LEON S.

KOWALCZYK

INTRODUCTION

Department of Chemical Engineering

University of Detroit U. S. A.

The designer of a chemical reactor must possess

a thorough knowledge of the subject and be equipped
with appropriate tools securing fast and reliable
results.
Basic knowledge of the reactor design problem
includes:
(1) Chemical Engineering Thermodynamics (heats
of reaction, chemical equilibria);
(2) Reaction kinetics for homogeneous and heterogeneous systems;
(3) Laws governing transport processes in the
presence of the chemical reaction;
(4) Fundamentals of optimization, stability and
control analysis; and
(5) Principles of process economics.
Designer's tools comprise:

DESIGN
OF THE CHEMICAL
REACTOR

(I) Imagination ;
(2) Ability to analyze and synthesize; and
(3) Sufficient knowledge of advanced mathematics
including the use of computers.
Design of the chemical reactor results in an
appropriate size of the reactor necessary for the
required rate of production at the best operating
conditions. The term best operating conditions
usually refers to:
(1) the most economic (profitable) commercial
operation based on,
(2) the best operating conditions with regard to
the stability and control of the process.
As a rule, then, the final size of the reactor is
controlled by the economy of the process.

The design of the chemical reactor is based on a detailed

study of the process, which together with appropriate design
procedures and analysis leads to the determination of reactor
size at the best operating conditions for the desired rate of
production. The detailed study of the process comprises: the
reaction kinetics; effect of physical phenomena accompanying
the reaction and interaction of the operating parameters on
the reaction rate and yield; materials of construction and

( 1 ) Presented at CHEMPOR` 75 held in Lisbon, 7-12 September

1975 at the Calouste Gulbenkian Foundation Center.
Papers presented at this International Chemical Engineering
Conference can be purchased directly from Revista Portuguesa
de Qumica (Instituto Superior Tcnico, Lisboa 1, Portugal)
at the following prices per volume sent by surface mail,
postage included (in Portuguese Escudos):

process economics. This study is followed by the formulation

of the process mathematical model which is used to calculate
the reactor size for selected set of process parameters. Finally,
the optimal reactor size is determined for the best operating
conditions based on the mathematical analysis of the model,
involving optimization, stability and control, and economic
considerations.

Rev. Port. Qum., 17, 15 (1975)

Whole set
Transport processes
Reaction engineering
Environmental engineering
Management studies

500
200
150
150
150

This paper was presented at the Reaction engineering section.

15

LEON S. KOWAL.CZYK

2 BASIC METHODS OF REACTOR DESIGN

Two approaches to the reactor design may be
recognized:
(1) The scaling-up method; and
(2) The mathematical model method.
The first method, being more expensive and less
reliable, is losing its applicability to the advantage
of the mathematical model method. Accordingly,
this paper deals exclusively with the second method
of reactor design.

Circulating
pump

Fig. 1
Batch reactor with external heating

(2) Semi-batch reactor; and

3 THE CHEMICAL REACTOR
For better comprehension of the further design
considerations, let's briefly consider essential characteristics of chemical reactors.
The chemical reactor is a vessel in which the
chemical reaction takes place. It is furnished with
heating and mixing equipment, and instruments
permitting control and maintenance of desirable
reaction conditions (temperature, pressure, flow
rate, initial reactant concentration).
Reactor classification may be based on various
methods of reactor operation and properties of
the reacting system.
The methods of reactor operation depend upon:
(1) Mode of reactant contacting;
(2) Heat exchange with the environment;
(3) Steadiness of operation;
(4) Stability of operation;
(5) Character of mass, heat and momentum distribution in the reacting system (reactor parameters);
and
(6) Degree of deviation from an ideal operation.
Properties of the reacting system that may be
used for reactor classification are the following:

(3) Flow reactors

no recycling
tubular
\ with recycling
single
g
CSTFR
\ in series

These basic types of homogeneous reactors are

shown in figs. 1 through 5. Descriptions of their
characteristics and operations may be found in
many texts on Chemical Reaction Engineering.
Similar methods of reactant contacting are used
in heterogeneous reactions. However, the application
of tubular flow reactors to heterogeneous systems
is more diversified. The heterogeneous tubular
reactor may have various geometric forms (tube,
kiln, tower, tunnel, conveyor) in which the solid
and fluid phase may contact in + ,
or  .
4 STEADINESS VS. STABILITY
In the steady state reactor operation, the physical
and chemical properties (temperature, pressure,
velocity, density, reaction, participant concentration,
pH, etc.) at any point within the reaction system
are independent of time. A good example of the
Water vapor

(1) Physical state of reaction participants;

(2) Presence or absence of a catalyst; and
(3) Type of the driving force of chemical reaction.
For the homogeneous systems we distinguish three
types of reactors with regard to the mode of reactant
contacting. They are :
(1) Batch reactor;
16

Fig. 2
Semibatch reactor for alcohol esterification
Rev. Port. Qum., 17, 15 (1975)

DESIGN OF THE CHEMICAL REACTOR

6 SELECTION OF THE REACTOR TYPE

Recycling pump

Fig. 3
Multi-tube flow reactor with recycling

steady state operation is given by a CSTFR at

constant temperature, pressure, feed rate and
feed composition. An unsteady-state operation is
exemplified by batch and semi-batch reactor operations as well as by starting up, shutting down or
purging of flow reactors.
The operation of reactor is stable when it is:

No formula is available for the selection of an

appropriate type of reactor and its mode of operation
for a given process as too many factors, such as
the rate of production, level of heat of reaction,
physical phases of reaction participants, type of
reaction, product quality, stability and flexibility
of operation, adaptability of equipment to other
purposes and process economics affect the choice
of the reactor type. However, the designer's experience, his engineering judgment and thorough
knowledge of advantages and disadvantages of
various types of reactors may be of great help in
making the final decision, which, however, will
primarily be based on the economic considerations
of the process.

(1) at steady state, and

(2) after experiencing small changes in process
variables, the reactor returns by itself to its original
state.
Accordingly, the reactor temporarily at transient
state may be stable.

Fig. 5
CSTFR's in series

5 CLASSIFICATION OF REACTORS BASED

ON THE DRIVING FORCE OF CHEMICAL
REACTION

A few primitive rules for the reactor selection are

listed below:

The occurrence of a chemical reaction may be

induced by various types of energy (driving force),
activating the dormant components of the system.
Accordingly, we recognize the:

Rate of production
Small:

moderate heat of reaction

high heat of reaction

(1) thermal,
(2) radiation,
(3) photochemical,
(4) biochemical, and
(5) nuclear reactors.

High:

batch reactor
semi-batch reactor

moderate heat of reaction

tubular flow, single

CSTFR

high heat of reaction

tubular flow with

recycle, CSTFR in
series, possibly with
intermittent cooling

Type of reaction
A good summary is given by LEVENSPIEL
Chapters 6 & 7.
Product

Fig. 4
CSTFR with a water cooling mantle

Rev. Port. Qum., 17, 15 (1975)

Product control
Better in flow reactors.
Cost of operation
Lower in flow reactors.
17

LEON S. KOWALCZYK

7 DESIGN BY THE MATHEMATICAL

MODEL METHOD

8 PRELIMINARY STUDY OF THE

PROBLEM

The mathematical model method seems to be the

most popular method of today's reactor design.
It is fast, relatively inexpensive and accurate when
reliable data on reaction kinetics, thermodynamics
and transport phenomena are available for the
process under consideration. The method begins
with a detailed theoretical and laboratory study
and is followed by the formulation of the mathematical model of the process, describing the model
dynamic behavior by a set of differential equations.
Analysis of these equations by advanced mathematical methods leads to the determination of the
optimal size of reactor operating at the lowest
cost for the required rate of production. Because
of the differential character of model equations,
the model may be directly subjected to the optimization, stability and control analysis. The increasing
popularity of the application of the mathematical
model to the reactor design is chiefly due to the
advent of electronic digital and analog computers
which enable the designer to study and analyze
the complex interactions of kinetic, thermodynamic
and transport phenomena taking place in the
chemical reactor.
The equations describing the mathematical model
are of semi-empirical nature, i.e., their form obtained
from the theoretical considerations of the process
or phenomena is well established and only the
constants and parameters and their changes associated with the variation of environment of the
reacting system must be determined by direct,
frequently independent experiment. In this way,
these equations have a physical and engineering
meaning and when applied to the mathematical
model, permit for extrapolation into unexplored
parameter regions and design the reactor, including
its stability and control, at any feasible condition.
The mathematical model design involves the
following steps:

The preliminary study of the problem comprises

laboratory experimentation and retrival and evaluation of library data in the following areas:

(1) Preliminary study of the problem;

(2) Formulation of the model and design procedure;
and
(3) Mathematical analysis of the model behavior
with regard to the reactor size, its optimization,
stability and control.
18

(1) Reaction kinetics;

(2) Thermodynamics;
(3) Mass, heat and momentum transport coefficients;
(4) Methods of reactant contacting for similar
processes;
(5) Methods of heat removal or addition;
(6) Materials of construction; and
(7) Process economics.
The purpose of the kinetic studies is to establish
stoichiometry and mechanism of reaction, character
of elementary steps in complex reactions, orders
and specific reaction rates of individual steps,
over-all rate expression, selectivity, effect of operational parameters (temperature, pressure, catalyst,
ratio of reactants, impurities, method of contacting)
and physical phenomena accompanying the reaction
on the rate of reaction and its yield.
The thermodynamic study leads to the values of
C, and C v of reacting system, heat of reaction,
heat of adsorption (gas-solid catalytic reactions),
energy of activation and equilibrium constant via
the free energy of reaction

OF= RTIn K
(I)
for various operational conditions.
The study of mass, heat and momentum transport
coefficients (diffusivity, thermal conductivity and
viscosity) in homogeneous and heterogeneous systems is necessary when a multiple gradient (dispersed
parameter) reactor model is used. Reliable data
must be established for the whole range of operational conditions. If they cannot be retrieved from
the literature or obtained from existing pilot or
commercial reactors, they must be carefully evaluated
by inter and/or extrapolation of existing data or
calculated from verified empirical formulae or
theoretical considerations [2]. The thermodynamic
and physical properties package data may also be
available from big computer systems [3].
Rev. Port. Qum., 17, 15 (1975)

DESIGN OF THE CHEMICAL REACTOR

In the study of heat removal or addition methods,

we are primarily concerned with design and operation
of appropriate heat exchange equipment. Design
of the heat transfer area for isothermal reactors
may represent quite a problem. Control of heat
removal and addition in isothermal and programmed
reactors must also be studied.
The study of constructional materials comprises
material availability, mechanical and thermal properties as well as chemical resistivity. Sometimes
search for new or less expensive materials may be
necessary.
Design activities must lead to the highest returns
or least expensive production. Accordingly, economic
optimization of reactor size for the desired rate
of production is of primary importance. The
designer, therefore, must be well acquainted with
cost of reactants, products and wastes; cost of
reactor with auxiliary equipment and control
instruments; and cost of reactor operation (energy,
labor, cooling water, depreciation, overhead, etc.).

which the elementary volume properties (point

properties) remain constant in time and space
(lumped parameter model). The distributed parameter model comprises the basic microscopic and
multiple gradient models, which are two-dimensional
for gradients in the axial and radial direction only.
The simplified gradient models (multiple gradient
models without radial gradients or without any
dispersion) are unidimensional models. The lump
parameter model is equivalent to the macroscopic
model and is zero dimensional.

9 FORMULATION OF THE
MATHEMATICAL MODEL

They describe the dynamic behavior of a reaction

model and reveal the relations among design
parameters with respect to mass, energy and
momentum transport within the reactor. The complexity of the design equations depends upon the
characteristics of the model. They reduce to a
simple set of algebraic equations for the steady
state macroscopic model.
The mathematical form of balance equations depends
upon several factors, such as:

A set of equations describing the dynamic behavior

of the chemical reactor is called the mathematical
model of the reactor. The equations themselves are
called the design equations and, in general, they
comprise the rate expression and the mass, heat,
and momentum balance equations. It should be
emphasized, however, that a complete description
of the mathematical model involves two sets of
equations :
(1) Design equations; and
(2) Boundary conditions, indicating the limits
within which the design equations may be used.
Two types of basic mathematical models are used
to represent the behavior of chemical reactors:

10 DESIGN EQUATIONS
The design equations for the chemical reactor
involve:
(1) The rate expression;
(2) Mass balances (total or for individual reaction
participants);
(3) Heat balances (for the total reactor); and
(4) Momentum balances.

(1) Character of variable change (difference, differential, or partial differential);

(2) Notation used (space coordinates, vector notation, cartesian tensor notation);
(3) System of coordinates used (cartesian rectangular, cylindrical, spherical); and
(4) Types of variables (regular, normalized or
dimensionless).

(1) Microscopic (multiple gradient); and

(2) Macroscopic.
The microscopic model is based on the properties
and behavior of the reaction system in an elementary
volume of the system and equivalently is called
the distributed parameter model. The macroscopic
model treats the reaction system as a whole in
Rev. Port. Quim., 17, 15 (1975)

11 RATE EQUATION
The rate equation expresses instantaneous rate of
chemical reaction per unit of reactor volume or
unit catalyst mass and is a basic component of mass
and heat balance equations.
19

LEON S. KOWALCZYK

The rate equation for homogeneous reactions is

usually expressed by the law of mass action. For
reactant A in the batch system, it will be:

For gas-solid fixed bed catalytic reactors

r=

(kinetic term)(potential term)

(absorption term)

n
dNA =
k ( NA ,
Vdt
\ V

V const.
(2)

dCA

dt

based on the Langmuir-Hinshelwood theory of

surface reactions.
For the gas-solid non-catalytic reactions, the rate
expression has a general form

kcCAn , V = const.
r=

(3)

where
N A moles of reactant A in the system
V volume of the system
CA concentration of A
t time
k specific reaction rate
n order of reaction

In complex, like free radical reactions, the rate

expression is based on the rate controlling step
and may have quite a complex form.
For the tubular flow reactor, the rate expression
converts to:
dCA

dFA/Fv
_

dV/Fv

dt

(6)

dFA
dV

= k^

FFA

'\n

J\

v
(4)

where
molal flow rate of A
F v volumetric flow rate through the reactor
V reactor volume, and
FA
FA

CA=

Fv

driving force
resistance

(7)

and depends upon the rate controlling step which

may be the gas film, diffusion through ash layer
or chemical reaction. In certain gas-solid reactions,
the rate of reaction may be controlled by the rate
of heat conduction through the ash layer.
In two immiscible fluid reactors, the chemical
reaction takes place at the fluid interface and the
rate depends upon the mass transfer rates in both
fluids and chemical reaction.

12 MICROSCOPIC TRANSPORT
BALANCES
The base for all balance alterations and adjustments
for practical purposes is the microscopic balance.
It assumes that the system is a continuum and
ignores the detailed molecular interactions. The
derivatives are based on mass, heat and momentum
changes in an elemental volume of the system.
The general form of microscopic transport balances
is given below:
Net _ Input through surfaces
cumulation]
[of elemental volume
ac
[

- Output through surfaces

of elemental
volume

For the CSTFR:

O CA = kcCA n

Generation inside
elemental
volume

(5)
where t is the time of residence.
The rate expression for heterogeneous reactions
may be more complex.
20

Consumption
inside
elemental
volume
(8)
Rev. Port. Qum., 17, 15 (1975)

DESIGN OF THE CHEMICAL REACTOR

A set of transport balance equations for reactant

A in a dilute binary system in a homogeneous

microscopic tubular flow reactor in cylindrical

coordinates is given in fig. 6.

Mass balance for reactant A in a dilute binary solution

v e aCA
aCAacA
aCA =
vZ
ao + az
at + ( v ` at + r
Transport through open surfaces by bulk flow

Accumulation

= DAB

I a
a2CA + a2CA
(r aCA + 1
r ar \ ar ) r2 ae 2 aZ2

+ CA

Transport through open surfaces by molecular

(I)

Generation

mass diffusion

Heat balance

C13

/ aT + Vr al'
at
ar

Vg a

+V

ae

al.
az

)_

Accumulation Transport through open surfaces by bulk flow (convection)

ar

Cr

aT
ar

+ r2

a2T
ae2

a2T

+ az

(2)

Transport through open surfaces by molecular heat diffusion

Consumption

(conduction)

Momentum balance for z component

C
Accumulation

avZ
+ v e avZ + vZ avz
v`
Or r ao az
at +

avZ

Transport through open surfaces by viscous diffusion

Op
az

a
L r 1 ar

1
av 1
+ 2
ar l
r
Z

a2vZ

a2vZ
ao2 + az ]+^ g=

Transport through open surfaces

(3)

Generation

Notation:

molar or mass density

V velocity

Cp specific heat
t time

viscosity

T temperature

D diffusivity

p pressure

thermal conductivity
CA concentration of reactant A

r radius
z axial direction of flow

Fig. 6
Microscopic transport balances in cylindrical coordinates for a tubular flow reactor
Rev. Port. Qum., 17, 15 (1975)

21

LEON S. KOWALCZYK

In deriving these equations the following assumptions

were made:
(1) Newtonian fluid;
(2) Negligible viscous dissipation;
(3) The same external force effect on each fluid
component;
D, k and C p .
(4) Constant

13 MULTIPLE GRADIENT TRANSPORT

BALANCES
Simplification of microscopic balances involving
modification of transport coefficients in balance
equations leads to the more realistic equations
which may be applied to reactor design purposes.
In the microscopic approach, these coefficients are
calculated by quantum mechanics, equilibrium and
non-equilibrium statistical mechanics and classical
mechanics. In the multiple gradient models, these
coefficients are empirical and must be determined
by direct experiment. Semi-empirical equation may
be of great value, when available. In the multiple
gradient model, one or more parameters (velocity,
temperature, concentration) may be dispersed in
the axial and/or radial direction. In the maximum
gradient model, only temperature and concentration
gradients in the axial direction are considered. All
radial gradients and dispersions are neglected.
The transport balances for the simplified gradient
models are based on the following general equation:
Net ( 2 ) rate of
Rate of
[accumulation (')

_ bulk flow
through open
surfaces

Net rate of flow

by molecular
transport through
open surfaces

Net rate of flow

by molecular
transport through
closed surfaces ( 3)

Net
rate of
generation
in reactor ( 4 )
(9)

22

G. F. FROMENT [4] divided fixed bed reactors into

two classes: pseudo-homogeneous and heterogeneous. For the homogeneous class, it was assumed
that the concentrations of a selected reaction
participant and temperatures in the fluid phase
and inside the solid particle are the same. In the
heterogeneous model, these parameters in the fluid
phase and inside the particle are different. Consequently, the balance equations for the first class
are similar to those for homogeneous systems, but
for the heterogeneous class the balance equations
are written separately for the fluid and solid
phase.
In the practical application to the reactor design,
the multiple gradient balance equations may be
considerably simplified.
The momentum balance in batch, semi-batch, short
tubular flow and CSTFR's has no significance and
can be safely omitted in design.
The basic design equations are derived for tubular
flow reactor. However, they may readily be adjusted
to batch and semi-batch reactors by deleting
insignificant terms.
Furthermore, the heat balance equations should be
omitted for isothermal design. Finally, for adiabatic
design, the term related to the heat exchange
with the environment should be dropped out.

14 BOUNDARY CONDITIONS
Differential equations are integrated within the
boundary conditions, which may be determined
from the problem statement or actual process.
They may be given or inferred from the state of
physical transport phenomena at the system boundaries. Written in the form of mathematical equations,
they indicate whether the dependent variables at
the boundaries are at equilibrium or whether some
transport of mass, heat and momentum takes
place.

(1) Accumulation = depletion.

(2) Net input = input-output
(3) Heat transport by conduction.
Mass transport by diffusion, osmosis, molecular sieves.
(4) Generation = consumption.
Rev. Port. Qum., 17, 15 (1975)

DESIGN OF THE CHEMICAL REACTOR

Another type of boundary condition is based on

the process rate at the boundary, expressed in the
form of an appropriate driving force divided by
resistance, with an interphase transport coefficient,
like in Newton's law of cooling.
The boundary conditions and/or end (initial and
final) conditions determine the realistic limits of
applicability of differential equations. In general,
we need a set of n boundary conditions for each
nth derivative with respect to time or space.
The selection of a coordinate system for design
equations may be influenced by boundary conditions.

15 MACROSCOPIC TRANSPORT
BALANCES
The simplest reactor model is the macroscopic
model. It is formulated in the following manner:
(1) Time is the only variable;
(2) No spatial gradients are included in the mathematical description of the model;
(3) Dependent variables such as concentrations,
temperature and velocity are not functions of the
position but are expressed as average values over
the reactor volume (batch, CSTFR) or inlet and
exit streams (tubular reactor);
(4) Flow is in one direction only.
These assumptions simplify considerably the mathematical description of the model, but also conceal
a lot of information, valuable to the design.
The general form of the transport balance equation
is given in eq. (10).

[ Rate of
IL accumulation

Net rate of
bulk flow
through open
_ surfaces

Net rate of flow

by molecular
transport through
_ closed surfaces

r Net
rate of
generation
in reactor
(10)

Rev. Port. Qum., 17, 15 (1975)

16 CALCULATION OF THE REACTOR

SIZE
16.1 DESIGN RESTRICTIONS
Before starting the reactor size calculations, the
designer must carefully consider limitations imposed
on the process variables and/or design parameters
by the character of the process.
These restrictions are of two types [5]:
(1) Internal, controlled by the designer; and
(2) External, imposed by agencies, outside of the
designer control (management, government or
nature).
Internal restrictions arise from the laws of conservation of mass, heat, and momentum; certain
well established design relations (shape of the
cylindrical reactor); standardized sizes of pipes,
metal sheets and thickness of heat insulation
materials; realistic values of operating parameters
(temperature, pressure, velocity, reactant ratio) and
properties of construction materials.
External limitations include rate of production,
product quality, legal product specification, steam
pressure and temperature of the cooling water.

16.2 DESIGN EQUATIONS

Calculation of the reactor size is based on the
product distribution equations and desired rate of
production. The product distribution equations
represent integrated forms of design equations and
show the relation between the degree of conversion
of a selected reactant and design parameters like
time of residence (tubular reactor length or catalyst
weight), temperature, etc. In the case of simple
operations, like a single isothermal CSTFR, this
relation reduces to the dependence of the exit
concentration from the time of residence.
It should be remembered that the design equations
involve a wide variety of equation types:
(1) Algebraic equations for certain macroscopic
models like the steady state isothermal CSTFR
(difference equations);
(2) Differential equations for batch and plug flow
tubular reactors;
23

LEON S. KOWALCZYK

(3) Differential-difference equations for . transient

operations of CSTFR's;
(4) Partial differential equations for semi-batch,
transient and microscopic and multiple gradient
reactors.
Although a dozen of analytical methods are available
to solve a set of ODE's, the numerical methods
and computers must be used in the case of more
complex reactions. The degree of difficulty is
rapidly increasing for transient and multiple gradient
models.
The exit conversion, time of residence and the
desired rate of production are the basic values,
necessary to calculate the reactor size at specified
operating conditions.

17 MATHEMATICAL ANALYSIS OF
DESIGN EQUATIONS
The purpose of mathematical analysis of design
equations is to better comprehend dynamic behavior
of the model and to establish the effect of interactions by process variables and parameters on
the yield of reaction, product quality, process
economics, reactor stability and control. Four
approaches to the analysis of design equations
will be discussed in the following sections. They are :
(1)
(2)
(3)
(4)

Process optimization;
Reactor stability;
Control; and
Statistical sensitivity analysis.

18 PROCESS OPTIMIZATION
The modern reactor design should lead to the
highest yield, highest product quality, least expensive
production or highest monetary returns. This may
be achieved by the application of the principles
of optimization in various stages of the design
project.
In engineering work we distinguish two types of
optimization criteria:
(1) Technical (yield, quality of product); and
24

(2) Economic (level of profit or investment).

It is easy to guess that the technical criteria are
nothing else but the disguised economic criteria.

19 APLLICATION TO REACTOR DESIGN

Even a brief description of methods and techniques
of optimization is far beyond the scope of this
paper. There are well presented methods and
techniques in many good books on this subject
and numerous papers published in the last decade.
The books of BEVERIDGE and SCHECHTER [5],
DENN [6] and DOUGLAS [7] are recommended.

20 OPTIMIZATION BY COMPUTER
SIMULATION
The computer approach to the study of dynamic
behavior of the reactor is called computer simulation. It is based on computer solution of design
equations leading to time-varying curves of concentrations, temperature, flow rates, initial concentration ratios, etc. and their analysis with
respect to the objective function [3].

21 CALCULATION OF THE OPTIMAL

REACTOR SIZE
The ultimate purpose of reactor design is to create
the best reactor for the desired production. Engineering optimization criteria are satisfied with the
smallest reactor producing at the highest rate the
high quality product. However, the production of
chemicals is unquestionably an economic process
and cost of the reactor and of its operation
is directly or indirectly related to the overall
economy of the plant. Consequently, the design
must also satisfy the economic criteria which may
involve the smallest unit cost of product, smallest
total cost of operation or highest gross profit of
the production unit.
To solve this problem, we have to formulate an
expression for the economic objective function
(reactor profitability) in terms of the capital invested,
operating costs, sales income, required production
Rev. Port. Qum., 17, 15 (1975)

DESIGN OF THE CHEMICAL REACTOR

rate, steady state mass and heat balances, and

values of input stream properties [1]. Optimal
solution of thus obtained equation leads to the
optimal value of conversion, which, in turn, may
be used to the optimal (economically) size of the
reactor.
Now, the optimal reactor must pass the test of
stability and control.

(4) Analysis via the transfer function (Laplace

transform space);
(5) Use of the Lyapunov methods; and
(6) Special techniques.

22 STABILITY ANALYSIS

Application of these methods to stability of chemical

reactors can be found in many fine texts on
stability. The matrix algebra method seems to be
quite popular [7]. If the optimum economic steady
state reactor design is found unstable even in
certain ranges of operation only, the change of
the operating conditions or application of an
appropriate controller is necessary.

When a system at equilibrium is subjected to the

action of a new force, the reaction of the system
may follow four ways:

23 PARAMETRIC SENSITIVITY

(1) The trajectory of the system motion, after a

small disturbance, tends to return to the original
state of the system and remains at it or in its
vicinity an infinite time. This system is called a
stable one.
(2) The system is asymptotically stable when the
trajectory of the system motion after a small
disturbance returns asymptotically to the original
state of the system.
(3) The system is asymptotically stable at large
when after any disturbance it returns asymptotically
to the origin.
(4) The unstable system, after disturbance, moves
away from the original state and never returns.
The stability of reactor is determined by the stability
analysis of the following phenomena related to the
design and operation of reactor:
(1) Stability of design equations;
(2) Stability of numerical solutions if the design
equations cannot be solved analytically;
(3) Parametric sensitivity;
(4) Multiplicity of steady states.
The stability of design equations may often be
determined by analytical methods. These involve:
(1) Input-output approach (graphical method);
(2) Use of matrix algebra;
(3) Use of the differential operator;
Rn'. Port. Qum., 17, 15 (1975)

A study of BiLous and AMUNDSON [8] revealed

that even small changes in reactor parameters,
such as wall temperature, feed dilution and wall
heat transfer coefficient may affect considerably
the reactor temperature profile and cannot be
ignored in the control design of a reactor. Provisions
should be made to keep the parameter in question
close to its critical value to obtain the desired rate
and quality of product.

24 MULTIPLICITY OF STEADY STATES

Certain combinations of feed composition and
reactor temperature lead to more than one steady
state solution. As a rule, the even numbered steady
states are unstable. The Van Heerden study of
CSTFR (I & EC, 45, 1242, 1952) revealed existence
of three steady states, whereas computations of
Copelovitz and Aris (CES, 25, 906, 1970) show
15 steady states for a catalyst particle.

25 STABILITY REGIONS
Existence of multiple steady states may considerably
complicate the reactor stability analysis. To ease
this problem, the stability region graphs are being
introduced. Such a graph, developed by McGovIN
25

LEON S. KOWALCZYK

1. 0
6
CA
CA o

0 6
.

0.4
0. 2
0

MIRK 10 i
1111EARIE
EA&
NM
mignams

22

24

30

26 2.8

32

Instability
regions

34

C P To
Feed condition CHCA
a

Fig. 7
Instability regions for a CSTFR

and PERLMUTTER [9] for the temperature dependent

CSTFR is shown in fig. 7. In this plot, the operating
conditions for the selected values of

The selection and installation of the control equipment involves the following steps:
(I) Selection of appropriate mode of control by
dynamic analysis of individual streams to be
controlled;
(2) Choice of control equipment: pneumatic or
electronic. Hence, the cost of equipment and its
maintenance might be a decisive factor;
(3) Location of appropriate controllers for particular
jobs;
(4) Adjustment of controller constants for the best
responses of the control system.

27 OPTIMAL CONTROL

CA o Initial concentration of reactant A

T o Feed temperature
C, Specific heat of the reacting system
F Volumetric rate of flow
OH Heat of reaction

Steady state optimal control. Let's suppose that

the stability analysis of our optimally designed
reactor indicates that the reactor is practically
stable and operates at steady state. However,
plant management wants to secure the highest
rate and quality of production and insists on elimination of time fluctuations in input streams like
feed composition and its temperature. In such
cases, the designer has to provide suitable controllers
to manipulate these inputs (control variables) in
order to maximize the reactor production and
ensure the highest profit for the operation.
Unsteady state optimal control. When the dynamic
stability analysis reveals that the optimal steady
state design is totally unstable or is unstable for a
fraction of time of operation, then steady state
optimal control must be used. It will be based
on the stability analysis of the system including
an appropriate controller.

26 REACTOR CONTROL

28 STATISTICAL SENSITIVITY ANALYSIS

The purpose of process control is to maintain the

desirable (optimal) producing conditions by means
of mechanical or electronic equipment. Two control
systems are in general use:

In the solution of design equations, we use numerical

values of design parameters like rate constants,
reaction orders, energies of activation, temperature
(in the case of isothermal reactors), initial conditions
and transport coefficients, whose reliability may
frequently arise some doubt. The statistical sensitivity analysis [10] explores the question of how

u
F,,Cp

in
CA
CA 0

vs. feed condition as

CpT o
OHCA o

coordinates are shown. In this plot

(1) The closed-loop system (feed-back control);

(2) The open-loop system (feed-forward control).
26

Rev. Port. Quim., 17, 15 (1975)

DESIGN OF THE CHEMICAL REACTOR

the parmeter uncertainties propagate into outputvariable uncertainties. It is based upon Taylor
series approximations for the variance and expected
value of the output variables of the model. The
first partial derivatives of the output variables
with respect to each parameter are determined from
the model structure. Determination of these derivatives and formation of the sensitivity matrix is the
essential part of the technique. The solution of the
sensitivity matrix leads to a set of linear equations,
indicating relationships between time and individual
parameters, which provide necessary information
to determine the variance of the output.
The statistical sensitivity analysis is limited to
reactors, modeled in terms of a set of nonlinear,
ordinary differential design equations.

BEVERIDGE, G. S. G. and SCHECHTER, R. S., Optimization: Theory and Practice, McGraw-Hill, New York,
1970, p. 27.
DENN, M. M., Optimization by Variational Methods,
McGraw-Hill, New York, 1969.
DOUGLAS, J. M., Process Dynamics and Control,
Vol. I, Prentice-Hall, New York, 1972.
BILOUS, O. and AMUNDSON, R. N., A. I. Ch. E. (Am.
Inst. Chem. Engrs.) J., 2, 117 (1956).
MCGOV]N,

G. and PERLUMTTER, W., A. I. Ch. E. (Am.

Inst. Chem. Engrs.) J., 17, 831 (1967) .

R. W., SCHAINKER, R. B. and DUCOT,

E. R., A. I. Ch. E. (Am. Inst. Chem. Engrs .) J., 21
ATHERTON,

441 (1975).

RESUMO

REFERENCES
[i] LEVENSPIEL, O., Chemical Reaction Engineering,
2nd ed., John Wiley and Sons, New York, 1972.
[2] HIMMELBLAU, D. M. and BISCHOFF, K. B., Process
Analysis and Simulation, John Wiley and Sons,
New York, 1968.
[3] MOTARD, R. L., SHACHAM, M. and ROSEN, E. M.,
A. I. Ch. E. (Am. Inst. Chem. Engrs.) J., 21, 417 (1975).

[4] FROMENT, G. F., Proc. 5th Europ./II Intern. Symp.

Chem. Reac. Eng., Amsterdam, 1975, p. A5-1.

Rev. Port. Qum., 17, 15 (1975)

O projecto do reactor qumico baseia-se num estudo pormenorizado do processo o qual, juntamente com as anlises e
clculos adequados, conduz determinao das dimenses
do reactor nas condies de operao ptimas e para o ritmo
de produo desejado. O estudo do processo compreende:
a cintica da reaco; efeito nas velocidades de reaco e
rendimento dos fenmenos fsicos que acompanham a reaco
e das interaces entre os parmetros operatrios; materiais
de construo e economia do processo. Depois deste estudo
formula-se um modelo matemtico do processo que se utiliza
no clculo das dimenses do reactor para um grupo de
parmetros do processo pr-seleccionado.
Finalmente faz-se o clculo da dimenso ptima do reactor
para as condies de operao ptima, com base na anlise
matemtica do modelo que envolve a considerao de aspectos
de optimizao, de estabilidade, de controle e econmicos.

27