6.

The Choice of a reactor for a particular reaction

The question is always asked about:
 how we choose the right type of reactor for the desired
production / reaction / ?.
 To answer the question we have to look into different factors affecting
the choice of reactor such as:
• The reaction type
• Operating temperature
• Planned scaled of production
• Cost of equipment and operation
• Working safety, stability and flexibility
• Equipment life expectancy
• Easiness of convertibility of the equipment
 No net formula is available to select the reactor, we can only use our
technical judgment and the affordability of the process including the
economic factor. 1
 Looking to the above factors, we can always choose the reactors based on
two criteria namely the character of the reactor and the general
knowledge about the kinetics of the type of reaction
6.1 Characteristics of individual reactor
 This is based on their process of operation namely batch or
continuous operation.
6.1.1 Batch operation
 often used for small scale production having long reaction times.
 The advantage of such type of reactor is the cost of the equipment and
flexibility of operation
 The disadvantage is the relatively extensive manual operation of
emptying, cleaning and refilling required.
 this process is difficult to use for automation. restricted only to produce
small amount of material
2
6.1.2 Continuous operation
Advantage
 can make products 24 hours a day,
 parameters such as temperature and product quality
controlling are easier.
 little supervision, suited for the application of automatic control

6.2 Kinetics behavior

 The main criteria which can be used to choose the type of an appropriate
reactor is the kinetic behavior which is governed by the reaction type.
 Reaction type is categorized in two distinct classifications namely single and
multiple reactions.

3
 6.2.1 Single reactions
The main characteristic of a single reaction is that it requires only one rate
expression to describe its kinetic behavior.
A  Product
 The product distribution of such reaction is fixed and to compare the reactor
type only depends on the size of a reactor or a unit output of a reactor which
is a function of conversion.
6.2.1.1 Comparison of batch, plug-flow and tank reactor

A) Batch and plug-flow reactor

Reactor size for specified conversion is obtained by using the material balance.
For constant density system, we have equations
C A dC
batch reactor : t  - A

C A,0 ( - rA )
C A dC
plug flow reactor :  F  - 
A
C A,0 ( - rA ) 4
 Both the equations show an element of fluid which reacts for the same
length of time in the batch and as well as in the plug flow reactor.
 This demonstrates that the same size of these reactors is needed to do a
given job and it is possible to relate the performance capabilities of these
reactors

B) Plug flow reactor and tank reactor

Generally ,
 the capacity of the tank of a CSTR is always smaller than that of a plug
flow reactor, if the reaction volume and the condition are identical in
both the cases.
 for a given duty the ratio of the size of tank and plug flow reactors
will depend on the form of a rate equation (i.e. order of reaction).

5
Again recalling modified material balance equation gives us,

 V   C A,0 V  C A,0 X A (6.1)

Tank :  F,T       
F   F  ( - rA )
 V,0 T  A,0 T
 V   C A,0 V  X A dX
(6.2)
Plug flow :  F, P        C A,0  A
F   F 
 V,0  P  A,0  P 0 ( - rA )
For a first order reaction
( rA )  kCA  kCA,0 (1 - X A )
The above equation (6.1) becomes
 C A,0 V  C A,0 X A XA
Tank :  F,T     
 F  kC A,0 (1  X A ) k (1  X A )
 A,0 T

For the nth – order reaction

(  rA )  kCn A  kCn A,0 (1 - X A ) n
Substituting these equations we can derives,
6
  C A,0 V  C A,0 X A XA (6.2.1)
 F,T     
 F 
 A,0 T
n
kC A,0 (1  X A ) n kC n 1A,0 (1  X A ) n
In the same manner it can also be formulated for the nth order reaction in the
plug-flow reactor
 C A,0 V  XA dX A
 F, P    
  (6.2.2)
n  1 n
F
 A,0  P 0 kC A,0 (1  X A )
To compare both the reactors, the equation (6.2.1.) and (6.2.2) are simplified
for n = 1 to give
 F,T k
Tank , X A  (6.3)
1   F,T k
- k F,P (6.4)
PFR , X A  1 - e
 The effect of the equations (6.3) and (6.4) are shown in the Figure 6.1. It is clear
from the figure that for any conversion the volume required is larger for the
CSTR.

 This is because of the nature of CSTR that the low reactant concentration is
taking place in the reactor
7
 From the Figure 6.1 one can see easily that for a small conversion there occurs
no significant change when we use CSTR or PFR. It implies that a volume
difference significantly occurs for higher conversions. As an example for
conversion say 80% or above, more than five as much volume is required for CSTR.
This is demonstrated in the Figure 6.1 depicting the ratio of volume of CSTR & PFR
versus conversion. By using the ratio of equation (6.2.1) and (6.2.2) we obtain
XA
 F kC n -1A,0  
1  X A n
 
m
XA
kC n -1A,0 dX A

F P

0 1  X A n

For

8
Figure 6.1 Comparison of performance of single mixed flow and plug
flow reactors for the nth-order reactions
9
Figure 6.2. Comparison of performance of mixed flow and plug flow reactors for
any reaction kinetics.
 For reactions with arbitrary but known rate the performance capabilities of
mixed and plug flow reactors are best illustrated in Fig. 6.2.
 The ratio of shaded and of hatched areas gives the ratio of space-times needed in
these two reactors.
 The rate curve drawn in Fig. 6.2 is typical of the large class of reactions whose rate
decreases continually on approach to equilibrium (this includes all nth-order
reactions, n > 0). For such reactions it can be seen that mixed flow always needs
a larger volume than does plug flow for any given duty.
10
 One can investigate the role of the order of reaction in reactor volume.
 For zero order reaction the reactor size is independent of the type
of reactor.
 However, the ratio of the volume increases with the order i.e.
higher volume of CSTR is required compared to that of PFR for
higher orders of reaction.

6.2.2 Multiple reactions

 Most reactions especially for organic industrial products, involve more

than one reaction. As for example, let us consider the following ethanol
manufacturing from ethylene reaction
k1
H 2C  CH 2  H 2SO 4  C2 H 5 0  S03 H
k2
H 2C  CH 2  C2 H 5 0  S03 H 
 (C2 H 5O ) 2 SO4

11
There are two basic types of multiple reactions, namely series and parallel.

Series Rxns k1 k2
A  B 
 C
k1
A  B  C
k2
B  C 
 D
 Multiple reactions are described more than one rate expression to
demonstrate its kinetic behavior. Due to the nature of the reaction, in
addition to the size of the reactor, the product distribution or yield
or selectivity of the product, are the important parameters which
describes the product to the maximum

 While designing for multiple reactions, our target is to maximize the wanted
product and minimize the unwanted product. To achieve such condition, the
most important factor to be considered, is the analysis of the product
distribution in terms of the yield and the selectivity of the wanted
product. 12
 Here we define the same by studying the so called selectively and yield
ratio. The instantaneous selectivity or differential selectivity ratio,
can be defined as the ratio of the rate formation of the desired
product to the formation of undesired product.

k1
Consider the reactions A  B desired  product (r )
D
k2
A 
 C unwanted  product (r )
U
The differential selectivity ratio for the reaction becomes
r
SC  U
r
D
Recalling the differential equation
 k1C A1
dC B
dt
 k 2C A 2
dC C
dt
13
Then the selectivity ratio becomes
r
CA 2 - 1
U dCC k2 (6.4.2)
SC   
r dCB k1
D
Our target therefore, would be to keep the selectivity ratio in equation
(6.4.2) as small as possible.
The ratio depends on the kinetic behavior of the reaction i.e.
 by describing the condition of the reaction
 the concentration of the reactant

6.2.2.1 Product distribution for parallel reaction

Consider the decomposition of a single reactant

k1
A  P      desired  product
k2
A 
 C      unwanted  product

14
 The different rate equation is given accordingly as

 k1C A1
dC P
r 
P dt
r  C  k 2C A1
dC
C dt
The differential selectivity ratio then becomes
U
r dCC k2  -  (6.7)
SC    CA 2 1
r dCP k1
D
 As we said our target is to keep the ratio, as small as possible so
that we can maximize the desired product.
 This can be regulated by using concentration of reactant A high or
low.
 This depends
 on the type of reactor used
 the kinetic behavior of the reaction
15
Generally based on the characteristics of the reactor
 for low concentration, we use CSTR
 for higher concentration we use PFR and BR.
To keep concentration of the reactant as low as possible through out
the reaction,
we have to maintain the conversion as high as possible
 or we have to increase the inert in the feed
 or decrease the pressure when we are working in the gas phase
system.
If we want to keep the reactant concentration high through out the
reaction, we have to use adverse effect to the above,
namely maintain low conversion,
remove inert from the feed
or increase the pressure in gas system.

16
kinetic behavior of the reaction to maximize the desired product using
equation (6.7) becomes .
1. 1   2
i.e. the desired reaction is of higher order than the unwanted reaction

 2 CA 2 - 1  2
dCC k k 1
SC 
dCP k1 k1 C  2 1
A

To keep the selectivity ratio, SC low

 we need higher reactant concentration and this is done in plug flow
or batch reactor which means a smaller size of a reactor

2. 1   2
i.e. the desired reaction is of lower order than the unwanted reaction

17
 CA 2 - 1
dCC k2
SC  
dCP k1
 A lower reactant concentration favors to form minimum selectivity
ratio. This is achieved in a CSTR which gives an effect of a large
reactor size. There is a conflicting demand of, in one side to keep that
ratio small and on the other a big size of reactor
In such case, analysis of economic justification is essential.
3. 1   2 i.e. the two reactions are of the same order
dCC k
SC   2
dCP k1
The product distribution here is fixed and unaffected by the type of
reactor used. This equation can vary only by changing the
temperature of the reaction and or by using a catalyst.

All parallel reactions follow the same reasoning

18
 6.2.2.2 Product distribution and reactor size for parallel reaction
 Depending on the kinetics product distribution may be different on the
type of reactor. The product distribution is expressed in terms of yield and
conversion. For the reaction
k1
A  P      desired  product
k2
A 
 C      unwanted  product
The Instantaneous Yield
r
YP  P  dCP (6.11)
- rA  dC A
Here YP = 1 when all A is converted into P.
 This demonstrates that instantaneous yield is specific function of CA
and the reactant concentration.
 concentration A varies throughout the reactor
19
 and this leads us to that instantaneous yield will also change with
position in the reactor.
In another word, it demonstrates also the size of a reactor
Overall yield ratio, which is the fraction of all the reacted A, is converted
into P. This terminology also known as local selectivity
(6.12)
 All P formed  C P,f C P,f
SP       Yin, reactor
 All A reacted  C A,0 - CA,f ( - CA )

Equation (6.12) represents the actual product distribution at the reactor

outlet and this one much concerns us and the proper averaging of Y
depends on the type of flow within the reactor.

The relation of equation (6.11) and (6.12) is often not a simple one and
depends among others such as the reactor type.

20
 Let us consider the reactor type.

6.2.2.2.1 Plug flow and batch reactor

the characteristics of these types of reactor
the reactant concentration changes progressively through the reactor

rP dCP
Taking equation (6.11) YP  
- rA  dC A
or dCP  - Y dCA
C A, f
Both sides integration gives (6.13)
C P,f  -  YdC A
C A,0
On the other side equation (6.12) can be rearranged into
CP,f  S P ( - CA )  S P C A,0 - CA,f  (6.12.1)

Equation (6.13) and (6.12.1) combines to obtain

C A, f
S P C A,0 - C A,f   -  YdC A (6.14)
C A,0 21
which gives, the overall yield in plug flow reactor
C A, f C A, f
1 1 (6.15)
[ S P ]P  -  Y dCA   Y dCA
C A,0 - CA,f C CA C
A,0 A,0
6.2.2.2.2 Mixed flow reactor, CSTR
Here the composition CA,f is every where and is constant throughout
the reactor, and then equation (6.14) for the overall yield in mixed flow
reactor becomes
SP m  Y evaluated at CA,f (6.16)

The overall yield from mixed and plug flow reactor converting A from CA,0
to CA,f are related by
dSP 
SP m    at CA, F (6.17)
 A
dC
C A, f
1
S P P   SP m dCA (6.18)
CA C 22
A,0
 Equation (6.17) and (6.18) allows us to predict the overall yield from one type
of reactor given by the overall yield from the other
6.2.2.2.3 Mixed flow reactor in series
One of the characteristics of mixed reactor in series is the reactant
concentration drops immediately to a low level at the entrance of each reactor
The overall yield is obtained for N series reactors by summing the equation
(6.12), which gives
Y1C A,0 - CA,1   Y2 C A,1 - CA,2   - - - -  YN C A, N 1 - CA, N   S P,mixed C A,0 - CA,1 
from which (6.19)
Y1C A,0 - CA,1   Y2 C A,1 - CA,2   - - - -  YN C A, N 1 - CA, N 
S P,mixed 
C A,0 - CA,1 
For any reactor type the exit concentration of the desired product is obtained
by using directly equation (6.12), thus
CP,f  S P (C A,0 - CA,f ) (6.12.1)

Since an integral and a summation are not equal in general, the instantaneous
yields obtained by batch or plug flow reactor are not the same to those obtained
from mixed and cascade reactors 23
 6.3 Operating temperature
In multiple reactions both the size of a reactor and product distribution might
be affected by operating temperature. Next we can observe the influence of
temperature for both the size of a reactor as well as the product distribution

for two competing multiple reactions of having the rate constants k1 and k2 , the
relative rate of such reaction is given
-E1 / RT
k 1 A1 e A1 (E 2 - E1 ) / RT
 - E 2 / RT
 e
k2 A2 e A2
One can easily see that the ratio changes with temperature depending on whether
E1 is greater or smaller than E2 , that is

k1
increases if E1  E 2
k2
when T rises
k1
decreases if E1  E 2
k2
24
This can be concluded that a general statement of a high temperature favors
the reaction of higher activation energies; a low temperature favors the
reaction of lower activation energies.
Having this in mind one can find the operating temperature for various types of
multiple reactions in favor of product distribution and reactor size

6.3.1 Product distribution and Temperature

Consider two kinds of multiple reactions

•Parallel reactions k1
A  P desired product
k2
A 
 C undesired product
For the same order of reaction, we have

dCP A1 (E 2 - E1) / RT
 e
dCC A2
25
Applying the above statement, the desired product becomes
if E1 > E2 using high T
if E1 < E2 using low T
 Consecutive reactions 1 2
A P  C
where P is desired and C is unwanted product having the same order
of reaction. The production of P can be written
k2
CP  k1  k2 - k1
  
C A,0  k2 
The production of P is increased if k1 / k2 is increased,

if E1 > E2 we use high T

if E1 < E2 we use low T

26
6.3.2 Reactor size and temperature
 Our target in multiple reactions is to have higher selectivity to the
desired product with a minimum size of a reactor.
 The minimum size (volume) will be achieved if the reaction rate is
maximum at all position in the reactor.
 The optimum temperature will be chosen to establish a maximum
rate at any conversion level.
It is, therefore, essential to investigate the relation of
kinetics (reaction order),
activation energy,
reactor size and
 temperature.
For the flow process the reactor size will be represented by space time
V
F 
FV,0
27
The influence of the value of activation energy can be easily seen by
analyzing its value of desired and unwanted reaction as we did
previously. Again to summarize this;
 In case if the activation energy is such that a desired product is
obtained at a higher temperature (E1 > E2) then we should use the
highest allowable temperature.
 The reason is that at higher temperature we obtain high rate of
reaction and consequently smaller size of a reactor

 When E1< E2, on the other hand, the desired product is obtained by
lowering the temperature.

 This causes the rate of reaction to be low and the size of the
reactor large.
These are a conflicting factor.
28
In such a case an alternative solution such as an intermediate temperature
should be found by using analytical method or by search procedures.
Alternative solution

i) parallel reactions E1
A 
 P desired product
E2
A  C undesired product
where (E1< E2), P desired product and C unwanted product
a) If F is of no concern, use the lowest allowable temperature and
with larger space time to favor the production P. This is best carried
out in mixed reactor, CSTR. The optimum temperature for the
maximum desirable product is obtained when
E2
T  R
ln  A2 F  2 - 1
E
  E1  29
 b) when a temperature progression is permitted to carry out in PFR,
for a given, F still it is better to use a rising temperature progression
than the best isothermal optimum. these conflicting factors can be
reduced by using an economic balance to use the appropriate
temperature and size of a reactor

ii) For the consecutive reaction

E1 E2
A  P  C
where (E1< E2), P is desired product and C is unwanted product
a) when  F is no concern use the lowest allowable temperature with
the proper reactor size,  F to maximize the desired product P
b) If a temperature progression is permitted to use PFR, then a falling
temperature progression is better than to use isothermal operation

Generally, for the series reaction, high intermediate product is obtained by using
PFR, and the end product is obtained by increasing temperature, which leads to
increase both k1 and k2. 30
31