Episode 47 : CONCEPTUAL DESIGN

OF CHEMICAL PROCESSES

SAJJAD KHUDHUR ABBAS

Ceo , Founder & Head of SHacademy
Chemical Engineering , Al-Muthanna University, Iraq
Oil & Gas Safety and Health Professional – OSHACADEMY
Trainer of Trainers (TOT) - Canadian Center of Human
Development
 INTRODUCTION
 Chemical process design is the application of chemical
engineering knowledge (chemical, physical and/or biological
transformations of raw materials) into products and economics in
the conceiving a chemical process plant to profitably
manufacture chemicals in a reliable and safe manner without
unduly affecting adversely the environment and society
 Chemical process plants are by nature large capital investment
projects that
 are expensive to build and operate
 have very long life times and
 manufacture specific chemicals
 Chemical process plants must be designed well to avoid large
financial losses over long periods of times due to inefficient
processes/poor operations
 INTRODUCTION
 Main design objectives of chemical processes:
 design of a grassroot plant or
 a retrofit design for existing chemical plants
 Complimentary objectives
 profitable, safe, reliable, flexible, controllable
and operable
 Not all of these objectives can be fulfilled
however and some trade offs must be made in
order to produce a practical design
 UNIT OPERATIONS
 In the past, chemical process plants are
designed using unit operations first
proposed by G.E. Davis in 1887
 Unit operations was formalised by A.D.
Little in 1915 as the defining principle of
chemical engineering
 The concept was earlier proposed by the
ancient alchemists, in the course of
transforming and purifying their chemicals
through a series of operations of heating,
distillation, evaporation etc.
 UNIT OPERATIONS
 New chemical process plants were then
designed by
 arranging the unit operations in the same
sequence as the original laboratory
methods
 increasing the size of equipment linearly
for greater capacity
 In the 40’s, it was realised that scaling-up
is not linear and pilot plant studies
needed to be done in order to determine
the correct scaling-up parameters
 UNIT OPERATIONS
 Up to the late 70’s, chemical process design was
still done by
 arranging unit operations in the sequence
proposed by the industrial chemists using block
diagrams and later PFDs
 performing the mass and energy balance

 sizing the individual equipment

 determining the economic viability of the plant

 Alternative PFDs were not easily generated due to

 the empirical nature of the chemical technology
 the large number of uncertain variables to be
determined all at once
 UNIT OPERATIONS
 Design parameters were determined in ad
hoc manner & specific for particular process
 No systematic method for generating
alternative PFDs and optimising them
 Short cut methods of designing heat and
mass transfer equipment already available
 Equipment costing methods have been fairly
developed using costing charts
 Possible integration and optimisation of unit
operations due to interconnections within the
chemical process system was not
understood
 PROCESS SIMULATION
 With powerful computers and better
understanding of thermodynamics in the late
60’s to early 80’s, computational and
optimisation methods were used in process
system engineering
 Since the 60’s, primitive process simulation
softwares were owned by large petrochemical
companies
 These were mainly the sequential modular
type where the unit operation modules were
solved one by one in the direction of mass
flow
 PROCESS SIMULATION
 Modular simulation consists of
 a top level of flowsheet topology where unit module are
sequenced, recycle and tear streams determined, and
convergence made,
 a middle level where the unit operations are modeled and
solved and
 a lower level where physical and thermodynamic models are
solved
 By the late 70’s, the solution of modular flowsheets
was significantly improved leading to simultaneous
modular flowsheets which are the basis of commercial
process simulation softwares such as
 ASPEN/PLUS from Aspen Technology Inc. and
 HYSYS from Hyprotech Ltd
 PROCESS SIMULATION
 Most process simulations use phase equilibrium
thermodynamic models including non-idealities in
both liquid and gas phases for their unit operation
models
 Popular models used are the equation of state
models for hydrocarbon mixtures and liquid
activity coefficients models for non-electrolyte,
non-ideal solutions
 Group contribution models such as UNIFAC are
becoming popular when no empirical vapour-liquid
equilibrium data is available
 Rate-based models are very well developed and
may well become more important when tray
efficiency could not account for non-ideal
 PROCESS SIMULATION

 In the 90’s, stoichiometric and equilibrium

reactor models are primitive with poor
handling of multiple reactions in
completely mixed and plug flow reactors
 Incorporation of rigorous generic models
for multi-phase industrial reactors is still a
long way off
 Some process simulator companies do
model these reactors for individual process
licence owners
 PROCESS SIMULATION
 The generic modeling of adsorption, membrane
and solid drying processes are not well developed
enough to be included in process simulations
 A shortcut method for the generic design of
adsorption columns presented by Wan Ramli Wan
Daud 2000b shows some promise
 Solids handling was neglected in process
simulation work
 It is now more important due to the increased
popularity of fluidised bed reactors and pneumatic
conveying
 PROCESS SIMULATION
 In the 80’s and 90’s significant improvement was
made in the equation-oriented process simulation
where the equations for all unit operations are
combined and solved simultaneously
 Allows specifications of certain design parameters
without having to solve another iterative loop
 The computational effort is reduced by the
exploitation of sparse matrices
 Succesful solution requires careful initialisation
based on users’ past experience
 It is used in quick on-line real time modelling and
optimisation where models are simpler and initial
points are taken from previous solutions
 PROCESS SIMULATION
 Both simulations require simultaneous solution of
large sets of non-linear equations which are mainly
based on Newton or quasi-Newton or Broyden
methods due to their good convergence properties
 Rapid solution of very large flowsheets can be
achieved by a suitable decomposition strategy
 by recycle tearing streams for the modular simulation
 by utilising powerful sparse matrix solvers for equation
oriented simulation
 Although process simulation is a powerful tool, it
is not possible to produce optimised design by
simply using it because the optimum configuration
and operating principle of the process plant could
only be produced by process synthesis
 PROCESS SYNTHESIS
 Contemporary process design method is an
iterative problem solving and optimisation method
using both heuristic and algorithmic methods
 Design method begins with the determination of
the design requirements and objectives which are
promulgated in either an economic or utlitarian
way
 A conceptual design is then produced through the
synthesis of several feasible alternative designs
and the rapid selection of the most viable of these
alternatives based on an economic performance
criterion without using rigorous performance
models of their operational principles
PROCESS SYNTHESIS
 PROCESS SYNTHESIS
 During synthesis, design variables or parameters
are selected or determined and optimised through
 Heuristics, intuitions and experience or
 algorithmic methods using shortcut performance models
of the equipment or chemical process
 Complex design problems are decomposed into
their constituent parts where
 each part is further synthesised,
 its performance is modelled on its operational principle
and
 its design variables or parameters are determined in a
similar manner
 while maintaining integral relationship with other
parts as well as with the overall design
 PROCESS SYNTHESIS
 First approach: Process synthesis can be solved by
mathematical modelling alone based on the
principles of process flowsheeting
 Assumes that technology emerges from science
linearly which is not true because scientific
knowledge on the physical phenomena in an
engineering artefact does not lead to knowledge on
the operating principles and design of the artefact
 The chemical process plant has a large number of
variables that are defined by a smaller number of
equations, with some inexplicable to deterministic
models and most highly non-linear
 Able to synthesise small plants where variables are
defined adequately by equal number of equations
unless efficient decomposition procedures are used
PROCESS SYNTHESIS
 PROCESS SYNTHESIS
 Second approach: Process synthesis could be
solved by expert knowledge obtained from
experience, intuition/insight and inspirations
 Expressed as heuristic rules/rules of thumbs
which set unknown parameters rapidly
 Some heuristics relate external performance
parameters with the operating variables of the
artefact simply and directly without complex
non-linear mathematical modelling
 PROCESS SYNTHESIS
 The synthesis problem is decomposed into an
heirarchy of decisions for generating and
exploring process alternatives starting from the
top down and considering a few design variables
at a time like peeling an onion
 Basic assumption : design parameters at the top
level also reflect design parameters further down
 Old alchemical maxim of the relationship between
the macrocosmos and the microcosmos: “what is
above so below”
 HEIRACHICAL PROCESS SYNTHESIS
First & Second Levels
 In Douglas version, after decomposition by
removing all the heat exchangers, the first level
involves use of heuristics to select
 Process Mode:
 Design Variables: Batch or continuous
 The second level involves construction of the
input-output structure of the process and targeting
the production rate by using heuristics:
 Whether the feed should be pretreated
 Destination of products
 Design variables : conversion of limiting reactant and
allowable purge concentration of excess reactant
 Economic potential of process:
Products sales less raw materials’ cost
HEIRACHICAL PROCESS SYNTHESIS
First & Second Levels
HEIRACHICAL PROCESS SYNTHESIS
Input-Output Structure
 HEIRACHICAL PROCESS SYNTHESIS
Input-Output
Toluene Structure:
Hydro-Dealkylation Process Destination Streams
Purge H , CH
2 4

H , CH Toluene Hydro- Benzene

2 4
Dealkylation Diphenyl
Toluene Process

Component Normal Boiling Light/Heavy Destination

Point(C)
Hydrogen -253 Light Recycle dan Purge
Methane -161 Light Recycle dan Purge
Benzene 80 Heavy Main Product
Toluene 111 Heavy Recycle
Diphenyl 253 Heavy Fuel
 HEIRACHICAL PROCESS SYNTHESIS
Input-Output
Toluene Structure:
Hydro-Dealkylation Process Material
P  n  2n Balance B 1 2

FT  n1
RG PG FH  FE  n1  n2

FG Toluene Hydro- PB PD  n2
Dealkylation
FT Process
PD FT  PB S B
PM  FM  n1

n1  PB S B
PM  1  y PH PG  1  y FH FG  PB S B 1 S B 
PG  FE  1  y FH FG  PB S B n2  PB
2S B
y PH  FE PG
PB 1 S B  1 S B  P
1 S B  P FH  y FH FG  FE   PB  FE  B
PG  FG  B
SB 2S B 2S B
2S B
1 1 y 1 S  2
FG  PH B PB
y FH  y PH S B
 HEIRACHICAL PROCESS SYNTHESIS
Input-Output
Toluene Structure:
Hydro-Dealkylation Process Material Balance
RG PG

PB = 265 kgmole h-1 benzene

FG Toluene Hydro- PB
Dealkylation
FT Process
PD

XB FG FT PH PM PD
(kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1)
0.1 312.49 266.13 31.31 281.75 0.56
0.2 312.64 266.35 31.33 281.99 0.68
0.3 312.84 266.67 31.37 282.31 0.83
0.4 313.13 267.12 31.42 282.77 1.06
0.5 313.58 267.81 31.50 283.49 1.41
0.6 314.34 268.99 31.63 284.70 1.99
0.7 315.82 271.27 31.90 287.06 3.13
0.8 319.51 276.97 32.55 292.94 5.98
0.9 336.48 303.20 35.56 320.02 19.10
 HEIRACHICAL PROCESS SYNTHESIS
Second Level Economic Potential
Toluene Hydro-Dealkylation Process

30000000

Excess Reactant
h 25000000
Economic Potential (RM)/Year
n) Concentration in
t
/ Purge Stream
M 20000000
R
(i
yph=0.1
15000000
yph=0.5
o yph=0.65
n 10000000
o yph=0.786
m
k
E 5000000
i
n
s 0
e 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
ot
P -5000000
Penukaran
Conversion Toluena
of Limiting Reactant

f PE 2  C B PB  C FD PD  C FP PG  CT FT  C H FG
 HEIRACHICAL PROCESS SYNTHESIS
Input-Output
Benzene Structure: Destination Streams
Alkylation Process
Propane and Propylene
As Fuel

Propylene Cumene

Benzene Benzene Alkylation

Process
P-diisopropyl Benzene
As Fuel
Benzene Recycle

Component Normal Boiling Light/Heavy Destination

Point (C)
C3H8 -42.1 Light Fuel
C3H6 -47.8 Light Fuel
C6H6 80.1 Heavy Recycle
C9H12 152.4 Heavy Main Product
C12H18 210.3 Heavy Fuel
 HEIRACHICAL PROCESS SYNTHESIS
Input-Output
Benzene Structure: Material
Alkylation Process P  n n
Balance
C 1 2

PG FP  n1  n2  y PP PG
FG PC FB  n1

FB Benzene Alkylation
PD  n2
PD
Process FP  y FP FG  PC S C  y PP PG
RB
1 S P
n1  C C
2S C
1 S P
n2  C C

FB 
1 S C PC 2S C
2S C y F Pr FG  y P Pr PG
1 S P
PD  C C
2S C
PC
FG 
S C y FP  y PP 1 y FP  1 y PP 
 HEIRACHICAL PROCESS SYNTHESIS
Input-Output
Benzene Structure: Material Balance
Alkylation Process
PG

FG PC PC = 104 kgmole h-1 cumene

FB Benzene Alkylation PD
Process

RB

XB FG PD FB PG
(kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1)
0.1 113.04 0.00 103.99 9.04
0.2 113.10 0.03 104.02 9.05
0.3 113.17 0.06 104.05 9.05
0.4 113.24 0.10 104.09 9.06
0.5 113.33 0.13 104.13 9.07
0.6 113.41 0.17 104.17 9.07
0.7 113.51 0.22 104.21 9.08
0.8 113.61 0.27 104.26 9.09
0.9 113.73 0.32 104.31 9.10
1.0 135.75 10.45 114.44 10.86
 HEIRACHICAL PROCESS SYNTHESIS
Second Level Economic Potential
Benzene Alkylation Process
45000000
40000000 Excess Reactant
Economic Potential (RM)/Year 35000000
Concentration in
Purge Stream
30000000
25000000 ypp=0.1
ypp=0.5
20000000 ypp=0.7
15000000 ypp=0.898

10000000
5000000
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
-5000000
Conversion of Limiting Reactant

f PE 2  CC PC  CFDIPB PDIPB  CFP PP  CP FP  C B FB

 HEIRACHICAL PROCESS SYNTHESIS
Third Level
 The third level involves the construction of the
reactor and recycle structures of the process by
using heuristics to decide on
 the number of reactor systems required
 their types (completely mixed or plug flow)
 operating modes and conditions and
 heat management
 number of recycle streams
 whether a gas recycle is required, and
 recycle flow rates as functions of conversion and mole or
recycle ratio
 Annual costs of reactors & compressors are
subtracted from economic potential at this level
HEIRACHICAL PROCESS SYNTHESIS
Third Level
HEIRACHICAL PROCESS SYNTHESIS
Recycle Structure
 HEIRACHICAL PROCESS SYNTHESIS
Recycle Structure  F 
Toluene Hydro-Dealkylation Process yFH FG  y PH RG  MR  FT 
 XT 
Purge P M y 1 1 yPH 1SB  2
RG  B  R  FH 
yFH  yPH 
Compressor H , CH
2 4 SB yPH  XT 
Vapour Recycle RG RT  1 XT FT
H , CH
2 4
Hydrogen Feed
Benzene Product
Separation
Toluene Feed Reactor & Purification
Diphenyl Product
System

Toluene Recycle RT
Component Normal Boiling Point (C) Light/Heavy Destination
Hydrogen -253 Light Recycle dan Purge
Methane -161 Light Recycle dan Purge
Benzene 80 Heavy Main Product
Toluene 111 Heavy Recycle
Diphenyl 253 Heavy Fuel
 HEIRACHICAL PROCESS SYNTHESIS
Recycle Structure
Benzene Alkylation Process
ylene
Propane & Prop
As Fuel

Propylene Reactor Separation Cumene

& Purification P-diisopropyl Benze ne
Benzene System As Fuel

Recycle RB
Benzene
RB  1 X P MR yPF FG

Component Normal Boiling Light/Heavy Destination

Point (C)
C3H8 -42.1 Light Fuel
C3H6 -47.8 Light Fuel
C6H6 80.1 Heavy Recycle
C9H12 152.4 Heavy Main Product
C12H18 210.3 Heavy Fuel
 HEIRACHICAL PROCESS SYNTHESIS
Adiabatic Temperature
• Energy Balance for Reactors N M K

n H Pc T T  F c T
j
m
rj
i1
i pi a m
k1
k pk m Tm 0
j1

• Adiabatic temperature N M N M

n H  Pc T T 0 Ta Tm njH Pc

m m
j rj i pi a m rj i pi
j1 i1 j1 i1
• For simple reaction A  B,
N
o  
 FA1 X A cPA FAX AcpB  T 25 
 j r  FAX AHr  FAcpA
M

Pc  FA1 X A cPA  FAX AcpB H  m

m
n
i1
i pi
j1  

Ta Tm 

X A H r  cpB cpA Tm 25
o

X Ac pB cpA cpA
cpATa Tm 
• The adiabatic coversion X Aa  
H  c
o
r pB cpA Ta 25 
M N
• In general i pi Ta 25
Xa Fc n H
a
j rj
i1 j1
 HEIRACHICAL PROCESS SYNTHESIS
Adiabatic Temperature
Toluene Hydro-Dealkylation Process
1800

k
it 1600

Adiabatic Temperature
a
b Molar Ratio
a
i
d 1400 MR=1
A
u
h MR=2
u 1200
S MR=3
MR=4
1000

800

600
0.1 0.2 0.30.4 0.5 0.6 0.7 0.8 0.9 1
Penukaran
Conversion Toluena
of Limiting Reactant
HEIRACHICAL PROCESS SYNTHESIS
Reactor Heat Management

1.0

XA

r=1
r = 10
r = 100
0
T

Isothermal Reactor for Non-Autocatalytic Irreversible Reaction

 HEIRACHICAL PROCESS SYNTHESIS
Reactor Heat Management

1.0 1.0

XA XA

r=1 r=1
r = 100 r = 100
r = 10 r = 10
0 0
T T
Endothermic Reaction Exothermic Reaction

Adiabatic Reactor for Non-Autocatalytic Irreversible Reaction

HEIRACHICAL PROCESS SYNTHESIS
Reactor Heat Management
Equilbrium Equilibrium

1.0 1.0

r=0 r=0
XA XA

r=1
r=1 r = 10 r = 100 r = 10 r = 100
0 0
T T
Endothermic Reaction Exothermic Reaction

Isothermal Reactor for Single Reversible Reaction

 HEIRACHICAL PROCESS SYNTHESIS
Reactor Heat Management
Equilibrium Adiabatic Curve Equilibrium Adiabatic Curve
1.0 1.0

XAf XAf

XA XA
r=1
r=0 r=0
r = 1 r = 10 r = 100 r = 10 r = 100
0 0
T T
Endothermic Reaction Exothermic Reaction

Adiabatic Reactor for Single Reversible Reaction

Compressor andPROCESS
HEIRACHICAL Reactor Sizing
SYNTHESIS
Toluene Hydro-Dealkylation Process
W  RG
nZRT1
P2 P1  1
n 1
n1 n T2 T1   P2 P1 n1 n

V  XA dX A V  X
   rA
  A
FAo C Ao 0 FAo C Ao  rA

F ln 1 1  X T 
V
k exp( E / RT )  o  y F H FG  y P H R G  F   o
0.5


XT Compressor Reactor Reactor Diameter

(kW) Volume Length (m)
(m3) (m)
0.1 3702.73 307.68 22.87 4.14
0.2 1787.160 314.99 22.87 4.19
0.3 1149.083 324.39 22.87 4.25
0.4 830.590 336.69 22.87 4.33
0.5 640.259 353.34 22.87 4.44
0.6 514.624 376.95 22.87 4.58
0.7 427.371 413.26 25.61 4.53
0.8 368.636 478.94 25.61 4.88
0.9 358.480 668.20 25.61 5.76
 HEIRACHICAL PROCESS SYNTHESIS
Third Level Economic Potential
Toluene Hydro-Dealkylation Process
I  7775.3 
0.82
I MSk W  W
R 2.18  m p  I
K pt  8.742.11 Fd    Cp 8150 
K rt   MSk   D1.066 L0.82 F F F
3I MSd     I MSd  3 

Material of Carbon Steel Carbon steel chromium- Stainless Steel

Construction molybdenum
Fm 1.00 2.15 3.75
Pressure 1.6 6.8 13.6 20.4 27.2 34.0 40.8 47.6 54.4 61.2 68.0
(Bar)

FP 1.00 1.05 1.15 1.20 1.35 1.45 1.6 1.8 1.9 2.3 2.5

Compressor Fd
Centrifugal compressor with electric motor 1.0
Centrifugal compressor with turbine 1.15
Reciprocating compressor with steam 1.07
Reciprocating compressor with electric motor 1.29
Reciprocating compressor with engine 1.82
 HEIRACHICAL PROCESS SYNTHESIS
Third Level Economic Potential
Toluene Hydro-Dealkylation Process

30000000

3 20000000
Economic Potential (RM)/Year

a
s Molar Ratio
r 10000000
A
i
m MR=2
o 0
n MR=3
k MR=4
o -10000000
E MR=5
si
e -20000000
n
t
o
P -30000000

-40000000
0 0.1 0.2 0.3
0.4 0.5 0.6 0.7 0.8 0.9 1
Penukaran
Conversion of Limiting Reactant

f PE 3  C B PB  C FD PD  C FP PG  CT FT  C H FG  K pt  K rt
 HEIRACHICAL PROCESS SYNTHESIS
Third Level Economic Potential
Benzene Alkylation Process
50000000

40000000
Economic Potential (RM)/Year

Molar Ratio
30000000
ypp=0.1
ypp=0.5
20000000
ypp=0.7
ypp=0.898
10000000

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

-10000000
Conversion of Limiting Reactant

f PE 2  C C PC  C FDIPB PDIPB  C FP PP  C P FP  C B FB  K rt
 HEIRACHICAL PROCESS SYNTHESIS
Fourth Level
 The fourth level involves the synthesis of the
separation structure of the flow sheet
 Reactor products are to be separated into a liquid
and a vapor phase by cooling, decompressing or
both with the main product in the liquid phase
because liquid purification technology like
distillation can produce very pure product
 The liquid stream is sent to the liquid separation
train consisting usually of a distillation train that
are sequenced using heuristics
 The vapor stream may be purged and the rest of
the vapor recovered &/or recycled, or be
condensed into liquid, which is sent to the liquid
separation train, and the rest of
uncondensible vapor recovered,
recycled or purged
 HEIRACHICAL PROCESS SYNTHESIS
Fourth Level
 Selection of simple or complex columns and order of
distillation columns sequence using heuristics
 Distillation columns design using short cut methods
e.g. Fenske-Underwood-Gilliland (FUG)
 For non-ideal and azeotropic distillation
 Identify azeotropes & alternative separators
 Select entrainers & identify feasible distillate & bottom
products compositions
 Design variables: pressure, temperature & product
recoveries at flash drum, absorbers, adsorbers,
membrane modules & distillation columns
 Annual costs of separation systems are added to the
economic potential at this level
HEIRACHICAL PROCESS SYNTHESIS
Fourth Level
HEIRACHICAL PROCESS SYNTHESIS
Separation Structure
HEIRACHICAL PROCESS SYNTHESIS
Separation Structure: Sub-cooled Liquid
HEIRACHICAL PROCESS SYNTHESIS
Separation Structure:
Both Superheated Vapor & Sub-cooled Liquid
HEIRACHICAL PROCESS SYNTHESIS
Separation Structure: Superheated Vapor
HEIRACHICAL PROCESS SYNTHESIS
Flashing to Separate Liquid and Vapour

 Dew point Ki  1
 xi   yi (3.23)
 Bubble point  yi   K i xi  1

(3.24)
 Flash calculation using Rachford-Rice method

zi   K i xi  1(3.25)
 xi
zi K i zi
xi  yi 
1 K i 1 1  K i 1
C C C
1 K i  zi
f     yi   xi   0
i1 i1 i1 1  K i 1
 HEIRACHICAL PROCESS SYNTHESIS
Flashing to Separate Liquid and Vapour
Toluene Hydro-Dealkylation Process
Reactor Product 3.3 bar dan 35C
Toluene Toluene Methane Hydrogen
Conversion (kgmole h-1) (kgmole h-1) (kgmole h-1)
XT
Top Product
0.1 2395.15 19562.99 13042.75 Toluene Methane Hydrogen Benzene Toluene Diphenyl
0.2 1065.41 9591.88 6395.489 Conversion (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1)
0.3 622.22 6270.52 4181.46 XT 19489.20 12986.6 17.24 51.29 1.8x10-5
0.4 400.67 4612.69 3076.54 0.1
0.5 267.81 3622.01 2416.55 0.2 9555.07 6367.45 16.96 22.43 2.1x10-5
0.6 179.32 2968.12 1981.40 0.3 6246.13 4162.88 16.75 12.93 2.6x10-5
0.7 116.26 2514.08 1680.23 0.4 4594.45 3062.64 16.48 8.19 3.2x10-5
0.8 69.24 2208.71 1480.45 0.5 3607.44 2405.44 16.22 5.38 4.2x10-5
0.9 33.69 2157.52 1463.81 0.6 2955.99 1972.15 15.98 3.55 5.9x10-5
0.7 2503.70 1672.30 15.82 2.28 9.2x10-5
0.8 2199.56 1473.44 15.77 1.35 1.7x10-4
0.9 2149.03 1457.23 16.56 0.69 5.9x10-4
Bottom Product
Toluene Metahne Hydrogen Benzene Toluene Diphenyl
Conversion (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1)
XT
0.1 73.79 56.20 247.76 2343.86 0.5636
0.2 36.81 28.04 248.04 1042.99 0.6766
0.3 24.39 18.58 248.25 609.29 0.8325
0.4 18.25 13.90 248.52 392.49 1.0580
0.5 14.58 11.11 248.78 262.43 1.4057
0.6 12.13 9.25 249.025 175.78 1.9925
0.7 10.39 7.93 249.18 113.98 3.1331
0.8 9.16 7.01 249.235 67.89 5.9827
0.9 8.49 6.58 248.44 33.00 19.0978
 1.75 bar dan 90C (a)

HEIRACHICAL PROCESS SYNTHESIS

Flashing to Separate Liquid and Vapour
Benzene Alkylation Process
1.75 bar dan 90C
Reactor Product
Propylene Propylene Propane Benzene DIPB
Conversion (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1)
XP
0.1 94.121 5.273 1975.874 0
0.2 83.711 5.276 936.505 0.030
0.3 73.292 5.279 590.055 0.062
0.4 62.863 5.283 416.837 0.096 Top Product
0.5 52.423 5.286 312.912 0.133 Propylene P
0.6 41.971 5.290 243.635 0.174 Conversion (kg
0.7 31.505 5.295 194.157 0.218 XP
0.8 21.023 5.300 157.055 0.266 0.1
0.9 10.522 5.305 128.205 0.320 0.2
0.99 1.256 6.333 137.868 10.451 0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.99
Bottom Product
Propylene Propylene Propane Benzene Cumene DIPB
Conversion (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1) (kgmole h-1)
XP
0.1 21.748 1.368 1780.262 102.717 0
0.2 11.622 0.835 777.299 101.639 0.029
0.3 8.551 0.704 472.044 101.134 0.061
0.4 7.136 0.686 331.385 101.047 0.096
0.5 6.289 0.725 251.894 101.221 0.133
0.6 5.653 0.812 200.999 101.557 0.173
0.7 5.134 0.979 166.004 102.037 0.217
0.8 4.617 1.309 140.564 102.632 0.266
0.9 3.762 2.087 121.025 103.300 0.319
0.99 1.184 6.019 137.592 103.970 10.450
HEIRACHICAL PROCESS SYNTHESIS
Separation Structure: Liquid Separation
HEIRACHICAL PROCESS SYNTHESIS
Sequencing of Simple Distillation Columns

2 1, 2 1
1

1, 2, 3 1, 2, 3

2, 3 3 3 2

Direct Sequence Indirect Sequence

Lightest First Heaviest First
HEIRACHICAL PROCESS SYNTHESIS
Sequencing of Complex Columns

1 2 1

1, 2, 3 1, 2, 3

2
3 3

Complex Columns: Complex Columns

Common Reboiler Common Condenser
HEIRACHICAL PROCESS SYNTHESIS
Sequencing of Complex Columns

1
1 1

1, 2, 3 2 1, 2, 3 1, 2, 3

3
3 3

Complex Columns: Complex Columns:

Both Top & Bottom Products Side Product Above or
of 1st Column as Feeds to Below Feed Point
2nd Column with
One Side Product
HEIRACHICAL PROCESS SYNTHESIS
Short-Cut Method for Multi-component Distillation
Fenske-Underwood-Gilliland (FUG)
• Fenske Equation to estimate minimum number of theoretical plate
ln LK 1 LK  HK 1 HK   
m   LK ,HK N  LK ,HK 1 
12

N min 
ln m
•
Rmin 
xD,LK x F ,LK   LK / HK x D,HK x F ,HK 
Underwood Equation to estimate minimum reflux ratio
R  1.2Rmin
 LK / HK 1
• Gilliland Equation to Estimate number of theoretical plates
N  N min   R  Rmin 0.5688 
 0.75 1   N 
N 1 
  R 1   2Nm
Eo  2.841
• Plate Efficiency: O’Connel Correlation  0.252
F

• Area of Condenser  T  T 
A  VH v
ln  dewc
U c Tcwo  Tcwi   Tdewc  Tcwo 
cwi
c

• Area of Reboiler
VH v
AR 
U R Ts  TdewR 
HEIRACHICAL PROCESS SYNTHESIS
Short-Cut Method for Multi-component Distillation
• Height of distillation tower H  0.69N Eo
0.5 12
• Diameter by using Fair Correlation for F  LM L  V     V 
  u f  C  L
 VM V   L   V
LV

C  FST FF FHAC F
1
FFT = (L/20)0.2

Distance between plate FF < 0.75

0.914 m
Cf (m/s)

0.610 m FHA = 1 if Ah/Aa > 0.1

0.1 0.457 m
0.305 m FHA = 5(Ah/Aa) + 0.5
0.229 m
if 0.06 > A h/A a > 0.1
0.127 m 12
 4VM V 
DT   
 0.6u f  1  Ad A  
 V 
0.01
0.01 0.1 1 10
F LV
HEIRACHICAL PROCESS SYNTHESIS
Preliminary PFD without Heat Exchangers
Toluene Hydro-Dealkylation Process
HEIRACHICAL PROCESS SYNTHESIS
Preliminary PFD without Heat Exchangers
Toluene Hydro-Dealkylation Process
Top Product
Design of stabilizer X R R Hydrogen T min
(kgmole h1)
Methane Benzene
(kgmole h-1) (kgmole h-1)
Toluene
(kgmole h-1)
0.1 0.354 0.530 56.194 73.780 2.478 0.000123
0.2 0.346 0.520 28.036 36.807 2.480 6.1x10-5
0.3 0.341 0.511 18.577 24.386 2.482 4.1x10-5
0.4 0.335 0.502 13.902 18.246 2.485 3.0x10-5
0.5 0.328 0.493 11.109 14.576 2.488 2.4x10-5
0.6 0.323 0.484 9.253 12.133 2.490 2.0x10-5
0.7 0.318 0.477 7.930 10.387 2.492 1.7x10-5
0.8 0.313 0.469 7.012 9.158 2.492 1.5x10-5
0.9 0.309 0.464 6.581 8.491 2.484 1.4x10-5

Bottom Product
XT Hydrogen Benzene Toluene Diphenyl
(kgmol j-1) (kgmol j-1) (kgmol j-1) (kgmol j-1)

XT Height Diameter Condenser Area Reboiler Area 0.1 0.0056 245.2841 2343.8599 0.5636
(m) (m) (m2) (m2) 0.2 0.0028 245.5596 1042.9852 0.6766
0.1 23.25 0.399 739.49 41.63 0.3 0.0019 245.7695 609.2914 0.8325
0.2 23.25 0.284 111.29 20.02 0.4 0.0014 246.0318 392.4874 1.0580
0.3 23.25 0.233 37.97 12.90 0.5 0.0011 246.2942 262.4312 1.4057
0.4 23.25 0.202 21.83 8.06 0.6 0.00093 246.5303 175.7762 1.9925
0.5 23.25 0.182 14.60 6.25 0.7 0.00079 246.6877 113.9812 3.1331
0.6 23.25 0.167 10.95 5.13 0.8 0.00070 246.7402 67.8907 5.9827
0.7 23.25 0.156 8.65 4.44 0.9 0.00066 245.9531 32.9967 19.0978
0.8 23.25 0.147 7.15 4.23
0.9 23.25 0.142 6.46 5.86
HEIRACHICAL PROCESS SYNTHESIS
Preliminary PFD without Heat Exchangers
Toluene Hydro-Dealkylation Process
Top Product
Design of benzene tower X R R T min Benzene
(kgmole h-1)
Toluene
(kgmole h-1)
0.1 7.898 11.846 245.210 0.701
0.2 3.845 5.767 245.486 0.312
0.3 2.503 3.754 245.696 0.182
0.4 1.835 2.752 245.958 0.117
0.5 1.436 2.154 246.220 0.078
0.6 1.173 1.759 246.456 0.053
0.7 0.990 1.486 246.614 0.034
0.8 0.864 1.296 246.666 0.020
0.9 0.799 1.199 245.879 0.0098

XT Height Diameter Condenser Area Reboiler Area

(m) (m) (m2) (m2)
Bottom Product 0.1 44.2 3.3 1291.64 327.61
XT Benzene Toluene Diphenyl Minimum No. of No. of 0.2 44.8 2.4 700.02 173.11
(kgmole h-1) (kgmole h-1) (kgmole h-1) no. of plates theoretical actual 0.3 45.4 2.0 494.61 122.70
plates plates 0.4 45.9 1.8 390.68 98.08
0.1 0.0736 2343.159 0.5636 19.1 38.2 59 0.5 46.4 1.6 328.70 84.06
0.2 0.0737 1042.673 0.6766 18.9 37.7 59 0.6 46.8 1.5 287.83 76.08
0.3 0.0737 609.109 0.8325 18.6 37.2 60 0.7 47.3 1.4 259.42 73.15
0.4 0.0738 392.370 1.0580 18.4 36.8 60 0.8 47.7 1.4 244.89 76.85
0.5 0.0744 262.353 1.4056 18.2 36.4 61 0.9 47.8 1.3 257.89 110.75
0.6 0.0740 175.724 1.9925 18.0 36.0 62
0.7 0.0740 113.947 3.1331 17.8 35.6 62
0.8 0.0740 67.870 5.9827 17.7 35.4 63
0.9 0.0738 32.987 19.0978 17.7 35.3 64
HEIRACHICAL PROCESS SYNTHESIS
Preliminary PFD without Heat Exchangers
Toluene Hydro-Dealkylation Process
Top Product
Design of toluene tower X R R T min Toluene
(kgmole h-1)
Diphenyl
(kgmole h-1)
0.1 0.0413 0.0621 2342.456 0.000168
0.2 0.0414 0.0621 1042.361 0.000202
0.3 0.0414 0.0621 608.927 0.000249
0.4 0.0415 0.0622 392.252 0.000316
0.5 0.0416 0.0624 262.274 0.000420
0.6 0.0418 0.0627 175.671 0.000596
0.7 0.0425 0.0637 113.913 0.000936
0.8 0.0450 0.0675 67.850 0.001788
0.9 0.0653 0.0979 32.977 0.005708

XT Height Diameter Condenser Area Reboiler Area

(m) (m) (m2) (m2)
Bottom Product 0.1 14.6 5.7 645.60 404.20
0.2 14.9 3.8 287.29 231.79
XT Toluene Diphenyl Minimum No. of No. of 0.3 15.2 2.9 167.84 169.53
(kgmole h-1) (kgmole h-1) no. of plates theoretical actual 15.5 2.3 108.12 137.31
0.4
plates plates 15.7 1.9 72.31 114.46
0.703 0.563 5.04 10.08 15
0.5
0.1 16.0 1.6 48.45 95.13
0.313 0.676 5.04 10.08 16
0.6
0.2 16.2 1.3 31.45 73.52
0.183 0.832 5.04 10.08 16
0.7
0.3 16.5 1.0 18.80 49.22
0.118 1.058 5.04 10.08 17
0.8
0.4 16.5 0.7 9.40 25.89
0.079 1.405 5.04 10.08 17
0.9
0.5
0.6 0.053 1.992 5.04 10.08 17
0.7 0.034 3.132 5.04 10.08 18
0.8 0.020 5.981 5.04 10.08 18
0.9 0.009 19.092 5.04 10.08 18
HEIRACHICAL PROCESS SYNTHESIS
Fourth Level Economic Potential
20000000

15000000

10000000
4

Economic Potential (RM)/Year

as
r
A 5000000
i

on 0 Molar Ratio
o 0 0.2 0.4 0.6 0.8 1
m
k -5000000
E
i yph=0.4
en -10000000 yph=0.1
s
ot -15000000 yph=0.2
P
yph=0.3
-20000000

-25000000

-30000000

-35000000

-40000000

-45000000
Penukaran
Conversion of Limiting Reactant
 HEIRACHICAL PROCESS SYNTHESIS
Fifth Level
 In the fifth level, need for heat exchanges is reconsidered
 Heat exchanger network (HEN) is optimized & integrated by pinch
analysis based on First & Second Law of Thermodynamics
 Targeting for minimum number of heat exchangers (Fisrt Law) and
minimum utility requirement (Second Law)
 Identification of Hot & Cold Streams
 Second Law: Minimum approach temperature difference: 10C
 First Law: Energy cascade diagram
 Second Law: Temperature-enthalpy & grand composite curves:
Identification of pinch temperature
 HEN synthesis above & below pinch temperature
 Optimization of HEN synthesis by stream splitting & removal
of loops
HEIRACHICAL PROCESS SYNTHESIS
Fifth Level
HEIRACHICAL PROCESS SYNTHESIS
Preliminary PFD with Heat Exchangers
H EIRACHICAL PROCESS SYNTHESIS
Preliminary PFD with Heat Exchangers
 HEIRACHICAL PROCESS SYNTHESIS
Hot & Cold Streams; Energy Cascade Diagram
Stream Stream Strea Tin Tout
No. Condition m (C) (C)
Entha C C
lpy / 250 Hot Cold 240
50,000 W
C Utility Utility
Temp. Int. 1
1 Hot 1000 250 120
2 Hot 4000 200 100 50,000
W
3 Cold 3000 90 150
4 Cold 6000 130 190 200 -40,000 W 190
FCp 1000 W/C 4000 W/C CTotal C 3000 W/C 6000 W/C Temp. Int. 2
250 240
1 70,000 W 10,000 W
Temp. Int. 1 = 50C -80,000 W
150 Temp. Int. 3 140
200 190 0W
40,000 W
2 Temp. Int. 4
Temp. Int. 2 = 40C 4 40,000 W
100 20,000 W 90
160 150 Temp. Int. 5
Temp. Int. 3 = 10C 60,000 W
150 140
3
Temp. Int. 4 = 30C
120 110
Temp. Int. 5 = 20C
100 90
 HEIRACHICAL PROCESS SYNTHESIS
Temperature-Enthalpy & Grand Composite Curves

250 250
Hot Enthalpy
Entalpi Panas
230 230

C 210 Cold Enthalpy

Entalpi Sejuk C) 210
o
Temperature (oC)

Temperature (oC)
) (
o
( 190 Shifted Cold
Entalpi Sejuk u 190
Enthalpy h
uh 170 Teranjak u
u S 170
S
150 150

130 130

110 110

90 90
0 100 200 300 400 500 600 0 20 40 60 80 100 120

Entalpi (kW)(kW) Entalpi (kW)

Enthalpy Enthalpy (kW)
HEIRACHICAL PROCESS SYNTHESIS
Integrated PFD
H EIRACHICAL PROCESS SYNTHESIS
Integrated PFD
 HEIRACHICAL PROCESS SYNTHESIS

Sixth Level
Poor process static & dynamic properties arise from using economic
viability for process selection causing off-spec products & excessive utilities
 Seider et al and Daud (2001) added a sixth level, where a plant-wide control
scheme is developed by using heuristics first introduced by Newell and Lee
 Selection of Control Variables:
 Heuristic 1: Select state variable representing
inventory that is not self
regulating
 Heuristic 2 Select state variable representing self regulating
inventory that transgress equipment’s limit or process condition
 Heuristik 3 Select state variable representing self
regulating inventory that interacts with another inventory
 Selection of Manipulated Variables:
 Heuristic 1: Select variable that acts directly with control variable
 Heuristic 2: Select variable that is more sensitive to control variable changes
 Heuristic 3: Select variable that acts vary fast
 Heuristic 4: Select variable that does not
interact with other control loops
 Heuristic 5: Select variable that does
not recycle any disturbance
HEIRACHICAL PROCESS SYNTHESIS
Sixth Level
HEIRACHICAL PROCESS SYNTHESIS
Mass & Energy Inventory Control: Reactor

FT
FCR

TCR
L
T
LT
TT

LCR
HEIRACHICAL PROCESS SYNTHESIS
Mass & Energy Inventory Control: Heater
HEIRACHICAL PROCESS SYNTHESIS
Distillation Control: Cut Control Top Product

PCR1
FCR1
L
P LCR1
LT1
PT1

FT1

FI1 R1

FT2

R2 FCR2

FT1
L

LT2 LCR2
HEIRACHICAL PROCESS SYNTHESIS
Distillation Control: Cut Control Bottom

PCR1

L LCR1
P FCR2
LT1
PT1

FT2

FI1 R1

FT3

R2 LCR2

LT2

FCR1

FT1
HEIRACHICAL PROCESS SYNTHESIS
Distillation Control: Product Quality Control

PCR1

FCR1 L
P QCR1
LT1
PT1 LCR1 F

FT1 QT1

T
TCR1 QCR2
TT1

L
LCR2
LT2

QT2
PROCESS SYNTHESIS
 PROCESS SYNTHESIS &
OPTIMISATION

 The third approach is the algorithmic method to

search for and optimise process alternatives
 Process synthesis involving heavy mathematical
modelling are decomposed efficiently due to very
large combinatorial flowsheet possibilities and
then optimised
 One approach is a tree search in the space of
design decisions where design decisions are
recorded at a node which can be backtracked to a
previous node & branched in different directions
 The solution is optimised by using mixed integer
linear programming (MILP)
 PROCESS SYNTHESIS &
OPTIMISATION
 Another method is the creation of a
superstructure of decisions containing most if
not all design alternatives and then using
mixed integer non linear programming (MINLP)
to optimise them
 Large superstructures might lead to very large
MINLP problems that might be unsolvable
 A viable alternative is to reduce the process
alternatives through the use of heuristics and
then optimise the reduced superstructure
using MINLP or MILP
 PROCESS SYNTHESIS &
OPTIMISATION
 The most popular non linear programming
algorithm used in process optimisation is
the successive quadratic programming
(SQP)
 requires less function evaluations

 does not require feasible points at

intemediate iterations and
 converges to an optimal solution from an
infeasible point.
 PROCESS SYNTHESIS &
OPTIMISATION

 Optimisation of reactor networks is

not very well developed mainly due to
the non-linear characteristics of
reacting systems
 Difficult to infer heuristic rules and
 Difficult to converge algorithmic
methods
 Novel method proposed by Glasser et
al. 1987 is to plot an attainable region
consisting of all the family of reactor
network solutions
 PROCESS SYNTHESIS &
OPTIMISATION
 It is sufficient to get the reactor network
at the boundary of the attainable region
because any interior point is simply the
mixture of the boundary points
 In two dimensional problems, the
reactors need to be continuous stirred
tank reactors (CSTR) and plug flow
reactors (PFR) only
 The remaining problem is the integration
of reactor networks with the separation
system
 CURRENT AND FUTURE
DEVELOPMENT
 More efforts should be devoted to the
generic modelling of
 adsorption
 membrane
 solid drying
 solids handling especially fluidisation and
pneumatic conveying
 Further work on integrating of process
control and process synthesis should be
developed using the structural control
matrix approach
 CURRENT AND FUTURE
DEVELOPMENT
 Important issues being neglected are
 safe design and operation and
 waste minimisation
 Heuristic approach of Kletz using keywords
like intensification, substitution, and
attenuation pioneered chemical process plant
design for safety
 Recently rapid risk analysis is used to
inhenrently safe design by Khan & Abbasia
2000
 A related issue is design for maintainability
 CURRENT AND FUTURE
DEVELOPMENT
 The minimum addition of chemical species
and their minimum production and
rejection in the mass exchange network
pioneered by El-Halwagi using the
minimum number of “mass exchangers”
can minimise wastes
 Flower et al first proposed the use of mass
exchange networks for waste minimisation
 Recently Noureldina & El-Halwagi reported
a mass exchange network-based method
for pollution prevention
 CURRENT AND FUTURE
DEVELOPMENT
 A method proposed recently by
Dantus & Higha is to evaluate source
reduction alternatives by
 economic performance including
waste related costs in an
environmental accounting
framework and
 the environmental impact of the
alternative
 CURRENT AND FUTURE
DEVELOPMENT
 A new method which is now becoming the trend is the
combination of
 economic objectives and
 life cycle assessment (LCA)-based
environmental objectives
 Uses goal programming to identify the Pareto surfa
of non inferior solutions
 More research should be directed at
incorporating waste minimisation and
environmental impact ideas in the
heuristics-based method of Douglas
 Thanks for Watching
Please follow me / SAJJAD KHUDHUR ABBAS