Professional Documents
Culture Documents
Modelling and Simulation of 1,2-Dichloroethane Production by Ethylene Oxychlorination in #Uidized-Bed Reactor
Modelling and Simulation of 1,2-Dichloroethane Production by Ethylene Oxychlorination in #Uidized-Bed Reactor
Abstract
A comprehensive reactor model for ethylene oxychlorination for the production of 1,2-dichloroethane in a #uidized-bed reactor is
developed. The model is based on the two-phase theory of #uidization and allows for the change in volumetric gas #ow rate in the
dense phase due to the change in number of moles accompanying the reaction. The model predictions compared favorably with the
industrial data obtained from the literature. The e!ect of di!erent parameters on the behavior of the system is also investigated. It has
been found that the bubble diameter, ethylene molar feed fraction, residence time, and height at minimum #uidization have signi"cant
e!ects on the reactor performance. 2001 Elsevier Science Ltd. All rights reserved.
0009-2509/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 0 9 - 2 5 0 9 ( 0 0 ) 0 0 2 6 8 - 2
622 S. M. Al-Zahrani et al. / Chemical Engineering Science 56 (2001) 621}626
& Feathers, 1976). Frequently, this purge may be vented Coe$cient for 1 1 1
" #
to the atmosphere or, if incineration is required, the size mass transfer K K K
@C @ C
of the incineration plank and the amount of support fuel ; D
used are much less than for air-based oxychlorinations K "4.5 KD #10.4 C
@ d d
(Wimer & Feathers, 1976; Reich, 1976; Markelo!, 1984).
The purpose of this study is to develop a mathematical e D;
K "6.78 KD @
model to describe the ethylene oxychlorination reaction C d
in a #uidized-bed reactor. In the proposed study, the
Coe$cient for ; o C o k C
rigorous theory of the two-phase #ow was used to model H "4.5 KD E NE #10.4 E E NE
heat transfer @B d d
the #uidized-bed reactor. The e!ects of operating and
design parameters were studied. The parameters include Heat transfer k o (o !o )gd
h "0.88 E E Q E N
e!ect of feed mole fractions, residence time, bubble dia- coe$cient U d k
N
meter, particle diameter, feed temperature, cooling
medium temperature, pressure, height at minimum #uidiz- Binary 10\.¹
(1/M #1/M )
D " G H
di!usivity HG P (<
#<
)
ation (H ), velocity at minimum #uidization (; ) on the 2 G H
KD KD
reactor performance. The present model was validated Di!usivity in (1!> )
using a real plant data reported in the literature. D " H
HK
mixture >
H
G D
G$H HG
2. Model development
Of the many models proposed in the past, the two- experimental evidence reported by Toor and Calder-
phase model which considers the #uidized-bed reactor to bank (1968) for small-size particles and high #ow rate.
consist of a bubble phase and a surrounding dense phase 3. The dense phase is assumed to be perfectly mixed and
has proved to be the most suitable model, for incorporat- uniform in temperature.
ing recent "ndings on the hydrodynamics of #uidization 4. An average value of the bubble size and hence an
(Werther, 1980). The hydrodynamic and transport prop- average value of the interphase exchange parameter is
erty correlations are listed on Table 1. The details of the used for the whole bed. The e!ective bubble size is
model is given elsewhere (Aljodai, 1999). taken as that which exists at 40% of the expanded bed
height (Fryer & Potter, 1972). This assumption is
2.1. Assumptions widely used (Wagialla, Helal & Elnashaie, 1991; El-
nashaie, Wagialla & Helal, 1991; Adris, 1989; Abshar,
1. The bubble gas is devoid of solids and is in plug #ow. 1994).
2. The extent of reaction in the bubble-cloud phase 5. The mass and heat transfer resistances between the
is negligible. This assumption is justi"ed by the particles and the dense-phase gas are negligible.
S. M. Al-Zahrani et al. / Chemical Engineering Science 56 (2001) 621}626 623
6. The ideal gas law applies to the gas phase in both Energy balance
phases.
o C Q (¹ !¹ )!o C Q (¹ !¹ )
7. The volumetric #ow rate through the bubble phase is E NE BD D E NE B B
constant. #h A (¹ !¹ )#A ; o C (¹ !¹ )[1!e\@F]
U U U B @ @ E NE D B
#<(1!d)(1!e)(!*H )r"0,
2.2. Reaction kinetics P
where
Several kinetic studies have been reported in the litera-
a"K /; , b"H /; .o .C .
ture for the oxychlorination route for 1,2-dichloroethane @B @ @B @ E NE
production (Carrubba, 1970; Fengqiu, Yongrong, Shunni
& Gatang, 1994; Bakshi et al., 1991; Dmitrieva, Bakshi
& Gelbshtein, 1991; Wachi & Asai, 1994). 3. Results and discussion
In this work, kinetic study reported by Wachi and Asai
(1994) for the process of 1,2-dichloroethane formation Sensitivity analysis was carried out to assess the vari-
was used. This kinetic states that while the reaction ous design, operating as well as hydrodynamic para-
exhibited "rst-order kinetics with regard to the concen- meters on the reactor performance. The reactor design
tration of cupric chloride, the dependency on ethylene and operating conditions used in the simulations are
concentration was interpreted by a Langmuir}Hinshel- listed in Table 2. The heat capacities, di!usivity, the
wood mechanism, however, the reaction sequence prob- density of the gas, and heat of reaction are functions of
ably proceeds via chlorination of ethylene by cupric temperature.
chloride. HCl and oxygen then regenerate the copper
salt. The reaction rate has the following form: 3.1. Validation of model
dN N N oxygen conversion and also the EDC production rate. It
G@ "A (K ) GB ! G@ ,
dh @ @B G Q Q is clear from Fig. 1 that as the ethylene molar #ow rate is
B @
increased, the HCl and O conversions and the EDC
at h"0, N "N and N "N .
G@ G@D GB GBD production increase. At low ethylene feed mole fractions
Energy balance (i.e., less than 0.1947) the ethylene conversion is high and
this is because of the excess availability of HCl and
d¹ oxygen and the shortage of ethylene quantity. After the
Qo C @ "!A (H ) (¹ !¹ ),
@ E NE dh @ @B @ @ B ethylene feed ratio reaches 0.1947, the HCl and oxygen
at h"0, ¹ "¹ and ¹ "¹ .
@ D B D Table 2
Design and operating data used in solving the reactor model
2.4. Mass and energy balances on dense phase
Design and operating data Value
The steady-state dense-phase mass and energy bal-
ances are given below: Bed height at min. #uidization 7.0 m
Bed diameter 3.4 m
Mass balance
Catalyst solid density 3075 kg/m
N N Catalyst particle density 1369 kg/m
N "N #A ; GB ! G@D [1!e\?&] Catalyst particle diameter 80 lm
GB GBD @ @ Q Q
B @ Feed temp. 460 K
Cooling medium temp. 360 K
# <(1!d)(1!e)r.
624 S. M. Al-Zahrani et al. / Chemical Engineering Science 56 (2001) 621}626
Table 3
Comparison between model predictions and pilot plant data at
HCl/C H "2
Reactor Inlet Plant Model
Res. Time, s 25 25
Pressure, kPa 400 400
HCl/C H (Feed) 2 2
Reactor Outlet Error%
Outlet Temp., K 498 498.18 #0.04%
C H Conversion 99.3% 99.43% #0.13%
HCl Conversion 98.3% 98.55% #0.25% Fig. 2. The e!ect of residence time on C H conversion.
The results reveal that the ethylene feed composition o Density of the gas, kg/m
E
has a very profound e!ect on the system performance. The o Density of solids, kg/m
Q
parametric sensitivity analysis has shown that the system
performance is particularly sensitive to residence time,
bubble diameter and height at minimum #uidization. References
The optimum ethylene conversion (i.e., 98%) and EDC
yield of more than 98% at 98% HCl conversion can be Abshar, M. E. (1994). Bifurcation, Stability and Chaos in yuidized bed
achieved by working at a residence time of 25 s, height at catalytic reactors. Ph.D. thesis, Univ. of Salford, Salford, UK.
Adris, A. M. (1989). Steam reforming of methane, MS. thesis., Univ. of
minimum #uidization of 7 m and bubble diameter of Salford, Salford, UK.
4.5 cm. Aljodai, A. (1999). Modelling and bifurcation of 1,2 dichloroethane pro-
duction by ethylene oxychlorination in yuidized bed reactor. MS.
Notations thesis. King Saud University, Saudi Arabia.
A Cross sectional area of the bubble phase, m Bakshi, Y. M., Gelbshtein, A. I., Gelperin, E. I., Dmitrieva, M. P.,
@ Zyshin, A. G., & Snagovshii, Y. S. (1991). Study of the mechanism and
d Bubble diameter, m kinetics of additive oxychlorination of olexns'. V. Kinetic model of the
C Concentration of cupric chloride in the cata-
! reaction. Translated from Kinetica Kataliz, 32, 663}672.
lyst Carrubba, R. V. (1970). Kinetics of the oxychlorination of ethylene.
C Concentration of ethylene in gas phase, Industrial and Engineering Chemistry, Process Design and Develop-
# ment, 9, 414}419.
mol/m
Cavaterra, E. (1988). Catalysts to make Dichloroethane. Hydrocarbon
C Molar heat capacity of gas, kJ/kg mol K
NE Processing, 67, 63}67.
C Molar heat capacity of solid, kJ/kg mol K
NQ Dmitrieva, M. P., Bakshi, Y. M., & Gelbshtein, A. I. (1991). Investiga-
h Distance along bed height, m tion of mechanism and kinetics of additive oxychlorination of ole-
H Total bed height, m "ns. IV. In#uence of copper containing salt catalyst with c-Al O .
H Height at minimum #uidization, m Support on pinetic parameters of individual stages of reaction.
KD Translated from Kneticka Kataliz, 32, 74}80.
h Bed to cooling surface heat transfer coe$c-
U Elnashaie, S. S., Wagialla, K. M., & Helal, A. M. (1991). The use of
ient, kJ/m s K mathematical and computer models to explore the applicability of
H Interphase heat transfer coe$cient between #uidized bed technology for highly exothermic catalytic reactions.
@B II. Oxidative dehydrogenation of ethyl benzene to styrene. Math-
bubble and dense phase based on bubble
phase volume, kJ/m s K ematical and Computer Modeling, 15, 43}54.
Fengqiu, C., Yongrong, Y., Shunni, R., & Gatang, C. (1994). Studies on
K Interphase mass transfer coe$cient between
@B ethylene oxychlorination, II. Reaction mechanism and kinetics.
bubble and dense phase based on bubble Petrochemical Technology, 23, 421}425.
phase volume, 1/s Fryer, E., & Potter, Q. E. (1972). Bubble size variation in two-phase
k Chemical reaction rate constant, 1/s models of #uidized bed reactors. Powder technology, 6, 312}322.
P Keane, D. P., Stobaugh, R. B., & Townsend, P. L. (1973). Vinyl chloride.
K Adsorption equilibrium constant, m/mol
? Hydrocarbon Processing, 52, 99}104.
N Molar #ow rate of component i in bubble
G@ Markelo!, R. G. (1984). Convert oxychlorination from air to oxygen.
phase, kg mol/s Hydrocarbon Processing, 63, 91}94.
N Inlet molar #ow rate of component j in Nawroski, J. S., & Velez, E. S. (1983). Oxychlorination of ethylene. Applied
G@D Industrial Catalysis, 1 (pp. 239}273). New York: Academic Press.
bubble phase, kg mol/s
N Molar #ow rate of component i in dense Reich, P. (1976). Air or oxygen for VCM. Hydrocarbon Processing, 55,
GB 85}89.
phase, kg mol/s Smallwood, P. V., Stephenson, M. W., Newman, M. W., & Bunten, I. J.
N Inlet molar #ow rate of component i in dense
GBD (1987). Vinyl chloride polymers (pp. 241}262). New York: Wiley.
phase, kg mol/s SRI Report. (1998). Process Economic Program, Stanford Research
Q Volumetric #ow in bubble phase, m/s Institute, Menlo Park, CA. Ethylene and derivatives, Vol. 1.
@ Toor, F. D., & Calderbank, P. H. (1968). Proceedings of the tripartite
Q Volumetric #ow in dense phase, m/s
B chemical engineering conference. Montreal, Que.: London Institu-
r Rate for reaction, kg mol/s m tion of Chemical Engineers.
R Gas constant, kJ/kg mol K Ullmann's Encyclopedia of Industrial Chemistry, (1986). Vol. 6, (5th
¹ Bubble phase temperature, K ed.). (pp. 263}271). New York: VCH.
@ Wachi, W., & Asai, Y. (1994). Kinetics of 1,2 Dichloroethane formation
¹ Dense phase temperature, K
B from ethylene and cupric chloride. Industrial Engineering and Chem-
¹ Feed gas temperature, K
D istry Research, 33, 259}264.
¹ Cooling medium temperature, K
U Wagialla, K. M., Helal, A. M., & Elnashaie, S. S. (1991). The use of
*H Heat of reaction, kJ/kg mol
P mathematical and computer models to explore the applicability of
; Bubble velocity, m/s #uidized bed technology for highly exothermic catalytic reactions. I.
@ Oxidative dehydrogenation of butene. Mathematical and Computer
; Minimum #uidization velocity, m/s
KD Modeling, 15, 17}31.
Werther, J. (1980). Mathematical Modelling of #uidized bed reactors.
Greek letters International Chemical Engineering, 20, 4.
d Fraction of bed consisting of bubbles Wimer, W. E., & Feathers, R. E. (1976). Oxygen gives low cost VCM.
e Void fraction Hydrocarbon Processing, 55, 81}84.