CENG 122 Vinyl Chloride Production Project

Analysis and Design of Separation Units

Introduction:
Polyvinyl chloride (PVC) is one of the worlds most versatile thermoplastics, having a wider
range of uses than any other plastic material. It can be made as a hard plastic, used in pipes and
window fixtures, or can be made small, to be used in clothing, hoses, flooring, etc. Not only is it
versatile, but it is cheap, nontoxic, and inert. The feedstock for PVC production is the vinyl
chloride monomer (VCM). Due to the extensive use of PVC, VCM production is one of the top
20 largest petrochemical industries in the world. Approximately 13 billion kg of VCM are
produced annually, mostly in the U.S. and China, and production is only expected to increase as
PVC demand increases.

Your Assignment:
Analyze the provided plant design to determine the amount of vinyl chloride product that can be
produced under the conditions provided. You will take various process streams, following
reactors, recycling, purging, etc., and determine the best way to separate and purify desired
chemical components. We will use our knowledge from CENG 122 to see if it is possible to
significantly increase the purity of our process streams or our final vinyl chloride product.

Project Goal:
The purpose of this project is to modify the basic assumptions and calculations we made
throughout your sophomore year in CENG 100, 102, and 113. You will compare how this plant
operates when separation is not perfect, and generally inefficient. Along with CENG 120, this
will be the last time you are exposed to this project - having now seen it from effectively every
angle in terms of chemical engineering.

Page !1 of 11
!
Process Description:
The first large-scale plants producing vinyl chloride utilized the hydrochlorination of acetylene.
However, since the mid 1950s, production from ethylene (as this precursor became more readily
available) has become the primary synthesis route. Now, 95% of all vinyl chloride is produced
from a series of reactions starting with chlorine, ethylene, and air. Ethylene (C2H4) and chlorine
(Cl2) directly react to produce 1,2-dichloroethane (C2H4Cl2, DCE). The by-product 1,1,2-
trichloroethan (C2H3Cl3, TCE), can also be formed, but is highly undesired. The DCE
intermediate is then sent to a cracking furnace where it is pyrolyzed to vinyl chloride monomer
(VCM) and hydrogen chloride (HCl). So the HCl does not go to waste, it is separated and sent to
another reactor where it reacts with additional fresh ethylene and oxygen to produce additional
DCE and water. The additional DCE is combined with the original DCE when fed to the
pyrolysis furnace. The product is separated, with the DCE recycled and the VCM collected as
product for later processing into PVC.
Process Steps:
Direct Chlorination of Ethylene:
Fresh ethylene gas is typically fed with purified chlorine gas (in slight excess) into the direct
chlorination reactor (R1), where the following reaction takes place:
Direct Chlorination (DC1): C2H4 + Cl2 C2H4Cl2 (1)
TCE By-Product Formation (DC2): C2H4 + 2 Cl2 C2H3Cl3 + HCl (2)
This reaction is typically conducted with an iron(III) chloride catalyst. The reaction is
exothermic and is typically cooled by use of cooling water.
In our system, the reactor operates at a constant P = 10 bar and T = 183 C.
The reaction goes almost to completion, thus requiring no separation or recycle.
Prior to entering the reactor, the streams are combined, pre-heated and compressed.
Oxychlorination of Ethylene:
Additional fresh ethylene is fed with air (or pure oxygen) and recycled HCl from later in the
process into the oxychlorination reactor (R2). The following reaction takes place:
Oxychlorination (OC1): C2H4 + 0.5 O2 + 2HCl C2H4Cl2 + H2O (3)
TCE By-Product Formation (OC2): C2H4Cl2 + HCl + 1/2O2 C2H3Cl3 + H2O (4)
Combustion (OC3): C2H4 + 3 O2 2 CO2 + 2 H2O (5)

Page !2 of 11
!
This reaction is highly exothermic and requires significant cooling to prevent runaway
temperatures and unwanted side-reactions.
The reaction takes place over a copper(II) chloride catalyst.
In our system, the reactor operates at a constant P = 10 bar and T = 260C.
The recycled HCl is provided entirely from the subsequent reactor.
The oxygen is fed such that the HCl is in 2% excess so all oxygen can be consumed.
In reality, some side-reactions inevitably take place that produce poly-chlorinated alkanes.

DCE Separation and Purification:

The effluent from R2 is sent through a separator to vent off N2 and any other gaseous inerts. A
minimal amount of reactants/products are lost at this point. This separator also collects and
recycles any solid catalyst from R2 that may have passed out of the reactor. Once the two reactor
effluent streams are combined, they are passed through a scrubbing column, where the products
come in contact with a dilute NaOH solution to remove chloral and various other chlorinated
byproducts. After the addition of a DCE recycle stream, the materials enter a series of separator
columns to remove all materials other than DCE and VCM.
NOTE: Because we are not accounting for all chlorinated and polymerized by-products, we will
not model the scrubbing and dehydration sections.

Pyrolysis of Dichloroethane:
The resulting reactor feed enters a pyrolysis furnace operating at a high temperature and pressure
where the DCE decomposes into vinyl chloride and HCl.
DCE Pyrolysis/VC Synthesis (P1): C2H4Cl2 C2H3Cl + HCl (6)
Vinyl Chloride Degradation (P2): C2H3Cl C2H2 + HCl (7)
De-chlorination (P3): C2H4Cl2 C2H4 + Cl2 (8)
Conversion is moderate requiring recycle of all unused DCE for higher overall conversion.
The reaction is highly endothermic, needing to be carried out in a fired heater. This heat is
provided by the combustion of methane.
In our system, the reactor operates at a constant P = 20 bar and T = 500C.
Since VCM is not very stable at the operating temperature, the residence time of the reactor is
kept low and the effluent stream must be immediately quenched.
Page !3 of 11
!
Final VCM Purification:
Subsequently, HCl is stripped out of the R3 effluent and recycled to the oxychlorination reactor.
The bottoms stream of the stripping column is sent to a series of distillation columns where the
final purified VCM product is collected. The DCE-rich bottoms streams are recycled to the
pyrolysis furnace. In our process, since we will finally learn about various separation techniques,
we will compare the single flash column presented in the diagram and replace it with something
that can yield better separation.

Energy Considerations:
Energy integration is of key importance in modern vinyl chloride synthesis plants. As you design
your separation columns, you will want to take the heat duty of the equipment into account.

Process Flow Diagram:

See the following pages for the plant design and a list of specific design features.

Page !4 of 11
!
 HCl Recycle (R)
Air (A)
Oxy-
Chlorination E2
Reactor (R2)
Vent
Separator
Gases (V)
Ethylene (F)
S
Direct
Chlorination VCM
Reactor (R1) E1 Dilute

Page !5 of 11
Chlorine (C) Product (P)
NaOH (N)

!
Note: Stream labels are provided in the diagram.
T G

Separator
DCE

Dryer &
E3

Scrubber

Stripper

Column
L RF
Cracking
Furnace (R3)
H B
W J
DCE Recycle (D)
Process Design Features:
Composition Exiting Reactor R2 (Oxychlorination Reactor) & entering Separator:

Species Reference Label Mole Fraction

C2H4 (ethylene) A 0.0675
HCl B 0.1289
O2 C 0.0949
N2 D 0.527699
1,2-dichloroethane (DCE) E 0.0899
1,1,2-trichloroethane (TCE) F 0.0004
H 2O G 0.0907
CO2 H 1E-06

The total flow rate exiting R2 and entering the Separator: 285 kmol/hr
Feed Conditions: T = 91C and P = 3 bar

Composition entering stripping column:

Species Reference Label Mole Fraction

1,2-dichloroethane (DCE) 1 0.21348
Vinyl Chloride (VC) 2 0.4648
HCl 3 0.3212
C2H2 (acetylene) 4 0.0005
C2H4 (ethylene) 5 1E-05
Cl2 6 1E-05

The total flow rate entering the Stripper: 270 kmol/hr

Conditions of stream exiting reactor R3: T = 578C and P = 30 bar
NOTE: Since VC is unstable at this temperature and pressure AND the stream needs to be
(mostly) liquid, you can drop the inlet temperature and pressure as much as you see fit. The only
limitations: (1) virtually all the DCE and VC must be fed in the liquid phase (there may be a little
in the vapor due to thermodynamics; and (2) you dont want the heat duty of the stripper to be
excessively large - hence dont cool the feed stream TOO MUCH.

Page !6 of 11
!
Composition entering final separation unit(s), labeled as Column (regardless of your results for
the stripping column):

Species Reference Label Mole Fraction

1,2-dichloroethane (DCE) 1 0.3117
Vinyl Chloride (VC) 2 0.6574
HCl 3 0.0152
N2 4 0.0157

The total flow rate entering the Column: 180 kmol/hr

Conditions of stream entering Column: T = 50C and P = 12 bar

PROJECT:
Model the three separation units separately in Aspen Plus. Answer the questions on the following
pages. Prior to Thanksgiving, a sheet will be provided that you can print out and simply write
down your final answer in the provided locations.

Most complicated calculations will be modeled in Aspen, requiring little to show for work.
For Aspen work, show a print-out of your process (called a Process Flow Diagram = PFD).
Include on that sheet all the relevant information: flow-rate, composition, temperature, and
pressure of each stream, as well as the heat duty of the unit.
For sizing calculations, provide hand-written sample calculations for each part. In other words,
for streams with seven components (as an example), write out and solve a species/material
balance for at least one component.

Page !7 of 11
!
GROUPS:
You are allowed (AND ENCOURAGED) to work on this project in groups as large as 5 or 6.
Groups CANNOT be any larger. Also, NO ONE can work alone. This is supposed to be a
collaborative effort, so you have to find at least one other person to work with. All group
members are expected to contribute equally and the work provided should label clearly which
person contributed to what parts.
Each group is expected to turn in a single completed project.

NOTE: LEGIBILITY AND CLEAR ORGANIZATION OF SOLUTIONS ARE NECESSARY

FOR PROPER GRADING.

WORTH: 10% of the overall course grade. Although this project shouldnt actually take a
considerable amount of time, DO NOT WAIT UNTIL THE LAST MINUTE TO START IT.
There are only a limited number of Aspen licenses (as youll probably already know going
through the quarter). If too many people try to sign in at the last minute to work on the project,
many people will not get access. So, as soon as you think you are capable of completing any part
of the project, get it done earlier, rather than later!

DUE: Thursday, December 1, beginning of class.

Page !8 of 11
!
PROJECT ANALYSIS:
Part I: Choosing a Thermodynamic Model
Choose a single thermodynamic model that you will use for all three separators. EXPLAIN
BRIEFLY WHY THAT IS LIKELY THE BEST MODEL TO USE.

Part II: Modeling the separator (to vent off the light gases after reactor R2)
Here, we will compare to see what is best for the desired separation:
Partial condensation (flash)? Distillation? Absorption?
Model the separation with each piece of equipment (separately) and choose which is likely the
best for your needs. If you can explain using logic why 1 should not be modeled, an explanation
will suffice (at best you can try to eliminate 1 option, not 2).
Goal: Removal of most the light gases with as little loss of DCE as possible. Partial removal of
TCE would be nice, but may not be possible with our models. Try to keep the heat duty low!
(1.) Choose your column type.
(2.) Choose a temperature or determine the temperature profile through your column. If you use
flash, just state the temperature used. Provide the plot (if relevant).
(3.) Choose an operating pressure. If you model pressure drop, mention it.
(4.) Provide the flow rate and composition of all streams
(5.) Size the column: number of stages, approximate overall (actual) height, diameter. If you
account for tray efficiency (if relevant), mention it here.
(6.) Provide the heat duty for your given conditions

Part III: Modeling the stripper

Here, we will compare to see if our original assumption of complete removal of HCl was
accurate. We will only be modeling a stripping column.
Model the stripping column, choosing one of 3 possible stripping agents: pure N2, pure O2, or air
(modeled as 21% O2 and 79% N2). You should be able to quickly determine if any of these yield
significantly different results. If not, use of air would likely be considerably cheaper than the
other options.
Goal: Removal of as much HCl and the light gasses, while losing as little of the DCE and VC as
possible. Try to keep the heat duty low!
Page !9 of 11
!
Caution: evaporation and heat/mass transfer could lead to unexpected cooling of some streams -
dont let them get unrealistically cool
(1.) Choose your stripping agent. Choose whatever feed conditions for this stream as you see fit.
(2.) Determine the feed conditions of the process stream, E3 on the diagram. This means T & P.
Again, choose wisely so you dont accidentally make the heat duty excessive. The feed must be
such that the DCE and VC are almost entirely liquid.
(3.) Determine the temperature profile through your column. Provide the plot.
(3.) Choose an operating pressure. If you model pressure drop, mention it.
(4.) Provide the flow rate and composition of all streams
(5.) Size the column: number of stages, approximate overall (actual) height, diameter. Include
tray efficiency here if you use it.
(6.) Provide the heat duty for your given conditions
(7.) You are unlikely to model the system to look anything like our ideal system in CENG 100. A
lot of HCl will remain in the liquid, and a lot of DCE & VC will be lost in the gas. Explain:
(a.) Why would excessive residual HCl going into the column be a problem?
(b.) Why would recycling excessive DCE and VC back to the beginning of the process be
a problem?
(c.) Any recommendations to further purify the process stream, before entering the
column? No models necessary - just some ideas.

Part IV: Modeling the column (to obtain a high purity VC product)
Here, we will compare to see what is best for the desired separation:
Partial condensation/reboiling (flash)? Distillation? Extraction? Other?
EXTRA CREDIT: A combination of these?
Model the separation with each piece of equipment (separately) and choose which is likely the
best for your needs. If you can explain using logic why 1 should not be modeled, an explanation
will suffice (but there should be at least 2 competing models).
Goal: Production of as high a purity VC product as possible at as high a flow rate as possible.
Note: getting 99.9% purity but only .0001 mol/s vs. 98% at 100 mol/s: which do you think is
better?

Page 10
! of 11
!
(1.) Choose your column type. If using extraction, what is the solvent & the inlet conditions? If
using absorption/stripping, what is the absorbing/stripping agent and inlet conditions?
(2.) Choose a temperature or determine the temperature profile through your column. If you use
flash, just state the temperature used. Provide the plot (if relevant).
(3.) Choose an operating pressure. If you model pressure drop, mention it.
(4.) Provide the flow rate and composition of all streams
(5.) Size the column: number of stages, approximate overall (actual) height, diameter. If you
account for tray efficiency (if relevant), mention it here.
(6.) Provide the heat duties for your given conditions

EXTRA CREDIT: In an effort to get even higher purity, or reduce the size of your column, or
reduce the heat duty (you may have to pick and choose which can be improved at the sacrifice of
the other(s)), you can have a separation sequence/train: a series of separation columns.
Up to you what to try, but some ideas:
(a.) (1) Small flash unit to remove trace elements + (2) Large flash or Distillation for VC/DCE
separation
(b.) 2 Distillation columns in series, taking the high purity VC stream from the first and sending
to the 2nd to purify further (likely to help reduce individual column size/heat duty)
(c.) Extraction of VC into a solvent that can be separated more easily in a distillation column
(d.) Extraction of DCE out of the process stream, to feed a higher purity VC stream into a
distillation column.

Note: For all of these Parts, there is no single correct answer. That includes the extra credit.
You just want to optimize separations.
However: once your flow rates are set, then sizing diameters has a specific protocol - that part is
not simply guess-and-check.

FINAL NOTE: When setting the number of stages for any equipment, you can choose to
determine this mathematically, or simply randomly choose a value. Just remember: you want to
get the best separation but the column and heat duties cant be too large/excessive.
Page !11 of !11