be used where a high degree of
agitation is necessary, possibly where a solid is held in suspension, where the
reaction requires a relatively low holding time, and where backmixing is not
detrimental to the yield.
If either the batch or the tubular type of reactor is chosen, the reactor size
and product distribution can be calculated by using the batch or
longitudinal-flow equations. For a stirred-tank continuous reactor, the
backmixing equations can be used. If a packed or baffled tower is used, then the
calculations must be made for both the longitudinal and back
mlxmg cases. Proper extrapolation must then be made from empirical data or
previous experience.

2 ;;~
 CHEMICAL-PROCESS KINETICS . ~ 6 R 2 55
Calculation of Product Distribution. Problem. Assume a process is to -~ produce the
hypothetical compound R from. the raw mater\a,li~~ ~:, C R It is a liquid-phase

.• \ \)-5\,,,
reaction, and the equations of the reaction ~rk'·.. n,.., 0 A+B-+R

A+R-+S ----_._-
Assume that S does not react further with A to form any more products in the
series. Figure 3-3 is a simplified flow diagram of the process. R is the desired
product, and all the S that is formed must be thrown away. The unreacted B
which passes through the reactor may be recovered and re
cycled, but there is a 6 per cent loss in the recovery process. Any un reacted A
which passes through the reactor is lost in the recovery process and cannot be
recycled. The reaction timl: is several hours, and a large reactor is' needed.
A catalyst has been discovered that will increase the reaction rate one
hundredfold. It is estimated that this would save $100,000 in the reactor cost.
You, as a process engineer, are asked to determine whether the catalyst should
be used in the process. For a preliminary comparison of economics, calculate
the raw-material costs of R as a function of mole ratio of B to A in the feed for
longitudinal reactors with and without cat alyst.
The production rate of the plant is 10 million lb of R per year. The raw
material costs are 10 and 20 cents per pound for A and B, respectively. The
values of K are 0.5 without catalyst and 1.5 with catalyst. Molecular weights
and physical properties of A, B, R, and S are available.
Solution (equations are taken from Table 3-5): Since R is the desired
product, there is no reason to use a recycle stream in the process and, thus, the
following equation can be used:
R = B' - B
1-K
This equation is combined with the material balance equations
Ao-A =R+2S
and 1 - B = R + S
and values of R are solved for at assumed values of B. The quantity R/CR +
S) will give the yield of the reaction, and from this the raw material cost can be
calculated.
This method of solution can be used for the case where no catalyst is used
(K = 0.5) and where the catalyst is used (K = 1.5). In both cases the loss of raw
materid B from the recycle stream per mole of product is 0.06 BIR. The
results of these calculations are plotted in Fig. 3-5. It can be seen that when no
catalyst is used the minimum raw-material cost is 20 cents per pound of R, but
when the catalyst is used, the minimum raw material cost is 23.5 cents per
pound of R. Thus, in this particular case,

r APAUCSt-l',:"
56 : .. r' l: ltNIT PROCESSES IN ORGANIC SY:NTHESIS
the use of the catalyst would result in an increased raw-material cost of
$350,000 per year.
The raw-material cost for the case of a reactor with no catalyst but with
complete backmixing is also plotted in Fig. 3-5 for comparison. It can be seen
that the raw-material cost is higher because of backmixing.

100

f
R. cost wllh catal wit
M. n yst 0.5 h
boclrm o
K
=

I
/
rR.M. cost with
catalyst K = I. 5
longitudinal

~
I
- ~- ..

- '
-
.:::~
f
:
- :
:
 -
LRM. cost with no
catalyst K= 05 longitudinal

90
80
70
60
CI::
50
'0
£!
40 ~

(I)
a.

C '" (I) 30 <.>

-:;;
0 <.>
:;i
20
Ii

Mole rotio __Q_

Ao

2 3 4 5 6 7 8 9 10 8

FIG. 3-5. Hypothetical raw-material costs related to use of a catalyst as a process variable.

Selection and Sizing of Catalytic Reactors. There are numerous exam· pIes
of catalytic vapor-phase reactions in the organic-chemical industry. Some
examples which are discussed in subsequent chapters are:
1. The vapor-phase reduction of nitrobenzene with hydrogen. 2. The
manufacture of hydrogen by the reaction of steam and natural gas over a nickel
catalyst followed by the shift reaction over an iron oxidE catalyst.
3. The hydrogenation of nitroxylene to xylidene.
4. The reduction of nitrobenzene to aniline.
5. The chlorination of C1-C4 paraffins.
6. The catalytic hydroammonolysis of carboxylic acids to form nitriles.
 CHEMICAL-PROC'ESS KINETICS 57

7. The catalytic vapor-phase amination of methanol.

8. The direct oxidation of ethylene to form ethylene oxide. 9. The synthesis
of methanol from carbon monoxide and hydrogen. 10. The hydrogenation of
olefinic hydrocarbons to produce high-octane fuel components for aviation use.
Since there are so many applications of vapor-phase catalytic reactions in
carrying out unit processes, the kinetics of this type of reaction are very
important.
One of the -most familiar of catalytic chemical reactors is the fixed-bed
reactor. Here the catalyst particles remain in a fixed position and the reacting
gases move through the catalyst bed. The catalyst particles are usually porous
pellets, either cylindrical or spherical, ranging from Ys-->i in. or more in
diameter. The physical structure of these pellets is usually such that the internal
pore surface is infinitely greater than the actual pellet surface. Thus, the actual
contact surface present is independent of pellet size.
Longitudinal versus Backmixing. Fixed catalyst beds may be classified
roughly into longitudinal and backmixing types.
Two disadvantages of reactors which have backmixing are: 1. For a specified
conversion, the amount of catalyst and the size of reactor required are much
greater than in the case of the longitudinal reactor.
2. The ultimate yield of desired product will not be so great as in the case of
the longitudinal reactor.
Fixed-bed reactors may be of either the longitudinal or the backmixing type,
but in most cases the longi~;udinal reactor is preferred because of the advantages
presented above. The catalyst chamb'er may consist of a series of relatively long
catalyst tubes operating in parallel.
The major disadvantages of the fixed-bed reactor are:
1. It is difficult to transfer heat into or out of the catalyst bed fast enough to
prevent large temperature gradients within the catalyst bed or even within the
catalyst pellets themselves.
2. Where the rate of diffusion through the pores of the catalYlst pellet is slow
compared with the reaction rate, it is not possible to utilize the total surface
area of the catalyst.
3. When the catalyst needs periodic regeneration, there is expense in volved in
taking the fixed-bed reactor out of service and of regenerating the catalyst.
In the fluidized-bed reactor, the catalyst bed is in the form of a fine powder
which is suspended by the upward motion of the reacting gases passing through
the bed. Difficulties which may occur in fluidization are slugging, channeling,
and attrition of the catalyst. In most cases, cir-
58 UNIT PROCE.SSES IN ORGANIC SYNTHE,SIS
cumventing these problems leads to a fluidized reactor which has a low
height: diameter ratio and thus allows backmixing of the gas.! Thus, one of
the major advantages of the fixed-bed reactors-that of longitudinal flow-is not
present in the more common types of fluidized-bed reactors.
Calculation of Reactor Size. Assume that the hypothetical reaction
 A+B-.R+S
is a vapor-phase reaction taking place on a solid catalyst. We can assume for
the purpose of illustration that the rate equation for a specific case could be
r = klPAPB
1 + k,PA + k'PR
If the above reaction were carried out in a longitudinal fixed-bed reactor,
operating i~othermally, at lOOO°F, what amount of catalyst would be re quired to
cause the reaction to be 95 per cent complete, using a stoichio metric feed
ratio and no recycle? The desired production rate is 1,000 lb of R per hour.
The reactor is to be operated at 2 atm and exhibits a neg ligible pressure drop.
Following is the general procedure for this type of problem: 1. Evaluate, or
look up, the values of the constants k!, k2, and k3 at the reaction temperature.
2. Calculate the values of PA and PB for various values of x from 0-0.95,
using a material balance.
3. Calculate the value of r at each value of x.
4. Plot 11r versus x.
5. Take the area under the curve of 11r versus x from x = 0 to x = 0.95 (the
direct conversion).
6. This area is the desired value of W IF, where F is the feed rate and W is
the required weight of catalyst for 95 per cent conversion. Since F is known,
calculate W.
Summary. The first three chapters of this book present a scientific ap proach
to some of the principles involved in the selection and design of chemical
reactors used in the unit processes. In many respects the selec tion of the
proper type of reactor for a unit process is still as much an art as it is a
science. It is recommended to the reader that, as he studies the unit processes
presented in the remainder of this book, he try to justify the type of reactor
used in each case.

REFERENCES
Corrigan, Thomas E., and Earle F. Young: Simple Reactor Design, Chern. Eng., 62 (8, 9, 10,
11, and 12) (1955).
1 GILLILAND and MASON, Ind. Eng. Chern., 41, 1191 (1949).
CHEMICAL-PROCESS KINETICS
59
)rrigan, Thomas E., and Earle F. Young: Reactor Design for Complex Reactions, Chem.
Eng., 63 (1, 2, and 3) (1956).
)rrigan, Thomas E., and W. Charles Mills: Catalytic Reactor Design, Chem. Eng., 63 (4, 5,
and 6) (1956).
)rrigan, Thomas E., and W. Charles Mills: Catalytic Reactor Design Problems, Chem.
Eng., 63 (7, 8, 9, and 10) (1956).
miels, Farrington: "Chemical Kinetics," Cornell University Press, Ithaca, New York, 1938.
ost, Arthur A., and Ralph G. Pearson: "Kinetics and Mechanism," John Wiley & Sons,
Inc., New York, 1953.
lIiland, E. R:and E. A. Mason: Gas and Solid Mixing in Fluidized Beds, Ind. Eng.
 Chem.,41 (6), 1191 (1949).
)ugen, O. A., and K. M. Watson: "Chemical Process Principles," pt. 3, John Wiley & Sons,
Inc., New York, 1947.
CHAPTER 4

NITRATION

LESTER P. KUHN, WILLIAM J. TAYLOR, JR., AND P. H. GROGGINS

I. INTRODUCTION
The nitration reaction serves to introduce one or more nitro groups (-N02)
into a reacting molecule. The nitro group may become attached to carbon to
form a nitroaromatic or nitroparaffinic compound. It may become attached to
oxygen to form a nitrate ester, or it may become at
tached to nitrogen to form a nitramine. In the nitration process the enter ing
nitro group may replace a number of different monovalent atoms or groups of
atoms. In this chapter we shall be concerned only with those nitrations in
which the nitro group replaces a hydrogen atom, since these are the reactions
of greatest technical importance. Among the reactions in which the nitro group
replaces atoms or groups other than hydrogen, the following are shown by way
of illustration. Certain alkyl halides can react with silver nitrate to form the
corresponding nitrate esters or with silver nitrite to form the corresponding
nitro compounds as shown below. RCI + AgNO, --> RONO, + AgCI
RCI + AgNO, --> RNO, + AgCI
In the aromatic series the sulfonic acid or acetyl groups can, in certain in
stances, be replaced as shown in the accompanying equations.
OH OH OH

O III
-SOJI RNO. OINU-NOI RNO. OINU-NOt

--> I --->. I
~ ~.
II.1
SOJI SOsH NO.
o
II

0
CCB.

1 RNO.
 -OMe CH.COO!-i
1
OMe ~M'
60 1
OMe
NOt
1
NITRATION 61

Nitration is one of the most important reactions in industrial synthetic organic

chemistry. Not only do nitration products find wide application as solvents,
dyestuffs, pharmaceuticals, and explosives, but they also serve as useful
intermediates for the preparation of other compounds, par
ticularly amines which are prepared by the reduction of the corresponding nitro
compound.

II. NITRATING AGENTS

A variety of reagents can be used to effect nitration. These inelude fuming,
concentrated, and aqueous nitric acid and mixtures of nitric acid with sulfuric
acid, acetic anhydride, acetic acid, phosphoric acid, and chloroform. Nitrogen
pentoxide, N20 5, and nitrogen tetroxide, N20 4, are also used in certain instances.
In order to make an intelligent choice of nitrating system for a particular
nitration, it is desirable to know what species are present in the various systems
and to understand the mechanism of the reaction under consideration.
The Nitryl Ion, NOt.1 The system nitrie acid-sulfuric acid, commonly known
as mixed acid, is the most important nitrating medium from a practical
standpoint and is probably also the best understood. There is a large body of
evidence that supports the belief that nitric acid exists in strong sulfuric acid as
the nitryl ion, N02+. From freezing-point depres
sion measurements, the Van't Hoff i factor (the number of particles gen erated by
one molecule of solute) of nitric acid in sulfuric acid is found to be 4, which
indicates that the ionization of nitric acid can be represented by the equation
HNO, + 2H,SO, --> NO,+ + H30+ + 2HSO,-
The i factors of ethyl nitrate, nitrogen pentoxide, and nitrogen tetroxide in
sulfuric acid are 5, 6, and 6, indicating that these substances ionize to form nitryl
ions according to the equations
C,H.ONO, + 3H,SO, --> NO,+ + H,O+ + C,H.OSO,H + 2HSO,-
N,O. + 3H,SO. __, 2NO,+ + H,O+ + 3HSO,-
N,O, + 3H,SO, --> NO+ + NO,+ + H,O+ + 3HSO,-

, Each of these solutions is a powerful nitrating agent.

Solutions of nitric acid have three different ultraviolet-absorption spec tra. In
dilute aqueous solution, its spectrum is that of the nitrate ion, N03-. In an inert
weakly polar solvent such as chloroform, its spectrum is the same as that of ethyl
nitrate, indicating that the nitric acid exists as unionized HN03• A third spectrum
is characteristic of sulfuric acid solutions of nitric acid and its esters, indicating
that here nitric acid does not exist as the nitrate ion or as unionized nitric acid.
 The Raman spec
1 In the literature, this ion is more frequently called the nitronium ion. Here, the
Chemical Abstracts nomenclature is used.
62 UNIT PROCESSES IN ORGANIC SYNTHESIS
trum of nitric acid in sulfuric acid is even more informative. The nitr ion,
being isoelcctronicl with carbon dioxide and the azide ion, Na-, shou be a
linear centrosymmetric ion; hence its Raman spectrum should consi of one
strong and polarized line. From the known vibrational Raman fr quencics of
CO2 and Na-, this line should have a frequency somewh grcatcr than 1,320
cm-I. Solutions of nitric acid in sulfuric acid and
perchloric acid have only one Raman line owing to nitric acid which occu at
1,400 cm-1 and which is polarized.
The fact that the ion derived from nitric acid in sulfuric acid is pOi tively
charged has been demonstrated by electrolysis experiments in whi( it was
found that the nitric acid migrates from the anode to the cathod Finally, it is
noteworthy that several crystalline nitryl salts have been is, lated and
characterized, e.g., N02+NOa-, N02+CI04-, and N02+HS20 7
The relationship between the amount of water present in sulfuric ac and the
per cent of nitric acid molecules that are ionized to form nitro ions is shown in
Fig. 4-l,2 In solutions weaker than 86 per cent sulfur acid, the ionization of
nitric acid is very slight but rapidly rises as tl sulfuric acid becomes more
concentrated. In about 94 per cent sulfur acid, the nitric acid is practically
completely ionized to nitryl ion.

100

::? 80 u
0
u
~
c

'0 60
cQ
~

.~
c Q 40
C
~
Q;
a._
20

86 88 90 92 94
Per cent H
2S04
FIG. 4-1. The effect of water on the ionization of nitric acid in sulfuric acid.

1 Molecules, radicals, and ions having the same number of atoms and valence electrons are

called isoelectronic. Species which are isoelectronic have similar shapes; thus since CO, and
Na- are linear and centrosymmetric, the nitryl ion is likely to be linear and centrosymmetric.
2 BONNER and WILLIAMS, Chemistry & Indu8try, 1951, 820.
NITRATION 63

Nitryl ions have also been detected spectroscopically in 100 per cent nitric
acid, but their concentration is very low. At -40°C, nitric acid is said to contain
1.2 per cent 1\02+, 1.7 per cent NO,,-, and 0.5 per cent H30+. Nitric acid in
acetic anhydride and nitric anhydride in 100 per cent nitric acid have been
shown to yield N02+ and N03-.
III. AROMATIC NITRATION
The nitration of aromatic compounds! can be represented by the equa
tion

ArH + HNO, ---> ArNO, + H20

Orientation. The nitration agent is an electrophilic reactant; hence re action
will be favored at the carbon atom of the aromatic ring where the electron
density is greatest. When the aromatic compound to be nitrated contains a
substituent, the nitro group can enter at the ortho, meta, or para positions. The
relative amounts of these isomeric products will de pend upon the substituent,
since the substituent has a profound effect on the electron densities around the
various carbon atoms. Certain substitu ents cause the electron density to be
greater at the ortho and para positions than at the meta positions; hence they
yield nitration products in which the ortho and para isomers predominate. Other
substituents cause the electron density to be greater at the meta positions than at
the ortho and para positions. These are therefore meta directing. The isomer
distribu tion arising from the nitration of various monosubstituted benzenes is
shown in Table 4-1.
TABLE 4-1. NITRATION OF VARIOqS MONOSUBSTITUTED BENZENES·

Percentage of
Group already present
Ortho Meta Para

F .................... 12.4 Trace 87.2 CI. .................... 30.1 Trace

69.9
Br ..................... 37.6 Trace 62.4 1 ..................... 41.1 Trace
58.7
CH •....................... 58.8 4.4 36.8 CH,CI. .................... 40.9 4.2
54.9 CHCl •..... 23.3 38.8 42.9 CCla ...... ............ 6.8 64.5 28.7
COOC.H6. ........... 28.3 68.4 3.3 COOH ........... 18.5 80.2 1.3
0.25
NO, ........... ............ 6.4 93.2 ---------------_. •
HOLLEMAN, Chem. Rev8., 1, 187 (1925).
1 For excellent discussions on aromatic nitration, see GILLESPIE and MILLER, Quart. 'lev.,

2, 277 (1948); INGOLD, "Structure and Mechanisms in Organic Chemistry," p. 269, Jornell
University Press, Ithaca, N.Y., 1053. .
64 UNIT PROCESSES IN ORGANIC SYNTHESIS
 Naphthalene Series. In the naphthalene series, it is possible to have two
different mononitro derivatives, viz., the alpha and beta compounds, also known
at I-nitronaphthalene and 2-nitronaphthalene. Upon nitration, the first nitro
group enters almost exclusively into the alpha or 1 position; a second nitro
group enters into position 5 or 8.

NO, NO,

Anthraqllinone Series. In the anthraquinone series, nitration with mixed acid

containing a slight excess over one equivalent of nitric acid results in the
formation of three nitro derivatives. These are the alpha substituted
l-nitroanthraquinone and the l,5-and 1,8-dinitroanthraqui nones. Unless the nitric
acid ratio and concentration are sufficient to en sure considerable dinitration,
some anthraquinone remains unreacted.
Theory of Aromatic Substitution. The development of the theory of orientation in
aromatic substitution, which was due largely to R. Robinson and C. K. Ingold
and their coworkers, is one of the most interesting chap ters in organic chemistry.
According to the theory, a substituent influences the electron density in two
important ways-by the inductive effect (-I when it attracts electrons and + I
whcn it repels electrons) and by the. mesomeric effect (-l\i or +M). The
inductive effect is associated with the dipole moment of the compound C6H5 - X.
If X is at the negative end of the dipole, it will draw electrons from the ring and
produce a - I effect, indicated by a straight arrow C6H5 --. X. If X is at the
positive end of the dipole, it will increase the electron density in the ring and pro
duce a + I effect. Groups which produce a - I effect in order of decreas ing
strength are -NMe3+, -N02, -COOEt, -halogen. Groups which produce a + I
effect are - 0- and alkyl. The + I effect causes all the posi tions in the ring to be
more reactive than in unsubstituted benzene, the ortho and para positions being
made somewhat more reactive than the meta. The - I effect reduces the reactivity
of all positions in the ring relative to benzene, the effect being greater in the ortho
and para positions, leaving the meta positions to be more reactive. Thus the
nitration of toluene produces predominantly ortho- and para-nitrotoluene, while
the nitration of nitrobenzene produces predominantly meta-dinitrobenzene.
NITRATION 65

The effect of a side chain between the substituent and the ring is to reduce the
effect of the substituent. This is illustrated in Table 4-2.

TABLE 4-2 PERcENTAQE OF META DERIVATIVES FORMED IN THE MONO NITRATION

 OF NITROBENZENE AND ITs SIDE-CHAIN HOMOLOGUES

O-CHzNOz O-CHtCHtNOt 13%

O-Nfh 93%
67%

Substituents which have an unshared pair of electrons can increase the

electron density in the ring by the mesomeric effect, +M, which is indi cated by
curved arrows. Other substituents can decrease the electron density in the ring
by th!) - M effect. These effects are illustrated in the accompanying diagrams.

{:o'\o R-N f \') I _

R
Substituents which show a +M effect activate all the positions in the ring
relative to unsubstituted benzene, the effect being more pronounced at the ortho
and para positions than at the meta positions. Similarly, the sub stituents
exhibiting the - M effect deactivate all the positions, the meta being less
deactivated than the ortho and para. Groups which show a + I and +M effect
will obviously facilitate substitution and be ortho-para directing, and groups
which have a -I and -M effect will make substitu tion more difficult and he meta
directing. When the two effects are in opposition, i.e., +1 and - M or - I and + M,
the net result will be more difficult to predict. The acetoxy, methoxy, and
acetamino groups all show the - I, + M effects. The + M effect is much more
powerful than the - I effect, and so phenyl acetate, anisole, and acetanilide are
readily nitrated and the products are predominantly the ortho-para nitro deriva
tives. The halogens als() produce the - I and + M effects, but in these cases the
two effects are ()f comparable importance. The halogenated ben zenes are
nitrated more ISlowly than benzene itself because of the - I ef fect, but the nitro
group ~nters the ortho and para positions because of the + M effect. Table 4-3
contains a classification of orienting substituents.
Ortho:Para Ratios. In the preceding paragraphs, the substituents were
classified as ortho-para directing or meta directing, the difference being
explained in terms of +1 and - I effects and + M and - M effects. The effect of a
substituent on the ratio of ortho to para nitration products is also explained by
thestl factors and an additional factor, the size of the substituent, which we shall
call thc steric factor.
A substituent will make the ortho positions less accessible to the incom ing
nitro group according to its size. The larger the substituent, the less
66 UNIT PROCESSES IN ORGANIC SYNTHESIS

TABLE 4-3. CLASSIFICATION OF ORIENTING SUBSTITUENTS

Electroni Example Effect on
c orientation
mechanis
m
 Ph<-R Ph <-- CH, Ortho, para

'" '" Ph<- COO- Ortho, para,

or meta
Ph<-R Ph--- NH,+ Meta
Ph-. R '" Ph -. Meta
'" Ph COOEt
,..--... Meta
Ph -> SMe2+
-.R ,..--... Ortho, para
Ph -> Cl
,..--... Ortho, para
,..--...
Ph -> OMe
Ph-.R ,..--... Ortho, para
Ph +- 0-
,..--...
Ph+-R

Type Reactivity +I ....... .... .. . Activation

+1, -M .. .......Activation or deactivation

-1. .............Deactivation -I, -M .........Deactivation Deactivation

-I, +M .........Deactivativn Activation

+1, +M .........Activation

accessible will the ortho positions be and the smaller will be the ortho:para
ratio of the products. Evidence for this effect can be seen from the results of
the mononitration of alkylbenzenes which are shown in Table 4-4. As
TABLE 4-4. PROPORTION OF PRODUCTS FORMED IN THE NITRATION OF ALKYLBENZENES

I PhCH, PhC.H. PhCH(CH,). PhC(CH,),

Ortho ................... 11.8

57 55 14
3.2 0 0
40 45 86

Meta ....... . . . . . . . . . . . . . 8.7 Para .................... 80

the alkyl group is increased in size, the yield of ortho-substituted product

decreases and the ortho:para ratio decreases. The rate of attack on the
individual nuclear positions of toluene and t-butylbenzene (the rate for one
position of benzene being taken as unity) presents the following compari
son:
CH, C(CH.).
I I
5.()5.5
420'7'
42
4·V·
1 . 2. ~ .5 O
75.
58.

The polar effects I and M influence the ortho: para ratio because they are
transmitted to the ortho and para positions to different extents. The inductive
effect is as strong or stronger at the ortho positions than at the para position,
whereas the mesomeric effect is much stronger at the para
NITRATION 67

than at the ortho positions. Thus a compound in which the + I effect is

dominant will, upon nitration, yield a larger ortho: para ratio of products than
will a compound in which the + M effect is dominant. Similarly, a
compound in which the - I effect is dominant will give a lower ortho: para
ratio of nitration products than will a compound in which the - M effect is
dominant. By way of illustration, the products of the nitration of phenyl
halides shown in Table 4-5 are cited. The - I effect decreases in

TABLE 4-5. NITRATION PRODUCTS OF PHENYL HALIDES

PhF PhCl
PhI
12 30
87 70

Per cent ortho 41

Per cent para. 59

going from PhF to PhI; hence the corresponding increase in yield of ortho
isomer.
The ortho: para ratio is also influenced by the nitration medium in a
manner which cannot be explained by the Robinson-Ingold theory. The
isomer distribution which results from the nitration of aniline and anilides in
several nitrating media is shown in Table 4-6.
The nitration of acetanilide with mixed acid yields nitroacetanilides in
which the ortho:para ratio is less than 0.1. When the nitration medium is
nitric acid, this ratio is 0.7, when acetyl nitrate in acetic anhydride is used, the
product is almost entirely o-nitroacetanilide. No satisfactory explana
tion has been given for these results. The 40-50 per cent yield of m-nitro
 aniline that results from the nitration of aniline in mixed acid or in nitric acid
can be explained in the following manner. In the strong acids, nitric and
sulfuric acid, aniline is largely ionized. ,
CeH.NH. + HA .... CeH.NH. + + A
TABLE 4-6. NITRATION OF ANILINE AND ANILIDES·

Nitration with 80 Per Cent HNO. in Gla.cial Acetic Acid

Per cent

Para Ortho Meta.

. 64 36
70 30
72 28
74 26
75 25
79 21
84 16

Ratio
Amine ortho:para
Toluenesulfonyl. ........ .' .......... .
Free amine ........................
Acetyl. " ... , ..... , ..... , ......... . • According to LAUlIIB, J. prakt. Chem., 137,
175 (1933).
Benzoyl. .......................... .
1 :1.8 1:2.3 1:2.6 1:2.8 1:3.0 1:3.8
Formyl. .......................... .
1:5.3
Chloroacetyl. ......................•
Amino oxalyl. ..................... .
68 UNIT PROCESSES IN ORGANIC SYNTHESIS

TABLE 4-6. NITRATION OF ANILINE AND A:,ILIDES (Continued)

Nitration with 94 Per Cent HNO, in Concentrated H2S04

Para Ortho
Amine Ratio
50 1 ortho:para
92 8
Free amine. 93 7
Acetyl. ...... . . . . . . . . . . . . . 94 6 1:11.5
. . Benzoyl. ...... . . . . . . . . . . 96 4 1:13.3
. .. ..... 98 2
98 2
Per ce

Formyl .... ....................... . Amino oxalyl.

...... 0·····.·····
Chloroacetyl. ... . . . . . . . . .. . ... . . .. .
Toluenesulfonyl. . . . . . . . . . . . Chloroacetyl ...................... . 73 27 Oxalyl.
... . ..... ' ..... " ........ . 85 15 Toluenesulfonyl
Nitration in lINO,
.................... . 88 12

Temperature, 20°C; time, 24 hr.

1:15.6
1 :24.0
1:49.0
1:49.0
Per cent
Amine

Ratio
ortho:para

1 :1.4
1: 1.5
1: 1.9
Para Ortho Meta 1:2.7
1:5.7
Aniline nitrate .............. . 56 4 40 Acetyl. . .
1:7.3
. . . ....... . 58 42 Benzoyl. .........................
. 60 40 Formyl. .......................... . 65 35

The anilinium ion is strongly deactivated because of the - I effect of the - NH3+
substituent. Its nitration will yield the meta isomer. The free amine will be
strongly activated because of the + M effect of the - NH2 group, and it will
yield the para isomer. In the relatively weak acid, acetic acid, the concentration
of the anilinium ion will be much less than in the stronger acids; hence no
significan.t amount of the meta isomer is formed.

IV. KINETICS AND MECHANISM OF AROMATIC NITRATION The kinetics of the

nitration reaction depend upon the reaction medium. Let us consider first the
reactions in strong sulfuric acid. Compounds which are nitrated at a
conveniently measurable rate in this system are those which have strong -I and
-M effects such as nitrobenzene, anthraqui none, and ethyl benzoate. The rate
of all these nitrations is proportional to the concentration of added nitric acid
and of organic substrate
Rate = k (HNO.) (ArH)
NITRATION 69

The effect of the amount of water present upon the reaction rate is shown in Fig.
4-2. The reaction rate rises sharply with increasing sulfuric acid concentration
and reaches a maximum at about 90 per cent H2S04 and then falls off at higher
 acid concentrations. It was first suggested by Westheimer and Kharasch 1 that
the rise in rate with increasing acid strength when the acid in less than 90 per
cent is due to the increase in concentration of nitryl ion. It was shown that the
equilibrium constant for the ionization of tris p-nitrophenylcarbinol increases in
the same way as the rate constant, as shown in Fig. 4-2. Since triphenylcarbinols
ionize in sulfuric acid to yield carbonium ions in a manner analogous to the for
mation of nitryl ions from nitric acid
(Ar)3COH + 2H2S04 -> (Ar)3C+ + H30+ + 2HSO.-

it was believed that both ionization constants would show a similar de pendence
on acid strength. This conclusion seems justified in view of the strong evidence
for the nitryl ion given in the paragraphs on nitrating agents.
+\r-----.------.------,-----,------,------,+I

00

:r:
0 0::

0 ..J 25'C (log k- f)

c>
::::. B. Dinitromesitylene at + !E -\ - \ c>
C. CH -C:::>·N(CH
3)3 +
0
3
...J

I. Ionization of
(N0 2·C::>- )3 COH
2. Ionization of
(Me3 N + -c:::>- )3 COH

-2
-2

70
Per cent H2S04

FIG. 4-2. The effect of water on the rate of nitration and on the ionization of triphenyl
carbinols.
1 KllARASCH and WESTHEIMER, J. Am. CJu:m. Sac., 68, 1871 (1946).
70 UNIT PROCESSES IN ORGANIC SYNTHESIS
The generally accepted mechanism which is compatible with the data may
be represented by the following series of reactions:
fast
 HNO. + 2H2SO, --> N02+ + H2SO.

H.O+ + 2HSO,- slow

(Step 1)

ArH + N02+ --> ArHN02+ (Step 2) (Step 3)

ArIINO,+ + HSO,- --> ArN02 +

There still remains to be explained the decrease in rate with increasing acid
strength when H2S04 > 90 per cent. It was originally thought that step 3 in the
above mechanism was also slow and that the decrease in rate was due to the
decrease in concentration of bisulfate ions which served to remove the proton
from the positively charged intermediate formed in step 2. This theory was
found t be untenable from experiments with isotopically labeled aromatic
substrates. If the removal of hydrogen were a slow step, then the nitration of
aromatic compounds in which the hy drogen was replaced by deuterium or
tritium should proceed more slowly. It has been found experimentally that the
tritium-containing compounds are nitrated at the same rate as the
hydrogen-containing substances, indi
cating that the proton-transfer step could not be rate determining. A plausible
explanation for the decrease in rate at higher acidity has been given by
Gillespie and MilIen.! An interaction occurs between the organic substrate and
the sulfuric acid which decreases the electron den sity in the ring and hence
probably a hydrogen-bond
decreases the reactivity. The interaction is
formation. The hydrogen bond between nitro
o benzene and sulfuric acid can
be pictured Ar./'

" O--HOSOaH. The

strength of a hydrogen bond increases with the acidity of the hydrogen donor,
which in this case is sulfuric acid. Decreasing the water content of the acid
would thus be expected to increase the interaction between acid and organic
substrate, which would result in greater withdrawal of electrons from the ring.
Nitration in Organic Solvents. In the organic solvents nitromethane or acetic
acid, with nitric acid in large excess, the kinetics of the nitration process
depend upon the aromatic compound being nitrated. Compounds such as
nitrobenzene or ethyl benzoate which possess strongly deactivating groups are
nitrated at a rate which is proportional to the concentration of the substrate,
i.e., the reaction is first order. Compounds which are more reactive than
benzene, such as toluene, xylene, and p-chloranisole, react at a rate which is
independent of the concentration of the substrate, i.e.,the reaction is zero order.
The rate of the reaction with all substrates which show zero-order kinetics is
the same. For substrates of intermediate re
i GILLESPIE and MILLEN, loco cit.
NITRATION 71
activity, there is a dependence of rate on concentration, the kinetics being
intermediate between first and zero order.
The most reasonable interpretation of these results is as follows. The
formation of nitryl ions probably takes place thus:
2HN03 -> H2N03+ + N03-
H2NO,+ -> H20 + N02+
The first step, which represents the transfer of a proton from one nitric acid
molecule to a.nother, is very rapid. The rate at which the second step, the
formation of the nitryl ion, takes place depends upon the medium. As we have
seen, in a strongly acidic, highly polar solvent like concentrated sul
furic acid, this takcs place very quickly. However, in less strongly acidic media
such as acetic acid or nitromethane, this step can be relatively slow. The
zero-order kinetics shown by highly reactive substrates in these solvents
indicate that the nitration step
N02+ + ArH -t ArNO,+ + H+
is fast compared 'with the rate of formation of N02+. In agreement with this
hypothesis is the fact that all these highly reactive compounds are nitrated at
the same rate, which is the rate of formation of the nitryl ion. In the nitration
of the aromatic substrates of low reactivity, the formation of the nitryl ion is
fast relative to the nitration step which is rate determin
ing. Each compound in this class reacts at its own characteristic rate. In the
nitration of compounds of intermediate reactivity, both steps occur at
comparable rates.
Nitration in Aqueous Nitric Acid. A similar pattern of behavior is found in
aqueous nitric acid.l Highly reactive substrates ~how zero-order kinet ics, and less
reactive compounds show first-order kinetics in about 40 mole per cent
aqueous nitric acid. The rate of nitration of the reactive com pounds is the same
as the rate of exchange of 0 18 between HNO!8 and H2016. This constitutes the
most convincing evidence for nitration by nitryl ion in aqueous solution. The
rate-determining step in each reaction is the formation of the nitryl ion. The
exchange of oxygen between nitric acid and water occurs in the following
steps:

2HNO~8 -...., H,NOl8+ + Nol8-

flUlt

slow

H2NOl8+ -_' H20'8 + NO:8+

The nitryl ion after being formed by the above reactions reacts with water
quickly:

Effect of Nitrous Acid on Nitrations. Nitrous acid or nitrogen dioxide in

certain instances exerts an inhibiting effect and in other instances exerts a
catalytic effect on aromatic nitrations. The inhibiting effect is observed
'BURTON and HALEVI, J. Chem. Soc., 1952,4917.
72 UNIT PROCESSES IN ORGANIC SYNTHESIS
in the nitration of compounds having no activating groups, which reactions are
necessarily carried out either in strong nitric acid or in mixed acid. In these
media the nitrous acid forms the nitrosyl ion, NO+, HNO, + HNO. ---> NO,- +
NO+ + H,O
HNO, + 2H,SO, ---> H,O+ + 2HSO,- + NO+

which decreases the concentration of nitryl ions (see equations, p. 61) and thus
reduces the reaction rate. The catalytic effect is observed in the nitration of
reactive substrates such as anisole or dimethylaniline, which are nitrated in
relatively weak nitric acid where the nitryl ion concentration is low. The
catalysis is due to the formation of a nitroso compound which is oxidized to
the nitro compound according to the equations
ArH + NO+ ---> ArNO + H+
ArNO + HNO, ---> ArNO, + HNO,

Because the nitrosyl ion is a much weaker electrophilic reagent than the nitryl
ion, it is able to react only with very reactive aromatic compounds such as
anisole or dimethylaniline.
Two conditions are therefore necessary for catalysis by nitrous acid: (1) The
substrate must be sufficiently reactive so as to be susceptible to attack by
nitrosyl ion, and (2) the reaction medium must be such that the concentration
of nitryl ions is very low, thus allowing the nitrosyl ions to compete favorably
for the substrate.
Oxynitration. An interesting reaction occurs between benzene and ap
proximately 50 per cent nitric acid containing 0.2 molar mercuric nitrate
which yields up to 85 per cent dinitrophenol and picric acid. This process is
known as oxynitration. It has been shown by Westheimer and co workers! that

this reaction occurs in the following steps: O-HgNOs ~ O-N=O

2NO")
I~~~~ (>50%)

1 WESTBEIMER, SEGAL, and SCHRAMM, J. Am. Chem. Soc., 69, 773 (1947).
 NITRATION 73

The benzene is initially converted to phenylmercuric nitrate which reacts with

nitrogen dioxide to yield nitrosobenzene. Each of these intermediates has been
isolated from the reaction mixture. The nitrosobenzene can react in two ways.
In nitric acid weaker than 50 per cent, it reacts with 2 moles of nitric oxide to
form phenyldiazonium nitrate, a reaction first discovered by Bamberger. The
diazonium salt is convertp,d by water to phenol, which is nitrated in steps to
the final products. In nitric acid of greater than 50 per cent concentration, the
nitrosobenzene is converted directly to p-nitro
phenol without going through the diazonium compound. The p-nitro phenol is
then nitrated further to give the dinitrophenol and picric acid.

v. NITRATION OF PARAFFINIC HYDROCARBONS

Gas-phase Reactions. In contrast to aromatic hydrocarbons which are
susceptible to attack by electrophilic reagents such as the nitryl ion, the
paraffins are quite inert to such reagents. The paraffins, on the other hand, are
susceptible to attack by certain atoms and free radicals. The nitration of these
compounds as practiced commercially is carried out in the vapor phase and at
temperatures of 350-450°C; it is undoubtedly a free-radical reaction. Nitric
acid of 70 per cent strength or less is generally used, although nitrogen dioxide
can also be used. A characteristic feature of reactions involving alkyl radicals
is the great variety of products formed. This is clearly shown by the nitration
of 2-methylpentane which yields all the possible mononitration products that
might be formed by breaking any one of the bonds present and introducing a
nitro group at the point of cleavage. The products are nitromethane,
nitroethane, 2-nitropropanc, 2-nitrobutane, I-nitroisobutane,
I-nitro-3-methylbutane, 2-nitro-3-mcth ylbutane.
NITRATION PRODUCTS OF ISOPENTANE \,
C

O~NC-C-C-C +C
C-C-C-C
C
4
,~-"
Substitution C-C-C-C
C----C----C----C 123 of H

~" FiSSion~1
+ Fission Fission
~~~ ~~~ ~~~ C

between, between between

C-C-C-C-NO,
C
C-NO, + c-<?-c-c c-<?-c + C-C-NO, C-NO, + C-C-C-NO,
74 UNIT PROCESSES IN ORGANIC SYNTHESIS
The nitration produces only mononitroparaffins and no significant amounts
of polynitro compounds. Although cleavage of the carbon skeleton occurs, as
shown above, no rearrangement of the carbon skeleton has been found to
occur.
The reaction is carried out by passing the reactants through the reaction
chamber in a flow system. The products are condensed and distilled. As a a
result of a systematic study, the following facts have emerged.!
1. There is an optimum temperature at which the highest yield is ob tained.
Using butane and concentrated nitric acid in molar ratio of 15: 1, and a
contact time of 1.6 sec, the results shown in the accompanying table were
obtained.
405 425
15 36
.1 2.9

Temperature, °C. . . . . . . . . . . . . . % Conversion of nitric acid. . . . . . . . . .

% Yield of
435
RNO,: based on butane..
22
1.4

2. The addition of oxygen increases the yield based on nitric acid but also
increases the oxidation of butane. These effects are shown in Fig. 4-3.
Oxygen also increases the yield of nitromethane and nitroethane and
., c:

:E
50 J2
e 0 '" .J:>
0. ~ '0 c Q)
'0
40 E
li;
a.

E :E - a.
0

~ ~o
'c ~o 0

c: ~ 0 'c
.~
.. > 20 c:

0
- c:
Q) Q)
uu

u
 ~ 10 10 ~ 0.. 0..

23
Moles 02/HN03

FIG. 4-3. The effect of oxygen on the nitration of butane. Butane: nitric acid ratio - 15. T -
425°0.
J BA.CHMAN, HA.SB, and ADDISON, J. Org. Chem., 17, 906 (1952).