UNIVERSITI SAINS MALAYSIA

SCHOOL OF CHEMICAL ENGINEERING

SEMESTER I, ACADEMIC SESSION 2022/2023

EKC316 SEPARATION PROCESS

INTEGRATED DESIGN PROJECT

REPORT

GROUP NUMBER : 31

GROUP MEMBERS : MAH HIN MAN 148703

: NURUL SYAFIQAH BINTI ARIFF 150134

: NURUL HUSNA BINTI KHAMIS 153868

: NUR AFZA NURINA BINTI ABDULLAH SANI 152967

LECTURER : ASSOC. PROF. DR. MOHAMAD ZAILANI ABU BAKAR

SUBMISSION DATE : 29/01/2023

 Table of Contents
1.0 Abstract ............................................................................................................................... 1
2.0 Introduction ........................................................................................................................ 2
3.0 Results and Discussion ....................................................................................................... 4
3.1. Mass Balance Review ................................................................................................. 4
3.2. Dew Point and Bubble Point Calculations .................................................................. 5
3.3. Manual Calculation of Short Cut Method (FUG) ....................................................... 7
3.4. DSTWU and RadFrac Simulations ........................................................................... 10
3.4.1. Short-Cut (DSTWU) Column ............................................................................ 10
3.4.2. Rigorous (RadFrac) Column .............................................................................. 13
3.5. Profiles Analysis ....................................................................................................... 17
3.6. Parametric Analysis (Sensitivity) .............................................................................. 19
3.6.1. Parametric Analysis for DSTWU Column ........................................................ 19
3.6.2. Parametric Analysis for RadFrac Column ......................................................... 20
4.0 Conclusion ........................................................................................................................ 21
5.0 References ........................................................................................................................ 23
Appendices .................................................................................................................................. i
1.0 Abstract

Distillation is arguably the most widely used separation method in the industry. A
multicomponent distillation separates a mixture of volatile substances based on their boiling
points, hence their relative volatilities with respect to the heavy key component. In this
integrated design assignment, our work is done based on distillation column C-105, which
recovers 99.5% of hydrogen chloride (light key) as top product and 99.5% vinyl chloride
(heavy key) as bottom product from a mixture consisting of ethylene, acetylene, 1,2-
dichloroethane and 1,1,2-trichloroethane in addition to the light key component as well as the
heavy key component. Manual calculation was done using the short-cut method also known as
Fenske-Underwood-Gilliland (FUG) method prior to simulation using Aspen Plus. Aspen Plus
was then utilised to simulate short-cut (DSTWU) column and rigorous (RadFrac) column; in
addition to the design of column internals, the analysis of three profiles (temperature, relative
volatility, molar flow rate) for the RadFrac simulated results, and the parametric analysis
(sensitivity) varying reflux ratio for DSTWU column, and varying reflux ratio and number of
stages for RadFrac column. As for the results, the minimum number of stages as well as the
actual number of stages were computed using both manual calculations (FUG) and short-cut
(DSTWU) distillation simulation in Aspen Plus. Both methods produce similar results, with
the minimum number of stages were 3.1166 and 3.3312 for FUG and DSTWU simulation
respectively; and the actual number of stages of 10.8555 and 9.3100 for FUG and DSTWU
simulation respectively. Next, the rigorous (RadFrac) distillation was simulated, and the
column parameters were investigated. With column diameter of 1.11673m and tray spacing of
0.6096m, a bubble-cap distillation column of column height of 4.8768m with only one tray
pass is recommended to be used to produce hydrogen chloride as top product. Next, profiles
analysis was done and discussed in Section 3.5. Lastly, a sensitivity analysis of the results was
performed, and higher number of stages and higher reflux ratio are recommended in order to
achieve a better separation.

1
2.0 Introduction

The IUPAC nomenclature for vinyl chloride is chloroethene with a formula of C2H3Cl or
H2C=CHCl. The molecular weight of chloroethene is 62.498 g/mol. Chloroethene has a
melting point of -153.7 °C and a boiling point of -13.4 °C. It is a colourless, flammable gas or
refrigerated liquid with a faintly sweet odour. It is highly toxic, flammable, and carcinogenic.
Chloroethene is mostly used to make Polyvinyl Chloride (PVC), which is subsequently used
to make pipes and bottles (PubChem, 2022). Chloroethene is mostly used to make Polyvinyl
Chloride (PVC), which is subsequently used to make pipes and bottles. (PubChem, 2022).
More than 10 million metric tonnes of vinyl chloride monomer (VCM) are produced globally
every year. China and the United States are among the top country that produced VCM (ChEBI,
2021). It has a high fire and explosion hazard as its boiling point is low. Thus, VCM will
undergo flash evaporation when it is exposed to the atmosphere. Besides, it can also form
peroxides when mixed with air which may explode. There are several health effects associated
with exposure to vinyl chloride in humans. Acute exposure may cause symptoms such as
dizziness, nausea, and ataxia meanwhile chronic exposure will lead to fatal respiratory failure
and cardiac arrhythmias (New Jersey Department of Health, 2015). There is also consistent
research done by scientists on the risk of liver cancer when in contact with VCM. (National
Cancer Institute, 2022). Direct chlorination, oxychlorination, purification and thermal cracking
of 1,2-dichloroethane (EDC), and purification of chloroethene monomer are the five common
procedures used to form chloroethene.

Figure 2: Structural Formula of Chloroethene (Alchetron, 2017)

Chloroethene can be produced through a few processes. For example, in direct chlorination,
homogeneous catalytic reaction of the liquid-phase feedstocks ethylene and chlorine produces
1,2-dichloroethane (EDC) in the presence of ferric chloride catalyst. Intimate mixing of

2
ethylene and chlorine in liquid EDC results in an exothermic reaction. The heat generated is
removed by cooling water or by operating the plug flow reactor at the boiling point of EDC to
enable vaporization of the pure product. The reaction temperature and pressure are 50°C and 1
atm. Slight excess of ethylene is favourable for complete chlorine conversion. Meanwhile,
oxychlorination process involves reaction between ethylene and hydrogen chloride, mixed with
oxygen, in the presence of an alumina-supported copper (II) chloride catalyst at 230°C and 1.5
atm. The disadvantages of oxychlorination process are that it forms carbon oxides and water
besides having a high impurity content. Thus, direct chlorination is more efficient and often
chosen. 1,2 Dichloroethane produced from direct chlorination and oxychlorination need to
undergo purification to obtain at least 99.5 wt% pure dry EDC. The thermal cracking process
then will run under high temperature and pressure to yield VCM. Finally, the VCM is purified
by passing through two different distillation columns. (Abu Bakar, 2022)

In this integrated project, we were tasked to design the C-105 distillation column. It is
used to remove 99.5% VCM as the bottom product and the column’s material is from stainless
steel. Distillation is a vital process that can cause a significant difference in the economic
operation of a process plant. The components in the column are separated based on their boiling
points. The component with the highest boiling point will be separated at the bottom of the
column as the bottom product, while that of lowest boiling point will form at the top of the
column as distillate (Internals, 2018).

As the feed contains several components, the basis of multicomponent distillation is

used in the design. All the components were listed based on their boiling points to determine
their key types, namely light key, heavy key, light-non key, and heavy-non key. The mass
balance of the C-105 unit is reviewed prior to any further calculations. Short cut method (FUG)
is used to manually calculate minimum the number of stages, minimum reflux number, and the
number of stages. Then, Aspen Plus is used to simulate short cut (DSTWU) and rigorous
(RadFrac) distillations. The profiles obtained which are temperature, relative volatility, and
molar flow rate are discussed in this report. Also, parametric analysis (sensitivity) is also done
for DSTWU column varying reflux ratio; and varying reflux ratio and number of stages for
RadFrac column.

3
3.0 Results and Discussion

Assumptions/Basis:

• 99.5% recovery of light key in distillate and 99.5% recovery of heavy key in bottom.
• Steady-state distillation.
• For computation using FUG method, feed is saturated liquid, q = 1.
• Feed temperature is the average of dew point temperature and bubble point temperature.
• Reflux ratio is 1.5 minimum reflux ratio.

3.1 Mass Balance Review

Pressure, 𝑃𝑃 =1.0132 bar

Table 3.1.1: Boiling Point of Component (Green, 2008)

Component Boiling Point [°C] Remark

Ethylene -103.90 LNK
Hydrogen chloride -85.05 LK
Acetylene -84.00 LNK
Vinyl chloride -13.40 HK
1,2-dichloroethane 83.47 HNK
1,1,2-trichloroethane 113.90 HNK

Based upon inspection of boiling points of each component, as well as the mass balance sheet
provided (attached as Appendix A), the revised mass balance table is provided below:

Table 3.1.2: Mass Balance Table for Revised Mass Balance

Feed (Stream 32) Distillate (Stream 33) Bottom (Stream 35)

Molar Molar Molar
Component
Flow Mole Flow Mole Flow Mole
Rate Fraction Rate Fraction Rate Fraction
[kmol/h] [kmol/h] [kmol/h]

Ethylene 0.11 1.22E-04 0.11000 3.44E-04 0.0000 0.00

1,2-
319.85 3.55E-01 318.25000 9.94E-01 1.6000 2.76E-03
dichloroethane

4
 Hydrogen
0.06 6.66E-05 0.06000 1.87E-04 0.0000 0.00
chloride

Vinyl chloride 318.91 3.54E-01 1.59455 4.98E-03 317.3155 5.47E-01

Acetylene 261.23 2.90E-01 0.00000 0.00 261.2300 4.50E-01

1,1,2-
0.11 1.22E-04 0.00000 0.00 0.1100 1.90E-04
trichloroethylene

3.2 Dew Point and Bubble Point Calculations

The relative volatility of each component in the feed can be calculated from the vapor pressure
of each component using the following relation [Perry’s]:

𝐶𝐶2
ln(𝑃𝑃) = 𝐶𝐶1 + + 𝐶𝐶3 ln(𝑇𝑇) + 𝐶𝐶4 𝑇𝑇 𝐶𝐶5 ; [𝑃𝑃][=][𝑃𝑃𝑃𝑃]
𝑇𝑇

Table 3.2.1: Vapour Pressure Constants for Each Individual Components (Green, 2008)

Component A B C D E
Ethylene 53.963 -2443.0 -5.5643 1.91E-05 2

Hydrogen chloride 104.270 -3731.2 -15.0470 3.13E-02 1

Acetylene 39.630 -2552.2 -2.7800 2.39E-16 6

Vinyl chloride 91.432 -5141.7 -10.9810 1.43E-05 2

1,2-dichloroethane 92.355 -6920.4 -10.6510 9.14E-06 2

1,1,2-trichloroethane 54.153 -6041.8 -4.5383 4.98E-18 6

Constant 𝐾𝐾𝑖𝑖 for each component is obtained by:

𝑃𝑃𝑖𝑖
𝐾𝐾𝑖𝑖 = ; 𝑃𝑃𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 101320𝑃𝑃𝑃𝑃
𝑃𝑃𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡
Relative volatility for each component, 𝛼𝛼𝑖𝑖 is given by:
𝐾𝐾𝑖𝑖
𝛼𝛼𝑖𝑖 =
𝐾𝐾ℎ𝑘𝑘
Whereas 𝐾𝐾ℎ𝑘𝑘 is for heavy key component (vinyl chloride).

5
The following equations are used to obtain dew point temperature at condenser and bubble
point temperature at reboiler:

Dew point:
𝑦𝑦𝑖𝑖 1 𝑦𝑦𝑖𝑖
∑𝑥𝑥𝑖𝑖 = ∑ = ∑ =1
𝐾𝐾𝑖𝑖 𝐾𝐾𝑐𝑐 𝛼𝛼𝑖𝑖

Rearranging the above equation gives:

𝑦𝑦𝑖𝑖
𝐾𝐾𝑐𝑐 = ∑
𝛼𝛼𝑖𝑖
Bubble point:
∑𝑦𝑦𝑖𝑖 = ∑𝐾𝐾𝑖𝑖 𝑥𝑥𝑖𝑖 = 𝐾𝐾𝑐𝑐 ∑𝛼𝛼𝑖𝑖 𝑥𝑥𝑖𝑖 = 1
Rearranging the above equation gives:
1
𝐾𝐾𝑐𝑐 =
∑𝛼𝛼𝑖𝑖 𝑥𝑥𝑖𝑖

The dew point and bubble point temperatures are obtained by trial and error using excel goal
seek until 𝐾𝐾𝑐𝑐 calculated is equal to 𝐾𝐾𝑐𝑐 of heavy key component. The results are summarized in
the Table 4 and Table 5 below.

Table 3.2.2: Dew Point Calculation for Condenser

Dew Point Temperature = 197.5432K = -79.6068°C
Mole
Component fraction Pi [Pa] Ki α y/α
(𝒙𝒙𝒊𝒊 )
Ethylene 3.44E-04 3.47E+05 3.42E+00 1.77E+02 1.94E-06

Hydrogen chloride 9.94E-01 1.37E+05 1.35E+00 6.98E+01 1.42E-02

Acetylene 1.87E-04 1.36E+05 1.34E+00 6.91E+01 2.71E-06

Vinyl chloride 4.98E-03 1.96E+03 1.94E-02 1.00 4.98E-03

1,2-dichloroethane 0.00 2.36E+00 2.33E-05 1.20E-03 0.00

1,1,2-trichloroethane 0.00 3.83E-01 3.78E-06 1.95E-04 0.00

6
 Table 3.2.3: Bubble Point Calculation for Reboiler
Bubble Point Temperature = 272.4064K = -0.7437°C
Mole
Component
fraction Pi [Pa] Ki α y/α
(𝒚𝒚𝒊𝒊 )
Ethylene 0.00 4.03E+06 3.98E+01 2.37E+01 0.00E+00

Hydrogen chloride 2.76E-03 2.51E+06 2.48E+01 1.48E+01 4.07E-02

Acetylene 0.00 2.60E+06 2.56E+01 1.53E+01 0.00

Vinyl chloride 5.47E-01 1.70E+05 1.68E+00 1.00 5.47E-01

1,2-dichloroethane 4.50E-01 2.71E+03 2.68E-02 1.59E-02 7.17E-03

1,1,2-trichloroethane 1.90E-04 6.84E+02 6.75E-03 4.02E-03 7.61E-07

3.3 Manual Calculation of Short Cut Method (FUG)

In order to obtain the average relative volatility of each component, the relative volatility of
each component in the feed is obtained and tabulated below.

Table 3.3.1: Relative Volatilities of Each Component in Feed

T = 232.9748K = -40.1752°C

Component Pi [Pa] Ki α

Ethylene 1.44E+06 1.42E+01 4.94E+01

Hydrogen chloride 7.56E+05 7.46E+00 2.59E+01

Acetylene 7.74E+05 7.64E+00 2.65E+01

Vinyl chloride 2.92E+04 2.88E-01 1.00

1,2-dichloroethane 1.62E+02 1.60E-03 5.55E-03

1,1,2-trichloroethane 3.25E+01 3.21E-04 1.11E-03

7
The average relative volatility of each component with respect to the heavy key component is
taken as the geometric mean of those of feed, top and bottom:

𝛼𝛼𝑖𝑖,𝐻𝐻𝐻𝐻 = (𝛼𝛼𝑖𝑖,𝐻𝐻𝐻𝐻(𝑡𝑡𝑡𝑡𝑡𝑡) 𝛼𝛼𝑖𝑖,𝐻𝐻𝐻𝐻(𝑏𝑏𝑏𝑏𝑏𝑏) 𝛼𝛼𝑖𝑖,𝐻𝐻𝐻𝐻(𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓) )1/3

The average relative volatilities of each component with respect to the heavy key component
are tabulated below:

Table 3.3.2: Average Relative Volatilities of Each Component Relative to Heavy Key
Component

Component α
Ethylene 59.1056
Hydrogen chloride 29.8652
Acetylene 30.3455
Vinyl chloride 1.0000
1,2-dichloroethane 0.0047
1,1,2-trichloroethane 0.0010

The average relative volatility of light key component relative to heavy key component is
29.8652.

Fenske Estimation of Minimum Number of Stages including Reboiler:

𝐿𝐿𝐿𝐿𝐷𝐷 𝐻𝐻𝐻𝐻𝐵𝐵
ln �
𝐻𝐻𝐻𝐻𝐷𝐷 � � 𝐿𝐿𝐿𝐿𝐵𝐵 �
𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 =
𝑙𝑙𝑙𝑙(𝛼𝛼�𝐿𝐿𝐿𝐿,𝐻𝐻𝐻𝐻 )

318.25 × 317.3155
� �
𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 = ln 1.5946 × 1.6
ln 29.8652

Solving for 𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 gives 𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 = 3.1166.

8
Underwood Estimation of Min. Reflux Number:

𝛼𝛼𝑖𝑖 𝑥𝑥𝑖𝑖,𝐹𝐹
1 − 𝑞𝑞 = ∑ � � − 𝑒𝑒𝑒𝑒 1
𝛼𝛼𝑖𝑖 − 𝜃𝜃

Table 3.3.3.1: Constants Used in Underwood Equation 1

Component Feed Mole Fraction αi

Ethylene 1.2219E-04 5.911E+01

Hydrogen chloride 3.5528E-01 2.987E+01

Acetylene 6.6647E-05 3.035E+01

Vinyl chloride 3.5424E-01 1.000E+00

1,2-dichloroethane 2.9017E-01 4.736E-03

1,1,2-trichloroethane 1.2219E-04 9.556E-04

Solving for 𝜃𝜃 gives 𝜃𝜃 = 1.9336.

𝛼𝛼𝑖𝑖 𝑥𝑥𝑖𝑖,𝐷𝐷
𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 + 1 = ∑ � � − 𝑒𝑒𝑒𝑒 2
𝛼𝛼𝑖𝑖 − 𝜃𝜃

Table 3.3.3.2: Constants Used in Underwood Equation 2

Component Distillate Mole Fraction αi

Ethylene 3.4373E-04 5.911E+01

Hydrogen chloride 9.9449E-01 2.987E+01

Acetylene 1.8749E-04 3.035E+01

Vinyl chloride 4.9827E-03 1.000E+00

1,2-dichloroethane 0.0000 4.736E-03

1,1,2-trichloroethane 0.0000 9.556E-04

Solving for 𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 gives 𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 = 0.05855

9
Gilliland Estimation of No. of Stages:

𝑁𝑁 − 𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 𝑅𝑅 − 𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 0.566

= 0.75 �1 − � � �
𝑁𝑁 + 1 𝑅𝑅 + 1

0.566
𝑁𝑁 − 3.1166 1.5(0.05855) − 0.05855
= 0.75 �1 − � � �
𝑁𝑁 + 1 1.5(0.05855) + 1

Solving for N gives 𝑁𝑁 = 10.8555

3.4 DSTWU and RadFrac Simulations

3.4.1 Short-Cut (DSTWU) Column

At first, DSTWU simulation was done using UNIFAC property methods. DSTWU simulation
is similar to FUG method, with the slight difference being DSTWU distillation simulates the
result using Winn equation instead of Fenske equation to calculate the minimum number of
stages. The input parameters and results are summarised in Table 3.4.1.1 and 3.4.1.2
respectively.

Table 3.4.1.1: Input Parameters for DSTWU Distillation Column

Parameter Values

Feed Temperature 232.9748K

Feed Pressure 1.0132 bar

Total Feed Flow Rate [kmol/hr] 900.27

Ethylene Molar Flow Rate [kmol/hr] 0.11

Hydrogen Chloride Molar Flow Rate [kmol/hr] 319.85

Acetylene Molar Flow Rate [kmol/hr] 0.06

Vinyl Chloride Molar Flow Rate [kmol/hr] 318.91

1,2-dichloroethane Molar Flow Rate [kmol/hr] 261.23

10
 1,1,2-trichloroethane Molar Flow Rate [kmol/hr] 0.11

Reflux Ratio -1.5 𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚

Light Key Component Recovery in Distillate 0.995

Heavy Key Component Recovery in Distillate 0.005

Pressure 1.0132 bar

Stage Pressure Drop 0 bar

Condenser and Reboiler Pressure 1.0132 bar

Figure 3.4.1.1: DSTWU Set Up

Table 3.4.1.2 Results Summary for DSTWU column using UNIFAC

Name C-105

Property method UNIFAC

Henry's component list ID

Electrolyte chemistry ID

Use true species approach for electrolytes YES

Free-water phase properties method STEAM-TA

Water solubility method 3

11
 Reflux ratio -1.5

Light key component recovery 0.995

Heavy key component recovery 0.005

Distillate vapor fraction 0

Minimum reflux ratio 0.159743758

Actual reflux ratio 0.239615637

Minimum number of stages 3.33116861

Number of actual stages 9.30995089

Feed stage 6.05863961

Number of actual stages above feed 5.05863961

Distillate temperature [°C] -85.0092066

Bottom temperature [°C] 0.439454053

Distillate to feed fraction 0.355465687

Name C-105

Table 3.4.1.3 Comparison between Manual Calculation and DSTWU

Parameter Manual Calculation DSTWU

Feed Temperature [°C] -40.1752 -40.1752

Top Temperature [°C] -79.6068 -85.0092

Bottom Temperature [°C] -0.7437 0.4394

Minimum reflux ratio 0.05855 0.1597

Actual reflux ratio 0.08783 0.2396

Minimum number of stages 3.1166 3.3312

Actual number of stages 10.8555 9.3100

12
Based on the comparison shown in Table 3.4.1.3, there are slight deviations in each parameter
obtained, causing the actual number of stages obtained using DSTWU simulation (N=9.3100)
to be slightly different than that of manual calculation (N=10.8555). The small discrepancy in
results of DSTWU simulation from manual calculation might be attributed to the different
degree of estimations used by the two methods. Besides, this might be caused by different
assumption, different basis and different database used. Therefore, the obtained data are valid
and can be applied for subsequent simulation using rigorous method (RadFrac).

3.4.2 Rigorous (RadFrac) Column

3.4.2.1 Simulation of RadFrac Column

RadFrac is the final general distillation unit operation. It can perform much more rigorous
calculation in Aspen Plus to perform simulation, sizing, and design of the tray within the
column (Hussain, n.d.). By using the actual reflux ratio, feed stage and actual number of stages
obtained from DSTWU to run the simulation along with some inputs as below

Table 3.4.2.1.1: Input Parameters for RadFrac Distillation Column

Parameter Value
Actual Reflux Ratio 0.239615637
Feed Stage 7
Actual Number of Stages 10
Distillate to Feed Ratio 0.355465687
Condenser Pressure 1.0132 bar
Stage Pressure Drop 0 bar

Figure 3.4.2.1.1: The RedFrac Setup

13
Table 3.4.2.1.2 Results Summary for RadFrac column using UNIFAC

Name C-105
Property method UNIFAC
Henry's component list ID
Electrolyte chemistry ID
Use true species approach for electrolytes YES
Free-water phase properties method STEAM-TA
Water solubility method 3
Number of stages 10
Condenser TOTAL
Reboiler KETTLE
Number of phases 2
Free-water NO
Top stage pressure [bar] 1.0132
Specified reflux ratio 0.239615637
Calculated molar reflux ratio 0.239615637
Calculated bottoms rate [kmol/hr] 580.254906
Calculated boilup rate [kmol/hr] 116.00656
Calculated distillate rate [kmol/hr] 320.015094
Condenser / top stage temperature [C] -84.4889649
Condenser / top stage pressure [bar] 1.0132
Condenser / top stage heat duty [cal/sec] -449494.076
Condenser / top stage reflux rate [kmol/hr] 76.6806206
Reboiler pressure [bar] 1.0132
Reboiler temperature [C] -7.06005877
Reboiler heat duty [cal/sec] 251312.2
Basis for specified distillate to feed ratio MOLE
Specified distillate to feed ratio 0.355465687
Basis for specified bottoms to feed ratio MOLE
Calculated molar boil-up ratio 0.199923445
Calculated mass boil-up ratio 0.139989746

14
Based on the result shown in Table 3.4.2.1.2, at actual number of stages of 10, the simulated
results are then used for proceeding column internal design.

3.4.2.2 Column Internals

The column internals using sieve-trayed column and bubble-cap column are simulated and
shown below.

Figure 3.4.2.2.1: Column Internals of Sieve-trayed Column

Figure 3.4.2.2.2: Summary of Column Internals of Sieve-trayed Column

15
 Figure 3.4.2.2.3: Column Internals of Bubble-cap Column

Figure 3.4.2.2.4: Summary of Column Internals of Bubble-cap Column

Based on Figure 3.4.2.2.2, sieve-trayed column exhibits weeping in stage 6 to stage 9.

Therefore, bubble-cap distillation column is suggested to be used as C-105. The internals of
the bubble-cap column is summarised in Figure 3.4.2.2.4 above. The simulation done shows
the diameter of the column is designed to be 1.11673m and the tray spacing is 0.6096m. The

16
column height is 4.8768m. The number of passes used in the internals design is 1 since our
column diameter is considered small; 2 or more passes are usually used in column of large
diameter in order to accommodate high liquid load. Also, it is recommended that the number
of passes should not exceed 2 as increasing the tray passes will shorten the path length, leading
to a decrease in tray efficiency. In addition, the cost of using multi-pass trays is high and may
negatively affect the process outcome since trays containing 2 or more passes are sensitive to
liquid and/or vapor mal-distribution if the flow is not symmetrical (“Plant Operation: Tray
Components,” n.d.).

3.5 Profiles Analysis

Figure 3.5.1: The Temperature Profile for Column C-105

Figure 3.5.1 depicts the temperature profile for the distillation column C-105 obtained from
the simulation by RadFrac in Aspen Plus. From the figure, it can be observed that the
temperature increases from Stage 1 (T = -84.4876 °C) which is at the top of the column to
Stage 9 (T = -7.0820 °C), at the bottom of the column. If we are to exclude the 1st stage and
final stage for condenser and reboiler, the temperature increases down the column from T = -
66.795 °C at Stage 2 to T = -25.8599 °C at stage 9. In both cases, the temperature generally
increases down the column. According to (Green & Perry, 2008), the temperature is the highest
at the bottom of the distillation column due to the reboiler duty, while the condenser duty is
accountable for the temperature being the lowest at the top section. The temperature at the
bottom should be lower than the boiling point of the heavier component while the temperature
at the top should be higher than the boiling point of the lighter component. These will help to

17
maintain the bottom product at liquid state and top product in gaseous phase (“Distillation
Fundamentals | Neutrium,” n.d.).

Figure 3.5.2: The Pressure Profile for Column C-105

Figure 3.5.2 illustrates the pressure profile down the distillation column C-105. It can be seen
that the pressure throughout the distillation column is constant. There is no pressure drop or
increase in pressure throughout the distillation column. Typically, there should be a pressure
gradient up the column, with the pressure at the bottom greater than that at the top. This
pressure gradient exists due to the liquid coming down the column impeding the vapor flow up
the column, thereby imposing a pressure loss on the flow. The pressure is held constant in
steady state distillation, while the temperature is varied to control the composition of the
product streams (“Distillation Fundamentals | Neutrium,” n.d.).

Figure 3.5.3: The Molar Flow Rate Profile for Column C-105

18
Figure 3.5.3 represents molar flow rate profile for the distillation column C-105 obtained from
the simulation by RadFrac in Aspen Plus. From the figure, it can be observed that the vapor
molar flow in the column remains almost at the same rate as it travels from stage 10 after being
released from the reboiler, up to stage 6. However, there is a sudden spike of the liquid flow
rate from stage 6 to stage 5. From then on, the vapor flow rate remains constant as it travels
upwards to stage 4 before it starts to increase slightly up to stage 2 even though the change is
negligible due to it being very small. The vapor molar flow rate starts to drop sharply as it flows
up to stage 1 due to the condensation of the vapor into liquid by the condenser. Meanwhile, the
liquid flows down the column from the top of the column at stage 1, at almost constant rate
until stage 5 before it starts to rise dramatically up to stage 6. This sudden spike is attributed to
the fact that stage 6 is the feed stage, and the vapor-liquid mixture feed is being fed into the
distillation column C-105, resulting in a sharp increase in both liquid and vapor flow rate. Then,
the liquid flow rate slowly increases until stage 9, then shows a decrease in stage 10. The
decrease in liquid flow rate is caused by the bottom product is being vaporised in the reboiler
and enters the column again. This is important to ensure the stage-by-stage equilibrium is
achieved (Ford, 2022).

3.6 Parametric Analysis (Sensitivity)

3.6.1 Parametric Analysis for DSTWU Column

Nomenclature:

TP1= ethylene; TP2 = BP1= hydrogen chloride; TP3 = acetylene; TP4 = BP2 = vinyl chloride;
BP1 = 1,2-dichloroethane; BP4 = 1,1,2-trichloroethane

Figure 3.6.1.1: Sensitivity Analysis Varying Reflux Ratio for DSTWU Column

19
For a distillation process, the defining criteria on whether the process is a well-designed process
is not only the degree of separation, but also construction cost, operating cost and maintaining
cost. These are related to condenser duty (CD), reboiler duty (RD) as well as the number of
stages (NS). Based on Figure 3.6.1.1, at 𝑅𝑅 = 1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 , CD and RD are at their lowest, which
means the operating cost and the maintenance cost are the lowest at 𝑅𝑅 = 1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 . However,
the number of stages decreases as the reflux ratio increases from 𝑅𝑅 = 1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 to 𝑅𝑅 = 1.
Higher number of stages will bring about a higher construction cost of the distillation column.
There is a trade-off to be made between construction cost and operating cost + maintenance
cost. Reflux ratio of 𝑅𝑅 = 1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 is chosen as it gives the best compromise between
construction cost and operating cost + maintenance cost.

3.6.2 Parametric Analysis for RadFrac Column

Figure 3.6.2.1: Sensitivity Analysis Varying Reflux Ratio for RadFrac Column

Figure 3.6.2.2: Sensitivity Analysis Varying Number of Stages for RadFrac Column

For RadFrac sensitivity analysis, the main parameters we have chosen to analyse are reflux
ratio (MOLE-RR) and the number of stages (NSTAGE). From Figure 3.6.2.1, as MOLE-RR
increases, the yield of the light key component (TP2=hydrogen chloride) at the top and that
of the heavy key component (BP2=VCM) at the bottom increase. This indicates that
increasing the reflux ratio is favourable for the separation process. However, the CD and RD

20
 also increase as MOLE-RR increases. This is a trade-off between product quality and the
operating cost. As product quality is of importance since product of higher purity will bring
in a lot for more revenue for the plant, the revenues made for producing products of higher
purity shall offset the increased operating cost due to increase in reflux ratio. Therefore, a
higher reflux ratio is recommended. Meanwhile, from Figure 3.6.2.2, as NSTAGE increases,
the yield of TP2 and BP2 also increase, meaning a higher number of stages favours the
distillation process and will achieve a higher degree of separation. At the same time, both the
magnitudes of CD and RD decrease as NSTAGE increases. This indicates that a higher
number of stages will bring about a lower operating cost, and this is beneficial for long term
operation. Thus, a column will higher number of stages should be used since the lower
operating cost in long term will offset the higher one-off construction cost of the column.

4.0 Conclusion

In order to separate high purity of vinyl chloride (99.5% VCM) as bottom product, from a
mixture containing ethylene, hydrogen chloride, acetylene, 1,2-dichloroethane and 1,1,2-
trichloroethane, a distillation column would be needed to achieve this desired yield. Therefore,
it is of utmost importance having a proper design of distillation column in order to achieve the
desired separation at optimum operating conditions and minimum cost.

Prior to any simulation done using Aspen Plus, a review of mass balance sheet provided was
done and a revised mass balance table was provided. Hydrogen chloride was chosen as the light
key component while vinyl chloride was chosen to be the heavy key component. Then, manual
calculation was done using the short-cut method which is also known as Fenske-Underwood-
Gilliland (FUG) method. The minimum number of stages and the actual number of stages
obtained using FUG method were obtained to be 3.1166 and 10.8555 respectively.

Then, with property method UNIFAC, Aspen Plus was used to simulate the short-cut (DSTWU)
distillation as well as the rigorous (RadFrac) distillation. The DSTWU simulation is similar to
the FUG method, but DSTWU distillation simulation method uses Winn equation to compute
the minimum number of stages instead of Fenske equation. The simulation results obtained
using DSTWU simulation gives minimum number of stages and actual number of stages with
values of 3.3312 and 9.3100 respectively. The small discrepancy in results of DSTWU
simulation from manual calculation might be attributed to the different degree of estimations

21
used by the two methods, different assumptions, different basis as well as different database
used. Then, by using the data obtained from DSTWU distillation simulation, RadFrac
distillation is simulated. The column to be used was determined to be bubble-cap distillation
column owing to no weeping occur in the column. The column diameter and the tray spacing
were designed to be 1.11673 m and 0.6096 m respectively. The column height is simulated to
be 4.8768m. Since the column diameter is considered small, the number of passes used in the
internals design is 1.

Next, optimisation of distillation column design can be done by conducting parametric analysis
(sensitivity) for both DSTWU and Radfrac model. The analysis was done by investigating the
effect of varying reflux ratio on the DSTWU model, and the varying reflux ratio and number
of stages for the RadFrac model. For the DSTWU model, at 𝑅𝑅 = 1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 , CD and RD are at
their lowest, which means the operating cost and the maintenance cost are the lowest at 𝑅𝑅 =
1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 . However, the number of stages decreases as the reflux ratio increases from 𝑅𝑅 =
1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 to 𝑅𝑅 = 1. Higher number of stages will bring about a higher construction cost of the
distillation column. Reflux ratio of 𝑅𝑅 = 1.5𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 is chosen as it gives the best compromise
between construction cost and operating cost. On the other hand, for Radfrac model, as the
reflux ratio increases, the yield of the heavy key component (BP2=VCM) at the bottom increase.
This indicates that increasing the reflux ratio is favourable for the separation process. However,
the condenser duty and reboiler duty also increase as MOLE-RR increases, which means higher
operating cost. As product of higher purity will generate far more revenue for the plant, the
revenues made for producing products of higher purity shall offset the increased operating cost.
So, a higher reflux ratio is recommended. Besides, as number of stages increases, the yield of
vinyl chloride also increases, meaning a higher number of stages favours the distillation process
and will achieve a higher degree of separation. At the same time, both the magnitudes of
condenser duty and reboiler duty decrease. This indicates that a higher number of stages will
bring about a lower operating cost, and this is beneficial for long term operation.

In short, the distillation column design parameters are always linked to operating cost.
Therefore, additional research should be done in the future in order to further optimise the
performance of a distillation column and lower the operating cost at the same time.

22
5.0 References

1. Abu Bakar, M.Z. (2022). Description of the Process.

2. Abu Bakar, M.Z. (2022). Process Flow Diagram (PFD).
3. Alchetron. (2017). Vinyl chloride, The Free Social Encyclopedia. Alchetron.com. [online].
Available at: https://alchetron.com/Vinyl-chloride [Accessed 21 Jan 2022].
ChEBI. (2021). chloroethene (CHEBI:28509). [online] Available at:
https://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI:28509 [Accessed 26 Jan. 2023].
4. Distillation Fundamentals | Neutrium [WWW Document], n.d. URL
https://neutrium.net/unit-operations/distillation-fundamentals/ [Accessed 22 Jan. 2023).
5. Ford, J. (2022) Distillation Columns – Internal Reflux Control, Control Engineering.
Available at: https://www.controleng.com/articles/distillation-columns-internal-reflux-
control/ (Accessed: January 28, 2023).
6. Green, D.W. and Perry, R.H. (2008) “13,” in Perry's Chemical Engineers' handbook. 8th
edn. New York etc.: McGraw-Hill.
7. Hussain, Y., n.d. Equilibrium Separation Column. Jordan University of Science (JUST)
[online] Available at:
https://www.just.edu.jo/~yahussain/files/Equilibrium%20Separation%20Columns.pdf
[Accessed 21 Jan. 2023]
8. Internals, A.P.T. (2018). What is Distillation and Why is it Important? [online] AMACS.
Available at: https://amacs.com/distillation/what-is-distillation-and-why-is-it-important-
2/#:~:text=The%20distillation%20process%20is%20used [Accessed 22 Jan. 2023].
9. National Cancer Institute. (2022). Vinyl Chloride - Cancer-Causing Substances - National
Cancer Institute. [online] Available at: https://www.cancer.gov/about-cancer/causes-
prevention/risk/substances/vinyl-
chloride#:~:text=Vinyl%20chloride%20exposure%20is%20associated [Accessed 26 Jan.
2023].
10. New Jersey Department of Health. (2015). Hazardous Substance Fact Sheet. [online]
Available at: https://nj.gov/health/eoh/rtkweb/documents/fs/2001.pdf [Accessed 26 Jan.
2023].
11. Plant Operation: Tray Components [WWW Document], n.d. URL
http://www.separationprocesses.com/Operations/POT_Chp02e.htm [aAccessed 22 Jan.
2023].

23
12. PubChem (2022). Vinyl chloride. [online] Available at:
https://pubchem.ncbi.nlm.nih.gov/compound/Vinyl-chloride [Accessed 21 Jan. 2023].

24
Appendices

A. Mass Balance Sheet Provided

STREAM 35

i
B. Stream Results of RadFrac Simulation

ii
C. Stream Results of RadFrac Simulation

iii