Energy Sources, Part A: Recovery, Utilization, and

Environmental Effects

ISSN: 1556-7036 (Print) 1556-7230 (Online) Journal homepage: http://www.tandfonline.com/loi/ueso20

CFD design and simulation of ethylene dichloride

(EDC) thermal cracking reactor

Afshin Davarpanah, Mojtaba Zarei, Kamran Valizadeh & Behnam Mirshekari

To cite this article: Afshin Davarpanah, Mojtaba Zarei, Kamran Valizadeh & Behnam
Mirshekari (2018): CFD design and simulation of ethylene dichloride (EDC) thermal cracking
reactor, Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, DOI:
10.1080/15567036.2018.1549133

To link to this article: https://doi.org/10.1080/15567036.2018.1549133

Published online: 26 Nov 2018.

Submit your article to this journal

Article views: 9

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ueso20
ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS
https://doi.org/10.1080/15567036.2018.1549133

CFD design and simulation of ethylene dichloride (EDC) thermal

cracking reactor
Afshin Davarpanaha, Mojtaba Zareib, Kamran Valizadehc, and Behnam Mirshekaria
a
Department of petroleum engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran;
b
Department of Chemical Engineering, Shahreza Branch, Islamic Azad University, Shahreza, Iran; cDepartment of
Chemical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran

ABSTRACT ARTICLE HISTORY

At high temperature, ethylene dichloride decomposes into vinyl chloride Received 24 April 2018
monomer and HCl by a complex reaction mechanism. Due to chemical Revised 15 October 2018
reaction type and heat transfer mechanism, design and simulation of Accepted 20 October 2018
cracking reactors are complicated. This process is very important from KEYWORDS
either chemical reaction or transport phenomena’s point of view. In this Thermal cracking reactor;
study, 2D simulation of cracking reactor is performed with FLUENT software. EDC; VCM; fluent software;
Simulation results were compared with the experimental data. Then, we COT
investigate the simulation results and effect of various parameters on the
process. Results show that with process progress and temperature increase
during reactor length, conversion and coke formation amount are
increased. At the end of the process, conversion is 52.69 that has good
agreement with industrial data. Addition of promoter to the process fluid
and fluid behavior in different coil outlet temperatures was studied; finally,
the effects of these factors are described in detail.

Introduction
From the 1960s up to now, the technology of reactor design and petrochemical furnaces has
gone through ups and downs and reached to the considerable achievements in the field of
design knowledge, arrangement of coils, and layout of the material used. Therefore, increase
of reactor's input heat flux, set of appropriate distribution, temperature reduction of the
reactor shell are considered as the energy consumption parameters in the simulation of
EDC reactor and furnaces. This design is based on the desired amount of converting feed
and selecting different products, and it requires an understanding of complex communication
and simultaneously heat transfer, fluid dynamics, mass transfer, and reaction kinetics in
a complicated mixture of hydrocarbon components. However, due to changes in the price
of each product, it is considered to be more profitable in the units and requires a high
understanding of the relationship between physical phenomena that affects the economic and
operational limits of the system. In this research, by focusing on simulation of reactor of
producing vinyl chloride monomer (VCM) through thermal cracking process, 95% of the
produced material need to produce PVC. Cracking process happened in the furnace that
provides the required heat for this process by burning flare gas in furnace (Andersson, Yuan,
and Sundén 2014; Shirmohammadi et al. 2015; Boudewijns et al. 2015; Cavalcanti et al. 2018;
Charitopoulos et al. 2018; Chen et al. 2017; Chong 2017). VCM is an important raw material
in the production of PVC, that was specifically produced by ethylene dichloride (EDC)

CONTACT Afshin Davarpanah afshindpe@gmail.com Department of petroleum engineering, Science and Research
Branch, Islamic Azad University, Tehran, Iran
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ueso.
© 2018 Taylor & Francis Group, LLC
2 A. DAVARPANAH ET AL.

thermal cracking process. Thermal cracking is the usual commercial method for producing
VCM. This process is commercially done within a specific temperature and pressure, due to
economic issues, including the small size of the reactors, optimization of heat transfer, and
ensuring the highest conversion rate. By careful study of reaction procedure and its stages, it
can be used as a better thought of reaction procedure and also the system can be moved on
an optimum operation. In thermal cracking, EDC is affected by various factors; in addition to
the original products, no predictable materials such as acetylene, ethylene, methyl chloride,
butadiene, benzene, and chlorinated hydrocarbons would be produced. Some of these com-
pounds were toxic and in addition to impacting on the environment and human health
they imposed high cost when deleted from the system. It is shown clearly in Figure 1
(Razmjoo et al. 2017; Edler et al. 2017; Farokhi and Birouk 2018; Greiner et al. 2015;
Rabbani et al. 2018; Higham et al. 2017; Hopf 2016).
According to this research, during the process, about 20–30 different chemical composi-
tions have been identified. Many of these combinations with EDC arrived in the coils, so it is
very important that the impure component were not included in the feed. For example,
butadiene with the concentration more than 100 ppm was considered as an impure compo-
nent, because removing these impurities from the VCM flow was very difficult (Hosseinpour,
Ahmadi, and Fatemi 2015; Kumar et al. 2016; Li et al. 2015; Pietrowicz, Kolasiński, and
Pomorski 2018; Rodemerck et al. 2016). Actual mechanism of thermal cracking for heavy oil
components are included numerous radical and molecular reaction in the reactor. The use of
all reactions and components in the model is impossible, and even solving equations using
modern computers takes a long time. For this reason, many efforts have been made by
researchers to use simplified reactions with logical methods. In modeling a thermal cracking
reactor, knowing feed specifications, temperature, inlet pressure, and reactor size is not only
enough, but what is important is first to understand the facts that occur during the
reactor. The mechanism of chemical reactions that occur in one thermal cracking reactor
thermal should be evaluated. Only in this way that we will be able to determine the
distribution and concentration of each output component at any point of the reactor, because
by having the reaction mechanisms, balance equations are written for each component and by
using mass and momentum balance equations, mathematical model of reactor will be pro-
posed and by this case effects of several parameters on distributing output products and
amount of conversion feed will be anticipated. In the case of pyrolysis reactions, three models
are proposed up to now (Sanz et al. 2016; Scharfe et al. 2016a, 2016b; Schulz et al. 2015;

Figure 1. General scheme of furnace EDC thermal cracking units.

 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 3

Tawai and Panjapornpon 2016; Won, Kim, and Kim 2015): experimental model, radical
model, and molecular model.

Methodology
Experimental model
This model has been widely administered by olefins producers. In this case that is the simplest
case, only overall mass and energy balance around unit will be done and model illustrated
regardless of reactions occurred within the reactor. Work procedure in this case was using
thermal cracking of a semi-industrial plant to obtain the required information. subsequently, an
experimental rule or semi-theory to obtain data from semi-industrial plant in industrial units
and plants were provided to check the reliability of simulation performances.

Radical model
Hydrocarbon thermal cracking was a highly endothermic phenomenon caused by applying heat between
carbon–carbon bonds and carbon–hydrogen bond of fractured hydrocarbons and as a result of heavy
hydrocarbons break up into hydrogen, methane, olefins, de-olefins, and other compounds. However, if
excessive heat is applied, efficiency was increased and heavy aromatic compounds and more coke were
formed. The steps of radical mechanism are described as follows: (i) beginning reactions: for starting
thermal cracking reactions, a lot of energy is consumed for breaking the carbon–carbon bond and forming
free radicals. (ii) Hydrogen adsorption reactions: one of the most important reactions in the mechanism of
thermal cracking, splitting hydrogen from molecules by an activated radical and as a result small molecules
and heavy radicals were being created. (iii) Isomerization reactions: in these reactions, heavy radicals break
from carbon–carbon bond and change to smaller radicals and unsaturated molecules. The beginning
reactions of isomerization compete with radical decomposition reactions. This reaction is caused by free
radicals in reactive position changes. (iv) The final stage: the final reactions were opposite of the beginning
reactions and conducted in two general ways. In the first stage, clashing of two radicals together forming
a larger molecule that is called combination reactions, and in the next stage by exchanging of hydrogen
between two radicals, two types of molecules appeared that is called disproportionation sharing reaction. (v)
Molecule reactions: in addition to the abovementioned radical reactions, molecular reactions are required
which are critical for the accuracy of the model.

Molecular model
In fact, the easiest way to explain the mechanism of thermal cracking reactions is by using
molecular models. In this way, we can avoid solving a series of problems due to differential
equations. To determine the kinetic parameters in this model, a set of experimental data with
high quality is needed (Tawai and Panjapornpon 2016; Won, Kim, and Kim 2015; Wong 2014;
Yang et al. 2017).

Procedure of coking model in the reactor

Generally speaking, in the cracking units, there are three different mechanisms to form coke in the
thermal reactor which are stated as follows:

● Coke formation which is caused by the surficial catalytic reactions with the existing hydro-
carbons in the gas phase.
● The formation of heavy aromatic compounds in the gas phase at high temperatures in
the furnace is caused by a chemical reaction, and subsequently, they are settled in a
4 A. DAVARPANAH ET AL.

physical form at low temperature. The compressed droplets are absorbed in the surface
and are converted to the coke in the dehydrogenase processes. This type of coke would
be in the form of film or spherical in the surface.
● Another type of coke is formed according to the temporal reaction of gas component
phases with the active materials on the surface. The gas components are estylene, de
oliphins, free radicals of methyl, ethyl, and vinyl which are considered as the principal
issues in the thermal cracking processes and subsequently leads to formation and
settlement of cokes at internal pipe surfaces.

Simulation analysis
Implementation process
In this section, simulation of EDC pyrolysis reactor by Fluent software is discussed. The first step in the
simulation of an issue by CFD method is to produce a geometrical definition of the proposed reactor.
Finally, to simulate the flow procedure in Fluent software, the properties of utilized mesh in the reactor were
entered in the simulator. Since then, the model was produced by Gambit software to model the mesh system.
Applied gridding in this model including tetrahedral elements in form of T-grid with non-structured type.
The number of used elements were 165,469. Desired model was drawn in Gambit software which is shown
in Figure 2. Furthermore, reactor geometry and mesh production profile are mentioned in Table 1.

Numerical methods by fluent software

Using the methods of Fluent software integral equations of mass and momentum durability, energy
equations and other scalars such as turbulence and equations of chemical interactions can be solved. The
range is divided into separate control volume using a computing grid. Integration of equations in
each volume controls to create algebraic equations with variables, such as velocity, pressure, temperature
and Conserved scalars like mass and energy. Equations integration are mostly analyzed algebraic equations
which included different variables such as velocity, pressure, temperature. The chemical components used
in this simulation analysis are explained in Table 2.
The thermodynamic properties of each formula in the furnace are explained in Table 3.

Results and discussion

Computational methods in software
In this part of study, the flowing model in the turbulent and standard equations were analytically
described. This model is one of the most famous and applicable models wth two equivalent
models of turbulent flow in a wide range of flow which is responsive with good approximation.
Used parameters in the standard model of K-є arementioned in Table 4.

Figure 2. View of the produced mesh in the software.

 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 5

Table 1. Reactor geometry and mesh production profile.

Coil length 95 meter
Length of each branch 13.5 meter
Coil diameter 0.134 meter
Radius of the bend 0.178 meter
Length of the bend 0.5 meter
Number of bend 9
Thickness of the wall 0.008 meter
Mesh interval size 0.03 meter
Total elements 165469
Examine quality mesh type Equiangular skew
Average quality range 0.2–0.4

Table 2. The chemical components that is used in the simulation analysis.

Molecular Components Radical Components
EDC Dichloride-Ethane R1 Cl*
VCM Vinyl-Chloride R2 CH2Cl-CH2*/CH3-CHCl*
HCl Hydrogen-Chloride R3 CH2Cl-CHCl*
CHCl3 Trichloride-Methane R4 CHCl2-CH2*
CCl4 Tetrachloride-Methane R5 CHCl=CH*/CH2=CCl*
EC Ethyl-Chloride R6 CH2Cl-CCl2*/CHCl2-CHCl*
Di Dichloride-Ethyl R7 CHCl2-CCl2*/CCl3-CHCl*
Tri Trichloride-ethylene R8 CCl3*
CP Chloroprene
C2H2 Acetylene
C6H6 Benzene
C4H6Cl2 Dichloride-butane
C Coke

Under relaxation factor

Solution control parameters, including variables that are convergence solution, play an important
role. Values used in this model are shown in Table 5.

Choose the best mesh independency network from the mesh

One of the important things that must be considered in the CFD calculation is independency of the mesh. In
order to determine the best mesh, the EDC conversion factors for five types of mesh are investigated, and the
results of evaluation of mesh to select the appropriate mesh are given in Table 6.
According to Table 6 and its comparison with the results of simulations and different meshes, it can be
seen that the number of meshes 165469 and 172351 had the same EDC conversion factor. The error between
the conversion factor of industrial and simulation was 8.8%. Consequently, the number of meshes did not
play a substantial role in the EDC conversion factor due to the simulated results from Table 6, and it was
considered as an independency parameter in the calculation of EDC conversion factor.

Temperature of process fluid in the length of the coil

Cracking reactions are highly endothermic. Thereby, to commence the reaction, the applied heat
was quickly absorbed in the system which is caused to increased the temperature. Moreover, as it is
evident in Figure 3, simulation results and industrial data were in a good agreement. . The highest
difference is about 13 degrees Kelvin. This difference is because of the following three reasons:
(i) Industrial data are measured by a thermocouple, it should be considered that we do not
accurately know that when these measurements are done. Measuring temperature in what time
duration and at which stage of furnace operation is very important, because by passing the time, we
6 A. DAVARPANAH ET AL.

Table 3. The thermodynamics properties of each formula in the furnace.

Reactions A[(cm3/mole)n−1s−1] n E(kj/mole)
Radical reaction chain initialization
EDC → R1 + R2 5.9 × 1015 1 342
CCl4 → R1 + R8 2.2 × 1012 1 230
Radical reaction chain propagation
EDC + R1 → HCl + R3 1.3 × 1013 2 7
EDC + R5 → VCM + R3 1.2 × 1013 2 34
EDC + R2 → EC + R3 1.0 × 1012 2 42
EDC + R4 → 1,1 + R3 5.0 × 1011 2 45
EDC + R6 → 1,1,2 + R3 2.0 × 1011 2 48
EDC + R7 → 1,1,1,2 + R3 1.0 × 1011 2 56
EDC + R8 → CHCl3 + R3 1.0 × 1012 2 63
VCM + R1 → R4 9.1 × 1010 2 0
VCM + R1 → HCl + R5 1.2 × 1014 2 56
VCM + R5 → CP + R1 5.0 × 1011 2 31
VCM + R4 → C4H6H2 + R1 2.0 × 1010 2 30
VCM + R2 → EC + R5 3.0 × 1011 2 61
R3 ↔VCM + R1 2.1 × 1014 1 84
R5 ↔C2H2 + R1 5.0 × 1014 1 90
R6 ↔Di + R1 2.0 × 1013 1 70
R7 ↔Tri + R1 2.5 × 1013 1 70
2C2H2 + R5 → C6H6 + R1 1.0 × 1014 2 20
EC + R1 → HCl + R2 1.7 × 1013 1 4
1,1 + R1 → HCl + R4 1.2 × 1013 1 6
1,1,2 + R1 → HCl + R6 1.7 × 1013 1 15
1,1,1,2 + R1 → HCl + R7 1.7 × 1013 2 17
CHCl3 + R1 → HCl + R8 1.6 × 1013 2 14
CCl4 + R5 → Di + R8 5.0 × 1011 2 33
CCl4 + R4 → 1,1,2 + R8 1.0 × 1012 2 33
CCl4 + R5 → 1,1,1,2 + R8 5.0 × 1011 2 33
Radical reaction chain termination
R2 + R1 → VCM + HCl 1.0 × 1013 2 13
R3 + R1 → Di + HCl 1.0 × 1013 2 12
R6 + R8 → Di + CCl4 1.0 × 1013 2 13
C2H2 + 2R1 → 2C + 2HCl 1.6 × 1014 2 70

Table 4. Parameters of K-є model.

Parameter Standard k-є
CD 0.09
C1 1.44
C2 1.92
σk 1
σε .31

Table 5. The amount of under relaxation factors.

pressure 0.1
Density 1
Body Force 1
Momentum 0.2
Turbulence Kinetic Energy 0.8
Turbulence Dissipation Rate 80.8
Turbulent Viscosity 1
 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 7

Table 6. Evaluation of mesh to select the appropriate mesh.

Number of mesh EDC conversion factor
98452 47.05
125893 49.1
148216 50.8
165469 52.69
172351 52.69
Experimental amount 48.4

have coke settling in the coils. However, due to the low heat transfer coefficient of coke, temperature
values in the thermocouple weren’t similar versus the time. So as we do not accurately measure the
time, there is the possibility of error. The profile of the simulated temperature is clearly shown in
Figure 3. The temperature profile for each coil is measured in the average form to be more matched
with the industrial data and minimize the errors. Moreover, we considered the location of thermo-
couple in the coil and its effects on the temperature measurements. Thereby, the simulation analysis
and industrial data had 50 degrees’ kelvin alteration. In Figure 4, 2D cross-sectional curve of
simulated temperature profile is schematically depicted for the EDC reactor.

EDC mass fraction changes during the length of the coil

By increasing temperature conversion factor, at the end of the reactor, conversion factor reaches
52.7%. EDC mass fraction in the entrance is 0.97 and in the outlet it is 0.46. EDC mass fraction in
the length of the reactor is shown in Figure 5.
2D cross-sectional curve of EDC mass fraction in the length of reactor is schematically depicted in
Figure 6.

VCM mass fraction changes during the length of the coil

At the beginning of the production process, the received heat from the furnace to increase the
temperature of process fluid had a low value. Next, this value will be increased gradually. VCM mass
fraction at the entrance is 0.018 and in the outlet it is 0.23. It should be noted that a small amount of
VCM at the entrance is the result of feedback, which is shown in Figure 7. Furthermore, 2D cross-
sectional curve of VCM mass fraction in the length of the reactor is shown in Figure 8.

780
770
TEMPERATURE( K )

760
750
740
730
720
710
700
690
0 50 100 150
LENGHT OF REACTOR(METER)
Figure 3. Simulated temperature profile of process fluid in the length of reactor.
8 A. DAVARPANAH ET AL.

Figure 4. 2D cross sectional curve of simulated temperature profile.

Coke concentration changes during the length of the reactor

In this part of study, the presence of acetylene in the formation of coke in the reactor were being
investigated. By reaction progress and increase in temperature, the amount of formed coke will be
increased. Coke concentration profile is shown in Figure 9. Furthermore, 2D cross-sectional
curve of coke concentration profile in the length of the reactor is shown in Figure 10.

Process fluid pressure in the length of the coil

At the beginning of the reasction, thecoil pressure drop is lower, but gradually with the progress of
the reaction and at the end of the reaction coil pressure drop will be increased. Several factors that
led to this decline include the following:
1. During the process, we will have a variety of reasons for forming coke. Due to the formation of
coke in the coil sidewall, the pressure drop had decreased.

1
EDC mass fraction

0.8

0.6

0.4

0.2

0
0 30 60 90 120 150
lenght of reactor(M)
Figure 5. EDC mass fraction in the length of reactor.
 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 9

Figure 6. 2D cross sectional curve of EDC mass fraction in the length of reactor.

0.25
VCM mass fraction

0.2

0.15

0.1

0.05

0
0 30 60 90 120 150
lenght of reactor(M)
Figure 7. VCM mass fraction in the length of reactor.

Figure 8. 2D cross sectional curve of VCM mass fraction in the length of reactor.
10 A. DAVARPANAH ET AL.

0.08

coke concentration
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
0 30 60 90 120 150
lenght of reactor(M)
Figure 9. Coke concentration profile in the length of reactor.

2. The total pressure drop due to nine bends in the processing system is considerable enough.
Pressure profile was shown in Figure 11.Furthermore, 2D cross sectional curve of process fluid
pressure profile is shown in Figure 12.

Fluid velocity change during the length of the reactor

However, coke formation in the coil had caused the decrease of coil effective area to flow the
fluid, the fluid volume was reached a plateau and the velocity will be increased. Velocity profiles
are plotted in Figure 13. Furthermore, 2D cross-sectional curve of process fluid velocity vectors
along the radius of the reactor is shown in Figure 14.

Investigate non-ideal case by considering process fluid

Considered that the process fluid is a real gas. As the result implies the assumption that the process
fluid is a real gas, the conversion factor will be less than ideal condition. EDC conversion factor is
described in Table 7.

Figure 10. 2D cross sectional curve of coke concentration profile in the length of reactor.
 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 11

2950
2900

pressure (KPa)
2850
2800
2750
2700
2650
2600
0 30 60 90 120 150
lenght of reactor(M)
Figure 11. Process fluid pressure profile.

Figure 12. 2D cross sectional curve of process fluid pressure profile.

3.5
velocity (meter per

3
2.5
second)

2
1.5
1
0.5
0
0 1 2 3 4 5 6 7
radius of coil(centimeter)
Figure 13. Process fluid velocity vectors along the radius of the reactor.

Simulate different coil outlet temperature (COT) in the EDC conversion factor
To simulate the coking model in the EDC reactor, different COTs and concentrations of cokes are
considered. The assumed COTs are 760, 770, and 780, and their impact on the temperature is schematically
12 A. DAVARPANAH ET AL.

Figure 14. 2D cross sectional curve of process fluid velocity vectors along the radius of the reactor.

800
COT=790 K
790
COT=780 K
780
COT=774 K
770
COT=770 K
Temperature

760
COT=760 K
750
740
730
720
710
700
690
0 30 60 90 120 150
Reactor Length
Figure 15. Effects of different COTs on the temperature.

Table 7. The amount of the EDC conversion factor in different equations.

Gas equation EDC conversion factor
Ideal gas 0.5269
Peng-Robinson 0.5142
Redlich-Kwong 0.5078
Soave-Redlich-Kwong 0.5045
Aungier-Redlich-Kwong 0.4975

depicted in Figure 15. As it is evident in Figure 15, for high values of COTs, the temperature has witnessed
higher values. In Figure 16, the influence of COTs on the EDC conversion factor is shown, and it is
illustrated that in the high values of COTs, the conversion factor has higher values, The results of this study
 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 13

0.7

0.65

EDC conversion factor

0.6

0.55

0.5

0.45

0.4
755 760 765 770 775 780 785 790 795
COT

Figure 16. Effects of different COTs on the EDC conversion factor.

0.016
0.014
Coke concentration

0.012
0.01
0.008
0.006
0.004
0.002
0
755 760 765 770 775 780 785 790
COT
Figure 17. Effects of different COTs on the coke concentration.

would be considered as the efficient technique to propose economic issues of reactor designation. The effect
of COTs on the coke concentration is shown in Figure 17.

Conclusion
In this research, by using FLUENT software for simulating the EDC, thermal cracking process was studied
in two-dimensional mode. According to proposed profiles, it was determined that by improving cracking
reaction and increasing inside temperature of the reactor, conversion factor will be increased and so the coke
settling in the sidewall of the coil will be increased. As described earlier, one of the effective factors in the
optimum operation of the system is to apply appropriate thermal profile on the coil area, because with
misappropriation of this profile or lower conversion factor than the anticipated amount, the amount of
formed coke was more than the allowable limit. Profound impacts of non-ideal assumptions in the fluids
processing were considered and it is concluded that measured conversion factor from the simulated analysis
is closer to the industrial data.
14 A. DAVARPANAH ET AL.

Recommendations for future

● To obtain more accurate results, more reactions can be considered in the simulation process and the
effect of different combinations should be examined.
● In this study, it was assumed that the amount of radiation heat transfer compared with the
conduction and convection is lower. By considering the heat radiation, more accurate results
can be obtained.
● Turbulence models such as the RNG (k-ε) and Realizable (k-ε) model can be applied and
results compared.
● One of the things that caused the failure is the lack of ash on the outer surface that leads to full
conduction of heat to the process fluid within the coil. By applying this strength in the overall
heat transfer coefficient, more accurate results can be achieved.

References
Andersson, M., J. Yuan, and S. Bengt. 2014. SOFC cell design optimization using the finite element method based CFD
approach. Fuel Cells 14 (2):177–88. doi:10.1002/fuce.201300160.
Boudewijns, T., M. Piccinini, P. Degraeve, A. Liebens, and D. V. Dirk. 2015. Pathway to vinyl chloride production via
dehydrochlorination of 1, 2-dichloroethane in ionic liquid media. ACS Catalysis 5 (7):4043–47. doi:10.1021/
acscatal.5b00736.
Cavalcanti, E. D. C., É. C. G. Aguieiras, P. R. Da Silva, J. G. Duarte, E. P. Cipolatti, R. Fernandez-Lafuente, J. A. C. Da
Silva, and D. M. G. Freire. 2018. Improved production of biolubricants from soybean oil and different polyols via
esterification reaction catalyzed by immobilized lipase from Candida rugosa. Fuel 215:705–13. doi:10.1016/j.
fuel.2017.11.119.
Charitopoulos, A. G., R. Visser, R. Eling, and C. I. Papadopoulos. 2018. Design optimization of an automotive
turbocharger thrust bearing using a CFD-based THD computational approach. Lubricants 6 (1):21. doi:10.3390/
lubricants6010021.
Chen, Z., V. M. Reddy, S. Ruan, N. A. K. Doan, W. L. Roberts, and N. Swaminathan. 2017. Simulation of MILD
combustion using Perfectly Stirred Reactor model. Proceedings of the Combustion Institute 36 (3):4279–86.
doi:10.1016/j.proci.2016.06.007.
Chong, S. 2017. Process Simulation for VCM Production. Chemical Engineering Process Simulation.
Edler, F., B. Geier, W. Reiter, J. Rieger, C. Spijker, and H. Raupenstrauch. 2017. Development of an atmosphere
particle kinetic model for particle reactions in a combustion Flash-Reactor using CFD-methods. Energy Procedia
120:540–47. doi:10.1016/j.egypro.2017.07.200.
Farokhi, M., and M. Birouk. 2018. A new EDC approach for modeling turbulence/chemistry interaction of the
gas-phase of biomass combustion. Fuel 220:420–36. doi:10.1016/j.fuel.2018.01.125.
Greiner, D., B. Galván, J. Périaux, N. Gauger, K. Giannakoglou, and G. Winter. 2015. Advances in evolutionary and
deterministic methods for design, optimization and control in engineering and sciences, Vol. 1. Switzerland: Springer.
Higham, M. D., M. Scharfe, M. Capdevila-Cortada, J. Perez-Ramirez, and N. Lopez. 2017. Mechanism of ethylene
oxychlorination over ruthenium oxide. Journal of Catalysis 353:171–80. doi:10.1016/j.jcat.2017.07.013.
Hopf, A. 2016. “Finding the perfect flow-port development using CFD topology optimization and adjoint solver.”
STAR Global Conference, Prague, Czech Republic.
Hosseinpour, M., S. J. Ahmadi, and S. Fatemi. 2015. Successive co-operation of supercritical water and silica-supported
iron oxide nanoparticles in upgrading of heavy petroleum residue: Suppression of coke deposition over catalyst. The
Journal of Supercritical Fluids 100:70–78. doi:10.1016/j.supflu.2015.02.013.
Kumar, A., J. B. Joshi, A. K. Nayak, and P. K. Vijayan. 2016. 3D CFD simulations of air cooled condenser-III:
Thermal–Hydraulic characteristics and design optimization under forced convection conditions. International
Journal of Heat and Mass Transfer 93:1227–47. doi:10.1016/j.ijheatmasstransfer.2015.10.048.
Li, Y., C. Zhang, Y. Liu, X. Hou, R. Zhang, and X. Tang. 2015. Coke deposition on Ni/HZSM-5 in bio-oil
hydrodeoxygenation processing. Energy & Fuels 29 (3):1722–28. doi:10.1021/ef5024669.
Pietrowicz, S., P. Kolasiński, and M. Pomorski. 2018. Experimental and numerical flow analysis and design optimiza-
tion of a fume hood using the CFD method. Chemical Engineering Research and Design 132:627–43. doi:10.1016/j.
cherd.2018.02.011.
Rabbani, E., A. Davarpanah, and M. Memariani. 2018. An experimental study of acidizing operation performances on
the wellbore productivity index enhancement. Journal Of Petroleum Exploration and Production Technology 1-11.
doi: 8 doi:10.1007/s13202-018-0441-8.
 ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS 15

Razmjoo, A., M. Qolipour, R. Shirmohammadi, S. M. Heibati, and I. Faraji. 2017. Techno-economic evaluation of
standalone hybrid solar-wind systems for small residential districts in the central desert of Iran. Environmental
Progress & Sustainable Energy 36 (4):1194-1207. doi:10.1002/ep.v36.4.
Rodemerck, U., S. Sokolov, M. Stoyanova, U. Bentrup, D. Linke, and E. V. Kondratenko. 2016. Influence of support
and kind of VOx species on isobutene selectivity and coke deposition in non-oxidative dehydrogenation of
isobutane. Journal of Catalysis 338:174–83. doi:10.1016/j.jcat.2016.03.003.
Sanz, S. G., L. McMillan, J. McGregor, J. A. Zeitler, N. Al-Yassir, S. Al-Khattaf, and L. F. Gladden. 2016. The
enhancement of the catalytic performance of CrOx/Al2O3 catalysts for ethylbenzene dehydrogenation through
tailored coke deposition. Catalysis Science & Technology 6 (4):1120–33. doi:10.1039/C5CY01157D.
Scharfe, M., P. A. Lira-Parada, A. P. Amrute, S. Mitchell, and J. Pérez-Ramírez. 2016a. Lanthanide compounds as
catalysts for the one-step synthesis of vinyl chloride from ethylene. Journal of Catalysis 344:524–34. doi:10.1016/j.
jcat.2016.10.026.
Scharfe, M., P. A. Lira-Parada, V. Paunović, M. Moser, A. P. Amrute, and J. Pérez-Ramírez. 2016b. Oxychlorination–
dehydrochlorination chemistry on bifunctional ceria catalysts for intensified vinyl chloride production. Angewandte
Chemie 128 (9):3120–24. doi:10.1002/ange.201510903.
Schulz, L. A., L. C. S. Kahle, K. H. Delgado, S. A. Schunk, A. Jentys, O. Deutschmann, and J. A. Lercher. 2015. On the
coke deposition in dry reforming of methane at elevated pressures. Applied Catalysis A: General 504:599–607.
doi:10.1016/j.apcata.2015.03.002.
Shirmohammadi, R., B. Ghorbani, and M. Hamedi, et al. 2015. Optimization of mixed refrigerant systems in low
temperature applications by means of group method of data handling (GMDH). Journal of Natural Gas Science and
Engineering 26:303–312.
Tawai, A., and C. Panjapornpon. 2016. Input–output linearizing control strategy for an ethylene dichloride cracking
furnace using a coupled PDE-ODE model. Industrial & Engineering Chemistry Research 55 (3):683–91. doi:10.1021/
acs.iecr.5b03759.
Won, D.-J., J. Kim, and J. Kim. 2015. Design optimization of duct-type AUVs using CFD analysis. Intelligent Service
Robotics 8 (4):233–45. doi:10.1007/s11370-015-0179-9.
Wong, R. 2014. “Advanced modelling of vinyl chloride monomer production via thermal cracking of ethylene
dichloride.”
Yang, M., Y. Dawei, M. Liu, L. Zheng, X. Zheng, Y. Wei, F. Wang, and Y. Fan. 2017. Optimization of MBR
hydrodynamics for cake layer fouling control through CFD simulation and RSM design. Bioresource Technology
227:102–11. doi:10.1016/j.biortech.2016.12.027.