Chemical Engineering Paper

I NTERNATIONAL J OURNAL OF C HEMICAL
R EACTOR E NGINEERING
Volume 3
2005
Article A14
Hydrogenation of Acetylene: Kinetic

Studies and Reactor Modeling
Navid Mostoufi
Ali Ghoorchian
Rahmat Sotudeh-Gharebagh
University of Tehran, mostoufi@ut.ac.ir

University of Tehran, alighoorchian@yahoo.com
University of Tehran, sotudeh@ut.ac.ir
ISSN 1542-6580
c
Copyright 2005
by the authors.
All rights reserved.
Brought to you by | University of Saskatchewan

Authenticated | 128.233.210.97
Download Date | 10/4/13 4:08 PM
Hydrogenation of Acetylene: Kinetic Studies and

Reactor Modeling
Navid Mostoufi, Ali Ghoorchian, and Rahmat Sotudeh-Gharebagh
Abstract
The kinetics of acetylene hydrogenation has been studied in a fixed bed reactor of a commercial Pd/Al2O3 catalyst. The experiments were carried out at 30,
50 and 70 o C with various feed compositions at atmospheric pressure. The experiments were repeated at 70 o C in the presence of the used catalyst to determine
the effect of the catalyst deactivation where the corresponding deactivation rate
constant was determined in order to predict the activity of the catalyst during each
run. Two well known kinetic models were used for a nearly similar catalyst to predict the experimental data of this work and none of them were found satisfactory.
A new model was then proposed to fit the experimental data. The hydrogenation
reactor was also simulated at industrial operating conditions with the proposed
kinetics for both plug and dispersion flows. The results of these simulations were
almost close to each other in most cases.
KEYWORDS: Acetylene hydrogenation, kinetics, modeling, plug and dispersion
flow, simulation, deactivation

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling
1. INTRODUCTION
Polyethylene has been a key product for many industries since 1960s. The feed of the
polymerization reactor, which comes from the olefin plant, is a mixture of hydrocarbons mainly
consisting of ethylene. An undesired impurity in the ethylene stream is acetylene at approximately
0.3 to 2% of the effluent of the olefin plant. This small amount of impurity may lead to
undesirable polymer properties. The amount of acetylene in the feed of ethylene polymerization
reactor should not exceed 2-3 ppm (Bos et al., 1993). There are two possible ways for eliminating
acetylene from ethylene: hydrogenation of acetylene to ethylene and separation of acetylene from
the main stream (Haehn et al., 1997, Brodzinski and Cybulski, 2000). The most common
industrial method of eliminating acetylene is hydrogenation, as the separation method is both
expensive and dangerous (Vincent and Gonzales, 2001). Thus, in order to reach the desired
amount of acetylene for the polymerization step, it is selectively hydrogenated to ethylene in an
adiabatic fixed bed catalytic reactor. The reaction should be selective because many other chain
reactions can occur during hydrogenation if it is not controlled. Almost all these reactions are not
desirable and have to be prevented.
There are three major reactions considered in this system (Bos et al., 1993; Westerterp et al.,
2002):
C2 H 2 + H 2
C2 H 4 + H 2
C2 H 2 + 2H 2
C2 H 4
C2 H 6
C2 H 6
(1)
(2)
(3)
Reactions 1 and 2 are the main reactions occurring in such systems. However, there are also some
side reactions which inevitably occur during acetylene hydrogenation. The most important side
reaction is the reaction 3, i.e., the direct hydrogenation of acetylene to ethane. Acetylene can also
be hydrogenated to heavier products, called green oils, by oligomerization. However, according to
Leviness et al. (1984) and Sarkani et al. (1984) even direct hydrogenation of acetylene to ethane
can be neglected.
In the industrial plants, the hydrogenation process is controlled in a multi-bed adiabatic
reactor. There are two methods to achieve this goal (Weiss, 1996):
a) Hydrogen is separated from the feed before entering the hydrogenation reactor and then is
added to the feed at a desired amount. This method, called tail-end hydrogenation, uses
hydrogen concentration as a controlling parameter in the reactor in addition to the feed
temperature.
b) The feed, which may contain up to 20% hydrogen, directly enters the hydrogenation
reactor. This type of process is called front-end hydrogenation. In this type of systems, the
only controlling parameter would be the feed inlet temperature.
Many catalysts have been studied for hydrogenation of acetylene. Catalysts based on nickel
sulfide (Anderson et al., 1948), nickel or zinc based catalysts on alumina or silica (Muller et al.,
1987; Kummerow et al., 1983), cadmium, calcium, barium, strontium or magnesium on Cr2O3
(Weisang and Engelhard, 1970) as well as copper on alumina, magnesia or silica (Taghavi et al.,
1978) have been used in some cases. However, it was found that palladium is the most selective
metal for acetylene hydrogenation and the most common commercially used catalyst is
Pd/alumina (Brodzinski and Cybulski, 2000; Vincent and Gonzalez, 2001; Godinez et al., 1995;

International Journal of Chemical Reactor Engineering
Vol. 3 [2005], Article A14
Vincent and Gonzalez, 2002). This catalyst is usually prepared either by ion exchange or by
precipitation in order to produce a low dispersion, low metal content supported catalyst (Vincent
and Gonzalez, 2001).
In spite of many studies carried out by the researchers to describe the mechanism of
hydrogenation of acetylene and its corresponding kinetic rate, almost no attention has been paid to
the modeling of the industrial reactor of acetylene conversion. The aim of this work is, therefore,
to assess the existing kinetics in the literature and then provide a complete description of the
acetylene hydrogenation reactor based on a theoretical model.
2. THEORY
2.1. Kinetics
The process of acetylene hydrogenation is consisted of adsorption of acetylene and hydrogen on
the catalyst surface, chemical reaction between the adsorbed species, and desorption of the
products from the surface (Vicent and Gonzalez, 2001). Bond (1962) proposed that since the
enthalpy of adsorption of acetylene is higher than that of ethylene, the surface coverage ratio of
acetylene to ethylene would be always high. Therefore, in this case it was expected that if a
mixture of acetylene and ethylene is used, hydrogenation of ethylene would not start until all the
acetylene in the mixture is consumed. However, the experiments conducted by Bos et al. (1993)
and Brodzinski and Cybulski (2000) indicated that this assumption is not realistic and
hydrogenation of ethylene cannot be completely prevented in any case. On the other hand, AlAmmar and Web (1978, 1979), Menshchikov et al. (1975) and Mc Gown et al. (1978) proposed
that the catalyst surface contains at least two different types of active sites. Furthermore,
Brodzinski and Cybulski (2000) proposed a model based on two active sites. They suggested that
these sites are created on the palladium surface by carbonaceous deposits. Some of these sites can
only take part in acetylene hydrogenation and others may be open to all the species in the gas
phase. Figure 1 show a simplified representation of active sites on the catalyst surface which is
proposed by Brodzinski and Cybulski (2000). As seen in this figure, different types of species are
adsorbed and react on different types of sites. According to Brodzinski and Cybulski (2000), a
type of site may exist which is too small for the species other than acetylene and hydrogen to be
adsorbed on. As compared to ethylene, acetylene is selectively hydrogenated on these sites by
hydrogen atoms which are also adsorbed on these sites.
Figure 1. Simplified representation of active sites on the catalyst surface (after Brodzineski and
Cybulski, 2000)

As mentioned above, researchers believe that different types of sites are formed on the surface of
the catalyst pellet due to the presence of carbonaceous deposits in the process of acetylene
hydrogenation. However, the exact source of these deposits is still in question. Based on the
assumption that the origin of formation of C4 oligomers (which are the precursors to carbon
deposits) is the acetylene adsorbed on the surface of catalyst, Al-Ammar and Webb (1978a,
1978b, 1979) suggested that these deposits are the source of deactivation and may contribute to
acetylene hydrogenation. Recently, Fasi et al. (2000) found that there are several types of surface
carbon and not all of them necessarily participate in the reaction. Larsson et al. (1998) showed
that it is the type of carbon not its amount that results in an increase in the selectivity to ethylene.
Therefore, carbon deposits can accept hydrogen at low temperatures and then release this
hydrogen at higher temperatures to participate in the hydrogenation procedure by a hydrogen
transfer mechanism (Vincent and Gonzalez, 2001). Based on this information, at least two
possible mechanisms may be suggested for this reaction system:
Mechanism 1: This mechanism is based on the reaction of acetylene in the form of vinylidene
with hydrogen molecules which are simultaneously adsorbed on the catalyst surface in a
competitive way. This is a Langmuir-Hinnshelwood mechanism and is the most common one
which has been proposed by almost all the researchers for such reaction system.
Mechanism 2: This mechanism involves the hydrogen transfer from carbonaceous deposits at
higher temperatures to the vinyl intermediate. This mechanism is called Al-Ammar mechanism.
These two proposed mechanisms are shown schematically in Figure 2. In addition to these
two main mechanisms, i.e., Langmuir-Hinshelwood and Al-Ammar mechanisms, other
mechanisms have been also considered for this system of reactions (Westerterp et al., 2002). Of
course, the temperature in industrial reactors of acetylene hydrogenation is les than 100 C in
which the only source of hydrogen comes from the dissociative adsorption of hydrogen followed
by direct hydrogenation over Pd sites. However, there is an alternative mechanism which starts to
happen at about 150 C which is much more selective towards ethylene formation than the low
temperature mechanism.
This mechanism involves hydride transfer from the growing
carbonaceous layer. This alternative mechanism becomes dominant at temperatures in excess of
175 C. Vincent and Gonzalez (2002) have addressed this point in considerable detail.
Figure 2. Representation of the two proposed mechanisms on catalyst surface (after Vincent and
Gonzalez, 2001)

Different kinetic models have been proposed based on each of the above described
mechanisms. Nevertheless, due to the complexity of the reactions in this system, none of the
proposed kinetics can be considered as the best, yet. Among them, the kinetic expressions
proposed by Boss et al. (1993), Brodzinski and Cybulski. (2000) and Menshchikov et al. (1975)
seem to be more acceptable and have been used by other researchers (Westerterp et al., 2002;
Vincent and Gonzalez, 2001). These kinetics expressions are given in Table 1.
Table 1. Kinetic models studied in this work
Kinetics Model
rC2 H 2 =
rC2 H 6 =
rC2 H 2 =
rC2 H 6 =
rC2 H 2 =
rC2 H 6 =
Reference
k1 PC2 H 2 PH 2
(1 + k 2 PC2 H 2 )(1 + k 3 PH 2 )
Bos et al. (1993)
k 4 PC2 H 4 PH 2
(1 + k 5 PC2 H 2 + k 6 PC2 H 4 ) 3
k1 PC2 H 2 PH 2
(1 + k 2 PC2 H 4 )(1 + k 3 PH 2 )
Menshchikov et al. (1975)
k 4 PC2 H 4 PH 2
(1 + k 5 PC2 H 4 )(1 + k 6 PH 2 )
k1 PC2 H 2 PH 2
(1 + k 2 PC2 H 4 )(1 + k3 PH 2 )
This work
k 4 PC2 H 4 PH 2
(1 + k5 PC2 H 4 )1.25 (1 + k 6 PH 2 )
2.2. Modeling
The acetylene hydrogenation system considered in this work consists of only Reaction (1) and (2).
All other side reactions are neglected. The industrial reactor of acetylene hydrogenation operates
at non-isothermal conditions. Therefore, in order to model such a reactor, the mass balance
equations have to be coupled with the energy balance equation and to be solved simultaneously.
Up to now most of the simulation studies in this field have been based on the plug flow
assumption for the reactor. Moreover, the few researchers, who have considered the dispersion
model, did not report temperature and concentration profiles in a large scale reactor or make a
comparison between these two models (Godinez et al., 1995; Szukiewicz et al., 1998). The
acetylene hydrogenation reactor has modeled by both models in this study.
The mass and energy balance equations, assuming the plug flow pattern for the gas, are as
follows:

dC A
=
dz
(1 )
rA
u
F
dT
CP
=
A
dz
(4)
H i ri
(5)
The second method of simulating this system is to take dispersion of the gas into
consideration. In this case, the mass balance equation should be rewritten as follows:
DA d 2C A
u dz 2
dC A
+
dz
(1 )
rA = 0
u
(6)
Thermal dispersion may be neglected in this case as the ratio of thermal dispersion coefficient to
mass dispersion coefficient is very low and the energy balance equation would be the same as the
previous case (Equation 5) in the modeling. Therefore, mass balance equations (Equation 6)
should be solved for all species together with the energy balance equation (Equation 5),
simultaneously. In order to solve the mass balance equation (Equation 6), two boundary
conditions are needed for each species. In this case, Dankwerts boundary conditions may be used
(Fogler, 1999) as given bellow:
at
at
z = 0; C A0 =
z = L;
Da
u
CA
z
z =0
+ C A (0 + , t )
CA
=0
z
(7)
(8)
Equations (5), (6), (7) and (8) form a set of boundary-value differential equations and could be
solved by the finite difference method (Constantinides and Mostoufi, 1999).
3. EXPERIMENTAL
3.1. Catalyst and Gases
The catalyst was Pd/Al2O3 with commercial name of G58-B from Sud-Chemie which is currently
used in many petrochemical complexes. Both new and used catalysts were employed in the
experiments. The used catalyst was acquired from an industrial reactor being in service for six
months before getting deactivated and taken out from the reactor.
The gases used in this work were 99.65% pure C2H2, 99.99% pure C2H4 and 99.99% H2 along
with 99.95% nitrogen. The latter was used as the diluting gas to prevent high conversion of
acetylene during the experiments. In order to obtain the desired concentration of hydrogen in the
mixture, hydrogen gas was premixed with nitrogen at a 1:9 ratio. To make such a premixed gas,
the container was first vacuumed and then filled by a calculated amount of hydrogen and then
slowly pressurized with nitrogen up to 6 barg.

3.2. Apparatus and Procedure

The experimental set-up for measuring the reaction rates of acetylene hydrogenation reactions is
shown in Figure 3. The 3 mm ID U-shaped micro reactor filled with 0.3 grams of finely
pulverized catalyst with a mesh of 180 to 300 m. Such size of the catalyst ensures the absence of
the effect of inter-particle heat and mass transfer resistances during the experiments in the
operating conditions of this work. Flow rates of the inlet streams were measured by three
rotameters. The reactor was placed in a warm water bath equipped with temperature controller
and heater. Compositions of both inlet and outlet streams of the reactor were analyzed by a gas
chromatograph (GC) equipped with a FID analyzer. At the beginning of each run, the feed was
analyzed by the GC before entering the reactor. During the experiments, the product gas from the
reactor was also conducted to the same GC for determining its composition after the reaction. The
feed flow rate varied between 30 to 110 mL/min and its composition was changed from high about
25% to less than 1% of acetylene content. The experiments were carried out at three different
temperatures, i.e., 30, 50 and 70 C.
Figure 3. Simplified schematic diagram of the experimental set up of this study. 1-C2H2 container
2-C2H4 container 3- H2+N2 container 4-Pressure regulator 5-Rotameter 6-Reactor 7Thermocouple and temperature indicator 8-Heater 9-Gas chromatograph.
In addition to the new catalyst, the experiments were repeated with the deactivated catalyst to
obtain a deactivation coefficient for this catalyst. The experiments for the old catalyst were
carried out only at 70C because as the catalyst becomes deactivated in the reactor, the feed
temperature is increased to counter this effect. The feed temperature used at industrial reactors at
such conditions is usually between 70 to 80C.

4. RESULTS AND DISCUSSION

4.1. Kinetics
Using the data of these experiments, the reaction rates of acetylene consumption and ethane
formation in each case were found to be as follows:
rC2 H 2 =
rC2 H 6 =
FC2 H 2 (in )
FC2 H 2 ( out )
m
FC2 H 6 ( out )
FC2 H 6 (in )
m
(9)
(10)
The calculations and discussions done below are based on these reaction rates.
Initially, the two well known kinetic models of Bos et al. (1993) and Menshchikov et al.
(1975) were considered as the base models and fitted the experimental data of this work to these
models to obtain new kinetic parameters for the catalyst employed in this study. It is worth
mentioning that in both models of Bos et al. (1993) and Menshchikov et al. (1975), the parameters
have been provided for a different type of catalysts than that studied in this work and as the
different types of catalysts may have different metal content and different porosities which
certainly affect the kinetic studies. Therefore, new kinetic parameters have to be obtained to fit
the new catalyst behavior. New reaction rate constants, evaluated from fitting the experimental
data of this work to the above mentioned kinetic models, are given in Table 2. An Arrhenius type
of temperature dependency is considered for the reaction rate constants as follows:
k i = ki , 0 exp
E a ,i
(11)
RT
The parity plot of the calculated reaction rates based on the kinetic model of Bos et al. (1993)
against the experimental reaction rates of this work are shown in Figures 4a and 4b for acetylene
consumption rate and ethane formation rate, respectively. The constants of the Bos et al. (1993)
model used in this figure are those reported in Table 2. It can be seen in Figure 4a that the model
of Bos et al. (1993) over-predicts the acetylene consumption rates obtained in this work.
According to this figure, the higher the temperature, the higher is the deviation of the model from
the reality. Figure 4b is the same comparison for ethane formation rate. It is evident in this figure
that data scattering is less than what was observed for acetylene consumption rate (Figure 4a).
Nevertheless, at lower temperatures the model of Bos et al. (1993) underestimates while at high
temperatures the model overestimates the reaction rate and the difference between calculated and
observed data is quit high.
The same comparison for the kinetic model of Menshchikov et al. (1975) is shown in Figures
5a and 5b. The constants of the Menshchikov et al. (1975) model used in this figure are those
reported in Table 2. Figure 5a compares the calculated and observed rates of acetylene
consumption. Although the data points in this figure are scattered, these points are also evenly
distributed around the unity line. This fact indicates that the kinetic model of Menshchikov et al.

0.7
0.9
(a)
(b)
0.8
0.6
Calculated Reaction Rates (mmol/kg cat.s)
30 C
50 C
70 C
0.7
0.6
0.5
0.4
0.3
0.2
30C
50C
70C
0.5
0.4
0.3
0.2
0.1
0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.1
0.2
0.9
0.3
0.4
0.5
0.6
0.7
Experimental ReactionRates (mmol/kg cat.s)
Experimental Reaction Rates (mmol/kg cat.s)
Figure 4. Parity plot of experimental reaction rates vs those calculated by the model of Bos et al.
(1993) (a) acetylene consumption rate (b) ethane formation rate
0.45
0.50
(b)
(a)
0.35
0.45
0.40
30 C
50 C
70 C
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.00
0.40
30 C
50 C
70 C
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.00
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
Figure 5. Parity plot of experimental reaction rates vs those calculated by the model of
Menshchikov et al. (1975) (a) acetylene consumption rate (b) ethane formation rate
(1975) for acetylene consumption rate is adequate, even though the experimental error has caused
the data points to be scattered. This point could be more clearly understood if one compares
Figure 5a with Figure 4a (corresponding to the same reaction rate with a different equation). In
fact, in Figure 4a the data points are scattered as well as biased toward higher values. Thus, even
if scattering of the data is contributed to the experimental error in Figure 4a, the kinetic model of
Bos et al. (1993) is inadequate to estimate the acetylene consumption rate properly. Therefore, it
can be concluded that the kinetic model of Menshchikov et al. (1975) with the newly calculated
parameters is an appropriate model for evaluating the acetylene consumption rate for the catalyst
employed in this work. Figure 5b is the same comparison for ethane formation rate. Although the
agreement between this model and the experimental ethane formation rates is considerably better
than that of the model of Bos et al. (1993) (compare Figure 4b with Figure 5b), however, still there

are some points where the agreement is not satisfactory. It should be noted that the agreement
between both Bos et al. (1993) model and Menshchikov et al. (1975) model is considerably better
for the points with lower acetylene content in terms of acetylene consumption rate than those
obtained with high acetylene content in the feed. Therefore, acetylene concentration of more than
20% in the feed, in both models would result in an unacceptable difference between calculated and
observed rates.
By considering all the above mentioned parity plots presented in Figures 4a, 4b, 5a and 5b, it
can be concluded that the model of Bos et al. (1993) cannot predict the reaction rates of the
catalyst employed in this study in the range and operating conditions of this study for either
acetylene consumption rate or ethane formation rate. However, the model of Menshchikov et al.
(1975) is able to predict the rate of acetylene consumption satisfactorily for the catalyst and
conditions of this study while its rate of ethane formation still needs to be improved. Therefore, a
new kinetic model is proposed here which consists of the acetylene consumption rate of the model
of Menshchikov et al. (1975), for with new parameters which have been obtained in this study,
and a new rate expression for ethane formation which better fits the experimental data. After
simplifying, this new model is given as follows:
48.01exp
rC2 H 2 =
668.6
404.3
1 + 584.59 exp
PC2 H 4 1 + 2.855 exp
PH 2
T
T
202.67 exp
rC2 H 6 =
146.8
PC2 H 2 PH 2
T
4784
PC2 H 4 PH 2
T
1502.7
1 + 0.0742 exp
PC2 H 4
T
1.25
400
1 + 2.89 exp
PH 2
T
(12)
(13)
This model is presented in its general from in Table 1 and its parameters are also given in Table 2.
It is worth noting that the power 1.25 in the denominator of Equation (13) is only a fitting
parameter and does not correspond to mechanism of the reaction.
A comparison between the calculated ethane formation rate and the experimental results of
this work is shown in Figure 6. It can be seen in this figure that there is a good agreement
between the model prediction and real reaction rates. The performance of the new rate equation,
presented in Equation (13), is considerably improved as compared to those proposed by Bos et al.
(1993) and Menshchikov et al. (1975). It seems that the reason that the new kinetic expressions
[Equations (12) and (13)] fit the experimental data of this work better than the previous data in the
literature is that the catalyst employed in this work (which is an industrial catalyst) slightly differs
from those employed by other researchers.

10
0.5
30 C
50 C
70 C
0.4
0.3
0.2
0.1
0
0
0.1
0.2
0.3
0.4
0.5
Figure 6. Parity plot of experimental ethane formation rates vs those calculated by the model of
this work.
4.2. Reactor Modeling

The two flow models coupled with each of the three kinetic models described in the Theory
section were solved for an industrial-scale reactor. The operating conditions considered for the
simulation are listed in Table 3. It is worth mentioning that in the industrial acetylene
hydrogenation units, two reactors in series are employed for complete conversion of acetylene in
the feed (Weiss, 1996). The values given in Table 3 are typical for the first hydrogenation reactor.
Results of this simulation are shown in Figures 7a-d in terms of profiles of temperature, acetylene
conversion, ethylene formation, and ethane formation along the reactor, respectively. In these
figures, the results of simulation of the reactor by the two flow models, i.e., plug flow and
dispersion flow, which are coupled with the three kinetic models, i.e., Bos et al. (1993),
Menshchikov et al. (1975) and the proposed model in this study are shown.
Parameter
C2H2
C2H4
C2H6
H2
Reactor length
Inlet temperature
Pressure
Table 3. Modeling conditions

Unit
%
%
%
%
m
K
bar(g)
Value
4
93
1
2
3
293
25
Figure 7a illustrates the temperature profiles of the reactor for the three kinetic models as
coupled with the two flow models. It can be seen in this figure that all these models predict
almost the same final temperatures for the reactor. In addition, the results of dispersion and plug

11
models are actually close to each other. The exit temperature of the reactor is about 360 to 365 K
according to all models which are close to the exit temperature of the product from the first
hydrogenation reactor in the industrial acetylene converting units.
50
370
(a)
Despersion
Plug
360
This Work
350
Acetylene Conversion (%)
Bos et al.
This Work
Temperature (K)
Menshchikov et al.
40
This Work
340
330
Bos et al.
320
310
This Work
30
20
10
Bos et al.
300
Bos et al.
0
290
0
0.5
1.5
2.5
0.5
1.5
2.5
Reactor Length (m)
Reactor Length (m)

25
(d)
(c)
Dispersion
Plug
Dispersion
Plug
2.5
20
Menshchikov et al.
Menshchikov et al.
Ethane Formation (%)
Ethylene Formation (%)
Menshchikov et al.
(b)
Plug
Dispersion
Menshchikov et al.
1.5
This Work
15
Bos et al.
10
This Work
5
This Work
Bos et al.
0.5
Menshchikov et al.
Bos et al.
0
0
0.5
1.5
2.5
0.5
Reactor Length (m)
1.5
2.5
Reactor Length(m)
Figure 7. Simulation results for different flow patterns combined with kinetic models investigated
in this work (a) temperature profiles (b) acetylene conversion profiles (c) ethylene
formation profiles (d) ethane formation profiles.
The corresponding acetylene conversion profiles are shown in Figure 7b. This conversion is
calculated from the following formula:
X C2 H 2 =
FC2 H 2 ( in )
FC 2 H 2
FC 2 H 2 ( in )
(14)
It is also seen in Figure 7b that the profiles are close to each other and so do the exit conversions.
This is an expected trend since all three kinetic models considered in this study provide quite the

12
same acetylene hydrogenation rates (see Table 1). This figure illustrates that only about half of
the acetylene is eliminated in the first hydrogenation reactor and the rest of this task remains to be
accomplished in the second reactor. The reason for not completing the conversion of acetylene in
a single reactor is controlling the temperature, as discussed in the introduction section and shown
in Figure 7a.
Ethylene formation can be calculated from:
X C2 H 4 =
FC2 H 4
FC2 H 4 (in )
FC 2 H 4 ( in )
(15)
The profiles of ethylene formation along the reactor are shown in Figure 7c. It can be seen in this
figure that the kinetic model of Menshchikov et al. (1975) predicts the highest ethylene formation
among the three models and the model developed in this work predicts the lowest. The difference
between the predictions of the three models observed in Figure 7c is due to the fact that in the
process of ethylene formation, two reaction rates (i.e., acetylene conversion and ethane formation)
are involved. Although all three kinetic models considered in this study provide almost the same
acetylene conversion rates, they are dissimilar in the rate of ethane formation (see Table 1).
Therefore, different profiles are obtained from each kinetic model for ethylene formation. This
figure also reveals that regardless of the kinetic model used in the simulation, the plug model
provides lower ethylene formations compared to the dispersion flow model. This is some thing
that can be expected because in dispersion flow the back mixing phenomena helps the conversion
of acetylene to be higher than that of plug flow. Consequently, the ethylene formation would be
also higher in this case.
Ethane formation is calculated from:
X C2 H 6 =
FC2 H 6
FC2 H 6 (in )
FC2 H 6 ( in )
(16)
Figure 7d shows the profiles of ethane formation along the reactor length for the models
considered in this work. It is clear in this figure that each kinetic model predicts a different ethane
formation rate as compared to another one. The discussions made for Figure 7c regarding the
difference of the three kinetic models in terms of ethane formation rate are also valid here. In fact,
the difference between these models, which is mainly originated from the difference in ethane
formation rate, shows up noticeably in this figure. Since the reaction rates proposed in this work
fits the experimental data better than the other two models (see Figures 4b, 5b and 6), the results
of simulation with the new model can be more trusted for the employed catalyst and operating
conditions of this simulation.
When the hydrogenation catalyst is used in an industrial reactor, it is gradually deactivated
until it reaches the point of inefficiency. At this point, it should be replaced with fresh catalyst.
However, up to this point the feed temperature is gradually being increased during the usage of the
catalyst to counter the effect of deactivation. This increase in the temperature can raise the
activity of the catalyst to some extent. Figure 8 demonstrates the effect of increasing the feed
temperature on exit acetylene concentration with catalysts of different activities for the simulation
parameters given in Table 3. The operating point of the fresh catalyst (acetylene concentration at

13
the reactor exit for the catalyst of the activity equal to unity) is also illustrated in the same figure.
It is obvious from this figure that in neither case, increasing the temperature can lead to an exit
acetylene concentration equal to that with a new catalyst being employed. In fact, the effect of
increasing the feed temperature on the performance of a reactor containing deactivated catalyst is
to decrease the acetylene concentration at the beginning, although such a concentration would not
reach the concentration equivalent to the fresh catalyst. Nevertheless, further increase in the feed
temperature even overturns this trend and results in decreasing the acetylene conversion in the
reactor. This is because of the reverse effect of temperature on the concentration of feed
components and the reaction rate. The higher the temperature, the higher will be the reaction rate
but at the same time the feed concentration would become lower as the temperature increases. It is
for this reason that increasing the feed temperature cannot be used as the only way of dealing with
deactivation of the catalyst during each run. Therefore, in practice, two reactors are used in series
in order to help the catalyst in the second reactor of the process to reach the desired concentration
of acetylene in the final product, in addition to increasing the feed temperature of the first reactor.
4.3. Catalyst Deactivation

As mentioned in the Experimental section, both active and deactivated catalysts were employed in
this study. The new catalyst was used to find the proper kinetic for the system and the deactivated
one was used to study the effect of using the catalyst for a long time and determine the
deactivation coefficient for the catalyst. This deactivation coefficient can be used for analyzing
the long term dynamic behavior of the acetylene hydrogenation unit and estimates the temperature
evolution of the feed to the reactor during the catalyst useful life.
0.04
0.038
a = 0.1
Exit Acetylene (kmole/s)
0.036
0.034
0.032
a = 0.3
0.03
a = 1.0
0.028
a = 0.5
0.026
a = 0.7
0.024
0.022
0.02
290
300
310
320
330
340
350
360
370
380
390
Feed Temperature (K)
Figure 8. Effect of temperature on altering the activity of deactivated catalyst.

The rate of reaction incorporating catalyst deactivation can be obtained as follows:
ri ,d = ri a (t )
(17)

14
from which the deactivation coefficient could be evaluated by using the experimental data of this
work. There is no explicit expression for the deactivation rate of this catalyst in the literature.
Therefore, although the future works might suggest a nonlinear relationship between the catalyst
deactivation rate and the fraction of active catalyst, a first-order deactivation rate is assumed in
this case:
da
= kd a
dt
(18)
or
a(t ) = e
kd t
(19)
The deactivated catalyst used in this study had been used in the corresponding industrial
process for six months for which the deactivation coefficient was found to be 0.25. Based on this
value, the deactivation rate constant is estimated to be
k d = 2.772 (0.067)
month
(20)
The figure in the parenthesis in Equation (20) is the standard deviation of the calculated
deactivation constant.
5. CONCLUSIONS
Selective hydrogenation of acetylene was studied in a fixed bed reactor of a commercial Pd/Al2O3.
Using the experimental data of this work and existing kinetic models from the literature, a new
kinetic expression for hydrogenation of acetylene was developed. The acetylene hydrogenation
reactor was simulated with different flow models (i.e., plug flow and dispersion flow models)
coupled with three different kinetic models (i.e., Bos et al., 1993; Menshchikov et al., 1975) and
the new model developed in this study). It has been shown that although the profiles along the
reactor length could be different, in most cases the differences between plug and dispersion flow
models are small in terms of reactor outlet quantities. The effect of deactivation of the catalyst
was studied experimentally with a used catalyst and the deactivation rate constant of the catalyst
was evaluated. It was demonstrated by simulation that it is necessary to employ two
hydrogenation reactors in series due to the following reasons:
(a) Practical temperature control of the reactor: Hydrogenation of the whole acetylene in the
feed would result in an unacceptable increase in the temperature of the outlet of the reactor if a
single reactor is to be employed.
(b) Reaching the desired exit concentration of acetylene: While the catalyst gets deactivated
over the time, it is not possible to overcome the deactivation of the catalyst only by increasing the
feed temperature, thus, a second reactor is needed to complete the process of acetylene
hydrogenation up to the desired exit acetylene concentration.

NOTATION
a
A
CA
Cp
DA
Ea
F
Fi
Hj
k
k0
kd
L
m
P
R
ri
ri,d
t
T
u
X
z
fraction of active catalyst

cross section area (m2)
concentration of component A (kmole/m3)
specific heat (J/kmole.K)
dispersion coefficient of component A (m2/s)
activation energy (J/kmole)
total molar flow rate of feed (kmole/s)
molar flow rate of species i (kmole/s)
heat of reaction of reaction j (J/kmole)
reaction rate constant
frequency factor
deactivation rate constant (s-1)
reactor length (m)
mass of catalyst (kg)
pressure (Pa)
gas consntant (J/kmole.K)
reaction rate of species i (kmole/kg cat.s)
reaction rate of species i for a deactivated catalyst (kmole/kg cat.s)
time (sec)
temperature (K)
superficial velocity (m/s)
conversion
distance along the reactor (m)
Greek Letters
bed voidage
catalyst density (kg/m3)
c
Subscripts
A
component A
d
deactivation
in
inlet
out
outlet
REFERENCES
Al-Ammar, A.S., Webb, G., Hydrogenation of acetylene over supported metal catalyst, Part I:
Adsorption of [14C] acetylene and [14C] ethylene on silica supported rhodium, iridium and
palladium and aluminum supported palladium, J. Chem. Soc. Faraday Trans. I, Vol. 74, 195-205
(1978).
Al-Ammar, A.S., Webb, G., Hydrogenation of acetylene over supported metal catalyst, Part II:
[14C] Tracer deactivation phenomena:, J. Chem. Soc. Faraday Trans. I, Vol. 74, 657-664 (1978).
Al-Ammar, A.S., Webb, G., Hydrogenation of acetylene over supported metal catalyst, Part III:
[14C] Tracer studied of the effects of the added ethylene and carbon monoxide on the reaction
catalyzed by silica-supported palladium, rhodium and indium, J. Chem. Soc. Faraday Trans. I,
Vol. 75, 1900-1911 (1979).

15
16
Anderson, J., Mc Alister, S.H., Derr, E.L., Peterson, W.H., Diolefines in alkylation feed stocks
conversion to monolefines by selective hydrogenation, Ind. Eng. Chem., Vol. 40, 2295-2301
(1948).
Bos, A.N., Bootsma, E.S, Foeth, F., Sleyster, H.W., Westertrep, K.R., A kinetic study of the
hydrogenation of ethyne and ethene on a commercial Pd/Al2O3 catalyst, Chem. Eng. Proc., Vol.
32, 53-63 (1993).
Brodzinski, A., Cybulski, A., The kinetic model of hydrogenation of acetylene-ethylene mixtures
over palladium surface covered by carbonaceous deposits, Appl. Catal. A, Vol. 198, 51-66
(2000).
Constantinides, A., Mostoufi, N., Applied Numerical Methods for Chemical Engineers with
MATLAB Applications, Prentice Hall PTR, Upper Saddle River, NJ (1999).
Fasi, A., Kiss, J.T., Torok, B., Palinco, I., The selectivity and activity determining role of
carbonaceous species and metal-metal oxide interface in metal-catalyzed hydrogenation and
isomerization reactions, Appl. Catal. A, Vol. 200, 189-200 (2000).
Fogler, H.S., Elements of Chemical Reaction Engineering, 3rd ed. Prentice Hall PTR, Upper
Saddle River, NJ (1999).
Godinez, C., Cabanes, A.L., Villora, G., Experimental study of the front-end selective
hydrogenation of steam-cracking C2-C3 mixture, Chem. Eng. Proc., Vol. 34, 459-468 (1995).
Haehn, P.C., Haidengger, E., Schdel, N., Revival of acetylenes within the petrochemical
industry, Hydrocarbon Eng., Vol. 1, 41-48 (1997).
Kummerow, I., Trimm, D.L., Wainwright, M.S., Studies of the catalytic hydrogenation of
acetylene impurities in a butadiene stream. Chemeca 83, Chem. Eng. Today, 103-109 (1983).
Larsson, M., Janson, J., Asplund, S., The role of coke in acetylene hydrogenation on Pd/PAl2O3, J. Catal., Vol. 178, 49-57 (1998).
Leviness, S., Nair, V., Weiss, A. H., Schay, Z. Guczi, L., Acetylene hydrogenation selectivity
control on PdCu/ Al2O3 catalysts, J. Molecular Catal., Vol. 25, 131-140 (1984).
Mc Gown, W.T., Kemball, C., Whan, D., Hydrogenation of acetylene in excess of ethylene on an
alumina supported palladium catalyst at atmospheric pressure in a spinning basket-reactor, J.
Catal., Vol. 51, 173-184 (1978).
Menshchikov, V.A., Falkovich, Y.G., Aerov, M.E., Hydrogenation kinetics of acetylene on a
palladium catalyst in the presence of ethylene, Kinet. Catal., Vol. 16, 1338-1355 (1975).
Muller, M., Deller, K., Volheim, G, Kuhn, W., Selective hydrogenation of acetylene in the
manufacturing of vinylchloride, Chem. Eng. Tech., Vol. 59, 645-647 (1987).
Srkny, A., Guczi, L., Weiss, A. H., On the aging phenomenon in palladium catalyzed acetylene
hydrogenation, Appl. Catal., Vol. 10, 369-388 (1984).

Szukiewicz, M., Kaczmarski, K., Petrus, R., Modeling of fixed-bed reactor: two models of
industrial reactor for selective hydrogenation of acetylene, Chem. Eng. Sci., Vol. 53, 149-155
(1998).
Taghavi, M.B., Pajonk, G., Teicher, S.J., Hydrognation slective en phase gazeuse de
lactylene en thylne en prsence de catalyseures au cuivre, dpos sur lalmine, de magnsie
et la silica, Bull. Soc. Chim. France, Vol. 7, 1302-1308 (1978).
Vincent, M.J., Gonzalez, R.D., A LangmuirHinshelwood model for a hydrogen transfer
mechanism in the selective hydrogenation of acetylene over a Pd/W-Al2O3 catalyst prepared by the
solgel method, Appl. Catal. A, Vol. 217, 143-156 (2001).
Vincent M.J., Gonzalez, R.D., Selective hydrogenation of acetylene through a short contact time
reactor, AIChE J., Vol. 48, 1257-1267 (2002).
Weisang, E., Engelhard, P.A., Nouveaux catalyseures dhydrogenation selective. Chim. Ind.
Genie. Chim., Vol. 103, 287-296 (1970).
Weiss, G., Modeling and control of an acetylene converter, J. Proc. Cont., Vol. 6, 7-15 (1996).
Westerterp, R., Bos, R., Wijngaarden, R., Kusters, W., Selective hydrogenation of acetylene in an
ethylene stream in an adiabatic reactor, Chem. Eng. Tech., Vol. 25, 529-539 (2002).

17
18

Table 2. Parameters of the kinetic models studied in this work

Parameter
Unit
Bos et al. (1993)
Menshchikov et al. (1975)
k1,0
kmole. kgcat -1.s -1. bar -2
33.39
77.09
k2,0
bar -1
5.1089
5052.8
k3,0
bar -1
3379.0
10.512
k4,0
kmole. kg cat -1.s -1. bar -2
10.1710-3
12.0310-5
k5,0
bar -1
44.635
11.509
k6,0
bar -1
0.0446
11.075
Ea,1
kJ/kmole
14638
-1220.46
Ea,2
kJ/kmole
-107.67
-5558.8
Ea,3
kJ/kmole
-3379.0
-3362.0
Ea,4
kJ/kmole
40354.0
28652.0
Ea,5
kJ/kmole
-33806.0
18425.0
Ea,6
kJ/kmole
-29400.0
4923.0
* The figures in the parentheses are standard deviation of calculated parameters.
This work*
77.09 (0.000353)
5052.8 (0.484)
10.512 (0.434)
4.0210-5 (5.1910-6)
9.4561 (0.0711)
10.5045 (0.0648)
-1220.46 (0.518)
-5558.8 (178.09)
-3362.0 (141.7)
39774.0 (0.076)
-12494.0 (261.5)
3326.6 (208.09)

Chemical Engineering Paper

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Chemical Engineering Paper

Hochgeladen von

Copyright:

Verfügbare Formate

I NTERNATIONAL J OURNAL OF C HEMICAL

Hydrogenation of Acetylene: Kinetic

University of Tehran, mostoufi@ut.ac.ir

University of Tehran, sotudeh@ut.ac.ir

Brought to you by | University of Saskatchewan

Hydrogenation of Acetylene: Kinetic Studies and

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

Brought to you by | University of Saskatchewan

International Journal of Chemical Reactor Engineering

Vol. 3 [2005], Article A14

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

Brought to you by | University of Saskatchewan

International Journal of Chemical Reactor Engineering

Vol. 3 [2005], Article A14

Bos et al. (1993)

Menshchikov et al. (1975)

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

Brought to you by | University of Saskatchewan

International Journal of Chemical Reactor Engineering

Vol. 3 [2005], Article A14

3.2. Apparatus and Procedure

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

4. RESULTS AND DISCUSSION

Brought to you by | University of Saskatchewan

International Journal of Chemical Reactor Engineering

Calculated Reaction Rates (mmol/kg cat.s)

Calculated Reaction Rates (mmol/kg cat.s)

Vol. 3 [2005], Article A14

Experimental ReactionRates (mmol/kg cat.s)

Experimental Reaction Rates (mmol/kg cat.s)

Calculated Reaction Rates (mmol/kg cat.s)

Experimental Reaction Rates (mmol/kg cat.s)

Experimental Reaction Rates (mmol/kg cat.s)

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

Brought to you by | University of Saskatchewan

International Journal of Chemical Reactor Engineering

Vol. 3 [2005], Article A14

Calculated Reaction Rates (mmol/kg cat.s)

Experimental Reaction Rates (mmol/kg cat.s)

4.2. Reactor Modeling

Table 3. Modeling conditions

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

Acetylene Conversion (%)

Reactor Length (m)

Reactor Length (m)

Ethane Formation (%)

Ethylene Formation (%)

Reactor Length (m)

Brought to you by | University of Saskatchewan

International Journal of Chemical Reactor Engineering

Vol. 3 [2005], Article A14

Brought to you by | University of Saskatchewan

Mostoufi et al.: Hydrogenation of Acetylene: Kinetic Studies and Reactor Modeling

4.3. Catalyst Deactivation

Exit Acetylene (kmole/s)

Feed Temperature (K)

Figure 8. Effect of temperature on altering the activity of deactivated catalyst.