Vous êtes sur la page 1sur 18

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305720565

Modeling and Simulation of Real Reactor

Technical Report · July 2016


DOI: 10.13140/RG.2.1.2687.7683

CITATION READS
1 2,892

1 author:

Md Firoz Kalam
Hochschule Bremerhaven
2 PUBLICATIONS   1 CITATION   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Desulfurization of Diesel Fuel View project

All content following this page was uploaded by Md Firoz Kalam on 29 July 2016.

The user has requested enhancement of the downloaded file.


Scientific Research
Modeling and Simulation of Real Reactor

Md. Firoz Kalam

Scientific Research Assistant


Institute of Thermal Process Engineering
Bremerhaven University of Applied Sciences
Bremerhaven, Germany
Table of Contents
1.0 Introduction .......................................................................................................................... 2
2.0 Objective .......................................................................................................................... 2
3.0 Background .......................................................................................................................... 2
3.1 Residence time distribution (RTD) .................................................................................. 2
3.1.1 Importance of RTD .................................................................................................... 2
3.1.2 Reactor modeling with the RTD:............................................................................... 3
3.2 Methods of Modeling .................................................................................................. 3
4.0 Measurement of RTD .......................................................................................................... 3
5.0 Tank In Series Model ........................................................................................................... 5
6.0 Axial Dispersion Model ....................................................................................................... 9
7.0 Reactor Performance .......................................................................................................... 11
7.1 Conversion ..................................................................................................................... 11
8.0 Discussion: ......................................................................................................................... 14
Above all, RTD data of tubular reactor is being satisfied for both the model. ........................ 14
9.0 Conclusion ......................................................................................................................... 15
Literature Cited ........................................................................................................................ 16

List of Table
Table 1 Tracer Data test ............................................................................................................. 4
Table 2 Exit age distribution data .............................................................................................. 5
Table 3 Calculations to determine tm and variance ................................................................... 7

List of Figure

Figure 1 Exit age distribution curve........................................................................................... 5


Figure 2 TIS model a) real system b) model system ................................................................. 6
Figure 3 Measured Vs Optimum Curve ..................................................................................... 8
Figure 4 Tank in series response to pulse tracer input for different number of tanks ............... 9
Figure 5 TIS model Vs Ideal CSTR ......................................................................................... 10
Figure 6 Tracer response curves in Axial Dispersion Model for different Bo numbers.......... 12
Figure 7 Conversion for different reactors............................................................................... 14
Figure 8 Conversion for different number of tanks ................................................................. 14
Tank in Series & Axial dispersion Model

1.0 Introduction

Most of the Chemical Industries, reactors which usually used are non ideal. So it’s really
important to know the modeling and simulation of real reactor in practice. To simulate real
reactor, it’s necessary to have Residence time distribution (RTD) of components. Residence
time distribution indicates how much time a particle spends time in reactor. For the
description of flow reactors, two different ideal reactor models i.e. Continuous Stirred tank
reactor (CSTR) and Plug flow Reactor (PFR) are used. CSTR assumes perfect mixing,
whereas PFR assumes no mixing. No real reactor has precisely the characteristics of either of
the two ideal reactors. For modeling reactor data tank-in-series (TIS) and axial dispersion
model (ADM) are widely used.

2.0 Objective

The main goal of this task is to modeling and simulating a non ideal plug flow reactor RTD
data by using Tank-In Series(TIS) and Axial Dispersion Model(ADM) and hence the
prediction of conversion of each model. Calculation has been done in Microsoft Excel 2007.

3.0 Background
3.1 Residence time distribution (RTD)

The performance of reactor depends upon the residence time of material in the reactor. In
case of plug flow reactor, the fluid or material flow in an orderly manner and all the materials
remains for equal time in the reactor. In batch reactor also all the fluid remains for equal time
in the reactor. This time spent by the fluid in the reactor is called the residence time.
Residence time distribution in a reactor is a characteristic of mixing that occurs in the
chemical reactor. The RTD characteristics of plug flow reactor and CSTR are quite different
because in PFR there is no axial mixing while in CSTR the content are uniformly mixed.
RTD offers a clue to the type of mixing occurring in the reactor [iv].

3.1.1 Importance of RTD

The knowledge of RTD is directly useful in vessel design, improvement in performance and
for the first order kinetics conversion calculations. Detailed knowledge of residence time
distribution with the understanding of the overall flow pattern helps in development of a
model of the system and this model can be used for the handling complicated kinetics. The

Chemical Reaction and Gasification | 2


Tank in Series & Axial dispersion Model

choice of RTD characterizing parameters is often a matter of balancing complicity against the
required degree of precision. In this regards, mixed model approach has gained importance in
RTD understanding. According to this concept, a real reactor is considered as consisting of
various flow regimes with known resistance time distribution (such as plug flow regions,
perfectly mixed regions) and various types of flow [iv].

3.1.2 Reactor modeling with the RTD:

There are many situations where the fluid in the reactor is neither well mixed nor
approximately plug flow. Hence a reactor is modeled in a number of ways. The RTD tells us
how long the various fluid elements have been in the reactor, but it does not specify the
exchange of matter between the fluid elements. The mixing of reacting species is one of the
major factors controlling the behavior of chemical reactors.
For the first order reactions knowledge of the length of time each molecule spends in the
reactor is all needed to predict conversion. Consequently mixing with the surrounding
molecules is not important. Therefore once the RTD is determined we can predict the
conversion that will be achieved in the real reactor provided that the specific reaction rate for
the first order reaction is known. For reactions other than first order knowledge of RTD is not
sufficient to predict conversion. In these cases the degree of mixing of molecules must be
known in addition to how long each molecule spends in the reactor. Hence models are
developed that account for the mixing of molecules inside the reactor [iv].

3.2 Methods of Modeling

There are two models significantly used to simulate non ideal reactors. One is one parameter
model and another is two parameter models. Example of One parameter model is Tank-In-
Series (TIS) and Axial Dispersion Model (ADM), and for two parameter model is CSTR with
dead volume and bypass model.

4.0 Measurement of RTD

The RTD is determined by injecting an inert chemical, molecule or atom, called a tracer, into
the reactor at some time t=0 and then measuring the tracer concentration, C, in the effluent
stream as a function of time.

Unfortunately, if the reaction rate is not first order, the RTD can’t be used directly to obtain
the conversion. Therefore considering first order reaction

Chemical Reaction and Gasification | 3


Tank in Series & Axial dispersion Model

Which is carried out in a 10cm diameter (D) tubular reactor 6.36m in length (L).The reaction
rate constant (K) is 0.25min− .The results of tracer test carried out on this reactor is given
below

t(min) 0 1 2 3 4 5 6 7 8 9 10 12 14
C(g/m^3) 0 1 5 8 10 8 6 4 3 2.2 1.5 0.6 0

Table 1 Tracer Data test

To obtain the parameters for the real reactor we have to first determine the residence time
distribution function. The quantity E (t) is called the residence-time distribution function. It is
the function that describes in a quantitative manner how much time different fluid elements
have spent in the reactor. The quantity E (t) dt is the fraction of fluid exiting the reactor that
has spent between time t and t+dt inside the reactor [iii]. We define the E (t

l e
E t = ………………………………….

Where,

E t = Residence time distribution

C = Pulse tracer signal

�� …………………… (2)
4
A = Area under the RTD curve=∫ ����

To obtain the area we divide the data into two regions 0-10minutes and 10-14 minutes
(referring Excel spreadsheet for calculation).The calculated E (t) and exit age distribution
curve are displayed

Chemical Reaction and Gasification | 4


Tank in Series & Axial dispersion Model

Integral function t(min) Cpulse(gm/m³) E(t)


1 0 0 0
4 1 1 0.0199867
2 2 5 0.0999334
4 3 8 0.1598934
2 4 10 0.1998668
4 5 8 0.1598934
2 6 6 0.1199201
4 7 4 0.0799467
2 8 3 0.05996
4 9 2.2 0.0439707
1 10 1.5 0.02998
4 12 0.6 0.011992
1 14 0 0

Table 2 Exit age distribution data

E(t)
0.250

0.200
E(t) min-1

0.150

0.100

0.050

0.000
0 2 4 6 8 10 12 14 16
t (min)

Figure 1 Exit age distribution curve

5.0 Tank In Series Model

In this model, RTD data otherwise known as tracer data was obtained from a pulse tracer
experiment carried out in a plug flow reactor. The result was used to determine the number of
equal sized ideal continuous stirred tank reactor (CSTR) arranged in series. The number of
ideal stirred tanks that will fit the RTD data of the non ideal plug flow reactor simply means
the number of ideal CSTR in series that will be equivalent to the non ideal PFR. This one
parameter model, for specificity can be regarded as ‘Tank In Series’ model. And the only

Chemical Reaction and Gasification | 5


Tank in Series & Axial dispersion Model

adjustable parameter to be extracted from residence time distribution RTD, data is the
number of tank n [v].

Figure 2 TIS model a) real system b) model system

For developing Tank in series model, it’s important to find the number of tanks theoretically.

=� ………………………………… (3)

Above expression represents the number of tanks necessary to model the real reactor as n
ideal tank in series. As the variance decreases number of tank increases.

Where,

n = number of tanks

τ = space time = t = mean residence time

σ = Variance

So we need to determine mean residence time and variance for calculating number of tanks in
series.

We know that,

α
Mean residence time, t = τ = ∫ t ∗ E t dt …………………(4)


And variance,� =∫ � ∗ � � �� ………………….(5)

The residence time and variance has been calculated using the generated values (from Excel)
displayed in following table

Chemical Reaction and Gasification | 6


Tank in Series & Axial dispersion Model

t(min) Cpulse(gm/m³) E(t) t*E(t) t²E(t)


0 0 0 0 0
1 1 0.0199867 0.019986676 0.019986676
2 5 0.0999334 0.199866755 0.399733511
3 8 0.1598934 0.479680213 1.43904064
4 10 0.1998668 0.799467022 3.197868088
5 8 0.1598934 0.799467022 3.99733511
6 6 0.1199201 0.71952032 4.317121919
7 4 0.0799467 0.559626915 3.917388408
8 3 0.05996 0.479680213 3.837441706
9 2.2 0.0439707 0.395736176 3.561625583
10 1.5 0.02998 0.299800133 2.998001332
12 0.6 0.011992 0.143904064 1.726848767
14 0 0 0 0

Table 3 Calculations to determine tm and variance

t = τ =5.1552298 min

σ =6.108482 min²

So the ideal number of tanks, n required for the Tank in series model calculated by using
above values

τ
n= σ = .

Now we need to measure the optimum number of tank which fit the RTD curve.

The equation of RTD for n CSTR in series is given below


− −
E t = − ! i
e i ………………………….(6)

Where,

τi =residence time for i number of tanks, i=1, 2, 3………100

Using solver we can find the optimum number of tanks.

Hence, n = 5.0017096≈

Therefore, measured and optimum RTD curve generated which is shown below

Chemical Reaction and Gasification | 7


Tank in Series & Axial dispersion Model

0.25

0.2

0.15
E(t)min-1

Measured
0.1
Optimum

0.05

0
0 2 4 6 8 10 12 14
t(min)

Figure 3 Measured Vs Optimum Curve

Using above formula we can also find the RTD curve for different number of tanks which is
shown below

0.8

0.7

0.6
n=1
0.5 n=2
E(t) min-1

n=5
0.4
Optimum
0.3 n=10
n=20
0.2
n=50

0.1 n=100

0
0 2 4 6 8 10 12 14 16
time, t (min)

Figure 4 Tank in series response to pulse tracer input for different number of tanks

Chemical Reaction and Gasification | 8


Tank in Series & Axial dispersion Model

Generated RTD curve for Tank in series model (TIS) in comparison with Ideal CSTR is
shown below

1.2

0.8 Ideal CSTR


Single reactor
n=2
E(t)

0.6
n=5
n=10

0.4 n=20
n=50
n=100
0.2

0
0 5 10 15
time (min)

Figure 5 TIS model Vs Ideal CSTR

6.0 Axial Dispersion Model

The dispersion model is also used to describe non ideal plug flow reactors. In this model,
there is an axial dispersion of the material, which is governed by analogy to Fick’s law of
diffusion. Every component in the mixture is transported through any cross section of the
reactor at a rate which resulting from molecular and convective diffusion. The ADM is
characterized by Bodenstein number (Bo) that cause deviation from ideal plug flow behavior.
The Bodenstein number is the single parameter of this Model. As Bo increases from 0 to �
the flow of the pattern also changes from complete mixing (CSTR) to no mixing (PFR).This
model gives different results for different boundary conditions[v].

Chemical Reaction and Gasification | 9


Tank in Series & Axial dispersion Model

The Axial Dispersion Model has the following analytical Solution

� − −�
C*=C*f= √ exp …………………………………(7)
�∗� 4∗�

Where,

C∗ = relative concentration

C = Electrical Signal

f = unknown factor

Bo = Bodenstein number

θ = dimensionless time =
τ

Considering Open vessel Boundary condition

8
= + ……………………………………(8)

By evaluating optimum Bodenstein number, it’s easier to find out the dispersion whether its
large dispersion or small dispersion.

Bo = .

We also know from definition of Bodenstin number

Bo =

Where,

U = Superficial velocity

L = Reactor length

D = Dispersion Coefficient

But is the vessel dispersion number which is reciprocally proportional to Bodenstein


number.

Hence, = . , which is greater than 0.01 means large extent of dispersion.

Chemical Reaction and Gasification | 10


Tank in Series & Axial dispersion Model

Predicted concentration curves for different Bodenstein number is shown below

3.5

Bo=1
2.5
Bo=5

2 Reactor Bo=7.5
C*

Bo=10
1.5 B0=20
B0=30
1
Bo=40

0.5 Bo=50
B0=100
0
0 0.5 1 1.5 2 2.5 3
Θ

Figure 6 Tracer response curves in Axial Dispersion Model for different Bo numbers

7.0 Reactor Performance

As mentioned in earlier residence time distribution RTD data is sufficient to predict the
reactor performance. Basically three parameters considered for evaluating reactor
performance, and these are conversion, yield and selectivity. Conversion means how much of
reactant has reacted .Yield represents the ratio of how much product produced to initial
substance. Selectivity is the ratio of yield to conversion (only in case of parallel reaction).

In this paper, conversion is taking in account for evaluating the reactor performance.

7.1 Conversion

For analyzing and comparison purposes the conversion is being calculated for Ideal PFR,
Ideal CSTR, TIS model and ADM model. The relations for calculating the conversion are
given below (refer the excel spreadsheet for calculation).

U i g = − + i

UI a PF = − e−

Chemical Reaction and Gasification | 11


Tank in Series & Axial dispersion Model

U I = −
+ i

B
4 x
U = − B ∗ −B ∗
+ x − − x

Where,

4
q=√ +

U i g = conversion in single ractor

UI a PF = conversion in Ideal PFR reactor

U I = conversion in Tank In Series Model

U = Conversion in Axial Dispersion Model

Predicted conversion for different reactors by using above formula are given below

Conversion
Types U(%)
Single Reactor 56.30912531
Ideal PFR 72.44007501
TIS 68.24291605
ADM 68.06024901

Table 3 Conversion for different reactors

The conversion for different number of tanks is displayed in the table below

Tanks Conversion
n U%
1 56.30912531
2 63.01862172
3 65.77416008
4 67.28032827
5 68.2306354
6 68.88508399
10 70.24778246
20 71.32039077
50 71.98632047
100 72.21219731
Table 4 Conversion for different number of tanks

Chemical Reaction and Gasification | 12


Tank in Series & Axial dispersion Model

Generalized graph by using both of the above tables

Conversion
80

70

60

50
U(%)

40

30

20

10

0
1
Singel CSTR 56.30912531
Ideal PFR 72.44007501
Tank in Series 68.24291605
Dispersion Model 68.06024901

Figure 7 Conversion for different reactors

80

75
conversion (X) %

70

65

60

55

50
0 10 20 30 40 50 60 70 80 90 100 110
number of reactors in series (n)

Figure 8 Conversion for different number of tanks

Chemical Reaction and Gasification | 13


Tank in Series & Axial dispersion Model

8.0 Discussion:

We have seen that we can apply both of these one parameter model (TIS and ADM) to
tubular reactors using the variance of RTD. For the first order reactions, the two models can
be applied with equal case. However, the tank-in series model is easier to use to obtain the
effluent concentration and conversion. However, we need to see whether the accuracy of
using the tank-in-series model over dispersion model. These two models are equivalent when
the Bodensten number (Bo) is equivalent to number of tanks in series, n, by the equation
n= + .Number of tanks calcuted from Bodenstein number is 4.77 which is very close to

the value of 4.35 calculated from equation (2).The reaction rate parameter is usually known
(i.e Da), but the Bodenstein number is not known because it depends on the flow and the
vessel. Hence we could find Bodenstein number by using either open vessel or closed vessel
conditions.

From the RTD data we require the . 9 ≈ number of tanks in series.For this number

of tanks, best fit curve obtained. And dispersion no. = . , which indicates large amount

of dispersion in open vessel condition.From the data, the number of tank increases as the
varience decreases.

An ideal CSTR has an exponential residence time distribution. In reality, it is impossible to


obtain such rapid mixing, especially on industrial scales where reactor vessels may range
between one and several tens of cubic meters, and hence the RTD of a real reactor will
deviate from the ideal exponential decay. This deviation from the ideal CSTR can be seen in
the Figure 5. The Figure 5 shows the exit age distribution for an ideal CSTR and TIS model
with different number of tanks.

Figure 7 and 8 represents the reactor performance in terms of conversion. Ideal PFR has more
conversion rate than any other model, which is usual. And TIS and ADM model almost has
same conversion. Figure 8 shows, conversion increases gradually for dramatically increased
number of tanks. Reactor performance could also be determined by yield and selectivity.

Above all, RTD data of tubular reactor is being satisfied for both the model.

Chemical Reaction and Gasification | 14


Tank in Series & Axial dispersion Model

9.0 Conclusion

A comparison of two modeling approaches for simulating the performance of tubular reactors
using the axial dispersion and tanks-in-series models is described.. The performance of the
reactors is described in terms of substrate conversion. The equation n = + is used to

correlate the parameter of the dispersion model (Bo) with that of the tanks-in-series model (n)
for the entire range of dispersion from plugs flow to CSTR. The predictions of the two
models agree well, especially at low dimensionless residence times and low Bodenstein
number. Practically, the predictions of the two models are essentially equivalent when the
above equation is used to relate their two parameters. However, the tanks-in-series model is
simpler and has computational advantages over the dispersion model, although its physical
basis is not as clear as that of the dispersion model.

Chemical Reaction and Gasification | 15


Tank in Series & Axial dispersion Model

Literature Cited

[i] Prof. Dr.-Ing. Wilfreid Schuetz, Hochschule Bremerhaven, Process Engineering and
Energy Technology Faculty, “Chemical Reaction and Gasification Lecture notes”, winter
semester 2014-15

[ii] Levenspiel O., Chemical reaction engineering, 2nd ed., 107-9, 253-99, John Wiley, New
York, N.Y. USA (1972)

[iii] Models for Non ideal reactors, University of Michigan


http://www.umich.edu/~essen/html/byconcept/chapter14.pdf

[iv] Kanse σitin and Dawande, “RTD Studies in Plug Flow Reactor and its Simulation with
Comparing σon Ideal Reactors”, Vol.1 (2),42-48,Feb(2012)

[V] Ibrahim M. Abu-Reesh and Basel F. Abu-Sharkh, “Comparison of Axial Dispersion and
Tanks-in-Series Models for Simulating the Performance of Enzyme Reactors”, Ind. Eng.
Chem. Res. 2003, 42, 5495-5505

Chemical Reaction and Gasification | 16

View publication stats