INDIVIDUAL REPORT - Keshrie Reddy1 PDF

Faculty of Engineering & the Built Environment
Department of Chemical Engineering
Bellville Campus
Project Title: Production of Styrene from Ethylbenzene
Course: BTech: Chemical Engineering
Module: PQD400S Chemical Engineering Process Design 4
Lecturer: Mr T. F. Madzimbamuto
Name and Surname: Keshrie Reddy
Student number: 216282772
Due Date: 9 September 2019
Declaration: I certify that this report is my own unaided work, except for the assistance
received by the teaching staff. I undertake not to pass this report on to any other student.
Date: ____________ Signed: ____________

Synopsis
This report includes a preliminary feasibility study and a compilation of the detailed designs of the
major equipment for the plant which is to produce 280 000 tonnes per year of styrene from
ethylbenzene.
A mini project was done to perform a preliminary evaluation on the economics of the production of
styrene by dehydrogenation of ethyl benzene. In order to achieve this, the operational conditions and
economic potential of this process was investigated. The conditions which yielded the highest profit
and met the reaction requirements (of adding steam necessary to add heat for the endothermic
reaction) was 900 K (temperature), 0.6 bar (pressure) and a steam to ethylbenzene ratio of 6:1. Under
these conditions, an ethyl benzene feed of roughly 43 000 kg/hr is required to produce 280 000 ton/yr
of styrene and a preliminary gross profit of approximately R199 million can be obtained annually. It
was therefore recommended to investigate further into this prospective plant.
Heat exchanger, E-501 as shown on the PFD of the plant attached in the Appendix, was designed. The
purpose of this heat exchanger is to exchanger heat from the hot high pressure steam fluid to the cold
hydrocarbon fluid – ethylbenzene with small amounts of styrene. This serves as part of the preheat
before the reactor. The designing of E-501 is discussed step by step, with calculations and all data
shown along with references. From the second iteration, it was found that with a two pass exchanger,
448 tubes would be needed with 8 mm outer diameter tubes, 1.83 m in length. The shell diameter was
calculated to be 0.33 m with baffle spacing of 0.1 m. Although a close baffle spacing will increase the
heat transfer coefficients, this would also cause a higher pressure drop in the exchanger, which was
calculated to be approximately 3630 kPa. The Purchase cost of the heat exchanger is 19 650 USD.
An isothermal reactor was chosen with a shell and tube design in order to help maintain a constant
optimum temperature, of 900 K, inside the tubes where the endothermic reaction occurs. The
isothermal reactor was also found to give higher conversion and be more economically operable. The
pressure drop of the reactor was found to be 0.09 bar while the optimum operating reactor pressure
was 0.4 bar. The percentage pressure drop was therefore 22.9%. Since the calculated pressure drop
along the reactor is greater than 10%, this design will not be economically operable. Different
dimensions can therefore be considered. Dp can be decreased, as long as the ratio of Dt/ Dp remains
greater than 15. The shell-side pressure drop of the reactor was however found to be 0.18(10 -5) bar.
With the reacting mixture being on the tube side of the reactor, the tube volume should be greater than
the reactor volume, as per Silla (2003). This was the case as the tube volume was 6.85 m3 and the
reactor volume was 6.77 m3.
i
A distillation column to be used in the styrene production plant was designed. In order to do this, the
composition of the feed stream to the column was determined along with the best operating conditions
for the column. The feed stream to the column contained ethylbenzene (13%) and styrene (87%) with
trace amounts of toluene. The mixture was therefore regarded as binary because according to Thakore
& Bhatt (2007:381), if the amounts of other components in the mixture are very low, the mixture can
be considered binary. The operating pressure and temperature was chosen to be 0.2 bar and 358 K. 0.2
bar in order to reduce the bubble point of the distillate (ethylbenzene) to 360 K. This lower pressure
will also reduce condensation costs of ethylbenzene in the condenser with cooling water at 303.15 K.
A sieve tray tower was selected as the best option as the column diameter was found to be greater than
1m and operated at low vacuum, where the use of trays improve vapour-liquid contact in large
diameter towers. Sieve trays are also the most economical choice and, according to Coulson &
Richardson (2005:548), the efficiency of a sieve plate used for an ethylbenzene-styrene system is
75%. In terms of the detailed tower design, the reflux ratio was found to be 18, the number of trays
needed was calculated to be 28 trays, the decided tray spacing was 0.5 m and the weir height was 10
mm. The top and bottom pressure specifications were 0.2 bar and 0.25 bar respectively while the
column diameter was calculated to be 7.7 m at the top of the column and 6.6 m at the base of the
column.
Pump P-501, the feed pump to plant 5, which pumps ethylbenzene from a storage tank (Tk-001) to a
mixer (M-501), was also sized. Carbon steel was selected as the material of construction, based on the
theory found in Sinnott & Towler (2008). To overcome the differences in pressure and height, static
pressure and dynamic loss was calculated. This information contributes to the pressure head that must
be supplied by the pump in a piping system. Energy balance was used to derive an equation from
which to plot the system curve. The system curve was plotted against the pump curve to find the best
point of operation. The flow rate and head for the system were therefore found to be 0.008 m3/s and
25 m/s respectively. A single stage centrifugal pump was selected based on the best operating
pressure head of the pump. The net positive suction head of this pump was calculated to be 13.4 m. A
characteristic curve for a centrifugal pump graph was used to estimate the pump efficiency and
impeller diameter, based on the pump’s capacity and head. The efficiency was found to be 61 % while
the required impeller diameter was found to be 130 mm. The pump power was then calculated to be
2.6 kW. The required pump power was divided by the pump efficiency to estimate the actual pump
power, of 4.16 kW.
ii
Nomenclature
Symbol/Abbreviation Description
𝐶𝑛 Concentration of species
𝐶𝑝𝐹 Specific heat capacity of feed
𝐹𝑛 Molar flow rate of species
𝐻𝑖 Henry’s law constant
𝑄̇ Volumetric flow rate
𝑄ℎ Duty of hot fluid
𝑄𝑐 Duty of cold fluid
𝑘𝑙𝑖𝑞 Thermal conductivity of a liquid
𝑚̇ Mass flow rate
𝑛̇ Molar flow rate
𝑟𝐴 Rate of reaction
𝑡𝐵𝑃 Bubble point temperature
𝑡𝐹 Feed temperature
𝑥𝑖 Liquid concentration
𝑦𝑖 Gas concentration
𝜀1 Extent of reaction 1 (Primary reaction)
𝜀2 Extent of reaction 2 (Side reaction)
𝜆𝐹 Latent heat of feed
∆𝑇𝐿𝑀 Log mean temperature difference
∆𝐻 Heat of vaporisation
∆𝑇 Change in temperature
∝ Average relative volatility
Ac Cross-sectional area
As Surface area
B Bottoms/Residue molar flow rate
cat Catalyst
Cp Heat capacity
CP Cost Price
D Distillate molar flow rate
De Equivalent diameter
do Outer diameter
e Pipe roughness
EB Ethylbenzene
EB(P) Ethyl Benzene reacting in primary reaction
EB(S) Ethyl Benzene reacting in side reaction
iii
EBin Moles Ethyl Benzene entering reactor
EBout Moles Ethyl Benzene exiting reactor
Eq. Equation
F Feed molar flow rate
f Fanning friction Factor
F To Molar flow rate of total moles
FEB0 Molar flow rate of initial amount of ethyl benzene
fi Tube side fouling factor
fo Shell side fluid fouling factor
Fsteam0 Molar flow rate of initial amount of steam
ft Feet
g Gravitational acceleration
G Mass flow rate
g Gram(s)
H Hydrogen
HD System head
hr Hour
JN Tube side heat transfer factor
K Thermal conductivity
K Kelvin
Keq Equilibrium constant at equilibrium conditions
kg Kilogram
kmol Kilomoles
L Length
lB Baffle spacing
M Methane
Mill Million
mol Moles
Mr Molar mass
MW Molecular weight
n Number of passes
ƞ Efficiency
NPSH Net positive suction head
Nr. Number
Nt Number of tubes
P Pressure
P1 Pressure at point 1
P2 Pressure at point 1
Patm Atmospheric pressure
iv
PFD Process Flow Diagram
PGP Preliminary Gross Profit
ppm(wt) Parts per million(weight)
Pt Tube pitch
Pv Vapour pressure
q Heat
Q Capacity
R Ideal gas constant
Re Reynolds number
S Solubility
S Styrene
SP Selling Price
T Temperature
TB Boiling point temperature
TC Critical temperature
Tci Cold fluid temperature in
Tco Cold fluid temperature out
TF Freezing point temperature
Thi Hot fluid temperature in
Tho Hot fluid temperature out
Tmax Maximum temperature
Tmin Minimum temperature
Tol Toluene
ton Tonnes
U0, ass Assumed transfer coefficient
Uo, calc Calculated transfer coefficient
v Velocity
V Volume
v1 Velocity at point 1
v2 Velocity at point 2
W Catalyst weight
W Water
X Conversion
xB Mole fraction of more volatile component in bottoms/residue
xD Mole fraction of more volatile component in distillate
y Mole fraction
yEB Mole fraction of Ethyl Benzene
yH Mole fraction of Hydrogen
v
Yr Year
yS Mole fraction of Styrene
Z1 Height at point 1 from datum
Z2 Height at point 2 from datum
zF Mole fraction of more volatile component in feed
𝐶𝑝𝑔 Heat capacity of gas
𝐶𝑝𝑙 Heat capacity of liquid
𝑥 Pressure ratio
𝜀 Extent of reaction
𝜇 Viscosity
𝜋 Pi
𝜌 Density
𝜏 Residence time
𝜑 Catalyst void fraction
vi
Table of Contents
Synopsis ......................................................................................................................................... i
Nomenclature ............................................................................................................................... iii
Chapter 1 MINI PROJECT ................................................................................................................ 1
Summary ....................................................................................................................................... 1
1.1. Problem statement .................................................................................................................. 1
1.2. Introduction ............................................................................................................................ 1
1.3. Data ........................................................................................................................................ 2
1.3.1 Calculated Data ................................................................................................................. 2
1.3.2. Material Balance .............................................................................................................. 4
1.3.3. Costing ............................................................................................................................ 5
1.4. Discussion ............................................................................................................................ 10
1.5. Conclusion............................................................................................................................ 12
1.6. Recommendations ................................................................................................................. 12
1.7. References ............................................................................................................................ 13
Chapter 2 PROCESS DESCRIPTION ............................................................................................. 14
2.1. Introduction .......................................................................................................................... 14

2.1.1. Feed preparation to the reactor ....................................................................................... 14
2.1.2. Ejector and coolers ......................................................................................................... 14
2.1.3. Flash Separation ............................................................................................................. 15
2.1.4. Distillation Columns ...................................................................................................... 15
Chapter 3 HEAT EXCHANGER DESIGN ...................................................................................... 16
Summary ..................................................................................................................................... 16
3.1. Introduction .......................................................................................................................... 16
3.2. Process Flow Diagram .......................................................................................................... 16
3.3. Design of Heat Exchanger ..................................................................................................... 17
3.3.1. Specification .................................................................................................................. 17
3.3.2. Physical Properties of Fluids .......................................................................................... 19
3.3.3. Assume Value of Overall Coefficient Uo,assume ........................................................... 21
3.3.4. Decide number of shell and tube passes. ......................................................................... 21
3.3.5. Determine heat transfer area required ............................................................................. 22
3.3.6. Decide type, tube size, material layout and assign fluids to shell or tube side .................. 22
3.3.7. Calculate number of tubes .............................................................................................. 23
3.3.8. Calculate shell diameter ................................................................................................. 23
3.3.9. Estimate tube-side heat transfer coefficient ..................................................................... 24
3.3.10. Decide baffle spacing and estimate shell-side heat transfer coefficient .......................... 24
3.3.11. Calculate overall heat transfer coefficient including fouling factors, Uo,calc .................... 26
3.3.12. Estimate tube- and shell-side pressure drops ................................................................. 26
3.3.13. Cost Estimation of Heat Exchanger .............................................................................. 27
3.4. Conclusion............................................................................................................................ 27
3.5. References ............................................................................................................................ 27
Chapter 4 REACTOR DESIGN ....................................................................................................... 28
Summary ..................................................................................................................................... 28
4.1. Problem Statement ................................................................................................................ 28
4.2. Introduction .......................................................................................................................... 28
4.3. Literature Review ................................................................................................................. 29
4.3.1. Continuous and Batch Processing ................................................................................... 29
4.3.2. Types of Reactors........................................................................................................... 29
4.3.3. Homogeneous and Heterogeneous Reactions .................................................................. 31
4.3.4. Adiabatic versus Non-adiabatic Reactor ......................................................................... 31
4.4. Design of Reactor ................................................................................................................. 34
4.4.1. Performance Equation .................................................................................................... 36
4.4.2. Initial Concentration of Ethylbenzene ............................................................................. 36
4.4.3. Reaction Rate ................................................................................................................. 37
4.4.4. Catalyst Weight .............................................................................................................. 39
4.4.5. Catalyst Bed Volume ..................................................................................................... 40
4.4.6. Space Time .................................................................................................................... 40
4.4.7. Reactor Volume ............................................................................................................. 40
4.4.8. Number of Tubes ........................................................................................................... 40
4.4.9. Reactor Shell Height ...................................................................................................... 40
4.4.10. Reactor Diameter ......................................................................................................... 41
4.4.11. Length to Diameter Ratio ............................................................................................. 41
4.4.12. Pressure Drop ............................................................................................................... 41
4.4.13. Tube Volume ............................................................................................................... 43
4.4.14. Shell-side Pressure Drop .............................................................................................. 43
4.5. Discussion ............................................................................................................................ 45
4.6. Conclusion............................................................................................................................ 47
4.7. Recommendation .................................................................................................................. 47
4.8. References ............................................................................................................................ 47
Chapter 5 SEPARATION UNITS DESIGN ..................................................................................... 49
Summary ..................................................................................................................................... 49
5.2. Introduction .......................................................................................................................... 49
5.3.1. Flash Column ................................................................................................................. 50
5.3.2. Distillation column ......................................................................................................... 52
5.4. Separation Column Design .................................................................................................... 54
5.4.1. Flash Column ................................................................................................................. 54
5.4.2. Distillation Column (T-502) ........................................................................................... 55
5.5. Discussion ............................................................................................................................ 69
5.6. Conclusion............................................................................................................................ 70
5.7. Recommendations ................................................................................................................. 71
5.8. References ............................................................................................................................ 71
Chapter 6 Pump Sizing .................................................................................................................... 73
Summary ..................................................................................................................................... 73
6.2. Introduction .......................................................................................................................... 73
6.3.1. Pump Sizing ................................................................................................................... 74
6.3.2. Types of Pumps.............................................................................................................. 74
6.3.3. Pump and System Curves ............................................................................................... 75
6.3.4. NPSH............................................................................................................................. 76
6.3.5. Pump Efficiency and Impeller Diameter ......................................................................... 76
6.3.6. Pump Power ................................................................................................................... 77
6.4. Pump Sizing Results and Discussion ..................................................................................... 77
6.4.1. System and Diagram ...................................................................................................... 77
6.4.2. Unit Spacing .................................................................................................................. 78
6.4.3. Fluid Properties .............................................................................................................. 79
6.4.4. Material of Construction and Pipe Diameter ................................................................... 80
6.4.5. Flow Rates ..................................................................................................................... 81
6.4.5. Friction Factor................................................................................................................ 81
6.4.6. System Curve ................................................................................................................. 82
6.4.7. Pump Curves .................................................................................................................. 85
6.4.8. Type of Pump ................................................................................................................ 85
6.4.9. NPSH............................................................................................................................. 86
6.4.10. Pump Efficiency and Impeller Diameter ....................................................................... 87
6.4.11. Power Required by the Pump ....................................................................................... 87
6.5. Conclusion............................................................................................................................ 87
6.6. List of References ................................................................................................................. 88
Appendix A ................................................................................................................................. 90
Appendix B ............................................................................................................................... 100
Appendix C ............................................................................................................................... 104
Appendix D ............................................................................................................................... 110
Appendix E ............................................................................................................................... 113
Appendix F................................................................................................................................ 122
List of Figures
Figure 1.1: Relationship between Fractional Selectivity to Toluene and Temperature ......................... 3
Figure 1.2: Relationship between Conversion and Pressure at Varying Temperatures with Steam Ratio
12:1 ................................................................................................................................................... 3
Figure 1.3: Relationship between Profit and Steam Ratio at Varying Pressures with Temperature
900K ................................................................................................................................................. 6
Figure 1.4: Relationship between Profit and Steam Ratio at Varying Pressures with Temperature
950K ................................................................................................................................................. 7
Figure 3.1: PFD of E-501 ................................................................................................................ 16
Figure 3.2: Temperature profile ....................................................................................................... 21
Figure 3.3: Segmental baffle (Coulson & Richardson, 2005:651) ..................................................... 24
Figure 4.1: Styrene selectivity (σ) as a function of ethylbenzene conversion (x_EB) (Savoretti et al.,
1998:210). ....................................................................................................................................... 32
Figure 4.2: Temperature profiles for adiabatic and non-adiabatic (isothermal) reactors (Savoretti et al.,
1998:210). ....................................................................................................................................... 33
Figure 4.3: Styrene selectivity profiles for points A and B of Figure 4.1 (Savoretti et al., 1998:210). 33
Figure 4.4: Inlet reactants temperature, considering the same EB feed flow rate (Savoretti et al.,
1998:211). ....................................................................................................................................... 34
Figure 4.5: Increased profit by using the isothermal reactor instead of adiabatic units Savoretti et al.,
1998:211). ....................................................................................................................................... 34
Figure 4.6: Material and energy data for reactor process streams. ..................................................... 35
Figure 4.7: Graphical representation of conversion versus rate of reaction data ................................ 39
Figure 5.1: Component flow rates for flash drum streams................................................................. 55
Figure 5.2: Vapour pressures of organic compounds up to 1 atm (1.013bar) ..................................... 58
Figure 5.3: Determination of optimum reflux ratio (Thakore & Bhatt, 2007:397) ............................. 61
Figure 5.4: McCabe-Thiele diagram for ethylbenzene-styrene system .............................................. 63
Figure 5.5: Column internals (Prieve, 2018) ..................................................................................... 66
Figure 5.6: Flooding velocity for sieve plates (Thakore & Bhatt, 2007:444) ..................................... 67
Figure 6.1: Centrifugal pump (CPL300B, 2017) .............................................................................. 74
Figure 6.2: Positive displacement pumps (CPL300B, 2017) ............................................................. 75
Figure 6.3: Basic pump curve plots (Morgan, 2016:42) .................................................................... 76
Figure 6.4: Pump performance chart (Kreider & Rabl, 1996:6) ........................................................ 77
Figure 6.5: Diagram of pump system ............................................................................................... 78
Figure 6.6: Inter-unit spacing recommendations for chemical plants ................................................ 79
Figure 6.7: Pipe friction versus Re and relative roughness Coulson & Richardson (2005:202) .......... 82
Figure 6.8: System curve and Pump Curve....................................................................................... 85
Figure 6.9: Characteristic curve for a centrifugal pump (Coulson, 1999:336) ................................... 87
List of Tables
Table 1.1: Flow Rate Summary for Process Streams for One Combination of Conditions. .................. 5
Table 1.2: Key for conditions of columns for tables below ................................................................. 7
Table 1.3: Varying Conditions and Data ............................................................................................ 8
Table 1.4: Gross Profit under Varying Conditions and Constant Inlet Feed of EB .............................. 8
Table 1.5: Stream Flow Rates for Reactor under Varying Conditions and Constant Inlet Feed of EB .. 9
Table 1.6: Stream Flow Rates for Separator under Varying Conditions and Constant Inlet Feed of EB9
Table 1.7: Stream Flow Rates for Toluene Tower (Column 1) under Varying Conditions and Constant
Inlet Feed of EB .............................................................................................................................. 10
Table 1.8: Stream Flow Rates for Styrene Tower (Column 2) under Varying Conditions and Constant
Inlet Feed of EB .............................................................................................................................. 10
Table 1.9: Stream Flow Rates for Fresh Plant Feed under Varying Conditions and Constant Inlet Feed
of EB............................................................................................................................................... 10
Table 3.1: Coefficients for specific heat capacity for compound in liquid phase ............................... 17
Table 3.2: Coefficients for specific heat capacity for compound in gas phase ................................... 18
Table 3.3: Coefficients for HVAP formula ....................................................................................... 18
Table 3.4: Physical properties of components in cold stream ............................................................ 19
Table 3.5: Physical properties of components in hot stream.............................................................. 20
Table 3.6: Summary of tube specifications ....................................................................................... 23
Table 3.7: Calculation of density and viscosity of stream based on mole fraction of species in fluid . 25
Table 3.8: Calculation of heat capacity and thermal conductivity of stream based on mole fraction of
species in fluid ................................................................................................................................ 25
Table 4.1: Conversion versus rate of reaction data ........................................................................... 38
Table 4.2: Coefficients of species for density formula ...................................................................... 42
Table 4.3: Coefficients of species for viscosity formula in gas phase ................................................ 42
Table 4.4: Fluid properties of Molten Salt (Serrano-Lopez, 2013) .................................................... 43
Table 5.1: Correlation constants (Yaws, 1999:400) .......................................................................... 51
Table 5.2: Henry’s law constant for compound in water – organic compounds (Yaws, 1999:416) .... 52
Table 5.3: Solubility of hydrocarbon in water .................................................................................. 54
Table 5.4: Liquid and vapour molar fractions of hydrocarbons ......................................................... 54
Table 5.5: Vapour pressures of Organic compounds up to 1 atm (Perry, 1998:2-68) ......................... 57
Table 5.6: Material balance for T-501 .............................................................................................. 58
Table 5.7: Experimental vapour-liquid equilibrium data for ethylbenzene and styrene (Jongmans et al.,
2011:3515) ...................................................................................................................................... 60
Table 5.8: Calculated values of K-values, relative volatility and minimum reflux ratio ..................... 60
Table 5.9: Variables used to calculate q ........................................................................................... 62
Table 5.10: Determining equation of rectifying line ......................................................................... 62
Table 5.11: Column diameter and Pressure data ............................................................................... 68
Table 6.1: Coefficients for density of ethylbenzene as a function of temperature (Yaws, 1999:185).. 79
Table 6.2: Coefficients for viscosity of ethylbenzene as a function of temperature (Yaws, 1999:478)80
Table 6.3: Coefficients for vapour pressure of ethylbenzene as a function of temperature (Yaws,
1999:159) ........................................................................................................................................ 80
Table 6.4: Fluid Properties of ethylbenzene ..................................................................................... 80
Table 6.5: Flow rates ....................................................................................................................... 81
Table 6.6: Typical pipe velocities and allowable pressure drops used to estimate pipe sizes.............. 81
Table 6.7: System data and specifications ........................................................................................ 82
Table 6.8: Velocity heads of pipe fittings and valves (Sinnott & Towler, 2008:249) ......................... 83
Table 6.9: Given data for pump curve and calculated data for system curve ..................................... 84
Table 6.10: Pump Characteristics (Silla, 2003:256) .......................................................................... 86
CHAPTER 1 MINI PROJECT
Summary
The aim of this project is to perform a preliminary evaluation on the economics of the production of
styrene by dehydrogenation of ethyl benzene. In order to achieve this, the operational conditions and
economic potential of this process was investigated. It was found that conditions of higher
temperatures and lower pressures favour the formation of styrene. However, side reactions also
become significant at higher temperatures. The conditions which yielded the highest conversion (a
temperature of 950 K, pressure of 0.4 bar and a steam ratio of 12:1), was not the conditions which
yielded the highest profit, due to the high costs of steam.
The conditions which yielded the highest profit and met the reaction requirements (of adding steam
necessary to add heat for the endothermic reaction) was 900 K (temperature), 0.6 bar (pressure) and a
steam to ethyl benzene ratio of 6:1. Under these conditions, an ethyl benzene feed of roughly 43 000
kg/hr is required to produce 280 000 ton/yr of styrene and a preliminary gross profit of approximately
R199 million can be obtained annually. It was therefore recommended to investigate further into this
prospective plant.
1.1. Problem statement

Styrene is manufactured and used worldwide, to produce synthetic materials. The most widely used
method to produce styrene is the dehydrogenation of ethyl benzene. The economics for the production
of styrene, by this method, is to be preliminarily evaluated, in order to determine whether this
production process is feasible. To achieve this, combinations of different operating conditions
(temperature, pressure and steam to ethyl benzene ratio) together with an approximation of the highest
preliminary gross profit attainable must be assessed. The best operating conditions and preliminary
profit to produce 280 000 ton/yr of styrene must then be reported.
1.2. Introduction
The chemical, styrene, is used to make synthetic rubber, polystyrene resins and latex. The process of
styrene production is over 5 decades old worldwide and new variations of established processes are
being developed. The feed stock for the most commercial styrene manufacture is ethyl benzene.
Dehydrogenation of ethyl benzene to produce styrene accounts for 85% of commercial production.
The reaction is carried out in the vapour phase with steam over a catalyst consisting primarily of iron
oxide (James & Caster, 2007:4).
The plant considered for styrene production is designed for the dehydrogenation of ethyl benzene.
This plant is required to produce 280 000 ton/yr of styrene. The feasibility and preliminary
profitability of the plant will be evaluated by only considering the cost of reactants and selling price of
1
products. The optimal operating conditions chosen should facilitate the maximum profit value for this
plant. An estimation of the plant’s optimum operating conditions, gross profit and economic potential
will be discussed.
It has been given that the temperature range of operation is within 800 – 950 K, the pressure range is
0.4 -1.4 bar and the steam to ethyl benzene ratio range is 6:1 – 12:1. The primary reaction is an
equilibrium limited reversible reaction, with the reaction proceeding to 80 % of its equilibrium value.
The fractional selectivity to toluene, along with the PFD and given costs and selling prices of the
reactants and products respectively are attached in Appendix A.
Optimum operating conditions will be determined by finding the conditions which allow for the
maximum conversion of raw materials to sellable products in combinations with the maximum
preliminary gross profit. A material balance, together with a range of different process conditions, in
different combinations, will be used to find the best operating conditions and economic potential of
this process.
1.3. Data
The given data for the project is attached in Appendix A along with sample calculations for data used
in section 1.3.1 below. The given data attached includes: the primary reversible reaction and side
reaction equations; temperature, pressure and steam ratio ranges which favour the forward reaction;
the selectivity at different temperatures, the catalyst used and the cost and selling prices of all
products and reactants. For this chapter, a simplified PFD of the process was also given and is
attached in Appendix A, along with stream information.
1.3.1 Calculated Data

Changing the three main variable reactor conditions, temperature, pressure and ratio of steam to ethyl
benzene, shows that different combinations of conditions gave different conversions, selectivity and
extents of reaction for the primary forward reaction and side reaction. The calculations used to
determine the effect operating conditions have on the above mentioned are attach in Appendix A
along with graphical representation of the change in conversion at a range of different conditions.
A high conversion of products to reactants is desired hence the conditions yielding this and the
relationship between the conversion and change in conditions were determined.
The role this data played in choosing the optimum process conditions, while taking preliminary gross
profit into account, will be discussed in section 1.4 (Discussion).
2
Fractional Selectivity to 0.14
0.12
y = 4E-08x3 - 0.0001x2 + 0.0837x - 23.43
0.1 R² = 1
Toluene
0.08
0.06
0.04
0.02
0
750 800 850 900 950 1000
Temperature (K)
Figure 1.1: Relationship between Fractional Selectivity to Toluene and Temperature
Figure 1.1 above shows that the fractional selectivity for toluene formation increases with an increase
in temperature. Given data, Table 11 in Appendix A, was used to plot the above graph to obtain the
relationship between temperature and selectivity. This relationship was used to calculate the effect
varying temperatures have on extent of reaction.
1.2
0.8
Conversion
Temperature: 800K
0.6
Temperature: 850K
0.4 Temperature: 900K

Temperature: 950K
0.2
0
0.3 0.5 0.7 0.9 1.1 1.3 1.5
Pressure (atm)
12:1
Figure 1.2 and Figures 6 to 12 attached in Appendix A, illustrate that conversion increases as the
temperature increases. According to Towler & Sinnott (2008:64), high conversion leads to the reactor
having low concentrations of at least one reactant and high concentrations of products, which includes
the products of the side reaction. As illustrated in Figure 1.1, the rise in temperature also increases
selectivity which increases the ratio of side products formed (methane and toluene) to the main
products formed (styrene and hydrogen). According to the costing data in Appendix A, styrene has a
higher selling price than methane and toluene. Therefore increasing the selectivity does not
3
necessarily lead to higher profits and high conversion may not be an indication of increased styrene
production or result in an increase in profit.
The graphs illustrating the relationship between conversion and the different operating conditions are
based on an ethyl benzene flow rate of 100 kmol/hr of the feed stream to the reactor. Figure 1.2
includes the highest conversion obtained from a specific combination of operating conditions,
compared to others – illustrated in Figures 6 - 12. The conditions yielding the highest conversion, at
0.94, is a temperature of 900 K, pressure of 0.4 bar and steam ratio of 12:1.
1.3.2. Material Balance

A basis of 100 kmol/hr of ethyl benzene was arbitrarily chosen to be entering the reactor. This
assumption was used to estimate the flow rates of the process streams of the plant at different
conditions by conducting a material balance. The system is assumed to be at steady state and due to
the occurrence of the side reaction, extent of reaction and conversion data; it is known that the total
mass of ethyl benzene entering the reactor does not dehydrogenate to form styrene. Conversion at all
the combinations of operating conditions considered were less than 1 (as shown in the graphs of
conversion mentioned above), showing that some of the reactants did not react, and re-enter the
reactor due to the recycle stream. Also, the occurrence of the side reactions means that some of the
ethyl benzene formed side products, instead of styrene.
It is assumed that there is 100% separation at the three phase separator (V-104), toluene tower (T-401)
and styrene tower (T-402) (Refer to Appendix A for PFD with equipment numbers and labels). This
assumption was made because the objective of this task is to only estimate whether or not there is a
possibility of preliminary gross profit in this styrene production process. For a more realistic monetary
analysis, the actual compositions of the streams must be taken into account.
The material balance was done around the reactor (see sample calculations in Appendix A). The flow
rates for process streams 1 to 10 (refer to Table 1.10 AppendixA) were calculated from the basis of
100 kmol/hr of ethyl benzene fed into the reactor (stream 2), temperature 800 K, pressure 0.4 bars and
steam to ethyl benzene ratio of 6:1. Table 1 below reflects the flow rates of all the streams in terms of
moles and mass. These flow rates are a function of reactor conditions, steam to ethyl benzene ratio
and the amount of ethyl benzene fed into the plant.
4
Table 1.1: Flow Rate Summary for Process Streams for One Combination of Conditions.
Substance
Total
Stream Units EB Steam Styrene Hydrogen Toluene Methane Waste water
kmol/hr 5.68E+01 6.00E+02 6.57E+02
1 kg/hr 6.03E+03 1.08E+04 1.68E+04
ton/yr 4.82E+04 8.65E+04 1.35E+05
kmol/hr 1.00E+02 6.00E+02 7.00E+02
2 kg/hr 1.06E+04 1.08E+04 2.14E+04
ton/yr 8.49E+04 8.65E+04 1.71E+05
kmol/hr 4.32E+01 5.62E+01 5.56E+01 5.62E-01 5.62E-01 6.00E+02 7.56E+02
3 kg/hr 4.59E+03 5.85E+03 1.12E+02 5.18E+01 9.02E+00 1.08E+04 2.14E+04
ton/yr 3.67E+04 4.68E+04 8.99E+02 4.14E+02 7.21E+01 1.23E+03 8.62E+04
kmol/hr 5.56E+01 5.62E-01 5.62E+01
4 kg/hr 1.12E+02 9.02E+00 1.21E+02
ton/yr 8.99E+02 7.21E+01 9.71E+02
kmol/hr 4.32E+01 5.62E+01 5.62E-01 1.00E+02
5 kg/hr 4.59E+03 5.85E+03 5.18E+01 1.05E+04
ton/yr 3.67E+04 4.68E+04 4.14E+02 8.40E+04
kmol/hr 6.00E+02 6.00E+02
6 kg/hr 1.08E+04 1.08E+04
ton/yr 8.65E+04 8.65E+04
kmol/hr 5.62E-01 5.62E-01
7 kg/hr 5.18E+01 5.18E+01
ton/yr 4.14E+02 4.14E+02
kmol/hr 4.32E+01 5.62E+01 9.94E+01
8 kg/hr 4.59E+03 5.85E+03 1.04E+04
ton/yr 3.67E+04 4.68E+04 8.36E+04
kmol/hr 4.32E+01 4.32E+01
9 kg/hr 4.59E+03 4.59E+03
ton/yr 3.67E+04 3.67E+04
kmol/hr 5.62E+01 5.62E+01
10 kg/hr 5.85E+03 5.85E+03
ton/yr 4.68E+04 4.68E+04
1.3.3. Costing
Costing is an important aspect in determining whether the process is feasible or not. If the preliminary
gross profit is negative or too low, then there would be no reason to invest further time into pursuing
this process as it is. The conditions which allowed for the highest gross profit, based on the 100
kmol/hr of ethyl benzene feed, was used to determine the mass flow rates of the ethyl benzene in at
stream 1(fresh feed) and styrene out at stream 10. This was then used to calculate the scaling factor
(1.24) to determine the amount of ethyl benzene to be fed into the plant so that a styrene output
capacity of 280 000 ton/yr is attained (a sample calculations for scaling factor is provided in
Appendix A). Table 1.13 in Appendix A provides the given costs of reactants and selling prices of the
products.
The data used to plot the graph in Figure 1.3 below indicates that the conditions yielding the highest
preliminary gross profit (R49 mill/yr), including steam injection, are: temperature 900 K, pressure 0.4
bar and steam ratio 6:1. Figure 1.4 shows that the highest preliminary gross profit (R89 mill/yr)
5
achievable without steam injection is under conditions of 950 K temperature and 0.4 bar pressure.
Although the process without steam injection yields the highest estimated preliminary profit overall,
those conditions were not chosen as the optimal conditions for this process. The reason for this will be
discussed further in section 1.4 (Discussion). Additional graphs which illustrate the effects of varying
operating conditions at temperatures 800 K and 850 K on the profits are attached in Appendix A.
Comparing these four “Profit versus Ratio” graphs, it can be seen that the profits increase with an
increase in temperature and decrease with an increase in pressure and steam injection. This will also
be explained in section 1.4. A sample calculation for the preliminary gross profit is attached in
Appendix A.
95000000
75000000
55000000
Pressure: 0.4 bar
Profit (R/hour)
35000000
Pressure: 0.6 bar
15000000 Pressure: 0.8 bar
-5000000 Pressure: 1.0 bar
-65000000
0 5 10 15
Ratio (kmol steam : 1 kmol EB)
Figure 1.3: Relationship between Profit and Steam Ratio at Varying Pressures with Temperature 900K
6
95000000
75000000
55000000
Pressure: 1.4 bar
Profit (R/hour)
35000000
Pressure: 0.4 bar

Pressure: 1.2 bar
-45000000
-65000000
0 5 10 15
Table 1.4 compares the estimated gross profits at the best conditions for producing 280 000 ton/yr,
keeping the ethyl benzene flow rate entering the reactor constant. Below are stream tables at the
conditions which allow for the best preliminary gross profit to be made without steam (column 1) and
without steam (column 2). The gross profit made by using the conditions which yield the highest
conversion is also shown (column 3) as these are the main profit results which will be discussed
further in section 1.4. The feed of ethyl benzene needed to produce the given capacity at these
conditions was found to be approximately 380 000 ton/yr.
Table 1.2: Key for conditions of columns for tables below

Column marked 1 Conditions for the highest profit without steam
Column marked 2 Conditions for the highest profit with steam
Column marked 3 Conditions for the highest conversion
7
Table 1.3: Varying Conditions and Data
Column: 1 2 3
Reactor Conditions
Temp K 950 900 950
Pressure bar 0.4 0.4 0.4
Ratio: Steam to 1 EB 0 6 12
Calculated Parameters
Selectivity unitless 0.130 0.060 0.130
Ke unitless 0.911 0.382 0.911
Conversion unitless 0.805 0.875 0.964
Extent of rxn 1 kmol/hr 291 337 348
Extent of rxn 2 kmol/hr 38 20 45
Table 1.3 states the reactor conditions and calculated parameters at which the 3 gross profits were
calculated in Table 1.4. The conditions under column 1yields the highest gross profit for this process
and happens to have zero steam injection. Since the steam is necessary in this reaction because it is
endothermic and required a source of heat, the highest gross profit with steam injection is also
considered - column 2 states these conditions. Column 3 states the conditions which yield the highest
conversion. This is added to the tables 1.3 and 1.4 to show that the highest conversion does not yield
the highest profit and will be discussed in section 1.4.
Table 1.4: Gross Profit under Varying Conditions and Constant Inlet Feed of EB
Gross Profit Calculation

Substance R/kg 1 2 3
Ethylbenzene 7.50 2.62E+05 2.84E+05 3.13E+05
Steam @ 800K 0.66 0 2.91E+04 5.82E+04
Waste water 0.06 0 2.65E+03 5.29E+03
Styrene 9.30 2.82E+05 3.26E+05 3.37E+05
Hydrogen 3.15 1.61E+03 2.01E+03 1.93E+03
Methane 1.29 7.83E+02 4.18E+02 9.37E+02
Toluene 6.60 2.30E+04 1.23E+04 2.75E+04
R/hr 4.54E+04 2.49E+04 -9.19E+03
Gross Profit
R/year 3.63E+08 1.99E+08 -7.35E+07
Tables 1.5 to 1.9 below show the individual stream flow rates for the 3 different conditions mentioned
above. With this, comparisons can be drawn between the stream flow rates under the 3 combinations
of operating conditions considered, for which the preliminary profits will be discussed. Refer to Table
1.10 in Appendix A for stream number descriptions.
8
Table 1.5: Stream Flow Rates for Reactor under Varying Conditions and Constant Inlet Feed of EB
Conditions: 1 2 3
REACTOR
Stream 2
Ethylbenzene kmol/hr 4.07E+02 4.07E+02 4.07E+02
Steam 0 2.44E+03 4.90E+03
Total kmol/hr 4.07E+02 2.85E+03 5.30E+03
ton/yr 3.46E+05 3.52E+05 7.06E+05
Stream 3
Styrene 2.90E+02 3.36E+02 3.48E+02
Hydrogen 2.53E+02 3.16E+02 3.03E+02
Methane 3.77E+01 2.02E+01 4.52E+01
Toluene 3.77E+01 2.02E+01 4.52E+01
Waste water 0 2.44E+03 4.90E+03
Total kmol/hr 6.98E+02 3.19E+03 5.65E+03
ton/yr 3.46E+05 6.98E+05 1.05E+06
Table 1.6: Stream Flow Rates for Separator under Varying Conditions and Constant Inlet Feed of EB
Three Phase Seperator

Stream 4
Hydrogen kmol/hr 2.53E+02 3.16E+02 3.02E+02
Methane 3.77E+01 2.02E+01 4.52E+01
Total kmol/hr 2.90E+02 3.36E+02 3.47E+02
ton/yr 8.93E+03 7.69E+03 1.07E+04
Stream 5
Styrene 2.90E+02 3.36E+02 3.47E+02
Toluene 3.77E+01 2.02E+01 4.52E+01
Total kmol/hr 4.07E+02 4.07E+02 4.07E+02
ton/yr 3.37E+05 3.38E+05 3.35E+05
Stream 6
Waste water kmol/hr 0 2.44E+03 4.89E+03
Total ton/yr 0 3.52E+05 7.05E+05
9
Table 1.7: Stream Flow Rates for Toluene Tower (Column 1) under Varying Conditions and Constant
Inlet Feed of EB
Toluene Tower
Stream 7
Toluene kmol/hr 3.77E+01 2.02E+01 4.52E+01
Total ton/yr 2.78E+04 1.49E+04 3.33E+04
Stream 8
Styrene 2.90E+02 3.36E+02 3.47E+02
Total kmol/hr 3.70E+02 3.87E+02 3.62E+02
ton/yr 3.09E+05 3.23E+05 3.02E+05
Table 1.8: Stream Flow Rates for Styrene Tower (Column 2) under Varying Conditions and Constant
Inlet Feed of EB
Styrene Tower
Stream 9
Total ton/yr 6.73E+04 4.34E+04 1.25E+04
Stream 10
Styrene kmol/hr 2.90E+02 3.36E+02 3.47E+02
Total ton/yr 2.42E+05 2.80E+05 2.89E+05
In Table 1.8, the highlighted number indicates the desired plant capacity under the chosen optimal
conditions.
Table 1.9: Stream Flow Rates for Fresh Plant Feed under Varying Conditions and Constant Inlet Feed of
EB
Fresh Plant Feed

Stream 1
Ethylbenzene kmol/h 3.28E+02 3.56E+02 3.93E+02
Steam 0 2.44E+03 4.89E+03
Total kmol/hr 3.28E+02 2.80E+03 5.28E+03
ton/yr 2.79E+05 6.55E+05 1.04E+06
1.4. Discussion
From Table 1.4 it was seen that the conditions yielding the best conversion does not yield the most
gross profit, as the process operating under these conditions generates a loss of R73.5 mill/yr. These
conditions were found to be 950 K temperature, 0.4 bar pressure and 12:1 steam ratio. Although
injecting steam at higher ratios increase conversion, this also increases the occurrence of the side
reaction. The expense of steam and waste water in the process therefore grows and the value of the
products formed at higher conversion does not compensate for the rise in costs. Also, the side
10
reactions become increasingly significant at the higher temperatures, where conversion is high. The
products of the side reaction, methane and toluene, do not have as high selling price as styrene, i.e.
they are not as valuable, and therefore the profitability of the process decreases. Also, the aim is not to
produce toluene and methane but the main product, styrene.
As shown in Tables 1.3 and 1.4, the highest preliminary gross profit was found to be at conditions of
950 K temperature and 0.4 bar pressure with no steam injected. These conditions were however not
chosen as the optimum conditions. Although more preliminary profit was shown here, these
conditions did not meet the energy requirement of the endothermic forward reaction. Steam is a
necessary addition to add the heat needed by the reaction, as there is no additional heat source at this
stage besides the steam. Hence, conditions involving steam injection, which resulted in the highest
preliminary gross profit was preferred and considered as the best operating conditions instead of this.
The optimum conditions chosen, which yield the best preliminary gross profits is a temperature of 900
K, 0.4 bar and a steam ratio of 6:1.These are a combination of temperature being high enough to
favour styrene production but low enough not to increase the selectivity of side reaction products (like
toluene) to styrene too much, as the side reaction also increases as the temperature increases. The
lower pressure also shifts the equilibrium to the right, favouring styrene production. The steam
injection flow rate here is low enough not to increase the costs of production (through steam and
waste water costs) too much but is still sufficient to provide the energy required by the endothermic
forward reaction. This added energy ensures that more of the desired main product, styrene, is formed.
This is in accordance with Towler & Sinnott (2008:64), which states that the optimum reactor
conditions usually favours low reactor conversion to give high selectivity for the desired products
when all costs are taken into account. The costs mentioned in the previous statement refer to reactants
that are converted to by-products and side products which must be recovered and purified for sale or
disposed as waste. In this process, the steam entering the reactor is removed as waste water and since
it is inert, it does not interfere in the reaction. It adds heat, limits the side reaction slightly and extends
catalyst life, which improve the process, therefore making it a significant addition. Ethyl benzene
which was not converted in the reactor are however recovered and recycled, reducing the waste and
costs of purchasing fresh reactant. Side products, toluene and methane, and by-product hydrogen, are
sold and thus add to the profits. These do however have to be recovered and separated from one
another as they are all present in the reactor output stream. This recovery process adds to the overall
costs of the plant, even though these costs were not considered to calculate the preliminary gross
profit.
It was assumed that the plant operates for 8000 hours per year. According to Sinnott (2005:143), this
is an appropriate estimation of plant operation time per annum, for a chemical plant, as the plant
11
attainment would range from 90% to 95% of the total hours in a year. This lost time costs the plant, as
the longer the plant remains out of operation, the more loss is incurred – because no products are
being manufactured to sell, which would have contributed to the profits. A plant shutdown is however
necessary for renewal of the Iron catalyst, cleaning equipment, maintenance and inspection.
Figures 3, 4, 13 and 14 clearly show that as the addition of steam increases, the profits decrease. From
the calculations, and visual representation of Figure 1.4, the conditions yielding the most profit,
mentioned above, for a plant capacity of 280 000 ton/yr, of styrene provides a maximum preliminary
gross profit of roughly R19.9 billion per year. As these are estimated values, it is important to note
that a margin of error of 30% is incurred and therefore should be taken into account.
1.5. Conclusion
It can be concluded that this rough analysis of the feasibility of styrene production is possibly
profitable, due to the income (selling price of products) exceeding the expenses (cost price of
reactants). At a plant capacity of 280 000 ton/yr of styrene, an ethyl benzene inlet fresh feed rate of
about 303 000 ton/yr is required. Also, estimated reactor operating conditions of 900 K temperature,
0.4 bar pressure and a steam ratio of 6:1 is necessary to achieve a preliminary profit while meeting the
energy requirements of the forward reaction to produce the desired product, styrene. A preliminary
gross profit of approximately R19.9 billion per annum can be made; therefore this production process
could possibly be economically feasible.
1.6. Recommendations
It is recommended that this process be investigated in further detail to attain results with a lesser
margin of error. This will provide a better indication of how profitable this process may be. The
positive preliminary profit of R 19.9 billion per annum shown is a good indication of economic
potential at this preliminary stage. This production process should be simulated using simulation
software in order to get more realistic data about the optimal operating conditions and economic
potential of the plant.
It is also recommended that an additional heat source be implemented to, along with the steam
injection. Heat energy is necessary because the primary reaction to produce styrene is an endothermic
reaction. It is also evident from the profit graphs (Figures 3 and 4) that the addition of steam decreases
the gross profits substantially. However, the steam also acts as diluent to shift equilibrium to the right
in favour of styrene and tends to limit side reactions and extend catalyst life. The additional heat
source could therefore be used in conjunction with a reduced steam injection, to minimise costs. For
example, the addition of a heat exchanger to provide heat to the reactor feed stream could be
investigated. The effects of different combinations of steam injection rates and heat transfer by the
exchanger could be examined to determine the best exchanger to use and optimum amount of heat to
12
add as a combination of steam and heat from the exchanger. Although the installation and
maintenance of the exchanger will add expenses to the process, it is possible that money could be
saved in the long term, therefore increasing profits without neglecting the forward reaction
requirements.
Since the current preliminary profitability is dependent on the value of the costs and selling prices at
present, it is also recommended that the most recent prices be used to calculate the profit. If the plant
has the appropriate storage facilities, ethyl benzene may be bought in access when the price decreases
and used within 3 days. As per the Agency for Toxic Substances and Disease Registry (2007), it takes
around 3 days for ethyl benzene to be broken down in air into other chemicals. This time frame will
also reduce the safety risk of storing this flammable substance for long periods of time. According to
Ethyl Benzene Product Stewardship Summary (2017), the appropriate storage facility for this
substance is a well-ventilated area at ambient temperature. It should also be away from any sources of
heat, including direct sunlight, and an ignition source.
1.7. References
 Agency for Toxic Substances and Disease Registry. (2017) Ethylbenzene. [Online]. Available:
http://www.atsdr.cdc.gov/toxfaq.html. [2019, February 18].
 Shell. 2017. Ethyl Benzene Product Stewardship Summary. [Online]. Available: https://www.
Shell.com/business-customers/chemicals/safe-product-handling-and transportation/product-st
eward shipsummaries/_jcr_content/par/expandablelist/expandablesection_460932334.stream
/1523032145901/e31a57e7c35269fe15f226262d7c5193c78dd9c867e6533bb468be24e8314fc6/
ethyl-benzene-pro duct-stewardship-summary-december-2017.pdf . [2019, February 18].
 James, D. H. & Castor, W. M. 2007. Styrene, Ullmann's Encyclopedia of Industrial Chemistry. 7th
edition. Texas: Wiley.
 Sinnott, R. K., Coulson, J. M., Richardson, J. F. 2005. Coulson & Richardson's chemical
engineering Series. Chemical Engineering Design. 4th Edition. Volume. 6. Oxford: Elsevier
Butterworth-Heinemann.
 Towler, G. & Sinnott, R. 2008. Chemical Engineering Design. Principles, Practice and Economics
of Plant and Process Design. London: Elsevier Butterworth-Heinemann.
13
CHAPTER 2 PROCESS DESCRIPTION
The process flow diagram (PFD) is attached in Appendix B. It also contains the equipment and stream
descriptions.
2.1. Introduction
This chapter describes the process flow diagram which has been developed for the production of
styrene from ethylbenzene. The purpose of the major equipment in the process is described in the
sections below and has been sectioned into four major sections. These sections include the preparation
of the feed to the reactor, the ejector and coolers, the flash separation and the distillation columns.
2.1.1. Feed preparation to the reactor

Fresh ethylbenzene feed enters at stream 1 (refer to PFD in Appendix B). It is then mixed with recycle
stream 35 so that stream 3 thereafter contains mostly ethylbenzene with a less styrene and toluene.
This then enters the shell side of heat exchanger E-501. Here the hydrocarbons are heated though
indirect contact with high pressure steam. This increase in temperature is intended to help increase the
temperature of the feed stream to the reactor (R-501). More heat is then added to the stream with the
introduction of steam from fired heater H-501. This is introduced to line 4, via a mixing valve, from
line 6, along which there is a PSV to help relieve some of the pressure that may build up on the line.
This heat is needed as the reaction to take place in R-501 is an endothermic reaction, so adding energy
will help drive the reaction to the right, however to prevent the reaction from occurring prematurely,
in the line, the mixture is not heated up all the way to the optimum condition. This mixture of steam
and hydrocarbons then enter R-501, which has a packed bed shell and tube reactor design. The
mixture is on the tube side with the catalyst, while molten salt is used as the heating fluid which adds
the necessary heat to the reactants.
2.1.2. Ejector and coolers

The products, with some reactants and steam, exit the reactor at stream 8 before entering cooler E-
502. This heat exchanger is in there to cool the feed to the ejector (J-501). According to Lieberman
(2014:243), the ejector can develop a better vacuum (or low pressure required), when the vapour
temperature to the ejector inlet is colder, caused by the reduced vapour flow to the ejector. The
products then enter the steam ejector, which is used to induce R-501’s low operating pressure.
Thereafter the hydrocarbon stream is condensed by exchanger E-503, so that it can be pumped by
pump (P-502) to exchanger E-504, just before the flash separator (V-501). Keeping this stream at
lower temperatures also helps to prevent the styrene from polymerising, which can occur
spontaneously at high temperatures. E-504 increases the temperature of the stream to that of the
operating temperature of V-501, while P-502 helps to increase the pressure of the fluid for V-501.
14
The valve before the flash drum is a check valve to prevent the backflow of the stream to the flash
drum.
2.1.3. Flash Separation

The stream enters V-501 where the different components are separated based on their volatilities. A
throttling valve is located at the entry of the vessel to ensure that the components flash within V-501
at reduced pressure. The water is separated containing trace amounts of hydrocarbon. It therefore gets
pumped to a treatment plant, to remove the organic molecules. The hydrogen and methane are the
lightest species in the mixture and therefore get separated from the other components by stream 14.
Stream 14 however does still also contain trace amounts of the other components in the mixture.
Stream 18 is comprised of mostly styrene and toluene, with little amounts of ethylbenzene. Along
stream 18, there is a valve to help decrease the pressure before the distillation columns.
2.1.4. Distillation Columns

At reduced pressure, stream 20 containing styrene, toluene and small amounts of ethylbenzene enter
the first distillation column (T-501). Here the styrene and ethylbenzene is separated from the toluene
which has a higher volatility. The condenser (E-506) and re-boiler (E-507) are used to improve
separation. The lover volatility components then move on as stream 28 to the second distillation tower
(T-502), where the styrene is separated from ethylbenzene, which is the higher volatility component
here. There is also a condenser (E-509) and re-boiler (E-510) here to increase the separation. As with
the toluene from T-501, the styrene from T-502 is pumped out of plant 5, while the unreacted
ethylbenzene from T-502 is pumped back to the feed stream via line the recycle steam (stream 35).
15
CHAPTER 3 HEAT EXCHANGER DESIGN
Summary
Heat exchanger, E-501 in Unit 500, styrene production unit, will be designed in this chapter. The
purpose of this heat exchanger is to exchanger heat from the hot high pressure steam fluid to the cold
hydrocarbon fluid – ethylbenzene with small amounts of styrene. This serves as part of the preheat
before the reactor. The designing of E-501 is discussed step by step, with calculations and all data
shown along with references.
3.1. Introduction
In a styrene plant, a heat exchanger must be designed to heat the feed stream of primarily
ethylbenzene, before the reactor. A shell and tube heat exchanger was chosen as the small tube
diameter makes the exchanger both economical and compact. The tubes are prevented from vibrating
by baffles holding the tube bundle and they are robust due to their shape. Although this type of
exchanger is likely to foul up faster, the fluids on the shell and tube side of the exchanger are
relatively clean. These fluids are high pressure steam and ethylbenzene with small amounts of styrene
and toluene. Compared to the other types of shell and tube exchangers, a floating head shell and tube
heat exchanger has more components in its construction, resulting in it being more costly. This design
however allows for free expansion of the tube bundle and allows the tube bundle to be removed from
the shell making cleaning both the inside and outside of the tubes possible (Mukherjee, 1998:5). A
pull-through floating head was chosen for this plant. It is not as costly as a normal floating head heat
exchanger and since it will have relatively clean service, ease of cleaning is not a major concern.
3.2. Process Flow Diagram
Figure 3.1: PFD of E-501
16
The cold hydrocarbon fluid in on the shell side and the hot steam fluid is on the tube side. This will be
explained further below. The ethylbenzene feed stream contains amounts of styrene and toluene due
to a recycle stream which joint with stream 2 upstream of the heat exchanger E-501. The temperature
of stream 5 will be calculated in the section below.
3.3. Design of Heat Exchanger
For the design of this heat exchanger, Kern’s Method has been used. The steps of the method are
included in sections and explained below.
3.3.1. Specification
In this section the duty is defined and the energy balance is used to determine the unspecified stream 5
(referring to Figure 3.1 above) outlet temperature. For the cold fluid (hydrocarbons), Equation 1 was
used to calculate the duty. Depending on the units of Cp, the mass flow rate or molar flow rate can be
used.
𝑄𝑐 = 𝑚̇ 𝐶𝑝∆𝑇 = 𝑛̇ 𝐶𝑝∆𝑇 (Equation 1) (Incropera, 2007:17)
The mass flow rates (𝑚̇) used in the calculations above were obtained by the mass balance in Chapter
1. There is a mixture of components in the cold fluid stream and each component has a different
specific heat capacity (Cp). The formulas from Yaws (1999:56) below was therefore used to calculate
the different Cp values are the given inlet temperature.
Heat capacity for liquid: 𝐶𝑝𝑙 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2 + 𝐷𝑇 3 (Equation 2)
Table 3.1 below contains the values of the coefficients used to calculate the specific heat capacity
using Equation 2 of the different compounds in the stream.
Table 3.1: Coefficients for specific heat capacity for compound in liquid phase
Species A B C D Tmin (K) Tmax

(K)
EB 1.02E+02 5.60E-01 -1.56E-03 2.01E-06 179 555
Water 9.21E+01 -4.00E-02 -2.51E+01 5.35E-07 273 615
Styrene 6.67E+01 8.41E-01 -2.16E-03 2.33E-06 244 583
Hydrogen 5.06E+01 -6.11E+00 3.09E-01 -4.15E-03 14 32
Methane -1.80E-02 1.20E+00 -9.87E-03 3.17E-05 92 172
Toluene 8.37E+01 5.17E-01 -1.49E-03 1.97E-06 179 533
17
Heat capacity for gas: 𝐶𝑝𝑔 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2 + 𝐷𝑇 3 + 𝐸𝑇 4 (Equation 3)
Table 3.2 below contains the values of the coefficients used to calculate the specific heat capacity
using Equation 3 of the different compounds in the stream.
Table 3.2: Coefficients for specific heat capacity for compound in gas phase
Species A B C D E Tmin Tmax

(K) (K)
EB -2.05E+01 5.96E-01 -3.08E-04 3.56E-08 1.24E-11 200 1500
Water 3.39E+01 -8.42E-03 2.99E-05 -1.78E-08 3.69E-12 100 1500
Styrene 7.12E+01 5.48E-02 6.48E-04 -6.99E-07 2.12E-10 100 1500
Hydrogen 2.54E+01 2.02E-02 -3.85E-05 3.19E-08 -8.76E-12 250 1500
Methane 3.49E+01 -4.00E-02 1.92E-04 -1.53E-07 3.93E-11 50 1500
Toluene -2.41E+01 5.22E-01 -2.98E-04 6.12E-08 -1.26E-12 200 1500
The fresh feed (stream 1) inlet temperature (Tci) was taken at 20CC, so the temperature of Tci was
taken to be the average temperature of steam 1 and 35 (recycle stream). It was specified that the outlet
temperature (Tco) of the cold stream should not exceed 5 oC lower than the inlet temperature
appropriate type of stream used for heating.
The specific heat capacity of the steam at temperature T hi obtained (explained below) was calculated
using Equation 2 and 3 above along with the corresponding coefficients for water from Tables 3.1 and
3.2. Ambient temperature of the water was taken to calculate the Cp of the water as it was in liquid
phase – using equation 2. The boiling point temperature of the water was used to calculate the Cp of
the water in gas phase (steam). Since the water gains energy when it changes from liquid to gaseous
phase, latent heat will also be added to calculate the duty. Equation 4 is used to calculate the heat of
vaporisation (HVAP) and the coefficients for this equation is listed in Table 3.3 below. Equation 4
and the data in Table 3.3 are from the textbook Yaws (1999:109).
𝑇
𝐻𝑉𝐴𝑃 = 𝐴(1 − 𝑇 )𝑛 (Unit: kJ/mol) (Equation 4)
𝐶
Table 3.3: Coefficients for HVAP formula
Species A Tc (K) N Tmin (K) Tmax (K) Tb (K)

EB 54.8 617 0.388 178 617 409
Water 52.1 647 0.321 273 647 373
Styrene 65.3 648 0.558 350 648 418
Hydrogen 0.659 33.2 0.38 14 33.2 20.4
Methane 10.3 191 0.265 90.7 191 112
18
Toluene 50.1 592 0.383 178 592 384
The pressure of the hps was specified to be 4237 kPa, according to the design brief. This was then
used to obtain the corresponding temperature (Thi) of the steam using an online software (TLV 2019).
The temperature of the hot fluid out (Tho) is unknown, so it was calculated by substituting the heat
gained by the cold fluid (-Qc) as the heat lost by the hot fluid (Qh) and solving the equation.
For the hot fluid (hps), 𝑄ℎ = 𝑚̇ 𝐶𝑝∆𝑇
It was decided that hps is the best fluid to use as it contains the adequate amount of heat to be
absorbed by the cold fluid mixture. Water also provides a cleaner service compared to using another
process fluid to heat the reactor feed which will reduce the rate of fouling.
3.3.2. Physical Properties of Fluids
Table 3.4: Physical properties of components in cold stream
Cold Stream inlet Mean outlet unit

Ethylbenzene
Temperature 297 410 522 K
Specific heat 0.184 0.198 0.212 kJ/kmolK
Thermal 0.130 0.0717 0.0132 kW/mK
conductivity
Density 866 745 625 kg/m3
Viscosity 0.000631 0.0553 0.11 kg/m.s
Styrene
Thermal 0.137 0.0768 0.0167 kW/mK
conductivity
Viscosity 0.000677 0.0559 0.111 kg/m.s
Toluene
Thermal 0.134 0.0748 0.0157 kW/mK
conductivity
Viscosity 0.000549 0.0614 0.122 kg/m.s
The thermal conductivity (k) for ethylbenzene, styrene and toluene was obtained using Equation 5
(Yaws, 1999:532), which is for liquid phase organic compounds, as all the species in the inlet stream
to the heat exchanger are in liquid phase. A sample calculation of 𝑘𝑙𝑖𝑞 is attached in Appendix C
along with a table of the coefficient values used in the formula for all species.
19
𝑇 2⁄
𝐿𝑜𝑔10 𝑘𝑙𝑖𝑞 = 𝐴 + 𝐵(1 − 𝐶 ) 7 (Equation 5)
After E-501, all the components in stream 4 are in gas phase. Therefore Equation 6 was used to
calculate the thermal conductivity of the species in the line. A table with the coefficients used in
Equation 6 for the different chemical species are attached in Appendix C.
𝑘𝑔𝑎𝑠 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2 (Equation 6)
To obtain the density, Equation 7 from Yaws (1999:198) was used. A table with the coefficient values
used in the formula for all species is attached in Appendix C, as well as a sample calculation.
𝑇 𝑛
𝐷𝑒𝑛𝑠𝑖𝑡𝑦 = 𝐴 × 𝐵 −(1−𝑇𝑐) (Equation 7)
The viscosity of the liquid was obtained by using Equation 8 from Yaws (1999:491). The viscosity of
the compounds in gas phase was obtained by using Equation 9 from the same textbook mentioned
above. The coefficients used in the calculations for each species, per phase, are attached in Appendix
C along with a sample calculation.
𝐵
𝐿𝑜𝑔10 𝜇𝑙𝑖𝑞 = 𝐴 + + 𝐶𝑇 + 𝐷𝑇 2 (Equation 8)
𝑇
𝜇𝑔𝑎𝑠 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2 (Equation 9)
Table 3.5: Physical properties of components in hot stream
Hot stream inlet Unit

HPS
Temperature 526.79 Keshrie
Specific heat 1.991 kJ/kgK
Thermal 0.0359
conductivity W/mK
Density 0.4205 kg/m3
Viscosity 0.00001794 kg/ms
The physical properties of the water vapour (steam) were obtained from Incropera (2007:945).
As mentioned above, Tho is calculated by equating Qc to Qh:
𝑄𝑐 = 113 kW = 𝑄ℎ
𝑄ℎ = 𝑚̇ 𝐶𝑝∆𝑇
113 = 2(kg/s) x 1.991(kJ/kgK) x (526.79 –Tho)K
20
Tho = -(113/(3.982) - 526.79)
= 498.41 K
3.3.3. Assume Value of Overall Coefficient Uo,assume
According to Coulson & Richardson (2005:637), a typical overall coefficient (U) for a shell and tube
heat exchanger vaporiser where the hot fluid is steam and the cold fluid is light organics is between
the range of 900 – 1200 W/m2OC.
Using Kern’s method to design the exchanger, the first assumption made for U is 900 W/ m2OC.
3.3.4. Decide number of shell and tube passes.
The log mean temperature difference, correction factor, F and mean temperature difference were also
determined at this stage.
Making the flow configuration of the fluids in the heat exchanger counter current, the temperature
model is represented by Figure 3.2 below.
Figure 3.2: Temperature profile
∆𝑇2 = 𝑇ℎ𝑜 − 𝑇𝑐𝑖 = 498.41 − 297 = 201.41 𝐾

∆𝑇1 = 𝑇ℎ𝑖 − 𝑇𝑐𝑜 = 526.79 − 522 = 4.79 𝐾
∆𝑇2 −∆𝑇1
∆𝑇𝐿𝑀 = ∆𝑇2 (Equation 10)
ln( )
∆𝑇1
201.41−4.79
= 201.41 = 52.59 𝐾
ln ( )
4.79
Log mean temperature difference is used when there is true counter current flow and applies in cases
of a double pipe and 1 to 1 exchanger with 1 shell and 1 tube pass in co-current or counter current
flow (Geankoplis, 1993:270).
21
In this case, a multiple-pass heat exchanger will be used, to increase hear transfer. It is therefore
necessary to calculate the correction factor (F T). This is used to obtain the correct mean temperature
drop by multiplying it by∆𝑇𝐿𝑀 .
𝑇ℎ𝑖 −𝑇ℎ𝑜
𝑍= 𝑇𝑐𝑜 −𝑇𝑐𝑖
(Equation 11)
526.79−498.41
= = 0.13
522−297
𝑇 −𝑇
𝑌 = 𝑇𝑐𝑜 −𝑇𝑐𝑖 (Equation 12)
ℎ𝑖 𝑐𝑖
522−297
= 526.79−297
= 0.98
∴ 𝐹𝑇 = 0.9 (using the chart attached in Appendix C)
3.3.5. Determine heat transfer area required
The heat transfer area required was calculated using this formula: Ao = q/Uass∆𝑇𝐿𝑀
113 000
𝐴𝑜 = = 2.39 𝑚2
900 × 52.59
3.3.6. Decide type, tube size, material layout and assign fluids to shell or tube side
Both steam and ethylbenzene give relatively clean service, which means cleaning is not a major issue.
Smaller tubes are therefore preferred and they will also give more compact heat exchangers, which
will therefore reduce costs. Triangular tube pattern allows for a more turbulent flow on the shell side
which increases heat transfer. Also since cleaning is not a concern, this is the preferred option in this
case. According to Coulson & Richardson (2005:646), the recommended tube pitch is 1.25 times the
tube diameter, especially if heavily fouling fluids are not being used.
Carbon steel tubes were chosen as the material of construction as the fluids are not excessively
corrosive and do not require a more corrosive resistant material. The tube outer diameter chosen was
16mm with a wall thickness of 1.6mm. The preferred tube length includes 1.83m and the optimum
tube length to shell diameter usually falls within the range of 5 to 10 (Coulson & Richardson
2005:646).
The steam will flow on the tube side and the hydrocarbons will flow in the shell side, as per the
reasons below. The steam is the hotter fluid which is required to give its heat to the cold fluid. Placing
the hotter fluid on the shell side reduces the heat it will lose through the shell of the exchanger. As
high pressure steam has a higher pressure than the hydrocarbons, according to Coulson & Richardson
22
(2005:660) it should be allocated to the tube side because high pressure tubes are cheaper than a
higher pressure shell.
Table 3.6: Summary of tube specifications
Tube pattern Triangular

Tube pitch 1.25 times tube diameter = 20 mm
Tube material Carbon steel
Tube diameter 16 mm
Tube wall thickness 1.6 mm
Tube length 1.83 m
Tube side fluid High pressure steam
Shell side fluid Light hydrocarbons
Shell diameter 0.4 m
3.3.7. Calculate number of tubes
Surface area if 1 tube: 𝐴𝑠 = 𝜋𝐷𝐿 = 𝜋(0.016)(1.83) = 0.092 𝑚 2

𝐴0 2.38
Number of tubes: 𝑁𝑡 = 𝐴𝑠
= 0.092 = 26 tubes and approximate to 30 tubes.
Therefore for 2 passes, tubes per pass = 30/2 = 15 tubes
Tube side velocity:

𝜋𝐷 2 𝜋(0.016)2
Cross sectional area: 𝐴𝑐 = 4
= 4
= 0.000201𝑚 2
Area per pass: 0.000201 × 15 = 0.0452 𝑚 2

2𝑘𝑔/𝑠
Volumetric flow: 𝑄̇ = 0.4205𝑘𝑔/𝑚3 = 4.756 𝑚 2 /𝑠
𝑄̇ 4.756
Tube side velocity: 𝐴 = 0.0452 = 105.23 m/s
𝑝𝑒𝑟 𝑝𝑎𝑠𝑠
Because the fluid used is at high pressure and is less dense with a low viscosity, a high velocity was
expected.
3.3.8. Calculate shell diameter

To calculate the tube bundle shell diameter, Equation 13 can be used where D b is bundle diameter
(mm), Nt is number of tubes and do is tube outer diameter. K and n are coefficients from Coulson &
Richardson (2005:649) and is attached in Appendix C.
𝑁
𝐷𝑏 = 𝑑𝑜 ( 𝐾𝑡 )1/𝑛 (Equation 13) (Coulson & Richardson, 2005:648)
30
= 16( )1/2.207
0.249
= 140.3 mm = 0.14 m
23
Shell bundle clearance Dc was then obtained from Figure 3.4 in Appendix C, which is from Coulson
& Richardson (2005:647). A pull-through floating head was chosen as the exchanger has relatively
clean service.
Finally, this gives a shell diameter of: 𝐷𝑠 = 𝐷𝑏 + 𝐷𝑐 = 140.3 mm + 90 mm = 230.3 mm = 0.23 m
3.3.9. Estimate tube-side heat transfer coefficient

𝜌𝑑𝑣 0.42(0.016)(105.23)
𝑅𝑒 = = = 39464.18 = 39 500
𝜇 0.00001794
Pr = 0.9955 (Incropera, 2007:945)
L/di = 1.83/(0.016-0.0016) =127.08
JN = 0.0035 from “figure 12.23” in Coulson & Richardson (2005:665). This figure is attached as
Figure 3.5 in Appendix C.
𝑁𝑢 = 𝐽𝑛. 𝑅𝑒. 𝑃𝑟 0.33
= 0.0035 × 39500 × 0.99550.33 = 138.04
𝑁𝑢𝑘 138.04×0.0359
ℎ𝑖 = 𝑑𝑖
= 0.016
= 309.73 𝑊/𝑚 2 𝐾
3.3.10. Decide baffle spacing and estimate shell-side heat transfer coefficient
Segmental baffle was selected for this exchanger, as shown in Figure 3.3. According to Coulson &
Richardson (2005:650), the optimum baffle spacing will usually be between 0.3 to 0.5 times that of
the shell diameter. Therefore, in this case the spacing will be: BS = 230.3 mm x 0.4 = 92.12 mm
(0.092 m). This should allow for good heat transfer without a pressure drop that is too high. Although
a close baffle spacing will increase the heat transfer coefficients, this would also cause a higher
pressure drop in the exchanger.
Figure 3.3: Segmental baffle (Coulson & Richardson, 2005:651)
Since the shell side fluid is a mixture, the Reynolds number must be obtained by getting the average
fluid properties based on the proportion, or fraction, of each fluid in the mixture. The mean values of
24
the properties listed in Table 3.4 above were used as the individual properties of each chemical
species below.
Table 3.7: Calculation of density and viscosity of stream based on mole fraction of species in fluid
kmol/hr mol frac 𝜌 𝜌 ×frac 𝜇 𝜇 ×frac

EB 611 0.998 745 744 0.0553 0.0552
S 1.12 0.002 521 0.952 0.0559 0.000102
T 0.11 0.00018 696 0.125 0.0614 1.102E-05
Total 612.23 1 745 0.0553
All the equations below were obtained from Coulson & Richardson (2005:684).
𝑃𝑡 −𝑑𝑜
Shell surface area: 𝐴𝑠 = 𝑃𝑡
× 𝐷𝑠 × 𝐵𝑠
20−16
= × 230.3 × 92.12 = 4243.05 𝑚𝑚 2 = 0.00424 𝑚 2
20
1.10
Shell equivalent diameter:De = (Pt 2 − 0.917𝑑𝑜 2 )
𝑑𝑜
1.10
= (202 − 0.917(16)2 ) = 113.608 𝑚𝑚 = 0.13 𝑚
16
Shell volumetric Flow rate: 𝑄̇ = 𝜌̇ =

𝑚 𝑛̇ ×𝑀𝑤 (611×106.18)+(1.12×104.16)+(0.11×92.15)
𝜌
= 744.654
65002.78 𝑚3 𝑚3
= 733.654
= 88.60 ℎ𝑟 = 0.0246 𝑠
𝑄̇ 0.0246
Shell side velocity: 𝑣𝑠 = 𝐴 = 0.00424 = 5.802 𝑚/𝑠
𝑠
𝜌𝑑𝑣 744.654(0.13)(5.802)
𝑅𝑒 = 𝜇
= 0.0553
= 10 151.29 = 10 200
Table 3.8: Calculation of heat capacity and thermal conductivity of stream based on mole fraction of
species in fluid
Species kmol/hr mol frac Cp Cp x frac K k x frac
EB 611 0.998 0.198 0.198 0.0717 0.0716
S 1.12 0.00183 0.190 0.000347 0.0768 0.00014
T 0.11 0.00018 0.167 2.99E-05 0.0748 1.34E-05
Total 612.23 1 0.198 0.0717
The average fluid properties based on mole fraction, of each fluid in the mixture in the table above
will be used to calculate the Prandtl number of the mixture.
25
𝜇×𝐶𝑝 0.0553×0.198
𝑃𝑟 = 𝑘
= 0.0717
= 0.153
𝑁𝑢 = 𝐽𝑛. 𝑅𝑒. 𝑃𝑟 0.33 = 0.0054 × 10 200 × 0.1530.33 = 29.64
JN = 0.0054 from Coulson & Richardson (2005:672). This figure is attached as Figure 3.6 in Appendix
C.
𝑁𝑢𝑘 29.64×0.0717
ℎ𝑜 = 𝐷𝑒
= 0.013
= 163.5 𝑊/𝑚2 𝐾
3.3.11. Calculate overall heat transfer coefficient including fouling factors, U o,calc
Figure 3.7, from Coulson & Richardson (2005:640) in Appendix C was used to obtain the fouling
factors of steam (oil free) and light organics. These were found to be 10 000 W/m2oC and 5000
W/m2oC respectively.
The conductivity of steel (at 600K) (𝑘𝑤 ) was obtained from Figure 3.4 attached in Appendix C
(Coulson & Richardson, 2005:662).
𝑑
𝑑𝑜 ln( 𝑜 )
1 1 1 𝑑𝑖 𝑑 1 1
= + + + ( 𝑜) ( + )
𝑈𝑜 ℎ𝑜 𝑓𝑜 2𝑘𝑤 𝑑𝑖 ℎ𝑖 𝑓𝑖
0.016
1 1 0.016 ln( ) 0.016 1 1
= 163.5 + 10000 + 2(36)
0.0144
+ (0.0144) (309.73 + 5000) = 0.01 𝑚 2 𝑂 𝐶/𝑊
Uo =100 W/m2K
This is much smaller than the initial guess of 900 W/m2K, therefore a new initial guess using Uo =100
W/m2K must be used until the assumed Uo deviated 30% at most, compared to the calculated Uo.
Using a new assumed overall transfer coefficient of 100 W/m2K and a tube diameter of 8 mm, a new
calculated overall transfer coefficient within 30% of the assumed one was found. This iteration
resulted in a heat exchanger consisting of 448 tubes, a required heat transfer area of 21 m2, shell
diameter of 0.33 m and baffle spacing of approximately 0.1 m.
3.3.12. Estimate tube- and shell-side pressure drops

Using the formula below, the new data used in the second iteration of the overall transfer coefficient,
was used to calculate the pressure drop.
𝐿 𝜌𝑢𝑡 2
∆𝑃𝑡 = 𝑁𝑝 [8𝑗𝑓 (𝑑 ) + 2.5] 2
𝑖
1.83 (745)42.22
= 2 [8(0.0001) ( ) + 2.5] = 3625 𝑘𝑃𝑎
0.008−0.0016 2
26
3.3.13. Cost Estimation of Heat Exchanger
Using the costing chart from Coulson & Richardson (2005,251), the purchase cost of the heat
exchanger was estimated using the materials of construction, pressure factor and type factor. Figure
3.9, attached in Appendix C, is the chart which was used to get the equipment cost using the new
calculated data, however the data is for 1998. The purchase cost was therefore found to be $13 000.
This price was then converted to the reflect that of recent times using the current CEPCI data
(Jenkins, 2019). The price, taking inflation into account, is therefore $19 650. The calculations for
these values are attached in Appendix C.
3.4. Conclusion
The assumed heat transfer coefficient should only be within 30% difference of the calculated heat
transfer coefficient. The results of the second iteration were therefore accepted. From the second
iteration, it was found that with a two pass exchanger, 448 tubes would be needed of 8 mm outer
diameter and 1.83 m in length. The shell diameter was calculated to be 0.33 m with baffle spacing of
0.1 m. Although a close baffle spacing will increase the heat transfer coefficients, this would also
cause a higher pressure drop in the exchanger, which was calculated to be approximately 3630 kPa.
The Purchase cost of the heat exchanger is 19 650 USD.
3.5. References
 Coulson, J., & Richardson, J. 2005. Chemical Engineering Design. Volume 6. 4th Edition. Jordan
hill: Elsevier Butterworth-Heinemann.
 CheGuide. 2016. LMTD Correction Factor Charts. [Online]. Available: https://cheguide.com/lmt
d_charts.html. [2019, April 8].
 Geankoplis, C. J. 1993. Transport Process and Unit Operations, Third Edition. Minnesota:
Prentice-Hall International, Inc.
 Incropera P. F., DeWitt D. P., Bergman, T. L., Lavine, A. S. 2007. Fundamentals of Heat and
Mass Transfer. Pheonix: John Wiley & Sons
 Jenkins, S. 2019. Chemical Engineering-Essentials for the CPI Professional. [Online]. Available:
https://www.chemengonline.com/2019-cepci-updates-january-prelim-and-december-2018-final/
[2019, August 30].
 Mukherjee, R. 1998. Shell and Tube Heat Exchangers: Effectively Design Shell-and-Tube Heat
Exchangers. Chemical Engineering Progress. New York: Engineers India Ltd.
 TLV. Calculator: Saturated Steam Table by Pressure. 2019. A Steam Specialist Company.
[Online] Available: https://www.tlv.com/global/TI/calculator/steam-table-pressure.html. [2019,
April 8].
 Yaws, C (1999). Chemiccal Properties Handbook. New York: Mc Graw-Hill.
27
CHAPTER 4 REACTOR DESIGN
Summary
An isothermal reactor is designed for the production of 280 000 ton/yr of styrene at 99% purity. The
styrene is produced from the dehydrogenation of ethylbenzene at 900 K, 0.4 bar and a steam to
ethylbenzene ratio of 6:1. An isothermal reactor was chosen with a shell and tube design in order to
help maintain a constant optimum temperature inside the tubes where the reaction occurs. The
isothermal reactor was also found to give higher conversion and be more economically operable. The
pressure drop of the reactor was found to be 0.09 bar while the optimum operating reactor pressure is
however 0.4 bar. The percentage pressure drop is therefore 22.9%. Since the calculated pressure drop
along the reactor is greater than 10%, this design will not be economically operable. Different
dimensions can therefore be considered. Dp can be decreased, as long as ratio Dt/ Dp > 15. The shell-
side pressure drop of the reactor was however found to be 0.18(10 -5) bar. With the reacting mixture
being on the tube side of the reactor, the tube volume should be greater than the reactor volume, as per
Silla (2003). This was the case as the tube volume was 6.85 m3 and the reactor volume was 6.77 m3.
4.1. Problem Statement

A styrene production unit must produce 280 000 ton/yr of styrene at a purity of 99%. A reactor is to
be designed to facilitate the dehydrogenation of ethylbenzene to produce styrene. Since this reaction is
endothermic, the nature of the reaction will cause the temperature to decrease. The temperature in the
reactor must be maintained at the optimal temperature in order to ensure the continuity of the reaction
as the reactor operates continually. The appropriate reactor must therefore be designed for this
process.
4.2. Introduction
In a Unit, defined as Unit 500, a reactor is to be designed to facilitate the formation of styrene from
the dehydrogenation of ethylbenzene. The reaction equation is given below. This reaction is
reversible, endothermic and nonequimolar according to the design brief. It also yields a side reaction
where methane and toluene are produced from the hydrogen, formed in the desired reversible
reaction, and ethylbenzene.
Primary Reaction: C8H10 ↔ C8H8 + H2
Ethyl benzene Styrene Hydrogen
Side Reaction: C8H10 + H2 → C7H8 + CH4
Ethyl benzene Hydrogen Toluene Methane
28
In previous chapters, design optimization variables such as the steam ratio and reactor inlet
temperature have been considered. In this chapter reactor size will be determined, as it is also one of
the main design optimisation variables of this process.
The design of the chemical reactor must satisfy certain requirements. The reaction kinetics is one such
requirement. The design must ensure sufficient residence time for the reaction to proceed to the
required degree of conversion. Another requirement is the mass transfer factors. The reaction rate of a
heterogeneous mixture may be controlled by the rates of diffusion of the reacting species; rather than
the chemical kinetics. Heat transfer factors such as the removal, or addition, of the heat of reaction
must also be considered, as well as the safety factors, i.e. the confinement of hazardous reactants and
products. The control of the reaction and the process conditions must be assessed as well (Coulson &
Richardson, 2005:481).
4.3. Literature Review
4.3.1. Continuous and Batch Processing

A continuous process is a process where the reactants are continuously fed to the reactor while the
products are continuously being removed. There is no accumulation, when the reactor is operating
under steady state conditions. A disadvantage is that it does not have the same flexibility as a batch
operation. Advantages include that this type of processing can operate at lower cost, and are preferred
for large scale operation.
All the reactants are added at the start of a batch process. The reaction then proceeds, causing the
compositions to change with time, until the required conversion is reached. The reaction is then
stopped and the product is removed. This type of process is suitable for small-scale production and for
processes where the same equipment is to be used to produce a range of different products, or grades
(Sinnott & Towler, 2008:627).
Semi-continuous or semi-batch processes refer to processes that are neither batch or continuous. In a
semi-continuous process, the process may be interrupted periodically – for example, to regenerate the
catalyst. In a semi-batch process, some of the reactant may be added or some of the product may be
removed as the reaction proceeds (Coulson & Richardson, 2005:482).
4.3.2. Types of Reactors

The flow pattern and manner of contacting phases is regarded as the reactor geometry, or “type of
reactor”. This includes the stirred tank reactor, tubular reactor, packed bed reactor (fixed and moving)
and fluidised bed (Coulson & Richardson, 2005:481).
29
4.3.1.1. Batch Reactor
A batch reactor has no inlet or outlet flow of reactants or products while the reaction occurs. When the
mixture is perfectly mixed, there is no variation in reaction rate throughout the reactor volume.
According to Fogler (2004:8), batch reactors include constant volume (variable-pressure) reactors and
constant pressure (variable-volume) reactors. An advantage of the batch reactor is the high
conversions that can be obtained by leaving the reactant in the reactor for long periods of time. The
disadvantages include the high labour costs per batch and the challenges which large scale production
brings (Fogler 2004:16).
4.3.1.2. Continuous Stirred Tank Reactor (CSTR)

CSTR or agitated reactors consist of a tank fitted with a mechanical agitator and cooling coils or a
jacket. Tank sizes range from small (a few litres) to large (a few thousand litres). These are used for
homogeneous and heterogeneous reactions involving liquids or liquid and gas, as well as reactions
involving finely suspended solids, held in suspension due to the agitation. CSTRs are suitable for
reactions where good mass or heat transfer is required. This reactor is normally run at steady state and
is well mixed. The exception is very rapid reactions, as this will limit the conversion that can be
achieved in one stage. The required power for agitation will depend on the degree of agitation needed
and will range from about 0.2kW/m3 for moderate mixing to 2kW/m3 for intense mixing (Sinnott &
Towler, 2008:628).
4.3.1.3. Plug Flow Reactor (PFR)

The PRF, or tubular reactor, consists of a cylindrical pipe and is usually operated at steady state. In
the tubular reactor, the reactants are continually consumed as they flow down the length of the
reactor. In general, these are used for gas phase reactions; however they are also suitable for liquid-
phase reactions. If high heat transfer rates are required, small-diameter tubes are used to increase the
surface area to volume ratio. Several tubes may be arranged in parallel, connected to a manifold or
fitted into a tube sheet in a similar arrangement to a shell and tube heat exchanger. For high-
temperature reactions, the tubes may be arranged in a furnace (Sinnot & Towler, 2008:629).
4.3.1.4. Packed Bed Reactor (PBR)

According to Coulson & Richardson (2005:629), the two basic types of PBRs are those where the
solid is a reactant and those where the solid is a catalyst. The amount of catalyst is important to the
rate of reaction. Therefore, for reactor design involving fluid-solid heterogeneous reactions, the
reaction rate is based on the mass of the solid catalyst rather than on the reactor volume - unlike
reactor design calculations involving homogeneous reactions. The reactor volume that contains the
catalyst is of secondary significance according to Fogler (2004:14). The size of industrial packed-bed
catalytic reactors, range from small tubes (a few centimetres in diameter) to large-diameter packed
beds. Packed-bed reactors are also used for gas and gas-liquid reactions. Heat transfer rates in large-
30
diameter packed beds are poor, therefore fluidized beds should be considered where high heat transfer
rates are required.
4.3.1.5. Fluidised Bed Reactor

The upward flow of the reacting fluid holds the solids up in suspension, which promotes good mixing,
and high heat and mass transfer rates. The solid in the fluidised bed may be a catalyst, among other
species. Advantages of this type of reactor include its higher heat transfer rate compared to a fixed
bed and that it can transport large amounts of solids as part of the reaction process (for example,
catalysts being moved to another vessel for regeneration). A disadvantage of fluidised bed reactors is
that they can only be used with particles that are relatively small in size. The design and scale-up of
large diameter reactors is still an uncertain process and design methods are largely empirical (Sinnot
& Towler, 2008:629).
4.3.3. Homogeneous and Heterogeneous Reactions

Homogeneous reactions are reactions where the reactants, products, and any catalyst used are in one
phase such as gaseous or liquid. According to Sinnott & Towler (2008:628), homogeneous gas-phase
reactors will always be operated continuously. On the other hand, liquid-phase reactors may be
continuous or batch. Tubular reactors are usually used for homogeneous gas-phase reactions. Both
tubular and stirred tank reactors are used for homogeneous liquid-phase reactions.
Heterogeneous reactions are reactions where two or more phases exist. A predominant concern in
these reactor designs is the promotion of mass transfer between the phases. The possible combinations
of phases include liquid-liquid, liquid-solid, liquid-solid-gas, gas-solid and gas liquid phases.
According to Sinnott & Towler (2008:628), a gas-sold phase reaction is where the solid may take part
in the reaction or act as a catalyst. The dehydrogenation of ethylbenzene to form styrene is classified
as a gas-solid reaction. Although all the reactants and products in the reactor will be in gas phase, the
iron catalyst will be in solid phase.
4.3.4. Adiabatic versus Non-adiabatic Reactor

In order to maintain the constant optimum temperature at which the reaction must occur, multiple
adiabatic reactors can be used with reheating stages between the reactors. According to Luyben
(2011:1231), high temperatures and low pressures are conducive to high conversion in the adiabatic
vapour phase reactors. Steam is mixed with the ethylbenzene feed before the reactors. This is done in
order to lower the partial pressure of ethylbenzene and increase the conversion. The reaction rates of
the by-products produced by the side reaction mentioned above increase with temperature and partial
pressures.
31
For non-adiabatic (isothermal) reactors, the desired conversion is achieved by changing the inlet
reactants temperature (Savoretti et al., 1998:209). For adiabatic reactors, each level of conversion is
obtained by varying the inlet temperatures – at the first and second bed for example. According to
Savoretti et al. (1998:209), the conversion increases with the decrease in selectivity, for both reactor
designs, as illustrated in Figure 4.1 below. A progressive increase in temperature causes this
behaviour but this is necessary to increase the conversion. Higher selectivity is shown by the
isothermal reactor compared to the adiabatic unit for the analysed range of ethylbenzene conversion.
The selectivity improvements range from 2.87% to 4.14%. It was found that by keeping the
ethylbenzene feed constant, any increase in selectivity led to a proportional increase in the production
rate of styrene (Savoretti et al., 1998:209).
Figure 4.1: Styrene selectivity (σ) as a function of ethylbenzene conversion (x_EB) (Savoretti et al.,
1998:210).
Figure 4.2 below shows the temperature profiles inside the two reactors for the same outlet
conversion, (points A and B from Figure 4.1). The corresponding selectivity profiles are shown in
Figure 4.3. The isothermal reactor showed a small temperature decrease near the reactor entrance
followed by a temperature increase up to the reactor outlet. The continuous heating along the reaction
path allows the desired conversion level to be reached with lower temperatures in the first half of the
reactor in comparison to the adiabatic bed. These lower temperatures lead to higher selectivities at
conditions of high concentration of ethylbenzene. In spite of the decrease in selectivity in the last part
of the reactor, the isothermal design shows higher outlet selectivity than the adiabatic beds, as shown
in Figure 4.3.
32
Figure 4.2: Temperature profiles for adiabatic and non-adiabatic (isothermal) reactors (Savoretti et al.,
1998:210).
Figure 4.3: Styrene selectivity profiles for points A and B of Figure 4.1 (Savoretti et al., 1998:210).
Figure 4.4 shows the inlet reactants temperature as a function of conversion. Since the ethylbenzene
feed rate and the steam consumption are the same for both designs used in Savoretti et al. (1998), the
inlet reactants temperature becomes the manipulated variable to achieve the desired conversion level.
For the given conversion, the isothermal reactor achieves the same conversion as the adiabatic reactor,
at temperatures up to 77 K lower than that of the adiabatic beds. In industry, the homogeneous
reactions become important at the inlet zone of the first adiabatic bed where high temperatures and the
highest ethylbenzene concentrations occur (Savoretti et al., 1998:211). This operating problem is
worsened by catalyst deactivation. In practice, the inlet temperature is progressively raised to avoid
excessive conversion losses, up to temperature levels for which the operation must be shutdown to
replace the catalyst. The isothermal reactor design would extend the catalyst life due to lower inlet
temperatures.
33
Figure 4.4: Inlet reactants temperature, considering the same EB feed flow rate (Savoretti et al.,
1998:211).
Under similar operating conditions, the isothermal reactor has to higher selectivity than the adiabatic
reactor. The difference between the income for selling styrene and the cost of the ethylbenzene shows
how the increase in selectivity influences the economy of the styrene plant. Assuming that, for a given
conversion and the same steam consumption, the recycle and separation costs are approximately the
same, Savoretti et al. (1998:211), states that the comparison of the economics for both designs allows
one to know the economical benefits of one reactor in comparison to the other. For a constant of
ethylbenzene inlet flow rate, the profit for each conversion is illustrated in Figure 4.5, for both reactor
designs. Profit (Pr) is equated to the economic function (𝜑) of an isothermal reactor (NA) minus that
of an adiabatic reactor (A).
Figure 4.5: Increased profit by using the isothermal reactor instead of adiabatic units Savoretti et al.,
1998:211).
4.4. Design of Reactor

The reaction in the reactor will be a heterogeneous reaction, as all the products and reactants will be in
gaseous phase while the catalyst used will be in a solid phase. A packed-bed flow reactor has been
selected as the best option as this will support the continuous processing in the production of styrene,
34
which includes the use of a catalyst. Also, an isothermal reactor has been chosen as it was found to be
the best option, as per the discussion below.
The optimum conditions chosen for the reactor was a temperature of 900 K, a pressure of 0.4 bar and
a steam to ethylbenzene ratio of 6:1. The conditions which yielded the highest conversion were shown
to produce a negative preliminary gross profit. The conditions which yielded a profit and a high
conversion was shown to yield a profit 17% lower than that of the chosen conditions, while having a
conversion that is only 6% higher than that of the chosen conditions mentioned above. Tables 4.7 and
4.8 with the data discussed above are attached in Appendix D. The calculations thereof are attached in
Chapter 1. An ethylbenzene feed to the reactor of 408 kmol/h was calculated to produce the desired
capacity of 280 000 ton/yr of styrene. This flow rate was therefore used in the calculations, and
calculations for this are included in chapter 1. Figure 4.6 below provides the material and energy data
for the inlet and outlet streams of the reactor. For the reactor in Figure 4.6, the catalyst, iron oxide, is
packed in the tubes and heat transfer fluid flows in the shell to add heat.
Figure 4.6: Material and energy data for reactor process streams.
35
Main Reaction: 𝐴 ↔ 𝐵 + 𝐶
Where A: ehtylbenzene
B: styrene
C: hydrogen
4.4.1. Performance Equation

According to Fogler (2004: 15), the performance equation for a packed-bed reactor is as follows:
𝑑𝑁𝐴
𝐹𝐴𝑜 − 𝐹𝐴 + ∫ 𝑟 ′𝐴 𝑑𝑊 =
𝑑𝑡
𝑑𝑁𝐴
No accumulation at steady state, so 𝑑𝑡
=0
𝑑𝐹𝐴
= −𝑟𝐴
𝑑𝑊
Including conversion, 𝐹𝐴 = 𝑑𝐹𝐴𝑜 𝑋
𝑑𝑋
𝐹𝐴𝑜 = −𝑟𝐴
𝑑𝑊
𝑑𝑋
∫ 𝐹𝐴𝑜 = ∫ 𝑑𝑊
−𝑟𝐴
𝑋
𝑑𝑋
𝑊 = 𝐹𝐴𝑜 ∫
0 −𝑟𝐴
4.4.2. Initial Concentration of Ethylbenzene

The reaction takes place in gas phase, therefore the ideal gas equation will be used to obtain
concentration.
Ideal gas equation: PV = nRT
𝑛 𝑃
Concentration: 𝐶𝐴𝑜 = 𝑉 = 𝑅𝑇
Pressure: 𝑃 = 𝑃𝐴𝑜 𝑦𝐴
𝑃𝐴𝑜 𝑦𝐴
Therefore, 𝐶𝐴𝑜 = 𝑅𝑇
36
1 bar = 0.986923 atm
0.986923 atm
(0.4 𝑏𝑎𝑟 × )(0.495)
𝐶𝐴𝑜 = 𝑏𝑎𝑟
0.082 𝑑𝑚 3 . 𝑎𝑡𝑚
( 𝑚𝑜𝑙. 𝐾 ) (900 𝐾)
= 0.003 𝑚𝑜𝑙/𝑑𝑚3
Change in volume, due to phase of reaction (gaseous).
1−𝑋
∴ 𝐶𝐴 = 𝐶𝐴𝑜 ( )
1 + 𝜀𝑋
𝐹𝐴𝑜 408
Extent: 𝜀 = 𝑦𝐴 𝛿 𝛿 = 1+1−1= 1 and 𝑦𝐴𝑜 = = = 0.143
𝐹𝑇 2860
1−𝑋
𝐶𝐴 = 0.003 ( )
1 + 0.143𝑋
4.4.3. Reaction Rate

The following reaction rate equations are given in the design brief and were obtained from Kent
(2007).
21.708
For forward reaction: 𝑟1 = 1.177 × 108 𝑒 (− 𝑅𝑇
)𝑃𝑒𝑏
7804
( )𝑃𝑠𝑡𝑦 𝑃ℎ𝑦𝑑
For reverse reaction: 𝑟−1 = 20.965 × 108 𝑒 𝑅𝑇
According to Fogler (2004:41), the rate of reaction can be expressed as a function of conversion.
Following an elementary rate law: −𝑟𝐴 = 𝑘𝐶𝐴
Stoichiometry is used to express concentration as a function of conversion.
Gas phase isothermal: T = T o
1−𝑋 𝑃
−𝑟𝐴 = 𝑘𝐶𝐴𝑜 (1+ 𝜀𝑋) 𝑃
𝑜
1−𝑋 𝑃
= 𝑘0.003 ( )
1 + 0.143𝑋 𝑃𝑜
For gas-phase reactions with pressure drop,
𝑃 𝑑𝑦 𝛼
𝑥=𝑃 and 𝑑𝑊
= − 2𝑥 (1 + 𝜀𝑋)
𝑜
37
To find reaction rate constant,
7804 1−𝑋
−𝑟𝐴 = 20.965 × 108 𝑒 ( 𝑅𝑇
)𝑃𝑠𝑡𝑦 𝑃ℎ𝑦𝑑
= 𝑘𝐶𝐴𝑜 ( )
1 + 𝜀𝑋
X = 0.87
𝐶𝐴𝑜 = 0.003 𝑚𝑜𝑙/𝑑𝑚 3
𝜀 = 0.143
7804
20.965 × 108 𝑒 ( 𝑅𝑇 ) 𝑃𝑠𝑡𝑦 𝑃ℎ𝑦𝑑
𝑘=
1−𝑋
𝐶𝐴𝑜 (1 + 𝜀𝑋)
7804
20.965 × 108 𝑒 (8314×900) (0.39477 × 0.4)(0.39477 × 0.007)
𝑘=
1 − 0.87
0.003 ( )
1 + (0.143 × 0.87)
𝑘 = 20.56 𝑠 −1
Plotting conversion (X) versus reaction rate inverse:
Table 4.1: Conversion versus rate of reaction data
X (-ra 1/-ra
0 0.06168 16.21271
0.1 0.054729 18.27173
0.2 0.047972 20.84549
0.3 0.0414 24.15462
0.4 0.035006 28.5668
0.5 0.028782 34.74384
0.6 0.022722 44.0094
0.7 0.01682 59.45201
0.8 0.01107 90.33722
0.9 0.005465 182.9929
1 0 #DIV/0!
38
rate of reaction Inverse (1/-ra) 200
180
160
140
120
100
80
60
40
20
0
0 0.2 0.4 0.6 0.8 1
Conversion (X)
Figure 4.7: Graphical representation of conversion versus rate of reaction data
According to Fogler (2004:45), the area under the graph above, will give the tubular reactor volume
necessary to produce a specific conversion. The reactor decided upon is however a backed bed
reactor.
4.4.4. Catalyst Weight

As shown in figure 4.6, the molar flow rate of ethylbenzene into the reactor is 408 kmol/hr
𝑘𝑚𝑜𝑙 1000𝑚𝑜𝑙 1ℎ𝑟

𝐹𝐴𝑜 = 408 × × = 113.33 𝑚𝑜𝑙/𝑠
ℎ𝑟 1𝑘𝑚𝑜𝑙 3600𝑠
Given in design brief: Bulk catalyst density: 𝜌 = 1282 kg/m3
Catalyst void fraction: 𝜑 = 0.4
𝑋
𝑑𝑋
0 −𝑟𝐴
𝑋
𝑑𝑋
1−𝑋
0 𝑘𝐶𝐴𝑜 (1 + 𝜀𝑋 )
𝐹𝐴𝑜 𝑋 𝑑𝑋
= ∫
𝑘𝐶𝐴𝑜 0 ( 1 − 𝑋 )
1 + 𝜀𝑋
0.87
𝐹𝐴𝑜 1
= (1 + 𝜀) [ln − 𝜀𝑋]
𝑘𝐶𝐴𝑜 1−𝑋 0
113.33 𝑚𝑜𝑙/𝑠 1
= 3
[(1 + 0.143) ln − 0.143𝑋]
(20.56 /𝑠)(0.003 𝑚𝑜𝑙/𝑑𝑚 ) 1−𝑋
39
𝑊 = 4056.145 ≈ 4057 𝑑𝑚 3
𝑘𝑔 𝑚3 3
∴ 𝑊𝑒𝑖𝑔ℎ𝑡 = 𝑉𝑜𝑙 × 𝜌 = 4057 𝑑𝑚 × 1282 3 × 3 = 5199.98 𝑘𝑔
𝑚 10 𝑑𝑚 3
∴ 𝐶𝑎𝑡𝑎𝑙𝑦𝑠𝑡 𝑤𝑒𝑖𝑔ℎ𝑡 = 5200 𝑘𝑔
4.4.5. Catalyst Bed Volume

𝑤𝑒𝑖𝑔ℎ𝑡 5200
𝑉𝑜𝑙𝑢𝑚𝑒 = = = 4.06 𝑚 3
𝑑𝑒𝑛𝑠𝑖𝑡𝑦 1282
4.4.6. Space Time

The equation below is from Fogler (2004:58).
3 103 𝑑𝑚 3
𝑊 𝑊. 𝐶𝐴𝑜 5200 𝑘𝑔 × 0.003 𝑚𝑜𝑙/𝑑𝑚 × 𝑚3
𝜏= = =
𝑣𝑜 𝐹𝐴𝑜 𝑚𝑜𝑙
(113.33 )
𝑠
𝜏 = 137.65 𝑘𝑔. 𝑠/𝑚 3
4.4.7. Reactor Volume

𝑉𝑐𝑎𝑡 4.06
𝑅𝑒𝑎𝑐𝑡𝑜𝑟 𝑣𝑜𝑙𝑢𝑚𝑛: 𝑉𝑟 = = = 6.77 𝑚 3
1 − 𝜑 1 − 0.4
4.4.8. Number of Tubes

A tube length of 4.9 m (16 ft) was chosen, as the preferred lengths of tubes, according to the TEMA
standard, are 6, 8,12, 16, 20 and 24 ft. According to Silla (2003), a tube diameter based on a PFR
design, must be assumed and the ratio of tube diameter to diameter of particle must be greater than 15.
Based on Huang (2015), the particle size of the catalyst (Dp) is 4 mm. A tube diameter (Dt) of 60 mm
was therefore chosen.
𝜋𝐷𝑡 2 𝜋(0.060)2
Volume of one tube: 𝑉𝑡 = 𝐴𝑐 𝐿𝑡 = 4
𝐿𝑡 = 4
4.9 = 0.014 𝑚 3
𝑉𝑟 6.77
Number of tubes: 𝑁𝑡 = = = 489 tubes
𝑉𝑡 0.014
4.4.9. Reactor Shell Height

According to Sinnott (2005), a range of 20-50% of the shell height is kept as allowance for reactor
height. An allowance of 25 % is therefore chosen for this reactor.
Reactor shell height: 𝐻𝑠 = 𝐿𝑡 + 𝐶𝑙𝑒𝑎𝑟𝑎𝑛𝑐𝑒 ℎ𝑒𝑖𝑔ℎ𝑡 + 𝐴𝑙𝑙𝑜𝑤𝑎𝑛𝑐𝑒 ℎ𝑒𝑖𝑔ℎ𝑡
40
= 4.9 + 2 + 0.25(4.9) = 8.125 = 8.13 𝑚
4.4.10. Reactor Diameter

The equation below will be used to calculate the reactor diameter. According to Sinnot (2005), the
following values are given for triangular tube pitch: K1 = 1.08, K2 = -0.9, K3 = 0.69 and K4 = -0.8.
(𝐷𝑠 − 𝐾1 )2 − 𝑃𝑡 (𝐷𝑠 − 𝐾1 )(𝑛𝐾1 + 𝐾4 )
𝑁𝑡 =
1.23 (𝑃𝑡 )2
n: Number of passes =1
𝑃𝑡 = 1.25 × 𝐷𝑡
(𝐷𝑠 − 1.08)2 − (1.25 × 0.06)(𝐷𝑠 − 1.08)(1.08 + (−0.8))

489 =
1.23 (1.25 × 0.06)2
𝐷𝑠2 − 2.16𝐷𝑠 + 1.1664 − 0.006075𝐷𝑠 + 0.02268

489 =
0.00691875
3.383 = 𝐷𝑠2 − 2.166075𝐷𝑠 + 1.18908
𝐷𝑠2 − 2.166075𝐷𝑠 − 2.19392 = 0
−𝑏±√𝑏 2−4𝑎𝑐
Using quadratic formula: 𝑥 = 2𝑎
2.167 ± √(−2.167)2 − 4(1)(−2.194)

𝐷𝑠 =
2(1)
∴ 𝐷𝑠 = 2.62 𝑚
4.4.11. Length to Diameter Ratio

According to Silla (2003), for a plug flow reactor, the length to diameter ratio is in the range of 3 - 5.
𝐻 8.13
Ratio: 𝐷𝑠 = 2.62 = 3.1
𝑠
This design is therefore acceptable.
4.4.12. Pressure Drop

The pressure drop of this reactor should be less than 10% of the operating pressure, in order for it to
be economically operable (Silla, 2003). The Ergun equation is used to calculate the pressure drop over
the length of the reactor (Fogler, 2004).
∆𝑃 150 𝜇 𝐺 (1 − 𝜑)2 𝐺 2 (1 − 𝜑)
= + 1.75
𝐿 𝐾. 𝑔. 𝜌 𝐷𝑝 2 𝜑3 𝐾. 𝑔. 𝜌 𝐷𝑝 𝜑3
41
𝑚 87600𝑘𝑔/ℎ×ℎ/3600𝑠
Mass velocity: 𝐺 = 𝐴.𝑁 = 𝜋(0.06)2
= 17.6 𝑘𝑔/𝑚 2 𝑠
𝑡 ×489
4
∆𝑃 150 𝜇 (17.6)(1 − 0.4)2 (17.6)2 (1 − 0.4)

= + 1.75
8.13 (0.87)(9.81)𝜌 (0.004)2 0.43 (0.87)(9.81)𝜌 (0.004) 0.43
The equation below, from Yaws (1999:198), was used to obtain the density. A table with the
coefficient values used in the formula for all species is attached in Appendix D.
𝑇 𝑛
𝐷𝑒𝑛𝑠𝑖𝑡𝑦 = 𝐴 × 𝐵−(1−𝑇𝑐 )
𝑇 𝑛
Ethylbenzene: 𝜌 = 𝐴 × 𝐵 −(1−𝑇𝑐)
297.972 0.2921
= 0.28889 × 0.26438−(1− 617.17 )
= 1.0754 g/ml
= 1.0754 kg/m3
Table 4.2: Coefficients of species for density formula

Units: g/mL A B n Tc Tmin Tmax
EB 0.289 0.264 0.292 617 178 617
The viscosity of the compounds in gas phase was obtained by using the equation below from Yaws
(1999:491).
Viscosity for liquid:
Ethylbenzene: 𝜇𝑔𝑎𝑠 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2
= −4.267 + (0.24735 × 521.79) + (−5.4264E − 05)(521.79)2
= 110 cP
= 0.11 kg/ms
Table 4.3: Coefficients of species for viscosity formula in gas phase

Units: cP A B C Tmin Tmax ngasT25 ngasTmin ngasTmax
EB -4.267 0.247 -5.43E-05 409 1000 --- 87.8 189
8.13 × 150 (0.11) (1.76)(1 − 0.4)2 8.13 × 1.75 × (1.76)2 (1 − 0.4)

∆𝑃 = +
(0.87)(9.81)(1075.4) (0.004)2 0.43 (0.87)(9.81)(1075.4)(0.004) 0.43
42
= 9155.93 𝑃𝑎 = 0.0916 bar
Operating pressure of reactor is 0.4 bar.
0.0916
Percentage pressure drop: 0.4
× 100 = 22.9 %
Since the calculated pressure drop along the reactor is greater than 10%, this design will not be
economically operable. Different dimensions can therefore be considered. D p can be decreased, as
long as ratio Dt/ Dp > 15.
4.4.13. Tube Volume

With the reacting mixture being on the tube side of the reactor, the tube volume should be greater than
the reactor volume (Silla, 2003).
Total tube volume: 𝑉𝑡 = 0.014 𝑚 3 × 489 𝑡𝑢𝑏𝑒𝑠 = 6.85 𝑚 3
The reactor volume was calculated above to be 6.77 𝑚 3
∴ 𝑉𝑡 > 𝑉𝑟
4.4.14. Shell-side Pressure Drop

The cross flow area is calculated using Kern’s method with the formula below. 5 baffles are assumed
for this reactor as it will increase the heat transfer by encouraging turbulent flow.
(𝑃𝑡 − 𝐷𝑡 )𝐷𝑠 𝑙𝐵
𝐴=
𝑃𝑡
𝐷𝑠
((1.25𝐷𝑡 ) − 𝐷𝑡 )𝐷𝑠 ( )
= 5
(1.25𝐷𝑡 )
2.62
((1.25 × 0.06) − 0.06)2.62 ( )
= 5 = 0.275 𝑚 2
(1.25 × 0.06)
As per the suggestion of the design brief, molten salt will be used as the heating fluid on the shell side
of the reactor. Table 4.4 below includes the fluid properties of the molten salt.
Table 4.4: Fluid properties of Molten Salt (Serrano-Lopez, 2013)
ρ: Density 2000 kg/m3

µ: Viscosity 0.00157 kg/m.s
Pr: Prandtl Number 5.024
k: Thermal Conductivity 0.5
43
From the material balance, Mass flow rate: m = 163.6 kg/h
From the energy balance from Chapter 2, the feed mass flow rate of the molten salt is:
kmol 𝑘𝑔 1ℎ
2448 ℎ
× 18.02 𝑘𝑚𝑜𝑙
× 3600 𝑠 = 12.25 𝑘𝑔/𝑠
𝑚𝑠 𝑚𝑠
Shell side mass velocity: 𝐺𝑠 = = 𝜋𝐷𝑠 2
𝐴
4
12.25
=
𝜋 2.62)2
(
4
12.25
= = 2.27 𝑘𝑔/𝑚 2 𝑠
5.39
𝐺 2.27
Shell side velocity: 𝑣𝑠 = 𝜌𝑠 = 2000 = 0.0011𝑚/𝑠
𝑠
Shell side equivalent diameter:

𝑃𝑡 1 𝑑2
4( × 0.87𝑃𝑡 − 𝜋 𝑜 )
2 2 4
𝑑𝑒 =
𝑑𝑜
𝜋 4
𝑃𝑡 = 1.25𝐷𝑡 = 0.075 𝑚
𝑑𝑜 = 0.06 𝑚
∴ 𝑑𝑒 = 0.088 m
Reynolds number for shell-side:

𝐺𝑠 𝑑𝑒
𝑅𝑒 =
𝜇
2.27 × 0.088
= = 127.24 ∴ 𝑇𝑢𝑟𝑏𝑢𝑙𝑒𝑛𝑡 𝑓𝑙𝑜𝑤
0.00157
The friction factor and heat transfer factor charts, attached in Appendix D, are the charts from which
the factors were obtained.
Ff = 0.2 jn = 0.04
Nusselt number:
𝜇
𝑁𝑢 = 𝑗ℎ 𝑅𝑒𝑃𝑟 1⁄3 ( )0.14
𝜇𝑤
44
𝜇
, 𝑎𝑠𝑠𝑢𝑚𝑒𝑑 𝑡𝑜 𝑏𝑒 = 1
𝜇𝑤
𝑁𝑢 = (0.04)(127.24)(5.024)1⁄3 (1)0.14
𝑁𝑢 = 8.72
Heat transfer co-efficient:

𝑁𝑢𝑘 8.72 × 0.5
ℎ= = = 49.55 𝑊/𝑚2 𝐾
𝑑𝑒 0.088
Therefore,
Pressure Drop:
𝐷𝑠 𝐿 𝜌𝑣𝑠2 𝜇 −0.14
∆𝑃𝑠 = 8𝑗ℎ ( )( ) ( )
𝑑𝑒 𝑙𝐵 2 𝜇𝑤
2.62 8.13 2000(0.0011)2

∆𝑃𝑠 = 8(0.04) ( )( ) (1)−0.14
0.088 2.62 2
5
∆𝑃𝑠 = 0.179 𝑃𝑎 = 0.18 × 10−5 𝑏𝑎𝑟
4.5. Discussion
For the reactants and products for the reaction which will be considered here, a batch reactor will not
be a practical choice as the reactants and products are in gas phase. Also, a batch reactor will be more
costly than using a reactor on a continuous or semi-continuous process. This process can be classified
as a semi-continuous process because the process will be interrupted periodically for catalyst
regeneration. According to the design brief, the catalyst maximum temperature is 600 K. Because the
chosen optimum temperature is 900 K, the catalyst will have to be replaced more regularly.
Due to the gas-solid heterogeneous reaction due to the use of the catalyst, a packed bed reactor has
been selected. The reaction, as mentioned, is endothermic which will cause a decrease in temperature
in the reactor. It is however desired to maintain a constant temperature in the reactor which promotes
a high conversion. This temperature is one of the optimal conditions, which are discussed below. To
maintain a constant temperature in the reactor, it was decided to load the catalyst in tubes in which the
reaction will occur, while a heating fluid flows on the shell side of the reactor – similar set up to a
shell and tube heat exchanger. The use of multiple tubes instead of one big tube was selected as it will
be more challenging to ensure that the heat from the fluid outside the large tube reaches the entire
cross sectional area of the tube, in which the reaction will occur. Using many tubes, with a smaller
tube diameter, and therefore cross sectional area, allows the heat to reach more of the reacting species
within the tubes.
45
The optimal reactor conditions, which give the highest conversion, is a temperature of 950 K, a
pressure of 0.4 bar and a steam to ethylbenzene ratio of 12:1. Because the steam ratio is great, the cost
of steam is so high that the preliminary gross profit becomes negative under these conditions – as
shown in the mini project. Low reactor temperature suppresses side reactions but require higher
ethylbenzene recycle in order to achieve the same styrene production rate. This in turn increases
separation costs. Higher steam ratios also suppress side reactions but increase furnace fuel costs and
steam supply costs (Luyben, 2011: 1231). The optimum conditions based on making the highest
preliminary gross profit was therefore shown to be 900 K, 0.4 bar and a 6:1 ratio. Taking the desired
purity of the styrene into account, the optimal conditions are now selected where the conversion is
relatively high at 87% but it is still possible to make a gross profit.
According to Savoretti et al. (1998), at temperatures up to 77 K lower than that of the adiabatic beds,
the isothermal reactor achieves the same conversion as the adiabatic reactor. High temperatures and
the highest ethylbenzene concentrations occur at the inlet zone of the first adiabatic bed therefore, in
industry, the homogeneous reactions become important at the inlet zone (Savoretti et al., 1998:211).
This operating issue is worsened by catalyst deactivation. In practice, the inlet temperature is
progressively raised to avoid excessive conversion losses, up to temperature levels for which the
operation must be shutdown to replace the catalyst. The isothermal reactor design would therefore
extend the catalyst life due to lower inlet temperatures, shown in Figure 4.4, and therefore reduce
costs associated with catalyst replacement.
Cost was also considered in the selection of the reactor resign. The effect that selectivity has on the
economy of the styrene plant can be seen from the difference between the income for selling styrene
and the cost of the ethylbenzene. As mentioned above, the isothermal reactor has to higher selectivity
than the adiabatic reactor. For a constant of ethylbenzene inlet flow rate, the profit for each conversion
is illustrated in Figure 4.5, for both reactor designs.
The pressure drop of the reactor was found to be 0.09 bar while the optimum operating reactor
pressure is however 0.4 bar. The percentage pressure drop is therefore 22.9%. Since the calculated
pressure drop along the reactor is greater than 10%, this design will not be economically operable.
Different dimensions can therefore be considered. Dp can be decreased, as long as ratio Dt/ Dp > 15.
The shell-side pressure drop of the reactor was however found to be 0.18(10 -5) bar. With the reacting
mixture being on the tube side of the reactor, the tube volume should be greater than the reactor
volume, as per Silla (2003). This was the case as the tube volume was 6.85 m3 and the reactor volume
was 6.77 m3.
46
4.6. Conclusion
The isothermal reactor was found to be the best option as it gives higher conversion at lower
temperatures compared to an adiabatic reactor, based on literature. This lower temperature also
prevents the catalyst from having to be replaced too regularly due to damage caused by high
temperatures. This would mean the plant would have to by shut down. Stopping production lowers
income as it reduces the amount of saleable products produced, besides the cost of purchasing and
loading new catalyst. A shell and tube design was chosen for the reactor as it allows for the uniform
constant temperature of the reacting mixture within the tubes. The length of the reactor was found to
be 8.13 m, the diameter was 2.62 m, the volume was 6.77 m3 and the catalyst bed volume was 4.06
m3.
4.7. Recommendation
The pressure drop of the reactor was found to be 0.09 bar while the optimum operating reactor
pressure is however 0.4 bar. The percentage pressure drop is therefore 22.9%. Since the calculated
pressure drop along the reactor is greater than 10%, this design will not be economically operable. It
is therefore recommended that different reactor dimensions be considered. Dp can be decreased, as
long as the Dt/ Dp ratio is greater than 15.
4.8. References
 Fogler, H. S., 2004. Elements of Chemical Reaction Engineering.3rd Edition. New Delhi:
Prentice-Hall.
 Huang, Y. 2015. Properties of B1 Catalyst. Research Gate. [Online]. Available:
https://www.researchgate.net/figure/Properties-of-the-B1-catalyst_tbl1_236931124 [24 April
2019]
 Kent, J.A. 2007. Kent and Riegel's Handbook of Industrial Chemistry and
Biotechnology. Volumes 1 & 2 (11th Edition). Springer: Verlag.
 Luyben, W. L. 2011. Design and Control of the Styrene Process. Ind. Eng. Chem. Res. 50: 1231–
1246.
 Savoretti, A. A., Borio, D.O., Bucala, V., Porra, J.A. 1998. Non-adiabatic radial-flow reactor for
styrene production. Chemical Engineering Science. 54: 205-213.
 Serrano-Lopez, R., Fradera, J., Cuesta-Lopez, S. 2013. Molten salts database for energy
applications.
 Silla, H., 2003. Chemical Process Engineering; Design and Economics. Huboken: USA.
47
 Sinnott R., Towler, G. 2008. Chemical Engineering Design. Principles, Practice and Economics
of Plant and Process Design. London: Elsevier.
 Sinnott, R. K., 2005. Chemical Engineering Design, Voume.6. 4th edition. Butter-worth
Heinemann: London.
48
CHAPTER 5 SEPARATION UNITS DESIGN
Summary
Polystyrene is manufactured from polymerised styrene and is one of the most widely used materials
because of its versatility. Dehydrogenation of ethylbenzene is a common method used to produce
styrene. A distillation column to be used in the styrene production plant was to be designed. In order
to do this, the composition of the feed stream to the column was determined along with the best
operating conditions for the column and reasons for the best choice of column (tray or packing). The
feed stream to the column contained ethylbenzene (13%) and styrene (87%) with trace amounts of
toluene. The mixture was therefore regarded as binary because according to Thakore & Bhatt
(2007:381), if the amounts of other components in the mixture are very low, the mixture can be
considered binary. The operating pressure and temperature was chosen to be 0.2 bar and 358 K. 0.2
bar in order to reduce the bubble point of the distillate (ethylbenzene) to 360 K. This lower pressure
will also reduce condensation costs of ethylbenzene in the condenser with cooling water at 303.15 K.
A sieve tray tower was selected as the best option as the column diameter was found to be greater than
1m and operated at low vacuum, where the use of trays improve vapour-liquid contact in large
diameter towers. Sieve trays are also the most economical choice and, according to Coulson &
Richardson (2005:548), the efficiency of a sieve plate used for an ethylbenzene-styrene system is
75%. In terms of the detailed tower design, the reflux ratio was found to be 18, the number of trays
needed was calculated to be 28 trays, the decided tray spacing was 0.5 m and the weir height was 10
mm. The top and bottom pressure specifications were 0.2 bar and 0.25 bar respectively while the
column diameter was calculated to be 7.7 m at the top of the column and 6.6 m at the base of the
column

Styrene can be polymerised to form polystyrene which is a versatile material with many uses. The
dehydrogenation of ethylbenzene is one of the most common methods used worldwide to produce
styrene. A distillation column to be used in the styrene production plant must be designed. In order to
do this, the composition of the feed stream to the column must be determined along with the best
operating conditions for the column and reasons for the best choice of column (tray or packing). The
detailed tower design, for a tray column for example, must include the reflux ratio, number of trays,
tray spacing, top and bottom pressure specifications, column diameter and weir height.
5.2. Introduction
Polystyrene is a multipurpose plastic used to produce an extensive variety of products for consumer
usage. As one of the most widely used plastics, several billions of kilograms of polystyrene are
produced every year, from styrene. The most common method of manufacturing this economically
49
important industrial chemical is through the dehydrogenation of ethylbenzene. As mentioned in
previous chapters, the dehydrogenation occurs in an isothermal reactor with catalyst before the
products enter a three phase separator.
The three phase separator mainly separates the lighter components (hydrogen and methane) from the
heavier hydrocarbons (toluene, ethylbenzene and styrene) as well as separating the waste water
containing a small amount of organic compounds. Two distillation columns are used in theis process.
The first column (T-501) separates the toluene (distillate) from the ethylbenzene and styrene which go
to the second column (T-502). A process flow diagram is attached in Appendix B. In T-502, the
ethylbenzene (distillate) is separated from the styrene which must have a purity of 99%. This report
will discuss the design of T-502 in detail.
In order to design the column, the composition of the feed stream will be determined, starting at the
three phase separator. From here the key components will be selected as well as the operating
conditions. Comparing the use of a packed column to that of a tray column, the better option for this
column’s specific needs will then be established. From here the degree of separation, reflux ratio,
theoretical number of stages, column diameter, column pressure drop and weir height are determined.
The calculations of the above are attached in Appendix E and the assumptions are explained below.
5.3.1. Flash Column

The stream numbers in this section refer to Figure 5.1 below.
After the reactor, stream 8 enters the flash column containing styrene, hydrogen, methane, toluene,
water and unreacted ethylbenzene. The flash column is used to separate the lighter hydrocarbons,
hydrogen and methane, from the heavier hydrocarbons, styrene and toluene. The operating
temperature and pressure creates the conditions under which the lighter components are in gas phase
and the heavier components are in liquid phase. The water is immiscible in the organic phase, as
stated by the project brief. It was also stated that it must be assumed that methane and hydrogen are
insoluble in streams 10 and 11, therefore all hydrogen and methane are only present in stream 9.
Raoult’s law was therefore used to determine the amount of ethylbenzene, styrene and toluene in gas
phase, under the operating conditions of the flash column. The remaining majority of the styrene,
toluene and ethylbenzene were then taken to be in liquid phase in stream 10.
Although the organics (styrene, toluene and ethylbenzene) and water are both in liquid phase in the
separator, the organic liquids are too non-polar to dissolve thoroughly in the polar water (Schaller,
2012). These liquids therefore form two separate layers and are separated in this way in the flash
drum. Although the properties of the water (stream 11) was assumed to be that of pure water, as stated
50
by the project brief, the costs for treatment of waste water will still be included under costing.
Therefore, the small amount of organics in the water was taken into account in the mass balance
calculations. To determine the amount of organics in the water, the solubility or the organics were
determined in order to apply Henry’s law. According to Sander (2015), Henry’s law states that the
amount of dissolved gas in liquid is proportional to its partial pressure above the liquid. Henry’s law
constant is the proportionality factor.
The solubilities of the hydrocarbons concerned are important because, according to Seader
(2010:164), Henry’s law is not applicable to gases with a high solubility, even when the partial
pressures are low. Since no widely accepted theory for gas-liquid solubilities exists, empirical
correlations and experimental data plots are used. These plots and the partial pressures of the solute
can be used to estimate the solubility in water at various pressures and for gas mixtures. This is
applicable provided that the mole fractions are low and no chemical reactions occur in the gas or
water (Seader, 2010:164).
The solubility of styrene in water is 0.03% according to ChemicalBook (2017). The solubility of
ethylbenzene and toluene in water at the operating temperature of the flash column was found using
the water solubility correlation formula from Yaws (1999:396), as shown below, where S is the
solubility in water in ppm(wt) and T is the temperature in K.
𝐵 𝐶
𝑙𝑜𝑔10 𝑆 = 𝐴 + 𝑇 + 𝑇2
The correlation constant values (A to C) for solubility of hydrocarbons in water are listed in Table 5.1
below. These values apply to conditions of saturation where the hydrocarbon is in equilibrium with
the water. For saturation, the pressure of the system is nearly equal to the total vapour pressure of
water and the hydrocarbon(s) (Yaws, 1999:396).
Table 5.1: Correlation constants (Yaws, 1999:400)

Species A B C
EB 13.37 -6643.67 9.899E05
Toluene 15.47 -7591.27 1.131E06
Data for Henry’s law constant for organic chemicals in water was found in Yaws (1999:403),
according to which, the results are intended for initial engineering and environmental impact studies.
For example, this law can be used to determine the environmental movement and fate of chemicals in
air and water. This translates into the equation below, to find the liquid concentration (x i), using the
gas concentration (yi), total pressure (P) and Henry’s law constant (Hi). Henry’s law constants for the
applicable hydrocarbons are tabulated in Table 5.2 below.
𝑃
𝑥𝑖 = 𝐻 𝑦𝑖
𝑖
51
Table 5.2: Henry’s law constant for compound in water – organic compounds (Yaws, 1999:416)
Species TF TB T H@T (atm/molFrac) H@T (atm/mol/m3)

EB 178.2 409.4 25 452.3 8.141E-03
Styrene 242.5 418.3 25 144.6 2.602 E-03
Toluene 178.2 383.8 25 352.9 6.352 E-03
Polymerisation of styrene is the conversion of styrene to polystyrene via a chemical process of
monomers joining together forming polymers. According to the project brief, to minimise
polymerisation of styrene product, the operating temperature of the flash unit should be between 50 –
100 oC (323.15 K - 373.15 K) and pressure between 0.5 and 1.5 bar.
5.3.2. Distillation column

Distillation is a method of fractionating components in a solution which are relatively volatile.
According to Geankoplis (1993:585) distillation involves, a volatile vapour and liquid phase which
are vaporised and then condensed. Distillation is the preferred choice in cases where the components
are relatively volatile and it is required to be separated in pure form. The separation becomes easier as
the relative volatilities increase. Liquid flows down the column while vapour flows counter-currently
up the column so that the two are brought into contact on plates or packing. The section below the
feed is known as the stripping section as the more volatile components are stripped from the liquid, in
this way increasing the concentration of the more volatile components in the rectifying section, above
the feed. Either all or some of the condensate from the condenser is returned to the top of the column,
known as reflux, which creates the liquid flow in the rectifying section. At the bottom of the column,
the reboiler vaporises some of the liquid which is returned to the column and creates the vapour flow
Although distillation is used worldwide and is very common in chemical industries, there is no
method that is one hundred percent reliable for distillation column design. All methods till date for
designing different distillation systems have limitations, so more than one method is generally used to
design a distillation column. Pilot plants are used for completely new mixtures prior to applying it at a
larger scale (Thakore & Bhatt, 2007:378).
There are several main components which constitute the distillation column. This includes a vertical
shell which houses the internals where the separation of liquid and vapour components occurs. The
internals include elements such as trays and/or packing, used to enhance the separation of
components. A reboiler is considered the driving force of the distillation column, according to
(Lieberman, 2014:75), and is responsible for the vaporisation needed for the distillation process. The
condenser cools and condenses the vapour leaving the top of the column while the reflux drum holds
said condensed vapour in order for the liquid to be recycled back to the column.
52
5.3.2.1. Tray and Packed Tower
A distillation column can either contain trays or packing. Different types of trays include bubble cap
trays, valve trays and sieve trays while different types of packing includes raschig rings, mesh packing
and ceramic packing. According to Thakore & Bhatt (2007:379), the selection criteria below can be
used to determine whether to use a tray tower or packed tower.
A packed tower provides lower pressure drop compared to a tray column, making it more suitable for
distillation under high vacuum conditions. Higher pressure drops can be detrimental as lower
pressures in the reboiler may lead to undesirable thermal cracking of the bottoms product. A packed
tower provides lower residence time for liquid, which is directly proportional to the rate of thermal
cracking – which increases exponentially with temperature. Therefore a reduction in the boiling point,
for a product which is highly sensitive to heat, improves the product quality.
A packed tower is also preferred for a tower with a diameter less than 0.15m, as the fabrication of the
trays for this diameter is difficult, and therefore more expensive. Adding to this, when there is packing
in a tower of a large diameter, the liquid has difficulty distributing. The use of packing (with plastic
packings for example) is more economical compared to trays of special metal alloys, for corrosive
systems.
Tray towers should be selected for very low liquid flow rates and for very low liquid to vapour ratios.
For systems with a higher ratio of liquid to vapour flow rates, it is preferred to use packed towers
which requires minimum wetting rate to work efficiently. Also, the use of trays are preferred if side
products are to be drawn from the distillation column.
5.3.2.2. Method for Design of Distillation Column

The following distillation column design method was adapted from Thakore & Bhatt (2007:380). Two
key components are to be selected from all the components of the feed mixture, if it contains more
than two components. The distillation column’s operating pressure is then chosen and the vapour-
liquid equilibrium data at this pressure obtained. Following this, the degree of separation required is
specified. For binary distillation the top and bottom product compositions is decided. For
multicomponent distillation, the distribution of key components in distillate and residue are chosen, as
well as the non-key components before completing the material balance.
The minimum reflux ratio is then determined along with the optimum reflux ratio. The number of
theoretical stages required for the desired separation must then be found before selecting the tower
type (tray or packing) and type of tray or packing. The next step is to find the diameter of the tower
and pressure drop across the tower and decide the top and bottom disengagement spaces. The tray
efficiency is then determined along with either the actual number of trays or height equivalent to
53
theoretical plate (HEPT) or height of packing in the case of a packed tower, where the liquid
distributor, redistributor and packing support would also need to be designed.
The material of construction of all the parts must be selected and an insulation system designed. The
condenser and reboiler is also designed and control methodology for the distillation column, line sizes
and locations of all nozzles are decided. Finally, a mechanical design of the distillation column,
including the design of the skirt support for the distillation column must be done.
5.4. Separation Column Design
5.4.1. Flash Column
5.4.1.1 Flash Column Separation

Table 5.3 below shows that the solubilities of the hydrocarbons are low in water, making Henry’s law
applicable, as discussed in the literature. The calculations of the solubilities are attached in Appendix
E.
Table 5.3: Solubility of hydrocarbon in water
Species Solubility (%)

Ethylbenzene 0.02
Styrene 0.03
Toluene 0.06
Table 5.4 below shows the mole fractions of ethylbenzene, styrene and toluene in liquid and vapour
phase within flash drum (V-501). These were calculated using the solubilities of each hydrocarbon
and the equation using Henry’s law constants at operating pressure, as stated in the literature. The
calculations of the fractions in Table 5.4 are attached in Appendix E.
Table 5.4: Liquid and vapour molar fractions of hydrocarbons
Species Ethylbenzene Styrene Toluene

Vapour fraction, (V) 1.73E-03 1.31E-04 1.15E-02
Liquid fraction in water (L11) 0.998 0.999 0.988
Liquid fraction (L10) 1.93E-04 3E-04 6.47E-04
Total 1 1 1
The flow rate of each component was calculated to be the values shown in Figure 5.1 below. The
hydrocarbons in vapour phase leave with the hydrogen and methane in stream 9 while stream 10
flows to the first distillation column (T-501) and stream 11 flows to a water treatment plant. Since
54
there are 3 three components in stream 10 which are to be separated in pure form, at least 2 distillation
columns are required for the separation (Thakore & Bhatt, 2007:381).
Figure 5.1: Component flow rates for flash drum streams
5.4.1.2. Flash Column Sizing

To size the flash drum, procedure and calculations from Wankat (2012:95) was used. Using the liquid
and vapour compositions and flow rates above, the flash drum was sized using an empirical
procedure. A vertical drum was selected as the flow rates are relatively low. Additional
disengagement area is therefore not needed, which according to Wankat (2012:97), is formed by
elongating the vessel.
The internal diameter of the drum was calculated to be 0.3 m when rounding up. This additional
length is added for safety reasons (Wankat, 2012:99). The total height was found to be 1.14 m and the
weight of the vessel, with a wall thickness of 4 mm, was found to be 408 N (42 kg). All the
calculations for theses dimensions are shown in detail in Appendix E.
5.4.2. Distillation Column (T-502)

When the feed is binary, it is possible to obtain a virtually pure top and bottom product. Is the feed
contains more than two components, only one virtually pure product can be obtained from the top or
bottom of the column and more than one column is necessary for separating the multicomponent feed
into its components (Coulson & Richardson, 2005:494).
55
According to the design brief, the distillation column that purified styrene is to be designed in detail.
This column must be designed to produce two product streams, ethylbenzene (tops) and styrene
(bottoms), from a single feed.
5.4.2.1. Key Component Selection

The feed mixture to T-502 contains mainly ethylbenzene and styrene and is therefore a binary
mixture. According to Thakore & Bhatt (2007:381), if the amounts of the other components in the
mixture are very low, like in the case of toluene in the mixture considered in this case, the mixture is
considered binary. The two key components will therefore be ethylbenzene as the light key because it
relatively more volatile and the heavy key component will be styrene as it is less volatile compared to
the ethylbenzene. Since the feed to T-502 is considered a binary mixture, there are no non-key
components to consider.
It is desired to keep the light key component out of the bottoms product and all components lighter
than the light key should not be present or be present in a negligible amount in the bottoms. This will
help to increase the purity of the bottoms product styrene, which is required to be 99% pure. In the
same way, the heavy key component is to be kept out of the top product.
5.4.2.2. Operating Pressure

According to Thakore & Bhatt (2007:382), the temperature of the cheapest cooling medium (for
example cooling water) decides the distillation column’s operating pressure. This is a common
practice as the cooling medium must be able to condense the distillate vapour.
Because the boiling point of ethylbenzene is 409.35 K under atmospheric pressure, the operating
pressure of the distillation column is decreased to decrease the bubble point. This allows the top
product (distillate) to be condensed by cooling water at 303.15 K used in the top reflux system. In
order for the bubble point of the distillate be closer to the cooling water temperature than 409.35 K,
but still under conditions in which it will vaporise, the column operating pressure is chosen to be 0.2
bar. According to the vapour pressure trend as a function of temperature (Figure 5.2), this will reduce
the bubble point of ethylbenzene to 358 K inside the column, which will facilitate condensation of
ethylbenzene in the condenser with cooling water at 303.15 K. Under these conditions, the bubble
point of styrene is still slightly higher than that of ethylbenzene, which will allow this less volatile
compound to remain in liquid phase.
As mentioned above, styrene tends to polymerise at high temperatures. According to Hui & Hamielec
(1972:749), industrial thermal processes for polystyrene usually operate between temperatures of
373.15 K to 473.15 K, while uninhibited styrene may even start to polymerise slowly at room
temperature. For heat sensitive material, it is advantageous to operate the distillation column under
vacuum, as it prevents thermal decomposition and reduces undesirable side reactions. In fact, to quote
56
Thakore & Bhatt (2007:382), “Distillation of ethyl benzene-styrene mixture under vacuum reduces
polymerization”.
Operating at vacuum also alters the vapour-liquid equilibrium data and facilitates the use of a cheaper
heating medium in the reboiler, such as saturated steam. Under vacuum conditions, a lower reflux
ratio or fewer stages, or even both, are required in systems where decreasing operating pressure,
increases relative volatility. Furthermore it provides safety when distilling hazardous or toxic material
and increases product recovery when dealing with maximum temperature constraints.
Also according to Thakore & Bhatt (2007:383), vacuum distillation presents the following
disadvantages mentioned below. It requires a larger diameter vapour line and if the condensation
temperature is reduced too much, it may result in additional costs caused by the cooling source
required for condensation. The possibility of contamination of the products, by air or other
components of the atmosphere, is increased and the efficiency of the main condenser is decreased by
the presence of air as a non-condensable.
5.4.2.3. Determining Vapour-Liquid Equilibrium data

Although it would be ideal to experimentally determine the vapour-liquid equilibrium (VLE) data for
this specific mixture under vacuum conditions, it is not possible in this case due to the theoretical
nature of the report. Literature was therefore used to obtain the data. The data in Perry (1998: 2-68)
was obtained experimentally and is tabulated in Table 5.5 below, for the two components in the
mixture. The data was graphed as shown in Figure 5.2 and the applicable values read from the graphs.
Table 5.5: Vapour pressures of Organic compounds up to 1 atm (Perry, 1998:2-68)
Compounds Ethylbenzene Styrene

Pressure Temperature
mmHg Bar degC K degC Keshrie
1 0.001 -10 263 -7 266
5 0.007 14 287 18 291
10 0.013 26 299 31 304
20 0.027 39 312 45 318
40 0.053 53 326 60 333
60 0.080 62 335 70 343
100 0.133 74 347 82 355
200 0.267 93 366 101 374
400 0.533 114 387 123 396
760 1.013 136 409 145 418
57
450
400
Temperature (K)
350
Ethylbenzene
300
Styrene
250
200
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Vapour pressure (bar)
Figure 5.2: Vapour pressures of organic compounds up to 1 atm (1.013bar)
5.4.2.4. Degree of Separation

Since we are dealing with a binary feed mixture, only this case (and not a multicomponent mixture)
will be considered for the design of this distillation column. For a binary mixture, x D (mole fraction of
more volatile component in distillate) and xB (mole fraction of more volatile component in bottoms) is
specified while considering the equilibrium limitations. The molar flow rate of the distillate (D) must
be found as well as that of the bottoms (B), using the material balance equations below. from Thakore
& Bhatt (2007: 385). (𝑧𝐹 is used as the feed can be saturated liquid, saturated vapour or a ,mixture of
the two.) Table 5.6 below shows the values of the three streams concerning T-501 needed to complete
the material balance equations, where ethylbenzene is the more volatile component in the mixture.
The calculations of these values are provided in Appendix E.
𝐹 =𝐷+𝐵 and 𝐹𝑧𝐹 = 𝐷𝑥𝐷 + 𝐵𝑥𝐵
Table 5.6: Material balance for T-501
Variable Value (kmol/hr)

F 388
D 26.7
B 370
xD 0.672
xB 0.092
zF 0.132
58
According to Coulson & Richardson (2005:494), only a sufficient amount of liquid is to be condensed
if the top product is required as a vapour. The condensed liquid should be enough to provide sufficient
reflux and this will be achieved using a partial condenser. This reflux will then be in equilibrium with
the vapour leaving the condenser. When the vapour is completely condensed to liquid, the top product
and the liquid returned to the column will have the same composition.
5.4.2.5. Minimum Reflux Ratio

The minimum reflux ratio (Rm) is the minimum reflux ratio where infinitely many trays are required
for the desired separation. As the reflux ratio is decreased, a reflux ratio where the desired separation
becomes possible is obtained. This is called the minimum reflux ratio and separations can be achieved
at any reflux ratio above Rm (Thakore & Bhatt, 2007:389).
For the binary case with ideal vapour-liquid equilibrium curve, and saturated liquid feed (therefor q =
1), the minimum reflux ratio can be calculated as follows:
1 𝑥 ∝(1−𝑥𝐷 )
𝑅𝑚 = ∝−1 [ 𝑧𝐷 − 1−𝑧𝐹
]
𝐹
∝ is the average relative velocity which is calculated by the dividing the equilibrium ratio (K-value)
of the more volatile component, by the K-value of the other component. Therefore, in this case the
relative volatility will be calculated as follows:
𝐾𝐸𝐵
∝= 𝐾𝑆
The K-value represents the tendency of a component to vaporise. According to Kister (1992: 3), the
K-value will be determined by the formula below:
𝑀𝑜𝑙𝑒 𝑜𝑓 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑖 𝑖𝑛 𝑣𝑎𝑝𝑜𝑢𝑟 𝑝ℎ𝑎𝑠𝑒

𝐾𝑖 =
𝑀𝑜𝑙𝑒 𝑜𝑓 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑖 𝑖𝑛 𝑙𝑖𝑞𝑢𝑖𝑑 𝑝ℎ𝑎𝑠𝑒
Table 5.7 provides the experimental vapour-liquid equilibrium data for ethylbenzene and styrene at
the design temperature and pressure of 358 K and 0.2 bar (20kPa) respectively, as discussed above. 𝑥1
from the table represents the mole fraction of component i in vapour phase while 𝑦1 represents the
mole of fraction of component i in liquid phase as a function of temperature and pressure. These
values were therefore used to calculate the K-values.
59
Table 5.7: Experimental vapour-liquid equilibrium data for ethylbenzene and styrene (Jongmans et al.,
2011:3515)
The table below gives the final K-values of the two components, the relative volatility and R m. The
calculations of these values are included in Appendix E.
Table 5.8: Calculated values of K-values, relative volatility and minimum reflux ratio
Variables Values
KEB 1.01
KS 0.81
∝ 1.25
Rm 18.4
5.4.2.6. Optimum Reflux Ratio

According to Thakore & Bhatt (2007:397), the optimum reflux ratio is the value of the reflux ratio
that results in the minimum total cost of distillation (annual operating and fixed costs). This is
dependent on the feed mixture, operating conditions and specified separation. If the desired purity of
the products changes then the value of the optimum and minimum reflux ratios changes as well.
Figure 5.3 below illustrates the optimum reflux ratio corresponding to the minimum total costs.
At Rm, the column would require an infinite number of trays or packing height. This means that the
fixed cost would be infinite while the operating cost would be at its minimum. The required tray
number or packing height will decreases as the reflux ratio increases, but this will also result in an
increases in the column diameter, condenser size, reboiler size and reflux pump size. Despite these
60
increases, the fixed cost will initially decrease, as shown in Figure 5.3, because the increase in reflux
ratio results in savings in tower height, which over compensates for other fixed costs. The fixed cost
however begins to increase as the reflux ratio continues to increase.
At a certain reflux ratio, the packing height or tray requirement approaches an almost constant value
which corresponds to the minimum height or trays requirement, while the diameter and heat
exchanger sizes increase. The fixed cost therefore passes through the minimum point, at a certain
reflux ratio value, and an increase in reflux ratio causes the fixed cost to further increase again. Unlike
with fixed cost, an increase in reflux ratio always increases the operating cost. Therefore, since the
annual cost in the sum of the operating and fixed costs, the optimum reflux ratio can be obtained when
the total annual cost passes through its minimum value as shown below.
Figure 5.3: Determination of optimum reflux ratio (Thakore & Bhatt, 2007:397)
5.4.2.6.1. Determination of Theoretical Stages

The theoretical stages are also known as the equilibrium stages which are required to achieve the
desired separation. For binary distillation, the Fenske equation can be used as a short cut method,
while the more detailed method is the McCabe-Thiele method. While the Fenske method can only be
used for binary distillation at total reflux, the McCabe-Thiele method also has limitations as it cannot
be used where heat losses are very significant. In cases where high purity of products is required, the
McCabe-Thiele method must either be used in addition to other supporting methods or more than one
graph must be drawn at different scales (Thakore & Bhatt, 2007:398). The McCabe Thiele method
will therefore be used for this system.
Making reference to the outline of the McCabe method from Thakore & Bhatt (2007:398), and data
gathered from the various sources previously mentioned in this report, the following stepwise
procedure was followed. Using the VLE (vapour-liquid equilibrium) data in Table 5.7, the VLE curve
in Figure 5.4 below was drawn at the operating 0.2 bar pressure of the column. The points of x D, zF
and xB were then marked on the x-axis and connected vertically to the diagonal 45o line. The q-line
61
was then drawn. The starting point of the line is x = y = zF, and the equation of the q-line is given
below (where q is the heat necessary for converting the feed from liquid to saturated vapour divided
by molar latent heat of vaporisation). The calculation of q, all the other variables tabulated in Table
5.9 and the calculation of the additional point on the q-line, using the q-formula, is included in
Appendix E.
𝑞 𝑧𝐹 𝜆𝐹 + 𝐶𝑝𝐹 (𝑡𝐵𝑃 − 𝑡𝐹 )
𝑦= 𝑥− 𝑤ℎ𝑒𝑟𝑒 𝑞=
𝑞−1 𝑞−1 𝜆𝐹
Table 5.9: Variables used to calculate q
Variable Value
𝜆𝐹 37179 kJ/kmol
𝐶𝑝𝐹 168 kJ/kmol
𝑡𝐵𝑃 365
𝑡𝐹 323.15
q 1.19
The rectifying operating line is then drawn on the graph. In order to get an infinite number of stages,
the operating line for both the feed and the rectifying section, have to intersect. Therefore, drawing a
line through the highest point of the feed equilibrium intersection will give a pinch point associated
with the infinite number of stages needed. The intercept of the operating line at minimum reflux is
then equal to the expression below, according to Thakore & Bhatt (2007:400).
𝑥𝐷
𝑅𝑚 +1
By drawing a straight line from the point where the x D line meets the 45o line, to where the q-line
meets the VLE line, the slope can be found and the intercept of the lie on the y-axis can be calculated
∆𝑦
to equate to the equation above. To find the slope, the following equation can also be used: 𝑚 = ∆𝑥 .
In order to get the value of the point where the q-line intercepts the VLE line, the equations of the two
lines were used where an input x value which yielded y-values with the least margin of difference was
used. The points used are tabulated below.
Table 5.10: Determining equation of rectifying line

x-value y-value
xD intercepts x=y 0.672 0.672
q-line intercepts VLE line 0.1407 0.1873

0.672−0.1873
𝑠𝑙𝑜𝑝𝑒 = 0.672−0.1497 = 0.91225
62
Straight line equation: 𝑦 = 𝑚𝑥 + 𝑐
Intercept: 𝑐 = 𝑦 − 𝑚𝑥 = 0.672 − (0.912 × 0.672) = 0.059
𝑥𝐷
∴ 0.05898 = 𝑅 ∴ 𝑅𝑚 = 10.39
𝑚 +1
Let R/Rm =1.8
∴ 𝑅 = 18.7
Stripping operating line must coincide with the intersection of the q-line and rectifying operating line
and the intersection of the x B line and the x-y line. The equation of the rectifying line was found to be
y = 0.034x + 0.949, using the same method as above. This equation was equated to the q-line equation
and solved to find a second set of coordinated for the stripping line for which to plot it on the graph.
Figure 5.4: McCabe-Thiele diagram for ethylbenzene-styrene system
From the graph above, it can be deduced that the feed point of T-502 is at tray 11. The number of
trays in the enriching section is 10 and that in the stripping section is 4, giving a total number of 14
theoretical trays.
According to Thakore & Bhatt (2007:401), the Robinson and Gilliland equations can be used tofind
the number of equilibrium stages needed to increase the purity of styrene (bottom product) from 9.2%
to 99%.
63
The Robinson-Gilliland correlation for the stripping section is the equation below. NS is the number of
ideal stages required from the reference point of x’ r to xB (which are the mole fractions of the more
volatile component at the reference point and in the bottoms product respectively). K' is the
equilibrium constant for the more volatile component while S’ is the slope of the stripping section
operating line.
𝐾′ 𝑥′
( ′ −1)( ′ 𝑟 −1)
𝑆 𝑥 𝐵
log[ 1 ]
(𝐾′ −1)
𝑆′
𝑁𝑆 = 𝐾′
+1
log ( )
𝑆′
The Robinson-Gilliland correlation for the enriching section is the equation below. Here N r is the
ideal number of stages required from the reference point of x r to x’D (which are the mole fractions of
the least volatile component at reference point and in the distillate respectively). S is the slope of the
enriching section operating line while K is the equilibrium constant of the least volatile component.
𝑥
(1−𝑆)+ ′𝑟 (𝑆−𝐾)
𝑥 𝐷
log[ ]
1−𝐾
𝑁𝑟 = 𝑆 −1
log( )
𝐾
The calculations using the two equations given are provided in Appendix E below. Using the above
equation for number of stages in stripping section, the number of theoretical stages needed to purify
the styrene as mentioned above was found to be 28.
5.4.2.7. Type of Tower

The use of trays versus packing in a distillation column is briefly discussed in the literature in section
5.3.2.1. When selecting the use of trays, various factors are to be taken into consideration for choosing
the type of tray to be used. This criteria includes cost, capacity, operating range, efficiency and
pressure drop. Three of the most commonly used trays in industry are sieve trays, valve trays and
bubble cap trays.
Considering mild steal as the material of construction, the cost ratio of bubble cap : valve : sieve trays
can be represented as 3:1.5:1, making bubble cap the most expensive and sieve the least expensive (a
third of the price of the bubble cap tray). This ratio will be somewhat different when another material
of construction is considered (Thakore & Bhatt, 2007:441).
At different tower capacities, the difference in tower diameter may not be large however considering a
constant flow rate (capacity), the column diameter will vary according to the type of tray used. The
order in which the tray will increase the column diameter is modified sieve tray (being the minimum),
sieve tray, valve tray and bubble cap tray (being the maximum) (Thakore & Bhatt, 2007:441).
64
Operating range is the range of liquid and vapour flow rates over which the tray works satisfactorily.
Sieve trays provide the least operating range, while bubble cap trays provide the most. The difference
in efficiency of the three trays mentioned above is very little, when operating with minimum turn
around or at the design flow rates. Turn around refers to the ratio of maximum to minimum flow rate.
In general however, bubble cap trays provide good efficiency for a wide range of turn around. Modern
chemical plants however have control systems which are precise and automated, which allows flow
rates to show minimal fluctuation.
Although packed towers are the better option for columns operating at vacuum, according to Thakore
& Bhatt, (2007:442), for a column designed to operate at a low vacuum (which is up to 700 torr or
0.93 bar), like 0.2 bar, a “tray tower can be selected to get other advantages”. One such advantage
includes improved vapour-liquid contact in large diameter towers. In order of providing the lowest to
highest pressure drop, the trays follow the order of sieve trays, valve trays then bubble cap trays.
Although bubble cap trays seem to be the best in terms of operation, according to Behr (1994:6), the
use of these trays results in a large pressure drop over the trays which means that two columns in
series will be necessary to maintain low reboiler temperatures. The most modern plants use packing in
place of trays which allows this separation to be achieved in one column and results in reduced
pressure drop, giving a lower bottom temperature and shorter residence time. Packing, is however also
preferred for a tower with a diameter less than 0.15 m, as the fabrication of the trays for this diameter
is difficult, and therefore more costly.
For the progression of this section of the report, calculations with regards to distillation columns with
sieve trays will be considered. The reason for the choice of type of column, being tray, and type of
tray, being sieve, will be elaborated upon in the discussion. The mechanisms to be implemented to
compensate for any short comings regarding this option, is also discussed in the discussion section of
the report.
5.4.2.8. Column Diameter and Pressure Drop

For a column with sieve trays, the column diameter is obtained based on the flooding velocity (VF).
The flooding velocity is the velocity at which flooding occurs in the column. The upper limit of the
vapour velocity through the column is fixed by flooding in the sieve tray column. The design velocity
through distillation columns should ideally be between 80% - 85% of the flooding velocity; however
it is accepted between 70% – 90% (Thakore & Bhatt, 2007:443).
Flooding in tray distillation columns can occur in either jet flooding or downcomer flooding. In the
column, froth of the liquid-vapour mixture forms on each tray from which vapour is separated and
rises upward to the liquid on the next above tray. When this froth mixture touches the tray above it, it
is called jet flooding. The froth is formed collectively by free flowing jets in which liquid droplets are
65
entrained. The free flowing jets are formed by vapour flowing through tray perforations after leaving
the orifice. Gravity pulls the downward flowing liquid in the column down from one tray to next tray
below it, although flows in the opposite direction of a pressure drop in the column. The liquid elevates
to a certain level inside the downcomer to compensate that for the fact that it is flowing against the
pressure drop. Downcomer flooding is when the liquid level in the downcomer, on any tray, rises
above the weir. Figure 5.5 below illustrates the tray, froth, downcomer, elevated liquid in the
downcomer and weir explained above.
Figure 5.5: Column internals (Prieve, 2018)
As stated by Thakore & Bhatt (2007:443), the equation below can be used to calculate the flooding
velocity through a sieve tray tower.
𝜎 0.2 𝜌 −𝜌 0.5
𝐿 𝑣
𝑉𝐹 = 𝐶𝑓 (0.02) (
𝜌𝑣
)
Cf represents the flooding constant, 𝜎 is the surface tension of the liquid and 𝜌𝑣 and 𝜌𝐿 are the
densities of the vapour and liquid respectively. 𝐶𝑓 is obtained from Figure 5.6 below, which also
requires the liquid-vapour flow factor, 𝐹𝐿𝑉 , obtained by the following equation where LW and VW is
the mass flow rate of liquid and vapour respectively.
𝐿 𝜌 0.5
𝐹𝐿𝑉 = 𝑉𝑊 (𝜌𝑣 )
𝑊 𝐿
66
Figure 5.6: Flooding velocity for sieve plates (Thakore & Bhatt, 2007:444)
Once 𝐹𝐿𝑉 is calculated, as shown in Appendix E, the tray spacing is assumed in order to use Figure
5.6 to obtain Cf. According to Coulson & Richardson (2005:556), plate spacing between 0.3 m to 0.6
m are normally used for columns with a diameter up to and exceeding 1 m. For the initial trail
calculation a tray spacing of 0.3 was assumed to find Cf using Figure5. (The diameter was however
found to be greater than 1m under this assumption. The plate spacing for the second trial was then
chosen to be 0.5 m because according to Thakore & Bhatt (2007:461), the 0.3 spacing is only
sufficient up until a column that is 1 m in diameter.)
Once Cf is obtained, it was plugged in to calculate the flooding velocity. The actual velocity through
the column could then be calculated, as shown in Appendix E, and used with the calculated
volumetric flow rate to calculate the net area (An) required at that section of the column (top or
bottom), using the equations below.
𝑉𝑀
Volumetric flow rate: 𝑄̇ = 𝜌 𝑎𝑣
𝑣
𝑄̇
Net area required: 𝐴𝑛 = 𝑣
To determine the column diameter, the cross sectional area (Ac) of the column must be found, as
shown in the following equation:
4×𝐴𝐶
𝐷𝑖 = √ 𝜋
Ac can be obtained by using the following simultaneous equations:
𝐴𝑑 = 𝑑𝐴𝑐 and 𝐴𝑛 = 𝐴𝑐 − 𝐴𝑑 ,
67
where 𝐴𝑑 is the downcomer area and d is assumed (0.08 – 0.12) because Ad ranges between 8% - 12%
of Ac (Thakore & Bhatt, 2007:444).
When calculating the diameter of the top and bottom of a distillation column, the overall method
described above will be the same, but the manner of determining the properties used in the equations
for the top and bottom will differ. For example, obtaining the top operating temperature and bottom
base temperature is different, as shown in Appendix E calculations.
The operating pressure at the top of the column was designed to be 0.2 bar. The operating pressure at
the column’s base was found by adding the pressure drop on the column to the 0.2 bar operating
pressure. The equation below, was obtained from Thakore & Bhatt (2007:460), and used to calculate
the pressure drop in the column:
∆𝑃𝑇 = 𝑁𝑎 × 𝑔 × 𝜌 × ℎ𝑡
Here Na represents the actual number of trays in the column, by dividing the theoretical number by the
assumed tray efficiency. Tray efficiency was obtained from literature, as mentioned above, for an
ethylbenzene-styrene system at 75%, and used as the first assumption. 𝑔 and 𝜌 is the gravitational
acceleration and liquid density of the bottoms product respectively. H t is the column height, which is
typically 20 times the column diameter and less, according to the project brief. Although the Ht:Di =
20:1 ratio can be increased to 30:1 if the column diameter is less than approximately 1 m, this
increases the costs. The extra costs are due to the need for adding extra supports to and/or around the
column.
The column diameter and pressures are summarised in Table 5.11.
Table 5.11: Column diameter and Pressure data
Column diameter at top 7.75 m

Column diameter at bottom 6.58 m
Pressure at top of column 0.2 bar
Pressure at bottom of column 0.246 bar
Pressure drop 0.046 bar
5.4.2.9. Weir Height

The weir was briefly mentioned and illustrated in Figure 5.5. The weir is important as it determines
the liquid height over the tray. According to Thakore & Bhatt (2007: 446), designing a high weir will
make the plate more efficient, but decrease the pressure drop. For columns operating under vacuum
conditions, the weir height is reduced and maintained between 6mm – 12mm, to maintain a low
68
pressure drop across the tray. Weir length is kept between 0.6 – 0.8 times the diameter of the column
and increasing the weir length also increases the column diameter. A decrease in the weir length
increases the net area of the tray, decreases the downcomer’s area and volume, increases its pressure
drop as well as elevates the downcomer’s liquid level. It is therefore important to balance the weir’s
height and length. Because this column has a design operating pressure of 0.2 bar, which is low
vacuum, a weir height of 10 mm has been selected.
5.5. Discussion
With regards to the flash drum, or three phase separator, the operating pressure and temperature was
determined to be 1 bar and 323.15 K. These were chosen in order for the lighter products to
effectively vaporise out of the relatively heavier hydrocarbons and for the water to remain in liquid
phase and also be separated at the bottom of the drum. This lower temperature and pressure also helps
to minimise the polymerisation of styrene in the vessel, which could cause complications in the
process, damage equipment and reduce the product yield. Using these conditions, good theoretical
separations of the different components were found through calculation, as shown in Figure 5.1.
Adding to the flash drum, a demister would enhance the vapor–liquid separator aspect by enhancing
the removal of liquid droplets entrained in the vapour stream. According to Coulson & Richardson
(2007:458), demisters can be used to improve the performance of separating vessels, like flash drums.
It can be implemented where the droplets are expected to be as small as 1𝜇𝑚, and where high
separating efficiencies are required for high quality vapour with regard to entrained liquids. Demisters
reduce the residence time required to separate a given liquid droplet size by reducing the volume and
associated cost of separator equipment.
Distillation was the chosen as the method of separation because the components are relatively volatile
and are required to be separated in pure form (styrene purity of at least 99%, according to the project
brief). The number of distillation columns required were determine to be two as there are three
components to be separated in high purity. The first distillation column (T-501) is to be located after
the flash unit (V-501) as it will separate the toluene (with 99% purity) from the styrene and
ethylbenzene. The second distillation column (T-502), will follow the first one as it will purify the
styrene by separating the ethylbenzene from it.
Sieve trays have been chosen as the trays to be used in T-501 because it is more economical and
would alter the column diameter the least if the plant capacity were to be increased. This therefore
provides some room if changes of that sort are to be made without having too much of a drastic
impact elsewhere. The fact that sieve trays provide the least operating range and only show almost
the same efficiency as the other tray types when operating at the design flow rates, has been taken into
account. The effect of the latter can therefore minimised as modern chemical plants have control
69
systems which are precise and automated, which results in minimal fluctuation in flow rates. This
control instrumentation will also assist in monitoring the process. According to Coulson &
Richardson (2005:548), the “Murphee plate efficiency” of a sieve plate used for an ethylbenzene-
styrene system is 75%.
The feed stream to the styrene distillation column contained ethylbenzene (13%) and styrene (87%)
with trace amounts of toluene. The mixture was therefore considered to be binary as, according to
Thakore & Bhatt (2007:381), if the amounts of the other components in the mixture are very low, the
mixture may be considered to be binary. The operating pressure and temperature was chosen to be 0.2
bar and 358 K. 0.2 bar in order to reduce the bubble point of the distillate (ethylbenzene) to 360 K.
This lower pressure will also reduce condensation costs of ethylbenzene in the condenser which uses
cooling water at 303.15 K.
In terms of the detailed distillation column design, the method was mostly adopted from Thakore &
Bhatt (2007), with other sources referenced along the way. From the calculations discussed above and
shown in Appendix E, the reflux ratio was found to be 18, the number of trays needed was calculated
to be 28 trays, the decided tray spacing was 0.5 m and the weir height was 10 mm. An inlet weir was
not included in the design, because it is stated by Thakore & Bhatt (2007:446) that it causes a
hydraulic jump. The top and bottom pressure specifications were 0.2 bar and 0.25 bar respectively
while the column diameter was calculated to be 7.7 m at the top of the column and 6.6 m at the base
of the column These finding are reasonable with comparison to other similar distillation column
designs.
5.6. Conclusion
It can be concluded that distillation is the preferred method of separation of ethylbenzene and styrene
because the components are relatively volatile and are required to be separated in pure form (styrene
purity of at least 99%, according to the project brief).
The selected operating pressure and temperature is 0.2 bar and 358 K. 0.2 bar and was chosen to
reduce the bubble point of the distillate (ethylbenzene) to 360 K and reduce condensation costs of
ethylbenzene in the condenser with cooling water at 303.15 K. A sieve tray tower was selected as the
best option. As mentioned above, the column diameter was found to be greater than 1m and operated
at low vacuum, where the use of trays improve vapour-liquid contact in columns with large diameters.
Sieve trays are also the most economical choice and, the efficiency of a sieve plate used for an
ethylbenzene-styrene system was stated to be 75%. In terms of the detailed tower design, the reflux
ratio was found to be 18, the number of trays needed was calculated to be 28 trays, the decided tray
spacing was 0.5 m and the weir height was 10 mm. The top and bottom pressure specifications were
70
0.2 bar and 0.25 bar respectively while the column diameter was calculated to be 7.7 m at the top of
the column and 6.6 m at the base of the column
5.7. Recommendations
It is recommended to add a demister pad to V-501. This will decrease the amount of water vapour
present in stream 9 containing hydrogen and methane and improve the quality of the stream. It is
desired to have as little water vapour mixed with these gases as they can be used as a source of natural
gas for combustion in the plant’s fired heater. Introducing water to the combustion reaction will
hinder the reaction. Also, water has the ability to corrode lines and equipment, especially if it was not
designed for carrying or holding water. The demister therefore has the ability to extend equipment
life, increase the amount of valuable products recovered and reduce the residence time needed for
separation.
It is recommended by CCLynch & Associates (2012), to waterproof the connection of the weir to the
channel. The joint between the channel and weir plate should hence be packed with “asphalt type
roofing compound” or chemically inert cement. It is also required for the crest of the weir to be kept
clean. “Fibres, stringy materials and larger particles” tend to adhere to the crest and should therefore
be periodically removed by purging accumulated silt and solids on the upstream side of the weir.
5.8. References
 Behr, A. 1994. Styrene Production from Ethylbenzene. Germany: Universitat Dortmund.
 Blackwell, W. W. 1984. Chemical Process Design on a Programmable Calculator. Chapter 3. New
York: McGraw Hill.
 CCLynch & Associates. 2012. Criteria for Proper Weir Design. [Online]. Available:
http://www.cclynch.com/criteria-for-proper-weir-design/ [2019, July 18].
 ChemicalBook. 2017. Styrene. ChemicalBook. [Online]. Available: https://www.chemical
book.com/ChemicalProductProperty_EN_CB3415111.htm. [2019. July, 2]
 Geankoplis, C. J. 1993. Transport Process and Unit Operations. Third Edition. Minnesota: Prentice-
Hall International, Inc.
 Hui, A. W., & Hamielec, A. E. 1972. Thermal polymerization of styrene at high conversions and
temperatures. An experimental study. Journal of Applied Polymer Science. 16.3: 749–769.
 Jongmans, M. T. G., Maassen, J. I. W., Luijks, A. J., Schuur, B., de Haan, A. 2011. Isobaric Low-
Pressure Vapor-Liquid Equilibrium Data for Ethylbenzene + Styrene + Sulfolane and the Three
Constituent Binary Systems. Journal of Chemical & Engineering Data. 56: 3510-3517.
 Kister, H. Z. 1992. Distillation Design. California: McGraw-Hill, Inc.
71
 Lieberman, N. P. & Lieberman, E. T. 2014. A Working Guide to Process Equipment. Fourth Edition.
New York: McGraw-Hill Education.
 Luyben, W. 2011. Design and Control of the Styrene Process. Industrial & Engineering Chemistry
Research. 50.3: 1231-1246.
 Prieve, D. 2018. Unit Operations of Chemical Engineering. ResearchGate. [Online]. Available:
https://www.researchgate.net/figure/Side-view-of-a-sieve-plate-tray-Taken-from-8-of-
Treybal_fig6_322364940. [2019, July 17]
 Sander, R. 2015. Compilation of Henry's law constants for water as solvent. Atmos. Chem. Phys.
4.15: 4399-4981.
 Schaller, C. 2012. Structure & Reactivity in Organic, Biological and Inorganic Chemistry. Structure-
Property Relationships. Mixtures that don't mix. [Online]. Available: https://employees.csbsju.edu
/cschaller/Principles%20Chem/imf/SPmixtures.htm. [2019, June 27].
 Seader, J., Henley, E., Roper, D. 2010. Separation Process Principle. 3rd ed. New York: wiley.
 Thakore, S. B. & Bhatt, B. I. 2007. Introduction to Process Engineering and Design. New Delhi:
Tata McGraw-Hill.
 Wankat, P. C. 2012. Separation Process Engineering. Includes Mass Transter Analysis. Third
Edition. New York: Prentice Hall.
 Yaws, C. 1999. Chemiccal Properties Handboek. New York: Mc Graw-Hill.
72
CHAPTER 6 PUMP SIZING
Summary
In this chapter, the sizing of pump P-501 is described and theoretically demonstrated. The
dehydrogenation of ethylbenzene occurs for the production of styrene in plant 5. Pump P-501 is a feed
pump to plant 5, which pumps ethylbenzene from a storage tank (Tk-001) to a mixer (M-501). Carbon
steel was selected as the material of construction, based on the theory found in Sinnott & Towler
(2008). To overcome the differences in pressure and height, static pressure and dynamic loss was
calculated. This information contributes to the pressure head that must be supplied by the pump in a
piping system. Energy balance was used to derive an equation from which to plot the system curve.
The system curve was plotted against the pump curve to find the best point of operation. The flow rate
and head for the system were therefore found to be 0.008 m3/s and 25 m/s respectively. A single
stage centrifugal pump was selected based on the best operating pressure head of the pump. The net
positive suction head of this pump was calculated to be 13.4 m. A characteristic curve for a
centrifugal pump graph was used to estimate the pump efficiency and impeller diameter, based on the
pump’s capacity and head. The efficiency was found to be 61 % while the required impeller diameter
was found to be 130 mm. The pump power was then calculated to be 2.6 kW. The required pump
power was divided by the pump efficiency to estimate the actual pump power, of 4.16 kW.

Fresh ethylbenzene feed is to be pumped from a storage tank to a mixer, where recycled ethylbenzene
recovered from the plant is to be mixed with the fresh feed. Along this line, fittings, elevation and
pipe roughness cause head loss. A pump must therefore be sized to overcome these losses and
transport liquid ethylbenzene from the storage tank to the mixer.
6.2. Introduction
There are different means of transporting fluids to from one point to another. According to Leeuw
(1995:1), gravity and pumps are the most commonly used methods. Pumps operate by increasing the
fluid’s static pressure, by supplying head or energy, to overcome friction losses in the system and
possibly elevation from a set point.
In the system being considered, the ethylbenzene fed to the plant for producing styrene is stored in a
storage tank. The storage tank is stored a distance away from the plant as ethylbenzene is a flammable
fluid [National Centre for Biotechnology Information, 2019]. According to GAPS Guidelines
(2015:1), which provides layout and spacing recommendations for chemical facilities, spacing
guidelines are intended to limit explosion overpressure and fire exposure damage as well as minimise
property loss. A pump is therefore required to transport the ethylbenzene along the distance, from the
storage tank to the plant. This pump will have to account for the head loss due to friction in the pipe
73
caused by internal wall roughness, fittings, elevation and change in pressure. The type of pump best
suited to accommodate the needs of the system will be chosen. It is also necessary to specify the
NPSH, pump efficiency, impeller diameter, pump power and speed. All of these must be taken into
account with the equipment and operating costs to obtain the optimum pump.
6.3.1. Pump Sizing

Pumps are devices which supply energy (or head) to a liquid in order to overcome head losses caused
by friction and to raise the level of the liquid. Sizing a pump involves correlating the pump’s flow and
pressure to that required by the system. Sizing a pump requires specifying the required outlet pressure
or the pump with a given system head (Morgan, 38:2016).
6.3.2. Types of Pumps

According to Sinnott & Towler (2008:243), dynamic pumps and positive displacement pumps are the
two general pump types that pumps can be classified as. A pump is selected based on the flow rate,
head required, type of fluid, frequency of use and power supply. A centrifugal pump, shown in Figure
6.1 below, is a type of dynamic pump. It is the most common used in the chemical process industry,
which other types being used only when there are special process conditions, like a high head. A
positive displacement is often used when additives are to be added to a process at low flow rates
(Sinnott & Towler, 2008:244). Positive displacement pumps include gear pumps, piston pumps and
vane pumps, shown in Figure 6.2 below.
Figure 6.1: Centrifugal pump (CPL300B, 2017)
74
Figure 6.2: Positive displacement pumps (CPL300B, 2017)
6.3.3. Pump and System Curves

The pump curve is a characteristic curve for a pump. It illustrates the head which the pump is able to
produce as a function of the volumetric flow rate. A typical pump curve has the flow rate on the x-
axis and the head on the y-axis as illustrated in Figure 6.3 below. The point of intersection between
the pump curve and y-axis is known as the “closed valve head” of the pump (Morgan, 2016:43). The
system curve illustrated the relationship between the head required by the system as a function of flow
rate. This curve is generated by calculating the system head over a range of expected process flow
rates. The point at which the pump and system curve intersects is known as the “duty point” or best
operating point (Morgan, 2016:43).
According to Sinnott & Towler (2008:254), the components to the pressure head that must be
supplied by the pump in a piping system include:
1. The static pressure. This overcomes the differences in pressure and height.
2. The dynamic loss caused by friction in the pipe, miscellaneous losses and the pressure loss through
equipment.
75
Figure 6.3: Basic pump curve plots (Morgan, 2016:42)
6.3.4. NPSH
According to Coulson & Richardson (2005:211), the inlet pressure to the pump must be high enough
as to prevent cavitation from occurring. Cavitation occurs with the formation of gas or vapour bubbles
in the pump casing. These bubbles form when the inlet pressure is below the vapour pressure of the
liquid being pumped.
NPSH is determined to avoid cavitation, as it characterises the likelihood of cavitation. When the
pressure at the pump is below the vapour pressure of the fluid, the fluid may start to boil. As the
bubbles of the vaporised fluid travel they collapse, creating zones of higher pressure. These pressure
fluctuations in the system lead to vibration and pitting of metals. This results in a decrease in pump
efficiency and structural damage to the pump which may lead to failure. In order to determine whether
a pump will be able to provide adequate head for the fluid to be pumped in a given system, the “Net
Positive Suction Head” (NPSH) must be determined.
Net positive suction head available (NPSHavail) is the pressure at the pump inlet, above the vapour
pressure of the liquid. This is expressed as head of liquid. Net positive head required (NPSHreqd) is
specified by the pump manufacturer and is a function of the design parameters of the pump (Coulson
& Richardson, 2005:211).
6.3.5. Pump Efficiency and Impeller Diameter

According to Kreider & Rabl (1996:6), the efficiency of the pump is greatest when the largest pump
impeller diameter is chosen. Pump efficiency decreases when smaller impellers are installed in a
pump because of the increased amount of fluid that slips through the space between the edges of the
impeller blades and the pump casing. Once the best point of operation has been determined using the
pump and system curve, this capacity and head can then be used with a pump performance chart to
obtain the pump efficiency and impeller diameter. Figure 6.4 below is one such pump performance
chart.
76
Figure 6.4: Pump performance chart (Kreider & Rabl, 1996:6)
6.3.6. Pump Power

In order to transport fluid from one point to another, energy is required. This energy must overcome
pipe wall friction losses, pipe fitting friction losses, losses due to process equipment, differences in
elevation and differences in pressure between the vessels between which the pump is pumping
(Sinnott & Towler, 2008:251).
Using the calculated system head, the pump power rating can be approximated. Although the pump
manufacturer provides the pump’s power rating and motor size, these are needed early in the design
process for the power cables to be sized (Morgan, 2016:42). The power required by the pump,
depends on the pump efficiency.
6.4. Pump Sizing Results and Discussion
6.4.1. System and Diagram

The system with the pump includes an atmospheric ethylbenzene storage tank (Tk-001), four 90o
elbows, unions where the pipes joined together and to the pump itself, pipe expansion (where the line
joined to the mixer), contraction (where the line joined the storage tank), the pump (P-501) and a
mixer which mixed the fresh ethylbenzene feed and recycled ethylbenzene. The system also included
one manual globe valve on the suction line and an automated flow controller gate valve on the
discharge line. A diagram of the system is included as Figure 6.5 below.
77
Z1
Z2
05-FC-01
Tk-001
Datum
05-FV-01 M-501
P-501
Figure 6.5: Diagram of pump system
The globe vale on the suction is used to shut the line if the pump needs to be serviced and is not
throttled as to not starve the pump, which could lead to cavitation. The gate valve on the discharge
line is automatically throttled depending on the amount of feed needed from the storage tank in
additions to the ethylbenzene feed flowing via the recycle line.
6.4.2. Unit Spacing

From the spacing recommendations set forth in GAPS Guidelines (2015:3), the inter-unit spacing
from an atmospheric storage tank to a process unit of moderate hazard, was found to be 250 feet (76.2
m). This was therefore taken to be the length of the pipe between the ethylbenzene storage tank and
process unit (mixer: M-501) to which the pump is required to pump the fluid. Figure 6.6 below was
the diagram used to obtain the distance mentioned above.
78
Figure 6.6: Inter-unit spacing recommendations for chemical plants
According to “Engineering Toolbox” (2004), the length of a single random pipe customarily ranges
between 5.5 m to 7.6 m. The amount of unions along the 76.2 m pipe from the storage tank to the
mixer is therefore 12. This includes the union of the lines to the suction and discharge of the pump.
6.4.3. Fluid Properties

The pump will be pumping ethylbenzene at a temperature of 303.15 K and pressure of 1 bar. This is
based on ambient conditions, as the chemical is in liquid phase under these conditions (Yaws,
1999:416) and can therefore be pumped. The density and viscosity of ethylbenzene at the temperature
stated above was calculated using formulas from Yaws (1999). The formulas and coefficients for the
formulas are given below. Other fluid properties which will be used in the pump sizing calculations is
the vapour density.
Table 6.1: Coefficients for density of ethylbenzene as a function of temperature (Yaws, 1999:185)
A B n TC Tmin Tmax ρ at 25o C

EB 0.2889 0.2644 0.2921 617.17 178.2 617.17 0.865
𝑇 𝑛
−(1− )
𝜌 = 𝐴𝐵 𝑇𝐶
79
Table 6.2: Coefficients for viscosity of ethylbenzene as a function of temperature (Yaws, 1999:478)
A B C D Tmin (K) Tmax (K) uliq@25C uliq@Tmax

EB -5.26 831 0.0108 -1.06E-05 210 617 0.629 0.05
𝐵
log10 𝜇𝑙𝑖𝑞 = 𝐴 + 𝑇 + 𝐶𝑇 + 𝐷𝑇 2 (Units:cP)
Table 6.3: Coefficients for vapour pressure of ethylbenzene as a function of temperature (Yaws,
1999:159)
A B C D E Temp. range (oC)

EB 36.2 -3340.2 -9.80 -1.18E-11 2.58E-06 178.15 – 617.17
𝐵
log10 𝑃 = 𝐴 + 𝑇 + 𝐶 log10 𝑇 + 𝐷𝑇 + 𝐸𝑇 2 (Units: mmHg)
Table 6.4 below is a summary of the fluid properties of the ethylbenzene to be pumped.
The calculations of the properties using the formulas given above are attached in Appendix F.
Table 6.4: Fluid Properties of ethylbenzene
Properties Value
Temperature 303 Keshrie
Density 861 kg/m3
Viscosity 0.0006 kg/m.s
Boiling Point 409 Keshrie
Phase Liquid
Molar mass (Yaws, 1999:271) 106 kg/mol
Vapour Pressure 1.7 kPa
6.4.4. Material of Construction and Pipe Diameter

According to Sinnott & Towler (2008:408), carbon steel is the cheapest metals, relative to metals like
copper, aluminium alloy, titanium and nickel. It is also available in a range of standard forms and
sizes, can be easily worked and welded and has good tensile strength and ductility. Carbon steel is
thus the most commonly used engineering material (Sinnott & Towler 2008:411). Carbon steel was
therefore selected as the material of construction of the pump and pipes surrounding it.
For a pipe of carbon steel, the following formula was adopted from Coulson & Richardson
(2005:220), to calculate the optimum pipe diameter:
𝑑𝑜𝑝𝑡𝑖𝑚𝑢𝑚 = 293𝐺 0.53 𝜌−0.37
80
The calculation for pipe diameter is included in Appendix F.
6.4.5. Flow Rates

The velocity of the fluid in the pipe was then calculated. From the given plant capacity of 280 000
tonnes per year, the design inlet molar flow rate was calculated to be approximately 350 kmol/hr in
Chapter 1. This was then converted to mass flow rate (G) using the molar mass. This was then divided
by the calculated density of the fluid to obtain the volumetric flow rate. Using the pipe diameter
calculated above, the cross sectional area of the pipe was found (Ac) and the velocity of the fluid
through the pipe was found by dividing the volumetric flow rate by the cross sectional area. The
formulas of the calculations and conversions described above as well as the flow rates in Table 6.5 are
attached in Appendix F below.
Table 6.5: Flow rates
Flow Rate Value

Mass flow rate (G) 10.3 kg/s
Volumetric flow rate (Q) 0.012 m3/s
Velocity (v) 2.11 m/s
The calculated velocity corresponds with the literature found in Coulson & Richardson (2005:217),
which states that the maximum velocity must be less than the critical (or sonic) velocity, as erosion is
likely to occur above this velocity. As shown in the table below, for ethylbenzene which is a non-
viscous pumped liquid, the velocity should be from 1 m/s to 3m/s.
Table 6.6: Typical pipe velocities and allowable pressure drops used to estimate pipe sizes
6.4.5. Friction Factor

With the calculated data, the Reynolds number (Re) could be obtained using the formula below
𝜌𝑑𝑣
𝑅𝑒 = 𝜇
As shown, Re was found to be greater than 2100, which indicated that there is turbulent flow in the
pipes (Geankoplis, 1993:89). The chart labelled Figure 6.7 was therefore used to find the fanning
friction factor. With a Reynolds number of approximately 270000 and a pipe roughness (𝜀/𝑑) of
0.001, the fanning friction factor was found to be 0.003.
81
Figure 6.7: Pipe friction versus Re and relative roughness Coulson & Richardson (2005:202)
6.4.6. System Curve

The table below contains the system data and specifications obtained from the calculations and
literature mentioned above. This was used to plot the system curve to find the best operating point of
the pump.
Table 6.7: System data and specifications
Variable Value
Z1 14 m
Z2 5m
v1 0 m/s
v2 2.22 m/s
P1 101.3 kPa
P2 110 kPa
Patm 121.3 kPa
Pv 1.7 kPa
F 0.003
L 76.2 m
D 0.083 m
G 9.81 m/s2
𝜌 861 kg/m3
The energy balance is simplified to Bernoulli’s equation. This can be used to derive a formula for the
system demand curve.
𝑃1 𝑣12 𝑃2 𝑣2 2
Bernouli equation: 𝑧1 + 𝜌𝑔
+ 2𝑔
+ ∑ 𝐻𝐿 = 𝑧2 + 𝜌𝑔
+ 2𝑔
+ 𝐻𝐷
𝑄̇
v=
𝐴
82
𝜋𝐷 2
A= 4
HL is the head loss of the system due to the friction caused by fittings and the inner pipe surface. The
gate valve has been taken as a quarter open in order to overestimate the friction that the pump will
have to overcome. If this is not done, the pump could shut off if the valve’s open position decreases
from 100% open.
4𝑓𝐿𝑣 2 𝑘𝑣 2
∑ 𝐻𝐿 = 𝐻𝑠 + ∑ 𝐻𝑓 = + ∑
2𝑔 2𝑔
∑ 𝐻𝑓 = ℎ𝑒𝑥𝑖𝑡 + 4ℎ90𝑜 𝑒𝑙𝑏𝑜𝑤𝑠 + 12ℎ𝑢𝑛𝑖𝑜𝑛 + ℎ1𝑜𝑝𝑒𝑛 𝑔𝑎𝑡𝑒 + ℎ 𝑔𝑙𝑜𝑏𝑒 + ℎ 𝑒𝑛𝑡𝑟𝑦

4
Table 6.8 below shows the values of the velocity heads for all the different valves and pipe fittings
represented in the equation above.
Table 6.8: Velocity heads of pipe fittings and valves (Sinnott & Towler, 2008:249)
The velocity at the surface of the liquid level in the tank and mixer are taken as zero, due to their
relatively large diameters. Tank Tk-001 is open to the atmosphere therefore P 1 is 101325 Pa, while the
pressure in the mixer is assumed to be 110000 Pa.
𝑃1 𝑃2 𝑣1 2 𝑣2 2
∴ 𝐻𝐷 = (𝑧1 − 𝑧2 ) + ( − ) + ( − ) + ∑ 𝐻𝐿
𝜌𝑔 𝜌𝑔 2𝑔 2𝑔
83
𝑃2 − 𝑃1 𝑣2 2 − 𝑣1 2 4𝑓𝐿𝑣 2 ∑ 𝐾 𝑣 2
= (𝑧1 − 𝑧2 ) + + + +
𝜌𝑔 2𝑔 𝐷 2𝑔 2𝑔
𝑃1 − 𝑃2 𝑣12 − 𝑣22 2𝑓𝐿 ∑ 𝐾 𝑣 2

= (𝑧1 − 𝑧2 ) + + +( + )( )
𝜌𝑔 2𝑔 𝐷 2 g
2
𝑄̇
𝑃1 − 𝑃2 𝑣12 − 𝑣22 2𝑓𝐿 ∑ 𝑘𝑛𝑟 ( )
𝐴
= ∆z + + +( + )
𝜌𝑔 2𝑔 𝐷 2 𝑔
( )
∆P ∆(𝑣 2 ) 2𝑓𝐿 ∑ 𝑘𝑛𝑟 𝑄̇ 2

= ∆z + + +( 2 + )( )
𝜌𝑔 2𝑔 𝐷𝐴 2𝐴2 𝑔
∆P ∆(𝑣 2 ) 2𝑓𝐿 ∑ 𝑘𝑛𝑟

= ∆z + + + 2 + 2 (𝑄̇2 )
𝜌𝑔 2𝑔 𝜋𝐷2 𝜋𝐷2
(𝐷𝑔 ( 4 ) 2𝑔 ( 4 ) )
101325 − 11000 0−0 2(0.003)(76.2) (0.5 + 4(0.7) + 12(0.04) + 16 + 8.5 + 1)

= (14 − 5) + + +( + ) (𝑄̇ 2 )
(861.03)(9.81) 2(9.81) 0.083(9.81)(0.0054) 2 2(9.81)(0.0054) 2
= 9 − 1.027 + 0 + (19369 + 15378)(𝑄̇2 )
∴ 𝐻𝐷 = 7.97 + 70747𝑄̇2
Therefore the system (demand) curve equation, in terms of Q is: 𝐻𝐷 = 7.97 + 70747𝑄̇2
The system curve generated from this equation was plotted against the pump curve to obtain the bent
point of operation. The data for the pump curve was given in the project brief and is included in the
table below.
Table 6.9: Given data for pump curve and calculated data for system curve
Given data Calculated data

Volumetric flow rate Differential head Velocity HD
3 3
m /s m /h kPa m m/s m
0 0 225 22.9 0 8
0.003 9 224 22.8 0.5 10
0.005 18 223 22.7 0.9 17
0.008 27 222 22.6 1.4 27
0.010 36 220 22.4 1.9 43
0.013 45 218 22.2 2.3 62
0.015 54 216 22.0 2.8 86
0.018 63 212 21.6 3.2 114
0.020 72 205 20.9 3.7 147
84
0.023 81 192 19.6 4.2 184
0.025 90 168 17.1 4.6 225
0.028 99 130 13.3 5.1 270
0.030 108 72 7.3 5.6 320
6.4.7. Pump Curves

Figure 6.8 below illustrates the system curve tabulated above and the pump curve. The data for the
pump curve was provided with the project brief. It can be seen from the graph that the best point of
operation, where the two curves intersect, is at a flow rate of 0.008 m3/s, giving a head of 25 m.
Figure 6.8: System curve and Pump Curve
6.4.8. Type of Pump

According to Morgan (2016:38), it is important to specify the type of pump required in the system, to
ensure that it is capable of delivering the system’s design flow rate at the given pressure. A single
stage centrifugal pump was selected, based on Table 6.10 below. The best operating pressure head of
the pump is approximately 25 m, which best suits the 500 ft (152.4 m) maximum pressure range from
the table.
85
Table 6.10: Pump Characteristics (Silla, 2003:256)
Silla (2003:259) states that approximately 95% of pumps used in the chemical industry are centrifugal
pumps. The liquid enters axially into the eye of the impeller and rotating blades imparts velocity to
the liquid. The kinetic energy of the liquid is then converted into pressure in the casing. Centrifugal
pumps are employed where it is required to deliver a high flow rate at a medium pressure.
Centrifugal pumps operate by transferring mechanical energy to the fluid in the form of increased
velocity head. This type of pump has a wide range of applications due to the low cost of in relation to
the amount of fluid it can pump, low maintenance costs, uniform flow, small floor space and
adaptability for use with a motor or turbine (Leeuw, 1995:1).
6.4.9. NPSH
The energy balance is used to derive a formula of NPSH. Point 1 is taken at a certain level from which
the liquid is to be pumped (datum) and point 2 is taken at the point of lowest pressure in the pump.
Point 2 is therefore the inlet chamber holding the impeller, as it is the location where cavitation is the
most likely to first occur. The formula below is the formula for NPSH, as given by Welty et al.
(2007:192).
𝑃𝑎𝑡𝑚 −𝑃𝑣
𝑁𝑃𝑆𝐻 = 𝜌𝑔
+ ∆𝑦 − ∑ ℎ𝐿
According to Welty (2007:192), the value of NPSH should be greater than the value obtained from the
pump performance curve at the same flow rate. This promotes proper pump installation. The purpose
of these ideas is to establish a maximum value for the height, from the datum point to the pump.
The NPSH of this pump was calculated to be 13.4 m and the calculation is attached in Appendix F.
86
6.4.10. Pump Efficiency and Impeller Diameter
Converting the flow rate capacity of 27 m3/s, found from the best point of operation, to litres per
second gives 7.5 l/s. Using this data and the head found, the efficiency of the pump was found on the
characteristic curve for a centrifugal pump, as shown on Figure 6.9 below. The efficiency read off the
graph is therefore 61%. The impeller diameter can also be read from the figure below. The impeller
diameter is therefore chosen be 130 mm. Although the point below falls in between 120 mm and 130
mm, the larger impeller diameter is chosen in order to account for a potential increase in the plant
capacity in the future, or on start up when there might not be ethylbenzene in the recycle line and this
pump will thus have to pump more fresh feed to the plant.
Figure 6.9: Characteristic curve for a centrifugal pump (Coulson, 1999:336)
6.4.11. Power Required by the Pump

The power required by the pump can be calculated by rearranging the formula given below (KSB,
2005:3).
𝜌𝑔 ∑ 𝐻𝐷 𝑄 𝜌𝑔 ∑ 𝐻𝐷 𝑄
ƞ= 𝑊
, ∴𝑃= ƞ
The power of the pump was found to be 2.6 kW. This was then divided by the efficiency to find the
pump’s actual power of 4.26 kW. The calculations thereof are attached in Appendix F. This value is
not far off from the value that can be obtained from the Figure 6.9 above.
6.5. Conclusion
The method used and theory behind the sizing of pump P-501 was described and demonstrated in this
report. The dehydrogenation of ethylbenzene occurs for the production of styrene in plant 5. Pump P-
87
501 is a feed pump to plant 5, which pumps ethylbenzene from a storage tank to a mixer. Carbon steel
was selected as the material of construction because of its availability in a range of standard forms and
sizes; it can be easily worked and welded and has good tensile strength and ductility. To overcome the
differences in pressure and height, the static pressure of the system was calculated. The dynamic loss,
which is caused by friction in the pipe, the miscellaneous losses, and the pressure loss through
equipment, was also calculated. This information contributes to the pressure head that must be
supplied by the pump in a piping system. Energy balance was used to derive an equation from which
to plot the system curve. The system curve was plotted against the pump curve to find the best point
of operation. The flow rate and head for the system were therefore found to be 0.008 m3/s and 25 m/s
respectively. A single stage centrifugal pump was selected due to the best operating pressure head of
the pump. The net positive suction head of this pump was calculated to be 13.4 m. A characteristic
curve for a centrifugal pump graph was used to estimate the pump efficiency and impeller diameter,
based on the pump’s capacity and head. The efficiency was found to be 61 % while the required
impeller diameter was found to be 130 mm. The above data was then used with the formula in section
6.4.11, to calculate the pump power which was found to be 2.6 kW. The actual pump power was
therefore estimated to be 4.16 kW, by dividing the required pump power by the pump efficiency.
6.6. List of References
 Engineering ToolBox, (2004). Pipe Lengths. [online] Available:
https://www.engineeringtoolbox.com /pipe-length-d_744.html [2019, 31 July].
 GAPS Guidelines. 2015. OIL AND CHEMICAL PLANT LAYOUT AND SPACING. GAPS2.5.2.
Connecticut: Global Asset Protection Services LLC.
 Geankoplis, C. J. 1993. Transport Process and Unit Operations, Third Edition. Minnesota:
Prentice-Hall International, Inc.
 Kreider, J. F. & Rabl. A. 1996. Fluid Flow Pump System. Available: http://www.eninepump.
com/zhichi/technical_6.pdf [2019, August 1].
 KSB. 2005. Technical information. Selecting Centrifugal Pumps. 4th edition. Germany: KSB
Aktiengesellschaft.
 Morgan, S. 2016. Pump Sizing: Bridging the Gap between Theory and Practice. American
Institute of Chemical Engineers.
 Welty, J. R., Wicks, C. E., Wilson, R. E., Rorrer, G. L. 2007. Fundamentals of Momentum, Heat
and Mass Transfer. 5th Edition. Oregon: Wiley
 Silla, H. 2003. Chemical Process Engineering Design and Economics. New Jersey: Marcel
Dekker, Inc.
88
 Sinnott R., Towler, G. 2008. Chemical Engineering Design. Principles, Practice and Economics of
Plant and Process Design. London: Elsevier.
 Yaws, C. 1999. Chemiccal Properties Handboek. New York: Mc Graw-Hill.
 National Centre for Biotechnology Information. 2019. PubChem Database. Ethylbenzene.
[Online]. Available: https://pubchem.ncbi.nlm.nih.gov/compound/Ethylbenzene (2019, 7 August).
 Leeuw, T. R.1995. Rating Pumps. Cape Town: Caltex.
 Coulson, J. M., Richardson, J. F., Backhurst, J. R., Harker, J. H. 1999. Fluid Flow, Heat Transfer
and Mass Transfer. Volume 1. Sixth Edition. Oxford: Butterworth Heineman.
89
Appendix A
Process Flow Information of Styrene Production
Ethylbenzene
steam 1 2
Hydrogen
Methane
7 Toluene
4
Styrene Reactor Three Phase
9
(R-401) Separator
(V-401)
3 5
Toluene
Tower (T-401) Styrene
Tower (T-402)
8
6
10 Styrene
Flow sheet for Styrene Production
Waste water
Figure 1.5: PFD of Styrene Production
Table 1.10: Stream Information
Stream Description Stream contents

number
1 Fresh feed stream to plant Ethyl benzene and steam
2 Feed to reactor including fresh feed and Ethyl benzene and steam
recycle stream
3 Reactor outlet stream Ethyl benzene, water, styrene, hydrogen,
methane and toluene
4 Top outlet stream of separator Hydrogen and methane
5 Feed to toluene tower Ethyl benzene, styrene, and toluene
6 Bottom outlet stream of separator Water
7 Top outlet stream of toluene tower Toluene
8 Inlet stream to styrene tower Ethyl benzene and styrene
9 Recycle stream Ethyl benzene
10 Bottom outlet stream of styrene tower Styrene
Assumptions and Utilised Data
90
Given Information:
 Primary reaction is reversible and limited by equilibrium

 Reactions occur at high temperatures (800K – 950K) and low pressures (0.4 – 1.4 bar) to shift
equilibrium to the right to favour styrene production
 Side reactions become significant at higher temperatures
 Assume that the only side reaction is hydrogenolysis of ethyl benzene to produce toluene and
methane
 Proprietary Iron catalyst is used to minimise side reactions
 Styrene production is run with mixture of ethyl benzene and steam
 Steam acts as diluent to shift equilibrium to the right in favour of styrene and tends to limit side
reactions and extend catalyst life
 Typical steam to ethyl benzene ratios entering reactor are 6:1 – 12:1
 The primary reaction proceeds to 80% of its equilibrium value
Table 1.11: Fractional selectivity to Toluene
Temperature (K) Fractional Selectivity to Toluene

800 0.01
850 0.03
900 0.06
950 0.13
Table 1.12: Molar mass for chemical species
Molar mass (kg/kmol)

EB 106.18
water 18.02
Styrene 104.16
Hydrogen 2.02
Methane 16.05
Toluene 92.15
91
Formulas and Sample Calculations
Reaction Equations
Primary Reaction: C8H10 ↔ C8H8 + H2
Ethyl benzene Styrene Hydrogen
Side Reaction: C8H10 + H2 → C7H8 + CH4
Ethyl benzene Hydrogen Toluene Methane
Material balance around Reactor
 Block Flow Diagram
EB(P) ↔ S + H(P)
Feed EB(S) + H(S) → M + To Product
𝐹𝐸𝐵0 𝐹𝐸𝐵1
𝐹𝑆𝑡𝑒𝑎𝑚0 𝐹𝑆1
𝐹𝐻1
𝐹𝑀1
𝐹𝑇𝑜𝑙1
𝐹𝑊𝑎𝑡𝑒𝑟1
First Assumptions of conditions used:
T = 800 K
P = 0.4 bar
Ratio of steam to EB: 6:1
 Calculation of total feed

If 𝐹𝐸𝐵0 = 100 kmol/hr then,
𝐹𝑆𝑡𝑒𝑎𝑚0 = 600 kmol/hr
92
Total Feed: 𝐹𝑇0 = 𝐹𝐸𝐵0 + 𝐹𝑆𝑡𝑒𝑎𝑚0 = 100 + 600
∴ 𝐹𝑇0 = 700 kmol/hr
 Calculation of equilibrium constant at temperature 800K
lnKeq = 15.5408 – 14852.6/T

14852.6
𝑒 𝑙𝑛𝐾𝑒𝑞 = 𝑒15.5408− 800
𝐾𝑒𝑞 = 0.0486
Calculation of Keq at temperature 800K
Kp = 0.8Keq
= 0.8(0. 0486) = 0.0389
 Conversion calculation
With steam:
−𝐾𝑃 𝐹𝑆𝑡𝑒𝑎𝑚0 + √ (𝐾𝑃 𝐹𝑆𝑡𝑒𝑎𝑚0 )2 +4 𝐹𝐸𝐵0 (𝑃+ 𝐾𝑃 )𝐾𝑃 𝐹𝑇

0
𝑋𝐸𝐵 =
2𝐹𝐸𝐵0 (𝑃 + 𝐾𝑃 )
−(0.0389)(600)+ √(0.0389×600)2+4 (100)(0.4+ 0.0389)(0.0389)(700)

𝑋𝐸𝐵 = 2(100) (0.4 +0.0389)
= 0.565
Without steam:
−𝐾𝑃 𝐹𝑆𝑡𝑒𝑎𝑚0 + √ (𝐾𝑃 𝐹𝑆𝑡𝑒𝑎𝑚0 )2 +4 𝐹𝐸𝐵0 (𝑃+ 𝐾𝑃 )𝐾𝑃 𝐹𝑇

0
𝑋𝐸𝐵 = 2𝐹𝐸𝐵0 (𝑃 + 𝐾𝑃 )
0+ √0+4 (100)(0.4+ 0.0389)(0.0389)(100)

𝑋𝐸𝐵 = 2(100) (0.4 +0.0389)
= 0.298
 Calculation of unreacted product, ethyl benzene
93
𝐹𝐸𝐵0 − 𝐹𝐸𝐵1
𝑋𝐸𝐵 =
𝐹𝐸𝐵0
𝐹𝐸𝐵1 = 𝐹𝐸𝐵0 − 𝑋𝐸𝐵 𝐹𝐸𝐵0
𝐹𝐸𝐵1 = 100 − (0.565)(100) = 43.456 kmol
 Material Balance
Assumption of perfect separation was made
Steady state is assumed therefore: Accumulation = 0 kmol
Accumulation =In - Out + Generation - Consumption
Out = In + Generation – Consumption
Ethyl benzene: 𝐹𝐸𝐵1 = 𝐹𝐸𝐵0 − 𝜉1 − 𝜉2 (Eq. 1)
Steam / water: 𝐹𝑆𝑡𝑒𝑎𝑚0 = 𝐹𝑊𝑎𝑡𝑒𝑟1 = 6(𝐹𝐸𝐵0 )
Hydrogen: 𝐹𝐻1 = 𝜉1 − 𝜉2
Styrene: 𝐹𝑆1 = 𝜉1
Methane: 𝐹𝑀1 = 𝜉2
Toluene: 𝐹𝑇1 = 𝜉2
𝐹𝑇𝑜𝑙1 𝜉
Fractional Selectivity to Toluene: 𝑆 = 𝐹 𝑆1
= 𝜉2
1
Simultaneous equations to find extent of reactions 𝜉1 and 𝜉2
At selected Temperature for this case (800 K),
𝜉2
𝑆 = 0.01 = 𝜉1
∴ 𝜉2 = 0.01𝜉1
So, substituting into (Eq. 1) 𝐹𝐸𝐵1 = 𝐹𝐸𝐵0 − 𝜉1 − 𝜉2
43.456 = 100 − 𝜉1 − 0.01𝜉1
1.01𝜉1 = 100 − 43.456
∴ 𝜉1 = 55.984 𝑘𝑚𝑜𝑙 and 𝜉2 = 0.560 𝑘𝑚𝑜𝑙
94
Therefore, at the above conditions
Ethyl benzene: 𝐹𝐸𝐵1 = 43.456 kmol
Steam / water: 𝐹𝑆𝑡𝑒𝑎𝑚0 = 𝐹𝑊𝑎𝑡𝑒𝑟1 = 600 kmol
Hydrogen: 𝐹𝐻1 = 55.424 kmol
Styrene: 𝐹𝑆1 = 55.984 kmol
Methane: 𝐹𝑀1 = 0.560 kmol
Toluene: 𝐹𝑇1 = 0.560 kmol
Converting moles to mass
𝑚𝑎𝑠𝑠 𝑚
𝑀𝑜𝑙𝑒𝑠 = : 𝑛=
𝑀𝑜𝑙𝑎𝑟 𝑚𝑎𝑠𝑠 𝑀𝑟
Gross Profit calculation
Table 1.13: Given Prices of Substances
Substance Price (R/kg) Description

Ethyl benzene R 7.50 Cost
Steam @ 800K R 0.66 Cost
Waste water R 0.06 Cost
Styrene R 9.30 Selling Price
Toluene R 6.60 Selling Price
Methane R 1.29 Selling Price
Hydrogen R 3.15 Selling Price
Preliminary Gross Profit = Selling Price – Cost Price
= [(Mass Styrene)out stream 10 × (SP Styrene)] + [(Mass Methane)out stream 4 × (SP Methane)] + [(Mass
Toluene)out stream 7 × (SP Toluene)] + [(Mass Hydrogen)out stream 10 × (SP Hydrogen)] – [(Mass EB)in
stream 1 × (CP EB)] – [(Mass Steam)in stream 1 × (CP Stream)] – [(Mass Waste Water)in stream 6 × (CP
Waste Water)]
= (5.85E+03)(9.30) + (9.02E+00)(1.29) + (5.18E+01)(6.60) + (1.12E+02)(3.15) - (6.03E+03)(7.50) –

(1.08E+04)(0.66) - (1.08E+04)(0.06)
= 2.2E+02 R/hr
95
Conversion to per year: PGP = 2.2E+02 x 8000 = 1.7E7 R/yr
Scaling Factor
𝑀𝑎𝑠𝑠 𝐸𝐵 𝑖𝑛 𝑀𝑎𝑠𝑠 𝐸𝐵 𝑖𝑛
(𝑀𝑎𝑠𝑠 𝑆𝑡𝑦𝑟𝑒𝑛𝑒 𝑜𝑢𝑡) = (𝑀𝑎𝑠𝑠 𝑆𝑡𝑦𝑟𝑒𝑛𝑒 𝑜𝑢𝑡 )
𝑏𝑎𝑠𝑒𝑑 𝑜𝑛 𝑎𝑠𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑏𝑎𝑠𝑒𝑑 𝑜𝑛 𝑔𝑖𝑣𝑒𝑛 𝑐𝑎𝑝𝑎𝑐𝑖𝑡𝑦
𝑀𝑎𝑠𝑠 𝐸𝐵 𝑖𝑛
(𝑀𝑎𝑠𝑠 𝐸𝐵 𝑖𝑛 )𝑏𝑎𝑠𝑒𝑑 𝑜𝑛 𝑔𝑖𝑣𝑒𝑛 𝑐𝑎𝑝𝑎𝑐𝑖𝑡𝑦 = ( ) . (𝑀𝑎𝑠𝑠 𝑆𝑡𝑦𝑟𝑒𝑛𝑒 𝑜𝑢𝑡)𝑏𝑎𝑠𝑒𝑑 𝑜𝑛 𝑔𝑖𝑣𝑒𝑛 𝑐𝑎𝑝𝑎𝑐𝑖𝑡𝑦
𝑀𝑎𝑠𝑠 𝑆𝑡𝑦𝑟𝑒𝑛𝑒 𝑜𝑢𝑡 𝑏𝑎𝑠𝑒𝑑 𝑜𝑛 𝑎𝑠𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛
100 × 106.2
=
56.2 × 104.2
× 280000000 ÷ 8000 × 104.2
= 1.8 × 280000000 ÷ 8000 × 104.2
= 546 kg/hr
Additional Graphs Relating Operating Conditions to Conversion
1.2
0.8
Conversion
Temperature: 800K
0.6
Temperature: 850K
0
0.3 0.8 1.3 1.8
Pressure (atm)
Figure 1.6: Relationship between Conversion and Pressure at Varying Temperatures with No Steam
96
1.2
0.8
Conversion
Temperature: 800K
0
0.3 0.8 1.3 1.8
Pressure (atm)
6:1
1.2
0.8
Conversion
Temperature: 800K

Temperature: 950K
0.2
0
0.3 0.5 0.7 0.9 1.1 1.3 1.5
Pressure (atm)
7:1
1.2
0.8
Conversion
Temperature: 800K
0
0.3 0.8 1.3 1.8
Pressure (atm)
8:1
97
1.2
0.8
Conversion
Temperature: 800K
0
0.3 0.8 1.3 1.8
Pressure (atm)
9:1
1.2
1
0.8
Conversion
Temperature: 800K
0.6
Temperature: 850K
0.4
Temperature: 900K
0.2
Temperature: 950K
0
0.3 0.5 0.7 0.9 1.1 1.3 1.5
Pressure (atm)
10:1
1.2
1
0.8
Conversion
Temperature: 800K
0.6
Temperature: 850K
0.4
Temperature: 900K
0
0.3 0.5 0.7 0.9 1.1 1.3 1.5
Pressure (atm)
11:1
98
Additional Graphs Relating Operating Conditions to Profit
95000000
75000000
Pressure: 0.4 bar
55000000
Profit (R/hour)

-25000000
Pressure: 1.2 bar
-45000000
0 5 10 15
95000000
75000000
Profit (R/hour)

-65000000
0 5 10 15
99
Appendix B
35
34
05-MV-01
2 3 4 7 P-506
1 M-501 E-501
R-501 15
Ethylbenzene P-501
Steam hms K-501
Hydrogen
cw Methane
14 E-506
05-PV-01 22 Toluene
21 25
fg V-502 24
8
6 P-504
23
cw Steam Steam cw
H-501 E-502
9 E-504 T-501 E-509
12 13 18 19 20
10 11 05-CV-01 V-501 32
ng E-505
E-503 31
J-501 P-502 27
A V-503
ms cw Steam
hps 5 E-507 33
26 Seam
28 T-502
30
29
16
E-508 37
Steam
Key P-505 E-510
A – Air 36
cw – Cooling water Styrene
E – Heat exchanger
fg – Flue gas 38
H – Heater 39
hms – Heating molten salt
hps – High pressure steam
J – Ejector P-507
ms – motive steam
M – Mixer 17
ng – natural gas
P – Pump Waste
T – Tower
V – Drum water
P-503
E-501 E-502 E-503 E-504 E-505 E-506 E-507 E-508 E-509 E-510 H-501 J-501 K-501 M-501 R-501 V-501 V-502 V-503
Feed Product Product V-501 T-501 T-501 T-501 T-502 T-502 T-502 Fired Ejector Gas Feed Reactor Flash Ovhd Ovhd
preheat cooler condenser preheat preheat condenser rebolier preheat condenser reboiler heater compress. mixer drum drum drum
P-501 P-502 P-503 P-504 P-505 P-506 P-507

EB feed V-501 Waste Toluene T-502 Recycle Styrene
pump feed pump water pump pump feed pump pump pump
100
Table 2.1: Equipment list for PFD
Equipment number Description

E-501 Feed pre-heat exchanger
E-502 Reactor product condenser
E-503 Reactor product cooler
E-504 V-501 preheater
E-505 Preheater to T-501
E-506 T-501 condenser
E-507 T-501 re-boiler
E-508 Preheater to T-502
E-509 T-502 condenser
E-510 T-502 re-boiler
H-501 Fired Heater
J-501 Ejector
K-501 Hydrogen & Methane Compressor
M-501 Mixer
V-501 Flash separator
V-502 T-501 reflux drum
V-503 T-502 reflux drum
T-501 Toluene distillation column
T-502 Styrene distillation column
R-501 Styrene reactor
P-501 Ethylbenzene feed pump
P-502 V-501 feed pump
P-503 waste water pump
P-504 Toluene pump
P-505 T-502 feed pump
P-506 Recycle stream pump
P-507 Styrene pump
101
Table 2.2: Stream number descriptions
Stream nr. Description

1 Fresh feed
2 Fresh feed after P-501
3 Feed w/ recycle
4 Feed after E-501
5 High pressure steam to furnace
6 Steam out of furnace
7 Reactor feed
8 Reactor outlet
9 Reactor outlet, after E-502
10 Feed to E-503
11 Process stream from E-503
12 Feed to E-504
13 Feed to flash V-501
14 Light gases from V-501
15 Compressed gases to storage
16 Water out of V-501
17 Waste water product out of P-503
18 Organic liquid out of V-501
19 Feed to E-505
20 Feed to T-501
21 T-501 top reflux before E-506
22 T-501 top product after E-506
23 Reflux to T-501
24 T-501 distillate from V-502
25 Toluene product
26 Bottoms product out of T-501
27 Bottom reflux into T-501
28 T-501 bottoms product to P-505
29 T-501 bottoms product to E-508
30 Feed to T-502
31 T-502 top reflux to E-509
32 T-502 top product after E-509
33 Reflux to T-502
102
34 Recycle stream before P-505
35 Recycle stream to M-501
36 T-502 bottom reflux out
37 Bottom reflux into T-502
38 Styrene product to P-502
39 Styrene Product
103
Appendix C
Sample calculations for Chapter 3 Heat Exchanger Design
Specific Heat Capacity:
Ethylbanzene: 𝐶𝑝𝑙 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2 + 𝐷𝑇 3
=102 + 0.56(297.972) + (−1.56𝑥10−3 )(297.972)2 +
(2.01𝑥10−6 )(297.972)3
= 183.5 J/kmolK
= 0.184 kJ/kmolK
Duty:
Ethyl benzene: 𝑄𝑐 = 𝑛̇ 𝐶𝑝∆𝑇𝑙𝑖𝑞𝑢𝑖𝑑 + 𝑛̇ ∆𝐻 + 𝑛̇ 𝐶𝑝∆𝑇𝑔𝑎𝑠
= [0.1697 x 0.184 x (409.15– 297.97)] + [54.8x(1-(409.14/617))0.388] + [0.1697(kmol/s) x
0.212 x ( 521.79-409.15)K]
= 3.47 + 35.93 + 3.24 = 42 kW
Styrene 𝑄𝑐 = 36.6 kW
Toluene 𝑄𝑐 = 33.6 kW
Total 𝑄𝑐 = 42 + 36.6 + 33.6 = 113 kW
(CheGuide, 2016)
Thermal Conductivity for liquid:

𝑇 2⁄
Ethylbenzene: 𝐿𝑜𝑔10 𝑘𝑙𝑖𝑞 = 𝐴 + 𝐵(1 − 𝐶 ) 7
𝑇 2⁄
𝐿𝑜𝑔10 𝑘𝑙𝑖𝑞 = (−1.7498) + 1.0437(1 − ) 7
617.17
𝑘𝑙𝑖𝑞 = 10−0.8853
= 0.13 W/mK
104
Table 3.9: Thermal conductivity formula coefficients of organic compounds in liquid phase
Units: A B C Tmin Tmax Kliq25C KliqTmin KliqTma

W/mK x
EB -1.75 1.04 617 178 586 0.130 0.155 0.0178
Styrene -1.70 1 648 243 616 0.137 0.146 0.0198
Methane -1.10 0.539 191 91 181 --- 0.221 0.0799
Toluene -1.67 0.977 592 178 562 0.134 0.160 0.0212
Thermal Conductivity for gas:

Ethylbenzene: 𝑘𝑔𝑎𝑠 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2
= −0.00797 + (4.0572E − 05 × 521.79) + (6.7289E − 08)521.792
= 0.0315 W/mK
Table 3.10: Thermal conductivity formula coefficients of organic compounds in gas phase
Units: A B C Tmin Tmax kgasT2 kgasTmin kgasTmax

W/mK 5
EB -0.00797 4.0572E- 6.7289E- 400 825 --- 0.01903 0.0713
05 08
Styrene -0.00712 4.5538E- 3.9529E- 273 973 0.00997 0.00826 0.07461
05 08
Toluene -0.00776 4.4905E- 6.4514E- 350 800 --- 0.01586 0.06945
05 08
Density
𝑇 𝑛
Ethylbenzene: 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 = 𝐴 × 𝐵−(1−𝑇𝑐)
297.972 0.2921
=0.28889 × 0.26438−(1− 617.17 )
= 1.0754 g/ml
= 1075.4 kg/m3
Table 3.11: Coefficients of species for density formula

EB 0.289 0.264 0.292 617 178 617
Styrene 0.194 0.263 0.286 648 243 648
Toluene 0.3 0.271 0.299 562 198 592
105
Viscosity for liquid:
𝐵
Ethylbenzene: 𝐿𝑜𝑔10 𝜇𝑙𝑖𝑞 = 𝐴 + 𝑇 + 𝐶𝑇 + 𝐷𝑇 2
830.65
= −5.2585 + 297.972 + (0.010784 × 297.972) + (−1.06E −
05)(297.972)2
𝜇𝑙𝑖𝑞 = 10−0.20024 = 0.63061𝑐𝑃
=0.631 × 10−3 𝑘𝑔/𝑚𝑠
Table 3.12: Coefficients of species for viscosity formula in liquid phase
Units: cP A B C D Tmin Tmax nliqT25 nliqTmax

EB -5.26 831 0.0108 -1.06E-05 210 617 0.629 0.05
Styrene -8.03 1267 0.0161 -1.35E-05 243 648 0.675 0.052
Toluene -5.16 811 0.0105 -1.05E-05 200 592 0.548 0.052
Viscosity for gas:

Ethylbenzene: 𝜇𝑔𝑎𝑠 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2
= −4.267 + (0.24735 × 521.79) + (−5.4264E − 05)(521.79)2
= 110 cP
= 0.11 kg/ms
Table 3.13: Coefficients of species for viscosity formula in gas phase
Units: A B C Tmin Tma ngasT25 ngasTmin ngasTmax

cP x
EB -4.267 0.247 -5.43E-05 409 1000 --- 87.8 189
Styrene -10 0.252 -3.79E-05 243 1000 61.7 48.9 204
Toluene 1.79 0.236 -9.35E-06 275 600 71.2 65.9 140
Table 3.14: Constants for use in Equation 13
Triangular tube pitch, pt = 1.25do

Nr. of passes 1 2 4 6 8
K 0.319 0.249 0.175 0.0743 0.0365
N 2.142 2.207 2.285 2.499 2.675
106
Figure 3.4: Shell Bundle Clearance
Figure 3.5: Tube side heat transfer factor
107
Figure 3.6: Shell side heat transfer factors, segmental baffles
Figure 3.7: Fouling factors (coefficients) typical values
Figure 3.8: Conductivity of metals
108
Figure 3.9: Diagram for costing hest exchanger (Coulson & Richardson, 2005:253)
Purchase Cost of Heat exchanger

The formula below is from Coulson & Richardson (2005,253):
Purchase cost = bare cost from figure x type factor x pressure factor = $10 000 x 1.3 x = $13 000
𝐶𝑜𝑠𝑡 𝐼
Effect of time on purchased equipment, due to inflation: 𝐶𝑜𝑠𝑡𝑎 = 𝐼𝑎 (Coulson & Richardson 2005:244)
𝑏 𝑏
𝐼𝑎 603.1
𝐶𝑜𝑠𝑡𝑎 = 𝐶𝑜𝑠𝑡𝑏 = $13 000 × = $19 650
𝐼𝑏 399
109
Appendix D
Table 4.7: Conversion at different reactor condition combinations
Conditions With steam: Ratios of steam to EB No
T (K) P (atm) 6 7 8 9 10 11 12 Steam
800 0.39 0.57 0.59 0.61 0.63 0.64 0.66 0.67 0.30
800 0.59 0.50 0.52 0.54 0.56 0.57 0.59 0.60 0.25
800 0.79 0.45 0.47 0.49 0.51 0.52 0.54 0.55 0.22
800 0.99 0.41 0.43 0.45 0.47 0.48 0.50 0.51 0.19
800 1.18 0.39 0.41 0.42 0.44 0.45 0.47 0.48 0.18
800 1.38 0.36 0.38 0.40 0.41 0.43 0.44 0.45 0.17
850 0.39 0.75 0.77 0.79 0.80 0.81 0.82 0.83 0.48
850 0.59 0.69 0.71 0.72 0.74 0.75 0.77 0.78 0.40
850 0.79 0.64 0.66 0.68 0.69 0.71 0.72 0.73 0.36
850 0.99 0.60 0.62 0.64 0.66 0.67 0.68 0.70 0.32
850 1.18 0.57 0.59 0.61 0.62 0.64 0.65 0.67 0.30
850 1.38 0.54 0.56 0.58 0.60 0.61 0.63 0.64 0.28
900 0.39 0.87 0.89 0.90 0.90 0.91 0.92 0.92 0.66
900 0.59 0.83 0.84 0.86 0.87 0.88 0.88 0.89 0.58
900 0.79 0.79 0.81 0.82 0.83 0.84 0.85 0.86 0.53
900 0.99 0.76 0.78 0.79 0.81 0.82 0.83 0.84 0.49
900 1.18 0.73 0.75 0.77 0.78 0.79 0.80 0.81 0.45
900 1.38 0.71 0.73 0.74 0.76 0.77 0.78 0.79 0.43
950 0.39 0.94 0.95 0.95 0.95 0.96 0.96 0.96 0.81
950 0.59 0.91 0.92 0.93 0.94 0.94 0.94 0.95 0.74
950 0.79 0.89 0.90 0.91 0.92 0.92 0.93 0.93 0.69
950 0.99 0.87 0.88 0.89 0.90 0.91 0.91 0.92 0.65
950 1.18 0.85 0.86 0.87 0.88 0.89 0.90 0.90 0.62
950 1.38 0.83 0.85 0.86 0.87 0.88 0.88 0.89 0.59
The selected optimal conditions are highlighted.
110
Table 4.8: Profits at different conversions, as shown by table 7 above
Conditions With steam: Ratios of steam to EB
T (K) P (atm) 6 7 8 9 10 11 12 No Steam
800 0.39 7.05E+07 4.05E+07 9.18E+06 -2.33E+07 -5.68E+07 -9.11E+07 -1.26E+08 1.71E+08
800 0.59 3.03E+07 1.75E+05 -3.12E+07 -6.37E+07 -9.71E+07 -1.31E+08 -1.66E+08 1.42E+08
800 0.79 2.85E+06 -2.77E+07 -5.94E+07 -9.20E+07 -1.25E+08 -1.60E+08 -1.94E+08 1.24E+08
800 0.99 -1.75E+07 -4.85E+07 -8.05E+07 -1.13E+08 -1.47E+08 -1.81E+08 -2.16E+08 1.11E+08
800 1.18 -3.35E+07 -6.48E+07 -9.72E+07 -1.30E+08 -1.64E+08 -1.99E+08 -2.34E+08 1.02E+08
800 1.38 -4.64E+07 -7.81E+07 -1.11E+08 -1.44E+08 -1.78E+08 -2.13E+08 -2.48E+08 9.46E+07
850 0.39 1.59E+08 1.27E+08 9.29E+07 5.79E+07 2.21E+07 -1.46E+07 -5.18E+07 2.62E+08
850 0.59 1.23E+08 9.21E+07 5.93E+07 2.54E+07 -9.42E+06 -4.51E+07 -8.14E+07 2.22E+08
850 0.79 9.63E+07 6.56E+07 3.34E+07 7.48E+04 -3.42E+07 -6.93E+07 -1.05E+08 1.97E+08
850 0.99 7.49E+07 4.44E+07 1.25E+07 -2.05E+07 -5.45E+07 -8.93E+07 -1.25E+08 1.78E+08
850 1.18 5.73E+07 2.69E+07 -4.90E+06 -3.78E+07 -7.16E+07 -1.06E+08 -1.41E+08 1.64E+08
850 1.38 4.25E+07 1.21E+07 -1.97E+07 -5.25E+07 -8.63E+07 -1.21E+08 -1.56E+08 1.53E+08
900 0.39 1.99E+08 1.63E+08 1.25E+08 8.69E+07 4.81E+07 8.84E+06 -3.08E+07 3.42E+08
900 0.59 1.76E+08 1.41E+08 1.05E+08 6.77E+07 2.99E+07 -8.41E+06 -4.72E+07 3.02E+08
900 0.79 1.56E+08 1.22E+08 8.72E+07 5.11E+07 1.41E+07 -2.35E+07 -6.17E+07 2.74E+08
900 0.99 1.40E+08 1.06E+08 7.20E+07 3.65E+07 7.16E+04 -3.70E+07 -7.47E+07 2.52E+08
900 1.18 1.25E+08 9.25E+07 5.85E+07 2.35E+07 -1.25E+07 -4.91E+07 -8.63E+07 2.35E+08
900 1.38 1.12E+08 8.00E+07 4.65E+07 1.18E+07 -2.37E+07 -6.01E+07 -9.70E+07 2.21E+08
950 0.39 1.70E+08 1.30E+08 8.99E+07 4.94E+07 8.66E+06 -3.23E+07 -7.35E+07 3.63E+08
950 0.59 1.58E+08 1.20E+08 8.02E+07 4.05E+07 3.10E+05 -4.02E+07 -8.09E+07 3.35E+08
950 0.79 1.48E+08 1.10E+08 7.13E+07 3.22E+07 -7.46E+06 -4.75E+07 -8.78E+07 3.12E+08
950 0.99 1.38E+08 1.01E+08 6.31E+07 2.44E+07 -1.47E+07 -5.43E+07 -9.43E+07 2.94E+08
950 1.18 1.29E+08 9.28E+07 5.54E+07 1.72E+07 -2.16E+07 -6.08E+07 -1.00E+08 2.78E+08
950 1.38 1.21E+08 8.52E+07 4.82E+07 1.04E+07 -2.80E+07 -6.69E+07 -1.06E+08 2.65E+08
111
Figure 4.8: Friction factor
Figure 4.9: Heat transfer factor
112
Appendix E
Solubility Calculations
𝐵 𝐶
Solubility Formula: 𝑙𝑜𝑔10 𝑆 = 𝐴 + 𝑇 + 𝑇2
𝐵 𝐶 −6644 989907
(13.365+ + )
Ethylbenzene: 𝑆 = 10(𝐴+𝑇 +𝑇2 ) = 10 323.15 323.152 = 192.94 𝑝𝑝𝑚(𝑤𝑡)
192.94 𝑝𝑝𝑚(𝑤𝑡)
Conversion: × 100% = 0.0193 = 0.02 %
1000000 𝑝𝑝𝑚(𝑤𝑡)
𝐵 𝐶 −7591 1000000
(15.471+ + )
Toluene: 𝑆 = 10(𝐴+𝑇+𝑇2 ) = 10 323.15 323.152 = 647.33 𝑝𝑝𝑚(𝑤𝑡)
647.33 𝑝𝑝𝑚(𝑤𝑡)
Conversion: 1000000 𝑝𝑝𝑚(𝑤𝑡) × 100% = 0.0647 = 0.06 %
kmol/hr of hydrocarbon in water (L11):

Ethylbenzene: 𝑆(%) × 𝐹𝐸𝐵 = 0.02 × 51.11 = 0.00987 𝑘𝑚𝑜𝑙/ℎ𝑟
Styrene: 𝑆(%) × 𝐹𝑆 = 0.03 × 336.49 = 0.1010 𝑘𝑚𝑜𝑙/ℎ𝑟
Toluene: 𝑆 (%) × 𝐹𝑇 = 0.06 × 20.20 = 0.0131 𝑘𝑚𝑜𝑙/ℎ𝑟
kmol/hr of hydrocarbon in vapour (V):

𝑃
𝑈𝑠𝑖𝑛𝑔 𝐻𝑒𝑛𝑟𝑦 ′ 𝑠 𝑙𝑎𝑤 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡: 𝑥𝑖 = 𝑦𝑖
𝐻𝑖
𝐻𝐸𝐵
Ethylbenzene: 𝑦𝐸𝐵 = 𝑥𝐸𝐵
𝑃
𝑎𝑡𝑚
452.3 0.00987 𝑚𝑜𝑙𝐹𝑟𝑎𝑐
𝑚𝑜𝑙𝐹𝑟𝑎𝑐
= 0.986923 𝑎𝑡𝑚 × = 0.0884 𝑘𝑚𝑜𝑙/ℎ𝑟
1 𝑏𝑎𝑟 × 51.11
𝑏𝑎𝑟
𝐻𝑆
Styrene: 𝑦𝑆 = 𝑃
𝑥𝑆
𝑎𝑡𝑚
114.56 0.101 𝑚𝑜𝑙𝐹𝑟𝑎𝑐
= 0.986923 𝑎𝑡𝑚 × = 0.0439 𝑘𝑚𝑜𝑙/ℎ𝑟
1 𝑏𝑎𝑟 × 336.62
𝑏𝑎𝑟
𝐻𝑇
Toluene: 𝑦𝑇 = 𝑥𝑇
𝑃
𝑎𝑡𝑚
352.9 0.0131 𝑚𝑜𝑙𝐹𝑟𝑎𝑐
= 0.986923 𝑎𝑡𝑚 × = 0.232 𝑘𝑚𝑜𝑙/ℎ𝑟
1 𝑏𝑎𝑟 × 20.20
𝑏𝑎𝑟
kmol/hr of hydrocarbon in liquid phase (L10):

Ethylbenzene: 𝐹𝐸𝐵 − 𝑉𝐸𝐵 − 𝐿11,𝐸𝐵 = 51.18 − 0.0884 − 0.00987 = 51.08 𝑘𝑚𝑜𝑙/ℎ𝑟
Styrene: 𝐹𝑆 − 𝑉𝑆 − 𝐿11,𝑆 = 336.62 − 0.0439 − 0.1010 = 336.48 𝑘𝑚𝑜𝑙/ℎ𝑟
Toluene: 𝐹𝑇 − 𝑉𝑇 − 𝐿11,𝑇 = 20.20 − 0.232 − 0.0131 = 19.95 𝑘𝑚𝑜𝑙/ℎ𝑟
113
Flash Drum Sizing Calculations
Vapour Density
𝑚 𝑃𝑀𝑎𝑣𝑔
𝜌𝑣 = =
𝑉 𝑅𝑇
𝑀𝑎𝑣𝑔. = 𝑥𝐸𝐵 𝑀𝑊𝐸𝐵 + 𝑥𝑠 𝑀𝑊𝑠 + 𝑥 𝑇 𝑀𝑊𝑇 + 𝑥𝑀 𝑀𝑊𝑀 + 𝑥𝐻 𝑀𝑊𝐻 + 𝑥𝑊 𝑀𝑊𝑊

0.088 0.044 0.232 20.20 316 0
= ( 367 ) (106) + ( 367 ) (104) + ( 367 ) (92.2) + ( 367 ) (16) + (367) (2.02) + (367) (18)
= 2.96 𝑔/𝑚𝑜𝑙
101325×2.69
∴ 𝜌𝑣 = 8.314×323.15×1000 = 11.18 𝑘𝑔/𝑚 3
Liquid Density
Liquid density formula and coefficients in the table below are from Yaws (1999:185).
Table 5.12: Density coefficients

EB 0.28889 0.26438 0.2921 617.17 178.2 617.17
Styrene 0.19383 0.26315 0.2857 648 242.54 648
Toluene 0.29999 0.27108 0.29889 561.79 198.18 591.79
Methane 0.15998 0.2881 0.277 190.58 90.67 190.58
Hydrogen 0.03125 0.3473 0.2756 33.18 13.95 33.18
Water 0.3471 0.274 0.28571 647.13 273.16 647.13
The table below shows the liquid densities of all the components entering the flash drum, using the
formula and coefficients stated above.
Table 5.13: Liquid densities
Substance 𝝆𝑳 (𝒌𝒈/𝒎𝟑 )
843
EB
580
Styrene
824
Toluene
0
Methane
0
Hydrogen
1004
Water
𝜌𝐿 = 𝑥𝐿 𝐸𝐵 𝜌𝐿 𝐸𝐵 + 𝑥𝐿 𝑠 𝜌𝐿 𝑠 + 𝑥𝐿 𝑇 𝜌𝐿 𝑇 + 𝑥𝐿 𝑀 𝜌𝐿 𝑀 + 𝑥𝐿 𝐻 𝜌𝐿 𝐻 + 𝑥𝐿 𝑊 𝜌𝐿 𝑊
51.1+0.00987 336.5+0.101 19.95+0.0131 0 0
= ( 407.5+2448 ) (843) + ( 407.5+2448 ) (580) + ( 407.5+2448 ) (824) + (407.5+2448) (0) + (407.5+2448) (0) +
2448
( ) (1004) = 950.14 𝑘𝑔/𝑚3
407.5+2448
Maximum permissible vapour velocity

𝜌𝐿 −𝜌𝑉
𝑢𝑝𝑒𝑟𝑚 = 𝐾𝑑𝑟𝑢𝑚 √ 𝜌𝑉
114
𝐾𝑑𝑟𝑢𝑚 = (𝑐𝑜𝑛𝑠𝑡. ) exp[𝐴 + 𝐵𝑙𝑛𝐹𝑙𝑣 + 𝐶 (𝑙𝑛𝐹𝑙𝑣 )2 + 𝐷(𝑙𝑛𝐹𝑙𝑣 )3 + 𝐸(𝑙𝑛𝐹𝑙𝑣 )4 ]
𝑊 𝜌
𝐹𝑙𝑣 = 𝑊𝐿 √ 𝜌𝑉
𝑉 𝐿
𝑊𝐿 = 𝑥𝐿 𝐸𝐵 𝑀𝑊𝐸𝐵 + 𝑥𝐿 𝑠 𝑀𝑊𝑠 + 𝑥𝐿 𝑇 𝑀𝑊𝑇 + 𝑥𝐿 𝑀 𝑀𝑊𝑀 + 𝑥𝐿 𝐻 𝑀𝑊𝐻 + 𝑥𝐿 𝑊 𝑀𝑊𝑊

51.1+0.00987 336.5+0.101 19.95+0.0131 0 0
=( ) (106) + ( ) (104) + ( ) (92.2) + ( ) (16.1) + ( ) (2.02) +
407.5+2448 407.5+2448 407.5+2448 407.5+2448 407.5+2448
2448
(407.5+2448) (18) = 30.27 𝑘𝑔/ℎ
𝑊𝑣 = 𝑥𝑣 𝐸𝐵 𝑀𝑊𝐸𝐵 + 𝑥𝑣 𝑠 𝑀𝑊𝑠 + 𝑥𝑣 𝑇 𝑀𝑊𝑇 + 𝑥𝑣 𝑀 𝑀𝑊𝑀 + 𝑥𝑣 𝐻 𝑀𝑊𝐻 + 𝑥𝑣 𝑊 𝑀𝑊𝑊

0.088 0.044 0.232 20.20 316 0
= ( 367 ) (106) + ( 367 ) (104) + ( 367 ) (92.2) + ( 367 ) (16) + (367) (2.02) + (367) (18)
=2.96 kg/hr
20.27 11.18
∴ 𝐹𝑙𝑣 = √ = 1.108
2.96 950.14
Table 5.14: Kdrum constants (Blackwell, 1984)

Constants for Kdrum formula
A B C D E
-1.87748 -0.81458 -0.18707 -0.0145229 -0.001014852
𝑓𝑡 𝑚
𝐾𝑑𝑟𝑢𝑚 = (𝑐𝑜𝑛𝑠𝑡. ) exp[𝐴 + 𝐵𝑙𝑛𝐹𝑙𝑣 + 𝐶 (𝑙𝑛𝐹𝑙𝑣 )2 + 𝐷(𝑙𝑛𝐹𝑙𝑣 )3 + 𝐸(𝑙𝑛𝐹𝑙𝑣 )4 ] = 0.14 = 0.043
𝑠 𝑠
𝜌𝐿 −𝜌𝑉
𝑢𝑝𝑒𝑟𝑚 = 𝐾𝑑𝑟𝑢𝑚 √
𝜌𝑉
950.14−11.18
= 0.043√ = 0.392 𝑚/𝑠
11.18
Cross Sectional Area of Vessel

𝑉(𝑀𝑉𝑉 ) 336.987×2.96
𝐴𝑐 = 𝑢 = 0.392×3600×11.18 = 0.0633 𝑚 2
𝑝𝑒𝑟𝑚 (3600)𝜌𝑉
Vessel Inner Diameter

4𝐴𝑐 4(0.0633)
𝐷=√ 𝜋
=√ 𝜋
= 0.284 ≈ 0.3 𝑚
Vessel Height
𝐻𝑡𝑜𝑡
𝑅𝑢𝑙𝑒 𝑜𝑓 𝑡ℎ𝑢𝑚𝑏: 𝐷
= 3 𝑡𝑜 5 (𝑊𝑎𝑛𝑘𝑎𝑡, 2012: 95)
𝐻𝑡𝑜𝑡
∴ 𝐿𝑒𝑡 =4
𝐷
∴ 𝐻𝑡𝑜𝑡 = 4 × 𝐷 = 1.14 𝑚
Weight of vessel
According to Sinnott & Towler (2005:1004), the total weight of the shell, excluding internal fittings,
115
can be estimated by the formula below:
𝑊𝑉 = 240𝐶𝑤 𝐷𝑚 (𝐻𝑉 + 0.8𝐷𝑚 )𝑡
Cw = 1.08 for vessels with only a few internal fittings (a factor to account for weight of internal
fittings)
t = vessel wall thickness (mm) wstimated to be 4mm
Dm = mean diameter of vessel:
𝐷𝑚 = 𝐷𝑖 + 𝑡 × 10−3
= 0.3 + 4 × 10−3 = 0.288 𝑚
∴ 𝑊𝑉 = 240(1.08)(0.0288)(1.14 + 0.8(0.288))(4 × 10−3 ) = 407.39 𝑁 ≈ 408 𝑁

Mass = weight x gravitational acceleration = 407.39 x 9.81 = 42 kg
Material Balance Calculations for T-502

 𝐹 =𝐷+𝐵
𝐹 = 51.08 + 336.48 = 387.56 𝑘𝑚𝑜𝑙/ℎ𝑟

𝐵 = 𝐵𝑠𝑡𝑦𝑟𝑒𝑛𝑒 + 𝐵𝐸𝑡ℎ𝑦𝑙𝑏𝑒𝑛𝑧𝑒𝑛𝑒
280 000 ton 1000 kg 1 year kg
Bstyrene = year
× ton
× 8200 hours × 104.16 kmol = 327.83 kmol/hr
And since 99 % purity of styrene is required, the amount above is let at 99% of the Bottoms product.
Bstyrene
∴ 𝐵 = Bstyrene + ( ) × 0.1 = 327.83 + 33.11 = 360.94 kmol/hr
0.99
𝐷 = (𝐹𝑠𝑡𝑦𝑟𝑒𝑛𝑒 − 𝐵𝑠𝑡𝑦𝑟𝑒𝑛𝑒 ) + (𝐹𝐸𝑡ℎ𝑦𝑙𝑏𝑒𝑛𝑧𝑒𝑛𝑒 − 𝐵𝐸𝑡ℎ𝑦𝑙𝑏𝑒𝑛𝑧𝑒𝑛𝑒 )

= (336.48 − 327.83) + (51.08 − 33.11) = 26.73 kmol/hr
 𝐹𝑧𝐹 = 𝐷𝑥𝐷 + 𝐵𝑥𝐵
51.08
𝑧𝐹 = 387.56 = 0.132
17.97
𝑥𝐷 = 26.73 = 0.672
33.11
𝑥𝐵 = = 0.092
360.94
Calculation of K-values:
𝑀𝑜𝑙𝑒 𝑜𝑓 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑖 𝑖𝑛 𝑣𝑎𝑝𝑜𝑢𝑟 𝑝ℎ𝑎𝑠𝑒
𝐾𝑖 = 𝑀𝑜𝑙𝑒 𝑜𝑓 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑖 𝑖𝑛 𝑙𝑖𝑞𝑢𝑖𝑑 𝑝ℎ𝑎𝑠𝑒
0.949
𝐾𝐸𝐵 = 0.937 = 1.013
1−0.949
𝐾𝑆 = 1−0.937 = 0.81
116
Average relative volatility:
𝐾𝐸𝐵 1.013
∝= = = 1.25
𝐾𝑆 0.81
Minimum reflux ratio:

1 𝑥 ∝(1−𝑥𝐷 )
𝑅𝑚 = ∝−1 [ 𝑧𝐷 − 1−𝑧𝐹
]
𝐹
1 0.672 1,25(1−0.672)
= 1,25−1 [0.132 − 1−0.132
] = 18.43 ≈ 19
Calculation of q:
𝜆𝐹 +𝐶𝑝𝐹 (𝑡𝐵𝑃 −𝑡𝐹 )
𝑞= 𝜆𝐹
𝑡𝐹 = 323.15 𝐾
367
366
365
364
Temperature (K)
363
362 X1 (liquid data)
361 y1 (vapour data)
360
359
358
357
0 0.2 0.4 0.6 0.8 1
x1, y1
Figure 5.7: Experimental vapour-liquid equilibrium EB-S data (at P=0.2 bar)
Using the T-x-y data from Table 7,the T-x-y graph above was drawn and at xF = 0.132,
𝑡𝐵𝑃 = 365 K.
323.15+364.5
Specific heat at 2
= 343.83 𝐾
Specific heat capacity:

The formula for the specific heat capacity of a liquid as a function of temperature was taken from
Yaws (1999:56). The formula and coefficients used in the formula are given below.
𝐶𝑝𝑙 = 𝐴 + 𝐵𝑇 + 𝐶𝑇 2 + 𝐷𝑇 3
Table 5.15: Coefficients for specific heat capacity formula

Species A B C D Tmin (K) Tmax (K)
EB 1.02E+02 5.60E-01 -1.56E-03 2.01E-06 179 555
Styrene 6.67E+01 8.41E-01 -2.16E-03 2.33E-06 244 583
117
𝐶𝑝𝐸𝐵 = (1.02E + 02) + (5.60E − 01)(343.83) + (−1.56E − 03)(343.83)2 + (2.01E − 06)(343.83)3
= 192 𝑘𝐽/𝑘𝑚𝑜𝑙𝐾
𝐶𝑝𝑆 = (6.67E + 01) + (8.41E − 01)(343.83 ) + (−2.16E − 03)(343.83 )2 + (2.33E − 06)(343.83 )3
= 165 𝑘𝐽/𝑘𝑚𝑜𝑙𝐾
∴ 𝐶𝑝𝐹 = 𝑧𝐹 × 𝐶𝑝𝐸𝐵 + (1 − 𝑧𝐹 ) × 𝐶𝑝𝑆
= 0.132 × 192 + (1 − 0.132) × 165 = 168 𝑘𝐽/𝑘𝑚𝑜𝑙𝐾
Latent heat:
The formula for the latent heat of vaporisation as a function of temperature was taken from Yaws
(1999:109). The formula and coefficients used in the formula are given below.
𝑇
𝜆 = 𝐴(1 − 𝑇 )𝑛
𝐶
Table 5.16: Coefficients for latent heat formula

Species A Tc (K) n Tmin (K) Tmax (K) Tb (K)
EB 54.8 617 0.388 178 617 409
Styrene 65.3 648 0.558 350 648 418
343.83 0.388
𝜆𝐸𝐵 = 54.8(1 − ) = 39.93 𝑘𝐽/𝑚𝑜𝑙
617
343.83 0.558
𝜆𝑆 = 65.3(1 − 648
) = 42.82 𝑘𝐽. 𝑚𝑜𝑙
𝜆𝐹 = 𝑧𝐹 × 𝜆𝐸𝐵 + (1 − 𝑧𝐹 ) × 𝜆𝑆 = 0.132 × 39.93 + (1 − 0.132) × 42.82 = 37.178.993 𝑘𝐽/𝑚𝑜𝑙

𝑘𝐽 1000𝑚𝑜𝑙 𝑘𝐽
= 37.178.993 𝑚𝑜𝑙 × 𝑘𝑚𝑜𝑙
= 37178.99 𝑘𝑚𝑜𝑙
Value of q
𝜆𝐹 +𝐶𝑝𝐹 (𝑡𝐵𝑃 −𝑡𝐹 ) 37178.99 + 168(365−323.15)
∴𝑞= 𝜆𝐹
= 37178.99
= 1.19
Point on the q-line

𝑞 𝐹 𝑧 1.19 0.132
𝑦 = 𝑞−1 𝑥 − 𝑞−1 = 1.19−1 (0.17) − 1.19−1 = 0.371
Number of Theoretical Stages

Theoretical Stages at stripping section:
𝐾′ 𝑥′
( 𝑆 ′ − 1) (𝑥 ′ 𝑟 − 1)
𝐵
log [ 1 ′
]
(𝐾 − 1)
𝑆′
𝑁𝑆 = +1
𝐾′
log ( 𝑆 ′ )
0.114−0
K' (slope of stripping section from xB to 0) = 0.084−0 = 0.1357
0.164−0.0917
S’= = 1.6
0.137−0.017
𝑥 ′ 𝑟 = 0.0917
𝑥′ 𝐵 = 0.11
118
0.1357 0.0917
( −1)( −1)
1.6 0.11 ]
log[ 1
(0.1357−1)
1.6
𝑁𝑆 = 0.1357 + 1 = 14
log( )
1.6
Total number of stages required for desired separation : 14 + 18 = 28
Column Diameter and Pressure Drop

Molar Flow rates
Top of column: L = D × R = 26.73 × 18.71 = 499.99 𝑘𝑚𝑜𝑙/ℎ𝑟
𝑉 = D × (R + 1) = 26.73 × (18.71 + 1) = 526.72 𝑘𝑚𝑜𝑙/ℎ𝑟
𝑘𝑚𝑜𝑙
Bottom of tower: 𝐿̅ = 𝐿 + (𝐹 × 𝑞) = 499.99 + (387.56 × 1.19) = 961.06 ℎ𝑟
𝑘𝑚𝑜𝑙
𝑉̅ = 𝑉 + (𝐹 × (𝑞 − 1)) = 526.72 + (387.56 × (1.19 − 1)) = 600.23
ℎ𝑟
Tower Diameter at the Top

Operating pressure: P = 0.2 bar
Operating temperature: 360 K from T-x-y diagram at y = xD =0.67
𝐿𝑤 𝐿 499.99
𝑉𝑤
= 𝑉 = 526.72 = 0.949
𝑛
𝑇
−(1− )
Liquid density: 𝜌 = 𝐴𝐵 𝑇𝐶 [Unit: kg/l] (Yaws, 1999:185)
Table 5.17: Coefficients for liquid density formula
A B N TC Tmin Tmax 𝜌 𝑎𝑡 25𝑜 𝐶
Styrene 0.29383 0.26315 0.28570 648 242.54 648 0.9
Ethylbenzene 0.28889 0.26438 0.29210 617.17 178.2 617.17 0.865
360 0.28570
𝜌𝐸𝐵 = 0.28889 × 0.26315−(1−648) = 809.35 kg/m3
1 1
Mixture of ethylbenzene and styrene: 𝜌𝑙 = 𝑥𝐷 1−𝑥𝐷 = 0.67 1−0.67 = 821.39 𝑘𝑔/𝑚3
+ +
𝜌𝐸𝐵 𝜌𝑠 809.35 847.21
Vapour density:
𝑃𝑉 = 𝑛𝑅𝑇
𝑛 𝑃
=
𝑉 𝑅𝑇
𝑚 𝑃𝑀𝑎𝑣𝑔
∴𝜌= 𝑉
= 𝑅𝑇
𝑀𝑎𝑣𝑔 = (𝑥𝐷 × 𝑀𝑟(𝐸𝐵) ) + (1 − 𝑥𝐷 × 𝑀𝑟(𝐸𝐵) ) = (0.67 × 106.18) + (1 − 0.67 × 104.16) = 105.52 𝑔/𝑚𝑜𝑙
20 × 105.52 𝑘𝑔
𝜌𝑣 = = 0.071 3
8.143 × 360 𝑚
Table 5.18: Density data summary for top of column
Compound Liquid density (kg/m3) Vapour density (kg/m3)
Styrene 847.21 --
Ethylbenzene 809.39 --
119
Mixture 821.39 0.071
Liquid-vapour flow factor

𝐿 𝜌 0.5 0.071 0.5
𝐹𝐿𝑉 = 𝑉𝑤 (𝜌𝑣 ) = 0.949 × (821.39) = 8.79 × 10−3 𝑚/𝑠
𝑤 𝐿
Assuming tray spacing of 0.5 m, using Figure 6, Cf = 0.1.
Surface tension:
𝑇 𝑛
𝜎 = 𝐴 (1 − 𝑇 ) [Unit: dynes/cm] (Yaws, 1999:212)
𝐶
Table 5.19: Coefficients for surface tension formula

Compound A Tc n Tmin Tmax 𝜎𝑎𝑡 25𝑂 𝐶
Ethylbenzene 66 617.17 1.268 178.15 617.17 28.59
Styrene 68.178 648 1.2222 273.15 648 32.1
Applying the formula, the following values were obtained:
Table 5.20: Surface tension data
Compound Surface tension Surface tension
(dyne/cm) (N/m)
Ethylbenzene 21.949 0.0219

Styrene 25.504 0.0255
Mixture 23.114 0.0255
Flooding Velocity:
𝜎 0.2 𝜌 −𝜌 0.5 0.0255 0.2 821.39−0.071 0.5
𝐿 𝑣
𝑉𝐹 = 𝐶𝑓 ( ) ( ) = 0.1 ( ) ( ) = 11.33 𝑚/𝑠
0.02 𝜌𝑣 0.02 0.071
Actual Vapour Velocity through Tower:

𝑣 = 0.85 × 𝑉𝐹 = 0.85 × 11.33 = 9.63𝑚/𝑠
Volumetric flow rate:

𝑉𝑀 526.72×105.52
𝑄̇ = 𝜌 𝑎𝑣 = 0.071
= 1438239.5 𝑚 3 /ℎ𝑟 = 399.51 𝑚 3 /𝑠
𝑣
Area
𝑄̇ 399.51
Net area: 𝐴𝑛 = 𝑣
= 8.63
41.48 𝑚 2
Downcomer area: 𝐴𝑑 = 𝑑𝐴𝑐 and 𝐴𝑛 = 𝐴𝑐 − 𝐴𝑑

Let d = 0.12 (d range: 0.08 – 0.12)
𝐴𝑛 41.48
∴ 𝐴𝑐 = = = 47.14 𝑚
𝑑 0.12
120
Inside Column diameter at the top:
4×𝐴𝐶 4×47,14
𝐷𝑖 = √ 𝜋
=√ 𝜋
= 7.75 𝑚
Pressure Drop
Operating pressure at column base = Operating pressure at top + Pressure drop in tower
Operating pressure at top = 0.2 bar = 20 kPa
Pressure drop in tower = ∆𝑃𝑇 = 𝑁𝑎 × 𝜌𝐸𝐵 × 𝑔 × ℎ𝑡
𝑡ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 28.14
𝑁𝑎 = 𝑡𝑟𝑎𝑦 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = 0.75
= 37.66 𝑡𝑟𝑎𝑦𝑠
∴ ∆𝑃𝑇 = 37.6 × 809.36 × 9.81 × 15.49 = 4632.77 𝑃𝑎 = 4.63 𝑘𝑃𝑎

∴ Operating pressure at column base ∶ 𝑃𝑏 = 𝑃 + ∆𝑃𝑇 = 20 + 4.63 = 24.63 𝑘𝑃𝑎 = 0.046 𝑏𝑎𝑟
Operating temperature at column base

When 𝑃𝑣 = 𝑃𝑏 then 𝑇 = 𝑇𝑏
Table 5.21: Coefficients for Antoine’s Equation (Yaws, 1999:65)
Compound A B C Tmin (C) Tmax (C)
S 6.94579 1443.88 210.039 -30.61 374.85
styrene 6.9571 1445.58 209.44 32 187
𝐵
Using Antoine’s equation: 𝑙𝑜𝑔𝑃𝑣 = 𝐴 − 𝑇+𝐶 and plugging in the coefficients above and 𝑃𝑣 = 𝑃𝑏 :
𝑇𝑏 = 371.9 K
Using these calculated operating conditions, for the bottom of the column, the same procedure for
finding the column diameter at the top is applied at the column base. These calculations yield the
following results:
Table 5.22: Results summary for column diameter at base calculation
Liquid density 836.52 kg/m3
Vapour density 0.083 kg/m3
Surface tension 0.024 N/m
Liquid-vapour flow factor at bottom 0.0159
Cf 0.079
Flooding velocity 8.23 m/s
Actual velocity 6.99 m/s
Volumetric flow rate 209.29 m3/s
Net area required at bottom 29.92 m
Inner cross-sectional area of column 34 m
Inside column diameter at base of column 6.58 m
121
Appendix F
Fluid Properties
Density
Table 6.11: Coefficients for density formula (Yaws, 1999)
A B n TC Tmin Tmax
EB 0.2889 0.2644 0.2921 617.17 178.2 617.17 0.865
𝑇 𝑛
−(1− )
𝜌 = 𝐴𝐵 𝑇𝐶
𝑇(303.15−273.15) 0.2921
−(1− ) 𝑔
𝜌𝐸𝐵 = (0.2889)(0.2644) 617.17 = 0.861 𝑚𝑙
𝑔 𝑘𝑔 106 𝑚𝑙 𝑘𝑔
0.861 × × = 861
𝑚𝑙 103 𝑔 𝑚3 𝑚3
Viscosity
Table 6.12: Coefficients for viscosity formula (Yaws, 1999)
A B C D Tmin (K) Tmax (K) uliq@25C uliq@Tmax
EB -5.26 831 0.0108 -1.06E-05 210 617 0.629 0.05
𝐵
log10 𝜇𝑙𝑖𝑞 = 𝐴 + + 𝐶𝑇 + 𝐷𝑇 2
𝑇
𝐵 831
(𝐴+ +𝐶𝑇+𝐷𝑇2 ) (−5.26+ +(0.0108)(303.15)+(−1.06E−05)(303.15)2 )
𝜇𝑙𝑖𝑞 = 10 𝑇 = 10 303.15 = 0.596 𝑐𝑃
𝑘𝑔
0.596 𝑐𝑃 ÷ 1000 = 0.0006 𝑚3
Vapour Pressure
Table 6.13: Coefficients for vapour pressure formula (Yaws, 1999)
A B C D E Temp. range (K)
EB 36.2 -3340.2 -9.80 -1.18E-11 2.58E-06 178.15 – 617.17
𝐵
log10 𝑃 = 𝐴 + 𝑇 + 𝐶 log10 𝑇 + 𝐷𝑇 + 𝐸𝑇 2
𝐵 2) −3340.2
+(−9.80) log10 𝑇+(−1.18E−11)𝑇+(2.58E−06)𝑇2 )
𝑃 = 10(𝐴+𝑇+𝐶 log10 𝑇+𝐷𝑇+𝐸𝑇 = 10(36.2+ 𝑇
𝑃𝑎
𝑃 = 12.7 𝑚𝑚𝐻𝑔 × 133 = 1699 𝑃𝑎 = 1.7 𝑘𝑃𝑎
𝑚𝑚𝐻𝑔
Flow Rates and Pipe Diameter

𝑛̇ 𝑘𝑚𝑜𝑙 𝑘𝑔 𝑘𝑔 𝑘𝑔
Mass flow rate: 𝐺 = 𝑀𝑟 = 350 ℎ𝑟
× 106.18 𝑘𝑚𝑜𝑙 = 37163 ℎ𝑟 = 10.3 𝑠
𝐺 10.3 𝑚3
Volumetric flow rate: 𝑄̇ = 𝜌 = 861
= 0.012 𝑠
0.53 −0.37
𝑑𝑜𝑝𝑡𝑖𝑚𝑢𝑚 = 293𝐺 𝜌 = 293𝐺(10.32)0.53 (861)−0.37 = 82.8 𝑚𝑚 = 0.083 𝑚
𝜋𝑑 𝜋(0.083)
𝐶𝑟𝑜𝑠𝑠 𝑠𝑒𝑐𝑡𝑖𝑜𝑛𝑎𝑙 𝑎𝑟𝑒𝑎: 𝐴𝐶 = 4
= 4
= 0.00539 𝑚 2
𝑄̇ 0.012 𝑚
𝑉𝑒𝑙𝑜𝑐𝑖𝑡𝑦: 𝑣 = 𝐴 = 0.00539 = 2.22 𝑠
𝐶
122
NPSH Calculation:
𝑃𝑎𝑡𝑚 −𝑃𝑣
𝑁𝑃𝑆𝐻 = 𝜌𝑔
+ (𝑧1 − 𝑧2) − ℎ𝐿
𝑃𝑎𝑡𝑚 −𝑃𝑣 101325−1699.46
𝜌𝑔
= (861.01)(9.81)
= 11.8 𝑚
𝑧1 − 𝑧2 = 14 − 5 = 9𝑚
∑ 𝐾 𝑣2 4𝑓𝐿𝑣 2 29.28 (2.22)2 4(0.003)(76.2)(2.22)2
ℎ𝐿 = ℎ𝑓 + ℎ𝑙 = 2𝑔
+ 𝐷𝑔
= 2(9.81)
+ (0.083)(9,81)
= 7.39 𝑚
∴ 𝑁𝑃𝑆𝐻 = 11.8 + 9 + 7.39 = 13.4 𝑚
Pump Power
𝜌𝑔 ∑ 𝐻𝐷 𝑄
𝑃= ƞ
(861.03)(9.81)(25.06)(0.008)
𝑃= (0.61)
= 2602.68 𝑊 = 2.6 𝑘𝑊
2.6
Actual Power = 0.61 = 4.26 𝑘𝑊
123

INDIVIDUAL REPORT - Keshrie Reddy1 PDF

Uploaded by

Copyright:

Available Formats

INDIVIDUAL REPORT - Keshrie Reddy1 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

INDIVIDUAL REPORT - Keshrie Reddy1 PDF

Uploaded by

Copyright:

Available Formats

Faculty of Engineering & the Built Environment

Department of Chemical Engineering

Project Title: Production of Styrene from Ethylbenzene

Course: BTech: Chemical Engineering

Module: PQD400S Chemical Engineering Process Design 4

Name and Surname: Keshrie Reddy

Student number: 216282772

Due Date: 9 September 2019

Date: ____________ Signed: ____________

2.1. Introduction .......................................................................................................................... 14

1.1. Problem statement

1.3.1 Calculated Data

Figure 1.1: Relationship between Fractional Selectivity to Toluene and Temperature

0.4 Temperature: 900K

1.3.2. Material Balance

-25000000 Pressure: 1.0 bar

Table 1.2: Key for conditions of columns for tables below

Gross Profit Calculation

Three Phase Seperator

Fresh Plant Feed

2.1.1. Feed preparation to the reactor

2.1.2. Ejector and coolers

2.1.3. Flash Separation

2.1.4. Distillation Columns

3.2. Process Flow Diagram

Figure 3.1: PFD of E-501

3.3. Design of Heat Exchanger

𝑄𝑐 = 𝑚̇ 𝐶𝑝∆𝑇 = 𝑛̇ 𝐶𝑝∆𝑇 (Equation 1) (Incropera, 2007:17)

Species A B C D Tmin (K) Tmax

Species A B C D E Tmin Tmax

Table 3.3: Coefficients for HVAP formula

Species A Tc (K) N Tmin (K) Tmax (K) Tb (K)

For the hot fluid (hps), 𝑄ℎ = 𝑚̇ 𝐶𝑝∆𝑇

3.3.2. Physical Properties of Fluids

Table 3.4: Physical properties of components in cold stream

Cold Stream inlet Mean outlet unit

Table 3.5: Physical properties of components in hot stream

Hot stream inlet Unit

As mentioned above, Tho is calculated by equating Qc to Qh:

113 = 2(kg/s) x 1.991(kJ/kgK) x (526.79 –Tho)K

3.3.3. Assume Value of Overall Coefficient Uo,assume

3.3.4. Decide number of shell and tube passes.

Figure 3.2: Temperature profile

∆𝑇2 = 𝑇ℎ𝑜 − 𝑇𝑐𝑖 = 498.41 − 297 = 201.41 𝐾

∴ 𝐹𝑇 = 0.9 (using the chart attached in Appendix C)

3.3.5. Determine heat transfer area required

Table 3.6: Summary of tube specifications

Tube pattern Triangular

3.3.7. Calculate number of tubes

Surface area if 1 tube: 𝐴𝑠 = 𝜋𝐷𝐿 = 𝜋(0.016)(1.83) = 0.092 𝑚 2

Therefore for 2 passes, tubes per pass = 30/2 = 15 tubes

Tube side velocity:

Area per pass: 0.000201 × 15 = 0.0452 𝑚 2

3.3.8. Calculate shell diameter

Finally, this gives a shell diameter of: 𝐷𝑠 = 𝐷𝑏 + 𝐷𝑐 = 140.3 mm + 90 mm = 230.3 mm = 0.23 m

3.3.9. Estimate tube-side heat transfer coefficient

Pr = 0.9955 (Incropera, 2007:945)

L/di = 1.83/(0.016-0.0016) =127.08

𝑁𝑢 = 𝐽𝑛. 𝑅𝑒. 𝑃𝑟 0.33

Date: Signed: