CH 6602 – Chemical Reaction Engineering I

Dr. P. Senthil Kumar

Associate Professor
Department of Chemical Engineering
 Unit V

The residence time distribution as a factor performance

Residence time functions and relationship between
them in reactor

Basic models for non-ideal flow

Conversion in non-ideal reactors
 Unit I – Non-Ideal Reactors
Ideal Reactors
1. PFR (no-axial mixing)
2. MFR (complete mixing)

Non-ideal reactors (real reactors)

The reactors always deviate some degree from these
ideal behaviors called real reactors.
 The deviation from ideal flow patterns may be
caused by

1. Stagnant regions in the vessel (dead zone)

2. Bypassing or short-circuiting of the fluid
3. Channeling
4. Recycling of fluid
5. Due to vortices and turbulences at inlet and outlet
 Factors that make up the contacting or flow pattern
are:

1. The RTD – residence time distribution of fluid flowing

through the vessel.
2. The state of aggregation of fluid and
3. The earliness and lateness of fluid mixing in the vessel
 Residence Time Distribution Function

• The time it takes a molecule to pass through a reactor is

called the residence time of the molecule in the reactor.
• It is clear that elements of fluid taking different routes
through the reactor may take different lengths of time to
pass through it.
• So there is a distribution of residence time of the fluid
material within the reactor.
• The distribution of these times for the stream of fluid
leaving the vessel is called the exit age distribution, E, or
the residence time distribution RTD of fluid.
• In an ideal PFR – all the molecules within the reactor
having the same length of time.
• In an ideal batch reactor - all the molecules within the
reactor having the same length of time
• The ideal PFR and BR are the only reactors in which all
the fluid elements in the reactors have the same
residence time.
• In all other reactors, the various molecules spend
different times inside the reactor, Ex: MFR
• The feed introduced into the MFR at any given time
completely mixed with the material already in the reactor.
• Some of the molecules entering the MFR leave it almost
immediately, because the material is continuously
removed from the reactor, other molecules remain in the
reactor almost forever because all the material is never
withdrawn at one time from the reactor. Many of the
molecules leave the reactor after spending a period of
time which is somewhere close to the mean residence
time.
• The distribution of residence time can significantly
affects the performance of reactor.
• The RTD of a reactor is a characteristic of the mixing
that occurs in the reactor.
 RTD Measurement

The RTD is determined experimentally by using an inert

chemical called a tracer (dyes and radioactive materials).
Characteristics of chemical tracer:
1. It should be non-reactive
2. It should be completely soluble in the system fluid
3. It should be easily detectable
4. Physical properties of the tracer should be similar to
those of the system and
5. It should not adsorb on the walls or others surfaces in
the vessel.
 Experimental Methods for Finding E

1. Pulse input
2. Step input
3. Periodic input
4. Random input
1.The Pulse Input Experiment
• The known quantity of tracer (M kg or moles) is suddenly
injected in one shot into the fluid entering the vessel in a
very short time period
• The tracer concentration in the exit stream is measured
as a function of time.
• In RTD analysis, the effluent concentration-time curve is
referred to as the C curve (Cpulse curve).
• The concentration of effluent stream increases with time,
reaches a maximum value and then falls, eventually
approaching zero.
The amount of tracer material, ∆M, leaving the vessel
between t and t+∆t is obtained by the material balance of
the tracer
∆M = C υ ∆t (1)
υ is the volumetric flow rate of the effluent from the vessel
∆M is the amount of material that has spent an amount of
time between t and ∆t in the vessel

Eqn (1) / M

∆M C υ C
= ∆t = ∆t ( 2)
M M M/υ
Where M is the total amount of tracer injected into the vessel
∆M / M is the fraction of material that has residence time
between t and t + ∆t in the vessel
Let us define E is
Cυ C
E= = (3)
M M/υ
∆M
= E ∆t ( 4)
M
whenever M is not known directly , it can be obtained by measuring
the tracer concentration in the effluent stream and summing all the
amounts of materials, ∆M , between time t = 0 and t = ∞.
Eqn (1) may be rewritten as
M ∞
∫ dM = ∫ C υ dt (5)
0 0
M ∞
∫ dM = υ ∫ C dt ( 6)
0 0
 ∞
M / υ = ∫ C dt (7)
0
Combining equations (3) and (7 )
C
E=∞ (8)
∫ C dt
0
∞ ∞
∫ C dt = ∑ Ci ∆t i =Area under the C curve = M / υ
0 0
E curve from C curve
Transforming an E curve into Eθ curve
 2. The Step Experiment
• Consider υ m 3 / s of fluid flowing through a vessel of volume V .
• Now at time t = 0 switch from ordinary fluid to fluid containing tracer
with tracer concentration Cmax and measure the outlet concentration
of tracer, Cstep , in the effluent stream as a function of time
 Cstep
F= Q Cmax = Co
Co
 Relationship between the F and E curves
• Consider that there is a steady flow of main fluid through
the vessel and at time t=0, we switch to tracer fluid and
measure the rising concentration of tracer fluid in the
effluent stream, the F curve.
• At any time t>0, the tracer fluid and only tracer fluid in
the effluent stream is younger than t. Therefore,
 fraction of tracer fluid   fraction of the effluent stream that has 
 = 
 in the effluent stream   been in the vessel for less than time t 

∞ ∞
t
 t 
F = ∫ E dt (1) Q ∫ E dt + ∫ E dt = 1 = ∫ E dt 
0  0 t 0 
∞
 fraction of the effluent stream that has 
1 − F = ∫ E dt =   ( 2)
t  been in the vessel for longer than time t 
∞ t
∫ E dt =1 − ∫ E dt (3)
t 0
∞
∫ E dt =1 − F ( 4)
t
F is the cummulative RTD function
On differentiating eqn (1) with respect to t , we get
dF
=E
dt

This is the relationship b / w E and F

Relation among the F,C,I and E curves in closed vessel

• The closed vessel being defined as one in which

material passes in and out by bulk flow only.
• Consider steady-state flow of fluid through a vessel and
a step function of tracer introduced into the fluid entering
the vessel.
• Suppose that the tracer is simply a second fluid
introduced into the vessel at time t=0 in place of the
original flowing fluid.
• Then at any time t or θ>0 a material balance for the
vessel gives
 Rate of   Rate of   Rate of tracer accumulation 
 = + 
 tracer i / p   tracer o / p   within the vessel 
d θ 
υ = υ F +  V ∫ I dθ
dt  0 
where υ = flow rate of second fluid into the vessel
F = fraction of second fluid in exit stream
d V θ 
1 = F +  ∫ I dθ
dt  υ 0 
d θ 
1 = F + t ∫ I d θ 
dt  0 
t dt
1= F + I Qθ = ; = dθ
t t
Also noting that at time θ
 fraction of second fluid   fraction of the exit stream that has 
 = 
 in the exit stream   been in the vessel for less than time θ 
θ
F = ∫ E dθ
0
similarly for a pulse tracer input we can show that
C=E
These expressions suffice to interrelate the F, C, I and E functions
at anytime θ
F + I =1
C=E
θ θ
F = 1 − I = ∫ E d θ = ∫ C dθ
0 0
dF dI
C=E= =−
dθ dθ
 The RTD in a PFR
• The distribution function in such reactor is a spike
of infinite height and zero width with area =1.
this spikes occurs at t=τ = V/v
Mathematically , this spike is represented as
Dirac delta function
E = δ(t − τ)
∞
Area under the curve = ∫ δ(t − τ) = 1
0
∞
Any integration with delta function = ∫ δ(t − τ) f (t ) dt = f ( τ)
0
 The RTD in a MFR – Pulse Input
Tracer i / p : Pulse i / p
A material balance of a tracer at any time t
after the tracer is injected is
i / p − o / p = accumulation
dC
0−υC= V
dt
C
dC 1t
∫ C = − τ ∫ dt
0 0
t
ln C = −
τ
t
−
C=e τ
∞ ∞ −t
∫ C(t ) dt = ∫ e τ dt
0 0
∞
∫ C(t ) dt = τ
0
C(t )
E(t ) = ∞
∫ C(t ) dt
0
t
−
e τ
E(t ) =
τ
 The RTD in a MFR – Step Input
Tracer i / p : Step i / p
A material balance of a tracer at any time t after the tracer is injected is
i / p − o / p = accumulation
dC
υC o − υ C = V
dt
dC
υ (C o − C) = V
dt
C
dC 1t
∫ (C − C) = τ ∫ dt
0 o 0
t
ln (Co − C) − ln Co = −
τ
(C o − C ) t
ln =−
Co τ
t
C −
=1− e τ = F
Co
 Problems
1. The concentration readings in the table represent
continuous response to pulse input into a closed
vessel which is to be used as a chemical reactor.
Calculate the mean residential time of fluid in the
vessel and tabulate and plot the Exit age
distribution, E.

Time, 0 5 10 15 20 25 30 35
min
Cpulse, 0 3 5 5 4 2 1 0
g/l
2. A large tank (860 liters) is used as a gas-liquid contactor. Gas
bubbles up through the vessel and out the top, liquid flows in at one
part and out the other at 5 liters/s. To get an idea of the flow pattern
of liquid in this tank a pulse of tracer (M = 150 gm) is injected at the
liquid inlet and measured at the outlet, as shown in Fig
(a) Is this a properly done experiment?
(b) If so, find the liquid fraction in the vessel.
(c) Determine the E curve for the liquid.
Mean Residence Time
1. The first moment of the RTD function, E, is the mean
residence time and is given by
∞
∫ t E dt ∞
t= 0
∞
= ∫ t E dt
0
∫ E dt
0

2. The second moment of RTD function, E, is taken about

the mean and is called a variance,
∞
σ = ∫ (t − t ) 2 E dt
2

The magnitude of variance indicates the spread of the

residence time distribution. The greater value of
variance indicates greater RTD spread.
 Problems
3. A continuous response to a delta function into a
closed vessel is given below. Tabulate and plot the
E(t) and E(θ) values determine the mean and the
variance.
Time, 0 5 10 15 20 25 30 35
min
Cpulse, 0 3 5 5 4 2 1 0
g/l
4. Sample of tracer was injected as a pulse into a
reactor and the effluent concentration is measured
as a function of time. The following date are
obtained:

t, min 0 1 2 3 4 5 6 7 8 9 10 12 14
C, g/m3 0 1 5 8 10 8 6 4 3 2.2 1.5 0.6 0

Construct the C and E curves and determine the

fraction of material leaving the vessel that has
spent between 1 and 5 min in the vessel and the
fraction of material leaving that has spent between
8.25 min and 10.5 min in the vessel.
 Conversion in non-ideal flow reactors

To evaluate the reactor behavior in general we have to

know four factors
1. Kinetics of the reaction
2. The RTD of fluid in the reactor
3. The earliness or lateness of fluid mixing in the reactor
4. Whether the fluid is micro or macro fluid
 States of aggregation of the flowing stream

Two extreme states of aggregation of the flowing

material based upon its nature are:

1. Microfluid – A fluid in which molecules are free to move

everywhere, collide, and intermix is called as microfluid.
It exhibits no segregation. Example: gases and thin
liquids.
2. Macrofluid – A fluid in which globules or aggregates,
each containing large number of molecules, of given
age do not mix with other globules is called as a
macrofluid. It exhibits complete segregation. Example:
very viscous liquids, noncoalescing droplets.
 Early and late mixing of fluids

The fluid elements of a single flowing stream can mix

with each other early or late in their flow through a
vessel.
• Consider that a second order reaction is carried out in
these two setups.
• In (a) the reactants starts at high concentration and
reacts away rapidly because n>1.
• In (b) the fluid drops immediately to a low concentration.
Since the rate of reaction drops more rapidly than does
the concentration you will end up with a lower
conversion.
• For microfluids
Late mixing favors reactions with n>1
Early mixing favors reactions with n<1
For n=1, the conversion is unaffected by mixing as for
first order reactions conversion is independent of
concentration dXA/dt = k (1-XA)
• For macrofluids, imagine little aggregates of fluid
residing for different lengths of time in the reactor.
• Each aggregate of macrofluid acts as its own little/tiny
batch reactor and reacts accordingly and therefore the
fluid elements have different compositions.
• The reaction time in any one of these tiny batch reactors
is equal to the time that aggregate spends in the reaction
environment.
• So in order to determine the mean composition in the
exit stream from the reactor we have to take into account
factors such as the kinetics and the RTD function E.
 First order reactions
Consider the following first order reactions :
A → Pr oducts
For a batch reactor
dC A
− = k CA
dt
C A = C Ao (1 − X A )
C AodX A = k C Ao (1 − X A ) dt
∞ ∞
dX A
= k dt X A = ∫ E dt − ∫ e −kt .E dt
(1 − X A ) 0 0
∞
− ln (1 − X A ) = kt X A = 1 − ∫ e −kt .E dt
X A = 1 − e − kt 0

∞ ∞
X A = ∫ X A .E dt = ∫ (1 − e − kt ).E dt
0 0
 Plug Flow Reactor
For a PFR , the RTD function is given by
E = δ ( t − τ)
∞ ∞
X A = ∫ X A .E dt = 1 − ∫ e −kt .E dt
0 0
∞
X A = 1 − ∫ e −kt .δ (t − τ) dt
0
∞
X A = 1 − ∫ δ (t − τ) e −kt dt
0
We have
∞
∫ δ (t − τ) f (t ) dt = f ( τ)
0

f (t ) = e −kt ∴ f ( τ) = e −kτ
X A = 1 − e −kτ for macrofluid
For a PFR we have
τ XA
dX A
= ∫
C Ao 0 − rA
XA
dX A
τ= ∫
0 k (1 − X A )
k τ = − ln (1 − X A )
X A = 1 − e−k τ for microfluid
Equations for macrofluid in a PFR is identical to the eqn obtained
for a microfluid .
∴ The deg ree of segregation does not have any effect on conversion for
first order reactions.
This occurs because the rate of change of conversion of the reacting
molecules for first order reaction does not depend on the concentration
of the reacting molecules.
 CSTR
The RTD function is given by
t
1 −
E= e τ
τ
∞
X A = 1 − ∫ e − kt .E dt
0
∞ − (1 / τ + k ) t
e
XA = 1 − ∫ dt
0 τ
 1  1 − ( k +1 / τ )t ∞
XA = 1 +   e [ ]
 k + 1/ τ  τ
0

τk
XA = for macrofluid
τ k +1
For CSTR the design equation is
τ X
= A
C Ao − rA
C Ao X A
τ=
k C Ao (1 − X A )
XA
τ=
k (1 − X A )
Re arranging we get ,
τk
XA = for microfluid
τ k +1
τ for macrofluid = τ for microfluid

So for the first order reactions inf ormation regarding only RTD is
sufficient
 Problems
5. The concentration readings given below represent a
continuous response to a pulse input into a closed vessel.
Time, 0 5 10 15 20 25 30 35
min
Cpulse, g/l 0 3 5 5 4 2 1 0

This vessel is to be used as a reactor for decomposition of

a liquid A with the stoichiometry A→ Products
and rate –rA=k.CA, k = 0.307 min-1.
(a). Estimate the fraction of reaction converted and
unconverted in the real reactor.
(b). Compare this with fraction converted and unconverted
in the PFR of the same size.
6. The dispersed non-coalescing droplets (CAo=2
mol/L) react (A→R, -rA=kCA2, k=0.5 L/mol-min) as
they pass through a contactor. Find the average
concentration of A remaining in the droplets leaving
the contactor if their RTD is given by the curve
7. A slug of dye is placed in the feed stream to a
stirred vessel operating at steady state. The dye
concentration in the effluent stream was monitored
as a function of time.
Time, 0 120 240 360 480 600 720 840 960 1080
sec
Tracer 0 6.5 12.5 12.5 10 5 2.5 1 0 0
Conc,
g/m3

(a) Determine the average residential time of the

flowing fluid and the F(t) curve for this system.
(b). Determine the variance
8. The concentration readings given below represent a
continuous response to a pulse input into a closed
vessel

Time, 0 1 2 3 4 5 6 7 8 9 10 12 14
sec
Cpulse, 0 1 5 8 10 8 6 4 3 2.2 1.5 0.6 0
g/m3

this vessel is to be used as a reactor for decomposition of

liquid A.
The reaction kinetics and stoichiometry are given by
-rA=k.CA k = 0.1 min-1
Calculate the conversion of A in the real reactor
 Models for Non ideal flow
• Models are useful to represent flow in the real reactor,
scale up and diagnosing the poor flow.
• When a single parameter is used to account for the non
ideal flow in the reactor is called one parameter model.
• This parameter is evaluated by analyzing the RTD
determined from the tracer used.
Types of models for non ideal (real) reactors
1. Dispersion model
2. Tanks-in-series model
3. Segregation model and
4. Compartment model
These models are used to predict the conversion from RTD
data
• The dispersion model
This model is used to account for deviation from ideal
plug flow and involves a modification of the ideal reactor
by imposing axial dispersion on plug flow.
• Tanks-in-series model
This model is used to model non ideal tubular reactor as
a series of equal sized stirred tanks.
• Compartment models
we consider real reactor to consists of different regions,
namely, plug flow region, mixed flow region and
dead/stagnant region. In this case, we model real
reactors using combinations of ideal reactors, dead
volume and bypassing.
Applications of dispersion and tanks-in-series model
• Turbulent flow in pipes
• Laminar flow in very long tubes
• Flow in packed beds
• Shaft kilns
• Long channels
• Screw conveyors etc.

Disadvantages
These models are not applicable to
• Laminar flow in short tubes or
• Laminar flow of viscous materials
Axial Dispersion
• Suppose an ideal pulse of tracer is introduced into the
fluid entering vessel.
• The pulse spreads as it passes through the vessel and
to characterize the spreading according to this model (as
shown in the figure).
• We assume a diffusion like process superimposed on
plug flow.
• We call this dispersion or longitudinal dispersion to
distinguish it from molecular dispersion.
• The dispersion coefficient D (m2/s) represents the
spreading process
Significance of D
• Large D means – rapid spreading of the tracer curve
• Small D means – slow spreading
• D=0 – No spreading hence plug flow
(D/uL) is the dimensionless group characterizing the
spread in the whole vessel.
∞
tCdt ∞ tC∆t
t=∫ =∑
0 Cdt 0 C∆t

∞
(t − t ) 2 Cdt ∞ t 2 Cdt 2 ∞
σ =∫
2
=∫ − t = ∫ (t − t ) 2 Edt
0 Cdt 0 Cdt 0

• These measures are directly linked by theory of D or

D/uL
 Dispersion Model
Consider a plug flow of fluid, on top of which is
superimposed some degree of back mixing, the
magnitude of which is independent of position within the
vessel. This condition implies that there exist no
stagnant pockets and no gross bypassing or short-
circuiting of fluid in the vessel. This is called the
dispersed plug flow model or dispersion model.
• With varying intensities of turbulence or intermixing the
predictions of this model should range from plug flow at
one extreme and mixed flow at the other.
• The reactor volume for this model will lie between those
calculated for plug flow and mixed flow.
• Intermixing of fluid flowing in the x-direction to be
described by ∂C ∂ 2C
=D
∂t ∂x 2
• Where D is the longitudinal or axial dispersion coefficient
uniquely characterize the degree of back mixing during
flow.
• Streamline flow of fluids through pipes, axial mixing is
mainly due to fluid velocity gradients, whereas radial
mixing is due to molecular diffusion alone.
• The basic differential equation representing this
dispersion model becomes
∂C  D  ∂ 2C ∂C
=  2 −
∂θ  uL  ∂z ∂z
• The dimensionless group (D/uL) called the vessel
dispersion number, is the parameter that measures the
extent of axial dispersion

Significance of D/uL
D
=0 − negligible dispersion, hence plug flow
uL
D
=∞ − L arg e dispersion, hence mixed flow
uL

Application
• Real packed beds and
• Long tubes (streamline flow)
Fitting the Dispersion Model for small extents of
Dispersion, D/uL=0.01 (Pulse Tracer)
∂C  D  ∂ 2C ∂C
=  2 −
∂θ  uL  ∂z ∂z
The solution of the above equation is
1  (1 − θ)2 
C= exp− 
2 π( D / uL )  4( D / uL ) 
 Large Deviation from Plug Flow D/uL>0.01 (Pulse Tracer)
1. Open boundary conditions
Either the flow is undisturbed as it passes the entrance and exit
boundaries
2. Closed boundary conditions
Will have plug flow outside the vessel up to the boundaries
Closed Vessel
Open Vessel
Small Deviation from Plug Flow (Step input)
Large Deviation from Plug Flow (Step input)
Correlations for Axial Dispersion

Experiments show that the dispersion model well represents flow in

packed beds and in pipes
 Problems
9. The closed vessel is well represented by the
dispersion model, calculate the vessel dispersion
number, D/uL. The C versus t tracer response of
this vessel is

Time, 0 5 10 15 20 25 30 35
min
Cpulse, g/l 0 3 5 5 4 2 1 0
 Chemical Reaction and Dispersion
• Steady state flow chemical reactor (closed vessel)
• Reactor length = L
• Constant velocity of fluid = u
• Axial mixing with a dispersion coefficient, D
Consider the reaction type A → Pr oducts
with the n th order rate equation − rA = k . CnA

The material balance for any reaction component

Input = Output + Disappearance by reaction + Accumulation

For component A

(out − in )bulk flow + (out − in )axial dispersion +

Disappearance by reaction + Accumulation =0 (1)
Entering by bulk flow = CA , l u S (mol / s )
Leaving by bulk flow = C A , l + ∆l u S
dN A  dC A 
Entering by axial dispersion = = − D S
dt  dl  l
dN A  dC A 
Leaving by axial dispersion = = − D S
dt  dl  l + ∆l
Disappearance by reaction = − rA V = −rA (S ∆l )

Substitute all the individual terms in Eqn(1)

  dC A   dC A  
−
u(C A , l + ∆l − C A , l )    
− D 
dl l + ∆l  dl l
 + ( −rA ) = 0 ( 2)
∆l  ∆l 
 
The basic limiting process of calculus
Q 2 − Q1 ∆Q dQ
lim l 2 → l1 = lim ∆l → 0 =
l 2 − l1 ∆l dl
So taking limits as ∆l → 0 in eqn ( 2)
 dC A   d 2C A 
 − D  2  + (k C A ) = 0
n
u (3)
 dl   dl 
In dimensionless form where z = l / L and τ = t = L / u = V / υ
D  d 2C A   dC A 
 2 
 −   − kτ C A =0
n
( 4)
uL  dz   dz 
or in terms of fractional conversion
D  d 2 X A   dX A  n −1
 2 
 −   − kτ C Ao (1 − X A ) n
=0 ( 5)
uL  dz   dz 
Where
kτ CnAo−1 = reaction rate group
D
= dispersion group
uL
n = reaction order
First order reaction
For vessels with any kind of entrance and exit conditions
the solution is
 1 uL 
4 a exp 
CA
= 1 − XA = 2 D 
C Ao  a uL   a uL 
(1 + a) exp
2
 − (1 − a) exp −
2

2 D   2 D
a = 1 + 4kτ( D / uL )
• For small deviations from plug flow D/uL becomes small,
the E curve approaches Gaussian, the above equation
becomes

• The size ratio needed for identical conversion is given by

• The exit concentration ratio for identical reactor size is

 Tanks-in-Series Model
• One parameter model used to represent non ideal flow in
tubular reactors.
• The parameter of this model is the number of tanks, N, in
series
• In modelling non ideal tubular reactor is to be considered
as a series of equal-sized stirred tanks
• First develop the RTD equation for three tanks in series
and then generalize it to N tanks in series to arrive at
equation that gives the number of tanks in series that
best fits the data
• Steady flow of fluid in and out (v m3/s) of the first three
ideal stirred tanks
• Volume (V1 m3)
• At time t=0, inject a pulse of tracer into the first tank.
• Tracer balance over the first tank at any time t

rate of disappearance of tracer = input rate − output rate

dC
V1 1 = 0 − v C1 (1)
dt
Where C1 is the tracer concentration in the exit stream
from the first tan k
t1 = V1 / v
dC1 v 1
= − dt = − dt
C1 V1 t1
Integrating with limits
t = 0 C1 = Co
t=t C1 = C1
C1
= e− t / t
Co
C1 = Co .e − t / t ( 2)
* Volumetric flow rate is constant ( v = v o )
* Equal − sized stirred tanks in series ( V1 = V2 = Vi )
* Mean residence time for all individual tanks
is identical (t1 = t 2 = t i )
Tracer balance over the second tank
dC 2
V2 = v C1 − v C2
dt
dC 2
Vi = v C1 − v C2
dt
dC2 v v
+ C2 = C1
dt Vi Vi
dC2 C2 C1
+ =
dt ti ti
Putting value of C1 from eqn( 2), we get
dC2 C2 Co .e − t / t i
+ = (3)
dt ti ti
The above eqn(3) is the first order differential eqn.
The solution of the eqn (3) is
Co t − t / t i
C2 = e ( 4)
ti
Similarly for the 3rd tank
Co t 2 − t / t i
C3 = 2
e (5)
2 (t i )
This eqn is the tracer concentration as a function of time
• The fraction of tracer material leaving the system of
three tanks that has been in the system b/w time t and
t+∆t

v C3 ∆t C3
E ∆t = = ∆t
No No / v
∞
We have, N o / v = ∫ C3dt
0

C3
E ∆t = ∞ . ∆t
∫ C3dt
0

Putting value of C3 from equation (5), we get

t2 −t / t i
E= . e ( 6)
2 (t i )2
Generalizing this method to a series of N equal − sized
tanks gives the RTD for the tanks in series, E
t N −1 . e −t / t i
E= (7 )
( N − 1)! (t i ) N
Multiplying by t i on both sides of the eqn (7)
t N −1 . t i −t / t i
ti E = e
( N − 1)! (t i ) N
N −1
t 1
ti E =   e −t / t i (8)
 ti  ( N − 1)!
For N tanks in series, total volume = N.Vi
t
ti = = mean residence time in one tank
N
where t represents total tank volume divided by flowrate v
N −1
tE  t  1
=  e − tN / t
N t/N ( N − 1)!
N −1
t NN
tE=  e − tN / t (9 )
t ( N − 1)!
E θi = t i E
( θi )N −1 − θ i
E θi = t i E = e (10)
( N − 1)!
N ( Nθ)N −1 − N θ
Eθ = ( N t i ) E = e (11)
( N − 1)!

θ i = t / t i = Nt / t
θ = t / t = t / Nt
 t
ti = , σ 2 = N (t i ) 2
N
2
( t )
t = N ti , σ 2 = N (t i ) 2 =
N
t θi = N , σ θ2i = N
t θ = 1, σ θ2 = 1 / N
• If we connect M tanks to already existing chain of N
tanks (same size) then the individual means and
variance are additive.

tN + tM = tM+N
σ 2N + σM
2
= σM
2
+N
If we introduce any one − shot tracer input into
N tanks (in series)
( ∆t )2
∆σ = σ − σ =
2 2
out
2
in
N
Therefore, the number of tanks in series is
2
1 t
N= 2 = 2
σθ σ
Conversion from Tanks-in-Series Model
For first order reaction in one tank
CA 1 1
= =
C Ao 1 + k t i 1 + k (t / N )
This eqn for the N equal sized tanks in series becomes
CA 1 1
= =
C Ao (1 + k t i ) N
(1 + k t / N )N

C Ao − C A C
We have, XA = =1− A
C Ao C Ao
1
XA = 1 −
(1 + k t i )N
1
or XA = 1 −
(1 + k t / N )
N
 V
ti =
N vo
For non − first order reaction, sequential mole balance
on each tank must be carried out
C Ai −1 − C Ai
ti =
− ri
 Problems
10. A first order liquid phase reaction is carried out in a
reactor for which the results of (pulse) tracer test
are given below. k= 0.25 min-1
Time, 0 1 2 3 4 5 6 7 8 9 10 12 14
min
Cpulse, 0 1 5 8 10 8 6 4 3 2.2 1.5 0.6 0
g/m3

Calculate the conversion using

(i). Ideal PFR
(ii). Ideal MFR
(iii). The tanks-in-series model
11. A reactor with dividing baffles is to be used to carry
the reaction A→R, -rA=0.05 CA, mol/l.min
Pulse test results are
Time, min 0 10 20 30 40 50 60 70
Cpulse, 0 1 5 8 10 8 6 4
g/m3

calculate the conversion

(i). Assume Plug Flow
(ii). Assume Mixed Flow
(iii). Tanks-in-series model
12. A pulse input to a vessel gives the results shown in
Figure

(a). Check the material balance with the tracer curve

to see whether the results are consistent or not.
(b). If the results are consistent, determine the mean
residential time and Volume of the reactor.
 Check your understanding
1. The reasons for non-ideality in the real reactors.
2. RTD means____________
3. Give the experimental methods for finding out exit age
distribution function, E
4. Pulse input will give _________ curve
5. Step input will give __________ curve
6. Example for zero parameter modelling _________
7. Examples for one parameter modelling _________
8. What is Dispersion number
9. Significance of Dispersion number
10. If the number of tanks are connected in series then the
reactor will be acting as like ________ Reactor
Answers:
1. Channeling, Recycling of fluid, Stagnant region
2. Residence Time Distribution
3. Pulse, Step, Periodic and Random input
4. C curve
5. F curve
6. Segregated Flow Model (SFM)
7. Dispersion Model and Tanks-in-series model
8. D/uL
9. D/uL=0 , plug flow: D/uL=∞ - mixed flow
10.Plug flow reactors
 Summary
In the first of CRE-2 we discussed
1. The non-ideal (real) reactors and its design – many
problems performed in this area.
2. The factors that make up the contacting or flow pattern
in the non-ideal reactors (RTD, state of aggregation of
fluid and earliness and lateness of fluid mixing in the
vessel) - many problems performed in this area.
3. Basic models used for non-ideal reactors (SFM,
Dispersion model and Tanks-in-series model) - many
problems performed in this area.
4. The conversions obtained from the non-ideal flow
reactors.