Chem. Prod. Process Model.

2020; ▪▪▪(▪▪▪): 20200061

Z. Moradi and M. Farsi*

Simulation of direct chlorination of ethylene

in a two-phase reactor by coupling
equilibrium, kinetic and population balance
models
https://doi.org/10.1515/cppm-2020-0061
Received June 16, 2020; accepted July 10, 2020; published online September 1, 2020

Abstract: The purpose of presented research is the mathematical simulation and sensitivity analysis of
ethylene dichloride synthesis (EDC) through direct chlorination of ethylene in a bubble column reactor at
steady state condition. In the first step, the reactor is heterogeneously simulated based on the energy and mass
balance equations by coupling the mass and energy, kinetic, equilibrium, and population balance models. In
the considered process, the gaseous ethylene and chlorine are dispersed and dissolved in the liquid medium
and converted to EDC at the presence of a homogeneous catalyst. The population balance model is applied to
calculate the heat and mass transfer area along the reactor. To investigate the accuracy of established model,
the results of simulation are compared with the plant data. It is confirmed that temperature, pressure, rate of
mass transfer, breakage, and coalescence phenomena change the bubble diameter and distribution in the
chlorination reactor. In the second step, the effects of operating pressure and temperature on the EDC pro-
duction rate are investigated by the developed model. In the third step, considering EDC production rate as the
cost function the optimal operating temperature of reactor is developed at steady state condition. Based on the
obtained results, the optimal operating temperature is 357 K and EDC production at the optimal condition is
23.79 mol s−1.

Keywords: bubble column reactor; direct chlorination of ethylene; equilibrium model; population balance
model.

1 Introduction
The ethylene dichloride is an organochloride intermediate chemical that used to produce vinyl chloride
monomer as the key chemical precursor to polyvinylchloride. Ethylene dichloride is also used as a scavenger
for lead in gasoline [1]. Industrially, the direct chlorination and oxychlorination of ethylene are two common
catalytic routes to produce ethylene dichloride [2]. In the direct chlorination process, the chlorine and ethylene
are dissolved in the liquid medium and react in the presence of a homogeneous catalyst. The direct chlori-
nation reaction can be carried out in the temperature range 20–70 °C as the low-temperature process or 100–
150 °C as the high-temperature. In the oxychlorination process, the gaseous ethylene, hydrogen chloride, and
oxygen react over the heterogenous cupric chloride catalyst in a fluidized bed reactor at the temperature range
200–300 °C [3]. In the commercial vinyl chloride production units, the oxychlorination and direct chlorination
processes are used in parallel. In the balanced process, the produced hydrogen chloride in the direct chlori-
nation of ethylene and EDC cracking units is used in the oxychlorination unit to produce EDC.

*Corresponding author: M. Farsi, Department of Chemical Engineering, School of Chemical and Petroleum Engineering, Shiraz
University, Shiraz, Islamic Republic of Iran, E-mail: [email protected]
Z. Moradi: Department of Chemical Engineering, School of Chemical and Petroleum Engineering, Shiraz University, Shiraz, Islamic
Republic of Iran
2 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene

Although the continuous stirred tank and bubble column reactors are commonly applied in the industrial
plants to produce EDC through direct chlorination of ethylene, application of bubble column reactors is more
popular due to excellent heat and mass transfer characteristics, efficient contact between the phases, easy and
accurate control of temperature, and low maintenance cost. However, back-mixing, short gas phase residence
time, and high pressure drop are the main disadvantage of bubble column reactors [4, 5]. Wachi and Morikawa
investigated the direct chlorination of ethylene and dichloroethane in a continuous stirred tank reactor with a
flat free gas–liquid interface [6]. They proposed reaction kinetics for direct chlorination reaction based on the
simultaneous absorption and reaction. Wachi and Morikawa experimented the direct chlorination of ethylene
in a pilot scale bubble column reactor at high temperature [7]. The temperature at the top of column was
controlled at 130 °C, which was almost equivalent to the boiling temperature of 1,2-dichloroethane. The results
showed that decreasing liquid velocity and applying excess ethylene to the reactor could improve chlorine
conversion in the considered process. When the liquid velocity exceeded 1.9 m s−1, the dichloroethane evap-
oration was depressed in the middle section of reactor. Orejas modeled the direct chlorination of ethylene at
low temperature condition in an industrial bubble column reactor by coupling equilibrium and kinetic models
[8]. The model was capable to predict the product specifications with acceptable accuracy. They claimed that
chlorine is more soluble in the liquid medium compared to ethylene. Abshhar investigated the ethylene
dichloride production in a gas-lift reactor [9]. They simulated the applied reactor based on the backflow cell
model. The simulation results suggested that the excess feeding of ethylene improves conversion of chlorine.
In this research, the direct chlorination of ethylene is mathematically simulated through the integration of
solubility, kinetic, and population balance models. Then, the effects of operating temperature and pressure on
EDC production rate is investigated. In the first section, the published researches on the modeling and
simulation of EDC synthesis through direct chlorination process are reviewed. The direct chlorination reactor is
described in section 2. The reaction kinetic model, the developed mathematical model, and applied numerical
solution are presented in sections 3 and 4, respectively. Ultimately, the simulation results are analyzed in
section 5 followed by presenting the remarkable results in section 6.

2 Process description
Figure 1 shows the structure of applied bubble column reactor in the EDC unit. The considered bubble column
reactor consists of two main parts including cooling and conversion sections. Gaseous ethylene and chlorine
dispersed in the conversion section through the considered distributor at the bottom. These reactants are
dissolved in the liquid medium and converted to EDC, ETC at the presence of homogenous ferric chloride
catalyst. The density difference between conversion and cooling sections allows the natural recirculation of
liquid medium in the reactor. The outlet hot liquid from the conversion section enters the cooling zone and
cooled down [8]. The cooling zone is a coil tube exchanger that the cooling water is circulated in the tubes and
produced heat in the reaction zone is removed from the system. The liquid product and gaseous effluent flow
out from the bottom and top of the reactor, respectively. Table 1 shows the feed specification and reactor design
data.

3 Process modeling
3.1 Kinetic model

As mentioned, the gaseous chlorine and ethylene are dissolved in the ethylene dichloride, and the reaction
takes place in the liquid phase at the presence of homogenous ferric chloride catalyst [10]. The considerable
side reaction in the EDC production process is the chlorination of EDC to ETC. The mechanism of main and side
reactions are as [11]:
 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene 3

Table : Feed specification and design data of direct chlorination reactor.

Parameter Value

Reactor Length of conversion zone (m) 

Diameter (m) .
Distributer diameter (mm) 
Feed Pressure (bar) .
Gas temperature (K) 
Liquid temperature (K) 
EDC flow rate in the inlet liquid (mol hr−) .
ETC flow rate in the inlet liquid (mol hr−) .
Oxygen flow rate in the inlet gas (mol hr−) .
EDC flow rate in the inlet gas (mol hr−) .
Chlorine flow rate in the inlet gas (mol hr−) .
Ethylene flow rate in the inlet gas (mol hr−) .

C2 H4 + Cl2 → C2 H4 Cl2 (1)

C2 H4 Cl2 + Cl2 → C2 H3 Cl3 + HCl (2)

3.2 Solubility model

Since the reactants at gaseous conditions are dissolved in the liquid and then converted to product, the φ − φ
approach is applied to calculate the equilibrium distribution of ethylene, chlorine, hydrogen chloride,
hydrogen sulfide EDC, ETC, and oxygen in the gas and liquid phases. In this approach, the fugacity of each
component is expressed as:
fiv  yi Pφvi (3)

fil  xi Pφli (4)

In general, the fugacity of each component in gas and liquid phases are equal at the phase equilibria condition
as:

yi Pφvi  xi P ∗i φli (5)

The equation (8) could be rearranged as:

yi P ∗i   φli
K e, i   (6)
xi P   φvi

The fugacity of each component is determined by the Soave–Redlich–Kowng equation of state that can
accurately describe both the liquid and vapor phases. The Soave–Redlich–Kwong equation of state is as [12]:
RT aα
P − (7)
V − b V(V + b)′

Where the parameters of equation are:

R2 T 2c
a  0.42748 (8)
Pc

RT c
b  0.08664 (9)
Pc
4 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene

0.5 2
⎛1 −  T  ⎠
⎛1 + 0.480 + 1.57ω − 0.17ω2 ⎝
α⎝ ⎞⎠
⎞ (10)
Tc

Based on the considered equation of state, the fugacity coefficient is explained as:
a Z+B
lnφ  Z − 1 − ln(Z − B) − ln  (11)
bRT B

The compressibility factor is:

Z 3 − Z 2 + A − B − B2 Z − AB  0 (12)

3.3 Reactor modeling

In general, the process simulation enables analysis, troubleshooting, and optimization of systems. The
theoretical modeling based on the mass and energy governing equations is one of the powerful procedures to
simulate chemical processes. However, the solution of complex and detailed theoretical models is time-
consuming, and adopting some assumptions could decrease problem difficulty. In this research, the below
assumptions are considered to simplify the developed model as:
– The flow regime of gas and liquid phases behaves as plug.
– The mass and heat diffusion in the axial direction is negligible compared to convection.
– The reactor is completely adiabatic.
– The reaction occurs in the liquid phase.

Based on the adopted assumptions and considering temperature and concentration gradient between phases,
the mass balance equation in the gas and liquid phases are:
F g dxi Nr
+ K G, i av k i xi − yi  − 1 − εg  ∑ ηj ri, j  0 (13)
Ac dz j1

F l dyi
− K G, i av k i xi − yi   0 (14)
Ac dz
where K G, i is the gas phase overall mass transfer coefficient and estimated based on the mass transfer coef-
ficient in the gas, liquid, and equilibrium constant. Based on the temperature gradient between phases, the
energy balance equation in the gas and liquid phases are:
C gp F g
− UT l − T g av  0 (15)
Ac

Clp F l dT l Nr
− UT g − T l av − 1 − εg  ∑ ηj ri, j (− Δ H)  0 (16)
Ac dz j1

The pressure drop in the reactor can be calculated based on the hydrostatic head as:
dP
 −g
1 − εg ρl + ρg εg  (17)
dz
In general, the volumetric flow rate of gas stream, heat and mass transfer area, and gas holdup change along
the bubble column reactor due to variation of temperature, hydrostatic pressure, and mass transfer between
gas and liquid phases. In this regard, the volumetric flow rate of gas stream is calculated based on the equation
of state considering the compressibility factor as:
 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene 5

Table : Used correlations for physical properties, mass and heat transfer coefficients.

Parameter Equation

Heat capacity of pure component [] C p ¼ C  þ C  T þ C  T  þ C  T 

Mixture heat capacity C pm ¼ ∑y i   C pi
C
μl ¼ eC  þ T þC  lnT þC  T
C
Liquid viscosity of pure component []
C
Gas viscosity of pure component [] μg ¼ C CT C
þ þ
T T

∑y i μi Mw : ​
The viscosity of mixture μm ¼ i
∑y i Mw :
​
i

CC T 
C 

C
Thermal conductivity in the gas phase [] k¼  
þ þ
T T

Thermal conductivity in the liquid phase [] kl ¼ C þ C T þ C T 

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
T : =Mi þ=Mj
The binary diffusion coefficient in the gas phase [] Dij ¼ : = = 
Pðυci þυcj Þ
pffiffiffiffiffiffiffi
T ϕ Mj
The binary diffusion coefficient in the liquid phase [] Dij ¼ : ×  ϑ:  μJ
! c; i i
Mixture diffusion mass transfer coefficient Dim ¼   ∑ Dyiji
 
i≠j :  :
D μ ρug dvs
Mass transfer coefficient gas-liquid phases [] k L ¼ : di;vsm ρDi; m μ

Mass transfer coefficient in the bubble [] k g ¼ : × R :

e S:
c ug
 
Overall mass transfer coefficient K y ¼ kg þ kmL

: : :

Gas holdup [] εg ¼ u:
g ρg ρl σ þ :
 =
Mean diameter of bubbles at the entrance of column DB ¼ gðρσD o
ρ Þ L G

Heat transfer coefficient [] h ¼ :R: : k

e Pr D
 n
 
Overall heat transfer coefficient U ¼ hg þ hl

dQg F g dT g T g dF g F g T g dP
 ZR   +   − 2  (18)
dx P dx P dx P dx
The supplementary equations to calculate the physical, chemical, and thermal properties and transfer co-
efficients are presented in Table 2.

3.4 Population balance model

In general changing operating pressure and temperature, mass transfer, breakage, and coalescence change the
bubble volume and transfer area in the system. Different theories have been developed to investigate the
hydrodynamic of a dispersed phase in a continuous phase such as Eulerian, Lagrange, and population balance
techniques [13]. By neglecting the bubble breakage due to low applied shear rate on bubbles, the changes in
bubble size distributions can be explained by the steady state form of population balance equations as:
∞
∂n(v, z) 1 v
u  ∫ β(v − w, w)n(w, z)n(v − w, z)dw − n(v, z) ∫ β(v, w)n(w, z)dw (19)
dz 2 w0 w0
6 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene

The first and second terms represent the birth of large bubbles due to the coalescing of small bubbles and the
death of those small bubbles, respectively. The collision frequency,  β, is explained as [14]:
2 2
2 3 2
⎝di ⎞
βdi , dj   180⎛ ⎠⎝⎛di ⎞⎝1 − di ⎞
⎠⎛ ⎠ eij (20)
3 3 3
dj dj dj

Manga and Stone suggested an empirical correlation for aggregation efficiency as [15].
1
eij  0.3δ1/2 + Bi δ6 (21)
2
The balance equations could be discretized into a set of ordinary differential equations in the axial direction as:
ΔN(lk ) 1 N N N
uk  ∑ ∑ βli , lj  ⋅ N(li ) ⋅ Nlj  − ∑ βlk , lj  ⋅ N(lk ) ⋅ Nlj  (22)
Δz 2 i1 j1 j1

In general, the analytical solution of population balance equations is difficult and some numerical methods
are proposed to solve the developed equations [16]. In this research, the class method as a popular and accurate
technique is considered to solve the balance equations [17]. The number of considered class and bubble
diameter range are 110 in the range 4–8 mm, respectively.

4 Model solution
In general, the developed model comprises a set of nonlinear differential and algebraic equations that should
be solved numerically considering an iterative procedure. The considered explicit procedure to solve the
developed model is as:
– Solve the solubility model and calculate the equilibrium constants based on the previous element data.
– Solve the population balance equations based on the previous element data and calculate the heat and
mass transfer area.
– Solve the mass and energy governing equations in the gas and liquid phase by Runge–Kutta four order
method.
– Resolve and update the solubility and population balance models based on the current element data.
– If the solubility and population balance results don’t change, go the next element.

It is confirmed that the considered reactor is simulated based on the presented data in Table 1 as the initial
condition of differential equations. The developed equations are programmed in MATLAB R2018b and performed
on a high-speed computer with an Intel Core i7 7700HQ processor running at 3.8 GHz using 16 GB of RAM.

5 Results and discussion

5.1 Process simulation

In this part, the simulation results of direct chlorination process are presented at steady state condition. Since
the developed model is used in the sensitivity analysis and optimization stages, proving the accuracy of model
Table : Comparison of simulation results and industrial data.

Simulation Plant data Absolute relative error (%)

Ethylene conversion (%) . . .

Chlorine conversion (%) . . .
EDC production (kg s−) . . .
ETC production (kg s−) . . .
 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene 7

Figure 1: Diagram of bubble column reactor.

is critical and shows the validity of simplified assumptions. Table 3 represents the comparison between
simulation results and plant data at the same operating condition. It appears that the developed model is
capable to predict the performance of direct chlorination process and the maximum relative error is 0.29%.
To investigate the system performance, the molar flow rate, and mole fraction of components, the tem-
perature of liquid and gas phases, bubble diameter and distribution are presented at steady state condition.
Figure 2 shows the chlorine molar flow rate in both gas and liquid phases along the bubble column reactor. It
appears that chlorine dissolves in the liquid and therefore chlorine flow rate decreases in the gas phase
continuously. In the liquid medium, the rate of chlorine transfer from gas to the liquid is dominant over the rare
of chlorine consumption, and chlorine concentration increases at the entrance of reactor. Decreasing chlorine
concentration in the gas phase decreases the rate of chlorine transfer toward the liquid and the chlorine
concentration decreases in the liquid medium due to a high rate of direct chlorination reaction. Based on the
results of simulation the total chlorine conversion in the reactor is 99.6% and chlorine content in the gas phase
approaches zero due to the high solubility of chlorine in the liquid.
Figure 3 shows the molar flow rate of ethylene dichloride in the gas and liquid phases along the reactor. It
appears that EDC content in the gas and liquid streams have maximum and minimum points, respectively.
Since the inlet gas stream is EDC lean and circulated liquid is EDC rich, a part of EDC in the liquid medium is
evaporated and transferred from liquid medium toward the gas at the entrance of reaction zone. Thus, EDC
content in the gas stream increases and approaches a maximum concentration. After that ethylene and
chlorine transfer from the gas phase toward the liquid increases EDC concentration in the gas stream.
Increasing EDC concentration in the gas phase changes the mass transfer direction and a part of EDC is
condensed. In the liquid medium, the evaporation of EDC is dominant over the EDC production through
chlorination reaction at the entrance of reactor, and EDC content decreases in the liquid medium. After a
minimum point, EDC content increases in the liquid phase due to the chlorination reaction and condensation
of EDC. Based on the simulation results the EDC flow rate in the liquid stream increases from 76.8 to
101.08 mol s−1 and it decreases from 1 to 0.51 mol s−1 in the gas phase. In general, the rate of EDC production in
the chlorination reactor is 23.79 mol s−1.
Figure 4 shows the ethylene trichloride molar flow rate as the main side-product in the direct chlorination
process. It appears that ETC concentration increases continuously in the liquid phase. Since ETC is more
8 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene

Figure 2: Chlorine molar flow rate in the gas and liquid

phases.

Figure 3: Ethylene dichloride molar flow rate in the gas and

liquid phases.

soluble in the liquid phase compared to the other components, the rate of ETC production is dominant over ETC
evaporation, and ETC content increases along the liquid medium. However, evaporation phenomena increases
ETC content immediately in the gas stream, and after a maximum point EDC concentration decreases due to
change in the mass transfer direction imposed by low ethylene and chlorine concentration in the gas stream.
Based on the simulation results the ETC flow rate in the liquid stream increases from 0.015 to 0.077 mol s−1 and
EDC content in the gas phase is negligible. In general, the rate of ethylene trichloride production and EDC
selectivity are about 0.062 mol s−1 and 99.7%, respectively.
Figure 5 shows the hydrogen chloride molar flow rate in the chlorination process. It appears that ETC
concentration increases continuously in the gas phase. In general, HCL is produced through chlorination of
ethylene dichloride in the liquid phase. Based on the presented results, the HCl molar flow rate increases in the
liquid phase and reaches a maximum point and then decreases continuously. It appears a part of produced
HCL in the liquid is evaporated and transferred to the gas phase at the entrance of reactor and HCL flow rate
increases in the gas and liquid phases. In the second part, decreasing chlorine concentration in the liquid
 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene 9

Figure 4: ETC molar flow rate in the gas and liquid phases.

Figure 5: HCL molar flow rate in the gas and liquid phases.

medium decreases the rate of HCL production. Thus, the low solubility of HCL in the liquid and low rate of EDC
chlorination decrease HCL content in the liquid phase in the second part of reactor. Based on the simulation
results, the rate of HCL production is about 0.062 mol s−1 that has been distributed between both phases.
Figure 6 shows the temperature profile in the liquid and gas phases. Since ethylene chlorination is an
exothermic reaction that occurs in the liquid medium, the temperature of liquid increases continuously along
the reactor. The inlet gas stream has a lower temperature compared to the liquid and decreases the rate of
temperature rise in the system. In general, a part of produced heat in the liquid phase is transferred to the gas
stream, and gas phase temperature increases along the reactor. Due to the direct contact heat transfer
mechanism, both gas and liquid phases approach to the equilibrium temperature of 342 K.
10 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene

Figure 6: Liquid and gas phase temperature

profiles in the gas and liquid phases.

Figure 7: Bubble size distribution along the reactor.

Figure 7 shows the bubble size distribution along the bubble column reactor. In general, temperature,
pressure, rate of mass transfer, breakage and coalescence phenomena change the bubble diameter and
distribution in the chlorination reactor. Since the rate of outlet mass from the gas phase is dominant over the
rate of inlet mass, it decreases bubble volume, while increasing temperature and decreasing pressure tend to
increase bubble volume. It appears that the size distribution of bubbles moves toward the smaller ones along
the reactor. Based on the simulation results, the mean diameter of inlet and outlet bubbles are 6.2 and 2.2 mm.

5.2 Sensitivity analysis

In this part, the effects of operating pressure and temperature on the EDC production rate are presented.
Figure 8a shows the effect of operating pressure on the EDC production rate in the chlorination process at the
pressure range of 2.6–3.3 bar. In general, pressure has an insignificant effect on the specification of liquid
 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene 11

(a)

Figure 8(a): The effect of operating pressure on the EDC

production rate.

(b)

Figure 8(b): The effect of operating temperature on the

EDC production rate.

phase. However, increasing operating pressure improves the solubility of chlorine and ethylene in the liquid
phase and could increase the reaction rate indirectly. On the other hand, it decreases the rate of EDC evapo-
ration in the system. Based on the simulation results, increasing operating pressure in the considered range
increases the rate of EDC production.
Figure 8b shows the effect of liquid temperature on the EDC production rate in the chlorination process at the
temperature range 335–370 K. Although increasing operating temperature enhances the rate of chlorination
reaction in the liquid phase, it decreases the solubility of chlorine and ethylene in the liquid medium. On the other
hand, it enhances the rate of EDC evaporation. Based on the simulation results, there is an optimum operating
temperature to improve the rate of EDC production. In general, increasing operating temperature from 330 to 357 K
increases EDC production rate, and then applying higher temperature decreases the EDC synthesis. Based on the
obtained results, the optimal operating temperature is 357 K and EDC production, selectivity, and chlorine con-
version at the optimal condition are 23.79 mol s−1, 99.8%, and 99.4%, respectively.
12 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene

6 Conclusion
In this research, the direct chlorination of ethylene in an industrial bubble column reactor was modeled and
simulated at steady state condition. The chlorination reactor was heterogeneously simulated based on the
mass and energy conservation laws by coupling kinetic, equilibrium, and population balance models. The
comparison between plant data and simulation results proved the accuracy of developed model and applied
assumptions. It appeared that the mean diameter of outlet bubbles decreases from 6.2 to 2.2 mm along the
reactor. The sensitivity analysis results showed that increasing pressure improved the solubility of chlorine
and ethylene in the liquid phase and increased the reaction rate indirectly. Based on the obtained results, the
optimal operating temperature was 357 K and EDC production, selectivity, and chlorine conversion at the
optimal condition were 23.79 mol s−1, 99.8%, and 99.4%, respectively.

Nomenclature
A Cross-section of the reactor (m2)
a The specific surface of the bubble (m2 m−3)
Cp Specific heat capacity (J mol−1 K−1)
D Bubble diameter (mm)
DiJ Liquid penetration coefficient (m2 s−1)
Dim Overall penetration coefficient (m2 s−1)
F Molar flow rate (mol s−1)
h Heat transfer coefficient (m s−1)
Ky,i Overall mass transfer coefficient (m s−1)
kg Mass transfer coefficient in the bubble (m s−1)
kl Mass transfer coefficient in liquid (m s−1)
ki Conductivity of component i (J m−1 s−1 K−1)
n Bubble number
P Total pressure (bar)
R Universal gas constant (J mol−1 K−1)
ri The reaction rate of component i (mol m−3 s−1)
T Temperature (K)
u The superficial velocity (m s−1)
U Heat transfer coefficient (W m−2 k−1)
xi Mole fraction of component i in the liquid phase (-)
yi Mole fraction of component i in the gas phase (-)
Z Compressibility factor
z Reactor length (m)

Greek letters
ρ Density (kg m−3)
μi Liquid viscosity (Pa s)
ΔH Heat of reaction for initiation and propagation reactions (J mol−1)
ε Gas holdup
β Collison frequency
σ Surface tension (N m−1)

Subscripts
g Gas phase
L Liquid phase
i Chemical species
 Z. Moradi and M. Farsi: Simulation of direct chlorination of ethylene 13

Author contribution: All the authors have accepted responsibility for the entire content of this submitted
manuscript and approved submission.
Research funding: None declared.
Conflict of interest statement: The authors declare no conflicts of interest regarding this article.

References
1. Kumar T, Mohsin R, Ghafir MFA, Kumar I, Wash AM. Concerns over use of leaded aviation gasoline (AVGAS) fuel. Chem Eng Trans
2018;63:181–6.
2. Khademi M, Angooraj Taghavi S. Optimization of ethylene oxychlorination fluidized-bed reactor using differential evolution
(DE) method. Sci Iran 2017;24:1253–63.
3. Montebelli A, Tronconi E, Orsenigo C, Ballarini N. Kinetic and modeling study of the ethylene oxychlorination to
1,2-dichloroethane in fluidized-bed reactors. Ind Eng Chem Res 2015;54:9513–24.
4. Jana SK, Das SK. Tapered bubble column using pseudoplastic non-Newtonian liquids–empirical correlation for pressure drop.
Chem Chem Technol 2017;11:327–32.
5. Kantarci N, Borak F, Ulgen KO. Bubble column reactors. Process Biochem 2005;40:2263–83.
6. Wachi S, Morikawa H. Liquid-phase chlorination of ethylene and 1,2-dichloroethane. J Chem Eng Jpn 1986;19:437–43.
7. Wachi S, Morikawa H. Chlorination of ethylene in a boiling bubble column reactor. J Chem Eng Jpn 1987;20:238–45.
8. Orejas JnA. Modelling and simulation of a bubble-column reactor with external loop: application to the direct chlorination of
ethylene. Chem Eng Sci 1999;54:5299–309.
9. Abashar M. Ethylene dichloride production in external-loop gaslift reactors. JKSUES 2004;16:179–201.
10. Kurta S, Mykytyn I, Khatsevich O, Ribun V. Mechanism of catalytic additive chlorination of ethylene to 1,2-dichloroethane.
Theor Exp Chem 2018;54:283–91.
11. Orejas JnA. Model evaluation for an industrial process of direct chlorination of ethylene in a bubble-column reactor with
external recirculation loop. Chem Eng Sci 2001;56:513–22.
12. Halder G. Introduction to chemical engineering thermodynamics. PHI Learning Pvt. Ltd; 2014.
13. Selma B, Bannari R, Proulx P. Simulation of bubbly flows: comparison between direct quadrature method of moments
(DQMOM) and method of classes (CM). Chem Eng Sci 2010;65:1925–41.
14. Laakkonen M, Alopaeus V, Aittamaa J. Validation of bubble breakage, coalescence and mass transfer models for gas–liquid
dispersion in agitated vessel. Chem Eng Sci 2006;61:218–28.
15. Aryafard E, Farsi M, Rahimpour M. Modeling and simulation of crude oil desalting in an industrial plant considering mixing
valve and electrostatic drum. Chem Eng Process Process Intensific 2015;95:383–9.
16. Kakhki NA, Farsi M, Rahimpour M. Effect of current frequency on crude oil dehydration in an industrial electrostatic coalescer. J
Taiwan Inst Chem E 2016;67:1–10.
17. Kumar S, Ramkrishna D. On the solution of population balance equations by discretization—I. A fixed pivot technique. Chem
Eng Sci 1996;51:1311–32.
18. Perry JH. Chemical engineers’ handbook. ACS Publications; 1950.
19. Poling BE, Prausnitz JM, O’connell JP The properties of gases and liquids. New York: Mcgraw-hill; 2001, vol. 5.
20. Reid RC, Prausnitz JM, Poling BE. The properties of gases and liquids; 1987.
21. Fukushima S, Kusaka K. Liquid-phase volumetric and mass-transfer coefficient, and boundary of hydrodynamic flow region in
packed column with cocurrent downward flow. J Chem Eng Jpn 1977;10:468–74.
22. Cho JS, Wakao N. Determination of liquid-side and gas-side volumetric mass transfer coefficients in a bubble column. J Chem
Eng Jpn 1988;21:576–81.
23. Specchia V, Baldi G. Pressure drop and liquid holdup for two phase concurrent flow in packed beds. Chem Eng Sci 1977;32:
515–23.
24. Farsi M, Jahanmiri A. Methanol production in an optimized dual-membrane fixed-bed reactor. Chem Eng Process: Process
Intensific 2011;50:1177–85.