1

Models for Nonideal Reactors

2
Some Guidelines
RTD
Model
Kinetic Data
Exit Concentration
Exit Conversion
1.The model must be mathematically tractable.
The equations used to describe a chemical reactor should be able to be solved without an
inordinate expenditure of human or computer time.

2.The model must realistically describe the characteristics of the nonideal reactor:
The phenomena occurring in the nonideal reactor must be reasonably described physically,
chemically, and mathematically.

3.The model must not have more than two adjustable parameters.
This constraint is used because an expression with more than two adjustable parameters
can be fitted to a great variety of experimental data, and the modeling process in this
circumstance is nothing more than an exercise in curve fitting.
A one-parameter model is, of course, superior to a two-parameter model if the one-
parameter model is sufficiently realistic.
To fair, however, in complex systems (e.g., internal diffusion and conduction, mass transfer
limitations) where other parameters may be measured independently, then more than two
parameters are quite acceptable.
3
4
One-Parameter Models
~Tanks-in-series model
modeling tubular reactors as a series of identically sized CSTRs
a parameter is the number of tanks, n
~Dispersion model
a modification of the ideal reactor by imposing axial dispersion on plug flow
a parameter is the dispersion coefficient
Two-Parameter Models
Using a combination of ideal reactors to model the real reactor
We could model the real reactor as
two ideal PBRs in parallel with the
two parameter being the fluid that
channels, v
b
, and the reactor dead
volume, V
D
. The reactor volume is
V=V
D
+V
S
with v
0
=v
b
+v
S
.
Considering a packed bed reactor with channeling
5
Tanks-in-Series (T-I-S) Model
We are first going to develop the RTD equation for three tanks in series (Figure 14-2)
and then generalize to n reactors in series to derive an equation that gives the number
of tanks in series that best fits the RTD data.
1
1
1
vC
dt
dC
V =
Considering a tracer pulse injected
into the first reactor of three equally
sized CSTRs in series
1 1
/ t
0
V / vt
0 1
e C e C C
t
= =
1
0
3 0
1 0 0
V
dt ) t ( C v
V / N C
}

= =
n /
V V V
v v
i
i 2 1
i 2 1
0
t = t
t = t = t
= =
=
2 1
2
i
vC vC
dt
dC
V =
i
/ t
i
0
i
2 2
e
C C
dt
dC
t
t
=
t
+
i
/ t
i
0
2
e
t C
C
t
t
=
i
/ t
2
i
2
0
3
e
2
t C
C
t
t
=
i
i
i
/ t
3
i
2
0
2
i
/ t 2
0
/ t 2
i
2
0
0
3
3
e
2
t
dt
2
e t C
e ) 2 /( t C
dt ) t ( C
) t ( C
) t ( E
t

t
t

t
=
t
t
= =
}
}
i
/ t
n
i
1 n
e
)! 1 n (
t
) t ( E
t

t
=
6
O

O
= t = O
n
1 n
e
)! 1 n (
) n ( n
) t ( E ) ( E
n /
i
t = t
n, the behavior of the system approaches
that of a plug-flow reactor
1 d ) ( E
d ) ( E d ) ( E 2 d ) ( E
d ) ( E ) 1 (
0
2
0 0 0
2
0
2
2
2
2
O O O =
O O + O O O O O O =
O O O =
t
o
= o
}
} } }
}

O
1 d ) ( E
1 d ) ( E
0
0
= O O O
= O O
}
}

n
1
1
n
)! 1 n (
)! 1 n (
n

1 d e
)! 1 n (
n

1 d e
)! 1 n (
) n ( n

1 d ) ( E
2 n
n
0
n 1 n
n
0
n
1 n
2
0
2 2
=
(

=
O O

=
O

O
O =
O O O = o
+

O +

O
}
}
}
2
2
2
1
n
o
t
=
o
=
O
7
For a first order reaction,
n
i
) k 1 (
1
1 X
t +
=
n v
V
0
i
= t
n may be a noninteger.
For reactions other than first order,
an integer number of reactors must be used.
8
Example 14-1
Show that X
T-I-S
=X
MM
for a first-order reaction
B A
k

Solution
For a first-order reaction,
MM seg
X X =
Therefore we only need to show X
seg
=X
T-I-S
.
For a first-order reaction in a batch reactor the conversion is
kt
e 1 X

=
} } }

= = =
0
kt
0
kt
0
dt ) t ( E e 1 dt ) t ( E ) e 1 ( dt ) t ( E ) t ( X X
Segregation model
error
2
t k
kt 1 e
2 2
kt
+ + =

} } }

=
|
|
.
|

\
|
=
0
2
2
0 0
2 2
dt ) t ( E t
2
k
dt ) t ( tE k dt ) t ( E
2
t k
kt X
9
}

t =
0
2
2
dt ) t ( E t
2
k
k X
2
0
2
2 2
0
2
0
2
0 0
2
0
2 2
dt ) t ( E t
2 dt ) t ( E t
dt ) t ( E dt ) t ( tE 2 dt ) t ( E t
dt ) t ( E ) t (
t =
t + t =
t + t =
t = o
}
}
} } }
}

2 2
0
2
dt ) t ( E t t + o =
}

) (
2
k
k X
2 2
2
t + o t =
Tanks in Series
n
k
n
1
1
1 X
|
.
|

\
|
t
+
=
n
2
2
t
= o
|
|
.
|

\
|
t
+ t t =
t

t
t =
|
|
.
|

\
|
+
t +
+
t
=
|
.
|

\
|
t
+ =

n 2
k
k
n 2
k
2
k
k
error
n
k
2
) 1 n ( n
k
n
n 1 1
k
n
1 1 X
2
2
2 2 2 2 2
2
2 2
n
) (
2
k
k X
2 2
2
o + t t =
MM seg S I T
X X X = =

10
Dispersion Model
In addition to transport by bulk flow,
UA
c
C, every component in the mixture is
transported through any cross section of
the reactor at a rate equal to [-D
a
A
c
(dC/dz)]
resulting from molecular and convective
diffusion.
By convective diffusion (i.e, dispersion) we mean either Aris-Taylor dispersion in
laminar flow reactors or turbulent diffusion resulting from turbulent eddies.
Radial concentration profiles for plug flow (a) and a representative axial and
radial profile for dispersive flow (b) are shown in Figure 14-1.
Some molecules will diffuse forward ahead of molar average velocity while others
will lag behind.
11
Figure 14-5 presents how dispersion causes the pulse to broaden as it moves down
the reactor and becomes less concentrated.
c T
T
a T
A UC
z
C
D F
|
.
|

\
|
+
c
c
=
The molar flow rate of tracer (F
T
) by
both convection and dispersion is
D
a
is the effective dispersion coefficient (m
2
/s)
U is the superficial velocity (m/s)
There is a concentration gradient on
both sides of the peak causing molecules
to diffuse away from the peak and thus
broaden the pulse. The pulse broadens
as it moves through the reactor.
A mole balance on the inert tracer T gives
t
C
A
z
F
T
c
T
c
c
=
c
c

t
C
z
) UC (
z
C
D
T T
2
T
2
a
c
c
=
c
c

c
c
C
T
vs. t
12
Flow, Reaction, and Dispersion
Balance Equations
A mole balance is taken on a particular component of the mixture (say, species A)
over a short length Az of a tubular reactor of cross section A
c
, to arrive at
0 r
dz
dF
A
1
A
A
c
= +
A
A
a
c
A
UC
dz
dC
D
A
F
+ =
0
U
r
dz
dC
dz
C d
U
D
A A
2
A
2
a
= +
A second-order ordinary differential equation.
Nonlinear when r
A
is other than zero or fist order.
A A
kC r =
0
U
kC
dz
dC
dz
C d
U
D
A A
2
A
2
a
=

+
l
z
,
C
C
0 A
A
= =
0 Da
d
d
d
d
Pe
1
2
2
r
=

a
D
U
sion ordisper diffusion ansportby Rateoftr
n convectio ansportby Rateoftr
Pe
k
ection Abyconv ansportof Rateoftr
action ofAbyre nsumption Rateofco
Da
l
= =
t = =
l is the characteristic length
Da:Damkohler number
Pe:Peclet number
13
letnumber fluidPec ) tube forempty (
D
Ud
), bed forpacked (
D
Ud
Pe
ber Pecletnum reactor
D
U
Pe
a
t
a
p
f
a
r
|
=
=
L
For open tubes
Pe
r
~ 10
6
, Pe
f
~ 10
4
For packed beds
Pe
r
~ 10
3
, Pe
f
~ 10
1
Boundary Conditions
closed-closed vessel
there is no dispersion or radial variation in
concentration either upstream (closed) or
downstream (closed) of the reaction section
open-open vessel
dispersion occurs both upstream (open) and
downstream (open) of the reaction section
A closed-open vessel boundary condition is one in which there is no dispersion in
the entrance section but there is dispersion in the reaction and exit section.
14
Closed-Closed Vessel Boundary Condition
0
-
0
+
z=0

F
A
) 0 ( C UA
z
C
D A ) 0 ( C UA
A c
0 z
A
a c A c
+
=

+
|
.
|

\
|
c
c
=
+
) 0 ( F ) 0 ( F
A A
+
=
A c
A
a c A
C UA
z
C
D A F +
c
c
=
At z=0
) 0 ( C
z
C
U
D
C
A
0 z
A a
0 A
+
=
+
|
.
|

\
|
c
c
=
+
0 A A
C ) 0 ( C =

At z=L
0
z
C
) L ( C ) L ( C
A
A A
=
c
c
=
+
Danckwerts boundary conditions
open-open vessel
) 0 ( C
z
C
U
D
) 0 ( C
z
C
U
D
A
0 z
A a
A
0 z
A a +
=

=
+
|
.
|

\
|
c
c
= +
|
.
|

\
|
c
c

+
15
Back to the Solution for a Closed-Closed System
0 Da
d
d
d
d
Pe
1
2
2
r
=

0
d
d
then 1 At
) 0 (
d
d
Pe
1
1 then 0 At
0 r
=

=
+
|
.
|

\
|

= =
+
+
r
r
2
r
2
r
0 A
AL
L
Pe / Da 4 1 q
) 2 / q Pe ( ) q 1 ( ) 2 / q Pe exp( ) q 1 (
) 2 / Pe exp( q 4
X 1
C
C
+ =
+
= = =
) 2 / q Pe ( ) q 1 ( ) 2 / q Pe exp( ) q 1 (
) 2 / Pe exp( q 4
1 X
r
2
r
2
r
+
=
outside the limited case of a first-order
reaction, a numerical solution of the
equation is required, and because this
is s split-boundary-value problem, an
iterative technique is required.
Finding D
a
and Peclet Number
1.Laminar flow with radial and axial molecular diffusion theory
2.Correlation from the literature for pipes and packed beds
3.Experimental tracer data
16
Dispersion in a Tubular Reactor with Laminar Flow

2 / t
t 2
2 / t 0
) t ( E
3
2

t >
t
t <
=
(
(

|
.
|

\
|
=
2
R
r
1 U 2 ) r ( u
The axial velocity varies in the radial direction according to the Hagen-Poiseuille equation:
U is the average velocity
In arriving at this distribution E(t), it was assumed that
there are no transfer of molecules in the radial direction
between streamlines.
7 / 8 t 8 / U 7 u 4 / R 3 r
2 / Ut 2 u 0 r
t = = =
t = = =
In addition to the molecules diffusing between
streamlines, they can also move forward or
backward relative to the average fluid velocity
by molecular diffusion (Ficks law).
17
The convective-diffusion equation for solute (e.g., tracer) transport in both the axial
and radial direction can be obtained.
(

c
c
+
|
.
|

\
|
c
c
c
c
=
c
c
+
c
c
2
2
AB
z
C
r
C
r
r r
1
D
z
C
) r ( u
t
C
Ut z * z =
(

c
c
+
|
.
|

\
|
c
c
c
c
=
c
c
+
|
.
|

\
|
c
c
2
2
AB
* z
* z
C
r
C
r
r r
1
D
* z
C
] U ) r ( u [
t
C
}
t
t
=
R
0
2
rdr 2 ) t , z , r ( C
R
1
) t , z ( C
2
2
* z
C
* D
* z
C
U
t
C
c
c
=
c
c
+
c
c
AB
2 2
AB
D 48
R U
D * D + =
D* is the Aris-Taylor dispersion coefficient
For laminar flow in a pipe
* D D
a

18
Correlations for D
a
Dispersion for Laminar and Turbulent Flow in Pipes
Dispersion in Packed Beds
d
t
is the tube diameter
Sc is the Schmidt number
d
p
is the particle diameter
c is the porosity
19
Experimental Determination of D
a
t
C
z
) UC (
z
C
D
T T
2
T
2
a
c
c
=
c
c

c
c
For the different types of boundary conditions
at the reactor entrance and exit, to solve for the
exit concentration as a function of dimensionless
time (O=t/t), and then relate D
a
, o
2
, and t.
Unsteady-State Tracer Balance

L
tU
,
L
z
,
C
C
0 T
T
= O = =
O c
+ c
=
c
c

c
c
2
2
r
Pe
1
Initial condition
0 ) 0 ( , 0 ) 0 , 0 ( C , 0 z , 0 t
T
= = > >
+ +
Mass of tracer injected, M is
}

=
0
T C
dt ) t , 0 ( C UA M
For a pulse input, C
T0

is defined as the mass
of tracer injected, M,
divided by the vessel
volume, V.
20
Solution for a Closed-Closed System

) 0 ( C
z
C
U
D
C
T
0 z
T a
0 T
+
=
+
|
.
|

\
|
c
c
=
+
At =0
1 ) 0 (
Pe
1
0
r
= +
|
|
.
|

\
|
c
c

+
+
0 =
c
c
At =1
0
z
C
A
=
c
c
t =
m
t
}

t
t
=
o
0
2
2 2
m
2
dt ) t ( E ) t (
1
t
) e 1 (
Pe
2
Pe
2
t
r
Pe
2
r r
2
m
2

=
o
RTD data
t
m
o
2
Pe
r
D
a
21
Solution for a Open-Open System

) t , 0 ( C
z
C
U
D
) t , 0 ( C
z
C
U
D
T
0 z
T a
T
0 z
T a +
=

=
+
|
.
|

\
|
c
c
= +
|
.
|

\
|
c
c

+
At =0
At =1
) t , 0 ( C ) t , 0 ( C
T T
+
=
) t , L ( C
z
C
U
D
) t , L ( C
z
C
U
D
T
L z
T a
T
L z
T a +
=

=
+
|
.
|

\
|
c
c
= +
|
.
|

\
|
c
c

+
) t , L ( C ) t , L ( C
T T
+
=
Consider the case when there is no variation in the dispersion coefficient for all z
and an impulse of tracer is injected at z=0 at t=0.
For long tubes (Pe>100) in
which the concentration
gradient at will be zero
(

O
O
O t
= = O
r
2
r
0 T
T
Pe / 4
) 1 (
exp
Pe / 2
1
C
) t , L ( C
) , 1 (
22
t
|
|
.
|

\
|
+ =
r
m
Pe
2
1 t
2
r r
2
2
Pe
8
Pe
2
+ =
t
o
t is based on the volume between z=0 and z=L (i.e., reactor
volume measured with a yardstick).
The mean residence time for an open system is greater than
that for a closed-system. The reason is that the molecules can
diffuse back into the reactor after they exit.
Case 1~space time t is known
2
r r
2
2
Pe
8
Pe
2
+ =
t
o
RTD data
Pe
r
t
|
|
.
|

\
|
+ =
r
m
Pe
2
1 t
t
m
o
2
t
m

check
Case 2~space time t is unknown
t
|
|
.
|

\
|
+ =
r
m
Pe
2
1 t
2
r r
2
2
Pe
8
Pe
2
+ =
t
o
4 Pe 4 Pe
8
4 Pe 4 Pe
Pe 2
t
r
2
r r
2
r
r
2
m
2
+ +
+
+ +
=
o
RTD data
t
m
o
2
Pe
r
t
|
|
.
|

\
|
+ =
r
m
Pe
2
1 t t

V

V
D
23
Sloppy Tracer Inputs
It is not always possible to inject a tracer pulse
cleanly as an input to a system because it takes a
finite time to inject the reactor.
When the injection does not approach a perfect
input (Figure 14-14), the differences in the variances
between the input and output tracer measurements
are used to calculate the Peclet number:
2
out
2
in
2
o o = o A
o
in
2
is the variance of the tracer measured at some point upstream (near the entrance)
o
out
2
is the variance of the tracer measured at some point downstream (after the exit)
r
2
m
2
Pe
2
t
=
o A
For an open-open system
24
Example 14-2
The first-order reaction
AB
is carried out in a 10-cm-diameter tubular reactor 6.36 m in length. The specific reaction rate is
0.25 min
-1
. The results of a tracer test carried out this reactor are shown in Table E14-2.1.
Calculate conversion using (a) the closed vessel dispersion model, (b) PFR, (c) the tanks-in-series
model, and (d) a single CSTR.
25
Solution
68 . 0
) 87 . 4 ( ) 3 . 1 1 ( ) 87 . 4 exp( ) 3 . 1 1 (
) 2 / 5 . 7 exp( ) 30 . 1 ( 4
1 X
2 2
=
+
=
(a)
) e 1 (
Pe
2
Pe
2
t
r
Pe
2
r r
2
m
2

=
o
min 15 . 5 dt ) t ( tE t
0
m
= =
}

2 2 2
m
0
2
0
2
m
2
min 10 . 6 ) 15 . 5 ( 63 . 32 t dt ) t ( E t dt ) t ( E ) t t ( = = = = o
} }

min g 50 dt ) t ( C
0
=
}

) e 1 (
Pe
2
Pe
2
23 . 0
) 15 . 5 (
1 . 6
r
Pe
2
r r
2

= = 5 . 7 Pe
r
=
29 . 1 ) min 25 . 0 min)( 15 . 5 ( k t k Da
1
m
= = = t =

87 . 4
2
) 3 . 1 )( 5 . 7 (
2
q Pe
30 . 1
5 . 7
) 29 . 1 ( 4
1
Pe
Da 4
1 q
r
r
= =
= + = + =
26
(b)
725 . 0 e 1 e 1 e 1 X
29 . 1 Da k
= = = =
t
35 . 4
1 . 6
) 15 . 5 (
n
2
2
2
= =
o
t
=
(c)
677 . 0
) 35 . 4 / 29 . 1 1 (
1
1
k
n
1
1
1 X
35 . 4 n
=
+
=
|
.
|

\
|
t
+
=
(d)
563 . 0
29 . 1 1
29 . 1
k 1
k
X =
+
=
t +
t
=
PFR: X=72.5%
Dispersion: X=68.0%
Tanks in series: X=67.7%
Single CSTR: X=56.3%
27
Tanks-in-Series Model Versus Dispersion Model
For first-order reactions, the two models can be applied with equal ease.
However, the tanks-in-series model is mathematically easier to use to obtain the
effluent concentration and conversion for reaction orders other than one and
for multiple reactions.
These two models are equivalent when the Peclet-Bodenstein number is related
to the number of tanks in series, n, by the equation
) 1 n ( 2 Bo = 1
2
Bo
n + =
Bo=UL/D
a
U is the superficial velocity
L is the reactor length
D
a
is the dispersion coefficient
75 . 4 1
2
5 . 7
1
2
Bo
n = + = + =
35 . 4
1 . 6
) 15 . 5 (
n
2
2
2
= =
o
t
=
For reactions other than first order,
X(n=4)<X<X(n=5)
28
Two-Parameter Models-Modeling Real Reactors
with Combinations of Ideal Reactors
Real CSTR Modeled Using Bypassing and Dead Space
A CSTR is believed to modeled as a combination of an ideal CSTR of volume V
s
, a
dead zone of volume V
d
, and a bypass with a volumetric flow rate v
b
(Figure 14-15)
We have used a tracer experiment to
evaluate the parameters of the model V
s

and v
s
. Because the total volume and
volumetric flow rate are known, once V
s

and v
s
are found, v
b
and V
d
can readily
be calculated.
29
Solving the Model System for C
A
and X
B A
Considering the first-order reaction
From a balance on species A around point 2
) v v ( C v C v C
Out In
s b A s As b 0 A
+ = +
=
0
b
v
v
= |
0
s As b 0 A
s b
s As b 0 A
A
v
v C v C
v v
v C v C
C
+
=
+
+
=
As 0 A A
C ) 1 ( C C | + | =
A mole balance on V
s
gives
0 V kC C v C v
s As As s 0 A s
=
0
b s
v
v
,
V
V
= | = o
Vk v ) 1 (
v ) 1 ( C
C
0
0 0 A
As
o + |
|
=
k ) 1 (
) 1 (
X 1
C
C
2
0 A
A
ot + |
|
+ | = =
The bypass stream and effluent
stream from the reaction volume
are mixed at point 2.
?
?
= |
= o
evaluated with tracer experiments
and RTD data
30
Using a Tracer to Determine the Model Parameters in CSTR-with-
Dead-Space-and-Bypass Model
We shall inject our tracer, T, as a positive step input. The unsteady-state
balance on the nonreacting tracer T in the reactor volume V
s
is
dt
dC
V
dt
dN
C v C v
on accumulati Out In
Ts
s
Ts
Ts s 0 T s
= =
=
0 T T
T
C C 0 Att
0 C 0 Att
= >
= <
A balance around junction point 2 gives
0
s Ts b 0 T
T
v
v C v C
C
+
=
0
0 b
s
v
V
v v
V V
= t
| =
o =
(

|
.
|

\
|
t o
|
=
t 1
exp 1
C
C
0 T
Ts
(

|
.
|

\
|
t o
|
| =
t 1
exp ) 1 ( 1
C
C
0 T
T
Ts 0 T T
C ) 1 ( C C | + | =
t
|
.
|

\
|
o
|
+
|
|
.
|

\
|
|
=
|
|
.
|

\
|

t 1
1
1
ln
C C
C
ln
T 0 T
0 T
t . vs
C C
C
ln
T 0 T
0 T
|
|
.
|

\
|

|
|
.
|

\
|
|
=
ot
|
=
1
1
ln ercept int
1
slope
31
Example 14-4
The elementary reaction
A+BC+D
is to be carried out in the CSTR shown schematically in Figure 14-15. There is both
bypassing and a stagnant region in this reactor. The tracer output for this reactor is shown
in Table E14-4.1. The measured reactor volume is 1.0 m
3
and the flow rate to the reactor is
0.1 m
3
/min. The reaction rate constant is 0.28 m
3
/kmolmin. The feed is equimolar in A and
B with an entering concentration of A equal to 2.0 kmol/m
3
. Calculate the conversion
that can be expected in this reactor (Figure E14-4.1).
C
T0
=2000 mg/dm
3
32
Solution
t
|
.
|

\
|
o
|
+
|
|
.
|

\
|
|
=
|
|
.
|

\
|

t 1
1
1
ln
C C
C
ln
T 0 T
0 T
25 . 1 I
1
1
= =
|
2 . 0
v
v
0
b
= = |
1
min 115 . 0 S
1

= =
ot
|
min 7
115 . 0
2 . 0 1
=

= ot
min 10
min / m 1 . 0
m 1
v
V
3
3
0
= = = t
7 . 0
min 10
min 7
= = o
33
A mole balance on V
s
gives
0 V r C v C v
s As As s 0 A s
= +
2
As Bs As As
kC C kC r = =
equimolar feed
0 V kC C v C v
s
2
As As s 0 A s
=
k 2
kC 4 1 1
C
s
0 A s
As
t
t + +
=
As 0 A A
C ) 1 ( C C | + | =
A mole balance around junction point 2 gives
min 7 . 8
08 . 0
7 . 0
v
V
min / m 08 . 0 min) / m 1 . 0 ( ) 2 . 0 1 ( v ) 1 ( v
m 7 . 0 m 1 7 . 0 V V
s
s
s
3 3
0 s
3 3
s
= = = t
= = | =
= = o =
3
3
3 3
As
m / kmol 724 . 0
min) kmol / m 28 . 0 min)( 7 . 8 ( 2
1 ) m / kmol 2 min)( kmol / m 28 . 0 min)( 7 . 8 ( 4 1 1
C =

+ +
=
3
3 3
As 0 A A
m / kmol 979 . 0
m / kmol 724 . 0 8 . 0 m / kmol 2 2 . 0
C ) 1 ( C C
=
+ =
| + | =
51 . 0
2
979 . 0
1 X = =
) STR anidealC ( 66 . 0 X =
34
Real CSTR Modeled as Two CSTRs with Interchange
In this particular model there is a highly agitated region in the vicinity of the
impeller; outside this region, there is a region with less agitation (Figure 14-17).
There is considerable material transfer
between the two regions. Both inlet and
outlet flow channels connect to the
highly agitated region. We shall model
the highly agitated region as one CSTR,
the quieter region as another CSTR,
with material transfer between the two.
Solving the Model System for C
A
and X
0
2
1
0 1
v
V
V ) 1 ( V
V V
v v
= t
o =
o =
| =
]} k ) 1 ( /[ { k 1
C
C
2
0 A
1 A
t o + | | ot + | +
=
2
2
0 A
1 A
] k ) 1 ( )[ k 1 (
] k ) 1 ( )[ k (
C
C
1 X
| t o + | ot + | +
| t o + | ot + |
= =
35
Using a Tracer to Determine the Model Parameters in a
CSTR with an Exchange Volume
2 T 1 1 T 1
2 T
2
2 T 1 1 T 0 2 T 1
1 T
1
C v C v
dt
dC
V : 2 actor Re
) C v C v ( C v
dt
dC
V : 1 actor Re
=
+ =
A mole balance on a tracer pulse injected at t=0 for each of the tanks is
C
T1
and C
T2
are the tracer concentration in
reactors 1 and 2, respectively, with initial
conditions C
T10
=N
T0
/V
1
and C
T20
=0.
2 T 1 T
2 T
1 T 2 T
1 T
C C
dt
dC
) 1 (
C ) 1 ( C
dt
dC
| | = o t
| + | = to
(

| + o
o o|

(

o o
| + o
=
o
+ | + o + | + o
=
|
|
.
|

\
|
t t
2
2 1
2 1
/ t m
2
/ t m
1
pulse
10 T
1 T
1
) 1 ( 4
1 1
) 1 ( 2
1
m , m
where
) m m (
e ) 1 m ( e ) 1 m (
C
C
1 2
0
2
1
0 1
v
V
V ) 1 ( V
V V
v v
= t
o =
o =
| =
?
?
= |
= o
36
Example 14-5
(a) Determine parameter o and | that can be used to model two CSTRs with interchange
using the tracer concentration data listed in Table E14-5.1.
(b) Determine the conversion of first-order reaction with k=0.03 min
-1
and t=40 min.
Solution
4 7 3 2
Te
4 5 3 2
Te
t 10 6 . 5 t 000438 . 0 t 129 . 0 t 3 . 17 921 C
min, 80 Fort
t 10 04 . 1 t 00146 . 0 t 642 . 0 t 6 . 59 2000 C
min, 80 Fort

+ =
>
+ =
s
C
Te
is the exit concentration of tracer determined experimentally.
37
2 T 1 T
2 T
1 T 2 T
1 T
C C
dt
dC
) 1 (
C ) 1 ( C
dt
dC
| | = o t
| + | = to
1 . 0
8 . 0
= |
= o
2 . 1 ) min 03 . 0 min)( 40 ( k
1
= = t

51 . 0
) 1 . 0 ( )] 2 . 1 )( 8 . 0 1 ( 1 . 0 )][ 2 . 1 )( 8 . 0 ( 1 . 0 1 [
) 1 . 0 ( )] 2 . 1 )( 8 . 0 1 ( 1 . 0 )][ 2 . 1 )( 8 . 0 ( 1 . 0 [
X
2
2
=
+ + +
+ +
=
) 7 . 0 X ( ) 55 . 0 X ( ) 51 . 0 X (
PFR CSTR el mod
= < = < =
38
Other Models of Nonideal Reactors Using CSTRs and PFRs
Figure 14-18(a) describes a real PFR or
PBR with channeling that is modeled as
two PFRs/PBRs in parallel. The two
parameters are the fraction of flow to the
reactors [i.e., | and (1-|)] and the fraction
volume [i.e., o and (1-o)] of each reactor.
Figure 14-18(b) describes a real PFR/PBR
has a backmix region and is modeled as a
PFR/PBR in parallel with a CSTR.
39
Figure 14-19 (a) and (b) show a real CSTR modeled as two CSTRs with
interchange. In one case, the fluid exits from the top CSTR (a) and in other
case the fluid exits from the bottom CSTR (b).
The parameter | represents the
interchange volumetric flow rate
and o the fractional volume of the
top reactor, where the fluid exits
the reaction system.
40
Closure
~In this section, models were developed for existing reactors to obtain a more
precise estimate of the exit conversion and concentration than estimates of
the examples given by the zero-order parameter models of segregation and
maximum mixedness.
~After completing this section, the student will use the RTD data and kinetic
rate law and reactor model to make predictions of the conversion and exit
concentrations using the tank-in-series and dispersion one-parameter
models.
~In addition, the student should be able to create combinations of ideal
reactors that mimic the RTD data and to solve for the exit conversions and
concentrations.
~Models of real reactors usually consist of combinations of PFRs, perfectly
mixed CSTRs, and dead spaces in a configuration that matches the flow
patterns in the reactor.
~For tubular reactors, the simple dispersion model has proven most popular.
41
~The parameters in the model, which with rare exception should not
exceed two in number, are obtained from the RTD data. Once the
parameters are evaluated, the conversion in the model, and thus in the
real reactor can be calculated.
~For typical tank-reactor models, this is the conversion in a series-
parallel reactor system. For the dispersion model, the second-order
differential equation must be solved, usually numerically.
~Analytical solution exist for first-order reactions, but as pointed out
previously, no model has to be assumed for the first-order system if
the RTD is available.
~Correlations exist for the amount of dispersion that might be expected
in common packed-bed reactors, so these systems can be designed
using the dispersion model without obtaining or estimating the RTD.
This situation is perhaps the only one where an RTD is not necessary
for designing a nonideal reactor.